Controlling distance, time and reactivity: Chemical principles of proximity labeling.

By generating short-lived reactive intermediates that covalently tag nearby biomolecules, proximity labeling (PL) has become a central strategy for spatial proteomics and for probing protein-protein interactions in living systems. However, key aspects of PL performance, including labeling radius, temporal resolution, and biological compatibility, are ultimately governed by how these intermediates are produced, confined, and quenched in situ. Here we use reactive-intermediate generation as an organizing chemical framework to compare major PL modalities. We focus primarily on proteome-centered PL systems, while noting that the same framework can extend to other biomolecular readouts. We discuss peroxide- and oxygen-driven platforms that form phenoxyl radicals and quinone electrophiles, ATP-coupled ligase approaches that transfer activated intermediates to proximal nucleophiles, and light-triggered photocatalytic systems that access carbenes or nitrenes, singlet oxygen or radical reactive oxygen species, and photoredox-uncaged electrophiles. Across these manifolds, we highlight the trade-offs that set operational boundaries and outline design principles for next-generation PL that is quantitative, minimally perturbative, and increasingly in vivo compatible.