"Medicine, Faculty of"@en . "Biochemistry and Molecular Biology, Department of"@en . "DSpace"@en . "UBCV"@en . "Liu, Xiaohu"@en . "2010-06-11T20:06:42Z"@en . "2010"@en . "Master of Science - MSc"@en . "University of British Columbia"@en . "Voltage-gated calcium channels (Cay) have functions ranging from regulating\nrelease of hormones and neurotransmitters, generating cardiac action potentials, and\nexcitation-contraction coupling. At nerve terminals, N- and P/Q- type Cavs convert the\naction potential into aC\u00C2\u00B2\u00E2\u0081\u00BA signal that in turn triggers neurotransmitter release.\nNeurotransmitter release requires several components, such as SNARE proteins.\nSNAREs, as well as many other presynaptic proteins, can interact with Cavs and inhibit\nthem by increasing their inactivation. The interaction is localized in the intracellular loop\nbetween domains II and III of the CL 1 subunit, in a domain termed \u00E2\u0080\u0098synprint\u00E2\u0080\u0099 (synaptic\nprotein interaction site). In this study, we tried to solve the structure of the synprint site\nby crystallography. To date, long needle-shape crystals were obtained; however, the\nquality of these crystals was not good enough for X-ray diffraction. in addition,\nisothermal titration calorimetry (ITC) was used to determine the interaction between\nSNARE protein syntaxinlA and the synprint site. It turned out that not any binding was\ndetected, suggesting that the interaction between SNARE proteins and the presynaptic\nCas, if at all present, is weak."@en . "https://circle.library.ubc.ca/rest/handle/2429/25586?expand=metadata"@en . "CRYSTALLOGRAPHIC INVESTIGATION AND CHARACTERIZATION OF THE INTERACTION BETWEEN PRESYNAPTIC VOLTAGE-GATED CALCIUM CHANNELS AND SNARE PROTEINS by Xiaohu Liu B.Sc., The University of British Columbia, 2007 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Biochemistry and Molecular Biology) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) January 2010 \u00C2\u00A9 Xiaohu Liu, 2010 Abstract Voltage-gated calcium channels (Cay) have functions ranging from regulating release of hormones and neurotransmitters, generating cardiac action potentials, and excitation-contraction coupling. At nerve terminals, N- and P/Q- type Cavs convert the action potential into a Ca2signal that in turn triggers neurotransmitter release. Neurotransmitter release requires several components, such as SNARE proteins. SNAREs, as well as many other presynaptic proteins, can interact with Cavs and inhibit them by increasing their inactivation. The interaction is localized in the intracellular loop between domains II and III of the CL 1 subunit, in a domain termed \u00E2\u0080\u0098synprint\u00E2\u0080\u0099 (synaptic protein interaction site). In this study, we tried to solve the structure of the synprint site by crystallography. To date, long needle-shape crystals were obtained; however, the quality of these crystals was not good enough for X-ray diffraction. in addition, isothermal titration calorimetry (ITC) was used to determine the interaction between SNARE protein syntaxinlA and the synprint site. It turned out that not any binding was detected, suggesting that the interaction between SNARE proteins and the presynaptic Cas, if at all present, is weak. 11 Table of Contents Abstract.ii Table of Contents iii List of Tables v List of Figures vi List of Abbreviations viii Acknowledgements X 1. Introduction 1 1.1 Voltage-gated calcium channels and synaptic transmission 1 1.1.1 Overview of voltage-gated calcium channels I 1.1.2 Presynaptic proteins in the synaptic vesicle exocytosis 12 1.1.3 Regulation of presynaptic Cas by SNARE proteins 16 1.2 Crystallization 24 1.2.1 Overview of protein crystallization 24 1.2.2 Vapor diffusion crystallization 25 1.2.3 Counter diffusion crystallization 27 1.2.4 Dialysis crystallization 27 1.2.5 Other crystallization techniques 28 2 Objectives 30 3. Materials and Methods 31 3.1 Cloning 31 3.1.1 Cloning of synprint domains from rabbit Ca2.l and bovine Ca2.2 31 3.1.2 Mutagenesis of the synprint constructs 36 3.1.3 Cloning of the fusion proteins 36 3.1.4 Cloning of syntaxinlA constructs 37 3.2 Protein expression and purification 38 3.2.1 Protein expression 38 3.2.2 Purification of the synprint domains from rabbit Ca2. 1 and bovine Ca2.2 ... 38 3.2.3 Purification of BovalBsynA8 mutants 41 111 3.2.4 Purification of MBP (GST)-(linker)-synA8EE fusion proteins 41 3.2.5 Purification of MBP (GST)-(linker)-synA8EE-His fusion proteins 42 3.2.6 Purification ofstx1189, synAl and synBi for TTC 42 3.2.7 Expression and purification of stxl9o2M for ITC 43 3.3 Crystallization 43 3.3.1 Vapor diffusion crystallization 43 3.3.2 Counter diffusion erystallzation 44 3.3.3 Dialysis crystallization 44 3.3.4 Seeding 44 3.4 Isothermal titration calorimetry (ITC) 45 3.5 Circular dichroism (CD) 45 4. Results 47 4.1 Crystallization of the synprint site from Ca2.1 and Ca2.2 47 4.1.1 Designing different synprint constructs from Ca2.1 and Ca2.2 47 4.1.2 Purification of different synprint constructs from Ca2. 1 and Ca2.2 48 4.1.3 CD experiment of the N- and C-terminal of the synprint site 51 4.1.4 Crystallization of different synprint constructs from Ca2. I and Ca2.2 51 4.2 Crystallization of synprint domains with surface engineering 54 4.2.1 Purification of Bovot 1 BsynA8C784A/E794A/E795A 55 4.2.2 Crystallization of synA8EE 57 4.2.3 Crystallization of synA8EE using different crystallization methods 61 4.3 Crystallization of fusion proteins 63 4.3.1 Purification of MBP-(linker)-synA8EE-His and GST-(linker)-synA8EE-His. 65 4.3.2 Crystallization of MBP-(linker)-synA8EE-His 65 4.4 Interaction between the synprint site and syntaxinlA 69 5. Discussion 81 5.1 Crystallization 81 5.2 Interaction between the synprint site and syntaxin 1A 83 References 86 iv List of Tables Table 1 Subunit composition and function of Cas Types 2 Table 2 Oligonucleotides used in this study 32 Table 3 Buffers of all the chromatography columns 40 V List of Figures Figure 1 Architecture of the voltage-gated calcium channel 4 Figure 2 The structure of the core of Cavf3 subunit-AID complex 7 Figure 3 The interaction sites of different regulatory proteins of Cav2 ul subunit 9 Figure 4 Structures of CaV1 and CaV2-Ca2+/CaM-IQ domain complexes 11 Figure 5 Cartoon representations of neuronal SNARE proteins and the structure of the SNARE complex 18 Figure 6 The structure of synaptotagmin 19 Figure 7 Sequence of the synprint site and interaction with SNARE proteins 22 Figure 8 Phase diagram of crystallization and vapor diffusion crystallization setups 26 Figure 9 Purification of the synprint domains 49 Figure 10 CD spectra of synAl (N-terminal) and synBi (C-terminal) 52 Figure 11 Spherulites and small needle-like crystals from crystallization trials of the synprint domains 53 Figure 12 Stability tests 56 Figure 13 Purification of synA8EE 58 Figure 14 The needle-like crystals obtained from synA8EE 60 Figure 15 Mass spectrum of synA8EE and the result of seeding experiment 62 Figure 16 Results of dialysis and counter diffusion crystallization 64 Figure 17 Hiload Q results of MBP-tagged synA8EE and GST-tagged synA8EE 66 Figure 18 Purification of MBP (GST)-linker-synA8EE-His 67 Figure 19 The needle-like crystals obtained from MBP-5A-synA8EE-His 70 Figure 20 The cartoon representation of the synprint site and syntaxin 1A 72 Figure 21 Purification of stxl-189 73 Figure 22 Purification of synAl 74 Figure 23 Purification of synB I 75 Figure 24 Purification of stxl9O-264 76 vi Figure 25 The interaction between domains of syntaxin TA and parts of the synprint site using ITC 77 Figure 26 Mass spectrometry of synA 1 and synB 1 and ITC result in the presence of glycerol 80 vii List of Abbreviations AH Change in Enthalpy AS Change in Entropy f3ME Beta-mercaptoethanol Boy Bovine CaM Calmodulin CaMKII Ca2+/calmoduiin-dependent protein kinase II Ca Voltage-gated calcium channel CD Circular dichroism CDI Ca2-dependent inactivation Co-IP Co-immunoprecipitation DMSO Dimethyl sulfoxide E. coli Escherichia coil ESI Electrospray ionization GST Giutathione-S-Transferase GPCR G-Protein-Coupied Receptor IPTG Isopropyl 13-D-1 Thiogalactopyranoside ITC Isothermal Titration Calorimetry Kd Dissociation Constant kDa kilo-Dolton LIC Ligation independent cloning MBP Maltose Binding Protein viii List of Abbreviations (Continued) MOLDI-TOF Matrix-Assisted Laser Desorption/Ionization-Time of Flight MWCO Molecular Weight Cut-Off NMR Nuclear Magnetic Resonance PCR Polymerase chain reaction p1 Isoelectric Point PKA Protein kinase A PKC Protein kinase C Rab Rabbit SDS-PAGE Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophorosis SNARE soluble N-ethylmaleimide-sensitive factor attachment protein receptors Stx Syntaxin 1A Synprint Synaptic protein interaction site synA8 Bovut 1 BsynA8C784A/E794A/E795A TEV Tobacco etch virus VDI Voltage-dependent inactivation ix Acknowledgements I would like to thank my supervisor, Dr. Filip Van Petegem, for giving me the opportunity to work on this project and all his help during the two and half years. I would also like to thank all the previous and present lab memebers for help. Finally, I would like to thank friends from Molday lab and Strynadka lab for kindly helping me with enzymes and antibodies, and friends from Accili lab and Ahern lab for allowing me to use some chemicals and instruments. x 1. Introduction 1.1 Voltage-gated calcium channels and synaptic transmission 1.1.1 Overview of voltage-gated calcium channels Virtually all excitable cells express plasma membrane voltage-gated calcium channels (Cas) that convert the electrical signals into the cellular activities. Rapid entry of Ca2 through Cas upon membrane depolarization regulates a wide range of physiological processes such as release of neurotransmitters, excitation-contraction coupling, hormone secretion, gene expression, and activation of calcium dependent enzymes (Sutton et al. 1999, Dolmetsch et al. 2001, Reid et al. 2003). To date, five different types of Cas have been identified based on their specific physiological functions, electrophysiological and pharmacological properties, and subcellular distributions (Table 1). They can also be grouped into the high-voltage activated channels (HVA) such as L-, PIQ-, N-, R-type channels which require large membrane depolarization to open and the low-voltage activated channels (LVA) such as the T-type channels that activate in response to smaller membrane depolarization (Catterall 2000). Channels can also undergo different states. Besides the simple \u00E2\u0080\u0098open\u00E2\u0080\u0099 and \u00E2\u0080\u0098closed\u00E2\u0080\u0099 states, the channels can reside in an \u00E2\u0080\u0098inactivated\u00E2\u0080\u0099 state, a state that prevents passage of ions even when a depolarizing signal is present. When the plasma membrane is depolarized, Cas open to allow the entry of Ca2,while they are closed as the plasma membrane is hyperpolarized. Channel inactivation typically occurs during membrane C a 2\u00C2\u00B1 c u rr e n t Pr im ar y Pr ev io us n a m e Sp ec ifi c C a 2+ c ha nn el ty pe lo cu liz fi on s o f n \u00E2\u0080\u0099jy sub un itc bl oc ke r Fu nc tio ns B C av l.1 L Sk el et al m u sc le s D H Ps Ex ci ta tio n- co nt ra ct io n co u pl in g Ca lc iu m ho m eo sta sis G en e re gu la tio n C a y 1.2 L Ca rd ia c m u sc le a D H Ps Ex ci ta tio n- co nt ra ct io n co u pl in g En do cr in e ce lls F- fo rm o n e se cr et io n N eu ro ns G en e re gu la tio n Ca y 1.3 L En do cr in e ce lls a1 0 D H Ps H or m on e se cr et io n N eu ro ns G en e re u la tio n C av l.4 L Re tin a a jF To ni c rie ur ot ra ns ni ilt er re le as e Ca v2 .I Pi Q N er ve te rm in als c v w -A ga to xi n N eu w lra !Is lIl ilt er re le as e D en dr ite s D en dr iti c C a 2+ tr an sie nt s C a\ 2. 2 N N er ve te rn iin al s a m a)- CT x-G VI A N eu ro tra ns m itt er re le as e De tid rit es D en dr iti c C a 2+ tr an sie nt s Ca v2 .3 R Ce ll bo di es a lE N on e C a 2td ep en d en t ac tio n po te nt ia is D en dr ite s N e rv e N eu ru ir an si ri it te r re le as e Te rm in als C av il T Ca rd ia c m u sc le tr iG N on e Re pe tit iv e rin g Sk el et al m u sc le Ne ur on s Ca v3 .2 T Ca rd ia c m u sc le a IR N on e Re pe tit iv e rin g N eu ro ns N eu ro ns C av i3 T a11 N on e Re pe tit iv e r in g depolarization, thus reducing channel availability. inactivation is an important process for channels to prevent excess Ca2 to enter the cells when the channels are open. Cas experience two types of inactivation: the intrinsic voltage-dependent inactivation (VDI) and the influx of calcium\u00E2\u0080\u0094triggeredCa2tdependent inactivation (CDI). Both inactivation processes are regulated by different modulators (such as auxiliary Caf3 subunits and calmodulin) and different signaling pathways (such as GPCR-mediated pathways and cAMP-dependent PKA) (Catterall 2000, Budde et al 2002, Halling et al 2006, Findeisen et al 2009). 1.1.1.1 Molecular structure of voltage-gated calcium channel HVA channels consist of the principle pore-forming ct subunit (Cacti) in association with two auxiliary subunits: a membrane-anchored, predominantly extracellular, disulfide-linked cL2\u00C3\u00B6 subunit (CacL2) and a cytoplasmic f3 subunit (Caf3). In addition, though it is still controversial, some HVA channels (e.g. Ca1. 1) are found to have another auxiliary transmembrane y subunit (Ca\u00E2\u0080\u0099y) (Kang and Campbell 2003; Moss et al 2002) (Figure 1). LVA channels also have an a1 subunit, but whether other auxiliary subunits are also present is still uncertain. However, recent studies show the possibility of 13 subunit and a2\u00C3\u00B6 subunit existing in T-type channels (Walsh et al 2009). The largest Caai subunit is approximately 1 9OkDa. The Caai subunit is the core of all functional Cas, defines the channel subtypes and incorporates the ion conduction pore, the voltage sensor, the gating apparatus and several sites that allow channel modulation (Catterall 2000, Kisilevsky and Zamponi 2008). The Caa1 subunit is composed of four 3 extra-cellular cytoplasm Figure 1 Architecture of the voltage-gated calcium channel. The pore-forming Caxl subunit consists of four homologous domains (I - IV), connected by cytosolic loops and flanked by cytosolic N- and C-termini. The cytoplasmic CaP subunit associates with the Caal subunit through the I\u00E2\u0080\u0094I! linker alpha interaction domain (AID). The Cact26 subunit is composed of the transmembrane 6 subunit and heavily glycosylated extracellular a2 subunit, which are linked by disulfide bonds. The Cay subunit is another transmembrane protein with cytoplasmic N- and C-termini. (Kisilevsky and Zamponi 2008) II Ill IV ? 1 7 NH3 4 homologous domains I-IV, each of which consists of six transmembrane segments (Si S6). Si to S4 are considered to be the voltage-sensing domain, whereas S5, S6 and the P loop between them are believed to form the conduction pore and confer Ca2 selectivity (Figure 1). The S4 segment contains characteristic positively charged residues (Arg and Lys). Upon membrane depolarization, these residues sense the potential change, and the S4 segment undergoes rotation and translation. Via the S4-S5 linker helix, the S4 motion exerts force on the major pore-forming S6 segment to \u00E2\u0080\u009Cpull\u00E2\u0080\u009D the channel open (Borjesson and Elinder 2008). The four pore-lining P loops between S5 and S6 contain an EEEE locus, which is believed to form the selectivity filter for Ca2 ions (Sather and McCleskey 2003). In addition, the cytoplasmic loops between domain I, II, III and IV (I-TI loop, Il-ITT loop and Ill-TV loop, respectively), and the N- and C-terminal regions provide crucial sites for regulatory and adaptor proteins, as discussed later. So far, ten different ci subunits have been identified, divided into three families (Cay 1, Ca2 and Ca3) (Catterall 2000, Lipscombe et al 2002, Bell et al 2004). The auxiliary intracellular Caf3 subunit, approximately 55 kDa, has multiple effects on Cas by increasing the channel expression, altering the membrane trafficking, and modulating the biophysical properties on the channel. This is also the only calcium channel subunit for which crystal structures are available (Chen et al 2004, Van Petegem et al 2004). To date, four different genes coding for CaI3 subunits have been cloned (Caf3 1, 132, 133 and 134). All the Caf3 subunits possess a common-core structure, which consist of a Src homology 3 (SH3) domain and a guanylate kinase (GK) domain (Figure 2). The GK domain of Ca13 associates with CacL1 through a high-affinity interaction with a region in the I-TI loop of Cacti (Figure 2) (Van Petegem et al 2004). CaP has several 5 effects on the channel: increased trafficking to the plasma membrane, hyperpolarization of the G-V curve, and modulation of the inactivation. (He et al 2007). To date, four distinct auxiliary cL2\u00C3\u00B6 subunits (x26-1, cL2\u00C3\u00B6-2, i26-3 and x26-4) have been described (Klugbauer et at 1999, Qin et at 2002). The extracellular fL2 domain and the transmembrane 6 domain are the products of the same gene, and the precursor polypeptide is cleaved post-translationally into ct2 and 6, which are linked by disulfide bonds (Jay et al 1991). The a2 domain is extensively glycosylated, which has been found to be essential for the stability of the interaction with the ct subunit (Gurnett et at 1997). The Cax2\u00C3\u00B6subunit assists in the channel trafficking, (Arikkath and Campbell 2003, Klugbauer et a! 2003), and alters activation and inactivation kinetics (Klugbauer et a! 1999, Hobom et at 2000, Yasuda et at 2004, Canti et at 2005). The ci26 subunit is one of the targets of anti-epileptic and anti-allodynic drugs gabapentin and its analogs (Field et at 2006). To date, eight different Cay subunits (Cay1- Cay8) have been cloned (Freise et al 2000). The Cay subunit is a glycoprotein with four transmembrane domains and intracellular N- and C- termini (Jay et at 1990). So far, only Cay1 has been proven to be part of Ca1.1 and Cai.2; however, the role of Cayl is still unclear (Curtis and Catterall 1985, Kuniyasu et al 1992). \u00E2\u0080\u0098y2-y8 have not been identified as the Cas subunits, but have been suggested to play a role in AMPA glutamate trafficking (Tomita et al 2003). 1.1.1.2 Modulation of voltage-gated calcium channels Ca2 is one of the essential second messengers for a variety of cellular activities; however, excessive intracellular Ca2 is very toxic. Therefore, tight regulation of Cas 6 to 1S6 y 1 N SH3 NK Figure 2 The structure of the core of Cav subunit-AID complex. The core protein is made up of a guanylate kinase (GK) domain (purple) and SH3 domain (green). The alpha interaction domain (AID) is shown in red, and the arrow indicates the direction in which the AID connects to transmembrane segment 1S6. (Van Petegem et al 2004) 7 that controls Ca2 entry is crucial for normal cell physiology. A large number ofproteins have been identified to regulate the activity of Cas such as channel gating and kinetics (Khanna ci\u00E2\u0080\u0099 al 2007). Here, three major modulations mediated by calmodulin, GPCRs and protein kinases are discussed (Figure 3). 1.1.1.2.1 Regulation of Cars by calmodulin Calmodulin (CaM) is one of the crucial regulators of HVA Cas, and it serves as the Ca2 sensor for positive and negative Ca2feedback control. CaM is a ubiquitous Ca2 binding protein that is composed of an N-terminal lobe and C-terminal lobe, and binding of Ca 2+ to any or all the sites can cause different conformational changes to interact with the target proteins (Hoeflich and IKura 2002, Clapham 2007). CaM interacts with Cas through an IQ motif (IQXXXRGXXXR) within the C- tail of the channels (Van Petegem et al 2005, Dunlap 2007, Kim et al 2008) (Figure 4). The binding between CaM and Cas drives two opposing Cas feedback modulations: calcium-dependent inactivation (CDI) and calcium-dependent facilitation (CDF) which is the process whereby increased basal Ca2 or repeated depolarization leads to increased Ca2 currents. CDI and CDF are induced by specific lobes of calmodulin. Ca2 binding to C-lobe of CaM (Ca2/C-lobe) controls CDI in Cay 1.2 but CDF in Cay 2.1; Ca2 binding to N-lobe (Ca2/N-lobe) governs CDI in Cay 2.1, 2.2 and 2.3 (Van Petegem et al 2005, Kim et al 2008). 8 Figure 3 The interaction sites of different regulatory proteins of Cav2 al subunit. Gfry binds to N-terminal, C-terminal and I-Il loop of CacL1 subunit. PKC phosphorylates I-TI ioop and Il-Ill ioop of Cact 1 subunit. SNARE proteins interacts the Cact 1 subunit at Il-Ill loop. Ca2+/CaM binds to TM domain (IQ-like domain) and CBD domain within C terminal. (Catterall and Few 2008) 9 1.1.1.1.2 G-protein modulation of Cas G-proteins usually exist in a heterotrimeric form, which consists of Gct, Gf3 and Gy. The activation of the G-protein by G-protein-coupled receptors (GPCR) causes dissociation between Go subunit and Gfry subunit, which can function as a modulator to trigger downstream events in the signaling pathways (Gether et at 2002, Perez and Kamik, 2005). G-proteins show inhibitory effect on Cas. The electrophysiological features of voltage-dependent inhibition are a marked current reduction, depolarizing shift of the activation curve, and slower kinetics in both channel activation and inactivation (Artim and Meriney, 2000). The underlying mechanism of voltage-dependent inhibition is the direct binding Gfry subunit to cytoplasmic regions ofal subunit of Ca2 that can render channels difficult to open (Herlitze et al 1996, Ikeda and Dunlap 1999). Three major interaction sites are the N-terminal region, the intracellular 1-lI loop, and the C-terminal tail of the Cact1 subunit. (Canti et al 1999, Zamponi et at 1997, Li et at 2004, Agler et al 2005). 1.1.1.1.3 Modulation of Cas by proteins kinases Three major kinases that regulate Cas are PKA, PKC andCa2/calmodulin- dependent protein kinase II (CaMKII). The activation of cAMP-dependent PKA is under GPCR regulation. The activity of both L-type Ca1.1 and Ca1.2 is upregulated by PKA (Cachelin et at 1983). PKA is recruited to L-type calcium channels by A kinase anchoring proteins (AKAP5), which are physically bound to the channels (Fraser et al 1998, Gray et at 1998). PKA also potentiates the neuronal Ca2.1 (Huang et at 1998). 10 I0 0 l.a 0 Ca2.1 CDF N _____ CDI Cav2.3 (R-type) 0 I0 U 0 I0 (b) Cav22 (N-type) r 0 l.0 I0 L I0 z 0 0 Ca2.2 NC (a) av2.1 (P10-type) C (c) (d CDI CavI .2 (L-type) Ca2.3 N C CDI Cal .2 COF NC CDI Figure 4 Structures of Ca1 and Ca2-Ca2/C M-IQ domain com?lexes. (a) structure of Ca 2.1-Ca/CaM IQ domain complex (b) structure of Ca 2.2-Ca k/CaM IQ domain complex (c) structure of Ca 2.3-Ca/CaM IQ domain complex (d) structure of Ca 1.2- Ca2/CaM IQ domain complex. The cartoon below the crystal structure is a schematic of the binding mode for each complex and lobe specific function. (Kim et al 2008) 11 PKC is a family of protein kinases, which can be activated by different second messengers (Nishizuka 1995). PKC plays a dual role in regulating Ca1.2. The activity of the channel is decreased by N-terminal phosphorylation, but increased by C-terminal phosphorylation (McGee et at 2004, Yang et al 2005). PKC potentiates Ca2. 1 and Ca2.2 by phosphorylation of the I-TI loop of the CacL1 subunit, thus preventing Gfry binding (Zamponi et at 1997). PKC also increases the activity of Ca2. I and Ca2.2 by phosphorylation of the 11-111 loop to dissociate SNARE protein binding and thus antagonize SNARE protein-mediated inhibition (more elaboration in next two sections) (Jarvis et al 2002). CaMKIT is a key downstream effector of Cas. Activation of CaMKII requires interacting with Ca2/CaM and autophosphorylation (Lou et at 1989, Hudmon et al 2002). CamMKII phosphorylates and interacts with the ctl subunit of Ca1 .2 to increase calcium-dependent facilitation (Hudmon et at 2005, Lee et at 2006). For Ca2.1, CaMKIT slows down voltage-dependent inactivation, but supportsCa2tdependent facilitation by binding instead of phosphorylating the eLi subunit of Ca2.i (Jiang et at 2008). CaMKII also increases the activity of LVA Ca3.2 by phosphorylating the Il-Ill loop of the CaeL1 subunit (Welshby et at 2003). 1.1.2 Presynaptic proteins in the synaptic vesicle exocytosis The release of neurotransmitters by synaptic vesicle exocytosis is a key event in neuronal communication, and it requires membrane fusion between vesicles and plasma membranes. The high speed ofCa2ttriggered synaptic transmission confers the ability of the nervous system to efficiently respond to the immense variety of stimuli. Therefore, synaptic vesicle exocytosis needs to be under precise regulation both spatially and temporally. There are several features in the exocytosis event. Firstly, the synaptic 12 vesicles filled with neurotransmitters are docked at the specialized sites of presynaptic membranes called active zones, and then a series of priming reactions leave the vesicles in a state that is ready for release. Next, an action potential triggers the opening of Cas. The resulting influx of Ca2 then triggers the fusion of the vesicle with the plasma membrane. To date, several presynaptic proteins have been found to be involved in the neurotransmitter release. Due to their involvement with Cas, we here describe the major points about SNAREs and synaptotagmin. 1.1.2.1 SNAREs SNAREs (soluble N-ethylmaleimide-sensitive factor attachment protein receptors) are involved in all the steps of exocytotic and endocytotic pathways in eukaryotes (Joseph and John 2003). SNARE proteins are believed to establish the basic frame for the fusion apparatus and provide the necessary force for membrane fusion (Sollner et al 1993). SNAREs can be divided into two groups: vesicle (v-) SNAREs located on the vesicle membranes and target (t-) SNAREs residing on the plasma membranes. The most distinctive and ubiquitous molecular action of SNAREs is the formation of a stable parallel four helix bundle, referred as SNARE core, by SNARE motifs provided by both t- and v-SNAREs (Figure 5b). According to the \u00E2\u0080\u009Czippering\u00E2\u0080\u009D hypothesis, the N-terminal parts of the SNARE motifs begin to interact, then SNARE domains coil towards the C terminus to form a trans-SNARE, and eventually, the carboxy-terminal linker regions and transmembrane domains of SNAREs bind to each other to assemble a cis-SNARE complex, resulting in membrane fusion (Montecucco et a! 2005, Sorensen et a! 2006, Stein et a! 2009). 13 In neuronal vesicle fusion, syntaxinlA and SNAP-25 serve as t-SNAREs, while synaptobrevin functions as a v-SNARE. SyntaxinlA contains an N-terminal three helical bundle Habc domain, a SNARE motif and a transmembrane segment (Figure 5a). The NMR studies and fluorescence correlation spectroscopy show that syntaxinlA switches between two conformations: in the closed conformation, the Habc domain partially interacts with the SNARE motif to prevent the formation of the SNARE complex, while in the open state, the SNARE motif is available for binding other proteins (Misura et al 2000, Margittai et al 2003, Chen et at 2008). SNAP-25 is an unfolded soluble protein, consisting of two SNARE motifs separated by a linker that has four palmitoylated cysteine residues, which are responsible for plasma membrane anchoring (Figure 5a). Synaptobrevin is composed of an unfolded N terminus, a SNARE motif and a transmembrane segment (Figure 5a). Upon binding, syntaxin, SNAP-25 and synaptobrevin undergo disorder-to-order transition to form a stable quatemary structure in 1:1:1 ratio (Figure 5b). Although the idea that SNAREs provide a basic frame for the membrane fusion is widely accepted, the mechanism underlying the membrane fusion driven by SNAREs is still under debate. One of several proposed mechanisms stated here presents how SNAREs could regulate the membrane fusion. Prior to formation of the SNARE complex, syntaxinlA exists in a highly-ordered cluster form (Sieber et al 2007, Lu et at 2008). SNAP-25 is believed to be bound to syntaxin by the interaction of the SNARE motif without significant alteration of the cluster morphology (Richman et al 2005). The advantage of forming clusters is to provide hot spots on the plasma membrane so that assembly of multiple SNARE complexes can take place simultaneously to trigger 14 membrane fusion. Increasing amounts of evidence show that successful membrane fusion relies on coordination of several SNARE complexes (Chen et al 1999, Hua and Scheller 2001). As the fusion process continues, multiple syntaxin/SNAP-25 clusters simultaneously recruit synaptobrevins to initiate the formation of SNARE complexes, resulting in gap junction-like structures. Lipid mixing then starts to take place to form a lipid stalk intermediate, eventually leading to a hemifusion state: the outer membranes are merged, whereas the inner membrane still remain intact (Reese ci\u00E2\u0080\u0099 al 2005, Yoon ci\u00E2\u0080\u0099 al 2006, Lu et al 2008). Finally, the fusion pore opens, which is a result of merging of the inner membrane and formation of ring-like cis-SNARE complexes. Most of the researchers agree that the formation of the SNARE complex itself can trigger the membrane fusion, which has been proven by in vitro reconstitution experiments. However, the fusion kinetics of in vitro reconstitution experiments is 1000 times slower than the fusion rate in synaptic neurotransmission (Weber et al 1998, Bowen et al 2004, Liu et al 2005). Therefore, the SNARE assembly alone is not an efficient fusogen. In addition, the timing for the synchronized neurotransmitter release needs to be strictly controlled: the membrane fusion event does not happen only until the entry ofCa2. Apparently, the random membrane fusion stimulated by SNARE complexes cannot serve this role. Hence, there must be more players involved in the process of synaptic neurotransmission. 1.1.2.2 Synaptotagmin There are strong evidences showing that synaptotagmin, a vesicle protein, acts as the calcium sensor for fusion in calcium-triggered synaptic vesicle exocytosis (Wang et 15 al 2006, Rizo et at 2006, Chapman 2008). Synaptotagmin consists of a transmembrane segment, two cytosolic C2 domains (C2A and C2B, which bind three and two respectively) separated by a linker region (Figure 6) (Rizo et al 2006). Upon Ca2 binding, both C2A and C2B domains function cooperatively and penetrate into the plasma membrane through electrostatic interaction with the phospholipid head groups (Bai et at 2000, Rhee et at 2005, Herrick et at 2006). This interaction not only brings the vesicle and plasma membrane into close proximity, but also creates a curved dimple that points toward the vesicles, which can reduce the energy barrier for fusion (Monck and Fernandez 1994, Hui et al 2009). The C2B domain interacts with the syntaxin lA/SNAP- 25 t-SNARE complex in aCa2tdependent manner, suggesting that synaptotagmin might initially help formation of the SNARE complex (Pang et at 2006, Bhalla et al 2006). More interestingly, synaptotagmin binds simultaneously to SNARE complex and membranes, indicating that upon Ca2 binding, synaptotagmin can cooperate with SNAREs to bring vesicle and plasma membrane into close proximity and accelerate membrane fusion (Arac et at 2006, Dai et al 2007). 1.1.3 Regulation of presynaptic Cas by SNARE proteins Ca2.1 and Ca2.2 are the main Ca2 channels expressed in the presynaptic terminals, and the Ca2 influx through these two channels is primarily responsible for synaptic transmission (Olivera et al 1994). Since neurotransmitter release is a steep exponential function of the Ca2 concentration, the regulation of presynaptic Ca2 channels provides a sensitive and efficient means to regulate neurotransmitter release (Mintz et al 1995). The neurotransmitter release occurs in two phases: a fast synchronous 16 component and a slow asynchronous component (Atluri and Regehr 1998). In the synchronous phase, the release is triggered by the timed Ca2 current, resulting in a large and fast synaptic transmission. The asynchronous release is driven by the residual Ca2 remaining in the nerve terminal after an action potential, and provides a basal amount of neurotransmitter release (Sabatini and Regehr 1996, Hagler and Goda 2001). 1.1.3.1 Functional effects of SNARE proteins on presynaptic Cas The close proximity between the primed vesicles and the presynaptic Cas is crucial for the fast synaptic transmission (Heidelberger et al 1994), and the SNARE proteins are one of the presynaptic proteins that serve this role (Mochida et al 1996). Using pull-down experiments, it was shown that syntaxinlA and SNAP-25 interact with Ca2. 1 and Ca2.2 through the synaptic protein interaction (synprint) site (Figure 1 and Figure 7a), which is localized on the Il-Ill loop of the Caii1 subunit (Rettig et al 1996, Sheng et al 1997, Kim et al 1997). Since SNAREs and vesicles are present at the active zones, this means that the Cay\u00E2\u0080\u0094SNARE interaction keeps the channel and vesicles in close proximity. Besides keeping the close proximity between the channels and the vesicles, syntaxinlA and SNAP-25 also modulate the activity of the presynaptic channels through interaction with the synprint site. In particular, the interaction between SNARE proteins and the presynaptic channels inhibits the calcium channel activity by stabilizing slow inactivation of the channel and causing a hyperpolarizing shift in the voltage-dependence of inactivation. It has been proposed that the inhibitory effect of the SNARE proteins on the channels prevents the 2+ . .entry of Ca when the vesicles are not ready to release. This interaction also 17 (a) Synaptobrevin Syntaxi n Habc domain Ii SNARE TM motif \u00E2\u0080\u0094IJ SNARE TM motif Figure 5 Cartoon representations of neuronal SNARE proteins and the structure of the SNARE complex. (a) v-SNARE Synaptobrevin consists of a SNARE motif and a transmembrane segment. t-SNARE syntaxin is composed of a Habc domain, a SNARE motif and a transmembrane segment. t-SNARE SNAP-25 is made up of two SNARE motifs. (b) SNARE motifs of synaptobrevin, syntaxin and SNAP-25 form a stable four-helical bundle. The Habc domain of syntaxin is not involved in the formation of SNARE complex. (Rizo et al 2006) SNAP-25 SNARE motif (b) SNARE motif 18 Synaptotagmin C2B Figure 6 The structure of synaptotagmin. Synaptotagmin consists of two C2 domains (C2A and C2B) and a transmembrane segment (cartoon representation). C2A domain binds three Ca2,while C2B binds two Ca2. (Rizo et a! 2006) TM C2A 19 occupies SNARE proteins from the formation of SNARE complex before the vesicle docking, which is suggested by the competition between synaptotagmin and the synprint site to syntaxinlA in aCa2-dependent manner. At low Ca2 levels (below 301.tM), syntaxin 1 A is proposed to interact strongly with the synprint site, while at higher Ca2 levels, syntaxin 1A shows a higher affinity for synaptotagmin (Sheng et al 1997). This 2+ . 2+ observation may suggest that upon Ca entry, syntaxmlA released from Ca channel is available for interacting with synaptotagmin and the other SNARE components to drive membrane fusion, and the channels relieved from the inhibition allow for more Ca2 entry. Inhibition of rapid, synchronous synaptic transmission accompanied with enhancement of asynchronous transmitter release by peptides containing the synprint site of Ca2.2 indicates that the binding of the synprint site to SNARE proteins may be essential in regulating docking and priming events for efficient synaptic transmission (Mochida et al 1996). BesidesCa2ttriggered exocytosis, there is another exocytotic event that is induced by membrane depolarization in aCa2tindependent fashion. A peptide containing the synprint site can decreaseCa2independent transmitter release by the peptide containing the synprint site of Ca2.2, implying that the interaction between the synprint site and SNARE proteins may transmit a voltage-dependent signal to the SNARE complex (Mochida et al 1998). The synprint site may also play a role in the localization of presynaptic Cas in the nerve terminals (Mochida et al 2003). The phosphorylation of the synprint site by PKC and CaMKII regulates the activity of presynaptic Cans (Yokoyama et al 1997, Yokoyama et al 2005). Phosphorylation by PKC reduces the binding affinity of syntaxinlA and reverses the regulatory effects (Yokoyama et al 2005). In addition, the interaction between syntaxinlA and Ca2.2 allows syntaxinlA to 20 promote G-protein inhibition of the channel by physically recruiting the Gfry subunit to the I-TI loop of the channel (Jarvis ci\u00E2\u0080\u0099 al 2001, Jarvis et al 2002). 1.1.3.2 Molecular determinates of the interaction between SNARE proteins and the synprint site The interaction between the synprint site and syntaxinlA and SNAP-25 shows isoform specificity and Ca2 dependence (Figure 7b) (Retting ci at 1996, Kim et at 1997). The differences in interactions between Ca2 channels and the SNARE proteins may contribute to the differences in the efficiency of synaptic transmission for Ca2 entry through these different channels and confer specific functional and regulatory properties on the process of neurotransmitter release. An in vitro binding assay has shown that both the N-terminal half (718-895) and the C-terminal half (832-963) of the Ca2.2 synprint bind to syntaxin and SNAP-25 (Yokoyama C T et at, 2005). Moreover, the overlapping region of two halves predicted to be collagen homolog is also required for the interaction. Mutagenesis experiments show that the specific peptide sequence LRASCEALY (781-789) and KTSASTPAGGEQDR (860-873) within the N-terminal and C-terminal regions are crucial for SNARE binding (Yokoyama C T et at, 2005). Deletion of the N-terminal half, the entire synprint site or the whole Il-Ill loop results in depolarizing shift of voltage-dependent inactivation. Furthermore, coexpression ofsyntaxinlA with those impaired channels fail to cause the same hyperpolarizing shift compared to the wildtype channels, but the inhibitory effect of syntaxin still significantly remains (Bezprozvanny I et al, 2000). This may imply that the 21 Ca2 .2 Ca2.1 Ca2. 1 Ca2.l Ca2 .2 Ca2.1. Ca2. 1 Ca2 .1 Ca2 .2 Ca2. 1 Ca72 .1 Ca2. 1 bovine human iaol rat rbal rabbit El bovine human iaol rat rbal rabbit SI bovine human isol rat rbal rabbit SI Cav2.1 (rbA) Cav2.1 (BI) Cav2.2 Binding Cadependenee Binding C?dependenee Binding Ca2dependenee SyntaxinlA - - + - + + SNAP-25 + - + - + + Figure 7 Sequence of the synprint site and interaction with SNARE proteins. (a) Sequence alignment of the synprint sites of different isoforms in Cav2.l and Cav2.2 channels. The first shaded region shows the putative coiled-coil. The second shaded region shows collagen homolog. The \u00E2\u0080\u009Cstart\u00E2\u0080\u009D and \u00E2\u0080\u009Cend\u00E2\u0080\u009D of the synprint sites are indicated by arrows. (Van Petegem 2007) (b) Interaction between the synprint site and SNARE protein shows isoform specificity and Ca2 dependence. (a) Ca2 .2 Ca2. 1 Ca2 .1 Ca2 .1 Ca2 .2 Ca2. 1 Ca2. 1 Ca2 .1 Ca2.2 Ca2. 1 Ca2. 1 Ca2 .1 end ofj bovine human isol rat rbal rabbit El bovine human iaol rat rbal rabbit El bovine human iaol rat rbal rabbit El start I1SEDWN GflpQGSVpKF$lp\u00E2\u0080\u0099SH.FFIVLTLFSNYTLLNVFLAIAVpNLANAQELTK r GEDW4 MY 4flQGGVSI3MfrFPMYFIVLTLFGNYTLtMVFLAIAVPMLAMAQELTK fr GEDWM MY EEflQGG4SMfrFfrjYFIVLTLFGNYTLLNVFLAIAVMLANAQELTX frSEDWMk MY GCQSybGfrFWIYFIVtTLFGMYTLLNVFLAIAVPNLAMAQELTK putative coiled-coil E fl E QKtALQKAKEVAEVSs.j4 AAN IA E E jIQKLALQKAKEVAEVS4 AAN IA K4KI42KPI4(PVWEQSjTflM E S [QKLALQKAKEVAEVS4 AAN IA KEW2KI42KP4C[VWEQ4IS[M r \u00E2\u0080\u00A2z bQKLALQKAKEVAEVS* AAN IA KEK42KPflVWEQStr.FM 411 [Li , rE51r en. an\u00E2\u0080\u0099 c El?. ALY EN AL -G ALY EM D KAAYT RA PT1\u00E2\u0080\u0099YA PS KASYA Wlst\u00E2\u0080\u0099nurne flTtTtJ - Gp Ia RPD ITMLDRPLVVk IZEMRNMMTMKSRAAEPT RPD tWMLDRPLVVPK2EMRMNMTMKS RPD tTHLDRPLVVbJ2EMRMMMTMKSRVAEPT RRMMWHRDRDKWAfl- AGEPDRADAPKAEGGE PkREERAR VDQRLGQQRAEDFLWQ*frMDRI4WH-SGSAGLDARRPWAGSQEAELSREGPfrGRE SD RAPEALtT*frRESj4D\u00E2\u0080\u0099F DARRAWPSSPERAPGREYGRESE VDQRLGQQRAEDFLWQAkJIHDRWDHSAHAAAGLDARRPWAGSQEAELSREHPYGRESD PRRSRSKEAAGPREARSERGRLGPWG MRfl SPEEAfRflP HHAREGS-LEQPGFWEGEA4PKAGWH R VMfl--GGSRESRSGSPRTS-9Zc4jI RH PQQREMAPPREHVPWDADP44LKAG4kP A TH V)4EGflP RN HQAREGG-LEPPGFWEGEAEKAGWH R AHGVGGSGGSRSQSPRTGTGflP RH HRHAPEPSREGAPGSKGEFflflifl3SSflGPREAESGEEPS Hflfl4KA NRRPGEE----GPEDKA*R)4444CSS14\u00E2\u0080\u0099ARGGEGESEGPD SGEjtRflN{GA RRRPGDE PDDRPE A RATjARAADGEGD DGE4CRj44iGP frHRRPGED----GPDDKAE R GSARSGEGEAEGPDGGGGGGGE4j{W{GP (b) 22 modulatory effect of syntaxin 1A is normally transmitted by the synprint site; however, in the absence of the synprint site, other regions of the channel may partially substitute for this site. Both the N-terminal Habc and the C-terminal H3 domain of syntaxin 1 A are implicated in interacting with the synprint site (Retting J et a!, 1996; Jarvis eta!, 2002). However, it seems that only the H3 domain and transmembrane segment of syntaxin 1 A have functional effects on channel properties. 1.1.3.3 Splicing variants of the synprint site The distinct Ca2 channel variants from regulated alternative splicing show the specialized channel functions to optimize calcium signaling in different regions of the brain. To date, two human Ca2.2 and two rat Ca2. I splice variants that lack synprint have been reported (Kaneko et al 2002, Rajapaksha et a! 2008). One of the common features for all four variants is that voltage-dependence inactivation is shifted in the depolarizing direction instead of hyperpolarization as a result of syntaxin or SNAP-25 modulation (Kaneko et a! 2002, Rajapaksha et a! 2008). This is reasonable because without the synprint site, the targeting site, SNARE proteins lose the modulatory effects on the Ca2channels. Interesting, two rat Ca2.1 variants are the predominant forms in neuroendocrine cells, which may reinforce the idea that the synprint site plays a role in targeting presynaptic Cas to nerve terminals (Rajapaksha et a! 2008). However, this hypothesis has been challenged by the fact that in snail, Ca2.1f2.2 lacking the entire synprint still localize in the nerve terminals (Spafford et al 2003). The observation suggests that other proteins such as the modular adaptor proteins CASK and Mint are also involved in channel targeting (Spafford et a! 2003). 23 1.1.3.4 Other interacting partners of the synprint site There are other proteins such as Cystein string protein (CSPu) (Swayne et al 2006), Huntingtin (htt) (Swayne et al 2005) and RIM (Coppola et al 2001) have also been proposed to interact with the synprint site. However, no direct functional effect on the channel has been shown. 1.2 Crystallization X-ray crystallography is an invaluable tool to study protein structures. Since the function of a protein depends on its structure, accurately determining protein structures allows for the study of protein functions at the atomic level and provides reliable answers to the structure-related questions. Here I will discuss a number of methods used to obtain protein crystals. 1.2.1 Overview of protein crystallization Crystallization is the process that forces a protein into well-ordered three dimensional arrays. The crystal formation involves three steps: nucleation (the initial event in which a small number of molecules become arranged and form a site for additional molecules to be deposited on), growth and cessation of growth. The crystallization process can be explained in the phase diagram (Figure 8a). For a protein to crystallize, it must first be in the state of supersaturation, in which the protein reaches or overcomes the solubility boundary and starts to become insoluble. However, high levels of supersaturation (precipitation zone) cause the protein to precipitate rather than forming 24 well-ordered crystals. The goal is to keep the protein in the labile zone where the nucleation occurs and then bring it to the metastable zone where the crystal grows. In the process of protein crystallization, both the environmental parameters and the intrinsic protein properties should be considered and well controlled. Suitable environmental parameters such as pH, temperature and ionic strength need to be adjusted for the optimal growth of crystal. Intrinsic properties include purity and homogeneity. Purity refers to lack of contamination. Homogeneity refers to lack of conformational heterogeneity (flexible domains and polymer formation) and lack of sequence heterogeneity (proteolysis or degradation). 1.2.2 Vapor diffusion crystallization The vapor diffusion technique is the most frequently used crystallization method. Typically, a small amount of protein (usually 1-10 p1) is mixed with an equal amount of crystallization solution containing for example buffer, salt and precipitant. The drop is suspended and sealed over the well solution which contains the crystallization solution. The different concentration of the salt and precipitant between the drop and the well solution drives water to evaporate from the drop until an equilibrium is established, in which the activity of the water in the drop and the well is equal (Figure 8b). 25 c 0 ct C) 0 c 0 Sitting Drop grease cove\u00E2\u0080\u0099sIlp or tape) V __ HO :q /, Figure 8 Phase diagram of crystallization and vapor diffusion crystallization setups. (a) The theoretical phase diagram. Crystallization only occurs in the supersaturation phase. Nucleation occurs in the labile zone, and crystal grows in metastable zone. Retrieved from http://molecularsciences.org/files/images/solubility-curve.gif. (b) sitting drop and hanging drop setups for vapor diffusion crystallization. Retrieved from http ://www.bio.fju.edu.tw/excel/contentO5/html/40.files!imageO 11 .gif. (a) (b) precipitant concentration Hanging brop protein 2tiI reswvo4r zi coversllp & ease protein ZpI 4 re$4rvc\u00E2\u0080\u0099 2u1 I I \u00E2\u0080\u009C\u00E2\u0080\u0098L__I i4rrrr,I 26 1.2.3 Counter diffusion crystallization In the conventional vapor diffusion approach, a supersaturation condition targets a single labile region for nucleation. Therefore, crystals grow out of one single condition. In the counter diffusion method, the protein and the precipitating agent are placed on the opposite ends of a capillary, and diffuse against each other, forming a gradient of conditions along the length of the capillary (Ng et al 2003). A single capillary can thus span a continuum of regions in the phase diagram. The continuous supersaturation gradient makes counter diffusion a good optimization method. One of the disadvantages is that counter diffusion in regular capillaries is not amenable to high throughput. However, chips (Fluidigm) and small counter diffusion plates (microlytics) have been developed to automate the procedure. 1.2.4 Dialysis crystallization The protein contained within the dialysis membrane is equilibrated against the precipitant in the surrounding solvent to slowly achieve the supersaturation for crystallization. Dialysis is the most effective way for crystallization and works by decreasing the ionic strength. If a protein is less soluble at low ionic strength, it may be possible to crystallize by dialyzing against a low-salt buffer or water to slowly remove the salt. Also, dialysis offers an easy way to screen for different conditions simply by moving the dialysis container from one condition to another. Unlike vapor diffusion, the protein concentration remains constant during dialysis. 27 1.2.5 Other crystallization techniques For a protein to crystallize, it must be in the supersaturation state (Figure 8a). In addition, nucleation is also necessary for crystallization. However, sometimes the levels of supersaturation that promote spontaneous nucleation are too high for the slow, accumulative growth of nice single 3D crystals, but often lead to showers of small crystals. Seeding is a powerful tool for crystal optimization, which separates the nucleation event from the growth process. The seed crystals are transferred from the original drops where they nucleated to a new experimental condition, in which the supersaturation level is high enough to support crystal growth, but low enough to prevent spontaneous nucleation (Bergfors 2007). Therefore, protein molecules can accumulate on a ready-made template in order for the crystal to grow bigger. Seeding is often used to reduce showers of crystals, to grow single crystals, to increase the size of crystals and to improve crystal quality. Crystallization of proteins fused to stabilizing proteins is a relatively new method. Usually, the fusion proteins have enhanced solubility and facilitate protein folding when overexpressed in an expression host. Several successful examples have proven the feasibility of fusion protein crystallization (Smyth et al 2003). One major problem with this method rises from the linker region between the protein tag and the protein of interest. A linker that is too short can cause misfolding of the protein, whereas a long linker increases the flexibility of the fusion protein, resulting in conformational heterogeneity or susceptibility to proteolysis. 28 Protein crystallization has been proven to be predominantly dependent on entropic effects (Derewenda and Vekilov 2006). Entropic cost from the ordering of protein molecules and the loss of side chain freedom disfavors the protein crystallization (Tidor and Karplus 1994). Burying flexible surface residues such as Lys and Glu at crystal contact regions significantly contribute to the loss of entropy. Therefore, mutating the surface residues with high conformational energy into small amino acids such as Ala would reduce the surface entropy, thus lowering the free energy barrier and promoting protein crystallization. 29 2 Objectives There are two experimental goals in this study. Firstly, in order to study the synprint site on the molecular level, we use crystallography to determine the three dimensional structure of the synprint site. Secondly, we use the biochemical technique isothermal titration Calorimetry (ITC) to further characterize the interaction between the synprint site and t-SNARE syntaxin 1 A 30 3. Materials and Methods 3.1 Cloning 3.1.1 Cloning of synprint domains from rabbit Ca2.1 and bovine Ca2.2 All the synprint domain constructs from rabbit Ca2. I and bovine Ca2.2 were cloned into the pET28HMT vector (Table 2). The templates for the inserts were rabbit Ca2. I full-length channel and bovine Ca2.2. The standard PCR protocol was the initial denaturation at 95\u00C2\u00B0C for 3 minutes and 30 cycles of: denaturation at 95\u00C2\u00B0C for 0.5 minutes, annealing at 55\u00C2\u00B0C for 0.5 minutes and elongation at 72\u00C2\u00B0C for 3 minutes, and I cycle to complete the elongation at 72\u00C2\u00B0C for 10 minutes. DMSO (Acros) at a final concentration of 2.5% was necessary for the successful PCR of all the bovine Ca2.2 constructs. Ligation independent cloning (LIC) was used to paste the PCR products into the expression vector. Briefly, the PCR products were verified on 1% agarose (Fisher) gels and extracted (Qiagen). The pET28HMT vector was digested by SspI (NEB) for 3 hours at 3 7\u00C2\u00B0C, verified on a 1% agarose gel and extracted. Both the inserts and the vector were subjected to T4 polymerase (Fermentas) treatment with dCTP (BioLabs) and dGTP (BioLabs), respectively, to create the long overhangs for 40 minutes at room temperature, followed by 20-minute heat inactivation of the polymerase at 75\u00C2\u00B0C. 2111 of T4-treated insert and vector were mixed together, and the overhangs were annealed for 10 minutes at room temperature. Finally, the ligation products were transformed into E. coil DH5a under kanamyciri selection. The LIC overhang sequence for the forward primer was tacttccaatccaatgca, and the LIC overhang sequence for the reverse primer was ttatccacttccaatgttatta. 31 Table 2 Oligonucleotides used in this study Construct Name Residue Primer Sequence Number 1 Rab a 1 AsynA (F) 722-8 76 tactcccaatccaatgcagagctcaccaa Rab a 1 AsynA (R) ttatccacttccaatgttaataccgccgcc 2 Rab a 1 AsynA 1 (F) 71 5-867 tacttccaatccaatgcagacaacctggccaatgcccag Rab a 1 AsynA 1 (R) ttatccacttccaatgttattaggcgtgggcgctggggtc ccg 3 Rab alAsynA2 (F) 722-851 tactcccaatccaatgcagagctcaccaa Rab a 1 AsynA2 (R) ttatccacttccaatgttattagtcctcggcccgctgctgc cc 4 Rab a 1 AsynA3(F) 743-876 tacttccaatccaatgcagacaacctggc caatgcccag Rab a lAsynA3(R) ttatccacttccaatgttattaggcgtgggcgctggggtc ccg 5 Rab a 1 AsynA4 (F) 743-851 tacttccaatccaatgcagacaacctggc caatgcccag Rab alAsynA4 (F) ttatccacttccaatgttattagtcctcggc ccgctgctgccc 6 Rab alAsynA5 (F) 715-826 tacttccaatccaatgcagacaacctggcc aatgcccag Rab a 1 AsynA5 (R) Ttatccacttccaatgttattacacgaccag cggccggtccaggtg 7 Rab a I AsynA7 (F) 738-821 tacttccaatccaatgcagccaaggaggtg gcggaagtg Rab a 1 AsynA7 (R) ttatccacttccaatgttattacacgaccagc ggccggtccaggtg 8 Rab a 1 AsynB 1 (F) 877-1004 tacttcca.atccaatgcagccgccggcctggacgcgcg C Rab a 1 AsynB 1 (R) ttatccacttccaatgttattagtccggcccctcggcctct cc 9 Bovsyncoil (F) 710-745 tacttccaatccaatgcagacaacctcgccaatgctcag Bovsyncoil (R) ttatccacttccaatgttattagctgacttcagccacttcttt ggc 10 Bova 1 BsynA 1 (F) 710-865 tacttccaatccaatgcagacaacctcgccaatgctcag Boval BsynA 1 (R) ttatccacttccaatcttattaccgccggc 1 1 Bova 1 BsynA 1 C784A (F) 710-865 gcagaacctgcgggccagcgctgaagcgctgtatagc g Bova I BsynA 1 C784A (R) cgctatacagcgcttcagcgctggcccccaggttctgc 12 Bova 1 BsynA4 (F) 738-846 ttatccacttccaatgttattagtcggcac Bova 1 BsynA4 (R) ttatccacttccaatgttattagtcggcac 32 Table 2 Oligonucleotides used in this study (continued) Construct Name Residue Primer Sequence ___________________________ Number 13 Bovct 1 BsynA8C784A (F) 710-846 tacttccaatccaatgcagacaacctgg ccaatgcccag BovcL I BsynA8C784A (R) ttatccacttccaatgttattagtcggcac 14 Bova 1 BsynA8C784A/E794A/E79 710-846 gatggaccccgcggcgcgcctgcgc 5A (a.k.a synA8EE) (F) Boviti BsynA8C784AIE794A/E79 tcgctatacagcgcttcacagctgg 5A (a.k.a synA8EE) (F) 15 Bova 1 BsynA8C784A/E794A/E79 710-846 cgagggcggagcggccggcgagggtgccg 5A/E820A (F) acc Bovo 1 BsynA8C784A/E794A/E79 ggccgggccttgccccccgcaggcccccgc 5A/E820A (R) gc 16 Bovct 1 BsynA8C784A/E794A/E79 710-846 tacttccaatccaatgcagacaacctcgccaat 5A/K737A/K739A/E740A (F) gctcaggagctgacc Bovct 1 BsynA8C784A!E794A/E79 ctgacttcagccactgctgcggccgcttgcag 5A/K737A1K739A/E740A (R) agcaag BovcL 1 BsynA8C784AIE794A/E79 cttgctctgcaagcggccgcagcagtggctg 5A/K737A1K739A/E740A (F\u00E2\u0080\u0099) aagtcag BovcL TBsynA8C784A/E794A/E79 ttatccacttccaatgttattagtcggcgtcggc 5A/K73 7A/K739A/E740A (R\u00E2\u0080\u0099) accctcgccggcctctcc 17 MBP-synA8EE (F) 710-846 tacttccaatccaatgcaatgaaaatcgaaga aggtaaac MBP-synA8EE (R) ccccggatccagtctgcgcgtctttcagggc MBP-synA8EE (F\u00E2\u0080\u0099) aaaaggatccgacaacctcgccaatgctcag MBP-synA8EE (R\u00E2\u0080\u0099) ttatccacttccaatgttattagtcggcac 18 MBP-3A-synA8EE (F) 710-846 tacttccaatccaatgcaatgaaaatcgaaga aggtaaae MBP-3A-synA8EE(R) ccceggatccagtctgcgcgtctttcagggc MBP-3A-synA8EE (F\u00E2\u0080\u0099) aaaaggatccgcagcagcagacaacct cgccaatgctcag MBP-3A-synA8EE (R\u00E2\u0080\u0099) ttatccacttccaatgttattagtcggcac 19 MBP-5A-synA8EE (F) 7 10-846 tacttccaatccaatgcaatgaaaatcgaaga aggtaaac MBP-5A-synA8EE (R) ccccggatccagtctgcgcgtctttcagggc MBP-5A-synA8EE (F\u00E2\u0080\u0099) aaaaggatccgcagcagcagcagcagacaa cctcgccaatgctcag MBP-5A-synA8EE (R\u00E2\u0080\u0099) ttatccacttccaatgttattagtcggcac 20 GST-synA8EE (F) 710-846 tacttccaatccaatgcaatgtcccctat actaggttattgg GST-synA8EE (R) ccccggatccatccgattttggaggatggtcc 33 Table 2 Oligonucleotides used in this study (continued) Construct Name Residue Primer Sequence Number GST-synA8EE (F\u00E2\u0080\u0099) aaaaggatccgacaacctcgccaatgctcag GST-synA8EE (R\u00E2\u0080\u0099) ttatccacttccaatgttattagtcggcac 21 GST-3A-synA8EE (F) 710-846 tacttccaatccaatgcaatgtcccctatactaggtt attgg GST-3A-synA8EE (R) ccccggatccatccgattttggaggatggtcc GST-3A-synA8EE (F\u00E2\u0080\u0099) aaaaggatccgcagcagcagacaacctcgccaa tgctcag GST-3A-synA8EE (R\u00E2\u0080\u0099) ttatccacttccaatgttattagtcggcac 22 GST-5A-synA8EE (F) 710-846 tacttccaatccaatgcaatgtcccctatactaggtt attgg GST-5A-synA8EE (R) ccccggatccatccgattttggaggatggtcc GST-5A-synA8EE (F\u00E2\u0080\u0099) aaaaggatccgcagcagcagacaacctcgccaa tgctcag GST-5A-synA8EE (R\u00E2\u0080\u0099) ttatccacttccaatgttattagtcggcac 23 MBP-synA8EE-His (F) 710-846 aaaacatatgaaaatcgaagaagg MBP-synA8EE-His (R) ccccggatccagtctgcgcgtctttcagggc MBP-synA8EE-His (F\u00E2\u0080\u0099) aaaaggatccgacaacctcgccaatgctcag MBP-synA8EE-His (R\u00E2\u0080\u0099) aaaactcgaggtcggcaccctcgc 24 MBP-3A-synA8EE-His (F) 710-846 aaaacatatgaaaatcgaagaagg MBP-3A-synA8EE-His (R) ccccggatccagtctgcgcgtctttcagggc MBP-3A-synA8EE-His (F\u00E2\u0080\u0099) aaaaggatccgcagcagcagacaacctcgccaa tgctcag MBP-3A-synA8EE-His (R\u00E2\u0080\u0099) aaaactcgaggtcggcaccctcgc 25 MBP-5A-synA8EE-His (F) 710-846 aaaacatatgaaaatcgaagaagg MBP-5A-synA8EE-His (R) ccccggatccagtctgcgcgtctttcagggc MBP-5A-synA8EE-His (F\u00E2\u0080\u0099) aaaaggatccgcagcagcagacaacctcgccaa tgctcag MBP-5A-synA8EE-His (R\u00E2\u0080\u0099) aaaactcgaggtcggcaccctcgc 26 GST-synA8EE-His (F) 710-846 aaaacatatgtcccctatactagg GST-synA8EE-His (R) ccccggatccatccgattttggaggatggtcc GST-synA8EE-His (F\u00E2\u0080\u0099) aaaaggatccgacaacctcgccaatgctcag GST-synA8EE-His (R\u00E2\u0080\u0099) aaaactcgaggtcggcaccctcgc 27 GST-3A-synA8EE-His (F) 710-846 aaaacatatgtcccctatactagg GST-3A-synA8EE-His (R) ccccggatccatccgattttggaggatggtcc GST-3A-synA8EE-His (F\u00E2\u0080\u0099) aaaaggatccgcagcagcagacaacctcgccaa tgctcag GST-3A-synA8EE-His (R\u00E2\u0080\u0099) aaaactcgaggtcggcaccctcgc 34 Table 2 Oligonucleotides used in this study (continued) 28 GST-5A-synA8EE-His (F) 710-846 aaaacatatgtcccctatactagg GST-5A-synA8EE-His (R) ccccggatccatccgattttggaggatggtcc GST-5A-synA8EE-His (F\u00E2\u0080\u0099) aaaaggatccgcagcagcagacaacctcgccaa tgctcag GST-5A-synA8EE-His (R\u00E2\u0080\u0099) aaaactcgaggtcggcacectegc 29 Syntaxin 1-189 (stxi -189) (F) 1-189 tacttccaatccaatgcagtatgaaggacc Syntaxin 1-189 (stxii 89) (R) ttatccacttccaatgttattacttcgagatgctgga gtcc 30 syntaxin 190-264 (StX19O2M) (F) 190-246 aaaaggatcccaggcccacagtgagatcg syntaxin 190-264 (stx1902M) (R) ccccctcgagctacttcctgcgcgccttgc 35 After plasmid extraction (Qiagen), several clones were sent for sequencing (Macrogen). 3.1.2 Mutagenesis of the synprint constructs Normal Quikchange (Stratagene) mutagenesis did not work for the synprint mutagenesis. Phusion site-directed mutagenesis (Finnzymes) was used instead according to the manufacturer\u00E2\u0080\u0099s guidelines. T4 ligase (NEB) was used to ligate the PCR products at room temperature for 5 minutes, followed by transformation. Another mutagenesis method was used to make the mutant of Bovc 1 BsynA8C784A/E794A1E795A1K737A/K739A/E740A (Table 2) using three sequential PCRs and using BovcdBsynAlC784A/E794A1E795A as a template. The first two PCRs were used to create the K737A1K739A1E740A mutations using primer 16F, 16R and 16F\u00E2\u0080\u0099, 16R\u00E2\u0080\u0099. The two PCR products were verified on a 1% agarose gel, extracted, and mixed together for the template of the third PCR. The third PCR was performed using primer 16F and 16R\u00E2\u0080\u0099. The PCR product was verified on a 1% agarose gel, extracted, and used as insert for LIC cloning into the pET28HMT vector. The standard PCR protocol, described above, was used in all the three PCRs. 3.1.3 Cloning of the fusion proteins The constructs of MBP and GST-(linker)-synA8EE were cloned into the pET28HT vector by LIC (Table 2). The insert consisted of an N-terminal MBP or GST tag fused to synA8EE. The MBP tag and the synA8EE were amplified with primer I 7F, 17R and 17F\u00E2\u0080\u0099, 17R\u00E2\u0080\u0099 respectively using the standard PCR protocol (Table 2). Both PCR products were verified by 1% agarose gel, extracted, and digested by BamHI (NEB) for 1 hour at 37\u00C2\u00B0C. The two BamHI-treated ends were ligated by T4 ligase at room temperature 36 for 2 hours. The ligation products were run on a 1% agarose gel. This generated a fused construct devoid of any linker. Two additional constructs of MBP-linker synA8EE with 3-alanine and 5-alanine codons were also made. Similarly, GST-linker-synA8EE constructs with different linkers were also made (Table 2). Finally, the gel-purified ligation product was cloned into pET28HT by LIC. The MBP and GST-(linker)-synA8EE-His constructs were cloned into the pET24a (Novagen) vector using NdeI and XhoI restriction sites. The MBP tag and the synA8EE were amplified with primer 23F, 23R and 23F\u00E2\u0080\u0099, 23R\u00E2\u0080\u0099 respectively using the standard PCR protocol (Table 2). Both PCR products were verified by 1% agarose gel, extracted, and digested by BamHl for 1 hour at 37\u00C2\u00B0C. The two BamHI-treated ends were ligated by T4 ligase at room temperature for 2 hours. The ligation products were run on a 1% agarose gel. The gel-purified ligation product was digested at 37\u00C2\u00B0C sequentially (16 hours for NdeI (NEB) and 16 hours for XhoI (NEB)).The insert and the vector were ligated using T4 ligase at room temperature for 2 hours, transformed into E. coli DH5ct, and plated on ampicillin. This generated a fused construct devoid of any linker. The other MBP-tagged and GST-tagged constructs were generated in a similar way. 3.1.4 Cloning of syntaxinlA constructs Mouse syntaxinlA (1-189) was cloned into the pET28HMT vector by LIC. The standard PCR protocol was used with a full-length mouse syntaxinlA as template. Mouse syntaxin 1 A (190-264) was cloned into pGEX6. 1 using the BamHl and XhoI restriction sites. The standard PCR protocol was used with a full-length mouse syntaxinlA clone as template. PCR products were verified on a 1% agarose, extracted, and digested at 37\u00C2\u00B0C in 37 sequential reactions (ihour for BamHI; 16 hours for XhoI). The vector and insert were ligated using T4 DNA ligase for 2 hours at room temperature, transformed in to E.coli DH5ct, and plated on ampicillin. 3.2 Protein expression and purification 3.2.1 Protein expression. All the proteins in this study were expressed in E. coil Rosetta (DE3)pLacl, and grown in 2xYT medium (Fisher) at 37\u00C2\u00B0C under continuous shaking unless stated otherwise. Typically, IL of culture was grown in 2.8L Fernbach flasks until 0D600 of 0.6, induced with 0.4mM IPTG (VWR), and grown further for 3-4 hours. Cells were harvested by centrifugation at 5,000 rpm for 15 minutes using the floor centrifuge (Beckman Coulter). Cells were lysed using Sonic Dismembrator Model 500 (Fisher) at 40% amplitude for 1 minute for four times. The cell lysis buffer contained 250mM KCI (Fisher), 10mM Hepes pH 7.4 (Fisher), 10% glycerol (Fisher), 100mM PMSF (Sigma), 25 mg/mi DNaseI (Sigma) and 25 mg/mI lysozyme (Pierce) unless stated otherwise. Centrifugation was performed at 35,000xg for 30 minutes at 4\u00C2\u00B0C to remove insoluble material. 3.2.2 Purification of the synprint domains from rabbit Ca2.1 and bovine Ca2.2 All the synprint domain constructs were cloned into the pET28HMT vector, containing, in sequence, a His6 tag, maltose binding protein (MBP) tag, and a cleavage site for TEV protease at the N-terminus of the synprint domains (Table 2). 38 A two-day purification protocol was used, and all the buffers used for chromatography are listed in Table 3. The clarified supernatant was loaded onto a poros MC (nickel affinity) column (Applied Biosystems) in buffer A and eluted in successive 6% and 60% steps of buffer B. The eluate was dialyzed (3.5K cutoff membrane tubing, VWR) in buffer A for 2 hours at room temperature, loaded onto an amylose (MBP affinity) column (BioLabs) in buffer A, and then eluted in buffer C. The eluate was collected and incubated with TEV protease (3 mg/mi, lmL) overnight at room temperature. The next day, the cleaved sample was run on a poros MC column to remove the His-MI3P tag, His TEV protease and any uncleaved proteins. The protein of interest was collected in the flow-through, and dialyzed in low ionic strength buffer D, F, or J, depending on the subsequent ion exchange column, for 1 hour at room temperature. The protein sample was then applied to either a Hiload SP (cation exchange) column (GE Healthcare) or a Hi load Q (anion exchange) column (GE Healthcare) for further polishing and eluted using a linear gradient of buffer E, G or K. The eluate was concentrated to approximately 500 iii by centrifugation at 4150 rpm at 4\u00C2\u00B0C using a Amicon Ultra centrifugal filter (3K cut-off Millipore) and loaded onto a superdex 200 (size exclusion) column (GE Heaithcare) in buffer A. The peak fractions were collected. The KCI concentration was reduced to 25mM in the final protein preparation by diluting the protein sample in 10mM Hepes pH 7.4. The protein concentration was determined by reading the absorbance at 280nm under denaturing conditions (Edeihoch, 1967). The final protein preparation was stored in the buffer containing 25mM KCI and 10mM Hepes pH 7.4 at a concentration of 10mg/mi at -80\u00C2\u00B0C. 39 Table 3 Buffers of all the chromatography columns Affinity column Equilibration/wash buffer Elution buffer Poros MC Buffer A: Buffer B: 25OmMKCJ 25OmMKCJ 10mM Hepes (Phosphate) 7.4 500mM Imidazole 7.4 or 8.0 or 10mM Tris 8.0 Amylose Buffer A: Buffer C: 250mMKC1 25OmMKC1 10mM Hepes (Phosphate) 7.4 10mM Hepes (Phosphate) or 10mM Tris 8.0 7.4 or 10mM Tris 8.0 10mM maltose Glutathion sepharose Buffer L: Buffer M: 250mM KC1, 50mM Tris 8.0, 10mM Tris 8.0 10mM reduced glutathione Ion exchang column Low ionic strength buffer High ionic strength buffer Hiload Q (anion Buffer D: Buffer E: exchange) 10mM KCI, 20mM MES 6.3 1M KC1, 20mM MES 6.3 Hiload SP (Cation exchange Buffer F: BufferG: 10mM KC1, 20mM Hepes 7.4 1M KCI, 20mM Hepes 7.4 Buffer H: Buffer!: 10mM KC1, 20mM Tris 8.0 1M KCI, 20mM Tris 8.0 BufferJ: Buffer K: 10mM KC1, 20mM Tris 8.8 1M KCI, 20mM Tris 8.8 Size exclusion column Equilibration/elution buffer Superdex200 Buffer A: 250mM KCI 10mM Hepes (Phosphate) 7.4 or 10mM Tris 8.0 40 3.2.3 Purification of BovalBsynA8 mutants The purification scheme of the BovctlBsynA8 mutants was the same as the two- day purification protocol described above with the following modifications: three additional columns, a shorter time of TEV cleavage and all the chromatography buffers contained Tris pH 8.0 as the buffer. On the first day, between the poros MC column and amylose column, the protein sample was dialyzed in buffer H at room temperature for 1 hour, loaded onto a Hiload Q column, and eluted using a linear gradient of buffer I. After the amylose column, the protein sample was concentrated to approximately 2 ml and run on a Hiload superdex 200 (size exclusion) column in buffer A overnight. On the second day, the peak fractions from the superdex 200 were collected and subjected to TEV cleavage for 3 hours at room temperature. Following the poros MC and Hiload Q columns, another amylose column was used to further remove the cleaved His-MBP tag. 3.2.4 Purification of MBP (GST)-(Iinker)-synA8EE fusion proteins A one-day protocol was used for the purification. The clarified supernatant was loaded onto a poros MC column in buffer A and eluted in successive 6% and 60% steps of buffer B. The eluate was dialyzed in buffer A supplementend with 14mM f3ME for 3 hours at room temperature along with TEV protease. The cleaved protein sample was then loaded on a second poros MC column, and the flow-through was collected in buffer A. The protein sample was applied to either an amylose column or a glutathione sepharose column (GE Healthcare) in buffer A depending on the affinity tag. The MEP or GST tagged proteins were then eluted in buffer C or L, respectively. After a one-hour dialysis in buffer F, the protein sample was loaded onto a Hiload Q column and eluted 41 using a linear gradient of buffer G. Since the fusion protein was sensitive to degradation, the protein sample was not prepared for crystallization. 3.2.5 Purification of MBP (GST)-(linker)-synA8EE-His fusion proteins The buffers used for chromatography all contained Tris pH 8.0. The clarified supematant was loaded onto a poros MC column in buffer A and eluted in successive 6% and 60% steps of buffer B. The eluate was dialyzed in buffer A for 2 hours at room temperature, loaded onto an amylose column in buffer A, and then eluted in buffer C. The eluate was directly loaded onto a poros MC column in buffer A and eluted in 60% buffer B. After a one-hour dialysis in buffer H, the protein sample was loaded onto a Hiload Q column, and eluted using a linear gradient of buffer I. The KC1 concentration was reduced as before (section 2.2.2). The final protein prep stored in the buffer containing 25mM KC1 and 10mM Tris pH 8.0 at a concentration of 30mg/mi at -80\u00C2\u00B0C. 3.2.6 Purification of stxlls9, synAl and synBi for ITC stx1189, synAl and synBi (Figure 20) were purified with the same purification protocol. The clarified supematant was loaded onto a poros MC column in buffer A and eluted in successive 6% and 60% steps of buffer B. The eluate was dialyzed in buffer A plus 14mM I3ME for 3 hours at room temperature along with TEV protease. The cleaved protein sample was then loaded on a second poros MC column, and the flow-through was collected in buffer A. synAl or synBi was dialyzed in buffer J, while stx1189 was dialyzed in buffer F. synAl or synBi was loaded onto a Hiload Q column and eluted using a linear gradient of buffer K. stx1189 was applied onto a Hiload Q column and 42 eluted using a linear gradient of buffer G. The protein preparations were stored with 20% glycerol at -80\u00C2\u00B0C prior to use. 3.2.7 Expression and purification ofstx190..264 for ITC 6L of cell culture was grown at 37\u00C2\u00B0C until the OD6reached 0.4, and then growth was continued at 25\u00C2\u00B0C until an 0D600 of 0.6. After induction with 0.4mM IPTG, the cell culture was grown at 25\u00C2\u00B0C overnight. The clarified supernatant was loaded onto a glutathione sepharose column in buffer L and eluted in buffer M. After dialysis in buffer H at room temperature for 1 hour, the protein sample was loaded onto a Hiload Q column and eluted in a linear gradient of buffer I. The eluate was collected and dialyzed in buffer A along with Prescission protease (kindly provided by the lab of Dr. N. Strynadka) for two days at 4\u00C2\u00B0C. The protein sample was applied onto a glutathione sepharose column in buffer L, and the flow-through was collected. The final protein preparation was stored in 20% glycerol at - 80\u00C2\u00B0C prior to use. All the chromatography buffers used contained Tris pH 8.0 and 14mM ME. 3.3 Crystallization 3.3.1 Vapor diffusion crystallization Both sitting drop and hanging drop setups were used in this study. The sitting drop was used for screening random conditions using the following crystallization kits from Qiagen: PEGs, Classics, Classics II, pH clear, pH clear II, JSCGH- and ProComplex in 96-well plates (low profile Greiner, Hampton Research). A multichannel pipette was 43 used to mix 1 il from 6ojil of the well solution with 1 jil of the protein drop. The plates were sealed with the crystal clear tape (Hampton Research). The hanging drop method was used for optimization in 24-well plates (VDX, Hampton Research). 1 jil of the protein sample was mixed with liii of the precipitant on a cover slip, and the cover slip was inverted over 500!Il of the precipitant solution. 3.3.2 Counter diffusion crystalization In the counter diffusion method, the Crystal Former (Microlytic) was used. The Crystal Former is a capillary device with 16 crystallization channels, and each provides access to two inlets for loading proteins and precipitant solutions. 0.5iil ofprotein was loaded into the right inlet, while 0.5jil of the precipitant solution was loaded into the left inlet. The inlets were sealed by crystal clear tape. 3.3.3 Dialysis crystallization Dialysis crystallization was carried out in 24-well plates. 5 j.tl of the protein samples were loaded into Siil dialysis buttons (Hampton Research) and sealed with a 1000 Da MWCO membrane. The dialysis buttons were placed in 1.5 ml of the well solution. The 24-well plates were sealed with tape. 3.3.4 Seeding In the seeding experiments, the natural fiber Seeding Tool (Hampton Research) was used to transfer small seeds from the spherulites or the needle clusters to a clear drop that was set up one day before by streaking a line in the drop. 44 3.4 Isothermal titration calorimetry (ITC) Both VP-ITC (MicroCal) and ITC200 (MicroCal) instruments were used to test the interaction between stx1189 and synA 1 or synB 1. Samples were concentrated and dialyzed in the buffers containing the different experimental conditions. When using VP ITC, samples were degassed for 5 minutes, and injections of lOjil ofstx1189 were titrated into 1.4 ml of synAl or synBi. When using ITC200, injections of 2il of stx1189 were titrated into 200 jil of synA I or synB 1. For all the measurements, stxl1g9was titrated into the experimental buffers as a control. The ITC200 instrument was used to test the interaction between StXI9O2M and synA I or synBl. The protein samples were dialyzed against 150mM KC1, 10mM Hepes pH 7.4, 20p.M CaC12 and 10mM f3ME. Injections of 2j.il ofstx190264 were titrated into 200il of synA 1 or synB 1. For all the measurements, stx190264 was titrated into the experimental buffer as a control. The protein concentration was determined by reading the absorbance at 280nm under denaturing conditions (Edelhoch, 1967). The experimental temperature was 25\u00C2\u00B0C. Data were processed with MicroCal Origin 7.0. 3.5 Circular dichroism (CD) The CD spectra were measured for synA 1 and synB 1 using a Jasco Model J-8 10 spectrometer (Jasco). Wavelength scans from 195 to 280 nm were taken at 25\u00C2\u00B0C in a 2 mm path length quartz cuvette. Both synAl and synBlspectra were measured at a protein concentration of lOjiM. The proteins were prepared in 10mM phosphate pH6.5. The 45 protein concentration was determined by reading the absorbance at 280nm under denaturing conditions (Edeihoch, 1967). 46 4. Results 4.1 Crystallization of the synprint site from Ca2.1 and Ca2.2 In the past 15 years, the interaction between the synprint site and SNAREs has been intensively studied by multiple biochemical and physiological techniques (Catterall and Few 2008, Kisilevisky and Zamponi 2008). However, some fundamental questions still remain, including how exactly the synprint site binds to SNARE proteins, why the interaction shows isoform specificity, and how the Ca2 dependence occurs. A detailed description of the interaction at the molecular level is needed to address these questions. In order to obtain a molecular template for further dissection of the interactions, we attempted to use protein crystallography to determine the three-dimensional structure of the synpririt site. 4.1.1 Designing different synprint constructs from Ca2.1 and Ca2.2 In crystallographic studies, designing suitable protein constructs is a crucial step in the success of the experiment. Based on sequence alignment of different isoforms from different species (Figure 7a), the synprint site can be divided into four parts: a putative coiled-coil domain, a conserved region, a region corresponding to homology with collagen, and a C-terminal non-conserved region. Seven amino acid residues (DNLANAQ) before the start of the proposed synprint site (indicated as \u00E2\u0080\u009Cstart arrow\u00E2\u0080\u009D) are also predicted to be part of the coiled-coil domain (Figure 7a). We therefore included this part in our study. Moreover, the entire synprint site was already shown to be sensitive to degradation, being cleaved rapidly into two parts. Given this experimental result, we also divided the synprint site into two halves: an N-terminal half that starts from the 47 coiled-coiled domain and ends at a \u00E2\u0080\u009Cproline\u00E2\u0080\u009D residue indicated as \u00E2\u0080\u009Cred arrow\u00E2\u0080\u009D in Figure 7a, and a C-terminal half that expands the rest of the non-conserved region. This experimental design strongly agrees with the previous predication that the synprint site consists of two domains (N-terminal and C-terminal) (Yokoyama et al, 2004). In addition, the C-terminus of Ca2.1 synprint contains 7 consecutive glycine residues, which has already been shown to be unstable to proteolysis. According to the information listed above, 11 synprint constructs from rabbit Ca2. 1 and bovine Ca2.2 have been made for crystallization trails (Table 2). 4.1.2 Purification of different synprint constructs from Ca2.1 and Ca2.2 All the synprint constructs were cloned into the pET28HMT vector, and were fused with a I-us6 tag, a maltose binding protein (MBP), and a specific protease cleavage site for the Tobacco etch virus protease. The result of the purification, as described in the materials and methods, is shown in Figure 9a. We successfully purified all of the synprint constructs listed in Table 2 except one construct, Rab alAsynA7. The protein seemed to precipitate after the second poros MC column. The reason is likely that the short construct did not fold properly, thus precipitating after removal of the MBP tag. The rest of the synprint constructs all remained stable throughout the entire purification and were well-behaved. However, all of them shared a small degradation problem, and we were unable to completely remove the degradation products (Figure 9a Lane 6). Nevertheless, the degree of degradation was different among these constructs. An interesting 48 (a) Figure 9 Purification of the synprint domains. Bovol bsynA8C784A was used as an example. Bands in Lane 2 to Lane 6 (a) represented the elution of the first poros MC column (b), the elution of the amylose column (c), the flow-through of the second poros MC column (d), the elution of the Hiload Q column (e) and the major peak of the superdex column (f). 72kDa .-I 55kDa\u00E2\u0080\u0094I 43kDa\u00E2\u0080\u0094\u00E2\u0080\u0099 34kDa\u00E2\u0080\u0094 26kDa \u00E2\u0080\u0094 l7kDa\u00E2\u0080\u0094 1 OkDa \u00E2\u0080\u0094 (b) rnAU 1500 1000 500 0 Ladder Poros MC Amylose Poros MC Hiload Q Superdex (c) ml inAU 200 300 4O0m 0 49 mAli 800 600 400 200 (d) (e\u00E2\u0080\u0099) mAU 15 10 0 50 100 150 200 ml 0 100 200 300 400 500 nil (1) mAli 40 30 20 10 10 30 ml 50 observation was that all the constructs with the additional seven amino acids (DNLANAQ) at the N-terminus seemed to be more stable than the others. We speculated that these residues were indeed part of the coiled-coil and that their presence could stabilize the structure of the entire protein. 4.1.3 CD experiment of the N- and C-terminal of the synprint site A CD experiment was performed to test the secondary structures of the N- terminal part (synAl) and the C-terminal part (synBi) of the synprint site. Figure 10 showed that the N-terminal part of synprint indeed showed some ci helical feature (48.4% x helix), which confirmed the prediction, while the C-terminal non-conserved part did not display any secondary structure but only random coils (13.4%a helix) (Greenfield and Fasman 1969). Therefore, we focused our efforts into crystallizing the N-terminal part because the C-terminal part without a well-organized structure would not have much chance to crystallize. 4.1.4 Crystallization of different synprint constructs from Ca2.1 and Ca2.2 All well-behaved proteins were used for random condition crystallization screening. Spherulites appeared for almost all of the synprint constructs (Figure 1 la). This is an indication that the synprint domains from both Ca2.1 and Ca2.2 had a good chance to be crystallized. Moreover, there were needle-like small crystals growing on the top of spherulites from BovulBsynA4C784A and BovCLIBsynA8C784A constructs (Figure 11 b). The cysteine to alanine mutation could prevent potential non-specific di 51 i:; (b) -4 c -6 CD -10 -12 -14 -16 Figure 10 CD spectra of synAl (N-terminal) and synBi (C-terminal). (a) CD spectrum of synA 1 showed c helical feature. (b) CD spectrum of synB 1 displayed random coils. (a) 0 -5 -10 -15 -20 -25 -30 wavelength (nm) 52 (a) (b) Figure 11 Spherulites and small needle-like crystals from crystallization trials of the synprint domains. (a) Spherulites or small crystalline materials. (b) The red arrow indicated the thin hair-like crystals growing on the spherulites. 53 sulfide bond formation. These small needle-like crystals tended to grow separately from each other, which was also a good sign for potential growth of single crystals. From these constructs, we discovered that the collagen homology region was important for crystallization. The needle-like crystals showed up in Bovu 1 BsynA4 C784A but not its homolog Rab xlAsynA4. The difference between Bovx1BsynA4 C784A and Rab x1AsynA4 was that BovcdBsynA4 C784A contained a collagen homology domain, but Rab zlAsynA4 contained a non-conserved sequence at this site (Figure 7a). However, unfortunately neither the spherulites nor the needle-like crystals appeared large enough for X-ray diffraction experiments. More effort was focused on fine screening of the two constructs (Bovx1BsynA4C784A and BowzlBsynA8C784A) around the conditions where the needle-like crystals showed up. However, the improvement in protein crystal quality was limited, and no single three-dimensional crystals were found in the fine screening. 4.2 Crystallization of synprint domains with surface engineering Protein crystallization predominantly depends on entropic effects. Mutational surface engineering, creating patches with low conformational entropy, is an effective tool to enhance the success of protein crystallization (Derewenda and Vekilov 2006). BovcdBsynA8C784A was chosen as the template for the surface engineering. There were three reasons for picking this synprint construct. Firstly, Bovct I BsynA8C784A starts from the sequence DNLANAQ, which we found to play a potential role in stabilizing the synprint domain. Secondly, both the N-terminal region (coiled-coil) and the C-terminal region (collagen homolog) of this construct are predicted to be alpha 54 helices, which could coil around one another. Thirdly, small needle-like crystals were observed for this construct (Section 3.1.3). The SER on-line server (http://nihserver.mbi.ucla.edu/SERI) was applied to predict the potential \u00E2\u0080\u009Chot spots\u00E2\u0080\u009D that are most suitable for mutations designed to enhance crystallizability by surface engineering. There were three hits: 737KAKE40,794EE5,and 820EAGE3. 4.2.1 Purification of BovcdBsynA8C784AIE794AIE795A Previous work has shown that degradation is a big issue for purification and crystallization of the synprint domains. To minimize degradation, three procedures were implemented. Firstly, a protease stability test was performed on synA8EE (short for BovctlBsynA8C784A/E794A/E795A) (Figure 12a). The basic pH 8.0 was chosen for protein purification. At the acidic pH 5.0, the degree of degradation was the smallest, but this condition was not chosen because the poros MC column would not be functional at this pH. Furthermore, given the p1 5.35 of synA8EE, the ion exchange columns would have poor separation. Secondly, the TEV cleavage time was shortened to 3 hours instead of overnight incubations. Thirdly, three additional columns were used for protein purification, and all the purification steps were performed as fast as possible, within two days (Section 2.2.3). As a result, one can see that this greatly reduced the amount of degradation (Lane 9 in Figure 13a versus Lane 6 in Figure 9a). In addition, the chromatogram of the second 55 (a) 72kDa\u00E2\u0080\u0094 \u00E2\u0080\u0094 55kDa\u00E2\u0080\u0094 43kDa- \u00E2\u0080\u0094 34kDa\u00E2\u0080\u0094 26kDa\u00E2\u0080\u0094 \u00E2\u0080\u0094 17kDa\u00E2\u0080\u0094 1 OkDa \u00E2\u0080\u0094 (b) 72kDa\u00E2\u0080\u0094 55kDa\u00E2\u0080\u0094 43kDa\u00E2\u0080\u0094 34kDa\u00E2\u0080\u0094 26kDa\u00E2\u0080\u0094i 1 7kDa \u00E2\u0080\u0094 Figure 12 Stability tests. (a) Stability test of synA8EE at different plls. (b) Stability test of MBP-5A-synA8EE at different pHs. pH3.O pH4.0p115.O pH 6.0 pH7.0 pH8.0 pH9.OpHlO.0 pH3.0 pH4.0 pH5.O pH 6.0 pH7.0 pH8.0 pH9.0 pHlO.0 56 size exclusion column displayed a single symmetrical peak, indicating that the protein was well-behaved (Figure 13i). 4.2.2 Crystallization of synA8EE synA8EE protein preparations were used to set up random condition crystallization screening. The best outcome from these screenings was needle-like small crystals growing on spherulites (Figure 14a). These needles were bigger and thicker than observed before. A lot of effort was put into the optimization of these synA8EE crystals. The best result from these fine screens was the appearance of long thin needle clusters sharing the same origin (Figure 14b). This was a large improvement. These long needle crystals (\u00E2\u0080\u0094O.2mm) were well separated from each other. However, the weakness of the needle crystals was that they were flexible and only grew in one dimension instead of three dimensions. Unfortunately, because of these problems, these needle crystals showed no visible diffraction at the CMCF beamline of the Canadian Light Source (Saskatoon Canada). There might be two underlying causes for this unsatisfying crystal quality. Firstly, the flexibility of the needle crystals suggested weak crystal contacts. Therefore, surface engineering of the other two \u00E2\u0080\u009Chot spots\u00E2\u0080\u009D was performed on synA8EE. However, synA8EE combined with the E820A surface mutation did not improve crystal quality. The synA8EE with K737A!K739A/E740A exhibited too much degradation during purification, indicating that the lysine to alanine mutations compromised the protein integrity. Secondly, we suspected that protein degradation was also influencing crystal 57 Figure 13 Purification of synA8EE. Bands in Lane 2 to Lane 9 (a) represented the elution of the first poros MC column (b), the elution of the HQ column (c), the elution of the amylose column (d), the major peak of of the superdex column (e), the elution of the second poros MC column (f), the elution of the Hiload Q column (g), the flow4hrough of the amylose column (h), and the peak of the superdex column (i). (a) 72kDa\u00E2\u0080\u0094 55kDa- 43kDa/ 34kDa\u00E2\u0080\u009D 26kDa 1 7kDa\u00E2\u0080\u0094_j lOkDa\u00E2\u0080\u0094 (b) cc) Ladder Poros Hiload Q Amylose superdex Poros Hiload Q Amylose Superdex mAU 500 400 300 200 100 50 100 150 200 250 300350 mAU 0 50 100 150 200 ml 58 (d) (e) mAU mAU\u00E2\u0080\u0099 1600 400! 300 200 1o0 0. (I) (g) mAU 120 80 40 0 (1) U) mAU mAU 80 120 60 80 40 40 59 800 400 0 50 100 150 200 250 ml 0 0 50 100 150 ml mAU 0 0 io 150 0 ml 0 50 ml (a) (b) Figure 14 The needle-like crystals obtained from synA8EE. (a) The short needle-like crystals from the random screen. (b) The long needle-like crystals after optimization. 60 quality. Mass spectrometry was used to identify the degradation sites (quadrupole ESI, UBC); however, it turned out that only one species representing synA8EE was identified(Figure I 5a), even though small degradation bands could even be seen on Coomassie-stained SDS-PAGE gel (Figure I 3a Lane 9). This unexpected result might be due to the fact that the signal of major species of synA8EE was so strong that the signals of minor species of degradation products were masked in the mass spectrum. Seeding is a very useful crystallographic tool for small crystals to grow high-quality crystals. Several conditions around the best crystallization hit for synA8EE were set up using lower concentrations of the protein. Seeding experiments were performed on the next day. However, along the streaking, only small needle clusters were observed without any improvement of crystal quality (Figure 1 5b). 4.2.3 Crystallization of synASEE using different crystallization methods If one crystallization method has been exhausted, the quality of crystals may be improved by using other crystallization techniques. 4.2.3.1 Crystallization of synA8EE using dialysis crystallization method In crystallographic studies, the existence of salt/buffer in the final protein preparation could always be a problem because one never knows if the salt/buffer prevents crystal formation. Obviously, H20 would be the best choice for the solvent. However, most of the time proteins need some salt and buffer for stability. 61 (b) 5259.9 0909 9.M1l 209(3051) Ml !Ev-03858,1t27] (G0,0750.1002395.0 0.0.133,933) C.1, (206219) 10\u00E2\u0082\u00AC\u00E2\u0080\u0099 \u00E2\u0080\u00A2 12500 14500 .iiiw Figure 15 Mass spectrum of synA8EE and the result of seeding experiment. (a) The mass spectrum of synA8EE. According to the mass spectrum, the protein preparation contained a single peak represented synA8EE. No other protein contaminates were I 310065, 151040 3.550) 150750 62 detected. (b) The result from seeding experiments. Only spherulites with small needle like crystals were observed. In this study, when the concentration of salt (KC1) was reduced below 10mM, synA8EE started to precipitate. This observation led to the idea that if the desalting process could be slowed down, it might be possible for synA8EE to crystallize during the process. Dialysis experiments were therefore set up (Figure 16a). In the horizontal direction, KC1 concentration was gradually increasing from 1mM to 100mM, and in the vertical direction, different pHs (3.0, 4.0, 5.0 and 6.0) were screened. In this screen, only spherulites were observed (Figure 16c). Therefore, this was unsuccessful for improving synA8EE crystallization. 4.2.3.2 Crystallization of synA8EE by counter diffusion crystallization method In the counter diffusion method, a gradient was formed of both protein and precipitant concentrations mixture is formed along the length of the capillary (Ng et al 2003). Therefore, the counter diffusion is a good optimization technique. In this study, the synA8EE protein was placed on the right side of the capillary, while the best condition with different pHs (3.0-10.0) was loaded on the left side of the caplillary (Figure 1 6b). However, only spherulites were observed (Figure 1 6d). Therefore, the counter diffusion method failed to grow better 3D crystals. 4.3 Crystallization of fusion proteins Previous studies have shown that large-affinity tags can help stabilize the proteins of interest for crystallization (Smyth et al 2003). In this study, both MBP and GST tags 63 (a) b) (c) (d) Figure 16 Results of dialysis and counter diffusion crystallization. (a) The setup of dialysis crystallization. (b) The setup of counter diffusion crystallization. (c) The spherulites were obtained from the dialysis crystallization. (d) The spherulite seen in the counter diffusion method. 64 a \u00E2\u0080\u0094 I \u00E2\u0080\u0094 were used as fusion tags for synA8EE, and 0, 3-alanine and 5-alanine linkers were used to test for the optimal length for crystallization. 4.3.1 Purification of MBP-(linker)-synA8EE-His and GST-(linker)-synA8EE-His Initially, the fusion proteins with only MBP or GST tags had degradation issues during purification (Figure 17). Therefore, we chose to put affinity tags (an MBP or GST tag and a His tag) on both ends, and used poros MC and amylase/glyutathione sepharose columns to select only intact fusion proteins (Section 2.2.5). As expected, this purification scheme indeed increased the purity and completely removed degradation products (Figure 1 8a). A protease stability test showed that the fusion protein was stable at all the tested pHs except pH3.0 (Figure 1 2b). The size exclusion column also confirmed that the fusion protein was well-behaved (Figure 1 8g). We observed that the GST-fusion proteins experienced more degradation than MBP-fusion proteins (Figure 17 and Figure 18). It appeared that instead of stabilizing the synA8EE, the GST-affinity tag interfered with the folding of synA8EE and made synA8EE more vulnerable to proteolysis. 4.3.2 Crystallization of MBP-(linker)-synA8EE-His All three MBP-fusion proteins with linkers of different lengths were successfully purified (3Omg/ml) and subjected to random screening. We observed a correlation between the linker length and quality and quantity of spherulites. The fusion protein 65 \u00E2\u0080\u0094 72kDa \u00E2\u0080\u00A2 \u00E2\u0080\u009455kD I! \u00E2\u0080\u009433kDa \u00E2\u0080\u00A2 \u00E2\u0080\u009434kDa \u00E2\u0080\u0094 26kDa Figure 17 Hiload Q results of MBP-tagged synA8EE and GST-tagged synA8EE. (a) The result of the MBP-tagged synA8EE from the Hiload Q column. Degradation was observed. (b) The result of the GST-tagged synA8EE from the Hiload Q column. The amount of degradation was more than that of MBP-tagged synA8EE. (c) The Hiload Q profile of MBP-tagged synA8EE. (d) The Hiload Q profile of GST-tagged synA8EE. (b)(a) 72kDa \u00E2\u0080\u0094 55kDa - \u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00E2\u0080\u0094 43kDa - 34kDa \u00E2\u0080\u0094 26kDa - ___ l7kDa \u00E2\u0080\u0094 \u00E2\u0080\u0094 (c) mAU 200 150 100 \u00C2\u00A30 0 (d) ,nAU 0 100 200 300 400 500 ml 66 Figure 18 Purification of MEP (GST)-Iinker-synA8EE-flis. (a) The purification of MBP-5A-synA8EE-His. Bands in Lane 2 to Lane 5 represented the elution of the first poros MC column (c), the elution of the amylase column (d), the elution of the second poros MC column (e) and the elution of the Hiload Q column (f). (b) a large amount of degradation was seen during the purification of GST-tagged fusion protein. (g) the superdex profile of the MBP-5A-synA8EE-His indicated that the protein was well- behaved. (a) 72kDa\u00E2\u0080\u0094\u00E2\u0080\u0099 S5kDa\u00E2\u0080\u0094\u00E2\u0080\u0099 43kDa\u00E2\u0080\u0094 34kDa\u00E2\u0080\u0094i 26kDa\u00E2\u0080\u0094 l7kDa\u00E2\u0080\u0094 (b) 72kDa\u00E2\u0080\u0094 55kDa \u00E2\u0080\u0094 43kDa\u00E2\u0080\u0094 34kDa\u00E2\u0080\u0094 26kDa \u00E2\u0080\u0094 67 (c) (d) - mAU mA 600 500 400 400 300 _j 0 50 100 150 200 250m1 (e) (f mAU znAU 500 400 A 400 300k 0o 200 200 100 100 --\u00E2\u0080\u0098 0 0 100 200 300 400 nil (g) mAU s0 40 30 10 0 0 10 15 20 25m1 100 zoo 50 100 150 200 Z50 68 without the linker gave spherulites for only 12 conditions, but no needle-like crystals. In contrast, the 5 alanine-linker fusion protein yielded 73 conditions with spherulites, and small needle-like crystals appeared in 17 conditions. The best results, obtained after fine screening of MBP-5A-synA8EE-His, are shown in Figure 19. However, the quality of these crystals is still poor and more optimization is needed. 4.4 Interaction between the synprint site and syntaxinlA It is well established that the modulation of the presynaptic voltage-gated calcium channels by SNARE proteins is mainly through the interaction between the synprint site and SNARE proteins. However, the precise binding sites are still under debate. For example, the Catterall group suggested that the H3 domain (SNARE motif) of syntaxin 1 A is responsible for interaction with the synprint site (Yokoyama C T et al, 2005), while the Zamponi lab showed that the Habc domain of syntaxinlA is sufficient for the binding (Jarvis S E et al, 2002). In this study, the biochemical technique isothermal titration calorimetry (ITC) was used to further map the binding sites between the synprint site and syntaxin 1 A. Isothermal titration calorimetry (ITC) is an exquisite biochemical tool to study protein-protein interactions by determining the thermodynamic parameters (enthalpy, entropy, stoichiometry). In this study, ITC was used to test if the N-terminal part of syntaxinlA stx1189 could bind to the N-terminal half (synAl from Rabbit Ca2.1) and the C-terminal half (synBl from Rabbit Ca2.1) of the synprint site (Figure 20). The purity 69 (a) (b) Figure 19 The needle-like crystals obtained from MBP-5A-synA8EE-His. (a) These needle-like crystals were short but thicker. (b) These needle-like crystals were long but thinner. 70 of the proteins used is shown in the figures 21-23. 5OOtM stx1189 was titrated into either 5OtM synAl or synBi. The S-shaped curves from the ITC profiles indicated that stx1189 bound to both synAl and synBi in 1:1 fashion with Kd152\u00C2\u00B1l2nM (AH-8.09\u00C2\u00B1O.26 xlO3cal/mol, AS=-O.859 cal/mol/K) and Kd= 1.8 1\u00C2\u00B1O.34uM (AH=r4.45\u00C2\u00B1O.O2x 1 & callmol, AS=-16.3 cal/mol/K), respectively (Figure 25 a and b). This result suggested two different aspects. Firstly, both the N-terminal half and the C-terminal half of the synprint site contributed to the interaction. Secondly, the N-terminal part of syntaxin 1 A containing Hai, domain itself showed strong affinity for the synprint region. However, the initial positive result was not reproducible using the original conditions (150mM KCI, 10mM Hepes pH 7.4, 20tM CaC12,and 10mM ME); moreover, the negative results suggested that there was no binding between stx1189 and the synprint site whatsoever. Then, a set of ITC experiments with different conditions was carried out to try to reproduce the initial positive result. The conditions were chosen to test which parameters might be crucial for the interaction, but which may have been recorded incorrectly. We changed the concentration of KC1 (10mM versus 250mM), and CaC12(0 versus 1M). All four fTC results were negative and showed no binding. The Chapman group reported that the non-specific binding of DNA to syntaxin IA, formed during lysis of the host cells, hindered syntaxin 1A from interacting with other protein partners. Therefore, we removed DNA by MnSO4precipitation. Unfortunately, we still obtained the same negative result. In previous puildown or Co-IP experiments showing interaction between syntaxin 1A and the synprint site, only crude cell lysates were used. Therefore, it is possible that some other proteins in the cell lysates assist in the binding. To address this point, relatively \u00E2\u0080\u009Cdirty\u00E2\u0080\u009D protein preparations of both synAl and stx1189 71 (a) Rabbit Cav2.1 synprint Figure 20 The cartoon representation of the synprint site and syntaxin 1A. (a)the different domains of the synprint site from rabbit Cav2. 1. (b)the intracellular part of syntaxin IA. 715 876 1004 (b) I SynAl SynBi I SyntaxinlA 27 146 189 264 I Nter Habc 1unI H3 72 niAU 60 (13) 72kDa\u00E2\u0080\u0094 I 55kDa_ 43k.Da \u00E2\u0080\u0094 34kDa 26kDa_ S. ,1 uiAU1 1500 10001 500 0 0 100 100 300 400 ml Ladder Poros MC Poros MC Hiload Q (c) (d) 60 40 0 20- 40 60 ml 50 100 -o ZOO ml 40 20 0 0 Figure 21 Purification of stxl-189. Bands in Lane 2 to Lane 4 (a) represented the elution of the first poros MC column (b), the flow-through of the second poros MC column (c), and the elution peak of the Hiload Q column, which is indicated by the arrow (d). 7\u00E2\u0080\u0099 (a) b) 72kDa \u00E2\u0080\u0094 S5kDa\u00E2\u0080\u0094 43kDa \u00E2\u0080\u0094 34kDa \u00E2\u0080\u0094 26kDa l7kDa\u00E2\u0080\u0094 Figure 22 Purification of synAl. Bands in Lane 2 to Lane 4 (a) represented the elution of the first poros MC column (b), the flow-through of the second poros MC column (c), and the elution of the Hiload Q column (d). mAU 3000 2000 1000 0 100 (c) ml (d) Ladder Poros MC Poros MC Hiload Q mAU 500 400 300 200 100 0 mAU, 150 100 50 0 100 0 50 100 150 200 250 200 300 ml ml 74 (a) (b) 72kDa \u00E2\u0080\u0094 55kD \u00E2\u0080\u0094 43kDa\u00E2\u0080\u0094 34kDa\u00E2\u0080\u0094 26kDa\u00E2\u0080\u0094 _ l7kDa\u00E2\u0080\u0094 (C) inAU 1200 800 400 0 Ladder Poros MC Poros MC Hi load Q mAU 2000 1500! 1000 500! 0 kL 0 Cd) 40 20 Figure 23 Purification of synBi. Bands in Lane 2 to Lane 4 (a) represented the elution of the first poros MC column (b), the flow-through of the second poros MC column (c), and the elution peak of the Hiload Q column (d). 100 200 300 ml 80 60 50 100 0 ml 0 50 100 150 200 ml 75 (a 72kDa\u00E2\u0080\u0094 33kDa - 34kDa\u00E2\u0080\u0094 26kDa \u00E2\u0080\u0094 l7kDa \u00E2\u0080\u0094 niAU 80 :: 20 20 0 100 200 300 10 S 0,n Figure 24 Purification of stxl9O-264. Bands in Lane 2 to Lane 4 (a) represented the elution of the glutathione sepharose column (b), the elution of the Hiload Q column (c), and the flow-through of the second glutathione sepharose column (d). mAU 3000 2000 1000 lOkDa \u00E2\u0080\u0094 LC) Ladder glutathione Hiload Q glutathione sepharose n 0 50 100 150 ml sepharose (d) L ml 0 40 80 120m1 76 (a) Time (mm) -10 0 1020 30405060 7080 901001l0120l3040 0.4 0.2 0.0 -0.2 - -0.4 (0 -0.6 -0.8 -1.0 1.2 . 9 -\u00E2\u0080\u00A2 0 N 2 I 4 / -6 (b) Time (mm) -100 1020304oso607o80901oo12034c15a60 0.0 -0.5 o -1.06) 8) 5 -1.50 -2.0 -2.5 0- 0.0 0.5 1 0 1.5 2.0 2.5 Molar Ratio Figure 25 The interaction between domains of syntaxin 1A and parts of the synprint site using ITC. (a) The interaction between stx1189 and synAl. (b) The interaction between stx1189 and synB 1. (c) The interaction between stx190-264 and synA 1. (d) The interaction between stx190264 and synB 1. _._ -- I J. N \u00E2\u0080\u0094II. CO C.) 8) 0 6) 0 E CO C.) C, 1) CO 00 05 10 1.5 2.0 Molar Ratio -2 -4 -6 -8 (d)(c) CO C-) 6) 0 6) 0 E CO C) C-) 8)(0 CO C.) Time (mm) 0 10 20 30 40 50 6 70 0.03 ! ithUiliil. -0.03 -0.04 -0.05. 0.4. CO C.) C 0 6) 0 CO 0 \u00E2\u0080\u0094 0.3. C CO 0.2 C o 0.1- 8) 0 E 0.0. CO C) - .0_I . . Na a a a \u00E2\u0080\u0094 S \u00E2\u0080\u00A2 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Molar Ratio 0.0 0.5 10 Molar Ratio 1.5 2.0 77 which had only experienced the poros MC column were used for ITC. The result indicated that there was no binding as well. Another experimental parameter that could affect the ITC result was the concentration of glycerol. The residual glycerol from the incomplete dialysis in the protein preparations might help the binding by maintaining the protein integrity. Thus, keeping glycerol in the protein preparations might solve the reproducibility problem. Indeed, a positive ITC result was seen by including 20% glycerol in the protein samples (Figure 26c). However, the Kd was only 7.94jiM, about 40 fold weaker compared to the initial ITC result. It is highly possible that this last ITC result is simply a false positive due to the heat of diluting glycerol. Another aspect that was under consideration was the effect of protein degradation. Previous studies have shown that presynaptic protein Munc 18-1 interacts with syntaxin 1AISNAP-25 heterodimer through the small N-terminal end (residues 1-27) of syntaxin 1A (Dulubova et al 2007). If the small N-terminal end of syntaxin IA was important for interacting with the synprint site, but cleaved off during purification, then it would explain the negative results. Therefore, mass spectrometry was used to determine if both synAl and stx1189 were intact (MALDI-TOF, NAPS). The result showed that the MS signals of both synAl and stx1189 were bigger than the calculated intact proteins (due to imprecision of MALDI for proteins this size), thus suggesting that proteolysis was not the cause of the latter negative results (Figure 26a and b). ITC was used to test if the C-terminal part of syntaxin stxl9o2M (the H3 domain), the coiled coil region bound to synAl or synBi. The purity of the proteins used is shown 78 in the figures 19, 20 and 22. The ITC results revealed that there was no interaction between 5txl2M and both halves of the synprint site (Figure 25 c and d). In all, there is no binding between the H3 domain of syntaxin 1A and the synprint site, and whether the N-terminal part of syntaxin IA interacts with the synprint site is still inconclusive. 79 w_ (c) Time (mAn) \u00E2\u0080\u00A2e C CS fCC 330 Sf4994 9Sf 000CeaCue .00 4.95 0.04 -0.00 -e AS - .15 -0 2 5 -0.2 \u00E2\u0080\u00A2 -CA -GA -C-. -1_u Figure 26 Mass spectrometry of synAl and synBi and ITC result in the presence of glycerol. (a) The molecular mass of synA 1 obtained from mass spectrometry was 19100.03 Da, greater than the theoretical molecular weight 19020 Da. (b) The molecular mass ofstx1189 obtained from mass spectrometry was 22172.0 Da, greater than the theoretical molecular weight 22071.5 Da (c) ITC result of the interaction between stx1189 and synAl in the presence of 20% glycerol. (b)(a) 00 0.5 2.0 1.5 2_C 2. Molar Ratio 80 5. Discussion 5.1 Crystallization The synprint site from Ca2. I and Ca2.2 has been proposed to be involved in channel gating, kinetics, and localization. Solving the synprint structure will provide further information at a molecular level to study how the synprint site is involved in channel regulation. In this study, 11 synprint domain constructs have been made, and 10 of them have been successfully purified and set up for crystallization trails except for the very minimal construct (RabcLlAsynA7). We did not focus on the C-terminal non-conserved region (Rab cdAsynBl) because both CD experiments (Figure lOb) and the sequence (absence of hydrophobic residues) (Figure 7a) suggested it to be unfolded. However, because the diversity, the C-terminal region may be responsible for the isoform specificity and Ca2 dependence of the synprint sites from different isoforms to interact with SNARE proteins (Figure 7). Unfortunately, our structural study was not able to provide the answer to these questions. The rest of the 9 constructs, made up of the different regions of the N-terminal part of the synprint site, were all well-behaved (Figure 90. Out of the 9 constructs, the best outcome came from the thin needle-like crystals (Figure 1 ib). Moreover, from these constructs, we observed that the conserved DNLANAQ sequence at the N-terminus appeared to help stabilize the synprint site, and the collagen homology region was important in protein crystallization. It was thought that the predicted u-helices of both the N-terminal (coiled-coil) and the C-terminal (the collagen homolog) could coil together and improve the stability of the protein. 81 It was suspected that the flexible surface residues and protein degradation might prevent the growth of single 3D crystals. In order to solve these two problems, surface engineering and a new two-day purification scheme with additional chromatography columns and different buffers were applied. Indeed, after minimizing the degradation effects and creating low-entropy patches on the protein surface, long, well-separated needle-like crystals from synA8EE were obtained (Figure 1 4b). However, the problem with these crystals was that they were flexible, displayed growth in one dimension, and did not diffract X rays. There might be several reasons for this result. Firstly, the flexibility of the needle-like crystals indicated that the crystal contacts were not strong. Therefore, further surface engineering might solve the problem. However, it turned out that mutating more surface residues did not help with crystallization. Secondly, the proteolysis issue still remained. Even though different procedures were used to protect the protein integrity, degradation products were still present in the final protein preparation (Figure 13a Lane 9). Furthermore, the long needle-like crystals appeared at pH3.0. According to the stability test, degradation still occurred at pH 3.0, and during the crystallization process, some proteins were still undergoing proteolysis, which could interfere with the crystal growth. Therefore, mass spectrometry was used to map the cleavage sites. Hopefully, by mutating the protease cleavage sites, the degradation issue could be resolved, and the crystal quality could be improved. Unfortunately, mass spectrometry was not able to identify any degradation products (Figure 15a). We also tried different crystallization techniques such as seeding, counter diffusion, and dialysis to obtain better crystals; however, none of them improved crystal quality. It thus seemed impossible to obtain good crystals of the synprint domains by themselves. 82 Therefore, we tried to crystallize the synprint site with attached protein tags (MBP or GST and His tags) because proteins tags can stabilize the structure of the protein of interest. Indeed, the protein degradation was substantially limited (Figure 1 8a). One major problem of this method is the conformational heterogeneity allowed by the flexible linker region. To solve this problem, we change the length of the linkers (no linker, 3 alanines and 5 alanines). We observed that the longer linkers resulted in crystals with improved quality. Given this trend, in the future, fusion proteins with even longer linkers should be tested. Moreover, if we cannot obtain high-quality crystals, we should try to solve the structure of the synprint site by NMR. 5.2 Interaction between the synprint site and syntaxin 1A It has been well established that the interaction between the synprint site and SNARE proteins (syntaxin IA and SNAP-25) is responsible for the modulation of presynaptic Cas by SNARE proteins (Bezprozvanny et al 2000, Yokoyama et al 2004, Evans and Zamponi 2006, Keith et al 2007). Moreover, researchers have proposed that the binding of the different domains of syntaxin IA to the synprint site regulates different types of channel behaviour. The interaction between the H3 domain (SNARE motif) of syntaxin IA and the synprint site is responsible for syntaxin lA-mediated channel inhibition (Rettig et al 1996, Bezprozvanny et al 2000). The interaction between the Habe domain of syntaxin 1A and the synprint site anchors syntaxin IA on the Ca2.2. Because of the Gfry subunit-H3domain interaction, syntaxin 1A recruits GI3y subunit to the Ca2 channel to promote G protein-mediated channel inhibition (Jarvis et a! 2002). 83 In this study, ITC experiments were used to map the binding sites between the synprint site and syntaxin 1A. The ITC experiment showed no binding between each half of the synprint site and the H3 domain of syntaxinlA (Figure 25c and d). However, in the previous study, a strong interaction was detected by pulldown assays (Sheng et al 1996). Perhaps, the pulidown/western blotting is a more sensitive biochemical technique than ITC. However, it is unlikely that ITC could not detect the reported nanomolar binding. One explanation is that either half of the synprint site shows weak affinity for the H3 domain of syntaxinlA, but the two halves bind cooperatively to the H3 domain. However, the ITC experiment using the full-length synprint site and the H3 domain of syntaxin also displayed the negative result (personal communication: Kelvin Lau). Therefore, it is reasonable to say that there is no strong interaction between the synprint site and the H3 domain of syntaxin. Whether the N-terminal part of sytnaxin 1 A interacted with the synprint site was inconclusive because our initial positive result could not be reproduced. The positive ITC result showed strong binding between both the N-terminal half and the C-terminal half of the synprint site to N-terminal part of syntaxinlA (Figure 25 a and b). The two S-shaped titration curves with 1:1 stoichiometry did not seem to be an artifact. However, the subsequent ITC using different experimental conditions all showed negative results. We put a lot of effort into finding a reasonable scientific explanation for the conflicting results, but to date it still remains inconclusive. Therefore, it is difficult to conclude whether the N-terminal part of syntaxin binds to the synprint site. Given these ITC results, it can be concluded that binding between the synprint site and syntaxinlA is not as strong as reported. However, the correlation between SNARE mediated channel inhibition and the interaction between SNAREs and the synprint site 84 has been proven by several research groups. Therefore, we believe that there is still binding between SNAREs and the synprint site, but it is too weak to be detected. In the nerve terminal, syntaxinlA and SNAP-25 are anchored on the plasma membrane to act as t-SNAREs for the vesicle fusion, and neuronal Ca2. 1 and 2.2 are \u00E2\u0080\u009Crestricted\u00E2\u0080\u009D in the active zones by several adaptor proteins, such as Mint 1 and CASK (Spafford et al 2003). The colocalization and high local concentration of SNAREs and Cas can thus increase the chances for them to interact, thereby allowing modulation of calcium channels through the synprint site. 85 References Agler, Hi., Evans, J., Tay, L.H., Anderson, M.J., Colecraft, H.M. & Yue, D.T. (2005). G protein-gated inhibitory module of N-type (ca(v)2.2) ca2+ channels. Neuron, 46(6), 891-904. Arac, D., Chen, X., Khant, H.A., Ubach, J., Ludtke, S.J., Kikkawa, M., Johnson, A.E., Chiu, W., S\u00C3\u00BCdhof, T.C., & Rizo, J. (2006). Close membrane-membrane proximity induced by Ca(2+)-dependent multivalent binding of synaptotagmin- 1 to phospholipids. Nat Struct Mol Biol,13(3),209-217. Arikkath, J. & Campbell, K.P. (2003). Auxiliary subunits: essential components of the voltage-gated calcium channel complex. Cuff Opin Neurobiol, 13(3), 298-307. Atluri, P.P. & Regehr, W.G. (1998). Delayed release of neurotransmitter from cerebellar granule cells. J Neurosci., 18(20),8214-8227. Artim, D.E. & Meriney, S.D. (2000). G-protein-modulated Ca(2+) current with slowed activation does not alter the kinetics of action potential-evoked Ca(2+) current. J Neurophysiol, 84(5), 2417-2425. Bai, J., Earles, C.A., Lewis, J.L., & Chapman, E.R. (2000). Membrane-embedded synaptotagmin penetrates cis or trans target membranes and clusters via a novel mechanism.J Biol Chem, 275(33), 25427-25435. Bell, T.J., Thaler, C., Castiglioni, A.J., Helton, T.D., & Lipscombe, D. (2004). Cell- specific alternative splicing increases calcium channel current density in the pain pathway. Neuron, 4 1(1), 127-138. Bergfors, T. (2007). Screening and optimization methods for nonautomated crystallization laboratories. Methods Mol Biol, 363,131-151. Bezprozvanny, I,, Scheller, R.H., & Tsien, R.W. (1995). Functional impact of syntaxin on gating of N-type and Q-type calcium channels. Nature, 378(6557), 623-626. Bezprozvanny, I., Zhong, P., Scheller, R.H., & Tsien, R.W. (2000). Molecular determinants of the functional interaction between syntaxin and N-type Ca2+ channel gating. Proc NatI Acad Sd U S A, 97(25), 13943-13948. Bhalla, A., Chicka, M.C., Tucker, W.C., & Chapman, E,R.(2006). Ca(2+)-synaptotagmin directly regulates t-SNARE function during reconstituted membrane fusion. Nat Struct Mol Biol, 13(4), 323-330. 86 Borjesson, S. I. and Elinder, F. (2008). Structure, function. and modification of the voltage sensor in voltage-gated ion channels. Cell Biochem Biophys, 53(3), 149-174. Bowen, M.E., Weninger, K., Brunger, A.T., & Chu, S. (2004). Single molecule observation of liposome-bilayer fusion thermally induced by soluble N-ethyl maleimide sensitive-factor attachment protein receptors (SNAREs). Biophys J, 87(5), 3569-3584. Budde,T., Meuth, S., & Pape, H.C. (2002). Calcium-dependent inactivation of neuronal calcium channels. Nat Rev Neurosci, 3(11), 873-883. Cachelin, A.B., de Peyer, J.E., Kokubun, S. & Reuter, 1-1. (1983). Ca2+ channel modulation by 8-bromocyclic AMP in cultured heart cells. Nature, 304(5925),462- 464. CaritI, C., Page, K. M., Stephens, G. J., Dolphin, A. C. (1999). Identification of residues in the N terminus of aiphaiB critical for inhibition of the voltage-dependent calcium channel by Gbeta gamma. J Neurosci, 19(16):6855-6864. CantI, C., Nieto-Rostro, M., Foucault, 1., Heblich, F., Wratten, J., Richards, M.W., Hendrich, J., Douglas, L., Page, K.M., Davies, A., & Dolphin, A.C. (2005). The metal-ion-dependent adhesion site in the Von Willebrand factor-A domain of alpha2delta subunits is key to trafficking voltage-gated Ca2+ channels. Proc NatI Acad Sci U S A, 102(32), 11230-11235. Chapman, E.R. (2008). How does synaptotagmin trigger neurotransmitter release?.Annu Rev Biochem, 77, 615-4 1. Catterall, W.A. (2000). Structure and regulation of voltage-gated Ca2+ channels. Annu Rev Cell Dev Biol, 16, 521-555. Catterall, W.A. & Few, A.P. (2008). Calcium channel regulation and presynaptic plasticity. Neuron, 59(6), 882-901. Chen, X., Lu, J., Dulubova, 1., & Rizo, J. (2008) NMR analysis of the closed conformation of syntaxin-1. J Biomol NMR, 41(1), 43-54. Chen, Y,A,, Scales, S.J., Patel, S.M., Doung, Y.C., & Scheller, R.H. (1999). SNARE complex formation is triggered by Ca2+ and drives membrane fusion. Cell, 97(2),165-174. Chen, Y.H., Li, M.H., Zhang, Y., He, L.L., Yamada, Y., Fitzmaurice, A., Shen, Y., Zhang, H., Tong, L., & Yang, J. (2004). Structural basis of the alphal-beta subunit interaction of voltage-gated Ca2+ channels. Nature, 429(6992), 675-680. 87 Clapham, D.E. (2007). Calcium signaling. Cell, 131(6),1047-1058. Coppola, T., Magnin-Luthi, S., Perret-Menoud, V., Gattesco, S., Schiavo, G., & Regazzi, R.J. (2001). Direct interaction of the Rab3 effector RIM with Ca2+ channels, SNAP-25, and synaptotagmin. J. Biol Chem, 276(35), 32756-32762. Curtis, B.M. & Catterall, W.A. (1985). Phosphorylation of the calcium antagonist receptor of the voltage-sensitive calcium channel by cAMP-dependent protein kinase. Proc Nati Acad Sci U S A, 82(8),2528-2532. Dai, H., Shen, N., Ara\u00C3\u00A7, D., & Rizo, J. (2007). A quatemary SNARE-synaptotagmin Ca2+-phospholipid complex in neurotransmitter release. J Mol Biol, 367(3),848-863. Derewenda, Z.S. &Vekilov, P,G. (2006). Entropy and surface engineering in protein crystallization. Acta Crystallogr D Biol Crystallogr, 62(Pt l),1 16-124. Dolmetsch, R.E., Pajvani,U., Fife, K., Spotts, J.M. & Greenberg, M.E. (2001). Signaling to the nucleus by an L-type calcium channel-calmodulin complex through the MAP kinase pathway. Science, 294(554 1), 333-339. Dolphin, A.C., Page, K.M., Berrow, N.S., Stephens, G.J. & CantI, C. (1999). Dissection of the calcium channel domains responsible for modulation of neuronal voltage- dependent calcium channels by G proteins. Ann N Y Acad Sci, 868, 160-174. Du, Y.L. & Lou, Y.Q. (1989). [Calcium antagonism of enpiperate on isolated rabbit aorta strips and guinea pig ileum]. Zhongguo Yao Li Xue Bao, 10(2),114-1 17. Dulubova, I., Khvotchev, M., Liu, S., Huryeva, I., SUdhof, T.C., & Rizo, J. (2007). Muncl8-1 binds directly to the neuronal SNARE complex. Proc NatI Acad Sci U S A, 1 04(8):2697-2702. Dunlap, K. (2007). Calcium channels are models of self-control. J Gen Physiol, 129(5), 379-383. Edeihoch, H. (1967).+ Spectroscopic determination of tryptophan and tyrosine in proteins. Biochemistry, 6(7): 1948-54 Evans, R.M., Zamponi, G.W. (2006). Presynaptic Ca2+ channels--integration centers for neuronal signaling pathways.Trends Neurosci, 29(11), 6 17-624. Field, M.J., Cox, P.J., Stott, E., Melrose, H., Offord, J., Su, T.Z., Bramwell, S., Corradini, L., England, S., Winks, J., Kinloch, R.A., Hendrich, J., Dolphin, A.C., Webb, T., & Williams, D. (2006). Identification of the alpha2-delta-1 subunit of voltage dependent calcium channels as a molecular target for pain mediating the analgesic actions of pregabalin. Proc Natl Acad Sci U S A, 103(46),17537-17542. 88 Findeisen, F., & Minor, D.L. Jr. (2009). Disruption of the 1S6-AID linker affects voltage gated calcium channel inactivation and facilitation. J Gen Physiol, 133(3),327-343 Fraser, I.D., Tavalin, S.J., Lester, L.B., Langeberg, L.K., Westphal, A.M., Dean, R.A., Marrion, N.V. & Scott, J.D. (1998). A novel lipid-anchored A-kinase Anchoring Protein facilitates cAMP-responsive membrane events. EMBO J, 17(8),2261-2272. Freise, D., Held, B., Wissenbach, U., Pfeifer, A., Trost, C., Himmerkus, N., Schweig, U., Freichel, M., Biel, M., Hofhiann, F., Roth, M., & Flockerzi, V. (2000). Absence of the gamma subunit of the skeletal muscle dihydropyridine receptor increases L-type Ca2+ currents and alters channel inactivation properties. J Biol Chem, 275(19), 14476-1448. Gether, U., Asmar, F., Meinild, A.K. & Rasmussen, S.G. (2002). Structural basis for activation of G-protein-coupled receptors. Pharmacol Toxicol, 9 1(6), 304-312. Gray, P.C., Scott, J.D. & Catterall, W.A. (1998). Regulation of ion channels by cAMP- dependent protein kinase and A-kinase anchoring proteins. Curr Opin Neurobiol, 8(3), 330-334. Greenfield, N. & Fasman, G.D. (1969).Computed circular dichroism spectra for the evaluation of protein conformation. Biochemistry. 8(10), 4108-4116. Gurnett, C.A., Felix, R., & Campbell, K.P. (1997). Extracellular interaction of the voltage-dependent Ca2+ channel alpha2delta and alphal subunits. J Biol Chem, 272(29), 18508-18512. Hagler, D.J. Jr. & Goda, Y.J. (2001). Properties of synchronous and asynchronous release during pulse train depression in cultured hippocampal neurons. Neurophysiol, 85(6),2324-2334. Hailing, D.B., Aracena-Parks, P., & Hamilton, S.L. (2006). Regulation of voltage-gated Ca2+ channels by calmodulin. Sci STKE, 318, en. He, L.L., Zhang, Y., Chen, Y.H., Yamada, Y., & Yang, J. (2007). Functional modularity of the beta-subunit of voltage-gated Ca2+ channels. Biophys J, 93(3), 834-845. Heidelberger, R., Heinemann, C., Neher, E., & Matthews, G. (1994). Calcium dependence of the rate of exocytosis in a synaptic terminal. Nature, 371(6497), 513- 515. Herlitze, S., Garcia, D.E., Mackie, K., Hille, B., Scheuer, T. & Catterall, W.A. (1996). Modulation of Ca2+ channels by G-protein beta gamma subunits. Nature, 380(657 1), 258-262. 89 Herrick, D.Z., Sterbling, S., Rasch, K.A., Hinderliter, A., & Cafiso, D.S. (2006). Position of synaptotagmin I at the membrane interface: cooperative interactions of tandem C2 domains. Biochemistry, 45(32), 9668-9674. Hobom, M., Dai, S., Marais, E., Lacinova, L., Hofmann, F., & Klugbauer, N. (2000). Neuronal distribution and functional characterization of the calcium channel alpha2delta-2 subunit. Eur J Neurosci, 12(4): 1217-1226. Hoeflich, K.P.& Ikura, M. (2002). Calmodulin in action: diversity in target recognition and activation mechanisms. Cell, 108(6),739-742. Hua, Y. & Scheller, R.H. (2001). Three SNARE complexes cooperate to mediate membrane fusion. Proc NatI Acad Sd U S A, 98(14), 8065-8070. Huang, C.C., Wang, S.). & Gean, P.W. (1998). Selective enhancement of P-type calcium currents by isoproterenol in the rat amygdale. J. Neurosci. 18, 2276\u00E2\u0080\u00942282. Hudmon, A. & Schulman, H. (2002). Structure-function of the multifunctional Ca2+/calmodulin-dependent protein kinase II. Biochem J, 364(Pt 3), 593-611. Hudmon, A., Schulman, H., Kim, J., Maltez, J.M., Tsien, R.W. & Pitt, G.S. (2005). CaMKII tethers to L-type Ca2+ channels, establishing a local and dedicated integrator of Ca2+ signals for facilitation. 3 Cell Biol, 171(3), 537-547. Hui, E., Johnson, C.P., Yao, J., Dunning, F.M., & Chapman, E.R. (2009). Synaptotagmin-mediated bending of the target membrane is a critical step in Ca(2+)-regulated fusion. Cell, 138(4), 709-721. Ikeda, S.R. & Dunlap, K. (1999). Voltage-dependent modulation ofN-type calcium channels: role of G protein subunits. Adv Second Messenger Phosphoprotein Res, 33, 131-151. Jarvis, S.E. & Zamponi, G.W. (2001). Distinct molecular determinants govern syntaxin lA-mediated inactivation and G-protein inhibition of N-type calcium channels. J Neurosci, 2 1(9), 2939-2948. Jarvis, S.E., Barr, W., Feng, Z.P., Hamid, J. & Zamponi, G.W. (2002). Molecular determinants of syntaxin 1 modulation of N-type calcium channels. J Biol Chem, 277(46), 44399-44407. Jay, S.D., Ellis, S.B., McCue, A.F., Williams, M.E., Vedvick, T.S., Harpold, M.M., Campbell, K.P. (1990). Primary structure of the gamma subunit of the DHP sensitive calcium channel from skeletal muscle. Science, 248(4954), 490-492. 90 Jay, S.D., Sharp, A.H., KahI, S.D., Vedvick, T.S., Harpold, M.M., & Campbell, K.P. (1991). Structural characterization of the dihydropyridine-sensitive calcium channel alpha 2-subunit and the associated delta peptides. J Biol Chem, 266(5), 3287-3293. Jiang, X., Lautermilch, N.J., Watari, H., Westenbroek, R.E., Scheuer, T., & Catterall, W.A. (2008). Modulation of CaV2. 1 channels by Ca2+/calmodulin-dependent protein kinase II bound to the C-terminal domain.Proc NatI Acad Sc, 105(1), 341- 346. Joseph, G.D. & John, G.F. (2003). What is the role of SNARE proteins in the membrane fusion. J. Physiol. Cell Physiol, 285 237\u00E2\u0080\u0094249. Kaneko, S., Cooper, C.B., Nishioka, N., Yamasaki, H., Suzuki, A., Jarvis, S.E., Akaike, A., Satoh, M., & Zamponi, G.W.( 2002). Identification and characterization of novel human Ca(v)2.2 (alpha 1B) calcium channel variants lacking the synaptic protein interaction site. J Neurosci, 22(1), 82-92. Kang, M.G., & Campbell, K.P. (2003). Gamma subunit of voltage-activated calcium channels. J Biol Chem, 278(24), 21315-21318. Keith, R.K., Poage, R.E., Yokoyama, C.T., Catterall, W.A., & Meriney, S,D. (2007). Bidirectional modulation of transmitter release by calcium channel/syntaxin interactions in vivo.J Neurosci, 27(2):265-269. Khanna, R., Zougman, A., & Stanley, E.F. (2007). A proteomic screen for presynaptic terminal N-type calcium channel (CaV2.2) binding partners. J Biochem Mol Biol, 40(3), 302-3 14. Kim, E.Y., Rumpf, C.H., Fujiwara, Y., Cooley, E.S., Van Petegem, F., & Minor, D.L. Jr. (2008). Structures of CaV2 Ca2+/CaM-IQ domain complexes reveal binding modes that underlie calcium-dependent inactivation and facilitation. Structure, 16(10), 1455-1467. Kim, D.K. & Catterall, W,A. (1997). Ca2+-dependent and -independent interactions of the isoforms of the aiphalA subunit of brain Ca2+ channels with presynaptic SNARE proteins. Proc NatI Acad Sci U S A, 94(26), 14782-14786. Kisilevsky, A.E. & Zamponi, G.W. (2008). Presynaptic calcium channels: structure, regulators, and blockers. Handb Exp Pharmacol,184, 45-75. Klugbauer, N., Lacinov\u00C3\u00A1, L., Marais, E., Hobom, M., & Hofmann F. (1999). Molecular diversity of the calcium channel alpha2delta subunit. J Neurosci, 19(2), 684-691. 91 Klugbauer,N., Marais, E., & Hofmann, F. (2003). Calcium channel alpha2delta subunits: differential expression, function, and drug binding. J Bioenerg Biomembr, 35(6), 639-647. Kuniyasu, A., Oka, K., Ide-Yamada, T., Hatanaka, Y., Abe, T., Nakayama, H., & Kanaoka, Y. (1992). Structural characterization of the dihydropyridine receptor- linked calcium channel from porcine heart. J Biochem, 112(2), 23 5-242. Lee, T.-S., Karl, R., Moosmang, S., Lenhardt, P., Klugbauer, N., Hofmann, F., et al.(2006). Calmodulin kinase II is involved in voltage-dependent facilitation of the L type Cavi .2 calcium channel: identification of the phosphorylation sites. J Biol Chem, 281, 25560\u00E2\u0080\u009425567. Li, B., Zhong, H., Scheuer, T. & Catterall, W.A. (2004). Functional role of a C-terminal Gbetagamma-binding domain of Ca(v)2.2 channels. Mol Pharmacol, 66(3), 761- 769. Lipscombe, D., Pan, J.Q., & Gray, A.C. (2002). Functional diversity in neuronal voltage gated calcium channels by alternative splicing of Ca(v)alphal. Mo! Neurobiol, 26(1), 2 1-44. Liu, T., Tucker, W.C., Bhalla, A., Chapman, E.R., & Weisshaar, J.C. (2005). SNARE- driven, 25-millisecond vesicle fusion in vitro. Biophys J, 89(4), 2458-2472. Lu, X., Zhang, Y., & Shin, Y. K. (2008). Supramolecular SNARE assembly precedes hemifusion in SNARE-mediated membrane fusion. Nat Struct Mol Biol, 15(7), 700- 706. Margittai, M., Widengren, J., Schweinberger, E., Schr\u00C3\u00B6der, G.F., Felekyan, S., Haustein, E., Konig, M., Fasshauer, D., Grubm\u00C3\u00BCller, H., Jahn, R., & Seidel, C.A. (2003). Single-molecule fluorescence resonance energy transfer reveals a dynamic equilibrium between closed and open conformations of syntaxin 1. Proc Nati Acad Sci,100(26),15516-15521. McGee, A.W., Nunziato, D.A., Maltez, J.M., Prehoda, K.E., Pitt, G.S. & Bredt, D.S.(2004). Calcium channel function regulated by the SH3-GK module in beta subunits. Neuron, 42(1), 89-99. Mintz, I.M., Sabatini, B.L., & Regehr, W.G. (1995). Calcium control of transmitter release at a cerebellar synapse. Neuron, 15(3),675-688. Misura, K.M., Scheller, R.H., & Weis, W.1. (2000). Three-dimensional structure of the neuronal-Seci-syntaxin Ia complex. Nature, 404(6776), 355-362. 92 Mochida, S., Sheng, Z.H., Baker, C., Kobayashi, H., & Catterall, W.A. (1996). Inhibition of neurotransmission by peptides containing the synaptic protein interaction site of N-type Ca2+ channels. Neuron, 17(4),781-788. Mochida, S., Westenbroek, R.E., Yokoyama, C.T., Zhong, H., Myers, S.J., Scheuer, T., Itoh, K., & Catterall, W.A. (2003). Requirement for the synaptic protein interaction site for reconstitution of synaptic transmission by P/Q-type calcium channels. Proc Nat! Acad Sci U S A,100(5), 28 19-2824. Mochida, S., Yokoyama, C.T., Kim, D.K., Itoh, K., & Catterall, W.A (1998). Evidence for a voltage-dependent enhancement of neurotransmitter release mediated via the synaptic protein interaction site of N-type Ca2+ channels. Proc Nati Acad Sd U S A, 95(24), 14523-14528. Montecucco, C., Schiavo, G., & Pantano, S. (2005). SNARE complexes and neuroexocytosis: how many, how close? Trends Biochem Sci, 30(7), 367-372. Monck, J.R. & Fernandez, J.M. (1994). The exocytotic fusion pore and neurotransmitter release. Neuron, 12(4),707-716. Moss, F.J., Viard, P., Davies, A., Bertaso, F., Page, K.M., Graham, A., Canti, C., Plumpton, M., Plumpton, C., Clare, J.J., & Dolphin, A.C. (2002). The novel product of a five-exon stargazin-related gene abolishes Ca(V)2.2 calcium channel expression. EMBO J, 21(7),1514-1523. Ng, J.D., Gavira, J.A., & Garcia-RuIz, J.M. (2003). Protein crystallization by capillary counterdiffusion for applied crystallographic structure determination. J Struct Biol, 142(1), 218-23 1. Nishizuka, Y. (1995). Protein kinase C and lipid signaling for sustained cellular responses. FASEB J, 9, 484-496. Norbert, Klugbauer., Lubica, Lacinov\u00C3\u00A1., Else, Marais., Muriel, Hobom., & Franz, Hofmann. (1999). Molecular Diversity of the Calcium Channel X2 Subunit. The Journal of Neuroscience, 19(2), 684-691. Olivera, B.M., Miljanich, G.P., Ramachandran, J., & Adams, M.E.(1994). Calcium channel diversity and neurotransmitter release: the omega-conotoxins and omega agatoxins. Annu Rev Biochem, 63, 823-867. Pang, Z.P., Shin, O.H., Meyer, A.C., Rosenmund, C., & S\u00C3\u00BCdhof, T.C. (2006). A gain-of function mutation in synaptotagmin- I reveals a critical role of Ca2+-dependent soluble N-ethy!maleimide-sensitive factor attachment protein receptor complex binding in synaptic exocytosis. J Neurosci, 26(48),12556-12565. 93 Perez, D.M. & Kamik, S.S. (2005). Multiple signaling states of G-protein-coupled receptors. Pharmacol Rev, 57(2), 147-161. Qin, N., Yagel, S., Momplaisir, M.L., Codd, E.E., & D\u00E2\u0080\u0099Andrea, M.R. (2002). Molecular cloning and characterization of the human voltage-gated calcium channel alpha(2)delta-4 subunit. Mol Pharmacol, 62(3),485-496. Rajapaksha, W.R., Wang, D., Davies, J.N., Chen, L., Zamponi, G.W., & Fisher, T.E. (2008). Novel splice variants of rat CaV2.1 that lack much of the synaptic protein interaction site are expressed in neuroendocrine cells. J Biol Chem, 283(23), 15997- 6003. Reid, C.A., Bekkers, J.M. & Clements, J.D. (2003). Presynaptic Ca2 channels: a functional patchwork. Trends Neurosci, 26(12), 683\u00E2\u0080\u0094687. Richman, R.W., Strock, J., Hams, M.D., Cabanilla, N.J., Lau, K.K., Siderovski, D.P., Divers\u00C3\u00A9-Pierluissi, M. (2005). RGS 12 interacts with the SNARE-binding region of the Cav2.2 calcium channel. J Biol Chem. ,280(2), 1521-1528. Reese, C., Heise, F., & Mayer, A. (2005). Trans-SNARE pairing can precede a hemifusion intermediate in intracellular membrane fusion. Nature, 436(7049), 410- 414. Rettig, J., Sheng, Z.H., Kim, D.K., Hodson, C.D., Snutch, T.P., & Catterall, W.A. (1996). Isoform-specific interaction of the aiphalA subunits of brain Ca2+ channels with the presynaptic proteins syntaxin and SNAP-25. Proc Natl Acad Sci U S A., 93(1 4),7363 -7368. Rhee, J.S., Li, L.Y., Shin, O.H., Rah, J.C., Rizo, J., SUdhof, T.C., & Rosenmund, C. (2005). Augmenting neurotransmitter release by enhancing the apparent Ca2+ affinity of synaptotagmin 1. Proc NatI Acad Sci U S A., 102(5 1), 18664-18669. Rizo, J., Chen, X., & Arac, D. (2006) Unraveling the mechanisms of synaptotagmin and SNARE function in neurotransmitter release. Trends Cell Biol, 16(7), 339-350. Sabatini, B.L. & Regehr, W.G. (1996). Timing of neurotransmission at fast synapses in the mammalian brain. Nature, 384(6605),170-172. Sather, W.A. & McCleskey, E.W. (2003). Permeation and selectivity in calcium channels. Annu Rev Physiol, 65, 133-159. Sheng, Z.H., Yokoyama, C.T., & Catterall, W.A. (1997). Interaction of the synprint site of N-type Ca2+ channels with the C2B domain of synaptotagmin I. Proc NatI Acad Sci U S A, 94(10), 5405-5410. 94 Sieber, J.J., Willig, K.I., Kutzner, C., Gerding-Reimers, C., Harke, B., Donnert, G.,Rammner, B., Eggeling, C., Hell, S.W., GrubmUller, H., & Lang, T. (2007). Anatomy and dynamics of a supramolecular membrane protein cluster. Science,317(5841),1072-1076. Smyth, D.R., Mrozkiewicz, M.K., McGrath, W.J., Listwan, P., & Kobe, B. (2003).Crystal structures of fusion proteins with large-affinity tags. Protein Sci, 12(7), 1313-1322. S\u00C3\u00B6llner, T., Bennett, M.K., Whiteheart, S.W., Scheller, R.H., & Rothman, J.E. (1993) Aprotein assembly-disassembly pathway in vitro that may correspond to sequential steps of synaptic vesicle docking, activation, and fusion. Cell,75(3), 409-4 18. S\u00C3\u00B8rensen, J.B., Wiederhold, K., MUller, E.M., Milosevic, I., Nagy, G., de Groot, B.L.,GrubmUller, H., & Fasshauer, D. (2006) Sequential N- to C-terminal SNARE complex assembly drives priming and fusion of secretory vesicles. EMBO J, 25(5),955-966. Spafford, J.D., Chen, L., Feng, Z.P., Smit, A.B., & Zamponi, G.W.(2003). Expression and modulation of an invertebrate presynaptic calcium channel alphal subunithomolog. J Biol Chem, 278(23), 21178-21187. Stein, A., Weber, G., Wahl, M.C., & Jahn, R. (2009). Helical extension of the neuronalSNARE complex into the membrane. Nature, 460(7254), 525-528. Sutton, K.G.,McRory, J.E.,Guthrie, H.,Murphy, T.H. & Snutch, T.P. (1999). P/Q-type calcium channels mediate the activity-dependent feedback of syntaxin- IA. Nature, 401(6755), 800-804. Swayne, L.A., Beck, K.E., & Braun, J.E.(2006). The cysteine string protein multimeric complex. Biochem Biophys Res Commun, 348(1), 83-9 1. Swayne, L.A., Chen, L., Hameed, S., Barr, W., Charlesworth, E., Colicos, M.A.,Zamponi, G.W., & Braun, J.E. (2005). Crosstalk between huntingtin and syntaxinIA regulates N-type calcium channels. Mol Cell Neurosci, 30(3), 339-351. Tidor, B. & Karplus, M. (1994). The contribution of vibrational entropy to molecular association. The dimerization of insulin. J Mol Biol, 238(3), 405-4 14. Tomita, S., Chen, L., Kawasaki, Y., Petralia, R.S., Wenthold, R.J., Nicoll, R.A., & Bredt,D.S. (2003). Functional studies and distribution define a family of transmembraneAMPA receptor regulatory proteins. J Cell Biol, 161(4),805-8I6. 95 Zamponi, G.W., Bourinet, E., Nelson, D., Nargeot, 3. & Snutch, T.P. (1997). Crosstalk between G proteins and protein kinase C mediated by the calcium channel aiphal subunit. Nature, 385(6615), 442-446. 97"@en . "Thesis/Dissertation"@en . "2010-05"@en . "10.14288/1.0070992"@en . "eng"@en . "Biochemistry and Molecular Biology"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "Attribution-NonCommercial-NoDerivatives 4.0 International"@en . "http://creativecommons.org/licenses/by-nc-nd/4.0/"@en . "Graduate"@en . "Crystallographic investigation and characterization of the interaction between presynaptic voltage-gated calcium channels and snare proteins"@en . "Text"@en . "http://hdl.handle.net/2429/25586"@en .