elife03348.xml

<?xml version="1.0" encoding="UTF-8"?><!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.1d1 20130915//EN"  "JATS-archivearticle1.dtd"><article article-type="research-article" dtd-version="1.1d1" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><front><journal-meta><journal-id journal-id-type="nlm-ta">elife</journal-id><journal-id journal-id-type="hwp">eLife</journal-id><journal-id journal-id-type="publisher-id">eLife</journal-id><journal-title-group><journal-title>eLife</journal-title></journal-title-group><issn publication-format="electronic">2050-084X</issn><publisher><publisher-name>eLife Sciences Publications, Ltd</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">03348</article-id><article-id pub-id-type="doi">10.7554/eLife.03348</article-id><article-categories><subj-group subj-group-type="display-channel"><subject>Research article</subject></subj-group><subj-group subj-group-type="heading"><subject>Biophysics and structural biology</subject></subj-group><subj-group subj-group-type="heading"><subject>Cell biology</subject></subj-group></article-categories><title-group><article-title>Common intermediates and kinetics, but different energetics, in the assembly of SNARE proteins</article-title></title-group><contrib-group><contrib contrib-type="author" id="author-14217"><name><surname>Zorman</surname><given-names>Sylvain</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/><xref ref-type="fn" rid="con1"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-14218"><name><surname>Rebane</surname><given-names>Aleksander A</given-names></name><xref ref-type="aff" rid="aff3"/><xref ref-type="aff" rid="aff4"/><xref ref-type="fn" rid="con2"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-14219"><name><surname>Ma</surname><given-names>Lu</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="fn" rid="con3"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-14220"><name><surname>Yang</surname><given-names>Guangcan</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="fn" rid="con4"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-14221"><name><surname>Molski</surname><given-names>Matthew A</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/><xref ref-type="fn" rid="con6"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-14222"><name><surname>Coleman</surname><given-names>Jeff</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/><xref ref-type="fn" rid="con5"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-17855"><name><surname>Pincet</surname><given-names>Frederic</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/><xref ref-type="aff" rid="aff5"/><xref ref-type="fn" rid="con7"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" corresp="yes" id="author-1828"><name><surname>Rothman</surname><given-names>James E</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/><xref ref-type="corresp" rid="cor1">*</xref><xref ref-type="other" rid="par-2"/><xref ref-type="fn" rid="con8"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" corresp="yes" id="author-13975"><name><surname>Zhang</surname><given-names>Yongli</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/><xref ref-type="corresp" rid="cor2">*</xref><xref ref-type="other" rid="par-1"/><xref ref-type="fn" rid="con9"/><xref ref-type="fn" rid="conf1"/></contrib><aff id="aff1"><institution content-type="dept">Department of Cell Biology</institution>, <institution>Yale University School of Medicine</institution>, <addr-line><named-content content-type="city">New Haven</named-content></addr-line>, <country>United States</country></aff><aff id="aff2"><institution content-type="dept">Nanobiology Institute</institution>, <institution>Yale University</institution>, <addr-line><named-content content-type="city">West Haven</named-content></addr-line>, <country>United States</country></aff><aff id="aff3"><institution content-type="dept">Department of Physics</institution>, <institution>Yale University</institution>, <addr-line><named-content content-type="city">New Haven</named-content></addr-line>, <country>United States</country></aff><aff id="aff4"><institution content-type="dept">Integrated Graduate Program in Physical and Engineering Biology</institution>, <institution>Yale University</institution>, <addr-line><named-content content-type="city">New Haven</named-content></addr-line>, <country>United States</country></aff><aff id="aff5"><institution content-type="dept">Laboratoire de Physique Statistique</institution>, <institution>UMR CNRS 8550 Associée aux Universités Paris 6 et Paris 7, Ecole Normale Supérieure</institution>, <addr-line><named-content content-type="city">Paris</named-content></addr-line>, <country>France</country></aff></contrib-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Brunger</surname><given-names>Axel T</given-names></name><role>Reviewing editor</role><aff><institution>Stanford University</institution>, <country>United States</country></aff></contrib></contrib-group><author-notes><corresp id="cor1"><label>*</label>For correspondence: <email>james.rothman@yale.edu</email> (JER);</corresp><corresp id="cor2"><label>*</label>For correspondence: <email>yongli.zhang@yale.edu</email> (YZ)</corresp></author-notes><pub-date date-type="pub" publication-format="electronic"><day>01</day><month>09</month><year>2014</year></pub-date><pub-date pub-type="collection"><year>2014</year></pub-date><volume>3</volume><elocation-id>e03348</elocation-id><history><date date-type="received"><day>13</day><month>05</month><year>2014</year></date><date date-type="accepted"><day>29</day><month>08</month><year>2014</year></date></history><permissions><copyright-statement>© 2014, Zorman et al</copyright-statement><copyright-year>2014</copyright-year><copyright-holder>Zorman et al</copyright-holder><license xlink:href="http://creativecommons.org/licenses/by/4.0/"><license-p>This article is distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/4.0/">Creative Commons Attribution License</ext-link>, which permits unrestricted use and redistribution provided that the original author and source are credited.</license-p></license></permissions><self-uri content-type="pdf" xlink:href="elife03348.pdf"/><abstract><object-id pub-id-type="doi">10.7554/eLife.03348.001</object-id><p>Soluble <italic>N</italic>-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs) are evolutionarily conserved machines that couple their folding/assembly to membrane fusion. However, it is unclear how these processes are regulated and function. To determine these mechanisms, we characterized the folding energy and kinetics of four representative SNARE complexes at a single-molecule level using high-resolution optical tweezers. We found that all SNARE complexes assemble by the same step-wise zippering mechanism: slow N-terminal domain (NTD) association, a pause in a force-dependent half-zippered intermediate, and fast C-terminal domain (CTD) zippering. The energy release from CTD zippering differs for yeast (13 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>) and neuronal SNARE complexes (27 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>), and is concentrated at the C-terminal part of CTD zippering. Thus, SNARE complexes share a conserved zippering pathway and polarized energy release to efficiently drive membrane fusion, but generate different amounts of zippering energy to regulate fusion kinetics.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.001">http://dx.doi.org/10.7554/eLife.03348.001</ext-link></p></abstract><abstract abstract-type="executive-summary"><object-id pub-id-type="doi">10.7554/eLife.03348.002</object-id><title>eLife digest</title><p>Many processes in living things need molecules to be transported within, or between, cells. For example, damaged or waste molecules are transported within a cell to structures that can break the molecules down, while nerve impulses are transmitted from one neuron to the next via the release of signaling molecules.</p><p>Cells—and the compartments within cells—are surrounded by membranes that act as barriers to certain molecules. Vesicles are small, membrane-enclosed packages that are used to transport molecules between different membranes; and in order to release its cargo, a vesicle must fuse with its target membrane. To fuse like this, the forces that act to push membranes away from one another need to be overcome. Proteins called SNARES, which are embedded in both membranes, are the molecular engines that power the fusion process. Once the SNARE proteins from the vesicle and the target membrane bind, they assemble into a more compact complex that pulls the two membranes close together and allows fusion to take place.</p><p>The final shape of an assembled SNARE complex is essentially the same for all SNARE complexes; however, it is not known whether all of these complexes fold using the same method. Now Zorman et al. have used optical tweezers—an instrument that uses a highly focused laser beam to hold and manipulate microscopic objects—to observe the folding and unfolding of four different types of SNARE complex. All four SNARE complexes followed the same step-by-step process: the leading ends of the SNARE proteins slowly bound to each other; the process paused; then the rest of the proteins rapidly ‘zippered’ together.</p><p>Zorman et al. revealed that, although the steps in the processes were the same, the energy released in the last step was different when different complexes assembled. This suggests that the energy released by the ‘zippering’ of different SNARE proteins is optimized to match the required speed of different membrane fusion events. Furthermore, Zorman et al. propose that the reason why the majority of energy is released in the later stages of complex assembly is because this is when the repulsion between the two membranes is strongest.</p><p>The discoveries of Zorman et al. will now aid future efforts aimed at understanding better how the numerous other proteins that interact with SNARE proteins regulate the process of membrane fusion.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.002">http://dx.doi.org/10.7554/eLife.03348.002</ext-link></p></abstract><kwd-group kwd-group-type="author-keywords"><title>Author keywords</title><kwd>SNAREs</kwd><kwd>optical tweezers</kwd><kwd>protein folding</kwd><kwd>membrane fusion</kwd><kwd>SNARE assembly</kwd><kwd>energy landscape</kwd></kwd-group><kwd-group kwd-group-type="research-organism"><title>Research organism</title><kwd><italic>E. coli</italic></kwd></kwd-group><funding-group><award-group id="par-1"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>GM093341</award-id><principal-award-recipient><name><surname>Zhang</surname><given-names>Yongli</given-names></name></principal-award-recipient></award-group><award-group id="par-2"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>DK027044</award-id><principal-award-recipient><name><surname>Rothman</surname><given-names>James E</given-names></name></principal-award-recipient></award-group><funding-statement>The funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication.</funding-statement></funding-group><custom-meta-group><custom-meta><meta-name>elife-xml-version</meta-name><meta-value>2</meta-value></custom-meta><custom-meta specific-use="meta-only"><meta-name>Author impact statement</meta-name><meta-value>The energy landscape of SNARE folding and assembly is optimized for efficient stage-wise membrane fusion.</meta-value></custom-meta></custom-meta-group></article-meta></front><body><sec id="s1" sec-type="intro"><title>Introduction</title><p>Soluble <italic>N</italic>-ethylmaleimide-sensitive factor attachment protein receptor (SNARE)-mediated membrane fusion is ubiquitous in eukaryotes and underlies numerous basic processes in humans, including neurotransmission, hormone secretion, and antibody production (<xref ref-type="bibr" rid="bib43">Sollner et al., 1993</xref>; <xref ref-type="bibr" rid="bib47">Sudhof and Rothman, 2009</xref>; <xref ref-type="bibr" rid="bib52">Wickner, 2010</xref>; <xref ref-type="bibr" rid="bib15">Jahn and Fasshauer, 2012</xref>). Malfunction of fusion has been associated with many important diseases such as neurological disorders and diabetes (<xref ref-type="bibr" rid="bib4">Burre et al., 2010</xref>; <xref ref-type="bibr" rid="bib45">Stockli et al., 2011</xref>). Consistent with their diverse functions and dysfunctions, these intracellular membrane fusion processes exhibit distinct kinetics and regulation (<xref ref-type="bibr" rid="bib18">Kasai et al., 2012</xref>). For example, fusion of synaptic vesicles occurs within 0.2 ms in response to the arrival of an action potential (<xref ref-type="bibr" rid="bib37">Sabatini and Regehr, 1996</xref>), whereas vacuole fusion in yeast is constitutive and lasts minutes (<xref ref-type="bibr" rid="bib52">Wickner, 2010</xref>). Although these diverse processes have long been identified, it is not fully understood how SNAREs specialize in membrane fusion and become adapted to and regulated for various fusion speeds.</p><p>SNAREs constitute a large family of proteins with highly conserved modular structures (<xref ref-type="bibr" rid="bib10">Fasshauer et al., 1998</xref>), including 38 SNARE proteins in humans. Each SNARE protein contains one or two defining SNARE motifs of around 60 amino acids in eight heptad repeats (<xref ref-type="fig" rid="fig1">Figure 1A</xref>). The motif is often connected to a C-terminal transmembrane domain via a short linker domain (LD of ∼10 a.a.). Complementary SNAREs are anchored to transport vesicles (v-SNAREs) and their targeted membranes (t-SNAREs) in disordered or partially disordered conformations. Their specific interactions lead to coupled folding and assembly into a stable parallel four-helix bundle, drawing the two membranes into close proximity for fusion (<xref ref-type="bibr" rid="bib43">Sollner et al., 1993</xref>; <xref ref-type="bibr" rid="bib47">Sudhof and Rothman, 2009</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). In the core of each SNARE bundle are 15 layers of hydrophobic amino acids and one middle layer of ionic amino acids. The ionic layer is formed by three glutamine residues (Q) and one arginine residue (R) from each of the SNARE motifs categorized as Q<sub>a</sub>, Q<sub>b</sub>, Q<sub>c</sub>, and R SNAREs (<xref ref-type="bibr" rid="bib10">Fasshauer et al., 1998</xref>; <xref ref-type="fig" rid="fig1">Figure 1B</xref>). Crystal structures show that the four-helix bundle structures are highly conserved in different SNARE complexes (<xref ref-type="bibr" rid="bib48">Sutton et al., 1998</xref>; <xref ref-type="bibr" rid="bib58">Zwilling et al., 2007</xref>; <xref ref-type="bibr" rid="bib44">Stein et al., 2009</xref>), which can be aligned to the angstrom level (<xref ref-type="bibr" rid="bib46">Strop et al., 2008</xref>).<fig-group><fig id="fig1" position="float"><object-id pub-id-type="doi">10.7554/eLife.03348.003</object-id><label>Figure 1.</label><caption><title>Chimeric SNARE construct and experimental setup used to study functional assembly of single SNARE complexes using dual-trap high-resolution optical tweezers.</title><p>(<bold>A</bold>) Modular parallel four-helix bundle structure of an assembled neuronal SNARE complex mediating membrane fusion. The SNARE complex contains different functional domains: an N-terminal domain (NTD), an ionic layer, a C-terminal domain (CTD), a linker domain (LD), two transmembrane domains, and other domains not shown here. (<bold>B</bold>) Diagram showing the chimeric SNARE construct and the experimental setup. Each SNARE complex contains one SNARE motif from the four highly conserved Q<sub>a</sub>, Q<sub>b</sub>, Q<sub>c</sub>, and R SNARE families. These motifs are joined into one protein through spacer sequences (dashed lines) to facilitate the single-molecule manipulation experiment. The same color coding for different SNARE proteins is used throughout this work. See <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref> for complete sequences of the chimeric SNAREs and <xref ref-type="fig" rid="fig1s2 fig1s3 fig1s4">Figure 1—figure supplements 2–4</xref> for minimal effects of the spacer sequences on the folding energy and kinetics of the SNARE complexes.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.003">http://dx.doi.org/10.7554/eLife.03348.003</ext-link></p></caption><graphic xlink:href="elife03348f001"/></fig><fig id="fig1s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.004</object-id><label>Figure 1—figure supplement 1.</label><caption><title>Amino acid sequences of the chimeric SNARE protein constructs used for the single-molecule manipulation study of SNARE assembly.</title><p>The four sets of SNARE proteins are neuronal SNAREs of rat syntaxin (SX) 1A, rat VAMP2, and mouse SNAP-25B, GLUT4 SNAREs of rat syntaxin 4A, rat SNAP-23, and rat VAMP2, endosomal SNAREs of rat syntaxin 13, mouse Vti1a, human Stx6, and mouse VAMP4, and yeast SNAREs of Sso1, Sec9, and Snc2. SNARE sequences within and between different rows are aligned in their SNARE motifs, in which amino acids in the 15 hydrophobic layers and the ionic layer are shown in red and green, respectively, with layer numbers (from −7 to +8) indicated above the sequences. The amino acids in the SNARE linker domains (LDs) are shown in bold. In yeast R SNARE Snc2, the amino acids deleted in the LD- and C-terminal domain (CTD)-truncated constructs (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1</xref>) are underlined by blue dashed lines. All natural cysteine residues were mutated to serine (underlined and in italic). The pulling sites (cysteine and lysine) are indicated in blue. The spacer sequences (Sp1–Sp3) are derived from an unstructured sequence found in a POU transcription factor. The Avi-tag as a substrate for biotinylation (on lysine) is underlined by a black line.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.004">http://dx.doi.org/10.7554/eLife.03348.004</ext-link></p></caption><graphic xlink:href="elife03348fs001"/></fig><fig id="fig1s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.005</object-id><label>Figure 1—figure supplement 2.</label><caption><title>The chimeric neuronal SNARE protein correctly folds into an expected four-helix SNARE bundle.</title><p>Circular dichroism spectrum of the chimeric SNARE protein. The presence of two local minima at 208 nm and 222 nm indicates a high content of alpha-helical structure in the protein. Based on the CD spectrum, we estimated that about 50% of the amino acids in the protein are in an alpha-helical configuration, consistent with a fully folded SNARE four-helix bundle in the chimeric protein.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.005">http://dx.doi.org/10.7554/eLife.03348.005</ext-link></p></caption><graphic xlink:href="elife03348fs002"/></fig><fig id="fig1s3" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.006</object-id><label>Figure 1—figure supplement 3.</label><caption><title>The chimeric neuronal SNARE protein folds into a homogenous SNARE four-helix bundle with an expected molecular weight.</title><p>Gel filtration profile of the chimeric protein. The purified chimeric SNARE protein was eluted from the column (Superdex 200 10/300 GL) in one major peak corresponding to a molecular weight of 57 kDa, consistent with a folded SNARE four-helix bundle.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.006">http://dx.doi.org/10.7554/eLife.03348.006</ext-link></p></caption><graphic xlink:href="elife03348fs003"/></fig><fig id="fig1s4" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.007</object-id><label>Figure 1—figure supplement 4.</label><caption><title>The chimeric t-SNARE protein supports lipid mixing between liposomes.</title><p>(<bold>A</bold>) Schematic representation of the chimeric t-SNARE and the v-SNARE used for the lipid-mixing assay. The chimeric t-SNARE protein is lipidated through a cysteine introduced to the C-terminus of syntaxin. (<bold>B</bold>) Comparison of membrane fusion activity mediated by the chimeric t-SNARE (blue) and the transmembrane domain truncated wild-type t-SNARE complex (black) in the fluorescence de-quenching assay. Both t-SNAREs were anchored to membrane through a unique cysteine at the C-termini of syntaxin. The fusion depends on formation of trans-SNARE complexes, as pre-incubation of t-SNARE proteoliposomes with the cytoplasmic domain of VAMP2 (CDV) inhibits membrane fusion (green). The reduced fusion activity of the chimeric t-SNARE compared to the wild-type t-SNARE was caused by the spacer sequence between the C-terminus of syntaxin and the cysteine residue (<xref ref-type="bibr" rid="bib30">McNew et al., 2000</xref>; <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref>).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.007">http://dx.doi.org/10.7554/eLife.03348.007</ext-link></p></caption><graphic xlink:href="elife03348fs004"/></fig></fig-group></p><p>The conserved sequences of SNAREs and their similar initial and final conformations implicate a conserved pathway of SNARE folding/assembly. However, the kinetics and energetics of SNARE folding have not been well characterized. It is notoriously difficult to study SNARE assembly using traditional ensemble-based experimental approaches due to the many states and pathways involved in the folding process, especially misassembled states (<xref ref-type="bibr" rid="bib51">Weninger et al., 2003</xref>; <xref ref-type="bibr" rid="bib36">Pobbati et al., 2006</xref>). In addition, functional SNARE assembly occurs in the presence of the opposing force imposed by membranes, which has a great impact on the kinetics and regulation of SNARE assembly (<xref ref-type="bibr" rid="bib47">Sudhof and Rothman, 2009</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). Although studies of SNARE assembly are facilitated by the use of soluble SNAREs isolated from membranes, the lack of an essential force load may complicate data interpretation regarding functional SNARE assembly. For example, whereas complexin can suspend assembly of trans-SNAREs in a partially zippered state (<xref ref-type="bibr" rid="bib19">Kummel et al., 2011</xref>; <xref ref-type="bibr" rid="bib23">Li et al., 2011</xref>; <xref ref-type="bibr" rid="bib28">Malsam et al., 2012</xref>), it cannot do so for isolated SNAREs (<xref ref-type="bibr" rid="bib7">Chen et al., 2002</xref>). Thus, new methods are required to better elucidate SNARE assembly.</p><p>Recently, we have applied high-resolution optical tweezers to quantitatively characterize the energetics and kinetics of neuronal SNARE folding for the first time (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). This single-molecule manipulation method allows measurement of the folding energy and kinetics of macromolecules under equilibrium conditions (<xref ref-type="bibr" rid="bib24">Liphardt et al., 2001</xref>). Furthermore, the external force applied to the SNARE complex mimics the opposing force from membranes (<xref ref-type="bibr" rid="bib22">Li et al., 2007</xref>; <xref ref-type="bibr" rid="bib25">Liu et al., 2009</xref>; <xref ref-type="bibr" rid="bib31">Min et al., 2013</xref>). Using this single-molecule method, we proved the long-standing hypothesis that neuronal SNAREs assemble by a zippering mechanism and discovered a half-zippered SNARE intermediate that plays a crucial role in the synchronized, calcium-triggered synaptic vesicle fusion (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). Step-wise SNARE zippering is initiated by slow association between N-terminal domains (NTDs) of t- and v-SNAREs. SNARE assembly then pauses in the half-zippered state in a force-dependent manner. Finally, the C-terminal domain (CTD) of the v-SNARE (VAMP2 or synaptobrevin) rapidly zippers along the pre-structured t-SNARE template to drive fast membrane fusion.</p><p>It is unknown whether other SNAREs assemble by the same zippering mechanism. Furthermore, it is not clear how SNARE assembly is adapted to efficient and versatile membrane fusion. It has been proposed that SNAREs generally assemble in an all-or-none manner without any partially folded intermediates (<xref ref-type="bibr" rid="bib15">Jahn and Fasshauer, 2012</xref>; <xref ref-type="bibr" rid="bib18">Kasai et al., 2012</xref>). It is argued that assembly of neuronal SNARE complexes occurs in a large energy gradient, and thus cannot be stopped halfway to form any partially assembled intermediates (<xref ref-type="bibr" rid="bib15">Jahn and Fasshauer, 2012</xref>). However, despite its fast speed, downhill SNARE assembly would be poorly coupled to membrane fusion, resulting in low energy efficiency of the SNARE engine. In contrast, many molecular engines tested at a single-molecule level have nearly 100% energy efficiency (<xref ref-type="bibr" rid="bib5">Bustamante et al., 2004</xref>). Based on the first law of thermodynamics, a mechanochemical process has 100% energy efficiency only when the process is reversible. Therefore, to maximize their energy efficiency, SNAREs are expected to fold in a relatively smooth energy landscape (<xref ref-type="bibr" rid="bib35">Onuchic and Wolynes, 2004</xref>) in the presence of the membrane load. This requires a close match between the energy landscape of SNARE assembly and the energy profile of membrane interactions. The energy opposing membrane fusion includes contributions from the long-ranged entropic membrane undulation, membrane deformation, and electrostatic interactions, and the short-ranged membrane dehydration and van der Waals interactions (<xref ref-type="bibr" rid="bib20">Leckband and Israelachvili, 2001</xref>). The strong short-ranged repulsion is the largest energy barrier for fusion and takes place within a few nanometers of membrane separation, thus constituting a hard core for fusion. To break this hard core, a SNARE complex is required to focus its folding energy on the membrane proximal C-terminus. Therefore, analogous to a car engine, an efficient SNARE engine is expected to change gears to meet increasing resistance as SNAREs fold towards membranes. However, it remains unclear whether such a gear-changing mechanism exists in SNARE assembly.</p><p>To address the above questions, we measured the folding energy and kinetics of four representative SNARE complexes at a single-molecule level, using high-resolution optical tweezers and a new chimeric SNARE design (<xref ref-type="fig" rid="fig1">Figure 1</xref>). These complexes mediate highly regulated exocytosis of neurotransmitters in pre-synaptic neurons (neuronal SNAREs: syntaxin 1, SNAP-25B, and VAMP2 or synaptobrevin) (<xref ref-type="bibr" rid="bib43">Sollner et al., 1993</xref>) and translocation of glucose transporter type 4 (GLUT4) in adipocytes or muscle cells (GLUT4 SNAREs: syntaxin 4, SNAP-23, and VAMP2) (<xref ref-type="bibr" rid="bib1">Bai et al., 2007</xref>; <xref ref-type="bibr" rid="bib45">Stockli et al., 2011</xref>). The complexes also affect constitutive fusion of endocytic vesicles to early endosome in mammals (endosomal SNAREs: syntaxin 13, Vti1A, syntaxin 6, and VAMP4) (<xref ref-type="bibr" rid="bib58">Zwilling et al., 2007</xref>) and fusion of post-Golgi vesicles to plasma membranes in yeast (yeast SNAREs: Sso1, Sec9, and Snc2) (<xref ref-type="bibr" rid="bib46">Strop et al., 2008</xref>). All four of these SNARE complexes were chosen for our study because they represent SNAREs in diverse evolutionary species, have different degrees of regulation, and mediate fusion with a speed ranging from 0.2 ms to 20 min (<xref ref-type="bibr" rid="bib18">Kasai et al., 2012</xref>). In addition, the crystal structures of neuronal, endosomal, and yeast SNARE complexes are available (<xref ref-type="bibr" rid="bib48">Sutton et al., 1998</xref>; <xref ref-type="bibr" rid="bib58">Zwilling et al., 2007</xref>; <xref ref-type="bibr" rid="bib46">Strop et al., 2008</xref>; <xref ref-type="bibr" rid="bib44">Stein et al., 2009</xref>), which facilitates derivation of their various assembly intermediates from our single-molecule measurements (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>), allowing us to compare the folding pathways and energy landscapes of different SNARE complexes.</p><p>Our results show that all four SNARE complexes assemble via the same zippering mechanism in three sequential steps: slow NTD association, fast CTD zippering, and finally rapid LD zippering. However, the CTD zippering energy of different SNARE complexes varies greatly and is highly concentrated at the C-terminus.</p></sec><sec id="s2" sec-type="results"><title>Results</title><sec id="s2-1"><title>Chimeric SNARE complex and experimental setup</title><p>To facilitate protein preparation and single-molecule experiments, we constructed new chimeric SNARE proteins in which three or four cognate SNARE proteins were joined into one polypeptide with the addition of two or three spacer sequences (<xref ref-type="fig" rid="fig1">Figure 1</xref>). Individual cytoplasmic SNARE sequences were truncated and regions that directly participate in SNARE complex formation were kept (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref>). To minimize their perturbation on the structure and dynamics of SNARE complexes, the spacer sequences were chosen to be unstructured and of proper length. Each chimeric SNARE protein consisted of a unique cysteine at the C-terminus of Q<sub>a</sub> SNARE and an Avi-tag at the C-terminus of R SNARE used to pull the single SNARE complex (<xref ref-type="fig" rid="fig1">Figure 1B</xref>).</p><p>We first examined the structural and functional integrity of the chimeric SNARE complexes. For this purpose, the recombinant proteins were purified and biotinylated in vitro. The expected helical bundles that formed were confirmed by circular dichroism spectra and gel filtration profiles (<xref ref-type="fig" rid="fig1s2 fig1s3">Figure 1—figure supplements 2 and 3</xref>). To test the function of the SNARE protein, we similarly made a chimeric neuronal t-SNARE protein and tested its ability to mediate lipid mixing with full-length VAMP2 (<xref ref-type="fig" rid="fig1s4">Figure 1—figure supplement 4</xref>). We found that the t-SNARE protein was as fusogenic as the wild-type cytoplasmic t-SNARE complex that is covalently linked to the membrane (<xref ref-type="bibr" rid="bib30">McNew et al., 2000</xref>). This result suggests that the spacer sequence between syntaxin and SNAP-25 does not significantly interfere with SNARE assembly and membrane fusion. Furthermore, the chimeric neuronal SNARE complex reveals folding energy and kinetics (see below) consistent with our recent reports based on a different SNARE construct in which syntaxin and VAMP2 were cross-linked at their N-termini by a disulfide bond (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). Taken together, the chimeric SNARE proteins mimic their corresponding SNARE complexes and can be used to facilitate the study of SNARE assembly at a single-molecule level. We refer to these proteins as SNARE complexes.</p><p>The SNARE complexes were cross-linked to a 2260 bp DNA handle (<xref ref-type="bibr" rid="bib6">Cecconi et al., 2005</xref>) and tethered to two polystyrene beads held in two optical traps of high-resolution optical tweezers (<xref ref-type="bibr" rid="bib32">Moffitt et al., 2006</xref>; <xref ref-type="bibr" rid="bib40">Sirinakis et al., 2011</xref>; <xref ref-type="fig" rid="fig1">Figure 1B</xref>). Single SNARE complexes were pulled from the C-termini of Q<sub>a</sub> and R SNAREs by moving one trap relative to another at a constant speed, typically 10 nm/s. The tension and extension of the protein-DNA tether were recorded at 10 kHz and used to derive protein folding energy and kinetics.</p></sec><sec id="s2-2"><title>Common intermediates and pathways of SNARE assembly</title><p>When pulled to a force up to 25 pN, all four SNARE-DNA tethers extended continuously in some force ranges, but discontinuously in other ranges (<xref ref-type="fig" rid="fig2">Figure 2A,B</xref>). The continuous extension increase was mainly caused by stretching of the semi-flexible DNA handle while the SNARE complex remained in the same folding state. The resultant force-extension curves (FECs) could generally be fit by the worm-like chain model of the DNA and polypeptide (<xref ref-type="bibr" rid="bib29">Marko and Siggia, 1995</xref>). In contrast, abrupt extension changes resulted from cooperative protein transitions between different states (<xref ref-type="fig" rid="fig2">Figure 2C</xref>). The FECs show that all four SNARE complexes sequentially unfolded via two reversible transitions and one or two irreversible unfolding steps. Compared to the FECs reported for the neuronal SNARE complex (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>) and confirmed by the detailed analysis described below, the second reversible transition (between state 2 and state 3) and the first irreversible transition (between state 3 and state 4) resulted from folding/unfolding transitions of CTD and NTD, respectively. Both transitions are energetically or kinetically distinct for each of the four SNARE complexes, as is demonstrated by non-overlapping distributions of the characteristic forces or different lifetimes associated with these transitions (<xref ref-type="fig" rid="fig3">Figure 3</xref>). In particular, NTD is mechanically more stable than CTD and unfolded generally after 10–10<sup>5</sup> CTD folding and unfolding transitions under our experimental conditions (<xref ref-type="fig" rid="fig2">Figure 2B</xref>).<fig-group><fig id="fig2" position="float"><object-id pub-id-type="doi">10.7554/eLife.03348.008</object-id><label>Figure 2.</label><caption><title>Four representative SNARE complexes assemble or disassemble via common intermediates and pathways.</title><p>(<bold>A</bold>) Force-extension curves (FECs) of the neuronal, GLUT4, endosomal, and yeast SNARE complexes. FECs were obtained by pulling the complexes (black) or relaxing them (gray). The reversible C-terminal domain (CTD) and linker domain (LD) folding/unfolding transitions are marked by blue solid and dashed ovals, respectively, whereas irreversible unfolding of the partially zippered SNARE complex is indicated by a red arrow. Continuous FEC regions can be fit by the worm-like chain model and represent different SNARE states numbered as in (<bold>C</bold>). Below ∼6 pN, deviation of some fits from the measured FECs corresponding to the unfolded complex (state 5) may be caused by intramolecular interactions or partial refolding of the complex. The t-SNARE state can be identified from some FECs, with a transient one (∼20 ms) marked by a cyan rectangle. (<bold>B</bold>) Time-dependent extension, force, and trap separation corresponding to the CTD and N-terminal domain (NTD) transition region in the FEC of the GLUT4 SNARE complex in A (marked by two magenta dots). In the upper and middle panels, the positions of different SNARE folding states are indicated by red dashed lines. About 90 CTD transitions occurred before NTD unzipping and reaching ∼18.6 pN equilibrium force (indicated by a red arrow in the middle panel). Here the equilibrium force for a two-state protein folding/unfolding process is defined as the average state forces (marked by dashed lines) under which the folded and the unfolded states are equally populated. Note that most NTD unzipping took place in the CTD-unfolded state (state 3). In the middle panel, the first CTD and the first NTD unzipping events during the pulling process are indicated by green dots and their time and force differences indicated. The time and force distributions are shown in <xref ref-type="fig" rid="fig3">Figure 3B,C</xref>. In the bottom panel, the separation between two optical traps was increasing at a speed of 10 nm/s to slowly pull the single SNARE complex. (<bold>C</bold>) Different SNARE assembly states partly derived from model-fitting of FECs shown in (<bold>A</bold>). Gray arrows indicate the pulling direction. Data associated with all FECs shown in this work were mean-filtered using 5 ms time window. See more FECs and their associated features in <xref ref-type="fig" rid="fig2s1 fig2s2 fig2s3 fig2s4">Figure 2—figure supplements 1–4</xref>.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.008">http://dx.doi.org/10.7554/eLife.03348.008</ext-link></p></caption><graphic xlink:href="elife03348f002"/></fig><fig id="fig2s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.009</object-id><label>Figure 2—figure supplement 1.</label><caption><title>Distinct linker domain and C-terminal domain transitions.</title><p>Force-extension curves (FECs) of yeast SNARE complexes containing the v-SNARE Snc2 truncated in the linker domain (LD) region or in both the LD region and part of the C-terminal domain (CTD) region (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref>). The CTD transition in the LD-truncated complex is marked by a solid blue oval. The cooperative SNARE reassembly is indicated by a black arrow.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.009">http://dx.doi.org/10.7554/eLife.03348.009</ext-link></p></caption><graphic xlink:href="elife03348fs005"/></fig><fig id="fig2s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.010</object-id><label>Figure 2—figure supplement 2.</label><caption><title>The neuronal t-SNARE complex as a transient unfolding intermediate of the half-zippered SNARE complex.</title><p>Force-extension curves (FECs) of a single neuronal SNARE complex corresponding to three pulling cycles. The inset shows close-up views of the unfolding process of the half-zippered SNARE complex in extension-time trajectories. The transient t-SNARE complex appeared in the second pulling cycle (indicated by a cyan rectangle), but was not discernible in the other two pulling cycles. Note that the C-terminal domain (CTD) transition seen in the second pulling cycle was significantly slower than those in the other cycles.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.010">http://dx.doi.org/10.7554/eLife.03348.010</ext-link></p></caption><graphic xlink:href="elife03348fs006"/></fig><fig id="fig2s3" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.011</object-id><label>Figure 2—figure supplement 3.</label><caption><title>The GLUT4 t-SNARE complex as a transient unfolding intermediate of the half-zippered SNARE complex.</title><p>Force-extension curves (FECs) of the same GLUT4 SNARE complex corresponding to three pulling cycles. The overlapping FECs (‘All’) and individual FECs from successive pulling cycles (#1–#3) were shifted along the x-axis for better comparison. The cooperative SNARE reassembly is indicated by a black arrow.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.011">http://dx.doi.org/10.7554/eLife.03348.011</ext-link></p></caption><graphic xlink:href="elife03348fs007"/></fig><fig id="fig2s4" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.012</object-id><label>Figure 2—figure supplement 4.</label><caption><title>The yeast t-SNARE complex is a stable unfolding intermediate of the half-zippered SNARE complex.</title><p>Force-extension curves (FECs) of a single yeast SNARE complex corresponding to two successive pulling cycles and their best fits by the worm-like chain model in the continuous phases. The cooperative SNARE reassembly is indicated by black arrows. The FECs corresponding to the completely unfolded SNARE complex are shown in gray.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.012">http://dx.doi.org/10.7554/eLife.03348.012</ext-link></p></caption><graphic xlink:href="elife03348fs008"/></fig></fig-group><fig id="fig3" position="float"><object-id pub-id-type="doi">10.7554/eLife.03348.013</object-id><label>Figure 3.</label><caption><title>Distinct transition kinetics and stabilities of SNARE C-terminal domain and N-terminal domain.</title><p>(<bold>A</bold>) Histogram distributions of the C-terminal domain (CTD) equilibrium force and the N-terminal domain (NTD) unzipping force for different SNARE complexes. The vertical axis shows the percentage of the event number in each bin. The average CTD equilibrium force (f<sub>1/2</sub>) or NTD unzipping force (f<sub>unzip</sub>) scored on the total numbers of transition events (N<sub>T</sub>) and single SNARE complexes (N<sub>m</sub>) are indicated, with the number in parenthesis designating the standard deviation of the mean. (<bold>B, C</bold>) Histogram distributions of the force and time differences of the first NTD and CTD unzipping events (<xref ref-type="fig" rid="fig2">Figure 2B</xref>). The average force difference (Δf<sub>unzip</sub>) or time difference (Δt<sub>unzip</sub>) is indicated. The distinct CTD and NTD transition kinetics are revealed by non-overlapping force distributions for neuronal, endosomal, and yeast SNARE complexes or significant force and time differences associated with the first unzipping events of CTD and NTD of the GLUT4 SNARE complex. Note that optical tweezers have a force measurement accuracy of 10% absolute forces between different single molecules and of &lt;0.1 pN relative forces within same single molecules (<xref ref-type="bibr" rid="bib32">Moffitt et al., 2006</xref>).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.013">http://dx.doi.org/10.7554/eLife.03348.013</ext-link></p></caption><graphic xlink:href="elife03348f003"/></fig></p><p>After the last irreversible unfolding event, the FECs obtained by pulling proteins to higher forces (&gt;25 pN) did not show any additional discontinuous extension changes (<xref ref-type="fig" rid="fig2">Figure 2A</xref>), indicating that the SNARE complexes had been completely unfolded. When relaxed, the SNARE complex remained unfolded until the force was dropped to ∼4 pN, leading to a large hysteresis in the FECs. Further relaxation of the complex to lower forces led to FECs overlapping with those of the FECs in the pulling phase, often with small and sudden extension drops manifesting cooperative reassembly of SNARE complexes (<xref ref-type="fig" rid="fig4">Figure 4</xref>). Additional cycles of pulling and relaxation generally revealed overlapping FECs, suggesting that the SNARE complexes could fully reassemble into nearly identical structures under our experimental conditions.<fig id="fig4" position="float"><object-id pub-id-type="doi">10.7554/eLife.03348.014</object-id><label>Figure 4.</label><caption><title>Overlapping force-extension curves obtained by repeatedly pulling a single neuronal or GLUT4 SNARE complex, revealing robust and common step-wise SNARE assembly and disassembly.</title><p>The overlapping force-extension curves (FECs) (designated by ‘All’) are shifted along the x-axis to reveal individual FECs corresponding to different pulling cycles (numbered). The cooperative reassembly events are indicated by black arrows. The neuronal SNARE-DNA tether broke in the third pulling cycle of the neuronal SNARE complex at the maximum pulling force. The GLUT4 SNARE complex unfolded at 2.5 pN force (red arrow) in the last pulling cycle, indicating that the complex was not properly assembled at the end of the fifth pulling cycle. Note that heterogeneity in SNARE zippering was observed, a phenomenon also seen in many single-molecule experiments (<xref ref-type="bibr" rid="bib26">Lu et al., 1998</xref>; <xref ref-type="bibr" rid="bib40">Sirinakis et al., 2011</xref>). The heterogeneity is manifested by changes in the rate and/or the equilibrium force of the C-terminal domain (CTD) transition detected in different pulling cycles of the same chimeric SNARE protein. For the single neuronal SNARE protein shown here, both equilibrium force and rate of the CTD transition are slightly lower in the first pulling cycle than in the following two cycles. More heterogeneity can be seen in <xref ref-type="fig" rid="fig2s2 fig2s3">Figure 2—figure supplements 2 and 3</xref>.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.014">http://dx.doi.org/10.7554/eLife.03348.014</ext-link></p></caption><graphic xlink:href="elife03348f004"/></fig></p><p>The reversible SNARE transitions could be better observed under approximately constant forces (<xref ref-type="fig" rid="fig5">Figure 5A</xref>, <xref ref-type="fig" rid="fig6">Figure 6A</xref>). In this case, the time-dependent extension change represented spontaneous folding/unfolding transition of the protein under tension due to thermal fluctuations. Both transitions in each of the four SNARE complexes were binary, as indicated by the two peaks in the histogram distributions of extension (<xref ref-type="fig" rid="fig5">Figure 5B</xref>, <xref ref-type="fig" rid="fig6">Figure 6B</xref>). The transitions remained cooperative at all forces tested, but were shifted to unfolded states at higher forces. Furthermore, the four SNARE complexes had similar average extension changes for both transitions (<xref ref-type="table" rid="tbl1">Table 1</xref>), implying that the same SNARE domains were involved in the observed transitions. Taken together, the results from experiments in variable and constant forces suggest that all four SNARE complexes follow similar pathways to assembly or disassemble via at least two intermediates.<fig id="fig5" position="float"><object-id pub-id-type="doi">10.7554/eLife.03348.015</object-id><label>Figure 5.</label><caption><title>Comparison of the two-state C-terminal domain transitions of four SNARE complexes.</title><p>(<bold>A</bold>) Force-dependent extension-time trajectories under approximately constant forces (f) revealing the unfolding probability (p) of C-terminal domain (CTD) as indicated. The idealized two-state transitions (red lines) were calculated based on a hidden Markov model (HMM). (<bold>B</bold>) Histogram distributions of the extensions shown in <bold>A</bold> (symbols) and their best fits with double-Gaussian functions (lines). For best comparison, the distributions for each SNARE complex were shifted along the x-axis to align them at the same average position of the unfolded CTD state. Distributions at different forces are color-coded as the corresponding extension traces in <bold>A</bold>. All the extension-time trajectories shown in this work were mean-filtered using a 1 ms time window. (<bold>C</bold>) CTD unfolding probabilities of four SNARE complexes. (<bold>D</bold>) The corresponding folding rates (hollow symbols) and unfolding rates (solid symbols) of CTD transitions. The best-fit unfolding probability (solid line), folding rate (dashed line), and unfolding rate (solid line) were obtained by non-linear least-squares fitting using a simplified energy-landscape model of SNARE assembly (‘Materials and methods’).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.015">http://dx.doi.org/10.7554/eLife.03348.015</ext-link></p></caption><graphic xlink:href="elife03348f005"/></fig><fig-group><fig id="fig6" position="float"><object-id pub-id-type="doi">10.7554/eLife.03348.016</object-id><label>Figure 6.</label><caption><title>Comparison of the two-state linker domain transitions of four SNARE complexes.</title><p>(<bold>A</bold>) Extension-time trajectories (black lines) and their best hidden Markov model (HMM) fits (red lines) showing fast binary transitions of linker domains (LDs) under constant forces. The force (f) and unfolding probability (p) are indicated. (<bold>B</bold>) Extension histogram distributions corresponding to the trajectories in <bold>A</bold> (symbols) and their best fits with double-Gaussian functions (red lines). (<bold>C</bold>) Force-dependent unfolding probability and transition rates (symbols) and their best fits (solid or dashed lines) of GLUT LD. Similar data corresponding to other SNARE complexes are shown in <xref ref-type="fig" rid="fig6s1">Figure 6—figure supplement 1</xref>.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.016">http://dx.doi.org/10.7554/eLife.03348.016</ext-link></p></caption><graphic xlink:href="elife03348f006"/></fig><fig id="fig6s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.017</object-id><label>Figure 6—figure supplement 1.</label><caption><title>Folding energy and kinetics of SNARE linker domains.</title><p>Unfolding probabilities (top panel) and folding rates (open symbols in bottom panel) or unfolding rates (solid symbols) of linker domains (LDs) in the yeast, endosomal, and neuronal SNARE complexes. The corresponding best fits based on the energy landscape model (‘Materials and methods’) are shown as solid or dashed lines.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.017">http://dx.doi.org/10.7554/eLife.03348.017</ext-link></p></caption><graphic xlink:href="elife03348fs009"/></fig><fig id="fig6s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.018</object-id><label>Figure 6—figure supplement 2.</label><caption><title>Minor effect of the spacer sequences in the chimeric SNARE proteins on the folding energy of SNARE complexes.</title><p>Predicted structures of the folded and unfolded states in linker domain (LD) transition and the accompanying extension change of the spacer sequence connecting syntaxin and SNAP-25 (red dashed line). The apparent LD folding energy measured by optical tweezers (Δ<italic>G</italic><sub><italic>app</italic></sub>) contains true LD folding energy (Δ<italic>G</italic><sub><italic>LD</italic></sub>) and the free energy change of the spacer sequence due to its extension change (Δ<italic>G</italic><sub><italic>sp</italic></sub>), that is,<disp-formula id="equ6"><label>(6)</label><mml:math id="m6"><mml:mrow><mml:mi>Δ</mml:mi><mml:msub><mml:mi>G</mml:mi><mml:mrow><mml:mi>a</mml:mi><mml:mi>p</mml:mi><mml:mi>p</mml:mi></mml:mrow></mml:msub><mml:mo>=</mml:mo><mml:mi>Δ</mml:mi><mml:msub><mml:mi>G</mml:mi><mml:mrow><mml:mi>L</mml:mi><mml:mi>D</mml:mi></mml:mrow></mml:msub><mml:mo>+</mml:mo><mml:mi>Δ</mml:mi><mml:msub><mml:mi>G</mml:mi><mml:mrow><mml:mi>s</mml:mi><mml:mi>p</mml:mi></mml:mrow></mml:msub><mml:mo>.</mml:mo></mml:mrow></mml:math></disp-formula></p><p>To correct for the true LD folding energy, we estimated the spacer energy change using a worm-like chain model. In the LD folded state, the spacer sequence is 65 a.a. long or 23.7 nm in contour length and has an extension of 11.8 nm, or the distance between point 1 (SNAP-25 R8) and point 2 (syntaxin R260). Thus, the ratio (<italic>r</italic>) of the extension to the contour length (<italic>l</italic>) is 0.5. Using <xref ref-type="disp-formula" rid="equ3">Equation 3</xref>, we calculated the free energy of the spacer sequence in the folded LD state to be 9 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>. In the unfolded LD state, the spacer sequence (including the unfolded LD sequence in syntaxin) becomes 73 a.a. long or 26.6 nm in contour length, with an extension of 10.6 nm (the distance between point 1 and point 3 at syntaxin A254). The free energy of the spacer in the unfolded LD state was similarly calculated to be 5.9 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>. Thus, the energy change of the spacer sequence upon LD folding is Δ<italic>G</italic><sub><italic>sp</italic></sub> = 3.1 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>. The apparent energy change derived from the singe-molecule measurement is Δ<italic>G</italic><sub><italic>app</italic></sub> = −6.6 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>. Thus the LD folding energy Δ<italic>G</italic><sub><italic>LD</italic></sub> = Δ<italic>G</italic><sub><italic>app</italic></sub> − Δ<italic>G</italic><sub><italic>sp</italic></sub> = −9.7 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>.</p><p>The two spacer sequences in the chimeric neuronal SNARE protein have no effect on the C-terminal domain (CTD) transition, because their extensions do not change in this process. The spacer sequence connecting SNAP-25 and VAMP2 (not shown) slightly decreases N-terminal domain (NTD) association energy by 4.3 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>. However, this reduced NTD stability does not alter its much greater lifetime than the CTD.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.018">http://dx.doi.org/10.7554/eLife.03348.018</ext-link></p></caption><graphic xlink:href="elife03348fs010"/></fig></fig-group><table-wrap id="tbl1" position="float"><object-id pub-id-type="doi">10.7554/eLife.03348.019</object-id><label>Table 1.</label><caption><p>Average equilibrium force, extension change, folding energy, and folding energy barrier and position associated with C-terminal domain and linker domain transitions of the four different SNARE complexes</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.019">http://dx.doi.org/10.7554/eLife.03348.019</ext-link></p></caption><table frame="hsides" rules="groups"><thead><tr><th rowspan="2">SNARE complex</th><th colspan="5">C-terminal domain</th><th colspan="5">Linker domain</th></tr><tr><th>Force (pN)</th><th>Extension change (nm)</th><th>Folding energy (<italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>)</th><th>Transition state energy<xref ref-type="tblfn" rid="tblfn1">*</xref> (<italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>)</th><th>Transition state position<xref ref-type="tblfn" rid="tblfn2">†</xref> (a.a.)</th><th>Force (pN)</th><th>Extension change (nm)</th><th>Folding energy (<italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>)</th><th>Transition state energy<xref ref-type="tblfn" rid="tblfn1">*</xref> (<italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>)</th><th>Transition state position<xref ref-type="tblfn" rid="tblfn2">†</xref> (a.a.)</th></tr></thead><tbody><tr><td>Neuron</td><td>16.2 (0.9)</td><td>7.2 (1.2)</td><td>−27 (4.7)</td><td>−5.5 (1.5)</td><td>17 (3)</td><td>8 (1)</td><td>4.7 (0.5)</td><td>−9.7 (1.6)</td><td>5.5 (1.5)</td><td>31 (1)</td></tr><tr><td>GLUT4</td><td>18.5 (1.8)</td><td>6.0 (0.9)</td><td>−23 (4.1)</td><td>−0.8 (1.0)</td><td>11 (2)</td><td>8.6 (0.9)</td><td>5.6 (1.1)</td><td>−12 (2.7)</td><td>2 (1.0)</td><td>30 (1)</td></tr><tr><td>Endosome</td><td>11.9 (0.9)</td><td>6.9 (0.4)</td><td>−16 (1.5)</td><td>2.1 (1.4)</td><td>12 (2)</td><td>6.3 (1.2)</td><td>5.1 (1.8)</td><td>−6.1 (2.4)</td><td>4.9 (1.5)</td><td>32 (2)</td></tr><tr><td>Yeast</td><td>10.1 (1.4)</td><td>5.8 (0.8)</td><td>−13 (2.5)</td><td>3.2 (1.5)</td><td>13 (2)</td><td>6.0 (1.6)</td><td>5.1 (1.2)</td><td>−5.7 (2.0)</td><td>3.6 (2.0)</td><td>32 (2)</td></tr></tbody></table><table-wrap-foot><fn id="tblfn1"><label>*</label><p>Here, negative energy indicates downhill protein folding (<xref ref-type="bibr" rid="bib55">Yang and Gruebele, 2003</xref>).</p></fn><fn id="tblfn2"><label>†</label><p>The number of the amino acids in the R SNARE C-terminal to the ionic layer (chosen as 0).</p></fn><fn><p>The equilibrium force and extension change were determined at an unfolding probability of 0.5 for the two-state processes. The standard deviations of the averages are shown in parenthesis. The equilibrium force distribution, the number of transitions, and the number of single molecules scored for C-terminal domain (CTD) transitions are shown in <xref ref-type="fig" rid="fig3">Figure 3</xref>. For parameters related to linker domain (LD) transitions, a total of 18, 35, 11, and 24 LD transitions in single neuronal, GLUT4, endosomal, and yeast SNARE complexes were scored, respectively.</p></fn></table-wrap-foot></table-wrap></p><p>To derive the structures of the intermediates observed in our experiments, we fit the continuous regions of the FECs using a quantitative model of the protein-DNA conjugate previously reported (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib53">Xi et al., 2012</xref>). In this model, the extension of the structured portion of the SNARE complex is force-independent, but varies as the SNARE complex changes its conformation (‘Materials and methods’). The model generally fit the measured FECs and extension changes obtained at constant forces well (<xref ref-type="fig" rid="fig2">Figure 2A</xref>). Extensive analysis revealed two common intermediates for the four SNARE complexes: the LD-unfolded state and the half-zippered state (<xref ref-type="fig" rid="fig2">Figure 2C</xref>). In the former, the SNARE LD was unfolded, while its CTD remained approximately intact. In the latter, the C-terminal half of the R SNARE was unzipped, whereas three Q SNARE motifs remained intact (<xref ref-type="bibr" rid="bib19">Kummel et al., 2011</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib21">Li et al., 2014</xref>). Specifically, neuronal, GLUT4, endosomal, and yeast SNARE complexes in the half-zippered state had their R SNAREs unzipped to −1, +3, +1, and +3 amino acids relative to the ionic layer, respectively, where the positive sign designates the C-terminal amino acids. The standard deviation of all positions was less than three amino acids (<xref ref-type="table" rid="tbl1">Table 1</xref>).</p><p>To further confirm the derived structures of the intermediate states, we truncated the LD or the CTD of the v-SNARE Snc2 in the yeast SNARE complex and repeated the pulling experiment. We found that LD truncation eliminated the LD but not the CTD transition, while the CTD truncation abolished both transitions (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1</xref>). These results support the inferred structures for the intermediate states. Finally, both CTD and LD folded more rapidly than similar coiled-coil proteins (<xref ref-type="bibr" rid="bib53">Xi et al., 2012</xref>), with their transition rates greater than 50 s<sup>−1</sup>, even at the equilibrium forces (<xref ref-type="fig" rid="fig5">Figure 5</xref>, <xref ref-type="fig" rid="fig6">Figure 6</xref>).</p><p>Further unzipping of the half-zippered states of all four SNARE complexes became irreversible and they remained unfolded for over 50 s under the slow relaxation conditions in our experiment (<xref ref-type="fig" rid="fig2">Figure 2A</xref>), indicating a large energy barrier for SNARE NTD association. Close inspection of the FECs showed that a fraction of half-zippered SNARE complexes, that is, 10%, 50%, and 30% for neuronal, GLUT4, and endosomal SNAREs, respectively, unfolded via a transient intermediate with a typical lifetime of less than 50 ms (<xref ref-type="fig" rid="fig2">Figure 2A,B</xref>, <xref ref-type="fig" rid="fig2s2 fig2s3">Figure 2—figure supplements 2 and 3</xref>). The yeast SNARE complex is special, because this additional intermediate appeared in 85% of the unfolding transitions of the half-zippered complex and generally lasted for more than 5 s (<xref ref-type="fig" rid="fig2s4">Figure 2—figure supplement 4</xref>). For all SNARE complexes, these intermediate states are located at an extension approximately halfway between the half-zippered states and the fully unfolded states, indicating their similar structures. Based on their relative extension positions, the intermediate states are estimated to be t-SNARE or Q SNARE complexes with ordered NTDs but disordered CTDs (<xref ref-type="fig" rid="fig2">Figure 2C</xref>).</p><p>In conclusion, all four SNARE complexes assemble or disassemble via three common intermediate states: the t-SNARE state, the half-zippered state, and the LD-unfolded state, and along similar folding pathways and kinetics, particularly slow NTD association and fast CTD and LD zippering.</p></sec><sec id="s2-3"><title>Energetics and kinetics of CTD zippering</title><p>The biggest difference between the four SNARE complexes lay in their CTD equilibrium forces (<xref ref-type="fig" rid="fig3">Figure 3A</xref>, <xref ref-type="fig" rid="fig5">Figure 5C</xref>), indicating different CTD folding energies. To quantify CTD folding energy and kinetics, we measured CTD transitions at different constant forces in their corresponding force ranges (<xref ref-type="fig" rid="fig5">Figure 5A</xref>). We analyzed each extension-time series using a two-state hidden Markov model (HMM) and determined the positions of the folded and unfolded CTD states and their corresponding fluctuations, the unfolding probability, and transition rates (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib53">Xi et al., 2012</xref>). The HMM-based analyses yielded idealized state transitions and extension histogram distributions that closely matched the corresponding experimental measurements (<xref ref-type="fig" rid="fig5">Figure 5A,B</xref>). The unfolding probability rises with the force increase in a sigmoidal manner (<xref ref-type="fig" rid="fig5">Figure 5C</xref>). The folding rate or unfolding rate decreases or increases approximately exponentially upon a force increase in the narrow force range tested (<xref ref-type="bibr" rid="bib5">Bustamante et al., 2004</xref>; <xref ref-type="fig" rid="fig5">Figure 5D</xref>). Both observations suggest a two-state CTD transition and the existence of a single major energy barrier corresponding to the transition state for the folding/unfolding process. The position of the transition state relative to the folded or unfolded state can be determined from the force-dependent transition rates.</p><p>We adopted a simplified energy landscape model to derive the energy and rate of SNARE folding at zero force (‘Materials and methods’) (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib53">Xi et al., 2012</xref>). Non-linear least-squares fitting of the model matched the experimental data well (<xref ref-type="fig" rid="fig5">Figure 5C,D</xref>), which revealed the free energy of the folded state and the transition state and their relative positions (<xref ref-type="table" rid="tbl1">Table 1</xref>). The CTD folding energy of neuronal, GLUT4, endosomal, and yeast SNARE complexes were −27 (±5; SD throughout the text) <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>, −23 (±4) <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>, −16 (±2) <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>, and −13 (±3) <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>, respectively. The CTD folding energy and the equilibrium rate (∼100 s<sup>−1</sup>) of the neuronal SNARE complex were very close to the energy (28 ± 3 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>) and the rate (∼160 s<sup>−1</sup>) reported earlier (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>), indicating that the spacer sequences in the chimeric construct used here have minimal effect on the folding energy and kinetics of the SNARE complex.</p><p>The binary CTD transition manifested the existence of an energy barrier and its associated transition state for CTD folding and unfolding in the presence of the external force. When extrapolated to zero force, the CTD folding energy barrier became minimal for endosomal and yeast SNARE complexes or disappears for neuronal and GLUT4 SNARE complexes (<xref ref-type="table" rid="tbl1">Table 1</xref>). In both scenarios, free energy of the transition states can still be defined (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib53">Xi et al., 2012</xref>). The transition states of four SNARE complexes are located between the third and sixth hydrophobic layers. The energy and position of the transition state is important for characterizing the energy landscape of SNARE folding described later in the text. The relatively small folding energy barrier suggests that the rate of SNARE-mediated fusion is not limited by the intrinsic rate of CTD folding (at zero force), and that the stability of the half-zippered state is strongly force-dependent. Any partially zippered trans-SNARE complexes involved in vesicle docking and priming are likely in strained states imposed by the membranes and regulatory proteins (<xref ref-type="bibr" rid="bib13">Guzman et al., 2010</xref>; <xref ref-type="bibr" rid="bib15">Jahn and Fasshauer, 2012</xref>).</p></sec><sec id="s2-4"><title>Energetics and kinetics of LD zippering</title><p>The folding and unfolding transitions of the LD in four SNARE complexes were similar. They were reversible, binary, and fast (<xref ref-type="fig" rid="fig6">Figure 6</xref>, <xref ref-type="fig" rid="fig6s1">Figure 6—figure supplement 1</xref>). Furthermore, the LD transitions occurred in narrow force ranges (6–8.6 pN), in contrast to the CTD transition (10.1–18.5 pN) (<xref ref-type="table" rid="tbl1">Table 1</xref>). Although the LD of the endosomal SNARE complex forms a four-helix bundle (<xref ref-type="bibr" rid="bib58">Zwilling et al., 2007</xref>; <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref>), rather than a two-stranded coiled coil as in the other three SNARE complexes, it has similar LD transition kinetics, associated extension change, and lower equilibrium force than its CTD. This observation corroborates the conclusion that LD is a domain distinct from CTD, even in the endosomal SNARE complex.</p><p>The energy and kinetics of LD zippering at zero force was obtained in a way similar to CTD, as previously described (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). For neuronal SNARE complexes, the new chimeric construct led to an equilibrium force of 8 (±1) pN for LD transition, compared to 12 (±2) pN previously measured for the same transition. Correcting for the minor effect of the spacer sequence added between syntaxin and SNAP-25 (<xref ref-type="fig" rid="fig6s2">Figure 6—figure supplement 2</xref>), we obtained LD zippering energy of −10 (±2) <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> for the neuronal SNARE complex, consistent with our previous measurement of −8 (±2) <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>. Similarly, we derived the zippering energy of LDs and their associated energy barriers for the other three SNARE complexes (<xref ref-type="table" rid="tbl1">Table 1</xref>). In the four SNARE complexes, LD zippering outputs less energy than CTD zippering. Thus, CTD zippering serves as the major power stoke for membrane fusion (<xref ref-type="bibr" rid="bib50">Walter et al., 2010</xref>).</p></sec><sec id="s2-5"><title>Different roles of SNARE zippering stages in membrane fusion</title><p>Our above analysis revealed a simplified folding energy landscape of each SNARE complex (<xref ref-type="fig" rid="fig7">Figure 7</xref>). To illustrate how such an energy landscape is adapted to stage-wise membrane fusion (<xref ref-type="fig" rid="fig7">Figure 7A</xref>), we calculated the energy landscape of SNARE assembly in the presence of membranes using a neuronal SNARE complex as an example (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). The interaction energy between membranes containing lipid-anchored t- and v-SNAREs has been measured by the surface forces apparatus (SFA) (<xref ref-type="bibr" rid="bib22">Li et al., 2007</xref>). The interaction as a function of membrane separation contains two exponentially decaying components with decay constants of 2.5 nm (<italic>d</italic><sub>1</sub>) and 6 nm (<italic>d</italic><sub>2</sub>). The short-ranged component represents membrane repulsion just before fusion, including membrane dehydration (<xref ref-type="bibr" rid="bib20">Leckband and Israelachvili, 2001</xref>), and the long-range component results from the steric repulsion between unfolded or partially unfolded t- and v-SNAREs before their association.<fig-group><fig id="fig7" position="float"><object-id pub-id-type="doi">10.7554/eLife.03348.020</object-id><label>Figure 7.</label><caption><title>Energy landscape of SNARE zippering perfectly meets the needs of membrane fusion.</title><p>(<bold>A</bold>) Cartoons of different assembly states of the trans-SNARE complex corresponding to the points indicated in <bold>B</bold>. (<bold>B</bold>) Free energy of membrane fusion per SNARE complex (red line), a single loaded trans-SNARE complex (blue), or a single unloaded SNARE complex (black) as a function of the distance between two membrane surfaces. The experimental and alternative energy landscapes are plotted in solid and dashed lines, respectively. For the unloaded SNARE complex, the free energy at each membrane distance represents the energy of the SNARE complex in the same zippering state as the trans-SNARE complex at that distance. The experimental energy landscape of the unloaded SNARE complex was derived from the measured energy at characteristic points (marked by circles) through interpolation. (<bold>C</bold>) Repulsive force between two membranes opposing their fusion. (<bold>D</bold>) Extension of all the unfolded amino acids in Q<sub>a</sub> and R SNAREs under membrane tension (gray in the left axis) or the folded portion of the SNARE complex (black) and zippering stage of the amino acids (A.A.) in R SNARE (red line in the right red axis). The amino acid number indicates the position of the amino acid relative to the ionic layer (0). The amino acids with negative numbers are in N-terminal domain (NTD) and those with positive numbers in C-terminal domain (CTD). At each amino acid number in the right axis, the R SNARE motif has assembled from its N-terminus (A.A. at −24) to the amino acid with this number. Note that NTD association is accompanied by a relatively small change in membrane distance compared to CTD zippering, because the extension decrease due to NTD folding is largely canceled by the extension increase of the folded t-SNARE.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.020">http://dx.doi.org/10.7554/eLife.03348.020</ext-link></p></caption><graphic xlink:href="elife03348f007"/></fig><fig id="fig7s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.03348.021</object-id><label>Figure 7—figure supplement 1.</label><caption><title>Estimation of the average forces generated by zippering of the N-terminal and C-terminal CTD of neuronal SNARE complex.</title><p>We estimated the C-terminal domain (CTD) zippering forces based on a simplified energy landscape model for a two-state process. The energy landscape is characterized by the experimentally measured energy of the unfolded state and the transition state and their associated positions in terms of extension. The folded state is chosen as a reference 0 here for energy and extension. We chose the average extension change upon CTD unzipping (7.2 nm, <xref ref-type="table" rid="tbl1">Table 1</xref>) as the length unit. Thus, the x-axis is the normalized extension and the energy shown in y-axis has a unit of 7.2 pN × nm. Based on our measurement, two-thirds of the folding energy is released upon zippering of the C-terminal one-third of the CTD, which leads to the energy landscape shown by the solid black line. The folded state, the transition state, and the unfolded state are located at positions of 0, 1/3, and 1 in the unit of total extension change of CTD transition. In this unit, the energy of the unfolded state is equal to the equilibrium force measured for CTD transition (∼16 pN). The average forces of the C-terminal one-third and the N-terminal two-thirds of CTD can be calculated based on the slopes of the corresponding energy changes, yielding 32 pN and 8 pN, respectively. The average force does not depend on the detailed energy landscape between the transition state and the folded or the unfolded state. Assuming an energy profile of <inline-formula><mml:math id="inf11"><mml:mrow><mml:mi>E</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:math></inline-formula> between extension position 0 and a, the average force in this region is defined as<disp-formula id="equ7"><mml:math id="m7"><mml:mtable columnalign="left"><mml:mtr><mml:mtd><mml:mrow><mml:mo>〈</mml:mo><mml:mi>f</mml:mi><mml:mo>〉</mml:mo></mml:mrow></mml:mtd><mml:mtd><mml:mo>≡</mml:mo><mml:mfrac><mml:mn>1</mml:mn><mml:mi>a</mml:mi></mml:mfrac><mml:mstyle displaystyle="true"><mml:mrow><mml:msubsup><mml:mo>∫</mml:mo><mml:mn>0</mml:mn><mml:mi>a</mml:mi></mml:msubsup><mml:mrow><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mfrac><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>E</mml:mi></mml:mrow><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>x</mml:mi></mml:mrow></mml:mfrac></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:mrow></mml:mstyle><mml:mtext>d</mml:mtext><mml:mi>x</mml:mi></mml:mtd></mml:mtr><mml:mtr><mml:mtd/><mml:mtd><mml:mo>=</mml:mo><mml:mfrac><mml:mrow><mml:msub><mml:mi>E</mml:mi><mml:mi>a</mml:mi></mml:msub></mml:mrow><mml:mi>a</mml:mi></mml:mfrac><mml:mo>,</mml:mo></mml:mtd></mml:mtr></mml:mtable></mml:math></disp-formula>where <inline-formula><mml:math id="inf8"><mml:mrow><mml:mi>f</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>≡</mml:mo><mml:mfrac><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>E</mml:mi></mml:mrow><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>x</mml:mi></mml:mrow></mml:mfrac></mml:mrow></mml:math></inline-formula> is the local force at position <italic>x</italic> and <inline-formula><mml:math id="inf9"><mml:mrow><mml:mi>E</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mn>0</mml:mn><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mn>0</mml:mn></mml:mrow></mml:math></inline-formula>, <inline-formula><mml:math id="inf10"><mml:mrow><mml:mi>E</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>a</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:msub><mml:mi>E</mml:mi><mml:mi>a</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula>. Thus the average force in a region can be simply determined by the slope of the line going through the energy points at the two ends of the region, regardless of the detailed energy profile in the region.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.03348.021">http://dx.doi.org/10.7554/eLife.03348.021</ext-link></p></caption><graphic xlink:href="elife03348fs011"/></fig></fig-group></p><p>We chose the membrane interaction energy (<italic>V</italic>) per SNARE complex versus the distance between two membrane surfaces at the sites of SNARE attachment (<italic>d</italic>) as<disp-formula id="equ1"><label>(1)</label><mml:math id="m1"><mml:mrow><mml:mi>V</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>d</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mfrac><mml:mrow><mml:msub><mml:mi>E</mml:mi><mml:mi>m</mml:mi></mml:msub></mml:mrow><mml:mrow><mml:mn>1</mml:mn><mml:mo>+</mml:mo><mml:mi>α</mml:mi></mml:mrow></mml:mfrac><mml:mrow><mml:mo>[</mml:mo><mml:mrow><mml:mi>exp</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mo>−</mml:mo><mml:mfrac><mml:mrow><mml:mi>d</mml:mi><mml:mo>−</mml:mo><mml:msub><mml:mi>d</mml:mi><mml:mi>c</mml:mi></mml:msub></mml:mrow><mml:mrow><mml:msub><mml:mi>d</mml:mi><mml:mn>1</mml:mn></mml:msub></mml:mrow></mml:mfrac></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>+</mml:mo><mml:mi>α</mml:mi><mml:mtext> </mml:mtext><mml:mi>exp</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mo>−</mml:mo><mml:mfrac><mml:mrow><mml:mi>d</mml:mi><mml:mo>−</mml:mo><mml:msub><mml:mi>d</mml:mi><mml:mi>c</mml:mi></mml:msub></mml:mrow><mml:mrow><mml:msub><mml:mi>d</mml:mi><mml:mn>2</mml:mn></mml:msub></mml:mrow></mml:mfrac></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow><mml:mo>]</mml:mo></mml:mrow><mml:mo>,</mml:mo><mml:mtext> </mml:mtext><mml:mi>d</mml:mi><mml:mo>≥</mml:mo><mml:msub><mml:mi>d</mml:mi><mml:mi>c</mml:mi></mml:msub><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>where <italic>E</italic><sub><italic>m</italic></sub> determines the energy barrier for membrane fusion per SNARE complex when membranes are brought to the minimal distance allowed by the molecular dimension of the fully folded SNARE four-helix bundle (<italic>d</italic><sub><italic>c</italic></sub> = 1 nm) (<xref ref-type="bibr" rid="bib48">Sutton et al., 1998</xref>; <xref ref-type="bibr" rid="bib44">Stein et al., 2009</xref>; <xref ref-type="fig" rid="fig7">Figure 7B</xref>). Below this critical distance, membrane fusion occurs irreversibly. The amplitude ratio of the two exponential components (<italic>α</italic>) was set to 0.5 based on the SFA measurement (<xref ref-type="bibr" rid="bib22">Li et al., 2007</xref>). The energy barrier for membrane fusion (<italic>N</italic> × <italic>E</italic><sub><italic>m</italic></sub>) and the exact number of SNARE complexes required for fusion (<italic>N</italic>) are under much discussion (<xref ref-type="bibr" rid="bib17">Karatekin et al., 2010</xref>; <xref ref-type="bibr" rid="bib33">Mohrmann et al., 2010</xref>; <xref ref-type="bibr" rid="bib49">van den Bogaart et al., 2010</xref>). To bypass these uncertainties, we chose <italic>E</italic><sub><italic>m</italic></sub> ≈ 50 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> per SNARE complex, consistent with our measured folding energy per neuronal SNARE complex and other estimations (<xref ref-type="bibr" rid="bib22">Li et al., 2007</xref>; <xref ref-type="bibr" rid="bib49">van den Bogaart et al., 2010</xref>).</p><p>The role of the LD in membrane fusion is not clear. Evidence suggests that LDs bind membranes and constitute parts of membrane anchors with SNARE transmembrane domains (<xref ref-type="bibr" rid="bib22">Li et al., 2007</xref>; <xref ref-type="bibr" rid="bib9">Ellena et al., 2009</xref>; <xref ref-type="bibr" rid="bib3">Borisovska et al., 2012</xref>). To simplify our calculations, we did not explicitly consider extension and energy contributions from LDs but assumed instead that membrane fusion occurs when CTD is fully zippered.</p><p>We computed the energy landscape of the loaded SNARE complex as the sum of the energy of the unloaded SNAREs, the entropic energy of the stretched and unfolded SNARE polypeptides calculated based on <xref ref-type="disp-formula" rid="equ3">Equation 3</xref> in ‘Materials and methods’, and the membrane interaction energy. At each SNARE zippering stage, an equilibrium membrane distance was calculated by equating the SNARE pulling force to the membrane repulsive force (<xref ref-type="fig" rid="fig7">Figure 7C,D</xref>). The calculated SNARE energy was plotted in <xref ref-type="fig" rid="fig7">Figure 7B</xref> as a function of the membrane distance. SNARE NTD association was initiated at the very N-termini of t- and v-SNAREs at a large distance of 12.5 nm (<xref ref-type="fig" rid="fig7">Figure 7B–D</xref>). The association was accompanied by coil-to-helix propagation of the partially disordered t-SNARE towards its C-terminus (<xref ref-type="bibr" rid="bib21">Li et al., 2014</xref>; <xref ref-type="fig" rid="fig7">Figure 7A</xref>, state ii). Further NTD zippering led to the half-zippered trans-SNARE complex at 9.7 nm (state iii) (<xref ref-type="bibr" rid="bib2">Bharat et al., 2014</xref>). Thus, NTD association occurs in a narrow distance range of 9.7–12.5 nm, where the membrane repulsive force is small (2–5 pN). Formation of the half-zippered state in the presence of membranes leads to a net energy release of ∼26 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> (the energy difference between state i and state iii) (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>), which can be used to dock or prime vesicles and prevent dissociation of the half-zippered trans-SNARE complex.</p><p>In contrast to NTD association, CTD zippering from the half-zippered state was directly and tightly coupled to membrane fusion (<xref ref-type="fig" rid="fig7">Figure 7B–D</xref>). The membrane-loaded half-zippered SNARE complex had an energy of ∼34 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> relative to its folded state, which consists of ∼27 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> CTD folding energy and ∼7 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> entropic energy stored in the stretched VAMP2 CTD. CTD zippering drew two membranes from 9.7 nm to 1 nm for fusion against a large average force. As a result, CTD zippering of the trans-SNARE complex reduced the energy barrier of membrane fusion (<xref ref-type="fig" rid="fig7">Figure 7A</xref>, state iv) to 7.4 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> per SNARE complex (the energy difference between state iv and state iii), consistent with a fusion rate of ∼600 s<sup>−1</sup>, where we assumed a maximum fusion rate of 10<sup>6</sup> s<sup>−1</sup> in the absence of any energy barrier (<xref ref-type="bibr" rid="bib55">Yang and Gruebele, 2003</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). Correspondingly, the metastable half-zippered SNARE complex in the absence of a force load was stabilized by the short-ranged membrane opposing force (<xref ref-type="fig" rid="fig7">Figure 7B</xref>) together with regulatory proteins, such as complexin (<xref ref-type="bibr" rid="bib47">Sudhof and Rothman, 2009</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib31">Min et al., 2013</xref>). In contrast, the optical trapping force used in our experiments was long-range, with a typical force constant of 0.1 pN/nm, compared to an average force constant of 7.1 pN/nm for the membrane force opposing CTD assembly. As a result, the half-zippered SNARE state was typically short-lived (&lt;0.2 ms) upon reassembly of the SNARE complex at the low forces favoring NTD association (<xref ref-type="fig" rid="fig4">Figure 4</xref>), because the trapping force opposing CTD zippering remained small immediately after NTD was zippered. Thus, the membrane's repulsive force was an integral component of SNARE assembly and regulation.</p><p>The exponential increase in the membrane repulsive force below 5 nm (from 17 pN to 60 pN, <xref ref-type="fig" rid="fig7">Figure 7C</xref>) required an increasing force output as CTD zippers toward its C-terminus. SNARE zippering indeed met this requirement by producing a high force in this region. Here, the magnitude of local force generated by SNARE zippering was equivalent to the slope of the energy landscape of the unloaded SNARE complex with respect to extension, which also represents the energy density (defined as the folding energy per unit length of R SNARE polypeptide chain zippered) distributed along the SNARE bundle. While zippering of the first two-thirds of CTD generated an average force of 8 pN, zippering of the last one-third of CTD produces an average force of up to 32 pN (<xref ref-type="fig" rid="fig7s1">Figure 7—figure supplement 1</xref>), well suited to counteracting the short-ranged membrane opposing force. The position where CTD changed its energy density (<xref ref-type="fig" rid="fig7">Figure 7B</xref>, at 5 nm) was intriguing, because it was close to the energy barrier of the trans-SNARE complex. This position overlap is no accident, because any force applied to the SNARE complex tilts the CTD zippering energy landscape of the unloaded SNARE complex toward the unfolded state (<xref ref-type="bibr" rid="bib5">Bustamante et al., 2004</xref>), which tends to make the density-changing point an energy barrier (for example, see Figure S9 in <xref ref-type="bibr" rid="bib53">Xi et al., 2012</xref>). It is this energy barrier that results in the binary CTD transition and the polarized CTD energy distribution.</p><p>To further corroborate the essential role of the polarized CTD energy distribution in membrane fusion, we calculated the energy landscape of the trans-SNARE complex based on an alternative energy landscape of SNARE zippering (<xref ref-type="fig" rid="fig7">Figure 7B</xref>, black dashed line). In this alternative landscape, the energy of CTD is enriched at its N-terminus, rather than its C-terminus, but with the same total CTD zippering energy. The half-zippered trans-SNARE complex is now greatly stabilized by membranes (at 8 nm of the blue dashed line), but nearly unable to fuse them, because the energy barrier for fusion increases to 24 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> (the energy difference between states at membrane distances of 1 nm and 8 nm) compared to 7.4 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> for the wild-type SNARE complex. In this case, significant energy from CTD zippering is not transmitted to membranes, but dissipated as heat. In addition, no additional energy barrier appears before fusion. Similarly, SNAREs alone with this alternative folding energy landscape are expected to zipper or unzip in a continuous manner in response to the force exerted by optical tweezers, in contrast to the observed cooperative two-state manner.</p><p>Various parameters for the membrane interaction energy have been reported for different model membranes (<xref ref-type="bibr" rid="bib20">Leckband and Israelachvili, 2001</xref>). To test how variation in membrane properties may change the requirement for the polarized energy distribution, we repeated our above calculations by changing the parameters in <xref ref-type="disp-formula" rid="equ1">Equation 1</xref> with <italic>L</italic><sub><italic>1</italic></sub> and <italic>α</italic> ranging from 1 nm to 3 nm and from 0 to 0.5, respectively. Although energy landscapes of the loaded SNARE complex quantitatively change with these parameters, the energy landscape with a C-terminal polarized energy distribution always led to a much lower energy barrier for fusion than the alternative energy landscapes with an N-terminal polarized energy distribution. Thus, a C-terminal polarized energy distribution is a general requirement for efficient membrane fusion.</p><p>Taken together, our observed binary CTD transition indicates that the polarized CTD energy distribution is essential for efficient membrane fusion. In contrast, continuous and progressive SNARE assembly leads to poor coupling to membrane fusion.</p></sec></sec><sec id="s3" sec-type="discussion"><title>Discussion</title><sec id="s3-1"><title>Role of the half-zippered SNARE state in membrane fusion</title><p>The identification of half-zippered intermediates in all four representative SNARE complexes suggests that SNARE complexes follow a common zippering mechanism to drive membrane fusion (<xref ref-type="bibr" rid="bib14">Hanson et al., 1997</xref>). This observation is not consistent with alternative mechanisms by which the entire SNARE four-helix bundle assembles in an all-or-none manner (<xref ref-type="bibr" rid="bib15">Jahn and Fasshauer, 2012</xref>; <xref ref-type="bibr" rid="bib18">Kasai et al., 2012</xref>) or in a continuous layer-by-layer manner. Instead, our data reveal two distinct cooperative assemblies of the NTD and CTD in the SNARE complex, which clarifies the detailed zippering kinetics.</p><p>The presence of a partially zippered SNARE complex has been supported by many experiments (<xref ref-type="bibr" rid="bib54">Xu et al., 1999</xref>; <xref ref-type="bibr" rid="bib37a">Schwartz and Merz, 2009</xref>; <xref ref-type="bibr" rid="bib50">Walter et al., 2010</xref>; <xref ref-type="bibr" rid="bib8">Diao et al., 2012</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib31">Min et al., 2013</xref>). Our work further demonstrates that the partially zippered complex is a half-zippered complex intrinsic to a SNARE complex and functionally important for membrane fusion. Mutations and truncations that alter the structure of the half-zippered neuronal SNARE complex and/or its folding energy and kinetics abolish membrane fusion (Ma L, Gao Y, Yang G, and Zhang YL, manuscript in preparation). Thus, such a half-zippered SNARE complex is required for fast and regulated synaptic vesicle fusion (<xref ref-type="bibr" rid="bib19">Kummel et al., 2011</xref>; <xref ref-type="bibr" rid="bib23">Li et al., 2011</xref>, <xref ref-type="bibr" rid="bib21">2014</xref>).</p><p>Our finding suggests that the half-zippered structure is more ancient in SNARE evolution than any regulators that target this structure, and may have more conserved function in membrane fusion than previously thought. We propose that the step-wise assembly enables reversible folding of SNARE complexes, as shown in our calculations. The step-wise and reversible assembly enhances not only the coupling between SNARE zippering and membrane fusion, but also the specificity of SNARE pairing. Furthermore, the half-zippered SNARE complexes may be the target of their cognate Sec1p/Munc18 (SM)-family proteins essential for SNARE-mediated membrane fusion (<xref ref-type="bibr" rid="bib38">Shen et al., 2007</xref>; <xref ref-type="bibr" rid="bib47">Sudhof and Rothman, 2009</xref>; <xref ref-type="bibr" rid="bib16">Jorgacevski et al., 2011</xref>; <xref ref-type="bibr" rid="bib27">Ma et al., 2013</xref>).</p></sec><sec id="s3-2"><title>Role of CTD zippering energy in the rate and mechanism of membrane fusion</title><p>For the neuronal SNARE complex, we and others suggested that assembly of the NTD and the CTD has distinct functions: while NTD assembly is responsible for vesicle docking and priming, CTD zippering directly drives membrane fusion (<xref ref-type="bibr" rid="bib50">Walter et al., 2010</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). Fusion of GLUT4-storage vesicles (GSVs) with the plasma membrane mediated by the GLUT4 SNARE complex appears to be very similar to fusion of synaptic vesicles, including distinct stages of vesicle docking, priming, fusion (<xref ref-type="bibr" rid="bib45">Stockli et al., 2011</xref>), and close CTD zippering energy. However, the docked GSVs take about 1 min to fuse after insulin triggering (<xref ref-type="bibr" rid="bib1">Bai et al., 2007</xref>). In this case, the observed fusion rate is probably limited by the slow NTD association, but not the fast CTD zippering. Thus, insulin may mainly regulate steps upstream of NTD association.</p><p>The CTD zippering energy puts a strong constraint on the detailed mechanism of membrane fusion, including on the number of SNARE complexes required for fusion (<italic>N</italic>). If the total CTD zippering energy of <italic>N</italic> trans-SNARE complexes is used to lower the energy barrier of membrane fusion (<italic>E</italic><sub><italic>b</italic></sub>) (<xref ref-type="bibr" rid="bib34">Montecucco et al., 2005</xref>; <xref ref-type="bibr" rid="bib33">Mohrmann et al., 2010</xref>), the fusion rate (<italic>k</italic>) should be an exponential function of the total zippering energy, that is, <italic>k = k</italic><sub><italic>0</italic></sub> <italic>×</italic> exp(<italic>N × E</italic><sub><italic>CTD</italic></sub> − <italic>E</italic><sub><italic>b</italic></sub>), where <italic>E</italic><sub><italic>CTD</italic></sub> is the CTD zippering energy per SNARE complex and <italic>k</italic><sub><italic>0</italic></sub> a pre-constant. The large difference of CTD zippering energy between either endosomal or yeast SNARE complex and neuronal SNARE complex (12 or 14 <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic>) suggests that more endosomal or yeast SNARE complexes than neuronal SNARE complexes may be required to mediate fusion (<xref ref-type="bibr" rid="bib33">Mohrmann et al., 2010</xref>; <xref ref-type="bibr" rid="bib52">Wickner, 2010</xref>; <xref ref-type="bibr" rid="bib39">Shi et al., 2012</xref>).</p><p>SNARE-mediated membrane fusion in vivo involves fixed numbers of SNARE complexes characteristic of different fusion processes, likely controlled by regulatory proteins (<xref ref-type="bibr" rid="bib34">Montecucco et al., 2005</xref>). This result is in contrast with reconstituted SNARE-mediated membrane fusion in vitro, in which the SNARE number is probably not controlled (<xref ref-type="bibr" rid="bib17">Karatekin et al., 2010</xref>; <xref ref-type="bibr" rid="bib39">Shi et al., 2012</xref>; <xref ref-type="bibr" rid="bib49">van den Bogaart et al., 2010</xref>). As a result, liposome–liposome fusion mediated by the four different SNARE complexes alone exhibit similar fusion rates (<xref ref-type="bibr" rid="bib38">Shen et al., 2007</xref>; <xref ref-type="bibr" rid="bib58">Zwilling et al., 2007</xref>; <xref ref-type="bibr" rid="bib56">Yu et al., 2013</xref>). Taken together, the difference in CTD zippering energy contributes to the large variation in the fusion rate mediated by SNARE complexes and indicates a different number of SNARE complexes required for different fusion processes in vivo.</p></sec><sec id="s3-3"><title>Significance of the polarized energy distribution in membrane fusion</title><p>A common feature of our derived energy landscapes for all tested SNARE complexes is their polarized energy distribution, with much higher energy density at the C-terminus of CTD. This distribution allows SNAREs to increase their force output as CTD zippering draws two membranes into close proximity. Thus, as specialized engines for membrane fusion, SNAREs contain a built-in automatic transmission system that adjusts their force output to accommodate the large force change required for membrane fusion. This system ensures efficient and tight coupling between SNARE zippering and membrane fusion. A mismatch between the force output from SNAREs and the load from membranes would inevitably lead to dissipation of the SNARE zippering energy into heat, reducing the efficiency or the rate of membrane fusion. Thus, membrane fusion requires SNAREs to ‘save the best for last’.</p><p>How does the SNARE complex focus its zippering energy to the C-terminus? <xref ref-type="bibr" rid="bib21">Li et al. (2014)</xref> have recently shown that the association of the N-terminal half of VAMP2 to the N-terminal t-SNARE triggers folding of the t-SNARE C-terminal domain (<xref ref-type="fig" rid="fig7">Figure 7A</xref>, state ii). The ordered t-SNARE then serves as a template for fast and energetic VAMP2 zippering (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). Furthermore, tight association between v- and t-SNAREs near the C-terminus of CTD is achieved by key amino acids in that region, including the highly conserved phenylalanine residue shared by the v-SNAREs in all four SNARE complexes (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref>). Substitution of the phenylalanine residue with alanine abolishes the binary CTD transition in vitro (Ma L, Gao Y, Yang G, and Zhang YL, manuscript in preparation) and exocytosis (<xref ref-type="bibr" rid="bib50">Walter et al., 2010</xref>). These observations strongly suggest that the polarized energy distribution of SNARE complexes is essential for membrane fusion.</p><p>In summary, as the molecular machine for membrane fusion, SNARE proteins share a working mechanism conserved from yeast to humans. They couple their step-wise folding/assembly to membrane fusion through a distinct half-zippered state. SNAREs contain a built-in transmission system that produces the highest forces at the very C-termini required for efficient membrane fusion. This unified mechanism provides a basis for dissecting the diverse functions of SNAREs in more detail.</p></sec></sec><sec id="s4" sec-type="materials|methods"><title>Materials and methods</title><sec id="s4-1"><title>SNARE sequences, purification, and biotinylation</title><p>Amino acid sequences of the SNARE constructs used in our study are listed in <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref>. The corresponding genes were codon-optimized, synthesized, subcloned into the protein expression pET-SUMO vector, and expressed in BL21(DE3) <italic>E</italic>scherichia <italic>coli</italic> cells as previously described (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). The proteins were purified using Ni-NTA resin (GE Healthcare Biosciences, Pittsburgh, PA) and biotinylated using biotin ligase (Avidity, Aurora, CO).</p></sec><sec id="s4-2"><title>High-resolution dual-trap optical tweezers</title><p>The tweezers were home-built and located in an acoustically isolated room with controlled temperature and air flow as previously described (<xref ref-type="bibr" rid="bib32">Moffitt et al., 2006</xref>; <xref ref-type="bibr" rid="bib40">Sirinakis et al., 2011</xref>). The machine was operated remotely through a computer interface written in LabVIEW (National Instruments, Austin, TX). The force and displacement measured by optical tweezers were calibrated by Brownian motion of polystyrene beads in optical traps before each single-molecule experiment. The beads were trapped in aqueous buffer in a microfluidic channel 0.2 mm in thickness, which was formed by sandwiching two coverslips with parafilm (<xref ref-type="bibr" rid="bib57">Zhang et al., 2012</xref>).</p></sec><sec id="s4-3"><title>Single-molecule protein folding experiment</title><p>The cysteine-containing SNARE complex was reduced by TCEP or DTT, treated with dithiodipyridine (DTDP), mixed with the thiol-containing DNA handle in a typical 20:1 protein:DNA molar ratio, and cross-linked to the DNA handle overnight. An aliquot of the protein-DNA conjugate was mixed with anti-digoxigenin-coated beads and injected into the microfluidic channel. One DNA-bound bead was caught by one optical trap, brought close to a streptavidin-coated bead held in another optical trap, and formed a single SNARE-DNA tether. The SNARE complex was then pulled at a uniform trap separation speed or held at an approximately constant force or trap separation. The single-molecule folding experiment was performed at room temperature (22°C) in phosphate-buffered saline. An oxygen scavenging system was added to prevent photo-damage of the SNARE-DNA tether (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>).</p></sec><sec id="s4-4"><title>Data analysis</title><p>Methods of data analysis are described in detail elsewhere (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib53">Xi et al., 2012</xref>) and are summarized here. The observed extension and energy changes contain contributions from the structured and unstructured parts of the SNARE protein as well as the DNA handle. The extension of the structured SNAREs was derived from the crystal structure of the SNARE complex and was assumed to be force-independent. The extensions of the unstructured polypeptide and the DNA handle were determined by the worm-like chain model (<xref ref-type="bibr" rid="bib29">Marko and Siggia, 1995</xref>; <xref ref-type="bibr" rid="bib42">Smith et al., 1996</xref>). Specifically, the extension (<italic>x</italic>) of a worm-like chain is related to the stretching force (<italic>F</italic>) and the contour length (<italic>l</italic>) by the Marko-Siggia formula<disp-formula id="equ2"><label>(2)</label><mml:math id="m2"><mml:mrow><mml:mi>F</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>r</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mfrac><mml:mrow><mml:msub><mml:mi>k</mml:mi><mml:mi>B</mml:mi></mml:msub><mml:mi>T</mml:mi></mml:mrow><mml:mi>P</mml:mi></mml:mfrac><mml:mrow><mml:mo>[</mml:mo><mml:mrow><mml:mfrac><mml:mn>1</mml:mn><mml:mrow><mml:mn>4</mml:mn><mml:msup><mml:mrow><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mn>1</mml:mn><mml:mo>−</mml:mo><mml:mi>r</mml:mi></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow><mml:mn>2</mml:mn></mml:msup></mml:mrow></mml:mfrac><mml:mo>+</mml:mo><mml:mi>r</mml:mi><mml:mo>−</mml:mo><mml:mfrac><mml:mn>1</mml:mn><mml:mn>4</mml:mn></mml:mfrac></mml:mrow><mml:mo>]</mml:mo></mml:mrow><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>where <inline-formula><mml:math id="inf1"><mml:mrow><mml:mi>r</mml:mi><mml:mo>=</mml:mo><mml:mrow><mml:mi>x</mml:mi><mml:mo>/</mml:mo><mml:mi>l</mml:mi></mml:mrow></mml:mrow></mml:math></inline-formula>, <italic>P</italic> is the persistence length of the polypeptide (0.6 nm) or DNA (30–50 nm), and <italic>k</italic><sub><italic>B</italic></sub><italic>T</italic> = 4.1 pN × nm the product of the Boltzmann constant and the room temperature. Extending a worm-like chain decreases its entropy. The associated energy increase can be obtained by integrating the force in <xref ref-type="disp-formula" rid="equ2">Equation 2</xref> with respect to the extension, yielding<disp-formula id="equ3"><label>(3)</label><mml:math id="m3"><mml:mrow><mml:mi>E</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mi>l</mml:mi><mml:mo>,</mml:mo><mml:mi>r</mml:mi></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mfrac><mml:mrow><mml:msub><mml:mi>k</mml:mi><mml:mi>B</mml:mi></mml:msub><mml:mi>T</mml:mi></mml:mrow><mml:mi>P</mml:mi></mml:mfrac><mml:mfrac><mml:mi>l</mml:mi><mml:mrow><mml:mn>4</mml:mn><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mn>1</mml:mn><mml:mo>−</mml:mo><mml:mi>r</mml:mi></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:mfrac><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mn>3</mml:mn><mml:msup><mml:mi>r</mml:mi><mml:mn>2</mml:mn></mml:msup><mml:mo>−</mml:mo><mml:mn>2</mml:mn><mml:msup><mml:mi>r</mml:mi><mml:mn>3</mml:mn></mml:msup></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>.</mml:mo></mml:mrow></mml:math></disp-formula></p><p>For the two-state transitions of LD and CTD, we determined the unfolding probability, transition rates, and average state extensions and forces at different trap separations based on the measured extension and force trajectories using a two-state hidden Markov model. Then, we constructed a force-dependent energy landscape model that relates these experimental measurements to model parameters, including free energy of the folded state and transition state and their associated positions. Finally, we fit this model to the experimental data and determined the folding energy, folding energy barrier, and their associated structures.</p><p>The unfolding energy (Δ<italic>G</italic>) of a protein can be measured based on the mechanical work to reversibly unfold the protein, that is,<disp-formula id="equ4"><label>(4)</label><mml:math id="m4"><mml:mrow><mml:mi>Δ</mml:mi><mml:mi>G</mml:mi><mml:mo>=</mml:mo><mml:msub><mml:mi>f</mml:mi><mml:mrow><mml:mn>1</mml:mn><mml:mo>/</mml:mo><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo>×</mml:mo><mml:mi>Δ</mml:mi><mml:mi>X</mml:mi><mml:mo>−</mml:mo><mml:mi>E</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mi>Δ</mml:mi><mml:mi>l</mml:mi><mml:mo>,</mml:mo><mml:mrow><mml:mrow><mml:mi>Δ</mml:mi><mml:mi>x</mml:mi></mml:mrow><mml:mo>/</mml:mo><mml:mrow><mml:mi>Δ</mml:mi><mml:mi>l</mml:mi></mml:mrow></mml:mrow></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>where <inline-formula><mml:math id="inf2"><mml:mrow><mml:msub><mml:mi>f</mml:mi><mml:mrow><mml:mn>1</mml:mn><mml:mo>/</mml:mo><mml:mn>2</mml:mn></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> is the measured equilibrium force, Δ<italic>X</italic> the corresponding extension change, and <italic>E</italic> the entropic energy of the unfolded polypeptide. Δ<italic>x</italic> and Δ<italic>l</italic> are the extension change and the contour length change, respectively, of the unfolded polypeptide associated with protein unfolding. However, neither Δ<italic>x</italic> nor Δ<italic>l</italic> in <xref ref-type="disp-formula" rid="equ4">Equation 4</xref> is directly measurable and both are related to Δ<italic>X</italic> in a model-dependent manner (see Equations 12 and 13 in <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). In addition, our experiments were not performed under exactly constant force, but constant trap separation for maximum spatiotemporal resolution (<xref ref-type="bibr" rid="bib41">Sirinakis et al., 2012</xref>). As a result, a SNARE domain in a two-state transition experiences slightly different average forces in the folded state (<inline-formula><mml:math id="inf3"><mml:mrow><mml:msub><mml:mi>f</mml:mi><mml:mn>1</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula>) and the unfolded state (<inline-formula><mml:math id="inf4"><mml:mrow><mml:msub><mml:mi>f</mml:mi><mml:mn>2</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula>) (<xref ref-type="fig" rid="fig2">Figure 2B</xref>). Nevertheless, the average of the two state forces <inline-formula><mml:math id="inf5"><mml:mrow><mml:mi>f</mml:mi><mml:mo>=</mml:mo><mml:mrow><mml:mrow><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:msub><mml:mi>f</mml:mi><mml:mn>1</mml:mn></mml:msub><mml:mo>+</mml:mo><mml:msub><mml:mi>f</mml:mi><mml:mn>2</mml:mn></mml:msub></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow><mml:mo>/</mml:mo><mml:mn>2</mml:mn></mml:mrow></mml:mrow></mml:math></inline-formula> remains constant, which we have simply referred to as force (<xref ref-type="bibr" rid="bib11">Gao et al., 2011</xref>; <xref ref-type="fig" rid="fig5 fig6">Figures 5 and 6</xref>). Therefore, we constructed a detailed energy landscape model to quantitatively account for the correlation between protein structural transitions and the observed extension changes.</p><p>We chose the contour length of the unfolded polypeptide directly pulled by optical traps, which is 0.365 nm per amino acid, as a reaction coordinate to describe the extension change and the energy landscape associated with SNARE folding and unfolding. Different structural models were used for LD and CTD transitions: in LD transition, the two helices in Q<sub>a</sub> and R SNAREs fold and unfold symmetrically, whereas in CTD transition, R SNARE folds and unfolds along the pre-structured t-SNARE template (<xref ref-type="bibr" rid="bib19">Kummel et al., 2011</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib21">Li et al., 2014</xref>). Using these structural models, we could fit the calculated extension to the measured FEC to determine the contour length parameter associated with each state. The total energy of the single-molecule system additionally includes the harmonic potential energy of two beads in optical traps. The folding energy of the structured part of the SNARE complex as a function of the contour length gives the folding energy landscape of SNARE folding. In our data analysis, this energy landscape was characterized by the free energy of the folded state and the transition state and their associated positions in the reaction coordinate, all relative to the unfolded state. Both energy and positions were chosen as model parameters first to calculate the total system energy and the extension of the SNARE-DNA tether. Then these calculations were used to further compute the opening probability based on the Boltzmann distribution and the folding and unfolding rates based on Kramer's theory, as well as the extension change, for the transition of each SNARE domain. These values from model predictions were fit against the corresponding experimental data by the non-linear least-squares method, which yielded the best-fit model parameters, including the energy of the folded state and the transition state.</p></sec><sec id="s4-5"><title>Energy landscapes of trans-SNAREs</title><p>In our NTD structural model, we incorporated a detailed mechanism for t-SNARE folding induced by NTD association (<xref ref-type="bibr" rid="bib21">Li et al., 2014</xref>). Because the kinetics of this coupled binding and folding process is unclear, we assumed that t-SNARE is gradually structured as VAMP2 starts to zipper from its N-terminus, and becomes fully structured when two-thirds of VAMP2 NTD has been zippered (<xref ref-type="fig" rid="fig7">Figure 7D</xref>). This forms a structure corresponding to the transition state of NTD association. Further zippering stabilizes NTD and forms the half-zippered state. Combined with the structural model for CTD transition, a complete model for assembly of the SNARE four-helix bundle was defined. This model also established the structure and the extension of the folded SNARE complex <italic>h</italic> as a function of the contour length <italic>l</italic>. The membrane distance <italic>d</italic> was determined by equating the SNARE pulling force to the membrane repulsive force, that is,<disp-formula id="equ5"><label>(5)</label><mml:math id="m5"><mml:mrow><mml:mi>F</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mo>−</mml:mo><mml:msup><mml:mi>V</mml:mi><mml:mo>′</mml:mo></mml:msup><mml:mrow><mml:mo>(</mml:mo><mml:mi>d</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>where <inline-formula><mml:math id="inf6"><mml:mrow><mml:mi>x</mml:mi><mml:mo>=</mml:mo><mml:mi>d</mml:mi><mml:mo>−</mml:mo><mml:mi>h</mml:mi><mml:mo>−</mml:mo><mml:msup><mml:mi>l</mml:mi><mml:mrow><mml:mrow><mml:mn>3</mml:mn><mml:mo>/</mml:mo><mml:mn>5</mml:mn></mml:mrow></mml:mrow></mml:msup><mml:msup><mml:mi>p</mml:mi><mml:mrow><mml:mrow><mml:mn>2</mml:mn><mml:mo>/</mml:mo><mml:mn>5</mml:mn></mml:mrow></mml:mrow></mml:msup><mml:mo>/</mml:mo><mml:mn>2</mml:mn></mml:mrow></mml:math></inline-formula> is the effective extension of the unfolded polypeptide with contour length <italic>l</italic> and <italic>V′</italic> the derivative of the membrane interaction energy (<xref ref-type="fig" rid="fig7">Figure 7D</xref>). The last term in the effective extension expression (<inline-formula><mml:math id="inf7"><mml:mrow><mml:msup><mml:mi>l</mml:mi><mml:mrow><mml:mrow><mml:mn>3</mml:mn><mml:mo>/</mml:mo><mml:mn>5</mml:mn></mml:mrow></mml:mrow></mml:msup><mml:msup><mml:mi>p</mml:mi><mml:mrow><mml:mrow><mml:mn>2</mml:mn><mml:mo>/</mml:mo><mml:mn>5</mml:mn></mml:mrow></mml:mrow></mml:msup><mml:mo>/</mml:mo><mml:mn>2</mml:mn></mml:mrow></mml:math></inline-formula>) corrects for the residual extension of the unfolded polypeptide in the absence of external force when one end of the polypeptide is attached to the membrane, which is estimated to be half of the Flory radius of a semi-flexible chain (<xref ref-type="bibr" rid="bib22">Li et al., 2007</xref>). Solving this non-linear equation at different contour lengths, we obtained the membrane distance <italic>d</italic> at any SNARE zippering stage. The energy landscapes of trans-SNAREs are the total energy of SNAREs (including the folding energy and the elastic energy of unfolded polypeptide) and membrane interaction energy as a function of the membrane distance (<xref ref-type="fig" rid="fig7">Figure 7B</xref>). The calculations were performed using Matlab codes that are available as source codes.</p></sec></sec></body><back><ack id="ack"><title>Acknowledgements</title><p>Aleksander A Rebane, the second author, performed exemplary work in accomplishing the complex data analyses herein reported. We thank Ying Gao for technical assistance and Jingshi Shen, Daniel Kummel, and Xinming Zhang for reading the manuscript. This work was supported by the NIH grants GM093341 to YZ and DK027044 to JER.</p></ack><sec sec-type="additional-information"><title>Additional information</title><fn-group content-type="competing-interest"><title>Competing interests</title><fn fn-type="conflict" id="conf1"><p>The authors declare that no competing interests exist.</p></fn></fn-group><fn-group content-type="author-contribution"><title>Author contributions</title><fn fn-type="con" id="con1"><p>SZ, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con2"><p>AAR, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con3"><p>LM, Acquisition of data, Drafting or revising the article, Contributed unpublished essential data or reagents</p></fn><fn fn-type="con" id="con4"><p>GY, Acquisition of data, Contributed unpublished essential data or reagents</p></fn><fn fn-type="con" id="con5"><p>JC, Acquisition of data, Contributed unpublished essential data or reagents</p></fn><fn fn-type="con" id="con6"><p>MAM, Analysis and interpretation of data, Contributed unpublished essential data or reagents</p></fn><fn fn-type="con" id="con7"><p>FP, Conception and design, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con8"><p>JER, Conception and design, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con9"><p>YZ, Conception and design, Analysis and interpretation of data, Drafting or revising the article</p></fn></fn-group></sec><ref-list><title>References</title><ref id="bib1"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bai</surname><given-names>L</given-names></name><name><surname>Wang</surname><given-names>Y</given-names></name><name><surname>Fan</surname><given-names>JM</given-names></name><name><surname>Chen</surname><given-names>Y</given-names></name><name><surname>Ji</surname><given-names>W</given-names></name><name><surname>Qu</surname><given-names>AL</given-names></name><name><surname>Xu</surname><given-names>PY</given-names></name><name><surname>James</surname><given-names>DE</given-names></name><name><surname>Xu</surname><given-names>T</given-names></name></person-group><year>2007</year><article-title>Dissecting multiple steps of GLUT4 trafficking and identifying the sites of insulin action</article-title><source>Cell Metabolism</source><volume>5</volume><fpage>47</fpage><lpage>57</lpage><pub-id pub-id-type="doi">10.1016/j.cmet.2006.11.013</pub-id></element-citation></ref><ref id="bib2"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bharat</surname><given-names>TA</given-names></name><name><surname>Malsam</surname><given-names>J</given-names></name><name><surname>Hagen</surname><given-names>WJ</given-names></name><name><surname>Scheutzow</surname><given-names>A</given-names></name><name><surname>Sollner</surname><given-names>TH</given-names></name><name><surname>Briggs</surname><given-names>JA</given-names></name></person-group><year>2014</year><article-title>SNARE and regulatory proteins induce local membrane protrusions to prime docked vesicles for fast calcium-triggered fusion</article-title><source>EMBO Reports</source><volume>15</volume><fpage>308</fpage><lpage>314</lpage><pub-id pub-id-type="doi">10.1002/embr.201337807</pub-id></element-citation></ref><ref id="bib3"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Borisovska</surname><given-names>M</given-names></name><name><surname>Schwarz</surname><given-names>YN</given-names></name><name><surname>Dhara</surname><given-names>M</given-names></name><name><surname>Yarzagaray</surname><given-names>A</given-names></name><name><surname>Hugo</surname><given-names>S</given-names></name><name><surname>Narzi</surname><given-names>D</given-names></name><name><surname>Siu</surname><given-names>SWI</given-names></name><name><surname>Kesavan</surname><given-names>J</given-names></name><name><surname>Mohrmann</surname><given-names>R</given-names></name><name><surname>Bockmann</surname><given-names>RA</given-names></name><name><surname>Bruns</surname><given-names>D</given-names></name></person-group><year>2012</year><article-title>Membrane-proximal tryptophans of synaptobrevin II stabilize priming of secretory vesicles</article-title><source>The Journal of Neuroscience</source><volume>32</volume><fpage>15983</fpage><lpage>15997</lpage><pub-id pub-id-type="doi">10.1523/JNEUROSCI.6282-11.2012</pub-id></element-citation></ref><ref id="bib4"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Burre</surname><given-names>J</given-names></name><name><surname>Sharma</surname><given-names>M</given-names></name><name><surname>Tsetsenis</surname><given-names>T</given-names></name><name><surname>Buchman</surname><given-names>V</given-names></name><name><surname>Etherton</surname><given-names>MR</given-names></name><name><surname>Sudhof</surname><given-names>TC</given-names></name></person-group><year>2010</year><article-title>α-synuclein promotes SNARE-complex assembly in vivo and in vitro</article-title><source>Science</source><volume>329</volume><fpage>1663</fpage><lpage>1667</lpage><pub-id pub-id-type="doi">10.1126/science.1195227</pub-id></element-citation></ref><ref id="bib5"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bustamante</surname><given-names>C</given-names></name><name><surname>Chemla</surname><given-names>YR</given-names></name><name><surname>Forde</surname><given-names>NR</given-names></name><name><surname>Izhaky</surname><given-names>D</given-names></name></person-group><year>2004</year><article-title>Mechanical processes in biochemistry</article-title><source>Annual Review of Biochemistry</source><volume>73</volume><fpage>705</fpage><lpage>748</lpage><pub-id pub-id-type="doi">10.1146/annurev.biochem.72.121801.161542</pub-id></element-citation></ref><ref id="bib6"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Cecconi</surname><given-names>C</given-names></name><name><surname>Shank</surname><given-names>EA</given-names></name><name><surname>Bustamante</surname><given-names>C</given-names></name><name><surname>Marqusee</surname><given-names>S</given-names></name></person-group><year>2005</year><article-title>Direct observation of the three-state folding of a single protein molecule</article-title><source>Science</source><volume>309</volume><fpage>2057</fpage><lpage>2060</lpage><pub-id pub-id-type="doi">10.1126/science.1116702</pub-id></element-citation></ref><ref id="bib7"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chen</surname><given-names>XC</given-names></name><name><surname>Tomchick</surname><given-names>DR</given-names></name><name><surname>Kovrigin</surname><given-names>E</given-names></name><name><surname>Arac</surname><given-names>D</given-names></name><name><surname>Machius</surname><given-names>M</given-names></name><name><surname>Sudhof</surname><given-names>TC</given-names></name><name><surname>Rizo</surname><given-names>J</given-names></name></person-group><year>2002</year><article-title>Three-dimensional structure of the complexin/SNARE complex</article-title><source>Neuron</source><volume>33</volume><fpage>397</fpage><lpage>409</lpage><pub-id pub-id-type="doi">10.1016/S0896-6273(02)00583-4</pub-id></element-citation></ref><ref id="bib8"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Diao</surname><given-names>J</given-names></name><name><surname>Grob</surname><given-names>P</given-names></name><name><surname>Cipriano</surname><given-names>DJ</given-names></name><name><surname>Kyoung</surname><given-names>M</given-names></name><name><surname>Zhang</surname><given-names>Y</given-names></name><name><surname>Shah</surname><given-names>S</given-names></name><name><surname>Nguyen</surname><given-names>A</given-names></name><name><surname>Padolina</surname><given-names>M</given-names></name><name><surname>Srivastava</surname><given-names>A</given-names></name><name><surname>Vrljic</surname><given-names>M</given-names></name><name><surname>Shah</surname><given-names>A</given-names></name><name><surname>Nogales</surname><given-names>E</given-names></name><name><surname>Chu</surname><given-names>S</given-names></name><name><surname>Brunger</surname><given-names>AT</given-names></name></person-group><year>2012</year><article-title>Synaptic proteins promote calcium-triggered fast transition from point contact to full fusion</article-title><source>eLife</source><volume>1</volume><fpage>e00109</fpage><pub-id pub-id-type="doi">10.7554/eLife.00109</pub-id></element-citation></ref><ref id="bib9"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ellena</surname><given-names>JF</given-names></name><name><surname>Liang</surname><given-names>BY</given-names></name><name><surname>Wiktor</surname><given-names>M</given-names></name><name><surname>Stein</surname><given-names>A</given-names></name><name><surname>Cafiso</surname><given-names>DS</given-names></name><name><surname>Jahn</surname><given-names>R</given-names></name><name><surname>Tamm</surname><given-names>LK</given-names></name></person-group><year>2009</year><article-title>Dynamic structure of lipid-bound synaptobrevin suggests a nucleation-propagation mechanism for trans-SNARE complex formation</article-title><source>Proceedings of the National Academy of Sciences of USA</source><volume>106</volume><fpage>20306</fpage><lpage>20311</lpage><pub-id pub-id-type="doi">10.1073/pnas.0908317106</pub-id></element-citation></ref><ref id="bib10"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Fasshauer</surname><given-names>D</given-names></name><name><surname>Sutton</surname><given-names>RB</given-names></name><name><surname>Brunger</surname><given-names>AT</given-names></name><name><surname>Jahn</surname><given-names>R</given-names></name></person-group><year>1998</year><article-title>Conserved structural features of the synaptic fusion complex: SNARE proteins reclassified as Q- and R-SNAREs</article-title><source>Proceedings of the National Academy of Sciences of USA</source><volume>95</volume><fpage>15781</fpage><lpage>15786</lpage><pub-id pub-id-type="doi">10.1073/pnas.95.26.15781</pub-id></element-citation></ref><ref id="bib11"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gao</surname><given-names>Y</given-names></name><name><surname>Sirinakis</surname><given-names>G</given-names></name><name><surname>Zhang</surname><given-names>YL</given-names></name></person-group><year>2011</year><article-title>Highly anisotropic stability and folding kinetics of a single coiled coil protein under mechanical tension</article-title><source>Journal of the American Chemical Society</source><volume>133</volume><fpage>12749</fpage><lpage>12757</lpage><pub-id pub-id-type="doi">10.1021/ja204005r</pub-id></element-citation></ref><ref id="bib12"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gao</surname><given-names>Y</given-names></name><name><surname>Zorman</surname><given-names>S</given-names></name><name><surname>Gundersen</surname><given-names>G</given-names></name><name><surname>Xi</surname><given-names>ZQ</given-names></name><name><surname>Ma</surname><given-names>L</given-names></name><name><surname>Sirinakis</surname><given-names>G</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name><name><surname>Zhang</surname><given-names>YL</given-names></name></person-group><year>2012</year><article-title>Single reconstituted neuronal SNARE complexes zipper in three distinct stages</article-title><source>Science</source><volume>337</volume><fpage>1340</fpage><lpage>1343</lpage><pub-id pub-id-type="doi">10.1126/science.1224492</pub-id></element-citation></ref><ref id="bib13"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Guzman</surname><given-names>RE</given-names></name><name><surname>Schwarz</surname><given-names>YN</given-names></name><name><surname>Rettig</surname><given-names>J</given-names></name><name><surname>Bruns</surname><given-names>D</given-names></name></person-group><year>2010</year><article-title>SNARE force synchronizes synaptic vesicle fusion and controls the kinetics of quantal synaptic transmission</article-title><source>The Journal of Neuroscience</source><volume>30</volume><fpage>10272</fpage><lpage>10281</lpage><pub-id pub-id-type="doi">10.1523/JNEUROSCI.1551-10.2010</pub-id></element-citation></ref><ref id="bib14"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hanson</surname><given-names>PI</given-names></name><name><surname>Roth</surname><given-names>R</given-names></name><name><surname>Morisaki</surname><given-names>H</given-names></name><name><surname>Jahn</surname><given-names>R</given-names></name><name><surname>Heuser</surname><given-names>JE</given-names></name></person-group><year>1997</year><article-title>Structure and conformational changes in NSF and its membrane receptor complexes visualized by quick-freeze/deep-etch electron microscopy</article-title><source>Cell</source><volume>90</volume><fpage>523</fpage><lpage>535</lpage><pub-id pub-id-type="doi">10.1016/S0092-8674(00)80512-7</pub-id></element-citation></ref><ref id="bib15"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Jahn</surname><given-names>R</given-names></name><name><surname>Fasshauer</surname><given-names>D</given-names></name></person-group><year>2012</year><article-title>Molecular machines governing exocytosis of synaptic vesicles</article-title><source>Nature</source><volume>490</volume><fpage>201</fpage><lpage>207</lpage><pub-id pub-id-type="doi">10.1038/nature11320</pub-id></element-citation></ref><ref id="bib16"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Jorgacevski</surname><given-names>J</given-names></name><name><surname>Potokar</surname><given-names>M</given-names></name><name><surname>Grilc</surname><given-names>S</given-names></name><name><surname>Kreft</surname><given-names>M</given-names></name><name><surname>Liu</surname><given-names>W</given-names></name><name><surname>Barclay</surname><given-names>JW</given-names></name><name><surname>Buckers</surname><given-names>J</given-names></name><name><surname>Medda</surname><given-names>R</given-names></name><name><surname>Hell</surname><given-names>SW</given-names></name><name><surname>Parpura</surname><given-names>V</given-names></name><name><surname>Burgoyne</surname><given-names>RD</given-names></name><name><surname>Zorec</surname><given-names>R</given-names></name></person-group><year>2011</year><article-title>Munc18-1 tuning of vesicle merger and fusion pore properties</article-title><source>The Journal of Neuroscience</source><volume>31</volume><fpage>9055</fpage><lpage>9066</lpage><pub-id pub-id-type="doi">10.1523/JNEUROSCI.0185-11.2011</pub-id></element-citation></ref><ref id="bib17"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Karatekin</surname><given-names>E</given-names></name><name><surname>Di Giovanni</surname><given-names>J</given-names></name><name><surname>Iborra</surname><given-names>C</given-names></name><name><surname>Coleman</surname><given-names>J</given-names></name><name><surname>O'Shaughnessy</surname><given-names>B</given-names></name><name><surname>Seagar</surname><given-names>M</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name></person-group><year>2010</year><article-title>A fast, single-vesicle fusion assay mimics physiological SNARE requirements</article-title><source>Proceedings of the National Academy of Sciences of USA</source><volume>107</volume><fpage>3517</fpage><lpage>3521</lpage><pub-id pub-id-type="doi">10.1073/pnas.0914723107</pub-id></element-citation></ref><ref id="bib18"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kasai</surname><given-names>H</given-names></name><name><surname>Takahashi</surname><given-names>N</given-names></name><name><surname>Tokumaru</surname><given-names>H</given-names></name></person-group><year>2012</year><article-title>Distinct initial SNARE configurations underlying the diversity of exocytosis</article-title><source>Physiological Reviews</source><volume>92</volume><fpage>1915</fpage><lpage>1964</lpage><pub-id pub-id-type="doi">10.1152/physrev.00007.2012</pub-id></element-citation></ref><ref id="bib19"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kummel</surname><given-names>D</given-names></name><name><surname>Krishnakumar</surname><given-names>SS</given-names></name><name><surname>Radoff</surname><given-names>DT</given-names></name><name><surname>Li</surname><given-names>F</given-names></name><name><surname>Giraudo</surname><given-names>CG</given-names></name><name><surname>Pincet</surname><given-names>F</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name><name><surname>Reinisch</surname><given-names>KM</given-names></name></person-group><year>2011</year><article-title>Complexin cross-links prefusion SNAREs into a zigzag array</article-title><source>Nature Structural &amp; Molecular Biology</source><volume>18</volume><fpage>927</fpage><lpage>933</lpage><pub-id pub-id-type="doi">10.1038/nsmb.2101</pub-id></element-citation></ref><ref id="bib20"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Leckband</surname><given-names>D</given-names></name><name><surname>Israelachvili</surname><given-names>J</given-names></name></person-group><year>2001</year><article-title>Intermolecular forces in biology</article-title><source>Quarterly Reviews of Biophysics</source><volume>34</volume><fpage>105</fpage><lpage>267</lpage><pub-id pub-id-type="doi">10.1017/S0033583501003687</pub-id></element-citation></ref><ref id="bib21"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Li</surname><given-names>F</given-names></name><name><surname>Kummel</surname><given-names>D</given-names></name><name><surname>Coleman</surname><given-names>J</given-names></name><name><surname>Reinisch</surname><given-names>KM</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name><name><surname>Pincet</surname><given-names>F</given-names></name></person-group><year>2014</year><article-title>A half-zippered SNARE complex represents a functional intermediate in membrane fusion</article-title><source>Journal of the American Chemical Society</source><volume>136</volume><fpage>3456</fpage><lpage>3464</lpage><pub-id pub-id-type="doi">10.1021/ja410690m</pub-id></element-citation></ref><ref id="bib22"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Li</surname><given-names>F</given-names></name><name><surname>Pincet</surname><given-names>F</given-names></name><name><surname>Perez</surname><given-names>E</given-names></name><name><surname>Eng</surname><given-names>WS</given-names></name><name><surname>Melia</surname><given-names>TJ</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name><name><surname>Tareste</surname><given-names>D</given-names></name></person-group><year>2007</year><article-title>Energetics and dynamics of SNAREpin folding across lipid bilayers</article-title><source>Nature Structural &amp; Molecular Biology</source><volume>14</volume><fpage>890</fpage><lpage>896</lpage><pub-id pub-id-type="doi">10.1038/nsmb1310</pub-id></element-citation></ref><ref id="bib23"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Li</surname><given-names>F</given-names></name><name><surname>Pincet</surname><given-names>F</given-names></name><name><surname>Perez</surname><given-names>E</given-names></name><name><surname>Giraudo</surname><given-names>CG</given-names></name><name><surname>Tareste</surname><given-names>D</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name></person-group><year>2011</year><article-title>Complexin activates and clamps SNAREpins by a common mechanism involving an intermediate energetic state</article-title><source>Nature Structural &amp; Molecular Biology</source><volume>18</volume><fpage>941</fpage><lpage>946</lpage><pub-id pub-id-type="doi">10.1038/nsmb.2102</pub-id></element-citation></ref><ref id="bib24"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Liphardt</surname><given-names>J</given-names></name><name><surname>Onoa</surname><given-names>B</given-names></name><name><surname>Smith</surname><given-names>SB</given-names></name><name><surname>Tinoco</surname><given-names>I</given-names><suffix>Jnr</suffix></name><name><surname>Bustamante</surname><given-names>C</given-names></name></person-group><year>2001</year><article-title>Reversible unfolding of single RNA molecules by mechanical force</article-title><source>Science</source><volume>292</volume><fpage>733</fpage><lpage>737</lpage><pub-id pub-id-type="doi">10.1126/science.1058498</pub-id></element-citation></ref><ref id="bib25"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Liu</surname><given-names>W</given-names></name><name><surname>Montana</surname><given-names>V</given-names></name><name><surname>Parpura</surname><given-names>V</given-names></name><name><surname>Mohideen</surname><given-names>U</given-names></name></person-group><year>2009</year><article-title>Single molecule measurements of interaction free energies between the proteins within binary and ternary SNARE complexes</article-title><source>Journal of Nanoneuroscience</source><volume>1</volume><fpage>120</fpage><lpage>129</lpage><pub-id pub-id-type="doi">10.1166/jns.2009.1001</pub-id></element-citation></ref><ref id="bib26"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lu</surname><given-names>HP</given-names></name><name><surname>Xun</surname><given-names>L</given-names></name><name><surname>Xie</surname><given-names>XS</given-names></name></person-group><year>1998</year><article-title>Single-molecule enzymatic dynamics</article-title><source>Science</source><volume>282</volume><fpage>1877</fpage><lpage>1882</lpage><pub-id pub-id-type="doi">10.1126/science.282.5395.1877</pub-id></element-citation></ref><ref id="bib27"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ma</surname><given-names>C</given-names></name><name><surname>Su</surname><given-names>L</given-names></name><name><surname>Seven</surname><given-names>AB</given-names></name><name><surname>Xu</surname><given-names>Y</given-names></name><name><surname>Rizo</surname><given-names>J</given-names></name></person-group><year>2013</year><article-title>Reconstitution of the vital functions of Munc18 and Munc13 in neurotransmitter release</article-title><source>Science</source><volume>339</volume><fpage>421</fpage><lpage>425</lpage><pub-id pub-id-type="doi">10.1126/science.1230473</pub-id></element-citation></ref><ref id="bib28"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Malsam</surname><given-names>J</given-names></name><name><surname>Parisotto</surname><given-names>D</given-names></name><name><surname>Bharat</surname><given-names>TAM</given-names></name><name><surname>Scheutzow</surname><given-names>A</given-names></name><name><surname>Krause</surname><given-names>JM</given-names></name><name><surname>Briggs</surname><given-names>JAG</given-names></name><name><surname>Sollner</surname><given-names>TH</given-names></name></person-group><year>2012</year><article-title>Complexin arrests a pool of docked vesicles for fast Ca2+-dependent release</article-title><source>The EMBO Journal</source><volume>31</volume><fpage>3270</fpage><lpage>3281</lpage><pub-id pub-id-type="doi">10.1038/emboj.2012.164</pub-id></element-citation></ref><ref id="bib29"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Marko</surname><given-names>JF</given-names></name><name><surname>Siggia</surname><given-names>ED</given-names></name></person-group><year>1995</year><article-title>Stretching DNA</article-title><source>Macromolecules</source><volume>28</volume><fpage>8759</fpage><lpage>8770</lpage><pub-id pub-id-type="doi">10.1021/ma00130a008</pub-id></element-citation></ref><ref id="bib30"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>McNew</surname><given-names>JA</given-names></name><name><surname>Weber</surname><given-names>T</given-names></name><name><surname>Parlati</surname><given-names>F</given-names></name><name><surname>Johnston</surname><given-names>RJ</given-names></name><name><surname>Melia</surname><given-names>TJ</given-names></name><name><surname>Sollner</surname><given-names>TH</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name></person-group><year>2000</year><article-title>Close is not enough: SNARE-dependent membrane fusion requires an active mechanism that transduces force to membrane anchors</article-title><source>The Journal of Cell Biology</source><volume>150</volume><fpage>105</fpage><lpage>117</lpage><pub-id pub-id-type="doi">10.1083/jcb.150.1.105</pub-id></element-citation></ref><ref id="bib31"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Min</surname><given-names>D</given-names></name><name><surname>Kim</surname><given-names>K</given-names></name><name><surname>Hyeon</surname><given-names>C</given-names></name><name><surname>Cho</surname><given-names>YH</given-names></name><name><surname>Shin</surname><given-names>YK</given-names></name><name><surname>Yoon</surname><given-names>TY</given-names></name></person-group><year>2013</year><article-title>Mechanical unzipping and rezipping of a single SNARE complex reveals hysteresis as a force-generating mechanism</article-title><source>Nature Communications</source><volume>4</volume><fpage>1705</fpage><pub-id pub-id-type="doi">10.1038/ncomms2692</pub-id></element-citation></ref><ref id="bib32"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Moffitt</surname><given-names>JR</given-names></name><name><surname>Chemla</surname><given-names>YR</given-names></name><name><surname>Izhaky</surname><given-names>D</given-names></name><name><surname>Bustamante</surname><given-names>C</given-names></name></person-group><year>2006</year><article-title>Differential detection of dual traps improves the spatial resolution of optical tweezers</article-title><source>Proceedings of the National Academy of Sciences of USA</source><volume>103</volume><fpage>9006</fpage><lpage>9011</lpage><pub-id pub-id-type="doi">10.1073/pnas.0603342103</pub-id></element-citation></ref><ref id="bib33"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Mohrmann</surname><given-names>R</given-names></name><name><surname>de Wit</surname><given-names>H</given-names></name><name><surname>Verhage</surname><given-names>M</given-names></name><name><surname>Neher</surname><given-names>E</given-names></name><name><surname>Sorensen</surname><given-names>JB</given-names></name></person-group><year>2010</year><article-title>Fast vesicle fusion in living cells requires at least three SNARE complexes</article-title><source>Science</source><volume>330</volume><fpage>502</fpage><lpage>505</lpage><pub-id pub-id-type="doi">10.1126/science.1193134</pub-id></element-citation></ref><ref id="bib34"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Montecucco</surname><given-names>C</given-names></name><name><surname>Schiavo</surname><given-names>G</given-names></name><name><surname>Pantano</surname><given-names>S</given-names></name></person-group><year>2005</year><article-title>SNARE complexes and neuroexocytosis: how many, how close?</article-title><source>Trends in Biochemical Sciences</source><volume>30</volume><fpage>367</fpage><lpage>372</lpage><pub-id pub-id-type="doi">10.1016/j.tibs.2005.05.002</pub-id></element-citation></ref><ref id="bib35"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Onuchic</surname><given-names>JN</given-names></name><name><surname>Wolynes</surname><given-names>PG</given-names></name></person-group><year>2004</year><article-title>Theory of protein folding</article-title><source>Current Opinion in Structural Biology</source><volume>14</volume><fpage>70</fpage><lpage>75</lpage><pub-id pub-id-type="doi">10.1016/j.sbi.2004.01.009</pub-id></element-citation></ref><ref id="bib36"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pobbati</surname><given-names>AV</given-names></name><name><surname>Stein</surname><given-names>A</given-names></name><name><surname>Fasshauer</surname><given-names>D</given-names></name></person-group><year>2006</year><article-title>N- to C-terminal SNARE complex assembly promotes rapid membrane fusion</article-title><source>Science</source><volume>313</volume><fpage>673</fpage><lpage>676</lpage><pub-id pub-id-type="doi">10.1126/science.1129486</pub-id></element-citation></ref><ref id="bib37"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sabatini</surname><given-names>BL</given-names></name><name><surname>Regehr</surname><given-names>WG</given-names></name></person-group><year>1996</year><article-title>Timing of neurotransmission at fast synapses in the mammalian brain</article-title><source>Nature</source><volume>384</volume><fpage>170</fpage><lpage>172</lpage><pub-id pub-id-type="doi">10.1038/384170a0</pub-id></element-citation></ref><ref id="bib37a"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Schwartz</surname><given-names>ML</given-names></name><name><surname>Merz</surname><given-names>AJ</given-names></name></person-group><year>2009</year><article-title>Capture and release of partially zipped trans-SNARE complexes on intact organelles</article-title><source>J. Cell Biol</source><volume>185</volume><fpage>535</fpage><lpage>549</lpage><pub-id pub-id-type="doi">10.1083/jcb.200811082</pub-id></element-citation></ref><ref id="bib38"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Shen</surname><given-names>J</given-names></name><name><surname>Tareste</surname><given-names>DC</given-names></name><name><surname>Paumet</surname><given-names>F</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name><name><surname>Melia</surname><given-names>TJ</given-names></name></person-group><year>2007</year><article-title>Selective activation of cognate SNAREpins by Sec1/Munc18 proteins</article-title><source>Cell</source><volume>128</volume><fpage>183</fpage><lpage>195</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2006.12.016</pub-id></element-citation></ref><ref id="bib39"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Shi</surname><given-names>L</given-names></name><name><surname>Shen</surname><given-names>QT</given-names></name><name><surname>Kiel</surname><given-names>A</given-names></name><name><surname>Wang</surname><given-names>J</given-names></name><name><surname>Wang</surname><given-names>HW</given-names></name><name><surname>Melia</surname><given-names>TJ</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name><name><surname>Pincet</surname><given-names>F</given-names></name></person-group><year>2012</year><article-title>SNARE proteins: one to fuse and three to keep the nascent fusion pore open</article-title><source>Science</source><volume>335</volume><fpage>1355</fpage><lpage>1359</lpage><pub-id pub-id-type="doi">10.1126/science.1214984</pub-id></element-citation></ref><ref id="bib40"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sirinakis</surname><given-names>G</given-names></name><name><surname>Clapier</surname><given-names>CR</given-names></name><name><surname>Gao</surname><given-names>Y</given-names></name><name><surname>Viswanathanc</surname><given-names>R</given-names></name><name><surname>Cairns</surname><given-names>BR</given-names></name><name><surname>Zhang</surname><given-names>YL</given-names></name></person-group><year>2011</year><article-title>The RSC chromatin remodeling ATPase translocates DNA with high force and small step size</article-title><source>The EMBO Journal</source><volume>30</volume><fpage>2364</fpage><lpage>2372</lpage><pub-id pub-id-type="doi">10.1038/emboj.2011.141</pub-id></element-citation></ref><ref id="bib41"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sirinakis</surname><given-names>G</given-names></name><name><surname>Ren</surname><given-names>Y</given-names></name><name><surname>Gao</surname><given-names>Y</given-names></name><name><surname>Xi</surname><given-names>Z</given-names></name><name><surname>Zhang</surname><given-names>Y</given-names></name></person-group><year>2012</year><article-title>Combined and versatile high-resolution optical tweezers and single-molecule fluorescence microscopy</article-title><source>The Review of Scientific Instruments</source><volume>83</volume><fpage>093708</fpage><pub-id pub-id-type="doi">10.1063/1.4752190</pub-id></element-citation></ref><ref id="bib42"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Smith</surname><given-names>CL</given-names></name><name><surname>Cui</surname><given-names>Y</given-names></name><name><surname>Bustamante</surname><given-names>C</given-names></name></person-group><year>1996</year><article-title>Overstretching B-DNA: the elastic response of individual double-stranded and single-stranded DNA molecules</article-title><source>Science</source><volume>271</volume><fpage>795</fpage><lpage>799</lpage><pub-id pub-id-type="doi">10.1126/science.271.5250.795</pub-id></element-citation></ref><ref id="bib43"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sollner</surname><given-names>T</given-names></name><name><surname>Whiteheart</surname><given-names>SW</given-names></name><name><surname>Brunner</surname><given-names>M</given-names></name><name><surname>Erdjument-Bromage</surname><given-names>H</given-names></name><name><surname>Geromanos</surname><given-names>S</given-names></name><name><surname>Tempst</surname><given-names>P</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name></person-group><year>1993</year><article-title>SNAP receptors implicated in vesicle targeting and fusion</article-title><source>Nature</source><volume>362</volume><fpage>318</fpage><lpage>324</lpage><pub-id pub-id-type="doi">10.1038/362318a0</pub-id></element-citation></ref><ref id="bib44"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Stein</surname><given-names>A</given-names></name><name><surname>Weber</surname><given-names>G</given-names></name><name><surname>Wahl</surname><given-names>MC</given-names></name><name><surname>Jahn</surname><given-names>R</given-names></name></person-group><year>2009</year><article-title>Helical extension of the neuronal SNARE complex into the membrane</article-title><source>Nature</source><volume>460</volume><fpage>525</fpage><lpage>528</lpage><pub-id pub-id-type="doi">10.1038/nature08156</pub-id></element-citation></ref><ref id="bib45"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Stockli</surname><given-names>J</given-names></name><name><surname>Fazakerley</surname><given-names>DJ</given-names></name><name><surname>James</surname><given-names>DE</given-names></name></person-group><year>2011</year><article-title>GLUT4 exocytosis</article-title><source>Journal of Cell Science</source><volume>124</volume><fpage>4147</fpage><lpage>4158</lpage><pub-id pub-id-type="doi">10.1242/jcs.097063</pub-id></element-citation></ref><ref id="bib46"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Strop</surname><given-names>P</given-names></name><name><surname>Kaiser</surname><given-names>SE</given-names></name><name><surname>Vrljic</surname><given-names>M</given-names></name><name><surname>Brunger</surname><given-names>AT</given-names></name></person-group><year>2008</year><article-title>The structure of the yeast plasma membrane SNARE complex reveals destabilizing water-filled cavities</article-title><source>The Journal of Biological Chemistry</source><volume>283</volume><fpage>1113</fpage><lpage>1119</lpage><pub-id pub-id-type="doi">10.1074/jbc.M707912200</pub-id></element-citation></ref><ref id="bib47"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sudhof</surname><given-names>TC</given-names></name><name><surname>Rothman</surname><given-names>JE</given-names></name></person-group><year>2009</year><article-title>Membrane fusion: Grappling with SNARE and SM proteins</article-title><source>Science</source><volume>323</volume><fpage>474</fpage><lpage>477</lpage><pub-id pub-id-type="doi">10.1126/science.1161748</pub-id></element-citation></ref><ref id="bib48"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sutton</surname><given-names>RB</given-names></name><name><surname>Fasshauer</surname><given-names>D</given-names></name><name><surname>Jahn</surname><given-names>R</given-names></name><name><surname>Brunger</surname><given-names>AT</given-names></name></person-group><year>1998</year><article-title>Crystal structure of a SNARE complex involved in synaptic exocytosis at 2.4 angstrom resolution</article-title><source>Nature</source><volume>395</volume><fpage>347</fpage><lpage>353</lpage><pub-id pub-id-type="doi">10.1038/26412</pub-id></element-citation></ref><ref id="bib49"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>van den Bogaart</surname><given-names>G</given-names></name><name><surname>Holt</surname><given-names>MG</given-names></name><name><surname>Bunt</surname><given-names>G</given-names></name><name><surname>Riedel</surname><given-names>D</given-names></name><name><surname>Wouters</surname><given-names>FS</given-names></name><name><surname>Jahn</surname><given-names>R</given-names></name></person-group><year>2010</year><article-title>One SNARE complex is sufficient for membrane fusion</article-title><source>Nature Structural &amp; Molecular Biology</source><volume>17</volume><fpage>358</fpage><lpage>364</lpage><pub-id pub-id-type="doi">10.1038/nsmb.1748</pub-id></element-citation></ref><ref id="bib50"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Walter</surname><given-names>AM</given-names></name><name><surname>Wiederhold</surname><given-names>K</given-names></name><name><surname>Bruns</surname><given-names>D</given-names></name><name><surname>Fasshauer</surname><given-names>D</given-names></name><name><surname>Sorensen</surname><given-names>JB</given-names></name></person-group><year>2010</year><article-title>Synaptobrevin N-terminally bound to syntaxin-SNAP-25 defines the primed vesicle state in regulated exocytosis</article-title><source>The Journal of Cell Biology</source><volume>188</volume><fpage>401</fpage><lpage>413</lpage><pub-id pub-id-type="doi">10.1083/jcb.200907018</pub-id></element-citation></ref><ref id="bib51"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Weninger</surname><given-names>K</given-names></name><name><surname>Bowen</surname><given-names>ME</given-names></name><name><surname>Chu</surname><given-names>S</given-names></name><name><surname>Brunger</surname><given-names>AT</given-names></name></person-group><year>2003</year><article-title>Single-molecule studies of SNARE complex assembly reveal parallel and antiparallel configurations</article-title><source>Proceedings of the National Academy of Sciences of USA</source><volume>100</volume><fpage>14800</fpage><lpage>14805</lpage><pub-id pub-id-type="doi">10.1073/pnas.2036428100</pub-id></element-citation></ref><ref id="bib52"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wickner</surname><given-names>W</given-names></name></person-group><year>2010</year><article-title>Membrane fusion: Five lipids, four SNAREs, three chaperones, two nucleotides, and a Rab, all dancing in a ring on yeast vacuoles</article-title><source>Annual Review of Cell and Developmental Biology</source><volume>26</volume><fpage>115</fpage><lpage>136</lpage><pub-id pub-id-type="doi">10.1146/annurev-cellbio-100109-104131</pub-id></element-citation></ref><ref id="bib53"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Xi</surname><given-names>ZQ</given-names></name><name><surname>Gao</surname><given-names>Y</given-names></name><name><surname>Sirinakis</surname><given-names>G</given-names></name><name><surname>Guo</surname><given-names>HL</given-names></name><name><surname>Zhang</surname><given-names>YL</given-names></name></person-group><year>2012</year><article-title>Direct observation of helix staggering, sliding, and coiled coil misfolding</article-title><source>Proceedings of the National Academy of Sciences of USA</source><volume>109</volume><fpage>5711</fpage><lpage>5716</lpage><pub-id pub-id-type="doi">10.1073/pnas.1116784109</pub-id></element-citation></ref><ref id="bib54"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Xu</surname><given-names>T</given-names></name><name><surname>Rammner</surname><given-names>B</given-names></name><name><surname>Margittai</surname><given-names>M</given-names></name><name><surname>Artalejo</surname><given-names>AR</given-names></name><name><surname>Neher</surname><given-names>E</given-names></name><name><surname>Jahn</surname><given-names>R</given-names></name></person-group><year>1999</year><article-title>Inhibition of SNARE complex assembly differentially affects kinetic components of exocytosis</article-title><source>Cell</source><volume>99</volume><fpage>713</fpage><lpage>722</lpage><pub-id pub-id-type="doi">10.1016/S0092-8674(00)81669-4</pub-id></element-citation></ref><ref id="bib55"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Yang</surname><given-names>WY</given-names></name><name><surname>Gruebele</surname><given-names>M</given-names></name></person-group><year>2003</year><article-title>Folding at the speed limit</article-title><source>Nature</source><volume>423</volume><fpage>193</fpage><lpage>197</lpage><pub-id pub-id-type="doi">10.1038/nature01609</pub-id></element-citation></ref><ref id="bib56"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Yu</surname><given-names>HJ</given-names></name><name><surname>Rathore</surname><given-names>SS</given-names></name><name><surname>Lopez</surname><given-names>JA</given-names></name><name><surname>Davis</surname><given-names>EM</given-names></name><name><surname>James</surname><given-names>DE</given-names></name><name><surname>Martin</surname><given-names>JL</given-names></name><name><surname>Shen</surname><given-names>JS</given-names></name></person-group><year>2013</year><article-title>Comparative studies of Munc18c and Munc18-1 reveal conserved and divergent mechanisms of Sec1/Munc18 proteins</article-title><source>Proceedings of the National Academy of Sciences of USA</source><volume>110</volume><fpage>E3271</fpage><lpage>E3280</lpage><pub-id pub-id-type="doi">10.1073/pnas.1311232110</pub-id></element-citation></ref><ref id="bib57"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhang</surname><given-names>Y</given-names></name><name><surname>Sirinakis</surname><given-names>G</given-names></name><name><surname>Gundersen</surname><given-names>G</given-names></name><name><surname>Xi</surname><given-names>Z</given-names></name><name><surname>Gao</surname><given-names>Y</given-names></name></person-group><year>2012</year><article-title>DNA translocation of ATP-dependent chromatin remodelling factors revealed by high-resolution optical tweezers</article-title><source>Methods in Enzymology</source><volume>513</volume><fpage>3</fpage><lpage>28</lpage><pub-id pub-id-type="doi">10.1016/B978-0-12-391938-0.00001-X</pub-id></element-citation></ref><ref id="bib58"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zwilling</surname><given-names>D</given-names></name><name><surname>Cypionka</surname><given-names>A</given-names></name><name><surname>Pohl</surname><given-names>WH</given-names></name><name><surname>Fasshauer</surname><given-names>D</given-names></name><name><surname>Walla</surname><given-names>PJ</given-names></name><name><surname>Wahl</surname><given-names>MC</given-names></name><name><surname>Jahn</surname><given-names>R</given-names></name></person-group><year>2007</year><article-title>Early endosomal SNAREs form a structurally conserved SNARE complex and fuse liposomes with multiple topologies</article-title><source>The EMBO Journal</source><volume>26</volume><fpage>9</fpage><lpage>18</lpage><pub-id pub-id-type="doi">10.1038/sj.emboj.7601467</pub-id></element-citation></ref></ref-list></back><sub-article article-type="article-commentary" id="SA1"><front-stub><article-id pub-id-type="doi">10.7554/eLife.03348.022</article-id><title-group><article-title>Decision letter</article-title></title-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Brunger</surname><given-names>Axel T</given-names></name><role>Reviewing editor</role><aff><institution>Stanford University</institution>, <country>United States</country></aff></contrib></contrib-group></front-stub><body><boxed-text><p>eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see <ext-link ext-link-type="uri" xlink:href="http://elifesciences.org/review-process">review process</ext-link>). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.</p></boxed-text><p>Thank you for sending your work entitled “SNARE assembly: common zippering intermediates and kinetics, different energetics” for consideration at <italic>eLife</italic>. Your article has been favorably evaluated by John Kuriyan (Senior editor), Axel Brunger (Reviewing editor), and three reviewers.</p><p>The Reviewing editor and the reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.</p><p>This manuscript reports single molecule optical trap force measurements for four different SNARE complexes, greatly extending previous work that was performed on the neuronal SNARE complex only. All four SNARE complexes show cooperative unfolding and refolding in the C-terminal half of the SNARE complex as well as an irreversible unfolding of the N-terminal half, but there are some differences in energetics. However, there are several major concerns:</p><p>1) The authors claim that the four SNARE complexes show common unfolding behavior, thus sharing a conserved zippering pathway. However, the unzipping behavior of the neuronal and endosomal SNARE complexes looks quite different from that of the GLUT4 and yeast SNARE complexes (shown in <xref ref-type="fig" rid="fig2">Figure 2A</xref> and <xref ref-type="fig" rid="fig2s1">Figure 2–figure supplement 1A</xref>). After the C-terminal half unzipping, the neuronal and endosomal SNARE complexes require a 5 to 10 pN higher force to unzip the N-terminal half (the irreversible transition), whereas in the GLUT4 and yeast SNARE complexes, the N-terminal half unzipping occurs almost simultaneously with the C-terminal half unzipping. This point is important because the authors stress that the C-terminal half of the SNARE motif works as a sort of independent domain in SNARE-complex assembly for all four SNARE complexes. To make this scenario possible, it is presumed that the N-terminal half is stably assembled, while the C-terminal half then shows zippering in a selective manner after a certain pause. The authors need to characterize the force range in which they see the N-terminal half unzipping for each type of the SNARE complex (the number of molecules studied should be specified), and compare this force distribution with that used for the CTD folding and unfolding (e.g., shown in <xref ref-type="fig" rid="fig4">Figure 4B</xref>). If these two force ranges (one for NTD unfolding and the other for CTD unfolding/refolding) are largely overlapping for a certain type of the SNARE complex, the SNARE-complex assembly may be essentially a one-step process in an all-or-none manner for that type of the SNARE complex.</p><p>2) All four SNARE complexes show a large hysteresis in their unfolding and refolding behavior. Thus, there is no guarantee that the N-terminal half unzipping and rezipping occurs at a same force level. Ideally, to confirm that the two-step model applies to the assembly process as well, re-assembly experiments should be performed. At the minimum, this limitation of the current experiments needs to be discussed.</p><p>3) The authors argue that the average force generated during one-third of CTD folding is 8 pN and zippering of the last one-third produces an average force up to 31 pN. Presumably, the authors reached these force values by taking derivatives of the energy landscape of the two-state model (in the zero-force case) that was also used to derive the folding energy value (if this is not the case, please provide the details of estimating these force values). An estimation of the folding energy value and position of the energy barrier seems reasonable, but the calculation of detailed force values is questionable using this simple two-state model. To derive the force value, one needs to know the local curvature of the energy landscape as demonstrated in Woodside et al. Science (2006) 314, 1001. Actually, it seems that the estimated force values do not well match with the experimental observation. For instance, the CTD of the neuronal SNARE complex shows active unfolding with 16 pN that is only half of the 31 pN force.</p><p>4) Statistical information is missing throughout the manuscript. For example, in <xref ref-type="fig" rid="fig3">Figure 3</xref> and <xref ref-type="fig" rid="fig4">Figure 4</xref>, how many independent SNARE complexes are studied for each data?</p><p>5) A concern is the effect of the spacers linking the different proteins into the chimeric constructs. The authors should provide evidence that the chimera likely zipper into the correct structure. More importantly, they should show that the zippering kinetics and/or pathway are unaffected. For example, the authors could investigate the effect of different spacer lengths or topologies on zippering kinetics.</p><p>6) The manuscript is very long and difficult read. There are many instances of incorrect English usage and/or confusing wording. While there is there is no formal word limit for this journal, it is not clear that the current length is warranted given the data presented. On the other hand, there are many instances where a more concise explanation would have improved the readability, and other instances where more details in the analysis would have been helpful. Such details should be provided in the Methods section. The authors are strongly encouraged to revise their text extensively to appeal to the broad readership of <italic>eLife</italic>.</p></body></sub-article><sub-article article-type="reply" id="SA2"><front-stub><article-id pub-id-type="doi">10.7554/eLife.03348.023</article-id><title-group><article-title>Author response</article-title></title-group></front-stub><body><p><italic>1) The authors claim that the four SNARE complexes show common unfolding behavior, thus sharing a conserved zippering pathway. However, the unzipping behavior of the neuronal and endosomal SNARE complexes looks quite different from that of the GLUT4 and yeast SNARE complexes (shown in</italic> <xref ref-type="fig" rid="fig2"><italic>Figure 2</italic>A</xref> <italic>and</italic> <xref ref-type="fig" rid="fig2s1"><italic>Figure 2–figure supplement 1A</italic></xref><italic>). After the C-terminal half unzipping, the neuronal and endosomal SNARE complexes require a 5 to 10 pN higher force to unzip the N-terminal half (the irreversible transition), whereas in the GLUT4 and yeast SNARE complexes, the N-terminal half unzipping occurs almost simultaneously with the C-terminal half unzipping. This point is important because the authors stress that the C-terminal half of the SNARE motif works as a sort of independent domain in SNARE-complex assembly for all four SNARE complexes. To make this scenario possible, it is presumed that the N-terminal half is stably assembled, while the C-terminal half then shows zippering in a selective manner after a certain pause. The authors need to characterize the force range in which they see the N-terminal half unzipping for each type of the SNARE complex (the number of molecules studied should be specified), and compare this force distribution with that used for the CTD folding and unfolding (e.g., shown in</italic> <xref ref-type="fig" rid="fig4"><italic>Figure 4</italic>B</xref><italic>). If these two force ranges (one for NTD unfolding and the other for CTD unfolding/refolding) are largely overlapping for a certain type of the SNARE complex, the SNARE-complex assembly may be essentially a one-step process in an all-or-none manner for that type of the SNARE complex</italic>.</p><p>We agree with the reviewers that distinct distributions of the CTD equilibrium force and the NTD unzipping force will help reinforce our conclusion that NTD and CTD assemble and disassemble in kinetically distinct steps. Following the reviewers’ suggestion, we have prepared the histogram distributions of the CTD equilibrium force and the NTD unzipping force for all four SNARE complexes and shown in a new <xref ref-type="fig" rid="fig2s2">Figure 2–figure supplement 2</xref>. Neuronal, yeast, and endosomal complexes exhibit mainly non-overlapping distributions, confirming distinct steps of CTD and NTD assembly.</p><p>GLUT4 SNARE complex has overlapping CTD and NTD force distributions. However, this overlap does not contradict the stage-wise assembly and disassembly of the GLUT4 SNARE complex. Strictly speaking, the CTD equilibrium force and the NTD unzipping force are different quantities and are not directly comparable. The equilibrium force is defined as the force under which the CTD has half unfolding probability (<xref ref-type="fig" rid="fig3">Figure 3C</xref>), whereas the NTD unzipping force is defined as the force that NTD first unzips. At a low pulling or force-loading speed, NTD may unzip before the pulling force reaches the CTD equilibrium force (see <xref ref-type="fig" rid="fig2s1">Figure 2–figure supplement 1</xref> for an example). However, in this case, CTD typically has already folded and unfolded for more than 30 times. If we compare the force and time difference at which CTD and NTD first unzip for each SNARE complex, the CTD always unzips earlier and at lower force than NTD for all four SNARE complexes (<xref ref-type="fig" rid="fig2s2">Figure 2–figure supplement 2</xref>). Therefore, the CTD and the NTD unzips in a sequential manner. This observation cannot be explained by an “all-or-none” assembly mechanism, because such a two-state process also implies a simultaneous and one-step unfolding of the whole SNARE bundle, which is not observed.</p><p>It must be pointed out that the differences between average CTD and NTD unzipping forces depend on the pulling speed, because the NTD unzipping process is not in equilibrium (<xref ref-type="bibr" rid="bib5">Bustamante et al., 2004</xref>). A lower pulling speed will decrease the force difference, mainly because the average NTD unzipping force decreases. Under a constant force, distinct CTD and NTD assembly or disassembly can still be revealed by their different lifetimes. The average lifetime of the folded CTD is generally less than 50 milliseconds under our experimental conditions (<xref ref-type="fig" rid="fig3">Figure 3D</xref>), whereas the average lifetime of the folded NTD is typically greater than 10 s in these experiments. Without a long NTD lifetime, it would be impossible for us to collect more than tens of thousands of CTD transition events at constant forces required to accurately determine the CTD opening probability and transition rates shown in <xref ref-type="fig" rid="fig3">Figure 3C–D</xref>. Thus, a non-overlapping force distribution is a sufficient, but not necessary condition for distinct NTD and CTD assembly and disassembly (consider the fact that all protein complexes dissociate even at zero force given sufficient time).</p><p>The lifetime difference between CTD and NTD becomes even greater for trans-SNARE complexes bridging two membranes. In this case, CTD zippering experiences a much greater opposing force than NTD zippering (<xref ref-type="fig" rid="fig5">Figure 5</xref>). As a result, the force-dependent half-zippered SNARE state is greatly stabilized by the apposed membranes.</p><p>In conclusion, our data are consistent with kinetically distinct NTD and CTD assembly and disassembly, but not simultaneous and one-step SNARE assembly. However, our data do not rule out cooperative interactions between NTD and CTD assembly. Instead, NTD assembly facilitates CTD assembly by contributing a coupling energy estimated to be 6 kT, as shown in our previous work (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>).</p><p>In the revised manuscript, we have therefore added two supplemental figures, <xref ref-type="fig" rid="fig2s1 fig2s2">Figure 2–figure supplement 1-2</xref>. These figures are referred to in the following sentences:</p><p>“The FECs show that all four SNARE complexes sequentially unfolded via two reversible transitions and one or two irreversible unfolding steps (<xref ref-type="fig" rid="fig2">Figure 2A</xref> and <xref ref-type="fig" rid="fig2s1 fig2s2">Figure 2–figure supplement 1–2</xref>).”</p><p><italic>2) All four SNARE complexes show a large hysteresis in their unfolding and refolding behavior. Thus, there is no guarantee that the N-terminal half unzipping and rezipping occurs at a same force level. Ideally, to confirm that the two-step model applies to the assembly process as well, re-assembly experiments should be performed. At the minimum, this limitation of the current experiments needs to be discussed</italic>.</p><p>We agree with the reviewers that it is technically difficult to reversibly unzip and re-zip the N-terminal half of the SNARE complex under constant forces. We once held neuronal SNARE complex at a constant force upon CTD unfolding for an overall 30 min and saw no re-folding, indicating a large energy barrier for NTD refolding as well as unfolding at high forces (&gt;12 pN). Consequently, we had to relax the complex to observe NTD refolding at a lower force (1-8 pN, see <xref ref-type="fig" rid="fig2s3">Figure 2–figure supplement 3</xref>, 5–7 and Figure in (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>)). This slow NTD association justifies the existence of the partially zippered neuronal SNARE complex, because de novo or all-or-none SNARE assembly will be too slow to match the speed (&lt;1 ms) required for the fast neurotransmission.</p><p>The SNARE reassembly processes observed in our experiments under low forces generally do not show any intermediates, because the lifetime of the half-zippered state is too short to be detected by our machine as is expected. As we reported before (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>) and modeled in <xref ref-type="fig" rid="fig5">Figure 5</xref>, the half-zippered state is force-dependent and stabilized by the short-ranged membrane opposing force. The state is stabilized, because its CTD zippering is opposed by large membrane repulsive force whereas its NTD association only meets small opposing force. Such a large force gradient cannot be mimicked by the linear force provided by optical tweezers to stabilize the half-zippered state as efficiently as the opposing membranes at the lower forces.</p><p>Finally, by crosslinking neuronal SNARE complex at a strategic position, we have recently been able to detect distinct reversible transitions in both NTD and CTD, confirming the stage-wise SNARE assembly and disassembly (Ma, L., Gao, Y., Yang, G., and Zhang, Y. L., manuscript in preparation).</p><p>To address the reviewers’ concern, we de-emphasized “assembly” by changing it to “assembly and disassembly”. In addition, following the reviewers’ suggestion and our above considerations, we added the following sentences:</p><p>“Correspondingly, the metastable half-zippered SNARE complex in the absence of a force load was stabilized by the short-ranged membrane opposing force (<xref ref-type="fig" rid="fig5">Figure 5B</xref>) together with regulatory proteins, such as complexin (<xref ref-type="bibr" rid="bib47">Sudhof and Rothman, 2009</xref>; <xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>; <xref ref-type="bibr" rid="bib31">Min et al., 2013</xref>). In contrast, the optical trapping force used in our experiments was long-range, with a typical force constant of 0.1 pN/nm, compared to an average force constant of 7.1 pN/nm for the membrane force opposing CTD assembly. As a result, the half-zippered SNARE state was typically short-lived (&lt;0.2 ms) upon reassembly of the SNARE complex at the low forces favoring NTD association (<xref ref-type="fig" rid="fig2s3">Figure 2–figure supplement 3</xref>), because the trapping force opposing CTD zippering remained small immediately after NTD was zippered. Thus, the membrane’s repulsive force was an integral component of SNARE assembly and regulation.”</p><p><italic>3) The authors argue that the average force generated during one-third of CTD folding is 8 pN and zippering of the last one-third produces an average force up to 31 pN. Presumably, the authors reached these force values by taking derivatives of the energy landscape of the two-state model (in the zero-force case) that was also used to derive the folding energy value (if this is not the case, please provide the details of estimating these force values). An estimation of the folding energy value and position of the energy barrier seems reasonable, but the calculation of detailed force values is questionable using this simple two-state model. To derive the force value, one needs to know the local curvature of the energy landscape as demonstrated in Woodside et al. Science (2006) 314, 1001. Actually, it seems that the estimated force values do not well match with the experimental observation. For instance, the CTD of the neuronal SNARE complex shows active unfolding with 16 pN that is only half of the 31 pN force</italic>.</p><p>Yes, we estimated the CTD zippering force based on a simplified energy landscape model for a two-state process. We have added Figure 5–figure supplement 2 to clarify and justify our estimation. Please see this figure as well as its legend for our replies to the reviewers’ comments.</p><p><italic>4) Statistical information is missing throughout the manuscript. For example, in</italic> <xref ref-type="fig" rid="fig3"><italic>Figure 3</italic></xref> <italic>and</italic> <xref ref-type="fig" rid="fig4"><italic>Figure 4</italic></xref><italic>, how many independent SNARE complexes are studied for each data</italic>?</p><p>To address this point, we added the required statistical information in the legend of <xref ref-type="table" rid="tbl1">Table 1</xref> and in <xref ref-type="fig" rid="fig2s2">Figure 2–figure supplement 2</xref>, including the number of transitions and single molecules scored. We have also added more statistical information when average values are given. Specifically, in the table legend, we added the following sentences in the main text:</p><p>“The equilibrium force distribution, the number of transitions and the number of single molecules scored for CTD transitions are shown in <xref ref-type="fig" rid="fig2s2">Figure 2–figure supplement 2</xref>. For parameters related to LD transitions, a total of 18, 35, 11, and 24 LD transitions in single neuronal, GLUT4, endosomal, and yeast SNARE complexes were scored, respectively.”</p><p>“Specifically, neuronal, GLUT4, endosomal, and yeast SNARE complexes in the half-zippered state had their R SNAREs unzipped to −1, +3, +1, and +3 amino acids relative to the ionic layer, respectively, whereby the positive sign designates the C-terminal amino acids. The standard deviation of all positions was less than 3 amino acids (<xref ref-type="table" rid="tbl1">Table 1</xref>).”</p><p>“Nonlinear least-squares fitting of the model matched the experimental data well (<xref ref-type="fig" rid="fig3">Figure 3C–D</xref>), which revealed the free energy of the folded state and the transition state and their relative positions (<xref ref-type="table" rid="tbl1">Table 1</xref>). The CTD folding energy of neuronal, GLUT4, endosomal, and yeast SNARE complexes were −27 (±5, s.d. throughout the text) k<sub>B</sub>T, −23 (±4) k<sub>B</sub>T, −16 (±2) k<sub>B</sub>T, and −13 (±3) k<sub>B</sub>T, respectively.”</p><p><italic>5) A concern is the effect of the spacers linking the different proteins into the chimeric constructs. The authors should provide evidence that the chimera likely zipper into the correct structure. More importantly, they should show that the zippering kinetics and/or pathway are unaffected. For example, the authors could investigate the effect of different spacer lengths or topologies on zippering kinetics</italic>.</p><p>Several lines of evidence, from both our experiments and theoretical calculations, show that the effect of the spacers on SNARE zippering energy and kinetics is minor and can be accounted for quantitatively. First, results from the CD measurement and gel filtration are consistent with a fully folded four-helix bundle (<xref ref-type="fig" rid="fig1s2 fig1s3">Figure 1–figure supplement 2–3</xref>); Second, the folding energy and kinetics of LD and CTD in neuronal SNARE complex match those of our earlier measurements based on a different SNARE construct without any long spacers used. In particular, the spacer sequence has minimal effort on CTD transition (<xref ref-type="fig" rid="fig3">Figure 3</xref>), because the extension of the spacers does not change in this process. Third, the spacer sequence introduced between syntaxin and SNAP-25 does not significantly affect the function of the t-SNARE to mediate membrane fusion in vitro (<xref ref-type="fig" rid="fig1s4">Figure 1–figure supplement 4</xref>). Fourth, results from the single-molecule manipulation experiments are generally homogenous among different molecules pulled and consistent with correctly folded four-helix bundle structures. If alternative folded conformations had occurred in our experiments, they could have been identified from their different assembly/disassembly kinetics (<xref ref-type="fig" rid="fig2s3">Figure 2–figure supplement 3</xref>). Fifth, for yeast SNARE complex, the derived LD and CTD transitions are corroborated by the experimental results from truncated SNARE complexes (<xref ref-type="fig" rid="fig2s4">Figure 2–figure supplement 4</xref>). Finally, the minor effect of the spacer sequences on SNARE assembly is confirmed by our theoretical calculations (Figure 4–figure supplement 2). The spacer sequence between t and v-SNAREs is expected to slightly destabilize NTD by ∼4.3 k<sub>B</sub>T due to stretching of the spacer, compared to ∼30 k<sub>B</sub>T folding energy of the NTD (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>). Because this destabilized NTD still has a much larger lifetime than CTD, our conclusion regarding distinct kinetics of NTD and CTD assembly is justified.</p><p>To address the reviewers’ concerns, in our revised manuscript, we have clarified the minor effect of the spacer sequences on SNARE transition energy and kinetics (Figure 4-figure supplement 2). In addition, we have emphasized this point in the main text as follows:</p><p>“Furthermore, the chimeric neuronal SNARE complex reveals folding energy and kinetics (see below) consistent with our recent reports based on a different SNARE construct in which syntaxin and VAMP2 were crosslinked at their N-termini by a disulfide bond (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>).”</p><p>“The CTD folding energy and the equilibrium rate (∼100 s<sup>−1</sup>) of the neuronal SNARE complex were very close to the energy (28 ± 3 k<sub>B</sub>T) and the rate (∼160 s<sup>−1</sup>) reported earlier (<xref ref-type="bibr" rid="bib12">Gao et al., 2012</xref>), indicating that the spacer sequences in the chimeric construct used here have minimal effect on the folding energy and kinetics of the SNARE complex.”</p><p>“Correcting for the minor effect of the spacer sequence added between syntaxin and SNAP-25 (Figure 4–figure supplement 2), we obtained LD zippering energy of −10 (±2) k<sub>B</sub>T for neuronal SNARE complex, consistent with our previous measurement of −8 (±2) k<sub>B</sub>T.”</p><p>Based on above extensive experimental evidence and theoretical calculations, we think our major conclusions are justified, and not affected by the chimeric SNARE constructs used in this work. Thus, we did not pursue studies to vary the length of the spacer sequence.</p><p><italic>6) The manuscript is very long and difficult read. There are many instances of incorrect English usage and/or confusing wording. While there is there is no formal word limit for this journal, it is not clear that the current length is warranted given the data presented. On the other hand, there are many instances where a more concise explanation would have improved the readability, and other instances where more details in the analysis would have been helpful. Such details should be provided in the Methods section. The authors are strongly encouraged to revise their text extensively to appeal to the broad readership of eLife</italic>.</p><p>Following the reviewers’ suggestion, we have shortened the main text and made revisions to add the necessary detail to the “Data Analysis” section of the Materials and Methods. The revised text has also been professionally edited.</p></body></sub-article></article>