Kawasaki disease: integrated bioinformatics and experimental validation reveal novel biomarkers KLHL42 and SOS2

Kawasaki disease (KD) is an acute self-limited vasculitis whose relationship with the gut microbial metabolite butyrate remains unclear. This study aimed to explore the association between Kawasaki disease (KD) and butyrate-related genes. Public databases were utilized to obtain KD-related transcriptomic datasets and butyrate-related genes. Differential expression analysis and weighted gene co-expression network analysis (WGCNA) were applied, identifying 51 candidate genes. Mendelian randomization analysis was conducted to evaluate causal relationships between genes and KD. Expression levels of key genes were validated using public datasets and clinical samples. A nomogram based on identified biomarkers was constructed and its predictive performance was assessed. Functional enrichment analysis was applied to explore biological pathways involved. Immune cell infiltration analysis and single-cell RNA sequencing (scRNA-seq) were used to characterize expression patterns of target genes across cell types and during differentiation. Differential expression analysis and WGCNA identified 51 candidate genes. Mendelian randomization revealed that SOS2 and GAS6 were risk factors for KD, while KLHL42 was a protective factor. Validation in public datasets and clinical samples showed that compared to controls, KLHL42 expression was downregulated and SOS2 expression was upregulated in KD patients. The nomogram constructed based on KLHL42 and SOS2 demonstrated good predictive performance. Functional enrichment analysis indicated that both KLHL42 and SOS2 were significantly associated with the chemokine signaling pathway and correlated closely with infiltration abundances of neutrophils and CD8⁺ T cells. Single-cell RNA sequencing analysis suggested that T cells, B cells, and monocytes were potentially key cell types in KD; expression levels of KLHL42 and SOS2 gradually decreased during T-cell differentiation but increased during B-cell and monocyte differentiation. This study identifies genes associated with butyrate-related host responses in Kawasaki disease, providing clues for further mechanistic investigation. This study employed transcriptomic analysis and Mendelian randomization to reveal the association characteristics between Kawasaki disease and genes related to butyrate. It identified KLHL42 and SOS2 as biomarkers, with SOS2 determined to be a risk factor and KLHL42 a protective factor, validated in clinical samples. This research provides novel therapeutic targets for investigating the pathological mechanisms of Kawasaki disease.