Construction and Bioinformatics Analysis of ceRNA Regulatory Networks in Idiopathic Pulmonary Fibrosis

Abstract

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive form of pulmonary fibrosis of unknown etiology. Despite ongoing research, there is currently no cure for this disease. Recent studies have highlighted the significance of competitive endogenous RNA (ceRNA) regulatory networks in IPF development. Therefore, this study investigated the ceRNA network associated with IPF pathogenesis. We obtained gene expression datasets (GSE32538, GSE32537, GSE47460, and GSE24206) from the Gene Expression Omnibus (GEO) database and analyzed them using bioinformatics tools to identify differentially expressed messenger RNAs (DEmRNAs), microRNAs (DEmiRNAs), and long non-coding RNAs (DElncRNA). For DEmRNAs, we conducted an enrichment analysis, constructed protein-protein interaction networks, and identified hub genes. Additionally, we predicted the target genes of differentially expressed mRNAs and their interacting long non-coding RNAs using various databases. Subsequently, we screened RNA molecules with ceRNA regulatory relations in the lncACTdb database based on the screening results. Furthermore, we performed disease and functional enrichment analyses and pathway prediction for miRNAs in the ceRNA network. We also validated the expression levels of candidate DEmRNAs through quantitative real-time reverse transcriptase polymerase chain reaction and analyzed the correlation between the expression of these candidate DEmRNAs and the percent predicted pre-bronchodilator forced vital capacity [%predicted FVC (pre-bd)]. We found that three ceRNA regulatory axes, specifically KCNQ1OT1/XIST/NEAT1-miR-20a-5p-ITGB8, XIST-miR-146b-5p/miR-31-5p- MMP16, and NEAT1-miR-31-5p-MMP16, have the potential to significantly affect IPF progression. Further examination of the underlying regulatory mechanisms within this network enhances our understanding of IPF pathogenesis and may aid in the identification of diagnostic biomarkers and therapeutic targets.

Keywords: Differential expression analysis; Pulmonary fibrosis; ceRNA; lncRNA; miRNA.

Fig. 1

The workflow chart of our study

Fig. 2

Visual presentation of differential gene screening results in the GSE32537 and GSE47460 dataset. a and d UMAP plot. b and e Volcano plots. c and f processed data boxplots. UMAP uniform manifold approximation and projection)

Fig. 3

Visual presentation of differential gene screening results in the GSE32538 and GSE24206 dataset. a and d UMAP plot. b and e Volcano plots. c and f processed data boxplots. UMAP uniform manifold approximation and projection)

Fig. 4

Venn diagrams of the DEmRNAs, a cross areas indicate the upregulated DEmRNAs, b cross areas indicate the downregulated DEmRNAs, and c cross areas indicate the common DEmRNAs, Enrichment analysis, PPI networks, and hub genes about DEmRNAS. d Construction of PPI network of DEmRNAs. e Identified hub genes. (DEmRNA differentially expressed mRNAs, PPI protein–protein interaction, KEGG Kyoto Encyclopedia of Genes and Genomes)

Fig. 5

miRNA-mRNA network. a Cross areas indicate the candidate DEmRNAs in the miRNA-mRNA network. b and c Bubble plot of BP, CC, MF and KEGG pathway analysis for candidate DEmRNAs in the miRNA-mRNA network. d Cross areas indicate the candidate DELncRNAs in the miRNA-lncRNA network. e The Volcano plots of the candidate DElncRNAs. (miRNA microRNA, mRNA messenger, BP biological processes, CC cellular components, MF molecular functions, KEGG Kyoto Encyclopedia of Genes and Genomes, lncRNA long non-coding RNA, DELncRNA differentially expressed lncRNA)

Fig. 6

a Ellipses represent miRNAs, triangles represent mRNAs, and rhomboids represent lncRNAs. b The Sankey plots of the ceRNA networks. Results of enrichment analysis of candidate miRNAs. (ceRNA competing endogenous RNA)

Fig. 7

a and b Top significantly enriched GO terms of candidate DEmiRNAs, including diseases and MFs. c Heatmap of the results of KEGG pathway analysis of candidate DEmiRNAs. (GO Gene Ontology, MF molecular functions, KEGG Kyoto Encyclopedia of Genes and Genomes)

Fig. 8

Correlation between the expression of candidate mRNAs and %predicted FVC (pre-bd). (%predicted FVC (pre-bd): % predicted pre-bronchodilator forced vital capacity)

Fig. 9

Expression levels of candidate DEmRNAs were assessed using qRT-PCR. (a and b): In 3T3 and A549 cells, the expression of ACSL1, FGG, SPRY4, and PEBP was downregulated in the TGF-β1 group compared with the Control group, while the expression of CLDN1, ITGB8, PDGFRA, MMP16, TRIM29, SFRP2, and COL3A1 was upregulated. (c): In C57BL mice, the expression of ACSL1, FGG, SPRY4, and PEBP was downregulated in the BLM group compared with the PBS group, while the expression of CLDN1, ITGB8, PDGFRA, MMP16, TRIM29, SFRP2, and COL3A1 was upregulated. (**P < 0.01, *** P < 0.001, **** P < 0.001). (qRT-PCR quantitative real-time reverse transcriptase polymerase chain reaction, PBS phosphate-buffered saline, BLM bleomycin)