Molecular Medicine Reports Molecular Medicine Reports 1791-2997 1791-3004 D.A. Spandidos 10.3892/mmr.2017.7656 mmr-16-06-8371 Articles Integrated analysis reveals candidate genes and transcription factors in lung adenocarcinoma ChenBaiwang * GaoShuhong * JiChangwei * SongGe Intensive Care Unit, Jining No. 1 People's Hospital, Jining, Shandong 272011, P.R. China Correspondence to: Dr Ge Song, Intensive Care Unit, Jining No. 1 People's Hospital, 6 Jiankang Road, Jining, Shandong 272011, P.R. China, E-mail: ge_song_doctor@sina.com

Contributed equally

 122017 28092017 16 6 8371 8379 29022016 23022017 Copyright © 2017, Spandidos Publications 2017

Lung adenocarcinoma is the most common type of non-small cell lung cancer in Asia. Therefore, it is important to improve understanding of the underlying transcriptional regulatory mechanisms involved. The present study aimed to identify potential candidate genes and transcription factors (TFs) associated with the disease. Four gene expression profiles were downloaded from the Gene Expression Omnibus database, which included 141 lung adenocarcinoma patients and 191 healthy controls. The differentially expressed genes (DEGs) were screened out and functional annotation was performed. In addition, TFs were identified and a global transcriptional regulatory network was constructed. Integrated analysis gave rise to a total of 1,238 DEGs in lung adenocarcinoma when compared with healthy tissues, including 970 upregulated and 268 downregulated DEGs. The six overexpressed outlier genes of ceruloplasmin, heparan sulfate 6-O-sulfotransferase 2, transmembrane protease serine 4, anillin actin binding protein, cellular retinoic acid binding protein 2 and cystatin SN may serve important roles in the development of lung adenocarcinoma. In addition, the downregulation of carbonic anhydrase 4 and S100 calcium binding protein A12 may render these effective diagnostic biomarkers. The results of the transcriptional regulatory network demonstrated that the hub nodes were sex determining region Y-box 10, Spi-B transcription factor and nuclear receptor subfamily 4 group A member 2. The four TFs, forkhead box D1, E74-like ETS transcription factor 5, homeobox A5 and kruppel-like factor 5, may warrant future investigations into their function in disease development. In conclusion, the present study provided for further studies a list of candidate genes and TFs for the detection and treatment of lung adenocarcinoma.

 lung adenocarcinoma candidate genes transcription factors integrated analysis Introduction

Lung adenocarcinoma is a malignant cancer and a primary subtype of non-small cell lung cancer (NSCLC), with the greatest incidence and the worst prognosis worldwide (1). In the majority of cases the development of lung adenocarcinoma is a multifactor and multistage process, which is associated with numerous genes (2).

In addition to the gene expression exhibited by cancer and healthy tissues, identification of the differential interactions between genes in the development of lung adenocarcinoma should be considered, as this may identify critical genes that may not otherwise be detectable (3,4). Transcription factors (TFs) bind to a specific region of the DNA sequence and consequently regulate the transcription of target genes (5,6). Transcriptional regulation is crucial for the development of lung adenocarcinoma (7). Therefore, it is important to construct gene regulatory networks that represent this (8).

Extensive investigation has been performed into the underlying mechanisms of lung adenocarcinoma. A number of studies have assessed gene expression in lung adenocarcinoma (9–12) or identified marker genes (13). Using computational methods, the potential associations between TFs and differentially expressed genes (DEGs) in the regulation of transcription in lung adenocarcinoma have been identified and a regulatory network was constructed (14). A previous study examined the underlying mechanisms of lung adenocarcinoma through the regulatory network using GSE2514 microarray data (7). A previous study on the synergistic regulation of microRNAs (miRNAs) and TFs have identified a variety of significant motifs (15). In addition, a miRNA-TF synergistic regulation network has been constructed (2).

However, the molecular mechanisms underlying lung adenocarcinoma remain to be fully elucidated. Therefore, analysis of the regulatory mechanism based on a large-scale study of genes associated with this disease is important to further the understanding of lung adenocarcinoma. The large body of biological data generated from gene expression profiles is a useful resource for understanding and deducing gene function (16).

The present study performed computational bioinformatics analysis of gene expression for the identification of potential transcriptional regulation associations between TFs and DEGs in lung adenocarcinoma and adjacent healthy tissue samples. The significantly enriched functions of these genes were investigated to further the understanding of the molecular mechanisms underlying lung adenocarcinoma. In addition, a transcriptional regulatory network was constructed.

 Materials and methods <sec> <title>Source of datasets

The transcriptome sequencing data from lung adenocarcinoma patients were downloaded from the Gene Expression Omnibus (GEO) repository (www.ncbi.nlm.nih.gov/geo/) (17). The following key words were used: [‘Lung Adenocarcinomas’ (MeSH Terms) or ‘Lung Adenocarcinomas’ (All Fields)] and ‘Homo sapiens’ (porgn) and ‘gse’ (Filter). In total, four datasets were obtained with the following accession numbers: GSE62949, GSE27262, GSE43458 and GSE32863. There were 191 cases and 141 controls enrolled in the present study. For each patient, the tumor and paired healthy tissue had been sequenced. The characteristics of the eligible datasets are summarized in Table I.

 Differential gene expression analysis

For all datasets, the gene expression level data for cancerous and healthy tissues were preprocessed by background correction and normalization. The limma package (bioconductor.org/packages/release/bioc/html/limma.html) (18) in R was used to analyze differential expression between lung adenocarcinoma and healthy tissues by using a paired Student's t-test. The P-value and false discovery rate (FDR) were obtained. Genes with Benjamini-Hochberg adjusted FDR<0.05 were reported as DEGs in the present study (19).

 Gene function annotation and pathway analysis

To assess the alterations in DEGs occurring at the cellular level, and the functional clustering of DEGs, the online Gene Ontology Enrichment Analysis and Visualization Tool (cbl-gorilla.cs.technion.ac.il/) (20,21) was used to identify and visualize the Gene Ontology (GO) database categories: Biology process, molecular function and cellular component (22). In addition, GeneCodis3 (genecodis.cnb.csic.es/analysis) (23–25) was used to conduct Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to investigate the functional roles and associations of the genes with varied expression in the analysis.

 Screening of potential TFs

To understand the regulatory mechanisms, the present study further analyzed TFs, which are essential for the regulation of gene activation or repression, in lung adenocarcinoma. The TFs in the human genome and the motifs of genomic binding sites were downloaded from the TRANSFAC® database (http://gene-regulation.com/pub/databases.html) (26), and the DEGs encoding TFs were identified. The TRANSFAC position weight matrix was used for gene promoter scanning to identify DEGs with the TF binding sites in the promoter region. Finally, the transcriptional regulatory networks were established and visualized using Cytoscape 3.0 software (www.cytoscape.org/) (27,28).

 RNA preparation and reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

A total of 4 patients with lung adenocarcinoma (mean age, 54±1.8) were recruited from Jining No. 1 People's Hospital between October 2015 and December 2015, including two men and two women. Tumors and corresponding healthy lung tissue samples were surgically resected, and immediately frozen in liquid nitrogen. Our study was approved by the ethics committee of Jining No. 1 People Hospital. The written informed consents were obtained from each of the patients.

Total RNA was extracted using TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer's protocol. RNA was reverse-transcribed using SuperScript® III Reverse Transcriptase (Invitrogen; Thermo Fisher Scientific, Inc.). RT-qPCR was performed using an ABI 7500 Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) and a Power SYBR® Green PCR Master mix (Invitrogen; Thermo Fisher Scientific, Inc.). qPCR reaction was performed under the following conditions: After denaturing for 10 min at 95°C, PCR was performed for 45 cycles of 95°C for 15 sec and 60°C for 1 min, followed by a 10 min incubation at 72°C. The results were analyzed using 2−ΔΔCq method (29). Student's t-test was performed to compare the gene expression in cancer and healthy tissue using Microsoft Excel. All reactions were analyzed triplicates. The human β-actin gene was used as endogenous control. The primers were listed in Table II.

 Results <sec> <title>Identification of differentially expressed genes

According to the inclusion criteria, four microarray datasets were obtained. Integrated analysis of these generated 1,238 DEGs with FDR<0.05 in lung adenocarcinoma compared with healthy tissues, including 970 upregulated and 268 downregulated DEGs. The top ten up and downregulated DEGs are presented in Table III.

 Functional enrichment analysis of DEGs

GO enrichment analysis of DEGs was performed to understand their biological functions. A total of 3 GO categories were investigated: Biological process, cellular component and molecular function. The results revealed that the significantly enriched GO terms for biological process were involved in the regulation of responses to stimulus (GO: 0048583; P=5.00E-07), regulation of defense responses (GO: 0031347; P=2.00E-06) and circulatory system processes (GO: 0003013; P=2.79E-06). In addition, the plasma membrane part (GO: 0044459; P=1.35E-05) was the significantly enriched GO term for cellular component. Notably, the significantly enriched GO term for molecular function was receptor activity (GO: 0004872; P=6.43E-07; Table IV).

KEGG pathway enrichment analysis indicated that extracellular matrix (ECM)-receptor interaction (FDR=1.05E-07) and cell cycle (FDR=1.18E-07) were significantly enriched. Furthermore, focal adhesion (FDR=9.43E-06), cell adhesion molecules (FDR=7.72E-05) and pathways in cancer (FDR=1.71E-04) were enriched (Table V).

 Construction of TF-target gene regulatory network for lung adenocarcinoma

To construct the TF-target gene regulatory network for lung adenocarcinoma, the TRANSFAC database was utilized to investigate TFs and their latent target genes. Differentially expressed TFs and latent target genes in lung adenocarcinoma were selected. A total of 40 differentially expressed TFs (27 upregulated and 13 downregulated) and 544 latent differentially expressed target genes were identified. The transcriptional regulatory network was subsequently constructed based on these findings. In the network, there were 36 TFs and 752 TF-target interactions (Fig. 1). The top ten TFs regulating the greatest number of downstream target genes were sex determining region Y-box 10 (SOX10), Spi-B transcription factor (SPIB), nuclear receptor subfamily 4 group A member 2 (NR4A2), forkhead box D1 (FOXD1), E74 like ETS transcription factor 5 (ELF5), homeobox A5 (HOXA5), kruppel like factor 5 (KLF5), estrogen related receptor α (ESRRA), sterol regulatory element binding transcription factor 1 (SREBF1) and REL proto-oncogene, NF-kB subunit (REL; Table VI).

 Validation of differentially expressed TFs and targets

Tumor and corresponding healthy lung tissue samples were used to validate the findings of the integrated analysis. Two TFs of SOX10 and HOXA5 were selected, where SOX10 had the highest number of downstream DEGs and HOXA5 was the primary significantly upregulated TF. In addition, solute carrier family 6 member 4 (SLC6A4) and sclerostin domain containing 1 (SOSTDC1) were two targets of SOX10 and they were listed in the top 50 significant DEGs. The transmembrane protease serine 4 (TMPRSS4) was a target of HOXA5 and it was listed in the top 300 significant DEGs. Therefore, SLC6A4, SOSTDC1 and TMPRSS4 were selected for validation. The RT-qPCR results demonstrated that the expression pattern of the selected genes was similar to that identified by the integrated analysis. SOX10 and TMPRSS4 were upregulated, whereas SLC6A4, SOSTDC1 and HOXA5 were downregulated in lung adenocarcinoma compared with the corresponding healthy lung tissue samples (Fig. 2). The significance of difference was slightly different, which was primarily due to the difference of sample number. There were 4 pairs samples used in the RT-qPCR experiment, whereas there were 191 cases and 141 controls in the integrated analysis.

 Discussion

Lung adenocarcinoma is the most common histological subtype of lung cancer. The present study investigated the molecular mechanisms underlying lung adenocarcinoma through the regulatory network using microarray datasets obtained from the GEO database. Integrated analysis of four microarray datasets identified a total of 1,238 DEGs (970 upregulated and 268 downregulated) in lung adenocarcinoma compared with healthy tissues. Functional annotation demonstrated that DEGs were closely associated with common pathways for cancers, including the cell cycle, p53 signaling pathway and pathways in cancer. In addition, ECM-receptor interactions, focal adhesion and cell adhesion molecules were significantly enriched, which may be closely associated with tumorigenesis in lung adenocarcinoma.

Of the top ten upregulated and downregulated DEGs, the majority were associated with the pathological process of lung adenocarcinoma. A previous study demonstrated that ceruloplasmin (CP) was overexpressed at a high frequency in lung adenocarcinoma compared with corresponding healthy lung tissues (30). Heparan sulfate 6-O-sulfotransferase 2 (HS6ST2) is significantly overexpressed in lung tumor tissues (31). TMPRSS4 expression has been associated with postoperative recurrence in patients with lung cancer (32). In addition, anillin actin binding protein (ANLN) has been reported to be essential for the formation or organization of actin cables in the cleavage furrow and serves an important role in cytokinesis (33). Suzuki et al (34) demonstrated that ANLN was overexpressed in the majority of primary NSCLCs, and the endogenous expression of ANLN in the nucleus was significantly associated with poor prognosis in patients with NSCLC. Cellular retinoic acid binding protein 2 (CRABP2) expression is markedly increased in lung adenocarcinoma (35) and the expression of cystatin SN (CST1) is closely associated with tumor metastasis properties in A549L6 cells (36). In the present study, CP, HS6ST2, TMPRSS4, ANLN, CRABP2 and CST1 were significantly upregulated in lung adenocarcinoma compared with corresponding healthy lung tissues, indicating that the overexpressed outlier genes may serve important roles in the development of lung adenocarcinoma.

In addition, three of the top ten downregulated genes have been reported previously. Carbonic anhydrase 4 is downregulated in lung adenocarcinoma (37). S100 calcium binding protein A12 (S100A12) is a proinflammatory marker that has the potential to be a diagnostic biomarker of NSCLC (38). A recent study reported that interleukin (IL)-6 is upregulated in lung adenocarcinoma and suggested that IL-6 may be a therapeutic target for the treatment of V-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog-driven lung adenocarcinoma (39). However, the present study revealed that IL-6 was downregulated in lung adenocarcinoma. Therefore, further studies are required.

Using the TRANSFAC database, 40 differentially expressed TFs were identified in lung adenocarcinoma and a transcriptional regulatory network was constructed. Based on the constructed transcriptional regulatory network, a set of crucial TFs, which had the highest number of downstream DEGs, were identified as being of interest, including SOX10, SPIB, NR4A2, FOXD1, ELF5, HOXA5, KLF5, ESRRA, SREBF1 and REL. A number of TFs may serve important roles in the development of lung adenocarcinoma.

The human Forkhead-box gene family consists of at least 43 members (40), including FOXD1, the loss of which may suppress cell proliferation and significantly increase the life expectancy of patients with NSCLC (41). A study using a transgenic mouse model of papillary lung adenocarcinomas revealed that ELF5 may cooperate with c-Myc to suppress and upregulate genes in cancer samples, which may serve an essential role in neoplastic transformation (42). Microarrays of invasion/metastasis lung adenocarcinoma cell lines revealed that HOXA5 may contribute to the suppression of metastasis in lung cancer via the regulation of cytoskeleton remodeling. In addition, KLF5 may promote the apoptosis of lung adenocarcinoma cells, potentially via the inhibition of cell proliferation and repair/activation of apoptosis pathway proteins (43). Therefore, the results of the present study may be useful for future investigations into the role of transcription factors in the development of this complex disease.

In conclusion, the present study generated a list of candidate genes and TFs for the future detection and treatment of lung adenocarcinoma. The results highlighted the potential mechanisms underlying human lung adenocarcinoma through the transcriptional regulatory network.

 References MinnaJDRothJAGazdarAFFocus on lung cancerCancer cell14952200210.1016/S1535-6108(02)00027-212086887 ZhaoNLiuYChangZLiKZhangRZhouYQiuFHanXXuYIdentification of biomarker and co-regulatory motifs in lung adenocarcinoma based on differential interactionsPLoS One10e0139165201510.1371/journal.pone.0139165264022524581687 BandyopadhyaySMehtaMKuoDSungMKChuangRJaehnigEJBodenmillerBLiconKCopelandWShalesMRewiring of genetic networks in response to DNA damageScience33013851389201010.1126/science.1195618211272523006187 LiuXLiuZPZhaoXMChenLIdentifying disease genes and module biomarkers by differential interactionsJ Am Med Inform Assoc19241248201210.1136/amiajnl-2011-00065822190040 LatchmanDSTranscription factors: An overviewInt J Biochem Cell Biol2913051312199710.1016/S1357-2725(97)00085-X9570129 MitchellPJTjianRTranscriptional regulation in mammalian cells by sequence-specific DNA binding proteinsScience245371378198910.1126/science.26671362667136 MengXLuPBaiHXiaoPFanQTranscriptional regulatory networks in human lung adenocarcinomaMol Med Rep6961966201210.3892/mmr.2012.103422895549 ChenCYChenSTFuhCSJuanHFHuangHCCoregulation of transcription factors and microRNAs in human transcriptional regulatory networkBMC Bioinformatics12Suppl 1S41201110.1186/1471-2105-12-S1-S41213425733044298 GarberMETroyanskayaOGSchluensKPetersenSThaeslerZPacyna-GengelbachMvan de RijnMRosenGDPerouCMWhyteRIDiversity of gene expression in adenocarcinoma of the lungProc Natl Acad Sci USA981378413789200110.1073/pnas.2415007981170759061119 Director's Challenge Consortium for the Molecular Classification of Lung AdenocarcinomaSheddenKTaylorJMEnkemannSATsaoMSYeatmanTJGeraldWLEschrichSJurisicaIGiordanoTJGene expression-based survival prediction in lung adenocarcinoma: A multi-site, blinded validation studyNat Med14822827200810.1038/nm.1790186416602667337 StearmanRSDwyer-NieldLZerbeLBlaineSAChanZBunnPAJrJohnsonGLHirschFRMerrickDTFranklinWAAnalysis of orthologous gene expression between human pulmonary adenocarcinoma and a carcinogen-induced murine modelAm J Pathol16717631775200510.1016/S0002-9440(10)61257-6163144861613183 ZhangWGongWAiHTangJShenCGene expression analysis of lung adenocarcinoma and matched adjacent non-tumor lung tissueTumori100338345201425076248 JiangHDengYChenHSTaoLShaQChenJTsaiCJZhangSJoint analysis of two microarray gene-expression data sets to select lung adenocarcinoma marker genesBMC Bioinformatics581200410.1186/1471-2105-5-8115217521476733 FuSPanXFangWDifferential co-expression analysis of a microarray gene expression profiles of pulmonary adenocarcinomaMol Med Rep10713718201424913908 LinCCChenYJChenCYOyangYJJuanHFHuangHCCrosstalk between transcription factors and microRNAs in human protein interaction networkBMC Syst Biol618201210.1186/1752-0509-6-18224138763337275 LiBQYouJChenLZhangJZhangNLiHPHuangTKongXYCaiYDIdentification of lung-cancer-related genes with the shortest path approach in a protein-protein interaction networkBiomed Res Int20132673752013237628323674655 BarrettTWilhiteSELedouxPEvangelistaCKimIFTomashevskyMMarshallKAPhillippyKHShermanPMHolkoMNCBI GEO: Archive for functional genomics data sets-updateNucleic Acids Res41Database IssueD991D995201323193258 RitchieMEPhipsonBWuDHuYLawCWShiWSmythGKlimma powers differential expression analyses for RNA-sequencing and microarray studiesNucleic Acids Res43e47201510.1093/nar/gkv007256057924402510 HardcastleTJGeneralized empirical Bayesian methods for discovery of differential data in high-throughput biologyBioinformatics32195202201626428289 EdenENavonRSteinfeldILipsonDYakhiniZGOrilla: A tool for discovery and visualization of enriched GO terms in ranked gene listsBMC Bioinformatics1048200910.1186/1471-2105-10-48191922992644678 EdenELipsonDYogevSYakhiniZDiscovering motifs in ranked lists of DNA sequencesPLoS Comput Biol3e39200710.1371/journal.pcbi.0030039173812351829477 AshburnerMBallCABlakeJABotsteinDButlerHCherryJMDavisAPDolinskiKDwightSSEppigJTGene ontology: Tool for the unification of biology. The gene ontology consortiumNat Genet252529200010.1038/75556108026513037419 Nogales-CadenasRCarmona-SaezPVazquezMVicenteCYangXTiradoFCarazoJMPascual-MontanoAGeneCodis: Interpreting gene lists through enrichment analysis and integration of diverse biological informationNucleic Acids Res37Web Server IssueW317W322200910.1093/nar/gkp416194653872703901 Tabas-MadridDNogales-CadenasRPascual-MontanoAGeneCodis3:A non-redundant and modular enrichment analysis tool for functional genomicsNucleic Acids Res40Web Server IssueW478W483201210.1093/nar/gks402225731753394297 Carmona-SaezPChagoyenMTiradoFCarazoJMPascual-MontanoAGENECODIS: A web-based tool for finding significant concurrent annotations in gene listsGenome Biol8R3200710.1186/gb-2007-8-1-r3172041541839127 MatysVFrickeEGeffersRGösslingEHaubrockMHehlRHornischerKKarasDKelAEKel-MargoulisOVTRANSFAC: Transcriptional regulation, from patterns to profilesNucleic Acids Res31374378200310.1093/nar/gkg10812520026165555 SmootMEOnoKRuscheinskiJWangPLIdekerTCytoscape 2.8: New features for data integration and network visualizationBioinformatics27431432201110.1093/bioinformatics/btq67521149340 ShannonPMarkielAOzierOBaligaNSWangJTRamageDAminNSchwikowskiBIdekerTCytoscape: A software environment for integrated models of biomolecular interaction networksGenome Res1324982504200310.1101/gr.123930314597658403769 LivakKJSchmittgenTDAnalysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) methodMethods25402408200110.1006/meth.2001.126211846609 WangKKLiuNRadulovichNWigleDAJohnstonMRShepherdFAMindenMDTsaoMSNovel candidate tumor marker genes for lung adenocarcinomaOncogene2175987604200210.1038/sj.onc.120595312386823 BackenACColeCLLauSCClampARMcVeyRGallagherJTJaysonGCHeparan sulphate synthetic and editing enzymes in ovarian cancerBr J Cancer9615441548200710.1038/sj.bjc.6603747174370112359940 ChikaishiYUramotoHKoyanagiYYamadaSYanoSTanakaFTMPRSS4 expression as a marker of recurrence in patients with lung cancerAnticancer Res36121127201626722035 OegemaKSavoianMSMitchisonTJFieldCMFunctional analysis of a human homologue of the Drosophila actin binding protein anillin suggests a role in cytokinesisJ Cell Biol150539552200010.1083/jcb.150.3.539109318662175195 SuzukiCDaigoYIshikawaNKatoTHayamaSItoTTsuchiyaENakamuraYANLN plays a critical role in human lung carcinogenesis through the activation of RHOA and by involvement in the phosphoinositide 3-kinase/AKT pathwayCancer Res651131411325200510.1158/0008-5472.CAN-05-150716357138 HanSSKimWJHongYHongSHLeeSJRyuDRLeeWChoYHLeeSRyuYJRNA sequencing identifies novel markers of non-small cell lung cancerLung Cancer84229235201410.1016/j.lungcan.2014.03.01824751108 CaiXLuoJYangXDengHZhangJLiSWeiHYangCXuLJinRIn vivo selection for spine-derived highly metastatic lung cancer cells is associated with increased migration, inflammation and decreased adhesionOncotarget62290522917201510.18632/oncotarget.4416260908684673208 ChenLZhuoDChenJYuanHScreening feature genes of lung carcinoma with DNA microarray analysisInt J Clin Exp Med812161121712015265501264612811 LimMYThomasPSBiomarkers in exhaled breath condensate and serum of chronic obstructive pulmonary disease and non-small-cell lung cancerInt J Chronic Dis20135786132013264648464590922 BrooksGDMcLeodLAlhayyaniSMillerARussellPAFerlinWRose-JohnSRuwanpuraSJenkinsBJIL6 Trans-signaling Promotes KRAS-Driven lung carcinogenesisCancer Res76866876201610.1158/0008-5472.CAN-15-238826744530 KatohMKatohMHuman FOX gene family (Review)Int J Oncol2514951500200415492844 NakayamaSSoejimaKYasudaHYodaSSatomiRIkemuraSTeraiHSatoTYamaguchiNHamamotoJFOXD1 expression is associated with poor prognosis in non-small cell lung cancerAnticancer Res35261268201525550559 CiribilliYSinghPSpanelRIngaABorlakJDecoding c-Myc networks of cell cycle and apoptosis regulated genes in a transgenic mouse model of papillary lung adenocarcinomasOncotarget63156931592201510.18632/oncotarget.5035264270404741625 WangCCSuKYChenHYChangSYShenCFHsiehCHHongQSChiangCCChangGCYuSLChenJJHOXA5 inhibits metastasis via regulating cytoskeletal remodelling and associates with prolonged survival in non-small-cell lung carcinomaPLoS One10e0124191201510.1371/journal.pone.0124191258758244396855

Constructed transcriptional regulatory network of lung adenocarcinoma. Red and green nodes denote upregulated and downregulated transcription factors, respectively. Blue nodes denote target genes.

Reverse transcription-quantitative polymerase chain reaction was performed to validate the transcription factors and their targets in the four patients with lung adenocarcinoma compared with the corresponding healthy tissue. Data are presented as the mean ± standard deviation. *P<0.05 vs. corresponding healthy lung tissue (control). SLC6A4, solute carrier family 6 member 4; SOSTDC1, sclerostin domain containing 1; TMPRSS4, transmembrane protease serine 4; SOX10, sex determining region Y-box 10; HOXA5, homeobox A5.

Characteristics of the individual GEO studies that produced the eligible datasets used in the present study.

GEO ID No. of samples (cancer:control) Platform Country Year
GSE62949 28:28 GPL8432 Illumina HumanRef-8 WG-DASL v3.0 USA 2015
GSE27262 25:25 GPL570 [HG-U133_Plus_2] Affymetrix Human Genome U133 Plus 2.0 Array China (Taiwan) 2013
GSE43458 30:80 GPL6244 [HuGene-1_0-st] Affymetrix Human Gene 1.0 ST Array [transcript (gene) version] USA 2013
GSE32863 58:58 GPL6480GPL6884 Illumina HumanWG-6 v3.0 expression beadchip USA 2012

The transcriptome sequencing data from patients with lung adenocarcinoma were downloaded from the GEO database. In total, four datasets were obtained with the following accession numbers: GSE62949, GSE27262, GSE43458 and GSE32863. Collectively, data from 332 individuals were obtained, comprising 141 patients with lung adenocarcinoma and 191 controls. For each patient, the tumor and paired healthy tissue had been sequenced. GEO, Gene Expression Omnibus.

Primers used in the present study.

Gene Forward (5′-3′) Reverse (5′-3′) Length (bp)
SLC6A4 CGTGCTCGCCGTGGTCAT CCCCGTGGCATACTCCTCC 100
SOSTDC1 CCCAGCAGCAACAGCACG CAGTTCCCGGCAACCCAC 105
TMPRSS4 ACACGGTGCAATGCAGACGA AGCCATAGCCCCAACTAACGA 169
SOX10 TCAGCGGCTACGACTGGACG CGTTGTGCAGGTGCGGGTA 156
HOXA5 TTCAACCGTTACCTGACCCGC TAAACGCTCAGATACTCAGGGACGG 183
β-actin CTGAAGTACCCCATCGAGCAC ATAGCACAGCCTGGATAGCAAC 223

Top ten upregulated and downregulated differentially expressed genes.

A, Upregulated
ID No. Symbol Log FC FDR
1356 CP 4.10E+00 2.33E-05
9245 GCNT3 3.91E+00 5.76E-05
90161 HS6ST2 6.29E+00 7.22E-04
56649 TMPRSS4 5.27E+00 8.23E-04
54443 ANLN 4.45E+00 1.12E-03
1382 CRABP2 3.93E+00 2.23E-03
9244 CRLF1 3.99E+00 4.52E-03
26585 GREM1 9.27E+00 7.46E-03
1469 CST1 7.34E+00 1.07E-02
7368 UGT8 3.91E+00 4.27E-02
B, Downregulated
ID No. Symbol Log FC FDR
7123 CLEC3B −4.68128 1.29E-15
762 CA4 −4.78845 1.28E-13
25928 SOSTDC1 −6.45818 4.20E-09
8547 FCN3 −7.00453 9.54E-08
80761 UPK3B −6.08962 1.09E-06
6532 SLC6A4 −6.64863 1.40E-06
4499 MT1M −3.70537 2.70E-05
3569 IL6 −3.92147 1.94E-04
6283 S100A12 −4.04627 3.93E-04
9173 IL1RL1 −5.20882 8.34E-03

FC, fold change; FDR, false discovery rate.

Enriched Gene Ontology database terms of differentially expressed genes.

A, Biological process
GO ID GO term No. of genes P-value
GO:0048583 Regulation of response to stimulus 66 5.00E-07
GO:0071310 Cellular response to organic substance 42 8.33E-07
GO:0070887 Cellular response to chemical stimulus 49 1.36E-06
GO:0032501 Multicellular organismal process 89 1.60E-06
GO:0031347 Regulation of defense response 31 2.00E-06
GO:0003013 Circulatory system process 14 2.79E-06
GO:0003008 System process 43 2.99E-06
GO:0050729 Positive regulation of inflammatory response   6 3.22E-06
GO:0007186 G-protein coupled receptor signaling pathway 32 5.06E-06
GO:1903034 Regulation of response to wounding 30 5.32E-06
GO:0044707 Single-multicellular organism process 87 5.80E-06
GO:0010033 Response to organic substance 65 6.71E-06
GO:0033993 Response to lipid 42 9.57E-06
GO:0002682 Regulation of immune system process 40 9.58E-06
GO:0042221 Response to chemical 75 1.04E-05
B, Molecular function
GO ID GO term No. of genes P-value
GO:0004872 Receptor activity 55 6.43E-07
GO:0038023 Signaling receptor activity 46 1.62E-06
GO:0060089 Molecular transducer activity 58 5.70E-06
GO:0004871 Signal transducer activity 51 2.22E-05
GO:0004888 Transmembrane signaling receptor activity 40 6.35E-05
GO:0005102 Receptor binding 42 2.57E-04
GO:0004908 Interleukin-1 receptor activity   2 4.14E-04
GO:0050998 Nitric-oxide synthase binding   2 5.68E-04
GO:0038187 Pattern recognition receptor activity   3 7.47E-04
GO:0008329 Signaling pattern recognition receptor activity   3 7.47E-04
GO:0050431 Transforming growth factor beta binding   5 9.29E-04
C, Cellular component
GO ID GO term No. of genes P-value
GO:0044459 Plasma membrane part 79 1.35E-05
GO:0005886 Plasma membrane 89 1.12E-04
GO:0031226 Intrinsic component of plasma membrane 52 1.38E-04
GO:0005887 Integral component of plasma membrane 48 3.53E-04
GO:0031526 Brush border membrane   3 5.25E-04
GO:0005576 Extracellular region 47 5.43E-04

GO, Gene Ontology database.

Top 15 enriched KEGG pathways of differentially expressed genes.

KEGG ID KEGG term Count FDR
hsa04512 ECM-receptor interaction 18 1.05E-07
hsa04110 Cell cycle 22 1.18E-07
hsa04115 p53 signaling pathway 14 5.46E-06
hsa04510 Focal adhesion 24 9.43E-06
hsa04514 Cell adhesion molecules   9 7.72E-05
hsa04670 Leukocyte transendothelial migration   9 7.72E-05
hsa03030 DNA replication   9 8.80E-05
hsa04974 Protein digestion and absorption 13 1.04E-04
hsa05200 Pathways in cancer 29 1.71E-04
hsa00512 Mucin type O-Glycan biosynthesis   8 1.71E-04
hsa04114 Oocyte meiosis   9 1.93E-04
hsa00250 Alanine, aspartate and glutamate metabolism   8 2.44E-04
hsa04530 Tight junction   7 2.63E-04
hsa04614 Renin-angiotensin system   6 2.99E-04
hsa04060 Cytokine-cytokine receptor interaction 24 3.43E-04

KEGG, Kyoto Encyclopedia of Genes and Genomes; FDR, false discovery rate; ECM, extracellular matrix.

Top ten transcription factors interacting with the greatest number of differentially expressed genes.

TF P-value Up/down Count Genes
SOX10 9.43E-02 Up 109 ANP32E, RIBC2, GRIP1, ADH1B, RETN, FEN1, HOXA5, PAFAH1B3, MELK, DDX6, BAIAP2L1, SYT12, TNNC1, DHFR, CLMN, PLAC9, FOXM1, CDC20, FAM3B, SLC6A4, CELSR3, KLHL12, ANGPTL7, TMEM100, TDO2, EYA2, IGSF10, SOSTDC1, ANLN, EGLN3, CLEC4E, RAMP1, ADAM8, CAPN10, KRT19, RBM3, C6, PABPC3, ARHGEF15, MEA1, KLF4, DENND2A, SPDEF, AQP5, TRIM24, RAD51AP1, GCGR, CBX5, VANGL1, SLCO5A1, IQGAP3, MARS, PTPRF, LILRA2, MYOM2, CD300LG, ECM2, CLDN12, KCNQ3, LPGAT1, CLEC4M, CX3CL1, CR2, HABP2, ATP8B3, MCM7, ATAD2, PTGFRN, CYP7B1, CDCA7, MAP7D2, DLG3, MUC1, RAB34, RRM2, CLDN4, GIPR, DPP3, ISLR, CENPF, ACP2, MORC2, PDIA6, UCN, XPO5, SUSD4, ACY3, PRPS2, BLM, TMEM45B, RBBP8, TNFRSF17, TMEM63C, VWA1, XPR1, PPP1R1B, PROK2, BRCA1, TMEM88, HMGA1, ACE2, PLOD2, TNFRSF9, PRR15, NMU, BCL9, PYCRL, ZBED2, CLDN2
SPIB 6.34E-02 Up 108 PANX2, PTPRF, CEP55, TLCD1, PBK, NIPSNAP1, ADAMTSL3, ARHGEF19, TSPAN15, ENTPD3, ATR, CSF3, GALNT7, HGD, UBE2T, NOS1, MR1, SLC27A2, CDCA2, T, GIMAP8, MAGED1, TOM1L1, PTPRH, PLXNB3, LYPD1, POLQ, RGS17, DPP6, HMGB3, SHMT2, CST4, CYP4B1, CTTN, FRK, ALG8, KL, PITX1, TARDBP, ERO1L, NEK6, CCNL2, WFIKKN1, RXRG, PSD3, RPN2, RAB26, RER1, PLA2G7, C6, GDF10, SMPDL3B, MRPS24, CELSR1, RHBDL2, DUOX1, PDZK1IP1, TIMP3, MRM1, ADRB2, TSPAN1, LOXL1, APLN, IGFBP3, PRSS16, GLRX2, NME1, SYT12, GALNT13, TMC5, RAD51AP1, POU2AF1, SAMD10, SFT2D2, UBE2J2, RHBDL1, SELE, MAPK8IP2, TMEM132A, GATA2, HN1, STEAP2, MAPKAPK2, FCN1, PSPH, MGRN1, TMPO, GPC3, GDPD1, GPR56, MARS, LRRTM2, GUCY1A2, GIMAP6, SLC41A2, CEACAM1, STK31, TNFRSF9, CCNA2, NCAPG2, HOXC6, AP1S1, DENND1B, GCGR, STIL, ARX, CD24, CXXC5
NR4A2 1.45E-02 Down 81 CABYR, AQP3, RTN4RL2, MORC2, DPP6, SEC14L2, CAND1, MMP7, TRHDE, PLEKHA8, ANGPTL1, DHTKD1, PLEKHA7, MAMDC2, NRIP3, FAM83A, ARHGAP26, LIPH, TMC4, RGS17, ENTPD6, NME1, GALNT10, PTPRF, MYOM2, LYPD1, DNAJC3, COX4I2, DNMT3B, GALNT7, AGR2, GJB1, TMEM97, MGRN1, AGT, SRD5A1, LIMK1, TMEM61, CEACAM1, MT2A, CYP2J2, SFT2D2, GRHL1, DSC2, PTGES, FHL5, CYP24A1, TMEM53, FAM81A, CREB3L4, OSMR, SLC7A10, CXCL13, DNASE1L3, GPR110, PDZK1IP1, AZGP1, RAMP1, FAM3B, SPOCK2, RCN3, MS4A1, NR4A3, C6, CXCL14, ADORA3, HIST1H2BK, MYOCD, ANKRD36, WIF1, RNPC3, NOSTRIN, DLK1, FCRL5, IL1RL1, PPIL2, FGF14, XPR1, NTRK3, SH2D3C, ARHGEF19
FOXD1 8.42E-02 Up 68 PAICS, TNPO1, PRR11, HRASLS, ZNF253, DSC2, COL22A1, EYA2, PRKDC, ZDHHC9, DDHD1, HSF2BP, DPP3, SPP1, ABCA4, BHMT2, ADH1B, AGT, PLA2G4A, DLG5, RANBP9, UHRF1, CLEC1A, TNS4, ADAM28, LDOC1, ANLN, NNT, ZNF567, CCNB1, GUCY1A2, NHS, ARSE, CTHRC1, CD24, PTGFRN, CHI3L1, CASC5, PYGB, ATF3, PMAIP1, HOXC6, HIST1H4J, LYPLA1, GPR25, KIAA2022, PRC1, HN1, GLRX2, IGFBP3, TSPAN15, PDK1, WDR66, ACP2, KPNA2, PCP4, B4GALNT3, SLC14A1, GPR160, HGD, FGF14, P4HA1, SH3GL3, OSMR, GRAMD3, LRRN3, KLHL17, BCHE
ELF5 5.48E-02 Up 59 PTPRB, RNF24, POLE2, PLXNB3, ANKRD36, TMEM106C, CLIC5, FA2H, APOBEC3B, SGPP2, GNG4, ZNF331, TSPAN18, DAP3, SORL1, FAM81A, TOM1L1, STRA6, LDB2, DAP, GTF2A1, VGLL3, LAD1, BTNL9, MMP7, PHF14, TESC, ABCA4, XRCC4, TARBP1, SMOX, SOCS7, GRHL1, STK32A, PLS1, PODXL2, VWF, TEX9, IGF2BP3, CD2AP, KIAA0319L, FLJ34503, PSMD11, CCT5, GCLC, CEP55, RAB11FIP4, LGR4, SNCA, IQGAP3, HSD17B6, SCYL3, CALCRL, EYA4, TSNARE1, PLEKHA6, RER1, SH2D1B, ADAM28
HOXA5 4.13E-04 Down 50 NR4A3, CENPF, CENPE, SERPINH1, CYP1B1, IBSP, ICA1, IL2RA, ITGB8, SNCA, SFPQ, ITGA11, CDO1, NDRG4, FANCF, SYTL2, VGLL3, PRELID2, MRPS23, FMO2, DNAJC3, ILDR1, DACH1, SLC28A3, CHMP4C, ADAM12, DCBLD1, TMPRSS4, ANKRD29, MAP3K13, KIF15, SRPK1, DOK5, COL17A1, MARK2, PC, SRM, CASC5, DPP4, ANXA9, PDGFC, SCN7A, CA10, SLCO1A2, FIGF, KCNQ5, UHMK1, GPR56, GRIP1, MAPKAPK2
KLF5 9.34E-02 Up 45 SORD, GPR56, PSD3, HMGA1, SMARCA4, CNFN, TLCD1, PKHD1L1, GYLTL1B, SEZ6L2, TNFRSF9, HSPB6, PTGDS, RECQL4, KRT19, MUC4, SLC4A2, ROBO4, ABCB1, DCBLD1, EEF1A2, PANX2, UHRF1, LDB2, PIAS3, MT2A, CAPN10, FAM65A, EPHB3, S100A4, SLC39A11, GLRX2, MEST, GDF10, UCHL1, GRIA1, HMGB3, HES6, RHOD, ARHGEF16, IL4I1, NR2C2, MRPS24, MCM8, KIAA1522
ESRRA 7.63E-02 Up 27 WFDC2, MFI2, PACSIN1, C1QTNF7, XDH, KCNQ5, RHBDL2, EPHB3, DEPDC1, STK39, CXXC5, SLC2A1, PRRT3, HKDC1, PPIF, STARD8, KLRD1, DLX3, SRPX, SNCA, TRIO, NNT, SEZ6L2, FAM111B, LRBA, NTRK3, REEP6
SREBF1 5.74E-02 Up 18 COL22A1, SLC28A3, DPP10, CIT, ATRX, EPHB2, CCBE1, TMEM87B, DNAJC3, ITGA11, KIF4A, B4GALNT3, TTK, PTPRZ1, ARHGEF15, COL5A1, ANKH, SLCO5A1
REL 2.16E-02 Up 14 CLEC1A, NUP155, LRP8, ANGPT1, ANKH, GHR, KCNQ5, EMP1, FOXA1, TEX9, SLC28A3, SLC30A7, GYLTL1B, LRRC15

Up, upregulated; down, downregulated; SOX10, sex determining region Y-box 10; SPIB, Spi-B transcription factor; NR4A2, nuclear receptor subfamily 4 group A member 2; FOXD1, forkhead box D1; ELF5, E74 like ETS transcription factor 5; HOXA5, homeobox A5; KLF5, kruppel like factor 5; ESRRA, estrogen related receptor α; SREBF1, sterol regulatory element binding transcription factor 1; REL, REL proto-oncogene, NF-κB subunit; TF, transcription factor.