LSCC tissue samples and adjacent normal tissues were collected from 45 patients with LSCC at the Otorhinolaryngology Head and Neck Surgery Biobank of Hebei Medical University (Shijiazhuang, China) from October, 2016 to March, 2018. Informed consent was obtained from all patients, none of whom had received chemotherapy or radiotherapy prior to surgery. The use of human tissues specimens was approved by and carried out according to the guidelines of the Ethics Committee of the Second Hospital of Hebei Medical University (Shijiazhuang, China). One part of the tissue specimens was placed into RNAlater solution (CoWin Biosciences, Beijing, China) and stored at -80°C for RNA extraction. The other part of the tissue specimens was fixed in 10% neutral formaldehyde solution, and paraffin blocks were routinely prepared and preserved. Tumor and normal adjacent tissues were confirmed by routine pathological diagnosis. Agilent SBC Human (4*180K) lncRNA Microarray (ID: 74348) was used to test the transcript expression profiles in 4 pairs of LSCC and normal tissues. The clinicopathological characteristics of the 45 paired specimens are presented in Table SI.

Three human LSCC cell lines (TU686, TU177 and AMC-HN-8) and 293T cells were purchased from BNBIO (Beijing, China) and preserved at the Otorhinolaryngology Head and Neck Surgery Biobank of Hebei Medical University. The TU686 and TU177 cells were cultured in RPMI-1640 medium (Gibco/Thermo Fisher Scientific, Inc., Waltham, MA, USA), supplemented with 10% fetal bovine serum (FBS; Gibco/Thermo Fisher Scientific, Inc.). The AMC-HN-8 and 293T cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco/Thermo Fisher Scientific, Inc.) supplemented with 10% FBS. The TU177 cells were treated with 10 ng/ml recombinant TGF-β (R&D Systems, Inc., Minneapolis, MN, USA) for 7 days and the medium was replenished every 2 days. All the cells were cultured at 37°C in a humidified 5% CO₂ incubator (Thermo Fisher Scientific, Inc.).

Total RNA was extracted from the tissues and cells using the the Eastep^®Super Total RNA Extraction kit (Promega, Madison, WI, USA), and the RNA integrity was evaluated by 1% agarose gel electrophoresis (containing DEPC; Bio-Rad Laboratories, Inc., Hercules, CA, USA). cDNA was synthesized using the Transcriptor First Strand cDNA Synthesis kit (Roche, Basel, Switzerland) following the manufacturer’s protocol. RT-qPCR was performed with GoTap^®qPCRMaster Mix (Promega) according to the manufacturer’s instructions using CFX96™ Real-Time PCR Detection Systems. The reaction conditions were as follows (two-step method): One cycle of pre-denaturation for 2 min at 95°C, followed by 40 cycles of 15 sec at 95°C for denaturation, and for annealing to extension, selecting the most suitable annealing temperature according to different primers for 60 sec. The relative expression levels were estimated with the 2^-ΔΔCq method. Each specimen was tested 3 times. The specific primer sequences are listed in Table SII.

Subcellular fractionation was performed with the PARIS™ Kit Protein and RNA Isolation System (Invitrogen/Thermo Fisher Scientific, Inc.) from LSCC cell lines according to the manufacturer’s instructions.

The 4 shRNAs specifically targeting MIR155HG and the control sh-NC were synthesized by Vigene Biosciences (Shandong, China). The pcDNA3.1-MIR155HG vector was synthesized by Sangon Biotech (Beijing, China). hsa-miRNA-155-5p mimic/inhibitor/negative controls were purchased from GenePharma (Shanghai, China) (Table SII). The cells were grown on 6-well plates and transfected using Lipofectamine 2000 (Invitrogen/Thermo Fisher Scientific, Inc.) when they reached 80% confluence, according to the manufacturer’s instructions. At 48 h following transfection, the cells were collected for RT-qPCR verification and further experiments.

Cell proliferation was detected by MTS assay. Cells were seeded in 96-well plates at 2×10³ per well following transfection for 24 h. Cell proliferation was evaluated using the CellTiter96^®AQueousOne Solution Cell Proliferation Assay kit (Promega) according to the manufacturer’s instructions. MTS reagent (20 μl) was added to 100 μl culture medium after seeding for 0, 24, 48, 72 and 96 h, and incubated in CO₂ incubator at 37°C for 2.5 h each time. The absorbance at 490 nm was measured using a Spark^® multimode microplate reader (Tecan, Männedorf, Switzerland).

The cells were seeded in 6-well plates at a density of 2×10³ per well following transfection for 24 h. After 12-14 days, the cells were washed twice with phosphate-buffered saline, fixed with paraformaldehyde and stained with 0.5% crystal violet solution at room temperature for 20 min. (Generay, Shanghai, China). The number of colonies (≥50 cells) was counted under a microscope (Olympus, Tokyo, Japan).

For migration assays, non-Matrigel-coated chambers (Corning Costar, Corning, NY, USA) with 8-μm pore membranes were used. A total of 1×10⁵ cells per well were seeded into the upper chamber and 650 μl medium (10% FBS) was added to the lower chamber. Following incubation in CO₂ incubator (37°C, 5% CO₂) (Thermo Fisher Scientific, Inc.) for 24 h, the residual cells on the upper surface of the membrane were removed, the membrane was fixed in 4% paraformaldehyde for 20 min, stained with 0.5% crystal violet solution for 20 min and examined under a microscope (CKX53; Olympus, Tokyo, Japan) in 5 randomly selected fields. For invasion assays, the upper chamber was pre-coated with 50 μl 1X Matrigel^® Basement Membrane Matrix (Corning Costar); the remaining steps were as described above for the migration assay. For the Transwell assays of TGF-β-treated and untreated cells, 5×10⁴ cells were seeded into the upper chamber.

A total of 16 male nude (BALB/c-nude) mice (age, 5 weeks; weight, 20-25 g) were obtained from Beijing Vital River Laboratory Animal Technology Co., Beijing, China; animal permit no.: [SCXY (Jing) 2016-0006]. The maintenance conditions for the mice were as follows: Temperature, 23-27°C; humidity, 40-60%; ventilation, 15 times/h; 12 h light/12 h dark cycle. The mice were fed standard laboratory food and water. The health and maintenance condition of the mice were monitored every 2 days. The mice were randomly divided into 2 groups (pcDNA3.1, and pcDNA3.1-MIR155HG; n=8 per group). The TU177 cells (5×10⁶) stably transfected with pcDNA 3.1 and pcDNA3.1-MIR155HG in 100 μl PBS were subcutaneously injected into the right flanks of the mice. Tumor volume was measured using a caliper at indicated time points and calculated using the following formula: V = 0.5 × length x width². After 4 weeks, the mice were sacrificed by cervical dislocation and the tumors were removed and weighed. The maximum diameter of a single tumor found was 17 mm and no mouse developed multiple tumors. All mice were in well body condition throughout the experiment. All animal experiments were performed at the Experimental Animal Center of the Fourth Hospital of Hebei Medical University according to the NIH guidelines, and were approved by the Institutional Animal Care and Use Committee of the Fourth Hospital of Hebei Medical University.

The potential downstream targets of miR-155-5p were predicted using TargetScan (http://www.targetscan.org/vert_72/) and miRwark (http://mirwalk.umm.uni-heidelberg.de/). SOX10 was predicted by the two databases and considered a potential target of miR-155-5p. The 3′-UTR of SOX10 mRNA containing the intact miRNA-155-5p recognition sequences was PCR-amplified and subcloned into the Nhel and Xbal sites of the pmirGLO vector (Youbio, Changsha, China). Mutation sites were promoted using the Site-Directed Mutagenesis kit (New England Biolabs, Inc., Ipswich, MA, USA). Mutation primers were designed in NEBase changer (http://nebasechanger.neb.com/) and synthesized by Sangon Biotech (Table SII).

Cell protein lysates were prepared using RIPA buffer (Solarbio, Beijing, China) followed by the addition of protease inhibitor cocktail (Promega) and PMSF (Solarbio), according to the manufacturers’ instructions. The protein concentration was measured using the BCA Protein Assay kit (Generay). Proteins were separated by 10% SDS-PAGE and then transferred to PVDF membranes (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The membranes were incubated with 5% non-fat milk (Becton-Dickinson and Company, Suzhou, China) for 2 h at room temperature and incubated overnight at 4°C with rabbit-anti-human SOX10 (molecular weight: 60 kDa) (1:2,000; cat. no. DF8009; Affinity, Cincinnati, OH, USA) and rabbit-anti-human GAPDH (molecular weight: 37 kDa) (1:5,000; cat. no. 10494-1-AP; Protein Tech Group, Inc., Chicago, IL, USA) antibodies. Subsequently, the membranes were incubated at room temperature with IgG (H+L) HRP (1:10,000; cat. no. RS0002; Ruiying Bio, Suzhou, China) for 1 h. Proteins were visualized with an enhanced ECL reagent (Vazyme, Nanjing, China) by ChemiDoc™ XRS+ (Bio-Rad Laboratories, Inc.).

The TU177 and 293T cells were seeded at a density of 3×10⁴ cells per well in 24-well plates. After 24 h, the cells were co-transfected with pmirGLO-SOX10-WT or pmirGLO-SOX10-MUT reporter plasmids, and miR-155-5p mimic or inhibitor or NC. Following transfection for 48 h, the relative luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s protocol. The luciferase activity of each well was normalized to Renilla luciferase activity.

The results were analyzed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism version 7 (GraphPad Software, Inc., La Jolla, CA, USA). The differences between 2 groups were analyzed with the Student’s t-test. Differences between >2 groups were determined by one-way ANOVA followed by Tukey’s post hoc test. Differences between the growth of different groups, two-way ANOVA was performed and followed by Tukey’s post hoc test. Pearson’s correlation analysis was used to analyze the correlation between the expression of MIR155HG and that of miR-155-5p, and between the expression of miR-155-5p and SOX10. Each experiment was performed at least 3 times. All data are presented as the means ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

To determine the potential role of lncRNAs in LSCC, microarray analysis was used to compare lncRNA expression levels between LSCC tissues and corresponding normal tissues. There were 3,073 differentially expressed lncRNAs in total (fold change ≥2, P<0.05), of which 1,967 were upregulated and 1,106 were downregulated in LSCC tissues compared with corresponding normal tissues (Fig. 1A). A miRNA host gene, MIR155HG, which was found to be upregulated in the microarray analysis, was selected for further analysis. To investigate the functional role of MIR155HG in LSCC, the expression of MIR155HG was initially analyzed in 45 pairs of LSCC tissues and corresponding adjacent non-tumor tissues by RT-qPCR. The expression level of MIR155HG was found to be significantly increased in the LSCC tissues compared with the adjacent normal tissues (Fig. 1B). The expression levels of MIR155HG in 3 LSCC cell lines (TU686, TU177 and AMC-HN-8) are shown in Fig. 1C. The association between MIR155HG expression and different clinicopathological characteristics in LSCC was further analyzed. High expression levels of MIR155HG were found to be closely associated with poor differentiation, lymph node metastasis and a more advanced TNM stage. However, there was no observed association between the expression of MIR155HG and age, alcohol consumption or smoking (Fig. 1D).

MIR155HG is located on chromosome 21q21.2, and contains 3 exons and 2 introns. miR-155-5p is located within the third exon of MIR155HG (Fig. 1E). The coding potential of MIR155HG was analyzed by using several prediction software programs, and the results revealed that the open reading frame (ORF) of MIR155HG was very short, containing only 82 amino acids (https://www.ncbi.nlm.nih.gov/orffinder/), and had no coding potential (http://lilab.research.bcm.edu/cpat/) (Fig. 1F and G). These results indicated that MIR155HG is a non-coding RNA. Furthermore, MIR155HG was found to be predominantly located in the nucleus of LSCC cells by subcellular fractionation assay (Fig. 1H).

To explore the biological function of MIR155HG, shRNAs against MIR155HG (sh-MIR155HG) were transfected into the LSCC cells to knock down endogenous MIR155HG expression. Due to the poor transfection efficiency, the TU686 cells, which had a relatively higher expression level of MIR155HG, were not selected for the in vitro assay. The TU177 and AMC-HN-8 cells, with a relatively higher transfection efficiency, were used as model cells for the follow-up experiments. sh-MIR155HG-1, which exhibited the most evident knockdown efficacy, was selected for use in further experiments (Fig. 2A).

MTS and colony formation assays were used to measure cell proliferation. MIR155HG knockdown suppressed TU177 and AMC-HN-8 cell proliferation (Fig. 2B). Consistent with the results of MTS assay, the results of colony formation assay also revealed that the downregulation of MIR155HG suppressed the colony-forming ability of the TU177 and AMC-HN-8 cells (Fig. 2C). These data provide evidence to suggest the tumor growth-promoting role of MIR155HG in vitro.

In order to examine the effects of MIR155HG on the malignant biological properties of LSCC cells, the cell migratory and invasive abilities were evaluated using Transwell Matrigel migration and invasion assays. The migration and invasion capacities of the TU177 and AMC-HN-8 cells were found to be markedly decreased when MIR155HG was knocked down (Fig. 2D and E). Taken together, these data demonstrated that MIR155HG suppressed the proliferation, migration and invasion of LSCC cells.

Subsequently, in order to assess the biological effects of MIR155HG overexpression, the pcDNA3.1-MIR155HG expression plasmid was transfected into the TU177 and AMC-HN-8 cells with relatively low expression levels of MIR155HG. pcDNA3.1 was used as a control. The relative expression level of MIR155HG in the pcDNA3.1-MIR155HG-transfected TU177 and AMC-HN-8 cells was markedly increased (Fig. 3A). MTS (Fig. 3B) and colony formation assays (Fig. 3C) demonstrated that MIR155HG upregulation promoted TU177 and AMC-HN-8 cell proliferation. Moreover, MIR155HG overexpression enhanced the migratory and invasive capacity of both cell lines (Fig. 3D and E). All these results indicated that the overexpression of MIR155HG increased the tumorigenicity of LSCC cells. To further verify the results of the in vitro experiments, animal experiments were conducted. A TU177 cell line stably transfected with pcDNA3.1 and pcDNA3.1-MIR155HG was constructed, and 5×10⁶ cells were subcutaneously injected into nude mice. The results demonstrated that the overexpression of MIR155HG significantly promoted tumor growth compared with the pcDNA3.1 group (Fig. 3F and G), whereas the volume and weight of the tumors were higher in the MIR155HG overexpression group (Fig. 3H and I).

As MIR155HG is the host gene of miR-155-5p, we further investigated whether miR-155-5p exerts the same effects as MIR155HG on LSCC. The expression of miR-155-5p was found to be significantly increased in the LSCC tissues compared with the adjacent normal tissues (Fig. 4A). The high expression level of miR-155-5p was found to be closely associated with the TNM stage, lymph node metastasis and poor differentiation. However, there was no association between the expression of miR-155-5p and age, alcohol consumption, or smoking (Fig. 4B).

To further determine whether miR-155-5p is associated with the progression of LSCC, miR-155-5p mimics and inhibitor were transfected into the TU177 and AMC-HN-8 cells (Fig. 4C). As shown in Fig. 4D-F, the overexpression of miR-155-5p markedly promoted the proliferation, migration and invasion of TU177 and AMC-HN-8 cells, whereas the downregulation of miR-155-5p markedly inhibited cell proliferation, migration and invasion. Collectively, these data suggest that miR-155-5p acts as an oncogenic miRNA by promoting LSCC cell proliferation, migration and invasion.

The expression level of miR-155-5p was found to highly correlate with the expression of MIR155HG (Fig. 5A). These findings suggested that miR-155-5p and its host gene, MIR155HG, may be co-expressed in LSCC. The downregulation of MIR155HG reduced the expression levels of miR-155-5p in the TU177 and AMC-HN-8 cells (Fig. 5B). However, the expression of MIR155HG was not affected by miR-155-5p knockdown or overexpression in LSCC cells (Fig. 5C). Overexpression of miR-155-5p may abrogate the inhibitory effect of MIR155HG knockdown on cell proliferation, migration and invasion (Fig. 5D-F). These results indicated that miR-155-5p may act synergistically with MIR155HG to promote LSCC progression.

To further explore the mechanisms underlying the effects of miR-155-5p on the prolif eration, migration and invasion of LSCC cells, bioinformatics tools, such as TargetScan and miRwalk, were used to search for potential target genes of miR-155-5p. Among these target genes, we focused on SOX10, a tumor suppressor gene that is involved in suppressing tumorigenicity (22) and EMT (23). TargetScan prediction revealed that the 3′UTR of SOX10 contains a conserved binding site for miR-155-5p (Fig. 6A). The expression level of SOX10 was markedly decreased in the LSCC tissues compared with the corresponding adjacent non-tumor tissues (Fig. 6B). The expression level of miR-155-5p negatively correlated with the expression of SOX10 in LSCC (Fig. 6C). The downregulation of miR-155-5p by transfection with miR-155-5p inhibitor markedly increased the transcriptional level and protein expression of SOX10, while the overexpression of miR-155-5p significantly decreased the transcriptional level and protein expression of SOX10 in LSCC cell lines (Fig. 6D and E). To confirm whether the SOX10 mRNA expression is regulated by miR-155-5p through direct binding to the 3′UTR of SOX10, a dual luciferase reporter assay was performed in the TU177 and 293T cells. The luciferase signal was reduced in the miR-155-5p mimic and pmirGLO-SOX10-3′UTR-wild-type co-transfected cells compared with the control group. However, co-transfection of miR-155 mimic and pmirGLO-SOX10-3′UTR-mutant-type did not reduce the luciferase signal compared with the control group. The luciferase signal was increased in the miR-155-5p inhibitor and pmirGLO-SOX10-3′UTR-wild-type, but not the mutant-type, co-transfected cells compared with the control group (Fig. 6F).

To investigate whether MIR155HG and miR-155-5p regulate the migration and invasion of LSCC cells via EMT, the TU177 cells were treated with 10 ng/ml of TGF-β for 7 days. The cells treated with TGF-β exhibited a change in morphology and acquired a spindle-shaped morphology (Fig. 7A). Furthermore, their migratory and invasive abilities were enhanced compared with the untreated cells (Fig. 7B). Subsequently, the molecular markers of EMT were analyzed by RT-qPCR. The levels of the mesenchymal markers, N-cadherin, Twist, Zeb1, Snail and vimentin, were found to be upregulated. However, we did not detect the expression of the epithelial marker, E-cadherin (after using several common primers of E-cadherin), possibly due to the fact that the expression abundance was too low to be monitored (Fig. 7C). These results suggested that the TU177 cells treated with TGF-β exhibited EMT-related characteristics. Moreover, to further determine whether MIR155HG and miR-155-5p were involved in TGF-β-induced EMT, their expression was detected by RT-qPCR analysis. The results indicated that the levels of MIR155HG and miR-155-5p were upregulated after the TU177 cells were treated with TGF-β (Fig. 7D). The knockdown of MIR155HG inhibited the expression of the mesenchymal markers, N-cadherin, vimentin, Twist and Zeb1, but did not affect the expression of Snail (Fig. 7E). miR-155-5p overexpression promoted the expression of the mesenchymal markers, N-cadherin, vimentin, Twist and Snail, but did not increase the expression of Zeb1 (Fig. 7F), whereas it partially reversed the inhibitory effect exerted by MIR155HG knockdown (Fig. 7G). Taken together, the above-mentioned results indicate that MIR155HG and miR-155-5p are EMT-related non-coding RNAs, and they synergistically promote EMT in LSCC cells.