LncRNA HOXA-AS3 promotes the progression of oral squamous cell carcinoma through sponging miR-218-5p

Objective The aim of the present study was to investigate the roles and molecular mechanism of long non-coding RNA (lncRNA) HOXA-AS3 in the progression of oral squamous cell carcinoma (OSCC). Methods The expression of HOXA-AS3 and miR-218-5p was detected in OSCC tissues and cells using quantitative real-time polymerase chain reaction (qRT-PCR). Cell Counting Kit-8 (CCK-8) and colony formation assays were used to examine the effects of HOXA-AS3 and miR-218-5p on the proliferation of OSCC cells. Luciferase reporter gene assay was used to conrm the directly binding condition between lncRNA HOXA-AS3 and miR-218-5p in OSCC cells. RNA immunoprecipitation assay was employed to verify the interaction between HOXA-AS3 and miR-218-5p. The relative expression of lncRNA HOXA-AS3 was observably upregulated in OSCC tissues and cell lines compared with the para-cancerous tissues and normal human oral keratinocyte (NHOK), respectively. Knockdown of HOXA-AS3 signicantly inhibited the proliferation and colony formation of OSCC cells. Bioinformatics analysis and luciferase reporter assay showed that HOXA-AS3 directly bound to miR-218-5p. Moreover, the expression of miR-218-5p was negatively regulated by HOXA-AS3, and there was an inverse correlation between them. Silencing miR-218-5p reversed the inhibitory effect of lncRNA HOXA-AS3 knockdown on the proliferative potential of OSCC cells.


Introduction
Oral squamous cell carcinoma (OSCC) is one of the most common head and neck squamous cell carcinomas and a public health threat. There are about 500,000 new cases in the world every year, and the incidence is increasing year by year [1][2][3]. Due to the reasons of quick progression, invasive growth, easy lymph node and distance metastasis, and high recurrence rate, the prognosis of OSCC is relatively poor, with an overall 5-year survival rate of less than 50% [4,5]. In recent years, despite synthetic serial treatments including of surgery, radiotherapy and chemotherapy have been widely applied in the patients of OSCC, the cure rate is one half only [6]. Therefore, an improved understanding of the molecular mechanisms underlying OSCC tumorigenesis help to provide novel insights into the pathogenesis of OSCC and thus improve the diagnostic and therapeutic methods. with 10% fetal bovine serum (FBS; Gibco, Waltham, MA, USA), 100 IU/mL of penicillin and 100 μg/mL of streptomycin at 37 °C incubator with 5% CO 2 .
shRNA containing the HOXA-AS3 interference sequence (sh-HOXA-AS3) and negative control (sh-NC) were purchased from GeneChem (Shanghai, PRC), and miR-218-5p mimics, anti-miR-218-5p and negative control (miR-NC) were purchased from RiboBio (Guangzhou, PRC). Transfection was performed using Lipofectamine® 2000 Reagent (Invitrogen, Waltham, MA, USA) according to the manufacturer's protocol in OSCC cells. The culture medium was replaced 6 h after transfection, and transfection e ciency was examined with the expression vector of red uorescent protein (RFP) at 48 h after the transfection.
RNA extraction and quantitative real time-polymerase chain reaction (qRT-PCR) Total RNA was extracted from OSCC tissues or cells using the TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. qRT-PCR was performed using the All-in-One™ miRNA qRT-PCR detection kit (GeneCopoeia, Rockville, MD, USA) for miR-218-5p and U6 as the internal control.

Cell proliferation analysis
Cell Counting Kit-8 reagent (Dojindo, Kumamoto, Japan) was used to measure the proliferation of OSCC cells. The OSCC cells were seeded into 96-well plates after transfection for 48 h. After culturing for 12 h, 24 h, 48 h and 72 h respectively, and then CCK-8 reagent was added to each well with 10 μl and further incubated for 4 h. The optical density (OD) value of each well was detected using an enzyme labeling instrument at 450 nm.

Colony formation assay
The OSCC cells were inoculated into 6-well plates with 200 cells in each well after transfection for 48 h.
Subsequently, the cells were cultured in the complete medium for 14 d. At the rst time, the medium was replaced after 5 d, and then replaced every 3 d. When the cell colonies formed, the medium was sucked dry. Then, the cells were washed twice with phosphate-buffered saline (PBS) and xed with 4% paraformaldehyde at 4 °C for 1 h. Next, the cells were stained with 0.1% crystal violet staining solution for 20 min. Finally, the number of cell colonies containing >50 cells in each well was calculated and photographed.

Luciferase reporter assay
Online Software Starbase v2.0 (http://starbase.sysu.edu.cn) was used to predict the target miRNAs of lncRNA HOXA-AS3. The wild-type HOXA-AS3 3'-UTR containing the miR-218-5p seed sequence fragment (HOXA-AS3 Wt) and mutant-type (HOXA-AS3 Mut) luciferase vectors were conducted. The OSCC cells were seeded into 48-well plates and were co-transfected with the luciferase vectors and miR-218-5p mimics or negative control using Lipofectamine® 2000 Reagent for 48 h. This assay was normalized with 0.05 μg of the RFP expression vector pDsRed2-N1 (Clontech, USA). Subsequently, cells were lysed with RIPA lysis buffer, and the luciferase activity and RFP intensity were detected with the F-4500 Fluorescence Spectrophotometer (Hitachi, Japan) according to the manufacturer's instructions.
RNA immunoprecipitation (RIP) assay RIP assay was used to detect the sponge function of HOXA-AS3 on miR-218-5p by using Magna RIPTM RNA Immunoprecipitation Kit (Millipore, Bedford, MA,USA). Brie y, the OSCC cells were transfected with miR-218-5p mimics, Vector-HOXA-AS3 or corresponding controls for 48 h, and then were lysed using the lysis buffer. Next, cell lysates were incubated with anti-Ago2 (Abcam, UK) or anti-IgG (Abcam, UK) and protein A/G magnetic beads. Finally, co-precipitated RNAs were detected by qRT-PCR.

Statistical analysis
Statistical analyse was performed using SPSS version 20.0 (IBM, Armonk, NY, USA) Data are presented as the mean ± standard deviation. Differences among multiple groups were analyzed by ANOVA (one-way) followed by Tukey t-test, and differences between the two groups were analyzed using the student's t-test. Correlation analysis between HOXA-AS3 and miR-218-5p expression was assessed using Pearson's correlation coe cient. The prognosis survival time of patients was evaluated using Kaplan-Meier analysis, and Log-rank test was used to examine the difference between different curves. P<0.05 was considered to indicate a statistically signi cant difference.

Results
LncRNA HOXA-AS3 was upregulated in OSCC tissues and cell lines, and signi cantly correlated with patient survival In this study, we examined the expression of lncRNA HOXA-AS3 in 38 paired human OSCC tissues and the corresponding para-carcinoma tissues, as well as OSCC cell lines using qRT-PCR assay. As shown in Figure 1A, the expression level of lncRNA HOXA-AS3 was signi cantly increased in human OSCC tissues compared with that in the corresponding para-carcinoma tissues. Meanwhile, the expression levels of lncRNA HOXA-AS3 in four OSCC cell lines were obviously higher than that of NHOK cells (Fig. 1B). The associations between lncRNA HOXA-AS3 expression and age, gender, clinical stage and distant metastasis of OSCC patients were shown in Table 1, indicating that patients with higher HOXA-AS3 expression had a higher clinical stage compared to those with lower HOXA-AS3 expression level. Moreover, as shown in Figure 1C, the lncRNA HOXA-AS3 expression in the 38 OSCC tissues was signi cantly associated with the overall survival of OSCC patients. The patients with low HOXA-AS3 expression had better overall survival compared to those with high expression. Furthermore, we found that downregulation of HOXA-AS3 decreased the colony formation ability of SCC-9 and CAL-27 cells by the colony formation assay (Fig. 2D). Thus, it was concluded that downregulation of lncRNA HOXA-AS3 suppressed the cell growth of OSCC.
MiR-218-5p was downregulated in human OSCC tissues and cell lines, and inversely correlated with HOXA-AS3 expression QRT-PCR was performed to detect expression level of miR-218-5p in OSCC cell lines, and 38 pairs of OSCC and para-cancerous tissues. As shown in Figure 4A, the expression level of miR-218-5p was markedly lower in OSCC cells than that of NHOK cells. Meanwhile, the expression of miR-218-5p was remarkably lower in OSCC tissues compared with that of para-cancerous tissues (Fig. 4B). And then, Pearson's correlation analysis showed there was a correlation between HOXA-AS3 and miR-218-5p expression in OSCC tissues. As shown in Figure 4C, the expression of miR-218-5p was inversely correlated with HOXA-AS3 expression level in OSCC tissues. Furthermore, the expression of miR-218-5p was markedly increased after HOXA-AS3 knockdown (Fig. 4D).

Overexpression of miR-218-5p inhibited the proliferation of OSCC cells
To investigate the role of miR-218-5p in the proliferation of OSCC cells, miR-218-5p mimics or miR-NC was transfected into SCC-9 and CAL-27 cells. As shown in Figure 5A, the expression level of miR-218-5p was signi cantly increased in miR-218-5p mimics group compared to that of miR-NC group both in SCC-9 and CAL-27 cells. Then, the proliferation of OSCC cells was measured using CCK-8 and colony formation assay, respectively. The data revealed that the proliferation ability of SCC-9 and CAL-27 cells in the miR-218-5p mimics group was obviously reduced when compared to miR-NC group (Fig. 5B-5D).
Together, overexpression of miR-218-5p inhibited the OSCC cells proliferation, which was consistent with the result of downregulation of HOXA-AS3.
Anti-miR-218-5p reversed the effect of lncRNA HOXA-AS3 knockdown on the proliferation of OSCC cells To further study the interaction between HOXA-AS3 and miR-218-5p in OSCC cells, anti-miR-218-5p was transfected into SCC-9 and CAL-27 cells with HOXA-AS3 silencing. The expression level of HOXA-AS3 in each group was measured by qRT-PCR. As shown in Figure 6A and 6B, the expression of HOXA-AS3 in the co-transfected with si-HOXA-AS3 and anti-miR-218-5p group was observably higher than that of the co-transfected with si-HOXA-AS3 and miR-NC group. Moreover, CCK-8 assay showed that HOXA-AS3 knockdown could signi cantly suppress the proliferative activity of SCC-9 and CAL-27 cells, but was further reversed by miR-218-5p knockdown ( Fig. 6C and 6D). As shown in Figure 6E and 6F, colony formation assay yielded the identical results.

Discussion
OSCC is a kind of malignant tumors that seriously threaten human health, and it has the characteristics of easy invasion and lymph node metastasis. Although great progress has been made in exploring new treatments, the prognosis of OSCC is still unsatisfactory due to its high malignancy [22]. Risk factors for OSCC include betel nut, tobacco, low-quality edible pigment, human papillomavirus infection, etc [23]. At present, the pathogenesis of OSCC is still not very clear, which may involve a multi-gene epigenetic and metabolic process, such as the loss of the function of cancer suppressor gene and activation of function of oncogene. [24,25]. Thus, the research on the molecular biological mechanism of OSCC, especially the detection and diagnosis of speci c early tumor markers, which is important for early diagnosis and improvement of prognosis of OSCC patients.
HOXA-AS3 is a novel lncRNA located in chromosome 7p15.2, and belongs to the clusters of HOX genes, a group of highly homologous transcription factors that regulate embryological development [26]. HOXA-AS3 interacts with Enhancer Of Zeste 2 (EZH2) and acts as an epigenetic switch that determines the lineage speci cation of mesenchymal stem cells [27]. At present, there has been only two published researches expounding the roles of lncRNA HOXA-AS3 in glioma and lung adenocarcinoma, respectively [20,21]. Wu et al. [20] reported that the expression of HOXA-AS3 was signi cantly increased in glioma tissues and cell lines, and knockdown of HOXA-AS3 inhibited the cell growth in vitro and vivo, promoted cell apoptosis, and impaired cell migration in glioma cells. Similarly, HOXA-AS3 was markedly upregulated in tissues and cells of lung adenocarcinoma, and promoted cancer cell progression [21]. However, there is no report on the function and molecular mechanism of HOXA-AS3 in OSCC. In this study, the expression levels of lncRNA HOXA-AS3 in OSCC tissues and cell lines were measured. We discovered that lncRNA HOXA-AS3 was more highly expressed in OSCC tissues and cells than that of in para-carcinoma tissues and NHOK cells, respectively. Moreover, the high expression of HOXA-AS3 was obviously correlated with pathological stage and overall survival of OSCC patients. To further investigate the function of lncRNA HOXA-AS3 on the biological performances of OSCC cells, HOXA-AS3 knockdown model was constructed in the SCC-9 and CAL-27 cells. The data revealed that knockdown of HOXA-AS3 suppressed the cell proliferation and growth in OSCC. This means that the high expression of HOXA-AS3 might be closely related to the progression of OSCC.
Currently, lncRNAs have been demonstrated to function as competing endogenous RNAs (ceRNA) by sponging miRNA and inhibiting intracellular miRNA function [28,29]. Therefore, establishing the interrelationship of lncRNA and its regulation miRNA may help further understanding of the molecular mechanism underlying tumor progression and provide potential therapeutic targets for the clinical treatment of tumors. Bioinformatics analysis predicted that there was a binding site of miR-218-5p in HOXA-AS3 3'UTR, and luciferase reporter assay con rmed the direct interaction in the SCC-9 and CAL-27 cells. Meanwhile, the expression of miR-218-5p was markedly decreased in OSCC tissues compared with para-carcinoma tissues, and was negatively correlated with HOXA-AS3 expression. RIP assay showed that HOXA-AS3 might work as a miR-218-5p sponge in OSCC. These results suggested that lncRNA HOXA-AS3 might in uence the function of OSCC cells via sponging and regulating the miR-218-5p.
Additionally, miR-218-5p was also illustrated to function as an anti-metastasis miRNA in non-small cell lung cancer [34], cervical cancer [35], and hepatocellular carcinoma [36]. Our study showed that the expression of miR-218-5p was signi cantly downregulated in OSCC cell lines, and overexpression of miR-218-5p remarkably inhibited the proliferation and colony formation of OSCC cells. Furthermore, to investigate whether HOXA-AS3 promoted the development of OSCC through regulating miR-218-5p, anti-miR-218-5p was transfected into SCC-9 and CAL-27 cells with knockdown of HOXA-AS3. The results revealed that knockdown of miR-218-5p could restore OSCC cell proliferation and colony formation activities after HOXA-AS3 silencing, suggesting that HOXA-AS3 might promote malignant progression of OSCC by inhibiting miR-218-5p.

Conclusion
In a word, our study showed that lncRNA HOXA-AS3 is signi cantly upregulated in OSCC and this high expression positively correlated with the pathological stage and poor prognosis of patients, which promotes the development of OSCC through sponging and inhibiting miR-218-5p.

Declarations
Ethics approval and consent to participate  Smoking: More than one cigarettes a day, sustaining more than 6 months Drinking: At least once a week, sustaining more than 6 months * Significant difference (P<0.05).