The origin of cellular life is rooted in the formation of a boundary that paradoxically separates yet couples the internal and external environments. Formed from amphipathic lipid molecules, the membrane is a highly dynamic structure whose barrier function and capacity for deformation are essential for cellular processes. Lipid biosynthesis, intracellular signaling, and vesicular and non-vesicular transport continuously remodel its composition and spatial organization. With eukaryogenesis came an explosion in membrane diversity, as organelles acquired a wide range of sizes, curvatures, and specialized functions. As membrane processes grew in complexity, membrane-remodeling machineries became essential for maintaining membrane architecture and supporting trafficking functions. Among these, the Endosomal Sorting Complex Required for Transport (ESCRT) machinery is thought to be one of the most evolutionarily conserved.
The ESCRT machinery remodels membranes across a wide range of spatial and temporal scales, from multivesicular body biogenesis to cytokinetic abscission, viral budding, and nuclear envelope reformation. In particular, ESCRT-III mediates membrane deformation and scission from inside the membrane neck, enabling so-called “reverse-topology” remodeling, in which the forming vesicle is oriented away from the cytoplasm. Curiously, in vitro reconstitutions have revealed that ESCRT-III polymers form on the outside of membrane necks and spontaneously favor the “normal” topology in model membrane systems. Thus, this thesis asks: what aspect of cellular membranes constrains ESCRT-III to mediate “reverse-topology” remodeling?
I propose a framework in which directional membrane bending by ESCRT-III emerges from membrane asymmetry. Lipid asymmetry is an intrinsic property of biological membranes that is not readily recapitulated in model systems. I show that lipid asymmetry, particularly sphingolipid asymmetry, creates membranes that are mechanically primed for inward budding and invagination.
In Saccharomyces cerevisiae, perturbations to phospholipid and sphingolipid asymmetry impair ESCRT-III–dependent intraluminal vesicle (ILV) formation and vacuolar sorting of ESCRT-dependent cargoes. Complementarily, enzymatic induction of sphingolipid asymmetry in vitro generates mechanically stressed membranes that, when coupled to ESCRT-III polymerization, drive inward invaginations in giant unilamellar vesicles. Using fluorescence lifetime imaging, I find that asymmetric membranes exhibit altered leaflet packing and tension, consistent with a mechanically primed state for inward bending. Finally, I show that leaflet-specific cargo crowding can bias membrane bending and tubulation in vitro, while synthetic cargo orientation in cells modulates sorting efficiency into ILVs.
Overall, this thesis proposes that ESCRT-III–dependent membrane remodeling is not solely determined by the protein machinery itself, but emerges from asymmetric mechanical stresses generated by lipids and cargo that provide the physical constraints for directional bending.
