Screening Of Differentially Expressed MiRNAs In NSCLC And Bioinformatics Analysis Of Their Target Genes

Objective:The aim of this study is to screen the differentially expressed microRNAs(miRNAs)in non-small cell lung cancer(NSCLC)and analyze the target genes regulated by these miRNAs by bioinformatics,so as to provide a theoretical basis for revealing the pathogenesis of NSCLC,provide new biomarkers for early diagnosis and treatments of lung cancer,and provide potential reference targets for drug research of lung cancer.Methods:Analysis was performed with the GEO database,datasets(GSE)about the differential expression of miRNAs in NSCLC were retrieved after entering the database,the species was set as human,the target datasets were extracted and the original data were downloaded by reading "title,summary,overall design,platforms".R language and affy package were used for quality control,limma package was used for calculating differential expression,and the results were visualized via drawing volcano diagram.The miRNAs with the most significant difference in expression was selected as the research objects,and their target genes were predicted by miRDB database.The intersection of the genes targeted by miRNA with the most significant up-regulation was obtained by Venny 2.1.0,and the Verm map was drawn.The miRNA with the most significant down-regulation was carried out by the same method.The union of the two intersection results was taken as the target genes for further study.David database was used for enrichment analysis of target genes,including GO analysis and KEGG signaling pathway analysis.Go analysis mainly includes three levels,namely molecular function(MF),cell composition(CC)and biological process(BP).Ggplot 2 package was used to draw bubble diagram to visualize the results.In order to screen out the hub genes,STRING database was used to analyze the protein-protein interaction produced by the expression of target genes,and the preliminary PPI network was obtained.The results were mapped to the Cytoscape 3.8.0 software.The degree of each node was calculated by using the Cytohubba plug-in.The PPI network was sorted according to the degree value,and the top five genes were selected as hub genes.Kaplan Meier plotter database was used to analyze the impact of hub genes on the overall survival(OS)of NSCLC,then drew the survival curve.In order to preliminarily predict the expression of hub genes that can significantly affect the prognosis of NSCLC,we used GEPIA2 database to analyze.Results:We obtained GSE17681 and GSE135918,including 46 samples.Samples in GSE17681 were from peripheral blood,and samples in GSE135918 were from cancer tissues and adjacent normal tissues.The results of differential expression analysis showed that 216 miRNAs were up-regulated and 167 down regulated in peripheral blood of NSCLC patients,and 366 miRNAs were up-regulated and 308 down regulated in cancer tissues.miR-199b-3p and miR-20a-5p were up-regulated and mir-19a-5p and mir-4471 were down regulated in peripheral blood and cancer tissues,respectively.There are 88 common target genes of miR-199b-3p and miR-20a-5p,and 32 common target genes of miR-19a-5p and miR-4471.After the combination of the two intersections,120 target genes were obtained,and these 120 target genes were used as follow-up research objects.Go analysis showed that on BP level,the target genes were mainly enriched in GO terms such as regulation of gene transcription,formation of dendrites of neuronal cells(P<0.05),on CC level mainly enriched in GO terms such as cytoplasm,cytosk eleton(P<0.05),and in the MF level mainly enriched in GO terms such as protein activation,transcriptional repression,mRNA 3'-UTR binding(P<0.05).KEGG pathway analysis showed that the target genes were mainly involved in the MAPK pathway,neurotrophin pathway,and so on(P<0.05).After plotting the PPI network,the genes with the top 5 degree scores were selected as hub genes,namely BDNF,IRS1,FBXO32,SORL1,NTRK2,among which BDNF had the highest degree score.Survival analysis showed that lung adenocarcinoma patients with high expression of IRS 1 and FBXO32 had worse OS prognosis(P<0.05),lung squamous carcinoma patients with high expression of BDNF had better OS prognosis(P<0.05),and NSCLC patients with high expression of SORL1 and NTRK2,whether adenocarcinoma or squamous cell carcinoma,had a better OS prognosis(P<0.05).Analysis of hub gene expression levels revealed that FBXO32 expression was elevated in tumor tissues compared with normal tissues in both adenocarcinoma and squamous cell carcinoma,and the difference was statistically significant(P<0.05),NTRK2 expression was significantly down regulated in adenocarcinoma and up-regulated in squamous cell carcinoma with significant differences(P<0.05),while the expression of the remaining genes was not significantly different compared with normal tissues.Conclusions:miR-199b-3p,miR-20a-5p,miR-19a-5p and miR-4471 showed the most significant differential expression changes in NSCLC and may be involved in the pathogenesis of NSCLC,and their common target genes are mainly involved in the regulation of gene transcription and MAPK signaling pathway,which may be the underlying mechanism of their involvement in NSCLC pathogenesis.The expression levels of BDNF,IRS1,FBXO32,SORL1,NTRK2 can have an impact on the overall survival prognosis of NSCLC,and they may play some roles such as oncogenes or tumor suppressors in the pathogenesis of NSCLC,but the specific effects are in contradiction with their expression levels in NSCLC,which need further experimental verification.These miRNAs and their target genes are also expected to be potential novel biomarkers applied in therapeutic drug research,diagnosis and prognosis evaluation.