Emergent spatiotemporal organization in the endosomal network: Quantitative analysis via lattice light-sheet microscopy

The endosomal network is the cell's supply chain, connecting it to the outside world by enabling the import and export of materials, as well as associated information. Cargo-containing vesicles termed endosomes make up this dynamic network, moving through the cell and undergoing chemical reactions that update endosomal identity, thereby constraining possible interaction partners and subsequent routes, and providing a readout of the location of an endosome within the overall process. The molecular mechanisms of this system are well-characterized, but understanding its collective behavior remains a challenge due to the inherent complexity: emergent properties of the transport network can only be quantified by observing thousands of individual vesicles (typically 100s of nm) engaging in rapid, stochastic behaviors (requiring ~1 s time resolution) distributed across multiple spatial scales across the entire cell (spanning 10s of µm). Limitations in fast volumetric imaging historically prevented direct whole-cell measurements of absolute numbers of events, but lattice light-sheet microscopy has the capacity to capture the dynamic, stochastic nature of this process across the full cell. Using this modality, we recently demonstrated that spatiotemporal organization emerges from stochastic interactions between endosomes of different identities in the conversion of very early to early endosomes, contrary to the conventional picture of orderly and isolated endosomal maturation. In combination with computational analysis routines to quantify complex endosomal dynamics over time (including motility, tubulation, collisions, conversions, fusion, and fission), this approach provides a transformative platform to map the endosomal system and quantitatively address hitherto intractable questions. We will discuss insights gleaned from the analysis of diverse events within single cells, and provide preliminary examples that extend this approach to multiple cells embedded within live tissue.