This article is mentioned in:

Introduction

Cancer-testis antigens (CTAs) are a group of antigens that are exclusively expressed in germline cells, such as testis and placental cells under normal physiological conditions. CTAs are also expressed in numerous human tumor cells of various histological origins (1). The first CTA identified in 1991 is termed melanoma-associated antigen 1 (MAGE-A1) (2). In 1994, Weber et al(3) demonstrated that decitabine, a demethylation agent, upregulates MAGE-A1 expression in melanoma cell lines. In 1996, the activation of MAGE-A1 in tumor cells was correlated with genome-wide demethylation (4). Subsequent to this, more CTAs were identified, including PRAME(5), NY-ESO-1(6) and SSX family antigens (7). Numerous studies concerning the immunogenicity of CTAs and the asssociation between CTAs and demethylation have been conducted (8–11). For example, in acute myeloid leukemia (AML) cell lines, NY-ESO-1 was upregulated by decitabine and thus decitabine-treated AML cells become susceptible to NY-ESO-1-specific T-cell cytotoxicity (8). Those results have suggested CTA to be a potential target for tumor immunotherapy. However, the majority of studies concerning CTAs and decitabine were limited to in vitro investigations and few CTAs have been demonstrated to be activated by decitabine in vivo by clinical decitabine treatment.

The nuclear RNA export factor 2 (NXF2) human gene was first identified in spermatogonia (12). Early studies on NXF2 were focused on its function as an mRNA exporter (13) and its involvement in male infertility (14–16). NXF2 has been found to be positive in 1.8% of invasive ductal carcinomas of the breast (17). Dubovsky et al(18) observed its involvement in chronic lymphocytic leukemia (CLL). Following screening of 22 CLL patients, two were observed to be positive for NXF2-specific IgG antibodies. The presence of these antibodies supported the hypothesis that NXF2 exhibits high enough immunogenicity to induce tumor-specific immune responses (18). However, no studies have been conducted concerning the expression pattern, regulation mechanism and clinical significance of NFX2 in acute leukemia. The aim of the current study was to investigate the expression and epigenetic regulation mechanism of NXF2 in acute leukemia cells, in vitro and in vivo.

Materials and methods

Cell lines and patient samples

Bone marrow and peripheral blood samples were collected from healthy donors and leukemia patients in the Hematology Department of the Chinese People’s Liberation Army General Hospital, (Beijing, China). Samples of testis were provided by the laboratory of the Urology Department, Peking University Third Hospital (Beijing, China). Written informed consent was obtained from all donors and patients. All experiments were approved by the ethics committee of the Chinese People’s Liberation Army General Hospital. Raji, Z-138, Hut-78, Jurkat, Molt-4, Kasumi-1, NB4, THP-1, U937 and K562 acute leukemia cell lines were obtained from the cell culture center of Peking Union Medical College (Beijing, China). Primary cells were separated from bone marrow or peripheral blood samples using the Ficoll-Paque method (19). Cell lines and primary acute leukemia cells were maintained in RPMI-1640 supplemented with 10% fetal calf serum, 1% penicillin/streptomycin and 1% glutamine. Cultures were maintained in a 5% CO2-humidified incubator at 37°C. Decitabine (5-aza-2′-deoxycytidine, Dacogen, DAC) was purchased from Xi’an Jansson Pharmaceutical, Ltd. (Xi’an, China). For demethylation treatment, decitabine was added to the in vitro culture system at a concentration of 1 μmol/l for 3 days for leukemia cell lines and 5 μmol/l for 3 days for primary leukemia samples (18). For clinical treatment, decitabine was administered alone at a dose of 20 mg/m2/day for 5 consecutive days (20).

RNA isolation, reverse transcription and semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was extracted from samples or cell lines with TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA), according to the manufacturer’s instructions. Total RNA was reverse transcribed to complementary DNA (cDNA) as previously described (21). Reverse transcription and semi-quantitative RT-PCR were conducted on a Veriti® Thermal Cycler (Applied Biosystems, Inc., Foster City, CA, USA). Primers used for semi-quantitative RT-PCR were: Forward: 5′-TGA AAC CCT GCA AGG AAA AC-3′ and reverse: 5′-GCA CTG AGG GAG TCC ACA AT-3′ for NXF2; and forward: 3′-GAG TCA ACG GAT TTG GTC GT-5′ and reverse: 3′-TTG ATT TTG GAG GGA TCT CG-5′ for glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The 25 μl PCR mixture was prepared with 12.5 μl 2X GoTaq Green Master mix (Promega Corporation, Fitchburg, WI, USA), cDNA, primers and nuclease-free water. The amplification procedure was as follows: 35 cycles of denaturation for 30 sec at 95°C, annealing for 50 sec at 60°C and extension for 50 sec at 72°C. PCR amplification products were analyzed on 1.5% agarose gels stained with ethidium bromide.

qPCR

qPCR was performed to quantify target genes in bone marrow samples using the Stratagene Mx3000P real-time qPCR system (Stratagene, La Jolla, CA, USA). Primers and probes used were: Forward: 3′-GAAGCCAGGCCAAAT GGA-5′, reverse: 3′-AGTCTGGGTCAAAGCGGAGAT-5′ and probe: 3′-FAM-ATGAACAAACGGTACAATGTCTCC CA-TAMRA-5′ for NFX2; and forward: CATACCAGGAAA TGAGCTTGACAA, reverse: CATACCAGGAAATGAGCT TGACAA and probe: 3′-FAM-CTCCTCTGACTTCAA CAGCGACACCCA-TAMRA-5′. Each qPCR reaction was conducted in 20 μl reaction volume with TaqMan universal master mix (Applied Biosystems, Inc.), 0.25 μM primers and probe and 20 ng cDNA. The amplification procedure was as follows: 40 cycles of denaturation for 15 sec at 95°C and annealing for 60 sec at 60°C. NXF2 mRNA expression was determined by 2−ΔΔCT relative to GAPDH.

Western blot analysis

Protein extracts from U937 and Raji cells were isolated using immunoprecipitation assay buffer (Sigma-Aldrich, St. Louis, MO, USA) and quantified to ensure equivalent protein loading. Polyacrylamide gel electrophoresis, tank-based transfer to Immobilon Hybond-C membranes (Amersham Pharmacia Biotech, Piscataway, NJ, USA) and immunodetection were performed as described (22). β-actin antibody was purchased from Abcam (Cambridge, MA, USA) and NXF2 antibody was purchased from Sigma-Aldrich. Signals were visualized using Immobilon Western Chemiluminescent horseradish peroxidase substrate (Millipore Corporation, Billerica, MA, USA) by exposure to films (Kodak, Rochester, NY, USA).

Firefly luciferase reporter constructs

Five promoter regions of the wild-type human NXF2 gene were amplified from one testis sample by PCR using specific primers (Fig. 3A). Primers were designed to contain KpnI and HindIII restriction enzyme sites. Sense primers used were: 5′-CGGGGTACCAGT GTGGGCTGAGGGTTG GA-3′, −1,000 to +28 bp for P1; 5′-CGGGGTACCGTAAGCATCCCCTGCTACACG-3′, −744 to +28 bp for P2; 5′-CGGGGTACCCAGGTGCCTGTAATG CCAGCT-3′, −512 to +28 bp for P3; 5′-CGGGGTACCCAG AGCGAGACGCCGTCT-3′, −400 to +28 bp for P4; 5′-CGG GGTACCGGCAGGCTTATAATCAGAACACCC-3′, −229 to +28 bp for P5; and the antisense primer was 5′-CCCAAG CTTCAAAGCAGTGGGGAGAGGAC-3′. The amplified promoter regions were cloned downstream of the firefly luciferase-coding region between KpnI and HindIII of a modified pGL3-control plasmid. Successful construction was confirmed by sequencing.

Figure 3 Activation of NXF2 was associated with CpG island demethylation in its promoter region. (A) Firefly luciferase reporter constructs. A typical CpG island located in the promoter region of the NXF2 gene. (B) Firefly luciferase assay indicated that the promoter region containing the CpG island is essential for NXF2 expression. (C) It was demonstrated that there are 14 CpG sites in the CpG island. (D) The CpG island was densely methylated in two normal bone marrow samples without NXF2 expression; however, less methylated in testis sample and in K562 cells with NXF2 expression. (E) The CpG island was densely methylated in Raji and U937 cells without NXF2 expression. Following decitabine treatment, the methylation level decreased in accordance with NXF2 activation. Black dots indicate methylated CpG sites and white dots indicate unmethylated CpG sites. NFX2, nuclear RNA export factor; CTR, control cells; DAC, decitabine-treated cells.

Transfection and luciferase assays

293T cells were plated in 24-well plates at 5×104 cells/well and grown overnight. Firefly luciferase reporter vector (500 ng) containing different promoter regions of NXF2 and 10 ng control vector pRL-TK (Promega Corporation) containing Renilla luciferase were cotransfected to 293 T cells in a final volume of 0.35 ml using SuperFect (Qiagen, Hilden, German). Cells were collected 48 h following transfection and luciferase activity was measured using a dual-reporter luciferase assay system (Promega Corporation).

Bisulfite modification and genomic sequencing

Genomic DNA was extracted from cells using the Wizard Genomic DNA Purification kit (Promega Corporation). DNA (1 μg) was modified with sodium bisulfite using the EpiTect Bisulfite kit (Qiagen). Primers for bisulfite-sequencing analysis were designed by MethPrimer software (San Francisco, CA, USA) (23) using the bisulfite-treated DNA as a template (Forward: 5′-GAGTTTTTAATTGTTTGTGTTGAG-3′ and reverse: 5′-GGTAGAGGTTGTATGAGATGGG-3′). PCR products were gel-purified and cloned into a pGEM-T vector (Promega Corporation). The inserted PCR fragments of individual clones were sequenced using an ABI PRISM DNA sequencer (Applied Biosystems, Inc.).

Results

Activation of NXF2 expression following decitabine treatment in acute leukemia cell lines

NXF2 expression was screened in two normal bone marrow (NBM) samples, two normal peripheral blood samples and two normal testis samples. The results showed that NXF2 was only positively expressed in the testis samples (Fig. 1A). NXF2 expression was screened in 10 acute leukemia cell lines, including five AML cell lines (Kasumi-1, NB4, THP-1, U937 and K562 cell lines; K562 cells are cells of the blastic transformation of chronic myeloid leukemia) and five acute lymphocytic leukemia (ALL) cell lines (Raji, Z-138, HuT-78, Jurkat and Molt-4), prior to and following decitabine treatment. The results showed that only K562 cells expressed NXF2 originally. However, following decitabine treatment, eight of the remaining nine acute leukemia cell lines demonstrated activation of NXF2 expression, with the exception of NB4 cells (Fig. 1B). This upregulation was confirmed in one AML cell line (U937) and one ALL cell line (Raji) by western blot analysis (Fig. 1C). Raji cells were then analyzed for the effect of different doses of decitabine on the activation of NXF2. The results showed that NXF2 activation was dose-dependent. Following treatment with 1 μM decitabine for 3 days, it was noted that NXF2 remained positive 28 days subsequent to the cessation of decitabine treatment (Fig. 1D).

Figure 1 Decitabine activated NXF2 expression in acute leukemia cell lines in vitro. (A) RT-PCR showed that NXF2 was originally expressed only in testis and not in bone marrow or peripheral blood samples from healthy donors. (B) Acute leukemia cell lines, including 5 acute myeloid leukemia cell lines (Kasumi-1, NB4, THP-1, U937 and K562) and 5 acute lymphocytic leukemia cell lines (Raji, Z-138, HuT-78, Jurkat and Molt-4) were screened for NXF2 expression prior to and following 1 μM decitabine treatment for 3 days in vitro. NXF2 expression was positive in K562 cells and was activated in the majority of other acute leukemia cell lines except NB4 following decitabine treatment. (C) In Raji and U937 cells, upregulation of NXF2 expression following decitabine treatment was confirmed by western blot analysis. β-actin was used as the positive control. (D) In Raji cells, NXF2 activation was decitabine dose-dependent and its expression lasted ≥28 days following the cessation of decitabine treatment. DAC, decitabine-treated cells; NFX2, nuclear RNA export factor 2.

Activation of NXF2 expression following decitabine treatment in primary acute leukemia cells in vitro and in vivo

Primary cells from 11 bone marrow samples, including four AML patients, four ALL patients and three healthy donors were collected. The primary cells were treated with 5 μM decitabine for 3 days in vitro. Relative NXF2 expression (Fig. 2A) was low in the majority of the samples. Following decitabine treatment, NXF2 mRNA expression in all acute leukemia samples was significantly upregulated. However, this upregulation was not observed in healthy donor samples. The results suggested that decitabine treatment activated NXF2 expression in primary acute leukemia cells in vitro.

Figure 2 Decitabine activated NXF2 expression in primary acute leukemia cells in vitro and in vivo. (A) Bone marrow samples from four AML patients, four acute lymphocytic leukemia patients and three healthy donors were collected and screened for NXF2 expression before and after decitabine treatment in vitro. NXF2 was activated in the majority of leukemia samples; however, no or only marginal upregulation was observed in healthy donor samples by qPCR. (B) NXF2 expression levels were analyzed in samples from nine AML or myelodysplastic syndrome patients before and after their first decitabine treatment. Patient characteristics are listed in Table I. Decitabine treatment was shown to activate NXF2 expression in acute leukemia cells in vivo. Values are presented as the mean ± SD of triplicates. NFX2, nuclear RNA export factor 2; AML, acute myeloid leukemia; ALL, acute lymphocytic leukemia; CTR, control cells; DAC, decitabine-treated cells.

In addition, another nine patients with AML or myelodysplastic syndrome, who received decitabine treatment clinically were selected. The samples were collected prior to and following the first cycle of decitabine treatment. The characteristics of these patients are listed in Table I. Samples were obtained from the bone marrow, with the exception of patient 6, which was a peripheral blood sample as the patient had hyperleukocytosis.

Table I Patient characteristics.

Table I

Patient characteristics.

Patient	Age (years)	Gender	Chromosome	Gene abnormalities	Diagnosis	Response	DAC cycles	Prognosis
1	61	M	46,XY	MLL-PTD(+)	MDS-RAEB	Failure	3	Progressed to AML after 3 cycles
2	75	M	46,XY,12p-inc[10]/, hypodiploid(43–45)	NUP98/HOXA9(+)	AML (MDS transformed)	SD	8	Still in DAC maintenance
3	38	M	46,XY	-	AML (MDS transformed)	SD	5	Quit DAC treatment after the 5th cycle
4	62	F	46,XX	-	AML (MDS transformed)	PD	3	Quit DAC treatment after the 3rd cycle
5	53	F	46,XX	GIT2/PDGFRB(+)	AML (CMML transformed)	PD	3	This patient received TKI treatment first. DAC was added in combination with TKI when her CMML progressed to AML. She quit DAC treatment after the 3rd cycle and received allo-HSCT
6	49	M	-	-	AML-M6	SD	7	Quit DAC treatment and received allo-HSCT after the 7th cycle
7	59	F	46,XY	-	AML (MDS transformed)	CR	14	CR after the 7th cycle, but relapsed after the 14th cycle, quit DAC treatment after relapse
8	64	M	46,XY	P15 hyper-methylated	AML (MDS transformed)	CR	12	CR after the 5th cycle, but relapsed after the 10th cycle, quit DAC treatment after relapse.
9	82	F	46,XX	-	AML (MDS	PR	10	PR after the 4th cycle, but transformed) turned to AML after the 7th cycle, another 3 cycles of DAC after progression did not work

[i] All samples were bone marrow samples except for patient 6. The samples collected from patient 6 were peripheral blood samples because the patient had hyperleukocytosis. M, male; F, female DAC, decitabine; MDS, myelodysplastic syndrome; RAEB, refractory anemia with excess blasts; AML, acute myeloid leukemia; CR, complete remission; PR, partial remission; SD, stable disease; PD, progressive disease; CMML, chronic myelomonocytic leukemia; TKI, tyrosine-kinase inhibitor; allo-HSCT, allogeneic hematopoietic stem cell transplantation; AML-M6, acute erythroid leukemia.

As the nine patients did not respond to the first cycle of decitabine, leukemia cells were observed in all of the samples. qPCR results showed that NXF2 expression was upregulated to varying degrees in all nine patients following decitabine treatment (Fig. 2B). This result suggests that decitabine treatment activated NXF2 expression in primary acute leukemia cells in vivo.

NXF2 expression correlates with CpG island hypomethylation in its promoter region

There was a typical CpG island located in the NXF2 promoter region spanning 282 bp and containing 14 CpG sites (Fig. 3A and C). By luciferase reporter construction, it was demonstrated that the promoter region containing the CpG island was essential for NXF2 expression (Fig. 3B). The methylation status of this CpG island in two NBM samples without NXF2 expression in one testis sample and in K562 cells with NXF2 expression was detected. The CpG island was densely-methylated in the two NBM samples (>90%), but only partially methylated in the testis sample and in the K562 cells (Fig. 3D). Raji and U937 cell lines were originally NXF2-negative and were analyzed for the methylation status of the CpG island prior to and following decitabine treatment. The methylation level of the CpG island decreased following decitabine treatment, concurrent with NXF2 activation (Fig. 3E). These results suggest that NXF2 expression was silenced by CpG island hypermethylation; thus, decitabine activated NXF2 expression by the demethylation mechanism.

Discussion

NXF2 is a CTA gene, however, there is limited information available concerning its expression pattern and the mechanism of NXF2 regulation. The present study showed that NXF2 expression was activated by a demethylating agent, decitabine, in acute leukemia cell lines and primary acute leukemia samples in vitro and in vivo. This study also showed that NXF2 expression is regulated by CpG island hypermethylation in its promoter region. To the best of our knowledge, this is the first study to demonstrate the mechanism of NXF2 regulation in acute leukemia cells.

Currently, certain CTAs have been demonstrated to be potential targets for immunotherapy against cancer due to their tumor specificity and strong immunogenicity. For example, Hunder et al(9) expanded autologous CD4+ T-cell clones specific for NY-ESO-1 (a CTA). Specific T cells were infused into a patient with refractory metastatic melanoma and a durable clinical remission was observed. In another study, Quintarelli et al(24) showed that cytotoxic T lymphocytes specific for PRAME, another type of CTA, target leukemic and leukemic-precursor cells. It was also demonstrated that NXF2 exhibits a high enough level of immunogenicity to induce immune responses in CLL patients (18). However, the application of CTA-specific immunotherapy is limited due to a relatively low expression in acute leukemia. It was suggested that this limitation was able to be overcome by demethylation treatment. Yan et al(25) demonstrated that, following demethylation treatment, the expression of PRAME in the ALL cell line (Raji cells) was upregulated. These Raji cells with greater PRAME expression showed increased sensitivity to killing by formerly low avidity PRAME-specific T-cell clones. However, the majority of similar studies are performed in vitro and few studies have observed that CTAs were activated in malignant cells in vivo by clinical decitabine treatment. In the present study it was demonstrated that NXF2 was activated in acute leukemia cells not only in vitro, but also in vivo by clinical decitabine treatment. It was hypothesized that NXF2 served as a novel target for immunotherapy against acute leukemia following clinical demethylation treatment. However, as no epitope sequence of NXF2 has yet been identified, it was difficult to determine the details of NXF2-specific immune responses following decitabine treatment. Thus, future studies are required to investigate the epitope of NXF2.

In conclusion, to the best of our knowledge, this was the first study to demonstrate that NXF2 is activated by decitabine in acute leukemia cells in vitro and in vivo; this activation was due to demethylation of the CpG island in the NXF2 promoter region. According to these results, it was hypothesized that NXF2 may serve as a novel clinical target for immunotherapy against acute leukemia.

Acknowledgements

This study was supported by grants from the National Basic Research Program of China (grant no. 2005CB522400); the National Natural Science Foundation of China (grant nos. 90919044, 30971297 and 81000221); the Capital Medical Development Scientific Research Fund (grant no. 2007–2040); the National Public Health Grand Research Foundation (grant no. 201202017); and the capital of the Public Health Project (grant no. Z111107067311070).

References