LINC01287 regulates tumorigenesis and invasion via miR‐298/MYB in hepatocellular carcinoma

Abstract Recently, it was reported that long non‐coding RNAs (lncRNAs) participated in promoting hepatocellular carcinoma (HCC) initiation and progression. Herein, we reported that the expression level of LINC01287 was elevated in HCC cell lines and tissues. LINC01287 down‐regulation inhibited HCC cells growth and invasion both in vitro and in vivo. LINC01287 exerted as a ceRNA and negatively regulated miR‐298 expression. MYB was identified as a downstream target of miR‐298. The miR‐298/MYB axis mediated LINC01287's effect on HCC. To the best of our knowledge, our findings provided the first evidence that LINC01287 functioned as an oncogene in HCC. LINC01287 may be a candidate prognostic biomarker and a target for new therapies in HCC patients.


| INTRODUCTION
Hepatocellular carcinoma (HCC) is the fifth most common cancer and third leading cause of cancer-related deaths worldwide. 1 Many patients that are diagnosed with HCC have already entered the late stage of the disease. Furthermore, metastasis after surgical resection and a high frequency of tumour recurrence contributes to a poor prognosis of HCC patients. 2 It is therefore important to develop novel strategies for the diagnosis and treatment of HCC.
About 2% of the human genome accounts for protein-coding genes. The majority of transcripts consists of non-coding RNAs (ncRNAs) 3 that can be grouped into the following classes depending on their transcript size: long non-coding RNAs (lncRNAs) and small ncRNAs. 4 LncRNAs are transcripts with a length greater than 200 nt.
Abnormal expression of lncRNAs often contributes to tumour initiation, growth and metastasis. 5 A number of articles have demonstrated that lncRNAs are dysregulated in many cancers, including HCC. 6 Although a great number of lncRNAs have been annotated, the role and molecular regulatory mechanisms of lncRNAs in HCC still need to be further clarified.
MicroRNAs (miRNAs) are a small and single-stranded RNAs, and they usually negatively regulate the expression of the down-stream gene via binding at the 3′-untranslated regions (UTRs). 7 Mounting documents demonstrate that abnormal expression of miRNAs is involved in tumorigenesis. 8 Interestingly, literature finds that lncRNAs could function as ceRNAs that compete for miRNA binding, thus derepressing the expression of miRNA-targeted mRNAs. 9 The lncRNA-miRNA-mRNA regulatory network is suggested to play a critical role in cancer tumorigenesis and progression. 10,11 We explored the role and underlying mechanism of LINC01287 in HCC. We also explore the interaction between LINC01287 and miR-298.

| Cell transfection, lentivirus production and transduction
The cell transfections were carried out using lipofectamine 2000 reagent (Invitrogen, Carlsbad, USA) according to manufactures instructions. siRNA against MYB (Ruibio, Guangzhou, China) and a non-targeting siRNA control (Ruibio) were used to knockdown gene expression.
shRNA sequences targeting LINC01287 were AAGCATTGTA-GACCTGGCTGCTGAA. The sequences were cloned into the pGFP-C-shLenti vector according to manufacturer's instructions (Origene, Rockville, MD, USA). Then, the viruses were packaged using 293T cells according to standard protocol. HCC cells were infected with virus particles plus 6 μg/mL Polybrene (Sigma-Aldrich, St. Louis, MO, USA).

| Quantitative real-time PCR assay
Total RNA was extracted from HCC cells using TRIzol reagent (Invitrogen) according to the manufacturer's instructions. The RNA was reverse transcribed to cDNA, followed by real-time PCR analyses.
The primers used in the study were listed in Table 1.
To determine the subcellular distribution of LINC01287, nuclear and cytoplasm fractions of cells were separated using the PARIS Kit (Life Technologies, Grand Island, NY, USA) according to the manufacturer's instructions. Then, the RNA was extracted from both fractions. Subsequently, real-time PCR assay was used to examine the expression ratios of specific RNA molecules between the nuclear and cytoplasm fractions.

| MTT, colony formation and cell cycle assays
For MTT assay, HCC cells were seeded in 96-well plates. After 24 hours, 5 mg/mL MTT was added to each well. The HCC cells were then incubated for another 6 hours. Subsequently, DMSO was added into each well and the absorbance was read at 490 nm under spectrophotometer.
For clone formation assay, HCC cells were seeded in 6-well culture plates. Two weeks later, the cells were fixed with paraformaldehyde, stained with haematoxylin solution and counted under a microscope.
For cell cycle assay, cells were harvested from culture dishes and washed three times by cold PBS. Then, the cells were fixed with 70% ice-cold ethanol at 4°C overnight. Finally, the cells were incubated with propidium iodide (supplemented with RNase A). The

| Boyden assay
For Boyden assay, HCC cells in serum-free DMEM were seeded in upper chambers, which were inserted into a 24-well plate. PRMI-1640 supplemented with 10% FBS were added into the lower chamber of each well. Twenty-four hours later, HCC cells that stayed on the upper surface of the membrane were removed and the cells that

| Western blot assay and immunofluorescence assay
For Western blot assay, the protein samples were transferred onto an PVDF membrane, followed by incubation for overnight in a 1:500 dilution of primary antibodies in 4°C. Subsequently, the membrane was incubated with HRP-conjugated rabbit or mouse for 1 hour at room temperature and then developed with a chemiluminescence reagent.
For immunofluorescence assay, cells were plated on culture slides. Twenty-four hours later when cells were attached to the culture slides, cells were rinsed with phosphate-buffered saline (PBS) three times and then were fixed in ice-cold methanol-acetone for 10 minutes. Subsequently, cells were blocked for 10 minutes in 5% BSA in PBS, followed by incubated with primary antibodies in PBS for 1.5 hours at room temperature. After washes with PBS three times, the slides were incubated for 40 minutes with secondary antibodies. After three washes, the slides were stained with 4-,6-diamidino-2-phenylindole (DAPI) for 10 minutes and were examined using an Olympus confocal imaging system.

| In situ hybridization assay
In situ detection of lncRNA was performed as previously described. 13

| Luciferase reporter assay
We

| In vivo tumour growth and invasion assay
All procedures involving animals were approved by the Institutional Committee on Animal Care at Gaozhou People Hospital. Huh7 cells were injected subcutaneously into both flanks of nude mice. Three mice were used in each group. Four weeks after implantation, the xenografts were removed from the mice and weighed. The tumour volume was calculated using the following formula: 4π/3 × (width/ 2) 2 × (length/2). Huh7 cells were used for invasion assay in vivo. The invasion assay was performed as previously described. 14 2.10 | Statistical analysis SPSS 13.0 (IBM, New York, USA) and GraphPad Prism 5.0 (California, USA) softwares were used for statistical analysis. The values are shown as the mean ± SEM. Analyses of different groups were performed using one-way ANOVA or two-tailed Student's t test. P < 0.05 was considered statistically significant.

| Detection of LINC01287 expression in HCC tissues along with cell lines
We focused a list of lncRNAs that were significantly elevated in HCC samples using online software program circlncRNAnet. 15 Because the expression level of LINC01287 was the highest among these lncRNAs and the function of LINC01287 has not been explored in HCC, we chose it for further study (Table S1). Subsequently, we examined the expression level of LINC01287 in 117 HCC patients. It was revealed that LINC01287 was up-regulated in HCC tissues, when compared with normal tissues ( Figure 1A). In addition, LINC01287 was up-regulated in late-stage HCC patients ( Figure 1B). In consistence, LINC01287 was elevated in HCC cell lines, compared with that in the normal human liver cell line LO2 ( Figure 1C). In addition, we confirmed that LINC01287 was up-regulated in HCC tissues when compared with the normal tissues using TCGA database ( Figure S1C, D). LINC01287 was mainly located in the cytoplasm, as determined by subcellular fractionation and real-time PCR assays ( Figure S1A).
The in situ hybridization assay further confirmed that LINC01287 was mainly located in the cytoplasm ( Figure S1B).
We analysed the relation between LINC01287 expression level and patients' clinicopathological features in  Figure 1D and E, P < 0.05).

| LINC01287 inhibition decreased HCC cell proliferation and invasion
We We further asked whether LINC01287 affected cell invasion in HCC. Inhibition of LINC01287 decreased HCC cell invasion, as revealed by Boyden assay (Figures 2F and S2D). In addition, inhibition of LINC01287 decreased MMP-2, MMP-3 and MMP-9 expression ( Figure 2G). E-cadherin expression was elevated, while N-cadherin expression and vimentin expression were down-regulated in sh-LINC01287 cells, compared to sh-ctrl cells ( Figure 2H).

| LncRNA LINC01287 negatively regulated miR-298 expression
LncRNAs usually exerted their function via acting as a sponge of miRNAs. 16 We thus asked whether there was an interaction between LINC01287 and miRNAs in HCC. We identified several microRNAs that may be regulated by LINC01287 via the online software miRDB (http://mirdb.org/miRDB/index.html) ( Figure S3A).
We examined the different expression of these microRNAs between sh-ctrl and sh-LINC01287 group in Huh7 cells. It was found that three microRNAs (miR-298, miR-4308 and miR-23c) altered significantly between sh-ctrl and sh-LINC01287 group ( Figure S3B). The binding sites of miR-298 on LINC01287 were indicated in Figure 3A.
MiRNAs directed their downstream targets through a ribonucleoprotein complex, which was called the RNA-induced silencing complex (RISC), the core component of which was Ago2. 17 RNA immunoprecipitation (RIP) assay was performed to examine whether LINC01287 and miR-298 were in the same RISC complex. We found that LINC01287 and miR-298 were enriched in Ago2 immunoprecipitates when compared to control IgG immunoprecipitates ( Figure 3B, P < 0.05). When LINC01287 was inhibited, the expression level of miR-298 was elevated in Huh7 cells ( Figure 3C, P < 0.05). Next, we tested whether miR-298 inhibited LINC01287 expression. The pre-

| LINC01287 regulated MYB expression through miR-298
We searched the potential targets of miR-298 using software Tar-getScan. These potential downstream genes were related to cell differentiation, proliferation, migration and apoptosis ( Figure S3H), as revealed by Kyoto Encyclopedia of Genes and Genomes and Gene oncology assays. MYB regulated cancer proliferation and invasion in HCC, and we chose it for further study. As indicated in Figure 4A, we cloned wild-type 3′-UTR region of MYB mRNA (including the predicted miR-298 recognition site) and the mutant type (excluded the predicted miR-298 recognition site into luciferase reporter plasmids). We observed that miR-298 decreased luciferase activity in the wild-type vector, but not that in the mutant type ( Figure 4B). Interestingly, both mRNA and protein expression of MYB was decreased by miR-298 treatment ( Figure 4C and D). In summary, these findings demonstrated that MYB was a downstream target of miR-298.
Next, we examined whether the miR-298/MYB axis mediated LINC01287's function. Western blot assay revealed that down-regulation of LINC01287 inhibited MYB expression. Nevertheless, LINC01287 down-regulation effect on MYB expression was abolished in the case of miR-298 inhibition ( Figure 4E). Interestingly, restoration of MYB could counteract LINC01287 down-regulation's effect on cell growth, cell cycle distribution and invasion ( Figure 4F-I).
In all, our findings revealed that LINC01287 could exert its function through the miR-298/MYB axis.

| LINC01287 inhibition decreased tumour growth in vivo
The effect of LINC01287 on tumour growth in vivo was tested.
HCC cell tumours grew more slowly when LINC01287 was inhibited ( Figure 5A). Compared with sh-ctrl group, the mean weight of xenograft tumours was less in the sh-LINC01287 group ( Figure 5B). The Ki-67 staining assay revealed that the proliferation index was lower in the sh-LINC01287 group, compared with sh-ctrl group ( Figure 5C).
Finally, LINC01287 down-regulation inhibited lung metastasis ability in vivo ( Figure 5D). In all, we revealed that LINC01287 promoted tumour growth and invasion in vivo.

| DISCUSSION
Recently, the role and the underlying mechanism of lncRNAs in cancer have been well documented. 18 Targeting the abnormal expression of lncRNAs may be effective in cancer treatment.
In the current study, we revealed that LINC01287 was up-regulated in HCC cell lines and tissues. LINC01287 down-regulation decreased cell proliferation, together with colony formation ability in vitro. In consistence, LINC01287 down-regulation decreased tumour growth in vivo. LINC01287 down-regulation contributed to cell cycle arrest in G1 stage. The G1/S phase checkpoint proteins (eg c-myc, cyclin D1 and CDK4) were altered when LINC01287 was inhibited. These data suggested that LINC01287 may lead to cell growth via affecting cell cycle progress. We also revealed that LINC01287 down-regulation inhibited GC cell invasion in vitro and decreased lung metastasis in vivo. The EMT plays vital role in promoting cancer cell invasion. 19 We thus asked whether LINC01287 was involved in EMT phenotype. It was revealed that LINC01287 down-regulation increased epithelial marker E-cadherin while decreased mesenchymal markers N-cadherin and vimentin expression. These data suggested that LINC01287 may promote EMT phenotype and thus lead to HCC cell invasion.
Previous study revealed that lncRNAs may act as endogenous molecular sponges to compete for miRNAs and negatively regulating miRNA expression. 20, 21 We identified several miRNAs that may In all, our data provided the first evidence that the LINC01287/ miR-298/MYB axis controlled cell growth and invasion in HCC cells.
Therapeutics that target LINC01287 may improve the treatment of HCC.

| CONCLUSIONS
In conclusion, our study demonstrated that LINC01287 was up-regulated in HCC and may function as a ceRNA to increase MYB expression by sponging miR-298, which consequently contributed to HCC growth and metastasis. Our findings indicated that LINC01287 could be a potential therapeutic target for HCC treatment.