Background
The Epstein-Barr virus (EBV) infects nearly all humans. Although EBV injection is asymptomatic in most individuals, EBV is closely associated with several types of malignancies, including lymphomas, nasopharyngeal carcinoma (NPC), gastric adenocarcinoma and gastric lymphoepithelioma-like carcinoma [
1‐
3]. Due to the tropism for B lymphocytes, the most common forms of EBV-associated lymphoproliferative disorders are B-cell lymphomas: Hodgkin lymphomas (HL), non-Hodgkin lymphomas (NHLs), including Burkitt lymphoma (BL) and diffuse large B-cell lymphoma (DLBCL) [
4]. EBV also plays a complex and multifaceted role in T/NK cell lymphomas [
5].
The virus is present in two stages: an active lytic state and latent state, and the virus can only be spread during the active state [
6]. During primary EBV infection, a portion of individuals, particularly adolescents, develop infectious mononucleosis (IM). EBV infection in vitro can transform resting B cells into immortalized lymphoblastoid cell lines (LCL).Under normal conditions, EBV is present in the latent state in the peripheral blood B cell compartment. EBV can be reactivated under certain circumstances. Reactivation of the viral replication plays an important role in the development of EBV-associated malignancy. Oral hairy leukoplakia is unequivocally due to lytic EBV infection [
7]. Except for hematological malignancies, the lytic form of EBV infection was also observed in malignant breast epithelial cells in certain cases of breast cancer [
8]. EBV reactivation-induced cell proliferation and migration caused the relapse of NPC [
9]. Elevated levels of EBV DNA is observed in EBV-positive lymphoma patients with active disease but not in EBV-positive patients in remission or those with EBV-negative tumors, indicating that the EBV viral load at lymphoma diagnosis is an indicator of disease activity and biological characteristics associated with a negative prognosis [
10].
Environmental factors serve to promote EBV-driven tumors, which are primarily of B cell origin but also of epithelial and NK or T cell origin. Immunosuppression or DNA-damaging agents, including chemotherapy, certain HDAC inhibitors, and radiation, can induce the reactivation of EBV replication [
11]. EBV infection in turn has been implicated in DNA damage. Phosphorylated H2AX, a reporter of DNA damage was increased in EBV-carrying cells in the absence of exogenous stimuli [
12].
Deregulation of the cellular Myc proto-oncogene is the hallmark of all types of Burkitt lymphoma. The Myc-mediated oncogenic signaling pathway regulates miRNAs, especially miR-17-92 clusters. miR-17-92 is transcribed as a polycistron, which is subsequently processed into 7 mature miRNAs: miR-17-3p and -5p, miR-18a, miR-19a, miR-19b, miR-20a and miR-92a. The miR-17-92 cluster is frequently amplified or overexpressed in lymphoma [
13]. Compared with the other miR-17-92 members,which showed distinct functional interplays with Myc, the role of miR-18a remains unexplained. In our previous study, miR-18a was shown to promote malignant progression of nasopharyngeal carcinoma, which is closely related to EBV infection [
14]. miR-18a shares a homologous sequence with EBV-encoded BART 5. In this study, we aimed to elucidate the roles of miR-18a in the tumorigenesis of lymphoma and determine whether they are mediated by EBV.
Methods
Cell culture and patients samples
Human Burkitt lymphomas cell lines P3HR-1, Raji(EBV positive)and BJAB(EBV negative) were cultured in RPMI-1640 (HyClone, Life Sciences, Logan, UT, USA) supplemented with penicillin G (100 U/mL), streptomycin (100 mg/mL) and 10% fetal calf serum. Cells were grown at 37 °C in a humidified atmosphere of 5% CO
2 and were routinely sub-cultured. Cells were obtained from Cell Bank of Cancer Research Institute, Central South University and authenticated by STR profiling. Patients diagnosed with lymphoma (
n = 100) were included in this study. Twenty non-cancerous individuals were enrolled as a control group. All cases enrolled in this study were identified at Xiangya hospital, Central South University, China. The clinical and laboratory characteristics of the cases are summarized in Table
1. The patients were informed of the sample collection and signed informed consent forms. The collection and use of samples were approved by the ethical review committees of Xiangya Hospital, Central South University.
Table 1
The clinical, laboratory characteristics of paitents
Gender |
Men | 65 | 35 | 30 | |
female | 35 | 16 | 19 | 0.5304 |
Age |
≥50 | 55 | 28 | 27 | |
< 50 | 45 | 23 | 22 | 1.000 |
Extranodal lymph node metastasis |
Yes | 71 | 43 | 28 | |
NO | 29 | 8 | 21 | 0.0039 |
Clinical stage |
I-II | 37 | 12 | 25 | |
III-IV | 63 | 39 | 20 | 0.0017 |
EBV(EBNA-1) |
+ | 51 | 42 | 9 | |
– | 49 | 9 | 40 | < 0.0001 |
Analysis of public datasets
The microRNA array datasets (GSE42906, GSE36926) were collected from the National Center for Biotechnology Information’s Gene Expression Omnibus (GEO, NCBI). Using GEO2R of PubMed (
http://www.ncbi.nlm.nih.gov/geo/geo2r/), we analyzed changes in the miRNA expression spectrum. Expression of the miR-17-92 cluster and EBV-encoded microRNA was analyzed and visualized by Multiple Experiment Viewer (MeV).
Immunohistochemistry and in situ hybridization
Lymphomas samples and normal lymph node tissues were fixed and embedded in paraffin wax. Next, 4- to 6-μm thick paraffin sections were defaced followed by hydration. The detailed procedures for the immunohistochemistry for EBNA1 and in situ hybridization for miR-18a have been described in our previous publication [
14]. Tissue sections were incubated with primary antibody EBNA1 (Novus Biologicals, Littleton, Colorado, USA) at 4 °C overnight in a humidified chamber. Finally, after dehydration and mounting, the sections were observed and imaged under a microscope (OLYMPUS BX-51, Japan). Goat serum and PBS were used instead of the first antibody as a negative control and blank control respectively. A semi-quantitative scoring criterion for IHC was used in which both the staining intensity and positive areas were recorded.
Cell cycle analysis
Cells were fixed with 70% ethanol overnight and were incubated in PBS containing 0.5 mg/ml RNase A (Takara, Japan) at 37 °C for 1 h. Cells were treated with propidium iodide at the final concentration of 5 μg/ml (Beyotime Biotechnology, Hangzhou, China) and subjected to flow cytometry analysis.
Quantitative real-time PCR
Total RNA was isolated from cells using Trizol® reagent (Invitrogen, Carlsbad, CA, USA). cDNA was synthesized from total RNA using the RevertAid First Strand cDNA Synthesis Kit (Fermentas, Waltham, MA, USA). To measure the expression of miR-18a, the microRNA qPCR Quantitation kit was purchased from Takara, Japan. Real-time PCR was performed using the Bio-Rad IQ™5 Multicolor Real-Time PCR detection System (Bio-Rad, Berkeley, CA, USA). The data were analyzed using iQ5 software. Relative EBV-related gene expression was determined and normalized to the expression of GAPDH or β-actin (for hypoxic conditions) using the 2ΔΔCt method. miR-18a expression was calculated and normalized to U6 using the 2ΔΔct method. The data are representative of the means of three experiments. Student’s t-test was applied to compare two or more values; p < 0.05 indicated that there was a significant difference. The qPCR protocol was as follows: 95 °C for 30 s and 40 cycles of 95 °C for 5 s and 60 °C for 30 s. A final melting curve analysis (60–95 °C) was conducted. The RT-PCR primers are listed below.
ERER1/2 forward:5’-GAGGTTTTGCTAGGGAGGAGA-3′.
ERER1/2 reverse:5’-CACCACCCGGGACTTGTA-3’.
EBNA1 forward: 5’-GAGAAGGGAGACACATCTGGA-3’.
EBNA1 reverse: 5’-CATGGTTATCACCCCCTCTTC-3’.
LMP1 forward:5’-CCCACTCTGCTCTCAAAACC-3’.
LMP1 reverse:5’-GTCCTGTGGGCCATTGTC-3’.
BZLF1 forward:5’-AGCTTATGCACCTCAGC-3’.
BZLF1 reverse:5’-CCGGCTTGGTTAGTCTGTTG-3’.
BMRF1 forward:5’-GCGAGGAAAAGGACATCGT-3’.
BMRF1 reverse:5’-CTTCACTTTCTTGGGGTGCT-3’.
BRLF1 forward:5’-GCTCAGGTCCATCTGTCCAC-3’.
BRLF1 reverse:5’-GGGAGATGGCTGACACTGTT-3’.
GAPDH forward:5’-AGCCACATCGCTCAGACAC-3’.
GAPDH reverse:5’-GCCCAATACGACCAAATCC-3′.
β-actin forward: 5’-ACTGGAACGGTGAAGGTGAC-3′.
β-actin forward: 5’-AGAGAAGTGGGGTGGCTTTT-3′.
Quantitative real-time EBV PCR was performed in samples collected from the media of cells using the Epstein-Barr virus DNA Quantitative Fluorescence Diagnostic Kit (Sansure Biotech, Hunan, China). Viral DNA was extracted, and PCR was performed according to the manufacturer’s instructions. The qPCR protocol was as follows: 94 °C for 5 min, followed by 45 cycles of 94 °C for 15 s and 57 °C for 30 s. The EBV copy number was calculated according to the standard curve.
EBV purification and fluorescence in situ hybridization
An EBV-specific probe was constructed from EBV and obtained from the productive EBV B-cell lineage P3HR-1, which has been describe in detail previously [
15,
16]. Briefly, the culture media were collected from P3HR-1, followed by repeated freezing and thawing for several times. After centrifuged for 20 min at 3000 RPM, the supernatant was collected and filtered with a 0.45-μm membrane filter. EBV viral DNA was extracted using the Qiagen QIAamp Virus MinElute Spin Kit (Qiagen, Valencia, CA, USA). PCR was performed to amplify the 3267 bp of EBV DNA probes. The DNA probes were labeled with biotin-dUTP by the random-primed labeling method (Roche, Mannheim, Germany). The slides were incubated with hybridization mixture containing 50 ng of probe and 5 μg of salmon sperm DNA overnight at 37 °C for 14 h. Following hybridization, the slides were incubated with avidin-FITC for 30 min at room temperature followed by incubation with anti-FITC. After washing, drying and mounting, the slides were examined under fluorescence microscopy (Olympus FSX100).
Western blotting
The procedures for western blotting have been described in our previous publication [
14]. Briefly, approximately 50 μg of protein were separated on SDS-PAGE and was transferred to PVDF membranes (Millipore, MA, USA). The membranes were incubated with the primary antibody overnight at 4 °C followed by a brief wash with PBST and incubation with secondary antibody for 1 h at 37 °C. An anti-GAPDH antibody control was purchased from Millipore (Billerica, MA, USA) and was used as a loading control. Finally, ECL solution (Millipore, MA, USA) was added to cover the blot surface, the signals were captured, and the intensity of the bands was quantified using the Bio-Rad ChemiDoc XRS+ system (Bio-Rad, CA, USA). The antibodies were used as follows: ATM, EBNA1 (Novus Biologicals, USA), γ-H2AX(Abcam, USA), β-actin (Santa Cruz Biotechnology, CA, USA) and GAPDH (cell signaling, USA).
In vitro cell proliferation assessment
The proliferation of lymphoma cells in suspension was measured using the CCK-8 assay (Beyotime Biotechnology, Hangzhou, China). The cell suspension was inoculated in a 96-well plate. After treatment, 10 μl of CCK-8 solution was added to each well and the plate was incubated for an additional 4 h. Next, the absorbance measured at 450 nm using a microplate reader. The experiment was repeated three times, and six parallel samples were measured each time.
Hypoxia and UV treatment
The cells were incubated in the hypoxic chamber with 0.1% oxygen, 95% N2 and 5% CO2 (Don Whitley Scientific, H35 hypoxystation) for 48–72 h. For UV treatment, cells were subjected to UV light (254 nm) for 10 min to cause DNA damages.
Immunofluorescence analysis
Cells were collected, washed and subsequently fixed in fixation buffer (4% paraformaldehyde in PBS). Next, the cells were permeabilized for 30 min with 0.25% Triton X-100 and blocked for 60 min with 5% BSA. Cells were incubated with anti-γ-H2AX antibody at a 1:800 dilution at 4 °C overnight, washed in TBST 3 times, and incubated with the secondary antibody FITC-conjugated IgG (BD biosciences, USA) at a dilution of 1:300 for 1 h. Next, cells were washed in TBST at room temperature, dried and mounted by mounting medium including the nucleus stain DAPI (Vector Laboratories). The cells were examined by fluorescence microscopy (Olympus FSX100).
Luciferase assay
Reporter constructs were generated in which the 3’UTR of ATM, wild-type or miR-18a binding-site mutant was cloned downstream of the luciferase open reading frame. Luciferase activity was measured using the Dual-Glo luciferase assay system (Promega). Renilla luciferase activity was normalized to the corresponding firefly luciferase activity. 293 cells were co-transfected with luciferase constructs and miR-18a mimics or mimics negativa control (mimics NC). The renilla constructs were also cotransfected as an internal control. Luciferase acticity was normalized to renilla luciferase activity. Data were presented as the mean ± SD of twice experiments with six replicates (Student’s t-test, *p < 0.05; **p < 0.01).
Alkaline comet assay
The alkaline comet assay was performed according to the procedure of alkaline Comet assay kit (Genmed Scientifics, USA). The detailed protocol has been previously published [
17]. Briefly, cells were pelleted and resuspended in 1 mL of PBS. Next, 10 μL of resuspended cells was mixed with an equal volume of prewarmed low melting- point agarose. The agarose-cell mixture was placed on fully frosted slide precoated with agarose and spread gently with a coverslip. After 10 min at 4 °C, the slides were immersed in precooled (4 °C) lysis solution for 80 min in a dark chamber. After soaking with electrophoresis buffer for 30 min, the slides were subjected to electrophoresis (25 V) for 30 min. Finally, the cells were stained with 10 μg/ml propidium iodide (Beyotime Biotechnology, ST511), and individual cells were viewed using an Olympus FSX100 fluorescence microscope.
Array CGH
The detailed protocol has been previously published [
17]. CGH was performed using Agilent SurePrint G3 Human Catalog 2x400K CGH Microarrays (Agilent, CA, USA). Genomic DNA of P3HR-1 or Raji cells transfected with miR-18a or mimics NC were extracted with the Universal Genomic DNA Kit (CW2298S, Cwbio, Beijing, China). The visualization and image analyses were performed using Feature Extraction (Agilent, Santa Clara, CA, USA).
Statistical analysis
Survival data were analyzed by Kaplan-Meier analysis. The log-rank test was used to determine the difference among survival curves according to miR-18a and EBNA1. Pearson χ2-test was used to discover the association between miR-18a and EBNA1. The real-time PCR data were represented as the means ± standard deviation.Statistical significance was determined using a 2-tail Student’s t-test. Statistical analyses were performed by SPSS11.0 and GraphPad prism. For cell cycle analysis, statistical significance for fold over miRNA negative control was determined using a Student’s t-test. All p values were two-sided, and a p value of less than 0.05 was considered to be significant.
Discussion
In this study, we performed a comprehensive analysis of expression of the miR-17-92 cluster and EBV status regarding lymphomas. Overexpression of miR-18a is observed in lymphomas tissue and EBV-infected B cells. We demonstrated that miR-18a can reactivate EBV through inhibiting the DNA damage response and induce genomic instability through EBV reactivation.
EBV established a persistent, asymptomatic, latent infection in a restricted pool of resting B cells [
6,
21]. In immunocompetent adults, the EBV viral load is stable for years. It is assumed that the latently infected memory B cells undergo periodic reactivation to produce infectious virus [
22]. EBV reactivation begins through expression of the immediate-early transcription factors BZLF1, which initiates a cascade of expression of early genes and late genes [
23]. The roles of EBV in the lytic or latent state in the development of lymphoma are not fully understood. B cell transformation after EBV infection in vitro is believed to be a latent infection. However, the in vitro model is not consistent with the viral gene expression programs observed in most EBV-positive B-cell tumors [
24]. Although the EBV latent gene-encoded products activate survival pathways to promote the progression of cancer and cellular transformation, an increased level of lytically infected cells may increase the likelihood of EBV-associated malignancies [
25,
26]. EBV viral load measurement is suggested to predict and monitor EBV-associated tumors, including nasopharyngeal carcinoma, post-transplant lymphoproliferative disorder, and Hodgkin’s disease [
27]. AIDS patients and organ transplant patients, who have a high risk of developing EBV-associated lymphomas, also have a high level of lytic as well as latent EBV infection [
26,
28]. A high level of EBV particles would be predicted to increase the number of latently infected B cells. The differentiation of EBV-infected memory B cells into plasma cells and acute stress represent two distinct pathways of EBV reactivation [
23]. Although this treatment is controversial, antiviral drugs that block virus replication and kill proliferating infected cells were administered to EBV-positive lymphoma patients [
29]. Our data showed that miR-18a caused a burst of lytic gene expression and concomitant transcription of EBV latent gene expression. miR-18a promoted the proliferation of EBV-positive lymphoma cells but not EBV-negative lymphoma cells, suggesting that miR-18a-induced EBV reactivation mediated the increase in lymphoma cell growth.
Although EBV infection is related to B cell malignancy, the slow course of cancer induction and infections worldwide suggest that EBV alone is rarely sufficient to cause lymphoma. EBV infection, endogenous and exogenous damages in DNA and host repair mechanisms play synergistic effects on tumorigenesis of lymphomas. Exposure to DNA damaging agents, such as certain chemotherapy, ultraviolet or hypoxia, can induce the reactivation of viral replication, causing fold increases in the EBV viral load [
6,
21]. The replication of EBV, in turn, increases replicative stress, similar to oncogene activation, leading to the induction of the DNA damage response (DDR). DDR emerges as a barrier to tumor progression responding to cellular DNA replicative stress [
30]. The host DDR, which is orchestrated by ATM and ATR kinase, senses EBV-induced oncogenic stress,causes cell cycle arrest or apoptosis and blocks long-term outgrowth of most infected cells [
19] [
31]. Inhibition of the DDR kinase ATM markedly increases the transformation efficiency of primary B cells [
19]. ATM kinase expression is reduced in EBV-associated nasopharyngeal carcinomas [
32]. It was reported that the DNA repair sensor ATM causes EBV lytic reactivation in Burkitt lymphoma cells in p53-dependent and non-dependent ways [
19]. The EBV latent-to-lytic switch is mediated by the viral proteins BZLF1, BRLF1 and BRRF1 [
11]. The ability of the EBV BRLF1 and BRRF1 proteins to induce lytic reactivation in EBV-infected AGS cells is ATM dependent. However, overexpression of BZLF1 induces lytic gene expression in the presence or absence of ATM activity [
19]. ATM enhances BZLF1 promoter activity in the context of the intact EBV genome [
11]. In our study, under normoxia, ATM induced the overexpression of BRLF1 and BMRF1, however, decreased the expression of BZLF1 in EBV-positive Raji cells. Under DNA damage stress, ATM decreased lytic and latent EBV gene expressions and reversed the induction of miR-18a on the EBV viral load. We speculated that ATM decreased EBV gene expression, including lytic or latent genes in the context of EBV- or DNA damage agents-induced genomic stress. As a host mechanism, miR-18a acts as a sensor to DNA damage stress induced by hypoxia, UV and EBV infection, inhibiting DDR and promoting genomic instability. miR-18a inhibited ATM-mediated DDR and increased the virus proliferation, increasing the growth of lymphomas and indicating the oncogenic effect of EBV is dependent on the host microenvironment. DNA damage that is not repaired leads to genomic instability and cancer. It is now generally accepted that oncoproteins encoded by tumor viruses can drive genomic instability and initiate tumorigenesis. EBV infection promotes genomic instability and telomere dysfunction [
12]. Repeated malarial infection helps EBV to cause lymphomas. However, it was reported that antibodies to fight malaria cause DNA damage that can lead to Burkitt’s lymphoma because of the ligation of inappropriate segments of chromosomes during antibody affinity maturation [
33]. Thus, host factors contribute to the genomic instability induced by viral replication.
There are three EBV-encoded microRNAs that share sequences with human miRNAs: EBV miRNA BART1-3p with human miRNA miR-29a/b/c; BART5-5p with miR-18a/b and BART22-3p with miR-520d-5p, and miR-524-5p [
14,
34]. EBV-encoded microRNAs have been reported to regulate human and viral transcripts. EBV can also alter host miRNA expression [
34]. BART miRNAs are present in all EBV-infected cells, with much higher expression in epithelial cells than in infected B lymphocytes. BART miRNAs act as repressors of EBV lytic replication and probably maintain a balance between the virus and its host [
35]. In our study, EBV can enhance the expression of miR-18a (Additional file
1: Figure S3). In contrast to the roles of BART miRNAs, the host miR-18a promoted EBV replication. Most EBV miRNAs co-target mRNAs with host miRNAs, in particular with members of the miR-17-92 miRNA cluster [
34]. Our study found that ATM is targeted by miR-18a in EBV-infected BJAB cells but not in EBV negative BJAB, indicating that EBV miRNAs co-target mRNAs with host miRNAs. Although sharing seed sequence identity, miR-18a and BART5-5p showed different potential to regulate expression of LMP1. BART5-5p can repress the expression of LMP1 and not miR-18a-5p [
34].