Nanodisc-cell fusion: control of fusion pore nucleation and lifetimes by SNARE protein transmembrane domains

Abstract

The initial, nanometer-sized connection between the plasma membrane and a hormone- or neurotransmitter-filled vesicle -the fusion pore- can flicker open and closed repeatedly before dilating or resealing irreversibly. Pore dynamics determine release and vesicle recycling kinetics, but pore properties are poorly known because biochemically defined single-pore assays are lacking. We isolated single flickering pores connecting v-SNARE-reconstituted nanodiscs to cells ectopically expressing cognate, "flipped" t-SNAREs. Conductance through single, voltage-clamped fusion pores directly reported sub-millisecond pore dynamics. Pore currents fluctuated, transiently returned to baseline multiple times, and disappeared ~6 s after initial opening, as if the fusion pore fluctuated in size, flickered, and resealed. We found that interactions between v- and t-SNARE transmembrane domains (TMDs) promote, but are not essential for pore nucleation. Surprisingly, TMD modifications designed to disrupt v- and t-SNARE TMD zippering prolonged pore lifetimes dramatically. We propose that the post-fusion geometry of the proteins contribute to pore stability.

Figure 1. Detection of single fusion pores between v-SNARE nanodiscs (vNDs) and flipped t-SNARE cells (tCells).

(a,b) Schematic of the setup and vND-tCell fusion. A pipette is sealed onto the tCell to isolate a patch of membrane (a). Unlike fusion of a closed vesicle with a voltage-clamped membrane, vND-tCell fusion establishes a direct ionic conduction pathway between the cell cytosol and the pipette solution (b). The schematic in b is approximately to scale. (c) Fusion pores are SNARE-induced. When the soluble cytoplasmic domain of the v-SNARE VAMP2 (CDV), or the tetanus neurotoxin light chain (TeNT) was included in the pipette solution, currents occurred much less frequently. Similarly, not including NDs in the pipette (no ND), using SNARE-free, empty NDs (eND), wild-type cells not expressing flipped t-SNAREs (wtCell), replacing Na⁺ with the larger NMDG⁺, or replacing wild-type VAMP2 with a mutant that was previously shown to induce docking but not fusion largely abolished currents. (***Indicates p < 0.001, t-test against vND-tCell). The number of patches/pores for each condition is indicated. (d) Example of a fusion pore current “burst”. Open sub-states are defined as having current <−0.25 pA (red dashed line) for at least 60 ms. Colored bars above the fusion pore indicate detected open periods. (e) Distribution of burst lifetimes, T_o, as defined in (d). The red, solid line is an exponential fit, with mean 5.8 s. (f ) Distribution of flicker numbers , and fitted geometric distribution (red solid line, y = p(1 − p)ⁿ, n = 0, 1, 2, 3, … with p = 0.754 ± 0.0129 (95% confidence interval). (g) Probability density function of open-pore conductances, G_po, averaged over 122 individual fusion pores from 50 cells. (h) Probability density function for open-pore radii. For (e–h), the mean ± S.E.M. were T_o = 5.8 ± 0.9 s, N_flicker = 12 ± 2, G_po = 152 ± 12 pS, and r_po = 0.60 ± 0.02 nm.

Figure 2. Fusion of vNDs with tCells elicits whole-cell currents.

(a) Schematic of the setup. A tCell is voltage-clamped at −70 mV in the whole-cell configuration and vNDs are puffed nearby from a pipette. (b) Fusion of vNDs elicited whole-cell currents that were much larger than the unitary fusion pore currents obtained in the cell-attached configuration depicted in Fig. 1. Application of vNDs in the presence of TeNT did not elicit appreciable currents. (c) Whole-cell charge transfer that resulted from application of vNDs alone, or vNDs in the presence of CDV or TeNT. (n = 8, 7, and 4 cells for vND alone, and with CDV and TeNT). (*Indicates p < 0.05, t- test against vND alone.)

Figure 3. Single-cell lipid mixing assay.

(a) Schematic of the approach. NDs containing 1% each of DiI and DiD lipid labels were pre-incubated with cells at 4 °C for 30 min to allow docking, but no fusion, with tCells. After rinsing twice with ice cold PBS to remove free NDs, PBS pre-heated to 37 °C was added and image acquisition was started and continued for 20 min using a spinning disc confocal microscope equipped with a temperature controlled stage set to 37 °C. For each image cycle, one frame recorded DiI fluorescence excited at 561 nm and the subsequent frame recorded DiD fluorescence excited at 647 nm. DiI fluorescence reports lipid mixing; upon fusion DiI and DiD are diluted in the plasma membrane and DiI is no longer quenched by DiD. At the labelling density used, DiD is not significantly self-quenched, so the DiD signal is proportional to the initial density of docked NDs. The ratio R of DiI-to-DiD fluorescence normalizes fusion signals for variations in docked ND density, temperature-induced fluorescence changes, and other instrumental and environmental factors. (b) Changes in DiI-to-DiD fluorescence ratio (ΔR), relative to the initial ratio (R_o), for SNARE-free NDs (eND), or NDs loaded with wild-type (vND) or docking-competent, fusion-incompetent VAMP2-4X (v4xND). 11, 8, and 6 dishes were analysed for vND, v4xND and eND conditions, respectively.

Figure 4. Zippering of v- and t-SNARE TMDs boosts pore nucleation but is not essential for it.

(a) Schematic of the VAMP2 modifications used. VAMP2 TMD residues I98, I102, and I106 that contact syntaxin1 TMD residues in a crystal were mutated to alanines that readily incorporate into α-helices (v3x). Alternatively, the TMD of VAMP2 was replaced with that of Bet1 (vBet1), or a lipid anchor long enough to span the entire bilayer (27) (maleimidopropionic acid solanesyl ester) (vC45). (b) The rate of pore nucleation was significantly reduced for all TMD-modified v-SNAREs, but still larger than the rate with SNARE-free NDs (eND). The number of patches/pores for each condition is indicated. (*, **, and ***indicate p < 0.05, p < 0.01, and p < 0.001, respectively, using the t-test against vND.) vND and eND data are copied from Fig. 1c for comparison with data from TMD-modified SNAREs.

Figure 5. Geometric constraints may contribute to pore lifetimes.

(a) Representative pore currents for NDs loaded with wild-type (vND) or v3x mutant VAMP2 (v3xND). Hypothetical post-fusion geometries of the VAMP2 proteins complexed with t-SNAREs are shown next to each trace. (b,c) Distribution of burst lifetimes (b) and their averages (c). Exponential fits (continuous lines) to data (staircase plots) are shown in (b). (d) Open-pore conductance fluctuations (variance) relative to mean, . TMD-modifications that reduced TMD zippering stabilized pore lifetimes. (***Indicates p < 0.001, using the 2-sample Kolmogorov-Smirnov test against eND). The number of patches/pores for every condition as in Fig. 4.