Abstract

Objective

   This article is dedicated to finding important genes related to the
   prognosis of lung adenocarcinoma (LUAD), looking for a new gene that
   may affect tumor radiosensitivity, and conducting basic experiments to
   verify the relationship between this gene and the radiosensitivity of
   LUAD.

Methods

   The gene expression profiles [31]GSE32863, [32]GSE33532, and
   [33]GSE43458 were obtained from NCBI-GEO. GEO2R and a Venn diagram were
   used to identify upregulated genes. STRING and Cytoscape were applied
   to develop a protein–protein interaction network (PPI) and analyze the
   modules. The Database for Annotation, Visualization and Integrated
   Discovery (DAVID) was used to process the GO and KEGG pathway analysis.
   The Kaplan Meier plotter and Gene Expression Profiling Interactive
   Analysis (GEPIA) were applied to get the significant prognostic
   information and differential expression between LUAD tissues and normal
   lung tissues. Western blotting and Q-PCR were used to detect the
   expression of PKMYT1 in tissues. Small interfering RNAs (siRNAs) were
   used to knockdown PKMYT1. The colony survival experiment was used to
   assess the effect of PMYT1 on the radiosensitivity of tumor cells. Cell
   cycle analysis was used to assess cell cycle distribution.

Results

   We identified 14 genes (PKMYT1, TTK, CHEK1, CDC20, PTTG1, MCM2, CDC25C,
   MCM4, CCNB1, CDC45, MAD2L1, CCNB2, BUB1, and CCNA2) that are important
   for LUAD and may be potential therapeutic targets. We confirmed that
   PKMYT1 is highly expressed in LUAD and firstly demonstrated that
   artificially silencing the expression of PKMYT1 can abrogate IR-induced
   G2/M phase arrest and increase the sensitivity of cancer cells to
   radiation.

Conclusion

   In summary, we obtained 14 core genes related to the poor prognosis of
   LUAD via bioinformatical analysis. We identified that PKMYT1 was
   significantly upregulated in LUAD tissues and firstly demonstrated that
   knockdown of PKMYT1 can eliminate the radiation-induced G2/M arrest,
   resulting in a lower survival rate for cells receiving radiation
   therapy. Our findings suggested that PKMYT1 is a promising target to
   improve the radiosensitivity of LUAD.

   Keywords: bioinformatics, PKMYT1, G2 checkpoint, lung adenocarcinoma,
   radioresistance

Introduction

   Lung cancer accounts for 18.4% of cancer deaths and remains the leading
   cause of cancer-related deaths ([34]Bray et al., 2018). Lung
   adenocarcinoma has become one of the most common types of lung cancer
   in recent decades ([35]Siegel et al., 2018). Radiotherapy is an
   effective treatment of LUAD, especially when some patients are not
   suitable for surgery. However, LUAD from different patients may display
   different degrees of radiation tolerance. Radioresistance is the main
   factor reducing the effectiveness of radiotherapy, resulting treatment
   failure ([36]Provencio et al., 2010; [37]Le Pechoux, 2011). Thus, it is
   important to conduct researches on developing radiosensitizers to treat
   this lethal disease.

   Bioinformatics can help us explore a lot of valuable clues, find
   meaningful genes, and conduct new research. In the present study,
   [38]GSE32863, [39]GSE33532, and [40]GSE43458 were analyzed through
   bioinformatics. Finally, we obtained fou14 genes (PKMYT1, TTK, CHEK1,
   CDC20, PTTG1, MCM2, CDC25C, MCM4, CCNB1, CDC45, MAD2L1, CCNB2, BUB1,
   and CCNA2), which were significantly enriched in the cell cycle.

   After an extensive literature review of these 14 genes, we found that
   PKMYT1 is a highly promising gene that may be closely related to tumor
   radiosensitivity. This is because studies have found that the
   abrogation of G2 checkpoint effectively reduced radiation-induced cell
   cycle arrest and increased tumor radiosensitivity ([41]Leijen et al.,
   2010; [42]De Witt et al., 2011; [43]Qin et al., 2014; [44]Busch et al.,
   2017). Additionally, PKMYT1 is currently considered to be the target of
   G2 checkpoint elimination and mitotic catastrophe ([45]Tse et al.,
   2009; [46]Schmidt et al., 2017) because PKMYT1 can prevent cells from
   transitioning from G2 to mitosis phase in two ways. One is inhibiting
   Cdk1 activity via phosphorylating Cdk1 at Thr14 and Tyr15, and the
   other is preventing the Cdk1-CycB complex from entering the nucleus by
   binding to the Cdk1-CycB complex and sequestering it in the cytoplasm
   ([47]Fattaey and Booher, 1997; [48]Wells et al., 1999; [49]Passer et
   al., 2003).

   However, there is no experimental evidence to prove the relationship
   between PKMYT1 and tumor radiosensitivity. We designed experiments to
   see if PKMYT1 can affect the resistance of LUAD cells to radiation,
   providing important evidence for PKMYT1 as a radiosensitizer in
   radiotherapy.

Materials and Methods

Bioinformatics Analysis

   The [50]GSE32863, [51]GSE33532, and [52]GSE43458 were obtained from
   NCBI-GEO, which included 40 LUAD tissues and 10 normal lung tissues, 80
   LUAD tissues and 30 normal lung tissues, and 58 LUAD tissues and 58
   normal lung tissues, respectively. GEO2R was used to identify
   upregulated genes with logFC > 0.5 and an adjusted P < 0.001. Venn
   diagram was used to select upregulated genes co-expressed in these
   three gene chips. A PPI network was established by STRING. Then, we use
   the Cytoscape to find potential relevance between those genes. Next, we
   used MCODE (Molecular Complex Detection) to obtain core genes. We
   processed BP, MF, CC, and KEGG pathway analyses (P < 0.05) using the
   Database for Annotation, Visualization and Integrated Discovery
   (DAVID). The Kaplan Meier plotter and Gene Expression Profiling
   Interactive Analysis (GEPIA) were applied to get the significant
   prognostic information (P < 0.05) of hub genes in LUAD and their
   differential expression between LUAD tissues and normal lung tissues (P
   < 0.05).

Human LUAD Specimens

   Human LUAD tissues with the matched normal adjacent specimens were
   obtained from The First Hospital of China Medical University. This
   study was approved by the Ethics Committee of The First Hospital of
   China Medical University.

Cell Culture

   LUAD cell lines: A549, H299, H1975, H1650, and H441 were cultured using
   Roswell Park Memorial Institute-1640 (RPMI-1640, Hyclone) medium
   containing 10% fetal bovine serum (FBS, Clarks) at 37°C.

siRNA and Transfection

   Small interfering RNAs (siRNAs) against PKMYT1 (siRNA#1 and siRNA#2)
   and negative control siRNAs (NC-siRNA), which were obtained from
   Shanghai GENECHEM Co., Ltd. (Jikai, Shanghai, China), were transfected
   into A549 and H1299 cells using Neofect reagent (ProbeGenne, Jiangsu,
   China) according to the manufacturer’s instructions.

Quantitative PCR (Q-PCR)

   Reverse transcription was carried out according to the instructions of
   PrimeScript^TMRT reagent Kit with GNA Eraser (Takara), and SYBR Premix
   Ex TaqII (Takara) was used for Q-PCR. We used several sequences: PKMYT1
   forward primer 5′-CATGGCTCCTACGGAGAGGT-3′ and reverse primer
   5′-ACATGGAACGCTTTACCGCAT-3′; β-actin forward primer
   5′-CCAGACAGCACTGTGTTGGCATA-3′, and reverse primer
   5′-ATGTTGCCCTAGACTTCGAGCAAG-3. The results were calculated using the
   2^–ΔΔCt method.

Western Blotting Assay

   The protein samples were loaded onto SDS-PAGE, transferred onto PVDF
   membranes, and immunoblotted with primary antibodies against PKMYT1
   (Abcam) and β-actin (CST) and peroxidase-labeled secondary antibodies
   (CST). The Enhanced Chemiluminescence System (OLYMPUS) was used to
   display immune response bands.

Irradiation Treatment and Clonogenic Survival Assay

   Transfected A549 and H1299 cells were irradiated with 0, 2, 4, 6, and 8
   Gy X-rays. After 12 days, the cells were fixed with 4% paraformaldehyde
   and stained with 0.1% crystal violet solution. A single- hit
   multi-target model formula: SF = 1–(1-e^–D/D0)^n was used to fit cell
   survival curves and calculate parameters (D0, Dq, k, N, and SER).

Cell Cycle Analysis

   Cells transfected with siRNA # 1, siRNA # 2, and NC-siRNA were
   irradiated with X-rays at different doses (0, 4Gy). After irradiation,
   the cells were further cultured for 24 h. The cells were trypsinized,
   washed with PBS, fixed with ice-cold 70% ethanol, and stored at 4°C for
   12–24 h. Prior to analysis by flow cytometry, the cells were washed
   with PBS, treated with 0.25 mg/ml RNase A, and 50 μg/ml propidium
   iodide for 30 min at 37°C. Next, the cells were subjected to cell cycle
   analysis using a fluorescence-activated cell sorter (BD FACS Calibur).

Statistical Analysis

   Statistical analysis was conducted by SPSS 20.0 software. The
   statistical significance between groups was determined using a
   two-tailed Student’s t-test. One-way ANOVA was used to compare
   different time points/dose within the same group. P < 0.05 were
   considered to indicate statistically significant differences.

Results

Identification of Upregulated Genes in LUAD

   We used GEO2R to filter 2707, 1082, and 2085 upregulated genes from
   [53]GSE33532, [54]GSE43458, and [55]GSE32863, respectively. We
   identified 324 commonly upregulated genes in the above datasets via
   Venn diagram software ([56]Figure 1).

FIGURE 1.

   [57]FIGURE 1
   [58]Open in a new tab

   324 commonly up-regulated genes were obtained from [59]GSE33532,
   GSE43458and GSE 32863 via GEO2R online tools and Venn diagram software.

Analyses of Gene Ontology and Pathway Enrichment

   DAVID was used to analyze the KEGG pathway and GO analysis of 324
   genes. The GO analysis showed that, for biological processes (BP),
   genes significantly enriched DNA unwinding involved in DNA replication,
   collagen fibril organization, microtubule-based movement, DNA
   replication initiation, mitotic spindle assembly, and mitotic metaphase
   plate congression. For molecular function (MF), genes were primarily
   enriched in ATP binding, microtubule motor activity, and ER retention
   sequence binding. For cellular component (CC), genes were particularly
   enriched in the kinesin complex, midbody, and chromosome passenger
   complex. The results of the KEGG pathway analysis indicated that the
   324 genes were significantly enriched in the cell cycle, biosynthesis
   of amino acids, and fructose and mannose metabolism ([60]Table 1).

TABLE 1.

   Gene ontology analysis of 324 genes.
   Categories Term Count % p-value FDR
   GOTERM_BP_DIRECT GO:0006268~DNA unwinding involved in DNA replication 5
   0.009586073 9.71E-06 0.015322831
   GO:0030199~collagen fibril organization 7 0.013420503 1.80E-05
   0.028344256
   GO:0007018~microtubule-based movement 9 0.017254932 2.75E-05
   0.043412072
   GO:0006270~DNA replication initiation 5 0.009586073 6.78E-04
   1.065044432
   GO:0090307~mitotic spindle assembly 5 0.009586073 0.001347575
   2.105625858
   GO:0007080~mitotic metaphase plate congression 5 0.009586073
   0.001568185 2.446354599
   GOTERM_CC_DIRECT GO:0005871~kinesin complex 9 0.017254932 2.72E-06
   0.003474135
   GO:0030496~midbody 11 0.021089361 2.88E-06 0.003685049
   GO:0032133~chromosome passenger complex 4 0.007668859 6.78E-05
   0.086752849
   GOTERM_MF_DIRECT GO:0005524~ATP binding 44 0.084357446 5.39E-06
   0.007214573
   GO:0003777~microtubule motor activity 7 0.013420503 6.96E-04
   0.927666683
   GO:0046923~ER retention sequence binding 3 0.005751644 0.00101958
   1.355274839
   KEGG_PATHWAY cfa04110:Cell cycle 17 0.032592649 2.86E-09 3.49E-06
   cfa01230:Biosynthesis of amino acids 11 0.021089361 1.68E-06
   0.002040968
   cfa00051:Fructose and mannose metabolism 6 0.011503288 3.68E-04
   0.446985975
   [61]Open in a new tab

PPI Network and Module Analysis

   A PPI network was constructed which included 267 nodes and 3161 edges.
   Then, a significant module with 72 nodes and 2357 edges was identified
   via MCODE ([62]Figure 2).

FIGURE 2.

   FIGURE 2
   [63]Open in a new tab

   (A) Construction of the protein–protein interaction (PPI) network. (B)
   Identification of the significant module.

Identification of Core Genes

   Kaplan Meier plotter was used to analyze the overall survival (OS) of
   72 hub genes. We identified that 58 genes had a significantly poor OS
   in LUAD ([64]Table 2 and [65]Figure 3). Next, we found that 53 of 58
   genes are highly expressed in LUAD tissues compared to normal lung
   tissues ([66]Table 3 and [67]Figure 4) via GEPIA.

TABLE 2.

   The overall survival analysis of the 72 hub genes.
   Categories Genes names
   Genes with significantly poor OS (P < 0.05) MAD2L1 PKMYT1 KIF2C TTK
   CCNA2 BUB1 KIF20A CKAP2L BIRC5 CCNB2 EXO1 CDCA3 CDKN3 CDC20 OIP5 HJURP
   KIF14 MCM2 MCM10 PRC1 MELK KIF15 NCAPG PTTG1 TK1 KIF4A UBE2C FOXM1
   AURKA KIFC1 CEP55 CDC25C TYMS ASF1B KIAA0101 EZH2 CENPF DLGAP5 CHEK1
   CDCA5 ZWINT ANLN POLQ CDC45 GINS2 TIMELESS SHCBP1 RFC4 NEK2 AURKB UBE2T
   CCNB1 TPX2 MCM4 TOP2A PBK ASPM CDCA8
   Genes without significantly poor OS (P > 0.05) KNTC1 HMMR RAD51 POLE2
   ECT2 RAD51AP1 STIL ATAD2 PLK4 CDK1 KIF11 MCM6 CDCA7 CKS1B
   [68]Open in a new tab

FIGURE 3.

   FIGURE 3
   [69]Open in a new tab

   A total of 58 of 72 genes had a significantly poor survival rate in
   LUAD via Kaplan meier plotter online tools (P < 0.05).

TABLE 3.

   Differential expression of 53 genes between LUAD and normal lung
   tissues.
   Categories Genes names
   Genes with significantly high expression in LUAD (P < 0.05) MAD2L1
   PKMYT1 KIF2C TTK CCNA2 BUB1 KIF20A CKAP2L BIRC5 CCNB2 EXO1 CDCA3 CDKN3
   CDC20 OIP5 HJURP MCM2 MCM10 PRC1 MELK KIF15 NCAPG PTTG1 TK1 KIF4A UBE2C
   FOXM1 AURKA KIFC1 CEP55 CDC25C TYMS ASF1B KIAA0101 CENPF DLGAP5 CHEK1
   CDCA5 ZWINT ANLN CDC45 GINS2 SHCBP1 NEK2 AURKB UBE2T CCNB1 TPX2 MCM4
   TOP2A PBK ASPM CDCA8
   Genes without significantly high expression in LUAD (P < 0.05) KIF14
   EZH2 POLQ TIMELESS RFC4
   [70]Open in a new tab

FIGURE 4.

   FIGURE 4
   [71]Open in a new tab

   A total of 53 of 58 genes are significantly upregulated in LUAD samples
   contrasted to normal lung samples (*P < 0.05).

KEGG Pathway Analysis

   The results of the KEGG pathway analysis indicated that the 53 genes
   were significantly enriched in cell cycle, oocyte meiosis, and
   progesterone-mediated oocyte maturation. Results showed that fourteen
   genes (PKMYT1, TTK, CHEK1, CDC20, PTTG1, MCM2, CDC25C, MCM4, CCNB1,
   CDC45, MAD2L1, CCNB2, BUB1, and CCNA2) enriched in cell cycle
   ([72]Table 4).

TABLE 4.

   KEGG pathway enrichment analysis of 53 genes.
   Term Count % p-value genes
   cfa04110:Cell cycle 14 0.15171218 2.29E-18 PKMYT1, TTK, CHEK1, CDC20,
   PTTG1, MCM2, CDC25C, MCM4, CCNB1, CDC45, MAD2L1, CCNB2, BUB1, CCNA2
   cfa04114:Oocyte meiosis 8 0.086692674 2.13E-08 CCNB2, MAD2L1, BUB1,
   PKMYT1, CDC20, AURKA, PTTG1, CDC25C
   cfa04914: Progesterone-mediated oocyte maturation 7 0.07585609 1.67E-07
   CCNB1, CCNB2, MAD2L1, BUB1, PKMYT1, CDC25C, CCNA2
   [73]Open in a new tab

   After an extensive literature review of these 14 genes, we found that
   PKMYT1 is a highly promising gene that may be closely related to tumor
   radiosensitivity. However, there is no experimental evidence to prove
   the relationship between PKMYT1 and tumor radiosensitivity. We
   therefore conducted basic experiments to explore whether PKMYT1 can
   affect tumor radiosensitivity.

PKMYT1 Is Highly Expressed in Tumors Contrasted to Normal Tissues Derived
From LUAD Patients

   [74]Figure 5A reveals that the protein levels of PKMYT1 were distinctly
   increased in LUAD tissues compared to matched normal lung tissues via
   western blotting. [75]Figure 5B shows that the mRNA levels of PKMYT1 in
   LUAD tissues were significantly higher than those in adjacent normal
   lung tissues by Q-PCR.

FIGURE 5.

   [76]FIGURE 5
   [77]Open in a new tab

   PKMYT1 is highly expressed in lung adenocarcinoma tissues. (A) Relative
   PKMYT1 protein levels in 24 pairs of lung adenocarcinoma tissues and
   their matched normal tissues were determined by western blotting. (B)
   Relative PKMYT1 mRNA levels in 33 pairs of lung adenocarcinoma tissues
   and their matched normal tissues were determined by Q-PCR. (C)
   Expression level of PKMYT1 in five lung adenocarcinoma cell lines. ***p
   < 0.001.

siRNA Mediated the Silence of PKMYT1

   The expression of PKMYT1 in LUAD cell lines, including A549, H299,
   H1975, H1650, and H441, was confirmed by WB ([78]Figure 5C). We next
   use the two most expressive ones (A549 and H1299 cells) for follow-up
   experiments. Both cells were infected with either control or PKMYT1
   siRNA. Then, the Q-PCR and western blotting analyses were conducted,
   and the data suggested that siRNAs can significantly inhibit the
   expression of PKMYT1 in RNA and protein levels ([79]Figures 6A,B).

FIGURE 6.

   [80]FIGURE 6
   [81]Open in a new tab

   Silencing efficiency of PKMYT1 in A549 and H1299 cells. (A) Western
   blotting assays detected the relative PKMYT1 protein levels in A549 and
   H1299 cells after transfecting with siRNA#1, siRNA#2, and NC-siRNA. (B)
   Q-PCR assays detected the relative PKMYT1 mRNA levels in A549 and H1299
   cells after transfecting with siRNA#1, siRNA#2, and NC-siRNA. *p <
   0.05, **p < 0.01.

Downregulation of PKMYT1 Is Significantly Associated With Radiosensitivity of
LUAD Cells

   The radiosensitivity of transfected A549 and H1299 cells was examined
   by colony formation assay. The data in [82]Figure 7 demonstrates that
   artificial silencing of PKMYT1 remarkably enhanced radiosensitizing
   properties in LUAD Cells.

FIGURE 7.

   FIGURE 7
   [83]Open in a new tab

   PKMYT1 silencing results in enhanced radiosensitivity of lung
   adenocarcinoma cells. (A) Representative images of colony formation in
   A549 and H1299 cells pretreated with SiRNA#1, SiRNA#2, or NC-SiRNA for
   48 h and then exposed to 2, 4, 6, or 8 Gy of X-ray irradiation. (B)
   Clonogenic cell survival curves were generated for A549 and H1299 cells
   treated with SiRNA#1, SiRNA#2, or NC-SiRNA after radiation. The
   survival data were normalized to those of the unirradiated control
   cells. (C) The single-hit multitarget model to calculate the
   radiobiological parameters of A549 and H1299 cell lines after
   downregulating PKMYT1 expression.

Downregulation of PKMYT1 Alters Cell Cycle Distribution and Abrogates
IR-Induced G2/M Phase Arrest

   As shown in [84]Figures 8, [85]9, silencing of PKMYT1 can decrease the
   G2/M cell proportion in A549 and H1299, and the difference was
   statistically significant (P < 0.05). In both cell lines, following
   treatment with silencing of PKMYT1 + radiation (4Gy of X-rays), the
   cells also accumulated in G2/M, but to a much lower level compared to
   the radiation control (p < 0.05), suggesting that the silencing of
   PKMYT1 can eliminate the radiation-induced G2/M arrest, thereby
   improving the radiosensitivity of tumor cells.

FIGURE 8.

   [86]FIGURE 8
   [87]Open in a new tab

   Downregulation of PKMYT1 abolishes irradiation-induced G2/M arrest in
   A549 cells.

FIGURE 9.

   [88]FIGURE 9
   [89]Open in a new tab

   Downregulation of PKMYT1 abolishes irradiation-induced G2/M arrest in
   H1299 cells.

Discussion

   The identification of a target to improve the radiosensitivity of LUAD
   is significant. In this article, we obtained 14 core genes related to
   the prognosis of LUAD by analyzing [90]GSE32863, [91]GSE33532, and
   [92]GSE43458. After an extensive literature review of these 14 genes,
   we found that PKMYT1 is a highly promising gene that may be closely
   related to tumor radiosensitivity. This is because PKMYT1 is currently
   considered to be the target of G2 checkpoint elimination and mitotic
   catastrophe ([93]Tse et al., 2009; [94]Schmidt et al., 2017) as it can
   prevent cells from transitioning from G2 to mitosis phase by affecting
   the activity of Cdk1 and the nuclear shuttle of the Cdk1-CycB complex
   ([95]Fattaey and Booher, 1997; [96]Wells et al., 1999; [97]Passer et
   al., 2003). Additionally, the abrogation of the G2 checkpoint
   effectively reduced radiation-induced cell cycle arrest and increased
   tumor radiosensitivity ([98]Leijen et al., 2010; [99]De Witt et al.,
   2011; [100]Qin et al., 2014; [101]Busch et al., 2017). However, the
   functional importance of PKMYT1 in tumor radiosensitivity has not yet
   been reported.

   In this study we demonstrated that PKMYT1 is highly expressed in LUAD
   tissues contrasted with normal lung tissues. We found that artificially
   silencing the expression of PKMYT1 can increase the sensitivity of LUAD
   cells to radiation. Cell cycle analysis found that downregulation of
   PKMYT1 alters cell cycle distribution and abrogates IR-induced G2/M
   phase arrest.

   Most cancer types fail to establish the G1 checkpoint, and this is
   usually because of mutation or deletion of the p53 tumor suppressor.
   Hence, cells deficient in p53 are more vulnerable to inactivation of
   the G2 checkpoint ([102]Passer et al., 2003). As a consequence,
   therapeutic ablation of G2 checkpoints may preferentially make
   p53-deficient tumor cells sensitive to DNA damaging treatments while
   retaining normal cells with intact p53 function ([103]Dixon and
   Norbury, 2002; [104]Meng et al., 2018). In addition, a very interesting
   point is that research has found that PKMYT1 is not necessary for the
   cell cycle of normal cells, but it has a rate-limiting function for the
   recovery of checkpoints after DNA damage ([105]van Vugt et al., 2004;
   [106]Chow and Poon, 2013). Inhibitors of components essential for the
   progression of cell cycle progression in normal cells may be limited by
   their greater toxicity to normal cells when subjected to anti-tumor
   therapy. Since PKMYT1 is relatively less important for normal cell
   cycle progression, PKMYT1 is a particularly attractive target for
   anticancer therapy.

Conclusion

   In summary, we obtained 14 core genes (PKMYT1, TTK, CHEK1, CDC20,
   PTTG1, MCM2, CDC25C, MCM4, CCNB1, CDC45, MAD2L1, CCNB2, BUB1, and
   CCNA2) related to the poor prognosis of LUAD via bioinformatical
   analysis. We identified that PKMYT1 was significantly upregulated in
   LUAD tissues and firstly demonstrated that knockdown of PKMYT1 resulted
   lower survival rate when cells receiving radiation therapy. Our
   findings suggested that PKMYT1 is a promising target to improve the
   radiosensitivity of lung adenocarcinoma.

Data Availability Statement

   The datasets generated for this study can be found in the data is
   available at NCBI GEO under accession numbers [107]GSE33532
   , [108]GSE32863, and [109]GSE43458.

Ethics Statememt

   This study was approved by the Ethics Committee of The First Hospital
   of China Medical University.

Author Contributions

   HL analyzed the data and drafted the manuscript. JL, YY, QQ, and GL
   provided suggestions and approved the final manuscript.

Conflict of Interest

   The authors declare that the research was conducted in the absence of
   any commercial or financial relationships that could be construed as a
   potential conflict of interest.

Abbreviations

   DAVID
          the Database for Annotation Visualization and Integrated
          Discovery

   GO
          Gene Ontology

   GEPIA
          the Gene Expression Profiling Interactive Analysis

   KEGG
          Kyoto Encyclopedia of Genes and Genomes

   LUAD
          lung adenocarcinoma

   PPI
          protein-protein interaction

   OS
          overall survival.

References