Open Access

Candidate genes and pathogenesis investigation for sepsis-related acute respiratory distress syndrome based on gene expression profile

  • Min Wang1,
  • Jingjun Yan1,
  • Xingxing He2,
  • Qiang Zhong1,
  • Chengye Zhan1 and
  • Shusheng Li1Email author
Biological Research201649:25

DOI: 10.1186/s40659-016-0085-4

Received: 6 January 2016

Accepted: 4 April 2016

Published: 18 April 2016

Abstract

Background

Acute respiratory distress syndrome (ARDS) is a potentially devastating form of acute inflammatory lung injury as well as a major cause of acute respiratory failure. Although researchers have made significant progresses in elucidating the pathophysiology of this complex syndrome over the years, the absence of a universal detail disease mechanism up until now has led to a series of practical problems for a definitive treatment. This study aimed to predict some genes or pathways associated with sepsis-related ARDS based on a public microarray dataset and to further explore the molecular mechanism of ARDS.

Results

A total of 122 up-regulated DEGs and 91 down-regulated differentially expressed genes (DEGs) were obtained. The up- and down-regulated DEGs were mainly involved in functions like mitotic cell cycle and pathway like cell cycle. Protein–protein interaction network of ARDS analysis revealed 20 hub genes including cyclin B1 (CCNB1), cyclin B2 (CCNB2) and topoisomerase II alpha (TOP2A). A total of seven transcription factors including forkhead box protein M1 (FOXM1) and 30 target genes were revealed in the transcription factor-target gene regulation network. Furthermore, co-cited genes including CCNB2-CCNB1 were revealed in literature mining for the relations ARDS related genes.

Conclusions

Pathways like mitotic cell cycle were closed related with the development of ARDS. Genes including CCNB1, CCNB2 and TOP2A, as well as transcription factors like FOXM1 might be used as the novel gene therapy targets for sepsis related ARDS.

Keywords

Acute respiratory distress syndrome Sepsis Differentially expressed mRNAs Functional enrichment analysis Pathway analysis Transcription factors

Background

Acute respiratory distress syndrome (ARDS) is a potentially devastating form of acute inflammatory lung injury as well as a major cause of acute respiratory failure [1]. Its development leads to a high short-term mortality rate and significant long-term consequences among survivors, such as physical and cognitive impairment [2]. This disease has represented an important and costly public health problem. The predisposing factors of ARDS are numerous and assorted, including sepsis, pneumonia, multiple blood transfusions, lung contusion, aspiration of stomach contents, and drug abuse or overdose [3]. Although researchers have made significant progresses in elucidating the pathophysiology of this complex syndrome over the years [4], the absence of a universal detail disease mechanism up until now has led to a series of practical problems for a definitive treatment [5].

Previous study has indicated that ARDS has a close relation with severe sepsis [6], but little is known about the detail differences between sepsis-related and non-sepsis-related ARDS. Clinical research reveals that sepsis-related ARDS has poorer recovery from lung injury, higher overall disease severity and higher mortality than non-sepsis-related ARDS [7], however, the mechanism leading to the development of ARDS is still unclear.

Over the past few years, considerable work has been done to test the contribution of genetic factors that may increase the risk of developing ARDS. A genome-wide association study by Wang et al. [8] have suggested that the BCL2-associated agonist of cell death (BAD) gene is a candidate gene associated with the development of ARDS. Besides, variants in more than 30 genes have been associated with ARDS [9]. For instance, common genetic variation in the angiopoietin-2 (Ang-2) gene is proved to be associated with increased risk of ARDS [10]. Gong et al. [11] suggested that the mannose binding lectin-2 (MBL-2) deficiency was associated with increased susceptibility to sepsis and ARDS. Additionally, based on the whole genome expression analysis, Kangelaris et al. [12] believe that the exploration of gene expression differences occurring early in the development of sepsis-related ARDS may further reveal the mechanisms of ARDS. Furthermore, based on the gene expression profiling on peripheral blood from ARDS patients, Dolinay et al. [13] found that the inflammasome pathway and its downstream cytokines play critical roles in ARDS development; Wang et al. [14] found that peptidase inhibitor 3 (PI3) may be a useful clinical marker for monitoring the early development of ARDS. Thus, the analysis of potential ARDS related genes and pathways based on gene expression profile may be a breakthrough for the further understanding of ARDS pathological mechanism.

In the present study, a bioinformatics analysis was performed based on a previous mRNA expression profile from patients with sepsis or sepsis-related ARDS, which was provided by Kangelaris et al. [12]. Via the investigation of disease related differentially expressed genes (DEGs) and pathways, we tried to explore the mechanism of the ARDS and to provide valid biological information for further investigation of this devastating disease.

Methods

Affymetrix microarray data

The mRNA expression profile of GSE66890 provided by Kangelaris et al. [12] was downloaded from a public functional genomics data repository Gene Expression Omnibus in National Center of Biotechnology Information, based on the platform of GPL6244 [HuGene-1_0-st] Affymetrix Human Gene 1.0 ST Array [transcript (gene) version] (Affymetrix Inc., Santa Clara, California, USA). This profile included 29 whole blood mRNA samples of patients with sepsis-related ARDS, and 28 whole blood mRNA samples of patients with sepsis alone.

Differential expression analysis

The oligo software [15] in R was used to preprocess the gene expression profile data. The CEL source files were performed background correction, quartile data normalization and calculating expression using robust multi-array average (RMA) algorithm [16] in affy (http://www.bioconductor.org/packages/release/bioc/html/affy.html). The DEGs between sepsis-related ARDS group and sepsis alone control group were analyzed using the limma package (available at http://www.bioconductor.org/packages/release/bioc/html/limma.html) in Bioconductor software [17]. The t test was used to identify the P value. fold change (FC) was calculated. P < 0.05 and |log2FC| ≥ 0.4 were defined to be statistically significant.

Functional enrichment analysis

The database for annotation, visualization and integrated discovery (DAVID, http://david.abcc.ncifcrf.gov/) [18] is a gene functional classification tool that provides a comprehensive set of functional annotation tools for investigators to understand biological meaning behind large list of genes. Gene Ontology (GO, http://www.geneontology.org) [19] function enrichment analysis were performed based on DAVID, which includes three categories: molecular function (MF), biological process (BP) and cellular component (CC). Kyoto encyclopedia of genes and genomes (KEGG, http://www.genome.ad.jp/kegg/) [20] is a database of biological systems which collects the genomic, chemical and systemic functional information. Reactome (http://www.reactome.org) [21] is a free pathway database that provides intuitive bioinformatics tool for basic research, genome analysis, modeling, systems biology and education. KEGG and Reactome pathway enrichment analyses were performed using DAVID as well. P value <0.05 was considered as threshold value for functional enrichment analyses.

Protein–protein interaction (PPI) network construction

Protein–protein interaction (PPI) network are central to most biological processes, which can help to uncover the generic organization principles of functional cellular networks [22]. Search tool for the retrieval of interacting genes/proteins (STRING) [23] is a biological database and web resource of known and predicted protein–protein interactions. In this study, proteins associated with DEGs were selected according to STRING database with combined score >0.4, and then PPI network was visualized using cytoscape (http://www.cytoscape.org/) [24]. Hub-proteins are small number of proteins with many interaction partners, which play an important role in PPI network [25].

Furthermore, to describe the importance of nodes in the PPI network, three methods including degree centrality [26], Betweenness centrality [27] and subgraph centrality [28] were introduced in the present study. The CytoNCA plugin [29] in cytoscape software was used for the calculation of three methods mentioned above. Furthermore, the modules in PPI network were explored using ClusterOne [30] in cytoscape software. P value <2.0E−7 was considered as threshold value for the analysis of modules.

Prediction analysis of transcription factors

To the further study the pathomechanism of ARDS, the analysis between transcription factors and their target genes obtained from PPI network was performed. IRegulon plugin [31] in cytoscape is used to detect transcription factors, motifs and their optimal sets of direct targets from a set of genes. In this study, iRegulon was used to analyze the transcription factors and their related target genes. The minimum identity between orthologous genes was 0.05, while the maximum false discovery rate on motif similarity was 0.001. The normalized enrichment score (NES) >5 was considered as threshold value for the selection of potential relationships.

Literature mining analysis of ARDS related genes

GenCLiP software (version 2.0, http://ci.smu.edu.cn/GenCLiP2.0/confirm_keywords.php) [32] is used to perform literature mining analysis for human genes and networks. In GenCLiP, the module of literature mining gene networks [32] can construct a gene-network for the input genes and generate sub-networks based on the user defined query terms, at the same time calculate the probability of random occurrence of the networks through random simulation. In the present study, the literature mining gene networks module in GenCLiP was used to analyze the co-cited network of the ARDS-related genes in the previous studies, and the input genes came from the key genes in the PPI network.

Resutls

Identification of DEGs

With thresholds of p value <0.05 and |log2FC| ≥ 0.4, a total of 122 up-regulated and 91 down-regulated DEGs were obtained in sepsis with ARDS group compared with sepsis group. The heat map of differentially expressed mRNAs was showed in Fig. 1.
Fig. 1

Heat map for the differentially expressed genes (DEGs). Green represents the low expression level of DEGs; red represents the high expression level of DEGs; blank represents the express is not significant

Functional enrichment analysis

GO functional enrichment analysis showed that the up-regulated DEGs were mainly involved in mitotic cell cycle (BP, GO: 0000278), cytoplasm (CC, GO: 0005737) and enzyme binding (MF, GO: 0019899). The down-regulated DEGs were mainly involved in positive regulation of response to stimulus (BP, GO: 0048584), spindle (CC, GO: 0005819) and signaling pattern recognition receptor activity (MF, GO: 0008329) (Table 1).
Table 1

Results of gene ontology functional enrichment analysis of differentially expressed genes in acute respiratory distress syndrome (ARDS) (Top 12 listed)

  

GO ID

Term

Count

P value

BP

Up-regulate

GO:0000278

Mitotic cell cycle

26

1.70E−09

GO:1903047

Mitotic cell cycle process

24

2.05E−09

Down-regulate

GO:0048584

Positive regulation of response to stimulus

20

1.31E−05

GO:0033029

Regulation of neutrophil apoptotic process

2

4.80E−05

CC

Up-regulate

GO:0005737

Cytoplasm

89

5.94E−07

GO:0005819

Spindle

11

1.56E−06

Down-regulate

GO:0005819

Spindle

5

0.0055

GO:0005829

Cytosol

20

0.0092

MF

Up-regulate

GO:0019899

Enzyme binding

26

4.22E−06

GO:0019900

Kinase binding

12

0.000143101

Down-regulate

GO:0008329

Signaling pattern recognition receptor activity

2

0.002

GO:0038187

Pattern recognition receptor activity

2

0.002

P value <0.05 was considered as threshold values of significant difference

BP biological process, MF molecular function, CC cellular component, GO gene ontology

The results of pathways enrichment analysis were listed in Table 2. KEGG pathway analysis showed that the up-regulated DEGs were mainly enriched in pathways like Cell cycle, and Hematopoietic cell lineage. The down-regulated DEGs were enriched in three pathways, including phagosome, cytosolic DNA-sensing pathway, and hematopoietic cell lineage. Reactome pathway analysis showed that the up-regulated DEGs were mainly enriched in pathways like mitotic prometaphase, and cell cycle, mitotic; the down-regulated DEGs were enriched in pathways like hydroxycarboxylic acid-binding receptors, innate immune System, and immune system.
Table 2

Results of KEGG (Kyoto encyclopedia of genes and genomes) and reactome pathway enrichment analysis of differentially expressed genes in acute respiratory distress syndrome ARDS (top 12 listed)

  

ID

Term

Count

P value

KEGG

Up-regulate

04110

Cell cycle

5

0.0035

04640

Hematopoietic cell lineage

4

0.006

04114

Oocyte meiosis

4

0.0138

Down-regulate

04145

Phagosome

4

0.0031

04623

Cytosolic DNA-sensing pathway

2

0.0216

04640

Hematopoietic cell lineage

2

0.0496

Reactome

Up-regulate

68877

Mitotic prometaphase

8

4.95E−06

69278

Cell cycle, mitotic

15

6.51E−06

2500257

Resolution of sister chromatid cohesion

7

2.81E−05

Down-regulate

3296197

Hydroxycarboxylic acid-binding receptors

2

6.93E−05

168249

Innate immune system

10

0.0003

168256

Immune system

12

0.0011

P value <0.05 was considered as threshold values of significant difference

PPI network analysis

With combined score >0.4, a total of 132 nodes with 290 protein interaction pairs were revealed. The PPI network was constructed based on the protein interaction pairs (Fig. 2). Top 20 genes (hub genes) with higher combined score that respectively evaluated by subgraph centrality, betweenness centrality and degree centrality were listed in Table 3. The results showed that cyclin B2 (CCNB2) had the highest combined score based on the subgraph centrality evaluation. Meanwhile, the topoisomerase II alpha (TOP2A) had the highest combined score in both betweenness and degree centrality evaluations. Furthermore, a sub-network module was obtained from the PPI network (Fig. 3). The result showed that there were 24 genes in the sub-network module, all of which were up-regulated. Interestingly, among the 24 genes, 20 could be found in Table 3.
Fig. 2

Protein-protein interaction network investigation. Red represents the up-regulated gene; Green represents the down-regulated gene

Table 3

Top 20 genes that evaluated by subgraph centrality, betweenness centrality and degree centrality respectively in the protein–protein interaction (PPI) network

Gene name

Subgraph centrality

Gene name

Betweenness centrality

Gene name

Degree centrality

CCNB2

92,020.02

TOP2A

4238.0317

TOP2A

29

CCNB1

91,113.23

HMMR

2151.6792

CCNB1

26

TOP2A

86,794.31

CCNB1

2114.8433

CCNB2

24

BUB1

85,489.84

VCAN

1703.794

BUB1

20

KIF11

72,242.97

BIRC5

1687.0967

BIRC5

20

BIRC5

65,156.9

TFRC

1503.5779

KIF11

18

CENPF

57,783.895

LRRK2

1472.6691

CENPF

15

NUSAP1

57,466.84

TYMS

977.0087

NUSAP1

15

DLGAP5

53,906.652

SPTA1

971.715

DLGAP5

15

NUF2

47,763.83

FLT3

961.2316

NUF2

13

PRC1

46,547.062

MPP7

923.03656

PRC1

13

NCAPG

43,847.594

MME

917.51965

NCAPG

13

CKS2

37,598.816

PNP

807.13245

RRM2

13

CEP55

34,042.55

TFPI

783.21027

CKS2

12

RRM2

32,993.395

RAD23A

767.7287

CEP55

11

TPX2

23,101.832

TUBB1

708.9906

TPX2

9

TYMS

18,167.875

MMP8

698.92865

TYMS

9

CDKN3

14,048.942

CCNB2

688.69775

TFRC

9

KIF14

11,749.522

DDB1

606.85

CDKN3

8

ASPM

11,499.243

LTF

585.95557

ASPM

8

Combined score >0.4 was considered as threshold values of significant difference

Fig. 3

Sub-network module extracted from protein–protein interaction network. Red represents the up-regulated gene

Transcription factor-target gene regulatory network analysis

The transcription factors of the top 35 genes with higher scores in Table 3 were predicted. With NES >4, a total of seven transcription factors [such as forkhead box protein M1 (FOXM1)] and 30 target genes [such as hyaluronan-mediated motility receptor (HMMR)] were revealed in the present regulatory network (Fig. 4).
Fig. 4

Regulatory network for transcription factors and target genes. Green circle represents the down-regulated genes; Red circle represents the up-regulated genes; Blue square represents the transcription factors

Literature mining analysis

The result of literature mining analysis revealed 14 genes that were revealed as the key ARDS related genes (Fig. 5). All these genes were differentially expressed in the present study. CCNB1 and CCNB2 had the highest co-cited times.
Fig. 5

Co-cited network for relations of differentially expressed genes in the previous studies. The number upon on the edge represents the frequency of co-cite for the relation

Discussion

Although intensively and continuously studies have been conducted in ARDS, its mortality is still as high as 30–40 % [33]. In the present study, a bioinformatics analysis between sepsis-related ARDS mRNA samples and human sepsis mRNA samples was performed to explore the mechanism of the ARDS. A total of 122 up-regulated and 91 down-regulated DEGs were obtained. The up- and down-regulated differentially expressed mRNAs were mainly involved in functions like mitotic cell cycle and pathway like cell cycle. PPI network of ARDS analysis revealed 20 hub genes such as CCNB1, CCNB2 and TOP2A. In addition, seven transcription factors were revealed in the transcription factor-target gene regulation network. Furthermore, 14 co-cited genes including CCNB2CCNB1 were revealed in literature mining analysis. These findings may contribute to understanding the development of sepsis-related ARDS.

In the present study, GO functional analysis showed that the up-regulated DEGs, such as CCNB1, CCNB2 and TOP2A, were most significantly assembled in BP related to mitotic cell cycle (GO: 0000278, P = 1.70E−09) (Table 1). Meanwhile, KEGG analysis also showed that the most significant pathway enriched by up-regulated DEGs was cell cycle (04110, count = 5, P = 0.0035) (Table 2). Study has reported that errors in mitosis can either kill a cell through apoptosis or cause mutations which may lead to disease [34]. Therefore, we speculated that some key factors might play important roles in the development of ARDS via taking part in mitosis cell cycle. When and how did these DEGs regulate the process of ARDS? The Reactome pathways analysis showed that mitotic prometaphase (68877, count = 8, P = 4.95E−06) (Table 2) was the most significant pathway enriched by the DEGs. GO analysis showed that both the up- and down-regulated DEGs were assembled in same function like spindle (GO: 0005819). Thus, we speculated that the DEGs might involve in the progression of ARDS via interfering the progress of spindle in premetaphase of mitotic. However, a further investigation is needed to clarify this speculation.

In this study, CCNB1 and CCNB2 were two outstanding ARDS-related genes based on the calculation of subgraph centrality in PPI network. Furthermore, the literature mining analyses showed that CCNB2CCNB1 had the highest co-cited times. As a mitotic cyclin, cyclin B is necessary for the progression of the cells into and out of M phase of the cell cycle [35]. An abnormal cytoplasmic cyclin B1 expression has been found to be associated with a specific T-cell response and cyclin B1-specific immune responses [36]. Importantly, increasing evidences indicate that the immune system plays a key role in lung diseases, including acute lung injury [37]. Activation of the innate immune response by binding of cell injury-associated endogenous molecules to pattern recognition receptors such as the Toll-like receptors on the lung epithelium and alveolar macrophages is now recognized as a potent driving force for ARDS [1]. Taken together, although there is no direct evidence that CCNB1 and CCNB2 are involved in ARDS, we speculated that CCNB1 and CCNB1 might have close relations in the development of ARDS.

Furthermore, TOP2A was also a hub gene had the highest score in betweenness centrality and degree centrality. TOP2A encodes a DNA topoisomerase that controls and alters the topologic states of DNA during transcription. Actually, the enzyme of TOP2A gene is an essential nuclear enzyme involved in processes such as chromosome condensation and chromatid separation during DNA transcription and replication [38]. Recent study reported that its encoding protein TOP2α is responsible for causing genomic DNA damage [39]. Interestingly, DNA damage is implicated in diverse pulmonary disorders, including acute lung injury [40]. Therefore, we speculated that the up-regulation of TOP2A in our study might have a potential relation in the development of ARDS.

The transcription factor-target gene regulation network analysis in this study revealed seven transcription factors including FOXM1. FOXM1 regulates the expression of a large array of G2/M-specific genes including CCNB2, and plays an important role in maintenance of chromosomal segregation and genomic stability [41]. In this study, HMMR was a target gene of FOXM1. HMMR (also identified as CD168) was originally discovered as a soluble protein that altered migratory cell behavior and bound to hyaluronan [42]. HMMR is less well studied than the main hyaluronan receptor of CD44 that has been examined in ARDS secondary to bleomycin injury. Priit et al. [43] have suggested that CD44 plays a role in resolving lung inflammation during the process of ARDS. Although the role of HMMR (CD168) in ARDS has not been studied before, we speculate that HMMR may be related with the progress of ARDS.

Despite of the results obtained above, there were some limitations in this study. Firstly, no verification experiments based on cells or tissues or joint analysis of expression profile data were performed to confirm our results, besides, the sample size was small. Secondly, due to the data themselves, there was no correction for multiple comparisons in the DEGs identification. Thirdly, the mRNA expression profile used in this study was extracted from blood leukocytes, and another important cells in ARDS, such as epithelial and endothelial cells of the lung had not been studied. Therefore, more investigations related to another cells in ARDS with experimental verification and diverse samples are needed in the further study.

Conclusions

In conclusion, the pathways like mitotic cell cycle were closed related with the development of sepsis related ARDS. Genes including CCNB1, CCNB2 and TOP2A, as well as transcription factors FOXM1 may be used as the novel gene therapy targets for ARDS.

Abbreviations

ARDS: 

acute respiratory distress syndrome

Ang-2

angiopoietin-2

VEGF

vascular endothelial growth factor

GEO: 

gene expression omnibus

RMA: 

robust multi-array average

GO: 

gene ontology

MF: 

molecular function

BP: 

biological process

CC: 

cellular component

KEGG: 

Kyoto encyclopedia of genes and genomes

PPI: 

protein-protein interaction

NES: 

normalized enrichment score

CCNB2

cyclin B2

TOP2A

topoisomerase II alpha

FOXM1

forkhead box protein M1

HMMR

hyaluronan-mediated motility receptor

CCNB1

cyclin B1

CCNB2

cyclin B2

Declarations

Authors’ contributions

MW participated in the design of this study, and performed the statistical analysis. JY, XH, QZ and CZ carried out the study, collected important background information, and bioinformatics analysis. SL conceived of this study, and participated in the design and drafted the manuscript. All authors read and approved the final manuscript.

Acknowledgements

None.

Competing interests

The authors declare that they have no competing interests.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Authors’ Affiliations

(1)
Department of Emergency and Intensive Care Unit, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology
(2)
Institute of Liver Diseases, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology

References

  1. Kollef MH, Schuster DP. The acute respiratory distress syndrome. N Engl J Med. 2012;332(14):27–37.Google Scholar
  2. Aeffner F, Bolon B, Davis IC. Mouse models of acute respiratory distress syndrome: a review of analytical approaches, pathologic features, and common measurements. Toxicol Pathol. 2015;43(8):1074–92.View ArticlePubMedGoogle Scholar
  3. Melmed S. Pathogenesis of pituitary tumors. Nat Rev Endocrinol. 2011;7(5):257–66.View ArticlePubMedGoogle Scholar
  4. Eaton S, Martin G. Clinical developments for treating ARDS. Expert Opin Investig Drugs. 2002;11(1):37–48.View ArticlePubMedGoogle Scholar
  5. de Luis Cabezon N, Sanchez Castro I, Bengoetxea Uriarte UX, Rodrigo Casanova MP, Garcia Pena JM, Aguilera Celorrio L. Acute respiratory distress syndrome: a review of the Berlin definition. Rev Esp Anestesiol Reanim. 2014;61(6):319–27.View ArticlePubMedGoogle Scholar
  6. Betrosian A, Papanikolaou M, Balla M, Georgiadis G. ARDS due to Yersinia enterocolitica sepsis in a patient with thalassemia major. Intensive Care Med. 1999;25(11):1338–9.View ArticlePubMedGoogle Scholar
  7. Sheu CC, Gong MN, Zhai R, Chen F, Bajwa EK, Clardy PF, Gallagher DC, Thompson BT, Christiani DC. Clinical characteristics and outcomes of sepsis-related vs non-sepsis-related ARDS. Chest. 2010;138(3):559–67.View ArticlePubMedPubMed CentralGoogle Scholar
  8. Wei Y, Su L, Lin X, Barnett I, Wurfel M, Christiani D, Wang Z. Genome-Wide Association Study and global expression profiling identified the bad gene as associated with acute respiratory distress syndrome (ards). Am J Respir Crit Care Med. 2015;191:A4354.Google Scholar
  9. Matthay MA, Song Y, Bai C, Jones KD. The acute respiratory distress syndrome in 2013. Translational Respir Med. 2013;1(1):1–6.View ArticleGoogle Scholar
  10. Su L, Zhai R, Sheu CC, Gallagher DC, Gong MN, Tejera P, Thompson BT, Christiani DC. Genetic variants in the angiopoietin-2 gene are associated with increased risk of ARDS. Intensive Care Med. 2009;35(6):1024–30.View ArticlePubMedPubMed CentralGoogle Scholar
  11. Gong MN, Zhou W, Williams PL, Thompson BT, Pothier L, Christiani DC. Polymorphisms in the mannose binding lectin-2 gene and acute respiratory distress syndrome. Crit Care Med. 2007;35(1):48–56.View ArticlePubMedPubMed CentralGoogle Scholar
  12. Kangelaris KN, Prakash A, Liu KD, Aouizerat B, Woodruff PG, Erle DJ, Rogers A, Seeley EJ, Chu J, Liu T, Osterberg-Deiss T, Zhuo H, Matthay MA, Calfee CS. Increased expression of neutrophil-related genes in patients with early sepsis-induced ARDS. Am J Physiol Lung Cell Mol Physiol. 2015;308(11):L1102–13.View ArticlePubMedGoogle Scholar
  13. Dolinay T, Kim YS, Howrylak J, Hunninghake GM, An CH, Fredenburgh L, Massaro AF, Rogers A, Gazourian L, Nakahira K. Inflammasome-regulated cytokines are critical mediators of acute lung injury. Am J Respir Crit Care Med. 2012;185(11):1225–34.View ArticlePubMedPubMed CentralGoogle Scholar
  14. Wang Z, Beach D, Su L, Zhai R, Christiani DC. A genome-wide expression analysis in blood identifies pre-elafin as a biomarker in ARDS. Am J Respir Cell Mol Biol. 2008;38(6):724–32.View ArticlePubMedPubMed CentralGoogle Scholar
  15. Carvalho BS, Irizarry RA. A framework for oligonucleotide microarray preprocessing. Bioinformatics. 2010;26(19):2363–7.View ArticlePubMedPubMed CentralGoogle Scholar
  16. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics. 2003;4(2):249–64.View ArticlePubMedGoogle Scholar
  17. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47.View ArticlePubMedPubMed CentralGoogle Scholar
  18. Alvord G, Roayaei J, Stephens R, Baseler MW, Lane HC, Lempicki RA. The DAVID Gene Functional Classification Tool: a novel biological module-centric algorithm to functionally analyze large gene lists. Genome Biol. 2007;8:R183.View ArticlePubMedPubMed CentralGoogle Scholar
  19. Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA. DAVID: database for annotation, visualization, and integrated discovery. Genome Biol. 2003;4(5):P3.View ArticlePubMedGoogle Scholar
  20. Kanehisa M, Goto S. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28(1):27–30.View ArticlePubMedPubMed CentralGoogle Scholar
  21. D’Eustachio P. Reactome knowledgebase of human biological pathways and processes. Methods Mol Biol. 2011;694:49–61.View ArticlePubMedGoogle Scholar
  22. Wu J, Vallenius T, Ovaska K, Westermarck J, Mäkelä TP, Wu Hautaniemi S, Wu J, et al. Integrated network analysis platform for protein-protein interactions. Nat Methods. 2009;6(1):75–7.View ArticlePubMedGoogle Scholar
  23. von Mering C, Huynen M, Jaeggi D, Schmidt S, Bork P, Snel B. STRING: a database of predicted functional associations between proteins. Nucleic Acids Res. 2003;31(1):258–61.View ArticleGoogle Scholar
  24. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498–504.View ArticlePubMedPubMed CentralGoogle Scholar
  25. He X, Zhang J. Why do hubs tend to be essential in protein networks? PLoS Genet. 2006;2(6):e88.View ArticlePubMedPubMed CentralGoogle Scholar
  26. Jeong H, Mason SP, Barabasi AL, Oltvai ZN. Lethality and centrality in protein networks. Nature. 2001;411(6833):41–2.View ArticlePubMedGoogle Scholar
  27. Goh KI, Oh E, Kahng B, Kim D. Betweenness centrality correlation in social networks. Phys Rev E: Stat, Nonlin, Soft Matter Phys. 2003;67(1 Pt 2):017101.View ArticleGoogle Scholar
  28. Estrada E, Rodriguez-Velazquez JA. Subgraph centrality in complex networks. Phys Rev E: Stat, Nonlin, Soft Matter Phys. 2005;71(5 Pt 2):056103.View ArticleGoogle Scholar
  29. Tang Y, Li M, Wang J, Pan Y, Wu FX. CytoNCA: a cytoscape plugin for centrality analysis and evaluation of protein interaction networks. Biosystems. 2015;127:67–72.View ArticlePubMedGoogle Scholar
  30. Bader GD, Hogue CW. An automated method for finding molecular complexes in large protein interaction networks. BMC Bioinformatics. 2003;4:2.View ArticlePubMedPubMed CentralGoogle Scholar
  31. Rekin’s Janky AV, Imrichová H, Van de Sande B, Standaert L, Christiaens V, Hulselmans G, Herten K, Sanchez MN, Potier D, Svetlichnyy D, Atak ZK, Fiers M, Marine JC, Aerts S. iRegulon: from a gene list to a gene regulatory network using large motif and track collections. PLoS Comput Biol. 2014;10(7):e1003731.View ArticlePubMedPubMed CentralGoogle Scholar
  32. Wang JH, Zhao LF, Lin P, Su XR, Chen SJ, Huang LQ, Wang HF, Zhang H, Hu ZF, Yao KT, Huang ZX. GenCLiP 2.0: a web server for functional clustering of genes and construction of molecular networks based on free terms. Bioinformatics. 2014;30(17):2534–6.View ArticlePubMedGoogle Scholar
  33. Ochiai R. Mechanical ventilation of acute respiratory distress syndrome. J Intensive Care. 2015;3(1):25.View ArticlePubMedPubMed CentralGoogle Scholar
  34. Diederichs S, Baumer N, Schultz N, Hamra FK, Schrader MG, Sandstede ML, Berdel WE, Serve H, Muller-Tidow C. Expression patterns of mitotic and meiotic cell cycle regulators in testicular cancer and development. Int J Cancer. 2005;116(2):207–17.View ArticlePubMedGoogle Scholar
  35. Ito M. Factors controlling cyclin B expression. Plant Mol Biol. 2000;43(5–6):677–90.View ArticlePubMedGoogle Scholar
  36. Ersvaer E, Zhang JY, McCormack E, Olsnes A, Ånensen N, M Tan E, Gjertsen BT, Bruserud Ø. Cyclin B1 is commonly expressed in the cytoplasm of primary human acute myelogenous leukemia cells and serves as a leukemia-associated antigen associated with autoantibody response in a subset of patients. Eur J Haematol. 2007;79(3):210–25.View ArticlePubMedGoogle Scholar
  37. Opitz B, van Laak V, Eitel J, Suttorp N. Innate immune recognition in infectious and noninfectious diseases of the lung. Am J Respir Crit Care Med. 2010;181(12):1294–309.View ArticlePubMedGoogle Scholar
  38. Lang AJ, Mirski SE, Cummings HJ, Yu Q, Gerlach JH, Cole SP. Structural organization of the human TOP2A and TOP2B genes. Gene. 1998;221(2):255–66.View ArticlePubMedGoogle Scholar
  39. Kamekawa H, Kurosawa A, Umehara M, Toyoda E, Adachi N. Endogenous factors causative of spontaneous DNA damage that leads to random integration in human cells. Gene Technol. 2013;2(1):1–5.View ArticleGoogle Scholar
  40. Barker GF, Manzo ND, Cotich KL, Shone RK, Waxman AB. DNA damage induced by hyperoxia: quantitation and correlation with lung injury. Am J Respir Cell Mol Biol. 2006;35(3):277–88.View ArticlePubMedPubMed CentralGoogle Scholar
  41. Laoukili J, Kooistra MR, Brás A, Kauw J, Kerkhoven RM, Morrison A, Clevers H, Medema RH. FoxM1 is required for execution of the mitotic programme and chromosome stability. Nat Cell Biol. 2005;7(2):10.View ArticleGoogle Scholar
  42. Ea T. Purification of a hyaluronate-binding protein fraction that modifies cell social behavior. Biochem Biophys Res Commun. 1982;108(3):8.Google Scholar
  43. Priit T, William R, Dianhua V, Jiurong L, Lauren C, Ellen P, Henson PM, Noble PW. Resolution of lung inflammation by CD44. Science. 2002;296(5565):155–8.View ArticleGoogle Scholar

Copyright

© Wang et al. 2016