Identification of Key Genes and Construction of miRNA-mRNA Regulatory Networks Associated with Glioma Grade by Bioinformatics Analysis.

INTRODUCTION The malignant transformation of Low-Grade Gliomas (LGGs) represents a critical clinical event, leading to intensified therapeutic regimens, higher morbidity, and premature death. This study aimed to identify Differentially Expressed Genes (DEGs) between LGGs and High-Grade Gliomas (HGGs) and to build a regulatory network involving microRNAs (miRNAs) and messenger RNAs (mRNAs) to explore the mechanisms driving the malignant progression of LGGs.

Methods

In this bioinformatics study, we obtained two microarray datasets, GSE30472 and GSE45921, from the Gene Expression Omnibus (GEO) database, where 24 LGG samples and 53 HGG samples from the Cancer Genome Atlas (TCGA), the Chinese Glioma Genome Atlas (CGGA), and the Human Protein Atlas (HPA) were included for analysis. After combining the datasets and correcting for batch effects, DEGs were identified. Further analyses included Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment, Protein-Protein Interaction (PPI) network construction, and key gene modeling for DEGs with specific biological phenotypes. Key genes were selected, and miRNA-mRNA regulatory networks were constructed using miRWalk 3.0, TargetScan, and miRDB. Univariate logistic regression was used to identify independent prognostic factors for cancer-specific survival, which were then used to construct the nomogram. The nomogram was validated by the concordance index (C-index), Receiver Operating Characteristic (ROC) curve, and calibration curves.

Results

A total of 16 mRNAs were found to be significantly differentially expressed (fold change >1.0, p < 0.05) between LGG and HGG samples. 16 hub genes, AADAT, ABLIM1, ACADSB, BABAM1, BCO2, BIRC5, C1QBP, CADM3, CALU, DAXX, DAZAP1, DBF4, DCTPP1, DEPDC1, AAK1, and C19orf47, were more prominently expressed in HGG than in LGG samples. Six regulatory pathways were identified, including hsa-let-7d-5p-CALU, hsa-let-7d-5p-AAK1, hsa-let-7e-5p-CALU, hsa-let-7e-5p-AAK1, hsamiR- 3666-DEPDC1, and hsa-miR-3666-AAK1, which are implicated in HGG pathogenesis. Finally, the dysregulated expression of the six regulatory miRNA-mRNA axes was validated by the data from the TCGA, CGGA, and HPA.

Discussion

The identified 16 hub genes and associated miRNA-mRNA axes (specifically, let-7d/let-7e to CALU/AAK1 and miR-3666 to DEPDC1/AAK1) point to potential molecular mechanisms underlying the malignancy of LGG into HGG and identify promising biomarkers and therapeutic targets. These findings require further functional validation in vitro and in vivo, as well as confirmation in larger clinical cohorts, to assess their causal role and clinical applicability.

Conclusion

Our research identified crucial hub genes and miRNA-mRNA regulatory pathways associated with glioma oncogenesis, shedding light on the mechanisms of malignant transformation in LGGs. Furthermore, these data may give us new insights into potential therapeutic targets for this devastating disease.