Cervical cancer tissue samples and para-carcinoma tissues were collected from 35 patients from the Gansu Provincial Hospital (Lanzhu, China) between 2015 and 2017 (the clinical characteristics are shown in Table I). Each patient provided informed consent and the study was approved by the Ethics Committee of The Gansu Provincial Hospital. All tissue samples were snap-frozen in liquid nitrogen and saved at −80°C.

Normal cervical epithelial cells (Ect1/E6E7) and two cervical cancer cell lines (SiHa and HeLa) were purchased from the American Type Culture Collection (Manassas, VA, USA) and maintained in RPMI-1640 medium which contained 10% fetal bovine serum (FBS). The cells were incubated in an atmosphere containing 5% CO₂ at 37°C.

miRNA-143 mimics, miRNA-143 inhibitor or GOLM1 small interfering RNA (siRNA) as well as the corresponding controls were obtained from Guangzhou RiboBio Co., Ltd. (Guangzhou, China) and transfected into human cervical cancer cell lines by Lipofectamine 2000 (Invitrogen: Thermo Fisher Scientific, Inc., Carlsbad, CA, USA) in line with the manufacturer's instructions. The cells were harvested 48 h post-transfection for further assays.

TRIzol reagent (Invitrogen: Thermo Fisher Scientific, Inc.) was used to isolate the total RNA from cervical cancer tissues and cells. cDNA was synthesized from 1 µg total RNA with a High Capacity cDNA Reverse Transcription kit (Applied Biosystems: Thermo Fisher Scientific, Inc., Foster City, CA, USA). RT-qPCR was carried out on CFX96 real-time PCR detection system (Bio-Rad Laboratories, Inc., Hercules, CA, USA) with SYBR Premix Ex Taq™ (Takara Biotechnology Co., Ltd., Dalian, China). The relative expression of mRNA was analysed using 2^−ΔΔCq (16) (Table II).

Transwell assays were conducted to value the cervical cancer cell invasion and migration capacities. A total of 1×10⁵ cells with transfection of miR-143 mimics or inhibitor were detached, suspended and seeded in the upper chamber. In the meantime, DMEM containing 10% FBS was added to the bottom chambers. The difference between invasion and migration assays was that the Transwell upper chambers were precoated with Matrigel for invasion assays. After incubation for 24 h, the cells that remained on the upper chambers were removed with cotton swabs whereas the cells attached to the lower chambers were fixed and stained, respectively, using formaldehyde (4%) and crystal violet (0.1%) for the detection of the images under a microscope (Olympus Corp., Tokyo, Japan) in five randomly selected fields.

The cells were lysed in chilled RIPA lysis buffer on ice, and then centrifuged at 12,000 × g for 20 min at 4°C and collected the supernatants. Protein concentrations were assessed with a BCA Protein Assay kit (Bio-Rad Laboratories, Inc.). Protein lysate was separated by 10% SDS-PAGE, followed by being electroblotted onto a PVDF membrane (DuPont, Boston, MA, USA), which had been blocked with 5% non-fat milk in TBST for 1 h at room temperature. Next, the membrane was incubated with the primary antibodies: anti-GOLM1 antibody (1:300, cat. no. HPA010638; Atlas Antibodies, Stockholm, Sweden) and anti-GAPDH antibody (1:3,000; cat. no. ab9485) at 4°C overnight, and then rinsed the membrane and incubated it with anti-rabbit IgG (1:4,000; cat. no. ab191866) (both from Abcam, Cambridge, UK) for 2 h at room temperature. Chemiluminescence detection system (Beyotime Institute of Biotechnology, Shanghai, China) was used to detect the results and the protein expression was normalized to GAPDH.

The luciferase reporter assays were carried out to evaluate whether GOLM1 was targeted by miRNA-143. Cervical cancer cells were co-transfected with the wild-type or mutant GOLM1 3′UTR luciferase reporter vectors and miRNA-143 by Lipofectamine™ 2000 (Invitrogen: Thermo Fisher Scientific, Inc.). After transfection (48 h), Dual-Luciferase Reporter Assay System (Promega Corp., Madison, WI, USA) was used to measure the luciferase activities.

All the experiments were performed at least 3 times. The SPSS 18.0 version (SPSS, Inc., Chicago, IL, USA) and GraphPad Prism 6 (GraphPad Software, Inc., La Jolla, CA, USA) were used to evaluate the statistical analysis. Students t test and one way ANOVA followed by Tukeys post hoc test were used to analyze two or multiple groups, respectively. Spearmans correlation analysis was used to determnine the correlation between miR-143 and GOLM1 expressions in cervical cancer tissues. P<0.05 was considered to be a statistically significant difference.

We first evaluated the expression of miR-143 and GOLM1 in cervical cancer tissue specimens. The results showed that the miRNA-143 expression in cervical cancer tissues was remarkably reduced in contrast with the normal tissue samples (Fig. 1A, P<0.01), whereas the GOLM1 mRNA expression in cervical cancer tissue samples was significantly increased (Fig. 1B, P<0.001). The expression of miRNA-143 and GOLM1 was then measured in cervical cancer cells (SiHa and HeLa). The results indicated that, compared to the control group, miRNA-143 expression in both SiHa and HeLa cells was significantly decreased (Fig. 1C, P<0.001), whereas the GOLM1 expression was significantly increased (Fig. 1D, P<0.001). Additionally, statistical analysis revealed that the miRNA-143 expression was negatively correlated with GOLM1 expression in the cervical cancer tissue samples (Fig. 1E). miRNA-143 suppresses cervical cancer cell invasion. HeLa and SiHa cells were transfected with miRNA-143 mimics or inhibitor, respectively. After 48 h, the transfection efficiency was measured. The results demonstrated that the miRNA-143 expression was dramatically upregulated in both SiHa and HeLa cells with transfection of miRNA-143 mimics, while the miRNA-143 expression level was downregulated in cervical cancer cells treated with miRNA-143 inhibitor (Fig. 2A and B, P<0.01). Additionally, cell invasion assay was performed using Transwell chambers. In the cell invasion assay, the overexpression of miRNA-143 decreased SiHa and HeLa cell invasion, whereas inhibition of the miRNA-143 expression enhanced SiHa and HeLa cell invasion (Fig. 2C-E, P<0.01). The results manifested that miR-143 inhibited cervical cancer cell invasion.

To explore the functions of miRNA-143 in metastasis, cell migration assay was also carried out by Transwell chambers. The overexpression of miRNA-143 significantly decreased SiHa and HeLa cell migration, whereas inhibiting the miRNA-143 expression increased SiHa and HeLa cell migration (Fig. 3A-C, P<0.01). This result indicated that miRNA-143 inhibited cervical cancer cell migration.

To investigate the mechanism of miRNA-143 in inhibiting cervical cancer, we continued to explore the potential targets of miRNA-143. In the present study, TargetScan was used to find accurate potential target genes of miRNA-143, and it predicted GOLM1 as a potential target of miRNA-143. The wild-type and mutant GOLM1 3′UTRs were generated with sequences presented in Fig. 4A. Luciferase reporter assay was performed on cervical cancer cells which were transfected with miRNA-143 mimics. Subsequently, cells were co-transfected with wild or mutant GOLM1 3′UTR vectors. The results revealed that the miRNA-143 mimics remarkably declined luciferase activity of the GOLM1 3′UTR-wt vector. Moreover, the luciferase activity of GOLM1 3′UTR-mut vector was not affected by miRNA-143 mimics (Fig. 4B and C, P<0.01). Our findings suggested that GOLM1 was directly targeted by miRNA-143. In addition, the GOLM1 protein expression was remar-kably reduced in cells with the transfection of miRNA-143 mimics and obviously increased in cells with the transfection of miRNA-143 inhibitor compared with their respective controls (Fig. 4D).

We sequentially studied whether GOLM1 was needed in regulating the miR-143 ability in cell invasion and migration of cervical cancer. GOLM1 siRNAs were transfected into SiHa and HeLa cells to knockdown endogenous GOLM1 and RT-qPCR was conducted to examine the GOLM1 mRNA expression. The results revealed that the GOLM1 expression in cervical cancer cells transfected with GOLM1 siRNAs was significantly reduced compared to the control group (Fig. 5A, P<0.001). Furthermore, Transwell assays were conducted to measure the invasion and migration of cervical cancer cells co-transfected with GOLM1 siRNA and miR-143 inhibitor. Our above results indicate that deletion of GOLM1 markedly reversed miR-143-medicated suppression of cell migration and invasion in cervical cancer cells (Fig. 5B-D, P<0.01).