Integrated Analysis of Competitive Endogenous RNA Networks in Acute Ischemic Stroke

Abstract

Background: Acute ischemic stroke (AIS) is a severe neurological disease with complex pathophysiology, resulting in the disability and death. The goal of this study is to explore the underlying molecular mechanisms of AIS and search for new potential biomarkers and therapeutic targets. Methods: Integrative analysis of mRNA and miRNA profiles downloaded from Gene Expression Omnibus (GEO) was performed. We explored differentially expressed genes (DEGs) and differentially expressed miRNAs (DEMirs) after AIS. Target mRNAs of DEMirs and target miRNAs of DEGs were predicted with target prediction tools, and the intersections between DEGs and target genes were determined. Subsequently, Gene Ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) pathway enrichment analyses, Gene set enrichment analysis (GSEA), Gene set variation analysis (GSVA), competitive endogenous RNA (ceRNA) (lncRNA-miRNA-mRNA) network, protein-protein interaction (PPI) network, and gene transcription factors (TFs) network analyses were performed to identify hub genes and associated pathways. Furthermore, we obtained AIS samples with evaluation of immune cell infiltration and used CIBERSORT to determine the relationship between the expression of hub genes and infiltrating immune cells. Finally, we used the Genomics of Drug Sensitivity in Cancer (GDSC) database to predict the effect of the identified targets on drug sensitivity. Result: We identified 293 DEGs and 26 DEMirs associated with AIS. DEGs were found to be mainly enriched in inflammation and immune-related signaling pathways through enrichment analysis. The ceRNA network included nine lncRNAs, 13 miRNAs, and 21 mRNAs. We used the criterion AUC >0.8, to screen a 3-gene signature (FBL, RPS3, and RPS15) and the aberrantly expressed miRNAs (hsa-miR-125a-5p, hsa-miR-125b-5p, hsa-miR-148b-3p, and hsa-miR-143-3p) in AIS, which were verified by a method of quantitative PCR (qPCR) in HT22 cells. T cells CD8, B cells naïve, and activated NK cells had statistical increased in number compared with the acute cerebral infarction group. By predicting the IC50 of the patient to the drug, AZD0530, Z.LLNle.CHO and NSC-87877 with significant differences between the groups were screened out. AIS demonstrated heterogeneity in immune infiltrates that correlated with the occurrence and development of diseases. Conclusion: These findings may contribute to a better understanding of the molecular mechanisms of AIS and provide the basis for the development of novel treatment targets in AIS.

Keywords: acute ischemic stroke; competitive endogenous RNA; differentially expressed genes; differentially expressed miRNAs; functional enrichment analyses; protein-protein interaction; qPCR analysis.

FIGURE 1

Flow chart of overall analysis.

FIGURE 2

Density plots of the dataset samples before and after correction. (A,B) The boxplot of GSE16561 dataset samples before and after correction after removing the inter-batch differences. (C,D) The boxplot of GSE110993 dataset samples before and after correction after removing the inter-batch differences.

FIGURE 3

PCA plots and differential expression of the samples of the data sets after correction. PCA plots of the GSE16561 (A) and GSE110993 (B) datasets after removing the inter-batch differences. Volcano plots of the GSE16561 (C) and GSE110993 (D) dataset; red plot represented upregulation, and blue plot represented downregulation. Heat maps of the GSE16561 (E) and GSE110993 (F) datasets. The color scale represented the abundance of gene expression. The darker the color shade, the higher expression level.

FIGURE 4

Intersected differentially expressed genes and target genes. (A) The intersection of DEGs and miRNA target genes. (B) The intersection of differentially expressed miRNAs and target miRNAs of DEGs.

FIGURE 5

GO/KEGG function enrichment analysis. (A) In GO biological function enrichment analysis, the X horizontal axis represents the proportion of DEGs enriched in GO term, and the color of the dot represents the adjusted p value: the redder the color, the smaller the adjusted p value; the bluer the color, the greater the adjusted p value. The size of the dot represents the amount of enriched mRNA. (B) In KEGG enrichment analysis, the X horizontal axis represents the proportion of DEGs, and the color of the dot represents the corrected p value. (C) GO function enrichment analysis upset chart. The horizontal axis represents the categories of term names enriched by DEGs, and the vertical axis represents the number of DEGs in this term. (D) KEGG function enrichment analysis Upset plot. (E) GO function enrichment analysis circos plot. (E) The outer circle is the information of the corresponding entry gene in the enrichment analysis, and the line is the corresponding enrichment term entry. (F) KEGG function enrichment analysis circos plot.

FIGURE 6

Pathway diagram. (A,B) Two pathway diagrams composed of two major networks are constructed using DEGs.

FIGURE 7

Gene Set Enrichment Analysis (GSEA). (A–J) GSEA enrichment analysis result sub-graph. The upper part of the graph represents the distribution of rank values of all genes after sorting, and the Signal2Noise algorithm is used by default. The lower part of the graph represents the line chart of the gene Enrichment Score, the horizontal axis is each gene in the gene set, and the vertical axis is the corresponding result.

FIGURE 8

GSVA analysis. (A) In the GSVA enrichment analysis of KEGG term entries, the color scale represents the abundance of gene expression, red represents up-regulation, and blue represents down-regulation. The darker the color shade, the higher is the expression level. (B) GSVA analysis of GO term entries.

FIGURE 9

ceRNA interaction and protein-protein interaction analysis. (A) ceRNA network diagram. In the network diagram, red indicates upregulation, blue indicates downregulation, squares indicate lncRNA, triangles indicate miRNA, and circles indicate mRNA. Sankey diagram (B). The three columns include lncRNAs, miRNAs, and mRNAs in order from left to right. The line colors represent different types of gene-gene interactions. (C) Diagram of interaction of differentially expressed proteins. Red indicates increased expression, blue indicates decreased expression, and color intensity indicates different degrees of u-regulation or downregulation. Orange represents the hub genes. (D) Diagram of hub-genes interaction in differentially expressed proteins. (E) MiRNAs targeting mRNAs interaction diagram. Red indicates upregulated expression, blue indicates downregulated expression, and color intensity indicates different degrees of upregulation or downregulation. Orange represents hub genes. (F) Diagram of hub genes interaction in targeted mRNAs.

FIGURE 10

ROC curve of key mRNA and miRNA. (A–G) ROC curve of mRNA and miRNA. The abscissa is specificity, and the ordinate is sensitivity (true positive rate), specificity = 1 (false positive rate) AUC is the area under the ROC curve enclosed by the coordinate axis.

FIGURE 11

Evaluation and visualization of immune cell infiltration. (A) Ungrouped immune cell infiltration map. (B) Immune cell infiltration map between acute cerebral infarction group and control group. (C) Correlation heat map of 22 types of immune cell infiltration. Blue and red indicate positive and negative correlations, respectively. The darker the color, the stronger is the correlation. (D) Immune cell infiltration map between a single sample of acute cerebral infarction group and control group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 12

Correlation between diagnostic markers and immune cell infiltration. (A–M) The linear regression of diagnostic markers and immune cell infiltration level. The horizontal axis indicates the immune cell infiltration level, and the vertical axis indicates the marker expression. The p value is the regression significance level, and R ² is the goodness-of-fit.

FIGURE 13

Network analyses of target genes and transcription factors. (A) The intersection of differentially expressed genes (DEGs), differential miRNA target genes, and transcription factors is determined using Venn diagram analysis. (B) Intersection molecules-miRNA network analysis. The inner ring is the intersection of DEGs and DEmiRTargetGenes, and the outer ring is miRNAs. Red indicates increased expression, blue indicates reduced expression, and color intensity indicates different degrees of upregulation or downregulation. (C) Network diagram of differential lncRNAs, miRNAs, and mRNAs. Red indicates upregulated expression; blue indicates downregulated expression.

FIGURE 14

Drug sensitivity analysis. The IC50 values of different drugs were determined in the control group and the acute cerebral infarction group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, *****p < 0.00001.

FIGURE 15

The relative expression of differentially expressed mRNA and miRNA in HT22. (A) The CCK-8 cell viability assay; (B) FBL (C) RPS3; (D) RPS15 (E) miRNA-143-3p (F) miRNA-148b-3p; (G) miRNA-125b-5p; (H) miRNA-125a-5p. The control group reflects the normal HT22 and the OGE/R group reflects the model group. *p < 0.05, **p < 0.01, ***p < 0.001.