Biophysical Mechanisms on the Pathway to Membrane Fusion

Fusion of membrane-enclosed compartments to mix or release contents is essential for intracellular trafficking, secretion, fertilization and development and is a critical step during cell invasion by enveloped viruses. In almost all intracellular fusion processes SNARE proteins are thought to exert the requisite forces, while viruses employ fusion proteins such as hemagglutinin of influenza and Env of HIV. The underlying biophysical mechanisms have been extensively explored in synthetic membrane systems where fusion is driven by divalent cations and membrane tension. Considerable evidence suggest that fusion pathways in these diverse systems may universally involve three principle steps: (1) tight adhesion providing intimate membrane contact; (2) hemifusion, where only outer leaflets of the apposing membranes are fused while the inner leaflets engage one another in a new bilayer region called the hemifusion diaphragm (HD); (3) HD rupture yielding complete fusion. This thesis elucidates for the first time the biophysical mechanisms that determine the hemifused state in synthetic membranes, and which must be contended with by biological fusion machines. In addition, a model is developed of SNARE-mediated membrane adhesion in SNARE-reconstituted large membrane systems in vitro. While hemifusion and adhesion are the focus here, our hemifusion analysis enables us to propose a simple mechanistic model for the transition to complete fusion.;We first examine the hemifused intermediate, a stage poorly understood even in the relatively simple protein-free systems which have been commonly studied over the last 35 years. Our general model describes hemifusion in vesicle systems driven by divalent cations or osmotically-generated tension. We find these forces drive expansion of a hemifusion connection between vesicles and thermodynamically favor HDs with areas typically a few percent of the vesicle areas. These predictions are in excellent quantitative agreement with recent experimental measurements. Calculated HD properties are presented for a range of additional experiments in the literature. A model is also developed of hemifusion-fission of vesicles with planar suspended bilayers (SBLs), a common experimental model of vesicle-plasma membrane fusion. In vesicle-SBL systems we find HDs grow to one half of the vesicle size. Precisely this behavior has been observed in experimental studies of cell-SBL hemifusion. The model shows that despite the large HD size, vesicle-SBL systems are intrinsically less fusogenic since high cation-generated vesicle tensions are dissipated in hemifusion equilibrium by the SBL. These predictions explain the osmotic gradient requirement universally reported for vesicle-SBL complete fusion.;The studies above addressed the equilibrium hemifused state. A model describing the dynamics of HD growth was then developed. Early- and long-time growth predictions are given which explain HD growth kinetics recently measured. Importantly, the results given here for time-dependent HD area and HD tension constitute the necessary information for the first quantitative model of the entire hemifusion-fusion pathway via HD rupture. We also discuss how the physical mechanisms established here may be relevant to biological systems where calcium spikes may locally generate high hemifusion driving forces.;Finally, we developed a model of the adhesion of giant vesicles reconstituted with SNARE proteins, a minimal experimental model of in vivo situations where large membranes adhere and fuse. Our model predicts that adhesion occurs through growth of a self-promoting; tightly-adhered patch. Once nucleated, the patch facilitates binding of SNARES provided by opposing membranes to form SNARE complexes. We find that SNARE complex mobility allows rapid patch growth. The realized growth velocity has power-law dependence on SNARE density. We find adhesion depends on the SNARE binding rate constant and SNARE complex diffusivity in contrast to the behavior predicted by prior models of systems where the biosticker-ligand complexes are immobile. The patch may provide a close contact region where SNARES can trigger fusion. Extending the model to a simple description of fusion, a broad distribution of fusion times is predicted. Increasing SNARE density accelerates fusion by boosting the patch growth velocity, thereby providing more complexes to participate in fusion. This quantifies the notion of SNARES as dual adhesion-fusion agents. These results show that controlled in vitro giant vesicle systems can probe fundamental properties of SNARE proteins such as the 2D complexation rate constant.