The present study was approved by the Ethics Committee of The First People's Hospital of Huzhou, Huzhou, China (approval no. 2019012), and informed consent was obtained from all patients.

A total of 240 NSCLC and matched para-carcinoma tissues (>5 cm away from the tumor) were obtained from patients who underwent surgical resections at the First People's Hospital of Huzhou (Huzhou, China) between January 2015 and December 2018. The NSCLC patients consisted of 113 females and 127 males (age range, 27–82 years; mean age, 63). Among them, an NSCLC tissue microarray (TMA), containing 189 pairs of tissues that were formalin-fixed paraffin-embedded (FFPE), was used for immunohistochemical (IHC) analysis, and 51 pairs of fresh tissues were used for reverse transcription-quantitative (RT-q)PCR and western blot analysis. These patients had not received chemotherapy or radiation therapy prior to surgical resection. According to the 8th edition of Union for International Cancer Control (UICC) (26,27), the pathological diagnosis was confirmed and classified by two certified pathologists (Professors Qilin Shi and Hui Xia, Department of Pathology, The First People's Hospital of Huzhou). All patients and clinical information were collected retrospectively and analyzed in this study, including sex, age, smoking history, tumor size, histological type, tumor differentiation, lymph node metastasis, cancer thrombus and TNM stage (Table I). Informed consent was obtained from all patients.

Kaplan-Meier Plotter (http://kmplot.com/analysis/) is a public web server used for the validation of the gene prognostic power using expression data from independent datasets (28). All the related data were downloaded from the Gene Expression Omnibus (GEO) database and The Cancer Genome Atlas (TCGA) (28,29). A total of 1,233 samples of nine independent datasets were integrated, including 1,100 samples in GEO datasets [GSE4573 (30), GSE14814 (31), GSE8894 (32), GSE19188 (33), GSE3141 (34), GSE31210 (35), GSE29013 (36) and GSE37745 (37)] and 133 samples in TCGA datasets (38). TMEM229A expression levels in patients with lung adenocarcinoma (n=601) and patients with squamous cell lung carcinoma (n=632) were collected using the Kaplan-Meier Plotter online bioinformatics datasets. Additionally, patients with lung adenocarcinoma and squamous cell carcinoma were divided into low or high TMEM229A expression groups, according to TMEM229A mRNA expression based on the median value. The overall survival was analyzed using the Kaplan-Meier method.

The normal human lung epithelial cell (BEAS-2B) and seven lung cancer cell lines (A549, H23, H226, 95D, H1975, PC-9 and H1299) were purchased from the American Type Culture Collection. Cells were cultured in DMEM (cat. no. SH30243.01; HyClone; Cytiva) or RPMI-1640 medium (cat. no. SH30809.01; HyClone; Cytiva) supplemented with 10% FBS (Gibco; Thermo Fisher Scientific, Inc.), 100 IU/ml penicillin and 100 µg/ml streptomycin (Sigma-Aldrich; Merck KGaA) in a humidified incubator under 5% CO₂ at 37°C. To inhibit the ERK signaling pathway, cells were treated with 100% dimethylsulfoxide (DMSO) as a negative control or PD98059 (ERK inhibitor; 20 µM; Beyotime Institute of Biotechnology) for 2 h prior to plasmid transfection in a humidified incubator under 5% CO₂ at 37°C.

To construct the full-length human TMEM229A proteins, the corresponding TMEM229A cDNA was cloned into a pcDNA3.1/myc-his vector (cat. no. V800-20; Invitrogen; Thermo Fisher Scientific, Inc.). In total, two specific small interfering (si)RNAs of TMEM229A (20 µM) and scrambled siRNA (referred to as si-Ctrl; 20 µM) were designed and synthesized by Guangzhou RiboBio Co., Ltd.. Once cells reached 60–70% confluence, cell transfections of different plasmids [pcDNA3.1/myc-his (referred to as vector), pcDNA3.1-TMEM229A/myc-his (referred to as overexpressed (OE)-TMEM229A), si-Ctrl, si-TMEM229A] were carried out using Lipofectamine^® 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions. The medium was replaced with complete medium and transfection was performed in a humidified incubator under 5% CO₂ at 37°C for 6 h. In a 6-well plate, 2 µg vector or OE-TMEM229A and 50 nM si-Ctrl or si-TMEM229A plasmids were added to each well. At 24 h after transfection, cells were harvested for the subsequent experiments. The siRNA sequences used in the experiment were as follows: si-TMEM229A#1 forward, 5′-GGAUGCGCCUCUACUUCUAdTdT-3′ and reverse, 5′-UAGAAGUAGAGGCGCAUCCdTdT-3′; si-TMEM229A#2 forward, 5′-CCUUCGUCUUCAAUUUCCUdTdT-3′ and reverse, 5′-AGGAAAUUGAAGACGAAGGdTdT-3′; and si-Ctrl forward, 5′-UUCUCCGAACGUGUCACGUTT-3′ and reverse, 5′-ACGUGACACGUUCGGAGAATT-3′.

The RTCA xCELLLigence system (ACEA Biosciences, Inc.; Agilent Technologies, Inc.) was used to uninterruptedly monitor cell morphology, proliferation and migration in vitro in a non-invasive manner, as previously described (39,40). A cell index was used to indicate the cell number and cell adhesion. As cells adhered to the surface of an E-plate and influenced the electrical impedance across the array, the xCELLLigence system provided an electronic record and converted it into the cell index. Cell proliferation and migration assays were performed according to the manufacturer's instructions (ACEA Biosciences, Inc.; Agilent Technologies, Inc.).

Briefly, for cell proliferation assays, 50 µl culture medium was added to measure the background, and then 6×10³ A549 cells, 1×10⁴ H23 cells or 1×10⁴ H1299 cells transfected with different plasmids were mixed with 100 µl culture medium and seeded into the E-plate. For cell migration and invasion assays, 165 µl culture medium was added to the lower chamber and 30 µl serum-free culture medium was added to the upper chamber, and then the E-plate was left to stand for 1 h in a humidified incubator under 5% CO₂ at 37°C. The cells were mixed with 100 µl serum-free culture medium and seeded into the E-plate. The data were recorded using xCELLLigence software 2.0 (ACEA Biosciences, Inc.; Agilent Technologies, Inc.) and analyzed using GraphPad Prism 5.0 (GraphPad Software, Inc.).

Cell proliferation was also assessed using a Cell Counting Kit-8 (CCK-8; Beyotime Institute of Biotechnology) assay, according to the manufacturer's instructions. Briefly, 6×10³ A549, 1×10⁴ H23 or 1×10⁴ H1299 cells transfected with different plasmids were seeded into 96-well plates and cultured for 24, 48 and 72 h at 37°C, followed by incubation with 10 µl of CCK-8 reagent for 1 h. The absorbance values were measured using a Spectra Max 190 reader (Molecular Devices, LLC) at 450 nm.

The transfected A549, H23 or H1299 cells were seeded into 6-well plates at a density of 1×10³ cells/well and cultured with culture medium containing 10% FBS (Gibco; Thermo Fisher Scientific, Inc.) for 10 days. Cells were fixed in 4% paraformaldehyde for 30 min at room temperature and followed by staining with 1% crystal violet stain solution (Beyotime Institute of Biotechnology) overnight at room temperature. Cell colonies were imaged under an inverted fluorescence microscope with a magnification of ×20 magnification (ZEISS Axio Vert.A1; Carl Zeiss AG) and counted if there were >50 cells in the assay.

The cell proliferative ability was also examined using a soft agar assay, as previously described (41,42). At 24 h-post-transfection, A549, H23 or H1299 cells were plated in 24-well plates, with the bottom layer containing 0.7% low-melting point agarose (cat. no. 16520100; Invitrogen; Thermo Fisher Scientific, Inc.). Cells (1,000-2,000 per well) were seeded into the medium with 0.35% agarose. After 14 days of incubation in a humidified incubator under 5% CO₂ at 37°C, cell colonies were imaged and counted if there were >50 cells in the assay under an inverted fluorescence microscope with a magnification of ×20 (ZEISS Axio Vert.A1; Carl Zeiss AG).

Cell migratory and invasive abilities were determined using a Transwell plate chamber (8 µm; Corning; Corning, Inc.), according to the manufacturer's protocols. For the migration assay, the transfected cells (5×10⁴) were mixed in 100 µl serum-free culture medium and seeded into the upper chamber, and 600 µl complete culture medium was added to the lower chamber. For the invasion assay, the upper chamber was firstly coated with Matrigel^® Matrix Basement Membrane (cat. no. 356234; BD Biosciences) for 4 h in a humidified incubator under 5% CO₂ at 37°C. Then, cells were seeded into the upper chamber, and complete medium was added to the lower chamber. Following 48 h incubation in a humidified incubator under 5% CO₂ at 37°C, the cells in the lower chamber were fixed in 4% paraformaldehyde for 30 min at room temperature and stained with 1% crystal violet stain solution (Beyotime Institute of Biotechnology) overnight at room temperature. The images were captured and the cells were decolorized with 30% acetic acid. The absorbance values were measured using a Spectra Max 190 reader (Molecular Devices, LLC) at 570 nm.

Total RNA from tissues and cells was extracted using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.), as previously described (43). Briefly, samples were treated with 20% chloroform, followed by incubation at room temperature for 5 min. Then, samples were centrifuged and the aqueous phase was transferred to a new tube, followed by the addition of an equal amount of isopropanol and incubation at room temperature for 10 min. The pellets were dissolved in nuclease-free water. cDNA was reverse transcribed using the PrimeScript RT reagent kit (Takara Biotechnology Co., Ltd.), according to the manufacturer's protocol. RT-qPCR was performed using UltraSYBR Green PCR Master mix (cat. no. CW0957H; CWBio) on an ABI 7500 system (Applied Biosystems; Thermo Fisher Scientific, Inc.). The thermocycling conditions used for the qPCR were as follows: Initial denaturation at 95°C for 10 min, followed by 40 cycles at 95°C for 15 sec and 60°C for 1 min. Gene expression was normalized to 18S ribosomal RNA (18sRNA), and the relative expression level was calculated using the 2^−ΔΔCq method (44). The primer sequences used in the experiment were as follows: TMEM229A forward, 5′-CGACCTACCCCGCTTTCTTTT-3′ and reverse, 5′-GCTCCCACCGTAGATGAAGAT-3′; N-cadherin forward, 5′-AGCCAACCTTAACTGAGGAGT-3′ and reverse, 5′-GGCAAGTTGATTGGAGGGATG-3′; E-cadherin forward, 5′-ATTTTTCCCTCGACACCCGAT-3′ and reverse, 5′-TCCCAGGCGTAGACCAAGA-3′; MMP2 forward, 5′-GATACCCCTTTGACGGTAAGGA-3′ and reverse, 5′-CCTTCTCCCAAGGTCCATAGC-3′; Snail1 forward, 5′-ACTGCAACAAGGAATACCTCAG-3′ and reverse, 5′-GCACTGGTACTTCTTGACATCTG-3′; and 18sRNA forward, 5′-GTAACCCGTTGAACCCCATT-3′ and reverse, 5′-CCATCCAATCGGTAGTAGCG-3′.

Tissues or cells were lysed in RIPA buffer (cat. no. P0013B; Beyotime Institute of Biotechnology) containing protease and phosphatase inhibitors (Beyotime Institute of Biotechnology). The total protein concentration was quantified with a BCA assay kit (Beyotime Institute of Biotechnology). An equal amount of total protein (30 µg) was loaded and separated using 8–12% SDS-PAGE. The proteins were transferred onto 0.45-µm PVDF membranes (EMD Millipore). The membranes were blocked with 5% non-fat milk at room temperature for 1 h, followed by incubation with primary antibodies overnight at 4°C. The membranes were subsequently washed with PBS (cat. no. SH30256.01; HyClone; Cytiva) containing 0.1% Tween-20, and incubated with the corresponding HRP-conjugated secondary antibodies [(goat-anti mouse IgG (H+L) antibody; 1:5,000; cat. no. A0216) or (goat-anti rabbit IgG (H+L) antibody; 1:5,000; cat. no. A0208) both from Beyotime Institute of Biotechnology] at room temperature for 1 h. The images were visualized using an ECL reagent (cat. no. P0018S; Beyotime Institute of Biotechnology) and measured using a Tanon 5200 system (Tanon Science & Technology Co., Ltd.). The relative expression level of proteins was normalized to β-actin using ImageJ vl.6.0 software (National Institutes of Health).

The primary antibodies used in the experiment were as follows: Anti-TMEM229A (1:500; cat. no. ab107780; Abcam); anti-E-cadherin (1:1,000; cat. no. 20874-1-AP; ProteinTech Group, Inc.); anti-N-cadherin (1:1,000; cat. no. 22018-1-AP; ProteinTech Group, Inc.); anti-MMP2 (1:1,000; cat. no. 10373-2-AP; ProteinTech Group, Inc.); anti-Snail1 (1:1,000; cat. no. 13099-1-AP; ProteinTech Group, Inc.); anti-ERK1/2 (1:2,000; cat. no. 4695; Cell Signaling Technology, Inc.); anti-phosphorylated (p)-ERK1/2 (1:2,000; cat. no. 4370; Cell Signaling Technology, Inc.); anti-AKT (1:2,000; cat. no. 4691; Cell Signaling Technology, Inc.); anti-p-AKT (1:2,000; Ser473; cat. no. 4060; Cell Signaling Technology, Inc.); anti-p-p38 (1:2,000; Thr180/Tyr182; cat. no. 4631; Cell Signaling Technology, Inc.); anti-p-JNK (1:2,000; Thr183/Tyr185; cat. no. 4668; Cell Signaling Technology, Inc.); anti-p-p65 (1:1,000; Ser536; cat. no. 3033; Cell Signaling Technology, Inc.), and anti-β-actin (1:5,000; cat. no. M1210-2; Hangzhou HuaAn Biotechnology Co., Ltd.).

The tissues were fixed with 10% formalin overnight at room temperature, and the FFPE tissues were sliced into 5-µm thickness sections, followed by dewaxing, hydrating and inactivating of endogenous peroxidase. Sections were then incubated with a polyclonal rabbit anti-TMEM229A antibody (1:500; cat. no. PA5-63294; Invitrogen; Thermo Fisher Scientific, Inc.) overnight at 4°C in a humidified chamber. The sections were subsequently washed with PBS (cat. no. SH30256.01; HyClone; Cytiva), and incubated with the SP Rabbit & Mouse HRP-conjugated secondary antibody kit (cat. no. CW2069S; CWBio) at room temperature for 1 h. Next, the sections were incubated with DAB (included in the aforementioned kit) at room temperature for 20 min. After washing with water, the sections were counterstained with hematoxylin staining solution (cat. no. C0107; Beyotime Institute of Biotechnology) at room temperature for 5 min. The TMA sections were further imaged and digitally scanned at a magnification of ×400 using a KF-PRO-005 platform (Ningbo Jiangfeng Bio-information Technology Co., Ltd.) into whole slide digital images. The scoring of the slides was conducted using HALO Multiplex IHC analysis software version v3.1.1076.308 (Indica Labs). The intensity of TMEM229A was scored as 1 (absent or weak), 2 (moderate) or 3 (high). The percentage of positive cells was assigned as 1 (<25%), 2 (25–50%) and 3 (>50%). The score of each slice was multiplied to acquire a final score of 1–9. A composite score of <3 was defined as TMEM229A low expression, and a score of ≥3 was defined as TMEM229A high expression.

All data are presented as the mean ± SEM of three independent experiments. The associations between TMEM229A expression and clinicopathological characteristics of patients were analyzed using the χ² test, Fisher's exact test and Pearson rank correlation analysis. Kaplan-Meier survival analysis was performed using the log-rank test. Statistical significance between the groups was determined using SPSS 21.0 software (IBM Corp.) with an unpaired Student's t-test or one-way ANOVA followed by a Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

To determine the expression pattern of TMEM229A in NSCLC, the mRNA expression level of TMEM229A was measured in 51 paired NSCLC and corresponding healthy tissues via RT-qPCR. Downregulation of TMEM229A was observed in 40 pairs of tissues, accounting for 78% of the total specimens examined (Fig. 1A). Consistent with the mRNA expression pattern, a reduced protein expression level of TMEM229A was identified in NSCLC tissues compared with the matched adjacent healthy tissues in 10/12 paired samples, as determined using western blotting (Fig. 1B). The low expression of TMEM229A protein in NSCLC tissues was further confirmed via IHC staining analysis (Fig. 1C). Moreover, TMEM229A expression was measured in BEAS-2B cells and seven NSCLC cell lines, including A549, H23, H1299, H226, 95D, PC-9 and H1975. The mRNA and protein expression levels of TMEM229A were decreased in several NSCLC cell lines (A549, H23, 95D, H226 and H1975) compared with BEAS-2B cells (Fig. 1D). Owing to the higher invasiveness of 95D, A549 and H23 were selected for the follow-up overexpressing experiments. Collectively, these data indicated that TMEM229A expression was downregulated in NSCLC.

To evaluate the clinical significance of TMEM229A in NSCLC, a NSCLC TMA containing 189 specimens was stained with TMEM229A antibody and scored in a standard manner, as previously described (45). The results indicated that 113 patients were classified into the low TMEM229A expression group, while 76 patients were classified into the high TMEM229A expression group (Fig. 1C and Table I). Further analysis of TMEM229A expression and the clinicopathologic characteristics revealed that low TMEM229A expression was associated with tumor differentiation (P=0.043), lymph node metastasis (P=0.008), cancer thrombus (P=0.012) and TNM stage (P=0.04). However, there were no significant associations with sex, age, smoking history, tumor size and histological type (Table I).

Subsequently, Kaplan-Meier Plotter was used to study the association between TMEM229A expression level and overall survival in patients with NSCLC. All the related data were downloaded from the GEO database, the EGA and TCGA. Patients with lung adenocarcinoma and those with squamous cell carcinoma were divided into low- or high-TMEM229A expression groups, according to TMEM229A mRNA expression. The results demonstrated that lung adenocarcinoma cases and squamous cell lung carcinoma cases with low TMEM229A expression had a poor prognosis with a shorter overall survival (P=0.0023 and P=0.0096, respectively; Fig. 1E and F). Thus, these data indicated that low TMEM229A expression was associated with poor prognosis in patients with lung adenocarcinoma and in patients with squamous cell lung carcinoma.

To investigate the effect of TMEM229A on cell proliferation, TMEM229A was overexpressed in A549 and H23 cells (Fig. 2A and B), which had relatively low expression levels of TMEM229A. Subsequently, the proliferative capacity of TMEM229A-overexpressing cells was measured using CCK-8 and RTCA assays. The results demonstrated that overexpression of TMEM229A significantly inhibited A549 and H23 cell proliferation in a time-dependent manner (Fig. 2C-F).

In addition, two TMEM229A siRNAs and si-Ctrl were transfected into H1299 cells, which demonstrated relatively high expression levels of TMEM229A. The RT-qPCR and western blot results revealed that TMEM229A expression was markedly decreased in the si-TMEM229A-transfected H1299 cells compared with si-Ctrl-transfected cells (Fig. S1A and B), and TMEM229A knockdown significantly promoted H1299 cell proliferation (Fig. S1C and D). Furthermore, the effects of TMEM229A overexpression or knockdown on cell proliferation were verified using cell colony and soft agar assays. The data indicated that overexpression of TMEM229A suppressed the number of cell colonies, while TMEM229A knockdown promoted these (Fig. 3A-D). Overall, these findings indicated that TMEM229A plays a crucial role in the proliferation of NSCLC cells.

The potential influence of TMEM229A overexpression or knockdown on cell migration and invasion was then examined. It was revealed that overexpression of TMEM229A significantly suppressed the migration and invasion of NSCLC cells, as determined using Transwell and RTCA assays (Fig. 2G-L), whereas knockdown of TMEM229A promoted the migratory and invasive abilities of H1299 cells compared with those in the control group (Fig. S1E-G).

Considering the critical role of EMT in the metastasis of NSCLC (46), the present study examined the role of TMEM229A on EMT, and detected the expression levels of E-cadherin, N-cadherin, MMP2 and Snail1. As demonstrated in Fig. 4A and B, TMEM229A overexpression significantly increased E-cadherin expression and reduced N-cadherin, Snail1 and MMP2 expression, while TMEM229A knockdown exerted the opposite effects. These data indicated that TMEM229A influenced cell migration and invasion by modulating the EMT phenotype.

To investigate the mechanisms via which TMEM229A promotes NSCLC progression, key factors of common signaling pathways were screened in NSCLC cells, including p38/MAPK, ERK, AKT, JNK and NF-κB. The expression levels of p-p38, p-ERK, p-AKT (Ser473), p-JNK and p-p65 were primarily analyzed in the present study. It was revealed that knockdown of TMEM229A upregulated p-ERK and p-AKT (Ser473) expression in NSCLC cells, while overexpression of TMEM229A exerted the opposite effect (Fig. 4B-F). Moreover, other signaling pathways showed no obvious changes (Fig. S2).

Subsequently, si-TMEM229A-transfected cells were treated with the ERK inhibitor PD98059, and the results demonstrated that PD98059 suppressed the TMEM229A-induced ERK phosphorylation, and partly reversed the expression levels of E-cadherin, N-cadherin, MMP2 and Snail1 (Fig. 5A). Moreover, the Transwell assay results indicated that the promoting effect on cell invasiveness caused by TMEM229A knockdown was also partially reversed by PD98059 (Fig. 5B).