Integrative single-cell and bulk transcriptomic profiling identify peripheral T-cell-associated candidate biomarkers in community-acquired pneumonia

Abstract

Background Community-acquired pneumonia (CAP) is associated with marked immune dysregulation, but the peripheral T-cell transcriptional alterations linked to CAP remain incompletely defined. We aimed to characterize CAP-associated T-cell heterogeneity and identify candidate blood biomarkers by integrating single-cell and bulk transcriptomic data. Methods GEO-processed whole-blood microarray expression data from 108 patients with CAP and 42 healthy controls (GSE65682), together with processed PBMC scRNA-seq data from five non-influenza CAP donors and four control donors (GSE164948), were analysed. Cell annotation, differential expression, disease contribution scoring, LASSO regression, and support vector machine-recursive feature elimination were used to identify candidate genes. Immune infiltration, enrichment, and pseudotime analyses were performed. Key findings were evaluated by qRT-PCR in PBMCs from an independent cohort of 36 CAP patients and 18 healthy controls. Results T cells showed the highest disease contribution score and were enriched for pathways related to antigen processing, phosphorylation, and immune regulation. Four candidate biomarkers, CD8B, GZMM, MYL12B, and TUBA1A, showed differential expression between CAP and controls. CIBERSORT suggested increased eosinophils and monocytes and reduced resting CD4+ memory T cells in CAP. Pseudotime analysis indicated that CAP-associated T cells accumulated at higher inferred pseudotime values. qRT-PCR confirmed significant upregulation of these four candidate biomarkers in CAP PBMCs; FCGR2A also showed CAP-associated expression changes. Conclusions Integrative transcriptomic profiling identified peripheral T-cell dysregulation in CAP and highlighted CD8B, GZMM, MYL12B, and TUBA1A as candidate blood biomarkers. These findings provide a basis for further multicenter validation and mechanistic studies.