Non-malignant (normal) and EOC tissues were obtained with informed patient consent and ethical approval from the Academic Medical Center Institutional Review Board of West China Second University Hospital, Sichuan University (Chengdu, China). The clinical tissues were obtained from the West China Second University Hospital. Experiments described were performed on samples obtained from the Academic Medical Center.

The ovarian cancer cell lines (A2780cp, A2780s, SKOV3 and CAOV3), the human ovarian epithelium cell line (HOEC) and the HEK293 cells were obtained from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (both Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA). All the cells were maintained in a humidified incubator containing 5% CO₂ at 37°C.

The lentivirus-based miR-18a expression system (lenti-miR-18a) and the negative control (lenti-NC) were purchased from GeneChem Co., Ltd. (Shanghai, China). Cell infection was performed according to the manufacturer's instructions. Cells were infected with lenti-miR-18a using a multiplicity of infection of 10. Puromycin (2 µg/ml) was added for selection at 72 h post-infection. Cell transfection was performed using FuGENE^® HD transfection reagent (Roche Diagnostics, Indianapolis, IN, USA) according to the manufacturer's instructions.

Briefly, cells were seeded in 6-well plates at a density of 2×10⁵ cells/well and cultured for 24 h to reach 70–80% confluence. Plasmids [2 µg; tumor protein p53-regulated inhibitor of apoptosis gene 1 (TRIAP1)-3′untranslated region (UTR), TRIAP1-3′UTR-mutant, IPMK-3′UTR and IPMK-3′UTR-mutant] were diluted in 100 µl DMEM without serum. A total of 5 µl FuGENE HD Transfection Reagent was added to the tubes containing diluted DNA. The suspension was mixed and incubated for 15 min at room temperature, and subsequently added to the 6-well plates.

Cell cycle assays were performed by flow cytometry. Briefly, the cells post-treatment were collected by centrifugation for 3 min at 1,200 × g at 4°C and resuspended in 1 ml cold PBS. Following fixation with 4 ml absolute ethanol, the cells were centrifuged for 3 min at 1,200 × g at 4°C and resuspended in 1 ml PBS. The cells were subsequently stained with 100 µl light sensitive propidium iodide (1 mg/ml; Beyotime Institute of Biotechnology, Beijing, China) and incubated for 5 min at room temperature prior to flow cytometry. A total of 5×10⁴ cells from each group were loaded and counted in each phase.

The miRWalk database (www.ma.uni-heidelberg.de/apps/zmf/mirwalk) and other programs, including miRanda (http://www.microrna.org/microrna/getGeneForm.do), Sanger miRDB (http://www.mirdb.org/miRDB/) and Targetscan (http://genes.mit.edu/tscan/targetscanS.html), were used variously for target prediction. The online tool miRWalk (version 2.0; www.umm.uni-heidelberg.de/apps/zmf/mirwalk/predictedmirnagene) was used to predict potential target mRNAs of miR-18a. To verify this bioinformatic analysis, the TRIAP1-3′UTR and the IPMK-3′UTR, each of which contains a single miR-18a binding site, were cloned downstream of the luciferase open reading frame. Meanwhile, TRIAP1-3′UTR mutant and IPMK-3′UTR mutant, which contain mutated miR-18a binding sites, were also introduced into the luciferase constructs. The 293 cells were infected with lenti-miR-18a, and puromycin was used to select the cells with stable expression. qPCR analysis confirmed that miR-18a was upregulated in the cells (data not shown). The luciferase constructs (TRIAP1-3′UTR, TRIAP1-3′UTR, IPMK-3′UTR and IPMK-3′UTR mutant) were transfected separately into cells that stably expressed miR-18a. Luciferase activity was determined 4 to 6 h later.

The pMiRluc-TRIAP1-3′UTR and pMiRluc-IPMK-3′UTR constructs containing miR-18a binding site were purchased from Fulengen Biotechnology Co. (Guangzhou, China). The luciferase reporter was co-transfected with lenti-NC (negative control) or lenti-miR-18a using FuGENE HD (Roche Diagnostics). Luciferase activity was measured 4 h after by luciferase reporter assay (Promega Corporation, Madison, WI, USA), and the values were normalized with activity of β-galactosidase.

Total RNA was extracted from each experimental group using TRIzol^® reagent (Invitrogen Life Technologies; Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions. RNA concentration was assessed spectrophotometrically at 260 nm (Thermo ND 2000; Thermo Fisher Scientific, Inc. Wilmington, DE, USA). RT was performed using a RT kit (Takara Bio, Inc., Otsu, Japan). PCR was subsequently performed using a Real Time PCR kit (Takara Bio, Inc.) on a Bio-Rad CFX96 thermal cycler (Bio-Rad Laboratories, Inc., Hercules, CA, USA), with U6 as an internal control. PCR conditions were as follows: Denaturation at 94°C for 2 min; 30 cycles of amplification at 94°C for 30 sec, annealing at 58°C for 30 sec, and extension at 72°C for 1 min. This was followed by a terminal elongation step at 72°C for 10 min. The Cq value of each PCR product was calculated, and the fold-change was analyzed, according to the protocols of previous studies (17). The primers for miR-18a and U6 were purchased from RiboBio Technology (Guangzhou, China; the sequences were not supplied due to the rules of the company). All experiments were performed in triplicate.

The Cell Counting Kit-8 (CCK-8) assay (Beyotime Institute of Biotechnology) was performed to detect cell viability according to the manufacturer's protocol. Briefly, 10 µl CCK-8 was added into the cell culture medium and incubated at 37°C for 4 h. Subsequently, absorbance was measured at 450 nm and each experiment was performed three times.

Following treatment, cells were lysed on ice for 20 min with radioimmunoprecipitation assay buffer (Beyotime Institute of Biotechnology). A total of 20 µg protein was separated by SDS-PAGE on a 8–12% gel and electronically transferred onto a polyvinylidene difluoride membranes (EMD Millipore, Billerica, MA, USA). Following blocking at 4°C overnight with 5% no-fat milk in TBT Tween 20, the membranes were incubated with the recommended dilutions of primary antibodies against TRIAP1 (dilution, 1:800; cat. no. LS-C346398-50; LifeSpan Biosciences, Inc., Seattle, WA, USA), IPMK (dilution, 1:500; cat. no. TA337886; OriGene Technologies, Inc., Beijing, China), AMP-activated protein kinase (AMPK) (dilution, 1:1,000; cat. no. 2603), phosphorylated (p)-AMPK (dilution: 1:800; cat. no. 2535) p53 (dilution, 1:2,000; cat. no. 2524), cleaved caspase 3 (CC3) (dilution, 1:500; cat. no. 9661), Akt (dilution, 1:1,000; cat. no. 13038), p-Akt (dilution, 1:1,000; cat. no. 4060), cyclin D1 (dilution, 1:800; cat. no. 2978; all Cell Signaling Technology, Inc., Danvers, MA, USA) and glyceraldehyde-3-phosphate dehydrogenase (dilution, 1:5,000; cat. no. SC47724; Santa Cruz Biotechnology, Inc. Dallas, TX, USA) for 1 h at 37°C. The membranes were subsequently incubated with the following secondary antibodies: Horseradish peroxidase (HRP)-conjugated anti-rabbit Immunoglobulin G (IgG; dilution, 1:10,000; cat. no. ZB2301); and the HRP-conjugated anti-mouse IgG antibody (dilution, 1:10,000; cat. no. ZB2305) both from OriGene Technologies, Inc. at room temperature for 2 h. Peroxidase-labeled bands were visualized using an enhanced chemiluminescence kit (EMD Millipore), according to the manufacturer's protocol.

TUNEL assay was performed using a DeadEnd™ Fluorometric TUNEL system (Promega Corporation) to detect apoptotic cells in A2780cp and A2780s cells and A2780cp tumor tissues, according to the manufacturer's instructions. Cell nuclei with green fluorescent staining were defined as TUNEL-positive and visualized using a fluorescence microscope (DTX500; Nikon Corporation, Tokyo, Japan). For quantification of the TUNEL-positive cells, the number of green fluorescent cells was counted in randomly selected fields (magnification, ×200). The cell nuclei were subsequently counter-stained with 4′,6-diamidino-2-phenylindole (Beyotime Institute of Biotechnology).

To establish the A2780cp intraperitoneal cancer model, A2780cp cells (8×10⁶) infected with lenti-miR-18a and lenti-NC (negative control), were injected into the intraperitoneal space of BALB/c nude female mice (6–8 weeks old; ~20 g). A total of 6 mice per group were used. The mice were obtained from the Animal Center of Sichuan University and housed at 26°C, with a 12-h light/dark cycle and ad libitum access to food and water. At 30 days after tumor cell injection, the mice were anesthetized using diethyl ether (100 mg/kg; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) and sacrificed. The dissected tumors were weighed. The tumors and viscera were examined grossly and microscopically stained with hematoxylin and eosin. Animal studies were performed with approval from with the Institutional Animal Care and Treatment Committee of Sichuan University.

Expression of PCNA was analyzed using mouse anti-human PCNA antibody (cat. no. I0710; Santa Cruz Biotechnology, Inc.). Paraffin-embedded tumor sections (3–5 µm) were mounted on 3-aminopropyl triethoxysilane-coated glass slides. Sections were deparaffinized in xylene, treated with a graded series of alcohol [100, 95 and 80% ethanol/double-distilled H₂O (v/v)] and rehydrated in PBS (pH 7.4). Antigen retrieval was performed by heating for 3 min in a pressure cooker with 0.1 mol/l citrate buffer (pH 6.0). Endogenous peroxide was blocked with 3% H₂O₂ for 10 min. Following PBS washes, slides were blocked with 5% normal goat serum (Beyotime Institute of Biotechnology) in PBS for 15 min at room temperature and incubated with primary anti-PCNA antibody (1:200) in blocking solution overnight at 4°C. All slides were subsequently incubated with biotin-conjugated goat anti-mouse or goat anti-rat secondary antibody (both dilution 1:200; cat. nos. SP9001 and SP9002, respectively; OriGene Technologies, Inc.) for 15 min at 37°C and with streptavidin-biotin complex at 37°C for 15 min. The immunoreaction was visualized using diaminobenzidine peroxide solution and the nuclei were counterstained with hematoxylin. All specimens were observed using Olympus BX600 microscopes and SPOT Fiex camera (both Olympus Corporation, Tokyo, Japan). The total number of cells and the number of PCNA-positive cells were counted, and the percentage of PCNA-positive cells was calculated.

All data were analyzed using one-way analysis of variance. Statistical analyses were performed using SPSS Statistics version 19.0 (IBM SPSS, Armonk, NY, USA). Values are expressed as mean ± standard error. P<0.05 was considered to indicate a statistically significant difference.

To determine the expression level of miR-18a in ovarian malignant and normal tissues, malignant ovarian and non-malignant (normal) tissues were collected and used for miR-18a detection by qPCR. As shown in Fig. 1A, miR-18a expression in ovarian malignant tissues was significantly downregulated compared with the normal ovarian tissues (P<0.001). Furthermore, the expression of miR-18a in several ovarian cancer cell lines and the normal ovarian cell line was determined. The results indicated that miR-18a expression was significantly lower in the ovarian cancer cell lines (A2780cp, A2780s, SKOV3 and CAOV3) compared with the non-malignant HOEC cell line (all P<0.01; Fig. 1B).

A lentivirus-based miR-18a expression system was established to overexpress miR-18a in two ovarian cancer cell lines (A2780cp and A2780s) to investigate the potential involvement of miR-18a in inhibiting the development of ovarian cancer. As shown in Fig. 2A and B, qPCR confirmed that miR-18a expression was significantly upregulated when miR-18a was overexpressed in A2780cp and A2780s cells compared with the NC (P<0.01). Analysis of cell viability by CCK-8 assay indicated that overexpression of miR-18a significantly inhibited cell growth in the ovarian cancer cell lines (A2780s and A2780cp) compared with the NC (P<0.01; Fig. 2C).

Flow cytometric analysis of the cell cycle indicated a significant reduction in the number of cells in the S-phase and a significant increase in the G0/G1 phase when miR-18a was overexpressed in the A2780cp and A2780s cells compared with the NC (P<0.01; Fig. 2D and E). However, no significant differences were observed in the G2/M phase between the two groups (Fig. 2D and E). Analysis of cell apoptosis by TUNEL assay indicated significant increases in the number of apoptotic cells when miR-18a was overexpressed in the A2780cp and A2780s cells (P<0.01; Fig. 2F and G).

In order to investigate the direct targets of miR-18a involved in cancer cell proliferation and apoptosis in human ovarian cells, a large number of potential target proteins in a database library were screened to identify potential microRNA binding seed sequences within the 3′-UTR. Bioinformatic analysis (Targetscan micro, RNA.org, microRNASeq) identified TRIAP1 and IPMK as candidate targets (Fig. 3A and B).

Luciferase expression in the cells overexpressing miR-18a and also expressing the TRIAP1-3′UTR construct was significantly reduced. The same finding was observed for the cells overexpressing miR-18a and also expressing the IPMK-3′UTR construct (P<0.01; Fig. 3C and D). No significant reduction in luciferase expression of cells transfected with miR binding site mutant plasmids was observed (Fig. 3C and D).

Western blotting was subsequently employed to determine the expression of TRIAP1 and IPMK proteins, and their downstream targets. Overexpression of miR-18a in A2780cp and A2780s cells markedly inhibited TRIAP1 and IPMK expression (Fig. 3E and F). The expression of cell cycle-associated proteins p-Akt, p-AMPK and cyclin D1 was also downregulated by miR-18a, whereas no significant reduction of Akt and AMPK expression was observed. Furthermore, miR-18a also significantly induced the expression of the apoptosis-associated protein p53, p21 and cleaved caspase 3 in the A2780cp and A2780s cells (Fig. 3E). Collectively, the results indicated that miR-18a directly targets TRIAP1 and IPMK, and hence regulates the expression of the downstream targets.

To further elucidate the effects of miR-18a on ovarian tumor growth in vivo, the A2780cp intraperitoneal mouse model was employed. The tumor was collected (Fig. 4A) and the weight was measured 30 days after the A2780cp injection. The mean tumor weight was 1.05±0.11 and 0.37±0.04 g in the NC and miR-18a group (P<0.01; Fig. 4B). Furthermore, the proliferative and apoptotic activity of the A2780cp tumors were determined by PCNA staining and TUNEL assay. When compared with the negative controls, lower numbers of PCNA-positive cells and higher numbers of apoptotic cells were found in the miR-18a group (P<0.01; Fig. 4C and D). Together, the results in the present study indicated that miR-18a may inhibit A2780cp intraperitoneal tumor growth in vivo by inhibiting proliferation and inducing apoptosis.