All the samples (20 paired NSCLC tissues and adjacent normal tissues) were obtained from patients suffering from NSCLC who operated at the second affiliated hospital of Harbin Medical University between 2011.5 and 2012.8. All these patients did not recieve any local or systemic treatment before operation. All collected tissue samples were immediately treated by liquid nitrogen, then stored at −80°C before RNA extraction. All patients signed their informed consent. The study has got approval from the Research Ethics Committee of Harbin Medical University, and thus meets the standards of the Declaration of Helsinki in its revised version of 1975 and its amendments of 1983, 1989, and 1996.

Normal human embryonic lung fibroblasts cell line (HELF) and several NSCLC cancer cell lines, including A549, SPC-A1, H1299, 95D, SK-MES-1, and NCI-H520, were used in this study. These cells were cultured in RPMI-1640 (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) or Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (FBS) at 37°C in humidified air with 5% CO₂.

The siRNA (small interfering RNA) sequences (GenePharma, Shanghai, China) were shown in Table I. Synthetic sequence-scrambled siRNA was used as negative controls (NC). Human E2F3 gene was ligated into pGCMV vector (GenePharma). The empty vector was used as NC. Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) was used to complete transfection in A549 and H1299 cells according to the manufacturer's protocol.

Different groups of cells were seeded into 6 wells plate at 1×10⁵ cells per well. The cell number was counted for three consecutive days and then growth curve was drawn according to the cell number at 24, 48 and 72 h. 1×10⁶ cells were prepared for cell cycle analysis. First, the cells were fixed with 70% cold ethanol. After 12 h at 4°C, the cells were washed with PBS and stained with propidium iodide (PI). Finally, flow cytometry (BD Biosciences, San Jose, CA, USA) was conducted to analyze cell cycle.

The wild-type and mutant 3′-UTR of human E2F3 or lncRNA NEAT1 were inserted into the pmiR-RB-REPORT TM (Guangzhou RiboBio Co., Ltd., Guangzhou, China) through XhoI and NotI sites. Co-transfected wild-type/mutant reporter plasmid and miR-Ribo TM negative control into A549 and H1299 cells by Lipofectamine 2000 (Invitrogen), and then measured luciferase activities using the Dual-Luciferase Reporter Assay system 48 h post-transfection according to the manufacturer's instructions (Promega Corp., Madison, WI, USA).

Total RNA in cells and tissues was extracted using Trizol reagent (Invitrogen). RNA was reversed into cDNA through reverse transcription kits (Guangzhou RiboBio Co., Ltd.). Then the levels of target mRNAs were detected using qRT-PCR by SYBR^® Premix Ex Taq™II (Takara Bio, Inc., Otsu, Japan) in the ABI PRISM^® 7300 real-time PCR system (Applied Biosystems, Foster City, CA, USA). GAPDH and U6 were used as endogenous controls. The primer sequences involved in this study were shown in Table II.

Total protein was obtained from the A549 and H1299 cells after 48 h of transfection using cell lysis reagent. BCA protein assay reagent kit (Thermo Fisher Scientific, Inc.) was used to detected the protein concentrations in supernatant. The protein was separated in 10% SDS-polyacrylamide gels, transferred to PVDF membranes (Hertfordshire, UK), incubated in 5% milk powder to block non-specific reaction, and then treated with antibodies overnight at 4°C, including anti-E2F3 (1:1,000 dilutions; Affinity), anti-cyclin D1 (1:1,000 dilutions; Affinity), anti-cyclinD2 (1:1,000 dilutions; Affinity), anti-CDK4 (1:1,000 dilutions; Affinity), anti-p21 (1:1,000 dilutions; Affinity), anti-p57 (1:1,000 dilutions; Affinity) and anti-GADPH (1:5,000 dilutions; Affinity). At last, the membranes were incubated with second antibody (conjugated to horseradish peroxidase, 1:12,000 dilutions) for 2 h and exposed to ECL Advance reagent (GE Healthcare Biosciences, Buckinghamshire, UK) for 2 min. The band was visualized with Fluor S Multimager and Quantity One 4.1 (Bio-Rad Laboratories, Hercules, CA, USA).

In the present study, all the data are presented as means ± SD. A two-tailed paired t-test was used to reflect the difference between groups. P<0.05 indicated that the difference was statistically significant.

In order to validate the differential expression, NEAT1 was detected in 20 paired NSCLC tissues and adjacent normal tissues with qRT-PCR. The result showed that NEAT1 expression was significantly up-regulated in cancerous tissues than normal counterparts (Fig. 1A). Compared with patients whose cancer tissue expressed less NEAT1 (≤2 folds of increase, n=6), patients whose cancer tissue expressed more NEAT1 (>2 folds of increase, n=14) had shorter overall survival according to Kaplan-Meier survival analysis (Fig. 1B). The results demonstrated that high levels of NEAT1 were associated with NSCLC poor prognosis. Upregulation of NEAT1 might play important roles in NSCLC.

We first detected NEAT1 expression in NSCLC cell lines, including A549, H1299, SPCA-1, 95-D, SK-MES-1 and NCI-H520 cell line. The result showed that the expression level of NEAT1 in NSCLC cell lines was higher than that in 16HBE cell line (Fig. 2A). Then, endogenous NEAT1 expression was down-regulated by approximately 80% through small interfering RNAs (siRNAs) in A549 and H1299 cells (Fig. 2B and C). Compared with blank cells, the proliferation ability was decreased in A549-siRNA1 and H1299-siRNA1 cells (Fig. 2D and E). Cell cycle analysis indicated that deletion of NEAT1 induced cell cycle arrest in G1 phase.

In view of the potential function of lncRNA as a ceRNA or a molecular sponge of miRNAs, we screened the miRNAs interacting with NEAT1 using the bioinformatics, starBase2.0. Among these miRNAs, miR-377-3p (MIMAT0000730) was selected for further study. We first verified miR-377-3p expression in A549 and H1299 cells, and it is significantly higher than 16HBE cell (Fig. 3A). Dual-luciferase reporter assay indicated that miR-377-3p could decrease the luciferase activity of pmirGLO-NEAT1-WT (wild-type), however, it had no affect on the pmirGLO-NEAT1-MUT (Fig. 3B and C). Targetscan, miRanda and PicTar were used to predict the possible targets of miR-377-3p, and found that E2F3 involving in cell cycle regulation harbors two conserved miR-377-3p cognate sites. Luciferase reporter assays indicated that luciferase activity was declined more than 50% after co-transfected with miR-377-3p and the reporter vector with E2F3-3′-WT (Fig. 3D and E). Up-regulating miR-377-3p in A549 and H1299 cells suppressed the expression of NEAT1 and E2F3 (Fig. 3F and G). Inhibition of NEAT1 resulted in down-regulation of E2F3 protein level (Fig. 3H).

Inhibiting NEAT1 in A549 cells suppressed the protein expression of cyclinD1, cyclinD2 and CDK4, whereas increased the protein levels of p21 and p57 (Fig. 4A). Consistently, over-expression of E2F3 in A549-siRNA1 cells reversed the expression of these proteins mentioned above (Fig. 4A). Over-expression of E2F3 also reversed the cell proliferation ability inhibited by decreased NEAT1 (Fig. 4B). So, E2F3 mediated the cell proliferation induced by NEAT1 by regulating cell cycle associated proteins expression.