Calcium-independent stimulation of membrane fusion and SNAREpin formation by synaptotagmin I

Abstract

Neurotransmitter release requires the direct coupling of the calcium sensor with the machinery for membrane fusion. SNARE proteins comprise the minimal fusion machinery, and synaptotagmin I, a synaptic vesicle protein, is the primary candidate for the main neuronal calcium sensor. To test the effect of synaptotagmin I on membrane fusion, we incorporated it into a SNARE-mediated liposome fusion assay. Synaptotagmin I dramatically stimulated membrane fusion by facilitating SNAREpin zippering. This stimulatory effect was topologically restricted to v-SNARE vesicles (containing VAMP 2) and only occurred in trans to t-SNARE vesicles (containing syntaxin 1A and SNAP-25). Interestingly, calcium did not affect the overall fusion reaction. These results indicate that synaptotagmin I can directly accelerate SNARE-mediated membrane fusion and raise the possibility that additional components might be required to ensure tight calcium coupling.

Figure 1.

Incorporation of Sytg I into labeled VAMP 2 liposomes accelerates fusion. (A) Reconstituted donor liposomes. Donor liposomes containing VAMP 2 alone (lane 1), Sytg I–VAMP 2 (lane 2), and Sytg I alone (lane 3) were analyzed by SDS-PAGE and stained with Coomassie blue. The positions of VAMP 2 and Sytg I are marked by arrows. (B) Fusion in the presence of Sytg I. Unlabeled acceptor liposomes (45 μl) containing syntaxin 1A/SNAP 25 (t-SNARE) were prewarmed to 37°C and mixed at time = 0 with prewarmed donor liposomes (5 μl) labeled with rhodamine and NBD lipids in the presence and absence of the cytosolic domain of VAMP 2 (cd–VAMP 2, aa 1–94, added in approximately equimolar amounts to the t-SNARE). The increase in NBD fluorescence was monitored for 2 h at 37°C, and the results were normalized to the maximum NBD fluorescence signal after addition of detergent (dodecylmaltoside). (C) Dose-dependent stimulation of fusion by Sytg I. Donor liposomes containing a constant amount of VAMP 2 but increasing amounts of Sytg I were incubated with acceptor t-SNARE liposomes, and the fusion was monitored and analyzed as described. Linear regression analysis (Cricket Graph III, curve fit) was performed using the first eight data points (first 14 min), and the initial rate of fusion (Δ % total fluorescence/Δ time) was plotted against the total amount of reconstituted Sytg I–VAMP 2.

Figure 1.

Incorporation of Sytg I into labeled VAMP 2 liposomes accelerates fusion. (A) Reconstituted donor liposomes. Donor liposomes containing VAMP 2 alone (lane 1), Sytg I–VAMP 2 (lane 2), and Sytg I alone (lane 3) were analyzed by SDS-PAGE and stained with Coomassie blue. The positions of VAMP 2 and Sytg I are marked by arrows. (B) Fusion in the presence of Sytg I. Unlabeled acceptor liposomes (45 μl) containing syntaxin 1A/SNAP 25 (t-SNARE) were prewarmed to 37°C and mixed at time = 0 with prewarmed donor liposomes (5 μl) labeled with rhodamine and NBD lipids in the presence and absence of the cytosolic domain of VAMP 2 (cd–VAMP 2, aa 1–94, added in approximately equimolar amounts to the t-SNARE). The increase in NBD fluorescence was monitored for 2 h at 37°C, and the results were normalized to the maximum NBD fluorescence signal after addition of detergent (dodecylmaltoside). (C) Dose-dependent stimulation of fusion by Sytg I. Donor liposomes containing a constant amount of VAMP 2 but increasing amounts of Sytg I were incubated with acceptor t-SNARE liposomes, and the fusion was monitored and analyzed as described. Linear regression analysis (Cricket Graph III, curve fit) was performed using the first eight data points (first 14 min), and the initial rate of fusion (Δ % total fluorescence/Δ time) was plotted against the total amount of reconstituted Sytg I–VAMP 2.

Figure 1.

Incorporation of Sytg I into labeled VAMP 2 liposomes accelerates fusion. (A) Reconstituted donor liposomes. Donor liposomes containing VAMP 2 alone (lane 1), Sytg I–VAMP 2 (lane 2), and Sytg I alone (lane 3) were analyzed by SDS-PAGE and stained with Coomassie blue. The positions of VAMP 2 and Sytg I are marked by arrows. (B) Fusion in the presence of Sytg I. Unlabeled acceptor liposomes (45 μl) containing syntaxin 1A/SNAP 25 (t-SNARE) were prewarmed to 37°C and mixed at time = 0 with prewarmed donor liposomes (5 μl) labeled with rhodamine and NBD lipids in the presence and absence of the cytosolic domain of VAMP 2 (cd–VAMP 2, aa 1–94, added in approximately equimolar amounts to the t-SNARE). The increase in NBD fluorescence was monitored for 2 h at 37°C, and the results were normalized to the maximum NBD fluorescence signal after addition of detergent (dodecylmaltoside). (C) Dose-dependent stimulation of fusion by Sytg I. Donor liposomes containing a constant amount of VAMP 2 but increasing amounts of Sytg I were incubated with acceptor t-SNARE liposomes, and the fusion was monitored and analyzed as described. Linear regression analysis (Cricket Graph III, curve fit) was performed using the first eight data points (first 14 min), and the initial rate of fusion (Δ % total fluorescence/Δ time) was plotted against the total amount of reconstituted Sytg I–VAMP 2.

Figure 6.

Sytg I–dependent acceleration of SNARE-mediated fusion is insensitive to calcium. (A) Kinetic profiles of membrane fusion of labeled Sytg I–VAMP 2 liposomes with unlabeled t-SNARE liposomes with 1 mM EGTA in the presence or absence of 2 mM CaCl₂. The assay was performed as in the Fig. 1 legend. (B) Binding of t-SNARE (Syntaxin 1A/ SNAP 25) to Sytg I. Full-length t-SNARE (1 μM) was incubated with immobilized Sytg I in the presence of either 1 mM EGTA (EGTA) or 1 mM CaCl₂ (Ca²⁺). t-SNARE binding was assayed by SDS-PAGE and with Coomassie blue staining. Controls lacking Sytg I failed to bind to t-SNARE (unpublished data). The positions of IgG, Sytg I, syntaxin 1A, and SNAP 25 are marked by arrows.

Figure 6.

Sytg I–dependent acceleration of SNARE-mediated fusion is insensitive to calcium. (A) Kinetic profiles of membrane fusion of labeled Sytg I–VAMP 2 liposomes with unlabeled t-SNARE liposomes with 1 mM EGTA in the presence or absence of 2 mM CaCl₂. The assay was performed as in the Fig. 1 legend. (B) Binding of t-SNARE (Syntaxin 1A/ SNAP 25) to Sytg I. Full-length t-SNARE (1 μM) was incubated with immobilized Sytg I in the presence of either 1 mM EGTA (EGTA) or 1 mM CaCl₂ (Ca²⁺). t-SNARE binding was assayed by SDS-PAGE and with Coomassie blue staining. Controls lacking Sytg I failed to bind to t-SNARE (unpublished data). The positions of IgG, Sytg I, syntaxin 1A, and SNAP 25 are marked by arrows.

Figure 2.

Soluble cd–Sytg I accelerates SNARE-dependent fusion. Unlabeled acceptor liposomes (45 μl) containing VAMP 2 were prewarmed to 37°C and mixed at time = 0 with prewarmed labeled donor liposomes (5 μl) containing t-SNARE in the presence and absence of the cytosolic domain of Sytg I (cd–Sytg I, ∼10 μM final concentration) and cd–VAMP 2 (see Fig. 1B legend). Note that in assays treated with cd–VAMP 2, the donor liposomes were pretreated with cd–VAMP 2 for 15 min before their mixing with acceptor liposomes. The increase in NBD fluorescence was monitored, and the results were normalized as before.

Figure 3.

Sytg I reconstituted into t-SNARE liposomes does not accelerate fusion. (A) Acceptor liposomes containing t-SNARE (lane 1) or Sytg I and t-SNARE (lane 2) were analyzed by SDS-PAGE and stained with Coomassie blue. The positions of syntaxin 1A, SNAP 25, and Sytg I are marked by arrows. (B) Acceptor liposomes (45 μl) were prewarmed to 37°C and mixed at time = 0 with prewarmed labeled donor liposomes (5 μl) containing VAMP 2. The increase in NBD fluorescence was monitored, and the results were normalized as before.

Figure 3.

Sytg I reconstituted into t-SNARE liposomes does not accelerate fusion. (A) Acceptor liposomes containing t-SNARE (lane 1) or Sytg I and t-SNARE (lane 2) were analyzed by SDS-PAGE and stained with Coomassie blue. The positions of syntaxin 1A, SNAP 25, and Sytg I are marked by arrows. (B) Acceptor liposomes (45 μl) were prewarmed to 37°C and mixed at time = 0 with prewarmed labeled donor liposomes (5 μl) containing VAMP 2. The increase in NBD fluorescence was monitored, and the results were normalized as before.

Figure 4.

Cleavage of the NH ₂ -terminal domain of syntaxin 1A does not effect Sytg I–dependent acceleration. (A) Steps in the in vitro fusion assay. First, the NRD (oval) of syntaxin 1A (represented here by the medium gray cylinder connected to the oval) moves to form an activated t-SNARE complex (SNAP 25 is represented by the two connected light gray cylinders). Next, VAMP 2 (single cylinder) interacts with the activated complex to form a fusion-competent docked SNARE complex. Finally, membrane fusion occurs accompanied by formation of the cis-SNARE complex. (B) Kinetic profiles of membrane fusion in the presence of Sytg I with t-SNARE missing the NRD. Unlabeled acceptor liposomes containing thrombin cleavable t-SNARE (N–t-SNARE) were treated with thrombin (tc) or buffer (Parlati et al., 1999). Fusion assays were performed as described in the Fig. 1 legend.

Figure 4.

Cleavage of the NH ₂ -terminal domain of syntaxin 1A does not effect Sytg I–dependent acceleration. (A) Steps in the in vitro fusion assay. First, the NRD (oval) of syntaxin 1A (represented here by the medium gray cylinder connected to the oval) moves to form an activated t-SNARE complex (SNAP 25 is represented by the two connected light gray cylinders). Next, VAMP 2 (single cylinder) interacts with the activated complex to form a fusion-competent docked SNARE complex. Finally, membrane fusion occurs accompanied by formation of the cis-SNARE complex. (B) Kinetic profiles of membrane fusion in the presence of Sytg I with t-SNARE missing the NRD. Unlabeled acceptor liposomes containing thrombin cleavable t-SNARE (N–t-SNARE) were treated with thrombin (tc) or buffer (Parlati et al., 1999). Fusion assays were performed as described in the Fig. 1 legend.

Figure 5.

Sytg I accelerates fusion-committed docking. (A) Acceptor t-SNARE liposomes (45 μl) were mixed with donor liposomes (5 μl) containing VAMP 2 with and without Sytg I at 4°C. After time equals x, cd–VAMP 2 (equimolar to t-SNARE) was added, and the reaction was allowed to incubate at 4°C until the final time point. The mixture was then warmed to 37°C, and the reaction is monitored as described in Fig. 1 legend. (B) Kinetic profiles of fusion from a time course of functionally docked VAMP 2 liposomes. The assay was performed as outlined in A. Note that the normalized data ignores the preassay fusion and an initial drop in fluorescence, which is due to temperature effects on the fluorescent probes (Chapman et al., 1995a), is typically observed. In addition, please note that the time course for SNAREpin formation is concentration dependent. (C) Kinetic profiles of fusion from a time course of functionally docked VAMP 2–Sytg I liposomes. The assay was performed as outlined in A.

Figure 5.

Sytg I accelerates fusion-committed docking. (A) Acceptor t-SNARE liposomes (45 μl) were mixed with donor liposomes (5 μl) containing VAMP 2 with and without Sytg I at 4°C. After time equals x, cd–VAMP 2 (equimolar to t-SNARE) was added, and the reaction was allowed to incubate at 4°C until the final time point. The mixture was then warmed to 37°C, and the reaction is monitored as described in Fig. 1 legend. (B) Kinetic profiles of fusion from a time course of functionally docked VAMP 2 liposomes. The assay was performed as outlined in A. Note that the normalized data ignores the preassay fusion and an initial drop in fluorescence, which is due to temperature effects on the fluorescent probes (Chapman et al., 1995a), is typically observed. In addition, please note that the time course for SNAREpin formation is concentration dependent. (C) Kinetic profiles of fusion from a time course of functionally docked VAMP 2–Sytg I liposomes. The assay was performed as outlined in A.

Figure 5.

Sytg I accelerates fusion-committed docking. (A) Acceptor t-SNARE liposomes (45 μl) were mixed with donor liposomes (5 μl) containing VAMP 2 with and without Sytg I at 4°C. After time equals x, cd–VAMP 2 (equimolar to t-SNARE) was added, and the reaction was allowed to incubate at 4°C until the final time point. The mixture was then warmed to 37°C, and the reaction is monitored as described in Fig. 1 legend. (B) Kinetic profiles of fusion from a time course of functionally docked VAMP 2 liposomes. The assay was performed as outlined in A. Note that the normalized data ignores the preassay fusion and an initial drop in fluorescence, which is due to temperature effects on the fluorescent probes (Chapman et al., 1995a), is typically observed. In addition, please note that the time course for SNAREpin formation is concentration dependent. (C) Kinetic profiles of fusion from a time course of functionally docked VAMP 2–Sytg I liposomes. The assay was performed as outlined in A.