Back

Single-cell RNA-seq reveals altered plasma cell subsets and decreased cytotoxicity of NK cells in patients with Kawasaki diseas

Zhou, Y.; Liu, Y.; Liu, L.; Shen, X.; Wen, X.; Lin, G.; Tang, L.; Li, Q.; Gan, J.; Li, R.; Shang, J.; Li, L.

2025-02-08 immunology
10.1101/2025.02.03.636324 bioRxiv
Show abstract

AimThe pathogenesis and therapeutic strategies of Kawasaki disease (KD) are worth further exploring. Using single-cell RNA sequencing (scRNA-seq), our study aimed to explore the landscape of peripheral blood mononuclear cells (PBMCs) in patients with KD. MethodsWe analyzed the scRNA-seq data of three patients with KD and three healthy controls from the GSE168732 dataset. Additionally, we explored the immune profiles and immune cell subtypes of patients with KD. ResultsSignificantly increased B cells and decreased NK cells were observed in the KD group compared with that in the control group. Furthermore, six B cell subsets were identified in the PBMCs of patients with KD. Particularly, plasma cells (CD9+ and S100A9+ B cell clusters) distinctly increased in the KD group. CD9+ B cell cluster was characterized by vascular endothelial growth factor (VEGF) signaling-associated marker genes and DEGs. Meanwhile the S100A9+ B cell cluster was characterized by platelet aggregation marker genes. As for NK cells, CD16+CD56dim NK cells were enriched in the KD group. Cytotoxicity-related activating receptors (KLRD1,KLRC3) and cytotoxicity-associated genes (PRF1, NKG7, GZMA, GZMH, GNLY) were downregulated in the KD group. ConclusionCD9+ and S100A9+ plasma cells may contribute to CALs in KD by upregulating VEGF and activating platelet. In addition, decreased cytotoxicity of NK cells mediated by KLRD1/KLRC3 receptor may play a vital role in the development of KD. Our results suggested that plasma cells and NK cells could be promising targets for the treatment of KD.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.