Background
Ovarian cancer (OC) is the sixth commonest diagnosed malignancy among women around world, the second most frequent gynecologic cancer, and also the most lethal tumor of female reproductive system [
1]. There are three major groups in ovarian cancers: epithelial, germ cell, and specialized stromal cell tumors in which epithelial ovarian cancers (EOCs) are the vast amount of ovarian cancers [
2]. Despite the fact that surgical techniques and adjunct therapies have been improved for years, little change has occurred in the survival rate of ovarian cancer since platinum-based therapy was spread over past 30 years [
3]. Manifesting chemical resistance, late detection and deficiency of targeted therapies for advanced ovarian cancers are considered to be the primary factor lead to poor prognosis [
3]. For these reasons, ovarian cancer has become one of the greatest clinical challenges [
4]. In order to develop new drug strategies or diagnostic biomarkers, a better appreciation of the molecular mechanism of ovarian cancer is needed [
5].
Since the association between aberrant DNA methylation patterns and malignancy was first found, its role in cancer development has been increasingly investigated [
6]. DNA methylation consists of the epigenetic mechanism connected to gene expression. DNA methylation is found on the cytosine residues of CG (CpG) dinucleotides [
4]. DNA methyltransferases (DNMTs) could catalyze the adjunction of a methyl group to the cytosine ring to form methyl cytosine in which
S-adenosylmethionine is recruited as a methyl donor [
4]. Both the hypermethylation inducing the silence of tumor suppressor genes and hypomethylation associated with genomic instability have been involved in cancer initiation and tumor progression [
7]. For example, aberrant methylation of CpG islands is a common epigenetic event occurred in EOC, which has the most variation type [
8]. Therefore, investigation on variation of tumor DNA methylation contributes to better understand the mechanism of cancer.
As a transcriptional factor,
FOXD3 is a member of the forkhead gene family and plays significant roles during development, cell maintenance and regulation of lineage specification [
9,
10]. What’s more, FOXD3 plays a significant role in tumor initiation and growth through other transcription factors like TWIST1 [
11]. Many studies have revealed the association between
FOXD3 and tumorigenesis. FOXD3 could inhibits non-small cell lung cancer growth [
12]. In addition, low expression of
FOXD3 contributes to poor prognosis in high-grade glioma patients [
13]. However, the function of
FOXD3 in ovarian cancer is still not explicit, which urges us to clarify its mechanism.
In this study, FOXD3 was demonstrated that the degree of methylation and expression in various ovarian cancer cells were changed compared to normal ovarian cells. Our results suggested that FOXD3 could affect tumor growth and aggressiveness in OC.
Methods
Bioinformatic analysis was based on GSE81224. The Chip Analysis Methylation Pipeline (ChAMP) package is a pipeline which not only integrates currently available 450k analysis methods but also offers its own novel functionality. Circular layout (cyclize package) is an efficient way to visualize huge amounts of genomic information.
Human tissue samples, cell lines
This study was approved by the institutional review board of The First Affiliated Hospital of University of South China and informed consent was obtained from all patients included in this study. Paired fresh OC tissues were collected from 25 patients who underwent OC resection without prior radiotherapy and chemotherapy in The First Affiliated Hospital of University of South China in 2018. These samples were snap-frozen in liquid nitrogen immediately after resection, and then stored at − 80 °C until needed. The SKOV3, OV90, HO8910 and HOSE cell lines were purchased from the BeNa Culture Collection (Shanghai, China). The OV90 and HO8910 cell lines were cultured in DMEM 1640 medium, SKOV3 and HOSE cultured in RPMI 1640 medium (Sigma-Aldrich Corp., St. Louis, MO, USA) containing 10% fetal bovine serum (Invitrogen) and incubated in a thermostat at 5% CO2, 37 °C.
Cell transfection
24 h before transfection, cells in logarithmic growth stage were digested with trypsin and resuspended with complete culture medium. Cell suspension was prepared by blowing and mixing with straw. 1 × 106 cells were seeded in each of the 6-well, and then cultured in incubator at 37 °C and 5% CO2 for 18 to 24 h, until cells reached 50–60% of the coverage rate. 3 h before transfection, the original medium was removed and replaced with a fresh basic medium without serum and antibiotics. Using liposome Lipofectamine 2000 (Life Technologies, USA) according to the kit instruction for transfection, and cultured at 37 °C and 5% CO2 conditions for 48 h.
Methylation-specific PCR
DNA extracted from tissue samples and cell lines was subjected to bisulfite modification to convert all unmethylated cytosines into uracils, leaving methylated cytosines unmodified. The bisulfite modification was carried out by using the CpGenome™ DNA modification kit (Chemicon International, Temecula, CA). MSP was performed using AmpliTaq Gold with primers specific for methylated and unmethylated sequences of the genes. MSP primers for each gene were listed in Table
1. The treated DNA was used immediately or stored at − 20 °C until use. The bisulfite modified DNA was subjected to PCR. Positive control methylated DNA samples for each gene examined was used. The conditions of ampify the bisulphite converted DNA by MSP primers was 95 °C 3 min (95 °C 10 s, 60 °C 30 s, 72 °C 20 s) 40 cycle, 72 °C 7 min, 4 °C ∞. Water blank was used as a negative control. PCR products were analyzed on 2.5% agarose gel and visualized under UV illumination.
Table 1
Sequences of MSP primers for qRT-PCR
MSP primers |
FOXD3 forward | 5′ GGTAGCGTTAGCGATATGTTC 3′ |
FOXD3 reverse | 5′ ACGTCGCTATCCTTCTCTTC 3′ |
Unmethylated primers |
FOXD3 forward | 5′ GGTAGTGTTAGTGATATGTTT 3′ |
FOXD3 reverse | 5′ ACATCACTATCCTTCTCTTC 3′ |
In vitro epigenetic drug treatment
The human ovarian cancer cell line SKOV3 and OV90 were treated with a culture medium containing a demethylating agent, 5-aza-2′-deoxycytidine (5-Aza-dC; Sigma-Aldrich, USA) dissolved with acetic acid in 1:1 ratio and subsequently added into the medium. And equivalent acetic acid was added into the negative control groups’ medium. For the 5-Aza-dC treatment, 5-Aza-dC (10 μM) was replenished daily for 72 h. The culture medium was altered every day. After harvesting the cells, RNA was extracted as described below for reverse transcription Quantitative real-time PCR (qRT-PCR), and proteins were extracted for the Western blotting analysis.
Quantitative reverse transcription polymerase chain reaction (qRT-PCR)
The total RNA from cells and tissue was extracted with the mirVana miRNA isolation kit (Ambion, USA) according to the manufacturer’s protocol. The cDNA was synthesized from total RNA using PrimeScript™ RT Reagent Kit with miRNAs specific RT primers (Applied Biosystems, Waltham, MA) (Takara, Dalian, China) in a total reaction volume of 10 μL in TPersonal Thermocycler (Biometra, Göttingen, Germany) by following the manufacturer’s instructions. Then, miRNA cDNA was quantified using SYBR Premix Ex Taq kit (Takara, Dalian, China) in a 20-μL reaction system (Applied Biosystems, Foster city, CA). The relative expression levels were evaluated by using the by 2
(−∆∆Ct) method. Primers for each gene were listed in Table
2.
Table 2
Sequences of primers for qRT-PCR
FOXD3 forward | 5′-GACGACGGGCTGGAAGAGAA-3′ |
FOXD3 reverse | 5′-GCCTCCTTGGGCAATGTCA-3′ |
β-Actin forward | 5′-GGACTTCGAGCAAGAGATGG-3′ |
β-Actin reverse | 5′-AGCACTGTGTTGGCGTACAG-3′ |
Western blotting
Western blotting was performed in tissue samples or cultured cells as indicated. The cells were lysed in buffer containing 1% NP40, 50 mM Tris, 5 mM EDTA, 1% sodium deoxycholate, 1% SDS, 1% Triton X-100, 10 mg/mL aprotinin, 1 mM PMSF, and 1 mg/mL leupeptin (pH = 7.5), supernatants were collected after spin and protein was measured by Bradford assay (Thermo, Waltham, MA, USA). Forty micrograms total proteins were resolved on SDS-PAGE. Following an electric transfer onto PVDF membranes, the blots with proteins were then blocked by 5% bovine serum albumin and incubated with appropriate primary antibodies at 4 °C overnight. The membranes were incubated by HRP conjugated secondary antibody, and signals were visualized by an enhanced ECL-based imaging system. The caspase 3 detected was active caspase 3. Antibodies used in the study include anti-FOXD3 antibody (ab178512, Abcam, USA), anti-beta Actin antibody (ab8227, Abcam), anti-Annexin A2 antibody (ab41803, Abcam), anti-Cleaved PARP1 antibody (ab4830, Abcam), anti-Casepase-3 antibody (ab13847, Abcam), and anti-GAPDH antibody (ab9484, Abcam). The graphs shown the representative images from three independent experiments. The results were compared with normal human ovarian cell line HOSE.
MTT assay
Cells were prepared into single cell suspensions with culture medium containing 10% fetal bovine serum. After 48 h of incubation at 37 °C and 5% CO2 in the constant temperature incubator, 10 μL MTT solution was added (5 mg/mL PBS, pH = 7.4) into each well. After incubate for 4 h, careful remove the culture supernatant. 100 μL medium was added into each hole and oscillate for 10 min to dissolve the crystal. 450 nm was selected to measure the light absorption values for each hole at the enzyme-linked immunoassay. The results were recorded and repeated three times.
Wound healing test
For wound healing test, the cells were plated in 6-well plates. The adherent cells were wounded by a 10 μL plastic pipette tip and then the scathing cells were rinsed with PBS and cultured with serum-free DMEM for another 24 h. There were four groups, including control group, p-vector group, p-FOXD3 group, and 5-Aza-dC group, where p-vector and p-FOXD3 represented empty plasmid vector group group and FOXD3 overexpression group respectively in the whole article. The wound closure in different groups was photographed and evaluated with the microscope. Finally, the wound healing area after 48 h was analyzed and statistically analyzed. The results were repeated three times.
SKOV3 and OV90 cells were re-suspended and seeded in 12-well plates at a density of 2000 cells/well, incubated for 2 weeks, and then stained with 0.5% crystal violet for 30 min. Excess dye was rinsed off twice with phosphate-buffered saline (PBS). Images were obtained using the computer software Quantity One® from Bio-Rad Laboratories, Inc.
Flow cytometric assays for apoptosis and cell cycle
For cell cycle, cells were digested by trypsin/EDTA (Gibco) and washed twice with ice-cold PBS. After fixed with 70% ethanol overnight at 4 °C, the cells were washed twice with PBS and then digested by 50 mg/mL RNase in 500 mL of PBS at 37 °C for 30 min. Next, the cells were stained with 20 mg/mL propidium iodide (PI) for 30 min at 37 °C. Finally, FACS Calibur flow analyzer (BD, USA) was used to detect the cell cycle, and FACS Diva (BD, USA) software was used to analyze the data. The experiment was repeated three times.
For apoptosis, endogenous apoptosis of the cells was used to monitor the changes. Cells were harvested using trypsin/EDTA and washed with PBS, and subsequently binding buffer was added to re-suspend the cells. Following incubation with Annexin-V and PI staining according to the manufacturer’s protocol (Bio-Vision), FACS Calibur flow cytometry was used to detect apoptosis, and FACS Diva software was used to analyze data. The experiment was repeated three times.
Xenograft experiments
6-week-old female BALB/c normal mice and nude mice were obtained from the Animal Center of the Chinese Academy of Medical Sciences (Beijing, China). For tumor-initiating assays, 6 BALB/c nude mice were used every sample. Tumor-initiating cell frequencies were calculated using extreme limiting dilution analysis. Tumor cells (1 × 10
6) were injected into 6-week-old BALB/c nude mice. All mice were raised for 4 weeks. The mice of 5-Aza-dC group were subcutaneously injected to the posterior flank with 1 mg/kg day 5-Aza-dC for the last 4 weeks every day [
14]. From the 1st week, every 5 days, tumor volume and tumor weight were calculated. All protocols for animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of The First Affiliated Hospital of University of South China.
Statistical analysis
SPSS 19.0 statistical software (Chicago, IL, USA) was used for statistical analysis. Quantitative data were expressed as mean ± standard deviation. A comparison of groups was analyzed by single factor ANOVA and qualitative data were expressed as the number of cases or percentage (%). A comparison of the groups was made using the χ2 test. Statistical significance was set at p < 0.05.
Discussion
In this study, FOXD3 was found hypermethylated in OC tissues compared to normal human ovarian tissues through the analysis of Illumina450 genome-wide methylation data. To further verify the FOXD3 methylation degree, RT-PCR was performed and suggested that FOXD3 was hypermethylated and its expression was downregulated in human OC.
DNA methylation is a critical mechanism of gene silence and control of gene expression, which is associated with diverse regulation of cellular process [
15]. As a transcriptionally epigenetic modification, abnormal DNA methylation including DNA hypermethylation in CpG islands of promoter regions frequently occur in OC and differ between histological subtypes [
16,
17]. It is known that there is a close connection between aberrant DNA methylation and transcriptional inhibition. While hypermethylation in CpG islands of promoters could decrease gene expression, hypomethylation increases gene expression [
18]. Thus, the mRNA levels of
FOXD3 detected by RT-PCR could confirm the correlation between
FOXD3 and OC. By figuring out the epigenetic marks that potentially mediated the genetic risk of OC, we could not only better understand the pathogenesis of OC but also enhance the increasing library of risk-related biomarkers for OC [
17].
FOXD3 which primarily identified in embryonic stem cells is a member of the FOX transcription factor family [
19]. It is reported that FOXD3 serves as a tumor suppressor in vast types of cancer. Yan et al. found that FOXD3 could suppress tumor growth in non-small cell lung cancer and Li et al. also revealed that FOXD3 could suppress colon cancer [
20,
21]. In the present study, we intended to figure out the underling mechanism of
FOXD3 in human OC. Generally, FOXD3 was shown to regulate the downstream microRNA [
22]. Combined with western blot, the level of FOXD3 was found to be downregulated in OC cell lines, which indicated that hypermethylation
FOXD3 could associate with a low expression level. Hence the
FOXD3 could serve as a critical factor in human OC.
High levels of promoter DNA methylation that lead to the silence of tumor suppressor genes are resulted from hyperactive DNMTs [
23]. DNMTs have been reported to be highly expressed in various types of tumors [
24,
25]. Notably, 5-Aza-dC is a DNMT inhibitor which causes a covalent entrapment of DNMT1 to decitabine-substituted DNA and loss of maintenance DNMT activity, subsequently reducing DNA methylation [
23]. 5-Aza-dC has also been shown to enjoy the anticancer efficacy on various cancers including ovarian cancer [
26,
27]. Furthermore, in order to clarify the pathological role of
FOXD3, we took 5-Aza-dC as epigenetic therapy on OC cell lines and
FOXD3 overexpression groups used as validation groups. 5-Aza-dC group and
FOXD3 overexpression groups suppressed cell migration and colony formation ability and thus induced apoptotic cell death compared with control group, which suggested that
FOXD3 could inhibit cell proliferation, migration and promote cell apoptosis in OC cell lines. Next, we further verified the pathological role of
FOXD3 in vivo. Similarly, 5-Aza-dC was used as an epigenetic therapy and
FOXD3 overexpression groups used as validation groups, the xenograft experiments revealed an obvious decrease of tumor weight and tumor volume in 5-Aza-dC group and
FOXD3 overexpression groups. These results showed that the restoration of expression
FOXD3 could suppress cancer cells proliferation in vitro as well as in vivo.
Conclusions
In summary, we firstly found that the hypermethylation of FOXD3 in OC by bioinformatic analysis. Next, we found that the reversed expression of FOXD3 in OC could suppress tumor growth. However, in current study there are fewer detected patients in total, so we need to increase the number of samples to confirm our results. In conclusion this study suggested that the FOXD3 could act as a potential therapeutic target in OC diagnosis and treatment.
Authors’ contributions
Substantial contribution to the conception and design of the work: GL, CC and JW; Analysis and interpretation of the data: HY, YT and PY; Drafting the manuscript: GL; Revising the work critically: YL and YL; Final approval of the work: All authors. All authors read and approved the final manuscript.