Investigation of SNARE-Mediated Exocytosis Controlled by Synaptic Regulatory Proteins Jaeil Shin Iowa State University

Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations

2014 Investigation of SNARE-mediated exocytosis controlled by synaptic regulatory proteins Jaeil Shin Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Biochemistry Commons

Recommended Citation Shin, Jaeil, "Investigation of SNARE-mediated exocytosis controlled by synaptic regulatory proteins" (2014). Graduate Theses and Dissertations. 13995. https://lib.dr.iastate.edu/etd/13995

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Investigation of SNARE-mediated exocytosis controlled by synaptic regulatory proteins

Jaeil Shin

A dissertation submitted to graduate faculty

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Biochemstry

Program of Study Committee:

Yeon-Kyun Shin, Major Professor

Guru Rao

Scott Nelson

Michael Shogren-Knaak

Edward Yu

Iowa State University

Ames, Iowa

2014

TABLE OF CONTENTS

ABSTRACT ...... vi CHAPTER 1: GENERAL INTRODUCTION ...... 1

1.1 Membrane fusion and exocytosis ...... 1

1.2 SNAREs ...... 1

1.3 SNARE-mediated neuroexocytosis ...... 3

1.4 Regulation of neurotransmitter release ...... 4

1.4.1 Synaptotagmin-1...... 4

1.4.2 Sec1/Munc18-1 (SM) protein family ...... 5

1.4.3 Complexin-1 ...... 6

1.5 Investigation apparatus ...... 7

1.5.1 Nanodisc as a model membrane: reconstitution of a single membrane protein ...... 7

1.5.2 Site-directed spin labeling (SDSL) and electron paramagnetic resonance (EPR) ...... 8

1.5.3 Single molecular fluorescence (Förster) resonance energy transfer (smFRET) ...... 9

1.6 References ...... 11

1.7 Figures and Legends ...... 19

CHAPTER 2: MULTIPLE CONFORMATIONS OF A SINGLE SNAREPIN BETWEEN TWO NANODISC MEMBRANES REVEAL DIVERSE PRE-FUSION STATES ...... 34

2.1 Abstract ...... 34

2.2 Introduction ...... 35

2.3 Results ...... 36

2.3.1 Reconstitution of the SNAREpin into two nanodiscs ...... 36 iii

2.3.2 The t- and v-SNAREs assemble into a parallel configuration between two nanodiscs, some with the frayed C-terminal end ...... 37

2.3.3 EPR confirms that both the fully assembled four-helix bundle and the half-zipped complex represent trans-SNAREpin conformations ...... 38

2.3.4 The real-time analysis of single SNAREpins reveals a subpopulation with highly fluctuating dynamic ...... 40

2.4 Discussion ...... 41

2.5 Materials and Methods ...... 43

2.5.1 Protein expression, purification, and fluorophore- or spin-labeling ...... 43

2.5.2 Reconstitution and purification of SNARE-incorporated nanodiscs ...... 45

2.5.3 Lipid mixing between v-discs and either t-discs or t-liposomes ...... 46

2.5.4 FRET measurements of trans-SNARE complex with TIR ...... 46

2.5.5 Purification and measurement of spin labeled trans-SNARE complex in EPR spectroscopy ...... 47

2.6 References ...... 49

2.7 Figures and Legends ...... 52

2.8 Supporting Information ...... 58

CHAPTER 3: SYNAPTOTAGMIN 1 IS AN ANTAGONIST FOR MUNC18-1 IN SNARE-ZIPPERING ...... 65

3.1 Abstract ...... 65

3.2 Introduction ...... 66

3.3 Results ...... 67

3.3.1 Munc18-1 promotes SNARE zippering without syt1, but not in the presence of syt1 .67

3.3.2 Syt1 and Munc18-1 are mutually antagonistic in SNARE-dependent lipid mixing ....69 iv

3.3.3 Munc18-1 has little effect on Ca2+-triggered content mixing ...... 70

3.4 Discussion ...... 71

3.5 Materials and Methods ...... 73

3.5.1 Plasmid constructs and site-directed mutagenesis ...... 73

3.5.2 Protein expression and purification ...... 74

3.5.3 Lipid mixture preparation ...... 74

3.5.4 Fluorophore-labeling of the single cysteine mutants ...... 75

3.5.5 Reconstitution and purification of t- and v-discs ...... 76

3.5.6 FRET measurements of trans-SNAREpins ...... 76

3.5.7 proteoliposomes reconstitution ...... 77

3.5.8 Bulk lipid mixing assay ...... 79

3.5.9 Single vesicle docking and lipid mixing assays ...... 79

3.5.10 Single-vesicle content mixing assay ...... 80

3.6 References ...... 81

3.7 Figures and Legends ...... 86

3.8 Supporting Information ...... 90

CHAPTER 4: THE ROLE OF COMPLEXIN-1 N-TERMINUS AS AN ACTIVE ASSISTANT IN Ca2+ TRIGGERED FUSION PORE FORMATION ...... 95

4.1 Abstract ...... 95

4.2 Introduction ...... 95

4.3 Results ...... 98

4.3.1 Cpx1 facilitates content mixing events in the presence of syt1 and Ca2+ ...... 98

4.3.2 SNARE motif in the trans-SNAREpin is fully assembled by cpx1 ...... 98

4.3.3 Cpx1 increases vesicle-vesicle lipid mixing ...... 99 v

4.3.4 N-terminal region of cpx1 actively induces fusion pore opening ...... 100

4.4 Discussion ...... 101

4.5 Materials and Methods ...... 104

4.5.1 Protein preparation ...... 104

4.5.2 Fluorophore-labeling of cysteine mutant proteins ...... 105

4.5.3 Preparation of phospholipid mixtures ...... 105

4.5.4 Reconstitution of SNARE incorporated vesicles ...... 106

4.5.5 Assembly of SNARE-incorporated nanodiscs ...... 107

4.5.6 Single-molecule lipid mixing assay ...... 108

4.5.7 FRET acquisition from the trans-SNAREpin between nanodiscs ...... 108

4.5.8 Content mixing assay ...... 110

4.6 Reference ...... 111

4.7 Figures and Legends ...... 115

4.8 Supporting Information ...... 121

CHAPTER 5: GENERAL SUMMARY ...... 123

5.1 General conclusion ...... 123

5.2 References ...... 127

ACKNOWLEDGEMENT ...... 129

ABSTRACT

In neurons, SNAREs mediate membrane fusion between synaptic vesicles and plasma membranes to transfer neurotransmitters under regulation of other synaptic proteins in the synapse. It has previously been thought that there are distinct stages in membrane fusion events.

First of all, synaptic vesicles are docked to the plasma membrane and trans-SNAREpin complexes are formed between two membranes while complete membrane fusion such as pore opening is prevented until the arrival of sufficient regulations. At the proper time, SNARE complexes form fully-extended cis-conformations, then consequently mix membranes into a continuous bilayer and open a pore for the neurotransmitter release.

In a normal synapse, the fusion event mediated by SNAREs must be regulated in timely fashion. Synaptotagmin-1 (syt1) is a Ca2+ sensor in the synaptic vesicle characterized by an N- terminal transmembrane domain, a flexible liner, and tandem C2 domains, C2AB, at the C- terminus. Syt1 and Ca2+ influx are well-known regulators for the docking of synaptic vesicles to the plasma membrane and also consequent fusion pore opening, although the specific mechanism is still unclear. SNARE-controlled membrane fusion events also depend on Sec1/Munc18 (SM) proteins. Munc18-1 is known to strongly bind to the closed syntaxin-1a conformation and

Munc13 may play a role in the structural transition of syntaxin-1a to facilitate the formation of the t-SNARE complex. Complexin1 (cpx1) is a small protein containing four distinct functional regions. It binds to SNARE complexes and assists syt1 by simultaneously activating and clamping the core fusion machinery. vii

In the current work described here, we have established a new platform of trans-

SNAREpin study tool using nanodiscs. SNARE-incorporated nanodiscs are able to trap trans- conformation of SNAREpin without further full fusion processes. Single-molecule fluorescence resonance energy transfer (smFRET) based on the total internal reflection microscopy (TIR) and electron paramagnetic resonance (EPR) were recruited to investigate structural information about trans-SNAREpin trapped in nanodiscs and SNARE-mediated fusogenic efficiencies in the presence of synaptic regulatory proteins. We first confirmed the presence of multiple conformations of trans-SNAREpin complex based on the new platform. We were then able to observe the role of each synaptic regulatory protein, such as Munc18-1 and cpx1, in the presence of syt1 during SNARE complex assembly and membrane fusion processes. Our results suggest that syt1 plays an overall role of regulation in SNARE-mediated synaptic vesicle fusion, including docking, membrane mixing, and fusion pore opening, while the other regulators

Munc18-1 and cpx1 may also serve specific functions during this process. Munc18-1 was able to accelerate the complete SNAREpin formation and SNARE-mediated lipid mixing only in the absence of syt1, while cpx1 maintained the same function independent of syt1. Furthermore, content-mixing assays confirmed cpx1 N-terminal region acted as an active assistant for syt1 in

Ca2+ triggered fusion pore-opening while Munc18-1 did not affect the result.

Considering the evolutionary-conserved function of SM proteins as activators in intracellular membrane fusion controlled by SNAREs and the experimental results in this work, it appears that lately-evolved membrane fusion systems such as synaptic exocytosis might fire

Munc18-1 and recruit new controllers, syt1 and cpx1, to perform temporal and spatial regulation of sophisticated signal transmission. 1

CHAPTER 1: GENERAL INTRODUCTION

1.1 Membrane fusion and exocytosis

Membrane fusion is one life’s most fundamental processes and is performed by merging two separated lipid membranes into a single continuous bilayer1. Although this universal reaction can vary in contact space and time by species and also organelles, all fusion events demonstrate similar underlying stages such as membrane contact, fusion between proximal leaflets, and opening a pore (Fig. 1)2. Membrane fusion events can also be categorized into three major types:

1) Extra- and intracellular fusion of pathogens with host cells. 2) Extracellular fusion of eukaryotic cells. 3) Intracellular fusion of organelles. Exocytosis is one type of intracellular fusion and is actuated either spontaneously or by specialized signals to transfer contents from one site to another. More specifically, in neurons the fusion process between synaptic vesicles and pre-synaptic plasma membranes is triggered by calcium ions (Ca2+) that exert neurotransmitter release following signal transmission3,4. These membrane-fusion events are mainly mediated by dynamic supramolecular assemblies involving conserved protein families.

1.2 SNAREs

SNARE (soluble N-ethyl-maleimide-sensitive factor attachment protein receptor) proteins are a large protein superfamily including tens of members occurring in yeast and

mammalian cells and identified as interacting proteins with SNAP (soluble N-ethyl-maleimide- sensitive factor attachment protein) and NSF (N-ethyl-maleimide-sensitive factor) simultaneously5-9. SNAREs can be divided into two main categories: v-SNAREs localized in cargo-containing vesicles, and t-SNAREs in the target membrane compartment. In presynaptic membranes, VAMP-2 (v-SNARE), also called synaptobrevin-2, is anchored to the synaptic vesicle via its C-terminal transmembrane domain10,11. Presynaptic plasma membranes, at the other site, harbor syntaxin-1a through its C-terminal transmembrane domain and also SNAP-25, are fundamentally attached to the membrane by palmitoylation of cysteine residues and bound to syntaxin-1a to form the binary complex (t-SNARE)12,13. All SNAREs contain a stretch of about

60-70 amino acid residues, called a motif domain, consisting of highly conserved ~8 heptad repeats14,15. The hydrophobic-coiled coil interface within the four helical bundle contains 16 layers contributed by the direct contact of a and d residue positions of each SNARE motif (Fig.

2)16. Throughout the SNARE superfamily, amino acid residues at the 0th layer are highly conservative as arginine (R) in VAMP-2 and glutamine (Q) in syntaxin-1a or SNAP-2517,18. The

SNARE motif and transmembrane domains in both VAMP-2 and syntaxin-1a are connected by

~10 amino acid linker regions. Additionally, syntaxin-1a contains an independently-folded helical domain known as the Habc domain at the N-terminus. The three-helical-bundle shape of the Habc domain folds back to the SNARE motif domain of syntaxin-1a to form a non-active closed conformation19,20. Presynaptic regulatory proteins such as Munc13/Munc18-1 have been proposed to interact and activate the dead-end syntaxin-1a (Fig. 3)21,22. A single SNARE motif contributed by VAMP-2 and syntaxin-1a, respectively, and two motifs from SNAP-25 spontaneously form a four helical coiled coil complex in solution generating the parallel cis-

SNAREpin conformation (Fig. 4)18,23. The complex conformation has been proposed as being initiated at the N-termini of SNARE motifs and zipped towards membrane-proximal C-termini24-

27. The primary role of SNAREs is to mediate membrane fusion as the minimal fusogenic machinery during an exocytosis process28,29. The repulsion of membranes at the contact site is believed to be overcome by the energy force occurring when SNAREs form stable helical bundles through their hydrophobic-favored interaction1,30. However, it is strongly believed that the SNARE complex has an intermediate conformation, a so-called trans-SNAREpin, in which two opposite transmembrane domains are anchored into an apposed membrane while N-terminal regions are zipped31-33.

1.3 SNARE-mediated neuroexocytosis

In neuroexocytosis, synaptic vesicles are brought to and fused with the presynaptic plasma membrane to release neurotransmitters34. The initiation of SNARE complex formation at the N-termini creates a docked synaptic-vesicle pool on the plasma membrane. Further SNARE complex formation, maintaining trans-conformation, brings two membranes into proximity and merges only the outer leaflet of each membrane constituting a hemifusion state (Fig. 5). At this point, the presynaptic region is filled with docked or partially fused synaptic vesicles that are ready to synchronously transfer signals (readily releasable pool (RRP)) upon arrival or by specific triggering such as a Ca2+ influx. Hemifused membranes resemble the partial SNARE complex in which N-termini are stably zipped, while C-termini of transmembrane domains are still anchored onto opposing membranes. It is widely accepted that there are cooperative

regulatory molecules in synapses to control membrane contact, hemifusion, and final pore opening (Fig. 6).

1.4 Regulation of neurotransmitter release

As minimal machinery for membrane fusion, SNAREs do not have a self-regulatory process for controlling full complex formation and membrane fusion. In physiology, however,

SNARE-mediated neuroexocytosis must be regulated to precisely release the signal. Although conventional in vitro studies show that SNAREs require more than just minutes of time to complete lipid mixing events, neurotransmitter release occurs within about 2 ms. Besides the minimal fusogenic machinery, other regulatory molecules must be involved to orchestrate timely-conducted signal transmission.

1.4.1 Synaptotagmin-1

Synaptotagmin-1 (syt1) is anchored to the synaptic vesicle via its single transmembrane domain at the N-terminus and senses the Ca2+ influx with tandem C2A and C2B domains at the

C-terminus, which is connected through its flexible linker region (Fig. 7)35-39. The loop region of both C2A and C2B domains (C2AB), especially in aspartate amino acids, are responsible for

Ca2+ binding40. The Ca2+ incorporating C2AB domain is capable of interaction with negatively- charged phospholipids such as phosphatidylserine (PS) or phosphatidylinositol 4, 5-bisphophate

41 (PIP2) in the plasma membrane . Based on recent studies, the specific interaction between syt1

and PIP2 facilitates the membrane fusion process, possibly contributing to t-SNARE clustering or membrane curvature formation42-44. Also, a asymmetric lipid composition study indicates that vesicular protein syt1 favorably interacts with the plasma membrane to exert its regulatory role in SNARE-mediated membrane fusion events45.

As a regulatory protein for the synaptic fusion, syt1 is also known to bind to the SNARE complex or the binary complex directly, and it is believed that these interactions lead to controlled fusion occurrence mediated by SNAREs40,43,44. Considering the zipping of the

SNARE complex from the N- to the C-terminus direction and the binding ability of syt1 to C- terminal SNARE complex in the presence of Ca2+, syt1 seems to fasten SNARE complex formation and result in accelerated membrane fusion46-48.

1.4.2 Sec1/Munc18-1 (SM) protein family

All SM proteins are ~60 kDa and expressed in the cytosol of eukaryotic cells as in yeast or vertebrates. In neurons, three SM proteins including Munc18-1, -2, and -3, have been identified as interacting at the plasma membrane49-52. Munc18-1 is considered to be a required component in the process of membrane fusion and exocytosis since the deletion of Munc18-1 abolishes vesicle exocytosis53,54. It has been asserted that Munc18-1 is able to bind its cognate trans-SNAREpin and accelerate SNARE-mediated membrane fusion and probably to speed up

SNARE-zipping fusion (Fig. 8a)21,55-57. The other documented property of Munc18-1 is to stabilize a ‘closed’ syntaxin-1a by capping the four helical bundles of syntaxin-1a contributed by

folding back N-terminal Habc and SNARE motif domains (Fig. 8b)58-61. It has been proposed that the specific interaction between Munc18-1 and syntaxin-1a leads to the isolation of syntaxin-1a from the binary complex with SNAP-25, thereby preventing the initiation of a

SNARE-mediated fusion event21. However, the acceleratory role of Munc18-1 for a fusion process is widely supported by in vitro vesicle fusion experiments and it is conserved in SM proteins throughout species58,62. One could note that the previously-reported enhancement of fusion by Munc18-1 was observed in the absence of syt1, which senses the Ca2+ to cause membrane fusion and pore opening, so it must be involved in experiments to determine the role of Munc18-1 in the evolved neuron system53.

1.4.3 Complexin-1

A family of evolutionarily-conserved ubiquitous proteins called complexin (cpx), enriched in neurons and competing with a-SNAP for binding to the SNARE complex, have been identified in species, suggesting a role in regulating SNARE function during membrane fusion63,64. Also, cpx-deficient neurons produce a phenocopy of syt1-deficient neurons with a suppression of fast synchronous exocytosis upon the arrival of a Ca2+ influx, supporting the idea that cpx functions as a syt1-dependent regulator of SNARE-mediated fusion65. To date, four isoforms of cpx have been identified (cpx1, 2, 3, and 4)66. Cpx1 and 2, ~85% identical in amino acid sequence, are cytosolic proteins in synapses, while cpx3 and4 are expressed at comparatively low levels in neurons. Henceforth, cpx1 will be discussed as a regulatory protein for the SNARE function in membrane fusion. The established model for the role of cpx1 is to

prime the synaptic vesicle to the plasma membrane as an activator and clamp the spontaneous release67-74. Cpx1 is a SNARE-specific binding protein (134 amino acid residues) containing two short α helices flanked by flexible N- and C-termini (Fig. 9)75. The distinct domains of cpx1 have been studied and proposed as exhibiting individual roles. The central more C-terminal α helix directly binds to the groove between VAMP-2 and syntaxin-1a in the SNARE complex or the binary complex of sytnaxin-1a and SNAP-2570,76. The N-terminal α helix is believed to compete with VAMP-2 for the binding site at the binary complex, and it therefore inhibits the fusion process mediated by SNAREs77. The N-terminal flexible domain activates a fusion step, especially for synchronous release but not for the clamp role69,70. The opposite C-terminal flexible domain is known to be related to priming and clamping roles but not for Ca2+ triggered release73. All distinct domains of cpx1 are required to provide sequential roles such as clamping, priming, and Ca2+ triggered release34. The clamping and activating role of cpx1 for fusion events is thought to be performed under cooperation with syt1 even though the mechanism is not well understood.

1.5 Investigation apparatus

1.5.1 Nanodisc as a model membrane: reconstitution of a single membrane protein

High-density lipoproteins (HDL) are protein-lipid particles that collect cholesterol, fatty acids, and lipids in the blood stream as spherical shapes and transport them to the liver for degradation (Fig. 10)78. Apolipoprotein A-1 (apoA-1) is the primary protein that scaffolds the

HDL structure by surrounding the lipid surface. The apo A-1 protein is the soluble protein and contains ~200 amino acid residues and consists of amphipathic α helices. The S. Sligar group generated discoidal bilayer lipids supported by purified apo A-1 proteins to produce homogenously-sized (~10 nm of the diameter) nanodiscs (Fig. 11)79. The application of nanodiscs is to incorporated proteins that contain transmembrane domain(s) used to investigate the biological function of proteins in their native membrane environment80-82. Throughout the study of this dissertation, the nanodisc is recruited to reconstitute single SNARE proteins,

VAMP-2, and syntaxin-1a to generate partially-assembled trans-SNAREpin in the presence of the native lipid molecules.

1.5.2 Site-directed spin labeling (SDSL) and electron paramagnetic resonance (EPR)

EPR spectroscopy, also called electron spin resonance (ESR) spectroscopy, measures the interaction of electron spins with a magnetic field. The energy difference ( ) between the high and low energy states of the electron spin is given by

Where h is Planck’s constant (6.626 x 10-34 J second), v is the frequency of the electromagnetic radiation, g is the splitting factor, is the Bohr magneton, and B is the strength of the external magnetic field83. The first-derivative spectra of EPR absorbance is recorded at constant microwave frequency and spectra peaks are split as a result of hyperfine interaction between the electron spin and the nearby nuclei (the number of the peaks = 2I +1, where I is the nuclear spin,

Fig. 12)84. One of the most useful applications of EPR is the measurement of the dynamic mobility of molecules like proteins, especially in membranes85. In SDSL, protein residues are replaced by cysteine and then labeled by the spin molecule one at a time, eventually mapping the entire area of interest86,87. The reagent of most widespread use is methanethiosulfonate, spin label (MTSSL), which is covalently labeled to the cysteine residue via disulfide bonding (Fig.

13)88.

In this dissertation, EPR was recruited to determine the local structure of the VAMP-2 protein when it formed partially-zipped trans-SNAREpin with the binary complex between two nanodiscs. The sharpness of the first-derivative EPR spectra from the individual residues harboring the spin label reflected the local structure stability.

1.5.3 Single molecular fluorescence (Förster) resonance energy transfer (smFRET)

FRET is a mechanism of energy transfer between two light-sensitive molecules serving as a donor and acceptor, respectively, and able to perform analysis of structure, dynamics, and interactions of biomolecules89-91. The FRET efficiency (E) is a function of the distance between donor and acceptor (R) , as shown in the equation

where R0 (the Förster radius) is the distance for which E equals 0.5. Consequently, FRET is extremely sensitive to a change of distance between fluorophore-labeled molecules, and can

therefore be used as a spectroscopic ruler92. FRET paired-molecules can be applied to a single molecule level (smFRET), combining with surface immobilized total internal reflection fluorescence microscopy (TIR)93,94. In a TIR setup, a totally-reflected specific laser source generates an evanescent field that excites the parallel area from the surface up to a distance of approximately 200 nm (Fig. 14). This specific field abolishes the noise and background signal to allow detection of single fluorophore-labeled biomolecules95. Another use of smFRET is to observe the real-time dynamics of fluorophore-labeled molecules. In the case of two interacting molecules at fluctuating distances, smFRET is able to measure kinetics based on donor and acceptor signals, as well as FRET values (Fig. 15)47,96.

Throughout this dissertation, smFRET provided information about assembled SNAREpin structure between fluorophore-labeled trans-SNAREpin or lipid-mixing events between fluorophore-labeled vesicles based on FRET assays. Additionally, content-mixing assays harboring single fluorophore-containing vesicles were determined using a TIR setup.

1.6 References

1. Monck, J.R. & Fernandez, J.M. The fusion pore and mechanisms of biological membrane fusion. Current Opinion in Cell Biology 8, 524-533 (1996).

2. Jahn, R., Lang, T. & Südhof, T.C. Membrane Fusion. Cell 112, 519-533 (2003).

3. Benfenati, F. & Valtorta, F. Neuroexocytosis. in Clostridial Neurotoxins, Vol. 195 (ed. Montecucco, C.) 195-219 (Springer Berlin Heidelberg, 1995).

4. Burgess, T.L. & Kelly, R.B. Constitutive and Regulated Secretion of Proteins. Annual Review of Cell Biology 3, 243-293 (1987).

5. Sollner, T., Bennett, M.K., Whiteheart, S.W., Scheller, R.H. & Rothman, J.E. A protein assembly-disassembly pathway in vitro that may correspond to sequential steps of synaptic vesicle docking, activation, and fusion. Cell 75, 409-18 (1993).

6. Clary, D.O., Griff, I.C. & Rothman, J.E. SNAPs, a family of NSF attachment proteins involved in intracellular membrane fusion in animals and yeast. Cell 61, 709-21 (1990).

7. Block, M.R., Glick, B.S., Wilcox, C.A., Wieland, F.T. & Rothman, J.E. Purification of an N-ethylmaleimide-sensitive protein catalyzing vesicular transport. Proc Natl Acad Sci U S A 85, 7852-6 (1988).

8. Chen, Y.A. & Scheller, R.H. SNARE-mediated membrane fusion. Nat Rev Mol Cell Biol 2, 98-106 (2001).

9. Söllner, T., Bennett, M.K., Whiteheart, S.W., Scheller, R.H. & Rothman, J.E. A protein assembly-disassembly pathway in vitro that may correspond to sequential steps of synaptic vesicle docking, activation, and fusion. Cell 75, 409-418 (1993).

10. Baumert, M., Maycox, P.R., Navone, F., De Camilli, P. & Jahn, R. Synaptobrevin: an integral membrane protein of 18,000 daltons present in small synaptic vesicles of rat brain. EMBO J 8, 379-84 (1989).

11. Trimble, W.S., Cowan, D.M. & Scheller, R.H. VAMP-1: a synaptic vesicle-associated integral membrane protein. Proc Natl Acad Sci U S A 85, 4538-42 (1988).

12. Oyler, G.A. et al. The identification of a novel synaptosomal-associated protein, SNAP- 25, differentially expressed by neuronal subpopulations. J Cell Biol 109, 3039-52 (1989).

13. Sollner, T. et al. SNAP receptors implicated in vesicle targeting and fusion. Nature 362, 318-24 (1993).

14. Jahn, R., Lang, T. & Sudhof, T.C. Membrane fusion. Cell 112, 519-33 (2003).

15. Weimbs, T. et al. A conserved domain is present in different families of vesicular fusion proteins: a new superfamily. Proc Natl Acad Sci U S A 94, 3046-51 (1997).

16. Poirier, M.A. et al. The synaptic SNARE complex is a parallel four-stranded helical bundle. Nat Struct Mol Biol 5, 765-769 (1998).

17. Jahn, R. & Scheller, R.H. SNAREs [mdash] engines for membrane fusion. Nat Rev Mol Cell Biol 7, 631-643 (2006).

18. Sutton, R.B., Fasshauer, D., Jahn, R. & Brunger, A.T. Crystal structure of a SNARE complex involved in synaptic exocytosis at 2.4 A resolution. Nature 395, 347-53 (1998).

19. Margittai, M., Fasshauer, D., Jahn, R. & Langen, R. The Habc Domain and the SNARE Core Complex Are Connected by a Highly Flexible Linker. Biochemistry 42, 4009-4014 (2003).

20. Fernandez, I. et al. Three-Dimensional Structure of an Evolutionarily Conserved N- Terminal Domain of Syntaxin 1A. Cell 94, 841-849 (1998).

21. Ma, C., Su, L., Seven, A.B., Xu, Y. & Rizo, J. Reconstitution of the Vital Functions of Munc18 and Munc13 in Neurotransmitter Release. Science 339, 421-425 (2013).

22. Hata, Y., Slaughter, C.A. & Sudhof, T.C. Synaptic vesicle fusion complex contains unc- 18 homologue bound to syntaxin. Nature 366, 347-51 (1993).

23. Stein, A., Weber, G., Wahl, M.C. & Jahn, R. Helical extension of the neuronal SNARE complex into the membrane. Nature 460, 525-8 (2009).

24. Pobbati, A.V., Stein, A. & Fasshauer, D. N- to C-Terminal SNARE Complex Assembly Promotes Rapid Membrane Fusion. Science 313, 673-676 (2006).

25. Weber, T. et al. SNAREpins: minimal machinery for membrane fusion. Cell 92, 759-72 (1998).

26. Lang, T. & Jahn, R. Core proteins of the secretory machinery. Handb Exp Pharmacol, 107-27 (2008).

27. Sudhof, T.C. & Rothman, J.E. Membrane fusion: grappling with SNARE and SM proteins. Science 323, 474-7 (2009).

28. McNew, J.A. et al. Close is not enough: SNARE-dependent membrane fusion requires an active mechanism that transduces force to membrane anchors. J Cell Biol 150, 105-17 (2000).

29. Weber, T. et al. SNAREpins: Minimal Machinery for Membrane Fusion. Cell 92, 759- 772 (1998).

30. Zimmerberg, J., Vogel, S.S. & Chernomordik, L.V. Mechanisms of membrane fusion. Annu Rev Biophys Biomol Struct 22, 433-66 (1993).

31. Gao, Y. et al. Single Reconstituted Neuronal SNARE Complexes Zipper in Three Distinct Stages. Science 337, 1340-1343 (2012).

32. Min, D. et al. Mechanical unzipping and rezipping of a single SNARE complex reveals hysteresis as a force-generating mechanism. Nat Commun 4, 1705 (2013).

33. Shin, J., Lou, X., Kweon, D.H. & Shin, Y.K. Multiple conformations of a single SNAREpin between two nanodisc membranes reveal diverse pre-fusion states. Biochemical Journal 459, 95-102 (2014).

34. Südhof, Thomas C. Neurotransmitter Release: The Last Millisecond in the Life of a Synaptic Vesicle. Neuron 80, 675-690 (2013).

35. Rizo, J. & Rosenmund, C. Synaptic vesicle fusion. Nat Struct Mol Biol 15, 665-674 (2008).

36. Lee, H.-K. et al. Dynamic Ca2+-Dependent Stimulation of Vesicle Fusion by Membrane- Anchored Synaptotagmin 1. Science 328, 760-763 (2010).

37. Südhof, T.C. THE SYNAPTIC VESICLE CYCLE. Annual Review of Neuroscience 27, 509-547 (2004).

38. Yoshihara, M. & Littleton, J.T. Synaptotagmin I Functions as a Calcium Sensor to Synchronize Neurotransmitter Release. Neuron 36, 897-908 (2002).

39. Chapman, E.R. Synaptotagmin: A Ca2+ sensor that triggers exocytosis? Nat Rev Mol Cell Biol 3, 498-508 (2002).

40. Chapman, E.R. How Does Synaptotagmin Trigger Neurotransmitter Release? Annual Review of Biochemistry 77, 615-641 (2008).

41. Martens, S., Kozlov, M.M. & McMahon, H.T. How Synaptotagmin Promotes Membrane Fusion. Science 316, 1205-1208 (2007).

42. van den Bogaart, G. et al. Membrane protein sequestering by ionic protein-lipid interactions. Nature 479, 552-555 (2011).

43. Kim, J.Y. et al. Solution single‐vesicle assay reveals PIP2‐mediated sequential actions of synaptotagmin‐1 on SNAREs. The EMBO Journal 31, 2144-2155 (2012).

44. Hui, E., Johnson, C.P., Yao, J., Dunning, F.M. & Chapman, E.R. Synaptotagmin- Mediated Bending of the Target Membrane Is a Critical Step in Ca2+-Regulated Fusion. Cell 138, 709-721 (2009).

45. Lai, Y. & Shin, Y.K. The importance of an asymmetric distribution of acidic lipids for synaptotagmin 1 function as a Ca2+ sensor. Biochemical Journal 443, 223-229 (2012).

46. Dai, H., Shen, N., Araç, D. & Rizo, J. A Quaternary SNARE–Synaptotagmin–Ca2+– Phospholipid Complex in Neurotransmitter Release. Journal of Molecular Biology 367, 848-863 (2007).

47. Choi, U.B. et al. Single-molecule FRET-derived model of the synaptotagmin 1-SNARE fusion complex. Nat Struct Mol Biol 17, 318-324 (2010).

48. Gerona, R.R.L., Larsen, E.C., Kowalchyk, J.A. & Martin, T.F.J. The C Terminus of SNAP25 Is Essential for Ca2+-dependent Binding of Synaptotagmin to SNARE Complexes. Journal of Biological Chemistry 275, 6328-6336 (2000).

49. Hata, Y., Slaughter, C.A. & Sudhof, T.C. Synaptic vesicle fusion complex contains unc- 18 homologue bound to syntaxin. Nature 366, 347-351 (1993).

50. Garcia, E.P., Gatti, E., Butler, M., Burton, J. & De Camilli, P. A rat brain Sec1 homologue related to Rop and UNC18 interacts with syntaxin. Proceedings of the National Academy of Sciences 91, 2003-2007 (1994).

51. Pevsner, J., Hsu, S.C. & Scheller, R.H. n-Sec1: a neural-specific syntaxin-binding protein. Proceedings of the National Academy of Sciences 91, 1445-1449 (1994).

52. Tellam, J.T., McIntosh, S. & James, D.E. Molecular Identification of Two Novel Munc- 18 Isoforms Expressed in Non-neuronal Tissues. Journal of Biological Chemistry 270, 5857-5863 (1995).

53. Verhage, M. et al. Synaptic Assembly of the Brain in the Absence of Neurotransmitter Secretion. Science 287, 864-869 (2000).

54. Schekman, R. Genetic and biochemical analysis of vesicular traffic in yeast. Current Opinion in Cell Biology 4, 587-592 (1992).

55. Shen, J., Tareste, D.C., Paumet, F., Rothman, J.E. & Melia, T.J. Selective activation of cognate SNAREpins by Sec1/Munc18 proteins. Cell 128, 183-95 (2007).

56. Deak, F. et al. Munc18-1 binding to the neuronal SNARE complex controls synaptic vesicle priming. J Cell Biol 184, 751-64 (2009).

57. Diao, J. et al. Single-Vesicle Fusion Assay Reveals Munc18-1 Binding to the SNARE Core Is Sufficient for Stimulating Membrane Fusion. ACS Chem Neurosci 1, 168-174 (2010).

58. Shen, J., Tareste, D.C., Paumet, F., Rothman, J.E. & Melia, T.J. Selective Activation of Cognate SNAREpins by Sec1/Munc18 Proteins. Cell 128, 183-195.

59. Misura, K.M.S., Scheller, R.H. & Weis, W.I. Three-dimensional structure of the neuronal-Sec1-syntaxin 1a complex. Nature 404, 355-362 (2000).

60. Pevsner, J. et al. Specificity and regulation of a synaptic vesicle docking complex. Neuron 13, 353-361.

61. Rickman, C., Medine, C.N., Bergmann, A. & Duncan, R.R. Functionally and Spatially Distinct Modes of munc18-Syntaxin 1 Interaction. Journal of Biological Chemistry 282, 12097-12103 (2007).

62. Yu, H. et al. Comparative studies of Munc18c and Munc18-1 reveal conserved and divergent mechanisms of Sec1/Munc18 proteins. Proceedings of the National Academy of Sciences 110, E3271-E3280 (2013).

63. McMahon, H.T., Missler, M., Li, C. & Südhof, T.C. Complexins: Cytosolic proteins that regulate SNAP receptor function. Cell 83, 111-119 (1995).

64. Takahashi, S. et al. Identification of two highly homologous presynaptic proteins distinctly localized at the dendritic and somatic synapses. FEBS Letters 368, 455-460 (1995).

65. Reim, K. et al. Complexins Regulate a Late Step in Ca2+-Dependent Neurotransmitter Release. Cell 104, 71-81 (2001).

66. Reim, K. et al. Structurally and functionally unique complexins at retinal ribbon synapses. The Journal of Cell Biology 169, 669-680 (2005).

67. Giraudo, C.G., Eng, W.S., Melia, T.J. & Rothman, J.E. A Clamping Mechanism Involved in SNARE-Dependent Exocytosis. Science 313, 676-680 (2006).

68. Tang, J. et al. A Complexin/Synaptotagmin 1 Switch Controls Fast Synaptic Vesicle Exocytosis. Cell 126, 1175-1187 (2006).

69. Xue, M. et al. Distinct domains of complexin I differentially regulate neurotransmitter release. Nat Struct Mol Biol 14, 949-958 (2007).

70. Maximov, A., Tang, J., Yang, X., Pang, Z.P. & Südhof, T.C. Complexin Controls the Force Transfer from SNARE Complexes to Membranes in Fusion. Science 323, 516-521 (2009).

71. Martin, J.A., Hu, Z., Fenz, K.M., Fernandez, J. & Dittman, J.S. Complexin Has Opposite Effects on Two Modes of Synaptic Vesicle Fusion. Current Biology 21, 97-105 (2011).

72. Hobson, R.J., Liu, Q., Watanabe, S. & Jorgensen, E.M. Complexin Maintains Vesicles in the Primed State in C. elegans. Current Biology 21, 106-113 (2011).

73. Kaeser-Woo, Y.J., Yang, X. & Südhof, T.C. C-Terminal Complexin Sequence Is Selectively Required for Clamping and Priming But Not for Ca2+ Triggering of Synaptic Exocytosis. The Journal of Neuroscience 32, 2877-2885 (2012).

74. Jorquera, R.A., Huntwork-Rodriguez, S., Akbergenova, Y., Cho, R.W. & Littleton, J.T. Complexin Controls Spontaneous and Evoked Neurotransmitter Release by Regulating the Timing and Properties of Synaptotagmin Activity. The Journal of Neuroscience 32, 18234-18245 (2012).

75. Chen, X. et al. Three-Dimensional Structure of the Complexin/SNARE Complex. Neuron 33, 397-409 (2002).

76. Weninger, K., Bowen, M.E., Choi, U.B., Chu, S. & Brunger, A.T. Accessory Proteins Stabilize the Acceptor Complex for Synaptobrevin, the 1:1 Syntaxin/SNAP-25 Complex. Structure 16, 308-320 (2008).

77. Yang, X., Kaeser-Woo, Y.J., Pang, Z.P., Xu, W. & Südhof, T.C. Complexin Clamps Asynchronous Release by Blocking a Secondary Ca2+ Sensor via Its Accessory α Helix. Neuron 68, 907-920 (2010).

78. Navab, M., Reddy, S.T., Van Lenten, B.J. & Fogelman, A.M. HDL and cardiovascular disease: atherogenic and atheroprotective mechanisms. Nat Rev Cardiol 8, 222-232 (2011).

79. Bayburt, T.H., Grinkova, Y.V. & Sligar, S.G. Self-Assembly of Discoidal Phospholipid Bilayer Nanoparticles with Membrane Scaffold Proteins. Nano Letters 2, 853-856 (2002).

80. Nath, A., Atkins, W.M. & Sligar, S.G. Applications of Phospholipid Bilayer Nanodiscs in the Study of Membranes and Membrane Proteins?? Biochemistry 46, 2059-2069 (2007).

81. Bayburt, T.H. & Sligar, S.G. Membrane protein assembly into Nanodiscs. FEBS Lett 584, 1721-7 (2009).

82. Borch, J. & Hamann, T. The nanodisc: a novel tool for membrane protein studies. Biol Chem 390, 805-14 (2009).

83. Klare Johann, P. Site-directed spin labeling EPR spectroscopy in protein research. in Biological Chemistry Vol. 394 1281 (2013).

84. Melanson, M., Sood, A., Török, F. & Török*, M. Introduction to spin label electron paramagnetic resonance spectroscopy of proteins. Biochemistry and Molecular Biology Education 41, 156-162 (2013).

85. Marsh, D. Electron spin resonance in membrane research: protein–lipid interactions from challenging beginnings to state of the art. European Biophysics Journal 39, 513-525 (2010).

86. Hubbell, W.L., McHaourab, H.S., Altenbach, C. & Lietzow, M.A. Watching proteins move using site-directed spin labeling. Structure 4, 779-783.

87. Cornish, V.W. et al. Site-specific incorporation of biophysical probes into proteins. Proceedings of the National Academy of Sciences 91, 2910-2914 (1994).

88. Berliner, L.J., Grunwald, J., Hankovszky, H.O. & Hideg, K. A novel reversible thiol- specific spin label: Papain active site labeling and inhibition. Analytical Biochemistry 119, 450-455 (1982).

89. Heyduk, T. Measuring protein conformational changes by FRET/LRET. Current Opinion in Biotechnology 13, 292-296 (2002).

90. Cheng, P.-C. The Contrast Formation in Optical Microscopy. in Handbook Of Biological Confocal Microscopy (ed. Pawley, J.B.) 162-206 (Springer US, 2006).

91. Hillisch, A., Lorenz, M. & Diekmann, S. Recent advances in FRET: distance determination in protein–DNA complexes. Current Opinion in Structural Biology 11, 201-207 (2001).

92. Stryer, L. & Haugland, R.P. Energy transfer: a spectroscopic ruler. Proceedings of the National Academy of Sciences 58, 719-726 (1967).

93. Ha, T. et al. Probing the interaction between two single molecules: fluorescence resonance energy transfer between a single donor and a single acceptor. Proceedings of the National Academy of Sciences 93, 6264-6268 (1996).

94. Weiss, S. Fluorescence Spectroscopy of Single Biomolecules. Science 283, 1676-1683 (1999).

95. Ha, T. Single-Molecule Fluorescence Resonance Energy Transfer. Methods 25, 78-86 (2001).

96. Jain, A., Liu, R., Xiang, Y.K. & Ha, T. Single-molecule pull-down for studying protein interactions. Nat. Protocols 7, 445-452 (2012).

1.7 Figures and Legends

Figure 1) General process of lipid bilayer fusion event.

(a) Two opposite membranes are brought into close proximity where electrostatic repulsions need to be overcome by external force (i.e. fusion proteins). (b) The boundary between the hydrophilic and hydrophobic portion of the bilayer is destabilized and the non- bilayer intermediate is formed between two membranes. (c) The formation of aqueous pore results in the formation of continuous bilayers.

Figure 2) Coiled coil SNARE complex formation.

(a) Organization of conserved SNARE complex. C alpha traces (grey), local helical axes (blue, red and green for VAMP-2, syntaxin-1a and SNAP-25, respectively), the superhelical axis (black), and layers (+8 to -7, C- to N-terminal direction) are shown. (b) Helical wheel representation of the SNARE complex contributed by VAMP-2, syntaxin-1a and SNAP-25 colored by yellow, green and red/blue, respecitvely. View is from the N-terminus and repeated heptad positions are labled a through g. (c) A structural model of SNARE complex.

Figure 3) A model of syntaxin-1a activation for SNARE-mediated membrane fusion by

Munc13/Munc18-1.

In the upper right panel, syntaxin-1a (yellow) is shown in a closed conformation bound to Munc18-1 and an open conformation bound to SNAP-25 (green). Inactivated closed syntaxin-1a- Munc18-1 heterodimer (upper left) turns into activated open conformation by the joining of Munc13 (lower left). By the triggering signal such as Ca2+, SNAREs finishes membrane fusion event (lower right).

Figure 4) SNARE domains and fully extended helical complex structure.

(a) Colorized domains of each SNARE protein. Syntaxin-1a and VAMP-2 contain single SNARE motif (red and blue) and transmembrane domain (yellow), connected by linker region (grey), while SNAP-25 contributes two SNARE motif domains (green). (b) Ribbon structure of the fully assembled SNARE complex including linkers and transmembrane domains.

Figure ) 5 Intermediate states in membrane fusion process.

(a) Two apposed membranes are in a close contact when fusion event is initiated. (b,c, and e) Destabilization of membrane surface in contact results in the merging of the proximal monolayers, which is often referred to as stalks or hemifusion state. (d and f) A transition from trans-bilayer to isotropic, micellar-like phase results in a pore formation and continuous bilayers.

Figure 6) SNARE-mediated neuroexocytosis.

(a) The synaptic vesicle is docked to the plasma membrane by the trans-SNARE complex formation between VAMP-2 and the binary complex of syntaxin-1a and SNAP-25 regulated by SM proteins (Munc13 and Mucn18-1). (b) Once SNAREs are partially assembled, cpx1 binds to further increase their priming. (c) The fusion pore opening is occurred by Ca2+ binding to syt1, which causes an interaction of syt1 with SNAREs and phospholipids. (d) After fusion pore opening, the resulting cis-SNARE complexes are disassembled by the NSF/SNAP ATPases and vesicles are recycled.

Figure 7) Structure of Ca2+ sensing synaptic vesicle protein, synaptotagmin 1 (syt1).

Domain structure of syt1, the C2A and C2B domains were rendered from references (Shao X, et al., 1998, Biochemistry and Fernandex I. et al., 2001, Neuron) using PyMOL. The remaining protein segments such as transmembrane domain (TMD) and linker region were added. Critical residues for Phospholipid membrane and SNARE complex binding are shown as K326/K327 in C2B and R233 in C2A / K366 in C2B, respectively. Ca2+ ions are depicted as red spheres.

Figure 8) Munc18-1 as a synaptic regulatory protein.

(a) Model for the interactions of Munc18-1 with synaptic SNARE proteins. The left panel indicates the binding of Munc18-1 to the closed conformation of syntaxin-1a that may prevent the proceeding of SNARE-mediated membrane fusion. The right diagram displays the binding of Munc18-1 to the assembled SNARE complex during the fusion reaction, which may prevent diffusion of SNARE complexes into the membrane. (b) The crystal structure of the Munc18-1- syntaxin-1a complex. Munc18-1 domains 1, 2, and 3 are defined as described (Misura et al., 2000, Nature) and colored blue, green, and yellow, respectively. The SNARE motif domain of syntaxin-1a is colored purple, and the N-terminal Habc domain is colored red.

Figure 9) Complexin 1 (cpx1) as a SNARE binding synaptic regulatory protein.

(a) Distinct cpx1 domains are colored blue, yellow, pink, and grey for N-terminus, accessory α helix, central α helix, and C-terminus. (b) Cartoon of the 3D structure of cpx1- SNARE complex (Chen X. et al., 2002, Neuron). The central α helix domain of cpx1 lies along the interface between VAMP-2 and syntaxin-1a making numerous antiparallel contacts with both. The accessory α helix continues away from the helical bundle toward the membrane even it lacks contacts with SNARE proteins.

Figure 10) The discoidal nanodisc formation in reverse cholesterol transport (RCT) pathway.

Lipid-free apo A-1 secreted by the liver binds to serum phospholipids to form nascent discoidal HDL particles, which interact with ATP-binding cassette transporter A1 (ABCA1). Cholesterol is esterified to cholesterol esters by LCAT to form mature spheroidal HDL particles. This cholesterol is subsequently taken up by the liver and excreted into the bile after conversion to bile acid.

Figure 11) Illustrations of nanodisc structures.

Nanodiscs composed of apo A-1 and phospholipid are shown in side (left) and top view (right). The two apo A-1 proteins are colored orange and blue.

Figure 12) Zeeman and hyperfine splitting for a nitroxide spin label (MTSSL).

In the presence of an external magnetic field (B), the energy levels of the unpaired electron spin split (Zeeman interaction) and the energy difference ( ) increases with increasing B field according to the resonance condition given in the text. With the hyperfine interaction between electron spin and nitrogen nucleus (I=1), the two energy levels are split further into three transitions.

Figure 13) Generation of the disulfide-linked nitroxide side chain R1.

The methanethiosulfonate spin label (MTSSL) is reacted with cysteine residue on the target protein to label nitroxide spin specifically.

Figure 14) The principle of total internal reflection fluorescence microscopy (TIR) setup.

(a) A beam of external light source is directed through a prism of high refractive index, which abuts a lower refractive index medium of glass or aqueous solution. If the light directs into the prism at higher than the critical angle, the beam will be totally internally reflected at the interface. (b) The reflection develops an evanescent wave at the interface by the generation of an electromagnetic field that permeates ~200 nanometers into the lower refractive index space. The light in this field is sufficiently strong enough to excite the fluorophores within shallow depth, which results in extremely low-level background noise intensity.

Figure 15) smFRET description.

(a) FRET efficiency, E, as a function of inter-dye distance (R). Donor dye excited directly by incident laser either emits or transfer energy to acceptor dye depending upon its proximity. (b) A exemplary traces of donor/acceptor intensities (green/red, top) and calculated FRET efficiency (bottom).

CHAPTER 2: MULTIPLE CONFORMATIONS OF A SINGLE

SNAREPIN BETWEEN TWO NANODISC MEMBRANES

REVEAL DIVERSE PRE-FUSION STATES

A paper published in Biochemical Journal, 2014, vol 459: 95-102

Jaeil Shin, Xiaochu Lou, Dae-Hyuk Kweon, and Yeon-Kyun Shin

2.1 Abstract

SNAREpins must be formed between two membranes to allow vesicle fusion, a required process for neurotransmitter release. Although its post-fusion structure has been well characterized pre-fusion conformations have been elusive. We used single molecule FRET and

EPR to investigate the SNAREpin assembled between two nanodisc membranes. The SNAREpin shows at least three distinct dynamic states, which might represent pre-fusion intermediates.

While the N-terminal half above the conserved ionic layer maintains a robust helical bundle structure the membrane-proximal C-terminal half shows either high FRET representing a helical bundle (45%), low FRET reflecting a frayed conformation (39%), or mid FRET revealing an yet unidentified structure (16%). It is generally thought that SNAREpins are trapped at a partially zipped conformation in the pre-fusion state, and complete SNARE assembly happens concomitantly with membrane fusion. However, our results show that the complete SNARE complex can be formed without membrane fusion, which suggests that the complete SNAREpin 35

formation could precede membrane fusion, providing an ideal access to the fusion regulators such as complexins and synaptotagmin-1.

2.2 Introduction

Neurotransmitter release at the synapse requires fusion of neurotransmitter-laden vesicles with presynaptic plasma membrane. Synaptic membrane fusion is mediated by the soluble N- ethylmaleimide sensitive factor attachment protein receptor (SNARE) complex1-3. To allow membrane fusion to happen target membrane (t-) SNAREs syntaxin-1a and SNAP-25 interact with vesicle membrane (v-) SNARE synaptobrevin-2 (or VAMP-2) to form the trans-SNARE complex or SNAREpin that bridges two membranes. Cognate SNARE motifs, which are conserved heptad-repeat sequences of 60-70 residues, mediate SNAREpin formation.

Structural studies have shown that SNAREs tend to spontaneously assemble into a parallel four stranded coiled coil. However, SNARE structures were determined with isolated

SNARE motifs4-7 or in a cis state without the repulsion force between two apposed membranes8.

It is thus unclear whether the four stranded coiled coil represents a pre-fusion state or simply a post-fusion state. A biochemical analysis9 as well as studies with mechanical tweezers10,11 raised the possibility of a partially zipped state in which membrane-distal N-terminal region is assembled while the membrane-proximal C-terminal region is frayed. It is however unknown how many intermediate states exist and what their conformational details are. 36

In this work, we employed a nano-sized membrane called nanodisc12-14 and reconstituted a single SNARE complex between two separate nanodiscs. The investigation of the dye-labeled

SNARE complex with single molecule (sm) FRET and site-directed spin labeling EPR reveal that there are three distinct conformational states for VAMP-2; (1) a C-terminal-frayed state, (2) a fully assembled state, and (3) a structurally uncharacterized state that appears to be in between the two two major states. Furthermore, EPR shows that the structural break occurs around the ionic layer.

2.3 Results

2.3.1 Reconstitution of the SNAREpin into two nanodiscs.

To investigate the structure and the dynamics of the SNAREpin using smFRET and EPR, we reconstituted VAMP-2 into a population of nanodiscs (v-discs) and the 1:1 binary complex of syntaxin-1a and SNAP-25 into a separate population of nanodiscs (t-discs) using a standard assembly method15. Briefly, all components required for the reconstitution, phospholipids,

SNAREs, and the belt-forming apoA1 protein were dissolved in the 50 mM sodium cholate solution. The rapid removal of the detergent by Bio-Beads SM-2 prompted self-assembly of proteonanodiscs (Fig. 1a). Assembled nanodiscs were purified using gel-filtration chromatography and the size and the shape were examined with transmission electron microscopy (TEM) (Fig. 1b). We also confirmed, using photobleaching, that most nanodiscs carry a single SNARE protein (Fig. S3). 37

For the spectroscopic investigation of the SNAREpin the nanodisc structure is desired to be sufficiently robust to hold off the fusion activity of the SNARE complex. When t- and v-discs were mixed we observed no significant lipid mixing (Fig. 1c), indicating that nanodiscs maintain the structural integrity despite SNARE complex formation. There is however the possibility that the lipid flow is restricted while the two bilayers are connected as it is the case with ‘restricted hemifusion’16. In contrast, lipid mixing occurred when v-discs were mixed with t-SNARE reconstituted liposomes, consistent with previously reported results13.

2.3.2 The t- and v-SNAREs assemble into a parallel configuration between two nanodiscs, some with the frayed C-terminal end.

For smFRET we prepared an N-terminal FRET pair VAMP-2 Q33C and syntaxin-1a

I203C labeled with Cy3 and Cy5, respectively. We also prepared a C-terminal FRET pair, Cy3- labeled VAMP-2 A72C and Cy5-labeled syntaxin-1a V241C (Fig. 2a). We immobilized the t- disc using the biotin lipid-to-streptavidin conjugation on the polyethylene glycol-coated quartz surface in a flow cell (Fig. 2b). The v-disc solution was flown in to the flow cell to allow

SNAREpin formation between two nanodiscs. To promote efficient SNARE complex formation we used the recombinant peptide derived from the C-terminal region of VAMP-2 (aa. 49-96), which is known to make the binary complex more compatible for VAMP-217.

When the SNAREpin with the N-terminal FRET pair I203C/Q33C (NN) was examined with total internal reflection (TIR) we found a single major FRET distribution picked at FRET 38

efficiency E∼0.8. Meanwhile, when the SNAREpin with the C-terminal FRET pair

V241C/A72C (CC) was examined we observed two distinct major FRET distributions, one centered at E∼0.8 (45.4±6.0%), the other near E∼0.1 (38.9±4.5%), and one minor distribution at

E∼0.45 (15.7±4.8%). However, when the out-of-register FRET pair I203C/A72C or

V241/Q33C (NC or CN) was examined we detected mostly low FRET and the distributions were centered at E∼0.1 or 0.35 with little population at high FRET. Together, the results show that v- and t-SNAREs assemble into a parallel configuration and the antiparallel orientation is negligible18,19.

Given that the SNAREpin is all parallel, it is interesting that the NN pair has mostly the high FRET population (E∼0.8), while the CC pair has the two major states at high (E~0.8) and low FRET (E∼0.1). Thus, the results imply that the CC pairs have one ordered and the other disordered and dynamic states (Fig. 2c). Although minor (16%) the distinct population centered at E∼0.45 is interesting. This state appears to be neither fully extended nor fully structured, which might indicate a possible new intermediate state.

2.3.3 EPR confirms that both the fully assembled four-helix bundle and the half-zipped complex represent trans-SNAREpin conformations.

To further characterize the structure of the SNAREpin with EPR, we prepared eleven single cysteine mutants of VAMP-2 around the conserved central ionic layer (Fig. S4a), labeled them with a nitroxide spin label MTSSL (S-(2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrol-3-yl) 39

methyl methanesulfonothioate spin label), and reconstituted them into nanodiscs (S-v-nanodiscs).

Proteonanodiscs carrying the unlabeled binary complex (t-nanodiscs) contain 6X His-tagged apoA1 and it was used to pull down the partner S-v-nanodisc to purify the spin labeled trans-

SNAREpins (Fig. 3a). The fusion activity of spin-labeled VAMP-2 was analyzed by the in vitro proteoliposome lipid mixing assay and we confirmed that all spin labeled VAMP-2 maintains the wild-type fusion activity (Fig. S4b).

Each S-v-nanodisc sample generated a narrow EPR spectrum reflecting unstructured conformation of VAMP-2 prior to complex formation with t-nanodisc20 (Fig. 3b, uncomplexed).

Meanwhile, when the cis complex, which was prepared by premixing spin-labeled VAMP-2 with t-SNAREs, was reconstituted into nanodiscs, EPR spectra showed significant line-broadening due to formation of a stable four-helix bundle (Fig. 3b, cis).

EPR spectra of spin labeled trans-SNAREpins showed broad line shapes for all N- terminal positions before the conserved ionic layer (residue R56). However, the EPR spectra for all C-terminal positions beyond the ionic layer exhibited two spectral components, one broad and the other sharp (Fig. 3b, trans). The broad spectral components may represent a well-structured coiled coil while the sharp components reflect the unstructured and freely moving conformation of VAMP-2.

A spectral decomposition analysis21,22 was used to separate two spectral components (Fig.

3b, decomposition) and the ratio of two spectral components was calculated. The ratio of unstructured component was low in the N-terminal region but reaches ∼30% for C-terminal positions after position 58. The transition from the structured to unstructured conformation 40

happens between residues 50 and 54 (Fig. 3c). Therefore, our EPR results show that for trans-

SNAREpin the N-terminal half of VAMP-2 above the conserved ionic layer is a well-structured coiled coil while the C-terminal half after the ionic layer is either fully structured (70%) or unstructured (30%), qualitatively consistent with the results from smFRET. We note that smFRET showed 39% the unstructured population, which was somewhat higher than the EPR estimate.

2.3.4 The real-time analysis of single SNAREpins reveals a subpopulation with highly fluctuating dynamics.

Both snapshot smFRET and static EPR analysis revealed three distinct conformational states of trans-SNAREpins; a fully assembled helix bundle, the half-zipped SNARE complex, and a structurally unknown mid FRET state. Next, we investigated the real-time dynamics of single FRET pairs to identify yet unidentified or transient dynamic substates. For NN FRET pairs, as expected, we observed stable high FRET time traces with little fluctuation (Fig. 4a).

The high stability of the FRET time traces indicates that the N-terminal region of the trans-

SNARE complex maintains a robust helical bundle.

In contrast, the analysis of single CC FRET pairs reveals a few different dynamic facets;

(1) time traces maintaining stable high FRET (E~0.8, 40.3%, Fig. 4b), (2) those maintaining stable low FRET (E~0.1, 39.7%, Fig. 4c), (3) those making transition between high and low

FRET (3.4%, Fig. 4d), (4) those maintaining stable mid FRET (E~0.45, 14.5%, Fig. 4e), and 41

finally, (5) traces fluctuating between multiple FRET states (2.1%, Fig. 4f). Stable high FRET may correspond to a fully assembled helical bundle while stable low FRET may represent fully unstructured conformation. Fluctuating traces may reflect the equilibrium conformational change between multiple states. Stable mid FRET traces are consistent with the structural intermediate state, which can transition to either the high or the low FRET state (Fig. S5). Thus, the real time analysis of smFRET pair confirmed the three conformational states identified with the static smFRET analysis without adding further intermediate state and revealed equilibrium fluctuations between three conformational states for some of them.

2.4 Discussion

For SNAREpins trapped between two nanodiscs we identified two major conformational states and one minor state; a fully-assembled helical bundle (45%), a half-zipped complex in which the C-terminal half of VAMP-2 is frayed as a random coil (39%), and a structurally unknown mid FRET state (16%). EPR revealed that for VAMP-2, the structural transition from a

α-helix to a random coil occurs right near the conserved ionic layer located at residue R56. It is likely that t-SNAREs syntaxin-1a and SNAP-25 maintains a complete three-stranded coiled coil in the half zipped complex10,11.

The real-time analysis of smFRET showed that the two major conformational states are mostly very stable with few transitions between the two within our experimental time window (2 min). The analysis however revealed a small subset of SNAREpins for which relatively fast transitions between multiple FRET states are allowed. How is it possible that a small subset 42

exhibits fast fluctuations despite the majority do not? To understand such different dynamic behaviors we must take into account the distance between two nanodiscs holding the SNAREpin, which sets the geometric constraints to the SNARE conformation. At sufficiently large distances between t- and v-nanodiscs the random coil conformation would be preferred for the C-terminal half of VAMP-2 and rezipping might not be allowed. For closely apposed nanodiscs, however, a fully assembled SNARE core would be energetically stable. We speculate that the fluctuation between two conformations is allowed only in a narrow window of reasonable inter-membrane distances.

Based on the experimental results, we propose a three-state model for trans-SNAREpins in the inter-membrane gap (Fig. 5). The N-terminal half above the ionic layer maintains a robust parallel four stranded coiled coil for all intermediates. The C-terminal half below the ionic layer, however, may have three different states: First, the C-terminal half of VAMP-2 is unstructured in a large inter-membrane gap (Fig. 5a). Second, it is structurally immobilized as it was indicated in

EPR spectra but not fully assembled to the four helix bundle yet (Fig. 5b). We speculate that the structure might be some kind of the out-of-register coiled coil that has been observed in other coiled coils23. Third, it remains as a fully assembled coiled coil between a closely apposed two membranes (Fig. 5c). With EPR we found that about 30% of SNAREpins hold the fully unstructured conformation while smFRET shows 39% are unstructured. The discrepancy might arise from the differences in nanodisc concentrations in two experiments. While we used only a

100 pmole range in sm FRET we used tens of mM for EPR that would create a clouded environment for nanodiscs, which in turn might force the inter-membrane gap between two nanodiscs to decrease. 43

The fact that the fully assembled coiled coil is the major conformation of the SNAREpin is surprising because it is generally thought that the four-stranded coiled coil is the ‘post-fusion’ conformation and in the ‘pre-fusion’ state it might have a partially assembled structure like one shown in Fig. 5a. In a Ca2+-triggered exocytosis we expect that the fusion machinery enters into the ‘primed state’ that is ready to fuse upon the rise of Ca2+ signal24. Therefore, we wonder if the fully assembled coiled coil represents the true primed state. The major Ca2+-sensor synaptotagmin-1 then interacts with the juxtramembrane region and/or the membrane to mediate membrane fusion25. We speculate that a priming agent such as complexin might play a role in holding the SNAREpins in the fully assembled coiled coil, instead of a partially assembled state26-28.

In summary, we characterized the structure and the dynamics of SNAREpin sandwiched between two nanodisc using smFRET and EPR. We found that while the N-terminal half maintains a robust coiled coil structure the C-terminal half have multiple dynamics states including the fully zipped, the half zipped, the partially zipped, and fluctuating states.

2.5 Materials and Methods

2.5.1 Protein expression, purification, and fluorophore- or spin-labeling.

All recombinant proteins were expressed in Escherichia coli BL21 (DE3) grown in LB medium at 37 °C to an A600 of 0.7. The protein expression was induced with 0.2 mM isopropyl-

β-D-thiogalactopyranoside (IPTG). Cells were then grown for another 5 hours at 25 °C for 44

soluble proteins, SNAP-25 and apoA1 or 16 hours at 16 °C for proteins with the transmembrane domain, syntaxin-1a and VAMP-2, respectively. Pellets were resuspended in phosphate buffered saline (PBS) or PBS with 0.3% triton (PBST) with the protease inhibitor, 4-(2-Ainoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF) and additional dithiothreitol (DTT) for single cysteine mutants of syntaxin-1a or VAMP-2. Cells were lysed by sonication, after which lysates were centrifuged at 13,000 rpm for 30 min at 4 °C to remove insoluble fractions. Supernatants containing glutathione S-transferase (GST)-fused recombinant proteins, syntaxin-1a, SNAP-25, and VAMP-2 were then incubated with glutathione sepharose beads (GE healthcare) while 6X

His-tagged SNAP-25 and apoA1 were incubated with Ni-NTA agarose beads (Qiagen) for 1 hour at 4 °C. After washing with PBS or PBST, GST-fusion proteins were eluted with thrombin

(0.02 unit/μl, Sigma) in PBS or PBS with 0.8% octyl β-D-glucopyranoside (PBS-OG) for membrane proteins and 6X His-tagged proteins were eluted with 200 mM imidazole (Sigma) in

PBS. The elution buffer for single cysteine mutant proteins contained 2 mM DTT. For pull down of the trans-SNAREpin sandwiched between two nanodiscs, His-tag-free apoA1 was obtained with thrombin cleavage of His-tag (Fig. S1). For the smFRET study of the trans-SNAREpin, single cysteine mutants of syntaxin-1a (I203C or V241C) and VAMP-2 (Q33C or A72C) were desalted with PD 10 column (GE healthcare) to eliminate free DTT in solution, then incubated with 10x molar excess of maleimide-derivative fluorophore, Cy5 or Cy3, respectively for overnight at 4 °C. Unreacted free fluorescence labels were removed by the PD 10 column and the labeling efficiency of each SNARE protein was measured with spectrophotometry

(Beckman). Extinction coefficients of Cy5 (250 000 M-1 cm-1 at 650 nm) and Cy3 (150 000 M-1 cm-1 at 552 nm) were used to calculate the concentration of the fluorophore. The detergent 45

compatible Lowry assay (DC assay, Bio Rad) was recruited to determine the concentration of

SNAREs. The labeling efficiency for each protein was 46% and 40% for syntaxin-1a I203C and

V241C and 55% and 62% for VAMP2-2 Q33C and A72C, respectively. For spin labeling with

MTSSL (Enzo Life Sciences) of VAMP-2 cysteine mutants, 10x molar excess of MTSSL was added to GST-VAMP-2 cysteine mutants on glutathione sepharose beads right after rapid washing with DTT free PBST before overnight incubation at 4 °C. Free MTSSL was washed out before eluting spin-labeled VAMP-2 with thrombin.

2.5.2 Reconstitution and purification of SNARE-incorporated nanodiscs.

To generate homogeneously-sized nanodiscs, phospholipid mixtures of 1-palmitoyl-2- oleoyl-sn-glycero-3-phosphocholine (POPC) and 1,2-dioleoyl-sn-glycero-3-phospho-L-serine

(DOPS) (Avanti Polar Lipids) with the molar ratio of 85:15 were dried and then resuspended in

T150 buffer (10 mM Tris-HCl, 150 mM NaCl, pH 7.4), which included 50 mM sodium cholate.

The SNARE protein and apoA1 protein were then added. The molar ratio for lipids:SNARE:apoA1 was 160:0.5:2. The self-assembly of SNARE-incorporated nanodiscs was initiated by a rapid removal of sodium cholate by treating the sample with SM-2 Bio-Beads. The t-SNARE or VAMP-2 incorporated nanodiscs were then purified through gel-filtration using

SuperdexTM200 GL 10/30 column (Amersham Biosciences) (Fig. S2)

2.5.3 Lipid mixing between v-discs and either t-discs or t-liposomes.

To determine whether v-discs fuse with either t-discs or t-liposomes, 1.5 molar % of each

1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl) (NBD-PE) and 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl)

(rhodamine-PE) (Avanti Polar Lipids) were included in v-discs. Both t-discs and t-liposomes were prepared with unlabeled phospholipids to dilute NBD-PE and rhodamine-PE when fusion occurred. The molar ratios of lipid:SNARE for nanodiscs and t-liposomes were 160:0.5 and

200:1, respectively. The v-discs were then mixed with 9x lipid molar excess of either t-discs or t- liposomes at 37 °C in a fusion buffer (10 mM HEPES, 100 mM KCl, pH 7.4). The reaction solution contained total 0.5 mM lipids. The NBD fluorescence intensity was measured at excitation and emission wavelengths of 465 and 530 nm, respectively with a fluorescence spectrophotometer (Cary Eclipse model, Varian). The maximum fluorescence intensity was obtained by adding 0.1% reduced Triton X-100 (Sigma).

2.5.4 FRET measurements of trans-SNARE complex with TIR.

To immobilize Cy5-labeled t-discs on the biotinylated PEG-coated quartz surface for smFRET, 0.5% of 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-biotinyl (biotinyl-PE,

Laysan Bio Inc.) was included into the phospholipid mixture for t-disc assembly. The quartz slide was assembled into a flow chamber and then covered with 0.2 mg/ml streptavidin. t-discs

(100~200 nM) were flown into the chamber and incubated with the 2 uM 49-96 peptide derived 47

from VAMP-2. 5x molar excess of Cy3-labeled v-discs were then added and incubated for 25 min to allow SNAREpin formation between two nanodiscs afterward. All TIR experiments were performed at room temperature (~25 °C) in the presence of oxygen scavenger system (0.4 %

(w/v) glucose (Sigma), 4 mM Trolox (Calbiochem), 1 mg/ml glucose oxidase (Sigma), 0.04 mg/ml catalase (Calbiochem)) in T150 buffer. smFRET measurements were carried out on a prism type TIR setup, which is based on inverted microscope (Eclipse Ti, Nikon) with the 100 ms or 200ms time resolution. Each Cy3 and Cy5 fluorescence signal from the chamber were collected by a water immersion lens (PlanApo VC 60x, Nikon) and separated by a dichroic mirror (T660lpxr, chroma), which had a threshold at 635 nm wavelength. Two separated signals were then transferred to separated images on an electron multiplying charged-coupled device camera (iXon3 DU897D, Andor Technology) as donor and acceptor signals, respectively. FRET efficiencies (E) were obtained by

( )

where FA and FD are the acceptor and donor fluorescence intensities.

2.5.5 Purification and measurement of spin labeled trans-SNARE complex in EPR spectroscopy.

We used the pull down method to purify spin labeled trans-SNAREpin in between nanodiscs. Spin-labeled VAMP-2 was incorporated into the His-tag-free v-nanodisc and the unlabeled t-SNARE was incorporated into the 6X His-tagged t-nanodisc, respectively. The t- 48

nanodiscs were attached to Ni-NTA sepharose beads and equal molar quantities of S-v-nanodiscs were then added and incubated for overnight at 4 °C to form spin labeled trans-SNAREpin.

After washing out of unbound free S-v-nanodiscs spin labeled trans-SNAREpins were eluted from the beads by imidazole. To obtain spin labeled cis–SNARE complex in the nanodisc, the mixture of syntaxin-1a, 6X His-tagged SNAP-25, and spin-labeled VAMP-2 were incubated with molar ratio of 1:2:1. Unbound syntaxin-1a and spin-labeled VAMP-2 were washed out using the Ni-NTA column and the complex was then incorporated into the nanodisc. Each spin- labeled VAMP-2 was reconstituted into nanodiscs for the EPR measurement. EPR spectra were collected in the first-derivative mode with 1 mW microwave power using the Bruker ESP 300 spectrometer equipped with a loop-gap resonator. The modulation amplitude was set at no greater than one-fourth of the line width. All EPR measurements were performed at room temperature.

2.6 References

1. Sollner, T. et al. SNAP receptors implicated in vesicle targeting and fusion. Nature 362, 318-324 (1993).

2. Weber, T. et al. SNAREpins: Minimal Machinery for Membrane Fusion. Cell 92, 759- 772 (1998).

3. Jahn, R. & Scheller, R.H. SNAREs [mdash] engines for membrane fusion. Nat Rev Mol Cell Biol 7, 631-643 (2006).

4. Sutton, R.B., Fasshauer, D., Jahn, R. & Brunger, A.T. Crystal structure of a SNARE complex involved in synaptic exocytosis at 2.4 A resolution. Nature 395, 347-53 (1998).

5. Poirier, M.A. et al. The synaptic SNARE complex is a parallel four-stranded helical bundle. Nat Struct Mol Biol 5, 765-769 (1998).

6. Antonin, W., Fasshauer, D., Becker, S., Jahn, R. & Schneider, T.R. Crystal structure of the endosomal SNARE complex reveals common structural principles of all SNAREs. Nat Struct Mol Biol 9, 107-111 (2002).

7. Strop, P., Kaiser, S.E., Vrljic, M. & Brunger, A.T. The Structure of the Yeast Plasma Membrane SNARE Complex Reveals Destabilizing Water-filled Cavities. Journal of Biological Chemistry 283, 1113-1119 (2008).

8. Stein, A., Weber, G., Wahl, M.C. & Jahn, R. Helical extension of the neuronal SNARE complex into the membrane. Nature 460, 525-528 (2009).

9. Sorensen, J.B. et al. Sequential N- to C-terminal SNARE complex assembly drives priming and fusion of secretory vesicles. EMBO J 25, 955-966 (2006).

10. Gao, Y. et al. Single Reconstituted Neuronal SNARE Complexes Zipper in Three Distinct Stages. Science 337, 1340-1343 (2012).

11. Min, D. et al. Mechanical unzipping and rezipping of a single SNARE complex reveals hysteresis as a force-generating mechanism. Nat Commun 4, 1705 (2013).

12. Bayburt, T.H., Carlson, J.W. & Sligar, S.G. Reconstitution and Imaging of a Membrane Protein in a Nanometer-Size Phospholipid Bilayer. Journal of Structural Biology 123, 37- 44 (1998).

13. Shi, L. et al. SNARE Proteins: One to Fuse and Three to Keep the Nascent Fusion Pore Open. Science 335, 1355-1359 (2012). 50

14. Krishnakumar, Shyam S. et al. Conformational Dynamics of Calcium-Triggered Activation of Fusion by Synaptotagmin. Biophysical Journal 105, 2507-2516 (2013).

15. Bayburt, T.H. & Sligar, S.G. Membrane protein assembly into Nanodiscs. FEBS Lett 584, 1721-7 (2009).

16. Chernomordik, L.V., Frolov, V.A., Leikina, E., Bronk, P. & Zimmerberg, J. The Pathway of Membrane Fusion Catalyzed by Influenza Hemagglutinin: Restriction of Lipids, Hemifusion, and Lipidic Fusion Pore Formation. The Journal of Cell Biology 140, 1369- 1382 (1998).

17. Pobbati, A.V., Stein, A. & Fasshauer, D. N- to C-Terminal SNARE Complex Assembly Promotes Rapid Membrane Fusion. Science 313, 673-676 (2006).

18. Bowen, M.E., Weninger, K., Brunger, A.T. & Chu, S. Single Molecule Observation of Liposome-Bilayer Fusion Thermally Induced by Soluble N-Ethyl Maleimide Sensitive- Factor Attachment Protein Receptors (SNAREs). Biophysical Journal 87, 3569-3584 (2004).

19. Weninger, K., Bowen, M.E., Choi, U.B., Chu, S. & Brunger, A.T. Accessory Proteins Stabilize the Acceptor Complex for Synaptobrevin, the 1:1 Syntaxin/SNAP-25 Complex. Structure 16, 308-320 (2008).

20. Brewer, K.D., Li, W., Horne, B.E. & Rizo, J. Reluctance to membrane binding enables accessibility of the synaptobrevin SNARE motif for SNARE complex formation. Proceedings of the National Academy of Sciences 108, 12723-12728 (2011).

21. Thorgeirsson, T.E., Russell, C.J., King, D.S. & Shin, Y.-K. Direct Determination of the Membrane Affinities of Individual Amino Acids†. Biochemistry 35, 1803-1809 (1996).

22. Zhang, Y., Su, Z., Zhang, F., Chen, Y. & Shin, Y.-K. A Partially Zipped SNARE Complex Stabilized by the Membrane. Journal of Biological Chemistry 280, 15595- 15600 (2005).

23. Xi, Z., Gao, Y., Sirinakis, G., Guo, H. & Zhang, Y. Single-molecule observation of helix staggering, sliding, and coiled coil misfolding. Proceedings of the National Academy of Sciences 109, 5711-5716 (2012).

24. Südhof, T.C. THE SYNAPTIC VESICLE CYCLE. Annual Review of Neuroscience 27, 509-547 (2004).

25. Honigmann, A. et al. Phosphatidylinositol 4,5-bisphosphate clusters act as molecular beacons for vesicle recruitment. Nat Struct Mol Biol 20, 679-686 (2013). 51

26. Schaub, J.R., Lu, X., Doneske, B., Shin, Y.-K. & McNew, J.A. Hemifusion arrest by complexin is relieved by Ca2+-synaptotagmin I. Nat Struct Mol Biol 13, 748-750 (2006).

27. Tang, J. et al. A Complexin/Synaptotagmin 1 Switch Controls Fast Synaptic Vesicle Exocytosis. Cell 126, 1175-1187 (2006).

28. Giraudo, C.G. et al. Alternative Zippering as an On-Off Switch for SNARE-Mediated Fusion. Science 323, 512-516 (2009).

2.7 Figures and Legends

Figure 1) Preparation of SNARE-incorporated proteonanodisc.

(a) Self-assembly of SNARE-incorporated nanodiscs. By rapid removal of sodium cholate (orange) from the mixture of the SNARE protein (blue/yellow), phospholipids (grey), and apoA1 (green) resulted in assembly of the proteonanodisc carrying the SNARE protein. (b) Purification of proteonanodiscs. SNARE-incorporated proteonanodiscs were collected from the earlier part of the elution volume (blue dashed square) than empty-nanodiscs (red spectrum) through the gel-filtration column and characterized by transmission electron microscopy (bottom). (c) Lipid mixing assays. v-nanodiscs containing NBD and rhodamine were mixed with unlabeled either t-liposomes (black trace) or t-nanodiscs (red trace).

Figure 2) Distributions of the FRET efficiency at the N- and C-terminal regions of trans-

SNAREpin.

(a) Fluorescent dye-labeled positions on VAMP-2 and syntaxin-1a, respectively. Q33C or A72C of VAMP-2 (blue ribbon) were labeled with Cy3 (light red saw-tooth circle) while I203C 54

or V241 of syntaxin-1a (red ribbon) were labeled with Cy5 (light blue saw-tooth circle). The α- carbon-to-α-carbon distance between two cysteine residues were indicated. The flexible linker region (gray) and transmembrane domain (yellow) are also included in the structure. (b) TIR setup for the FRET detection on the SNAREpin sandwiched between two nanodiscs. t-nanodisc carrying Cy5-labeled syntaxin-1a was immobilized on the quartz surface through biotin (dark red square) -streptavidin (purple, cross shape) conjugation and allowed to interact with v- nanodisc carrying Cy3-laveled VAMP-2. (c) Distributions of the FRET efficiencies for the NN, CC, and out-of-register NC and CN FRET pairs. The Gaussian fitting lines also are included (black line).

Figure 3) EPR analysis of spin labeled trans-SNAREpins.

(a) Purification of nitroxide-labeled trans-SNAREpins sandwiched between two nanodiscs using pull-down method with imidazole elution (top). TEM images of sandwiched nanodiscs mediated by trans-SNAREpins (bottom, yellow dashed circle). (b) The first derivative EPR spectra of each eleven spin-labeled VAMP-2 in specific conformations. S-v-nanodisc only (uncomplexed), pre-assembled cis-SNAREpin in nanodisc (cis), purified spin labeled trans- SNAREpin in nanodiscs (trans), and complexed fraction in trans-SNAREpin resulted from spectral decomposition (decomposition, red). (c) Unstructured fraction of each eleven VAMP-2 in cis-SNAREpin (black closed circle and line) and trans- (red closed circle and line) conformation with the breaking point at right near the ionic layer. 56

Figure 4) Real time traces of FRET in trans-SNARE complex conformation.

All traces for donor and acceptor signals under donor-exciting 532 nm laser source, unless otherwise indicated, were in the top panel and the calculated FRET traces are in the bottom panel. (a) The NN FRET pair (Q33Cy3 of VAMP-2 and I203Cy5 of syntaxin-1a) in trans-SNAREpin shows stable high FRET (blue open circle and line) followed by acceptor photobleaching (red open circle and line) and the recovery of the donor fluorescence (green open circle and line). The CC FRET pair (A72Cy3 of VAMP-2 and V241Cy5) show 5 different time traces; b) stable high FRET, c) stable low FRET and co-localized two fluorophores were confirmed by switching laser sources between acceptor-exciting 635 nm (red rectangle) and donor-exciting 532 nm (green rectangle), d) transition between high and low, E) stable mid FRET, and F) relatively fast fluctuation between multiple FRET states.

Figure 5) A proposed model of multiple conformations for the SNAREpin.

When a synaptic vesicle with VAMP-2 (blue) is recruited into the t-SNARE complex of SNAP-25 (green) and syntaxin-1a (red) on the presynaptic plasma membranes, docking is initiated by the N-terminal half interaction between VAMP-2 (blue, rod shape) and t-SNAREs while C-terminal half of VAMP-2 (blue line) is remained as unbound (a). As the vesicle gets closer to the membrane within a certain distance, the C-terminal VAMP-2 forms hypothetical intermediate structure (distorted blue line) (b). Afterwards, in close enough contact between two membranes, SNARE proteins form fully assembled four-helical bundle structure and wait for the action of synaptotagmin-1 and Ca2+ for Ca2+-triggered vesicle fusion (c).

2.8 Supporting Information

Content: Supplementary Figure and Legends S1-S5

Figure S1) A Construct of apoA1 protein.

The expressed apoA1 construct includes 6X-His tag (light blue) at the N-terminus followed by thrombin cleavage site (orange), and membrane scaffolding apoA1 protein (green). 59

Figure S2) Confirmation of v- or t-nanodisc assembly.

Separately assembled v-, t-, and empty-nanodiscs were purified by the Superdex 200 10/300 GL column. Either v- (blue trace) or t-discs (green trace) came earlier than empty-discs (red trace) or apoA1 protein (black trace). Purified v- and t-discs were then further analyzed by SDS-PAGE (inset) to confirm the incorporation of SNAREs in nanodiscs.

Figure S3) Distribution of the number of SNARE proteins embedded in a single nanodisc obtained from counting photobleaching steps. 61

(a and b) Cy3-labeled VAMP2 (Q33C or A72C) v-discs or (c and d) Cy5-labeled syntaxin-1a (I203C or V241C) t-discs were immobilized on the PEG-coated quartz surface in T150 buffer. Direct excitation of each fluorophore with 532 nm or 635 nm laser resulted in single or multiple steps of photobleaching (top). Population was plotted versus the number of photobleaching steps (bottom).

Figure S4) Lipid mixing activities of spin-labeled VAMP-2 mutants.

(a) Each residue of the single cysteine mutation in VAMP-2 is presented (yellow stick) in cis-SNARE complex structure. Interface residues of VAMP-2 for coiled-coil formation were not selected to minimize the effect of spin labeling. For clarity, all cysteine mutants are indicated in a single VAMP-2 (blue ribbon) while syntaxin-1a (red) and SNAP-25 (green) are pictured as cartoon helices. (b) Each spin-labeled or wild-type VAMP-2 was reconstituted into NBD- and rhodamine- containing liposomes (v-liposomes), whereas t-SNARE was reconstituted into unlabeled liposome (t-liposomes). v-liposomes and t-liposomes were mixed in fusion buffer at 63

37 °C to allow lipid mixing. The normalized NBD fluorescence intensity of each spin-labeled v- liposome was compared to that of wild-type VAMP-2.

Figure S5) Transition from mid FRET to either high or low FRET in CC pair.

All traces of donor or acceptor signals were obtained under donor-exciting 532 nm laser source (green rectangle) or acceptor-exciting 635 nm laser source (red rectangle) (top panel) and the FRET efficiencies were calculated as mentioned in the text (bottom panel). (A) The single transition from the intermediate mid FRET to high FRET state. (B) The single transition from the intermediate mid FRET to low FRET state. 65

CHAPTER 3: SYNAPTOTAGMIN 1 IS AN ANTAGONIST FOR

MUNC18-1 IN SNARE-ZIPPERING

Xiaochu Lou*, Jaeil Shin*, Yoosoo Yang*, Jaewook Kim, and Yeon-Kyun Shin

Submitted to Nature Structure and Molecular Biology in 2014

(*: equally contributed to this work)

3.1 Abstract

In neuroexocytosis, SNAREs and Munc18-1 may consist of the minimal membrane fusion machinery. Consistent with this notion, we observed, using single molecule fluorescence assays, that Munc18-1 stimulates SNARE zippering and SNARE-dependent lipid mixing in the absence of a major Ca2+-sensor synaptotagmin 1(syt1), providing the structural basis for the conserved function of SM proteins in exocytosis. However, when full-length syt1 is present no enhancement of SNARE zippering and no acceleration of Ca2+-triggered content mixing by

Munc18-1 are observed. Thus, our results show that syt1 acts as an antagonist for Munc18-1 in

SNAREs zippering and fusion pore opening. Although the Sec1/Munc18 family may serve as part of the fusion machinery in other exocytotic pathways, Munc18-1 may have evolved to play a different role such as regulating syntaxin 1a in neuroexocytosis.

3.2 Introduction

Essential to the functional connectivity in the central nerve system is neurotransmitter release at synapses which requires fusion of vesicles to the presynaptic plasma membrane.

Vesicle fusion is an energetically costly process because it needs to overcome the energy barrier for merging two stable membranes to a single bilayer1,2.

The required fusion energy would have to be provided by the protein machinery. It is thought that SNAREs and Munc18-1 constitute the minimal fusion machinery3,4. The SNARE complex formed between vesicle (v-) SNARE and target plasma membrane (t-) SNAREs is considered the core of the fusion machine3-11. Further, the binding of Munc18-1 to the SNARE complex would provide an additional energetic boost in driving vesicle fusion12-14.

There is ample evidence that SNAREs belong to the minimal fusion machinery.

Treatment of the presynapse with the clostridial toxins, which specifically cleave SNARE proteins, abolishes neurotransmitter release completely15-18. Furthermore, proteoliposomes reconstituted with SNAREs only support lipid mixing, demonstrating that SNAREs alone can drive membrane fusion8,19-24. Similarly, it was shown that Munc18-1 accelerates SNARE- mediated proteoliposome fusion, supporting the notion that Munc18-1 is part of the minimal fusion machine12,25.

Meanwhile, Munc18-1 appears to have another important function to regulate the

SNARE assembly. Munc18-1 binds to the Habc domain of t-SNARE syntaxin 1a and prevents sytaxin 1a from the premature binding to another t-SNARE SNAP-25 that might lead to a nonproductive t-SNARE complex26-30. 67

Although the Munc18-1 function to protect syntaxin 1a is supported by a variety of evidence its role as part of fusion machine is debatable. The proteoliposome fusion assay12,25, on which this proposition relies heavily, did not include synaptotagmin 1 (syt1), a major Ca2+- sensor for synaptic vesicle fusion31-33. Syt1 is a vesicular protein, consisting of tandem Ca2+- binding C2 domains and a transmembrane helix34. Syt1 interacts with the core SNARE complex as well as phospholipids35-39. Provided that syt1 cooperates with the SNARE complex intimately during the moment of fusion it is difficult to envision how Munc18-1 might gain access to the core complex simultaneously40.

In this work, we investigated, using single molecule (sm) FRET, the conformational changes of a trans-SNARE complex (or SNAREpin) assembled between two nanodisc membranes41 induced by Munc18-1. We also studied the effect of Munc18-1 on SNARE- dependent proteoliposome lipid mixing and on the Ca2+-triggered fusion pore opening in well- defined in vitro settings39,42 to dissect the Munc18-1 function in the presence of syt1. Our results show that although Munc18-1 has the capacity to stimulate SNARE complex formation and

SNARE-dependent lipid mixing syt1 largely negates such positive effects on membrane fusion, suggesting that syt1 acts an antagonist for Munc18-1.

3.3 Results

3.3.1 Munc18-1 promotes SNARE zippering without syt1, but not in the presence of syt1.

It has been previously shown that Munc18-1 can bind to the SNARE four helix bundle but it has not yet been demonstrated that such binding actually stimulate SNARE zippering43. To 68

investigate effects of Munc18-1 on the conformation of the trans-SNAREpin with smFRET we prepared an N-terminal FRET pair VAMP2 Q33C and syntaxin 1a I203C labeled with the donor dye Cy3 and the acceptor dye Cy5 (NN), respectively. We also prepared a C-terminal FRET pair,

Cy3-labeled VAMP2 A72C and Cy5-labeled syntaxin 1a V241C (CC) (Fig. 1a). For the detection of FRET with total internal reflection (TIR) the t-SNARE-reconstituted nanodiscs (t- discs) were tethered on the imaging surface using the streptavidin-to-biotin lipid conjugation.

The v-SNARE reconstituted nanodiscs (v-discs) were then flown into the flow cell to allow docking and trans-SNAREpin formation (Fig. 1b)41.

For the SNAREpin with the NN FRET pair we observed a dominant FRET distribution peaked at the FRET efficiency E∼0.8 (Fig. 1c, left), indicating that the helical structure is robust at the N-terminal region. We observed a small low FRET population (~8%) that might reflect non-specific binding to the surface (Fig. 1c, left). The possibility of antiparallel SNARE assembly44 could be ruled out since out-of-register combinations (NC and CN) dominantly populated mid-FRET (E~0.4) (Fig. S1). In contrast, the SNAREpin with the CC FRET pair gave two major low and high FRET distributions with some population in the mid-FRET (Fig. 1c, upper-right, see the ref41 for the discussion of the mid-FRET population). The low FRET peak might reflect a half-zipped SNAREpin, while the high and mid FRET peaks represent a fully- zipped SNAREpin41. When Mun18-1 (1 µM) was added, for the CC FRET pair we observed the shift of the FRET distribution from low to high, showing that Munc18-1 promoted formation of fully-zipped SNAREpin (Fig. 1c, lower-right, Fig. 1e, and Fig. S2).

Now, to find out if such promotion of SNARE zippering by Munc18-1 still stands in the presence of syt1 we incorporated syt1 to v-disc in the molar ratio of 1:1 to VAMP2. With syt1 69

we still observe a dominant high FRET distribution for the NN FRET pair (Fig. 1d, upper- left), identical to that observed in the absence of syt1. Interestingly, however, we did not observe any change in the distribution of the CC FRET population at all (Fig. 1d right, and Fig. 1f). Thus, the results suggest that syt1 works as an antagonist to Munc18-1 for SNAREpin binding, which abrogates the enhancement of SNARE zippering by Munc18-1.

3.3.2 Syt1 and Munc18-1 are mutually antagonistic in SNARE-dependent lipid mixing.

It has been consistently shown, using the in vitro bulk or single vesicle lipid mixing assay, that Munc18-1 enhances SNARE-dependent lipid mixing (Fig. S3a)12,14. We ask if such stimulation of lipid mixing still exists in the presence of syt1.

To measure lipid mixing we immobilized vesicles carrying t-SNARE on the imaging surface (t-vesicles) and vesicles carrying v-SNARE (v-vesicles) was added to the flow cell to allow vesicle docking (Fig. 2a). The lipid dyes DiI and DiD were separately incorporated into t- and v-vesicles, respectively, to detect FRET due to lipid mixing (Fig. 2a). Without Munc18-1 lipid mixing was slow: only around 20% of docked vesicles showed lipid mixing after 30 min incubation (Fig. 2b, f). However, when Munc18-1 was present lipid mixing was accelerated significantly: 50% of docked vesicles were lipid-mixed (Fig. 2c, f), consistent with previously reported results14.

When syt1 was incorporated into the v-vesicles in the molar ratio of 1:1 to VAMP2 we observed great enhancement of lipid mixing. After 30 min incubation nearly all docked vesicles were lipid-mixed (Fig. 2d, f). The stimulation of SNARE-dependent lipid mixing by syt1 was 70

previously reported and not surprising39. What is surprising, though, is, when we added Munc18-

1 the stimulation of SNARE-dependent lipid mixing by syt1 is much reduced (Fig. 2e). We observed only about 60% of docked vesicles show lipid mixing (Fig. 2e, f), suggesting that

Munc18-1 is antagonistic to syt1 in SNARE-dependent lipid mixing.

3.3.3 Munc18-1 has little effect on Ca2+-triggered content mixing.

Although SNAREs alone as well as SNAREs together with syt1 can support lipid mixing, they are not effective in driving content mixing or fusion pore formation unless Ca2+ is present39,45. Therefore, in vitro Ca2+-triggered content mixing assay offers more stringent test for the competition between Munc18-1 and syt1 to gain access to the SNAREpin.

To measure content mixing we incorporated sulforhodamine B into v-vesicles in an in vitro setup depicted in Fig.2a except for lipid dyes (Fig. 3a). After vesicle docking Ca2+ (500

µM) was flown in to the flow cell and the intensity jump of the fluorescence signal due to dequenching was detected as a signal for content mixing45. The cumulative time plot shows that approximately 13.5% of docked vesicle show content mixing with a half time of 30 sec (Fig. 3b).

Interestingly, addition of Munc18-1 showed little change in the cumulative time plot in the range of 100-1000 nM, although there was some reduction of content mixing at very high concentration of Munc18-1 (Fig. 3b). Thus, our result demonstrates that Munc18-1 does not stimulate Ca2+-triggered fusion pore formation, suggesting that the SNAREpin is not accessible to Munc18-1 in the presence of syt1. 71

We note that in this assay vesicle docking became much reduced as the Munc18-1 concentration was increased (Fig. 3c and Table S2). We interpret this result as the consequence of dissociation of heterodimeric t-SNARE into the individual components syntaxin 1a and

SNAP-25 (ref. 30), which would reduce vesicle docking mediated by the interaction between syt1 and t-SNARE35,38.

3.4 Discussion

In this work, we observed, using smFRET and single vesicle lipid mixing, that Munc18-

1’s positive influence on SNARE zippering and lipid mixing is robust in the absence of syt1.

Surprisingly, however, when syt1 is present the stimulation of SNARE zippering by Munc18-1 disappears completely while its effect on lipid mixing is reduced significantly. Most importantly, we did not observed any acceleration of content mixing or fusion pore opening by Munc18-1.

These findings demonstrate that syt1 competes with Munc18-1 for the access to the SNAREpin and acts as an antagonist for Munc18-1 in SNAREpin binding.

Neurotransmitter release is tightly regulated by Ca2+ and happens in less than 1 msec upon the Ca2+ influx46,47. Thus, it appears to be necessary for Ca2+-sensor syt1 to gain intimate access to the SNAREpin32. In fact, it is shown that syt1 has the capacity to bind the SNARE core and negatively charged lipids clustered at the immediate vicinity of the SNARE complex48. With the necessity of the syt1’s presence at the nearest neighbor of the SNAREpin it is hard to envision how Munc18-1 gains full access to the SNAREpin. Meanwhile, Munc18-1 binding to 72

the SNARE core is relatively weak with the binding constant of ~1 µM49 which may not be sufficiently strong to hold onto the SNARE core in competition with syt1.

Alternatively, Munc18-1’s binding to the Habc domain of syntaxin 1a is much stronger with the binding constant of ∼10 nM28,50. This binding stabilizes the ‘closed’ form of syntaxin 1a, which does not allow premature binding of syntaxin 1a to SNAP-25 (ref. 30). The controlled binding between syntaxin 1a and SNAP-25 is necessary because their freely binding likely leads to the formation of the 2:1 complex which is known to be the non-productive dead-end product51,52. Recently, it was shown that Munc13 plays a role in relieving syntaxin 1a from the inhibitory Munc18-1 capping29,30.

The notion that Munc18-1 is part of the minimal fusion machinery is largely based on two experimental observations. Firstly, the knockout of Munc18-1 abolishes neurotransmitter release completely35,53,54. The severity of the knockout phenotype could be explained equally well by the losing of inhibitory capping on syntaxin 1a. Secondly, Munc18-1 has the capacity of accelerating SNARE-dependent proteoliposome fusion significantly12,25. The caveats here are that syt1 is not included in those experiments. Our results argue that Munc18-1 may not be part of the minimal fusion machinery and it rather plays an important regulatory role in controlling syntaxin 1a binding to SNAP-25.

There are however many exocytotic systems where vesicle fusion is either not regulated by Ca2+ and therefore, syt1-like molecules are not involved55,56. Also, there are systems where the Habc-like domain is not present in syntaxin 1a-analogs57,58. In such cases, the

Sec1/Munc18’s main function may be to stimulate vesicle fusion via its binding to the

SNAREpin. The analysis by Shen et al. suggests that the SNAREpin binding may be the 73

evolutionarily conserved function for the Sec1/Munc18 family25. However, at the top of the evolution may be neuroexocytosis. Thus, we speculate that the evolutionary pressure to implement the tight Ca2+ control of vesicle fusion might the Muc18-1’s role must have been diverged elsewhere.

3.5 Materials and Methods

3.5.1 Plasmid constructs and site-directed mutagenesis.

DNA sequences encoding rat syntaxin-1a (amino acids 2-288 with three native cysteines

C145, C271, and C272 replaced by alanines), rat VAMP2 (amino acids 1-116 with C103 replaced by alanine), and rat SNAP-25 (amino acids 1-206 with four native cysteines C85, C88,

C90, and C92 replaced by alanines) were inserted into the pGEX-KG vector as N-terminal glutathione S-transferase (GST) fusion proteins. Full-length rat synaptotagmin 1 (Syt1, amino acids 50-421 with four native cysteines C74, C75, C77 and C79 replaced by alanines and another

C82 replaced by serine) and full-length rat Munc18-1 was inserted into pET-28b vector as a C- terminal His-tagged protein. We used the Quick Change site-directed mutagenesis kit

(Stratagene) to generate all mutants including syntaxin-1a I203C and V241C and VAMP2 Q33C and A72C. DNA sequences were confirmed by the Iowa State University DNA Sequencing

Facility.

3.5.2 Protein expression and purification.

Protein expression and purification were described previously41,45. Briefly, all recombinant proteins were expressed in E. coli BL21 (DE3). GST-tagged proteins, syntaxin-1a,

SNAP-25, and VAMP2 were purified by affinity chromatography using Glutathione−Agarose beads (Sigma-Aldrich) and cleaved from beads with thrombin (0.02 unit μl-1, Sigma-Aldrich) in

PBS or PBS with 0.8% (w/v) octyl β-D-glucopyranoside (PBS-OG) for membrane proteins. His- tagged proteins, apoA1, synaptotagmin 1 (syt1), and Munc18-1 were purified by Ni-NTA agarose beads (Qiagen). His-tagged apoA1 and Munc18-1 were eluted with 200 mM imidazole

(Sigma-Aldrich) in PBS. Munc18-1 was further dialysed twice in 2 L PBS buffer at 4 °C overnight after elution. Syt1 was eluted with 25 mM HEPES pH 7.4 with 400 mM KCl, 250 mM imidazole, 0.8% OG, and 1 mM EDTA.

3.5.3 Lipid mixture preparation.

The lipid molecules used in this study are 1,2-dioleoyl-sn-glycero-3-phospho-L-serine

(DOPS), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), phosphatidylinositol-4,5- bisphosphate (PIP2, from porcine brain), cholesterol, 1,2-dipalmitoyl-sn-glycero-3- phosphoethanolamin-N-(biotinyl) (biotin-DPPE), and 1,2-distearoyl-sn-glycero-3- phosphoethanolamine-N-[biotinyl(polyethylene glycol)-2000] (biotin-PEG-DSPE). All lipids were obtained from Avanti Polar Lipids. 1,1 ′ -Dioctadecyl-3,3,3 ′ ,3 ′ -

Tetramethylindocarbocyanine Perchlorate (DiI), 1,1 ′ -Dioctadecyl-3,3,3 ′ ,3 ′ - 75

Tetramethylindodicarbocyanine Perchlorate (DiD), and sulforhodamine B were obtained from

Invitrogen. The ideal amount of lipids were first mixed and then completely dried for further using.

3.5.4 Fluorophore-labeling of the single cysteine mutants.

The fluorophore-labeling of the single cysteine mutants of syntaxin-1a and VAMP2 were described previously41,59. Briefly, the mutants were purified through the way as the same as wild- type syntaxin-1a and VAMP2, except that the cleavage buffer for single cysteine mutant proteins contained 2.5 mM DTT. For the smFRET study of the trans-SNAREpin, single cysteine mutants of syntaxin-1a (I203C or V241C) and VAMP-2 (Q33C or A72C) were desalted with the PD-10 column (GE healthcare) to eliminate free DTT after cleavage then incubated with 10x molar excess of maleimide-derivative fluorophore, Cy5 and Cy3 (GE healthcare), respectively, at 4 °C overnight. Unreacted free fluorescence labels were removed by the PD-10 column and the labelling efficiency of each mutant was measured with spectrophotometry (Beckman). Extinction coefficients of Cy5 (250, 000 M-1 cm-1 at 650 nm) and Cy3 (150, 000 M-1 cm-1 at 552 nm) were used to calculate the concentration of the fluorophore. The detergent compatible Lowry assay

(DC assay, Bio-Rad) was used to determine the concentration of SNAREs. The labelling efficiencies were 46% and 40% for syntaxin-1a I203C and V241C and 55% and 62% for

VAMP2 Q33C and A72C, respectively.

3.5.5 Reconstitution and purification of t- and v-discs.

To mimic the physiological lipid compositions of synaptic membranes the mixture of

POPC, DOPS, cholesterol, PIP2, and biotin-PEG-DSPE with the molar ratio of 62.9:15:20:2:0.1 for t-discs and the mixture of POPC, DOPS, and cholesterol with the molar ratio of 75:5:20 for v-discs were dried and resuspended inTris150-EDTA buffer (10 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, pH 7.4). 3 μl of each 50 mM lipid mixture was dissolved by final 50 mM sodium cholate then t- (the binary complex of syntaxin-1a and SNAP-25) or v-SNARE (VAMP-2 with or without syt1 with the molar ratio of 1:1) and apoA1 protein were added to the solubilized lipid mixture. The molar ratio between lipid, SNARE(s), and apoA1 was 160:0.5:2. The self-assembly of SNARE-incorporated nanodiscs was initiated by a rapid removal of sodium cholate by treating the sample with 50% (w/v) SM-2 Bio-Beads. The t- or v-discs were then purified through gel-filtration using SuperdexTM200 GL 10/30 column (Amersham Biosciences)

3.5.6 FRET measurements of trans-SNAREpins.

After coating the quartz surface with a solution of methoxy-polyethylene glycol (mPEG) and biotin-PEG molecules (100:1), the quartz slide was assembled into a flow chamber and coated with streptavidin (0.2 mg ml-1). 100~200 nM t-discs (containing Cy5-labeled sytnaxin 1a on the position of I203C or V241C) were immobilized on the surface then 1 μM Munc18-1 or equal volume buffer were flown in and the sample was incubated for 10 min at room temperature

(~25 °C). Five times molar excess of v-discs (containing Cy3-labeled VAMP2 on the position of

Q33C or A72C without or with syt1) with 1 μM Munc18-1 or equal volume buffer were then 77

added to the flow chamber and the sample was incubated for 25 min at 37 °C to allow trans-

SNAREpin formation. All TIR experiments were performed at room temperature in the presence of oxygen scavenger system (0.4 % (w/v) glucose (Sigma), 4 mM Trolox (Calbiochem), 1 mg/ml glucose oxidase (Sigma), 0.04 mg/ml catalase (Calbiochem)) in Tris150-EDTA buffer. The smFRET measurements were carried out on a prism-type TIR setup, which is based on inverted microscope (IX71, Olympus) with laser exposure time of 200 ms. Each Cy3 and Cy5 fluorescence signal from the chamber was collected by a water immersion lens (UPlanSApo

60x/1.20w, Olympus) and separated by a dichroic mirror (T660lpxr, chroma), which had a threshold at 635 nm wavelength. Two separated signals were then transferred to separated images on an electron multiplying charged-coupled device camera (iXon DU897E, Andor

Technology) as donor and acceptor signals, respectively. FRET efficiencies (E) were obtained by

( )

where IA and ID are the acceptor and donor fluorescence intensities.

3.5.7 proteoliposomes reconstitution.

For the bulk and single lipid mixing assay, the molar ratios of lipids were 15:64:20:1:0.1

(DOPS:POPC:cholesterol:DiI:Biotin-DPPE) for the t-SNARE-reconstituted (t-)vesicles, and

5:73:20:2 (DOPS:POPC:cholesterol:DiD) for the v-SNARE-reconstituted (v-)vesicles, respectively. The lipid mixture was first completely dried and then hydrated by dialysis buffer

(25 mM HEPES, pH 7.4, 100 mM KCl). After five freeze–thaw cycles, protein-free large 78

unilamellar vesicles (~100 nm in diameter) were prepared by extrusion through a 100 nm polycarbonate filter (Whatman). For membrane reconstitution, SNARE proteins and syt1 were mixed with protein free vesicles at the protein to lipid molar ratio of 1:200 for each protein component (this ratio was kept for all experiments including the single vesicle content mixing assay) with ~0.8% OG in the dialysis buffer at 4°C for 15 min. The liposome/protein mixture was diluted 2 times with dialysis buffer for t-vesicles, and then the diluted t-vesicles were dialyzed in 2 L of dialysis buffer at 4 °C overnight. For v-vesicles, the mixture was diluted twice with dialysis buffer containing 1 mM EDTA and dialyzed in 2 L of dialysis buffer with EDTA at

4 °C overnight. Details for reconstitution were discussed in our previous work39,45.

For the single vesicle content mixing assay with the small sulforhodamine B content indicator, the lipid compositions were as the same as those used in the single vesicle docking assay except that the fluorescent lipid dyes (DiI and DiD) were replaced by the equal amount of

POPC and 2% PIP2 was incorporated into the t-vesicles. The lipid mixture was first completely dried and then hydrated with dialysis buffer, but a population of vesicles to make v-vesicles was hydrated in the presence of 20 mM sulforhodamine B. The overall vesicle preparation and protein reconstitution process was the same as above except that v-vesicles were always kept in the 20 mM sulforhodamine B prior to dialysis overnight. Free sulforhodamine B was removed using the PD-10 desalting column (GE healthcare) after dialysis. Details for reconstitution were discussed in our previous work39,45.

3.5.8 Bulk lipid mixing assay.

Reconstituted t- and v-vesicles were mixed at a ratio of 1:1. The final lipid concentration was 0.1 mM. The fluorescence intensity was monitored in two channels with the excitation wavelength of 530 nm and emission wavelengths of 570 and 670 nm for donor DiI and acceptor

DiD, respectively. Fluorescence changes were recorded with the Varian Cary Eclipse model fluorescence spectrophotometer using a quartz cell of 100 μL with a 2 mm path length. All measurements were performed at 35 °C.

3.5.9 Single vesicle docking and lipid mixing assays.

The t-vesicles with the final lipid concentration of 1 μM were flown into the chamber and immobilized on the PEG-coated surface through streptavidin-to-biotin lipid conjugation through

30-minute incubation at room temperature (~25 C). After two rounds of washing with 200 l dialysis buffer, 1 μM Munc18-1 or equal volume of dialysis buffer was flown into the chamber and the sample was incubated for another 30 minutes at room temperature. Then, the v-vesicles

(1 ~ 3 μM) with or without 1 μM Munc18-1 were injected into the flow chamber and the sample was again incubated for 30 minutes at 37 C. After washing off free v-vesicles using dialysis buffer containing 1 μM Munc18-1 or equal volume of dialysis buffer, movies were acquired by taking 100 consecutive frames with the 100 ms exposure time from five randomly chosen imaging areas (4590 m2). The first 60 frames were taken by using 532 nm laser source to calculate the FRET efficiency and the following 40 frames were taken by using 635 nm laser 80

excitation to check the binding of v-vesicles (including docked and non-specific binding of v- vesicles. The non-specifically bound v-vesicles were excluded from the analysis).

3.5.10 Single-vesicle content mixing assay.

The t-SNARE vesicles (125 μM) were immobilized on the PEG-coated surface through streptavidin-to-biotin lipid conjugation for 20 min. Following several rounds of washing with

200 μL of dialysis buffer, the v-vesicles containing 20 mM SRB (10 μM) were injected into the flow chamber and the sample were incubated at room temperature for 10 min for docking. After washing out the unbound v-vesicles with dialysis buffer containing 1mM EDTA, additional 20 min incubation was followed by 500 μM of Ca2+ injection using motorized syringe pump. A stepwise jump in fluorescence intensity is detected with fusion pores, due to the dequenching of

SRB39,45. The fluorescent signals from the immobilized individual vesicle pairs were obtained from TIF fluorescence microscope. The details of TIR fluorescence microscope imaging and single molecule data analysis have been reported in our previous work60.

3.6 References

1. Chernomordik, L.V. & Kozlov, M.M. Protein-lipid interplay in fusion and fission of biological membranes. Annu Rev Biochem 72, 175-207 (2003).

2. Martens, S. & McMahon, H.T. Mechanisms of membrane fusion: disparate players and common principles. Nat Rev Mol Cell Biol 9, 543-56 (2008).

3. Sudhof, T.C. & Rothman, J.E. Membrane fusion: grappling with SNARE and SM proteins. Science 323, 474-7 (2009).

4. Rizo, J. & Sudhof, T.C. The membrane fusion enigma: SNAREs, Sec1/Munc18 proteins, and their accomplices--guilty as charged? Annu Rev Cell Dev Biol 28, 279-308 (2012).

5. Sollner, T. et al. SNAP receptors implicated in vesicle targeting and fusion. Nature 362, 318-24 (1993).

6. Sutton, R.B., Fasshauer, D., Jahn, R. & Brunger, A.T. Crystal structure of a SNARE complex involved in synaptic exocytosis at 2.4 A resolution. Nature 395, 347-53 (1998).

7. Poirier, M.A. et al. The synaptic SNARE complex is a parallel four-stranded helical bundle. Nat Struct Biol 5, 765-9 (1998).

8. Weber, T. et al. SNAREpins: minimal machinery for membrane fusion. Cell 92, 759-72 (1998).

9. Brunger, A.T. Structure and function of SNARE and SNARE-interacting proteins. Q Rev Biophys 38, 1-47 (2005).

10. Jahn, R. & Scheller, R.H. SNAREs--engines for membrane fusion. Nat Rev Mol Cell Biol 7, 631-43 (2006).

11. Stein, A., Weber, G., Wahl, M.C. & Jahn, R. Helical extension of the neuronal SNARE complex into the membrane. Nature 460, 525-8 (2009).

12. Shen, J., Tareste, D.C., Paumet, F., Rothman, J.E. & Melia, T.J. Selective activation of cognate SNAREpins by Sec1/Munc18 proteins. Cell 128, 183-95 (2007).

13. Deak, F. et al. Munc18-1 binding to the neuronal SNARE complex controls synaptic vesicle priming. J Cell Biol 184, 751-64 (2009).

14. Diao, J. et al. Single-Vesicle Fusion Assay Reveals Munc18-1 Binding to the SNARE Core Is Sufficient for Stimulating Membrane Fusion. ACS Chem Neurosci 1, 168-174 (2010).

15. Schiavo, G. et al. Tetanus and botulinum-B neurotoxins block neurotransmitter release by proteolytic cleavage of synaptobrevin. Nature 359, 832-5 (1992).

16. Blasi, J. et al. Botulinum neurotoxin A selectively cleaves the synaptic protein SNAP-25. Nature 365, 160-3 (1993).

17. Keller, J.E. & Neale, E.A. The role of the synaptic protein snap-25 in the potency of botulinum neurotoxin type A. J Biol Chem 276, 13476-82 (2001).

18. Breidenbach, M.A. & Brunger, A.T. New insights into clostridial neurotoxin-SNARE interactions. Trends Mol Med 11, 377-81 (2005).

19. Pobbati, A.V., Stein, A. & Fasshauer, D. N- to C-terminal SNARE complex assembly promotes rapid membrane fusion. Science 313, 673-6 (2006).

20. Bhalla, A., Chicka, M.C., Tucker, W.C. & Chapman, E.R. Ca(2+)-synaptotagmin directly regulates t-SNARE function during reconstituted membrane fusion. Nat Struct Mol Biol 13, 323-30 (2006).

21. Stein, A., Radhakrishnan, A., Riedel, D., Fasshauer, D. & Jahn, R. Synaptotagmin activates membrane fusion through a Ca2+-dependent trans interaction with phospholipids. Nat Struct Mol Biol 14, 904-11 (2007).

22. Yoon, T.Y. et al. Complexin and Ca2+ stimulate SNARE-mediated membrane fusion. Nat Struct Mol Biol 15, 707-13 (2008).

23. Lee, H.K. et al. Dynamic Ca2+-dependent stimulation of vesicle fusion by membrane- anchored synaptotagmin 1. Science 328, 760-3 (2010).

24. Shi, L. et al. SNARE proteins: one to fuse and three to keep the nascent fusion pore open. Science 335, 1355-9 (2012).

25. Yu, H. et al. Comparative studies of Munc18c and Munc18-1 reveal conserved and divergent mechanisms of Sec1/Munc18 proteins. Proc Natl Acad Sci U S A 110, E3271- 80 (2013).

26. Fernandez, I. et al. Three-dimensional structure of an evolutionarily conserved N- terminal domain of syntaxin 1A. Cell 94, 841-9 (1998).

27. Dulubova, I. et al. A conformational switch in syntaxin during exocytosis: role of munc18. EMBO J 18, 4372-82 (1999).

28. Misura, K.M., Scheller, R.H. & Weis, W.I. Three-dimensional structure of the neuronal- Sec1-syntaxin 1a complex. Nature 404, 355-62 (2000).

29. Ma, C., Li, W., Xu, Y. & Rizo, J. Munc13 mediates the transition from the closed syntaxin-Munc18 complex to the SNARE complex. Nat Struct Mol Biol 18, 542-9 (2011).

30. Ma, C., Su, L., Seven, A.B., Xu, Y. & Rizo, J. Reconstitution of the vital functions of Munc18 and Munc13 in neurotransmitter release. Science 339, 421-5 (2013).

31. Fernandez-Chacon, R. et al. Synaptotagmin I functions as a calcium regulator of release probability. Nature 410, 41-9 (2001).

32. Chapman, E.R. Synaptotagmin: a Ca(2+) sensor that triggers exocytosis? Nat Rev Mol Cell Biol 3, 498-508 (2002).

33. Rizo, J. & Rosenmund, C. Synaptic vesicle fusion. Nat Struct Mol Biol 15, 665-74 (2008).

34. Perin, M.S., Brose, N., Jahn, R. & Sudhof, T.C. Domain structure of synaptotagmin (p65). J Biol Chem 266, 623-9 (1991).

35. de Wit, H. et al. Synaptotagmin-1 docks secretory vesicles to syntaxin-1/SNAP-25 acceptor complexes. Cell 138, 935-46 (2009).

36. Hui, E., Johnson, C.P., Yao, J., Dunning, F.M. & Chapman, E.R. Synaptotagmin- mediated bending of the target membrane is a critical step in Ca(2+)-regulated fusion. Cell 138, 709-21 (2009).

37. Lai, Y. & Shin, Y.K. The importance of an asymmetric distribution of acidic lipids for synaptotagmin 1 function as a Ca2+ sensor. Biochem J 443, 223-9 (2012).

38. Kim, J.Y. et al. Solution single-vesicle assay reveals PIP2-mediated sequential actions of synaptotagmin-1 on SNAREs. EMBO J 31, 2144-55 (2012).

39. Lai, Y., Lou, X., Jho, Y., Yoon, T.Y. & Shin, Y.K. The synaptotagmin 1 linker may function as an electrostatic zipper that opens for docking but closes for fusion pore opening. Biochem J 456, 25-33 (2013).

40. Shin, Y.K. Two gigs of Munc18 in membrane fusion. Proc Natl Acad Sci U S A 110, 14116-7 (2013). 84

41. Shin, J., Lou, X., Kweon, D.H. & Shin, Y.K. Multiple conformations of a single SNAREpin between two nanodisc membranes reveal diverse pre-fusion states. Biochem J 459, 95-102 (2014).

42. Kyoung, M., Zhang, Y., Diao, J., Chu, S. & Brunger, A.T. Studying calcium-triggered vesicle fusion in a single vesicle-vesicle content and lipid-mixing system. Nat Protoc 8, 1-16 (2013).

43. Dulubova, I. et al. Munc18-1 binds directly to the neuronal SNARE complex. Proc Natl Acad Sci U S A 104, 2697-702 (2007).

44. Weninger, K., Bowen, M.E., Chu, S. & Brunger, A.T. Single-molecule studies of SNARE complex assembly reveal parallel and antiparallel configurations. Proc Natl Acad Sci U S A 100, 14800-5 (2003).

45. Lai, Y. et al. Fusion pore formation and expansion induced by Ca2+ and synaptotagmin 1. Proc Natl Acad Sci U S A 110, 1333-8 (2013).

46. Jahn, R. & Fasshauer, D. Molecular machines governing exocytosis of synaptic vesicles. Nature 490, 201-7 (2012).

47. Borst, J.G. & Sakmann, B. Calcium influx and transmitter release in a fast CNS synapse. Nature 383, 431-4 (1996).

48. Honigmann, A. et al. Phosphatidylinositol 4,5-bisphosphate clusters act as molecular beacons for vesicle recruitment. Nat Struct Mol Biol 20, 679-86 (2013).

49. Xu, Y., Su, L. & Rizo, J. Binding of Munc18-1 to synaptobrevin and to the SNARE four- helix bundle. Biochemistry 49, 1568-76 (2010).

50. Burkhardt, P., Hattendorf, D.A., Weis, W.I. & Fasshauer, D. Munc18a controls SNARE assembly through its interaction with the syntaxin N-peptide. EMBO J 27, 923-33 (2008).

51. Fasshauer, D., Otto, H., Eliason, W.K., Jahn, R. & Brunger, A.T. Structural changes are associated with soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor complex formation. J Biol Chem 272, 28036-41 (1997).

52. Xiao, W., Poirier, M.A., Bennett, M.K. & Shin, Y.K. The neuronal t-SNARE complex is a parallel four-helix bundle. Nat Struct Biol 8, 308-11 (2001).

53. Gerber, S.H. et al. Conformational switch of syntaxin-1 controls synaptic vesicle fusion. Science 321, 1507-10 (2008).

54. Verhage, M. et al. Synaptic assembly of the brain in the absence of neurotransmitter secretion. Science 287, 864-9 (2000).

55. Schekman, R. Genetic and biochemical analysis of vesicular traffic in yeast. Curr Opin Cell Biol 4, 587-92 (1992).

56. Bonifacino, J.S. & Glick, B.S. The mechanisms of vesicle budding and fusion. Cell 116, 153-66 (2004).

57. Dulubova, I., Yamaguchi, T., Wang, Y., Sudhof, T.C. & Rizo, J. Vam3p structure reveals conserved and divergent properties of syntaxins. Nat Struct Biol 8, 258-64 (2001).

58. Dulubova, I. et al. How Tlg2p/syntaxin 16 'snares' Vps45. EMBO J 21, 3620-31 (2002).

59. Lai, Y., Lou, X., Wang, C., Xia, T. & Tong, J. Synaptotagmin 1 and Ca2+ drive trans SNARE zippering. Sci Rep 4, 4575 (2014).

60. Diao, J. et al. A single vesicle-vesicle fusion assay for in vitro studies of SNAREs and accessory proteins. Nat Protoc 7, 921-34 (2012).

3.7 Figures and Legends

Figure 1) Effects of Munc18-1 on the conformation of the trans-SNAREpin detected by smFRET

(a) Fluorescent dye-labeled positions on syntaxin 1a and VAMP2. I203C or V241C of syntaxin 1a was labeled with Cy5 (light blue saw-tooth circle) and Q33C or A72C of VAMP2 was labeled with Cy3 (light blue saw-tooth circle). The Habc domain (white) of syntaxin 1a is depicted broken to indicate the longer length than the SNARE motif (red). (b) The TIR setup for the smFRET detection of the trans-SNAREpin conformation. The t-discs carrying Cy5-labeled syntaxin 1a (red and white) and SNAP-25 (green) was immobilized on the quartz surface through biotin (yellow square) and streptavidin (purple, cross-shaped) conjugation and then allowed to interact and form the trans-SNAREpin with the v-discs carrying Cy3-labeled VAMP2 (blue) with or without synaptotagmin 1 (syt1) (not shown here). Distributions of FRET efficiencies for the NN and CC pairs in the absence (upper panel) or presence (lower panel) of Munc18-1 without syt1 (c) or with syt1 (d), respectively. The Gaussian fittings are included (black lines). The relative changes in the ratio of high- to low-FRET populations for the CC pairs in the presence (red) or absence of Munc18-1 (gray) without syt1 (e) and with syt1 (f), respectively. Error bars denote the S.D. of three independent experiments. Areas under Gaussian fittings were used to calculate the high-to-low FRET population ratios.

a b c f

d e

Figure 2) Munc18-1 and syt1 are mutually antagonistic in SNARE-dependent lipid mixing.

(a) Schematic diagram of the single-vesicle docking and lipid mixing assay. T-vesicles reconstituted with SNAP-25/syntaxin 1a were immobilized on the surface of the flow cell. V- vesicles reconstituted with VAMP2 or VAMP2 together with syt1 (VAMP2:syt1=1:1) were flown into the flow cell with or without 1 μM Munc18-1. (b-e) Distributions of FRET efficiencies between immobilized t-vesicles and VAMP2-vesicles (b), VAMP2-vesicles in the presence of 1 μM Munc18-1 (c), VAMP2/syt1-vesicles (d), and VAMP2/syt1-vesicles in the presence of 1 μM Munc18-1 (e). The Gaussian fitting were shown to indicate the low- and high- FRET populations. (f) Lipid-mixing efficiencies quantified based on the percentages of high- FRET peaks (E~0.5–1). Error bars denote the S.D. of three independent experiments. 89

a b c

Figure 3 Effects of Munc18-1 on Ca2+-triggered fusion pore opening in in vitro content mixing assay.

(a) Schematics of the in vitro content mixing assay. (b) Quantitative comparison of cumulative time content mixing percentage of the total docked population at various concentrations of Munc18-1 (Ca2+ injection at 10s). The control, without Munc18-1, is depicted with red circles. (c) In vitro single vesicle docking. Individual v-vesicles that tethered onto the t- vesicles on the imaging area were counted. The experiments were performed by incubating the samples in the presence of the specified Munc18-1 concentrations. The data were normalized against the control which was obtained in the absence of Munc18-1. Data in (b) and (c) were represented as mean ± S.D. of three independent experiments.

3.8 Supporting Information

Content: Supplementary Figure and Legends S1-S3, Table S1-S2

a b

Figure S1) The conformation of parallel SNARE assembly between nanodiscs.

Distribution of FRET efficiencies for the out-of-register NC (a, syntaxin 1a V241C- VAMP2 Q33C) and CN (b, syntaxin 1a I203C-VAMP2 A72C) pairs. The Gaussian fittings are included (black lines).

a b

c d

e f

Figure S2) Increased fully assembled SNAREpin dependent on Munc18-1.

Distributions of the FRET efficiencies for the CC pair in the absence (a) or presence of various concentrations of Munc18-1 (b, c, and d). The relative changes in the low-FRET peak for the CC pairs (e). FRET distribution of the CC pair in the presence of 2 µM BSA as a negative control (f). 92

a b

Figure S3) Munc18-1 behaves opposite ways depending on the absence or the presence of syt1.

(a and b) Bulk lipid mixing assays in the absence (a) or presence of syt1 (b). The change of the FRET efficiency (E), which was calculated by the formula: E = IDiD/(IDiD+IDiI), represents lipid mixing. The black traces represent the changes in the fluorescence intensity without Munc18-1 and the red traces represent those with Munc18-1.

Table S1) Single vesicle lipid mixing assay.

Table S2) Numbers of total docked vesicles for content mixing using Sulforhodamine B. 95

CHAPTER 4: THE ROLE OF COMPLEXIN-1 N-TERMINUS AS

AN ACTIVE ASSISTANT IN Ca2+ TRIGGERED FUSION PORE

FORMATION

Jaeil Shin, Xiaochu Lou, and Yeon-Kyun Shin

In preparation of manuscript

4.1 Abstract

Complexin-1 (cpx1) is a presynaptic protein known to interact with the SANRE complex to regulate synaptotagmin-1 (syt1) and Ca2+-dependent neuroexocytosis processes. The general notion of cpx1’s role in synchronized neurotransmitter release is to clamp SNARE-mediated fusion and facilitate it upon the arrival of a Ca2+ signal. The combined study of SNAREpin formation, SNARE-mediated lipid mixing, and content-mixing in the presence of selectively- truncated cpx1 revealed that cpx1 fastened the SNARE zipping and consequently increased lipid mixing, but produced no further increase in content mixing without Ca2+ influx. More interestingly, the N-terminal region of cpx1 performed the role of an active assistant, not a passive clamp, in Ca2+ triggered content mixing.

4.2 Introduction

In neuroexocytosis, synaptic vesicles must fuse to the plasma membrane and open a pore to release the neurotransmitter1-3. SNARE proteins have been thought to overcome the energy 96

barrier of membrane repulsion by providing the energy force generated when stable four-helical

SNARE complexes are formed4,5. This complex formation is initiated at the N-terminal region of

SNARE complex and favors finish zipping in the C-terminal direction2,6-8. By providing the favorable-complex formation property and its driving force to fuse membranes, SNARE is believed to constitute minimal fusogenic machinery2,9.

However, in the physiological environment, this spontaneous process is precisely governed by synaptic regulatory proteins such as synaptotagmin-1 (syt1) and complexin-1 (cpx1) in transferring the signal with correct timing10-14. As a Ca2+ sensor, syt1 consists of Ca2+ binding the C2AB domain with the transmembrane domain, anchoring syt1 to the synaptic vesicles15-18.

It is known to finalize the pore opening by interacting with the SNARE complex and lipid molecules upon the arrival of Ca2+13,14,19. The mechanism of cpx1 is still controversial, but is believed to prime and clamp synaptic vesicles to the plasma membrane until syt1 and Ca2+ arrive at the active spot to finally accelerate the pore-opening process20-23. How does cpx1 elevate the pore-opening process? Cpx1 is composed of 4 distinct domains (134 amino acids) and their postulated roles have been reported with through vivo and in vitro experiments. The central helix domain (cntH, 49-70) binds to the groove between VAMP-2 and syntaxin-1a in the assembled

SNARE complex in an antiparallel direction and thereby stabilizes it24-26. The accessory helix domain (accH, 27-48) may act as a clamp for SNAREpin conformation, possibly by competing with VAMP-2 at the binding site of the binary complex of syntaxin-1a and SNAP-25 (t-

SNARE)27,28. Almost half the length of cpx1 at the C-terminal (Ct, 71-134) is relatively unstructured and its function is not clear. Recently it has been reported that Ct possibly interacts with synaptic vesicles via its putative amphipathic residues and is thereby involved in vesicle 97

priming29,30. Another unstructured N-terminal domain (Nt, 1-26) is believed to provide a role in synchronized fusion pore-opening upon the arrival of calcium31,32. Although the distinct domains of cpx1 apparently play individual roles, there has been to date no plausible hypothesis for cpx1 function in SNARE-mediated neuroexocytosis33.

We used a single-molecule (sm) content-mixing assay to investigate the role of cpx1 in the pore-opening process and confirm its acceleratory function34. The structural information based on smFRET using trans-SNAREpin between nanodiscs revealed that cpx1 induced complete assembly of the core SNARE complex. Furthermore, the single vesicle-vesicle lipid- mixing efficiency positively increased in the presence of cpx1. However, N-terminally truncated cpx1 fragments were unable to elevate content mixing while the high efficiencies of complete

SNAREpin formation and lipid-mixing were still maintained. Meanwhile, C-terminal truncated cpx1 increased content-mixing efficiency, but enhanced neither the SNAREpin complex formation nor the lipid-mixing efficiency.

This result may justify the new mechanism of cpx1 during SNARE-mediated membrane fusion and neurotransmitter release. Priming of synaptic vesicles is mediated by cntH and Ct of cpx1, as known, but Nt of cpx1 may also act as an independent pore-opening assistant, not a passive blocker, in the Ca2+ triggered release event.

4.3 Results

4.3.1 Cpx1 facilitates content mixing events in the presence of syt1 and Ca2+

The well-established content-mixing assay procedure was introduced to confirm the role of cpx1 as an accelerator in the SNARE-mediated fusion-pore opening process in the presence of syt1 and Ca2+ 23,35. Unlabeled vesicles harboring the binary complex of syntaxin-1a and SNAP-

25 (t-vesicles) were immobilized on the surface of the flow cell via avidin-biotin interaction.

Other vesicles containing sulforhodamine B (SRB) as a concentration-dependent self-quenching fluorophore inside the vesicle and VAMP-2 and syt1 on the membrane (v-vesicles) were introduced to the flow chamber with or without 1 uM cpx1 and reacted for 10 min at room temperature (Fig. 1a). After washing unbound v-vesicles, another 20 min of incubation at room temperature allowed spontaneous content-mixing. Finally, 500 uM of Ca2+ was injected to initiate syt1 dependent content-mixing events, and the sudden increase of SRB intensity due to the dilution of self-quenching fluorophore was considered to be evidence of content-mixing. As shown previously, content-mixing probability was increased by the cpx1, indicating that it is a final fusion accelerator (Fig.1b).

4.3.2 SNARE motif in the trans-SNAREpin is fully assembled by cpx1

A trans-SNAREpin model was recruited to determine the structure of SNAREpin under the control of cpx136,. Truncated cpx1 fragments were also produced and applied to the trans-

SNAREpin complex to investigate the domain-specific role of cpx1 on SNAREpin assembly 99

(Fig. 2a). The binary complex of acceptor fluorophore Cy5-labeled at the N- (I203C) or C- terminal (V241C) of syntaxin-1a and SNAP-25 (t-SNARE) was incorporated into nanodiscs (t- discs). They were immobilized on the flow cell mediated by avidin and biotin interaction, then incubated with nanodiscs harboring donor fluorophore Cy3-labeled VAMP-2 at the N- (Q33C) or C-terminal (A72C) region and with syt1 (v-discs) (Fig. 2b). Consistent with previous results, an N-terminal pair of SNAREpins (NN pair) formed a robustly-assembled structure, while a C- terminal pair (CC pair) contained at least three major distributions of conformations, although syt1 was added to v-discs36. The NN pair showed no distribution change in the presence of cpx1 as expected; however, a high-FRET population of CC pairs increased, indicating a stably- assembled conformation independent of syt1 (Fig. 2c and Fig. S1). N-terminally truncated cpx1 fragments, cpx27-134 and cpx41-134, also resulted in an increased high FRET portion while a C- terminal truncated cpx1 (cpx1-75) did not affect the FRET distribution at all (Fig. 2d). The docking probability was calculated based on the ratio of bound v-disc numbers and total t-disc numbers, with the result that the cntH and Ct of cpx1 induced binding between v-discs and t- discs while C-terminally truncated cpx1 reduced (Fig. 2e).

4.3.3 Cpx1 increases vesicle-vesicle lipid mixing

We asked if facilitated SNARE assembly under the force of cpx1 also increased membrane fusion efficiency. Vesicle-vesicle lipid mixing assay in a single molecule level was applied to determine the role of cpx1 on the SNARE-mediated lipid-mixing event. Vesicles carrying t-SNAREs were immobilized on the flow-cell surface (t-vesicles), then incubated with 100

other vesicles harboring VAMP-2 with or without syt1 (v-vesicles) in the presence or absence of cpx1 to allow the docking event (Fig. 3a). To generate the FRET between two vesicles, lipidic- fluorophore DiI and DiD were individually incorporated into each vesicle. Lipid mixing was basically enhanced by syt1 in v-vesicles, consistent with a previous study23. The additional cpx1 in the interacting vesicle pairs shifted FRET efficiency to a higher portion independent of syt1, indicating that cpx1 itself was an accelerator of SNARE-mediated membrane fusion (Fig. 3b).

Correlated to the effect of cpx1 fragments on the SNAREpin formation, lipid-mixing efficiency was also increased by N-terminally truncated cpx27-134 and cpx41-134, while cpx1-75 did not change the FRET distribution as much as the other fragments (Fig. 3c).

4.3.4 N-terminal region of cpx1 actively induces fusion pore opening

We confirmed that cpx1, without an N-terminal part, induced a fully-assembled SNARE motif region followed by enhanced membrane-fusion efficiency. To determine how cpx1 and its fragments also drive the final step of SNARE-mediated membrane fusion such as pore-opening, the content-mixing assay setup was applied once again with cpx1 and truncated cpx1 fragments.

Unlabeled t-vesicles were placed on the imaging surface via the interaction between avidin and biotin, then SRB containing v-vesicles with or without cpx1 were injected as shown in Fig 1a.

After spontaneous content-mixing, Ca2+ triggered content-mixing events in the presence of different versions of cpx1 fragments were recorded. As already documented, full-length cpx1 facilitated the content-mixing process. However, N-terminally truncated cpx1 fragments: cpx27-

134 and cpx41-134, were not able to increase content-mixing although they provided the same effect 101

as full-length cpx1 in the SNAREpin assembly and lipid-mixing experiments (Fig. 4a). This disagreement may imply an intermediate state such as hemi-fused membranes forced by the fully-assembled trans-SNAREpin, merging outer leaflets of membranes while inner leaflets are still separated. In contrast, a C-terminally truncated cpx1 fragment, cpx1-75, increased content- mixing at a slower initial rate while affecting neither the fully assembled trans-SNAREpin conformation nor the lipid-mixing efficiency. We interpret the N-terminal region of cpx1 to be the most important one in regulating the synchronized fusion pore-opening event triggered by syt1 and Ca2+, and it may be able to assist the event actively as a supporter and not a passive blocking clamp as proposed by other groups. The docked v-vesicle numbers were also increased by cntH and Ct of cpx1, while Nt and accH decreased their probability in a manner consistent with the single SNARE docking probability gained from nanodisc experiment (Fig. 4b and Fig.

S2).

4.4 Discussion

The current study of trans-SNAREpin structure based on smFRET and nanodisc setup revealed that a core SNARE complex was fully assembled in the presence of cpx1. Also, vesicle- vesicle lipid-mixing using smFRET indicated that cpx1 drove vesicles to a highly-mixed state.

Even though N-terminally truncated cpx1 fragments (cpx27-134 and cpx41-134), including common regions, cntH and Ct of cpx1 were still able to induce complete assembly of SNAREpin and lipid mixing, they did not produce an increase in content-mixing events upon the arrival of Ca2+. Also, cpx1-75 did not increase SNAREpin assembly or lipid mixing as much as the other cpx1 102

fragments did even though it contained a SNAREpin stabilizing region, cntH of cpx1. We interpret this to mean that Ct of cpx1 may interact with synaptic vesicles and increase the local concentration of cntH of cpx1 at the active spot. Taken together, the fully-assembled core

SNAREpin complex induced by cpx1 (mainly by cntH and Ct) brings two membranes into close proximity then consequently merges them without pore formation until the arrival of a Ca2+ signal. This partially-fused state may imply the hemifusion of membranes37,38. SNARE might be the hemifusogenic machinery and cpx1, especially cntH and Ct, speeds up this process.

Previously-reported studies also indicated that cntH of cpx1 bound and stabilized the fully assembled SNAREpin and that Ct of cpx1 is implicated in synaptic vesicle-priming events via its potential membrane-binding region containing amphipathic residues24-26,29,30.

During the pore-opening step, it is evident that the role of cpx1 Nt is that of an accelerator; the mechanism is, however, somehow elusive31,39. It has been suggested that Nt of cpx1 acts as a final defender against spontaneous release, possibly through antiparallel interaction with the C-terminal region of a SNARE complex31. In this work, however, the spontaneous content-mixing event was not increased in the presence of cpx27-134 or cpx41-134, indicating that Nt of cpx1 was not required for the blocking of full-fusion. Meanwhile, cpx1-75 demonstrated a similar facilitating effect on content-mixing although the rate was a bit lower than for full-length cpx1, while there was almost no difference in SNAREpin assembly and lipid mixing. We concluded that Nt of cpx1 actively supports the pore-opening process mediated by syt1 and Ca2+ rather than passively leaves the site. However, the reported in vivo studies clearly showed that deletion of N-terminal cpx1 increased the spontaneous release, which supports its blocking role39. Even though our in vitro experiment set-up was performed with only minimal 103

participants involved in SNARE-mediated fusion events, it is believed to be a suitable platform for understanding the role of each molecule.

The docking probabilities between v- and t-discs or v- and t-vesicles were increased by the additional cpx1 or cpx1 fragments that contained both cntH and Ct, while the cpx1 fragment harboring Nt and accH reduced only the docking percentage. As a competitor for VAMP-2 at the binding site in the binary complex, accH possibly inhibits the SNARE binding process. However, the presence of full-length cpx1 and cpx27-134 that also contains accH did not result in reduced docking. Accordingly, we conclude that the individual role of each cpx1 domain has a running priority and the effect of accH as a SNARE-binding inhibitor is masked by Ct or by cntH and Ct of cpx1 cooperatively.

Based on the current work, we speculate on using a model in which the synaptic vesicle containing VAMP-2 and syt1 docks to the plasma membrane and N-terminal zippering of

SNAREpin is initiated (Fig 5a). In the presence of cpx1, Ct of cpx1 binds to the lipid (favorably to synaptic vesicles) and increases the local concentration of cntH of cpx1. The core SNARE complex formation is stabilized by cntH of cpx1 and results in vesicle priming (Fig. 5b). At this state, lipid mixing is efficiently processed while the two transmembrane domains of syntaxin-1a and VAMP-2 are still separated, consequently generating a hemifused state (Fig 5b). The conformational change of syt1 by Ca2+ binding to C2AB domain finally executes pore-opening to release neurotransmitters and is supported by Nt of cpx1 (Fig 5c). At the final step for pore opening, two transmembrane domains (and also membrane-proximal regions) of VAMP-2 and syntaxin-1a must interact with one another to form the extended SNARE complex and the full- 104

fused membranes. The N-terminal domain of cpx1 as a pore-opening helper is possibly involved in the final complex formation between transmembrane domains of SNAREs, and this interaction should be addressed by further study to unravel the mechanism of cpx1 in neurotransmitter release.

4.5 Materials and Methods

4.5.1 Protein preparation

Proteins used in this study were expressed in E. coli BL21 (DE3) and purified with the general protocol through either GST (Sigma-Aldrich) or Ni-NTA (Qiagen) agarose beads34,40.

GST-tagged proteins, syntaxin-1a, SNAP-25, VAMP-2, and all complexin proteins (cpx1, cpx27-

134, cpx41-134, and cpx1-75) were bound to GST agarose beads in PBS or PBS with 0.8% octyl β-D- glucopyranoside (PBS-OG) for membrane proteins. After washing out the non-specific bindings from the beads, proteins were eluted by the cleavage enzyme thrombin (0.02 unit μl-1, Sigma-

Aldrich). His-tagged proteins apoA1 and syt1 were attached to the Ni-NTA beads in PBS and

HEPES buffer consisting of 25 mM HEPES pH 7.4 with 400 mM KCl, 25 mM imidazole, 0.8%

OG, and 1 mM EDTA. After thoroughly washing the beads, apoA1 protein was eluted by 200 mM Imidazole in PBS, while syt1 was obtained using the HEPES buffer with an additional 250 mM of imidazole. All proteins containing a cysteine residue for further fluorophore labeling were eluted in the presence of 2.5 mM DTT as a reductant.

105

4.5.2 Fluorophore-labeling of cysteine mutant proteins

The fluorophore-labeling of the single cysteine mutants of syntaxin-1a and VAMP-2 was previously described40,41. Briefly, single cysteine mutants of syntaxin-1a (I203C or V241C) and

VAMP-2 (Q33C or A72C) were prepared in a DTT-free state by using a PD-10 (GE healthcare) desalting column, then labeled by maleimide-derivative fluorophore, either Cy5 or Cy3, The free fluorophores were removed by the PD-10 column after complete interaction and the labeling efficiencies were determined with spectrophotometry (Beckman). Extinction coefficients of Cy5

(250, 000 M-1 cm-1 at 650 nm) and Cy3 (150, 000 M-1 cm-1 at 552 nm) were used to calculate the concentration of fluorophore. A detergent-compatible Lowry assay (DC assay, Bio-Rad) was used to determine the concentration of SNAREs. The labeling efficiencies were 46% and 40% for syntaxin 1a I203C and V241C and 55% and 62% for VAMP2 Q33C and A72C, respectively.

4.5.3 Preparation of phospholipid mixtures

The lipid molecules used in this study were 1-palmitoyl-2-oleoyl-sn-glycero-3- phosphocholine (POPC), 1,2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS), cholesterol, phosphatidylinositol-4,5-bisphosphate (PIP2, from porcine brain), 1,2-dipalmitoyl-sn-glycero-3- phosphoethanolamin-N-(biotinyl) (biotin-DPPE), and 1,2-distearoyl-sn-glycero-3- phosphoethanolamine-N-[biotinyl(polyethylene glycol)-2000] (biotin-PEG-DSPE). All lipids were obtained from Avanti Polar Lipids. 1,1′-Dioctadecyl-3,3,3′,3′-Tetra-methy-lindo-carbo- cyanine Perchlorate (DiI), 1,1′-Dioctadecyl-3,3,3′,3′-Tetra-methy-lindo-di-carbo-cyanine

Perchlorate (DiD), and sulforhodamine B (SRB) were purchased from Invitrogen. Lipid 106

molecules dissolved in chloroform were mixed in a glass tube based on the required ratio, then dried completely by nitrogen gas for subsequent purposes such as vesicle or nanodisc construction.

4.5.4 Reconstitution of SNARE incorporated vesicles

For the single-lipid mixing assay, the molar ratios of lipids were 61.9:15:20:1:2:0.1

(POPC:DOPS:cholesterol:DiI:PIP2:Biotin-DPPE) for the t-vesicles, and 74:5:20:1

(POPC:DOPS: cholesterol:DiD) for the v-vesicles, respectively. The lipid mixture was first completely dried and then hydrated using a dialysis buffer (25 mM HEPES, pH 7.4, 100 mM

KCl). After five freeze–thaw cycles, protein-free large unilamellar vesicles (~100 nm in diameter) were prepared by extrusion through a 100 nm polycarbonate filter (Whatman). For SNAREs and syt1 reconstitution in vesicles, the binary complex of syntaxin-1a and SNARE-25 were mixed with protein free t-vesicles, while VAMP-2 with or without syt1 was mixed with v-vesicles in the presence of 0.8% OG. The lipid-to-protein ratio was 200:1 in each case and the mixtures were diluted 2 times with the dialysis buffer. For v-vesicle reconstitution, a 1 mM EDTA was maintained in all processes34,42.

For the single-vesicle content-mixing assay with a small sulforhodamine B content indicator, the lipid compositions were the same as those used in the single lipid-mixing assay except that the fluorescent lipid dyes (DiI and DiD) were replaced by an equal amount of POPC.

The lipid mixtures were first completely dried and then hydrated with the dialysis buffer, but the one for v-vesicles was hydrated in the presence of additional 20 mM SRB. The overall vesicle 107

preparation and protein-reconstitution process was the same as above except that 20 mM SRB was maintained prior to each dialysis step for v-vesicle reconstitution. Free SRB was removed using the PD-10 desalting column (GE healthcare) after dialysis. Details for reconstitution have been discussed in our previous work34,42.

4.5.5 Assembly of SNARE-incorporated nanodiscs

To mimic the physiological lipid compositions of synaptic membranes, a mixture of

POPC, DOPS, cholesterol, PIP2, and biotin-PEG-DSPE with a molar ratio of 62.9:15:20:2:0.1 for t-discs and a mixture of POPC, DOPS, and cholesterol with a molar ratio of 75:5:20 for v-discs was dried and hydrated in a Tris150-EDTA buffer (10 mM Tris-HCl, 150 mM NaCl, 1 mM

EDTA, pH 7.4). 3 μl of each 50 mM lipid mixture was dissolved in 50 mM sodium cholate, then a binary complex of Cy5-labeled syntaxin 1a with SNAP-25 or Cy3-labeled VAMP-2 with or without syt1 was added. Finally, apoA1 proteins were added to each solubilized mixture. The molar ratio of lipid, SNARE(s), and apoA1 was 160:0.5:2 while the ratio of VAMP-2 and syt1 was 1:1. Self-assembly of SNARE-incorporated nanodiscs was initiated by rapid removal of sodium cholate through treating the sample with 50% (w/v) SM-2 Bio-Beads. The t- or v-discs were then purified through gel-filtration using a SuperdexTM200 GL 10/30 column (Amersham

Biosciences)36.

108

4.5.6 Single-molecule lipid mixing assay

T-vesicles with a final lipid concentration of 1 μM were flown into the imaging cell and immobilized on a PEG-coated surface through avidin-to-biotin conjugation. After two rounds of washing with a 200 l dialysis buffer, 1 μM cpx1 (or various cpx1 fragments) or equal volume of dialysis buffer were injected into the cell and incubated for 10 minutes at room temperature.

Then the v-vesicles (1 ~ 3 μM) with or without 1 μM cpx1 were injected into the flow chamber and incubated for another 30 min at 37 C to allow SNARE-mediated lipid mixing to occur.

After washing off free v-vesicles using the dialysis buffer, emission of each fluorophore intensity by an incoming 532 nm laser source was recorded from a random imaging area (4590 m2).

The first 80 frames were recorded by the 532 nm laser source to acquire the FRET efficiency and the following 10 frames were recorded using 635 nm to specifically discriminate bound v- vesicles from t-vesicles.

4.5.7 FRET acquisition from the trans-SNAREpin between nanodiscs

After coating the quartz surface with a solution of methoxy-polyethylene glycol (mPEG) and biotin-PEG molecules (100:1), the quartz slide was assembled into a flow chamber and coated with streptavidin (0.2 mg ml-1). 100~200 nM t-discs (containing Cy5-labeled sytnaxin 1a on the position of I203C or V241C) were immobilized on the surface, then 1 μM cpx1 (or various cpx1 fragmnets) or an equal volume of buffer were flown in and the sample was incubated for 10 min at room temperature. A five-time molar excess of v-discs (containing Cy3- 109

labeled VAMP2 on the position of Q33C or A72C without or with syt1) with 1 μM cpx1 or equal volume of buffer was then added to the imaging chamber and the sample was incubated for 25 min at room temperature to allow for trans-SNAREpin formation. A 532 nm wavelength laser was used to excite the donor fluorophore (Cy3) and obtain the emission intensities of donor only or the acceptor (Cy5) excited by the emission wavelength of the donor when two fluorophores are in close proximity. All TIR experiments were performed at room temperature in the presence of an oxygen scavenger system (0.4 % (w/v) using glucose (Sigma), 4 mM Trolox (Calbiochem),

1 mg/ml glucose oxidase (Sigma), and 0.04 mg/ml catalase (Calbiochem)) in a Tris150-EDTA buffer. The smFRET measurements were carried out on a prism-type TIR setup based on an inverted microscope (IX71, Olympus) with laser exposure time of 200 ms. Each Cy3 and Cy5 fluorescence signal from the chamber was collected by a water immersion lens (UPlanSApo

60x/1.20w, Olympus) and separated by a dichroic mirror (T660lpxr, chroma) with a threshold at a 635 nm wavelength. The two separated signals were then transferred to separate images on an electron-multiplying charged-coupled device camera (iXon DU897E, Andor Technology) as donor and acceptor signals, respectively. FRET efficiencies (E) were obtained using

( )

where IA and ID are the acceptor and donor fluorescence intensities.

110

4.5.8 Content mixing assay

The unlabeled t-vesicles (125 μM) were immobilized on the PEG-coated surface using avidin-to-biotin conjugation. Following several rounds of washing with 200 μL of dialysis buffer, the v-vesicles encapsulating 20 mM SRB (10 μM) with 2 uM cpx1 (or various cpx1 fragments) or equal amount of dialysis buffer were injected into the flow chamber and incubated for 10 min at room temperature to allow vesicle-vesicle docking. After washing out the unbound v-vesicles with a dialysis buffer containing 1mM EDTA, an additional 20 min of incubation was used to allow spontaneous content mixing. Using the motorized syringe pump, 500 μM Ca2+ was injected into the flow cell to produce vesicle pore opening while the trace of SRB intensity, excited by a 532 nm wavelength laser source, was recorded. A stepwise jump of de-quenched fluorescence SRB intensity was counted when a fusion pore was formed34,42. The details of TIR fluorescence-microscope imaging and single molecule data analysis have been reported in our previous work43.

111

4.6 Reference

1. Sollner, T. et al. SNAP receptors implicated in vesicle targeting and fusion. Nature 362, 318-324 (1993).

2. Weber, T. et al. SNAREpins: Minimal Machinery for Membrane Fusion. Cell 92, 759- 772 (1998).

3. Jahn, R. & Scheller, R.H. SNAREs [mdash] engines for membrane fusion. Nat Rev Mol Cell Biol 7, 631-643 (2006).

4. Martens, S. & McMahon, H.T. Mechanisms of membrane fusion: disparate players and common principles. Nat Rev Mol Cell Biol 9, 543-56 (2008).

5. Chernomordik, L.V. & Kozlov, M.M. Protein-lipid interplay in fusion and fission of biological membranes. Annu Rev Biochem 72, 175-207 (2003).

6. Pobbati, A.V., Stein, A. & Fasshauer, D. N- to C-Terminal SNARE Complex Assembly Promotes Rapid Membrane Fusion. Science 313, 673-676 (2006).

7. Lang, T. & Jahn, R. Core proteins of the secretory machinery. Handb Exp Pharmacol, 107-27 (2008).

8. Sudhof, T.C. & Rothman, J.E. Membrane fusion: grappling with SNARE and SM proteins. Science 323, 474-7 (2009).

9. McNew, J.A. et al. Close is not enough: SNARE-dependent membrane fusion requires an active mechanism that transduces force to membrane anchors. J Cell Biol 150, 105-17 (2000).

10. McMahon, H.T., Missler, M., Li, C. & Südhof, T.C. Complexins: Cytosolic proteins that regulate SNAP receptor function. Cell 83, 111-119 (1995).

11. Takahashi, S. et al. Identification of two highly homologous presynaptic proteins distinctly localized at the dendritic and somatic synapses. FEBS Letters 368, 455-460 (1995).

12. Chapman, E.R. How Does Synaptotagmin Trigger Neurotransmitter Release? Annual Review of Biochemistry 77, 615-641 (2008).

13. Kim, J.Y. et al. Solution single‐vesicle assay reveals PIP2‐mediated sequential actions of synaptotagmin‐1 on SNAREs. The EMBO Journal 31, 2144-2155 (2012). 112

14. Hui, E., Johnson, C.P., Yao, J., Dunning, F.M. & Chapman, E.R. Synaptotagmin- Mediated Bending of the Target Membrane Is a Critical Step in Ca2+-Regulated Fusion. Cell 138, 709-721 (2009).

15. Rizo, J. & Rosenmund, C. Synaptic vesicle fusion. Nat Struct Mol Biol 15, 665-674 (2008).

16. Südhof, T.C. THE SYNAPTIC VESICLE CYCLE. Annual Review of Neuroscience 27, 509-547 (2004).

17. Yoshihara, M. & Littleton, J.T. Synaptotagmin I Functions as a Calcium Sensor to Synchronize Neurotransmitter Release. Neuron 36, 897-908 (2002).

18. Chapman, E.R. Synaptotagmin: A Ca2+ sensor that triggers exocytosis? Nat Rev Mol Cell Biol 3, 498-508 (2002).

19. Martens, S. & McMahon, H.T. Mechanisms of membrane fusion: disparate players and common principles. Nat Rev Mol Cell Biol 9, 543-556 (2008).

20. Reim, K. et al. Complexins Regulate a Late Step in Ca2+-Dependent Neurotransmitter Release. Cell 104, 71-81 (2001).

21. Tang, J. et al. A Complexin/Synaptotagmin 1 Switch Controls Fast Synaptic Vesicle Exocytosis. Cell 126, 1175-1187 (2006).

22. Yoon, T.-Y. et al. Complexin and Ca2+ stimulate SNARE-mediated membrane fusion. Nat Struct Mol Biol 15, 707-713 (2008).

23. Lai, Y. et al. Fusion pore formation and expansion induced by Ca2+ and synaptotagmin 1. Proceedings of the National Academy of Sciences 110, 1333-1338 (2013).

24. Chen, X. et al. Three-Dimensional Structure of the Complexin/SNARE Complex. Neuron 33, 397-409 (2002).

25. Bowen, M.E., Weninger, K., Ernst, J., Chu, S. & Brunger, A.T. Single-Molecule Studies of Synaptotagmin and Complexin Binding to the SNARE Complex. Biophysical Journal 89, 690-702 (2005).

26. Weninger, K., Bowen, M.E., Choi, U.B., Chu, S. & Brunger, A.T. Accessory Proteins Stabilize the Acceptor Complex for Synaptobrevin, the 1:1 Syntaxin/SNAP-25 Complex. Structure 16, 308-320 (2008).

113

27. Lu, B., Song, S. & Shin, Y.-K. Accessory α-Helix of Complexin I Can Displace VAMP2 Locally in the Complexin–SNARE Quaternary Complex. Journal of Molecular Biology 396, 602-609 (2010).

28. Yang, X., Kaeser-Woo, Y.J., Pang, Z.P., Xu, W. & Südhof, T.C. Complexin Clamps Asynchronous Release by Blocking a Secondary Ca2+ Sensor via Its Accessory α Helix. Neuron 68, 907-920 (2010).

29. Seiler, F., Malsam, J., Krause, J.M. & Söllner, T.H. A role of complexin–lipid interactions in membrane fusion. FEBS Letters 583, 2343-2348 (2009).

30. Kaeser-Woo, Y.J., Yang, X. & Südhof, T.C. C-Terminal Complexin Sequence Is Selectively Required for Clamping and Priming But Not for Ca2+ Triggering of Synaptic Exocytosis. The Journal of Neuroscience 32, 2877-2885 (2012).

31. Xue, M. et al. Binding of the complexin N terminus to the SNARE complex potentiates synaptic-vesicle fusogenicity. Nat Struct Mol Biol 17, 568-575 (2010).

32. Xue, M. et al. Distinct domains of complexin I differentially regulate neurotransmitter release. Nat Struct Mol Biol 14, 949-958 (2007).

33. Südhof, Thomas C. Neurotransmitter Release: The Last Millisecond in the Life of a Synaptic Vesicle. Neuron 80, 675-690 (2013).

34. Lai, Y. et al. Fusion pore formation and expansion induced by Ca2+ and synaptotagmin 1. Proc Natl Acad Sci U S A 110, 1333-8 (2013).

35. Kyoung, M. et al. In vitro system capable of differentiating fast Ca2+-triggered content mixing from lipid exchange for mechanistic studies of neurotransmitter release. Proceedings of the National Academy of Sciences 108, E304-E313 (2011).

36. Shin, J., Lou, X., Kweon, D.H. & Shin, Y.K. Multiple conformations of a single SNAREpin between two nanodisc membranes reveal diverse pre-fusion states. Biochemical Journal 459, 95-102 (2014).

37. Chernomordik, L.V. & Kozlov, M.M. Membrane Hemifusion: Crossing a Chasm in Two Leaps. Cell 123, 375-382 (2005).

38. Yoon, T.Y. & Shin, Y.K. Progress in understanding the neuronal SNARE function and its regulation. Cellular and Molecular Life Sciences 66, 460-469 (2009).

39. Maximov, A., Tang, J., Yang, X., Pang, Z.P. & Südhof, T.C. Complexin Controls the Force Transfer from SNARE Complexes to Membranes in Fusion. Science 323, 516-521 (2009). 114

40. Shin, J., Lou, X., Kweon, D.H. & Shin, Y.K. Multiple conformations of a single SNAREpin between two nanodisc membranes reveal diverse pre-fusion states. Biochem J 459, 95-102 (2014).

41. Lai, Y., Lou, X., Wang, C., Xia, T. & Tong, J. Synaptotagmin 1 and Ca2+ drive trans SNARE zippering. Sci Rep 4, 4575 (2014).

42. Lai, Y., Lou, X., Jho, Y., Yoon, T.Y. & Shin, Y.K. The synaptotagmin 1 linker may function as an electrostatic zipper that opens for docking but closes for fusion pore opening. Biochem J 456, 25-33 (2013).

43. Diao, J. et al. A single vesicle-vesicle fusion assay for in vitro studies of SNAREs and accessory proteins. Nat Protoc 7, 921-34 (2012).

115

4.7 Figures and Legends

a b

Figure 1) Cpx1 increases Ca2+ triggered content mixing efficiency.

(a) Schematic diagram of the in vitro content mixing setup. T-vesicles carrying the binary complex of syntaxin-1a/SNAP-25 were immobilized on the flow cell by the conjugation of streptavidin (purple, cross-shaped) and biotin (yellow square). The concentration dependent self- quenching fluorophore, sulforhodamine B (dark purple), in the v-vesicle emits increased intensity (light orange) when it is diluted in larger volume by opening a pore with the immobilized empty t-vesicle. (b) Quantitative comparison of cumulative time content mixing percentage of the total docked population in the presence (red) or absence of cpx1 (black).

116

a b

e d

Figure 2) Effect of cpx1 on the conformation of the trans-SNAREpin detected by smFRET.

(a) Purified cpx1 fragments, which truncate specific regions. Full-length cpx1 contains; N-terminal region (Nt, red), accessory a helix (accH, white), central a helix (cntH, yellow), and C-terminal region (Ct, broken green) while the other truncated versions miss one or two distinct parts. (b) smFRET setup for transSNAREpin conformation study using fluorescent dye-labeled syntaxin-1a and VAMP-2. I203C or V241C of syntaxin-1a was labeled with Cy5 (light blue saw- tooth circle) and Q33C or A72C of VAMP-2 was labeled with Cy3 (light blue saw-tooth circle). The Habc domain (white) of syntaxin 1a is depicted broken to indicate the longer length than the SNARE motif (red). The t-discs carrying Cy5-labeled syntaxin-1a (red and white) and SNAP-25 117

(green) was immobilized on the quartz surface through biotin (yellow square) and streptavidin (purple, cross-shaped) conjugation and then allowed to interact and form the trans-SNAREpin with the v-discs carrying Cy3-labeled VAMP-2 (blue) with or without syt1 (brown). (c) FRET distribution of trans-SNAREpin at the N- (left) and C-terminal part (right) in the presence of syt1 (top) and additional cpx1 (bottom). (d) FRET distribution of CC pair in the presence of cpx1 fragments; cpx27-134, cpx41-134, and cpx1-75. (e) Percentage of paired v-disc populations with immobilized t-discs in the presence of cpx1 fragments.

118

a b

Figure 3) Effect of cpx1 fragment on SNARE-mediated single vesicle mixing assay.

(a) Schematic diagram of the single-vesicle docking and lipid mixing assay. T-vesicles reconstituted with syntaxin-1a/SNAP-25 were immobilized on the surface of the flow cell via the conjugation of streptavidin (purple, crossed shape) and biotin (yellow square). V-vesicles reconstituted with VAMP-2 or VAMP-2 together with syt1 (VAMP2:syt1=1:1) were flown into the flow cell with or without 1 μM cpx1 to allow docking and lipid mixing event. (b) FRET distribution of SNARE-mediated single lipid mixing assay excluding (top panel) or including syt1 (bottom panel) in v-vesicles. V-vesicle incorporated syt1 itself was able to enhance lipid 119

mixing efficiency (left panel). In the presence of cpx1, lipid mixing events were accelerated independent of syt1 presence (right panel). (c) N-terminally truncated cpx1 fragments (cpx27-134 and cpx41-134) increased lipid mixing as much as full-length cpx1 whereas C-terminally truncated cpx1 (cpx1-75) showed no effect on the mixing process.

120

a b

Figure 4) Effect of specific region truncated cpx1 on the Ca2+ triggered content mixing efficiency.

(a) Cumulative time content mixing percentage of the total docked populations in the presence of various cpx1 fragment versions. N-terminally truncated cpx27-134 and cpx41-134 did not affect the efficiency of fusion pore opening while full-length cpx1 and C-terminally truncated cpx1-75 enhanced it. (b) The normalized docking populations of the v-vesicles to the immobilized t-vesicles in the presence of various cpx1 fragments. The full-length cpx1 and N- terminally truncated fragments (cpx27-134 and cpx41-134) increased docking probability while C- terminally truncated cpx1-75 reduced it.

121

4.8 Supporting Information

Content: Supplementary Figure and Legends S1-S2

Figure S1) Effect of cpx1 on the SNAREpin formation independent of syt1.

FRET distribution of trans-SNAREpin between nanodiscs without syt1 in the absence (a) or presence of 1 uM cpx1 (b).

122

Figure S2) Effect of cpx1 fragments on the docking population of v-vesicles to the immobilized t-vesicles.

The averaged numbers (from three different experiment sets) of docked v-vesicles on the immobilized t-vesicles indicate the increased binding affinity between two vesicles by full-length cpx1 and N-terminally truncated cpx1 fragments (cpx27-134 and cpx41-134) and decreased binding by C-terminal truncated cpx1-75.

123

CHAPTER 5: GENERAL SUMMARY

5.1 General conclusion

It has been widely accepted that SNARE proteins are the minimal machinery required to merge opposing membranes and release contents from one site to the other1,2. In neuroexocytosis,

SNAREs form energetically a stable four-helix complex contributing one helix each from syntaxin-1a and VAMP-2 and two helices from SNAP-25. The force generated by the complex formation is believed to overcome electrostatic repulsion between apposed membranes and ultimately fuse them into a single layer3,4. Conventional lipid-mixing experiments have proven that SNAREs are fusogenic molecules5, but fusion events in physiology must be properly controlled to transfer neuro-signals at just the right time. Although isolated SNAREs spontaneously form the extended cis-SNARE complex, there is an intermediate complex state, the so-called trans-SNAREpin conformation, when VAMP-2 and syntaxin-1a have been anchored to each membrane6-8. Presynaptic regulatory proteins, mainly syt1, Munc18-1/Munc13, and cpx1 play the role of shifting the equilibrium9.

Through this dissertation we have paved the way for understanding the structure information of trans-SNAREpin sandwiched in nanodiscs8. Furthermore, this new platform, together with the vesicle-vesicle lipid mixing and content release assay in a single molecular level, allowed examination of the cooperative role of presynaptic regulatory proteins during

SNARE-mediated membrane fusion. 124

trans-SNAREpin has been investigated for decades but the isolation of membrane- anchored trans-SNAREpin complexes was not possible because of difficulty of sample preparation. Recently, Gao et al. and Min et al. contributed knowledge of the structure information of trans-SNAREpin using external force, but crucial components, lipid membranes, were not involved in those investigations. We assembled the intermediated trans-SNAREpin anchoring each transmembrane domain of VAMP-2 and syntaxin-1a to opposing nanodiscs8.

Unpaired electron spins labeled at the motif region of VAMP-2 suggested through the EPR experiment that the conserved ionic layer was the breaking point of trans-SNAREpin.

Furthermore, the study of fluorescence-labeled VAMP-2 and syntaxin-1a at the N- and C- terminal SNARE motifs with smFRET convinced us that there were multiple conformations of trans-SNAREpin.

By recruiting trans-SNAREpin trapped in nanodiscs along with well-developed vesicle- vesicle lipid and content mixing assays, we explored the role of the presynaptic regulatory proteins syt1, Munc18-1, and cpx110,11. Early in the evolution of the process, Munc18-1 expedited a SNARE-mediated membrane-fusion progress, probably by stabilization of four- helical SNARE complexes12-15. This conserved acceleratory function of Munc18-1 was maintained in our investigation using a SNARE-mediated lipid-mixing assay. Additionally, fluorescence-labeled trans-SNAREpin revealed that Munc18-1 also aided full SNARE complex formation. Intriguingly, however, these effects of Munc18-1 disappeared in the presence of syt1, which also interacts with SNARE complex16-18. Therefore, content-mixing events, indicating the final step of exocytosis, were not affected by Munc18-1 when syt1 was concurrently presented. 125

We concluded that Munc18-1 was defeated by syt1 and lost its place in the SNARE-functioning zone.

Our next interesting activity regarding the regulatory system for neuroexocytosis was replenishment of Munc18-1 as a supporter of SNARE assembly and fusion event. Without the calcium influx, cpx1 hastened core SNARE complex formation and lipid mixing in the presence or absence of syt1 without opening a pore between two membranes. We hypothesized that the hemi-fused membrane opened a pore and experienced content mixing upon arrival of calcium with the support of cpx1. Truncated cpx1 fragments in these investigations revealed that cntH and Ct of cpx1 were required to assemble the robust core SNARE complex and also to potentially form the hemi-fusion state19. Fascinatingly, N-terminally truncated cpx1 fragments did not increase the spontaneous content-mixing events although they maintained hypothesized hemi-fusion. Only full-length cpx1 or C-terminally truncated cpx1 boosted up the content- mixing probability. The results led us to visualize the role of cpx1 Nt as an active assistant for the pore-opening process initiated by syt1 and calcium.

As the consequence of the evolution in SNARE-mediated neurotransmitter release toward explicit transmission, synaptic vesicles recruited precise regulation systems, including syt1,

Munc18-1, and cpx1. Because the preference of Munc18-1 is to grab the four-helical bundle, it binds to closed syntaxin-1a or stabilized ternary SNARE complex. The role of Munc18-1 in neurons might be the isolation of closed syntaxin-1a from SNAP-25. Munc13 will be included to release Munc18-1 and initiate the binary complex formation15. Cpx1 assumes the role of

Munc18-1 as a fastener of the SNARE complex in the presence of syt1, thereby producing a 126

hemi-fused membrane until the Ca2+ influx. At the required instant, the local concentration of

Ca2+ is increased by uptake via Ca2+ channels then leads the conformational change of syt1 and initiates fully-extended SNARE complex formation to open a pore. The final stage is promoted by Nt of cpx1.

In the hemi-fused membrane state, the uncomplexed parts of SNAREs are the membrane- proximal and transmembrane region of each VAMP-2 and syntaxin-1a. The last-minute event in pore formation is the assembly of these two C-terminal parts, and we anticipate that Nt of cpx1 scaffolds this final event.

127

5.2 References

1. McNew, J.A. et al. Close is not enough: SNARE-dependent membrane fusion requires an active mechanism that transduces force to membrane anchors. J Cell Biol 150, 105-17 (2000).

2. Weber, T. et al. SNAREpins: Minimal Machinery for Membrane Fusion. Cell 92, 759- 772 (1998).

3. Sutton, R.B., Fasshauer, D., Jahn, R. & Brunger, A.T. Crystal structure of a SNARE complex involved in synaptic exocytosis at 2.4 A resolution. Nature 395, 347-53 (1998).

4. Stein, A., Weber, G., Wahl, M.C. & Jahn, R. Helical extension of the neuronal SNARE complex into the membrane. Nature 460, 525-8 (2009).

5. Jérôme, V. & Pessin, J. SNARE-Mediated Fusion of LIposomes. in Membrane Trafficking, Vol. 457 (ed. Vancura, A.) 241-251 (Humana Press, 2008).

6. Gao, Y. et al. Single Reconstituted Neuronal SNARE Complexes Zipper in Three Distinct Stages. Science 337, 1340-1343 (2012).

7. Min, D. et al. Mechanical unzipping and rezipping of a single SNARE complex reveals hysteresis as a force-generating mechanism. Nat Commun 4, 1705 (2013).

8. Shin, J., Lou, X., Kweon, D.H. & Shin, Y.K. Multiple conformations of a single SNAREpin between two nanodisc membranes reveal diverse pre-fusion states. Biochemical Journal 459, 95-102 (2014).

9. Südhof, Thomas C. Neurotransmitter Release: The Last Millisecond in the Life of a Synaptic Vesicle. Neuron 80, 675-690 (2013).

10. Kyoung, M. et al. In vitro system capable of differentiating fast Ca2+-triggered content mixing from lipid exchange for mechanistic studies of neurotransmitter release. Proceedings of the National Academy of Sciences 108, E304-E313 (2011).

11. Lai, Y. et al. Fusion pore formation and expansion induced by Ca2+ and synaptotagmin 1. Proceedings of the National Academy of Sciences 110, 1333-1338 (2013).

12. Shen, J., Tareste, D.C., Paumet, F., Rothman, J.E. & Melia, T.J. Selective activation of cognate SNAREpins by Sec1/Munc18 proteins. Cell 128, 183-95 (2007).

13. Deak, F. et al. Munc18-1 binding to the neuronal SNARE complex controls synaptic vesicle priming. J Cell Biol 184, 751-64 (2009). 128

14. Diao, J. et al. Single-Vesicle Fusion Assay Reveals Munc18-1 Binding to the SNARE Core Is Sufficient for Stimulating Membrane Fusion. ACS Chem Neurosci 1, 168-174 (2010).

15. Ma, C., Su, L., Seven, A.B., Xu, Y. & Rizo, J. Reconstitution of the Vital Functions of Munc18 and Munc13 in Neurotransmitter Release. Science 339, 421-425 (2013).

16. Chapman, E.R. How Does Synaptotagmin Trigger Neurotransmitter Release? Annual Review of Biochemistry 77, 615-641 (2008).

17. Kim, J.Y. et al. Solution single‐vesicle assay reveals PIP2‐mediated sequential actions of synaptotagmin‐1 on SNAREs. The EMBO Journal 31, 2144-2155 (2012).

18. Hui, E., Johnson, C.P., Yao, J., Dunning, F.M. & Chapman, E.R. Synaptotagmin- Mediated Bending of the Target Membrane Is a Critical Step in Ca2+-Regulated Fusion. Cell 138, 709-721 (2009).

19. Chernomordik, L.V. & Kozlov, M.M. Membrane Hemifusion: Crossing a Chasm in Two Leaps. Cell 123, 375-382 (2005).

129

ACKNOWLEDGEMENT

I would like to express the gratitude to my advisor Professor Dr. Yeon-Kyun Shin, you have been an excellent mentor for me. I would like to appreciate you for encouraging my research and for allowing me to grow as an independent research scientist.

I would also like to thank all my committee members, Dr. Guru Rao, Dr. Michael

Shogren-Knaak, Dr. Scott Nelson, and Dr. Edward Yu for serving as my program of study committee members. I also want to thank you for your valuable comments and suggestions during my graduate course. In addition, special thanks to administrative staffs of the department,

Kelly Yohnke, Connie Garnett, Pete Lelonek, Desi Gunning, and Stacey Poling. All of you have supported in different ways to make me survived in Iowa state university and achieved Ph.D degree finally.

I would like to express my grateful feeling to my current lab mates, Xiaochu Lou, Jae-

Kyun Song, Linxaing Yin, Yicheng (Zoie) Zhu, and Brenden Hawk and also past mates, Dr.

Jiansong Tong, Dr. Bin Lu, Dr. Jinyoung Chang, Dr. Shuang Song, Dr. Ying Lai, and Dr. Sunae

Kim for sharing scientific ideas and supporting experiment skills. Without all of you, I could not have successfully settled down and made good progress in Dr. Shin’s lab.

A special thanks to my family. Words cannot express how grateful I am to my mother-in law, Hyo Ryeon Kim, father-in law, Ki Heung Yang, my mother, Young Ja Ko, and father, Ki

Young Shin for supporting me in the all aspects. I would also like to thank my brother-in law,

Hyun Suk Yang. You have been always showing your enthusiasm in your career and it 130

stimulates me to realize how valuable my work is also. My siblings, Jae Gwan Shin and Sun Hee

Shin couldn’t be replaced by any other people. I always wish you have endless happiness with your family: Eun Jung Park, Su Ho Shin, Su Jeong Shin, Tae Ho Shin, Jae Hong Kim, and

Terrian Gunwoo Kim.

At the end I would like to express the biggest appreciation to my beloved wife Dr.

Yoosoo Yang who has willingly been spending her valuable time with me. I promise you that I will be the indispensable partner of you for the whole life.