Background
The development of hepatocellular carcinoma (HCC) is closely associated with infection with the hepatitis B virus (HBV) [
1]. However, the molecular mechanisms involved in HBV-mediated hepatocarcinogenesis remain unclear. The current study suggests that osmotic stress plays an important role in the development of inflammation and tumors [
2]. The liver, kidneys, gastrointestinal tract and other tissues and organs are significantly differentially affected by exposure to an environment with moderate changes in osmotic pressure [
3]. In rat livers with hypotension, the liver cells swell and become hypotonic, which activates mitogen-activated protein kinase (MAPK) pathways and inhibits proteolysis [
4]. However, liver cells continue to shrink and become dehydrated while in a hyperosmolar state, resulting in the activation of CD95 and the eventual induction of apoptosis [
5]. Further research into the relationship between osmotic stress and apoptosis in liver cells indicated that hepatocyte apoptosis can be inhibited by activating the integrin/Src/p38 MAPK signaling pathways in cells in a hypotonic state, whereas hepatocyte apoptosis can be promoted by activating NOX/ROS/CD95 while cells are in a hyperosmolar state [
6,
7]. The liver cells of patients with chronic hepatitis B virus infections reside in a long-term hypertonic environment resulting from inflammatory infiltration, which leads to the accumulation of mutations and malignant representation in liver cells. These phenomena result from the rapid cellular regeneration induced by the environment of sustained necrosis. Therefore, osmotic stress in liver tissues may be associated with the occurrence and development of HBV-associated HCC.
Nuclear factor of activated T-cells 5 (NFAT5), also known as tonicity-responsive enhancer-binding protein (TonEBP), is a transcription factor that is crucial for cellular responses to hypertonic stress [
8]. Unlike other members of the NFAT family (NFAT1-NFAT4), NFAT5 is not regulated by Ca
2+ and, thus, is defined as the calcium-independent phosphatase calcineurin [
9]. Under high osmotic pressure, NFAT5 promotes the transcription of osmotic pressure protection genes, such as aldose reductase (AR), taurine transport protein (TauT), Na
+-dependent myo-inositol cotransporter (SMIT) and heat shock protein 70 (HSP70), thereby protecting cells from injury resulting from high osmotic pressure [
10‐
12]. It has been suggested that NFAT5 plays important roles in different cell types and tissues, in an at least partly tonicity-independent manner, during embryonic development, cell differentiation, inflammatory processes, and cellular stress responses [
13]. Recent studies have shown that NFAT5 is involved in the pathogenesis of multiple cancers, including non-small cell lung cancer, leiomyoma, breast cancer, melanoma and renal carcinoma [
14‐
17]. However, the role of NFAT5 in HBV-associated HCC has never previously been investigated.
In this study, we investigated the mechanism whereby HBV affects NFAT5 by separating the upstream pathway and convergent downstream pathways of NFAT5 in hepatoma cells. Our results showed that HBV could suppressed NFAT5 expression by inducing hypermethylation of the AP1-binding site in the NFAT5 promoter and inhibiting miR-30e-5p/MAP4K4/DARS2 pathway in hepatoma cells. DARS2, as a downstream target gene of NFAT5, promoted HCC tumorigenesis by accelerating cell cycle progression and attenuating cell apoptosis. Our findings provide a novel biomarker for early HCC diagnosis and a novel therapeutic target in HCC.
Methods
Patient samples and clinical data
HCC tissues were collected in our hospital during hepatectomies of HCC patients who were first diagnosed with HCC and had not received any prior treatment. The patients and their families agreed to donate the HCC tissue for our research. The HCC tissues were collected and immediately stored at −80 °C in an RNA protective reagent. Clinical data were collected from EMR(electronic medical record) of our department. Detailed clinical data are listed in Table
1 and Additional file
1: Table S1. Follow-up visits for the survival analysis occurred for up to 80 months for NFAT5, or 40 months for DARS2 until recurrence or death.
Cell culture and transfection
Cell lines, including the human normal liver cell L02 and the hepatoma cell lines Huh 7, HepG2 and Hep3B cells were obtained from the Cell Bank of the Shanghai Institute of Cell Biology at the Chinese Academy of Science (Shanghai, China), where they were characterized by mycoplasma detection, DNA fingerprinting, isozyme detection, and determination of cell viability. The HepG2.2.15 cell line was derived from HepG2 cells and stably expresses HBV(Genotype D, Serotypeayw, U95551), which was used as an HBV replication model. The stable cell lines were maintained in DMEM containing 400 μg/ml G418. The plasmid pCMV-HBV-1.3, which expresses HBV (genotype C, serotype adr, FJ899793), was a gift from Dr. Ying Zhu (State Key Laboratory of Virology, College of Life Sciences, Wuhan University, China). All cells were incubated at 37 °C in a humidified atmosphere containing 5% CO2. MiR-30e-5p mimics, NFAT5 siRNA, and DARS2 siRNA were transfected into cells using GenMute siRNA transfection reagent (SignaGen Laboratories, USA). The NFAT5 plasmid was transfected using a Lipojet in vitro siRNA and DNA transfection kit (SignaGen Laboratories, USA). Subsequent experiments were performed 48 h after transfection. Cells were treated with the c-MYC specific inhibitor 10,058-F4 (Selleckchem, USA) at a concentration of 50 mM for 48 h.
Real-time quantitative PCR
RNA was extracted from tissues and cells by TRIzol reagent (Biosharp, China). The concentration and quality of the total RNA were examined using a Nanodrop 2000 Spectrophotometer. The concentration was between 800 ng/μl and 2500 ng/μl, with the A260/280 between 1.8 and 2.0. MicroRNA reverse transcription was performed using a miRNA cDNA Synthesis Kit with Poly (A) Polymerase Tailing (abm, Canada). The mRNA was reverse-transcribed with PrimeScript RT Master Mix (Takara, Japan). The miR-30e-5p relative expression in HCC tissue was calculated as log2 (2-ΔΔCT) based on the threshold cycle and normalized to U6 expression. The mRNA relative expression was calculated as 2-ΔΔCT based on the threshold cycle and normalized to β-actin expression. RT-qPCR was performed on a Bio-Rad iQ5 instrument using SYBR green mix (Toyobo, Japan).
Protein extraction and western blot analysis
Total protein was extracted by lysis cells in RIPA and 1% PMSF for 30 min on ice, and the lysates were then centrifuged at 12,000 rpm and 4 °C for 15 min. The supernatants were collected and boiled at 95 °C for 5 mins with 5X SDS loading buffer. The total protein quantity was measured by a BCA assay. An appropriate volume of each sample was loaded onto an SDS-PAGE gel. After SDS-PAGE, the proteins were transferred to a PVDF membrane. The PVDF membrane was incubated in blocking buffer (5% skim milk in TBS-T) for 2 h at RT. The membrane was then incubated with a primary antibody at 4 °C overnight. The next day, the membrane was washed with TBS-T and then incubated with a secondary antibody at 4 °C for 1–2 h. Finally, the PVDF membrane was rinsed, and the signals were developed using ECL methods.
ChIP-PCR and ChIP-seq
The ChIP assay was performed using the Magna ChIP-Seq™ Chromatin Immunoprecipitation and Next Generation Sequencing Library Preparation Kit (17–1010, Millipore, USA). Next-generation sequencing was performed by Orbiotech (China). The ChIP-PCR primers were as follows: NFAT5 promoter forwwad primer 5′- CCCAACCCTAGCACTCCAA-3′ and NFAT5 promoter reverse primer 5′- ATATTGATGCCAGTAGCCA CG-3′. The antibodies for ChIP were as follows: NFAT5 antibody (HPA069711, Sigma-Aldrich, USA) and c-MYC antibody (ChIP Grade, ab32, Abcam, USA).
Luciferase reporter assay and plasmid construction
A 2000-bp promoter construct of the NFAT5 gene, corresponding to the sequence from nt −2000 to 0 (relative to the transcriptional start site) of the 5′-flanking region of the human NFAT5 gene, was generated from human genomic DNA by PCR. The PCR product was cloned into the Kpnl and Nhel sites of the pGL3-Basic vector. Efficiency of construct was confirmed by DNA sequencing. The 5-flanking deletion constructs of the NFAT5 promoters, NFAT5-promoter-(50 bp, 237 bp, 409 bp, 741 bp, 1012 bp, 1531 bp) were similarly generated by PCR, using the NFAT5-promoter-2000 bp construct as a template. The pBlue-HBV plasmids were transfected into Huh 7 cells. MiR-30e-5p mimics and negative control mimics were transfected into Hep3B cells. The wild-type and mutant MAP4K4 3’UTR sequences are displayed in Supplimentary data. The MAP4K4 3’UTR or NFAT5 promoter was amplified and cloned into the pGL3-reporter vector containing the firefly luciferase gene. Co-transfection is proceede with firefly luciferase report gene, mimics or pBlue-HBV and pRL-TK plasmid with Renilla luciferase gene. The firefly and Renilla luciferase activities were detected 48 h after transfection. The firefly luciferase activities were normalized to Renilla luciferase activities.
Electrophoresis mobility shift assay (EMSA)
EMSA was performed using a nonradioactive EMSA kit (PIERCE, cat:89,880), The 5′ end of the oligonucleotides was biotin-labeled. Probe was labeled and purified at first, The sample binding reaction system:ddH2O Water 5.8ul, 10X binding buffer 1.5ul, polydI:dC 2ul, sample 4.7ul, Bio-AP1 probes 1ul. The Super-shift reaction system: ddH2O Water 5.8ul, 10X binding buffer 1.5ul, polydI:dC 2ul, sample 4.7ul, Bio-AP1 probes 1ul, the AP1 specific antibody 4.0ul.The reaction procedure was accorded to the manufacturer’s instructions. Using an Imager apparatus (Alpha Innotech, San Leandro, CA) to obtain images.
Quantitative Methylation changes of NFAT5 promoter
Genomic DNA was isolated from all samples using the Cell genomic DNA extraction Kit (Generay, cat: GK 0122). An Epitect bisulfite kit (Qiagen AG, Basel, Switzerland) was used to perform bisulfite conversion of the genomic DNA. The shrimp alkaline phosphatase (Sequenom, San Diego) was used to remove unincorporated dinucleotide triphosphates.2ul of PCR product was used as a template for the transcription reaction, which was performed by the following PCR. Then T cut/RNase A digestive response procedures was performed. Resin Purification before performing MALDI-TOF MS analysis.The RNase A-treated product was robotically dispensed onto a silicon matrix of preloaded chips (SpectroCHIP; Sequenom, San Diego). The EpiTYPER software version 4.0 was used to detect the methylation ratios of the spectra.
Immunohistochemical staining (IHC)
IHC for NFAT5 and DARS2 was performed on HCC and para-tumor tissues of patients at Zhongnan Hospital of Wuhan University. The antibodies for IHC were as follows: NFAT5 antibody (HPA069711, Sigma-Aldrich, USA) and DARS2 antibody (ab154606, Abcam, USA).
Flow cytometry assay for apoptosis and cell cycle
Cells for the FCM apoptosis assay were stained with Annexin-V FITC/PI. FCM was performed 48 h after transfection. The apoptotic rate was calculated as the total sum of the acute apoptotic rate and the terminal apoptotic rate. Cells for the FCM cell cycle assay were stained with PI. FCM was performed 48 h after transfection. Cell counts in each phase of the cell cycle are displayed along with the proportion of the total cell counts.
Statistical analysis
A paired t-test was performed to assess the variance differences between miR-30e-5p and DARS2 expression in tumor and non-tumor tissues for statistical significance. An unpaired t-test was performed to measure the significance of continuous data. A chi-square test was used to determine the relationship between the clinical data and DARS2 expression. ROC curve generation, Kaplan-Meier survival analysis and COX regression analysis were performed by SPSS 21. Statistical significance was indicated by P < 0.05 (*), P < 0.01 (**), and P < 0.0001 (***). All experiments were repeated in triplicate.
Discussion
HBV is the most common hepatitis virus and causes chronic infections in the human liver. Recent findings suggest that osmotic stress plays an important role in inflammation and tumorigenesis [
2]. NFAT5/TonEBP/OREBP is the only known osmotic pressure-sensitive transcription factor in mammals and plays a vital role in enhancing cell survival, migration and proliferation, vascular remodeling, tumor invasion and angiogenesis [
21]. NFAT5 has been shown to be involved in the pathogenesis of multiple cancers. Knocking out NFAT5 reduces the invasiveness of melanoma, and the tumor suppressor gene miR-211 is thought to downregulate the expression of NFAT5 in melanoma [
22]. NFAT5 is located at a central node that promotes metastasis in melanoma [
12]. Some studies on breast and colon cancer have confirmed that the α6β4 integrin cluster promotes the expression of NFAT5 and the invasiveness and migratory capabilities of tumor cells [
23]. NFAT5 may therefore promote tumor cell migration, leading to the formation of various new tumors.
Here, we present the first evidence showing that NFAT5 is involved in the proliferation of HBV-associated HCC. However, we found that NFAT5 has a completely different function in HCC than in other types of human cancers, including breast cancer, colon carcinoma, lung adenocarcinoma, renal cell carcinoma and melanoma [
2,
3,
16,
22,
24]. The key finding of the current study is that NFAT5 acts as a tumor suppressor by inhibiting cell cycle progression and promoting tumor cell apoptosis in vitro. We demonstrated the presence of a clearly positive correlation between the level of NFAT5 protein expression and OS in patients with HBV-associated HCC, suggesting that NFAT5 is a cancer suppressor gene. We further investigated the mechanism whereby HBV affects NFAT5 through separating the upstream pathway and convergent downstream pathways of NFAT5 in hepatoma cells. We found that HBV suppressed NFAT5 expression by inducing hypermethylation of the AP1-binding site in the NFAT5 promoter. Though HBV doesn’t encode any methylases, however HBV infection can stimulate the overexpression of DNMTs, particularly DNMT1, DNMT3A and DNMT3B [
25]. We speculate the overexpression of DNMTs may result in the hyper-methylation of the AP1-binding site in the NFAT5 promoter. In addition HBV also inhibited NFAT5 through miR-30e-5p targeting of the MAP4K4 signaling pathway. DARS2 promoted HCC tumorigenesis by accelerating cell cycle progression and attenuating cell apoptosis, as a downstream target gene of NFAT5. Our results suggest that the upregulation of DARS2 by HBV promotes hepatocarcinogenesis through the miR-30e-5p/MAP4K4/NFAT5 pathway.
MicroRNAs (miRNAs) comprise a group of small noncoding RNAs regulating gene expression at the posttranslational level, thereby participating in fundamental biological processes, including cell proliferation, differentiation, and apoptosis [
26]. Mounting evidence indicates that dysregulation of miRNAs plays important roles in HBV infection and HBV-associated HCC [
27‐
30]. Here, by screening a series of miRNAs based on reports in the literature, we identified miR-30e-5p as an effective reinforcer of NFAT5. We discovered that miR-30e-5p was downregulated in HBV-associated HCC, acting as a tumor suppressor, and that HBV could suppress miR-30e-5p expression. Several studies support our findings. For example, miR-30e is expressed at significantly lower levels in the sera of HCC patients than in healthy volunteers, suggesting serum miR-30e as a novel noninvasive biomarkers of hepatocellular carcinoma [
31]. miR-30e inhibits the proliferation of hepatoma cells through directly targeting the 3′-UTR of P4HA1 mRNA [
32]. However, NFAT5 is not a target gene of miR-30e-5p according to prediction using TargetScan. The existing literature indicates that the p38/MAPK pathway is associated with the NaCl-induced nuclear translocation of NFAT5 [
18,
33,
34]. Moreover, the TargetScan database indicated that MAP4K4 is a target gene of miR-30e-5p, which also participates in the MAPK signaling pathway. We hypothesized that the MAPK signaling pathway might be involved into the correlation between NFAT5 and miR-30e-5p. Thus, we performed Western blot assays of components of the MAPK signaling pathway under miR-30e-5p overexpression. The results showed that miR-30e-5p inhibited MAP4K4 expression and thereby suppressed the downstream protein c-MYC by decreasing the level of phosphorylated ERK1/2 in Hep3B cells. We also found that miR-30e-5p inhibited MAP4K4 by binding to the 3128 bp–3135 bp positions of the MAP4K4 3′-UTR in luciferase reporter assays, which indicated that MAP4K4 is a direct target gene of miR-30e-5p. We then focused on c-MYC, since it can bind to the NFAT5 promoter according to ALGGEN PROMO, and c-MYC is downstream of the MAPK signaling pathway according to the KEGG database. We next studied the mechanism of interaction between c-MYC and NFAT5 via ChIP-PCR. The results showed that c-MYC inhibited NFAT5 transcription by binding to the NFAT5 promoter (positions −observed to be upregulated when we treated Hep3B cells with 10,058-F4, a c-MYC inhibitor, indicating that c-MYC inactivates NFAT5 transcription via binding to its promoter. These results suggest that HBV inhibits NFAT5 expression through miR-30e-5p targeting of the MAP4K4 signaling pathway.
To further study the molecular mechanism underlying the promotion of hepatoma cell apoptosis by NFAT5, we identified targeted proteins associated with NFAT5 function to illustrate the role of the downstream pathway of NFAT5 in HCC. The combination of ChIP and next-generation sequencing is a biochemical technique used to identify direct gene targets of a factor of interest in a genome-wide, unbiased manner and to elucidate their functional significance [
35]. Novel targets of NFAT5 were revealed using ChIP-sequencing, as shown in Table 2. Among these targeted genes, DUSP5P and UBE2MP1 are lncRNAs and were therefore excluded from further analysis. A literature review of the other 5 targets showed that some of these genes have been previously studied in tumor diseases. RNF126 promotes the proliferation and viability of tongue cancer by regulating the AKT signaling pathway [
36]. RNF126 was found to positively regulate BRCA1 by directly interacting with E2F1 for homologous recombination in breast and ovarian cancer [
37]. WRNIP1 is involved in cell cycle progression, and its phosphorylation is reduced by FGFR1OP in lung cancer [
38]. However, DARS2 has not been previously studied in any tumor disease. The DARS2 gene is located on chromosome 1q25.1 and encodes mitochondrial aspartyl-tRNA synthetase, which is important for the mitochondrial unfolded protein response (UPRmt) [
39]. Mutations in the DARS2 gene are associated with leukoencephalopathy (LBSL) [
19,
20]. To study the relationship between NFAT5 and DARS2 in HBV-associated HCC, we investigated the effect of HBV on NFAT5 and DARS expression using HepG2.2.15 cells carrying an integrated fragment of HBV genomic DNA and HepG2 cells that did not carry HBV genomic DNA. The results showed that HBV upregulated DARS2 expression and inhibited NFAT5 expression simultaneously. We next found that NFAT5 negatively regulated DARS2, and the inhibition of tumorigenesis by NFAT5 on was executed through DARS2 in different hepatoma cell lines. Subsequent experiments showed that DARS2 expression was higher in HCC tissues than in para-tumor tissues and that DARS2 regulated the cell cycle progression and apoptosis of HCC cells. The present study revealed that aberrant expression of DARS2 contributed to HCC development, and DARS2 may be a potential target for the treatment and diagnosis of HCC. Although DARS2 has not been identified in previous tumor studies, its biological function is well documented. DARS2 depletion leads to severe deregulation of mitochondrial protein synthesis, followed by a large mitochondrial respiratory chain (MRC) deficit [
40], indicating that DARS2 upregulation could contribute to higher mitochondrial efficiency. The mitochondrial-dependent apoptosis pathway is initiated by cyclophilin-D, leading to reduction of the mitochondrial membrane potential (MMP) and, ultimately, opening of the mitochondrial permeability transition pore (mPTP) [
41]. A recent study showed that ECHS1 binding to HBsAg decreased MMP, inducing mitochondrial-dependent apoptosis [
42]. Our clinical data indicated that DARS2 was associated with HBV infection, and DARS2 might therefore be associated with ECHS1. Thus, we hypothesize that DARS2 positively regulates mitochondrial function and induces MMP, contributing to the growth of HCC cells. However, this hypothesis requires verification through further research. The available evidence suggests that DARS2 plays an essential role in mitochondrial function. However, the mechanism underlying the regulation of mitochondrial function by DARS2 remains unclear. Overall, how DARS2 promotes HCC tumorigenesis is a topic that deserves further study.
In summary, the present study illustrates that the upregulation of DARS2 by HBV promotes hepatocarcinogenesis through the miR-30e-5p/MAPK/NFAT5 pathway. Furthermore, NFAT5 acts as a tumor suppressor in HBV-associated HCC tissues by suppressing DARS2 expression. As aberrant expression of DARS2 contributes to HCC development, DARS2 may be a potential target for the treatment and diagnosis of HCC. Our results suggest that two pathway of inhibition on NFAT5 expression mediated by HBV may play an important role in the progression of HBV-associated HCC. These findings provide new insights that increase our understanding of the molecular mechanisms involved in the development of HBV-associated HCC via the miR-30e-5p/MAPK/NFAT5/DARS2 pathway.