Atypical memory B cell clonal expansion and inflammatory programs associate with platelet-activating antibody development in COVID-19
Nathan Witman, Mei Yu, Yuqi Zhang, Kexin Gai, Yuhong Chen, Lu Zhou, Christine Nguyen, Wen Zhu, Yongwei Zheng, Shawn Jobe, Mary Beth Graham, Weiguo Cui, Demin Wang, Renren Wen
Nathan Witman, Mei Yu, Yuqi Zhang, Kexin Gai, Yuhong Chen, Lu Zhou, Christine Nguyen, Wen Zhu, Yongwei Zheng, Shawn Jobe, Mary Beth Graham, Weiguo Cui, Demin Wang, Renren Wen
View: Text | PDF
Research Article Immunology Vascular biology

Atypical memory B cell clonal expansion and inflammatory programs associate with platelet-activating antibody development in COVID-19

Abstract

Patients with COVID-19 who develop platelet-activating antibodies represent a subset at heightened thrombotic risk, yet the immune features associated with this response remains to be defined. We applied single-cell RNA-seq of B and T cells, single B cell V(D)J-seq, and plasma cytokine and chemokine analysis to define immune signatures distinguishing patients who did (PEA+) or did not (PEA–) develop these antibodies. Patients positive for PEA showed prominent transcriptional enrichment of inflammatory, antigen presentation, and B cell receptor signaling pathways within antigen-experienced B cell subsets. Expanded B cell clones in patients positive for PEA were disproportionately enriched within atypical memory B cells and exhibited upregulated IFN-γ–response signatures, increased proliferative mutational patterns, limited class switching, and a significant overrepresentation of RKH/Y5 heavy-chain motifs associated with platelet-activating antibodies, consistent with an extrafollicular-biased response. Parallel T cell profiling revealed IL-12 pathway enrichment across most T cell subsets, increased IFN-γ transcription, and elevated plasma levels of Th1-associated cytokines in patients positive for PEA. Collectively, these data highlight a coordinated inflammatory environment marked by Th1-skewed T cell activation and selective expansion of atypical memory B cell clones carrying RKH/Y5 motifs, defining immunologic features associated with platelet-activating antibody development in COVID-19.

Authors

Nathan Witman, Mei Yu, Yuqi Zhang, Kexin Gai, Yuhong Chen, Lu Zhou, Christine Nguyen, Wen Zhu, Yongwei Zheng, Shawn Jobe, Mary Beth Graham, Weiguo Cui, Demin Wang, Renren Wen

×

Figure 5

Paralleled upregulation of inflammatory programs in T cells from PEA+ patients.

Options: View larger image (or click on image) Download as PowerPoint
Paralleled upregulation of inflammatory programs in T cells from PEA+ pa...
CD4+ T cells were FACS-sorted from PBMCs of 4 PEA+ and 5 PEA– patients, followed by scRNA-seq. (A) UMAP visualization of CD4+ T cells. UMAP of 67,303 CD4+ T cells, colored by 8 clusters based on 5′ gene expression and annotated by subset identity. (B) Dot plot showing the expression of key marker genes used to validate the cluster annotations. (C) UMAP plots split by PEA status. UMAP plots show PEA+ (n = 30,930) and PEA– (n = 36,373) cell distributions. (D) CD4+ T cell population distribution by PEA status, with PEA+ and PEA–. Significance was assessed by Wilcoxon rank-sum test. (E) GSEA on pseudobulk-adjusted DESeq2 gene rankings. Genes were ranked by log2 fold change, and enrichment was assessed using the MSigDB KEGG database. Comparative analysis of transcriptomes in CD4 T cells from PEA+ and PEA– patients was performed for the indicated pathways. Color gradient indicates FDR-adjusted q values. *q < 0.25, **q < 0.1, ***q < 0.05, ****q < 0.01. (F) Overlapping KEGG pathways in atypical memory B cells and Th1-GzmB T cells from PEA+ patients. Venn diagram shows shared pathway enrichment between the 2 subsets types in PEA+ patients. (G) GSEA on pseudobulk-adjusted DESeq2 gene rankings. Genes were ranked by log2 fold change, and enrichment was assessed using the MSigDB KEGG database. Comparative analysis of transcriptomes in CD4 T cells from PEA+ and PEA– patients was performed for the indicated pathways. Color gradient indicates FDR-adjusted q values. *q < 0.25, **q < 0.1, ***q < 0.05, ****q < 0.01.