This article is mentioned in:

Introduction

Laryngeal squamous cell carcinoma (LSCC) is an aggressive form of head and neck malignancy (1). In spite of developments in treatment techniques, including surgery, chemotherapy and radiation therapy, the survival rate has remained poor over recent years (2,3). Therefore, a clearer comprehension of the progression of LSCC is urgently required in order to develop an effective therapeutic approach to address this type of cancer.

Recently, the function of non-coding RNAs (ncRNAs) in LSCC have been given extensive research focus (4). ncRNAs, which have no or limited protein-coding capacity, include short ncRNAs (miRNAs) and long ncRNAs (lncRNAs) (5). Research has indicated that numerous miRNAs are involved in LSCC progression and serve as tumor suppressors or oncogenes via binding to the 3′-untranslated regions 3′-UTRs of target mRNAs (6,7). lncRNAs, a family of transcripts measuring more than 200 nucleotides in length, have been reported to be involved in the occurrence and development of cancer (8). Numerous lncRNAs have been identified as vital players in the progression of LSCC via modulation of proliferation, the cell cycle, apoptosis, as well as the ability to invade, migrate and metastasize (9,10). Based on miRNAs and lncRNAs that are multifunctional in LSCC, certain miRNAs or lncRNAs could be utilized to serve as novel diagnostic markers and therapeutic agents for LSCC (11).

The X-inactive-specific transcript (XIST), one of the first discovered lncRNAs in mammals, was reported to be associated with cell differentiation, cell proliferation and genome maintenance in human cells (12). Accumulating evidence suggests an aberrant regulation of XIST in numerous different human malignancies, aside from being outlined in the initiation and development of tumors (13,14). The expression levels, functions and underlying molecular mechanisms that are associated with XIST in LSCC progression remain to be elucidated. The present study assessed the expression of XIST in samples from patients with LSCC, along with its significance at a clinical level, and an assessment of its function and underlying molecular mechanism.

Materials and methods

Human tissue samples

A total of 48 LSCC tissue samples along with the adjacent healthy tissue were sourced from patients (26 males and 22 females; mean age: 53.5±2.1 years) who were subjected to partial or total laryngectomy between March 2016 and April 2017 at the Department of Otorhinolaryngology-Head and Neck Surgery, The First Hospital of Jilin University (Changchun, Jilin, China). The patients that received radiotherapy, chemotherapy or other therapy were excluded from the present study. The samples were rapidly frozen in liquid nitrogen following surgery, and then taken to the laboratory under freezing conditions and stored at −80°C until use. Prior to the collection of samples, written informed consent was obtained from all patients. The use of tissue samples was approved by the Ethics Committee of Jilin University (Changchun, Jilin, China).

Culture of cells and transfection

The Cell Bank of the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China) was the source of human LSCC cell line TU212, and these cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) plus 10% fetal bovine serum (FBS; Hyclone; Thermo Fisher Scientific, Inc.), penicillin (100 U/ml) or streptomycin (100 µg/ml). Cells were incubated in a humidified incubator at 37°C and 5% CO2.

Short hairpin RNA (shRNA) directed against XIST (sh-XIST) and scrambled shRNA control (sh-NC) were synthesized and inserted into pGPH1/Neo. The vector and the following: miR-144 mimic (miR-144), scrambled miRNA negative control (miR-NC), miR-144 inhibitor (anti-miR-144), and scrambled inhibitor control (anti-miR-NC) were sourced at GenePharma Co., Ltd. (Shanghai, China). Transient transfection of TU212 cells was performed with one of the aforementioned mimics, inhibitors or plasmids using Lipofectamine® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The transfection efficiency was determined at 24 h after transfection. The selection of sh-XIST and sh-NC stable transfectants was performed with 800 µg/ml neomycin (Sigma-Aldrich; Merck KGaA).

RNA isolation and reverse transcription (RT) PCR analysis

TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) was used to extract total RNA from all samples and cell lines. The quality and concentration of RNA were assessed using a NanoDrop Spectrophotometer (ND-2000; Thermo Fisher Scientific, Inc.). PrimeScript™ RT reagent kit (Takara) or miRNA cDNA synthesis kit (CWBIO) was utilized for complementary DNA (cDNA) synthesis. The SYBR Premix Ex Taq II (Takara) or the miRNA qPCR Assay kit was used for quantitative (q)PCR with a 7900HTfast Real-time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). U6 was used as the control for miRNA; GADPH was used as the endogenous control for lncRNA/mRNA. The 2−ΔΔCq method was used to obtain the relative expression levels (15). Table I lists the primer sequences used in the present study.

Table I. Reverse transcription PCR primers used for mRNA expression analysis.

Table I.

Reverse transcription PCR primers used for mRNA expression analysis.

Target gene	Primer (5′-3′)
U6	F-TCCGATCGTGAAGCGTTC
	R-GTGCAGGGTCCGAGGT
miR-144	F-GGGAGATCAGAAGGTGATT
	R-GTGCAGGGTCCGAGGT
XIST	F-CTCTCCATTGGGTTCAC
	R-GCGGCAGGTCTTAAGAGATGA
IRS1	F-AGAACGAGAAGAAGTGGCGG
	R-GCCTTTGCCCGATTATGCAG
GAPDH	F-AAGGTGAAGGTCGGAGTCAA
	R-AATGAAGGGGTCATTGATGG

[i] F, forward; R, reverse. U6, RNA, U6 small nuclear 1; XIST, X inactive-specific transcript; IRS1, insulin receptor substrate 1; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Cell proliferation and colony formation assays

The proliferation of cells was assessed with a CellTiter96® Aqueous One Solution Cell Proliferation kit (MTS; Promega Corp.). Briefly, 1×103 transfected cells/well were seeded into 96-well plates. At the indicated time, the addition of 20 µl MTS was performed followed by incubation for 120 min at 37°C. Measurement of the absorbance at 490 nm was perceived using spectrophotometry (Synergy2; BioTek Instruments, Inc.).

For cell colony assay, 6-well culture plates were used to seed stable XIST-silenced TU212 cells at 500 cells/well and cultured at 37°C in DMEM with 10% FBS. After 10 days of culture, the colonies were fixed with ethanol (Sigma Aldrich; Merck KGaA) for 10 min and subjected to 0.1% crystal violet (Sigma Aldrich; Merck KGaA) staining for 30 min at 25°C. A light microscope (magnification ×200; Olympus Corp.) was used to capture images and count the colonies manually.

Apoptosis assay

For apoptosis assays, the harvesting of transfected cells was performed followed by staining with FITC-Annexin V and propulsion iodide (PI) for 10 min at 25°C using an Annexin V-FITC/PI kit (BD Pharmingen). Apoptosis was detected using a FACSCalibur (BD Biosciences). The apoptosis rate was assessed using Flowjo software 7.6.1 (Tree Star Corp).

Wound healing assay

Transfected cells were seeded in 24-well plates until the cell density reached >90%. Scratch wounds were created with the tip of a 100 µl pipette. In order to decrease the influence of the apoptotic rate, the cells were cultured in fresh serum-free medium at 37°C in 5% CO2 for 24 h. Images were captured at randomly selected fields at 0 and 24 h after the scratch was created using a light microscope (magnification ×200; Olympus Corp.).

Transwell invasion assay

Matrigel invasion assays were used for invasion. The upper chamber of the BioCoat Matrigel Invasion Chamber (BD Biosciences) was seeded with 1×105 transfected cells in 100 µl DMEM lacking serum. The lower chambers received a medium containing 10% FBS. Following incubation for 48 h, the cells that entered the lower chambers were subjected to fixation with 70% ethanol for 10 min and crystal violet staining (0.1%) for 15 min at 25°C. These cells were imaged and enumerated across five randomly selected fields with a light microscope (magnification, ×200).

Bioinformatics predictions and luciferase reporter assays

Putative binding sites of XIST and miR-144 were predicted using the starBase v2.0 database, a public algorithm (16). This putative binding site, wild-type (Wt), and its mutated (Mut) sequence were subjected to subcloning in a pmirGLO Dual-luciferase vector (Promega Corporation). This yielded Wt-XIST and Mut-XIST recombinant vectors, respectively. TU212 cells were transfected with the Wt/mut-XIST reporter vector and the miR-144/miR-NC mimics in a 24-well plate using Lipofectamine® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). Dual-Luciferase Reporter Assay system (Promega Corporation) was utilized to measure the activity of the reporter after 48 h while normalization was in reference to Renilla luciferase activity, according to the manufacturer's protocol.

Western blot analysis

RIPA lysis buffer (Beyotime Institute of Biotechnology) was utilized to extract total proteins whose concentration was estimated with a BCA Protein Assay kit (Beyotime Institute of Biotechnology). Of these extracted samples, 30 µg was loaded per lane and separated via SDS-PAGE (10% gel), followed by transfer to polyvinylidene fluoride (PVDF) membranes (EMD Millipore). Subsequently, blocking of these membranes was performed for 2 h using 5% skimmed milk, followed by overnight incubation at 4°C with primary antibodies: Anti-IRS1 (dilution 1:1,000; cat no. sc-8038), anti-PI3K (dilution 1:1,000; cat no. sc-365290), anti-AKT (dilution 1:1,000; cat no. sc-5298), anti-phosphorylated (p)-PI3K (dilution 1:500; cat no. sc-1637) anti-p-AKT (dilution 1:500, cat no. sc-514032) and GAPDH (dilution 1:3,000; cat no. sc-47724). Secondary antibodies (anti-mouse; dilution 1:5,000; cat no. sc-516102) conjugated to horseradish peroxidase (HRP) were added for 2 h at room temperature. All antibodies were obtained from Santa Cruz Biotechnology Inc. Observation of the western blotting images was achieved using enhanced chemiluminescence (ECL) detection reagent on a Bio-Rad ChemiDoc MP system (Bio-Rad laboratories). ImageJ software version 1.46 (National Institutes of Health) was used to measure the density of the protein bands.

Animal experiments

The Experimental Animal Center of Jilin University (Changchun, Jilin, China) provided the 5- to 6 week-old male BALB/c mice (18–20 g; n=10). Mice were housed in specific pathogen-free conditions (SPF) adhering to standard practices with a fixed temperature and humidity level. The protocols received approval from the Institutional Animal Care and Use Committee of Jilin University. A total of 2×106 of TU212 cells (100 µl) were injected into the dorsal scapula region of all the animals. Random assignment of these animals was performed 10 days post-injection, separating the mice into two groups (n=5). The mice were subjected to weekly injections over 21 days. Animals in the test received 100 µl stable XIST-depletion TU212 cells (2×106 cells), while the controls received 100 µl TU212 cells (2×106 cells) stably transfected with the sh-NC plasmid. Calipers were utilized to measure the tumor size on a weekly basis in order to calculate the tumor volume according to the following formula: Volume = (length × width2 ×0.5). After 1 week of treatment, all mice were euthanized by intraperitoneal injection of 200 mg/kg pentobarbital, and then the tumors were excised and weighed. Solid tumors were stored at −80°C until subsequent tests.

Statistical analysis

Data are presented as the mean ± standard deviation and were analyzed using SPSS software (version 18.0; SPSS, Inc.). Student's t-test or one-way ANOVA followed by the Tukey's post hoc test was applied in order to analyze the differences between/among groups. The correlation of XIST and miR-144 or IRS1 in tissue samples was assessed using Pearson's correlation coefficient. P<0.05 was considered to indicate a statistically significant difference.

Results

Expression of XIST is increased in LSCC samples

The present study initially detected the expression of XIST in 48 pairs of LSCC specimens and adjacent normal samples using RT-qPCR. Upregulation of XIST was observed in LSCC samples in comparison with the adjacent normal tissues (Fig. 1A). In addition, this increase in XIST demonstrated a positive association with advanced TNM stage and lymph node metastasis (Fig. 1B and C). These observations are suggestive of the involvement of XIST in LSCC progression.

Figure 1. XIST is upregulated in LSCC tissues. (A) The relative expression of XIST in LSCC tissues (n=48) compared with corresponding nontumor tissues (n=48) was assessed by reverse transcription-quantitative PCR and normalized to GAPDH expression. (B) XIST expression in LSCC tissues with different TNM stage. (C) XIST expression in LSCC tissues with or without lymph node metastasis. **P<0.01. LSCC, laryngeal squamous cell carcinoma; XIST, X inactive-specific transcript; TNM, Tumor-Node-Metastasis.

Knockdown of XIST suppresses LSCC proliferation with increased apoptosis

In order to investigate the involvement of XIST in LSCC, the expression of XIST was decreased by transfection with sh-XIST. As presented in Fig. 2A, XIST expression was successfully and significantly suppressed in the TU212 cells by transfection with sh-XIST. This was followed by an assay to test the effect of XIST on the proliferation of cells. MTS assay demonstrated that the proliferation of TU212 was suppressed to a significant degree following XIST knockdown (Fig. 2B). A concomitant decrease in colonies was in lieu of this result (Fig. 2C). This was followed by flow cytometry, which examined the effect of the knockdown on apoptosis. As presented in Fig. 2D, knockdown of XIST significantly increased the rate of apoptosis in comparison with the cells transfected with sh-NC. The observations are suggestive of compromised proliferation and enhanced apoptosis in LSCC by silencing of XIST.

Figure 2. Knockdown of XIST inhibits LSCC cell proliferation and promotes cell apoptosis. (A) Reverse transcription-quantitative PCR was used to assess the level of XIST in TU212 cells transfected with plasmid sh-XIST or sh-NC. (B) MTS assay was used to evaluate the effect of XIST silencing on TU212 cell proliferation. (C) Relative colony formation was determined in TU212 cells transfected with plasmid sh-XIST or sh-NC. (D) Cell apoptosis was determined in TU212 cells transfected with plasmid sh-XIST or sh-NC by flow cytometry. *P<0.05, **P<0.01. LSCC, laryngeal squamous cell carcinoma; XIST, X inactive-specific transcript; NC, negative control.

XIST knockdown affects the migration and invasion ability of LSCC cells

In order to elucidate the mechanism underlying the effects of XIST on LSCC, the effect that this lncRNA had on the ability of LSCC cells to migrate and invade was assessed in the present study. The wound healing assay revealed that sh-XIST-transfected TU212 cells exhibited a distinct decrease in the ability to migrate in comparison with the control (sh-NC) (Fig. 3A). Among the same cells, the Transwell assay revealed a lowered ability to invade following XIST-knockdown than in the sh-NC group (Fig. 3B).

Figure 3. Knockdown of XIST suppresses LSCC migration and invasion. (A) Wound healing assay was performed to investigate the effect of XIST silencing on TU212 cell migratory capacity. (B) Matrigel invasion assay was performed to evaluate the effect of XIST silencing on TU212 cell invasive capacity. **P<0.01. LSCC, laryngeal squamous cell carcinoma; XIST, X inactive-specific transcript; NC, negative control.

XIST directly targets miR-144 in LSCC

Accumulating evidence has suggested that regulation of miRNA expression is mediated by lncRNAs that function as a sponge for miRNAs (17,18). Thus, the StarBase V2.0 software (http://starbase.sysu.edu.cn/) was utilized to search for putative miRNAs that were targeted by XIST. This led to the selection of miR-144 based on its biological role in LSCC (19,20). As presented in Fig. 4A, the XIST transcript contains a potential site for binding miR-144. To confirm this, a dual luciferase reporter assay was performed in the present study. The assay revealed that the luciferase activity was markedly lowered by miR-144 mimic transfection in the case of Wt-XIST (P<0.01; Fig. 4B). Furthermore, the level of miR-144 in TU212 was conspicuously increased by XIST knockdown (Fig. 4C). To test transfection efficiency of miR-144 mimics or inhibitor in TU212 cells, we examined the expression of miR-144 in this cell line by RT-qPCR. We found that miR-144 overexpression resulted in a significant decrease in the level of XIST, whereas the opposite was observed during the decrease in the level of miR-144 (Fig. 4D). Transfection with miR-144 mimics significantly increased miR-144 expression, while transfection with miR-144 inhibitor decreased miR-144 expression in the TU212 cells (Fig. 4E). The present study then investigated the correlation between the expression of miR-144 and XIST in LSCC samples. The results revealed a distinct downregulation of miR-144 in LSCC tissues compared with the adjacent normal tissues (Fig. 4F), while the association between XIST and miR-144 was demonstrated to be significantly negative (r=−0.536; P<0.001; Fig. 4G). Such observations are implicative of XIST directly targeting miR-144 in LSCC.

Figure 4. miR-144 is a target of XIST in LSCC cells. (A) The predicted binding and mutant sites between miR-144 and XIST are presented. (B) Luciferase activity was detected in TU212 cells following co-transfection with miR-144/miR-NC mimic and Wt-XIST or Mut-XIST reported plasmid. (C) miR-144 expression was assessed in TU212 cells transfected with sh-XIST or sh-NC by RT-qPCR. (D) XIST expression was assessed in TU212 cells transfected with miR-144/miR-NC or miR-144 inhibitor/anti-miR-NC by RT-qPCR. (E) miR-144 expression was examined in TU212 cells transfected with miR-144/miR-NC or miR-144 inhibitor/anti-miR-NC by RT. (F) miR-144 expression was examined in LSCC tissues and adjacent normal tissues (n=48) by RT-qPCR. (G) Correlation between XIST expression and miR-144 expression in LSCC tissues was analyzed by Pearson's correlation analysis. **P<0.01. Wt, wild-type; Mut, mutant-type; RT-qPCR, reverse transcription-quantitative PCR; LSCC, laryngeal squamous cell carcinoma; XIST, X inactive-specific transcript; NC, negative control.

miR-144 mediates the tumor-suppressive effects of XIST depletion on LSCC cells

Transfection of TU212 was performed using a sh-XIST plasmid, and another with a miR-144 inhibitor in order to investigate whether the aforementioned effects involved miR-144. Following transfection, the respective assays were performed to assess the ability of the cells to proliferate, form colonies, migrate and invade. The level of miR-144 was lowered distinctly in those cells that received the sh-XIST and miR-144 inhibitor (anti-miR-144) in comparison with the sh-XIST group, as demonstrated by RT-qPCR (Fig. 5A). In addition, the presence of the miR-144 inhibitor caused a reversal in part of the effects seen with XIST depletion, particularly in terms of the ability of the cells to proliferate, form colonies, undergo apoptosis, migrate and invade (Fig. 5B-F). These results are suggestive of mediation of the miR-144 involved in the manifestations induced by knockdown of XIST in LSCC.

Figure 5. Inhibition of miR-144 reverses the effect on proliferation, colony formation, apoptosis, migration and invasion of LSCC cells mediated by XIST silencing. (A) miR-144 expression was assessed in TU212 cells transfected with sh-NC, sh-XIST and sh-XIST+miR-144 inhibitor (anti-miR-144). (B-F) Cell proliferation, colony formation, apoptosis, migration and invasion were determined in TU212 cells transfected with sh-NC, sh-XIST, and sh-XIST+miR-144 inhibitor (anti-miR-144). *P<0.05, **P<0.01. LSCC, laryngeal squamous cell carcinoma; XIST, X inactive-specific transcript; NC, negative control.

XIST regulates insulin receptor substrate 1 (IRS1) expression and the PI3K/AKT signaling pathway via inhibition of miR-144

IRS-1, a known oncogene, was reported to serve as a direct target of miR-144 in a previous study (19). Thus, the present study investigated the role of this gene in the XIST regulation of IRS1 by regulating miR-144 in LSCC cells. TU212 cells were transfected with sh-NC, sh-XIST and sh-XIST+miR-144 inhibitor individually, followed by analysis of expression (at both the RNA and protein levels with the appropriate assays). The results revealed that XIST knockdown significantly inhibited the levels of its mRNA (Fig. 6A) and protein (Fig. 6B) in the TU212 cells, while miR-144 inhibitor reversed these trends. IRS1 was reported to be involved in regulating the PI3K/AKT signaling pathways (21,22). The present study investigated whether XIST affects activation of the PI3K/AKT signaling pathway mediated by miR-144. The western blot analysis revealed that XIST knockdown significantly inhibited activation of the PI3K/AKT pathway in TU212 cells (Fig. 6B), while miR-144 inhibitor demonstrated a reverse trend. The present study further investigated the correlation between XIST and IRS1 in LSCC clinical samples. A conspicuous upregulation of IRS1 was observed (Fig. 6C), which revealed a positive correlation with that of XIST in the LSCC tissues (r=0.519; P<0.001; Fig. 6D). Collectively, these results highlight that XIST modulated IRS1 expression and the PI3K/AKT signaling pathway by regulating miR-144 in LSCC cells.

Figure 6. XIST regulates IRS1 expression and the PI3K/AKT signaling pathway via inhibition of miR-144. (A) IRS1 mRNA expression was determined in TU212 cells transfected with sh-NC, sh-XIST and sh-XIST+anti-miR-144. (B) IRS1, PI3K, p-PI3K, AKT and p-AKT protein levels were assessed in TU212 cells transfected with sh-NC, sh-XIST and sh-XIST+anti-miR-144. (C) IRS1 mRNA expression was examined in LSCC tissues and adjacent normal tissues (n=48) by reverse transcription-quantitative PCR. (D) Correlation between XIST expression and IRS1 expression in LSCC tissues was analyzed by Pearson's correlation analysis. *P<0.05, **P<0.01. LSCC, laryngeal squamous cell carcinoma; IRS1, insulin receptor substrate 1; XIST, X inactive-specific transcript; NC, negative control.

XIST knockdown inhibits tumorigenesis in a mouse model

TU212 cells were inoculated in mice in order to construct a xenograft mouse model and study the role of XIST in LSCC in vivo. The animals that received sh-XIST cells exhibited a conspicuous lower xenograft tumor volume in comparison with those that received sh-NC cells (Fig. 7A). In addition, the average dimensions (size/weight) of tumors in the sh-XIST-inoculated group were comparatively and distinctly lower than these parameters in the sh-NC-inoculated group (Fig. 7B and C). The levels of XIST, miR-144 and IRS1 were analyzed in the tumors. The levels of XIST and IRS1 were significantly lower (Fig. 7D and E), and that of miR-144 was higher in the sh-XIST group (Fig. 7F) as demonstrated by RT-qPCR.

Figure 7. Knockdown of XIST inhibits LSCC tumorigenesis in vivo. (A) The tumor volume was calculated every 7 days after treatment with sh-XIST or sh-NC. (B) The representative images of the xenograft tumors. (C) The tumor weight. (D-F) Reverse transcription-quantitative PCR was used to analyze the levels of XIST, IRS-1 and miR-144 in the xenograft tumor. *P<0.05, **P<0.01. LSCC, laryngeal squamous cell carcinoma; IRS1, insulin receptor substrate 1; XIST, X inactive-specific transcript; NC, negative control.

Discussion

The vital roles of long non-coding RNAs (lncRNAs) in oncogenesis, as well as in the advancement of tumors, have been demonstrated in a number of research articles. This is suggestive of their application as markers for diagnosis, prognosis or therapeutic intervention in tumors such as laryngeal squamous cell carcinoma (LSCC) (9,10). For example, Yang et al demonstrated that LOC554202 promotes LSCC progression via miR-31 (23). Another group led by Shen highlighted inhibition of LSCC by another lncRNA AC026166.2-001 via miR-24-3p/p27 (24). Qu et al demonstrated that HOXA11-AS plays an oncogenic role in LSCC by promoting the ability of cells to proliferate, migrate and invade (25). Such examples reported are indicative of the use of lncRNAs as markers or targets in patients with LSCC. This calls for a thorough understanding of the functions and mechanisms associated with lncRNAs in LSCC, which is crucial for addressing this malignancy.

lncRNA X inactive-specific transcript (XIST) demonstrates increased levels to function as an oncogene in multiple types of cancer, such as non-small lung cancer, hepatocellular carcinoma, retinoblastoma, osteosarcoma, thyroid cancer, colon cancer, bladder cancer, glioma and gastric cancer (26–34). On the contrary, in breast cancer and ovarian cancer (35,36), XIST expression was found to be downregulated and to serve as a tumor suppressor. Such contradictory observations imply that XIST can function either as an oncogene or tumor suppressor dependent on the type of tumor. As the function of XIST in the incidence, as well as the development of LSCC, lacked clarity, the present study aimed to address these aspects. LSCC samples exhibited higher levels of XIST than the levels noted in the adjacent normal tissues, and this increase was associated strongly and positively with the TNM stage and lymph node metastasis status. The ability of LSCC cells to proliferate, form colonies, migrate and invade was markedly lower, while the apoptosis ratio was higher following XIST knockdown. The use of mouse models revealed that XIST depletion caused a compromise in tumor growth. Taken together, the oncogenic functioning of XIST in LSCC is suggested.

Research shows that lncRNAs function as competing endogenous RNAs (ceRNAs) in order to modulate the expression and functioning of miRNAs, which in turn bind mRNAs and are associated with the origin and advancements of tumors (17,18). The ceRNA aspect of XIST to sponge miRNAs, such as miR-141 (26), miR-367 (26), miR-124 (28), miR-181a (27), miR-195-5p (29), miR-34a (31), miR-139-5p (32) and miR-429 (32) in numerous cancer samples has been studied. The present study revealed an association between XIST and miR-144 through biological analyses, which was further confirmed by luciferase reporter experiments. miR-144 works as a tumor suppressor in LSCC (19,20). In lieu of earlier work (19), a lowered level of miR-144 in LSCC samples in comparison with that of adjacent normal tissues was demonstrated in the present study. A point to be noted is an inverse association of miR-144 with XIST in LSCC. Furthermore, miR-144 overexpression caused inhibition of XIST, while higher XIST was observed when miR-144 was inhibited in TU212 cells. When XIST was silenced, the levels of miR-144 were increased. Furthermore, miR-144 inhibition rescued the inhibitory effect on LSCC progression caused by the absence of XIST. The observations reported are indicative of the miR-144 sponging function of XIST to promote LSCC progression.

Research has demonstrated that target genes of miRNAs could be regulated by lncRNAs by sponging its target miRNAs (37). A recent study revealed that miR-144 plays an anticancer role in LSCC by targeting insulin receptor substrate 1 (IRS1) (19). IRS1, a docking protein, functions as an oncogene in multiple types of cancer by promoting tumor progression (38), and also regulates the downstream PI3K/AKT and MAPK circuit (21,22,39). Therefore, the involvement of IRS1 in the XIST-miR-144 association in LSCC was investigated in the present study. XIST depletion causes a distinct inhibition of IRS1 expression in TU212 cells, while miR-144 inhibitor reversed this trend. Furthermore, it was also revealed that XIST knockdown significantly inhibited activation of the PI3K/AKT signaling pathway in TU212 cells, while miR-144 inhibitor reversed this trend. The present study further investigated the correlation between XIST and IRS1 in LSCC clinical samples, and reported a positive association in LSCC. The aforementioned results suggest that XIST modulates IRS1 expression and the PI3K/AKT signaling pathway by regulating miR-144 in LSCC cells.

There were limitations to the present study. First, additional LSCC tissue samples are required in order to further investigate the clinical significance of XIST in LSCC. Secondly, to test the biological role of XIST in LSCC, two or more LSCC cell lines should be used. Thirdly, XIST could regulate more miRNAs or target genes; thus, further experiments should be performed in order to fully understand the molecular mechanism underlying XIST in LSCC.

In summary, XIST is significantly upregulated in human LSCC samples, and increased XIST expression is associated with TNM stage and lymph node metastasis. Depletion of XIST adversely affected the ability of LSCC to proliferate, form colonies, migrate and invade in vitro, and compromised growth in mouse models via regulation of the miR-144/IRS1 axis. The present study is suggestive of investigating XIST as a tool for the therapeutic intervention of LSCC.

Acknowledgements

Not applicable.

Funding

No funding was received.

Availability of data and materials

The datasets used during the present study are available from the corresponding author upon reasonable request.

Authors' contributions

XW and XYC designed the present study. CLC performed the experiments. YNL analyzed the data and XYC wrote the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The present study was approved by the Ethics Committee of Jilin University (Changchun, Jilin, China) in accordance with the Declaration of Helsinki (2000) and written informed consent was obtained from all participants.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References