Silencing of LINC00461 enhances radiosensitivity of lung adenocarcinoma cells by down‐regulating HOXA10 via microRNA‐195

Abstract Lung adenocarcinoma is recognized as one of the most recurrent tumours in adults. Long non‐coding RNAs (lncRNAs) are non–protein‐coding transcripts and have been demonstrated to regulate biological functions during tumorigenesis. Our study aims to investigate the underlying molecular mechanisms of LINC00461/microRNA‐195 (miR‐195)/HOXA10 responsible for its involvement in lung adenocarcinoma. We firstly selected differentially expressed lncRNAs and genes by the Cancer Genome Atlas (TCGA) database and Gene Expression Omnibus (GEO). The functional role of LINC00461 in lung adenocarcinoma was then determined using ectopic expression, knockdown and reporter assay experiments. Besides, we detected the expression profiles of LINC00461, miR‐195, HOXA10 and apoptosis‐ and invasion‐related genes. Cell proliferation, migration and invasion were evaluated. In vivo tumour formation ability was analysed. Overexpressed LINC00461 and HOXA10 but down‐regulated miR‐195 were observed in primary and metastatic lung adenocarcinoma. LINC00461 negatively regulated miR‐195, while miR‐195 negatively regulated HOXA10. Forced LINC00461 expression decreased expression of miR‐195 and Bax, increased expression of HOXA10, MMP‐2, MMP‐9 and Bcl‐2, promoted cell proliferation, migration and invasion as well as tumour formation, and enhanced radiosensitivity of lung adenocarcinoma cells. However, these effects were reversed by lentivirus‐mediated miR‐195–forced expression, thereby suggesting that miR‐195 could antagonize the harmful effect of LINC00461 on lung adenocarcinoma cells. Collectively, the present study provides evidence supporting the inhibitory effect of LINC00461 silencing on lung adenocarcinoma, which suppresses lung adenocarcinoma cell migration, invasion and radiosensitivity via HOXA10 by binding to miR‐195, which provides a promising basis for the targeted intervention treatment for human lung adenocarcinoma.


| INTRODUC TI ON
Lung cancer is among the most prevalent malignancies worldwide associated with a clinically high overall cancer-related mortality. 1 Lung adenocarcinoma, presenting as epithelial cancer of glandular origin, has the highest prevalence of all types of non-small cell lung cancer (NSCLC). 2 In addition, lung adenocarcinoma accounts for 40% of lung cancer cases, accounting for over 500,000 yearly deaths worldwide. 3 Approximately 25%-30% of patients with NSCLC have contracted a localized disease because of surgical intervention for curative purposes; unfortunately, the 5-year survival rate still remains to be poor, and 40%-70% of these patients are at risk of a susceptible systemic disease with or without local relapses and could eventually die from it. 4 In order to improve the prognosis of patients with lung adenocarcinoma, the identification of molecular mechanism is required.
Among the non-coding RNAs, long non-coding RNAs (ln-cRNAs) are transcripts with more than 200 nucleotides in length. 5 Aberrant lncRNA expression is involved in various types of human cancers, including breast cancer, prostate cancer and colorectal cancer. 6 LINC00461, a human homolog of the mouse lncRNA C130071C03Rik, is reported to present with an overexpressed expression profile in human glioma tissues. 7 Furthermore, LINC00461 was predicted to be up-regulated in lung adenocarcinoma in the Cancer Genome Atlas (TCGA). LncRNAs binding with microRNAs (miRNAs or miRs) inhibit expression of miRs by binding to their target sites, thereby regulating the protein expression. 8 Moreover, a biological prediction website RNA22 predicted LINC00461 to directly regulate miR-195. MiR-195, located on the chromosome 17p13.1, is a member of the miR-16/15/195/424/497 family. 9 Besides, miR-195 has been reported to be involved in tumorigenesis as a tumour suppressor. 10 High miR-195 level has been used as a tumour predictor for the prognosis of non-smoking women with lung adenocarcinoma. 11 A bioinformatics website microRNA.org then further revealed that homeobox A10 (HOXA10) was a potential target gene of miR-195. HOXA10 belongs to the homeobox gene family that is well conserved during evolution and plays a vital role in several biological processes. 12 A competing endogenous RNA network analysis showed that lncRNA ENSG00000240990 competed with HOXA10 to affect the prognosis of lung adenocarcinoma patients. 13 Furthermore, HOXA10 with recurrent up-regulation has been documented in human lung cancer cells and tissues. 14 Base on the aforementioned literature, we hypothesized that LINC00461 may be involved in lung adenocarcinoma by regulating HOXA10 via miR-195. Therefore, this study was planned to explore the regulatory mechanism of the LINC00461/miR-195/HOXA10 regulatory network in lung adenocarcinoma.

| Microarray-based gene expression profiling
TCGA (http://cance rgeno me.nih.gov/) was used to retrieve gene expression datasets related to lung adenocarcinoma, and the transcriptome profiling data containing the package edgeR of R was analysed using differential analysis. 15 False-positive discovery (FDR) correction was performed on the p-value with package multitest. FDR < 0.05 and |log2 (fold change)| > 2 were set as the threshold to screen DEGs Affy package of R software was employed for background correction and normalization of each chip data. 16 The linear empirical Bayes statistical method in the Limma installation package, in combination with the traditional t test, was employed for non-specific filtration of the expression data, in order to screen the differentially expressed RNAs and genes. 17 The miRNAs interacting with specific lncRNA and gene were determined using RNA22 (https ://cm.jeffe rson.edu/rna22/ ).

| Dual-luciferase reporter gene assay
A biological prediction website RNA22 was employed to predict the target relationship between LINC00461 and miR-195. The 3'untranslated region (3'UTR) of LINC00461 was amplified, and PCR products were subcloned and ligated into the pmirGLO (Promega) using the endonuclease sites SpeI and Hind III to collectively construct pMIR-LINC00461-wild-type (Wt). Then, the LINC00461 binding site mutant homeoboxa10, long non-coding RNA LINC00461, lung adenocarcinoma, microRNA-195, radiosensitivity (Mut) (LINC00461-Mut: GACCAGGGACGCTGCTC.) was predicted using the target gene database, and the recombinant vector was constructed by the T4 DNA Ligase. MiR-195 mimic and negative control (NC) were, respectively, cotransfected using the luciferase reporter vector into NCI-H1650 cells (with Renilla luciferase vector pRL-TK [Takara Biotechnology Ltd] as internal control). After 48 hours, the cells were collected and lysed, and the relative luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega). The experiment was performed three times independently.
A biological prediction website microRNA.org was employed to predict the target relationship between miR-195 and HOXA10. The 3'UTR of HOXA10 was amplified, and the PCR products were subcloned and ligated into pmirGLO (Promega) using the endonuclease sites SpeI and Hind III to conjointly construct pMIR-HOXA10-Wt.
Then, the HOXA10 binding site Mut (auUAAUAUUGUAAACGAC-CUg) was predicted by the target gene database and the recombinant vector was constructed using T4 DNA Ligase. MiR-195 mimic and NC were, respectively, cotransfected with the luciferase reporter vector into NCI-H1650 cells (with Renilla luciferase vector pRL-TK [Takara Biotechnology Ltd] as internal control). After 48 hours, the cells were collected and lysed, and the relative luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega).
The experiment was performed three times independently.

| RNA fluorescence in situ hybridization (FISH)
FISH technique was employed to identify the subcellular localization of LINC00461 in the cells. According to the instructions of Ribo™ lncRNA FISH Probe Mix (Red) (Ribo Biological), coverslips were placed in 6-well plates, and cells in logarithmic growth phase were seeded in the plates for 1 d to facilitate cell confluence to 80%. Then, the coverslips were removed, rinsed with phosphate-buffered saline (PBS), fixed using 1 mL of 4% paraformaldehyde, followed by the addition of protease K (2 μg/mL), glycine and acetylation reagent, and then finally incubated in 250 μL pre-hybridization solution for 1 hour at 42ºC. Next, the pre-hybridization solution was removed, and 250 μL of pre-hybridization solution containing probes (300 ng/ mL) was added the samples for overnight incubation at 42°C. Then, the coverslips were rinsed 3 times with phosphate-buffered saline

| RNA pull-down assay
Using the Magnetic RNA-Protein Pull-Down Kit (Pierce), 1 μg of biotin-labelled RNA LINC00461 was added into Eppendorf (EP) tubes, added with 500 μL of Structure Buffer at 95ºC for 2 minutes and incubated on ice for 3 minutes. Following suspension, 50 μL of magnetic beads was added to the EP tube for overnight incubation at 4ºC, followed by 1610 g centrifugation for 3 min. Afterwards, the precipitate was rinsed with 500 μL RIP 3 times, and then, 10 μL of cell lysate was added, and placed at room temperature for 1 hour.
Then, the magnetic beads RNA-protein complex was centrifuged to collect the supernatant followed by rinsing with 500 μL of RIP Wash Buffer 3 times. The 10 μL of supernatant was used as the Input of protein. The protein concentration was measured, and the protein expression was detected by conducting Western blot analysis. The experiment was performed three times independently.

| RNA binding protein immunoprecipitation (RIP)
The RIP kit (Millipore) was used to identify the binding between  at room temperature for 20 minutes, which was added into the well in a drop wise manner and mixed gently. Next, the cells were preserved in a 5% CO 2 incubator at 37ºC for 5-6 hours, and the complete medium replacement was conducted to sustain further culture for 24-48 hours. The transfection efficiency was determined and the follow-up experiments were conducted upon reaching over 70% transfection efficiency.

| Isolation of mRNA and reverse transcription quantitative polymerase chain reaction (RT-qPCR)
Trizol Reagent (Invitrogen) was used to extract total RNA, and the reverse transcription was conducted using the PrimeScript™ RT reagent Kit (RRO37A, TaKaRa Biotechnology Co. Ltd). Fluorescence quantitative PCR instrument (ABI 7500, Applied Biosystems) was employed for amplification of the target genes and reference genes. U6 was regarded as the internal reference for miR-195, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was regarded as the internal reference for others. On the basis of the 2 −ΔΔCT method, the expression of the genes was calculated. 18 The experiment was performed three times independently. Primer sequences are shown in Table 1. as the internal reference. Then, the membrane was visualized using enhanced chemiluminescence (WBKLS0100, Millipore), followed by analysis using the Image J software (Bio-Rad Laboratories). The experiment was performed three times independently.

| Radiotherapy and 3-(4, 5-dimethylthiazol-2yl)-2, 5-diphenyltetrazolium bromide (MTT) assay
The cells in the logarithmic growth phase were seeded in a 96-well plate at a concentration of 1 × 10 5 cells/mL. Cells were then allocated into the irradiation group and the control group, with 4 groups in each group, 8 wells in each group and 100 μL in each well, and cultured in cell incubator of 5% CO 2 at 37°C. Cells were irradiated with 6 MV X-ray using a linear accelerator, with X-ray dose of 1, 1.5, 2 and 2.5 Gy to isocenter for each group, respectively. The dose rate was 100 Mu/min, and the irradiation field was 10 cm × 10 cm. The aforementioned steps were performed to optimize the dose of radiation.
After grouping, the cells underwent the same treatment except for the X-ray dose of 2 Gy to isocenter. The control group received no treatment. Next, 2 mg/mL MTT liquid was added in each group group/A490 of the control group × 100%. The experiment was performed three times independently.

| Scratch test
Six-well plates of six groups were selected and grouped in strict accordance with the corresponding cells. Behind each plate, a line was drawn every 0.5-1 cm across the well using a marker pen. A minimum of five lines was drawn per well. Then, each well was added with approximately 5 × 10 5 cells for overnight culture. On the second day, the pipette tip was used to vertically scratch the line. Cells were rinsed with PBS for three times with the removal of the scratchedout cells and cultured with serum-free medium in an incubator with 5% CO 2 at 37ºC followed by observation at 0 and 24 hours under an inverted microscope (×40). Photographs were obtained, and scratch distance was measured. The experiment was performed three times independently.

| Transwell assay
The apical chamber of each chamber was added with 60 μL of  The experiment was performed three times independently. was an indicative of significant difference.

| LINC00461 is overexpressed in lung adenocarcinoma
TCGA database demonstrated a high expression of LINC00461 and HOXA10 in lung adenocarcinoma ( Figure 1A expression was higher in H1299, A549, PC9 and LTEP-A-2 cell lines than those observed in the NCI-H1650 cells (P < .05) ( Figure 1E).
Therefore, the NCI-H1650 and PC9 cell lines were selected for subsequent experimentation.

| LINC00461 up-regulated HOXA10 by binding to miR-195
Based on the bioinformatics website, we found that there was a specific binding point between miR-195 and LINC00461 (Figure 2A).
Dual-luciferase reporter gene assay ( Figure 2B) showed that when cotransfected with pLINC00461-Wt, the luciferase activity of miR-195 mimic group was lower than that of NC group (P < .05), while when cotransfected with pLINC00461-Mut, no significant difference in the luciferase activity was observed (P > .05). The results suggested that LINC00461 competitively bound to miR-195.
Furthermore, HOXA10 was predicted as a potential target gene

| Silencing LINC00461 inhibited the expression of invasion/apoptosis-related factors
The

| Silencing LINC00461 and overexpressed miR-195 inhibited proliferation, migration and invasion but enhanced radiosensitivity of lung adenocarcinoma cells
The results of radiotherapy pre-test showed ( Figure 4A) that the cell survival rate of the 1.0 Gy and 1.5 Gy groups was high, indicating that the radiosensitivity was not obvious, while the survival rate of 2.5 Gy group was low, thereby suggesting that the high dose was not suitable for radiotherapy. Therefore, the 2.0 Gy group was used in the following experiments. We then continued the experiment to identify the influence on cell functions in the NCI-H1650 and PC9 cell lines. As shown in Figure 4B-G, among LINC00461-related groups, there was no significant difference among the control and shRNA-NC groups (P > .05). In comparison with the control group, cell proliferation, migration and invasion abilities of the sh-LINC00461 group were decreased while the radiosensitivity was elevated (P < .05). A contradicting trend was observed in the LINC00461 + mimic NC group (P < .05). In comparison with the LINC00461 + mimic NC group, cell proliferation, migration and invasion abilities were decreased in the LINC00461 + miR-195 mimic group while the radiosensitivity was markedly increased (P < .05). As shown in Figure 4H

| D ISCUSS I ON
Lung adenocarcinoma denotes the most prevalent subtypes of NSCLC affecting millions of population around the globe. 19 Recently, the functionality of lncRNAs has been extensively reported in lung adenocarcinoma. For instance, lncRNA FEZF1-AS1 is related to F I G U R E 2 LINC00461 up-regulates HOXA10 by binding to miR-195. A, LINC00461 is presumed to bind to the 3'untranslated regions (3'UTR) of miR-195 using RNA22. B, A combination of miR-195 mimic and pLINC00461-Wt significantly reduces the luciferase activity, suggesting that LINC00461 binds to the 3'UTR of miR-195 (*P < .05, vs the control group). C, miR-195 is presumed to bind to the 3'untranslated regions (3'UTR) of HOXA10 using RNA22; D, a combination of miR-195 mimic and pHOXA10-Wt significantly reduces the luciferase activity, suggesting that LINC00461 binds to the 3'UTR of HOXA10 (*P < .05, vs the control group); E, LINC00461 is predominantly localized in the cytoplasm observed by FISH test. F, Ago2 expression is detected by Western blot analysis (*P < .05, vs the IgG group). G, LINC00461 directly bound to Ago2 by RIP assay. H, LINC00461 binds to miR-195, determined by RNA pull-down test (*P < .05, vs the miR-195-Wt group). The results of luciferase activity, Western blot analysis and RNA pull-down test were expressed as mean value ± standard deviation. Comparisons between two groups were conducted by t test. The experiment was run in triplicate independently. The comparisons among three groups were conducted by one-way ANOVA. miR-195, microRNA-195; ANOVA, analysis of variance; Wt, wild-type; HOXA10, Homeobox A10; and Mut, mutant type In the following experiments, miR-195 was observed to be down-regulated but HOXA10 was up-regulated in lung adenocarcinoma. In addition, lung adenocarcinoma cells showed decreased HOXA10 and increased miR-195 following LINC00461 silencing.
When miR-195 expression was decreased by inhibitor, HOXA10 expression as well as cell proliferation, invasion, migration and radiosensitivity were promoted. MiR-195 expression has been reported to be often decreased in various types of cancers, such as hepatocellular carcinoma. 10 In addition, a prior study highlighted the functionality of miR-195 as a tumour suppressive miRNA in NSCLC. 26 31 We also gathered evidence implicating that the down-regulated LINC00461 enhanced radiosensitivity, which is consistent with the findings of a previous study demonstrating that the lower levels of Bcl-2 and higher levels of Bax were indicative of enhanced radiosensitivity. 32 In addition, inhibition of MMP-2 was also found to enhance radiosensitivity in lung cancer cells. 33

| CON CLUS ION
Our study showed that LINC00461 functioned as a competing endogenous RNA to regulate HOXA10 expression by binding to miR-195 in lung adenocarcinoma cells. Furthermore, silencing of LINC00461 and restoration of miR-195 expression inhibited cell growth and induced apoptosis in lung adenocarcinoma cells.
Altogether, our findings provided an insight on the potential of LINC00461 as a sensitizer for lung adenocarcinoma radiotherapy.
Nevertheless, these data should be further validated in independent cohorts and prospective trials. In addition, further studies are still required to study whether miR-195 and LINC00461 expressions are associated with the presence of KRAS and EGFR in lung adenocarcinoma and to better elucidate the modulatory role of KRAS and EGFR in NSCLC.

ACK N OWLED G EM ENTS
The authors want to show their appreciation to reviewers for their helpful comments.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest concerning this study.

AUTH O R S CO NTR I B UTI O N
JQH participated in the study design and experimental work. YJW, HMZ and XDL participated in sample collection and data analysis.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.