Identification of key genes in human airway epithelial cells in response to respiratory pathogens using microarray analysis

Abstract

Background: Airway epithelium is the primary target for pathogens. It functions not only as a mechanical barrier, but also as an important sentinel of the innate immune system. However, the interactions and processes between host airway epithelium and pathogens are not fully understood.

Results: In this study, we identified responses of the human airway epithelium cells to respiratory pathogen infection. We retrieved three mRNA expression microarray datasets from the Gene Expression Omnibus database, and identified 116 differentially expressed genes common to all three datasets. Gene functional annotations were performed using Gene Ontology and pathway analyses. Using protein-protein interaction network analysis and text mining, we identified a subset of genes functioned as a group and associated with infection, inflammation, tissue adhesion, and receptor internalization in infected epithelial cells. These genes were further identified in BESE-2B cells in response to Talaromyces marneffei by Real-Time quantitative PCR (qRT-PCR). In addition, we performed an in silico prediction of microRNA-target interactions and examined our findings.

Conclusions: Using bioinformatics analysis, we identified several genes that may serve as biomarkers for the diagnosis or the surveillance of early respiratory tract infection, and identified additional genes and miRNAs that warrant further fundamental experimental research.

Keywords: Bioinformatics analysis; Biomarker; Human airway epithelial cell; Microarray analysis; Respiratory pathogen.

Fig. 1

DEGs in three mRNA microarray datasets identified using GEO2R (P <  0.05). DEGs between normal and infected airway epithelial cells from the GSE3397 (n = 2033), GSE6802 (n = 1241), and GSE48466 (n = 12,950) datasets were identified, and 116 genes were differentially expressed in all three datasets

Fig. 2

The differential expressed protein–protein interaction network and network modules. a Protein and protein interaction (PPI) pairs of the 116 DEGs were constructed using the STRING database, and 71 PPI pairs were derived. b Modules of the PPI network were analyzed using Cytoscape plugin DCOME. Three significant modules containing 12 hub genes were identified. Module 1 comprise CTSS, NOTCH4, IL8, CREB1, TCF3, and SERPINA1, module 2 comprise PTGER3, RGS4, and OPRM1, and module 3 comprise MPP6, FGFR1, and NSUN3

Fig. 3

Relative expression level of hub genes in BEAS-2B cells in response to Talaromyces marneffei. The expression level of mRNAs was performed using qRT-PCR. Results were shown as mean ± SD, * P <  0.05

Fig. 4

The linear relationship between the hub genes retrieved using COREMINE. Ten hub genes were associated with tissue adhesion, receptor internalization, inflammation and infection. The thicker the line, the closer the connection between the two ends