Co-expression network analysis reveals the pivotal role of mitochondrial dysfunction and interferon signature in juvenile dermatomyositis

• Published in: PeerJ (San Francisco, CA) Vol. 8; p. e8611
• Main Authors: Zhong, Danli, Wu, Chanyuan, Bai, Jingjing, Xu, Dong, Zeng, Xiaofeng, Wang, Qian
• Format: Journal Article
• Language: English
• Published: United States PeerJ. Ltd 18.02.2020; PeerJ, Inc; PeerJ Inc
• Subjects: Biochemistry; Bioinformatics; Biological response modifiers; Connectivity; Data collection; Datasets; Deferoxamine; Dermatology; Dermatomyositis; Digital map services; Disease; Diseases; DNA microarrays; Felodipine; Gene expression; Genes; Genomes; Hub genes; Inflammation; Interferon; Internal Medicine; Juvenile dermatomyositis; Mitochondria; Physiological aspects; Principal components analysis; Protein interaction; Protein-protein interactions; Rapamycin; Rheumatology; Small-molecule compounds; Statistics; Trichostatin A; Valproic acid; Weighted gene co-expression network analysis; United States > US; Juvenile dermatomyositis; Hub genes; Small-molecule compounds; Weighted gene co-expression network analysis
• DOI: 10.7717/peerj.8611
• ISSN: 2167-8359

Juvenile dermatomyositis (JDM) is an immune-mediated disease characterized by chronic organ inflammation. The pathogenic mechanisms remain ill-defined. Raw microarray data of JDM were obtained from the gene expression omnibus (GEO) database. Based on the GSE3307 dataset with 39 samples, weighted correlation network analysis (WGCNA) was performed to identify key modules associated with pathological state. Functional enrichment analyses were conducted to identify potential mechanisms. Based on the criteria of high connectivity and module membership, candidate hub genes were selected. A protein-protein interaction network was constructed to identify hub genes. Another dataset (GSE11971) was used for the validation of real hub genes. Finally, the real hub genes were used to screen out small-molecule compounds via the Connectivity map database. Three modules were considered as key modules for the pathological state of JDM. Functional enrichment analysis indicated that responses to interferon and metabolism were dysregulated. A total of 45 candidate hub genes were selected according to the pre-established criteria, and 20 genes could differentiate JDM from normal controls by validation of another external dataset (GSE11971). These real hub genes suggested the pivotal role of mitochondrial dysfunction and interferon signature in JDM. Furthermore, drug repositioning highlighted the importance of acacetin, helveticoside, lanatoside C, deferoxamine, LY-294002, tanespimycin and L01AD from downregulated genes with the potential to perturb the development of JDM, while betonicine, felodipine, valproic acid, trichostatin A and sirolimus from upregulated genes provided potentially therapeutic goals for JDM. There are 20 real hub genes associated with the pathological state of JDM, suggesting the pivotal role of mitochondrial dysfunction and interferon signature in JDM. This analysis predicted several kinds of small-molecule compounds to treat JDM.