Long non‐coding RNA VPS9D1‐AS1 enhances proliferation, invasion, and epithelial‐mesenchymal transition in endometrial cancer via miR‐377‐3p/SGK1

Abstract

Endometrial cancer (EC) is a kind of gynecologic malignancy with a rising incidence rate. This study aimed to explore the role of VPS9D1 antisense RNA1 (VPS9D1‐AS1) in EC. The expression of VPS9D1‐AS1, microRNA (miR)‐377‐3p, and serum and glucocorticoid‐regulated kinase 1 (SGK1) was detected by Quantitative Real‐Time PCR (qRT‐PCR). Cell proliferation, invasion and epithelial‐mesenchymal transition (EMT) were determined by cell counting kit‐8 (CCK‐8), 5‐Ethynyl‐2′‐Deoxyuridine (EdU) transwell, and western bolt. VPS9D1‐AS1 was predicted to sponge miR‐377‐3p via Starbase, and verified by luciferase reporter, RNA binding protein immunoprecipitation (RIP), and RNA pull‐down experiments. The clinical characteristics of VPS9D1‐AS1, miR‐377‐3p, and SGK1 were analyzed. The role of VPS9D1‐AS1 on EC tumorigenesis was assessed in xenografted nude mice. VPS9D1‐AS1 was upregulated in EC cells and tissues. Interference of VPS9D1‐AS1 inhibited growth, invasion, and EMT of EC cells. Mechanically, VPS9D1‐AS1 was a molecular sponge of miR‐377‐3p, and overexpression of miR‐377‐3p reversed VPS9D1‐AS1‐induced EC cells proliferation, invasion, and EMT. Moreover, SGK1 was confirmed to bind with miR‐377‐3p. Furthermore, overexpression of SGK1 alleviated sh‐VPS9D1‐AS1‐caused effects on EC cells. High level of VPS9D1‐AS1 and SGK1, or low miR‐377‐3p expression predicted a poor prognosis. The expression of the three genes was correlated with lymph node metastasis, pathological stage, and International Federation of Gynecology and Obstetrics (FIGO) stage, but not associated with age, ER, and PR expression. Interestingly, knockdown of VPS9D1‐AS1 suppressed EC tumor growth in mice. VPS9D1‐AS1 promoted cell invasion, proliferation, and EMT via modulating miR‐377‐3p/SGK1 axis, which provided new options for therapeutic strategies of EC.

INTRODUCTION

Endometrial cancer (EC) is a kind of gynecologic malignancy with a rising incidence rate.^1,2 Based on histopathological and endocrine factors, EC has been sorted into two types, namely, estrogen-dependent type I and non-estrogen-dependent type II.³ The standard treatments for EC in clinic are surgery and radiation, which depend on the risk of disease recurrence.^4,5 Unfortunately, some patients with advanced recurrent disease or stage endometrial cancer do not respond well to the standard therapies. Given that the prognosis of EC is unsatisfactory, the underlying mechanisms of EC remain to be understood. Hence, it is avid need to advance predictive biomarkers or therapeutic targets for EC.

Emerging studies have confirmed long non-coding RNA (lncRNA) is associated with regulating EC tumorigenesis. LncRNA NEAT1 enhances EC cell migration, invasion, and proliferation via modulating the miR-144-3p/EZH2 pathway.⁶ LncRNA HEIH increases paclitaxel tolerance in EC cells through activating MAPK signaling.⁷ LncRNA ROR accelerates proliferation of EC cells via regulating Notch1 pathway.⁸ VPS9D1-AS1 is an antisense RNA with 1637 nucleotides. It is first reported to be modulated by c-Myc in colon cancer, thus designated as c-Myc-upregulated lncRNA (MYU).⁹ More and more studies discover that VPS9D1-AS1 participates in tumorigenesis. For instance, it sponges miR-1301–3p to inhibit apoptosis and enhance growth of colon adenocarcinoma cells.¹⁰ VPS9D1-AS1 acts on the Wnt/β-catenin axis that enhances the development of esophageal squamous cell carcinoma.¹¹ Moreover, VPS9D1-AS1 modulates HuR/CDK4 signaling pathway to facilitate hepatocellular carcinoma progression.¹² Nevertheless, its role and mechanism in EC remain unclear.

This study aimed to explore the role of VPS9D1-AS1 in EC. The detailed role and underlying mechanism of VPS9D1-AS1 in the proliferation, invasion, and EMT of EC were explored both in vitro and in vivo in this study. In addition, the clinical characteristics of VPS9D1-AS1, miR-377-3p, and SGK1 were also analyzed. We hope our data can provide a new biomarker and therapeutic target for EC.

METHODS AND MATERIALS Patients data collection

The EC tissues and adjacent paracancerous tissues were acquired from EC patients at Hunan Provincial Tumor Hospital, which were consistent with the Declaration of Helsinki and approved by the Ethics Committee of Hunan Provincial Tumor Hospital. Informed consent was signed by all participators. The EC tissues (n = 92) and paracancerous tissues (n = 92) were collected from surgery, rapidly frozen, and preserved at −80°C.

Cell culture and transfection

The AN3CA, HEC-1-B, Ishikawa, HEC-1-A, and human normal endometrial cells were maintained in DMEM (Solarbio, Beijing, China) with additional 10% fetal bovine serum (FBS) (Gibco, Rockville, MD, USA) with 5% CO₂ at 37°C.

Small hairpin RNA (shRNA) for VPS9D1-AS1 (sh-VPS9D1-AS1-1, sh-VPS9D1-AS1-2 or sh-VPS9D1-AS1-3) or sh-negative control (NC) were designed and inserted into LV5-CMV-GFP-Puro lentiviral vector (GenePharma, Shanghai, China) to establish the stable transfection cells. The VPS9D1-AS1 sequences were inserted into pcDNA3 plasmid to overexpress VPS9D1-AS1. The upregulation of SGK1 was achieved by clone and insertion into pcDNA3 vector (designated as SGK1). Then, the pcDNA3-VPS9D1-AS1 (oe-VPS9D1-AS1), pcDNA3 (oe-NC), sh-VPS9D1-AS1 or sh-NC, and mimic NC, miR-377-3p mimic, pc-NC, SGK1, inhibitor-NC or miR-377-3p-inhibitor were co-transfected with Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA).

Quantitative real-time PCR (qRT-PCR) assay

Cells or tissues RNA was isolated with TRIzol (Thermo Fisher Scientific, Waltham, MA, USA) and subsequently reversely transcribed with RevertAid™ First Strand cDNA Synthesis Kit (Thermo Fisher Scientific). Thereafter, the qRT-PCR was detected by SYBR Green PCR Kit (Takara, Dalian, China). The expression levels were analyzed with 2^−ΔΔCt method. The primers were as follows: VPS9D1-AS1 (upstream 5′-AGCTTTCCTCCTTCATCGGA-3′ and downstream 5′-TGGCTTGCAGGGAAAACAC-3′; miR-377-3p (upstream 5′-ATCACACAAAGGCAACTTTTGT-3′ and downstream 5′-GGTGCAGGGTCCGAGGTAT-3′); Serum and glucocorticoid-induced protein kinase 1 (SGK1) (upstream 5′-GAGGGAGCGCTGCTTCCT-3′ and downstream 5′-ACCCAAGGCACTGGCTATTTC-3′); GAPDH (upstream 5′-CAGCTAGCCGCATCTTCTTTT-3′ and downstream 5′-GTGACCAGGCGCCCAATAC-3′); U6 (upstream 5′-CTCGCTTCGGCAGCACA-3′ and downstream 5′-AACGCTTCACGAATTTGCGT-3′).

Cell counting Kit-8 (CCK-8) assay

The cell viability was measured by CCK-8 assay (CCK-8, Sigma, St. Louis, MO, USA) based on the operation manual. The absorbance was determined at 450 nm by a microplate reader (Thermo Fisher Scientific).

5-Ethynyl-2′-deoxyuridine (EdU) assay

The EDU proliferation experiment was carried out to detect the cell proliferation. Following the transfection, cells were stained with 50 μM EdU for 2 h and subsequently stained with DAPI solution (Sigma). The EDU-positive cells numbers were then observed and counted by a fluorescence microscopy (Olympus, Tokyo, Japan).

Transwell invasion assay

The cell invasion was detected by Transwell assay. Approximately 5 × 10⁴ cells were cultured in upper chambers supplemented with matrix and FBS-free medium. Then, lower chambers were added with medium including 10% FBS. Following the cultivation of 24 h, cells in the lower chamber were immobilized with 4% paraformaldehyde, stained with 0.1% crystal violet, and captured with a microscope (Olympus).

Western blot analysis

Total proteins of tissues or cells were enriched after lysed in lysis buffer. Then, proteins were dissolved by SDS-PAGE and electrically transferred onto PVDF membranes. The antibodies included anti-E-cadherin (1/700; ab231303), anti-SGK-1 (1/1500; ab43606), anti-vimentin (1/1000; ab45939), anti-N-cadherin (1/10000; ab76011), anti-GAPDH (1/2500; ab9485, all in Abcam, Cambridge, UK), and peroxidase-conjugated anti-mouse or rabbit IgG (1/5000; Beyotime, Shanghai, China). Protein expression was visualized by an ECL assay (Beyotime).

Fluorescence in situ hybridization (FISH) assay

The location of VPS9D1-AS1 was analyzed with a FISH kit (Genepharma). The special probe for VPS9D1-AS1 (Genepharma) was designed, synthesized, amplified, and then inserted into the pMD-18 T vector. After being labeled with fluorescein isothiocyanate (FITC)-UTP (Roche, Switzerland), the samples were hybridized with probes and washed with saline-sodium citrate solution. Then, the slides were rinsed with phosphate buffer saline (PBS, Beyotime), stained with DAPI for 10 min, and then evaluated under a fluorescence microscope (Olympus).

Luciferase reporter assay

The binding sites between VPS9D1-AS1 and miR-377-3p, and the binding sites between miR-377-3p and 3′-UTR of SGK1 mRNA were forecasted by Starbase (starbase.sysu.edu.cn/starbase2/index.php). The normal sequences of VPS9D1-AS1 (VPS9D1-AS1-Wt) and SGK1 (SGK1-Wt), and the mutant sequences of VPS9D1-AS1 (VPS9D1-AS1-Mut) and SGK1 (SGK1-Mut) were generated, inserted into pGL3 luciferase vector, and then co-transfected with mimic NC or miR-377-3p mimic by Lipo3000 (Invitrogen). Following transfection of 48 h, cells were harvested to analyze the luciferase activity by a microplate reader (Thermo Fisher Scientific).

RNA pull-down assay

Ishikawa or AN3CA cells were infected with biotinylated miR-337-3p-Wt, miR-337-3p-Mut or probe NC (Invitrogen) for 48 h. Then, cells were lysed and hatched with streptavidin magnetic beads (Invitrogen). Finally, the expression level of VPS9D1-AS1 was examined by qRT-PCR.

RNA binding protein immunoprecipitation (RIP) analysis

RIP detection was performed by a Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Billerica, MA, USA) to analyze the binding of VPS9D1-AS1 and miR-377-3p. Briefly, Ishikawa or AN3CA cells were treated with RIPA lysis buffer and then hatched IgG (negative control) or human anti-Ago2 antibody conjugated with magnetic beads. Following the digestion, RNAs immunoprecipitated from samples were separated and subjected to qRT-PCR assay.

Animal

Animal experiments were in accordance with the Guide for the Care and Use of Laboratory Animals and approved by the Ethics Committee of Hunan Provincial Tumor Hospital. 15 six-week-old BALB/c nude mice (Nanjing Junke, China) were maintained 1 week for adaptive feeding. Then, mice were distributed into three groups (n = 5), including sh-NC + inhibitor-NC, sh-VPS9D1-AS1 + inhibitor-NC, and sh-VPS9D1-AS1 + miR-377-3p inhibitor. Mice in sh-VPS9D1-AS1 + miR-377-3p inhibitor group were inoculated with 4 × 10⁶ Ishikawa cells infected with sh-VPS9D1-AS1 and miR-377-3p inhibitor, while mice in sh-VPS9D1-AS1 + inhibitor-NC group were injected with the same dose of cells transfected with sh-VPS9D1-AS1 and inhibitor-NC. Additionally, mice in sh-NC + inhibitor-NC were obtained the equal dose of cells transfected with sh-NC and inhibitor-NC. Tumor weight and volume were detected every week. Tumor volume was calculated according to the following formula: volume = 1/2 × length × width². After 6 weeks, the animals were devoted by intraperitoneal injection of sodium pentobarbital (120 mg/kg), and tumors were isolated, weighed, and stored for future histopathological determination.

Immunohistochemical (IHC) assay

Tumor tissues were embedded, cut, deparaffinized, hydrated, antigen retrieved, and incubated with primary antibody including Ki-67 (1/1000, ab15580), N-cadherin (1/600, ab76011), and vimentin (1/500, ab45939, all in Abcam). Then, slices were maintained with anti-rabbit IgG at room temperature for 20 min. Ultimately, the sections were stained with DAB and the pictures were captured by a light microscope (Olympus).

TUNEL assay

An in situ cell death detection kit (Roche, Budapest, Hungary) was employed for the TUNEL assay based on the operating instructions. In brief, paraffin-embedded sections were cleared in xylene, dehydrated with the graded concentrations of ethanol, administrated with proteinase K for 15 min at room temperature, fixed with 10% formalin at room temperature for 10 min, and introduced with 2% H₂O₂ for 5 min at room temperature. Then, sections were incubated with TUNEL reaction mixture for 1 h at 37°C and treated with the biotinylated antibody, separately. The apoptosis was calculated by TUNEL positive cells/the total cells number.

Statistical analysis

GraphPad Prism 7 (GraphPad, USA) was used to analyze the data and figures. The comparison was analyzed using Student's t tests between two groups, and using one-way ANOVA analysis followed by Dunnett's post-hoc test among three groups. The survival assessment was conducted by Kaplan–Meier analysis with log-rank test. Pearson's correlation analysis was applied to determine the relevance among VPS9D1-AS1 and miR-377-3p, miR-377-3p and SGK1, as well as VPS9D1-AS1 and SGK1. p < 0.05 signified the statistical significance.

RESULTS VPS9D1-AS1 was upregulated in EC tissues and cells

The expression level of VPS9D1-AS1 in uterine corpus endometrial carcinoma (UCEC) was markedly upregulated based on the TCGA database (Figure 1A). Relative to paracancerous tissues, VPS9D1-AS1 level was prominently enhanced in EC tissues (Figure 1B). Moreover, with the cut-off value calculated, the 92 EC patients were divided into a high-VPS9D1-AS1 expression group (n = 46) and a low-VPS9D1-AS1 expression group (n = 46). High expression level of VPS9D1-AS1 predicted a poor prognosis (Figure 1C). Additionally, the VPS9D1-AS1 level was strongly correlated with lymph node metastasis, pathological stage, and International Federation of Gynecology and Obstetrics (FIGO) stage, but not associated with the age, as well as ER and PR expressions (Table 1). As expected, the VPS9D1-AS1 expression was also significantly upregulated in EC cells, including AN3CA, HEC-1-A, Ishikawa, and HEC-1-B cells (Figure 1D). Notably, AN3CA and Ishikawa cells exhibited higher expression of VPS9D1-AS1 than HEC-1-A and HEC-1-B cells, thus the two types of cells were employed for following experiments.

FIGURE 1. VPS9D1-AS1 is highly expressed in endometrial cancer (EC) tissues and cells. (A) The expression level of VPS9D1-AS1 in uterine corpus endometrial carcinoma (UCEC) was analyzed based on the TCGA database. (B) The expression level of the VPS9D1-AS1 in EC tissues and paracancerous tissues was measured by qRT-PCR, indicating that VPS9D1-AS1 was highly expressed in EC tissues. (C) Association between VPS9D1-AS1expression and overall survival of EC patients was analyzed by Kaplan–Meier curve test. (D) The qRT-PCR assay suggested that the transcription of VPS9D1-AS1 was upregulated in various EC cell lines (AN3CA, HEC-1-A, HEC-1-B and Ishikawa) compared to endometrial cells. *p [less than] 0.05; **p [less than] 0.01; ***p [less than] 0.001. All assays were performed in triplicate.

TABLE 1 Correlation between VPS9D1-AS1 expression and clinicopathological parameters of endometrial cancer patients

[Table omitted. See PDF]

^*p < 0.05 vs High VPS9D1-AS1 expression.

Knockdown of VPS9D1-AS1 inhibited proliferation, invasion, and EMT of EC cells

Compared to sh-NC group, all three shRNAs targeting VPS9D1-AS1 markedly reduced VPS9D1-AS1 levels in AN3CA and Ishikawa cells. Among them, the suppressive effect of sh-VPS9D1-AS1-2 was better than that of sh-VPS9D1-AS1-1 and sh-VPS9D1-AS1-3 (Figure 2A), thus sh-VPS9D1-AS1-2 was used for subsequent studies. Transfection of sh-VPS9D1-AS1-2 into AN3CA and Ishikawa cells markedly declined the proliferation relative to sh-NC group (Figure 2B). In addition, Edu staining assay proved that knockdown of VPS9D1-AS1 reduced EDU-positive cells in both AN3CA and Ishikawa cells, which also suggested that interference of VPS9D1-AS1 inhibited EC cell proliferation (Figure 2C). Additionally, knockdown of VPS9D1-AS1 suppressed the invasion of AN3CA and Ishikawa cells (Figure 2D). Moreover, downregulation of VPS9D1-AS1 increased E-cadherin with decreased level of vimentin and N-cadherin in both AN3CA and Ishikawa cells, indicating that knockdown of VPS9D1-AS1 alleviated EMT in EC cells (Figure 2E). There was no statistical difference in all the above-mentioned indicators between sh-NC group and control group (Figure 2B–E). Thus, downregulation of VPS9D1-AS1 inhibited invasion, growth, and EMT of EC cells.

FIGURE 2. Knockdown of VPS9D1-AS1 inhibits proliferation, invasion, and EMT of endometrial cancer (EC) cells. (A) Knockdown efficiency of VPS9D1-AS1 by corresponding shRNAs was detected by qRT-PCR in AN3CA and Ishikawa cells. (B) The CCK-8 assay was used to evaluate cell proliferation after silencing of VPS9D1-AS1 in AN3CA and Ishikawa cells. (C) EDU assay was used for the detection of the cell proliferation of AN3CA and Ishikawa cells after transfection. (D) Effects of VPS9D1-AS1 knockdown on the invasion of AN3CA and Ishikawa cells. A transwell assay was used to measure the cell invasion ability. (E) The expressions of EMT markers (E-cadherin, N-cadherin and vimentin) were evaluated by western blot. The data were expressed after being normalized to GAPDH. *p [less than] 0.05; **p [less than] 0.01. All assays were performed in triplicate.

VPS9D1-AS1 was identified to sponge miR-377-3p

From Figure 3A, VPS9D1-AS1 was primarily distributed in the cytoplasm. In addition, VPS9D1-AS1 was predicted to harbor an miR-377-3p binding site based on the bioinformatics databases (starbase.sysu.edu.cn/starbase2/index.php) (Figure 3B). The luciferase activity was dramatically attenuated by co-transfection with VPS9D1-AS1-Wt and miR-377-3p rather than co-transfection with VPS9D1-AS1-Mut and miR-377-3p into both AN3CA and Ishikawa cells (Figure 3C), suggesting that VPS9D1-AS1 was a direct target of miR-377-3p. Additionally, compared to the Bio-probe NC or Bio-miR-377-3p-Mut, Bio-miR-377-3p-Wt showed significantly higher enrichment of VPS9D1-AS1 in both AN3CA and Ishikawa cells (Figure 3D). Moreover, compared IgG, Ago2 could enrich more VPS9D1-AS1 and miR-377-3p in both AN3CA and Ishikawa cells (Figure 3E). Furthermore, miR-377-3p was lowly expressed in EC tissues relative to that in paracancerous tissues (Figure 3F). By calculated the cut-off value, the 92 EC patients were divided into a high-miR-377-3p group (n = 46) and a low-miR-377-3p group (n = 46), in which low expression of miR-377-3p predicted a poor prognosis (Figure 3G). More importantly, the expression level of miR-377-3p was consistently notably related in lymph node metastasis, pathological stage, and FIGO stage, but was not correlated with age, ER, and PR expressions (Table 2). Additionally, miR-377-3p expression was negatively relevant in the expression level of VPS9D1-AS1 in EC tissues (Figure 3H). EC cells also exhibited lower expression of miR-377-3p expression than endometrial cells (Figure 3I). Interestingly, transfection with sh-VPS9D1-AS1 elevated the expression level of miR-377-3p in both AN3CA and Ishikawa cells (Figure 3J). Therefore, these findings indicated that VPS9D1-AS1 could sponge miR-377-3p.

FIGURE 3. VPS9D1-AS1 is identified AS sponge of miR-377-3p. (A) The in situ expression of VPS9D1-AS1 in Ishikawa cells. The green fluorescent signal was from the FITC-VPS9D1-AS1 probe. The blue fluorescent signal was from nuclear DNA counterstained with DAPI (bar = 50 μm). (B) Binding sites were predicted between VPS9D1-AS1 and miR-377-3p, and sequences of miR-377-3p and wild-type VPS9D1-AS1 (VPS9D1-AS1 WT) and its mutant (VPS9D1-AS1 MUT) were exhibited. (C) A dual luciferase activity assay was performed to evaluate the binding ability of VPS9D1-AS1-Wt and VPS9D1-AS1-Mut to miR-377-3p. VPS9D1-AS1-Wt could target miR-377-3p, whereas VPS9D1-AS1-Mut lost this ability in both AN3CA and Ishikawa cells. (D) RNA pull-down assay was performed to use Bio-probe NC, Bio-miR-377-Wt, and Bio-miR-377-Mut to pull-down miR-377-3p. qPCR was used to determine the expression of VPS9D1-AS1 in each pull-down group. (E) RIP assay indicated that Ago2 protein enriched more VPS9D1-AS1 than IgG. (F) The miR-377-3p level in EC tissues and paracancerous tissues was measured by qRT-PCR, indicating that miR-377-3p was downregulated in EC tissues. (G) Low miR-377-3p expression predicted a poor prognosis. (H) Correlation analysis of the relationship between VPS9D1-AS1 and miR-377-3p expression in EC tissues. (I) The miR-377-3p expression was diminished in EC cells. (J) The miR-377-3p expression was increased in both AN3CA and Ishikawa cells transfected with sh-VPS9D1-AS1. *p [less than] 0.05; **p [less than] 0.01; ***p [less than] 0.001. All assays were performed in triplicate.

TABLE 2 Correlation between miR-377-3p expression and clinicopathological parameters of endometrial cancer patients

[Table omitted. See PDF]

^*p < 0.05 vs High miR-377-3p expression.

Overexpression of miR-377-3p reversed VPS9D1-AS1-induced EC cells proliferation, invasion, and EMT

As demonstrated in Figure 4A, miR-377-3p level was suppressed by oe-VPS9D1-AS1 in Ishikawa cells, which could be rescued with transfection of miR-377-3p mimics. Moreover, overexpression of VPS9D1-AS1 promoted EC cells proliferation, whereas miR-377-3p mimic reversed this change (Figure 4B). In addition, Edu staining assay suggested that oe-VPS9D1-AS1 increased Edu-positive cells and invasion of Ishikawa cells, which was counteracted by miR-377-3p mimic (Figure 4C,D). Additionally, oe-VPS9D1-AS1 descended E-cadherin and ascended vimentin and N-cadherin expression, indicating that VPS9D1-AS1 induced EMT. However, this alteration in the relative protein expression of E-cadherin, vimentin, and N-cadherin was reversed with miR-377-3p mimic (Figure 4E). In summary, these data implied that upregulation of miR-377-3p reversed oe-VPS9D1-AS1-induced the attenuation of EC cells growth, invasion, and EMT.

FIGURE 4. Overexpression of miR-377-3p reverses VPS9D1-AS1-induced endometrial cancer (EC) cells proliferation, invasion and EMT. (A) The oe-NC or oe-VPS9D1-AS1 and mimic NC or miR-377-3p mimic were co-transfected in Ishikawa cells. The miR-377-3p level in each group was determined by qRT-PCR. (B) The Ishikawa cell proliferation in each group was detected by CCK-8 assay. (C) The Edu staining assay indicated that oe-VPS9D1-AS1 promoted Ishikawa cell proliferation, which was reversed by miR-377-3p mimic. (D) The oe-VPS9D1-AS1 enhanced invasion of Ishikawa cells, while was reversed by miR-377-3p mimic. (E) The oe-VPS9D1-AS1 promoted the EMT process in Ishikawa cells, whereas these changes were rescued by miR-377-3p mimic. The data were expressed after being normalized to GAPDH. *p [less than] 0.05 versus oe-NC + mimic NC. #p [less than] 0.05 versus oe-VPS9D1-AS1 + mimic NC. All assays were performed in triplicate.

SGK1 directly targeted miR-377-3p

The binding sites were found between miR-377-3p and SGK1 according to the complementary base pairing (Figure 5A). Ishikawa cells co-transfected with miR-37-3p mimic and SGK1-Wt notably decreased the luciferase activity compared to co-transfection of miR-37-3p mimic and SGK1-Mut (Figure 5B), indicating the direct interaction of miR-37-3p and 3′-UTR of SGK1 mRNA. Also, miR-377-3p mimic significantly reduced SGK1 protein level in Ishikawa cells (Figure 5C). Thus, oe-VPS9D1-AS1 expectedly increased the protein expression level of SGK1, while this enhancement was inverted by miR-377-3p mimic (Figure 5D). Additionally, SGK1 expression was highly expressed in EC tissues relative to paracanerous (Figure 5E), which predicted a poor prognosis (Figure 5F). Consistently, the expression of SGK1 was significantly relevant in lymph node metastasis, pathological stage, and FIGO stage, but was not correlated with age, as well as ER and PR expressions (Table 3). As expected, SGK1 expression was positively associated with VPS9D1-AS1 expression and negatively involved in miR-377-3p expression in EC patients (Figure 6A and B). The expression of SGK1 was also upregulated in EC cells relative to that in endometrial cells (Figure 6C). To further explore the modulatory role of VPS9D1-AS1 in SGK1 expression, the SGK1 was overexpressed in Ishikawa cells (Figure 6D,E). Overexpression of SGK1 rescued the sh-VPS9D1-AS1-induced the relative protein of SGK1 (Figure 6F). Interestingly, interference of VPS9D1-AS1 decreased the growth of EC cells, whereas overexpression of SGK1 reversed this change (Figure 6G). Also, upregulation of SGK1 enhanced the sh-VPS9D1-AS1-induced the invasion of Ishikawa cells (Figure 6H). Additionally, sh-VPS9D1-AS1 reduced the expression level of vimentin and N-cadherin and increased the E-cadherin expression, which were antagonized with overexpression of SGK1 (Figure 6I), indicating that upregulation of SGK1 enhanced sh-VPS9D1-AS1-induced EMT process in Ishikawa cells. These results totally demonstrated that SGK1 was a direct target of miR-377-3p and that high-expression of SGK1 antagonized sh-VPS9D1-AS1-imduced the suppression of EC cells growth, invasion, and EMT.

FIGURE 5. SGK1 is a direct target of miR-377-3p. (A) Predicted binding site between miR-377-3p and 3′-UTR of SGK1. (B) The miR-377-3p interacted with 3′-UTR of SGK1 mRNA by directly targeting verified by luciferase reporter assay. *p [less than] 0.05 versus mimic-NC. (C) The SGK1 expression detected by qRT-PCR was suppressed by miR-377 mimic in Ishikawa cells. *p [less than] 0.05 versus mimic-NC. (D) After co-transfected with oe-VPS9D1-AS1 and miR-377-3p mimic, the SGK1 protein expression was determined by western blot. The data were expressed after being normalized to GAPDH. *p [less than] 0.05 versus oe-NC + mimic NC. #p [less than] 0.05 versus oe-VPS9D1-AS1 + mimic NC. (E) The expressions of SGK1 in EC tissues and paracancerous tissues were measured by qRT-PCR, indicating that SGK1 was upregulated in EC tissues. *p [less than] 0.05. (F) High SGK1 expression predicted a poor prognosis. All assays were performed in triplicate.

TABLE 3 Correlation between SGK1 expression and clinicopathological parameters of endometrial cancer patients

[Table omitted. See PDF]

^*p < 0.05 vs High SGK1 expression.

FIGURE 6. VPS9D1-AS1 regulates the progress of endometrial cancer (EC) cells via miR-377-3p/SGK1 axis. (A and B) The correlation analysis of the relationship between VPS9D1-AS1 and SGK1 expressions (A) and miR-377-3p (B) in EC tissues. (C) The miR-377-3p expression was elevated in EC cells based on qRT-PCR. *p [less than] 0.05 versus Endometrial cells. (D and E) The efficiency of SGK1 overexpressed in Ishikawa cells. *p [less than] 0.05 versus Control. (F) Western blot confirmed sh-VPS9D1-AS1 decreased SGK1 expression. *p [less than] 0.05 versus sh-NC + pc-NC. #p [less than] 0.05 versus sh-OIP5-AS1-2 + pc-NC. (G) The cell proliferation was detected by CCK-8 assay. *p [less than] 0.05 versus sh-NC + pc-NC. #p [less than] 0.05 versus sh-OIP5-AS1-2 + pc-NC. (H) Transwell assay was used to measure the effect of overexpression of SGK1 on cell invasion in Ishikawa cells transfected with sh-VPS9D1-AS1. *p [less than] 0.05 versus sh-NC + pc-NC. #p [less than] 0.05 versus sh-OIP5-AS1-2 + pc-NC. (I) The EMT markers including E-cadherin, N-cadherin and vimentin were evaluated by western bolt. The data were expressed after being normalized to GAPDH. *p [less than] 0.05 versus sh-NC + pc-NC. #p [less than] 0.05 versus sh-OIP5-AS1-2 + pc-NC. All assays were performed in triplicate.

Silence of VPS9D1-AS1 inhibited EC tumor growth in vivo

Although no statistical differences were discovered in body weight (Figure 7A), knockdown of VPS9D1-AS1 observably decreased tumor volume and weight (Figure 7B,C). Interference of miR-377-3p markedly neutralized sh-VPS9D1-AS1-induced the reduction of tumor volume and weight (Figure 7B,C). In line with the in vitro results, application of sh-VPS9D1-AS1 prominently diminished the expression of VPS9D1-AS1 and SGK1 with enhanced the miR-377-3p level, which were reversed by the silence of miR-377-3p (Figure 7D). Consistently, western blot confirmed the alterations in the relative levels of SGK1 protein in mice (Figure 7E). Furthermore, IHC analysis of xenografted tumors displayed that sh-VPS9D1-AS1 notably decreased the level of Ki-67, vimentin, and N-cadherin, which was antagonized by the miR-377-3p inhibitor (Figure 7F). Also, TUNEL staining revealed that knockdown of miR-377-3p observably reduced the sh-VPS9D1-AS1-induced apoptosis rate (Figure 7F,G). In conclusion, these results demonstrated that knockdown of VPS9D1-AS1 suppressed EC tumor growth through regulating miR-377-3p/SGK1 axis in situ.

FIGURE 7. Silence of VPS9D1-AS1 inhibits endometrial cancer tumor growth in vivo. The female BALB/c nude mice were injected with Ishikawa cells infected with sh-NC, sh-VPS9D1-AS1, miR-377-3p inhibitor, or inhibitor-NC, and monitored for continuous 6 weeks. (A) The body weight was measured every week for 6 weeks. (B) The tumor volume was measured every week for 6 weeks. (C) The tumors were removed, captured, and then weighed. (D) The mRNA levels of VPS9D1-AS1, miR-377-3p and SGK1 in tumor tissues were detected by qRT-PCR. (E) The relative protein expression of SGK1 was determined by western blot. The data were expressed after being normalized to GAPDH. (F) IHC staining of Ki-67, N-cadherin and vimentin in tumor tissues, as well as the determination of apoptosis by TUNEL. (G) The apoptosis rate was exhibited by the histogram. *p [less than] 0.05 versus sh-NC + inhibitor-NC. #p [less than] 0.05 versus sh-VPS9D1-AS1 + inhibitor-NC. All assays were performed in sex triplicate.

DISCUSSION

In the current study, we confirmed that VPS9D1-AS1 promoted EC cells proliferation, invasion, and EMT by regulating miR-377-3p/SGK1 axis, which provided a new target for EC therapy. VPS9D1-AS1 expression is strongly increased in diverse cancers, such as colon adenocarcinoma,¹⁰ esophageal squamous cell carcinoma,¹¹ hepatocellular carcinoma,¹² and acute myeloid leukemia,¹³ which indicates that VPS9D1-AS1 can act as an oncogene. Here, the expression level of VPS9D1-AS1 was consistently upregulated in EC tissues and cells, which also suggested that VPS9D1-AS1 could be a biomarker of EC. Moreover, the upregulation of VPS9D1-AS1 was involved in a poor prognosis in EC patients. Similar outcomes were also observed in colorectal cancer,¹⁴ non-small cell lung cancer (NSCL),¹⁵ and esophageal squamous cell carcinoma.¹¹ Furthermore, clinical trials exhibited that the expression level of VPS9D1-AS1 was obviously related in pathological stage, lymph node metastasis, and FIGO stage. Hence, these results clarified upregulation of VPS9D1-AS1 was tightly related in poor prognosis in EC.

High expression of VPS9D1-AS1 has been shown to promote the detrimental process of several tumors. Han et al.¹⁶ reported that upregulation of VPS9D1-AS1 accelerates the growth, migration, and invasion but represses apoptosis of NSCL. Another study also confirms the same outcomes in colorectal cancer.¹⁷ Additionally, overexpression of VPS9D1-AS1 enhances growth, invasion, and migration of esophageal squamous cell carcinoma.¹¹ In accordance with these findings, our results also manifested high-expression of VPS9D1-AS1 facilitated the proliferation, migration, and EMT of EC, which were notably ameliorated with application of shRNA targeting VPS9D1-AS1. EMT, excessive proliferation, and invasion are believed to be significantly involved in the malignant process of tumors.¹⁸ Thus, our data indicated that VPS9D1-AS1 might be a therapeutic target of EC.

LncRNAs as a sponge target to downstream miRNAs to modulate the progress of tumors.¹⁹ VPS9D1-AS1 was predicted to sponge miR-377-3p, which was further verified by luciferase reporter as well as RIP and RNA pull-down assays in the present study. MiR-377-3p has also been confirmed to be an oncogene that involves in the prognosis of tumors. Downregulation of miR-377-3p indicates poor prognosis in cervical carcinoma,²⁰ and osteosarcoma.²¹ The analogical findings were also discovered in EC in our study, and the clinical trials displayed that the expression level of miR-377-3p was obviously relevant in pathological stage, lymph node metastasis, and FIGO stage. Furthermore, miR-377-3p expression was negatively related in VPS9D1-AS1 expression in EC tissues. VPS9D1-AS1 sponging different miRNAs to regulate the tumor progression has been shown in many cancers. For instance, VPS9D1-AS1 modulates growth, migration, invasion, and apoptosis through targeting miR-532-3p in NSCL.¹⁶ Upregulation of VPS9D1-AS1 enhances the cell viability, invasion, and migration and declines the apoptosis by binding miR-525-5p in colorectal cancer.¹⁷ VPS9D1-AS1 sponging miR-184 modulates the growth and migration of prostate cancer cells.²² Consistently, the results in the study confirmed that overexpression of miR-377-3p reversed VPS9D1-AS1-induced EC cells proliferation, invasion, and EMT. Therefore, all the findings suggested that VPS9D1-AS1 modulated the growth, invasion, and EMT through sponging miR-377-3p in EC cells.

In addition, miRNAs can regulate the targets' genes expression via binding to 3′-UTR of targets.²³ Bioinformatics prediction showed that SGK1 was a potential target of miR-377-3p, which was confirmed by luciferase reporter assay. Overexpression of miR-377-3p reduced both the mRNA and protein levels of SGK1, which verified a negative interaction between miR-377-3p and SGK1. SGK1, as a serine/threonine-kinase, contains a highly conserved kinase structure.²⁴ A recent review has summarized that the expression, clinical significance, and prognosis of SGK1 in the extensiveness tumors.²⁵ Our data also showed that SGK1 expression was highly expressed in EC, which predicted the poor prognosis. Clinical value indicated that the expression level of SGK1 was similarly involved in pathological stage, lymph node metastasis, and FIGO stage. Multiple functional roles of SKG1 involve in tumors, including tumorigenesis, proliferation and apoptosis, invasion and migration, metabolism, chemo- and radio-resistance, and tumor microenvironment.²⁵ Knockdown of SGK1 significantly reduces myeloma cells proliferation.²⁶ In addition, SGK1 inhibition blocks tumor progression in a preclinical hepatocellular carcinoma.²⁷ In the current study, overexpression of SGK1 counteracted sh-VPS9D1-AS1-induced the inhibition of EC cells proliferation, invasion, and EMT. Combined with the data that overexpression of VPS9D1-AS1 increased the relative protein level of SGK1, we concluded that VPS9D1-AS1 sponged miR-377-3p and decreased miR-377-3p expression, and then increased SGK1 level in EC, thereby eventually regulating the progress of EC.

The repressive effect of the VPS9D1-AS1 silence has been widely demonstrated in a transplanted mouse model of various cancers, such as esophageal squamous cell carcinoma,¹¹ hepatocellular carcinoma,¹² acute myeloid leukemia,²⁸ colorectal cancer,¹⁷ and NSCL.¹⁶ Likewise, knockdown of VPS9D1-AS1 observably decreased tumor volume and weight, which were markedly neutralized by the interference of miR-377-3p. In addition, application of sh-VPS9D1-AS1 prominently enhanced the miR-377-3p expression with the diminished expression SGK1, which was reversed by the silence of miR-377-3p. Thus, the VPS9D1-AS1/miR-377-3p/SGK1 signaling axis was also confirmed in vivo. Furthermore, sh-VPS9D1-AS1 notably suppressed the Ki-67 level, an effective indicator of the proliferation of tumors,²⁹ and the expression of vimentin and N-cadherin, which was antagonized by the treatment of miR-377-3p inhibitor. Totally, all these data elaborated that downregulation of VPS9D1-AS1 inhibited EC tumor growth via modulating miR-377-3p/SGK1 axis in vivo.

In summary, the current study demonstrated VPS9D1-AS1 promoted cell invasion, proliferation, and EMT in EC via regulating miR-377-3p/SGK1 axis, which provided new options for therapeutic strategies for EC. However, some drawbacks of the present study need to be addressed. First, more plenitude cohort of clinical data should be included in our further study. Secondly, the regulative role of VPS9D1-AS1 via miR-377-3p/SGK1 axis need more detailed researches in our subsequent assays. Additionally, Lin et al²⁸ showed that VPS9D1-AS1 knockdown inhibits the MEK/ERK signaling pathway in acute myeloid leukemia. Ma et al¹¹ revealed that VPS9D1-AS1 promotes esophageal squamous cell carcinoma progression through the Wnt/β-catenin signaling pathway. Thus, the roles of VPS9D1-AS1 may be involved in the signaling pathway, such as MEK/ERK and Wnt/β-catenin signaling pathway, which can be further discussed in the following study. In brief, our present results can lay a theoretical foundation for the diagnosis and treatment of EC.

ACKNOWLEDGMENTS

We thank Dr. Xuan Zhou in our department for providing helpful suggestions during experiment.

CONFLICT OF INTEREST

All authors declare no conflict of interest.