The collection of clinical samples were approved by the Research Ethics Committee of Renmin Hospital of Wuhan University. All patients enrolled in this study were pathologically confirmed as cervical cancer and were provided with written consent and agreed to donate their tissue samples for research use. These clinical samples were stored in liquid nitrogen before extracting RNAs.

Cervical cancer cell lines including C33A, HeLa, Caski and SiHa were cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Carlsbad, CA, USA) supplemented with fetal bovine serum (10%, FBS; HyClone Laboratories, Logan, UT, USA) and penicillin/streptomycin (1%; Invitrogen). Human normal cervical epithelial cells (NCEC) were derived from healthy female cervical tissue and were cultured in serum-free medium (Invitrogen) along with EGF, bovine pituitary extract and penicillin/streptomycin (1%; Invitrogen). All cells were maintained in cell incubator at 37°C with 5% CO₂. miR-187 mimic and inhibitor were obtained from Shanghai GenePharma Co., Ltd. (Shanghai, China). FGF9 vector and FGF9 siRNA were from Guangzhou Ruibo Biological Technolog, Co., Ltd. (Guangzhou, China). All these vectors were transfected into cervical cancer cells using Lipofectamine 2000 (Invitrogen) based on the provided protocols. The transfection efficiency of miR-187 mimic and inhibitor were confirmed by evaluating miR-187 level with qRT-PCR. The transfection efficiency of FGF9 vector and siRNA were confirmed by evaluating FGF9 level with western blot analysis.

The RNA in cervical cancer tissues and cell lines was isolated with TRIzol reagent (Invitrogen) according to the manufacturer's instructions. Reverse transcription of miRNA and mRNA were performed with miScript II RT kit (Qiagen, Hilden, Germany) and QuantScript RT kit (Tiangen Biotech, Co., Ltd., Beijing, China), respectively. SYBR-Green PCR kit (Qiagen) was used for RT-PCR quantification. Relative miR-187 level and FGF mRNA level were calculated using 2^−ΔΔCt method after normalization to U6 and GAPDH, respectively.

Cellular proteins from cervical cancer cells were extracted with RIPA buffer. Isolated proteins were subjected to electrophoresis on 4–20% SDS-PAGE gels and transferred to PVDF membranes. After blocked with 5% non-fat milk/TBST (Tris-buffered saline Tween-20), the membranes were incubated with the primary antibody including GAPDH (1:2,000; Santa Cruz Biotechnology, Santa Cruz, CA, USA), FGF9 (1:1,000; Santa Cruz Biotechnology) and caspase-3 (1:1,000; Santa Cruz Biotechnology) at 4°C overnight and incubated with HRP-conjugated secondary antibody at room temperature for 2 h. The protein bands were detected and visualized with the ECL reagent (Beyotime Institute of Biotechnology, Haimen, China).

For viability assay, cervical cancer cells transfected with corresponding vectors were seeded in 96-well plates (5×10³ cells/well). After cell seeding (0, 24, 48 and 72 h) MTT reagent was added to each well and the cells were incubated at 37°C for 4 h. After removing the culture medium, cervical cancer cells were solubilized in 150 µl dimethyl sulfoxide (DMSO) and subjected to colorimetric analysis (wavelength, 490 nm). For proliferation assay, cervical cancer cells transfected with corresponding vectors were seeded in 6-well plates and maintained in cell incubators for 2 weeks. Two weeks later, the cell colonies were stained with crystal violet solution. The number of cell colonies were counted and compared between groups.

To evaluate the apoptosis of cervical cancer cells, the flow cytometry assay was performed to evaluate the percentage of apoptotic cells. In brief, cervical cancer cells were harvested and re-suspended in phosphate-buffered saline (PBS), and then stained with Annexin V detection kit, and subjected to FACScan analysis. Additionally, cellular level of caspase-3 which is a marker of cell apoptosis was evaluated with western blot analysis.

Cervical cancer cells seeded in 24-well plates were transfected with 200 ng of miR-187 mimic or inhibitor or the corresponding control vector along with 50 ng of wild-type (WT) or mutant (MT) 3′-UTR of FGF9 mRNA. Forty-eight hours after the transfection, these cells were collected and luciferase activity was detected with a Dual-luciferase reporter assay system following the manufacturer's instructions (Promega, Madison, WI, USA).

To investigate the in vivo influence of miR-187 on cervical cancer cells, C33A cells transfected with miR-187 mimic or control vector were subcutaneously inoculated into nude mice. Twenty-eight days after the cell inoculation, the mice were sacrificed and the formed subcutaneous tumors were subjected to Ki-67 staining. Tumor volumes were calculated every 7 days based on the following formula: length × width²/2. The protocols for in vivo experiments were approved by the Animal Research Protection Committee of Renmin Hospital of Wuhan University.

Statistical analyses were performed using the GraphPad and SPSS 15.0 software. Comparisons between the groups were performed using the t-test and the χ² test. Overall survival and progression-free survival analysis were performed using the Kaplan-Meier method for plotting and the log-rank test for comparison. All differences were regarded as statistically significant at the level of P<0.05.

Using qRT-PCR, miR-187 expression levels were evaluated in 60 pairs of cervical cancer tissues and the corresponding non-tumor tissues, as well as in cervical cancer cell lines. Compared with the non-tumor tissues, cervical cancer tissues showed significantly decreased level of miR-187 (P<0.05; Fig. 1A). Moreover, the level of miR-187 was elevated with the progression of the stage of cervical cancer (P<0.05; Fig. 1B). Additionally, miR-187 level in all cervical cancer cell lines including Caski, SiHa, HeLa and C33A was significantly decreased compared with that in the normal cervical epithelial cells (NCEC) (P<0.05; Fig. 1C).

To explore the clinical significance of miR-187 level in cervical cancer, we investigated the correlation of miR-187 expression level with the prognosis of cervical cancer patients. As shown in Fig. 2A, patients with low level of miR-187 had significantly decreased rate of overall survival (P<0.05; Fig. 2A). The data of progression-free survival demonstrated that low level of miR-187 was associated with decreased rate of progression-free survival (P<0.05; Fig. 2B), thus indicating that miR-187 could serve as a biomarker for cervical cancer patients.

As the miR-187 level was highest in Caski cells and lowest in C33A cells, we chose Caski cells for knockdown experiments and C33A for overexpression experiments. As shown in Fig. 3A, miR-187 mimic significantly increased the expression level of miR-187 in C33A cells (P<0.05). The MTT assay showed that miR-187 overexpression significantly decreased the cell viability of C33A cells (P<0.05; Fig. 3B). Colony formation assay showed that miR-187 overexpression significantly decreased the number of formed cell colonies of C33A cells (P<0.05; Fig. 3C). On the other hand, flow cytometry assay showed miR-187 overexpression increased the percentage of apoptotic cells (P<0.05; Fig. 3D). The western blot analysis for caspase-3 demonstrated that miR-187 overexpression significantly increased the caspase-3 level in C33A cells (P<0.05; Fig. 3E). Furthermore, we performed miR-187 knockdown in Caski cells. miR-187 inhibitor significantly decreased the level of miR-187 in Caski cells (P<0.05; Fig. 4A). miR-187 knockdown significantly increased the cell viability (P<0.05; Fig. 4B) and cell colony number (P<0.05; Fig. 4C) while decreased the percentage of apoptotic cells (P<0.05; Fig. 4D) and caspase-3 level (P<0.05; Fig. 4E) in Caski cells.

To confirm the in vitro effects of miR-187 on C33A cells, we performed subcutaneous injection experiments for C33A cells. As shown in Fig. 5A and B, forced expression of miR-187 in C33A cells inhibited the tumor growth in nude mice (P<0.05). Furthermore, we performed Ki-67 staining for the formed tumors. The results of Ki-67 staining showed that miR-187 overexpression significantly decreased the number of Ki-67 positive cells in the tumors (P<0.05; Fig. 5C and D).

miRNA regulated cell proliferation, apoptosis and other cellular processes by interacting with the 3′-UTR site of targeted genes. FGF9, which had the complementary sequence for miR-187 binding (Fig. 6A), was one of the predicted genes of miR-187 based on the data of online database. We then performed luciferase assay for wild-type and mutant FGF9 3′-UTR to investigate whether miR-187 could interact with FGF9 3′-UTR. The results of luciferase assay showed that miR-187 overexpression significantly decreased the luciferase activity of wild-type FGF9 3′-UTR while miR-187 knockdown significantly increased that of wild-type FGF9 3′-UTR (P<0.05; Fig. 6B). Altering miR-187 level did not affect the luciferase activity of mutant FGF9 3′-UTR (Fig. 6B). Furthermore, qRT-PCR and western blot analysis showed miR-187 overexpression significantly reduced the mRNA (P<0.05; Fig. 6C) and protein (P<0.05; Fig. 6D) level of FGF9 in C33A cells. miR-187 knockdown significantly increased the mRNA (P<0.05; Fig. 6E) and protein (P<0.05; Fig. 6F) level of FGF9 in Caski cells.

To confirm whether FGF9 was involved in the biological function of miR-187 in cervical cancer cells, we performed rescue experiments for FGF9 in cervical cancer cells. For C33A cells overexpressing miR-187, we used FGF9 vector to overexpress FGF9 in these cells. FGF9 vector significantly increased FGF9 expression in C33A cells overexpressing miR-187 (P<0.05; Fig. 7A). FGF9 overexpression reversed the inhibitory effects of miR-187 on cell viability (P<0.05; Fig. 7B) and colony formation (P<0.05; Fig. 7C) while blocked the promoting effects of miR-187 on apoptosis (P<0.05; Fig. 7D) and caspase-3 level (P<0.05; Fig. 7E). For Caski cells with miR-187 knockdown, we used FGF9 siRNA to inhibit FGF9 expression in these cells. FGF9 siRNA significantly reduced the expression of FGF9 in Caski cells with miR-187 knockdown (P<0.05; Fig. 8A). FGF9 knockdown in these cells significantly blocked the promoting effects of miR-187 knockdown on cell viability (P<0.05; Fig. 8B) and colony formation (P<0.05; Fig. 8C) while reversed the inhibitory effects of miR-187 knockdown on apoptosis (P<0.05; Fig. 8D) and caspase-3 level (P<0.05; Fig. 8E).