Long noncoding RNA OIP5-AS1 targets Wnt-7b to affect glioma progression via modulation of miR-410

The present study was undertaken to investigate the underlying mechanisms of long noncoding RNA OIP5-AS1 via regulating miR-410 to modulate Wnt-7b in the progression of glioma. To address this problem, we measured the expression of OIP5-AS1 and miR-410 in glioma tissues by qRT-PCR. Glioma U87 cells were transfected with OIP5-AS1 siRNA or miR-410 inhibitors. The targeting relationships among miR-410, OIP5-AS1 and Wnt-7b were verified by luciferase reporter assays. Western blotting was employed to determine the expression of Wnt-7b/β-catenin pathway-related proteins, while MTT, flow cytometry, Transwell assays and wound-healing assays were used to measure the biological characteristics of glioma cells. The results showed that OIP5-AS1 expression was higher and miR-410 was lower in glioma tissues. Luciferase reporter assays confirmed a targeting relationship between OIP5-AS1 and miR-410, as well as between miR-410 and Wnt-7b. Silencing OIP5-AS1 reduced cell proliferation, invasion and migration of glioma U87 cells and led to depressed expression levels of miR-410, Wnt-7b, p-β-catenin, GSK-3β-pS9, c-Myc and cyclin D1. Furthermore, down-regulation of OIP5-AS1 induced G0/G1 phase cell cycle arrest and apoptosis of glioma cells. Inhibitors of miR-410 abolished the biological effects of OIP5-AS1 siRNA in glioma cells. In vivo, OIP5-AS1 knockdown also inhibited tumor growth. Taken together, this research suggested that silencing OIP5-AS1 may specifically block the Wnt-7b/β-catenin pathway via targeted up-regulating miR-410, thereby inhibiting growth, invasion and migration while promoting apoptosis in glioma cells.


Introduction
Glioma, as a very common malignant tumor in the human central nervous system, features rapid growth, strong invasion, high mortality and high disability rates, accounting for approximately 60% of adult primary brain tumors [1,2]. Although great advances have been made in comprehensive therapies of glioma, including radiotherapy, chemotherapy and surgery, the clinical management of glioma still faces many challenges. In particular, glioma is usually found in hidden areas of the brain and is diagnosed at advanced/late stages, possibly causing difficulties for thorough resection and displaying resistance to both chemotherapy and radiotherapy [3,4]. The average survival time of patients diagnosed with glioma was only approximately 14 months with a 5-year survival rate of no more than 5% [5]. Therefore, it is necessary to study glioma progression at the molecular level and to identify new therapeutic approaches [6]. Currently, with the deepening understanding and application of molecular targeted therapy for glioma, the role of long noncoding RNA (lncRNA) in glioma development has attracted clinical attention [7]. LncR-NAs, consisting of more than 200 nucleotides without protein-coding potential [8], have been suggested to be substantially involved in the development and progression of glioma. For example, lncRNA TUG1 was identified as a tumor suppressor by Li et al. [9] to be down-regulated in glioma tissues, and it has the potential to be applied for the treatment of glioma. Another group reported that overexpressed lncRNA H19 in glioma tissues was negatively linked to the glioma prognosis and that silencing H19 inhibited the growth of glioma cells by modulating miR-675 [10].
OPA-interacting protein 5 antisense transcript 1 (OIP5-AS1), a member of the lncRNA family, is located on the human chromosome 15q15.1 and is evolutionarily conserved in vertebrates [11]. According to the study by Kim et al. [12], OIP5-AS1, a competing endogenous RNA (ceRNA) for HuR, suppressed the growth of the cervical cancer cell line HeLa. Yang et al. [13] also showed that OIP5-AS1 regulated the expression of miR-410 and modulated its target gene KLF10, thus affecting the downstream PTEN/PI3K/AKT signaling pathway in multiple myeloma by facilitating cell proliferation, cycle progression and apoptosis inhibition. Meanwhile, as illustrated by the study of Chen et al. [14], miR-410 acts on its target gene MET to effectively regulate the proliferation and invasion of glioma cells. Actually, miR-410 has been proven to specifically inhibit Wnt-7b expression, thereby influencing the proliferation and invasion of oral squamous cell carcinoma cells [15]. Furthermore, OIP5-AS1 is important for controlling neurogenesis during development, because it is highly expressed in the nervous system, as well as some types of human cancers [16]. Limited data are available concerning the possible association between OIP5-AS1 and glioma; however, Liu et al. [17] reported that PIWIL3/piR-30188/OIP5-AS1 regulated the biological behavior of glioma cells via the miR-367-3p/CEBPA/TRAF4 pathway. Therefore, after identifying targeting relationships between OIP5-AS1 and miR-410 and between miR-410 and Wnt-7b using a biological information prediction website, we investigated whether OIP5-AS1 regulated miR-410 to target Wnt-7b and affect glioma development.

Ethics statement
The study obtained approval of the Ethics Committee of The Second Hospital of Hebei Medical University and informed consent was obtained from all patients prior to the study. Furthermore, the animal experiments were approved by the Ethics Committee for Laboratory Animals, and all procedures were performed strictly in accordance with the Guide for the Care and Use of Laboratory Animals issued by the National Institutes of Health (NIH) [18].

Study subjects
From October 2016 to October 2017, tissue samples were obtained from 65 glioma patients who underwent surgical resection at the Department of Neurosurgery in our hospital; diagnoses were confirmed by the pathological examination. Hematoxylin-eosin (HE) staining was used to observe glioma morphology ( Figure 1A). Patients had not received any other therapies prior to surgery, and complications such as other malignant tumors, serious systemic infections and other severe systemic diseases were ruled out. The patients had a mean age of 46.23 + − 12.58 years, including 40 males and 25 females. Based on the pathological classification of tumors of the central nervous system published by WHO in 2007 [19], there were 29 cases of low-grade glioma (grade I/II), including 12 cases of grade I and 17 cases of grade II, and 36 cases of high-grade glioma (grade III/IV), including 19 cases of grade III and 17 cases of grade IV. Non-neoplastic brain tissue samples were obtained from 10 adult patients with craniocerebral injuries who underwent partial resections of brain tissues to reduce intracranial pressure. All samples were preserved at -80 • C for subsequent experiments.
First, U87 cells collected during the logarithmic growth phase were digested, centrifuged, made into single cell suspensions with antibiotic-free medium, and seeded into six-well plates at 2 × 10 5 cells/well before 12-24 h of incubation. Cells were cultured in complete medium at least 24 h before transfection. A total of 100 nM of OIP5-AS1 siRNA, siRNA NC or miR-410 inhibitors were diluted separately in 200 μl of Opti-MEM medium. After incubation for 15 min at room temperature, the oligonucleotides were added, and the cells were incubated for 48 h. When cells were adherent to the bottom and walls of the plate and cell confluence reached 50-70%. Lipofectamine-2000 (Invitrogen) was utilized for the transfection of U87 cells according to the manufacturer's protocol.

Dual-luciferase reporter assay
OIP5-AS1 and wnt-7b 3 -UTR fragments containing the binding site of miR-410 were amplified to construct wild-type plasmids OIP5-AS1-wt and Wnt-7b-wt, expressing the binding sites of miR-410 to OIP5-AS1 and wnt-7b, respectively, as well as mutant type plasmids OIP5-AS1-mt and Wnt-7b-mt, expressing the binding sites of miR-410 to OIP5-AS1 and wnt-7b, respectively. During the experiment, the human embryonic kidney cell line HEK 293T cells (purchased from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences, Shanghai, China) were obtained at the logarithmic growth phase, which were seeded to the 24-well plate and cultured in the DMEM medium with 10% FBS. When the cell confluence rose to 80%, cells were transfected by following the manual of the Lipofectamine 2000 (Invitrogen) reagent kit. Approximately 48 h after transfection, a dual-luciferase reporter gene assay kit (Promega) was used for detection before calculating the ratio of the firefly luciferase activity to Renilla luciferase activity. The experiment was performed three times to obtain mean values.

Expression levels of OIP5-AS1, miR-410 and Wnt-7b measured by qRT-PCR
The TRIzol reagent kit (Invitrogen, U.S.A.) was employed for the extraction of total RNA from glioma tissues, normal tissues and cells in each group; primer sequences were synthesized by Sangon Biotech (Shanghai) Co., Ltd. (Table 1). Next, total RNA (2 μg) was reverse-transcribed into cDNA with a Reverse Transcription Kit (Takara, Biochemical, Tokyo, Japan). In the subsequent step, the cDNA was amplified using a SYBR Premix Ex Taq II (Perfect Real-Time) kit (Akara, Biochemical, Tokyo, Japan). U6 was the internal reference gene for miR-410 and GAPDH was used for OIP5-AS1. The 2 − C t method was applied for the calculation of the relative expression of target genes. Each experiment was repeated three times independently.

Western blotting analysis
Cells were disrupted by RIPA lysis and the concentration of proteins was calculated by determination with a bicinchoninic acid (BCA) protein assay (Thermo, U.S.A.). Next, protein separation was achieved by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and proteins were transferred onto nitrocellulose membranes. Then, the membranes were placed in a refrigerator for overnight reaction at 4 • C with primary antibodies, as follows: Wnt-7b (

Cell proliferation assessed by MTT assay
Cells collected at the logarithmic growth phase were digested with trypsin and adjusted for cell density until it was 1 × 10 5 cells/ml. At 24, 48, 72 and 96 h after the cells were seeded in 96-well plates, 20 μl of MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) (Sigma, U.S.A.) solution (5 mg/ml) was added to each well for 4 h of incubation. Next, the solution in each well was removed, and 150 μl of dimethyl sulfoxide (DMSO) solution (Sigma, U.S.A.) was added into each well for 10 min of oscillation. Finally, an enzyme-linked immunosorbent assay (ELISA) reader (BIO-RAD, Cal, U.S.A.) was employed to measure the OD (optical density) value of each well under the wavelength of 490 nm. To obtain the mean value, the experiment was repeated three times independently.

Cell invasion detected by Transwell assay
After overnight thawing, Matrigel was diluted by DMEM medium three times its own volume and plated into the upper chambers of Transwell systems. Approximately 2 × 10 4 cells were inoculated into each well of the upper chamber, while 0.5 ml of DMEM medium containing 10% FBS was added to each well of the lower chamber simultaneously. After incubation for 24 h, the cells were fixed with paraformaldehyde and stained for half an hour with 0.1% Crystal Violet. Next, cells that did not go through the upper chamber were washed away with 0.1 ml of phosphate buffer saline (PBS). Finally, an optical microscope was used for the observation, counting and imaging. The experiment was repeated three times and the mean values were obtained.

Cell migration measured by wound-healing assay
A marker pen was used to draw three parallel lines on the back of the 12-well plate, and then 1 × 10 4 cells were seeded to each well. When cells evenly covered each well, a pipette tip was employed to scratch lines vertical to the three lines on the back of the plate, and 0.1 ml of PBS buffer was used to wash away the shed cells. An optical microscope was used for the observation and imaging of cells at 0 and 48 h of incubation in the culture medium. The experiment was repeated three times to obtain mean values.

Cell cycle and apoptosis evaluated by flow cytometry
After transfection for 48 h, cells were collected, digested with 0.25% trypsin and adjusted to the density of 1 × 10 6 cells/ml. Next, 1 ml of cell suspensions were centrifuged for 10 min at the rate of 1500 rpm; the supernatants were removed, and 2 ml of PBS buffer was added for another round of centrifugation. Then, the supernatants were discarded, and precooled 70% ethanol was added for the overnight fixation of cells at 4 • C. On the next day, 50 μg of RNAase-containing propidium iodide (PI) was added into 100 μl of cell suspension for 30 min of reaction without exposure to light, filtered by a 100-mesh nylon filter screen.

Statistical analysis
All data were processed and analyzed by using the statistics software SPSS 20.0 (SPSS, Inc, Chicago, U.S.A.). Measurement data were presented by mean + − standard deviation (x + − s). Mean values between two groups were compared using independent samples t-tests, while mean values among several groups analyzed by one-way ANOVA. Intergroup comparisons were tested by the least significant difference-test (LSD-t), and correlation analysis was conducted using the Pearson correlation coefficient. P<0.05 was regarded as statistically significant.

Expression of OIP5-AS1 and miR-410 in glioma tissues
The expression levels of OIP5-AS1 and miR-410 in glioma tissues and normal tissues were evaluated by qRT-PCR, and the data in human tissues showed that the expression levels of OIP5-AS1 were up-regulated, but the expression levels of miR-410 were down-regulated in glioma tissues compared with normal tissues (both P<0.05, Figure 1B). According to the correlation analysis results, expression levels of OIP5-AS1 negatively correlated to levels of miR-410 (r = −0.645, P<0.05, Figure 1C). Furthermore, expression levels of OIP5-AS1 and miR-410 were significantly associated with the pathological grading of glioma, because OIP5-AS1 was notably more highly expressed in high-grade glioma tissues than in low-grade tissues (P<0.05, Figure 1D); however, miR-410 levels were lower in high-grade glioma than in low-grade tissues (P<0.05, Figure 1E).

The targeting relationship among miR-410, OIP5-AS1 and Wnt-7b
OIP5-AS1 is a molecular sponge that regulates miR-410, as predicted by Starbase 2.0. With the use of online prediction software Target Scan, we identified the binding site between Wnt-7b and miR-410 (Figure 2A,B). According to the dual-luciferase activity assay, miR-410 mimics did not significantly differ from the NC group with respect to the effect on the luciferase activity of OIP5-AS1-mt and Wnt-7b-mt (all P>0.05); however, miR-410 mimics significantly   down-regulated the luciferase activity of wild-type OIP5-AS1-wt and Wnt-7b-wt (all P<0.05, Figure 2C,D). These results suggested that there was a targeting relationship between OIP5-AS1 and miR-410, as well as between miR-410 and Wnt-7b.

Expression of Wnt-7b/β-catenin signaling pathway-related proteins
We then examined the protein expression levels in U87 glioma cells and found that there was no difference between the NC and Blank groups with respect to expression levels of each protein, and the expression levels of total GSK-3β and total β-catenin were not significantly different in all groups (all P>0.05). Furthermore, protein expression levels of Wnt-7b, p-β-catenin, GSK-3β-pS9, c-Myc and cyclin D1 were lower in the OIP5-AS1 siRNA group (all P<0.05) but were remarkably higher in the miR-410 inhibitors group compared with the Blank group (all P<0.05). Moreover, cells in the siRNA+ inhibitors group were dramatically lower in terms of expression of miR-410 and increased in terms of expression of Wnt-7b/β-catenin pathway-related proteins when compared with cells in the OIP5-AS1 siRNA group (Figure 3, all P<0.05).

Comparison of cell proliferation among different groups
Proliferation of U87 glioma cells in each group was evaluated using the MTT assay ( Figure 4). First, there was no significant difference between the Blank and NC groups with respect to proliferation at various time points (P>0.05). Proliferation was apparently inhibited in the OIP5-AS1 siRNA group and was remarkably enhanced in the miR-410 inhibitors group compared with cells in the Blank group (all P<0.05). On the other hand, in comparison with the OIP5-AS1 siRNA group, the siRNA+ inhibitors group increased dramatically in terms of proliferation (all P<0.05), suggesting that miR-410 inhibitors reversed the inhibitory effect of OIP5-AS1 siRNA on proliferation.

Cell cycle and cell apoptosis of different groups
We further investigated the mechanism of transfection of the OIP5-AS1 on cell cycle and apoptosis in U87 human glioma cells. There were no differences between the NC and Blank groups with respect to the proportion of G 0 /G 1 phase, S phase and G2 phase (all P>0.05). Compared with cells in the Blank group, those in the OIP5-AS1 siRNA group showed a significantly elevated proportion of G 0 /G 1 phase, a reduced proportion of S phase, and apparently increased apoptosis rates (all P<0.05, Figure 5). Cells in the miR-410 inhibitors group were significantly reduced in terms of proportion of G 0 /G 1 phase cells and apoptosis rates, with elevated proportions of S phase cells (all P<0.05). On the other hand, by comparison with cells in the OIP5-AS1 siRNA group, those in the siRNA+ inhibitors group showed substantially fewer cells in the G 0 /G 1 phase and lower apoptosis rates (all P<0.05), with an increased proportion of S phase cells (all P<0.05). No significant differences were found in terms of the proportion of G2 phase in all groups (all P>0.05).

Comparison of cell invasion and migration among different groups
Transwell and wound-healing assays were employed to assess invasion and migration of U87 human glioma cells, respectively. As shown in Figure 6, the NC group showed no observable difference from the Blank group in terms of the number of invasive cells and the wound-healing rate (P>0.05). However, the number of invasive cells and the wound-healing rate were substantially lower in the OIP5-AS1 siRNA group and significantly higher in the miR-410 inhibitors group (all P<0.05). Furthermore, the siRNA+ inhibitors group was statistically higher than the OIP5-AS1 siRNA group with respect to these two indexes (all P<0.05).

Effect of OIP5-AS1 and miR-410 on the tumorigenicity of glioma cells in vivo
As demonstrated by the experiment of tumorigenicity in an in vivo nude mice model, the nude mice in the NC group were not significantly different in terms of tumor volume or weight at various time points compared with those of the Blank group (all P>0.05). For the nude mice in the OIP5-AS1 siRNA group, their tumor growth was inhibited with significantly reduced tumor volume and weight when compared with mice in the Blank and NC groups (all P<0.05); the miR-410 inhibitors group was completely opposite to the OIP5-AS1 siRNA group in terms of the changes in these indexes (all P<0.05, Figure 7A,B). HE staining of tumors in the Blank and NC groups showed typical histological   features of glioma, including high cellular density and an infiltrative growth pattern; the changes were substantial in miR-410 inhibitors group. In the OIP5-AS1 siRNA group, necrotic foci were observed, tissue alignment became irregular, and glioma cellular density decreased significantly compared with the other four groups ( Figure 7C).

Discussion
We found that OIP5-AS1 was significantly increased and miR-410 was decreased in glioma tissues, displaying a negative correlation relationship between OIP5-AS1 and miR-410 expression levels. As documented previously, OIP5-AS1 may act as a ceRNA to combine with HuR, thereby inhibiting the expression of HuR to participate in the phenotypic regulation of tumor cells [12], and lncRNA is increasingly identified as expressed aberrantly in tumors. Similarly, a study by Deng et al. [20] reported that OIP5-AS1 exerted oncogenic roles when highly expressed in lung adenocarcinoma tissues and cells via competitively binding to miR-448 to affect proliferation, migration and invasion. Furthermore, miR-410 was down-regulated in human glioma tissues, acting as a tumor suppressor to greatly affect glioma growth by down-regulating its target gene MET [14]. Therefore, we hypothesized that OIP5-AS1 may act as an oncogene and miR-410 may act as a tumor suppressor in glioma. We also found that the expression of OIP5-AS1 and miR-410 had a good deal to do with the pathological grading of glioma. To be specific, higher pathological gradings of glioma patients demonstrated higher expression levels of OIP5-AS1 and lower expression levels of miR-410, suggesting that OIP5-AS1 and miR-410 were closely related to the pathogenesis and development of glioma.
To investigate the regulation of OIP5-AS1 and miR-410 on the growth of glioma, we chose U87 glioma cells for transfection in vitro. We found that silencing OIP5-AS1 using siRNAs in U87 glioma cells inhibited cell growth via effectively suppressing proliferation, invasion and migration capabilities, and promoting apoptosis, as well as inducing G0/G1 phase cell cycle arrest. Consistent with our finding, Naemura et al. [11] also reported that silencing of OIP5-AS1 modulated the cell cycle and thereby regulated the proliferation of cervical cancer HeLa cells. In the present study, the trend of the miR-410 inhibitors group was observed to be completely opposite to that of the OIP5-AS1 siRNA group, suggesting that blocking miR-410 reversed the inhibitory role of OIP5-AS1 siRNA in terms of growth and metastasis. More importantly, our dual-luciferase assay confirmed the targeting relationship between OIP5-AS1 and miR-410, suggesting that OIP5-AS1 may play roles in glioma pathogenesis and progression by modulation of miR-410.
Expression of Wnt-7b/β-catenin signaling pathway-related proteins was determined to further elucidate the underlying mechanism of OIP5-AS1 in glioma. Expression levels of Wnt-7b, p-β-catenin, GSK-3β-pS9, c-Myc and Cyclin D1 were dramatically down-regulated and the expression of miR-410 was up-regulated in the OIP5-AS1 siRNA group. Nevertheless, cells in the miR-410 inhibitors group showed a completely opposite trend in the changes of these indexes to the OIP5-AS1 siRNA group, showing that inhibition of miR-410 reversed the effect of OIP5-AS1 siRNA to activate the activity of the Wnt-7b/β-catenin pathway. Similarly, as illustrated by Shiah et al. [15], miR-410 attenuated the Wnt-β-catenin pathway in oral squamous cell carcinoma cells by targeting Wnt-7b, an activator of the Wnt-β-catenin pathway. Furthermore, Wnt-7b levels were markedly lower in glioma tissues than in nontumor tissues, as illustrated Zhang et al [21]. Notably, the Wnt7b signaling pathway was shown to regulate distinct glioma-vascular interactions and tumor microenvironments [22]. Using the dual-luciferase reporter assay, we also confirmed that Wnt-7b was indeed the target gene of miR-410, suggesting that silencing OIP5-AS1 may affect growth and metastasis of U87 glioma cells via targeted regulation of Wnt-7b by miR-410. Wnt-7b serves as an important agonist of the Wnt/β-catenin signaling pathway [23,24], possibly preventing the phosphorylation and degradation of β-catenin induced by GSK-3β inhibition in cytoplasm; the accumulated β-catenin would translocate to the nucleus to bind with T-cell factor/lymphoid enhancer factor and then affect the expression of Wnt target genes, including cyclin D1 and c-Myc, eventually promoting tumor pathogenesis [25,26]. β-Catenin is the core member of the Wnt pathway that has been shown to be highly expressed in high-grade glioma and poorly expressed in low-grade astrocytoma, suggesting that β-catenin expression is associated with the degree of malignancy in glioma [27]. There was evidence that the knockdown of β-catenin greatly changed the growth and cell cycle distribution in glioma, inhibiting the proliferation and growth of glioma cells [28]. Moreover, c-Myc is the downstream target gene of the pathway and its enhanced expression was shown to be closely related to the development and progression of glioma [29]. In glioma cells, inhibition of cyclin D1 blocked progression of the cell cycle, inhibited proliferation and induced apoptosis [30,31]. Based on all this evidence, we may conclude that OIP5-AS1 siRNA specifically inhibited the Wnt-7b/β-catenin pathway and the downstream genes cyclin D1 and c-Myc, resulting in cell cycle arrest, thereby inhibiting cell proliferation, invasion and migration, and promoting apoptosis. Finally, similar results were also observed in the nude mice tumorigenesis model, where tumor volume and weight were significantly reduced and the growth was apparently inhibited in mice treated with OIP5-AS1 siRNA, further confirming the inhibitory effect of OIP5-AS1 siRNA on glioma growth. Therefore, silencing OIP5-AS1 may inhibit the Wnt-7b/β-catenin pathway by up-regulating miR-410, thereby suppressing glioma growth, invasion and migration and promoting apoptosis, deepening understanding of the molecular mechanism of glioma.