OSCC cell lines (Cal-27 and SCC-9; American Type Culture Collection) were maintained in DMEM (Gibco; Thermo Fisher Scientific, Inc.), supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) and antibiotics (100 U/ml penicillin and 100 µg/ml streptomycin) in an incubator at 37°C with 5% CO₂.

Small interfering (si)RNAs targeted against DLEU1 (DLEU1 siRNA1, DLEU1 siRNA2 and DLEU1 siRNA3; 10 nM) and a negative control (NC) siRNA (siRNA-NC) were purchased from Guangzhou RiboBio Co., Ltd., and were transfected into OSCC cells (5×10³) using Lipofectamine^® 2000 (Thermo Fisher Scientific, Inc.). Then, cells were incubated at 37°C for 6 h and the transfection efficiency was determined via reverse transcription-quantitative PCR (RT-qPCR). After 24 h of incubation, transfected cells were used for subsequent experiments. The siRNA sequences were as follows: siRNA-NC, 5′-UUCUCCGAACGUGUCACGUTT-3′; DLEU1 siRNA1, 5′-GGAAUGAAGCAACUGAGAUUU-3′; DLEU1 siRNA2, 5′-GGGTTACGATTGCCCAGAT-3′; and DLEU1 siRNA3, 5′-CGTTAAGGTTCCGGACGAC-3′.

The NC, miR-149-5p agomir and miR-149-5p antagomir were synthesized and obtained from Shanghai GenePharma Co., Ltd. Cells (5×10³) were transfected with 10 nM control agomir (NC), miR-149-5p agomir or miR-149-5p antagomir for 24 h with Lipofectamine 2000 at 37°C. After 24 h of transfection, transfected cells were used for subsequent experiments. Meanwhile, the concentration of miR-149-5p agomir/antagomir (50 nM) was selected according to the instructions of the manufacturer. The sequences were as follows: miR-149-5p agomir, 5′-UCUGGCUCCGUGUCUUCACUCCC-3′; miR-149-5p antagomir, 5′-GGGAGUGAAGACACGGAGCCAGA-3′; and NC, 5′-UUCUCCGAACGUGUCACGUTT-3′.

In total, 10 pairs of OSCC samples and adjacent normal tissues (~2 cm from tumor) were collected from the Affiliated Stomatological Hospital of Guizhou Medical University (Guiyang, China) between April 2018 and April 2019. The patients were informed of the purpose of the experiments and provided written informed consent. The clinical information of the patients is listed in Table I. Patients were diagnosed with OSCC following the American College of Rheumatology classification criteria for the diagnosis of OSCC (20). Meanwhile, the samples were used for investigation of DLEU1 and CDK6 levels. The present study was approved by the Institutional Ethical Committee of Affiliated Stomatological Hospital of Guizhou Medical University (approval no. ASHGMU20190523).

Total RNA was extracted from OSCC cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Total RNA was reverse transcribed into cDNA using the PrimeScript RT reagent kit (Takara Bio, Inc.) according to the manufacturer's instructions. Then, RT-qPCR was performed using a SYBR^® Premix Ex Taq™ II kit (Takara Bio, Inc.) on a 7900HT system (Applied Biosystems; Thermo Fisher Scientific, Inc.) according to the following conditions: 60°C for 1 min, 90°C for 15 min, followed by 40 cycles of application at 90°C for 15 sec and 55°C for 60 sec. The primers used were as follows: U6 forward, 5′-CGCTTCGGCAGCACATATAC-3′ and reverse, 5′-AAATATGGAACGCTTCACGA-3′; DLEU1 forward, 5′-GGTCCACGGCACGTTAACA-3′ and reverse, 5′-CCAATTGAAGGCCTTAAGG-3′; miR-149-5p forward, 5′-TGCGCTAGCAGCGGGAACAGTTC-3′ and reverse, 5′-CCAGTGCAGGGTCCGAGGTATT-3′; and β-actin forward, 5′-GTCCACCGCAAATGCTTCTA-3′ and reverse, 5′-TGCTGTCACCTTCACCGTTC-3′. The relative expression level was quantified by normalizing to β-actin or U6 using the 2^−ΔΔCq method (21).

The survival curve was calculated based on the data from The Cancer Genome Atlas (TCGA). The result shown here is in part based upon data generated by the TCGA Research Network (https://www.cancer.gov/tcga). In addition, the data from TCGA were analyzed using Gene Expression Profiling Interactive Analysis (GEPIA) as previously described (22).

The potential downstream miRNA of DLEU1 was predicted using StarBase (http://starbase.sysu.edu.cn/index.php). In addition, the target mRNA of miR-149-5p was predicted using TargetScan (http://www.targetscan.org/vert_72/), miRDB (http://www.mirdb.org/) and miRWalk (http://mirwalk.umm.uni-heidelberg.de/).

SCC-9 or Cal-27 cells (5.0×103 cells/well) were transfected with siRNA-NC, DLEU1 siRNA2 or DLEU1 siRNA3 for 0, 24, 48 or 72 h. Then, 10 µl CCK-8 reagent (Beyotime Institute of Biotechnology) was added into each well, and the plate was incubated for 2 h at 37°C. The absorbance was detected at 450 nm using a microplate reader (Bio-Rad Laboratories, Inc.).

OSCC cells were plated onto a 96-well plate at a density of 5.0×10³ cells/well. Following incubation, cells were transfected with siRNA-NC or DLEU1 siRNA2 for 72 h. After that, cells were fixed with 4% paraformaldehyde for 20 min at room temperature and blocked with 2% bovine serum albumin (Beyotime Institute of Biotechnology) at room temperature for 30 min. Subsequently, cells were incubated with rabbit monoclonal antibody anti-Ki67 (1:100; cat. no. ab92742; Abcam) at 4°C overnight. Then, cells were incubated with anti-rabbit IgG secondary antibody (1:1,000; cat. no. ab150077; Abcam) for 1 h at room temperature. The results were observed under a fluorescence microscope (Olympus BX53; Olympus Corporation).

For the cell migration assay, 2×10⁵ OSCC cells were seeded into the upper chambers of the 24-well plates in 200 µl serum-free DMEM (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 0.2% bovine serum albumin (Beyotime Institute of Biotechnology). The lower chambers contained RPMI-1640 medium supplemented with 1% FBS. After incubation for 24 h at 37°C, the non-migrating cells were gently removed from the upper side of each chamber with a cotton swab, while the cells that had migrated were fixed with 95% alcohol for 10 min at room temperature and stained with 0.1% crystal violet (Sigma-Aldrich; Merck KGaA) for 5 min at room temperature. Finally, cells were counted under an inverted light microscope (Olympus Corporation) at ×400 magnification.

For the cell invasion assay, the upper chamber was pre-treated with 100 µl Matrigel (BD Biosciences) for 4 h at 37°C. Subsequently, transfected OSCC cells (2×10⁵) were seeded into the upper chambers with serum-free DMEM, while the media in the lower chambers was supplemented with 1% FBS. After incubation for 24 h, the Transwell chambers were fixed with 95% alcohol for 10 min at room temperature and stained with 0.1% crystal violet for 5 min at room temperature. The Transwell chamber was observed and the invaded cells were counted under a microscope.

Cells were trypsinized and resuspended in binding buffer. Then, cells were stained with 5 µl Annexin V-FITC (BD Biosciences) and propidium iodide (PI; BD Biosciences) in the dark at 37°C for 30 min. Flow cytometry (FACScan™; BD Biosciences) was applied to analyze the apoptosis rate using CellQuest™ software (version 5.1; BD Biosciences).

OSCC cells were lysed in RIPA lysis buffer (Nanjing KeyGen Biotech Co., Ltd.), and the protein concentration was determined using a BCA Assay kit (Beijing Solarbio Science & Technology Co., Ltd.). Proteins (30 µg per lane) were separated via SDS-PAGE (10% gel), and separated proteins were then transferred onto PVDF membranes (Thermo Fisher Scientific, Inc.). The membranes were blocked in 5% non-fat dried milk in TBS with Tween-20 (10%) for 1 h at room temperature. Then, the PVDF membranes were incubated at 4°C overnight with the following primary antibodies: Anti-Bax (cat. no. ab182733; 1:1,000), anti-cleaved caspase-3 (cat. no. ab49822; 1:1,000), anti-Bcl-2 (cat. no. ab196495; 1:1,000), anti-CDK6 (cat. no. ab131469; 1:1,000), anti-CDK4 (cat. no. ab68266; 1:1,000), anti-cyclin-dependent kinase inhibitor p27 (p27 Kip1; cat. no. ab32034; 1:1,000) and anti-β-actin (cat. no. ab8226; 1:1,000). Then, the membrane was incubated with horseradish peroxidase-labeled goat anti-rabbit secondary antibody (cat. no. ab7090; 1:5,000) at room temperature for 1 h. All antibodies were obtained from Abcam. Enhanced chemiluminescence reagent (Thermo Fisher Scientific, Inc.) was used to visualize the protein bands. ImageJ software (version 2.0; National Institutes of Health) was used to quantify the intensity of the bands.

CDK6 3′-untranslated region (3′UTR) containing the putative binding site of miR-149-5p (CDK6 WT 3′UTR) and the CDK6 3′UTR with the mutation binding site (CDK6 MT 3′UTR) were synthesized by Shanghai GenePharma Co., Ltd. The mutation was generated using a site directed mutagenesis kit (Promega Corporation). The partial sequences of DLEU1 were synthesized by Shanghai GenePharma Co., Ltd. These were then cloned into the vectors (pmirGLO; Promega Corporation). Subsequently, DLEU1 (WT/MT) and CDK6 (WT/MT) were transfected into OSCC cells using Lipofectamine 2000. After 48 h of transfection, the relative luciferase activity was detected using the Dual-Luciferase Reporter kit (Promega Corporation). The data were normalized to Renilla luciferase activity.

Probe-control or probe-DLEU1 was transcribed and labeled with a Biotin RNA Labeling Mix (Roche Diagnostics). Cells were lysed with Poly-lysis buffer (ELK bioscience), washed with PBS and centrifuged at 1000 × g for 5 min at 4°C. Secondary structure formation in the biotin-labeled RNAs was induced with RNA structure buffer (Thermo Fisher Scientific, Inc.). Streptavidin beads (75 µl; Thermo Fisher Scientific, Inc.) were washed and incubated overnight. After that, streptavidin bead-RNA complexes were obtained by separating the mixture. Then, cell lysates (5×10⁷ cells) were added to the complexes and incubated for 1 h. Following incubation, the mixture was separated again, and the supernatant of cell lysates was utilized to detect the enrichment of miR-149-5p. RT-qPCR was used to detect the enrichment of miR-149-5p.

Cell cycle distribution was assessed as previously described (23). Briefly, OSCC cells (5×10⁵) were harvested, fixed with 75% ethanol on ice for 20 min, permeabilized with 0.25% Triton X-100 and stained with Pharmingen PI/RNase (BD Biosciences). After incubation for 15 min at 4°C, cells were analyzed using a flow cytometer (BD FACSAria III; BD Biosciences) and ModFit (version 3.0; Verity Software House, Inc.). The data were quantified using FlowJo software (version 3.0; FlowJo, LLC).

Data are presented as the mean ± SD. The CCK-8 assay was performed five times. Immunofluorescence staining, RT-qPCR, western blotting, flow cytometry and Transwell invasion assays were repeated in triplicate. In addition, all other experiments were repeated three times. One-way analysis of variance and Tukey's post hoc tests were used for comparisons between ≥3 groups. A paired Student's t-test was used for comparisons between tumor tissues and adjacent normal tissues of the same patients, while an unpaired Student's t-test was used for comparisons between unpaired groups. P<0.05 was considered to indicate a statistically significant difference.

To detect the role of DLEU1 in OSCC, TCGA data were analyzed. As presented in Fig. 1A, DLEU1 level was negatively associated with the survival rate of patients with OSCC. In addition, DLEU1 was closely associated with patients with advanced stage OSCC (Fig. 1B). Furthermore, the expression of DLEU1 was significantly higher in OSCC tissues (Fig. 1C). Taken together, the expression of DLEU1 was upregulated in OSCC.

In order to evaluate the efficacy of cell transfection, RT-qPCR was performed. The data indicated that the expression of DLEU1 was significantly downregulated in OSCC cells when transfected with DLEU1 siRNAs (Fig. 2A and B). In addition, knockdown of DLEU1 significantly suppressed the cell viability of OSCC cells (Fig. 2C and D). DLEU1 siRNA2 had the most significant inhibitory effects on the expression of DLEU1, thus DLEU1 siRNA2 was used in subsequent investigations. Furthermore, the results of the Ki67 staining revealed that DLEU1 siRNA2 significantly inhibited the proliferation of OSCC cells (Fig. 2E and F), and it was revealed that Cal-27 cells were more susceptible to DLEU1 siRNA2 treatment. Thus, Cal-27 cells were selected for use in the subsequent analyses. Overall, silencing of DLEU1 significantly inhibited the proliferation of OSCC cells.

In order to investigate the underlying mechanism by which DLEU1 mediated the proliferation of OSCC cells, StarBase was used. As indicated in Fig. 3A, DLEU1 had a putative miR-149-5p targeting site. In addition, miR-149-5p is known to regulate the tumorigenesis of OSCC (19). Thus, miR-149-5p was selected for further analysis. The level of miR-149-5p in OSCC cells was upregulated by miR-149-5p agomir, but was downregulated by miR-149-5p antagomir (Fig. 3B). This result suggested that miR-149-5p agomir or antagomir was stably transfected into OSCC cells. Co-transfection of the WT-DLEU1 vector with miR-149-5p agomir resulted in a significant decrease in the luciferase activity when compared with the MT-DLEU1 vector (Fig. 3C). Furthermore, the expression of miR-149-5p in OSCC cells was significantly increased by the knockdown of DLEU1, while this was completely reversed by miR-149-5p antagomir (Fig. 3D). In addition, the results of the pull-down assay suggested that DLEU1 bound to miR-149-5p directly in OSCC cells (Fig. 3E). Overall, DLEU1 could sponge miR-149-5p in OSCC.

In order to further investigate the underlying mechanism by which DLEU1 regulated the progression of OSCC, flow cytometry was used. As indicated in Fig. 4A, DLEU1 siRNA2 significantly induced cell apoptosis, which was partially reversed by miR-149-5p antagomir (Fig. 4A). In addition, the migration and invasion of OSCC cells was significantly inhibited following knockdown of DLEU1, and this was also reversed by the miR-149-5p antagomir (Fig. 4B). Meanwhile, the expression levels of Bax and cleaved caspase-3 in OSCC cells were significantly increased by DLEU1 siRNA2 (Fig. 4C-E). By contrast, DLEU1 siRNA2 significantly inhibited the expression of Bcl-2 in OSCC cells (Fig. 4C and F). However, the effects of DLEU1 knockdown on these three proteins were significantly reversed by miR-149-5p antagomir (Fig. 4C-F). Knockdown of DLEU1 inhibited the tumorigenesis of OSCC via mediation of miR-149-5p.

In order to identify the downstream target of miR-149-5p, TargetScan, miRDB and miRWalk were used in the present study. As presented in Fig. 5A, CDK6 was demonstrated to be the potential target of miR-149-5p using these three online tools. The co-transfection of the WT-CDK6 vector with miR-149-5p agomir significantly decreased the luciferase activity when compared with the MT-CDK6 vector (Fig. 5B). In addition, the level of CDK6 was significantly higher in OSCC tissues, compared with adjacent normal tissues (Fig. 5C). Furthermore, the protein expression of CDK6 in OSCC cells was significantly inhibited by miR-149-5p agomir (Fig. 5D and E). In summary, miR-149-5p directly targeted CDK6 in OSCC cells.

In order to determine cell cycle distribution, flow cytometry was performed in the present study. The data revealed that DLEU1 knockdown significantly induced G1 cell cycle arrest in OSCC cells, which was partially rescued by miR-149-5p antagomir (Fig. 6A). In addition, the expression of CDK6 in OSCC cells was significantly inhibited by DLEU1 siRNA2, while this phenomenon was partially reversed by miR-149-5p antagomir (Fig. 6B and C). Meanwhile, the expression levels of CDK4 and p27 Kip1 in OSCC cells were not affected by the knockdown of DLEU1 or the addition of miR-149-5p antagomir (Fig. 7A-C). In summary, knockdown of DLEU1 induced G1 cell cycle arrest in OSCC cells via mediation of CDK6.