A lever hypothesis for Synaptotagmin-1 action in neurotransmitter release and Studies of Synaptotagmin-1 action by all-atom molecular dynamics simulations

Abstract 1:

Neurotransmitter release is triggered in microseconds by Ca2+-binding to the Synaptotagmin-1 C2-domains and by SNARE complexes that form four-helix bundles between synaptic vesicles and plasma membranes, but the coupling mechanism between Ca2+-sensing and membrane fusion is unknown. Release requires extension of SNARE helices into juxtamembrane linkers that precede transmembrane regions (linker zippering) and binding of the Synaptotagmin-1 C2B domain to SNARE complexes through a ‘primary interface’ comprising two regions (I and II). The Synaptotagmin-1 Ca2+-binding loops were believed to accelerate membrane fusion by inducing membrane curvature, perturbing lipid bilayers or helping bridge the membranes, but SNARE complex binding through the primary interface orients the Ca2+-binding loops away from the fusion site, hindering these putative activities. To clarify this paradox, we have used NMR and fluorescence spectroscopy. NMR experiments reveal that binding of C2B domain arginines to SNARE acidic residues at region II remains after disruption of region I, and that a mutation that impairs spontaneous and Ca2+-triggered neurotransmitter release enhances binding through region I. Moreover, fluorescence assays show that Ca2+ does not induce dissociation of synaptotagmin-1 from membrane-anchored SNARE complex but causes reorientation of the C2B domain. Based on these results and electrophysiological data described in Toulme et al. (https://doi.org/10.1073/pnas.2409636121), we propose that upon Ca2+ binding the Synaptotagmin-1 C2B domain reorients on the membrane and dissociates from the SNAREs at region I but not region II, acting remotely as a lever that pulls the SNARE complex and facilitates linker zippering or other SNARE structural changes required for fast membrane fusion.

Abstract 2:

Neurotransmitter release is triggered in microseconds by the two C2 domains of the Ca2+ sensor Synaptotagmin-1 and by SNARE complexes, which form four-helix bundles that bridge the vesicle and plasma membranes. The Synaptotagmin-1 C2B domain binds to the SNARE complex via a ‘primary interface’, but the mechanism that couples Ca2+-sensing to membrane fusion is unknown. Widespread models postulate that the Synaptotagmin-1 Ca2+-binding loops accelerate membrane fusion by inducing membrane curvature, perturbing lipid bilayers or helping bridge the membranes, but these models do not seem compatible with SNARE binding through the primary interface, which orients the Ca2+-binding loops away from the fusion site. To test these models, we performed molecular dynamics simulations of SNARE complexes bridging a vesicle and a flat bilayer, including the Synaptotagmin-1 C2 domains in various configurations. Our data do not support the notion that insertion of the Synaptotagmin-1 Ca2+ binding loops causes substantial membrane curvature or major perturbations of the lipid bilayers that could facilitate membrane fusion. We observed membrane bridging by the Synaptotagmin-1 C2 domains, but such bridging or the presence of the C2 domains near the site of fusion hindered the action of the SNAREs in bringing the membranes together. These results argue against models predicting that Synaptotagmin-1 triggers neurotransmitter release by inducing membrane curvature, perturbing bilayers or bridging membranes. Instead, our data support the hypothesis that binding via the primary interface keeps the Synaptotagmin-1 C2 domains away from the site of fusion, orienting them such that they trigger release through a remote action.

Methods
MD simulations. All-atom MD simulations were performed using Gromacs with the CHARMM36 force field. Solvation and ion addition for system setup were performed at the BioHPC supercomputing facility of UT Southwestern. Minimizations, equilibration steps and production molecular dynamics (MD) simulations were carried out on Frontera at the Texas Advanced Computing Center (TACC). Pymol (Schrödinger, LLC) was used for system design, manual manipulation and system visualization. The methodology used to set up the systems and run MD simulations was analogous to that described previously (1, 2). Systems were energy minimized using double precision, whereas the default mixed precision was used in all MD simulations. The systems were heated to the desired temperature running a 1 ns simulation in the NVT ensemble with 1 fs steps, and then equilibrated to 1 atm for 1 ns in the NPT ensemble with isotropic Parrinello-Rahman pressure coupling and 2 fs steps. NPT production MD simulations were performed for the times indicated in Table S1 using 2 fs steps, isotropic Parrinello-Rahman pressure coupling and a 1.1 nm cutoff for non-bonding interactions. Three different groups of atoms were used for Nose-Hoover temperature coupling: i) protein atoms; ii) lipid atoms; and iii) water and KCL ions. Periodic boundary conditions were imposed with Particle Mesh Ewald (PME) summation for long-range electrostatics.

NMR spectroscopy. All NMR spectra were acquired at 25 °C on Agilent DD2 spectrometers equipped with cold probes operating at 600 or 800 MHz. Titrations of WT and mutant 2H,15N-labeled Syt1 C2B domain specifically 13CH3-labeled at the Ile d1 and Met methyl groups (referred to as 15N-C2B for simplicity) with SNARE complex four-helix bundle bound to a fragment spanning residues 26-83 of complexin-1 (referred to as CpxSC) were performed as described in (3). Specific 13CH3 labeling was performed for acquisition of 1H-13C heteronuclear multiple quantum (HMQC) spectra with these samples, although no such spectra are described here. Because the R322E/K325E mutation increases the stability of the C2B domain and slows down H/D exchange, some amide groups in the b-strands are not fully protonated after expression in D2O and purification in buffers containing H2O, resulting in signal loss for the corresponding peaks. All newly purified C2B mutants bearing the R322E/K325E mutation were incubated at RT for one week or at 37 °C for 15 hours to facilitate full exchange, but some amide groups were still not fully exchanged after this procedure. The titrations were performed in NMR buffer containing 20 mM HEPES (pH 7.4), 100 mM KCl, 1 mM EDTA, 1 mM TCEP, 10% D2O and protease inhibitor cocktail [which contained 1 mM Antipain Dihydrochloride (Thermo Fischer Scientific: 50488492), 20 mM Leupeptin (Gold Bio: L01025) and 0.8 mM Aprotinin (Gold Bio: A655100)]. A 1H-15N TROSY-HSQC spectrum was acquired first for 32 μM isolated 15N-C2B domain and additional 1H-15N TROSY-HSQC spectra were acquired after adding increasing concentrations of CpxSC to the sample, resulting in gradual dilution of the 15N-C2B domain. The protein concentrations of each titration step for each mutant are indicated in the legends of Fig. 4 and S15. Soluble SNARE complex was assembled as described in (3), concentrated at room temperature to a concentration above 250 μM using a 30 kDa centrifugation filter (Amicon) and exchanged into NMR buffer using Zeba Spin Desalting Columns, 7K MWCO, 10 mL (Thermo Fisher). Complexin-1 (26–83) was also concentrated above 250 μM using a 3 kDa centrifugation filter (Amicon) and exchanged into NMR buffer. SNARE-Complexin-1 (26–83) complex was preassembled with 20% excess Complexin-1 (26–83) before mixing with 15N-C2B domain. Total acquisition times ranged from 3.5 to 87.5 hr, depending on the sensitivity of the spectra. All NMR data were processed with NMRPipe and analyzed with NMRViewJ.

Bimane fluorescence quenching assay. All fluorescence emission scans were collected on a PTI Quantamaster 400 spectrofluorometer (T-format) at room temperature with slits set to 1.25 mm. For tryptophan-induced bimane fluorescence quenching assays, we used SNARE complexes anchored on nanodiscs as described (3) and labeled with bimane at position R59C of SNAP-25_N. The lipid composition of nanodiscs was 84% POPC, 15% DOPS, 1% PIP2. The experiments were performed in 25 mM HEPES pH 7.4 100 mM KCl 0.1 mM TCEP 2.5 mM MgCl2, 2 mM ATP 1 mM EGTA containing 1.5 μM BSA to prevent sample binding to the cuvette.  Fluorescence emission spectra (excitation at 380 nm) were acquired for samples containing 1 mM SC-59-bimane-nanodiscs alone or in the presence of 4 mM C2A-T285W without or with 2.0 mM CaCl2 (1.0 mM free Ca2+).

FRET assays. All fluorescence emission scans were collected on a PTI Quantamaster 400 spectrofluorometer (T-format) at room temperature with excitation at 550 nm and slits set to 1.25 mm. Each sample contained 0.125 μM C2AB-SNARE-complex-liposomes in 25 mM HEPES pH 7.4 100 mM KCl buffer, 1 mM EGTA, 2.5 mM MgCl2, 2 mM ATP, 1.5 μM BSA, and 0.15 μM full-length Complexin-1. Mg-ATP and complexin-1 were included to hinder non-specific interactions. To examine the reproducibility of Ca2+-induced changes in FRET, spectra were acquired for separate samples prepared under identical conditions before and after addition of 2.0 mM CaCl2 (1.0 mM free Ca2+). Because the Syt1 C2AB fragment and SNARE complex were fused covalently via the 37 aa linker and the SNARE complex is resistant to SDS, control spectra to measure the maximum donor fluorescence observable without FRET were acquired after incubation for 5 minutes at 37 oC with 0.4 µM NSF and 2 µM aSNAP (to disassemble the SNARE complex) in 25 mM HEPES pH 7.4 100 mM KCl buffer containing 1 mM EGTA, 2.5 mM MgCl2, 2 mM ATP, 1.5 μM BSA and 1% BOG. The presence of detergent proved to be necessary to recover maximum signal because a subset of the C2AB-SNARE-complexes get reconstituted inside the liposomes and hence are inaccessible to the disassembly machinery.

References

(1) J. Rizo, L. Sari, Y. Qi, W. Im, M. M. Lin, All-atom molecular dynamics simulations of Synaptotagmin-SNARE-complexin complexes bridging a vesicle and a flat lipid bilayer. Elife 11, e76356 (2022).

(2) J. Rizo, L. Sari, K. Jaczynska, C. Rosenmund, M. M. Lin, Molecular mechanism underlying SNARE-mediated membrane fusion enlightened by all-atom molecular dynamics simulations. Proc Natl Acad Sci U S A 121, e2321447121 (2024).

(3) R. Voleti, K. Jaczynska, J. Rizo, Ca(2+)-dependent release of Synaptotagmin-1 from the SNARE complex on phosphatidylinositol 4,5-bisphosphate-containing membranes. Elife 9, e57154 (2020).