The human gingival cancer cell line Ca9-22 was obtained from the Japanese Collection of Research Bioresources Cell Bank and the human tongue cancer-derived cell line HSC-4 was obtained from the RIKEN BioResource Center. As it has been reported that some stocks of Ca9-22 are contaminated with MSK-922 (19), short tandem repeat analysis was performed for the Ca9-22 cell line by BEX Co., Ltd., and it was confirmed to be authentic. Ca9-22 cells were cultured in minimal essential medium (Nacalai Tesque, Inc.) supplemented with 600 mg/l glutamine and 10% heat-inactivated fetal bovine serum (FBS; Nichirei Biosciences, Inc.). HSC-4 cells were maintained in RPMI-1640 medium (Nacalai Tesque, Inc.) supplemented with 10% heat-inactivated FBS. Both media contained 100 IU/ml penicillin (Thermo Fisher Scientific, Inc.) and 100 mg/ml streptomycin (Thermo Fisher Scientific, Inc.). Cells were maintained at 37°C in an atmosphere containing 95% air and 5% CO₂.

To assess the effects of expression levels of TFAP2E on the survival rate of patients with OSCC, Kaplan-Meier survival analysis, followed by log-rank test for statistical analysis, was performed using the online cBioPortal tool (http://www.cbioportal.org). Sample data from 322 patients in The Cancer Genome Atlas (TCGA) HNSCC Firehose Legacy data set, including gene expression levels in OSCC tissues and survival period, were obtained from cBioPortal.

Ca9-22 and HSC-4 cells were seeded and cultured for 24 h before transfecting with 10 nM siRNA using Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions at room temperature. Anti-TFAP2E siRNA (cat. no. s50548; Thermo Fisher Scientific, Inc.), a control siRNA (si-N/C; cat. no. 4390843; Thermo Fisher Scientific, Inc.) and anti-TP53 siRNA (cat. no. sc-29435; Santa Cruz Biotechnology, Inc.) were used in the present study.

To analyze the cell proliferation rate, cells were seeded into 96-well culture plates at a density of 5×10³ cells/well and transfected with siRNAs after 24 h. After 1, 2, 3, 4 and 5 days, the proliferation of transfected cells was measured using the standard water-soluble tetrazolium salt (WST)-8 assay with Cell Count Reagent CF (Nacalai Tesque, Inc.). Briefly, culture medium was replaced with fresh medium containing 10 µl of WST8 solution, and the cells were cultured for 1 h, followed by measurement of the absorbance at 450 nm using a plate reader (Spectra Max ABS plus; Molecular Devices, LLC).

Cell viability was also analyzed by cell staining. Briefly, cells were seeded into 6-well culture plates at a density of 5×10³ cells/well. A total of 24 h after seeding, cells were transfected with siRNAs and were cultured for 10 days. In the middle of culture, i.e. 5 days after siRNAs transfection, the medium was replaced and siRNA transfection was performed again. Cells were fixed and stained using a Diff-Quick Stain Kit (Sysmex Corporation) at room temperature. Briefly, cells were fixed with 99% methanol for 10 min and stained with Diff-Quick solution II for 5 min, followed by washing with distilled water.

For the analysis of drug sensitivity, cells were seeded into 96-well culture plates at a density of 5×10³ cells/well and transfected with siRNAs after 24 h. In addition, 1, 5 and 10 µM cisplatin (CDDP; MilliporeSigma) or 25, 50 and 100 µM of H₂O₂ (Nacalai Tesque, Inc.) were added 24 h after the transfection, followed by culture for additional 24 h and analysis of cell viability using WST-8 assay.

The cells were plated at a density of 2×10⁴ cells/ml in culture dishes 60 mm in diameter (5 ml/dish) and were transfected with siRNAs 24 h later. A total of 3 days after transfection, both floating and attached cells were collected by centrifugation, washed in PBS and fixed in 70% ethanol at −20°C overnight for fluorescence-activated cell sorting (FACS) analysis. The cells were washed in PBS before incubating in PBS containing 0.1% FBS, 25 µg/ml propidium iodide and 200 µg/ml RNase A for 15 min at room temperature.

Cell cycle progression was monitored in the presence of nocodazole, which prevents cells from completing mitosis (20). The cells were plated and transfected with siRNAs as aforementioned. After 2 days of culture, the medium was replaced with fresh medium containing 100 ng/ml nocodazole (FUJIFILM Wako Pure Chemical Corporation). The cells were harvested every 3 h and fixed in 70% ethanol at −20°C overnight. The cells were washed in PBS containing 0.5% FBS, followed by incubation in PBS containing 0.5% FBS, 2.5 µg/ml propidium iodide, 200 µg/ml RNase A and anti-phosphorylated histone H3 serine 28 (p-H3Ser28) conjugated with Alexa 647 (cat. no. 641006; BioLegend, Inc.) for 1 h at room temperature.

All of the aforementioned cells were subjected to FACS analysis using a Gallios Flow Cytometer (Beckman Coulter, Inc.) and data analysis was performed using Kaluza Analysis Software Ver. 2.1 (Beckman Coulter, Inc.). The percentage distribution of cells in distinct cell cycle phases was calculated based on Michael H. Fox algorithm (21).

Ca9-22 and HSC-4 cells were plated at a density of 2×10⁴ cells/ml in culture dishes 60 mm in diameter (5 ml/dish) and were transfected with siRNAs 24 h after seeding. Total RNA was extracted from cells 2 or 5 days after siRNA transfection, using RNeasy mini kits (Qiagen GmbH) and cDNA was synthesized by RT using an iScript cDNA synthesis system (Bio-Rad Laboratories, Inc.), according to the manufacturers' instructions. qPCR for TFAP2E and 18S rRNA, which was used as a housekeeping gene (22), was performed using the TaqMan Pre-Developed Assay Reagents Hs00698734_m1 and Hs99999901_s1 for TFAP2E and 18S rRNA (Thermo Fisher Scientific, Inc.) respectively, and Premix Ex Taq Perfect Real Time (Takara Bio, Inc.). qPCR conditions were as follows: Initial denaturation at 95°C for 30 sec, followed by 40 cycles at 95°C for 5 sec and 60°C for 30 sec. qPCR for TP53 was performed using SYBR Premix Ex Taq™ (Takara Bio, Inc.) with the following primers: Sense 5′-CCCCTCCTGGCCCCTGTCATCTTC-3′ and antisense 5′-GCAGCGCCTCACAACCTCCGTCAT-3′. qPCR conditions were as follows: Initial denaturation at 95°C for 30 sec, followed by 40 cycles at 95°C for 5 sec, 56°C for 10 sec and 72°C for 30 sec. Measurements were performed in triplicate. Data processing was performed using a standard curve-based method (23). A mixture of cDNA generated from the total RNA of Ca9-22 and HSC-4 cells was used to obtain a standard curve for each gene.

Ca9-22 and HSC-4 cells were plated at a density of 2×10⁴ cells/ml in culture dishes 60 mm in diameter (5 ml/dish) and were transfected with siRNAs 24 h after seeding. Cells were collected 2 days after transfection with siRNAs. For the analysis of cell cycle-related proteins, cells were treated with 100 ng/ml nocodazole 2 days after the transfection and collected 0, 3, 6, 9 and 12 h after nocodazole addition. Cells were lysed in RIPA buffer containing a protease and phosphatase inhibitor cocktail (Nacalai Tesque, Inc.), before passing through a 1-ml syringe with a 27-G needle. Protein concentrations in the lysates were measured using a Bio-Rad DC kit (Bio-Rad Laboratories, Inc.). The lysates containing 10 µg protein were separated by SDS-PAGE on 4–12% gels, followed by electroblotting onto Immobilon-P membranes (MilliporeSigma). Membranes were blocked with Blocking-one (Nacalai Tesque, Inc.) overnight at 4°C and incubated with anti-TFAP2E (cat. no. 29-175; ProSci, Inc.; 1:1,000), anti-Cyclin B1 (D-11; cat. no. sc-7393; Santa Cruz Biotechnology, Inc.; 1:1,000), anti-histone H3 (cat. no. 9715; Cell Signaling Technology, Inc.; 1:1,000), anti-p-H3Ser28 (cat. no. 9713; Cell Signaling Technology, Inc.; 1:1,000), anti-GAPDH (cat. no. ab9485; Abcam.; 1:1,000) and anti-TP53 (DO-1; cat. no. sc-126; Santa Cruz Biotechnology, Inc.; 1:1,000) antibodies at 4°C. After incubation for 24 h, the membranes were washed in Tris-buffered saline (TBS) containing 0.1% Tween 20 (TBS-T), followed by incubation with the appropriate horseradish peroxidase-conjugated secondary antibodies (cat. nos. NA934V and NA931V; Cytiva; 1:2,000) for 1 h at room temperature. The membranes were washed extensively with TBS-T and the results were visualized using Chemi-Lumi-One Super (Nacalai Tesque, Inc.) and an ImageQuant™ 800 system (Danaher Corp.). Experiments were performed at least three times and representative blots are shown. The band density of histone H3 (total H3) and p-H3Ser28 was semi-quantified using ImageJ (ver. 1.48; National Institutes of Health) (24) to normalize signal intensity of p-H3Ser28 to total H3.

Statistical analyses to examine the significance of the differences between two groups were performed using Student's unpaired t-test. One-way ANOVA followed by post-hoc Tukey's test was performed to examine the significance of the differences among multiple groups. All statistical analyses were performed using JMP software ver. 11.2 (SAS Institute, Inc.). Data are presented as the mean ± SD of at least three independent experiments. In all analyses, P<0.05 was considered to indicate a statistically significant difference.

Kaplan-Meier survival analysis of TCGA public microarray data sets was performed for 322 cases, including 128 cases of OSCC in the oral tongue, 27 in the base of the tongue, 62 in the floor of the mouth, 22 in the buccal mucosa, 7 in the hard palate, 3 in the lips and 73 in the oral cavity. The patients were divided to low or high expression groups using the median expression level of TFAP2E as a cutoff. Kaplan-Meier analysis showed that low TFAP2E expression was significantly associated with a shorter overall survival in patients with OSCC (Fig. 1). These results strongly suggested that TFAP2E plays a suppressive role in the malignant progression of OSCC.

To assess the hypothesis that TFAP2E acts as a tumor suppressor in OSCC, the present study examined the effects of TFAP2E knockdown on human OCSS-derived Ca9-22 and HSC-4 cells using anti-TFAP2E siRNA. RT-qPCR and western blotting confirmed that Ca9-22 and HSC-4 cells transfected with anti-TFAP2E siRNA exhibited reduced endogenous TFAP2E expression compared with in the control group (Fig. 2A and B). The standard WST-8 cell survival assay showed that TFAP2E knockdown resulted in a marked increase in cell proliferation (Fig. 2C and D). Consistent with these results, the number of viable TFAP2E-depleted cells was substantially greater than that in the control group (Fig. 2E and F).

The present study also examined whether TFAP2E knockdown affected cell proliferation or cell cycle progression by analyzing the cell cycle distribution of Ca9-22 and HSC-4 cells transfected with anti-TFAP2E siRNA or control siRNA using FACS. There were no significant differences in either cell cycle distribution or the proportion of dead cells, as indicated by sub-G₁ DNA content, between TFAP2E-knockdown and control cell groups (Fig. 3). In addition, knockdown of TFAP2E had a negligible effect on the resistance of the cells to CDDP and H₂O₂ (Fig. S1), indicating that knockdown of TFAP2E may accelerate cell growth via some mechanism other than augmenting stress resistance in the cells.

To investigate the mechanisms by which TFAP2E depletion promotes cell proliferation, the present study analyzed the cell cycle progression rate. Because anti-TFAP2E siRNA did not work when cells were treated with a double thymidine block to achieve synchronization to the late G₁ phase (data not shown), cell cycle progression was analyzed using the previously reported method for asynchronous cells (25). In this method, cell cycle progression was monitored in the presence of nocodazole. As nocodazole prevent cells from completing mitosis, cell cycle progression rate could be analyzed by measuring the accumulation rate of G₂/M cells. FACS analysis showed that the proportion of cells in the G₀/G₁ phase decreased and that of cells in the G₂/M phase increased in a time-dependent manner (Figs. 4A, B and S2); however, there were few significant differences between control and TFAP2E-knockdown cells. The G₂ and M phases could not be distinguished based on the DNA content of the cells, since cells in both G₂ and M phase have 4N DNA content (tetraploidy); therefore, the present study also monitored the population of cells positive for p-H3Ser28, which is a reliable marker of the early M phase (26). The proportion of p-H3Ser28-positive cells increased over time, and the rate of increase was significantly higher in TFAP2E-depleted Ca9-22 cells compared with that in the control group (Figs. 4C and S3A); similar results were observed in HSC-4 cells (Figs. 4D and S3B). Consistent with these results, western blotting demonstrated that the accumulation of p-H3Ser28 and cyclin B1, an alternative molecular marker for the G₂/M phase, occurred earlier in TFAP2E-knockdown cells than in control cells (Fig. 5). Collectively, these results suggested that TFAP2E may serve a role in regulating the G₂/M transition in OSCC cells. As a tumor suppressor gene TP53 is a key molecule in regulation of the G₂/M transition; therefore, the present study examined whether depletion of TFAP2E increased cell proliferation via suppressing TP53 expression. The data showed that knockdown of TFAP2E resulted in downregulation of TP53 at the protein level, but not at the mRNA level, in both Ca9-22 and HSC-4 cells (Fig. S4A and B). However, knockdown of TP53 using siRNA suppressed, rather than increased, the viability of both cells (Fig. S4C-F). These results indicate that accelerated proliferation of TFAP2E-knockdown cells could not owe to downregulation of the TP53 protein.