Identification of Tryptophan Metabolism-Related Biomarkers in Kawasaki Disease

<b>Background:</b><b> </b>Kawasaki disease (KD) is an acute febrile vasculitis that predominantly affects children under five years of age. Although immune dysregulation is known to contribute to KD pathogenesis, the role of tryptophan metabolism (TM) in KD remains largely unexplored. This study aimed to identify TM-related biomarkers in KD, offering potential targets for diagnosis and therapy.<b>Methods:</b>Transcriptomic datasets related to KD were obtained from publicly available databases. Candidate genes associated with TM were screened through differential expression analysis and weighted gene co-expression network analysis (WGCNA). Key biomarkers with strong biological relevance were further identified using protein-protein interaction (PPI) networks, three machine learning algorithms, and receiver operating characteristic (ROC) curve analysis. Immune infiltration characteristics were examined in KD samples. The expression levels of candidate biomarkers were validated by reverse transcription quantitative PCR (RT-qPCR). A vascular endothelial cell inflammation model was used to explore the anti-inflammatory effects of saxagliptin targeting key genes.<b>Results:</b>A total of 234 TM-related candidate genes were identified in KD. Four core genes-SPI1, FCER1G, ITGAX, and NCF2-were consistently identified across three machine learning models and showed high diagnostic performance in ROC analysis. These biomarkers are significantly involved in immune-related pathways, including NOD-like receptor signaling. Immune profiling revealed substantial alterations in the immune microenvironment of KD patients compared with healthy controls. The expression of the four biomarkers was significantly elevated in KD clinical samples, consistent with bioinformatic predictions. In vitro, saxagliptin targeting SPI1 and NCF2 significantly suppressed inflammatory cytokine production, including IL-1β, IL-6, and TNF-α, in inflamed vascular endothelial cells.