E2F-1 targets miR-519d to regulate the expression of the ras homolog gene family member C

E2F1 (E2F transcription factor 1) can act as a tumor suppressor or oncogene. We report the molecular mechanism of E2F1 in ovarian carcinoma tumorigenesis and progression. E2F1 expression levels in ovarian carcinoma tissue were examined by immunohistochemistry. After E2F1 plasmid transfection and E2F1-microRNA-519d (miR-519d)/si-RhoC (Ras homolog gene family member C) co-transfection, ovarian cancer cell phenotypes and the related molecules were examined in vitro and in vivo. E2F1 was overexpressed in type I and type II ovarian carcinoma as compared to normal ovary tissues and normal fallopian tube tissues, respectively. E2F1 overexpression promoted cell proliferation, G1–S progression, survival, migration, and invasion in vitro; miR-519d or siRhoC co-transfection reversed E2F1 oncogenic effects. E2F1 overexpression promoted tumor growth in vivo; miR-519d overexpression inhibited it. E2F1 overexpression increased RhoC, Bcl-2, cyclin D1, survivin, MMP2 (matrix metalloproteinase 2), MMP9, STAT3 (signal transducer and activator of transcription 3), and HuR (ELAV-like RNA-binding protein 1) expression; miR-519d overexpression decreased their expression. E2F1 downregulated miR-519d directly and miR-519d downregulated RhoC directly. Conversely, miR-519d directly downregulated E2F1, There is a direct repressive regulatory loop between E2F1 and miR-519d. We provide evidence that E2F1/miR-519d/RhoC is a promising signaling pathway for diagnosing and treating ovarian carcinoma.


INTRODUCTION
Epithelial ovarian carcinoma is the most lethal gynecological malignancy, and has consistently low 5-year relative survival rates [1][2][3][4][5].Difficult early diagnosis and frequent recurrence and metastasis lead to the high mortality rate of epithelial ovarian carcinoma.Exploring the underlying molecular mechanisms of epithelial ovarian carcinoma is essential for developing effective diagnostic and treatment strategies.

The correlation between E2F1 expression and ovarian carcinoma pathogenesis and aggressiveness
Immunohistochemical analysis of the clinical relevance of E2F1 expression in ovarian carcinoma showed that E2F1 expression was statistically significantly elevated in type I tumors (Table 1) compared to normal ovarian tissues (Figure 1A, 1D) and was positively associated with International Federation of Gynecology and Obstetrics (FIGO) stage I-II and III-IV disease.E2F1 expression was also significantly elevated in type II tumors (Table 2) compared to normal fallopian tube tissues (Figure 1B, 1E).The findings suggest that E2F1 has potent oncogenic effects in ovarian carcinoma (Figure 1C, 1F) of different type and origin.
In the nude mouse xenograft model, mice injected with cells stably transfected E2F1 plasmid had a dramatic induction of tumor size (Figure 3K, P < 0.05) and faster tumor growth than the control mice (Figure 3L, 3M, P < 0.05), suggesting that E2F1 promotes tumorigenesis and growth.
The above findings reveal that a direct repressive regulatory loop exists between E2F1 and miR-519d.
We generated a luciferase reporter gene containing the human RhoC mRNA 3′ UTR predicted to have potential miR-519d binding sites.Co-transfecting HEK293 cells with RhoC-luciferase reporter plasmid plus miR-519d mimic significantly reduced luciferase activity (Figure 5D, P < 0.05).Furthermore, luciferase activity was fully www.impactjournals.com/oncotargetrestored in the mutant and control groups, demonstrating that RhoC mRNA is a direct target of miR-519d.
The nude mouse xenograft model showed that mice injected with cells stably expressing miR-519d had obviously reduced tumor size (Figure 5E, P < 0.05) and tumor growth than the control mice (Figure 5F, 5G, P < 0.05).Immunohistochemistry showed that E2F1 and RhoC expression were downregulated in the tumor xenograft tissues of the miR-519d group as compared to the control (Figure 5H).
The immunohistochemistry assay showed that E2F1 and RhoC expression were upregulated in the tumor xenograft tissue from nude mice of the E2F1 group when compared to the control group (Figure 6E, P < 0.05).The findings support the premise that E2F1 promotes ovarian carcinogenesis and growth by upregulating the expression of RhoC and Bcl-2, cyclin D1, survivin, MMP2, MMP9, STAT3, and HuR.
MiR-519d triggered cellular senescence by repressing HuR expression.The senescence detection kit showed that silencing E2F1 increased SA-β-gal (senescence-associate β-galactosidase) activity and triggered senescence as compared to the control groups (Figure 6F, P < 0.05).The data indicate that silencing E2F1 directly targeted miR-519d to downregulate HuR, inducing cellular senescence.

Effects of E2F1 and miR-519d co-transfection on ovarian carcinoma cells
A2780 and OVCAR3 cells were co-transfected with E2F1 plasmid and miR-519d mimic or with E2F1 siRNA and miR-519d inhibitor.The MTT, cell cycle, apoptosis, wound healing, and cell invasion assays showed that E2F1 plasmid and miR-519d mimic co-transfection reversed the effect of E2F1 alone, causing severe growth retardation (Figure 7A, P < 0.05), high levels of apoptosis (Figure 7C, P < 0.05), and reduced migration (Figure 7D, P < 0.05) and invasive ability (Figure 7E, P < 0.05) as compared to the control and mock cells, but had the same effect as E2F1 in promoting G1-S progression (Figure 7B, P < 0.05).E2F1 siRNA and miR-519d inhibitor co-transfection partly counteracted the effects of silencing E2F1 on cell proliferation, the cell cycle, apoptosis, migration, and invasion, although the cells still exhibited slower growth (Figure 7F, P < 0.05), G1 arrest (Figure 7G, P < 0.05), increased apoptosis (Figure 7H, P < 0.05), decreased migration (Figure 7I, P < 0.05) and invasive ability (Figure 7J, P < 0.05) as compared with the control and mock cells.
These results demonstrate that miR-519d plays a central role in E2F1 promotion of ovarian carcinoma cell proliferation, apoptosis, and cell migration and invasion, but not on the cell cycle.

Effects of E2F1 and RhoC co-transfection on ovarian carcinoma cells
A2780 and OVCAR3 cells were co-transfected with E2F1 plasmid and RhoC siRNA or with E2F1 siRNA and RhoC plasmid.Cell functional assays were performed to analyze the effects on A2780 and OVCAR3 cell proliferation, cycle, apoptosis, migration, and invasion.E2F1 plasmid and RhoC siRNA cotransfection reversed the effect of E2F1 alone, causing severe growth retardation (Figure 8A, P < 0.05), higher levels of apoptosis (Figure 8C, P < 0.05), and reduced cell migration (Figure 8D, P < 0.05) and invasive ability (Figure 8E, P < 0.05), but had the same effect in promoting G1-S progression (Figure 8B, P < 0.05) when compared with the control and mock cells.E2F1 siRNA and RhoC plasmid co-transfection partly counteracted the effects of E2F1 silencing on cell proliferation, the cell cycle, apoptosis, and cell migration and invasion, although the cells still exhibited slower growth (Figure 8F, P < 0.05), G1 arrest (Figure 8G, P < 0.05), increased apoptosis (Figure 8H, P < 0.05), and decreased migration (Figure 8I, P < 0.05) and invasive ability (Figure 8J, P < 0.05) when compared with the control and mock cells.
These results demonstrate that RhoC plays a central role in the oncogenic regulation by E2F1/miR-519d on ovarian carcinoma cell proliferation, apoptosis, migration, and invasion, but not on the cell cycle.

DISCUSSION
According to the most recent National Comprehensive Cancer Network (NCCN) guidelines, epithelial ovarian cancer is grouped into type I ovarian cancer (mostly originating from the ovary) and type II ovarian cancer (mostly originating from the oviduct) [31].We compared E2F1 expression levels in 59 cases of type I tumors and 17 normal ovarian tissues, and E2F1 expression was significantly elevated in the ovarian carcinoma tissues   and was positively associated with FIGO stage I-II and III-IV disease.We also compared E2F1 expression levels in 96 cases of type II tumors and 23 fallopian tube tissues, and E2F1 expression was also significantly elevated in the ovarian carcinoma tissues.The finding is consistent with the study by Suh and colleagues [27].We found that E2F1 promoted ovarian carcinoma cell proliferation, G1-S progression, survival, migration, and invasion in vitro and    promoted ovarian carcinoma tumorigenesis and tumor growth in vivo.We also observed that E2F1 increased Bcl-2, cyclin D1, survivin, MMP2, and MMP9 expression in ovarian carcinoma, but did not alter p53 expression.These results are similar to the oncogenic role of E2F1 in small cell lung cancer [28] and clear cell renal cell carcinoma [29], but differ from the suppressive effect of E2F1 in gastric cancer [17][18][19] and Hodgkin lymphoma [25].Our findings suggest that E2F1 functions as an oncogene in ovarian carcinoma in part by upregulating Bcl-2, cyclin D1, survivin, MMP2, and MMP9 expression.
To explore the underlying molecular mechanism of the oncogenic effects of E2F1 on ovarian carcinoma, we focused on microRNAs (miRNAs).MiRNAs are small, non-coding, single-stranded, and evolutionarily conserved RNAs that regulate gene expression by binding directly to complementary sites in the 3ʹ UTR of target mRNAs; the process plays a crucial role in tumor development and progression [32][33].Previous studies have proved that transcription factors act together with miRNAs in regulating the target genes [34].
The dual luciferase reporter assay confirmed the prediction that the miR-519d promoter sequence contains an E2F1 binding site.E2F1 overexpression suppressed miR-519d mRNA expression, proving that E2F1 targets the miR-519d promoter directly to decrease its expression.MiR-519d belongs to the chromosome 19 miRNA cluster (C19MC), and acts as a tumor suppressor in breast cancer, chondrosarcoma, osteosarcoma, and hepatocellular carcinoma [35][36][37][38], and it is significantly downregulated in ovarian cancer cell lines and tissues [39].MiR-519d decreased ovarian cancer cell proliferation and sensitized ovarian cancer cells to cisplatin-induced cell death [40].We conclude that miR-519d has the same suppressive effects on ovarian carcinoma based on the significantly decreased tumor size and growth in the nude mouse xenograft model.MiR-519d suppresses trophoblast cell invasion and migration by downregulating MMP2 [41], suppresses breast cancer by inhibiting STAT3 expression [35], and inhibits tumor metastasis and MMP2 expression in chondrosarcoma and osteosarcoma [36][37].MiR-519d reduces cell proliferation and inhibits tumorigenesis largely by decreasing the levels of the RNA-binding protein HuR to trigger senescence in ovarian carcinoma cells [42][43][44].We found that miR-519d transfection decreased MMP2, MMP9, STAT3, and HuR expression in ovarian carcinoma cells; at the same time, E2F1 increased the expression of these genes.We found that silencing E2F1 triggered senescence in ovarian carcinoma cells, supporting the premise that E2F1 targets miR-519d directly, decreasing its expression.We cotransfected ovarian carcinoma cells with E2F1 plasmid and miR-519d mimic, and found that miR-519d greatly reversed the effect of E2F1, revealing that miR-519d plays a central role in the oncogenic effects of E2F1 on ovarian carcinogenesis and development.
We found that RhoC is a putative and direct target of miR-519d, which the dual luciferase reporter assay confirmed.We observed that miR-519d transfection decreased RhoC expression and that E2F1 overexpression increased RhoC expression in ovarian carcinoma cell lines; RhoC expression was decreased in miR-519d tumor xenograft tissue and was increased in E2F1 tumor xenograft tissue when compared with their respective controls.RhoC is a small G protein/guanosine triphosphatase closely involved in tumor invasion and metastasis [45][46][47], and could serve as a good biomarker of ovarian carcinoma differentiation and progression [48][49].We found that E2F1 plasmid and RhoC siRNA co-transfection greatly reversed the oncogenic role of E2F1 in ovarian carcinoma cell phenotype, revealing that RhoC plays a central role in the oncogenic regulation of E2F1/miR-519d on ovarian carcinogenesis and development.The above findings demonstrate that E2F1 promotes ovarian carcinoma by directly targeting miR-519d, suppressing it, which in turn targets RhoC expression, downregulating it.Conversely, miR-519d mimic decreased E2F1 expression.The prediction that miR-519d would target the E2F1 mRNA 3′ UTR was in agreement with the results of the dual luciferase reporter assay.Furthermore, E2F1 expression was lower in the miR-519d xenograft group, proving that miR-519d also targets E2F1 to downregulate its expression.These findings indicate an E2F1-miR-519d direct repressive regulatory loop.
In the E2F family of transcription factors, E2F2, E2F3, E2F4, and E2F8 expression is increased and may play oncogenic roles in ovarian carcinoma [26,30,[50][51], and exert the same effects as E2F1.We searched target prediction websites and found that in addition to E2F1, the miR-519d promoter also contains E2F4 binding sites.Taken together, we hypothesize that E2F4 could have the same oncogenic effects as E2F1 on ovarian carcinoma by suppressing miR-519d.However, further research is needed to confirm this.
In conclusion, our study identifies a novel molecular mechanism for E2F1, whereby it targets miR-519d to upregulate RhoC, Bcl-2, cyclin D1, survivin, MMP2, MMP9, STAT3 and HuR expression, promoting tumorigenesis and progression in ovarian carcinoma.The E2F1/miR-519d/RhoC signaling pathway is of significant importance and shows promise as a target for diagnosing and treating ovarian carcinoma.

Ovarian carcinoma specimens
According to the latest NCCN guidelines, ovarian cancer were grouped into type I ovarian cancer (most origin from ovaries) and type II ovarian cancer (most origin from oviducts).Consisting of type I tumors 59 cases and type II tumors 96 cases, 17 normal ovarian tissues cases and 23 fallopian tube tissues cases were collected from patients undergoing surgical resection at the Department of Gynecology of the First Affiliated Hospital of China Medical University (Shenyang, Liaoning, China) between 2003 and 2014.The tumor specimens were microscopically confirmed by pathologists.None of the patients had received preoperative chemotherapy or radiotherapy.Informed consent was obtained from all subjects, the study was approved by the China Medical University Ethics Committee, and all specimens were handled and made anonymous according to ethical and legal standards.The average age at surgery was 54.3 years (range, 18-76 years).Each ovarian carcinoma specimen was evaluated according to the 2014 FIGO staging system.The histological architecture of ovarian carcinoma was defined in terms of World Health Organization classification.

MTT assay
Ninety-six well plates were seeded with 3 × 103 cells per well.At designated time points following transfection (0 h, 24 h, 48 h, and 72 h), 20 μL of 5 mg/mL MTT (Sigma,USA) was added into every well and cells incubated at 37°C for 4 hours.The liquid was then removed and l50 μL of dimethylsulfoxide (DMSO, Sigma, USA) added to dissolve the precipitated formazan.Following shaking for 10 min, the optical density (OD) was measured at 490 nm on a microplate spectrophotometer (Bio-Tek Instruments, Winooski, USA).

Cell cycle analysis
More than 1 × 104 cells were trypsinized, collected, washed twice with phosphate buffered saline (PBS), and fixed in 70% ethanol at −20°C for at least 12 h.The cells were washed twice with PBS and incubated with 400 µL RNase (0.25 mg/mL) at 37°C for 1 h, then resuspended in 100 µL Propidium Iodide(PI, KeyGen, NanJing China) and incubated at 4°C in the dark for 30 min.The PI cell cycle profile was detected by flow cytometry.

Cell apoptosis assay
Cells were harvested at 1500 rpm for 5 min and washed twice with cold PBS.They were then resuspended in a mix of 50 µL binding buffer, 5 μL annexin V-FITC and 5 μL PI (KeyGen, China).Samples were gently vortexed and incubated in the dark for 15 min at room temperature.Two hundred microliters of binding buffer was added to each tube and samples were examined by flow cytometer within 1 h.

Wound healing assay
Cells were seeded into 6-well plates and cultured for approximately 24 h to reach 80% confluence in the presence of mitomycin C(10 ug/ml, Sigma, USA).The monolayer was then scratched with a 200 μL pipette tip to create a wound.The 6-well plates were washed twice with PBS, and cells cultured in FBS-free medium.Cells were transfected with the miRNA or siRNA (or controls) at the same time.Prior to the wound healing assay, the width of each wound was examined by microscope and photographed using Image J software (National Institutes of Health, Bethesda, USA) at 0 h, 24 h and 48 h after wounding.Cell migration was measured by subtracting the wound width at 24 h or 48 h from that at 0 h.

Cell invasion assay
Cell invasion assays were performed using Transwell chambers (24-well inserts; 8 µm-pore size; BD Bioscience, San Jose, CA, USA) pre-coated with 40 μL of Matrigel basement membrane matrix (Matrigel at a dilution of 1:10) incubated for 4 h at 37°C in 5% CO2.Two hundred microliters of serum-free medium containing 5 × 104 cells/well of transfected or nontransfected cells were placed in the upper chamber in the presence of mitomycin C (10 ug/ml, Sigma, USA), and the lower chambers filled with 600 μL of the same complete medium.The plates were incumanuscripting cells at the bottom of the upper chamber were washed twice with PBS.Cells on the underside of the filters were examined and counted using an Olympus fluorescence microscope (Tokyo, Japan).Cells were counted three times, and each experiment was repeated in triplicate.

Cell senescence
Cells were washed once with PBS, 1 ml of β-galactosidase staining fixative was added and fixed at www.impactjournals.com/oncotargetroom temperature for 15 minutes, Then cells were washed three times with PBS for 3 minutes each, after that 1 ml of working solution was added to each well, incubated overnight at 37°C, then observed cells under ordinary light microscope.

Real-time RT-PCR
Total RNA extraction from ovarian carcinoma cell lines and ovarian tissues was performed with 1 mL TRIzol reagent (Takara, Shiga, Japan) and the total RNA was reverse-transcribed to complementary DNA (cDNA) using the avian myeloblastosis virus reverse transcriptase and random primers (Takara, Shiga, Japan) according to the manufacturer's instructions.Real-time PCR amplification of the cDNA was performed in 20 μL reactions according to the SYBR Premix Ex Taq II kit (Takara, Shiga, Japan) protocol; 18S rRNA was used as the internal control.Hairpin-it microRNA Normalization RT-PCR Quantitation (GenePharma) was used to check E2F-1.The oligonucleotide primers for PCR were based on GenBank sequences.The primers used for E2F-1 and GAPDH were as follows: E2F-1 forward, 5-CCATCCAGGAAAAGGTGTGAA-3 and reverse, 5-AGCGCTTGGTGGTCAGATTC-3′.

Dual luciferase reporter assay
The plasmid were designated pGL3-RhoC, pGL3-miR-519d and pGL3-E2F1.The pGL3-RhoC mutant plasmid, pGL3-miR-519d mutant plasmid and pGL3-E2F1 mutant plasmid containing nucleotide mutations in the predicted binding site based on the the website searching in GenBank and inserted into the downstream region of the firefly luciferase reporter (Promega, Madison, WI, USA).HEK293T Cells were seeded at 1 × 105 cells per well in 24-well plates, and cultured until 70% confluent.Cells were then co-transfected with the the pGL3-plasmid and relate plasmid or mimic (50 nM, Ribobio Biotechnology); pGL3-mutant, plasmid-mutant and control (400 ng, Promega) using Lipofectamine 2000.After 48 hours cells were lysed using cell lysis buffer (Cell Signaling, Boston, USA) and luciferase activity was measured using the Dual Luciferase Reporter Assay System (Promega BioSciences, San Luis Obispo, USA) according to the manufacturer's instructions.The luciferase activities were normalized to that of Renilla luciferase.The results were expressed as the means ± SD of at least three independent experiments.

Immunohistochemistry
Consecutive tissue sections were deparaffinized with xylene, rehydrated with alcohol, and subjected to antigen retrieval by heating in target retrieval solution (Dako) for 15 min in a microwave oven (Oriental Rotor).The sections were quenched with 3% hydrogen peroxide for 20 min to block endogenous peroxidase activity.Nonspecific binding was prevented by adding 5% bovine serum albumin for 5 min.The sections were incubated at 4°C overnight with anti-E2F1 or anti-RhoC antibodies, and then incubated with HRP-conjugated anti-rabbit antibodies (Dako) for 2 h.After each treatment, the slides were washed three times with TBST for 5 min, and the binding sites were visualized with 3, 3′-diaminobenzidine.After counterstaining with Mayer's hematoxylin, the sections were dehydrated, cleared and mounted.Negative controls were prepared by omitting the primary antibody.

Statistical analysis
Statistical evaluation was performed using the Spearman correlation test to analyze the rank data and www.impactjournals.com/oncotarget the Mann-Whitney U test to differentiate the means of different groups.Cox's proportional hazards model was employed for multivariate analysis.A p-value of < 0.05 was considered statistically significant.SPSS 17.0 (SPSS, Chicago, IL, USA) software was employed to analyze all data.

Figure 1 :
Figure 1: E2F-1 expression is correlated with pathogenesis and aggressiveness of ovarian carcinoma.E2F-1 expression in normal ovarian tissue (A and D) oviduct tissues (B and E) and ovarian carcinoma tissues (C and F) which was detected by immunohistochemistry.

Figure 3 :
Figure 3: Effects of E2F1 on ovarian carcinoma in vitro and in vivo.Overexpression of E2F-1 (A) increased the rate of growth, (B) promoted G1/S progression, (C) decreased levels of apoptosis, (D) increased migration and (E) invasion ability.Knockdown of E2F1 exhibit significantly (F) decreased growth, (G) increased G1 arrest and (H) increased apoptosis, (I) lower migration and (J) invasion ability.Nude mice injected with cells stably expressing E2F-1 showed a dramatic increase in tumor size (K) and tumor xenograft (L) growth and (M) volum in vivo.Results are representative of three separate experiments; data are expressed as the mean ± standard deviation.*p < 0.05.

Female 4 -
5 week-old, 18~20 g BALB/c nude mice were obtained from Vital River Laboratories (Beijing, China) and housed in a SPF grade environment.Subcutaneous tumor xenografts were established via injection of 1 × 10 7 A2780 cells suspended in 200 μL PBS with (E2F-1 group or miR-519d group) or without (mock group) LV-hsa-E2F-1 transfection or LV-hsa-miR-519d transfection (Genechem, Shanghai, China) directly hypodermic in.All animal research protocols were performed following the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the China Medical University Animal Care and Use Committee.

Table 2 : Relationship between E2F1 expression and clinicopathological features of Type II ovarian carcinomas
Bold means P < 0.05.Immunohistochemical analysis of the clinical relevance of E2F1 expression in ovarian carcinoma showed that E2F1 expression was significantly elevated in type II tumors compared to normal fallopian tube tissues. www.impactjournals.com/oncotarget