LncRNA SNHG8 induces ovarian carcinoma cells cellular process and stemness through Wnt/ β -catenin pathway

Abstract

Ovarian carcinoma ranks fifth in the leading causes of cancer-relevant deaths among the female, with the highest fatality rate in all gynecological malignant tumors and the rising incidence worldwide. Mounting evidence has unveiled that lncRNAs are implicated in the tumorigenesis and cancer development. Several studies have proven the carcinogenic role of SNHG8 in various malignancies, but the physiological functions of SNHG8 in ovarian carcinoma need more detailed explanations. The present study certified that inhibition of SNHG8 executed suppressive activities in ovarian carcinoma by obstructing cell proliferation, migration, EMT process and stemness as well as driving cell apoptosis. Moreover, SNHG8 bound with CAPRIN1 and positively modulated the expression of CAPRIN1. Further experiments manifested that CTNNB1 and Axin1 displayed a binding affinity with CAPRIN1. Knockdown of CAPRIN1 promoted the mRNA degradation of CTNNB1 and Axin1. Finally, we corroborated that CTNNB1 (or Axin1) ectopic expression or activation of Wnt/ $\beta$ -catenin pathway abrogated the effects of SNHG8 downregulation on the cellular process of ovarian carcinoma cells. To summarize, SNHG8 acted as an oncogene in ovarian carcinoma via targeting Wnt/ $\beta$ -catenin pathway, providing a new insight into understanding ovarian carcinoma at the molecular level.

1. Introduction

Ovarian carcinoma ranks fifth in the leading causes of cancer-relevant deaths among the female, with the highest fatality rate in all gynecological malignant tumors and the rising incidence worldwide [1, 2]. For ovarian carcinoma patients, late diagnosis generally occurs due to insufficient specific symptoms and screening methods [3, 4]. Once diagnosed, patients with ovarian carcinoma often miss out on the surgery [5]. Despite the continual improvement in ovarian carcinoma therapies, the survival remains limited in advanced ovarian carcinoma patients, and the survival rate within five years of diagnosis of sufferers with distant metastasis is only 28.9% [6]. In consequence, it is imperative to conduct more research to understand the latent mechanisms of ovarian carcinoma.

Long noncoding RNAs (lncRNAs) are defined as a subset of noncoding RNAs which are longer than 200 nucleotides and featured with limited or no protein-coding capacity [7]. Accumulating paper has proved that lncRNAs are involved in the tumorigenesis and cancer development by regulating biological activities, such as cell proliferation, invasion, apoptosis and differentiation [8, 9]. For example, lncRNA NEAT1 facilitates ovarian carcinoma cell metastasis through targeting miR-382-3p/ROCK1 axial [10]. Novel lncRNA OTUD6B-AS1 correlates with unfavorable prognosis and retards the proliferation of clear cell renal cell carcinoma [11]. LncRNA SNHG20 facilitates cell proliferation and apoptosis of glioma by regulation of PTEN/PI3K/AKT signaling pathway [12]. According to recent investigations, the carcinogenesis of SNHG8 has been certified in multiple malignancies, including non-small-cell lung cancer [13], gastric carcinoma [14], pancreatic adenocarcinoma [15]. Nevertheless, the physiological functions of SNHG8 in ovarian carcinoma need more detailed explanations.

Wnt signaling plays a pivotal role during organ development and tissue homeostasis in several diseases, including tumors [16, 17]. It is well accepted that Wnt signaling is diversified into two pathways, namely the canonical and noncanonical pathway [18]. Besides, LEF-1 and $\beta$ -catenin serve as key regulators in the canonical Wnt pathway through mediating the transcription of target genes vital for the development of cancer [19, 20]. An extensive body of studies confirmed that Wnt/ $\beta$ -catenin regulates the processes of cell proliferation, metastasis and resistance to drugs in human cancer [21, 22]. Worth mentioning, the association between the ovarian carcinoma and Wnt/ $\beta$ -catenin pathway has been explored [23]. However, whether Wnt/ $\beta$ -catenin pathway mediates the influences of SNHG8 on ovarian carcinoma progression remains elusive.

Therefore, the purpose of current work was to identify the functions of SNHG8 on the cellular process of ovarian carcinoma cells and the mechanism through which SNHG8 might regulate Wnt/ $\beta$ -catenin pathway.

2. Materials and methods

2.1 Cell culture and transfection

Human normal fallopian tube epithelial cell line FTE187 and ovarian carcinoma cell lines (SKOV3, ES2, CaOV3 and HG-SOC) were offered by American Type Culture Collection (ATCC, USA). All cells were cultivated in DMEM (Gibco, Carlsbad, CA, USA) or RPMI-1640 medium (Invitrogen, Thermo Fisher Scientific, USA) complemented with 10% fetal bovine serum (Invitrogen, USA), 100 $\mu$ g/mL streptomycin and 100 unit/mL penicillin at 37 ${}^{\circ }$ C in a humid atmosphere containing 5% CO ${}_2$ . Short hairpin RNAs (shRNAs) against SNHG8 or CAPRIN1 and nontargeting control shRNAs (shNCs) were commercially acquired from Origene (Shanghai, China). The CTNNB1 (or Axin1) sequences were inserted into the pcDNA3.1 vector (Genepharma, China) for overexpression of CTNNB1 (or Axin1) and the empty pcDNA3.1 plasmid was adopted as a negative control. Cell transfection was performed with the utilization of Lipofectamine 2000 (Invitrogen) obeying the instruction of the manufacturer.

2.2 Cell counting Kit-8 (CCK-8) assay

Cell viability was estimated by Cell Counting Kit-8 (CCK-8; Beyotime Institute of Biotechnology, Shanghai, China). Transfected cells were cultured on a 96-well plate at a density of 2 $\times$ 10 ${}^3$ /well. 0, 24, 48, 72, or 96 h later, 10 $\mu$ L CCK-8 reaction solution was supplemented for another 4 h of incubation at 37 ${}^{\circ }$ C. The absorbance was observed at 450 nm with a microplate reader (Thermo Fisher Scientific, Inc., Waltham, MA, USA).

2.3 5-ethynyl-2’-deoxyuridine (EdU) assay

An EdU incorporation assay kit (Ribobio, China) was employed for evaluation of cell proliferation. In brief, 2 $\times$ 10 ${}^4$ cells were treated with EdU for 2 h. Afterwards, 2 mg/ml glycine was added and then cells were permeabilized with 0.5% Triton X-100 for 20 min, rinsed by PBS, stained with Apollo staining reaction buffer for 30 min. DAPI stained the nuclei of cells and the EdU incorporation rate was calculated under a fluorescence microscope.

2.4 Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA extraction from cells was carried out with TRIzol (Invitrogen) under the recommendations of the manufacturer. The PrimeScript RT reagent kit (Promega, USA) was applied to synthesize cDNA. Then, PCR was conducted with a SYBR Green PCR kit (Takara, Japan) in an ABI 7500 real-time PCR system (Applied Biosystems, USA). The expression of genes was quantified by the 2 ${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{CT}}$ method. Analysis of each sample was conducted in triplicate and $\beta$ -actin was utilized as the internal control. Primers were designed and shown as follows: SNHG8 (forward: 5’-CCCGAGAA CCGTCAGTTTGA-3’, reverse: 5’-ACACCCGTTTCC CCAACTAC-3’); CTNNB1 (forward: 5’-CATCTACAC AGTTTGATGCTGCT-3’, reverse: 3’-GCAGTTTTGT CAGTTCAGGGA-5’); CAPRIN1 (forward: 5’-TCTCG GGGTGATCGACAAGAA-3’, reverse: 3’-CCCTTTGT TCATTCGTTCCTGG-5’); $\beta$ -actin (forward: 5’-AGCG AGCATCCCCCAAAGTT-3’, reverse: 3’-GGGCACG AAGGCTCATCATT-5’).

2.5 Wound-healing assay

Forty-eight h after transfection, cells were seeded into 6-well plate and grown to approximately 85% confluence. Then, the sterile pipette was applied to scratch the bottom of the culture plate. After washing by PBS buffer, cells were added with serum-free DMEM and maintained at 37 ${}^{\circ }$ C for 48 h. The images of cell scratch were recorded via a microscope at 0 h and 48 h after scratch.

2.6 Cell migration

Cell migration was detected by transwell assay employing transwell chambers (8 $\mu$ m pore size, Corning, USA). After transfection, cells with serum-free medium were seeded in the upper chamber and the medium containing 10% FBS was added to the lower chamber. Forty-eight h of incubation later, 4% paraformaldehyde was applied to fix cells on the lower surface and then cells were stained by 0.1% crystal violet. The number of migrated cells was quantified under a microscope.

2.7 Sphere formation assay

As previously described [24], cell self-renewal capacity was evaluated by this assay. The single cell suspension was prepared by enzymatic disassociation, filtrated using a 40 $\mu$ m cell strainer, plated into 96-well ultralow-attachment plates (Corning, Tewksbury, MA, USA) and then grown in serum-free DMEM/F12 media supplemented with 20 ng/ml hEGF (Life Technologies), bFGF (10 ng/ml, Sigma), heparin (4 $\mu$ g/ml, Sigma), 1% methylcellulose (R&D), and B27 Supplement (Thermo-Fisher Scientific). After 10–14 days of incubation, spheres were fixed and stained by crystal violet, and the number of spheres was counted for statistical analysis.

2.8 Flow cytometry analysis

Stem cell marker CD133 was measured in transfected cells by flow cytometry analysis as previously described [25].

2.9 Caspase-3 activity assay

The caspase-3 activity kit (Beyotime Institute of Biotechnology) was adopted for the analysis of caspase activity. Total protein from cells was treated with caspase-3 substrate Ac-DEVD-pNA at 37 ${}^{\circ }$ C for 2 h and caspase-3 activity was detected by a VERSAmax microplate reader (Molecular Devices, LLC) at wavelength of 405 nm.

2.10 Western blot

RIPA lysis buffer (Beyotime, China) supplemented with protease inhibitors was utilized to extract proteins from cells, and the BCA protein assay (Pierce, USA) was performed to determine protein concentrations. Equal amounts (40 $\mu$ g) of cell protein lysates were isolated by 10% SDS-PAGE and subsequently transferred to the PVDF membranes (Millipore, USA). The membranes were sealed in 5% skim milk for 1 h and probed with primary antibodies of E-cadherin (ab76055, Abcam), N-cadherin (#13116, Cell Signaling Technology), SNAIL (#3879, Cell Signaling Technology), ALDH1 (ab23375, Abcam), Nanog (Ab62734, Abcam), Nanog (ab109250, Abcam), OCT4 (ab18976, Abcam), SOX2 (ab93689, Abcam), CD44 (ab157107, Abcam), CD133 (ab19898, Abcam), CAPRIN1 (HPA018126, Atlas Antibodies), $\beta$ -catenin (ab22656, Abcam), CyclinD1(ab190564, Abcam), c-myc (ab17356, Abcam), CTNNB1 (ab32572, Abcam), Axin1 (ab115205, Abcam), Axin2 (ab32197, Abcam) and $\beta$ -actin (#3700, Cell Signaling Technology) at 4 ${}^{\circ }$ C overnight. Subsequently, membranes underwent 2 h of incubation with an HRP-conjugated secondary antibody at room temperature. The blots were visualized by with an EasyBlot ECL detection system (Sangon, China) and $\beta$ -actin served as a loading control.

2.11 RNA precipitation

To figure out the possible proteins that bind with SNHG8, biotin-labeled SNHG8 RNA probe was employed in RNA precipitation assay. Then PierceTM Silver Stain Kit (Fisher Scientific) was utilized to conduct silver staining or western blot as previously described [26].

2.12 RNA immunoprecipitation (RIP)

The Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Bedford, MA, USA) was applied to conduct RIP assay in light of the suggestion of the manufacturer. Cell lysates were incubated in RIP buffer supplemented with magnetic beads coated with anti-CAPRIN1 (Abcam) antibody or negative control IgG (Millipore). The immunoprecipitated RNA was obtained with proteinase K and purified RNA was analyzed by qRT-PCR.

Figure 1.

Silencing of SNHG8 impedes cell proliferation and contributes to cell apoptosis. (A) SNHG8 expression was analyzed by qRT-PCR in ovarian carcinoma cells (SKOV3, ES2, CaOV3 and HG-SOC) and normal cells FTE187. (B) Transfection efficiency was verified by qRT-PCR. (C–D) Cell proliferation of SKOV3 and CaOV3 cells was tested byCCK-8 and EdU assays. (E) Caspase-3 activity assay was applied to analyze cell apoptosis. ${}^{\ast }P<$ 0.05, ${}^{\ast \ast }P<$ 0.01, ${}^{\ast \ast \ast }P<$ 0.001.

Figure 2.

SNHG8 suppression restrains the migration, EMT and stemness of ovarian carcinoma. (A) The migrated ability of SKOV3 and CaOV3 cells was assessed by wound healing assay. (B) Transwell assay for cell migration in transfected cell lines. (C) Western blot was applied to estimate the expression of EMT process-related proteins. (D–E) The sphere formation ability was detected by sphere formation assay and western blot. (F) Percentage of CD133 in SNHG8-downregulated cells was measured. ${}^{\ast \ast }P<$ 0.01.

2.13 RNA pull-down assay

Full-length sense and antisense of SNHG8 and CTNNB1 RNA were transcribed using MEGAscript ${}^{\text{TM}}$ T7 Transcription Kit (Thermo Fisher Scientific, Waltham, MA, USA). Pierce ${}^{\text{TM}}$ RNA 3’ End Desthiobiotinylation Kit (Thermo Fisher Scientific) was utilized for the biotinylation of transcripts. The pull-down assay was performed with Pierce ${}^{\text{TM}}$ Magnetic RNA-Protein Pull-Down Kit (Thermo Fisher Scientific) following the manufacturer’s suggestion. After transfection with biotinylated SNHG8 or CTNNB1 or Axin1, cell lysates were harvested and incubated with M-280 streptavidin magnetic beads (Invitrogen). The abundance of CAPRIN1 in complex was examined by western blot.

2.14 Analysis of mRNA stability

After transfection, cells were collected and supplemented with the transcriptional inhibitor 5 $\mu$ g/ml Actinomycin D (Sigma). CTNNB1 or Axin1 level was evaluated utilizing qRT-PCR at 0, 2, 4, 6, 8 h after Actinomycin D treatment.

2.15 Statistical analysis

All results were shown as the mean $\pm$ standard deviation (SD) of three independent experiments. SPSS 20.0 software (IBM, USA) using Student’s $t$ test or ANOVA with Tukey’s post hoc test processed the data acquired from the experiments. $P<$ 0.05 was considered to indicate statistical significance.

3. Results

3.1 Silencing of SNHG8 impedes cell proliferation and contributes to cell apoptosis

In order to estimate the physiological function of SNHG8 in ovarian carcinoma, we first detected the expression of SNHG8 in normal cells and ovarian carcinoma cells. qRT-PCR analysis demonstrated that SNHG8 was pronouncedly elevated in ovarian carcinoma cells compared to normal FTE187 cells (Fig. 1A). Thereafter, SKOV3 and CaOV3 cells were transfected with shRNAs targeting SNHG8, leading to the overt downregulation of SNHG8 expression (Fig. 1B). CCK-8 and EdU assays were adopted for evaluation of cell proliferation in response to SNHG8 inhibition and knockdown of SNHG8 resulted in a remarkable decrease of cell proliferation (Fig. 1C and D). In concert with these findings, cell apoptosis rate was notably enhanced in the face of SNHG8 depletion (Fig. 1E). In a word, our results indicate that inhibition of SNHG8 exerts inhibitive functions in ovarian carcinoma by retarding cell proliferation and inducing cell apoptosis.

Figure 3.

SNHG8 regulates the expression of CAPRIN1 by binding to it. (A) RNA-pull down assay confirmed the binding between SNHG8 and CAPRIN1. (B) The level of CAPRIN1 in cell lines was examined by qRT-PCR assay. (C) The association between SNHG8 and CAPRIN1 was confirmed by RIP assay. (D–E) The mRNA and protein expression of CAPRIN1 was measured by qRT-PCR and western blot respectively. ${}^{\ast }P<$ 0.05, ${}^{\ast \ast }P<$ 0.01, ${}^{\ast \ast \ast }P<$ 0.001.

3.2 SNHG8 suppression restrains the migration, EMT process and stemness of ovarian carcinoma

We subsequently investigated the impacts of SNHG8 on cell migration. The wound healing assay manifested that SNHG8 downregulation caused slower scratch wound closure in SKOV3 and CaOV3 cells, suggesting that depletion of SNHG8 inhibited the migration of ovarian carcinoma cells (Fig. 2A). This result was further confirmed by the transwell assay (Fig. 2B). Considering that epithelial-mesenchymal transition (EMT) process plays a vital role in cell metastasis progression of malignancy [27], we evaluated the EMT process in transfected cells. In western blot assay, an evident increase in the expression of epithelial marker E-cadherin and an obvious diminution in the expression of mesenchymal markers (N-cadherin and SNAIL) was observed when SNHG8 was silenced (Fig. 2C). Furthermore, given that cancer cells undergoing EMT process exhibit stem cell properties [28], sphere formation assay was conducted. Results presented that inhibition of SNHG8 repressed the sphere formation ability (Fig. 2D). Concordant with these data above, we discovered that the expression levels of stem factors (ALDH1, Nanog, SOX2, OCT4, CD44, CD133) were dropped due to SNHG8 knockdown (Fig. 2E). Additionally, stem cell marker CD133 was evaluated in SNHG8-downregulated cells through flow cytometry analysis. CD133 positivity was obviously decreased after depletion of SNHG8 (Fig. 2F). On the whole, we draw a conclusion that downregulation of SNHG8 impairs ovarian carcinoma cell migration, EMT process and stemness.

3.3 SNHG8 regulates the expression of CAPRIN1 by binding to it

Pull-down silver staining was carried out to explore the proteins that CAPRIN1 could interact with SNHG8 (Fig. S1A). By the assistance of starBase database, we found that CAPRIN1 may possess the potential binding sites with SNHG8. Then western blot assay elucidated that CAPRIN1 was pulled down by SNHG8 and had no response to antisense SNHG8 (Fig. 3A). As CAPRIN1 expression in ovarian carcinoma cells was upregulated (Fig. 3B), CAPRIN1 was chosen for in-depth study on the mechanism of SNHG8. Besides, RIP assay showed that SNHG8 was predominantly abundant in CAPRIN1 immunoprecipitate compared with IgG group (Fig. 3C). Given that CAPRIN1 bound with SNHG8, we explored the association between SNHG8 and CAPRIN1. Results of qRT-RCR and western blot proofed that SNHG8 suppression led to a significant attenuation of CAPRIN1 mRNA and protein levels in SKOV3 and CaOV3 cells (Fig. 3D and E). Jointly, SNHG8 binds with CAPRIN1 and positively modulates the expression of CAPRIN1.

Figure 4.

CAPRIN1 promotes the stability of CTNNB1. (A) qRT-PCR detected CTNNB1 level in ovarian carcinoma cells and normal fallopian tube epithelial cells. (B–C) RIP and RNA-pull down assays examined whether CTNNB1 bound with CAPRIN1. (D–E) The expression of CAPRIN1 and CTNNB1 was quantified by qRT-PCR. ${}^{\ast }P<$ 0.05, ${}^{\ast \ast }P<$ 0.01, ${}^{\ast \ast \ast }P<$ 0.001.

Figure 5.

SNHG8 facilitates ovarian carcinoma via activation of Wnt/ $\beta$ -catenin pathway. (A) Transfection efficiency was evaluated by qRT-PCR analysis. (B) The protein levels of key target genes in Wnt/ $\beta$ -catenin pathway was measured by western blot. (C–D) CCK-8 and EdU assays were adopted for detection of cell proliferation. (E) Cell apoptosis of ovarian carcinoma was assessed by caspase-3 activity assay. (F) Transwell assay was conducted to determine cell migration. (G) EMT process was estimated by western blot. (H) The number of spheres was examined by sphere formation assay. ${}^{\ast \ast }P<$ 0.01.

3.4 CAPRIN1 promotes the stability of CTNNB1 and Axin1

To further understand the molecular mechanism of SNHG8, the proteins that were pulled down by SNHG8 were submitted to GO analysis. The results suggested that SNHG8 was highly possible to influence Wnt signaling pathway and mRNA stability regulation (Fig. S1B). Given that CTNNB1 and Axin1 are Wnt signaling pathway-related genes and were predicted by starBase to be regulated by CAPRIN1 (Fig. S1C), therefore, CTNNB1 and Axin1 were chosen to be studied. Further, the expression of CTNNB1 and Axin1 in ovarian carcinoma cells was upregulated in comparison with that in normal FTE187 cells (Fig. 4A, Fig. S1D). RIP and RNA-pull down experiments were then carried out to verify the relationship between CAPRIN1 and CTNNB1 or Axin1. Our findings revealed that CTNNB1 and Axin1 were enriched by anti-CAPRIN1 antibody (Fig. 4B, Fig. S1E). Afterwards, CAPRIN1 and Axin1 can be only detected in the pull-down complex of CTNNB1 but not in the antisense CTNNB1 group, implying that CTNNB1 displayed a binding affinity with CAPRIN1 (Fig. 4C, Fig. S1F). Considering that CAPRIN1 is known as a RNA-binding protein which is participated in the regulation of target genes [29], we assessed the influences of CAPRIN1 on CTNNB1 and Axin1 expression. CAPRIN1 expression was knocked down and qRT-PCR confirmed the efficiency of transfection (Fig. 4D). Subsequently, ovarian carcinoma cells were treated with RNA synthesis inhibitor Actinomycin D and results illuminated that the mRNA degradation of CTNNB1 and Axin1 were quicker owing to CAPRIN1 depletion, implying that absence of CAPRIN1 restrained the CTNNB1 mRNA stability (Fig. 4E, Fig. S1G). Furthermore, CAPRIN1 was overexpressed in SKOV3 and CaOV3 cells by the transfection of pcDNA3.1/CAPRIN1 (Fig. S1H). Finally, it was validated that the suppressed mRNA stability, mRNA and protein level of CTNNB1 and Axin1 in sh-SNHG8#1-transfected cells were reversed by CAPRIN1 overexpression (Fig. S1I–J). Taken together, CAPRIN1 inhibition enhances the mRNA degradation of CTNNB1 and Axin1.

3.5 SNHG8 facilitates ovarian carcinoma via activation of Wnt/ $\beta$ -catenin pathway

Given that CTNNB1 is a protein that related with Wnt/ $\beta$ -catenin pathways, we hypothesized that SNHG8 executed its carcinogenic role in ovarian carcinoma by regulating Wnt/ $\beta$ -catenin pathway. CTNNB1 (or Axin1) was overexpressed after being transfected with pcDNA3.1/CTNNB1 (or pcDNA3.1/Axin1) vectors (Fig. 5A, Fig. S2A) and then we estimated Wnt/ $\beta$ -catenin pathway by examining $\beta$ -catenin, Cyclin D1, c-myc, Axin2 expression. Our data unfolded that suppression of SNHG8 decreased the levels of Wnt/ $\beta$ -catenin signaling target genes, whereas overexpression of CTNNB1 (or Axin1) or activation of Wnt/ $\beta$ -catenin pathway (using an agonist of Wnt/ $\beta$ -catenin LiCl) abrogated the effects of SNHG8 knockdown on the Wnt/ $\beta$ -catenin pathway effectors (Fig. 5B, Fig. S2B). The CCK-8 and EdU assays proofed that the cell proliferative ability diminished by SNHG8 deficiency was recuperated in the face of CTNNB1 (or Axin1) ectopic expression or activation of Wnt/ $\beta$ -catenin pathway (Fig. 5C and D, Fig. S2C–D). Consistently, treatment of LiCl or upregulation of CTNNB1 (or Axin1) blocked the promotion of cell apoptosis caused by SNHG8 deficiency (Fig. 5E, Fig. S2E). In addition, the impaired cell migration, EMT process and cell stemness in SNHG8 depleted cells was rescued by CTNNB1 (or Axin1) overexpression or LiCl treatment (Fig. 5F–H, Fig. S2F–H). To sum up, it was proved that SNHG8 functions as an oncogene in ovarian carcinoma through Wnt/ $\beta$ -catenin pathway.

4. Discussion

Ovarian carcinoma ranks as the eighth most prevalent malignancy and becomes most fatal gynecological cancer in women [30]. Ovarian carcinoma is disproportionately lethal in view of lacking sophisticated early diagnosis approaches [31]. Although there are advances in treating ovarian carcinoma, the 5-year survival rate of patients is still unsatisfactory [32]. Hence, our study intended to explore the potential mechanism of ovarian carcinoma so as to find out a potent target for better therapy.

It is evident from increasing investigations that lncRNAs function as core participants in regulation of biological processes in various tumors, including ovarian carcinoma [33, 34, 35]. Mounting evidence has implied the oncogenic functions of SNHG8 on several malignancies. For instance, lncRNA SNHG8 enahnces the malignancy of hepatocellular carcinoma by sponging miR-149-5p [36]. LncRNA SNHG8/miR-152/c-MET axis facilitates the development of endometrial carcinoma [37]. lncRNA SNHG8 acts as an oncogene in Epstein-Barr virus-associated gastric carcinoma [38]. Nonetheless, the impacts of SNHG8 on ovarian carcinoma remain to be probed. Our work illustrated the upregulated SNHG8 in ovarian carcinoma cells and silencing of SNHG8 elicited suppressive influences by hampering cell proliferation, migration, EMT process and stemness whereas promoting cell apoptosis.

Cell cycle associated protein 1 (CAPRIN1), a 709 amino acid protein located in the cytoplasm, is a member of the conserved protein family observed throughout vertebrates [39, 40]. And it was detected to be the most enriched protein to be pulled down by biotinylated SNHG8. Previous paper has delineated the critical role of CAPRIN1 in the development of diverse cancers. For example, enhanced caprin1 level could predict unfavorable prognosis in hepatocellular carcinoma [41]. Caprin-1 exerted its oncogenic functions on osteosarcoma tumor [42]. microRNA-223 suppresses the malignancy of breast cancer via targeting Caprin-1 [43]. Results from experiments disclosed that CAPRIN1 was upregulated in ovarian carcinoma and bound with SNHG8. Furthermore, SNHG8 positively modulated the expression of CAPRIN1. More importantly, GO analysis of the proteins that were pulled down by SNHG8 implied that SNHG8 might regulate Wnt signaling pathway and mRNA stability. Besides, CAPRIN1 has been reported to act as an RNA-binding protein [44]. Considering that CTNNB1 and Axin1 are Wnt signaling pathway-related genes and were probed by starBase to be regulated by CAPRIN1, therefore, the interaction between CAPRIN1 and CTNNB1 or Axin1 was further explored. Our findings manifested that CAPRIN1 displayed a binding affinity with CTNNB1 and Axin1 as well as promoted the mRNA stability of CTNNB1 and Axin1.

Wnt/ $\beta$ -catenin pathway has been validated to participate in regulating cancer development by affecting biological activities [45, 46]. Considering that $\beta$ -catenin is of immense importance in the canonical Wnt pathway [47], we hypothesized that SNHG8 executed its carcinogenic role in ovarian carcinoma through $\beta$ -catenin-dependent Wnt pathway. Results of rescue experiments suggested that enforced expression of CTNNB1 (or Axin1) or activation of Wnt/ $\beta$ -catenin pathway compensated for SNHG8-mediated cell proliferation, apoptosis, migration, EMT process and stemness in ovarian carcinoma.

In conclusion, this study clarified the carcinogenic role of SNHG8 in ovarian carcinoma and that SNHG8 enhances the malignancy of ovarian carcinoma via activation of Wnt/ $\beta$ -catenin pathway, shedding a light into understanding ovarian carcinoma.