Background: Acne is the most common skin inflammatory condition. The pathogenesis of acne is not fully understood.
Aims: We performed weighted gene co-expression network analysis (WGCNA) to select acne-associated genes and pathways.
Methods: GSE53795 and GSE6475 datasets including data from lesional and nonlesional skin of acne patients were downloaded from the NCBI Gene Expression Omnibus. Differentially expressed genes (DEGs) in lesions were identified following a false discovery rate <0.05 and | log2 fold change | ≥0.5. DEG-associated biological processes and pathways were identified. WGCNA analysis was performed to identify acne-associated modules. DEGs in the acne-associated modules were used for protein-protein interaction (PPI) network construction and Gene Set Enrichment Analysis (GSEA). Acne-associated candidate DEGs and pathways were identified together with items in the Comparative Toxicogenomics Database (CTD).
Results: A total of 2,140 and 1,190 DEGs were identified in GSE53795 and GSE6475 datasets, respectively, including 716 overlapping DEGs with similar expression profiles in the two datasets, which were clustered into 10 consensus modules. Two modules (brown and turquoise, 359 genes) were associated with acne phenotype. Of these 359 DEGs, 254 were enrolled in the PPI network. GSEA showed that these DEGs were associated with chemokine signaling pathway, cytokine-cytokine receptor interaction, and natural killer cell-mediated cytotoxicity. After identification in CTD, one pathway Cytokine-cytokine receptor interaction and 24 acne-associated DEGs, including IL1R1, CXCL1, CXCR4, CCR1, CXCL2 and IL1β, were identified as candidates associated with acne.
Conclusion: Our results highlight the important roles of the proinflammatory cytokines including IL1β, CXCL1, CXCL2, CXCR4, and CCR1 in acne pathogenesis or therapeutic management.
Keywords: Acne; Inflammation; Microarray data; Pathway; Weighted gene co-expression network analysis.
© 2019 S. Karger AG, Basel.