Rapid CE-MS with Real-Time Eco-AI Resolves Proteomic Heterogeneity Among Single Human Neutrophils

Abstract

Single-cell proteomics by mass spectrometry is advancing rapidly, yet throughput and sensitivity remain limiting--particularly for small, protein-poor cell types such as neutrophils. As the most abundant circulating leukocytes in humans, neutrophils are central to immune defense and inflammation, but their proteomes comprehensive single-cell level characterization has only concurrently been reported1 and remains limited. Here, we introduce a rapid capillary electrophoresis-mass spectrometry (CE-MS) platform, which integrates electrophoresis-correlative real-time data acquisition with sub-7-minute separations and artificial intelligence (AI)-based data processing software to achieve deep, high-throughput profiling. Using single-cell-equivalent HeLa digests, the Rapid Eco-AI platform identified [~]1,350 proteins from 300 pg and [~]835 proteins from 75 pg of input-- approaching the complexity of a mammalian cell proteome. Applied to freshly isolated human neutrophils, the workflow identified 151 proteins from [~]2 pg of material, [~]3% of the total cell proteome. Analysis of 13 individual cells revealed marked functional heterogeneity across pathways of degranulation, neutrophil extracellular trap (NET) formation, chemotaxis, and innate immunity, with hierarchical clustering resolving at least four distinct proteomic subtypes. These results establish Rapid Eco-AI as a sensitive, scalable, and broadly applicable CE-MS approach for immune-cell phenotyping at single-cell and subcellular resolution, facilitating new research opportunities in systems immunology and clinical proteomics.