GEPIA (gepia.cancer-pku.cn/index.html) was used to screen genes overexpressed in cervical squamous cell cancer, developed survival curve, and automatically calculate Log-rank and HR. Additionally, targetscan (https://www.targetscan.org/vert_72/), mirDIP (http://ophid.utoronto.ca/mirDIP/), and miR-Gator (http://mirgator.kobic.re.kr/), to suggest miRNA with binding sites to target genes.

The present study obtained 19 normal cervical tissues, 88 cervical squamous cell carcinoma tissues and 8 cervical adenocarcinoma tissues from hospitalized patients in the Third Affiliated Hospital Guangzhou Medical University (Guangzhou, China) between 2014–2019. Patients diagnosed with metastatic cervical cancer or other malignant tumors, or with a history of radiotherapy and chemotherapy before surgery were excluded. The patients agreed to use the tissue removed during surgery for general scientific research and signed an informed consent form. The Ethics Committee of the Third Affiliated Hospital Guangzhou Medical University approved the present study (approval no. 2019-037). The tissues were scored according to the degree of staining and the percentage of positively stained cells, and the final scores were added.

The HPV16-positive cervical cancer SiHa cell line (National Infrastructure of Cell Line Resource) was cultured in RPMI1640 medium (HyClone; Cytiva), and the human cervical epithelial immortalized H8 (Jennio Biotech Co., Ltd.), HPV16-positive cervical cancer Caski (National Infrastructure of Cell Line Resource) and HPV18-positive cervical cancer HeLa (Pricella^®; Wuhan Elabscience Biotechnology Co., Ltd.) cell lines were cultured in DMEM medium (HyClone; Cytiva), containing 10% fetal bovine serum (Shanghai Excell Biological Technology Co., Ltd.). The cells were plated in petri dishes and cultured in an incubator at 37°C and 5% CO₂ atmosphere.

siRNAs (Sigma-Aldrich; Merck KGaA) against target genes were used to knockdown the expression of the target genes, and Lipofectamine™ 3000 (Thermo Fisher Scientific, Inc.) was used as a cell transfection reagent. A total of 50 nM siRNA was transfected into SiHa and HeLa cells at 37°C for 48 h immediately before subsequent experiment. The target genes and corresponding siRNA sequences were as follows: CHMP4C, 5′-CAGAUUGAUGGCACACUUUdTdT-3′ (forward) and 5′-AAAGUGUGCCAUCAAUCUGdTdT-3′ (reverse); HPV16 E6, 5′-GAGUAUAGACAUUAUUGUUdTdT-3′ (forward) and 5′-AACAAUAAUGUCUAUACUCdTdT-3′ (reverse); and HPV18 E6, 5′-CAGACUCUGUGUAUGGAGAdTdT-3′ (forward) and 5′-UCUCCAUACACAGAGUCUGdTdT-3′ (reverse). The scrambled siRNA sequences used for the control were as follows: 5′-UUCUCCGAACGUGUCACGUTTdTdT-3′ (forward) and 5′-ACGUGACACGUUCGGAGAATTdTdT-3′ (reverse).

Full-length cDNAs of human CHMP4C were synthesized by Guangzhou Dahong Biotechnology Co., Ltd. and cloned into the expression vector pCMV-MCS-3Flag (Guangzhou Dahong Biotechnology Co., Ltd) (Fig. 1). When cell confluence reached 80–90%, the plasmids were transfected using Lipofectamine 3000 and p3000 reagents (Thermo Fisher Scientific, Inc.), following the manufacturer's protocol. A total of 50 nM microRNA (miRNA/miR) and negative control mimics (Guangzhou RiboBio Co., Ltd.) were transfected into SiHa and HeLa cells using Lipofectamine 3000 reagent. A total of 48 h after transfection at 37°C, the transfection efficiency was immediately assessed using reverse transcription-quantitative PCR (RT-qPCR). The sequences of the miR-543 mimic were 5′-AAACAUUCGCGGUGCACUUCUU-3′ (sense), 5′-UUUGUAAGCGCCACGUGAAGAA-3′ (antisense). The sequences of the negative control mimic were 5′-UUUGUACUACACAAAAGUACUG-3′ (sense), 5′-AAACAUGAUGUGUUUUCAUGAC-3′ (antisense).

After the SiHa and HeLa cells were digested and collected, 3×10³ cells and 100 µl medium were added to each well in a 96-well plate. After the cells had adhered, the experimental group was transfected with CHMP4C siRNA, whereas the negative control group was transfected with scramble siRNA. A total of 20 µl MTT solution was then added to each well, followed by incubation at 37°C for 4 h. The liquid was then aspirated from the well, and 150 µl dimethyl sulfoxide was added. Finally, a microplate reader (BioTek; Agilent Technologies, Inc.) was used to measure the optical density value of each well at 490 nm.

SiHa and HeLa cells were inoculated in a six-well plate. When the cell density reached ~80%, a 200 µl pipette tip was used to scratch the cells, and then serum-free medium was added. Images of the scratches were captured under a microscope (Olympus IX73; fluorescence microscope; Olympus Corporation and Leica DMi1; light microscope; Leica Microsystems GmbH) and then processed using cellSens Version 2.1 (Olympus Corporation) and Leica Application Suite X Version 4.12 software (Leica Microsystems GmbH). The wound area was measured using Image J software Version 1.52a (National Institutes of Health). The following formula was used to assess the level of wound-healing: Mobility=(starting area-final area)/starting area ×100%.

The transfected SiHa and HeLa cells were collected using EDTA-free trypsin and washed once with PBS. The cells were resuspended in 100 µl binding buffer (1X) and 5 µl Annexin V-FITC/PI staining solution (BD Biosciences) and placed in the dark. After the addition of 400 µl binding buffer (1X) to each tube, apoptosis was detected using flow cytometry (BD FACScanto™ II Clinical Flow Cytometry System; BD Biosciences, and Attune NxT Flow Cytometer; Thermo Fisher Scientific, Inc.). FlowJo software (version 10.8; BD Biosciences) and Attune NxT Flow Cytometry Software (version 2.6; Thermo Fisher Scientific, Inc.) were used for analysis.

Matrigel (BD Bioscience) was diluted at a 1:15 ratio with cold serum-free medium (RPMI-1640 or DMEM; HyClone; Cytiva) refrigerated at 4°C and 30 µl was plated into the filters of the Transwell chambers (Corning, Inc.). Then the chambers were placed in a 24-well plate in a 37°C incubator for 4 h. Cells were suspended at a density of 4×10⁴ in 200 µl serum-free medium (RPMI1640 or DMEM, HyClone; Cytiva) and plated in the upper chamber, whilst 600 µl medium supplemented with 10% fetal bovine serum (Shanghai Excell Biological Technology Co., Ltd. Bio) was added to the lower chamber. After incubation at 37°C for 48 h, the chamber was fixed with 4% formaldehyde for 30 min and stained with 1% crystal violet for 15 min both at room temperature. Photographs were captured under a microscope (Olympus IX73, fluorescence microscope, Olympus Corporation), and the invasive cells were manually counted. Each experiment was performed in triplicate, and images of 3 fields of view were captured for each replicate.

The normal and cervical cancer tissues were fixed in 4% paraformaldehyde (at room temperature for 24 h, embedded in paraffin wax, and cut into 4 mm slices. The paraffin-embedded tissue sections were heated in a 65°C incubator for 4 h and dewaxed with xylene (Foshan Bodi Chemical Reagent Co., Ltd) for 20 min twice at room temperature. Then they were immersed into 100% alcohol (Foshan Bodi Chemical Reagent Co., Ltd) for 5 min, 95% alcohol for 2 min, 85% alcohol for 2 min, 75% alcohol for 1 min, and finally turned into distilled water for 3 min 3 times, so as to fully hydrate. The sections were immersed in an EDTA antigen retrieval solution and heated in a pressure cooker for 5 min with the cover unlocked. Then the cooker was locked, removed from the heat source 3 min after the pressure reaches the highest. After then, the sections allowed to cool at room temperature. After inactivation of endogenous catalase activity using 3% hydrogen peroxide, the sections were blocked with 5% BSA (BEIJING SOLARBIO Technology Co., Ltd.) in a 37°C incubator for 40 min. The sections were then incubated with anti-CHMP4C antibodies (1:400; #GTX122876; GeneTex, Inc.) overnight at 4°C. The following day, the sections were re-warmed at room temperature for 30 min and washed three times with PBS. Horseradish peroxidase-conjugated anti-rabbit antibodies (1:800; cat. no. ab97051; Abcam) were then added and the sections were incubated for 45 min in a 37°C incubator. After diaminobenzidine coloration and hematoxylin staining at room temperature for 10 min, the sections were dehydrated, sealed with a neutral resin and imaged under a microscope (Leica Microsystems GmbH).

As the presence of a binding site between CHMP4C and miR-543 was predicted (data not shown), a CHMP4C 3′-untranslated region (3′-UTR) containing a miR-543-binding site sequence was designed. 293T cells (Pricella^®; Wuhan Elabscience Biotechnology Co., Ltd.) were seeded into 24-well plates and transfected with 600 ng SV40-dual-luciferase vector (Shanghai Genechem Co., Ltd.) expressing a wild type CHMP4C 3′-UTR using Lipofectamine™ 3000 (Thermo Fisher Scientific, Inc.), whilst the mutant type acted as a control. At the same time, 50 nM miR-543 or negative mimics (Shanghai Genechem Co., Ltd.) were co-transfected. The sequence of the miR-543 mimics were 5′-AAACAUUCGCGGUGCACUUCUU-3′ (sense) and 5′-AAGAAGUGCACCGCGAAUGUUU-3′ (antisense). The sequences of the negative control mimic were 5′-UUUGUACUACACAAAAGUACUG-3′ (sense), 5′-AAACAUGAUGUGUUUUCAUGAC-3′ (antisense). After 48 h, the Dual-Luciferase^® Reporter Assay System (Promega Corporation) was used to detect the level of luciferase activity. The relative luciferase intensity was normalized to renilla luciferase activity.

Total RNA was extracted from SiHa and HeLa cervical cancer cells using the TRIzol reagent (Takara Bio, Inc.). After separation, precipitation, washing and dissolution, the concentration of RNA was detected using a spectrophotometer [Unico (Shanghai) Instrument Co., Ltd.]. For mRNA analysis, 2 µg total RNA was reverse-transcribed to synthesize complementary (c)DNA using random primers and avian myeloblastosis virus transcriptase (PrimeScript™ RT reagent Kit (Perfect Real Time); Takara Bio, Inc.). The reverse transcription program was set at 37°C for 15 min, 85°C for 5 min, and finally cooled to 4°C. qPCR amplification of the cDNA was then performed using the SYBR^® Premix Ex Taq™ II kit (Takara Bio, Inc.) with 18S as an internal control. The amplification conditions were as follows: Initial denaturation at 95°C for 30 s, followed by 40 cycles of 95°C for 5 s and an 60°C for 30 sec. The Hairpin-it™ miRNAs qPCR Quantitation Kit (Shanghai GenePharma Co., Ltd.) was used to detect miRNA expression according to the manufacturer's protocols, and U6 was used as the internal reference. The relative expression of CHMP4C or miR-543 was calculated using the 2^−ΔΔCq method (11). The sequences of the primers used were as follows: CHMP4C, 5′-GAGAAGCCCTGGAGAAC-3′ (forward) and 5′-CACCAAAGCCAACCC-3′ (reverse); 18S, 5′-CTTAGTTGGTGGAGCGATTTGTC-3′ (forward) and 5′-CGGACATCTAAGGGCATCACA-3′ (reverse); miR-543, 5′-GAGAAGTTGCCCGTGTT-3′ (forward) and 5′-CGCGAATGTTTCGTCA-3′ (reverse); and U6, 5′-CGCTTCGGCAGCACATATAC-3′ (forward) and 5′-AAATATGGAACGCTTCACGA-3′ (reverse).

SiHa and HeLa cell proteins were extracted using RIPA lysis buffer (Shanghai Bestbio Biotechnology Co., Ltd,) after transfection, and protein determination was performed by the BCA method. Proteins of different molecular weights were separated using 10% SDS-PAGE in which 40 µg protein was loaded to each lane, and subsequently electrotransferred to Hybond membranes (MilliporeSigma; Merck KGaA). Subsequently, 3% BSA (BEIJING SOLARBIO TECHNOLOGY CO., LTD) was used to block non-specific sites at room temperature for 2 h, and anti-CHMP4C (1:1,000; #16256-1-AP; Proteintech Group, Inc.), anti-Bcl-XL (1:2,000; #10783-1-AP; Proteintech Group, Inc.), anti-Bcl2 (1:500; #66799-1-Ig; Proteintech Group, Inc.), anti-Survivin (1:1,000; #10508-1-AP; Proteintech Group, Inc.), anti-Caspase-7 (1:1,000; #27155-1-AP; Proteintech Group, Inc.) and anti-HPV16/18 E6 (1:50; #sc-460; Santa Cruz Biotechnology, Inc.) antibodies were added and incubated overnight at 4°C. Anti-β-actin (1:10,000; #66009-1-Ig; Proteintech Group, Inc.), GAPDH (1:1,000; #60004-1-Ig; Proteintech Group, Inc.) or anti-α-tubulin (1:1,000; #80762-1-RR; Proteintech Group, Inc.) antibodies were used as controls. On the second day, the membrane was incubated with anti-rabbit IgG (1:10,000; cat. no. #SA00001-2; Proteintech Group, Inc.) or anti-mouse (1:10,000; #SA00001-1; Proteintech Group, Inc.) antibodies at room temperature for 2 h. The protein bands were visualized using the Enhanced Chemiluminescence kit (SuperSignal™ West Pico PLUS Chemiluminescence substrate; Thermo Fisher Scientific, Inc.).

SPSS 24.0 software (IBM Corp.) was used to perform the statistical analysis. All data are expressed as the mean ± standard deviation. The immunohistochemistry results were analyzed using Fisher's exact test. Comparisons between the experimental and control groups were performed using an unpaired double-tailed t-test, and multiple groups were compared using one-way ANOVA and Dunnett's posttest. P<0.05 was considered to indicate a statistically significant difference.

Bioinformatics analysis was used to assess the role of CHMP4C in cervical cancer. The GEPIA database and log-rank tests demonstrated that a higher CHMP4C expression may have a worse clinical prognosis than those with a lower CHMP4C expression (Fig. 2A). Following this, immunohistochemistry was performed using normal cervical and cervical cancer tissues to further assess the expression of CHMP4C in cervical cancer. The results revealed that the expression level of CHMP4C in malignant cervical tumor tissues was significantly higher than that in normal tissues (Fig. 2B-D; Table SI; P<0.05). This finding indicates that CHMP4C may participate in cervical cancer tumorigenesis. Moreover, the level of CHMP4C was positively associated with the age of the patients (Table SII; P<0.05).

Firstly, expression of CHMP4C in each HPV-positive cervical cancer cell line was detected using RT-qPCR, and SiHa and HeLa cells were selected as the experimental objects as they had the highest expression of CHMP4C out of the cervical cancer cell lines assessed (Fig. 3A). Furthermore, western blotting demonstrated that the expression level of CHMP4C was significantly higher in cervical cancer SiHa and HeLa cells compared with that in the normal cervical H8 cells (Fig. 3B).

To evaluate the biological function of CHMP4C, SiHa and HeLa cells were transfected with si-CHMP4C or si-Scramble, and the transfection efficiency was determined using RT-qPCR. The results demonstrated that the expression level of CHMP4C significantly decreased following siRNA transfection, compared with the control (Fig. 4A and B; P<0.05). Silencing of CHMP4C expression was also associated with a significant decrease in cell proliferation, compared with the control (Fig. 4C and D; P<0.05). This indicates that CHMP4C promotes cell proliferation. Moreover, the cell apoptosis assay revealed that after 48 h of transfection, the si-CHMP4C cells had a significantly higher apoptosis rate than the control cells (Fig. 4E and F; P<0.05). A significant reduction in cell migration (Fig. 4G and H; P<0.05) and invasion (Fig. 4I and J; P<0.05) was also demonstrated in cells with downregulated CHMP4C expression, in comparison with the control cells. Conversely, the plasmid was transfected in the SiHa and HeLa cells to overexpress CHMP4C and assess the transfection efficiency of the plasmid using RT-qPCR (Fig. 5A and B; P<0.05). The results revealed that, in comparison with the vector control, overexpression of CHMP4C resulted in a significant increase in cell proliferation (Fig. 5C and D; P<0.05) and migration (Fig. 5E and F; P<0.05), and a significant reduction in the rate of cell apoptosis (Fig. 5G and H; P<0.05). The observed changes in phenotype indicate that CHMP4C acts as a proto-oncogene in cervical cancer cells.

The occurrence of cervical cancer is related to HPV (namely, high-risk HPV16 and HPV18) infection. Viral DNA is integrated into the host genome to encode oncogenic proteins, such as E6, which interfere with the cell cycle and apoptosis via regulation of Bcl2 and Bcl-XL (12–14). Western blotting showed that the expression of Bcl2, Bcl-xL, and Survivin was decreased when CHMP4C expression was silenced, but Caspase-7 expression was increased (Fig. 6A). Therefore, CHMP4C increased the susceptibility of cervical cancer through the regulation of Bcl2, Bcl-XL, Survivin and Caspase-7. Furthermore, to determine whether CHMP4C expression was associated with cervical HPV infection, HPV16 E6 expression was silenced in SiHa cells, and HPV18 E6 expression was silenced in HeLa cells through siRNA transfection. This resulted in a significant decrease in the expression of CHMP4C, compared with the controls (Fig. 6B-D; P<0.05), and demonstrated that HPV infection may lead to increased expression of CHMP4C.

E6 induction of the expression of CHMP4C was assessed. A previous study reported that E6 could inhibit miR-543 expression in human foreskin keratinocytes (HFKs) (15). Accordingly, the findings of the present study demonstrated significant increases in miR-543 expression after E6 siRNA transfection in cervical cancer cells, compared with that of the controls (Fig. 7A and B; P<0.05). Moreover, the results of bioinformatics analysis revealed that miR-543 could bind with the 3′-UTR of CHMP4C (Fig. 7C). In addition, the dual-luciferase reporter assay demonstrated that the relative luciferase activity was significantly reduced in the group co-transfected with wild-type CHMP4C 3′-UTR and miR-543, compared with that of the control groups (Fig. 7D; P<0.05). These results imply that miR-543 directly targets the 3′-UTR of CHMP4C. Furthermore, after overexpression of miR-543 by transfection of the miR-543 mimics in SiHa and HeLa cells, western blotting indicated that CHMP4C expression was significantly reduced compared with that in control cells (Fig. 7E-G; P<0.05).