Long non‐coding RNA SNAI3‐AS1 promotes the proliferation and metastasis of hepatocellular carcinoma by regulating the UPF1/Smad7 signalling pathway

Abstract Emerging evidence has indicated that deregulation of long non‐coding RNAs (lncRNAs) can contribute to the progression of human cancers, including hepatocellular carcinoma (HCC). However, the role and exact mechanism of most lncRNAs in tumours remains largely unknown. In the current study, we found a novel long non‐coding RNA termed SNAI3‐AS1 which was generally up‐regulated in HCC tissues compared with normal control. Higher expression of SNAI3‐AS1 was significantly correlated with shorter overall survival of HCC patients. Knockdown of SNAI3‐AS1 inhibited the proliferation and metastasis of HCC cells in vitro, whereas overexpression of SNAI3‐AS1 promoted the proliferation and metastasis of HCC cells. Further investigations showed that SNAI3‐AS1 could affect HCC tumorigenesis by binding up‐frameshift protein 1 (UPF1), regulating Smad7 expression and activating TGF‐β/Smad pathway. Functionally, SNAI3‐AS1 promoted HCC growth and metastasis by inducing tumour epithelial to mesenchymal transition (EMT). Taken together, these findings showed that SNAI3‐AS1 promotes the progression of HCC by regulating the UPF1 and activating TGF‐β/Smad pathway.


| INTRODUC TI ON
Hepatocellular carcinoma (HCC) is the dominant histological type of primary malignancies in liver and the second highest leading cause of cancer-related death worldwide. 1 Although great progress has been made in the diagnosis and treatment of HCC, it still has a high post-operative recurrence rate due to tumour metastasis and chemoresistance. 2,3 Therefore, it is urgent to explore potential biomarkers and molecular mechanisms of HCC tumorigenesis, which contribute to early diagnosis and effective therapy.
In recent years, the role of protein-coding genes in the development and progression of HCC has been extensively studied, and some molecular markers that can be used to determine the prognosis of HCC have been explored. [4][5][6] However, the molecular mechanism of HCC has not yet been fully elucidated and needs to be further explored. Long non-coding RNAs (lncRNAs) are a class of RNA transcripts which are more than 200 nucleotides in length and have no protein-coding potential. New evidence showed that lncRNAs play important regulatory roles in tumorigenesis or cancer progression, hence, lncRNAs have gained more and more attention and may present new opportunities for disease diagnosis and treatment. [7][8][9] Meanwhile, the number of lncRNAs has been found to be aberrantly expressed in multiple human cancers, including HCC. [10][11][12][13][14] These aberrant expressed lncRNAs regulate a wide range of important pathophysiological processes which associated with tumorigenesis, metastasis, prognosis. As a new type of regulatory RNA molecule, lncRNA has a diverse subcellular location and plays important roles in many aspects of cell activity. LncRNAs interfere gene expression in transcriptional or posttranscriptional process by direct or indirect ways. 15,16 Therefore, studying the role of lncRNAs in HCC may help to further understand HCC carcinogenesis.
In this study, we identified a novel HCC related lncRNA, termed

| Clinical samples and cell lines
A total of 46 pairs fresh HCC tissue specimens and matched adjacent non-malignant tissues (3-5 cm distal to the edge of tumour) were collected from patients underwent resection of primary HCC at the First Affiliated Hospital of Xi'an Jiaotong University. None of patients received any chemotherapy or radiotherapy treatments before surgery. Written informed consent from all patients and approval of the Hospital Ethic Committees was obtained.
Kaplan-Meier and log-rank analyses were used for survival analysis. All human HCC cells involved and immortalized human hepatic cell LO2 were purchased from the Type Culture Collection of the Chinese Academy of Sciences. All cells were cultured in DMEM/ high glucose (Hyclone), supplemented with 10% foetal bovine serum (FBS, Gibco) and 100 μg/mL streptomycin and 100 U/mL penicillin (Hyclone), and maintained at 37°C in humidified incubator with 5% CO 2 .

| RNA isolation, and quantitative real-time PCR
The total RNA was extracted from tissues or cultured cells according to the instruction of Trizol Reagent (Invitrogen). Then, the cDNAs were synthesized following the protocol of PrimeScript™ RT Master Mix (Takara). Quantitative real-time PCR was conducted using SYBR Premix Ex TaqTM II (Takara) on Thermal Cycler CFX6 System (BioRad). β-actin was used as internal reference. The relative expression of genes was calculated using the 2 −ΔΔCt method. The primers used in this study were presented in Supplementary Table 1.

| Transfection of cell lines
A shRNA targeting SNAI3-AS1 was designed by Genechem to construct stable SNAI3-AS1 knockdown cell lines. SNAI3-AS1 and UPF1 were cloned into the expression vector pCMV (Invitrogen) for over-
After transfection for 24 hours, 48 hours, 72 hours, 96 hours, 10 μL of 5 mg/mL MTT was added into each well and then cultured for 4 hours in incubator. The supernatant was then discarded and 150 μL of DMSO was added to dissolve the crystal. The optical density (OD) was measured by EnSpire Multimode Plate Reader (PerkinElmer) at 490 nm. Triplicate experiments were performed for each assay.

| Colony formation assay
Transfect the cells with the indicated reagents. After 24 hours of routine incubation, cells were re-plated at a density of 500 cells/6 cm plates and maintained in DMEM. After 2 weeks, when clones formed by single cell possessed at least 50 cells, the clones were fixed with methanol and stained with 0.1% crystal violet in PBS for 15 minutes.
The colony formation was determined by counting the number of stained colonies.

| Transwell assay
After 48 hours of transfection, cells were respectively seeded into 8 μm

| Wound healing assay
Appropriate cell density is required which need to attain 90% after 24 hours of transfection in 6-well plates. Wound was scratched by 10 μL sterile tip and then washed off the floated cells. The wound size was measured and photographed at 0 hour, 24 hours and 48 hours.

| Western blot analysis
The transfected cells were lysed by using RIPA (Beyotime) supplemented with proteinase and phosphatase inhibitors. Protein samples were loaded for electrophoresis (5% gel for concentration and 10% for separation), and then transferred onto 0.45 μm or 0.22 μm pore size PVDF membrane (Merck Millipore). After blocking with 5% non-fat milk for 1 hour, the membrane was incubated with specific primary antibodies (Supplementary Table 2) at 4°C overnight. The next day washed the membrane and then incubated with secondary antibodies (Zhuangzhi Biology, dilution rate of 1:5000) for 1 hour at room temperature. Proteins bands were detected by using ECL immunoblotting kit (Millipore) according to the manufacturer's protocol.

| Luciferase reporter assay
Bioinformatics tools were used to analyse the SNAI3-AS1 binding sites on UPF1. Plasmid GV208-SNAI3-AS1-WT (Genechem) was constructed by inserting the sequence of SNAI3-AS1 into the GV208 vector, and plasmid GV208-SNAI3-AS1-Mut (Genechem) was established by inserting SNAI3-AS1 sequence with the UPF1 binding site muted by site-specific mutagenesis. Cells were cultured in 96-well plates and proliferated to 60%-80% confluence before transfection. Plasmid SNAI3-AS1-WT or Mut together with UPF1 plasmid or negative control were cotransfected using the Lipofectamine 2000 reagent (Invitrogen). After transfection for 48 hours, the Dual Luciferase Assay Kit was conducted to examine the luciferase activity according to the manufacturer's instructions. Renilla luciferase activity was used as control.

| RNA immunoprecipitation (RIP)
RIP assays were performed using a Millipore EZ-Magna RIP RNA-Binding Protein Immunoprecipitation kit (Millipore) according to the manufacturer's protocol. Antibodies used for RIP included rabbit polyclonal IgG (Millipore) and antibodies to UPF1 (Abcam). RIP-PCR was performed as qRT-PCR using total RNA as input controls.

| Immunofluorescence (IF)
Cells were cultured on glass coverslips for 24 hours to confluence, then fixed in 4% paraformaldehyde at room temperature for 15 minutes, and then permeabilized using 0.5% Triton X-100 and blocked for 1 hour with 10% goat serum. Following incubated cells with primary antibody 4°C overnight and secondary antibodies with a appropriate dilution for 1 hour (Supplementary Table 2). After washing with PBS, DAPI was applied for nucleus staining. Images were collected using invert fluorescent microscope (Leica). Antibodies used in this study were presented in Supplement Table 2.

| Statistical analysis
All data were analysed using SPSS 23.0 and Graphpad Prism 7.0. The difference between two groups was compared by Student's t test.
Correlation between two groups was analysed using Pearson's correlation coefficient analysis. The Kaplan-Meier test was used to assess prognosis. A value of P < 0.05 was considered significant.

| SNAI3-AS1 is up-regulated in HCC and correlated with poor progression
To investigate the roles of SNAI3-AS1 in HCC, SNAI3-AS1 expression was examined in HCC tissues and cell lines. QRT-PCR revealed that expression of SNAI3-AS1 was significantly elevated in 69.5%

| Knockdown of SNAI3-AS1 represses the proliferation, invasion and migration of HCC cells in vitro
In vitro experiments were performed to determine the effect of SNAI3-

| SNAI3-AS1 overexpression promotes the proliferation, invasion and migration of HCC cells in vitro
We further examined the role of SNAI3-AS1 by assessing the ef-

| SNAI3-AS1 promoted HCC tumorigenesis by interacts with UPF1
Recently, several studies have found that many lncRNAs are involved in multiple regulation pathways through their interaction with RNA-binding proteins (RBPs). To test this hypothesis, we The result showed that SNAI3-AS1 contains a potential binding site for UPF1. UPF1 is known to interact with many RNA substrates and promote mRNA stability. 17 Next, we constructed luciferase reporter vectors of SNAI3-AS1, and luciferase reporter assay result showed cotransfection cells with pCMV-SNAI3-AS1-WT and UPF1 vector significantly inhibited luciferase reporter activity, however, pCMV-SNAI3-AS1-Mut in UPF1 putative targeting sites did not resulted in these effects ( Figure 4A). To further validate the interaction between SNAI3-AS1 and UPF1, we performed RNA immunoprecipitation (RIP) with an antibody against UPF1 using cell extracts from HepG2 and Hep3B HCC cell lines.
We observed an enrichment of SNAI3-AS1 with UPF1 antibody as compared to the non-specific antibody (IgG control; Figure 4B).
Meanwhile, Pearson's correlation analysis suggested that UPF1 expression was inversely correlated with SNAI3-AS1 in HCC tissues ( Figure 4C), and knockdown of SNAI3-AS1 could increase UPF1RNA and protein levels in HepG2 and Hep3B cells ( Figure 4D,E). We further studied the roles of SNAI3-AS1 and UPF1 in HCC cell invasion using transwell invasion assay. The results showed that UPF1 suppressed HepG2 and Hep3B cells invasion, whereas SNAI3-AS1 promoted invasion. However, the inhibitory effect of SNAI3-AS1-shRNA on HCC cell invasion could be partially restored by UPF1 inhibition ( Figure 4F,G). These observations suggest that SNAI3-AS1 knockdown could suppress HCC cell invasion by regulating UPF1 expression.

| SNAI3-AS1 induced EMT in HCC cells
Since increasing evidence identified that tumour migration and invasion is tightly involved in EMT, so we speculated that SNAI3- significantly reduced the expression of two matrix metalloproteinases MMP-2 and MMP-9 that are closely correlated with metastasis ( Figure 5E). Overall, these data indicated that SNAI3-AS1 induced EMT in HCC cells.

Clinical factors
No. of cases   in endogenous UPF1 expression ( Figure 6A). Western blotting confirmed that Smad7 expression level was up-regulated when UPF1 was silenced in HCC cells ( Figure 6B). While the Smad7 expression level was down-regulated when UPF1 was overexpressed in HCC cells ( Figure 6C). Pearson's correlation analysis showed that Smad7 mRNA expression was significantly negatively correlated with UPF1 in HCC tissues ( Figure 6D). Previous studies had demonstrated the key role of Smad7 in the TGF-β pathway. As shown in our study, UPF1 could affect the expression of Smad7 in HCC. Therefore, we speculated that SNAI3-AS1 could affect the TGF-β pathway by regulating UPF1. As shown in Figure 6E, SNAI3-AS1 knockdown significantly decreased phosphorylation of Smad2/3, whereas total Smad2/3 expression level was similar between groups. In addition, the rescue experiments were conducted to prove that SNAI3-AS1 promotes tumour EMT by regulating UPF1. As shown before, knockdown of SNAI3-AS1 increased epithelial marker E-cadherin and decreased mesenchymal markers Ncadherin and vimentin, however, these effects could be partially restored by UPF1 inhibition ( Figure 6F). All these results demonstrated that SNAI3-AS1 promotes HCC tumorigenesis by binding UPF1, regulating Smad7 expression, and inducing activation of the TGF-β/Smad pathway (Figure 7).

| D ISCUSS I ON
Genomic studies revealed that less than 2% of the human genome has protein-coding functions. 18 High-resolution microarrays and massively sponge of let-7c-5p, which indicated that CDKN2B-AS1 may be a potential prognostic biomarker and a candidate target for HCC therapy. 27 In the present study, we focused on lncRNA SNAI3-AS1

| CON CLUS ION
In conclusion, the current study demonstrated that highly ex- mechanism on tumorigenesis of HCC, but it is worthwhile to reveal that SNAI3-AS1 may be a novel prognostic factor and potential therapeutic target for HCC.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA ACCE SS I B I LIT Y
The data in this study are available.