Deep Profiling of the Single-Cell Proteome

The proteome encodes the functional state of a cell with a resolution that transcripts cannot fully capture, yet single-cell and spatial proteomics remain constrained by limited sensitivity, throughput, and analytical scalability. This thesis presents a comprehensive framework for deep profiling of the single-cell proteome, spanning methodological development, biological application, and technological innovation. In Part I, a robust pipeline for single-cell proteomics is established through systematic optimization of sample preparation, including deparaffinization, lysis and extraction, and miniaturized digestion workflows in high-throughput multi-well plate formats. These advances enable efficient protein recovery from both fluorescence-activated cell sorting (FACS) isolated single cells and laser capture microdissection (LCM) tissue specimens. Complementary complexity reduction strategies and scpViz, an open-source platform for visualization and analysis of sparse single-cell proteomic datasets, further improve sensitivity and data interpretability. Together, these approaches enable quantification of approximately 2,500 proteins from single cells, with recent benchmarks on next-generation instrumentation exceeding 4,000 proteins per cell. In Part II, the developed framework is applied to interrogate biological systems at cellular and tissue scales. In a colorectal cancer model, single-cell proteomics reveals proteomic heterogeneity between parental and drug-resistant lines, identifying divergent response programs undetectable at bulk resolution, and implicating loss of signalling plasticity as a hallmark of resistance. In the nervous system, spatially resolved LCM proteomics uncovers region-specific protein dysregulation at near single-cell resolution. Additional investigations into developmental systems demonstrate the capacity to capture spatiotemporal proteomic dynamics across biological contexts. In Part III, a nanoDESI-based platform is developed toward subcellular spatial proteomics. A first-principles model of fluid and analyte transport provides quantitative scaling laws governing extraction efficiency as a function of probe geometry and flow parameters. This theoretical framework is coupled with nanoDESI probe microfabrication and a custom instrumentation platform enabling controlled, high-resolution sampling. Collectively, this thesis integrates experimental, computational, and theoretical approaches to enable deep, spatially resolved profiling of the single-cell proteome, providing both the experimental infrastructure and the physical framework necessary for next-generation proteomics technologies.