Objectives: Glioblastoma multiforme (GBM) is defined as the most frequent and lethal form of the primary brain tumors in the central nervous system (CNS) in adults. Recent studies have focused on the identification of the new targets for the diagnosis and treatment of GBM and resulted in great interest for miRNAs due to their regulatory effects in cancer pathogenesis. Thus, we aimed to characterize novel molecular biomarkers for GBM by computational analysis.Methods: 118 miRNAs that are clinically related with glioblastoma and proven by experimentally were exported through miRTarBase database. 1016 genes projected by these 118 miRNAs were determined via ComiR database. Subsequently, the genes with transcribed ultraconserved regions (T-UCRs) in their exonic regions were designated and the genes which have potential competing endogenous RNA (ceRNA) activities were extracted. Genes with remarkable expression profile differences between glioblastoma and normal brain tissues among ceRNAs that are associated with glioblastoma involving T-UCR were identified.
Results:The statistical analysis of the correlation between PBX3 and NRXN3 genes and glioblastoma was carried out by Spearman correlation test. PBX3 and NRXN3 expression was significantly higher and lower in glioblastoma than in normal brain tissues, respectively. On the other hand, the other genes did not have any remarkable differential expression pattern.
Conclusion:Based on the findings of the current study, it is determined that NRXN3 acts as a tumor suppressor gene and NRXN3 gene is downregulated in GBM. PBX3 gene functions as an oncogene and is upregulated in GBM.