Longitudinal single-cell atlas to identify B cell subpopulations and states associated with human vaccine responses 2309736

Abstract

Abstract Introduction Vaccine responses vary across individuals, with specific B cell subpopulations driving humoral immune response. Bulk transcriptomic studies have identified pan-vaccine predictive signatures, but have difficulty determining which cell subsets drive these signals. Single-cell approaches offer the cellular resolution to link signatures to specific subpopulations, and recent study have successfully associated B cell subsets with vaccine responsiveness. However, most single-cell vaccine studies use small cohorts (< 10 subjects) with insufficient power to reliably detect rare cell states like functionally distinct plasma cell subsets and subtypes of memory B cells. Moreover, inconsistent cell annotations hinder cross-study comparisons. A large-scale, standardized B cell atlas across multiple vaccines would facilitate detection of rare response-associated populations and establish reference subset definitions. Methods We curated and processed nine single-cell RNA-seq studies (53 subjects) following SARS-CoV-2, seasonal influenza, and malaria vaccinations spanning baseline through various timepoints post-vaccination. All datasets were processed through a standardized pipeline and integrated to preserve biological variation while correcting technical batch effects. Hierarchical celltype annotation was applied to identify B cell subpopulations. Results The integrated atlas comprises 447,493 B cells spanning naïve (50-70%), memory (25-40%), activated, and plasma cell populations in PBMC samples across three different vaccine responses. We observe plasmablast enrichment post-vaccination consistent with peak antibody responses across many vaccines. Conclusion This multi-vaccine B cell atlas enables systematic profiling of rare response-associated subpopulations not detectable in small-cohort studies. Integration of transcriptional states with antibody outcomes may reveal cellular predictors of immune response. Funding Source This work is supported through the NIH U19AI089992 grant. Topic Categories Computational and Systems Immunology (COMP)