The GSE19804, GSE19188, GSE30219, GSE27262 data sets from the Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo) and The Cancer Genome Atlas (TCGA) TANRIC database (http://ibl.mdanderson.org/tanric/_design/basic/query.html) were analysed using bioinformatic methods. The gene expression analyses were assessed using a linear model with Limma package in R software (version 3.3.2) (14). Differentially expressed genes were identified using the threshold absolute fold change (FC) >1.5 and false discovery rate <0.05 as cut-offs.

For RNA-Seq, transcript alignments were conducted by HISAT2 (15) (version 2.0.5), assembly and quantification were conducted by StringTie (16,17) (version 1.3.3b), and differential expression analyses were conducted based on the negative binomial distributions with edgeR package in R software (version 3.18.1). Differential expression genes were identified as the threshold |log2(FC)| >1 and FDR <0.05. RNA transcriptome sequencing was performed by Novogene Bioinformatics Technology Co. Ltd.

A total of 35 paired LAD and normal lung tissues were obtained from patients (10 females and 25 males) who underwent primary surgical resection at the Jinling Clinical Medical College of Nanjing Medical University (Nanjing, China) from January 2015 to December 2016. The median age was 64 years and the range from 38 to 83 years. No treatments were conducted in these patients prior to surgery, and all tissues were immersed in RNALater stabilization solution (Qiagen GmbH) and stored at −80°C. All patients with LAD were diagnosed according to histopathological evaluation, which was based on the 2011 International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society: international multidisciplinary classification of lung adenocarcinoma (18). The study protocol and methods were approved by the Research Ethics Committee of Jinling Clinical Medical College of Nanjing Medical University. All of the participants provided written informed consent form and agreed to the use of their samples in scientific research.

A total of 3 LAD H1299, A549, PC9 and 16HBE cell lines were purchased from the Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences. H1299 and A549 cells were maintained in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.). PC9 and 16HBE cells were cultured in Dulbecco's modified Eagle's medium (Gibco; Thermo Fisher Scientific, Inc.). Both media were supplemented with 10% foetal bovine serum (FBS, HyClone; GE Healthcare Life Sciences). All of the cells were grown in a humidified atmosphere at 37°C with 5% CO₂.

Total RNA was isolated from cultured LAD cells or frozen tissue samples using TRIzol^® reagent (Thermo Fisher Scientific, Inc.). cDNA was synthesized using a Reverse Transcription kit (Takara Biotechnology Co., Ltd.). RT-qPCR assays were performed using a SYBR Premix Ex Taq II kit (Takara Biotechnology Co., Ltd.), and the target gene expression values were normalized to the expression of GAPDH. The results were analysed based on the comparative cycle threshold (Cq) (2^−ΔCq or 2^−ΔΔCq) methods (19). The primer sequences are listed in Table SI.

Small interfering RNAs (siRNAs; si-linc00467 1#, si-linc00467 2# and si-linc00467 3#) and plasmid vectors (pcDNA3.1 and pcDNA3.1-linc00467) were designed and synthesized by Shanghai GenePharma Co., Ltd., and the empty pcDNA3.1 vector was used as the control. The concentrations of siRNA and plasmids were 50 µM and 2 µg/ml, respectively. siRNAs were transfected into cells using HiPerFect Transfection Reagent (Qiagen GmbH). The plasmid vectors were transfected into cells using X-tremeGENE HP DNA transfection reagent (Roche Diagnostics). After 48 h of incubation in a humidified atmosphere at 37°C with 5% CO₂, the cells were harvested for subsequent experiments. Short hairpin RNA (shRNA) lentiviruses (linc00467 shRNA, control shRNA) were also designed and synthesized by Shanghai GenePharma Co., Ltd. The LV3 lentiviral vector (Shanghai GenePharma Co., Ltd.) to express shRNA via the H1 promoter. Synthetic oligonucleotide primers [forward (5′-GATCCGCTGGCAAATATGAAGGTATTCAAGAGATACCTTCATATTTGCCAGCTTTTTTG-3′) and reverse (5′AATTCAAAAAAGCTGGCAAATATGAAGGTATCTCTTGAATACCTTCATATTTGCCAGCG-3′)], and the shRNA sequence expressed by the vector were sheared to form a target sequence 5′-GCTGGCAAATATGAAGGTA-3′. A fragment targeting linc00467 (5′-GCTGGCAAATATGAAGGTA-3′) inserted into a lentiviral vector was used as linc00467-shRNA, and a non-targeting fragment (5′-TTCTCCGAACGTGTCACGT-3′) inserted into a lentiviral vector was used as control-shRNA. The shRNA lentiviruses were transfected into cells in the presence of 2 µg/ml Polybrene, and linc00467-shRNA-transfected cells were screened with 1, 2, 4 or 8 µg/ml puromycin for 8 days post-transfection. siRNA was used in the preliminary functional experiments, whereas the experiments investigating the mechanisms of action required longer culture times. Therefore, shRNA was used in the later mechanism studies. The primer sequences are listed in Table SII.

Cytoplasmic and nuclear RNA were separated and purified using a PARIS kit (Thermo Fisher Scientific, Inc.), following the manufacturers' protocol. The buffer was pre-cooled using ice and added to the cultured cells; lysate was then collected in an enzyme-free Eppendorf (EP) tube, and cytoplasmic lysate was collected by centrifugation at 10,000 × g for 1 min at room temperature into an enzyme-free EP tube. Cell disruption buffer was then added, and the nuclear lysate was collected by centrifugation at 10,000 × g for 1 min at room temperature into an enzyme-free EP tube. An equal volume of 2X lysis/binding solution (Invitrogen; Thermo Fisher Scientific, Inc.) and absolute ethanol were added into each tube, following which the mixture was transferred into an enzyme-free EP tube and centrifuged at 10,000 × g for 1 min at room temperature. The liquid flow-through was then discarded, 3 wash steps were performed and 95–100°C Elution Solution (Qiagen GmbH). was added to obtain RNA. GAPDH and U6 were used as internal controls, and RT-qPCR was used to measure the expression of linc00467 in the nucleus and cytoplasm.

An MTT assay was used to measure cell proliferation using a Cell Proliferation Reagent kit I (Roche Diagnostics). After 48 h of transfection, the cells were plated on 96-well plates at a density of 3,000 cells per well. The 96-well culture plates were taken at 0, 24, 48, 72 and 96 h, respectively, and proliferation was detected at the corresponding time points. MTT (20 µl) was added into each well, and the cells were incubated for 4 h in a humidified atmosphere containing 5% CO₂ at 37°C. The reaction was then terminated by removal of the supernatant, and 150 µl dimethyl sulfoxide (DMSO) solution was added to dissolve the purple formazan crystals. The plate was shaken for 20 min to promote the dissolution of the crystals. The absorbance was measured at a wavelength of 490 nm using a microplate reader, and cell proliferation was analysed based on the absorbance value. The clonogenic ability of the LAD cells was examined using the colony formation method. The transfected cells were digested with 0.25% trypsin and thoroughly pipetted into a single-cell suspension. Following cell counting, the cells were densely seeded (1,000 cells/well) in a 6-well culture plate, evenly dispersed and plated at 37°C. After 4 days, the supernatant was discarded and replaced with new complete medium. When the colonies were visible to the naked eye, the culture was terminated. The supernatant was aspirated, the plates were carefully rinsed with PBS, and cells were fixed with 4% paraformaldehyde for 15 min at room temperature. Subsequent to removal of the fixative, 1 ml 0.1% crystal violet stain solution was added into each well for 15 min at room temperature, following which the stain was discarded. The remaining stain was washed off with PBS, and the plates were dried at room temperature. The number of colonies >10 cells was counted using an optical light microscope at ×100 magnification (Olympus, Tokyo, Japan), and statistical analyses were performed.

The cells were densely seeded (2×10⁵ cells/well) in a 6-well culture plate. H1299, A549 and PC9 cells transfected with si-NC/si-linc00467 or empty vector/pcDNA3.1-linc00467 were collected 48 h after transfection. The cells were diluted with the cell binding buffer (BD Pharmingen; BD Biosciences) and stained with Annexin V-allophycocyanin (APC) and 7 amino-actinomycin (7-AAD) in the dark at room temperature for 15 min, prior to being analysed with a flow cytometer (FACScan^®; BD Biosciences). A tube containing only Annexin V-APC and 7-AAD was used as the negative control. The data were analysed using CellQuest software (version 6.0 (BD Biosciences).

Transwell chambers (EMD Millipore) were used for cell migration and invasion assays. For the migration assay, 3×10⁴ cells in 300 µl serum-free medium were seeded into the top chamber of each insert, and 700 µl medium supplemented with 10% FBS was added into the lower chamber. For the invasion assay, the upper chamber membrane was first uniformly coated with Matrigel (BD Biosciences) at 37°C for 24 h. Then, 1×10⁵ cells in 300 µl serum-free medium were placed into the top chamber of each insert, and 700 µl medium supplemented with 10% FBS was added into the lower chamber. Following incubation of the A549 and H1299 cells at 37°C for 48 h, and the PC9 cells were incubated at 37°C for 24 h, the cells in the top chamber were removed, and cells that had migrated or invaded through the membrane were stained for 15 min with 0.1% crystal violet at 37°C. The stained cells were counted under a light microscope (magnification, ×100; Olympus Corporation).

The probability of an interaction between linc00467 and RBPs was determined using an lncRNA prediction website (http://annolnc.cbi.pku.edu.cn/index.jsp). The sequence of linc00467 was entered into the AnnoLnc database and into the protein interaction page; from this, the interaction score between lncRNA and protein could be predicted.

Cells were collected, washed twice with ice-cold PBS and lysed in 100 µl RIPA buffer (Servicebio) supplemented with protease inhibitor (Sigma-Aldrich; Merck KGaA) for 30 min, and a bicinchoninic protein assay kit (CWBIO Corporation) was used to detect the protein concentration. Total protein samples (20 µg) were electrophoresed on 10% SDS-PAGE and transferred onto polyvinylidene fluoride membranes (EMD Millipore). Following blocking with 5% milk for 1 h at room temperature, the membranes were incubated overnight at 4°C with primary antibody EZH2 (1:1,000; ab191250; Abcam) and HTRA3 (1:1,000; ab227463; Abcam); GAPDH was used as the loading control (1:3,000; ab181602; Abcam). The blots were extensively washed five times with TBST (0.5% Tween-20) and incubated with horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (1:3,000; ab150081; Abcam) for 1 h at room temperature. Following washing, an ECL chromogenic substrate and densitometry using Quantity One software (version 4.6.9, Bio-Rad Laboratories, Inc.) were used to measure the blots.

The RNA immunoprecipitation (RIP) assay was performed using a Magna RIP RNA-Binding Protein Immunoprecipitation kit (17–701; EMD Millipore) in accordance with the manufacturer's protocol. Whole-cell extract from A549 and H1299 cells was used for immunoprecipitations by using RIP lysis buffer (CS203176; EMD Millipore). For each immunoprecipitation reaction, 50 µl protein A Sepharose beads and 5 µg antibody against histone-lysine N-methyltransferase EZH2 (ab191250; Abcam) were used. Immunoprecipitation was performed using a relevant antibody to the DNase-treated extract and incubating at 4°C overnight. Then samples were sequentially treated with Proteinase K and RQ1 RNase-free DNase to digest protein and genomic DNA, respectively. Purified RNA samples were re-suspended in water and stored at 80°C. Finally, the purified RNA was used for RT-qPCR.

A549 and H1299 cells were cross-linked with 1% formaldehyde for 10 min at room temperature. Crosslinking was halted by addition of 125 mM glycine. Pellets were washed twice with complete proteinase inhibitor cocktail (Roche Diagnostics) in PBS, and then resuspended in 400 µl SDS lysis buffer (Sigma-Aldrich; Merck KGaA) for 30 min. Then, 2% of supernatant was saved as input and 100 µl of supernatant was incubated with 5 µg EZH2 (ab191250; Abcam) or IgG (ab172730; Abcam) antibodies for 16 hours at 4°C. Then 30 µl of protein G magnetic beads (Thermo Fisher Scientific, Inc.) was added to each IP reaction and incubated for 2 h at 4°C with rotation. Immunoprecipitates were then processed with a series of washes and the cross-linking reversed. The DNA was purified by phenol/chloroform/isoamyl extraction and RT-qPCR was performed as aforementioned. The sequences of the ChIP primers for the HTRA3 promoter region are listed in Table SI.

All of the statistical analyses were performed using SPSS 18.0 software (SPSS, Inc.). P<0.05 was considered to indicate a statistically significant difference, and P<0.01 was considered to indicate a highly statistically significant difference. Data are presented as mean ± standard deviation. A two-tailed Student's t-test was applied to compare differences between two groups. The comparison of multiple groups were performed using one-way analysis of variance, and the post-hoc multiple comparisons were performed using the Least Significant Difference test.

A total of 4 microarray datasets (GSE19804, GSE19188, GSE30219 and GSE27262) obtained from the GEO were used to analyse linc00467 expression in lung cancer tissues and normal tissues. It was identified that linc00467 was upregulated in the 4 data sets (Fig. 1A-D). In addition, TCGA data also indicated that linc00467 expression was upregulated in lung cancer tissues compared with normal tissues (Fig. 1E).

To validate the GEO and TCGA results, RT-qPCR assays were used to measure linc00467 expression in 35 paired LAD and adjacent normal lung tissues. The relative expression of lncRNA was calculated using the Cq value (2^−ΔCq) method. The expression level of linc00467 in LAD tissues was increased compared with that in adjacent normal lung tissues (P<0.001; Fig. 2A). Then, the association between linc00467 expression and patients' clinical characteristics were evaluated (Table I). The linc00467 expression levels were significantly increased in patients with larger tumours compared with smaller tumours (P=0.013; Fig. 2B), and in those with more advanced TNM stages (P=0.021; Fig. 2C). However, there were no significant associations between linc00467 expression and other clinical parameters, including age, sex, differentiation, lymph node metastasis and smoking history.

To investigate the function of linc00467 in LAD cells, its expression levels in the human bronchial epithelial 16HBE cell line and in 3 human LAD A549, PC9 and H1299 cell lines were assessed using RT-qPCR. A549 and H1299 cells expressed increased levels of linc00467 compared with 16HBE cells, while the PC9 cells expressed relatively decreased linc00467 levels compared with A549 and H1299 cells (Fig. 3A). Therefore, linc00467 was knocked down with siRNA in the A549 and H1299 cell lines and overexpressed in PC9 cell lines. A localization assay for linc00467 in cells indicated that linc00467 was localized in the nucleus and cytosol but was localized to a greater extent in the nucleus (Fig. 3B-D). The results of the linc00467 knockdown assay in the A549 and H1299 cells by transfection with siRNAs are presented in Fig. 3E and F; ‘si-linc00467 1#’ exhibited the best knockdown efficiency, so it was used for subsequent functional experiments. linc00467 expression was upregulated in PC9 cells by transfection with pcDNA3.1-linc00467 (Fig. 3G).

The role of linc00467 in LAD development was next examined. linc00467 was knocked down in A549 and H1299 by transfection with siRNAs and overexpressed in PC9 with pcDNA3.1-linc00467. The MTT assays indicated that A549 and H1299 cell viability was decreased following knockdown of linc00467 (Fig. 4A and B), while cell viability was increased in pcDNA3.1-linc00467-treated PC9 cells (Fig. 4C). The colony formation ability of A549 and H1299 cells was consistently and significantly impaired following linc00467knockdown, while linc00467 overexpression increased PC9 cell colony formation ability (Fig. 4D-F). Overall, linc00467 served as an oncogene to promote LAD cell proliferation. Flow cytometric analyses indicated that linc00467 knockdown significantly increased the levels of apoptosis in A549 and H1299 cells (Fig. 5A and B). By contrast, linc00467 overexpression decreased the levels of apoptosis in PC9 cells (Fig. 5C). The results confirmed that linc00467 promoted LAD cell proliferation by affecting LAD cell apoptosis.

As linc00467 was confirmed to promote LAD cell proliferation in vitro, its effects on LAD cell migration and invasion was additionally investigated. A Transwell assay was used to evaluate the biological effect of linc00467 on the LAD cell metastasis. linc00467 knockdown significantly suppressed the migratory and invasive abilities of A549 and H1299 cells (Fig. 6A and B), while linc00467 overexpression increased the migratory and invasive abilities of PC9 cells (Fig. 6C). These results suggested that linc00467 promoted the migration and invasion of LAD cells.

linc00467 was knocked down in H1299 cells by transfection with shRNA lentiviruses. The knockdown efficiency of shRNA was verified by RT-qPCR (Fig. 7A and B). RNA transcriptome sequencing was used to identify genes that were differentially expressed between linc00467-knockdown H1299 and control cells. A total of 625 differentially expressed transcripts were identified using database analysis (471 upregulated and 154 downregulated transcripts, |log2 (FC)| >1 and FDR<0.05) (Fig. 8A and B). RT-qPCR was used to verify the expression of 6 randomly selected genes expression in A549 and H1299 cells, and the results were consistent with the RNA transcriptome sequencing results (Fig. 8C and D).

RNA transcriptome sequencing indicated that HTRA3 was upregulated to a greater extent in linc00467-depleted H1299 cells compared with the control cells. It was then additionally verified that HTRA3 exhibited higher mRNA expression in linc00467-knockdown A549 cells and H1299 cells compared with the control cells using RT-qPCR, and protein levels were correspondingly elevated (Fig. 9A-D).

lncRNAs are able to bind RNA-binding proteins (RBPs) to regulate the expression of downstream genes. Therefore, the present study investigated whether linc00467 regulated HTRA3 expression by binding to RBPs. The probability of an interaction between linc00467 and RBPs was first determined using an lncRNA prediction website (http://annolnc.cbi.pku.edu.cn/index.jsp). The sequence of linc00467 was input into the AnnoLnc database, and entered into the protein interaction page; subsequently, the interaction score between lncRNA and protein could be predicted. The interaction score for linc00467 and EZH2 was 96.0705, close to the maximum value of 100. Whether linc00467 repressed HTRA3 transcription was next investigated by recruiting EZH2 to the HTRA3 promoter. EZH2 expression was first knocked down in A549 and H1299 cells using transfection with siRNA (Fig. 10A and B). In addition, western blot analysis was performed to measure HTRA3 protein expression levels when EZH2 was downregulated. The results indicated that HTRA3 protein expression levels were increased in the EZH2-downregulated group compared with in the control group (Fig. 10C and D). Then, an RIP assay was performed to directly address whether linc00467 and EZH2 are binding partners. The RIP assay suggested that linc00467 was able to directly bind EZH2 in A549 and H1299 cells (Fig. 10E and 10F). In addition, a ChIP assay was used to verify that EZH2 bound to the promoter of HTRA3 (Fig. 10G). The results demonstrated that linc00467 repressed HTRA3 expression by binding with EZH2.

Rescue experiments were used to examine the effect of inhibiting HTRA3 expression in linc00467-knockdown A549 and H1299 cells. The rescue experiments demonstrated that cell proliferation, migration and invasion were inhibited in linc00467-knockdown A549 and H1299 cells, whereas knockdown of HTRA3 partially reversed all of these effects (Fig. 11A-D).

The expression level of HTRA3 in LAD tissues was decreased compared with the adjacent normal lung tissues (P<0.001; Fig. 12A). The association between HTRA3 expression and LAD clinical characteristics was additionally evaluated (Table II). Decreased HTRA3 expression was associated with larger tumour sizes (P=0.018; Fig. 12B), increased lymph node metastasis (P=0.031; Fig. 12C) and advanced TNM stages (P=0.012; Fig. 12D).