The present study was performed according to the principles of the Declaration of Helsinki and approved by the Ethics Committees of The Third Affiliated Hospital of Sun Yat-Sen University. Written informed consent for research purposes was also provided by each participant. Cervical cancer tissues and corresponding adjacent normal tissues were collected from 65 patients who underwent surgical resection at The Third Affiliated Hospital of Sun Yat-sen University between September 2014 and January 2016. All the patients did not receive prior radiotherapy or chemotherapy. All tissue samples were stored in liquid nitrogen until use.

Cervical cancer cell lines (HeLa, C-33A, SiHa and Ca-Ski) and a human normal cervical epithelial cell line (Ect1/E6E7) were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). Cervical cancer cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 IU/ml penicillin and 100 µg/ml streptomycin (Gibco, Grand Island, NY, USA). Ect1/E6E7 cells were maintained in keratinocyte serum-free medium (Gibco) containing 0.1 ng/ml human recombinant epithelial growth factor, 0.05 mg/ml bovine pituitary extract, 100 IU/ml penicillin, and 100 µg/ml streptomycin (Gibco) at 37°C in a humidified incubator with 5% CO₂.

The miR-433 mimics and negative control miRNA mimics (miR-NC) were obtained from GeneCopoeia (Guangzhou, China). MTDH-targeted small interfering RNA (si-MTDH) and the negative control siRNA (si-NC) were chemically synthesized by GenePharma Co., Ltd. (Shanghai, China). MTDH overexpressed vector (pCDNA3.1-MTDH) and corresponding blank vector (pCDNA3.1) were obtained from the Chinese Academy of Sciences (Changchun, China). Cells were seeded into 6-well plates 18–24 h before transfection. Following the protocols for the use of Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), miR-433 mimics (60 nM), miR-NC (60 nM), si-MTDH (60 nM), si-NC (60 nM), pcDNA3.1-MTDH (2 mg/ml) or pcDNA3.1 (2 mg/ml) was transfected into the cells. After incubation for 6 h, the culture medium was replaced with fresh DMEM with 10% FBS. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and western blotting were performed to determine the transfection efficiency.

Total RNA was isolated from tissue samples or cells using TRIzol (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions. For miR-433 expression, reverse transcription was performed using TaqMan MicroRNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA). Quantitative PCR was performed to detect the miR-433 expression level using TaqMan MicroRNA PCR kit (Applied Biosystems). To quantify MTDH mRNA expression, PrimeScript RT reagent kit was used to synthesize cDNA, which was then amplified using SYBR Premix Ex Taq™ kit (both from Takara Bio, Dalian, China). GAPDH and U6 were used for normalization of MTDH mRNA and miR-433, respectively. Primers used in the present study were purchased from Guangzhou RiboBio Co., Ltd. (Guangzhou, China) and shown in Table I. Each sample was performed in triplicate, and relative expression changes were calculated using the 2^−ΔΔCt method (29).

Cell proliferation was determined using the MTT assay (Sigma, St. Louis, MO, USA). At 24 h post-transfection, transfected cells were collected and seeded into 96-well plates at a density of 3.0×10³/well. The plates were incubated for 0, 24, 48 or 72 h after transfection. At each time point, cells were treated with 20 µl MTT assay reagent (5 mg/ml) for additional 4 h. The supernatant was removed, and 150 µl of dimethyl sulfoxide (Sigma) was added into each well. Cellular proliferation was determined by detecting the optical density (OD) at a wavelength of 490 nm. Each assay was performed in triplicate and repeated 3 times.

The Matrigel invasion chambers were utilized to assess cell invasion ability (8 µm; Corning, Cambridge, MA, USA). At 48 h post-transfection, cells were incubated with FBS-free culture medium. On the following day, cells were harvested and suspended in FBS-free culture medium. Cells (1×10⁵) were placed intothe upper chambers, and the lower chambers were filled with DMEM containing 10% FBS. After incubation for 48 h, the cells remaining on the top of the chambers were removed. Cells that invaded to the bottom of the membranes were fixed, stained with 0.5% crystal violet and washed. The invasive cells in at least 5 randomly selected fields were photographed and counted under an inverted microscope (Olympus Corp., Tokyo, Japan). The present study was performed in triplicate and repeated 3 times.

The cell apoptosis rate was determined using the Dead Cell Apoptosis Kit with Annexin V Alexa Fluor™ 488 and propidium iodide (PI) (catalog no. V13241; Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer's instructions. Subsequent to a 72-h incubation, transfected cells were harvested. After washing 3 times with ice-cold phosphate-buffered saline (PBS), the transfected cells were fixed in 80% ice-cold ethanol in PBS. Subsequently, the cells were resuspended in 100 µl Annexin-binding buffer and incubated with 5 µl Annexin V-FITC and 3 µl PI (50 µg/ml). After incubation at room temperature in the dark for 20 min, cell apoptosis was examined using flow cytometry and analyzed using FACSCalibur and CellQuest software (Beckman Coulter, Inc., Miami, FL, USA).

TargetScan (http://www.targetscan.org/index.html) and miRanda (http://www. microrna.org/microrna/) were adopted to analyze the potential targets of miR-433 (30). ‘Human’ was selected as the species, and ‘miR-433’ was entered. Putative miRNA-mRNA interaction was based on the total context score. The more negative the total context score, the higher the probability of miRNA-mRNA binding. Relevant targets predicted by all 2 databases were chosen for laboratory experimentation.

Luciferase reporter plasmids, pMIR-Report-MTDH-3′-UTR-wild-type (Wt) and pMIR-Report-MTDH-3′-UTR-mutant (Mut), were synthesized and confirmed by GenePharma. Cervical cancer cells were co-transfected with pMIR-Report-MTDH-3′-UTR-Wt or pMIR-Report-MTDH-3′-UTR-Mut and miR-433 mimics or miR-NC using Lipofectamine 2000. After incubation for 48 h, luciferase reporter assays were conducted using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA) following the manufacturer's instructions. Firefly luciferase activity was used as an internal control for Renilla luciferase activity. All experiments were carried out in triplicate and repeated 3 times.

Total cell lysates were prepared by incubating cells in RIPA buffer (Beyotime, Shanghai, China) on ice for 1 h. The concentration of protein was determined using the BCA protein assay kit according to the manufacturer's protocol (Pierce Biotechnology, Inc., Rockford, IL, USA). Equal amounts of proteins were separated on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels, transferred onto polyvinylidene difluoride membranes (Millipore, Billerica, MA, USA), blocked with 5% skim milk in Tris-buffered saline containing 0.05% Tween-20 (TBST), and incubated with primary antibodies overnight at 4°C as follows: mouse anti-human monoclonal MTDH (sc-517220; 1:1,000 dilution), mouse anti-human monoclonal p-AKT (sc-271966; 1:1,000 dilution), mouse anti-human monoclonal AKT (sc-56878; 1:1,000 dilution), rabbit anti-human monoclonal p-β-catenin (ab75777; 1:1,000 dilution) (all from Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse anti-human monoclonal β-catenin antibody (ab22656; Abcam, Cambridge, UK, USA), and mouse anti-human monoclonal GAPDH (sc-32233; 1:1,000 dilution; Santa Cruz Biotechnology). In the following steps, membranes were washed with TBST and probed with corresponding horseradish peroxidase (HRP)-conjugated secondary antibody (1:5,000 dilution; Santa Cruz Biotechnology) for 2 h at room temperature. The protein bands were detected using enhanced chemiluminescence (ECL) solution (Pierce Biotechnology, Inc.). GAPDH was used as an internal control.

All data are expressed as mean ± SD, and were compared with two-tailed Student's t-test or one way ANOVA using SPSS 19.0 statistical software (SPSS, Inc., Chicago, IL, USA). Student-Newman-Keuls (SNK) was used to compare differences between 2 groups in the multiple group study. P<0.05 was considered to indicate a statistically significant result.

In the present study, we first examined miR-433 expression in cervical cancer and corresponding adjacent normal tissues by RT-qPCR. Our results showed much higher miR-433 expression in the adjacent normal tissues than that observed in the cervical cancer tissues (Fig. 1A; P<0.05). We then investigated the association between miR-433 and clinicopathological features in cervical cancer. The median miR-433 expression level (median=0.8902) was regarded as a cut-off to divide all cervical cancer patients into either the miR-433 low-expression group (n=33) or miR-433 high-expression group (n=32). As shown in Table II, low miR-433 expression level was significantly correlated with tumour size (P=0.018), FIGO stage (P=0.031), lymph node (P=0.001) and distant metastases (P=0.032). However, no significant correlations were noted between miR-433 expression and age (P=0.685), histology (P=0.526), HPV infection (P=0.247) and family history of cancer (P=0.163).

We further measured expression levels of miR-433 in cervical cancer cell lines (HeLa, C-33A, SiHa and Ca-Ski) and human normal cervical epithelial cell line (Ect1/E6E7). As shown in Fig. 1B, the expression level of miR-433 was downregulated in cervical cancer cell lines compared with that noted in the Ect1/E6E7 cells (P<0.05).

Downregulation of miR-433 in cervical cancer prompted us to explore whether miR-433 acts as a tumour suppressor in cervical cancer. Therefore, we examined the effects of miR-433 overexpression on cervical cancer cells. HeLa and SiHa cells, which expressed relatively low miR-433 expression among the 4 examined cervical cancer cell lines, were transfected with miR-433 mimics or miR-NC. Expression levels determined through RT-qPCR confirmed marked upregulation of miR-433 in the HeLa and SiHa cells transfected with miR-433 mimics (Fig. 2A and B; P<0.05). Firstly, we measured cellular proliferation using MTT assay after transfection of HeLa and SiHa cells with miR-433 mimics or miR-NC. The results showed that restoration of the expression of miR-433 significantly decreased proliferation of the HeLa and SiHa cells compared with that noted in the miR-NC group (Fig. 2C and D; P<0.05). HeLa and SiHa cells were transfected with miR-433 mimics or miR-NC; and cell invasion assay was performed to evaluate the effects of miR-433 restoration on the invasion of these cells. Invasion capabilities of HeLa and SiHa cells were significantly suppressed when cells were transfected with the miR-433 mimics (Fig. 2E and F; P<0.05). Flow cytometric analysis was used to determine the apoptosis rates of the HeLa and SiHa cells transfected with the miR-433 mimics or miR-NC. The results revealed that at 72 h after transfection, the apoptosis rate was significantly increased in the HeLa and SiHa cells transfected with miR-433 mimics compared with that noted in the miR-NC controls (Fig. 2G and H; P<0.05). These results indicate that miR-433 acts as a tumour suppressor in cervical cancer.

To further elucidate the underlying mechanisms involved in the tumour-suppressive roles of miR-433 in cervical cancer cells, bioinformatic analysis was performed to search for potential downstream targets of miR-433. Hundreds of candidate targets were predicted, such as PAK4, PAX6, MACC1, MTDH, CREB1 and Notch1. Among these candidate targets, MTDH was selected for further target identification (Fig. 3A); this gene was found to be upregulated in cervical cancer tissues and to contribute to cervical cancer occurrence and progression (27,28). We employed a luciferase reporter assay to ascertain whether MTDH is a direct target gene of miR-433. HeLa and SiHa cells were cotransfected with pMIR-Report-MTDH-3′-UTR-wild-type (Wt) or pMIR-Report-MTDH-3′-UTR-mutant-type (Mut) and miR-433 mimics or miR-NC. The results showed that miR-433 introduction decreased luciferase activity of pMIR-Report-MTDH-3′-UTR-Wt, but not pMIR-Report-MTDH-3′-UTR-Mut (Fig. 3B; P<0.05).

We further investigated whether miR-433 overexpression can regulate MTDH expression in cervical cancer cells. Results of RT-qPCR and western blot analysis showed that ectopic expression of miR-433 in HeLa and SiHa cells significantly reduced MTDH expression at both mRNA and protein levels (Fig. 3C and D; P<0.05). Collectively, these results suggest that MTDH is a direct downstream target of miR-433 in cervical cancer.

To further explore the association between miR-433 and MTDH, MTDH expression was detected in cervical cancer and corresponding adjacent normal tissues. Data from RT-qPCR and western blot analysis demonstrated significantly increased expression of MTDH in cervical cancer tissues compared with that in corresponding adjacent normal tissues (Fig. 4A and B; P<0.05). We also evaluated the correlation between MTDH mRNA and miR-433 expression level in cervical cancer tissues and Spearman's correlation analysis indicated a significantly negative correlation between miR-433 and MTDH mRNA expression among the cervical cancer tissues (Fig. 4C; r=−0.7038; P<0.0001).

To investigate whether downregulation of MTDH expression exhibits tumour-suppressive functions similar to those of miR-433 overexpression in cervical cancer, HeLa and SiHa cells were transfected with si-MTDH to genetically knock down endogenous MTDH expression (Fig. 5A; P<0.05). Next, MTT assay, cell invasion assay and flow cytometric analysis were conducted in HeLa and SiHa cells transfected with si-MTDH or si-NC. The results showed that MTDH knockdown exhibited tumour-suppressive roles similar to those of miR-433 overexpression in cervical cancer cell proliferation (Fig. 5B and C; P<0,05), invasion (Fig. 5D and E; P<0,05) and apoptosis (Fig. 5F and G; P<0,05) and further suggest that MTDH is a functional downstream target of miR-433 in cervical cancer.

To further evaluate whether MTDH mediates the effects of miR-433, which affects cervical cancer cell proliferation, invasion and apoptosis, rescue experiments were performed, and miR-433 mimics with or without pcDNA3.1-MTDH were transfected into HeLa and SiHa cells. Western blot results confirmed that MTDH expression was recovered in the miR-433 mimic-transfected cells after being transfected with pcDNA3.1-MTDH (Fig. 6A; P<0.05). Rescue experiments revealed that MTDH overexpression markedly reversed the effects of miR-433 overexpression in regards to proliferation (Fig. 6B and C; P<0,05), invasion (Fig. 6D and E; P<0,05) and apoptosis (Fig. 6F and G; P<0,05) of HeLa and SiHa cells. These results indicate that miR-433 exert its tumour-suppressing roles in cervical cancer cells, at least in part, by suppressing MTDH.

MTDH was previously reported to play essential roles inthe regulation of the AKT and β-catenin pathways (31,32). Thus, we detected expression levels of p-AKT, AKT, p-β-catenin and β-catenin in HeLa and SiHa cells after transfection with miR-433 mimics or miR-NC. As shown in Fig. 7, restoration of expression of miR-433 decreased p-AKT and p-β-catenin expressions in the HeLa and SiHa cells. However, this restored expression did not affect total AKT and β-catenin expression. We also noted recovered expression levels of p-AKT and p-β-catenin in the miR-433 mimic-transfected HeLa and SiHa cells cotransfected with pcDNA3.1-MTDH. These results indicate that miR-433 exerts tumour-suppressing roles in cervical cancer cells by directly targeting MTDH and affecting downstream AKT and β-catenin pathways.