Construction of Potential Idiopathic Pulmonary Fibrosis Related MicroRNA and Messenger RNA Regulatory Network
Idiopathic pulmonary fibrosis (IPF) is a chronic and progressive lung disease characterized by pulmonary fibrosis of unknown etiology, with pathological manifestations of usual interstitial pneumonia. The prognosis for IPF remains poor, with a median survival time of approximately 2 to 3 years after diagnosis. Given the severity of the disease, it is crucial to explore and develop effective therapeutic modalities. MicroRNAs (miRNAs) play a pivotal role in RNA silencing and post-transcriptional gene expression regulation through base pairs of intramolecular complementary sequences of messenger RNA (mRNA). Each miRNA can target multiple genes, and several miRNAs can co-regulate a single gene. These molecules are particularly significant in organ fibrosis, including IPF, and may be related to the pathogenesis of the disease. Constructing a potential miRNA-mRNA regulatory network for IPF can help reveal a more comprehensive molecular mechanism of miRNAs in this condition.
To identify differentially expressed miRNAs (DE-miRNAs) in IPF, three microarray datasets (GSE75647, GSE27430, and GSE13316) from the National Center for Biotechnology Information Gene Expression Omnibus (GEO) Dataset were selected. Fourteen up-regulated candidate DE-miRNAs (miR-31, miR-493, miR-382, miR-410, miR-432, miR-654-3p, miR-127-3p, miR-487b, miR-409-3p, miR-495, miR-369-5p, miR-299-5p, miR-409-5p, miR-154) and six down-regulated candidate DE-miRNAs (miR-184, miR-338-3p, miR-203, miR-326, miR-375, miR-30b) were identified.
FunRich software, an open-access standalone functional enrichment and interaction network analysis tool, was used to predict upstream transcription factors (TFs) of the candidate DE-miRNAs. The top ten TFs for up-regulated DE-miRNAs were POU2F1, EGR1, NOBOX, MEF2A, BARHL1, PDX1, PORA, SP1, HOXA3, and SP4. The top ten TFs for down-regulated DE-miRNAs were EGR1, SP1, ZFP161, POU2F1, FOXD3, SP4, MAFB, MEF2A, NKX2-1, and ARID3A.
The miRNet database, an integrated platform linking miRNAs, targets, and functions, was utilized to predict the downstream target genes of the candidate DE-miRNAs. The downstream target genes for the up-regulated DE-miRNAs included 1285 genes, while the downstream target genes for the down-regulated DE-miRNAs included 1411 genes.
A microarray dataset (GSE92592) focusing on mRNA expression in the GEO Dataset was selected for subsequent analysis. This led to the identification of 1160 up-regulated DE-mRNAs and 1427 down-regulated DE-mRNAs.
An inverse relationship between miRNA and mRNA target gene is widely acknowledged. A combined analysis of 1427 down-regulated DE-mRNAs and 1285 target genes of up-regulated DE-miRNAs resulted in the identification of 49 candidate target genes for up-regulated DE-miRNAs. Similarly, a combined analysis of 1160 up-regulated DE-mRNAs and 1411 target genes of down-regulated DE-miRNAs identified 53 candidate target genes for down-regulated DE-miRNAs.
The Enrichr database, a comprehensive gene set enrichment analysis web server, was used to perform gene ontology (GO) function enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of the screened candidate target genes of DE-miRNAs.
GO biological process (BP) analysis showed that down-regulated candidate target genes (candidate target genes of up-regulated DE-miRNAs) were significantly enriched in negative regulation of DNA binding, regulation of adiponectin secretion, positive regulation of cell migration by vascular endothelial growth factor signaling pathway, and so on. GO molecular function (MF) analysis showed that candidate target genes of up-regulated DE-miRNAs were significantly enriched in transcription corepressor binding, G-protein coupled receptor activity, postsynaptic density protein-95, Drosophila disc large tumor suppressor, and zonula occludens-1 (PDZ) domain binding, and so on. GO cellular component (CC) analysis showed that candidate target genes of up-regulated DE-miRNAs were significantly enriched in integral component of plasma membrane, clathrin-coated endocytic vesicle membrane, clathrin-coated endocytic vesicle, and so on.
GO BP analysis showed that up-regulated candidate target genes (candidate target genes of down-regulated DE-miRNAs) were significantly enriched in positive regulation of gene expression, positive regulation of nucleic acid-templated transcription, negative regulation of neuron apoptotic process, and so on. GO MF analysis showed that candidate target genes of down-regulated DE-miRNAs were significantly enriched in metalloendopeptidase activity, TF activity/RNA polymerase II core promoter proximal region sequence-specific binding, transcriptional activator activity/RNA polymerase II core promoter proximal region sequence-specific binding, and so on. GO CC analysis showed that candidate target genes of down-regulated DE-miRNAs were significantly enriched in platelet alpha granule membrane, platelet alpha granule, dendrite, and so on.
KEGG pathway enrichment analysis was further conducted for candidate target genes of DE-miRNAs. Candidate target genes of up-regulated DE-miRNAs (down-regulated candidate target genes) were significantly enriched in pathways in cancer, signaling pathways regulating pluripotency of stem cells, breast cancer, and so on. Candidate target genes of down-regulated DE-miRNAs (up-regulated candidate target genes) were significantly enriched in bladder cancer, central carbon metabolism in cancer, transcriptional misregulation in cancer, and so on.
The candidate target genes were mapped into the STRING database, a web server to retrieve and display the repeatedly occurring neighborhood of a gene. It was found that 44 candidate target genes of up-regulated DE-miRNAs in the network could have protein interaction and 116 edges represented the interaction between proteins, while 51 candidate target genes of down-regulated DE-miRNAs in the network could have protein interaction and 209 edges represented the interaction between proteins.
The nodes and edges data were input into Cytoscape 3.6.0 software, and the “Network Analyzer” tools were used for topology analysis, leading to the identification of hub genes. The top 12 hub genes in the target genes of up-regulated DE-miRNAs were VEGFA, FOS, IFNG, MAPK4, MAP2, FABP4, PTGDR, GRM3, KLF6, SPRY4, PER2, and HEY1; the top 12 hub genes in the target genes of down-regulated DE-miRNAs were MMP9, SOX2, MMP2, CDH2, TOP2A, HIF1A, MMP1, IGFBP5, CALU, HMGA2, ITGB3, and CLU. According to the miRNA and candidate target gene pairs analyzed above, a link between miRNA and hub genes was established.
The potential IPF related miRNA-mRNA regulatory network includes miR-31-5p-SPRY4, miR-127-3p-MAPK4, miR-382-5p-PTGDR, miR-369-5p-FABP4, miR-409-3p-IFNG, miR-410-3p-KLF6 VEGFA/HEY1, miR-495-3p-VEGFA/PER2, miR-493-5p-FOS, miR-493-3p-MAP2, and miR-487b-3p-GRM3 of up-regulated miRNA and down-regulated mRNA regulatory network; miR-203a-3p-IGFBP5/MMP1/TOP2A, miR-375-CALU/SOX2, miR-326-CLU/HMGA2, and miR-338-3p-CDH2/HIF1A/ITGB3/MMP2/MMP9 of down-regulated miRNA and up-regulated mRNA regulatory network. These regulatory networks have been relatively underreported, especially in IPF-related research.
For further targeted study, the GEO Dataset was used to detect the expression levels of the 24 hub genes. The GSE72073 dataset was selected for subsequent analysis. The results showed that in GSE72073 dataset the expression level of VEGFA was significantly lower in IPF tissues than that in normal tissues, and the expression levels of SOX2, MMP2, CDH2, TOP2A, HIF1A, MMP1, CALU, and ITGB3 were significantly higher in IPF tissues than those in normal tissues. Based on the preliminary validation, a more accurate potential miRNA-mRNA regulatory network contributing to IPF was established, including miR-410-3p/miR-495-3p-VEGFA, miR-375-SOX2/CALU, miR-338-3p-MMP2/CDH2/HIF1A/ITGB3, and miR-203a-3p-TOP2A/MMP1 regulatory pathways, which could be further studied in clinical and basic experiments, combining the predicted TFs, GO enrichment analysis, and KEGG pathways analysis above.
In conclusion, this study reveals a comprehensive potential mechanism of miRNA-mRNA regulatory axes in the pathogenesis of IPF and establishes a potential IPF-related miRNA-mRNA regulatory network, which may help to understand the underlying mechanism and open new research pathways for IPF.
doi.org/10.1097/CM9.0000000000001276
Was this helpful?
0 / 0