Long non-coding RNA DSCR8 acts as a molecular sponge for miR-485-5p to activate Wnt/β-catenin signal pathway in hepatocellular carcinoma

Previous evidences reveal that long non-coding RNA (lncRNA) down syndrome critical region 8 (DSCR8) involves in the progression of multiple cancers. However, the exact expression, function, and mechanism of DSCR8 in hepatocellular carcinoma (HCC) remain uncovered. In this study, real-time PCR in HCC tissues and cell lines indicated that DSCR8 expression was upregulated, while miR-485-5p was downregulated. MTT assay, plate clone formation, Edu assay, flow cytometry, and in vivo experiments indicated that DSCR8 promoted HCC cell proliferation and cycle, whereas accelerated cell apoptosis. Luciferase reporter gene assay, RIP assay, and rescue experiments demonstrated that DSCR8 functioned as a competing endogenous RNA (ceRNA) by sponging miR-485-5p in HCC cells. Furthermore, gain- and loss-of-function studies showed that miR-485-5p activated Wnt/β-catenin signal pathway by targeting Frizzled-7 (FZD7). Moreover, DSCR8 activated Wnt/β-catenin signal pathway to promote HCC progression by DSCR8/miR-485-5p/FZD7 axis. Statistical analysis revealed that DSCR8 and miR-485-5p were closely related to some malignant clinicopathological features and 5-year survival rates of HCC patients. Taken together, the present study reports for the first time that DSCR8 activates Wnt/β-catenin signal pathway to promote HCC progression by DSCR8/miR-485-5p/FZD7 axis. The findings provide promising and valuable strategies for targeted therapy of HCC.


Introduction
As one of the most common cancers in the world, hepatocellular carcinoma (HCC) has characteristics of high morbidity and high mortality [1][2][3] . In the past decades, though researchers have been long committed to identifying the potential therapeutic targets to improve the diagnosis and treatment levels for HCC, the outcomes of HCC patients remain unsatisfactory 2 . Thus, it is important for us to discover some novel and practical therapeutic targets for HCC.
In recent years, non-coding RNAs, including long noncoding RNAs (lncRNAs) and microRNA (miRNAs), have been largely reported in studies about cancers, including HCC 4,5 . In our previous studies, we found that some lncRNAs, such as CASC2 6 , TUSC7 7 , and Ftx 8 , act as competing endogenous RNAs (ceRNAs) to regulate HCC cells' migration, invasion, proliferation, apoptosis, and so on. For example, we found that lncRNA CASC2 exerts its inhibitory effects on HCC cells through CASC2/miR-367/ FBXW7 pathway 6 . And we also found that lncRNA TUSC7 acts as a molecular sponge for miR-10a to suppress HCC cells' migration and invasion 7 . It is worth noting that, recently, lncRNA down syndrome critical region 8 (DSCR8) has been found to be dysregulated in uterine cancer and melanoma 9,10 . In these cancers, DSCR8 is highly expressed and might be potential prognostic indicators and therapeutic targets 9,10 . However, the expression and functions of DSCR8 in HCC remain unknown. MiR-485-5p has been identified as an antioncogene in HCC, which is involved in multiple biological and pathological processes of HCC 11,12 . However, the underlining mechanisms of miR-485-5p remain to be further explored.
Frizzled-7 (FZD7) is one of the receptors for Wnt signaling pathway 13 . It has been strongly confirmed that FZD7 is highly expressed in multiple cancers, including HCC [14][15][16] . And overexpressed FZD7 promotes the progression of cancers by inducing the activation of Wnt signaling pathway 13,17 . Recently, Wu J et al. found that miR-485-5p represses invasion and proliferation of melanoma cells by targeting FZD7 18 . However, whether FZD7 is regulated by miR-485-5p in HCC is still uncovered.
In the present study, we attempted to explore the expression, clinical significance, functions, and potential mechanisms of DSCR8 in HCC. DSCR8 was determined to be highly expressed in HCC. Gain-and loss-of-function analysis revealed that DSCR8 promoted cell proliferation and cell cycle, whereas suppressed cell apoptosis in HCC. Furthermore, the relationships among DSCR8, miR-485-5p, FZD7, and Wnt/β-catenin signal pathway in HCC cells were investigated. We found that DSCR8 activated Wnt/β-catenin signal pathway to promote HCC progression by DSCR8/miR-485-5p/FZD7 axis. DSCR8 and miR-485-5p were closely related to some malignant clinicopathological features and prognosis of HCC patients. In conclusion, DSCR8/miR-485-5p/FZD7 signal pathway may provide a novel and promising treatment strategy for HCC.

Expression and clinical significance of DSCR8 in HCC
The expression of DSCR8 in HCC tissues was detected by real-time PCR. And we found that the median expression of DSCR8 was much higher in HCC tissues than that in non-tumor tissues (P < 0.001, Fig. 1a). Then we analyzed the data from GEO dataset (GSE54236). Interestingly, the data displayed that DSCR8 was frequently upregulated in HCC tissues (P < 0.001, Fig. 1b), which was consistent with our finding. Additionally, DSCR8 was overexpressed in HCC cell lines compared to that in human normal liver cell (LO2) (P < 0.05, P < 0.001, respectively, Fig. 1c). Taken together, we conclude that DSCR8 might be an oncogene in HCC.

DSCR8 promotes cell proliferation and cell cycle and inhibits cell apoptosis in HCC
We increased the DSCR8 expression of Hep3B cells by pcDNA/DSCR8 and decreased the DSCR8 expression of Huh7 cells by sh-DSCR8. As shown in Fig. 2a, the transfection efficiencies were validated by real-time PCR. Then 3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide (MTT) assay, 5-ethynyl-2¢-deoxyuridine (Edu) assay, plate clone formation assay, and flow cytometry for detection of cell cycle were conducted to assess the effects of DSCR8 on cell proliferation and cell cycle in HCC cells. Results showed that pcDNA/DSCR8 accelerated the proliferation (Fig. 2b-d) and cell cycle (Fig. 2e) of Hep3B cells, while sh-DSCR8 inhibited the proliferation ( Fig. 2b-d) and cell cycle (Fig. 2e) of Huh7 cells. In addition, flow cytometry for detection of cell apoptosis revealed that pcDNA/DSCR8 repressed apoptosis of Huh7 cells (Fig. 2f), while sh-DSCR8 had the contrary effect on Hep3B cells' apoptosis ( Fig. 2f). Thus we conclude that DSCR8 promotes proliferation and cell cycle, while induces apoptosis in HCC cells.

DSCR8 promotes HCC growth in vivo
Next, subcutaneous tumor models were conducted to explore the effects of DSCR8 in vivo. We found that the sizes and weights of tumor nodes were increased compared by pcDNA/DSCR8 compared to control group ( Fig.   Fig. 1 The expression of DSCR8 in HCC tissues and cell lines. a The real-time PCR results from our patients cohort revealed that DSCR8 was significantly upregulated in HCC tissues (T, n = 75) than that in normal tissues (NT, n = 75). b Data from GEO dataset (GSE54236) showed that the expression of DSCR8 in HCC tissues (n = 81) was significantly higher than that in normal tissues (n = 80). c DSCR8 expression was upregulated in HCC cell lines compared to normal hepatocyte (LO2). *P < 0.05, ***P < 0.001 3a), while sh-DSCR8 resulted in decreases in the sizes and weights of tumor nodes (Fig. 3b). Immunohistochemistry for Ki-67 and terminal deoxinucleotidyl transferasemediated dUTP-fluorescein nick end labeling (TUNEL) staining in the xenografted tissues revealed that pcDNA/ DSCR8 increased the proportion of Ki-67-positive cells (Fig. 3c) and reduced the proportion of apoptotic cells (Fig. 3e), while sh-DSCR8 had the contrary effects (Fig. 3d, f). Thus these findings demonstrate that DSCR8 promotes HCC growth both in vitro and in vivo.

DSCR8 acts as a molecular sponge for miR-485-5p in HCC cells
Previous evidences reveal that lncRNAs is capable of acting as a molecular sponge for miRNAs to participate in the cancer progression 19 . And to explore whether DSCR8 could act as a molecular sponge for miRNAs in HCC cells, RNA fluorescence in situ hybridization (RNA-FISH) for the subcellular localization of DSCR8 was conducted. The result indicated that DSCR8 was localized both in the cell nuclear and cytoplasm ( Supplementary Fig. 1), which suggested that DSCR8 could act as a molecular sponge for miRNAs. Then we attempted to determine the most potential miRNA that can be regulated by DSCR8 via applying bioinformatics tools (microRNA.org, TargetScan, and miRDB). The data suggested that there were putative binding sites between DSCR8 and miR-485-5p (Fig. 4a). In addition, miR-485-5p has been identified as an anti-oncogene in HCC 11,12 . Thus we focused on miR-485-5p. By performing real-time PCR in all of the 75 paired HCC tissues and the adjacent non-tumor tissues, we confirmed that miR-485-5p was significantly underexpressed in HCC tissues (P < 0.001, Fig. 4b). Pearson correlation analysis indicated that DSCR8 expression was negatively related to miR-485-5p expression in HCC tissues (r = −0.8848, P < 0.001, Fig. 4c). In addition, the expression of miR-485-5p was underexpressed in HCC cell lines compared to LO2 cells (P < 0.001, respectively, Fig. 4d). Furthermore, miR-485-5p was negatively regulated by DSCR8 in Huh7-pcDNA/DSCR8 and Hep3B-sh-DSCR8 cells (Fig. 4e). We altered the miR-485-5p expression levels of Huh7 and Hep3B cells by miR-485-5p mimics and inhibitors (Fig. 4f). Interestingly, the expression of DSCR8 was negatively regulated by miR-485-5p (Fig. 4g). Furthermore, luciferase reporter gene assay revealed that miR-485-5p directly targeted 3′UTR of DSCR8-wt to negatively regulate the luciferase activity of DSCR8-wt-3′UTR, rather than 3′UTR of DSCR8-mut (Fig. 4h). Then anti-Ago2 RNA immunoprecipitation (RIP) assay was conducted with miR-485-5p mimics, and the data revealed that both DSCR8 and miR-485-5p were enriched in pulled down Ago2 protein (Fig. 4i). Taken
miR-485-5p mediates the effects of DSCR8 on proliferation, cell cycle, and apoptosis of HCC cells Next, we conducted rescue experiments to explore whether miR-485-5p mediated the effects of DSCR8 on cell proliferation, cell cycle, and cell apoptosis in HCC cells. Rescue experiments in Hep3B cells revealed that miR-485-5p mimics receded the promotion effect of pcDNA/DSCR8 on cell proliferation (Fig. 5a-c) and cell cycle (Fig. 5d) and reversed the inhibitory effect of DSCR8 on cell apoptosis (Fig. 5e). In contrast, miR-485-5p inhibitors reversed the inhibitory effect of sh-DSCR8 on cell proliferation ( Fig. 5a-c) and cell cycle (Fig. 5d) in Huh7 cells while weakened the promotion effect on apoptosis (Fig. 5e). Thus we demonstrate that miR-485-5p mediates the effects of DSCR8 on proliferation, cell cycle, and apoptosis of HCC cells.

Discussion
Growing evidences confirm that non-coding RNAs, especially lncRNAs and miRNAs, have considerable potential value for improving diagnostic and therapeutic levels for HCC 20,21 . Recently, researchers have identified numerous lncRNAs and miRNAs in HCC, such as lncRNA HNF1A-AS1 22 , LncRNA-NEF 22 , miRNA-874 23 , and so on. In these researches, abnormally expressed noncoding RNAs involve in various cytological behaviors of HCC cells 22,23 , which suggests that lncRNAs and miRNAs deserve further studies. LncRNA DSCR8 is a newly identified non-coding RNA in uterine cancer and melanoma, which might be potential prognostic indicators and therapeutic targets in these cancers 9,10 . However, the expression and functions of DSCR8 in HCC remain uncovered. In the present study, we reported the expression, functions, and potential mechanism of DSCR8 in HCC for the first time. Our study revealed that DSCR8 was overexpressed in HCC tissues and cell lines, which was consistent with the data from GEO dataset. These findings suggests that DSCR8 may be a potential oncogene in HCC.
Then we explored the functions of DSCR8 on cell proliferation, cell cycle, and cell apoptosis in HCC by gain-and loss-of-function experiments. As we expected, upregulated DSCR8 promoted cell proliferation and cell cycle and repressed cell apoptosis, whereas silenced DSCR8 had the contrary effects. Besides, we confirmed the functions of DSCR8 in HCC by establishing nude mice models. These findings demonstrates that DSCR8 plays important roles in HCC growth.
Next, we explored the potential mechanism of DSCR8 in HCC cells. It is well known that lncRNAs exert their influences through multiple ways among which acting as ceRNAs is a very general and an important path for lncRNAs to exert the influences. First, we conducted RNA-FISH assay to explore the subcellular location of DSCR8 in HCC cells. The result indicated that DSCR8 was localized both in the cell nuclear and cytoplasm, which suggested that DSCR8 could act as a molecular sponge for miRNAs. Then, based on the analysis results from bioinformatics tools and previous studies 11, 12 , we speculated that DSCR8 might act as the sponge for miR-485-5p in HCC cells. Interestingly, we found that miR-485-5p was significantly underexpressed in HCC tissues and cell lines. And there existed a negative correlation between DSCR8 expression and miR-485-5p expression in HCC. In addition, the expression of miR-485-5p was negatively regulated by DSCR8 in HCC cells. Meanwhile, miR-485-5p also negatively regulated DSCR8. Subsequently, luciferase reporter gene and anti-Ago2 RIP revealed that DSCR8 directly targeted miR-485-5p in HCC cells. What is more, rescue experiments in in vitro experiments manifested that miR-485-5p was a mediator for DSCR8 in HCC cells. These data strongly demonstrate that DSCR8 may act as a sponge for miR-485-5p in HCC cells.
Furthermore, we attempted to explore the downstream target of miR-485-5p in HCC cells, which may mediate DSCR8/miR-485-5p axis. Then bioinformatics tools combined with previous studies 18 were employed for comprehensive analysis. Based on the bioinformatics tools, we found that FZD7, a very critical receptor for activation of Wnt/β-catenin signal pathway 13 , may be one of the targets of miR-485-5p. Besides, it has been reported that FZD7 is a target of miR-485-5p in melanoma cells 18 , and FZD7 is an oncogene in HCC 24 . Results from our patients' cohort confirmed that FZD7 was significantly upregulated in HCC tissues than that in normal tissues, which was consistent with the data from database UAL-CAN (http://ualcan.path.uab.edu/index.html). Then luciferase reporter gene indicated that miR-485-5p targeted FZD7. In addition, real-time PCR and western blot for rescue experiments showed that miR-485-5p negatively regulated the expression of FZD7, the accumulation of both cytoplasmic and nuclear β-catenin, and the expression of Wnt/β-catenin pathway downstream targets c-Myc and cyclin D1 in HCC cells. The above data revealed that miR-485-5p directly targeted FZD7 to inhibit Wnt/βcatenin signal pathway in HCC cells. Our above findings revealed that miR-485-5p was a mediator for DSCR8 in HCC cells; then we attempted to explore the relationships among DSCR8, miR-485-5p, FZD7, and Wnt/β-catenin pathway. In the subsequent rescue experiments, we measured the expression changes of FZD7, the accumulation of both cytoplasmic and nuclear β-catenin, and the expression of c-Myc and cyclin D1. We found that miR-485-5p or FZD7 mediated DSCR8-induced activation of Wnt/β-catenin pathway. Thus we conclude that DSCR8 activates Wnt/β-catenin signal pathway to promote HCC progression by DSCR8/miR-485-5p/FZD7 axis. Clinically, we found that DSCR8 and miR-485-5p were closely related to tumor size, TNM stage, venous invasion, 5-year overall survival, and 5-year disease-free survival of HCC patients.
Herein, the present study determines the expression, clinical significance, and functions of lncRNA DSCR8 in HCC for the first time. In addition, we found that DSCR8 activates Wnt/β-catenin signal pathway to promote HCC progression by DSCR8/miR-485-5p/FZD7 axis (Fig. 9). Our study provides a novel potential targeted therapy strategy for HCC.

Tissue samples
HCC tissue samples and adjacent non-tumor tissue samples, which were histopathologically confirmed, were collected from 75 HCC patients who underwent surgery at the First Affiliated Hospital of Xi'an Jiaotong University from January 2009 to December 2011. All of the patients did not receive chemotherapy or radiotherapy before surgery. All of the samples were stored at −80°C. Our study got approval from the Ethics Committees of the First Affiliated Hospital of Xi'an Jiaotong University, and informed consent was obtained from all of the patients.

Real-time PCR
The assay was conducted according to the protocols described in our previous studies 7 . All RNA was extracted based on the protocol of TRIzol reagent (Invitrogen, Carlsbad, CA, USA). In brief, for detection of miRNA expression, cDNA was synthesized using the qScript microRNA cDNA Synthesis Kit (Quantabio, Beverly, MA)  Detection of cell proliferation, cell cycle, and cell apoptosis MTT assay, plate clone formation assay, Edu assay, flow cytometry for cell cycle, and cell apoptosis were conducted according to the protocols described in our previous studies 25,26 . Briefly, with respect to MTT assay, proliferation of cells was measured by MTT solution (5 mg/ml; Beyotime Institute of Biotechnology, Shanghai, China) at 24, 48, and 72 h, respectively. With respect to plate clone formation assay, 1 × 10 3 cells were seeded in 6well plates. The cells were mixed and then cultured for 2 weeks in culture medium with 10% FBS. Clusters containing ≥30 cells were counted as a single colony. With respect to Edu assay, Cell-Light™ EdU Apollo®488 In Vitro Imaging Kit (#C10310-3, RiboBio Co., LTD, Guangzhou, China) was used. With respect to flow cytometry for cell cycle and cell apoptosis, PE Annexin V Apoptosis Detection Kit I (#559763, Becton Dickinson bioscience, San Jose, CA, USA) and PI/RNase Staining Buffer (#550825, Becton Dickinson bioscience, San Jose, CA, USA) were used. All of the above experiments were performed according to the manufacturer's protocols.