Retracted: Long noncoding RNA nuclear enriched abundant transcript 1 impacts cell proliferation, invasion, and migration of glioma through regulating miR-139-5p/ CDK6

2.1 Human tissue samples

A total of 78 pairs of glioma tissues and adjacent tissues were collected from 78 glioma patients who were undergoing surgical treatment at the Affiliated Hospital of Xuzhou Medical University during March 2012 to December 2013. Pathological examination was conducted to confirm the disease. All the patients were received surgery, and without radiation treatment or chemotherapy. Specimens obtained distant from the tumor at least 3 cm were used as adjacent tissue specimens. Tumor tissue specimens were diagnosed by pathological examination. All experiments were approved by the local Ethics Committee of the Affiliated Hospital of Xuzhou Medical University. Informed consent was obtained from all patients in the current study.

2.2 Microarray analysis

Microarray data set was obtained from Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/geo, accession number GSE4290). The data set GSE4290 includes 23 nontumor samples and 81 glioblastoma samples. Affymetrix Human Genome U133 Plus 2.0 Array was used for analysis. Differentially expressed lncRNAs and mRNAs were identified using a t test (p < 0.05) combined with fold change (FC; log₂ (FC) > 2 for upregulated lncRNAs and mRNAs, and log₂ (FC) < −2 for downregulated lncRNAs and mRNAs). These differentially expressed lncRNAs was visualized by a heat map and volcano analysis using R (https://www.r-project.org/).

2.3 Cell culture and cell transfection

Glioma cell lines U251, SHG-44, TJ905 and human astrocyte cell line HA were obtained from BeNa Culture Collection (Shanghai, China). HEK-293T cells were purchased from the stem cell bank of Chinese Academy of Sciences. The glioma cell lines U251, SHG-44, and TJ905 and human astrocyte cell line HA (BeNa Culture Collection) were cultured in high glucose Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT) and 1% penicillin-streptomycin (HyClone). All the cells were seeded in the 24-well plates and incubated at 37°C in the humidified incubators with 5% CO₂. MiR-139-5p mimics (50 nM; GenePharma, Shanghai, China) were transiently transfected into glioblastoma cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) and negative control (NC) (50 nM; GenePharma, Shanghai, China) group was transfected with meaningless sequences. LncRNA transfection was divided into NC (50 nM) group, si-NEAT1#1 (50 nM) group (the sequence was 5′-GGAGGAGTCAGGAGGAATA-3′) and si-NEAT1#2 (50 nM) group (the sequence was 5′-GATCCCTAAGCTGTAGAACAT-3′). The CDK6-expressing vector (1 μg, pCMV-CDK6) was constructed (GenePharma) and transfected to overexpress CDK6 mRNA expression level. Medium was replaced after 24 hr and cell culture then continued. Cell transfection efficiency was detected after 24-hr transfection.

2.4 qRT-PCR

Total RNA of tissue samples or cell samples was extracted using Trizol reagent (Invitrogen). Complementary DNA (cDNA) was reverse transcribed according to the manufacturer's instructions. MiR-139-5p was tested using miRNAs Qper Quantitation Kit (Invitrogen), mRNA CDK6 and lncRNA NEAT1 were tested using RT-PCR Kit (Invitrogen), mRNAs and lncRNAs were normalized using GAPDH and miRNAs were normalized using U6. Primers were listed in Table 1. Data were analyzed by StepOne software after reaction. The 2^−ΔΔCt method was adopted and applied to calculate the relative expression.

2.5 Western blot

Cells were collected in centrifuge tubes after transfection and washed twice by phosphate buffered solution (PBS) equilibrium liquid. Denatured protein was electrophoresed on an 8%–10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to nitrocellulose membrane. The membrane was blocked in 5% nonfat milk and incubated with diluted primary antibodies at 4°C overnight followed by incubation for 2 hr at room temperature with a secondary antibody. Immunoblots were visualized by enhanced chemiluminescence (ECL kit, Santa Cruz Biotechnology, CA). And the relative integrated density values were calculated by Quantity One Software (Bio-Rad, Hercules, CA). The primary antibody was rabbit monoclonal to CDK6 (1:50000, #ab124821; Cambridge, MA), and the secondary antibody was goat anti-rabbit IgG H&L (HRP; 1:2000, #ab6721, Cambridge, MA, USA).

2.6 Immunohistochemistry

Tumor tissue samples were fixed in 4% paraformaldehyde, and then were dehydrated, embedded in paraffin. Serial 4 μm thick sections were deparaffinaged by xylol, rehydrated, and treated with peroxide in 100% methanol for 30 min to inhibit endogenous peroxidase activity. Slides were heated for antigen retrieval in 10 mM of sodium citrate (pH = 6.0). Endogenous peroxidase activity was blocked by 0.3% hydrogen peroxide solution. Sections were washed with PBS (pH = 7.4) again and incubated in 5% bovine serum albumin (Sigma, St. Louis, MO) for 30 min to block nonspecific antibody binding. The primary antibody discussed above (1:250) was incubated at 4°C overnight. Secondary antibody goat anti-rabbit IgG H&L (HRP; 1:1000, #ab6721, Cambridge, MA) were incubated after being washed with PBS. 3,3′-diaminobenzidine (DAB) (Beyotime Institute of Biotechnology, Jiangsu, China) systems were used for detection and hematoxylin was used to stain. OLYMPUS BX-41 microscope (Olympus, Center Valley, PA) was used to observe.

2.7 MTT assay

Cells were seeded into 96-well culture plates and then added with MTT kit (20 μl, 5 mg/ml; Sigma) to incubate for 4 hr. The reaction was terminated by removal of the supernatant and addition of 100 μl dimethyl sulfoxide (DMSO). The optical density (OD) value of each well was measured at 490 nm.

2.8 Cell cycle analysis by flow cytometry

Cell cycle was tested by cell cycle kit (KeyGen, Nanjing, China) according to the manufacturer's instructions. Cells were washed with cold PBS and collected after centrifuging for 5 min with 1000 rpm. Cells were fixed with 75% ethanol overnight at 4°C in the fridge and washed twice with PBS. Then the cells were incubated with 100 μl RnaseA for 30 min at room temperature in the dark. Flow cytometry was used for test after cells were added 50 μl PI and kept in the dark for 1 hr.

2.9 Cell apoptosis assay by flow cytometry

Apoptotic cells were tested by apoptosis detection kit (KeyGen, Nanjing, China) according to the manufacturer's instructions. Cells under log phase were obtained. The group only added with Lipofectamine 3000 reagent (Invitrogen) was used as blank control group. 400 μl binding buffer was used to resuspend cells after cells were collected. Cells were added with isometric Annexin V-FITC and propidium iodide (PI; 3–5 μl) and incubated for 10–15 min at room temperature in the dark. Cells which had no obvious cell aggregates were tested using the flow cytometry and analyzed by Cell Quest software.

2.10 Invasion and migration ability by transwell assay

All the reagents and equipment were put on the ice to precool. Transwell chambers were put in 24-well plate and smeared with 50 μl (0.2 μg/μl) Matrigel (BD Biosciences, Franklin Lakes, NJ) and then incubated for 15 min at 37°C. Serum-free medium was used to dilute cells at a density of 2.5 × 10⁴ cells/ml after digesting, centrifuging and counting cells. Cell suspension was seeded in the upper chamber 200 μl/well. 500 μl culture medium containing 10% FBS was seeded in the lower chamber. The cells were fixed with paraformaldehyde and stained with crystal violet for 15 min. The noninvasive cells were erased with cotton swabs. After drying out, the cells that invaded through the membrane and stuck to the lower surface of the membrane were counted in four random fields (×200) under a light microscope.

Cell migration assay was basically the same as invasion assay discussed above. However, transwell chamber need not to be coated with Matrigel. Besides, the number of cells seeded in the chamber was different. The density of U251 cells in cell migration assay was 1.5 × 10⁵ cells/well.

2.11 Dual-luciferase reporter assay

CDK6 3′untranslated region (UTR) and NEAT1 cDNA were amplified by PCR and CDK6 3′UTR segments and NEAT1 cDNA fragment were obtained and separated by gel electrophoresis. CDK6 3′UTR mutant vector and NEAT1 mutant vector were structured. Finally, restriction analysis and sequencing were performed to verify recombinant plasmids. Plasmid extraction was performed on CDK6/NEAT1 3 'UTR dual-luciferase reporter. Plasmid DNA (0.2 μg) was mixed with FugeneHD (0.3 μl; Roche Applied Science, Mannheim, Germany). Each group had three wells and NC group was set. 0.15 μg sensor reporter genes were diluted by 30 μl Opti-MEM (Life Technologies, Gaithersburg, MD) and mixed with 0.9 μl FugeneHD and then placed at room temperature for 15 min. Into each well, 15 μl of the above mixture was added. Cells were cultivated and transfected for 48 hr. Lysis buffer was diluted by double distilled water and added into each well. Lysis buffer centrifugation sedimentation was collected after 1-hr shaking. Firefly luciferase and sea cucumber luciferase substrate were added into the sediment. ELIASA was used to detect fluorescence intensity, and then analytical data were recorded.

2.12 Xenograft model

All the experiments were carried out with the approval of animal care and Committee of the Affiliated Hospital of Xuzhou Medical University. The precipitation of U251 cells at logarithmic phase were sustained after conventional digestion and centrifugation and washed three times with PBS buffer solution. The cells were resuspended using DMEM medium (5 × 10⁷/ml). Four-week-old BALB/c nude mice were selected and cultured in specific pathogen-free (SPF) level environment. The glioma cells were subcutaneously injected into the left/right flanks of the mice (2 × 10⁶ cells/mouse). The diameters of the tumors were measured every 3–5 days after subcutaneous injection of the glioma cells, and the tumor volumes were estimated. After 5–6 weeks, mice were killed and tumor growth was photographed. Volumes and weights of tumors separated from subcutaneous tissues were recorded.

2.13 Statistical analysis

Data were analyzed using GraphPad Prism 6.0 software (GraphPad Software Inc., CA), and were presented as mean ± standard deviation. Difference between two groups was compared using Student's t test. The difference among three groups and above was estimated using one-way analysis of variance. All experiments were repeated at least three times. p < 0.05 was considered to be statistically significant.

3.1 NEAT1 was high expressed in glioma tissues and cells

According to the microarray analysis, we found that there existed 71 high-expressed lncRNAs and 92 low-expressed lncRNAs in glioma tissues (Figure 1a). Of all these lncRNAs, lncRNA NEAT1 was significantly upregulated, with an average increase of 2.2 times (adj. p. Val = 6.99E-13). We respectively selected the top 10 lncRNAs which was the most significantly upregulated or downregulated as shown in Figure 1b. Furthermore, quantitative RT-PCR (qRT-PCR) also revealed that NEAT1 presented a considerably higher expression in glioma tissues compared with the adjacent tissues (p <  0.01; Figure 1c). Meanwhile, the endogenous expressions of NEAT1 in human astrocytes cell line HA and glioma cell lines U251, TJ905, SHG-44 were tested by qRT-PCR. Compared with that in human astrocytes HA, the expression of NEAT1 was remarkably higher in glioma cell lines (p < 0.01), especially in U251 (p < 0.001; Figure 1d). Therefore, U251 glioma cell line was chosen to perform the subsequent experiments.

3.2 NEAT1 promoted cell proliferation, invasion, and migration of glioma and inhibited cell apoptosis

NEAT1 expression of U251 cells was tested by qRT-PCR to verify the transfection efficiency. QRT-PCR result displayed that in the U251 cell line, NEAT1 expression was dramatically downregulated in both si-NEAT1#1 and si-NEAT1#2 groups compared with in NC group (p < 0.05; Figure 2a). NEAT1 expression had the similar trend in the SHG-44 cell line (p < 0.05; Figure 2c, Supporting Information Table S1). Furthermore, MTT assay also indicated that the cell viability in si-NEAT1#1 group and si-NEAT1#2 group was considerably lower than that in NC group since the second day, suggesting NEAT1 knockdown could repress U251 glioma cell proliferation ability (p < 0.05; Figure 2b). Both si-NEAT1#1 group and si-NEAT1#2 had parallel inhibitory effects on cell viability (p < 0.05; Figure 2d).

In addition, U251 and SHG-44 cell cycle and apoptosis condition were analyzed by flow cytometry assay. Compared with NC group, the proportion of cells of the S phase in both si-NEAT1#1 and si-NEAT1#2 groups remarkably decreased, whereas the proportion of cells of the G0/G1 phase increased (p < 0.05; Figure 3a). Furthermore, we found that the cell apoptosis was significantly enhanced after NEAT1 knockdown in comparison with NC group (p < 0.01; Figure 3b). In invasion and migration experiments, the number of U251 cells and SHG-44 cells in si-NEAT1 group was obviously fewer than in NC group (p < 0.05; Figure 4a,b). All the above results indicated that downregulation of NEAT1 could inhibit U251 cells and SHG-44 cells proliferation, invasion and migration and promote cell apoptosis. Because si-NEAT1#2 had better inhibitory effects and U251 cell line had more obvious differences, si-NEAT1#2 and U251 cell line were chosen for the subsequent experiments.

3.3 miR-139-5p was the target of lncRNA NEAT1

QRT-PCR results indicated that the expression of miR-139-5p in glioma tissues was significantly lower than that in adjacent normal tissues (p < 0.05; Figure 5a, Supporting Information Table S1). As presented in Figure 5b (Supporting Information Table S1), the expression of miR-139-5p was negatively correlated with the expression of NEAT1 (p < 0.05). After silencing NEAT1 or overexpressed miR-139-5p, the expression of NEAT1 in si-NEAT1 group was dramatically downregulated in contrast to NC group (p < 0.01; Figure 5c). When overexpressed miR-139-5p, the expression of NEAT1 had no obvious difference compared with the NC group (p > 0.05; Figure 5c). Differentially expression levels of miR-139-5p in U251 cells transfected with si-NEAT1 or miR-139-5p mimics were shown in Figure 5d. It was reflected that the expression of miR-139-5p was significantly upregulated in both si-NEAT1 group and miR-139-5p mimics group contrast to the NC group (p < 0.05). These results suggested that NEAT1 downregulated the expression of miR-139-5p. The binding relationship between NEAT1 and miR-139-5p was predicted through bioinformatics analysis (Figure 5e). After cotransfection of NEAT1-wt and miR-139-5p mimics into HEK-293T cells, the luciferase ratio remarkably declined (P < 0.01; Figure 5f), whereas there was no significant difference between scrambles group and NEAT1-mut group after cotransfection of NEAT1-mut and miR-139-5p mimics (p > 0.05).

3.4 miR-139-5p suppressed cell proliferation, invasion, and migration of glioma and accelerated cell apoptosis

MiR-139-5p expression of U251 cells was tested by qRT-PCR to verify the transfection efficiency. As displayed in Figure 6a, miR-139-5p mimics considerably upregulated miR-139-5p expression, leading to overexpression of miR-139-5p (p < 0.01). MTT assay revealed that the viability of U251 cells transfected with miR-139-5p mimics was remarkably repressed since the second day, in comparison with NC group (p < 0.01, Figure 6b). The proportion of the S phase dramatically reduced whereas the proportion of G0/G1 phase increased in the miR-139-5p group compared with that in the NC group (p < 0.01; Figure 6c). Furthermore, the flow cytometry revealed that the apoptosis rate in the miR-139-5p group considerably increased compared with that in the NC group (p < 0.05; Figure 6d).

Additionally, the transwell results illustrated that miR-139-5p overexpression exerted suppressive influence on glioma cell migration and invasion (Figure 7a,b; p <  0.01). Taken together, miR-139-5p inhibited glioma cells proliferation, invasion and migration and promoted cell apoptosis.

3.5 CDK6 was highly expressed in glioma tissues and cells

Microarray analysis revealed that there were 2,002 mRNAs with high expression and 2,348 mRNAs with low expression in glioma tissues compared to adjacent tissues. CDK6 was one of the highly expressed mRNAs. Compared with adjacent tissues, CDK6 was significantly upregulated in glioma tissues (adj.p.Val = 1.65E-07; Figure 8a,b). The difference in expression of CDK6 mRNA in glioma tissues and adjacent normal tissues was also detected by qRT-PCR. The expression of CDK6 in glioma tissues was significantly higher than that in adjacent tissues (p < 0.05; Figure 8d, Supporting Information Table S1). The immunohistochemistry results suggested that the expression of CDK6 protein was higher in glioma tissues than that in adjacent tissues (p < 0.05, Figure 8c). Pearson's correlation coefficient analysis indicated that miR-139-5p was negatively correlated with CDK6 in terms of expression (p < 0.05; Figure 8e, Supporting Information Table S1).

3.6 CDK6 was the target of miR-139-5p

Bioinformatics analysis was used to predict the target relationship between CDK6 and miR-139-5p. The binding sites were shown in Figure 9a. Dual-Luciferase reporter gene assay showed that there were significant differences between the miR-139-5p group and the NC group in luciferase activity when HEK-293T cells were transfected with CDK6-wt (p <  0.05; Figure 9b). However, there was no significant difference between the miR-139-5p group and the NC group after the mutation of CDK6 3′UTR (p > 0.05). The mRNA expression of CDK6 in glioma cells U251 detected by real-time quantitative PCR was shown in Figure 9c (p < 0.01). Compared with the NC group, the mRNA expression of CDK6 in the miR-139-5p group significantly decreased after transfected with miR-139-5p mimics, suggesting miR-139-5p could regulate the CDK6 mRNA expression. The expression levels of CDK6 protein both in the miR-139-5p group and the NC group were further determined by western blot, which revealed that the miR-139-5p overexpression could significantly down-regulate the expression level of CDK6 protein in U251 cells (p <  0.01; Figure 9d).

3.7 NEAT1 and CDK6 promoted cell proliferation, invasion, and migration of glioma and suppressed cell apoptosis

The expression of CDK6 was positively correlated with NEAT1 expression in U251 cells (p < 0.01; Figure 10a, Supporting Information Table S1). The experiments were divided into four groups, NC group, si-NEAT1 group, CDK6 group, and si-NEAT1+CDK6 group, respectively. The results of MTT indicated that the cell viability after transfection with CDK6 was significantly strengthened, while silencing NEAT1 could considerably suppress cell proliferation ability (p < 0.05; Figure 10b). There was no significant difference between the si-NEAT1+CDK6 group and the NC group in terms of cell viability (p > 0.05). The cell cycle and apoptosis condition were detected by flow cytometry assay, of which the results displayed the glioma cells arrested at G0/G1 phase after transfection with CDK6 remarkably decreased, while NEAT1 knockdown could further induced cell cycle arrest in G0/G1 phase (p < 0.01; Figure 10c). For apoptosis condition, flow cytometry results exhibited that CDK6 mimics inhibited cell apoptosis, whereas NEAT1 knockdown exerted facilitative influence on cell apoptosis (p < 0.01; Figure 10d). There was no significant difference between si-NEAT1+CDK6 group and NC group (p < 0.05).

In addition, the change of cell metastasis post-transfection was observed through transwell assay. The cell migration and invasion results both indicated that the number of migratory or invasive cells after silencing NEAT1 considerably decreased compared with the NC group, whereas upregulation of CDK6 expression contributed to cell mobility and invasiveness (p < 0.05; Figure 11a,b). Nonetheless, no significant difference was detected between the si-NEAT1+CDK6 group and the NC group (p > 0.05). These results strongly suggested that NEAT1 could regulate CDK6 to promote the invasion and migration of glioma cells.

3.8 NEAT1 knockdown restrained glioma growth through regulating CDK6 in vivo

The tumors in mice of the four groups with U251 cells transfected with si-NEAT1, CDK6, si-NEAT1+CDK6 and NC were taken and measured respectively (Figure 12a). Nude mouse experiment results indicated that the tumor volume of mice in CDK6 group was significantly larger than that in the NC group, whereas that in si-NEAT1 group presented the opposite trend. Likewise, after transfection with si-NEAT1, the tumor weight of mice remarkably decreased, whereas that of mice transfected with CDK6 mimics significantly increased (p < 0.05; Figure 12b,c). The tumor volume and weight of si-NEAT1+CDK6 group were equivalent to that of NC group (p > 0.05). The tumor volume and weight of mice in the si-NEAT1 group were minimal compared with that in NC group. The NEAT1 expression in the tumor of si-NEAT1 and si-NEAT1+CDK6 groups was significantly reduced compared with the NC group (p < 0.05; Figure 12d). In addition, the expression of miR-139-5p in the si-NEAT1 group and si-NEAT1+CDK6 group was higher than that in the NC group (p < 0.05; Figure 12e). No obvious difference of the expression levels of NEAT1 and miR-139-5p between NC group and CDK6 group (p > 0.05; Figure 12d,e). Moreover, CDK6 expression of CDK6 group increased while that in the si-NEAT1 decreased (p < 0.05; Figure 12f). There was no remarkable difference between the NC group and si-NEAT1+CDK6 group (p > 0.05).