OncoTargets and Therapy Dovepress open access to scientific and medical research

Open Access Full Text Article ORIGINAL RESEARCH Bioinformatics Analysis to Reveal Potential Differentially Expressed Long Non-Coding RNAs and Associated with Tumour Metastasis in Lung Adenocarcinoma

This article was published in the following Dove Press journal: OncoTargets and Therapy

Xiaojun Yao1,2 Background: Due to the onset of metastases, the survival rate of lung adenocarcinoma Hongwei Zhang2 (LUAD) is still low. In view of this, we performed this study to screen metastasis-associated Shujun Tang2 genes and lncRNAs in LUAD. fi Xinglong Zheng2 Methods: The mRNA and lncRNA expression pro les of 185 metastatic LUAD and 217 Liangshuang Jiang 1 non-metastatic LAUD samples were retrieved from the TCGA database and included in this study. The differentially expressed mRNAs (DEmRNAs) and lncRNAs (DElncRNAs) 1 Department of Thoracic Surgery, The between metastatic samples and non-metastatic samples of LAUD, as well as the cis nearby- Public Health Clinical Center of Chengdu, Chengdu, Republic of China; targeted DEmRNAs of DElncRNAs and the DElncRNA-DEmRNA co-expression network, 2Department of Thoracic Surgery, were obtained. Quantitative real-time polymerase chain reaction (qRT-PCR) was used to Meishan Cancer Hospital, Chengdu, detect the expression levels of selected DEmRNAs. Survival analysis of selected People’s Republic of China DElncRNAs and DEmRNAs was performed. Results: In total, 1351 DEmRNAs and 627 DElncRNAs were screened between the LUAD primary tissue samples and metastatic samples. Then, 194 DElncRNA-nearby-targeted DEmRNA pairs and 191 DElncRNA-DEmRNA co-expression pairs were detected. Except for RHCG and KRT81, the expression of the other six DEmRNAs in the qRT-PCR results generally exhibited the same pattern as that in our integrated analysis. The expression of CRHR2, FAM83A-AS1, FAM83A and Z83843.1 was significantly correlated with the over- all survival time of patients with metastatic LUAD. Conclusion: We speculate that two interaction pairs (FAM83A-AS1-FAM83A and Z83843.1-MATR3) and four genes (CRHR2, UGT2B15, CHGB and NEFL) are closely associated with the metastasis of LUAD. Keywords: lung adenocarcinoma, LUAD, metastasis, TCGA, long non-coding RNA, lncRNA

Introduction Lung cancer is a prevalent malignancy worldwide and is divided into two main Correspondence: Liangshuang Jiang histological subtypes: small-cell lung carcinoma (SCLC) and non-small-cell lung Department of Thoracic Surgery, The Public Health Clinical Center of Chengdu, carcinoma (NSCLC); the latter includes large cell lung cancer, lung squamous cell No. 18, Jingjusi Road, Jinjiang District, carcinoma (LSCC) and lung adenocarcinoma (LUAD), and LUAD is one of the ’ Chengdu 610061, People s Republic of 1 fi China most common types of lung cancer. LUAD is dif cult to diagnose due to its small Tel +86-18980070581 tumour size, rapid invasion and metastasis in the early stage.2 Despite the use of Fax +86-28-84537477 Email [email protected] chemotherapy and radiotherapy for LUAD, the survival rate is still low, and LUAD submit your manuscript | www.dovepress.com OncoTargets and Therapy 2020:13 3197–3207 3197 DovePress © 2020 Yao et al. This work is published and licensed by Dove Medical Press Limited. The full terms of this license are available at https://www.dovepress.com/terms.php – http://doi.org/10.2147/OTT.S242745 and incorporate the Creative Commons Attribution Non Commercial (unported, v3.0) License (http://creativecommons.org/licenses/by-nc/3.0/). By accessing the work you hereby accept the Terms. Non-commercial uses of the work are permitted without any further permission from Dove Medical Press Limited, provided the work is properly attributed. For permission for commercial use of this work, please see paragraphs 4.2 and 5 of our Terms (https://www.dovepress.com/terms.php). Yao et al Dovepress accounts for 90% of cancer-related deaths.3 Importantly, Functional Annotation and – the underlying mechanisms of LUAD cell invasion and Protein Interaction (PPI) Network metastasis are still not fully understood.4 Hence, a better understanding of metastatic progression in LUAD may Construction of DEmRNAs Ontology (GO) classification and Kyoto Encyclopedia result in improved treatment for patients. of Genes and Genomes (KEGG) pathway enrichment ana- Recently, long non-coding RNAs (lncRNAs) have lysis of DEmRNAs were conducted with CPDB (http:// become novel and popular topics in the field of cancer cpdb.molgen.mpg.de/CPDB). A value of p <0.01wasset treatment. Genome-wide transcriptomic studies over the as the cut-off for significance. The top 100 up- and down- past decade have generated a large quantity of informa- tion about lncRNAs, facilitating the understanding of the regulated DEmRNAs were searched with BioGrid (http:// aetiology of diverse cancers, including LUAD, at the www..org/database/DB-0184). Then, PPI networks molecular level.5 It has been reported that highly were constructed with Cytoscape software (version 3.6.1, expressed SPRY4-IT1 promotes tumour cell migration http://www.cytoscape.org). and invasion in LUAD.6 A group of seven-lncRNAs was reported to predict survival in the early stage of Cis Nearby-Targeted DEmRNAs of the LUAD.7 DKFZP434 L187 and LOC285548 may serve DElncRNAs 8 as prognostic markers for LUAD patients. Investigation DEmRNAs transcribed within a 100-kb window upstream or of disease-associated genes can improve the understand- downstream of DElncRNAs, which were defined as cis ing of disease aetiology and development, thereby facil- nearby-targeted DEmRNAs of DElncRNAs, were searched itating the design and development of novel preventive to obtain the targeted DEmRNAs of DElncRNAs with cis- and treatment strategies. regulatory effects. GO classification and KEGG enrichment In this work, we performed a study, involving 185 analysis of cis nearby-targeted DEmRNAs of DElncRNAs metastatic LUAD and 217 non-metastatic LAUD sam- were performed by using CPDB with p-value < 0.01. ples collected from the TCGA database, to investigate metastasis-associated genes and lncRNAs in LUAD. In addition to the acquisition of differentially expressed DElncRNA-DEmRNA Co-Expression mRNAs (DEmRNAs) and lncRNAs (DElncRNAs), Network identification of cis nearby-targeted DEmRNAs of Pearson’s correlation coefficient (PCC) of each DElncRNA- DElncRNAs and construction of DElncRNA-DEmRNA DEmRNA pair was calculated. DElncRNA-DEmRNA pairs co-expression network were performed. In doing this with an absolute value of PCC > 0.7 and p-value < 0.01 were work, we expected to contribute to expanding the defined as co-expressed DElncRNA-DEmRNA pairs. By knowledge on LUAD. using Cytoscape, DElncRNA-DEmRNA co-expression net- works were constructed. Materials and Methods Data Acquisition and Analysis of mRNA Quantitative Real-Time Polymerase Chain and lncRNA Expression Profiles Reaction (qRT-PCR) Validation The mRNA and lncRNA expression data, including data Thirteen samples were collected from 7 patients with non- for 185 metastatic LUAD and 217 non-metastatic LAUD metastatic LUAD and 6 patients with metastatic LUAD. We samples, were collected from the TCGA database (http:// obtained written informed consent from every participant. tcga-data.nci.nih.gov/). Table 1 displays the clinical char- The present study was approved by the Medical Ethics acteristics of all these patients. With DESeq2 (http://bio Committee of Meishan Cancer Hospital (201,908) and in conductor.org/packages/DESeq2/) in R version 3.5.1, both accordance with the 1964 Helsinki declaration and its later DEmRNAs and DElncRNAs between metastatic LUAD amendments or comparable ethical standards. Total RNA and non-metastatic LAUD were acquired with false dis- was isolated with Trizol reagent (Invitrogen, USA). The covery rate (FDR) < 0.01. By using the R package “pheat- qRT-PCR reactions were performed based on SuperReal map”, hierarchical clustering analysis of DEmRNAs and PreMix Plus (Invitrogen, USA) in an ABI 7500 Real-time DElncRNAs was conducted. PCR Detection System. With the 2-ΔΔCt method, relative

3198 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2020:13 DovePress Dovepress Yao et al

Table 1 Clinical Characteristics of LUAD Patients

Metastatic (n=185) Non-Metastatic (n=217) p-value Overall (n=402)

Age 0.651 Mean (SD) 64.5 (9.80) 65.3 (10.3) 64.9 (10.1) Median [Min, Max] 65.0 [40.0, 87.0] 67.0 [33.0, 86.0] 66.0 [33.0, 87.0] Missing 7 (3.8%) 12 (5.5%) 19 (4.7%)

Race 0.251 American Indian or Alaska Native 1 (0.5%) 0 (0%) 1 (0.2%) Asian 3 (1.6%) 3 (1.4%) 6 (1.5%) Black or African American 21 (11.4%) 15 (6.9%) 36 (9.0%) White 128 (69.2%) 166 (76.5%) 294 (73.1%) Missing 32 (17.3%) 33 (15.2%) 65 (16.2%)

Gender 0.519 Female 92 (49.7%) 116 (53.5%) 208 (51.7%) Male 93 (50.3%) 101 (46.5%) 194 (48.3%)

Pathologic tumor < 0.001 T1 36 (19.5%) 81 (37.3%) 117 (29.1%) T2 117 (63.2%) 114 (52.5%) 231 (57.5%) T3 17 (9.2%) 17 (7.8%) 34 (8.5%) T4 13 (7.0%) 5 (2.3%) 18 (4.5%) TX 2 (1.1%) 0 (0%) 2 (0.5%)

Pathologic node <0.001 N0 11 (5.9%) 217 (100%) 228 (56.7%) N1 95 (51.4%) 0 (0%) 95 (23.6%) N2 74 (40.0%) 0 (0%) 74 (18.4%) N3 2 (1.1%) 0 (0%) 2 (0.5%) NX 3 (1.6%) 0 (0%) 3 (0.7%)

Pathologic_metastasis <0.001 M0 123 (66.5%) 217 (100%) 340 (84.6%) M1 25 (13.5%) 0 (0%) 25 (6.2%) MX 36 (19.5%) 0 (0%) 36 (9.0%) Missing 1 (0.5%) 0 (0%) 1 (0.2%)

Smoking 0.772 Current reformed smoker for >15 years 43 (23.2%) 57 (26.3%) 100 (24.9%) Current reformed smoker for ≤15 years 62 (33.5%) 70 (32.3%) 132 (32.8%) Current reformed smoker, duration not specified 1 (0.5%) 2 (0.9%) 3 (0.7%) Current smoker 44 (23.8%) 46 (21.2%) 90 (22.4%) Non-smoker 25 (13.5%) 38 (17.5%) 63 (15.7%) Missing 10 (5.4%) 4 (1.8%) 14 (3.5%)

Stage <0.001 Stage I 0 (0%) 183 (84.3%) 183 (45.5%) Stage II 79 (42.7%) 25 (11.5%) 104 (25.9%) Stage III 79 (42.7%) 5 (2.3%) 84 (20.9%) Stage IV 25 (13.5%) 0 (0%) 25 (6.2%) Missing 2 (1.1%) 4 (1.8%) 6 (1.5%)

Abbreviation: LUAD, lung adenocarcinoma.

OncoTargets and Therapy 2020:13 submit your manuscript | www.dovepress.com 3199 DovePress Yao et al Dovepress gene expression was determined. Human GAPDH and Table 3 Top 10 Up- and Down-Regulated DElncRNAs in Tumor ACTB were treated as endogenous controls in the analysis. Tissues Between Metastatic LUAD and Non-Metastatic LUAD Symbol p-value FDR Regulation

Survival Analysis of Selected DElncRNAs AP005131.6 7.72E-14 9.77E-11 Up and DEmRNAs Z83843.1 3.87E-13 3.92E-10 Up To further investigate the prognostic value of selected ENSG00000279038 1.17E-12 9.90E-10 Up AC004594.1 1.94E-12 1.38E-09 Up DElncRNAs and DEmRNAs, survival analysis was per- AL022323.4 2.18E-12 1.38E-09 Up formed by using the survival package (https://cran.r-pro AC087521.2 3.68E-11 1.78E-08 Up ject.org/web/packages/survival/index.html)inR. MAL2-AS1 3.86E-11 1.78E-08 Up ENSG00000272367 4.36E-11 1.84E-08 Up Results AL139022.1 6.48E-11 2.53E-08 Up fi AC005070.3 7.65E-11 2.77E-08 Up Identi cation of DEmRNAs and ERVH48-1 1.03E-19 5.21E-16 Down DElncRNAs in Tumour Tissues Between LINC00707 3.25E-14 7.93E-11 Down AC022784.1 4.70E-14 7.93E-11 Down Metastatic LUAD and Non-Metastatic CASC15 1.12E-11 6.28E-09 Down LAUD LINC00592 1.79E-10 5.33E-08 Down In total, 1351 DEmRNAs (882 up- and 469 down-regulated ENSG00000230439 1.48E-09 3.25E-07 Down DEmRNAs) and 627 DElncRNAs (493 up- and 134 down- AC019171.1 1.48E-08 1.98E-06 Down FAM83A-AS1 3.23E-08 3.64E-06 Down regulated DElncRNAs) were screened out between the AL590428.1 5.28E-08 5.14E-06 Down LUAD primary tissue samples and metastatic samples. The H19 5.42E-08 5.18E-06 Down top 10 up- and down-regulated DEmRNAs and DElncRNAs Abbreviations: DElncRNAs, differentially expressed lncRNAs; LUAD, lung ade- are shown in Tables 2 and 3, respectively. Hierarchical clus- nocarcinoma; FDR, false discovery rate. tering analysis of top 100 up- and down-regulated DEmRNAs and DElncRNAs is shown in Figure 1A and B, respectively. Functional Annotation and PPI Network of DEmRNAs Table 2 Top 10 Up- and Down-Regulated DEmRNAs in Tumor Regulation of cellular process (p = 8.84E-06), aromatic com- Tissues Between Metastatic LUAD and Non-Metastatic LUAD pound biosynthetic process (p = 1.05E-04), extracellular ID Gene p-value FDR Regulation matrix (p = 4.31E-07), ion binding (p = 4.86E-12) and cation fi 5225 PGC 4.29E-31 7.06E-27 Up binding (p = 4.56E-10) were the signi cantly enriched GO 213 ALB 6.90E-21 5.67E-17 Up terms in metastatic samples (Figure 2A–C). Glycolysis/ 8707 B3GALT2 3.66E-18 1.51E-14 Up Gluconeogenesis (p = 7.04E-04), p53 signaling pathway (p = 1395 CRHR2 6.04E-18 1.99E-14 Up 2.29E-03) and DNA replication (p = 2.77E-03) were signifi- 11,197 WIF1 2.23E-17 6.10E-14 Up cantly enriched KEGG pathways in metastatic samples 7366 UGT2B15 2.08E-13 3.10E-10 Up (Figure 2D). The PPI network included 102 nodes and 81 200,010 SLC5A9 4.40E-13 5.17E-10 Up 405,754 ERVFRD-1 6.82E-13 7.47E-10 Up edges. ALB (degree = 19), MATR3 (degree = 12) and 56,000 NXF3 7.81E-13 7.55E-10 Up CYLD (degree = 6) were three hub of the PPI network 2315 MLANA 1.03E-12 8.93E-10 Up (Figure 3). 1114 CHGB 1.85E-20 1.01E-16 Down 7348 UPK1B 2.66E-17 6.24E-14 Down Cis Nearby-Targeted DEmRNAs of the 50,632 CALY 1.88E-15 3.87E-12 Down 144,568 A2ML1 6.24E-15 1.14E-11 Down DElncRNAs 51,458 RHCG 7.30E-14 1.20E-10 Down A total of 194 DElncRNA-nearby-targeted DEmRNA pairs, 653,140 FAM228A 2.29E-13 3.14E-10 Down involving in 156 DElncRNAs and 166 DEmRNAs, were 4747 NEFL 3.09E-13 3.91E-10 Down detected (Figure 4). Modification by symbiont of host mor- 79,570 NKAIN1 7.48E-13 7.55E-10 Down phology or physiology (p = 5.42E-04), regulation of cellular 3887 KRT81 8.41E-13 7.68E-10 Down 171,177 RHOV 1.38E-12 1.13E-09 Down metabolic process (p = 9.76E-04), nucleus (p = 2.10E-04), SUMO ligase activity (p = 6.86E-04) and DNA binding (p = Abbreviations: DEmRNAs, differentially expressed mRNAs; LUAD, lung adeno- carcinoma; FDR, false discovery rate. 1.01E-03) were the significantly enriched GO terms

3200 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2020:13 DovePress Dovepress Yao et al

Figure 1 DElncRNAs and DEmRNAs in tumor tissues between metastatic LUAD and non-metastatic LAUD. (A) and (B) displayed hierarchical clustering results of top 100 DEmRNAs and DElncRNAs in tumor tissues between metastatic LUAD and non-metastatic LAUD, respectively. Row and column represented DEmRNAs/DElncRNAs and tissue samples, respectively. The color scale represented the expression levels.

(Figure S1A–C). Cholinergic synapse (p = 2.40E-03) and (p = 4.38E-05) and organic cyclic compound binding Aldosterone synthesis and secretion (p = 8.35E-03) were the (p = 6.63E-05) were the significantly enriched GO terms significantly enriched KEGG pathways (Figure S1D). (Figure S2A–C). P53 signaling pathway (p = 5.59E-04), Cell cycle (p = 6.39E-04) and Amphetamine addiction (p = DElncRNA-DEmRNA Co-Expression 6.48E-03) were the significantly enriched KEGG pathways Network (Figure S2D). A total of 191 DElncRNA-DEmRNA co-expression pairs including 56 DElncRNAs and 108 DEmRNAs were QRT-PCR Validation obtained with an absolute value of PCC > 0.7 and p < Eight DEGs, including CHGB, CRHR2, A2ML1, RHCG, 0.01 (Figure 5). segregation (p = 4.43E-14), UGT2B15, NEFL, KRT81 and RHOV, were selected for mitotic cell cycle process (p = 1.52E-13), chromosomal qRT-PCR validation. In our integrated analysis, CRHR2 region (p = 3.90E-09), heterocyclic compound binding and UGT2B15 were up-regulated while the other six

OncoTargets and Therapy 2020:13 submit your manuscript | www.dovepress.com 3201 DovePress Yao et al Dovepress

Figure 2 Significantly enriched GO terms and KEGG pathways of DEmRNAs in tumor tissues between metastatic LUAD and non-metastatic LAUD. (A) BP,biological process; (B) CC, cellular component; (C) MF,molecular function; (D) KEGG pathways. The x-axis shows counts of DEmRNAs enriched in GO terms or KEGG pathways and the y-axis shows GO terms or KEGG pathways. The color scale represented -log p-value.

DEGs were down-regulated in metastatic LUAD. Except on chromosome 8, locus q24.13.9 Numerous studies have for RHCG and KRT81, the expression of the others in the linked FAM83A to lung cancer in recent years. A previous qRT-PCR results generally exhibited the same pattern as study revealed that FAM83A was associated with metas- that in our integrated analysis (Figure 6). tasis and recurrence in lung cancer.10 Li et al identified that FAM83A was overexpressed in both smoking and non- Survival Analysis smoking groups of patients with LUAD.11 Zhang et al Survival analysis was performed to evaluate the prognostic indicated that overexpression of FAM83A may predict value of two DElncRNAs (FAM83A-AS1 and Z83843.1) and poor prognosis in LUAD patients.9 Wang et al demon- six DEmRNAs (FAM83A, MATR3, CRHR2, UGT2B15, strated that increased FAM83A was negatively associated CHGB and NEFL). On the basis of the results of survival with poor survival in LUAD, which was consistent with analysis, the expression of CRHR2 (p = 0.046), FAM83A- our survival analysis result.12 FAM83A-AS1, a natural AS1 (p =0.001), FAM83A (p =0.0021) and Z83843.1 antisense transcript (NAT) RNA, was reported to promote (p =0.036) was significantly correlated with the overall sur- cell proliferation, migration, invasion and the epithelial- vival time of patients with metastatic LUAD (Figure S3A– mesenchymal transition (EMT) in LUAD.13 Shi et al D), while the expression of CHGB (p = 0.0051), CRHR2 (p = reported that FAM83A-AS1 could promote LUAD by 0.047), FAM83A (p = 0.044) and NEFL (p < 0.0001) was elevating FAM83A expression.14 In this analysis, significantly correlated with the overall survival time of FAM83A-AS1 was significantly down-regulated in the patients with non-metastatic LUAD (Figure S3E–H). Of metastatic LUAD group compared with the non- note, CRHR2 and FAM83A were involved not only in meta- metastatic LUAD group, and FAM83A was a nearby- static LUAD but also in non-metastatic LUAD. targeted and co-expressed DEmRNA of FAM83A-AS1. These results may indicate the critical role of the Discussion FAM83A-AS1-FAM83A interaction pair in the metastatic Although LUAD is a common type of lung cancer, its progression of LUAD. mechanisms underlying its oncogenesis and progression Matrin 3 (MATR3) encodes a nuclear matrix protein remain unclear. This present work utilized an analysis to that is widely expressed in various tissues and can bind to identify key dysregulated lncRNAs and genes associated DNA and RNA via zinc finger domains and RNA recogni- with metastatic progression in LUAD. tion motifs.15 Mutation in MATR3 was associated with Family with sequence similarity 83 member amyotrophic lateral sclerosis (ALS).16 Yang et al demon- A (FAM83A), a member of the FAM83 family, is located strated that highly expressed MATR3 was associated with

3202 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2020:13 DovePress Dovepress Yao et al

RCN1 TTYH3 DDX17 RBM12B MON2 DDX5 GFI1B CXorf23 PTPRN MFSD8 PTBP2 ARRB2 MATR3 AIMP2 DNAAF5 FAM208A TRIM13 KRT78 S100A9 PTBP1 SHC1

RC3H1 CYLD S100A8

CABLES1 NCCRP1 NR5A2 KRT14 GPR18 CHGB TK1 ARG1 TTN YWHAG NEFL GRAP2 ALB CLCA2 AGA ATM PHC3 KRT6B DICER1 KRT5 CTSL

CAMTA1 APOA1

SAA2 ABCA6 LBP

SNX22 P3H4 IKZF1 ADM CAPN6 BCL2L10 FGF12 DPYS B3GALT2 OLFM4 NEDD4L LRFN4 BCL2 TMEM41B

GNPTAB BMP7

IGFBP1 CPLX2 SYT5 MGAT5B ABCC2 LRP2BP PI3 ABCC9 TMED6 A2ML1 COLGALT2

SUSD5 STX1A PALM2 FAM83A TUBB3 LRP2 CSTA LDHA KCNJ8 GPATCH2L C1QTNF3

ANLN ZDHHC17 SYNE1 BCL2L13 GAPDH VAPA MRPS24

FSCN1 CCDC7 MUSK UPK1B ANKRD36 VPS13C DDX56

Figure 3 Protein–protein interaction (PPI) networks. The red and blue ellipses represented proteins encoded by up- and down-regulated DEmRNAs in tumor tissues between metastatic LUAD and non-metastatic LAUD. Ellipses with black border were DEmRNAs derived from top 10 up- and down-regulated DEmRNAs in tumor tissues between metastatic LUAD and non-metastatic LAUD.

OncoTargets and Therapy 2020:13 submit your manuscript | www.dovepress.com 3203 DovePress Yao et al Dovepress

ZNF384 USP3-AS1 VIPR1-AS1 ZNF443 AC090948.1 ZNF69 ZNF14

ZNF564 NOP2 HERC1 NKTR AC006064.2 RFTN1 AC090948.3 AC008770.3 ZNF700 AC011477.2 ZNF506 AC010422.4 ZNF490 NCAPD2 GAPDH AC006064.3 AC015914.1 AC006059.1 PLEKHG6 ZNF709 AC005840.2 MRPL51 AC090948.2 ZNF763 ZNF253 ZNF791 DAPK2 CCDC13

LTBR

LARP4B AC005632.2 AC004039.1 AF131215.6 AC104695.2 AC011503.1 AC124312.5 AL158163.2 AC022509.2 PRICKLE2 ZBTB25

OTOA ITPR2 PSMD6-AS2 AC012557.1 AL049869.3 AL359878.1 KIF3A XKR6 FOSL2 ZNF254 SNURF BBIP1 AC092375.2

PWAR6 AL158163.1 AC024145.1 ATXN7 HSPA2 WDR37 NPIPB4 TH2LCRR AF131215.5 AC093690.1 AC073534.1

FANCI AC005899.7 AC068228.1 RCAN2 VPS18 PURA PIAS2 DCUN1D5 AC036103.1 PLIN3

MIR9-3HG RHBDL3 FAM83A AL035701.1 AC020661.1 AC011379.2 AC090241.3 AP001486.2 GANC AC027319.1

TICAM1 RHCG AC005899.8 FAM83A-AS1 ENPP4 RHOV IGIP ST8SIA5 MMP13 AC012651.1

RIMKLB ZNF451 AGO3 AP000695.2 AL159169.2 SEZ6 MAMDC2 SLFN12L AC116366.2 AVEN

AC015911.3 C5orf56 AC092490.1 ZNF451-AS1 AL138787.2 MORC3 ZDHHC21 AC024267.4

SLFN14 AC116366.1 AC009268.2 MAMDC2-AS1 COL8A2 AP000695.1 AL159169.3 PIPOX A2ML1 BEND6

SBSPON CCDC7 LRRD1 SESN1 SEC14L3 PKMYT1 NPTX1 LRRC15 PHACTR2 KMT2C INTS1 NEDD4L BIRC6 AC079907.1 PTGER3

AC102953.2

AL133243.3 DRAM1 ZRANB2-AS1 AC022893.1 ITGB1-DT AC000120.1 AL390208.1 AC004832.5 AC108134.3 AC124319.2 LINC00887 AL031320.2 AC006017.1 MAFK AC090236.2

IL12B TRIP13 MDM4 GCLC ZC3H7A HTR4 WIF1 ZNF91 MECOM SLC35G2 TUBG1 HHIPL2 CLHC1 KLHL33 KRT18 ARID4A BAIAP2L1 SLC4A5

AC008691.1 AC116351.2 AL512306.3 LINC01564 AC007216.3 AC114939.1 AC026124.2 AC010300.1 AC074033.1 AC096992.2 AC067852.2 AL513314.2 AC012358.1 AL355075.2 AC068888.2 PSMA3-AS1 AC093799.1 DCTN1-AS1

VAPA TSPYL2 KIAA1551 GPR98 RNASEH1 IL11RA APOA1 ZNF33A MMP21 PRR11 SLC25A36 CNTN4 LRMP ZNF217 LRP2BP SNX30 ZNF652 CAPN7

AC006238.1 KANTR AC016957.2 LUCAT1 AC114810.1 AL162231.4 APOA1-AS AL117339.4 EDRF1-AS1 AC099850.3 AC108727.1 AC024060.1 AC023510.2 AL157838.1 AC112722.1 AL139041.1 AC091180.2 SH3BP5-AS1

PA2G4 CBX7 PTPRB SLC2A1 ACHE ADM C1QTNF7 ARHGEF4 NEURL1 EPHX1 AK1 NWD1 ATF7IP2 MFSD8 DMXL2 CYP2A6 DAAM2

AC034102.8 AL031846.2 AC025569.1 SLC2A1-AS1SLC12A9-AS1 AC080023.1 AC098829.1 AC133785.1 AL121929.2 AC099066.2 AL162586.1 AC024075.3 AC027277.2 AC093591.2 AC066613.1 AC092071.1 AL590999.1

TVP23C LRRC66 DRC7 KANSL1 RAB40B TM4SF19 ZNF710 DNAH7 SMAD6 PAX9 TTLL10 CREB3L1 AC007342.5 ZC3H12A CAMK4 ANKRD12

CHD9

NDUFV2-AS1 AC005838.2 AC027271.1 AC092118.1 CR936218.1 AC124283.3 TM4SF19-AS1 AC087284.1 AC114760.2 AC110048.2 AL079303.1 TTLL10-AS1 LINC02489 AC007342.4 LINC01137 STARD4-AS1

HMGN3 KPNA2 TMEM56 WDR4 CACNA1D MAP3K2 LMOD3 LINC01615 SUSD2 LCA5 C12orf76 THBS2 AC007546.1 AC253536.3 AL451064.1 HMGN3-AS1 AC134407.1 AC092802.2 ERVH48-1 AC012467.1 AC010976.1 AC109587.1 LINC02544

Figure 4 DElncRNA-nearby DEmRNA interaction networks. The inverted triangles and ellipses represent DElncRNAs and their nearby DEmRNAs in tumor tissues between metastatic LUAD and non-metastatic LAUD, respectively. Red and blue color represent up- and down-regulated DEmRNAs/DElncRNAs in metastatic LUAD tumor tissues compared with non-metastatic LAUD, respectively. Inverted triangles and ellipses with black border were DElncRNAs/DEmRNAs derived from top 10 up- and down- regulated DElncRNAs/DEmRNAs. poor survival of patients with neuroblastoma and that it most DEmRNAs, including MATR3, in the DElncRNA- may enhance neuroblastoma progression by interacting DEmRNA co-expression network. The discovery of the with lncRNA SNHG1.17 Nho et al suggested that Z83843.1-MATR3 interaction pair reminders us to focus MATR3 was involved in carcinogenesis progression by on its role in the development of LUAD. regulating apoptosis and colony formation in oral squa- Corticotropin-releasing hormone receptor 2 (CRHR2), mous cell carcinoma (OSCC).15 MATR3 was a hub gene located at 7p21-p15, consists of 12 exons, and alternative in the PPI network in our study, although there is no splicing of the first exon gives rise to three isoforms.18 The previous report linking MATR3 with lung cancer. On the protein encoded by this gene is a G-protein coupled recep- other hand, Z83843.1 was significantly up-regulated in the tor that belongs to the subfamily of corticotropin releasing metastatic LUAD group compared with the non-metastatic hormone receptors, with nearly 71% amino acid sequence LUAD group, which was also an lncRNA that covered similarity with CRHR1, and is known as a receptor of

3204 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2020:13 DovePress Dovepress Yao et al

SLFN12L ENSG00000239665 UBAP1L C5orf56 GIMAP5 N4BP2L2-IT2 SH3BP5-AS1

N4BP2L2 GAB3 AC116366.1 AC009133.3 MDM4 GPR18 Z97832.2 TSSK4 CD96 ZBTB37 LINC00861 ZNF80 NLRC3

HNRNPU-AS1 GRAP2 CHRM3-AS2 NEAT1 AC010761.3 SLFN14 TESPA1

LHX4 CAMK4

NKTR ENSG00000272367

ENSG00000277877 ABCD2 ITK AC015911.3 THEMIS GPR174 SEMA6A-AS1 ZNF831 LINC00996

AC010976.1 SREK1 LINC01934 IKZF1 OCLM AC021078.1 SCML4 LINC01215 INO80D ANKRD36 CD226 BTLA MATR3 ANKHD1

EBLN2 Z83843.1 KLRD1 LINC00402 CLDN20 AC007390.2 MALAT1

FO393401.1 AC090154.1 AC092881.1

ERVW-1 GCNT7 AC005070.3

MYSM1

TUBA1B FEN1

PRR11

HMMR DTL MYBL2

TRIP13

COL11A1 AF131215.6 C7 FAM83A-AS1 AC012467.1 MIR143HG INCENP AC099850.3 CENPW MCM6

FAM83A LINC01615 XKR6 CADM3-AS1 ORC1 CACNA1D PDE5A MKI67 HJURP AC068228.1 KPNA2 THBS2 AF131215.5 ADH1B

NCAPD2 AL590999.1 AL035701.1 AL121929.2 PSMA3-AS1

CEP55 LINC02544 ANLN

DAAM2 ENPP4 NEURL1 ARID4A RRM2 ITGA11

GTSE1 EXO1 AC066613.1 AC015914.1 ZNF451-AS1 AL079303.1 AC011477.2 PSMD6-AS2 ZEB2-AS1 PWAR6 AL359878.1

PLK1 CDCA5 DMXL2 DAPK2 ZNF451 PAX9 ZNF506 NR2C2 ZEB2 SNURF LARP4B

SPC25 CDC6 NUSAP1 LINC02489 TM4SF19-AS1 AC133785.1 AC253536.3 AL157838.1 AL049869.3 TTLL10-AS1 SLC2A1-AS1 ARHGAP11A KIF4A CDT1 TMPO-AS1

CCNB1 ARHGEF4 ZBED6 ZNF217 NFAT5 TTLL10 SLC2A1 PRC1 CREB3L1 TM4SF19

FOXM1 OIP5 FANCI

KIF11

Figure 5 DElncRNA-DEmRNA co-expression networks. The inverted triangles and ellipses represent DElncRNAs and their co-expressed DEmRNAs in tumor tissues between metastatic LUAD and non-metastatic LAUD, respectively. Red and blue color represent up- and down-regulated DEmRNAs/DElncRNAs in tumor tissues between metastatic LUAD and non-metastatic LAUD, respectively. Inverted triangles with black border were DEmRNAs/DElncRNAs derived from top 10 up- and down-regulated DEmRNAs/DElncRNAs. corticotropin-releasing hormone (CRH).18 A correlation retinol metabolism.20 Chen et al reported that UGT2B15 between the rs2267716 polymorphism in CRHR2 and was up-regulated in gastric cancer (GC), which may be an hepatocellular carcinoma (HCC) in patients with chronic oncogene in GC.21 Decreased expression levels of hepatitis C virus (HCV) or HBV infection was proposed.18 UGT2B15 were detected in prostate cancer, and were Rodriguez et al reported that decreased CRHR2 promotes associated with lymph node metastases.22 In this study, tumour growth and EMT in colorectal cancer (CRC).19 both CRHR2 and UGT2B15 were significantly up- UDP glucuronosyltransferase family 2 member B15 regulated in metastatic LUAD samples, which may sug- (UGT2B15) encodes a member of the uridine diphosphate- gest that high expression of CRHR2 and UGT2B15 is glucuronosyltransferase (UGT) family, and is involved in involved LUAD metastasis. several metabolic pathways, such as, pentose and glucur- Chromogranin B (CHGB) encodes a tyrosine-sulfated onate interconversion, steroid hormone biosynthesis, and secretory protein that is expressed abundantly in

OncoTargets and Therapy 2020:13 submit your manuscript | www.dovepress.com 3205 DovePress Yao et al Dovepress

Figure 6 The qRT-PCR results of selected DEmRNAs. The x-axis represents the DEmRNAs and the y-axis represents log2 (fold change).

peptidergic endocrine cells and neurons and was first size for validation. Due to this limitation, although the described in pheochromocytoma in 1987.23,24 Decreased qRT-PCR results suggested that our integrated analysis cytoplasmic CHGB expression was found in primary results were generally convincing, the validation results tumours of patients with the pancreatic neuroendocrine were not significant. More samples will be collected and tumour (PNET) phenotype in Von Hippel-Lindau (VHL) more in deep research on functional experiments will be syndrome with metastatic disease compared with those included in our future work. without metastatic disease.25 CHGB was reported to be associated with prolactin pituitary (PRL) tumour Disclosure 26 metastasis. Stenman et al suggested that CHGB may The authors report no conflicts of interest in this work. serve as a possible preoperative and postoperative marker for pheochromocytomas (PCCs) and abdominal paragan- References gliomas (PGLs) with potential for aggressive behaviour.23 1. Qin Y, Zhou X, Huang C, et al. Lower miR-340 expression predicts Neurofilament light (NEFL) encodes the light chain neu- poor prognosis of non-small cell lung cancer and promotes cell rofilament protein, which functionally maintains the neu- proliferation by targeting CDK4. Gene. 2018;675:278–284. doi:10.10 16/j.gene.2018.06.062 ronal caliber and plays a role in intracellular transport to 2. Spiro SG, Tanner NT, Silvestri GA, et al. Lung cancer: progress in 27 axons and dendrites. It is located at chromosome 8p21, diagnosis, staging and therapy. Respirology. 2010;15(1):44–50. a region reported to be enriched with tumour suppressor doi:10.1111/j.1440-1843.2009.01674.x 28 3. Zhang P, Sun J, Kai J, et al. ASAP3 is a downstream target of genes (TSGs). Lower NEFL expression was found in HIF-1alpha and is critical for progression of lung adenocarcinoma. lymph node metastases than in paired primary breast can- Onco Targets Ther. 2019;12:5793–5803. doi:10.2147/OTT.S199603 29 4. Ren W, Mi D, Yang K, et al. The expression of hypoxia-inducible cer tissues. Oh et al reported that the expression level of factor-1alpha and its clinical significance in lung cancer: a systematic NEFL was greatly decreased and may be associated with review and meta-analysis. Swiss Med Wkly. 2013;143:w13855. the course of colorectal cancer (CRC) progression and 5. Wang Y, Wu N, Liu J, Wu Z, Dong D. FusionCancer: a database of 30 cancer fusion genes derived from RNA-seq data. Diagn Pathol. liver metastasis. In addition, methylation of NEFL was 2015;10:131. doi:10.1186/s13000-015-0310-4 reported to be a novel mechanism for NSCLC invasion 6. Zhang X, Chi Q, Zhao Z. Up-regulation of long non-coding RNA 31 SPRY4-IT1 promotes tumor cell migration and invasion in lung and metastasis. Notedly, both CHGB and NEFL were adenocarcinoma. Oncotarget. 2017;8(31):51058–51065. doi:10.1863 significantly down-regulated in metastatic LUAD samples. 2/oncotarget.16918 Hence, we speculated that CHGB and NEFL may contri- 7. Chen M, Liu B, Xiao J, Yang Y, Zhang Y. A novel seven-long non-coding RNA signature predicts survival in early stage lung bute to LUAD progression and are involved in the metas- adenocarcinoma. Oncotarget. 2017;8(9):14876–14886. doi:10.1863 tasis of LUAD. 2/oncotarget.14781 8. Li L, Feng T, Qu J, et al. LncRNA expression signature in prediction In conclusion, a total of 1351 DEmRNAs and 627 of the prognosis of lung adenocarcinoma. Genet Test Mol DElncRNAs were detected in LUAD metastatic samples Biomarkers. 2018;22(1):20–28. doi:10.1089/gtmb.2017.0194 compared with non-metastatic samples. We emphasized 9. Zhang JT, Lin YC, Xiao BF, Yu BT. Overexpression of family with sequence similarity 83, member A (FAM83A) predicts poor clinical the crucial roles of two interaction pairs (FAM83A-AS1- outcomes in lung adenocarcinoma. Med Sci Monit. 2019;25: FAM83A and Z83843.1-MATR3) and four genes 4264–4272. doi:10.12659/MSM.910804 10. Li Y, Dong X, Yin Y, et al. BJ-TSA-9, a novel human tumor-specific (CRHR2, UGT2B15, CHGB and NEFL) in the metastasis gene, has potential as a biomarker of lung cancer. Neoplasia. 2005;7 of LUAD. Our study had the limitation of a small sample (12):1073–1080. doi:10.1593/neo.05406

3206 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2020:13 DovePress Dovepress Yao et al

11. Li Y, Xiao X, Ji X, Liu B, Amos CI. RNA-seq analysis of lung 22. Grosse L, Paquet S, Caron P, et al. Androgen glucuronidation: an adenocarcinomas reveals different gene expression profiles between unexpected target for androgen deprivation therapy, with prognosis smoking and nonsmoking patients. Tumor Biol. 2015;36(11):89 and diagnostic implications. Cancer Res. 2013;73(23):6963–6971. 93–9003. doi:10.1007/s13277-015-3576-y doi:10.1158/0008-5472.CAN-13-1462 12. Wang Y, Lu T, Wo Y, et al. Identification of a putative competitive 23. Stenman A, Svahn F, Hojjat-Farsangi M, et al. Molecular profiling of endogenous RNA network for lung adenocarcinoma using TCGA pheochromocytoma and abdominal paraganglioma stratified by the datasets. Peer J. 2019;7:e6809. doi:10.7717/peerj.6809 PASS algorithm reveals chromogranin B as associated with histologic 13. Xiao G, Wang P, Zheng X, Liu D, Sun X. FAM83A-AS1 promotes prediction of malignant behavior. Am J Surg Pathol. 2019;43 lung adenocarcinoma cell migration and invasion by targeting (3):409–421. doi:10.1097/PAS.0000000000001190 miR-150-5p and modifying MMP14. Cell Cycle. 2019;18:29 24. Benedum UM, Lamouroux A, Konecki DS, et al. The primary struc- – 72 2985. ture of human secretogranin I (chromogranin B): comparison with 14. Shi R, Jiao Z, Yu A, Wang T. Long noncoding antisense RNA chromogranin A reveals homologous terminal domains and a large FAM83A-AS1 promotes lung cancer cell progression by increasing intervening variable region. EMBO J. 1987;6(5):1203–1211. doi:10. – FAM83A. J Cell Biochem. 2019;120(6):10505 10512. doi:10.1002/ 1002/j.1460-2075.1987.tb02355.x jcb.28336 25. Weisbrod AB, Zhang L, Jain M, Barak S, Quezado MM, Kebebew E. 15. Nho SH, Yoon G, Seo JH, et al. Licochalcone H induces the apop- Altered PTEN, ATRX, CHGA, CHGB, and TP53 expression are tosis of human oral squamous cell carcinoma cells via regulation of associated with aggressive VHL-associated pancreatic neuroendo- matrin 3. Oncol Rep. 2019;41(1):333–340. doi:10.3892/or.2018.6784 crine tumors. Horm Cancer. 2013;4(3):165–175. doi:10.1007/s126 16. Johnson JO, Pioro EP, Boehringer A, et al. Mutations in the Matrin 3 72-013-0134-1 gene cause familial amyotrophic lateral sclerosis. Nat Neurosci. 26. Zhang W, Zang Z, Song Y, Yang H, Yin Q. Co-expression network 2014;17(5):664–666. doi:10.1038/nn.3688 analysis of differentially expressed genes associated with metastasis 17. Yang TW, Sahu D, Chang YW, Hsu CL, Hsieh CH. RNA-binding – proteomics reveals MATR3 interacting with lncRNA SNHG1 to in prolactin pituitary tumors. Mol Med Rep. 2014;10(1):113 118. enhance neuroblastoma progression. J Proteome Res. 2019;18 doi:10.3892/mmr.2014.2152 (1):406–416. 27. Leermakers FA, Zhulina EB. How the projection domains of 18. Gu X, Qi P, Zhou F, et al. An intronic polymorphism in the NF-L and alpha- determine the conformations of fi corticotropin-releasing hormone receptor 2 gene increases suscept- NF-M and NF-H in neuro laments. Eur Biophys J. 2010;39 – ibility to HBV-related hepatocellular carcinoma in Chinese (9):1323 1334. doi:10.1007/s00249-010-0585-z population. Hum Genet. 2010;127(1):75–81. doi:10.1007/s00439- 28. Huang Z, Zhuo Y, Shen Z, et al. The role of NEFL in cell growth and 009-0750-6 invasion in head and neck squamous cell carcinoma cell lines. J Oral 19. Rodriguez JA, Huerta-Yepez S, Law IK, et al. Diminished expression Pathol Med. 2014;43(3):191–198. doi:10.1111/jop.12109 of CRHR2 in human colon cancer promotes tumor growth and EMT 29. Feng Y, Sun B, Li X, et al. Differentially expressed genes between via persistent IL-6/Stat3 signaling. Cell Mol Gastroenterol Hepatol. primary cancer and paired lymph node metastases predict clinical 2015;1(6):610–630. doi:10.1016/j.jcmgh.2015.08.001 outcome of node-positive breast cancer patients. Breast Cancer Res 20. Kudryavtseva AV, Lukyanova EN, Kharitonov SL, et al. Treat. 2007;103(3):319–329. doi:10.1007/s10549-006-9385-7 Bioinformatic identification of differentially expressed genes asso- 30. Oh BY, Cho J, Hong HK, et al. Exome and transcriptome sequencing ciated with prognosis of locally advanced lymph node-positive pros- identifies loss of PDLIM2 in metastatic colorectal cancers. Cancer tate cancer. J Bioinform Comput Biol. 2019;17(1):1950003. doi:10.11 Manag Res. 2017;9:581–589. doi:10.2147/CMAR.S149002 42/S0219720019500033 31. Shen Z, Chen B, Gan X, et al. Methylation of neurofilament light 21. Chen X, Li D, Wang N, et al. Bioinformatic analysis suggests that polypeptide promoter is associated with cell invasion and metastasis UGT2B15 activates the HippoYAP signaling pathway leading to the in NSCLC. Biochem Biophys Res Commun. 2016;470(3):627–634. pathogenesis of gastric cancer. Oncol Rep. 2018;40(4):1855–1862. doi:10.1016/j.bbrc.2016.01.094 doi:10.3892/or.2018.6604

OncoTargets and Therapy Dovepress Publish your work in this journal OncoTargets and Therapy is an international, peer-reviewed, open agents and protocols on patient perspectives such as quality of life, access journal focusing on the pathological basis of all cancers, adherence and satisfaction. The manuscript management system is potential targets for therapy and treatment protocols employed to completely online and includes a very quick and fair peer-review improve the management of cancer patients. The journal also system, which is all easy to use. Visit http://www.dovepress.com/ focuses on the impact of management programs and new therapeutic testimonials.php to read real quotes from published authors.

Submit your manuscript here: https://www.dovepress.com/oncotargets-and-therapy-journal

OncoTargets and Therapy 2020:13 submit your manuscript | www.dovepress.com 3207 DovePress