Long non‐coding RNA SNHG6 promotes cell proliferation and migration through sponging miR‐4465 in ovarian clear cell carcinoma

Abstract Dysregulation of small nucleolar RNA host gene 6 (SNHG6) exerts critical oncogenic effects and facilitates tumourigenesis in human cancers. However, little information about the expression pattern of SNHG6 in ovarian clear cell carcinoma (OCCC) is available, and the contributions of this long non‐coding RNA to the tumourigenesis and progression of OCCC are unclear. In the present study, we showed via quantitative real‐time PCR that SNHG6 expression was abnormally up‐regulated in OCCC tissues relative to that in unpaired normal ovarian tissues. High SNHG6 expression was correlated with vascular invasion, distant metastasis and poor survival. Further functional experiments demonstrated that knockdown of SNHG6 in OCCC cells inhibited cell proliferation, migration and invasion in vitro as well as tumour growth in vivo. Moreover, SNHG6 functioned as a competing endogenous RNA (ceRNA), effectively acting as a sponge for miR‐4465 and thereby modulating the expression of enhancer of zeste homolog 2 (EZH2). Taken together, our data suggest that SNHG6 is a novel molecule involved in OCCC progression and that targeting the ceRNA network involving SNHG6 may be a treatment strategy in OCCC.


| INTRODUC TI ON
Epithelial ovarian cancer (EOC) is the most lethal gynaecological malignancy in the world, and its incidence has increased in the last decade. [1][2][3] Of the several subtypes of EOC, ovarian clear cell carcinoma (OCCC) represents 5%-25% of all EOCs depending on geographic location, and gene expression studies support the idea that OCCC is distinct from other EOCs, with a poorer prognosis due to a lower response rate to anticancer drug treatment. [3][4][5] Identifying novel therapeutic targets and establishing new treatment strategies for OCCC is thus important.
In recent years, long non-coding RNAs (lncRNAs) have been reported as a category of non-coding RNAs with a length of greater than 200 nucleotides that do not encode proteins, and the emerging impact of lncRNAs in cancer initiation and progression has been discovered in diverse types of cancer, including OCCC. 6 is a novel molecule involved in OCCC progression and that targeting the ceRNA network involving SNHG6 may be a treatment strategy in OCCC.

K E Y W O R D S
cell migration, EZH2, miR-4465, ovarian clear cell carcinoma, SNHG6 evidence indicates that lncRNAs are involved in a variety of biological processes such as cell apoptosis, proliferation, metastasis, chemotherapeutic drug resistance and differentiation, suggesting that lncRNAs can be useful diagnostic markers or therapeutic agents for cancers. 6,9 For example, overexpression of MALAT1 and PVT1 promotes cancer cell proliferation and survival in gastrointestinal tumours. [10][11][12] However, lncRNA-p21 was reported to be down-regulated and to suppress the growth and metastasis of cancer cells. 13,14 Currently, some small nucleolar RNAs (snoRNAs), a subclass of lncRNAs, exhibit differential expression patterns in various human cancers and demonstrate the ability to affect cell transformation, tumourigenesis and metastasis. 15,16 Small nucleolar RNA host gene 6 (SNHG6), also known as U87HG, is a recently identified lncRNA shown to be a potential oncogene involved in cell proliferation and epithelial-mesenchymal transition (EMT) progression in many cancers. [17][18][19][20][21][22] However, the activities of SNHG6 related to OCCC tumourigenesis have not been well characterized, prompting us to explore the role of SNHG6 in human OCCC. In this study, we revealed that SNHG6 was overexpressed in OCCC tissues and that this lncRNA promoted cell proliferation, migration and invasion in vitro as well as tumour growth in vivo. Mechanistically, SNHG6 facilitated the development of OCCC by sponging miR-4465, which targets enhancer of zeste homolog 2 (EZH2). In summary, our study revealed the role of SNHG6 and first revealed that SNHG6 could sponge miR-4465 in OCCC.

| Human samples and tissue handling
This study was conducted with the understanding and written consent of each individual. The study methodologies conformed to the standards established by the Declaration of Helsinki. All human tissues were collected using the protocols approved by the Human

| Cell lines and culture conditions
Cells were obtained as follows: the HEK-293T and ES-2 cell lines were purchased from the Cell Bank of Type Culture Collection (CBTCC, Chinese Academy of Sciences, Shanghai, China). The RMG-1, TOV21G, OVCA420 and OVISE cell lines were purchased from Jingdu Biotech (Shanghai, China). All six cell lines were cultured in DMEM (HyClone). All media were supplemented with 10% foetal bovine serum (10% FBS, Gibco), 100 U/mL penicillin (Gibco) and 100 mg/mL streptomycin (Gibco), and all cell lines were maintained at 37°C in a humidified atmosphere of 5% CO 2 .

| RNA isolation, reverse transcription and quantitative real-time PCR
As described in previous reports, total RNA was extracted from the tissue samples and cell lines using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. Reverse transcription (RT) and quantitative real-time PCR (qRT-PCR) kits (Takara, Dalian, China) were utilized to evaluate the mRNA expression levels of the indicated genes. PCR primers were designed and synthesized using a primer design tool (Vector NTI; The primers used in this study are listed in Table S1). The relative quantification value for each target gene was expressed as 2 −ΔΔCT . β-Actin was used as the internal reference for the mRNA expression, and U6 was used as the internal reference for miRNA expression.  Table S2. miR-4465 mimics were purchased from Synbio Technologies (Suzhou, China).

| Cell proliferation assays
A total of 2 × 10 3 cells per well were seeded in 96-well plates 24 hours before the experiment. TOV21G and RMG-1 cells were transfected with targeted siRNAs or scrambled siRNA. Proliferation was measured using a Cell Counting Kit-8 (CCK-8) kit (Dojindo, Japan) according to the manufacturer's protocol. All experiments were performed in triplicate. Cell proliferation curves were plotted using the absorbance at each time point.

| Colony formation assay
Cells were digested with trypsin into single-cell suspensions at 48 hours after transfection. For the colony formation assay, a sample of 2000 cells was plated into six-well plates and incubated in the appropriate medium supplemented with 10% FBS at 37°C. After 2 weeks, cells were fixed and stained with 0.1% crystal violet, and visible colonies were manually counted. Triplicate wells were measured for each treatment group.

| Cell wound healing and invasion assays
For the wound healing assay, cells were seeded into six-well plates and allowed to grow to 90%-95% confluence. A single scratch wound was created 6 hours after siRNA transfection. Cells were washed with PBS to remove cell debris, supplemented with serum-free medium and monitored. Images were captured by phase contrast microscopy at 0, 24, 36 and 48 hours after wounding. The experiments were repeated independently in triplicate.

| Western blotting
Cells were lysed in RIPA buffer (Sigma-Aldrich) supplemented with protease inhibitor (Roche, Basel, Switzerland) and phosphatase inhibitor (Roche). The protein concentration was measured using a BCA protein assay kit (Thermo Scientific, USA). Rabbit anti-EZH2, anti-MMP2, anti-MMP9, anti-N-Cadherin and anti-E-Cadherin primary antibodies were purchased from Cell Signaling Technology.

| Subcellular fractionation
Separation of the nuclear and cytosolic fractions of TOV21G and RMG-1 cells was performed with a PARIS kit (Life Technologies; Thermo Fisher Scientific, Inc), according to the manufacturer's protocol.

| Statistical analysis
All statistical analyses were performed with spss 18.0 (IBM, SPSS, Chicago, IL). The significance of the differences between the groups was estimated using Student's t test, the chi-squared test or the Wilcoxon test as appropriate. The overall survival (OS) and DFS rates were calculated using the Kaplan-Meier method with the log-rank test for comparison. A value of P < 0.05 indicated a significant difference.

| SNHG6 is up-regulated in OCCC and is associated with poor prognosis
To explore the role of SNHG6 in OCCC, we first examined its expression in 48 patients with OCCC using qRT-PCR. The results showed higher SNHG6 expression in OCCC tissues than in normal ovarian tissues (P = 0.0139; Figure 1A). In addition, clinicopathological correlation analysis was conducted, and the 48 OCCC patients were divided into the high-and low-expression groups based on the median expression value. As shown in Table 1, high-SNHG6 expression was strongly associated with vascular invasion and distant metastasis, suggesting that SNHG6 may be involved in OCCC invasion and metastasis. The OS and progression-free survival (PFS) curves were then plotted according to the SNHG6 expression level using the Kaplan-Meier method. As shown in Figure 1B and 1, high-SNHG6 expression was significantly correlated with both shorter OS (P = 0.039) and shorter PFS (P = 0.032) times.

| SNHG6 promotes the proliferation of OCCC cells
Aiming to identify a function of SNHG6, we profiled its expression in a panel of OCCC cell lines. SNHG6 expression was then down-or up-regulated in TOV21G, RMG-1, ES-2 and OVCA420 cell lines in accordance with their high-or low-endogenous SNHG6 expression ( Figure 2A). Transfection with SNHG6 siRNAs or the overexpression plasmid significantly down-or up-regulated the expression of SNHG6 respectively ( Figure 2B and 2). As demonstrated by the CCK-8 assays, two of the three independent siRNAs targeting SNHG6 significantly decreased the proliferation of TOV21G and RMG-1 cells ( Figure 2D), whereas the overexpression of SNHG6 led to a significant increase in the growth of ES-2 and OVCA420 cells ( Figure 2E). Consistent with these results, repression of SNHG6 significantly inhibited the colony-forming ability of OCCC cells ( Figure 2F), whereas SNHG6 overexpression accelerated OCCC cell colony formation compared with that in the control groups ( Figure 2G).

| SNHG6 enhances the invasion and migration of OCCC cells
To determine whether SNHG6 promotes OCCC cell invasion and migration, we performed Transwell and wound healing assays. Knockdown of SNHG6 resulted in a decrease in the number of invaded cells relative to the number of invaded control cells (P < 0.05; Figure 3A); conversely, SNHG6 induction increased the invasive ability of ES-2 and OVCA420 cells (P < 0.05; Figure 3B). Moreover, the migration capacity was decreased when SNHG6 was down-regulated (P < 0.05; Figure 3C). The migration assay results were verified by pcDNA3.1-SNHG6 transfection (P < 0.05; Figure 3D). Accordingly, the Western blotting results showed that SNHG6 knockdown down-regulated the expression of invasion-promoting proteins such as N-cadherin, MMP2 and MMP9 and up-regulated the expression of the invasion-suppressing protein E-cadherin ( Figure 3E). Collectively, these results suggest that SNHG6 may enhance OCCC cell migration and invasion.

| SNHG6 knockdown inhibits the growth of OCCC xenografts in vivo
To provide in vivo evidence for the oncogenic role of SNHG6 in OCCC, we established cell lines with stable SNHG6 knockdown for use in a xenograft mouse model; the knockdown efficiency is shown in Figure 4A.
Ten mice were injected subcutaneously with TOV21G-NC and TOV21G-shRNA cells, and all developed detectable tumours ( Figure 4B).
However, knockdown of SNHG6 markedly reduced the increase in the  tumour volumes in the xenograft mouse model ( Figure 4C). Additionally, the size and weight of SNHG6 knockdown tumours were lower than those of NC tumours ( Figure 4D and 4). In addition, Ki-67 staining, which represents the proliferation index, was lower in the SNHG6 knockdown groups than in the control groups ( Figure 4F). Therefore, SNHG6 may promote OCCC tumourigenesis both in vitro and in vivo.

| SNHG6 functions as a sponge for miR-4465 in OCCC cells
To understand the mechanism by which SNHG6 contributes to the malignant phenotypes of OCCC cells, we assessed SNHG6 localization, because the activities of lncRNAs mainly depend on their subcellular distribution. Analysis of the cytoplasmic and nuclear RNA fractions of OCCC cells revealed that SNHG6 was localized preferentially in the cytoplasm ( Figure 5A). Previous literature reported that cytoplasmic lncRNAs can bind directly to miRNAs and function as sponges or as competing endogenous RNAs (ceRNAs) to control the availability of miRNAs for binding to their target mRNAs.

| SNHG6 influences the expression of the miR-4465 target gene EZH2
To identify the targets of SNHG6 ceRNA, target prediction tools (TargetScan; http://www.targe tscan.org/vert_50) were used and a literature review was conducted to evaluate the potential miR-4465 target genes. Among these potential target genes, EZH2 was selected as the predicted target. Enhancer of zeste homolog 2 is also considered an important factor in EMT, which is associated with tumour growth and metastasis. We first examined whether SNHG6 could influence EZH2 expression and found that depletion of SNHG6 down-regulated EZH2 mRNA and protein expression in TOV21G and RMG-1 cells ( Figure 6A and 6), suggesting that SNHG6 influences EZH2 by sponging miR-4465. To further assess whether EZH2 was a direct target of miR-4465, luciferase reporter plasmids containing the WT and MUT EZH2 binding sites were designed ( Figure 6C). Cotransfection of the luciferase reporter plasmid containing the WT EZH2 binding site with miR-4465 mimics into HEK-293T cells contributed to a decrease in reporter activity ( Figure 6D). In addition, EZH2 expression was decreased by miR-4465 mimics in TOV21G and RMG-1 cells ( Figure 6E). These results suggest that SNHG6 promotes OCCC cell proliferation and invasion through reducing miR-4465 binding to EZH2 mRNA.
F I G U R E 3 Small nucleolar RNA host gene 6 (SNHG6) enhances cell invasion and migration in vitro. (A,B) The invasion potential of cells transfected with the SNHG6 small interfering RNAs (siRNAs) or the overexpression plasmid was assessed using a Transwell assay. The scale bar represents 50 μm. (C,D) The migration ability of cells with altered SNHG6 expression was evaluated using a wound healing assay; images of TOV21G, RMG-1, ES-2 and OVCA420 cells were captured at 0 and 48 h post wounding. The scale bar represents 200 μm. E, Western blot analysis was used to determine the N-cadherin, E-Cadherin, MMP2 and MMP9 expression levels. β-Actin or cytoplasmic 1 (ACTIN) was used as the reference. *P < 0.05, **P < 0.01, ***P < 0.001 vs negative control (NC)

| SNHG6 facilitates tumour proliferation and metastasis via miR-4465
As we found that SNHG6 directly binds to miR-4465, we next investigated the coregulation of OCCC cell proliferation and metastasis by SNHG6 and miR-4465. The CCK-8 and colony formation assay results showed that the proliferation of TOV21G and RMG-1 cells was increased when SNHG6 expression was up-regulated and inhibited when miR-4465 was overexpressed; however, the promotive effect of the SNHG6 overexpression plasmid on OCCC cell growth was partially restored by transfection with miR-4465 mimics (P < 0.05; Figure 7A and 7). In addition, the Transwell assay F I G U R E 5 Small nucleolar RNA host gene 6 (SNHG6) functions as a sponge for miR-4465. A, The cytoplasmic and nuclear RNA fractions were isolated from TOV21G and RMG-1 cells. Small nucleolar RNA host gene 6 was distributed mainly in the cytoplasm. β-Actin was the cytoplasmic internal control, and U6 was the nuclear internal control. The values are presented as the means ± SEMs. B, Venn diagram shows the results of the combination analysis to identify potential targets of SNHG6 (miRcode: http://www.mirco de.org/; starBase V2.0: http:// starb ase.sysu.edu.cn/mirLn cRNA.php). C, The relative miR-4465 levels in TOV21G and RMG-1 cells with SNHG6 knockdown are shown. D, The binding region between miR-4465 and SNHG6 was predicted, and diagrams of the luciferase reporter plasmids containing the wild-type (WT) (SNHG6) or mutant SNHG6 (MUT-SNHG6) sequence are shown. E, Small nucleolar RNA host gene 6 cDNA containing the putative miR-4465 binding site or the corresponding mutant sequence was cloned downstream of the luciferase gene in the pGL3-Basic vector; the resulting plasmid was designated RLuc-SNHG6. Luciferase reporter plasmids containing the WT or mutant SNHG6 sequence were then cotransfected into HEK-293T cells along with miR-4465 mimics in parallel with an empty plasmid vector. Luciferase activity was determined using a dual luciferase assay and is shown as the relative luciferase activity normalized to Renilla luciferase activity. F, The differential expression of miR-4465 in ovarian clear cell carcinoma (OCCC) tissues (n = 48) and unpaired normal ovarian tissues (n = 44) was analysed. G, Pearson correlation curves are shown, revealing the negative relationship between SNHG6 and miR-4465 in OCCC. **P < 0.01, ***P < 0.001 vs negative control (NC). ns, not significant results showed that cell invasion was inhibited after transfection with miR-4465 mimics, suggesting the suppressive effect of miR-4465 on tumour invasion. As expected, the pro-metastatic effect of SNHG6 in TOV21G and RMG-1 cells was rescued by cotransfection with miR-4465 mimics (P < 0.05; Figure 7C).

| D ISCUSS I ON
Moreover, we confirmed that SNHG6 promoted OCCC cell proliferation and metastasis, which was restored by miR-4465 mimics. These data suggest that SNHG6 exerts its pro-tumour effects at least in part by regulating miR-4465 expression.
In conclusion, this study is the first to investigate a potential mechanism of the SNHG6/miR-4465 axis in OCCC progression.
Increased expression levels of SNHG6 are associated with tumour progression and are inversely correlated with prognosis. In addition, SNHG6 functions as a ceRNA, regulating EZH2 expression by competitively binding miR-4465. These findings provide mechanistic insight into the role of SNHG6 in promoting OCCC metastasis and suggest that SNHG6 is an important prognostic factor and therapeutic target.

ACK N OWLED G EM ENTS
This work was supported in part by grants from the National Natural

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