The present study consisted of 20 samples of HG-SOC and 12 normal fallopian tube (FT) tissues. All samples were collected between March and June 2013 at Qilu Hospital during surgery and were immediately stored at −80°C. For the use of these samples, signed informed consent was obtained. All the diagnoses were confirmed by at least two pathologists.

Human ovarian cancer cell lines A2780 and SKOV3 were originally obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). Roswell Park Memorial Institute (RPMI)-1640 medium was purchased from Gibco-BRL (Rockville, MD, USA). Fetal bovine serum (FBS) was supplied by Haoyang Biological Manufacture Co. Ltd. (Tianjin, China). The human ovarian cancer cell lines A2780 and SKOV3 were cultured in RPMI-1640 medium supplemented with 10% FBS and 100 U penicillin-streptomycin at 37°C in a humidified atmosphere containing 5% CO₂.

miRNA mimics and miRNA inhibitors (2′-O-methyl modified) for in vitro transfection and its negative control (miR-nc) were designed and synthesized by RiboBio (Guangzhou, China). A2780 (3.5×10⁵) and SKOV3 (3.0×10⁵) cells were seeded into 6-well plates and incubated overnight. Transfection was performed using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. After 6 h, the medium was replaced with fresh medium. During transfection, the medium was antibiotic-free.

The 3′-untranslated region (3′UTR) of ZEB1 mRNA containing the putative miR-1236-3p binding site was cloned into the pGL3 vector according to the manufacturer’s instructions. The forward primer (5′-CGAGCT CGACAGCACAGAGCAGGAA-3′) and the reverse primer (5′-CCCTCGAGTAGTTAGCACGGGTTGGA-3′) were used to amplify the 3′UTR of ZEB1. The forward primer contained a Sac1 restriction site and the reverse primer contained an Xho1 site. The binding site of miR-1236-3p in the ZEB1 3′UTR was mutated by using primers: F, 5′-CAGGCGCTTAAAGG AAGCTGATTAAT-3′ and R, 5′-AGCGCCTGTATTGTTGC TCTCTGAGT-3′. SKOV3 cells were plated at 1.5×10⁴/well in a 96-well plate one day before transfection. Cells were co-transfected with 50 ng of wild-type or mutant ZEB1 3′UTR, and 5 pmol of miR-1236-3p mimics or negative control. After 48 h, luciferase activity was measured using the Dual-Luciferase Reporter assay system (Promega).

Total RNA was extracted from cells using TRIzol reagent (Invitrogen) according to the manufacturer’s protocol. Then cDNA was synthesized using the PrimeScript RT reagent kit (Takara Biotechnology Co., Ltd., Dalian, China). Quantitative real-time PCR (qRT-PCR) was performed with the SYBR Premix Ex Taq (Takara) by using the Applied Biosystems StepOnePlus™ Real-Time PCR System according to the manufacturer’s protocol. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the endogenous control.

The cells were washed 3 times with PBS chilled to 4°C and lysed on ice in RIPA buffer (Shenneng Bocai, Shanghai, China) containing protease inhibitors (1 mM). The density of the protein was measured using the BCA protein assay kit (Merck, Darmstadt, Germany). The same amount of protein was separated by 5–10% SDS-PAGE and then transferred to a PVDF membrane (Millipore) using a semi-dry blotting apparatus (Bio-Rad Laboratories, Hercules, CA, USA). Membranes were blocked with 5% (w/v) non-fat milk in Tris-buffered saline with Tween-20 (TBST) (100 mM NaCl, 50 mM Tris and 0.1% Tween-20) at room temperature for 1 h. After the blocking step, the membranes were incubated overnight with primary antibodies (Cell Signaling Technology, Beverly, MA, USA) at 4°C. The membranes were then washed 3 times with TBST and incubated with horseradish peroxidase-labeled secondary antibodies for 2 h. Signals were detected with an ECL system (Merck) according to the manufacturer’s instructions. GAPDH was used as the loading control.

Invasion assays were performed in a Transwell system (24-wells, 8-μm pore size) coated with 2 mg/ml of Matrigel (both from BD Biosciences, Bedford, MA, USA). First, the cells were transfected with miR-1236-3p. After 48 h, the transfected cells were digested with trypsin, and 1.0×10⁵ cells were suspended in serum-free RPMI-1640 and seeded into the upper chamber of each well. The lower chamber was filled with RPMI-1640 supplemented with 20% FBS. Then the cells were fixed with formaldehyde, permeabilized with 100% methanol and stained with 0.5% crystal violet. The number of cells that had attached to the lower surface of the membrane were counted in 6 random fields under a light microscope and analyzed statistically. The cell migration assays were performed in a similar manner, except that the upper chambers were Matrigel-free.

All data are expressed as means ± standard deviation (SD). Student’s t-test (paired, 2-tail) was employed to analyze the significance of two groups. P-value <0.05 was considered to indicate a statistically significant result. All of the experiments were repeated at least 3 times.

To confirm the expression of miR-1236-3p in HG-SOC tissue samples, we employed qRT-PCR to quantify and compare the miR-1236-3p expression levels in the HG-SOC (n=20) and FT (n=12) tissue samples. As shown in Fig. 1A, the expression of miR-1236-3p was generally decreased in the HG-SOC when compared to the FT samples (P<0.001). This suggests that miR–1236-3p acts as a tumor suppressor in HG-SOC.

To explore the influence of miR-1236-3p on ovarian cancer cells, we transfected SKOV3 and A2780 cells with miR-1236-3p mimics or inhibitors. As shown in Fig. 1C, SKOV3 cells treated with miR-1236-3p mimics presented a cobblestone-like appearance, while cells treated with the negative control exhibited a spindle-like morphology. We then inhibited the expression of miR-1236-3p in A2780 cells. A2780 cells transfected with the miR-1236-3p inhibitors were narrower than cells transfected with the negative control (Fig. 1D). As known, the classic morphological change correlated with EMT is the conversion from a rounded, epithelial-like form to a spindle-shaped, mesenchymal form (8,9). These changes indicated that the manipulation of miR-1236-3p inhibited or stimulated the process of EMT.

To further confirm the role of miR-1236-3p in ovarian cancer cells, we used a two-chamber assay to evaluate the migratory and invasive capacities of SKOV3 and A2780 cells treated with miR-1236-3p mimics or inhibitors. As shown in Fig. 2, increased miR-1236-3p significantly suppressed migration and invasion of the ovarian cancer cells. In contrast, decreased miR-1236-3p promoted these abilities. Our results showed that miR-1236-3p greatly influenced migration and invasion of both ovarian cancer cell lines.

Based on the above findings, we hypothesized that miR-1236-3p regulates genes associated with EMT. Using online miRNA target prediction tools, such as microRNA.org, we found that the 3′UTR of ZEB1 mRNA contained a putative binding site for miR-1236-3p. To investigate the ability of miR-1236-3p to bind and regulate the 3′UTR of ZEB1, we performed the luciferase reporter assay. Wild-type or the mutant ZEB1 3′UTR sequence was cloned into the pGL3 vector (Fig. 3A). SKOV3 cells were co-transfected with miR-1236-3p (mimics or negative control) and the pGL3 vector (wild-type or mutant). After 48 h, cells were lysed to measure the luciferase activity using the Dual-Luciferase Reporter assay system. Overexpression of miR-1236-3p significantly reduced the luciferase activity in the wild-type (P<0.001) but not the mutant ZEB1 3′UTR (Fig. 3B). This demonstrated that ZEB1 was directly targeted by miR-1236-3p.

We then tested whether miR-1236-3p modulates the expression of ZEB1 and EMT-related genes in ovarian cancer. First, the SKOV3 and A2780 cell lines were transfected with miR-1236-3p mimics or miR-nc. The transfected cells were analyzed by qRT-PCR and western blotting. As shown in Fig. 4A and B, ZEB1 was downregulated by miR-1236-3p mimics at both the protein and mRNA levels. We also detected the expression of EMT-related markers E-cadherin and N-cadherin. The results showed that the expression of E-cadherin was upregulated and the N-cadherin expression was downregulated. Second, we downregulated the expression of miR-1236-3p in the two cell lines by using miR-1236-3p inhibitors. As expected, the expression of ZEB1 and EMT-related genes showed an opposite pattern at the protein level (Fig. 4C) and mRNA level (Fig. 4D) compared to the previous assay. As known, a switch from epithelial marker E-cadherin to mesenchymal marker N-cadherin is a classical molecular change during EMT (8,9). Thus, our results indicated that miR-1236-3p may regulate the process of EMT.

Finally, we tested whether miR-1236-3p-induced ZEB1 suppression confirmed in our research was clinically relevant. As shown in Fig. 1B, we found that the expression of ZEB1 protein was generally upregulated in HG-SOC (n=8) when compared to that in the FT (n=8) samples, and its expression was inversely correlated with miR-1236-3p. Taken together, our findings revealed that miR-1236-3p downregulation induced overexpression of ZEB1, and consequently influenced migration and invasion of HG-SOC cells.