This article is mentioned in:

Introduction

Nasopharyngeal carcinoma (NPC) is a human epithelial malignancy with high incidence in southern China, particularly, in the Cantonese population. The Epstein-Barr virus (EBV) was the first human tumor virus to be identified and has yielded significant insights into the pathogenesis of cancer. Latent EBV infection is one of the most important factors for NPC in endemic areas. NPC is associated with EBV latent infection type II and the expression of EBV latent membrane protein (LMP) 1, EBV-LMP2A, EBV-LMP2B, EBV nuclear antigen (EBNA) 1, EBV-encoded RNA (EBER) 1, EBER2 and the BamHI A rightward transcripts (BARTs) (1–3). Metastasis to the neck lymph nodes is the predominant initial symptom in NPC. In addition, recurrence and distant metastasis after radiotherapy are known to occur in NPC. Therefore, it is imperative to identify novel diagnostic and therapeutic approaches.

Loss of heterozygosity (LOH) on chromosomes 13q and 14q are common genetic events, and putative tumor-suppressor genes located in these regions may be involved in the development of NPC (4,5). Evidence from LOH studies showed that NPC-related tumor-suppressor genes may be present at 14q24-q32 (5). LOH on chromosome 14q was detected in 80% of NPC tumors with high-frequency LOH loci clustered to 14q11-q13, 14q21-q24 and 14q32 (5). The GALC gene is located at 14q31, has 17 exons and 16 introns of an approximate size of 60 kb (6). Aberrant methylation in the promoters of tumor-related genes is closely associated with epigenetically mediated gene silencing (7–14). DNA hypermethylation in the promoter region has been reported in NPC (7–10). Hypermethylation of the 9p21 chromosomal region genes (p14, p15 and p16) has been observed in the peripheral blood of leukemic patients (11). DNA methylation in cancer-related genes can be a potential biomarker and therapeutic target for prostate cancer (12). It has been reported that p16 promoter methylation is a promising potential biomarker for the early diagnosis of gastric cancer (13). A potential biomarker panel assessing the hypermethylation of six gene promoters holds great promise as a diagnostic test for high-grade (type II) serous ovarian cancer (14).

Therefore, we aimed to investigate the possible tumor-suppressor effect of the GALC gene in EBV-associated NPC, and identified that the GALC gene is downregulated by promoter hypermethylation in the NPC cell line, CNE-2Z.

Materials and methods

Specimens

The World Health Organization has classified NPC into different histological types [keratinizing squamous cell carcinoma (type I), differentiated non-keratinizing carcinoma (type II) and undifferentiated non-keratinizing carcinoma (type III)] based on the tumor cell appearance on light microscopy (15). Type II and III NPC are predominantly EBV-positive.

Specimens of type III NPC and chronic nasopharyngitis were obtained from patients diagnosed with these conditions between January 2008 and August 2012 at the Affiliated Hospital of Guangdong Medical College and Gaozhou People's Hospital (Guangdong, China). Informed consent was obtained from all the participants. None of the NPC patients were previously treated with radiotherapy or chemotherapy. Of the 41 NPC patients whose specimens were used for immunohistochemical analysis, 7 were at TNM stage II, 21 were at stage III and 13 were at stage IV. There were 35 men and 6 women in the NPC group, with ages ranging from 31 to 86 years (median age, 54 years; mean age, 52 years). Of the 15 chronic nasopharyngitis patients, 9 were male and 6 were female, with ages ranging from 15 to 79 years (median age, 47 years; mean age, 45.9 years).

EBER in situ hybridization

The sections of 41 NPC tumor specimens were dewaxed in xylene and digested with proteinase K for 16 min. The sections were then rehydrated in serial graded ethanol washes (70, 95 and 100%) and hybridized for 2 h at 37°C with digoxigenin-labeled EBER probes (ZSGB-BIO, China). The hybridization products were detected using anti-digoxigenin-horseradish peroxidase (ZSGB-BIO, Beijing, China) for 30 min at 37°C, followed by 3 min of incubation in diaminobenzidine solution for the detection of EBERs. The slides were counterstained with Mayer's hematoxylin and positive staining was observed as brown granules at the site of hybridization on light microscopy. Only specimens with signals within the tumor cell nuclei were considered to be positive.

Immunohistochemical staining

Tissue samples were fixed in 10% buffered formalin, dehydrated, embedded in paraffin and cut into 4-μm sections. The samples were evaluated by two experienced pathologists. Sections with known high and no GALC expression were used as positive and negative controls, respectively. The sections incubated with 0.01 mol/l phosphate-buffered saline (PBS) instead of primary antibody served as the blank control. A streptavidin-peroxidase (SP) kit was purchased from Beijing Zhongshan Golden Bridge Biotechnology, Co. (Beijing, China). Rabbit anti-human GALC polyclonal antibody was purchased from the Proteintech Group, Inc. (Chicago, IL, USA) (working dilution, 1:200). Procedures were implemented according to the manufacturer's instructions.

Evaluation of in situ hybridization and immunohistochemical staining

For each section, five low magnification fields (×100) were randomly selected under a microscope, and the proportions of positively stained cells per field were determined under higher magnification (×400). The results were expressed as the mean percentage of positively stained cells in the five fields. GALC was specifically located in the cytoplasm. The staining intensity (A) was classified as follows: negative, 0 points; weak, 1 point; moderate, 2 points; and strong, 3 points. The percentage of positive cells (B) was classified as: 0%, 0 points; 1–25%, 1 point; 26–50%, 2 points; 51–75%, 3 points, and >75%, 4 points. The product of A and B yielded the final score, which was classified as follows: 0–1, negative (−) and ≥2, positive (+). The sections were evaluated by three pathologists who were blinded to the study design.

Cell culture and drug treatment

The NPC CNE-2Z cell line which has low differentiation, was obtained from the Department of Pathology, Guangdong Medical College. The CNE-2Z cells were cultured in RPMI-1640 with 10% fetal bovine serum (FBS; HyClone, Laboratories, Logan, UT, USA), 0.1 μM penicillin and 0.1 μM streptomycin under 5% CO2 at 37°C. For drug treatment, exponentially growing CNE-2Z cells were seeded at a density of 2×105 cells/well into 6-well plates. The cells were treated with different doses (0, 5, 10, 20, 40 and 60 μM/l) of the demethylation agent 5-Aza-2′-deoxycytidine (5-Aza-dC; Sigma, St. Louis, MO, USA) for 48 h. The cells were collected for the detection of GALC protein expression using western blotting and immunofluorescent assay, GALC mRNA expression using reverse transcriptase-polymerase chain reaction (RT-PCR) assay, GALC gene promoter methylation using bisulfite sequencing PCR (BSP), cell proliferation using Cell Counting kit (CCK)-8 assay, and cell migration using scratch and Transwell migration assays.

Quantitative RT-PCR

Total RNA was extracted from the cultured CNE-2Z cells using RNAiso Plus (Takara, Japan). Reverse transcription was performed with the PrimeScript™ RT reagent kit (Takara). The primers used for the PCR assay were: GALC, 5′-AGGGAACCTCACCATCATCA-3′ (forward) and 5′-GCTGTCAAGGAGCCATAGAGAA-3′ (reverse); and GAPDH, 5′-AGGTCGGAGTCAACGGATTTG-3′ (forward) and 5′-GTGATGGCATGGACTGTGGT-3′ (reverse). PCR was performed in a LightCycler 480 fluorescent quantization PCR thermal cycler (Roche, Switzerland). Quantitative RT-PCR was performed with SYBR® Premix Ex Taq™ II (Takara) at 95°C for 30 sec, followed by 40 cycles of 95°C for 5 sec and 60°C for 30 sec, after which a melt-curve analysis was performed once from 60 to 95°C, followed by 1 cycle at 50°C for 30 sec. The level of GAPDH mRNA was used as an internal control. The reactions were performed in triplicate. The relative amounts of GALC mRNA were calculated in accordance with the 2−ΔΔCt method.

BSP

Bisulfite modification of genomic DNA converts unmethylated deoxycytidine to uracil (read as thymidine), however, methylated deoxycytidine remained unmodified. A 372-bp region within the promoter (−158 to +214 nucleotides relative to the transcriptional start site) containing the CpG island was PCR-amplified from the bisulfite-treated DNA. The PCR fragments were cloned and sequenced in 10 clones, thereby allowing identification of the methylation status of each of the 37 specific CpG sites in the GALC gene. The methylation primers of the GALC gene promoter region used for the BSP assay were: 5′-TTG(T/C)GTTAAAGTGTTTTATTA GGTGA-3′ (forward) and 5′-CAAC(A/G)CACACAACAAC AAAAACA-3′ (reverse).

Western blot analysis

Total cellular proteins of the harvested CNE-2Z cells were extracted. The standard western blotting protocol was applied. The specific primary antibody anti-GALC (working dilution, 1:200; Proteintech) and a horseradish peroxidase-conjugated secondary antibody (Mai Bio Co., Ltd, Zhejiang, China) were used. Specific bands were visualized using enhanced chemiluminescence reagents and recorded on film.

Immunofluorescent staining

CNE-2Z cells were fixed in 4% paraformaldehyde for 15 min at room temperature, blocked and incubated with the primary antibody anti-GALC (working dilution, 1:200), the secondary antibody Rhodamine-conjugated goat anti-rabbit IgG (both from Proteintech) and 4′,6-diamidino-2-phenylindole (DAPI; Sigma). In each experiment, the immunopositive cells in five randomly selected fields were counted and 500–1,000 cells were counted in total.

Cell viability

The proliferation of CNE-2Z cells was determined using the CCK-8 assay, according to the manufacturer's instructions (Beyotime Institute of Biotechnology, Shanghai, China). Exponentially growing CNE-2Z cells (4×103 cells/well) in 100 μl medium were seeded in 96-well plates. After 12 h, the medium in each well was replaced with media containing different concentrations of 5-Aza-dC (0, 5, 10, 20, 40 and 60 μM/l), and the plates were incubated for 48 h. The supernatant was removed, and 200 μl of CCK-8 and RPMI-1640 mixed medium was added into each well, followed by incubation for 2 h at 37°C in the dark. The absorbance of CCK-8-containing medium was measured at a wavelength of 450 nm (reference wavelength, 630 nm) using a microplate reader (Bio-Rad, Hercules, CA, USA). Cell proliferation was calculated as: Viability = (ODtest group - ODblank group)/(ODcontrol group - ODblank group). The results were presented as mean ± standard deviation (SD). Three independent experiments, each in triplicate, were conducted.

Cell migration

In vitro scratch and Transwell chamber assays were performed to assess the migration of CNE-2Z cells. Exponentially growing CNE-2Z cells (8×105 cells/well) were seeded in 6-well plates. When the cells were cultured to 90% confluency, each well was manually scratched three times with a 200-μl pipette tip, washed with PBS three times and incubated at 37°C with media containing different concentrations of 5-Aza-dC (0, 20 and 40 μM/l). Cell images were recorded at 0 and 24 h. The distance between two cell edges was analyzed using Image-Pro software.

The Transwell system (24 wells, 8-μm pore size with polycarbonate membrane; Corning Costar, NY, USA) not coated with Matrigel was used for the migration assay. The CNE-2Z cells were incubated with media containing different concentrations of 5-Aza-dC (0, 20 and 40 μM/l) at 37°C for 48 h. A total of 4×104 cells were suspended in 100 μl serum-free medium and were seeded in the upper chambers. Then, 600 μl RPMI-1640 containing 10% FBS was added to the lower chamber. After 24 h, the cells remaining in the upper chambers were removed with a cotton swab, while the cells attached to the lower surface were fixed with methanol and stained with hematoxylin and eosin (H&E). The cells migrating to the lower side in five randomly selected fields were counted by light microscopy, and the total number of migrating cells was determined and analyzed statistically. Each experiment was performed in triplicate.

Statistical analysis

Statistical analyses were performed using SPSS 17.0. One-way analysis of variance (ANOVA) was performed to analyze the data from the CCK-8, scratch and Transwell chamber assays and RT-PCR. The χ2 and the Fisher's exact tests, and the logistic regression analysis were used to compare the rates between the different groups and to test the correlation between two factors. P<0.05 was considered to indicate a statistically significant result.

Results

EBER and GALC protein expression in chronic nasopharyngitis and NPC

A high expression of EBER was detected in the specimens from the 41 NPC patients. EBER expression was not correlated with different clinicopathological parameters, such as age, gender, lymph node metastasis and TNM stage (P>0.05).

EBER and GALC protein expression in the columnar epithelium of the nasopharyngeal mucosa and in NPC

No EBER expression was found in the columnar epithelial of the nasopharyngeal mucosa obtained from the 15 individuals in the control group. By contrast, 92.68% of the NPC specimens from the 41 NPC patients exhibited EBER expression. The immunohistochemical SP method was used to measure the expression of GALC protein. GALC protein expression was detected in 24.39% of NPC specimens, which had a high rate of EBV latent infection. GALC protein expression was detected in 60.00% of columnar epithelia of chronic nasopharyngitis specimens with no EBV latent infection. The difference between these two groups was statistically significant (P<0.05) (Table I).

Table I EBER and GALC protein expression in the columnar epithelium of the nasopharyngeal mucosa and in NPC.

Table I

EBER and GALC protein expression in the columnar epithelium of the nasopharyngeal mucosa and in NPC.

Groups	EBERa			GALCb
	n	−	+(%)	−	+(%)
Nasopharyngeal	15	15	0 (0)	6	9 (60.00)
mucosal
columnar
epithelium
NPC cells	41	3	38 (92.68)	31	10 (24.39)

a Fisher's exact test, P<0.01.

b χ2=6.212, P<0.05. EBER, EBV-encoded RNA; NPC, nasopharyngeal carcinoma.

Relationship between GALC protein expression and clinicopathological parameters of NPC

The GALC protein expression was not correlated with age, gender and TNM stage (P>0.05) (Table II). GALC protein expression occurred significantly more frequently in tumors without lymph node metastasis than in tumors with metastasis (P<0.05) (Table II). Logistic regression showed that GALC protein expression protected against lymph node metastasis (P<0.05) (Table III).

Table II Relationship between GALC protein expression and clinicopathological parameters in NPC.

Table II

Relationship between GALC protein expression and clinicopathological parameters in NPC.

Clinicopathological parameters	GALC
	n	−	+(%)	P-value
Age (years)
<40	5	5	0 (0)	0.31
≥40	36	26	10 (27.77)
Gender
Male	35	26	9 (25.71)	1
Female	6	5	1 (16.66)
Lymph node metastasis
No	12	6	6 (50.00)	0.040a
Yes	29	25	4 (13.79)
TNM stage
II	7	3	4 (57.14)	0.083
III	21	17	4 (19.04)
IV	13	11	2 (15.38)

a Statistically significant. NPC, nasopharyngeal carcinoma.

Table III Logistic regression analysis of GALC protein expression and lymph node metastasis in NPC (n=41).

Table III

Logistic regression analysis of GALC protein expression and lymph node metastasis in NPC (n=41).

Selected factor	β	SE	Wald	P-value	Exp (β)	95% CI
GALC	−1.8326	0.7895	5.3877	0.020	0.160	0.034–0.752

[i] CI, confidence interval; NPC, nasopharyngeal carcinoma.

Effect of different 5-Aza-dC concentrations on GALC gene methylation status of CNE-2Z cells

To confirm that hypermethylation of the 37 CpG sites in the promoter region (−158 to +214) of the GALC gene decreased after 5-Aza-dC treatment, CNE-2Z cells were treated with different concentrations of 5-Aza-dC (0, 5, 10, 20, 40 and 60 μM/l) for 48 h, and the methylation status of the GALC gene promoter was analyzed using bisulfite genomic sequencing in the control (0 μM/l) and treated samples (Fig. 3). In the control sample, CpG sites were 95.9% methylated, and in the treated samples the sites were 86.5, 81.9, 50.8, 71.6 and 70.3% methylated at the concentrations of 5, 10, 20, 40 and 60 μM/l, respectively.

Figure 3 Methylation status of the GALC gene promoter region in CNE-2Z cells treated with different concentrations of 5-Aza-dC (0, 5, 10, 20, 40 and 60 μM/l) for 48 h. The methylation status of the 37 CpG sites of the GALC gene promoter region is shown. (○, unmethylated CpG site; •, methylated CpG site). 5-Aza-dC, 5-Aza-2′-deoxycytidine.

Effect of different concentrations of 5-Aza-dC on GALC mRNA expression in CNE-2Z cells

To provide evidence for a possible role of DNA methylation in the regulation of GALC expression, we analyzed GALC mRNA levels in CNE-2Z cells using RT-PCR in the control and treated groups (Fig. 4). The results showed that GALC mRNA levels were significantly higher in the treated groups than those in the control group (P<0.05). GALC mRNA levels in the 5, 10, 40 and 60 μM/l groups were significantly lower than that in the 20 μM/l group (P<0.01), while the levels in the 5 and 10 μM/l groups were significantly lower than that in the 40 μM/l group (P<0.05). GALC mRNA levels in the 5, 10 and 40 μM/l groups did not significantly differ from that in the 60 μM/l group (P>0.05).

Figure 4 GALC mRNA expression detected using RT-PCR analysis. The control sample and the CNE-2Z cells were treated with different concentrations of 5-Aza-dC (5, 10, 20, 40 and 60 μM/l) for 48 h, and then subjected to RT-PCR to analyze the GALC mRNA expression levels. The GALC mRNA expression levels in the treated samples (20, 40 and 60 μM/l) significantly differed from those in the control samples. Statistical analysis was performed using one-way ANOVA, followed by the significant difference (Dunnett's T3) test. Data are presented as means ± SD, n=3. *P<0.05, treated groups vs. control. #P<0.01, 5, 10, 40 and 60 μM/l vs. 20 μM/l. &P<0.05, 5 and 10 μM/l vs. 40 μM/l. 5-Aza-dC, 5-Aza-2′-deoxycytidine.

GALC protein expression in CNE-2Z cells treated with different concentrations of 5-Aza-dC and detected using western blotting and immunofluorescence

To provide evidence for a possible role of DNA methylation in the regulation of GALC expression, we analyzed GALC protein expression using western blotting and immunofluorescence analyses before and after treatment with different concentrations of 5-Aza-dC (5, 10, 20, 40 and 60 μM/l) for 48 h (Fig. 5). The results showed that GALC protein expression was absent in the control group but present in the cells treated with 5-Aza-dC. GALC protein expression was highest in the 20 μM/l group.

Figure 5 GALC protein expression in CNE-2Z cells was analyzed by western blotting and immunofluorescence analysis, before and after treatment with different concentrations of 5-Aza-dC (5, 10, 20, 40 and 60 μM/l) for 48 h. (A) GALC protein expression was absent in the control group but present in the cells treated with different concentrations of 5-Aza-dC. β-actin protein was used as a positive control. (B) Immunofluorescent staining for GALC (red) and DAPI (green, nuclei). The inset is a high-magnification view. Scale bars, 50 μm. 5-Aza-dC, 5-Aza-2′-deoxycytidine.

These results indicated a correlation between the methylation status of the GALC gene and the expression level of GALC mRNA or protein. Data suggest that promoter methylation suppresses GALC gene expression, and 5-Aza-dC-induced demethylation of CpG sites upregulates GALC gene expression in the NPC CNE-2Z cell line.

Effect of different concentrations of 5-Aza-dC on CNE-2Z cell proliferation

CCK-8 was used to examine the proliferation of CNE-2Z cells treated with different concentrations of 5-Aza-dC for 48 h (Fig. 6). The results showed that cell viability was significantly decrased in the 20, 40 and 60 μM/l groups than in the control group (P<0.05). The cell viability in the 5 and 10 μM/l groups significantly differed from that in the 20 μM/l group (P<0.05), while in the 5, 10 and 40 μM/l groups cell viability significantly differed from that in the 60 μM/l group (P<0.05). The viability in the 40 and 60 μM/l groups did not significantly differ from that the in 20 μM/l group (P>0.05).

Figure 6 Viability of CNE-2Z cells treated with different concentrations of 5-Aza-dC for 48 h was determined using CCK-8 assay. Statistical analysis was performed by one-way ANOVA, followed by the least significant difference (LSD) test. Data presented as means ± SD, n=3. *P<0.05, 20, 40 and 60 μM/l vs. control. #P<0.05, 5 and 10 μM/l vs. 20 μM/l. &P<0.05, 5, 10 and 40 μM/l vs. 60 μM/l. 5-Aza-dC, 5-Aza-2′-deoxycytidine.

Cell migration and scratch assay

After the CNE-2Z cells were treated with 20 or 40 μM/l 5-Aza-dC for 24 h, the relative cell migration distance was significantly higher in the treatment group than in the control group. Moreover, the cell migration ability decreased with an increasing concentration of 5-Aza-dC, and the differences between the treatment and control groups were statistically significant (P<0.01). However, no significant difference was found between the 20 and 40 μM/l group (P>0.05; Fig. 7).

Figure 7 (A) Scratch assay was used to detect the migration of CNE-2Z cells treated with different concentrations of 5-Aza-dC for 24 h. Representative images of the cellular migration distance are shown with an original magnification of ×40. (B) Statistical analysis was performed by one-way ANOVA, followed by the significant difference (LSD) test. Data presented as means ± SD, n=3. *P<0.01, 20 and 40 μM/l vs. control. 5-Aza-dC, 5-Aza-2′-deoxycytidine.

Cell migration and Transwell chamber assay

After the CNE-2Z cells were treated with 20 or 40 μM/l 5-Aza-dC for 24 h, the number of cells that migrated to the lower side was significantly lower in the treatment groups than in the control group (P<0.01). In addition, the number of the cells that migrated to the lower side was lower in the 20 μM/l group than that in the 40 μM/l group, although the difference was not statistically significant (P>0.05; Fig. 8).

Figure 8 (A) Transwell chamber assay was used to detect the migration of CNE-2Z cells treated with different concentrations of 5-Aza-dC for 24 h. Representative images of H&E staining cells are shown with an original magnification of ×200. (B) Statistical analysis was performed by one-way ANOVA, followed by the significant difference (LSD) test. Data presented as means ± SD, n=3.*P<0.01, 20 and 40 μM/l vs. control. 5-Aza-dC, 5-Aza-2′-deoxycytidine.

Discussion

Latent EBV infection is one of the most important factors for NPC in endemic areas. The transcription of EBERs is present in most NPC tumors (1–3). In the present study, no EBER expression was found in the columnar epithelial of the nasopharyngeal mucosa obtained from the 15 individuals in the control group. By contrast, 92.68% of the NPC specimens from the 41 NPC patients exhibited EBER expression. The immunohistochemical SP method was used to measure the expression of GALC protein. GALC protein expression was detected in 24.39% of NPC specimens, which had a high rate of EBV latent infection. GALC protein expression was detected in 60.00% of columnar epithelial of chronic nasopharyngitis specimens which had no EBV latent infection. The difference between these two groups was statistically significant (P<0.05). Subsequently, the 41 NPC specimens were divided into a lymph node metastasis and non-lymph node metastasis groups according to the TNM stage. Statistical analyses showed that the rates of the GALC protein expression were 13.79 and 50.00%, respectively, in these two groups and the difference was statistically significant (P<0.05). Logistic regression showed that the expression of GALC protein was a protective factor against lymph node metastasis of NPC (P<0.05). Compared with the normal columnar epithelial of the nasopharyngeal mucosa, the NPC cells showed a decreased expression of GALC protein. The results of the present study were in accordance with the results reported by Görögh for laryngeal squamous cell carcinoma as compared with the corresponding normal mucosa (16). The findings of the present study suggested that the downregulation of GALC protein expression in the NPC cells may be associated with EBV latent infection and the lymph node metastasis of NPC.

A study on EBV+ gastric carcinoma has suggested that EBV-infected cells acquire extensive methylation that silences multiple tumor-suppressor genes and rapidly transforms into cancer cells, without going through a precancerous stage of EBV infection or methylation accumulation (17). EBV+ gastric cancers show distinct methylation patterns that are likely attributable to EBV infection (18). The host defense system induces aberrant methylation in the host cellular genome (18). The host cells themselves have a methylation mechanism and, besides EBV infection, also triggers high methylation epigenotype, although this may be rare and in most cases is triggered by EBV infection (19).

LOH on chromosome 14q is detected in 80% of NPC tumors, with high-frequency LOH loci clustered at 14q11-q13,14q21-q24 and 14q32, while no LOH has been found at 14q31 (5). The GALC gene is located at 14q31, has 17 exons and 16 introns of ~60 kb. The deficiency of galactocerebrosidase (GALC) leads to the accumulation of psychosine, which is responsible for globoid cell leukodystrophy (Krabbe disease) in humans and certain animals. This enzyme catalyzes the lysosomal hydrolysis of specific galactolipids, including galactosylceramide (galactocerebroside) and galactosylsphingosine (psychosine). The 5′-untranslated region is GC-rich, containing no perfect CAAT or TATA sequences (6). Neither mutation nor other alterations of the GALC gene promoter sequence have been detected (16). Since the GALC gene is located at 14q31, no LOH has been found at 14q31 and neither mutation nor other alterations of the GALC gene promoter sequence have been detected. Therefore, we speculated that the GALC gene in NPC may be a tumor-suppressor gene, and downregulation of GALC protein may be associated with GALC gene promoter hypermethylation. Methylation of CpG sites is reversible with pharmacological demethylation using epigenetic agents such as 5-Aza-dC. Demethylation induced re-activation and expression of the genes. The NPC CNE-2Z cell line was EBV+ when it was established and now, after many years of in vitro culture, the CNE-2Z cell line is EBER− and LMP1−. In the present study, we found that the CNE-2Z cell line exhibits a low level GALC mRNA expression and does not express GALC protein. In addition, hypermethylation (95.9%) was found at the CpG sites of the GALC gene promoter in the CNE-2Z cell line. Treatment and control groups were used in the present study. The CNE-2Z cells in the control group were cultured for 48 h, while the cells in the treatment groups were treated with different concentrations (5, 10, 20, 40 and 60 μM/l) of 5-Aza-dC for 48 h. Then, the GALC mRNA and protein expression, GALC gene promoter methylation, and cell viability in each group were measured. In addition, the migration ability of the cells in the control group and the 20 and 40 μM/l treatment groups were also measured. The results showed that compared with the control group, the treatment groups showed a significantly lower rate of promotor methylation at the CpG sites in the GALC gene, and these differences were more pronounced in the 20, 40 and 60 μM/l treatment groups. The expression of GALC mRNA was significantly higher in the treated than in the control groups (P<0.05), and the expression in the 20 μM/l treated group was the highest, which was significantly different from that in the other groups (P<0.01). GALC protein expression was detected in all the treatment groups but not in the control group. Compared with the control group, the cell viability in the 20, 40 and 60 μM/l treatment groups was significantly lower (P<0.05), suggesting that the cell proliferation ability decreased following treatment. The scratch assay showed that the migration distance of the cells in the 20 and 60 μM/l treatment groups was significantly higher and the cell-migration ability was significantly lower than the corresponding values in the control group (P<0.01). The difference between the 20 and 60 μM/l treatment groups was not statistically significant (P>0.05). Transwell chamber assay showed that significantly fewer cells migrated to the lower side in the 20 and 60 μM/l treatment groups than in the control group (P<0.01). The cell migration ability was also significantly lower in the treatment groups. The difference between the 20 and 60 μM/l treatment groups was not statistically significant (P>0.05). These findings demonstrated that the silencing of the GALC gene in the CNE-2Z cells was closely associated with the hypermethylation of the CpG sites in the promoter, while demethylation effectively inhibited the proliferation and migration of the CNE-2Z cells. The results of the present study were in accordance with previous findings (20,21).

Experiments on DNA methylation of cancer-related genes have suggested that promoter hypermethylation is a potential biomarker for the early diagnosis and a therapeutic target in cancer (12–14). We report that GALC is downregulated in NPC, and that promoter hypermethylation contributes to the loss of GALC expression. To the best of our knowledge, this is the first study demonstrating GALC gene promoter hypermethylation in NPC. Our functional studies support the putative tumor-suppressor effect of GALC in the NPC CNE-2Z cell line. The present findings suggest that GALC gene promoter hypermethylation is a potential epigenetic biomarker for the early diagnosis and prediction of metastasis of EBV-associated NPC.

Abbreviations:

Acknowledgments

This study was supported by the National Basic Research Program of China (973 Program, no. 2011CB504800), and the 2013 Scientific Research Foundation of Guangdong Medical University, Zhanjiang, China (no. M2013033).

References