Weighted gene co-expression network analysis reveals specific modules and biomarkers in Parkinson’s disease

Abstract

BackgroundParkinson’s disease (PD) ranks as the second most frequently occurring neurodegenerative disease. The precise pathogenic mechanism of this disease remains unknown. The aim of the present study was to identify the biomarkers in PD and classify the primary differentially expressed genes (DEGs). MethodsThe present study searched for and downloaded mRNA expression data from the Gene Expression Omnibus database to identify differences in mRNA expression in the substantia nigra (SN) and blood of patients with PD and healthy controls. In addition, in order to investigate the biological functions of the classified dysregulated genes, the present study utilized Gene Set Enrichment Analysis (GSEA), Gene Ontology (GO), reverse transcription-quantitative PCR (RT-qPCR), gene co-expression network analysis and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. A receiver operating characteristic (ROC) curve was applied to assay TMEM243 as a diagnostic marker. ResultsBetween PD and controls in GSE20292, the present study identified 1862 DEGs. Using the weighted gene co-expression network analysis, the present study identified 15 modules in PD. The module preservation analysis revealed that the tan, blue and green-yellow modules were the most stable. KEGG pathway analysis revealed that five DEGs in the black module were significantly enriched in the ubiquitin-mediated proteolysis pathway, nucleotide excision repair pathway, mismatch repair pathway. The present study selected 303 genes with high connectivity in blue, green-yellow and tan modules as hub genes, where 58 were differentially expressed in both the GSE20292 and GSE54536 datasets. In the SN and blood, 11 genes exhibited the same trend of expression. Furthermore, in the blood samples of patients with PD, the results displayed a significant upregulation of TMEM243. The expression levels of CCR4, CAMK1D, ACTR1B and SPSB3 increased, while both the levels of INA and PSMD4 decreased. These findings are consistent with the bioinformatics analysis results but are not statistically significant. TMEM243 can be considered as a diagnostic biomarker (area under the curve = 0.694; sensitivity, 80 %; specificity, 56 %; P < 0.018). ConclusionTMEM243 was distinctly upregulated in the blood samples of patients with PD, as validated via RT-qPCR, and was highly sensitive, revealing its potential as a biomarker for the future diagnosis of PD.