Introduction
Epithelial ovarian cancer is the fourth most lethal cancer among women and the leading cause of gynecological cancer deaths worldwide. The five-year survival rate for epithelial ovarian cancer patients is only 44% in the USA [
], because most of these patients are diagnosed at an advanced stage with widely omental metastases [
,
]. However, the molecular mechanisms of metastasis remain ambiguous. A better understanding of these mechanisms is especially needed for the early diagnosis and enhanced treatment of ovarian cancer.
MicroRNAs (miRNAs) are a class of evolutionarily conserved small non-coding RNAs which negatively modulate gene expression by pairing with the 3′-UTR of target mRNAs to direct their posttranslational repression [
]. miRNAs play vital roles in numerous biological processes, such as the regulation of cell cycle, differentiation, apoptosis, and angiogenesis, as well as cancer initiation and progression [
–
]. Many studies have shown that miRNAs are highly dysfunctional and deregulated in various tumors, either as an onco-miRNA or as a suppressor miRNA. miR-9 acts as a metastasis-associated miRNA to promote many cancers to acquire malignant phenotypes, resulting in cancer progression and poor prognosis. These cancers include breast cancer [
], esophageal squamous cell carcinoma [
], hepatocellular cancer [
], colorectal cancer [
], and melanoma [
]. Other studies have shown that miR-9 was downregulated in endometriod and clear cell ovarian cancer [
] and recurrent ovarian serous cancer compared with the primary tumor [
]. We have previously demonstrated that upregulated miR-9 was associated with improved prognosis, longer survival rates, and cisplatin sensitivity [
]. These results demonstrated that the role of miR-9 remains unclear and needs further verification.
Epithelial‒mesenchymal transition (EMT) is a multi-faced transdifferentiation program that enables tumor cells acquire malignancy-associated phenotypes and important to tumor aggression and metastasis [
]. Recently, numerous studies have documented that miRNAs may be involved in the EMT process.
In our present study, we investigated the function of miR-9 in primary ovarian cancer tissue compared with paired metastatic ones. miR-9 was upregulated in metastatic sites and had a reverse correlation with E-cadherin expression in vitro and patient samples. Therefore, we hypothesized that miR-9 may be involved in the metastasis of ovarian cancer via the regulation of E-cadherin expression.
Materials and methods
Tissue samples
A total of 25 paired formalin-fixed and paraffin-embedded ovarian serous tumor samples were collected at Tongji Hospital (Wuhan, Hubei, China). The participants provided informed consent between January 2015 and July 2015. All patients underwent debulking and subsequent platinum-centered chemotherapy. The protocol was approved by the Ethics Committee of Tongji Hospital.
Cell culture
Human ovarian cancer cell lines SKOV3 and A2780 were purchased from the American Type Culture Collection (Manassas, VA, USA). SKOV3 and A2780 cells were cultured in Macoy’5A medium or RPMI 1640 medium, respectively (Gibco, USA), supplemented with 10% fetal bovine serum (Gibco, USA), penicillin (100 units/ml), and streptomycin (100 µg/ml) at 37 °C in a humidified atmosphere containing 5% CO2.
Reagents and miRNA transfections
miR-9 mimics or inhibitor was purchased from Ribobio Co. Ltd, Guangzhou, China. miR-9 mimics (50 nmol/L) or miR-9 inhibitor (100 nmol/L) was transfected into cells with Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions.
Plasmid construction and luciferase assay
The full-length 3′-UTR of E-cadherin from the genomic DNA of a normal patient was amplified using PCR, and the potential miR-9 binding site in the 3′-UTR of E-cadherin was mutated by the overlap extension PCR. Wild-type and mutant 3′-UTRs of E-cadherin were ligated into the Psi-Check2 plasmid (Promega) at the
XhoI and
NotI sites directly downstream from the renilla luciferase coding sequence. The authenticity and orientation of the inserts were confirmed by sequencing as previously described [
]. SKOV3 cells were seeded at 1.0 × 10
4/well in a 96-well plate at 24 h before transfection. Cells were co-transfected with miR-9 mimics or NC and wild-type or mutant E-cadherin 3′-UTR plasmids using Lipofectamine
TM 3000 (Invitrogen). After 48 h, luciferase assay was performed using the Dual-Luciferase Reporter assay system (Promega).
miRNA target prediction
Algorithms, such as Pictar (http://pictar.mdc-berlin.de/), Targetscan (http://www.targetscan.org/), miRanda (http://www.microrna.org/microrna/home.do), RNA22 (http://cbcsrv.watson.ibm.com/rna22.html), and miRWalk (http://www.umm.uni-heidelberg.de/apps/zmf/mirwalk/), were used to identify the target genes to predict miRNA potential targets [
,
].
RNA and miRNA extraction
Total RNAs including miRNAs were extracted from primary ovarian cancer tissue or formalin-fixed and paraffin-embedded tumor samples using miRNeasy FFPE kit (Qiagen Inc., Valencia, CA, USA) following manufacturer’s instructions. The levels of mature miR-9 were determined by Bulge-LoopTM miRNA qRT-PCR Primer Set (Ribobio) with SYBR Green quantitative real-time PCR. The miR-9 levels were normalized to those of U6 snRNA. Gene mRNA was extracted using RNAprep pure cell kit (TIANGEN Biotech CO., LTD, China), and the expression of mRNA was determined by qRT-PCR using the PrimeScript RT reagent kit and SYBR Premix EX Taq (TaKaRa) according to the manufacturer’s instructions. The expression of gene mRNA was normalized to that of b-actin by 2-DDCt method.
Wound healing assay
SKOV3 cells were first transfected with miR-9 mimics or inhibitor for 12 h, and then wound healing assay was performed according to the standard protocol [
].
Migration and invasion assay
Migration assay was performed using a migration chamber (Corning, New York, NY, USA) following the manufacturer’s instructions. Matrigel invasion assay was performed using membranes coated with Matrigel matrix (BD Science, Sparks, MD, USA). First, the cells were transfected with miR-9 mimics or inhibitor. After 12 h, the transfected cells were digested, and single cell suspensions (1×10
5 cells/well) were seeded into the upper chambers and allowed to invade for another 12 h at 37 °C in a CO
2 incubator. Then, cells on the lower side of the chamber were fixed, stained with 0.1% crystal violet for 15 min and counted using a light microscope according to the published criteria [
].
Immunofluorescence (IF) assay
After cells were transfected with miR-9 mimics for 48 h, cells were fixed with paraformaldehyde 4% and incubated with the primary antibodies E-cadherin (1:50, Cell Signaling Technology, Danvers, MA, USA), N-cadherin (1:50, Cell Signaling Technology, Danvers, MA, USA) and vimentin (1:100, Cell Signaling Technology, Danvers, MA, USA) overnight at 4 °C. The next day, cells were first washed with PBS and incubated with secondary antibody conjugated to FITC or CY3 (Invitrogen, Carlsbad, CA, USA). The nuclei were stained with DAPI. Images were acquired using a laser scanning confocal microscope (Olympus).
Western blot
SKOV3 or A2780 cells were seeded in a 6-cm culture plate. After cell density of approximately 30%–50% was reached, miR-9 mimics or inhibitor was transfected and continued to be cultured for another 48 h. Then, the cells were collected and lysed by radioimmunoprecipitation assay buffer. The primer antibody was E-cadherin (1:1000, item number 3195, Cell Signaling Technology, Danvers, MA, USA). The primer antibody was GAPDH (1:5000, item number 60004, Proteintech Group, Inc., Wuhan, China). The subsequent procedure was performed as described previously [
]. The chemiluminescence was detected by a Bio-Rad Imaging system (Bio-Rad, Hercules, CA, USA).
Statistical analysis
All data are expressed as mean±SD. Each experiment was repeated at least thrice independently. Statistical significance of differences was analyzed by two-tailed Student’s t-test or one-way ANOVA. P value less than 0.05 was considered statistically significant. All statistical analyses were performed using SPSS 17.0 (SPSS Inc., Chicago, IL, USA).
Results
miR-9 is upregulated in metastatic ovarian cancer tissue
The expression of miR-9 at the metastatic sites in ovarian cancer patients was compared with the paired primary site tissue. A total of 25 paired formalin-fixed and paraffin-embedded ovarian serous tumor samples were collected to quantify the expression of miR-9 using qRT-PCR. The average expression of miR-9 was significantly higher at the metastatic sites than at their paired primary sites (P<0.05) (Fig. 1A). Furthermore, five paired fresh patient-derived ovarian cancer cell lines diagnosed with high-grade serous ovarian cancer were also detected by qRT-PCR. The results showed miR-9 expression of the metastatic sites was upregulated by at least 6-fold (Fig. 1B).
miR-9 level may be negatively correlated with the E-cadherin expression but positively correlated with vimentin expression
The different expression levels of E-cadherin and vimentin between ovarian cancer metastatic site tissue and paired primary tissue was confirmed. qRT-PCR was performed, and the results showed that the E-cadherin expression was generally downregulated at the metastatic sites than at the primary sites (Fig. 1C). By contrast, vimentin expression was remarkably upregulated at the metastatic sites (Fig. 1D), in accordance with miR-9 expression between the metastatic and primary sites. These results indicated that miR-9 was upregulated and may have a negative correlation with E-cadherin.
E-cadherin is a direct target of miR-9
Bioinformatics analysis tools were used to search for the potential target of miR-9. The results show that E-cadherin was a putative candidate. SKOV3 and A2780 cells were used to detect E-cadherin expression after miR-9 mimics or inhibitor transfection to test whether E-cadherin was regulated by miR-9. SKOV3 and A2780 cells were transfected with miR-9 mimics for 36 h. qRT-PCR results showed that miR-9 expression was significantly upregulated compared with transfection with miR-9 NC (Fig. 2A). Meanwhile, the E-cadherin expression was found to be considerably downregulated when miR-9 mimics was transiently transfected into SKOV3 and A2780 cells after 36 h of detection through qRT-PCR and 48 h through Western blot separately (Fig. 2B and 2C). By contrast, miR-9 expression was significantly downregulated when transfected with miR-9 inhibitor (data not shown), and the above effects were also restored as shown by qRT-PCR and Western blot results (Fig. 2D and 2E). Next, a luciferase reporter assay was performed to confirm whether E-cadherin was a direct target of miR-9. Wild-type or mutant-type E-cadherin 3′-UTR sequence was cloned into the psi-check 2 vector. SKOV3 cells were co-transfected with miR-9 mimics and the psi-check 2 plasmid (wild-type or mutant) for 48 h. The dual-luciferase reporter assay showed that the luciferase activity was significantly reduced in the wild-type E-cadherin 3′-UTR but not the mutant type (Fig. 2F and 2G). These results demonstrated that E-cadherin was directly targeted by miR-9.
miR-9 promotes ovarian cancer cell migration and invasion
SKOV3 and A2780 cells were transfected with miR-9 mimics for 24 h to elucidate the function of miR-9 in cancer cell metastasis. The cells became scattered and displayed a spindle-like or fibroblast morphology, but these effects were not observed when the cells were transfected with miR-9 inhibitor (Fig. 3A and 3B). These phenomena implied that miR-9 can promote cell motility. Wound healing assay showed that the ability of migration in SKOV3 cells was enhanced after miR-9 mimic transfection for 24 h (Fig. 3C). Then, we used Transwell assay to evaluate the migration and invasion capacities of the SKOV3 cells. Fig. 3D shows that miR-9 mimics could significantly enhance the migration and invasion of ovarian cancer cells. By contrast, the migration and invasion ability of SKOV3 cells were decreased when transfected with miR-9 inhibitor. Migration and invasion assay was also conducted in A2780 cells (data not shown). These results revealed that miR-9 may promote ovarian cancer cell motility.
miR-9 regulates EMT-related genes to promote tumor cell metastasis
SKOV3 and A2780 cells were transfected with miR-9 mimics for 36 h to further confirm whether miR-9 could modulate the EMT-related genes to enhance ovarian cancer metastasis. Then, qRT-PCR showed that E-cadherin expression was downregulated, while mesenchymal markers, such as N-cadherin, vimentin, b-catenin, and MMP-9, were inversely upregulated (Fig. 4A and 4B). Furthermore, IF was detected to explore the molecular mechanism and subcellular localization of miR-9-E-cadherin axis in promoting metastasis. IF results indicated that the intensity of E-cadherin was reduced while the intensity of N-cadherin or vimentin was increased after miR-9 mimics were transfected in A2780 or SKOV3 cells for 48 h, respectively (Fig. 4C and 4D). EMT is a vital molecular change in tumor mobility and metastasis [
,
]. These results demonstrated that miR-9 may regulate EMT-related genes to promote ovarian cancer cell metastasis.
Discussion
Epithelial ovarian cancer is a lethal and highly metastatic disease with the highest mortality and morbidity in all gynecologic cancers in the developed world with approximately 14 030 deaths from the disease in 2013 [
]. The majority of epithelial ovarian cancer patients will eventually relapse at metastatic sites. EMT plays a key role in the metastasis of tumor cells, and this process possesses a multi-faceted transdifferentiation program that enables tumor cells to acquire malignancy-associated phenotypes [
]. Numerous studies have indicated that miRNAs, which are important gene regulators acting as oncomiRs or anti-oncomiRs, are increasingly involved in regulating the malignant tumor progression. EMT was also found to be regulated by metastamiRs [
], which exert important roles in various processes of metastasis. In this study, miR-9 was identified to be a metastamiR in ovarian serous cancer. miR-9 is selectively expressed in neural tissue under normal conditions and exhibits a regulator and prodifferentiation function [
]. miR-9 was first found to be upregulated in brain tumors than in tumors of other histological types [
]. The enhancement of metastasis of miR-9 overexpression has also been found in hepatocellular carcinoma [
], head and neck squamous cell carcinoma [
], colon cancer [
], esophageal squamous cell carcinoma [
], breast cancer [
,
], and cervical cancer [
]. This result suggests the role of a promoter or onco-miR in tumor development and progression. However, miR-9 is also downregulated in gastric cancer [
], pancreatic cancer [
], and ovarian cancer [
], indicating that a role of anti-miR in tumor progression. Interestingly, Ma
et al.reported that miR-9 acted as a metastamiR to promote breast cancer metastasis [
], while Lwhmann
et al. noted that miR-9 was also transcriptionally downregulated in early breast cancer development [
]. Moreover, in gastric tumor, Inoue
et al. reported that miR-9 was upregulated [
], while Zheng
et al. indicated that miR-9 functioned as a suppresser miRNA and was downregulated in gastric tumor progression [
]. The heterogeneity of these tumors or the limited specimens may have caused the highly diverse miR-9 expression even in similar tumors. Whether miR-9 functions as a promoter or suppressor in the progression of various tumors remains obscure. A previous study has demonstrated that TGF-β upregulation of miR-182 expression may promote gallbladder cancer metastasis by targeting CADM1 [
]. Another study revealed that miR-182 may potentiate TGF-b-induced EMT and metastasis of cancer cells by directly targeting SMAD7 [
]. TGF-bsignaling pathway plays vital roles in EMT and metastasis of cancer cells. However, the detailed mechanism of miR-9 in promoting the metastasis of the ovarian cancer cell should be further illuminated.
In this study, we demonstrated that the average level of miR-9 was significantly upregulated in 25 metastatic sites of ovarian cancer patients compared with paired primary sites. Furthermore, we also confirmed that miR-9 expression was negatively associated with E-cadherin expression but positively correlated with vimentin expression. Functional study indicated that miR-9 overexpression could increase ovarian cancer cell migration and invasion, which were confirmed by wound healing and Transwell assays. Similar to other studies which used informatics analysis tools, E-cadherin was predicted to be directly targeted by miR-9, which was further confirmed through luciferase reporter assay. qRT-PCR and Western blot experiment showed that E-cadherin expression was downregulated when cells were transfected with miR-9 mimics but was upregulated when endogenous miR-9 was transiently inhibited. IF results indicated that the intensity of E-cadherin was reduced in miR-9 overexpressed ovarian cancer cells, contrary to the enhanced intensities of N-cadherin and vimentin. These results confirmed that miR-9 was involved in the regulation of ovarian cancer cell metastasis by directly targeting E-cadherin.
Thus, a high level of miR-9 was found at the ovarian cancer metastatic sites compared with their paired primary sites. We also demonstrated that miR-9 could promote metastasis in ovarian serous cancer by directly targeting E-cadherin 3′-UTR. Manipulation of miR-9 may regulate the migration and invasion of ovarian cancer cells. Thus, miR-9 may be a potential target for the prediction and treatment of serous ovarian cancer.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(363KB)
