Welcome international healthcare professionals

This site is no longer supported and will not be updated with new content. You are welcome to browse and download all content already included in the site. Please note you will have to register your email address to access the site.

You are here

DDX43 promoter is frequently hypomethylated and may predict a favorable outcome in acute myeloid leukemia

Leukemia Research


DEAD box polypeptide 43 (DDX43), a cancer/testis antigen (CTA), has been found to be overexpressed in various solid tumors and some hematologic malignancies. In the present work hypomethylation of the DDX43 gene was detected in 15% (32/214) of primary acute myeloid leukemia (AML) using real-time quantitative methylation-specific PCR (RQ-MSP). The level of DDX43 expression was correlated with DDX43 hypomethylation (R = 0.277, P = 0.014). Moreover, bisulfite sequencing confirmed the significant correlation between the methylation density and the level of DDX43 hypomethylation. Additionally, restoration of DDX43 expression in the K562 cell line by 5-aza-2′-deoxycytidine treatment confirmed a direct contribution of methylation in regulating the DDX43 gene. DDX43 hypomethylation was observed more frequently in favorable group (21.4%) and intermediate group (15.8%) than in poor group (0%) (P = 0.009). AML patients with DDX43 hypomethylation had a better overall survival (median not obtained) than those with DDX43 methylation (median 8 months, 95% confidence interval 5.6–10.4 months) (P = 0.014). In summary, the DDX43 gene is activated by promoter hypomethylation and DDX43 hypomethylation may be a favorable prognostic factor in AML.

Keywords: Acute myeloid leukemia, DDX43, Hypomethylation, Prognosis.

1. Introduction

Acute myeloid leukemia (AML) is a group of common disorders derived from haemopoietic progenitor cells, which lose the ability to differentiate normally and to respond to normal regulators of proliferation [1] . So far, it is well-known that genetic abnormalities play a pivotal role in the pathogenesis of AML [2] and [3]. Recently, it has been demonstrated that epigenetic aberrations, such as alterations in the DNA methylation, the histone modification patterns and miRNA expression are also involved in the development of leukemia [4], [5], [6], and [7]. Aberrant methylation of numerous genes, such as p15, p73, HIC1, RARβ2, DAPK1, SOCS-1, E-cadherin, CTNNA1, and ER have been identified in AML [8] and [9].

Cancer/testis antigens (CTAs) provide promising targets for cancer-specific immunotherapy due to their expression in a broad spectrum of cancers and limited expression in normal tissues such as testis and placenta [10] . A new member of the DEAD (Asp-Glu-Ala-Asp)-box family of helicases, DDX43 (DEAD box polypeptide 43, also known as HAGE), was first identified together with sarcoma antigen (SAGE) as a tumor-specific CTA gene in a human sarcoma cell line [11] . So far, DDX43 has been found with overexpression in various solid tumors such as salivary gland, colon, brain, lung and prostate cancers and hematologic malignancies (e.g., chronic myeloid leukemia and multiple myeloma) [10], [12], [13], [14], [15], and [16]. Recently, it was also shown that the abnormal hypomethylation of DDX43 promoter was present in chronic myeloid leukemia (CML) and was associated with poor outcome [15] . However, the pattern of DDX43 methylation has not been studied in AML. In this study, we aimed to analyze the methylation status of DDX43 promoter and its clinical implications in primary AML patients.

2. 2 Materials and Methods

2.1. Patients and Samples

Two hundred and fourteen patients with primary AML presented at the Affiliated People's Hospital of Jiangsu University were selected for this investigation based on the availability of stored leukemic cells. Cytogenetic data were acquired from 201 patients. The diagnosis and classification of AML patients were made according to the revised French–American–British (FAB) classification and the 2008 World Health Organization proposal [17] and [18]. The risk grouping was taken according to the cytogenetical and molecular abnormalities [19] . Clinical and laboratory features of all patients were summarized in Table 1 . After obtaining informed consent, the bone marrow (BM) aspirates from all patients were collected at the time of diagnosis. BM specimens obtained from 24 iron deficiency anemia (IDA) individuals were used as controls. Bone marrow mononuclear cells (BMNCs) were separated by density gradient centrifugation using Ficoll solution and washed twice with PBS.

Table 1 The hypomethylation of DDX43 gene promoter in patients with AML.

Patient's parameter The status of DDX43 methylation
Hypomethylated (n = 32) Methylated (n = 182) Total (n = 214) P-value
Age (years) a 48.5 (11–83) 47 (2–93) 47 (2–93) 0.870
Sex (male/female) 24/8 99/83 123/91 0.033
WBC (×109/l) a 17.2 (0.9–528.0) 14.1 (0.7–528.0) 14.1 (0.7–528.0) 0.996
Hemoglobin (g/l) a 78 (40–126) 71 (31–147) 74 (31–147) 0.233
Platelets (×109/l) a 35 (3–157) 38 (4–447) 38 (3–447) 0.909
FAB, no.       0.019
 M1 3 23 26  
 M2 6 80 86  
 M3 10 21 31  
M4 7 32 39  
M5 4 20 24  
M6 2 6 8  
Risk grouping       0.009
Favorable 15 55 70  
Intermediate 16 85 101  
Poor 0 30 30  
Karyotyping       0.012
Normal 13 74 87  
t(8;21) 3 24 27  
inv(16) 0 1 1  
t (15;17) 10 18 28  
+8 0 7 7  
−5/5q-, −7/7q- 0 11 11  
t(11q23) 0 5 5  
t(9;22) 0 4 4  
Complex 0 9 9  
NPM1       1.000
Mutant 3 17 20  
Wild-type 29 162 191  
FLT3-ITD       1.000
Mutant 1 3 4  
Wild-type 31 176 207  
IDH1       1.000
Mutant 1 4 5  
Wild-type 31 175 206  
IDH2       0.698
Mutant 1 11 12  
Wild-type 31 168 199  
IDHI and IDH2       0.476
Mutant 1 15 16  
Wild-type 31 164 195  
DNMT3A       0.697
Mutant 1 12 13  
Wild-type 31 167 198  
DDX43 transcripts (%) a 977.96 (59.00–9843.00) 0.02 (0.00–128.00) 0.61 (0.00–9843.00) <0.001

a Median (range).

WBC, white blood cells; FAB, French–American–British classification; AML, acute myeloid leukaemia.

The follow-up data were obtained for 124 cases. The median follow-up duration of the patients was 8 months (range, 1–73 months).

2.2. 5-aza-2′-Deoxycytidine Treatment

The leukemic cell line K562 was plated at a density of 1 × 106/ml and was cultured in 5 ml RPMI 1640 medium at 37 °C in a humidified atmosphere containing 5% CO2. 5-aza-2′-deoxycytidine (DAC) (Sigma-Aldrich, Steinheim, USA) diluted in dimethyl sulfoxide (DMSO) was added in four flasks of K562 cells once a day at the same time at different final concentrations of 0.1 μM, 1 μM, 10 μM and 50 μM as experimental group. K562 cells only added with DMSO were set as the DAC controls, and the volume of DMSO was the same as that used in diluting DAC. K562 cells without treatment were also used as the control. All cells were cultured until harvested for extraction of RNA and DNA.

2.3. Real-Time Quantitative Methylation-Specific PCR

The basic methods were briefly described as previously reported [20] . Genomic DNA was obtained from BMNCs using DNA Purification Kit (Gentra, Minneapolis, MN, USA). 1 μg of genomic DNA was sodium bisulphite-modified as described in manufacturer instruction using the CpGenome™ DNA Modification kit (Chemicon, Ternecula, CA, USA). Modified DNA was used for real-time quantitative methylation-specific PCR (RQ-MSP) immediately or stored at −80 °C at once until analyzed.

RQ-MSP was performed using methylation-specific and unmethylation-specific primers and ALU repetitive sequence in a 7300 Thermal Cycler (Applied Biosystems, Foster City, CA, USA). Primer sequences for the methylated (M) RQ-MSP reaction were 5′-GGAGGAGTTTTTAAGGTTTTTACGT-3′ (forward) and 5′-GACAATTCC TCGTAACCAACG-3′ (reverse), and primer sequences for the unmethylated (U) RQ-MSP reaction were 5′-GGAGGAGTTTTTAAGGTTTTTATGT-3′ (forward) and 5′-ACAACAATTCCTCATAACCAACAA -3′ (reverse). ALU repetitive sequence was used as reference sequence. Each PCR reaction included 2 μl of modified DNA, MgCl2 2.0 mmol/l, 10 × PCR buffer, 1U Taq DNA polymerase (MBI Fermentas, Amherst, NY, USA), 200 nmol/l of each primer, 200 μmol/l dNTP, 20 × EvaGreen 1.2 μl, and 50 × ROX 0.5 μl. Both methylated and unmethylated PCR running protocols consisted of an initial denaturation step of 4 min at 94 °C, followed by a amplification program of 40 cycles at 94 °C for 30 s (denaturation), 61 °C for 30 s (anneaing), 72 °C for 30 s (extension), 82 °C for 30 s (data collection), and finally, a melting program of one cycle at 95 °C for 15 s, 60 °C for 60 s, 95 °C for 15 s and 60 °C for 15 s. Distilled water without DNA was used as negative control as well as recombined methylated and unmethylated DDX43 plasmids were positive controls for each set of PCR. The normalized ratio (Nunmethylation-DDX43) was calculated in relation to the reference ALU sequence [21] and was used to assess the degree of methylation of DDX43 promoter in samples. The standard curves were established using recombined methylated and unmethylated DDX43 plasmids as well as ALU plasmids from 1 × 108 copies/μl to 10 copies/μl before determining the cutoff level of specific fluorescence for the three corresponding sequences in samples. Nunmethylation-DDX43 was calculated according to the following formula. Nunmethylation-DDX43 = (Eunmethylation-DDX43)ΔCTunmethylation-DDX43(control-sample) ÷ (EALU)ΔCTALU(control-sample). CT was the cycle number when the fluorescent signal intensity of detected gene in each reaction reaches to the setting threshold. ΔCT was the difference in CT value between control and sample, either for target (unmethylated DDX43 sequences) or reference sequences. E was the amplification efficiency of each PCR reaction and was computed as E = 10(−1/slope). Each sample was detected twice.

2.4. Real-Time PCR

Total RNA was extracted using Trizol reagent (Invitrogen, Carlsbad, CA, USA) in accordance with the manufacturer's standard method from BMNCs. Reverse transcription reaction was carried out with 2 μg total RNA in a final volume of 40 μl, containing 10 μmol/l of random primers, 0.5 mmol/l of dNTPs, 200 U of MMLV reverse transcriptase, 10 mmol/l of dithiothreitol and 25 U of RNase inhibitor. After cDNA synthesis, real time quantitative PCR (RQ-PCR) was performed to detect mRNA expression of DDX43 gene using of the specific primers [15] . Specificity of amplification products was confirmed by melting curve analysis. The mRNA abundance of DDX43 gene was calculated relative to the expression of the housekeeping gene, ABL1.

2.5. Agarose Gels Electrophoresis

Both PCR products of DDX43 methylation and DDX43 expression were analyzed on 2% agarose gels and visualized under UV illumination after staining with ethidium bromide.

2.6. Bisulfite-Sequencing

To further determine the comprehensive methylation status of the DDX43 promoter and exon 1 in AML patients, bisulfite-modified DNA sequencing was performed in five DDX43-hypomethylated and five DDX43-methylated AML samples in accordance with the result of RQ-MSP. One microgram of genomic DNA was modified as above described. The specific sequencing PCR primers were as previously reported [20] . The 307-bp PCR products, which covered 21 CpG sites (+12bp∼ + 250 bp relative to the DDX43 translation start site) within a CpG-rich region located in DDX43 promoter and exon 1, were cloned into pMD®.19-T Vector (TaKaRa, Dalian, China). 9 to 10 independent colonies of each sample were sequenced.

2.7. Mutations Analysis

NPM1, IDH1 (R132), IDH2 (R140 and R172) and DNMT3A mutations were detected by high-resolution melting analysis (HRMA) as reported previously [22], [23], and [24]. Briefly, genomic DNA samples were amplified using gene-specific primers. Mutation scanning was performed for PCR products using HRMA with the LightScanner™ platform (Idaho Technology Inc. Salt Lake City, Utah). All positive samples were directly DNA sequenced to confirm the results of HRMA. FLT3 internal tandem duplication (ITD) was detected using direct DNA sequencing [25] .

2.8. Statistical Analysis

Statistical analysis was performed using the SPSS 17.0 software package (SPSS, Chicago, IL, USA). Pearson chi-square analysis and Fisher exact test were carried out to compare the difference of categorical variables between patients group. Mann–Whitney's U-test was carried out to compare the difference of continuous variables between patient groups. The correlation between the frequency of DDX43 promoter hypomethylation and the clinical and hematologic parameters was analyzed with Spearman's rank correlation. Survival was analyzed according to the Kaplan-Meier method and differences in the distribution were evaluated by means of the log-rank test. A multivariate analysis using the Cox regression backward stepwise likelihood ratio was also performed. Hazard ratios (HRs) were given within 95% confidence intervals (CIs). For all analyses, the P-values were two-tailed, and a P-value of less than 0.05 was considered statistically significant.

3. Results

3.1. The Methylation Status of DDX43 Promoter in Controls

The maximal sensitivity of RQ-MSP was 100 copies/μl. The sample was considered as positive if the CT value of the sample was clearly outside the CT value range of negative controls (e.g. at least one cycle lower than the lowest CT value of the non-specific amplification) and within certain distance (e.g. four cycles) from the maximal sensitivity of 100 copies/μl, otherwise the sample was considered as negative. The control sample which showed the lowest difference in CT between the target (unmethylated DDX43 sequences) and the reference (ALU sequences) sequences was used as control/calibrator sample for quantification of the DDX43 unmethylation in both controls and AML patients [15] . It was considered as 100% of DDX43 unmethylation.

In this study, we analyzed DDX43 unmethylation level in 24 IDA samples. The presence of specific unmethylated products in the melting curves confirmed 23 specimens displayed slight unmethylation of DDX43. The CT values for DDX43 unmethylation in controls were 27.14–35.01 (30.18 ± 2.13). Nunmethylation-DDX43 ratio of all controls was 0–100% (31.78 ± 27.00%). Thus, a Nunmethylation-DDX43 ratio equal to or above 112.78% (determined as the mean plus 3S.D.) was chosen to define the DDX43 hypomethylation in controls and AML samples.

3.2. Methylation Status of DDX43 Promoter and Its Correlation with Clinical Characteristics in AML

DDX43 hypomethylation was determined in 32 of 214 (15.0%) de novo AML patients. The representative results of RQ-MSP are shown in Fig. 1 . A significant correlation was observed between the frequency of DDX43 hypomethylation and patients’ sex. The frequency of DDX43 hypomethylation in male (24/123, 19.5%) was higher than that in female (8/91, 8.8%), the difference was statistically significant (P = 0.033). There was no significant difference in the age, white blood cell count, haemoglobin concentration and platelet count between the patients with and without DDX43 hypomethylation ( Table 1 ).


Fig. 1 Electrophoresis results of RQ-PCR and RQ-MSP products of DDX43 gene in AML patients. A: DDX43 expression; B: ABL expression; C: DDX43 unmethylation; D: DDX43 methylation; and E: ALU. 1: 100 bp DNA Ladder; 2–8: AML; 9: IDA; 10: positive control; 11: negative control.

3.3. Association of DDX43 Hypomethylation with DDX43 Overexpression in AML

The level of DDX43 expression was examined in 78 AML patients with available mRNA. A positive correlation was observed between the level of DDX43 expression and DDX43 unmethylation (R = 0.277, P = 0.014). DDX43 hypomethylated patients (n = 9) had significantly higher level of DDX43 expression (median 977.96%) than those DDX43 methylated (n = 69) (median 0.02%) (P < 0.001) and than controls (n = 24) (median 0.00%) (P < 0.001) ( Table 1 ).

3.4. Analysis on the Density of DDX43 Methylation by Bisulfite-Sequencing

In order to confirm the results of RQ-MSP, we evaluated the methylation density of 21-CpGs in DDX43 promoter in five DDX43-hypomethylated and five DDX43- methylated AML samples according to RQ-MSP. The results of RQ-MSP and bisulfite-sequencing were highly correlated (R = −0.915, P < 0.001) ( Fig. 2 ) ( Table 2 ).


Fig. 2 Methylation density of DDX43 promoter and exon 1 identified in five DDX43-hypomethylated and five DDX43-methylated AML patients by bisulfite sequencing. White cycle: unmethylated CpG dinucleotide; Black cycle: methylated CpG dinucleotide. a: M5b; b: M5b; c: M4b; d: M3; e: M3; f: M4a; g: M3; h: M2; i: M4a; and j: M2a. The results of RQ-MSP and bisulfite-sequencing were highly negatively correlated (R = −0.915, P < 0.001).

Table 2 The results of five DDX43-hypomethylated and five DDX43-methylated AML patients for bisulfite sequencing.

No. DDX43 hypomethylation DDX43 methylation density (%)
A 6.79 13.8
B 6.53 14.7
C 4.31 50.0
D 3.14 57.1
E 3.45 63.8
F 0.02 88.1
G 0.37 91.4
H 0.25 95.2
I 0.04 97.1
J 0.01 98.6

3.5. Analysis on the Cell Line Treated with DAC

The level of unmethylated DDX43 promoter increased in a dose-dependent manner and a dose-related in K562 cells treated with DAC. Accordingly, DDX43 expression was significantly up-regulated after DAC treatment ( Fig. 3 ).


Fig. 3 Electropheresis results of RQ-PCR and RQ-MSP of K562 cells before and after 4-day treatment with DAC at different concentrations. A: DDX43 expression; B: ABL expression; C: DDX43 unmethylation; D: DDX43 methylation; and E: ALU. 1: 100 bp DNA Ladder; 2: 0.1 μM DAC; 3: 1 μM DAC; 4: 10 μM DAC; 5: 50 μM DAC; 6: 0.1 μM DAC control; 7: 1 μM DAC control; 8: 10 μM DAC control; 9: 50 μM DAC control; 10: non-treated K562 control; 11: positive control; 12: negative control.

3.6. Association Between DDX43 Promoter Hypomethylation and AML Subtypes

Aberrant hypomethylation of DDX43 promoter could be observed in every subtype. DDX43 hypomethylation frequency was M3 (32.3%) > M6 (25.0%) > M4 (17.9%) > M5 (16.7%) > M1 (11.5%) > M2 (7.0%) among FAB subtypes, and the difference was statistically significant (P = 0.019) ( Table 1 ).

3.7. Association of DDX43 Promoter Hypomethylation with Karyotypes and Molecular Abnormalities

The frequency of DDX43 promoter hypomethylation was associated with risk groups in AML ( Table 1 ). DDX43 hypomethylation was observed more frequently in favorable group (21.4%) and intermediate group (15.8%) than in poor group (0%) ( Table 1 ). For further analysis, the DDX43 hypomethylation was observed in patients with t(15;17) (10/28, 35.7%), normal cytogenetics (13/87, 14.9%) and t(8;21) (3/27, 11.1%). However, the DDX43 hypomethylation was not found in any patients with 11q23 rearrangement, complex abnormalities, t(9;22), −5/5q- or −7/7q- ( Table 1 ).

3.8. Association Between DDX43 Hypomethylation and Prognosis

The estimated 50% survival time of all patients was 8 months (95% confidence interval 4.4–11.6 months). Univariate analysis showed that the patients with DDX43 hypomethylation had a better overall survival (OS) (median not obtained) than those with DDX43 methylation (median 8 months, 95% confidence interval 5.6–10.4 months) (P = 0.014, Fig. 4 ). Furthermore, multivariate analysis identified DDX43 hypomethylation has a tendency to be an independent prognostic factor (P = 0.061) besides the age and risk grouping ( Table 3 ).


Fig. 4 Overall survival of AML patients according to Kaplan–Meier analysis.

Table 3 Multivariate analyses of prognostic factors for overall survival in AML.

  P-value HR 95% CI for HR
Sex (male/female) 0.729 0.916 0.557–1.505
Age (<60 versus >60 years) <0.001 2.410 1.498–3.878
WBC (×109/L) (<30 versus >30) 0.153 1.428 0.876–2.327
Hemoglobin (g/L) (<110 versus >110) 0.988 1.005 0.502–2.012
Platelet (×109/L) (<100 versus >100) 0.670 1.130 0.644–1.981
Risk grouping (favorable/intermediate/poor) <0.001 2.042 1.457–2.863
DDX43 hypomethylation (yes versus no) 0.061 0.375 0.135–1.044

WBC, white blood cells; HR, hazard ratio; CI, confidence interval.

3.9. Association of DDX43 Hypomethylation with Gene Mutations

NPM1, FLT3-ITD, IDH1/IDH2 and DNMT3A mutations were detected in 211 AML patients. No significant difference was observed in the frequency of DDX43 hypomethylation between patients with and without these mutations ( Table 1 , P > 0.05).

4. Discussion

Aberrant methylation of tumor suppressor gene promoters has been identified as an important mechanism contributing to the pathogenesis and progression of neoplasms of different tissue types including AML, however, oncogene methylation profiles have seldom been studied in malignancies. Global DNA hypomethylation and gene-specific hypomethylation are common molecular alterations in human cancers and play an important role in cancer [26] . DNA hypomethylation may be responsible for the development of malignancies through the following mechanisms: generation of chromosomal instability, reactivation of transposable elements, and loss of imprinting [27] and [28]. A large fraction of human cancers have shown gene-specific hypomethylation, such as BAGE and MAGE-A1 in colon cancer, XAGE-1 in gastric cancer, GAGE in endometrial cancer [26] , and PRAME in myelodysplastic syndrome [21] . In this study, we investigated the hypomethylation pattern of the DDX43 promoter and its clinical relevance in AML.

Although DDX43 gene is suggested as a candidate oncogene due to its high expression in cancers, but details regarding its clear physical function has been rare. DDX43 is a new member of RNA helicase family, which drive RNA metabolism including transcription, pre-mRNA splicing, ribosome biogenesis, translation initiation/elongation, and RNA decay [29] . Previous studies have implicated that certain RNA helicases such as DDX1, DDX5 (p68), DDX53 (CAGE) play a significant role in tumor cell development and proliferation in many kinds of cancers [30], [31], and [32]. Furthermore, DDX5 has been suggested to act in cell proliferation and cancer by regulating of H-RAS expression [33] . Recently, Linley et al. has demonstrated that DDX43 promotes melanoma-initiating cell (MMIC)-dependent tumorigenesis by up-regulating N-RAS expression with a concomitant oncogenic activation of NRAS/ERK/AKT signaling pathway [34] . We found that the hypomethylation of DDX43 promoter was frequently present in each subtype of AML patients but was absent in controls and DDX43 expression was regulated by promoter methylation, indicating hypomethylation of DDX43 promoter may be associated with the pathogenesis of AML. Roman-Gomez et al. identified that DDX43 hypomethylation was associated with disease progression and poor outcome in CML. However, we found that DDX43 hypomethylation was associated with favorable/intermediate-risk groupings in AML. Moreover, univariate analysis identified DDX43 hypomethylation as a favorable prognostic factor and multivariate analysis also identified DDX43 hypomethylation has a tendency to be a favorable prognostic factor in AML. Furthermore, our previous study showed that DDX43 hypomethylation was also associated with better outcome in MDS [20] . These results suggest that DDX43 gene may play a different role in different diseases. During the development and progression of acute leukemia, DDX43 may be activated by promoter hypomethylation and inhibit the proliferation of specific leukemic cells. Further functional study should be needed to determine its significance in acute leukemia. There is a true case in the different roles of PRAME gene in leukemia. In normal hematopoietic progenitors, forced PRAME expression inhibited myeloid differentiation independent on all-trans retinoic acid (ATRA) but did not alter proliferation [35] . PRAME protein overexpression promoted proliferation and inhibited myeloid differentiation only in ATRA-responsive leukemic cell lines such as HL-60 and NB4. However, PRAME silencing promoted the proliferation of both K562 cell line and primary CML cells [35] and [36].

There was a significant sex difference in the patients with and without DDX43 hypomethylation. We analyzed the association of the sex grouping and the cytogenetic features, but no significant correlation was observed. The underlying mechanism for gender discrepancy is unknown. This phenomenon might be explained by the fact that smokers in China are mainly male. Cigarette smoking can affect the epigenome and might mediate risk for diseases and cancers [37] . Recurrent mutations of several genes (TET2, DNMT3A, and IDH1/IDH2) that regulate cytosine methylation have been recently found in AML [38] . Somatic mutations of these genes alter specific biochemical functions of their gene products and may result in abnormal regulation of gene expression through altered epigenetic mechanisms. The association between the methylation status of DDX43 promoter and the mutations in three methylation modifiers was analyzed. However, no significant difference was observed in DDX43 hypomethylation between patients with and without IDH1/IDH2 or DNMT3A mutations. Further analysis also did not find the association of DDX43 hypomethylation with IDH1/IDH2 or DNMT3A mutations in patients with normal karyotypes. There was also no significant difference between two prognosis-associated mutations of NPM1 or FLT3-ITD and DDX43 hypomethylation.

Immunotherapy is a promising choice in malignancies. However, the heterogeneity of expression of the targeted antigens has been a potential difficulty in developing tumor vaccines [39] . CTAs were originally identified as proteins containing epitopes that induced cell-mediated immune responses in cancer patients and are potentially ideal targets for tumor-specific immunotherapy due to their specific distribution of expression in cancers. DDX43 may well be an ideal tumor vaccine since it has been identified strongly immunogenic as a whole antigen [40] and several MHC class I/II DDX43-derived immunogenic peptides are found [41] . Moreover, the expression of DDX43 can be induced by demethylating agent 5-aza-2ʹ-deoxycytidine, which is now an important drug used in the treatment of MDS and refractory AML. Thus, DDX43-targeted immunotherapy may be promising as a novel strategy to augment the efficacy of DNA methylation inhibitors, especially in old patients who cannot tolerate the toxic reaction of chemotherapy.

In conclusion, the present study shows for the first time that the DDX43 promoter is frequently hypomethylated in AML and DDX43 hypomethylation is a favorable prognostic factor in AML.

5. Conflict of Interest Statement

The authors declare no conflict of interest.


This study was supported by the National Natural Science Foundation of China (81270630, 81172592), 333 Project of Jiangsu Province (BRA2011085, BRA2013136), Science and Technology Special Project in Clinical Medicine of Jiangsu Province (BL2012056), Clinical Medical Science and Development Foundation of Jiangsu University (JLY20120011, JLY20120013), Innovation Project of Graduate Student in Jiangsu Province's Ordinary University (CXLX12_0673) and Social Development Foundation of Zhenjiang (SH2011056, SH2013082, FZ2011054).

LJ, QJ, DZQ and QW designed the study, analyzed the data and wrote the paper; CQ, YJ and MJC performed all experiments; CXX, AC and TCY were involved in the delivery of the clinical data; XDS and MYJ contributed to the technique support and writing of the manuscript.


  • [1] E. Estey, H. Döhmer. Acute myeloid leukaemia. Lancet. 2006;368:1894-1907 Crossref.
  • [2] L.M. Kelly, D.G. Gilliland. Genetics of myeloid leukemia. Annu Rev Genomics Hum Genet. 2002;3:179-198 Crossref.
  • [3] T.C. Lin, H.A. Hou, W.C. Chou, D.L. Ou, S.L. Yu, H.F. Tien, et al. CEBPA methylation as a prognostic biomarker in patients with de novo acute myeloid leukemia. Leukemia. 2011;25:32-40
  • [4] E. Jost, O. Galm. EHA scientific workshop report: the role of epigenetics in hematological malignancies. Epigenetics. 2007;2:71-79 Crossref.
  • [5] M. Fabbri, R. Garzon, M. Andreeff, H.M. Kantarjian, G. Garcia-Manero, G.A. Calin. MicroRNAs and noncoding RNAs in haematological malignancies: molecular, clinical and therapeutic implications. Leukemia. 2008;22:1095-1105 Crossref.
  • [6] S. Yendamuri, G.A. Calin. The role of microRNA in human leukaemia: a review. Leukemia. 2009;23:1257-1263 Crossref.
  • [7] J. Chen, O. Odenike, J.D. Rowley. Leukaemogenesis: more than mutant genes. Nat Rev Cancer. 2010;10:23-36 Crossref.
  • [8] J. Boultwood, J.S. Wainscoat. Gene silencing by DNA methylation in haematological malignancies. Br J Haematol. 2007;138:3-11 Crossref.
  • [9] Y. Ye, M.A. McDevitt, M. Guo, W. Zhang, O. Galm, S.D. Gore, et al. Progressive chromatin repression and promoter methylation of CTNNA1 associated with advanced myeloid malignancies. Cancer Res. 2009;69:8482-8490 Crossref.
  • [10] S.P. Adams, S.S. Sahota, A. Mijovic, B. Czepulkowski, R.A. Padua, G.J. Mufti, et al. Frequent expression of HAGE in presentation chronic myeloid leukaemias. Leukemia. 2002;16:2238-2242 Crossref.
  • [11] V. Martelange, C. De Smet, E. De Plaen, C. Lurquin, T. Boon. Identification on a human sarcoma of two new genes with tumor-specific expression. Cancer Res. 2000;60:3848-3855
  • [12] H. Nagel, R. Laskawi, H. Eiffert, T. Schlott. Analysis of the tumour suppressor genes, FHIT and WT-1, and the tumour rejection genes, BAGE, GAGE-1/2, HAGE, MAGE-1, and MAGE-3, in benign and malignant neoplasms of the salivary glands. Mol Pathol. 2003;56:226-231 Crossref.
  • [13] M.G. Mathieu, A.J. Linley, S.P. Reeder, C. Badoual, E. Tartour, R.C. Rees, et al. HAGE, a cancer/testis antigen expressed at the protein level in a variety of cancers. Cancer Immun. 2010;10:2
  • [14] M. Condomines, D. Hose, P. Raynaud, M. Hundemer, J. De Vos, M. Baudard, et al. Cancer/testis genes in multiple myeloma: expression patterns and prognosis value determined by microarray analysis. J Immunol. 2007;178:3307-3315
  • [15] J. Roman-Gomez, A. Jimenez-Velasco, X. Agirre, J.A. Castillejo, G. Navarro, E. San Jose-Eneriz, et al. Epigenetic regulation of human cancer/testis antigen gene, HAGE, in chronic myeloid leukemia. Haematologica. 2007;92:153-162 Crossref.
  • [16] A.P. Liggins, S.H. Lim, E.J. Soilleux, K. Pulford, A.H. Banham. A panel of cancer-testis genes exhibiting broad-spectrum expression in haematological malignancies. Cancer Immun. 2010;10:8
  • [17] J.M. Bennett, D. Catovsky, M.T. Daniel, G. Flandrin, D.A. Galton, H.R. Gralnick, et al. Proposed revised criteria for the classification of acute myeloid leukaemia. A report of the French–American–British cooperative group. Ann Intern Med. 1985;103:620-625 Crossref.
  • [18] S.H. Swerdlow, E. Campo, N.L. Harris, E.S. Jaffe, S.A. Pileri, H. Stein (Eds.) et al. WHO classification of tumours of haematopoietic and lymphoid tissues (IARC Press, Lyon, France, 2008)
  • [19] National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: acute myeloid leukemia version 2.2012. (National Comprehensive Cancer Network, Fort Washington, 2012)
  • [20] Q. Chen, J. Lin, D.M. Yao, J. Qian, W. Qian, J. Yang, et al. Aberrant hypomethylation of DDX43 promoter in myelodysplastic syndrome. Br J Haematol. 2012;158:293-296 Crossref.
  • [21] J. Qian, Z.H. Zhu, J. Lin, D.M. Yao, Y. Li, J. Yang, et al. Hypomethylation of PRAME promoter is associated with poor prognosis in myelodysplastic syndrome. Br J Haematol. 2011;154:153-155 Crossref.
  • [22] J. Qian, D.M. Yao, J. Lin, W. Qian, C.Z. Wang, H.Y. Chai, et al. U2AF1 mutations in Chinese patients with acute myeloid leukemia and myelodysplastic syndrome. PLoS One. 2012;7:e45760 Crossref.
  • [23] J. Lin, D.M. Yao, J. Qian, Q. Chen, W. Qian, Y. Li, et al. IDH1 and IDH2 mutation analysis in Chinese patients with acute myeloid leukemia and myelodysplastic syndrome. Ann Hematol. 2012;91:519-525 Crossref.
  • [24] J. Lin, D.M. Yao, J. Qian, Q. Chen, W. Qian, Y. Li, et al. Recurrent DNMT3A R882 mutations in Chinese patients with acute myeloid leukemia and myelodysplastic syndrome. PLoS One. 2011;6:e26906 Crossref.
  • [25] P.D. Kottaridis, R.E. Gale, M.E. Frew, G. Harrison, S.E. Langabeer, A.A. Belton, et al. The presence of a FLT3 internal tandem duplication in patients with acute myeloid leukemia (AML) adds important prognostic information to cytogenetic risk group and response to the first cycle of chemotherapy: analysis of 854 patients from the United Kingdom Medical Research Council AML 10 and 12 trials. Blood. 2001;98:1752-1759 Crossref.
  • [26] I.P. Pogribny, F.A. Beland. DNA hypomethylation in the origin and pathogenesis of human diseases. Cell Mol Life Sci. 2009;66:2249-2261 Crossref.
  • [27] A.S. Wilson, B.E. Power, P.L. Molloy. DNA hypomethylation and human diseases. Biochim Biophys Acta. 2007;1775:138-162 Crossref.
  • [28] M. Esteller. Epigenetics in cancer. N Engl J Med. 2008;358:1148-1159 Crossref.
  • [29] O. Cordin, J. Banroques, N.K. Tanner, P. Linder. The DEAD-box protein family of RNA helicases. Gene. 2006;367:17-37 Crossref.
  • [30] L.C. Amler, J. Schürmann, M. Schwab. The DDX1 gene maps within 400 kbp 5ʹ to MYCN and is frequently coamplified in human neuroblastoma. Genes Chromosomes Cancer. 1996;15:134-137 Crossref.
  • [31] F.V. Fuller-Pace, H.C. Moore. RNA helicases p68 and p72: Multifunctional proteins with important implications for cancer development. Future Oncol. 2011;7:239-251 Crossref.
  • [32] E. Por, H.J. Byun, E.J. Lee, J.H. Lim, S.Y. Jung, I. Park, et al. The cancer/testis antigen CAGE with oncogenic potential stimulates cell proliferation by up-regulating cyclins D1 and E in an AP-1- and E2F-dependent manner. J Biol Chem. 2010;285:14475-14485 Crossref.
  • [33] M. Camats, S. Guil, M. Kokolo, M. Bach-Elias. p68 RNA helicase (DDX5) alters activity of cis- and trans-acting factors of the alternative splicing of H-Ras. PLoS One. 2008;3 e2926
  • [34] A.J. Linley, M.G. Mathieu, A.K. Miles, R.C. Rees, S.E. McArdle, T. Regad. The helicase HAGE expressed by malignant melanoma-initiating cells is required for tumor cell proliferation in vivo. J Biol Chem. 2012;287:13633-13643 Crossref.
  • [35] V.G. Oehler, K.A. Guthrie, C.L. Cummings, K. Sabo, B.L. Wood, T. Gooley, et al. The preferentially expressed antigen in melanoma (PRAME) inhibits myeloid differentiation in normal hematopoietic and leukemic progenitor cells. Blood. 2009;114:3299-3308 Crossref.
  • [36] N. Tajeddine, J.L. Gala, M. Louis, M. Van Schoor, B. Tombal, P. Gailly. Tumor-associated antigen preferentially expressed antigen of melanoma (PRAME) induces caspase-independent cell death in vitro and reduces tumorigenicity in vivo. Cancer Res. 2005;65:7348-7355 Crossref.
  • [37] K.W. Lee, Z. Pausova. Cigarette smoking and DNA methylation. Front Genet. 2013;4:132
  • [38] H. Dombret. Gene mutation and AML pathogenesis. Blood. 2011;118:5366-5367 Crossref.
  • [39] F. Meklat, Z. Li, Z. Wang, Y. Zhang, J. Zhang, A. Jewell, et al. Cancer-testis antigens in haematological malignancies. Br J Haematol. 2007;136:769-776 Crossref.
  • [40] C.L. Riley, M.G. Mathieu, R.E. Clark, S.E. McArdle, R.C. Rees. Tumour antigen-targeted immunotherapy for chronic myeloid leukaemia: is it still viable. Cancer Immunol Immunother. 2009;58:1489-1499 Crossref.
  • [41] M.G. Mathieu, A.J. Knights, G. Pawelec, C.L. Riley, D. Wernet, F.A. Lemonnier, et al. HAGE, a cancer/testis antigen with potential for melanoma immunotherapy: identification of several MHC class I/II HAGE-derived immunogenic peptides. Cancer Immunol Immunother. 2007;56:1885-1895 Crossref.


a Laboratory Center, Affiliated People's Hospital of Jiangsu University, Zhenjiang, People's Republic of China

b Department of Haematology, Affiliated People's Hospital of Jiangsu University, Zhenjiang, People's Republic of China

c State Key Laboratory of Experimental Hematology, Institute of Hematology & Blood Diseases Hospital, CAMS & PUMC, Tianjin, People's Republic of China

lowast Corresponding author. Department of Haematology, Affiliated People's Hospital of Jiangsu University, 8 Dianli Rd., Zhenjiang, People's Republic of China. Tel.: +86 511 88915303; fax: +86 511 85234387.

1 These authors contributed equally as first authors.