Identification of biomarkers and construction of a microRNA-mRNA regulatory network for ependymoma using integrated bioinformatics analysis

Abstract

Ependymomas (EPNs) are one of the most common types of malignant neuroepithelial tumors. In an effort to identify potential biomarkers involved in the pathogenesis of EPN, the mRNA expression profiles of the GSE25604, GSE50161, GSE66354, GSE74195 and GSE86574 datasets, in addition to the microRNA (miRNA/miR) expression profiles of GSE42657 were downloaded from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) and differentially expressed miRNAs (DEMs) between EPN and normal brain tissue samples were identified using the Limma package in R and GEO2R, respectively. Functional and pathway enrichment analyses were conducted using the Database for Annotation, Visualization and Integrated Discovery. A protein-protein interaction network was constructed using the Search Tool for Retrieval of Interacting Genes database, which was visualized using Cytoscape. The targeted genes of DEMs were predicted using miRWalk2.0 and a miRNA-mRNA regulatory network was constructed. Following analysis, a total of 948 DEGs and 129 DEMs were identified. Functional enrichment analysis revealed that 609 upregulated DEGs were significantly enriched in 'PI3K-Akt signaling pathway', while 339 downregulated DEGs were primarily involved in 'cell junction' and 'retrograde endocannabinoid signaling'. In addition, 6 hub genes [cyclin dependent kinase 1, CD44 molecule (Indian blood group) (CD44), proliferating cell nuclear antigen (PCNA), MYC, synaptotagmin 1 (SYT1) and kinesin family member 4A] and 6 crucial miRNAs [homo sapiens (hsa)-miR-34a-5p, hsa-miR-449a, hsa-miR-106a-5p, hsa-miR-124-3p, hsa-miR-128-3p and hsa-miR-330-3p] were identified as biomarkers and potential therapeutic targets for EPN. Furthermore, a microRNA-mRNA regulatory network was constructed to highlight the interactions between DEMs and their target DEGs; this included the hsa-miR-449a-SYT1, hsa-miR-34a-5p-SYT1, hsa-miR-330-3p-CD44 and hsa-miR-124-3p-PCNA pairs, whose expression levels were confirmed using reverse transcription-quantitative polymerase chain reaction. In conclusion, the present study may provide important data for the investigation of the molecular mechanisms of EPN pathogenesis.

Keywords: differentially expressed gene; differentially expressed miRNA; ependymoma; module; regulatory network.

Figure 1.

Venn diagram of differentially expressed genes among the five datasets.

Figure 2.

Significantly enriched Reactome pathway analysis of differentially expressed genes.

Figure 3.

Protein-protein interaction regulatory network of DEGs in ependymoma. Nodes represent DEGs. Red nodes indicate upregulated DEGs and green nodes indicated downregulated DEGs. Nodes with higher degree values are depicted with larger shapes. Edges/lines stand for the regulatory association between any 2 nodes. DEGs, differentially expressed genes.

Figure 4.

Top 5 modules in the protein-protein interaction network for DEGs. (A) Module 1. (B) Module 2. (C) Module 3. (D) Module 4. (E) Module 5. Nodes represent DEGs. Edges/lines stand for the regulation association between any 2 nodes. Red and green nodes represent upregulated and downregulated genes, respectively. DEGs, differentially expressed genes.

Figure 5.

Regulatory network of upregulated miRNAs and downregulated DEGs in ependymoma. The red triangles represent the upregulated miRNAs. The green circles indicate downregulated DEGs. miRNA, microRNA; DEGs, differentially expressed genes.

Figure 6.

Regulatory network between downregulated miRNAs and upregulated DEGs in ependymoma. The green triangles represent downregulated miRNAs. The red circles indicate upregulated DEGs. miRNA, microRNA; DEGs, differentially expressed genes.

Figure 7.

Boxplots of association analyses. Boxplots of (A) SYT1, (B) CD44 and (C) PCNA in the GSE25604, GSE50161, GSE66354, GSE74195 and GSE86574 datasets. (D) Boxplots of hsa-miR-449a, hsa-miR-34a-5p, hsa-miR-330-3p and hsa-miR-124-3p from the GSE42657. *P<0.05 and **P<0.01. SYT1, synaptotagmin 1; CD44, CD44 molecule (Indian blood group); PCNA, proliferating cell nuclear antigen; hsa, homo sapiens; miR, microRNA.

Figure 8.

ROC analyses for efficacy evaluation. ROC analyses of (A) SYT1, (B) CD44 and (C) PCNA in the GSE25604, GSE50161, GSE66354, GSE74195 and GSE86574 datasets. (D) ROC analyses of hsa-miR-449a, hsa-miR-34a-5p, hsa-miR-330-3p and hsa-miR-124-3p from the GSE42657. AUC, area under the curve; hsa-miR, homo sapiens microRNA; ROC, receiver operating characteristic; SYT1, synaptotagmin 1; CD44, CD44 molecule (Indian blood group); PCNA, proliferating cell nuclear antigen; hsa, homo sapiens; miR, microRNA.

Figure 9.

Boxplots of hub genes and miRs analyzed using reverse transcription-quantitative polymerase chain reaction. Significance between normal (n=10) and EPN tissues (n=10) for genes and miRs, including (A) CD44, (B) PCNA, (C) SYT1, (D) hsa-miR-124-3p, (E) hsa-miR-330-3p, (F) hsa-miR-34a-5p and (G) hsa-miR-449a, was determined using an independent sample t-test. **P<0.01. miRs, microRNAs; hsa, homo sapiens; EPN, ependymoma; CD44, CD44 molecule (Indian blood group); PCNA, proliferating cell nuclear antigen; SYT1, synaptotagmin 1.