Five immune-gene-signatures participate in the development and pathogenesis of Kawasaki disease.

OBJECTIVE

To screen for immune genes that play a major role in Kawasaki disease and to investigate the pathogenesis of Kawasaki disease through bioinformatics analysis.

METHODS

Kawasaki disease-related datasets GSE18606, GSE68004, and GSE73461 were downloaded from the Gene Expression Omnibus database. Three microarrays were integrated and standardized to include 173 Kawasaki disease samples and 101 normal samples. The samples were analyzed using CIBERSORT to obtain the infiltration of 22 immune cells and analyze the differential immune cells in the samples and correlations. The distribution of the samples was analyzed using principal component analysis (PCA). Immune-related genes were downloaded, extracted from the screened samples and analyzed for differential analysis (different expression genes [DEG]) and weighted gene co-expression network analysis (WGCNA). We constructed coexpression networks, and used the cytohobbe tool in Cytoscape to analyze the coexpression networks and select the immune genes that played a key role in them.

RESULTS

Immune cell infiltration analysis showed that B cells naive, T cells CD8, natural killer (NK) cells activated, and so forth were highly expressed in normal samples. T cells CD4 memory activated, monocytes, neutrophils, and so forth were highly expressed in Kawasaki disease samples. PCA results showed a significant difference in the distribution of normal and Kawasaki disease samples. From the screened samples, 97 upregulated and 103 downregulated immune-related genes were extracted. WGCNA analysis of DEG yielded 10 gene modules, of which the three most relevant to Kawasaki disease were red, yellow, and gray modules. They were associated with cytokine regulation, T-cell activation, presentation of T-cell receptor signaling pathways, and NK cell-mediated cytotoxicity. CXCL8, CCL5, CCR7, CXCR3, and CCR1 were identified as key genes by constructing a coexpression network.

CONCLUSION

Our study shows that we can distinguish normal samples from Kawasaki disease samples based on the infiltration of immune cells, and that CXCL8, CCL5, CCR7, CXCR3, and CCR1 may play important roles in the development of Kawasaki disease.