University of Southern California Dissertations and Theses

The structure and function of membrane curving proteins on different membrane shapes and their regulation by post-translational modifications

(USC Thesis Other)

The structure and function of membrane curving proteins on different membrane shapes and their regulation by post-translational modifications

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content I THE STRUCTURE AND FUNCTION OF MEMBRANE CURVING PROTEINS ON DIFFERENT MEMBRANE SHAPES AND THEIR REGULATION BY POST- TRANSLATIONAL MODIFICATIONS Thesis by Mark Robert Ambroso A Dissertation Presented to FACULTY OF THE USC GRADUATE SCHOOL AND KECK SCHOOL OF MEDICINE UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Degree of DOCTOR OF PHILOSOPHY (GENETICS, MOLECULAR AND CELLULAR BIOLOGY) August, 2015 II Dedication Thank you to my wife, for supporting me in my decisions. Thank you to my parents, for making it all possible. And thank you to whatever force created this silly and wonderful world for us to study. III Acknowledgements I would like to thank the NIH (Grant GM063915) and the National Multiple Sclerosis Society for their financial support. For his direction, leadership, and discussions I thank Dr. Ralf Langen. I will sorely miss working in such a positive research laboratory. I would like to acknowledge Dr. J. Mario Isas for providing the technical expertise, mentorship, and friendship that made this work possible and my Ph.D. an experience I will never forget. I will miss our discussions on life, spirituality, and science. I would also like to acknowledge Dr. Jobin Varkey, Dr. Balachandra Hegde, Dr. Nitin Pandey, Dr. Charles Bugg, Prabhavati Hegde, Natalie Kegulian, Alan Okada, and Kazuki Teranishi for their contributions to my work and for creating an incredibly productive and enjoyable laboratory environment. I would like to express my gratitude to collaborators for their kindness and dedication, including Dr. Jonah Chan, Dr. Jeannie Chen, Dr. Martin Kast, Dr. Songi Han, and Dr. Doug Rees. For their time and energy spent engaging in discussions and developing concepts I would like to acknowledge my committee members Dr. Ansgar Siemar, Dr. Curtis Okamoto, Dr. Tobias Ulmer, and Dr. Ian Haworth. I would also like to thank USC for promoting excellence in their sciences through a positive and collaborative environment. IV Table of Contents Dedication II Acknowledgements III List of Figures and Tables IX Abstract XII Chapter 1: Introduction 1 1.1 Membrane Curvature 1 1.1.1 Membrane Curvature in the Cell 1 1.1.2 Membrane Curvature Nomenclature 3 1.1.3 Membrane Curving Proteins and Domains 6 1.2 SDSL and EPR as the Method of Choice 11 1.3 SDSL and EPR Methodology 15 1.3.1 Mobility 15 1.3.2 Accessibility 16 1.3.3 Distances 19 1.4 Using SDSL and EPR Methodology to Determine how BAR Proteins Curve Membranes 20 1.4.1 Endophilin A1 on Tubes and Vesicles 23 1.4.2 Amphiphysin on Tubes and Vesicles 25 1.4.3 Regulation of Curvature through Post-Translational Modifications 26 1.5 Takeaway 27 Chapter 1: References 28 Chapter 2: Tubulation by Amphiphysin Requires Concentration-Dependent Switching from Wedging to Scaffolding 36 Abstract 37 2.1 Introduction 38 2.2 Results 41 2.2.1 Amphiphysin Vesicle Binding is Mediated by Shallow Insertion of its Amphipathic N-terminal Helix 41 2.2.2 On Tubes, the BAR Domain Comes into Contact with the Membrane and N Termini Move Deeper into the Bilayer 50 V 2.2.3 Tubulation and BAR Domain Scaffolding are Dependent on Protein Concentration 54 2.3 Discussion 59 2.4 Experimental Procedures 67 2.4.1 Generation of Amphiphysin Mutants and Spin Labeled Derivatives 67 2.4.2 Vesicle Preparation and Amphiphysin Membrane Interaction 67 2.4.3 Acquisition and Analysis of EPR Data 68 2.4.4 Electron Microscopy 69 References 70 Chapter 3: Endophilin A1 Induces Different Membrane Shapes Using a Conformational Switch that is Regulated by Phosphorylation 76 Abstract 77 3.1 Introduction 78 3.2 Results 81 3.2.1 On Tubes, N-Terminal α-helices Insert Deeply into the Acyl Chain Region of the Lipid Membrane 81 3.2.2. On Tubes, Insert Region Inserts Deeply into Acyl Chain Region and BAR Domain Contacts the Lipid Headgroups 86 3.2.3. Phosphomimetic Mutation S75D Destabilizes Tubes by Preventing Deep Insertion of the Insert Helix. 92 3.3 Discussion 94 3.4 Experimental Procedures 100 3.4.1 Rat Endophilin A1 Protein Purification and Spin Labeling 100 3.4.2 Lipid and Tube Sample Preparation 100 3.4.3 Titration Assay for Detecting Membrane Binding 101 3.4.4 Electron Paramagnetic Resonance Continuous Wave and Power Saturation Measurements 102 3.4.5 Pulse EPR and Distance Analysis 102 VI References 104 Chapter 4: Lysine Acetylation is a Mechanism for Regulating the Interaction between Proteins and Membranes 109 Abstract 110 4.1 Introduction 111 4.2 Results 113 4.2.1 Lysine Acetylation is a Prevalent Modification in Membrane Binding Regions of Known Membrane-Interacting Domains 4.2.1.1 Strategy for Determining the Propensity of Acetylation in Membrane Binding Regions 113 4.2.1.2 BAR Proteins have Increased Levels of Lysine Acetylation in their Membrane Binding Regions 116 4.2.1.3 PX Domains have Increased Levels of Acetylation in their Membrane Binding Regions 120 4.2.1.4 PH Domains do not have Increased Levels of Acetylation in their Membrane Binding Regions 120 4.2.1.5 C2 Domains have Increased Levels of Acetylation in their Membrane Binding Regions 121 4.2.1.6 EHD Family Members have Increased Levels of Acetylation in their Membrane Binding Regions 126 4.2.2 Acetylation-mimicking Mutations in Amphiphysin, EHD2 and Synaptotagmin I Reduces their Ability to Shape Membranes by Reducing their Membrane Affinity 128 4.2.3 Acetylation-mimicking Mutations of Amphiphysin and EHD2 alter their Localization in Cells 133 4.3 Discussion 137 4.4 Experimental Procedures 140 4.4.1 Bioinformatic Analysis of Protein Acetylation 140 4.4.2 Generation of Protein Constructs and Mutants 140 VII 4.4.3 Cell Culture, Transfection and Confocal Microscopy 141 4.4.4 Vesicle Preparation 142 4.4.5 Acquisition and Analysis of EPR data 143 4.4.6 Electron Microscopy 144 References 146 Chapter 5: Myelin Basic Protein is a Membrane Curving Protein Inhibited by Citrullination 152 Abstract 153 5.1 Introduction 154 5.2 Results 157 5.2.1 MBP is a Potent Inducer of Membrane Curvature 157 5.2.2 MBP-coated Tubules Exhibit Vesicle-Wrapping Properties and Generate Multi-layered Networks 160 5.2.3 MBP Generates Membrane Curvature in a Similar Fashion to Known Membrane Curving Proteins 163 5.2.4 Electrostatic Interactions Regulate the Level of Membrane Curvature MBP 165 Induces 5.2.5 Disease State MBP Generates a Reduced Level of Membrane Curvature 167 5.3 Discussion 169 5.4 Methods 171 5.4.1 MBP Expression and Purification 171 5.4.2 Vesicle Preparation 171 5.4.3 Electron Microscopy 171 5.4.4 Dye Leakage Assay 172 References 173 Appendix I: Hydration Dynamics as an Intrinsic Ruler for Refining Protein Structure at Lipid Membrane Interfaces 178 VIII Abstract 179 Significance 180 Introduction 181 Results and Discussion 185 Approach to Quantify Local Hydration Dynamics at Biomolecular Interfaces 185 Probing Hydration Dynamics as a Function of Distance Along the Bilayer Normal 189 Topology and Immersion Depth of α-synuclein on a Lipid Membrane Surface 194 Conclusion 202 Materials and Methods 204 Preparation of Spin-Labeled α-synuclein 204 Preparation of Spin-Labeled Annexin B12 205 ODNP Experiments 205 References 207 Concluding Remarks and Discussion 212 References 217 IX List of Figures and Tables Figure 1.1 Membrane Curvature in the Cell. 2 Figure 1.2 Geometric Considerations of Membrane Curvature. 4 Figure 1.3 Isotropic and Anisotropic Membrane Curvature 5 Table 1.1 Peripheral Membrane Binding Domains 7 Figure 1.4 Protein Scaffolding and Wedging 9 Figure 1.5 Site-Directed Spin Labeling with MTSL 14 Figure 1.6 EPR Spectra Mobility and Principles of Membrane Accessibility 18 Figure 1.7 BAR Protein Family Structures and Diversity 21 Figure 1.8 Curvature-Specific Involvement of BAR Proteins in Endocytosis 22 Figure 2.1 Amphiphysin Binds Vesicles with its N terminus and not its BAR Domain 45 Figure S2.1 EPR Continuous Wave Spectra, as well as NiEDDA and O 2 Accessibilities, of Spin Labeled Derivatives of Amphiphysin Bound to 100 nm Vesicles 48 Figure 2.2 The N Terminus Submerges into the Acyl Chain Region on Tubes 52 Figure S2.2 EPR Continuous Wave Spectra, as well as NiEDDA and O 2 Accessibilities, of Spin Labeled Derivatives in the N Terminus of Amphiphysin Bound to Lipid Tubes 53 Figure 2.3 The BAR Domain Adheres its Concave Surface to the Membrane on Tubes 56 Figure S2.3 Spin Label Mobility and Accessibility for Spin Labeled Amphiphysin Bound to Tubes or to Vesicles 57 Figure 2.4 Model of how Amphiphysin Stabilizes Different Types of Membrane Curvature 66 Figure 3.1 Endophilin A1-Induced Tubulation 83 Figure 3.2 N-terminal Helices Penetrate Deeply into the Membrane 84 Figure S3.1 Depth Calibration Using Spin Labeled Lipids 86 Figure 3.3 Concerted Movement of BAR Domain and Insert Region Toward the Membrane 88 Figure S3.2 CW EPR and DEER Measurements of Endophilin A1’s Insert Region 89 X Figure S3.3 O 2 and NiEDDA Accessibilities for Insert Region Residues 91 Figure 3.4 Phosphomimetic S75D Mutation Destabilizes Tubes by Reducing Membrane Immersion Depth of the Insert Region 93 Figure 3.5 Schematic Illustration of Endophilin A1 Tube and Vesicle Binding and its Modulation by Phosphorylation 98 Figure 4.1 BAR and PX Domains are Commonly Acetylated, but not PH Domains 118 Figure 4.2 Sequence Alignment of PX Domains from Sorting Nexins to p40 phox 122 Figure 4.3 Lysine Acetylation is Localized to Membrane Binding Regions of C2 Domains 124 Figure 4.4 The Eps15-homology EHD1-4 Proteins have Localized Lysine Acetylation in their Membrane Binding Regions 127 Figure 4.5 Acetylation-Mimicking Mutations Inhibit Membrane Remodeling and Reduce Membrane Affinity 131 Figure 4.6 Acetylation-Mimicking Mutations Alter how Amphiphysin and EHD2 Interact with Cellular Membranes 135 Figure 5.1 MBP Alone can Remodel Large Myelin-Like Vesicles into Different 158 Structures Figure 5.2 Lipid Tubes formed by MBP Exhibit the Ability to Wrap Vesicles 161 Figure 5.3 MBP can Generate Multi-Layer Wraps of Membrane Tubes 162 Figure 5.4 MBP Disrupts the Integrity of Model Membranes 164 Figure S5.1 Vesiculation of 100 nm Diameter Vesicles 164 Figure 5.5 Correlations between Membrane Leakage and MBP’s Net Charge 166 Figure 5.6 Citrullination-Mimicking Mutations Reduce the Ability of MBP to Induce Curvature on Model Membranes 168 Appendix Figure A1 188 Appendix Figure A2 191 Appendix Table S1 and S2 192 Appendix Figure S1 193 Appendix Figure S2 193 Appendix Figure A3 197 Appendix Figure A4 198 XI Appendix Table S3 199 Appendix Table S4 200 Appendix Figure S3 201 Appendix Figure S4 201 Appendix Figure S5 202 XII ABSTRACT Membrane curvature is an essential biophysical property utilized within cells to execute their most basic functions. Recent advances in this field have shown proteins capable of shaping membranes to be the major driving force behind these remodeling events. These proteins are highly regulated by cells and are recruited to remodeling events in a location, time, and function specific manner. The Bin/Amphiphysin/Rvs (BAR) superfamily of proteins are banana-shaped homodimers which are involved in remodeling membranes during endocytosis. BAR proteins endophilin A1 and amphiphysin have been shown to recruit important endocytosis-specific co- factors to the membrane as well as directly act in membrane remodeling during this process. In vitro, these proteins are observed to bind large vesicles and convert them into highly curved structures such as small vesicles or cylindrical tubes. While previous studies have shown these proteins to be able to curve membranes, the underlying mechanisms of this process are not fully understood. Furthermore, how cells regulate the shape or curvature a protein induces is unclear. The central thesis of this work is using biophysical tools to elucidate the structures and mechanisms protein use to bend membranes as well as to evaluate how these mechanisms may be regulated in the cell. Our initial studies evaluated the shape-dependent structures that membrane curving proteins endophilin A1 and amphiphysin have on lipid vesicles or tubes using site-directed spin labeling and electron paramagnetic resonance. These studies revealed that these proteins use different mechanisms and structures to generate distinct membrane shapes. We additionally investigated the affect of known post-translation modifications of these proteins. Phosphorylation of endophilin A1 was found to regulate the membrane shape this protein generates. Using a proteomics approach, lysine acetylation, a modification previously not considered to be important for protein-membrane interactions, is observed to be localized in XIII membrane-binding regions of peripheral membrane proteins. Biophysical characterization of acetylation-mimicking mutations in proteins amphiphysin, EHD2, and synaptotagmin show that this modification is capable of regulating the type of curvature these proteins generate in cells and in vitro by reducing their overall affinity for the membrane. While the field is rapidly growing, new membrane curving proteins and processes are still being discovered. Myelin, defined as concentric membrane wraps enveloping neuronal axons, are an essential component of the mammalian nervous system. How these membranes wrap and remain tightly curved around an axon is not completely understood. Loosely compacted myelin is a hallmark of the neurodegenerative disease Multiple Sclerosis. Myelin is mostly composed of lipids and two proteins, myelin basic protein and the proteolipid protein. Myelin basic protein is known as the executive protein in myelination and knockout results in loosely compacted and defective myelin wraps. Using a biophysical characterization of myelin basic protein, we find this protein to be able to induce curvature similar to that of other known membrane curving proteins. Myelin basic protein-induced tubes exhibit intermembrane adhesion and the ability to wrap neighboring vesicles, or form extensive multi-layered tube networks. Finally, we find that the Multiple-sclerosis related aberrant post-translational modification of myelin basic protein results in a decreased ability of this protein to generate curvature. We therefore conclude that this protein is a membrane curving protein and that this function may be inhibited in disease. 1 CHAPTER 1 INTRODUCTION 1.1 Membrane Curvature 1.1.1 Membrane Curvature in the Cell Membranes act as partitions in order to establish cellular compartments and the ability to regulate and remodel their shape is a vital aptitude in cellular functioning. Highly coordinated membrane remodeling events are observed in essential processes including endo- and exocytosis, intracellular membrane trafficking, organelle shape, viral budding, phagocytosis, autophagy, and cell movement (Doherty and McMahon, 2009; McMahon and Boucrot, 2015). In certain cases such as synaptic endocytosis, cells must regulate the uptake of synaptic vesicles carrying neurotransmitters often on millisecond timescales (Watanabe et al., 2013). This process involves the careful generation pre-endocytotic invaginations, sphere-like intermediates connected to the plasma membrane by a cylindrical neck, and scission of the cylindrical necks in order to release vesicles into the cytosol. The diversity of structures needed during this processes indicates that cells are capable of orchestrating specific membrane shapes with precise regulation and direction (Figure 1.1). In fact, the disruption of these faculties results in the generation of neurological diseases and muscular myopathies (Bergmann et al., 2003; De Camilli et al., 1993; Matta et al., 2012; Nicot et al., 2007). The field of membrane curvature is relatively young and the precise mechanisms that underlie how cells regulate this process are still not fully understood. Determining the structure of membrane curving proteins bound to different membrane shapes would elucidate how proteins can generate different levels of curvature. While difficult for X-ray crystallography and NMR, site-directed spin labeling (SDSL) and electron paramagnetic resonance (EPR) is an extremely powerful technique for studying the structures of proteins bound to different membrane shapes. 2 Figure 1.1 Membrane Curvature in the Cell. Cellular membranes are highly curved and undergo remodeling. Examples of areas of the cell which undergo high degrees of membrane curvature are highlighted in red. Figure is adapted from McMahon and Gallop, 2005. 3 1.1.2 Membrane Curvature Nomenclature Membrane curvature can be described using basic geometric considerations (Lipowsky, 2014). Membrane bending is relative to an observer looking along the plane of a lipid bilayer (Figure 1.2a). When the membrane bends below the plane of the monolayer or bilayer it is described as negative curvature while bending above the plane as positive curvature. A significant contributor to the curvature state a bilayer spontaneously assumes often results from its lipid composition (Figure 1.2b). Lipid bilayers are a composite of many different lipids with distinct biochemical properties, including different acyl-chain lengths, saturation levels, and headgroups. The summary of these properties defines the spontaneous curvature of that lipid and the resulting combination of these individual lipids lends a specific amount of effective spontaneous curvature to the overall bilayer (Lipowsky, 2014; Yandrapalli et al., 2014). As a majority of cellular membranes are asymmetric in lipid content (Epand, 2015) a net spontaneous curvature is generated across the bilayer. The resulting shape derived from the spontaneous curvature of the composite of lipids depends on the directionality of the curvature produced (Bobrovska et al., 2013). Isotropic distributions of curvature would result in spherical shapes, where the curvature is distributed in all 360°. Anisotropic curvature, or curvature in a single direction, results in the generation of cylindrical shapes where the direction of curvature is perpendicular to the axis of no curvature (Figure 1.3). Due to the apparent complexity of generating specific types of membrane curvature it is clear that the cell must be capable of selecting, regulating, and coordinating different mechanisms. 4 Figure 1.2 Geometric Considerations of Membrane Curvature. a) Curvature can be described relative to the compartment that it extends towards or away from. Positive curvature is typically described as the bending of the membrane away from the inside compartment and negative curvature towards. In this figure, blue scaffolds represent how proteins can facilitate this curvature. b) The intrinsic shape of individual lipids can influence the spontaneous curvature of a bilayer. Asymmetric lipid compositions can lead to positive (middle) or negative (bottom) curvature, while symmetric compositions lend zero spontaneous curvature (top). Green and red lipids headgroups represent the outer and inner leaflets respectively. Figures are adapted from Kirchhausen et al., 2012 (a) and Yandrapalli et al., 2014 (b). 5 Figure 1.3 Isotropic and Anisotropic Membrane Curvature. The directionality of curvature has a major impact on the shape it generates. A shape is generated based on a direction of curvature (red arrows) and/or a direction of no curvature (yellow arrows). a) Spheres are generated by isotropic curvatures curvature is distributed in all directions. b) Cylindrical tubes are generated when curvature is in a single direction perpendicular to a direction of no curvature. 6 1.1.3 Membrane Curving Proteins and Domains While lipids play a significant role in the shape of cellular membranes, it has become increasingly clear that protein-membrane interactions are the primary driving force behind a majority of complex and highly-coordinated membrane events (Doherty and McMahon, 2009; Drin and Antonny, 2010; Farsad and Camilli, 2003; Itoh et al., 2005; Kirchhausen, 2012; McMahon and Boucrot, 2015; McMahon and Gallop, 2005; Qualmann et al., 2011; Rao and Haucke, 2011; Richard et al., 2011; Soda et al., 2012; Yu and Schulten, 2013). Entire protein families of membrane-curving proteins have been discovered and observed to remodel membranes in vitro (Ambroso et al., 2014; Cui et al., 2009; Farsad et al., 2001; Gallop et al., 2006; Lai et al., 2012; Mim et al., 2012; Mizuno et al., 2010; Peter et al., 2004; Shah et al., 2014; Sorre et al., 2012; Varkey et al., 2010; Westphal and Chandra, 2013; Zhu et al., 2012) and/or in vivo (Ferguson et al., 2009; Meinecke et al., 2013; Milosevic et al., 2011; Sundborger et al., 2011), including the BIN/Amphiphysin/Rvs (BAR) synaptotagmin, synuclein, and Eps15- homology families. Highly promiscuous protein domains capable of binding and potentially curving membranes have been characterized and observed in proteins spanning the proteome, such as BAR, C1, C2, pleckstrin-homology (PH), PX proteins, ENTH/epsin, ANTH, FYVE, and Tubby (Table 1.1) (Stahelin et al., 2014). Most of these proteins have been shown or are proposed to be capable of generating multiple membrane shapes in vitro such as small vesicles or lipid tubes. Importantly, knockout of membrane binding proteins or mutation of membrane curving domains has been shown to inhibit essential life processes (Bai et al., 2010; Bergmann et al., 2003; Matta et al., 2012; Nicot et al., 2007). 7 Table 1.1 Peripheral Membrane Binding Domains. Peripheral membrane-binding protein domains are listed respective to their crystal structures. Most of these domains have been implicated in binding specificity to specific phosphoinositols as described. References to studies which have characterized the lipid binding areas of these domains are provided on the right. Figure is adapted from Stahelin et al., 2014. 8 There are several mechanisms that have been proposed for how proteins bend membranes, including membrane crowding (Stachowiak et al., 2012), bilayer couple (Farsad and Camilli, 2003), amphipathic insertions (wedging) (Drin and Antonny, 2010), and scaffolding (Qualmann et al., 2011). Of these mechanisms amphipathic insertions and scaffolding are by far the most thoroughly studied and are a focus of our work (Figure 1.4) (Mim and Unger, 2012). Amphipathic helices are defined as helical segments whose revolving amino acid sequence aligns hydrophilic residues on one side of the helical axis and hydrophobic residues on the opposing side. The polar nature across the axis of the helical segment allows it to associate one side with the charge lipid headgroups and the other into the hydrophobic acyl chains. As the amphipathic helix inserts into a leaflet it begins to take up space and wedge apart nearby lipids at a magnitude dependent on the depth it inserts (Campelo et al., 2008). Shallow depths asymmetrically wedge apart the lipid headgroups and not the acyl chains of a single leaflet, generating a maximal wedging force. Helices embedded in the middle of the membrane symmetrically wedge apart acyl chains from both leaflets, generating minimal spontaneous curvature. While a single amphipathic insertion may not generate enough curvature to generate shapes on a large membrane, increasing densities of protein insertions are capable of remodeling flat membranes into cylindrical tubes or small spherical vesicles as evident by previous work from our group on synucleins (Varkey et al., 2010) and a subsequent study by an independent group (Westphal and Chandra, 2013). 9 Figure 1.4 Protein Scaffolding and Wedging. Proteins can bend membranes by forcing lipids to conform to their intrinsic shape through a scaffolding mechanism (top). Proteins can also insert amphipathic domains into the membrane thereby generating a splitting force through a wedging mechanism (bottom). Figure adapted from Mim and Unger, 2012. 10 As an alternative to shaping membranes through insertions, proteins can also bind membranes as a scaffold and force them to conform to their intrinsic shape (Figure 1.4). This interaction is typically mediated through electrostatic interactions between positively charged residues lining the membrane-interacting face of the scaffolding protein and negatively charged lipid headgroups. N-BAR proteins are banana-shaped scaffolds which bind the membrane through electrostatic interactions between positive residues on their concave surface and anionic membranes, thereby generating a positive curving force (Qualmann et al., 2011). Alternatively, charged residues line the convex surface of I-BAR proteins and generate negative curvature (Saarikangas et al., 2009). Similar to amphipathic helices, scaffolds have to be recruited in certain protein densities in order to generate a significant amount of membrane curvature (Sorre et al., 2012; Zhu et al., 2012). How both amphipathic insertions and scaffolding are utilized by proteins and cells to regulate the shape of membranes is a primary focus of our studies. While some proteins utilize a single mechanism to bend membranes, others have the ability to use a combination of mechanisms. α-synuclein folds into an extended amphipathic helix on large vesicles (Jao et al., 2008; Varkey et al., 2010) and are believed to work solely through a wedging mechanism. N-BAR proteins endophilin A1 and amphiphysin have been found to utilize both amphipathic insertions as well as a scaffolding domain to generate membrane curvature (Ambroso et al., 2014; Bai et al., 2010; Boucrot et al., 2012; Gallop et al., 2006; Jao et al., 2010; Jung et al., 2010; Peter et al., 2004). The interplay between these two mechanisms has been proposed to determine the type of curvature BAR proteins generate (Boucrot et al., 2012). Furthermore, recent studies from our lab have shown that the membrane bound structures of N-BAR proteins amphiphysin and endophilin depends on the shape it generates (Ambroso et al., 2014; Isas et al., 2015). Therefore, to understand how proteins 11 generate different membrane shapes it is imperative to be able to study their three-dimensional structure on distinct membrane shapes. Many different membrane binding or curving proteins have been discovered but many more are still being uncovered. Myelin basic protein (MBP) is an intrinsically disordered protein known as the executive protein in axon myelination. MBP knockout results in loosely compact myelin. Previous work has shown that MBP potently binds membranes and is proposed to act as a membrane fusing agent. However, MBP shares many characteristics with known membrane curving proteins such as α-synuclein. A component of this thesis is using biophysical tools to define and characterize membrane curving proteins in a non-structural based approach, such as what has been performed for α-synuclein (Varkey et al., 2010; Westphal and Chandra, 2013). 1.2 SDSL and EPR as the Method of Choice While cell biology-based approaches have identified a number of protein families responsible for regulating the geometry of membranes, the structural mechanisms underlying how a protein can generate membrane curvature are still not fully understood. Despite representing 20-30% of the all proteins in most organisms and 40% of drug targets, only a small number of structures of membrane bound proteins have been elucidated by X-ray crystallography or NMR (Carpenter et al., 2008). Crystallizing proteins in the presence of lipids or bound to specific membrane shapes is extremely difficult. Analysis by liquid-state NMR requires the use of detergents or membrane-mimicking systems such as nanolipoprotein particles or bicelles instead of larger and more physiological membrane shapes (Raschle et al., 2010). Solid-state NMR has fewer limitations in using large membranes but suffers from decreasing resolution and increasing experimental cost as the size of a complex increases (Opella, 2013; Tang et al., 2013). 12 SDSL in combination with EPR is an amenable technique for studying the three-dimensional structure of proteins bound to large membranes of specific shapes. These data can be used as structural constraints for building 3-D models of membrane bound proteins. The basis of SDSL and EPR relies on the covalent modification of an amino acid side- chain with a small paramagnetic spin label containing a single unpaired electron. Spin labels are commercially available and although there are many different kinds the most commonly utilized versions are thiol-reactive conjugates capable of being linked to proteins via cysteine residues. The most well characterized protein spin label, [1-oxy-2,2,5,5-tetramethyl-pyrroline-3-methyl]- methane-thiosulfonate, covalently attaches to cysteine side chains and forms the side chain R1 (Figure 1.5) (Oh et al., 2000). There also exists an array of synthesized lipids containing spin labels at specific locations along the acyl chains, the lipid phosphates, and the lipid headgroups. The ability to gather information from both the protein as well as the lipid environment is a major component of why SDSL and EPR can be such a powerful tool for determining membrane protein structure. Special consideration must be taken in deciding whether a target protein is amendable to study by SDSL and EPR. The strategy of this technique relies on being able to direct the spin label to a specific side chain of a protein through conjugation with cysteine(s). Therefore, it is most practical to either recombinantly express or synthetically generate the protein of interest. To generate constructs containing a single cysteine it is often required to mutate native cysteines to other residues, commonly to an alanine. Proteins with many cysteines or whose structure or function relies on the presence of native cysteines are not ideal targets for this method of study. Using target-protein specific functional assays and biophysical methods, one should also verify that cysteine mutagenesis does not significantly alter its endogenous structure and/or function. 13 For proteins judged to be amenable to study by EPR, site-directed mutagenesis using standard PCR and QuikChange (Stratagene) techniques is used to mutate native cysteines to alanines and residues of interest to cysteines, one at a time. All plasmids are confirmed to contain the desired mutation by DNA sequencing. A plasmid containing the cDNA of the resulting construct as well as genes encoding antibiotic resistance are transformed into E. coli, grown in LB media in the presence of the appropriate antibiotic, and expressed by adding a metabolite that induces transcription of the plasmid. By systematically mutating residues along a protein‘s sequence, one at a time to a cysteine, spin labeling these residues, and making EPR measurements, large-scale 3-D protein structures and models can be determined. This process takes into account a variety of information that can be determined from EPR spectra of spin labeled proteins, including the mobility of spin labeled residues in solution and when bound to membranes. Differences between the two aforementioned spectra can elucidate which residues come into contact with the membrane and thereby changing their mobility (Hubbell et al., 1998). These membrane bound residues can then be investigated for their depth in the membrane by measuring their accessibility to reagents that partition in the membrane (Altenbach et al., 1994). Finally, the distance between intra- and intermolecular spin labels can be measured and have been used to determine how well crystallographic data applies to proteins in solution (Jeschke, 2002). By computationally refining data gathered from the mobility, membrane accessibility, and distance information from many different spin labeled sites, the 3-D structure of proteins bound to membranes with distinct membrane shapes can be elucidated. 14 Figure 1.5 Site-Directed Spin Labeling with MTSL. MTSL (I) is the spin labeled analog most commonly used in site-directed spin labeling of proteins. It is covalently attached to cysteine residues through a disulfide bond and contains an unpaired electron on the nitroxide. The covalently attached label is typically referred to as R1. Figure adapted from Oh et al., 2000. 15 1.3. SDSL and EPR Methodology EPR measurements of spin labeled protein or lipid are capable of providing information about the local environment, accessibility to polarity-sensitive reagents, or even their proximity from neighboring spin labels. 1.3.1 Mobility The local structure of R1 is encoded in the line shape of its continuous wave X-band EPR spectrum, which is recorded as a first derivative of the absorption spectrum to improve the signal-to-noise ratio. The nitroxide moiety has three different spin states (different magnetic moments) and undergoes hyperfine splitting, resulting in a three-line EPR spectrum. R1 is highly sensitive to its environment and is affected by backbone dynamics and tertiary interactions. A highly mobile R1 will give rise to lines qualitatively narrow in width whereas low mobility causes line broadening (Figure 1.6a) (Margittai and Langen, 2008). The resulting spectra can provide information on the local environment, secondary structure, or even tertiary and quaternary packing of R1. Quantification of the line broadening is typically done through measuring the width of the spectrum using the second moment (<H 2 >) as well as the inverse of the central line width (ΔH 0 -1 ). In fact, these two mobility parameters are capable of distinguishing whether an R1 is in buried, helix surface, loop, or membrane contacting positions (Apostolidou et al., 2008; Gallop et al., 2006; Isas et al., 2002, 2003; Jao et al., 2004; Langen et al., 2000; Mchaourab et al., 1996; Shah et al., 2014). Systematic studies combining X-ray crystallography (Langen et al., 2000), mutagenesis (Columbus et al., 2001), and spectral simulations (Mchaourab et al., 1996) have further correlated local structure to R1 mobility in α- helical proteins. Comparison of EPR spectrum before and after the addition of lipid membranes 16 can additionally elucidate membrane binding residues as well as changes in protein structure. This principle has been utilized to determine which residues on the concave surface of the BAR domain contact the membrane in the case of amphiphysin and endophilin (Ambroso et al., 2014; Isas et al., 2015), global structural changes in annexins (Isas et al., 2002; Langen et al., 1998), and the structuring of amphipathic helices in the membrane as in the case of α-synuclein (Jao et al., 2004) and the N-terminal helices of N-BAR proteins (Ambroso et al., 2014; Gallop et al., 2006; Isas et al., 2015). 1.3.2 Accessibility While the exact meaning of a change in EPR spectrum upon addition of membranes cannot be determined by mobility alone, in combination with accessibility measurements detailed structural information can be obtained. These measurements involve monitoring the relaxation properties of R1 in the presence of paramagnetic colliders as a function of microwave power. As described by the Maxwell-Boltzmann distribution, the excitation of electrons from low to high energy levels by increasing microwave power eventually reaches saturation. Paramagnetic colliders increase the rate by which high energy state electrons relax into the lower state, thereby increasing the power at which saturation occurs. Therefore, the increase in microwave power required to reach saturation is directly correlated to the accessibility of R1 to paramagnetic colliders. By comparing the accessibility of R1 to a paramagnetic collider which prefers the aqueous solution, O 2 , versus that of one which prefers the hydrophobic lipid bilayer, Ni(II) ethylenediaminediacetate (NiEDDA), a relative membrane accessibility can be determined (Figure 1.6b) (Altenbach et al., 1994). By repeating these measurements for many different sites 17 in the same protein, the 2-D profile of the membrane bound domain can be modeled. The relative accessibility can also be extrapolated to depth using lipids containing spin labels along their acyl chains and headgroups. A small amount of these lipids are added to membrane vesicles and their corresponding accessibility measurements are measured while bound to the protein-membrane complex being studied (Frazier et al., 2002, 2003). The lipid system can be calibrated from the lipid headgroups to the bottom of the acyl chains, around a depth range of -5 to 20 Å below the phosphates. This information has been used to determine how N-BAR proteins endophilin and amphiphysin insert their amphipathic N-terminal helices at different depths depending on the membrane shape they make. Recent advances have been extended the resolution above the lipid phosphates to approximately 30 Å using an NMR/EPR technique which can quantify the translational diffusion of hydration water surrounding a membrane (Cheng et al., 2013). It is therefore possible to obtain structural information for protein domains embedded in the bilayer as well as domains which are relatively remote from the membrane. Additionally, R1 accessibility can distinguish whether changes in mobility are due to direct or indirect contact with the membrane. Secondary structural information can be elucidated by systematically scanning a protein sequence and determining a periodicity. Membrane-bound amphipathic helices have been shown to have oscillating accessibilities to O 2 resembling the canonical periodicity of an α-helix (~3.6 amino acids per turn) (Ambroso et al., 2014; Apostolidou et al., 2008; Jao et al., 2008, 2010; Lai et al., 2012). Alternatively, a membrane-penetrating N-terminal domain of EHD2 lacked this periodicity and was determined to bind the membrane in a non- helical ‗loop‘ (Shah et al., 2014). 18 Figure 1.6 EPR Spectra Mobility and Principles of Membrane Accessibility. a) The EPR spectra of a fast moving spin label displays narrow line widths. Spin labels in structured regions of a protein or in regions making contact with other proteins or membranes results in EPR spectra with broadened lines. b) An illustration of the preferential partitioning of paramagnetic colliders O 2 and NiEDDA. By measuring the relative accessibility of a spin label to these two reagents a relative membrane depth can be inferred. Figure was adapted from Margittai and Langen, 2008. 19 1.3.3 Distances Line-broadening affects can also be due to dipolar interactions between spin labels in close proximity of each other (0-20 Å). Using frozen samples, spectral broadening can be directly correlated to dipolar interactions. The dipolar effect is proportional to 1/r 3 and Fourier deconvolution of the spectrum can determine the distance between the spin labels (Hubbell et al., 1998). At room temperature, dipolar interactions are often averaged out due to the fast motion of the system. For slowly tumbling samples, dipolar interactions can be deconvoluted using rigid lattice methods and distances can be measured. These methods can be used to analyze intramolecular distances for proteins containing two R1 sites, or intermolecular distances between singly labeled proteins. Longer range distances (20-80 Å) can be measured using double electron-electron resonance (DEER) and pulse EPR (Jeschke, 2002; Jeschke and Polyhach, 2007). The longer the distance between spin labels the greater time it takes for them to undergo dipolar coupling. In order to observe long range interactions between spin labels, samples are subjected to an observer and pump frequency. With the observer frequency being applied to the EPR spectrum region being monitored, the pump frequency is added outside of this region. Changes in the EPR spectrum due to the pump frequency are observed as a spin echo. By observing the echo intensity versus time and applying a Fourier transform, distances can be determined. By strategically and systematically generating proteins containing two cysteine mutations global distance information can be used to three dimensionally define a proteins structure or verify data obtained from crystallography (Hubbell et al., 1998; Jao et al., 2008, 2010; Der-Sarkissian et al.; Shah et al., 2014). Alternatively, distances between singly labeled proteins can be used to determine quaternary structures and interactions between proteins. 20 1.4 Using SDSL and EPR Methodology to Determine how BAR Proteins Curve Membranes BAR proteins represent one of the largest and well studied membrane curving protein superfamilies (Rao and Haucke, 2011). Through extensive and meticulous study, individual members of the family have been shown to be important at different steps in membrane processes (Qualmann et al., 2011). This is likely to be due to the fact that while all BAR proteins share a scaffolding domain, the exact shape and curvature of the membrane binding interfaces tend to be slightly different between family members (Figure 1.7). It has therefore been proposed that BAR domains which are more flat in nature, such as FCHo2, initiate the initial positive curvature required for endocytosis through creating slight invaginations from the otherwise flat plasma membrane (Figure 1.8). In later stages of endocytosis the more highly curved BAR proteins endophilin and amphiphysin are recruited to carry out processes of more extreme curvature, such as the generation and scission of the cylindrical endocytotic necks. While these proteins have been observed to be important for endocytosis, the mechanistic processes underlying how these proteins generate curvature is still not fully understood. 21 Figure 1.7 BAR Protein Family Structures and Diversity. a) BAR domains create their membrane binding surface (black line) through homodimerization. Monomers are displayed in orange and teal. On average, BAR domains are anywhere from 10-20 nm in length and with variable degrees of curvature. Several subdomains of BAR proteins are also characterized by domains outside of the BAR domain, such as the sorting nexins which contain PX domains or the N-BAR domains which have N-terminal helices that bind membranes. b) F-BAR domains tend to have more flat degrees of curvature compared to that of traditional BAR domains such as arfaptin or endophilin. c) PinkBAR proteins have flat membrane binding surfaces while (d) I-BAR proteins bind using their convex surface. Figure adapted from Mim and Unger, 2012. 22 Figure 1.8 Curvature-Specific Involvement of BAR Proteins in Endocytosis. Different BAR proteins generate different levels of curvature and are recruited during different stages of endocytosis. BAR domains which prefer to bind lower degrees of curvature are first recruited to the membrane while BAR domains capable of stabilizing and producing high degrees of curvature are recruited at end stages. Importantly, BAR proteins have been shown to be capable of generating different membrane shapes, indicating that the type of curvature these proteins generate in vivo is highly regulated. Figure adapted from Qualmann et al., 2011. 23 1.4.1 Endophilin A1 on Tubes and Vesicles Endophilin A1 is an N-BAR protein involved in endocytosis as both a membrane curvature generator (Bai et al., 2010; Farsad et al., 2001; Gallop et al., 2006; Masuda et al., 2006) and as a recruiter of synaptojanin (Milosevic et al., 2011; Schuske et al., 2003; Soda et al., 2012; Verstreken et al., 2003) and dynamin (Meinecke et al., 2013; Ringstad et al., 1999; Sundborger et al., 2011). Most recently it was shown that endophilin controls a clathrin- independent tubulovesicular endocytic pathway which mediates uptake of G-protein-coupled receptors (Boucrot et al., 2015). It has also been shown to form both small vesicles and lipid tubes in vitro (Ambroso et al., 2014; Farsad et al., 2001; Gallop et al., 2006; Jao et al., 2010; Mim et al., 2012; Mizuno et al., 2010) although by which mechanisms is unclear. To understand how endophilin A1 binds and curves membranes, our lab utilized an SDSL, EPR, and computational analysis to determine the 3-D structure of endophilin A1 bound to lipid vesicles made from total brain lipids (Jao et al., 2010). DEER distances between sites in each monomer of the homodimerized BAR domain revealed the BAR domain to retain the structure resolved by X-ray crystallography when bound to small vesicles or in solution (Gallop et al., 2006). Bound to vesicles, spin labeled sites on the concave and convex surfaces of the BAR domain did not exhibit changes in spectral mobility or increased accessibility to O 2 , suggesting that the BAR domain is not the primary membrane-binding region when endophilin binds vesicles. Mobility data of the N-terminus and insert region of endophilin A1 which were unresolved in the crystal structure were used to show these regions become immobilized when endophilin A1 binds small vesicles (Gallop et al., 2006; Jao et al., 2010). A periodic accessibility to O 2 and NiEDDA showed the N-termini and insert region of endophilin A1 to form amphipathic helices. Using spin labeled lipids to generate a depth calibration, the N-terminal and insert region‘s helical 24 backbones were extrapolated to be at the level of the lipid phosphates and cemented as the main membrane-binding regions of endophilin A1 on small vesicles. To determine the conformation of the vesicle bound insert region helices, distance measurements between the insert region and BAR domain as well as between the insert regions in the same BAR domain were obtained and used as restraints in computational refinement. These distances were used in combination with accessibility data from the entire protein to develop a 3-D model of endophilin bound to small vesicles, and calculated the most likely model of endophilin‘s insert regions to be anti-parallel and perpendicular to the axis of the BAR domain (Jao et al., 2010). We therefore concluded that endophilin A1 binds small vesicles using its amphipathic wedges and not with its scaffolding BAR domain. The same analysis used on vesicle bound endophilin A1 was used to determine whether it uses a different structure to bind and form membrane tubes (Ambroso et al., 2014). Using a model membrane lipid composition of 2:1 DOPG:DOPE, homogenous endophilin A1-coated tube preparations were isolated and used for EPR experimentation. Accessibility measurements and lipid calibration of tube bound endophilin A1 revealed that the several spin sites on the concave surface of the BAR domain come into close contact with the tube membrane. It therefore seems that BAR domains only scaffold membranes when generating tubes and not when binding vesicles. A similar increase in O 2 accessibility was observed for the N-termini and insert helices, suggesting that endophilin A1 inserts its helices significantly deeper into the lipid bilayer of membrane tubes. As the wedging force of an amphipathic insertion is dependent on its depth in the membrane (Campelo et al., 2008), it seems that N-BAR proteins vary the depth of their amphipathic helices to generate different shapes. Distance measurements between sites in the insert region helices revealed them to retain the same anti-parallel and perpendicular to the 25 BAR domain conformation observed on vesicles. The structural differences SDSL and EPR can detect in how endophilin A1 binds small vesicles or tubes allows for us to understand how a single protein can generate distinct types of curvature. 1.4.2 Amphiphysin on Tubes and Vesicles A subsequent study was performed for the endophilin A1 homolog amphiphysin, which contains an BAR domain and N-terminus but does not have insert region helices (Peter et al., 2004). It has been suggested to participate in endocytosis and T-tubule formation (Butler et al., 1997; De Camilli et al., 1993; David et al., 1994; Kukulski et al., 2012; Lee et al., 2002; Meinecke et al., 2013; Razzaq et al., 2001). Knockout of amphiphysin in D. melanogaster destabilizes the T-tubule networks and mutated forms in humans are implicated in muscular diseases including centronuclear myopathies (Böhm et al., 2014; Claeys et al., 2010; Nicot et al., 2007; Razzaq et al., 2001). Like endophilin A1, amphiphysin has been shown in vitro to generate both vesicles and tubes (Isas et al., 2015; Peter et al., 2004). To determine if amphiphysin uses different mechanisms to generate distinct membrane shapes similar to what we observed for endophilin A1, an extensive SDSL and EPR study involving over 63 different spin labeled sites was utilized. Unlike our study on endophilin A1 which only analyzed 5 spin labeled sites in the BAR domain, this study involved over 49. Analysis of mobility spectra for residues lining the concave surface of the BAR domain of amphiphysin in solution or bound to small vesicles revealed minor changes and accessibility measurements showed that none of these residues penetrate into the vesicle membrane. Similar to endophilin A1, spin labeled sites in the N- terminal region of amphiphysin exhibited changes in mobility between solution and binding small vesicles. Oscillating O 2 accessibilities suggested they form amphipathic helices and lipid calibration showed their helical axis to penetrate to the level of the lipid phosphates. Taken 26 together these data suggest that amphiphysin binds vesicles predominantly through a wedging mechanism with its N-terminus and not by a scaffolding mechanism through its BAR domain. When the same analysis was performed for amphiphysin bound to tubes (Isas et al., 2015), sites along the entire concave surface were found to have levels of O 2 accessibility corresponding to direct membrane contact. One pocket of residues lining the tip regions of the BAR domain were found to deeply penetrate into the acyl chains of the bilayer. Increased O2 accessibilities were also measured for N-terminal spin sites and lipid calibration showed that they penetrate significantly deeper on tubes compared to vesicle membranes. The finding that amphiphysin and endophilin, two proteins of a massive protein-family, use similar structures and mechanisms to generate distinct membrane shapes suggests that these findings may be generalizable to a wide-variety of membrane curving proteins. 1.4.3 Regulation of Curvature through Post-translational Modifications While we have discovered that BAR proteins endophilin and amphiphysin use distinct mechanisms and structures to produce different membrane shapes, it is still unclear how these are controlled in the cell. Post-translational modifications act as molecular switches in a number of cellular processes and are a major mechanism by which cells control proteins. Many membrane-curving proteins and domains contain phosphorylation, methylation, and lysine acetylation. Phosphorylation has been shown to affect the way certain membrane curving proteins function, including ACAP4 (Zhao et al., 2013) and syndapin (Quan et al., 2012), and is shown in this thesis to directly regulate the membrane shape endophilin A1 induces. The affect PTMs can have on how other BAR proteins or lipid binding domains has not been studied. Lysine acetylation in specific is highly localized in membrane binding domains. This 27 modification negates the positive charge normally carried by the lysine side chain. Due to the fact that the addition of a negative charge to endophilin affects the way it interacts with membranes, we hypothesize that the negation of a positive charge may also have a similar effect. In this thesis we explore the increased prevalence of acetylation in lipid binding domains (Chapter 4). Using proteomics in combination with biophysical experimentation, we show that lysine acetylation is a common PTM in lipid binding regions of peripheral membrane proteins that can regulate how these proteins interact with membranes in vitro and in the cell. 1.5 Takeaway Membrane curvature is an essential biophysical property utilized within cells to execute their most basic functions. In this thesis, we use SDSL and EPR to elucidate how BAR proteins endophilin and amphiphysin induce different membrane shapes using distinct mechanisms and structures. The role post-translational modifications phosphorylation and acetylation play in regulating this function is investigated for BAR proteins and a large number of other membrane binding domains. Finally, we expand the characterization of the many functions of MBP to being a membrane curving protein, a function regulated by the disease-related PTM citrullination. 28 CHAPTER 1 REFERENCES Altenbach, C., Greenhalgh, D.A., Khorana, H.G., and Hubbell, W.L. (1994). A collision gradient method to determine the immersion depth of nitroxides in lipid bilayers: application to spin-labeled mutants of bacteriorhodopsin. Proc. Natl. Acad. Sci. U. S. A. 91, 1667– 1671. Ambroso, M.R., Hegde, B.G., and Langen, R. (2014). Endophilin A1 induces different membrane shapes using a conformational switch that is regulated by phosphorylation. Proc. Natl. Acad. Sci. 111, 6982–6987. Apostolidou, M., Jayasinghe, S.A., and Langen, R. (2008). Structure of alpha-helical membrane- bound human islet amyloid polypeptide and its implications for membrane-mediated misfolding. J. Biol. Chem. 283, 17205–17210. Bai, J., Hu, Z., Dittman, J.S., Pym, E.C.G., and Kaplan, J.M. (2010). Endophilin functions as a membrane-bending molecule and is delivered to endocytic zones by exocytosis. Cell 143, 430–441. Bergmann, C., Zerres, K., Senderek, J., Rudnik-Schoneborn, S., Eggermann, T., Häusler, M., Mull, M., and Ramaekers, V.T. (2003). Oligophrenin 1 (OPHN1) gene mutation causes syndromic X-linked mental retardation with epilepsy, rostral ventricular enlargement and cerebellar hypoplasia. Brain J. Neurol. 126, 1537–1544. Bobrovska, N., Góźdź, W., Kralj-Iglič, V., and Iglič, A. (2013). On the role of anisotropy of membrane components in formation and stabilization of tubular structures in multicomponent membranes. PloS One 8, e73941. Böhm, J., Biancalana, V., Malfatti, E., Dondaine, N., Koch, C., Vasli, N., Kress, W., Strittmatter, M., Taratuto, A.L., Gonorazky, H., et al. (2014). Adult-onset autosomal dominant centronuclear myopathy due to BIN1 mutations. Brain J. Neurol. Boucrot, E., Pick, A., Çamdere, G., Liska, N., Evergren, E., McMahon, H.T., and Kozlov, M.M. (2012). Membrane fission is promoted by insertion of amphipathic helices and is restricted by crescent BAR domains. Cell 149, 124–136. Boucrot, E., Ferreira, A.P.A., Almeida-Souza, L., Debard, S., Vallis, Y., Howard, G., Bertot, L., Sauvonnet, N., and McMahon, H.T. (2015). Endophilin marks and controls a clathrin- independent endocytic pathway. Nature 517, 460–465. Butler, M.H., David, C., Ochoa, G.C., Freyberg, Z., Daniell, L., Grabs, D., Cremona, O., and De Camilli, P. (1997). Amphiphysin II (SH3P9; BIN1), a member of the amphiphysin/Rvs 29 family, is concentrated in the cortical cytomatrix of axon initial segments and nodes of ranvier in brain and around T tubules in skeletal muscle. J. Cell Biol. 137, 1355–1367. De Camilli, P., Thomas, A., Cofiell, R., Folli, F., Lichte, B., Piccolo, G., Meinck, H.M., Austoni, M., Fassetta, G., Bottazzo, G., et al. (1993). The synaptic vesicle-associated protein amphiphysin is the 128-kD autoantigen of Stiff-Man syndrome with breast cancer. J. Exp. Med. 178, 2219–2223. Campelo, F., McMahon, H.T., and Kozlov, M.M. (2008). The hydrophobic insertion mechanism of membrane curvature generation by proteins. Biophys. J. 95, 2325–2339. Carpenter, E.P., Beis, K., Cameron, A.D., and Iwata, S. (2008). Overcoming the challenges of membrane protein crystallography. Curr. Opin. Struct. Biol. 18, 581–586. Cheng, C.-Y., Varkey, J., Ambroso, M.R., Langen, R., and Han, S. (2013). Hydration dynamics as an intrinsic ruler for refining protein structure at lipid membrane interfaces. Proc. Natl. Acad. Sci. U. S. A. 110, 16838–16843. Claeys, K.G., Maisonobe, T., Böhm, J., Laporte, J., Hezode, M., Romero, N.B., Brochier, G., Bitoun, M., Carlier, R.Y., and Stojkovic, T. (2010). Phenotype of a Patient with Recessive Centronuclear Myopathy and a Novel Bin1 Mutation. Neurology 74, 519–521. Columbus, L., Kálai, T., Jekö, J., Hideg, K., and Hubbell, W.L. (2001). Molecular motion of spin labeled side chains in alpha-helices: analysis by variation of side chain structure. Biochemistry 40, 3828–3846. Cui, H., Ayton, G.S., and Voth, G.A. (2009). Membrane binding by the endophilin N-BAR domain. Biophys. J. 97, 2746–2753. David, C., Solimena, M., and De Camilli, P. (1994). Autoimmunity in stiff-Man syndrome with breast cancer is targeted to the C-terminal region of human amphiphysin, a protein similar to the yeast proteins, Rvs167 and Rvs161. FEBS Lett. 351, 73–79. Doherty, G.J., and McMahon, H.T. (2009). Mechanisms of endocytosis. Annu. Rev. Biochem. 78, 857–902. Drin, G., and Antonny, B. (2010). Amphipathic helices and membrane curvature. FEBS Lett. 584, 1840–1847. Epand, R.M. (2015). Introduction to membrane lipids. Methods Mol. Biol. Clifton NJ 1232, 1–6. Farsad, K., and Camilli, P.D. (2003). Mechanisms of membrane deformation. Curr. Opin. Cell Biol. 15, 372–381. 30 Farsad, K., Ringstad, N., Takei, K., Floyd, S.R., Rose, K., and De Camilli, P. (2001). Generation of high curvature membranes mediated by direct endophilin bilayer interactions. J. Cell Biol. 155, 193–200. Ferguson, S.M., Ferguson, S., Raimondi, A., Paradise, S., Shen, H., Mesaki, K., Ferguson, A., Destaing, O., Ko, G., Takasaki, J., et al. (2009). Coordinated actions of actin and BAR proteins upstream of dynamin at endocytic clathrin-coated pits. Dev. Cell 17, 811–822. Frazier, A.A., Wisner, M.A., Malmberg, N.J., Victor, K.G., Fanucci, G.E., Nalefski, E.A., Falke, J.J., and Cafiso, D.S. (2002). Membrane Orientation and Position of the C2 Domain from cPLA2 by Site-Directed Spin Labeling†. Biochemistry 41, 6282–6292. Frazier, A.A., Roller, C.R., Havelka, J.J., Hinderliter, A., and Cafiso, D.S. (2003). Membrane- Bound Orientation and Position of the Synaptotagmin I C2A Domain by Site-Directed Spin Labeling†. Biochemistry 42, 96–105. Gallop, J.L., Jao, C.C., Kent, H.M., Butler, P.J.G., Evans, P.R., Langen, R., and McMahon, H.T. (2006). Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J. 25, 2898–2910. Hubbell, W.L., Gross, A., Langen, R., and Lietzow, M.A. (1998). Recent advances in site- directed spin labeling of proteins. Curr. Opin. Struct. Biol. 8, 649–656. Isas, J.M., Langen, R., Haigler, H.T., and Hubbell, W.L. (2002). Structure and Dynamics of a Helical Hairpin and Loop Region in Annexin 12: A Site-Directed Spin Labeling Study. Biochemistry 41, 1464–1473. Isas, J.M., Patel, D.R., Jao, C., Jayasinghe, S., Cartailler, J.-P., Haigler, H.T., and Langen, R. (2003). Global structural changes in annexin 12. The roles of phospholipid, Ca2+, and pH. J. Biol. Chem. 278, 30227–30234. Isas, J.M., Ambroso, M.R., Hegde, P.B., Langen, J., and Langen, R. (2015). Tubulation by amphiphysin requires a concentration-dependent switching from wedging to scaffolding. Structure in press. Itoh, T., Erdmann, K.S., Roux, A., Habermann, B., Werner, H., and De Camilli, P. (2005). Dynamin and the actin cytoskeleton cooperatively regulate plasma membrane invagination by BAR and F-BAR proteins. Dev. Cell 9, 791–804. Jao, C.C., Der-Sarkissian, A., Chen, J., and Langen, R. (2004). Structure of membrane-bound α- synuclein studied by site-directed spin labeling. Proc. Natl. Acad. Sci. U. S. A. 101, 8331–8336. 31 Jao, C.C., Hegde, B.G., Chen, J., Haworth, I.S., and Langen, R. (2008). Structure of membrane- bound alpha-synuclein from site-directed spin labeling and computational refinement. Proc. Natl. Acad. Sci. U. S. A. 105, 19666–19671. Jao, C.C., Hegde, B.G., Gallop, J.L., Hegde, P.B., McMahon, H.T., Haworth, I.S., and Langen, R. (2010). Roles of amphipathic helices and the bin/amphiphysin/rvs (BAR) domain of endophilin in membrane curvature generation. J. Biol. Chem. 285, 20164–20170. Jeschke, G. (2002). Distance Measurements in the Nanometer Range by Pulse EPR. ChemPhysChem 3, 927–932. Jeschke, G., and Polyhach, Y. (2007). Distance measurements on spin-labelled biomacromolecules by pulsed electron paramagnetic resonance. Phys. Chem. Chem. Phys. PCCP 9, 1895–1910. Jung, A.G., Labarrera, C., Labarerra, C., Jansen, A.M., Qvortrup, K., Wild, K., and Kjaerulff, O. (2010). A mutational analysis of the endophilin-A N-BAR domain performed in living flies. PloS One 5, e9492. Kirchhausen, T. (2012). Bending membranes. Nat. Cell Biol. 14, 906–908. Kukulski, W., Schorb, M., Kaksonen, M., and Briggs, J.A.G. (2012). Plasma membrane reshaping during endocytosis is revealed by time-resolved electron tomography. Cell 150, 508–520. Lai, C.-L., Jao, C.C., Lyman, E., Gallop, J.L., Peter, B.J., McMahon, H.T., Langen, R., and Voth, G.A. (2012). Membrane binding and self-association of the epsin N-terminal homology domain. J. Mol. Biol. 423, 800–817. Langen, R., Isas, J.M., Luecke, H., Haigler, H.T., and Hubbell, W.L. (1998). Membrane- mediated Assembly of Annexins Studied by Site-directed Spin Labeling. J. Biol. Chem. 273, 22453–22457. Langen, R., Oh, K.J., Cascio, D., and Hubbell, W.L. (2000). Crystal structures of spin labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra in terms of structure. Biochemistry 39, 8396–8405. Lee, E., Marcucci, M., Daniell, L., Pypaert, M., Weisz, O.A., Ochoa, G.-C., Farsad, K., Wenk, M.R., and De Camilli, P. (2002). Amphiphysin 2 (Bin1) and T-tubule biogenesis in muscle. Science 297, 1193–1196. Lipowsky, R. (2014). Remodeling of membrane compartments: some consequences of membrane fluidity. Biol. Chem. 395, 253–274. 32 Margittai, M., and Langen, R. (2008). Fibrils with parallel in-register structure constitute a major class of amyloid fibrils: molecular insights from electron paramagnetic resonance spectroscopy. Q. Rev. Biophys. 41, 265–297. Masuda, M., Takeda, S., Sone, M., Ohki, T., Mori, H., Kamioka, Y., and Mochizuki, N. (2006). Endophilin BAR domain drives membrane curvature by two newly identified structure- based mechanisms. EMBO J. 25, 2889–2897. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., De Bock, P.-J., Morais, V.A., Vilain, S., Haddad, D., et al. (2012). LRRK2 controls an EndoA phosphorylation cycle in synaptic endocytosis. Neuron 75, 1008–1021. Mchaourab, H.S., Lietzow, M.A., Hideg, K., and Hubbell, W.L. (1996). Motion of Spin-Labeled Side Chains in T4 Lysozyme. Correlation with Protein Structure and Dynamics†. Biochemistry 35, 7692–7704. McMahon, H.T., and Boucrot, E. (2015). Membrane curvature at a glance. J. Cell Sci. 128, 1065–1070. McMahon, H.T., and Gallop, J.L. (2005). Membrane curvature and mechanisms of dynamic cell membrane remodelling. Nature 438, 590–596. Meinecke, M., Boucrot, E., Camdere, G., Hon, W.-C., Mittal, R., and McMahon, H.T. (2013). Cooperative recruitment of dynamin and BIN/amphiphysin/Rvs (BAR) domain- containing proteins leads to GTP-dependent membrane scission. J. Biol. Chem. 288, 6651–6661. Milosevic, I., Giovedi, S., Lou, X., Raimondi, A., Collesi, C., Shen, H., Paradise, S., O‘Toole, E., Ferguson, S., Cremona, O., et al. (2011). Recruitment of endophilin to clathrin-coated pit necks is required for efficient vesicle uncoating after fission. Neuron 72, 587–601. Mim, C., and Unger, V.M. (2012). Membrane curvature and its generation by BAR proteins. Trends Biochem. Sci. 37, 526–533. Mim, C., Cui, H., Gawronski-Salerno, J.A., Frost, A., Lyman, E., Voth, G.A., and Unger, V.M. (2012). Structural basis of membrane bending by the N-BAR protein endophilin. Cell 149, 137–145. Mizuno, N., Jao, C.C., Langen, R., and Steven, A.C. (2010). Multiple Modes of Endophilin- mediated Conversion of Lipid Vesicles into Coated Tubes. J. Biol. Chem. 285, 23351– 23358. Nicot, A.-S., Toussaint, A., Tosch, V., Kretz, C., Wallgren-Pettersson, C., Iwarsson, E., Kingston, H., Garnier, J.-M., Biancalana, V., Oldfors, A., et al. (2007). Mutations in 33 amphiphysin 2 (BIN1) disrupt interaction with dynamin 2 and cause autosomal recessive centronuclear myopathy. Nat. Genet. 39, 1134–1139. Oh, K.J., Altenbach, C., Collier, R.J., and Hubbell, W.L. (2000). Site-directed spin labeling of proteins. Applications to diphtheria toxin. Methods Mol. Biol. 145, 147–169. Opella, S.J. (2013). Structure Determination of Membrane Proteins by Nuclear Magnetic Resonance Spectroscopy. Annu. Rev. Anal. Chem. 6, 305–328. Peter, B.J., Kent, H.M., Mills, I.G., Vallis, Y., Butler, P.J.G., Evans, P.R., and McMahon, H.T. (2004). BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 303, 495–499. Qualmann, B., Koch, D., and Kessels, M.M. (2011). Let‘s go bananas: revisiting the endocytic BAR code. EMBO J. 30, 3501–3515. Quan, A., Xue, J., Wielens, J., Smillie, K.J., Anggono, V., Parker, M.W., Cousin, M.A., Graham, M.E., and Robinson, P.J. (2012). Phosphorylation of syndapin I F-BAR domain at two helix-capping motifs regulates membrane tubulation. Proc. Natl. Acad. Sci. U. S. A. 109, 3760–3765. Rao, Y., and Haucke, V. (2011). Membrane shaping by the Bin/amphiphysin/Rvs (BAR) domain protein superfamily. Cell. Mol. Life Sci. 68, 3983–3993. Raschle, T., Hiller, S., Etzkorn, M., and Wagner, G. (2010). Nonmicellar systems for solution NMR spectroscopy of membrane proteins. Curr. Opin. Struct. Biol. 20, 471–479. Razzaq, A., Robinson, I.M., McMahon, H.T., Skepper, J.N., Su, Y., Zelhof, A.C., Jackson, A.P., Gay, N.J., and O‘Kane, C.J. (2001). Amphiphysin is necessary for organization of the excitation-contraction coupling machinery of muscles, but not for synaptic vesicle endocytosis in Drosophila. Genes Dev. 15, 2967–2979. Richard, J.-P., Leikina, E., Langen, R., Henne, W.M., Popova, M., Balla, T., McMahon, H.T., Kozlov, M.M., and Chernomordik, L.V. (2011). Intracellular curvature-generating proteins in cell-to-cell fusion. Biochem. J. 440, 185–193. Ringstad, N., Gad, H., Löw, P., Di Paolo, G., Brodin, L., Shupliakov, O., and De Camilli, P. (1999). Endophilin/SH3p4 is required for the transition from early to late stages in clathrin-mediated synaptic vesicle endocytosis. Neuron 24, 143–154. Saarikangas, J., Zhao, H., Pykäläinen, A., Laurinmäki, P., Mattila, P.K., Kinnunen, P.K.J., Butcher, S.J., and Lappalainen, P. (2009). Molecular mechanisms of membrane deformation by I-BAR domain proteins. Curr. Biol. CB 19, 95–107. 34 Der-Sarkissian, A., Jao, C.C., Chen, J., and Langen, R. (2003) Structural Organization of α- Synuclein Fibrils Studied by Site-directed Spin Labeling. J Biol. Chem.278, 37530- 37535. Schuske, K.R., Richmond, J.E., Matthies, D.S., Davis, W.S., Runz, S., Rube, D.A., van der Bliek, A.M., and Jorgensen, E.M. (2003). Endophilin is required for synaptic vesicle endocytosis by localizing synaptojanin. Neuron 40, 749–762. Shah, C., Hegde, B.G., Morén, B., Behrmann, E., Mielke, T., Moenke, G., Spahn, C.M.T., Lundmark, R., Daumke, O., and Langen, R. (2014). Structural Insights into Membrane Interaction and Caveolar Targeting of Dynamin-like EHD2. Structure 1993 22, 409–420. Soda, K., Balkin, D.M., Ferguson, S.M., Paradise, S., Milosevic, I., Giovedi, S., Volpicelli- Daley, L., Tian, X., Wu, Y., Ma, H., et al. (2012). Role of dynamin, synaptojanin, and endophilin in podocyte foot processes. J. Clin. Invest. 122, 4401–4411. Sorre, B., Callan-Jones, A., Manzi, J., Goud, B., Prost, J., Bassereau, P., and Roux, A. (2012). Nature of curvature coupling of amphiphysin with membranes depends on its bound density. Proc. Natl. Acad. Sci. U. S. A. 109, 173–178. Stachowiak, J.C., Schmid, E.M., Ryan, C.J., Ann, H.S., Sasaki, D.Y., Sherman, M.B., Geissler, P.L., Fletcher, D.A., and Hayden, C.C. (2012). Membrane bending by protein-protein crowding. Nat. Cell Biol. 14, 944–949. Stahelin, R.V., Scott, J.L., and Frick, C.T. (2014). Cellular and molecular interactions of phosphoinositides and peripheral proteins. Chem. Phys. Lipids 182, 3–18. Sundborger, A., Soderblom, C., Vorontsova, O., Evergren, E., Hinshaw, J.E., and Shupliakov, O. (2011). An endophilin-dynamin complex promotes budding of clathrin-coated vesicles during synaptic vesicle recycling. J. Cell Sci. 124, 133–143. Tang, M., Comellas, G., and Rienstra, C.M. (2013). Advanced Solid-State NMR Approaches for Structure Determination of Membrane Proteins and Amyloid Fibrils. Acc. Chem. Res. 46, 2080–2088. Varkey, J., Isas, J.M., Mizuno, N., Jensen, M.B., Bhatia, V.K., Jao, C.C., Petrlova, J., Voss, J.C., Stamou, D.G., Steven, A.C., et al. (2010). Membrane curvature induction and tubulation are common features of synucleins and apolipoproteins. J. Biol. Chem. 285, 32486– 32493. Verstreken, P., Koh, T.-W., Schulze, K.L., Zhai, R.G., Hiesinger, P.R., Zhou, Y., Mehta, S.Q., Cao, Y., Roos, J., and Bellen, H.J. (2003). Synaptojanin is recruited by endophilin to promote synaptic vesicle uncoating. Neuron 40, 733–748. 35 Watanabe, S., Rost, B.R., Camacho-Pérez, M., Davis, M.W., Söhl-Kielczynski, B., Rosenmund, C., and Jorgensen, E.M. (2013). Ultrafast endocytosis at mouse hippocampal synapses. Nature 504, 242–247. Westphal, C.H., and Chandra, S.S. (2013). Monomeric synucleins generate membrane curvature. J. Biol. Chem. 288, 1829–1840. Yandrapalli, N., Muriaux, D., and Favard, C. (2014). Lipid domains in HIV-1 assembly. Virology 5, 220. Yu, H., and Schulten, K. (2013). Membrane sculpting by F-BAR domains studied by molecular dynamics simulations. PLoS Comput. Biol. 9, e1002892. Zhao, X., Wang, D., Liu, X., Liu, L., Song, Z., Zhu, T., Adams, G., Gao, X., Tian, R., Huang, Y., et al. (2013). Phosphorylation of the Bin, Amphiphysin, and RSV161/167 (BAR) domain of ACAP4 regulates membrane tubulation. Proc. Natl. Acad. Sci. U. S. A. 110, 11023– 11028. Zhu, C., Das, S.L., and Baumgart, T. (2012). Nonlinear sorting, curvature generation, and crowding of endophilin N-BAR on tubular membranes. Biophys. J. 102, 1837–1845. 36 CHAPTER 2 Tubulation by amphiphysin requires concentration-dependent switching from wedging to scaffolding J. Mario Isas 1,* and Mark R. Ambroso 1,* , Prabhavati B. Hegde 1 , Jennifer Langen 2 , Ralf Langen 1,¥ 1 Zilkha Neurogenetic Institute, University of Southern California, Los Angeles, California 90033, USA. 2 Department of Biological Sciences, University of Southern California, Los Angeles, CA 90089-2910, USA. *J.M.I. and M.R.A. contributed equally to this work. ¥ Corresponding Author This work was originally published in Structure (2015), in press and formatted according to dissertation guidelines. EPR measurements and writing of the manuscript was performed by the author. 37 CHAPTER 2 ABSTRACT BAR proteins are involved in a variety of membrane remodeling events, but how they can mold membranes into different shapes remains poorly understood. Using EPR, we find that vesicle binding of the N-BAR protein amphiphysin is predominantly mediated by the shallow insertion of amphipathic N-terminal helices. In contrast, the interaction with tubes involves deeply inserted N-terminal helices together with the concave surface of the BAR domain, which acts as a scaffold. Combined with the observed concentration dependence of tubulation and BAR domain scaffolding, the data indicate that initial membrane deformations and vesicle binding are mediated by insertion of amphipathic helical wedges, while tubulation requires high protein densities at which oligomeric BAR domain scaffolds form. In addition, we identify a pocket of residues on the concave surface of the BAR domain that deeply insert into tube membrane. Interestingly, this pocket harbors a number of disease mutants in the homologous amphiphysin 2. 38 2.1 Introduction The remodeling of cellular membranes is controlled by proteins that sense, stabilize or induce membrane curvature (Rao and Haucke, 2011; Zimmerberg et al., 2006). Examples of membrane remodeling include vesicle budding and fusion events as well as the formation of cylindrical tubes in the cell. BAR domain-containing proteins have recently risen to prominence as they are involved in a wide range of membrane remodeling events. Amphiphysin is an N-BAR protein involved in endocytosis as well as T-tubule formation (Butler et al., 1997; David et al., 1994; De Camilli et al., 1993; Kukulski et al., 2012; Lee et al., 2002; Meinecke et al., 2013; Razzaq et al., 2001). The deletion of amphiphysin in Drosophila destabilizes the T-tubule network and mutations in humans have been shown to cause muscle diseases including centronuclear myopathy (Böhm et al., 2014; Claeys et al., 2010; Nicot et al., 2007; Razzaq et al., 2001). Endocytosis as well as T-tubule formation involves the shaping of membranes into curved entities, but while T-tubule formation results in the generation of elongated tubular structures, endocytosis is a more dynamic process that leads to the formation of curved vesicles. The ability of N-BAR proteins to partake in the formation or stabilization of different types of membrane curvature is also seen in vitro. Amphiphysin and the related N-BAR protein endophilin are capable of forming bilayer tubes of different diameter as well as cylindrical micelles and small highly curved vesicles (Farsad et al., 2001; Mizuno et al., 2010; Mim et al., 2012; Peter et al., 2004). Two of the main mechanisms that have been discussed for the formation of such structures are scaffolding and the wedging of amphipathic helices. The first evidence that amphiphysin and endophilin might be able to promote membrane curvature via scaffolding came from structural studies which showed that these proteins are banana-shaped dimers (Gallop et al., 2006; Peter et al., 2004; Weissenhorn, 2005). As shown 39 with the example of amphiphysin (Figure 2.1A), these dimers have a curvature complementary to that of the tubes they form. Moreover, the dimers are known to further associate into larger oligomeric networks on tubes (Mim et al., 2012; Mizuno et al., 2010; Takei et al., 1999; Yin et al., 2009). It has, therefore, been proposed that these proteins act via a scaffolding mechanism in which rigid BAR domain oligomers impose their shape onto the membrane and promote positive curvature. The concept of using scaffolding for controlling membrane curvature has since been expanded to other types of BAR domain proteins, such as F-BAR and I-BAR proteins (Arkhipov et al., 2009; Frost et al., 2008; Henne et al., 2007; Itoh et al., 2005; Pykäläinen et al., 2011; Saarikangas et al., 2009; Shimada et al., 2007; Yu and Schulten, 2013). However, membrane curvature induction by amphiphysin and endophilin does not solely rely on their BAR domains as other regions outside of the BAR domain are critical for tubulation as well. For example, mutations in the N-terminal regions of amphiphysin and endophilin as well as mutations in an insert region that is present in endophilin but not in amphiphysin, have been shown to significantly inhibit tubulation (Farsad et al., 2001; Gallop et al., 2006; Itoh and De Camilli, 2006; Peter et al., 2004). Site directed spin labeling (SDSL) studies using electron paramagnetic resonance (EPR) (Gallop et al., 2006; Jao et al., 2010) as well as computational studies (Blood and Voth, 2006; Blood et al., 2008; Cui et al., 2009, 2013; Mim et al., 2012) were able to verify that all of these regions fold up into amphipathic helical structures that are thought to act as wedges in the membrane. Work on the Parkinson‘s disease protein α-synuclein has shown that helical wedges can be sufficient for inducing tubulation, demonstrating that scaffolding is not a prerequisite for membrane bending (Varkey et al., 2010; Westphal and Chandra, 2013). Interestingly, endophilin‘s BAR domain was found to be at a significant distance away from the membrane under vesiculating conditions suggesting that the amphipathic helices might be the 40 predominant curvature-stabilizing entities (Jao et al., 2010). In order to address the apparent discrepancy with respect to the importance of scaffolding and helix insertion, a recent study used using SDSL and EPR found that endophilin employs different structures and mechanisms to generate vesicles or tubes (Ambroso et al., 2014). The central finding was that the N-terminal and insert region helices were more deeply embedded in tube than vesicle membranes. Based on membrane proximity measurements of two residues, it was also suggested that this helical movement is coupled to a shift of the BAR domain towards the membrane. This movement could significantly enhance the scaffolding activity of the BAR domain. However, the precise orientations of endophilin‘s BAR domain with respect to vesicle or tube membranes remain unknown. Moreover, a recent study on the BAR protein ACAP1 suggested that tube formation is not directly mediated by the BAR domain but by membrane interaction with its pleckstrin homology domain (Pang et al., 2014). Thus, tube formation may not require BAR domain- dependent scaffolding. Amphiphysin is different from endophilin and ACAP1 as it lacks insert helices or PH domains. We therefore set out to test (a) whether the BAR domain of amphiphysin is involved in scaffolding and (b) whether amphiphysin uses different mechanisms to interact with vesicles or tubes. In order to obtain detailed structural information for the membrane interaction of the BAR domain and N-terminal helix of amphiphysin (henceforth referred to as amphiphysin), we performed SDSL for 49 spin labeled sites in the BAR domain and 14 sites in the N-terminus. We then used EPR of these spin labeled derivatives to investigate the structures of amphiphysin bound to vesicles or tubes. In support of a scaffolding mechanism, we find that the entire concave surface of the BAR domain comes into direct contact with the tube membrane. Residues on this surface exhibit significant membrane interaction with all sites at least penetrating into the lipid headgroup region. The most deeply inserted residues (144, 147, and 41 151) were found in a pocket that even inserts into the acyl chain region of the membrane. Interestingly, both this pocket and the N-terminal helix, which we find to differentially interact with tubes and vesicles, are in the immediate vicinity of known familial disease mutants (Böhm et al., 2014; Claeys et al., 2010; Nicot et al., 2007; Razzaq et al., 2001). In contrast to tubes, scaffolding does not seem to be important on vesicles where the BAR domain only minimally contacts the membrane and where membrane interaction is predominantly mediated by the N- terminal helices. 2.2 Results 2.2.1 Amphiphysin vesicle binding is mediated by shallow insertion of its amphipathic N- terminal helix In order to compare the structures of vesicle and tube bound amphiphysin, we sought conditions that resulted in clean preparations of amphiphysin either bound to vesicles or tubes. We found that amphiphysin both vesiculated and tubulated vesicles containing the commonly used total brain lipids. While it was possible to enrich for vesicles and tubes, it was difficult to obtain clean preparations containing only tubes or vesicles. This problem was further compounded by significant batch-to-batch variations in the total brain lipid extracts. Thus, we examined a large number of conditions and found two different defined lipid compositions that resulted in amphiphysin stably bound to either vesicles or tubes. The structure of amphiphysin bound to the respective structures was then studied using SDSL and EPR spectroscopy. Toward this end, we introduced spin labels, one amino acid at a time, at 14 selected sites in the N-terminal region of amphiphysin (not resolved in the crystal structure) and 49 sites throughout the BAR domain 42 (Figure 2.1A). First, we examined the changes in structure and membrane topography that occur as amphiphysin binds to vesicles containing a mixture of phosphatidylserine and cholesterol (Figure 2.1B). The EPR spectra for N-terminally labeled sites revealed that this region is unfolded in solution (Figure S2.1A) explaining why the N-terminus was not resolved in the crystal structure. The change in the EPR spectra upon vesicle binding, however, reveals reduced mobility and ordering of the N-terminus (Figure S2.1B). In order to determine the orientation of the N-terminus with respect to the vesicle membrane, we measured the membrane immersion depths of each spin labeled site through accessibility measurements (Figure S2.1C and S2.1D). These measurements take advantage of the well established collision gradients that increase toward the center of the membrane in the case of hydrophobic O 2 , while the opposite is observed for collisions with hydrophilic NiEDDA (Altenbach et al., 1994). The data can be conveniently summarized with the depth parameter Ф, which is defined by Ф=ln(ΠO 2 /ΠNiEDDA). Larger Ф values indicate increasing membrane immersion depths (Altenbach et al., 1994; Frazier et al., 2003). As illustrated in Figure 2.1C, Ф oscillates as a function of N-terminal sequence with a periodicity that indicates that the N-terminus folds into an α-helix upon membrane binding (Hubbell et al., 1998). When plotted onto a helical wheel (Figure 2.1D), all lipid exposed residues (local maxima in Ф plot, red) are on one face, while all solvent exposed residues (local minima in Ф plot, blue) are on the other, suggesting the helix is amphipathic. Based upon calibration using spin labeled lipids (see methods, Figure S2.1E), we find that the deepest lipid exposed sites are located at immersion depths of 7 to 10 Å. Considering the extension of the R1 side chain (~7 to 10 Å from the center of an α-helix) (Langen et al., 2000), the N-terminal helix is parallel to the membrane with its center approximately at the level of the head group 43 phosphates, where it is well positioned to act as a molecular wedge (Figure 2.1E) (Campelo et al., 2008, 2010; Drin and Antonny, 2010). Next we investigated the structure and membrane proximity of the BAR domain on vesicles. Comparison of the EPR spectra for spin labeled derivatives bound to vesicles or in solution revealed that mobility changes between the two states were highly localized (Figure S2.1F and S2.1G). These changes are summarized in Figure 2.1F using the central line width, a commonly employed mobility parameter (Jao et al., 2004). Little or no mobility changes were observed at sites near the convex side of the protein as well as at the more central regions along the concave surface of the BAR domain. In contrast, the strongest immobilization was observed near the tip region of amphiphysin for residues 146, 148, 150-153, 168, 170-174, and 176. These residues are located at the end of helix 2 and the beginning of helix 3, respectively (Figure 2.1A and 1H). In solution, these regions have multi-component EPR spectra indicative of ordered and disordered structure consistent with frayed helical ends. Upon vesicle binding these regions become much more ordered. The high mobility in the intervening loop region (residues 156 to 167) remains. When accessibility measurements were performed, no labeling site in the BAR domain gave rise to strongly enhanced O 2 accessibilities (Figure 2.1G; Table S2.1). Ф values for all sites ranged between -0.9 and -2, values that are typically observed for soluble proteins in aqueous solution under these conditions (Shah et al., 2014). Importantly, all sites lie outside the region for which immersion depth is calibrated, from the interior of the bilayer to ~5 Å above the phosphate level (Figure 2.1G, dashed line; Figure S2.1E). This lack of calibration precludes us from directly determining the precise location of the BAR domain relative to the membrane, however the data still allow us to estimate a lower limit for the distance between the BAR domain and the vesicle membrane. Considering that all sites lay outside the calibrated depth 44 range, their nitroxide moieties must be at least 5 Å above the phosphate level. Many labeled sites include residues on the concave surface of the BAR domain and are predicted to project directly toward the membrane. Since the backbone is at least another 7 Å away from the nitroxide moiety, we can therefore estimate that the backbone to phosphate distance in these cases must be at least 12 Å. Thus, if the BAR domain contacts the membrane, such contacts must be limited to interactions with the more distal region of the headgroups. Considering the different geometries of the modestly curved membrane and the highly curved BAR domain, such contacts could involve the tip region of the BAR domain (Figure 2.1H), which experiences a significant reduction in mobility upon vesicle binding. However, we cannot exclude the possibility that vesicle interaction causes an ordering of the frayed helical ends via alternative mechanisms that do not require physical contact of this region with the membrane. 45 Figure 2.1 Amphiphysin binds vesicles with its N-terminus and not its BAR domain. (A) Crystal structure of homodimer drosophila amphiphysin (pdb 1URU) with individual subunits in blue and teal. Yellow α-carbons highlight residues in the BAR domain which were spin labeled (sequence numbers are provided for select sites). The N-terminal regions of amphiphysin are unresolved in the crystal structure and not shown. (B) Negative stain electron microscopy image of amphiphysin bound to 100 nm vesicles composed of a 4:1 weight to weight ratio of 1- palmitoyl-2-oleoyl-sn-glycero-3-phospho-L serine and cholesterol. Scale bar = 500 nm. (C) The depth parameter Ф for vesicle bound amphiphysin is plotted versus N-terminal spin labeling position. (D) Helical wheel representation of the N-terminal region of amphiphysin. More deeply membrane inserted residues (red) and more solvent exposed residues (blue) are color coded as in (C). (E) Model of amphiphysin‘s N-terminal helix bound to a vesicle membrane. The helix (gray) was manually placed according to immersion depths obtained from Ф values in Figure 2.1C and the calibration in Figure S2.1E (see text). Single phospholipids (orange) as well as the level of the lipid headgroups (black line) and phosphates (gray line) are displayed for reference. (F) The change in central line width upon transitioning from solution to the vesicle bound form of amphiphysin is plotted versus spin labeled position. Positive changes indicate ordering upon vesicle binding. Sites with the largest changes (cutoff arbitrarily set at 1 Gauss) are highlighted (green). (G) The depth parameter Ф for spin labeled sites in amphiphysin bound to 100 nm vesicles plotted versus labeled position. The ΠO 2 and ΠNiEDDA values used to calculate Ф are given in Figure S2.1D and Table S2.1. The Ф value obtained for a lipid moiety with a spin label at the position of the lipid headgroup (see methods) is displayed for reference (black dashed line). (H) Model of amphiphysin‘s BAR domain bound to 100 nm vesicles. The N-termini are 46 omitted for simplicity. Residues in the BAR domain that undergo line width changes greater than 1 Gauss upon binding vesicles are colored green as in Figure 2.1F. The location of the BAR domain relative to the membrane was estimated as described in the text. The level of the headgroups (black line) and lipid phosphates (gray line) are displayed for reference. Error bars represent SD, n = three independent experiments. 47 Figure 2.1 continued 48 Figure S2.1 EPR continuous wave spectra, as well as NiEDDA and O 2 accessibilities, of spin labeled derivatives of amphiphysin bound to 100 nm vesicles. (A) Continuous wave (CW) EPR spectra of spin labeled amphiphysin containing spin label R1 at position 13 in solution (black) and bound to vesicles (red). Sharp spectral lines indicate high spin label mobility in solution. The broadened spectra for the vesicle bound form indicate reduced mobility. All other N-terminal labeling sites had similar spectra in solution indicating an unfolded structure. (B) The CW EPR spectra of N-terminally spin labeled derivatives of amphiphysin when bound to vesicles. (C) The N-terminal amino acid sequence of Drosophila melanogaster amphiphysin. (D) Accessibility to oxygen (ΠO 2 ) and NiEDDA (ΠNiEDDA) as function of labeling position for vesicle bound amphiphysin. The periodicity in the respective accessibilities is indicative of α- helical secondary conformation (Hubbell et al., 1998) and the out-of phase periodicity indicates that this helical structure is amphipathic. (E) Lipids containing spin labels at known bilayer immersion depths (Bretscher et al., 2008; Dalton et al., 1987) were analyzed for accessibility to O 2 and NiEDDA in vesicles bound by non-spin labeled amphiphysin. These values were used to calibrate (Ф) with respect to membrane immersion depth (d) in Å (Altenbach et al., 1994). The values were fit to a hyperbolic tangent function (black line), Ф= A tanh[B(x-C)] + D, as previously described (Frazier et al., 2002), where A= 2.7, B= 0.08, C= 14, D= 2.5, and x is depth given in Å. Error bars represent SD, n = at least three independent experiments. (F-G) CW EPR spectra of vesicle bound (red) and soluble (black) amphiphysin spin labeled in the BAR domain at the indicated positions, which are also shown in Figure 2.1A. High or low spectral amplitudes were rescaled using the indicated scaling factor (e.g., ―2x‖) for space considerations. Scan width is 100 gauss. 49 Figure S2.1 continued 50 2.2.2 On tubes, the BAR domain comes into contact with the membrane and N termini move deeper into the bilayer In order to compare the structure of vesicle bound amphiphysin with that of the tube bound protein, we repeated the experiments under tubulating conditions using phosphatidylglycerol and phosphatidylethanolamine containing vesicles. Under the conditions used, the membranes became completely tubulated with a well defined, highly oligomeric protein coat (Figure 2.2A and 2.2B). The EPR spectra of sites in the N-terminal region revealed that this region also becomes ordered on tubes (Figure S2.2A and S2.2B). Moreover, the periodicity in the accessibility data (Figures 2.2C and S2.2C) again indicated the formation of α-helical structure. When compared to the vesicle bound form, however, much larger Ф values were obtained for the N-terminus on tubes indicating deeper membrane insertion. According to the calibration (Figure S2.2D), the most deeply inserted residues are on average ~16 Å below the phosphate level. Taking into account the 7-10 Å length of the spin labeled moiety from the center of the α-helix (Langen et al., 2000), we approximate the helical wedges to be driven more deeply into the tube membrane with their centers penetrating about 6-9 Å into the acyl chain region (Figure 2.2D). Having established that the location of the N-terminal helices differs between vesicles and tubes, we next wanted to determine whether similar changes might also occur in the BAR domain. As in the case of vesicle binding, tubulation caused spectral changes for spin labeled sites in the BAR domain that were highly dependent on where the label was introduced (Figure S2.3A and S2.3B). Again, the EPR spectral changes were mainly limited to sites located on the concave surface. However, there were also some distinctive differences between the spectra of the vesicle and tube bound proteins. A comparison of the central line widths revealed that the mobility of the vesicle and tube bound states differed most for sites on the concave surface 51 (Figure S2.3C and S2.3D). The most pronounced differences were observed for residues 58, 97, 113, 133, 151, 154, 158 and 171, most of which are spread out over the presumed membrane interaction surface. This is in contrast to the more localized mobility changes observed mainly in the tip region for the vesicle bound protein. To test whether this enhanced ordering of residues on the concave surface is accompanied by altered membrane proximity of the BAR domain, we again performed O 2 and NiEDDA accessibility measurements. The respective accessibilities are given in Table S2.1 and are summarized by the Ф values in Figure 2.3A. As in the case of vesicle bound amphiphysin, sites on the convex surface again resulted in negative Ф values indicating that these regions do not come into direct contact with the membrane. However when labeling occurred at the concave surface, more positive Ф values could be seen indicating membrane proximity or membrane penetration for those sites. The largest Ф values were obtained for residues 58, 133, 144, 147, 148, 151, 154, 170, and 171 indicating that residues on the concave surface of the BAR not only become more ordered but that they also come into closer proximity to the membrane. These enhanced Ф values are not the consequence of a previously observed steric exclusion of NiEDDA (Isas et al., 2002) but are instead due to enhanced accessibility to O 2 (Figure S2.3E). While the accessibility values for most of these sites are consistent with a location near the lipid headgroup level, a cluster of residues (144, 147, and 151) penetrates more deeply (nitroxide moiety at 2-7 Å immersion depth) into the acyl chain region (Figures 2.3B and S2.2D). 52 Figure 2.2 The N-terminus submerges into the acyl chain region on tubes. (A) Lipid tubes formed from large vesicles composed of a 2:1 weight to weight ratio of 1-palmitoyl-2-oleoyl-sn- glycero-3-[phospho-RAC-(1-glycerol)] and 1-palmitoyl-2-oleoyl-sn-glycero-3- phosphoethanolamine after incubation with amphiphysin at a 1:10 (protein:lipid) weight to weight ratio observed by negative stain electron microscopy. Scale bar = 500 nm. (B) High magnification reveals a striated amphiphysin protein coat around a lipid tube. Scale bar = 50 nm. (C) The depth parameter Ф as a function of residue number in the N-terminus on tubes. O 2 (red) and NiEDDA (blue) accessible residues are color coded similarly to the helical wheel in Figure 2.1D. Ф values obtained for vesicle bound amphiphysin (gray dashed line) from Figure 2.1C are plotted for comparison. (D) Model of amphiphysin‘s N-terminus bound to tubes. The helix was manually placed according to the calibration (Figure S2.2D) and depth measurements (Figure 2.2C) as described in the text. Single lipids (orange) are displayed for reference. The plane of the headgroups (black line) and lipid phosphates (gray line) are displayed for reference. Error bars represent SD, n = three independent experiments. 53 Figure S2.2 EPR continuous wave spectra, as well as NiEDDA and O 2 accessibilities, of spin labeled derivatives in the N-terminus of amphiphysin bound to lipid tubes. (A) CW EPR spectra of amphiphysin R1-labeled at position 13 in solution (black) and bound to tubes (red). (B) CW EPR spectra of tube bound amphiphysin labeled with R1 at the indicated N- terminal sites. (C) Accessibility to O 2 (ΠO 2 ) and NiEDDA (ΠNiEDDA) as a function of labeling position for tube bound amphiphysin. (D) Accessibility to O 2 and NiEDDA of spin labeled lipids in tubes formed by non-spin labeled amphiphysin were used to calibrate Ф with respect to immersion depth (Altenbach et al., 1994). The values were fit to a hyperbolic tangent function (black line), Ф= A tanh[B(x-C)] + D, where A= 3.1, B= 0.04, C= 10, D= 1.2, and x is depth given in Å. Error bars represent SD, n = at least three independent experiments. 54 2.2.3 Tubulation and BAR domain scaffolding are dependent on protein concentration Recent studies using mechanically pulled nanotubes found two distinct ways in which amphiphysin or endophilin can stabilize membrane curvature (Sorre et al., 2012; Zhu et al., 2012). At dilute protein conditions, the effect of these proteins was relatively modest, but at Figure S2.2 continued 55 higher protein-to-lipid ratios they became more potent and stabilized tubes similar in size to those generated here. In order to test how different protein concentrations might affect amphiphysin membrane interaction, we repeated the tubulation experiments at four fold lower protein-to-lipid ratios. According to negative stain electron microscopy, these conditions reduced overall tubulation and more vesicular structures were present (Figure 2.3C). As expected from the more abundant presence of vesicles, accessibility measurements resulted in a significant reduction of Ф values indicating reduced interaction of the BAR domain with the membrane (Figure 2.3D). These experiments suggest that a threshold protein-to-lipid ratio is required in order for stable tubulation to occur. Inasmuch as tubulation coincides with the formation of an oligomeric protein coat around the tubes, it is likely that the switch to scaffolding is coupled to oligomer formation. The finding that decreasing protein-to-lipid ratios cause reduced tubulation as well as reduced Ф values also served as an important control. Our analysis of tube and vesicle-bound protein in Figures 2.1, 2.2, 2.3A and 2.3B used two different lipid compositions, while the data in Figure 2.3C and 2.3D used the same lipid composition. Thus, the strong effects on the Ф values in Figure 2.3D must be due to the decreased yield of tubulation and not lipid composition. To further address this issue, we also investigated the Ф values from selected amphiphysin derivatives bound to total brain lipid membranes. Again we observed the same trend that preparations enriched for tubes had more positive Ф values than those enriched for vesicles (data not shown). 56 Figure 2.3 The BAR domain adheres its concave surface to the membrane on tubes. (A) The depth parameter, Ф, is plotted as function of the labeling position within the BAR domain of tube bound amphiphysin. The black dashed line corresponds to a headgroup location approximately 5 Å above the phosphate level (see methods; Figure S2.2D). Residues with Ф values equal to or greater than that of the dashed line are in red, while residues with smaller values are in blue. Ф values obtained for select residues of vesicle bound amphiphysin (gray dashed line) from Figure 2.1G are plotted for comparison. (B) Model of amphiphysin‘s BAR domain bound to tubes with the N-terminal helices omitted for simplification. The level of the headgroups (black line) and lipid phosphates (gray line) on a lipid leaflet (orange) are schematically displayed for reference. α-carbons are color coded as in Figure 2.3A. (C) Negative stain electron microscopy of a mixture of lipid tubes and small vesicles formed from large vesicles after incubation with amphiphysin at a 1:40 (protein:lipid weight ratio). Scale bar = 500 nm. (D) Depth parameter (Ф) for select sites on the BAR domain when bound to tubes at various protein to lipid weight ratios; 1:10 (black) and 1:40 (gray). Ф values taken from Figure 2.1G for the same sites when bound to 100 nm vesicles (white) are also shown for reference (protein to lipid molar ratio of 1:10). Decreasing Ф values represent decreasing exposure of these sites to the membrane environment. Error bars represent SD, n = three independent experiments. 57 Figure S2.3, related to Figure 3. Spin label mobility and accessibility for spin labeled amphiphysin bound to tubes or to vesicles. (A-B) CW EPR spectra of amphiphysin derivatives spin labeled in the BAR domain at the indicated positions in solution (black) or bound to tubes (red). High or low spectral amplitudes were rescaled using the indicated scaling factor (e.g., ―2x‖) for space considerations. Scan width is 100 gauss. (C) The percent change in CW EPR spectral central line width for spin labeled amphiphysin derivatives bound to tubes or to vesicles is plotted versus residue number. Derivatives with a percent change in central line width greater than or equal to 10% (dashed line) are highlighted (green). (D) Crystal structure of drosophilia amphiphysin dimer (pdb, 1URU) indicating spin labeled derivatives with percent central line width changes greater than 10% in green as determined in (C). Residue numbers are given for only one subunit of the dimer. (E) Plot of ΠO 2 versus ΠNiEDDA for spin labeled sites in the BAR domain bound to tubes (black) or vesicles (red). Prior studies have shown that the accessibilities to O 2 and NiEDDA are linearly related for proteins in solution (Isas et al, 2002). This is indeed the case for all sites in the BAR domain upon vesicle binding, in agreement with the notion that the BAR domain is too far from Figure 2.3 continued 58 the membrane to experience enhanced O 2 accessibility. In contrast, accessibility values for several tube bound derivatives do not fall into the same diagonal region delineated by the dashed lines. A number of sites fall below the dashed line as a consequence of enhanced O 2 accessibility due to membrane proximity. The plot also shows that the enhanced Ф values in Figure 2.3A are not due to the ―excluded volume effect‖. In very rare cases, enhanced Ф values can be the consequences of the ―excluded volume effect‖ in exceedingly immobilized regions (Isas et al., 2002). In such a scenario, sites may be very inaccessible to both O 2 and NiEDDA, but due to its smaller size, accessibility to O 2 is non-zero while accessibility to the larger molecule of NiEDDA is near zero. The plot indicates that this is not the case here. 59 2.3 Discussion: In this study, we set out (a) to determine what mechanisms amphiphysin uses to bend lipid membranes and (b) to provide a detailed structural analysis of the BAR domain on tubes and vesicles. Our structural analysis revealed that amphiphysin uses different structures and mechanisms to interact with vesicle or tube membranes. The BAR domain only weakly contacts Figure S2.3, continued 60 the vesicle membrane. Vesicle interaction reduces the mobility at the tip regions of the BAR domain but even these residues do not experience significant membrane immersion. Thus, if there is any interaction between the BAR domain and the membrane, it is likely limited to contacts with the more distal portions of the lipid head groups. In contrast, tube binding involves the entire concave surface of the BAR domain. Residues throughout this surface are in direct contact with the membrane and penetrate deeply into the headgroup region and in some cases even into the acyl chain region. This downward movement allows the BAR domain to impart its own curved structure onto the membrane and act as a scaffold. Scaffolding is further aided by the formation of specific oligomeric structures that wrap around the tubes (Figure 2.2B) (Mim et al., 2012; Mizuno et al., 2010; Takei et al., 1999). We also find that the N-terminus is capable of forming three distinct structures: (a) a random coil in solution, (b) an amphipathic helix shallowly inserted into vesicles, or (c) an amphipathic helix deeply inserted into the tube membrane. The movement of the BAR domain toward the tube membrane allows for numerous positively charged residues on its concave surface to engage in electrostatic interactions with negative charges on the lipid headgroups. This interaction is likely to contribute a significant amount of energy towards the formation of a tight scaffold. Although the entire concave surface is in contact, the binding interactions are not perfectly uniform. Interestingly, we identified a pocket of amino acids (~144 to 151) with reduced mobility that more deeply penetrate into the tube membrane than other amino acids on the concave surface of the BAR domain (Figure 2.3A and 3B). The importance of this region is further illustrated by the fact that the homologous region in amphiphysin 2 harbors mutations found in familial forms of centronuclear myopathy (Claeys et al., 2010; Nicot et al., 2007). These D151N and R154Q mutations (corresponding to position 61 D146 and R149 in the present study) are thought to inhibit tubulation in vivo as well as in vitro (Claeys et al., 2010; Nicot et al., 2007; Wu et al., 2014). One possibility is that this membrane insertion pocket makes specific lipid contacts. In fact, a recent study on F-BAR proteins identified a conserved lipid binding site on the concave membrane binding surface of these proteins (Moravcevic et al., 2015). A specific coordination of lipids is consistent with the strong immobilization observed in the EPR spectra for this region, as prior studies on annexins have shown that specific lipid coordination results in pronounced immobilization (Isas et al., 2002). In principle, protein-protein contacts, such as contacts with N-terminal helices, could contribute to the observed immobilization as well. For example, one could envisage an interaction between N- terminal helices from one dimer with a BAR domain from an adjacent dimer. Such an interaction could stabilize the oligomeric coats on tubes and couple the movement of the BAR domain to the movement of the N-terminal helices. However, such contacts were not resolved in cryoEM reconstructions of endophilin tubes (Mim et al., 2012; Mizuno et al., 2010) and additional studies will be needed to test for this possibility. Regardless of the precise mechanism, geometric considerations may explain why oligomerization is more pronounced on tubes. While a vesicle is isotropically curved (curved in 3 dimensions), a tube is curved anisotropically with curvature around but not along its axis. We would therefore expect vesicle-associated BAR proteins to follow the isotropic curvature of the vesicle and to be oriented in various directions. In contrast, tube-bound BAR proteins are much more likely to be aligned in similar orientations in order to stabilize the anisotropic curvature. Oligomerization should greatly facilitate such an alignment. The finding that amphiphysin‘s N-terminal helices are the primary membrane interacting region on vesicles is consistent with a recent microscopy study that found the BAR domain alone to possess little curvature sensitivity toward vesicles; rather membrane curvature sensing was 62 mainly mediated by the N-terminal helices (Bhatia et al., 2009). The shallow insertion of amphiphysin‘s N-terminal helices is a common conformation also observed for vesicle binding of amphipathic helices from epsin, α-synuclein and endophilin (Gallop et al., 2006; Jao et al., 2008, 2010; Lai et al., 2012). A computational study indicated that shallowly inserted helices selectively wedge into the head groups (Campelo et al., 2010), thereby stabilizing curvature in a manner akin to the spontaneous curvature effect of lipids with large headgroups and small acyl chains. On tubes, however, the N-terminal helices submerge beyond the lipid phosphates and into the acyl chain region where they produce a reduced amount of spontaneous curvature (Campelo et al., 2008, 2010). In addition to no longer applying pressure exclusively to the headgroup region, the N-terminal helices may also push the acyl chain regions apart. This feature may be beneficial for tubulation by allowing amphiphysin to compensate for lipid vacancies that occur as a target vesicle is being remodeled into a tubular structure, where the surface area of the outer leaflet increases significantly, requiring additional lipids or protein to fill out the leaflet. The ability to take up space in the outer leaflet may be particularly important in cases where lipid flip-flop is slow or inefficient such as in the case of membranes with low cholesterol content. The deep insertion of the amphipathic helices may also alter the structural organization of nearby lipids. The interaction of the amphipathic helices with adjacent lipids is likely to cause a bending of these lipids, which in turn will cause local membrane thinning. This bending could occur for two reasons, 1) the negatively charged moieties of the lipid headgroups are expected to bend around the hydrophilic surface of the amphipathic helices to interact with positively charged residues (Ambroso et al., 2014), and 2) the acyl chains are likely to bend around the hydrophobic surface of the helices to take up otherwise empty space between the helices and the center of the bilayer. Inasmuch as membrane thickness is inversely related to rigidity, local membrane 63 thinning would facilitate membrane bending. Moreover, the thinning of the membrane could make protein oligomerization energetically favorable. Close spatial proximity of two helical wedges brings two locally thinned membranes together, limiting the number of energetically unfavorable membrane thickness transitions. Such matching of membrane thickness has previously been found to be an important factor in clustering transmembrane proteins (Haselwandter and Phillips, 2013; Haselwandter and Wingreen, 2014). It should be noted that disease mutations found in centronuclear myopathy are not limited to the BAR domain of amphiphysin 2 as several mutants have been mapped to the N-terminus (K21del and R24C; K16 and R19 respectively for the structurally homologous amphiphysin used in this study) (Böhm et al., 2014; Claeys et al., 2010; Nicot et al., 2007). The interesting similarity between the mutations in the BAR domain and the N-terminus is that they all become more deeply inserted in the tube bound state. How the mutations affect these conformational transitions remains to be tested. A prior study found that increased length or number of helical wedges promoted vesiculation while enhanced scaffolding promoted tubulation (Boucrot et al., 2012). Our finding that vesiculation is mainly mediated by the insertion of amphipathic helices while tubulation is accompanied by enhanced scaffolding is in good agreement with this notion. However, our data also reveal an important detail, namely that the overall length and number of helices is not the only consideration. This is because the same protein has the ability to switch between different structural states that either rely on wedging or a combination of wedging and scaffolding. A dichotomy of mechanisms used by amphiphysin to interact with membranes is further consistent with previous work in which amphiphysin was found to mainly sense curvature in conditions of low protein density while inducing curvature at higher protein densities (Sorre et al., 2012). Our 64 structural data indicate that the enhanced ability to cause extensive tubulation at higher protein densities is caused by movements of the BAR domain and the amphipathic helices toward the membrane that occur as the oligomeric scaffolds are formed. Taking all of these data together, it appears likely that the initial membrane interactions of amphiphysin at low protein densities are predominately mediated by the wedging of amphipathic helices (Figure 2.4A). Such wedging could then create an initial membrane bending. As the membrane bending increases and amphiphysin reaches a threshold density, the BAR domains undergo a concerted structural reorganization that brings them closer to the membrane, allows them to form an oligomeric scaffold, and lets their N-terminal helices insert more deeply into the membrane (Figure 2.4B). Our study is consistent with previous studies (Boucrot et al., 2012; Sorre et al., 2012) which indicate that a minimum threshold of amphiphysin density is required for extensive tubulation. That does not mean that continuously raising the protein density will only generate more tubes. In fact, Boucrot et al. find vesiculation to occur at very high protein densities of BAR proteins, and we made similar observations for α-synuclein-dependent membrane remodeling (Varkey et al., 2010, 2013). The different structures and mechanisms used to bind tubes and vesicles may also have implications for the regulation of membrane curvature in vivo. By preferentially stabilizing the vesicle or tube bound form of the protein it may be possible to guide membrane remodeling in vivo. Recent studies suggests that phosphorylation of endophilin at position S75 favors the vesicle bound form of this protein (Ambroso et al., 2014), and is a regulatory mechanism in synaptic endocytosis (Matta et al., 2012). Thus, the use of post-translational modifications may be an effective means for controlling different types of membrane curvature. Interestingly, amphiphysin has also been found to contain a phosphorylation site in its N-terminus (Hornbeck 65 et al., 2012). Future studies will have to show whether post-translational modifications are used to regulate which types of membrane curvature amphiphysin generates in vivo. 66 Figure 2.4 Model of how amphiphysin stabilizes different types of membrane curvature. (A) Upon contact with the membrane, the N-termini of amphiphysin are able to fold into amphipathic α-helices that embed (red is used for membrane exposed sites) into the membrane at the level of the lipid phosphates while the BAR domain remains distant (blue α-carbons represent solvent exposed spin sites). This state can either induce positive curvature by wedging apart lipid headgroups or sense already highly-curved membranes by stabilizing the existing packing defects in their outer leaflet. (B) At higher protein densities, amphiphysin oligomerizes, its BAR domain moves closer to the membrane and its helices insert more deeply into the membrane as tubes are formed. 67 2.4 Experimental Procedures 2.4.1 Generation of amphiphysin mutants and spin labeled derivatives. The plasmid containing His 6 -tagged N-BAR domain (a.a. 1–244) of Drosophila Amphiphysin (Peter et al., 2004) was kindly provided by Dr. Harvey McMahon (Medical Research Council). In order to allow specific labeling of cysteine residues, C66 and C82 were mutated to alanine by site- directed mutagenesis (QuikChange, Stratagene). Single cysteine mutants were then introduced and verified by sequencing. Proteins were expressed as previously described (Peter et al., 2004) in E. coli BL21 (DE3) and purified using nickel-nitrilo-triacetic acid–agarose, followed by superdex 200 gel filtration. Remaining impurities were removed using mono S cation exchange chromatography with a low salt buffer A (20 mM hepes pH 7.4, 1 mM dithiothreitol (DTT)) and elution buffer B (20 mM hepes pH 7.4, 2 M NaCl and 1 mM DTT). Protein concentrations were determined by absorbance at 280 nm using an extinction coefficient ε = 21860 M -1 cm -1 . Immediately prior to spin labeling, DTT was removed via size exclusion (PD-10 (GE)) columns equilibrated in 20 mM hepes, pH 7.4, 500 mM NaCl. Spin label (1-oxyl-2,2,5,5 tetramethyl-Δ3-pyrroline-3-methylmethanethiosulfonate) was incubated with protein in a 3 to 5 fold molar excess at room temperature for one hour, or alternatively, at 4° C overnight. Unreacted spin label was removed using PD-10 columns as described above. 2.4.2 Vesicle preparation and amphiphysin membrane interaction. The following synthetic lipids were used: 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE), 1-palmitoyl- 2-oleoyl-sn-glycero-3-phospho-L serine (POPS), 1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho- RAC-(1-glycerol)] (POPG), and cholesterol (Avanti Polar Lipids, Alabaster, AL). For tubulation experiments, large multi-lamellar vesicles were prepared by vortexing the dried lipid film 68 containing POPG/POPE (2:1, weight:weight) in buffer A. The various labeled and unlabeled forms of amphiphysin were incubated with vesicles at a protein-to-lipid ratio of 1:10 (weight:weight) wherein the protein was added last. In each case, tubulation was verified using negative-stain transmission electron microscopy. For binding small vesicles, extruded vesicles containing POPS and cholesterol were made by mixing POPS/Cholesterol (4:1, weight:weight) and vortexing the dried lipid in 20 mM hepes, pH 7.4, 150 mM NaCl. The vesicles were then treated to 10 cycles of freezing and thawing and then extruded using a mini-extruder with a 100 nm cutoff polycarbonate membrane (Avanti Polar Lipids, Alabaster, AL). For vesicle binding, protein and lipid were mixed at a 1:10 weight to weight ratio and incubated at room temperature for 20 minutes. Again, electron microscopy was used to assay sample homogeneity. Under these conditions, vesicles stayed intact and did not show any significant change in size. 2.4.3 Acquisition and analysis of EPR data. Continuous wave EPR spectra were recorded using a Bruker EMX spectrophotometer fitted either with an ER4119HS resonator or a Bruker dielectric resonator. The latter was also used for all power saturation experiments. The scan width for all EPR spectra is 100 Gauss. For EPR experiments, tube bound amphiphysin was harvested by centrifugation at 16,000 g in a micro centrifuge and vesicle bound amphiphysin at 120,000 g in an ultra centrifuge (Beckman Coulter Inc., Brea, CA). Pellets were taken up in Quartz capillaries (VitroComInc., New Jersey) for recording continuous wave EPR spectra and in TPX capillaries for accessibility measurements. Accessibilities to O 2 and NiEDDA (ΠO 2 and ΠNiEDDA) were measured by power saturation at room temperature (Altenbach et al., 1994). O 2 measurements were performed with samples equilibrated with air and NiEDDA measurements were performed using exogenously added NiEDDA to a final concentration of 10 mM. The membrane immersion depth for lipid-exposed residues was calculated from the depth parameter 69 Φ (Altenbach et al., 1994). Φ was calibrated for depth by doping the lipid mixtures mentioned above with 1% of 1-palmitoyl-2-DOXYL-stearoylsn-glycero-3-phosphocholine (Avanti PolarLipids, Alabaster, AL) spin-labeled at positions 5, 7, 10 and 12 on the acyl chains as well as the tempo labeled (1-palmitoyl-2-oleoyl-sn-glycero-3-phospho(tempo)choline) derivative containing a spin label in the headgroup region (Figure S2.1E and S2.1D). 2.4.4 Electron microscopy. Small aliquots (~10 μL) of tube or vesicle bound amphiphysin sample were incubated with carbon-coated formvar films mounted on copper grids (Electron Microscopy Services, Hatfield) for 5 minutes. Excess liquid was removed using filter paper and the grids were immediately stained with 1% uranyl acetate for 1 minute, rinsed thrice with 10 μL additional 1% uranyl acetate, and subsequently dried. A JEOL 1400 transmission electron microscope was used for specimen observation at 100 kV. 70 CHAPTER 2 REFERENCES Altenbach, C., Greenhalgh, D.A., Khorana, H.G., and Hubbell, W.L. (1994). A collision gradient method to determine the immersion depth of nitroxides in lipid bilayers: application to spin-labeled mutants of bacteriorhodopsin. Proc. Natl. Acad. Sci. U. S. A. 91, 1667– 1671. Ambroso, M.R., Hegde, B.G., and Langen, R. (2014). Endophilin A1 induces different membrane shapes using a conformational switch that is regulated by phosphorylation. Proc. Natl. Acad. Sci. 111, 6982–6987. Bhatia, V.K., Madsen, K.L., Bolinger, P.-Y., Kunding, A., Hedegård, P., Gether, U., and Stamou, D. (2009). Amphipathic motifs in BAR domains are essential for membrane curvature sensing. EMBO J. 28, 3303–3314. Blood, P.D., and Voth, G.A. (2006). Direct observation of Bin/amphiphysin/Rvs (BAR) domain- induced membrane curvature by means of molecular dynamics simulations. Proc. Natl. Acad. Sci. U. S. A. 103, 15068–15072. Blood, P.D., Swenson, R.D., and Voth, G.A. (2008). Factors influencing local membrane curvature induction by N-BAR domains as revealed by molecular dynamics simulations. Biophys. J. 95, 1866–1876. Böhm, J., Biancalana, V., Malfatti, E., Dondaine, N., Koch, C., Vasli, N., Kress, W., Strittmatter, M., Taratuto, A.L., Gonorazky, H., et al. (2014). Adult-onset autosomal dominant centronuclear myopathy due to BIN1 mutations. Brain J. Neurol. Boucrot, E., Pick, A., Çamdere, G., Liska, N., Evergren, E., McMahon, H.T., and Kozlov, M.M. (2012). Membrane fission is promoted by insertion of amphipathic helices and is restricted by crescent BAR domains. Cell 149, 124–136. Bretscher, L.E., Buchaklian, A.H., and Klug, C.S. (2008). Spin-labeled lipid A. Anal. Biochem. 382, 129–131. Butler, M.H., David, C., Ochoa, G.C., Freyberg, Z., Daniell, L., Grabs, D., Cremona, O., and De Camilli, P. (1997). Amphiphysin II (SH3P9; BIN1), a member of the amphiphysin/Rvs family, is concentrated in the cortical cytomatrix of axon initial segments and nodes of ranvier in brain and around T tubules in skeletal muscle. J. Cell Biol. 137, 1355–1367. Campelo, F., McMahon, H.T., and Kozlov, M.M. (2008). The hydrophobic insertion mechanism of membrane curvature generation by proteins. Biophys. J. 95, 2325–2339. 71 Campelo, F., Fabrikant, G., McMahon, H.T., and Kozlov, M.M. (2010). Modeling membrane shaping by proteins: focus on EHD2 and N-BAR domains. FEBS Lett. 584, 1830–1839. Claeys, K.G., Maisonobe, T., Böhm, J., Laporte, J., Hezode, M., Romero, N.B., Brochier, G., Bitoun, M., Carlier, R.Y., and Stojkovic, T. (2010). Phenotype of a Patient with Recessive Centronuclear Myopathy and a Novel Bin1 Mutation. Neurology 74, 519–521. Cui, H., Ayton, G.S., and Voth, G.A. (2009). Membrane binding by the endophilin N-BAR domain. Biophys. J. 97, 2746–2753. Cui, H., Mim, C., Vázquez, F.X., Lyman, E., Unger, V.M., and Voth, G.A. (2013). Understanding the role of amphipathic helices in N-BAR domain driven membrane remodeling. Biophys. J. 104, 404–411. Dalton, L.A., McIntyre, J.O., and Fleischer, S. (1987). Distance estimate of the active center of D-beta-hydroxybutyrate dehydrogenase from the membrane surface. Biochemistry 26, 2117–2130. Drin, G., and Antonny, B. (2010). Amphipathic helices and membrane curvature. FEBS Lett. 584, 1840–1847. Farsad, K., Ringstad, N., Takei, K., Floyd, S.R., Rose, K., and De Camilli, P. (2001). Generation of high curvature membranes mediated by direct endophilin bilayer interactions. J. Cell Biol. 155, 193–200. Frazier, A.A., Roller, C.R., Havelka, J.J., Hinderliter, A., and Cafiso, D.S. (2003). Membrane- Bound Orientation and Position of the Synaptotagmin I C2A Domain by Site-Directed Spin Labeling†. Biochemistry 42, 96–105. Frost, A., Perera, R., Roux, A., Spasov, K., Destaing, O., Egelman, E.H., De Camilli, P., and Unger, V.M. (2008). Structural basis of membrane invagination by F-BAR domains. Cell 132, 807–817. Gallop, J.L., Jao, C.C., Kent, H.M., Butler, P.J.G., Evans, P.R., Langen, R., and McMahon, H.T. (2006). Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J. 25, 2898–2910. Haselwandter, C.A., and Phillips, R. (2013). Connection between Oligomeric State and Gating Characteristics of Mechanosensitive Ion Channels. PLoS Comput Biol 9, e1003055. Haselwandter, C.A., and Wingreen, N.S. (2014). The role of membrane-mediated interactions in the assembly and architecture of chemoreceptor lattices. PLoS Comput. Biol. 10, e1003932. 72 Henne, W.M., Kent, H.M., Ford, M.G.J., Hegde, B.G., Daumke, O., Butler, P.J.G., Mittal, R., Langen, R., Evans, P.R., and McMahon, H.T. (2007). Structure and analysis of FCHo2 F- BAR domain: a dimerizing and membrane recruitment module that effects membrane curvature. Structure1993 15, 839–852. Hornbeck, P.V., Kornhauser, J.M., Tkachev, S., Zhang, B., Skrzypek, E., Murray, B., Latham, V., and Sullivan, M. (2012). PhosphoSitePlus: a comprehensive resource for investigating the structure and function of experimentally determined post-translational modifications in man and mouse. Nucleic Acids Res. 40, D261–D270. Hubbell, W.L., Gross, A., Langen, R., and Lietzow, M.A. (1998). Recent advances in site- directed spin labeling of proteins. Curr. Opin. Struct. Biol. 8, 649–656. Isas, J.M., Langen, R., Haigler, H.T., and Hubbell, W.L. (2002). Structure and Dynamics of a Helical Hairpin and Loop Region in Annexin 12: A Site-Directed Spin Labeling Study. Biochemistry 41, 1464–1473. Itoh, T., Erdmann, K.S., Roux, A., Habermann, B., Werner, H., and De Camilli, P. (2005). Dynamin and the actin cytoskeleton cooperatively regulate plasma membrane invagination by BAR and F-BAR proteins. Dev. Cell 9, 791–804. Jao, C.C., Der-Sarkissian, A., Chen, J., and Langen, R. (2004). Structure of membrane-bound α- synuclein studied by site-directed spin labeling. Proc. Natl. Acad. Sci. U. S. A. 101, 8331–8336. Jao, C.C., Hegde, B.G., Chen, J., Haworth, I.S., and Langen, R. (2008). Structure of membrane- bound alpha-synuclein from site-directed spin labeling and computational refinement. Proc. Natl. Acad. Sci. U. S. A. 105, 19666–19671. Jao, C.C., Hegde, B.G., Gallop, J.L., Hegde, P.B., McMahon, H.T., Haworth, I.S., and Langen, R. (2010). Roles of amphipathic helices and the bin/amphiphysin/rvs (BAR) domain of endophilin in membrane curvature generation. J. Biol. Chem. 285, 20164–20170. Kukulski, W., Schorb, M., Kaksonen, M., and Briggs, J.A.G. (2012). Plasma membrane reshaping during endocytosis is revealed by time-resolved electron tomography. Cell 150, 508–520. Lai, C.-L., Jao, C.C., Lyman, E., Gallop, J.L., Peter, B.J., McMahon, H.T., Langen, R., and Voth, G.A. (2012). Membrane binding and self-association of the epsin N-terminal homology domain. J. Mol. Biol. 423, 800–817. Langen, R., Oh, K.J., Cascio, D., and Hubbell, W.L. (2000). Crystal structures of spin labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra in terms of structure. Biochemistry 39, 8396–8405. 73 Lee, E., Marcucci, M., Daniell, L., Pypaert, M., Weisz, O.A., Ochoa, G.-C., Farsad, K., Wenk, M.R., and De Camilli, P. (2002). Amphiphysin 2 (Bin1) and T-tubule biogenesis in muscle. Science 297, 1193–1196. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., De Bock, P.-J., Morais, V.A., Vilain, S., Haddad, D., et al. (2012). LRRK2 controls an EndoA phosphorylation cycle in synaptic endocytosis. Neuron 75, 1008–1021. Meinecke, M., Boucrot, E., Camdere, G., Hon, W.-C., Mittal, R., and McMahon, H.T. (2013). Cooperative recruitment of dynamin and BIN/amphiphysin/Rvs (BAR) domain- containing proteins leads to GTP-dependent membrane scission. J. Biol. Chem. 288, 6651–6661. Mim, C., Cui, H., Gawronski-Salerno, J.A., Frost, A., Lyman, E., Voth, G.A., and Unger, V.M. (2012). Structural basis of membrane bending by the N-BAR protein endophilin. Cell 149, 137–145. Mizuno, N., Jao, C.C., Langen, R., and Steven, A.C. (2010). Multiple Modes of Endophilin- mediated Conversion of Lipid Vesicles into Coated Tubes. J. Biol. Chem. 285, 23351– 23358. Moravcevic, K., Alvarado, D., Schmitz, K.R., Kenniston, J.A., Mendrola, J.M., Ferguson, K.M., and Lemmon, M.A. (2015). Comparison of Saccharomyces cerevisiae F-BAR Domain Structures Reveals a Conserved Inositol Phosphate Binding Site. Structure 1993. Nicot, A.-S., Toussaint, A., Tosch, V., Kretz, C., Wallgren-Pettersson, C., Iwarsson, E., Kingston, H., Garnier, J.-M., Biancalana, V., Oldfors, A., et al. (2007). Mutations in amphiphysin 2 (BIN1) disrupt interaction with dynamin 2 and cause autosomal recessive centronuclear myopathy. Nat. Genet. 39, 1134–1139. Pang, X., Fan, J., Zhang, Y., Zhang, K., Gao, B., Ma, J., Li, J., Deng, Y., Zhou, Q., Egelman, E.H., et al. (2014). A PH Domain in ACAP1 Possesses Key Features of the BAR Domain in Promoting Membrane Curvature. Dev. Cell. Peter, B.J., Kent, H.M., Mills, I.G., Vallis, Y., Butler, P.J.G., Evans, P.R., and McMahon, H.T. (2004). BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 303, 495–499. Pykäläinen, A., Boczkowska, M., Zhao, H., Saarikangas, J., Rebowski, G., Jansen, M., Hakanen, J., Koskela, E.V., Peränen, J., Vihinen, H., et al. (2011). Pinkbar is an epithelial-specific BAR domain protein that generates planar membrane structures. Nat. Struct. Mol. Biol. 18, 902–907. 74 Rao, Y., and Haucke, V. (2011). Membrane shaping by the Bin/amphiphysin/Rvs (BAR) domain protein superfamily. Cell. Mol. Life Sci. 68, 3983–3993. Razzaq, A., Robinson, I.M., McMahon, H.T., Skepper, J.N., Su, Y., Zelhof, A.C., Jackson, A.P., Gay, N.J., and O‘Kane, C.J. (2001). Amphiphysin is necessary for organization of the excitation-contraction coupling machinery of muscles, but not for synaptic vesicle endocytosis in Drosophila. Genes Dev. 15, 2967–2979. Saarikangas, J., Zhao, H., Pykäläinen, A., Laurinmäki, P., Mattila, P.K., Kinnunen, P.K.J., Butcher, S.J., and Lappalainen, P. (2009). Molecular mechanisms of membrane deformation by I-BAR domain proteins. Curr. Biol. CB 19, 95–107. Shah, C., Hegde, B.G., Morén, B., Behrmann, E., Mielke, T., Moenke, G., Spahn, C.M.T., Lundmark, R., Daumke, O., and Langen, R. (2014). Structural Insights into Membrane Interaction and Caveolar Targeting of Dynamin-like EHD2. Structure 1993 22, 409–420. Shimada, A., Niwa, H., Tsujita, K., Suetsugu, S., Nitta, K., Hanawa-Suetsugu, K., Akasaka, R., Nishino, Y., Toyama, M., Chen, L., et al. (2007). Curved EFC/F-BAR-domain dimers are joined end to end into a filament for membrane invagination in endocytosis. Cell 129, 761–772. Sorre, B., Callan-Jones, A., Manzi, J., Goud, B., Prost, J., Bassereau, P., and Roux, A. (2012). Nature of curvature coupling of amphiphysin with membranes depends on its bound density. Proc. Natl. Acad. Sci. U. S. A. 109, 173–178. Takei, K., Slepnev, V.I., Haucke, V., and De Camilli, P. (1999). Functional partnership between amphiphysin and dynamin in clathrin-mediated endocytosis. Nat. Cell Biol. 1, 33–39. Varkey, J., Isas, J.M., Mizuno, N., Jensen, M.B., Bhatia, V.K., Jao, C.C., Petrlova, J., Voss, J.C., Stamou, D.G., Steven, A.C., et al. (2010). Membrane curvature induction and tubulation are common features of synucleins and apolipoproteins. J. Biol. Chem. 285, 32486– 32493. Varkey, J., Mizuno, N., Hegde, B.G., Cheng, N., Steven, A.C., and Langen, R. (2013). α- Synuclein oligomers with broken helical conformation form lipoprotein nanoparticles. J. Biol. Chem. 288, 17620–17630. Westphal, C.H., and Chandra, S.S. (2013). Monomeric synucleins generate membrane curvature. J. Biol. Chem. 288, 1829–1840. Wu, T., Shi, Z., and Baumgart, T. (2014). Mutations in BIN1 associated with centronuclear myopathy disrupt membrane remodeling by affecting protein density and oligomerization. PloS One 9, e93060. 75 Zhu, C., Das, S.L., and Baumgart, T. (2012). Nonlinear sorting, curvature generation, and crowding of endophilin N-BAR on tubular membranes. Biophys. J. 102, 1837–1845. 76 CHAPTER 3 Endophilin A1 induces different membrane shapes using a conformational switch that is regulated by phosphorylation Mark R. Ambroso 1 , Balachandra G. Hegde 2 , Ralf Langen 1, * 1 Department of Biochemistry and Molecular Biology, Zilkha Neurogenetic Institute, Keck School of Medicine, University of Southern California, Los Angeles, CA, 90033. 2 Present address: Post Graduate Department of Physics, Rani Channamma University, Vidyasangama, Belagavi - 591156 Karnataka India *Corresponding Author This work was originally published in the Proceeding of the National Academy of Sciences 111, 6982–6987, 13 May 2014 and formatted according to dissertation guidelines. Experimentation and manuscript writing was performed by the author. 77 CHAPTER 3 ABSTRACT Membrane remodeling is controlled by proteins that can promote the formation of highly-curved spherical or cylindrical membranes. How a protein induces these different types of membrane curvature and how cells regulate this process is still unclear. Endophilin A1 is a protein involved in generating endocytotic necks and vesicles during synaptic endocytosis and can transform large vesicles into lipid tubes or small and highly-curved vesicles in vitro. Using electron microscopy and electron paramagnetic resonance of endophilin A1, we find that tubes are formed by a close interaction with endophilin A1‘s BAR domain and deep insertion of its amphipathic helices. In contrast, vesicles are predominantly stabilized by the shallow insertion of the amphipathic helical wedges with the BAR domain removed from the membrane. By showing that the mechanism of membrane curvature induction is different for vesiculation and tubulation, these data also explain why previous studies arrived at different conclusions with respect to the importance of scaffolding and wedging in the membrane curvature generation of BAR proteins. The Parkinson‘s disease-associated kinase LRRK2 phosphorylates S75 of endophilin A1, a position located in the acyl chain region on tubes and the aqueous environment on vesicles. We find that the phosphomimetic mutation S75D favors vesicle formation by inhibiting this conformational switch, acting to regulate endophilin A1-mediated curvature. As endophilin A1 is part of a protein superfamily, we expect these mechanisms and their regulation by post-translational modifications to be a general means for controlling different types of membrane curvature in a wide range of processes in vivo. 78 3.1 Introduction Numerous cellular remodeling events are controlled by proteins that can regulate membrane shape (Rao and Haucke, 2011). In the example of synaptic endocytosis, proteins must engender invagination, drive pit formation, stabilize neck structures, and ultimately cause fission. These steps are executed through spatial and temporal application of membrane-altering proteins that contribute a range of curvatures (McMahon and Gallop, 2005, Qualmann et al., 2011). Moreover, misregulated expression or post-translational modification of these proteins is implicated in a number of diseases (Bergmann et al., 2003; Matta et al., 2012; Nicot et al., 2007). An increasing body of evidence suggests that endophilin plays an essential role in synaptic endocytosis by recruiting cofactors such as dynamin (Meinecke et al., 2013; Ringstad et al., 1999; Sundborger et al., 2011) and synaptojanin (Milosevic et al., 2011; Schuske et al., 2003; Soda et al., 2012; Verstreken et al., 2003) as well as by inducing membrane curvature (Bai et al., 2010; Farsad et al., 2001; Gallop et al., 2006; Masuda et al., 2006). It has been observed to localize to the synaptic vesicle pool and endocytotic neck regions in vivo and generate small highly-curved vesicles and lipid tubes from large vesicles in vitro (Ferguson et al., 2009; Meinecke et al., 2013; Milosevic et al., 2011; Sundborger et al., 2011). It has thus been proposed that a main component of endophilin‘s function in vivo is its ability to regulate membrane curvature. In general, proteins generate curvature through several mechanisms; by forcing membranes to conform to their own intrinsic protein-shape (scaffolding) (Qualmann et al., 2011), inserting amphipathic segments into the lipid bilayer and thus generating a wedging force (Drin and Antonny, 2010), protein crowding (Stachowiak et al., 2012), and by a less considered mechanism of space-filling where protein insertions alleviate packing differences between the 79 inner and outer leaflet of the membrane (bilayer couple) (Farsad and Camilli, 2003). It is still unclear whether endophilin A1 utilizes a combination of mechanisms to generate the various types of membrane curvature or whether different mechanisms are used to cause vesiculation or tubulation. The structure of endophilin A1 was first elucidated through crystallography and found to contain a crescent-shaped BIN/amphiphysin/Rvs (BAR) domain (Gallop et al., 2006; Masuda et al., 2006; Weissenhorn, 2005) (Figure 3.1A, pdb 2C08). Mutational analyses of endophilin A1 have discovered that the loss of positively charged amino acids on the concave surface of the BAR domain inhibits the ability of endophilin A1 to curve lipid membranes (Gallop et al., 2006; Peter et al., 2004). The natural shape of the BAR domain implies that endophilin binds membranes and induces its intrinsic curvature through a scaffolding mechanism (Arkhipov et al., 2009; Blood and Voth, 2006; Blood et al., 2008). Surprisingly, we found that when bound to highly-curved vesicles (similar in size to synaptic vesicles), endophilin A1‘s BAR domain resides at a significant distance from the membrane (Jao et al., 2010), where the scaffolding mechanism is not likely to be very strong. Instead, we found that the main contacts were made by the N-terminal (H0) (residues 1-20) and insert region (residues 60-87) helices of endophilin A1, which are only partially resolved in the crystal structure (Gallop et al., 2006; Jao et al., 2010). These regions form amphipathic α-helices which embed in the membrane at the level of the lipid phosphates. The locations of the helices are optimized for curvature sensing (Bhatia et al., 2009) and inducing (Campelo et al., 2008; Mim and Unger, 2012) through the wedging mechanism. In fact, helices alone can be sufficient to bend membranes, as shown by studies of α- synuclein (Jao et al., 2008; Varkey et al., 2010; Westphal and Chandra, 2013). 80 As of now, high-resolution structural data of endophilin bound to tubes has been limited. Studies using cryo-electron microscopy (cryoEM) in conjunction with computational modeling have shown that BAR domains take up specific high-density oligomeric states that appear to stabilize tube structures by acting as scaffolds (Cui et al., 2013; Mim et al., 2012; Mizuno et al., 2010). However, scaffolding would likely be ineffective if tube-bound endophilin had the same remote membrane binding as observed on vesicles. For these reasons, the main goal of this study was to investigate whether endophilin A1 uses different structures and mechanisms in generating tubes than in forming vesicles. We hypothesize that defining these two states will provide insight into not only how one protein can generate multiple types of membrane curvature but how they might be regulated in the cell. We find that in comparison to the structure on liposomes (Gallop et al., 2006; Jao et al., 2010), tube-bound endophilin shifts the concave surface of its BAR domain close to the membrane, and the H0 and insert regions, concertedly, submerge below the lipid phosphates and into the acyl chains. These data suggest a more pronounced role of scaffolding. The concerted downward movement of the helices appears to be important for tubulation. We noticed that S75, a phosphorylation site targeted by Parkinson‘s disease (PD)-associated LRRK2 kinase (Matta et al., 2012), is shallowly inserted on vesicles but deeply on tubes. We hypothesized that the introduction of a negative charge would create a large energetic cost for deeply inserting S75 in the tube conformation. In fact, we found phosphomimetic mutation S75D to destabilize tubulation through favoring shallow insertion of the insert helices and not by a decrease in membrane association. It is possible that LRRK2-mediated phosphorylation of S75 could act as an important regulator for how endophilin structurally interacts with and curves membranes. As mutations that constitutively activate LRRK2 represent the most common form of inherited PD, 81 determining how phosphorylation alters the structure and function of endophilin A1 is of significant importance. 3.2 Results 3.2.1 On tubes, N-terminal α-helices insert deeply into the acyl chain region of the lipid membrane. In order to investigate the structure of tube-bound endophilin A1, we first optimized conditions for tubulation. According to transmission electron microscopy (Figure 3.1B-J), we were able to generate homogeneous tubes with an average diameter of 35 nm and a typical length of 10 μm that were stable for at least twenty-four hours. Then, we used these conditions in site- directed spin labeling studies aimed at determining the local structure and location of the H0 when endophilin A1 is bound to tubes. Endophilin A1 was spin labeled at select sites in its H0, one at a time, and these derivatives were confirmed to tubulate vesicles under optimized conditions. Continuous wave (CW) electron paramagnetic resonance (EPR) spectra of these spin labeled positions (R1 denotes presence of a spin label) indicate an ordering of the H0 upon tube formation (Figure 3.2A). Thus H0, which is unfolded in solution, becomes structured on lipid tubes. A spin label‘s depth in the lipid bilayer can be detected using a collision-gradient method which measures the relative accessibility of the label to lipophilic oxygen and lipophobic NiEDDA (Altenbach et al., 1994). To determine the immersion depth of H0 in a tube-bound state, we measured the accessibility of the spin labeled derivatives to O 2 (ΠO 2 ) and NiEDDA (ΠNiEDDA) and calculated the depth parameter, Ф (Ф=ln(ΠO 2 /ΠNiEDDA)) (Altenbach et al., 1994). Increasing depth of a spin label in the lipid bilayer will cause exposure to high levels of oxygen and minimal levels of NiEDDA, resulting in an increased Ф value, and alternatively, 82 solvent exposed spin labels will give low to negative Ф values. Positive Ф values were observed for sites on the hydrophobic face and low to negative Ф values for the sites on the hydrophilic face of the H0 helices (Figure 3.2B). After calibration (Figure S3.1), we converted Ф values into depth and found that labeled sites lining the hydrophobic face of the H0 helices penetrate deeper into tubes than into vesicles (Figure 3.2C). The average immersion depth of these sites was 18 Å. Inasmuch as the nitroxide moiety is typically 7-10 Å from the center of the helix (Langen et al., 2000), we can estimate that the depth of the H0 helix is on the order of 8-11 Å below the lipid phosphates (Figure 3.2D). The H0 region was previously shown to be necessary for stable oligomerization of endophilin on tubes and cross linking data suggested that this could be mediated by an interaction between neighboring H0 from different dimers (Mim et al., 2012). To examine whether this stabilization was mediated by a direct contact between neighboring H0 helices, residues lining either side of the amphipathic H0 helix were selected and examined for the presence of spin-coupling while bound to tubes (Figure 3.2E). No presence of spin-coupling or line-shape changes upon dilution of these spin labeled mutants with cysless analogs was observed, suggesting that the labeled sites on the helices do not stably reside within 20 Å of each other. We therefore used double electron-electron resonance (DEER) which has a significantly longer range (up to 60-70 Å). Intermolecular distances were measured using endophilin labeled at position 7 or 12, which are close to the center of the H0 helix. We found broad distance distributions with peaks around 30 to 40 Å suggesting that neighboring H0 helices may not have one unique orientation to each other and are unlikely to be in direct contact (Figure 3.2F and G). 83 Figure 3.1. Endophilin A1-induced tubulation. (A) The crystal structure of rat endophilin A1 (pdb 2C08) dimer (subunits are colored yellow or blue). The N-termini (H0) and insert region are schematically illustrated as cylinders to indicate their ability to become helical upon membrane binding (Gallop et al., 2006; Jao et al., 2010; Mim et al., 2012). Lipid tubes generated from large vesicles incubated with spin labeled (R1) endophilin A1 derivatives are visualized by negative stain transmission electron microscopy. Representative examples are shown for (B-D) N- terminal derivatives; 5R1, 6R1, and 13R1 respectively. (E-G) Insert region derivatives; 70R1, 71R1, and 77R1 respectively. (H-J) BAR domain derivatives; 108R1, 159R1, and 247R1 respectively. All samples were screened for thorough tubulation prior to further experimentation. Scale bars are set to 500 nm. 84 Figure 3.2. N-terminal helices penetrate deeply into the membrane. (A) Continuous wave (CW) EPR spectra of spin labeled (R1) endophilin A1 derivatives bound to tubes (black). The EPR spectrum of position 4 in solution (red) is representative of other N-terminal sites and is shown at half amplitude. (B) Helical wheel depiction of H0 showing a hydrophilic (purple) and hydrophobic (yellow) face. Measured Ф values are shown for select residues. (C) Ф values were converted into immersion depths for sites on the hydrophobic face of H0 either on tubes (black) or in a previously elucidated vesicle-bound (grey) state (Gallop et al., 2006). Depths represent the location of the nitroxide label which is typically 7-10 Å from the center of the α-helix (Langen et al., 2000). (D) A schematic model of the H0 helices on small vesicles (grey helix) or tubes (colored helix as in B) relative to the lipid headgroups (grey) and acyl chains (light grey). The helices were manually placed according to the observed average immersion depth of lipid exposed sites, taking into account the length of the nitroxide side chain. (E) CW EPR spectra of select endophilin A1 derivatives on tubes either fully labeled (black) or mixed with 3 fold excess of unlabeled protein (red). The overlay of the respective spectra indicates the absence of significant spin-spin interactions. (F and G) Baseline subtracted time-evolution data (black, left) from DEER experiments of the indicated tube-bound endophilin derivatives were subjected to Tikhonov regularization (red) resulting in the shown distance distributions (right). Error bars represent SD, n = three independent experiments, Student t-test: *p<0.005. 85 Figure 3.2, continued 86 Figure S3.1. Depth calibration using spin labeled lipids. Accessibility to O 2 and NiEDDA of spin labeled lipids under endophilin A1-induced tubulating conditions (see methods) are summarized by Ф. Doxyl-labels (black squares) were used for positions 5, 7, 10, and 12 along the acyl chain. TempoPC was used to introduce a label in the headgroup region (substituting an N-methyl group on the choline headgroup). Spin label immersion depths were taken from previous studies (Altenbach et al., 1994; Bretscher et al., 2008; Dalton et al., 1987). The points were fit using the following equation, Å= 7.863Ф – 2.829 (R 2 = 0.992). A generic phospholipid structure is plotted alongside the plot for scale. Positive values indicate deeper immersion depths. 3.2.2 On tubes, insert region inserts deeply into acyl chain region and BAR domain contacts the lipid headgroups. Next, we investigated whether endophilin A1 regions other than H0 differently interact with tubes and vesicles. Our prior studies on vesicle-bound endophilin A1 revealed that residues 63-75 of the insert region take up an α-helical conformation (Jao et al., 2010). In the dimer these 87 helices are antiparallel to one another and largely perpendicular to the long axis of the BAR domain (Figure 3.1A). Thus, we first investigated whether the insert helices retain this orientation upon tubulation. DEER distances between singly-labeled mutants in the insert region were similar albeit slightly longer than those previously reported for vesicle-bound protein (Figure 3.3A-C and Figure S3.2A). Thus, the anti-parallel and α-helical structure is retained under tubulating conditions. Moreover, CW EPR spectra of spin labeled mutants 63-79 showed signs of ordering comparable to that seen for the same region bound to vesicles (Figure S3.2B and C). To determine whether the insert helices immerse themselves more deeply into the bilayer of tubes, accessibility-based depth measurements were performed by replacing each side chain with R1, one amino acid at a time. The Ф values display a periodicity indicative of an α-helix and fit well onto a helical wheel; high Ф values fall onto one surface (hydrophobic face, yellow) and low Ф values onto the opposite surface (hydrophilic face, purple) (Figure 3.3D and E). After calibration, the depth of the spin labeled sites suggest a helical axis 5-8 Å deep in the membrane (Figure 3.3F). Residues 76-79 retain a periodicity consistent with an α-helical structure, but the variation in depth is no longer as pronounced (Figure S3.3). These residues are likely to be in a transition region between a fully helical and a loop structure and may represent an additional helix-like structure not present on vesicles. The constant immersion depth of the membrane facing residues also indicates that the insert helices are parallel to the membrane and not upward- tilted as on vesicles (Figure 3.3G). Having established that the insert regions as well as the H0 helices move deeper into the bilayer, we wanted to investigate whether the BAR domain might also undergo similar movements. According to the crystal structure, residues 63-67 of the insert region are directly adjoined to the BAR domain suggesting that the movement of the insert helices might be directly 88 coupled to that of the BAR domain. To test this notion, spin labeled sites previously used for vesicle-bound endophilin (Jao et al., 2010) were subjected to accessibility measurements (Figure 3.3H and I). On tubes, spin labeled residues located on the concave surface, 159 and 166, showed increased Ф values while all sites on the convex surfaces retained highly negative Ф values and lie outside our depth calibration, and are therefore compared by Ф. However, spin labeled residues 159 and 166 lie inside our calibration range with values located approximately 4-6 Å above the level of the lipid phosphates. Thus, tubulation causes a concerted movement of the BAR domain together with the H0 and insert region towards the membrane (Figure 3.3J). Figure 3.3. Concerted movement of BAR domain and insert region toward the membrane. (A) Baseline subtracted time-evolution data (black) from a DEER experiment of tube-bound 64R1 subjected to Tikhonov regularization (red) with (B) resulting distance distribution. (C) A comparison between intra-dimer distances on vesicles (Jao et al., 2010) and on tubes. (D) Local Ф maxima (yellow) and minima (purple) fall onto a hydrophobic or hydrophilic face of a helical wheel. (E) Ф as function of labeling position on tubes (solid, colored as in D) and on vesicles (Jao et al., 2010) (dashed). (F) Immersion depth after calibration for sites on the hydrophobic face of the insert region on vesicles (Jao et al., 2010) (grey) and on tubes (black). (G) A schematic model of the insert region on vesicles (grey helix) and tubes (colored as in D) relative to the lipid headgroups (dark grey, negative depth values) and the acyl chains (light grey, positive depth values). The helices were manually placed as described in Figure 3.2. The phosphorylation site S75 moves from the acyl chain environment to the aqueous environment (illustrated phosphate groups). (H) Crystal structure of rat endophilin A1 dimer (pdb 2C08) showing the locations of spin labeled sites. (I) Bar graph comparing Ф values measured for sites 89 on the concave and convex surfaces of the BAR domain when bound to tubes (black) or vesicles (grey). (J) Schematic illustration of the location of the BAR domain relative to the bilayer when bound to vesicles (left) or tubes (right). Error bars represent SD, n represents at least three independent experiments, Student t-test: *p<0.01. Figure S3.2. CW EPR and DEER measurements of endophilin A1’s insert region. (A) CW EPR spectra of rat endophilin A1 spin labeled (R1) at the indicated positions bound to tubes (black) or small and highly curved vesicles (red). Similar but not identical spectra were obtained Figure 3.3, continued 90 in both cases. Scan width is 100 gauss. (B) CW EPR spectra of rat endophilin A1 derivatives spin labeled at the indicated positions in solution (red) and bound to tubes (black). The EPR spectra of certain indicated derivatives in solution (red) are shown at half amplitude for space considerations. Scan width is 150 gauss. (C) Baseline subtracted time-evolution data (left panels, black) from DEER experiments of spin labeled endophilin A1 derivatives bound to tubes. Labeled derivatives were diluted with nonparamagnetically labeled derivatives to reduce intermolecular signals. Data were subjected to Tikhonov regularization (left panels, red), resulting in the corresponding distance distribution (right panels). Figure S3.2, continued 91 Figure S3.3. O 2 and NiEDDA accessibilities for insert region residues. Accessibility to paramagnetic colliders O 2 (ΠO 2 ) and NiEDDA (ΠNiEDDA) of the indicated spin labeled sites from tube-bound endophilin A1. A pronounced out-of-phase periodicity is observed which is typically observed for membrane exposed amphipathic α-helices (Hubbell et al., 1998). Figure S3.3, continued 92 3.2.3 Phosphomimetic mutation S75D destabilizes tubes by preventing deep insertion of the insert helix. It is still unclear how endophilin A1 transitions between tubulation and vesiculation in vivo. A recent report elucidated a direct link between mutations in LRRK2 kinase and the increased phosphorylation of endophilin A1 at position S75 (Matta et al., 2012). Our structural data show that this position is located in a region that takes up a very different location with respect to the membrane when on tubes or vesicles (Figure 3.3G). While the S75 side chain was determined to reach above the lipid headgroups and into the aqueous environment on vesicles (Jao et al., 2010), it is deeply embedded in the acyl chain region upon tubulation. Given that the S75 side chain faces straight out of the membrane and approximating a phosphoserine side chain to be 4 Å long, one could estimate the negative charge to be around the lipid phosphates on tubes but well above the lipid headgroups on vesicles. We therefore hypothesized that S75 phosphorylation might preferentially destabilize the tube-bound conformation and favor the vesicle-bound structure. To test this, we first aimed to verify whether phosphorylation affects the overall membrane association of endophilin in vitro. Using a phosphomimetic mutation S75D in combination with spin labeled derivative 74R1, we titrated both S75-74R1 and S75D-74R1 with lipid vesicles (Figure 3.4A and B) and monitored binding via amplitude changes in EPR CW spectra (Figure 3.4C). No difference in binding saturation was observed using our optimized lipid composition, although a noticeable albeit small (p>0.1) decrease was observed for binding to a previously used lipid system (Matta et al., 2012). To determine whether the phosphomimetic mutation affects the structure of endophilin A1, tube-bound S75D-74R1 and S75D-63R1 mutants were investigated by CW EPR. These labeled positions were selected to measure positions near both ends of the helix. Moreover, spin-spin interactions in the spectra of 63R1 are a useful 93 determinant of endophilin dimerization (Jao et al., 2010). In comparison to their respective S75 counterparts, S75D-63R1 and S75D-74R1 had comparable CW EPR spectra but significantly reduced Ф values (Figure 3.4D and E). These data suggest that the mutants remain dimerized, membrane associated, and in a similar local structure but at decreased depths within the bilayer. We then investigated whether the modification could affect endophilin‘s ability to form tubes in our optimized tubulation system. While the spin labeled derivatives containing S75 (wt) produced tubes that were stable for days to weeks, S75D-63R1 and S75D-74R1 generated more vesiculation (Figure 3.4F-I). Similarly, comparison of unlabeled wt and S75D endophilin A1 constructs revealed that phosphomimetic mutants produce tubes with decreased stability and increased amounts of small vesicles (Figure 3.4J and K). In summary, the phosphomimetic modification at position S75 inhibits stable tubulation by shifting the insert region to shallower depths within the lipid bilayer. Figure 3.4. Phosphomimetic S75D mutation destabilizes tubes by reducing membrane immersion depth of the insert region. CW EPR spectra of S75D-74R1 incubated with 0 mM (black), 2.5 mM (purple), and 10 mM (teal) of (A) vesicles composed of a 5:2:1:1 molar ratio of L-α-phosphatidylcholine, L-α-phosphatidylethanolamine, L-α-phosphatidylserine, and cholesterol (Matta et al., 2012) or (B) vesicles composed of DOPG:DOPE (2:1). (C) Spectral amplitudes for S75-74R1 (solid lines) and S75D-74R1 (dashed lines) from the experiments in A (black lines) or B (blue lines) are plotted as function of lipid concentration. (D) CW EPR spectra of tube-bound endophilin A1 labeled at positions 63 or 74 with (black) or without (red) the S75D mutation. All spectra are normalized to the same number of spins. (E) Ф values of 63R1 and 74R1 with (black) and without (grey) the S75D mutation incubated with DOPG:DOPE (2:1). 94 Negative stain electron microscopy shows S75-63R1 forming stable tubes after twenty-four hours (F) while S75D-63R1 produces mainly vesicular structures (G). Similar results were obtained for S75-74R1 (H) and S75D-74R1 (I) as well as S75 (J) and S75D (K) in a wt background. Scale bars are 0.5 μm. Error bars represent SD, n represents at least three independent experiments, Student t-test: *p<0.005. 3.3 Discussion When compared to its vesicle-bound form (Gallop et al., 2006; Jao et al., 2010), different regions of tube-bound endophilin undergo a concerted structural reorganization that brings the BAR domain closer to the membrane and inserts the amphipathic helices more deeply into the acyl chain region. This orientation appears to be important as the introduction of a phosphomimetic negative charge at residue 75 destabilizes the more deeply embedded tube- bound structure in favor of other mainly vesicular forms, potentially representing a regulatory mechanism responsible for guiding endophilin A1‘s curvature state. The closer membrane proximity and direct contact of the BAR domain on tubes suggests a much more pronounced scaffolding effect. This present data together with our previous study on vesicle-bound endophilin A1 suggests that there are two very different mechanisms by which vesicles or tubes are generated (Figure 3.5A and B). While the membrane curvature of vesicles is Figure 3.4, continued 95 predominantly stabilized by the insertion of endophilin‘s helical wedges, tubes are stabilized through a combination of scaffolding and helix insertion. Whereas the present study revealed two structural states, previous studies on endophilin A1 and the related amphiphysin identified two distinct functional modes of membrane interaction (Sorre et al., 2012; Zhu et al., 2012). Under high protein density conditions, both N-BAR proteins behaved as potent inducers of membrane curvature and generated tubes similar in size to those observed here. In contrast, low protein density conditions had a smaller effect on membrane curvature induction. In agreement with cryoEM data (Mim et al., 2012; Mizuno et al., 2010), which showed a highly oligomeric protein coat around lipid tubes, it was suggested that increasing protein concentrations promoted the assembly of the protein scaffold (Sorre et al., 2012; Zhu et al., 2012). The differential modes of membrane binding are consistent with the observed structural rearrangements in tube-bound endophilin which, as we find, do not only polymerize the BAR domains but also bring them closer to the membrane. Another recent study investigated the binding of endophilin and amphiphysin to intact vesicles of varying diameters. Interestingly, the curvature sensitivity solely relied on the amphipathic helices and not the BAR domain (Bhatia et al., 2009). This result would have been difficult to rationalize with a model in which the BAR domain is in close contact with the membrane. Rather, this finding is consistent with a model in which the BAR domain only moves close to the membrane when it engages in scaffolding and oligomerizing on tubes. While it is clear that the H0 helices play a critical role in oligomerization (Mim et al., 2012), the underlying mechanisms are still unknown. Our data are inconsistent with a model in which neighboring helices are directly aligned and in physical contact with each other. However, we cannot exclude models in which the neighboring helices are severely staggered or in which 96 helices jointly coordinate a lipid bridge between them. The latter may be more likely since the sides of the helices are lined with positive charges. This would cause repulsion in the case of direct interaction between the helices but would favor association with negatively charged lipids. Considering the shallow immersion depth typically observed by EPR for most membrane-bound amphipathic helices (Drin and Antonny, 2010; Drin et al., 2007; Jao et al., 2008, 2010; Lai et al., 2012), it was surprising to discover the deeply inserted amphipathic helices on tubes. This movement brings the C α of lysine residues deep into the acyl chain region (~7-10 Å). Although the membrane environment is an unfavorable location for lysines, it is likely that such side chains snorkel (Strandberg and Killian, 2003) in order to form a salt bridge with lipid phosphates. Given that an extended lysine side chain is 6.5 Å long and amine- phosphate bonds to be approximately 3 Å, the submerged lysines are within range to snorkel to the lipid phosphates. This interaction could be facilitated further by a movement of the phosphate toward the lysine side chain. The deeper insertion of helical wedges is also likely to have some functional consequences. The downward movement of the helices reduces their ability to push the headgroups apart, thereby decreasing the spontaneous curvature effects (Campelo et al., 2010) (Figure 3.5C). A prior study, which combined mutagenesis and computational methods, found that strong spontaneous curvature contributions of helical wedges at high densities promoted vesiculation while enhanced scaffolding favored tubulation (Boucrot et al., 2012). According to this notion, the reduced spontaneous curvature of more deeply inserted helices may therefore be beneficial to tube stability and reduce vesiculation. In addition, the increased helix immersion depths on tubes also promote acyl chain separation (Figure 3.5C). By taking up additional space in the outer leaflet, the helices may compensate for the increasing imbalance between the surface areas of the inner- and outer leaflet 97 during tubulation. Such a bilayer couple-like mechanism could be especially important in membranes of low cholesterol where there is little flip-flop (Campelo et al., 2008). Interestingly, vesicle fusion events result in transleaflet flip flop (Kato et al., 2002; Lee et al., 2000). If similar lipid redistribution occurs during the reverse vesiculation process, the need for space filling may not be as important. Regardless of the exact reasons, our study on S75 phosphorylation suggests that helix depth can be used to regulate membrane curvature. As endophilin A1 is a member of a populous protein family with similar domains and functions, it is possible that the mechanisms we find here are applicable to a wide range of proteins and membrane processes. Importantly, the ability to switch between the respective structural states might be an effective means for regulating the type of membrane curvature is induced in vivo. One such regulatory mechanism could be phosphorylation. The S75 phosphorylation site is located in an aqueous environment on vesicles while it is just below the phosphate level on tubes, suggesting that phosphorylation would be more destabilizing for the latter (Figure 3.3G). Indeed, S75D mutants decreased tube stability and enhanced vesiculation. The addition of negative charges near the C-terminus of the insert region (K76E, R78E) has also been shown in previous studies to destabilize tubes (Gallop et al., 2006). Thus, our data provide a structural basis for how the addition of a negative charge to amphipathic domains can act as a molecular switch to toggle between the formations of tubes and other curved structures. Since negative charges can be easily added by phosphorylation, the modification could be used to control different types of membrane curvature in vivo. Interestingly, other post-translational modification sites are located within regions found to undergo substantial movements in the present study (Kaneko et al., 2005). One of these sites is located in the H0 helix of endophilin A1 and has been shown to impact receptor internalization. There is also evidence of a 98 phosphorylation site in the N-termini of amphiphysin, suggesting that other BAR proteins or membrane-curving proteins may be regulated by phosphorylation. It may, therefore, be possible that the mechanism identified for S75 phosphorylation may be more generally applicable. The ability to regulate between tubulation and vesiculation may be important for endophilin‘s function at the endocytic neck which is transiently generated and ultimately destabilized. In fact, the inability to toggle between the phosphorylated and unphosphorylated versions of endophilin has been shown to inhibit endocytosis (Matta et al., 2012). It remains to be tested whether misregulation of this process has a direct impact in PD pathogenesis. Figure 3.5. Schematic illustration of endophilin A1 tube and vesicle binding and its modulation by phosphorylation. (A) The incubation of endophilin A1 and large lipid vesicles can result in vesiculation or tubulation. (B) On small vesicles (left), endophilin A1 predominantly uses its amphipathic helices (red pentagons) rather than the BAR domain for membrane binding. The location of the helices is optimized for stabilizing membrane curvature by wedging into the headgroup region (dark grey) and thereby generating a splitting force between neighboring lipids (C, blue lines). Endophilin A1 binds tubes (A, right) in a highly oligomeric, anisotropic manner (Mim et al., 2012; Mizuno et al., 2010) and moves its amphipathic helices deeper into the acyl chains (B, right), filling more space within the acyl chain region (light grey) which more optimally pushes entire lipids apart (C, right). The difference in lipid area the helices take up in the membrane is greater for tubes (d t ) than for vesicles (d v ). Simultaneously the BAR domain moves into contact with the lipid headgroups (B, right). For simplicity only the outer leaflet of the membrane is shown in B. LRRK2 phosphorylation introduces a negative charge at S75, a site 99 submerged in the acyl chain region of tubes. This modification destabilizes tubes and favors vesicles as illustrated in part A. Figure 3.5, continued 100 3.4 Experimental procedures 3.4.1 Rat endophilin A1 protein purification and spin labeling. Cysteine and S75D mutations were introduced using a QuikChange Site-Directed Mutagenesis kit (Stratagene, La Jolla). All DNA constructs were examined for accuracy by DNA sequencing. Recombinant rat endophilin A1 was expressed and purified as previously described (Gallop et al., 2006; Jao et al., 2010). In short, proteins were expressed in BL21 cells for 16 hours at 18 °C. Lysed cells were centrifuged at 40,000 g for 20 minutes at 4 °C, and the supernatant was bound to glutathione beads for 1 hour at 4 °C. The beads were washed with an excess of 20 mM HEPES-HCl, pH 7.4, 150 mM NaCl, 1 mM DTT, 1 mM EDTA. The GST tag was cleaved by incubation with thrombin protease (Sigma Aldrich) for 16 hours at 4 °C. Cleaved proteins were further purified by anion exchange chromatography and Superdex 200 gel filtration. Purified protein was reacted with 8x molar excess of the spin label MTSL (1-oxy-2,2,5,5-tetramethyl-d-pyrroline-3-methyl)- methanethiosulfonate. Excess spin label was removed using PD10 columns (GE Healthcare). Protein concentrations were quantified using UV absorbance at 280 nm and ε=5,240. 3.4.2 Lipid and tube sample preparation. Optimized conditions for tube formation were found using lipid vesicles composed of 1,2-Dioleoyl-sn-Glycero-3-[Phospho-rac-(1-glycerol)](Sodium Salt) (DOPG) and 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE) combined in a 2:1 mass ratio respectively (Avanti Polar Lipids, Alabaster, Alabama). The chloroform mixture was dried by a constant stream of N 2 and desiccated overnight. The lipid film was then resuspended into buffer (20 mM Hepes, pH 7.4) to a final concentration of 4 mg/mL. Each sample was prepared with 100 μg of lipid mixture and 50 μg of spin labeled protein in approximately 50 μl total volume (1:70 protein:lipid molar ratio) and allowed to react for 20 minutes at room temperature. Dilution of spin labeled endophilin A1 to 25% during experiments that tested for 101 spin-spin coupling was done through the addition of 3 fold excess wild type endophilin A1 with its single cysteine mutated to serine (C108S) and 3 fold excess lipid mixture. This mutation does not alter the ability of endophilin to tubulate vesicles as evidence by negative-stain transmission electron microscopy (Gallop et al., 2006). The tube-bound and unbound proteins were separated by centrifugation at 16,000 g for 15 minutes at 37 °C, removing the supernatant, and resuspending the pellet with 6 μL of buffer (20 mM Hepes, pH 7.4, 150 mM NaCl). Tube formation was confirmed for each sample through negative-stain electron microscopy. Aliquots (4 μL) of each sample were loaded onto a Formvar Copper Film (Electron Microscopy Sciences, Hatfield, PA), stained with 1% uranyl acetate for 60-90 seconds, and visualized using a Jeol JEM-1400 transmission electron microscope. Wt and S75D mutant samples were done side by side using the conditions described above. 3.4.3 Titration assay for detecting membrane binding. Lipid titration assays were either performed using the conditions described above or with a previously used lipid composition, 5:2:1:1 molar ratio of L-α-phosphatidylcholine, L-α-phosphatidylethanolamine, L-α- phosphatidylserine, and cholesterol (Matta et al., 2012). The latter composition was rehydrated in buffer (20 mM Hepes, pH 7.4, 150 mM NaCl), made into small unilamellar vesicles using a tip sonicator, and centrifuged for 10 minutes at 13,200 rpm to remove metallic debris. For both lipid compositions, increasing amounts of lipid were incubated with 20 μM spin labeled endophilin derivatives and monitored by continuous wave electron paramagnetic resonance. All samples were measured using the same number of scans on a Bruker EMX spectrometer fitted with a Bruker ER4119HS resonator (12.7 mW power), and amplitudes were obtained from the signal measured by WinAcquisit (Bruker Biospin). 102 3.4.4 Electron paramagnetic resonance continuous wave and power saturation measurements. EPR spectra as well as power saturation experiments were obtained using a Bruker EMX spectrometer fitted with a dielectric resonator. Spectra were obtained at 1.85 mW incident microwave power. Power saturation measurements were typically performed using a range of 1.26 to 63.46 mW incident microwave power. In the case of high NiEDDA accessibilities experiments were performed using a range of 1.26 to 100.58 mW incident microwave power. Scan width is 150 gauss. O 2 and NiEDDA accessibilities were obtained by power saturation as previously described (Altenbach et al., 1994). Briefly, accessibility to O 2 was measured with samples equilibrated with air, and accessibility to NiEDDA by adding 10 mM NiEDDA and flushing the O 2 out of the sample by applying a constant stream of N 2 . The depth parameter, Ф, was calculated using the relationship Ф= ln[Π(O 2 )/Π(NiEDDA)]. Ф values less than -1.5 represent bulk solvent exposure and lie outside the range of calibration (Figure S3.1). Calibration of immersion depth in relation to accessibility was measured using 1-palmitoyl-2- stearoyl-(n-DOXYL)-sn-glycero-3-phosphocholine and 1-palmitoyl-2-oleoyl-sn-glycero-3- phospho(tempo)choline (Avanti Polar Lipids). The following linear relation was determined between depth and Ф for the nitroxide moiety: [Å]=7.86Ф-2.83 (Figure S3.1). Depth values reported for vesicle-bound endophilin were obtained using a previously reported depth calibration for this system (Jao et al., 2010). 3.4.5 Pulse EPR and distance analysis. For intermolecular distance measurements between single cysteine mutants in the N-terminus (H0) of endophilin A1, samples were prepared as described above and with 100% of the protein spin labeled to enhance signals from neighboring dimers. Intramolecular distance measurements between single cysteine spin labeled mutants in the insert region of endophilin A1 (between subunits of the dimer) were prepared as described 103 above, but were constituted with 2 fold excess of cysless endophilin A1 (C108S) in order to minimize intermolecular signals. Centrifuged sample pellets were resuspended in cryoprotectant buffer containing 20-25% sucrose prior to loading into a quartz capillary and flash-freezing. All samples were flash frozen, and data were acquired at 78 K. A Bruker Elexsys E580 X-band pulse EPR spectrometer fitted with a 3 mm split ring (MS-3) resonator, a temperature controller (ITC503S; Oxford Instruments), and a continuous flow helium cryostat (CF935; Oxford instruments) was used for all four-pulse double electron-electron resonance (DEER) experiments. The data were fitted using Tikhonov regularization (Chiang et al., 2005) as implemented in the DEERAnalysis2011 package (Jeschke et al., 2006). The background contribution from nonspecific interaction was subtracted using a fit dimensionality model. 104 CHAPTER 3 REFERENCES Altenbach, C., Greenhalgh, D.A., Khorana, H.G., and Hubbell, W.L. (1994). A collision gradient method to determine the immersion depth of nitroxides in lipid bilayers: application to spin-labeled mutants of bacteriorhodopsin. Proc. Natl. Acad. Sci. U. S. A. 91, 1667– 1671. Arkhipov, A., Yin, Y., and Schulten, K. (2009). Membrane-bending mechanism of amphiphysin N-BAR domains. Biophys. J. 97, 2727–2735. Bai, J., Hu, Z., Dittman, J.S., Pym, E.C.G., and Kaplan, J.M. (2010). Endophilin functions as a membrane-bending molecule and is delivered to endocytic zones by exocytosis. Cell 143, 430–441. Bergmann, C., Zerres, K., Senderek, J., Rudnik-Schoneborn, S., Eggermann, T., Häusler, M., Mull, M., and Ramaekers, V.T. (2003). Oligophrenin 1 (OPHN1) gene mutation causes syndromic X-linked mental retardation with epilepsy, rostral ventricular enlargement and cerebellar hypoplasia. Brain J. Neurol. 126, 1537–1544. Bhatia, V.K., Madsen, K.L., Bolinger, P.-Y., Kunding, A., Hedegård, P., Gether, U., and Stamou, D. (2009). Amphipathic motifs in BAR domains are essential for membrane curvature sensing. EMBO J. 28, 3303–3314. Blood, P.D., and Voth, G.A. (2006). Direct observation of Bin/amphiphysin/Rvs (BAR) domain- induced membrane curvature by means of molecular dynamics simulations. Proc. Natl. Acad. Sci. U. S. A. 103, 15068–15072. Blood, P.D., Swenson, R.D., and Voth, G.A. (2008). Factors influencing local membrane curvature induction by N-BAR domains as revealed by molecular dynamics simulations. Biophys. J. 95, 1866–1876. Boucrot, E., Pick, A., Çamdere, G., Liska, N., Evergren, E., McMahon, H.T., and Kozlov, M.M. (2012). Membrane fission is promoted by insertion of amphipathic helices and is restricted by crescent BAR domains. Cell 149, 124–136. Bretscher, L.E., Buchaklian, A.H., and Klug, C.S. (2008). Spin-labeled lipid A. Anal. Biochem. 382, 129–131. Campelo, F., McMahon, H.T., and Kozlov, M.M. (2008). The hydrophobic insertion mechanism of membrane curvature generation by proteins. Biophys. J. 95, 2325–2339. Campelo, F., Fabrikant, G., McMahon, H.T., and Kozlov, M.M. (2010). Modeling membrane shaping by proteins: focus on EHD2 and N-BAR domains. FEBS Lett. 584, 1830–1839. 105 Chiang, Y.-W., Borbat, P.P., and Freed, J.H. (2005). The determination of pair distance distributions by pulsed ESR using Tikhonov regularization. J. Magn. Reson. 1997 172, 279–295. Cui, H., Mim, C., Vázquez, F.X., Lyman, E., Unger, V.M., and Voth, G.A. (2013). Understanding the role of amphipathic helices in N-BAR domain driven membrane remodeling. Biophys. J. 104, 404–411. Dalton, L.A., McIntyre, J.O., and Fleischer, S. (1987). Distance estimate of the active center of D-beta-hydroxybutyrate dehydrogenase from the membrane surface. Biochemistry 26, 2117–2130. Drin, G., and Antonny, B. (2010). Amphipathic helices and membrane curvature. FEBS Lett. 584, 1840–1847. Drin, G., Casella, J.-F., Gautier, R., Boehmer, T., Schwartz, T.U., and Antonny, B. (2007). A general amphipathic α-helical motif for sensing membrane curvature. Nat. Struct. Mol. Biol. 14, 138–146. Farsad, K., and Camilli, P.D. (2003). Mechanisms of membrane deformation. Curr. Opin. Cell Biol. 15, 372–381. Farsad, K., Ringstad, N., Takei, K., Floyd, S.R., Rose, K., and De Camilli, P. (2001). Generation of high curvature membranes mediated by direct endophilin bilayer interactions. J. Cell Biol. 155, 193–200. Ferguson, S.M., Ferguson, S., Raimondi, A., Paradise, S., Shen, H., Mesaki, K., Ferguson, A., Destaing, O., Ko, G., Takasaki, J., et al. (2009). Coordinated actions of actin and BAR proteins upstream of dynamin at endocytic clathrin-coated pits. Dev. Cell 17, 811–822. Gallop, J.L., Jao, C.C., Kent, H.M., Butler, P.J.G., Evans, P.R., Langen, R., and McMahon, H.T. (2006). Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J. 25, 2898–2910. Hubbell, W.L., Gross, A., Langen, R., and Lietzow, M.A. (1998). Recent advances in site- directed spin labeling of proteins. Curr. Opin. Struct. Biol. 8, 649–656. Jao, C.C., Hegde, B.G., Chen, J., Haworth, I.S., and Langen, R. (2008). Structure of membrane- bound alpha-synuclein from site-directed spin labeling and computational refinement. Proc. Natl. Acad. Sci. U. S. A. 105, 19666–19671. Jao, C.C., Hegde, B.G., Gallop, J.L., Hegde, P.B., McMahon, H.T., Haworth, I.S., and Langen, R. (2010). Roles of amphipathic helices and the bin/amphiphysin/rvs (BAR) domain of endophilin in membrane curvature generation. J. Biol. Chem. 285, 20164–20170. 106 Jeschke, G., Zimmermann, H., and Godt, A. (2006). Isotope selection in distance measurements between nitroxides. J. Magn. Reson. 1997 180, 137–146. Kaneko, T., Maeda, A., Takefuji, M., Aoyama, H., Nakayama, M., Kawabata, S., Kawano, Y., Iwamatsu, A., Amano, M., and Kaibuchi, K. (2005). Rho mediates endocytosis of epidermal growth factor receptor through phosphorylation of endophilin A1 by Rho- kinase. Genes Cells 10, 973–987. Kato, N., Nakanishi, M., and Hirashima, N. (2002). Transbilayer Asymmetry of Phospholipids in the Plasma Membrane Regulates Exocytotic Release in Mast Cells. Biochemistry 41, 8068–8074. Lai, C.-L., Jao, C.C., Lyman, E., Gallop, J.L., Peter, B.J., McMahon, H.T., Langen, R., and Voth, G.A. (2012). Membrane binding and self-association of the epsin N-terminal homology domain. J. Mol. Biol. 423, 800–817. Langen, R., Oh, K.J., Cascio, D., and Hubbell, W.L. (2000). Crystal structures of spin labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra in terms of structure. Biochemistry 39, 8396–8405. Lee, D.-S., Hirashima, N., and Kirino, Y. (2000). Rapid transbilayer phospholipid redistribution associated with exocytotic release of neurotransmitters from cholinergic nerve terminals isolated from electric ray Narke japonica. Neurosci. Lett. 291, 21–24. Masuda, M., Takeda, S., Sone, M., Ohki, T., Mori, H., Kamioka, Y., and Mochizuki, N. (2006). Endophilin BAR domain drives membrane curvature by two newly identified structure- based mechanisms. EMBO J. 25, 2889–2897. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., De Bock, P.-J., Morais, V.A., Vilain, S., Haddad, D., et al. (2012). LRRK2 controls an EndoA phosphorylation cycle in synaptic endocytosis. Neuron 75, 1008–1021. McMahon, H.T., and Gallop, J.L. (2005). Membrane curvature and mechanisms of dynamic cell membrane remodelling. Nature 438, 590–596. Meinecke, M., Boucrot, E., Camdere, G., Hon, W.-C., Mittal, R., and McMahon, H.T. (2013). Cooperative recruitment of dynamin and BIN/amphiphysin/Rvs (BAR) domain- containing proteins leads to GTP-dependent membrane scission. J. Biol. Chem. 288, 6651–6661. Milosevic, I., Giovedi, S., Lou, X., Raimondi, A., Collesi, C., Shen, H., Paradise, S., O‘Toole, E., Ferguson, S., Cremona, O., et al. (2011). Recruitment of endophilin to clathrin-coated pit necks is required for efficient vesicle uncoating after fission. Neuron 72, 587–601. 107 Mim, C., and Unger, V.M. (2012). Membrane curvature and its generation by BAR proteins. Trends Biochem. Sci. 37, 526–533. Mim, C., Cui, H., Gawronski-Salerno, J.A., Frost, A., Lyman, E., Voth, G.A., and Unger, V.M. (2012). Structural basis of membrane bending by the N-BAR protein endophilin. Cell 149, 137–145. Mizuno, N., Jao, C.C., Langen, R., and Steven, A.C. (2010). Multiple Modes of Endophilin- mediated Conversion of Lipid Vesicles into Coated Tubes. J. Biol. Chem. 285, 23351– 23358. Nicot, A.-S., Toussaint, A., Tosch, V., Kretz, C., Wallgren-Pettersson, C., Iwarsson, E., Kingston, H., Garnier, J.-M., Biancalana, V., Oldfors, A., et al. (2007). Mutations in amphiphysin 2 (BIN1) disrupt interaction with dynamin 2 and cause autosomal recessive centronuclear myopathy. Nat. Genet. 39, 1134–1139. Peter, B.J., Kent, H.M., Mills, I.G., Vallis, Y., Butler, P.J.G., Evans, P.R., and McMahon, H.T. (2004). BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 303, 495–499. Qualmann, B., Koch, D., and Kessels, M.M. (2011). Let‘s go bananas: revisiting the endocytic BAR code. EMBO J. 30, 3501–3515. Rao, Y., and Haucke, V. (2011). Membrane shaping by the Bin/amphiphysin/Rvs (BAR) domain protein superfamily. Cell. Mol. Life Sci. 68, 3983–3993. Ringstad, N., Gad, H., Löw, P., Di Paolo, G., Brodin, L., Shupliakov, O., and De Camilli, P. (1999). Endophilin/SH3p4 is required for the transition from early to late stages in clathrin-mediated synaptic vesicle endocytosis. Neuron 24, 143–154. Schuske, K.R., Richmond, J.E., Matthies, D.S., Davis, W.S., Runz, S., Rube, D.A., van der Bliek, A.M., and Jorgensen, E.M. (2003). Endophilin is required for synaptic vesicle endocytosis by localizing synaptojanin. Neuron 40, 749–762. Soda, K., Balkin, D.M., Ferguson, S.M., Paradise, S., Milosevic, I., Giovedi, S., Volpicelli- Daley, L., Tian, X., Wu, Y., Ma, H., et al. (2012). Role of dynamin, synaptojanin, and endophilin in podocyte foot processes. J. Clin. Invest. 122, 4401–4411. Sorre, B., Callan-Jones, A., Manzi, J., Goud, B., Prost, J., Bassereau, P., and Roux, A. (2012). Nature of curvature coupling of amphiphysin with membranes depends on its bound density. Proc. Natl. Acad. Sci. U. S. A. 109, 173–178. Stachowiak, J.C., Schmid, E.M., Ryan, C.J., Ann, H.S., Sasaki, D.Y., Sherman, M.B., Geissler, P.L., Fletcher, D.A., and Hayden, C.C. (2012). Membrane bending by protein-protein crowding. Nat. Cell Biol. 14, 944–949. 108 Strandberg, E., and Killian, J.A. (2003). Snorkeling of lysine side chains in transmembrane helices: how easy can it get? FEBS Lett. 544, 69–73. Sundborger, A., Soderblom, C., Vorontsova, O., Evergren, E., Hinshaw, J.E., and Shupliakov, O. (2011). An endophilin-dynamin complex promotes budding of clathrin-coated vesicles during synaptic vesicle recycling. J. Cell Sci. 124, 133–143. Varkey, J., Isas, J.M., Mizuno, N., Jensen, M.B., Bhatia, V.K., Jao, C.C., Petrlova, J., Voss, J.C., Stamou, D.G., Steven, A.C., et al. (2010). Membrane curvature induction and tubulation are common features of synucleins and apolipoproteins. J. Biol. Chem. 285, 32486– 32493. Verstreken, P., Koh, T.-W., Schulze, K.L., Zhai, R.G., Hiesinger, P.R., Zhou, Y., Mehta, S.Q., Cao, Y., Roos, J., and Bellen, H.J. (2003). Synaptojanin is recruited by endophilin to promote synaptic vesicle uncoating. Neuron 40, 733–748. Weissenhorn, W. (2005). Crystal structure of the endophilin-A1 BAR domain. J. Mol. Biol. 351, 653–661. Westphal, C.H., and Chandra, S.S. (2013). Monomeric synucleins generate membrane curvature. J. Biol. Chem. 288, 1829–1840. Zhu, C., Das, S.L., and Baumgart, T. (2012). Nonlinear sorting, curvature generation, and crowding of endophilin N-BAR on tubular membranes. Biophys. J. 102, 1837–1845. 109 CHAPTER 4 Lysine Acetylation is a Mechanism for Regulating the Interaction between Proteins and Membranes Mark R. Ambroso 1 , Alan Okada 1 , Daniel Merken 2 , Arthur Alves Melo 3 , Oliver Daumke 3 , Karen Chang 4 , Ian S. Haworth 2 , and Ralf Langen 1,¥ 1 Zilkha Neurogenetic Institute, Department of Biochemistry and Molecular Biology, University of Southern California, Los Angeles, California 90033, USA, 2 Department of Pharmacology and Pharmaceutical Sciences, University of Southern California, Los Angeles, California 90089, USA, 3 Max-Delbrück-Center for Molecular Medicine, Crystallography, Robert-Rössle-Straße 10, 13092 Berlin, Germany; Institute of Chemistry and Biochemistry, Freie Universität Berlin, Takustraße 6, 14195 Berlin, Germany; Institute of Medical Physics and Biophysics, Charité, Charitéplatz 1, 10117 Berlin, Germany, 4 Zilkha Neurogenetic Institute and Department of Cell & Neurobiology, Keck School of Medicine, University of Southern California, Los Angeles, California, United States of America ; Neuroscience Graduate Program, University of Southern California, Los Angeles, California, United States of America ¥ Corresponding Author: Ralf Langen, Tel.: 323-442-1323 Fax: 323-442-4404. Email: Langen@usc.edu 110 CHAPTER 4 ABSTRACT Post-translational modifications are commonly used to regulate protein function in vivo including protein-membrane interactions. Lysine acetylation has been well-studied as a modulator of protein-DNA interactions but its importance in regulating proteins outside of the field of epigenetics has been limited. Using existing modification and structural databank information, we find lysine acetylation to be highly localized in membrane versus non-membrane binding regions of BAR, PX, C2, and Eps15-homology lipid-binding domains, but not in PH domains. Acetylation of BAR, Eps15-homology, or C2 domain containing proteins amphiphysin, EHD2, or synaptotagmin 1, respectively, inhibits their ability to remodel liposomes and/or reduces overall membrane affinity. Amphiphysin and EHD2 are known to produce distinctive membrane-associated phenotypes when expressed in cell culture and acetylation-mimicking mutations abolish this localization. Taken together, we show that lysine acetylation is previously unconsidered and potent mechanism for regulating protein-membrane interactions. 111 4.1 Introduction Cells can potently regulate a protein‘s function by enzymatically modifying its amino acid side-chains. Post-translational modifications (PTMs) act via a variety of mechanisms in order to influence virtually all cellular processes. Lysine acetylation is the fourth most common PTM (Prabakaran et al., 2012) best known for its ability to regulate protein-DNA (Gräff and Tsai, 2013) and protein-protein interactions (Choudhary et al., 2014), but little has been done to evaluate how it can regulate protein-membrane interactions. Proteins typically associate with membranes through a combination of electrostatic and hydrophobic driving forces between their amino acids and the phospholipid bilayer (Lemmon, 2008). Acetylation of the lysine side chain removes a positive charge and adds an acetyl group, changing the side chain from being hydrophilic to hydrophobic. This mechanism has been shown for how acetylation can modulate histone-DNA interactions by reducing the electrostatic affinity between lysines and the negatively charged phosphate-deoxyribose backbone (Hong et al., 1993). As this regulatory process relies on the same principles which regulate how proteins interact with membranes, it is plausible that this PTM may therefore be an effective means for regulating membrane association. However, this concept has yet to be studied. Membrane-binding proteins typically contain lipid-binding domains (LBDs), such as BIN/amphiphysin/RVS (BAR), pleckstrin homology (PH), phox-homology (PX), Eps15- homology (EHD), and C2 domains (Daumke et al., 2007; Lemmon, 2008; Qualmann et al., 2011). These LBDs typically facilitate the localization of the protein to specific membranes within the cell in order to carry out their function, either by simply interacting with the membrane, remodeling membranes into specific shapes, or by acting to recruit the protein in 112 order for other domain within the protein to carry out their function. Post-translation modifications have been observed to be a potent means for regulating membrane binding proteins in vivo. The membrane affinity of BAR proteins syndapin and ACAP4 is affected by phosphorylation (Takeda et al., 2013; Zhao et al., 2013). LRRK2, a kinase constitutively activated in inheritable Parkinson‘s disease, phosphorylates endophilin A1 in a regulatory cycle and the aberrant phosphorylation of endophilin directly affects its ability to contribute to membrane remodeling in synaptic endocytosis (Matta et al., 2012). The authors of this study contributed the defect in synaptic endocytosis to a reduced membrane affinity of the phosphorylated protein. Interestingly, the phosphorylation site in endophilin A1 is neighboring a lysine acetylation site. Acetylation has been shown to affect the specificity PTEN has for phosphatidylinositol 3,4,5-triphoshpate by modifying lysines in the catalytic cleft (Okumura et al., 2006). However, the idea that acetylation may be used as a general mechanism for regulating how proteins interact with membranes has not been investigated. Here we show that lysine acetylation is a prevalent modification for lysine residues within the membrane binding regions of common membrane binding domains. We find that lysine acetylation can alter how BAR protein amphiphysin, Eps15-homology protein EHD2, and the C2 protein synaptotagmin I interact with membranes by reducing their binding affinity. In addition, our data shows that acetylation-mimicking constructs of EHD2 and amphiphysin localize differently in cells compared to wild type. It therefore seems plausible that lysine acetylation can act as a mechanism for regulating protein-membrane interactions in vivo. 113 4.2 Results 4.2.1 Lysine Acetylation is a Prevalent Modification in Membrane Binding Regions of Known Membrane-Interacting Domains 4.2.1.1 Strategy for Determining the Propensity of Acetylation in Membrane Binding Regions Although PTMs have been extensively studied for their affects on certain proteins functions, their affect on protein-membrane interactions has been cursorily examined. In order to determine if certain PTMs are localized in proteins that interact with membranes, we performed a preliminary screen in which a handful of peripheral membrane proteins with well characterized membrane binding regions and crystal structures were referenced against a PTM databank (Hornbeck et al., 2012). A wide-variety of modifications were found in membrane interacting regions including lysine acetylation. This preliminary finding was surprising, as lysine acetylation has been extensively studied for its effect on DNA-histone or protein-protein interactions but very few studies have examined the affect this modification has on protein- membrane interactions. The significance of this observation was not initially clear as our initial screen only included a small pool of different proteins. We therefore devised methods to expand our analysis to a large number of peripheral membrane proteins in order to measure if lysines in membrane binding regions have a higher prevalence of membrane acetylation. In order to do so, we had to devise methods that allowed us to overcome several limiting factors. First, the structure and membrane binding motifs of many peripheral membrane proteins have yet to be determined, making differentiation of membrane or nonmembrane binding lysines impossible. Second, many 114 peripheral membrane proteins such as the annexins can refold into several different protein conformations thereby creating difficulties in defining a concrete membrane binding region (Hegde et al., 2006; Isas et al., 2003; Kim et al., 2005; Langen et al., 1998). Therefore, we chose to limit our analysis to membrane binding domains which are relatively static and possess structural homology across family members, such as the BAR, C2, PX, EHD, and PH domains. These boundaries allow us to generalize the membrane binding information known for a few of the domain members to the broader domain family either by structure or sequence homology. For example, membrane interacting residues in BAR proteins endophilin A1 and amphiphysin have been extensively characterized by crystallography (Gallop et al., 2006; Masuda et al., 2006; 2004; Weissenhorn, 2005), mutational analysis (Ferguson et al., 2009; Fernandes et al., 2008; Jung et al., 2010; Mim et al., 2012), and site-directed spin labeling (SDSL) and electron paramagnetic resonance (EPR) (Ambroso et al., 2014; Gallop et al., 2006; Isas et al., 2015; Jao et al., 2010). Due to the highly static nature of the BAR domain which does not undergo major conformational changes upon binding lipids (Henne et al., 2007; Jao et al., 2010), other BAR proteins membrane binding regions which have not undergone the same rigorous analysis can be inferred from their homology to endophilin A1 and amphiphysin or alternatively to a more homologous BAR protein which has been thoroughly studied. While we can draw off of previous studies which provide a near comprehensive characterization of the membrane binding sites in these model domains, not every residue in each domain has been analyzed. Therefore, for purposes of simplicity, residues which have not been previously studied but are close to membrane binding residues in both sequence and spatial proximity were included as membrane binding residues. For example, while not every single residue along the concave surface of amphiphysin‘s BAR domain has been measured by EPR membrane-accessibility measurements, 115 numerous intermittent sites along this surface are membrane exposed and suggest that intervening sites are as well (Isas et al., 2015). Considering that the lysine side chain can extend up to 7 Å and would likely interact with the more distal portions of negatively charged lipids, it is reasonable to include residues that are in nearby proximity to known membrane binding residues. For our analysis of proteins with existing crystal structures, acetylation data (www.phosphosite.org; Hornbeck et al., 2012) and structure files (www.rcsb.org) were downloaded in bulk and scripts were generated to correlate the modification data into the .pdb files by marking acetylated lysines with a specific color. Modified .pdb files were then processed through a structural homology algorithm (generated by the Haworth group), sorted into individual clusters, and three-dimensionally aligned to one another in PyMOL. These clusters serve as the format by which acetylated lysines in many different proteins could be observed in context of their tertiary structure. It also allowed for proteins whose membrane binding regions are not well defined to be inferred by structural homology with well-characterized lipid binding domains. Importantly, false positives were eliminated during this step of the analysis due to their lack of structural homology. We then used these clusters to manually identify the differently colored acetylation sites present in membrane or nonmembrane binding regions. Acetylated lysines with 2 Å of a membrane binding residue were considered to be in a membrane binding region and all others to be outside of this region. This analysis was done manually and modifications were verified by cross-referencing hits with the Phosphosite.org website (Hornbeck et al., 2012). The number of acetylated lysines in membrane or nonmembrane binding regions were tallied and normalized according to the total amount of lysines in each region counting only residues elucidated in the .pdb file. 116 To analyze protein domains that only have a few existing crystal structures we relied on sequence homology to align the protein domains and therefore their membrane binding regions. Acetylated lysines were gathered from www.Phosphosite.org and manually annotated onto their sequence alignments. 4.2.1.2 BAR Proteins have Increased Levels of Lysine Acetylation in their Membrane Binding Regions BAR proteins are best known for their contributions to endocytosis (Qualmann et al., 2011; Rao and Haucke, 2011). Of the BAR proteins that have been crystallized, thirteen had at least one acetylated lysine in their BAR domain. These included several different subclasses of BAR proteins, including N-BAR (AMPH, BIN1 and 2, SH3GL2), F-BAR (FCHo2, FNBP1, TRIP10, PACSIN 1 and 2), PX-BAR (SNX9, SNX33) and I-BAR (BAIAP2). The membrane binding regions of these domains were determined using previously published data as referenced below. Using structural homology alignments in PyMOL, our analysis revealed 15% of lysines in membrane binding regions of these BAR proteins have been observed to be acetylated compared to 0.59% for lysines in nonmembrane binding regions (Figure 4.1a). These lysines were typically localized to the concave surface of the N-BAR or F-BAR homodimer where these proteins are known to interact with negatively charged membranes via electrostatic interactions (Ambroso et al., 2014; Farsad et al., 2001; Frost et al., 2008, 2008; Gallop et al., 2006; Henne et al., 2007; Isas et al., 2015; Itoh et al., 2005; Masuda et al., 2006; Peter et al., 2004; Qualmann et al., 2011; Rao and Haucke, 2011; Shimada et al., 2007). The I-BAR protein IRSp53 (Gene name: BAIAP3), a BAR protein who binds membranes using its convex surface instead of its concave surface like most other BAR proteins, has localized acetylation on its membrane convex surface but not its opposing concave side (Saarikangas et al., 2009). N-BAR proteins were also observed 117 to have highly localized acetylation in the N termini regions known to be important for membrane interaction and curvature generation (Ambroso et al., 2014; Gallop et al., 2006; Isas et al., 2015; Jao et al., 2010). This data indicated that lysines in important membrane interacting regions of BAR proteins have a higher probability of being acetylated. 118 Figure 4.1 BAR and PX Domains are Commonly Acetylated, but not PH Domains. Percentage of acetylated lysines in membrane binding regions (%K AC Membrane ) and non- membrane binding regions (%K AC Non-membrane ) of proteins belonging to (A) BAR, (B) PX, or (C) PH domain families listed by gene name and their respective structure file (.pdb). Percentages represent the number of acetylated residues out of the total number of lysines in their respective regions. An example structure for each domain family is provided to illustrate membrane binding residues (blue α-carbons) and non membrane binding regions (gray α-carbons). These regions were determined using data from previous biophysical studies as discussed in the text. In the case of the BAR family which contains several different subdomains (N-BAR, F-BAR, I-BAR), multiple studies and structures were used to define their respective membrane binding regions. A red line is used to illustrate the approximate level of the lipid headgroups in respect to the modeled domains. 119 Figure 4.1, continued 120 4.2.1.3 PX Domains have Increased Levels of Acetylation in their Membrane Binding Regions PX domains are known to specifically bind phosphoinositols and are the defining domain of the sorting nexin family (Carlton et al., 2005). Similar to the FYVE domains, PX domains are known to bind membranes through a combination of electrostatic interactions, headgroup binding, and membrane insertion (Bravo et al., 2001). Crystallography, EPR, and mutational studies on p40 phox , p47 phox , and Vam7p have been used to characterize the lipid binding residues in PX domains (Bravo et al., 2001; Lee et al., 2006; Lomize et al., 2007; Stahelin et al., 2003). A model of membrane bound p40 phox (Gene name: NCF4) is shown in Figure 4.2. Out of the PX domains that have existing crystallographic data nine have been shown to contain acetylated lysines. Only 3.7% of lysines in nonmembrane binding residues have been observed to be acetylated compared to that of 21.72% of lysines in membrane binding regions (Figure 4.1b). Modified lysines tended to localize to residues in the α1, PxxP, and α2 regions known to be important for binding phosphoinositols (Ellson et al., 2002). The same was observed for the PX domains of sorting nexins 2, 6, 25, 29, and 30 when they were aligned by sequence homology to p40 phox and sorting nexins with crystal structures; 1, 5, 9, and 14 (Figure 4.2). 4.2.1.4 PH Domains do not have Increased Levels of Acetylation in their Membrane Binding Regions PH domains specifically recognize phosphoinositidyl-3,4,5-triphosphate and work to localize proteins to the plasma membrane (Lemmon et al., 1996). The lipid binding regions of the PH domain containing protein GRP1 have been characterized through EPR, NMR, and crystallography (Chen et al., 2012; He et al., 2008; Lai et al., 2013; Lumb et al., 2011; Pilling et 121 al., 2011). Out of PH domains that have been crystallized ten had been observed to be acetylated. Our analysis revealed that acetylation was observed to be equally dispersed in lysines inside and outside of the membrane binding regions (Figure 4.1c). These results suggest that acetylation is not specifically targeted to membrane regions of this domain. 4.2.1.5 C2 Domains have Increased Levels of Acetylation in their Membrane Binding Regions The membrane binding regions of C2 domains include the Ca 2+ coordinating loops and a β-binding site which specifically coordinates phosphoinositols using cationic residues (Kuo et al., 2009; Li et al., 1995). These regions have been characterized in studies on membrane bound synaptotagmin I or cytosolic phospholipase A2 (Bai et al., 2004; Brose et al., 1992; Frazier et al., 2002, 2003; Hui et al., 2011; Kuo et al., 2009; Li et al., 1995; Martens et al., 2007; Rufener et al., 2005; Stahelin et al., 2014; Ward et al., 2013). Many different proteins contain C2 domains that either share close sequence homology with synaptotagmin‘s C2A or C2B domain, but very few have been crystallized. We therefore used sequence homology to align many different C2 domain protein sequences to that of either synaptotagmin C2A (residues 157-245) or C2B (residues 287-378) and annotated these sequences with known sites of acetylation. Our analysis showed that lysines in the beta binding sites or Ca 2+ binding loops of the C2A and C2B domains had higher densities of modified lysines (Figure 4.3). In fact, 37.85% of lysines in these membrane binding regions have been observed to be acetylated compared to 7.64% in nonmembrane binding regions. 122 Figure 4.2 Sequence Alignments of PX Domains from Sorting Nexins to p40 phox . The structure of p40 phox (1H6H.pdb) with membrane binding regions highlighted in blue and nonmembrane binding in gray, as determined by previous studies (Bravo et al., 2001; Lee et al., 2006; Lomize et al., 2007; Stahelin et al., 2003). The relative level of the lipid headgroups was manually placed and is represented by the red line. The sequence of p40phox (residues 19-140) was used as a model template to align the sequences of PX domains from members of the sorting nexin family. The residues which are part of the membrane binding region are highlighted by the blue line below the numbering of p40 phox ‘s sequence. Acetylation sites are marked in green. 123 Figure 4.2, continued 124 Figure 4.3 Lysine acetylation is localized to membrane binding regions of C2 domains. The crystal structure of human synaptotgamin 1 C2A (A) and C2B (B) (2R83.pdb) with regions of the Ca 2+ and phosphoinositol binding sites highlighted with blue α-carbons. The linker region is omitted for simplicity. (C) Sequence alignments of C2 domains by gene name. While many different proteins contain C2 domains, not all contain both a C2A and C2B domain. Therefore, alignments were done to human synaptotagmin 1 (a.a. 150-331) and those with >50% sequence homology to either C2A or C2B were included for analysis. Acetylation sites are highlighted by a green box. Membrane binding regions of C2A and C2B are also signified using colored lines below the synaptotagmin 1 sequence. 125 Figure 4. 3, continued 126 4.2.1.6 EHD Family Members have Increased Levels of Acetylation in their Membrane Binding Regions EHDs make up a family of dynamin-related adenosine triphosphatases. They are implicated in transportation of exit cargo from the endocytotic recycling compartment to the plasma membrane (George et al., 2007) as well as regulating caveolae dynamics (Morén et al., 2012). The N terminus and tip regions of EHD2 were experimentally determined to be the primary membrane binding sites using crystallography, mutational analysis, and EPR (Daumke et al., 2007; Shah et al., 2014). As EHD2 is the only family member that has been crystallized, sequence alignments to the EHD1, 3 and 4 were performed and annotated with known acetylation modifications (Figure 4.4). No Acetylation was observed in the N termini of the EHDs but increased levels were observed in the tip regions. Lysines in this region had a 45.83% probability of being acetylated compared to 11.67% of lysines outside of this region. This analysis revealed that lysines in membrane binding regions of BAR, PX, C2, and EHD proteins, but not PH domains, have a significantly higher probability of being acetylated compared to lysines in nonmembrane binding regions. Averaged together, 26.35% of lysines in membrane binding regions have been observed to be acetylated compared to 5.9% for lysines outside of these regions. These results suggest that lysine acetylation may play an important role in regulating how some lipid binding proteins interact with membranes. 127 Figure 4.4 The Eps15-homology EHD1-4 proteins have localized lysine acetylation in their membrane binding regions. (A) The crystal structure of EHD2 (4CID.pdb) with membrane binding regions labeled in blue. A red line was manually placed to represent the level of the lipid headgroups. (B) EHD 1, 3 and 4 exhibit a high degree of sequence homology to residues 302-419 of EHD2. Residues within this sequence known to bind membranes are highlighted by a blue bar below EHD2‘s sequence numbering. Acetylation sites are highlighted by a green box. 128 4.2.2 Acetylation-mimicking Mutations in Amphiphysin, EHD2, and Synaptotagmin 1 Reduces their Ability to Shape Membranes by Reducing their Membrane Affinity To gain a better understanding of how acetylation modulates the interaction of lipid binding domains with membranes we tested their impact in vitro. Three candidates were chosen to represent their respective domains, amphiphysin (BAR), EHD2 (EHDs), and synaptotagmin I (C2). All three proteins have been previously shown by our group or others to bind, tubulate, and vesiculate anionic liposomes (Hui et al., 2011; Isas et al., 2015; Peter et al., 2004; Shah et al., 2014). In order to recapitulate lysine acetylation sites in membrane binding regions, recombinant constructs of these candidate proteins were generated with lysines mutated to glutamines, a common mechanism for mimicking acetylation (Zhou et al., 2012). The specific mutations chosen are described for their respective proteins below. These and wild type constructs were recombinantly expressed, purified, incubated with anionic liposomes, and observed for their ability to remodel membranes by negative-stain transmission electron microscopy (nsTEM) and/or for analysis by SDSL and EPR. The N-BAR domain (residues 1-244) of amphiphysin (Gene name: AMPH) has been shown to potently induce membrane curvature of liposomes in vitro using its N-terminal amphipathic helices and concave BAR surface (Isas et al., 2015; Peter et al., 2004). The N terminus has also been observed to contain several acetylated lysines (Hornbeck et al., 2012). We therefore created acetylation-mimicking mutations K5Q K15Q and examined its ability to remodel vesicles compared to that of wild type. Incubation with an optimized lipid composition for forming lipid tubes, POPG:POPE, in a 1:70 (protein:lipid) molar ratio (Isas et al., 2015) results in the near complete transformation of liposomes into lipid tubes consisting of a constant diameter averaging ~40 nm (Figure 4.5a). Tubulation was significantly inhibited in the K5Q 129 K15Q construct and the small amounts of tubes produced have a ‗bead on a string‘ like morphology (Figure 4.5b). We then generated constructs containing K5R K15R mutations to examine whether the positive charge of the arginine side chain could rescue amphiphysin‘s ability to tubulate these liposomes. This construct generated tubes in high similarity to those produced by wild type (Figure 4.5c). When these experiments were repeated using a lipid composition that does not facilitate amphiphysin‘s ability to tubulate vesicles as strongly (2:1 wt/wt POPS:POPC), wild type and arginine mutants were capable of generating a mixed sample of small vesicles (20-30 nm in diameter) and tubes while the acetylation-mimicking constructs produced no tubulation and minimal amounts of vesiculation (Figure 4.5d-f). We have previously shown that phosphorylation of an amphipathic helix in the amphiphysin homolog endophilin A1 can change the structure that this BAR protein uses to remodel vesicles without reducing the protein‘s affinity for the membrane (Ambroso et al., 2014). To examine whether the acetylation-mimicking mutations affect amphiphysin‘s ability to remodel vesicles is due to changes in membrane affinity or local structure, we generated constructs with a cysteine at residue 20 with or without the K5Q K15Q mutations for analysis by SDSL and EPR. Position 20C was chosen since it is in close proximity to the K5Q K15Q mutations and has been observed to be part of a membrane-inserted N-terminal amphipathic helix on both vesicles and tubes (Isas et al., 2015). These constructs were spin labeled, titrated with increasing concentrations of liposomes, and monitored for change in overall EPR signal amplitude (Figure 4.5g) similar to what has been done previously for spin labeled endophilin A1 (Ambroso et al., 2014). A comparison of the lipid saturation curve between spin labeled wild type and K5Q K15Q amphiphysin reveals that more lipid is required to saturate the spin label in the acetylation-mimicking construct suggesting its affinity for the membrane is reduced. 130 Therefore, it seems that N-terminal acetylation of amphiphysin can reduce its affinity for membranes and thereby inhibit its ability to remodel vesicles. To examine the effect of acetylation of EHD2 and synaptotagmin I (Gene name: SYT1) we generated constructs which mimic known acetylation sites in membrane binding regions of these proteins. Position K324 in EHD2 has been previously shown to deeply penetrate into the bilayer on small vesicles (Shah et al., 2014) and is a site of acetylation (Hornbeck et al., 2012). Similarly, K237 of synaptotagmin I is located in a Ca 2+ binding loop which contacts the membrane (Frazier et al., 2003). Acetylation-mimicking constructs at these sites were generated with or without a nearby cysteine for analysis by SDSL and EPR. EHD2 spin labeled at position 321 with or without the K324Q mutation was titrated with folch liposomes, a lipid system previously optimized for this protein (Shah et al., 2014), and monitored for changes in EPR signal amplitude (Figure 4.5h). This same analysis was performed for synaptotagmin I spin labeled at position 227 with or without a K237Q mutation and titrated with a previously optimized lipid composition (Frazier et al., 2003) (Figure 4.5i). In both cases, the acetylation- mimicking mutations reduced overall membrane affinity. Additionally, tubulation was only observed by nsTEM for wild type EHD2 and not the K324Q mutant when the protein was incubated with POPG:POPE (2:1 wt/wt) liposomes at a 1:2 protein to lipid ratio (Figure 4.5j and k). Therefore, acetylation of lysine residues in membrane binding regions of all three protein domains reduced their overall membrane affinity as well as the ability of amphiphysin and EHD2 to remodel negatively charged membranes. 131 Figure 4.5. Acetylation-Mimicking Mutations Inhibit Membrane Remodeling and Reduce Membrane Affinity. nsTEM images of POPG:POPE (2:1 wt/wt) liposomes incubated with wt (a), K5Q K15Q (b), or K5R K15R (c) amphiphysin N-BAR domain at a 1:70 protein to lipid molar ratio. Similar experiments were repeated using POPS:POPC (2:1 wt/wt) liposomes incubated with wt (d), K5Q K15Q (e), or K5R K15R (f) amphiphysin. Spectral amplitudes from spin labeled protein as a function of lipid concentration with (dashed line) or without (solid line) the indicated acetylation-mimicking mutation for amphiphysin spin labeled at position 20 (g), EHD2 spin labeled at position 321 (h), or synaptotagmin I spin labeled at position 227 (i). nsTEM images of POPG:POPE (2:1 wt/wt) liposomes incubated with wt (j) or K324Q (k) EHD2 at a 2:1 protein to lipid molar ratio. Scale bars = 500 nm. Error bars represent standard deviation, n= at least 3 independent measurements. 132 Figure 4.5, continued 133 4.2.3 Acetylation-mimicking mutations of amphiphysin and EHD2 alter their localization in cells In order to evaluate whether the acetylation of lysines within membrane-binding regions plays an important role in how proteins bind cellular membranes, we studied the effects of acetylation-mimicking mutations of amphiphysin and EHD2. We developed constructs containing glutamine mutations at position K15Q and K5Q K15Q for amphiphysin as well as K324 and/or K328 for EHD2 tagged with a green-fluorescent protein (GFP). In the case of amphiphysin, it has previously been demonstrated that cellular expression of GFP-tagged amphiphysin produces tubulated networks (Peter et al., 2004). Consistent with these findings, we also found tubular networks in cultured COS-7 cells transfected with the BAR domain of wild type C-terminally-GFP-tagged amphiphysin (wt-Amphi-GFP) (Figure 4.6a). We next overexpressed the single (15Q-Amphi-GFP) or double (5Q15Q-Amphi-GFP) mutant of amphiphysin (Figure 4.6b and c). We found that these mutants were incapable of forming a tubular network and instead distributed throughout the cytoplasm. Surface plots, which reflect the relative fluorescent intensity by area in the middle of a cell, were processed for cells in Figures 4.6a-b and are shown in Figures 4.6d-f. These plots reveal the significant difference in cellular localization between the constructs. Some cells in both wt-Amphi-GFP and 5Q15Q- Amphi-GFP conditions developed clumping within the cytoplasm. This was observed to become more intense over time, likely representing an artifact of overexpression. Cellular expression of wild type N-terminally GFP-tagged EHD2 (wt-EHD2-GFP) produces a phenotype wherein EHD2 subcellular distribution is largely confined to the cellular membrane and the cytoplasmic region immediately adjacent in HeLa cells (Daumke et al., 2007) (Figure 4.6g). Intriguingly, transfection of the single mutant (324Q-EHD2-GFP) or double 134 mutant (324Q328Q-EHD2-GFP) produces essentially identical phenotypes characterized by cytoplasmic distribution (Figure 4.6h and i). This change is highlighted in the relative fluorescence density plots in Figures 4.6j-l. These findings suggest that acetylation directly affects how lipid binding proteins such as amphiphysin and EHD2 bind cellular membranes. 135 Figure 4.6 Acetylation-mimicking Mutations Alter how Amphiphysin and EHD2 Interact with Cellular Membranes. Fluorescent confocal imaging of C-terminally GFP- tagged constructs of the N-BAR domain of amphiphysin in the absence (a) or presence of an acetylation-mimicking mutation K15Q (15Q-Amphi-GFP) (b) or K5Q K15Q (5Q15Q- Amphi-GFP) (c) transfected into COS-7 cells and imaged after 24 hours. d-f) Surface area plots obtained at the focal plane of the center of the cells imaged in a-c. N-terminally GFP- tagged constructs of EHD2 in the absence (g) or presence of K324Q (h) or K324Q K328Q (i) acetylation-mimicking mutation were transfected into COS-7 cells. j-l) Surface area plots of cells in g-i were obtained from images at the focal plane of the middle of the cell. 136 Figure 4.6, continued 137 4.3 Discussion In this manuscript we have identified lysine acetylation as a mechanism for regulating how lipid binding proteins interact with membranes. We show for a large number of proteins in the C2, BAR, PX, and EHD domains that lysine acetylation is localized to membrane binding regions. In contrast, the phosphoinositol binding PH domain was observed to be acetylated in equal proportions inside or outside of their membrane binding regions, suggesting that regulation by acetylation may be specific to certain domains. By mimicking acetylation sites found in BAR protein amphiphysin, C2 protein synaptotagmin, and Eps15-homology protein EHD2 with glutamines, we show that these modifications reduce membrane binding affinity and influence the ability of these proteins in forming specific membrane shapes. GFP-tagged constructs of amphiphysin and EHD2 show that acetylation significantly affects the way they interact with membranes in the cell. These findings suggest that lysine acetylation is a PTM capable of directly regulating how proteins interact with membranes. To determine if lysine acetylation is a common mechanism for regulating membrane proteins we utilized a proteomic and bioinformatic approach to determine that lysines in membrane binding regions are more likely to be acetylated. This approach is limited by the relatively small number of proteins which have crystal structures, lipid binding domains extensively characterized, and existing PTM data which may not include all relevant modifications. Still, we were able to identify significant differences in the prevalence of lysine acetylation in membrane versus nonmembrane binding regions of BAR, PX, C2, and EHD domains. It is important to note that our approach was able to detect a lipid binding domain which does not have significant acetylation in its membrane binding regions; the PH domain. This suggests that our analysis has a certain level of specificity. Future work will be performed 138 to expand this analysis to a broader range of proteins in hope that it can elucidate lipid binding domains which may or may not be regulated by acetylation. Proteins are thought to interact with lipid membranes through a combination of electrostatic and hydrophobic interactions. Anionic lipid headgroups interact with positively charged side chains while the acyl chains shield the hydrophobic side chains from the aqueous environment. The combination of these two forces likely lends to the overall affinity a protein has for the membrane. Augmentation of the charge or hydrophobicity of a protein‘s side chains is likely to have an effect on the way a protein interacts with a membrane. In fact, charge modulation of an amino acid side chain in a membrane binding region of endophilin A1 was recently shown to change how this protein binds and interacts with liposomes (Ambroso et al., 2014). The addition of an acetyl group to a lysine side chain negates the positive charge on the amine and creates a more hydrophobic side chain. It is therefore reasonable to assume that lysine acetylation could act as a potent modulator of protein membrane interactions. It is important to note, however, that the impact acetylation would have is dependent on the type of protein. In this study, we analyzed the impact of this modification on three different proteins and found their overall membrane affinities to be reduced. However, we cannot exclude the possibility that lysine acetylation may increase or have no affect on membrane binding affinities in other proteins. Our study shows that a single acetylation site is sufficient to completely alter the phenotype EHD2 and amphiphysin have when expressed in cells. This suggests that acetylation can potently affect the way these proteins interact with cellular membranes. As lysine acetylation is a reversible modification, its affect on protein-membrane interactions could be easily toggled on or off at specific times during membrane remodeling processes. Future studies will have to 139 reveal if acetylation is used directly as a regulatory process in vivo. It is also a focus of future studies to examine whether mutation of acetylation sites play a role in disease. 140 4.4 Experimental Procedures 4.4.1 Bioinformatic analysis of protein acetylation Structure files (.pdb) where downloaded in bulk off of the www.rcsb.org website using advanced keyword searches and by selecting to only retrieve representatives at 50% sequence identity. Acetylation data was acquired in bulk from www.phosphosite.org (Hornbeck et al., 2012) in December of 2014. Python scripts were generated which could identify the .pdb id and correlate it with the gene name used in the acetylation files. Additional scripts were used to create files which would label the acetylation sites observed for residues in the .pdb files to be colored in PyMOL (PyMOL Molecular Graphic System, Version 1.7.4 Schrödinger, LLC). These files were then subjected to a structural homology algorithm created in the Haworth lab and placed into clusters based on levels of homology. These clusters were visualized in PyMOL and used to identify whether colored acetylation sites were in membrane or nonmembrane binding regions. Sequence alignments were initially performed using ClustalOmega (Sievers et al., 2011) and prepared for publication using ESPript-http://espript.ibcp.fr (Robert and Gouet, 2014). 4.4.2 Generation of Protein Constructs and Mutants The original plasmids and cDNA encoding mouse EHD2 (a.a. 1-404), rat synaptotagmin- 1 (a.a. 80-421), and Drosophila amphiphysin (a.a. 1-247) were gifts from Dr. Oliver Daumke, Dr. Greg Schiavo, and Dr. Harvey McMahon respectively. An N-terminally GFP-tagged construct of EHD2 was also provided by Dr. Oliver Daumke. The N-BAR domain of amphiphysin was cloned into a pEGFP-N1 vector using 5‘ NheI and 3‘ XhoI cut sites. Lysine to glutamine mutations were made to this construct following Quikchange (Agilent) site-directed 141 mutagenesis. Constructs made for recombinant expression were purified as previously described (Coppola et al., 2001; Isas et al., 2015; Peter et al., 2004; Shah et al., 2014). In brief, all proteins were expressed in E. coli BL21 (DE3) (New England Biolabs). EHD2 and amphiphysin were purified using nickel-nitrilo-triacetic acid–agarose, followed by gel filtration with a superdex 200 column, and in some cases remaining impurities were removed using mono S cation exchange chromatography with a low salt buffer A (20 mM hepes pH 7.4, 1 mM dithiothreitol (DTT)) and elution buffer B (20 mM hepes pH 7.4, 2 M NaCl and 1 mM DTT). Synaptotagmin was purified by immobilizing the protein on glutathione-agarose followed by extensive washing. Synaptotagmin was eluted off the beads by thrombin cleaving of the GST tag from the protein. Protease and other impurities were removed using a mono Q column (GE). Protein concentration was determined by absorbance (280 nm) using an extinction coefficient ε = 21860 M -1 cm -1 . The purified samples were flash frozen and stored at -80° C. Spin label (1-oxyl-2,2,5,5 tetramethyl-Δ3-pyrroline-3-methylmethanethiosulfonate) was incubated in a 5 to 10 fold molar excess of protein immediately following the removal of DTT using size exclusion chromatography (PD-10 column (GE)), and left to react at 4° C overnight. Excess spin label was removed using PD-10 columns. 4.4.3 Cell culture, Transfection and Confocal Microscopy HeLa and COS-7 cell lines were cultured in Dulbecco‘s Modified Eagle‘s Medium (DMEM) supplemented with 100 U/mL penicillin G, 100 μg/mL streptomycin, 4.5 g/L glucose, sodium pyruvate (Cellgro, Manassas, VA) and 10% heat inactivated fetal bovine serum (Invitrogen) at 37 °C in humidified at with 5% CO 2 . 142 Following three washes with phosphate buffered saline solution, HeLa or COS-7 cells lines were trypsin digested and the cell suspensions were centrifuged at 1,000 x g for 5 minutes. Cells were recovered in fresh media and plated on custom, #1 thickness, glass-bottomed coverslips and allowed to recover for 24 hours before transfection. Expression of cDNA constructs was induced using Lipofectamine 2000 (Invitrogen) and 1.2 μg of cDNA plasmid according to manufacturer protocol. Cells were imaged either 6, 12, 24 or 48 hours following transfection with an Olympus IX-83 confocal microscope using an UPLFN 100x oil immersion objective (NA: 1.30). eGFP fluorescence was excited using a 488 laser and light was collected through the objective. Images at various time points were acquired for analysis, which was performed using ImageJ software from the NIH (version 1.48). 4.4.4 Vesicle Preparation The initial preparation of vesicles was the same for all lipid compositions used in this study. Lipid stocks were suspended in chloroform and mixed to the desired molar or weight proportions in organic solvent, dried under a stream of N 2 gas, and dried overnight in a desiccator. For nsTEM studies with amphiphysin and EHD2, multilamellar vesicles (MLVs) of 2:1 wt/wt 1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho-RAC-(1-glycerol)] (POPG) and 1- palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE) were resuspended in buffer A (20 mM Hepes, pH 7.4) to 4 mg/mL and used immediately. Amphiphysin was also tested for its interaction with MLVs of 2:1 wt/wt 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) were resuspended in buffer A to 4 mg/mL. 143 For EPR-based lipid titration experiments of amphiphysin, we used MLVs composed of 2:1 wt/wt POPG:POPE. For saturating EHD2, we used a previously optimized lipid system of bovine total brain lipids, folch I (Sigma) (Shah et al., 2014). These lipids were suspended in 20 mM Hepes, pH 7.4, 100 mM NaCl, and made into small unilamellar vesicles (SUVs) using a tip sonicator. Synaptotagmin I was previously studied by EPR on vesicles composed of a 3:1 molar ratio of POPC:POPS, and extruded through polycarbonate filters with a 0.1 μm pore diameter (Frazier et al., 2003). These vesicles were suspended in 20 mM Hepes, pH. 7.4, 100 mM NaCl, and 1 mM Ca 2+ . For all experiments, protein concentration was held consistent at 10 μM and the amount of lipids added was varied. 4.4.5 Acquisition and Analysis of EPR Data Continuous wave (CW) EPR spectra were recorded for samples placed into Quartz capillaries (VitroComInc., New Jersey) using a Bruker EMX spectrophotometer fitted with an ER4119HS resonator. For lipid titration experiments, CW EPR spectra amplitudes were recorded for samples of spin labeled protein in 20 mM Hepes, pH 7.4, 100 mM NaCl buffer, (as well as 1 mM Ca 2+ in the case of synaptotagmin), or with the addition of varying concentrations of protein. The values were then normalized relative to the protein‘s CW EPR spectral amplitude difference between the protein alone in solution and at supersaturating conditions. Unlike the spin labeled constructs of amphiphysin and EHD2, synaptotagmin I spin labeled at position 227 did not undergo major changes in signal amplitude, similar to what was observed for this region of synaptotagmin previously (Frazier et al., 2003). However, increasing concentrations of lipids significantly increased an immobilized component in the low field transition line. Therefore, the ratio of the amplitude of this peak and that of the central line width was used to plot the affect of 144 increasing lipid concentrations and normalized to its highest (saturating conditions) and lowest (in solution) values obtained. 4.4.6 Electron Microscopy Carbon-coated formvar films mounted on copper grids (Electron Microscopy Services, Hatfield) were suspended on small aliquots of protein-membrane samples for 10 minutes. Excess liquid was removed using filter paper and grids were subsequently subjected to a two minute incubation on a droplet of 1% uranyl acetate was used to stain the sample-coated grids. A JEOL 1400 transmission electron microscope was used for specimen observation at 100 kV. 145 Acknowledgements: Author contributions: M.R.A., A.O., D.M., K.C. performed research. M.R.A. and A.O. wrote the manuscript. 146 CHAPTER 4 REFERENCES Ambroso, M.R., Hegde, B.G., and Langen, R. (2014). Endophilin A1 induces different membrane shapes using a conformational switch that is regulated by phosphorylation. Proc. Natl. Acad. Sci. 111, 6982–6987. Bai, J., Tucker, W.C., and Chapman, E.R. (2004). PIP2 increases the speed of response of synaptotagmin and steers its membrane-penetration activity toward the plasma membrane. Nat. Struct. Mol. Biol. 11, 36–44. Bravo, J., Karathanassis, D., Pacold, C.M., Pacold, M.E., Ellson, C.D., Anderson, K.E., Butler, P.J., Lavenir, I., Perisic, O., Hawkins, P.T., et al. (2001). The crystal structure of the PX domain from p40(phox) bound to phosphatidylinositol 3-phosphate. Mol. Cell 8, 829– 839. Brose, N., Petrenko, A.G., Sudhof, T.C., and Jahn, R. (1992). Synaptotagmin: a calcium sensor on the synaptic vesicle surface. Science 256, 1021–1025. Carlton, J., Bujny, M., Rutherford, A., and Cullen, P. (2005). Sorting nexins--unifying trends and new perspectives. Traffic 6, 75–82. Chen, H.-C., Ziemba, B.P., Landgraf, K.E., Corbin, J.A., and Falke, J.J. (2012). Membrane Docking Geometry of GRP1 PH Domain Bound to a Target Lipid Bilayer: An EPR Site- Directed Spin-Labeling and Relaxation Study. PLoS ONE 7, e33640. Choudhary, C., Weinert, B.T., Nishida, Y., Verdin, E., and Mann, M. (2014). The growing landscape of lysine acetylation links metabolism and cell signalling. Nat. Rev. Mol. Cell Biol. 15, 536–550. Coppola, T., Magnin-Lüthi, S., Perret-Menoud, V., Gattesco, S., Schiavo, G., and Regazzi, R. (2001). Direct Interaction of the Rab3 Effector RIM with Ca2+Channels, SNAP-25, and Synaptotagmin. J. Biol. Chem. 276, 32756–32762. Daumke, O., Lundmark, R., Vallis, Y., Martens, S., Butler, P.J.G., and McMahon, H.T. (2007). Architectural and mechanistic insights into an EHD ATPase involved in membrane remodelling. Nature 449, 923–927. Ellson, C.D., Andrews, S., Stephens, L.R., and Hawkins, P.T. (2002). The PX domain: a new phosphoinositide-binding module. J. Cell Sci. 115, 1099–1105. 147 Farsad, K., Ringstad, N., Takei, K., Floyd, S.R., Rose, K., and De Camilli, P. (2001). Generation of high curvature membranes mediated by direct endophilin bilayer interactions. J. Cell Biol. 155, 193–200. Ferguson, S.M., Ferguson, S., Raimondi, A., Paradise, S., Shen, H., Mesaki, K., Ferguson, A., Destaing, O., Ko, G., Takasaki, J., et al. (2009). Coordinated actions of actin and BAR proteins upstream of dynamin at endocytic clathrin-coated pits. Dev. Cell 17, 811–822. Fernandes, F., Loura, L.M.S., Chichón, F.J., Carrascosa, J.L., Fedorov, A., and Prieto, M. (2008). Role of helix 0 of the N-BAR domain in membrane curvature generation. Biophys. J. 94, 3065–3073. Frazier, A.A., Wisner, M.A., Malmberg, N.J., Victor, K.G., Fanucci, G.E., Nalefski, E.A., Falke, J.J., and Cafiso, D.S. (2002). Membrane Orientation and Position of the C2 Domain from cPLA2 by Site-Directed Spin Labeling†. Biochemistry 41, 6282–6292. Frazier, A.A., Roller, C.R., Havelka, J.J., Hinderliter, A., and Cafiso, D.S. (2003). Membrane- Bound Orientation and Position of the Synaptotagmin I C2A Domain by Site-Directed Spin Labeling†. Biochemistry 42, 96–105. Frost, A., Perera, R., Roux, A., Spasov, K., Destaing, O., Egelman, E.H., De Camilli, P., and Unger, V.M. (2008). Structural basis of membrane invagination by F-BAR domains. Cell 132, 807–817. Gallop, J.L., Jao, C.C., Kent, H.M., Butler, P.J.G., Evans, P.R., Langen, R., and McMahon, H.T. (2006). Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J. 25, 2898–2910. George, M., Ying, G., Rainey, M.A., Solomon, A., Parikh, P.T., Gao, Q., Band, V., and Band, H. (2007). Shared as well as distinct roles of EHD proteins revealed by biochemical and functional comparisons in mammalian cells and C. elegans. BMC Cell Biol. 8, 3. Gräff, J., and Tsai, L.-H. (2013). Histone acetylation: molecular mnemonics on the chromatin. Nat. Rev. Neurosci. 14, 97–111. He, J., Haney, R.M., Vora, M., Verkhusha, V.V., Stahelin, R.V., and Kutateladze, T.G. (2008). Molecular mechanism of membrane targeting by the GRP1 PH domain. J. Lipid Res. 49, 1807–1815. Hegde, B.G., Isas, J.M., Zampighi, G., Haigler, H.T., and Langen, R. (2006). A novel calcium- independent peripheral membrane-bound form of annexin B12. Biochemistry 45, 934– 942. Henne, W.M., Kent, H.M., Ford, M.G.J., Hegde, B.G., Daumke, O., Butler, P.J.G., Mittal, R., Langen, R., Evans, P.R., and McMahon, H.T. (2007). Structure and analysis of FCHo2 F- 148 BAR domain: a dimerizing and membrane recruitment module that effects membrane curvature. Structure 1993 15, 839–852. Hong, L., Schroth, G.P., Matthews, H.R., Yau, P., and Bradbury, E.M. (1993). Studies of the DNA binding properties of histone H4 amino terminus. Thermal denaturation studies reveal that acetylation markedly reduces the binding constant of the H4 ―tail‖ to DNA. J. Biol. Chem. 268, 305–314. Hornbeck, P.V., Kornhauser, J.M., Tkachev, S., Zhang, B., Skrzypek, E., Murray, B., Latham, V., and Sullivan, M. (2012). PhosphoSitePlus: a comprehensive resource for investigating the structure and function of experimentally determined post-translational modifications in man and mouse. Nucleic Acids Res. 40, D261–D270. Hui, E., Gaffaney, J.D., Wang, Z., Johnson, C.P., Evans, C.S., and Chapman, E.R. (2011). Mechanism and function of synaptotagmin-mediated membrane apposition. Nat. Struct. Mol. Biol. 18, 813–821. Isas, J.M., Patel, D.R., Jao, C., Jayasinghe, S., Cartailler, J.-P., Haigler, H.T., and Langen, R. (2003). Global structural changes in annexin 12. The roles of phospholipid, Ca2+, and pH. J. Biol. Chem. 278, 30227–30234. Isas, J.M., Ambroso, M.R., Hegde, P.B., Langen, J., and Langen, R. (2015). Tubulation by amphiphysin requires a concentration-dependent switching from wedging to scaffolding. Structure in press. Itoh, T., Erdmann, K.S., Roux, A., Habermann, B., Werner, H., and De Camilli, P. (2005). Dynamin and the actin cytoskeleton cooperatively regulate plasma membrane invagination by BAR and F-BAR proteins. Dev. Cell 9, 791–804. Jao, C.C., Hegde, B.G., Gallop, J.L., Hegde, P.B., McMahon, H.T., Haworth, I.S., and Langen, R. (2010). Roles of amphipathic helices and the bin/amphiphysin/rvs (BAR) domain of endophilin in membrane curvature generation. J. Biol. Chem. 285, 20164–20170. Jung, A.G., Labarrera, C., Labarerra, C., Jansen, A.M., Qvortrup, K., Wild, K., and Kjaerulff, O. (2010). A mutational analysis of the endophilin-A N-BAR domain performed in living flies. PloS One 5, e9492. Kim, Y.E., Isas, J.M., Haigler, H.T., and Langen, R. (2005). A helical hairpin region of soluble annexin B12 refolds and forms a continuous transmembrane helix at mildly acidic pH. J. Biol. Chem. 280, 32398–32404. Kuo, W., Herrick, D.Z., Ellena, J.F., and Cafiso, D.S. (2009). The Calcium-Dependent and Calcium-Independent Membrane Binding of Synaptotagmin 1: Two Modes of C2B Binding. J. Mol. Biol. 387, 284–294. 149 Lai, C.-L., Srivastava, A., Pilling, C., Chase, A.R., Falke, J.J., and Voth, G.A. (2013). Molecular Mechanism of Membrane Binding of the GRP1 PH Domain. J. Mol. Biol. 425, 3073– 3090. Langen, R., Isas, J.M., Luecke, H., Haigler, H.T., and Hubbell, W.L. (1998). Membrane- mediated Assembly of Annexins Studied by Site-directed Spin Labeling. J. Biol. Chem. 273, 22453–22457. Lee, S.A., Kovacs, J., Stahelin, R.V., Cheever, M.L., Overduin, M., Setty, T.G., Burd, C.G., Cho, W., and Kutateladze, T.G. (2006). Molecular mechanism of membrane docking by the Vam7p PX domain. J. Biol. Chem. 281, 37091–37101. Lemmon, M.A. (2008). Membrane recognition by phospholipid-binding domains. Nat. Rev. Mol. Cell Biol. 9, 99–111. Lemmon, M.A., Ferguson, K.M., and Schlessinger, J. (1996). PH domains: diverse sequences with a common fold recruit signaling molecules to the cell surface. Cell 85, 621–624. Li, C., Ullrich, B., Zhang, J.Z., Anderson, R.G.W., Brose, N., and Südhof, T.C. (1995). Ca2+- dependent and -independent activities of neural and non-neural synaptotagmins. Nature 375, 594–599. Lomize, A.L., Pogozheva, I.D., Lomize, M.A., and Mosberg, H.I. (2007). The role of hydrophobic interactions in positioning of peripheral proteins in membranes. BMC Struct. Biol. 7, 44. Lumb, C.N., He, J., Xue, Y., Stansfeld, P.J., Stahelin, R.V., Kutateladze, T.G., and Sansom, M.S.P. (2011). Biophysical and Computational Studies of Membrane Penetration by the GRP1 Pleckstrin Homology Domain. Structure 1993 19, 1338–1346. Martens, S., Kozlov, M.M., and McMahon, H.T. (2007). How Synaptotagmin Promotes Membrane Fusion. Science 316, 1205–1208. Masuda, M., Takeda, S., Sone, M., Ohki, T., Mori, H., Kamioka, Y., and Mochizuki, N. (2006). Endophilin BAR domain drives membrane curvature by two newly identified structure- based mechanisms. EMBO J. 25, 2889–2897. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., De Bock, P.-J., Morais, V.A., Vilain, S., Haddad, D., et al. (2012). LRRK2 controls an EndoA phosphorylation cycle in synaptic endocytosis. Neuron 75, 1008–1021. Mim, C., Cui, H., Gawronski-Salerno, J.A., Frost, A., Lyman, E., Voth, G.A., and Unger, V.M. (2012). Structural basis of membrane bending by the N-BAR protein endophilin. Cell 149, 137–145. 150 Morén, B., Shah, C., Howes, M.T., Schieber, N.L., McMahon, H.T., Parton, R.G., Daumke, O., and Lundmark, R. (2012). EHD2 regulates caveolar dynamics via ATP-driven targeting and oligomerization. Mol. Biol. Cell 23, 1316–1329. Okumura, K., Mendoza, M., Bachoo, R.M., DePinho, R.A., Cavenee, W.K., and Furnari, F.B. (2006). PCAF modulates PTEN activity. J. Biol. Chem. 281, 26562–26568. Peter, B.J., Kent, H.M., Mills, I.G., Vallis, Y., Butler, P.J.G., Evans, P.R., and McMahon, H.T. (2004). BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 303, 495–499. Pilling, C., Landgraf, K.E., and Falke, J.J. (2011). GRP1 PH Domain, Like AKT1 PH Domain, Possesses a Sentry Glutamate Residue Essential for Specific Targeting to Plasma Membrane PI(3,4,5)P3. Biochemistry 50, 9845–9856. Prabakaran, S., Lippens, G., Steen, H., and Gunawardena, J. (2012). Post-translational modification: nature‘s escape from genetic imprisonment and the basis for dynamic information encoding. Wiley Interdiscip. Rev. Syst. Biol. Med. 4, 565–583. Qualmann, B., Koch, D., and Kessels, M.M. (2011). Let‘s go bananas: revisiting the endocytic BAR code. EMBO J. 30, 3501–3515. Rao, Y., and Haucke, V. (2011). Membrane shaping by the Bin/amphiphysin/Rvs (BAR) domain protein superfamily. Cell. Mol. Life Sci. 68, 3983–3993. Robert, X., and Gouet, P. (2014). Deciphering key features in protein structures with the new ENDscript server. Nucleic Acids Res. 42, W320–W324. Rufener, E., Frazier, A.A., Wieser, C.M., Hinderliter, A., and Cafiso, D.S. (2005). Membrane- bound orientation and position of the synaptotagmin C2B domain determined by site- directed spin labeling. Biochemistry 44, 18–28. Saarikangas, J., Zhao, H., Pykäläinen, A., Laurinmäki, P., Mattila, P.K., Kinnunen, P.K.J., Butcher, S.J., and Lappalainen, P. (2009). Molecular mechanisms of membrane deformation by I-BAR domain proteins. Curr. Biol. CB 19, 95–107. Shah, C., Hegde, B.G., Morén, B., Behrmann, E., Mielke, T., Moenke, G., Spahn, C.M.T., Lundmark, R., Daumke, O., and Langen, R. (2014). Structural Insights into Membrane Interaction and Caveolar Targeting of Dynamin-like EHD2. Structure 1993 22, 409–420. Shimada, A., Niwa, H., Tsujita, K., Suetsugu, S., Nitta, K., Hanawa-Suetsugu, K., Akasaka, R., Nishino, Y., Toyama, M., Chen, L., et al. (2007). Curved EFC/F-BAR-domain dimers are joined end to end into a filament for membrane invagination in endocytosis. Cell 129, 761–772. 151 Sievers, F., Wilm, A., Dineen, D., Gibson, T.J., Karplus, K., Li, W., Lopez, R., McWilliam, H., Remmert, M., Söding, J., et al. (2011). Fast, scalable generation of high ‐quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7, 539. Stahelin, R.V., Burian, A., Bruzik, K.S., Murray, D., and Cho, W. (2003). Membrane Binding Mechanisms of the PX Domains of NADPH Oxidase p40 phox and p47 phox. J. Biol. Chem. 278, 14469–14479. Stahelin, R.V., Scott, J.L., and Frick, C.T. (2014). Cellular and molecular interactions of phosphoinositides and peripheral proteins. Chem. Phys. Lipids 182, 3–18. Takeda, T., Robinson, I.M., Savoian, M.M., Griffiths, J.R., Whetton, A.D., McMahon, H.T., and Glover, D.M. (2013). Drosophila F-BAR protein Syndapin contributes to coupling the plasma membrane and contractile ring in cytokinesis. Open Biol. 3. Ward, K.E., Bhardwaj, N., Vora, M., Chalfant, C.E., Lu, H., and Stahelin, R.V. (2013). The molecular basis of ceramide-1-phosphate recognition by C2 domains. J. Lipid Res. 54, 636–648. Weissenhorn, W. (2005). Crystal structure of the endophilin-A1 BAR domain. J. Mol. Biol. 351, 653–661. Zhao, X., Wang, D., Liu, X., Liu, L., Song, Z., Zhu, T., Adams, G., Gao, X., Tian, R., Huang, Y., et al. (2013). Phosphorylation of the Bin, Amphiphysin, and RSV161/167 (BAR) domain of ACAP4 regulates membrane tubulation. Proc. Natl. Acad. Sci. U. S. A. 110, 11023– 11028. Zhou, B.-R., Feng, H., Ghirlando, R., Kato, H., Gruschus, J., and Bai, Y. (2012). Histone H4 K16Q mutation, an acetylation mimic, causes structural disorder of its N-terminal basic patch in the nucleosome. J. Mol. Biol. 421, 30–37. 152 CHAPTER 5 Myelin Basic Protein is a Membrane Curving Protein Inhibited by Citrullination Mark Robert Ambroso 1 , Jonah R. Chan 2 and Ralf Langen 1, * 1 Zilhka Neurogenetic Institute, University of Southern California, Los Angeles, California 90033 USA. 2 Department of Neurology and Program in Neuroscience, University of California, San Francisco, San Francisco, California, USA. *Corresponding Author: Ralf Langen, Zilkha Neurogenetic Institute, 1501 San Pablo St., 121 ZNI, Los Angeles, CA 90033. Tel.: 323-442-1323 Fax: 323-442-4404. Email langen@usc.edu 153 CHAPTER 5 ABSTRACT Myelin basic protein (MBP) is a major component of myelin required for the formation of healthy myelin wraps. MBP is an intrinsically disordered peptide which structures into three segmented α-helices on membranes. While MBP has been proposed to act as an intermembrane fusing agent between leaflets of healthy myelin, the exact function of MBP in the myelination process is not fully understood. In this study, we show that recombinant human MBP curves lipid membranes in a manner akin to other intrinsically disordered peptides which structure into α-helices on membranes such as the synucleins. We observe MBP to generate cylindrical tubes when incubated with lipid vesicles composed of a myelin-like lipid composition. We observe MBP-coated tubes to wrap or adhere to neighboring vesicles and tubes. Under certain conditions MBP-induced tubulation can generate large multi-layer networks of membrane. Finally, the reduction of the net positive charge of MBP by post-translational modification, including the Multiple sclerosis (MS)-related modification citrullination, results in its decreased potency in curvature generation. It therefore may be possible that MBP acts both to bend and to fuse membrane leaflets in myelin and that these functions are perturbed in disease. 154 5.1 Introduction A large number of cellular events require the remodeling of lipid membranes (McMahon and Gallop, 2005; Qualmann et al., 2011). These events are most often executed by proteins capable of curving lipid membranes into specific shapes. Mutations and aberrant post-translation modification (PTM) of proteins responsible for these events are implicated in disease (Bergmann et al., 2003; Matta et al., 2012; Nicot et al., 2007). While the importance of membrane curvature in processes such as endo- and exocytosis is well studied (Doherty and McMahon, 2009), its importance in alternative events involving the organization and shape of membranes, such as axonal myelination, has yet to be determined. Rapid nerve signaling and saltatory conduction require axons to be tightly wrapped by layers of myelin membrane (Buttermore et al., 2013). The canonical features of nervous system myelin are the regularly spaced major dense and intraperiod lines, representing condensed cytoplasmic myelin membrane and apposed outer membrane respectively (Yoshikawa, 2001). The total composition of nervous system myelin is largely made up of lipids and two proteins, the proteolipid protein and myelin basic protein (MBP). MBP is a significant component of the major dense line and MBP-knockout results in loosely compacted myelin sheaths and reduced neuronal capacities in mice (Nave, 1994). MBP has long been known as a lipid aggregator (Cheifetz and Moscarello, 1985; Mac Millan et al., 2000; Sridhara et al., 1984) and proposed to function as a membrane fusing agent between layers of compact myelin. It has also been shown to interact with cytoskeletal assemblies, SH3-domains, and Fyn-mediated signaling pathways (Harauz et al., 2009). Recently it has been suggested that MBP plays an additional role in myelination by acting as a molecular sieve restricting protein diffusion into the compact myelin (Aggarwal et al., 2011). 155 The most common splice variant of MBP in healthy myelin is an 18.5 kDa intrinsically disordered and highly basic peptide. Biophysical studies have shown this peptide to bind membranes at high affinities and aggregate vesicles (Mac Millan et al., 2000). MBP has been alternatively observed under specific protein-to-lipid ratios and levels of ionic strength MBP has been observed to convert large vesicles into small uniformly-sized vesicles (Boggs et al., 1999; Sridhara et al., 1984), a function common to many membrane curving proteins (Bhatia et al., 2009; Boucrot et al., 2012; Gallop et al., 2006; Lai et al., 2012; Varkey et al., 2010). MBP binds membranes and structures three separate regions of its sequence into α-helices. Similarly, both islet amyloid polypeptide (Apostolidou et al., 2008) and α-synuclein (Jao et al., 2008) are intrinsically disordered membrane-curving proteins known to fold into α-helical structures upon interaction with membranes. There are several well-characterized mechanisms by which proteins generate membrane curvature, including scaffolding (Qualmann et al., 2011), wedging (Drin and Antonny, 2010), protein-crowding (Stachowiak et al., 2012), and bilayer couple (Farsad and Camilli, 2003). The wedging mechanism involves the insertion of a protein segment into the membrane thereby increasing the local packing density and generating a lateral dispersion force. The most well studied protein wedge is the amphipathic helix (Drin and Antonny, 2010). The arrangement of amino acids in amphipathic helices forms opposing hydrophobic and hydrophilic faces across the plane of the helical axis. This conformation allows for the hydrophobic side to associate into the acyl chains of the bilayer and the hydrophilic face with the polar lipid headgroups. As a helix inserts asymmetrically into a leaflet of a bilayer it generates a lateral splitting force across the plane of the membrane. The stressed leaflet then must bend to accommodate the insertion. This affect has been calculated to be correlated to the depth of the helical insertion (Campelo et al., 156 2008). Structural studies on MBP have elucidated its membrane binding α-helices to be amphipathic and insert into the membrane at similar depths to the amphipathic helices of known membrane-curving proteins amphiphysin, endophilin, α-synuclein, and epsin (Ambroso et al., 2014; Campelo et al., 2010; Gallop et al., 2006; Isas et al., 2015; Jao et al., 2008, 2010; Lai et al., 2012; Varkey et al., 2010). These proteins are capable of vesiculation as well as tubulation, or the generation of cylindrical lipid shapes with radial diameters of 20-100 nm (Ambroso et al., 2014; Boucrot et al., 2012; Cui et al., 2013; Farsad et al., 2001; Mim et al., 2012; Mizuno et al., 2010; Peter et al., 2004; Varkey et al., 2010; Zhu et al., 2012). Inasmuch as MBP shares many characteristics with known membrane curving proteins, this function has yet to be tested. The importance of this protein in maintaining healthy myelin is further solidified by the fact that degree of aberrant post-translation modification of MBP has been correlated to the severity of MS (Musse and Harauz, 2007). Citrullination of arginines results in the loss of a positive charge and has therefore been postulated to affect the way MBP electrostatically interacts with negatively charged membranes. Structural comparison between membrane bound MBP when citrullinated or uncitrullinated revealed the modification to decrease the depth by which MBP penetrates into membrane, thereby exposing an immunodominant epitope (Musse et al., 2006). Interestingly, charge-modulation and differential insertion depths of amphipathic helices have been observed to directly affect the manner in which they shape membranes (Ambroso et al., 2014; Zhao et al., 2013). In this study we show that MBP is a potent membrane curving protein capable of not only inducing vesiculation of large vesicles but also membrane tubes. Fluorescent leakage experiments show a dose-dependent increase in the disruption of model membrane integrity, similar to other known membrane curving proteins (Varkey et al., 2010). Interestingly, when 157 MBP is added to dried lipids films and thereby exposed to both the inner and outer leaflets of the membrane, it can generate lipid tubes capable of ‗wrapping‘ neighboring vesicles and tubes. Finally, citrullination-mimicking constructs of MBP which recapitulate the most common form of MBP found in MS patients (Bates et al., 2002) are less potent curvature-inducers compared to that of the wild type protein. 5.2 Results 5.2.1 MBP is a potent inducer of membrane curvature. To determine whether MBP is capable of generating membrane curvature we incubated MBP with lipid vesicles (Figure 5.1, top left) under a variety of conditions. Using negative-stain transmission electron microscopy (TEM) we observed several membrane binding modes of MBP. When MBP was exogenously added to large multilamellar vesicles composed of a lipid composition mimicking healthy myelin at low protein concentrations (molar protein:lipid ratio < 1:500) we observed lipid aggregates on the order of several microns (Figure 5.1, top right). In contrast, high protein concentrations (molar protein:lipid > 1:50) caused the large multilamellar vesicles to be vesiculated into smaller vesicles with an average diameter of 30 nm (Figure 5.1, middle left). Close examination revealed a tendency of neighboring vesicles to form flat interaction surfaces (Figure S5.1A). At concentrations between the aggregation and vesiculation states (protein:lipid 1:100-350), a mixture of aggregates and vesicles were observed alongside a small population of membrane tubes with an average diameter of 40 nm (Figure 5.1, middle right). Membrane tubes are a common membrane shape generated by proteins capable of inducing membrane curvature (Ambroso et al., 2014; Boucrot et al., 2012; Cui et al., 2013; Farsad et al., 2001; Mim et al., 2012; Mizuno et al., 2010; Peter et al., 2004; Varkey et al., 2010; Zhu et al., 2012), suggesting that MBP may share a similar function. 158 Figure 5.1. MBP alone can remodel large myelin-like vesicles into different structures. Top left: lipid vesicles with a lipid composition similar to that of healthy myelin in the absence of protein as observed by negative-stain electron microscopy. Top right: incubating low amounts of MBP with pre-solubalized myelin-like lipids results in the formation of large protein-vesicle aggregates. Middle left: increased protein to lipid molar ratios (1:50) results in the remodeling of large vesicles into small vesicles. Adding exogenous MBP to solubalized lipid vesicles with the same lipid composition used in the previous experiments resulted in membrane tubulation at specific protein to lipid molar ratios; 1:200 (middle right), 1:100 (bottom left), and 1:50 (bottom right). The scale bar = 500 μm. 159 5.2.2 MBP-coated tubules exhibit vesicle-wrapping properties and generate multi-layered Figure 5.1, continued 160 5.2.2 MBP-coated Tubules Exhibit Vesicle-Wrapping Properties and Generate Multi- Layered Networks. While mild levels of membrane tubulation were observed for MBP externally added to vesicles, it is unclear what affect MBP would have if it was exposed to both the inner and outer leaflet of these myelin-like vesicles. To examine this, dried films of myelin-like vesicles were suspended in a buffer solution (20 mM Hepes, pH 7.4; 100 mM NaCl) containing MBP. Under these conditions, MBP generated extensive membrane tubulation with average diameters of around 40 nm (Figure 5.2 top left). Similar to the vesicles formed at high protein:lipid molar ratios, MBP-induced tubes were observed to have a high affinity for adhering to each other or even wrapping around small vesicles (Figure 5.2 top right). Tubes were observed to prefer wrapping around vesicles with diameters less than or equal to 150 nm (Figure 5.2 bottom). These data suggest that MBP both generates curvature and acts in intermembrane adhesion. Under lower protein density (1:350 protein:lipid molar ratio), MBP solubalization of lipid films induced tubulation with a high propensity to form into large and closely adhered ‗wraps‘ (Figure 5.3 top left). These networks often formed around vesicles and extended outward in a circular fashion (Figure 5.3 top right). High magnification of these networks revealed a dense stain line and a lighter stain line signifying the lipid bilayer and the coat of MBP protein respectively (Figure 5.3 bottom). The combination of MBP‘s ability to generate membrane curvature and cause intermembrane adhesion may play a significant role in the maintenance of myelin sheaths. 161 Figure 5.2. Lipid tubes formed by MBP exhibit the ability to wrap vesicles. (A-C) MBP in buffer was used to solubalize dried lipid films with a lipid composition mimicking that of healthy myelin at a 1:100 protein to lipid molar ratio and observed using negative-stain electron microscopy. Small vesicles observed to be encircled by the lipid tubes are highlighted with arrows. Scale bar represents 200 μm. 162 Figure 5.3 MBP can generate mutli-layer wraps of membrane tubes. (A-C) Negative-stain electron microscopy of MBP incubated with dried lipid films with a lipid composition mimicking that of healthy myelin at a 1:350-500 protein to lipid molar ratio. Lipid tubes consisting of similar thickness closely adhere to one another. Scale bar represents 500 μm. Upon close examination a distinct protein coat is observed for closely adhered tubes (C). Scale bar represents 100 μm. 163 5.2.3 MBP generates membrane curvature in a similar fashion to known membrane curving proteins. To quantitatively measure the level of membrane curvature generated by MBP we utilized a well established fluorescence leakage assay (Varkey et al., 2010). At high protein concentrations (protein:lipid 1:10) only minimal vesicle leakage (~5%) was observed for myelin- like vesicles (Figure 5.4A) suggesting that the membrane integrity remains largely intact in this lipid system. However, when similar experiments were done using simplified lipid system containing high percentages of negatively charged lipids (66:33 PS:PC molar ratio), MBP was shown to potently disrupt membrane integrity in a concentration dependent manner (Figure 5.4B). nsTEM of samples from the fluorescent leakage assay reveal the 100 nm vesicles used in the study to be vesiculated into smaller vesicles (Figure S5.1). Similar disruption of model membrane integrity has been observed for the known membrane curving protein α-synuclein (Varkey et al., 2010). 164 Figure 5.4 MBP disrupts the integrity of model membranes. Lipid vesicles (66:33 PS:PC molar ratio) extruded to 100 nm in diameter and containing the fluorophore 8-aminonaphthalene- 1,3,6-trisulfonate (ANTS) and fluorophore-quencher p-xylene-bis-pyridinium bromide (DPX) were incubated with MBP in a 1:200 (magenta), 1:100 (blue), 1:50 (red), and 1:25 (black) protein:lipid molar ratio. Disruption of the lipid vesicle integrity causes its contents to diffuse into the extravesicular solution, thereby distancing the quencher and fluorophore from one another and increasing fluorescence. Error bars represent SD, n= 3. Figure S5.1 Vesiculation of 100 nm diameter vesicles 165 5.2.4 Electrostatic interactions regulate the level of membrane curvature MBP induces. It has been previously reported that MBP‘s interaction with membranes is dependent on ionic strength and the net charge of MBP (Boggs et al., 1997). To examine whether the charge of MBP affects its ability to remodel membranes, we used a well established cation exchange protocol (Chou et al., 1976) to separate differently charged analogs of MBP from bovine tissue (bMBP). We then tested the most charged fraction with the least charge in the quantitative fluorescent leakage assay. Compared to the least charged fraction, the most charged fraction of bMBP consistently caused a higher percentage of membrane disruption at multiple different protein to lipid molar ratios (Figure 5.5A). Thus the charge of the membrane and of MBP itself regulates MBP‘s ability to shape and remodel lipids. Moreover, negative-stain TEM imaging revealed these samples to be highly vesiculated confirming that the loss of membrane integrity corresponded to the induction of membrane curvature (Figure 5.5B). 166 Figure 5.5. Correlations between membrane leakage and MBP’s net charge. Differently charged fractions of MBP from tissue extracts were separated using cation exchange chromatography (Chou et al., 1976) and incubated with dye-containing multilamellar lipid vesicles (66:33 PG:PE molar ratio) and monitored for increases in fluorescence over time. 167 5.2.5 Disease state MBP generates a reduced level of membrane curvature. While we have already established that electrostatic interactions as well as membrane and protein charge are important modulators of MBP‘s ability to induce curvature, it is still unclear how these interactions are related to disease. To investigate whether aberrant post-translational modification of MBP known to be associated with MS progression affects the protein‘s ability to generate curvature, we generated an established citrulline-mimick peptide construct (MBP-C6) and thereby removing six positive charges from MBP (Figure 5.6A) (Bates et al., 2002). Importantly, recombinant MBP has been previously found to have similar characteristics to MBP extracted from tissue (Bates et al., 2000). We then tested this constructs ability to remodel lipid membranes using the fluorescent leakage assay (Figure 5.6B). MBP-C6 caused less total fluorescent leakage than the wt construct (Figure 5.6C). Thus, a reduction in charge in the citrullination mimicks inhibits membrane tubulation but decreases MBP‘s capacity to disrupt membrane integrity. 168 Figure 5.6. Citrullination-mimicking mutations reduce the ability of MBP to induce curvature on model membranes. (A) A sequence comparison (Clustal Omega) of the constructs used in our experiments; wild type human MBP (C1) and a construct with six arginines mutated to glutamines (C6). Arg-Gln mutations are highlighted in red. The arginines were mutated to mimick the arigines most commonly citrullinated in MS (Bates et al., 2002). (B) Time-dependent increase in fluorescence from dye-containing vesicles after incubation with the two different constructs of MBP discussed in (A). (C) The total maximum increase in fluorescence from dye-containing vesicles after incubation with the C1 or C6 construct. Error bars represent SD, n=3 independent experiments, *p<0.05. 5.3 Discussion 169 5.3 Discussion In this study, we show that MBP is a potent membrane curving protein capable of inducing aggregation, vesiculation, and tubulation in both model and myelin-mimicking lipids systems. A dose-dependent relationship of membrane curvature and increased MBP protein:lipid molar ratio was observed. These results are in good agreement with those observed for α- synuclein, an intrinsically disordered membrane curving protein that forms an amphipathic helix when associating with membranes (Jao et al., 2008; Varkey et al., 2010, 2013; Westphal and Chandra, 2013). MBP-coated tubes were found to be capable of wrapping vesicles and self- association into large patterned networks. Analysis of decreasingly charged constructs of MBP and membrane lipids suggests that MBP‘s ability to induce curvature is dependent on its electrostatic interactions with the membrane. Our data also suggests that MS-associated citrullination-mimicks of MBP generate decreased amounts of membrane curvature compared to wild type. MBP‘s ability to generate membrane curvature may therefore be an important function in generating healthy myelin sheaths. Small-angle X-ray scattering and NMR data suggest that MBP binds membranes by inserting three-different amphipathic helical segments into the interfacial region between the hydrophobic and aqueous phases. Detailed structural characterization of the individual amphipathic helical segments of MBP has shown that T33-D46 (Libich and Harauz, 2008), V83- T92 (Ahmed et al., 2012; Musse et al., 2006), and Y142-L154 (Homchaudhuri et al., 2010) insert into the membrane at the layer of the lipid phosphates. This conformation is highly similar to the structure of membrane bound amphipathic helices from known membrane-curving proteins Epsin, N-BAR proteins, and α-synuclein. At these depths, these helices have been predicted to generate optimal levels of spontaneous curvature (Campelo et al., 2008). An important 170 distinction between the amphipathic helices formed by MBP and other membrane curving proteins is the helical extension. While proteins like synuclein form a single extended helix on membranes larger than 50 nm, MBP consists of three distinct helices connected by flexible linkers. This suggests that MBP is more likely to induce isotropic curvature as its helices can adopt a large number of different alignments in respect to one another. Still, in this study we were able to find conditions in which MBP induced membrane tubes which are generated by anisotropic curvature. One of MBP‘s most defining characteristics is its highly basic nature (Harauz et al., 2009). The net charge of MBP is regulated in vivo by charge-modifying post-translational modifications. The manipulation of MBP‘s charge has been thoroughly investigated in vitro and observed to regulate its ability to aggregate and fuse membranes (Boggs et al., 1997). We observe here that MBP induces increasing amounts of membrane curvature with increasing overall positive charge. Analysis of MBP extracted from MS patients has shown a correlation between the severity of disease and MBP arginine citrullination. Our studies show that citrulline- mimicking constructs of MBP results in decreased ability to curve membranes. Future studies are required to determine whether MBP induced curvature and its regulation by citrullination plays an important role in myelination and disease. 171 5.4 Methods 5.4.1 MBP expression and purification Gene blocks encoding the 18.5 kDa isoform of human MBP (C1) and (C6) (Bates et al., 2002) were purchased from GenScript and inserted into a pET28b(-) vector with a C-terminal His 6 tag using 5‘ NcoI and 3‘ XhoI cut sites. Protein expression and purification of MBP (C1) and (C6) was performed as previously described (Bates et al., 2000, 2002) with a few notable adjustments. Plasmids were transformed into E. coli BL21(DE3)pLysS cells, expanded to an OD of 0.8, and induced with 1 mM IPTG for 16 hours at 22 °C. Cell pellets were dissolved in 6 M urea solution and proteins were purified on nickel-nitrilo-triacetic acid-agarose beads, dialyzed into a low ionic strength buffer, and further purified on a CM52 cation exchange column. Proteins were refolded through slow dialysis into 20 mM Hepes, pH 7.4 buffer, lyophilized, and stored at -20 °C. 5.4.2 Vesicle preparation All of the following lipids were used in this study: 1-palmitoyl-2-oleoyl-sn-glycero-3- phosphoethanolamine (POPE), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS), 1- palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), and 1-palmitoyl-2-oleoyl-sn-glycero- 3-[phospho-RAC-(1-glycerol)] (POPG), cholesterol, sphingomyelin, and phosphoinositol (Avanti Polar Lipids Inc.). Myelin-like vesicles were composed of 44% cholesterol, 27% POPE, 13% POPS, 11% POPC, 3% sphingomyelin, 2% phosphoinositol as previously described (Bates et al., 2004). Model lipid vesicles were made from a molar ratio of 2:1 PS/PC or PG/PE. Lipids were resuspended in a 20 mM Hepes, pH 7.4 buffered solution. Multi-lamellar vesicles used in electron microscopy studies were untreated beyond resuspension of lipids films. 172 5.4.3 Electron microscopy Small aliquots of samples (~10 μl) were dried onto carbon-coated formvar films mounted on copper grids (Electron Microscopy Sciences) and incubated for 5 to 20 minutes. The grids where then negatively stained using droplets of 1% uranyl acetate. Grids were mounted into a JEOL 1400 transmission electron microscope accelerated to 100 kV. 5.4.4 Dye leakage assay Leakage vesicle assays were performed as previously described (Varkey et al., 2010). In short, dried lipid films composed of a myelin-like lipid composition (see above), 2:1 POPG:POPE, or 2:1 POPS:POPC were resolubalized in 9 mM 8-aminonaphthalene-1,3,6-trisulfonic acid (ANTS) and 25 mM p-xylene-bis-pyridinium bromide (DPX) (Invitrogen). Resuspended lipids were extruded for twenty iterations through polycarbonate membrane filters with cutoff pores of either 1000 or 100 nm in size (Avanti Polar Lipids inc.). Extravesicular dye was removed by applying the vesicle solution to columns fitted with superdex 200 gel filtration resin (GE). All data were normalized to the maximal fluorescence of each experiment, measured by the addition of Triton- X 100 to a final concentration of 0.04% (Sigma). Fluorescence was measured using a JASCO FP-6500 fluorometer with the excitation and emission wavelengths set to 380 and 520 nm, respectively. 173 CHAPTER 5 REFERENCES Aggarwal, S., Yurlova, L., Snaidero, N., Reetz, C., Frey, S., Zimmermann, J., Pähler, G., Janshoff, A., Friedrichs, J., Müller, D.J., et al. (2011). A size barrier limits protein diffusion at the cell surface to generate lipid-rich myelin-membrane sheets. Dev. Cell 21, 445–456. Ahmed, M.A.M., De Avila, M., Polverini, E., Bessonov, K., Bamm, V.V., and Harauz, G. (2012). Solution nuclear magnetic resonance structure and molecular dynamics simulations of a murine 18.5 kDa myelin basic protein segment (S72-S107) in association with dodecylphosphocholine micelles. Biochemistry 51, 7475–7487. Ambroso, M.R., Hegde, B.G., and Langen, R. (2014). Endophilin A1 induces different membrane shapes using a conformational switch that is regulated by phosphorylation. Proc. Natl. Acad. Sci. 111, 6982–6987. Apostolidou, M., Jayasinghe, S.A., and Langen, R. (2008). Structure of alpha-helical membrane- bound human islet amyloid polypeptide and its implications for membrane-mediated misfolding. J. Biol. Chem. 283, 17205–17210. Bates, I.R., Matharu, P., Ishiyama, N., Rochon, D., Wood, D.D., Polverini, E., Moscarello, M.A., Viner, N.J., and Harauz, G. (2000). Characterization of a recombinant murine 18.5-kDa myelin basic protein. Protein Expr. Purif. 20, 285–299. Bates, I.R., Libich, D.S., Wood, D.D., Moscarello, M.A., and Harauz, G. (2002). An Arg/Lys-- >Gln mutant of recombinant murine myelin basic protein as a mimic of the deiminated form implicated in multiple sclerosis. Protein Expr. Purif. 25, 330–341. Bates, I.R., Feix, J.B., Boggs, J.M., and Harauz, G. (2004). An Immunodominant Epitope of Myelin Basic Protein Is an Amphipathic α-Helix. J. Biol. Chem. 279, 5757–5764. Bergmann, C., Zerres, K., Senderek, J., Rudnik-Schoneborn, S., Eggermann, T., Häusler, M., Mull, M., and Ramaekers, V.T. (2003). Oligophrenin 1 (OPHN1) gene mutation causes syndromic X-linked mental retardation with epilepsy, rostral ventricular enlargement and cerebellar hypoplasia. Brain J. Neurol. 126, 1537–1544. Bhatia, V.K., Madsen, K.L., Bolinger, P.-Y., Kunding, A., Hedegård, P., Gether, U., and Stamou, D. (2009). Amphipathic motifs in BAR domains are essential for membrane curvature sensing. EMBO J. 28, 3303–3314. Boggs, J.M., Yip, P.M., Rangaraj, G., and Jo, E. (1997). Effect of posttranslational modifications to myelin basic protein on its ability to aggregate acidic lipid vesicles. Biochemistry 36, 5065–5071. 174 Boggs, J.M., Rangaraj, G., Koshy, K.M., Ackerley, C., Wood, D.D., and Moscarello, M.A. (1999). Highly deiminated isoform of myelin basic protein from multiple sclerosis brain causes fragmentation of lipid vesicles. J. Neurosci. Res. 57, 529–535. Boucrot, E., Pick, A., Çamdere, G., Liska, N., Evergren, E., McMahon, H.T., and Kozlov, M.M. (2012). Membrane fission is promoted by insertion of amphipathic helices and is restricted by crescent BAR domains. Cell 149, 124–136. Buttermore, E.D., Thaxton, C.L., and Bhat, M.A. (2013). Organization and maintenance of molecular domains in myelinated axons. J. Neurosci. Res. 91, 603–622. Campelo, F., McMahon, H.T., and Kozlov, M.M. (2008). The hydrophobic insertion mechanism of membrane curvature generation by proteins. Biophys. J. 95, 2325–2339. Campelo, F., Fabrikant, G., McMahon, H.T., and Kozlov, M.M. (2010). Modeling membrane shaping by proteins: focus on EHD2 and N-BAR domains. FEBS Lett. 584, 1830–1839. Cheifetz, S., and Moscarello, M.A. (1985). Effect of bovine basic protein charge microheterogeneity on protein-induced aggregation of unilamellar vesicles containing a mixture of acidic and neutral phospholipids. Biochemistry 24, 1909–1914. Chou, F.C., Chou, C.H., Shapira, R., and Kibler, R.F. (1976). Basis of microheterogeneity of myelin basic protein. J. Biol. Chem. 251, 2671–2679. Cui, H., Mim, C., Vázquez, F.X., Lyman, E., Unger, V.M., and Voth, G.A. (2013). Understanding the role of amphipathic helices in N-BAR domain driven membrane remodeling. Biophys. J. 104, 404–411. Doherty, G.J., and McMahon, H.T. (2009). Mechanisms of endocytosis. Annu. Rev. Biochem. 78, 857–902. Drin, G., and Antonny, B. (2010). Amphipathic helices and membrane curvature. FEBS Lett. 584, 1840–1847. Farsad, K., and Camilli, P.D. (2003). Mechanisms of membrane deformation. Curr. Opin. Cell Biol. 15, 372–381. Farsad, K., Ringstad, N., Takei, K., Floyd, S.R., Rose, K., and De Camilli, P. (2001). Generation of high curvature membranes mediated by direct endophilin bilayer interactions. J. Cell Biol. 155, 193–200. Gallop, J.L., Jao, C.C., Kent, H.M., Butler, P.J.G., Evans, P.R., Langen, R., and McMahon, H.T. (2006). Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J. 25, 2898–2910. 175 Harauz, G., Ladizhansky, V., and Boggs, J.M. (2009). Structural Polymorphism and Multifunctionality of Myelin Basic Protein. Biochemistry 48, 8094–8104. Homchaudhuri, L., De Avila, M., Nilsson, S.B., Bessonov, K., Smith, G.S.T., Bamm, V.V., Musse, A.A., Harauz, G., and Boggs, J.M. (2010). Secondary Structure and Solvent Accessibility of a Calmodulin-Binding C-Terminal Segment of Membrane-Associated Myelin Basic Protein. Biochemistry 49, 8955–8966. Isas, J.M., Ambroso, M.R., Hegde, P.B., Langen, J., and Langen, R. (2015). Tubulation by amphiphysin requires a concentration-dependent switching from wedging to scaffolding. Structure in press. Jao, C.C., Hegde, B.G., Chen, J., Haworth, I.S., and Langen, R. (2008). Structure of membrane- bound alpha-synuclein from site-directed spin labeling and computational refinement. Proc. Natl. Acad. Sci. U. S. A. 105, 19666–19671. Jao, C.C., Hegde, B.G., Gallop, J.L., Hegde, P.B., McMahon, H.T., Haworth, I.S., and Langen, R. (2010). Roles of amphipathic helices and the bin/amphiphysin/rvs (BAR) domain of endophilin in membrane curvature generation. J. Biol. Chem. 285, 20164–20170. Lai, C.-L., Jao, C.C., Lyman, E., Gallop, J.L., Peter, B.J., McMahon, H.T., Langen, R., and Voth, G.A. (2012). Membrane binding and self-association of the epsin N-terminal homology domain. J. Mol. Biol. 423, 800–817. Libich, D.S., and Harauz, G. (2008). Backbone Dynamics of the 18.5 kDa Isoform of Myelin Basic Protein Reveals Transient α-Helices and a Calmodulin-Binding Site. Biophys. J. 94, 4847–4866. Mac Millan, S.V., Ishiyama, N., White, G.F., Palaniyar, N., Hallett, F.R., and Harauz, G. (2000). Myelin basic protein component C1 in increasing concentrations can elicit fusion, aggregation, and fragmentation of myelin-like membranes. Eur. J. Cell Biol. 79, 327– 335. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., De Bock, P.-J., Morais, V.A., Vilain, S., Haddad, D., et al. (2012). LRRK2 controls an EndoA phosphorylation cycle in synaptic endocytosis. Neuron 75, 1008–1021. McMahon, H.T., and Gallop, J.L. (2005). Membrane curvature and mechanisms of dynamic cell membrane remodelling. Nature 438, 590–596. Mim, C., Cui, H., Gawronski-Salerno, J.A., Frost, A., Lyman, E., Voth, G.A., and Unger, V.M. (2012). Structural basis of membrane bending by the N-BAR protein endophilin. Cell 149, 137–145. 176 Mizuno, N., Jao, C.C., Langen, R., and Steven, A.C. (2010). Multiple Modes of Endophilin- mediated Conversion of Lipid Vesicles into Coated Tubes. J. Biol. Chem. 285, 23351– 23358. Musse, A.A., and Harauz, G. (2007). Molecular ―negativity‖ may underlie multiple sclerosis: role of the myelin basic protein family in the pathogenesis of MS. Int. Rev. Neurobiol. 79, 149–172. Musse, A.A., Boggs, J.M., and Harauz, G. (2006). Deimination of membrane-bound myelin basic protein in multiple sclerosis exposes an immunodominant epitope. Proc. Natl. Acad. Sci. U. S. A. 103, 4422–4427. Nave, K.A. (1994). Neurological mouse mutants and the genes of myelin. J. Neurosci. Res. 38, 607–612. Nicot, A.-S., Toussaint, A., Tosch, V., Kretz, C., Wallgren-Pettersson, C., Iwarsson, E., Kingston, H., Garnier, J.-M., Biancalana, V., Oldfors, A., et al. (2007). Mutations in amphiphysin 2 (BIN1) disrupt interaction with dynamin 2 and cause autosomal recessive centronuclear myopathy. Nat. Genet. 39, 1134–1139. Peter, B.J., Kent, H.M., Mills, I.G., Vallis, Y., Butler, P.J.G., Evans, P.R., and McMahon, H.T. (2004). BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 303, 495–499. Qualmann, B., Koch, D., and Kessels, M.M. (2011). Let‘s go bananas: revisiting the endocytic BAR code. EMBO J. 30, 3501–3515. Sridhara, S.I., Epand, R.M., and Moscarello, M.A. (1984). Phospholipid vesicle aggregation induced by human myelin basic protein. Neurochem. Res. 9, 241–248. Stachowiak, J.C., Schmid, E.M., Ryan, C.J., Ann, H.S., Sasaki, D.Y., Sherman, M.B., Geissler, P.L., Fletcher, D.A., and Hayden, C.C. (2012). Membrane bending by protein-protein crowding. Nat. Cell Biol. 14, 944–949. Varkey, J., Isas, J.M., Mizuno, N., Jensen, M.B., Bhatia, V.K., Jao, C.C., Petrlova, J., Voss, J.C., Stamou, D.G., Steven, A.C., et al. (2010). Membrane curvature induction and tubulation are common features of synucleins and apolipoproteins. J. Biol. Chem. 285, 32486– 32493. Varkey, J., Mizuno, N., Hegde, B.G., Cheng, N., Steven, A.C., and Langen, R. (2013). α- Synuclein oligomers with broken helical conformation form lipoprotein nanoparticles. J. Biol. Chem. 288, 17620–17630. Westphal, C.H., and Chandra, S.S. (2013). Monomeric synucleins generate membrane curvature. J. Biol. Chem. 288, 1829–1840. 177 Yoshikawa, H. (2001). Myelin-associated oligodendrocytic basic protein modulates the arrangement of radial growth of the axon and the radial component of myelin. Med. Electron Microsc. 34, 160–164. Zhao, X., Wang, D., Liu, X., Liu, L., Song, Z., Zhu, T., Adams, G., Gao, X., Tian, R., Huang, Y., et al. (2013). Phosphorylation of the Bin, Amphiphysin, and RSV161/167 (BAR) domain of ACAP4 regulates membrane tubulation. Proc. Natl. Acad. Sci. U. S. A. 110, 11023– 11028. Zhu, C., Das, S.L., and Baumgart, T. (2012). Nonlinear sorting, curvature generation, and crowding of endophilin N-BAR on tubular membranes. Biophys. J. 102, 1837–1845. 178 APPENDIX I Hydration Dynamics as an Intrinsic Ruler for Refining Protein Structure at Lipid Membrane Interfaces Chi-Yuan Cheng a , Jobin Varkey b , Mark R. Ambroso b , Ralf Langen b , Songi Han a * a Department of Chemistry and Biochemistry, University of California, Santa Barbara, Santa Barbara, CA, 93106. b Department of Biochemistry and Molecular Biology, Zilkha Neurogenetic Institute, Keck School of Medicine, University of Southern California, Los Angeles, CA, 90033. *Corresponding Author This work was originally published in the Proceeding of the National Academy of Sciences 110, 16838-16843, 15 Oct 2013 and formatted according to dissertation guidelines. Annexin B12 sample preparation and a portion of the annexin B12 ODNP measurements were performed by Mark Ambroso. 179 ABSTRACT Knowing the topology and location of protein segments at water–membrane interfaces is critical for rationalizing their functions, but their characterization is challenging under physiological conditions. Here, we debut a unique spectroscopic approach by using the hydration dynamics gradient found across the phospholipid bilayer as an intrinsic ruler for determining the topology, immersion depth, and orientation of protein segments in lipid membranes, particularly at water– membrane interfaces. This is achieved through the site-specific quantification of translational diffusion of hydration water using an emerging tool, 1 H Overhauser dynamic nuclear polarization (ODNP)-enhanced NMR relaxometry. ODNP confirms that the membrane-bound region of α-synuclein (αS), an amyloid protein known to insert an amphipathic α-helix into negatively charged phospholipid membranes, forms an extended α-helix parallel to the membrane surface. We extend the current knowledge by showing that residues 90–96 of bound αS, which is a transition segment that links the α-helix and the C terminus, adopt a larger loop than an idealized α-helix. The unstructured C terminus gradually threads through the surface hydration layers of lipid membranes, with the beginning portion residing within 5–15 Å above the phosphate level, and only the very end of C terminus surveying bulk water. Remarkably, the intrinsic hydration dynamics gradient along the bilayer normal extends to 20–30 Å above the phosphate level, as demonstrated with a peripheral membrane protein, annexin B12. ODNP offers the opportunity to reveal previously unresolvable structure and location of protein segments well above the lipid phosphate, whose structure and dynamics critically contribute to the understanding of functional versatility of membrane proteins. 180 SIGNIFICANCE The structural properties of a membrane-associating protein at the water–membrane interfaces are intimately linked to its biological function, but they are difficult to characterize using existing biophysical tools. We identify the existence of a distinct intrinsic gradient of water diffusion across the lipid bilayer, encompassing a thick surface hydration layer above the lipid membrane surface, and debut an approach to exploit this gradient as a highly sensitive ruler to determine the topology, immersion depth, and location of a membrane associating protein, including the segments residing well above the membrane surface, that are otherwise difficult to resolve. This study demonstrates a potential of a broadly applicable approach for the structure–dynamics– function study of membrane proteins, membrane systems, and beyond. 181 Introduction Membrane proteins (MPs) constitute more than 25% of all human proteins. They not only represent the largest class of drug targets, but they also perform essential functions ranging from sensing, signaling, to transporting ions and solutes (Rosenbaum et al., 2009). MP function is critically associated with its structure, not only the conformations within the hydrophobic core of the bilayer but also of the protein segments located at or near the hydrophilic lipid headgroup region (Cho et al., 2005; von Heijne et al., 2006; Rosenbaum et al., 2009; White et al., 2008). This water–membrane interface comprises a large portion ( ∼20%) of the total MP in the lipid bilayer. Many functional domains of MPs are found to extend outward from the outer leaflet of the lipid bilayer into the water–membrane interface, exposing a heterogeneous environment capable of mediating a wide range of noncovalent interactions in the extracellular matrix. For instance, the membrane-associating domain of peripheral MP is essential for several key cellular functions by regulating the dynamics and morphology of cellular membranes (Lemmon et al., 2008). These functional domains of MPs are commonly highly disordered, display heterogeneous conformations, are often associated with parts of the cell matrix, or transiently bind to a cell surface in crowded environments (Cho et al., 2005). Thus, it is difficult to study the structure and dynamics of these protein segments at water–membrane interfaces by using traditional analytical tools. For instance, although X-ray crystallography and solution-state NMR spectroscopy offer atomic-level information of MPs, they have limitations associated with the size, dynamics, and complexity of MPs in membrane-mimetic systems (Raschle et al., 2010). Electron paramagnetic resonance (EPR) spectroscopy combined with site-directed spin labeling (SDSL) is an alternative approach for accessing local structural and dynamic information of MPs by means of probing dynamics and coupling of stable nitroxide radicals, 182 termed spin labels, attached to biomolecular sites of interest. Its unique benefit is that the measurement is not intrinsically limited by the size or complexity of protein and its environments (Altenbach et al., 2001). Continuous wave (CW) EPR line shape analysis can track conformational changes of protein segments and their interactions with other biomolecular assemblies (Hubbell et al., 2000). Pulsed dipolar EPR methods provide intra- or interprotein distances between 2- and 7-nm length scales in various biological environments (Jeschke et al., 2012). Moreover, the local protein structure with respect to the lipid bilayer surface can be determined from the collision frequency (or solvent accessibility) of a single nitroxide radical labeled on consecutive protein residues to paramagnetic reagents, such as nickel- ethylenediaminediacetic acid (Ni-EDDA) (water-soluble) and O 2 (lipid-soluble), via power saturation EPR measurements (Altenbach et al., 1994). Although this collision gradient approach has been well established to determine the immersion depth and topology of various MP structural elements in lipid bilayers (Altenbach et al., 1994; Chakrapani et al., 2008; Frazier et al., 2002, 2003; Jao et al., 2004, 2008; Isas et al., 2004, 2005; Perozo et al., 1998), it has lower or, in some cases, no resolution when the protein segment is located more than 5 Å away from the lipid phosphate toward the solvent (Frazier et al., 2002, 2003). This limitation is because the collision frequency of nitroxide radicals with paramagnetic reagents measured by this method relies on their concentration gradient along the bilayer, which vanishes in the region of a distance >5 Å above the phosphate level. Thus, the collision frequency becomes indistinguishable from that in bulk water. As MPs‘ domains close to water–membrane interfaces are highly tuned to their functional role, alternative and/or more sensitive methods for characterizing their structural features and locations are needed. 183 A synthetic lipid bilayer comprises well-defined hydrophobic dimensions, with a substantially steep gradient of water density connecting the surface hydration layers of the membrane and a nearly dehydrated core of hydrocarbon bilayer (Wimley et al., 1996). A less obvious, but equally important, characteristic is the differential hydration dynamics across the lipid bilayer that is determined by the interfacial energy between hydration water and the lipid molecules, as well as the spatial confinement of hydration water within the bilayers. Given that these properties are thought to play a key role in tuning the structure and function of MPs in the lipid membrane (MacCallum et al., 2011), it motivates us to closely investigate the diffusion characteristics of hydration water across the lipid bilayer. Although it is now well accepted that the water diffusivity within the interior and near the surface of the lipid bilayer is significantly (5- to 11-fold) retarded (Hodges et al., 1997; Kausik et al., 2011; Tobias et al., 1997) compared with that of bulk water (i.e., 2.3 × 10 −9 m 2 /s at 25 °C), the mechanism by which water passively diffuses across the bilayer is not well established. Although water diffusivity is assumed to be homogeneous across the bilayer interior in some studies (Finkelstein et al., 1976; Reeves et al., 1970), recent computational works have suggested local heterogeneities in water diffusivity across the bilayer, possibly mediated by transient pores (Bemporad et al., 2004). Diverging viewpoints are further fueled by the sparsity of direct experimental measurements. Among others, an experimental observation of differential hydration dynamics across the lipid bilayer has been reported on large unilamellar vesicles (LUVs, 200 nm in diameter) made of 1,2- dioleoyl-3-trimethylammonium-propane (DOTAP), countering the homogeneous diffusion model (Kausik et al., 2011). These measurements were enabled by Overhauser dynamic nuclear polarization (ODNP)-enhanced NMR relaxometry that probes the local translational diffusion of hydration water around the specific sites of biological samples under physiological conditions 184 (Armstrong et al., 2007, 2009, 2011; Franck et al., 2013; Hausser et al., 1968; Hussain et al., 2013; Ortony et al., 2011). In this work, we extend this finding by demonstrating that the gradient of water diffusion not only exists within the bilayer interior, but also spans about 20–30 Å above the phosphate level, eventually interfacing bulk water. This finding is consistent with recent studies identifying the existence of a thick hydration shell (>20 Å thickness) on biomolecular surfaces (Ebbinghaus et al., 2007). Based on this observation, we debut a unique approach that employs this extended gradient of hydration dynamics as a sensitive and high-resolution molecular ruler for probing conformation and location of MPs within the bilayer and near the water–membrane interface, extending to the distance of 20–30 Å above the phosphate level. In this work, α-synuclein (αS), a 14-kDa amyloid protein implicated in Parkinson disease (PD), was used as an illustrative system to showcase the herein introduced approach. It has been known that its amphipathic N terminus (residues 1–60) can transform into an extended α-helix upon binding to a negatively charged small unilamellar vesicle (SUV) surface, whereas the C terminus (residues 96–140) remains unstructured and does not bind to the lipid membrane (Bodner et al., 2009; Ferreon et al., 2009; Jao et al., 2004, 2008; Lokappa et al., 2011; Robotta et al., 2011). The residues 61–95 of αS contain a highly hydrophobic self-aggregating sequence that is essential for αS fibrillation (Fink et al., 2006). It is generally thought that the interaction between different αS aggregated species and the cell membrane is directly implicated in the pathological and physiological roles of PD (Fink et al., 2006; Outeiro et al., 2003; Varkey et al., 2010). However, little is known about the conformational states of αS in the disease mechanism of PD, fueled by challenges in characterizing the structural states of membrane-bound proteins in the aggregation processes. Although the C terminus does not seem to play a key role in the αS fibrillation, recent studies 185 have suggested that it may be critical for preventing the formation of rapid αS filament assemblies (Bertoncini et al., 2005; Dedmon et al., 2005; Levitan et al., 2011). Our unique approach demonstrates that the local hydration dynamics at the αS protein–membrane interfaces can sensitively register the relative position of the protein residues that are embedded within the lipid bilayer, as well as located more than 5 Å above the lipid phosphate. Importantly, this result implies that the intrinsic hydration dynamics gradient along the bilayers must dominate over the intrinsic hydration dynamics of the protein surface. We verify this to be the case by studying a model peripheral MP, annexin B12 (Cartailler et al., 2000; Isas et al., 2004, 2005), whose surface hydration dynamics are found to scale with distance up to 30 Å from the phosphate level toward bulk water, suggesting the potential generality of this method. Thus, ODNP is uniquely applicable to characterizing the structural features of MPs‘ regions that reside at the water– membrane interface. By integrating this powerful tool with other mature biophysical tools, such as power saturation EPR and CW EPR line shape analysis, previously inaccessible questions can now be addressed, such as whether a MP segment is bound to the surface or immersed in the bilayer, whether the structural motifs of a MP orient along the membrane surface, whether hydration water resides between the MP segment and the membrane surface, and what part of the unstructured MP segments resides close to the membrane surface or extends into bulk water. Results and Discussions Approach to quantify local hydration dynamics at biomolecular interfaces ODNP measurements were used to quantify the diffusion dynamics of hydration water around nitroxide spin labels tethered to the specific sites of protein or lipid membrane systems, while the local conformational dynamics at the same sites of protein or lipid membrane were concurrently 186 evaluated by CW EPR line shape analysis. ODNP relies on extracting the efficiency of dipolar relaxation between the electron spin of a nitroxide radical and the nuclear spin of nearby water protons. The efficiency of the electron–nuclear cross-relaxation is critically modulated by the water diffusivity within 5–10 Å around the spin label (Armstrong et al., 2007, 2009; Franck et al., 2013). The distance-dependent dipolar coupling between the water protons and the electron spin of the radical functionalized at biomolecular interfaces permits the selective characterization of the diffusivity of hydration water within 5–10 Å (encompasses two to four hydration layers) around the tethered spin label, whose dynamic characteristics is defined by the nearby biomolecular surface. Given that the frequency for probing hydration dynamics by NMR relaxometry is extended from the traditional upper limit of hundreds of megahertz to 10 GHz, and given that the efficiency of electron–nuclear cross-relaxation yields up to 300-fold NMR signal amplification, ODNP is rendered much more sensitive than conventional NMR relaxometry methods. In ODNP, the electron–proton dipolar cross-relaxation that is critically mediated through the translational diffusion of the hydration water is driven by the saturation of the EPR transition, giving rise to NMR signal enhancement (Armstrong et al., 2007, 2009; Franck et al., 2013; Hausser et al., 1968). At extrapolated maximum saturation, the NMR signal enhancement can be expressed as (Armstrong et al., 2007, 2009; Franck et al., 2013; Hausser et al., 1968) where ξ is the coupling factor, f is the leakage factor accounting for paramagnetic-enhanced proton relaxation over all proton relaxation mechanisms, s max is the maximum saturation factor for the electron spin (i.e., s max = 1 for full saturation achieved for slow-tumbling molecular systems) (Armstrong et al., 2007), and | γe/ γN| is the ratio of the gyromagnetic ratios of the 187 electron and proton spins, giving 658. ξ is the key parameter describing the efficiency of electron–proton cross-relaxation and can be obtained directly from the measurement of E max and the proton T 1 relaxation rates in the presence and absence of the paramagnetic radicals (Armstrong et al., 2007, 2009; Franck et al., 2013). The translational correlation time, τ, of hydration water around the radical can then be extracted from ξ, when explicit spectral density functions are modeled (Hausser et al., 1968), as described in SI Text. In this study, ODNP is performed using a modified X-band CW EPR spectrometer at a magnetic field of 0.35 T and at 25 °C, irradiating with moderate microwave powers (up to ∼4 W) at the electron Larmor frequency of ωe = 9.8 GHz. The choice of ωe of 9.8 GHz is critical, as the electron–nuclear cross-relaxation near this frequency is precisely modulated by the fast translational diffusion of loosely bound hydration water within the distance of closest approach (d) between the electron and 1 H at biomolecular interfaces with translational correlation time, τ, on the order of tens to hundreds of picosecond, given the translational diffusivity of water,D, following D ∼ d 2 /τ (Armstrong et al., 2007, 2009; Franck et al., 2013; Hausser et al., 1968). The range of τ accessible by ODNP at ωe = 9.8 GHz covers the dynamic range for hydration water surrounding solvent-exposed sites, as well as to some extent buried sites of proteins and other biomacromolecules (Ortony et al., 2011). To qualitatively compare the hydration dynamics in different biological environments, we define the retardation factor of hydration dynamics, ρt = 〈τ 〉/ τ bulk , which is the average translational correlation time of the local hydration water around a tethered spin label divided by that of bulk water (i.e., τ bulk = 33 ps, measured by ODNP at 0.35 T using a small nitroxide radical dissolved in water) (Bennati et al., 2010; Franck et al., 2013). As illustrated in Figure 1, the coupling between the electron spin and water proton is shown to be sensitively modulated within the dynamic range of hydration water at solvent- 188 exposed surfaces of proteins ( ρt = 2–5) (Armstrong et al., 2011; Jansson et al., 2009; Polnaszek et al., 1984; Russo et al., 2004), as well as on the surfaces or within cores of lipid bilayers ( ρt = 5–11) (Hodges et al., 1997; Kausik et al., 2011; Tobias et al., 1997), affording the use of ODNP to probe conformational changes or interactions between proteins and lipid membranes by monitoring the changes in the local hydration dynamics at molecular interfaces. Crucially, biological samples at dilute concentrations (approximately tens of micromolars), of minute volumes (approximately a few microliters), and in an environment of excess water, lipids and other biological constituents at physiological temperature are experimentally accessible. Similar to CW EPR spectroscopy, ODNP is not fundamentally limited by the molecular weight, size, or the complexity of the sample system, permitting the study of site-specific hydration dynamics of MPs in lipid membrane environments under physiological conditions. Figure A1. Coupling factor (ξ) is presented as a function of translational correlation time (τ) of hydration water at 0.35 T and at 25 °C. Retardation factor ( ρt) of hydration water is referenced to the bulk water diffusivity. The distance (xi) with respect to the phosphate level and its relation with the retardation factor are shown in the Inset. 189 Probing Hydration Dynamics as a Function of Distance Along the Bilayer Normal. To obtain the diffusion profile of water penetration into the fully hydrated lipid bilayer composed of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and 1-palmitoyl-2- oleoyl-sn-glycero-3-phospho-L-serine (POPS) in SUVs, we measured the ρt value of hydration water at distances above the lipid phosphate and at different bilayer depths using lipid spin probes: 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho(tempo)choline (TEMPO-PC) with a nitroxide radical attached to the choline moiety of PC and doxyl stearic acids (DSAs) with a nitroxide radical attached at various positions along the alkyl chains. The distance from the nitrogen of the nitroxide to the phosphate in TEMPO-PC was estimated to be ∼5 Å based on molecular models (Farahbakhsh et al., 1992), whereas the immersion depths of the lipid spin probes were estimated from X-ray diffraction (Dalton et al., 1987). The correlation between the distance from the lipid phosphate to the nitroxide of the lipid spin probes, xi, and the herein measured retardation factors, ρt, at the given spin-labeled sites of bilayers is presented in Figure 2, where a positive xi value signifies the position below the phosphate group toward the bilayer and a negative xi value represents the position above the phosphate group toward bulk water. The experimental ODNP data in Figure 2 are presented in Tables S1 and S2. We empirically found that the natural logarithm of the ρt value of the POPC/POPS bilayer increases approximately linearly with the immersion depth of the nitroxide spin probe within the bilayers. Most significantly, the surface hydration dynamics probed by TEMPO-PC at around 5 Å above the phosphate group of the lipid bilayer (Farahbakhsh et al., 1992) is still 6.8-fold retarded compared with bulk water diffusivity, thereby providing a significant dynamic range for resolving the location of protein segments within the relatively thick hydration layers on the lipid membrane 190 surfaces. In contrast, the solvent accessibility measured by power saturation EPR loses contrast when the spin label is >5 Å away from the phosphate level (Frazier et al., 2002, 2003). To test the generality and the validity of this intrinsic hydration dynamics gradient-based ruler along the bilayer normal at distances of >5 Å above the lipid phosphate, we examined another membrane-associating protein—a Ca 2+ -dependent peripheral MP, annexin B12. The membrane-bound form of annexin B12 has a convex shape on phosphatidylserine (PS)-rich membrane surfaces with a known tertiary structure (Cartailler et al., 2000), whose membrane binding is mediated by Ca 2+ ions (Isas et al., 2004, 2005). The crystallographic data suggest that the most distant portion of the membrane-bound annexin B12 is around xi = −30 Å (above the phosphate), whereas its deepest residue is xi ∼2 Å (below the phosphate) (Cartailler et al., 2000), as illustrated in Figure S1. Figure 2 presents a plot of the retardation factors at selected water- exposed residues of annexin B12 bound on the LUV surface composed of POPC and POPS with varying distance, xi, with respect to the phosphate level. Interestingly, we revealed that the natural logarithm of the retardation factor again monotonically decreases with the distance away from the bilayer surface, whereas a simple linear regression of the measured data is not suitable, especially at distant water-exposed sites of the membrane-bound annexin B12. Rather, the distance dependence of the ln ρt value across the protein surface along bilayer normal follows a hyperbolic tangent function edged by a significant retardation factor up to xi = −30 Å (solid curve in Figure 2), whereas an approximately linear relation was only found between xi = −5 and −25 Å (dotted line in Figure 2). To understand whether Ca 2+ influences the absolute values of hydration dynamics on the POPC/POPS bilayer surface studied herein, we compare the ODNP data in the presence and absence of Ca 2+ . We found that upon addition of 1 mM CaCl 2 to the POPC/POPS LUV solution, the hydration dynamics measured at the bare membrane surface 191 substantially slows by 54% (τ increases from 257 to 474 ps), likely because Ca 2+ rigidifies the PS and neighboring PC headgroups (Boettcher et al., 2011). All experimental data yielding ρt values are presented in Table S2. This finding suggests that, although a qualitatively similar hydration dynamics gradient was found in different lipid bilayer systems, the absolute diffusivities of the hydration water alter significantly, likely depending on the lipid composition, lipid membrane curvature, charge density of the phospholipid headgroups, buffer condition, and the presence of ions or osmolytes. For instance, a similar linear correlation between the retardation factor and immersion depth of lipid bilayers was observed in different bilayer systems, regardless of their surface charges or vesicle sizes, whereas their absolute water diffusivities are quite different (Figure S2). Detailed investigation into this dependence merits future studies, in particular considering the potential effect of membrane curvature on tuning the hydration dynamics on the membrane surface (Kunding et al., 2011) and the interactions of curvature-sensitive MPs with lipid membranes. Figure A2. Distance dependence of retardation factor ( ρt) at specific sites of small unilamellar vesicles composed of POPC:POPS (7:3) (□) and annexin B12 bound on the surface of large unilamellar vesicles composed of POPC:POPS (1:2) in the presence of 1 mM Ca 2+ (●) at 25 °C. Error bars represent SDs. The fit parameters are summarized in supporting information. 192 193 Figure S1. X-ray crystal structure of membrane-bound annexin B12 (PDB ID code 1DM5) in the presence of Ca 2+ . The residues studied in this work are labeled in yellow. The distances of selected residues with respect to the lipid phosphate are reported in Table S2. Figure S2. Retardation factor (ρt) at the site-specific nitroxide radical of lipid bilayers vs. the distance from the phosphate group to the nitroxide radicals of phospholipids or detergents in POPC/POPS small unilamellar vesicles (25 nm diameter) and DOTAP large unilarmellar vesicles (200 nm diameter) (Kausik et al., 2011) at 25 °C. These vesicle samples are in deionized water in the absence of Ca 2+ . 194 Topology and immersion depth of α-synuclein on a lipid membrane surface The approximately linear relation found between the ln ρt value and corresponding distance with respect to the phosphate group, xi, can be used as a calibration curve to determine the relative distance of specific protein residues in the lipid bilayer. Here, we use αS as an example to examine whether the ODNP-derived hydration dynamics gradient can indeed provide structural information of a membrane-associating protein with higher sensitivity and spatial resolution. Previous studies have revealed that when αS binds to intact negatively charged SUVs, its amphipathic repeats comprising ∼90 N-terminal residues form an extended α-helix, located parallel to the membrane surface, whereas the disordered C terminus has not been observed to directly interact with the lipid membrane (Bodner et al., 2009; Ferreon et al., 2009; Jao et al., 2004, 2008; Lokappa et al., 2011; Robotta et al., 2011). We first measured hydration dynamics at selected sites of monomeric αS between residues 76 and 108 in the absence of lipid membranes, as shown in Figure 3. The results show that the ρt value exhibits a small dispersion between 2.4 and 4.0, suggesting that hydration water is relatively freely, and roughly evenly, diffusing near the protein surface in the unbound state. This result displays typical characteristics of hydration water around solvent-exposed protein surfaces with ρt values found between 2 and 5 (Armstrong et al., 2011; Jansson et al., 2009; Polnaszek et al., 1984; Russo et al., 2004), confirming that αS in solution is in an overall unstructured state, where each residue is significantly exposed to bulk water. Additionally, the retardation factor of membrane-bound αS at consecutive spin-labeled sites from residue 76–108, as well as two additional residues 124 and 136 of the C-terminal end, were examined (Figure 3). The first observation is that all sites, except residues 124 and 136, display measurably higher retardation factors ( ρt = 5.4–9.7) in the membrane-bound states compared 195 with their unbound states, implying the existence of the interactions between membrane surfaces and the C-terminal segment of αS that was previously assumed to be simply water exposed (Jao et al., 2004, 2008). The ODNP data of unbound and bound αS, together with representative plots of 1 H NMR signal enhancements vs. microwave power, are summarized in Tables S3 and S4 and Figure S3. By examining the periodicity of the hydration dynamics along the protein sequence in quantitative detail, we found the retardation factor between residues 76 and 90 to exhibit periodic oscillation every 3.48 ± 0.08 residues, as extracted from the least-squares fits to a cosine function, constituting close to an ideal α-helix structure (with 3.67 residues per helical turn or 11 residues per three turns) (Jao et al., 2004, 2008). We then carefully evaluated the periodicity by using the α-helix periodicity index, αPI, based on a harmonic analysis with discrete Fourier transformation (Figure S4) (Cornette et al., 1987; Donnelly et al., 1994). This method has been previously used to facilitate the secondary structure assignment of MPs by power saturation EPR (Chakrapani et al., 2008, Perozo et al., 1998). The protein segment with αPI ≥ 2 is considered to be an α-helical structure with statistical significance (Cornette et al., 1987; Donnelly et al., 1994). We found an αPI value of 3.38 for the retardation factors of membrane-bound αS between residues 76 and 90, confirming a significant α-helical structure. Interestingly, the retardation factors of membrane-bound αS between residues 90 and 96, representing the transition region between the α-helix and the disordered C terminus, yield an αPI value of 0.13, indicating these residues are not part of an α-helix. It is likely that the hydrophobic residues (Phe-94 and Val-95) of this domain facilitate the penetration below the phosphate level to form a more extended loop than an α-helix. 196 The positions of membrane-bound αS residues with respect to the lipid phosphate can be determined by referencing their retardation factors against the calibration curve obtained from lipid spin probes, as shown in Figure 2. We observe that the immersion depth of the helical bottom lies in the range of 8–10 Å, with the center of the helix located 1–3 Å below the phosphate level (Figure 4). This result is in excellent agreement with the first observation of this shallow insertion into bilayers made by power saturation EPR (Jao et al., 2004, 2008), where the immersion depth of the bottom portion of the α-helix was estimated at 11 Å and the center of the helix was at 1–4 Å below the lipid phosphate (Jao et al., 2004, 2008). Our findings are also consistent with recent fluorescence and neutron reflectometry results, reporting an immersion depth of the α-helical bottom of around 9–14 Å (Pfefferkorn et al., 2012). Interestingly, we found that the hydration dynamics between residues 98 and 108, which are thought to be part of an entirely bulk water-exposed C terminus, exhibit still significantly retarded hydration dynamics. Their retardation factors are distinctly different from those of the same αS residues free in solution. This observation implies that this segment still resides closely within the surface hydration layers spanning 5–15 Å above the lipid phosphate group and experiences a more stable network of hydration water through which water diffusion is significantly retarded compared with in bulk water. Interestingly, 9 of the 11 amino acids between residues 98 and 108 are either polar or negatively charged, making this segment highly favorable for binding to positively charged trimethyl ammonium groups of the choline moieties that protrude to the solvent-exposed headgroup region of the bilayer. Noteworthily, the ρt values at residues 124 and 136 of membrane-bound αS, which are much closer to the end of the C terminus, display unaltered hydration dynamics within errors to those of the unbound state. This finding solidifies the idea that the C-terminal region, following the extended α-helical segment, 197 gradually winds out and dangles outward to the solvent, where only the very end of the C terminus is fully solvated in bulk water (Figure 4). This finding implies that the transition region still resides within the surface hydration layers of the outer leaflet of the lipid membrane. Moreover, ODNP reveals that the C terminus of membrane-bound αS displays a distinct landscape of hydration dynamics that may have functional implications. Its disordered and flexible structural configuration could regulate the interaction with constituents in solution, such as ligands, proteins, or metals, implicated in modulating the biological functions of αS (Eliezer et al., 2001; Ulmer et al., 2005). Furthermore, a recent study showed that the highly charged C terminus of free αS can efficiently inhibit its fibril formation driven by electrostatic repulsions (Levitan et al., 2011). Our result demonstrates that the beginning portion of the C terminus is in close proximity to the membrane surface—this possibly tucks away the αS and slows its aggregation by preventing free inter-αS interactions. ODNP could be used in future studies to track the conformational changes of αS or other amyloid proteins on membrane surfaces during their aggregation process. Figure A3. Retardation factor ( ρt) at various residues of membrane-bound and free α-synuclein measured by ODNP. The distance (xi) of nitroxide radicals with respect to the phosphate group of the POPC/POPS bilayers is presented. Error bars represent SDs. The red curves are least- squares fits of data to a cosine function. The representative phospholipids and lipid spin probes are shown in the background. Nitroxide radicals of lipid spin probes are represented in red balls. 198 Figure A4. Structural representation of topology and immersion depth of membrane-bound α- synuclein. The red ribbon was the topology obtained from ODNP measurements, whereas the gray ribbon was observed from the literature (Jao et al., 2004, 2008). Black solid line: phosphate level; black dashed line: center of the helix; black dotted line: bottom of the helix. Figure A3, continued 199 200 201 Figure S3. Representative plots of 1H NMR signal enhancement as a function of microwave power, with extrapolation to the infinite power to obtain Emax value. The examples demonstrate the difference in membrane-bound and unbound state of α-synuclein with the 2,2,5,5- tetramethylpyrroline-3-yl-methanethiosulfonate (MTSL) spin label at 77C (A, membrane-bound domain) and 101C (B, C terminus) of α-synuclein. Figure S4. The Fourier transform power spectra P(ω) of retardation factor between (A) residues 76–90 and (B) residues 90–96. The power spectra with a major peak at around 105° of angular frequency is consistent with the α-helical periodicity index (αPI) of the sequence. αPI ≥ 2 shows that the secondary structure assignment for α-helix is statistically significant. 202 Figure S5. Chemical structures of lipid samples (phospholipids and lipid spin labels) used in this work and previous work (Kausik et al., 2011). Conclusion In this work, we introduce a unique tool to determine and refine the topology and the immersion depth of membrane-associating proteins using the intrinsic diffusion gradient of hydration water across a lipid bilayer as a sensitive ruler. We characterize the dependence of the local water diffusivity on the distance across the bilayer normal in bare lipid membranes and on the surface of membrane-bound annexin B12. We find the intrinsic hydration dynamics gradient to extend around 20–30 Å above the phosphate level and to dominate over that of the protein surface, both of which are prerequisites for the general applicability of this intrinsic ruler. However, the absolute diffusivities of hydration water is expected to vary with membrane surface properties 203 (e.g., membrane curvature, roughness, or charge density), buffer conditions, and to some extent, the structural nature of protein embedded, requiring a proper experimental calibration. We demonstrate the application of this intrinsic ruler method on the example of membrane-bound αS that we found to form an extended α-helix between residues 76 and 90, with its long axis of the helical center slightly embedded 1–3 Å below the lipid phosphates, in good agreement with earlier findings (Jao et al., 2004, 2008). We further observed that the beginning portion of the C terminus is located within the distinctly characteristic hydration layers off the membrane surface, whereas the very end of the C terminus is entirely exposed to bulk water. In general, by exploiting the predetermined relationship between the retardation of hydration dynamics and its corresponding distance with respect to the lipid phosphate, ODNP provides a critical supplement to structural biology tools to characterize the topology, immersion depth, and orientation of membrane-active biological constituents, including transmembrane proteins, amyloid fibrils, and membrane-active peptides. A key strength of this approach is that the measurements are carried out in solution under physiological conditions, while the constituents can be associated with lipid membranes with virtually any size or complexity. Overall, ODNP provides a unique capability with enhanced resolution, site specificity, and sensitivity for identifying the structure and location of MPs, especially with finer resolution in the interfacial volume at distances up to 20– 30 Å above the phosphate level. We believe ODNP will offer new opportunities for deepening the understanding of MP structure, organization, and function. 204 Materials and Methods Preparation of Spin-Labeled α-Synuclein. Single cysteine mutants of α-synuclein were expressed in BL21(DE3)pLysS Escherichia coli cells, and the cell pellet was resuspended in lysis buffer (100 mM Tris ⋅HCl, 300 mM NaCl, 1 mM EDTA, pH 8) (Jao et al., 2004, 2008). After boiling the cell lysate and precipitating with HCl, the supernatant was dialyzed against a buffer (20 mM Tris ⋅HCl, 1 mM EDTA, and 1 mM DTT, pH 8). Two rounds of anion exchange chromatography were performed, and proteins were eluted with a salt gradient of 0–1 M NaCl. DTT was added to the samples to a final concentration of 1 mM for 30 min. DTT was then removed by size exclusion PD-10 columns (GE Healthcare) in buffer A (10 mM Hepes and 100 mM NaCl, pH 7.4). A fivefold molar excess of the 2,2,5,5-tetramethylpyrroline-3-yl- methanethiosulfonate (MTSL) spin label (Toronto Research Chemicals) was incubated with the protein samples for 1 h at 25 °C. Excess spin label was removed by buffer exchange using PD-10 columns. Spin-labeled proteins were concentrated using Amicon Ultra centrifugal filters (Millipore). Protein concentration was quantified using Micro BCA protein assay kit (Thermo Scientific). The SUVs composed of POPC and POPS (7:3, wt:wt) were prepared in buffer A using a sonication method. The average diameter of SUV was about 25 nm, as measured by dynamic light scattering (Zetasizer Nano ZS; Malvern Instruments). The following lipid spin probes were used for ODNP measurements: TEMPO-PC and n-doxyl stearic acid (n = 5, 7, 10, 12); 62.5 mM lipid and 500 μM spin probe were used. The chemical structure of the lipids and lipid spin probes used in this work are summarized in Figure S5. Proteins were incubated with SUVs at a 1:250 protein to lipid molar ratio in buffer A at 25 °C for 1 h, followed by several rounds of washes using a 100-kDa cutoff centrifugal filter unit (Millipore) to remove the unbound proteins. 205 Preparation of Spin-Labeled Annexin B12. The expression and purification of the annexin B12 used in this study were previously described (Isas et al., 2004, 2005). The preparation of MTSL spin-labeled annexin B12 cysteine mutants is similar to that of spin-labeled αS, except equilibrated in buffer B (20 mM Hepes, pH 7.4, and 100 mM NaCl). LUVs composed of POPS and POPC in a 2:1 molar ratio were prepared in buffer B through an extrusion method using 1,000-nm filters. Spin-labeled annexin B12 (150 μg) was incubated with 1.5 mg of LUVs and 1 mM CaCl 2 and allowed to react for at least 15 min at 25 °C. Lipids were pelleted at 16,000 × gfor 20 min, the supernatant was removed, and the pellet was resolubilized in 6 μL of buffer B and subjected to ODNP measurements. The distances between the lipid phosphates and the selected residues of membrane-bound annexin B12 were approximated based on a molecular model of annexin B12‘s crystal structure on a membrane surface (Cartailler et al., 2000) using Discovery Studio Client 3.0 (Accelrys) and known depths of its membrane-binding surface (Isas et al., 2004, 2005). ODNP Experiments. 1 H ODNP experiments were performed at a 0.35-T electromagnet, operating at a 14.8-MHz 1 H Larmor frequency and at 9.8-GHz electron Larmor frequency. A 3.5-μL sample was loaded in a 0.6-mm-inner-diameter quartz capillary tube (Fiber Optic Center), and both ends were sealed with bee wax. The capillary was mounted on a homebuilt NMR probe with a U-shaped RF coil. The EPR signal was then acquired by Bruker X-band EMX CW EPR spectrometer with a dielectric microwave resonator (Bruker ER-4123D) at 25 °C. During the ODNP experiments, the center field of nitroxide hyperfine transition lines was irradiated continuously at various microwave powers while the 1 H NMR signal was recorded. Dry air continuously flowed through the EPR cavity at a flow rate of 8 L/min to prevent the sample heating. 1 H T 1 relaxation measurements were performed by an inversion-recovery pulse sequence 206 operated by a Bruker Avance spectrometer at a 0.35-T electromagnet. The theory and instrumentation of ODNP are summarized in the SI Text. 207 REFERENCES Altenbach C, Greenhalgh DA, Khorana HG, Hubbell WL (1994) A collision gradient method to determine the immersion depth of nitroxides in lipid bilayers: Application to spin-labeled mutants of bacteriorhodopsin. Proc Natl Acad Sci USA 91(5):1667–1671. Altenbach C, Oh KJ, Trabanino RJ, Hideg K, Hubbell WL (2001) Estimation of interresidue distances in spin labeled proteins at physiological temperatures: Experimental strategies and practical limitations. Biochemistry 40(51):15471–15482. Armstrong BD, Han S (2007) A new model for Overhauser enhanced nuclear magnetic resonance using nitroxide radicals. J Chem Phys 127(10):104508. Armstrong BD, Han S (2009) Overhauser dynamic nuclear polarization to study local water dynamics. J Am Chem Soc 131(13):4641–4647. Armstrong BD, et al. (2011) Site-specific hydration dynamics in the nonpolar core of amolten globule by dynamic nuclear polarization of water. J Am Chem Soc 133(15):5987–5995. Bemporad D, Essex JW, Luttmann C (2004) Permeation of small molecules through a lipid bilayer: A computer simulation study. J Phys Chem B 108(15):4875–4884. Bennati M, Luchinat C, Parigi G, Türke MT (2010) Water 1H relaxation dispersion analysis on a nitroxide radical provides information on the maximal signal enhancement in Overhauser dynamic nuclear polarization experiments. Phys Chem Chem Phys 12(22):5902–5910. Bertoncini CW, et al. (2005) Release of long-range tertiary interactions potentiates aggregation of natively unstructured alpha-synuclein. Proc Natl Acad Sci USA 102(5): 1430–1435. Bodner CR, Dobson CM, Bax A (2009) Multiple tight phospholipid-binding modes of alpha- synuclein revealed by solution NMR spectroscopy. J Mol Biol 390(4):775–790. Boettcher JM, et al. (2011) Atomic view of calcium-induced clustering of phosphatidylserine in mixed lipid bilayers. Biochemistry 50(12):2264–2273. Cartailler JP, Haigler HT, Luecke H (2000) Annexin XII E105K crystal structure: Identification of a pH-dependent switch for mutant hexamerization. Biochemistry 39(10): 2475–2483. Chakrapani S, Cuello LG, Cortes DM, Perozo E (2008) Structural dynamics of an isolated voltage-sensor domain in a lipid bilayer. Structure 16(3):398–409. Cho WH, Stahelin RV (2005) Membrane-protein interactions in cell signaling and membrane trafficking. Annu Rev Biophys Biomol Struct 34:119–151. Cornette JL, et al. (1987) Hydrophobicity scales and computational techniques for detecting amphipathic structures in proteins. J Mol Biol 195(3):659–685. 208 Dalton LA, McIntyre JO, Fleischer S (1987) Distance estimate of the active center of Dbeta- hydroxybutyrate dehydrogenase from the membrane surface. Biochemistry 26(8):2117– 2130. Dedmon MM, Lindorff-Larsen K, Christodoulou J, Vendruscolo M, Dobson CM (2005) Mapping long-range interactions in alpha-synuclein using spin-label NMR and ensemble molecular dynamics simulations. J Am Chem Soc 127(2):476–477. Donnelly D, Overington JP, Blundell TL (1994) The prediction and orientation of alpha-helices from sequence alignments: The combined use of environment-dependent substitution tables, Fourier transform methods and helix capping rules. Protein Eng 7(5):645–653. Ebbinghaus S, et al. (2007) An extended dynamical hydration shell around proteins. Proc Natl Acad Sci USA 104(52):20749–20752. Eliezer D, Kutluay E, Bussell R, Jr., Browne G (2001) Conformational properties of alpha- synuclein in its free and lipid-associated states. J Mol Biol 307(4):1061–1073. Farahbakhsh ZT, Altenbach C, Hubbell WL (1992) Spin labeled cysteines as sensors for protein- lipid interaction and conformation in rhodopsin. Photochem Photobiol 56(6): 1019–1033. Ferreon ACM, Gambin Y, Lemke EA, Deniz AA (2009) Interplay of alpha-synuclein binding and conformational switching probed by single-molecule fluorescence. Proc Natl Acad Sci USA 106(14):5645–5650. Fink AL (2006) The aggregation and fibrillation of alpha-synuclein. Acc Chem Res 39(9):628– 634. Finkelstein A (1976) Water and nonelectrolyte permeability of lipid bilayer membranes. J Gen Physiol 68(2):127–135. Franck JM, Pavlova A, Scott JA, Han S (2013) Quantitative cw Overhauser effect dynamic nuclear polarization for the analysis of local water dynamics. Prog Nucl Magn Reson, 10.1016/j.pnmrs.2013.06.001. Frazier AA, et al. (2002) Membrane orientation and position of the C2 domain from cPLA2 by site-directed spin labeling. Biochemistry 41(20):6282–6292. Frazier AA, Roller CR, Havelka JJ, Hinderliter A, Cafiso DS (2003) Membrane-bound orientation and position of the synaptotagmin I C2A domain by site-directed spin labeling. Biochemistry 42(1):96–105. Hausser KH, Stehlik D (1968) Dynamic nuclear polarization in liquids. Adv Magn Reson 3:79– 139. 209 von Heijne G (2006) Membrane-protein topology. Nat Rev Mol Cell Biol 7(12):909–918. Hodges MW, Cafiso DS, Polnaszek CF, Lester CC, Bryant RG (1997) Water translational motion at the bilayer interface: An NMR relaxation dispersion measurement. Biophys J 73(5):2575–2579. Hubbell WL, Cafiso DS, Altenbach C (2000) Identifying conformational changes with site- directed spin labeling. Nat Struct Biol 7(9):735–739. Hussain S, Franck JM, Han S (2013) Transmembrane protein activation refined by sitespecific hydration dynamics. Angew Chem Int Ed Engl 52(7):1953–1958. Isas JM, Langen R, Hubbell WL, Haigler HT (2004) Structure and dynamics of a helical hairpin that mediates calcium-dependent membrane binding of annexin B12. J Biol Chem 279(31):32492–32498. Isas JM, et al. (2005) Calcium- and membrane-induced changes in the structure and dynamics of three helical hairpins in annexin B12. Biochemistry 44(50):16435–16444. Jansson H, Kargl F, Fernandez-Alonso F, Swenson J (2009) Dynamics of a protein and its surrounding environment: A quasielastic neutron scattering study of myoglobin in water and glycerol mixtures. J Chem Phys 130(20):205101. Jao CC, Der-Sarkissian A, Chen J, Langen R (2004) Structure of membrane-bound alphasynuclein studied by site-directed spin labeling. Proc Natl Acad Sci USA 101(22):8331–8336. Jao CC, Hegde BG, Chen J, Haworth IS, Langen R (2008) Structure of membrane bound alpha- synuclein from site-directed spin labeling and computational refinement. Proc Natl Acad Sci USA 105(50):19666–19671. Jeschke G (2012) DEER distance measurements on proteins. Annu Rev Phys Chem 63:419–446. Kausik R, Han S (2011) Dynamics and state of lipid bilayer-internal water unraveled with solution state 1H dynamic nuclear polarization. Phys Chem Chem Phys 13(17):7732– 7746. Kunding AH, et al. (2011) Intermembrane docking reactions are regulated by membrane curvature. Biophys J 101(11):2693–2703. Lemmon MA (2008) Membrane recognition by phospholipid-binding domains. Nat Rev Mol Cell Biol 9(2):99–111. Levitan K, et al. (2011) Conserved C-terminal charge exerts a profound influence on the aggregation rate of α-synuclein. J Mol Biol 411(2):329–333. 210 Lokappa SB, Ulmer TS (2011) Alpha-synuclein populates both elongated and broken helix states on small unilamellar vesicles. J Biol Chem 286(24):21450–21457. MacCallum JL, Tieleman DP (2011) Hydrophobicity scales: A thermodynamic looking glass into lipid-protein interactions. Trends Biochem Sci 36(12):653–662. Ortony JH, et al. (2011) Probing the hydration water diffusion of macromolecular surfaces and interfaces. New J Phys 13(1):015006. Outeiro TF, Lindquist S (2003) Yeast cells provide insight into alpha-synuclein biology and pathobiology. Science 302(5651):1772–1775. Perozo E, Cortes DM, Cuello LG (1998) Three-dimensional architecture and gating mechanism of a K+ channel studied by EPR spectroscopy. Nat Struct Biol 5(6):459–469. Pfefferkorn CM, et al. (2012) Depth of α-synuclein in a bilayer determined by fluorescence, neutron reflectometry, and computation. Biophys J 102(3):613–621. Polnaszek CF, Bryant RG (1984) Nitroxide radical induced solvent proton relaxation: Measurement of localized translational diffusion. J Chem Phys 81(9):4038–4045. Raschle T, Hiller S, Etzkorn M, Wagner G (2010) Nonmicellar systems for solution NMR spectroscopy of membrane proteins. Curr Opin Struct Biol 20(4):471–479. Reeves JP, Dowben RM (1970) Water permeability of phospholipid vesicles. J Membr Biol 3(2):123–141. Robotta M, et al. (2011) Direct evidence of coexisting horseshoe and extended helix conformations of membrane-bound alpha-synuclein. ChemPhysChem 12(2):267–269. Rosenbaum DM, Rasmussen SG, Kobilka BK (2009) The structure and function of G-protein- coupled receptors. Nature 459(7245):356–363. Russo D, Hura G, Head-Gordon T (2004) Hydration dynamics near a model protein surface. Biophys J 86(3):1852–1862. Tobias DJ, Tu KC, Klein ML (1997) Atomic-scale molecular dynamics simulations of lipid membranes. Curr Opin Colloid Interface Sci 2(1):15–26. Ulmer TS, Bax A, Cole NB, Nussbaum RL (2005) Structure and dynamics of micellebound human alpha-synuclein. J Biol Chem 280(10):9595–9603. Varkey J, et al. (2010) Membrane curvature induction and tubulation are common features of synucleins and apolipoproteins. J Biol Chem 285(42):32486–32493. 211 White SH, von Heijne G (2008) How translocons select transmembrane helices. Annu Rev Biophys 37:23–42. Wimley WC, White H (1996) Experimentally determined hydrophobicity scale for proteins at membrane interfaces. Nat Struct Biol 3(10):842–848. 212 CONCLUDING REMARKS AND DISCUSSION The structures and regulatory mechanisms underlying how proteins generate curvature in the cell are slowly becoming clear. In this thesis, we evaluate how proteins known to be important for membrane remodeling-dependent processes in the cell are capable of bending membranes. It is important to understand how these proteins generate membrane curvature in processes such as synaptic endocytosis because the breakdown of this function can lead to neurodegeneration. To accomplish a structural characterization of proteins bound to different membrane shapes we utilized a SDSL and EPR approach (Chapter 1). We find BAR proteins endophilin A1 and amphiphysin to generate lipid vesicles or lipid tubes using different membrane-bound structures and mechanisms (Chapters 2 and 3). On vesicles, these proteins bind membranes using their helical N termini and not their BAR domain. On tubes, BAR proteins scaffold the membrane with the BAR domain and deeply insert their helical N termini. Furthermore, we find that the membrane interaction of these proteins, and many more membrane binding domains, can be regulated by post translational modifications. Phosphorylation was observed to directly regulate the shape that endophilin A1 produces (Chapter 3). Lysine acetylation, a previously unconsidered modification for regulating how proteins interact with membranes, was found to be highly localized in membrane-binding regions of peripheral membrane proteins (Chapter 4). Using acetylation-mimicking mutations, we find that this modification can significantly change the way multiple protein domains interact with membranes in vitro and in the cell. This thesis also includes a characterization of MBP as a membrane curving protein, a function previously unassociated with this protein (Chapter 5). Post-translational modification by arginine citrullination of MBP results in decreased membrane interaction. Therefore, the overarching theme to this thesis is that membrane curving proteins 213 utilize different mechanisms to produce different shapes and that these mechanisms are regulated by post-translational modification. There are several systems suggested to explain how proteins bend membranes (McMahon and Gallop, 2005), but the predominant mechanisms are helical wedging and protein scaffolding (Drin and Antonny, 2010; Qualmann et al., 2011). In our structural characterization of BAR proteins endophilin A1 and amphiphysin bound to vesicles or tubes, find that these two mechanisms have context-dependent importance. Based on the crystal structures which do not resolve the amphipathic N terminus of amphiphysin or the N terminus and insert helices of endophilin A1, these proteins were first hypothesized to bend membranes using their curved BAR domain as a scaffold (Gallop et al., 2006; Peter et al., 2004). As described in chapters 2 and 3, the BAR domain of endophilin A1 and amphiphysin were determined by EPR accessibility measurements to be relatively distal from the membrane when bound to vesicles. Similar measurements of the N terminus of amphiphysin as well as the N terminus and insert helices of endophilin A1 determined that these regions fold into amphipathic helices and wedge themselves shallowly into the membrane. When similar experiments were performed on tubes, the BAR domains bound the membrane with their concave surface and the amphipathic helical regions submerged into the acyl chains of the bilayer. These experiments revealed that BAR proteins endophilin A1 and amphiphysin induce vesicles using helical wedging and not scaffolding, while forming tubes by engaging their scaffolds to the membrane. Therefore, BAR proteins use different mechanisms to generate distinct membrane shapes. The variable depth at which the helical wedges submerge also affects the level of curvature they produce, as predicted by computational analysis (Campelo et al., 2008). 214 While scaffolding was found to be important for how BAR proteins generate membrane tubes, non BAR domain containing proteins such as α-synuclein have been observed to generate tubes using a long extended helical wedge (Mizuno et al., 2012; Varkey et al., 2010). This indicates that scaffolding is not required for the generation of membrane tubes for all proteins. MBP is a similar protein capable of folding into amphipathic helical wedges on membranes as well as inducing the formation of membrane tubes (Chapter 5). Unlike the BAR proteins, however, the characterization of MBP in this thesis was limited to defining its ability to induce curvature and a detailed SDSL and EPR structural analysis was not implemented. Therefore, we cannot conclude as to what exact mechanism MBP uses to induce curvature. Nonetheless, MBP has been shown by other groups to contain amphipathic helical segments and our analysis in chapter 5 shows that it shares many characteristics with known membrane curving proteins such as synucleins and IAPP. The structural characterization of endophilin A1 has not only revealed that different structures and mechanisms are used to generate different shapes but that post-translational modification may directly regulate these mechanisms. Vesiculation was found to be favored over tubulation in a phosphomimetic mutation at the physiological phosphorylation site S75 (Chapter 3). We propose that the addition of a negative charge at residue S75, which is positioned in the membrane on tubes, creates a highly unfavorable structure. This negative charge is more likely to be stable in the aqueous environment and at depths that more closely resemble the structure observed for endophilin bound to small vesicles. Work from another group found that this phosphorylation site is part of a regulation cycle and that aberrant modification significantly inhibited endocytosis (Matta et al., 2012). Therefore, phosphorylation of endophilin A1 regulates its ability to curve membranes in vitro and affects its function in vivo. To discover if other post- 215 translational modifications are localized in important membrane-binding regions of membrane curving proteins we utilized the post-translational modifications data bank (Hornbeck et al., 2012). Lysine acetylation, a modification that has not been investigated for its role in protein- membrane interactions, was found to be highly localized in important membrane binding regions of the lipid binding domains of BAR, PX, C2, and Eps15-homology domains (Chapter 4). Similar to phosphorylation, lysine acetylation modifies the charge of the protein and may modify the electrostatic interactions between protein domains and the membrane. In chapter 4, we show that acetylation-mimicking mutations directly affect the interaction various lipid binding domains have with membranes. Therefore, this thesis provides the first evidence that lysine acetylation may be a mechanism by which wide-range of membrane binding proteins are regulated in the cell. These post-translational modifications may also play important roles in disease. MBP is aberrantly post-translational modified in Multiple Sclerosis, where an increase in citrullination is observed. Using citrullination-mimicking mutations (Bates et al., 2002), we found that MBP‘s ability to induce curvature is inhibited by these modifications. Taken with previous evidence showing that aberrant post-translational modification of endophilin is implicated in Parkinson‘s disease, it seems that post-translational modifications may be a vital means by which membrane-curving proteins are regulated in the cell. Due to the novelty of finding lysine acetylation to be important for protein-membrane interactions, there are many potential future studies that can be based on this work. Future in vivo analysis will have to determine whether lysine acetylation is a mechanism that is commonly used to regulate a wide-range of membrane-remodeling events. Lysine mutations or aberrant modification of membrane curving proteins may also play significant roles in disease and is a focus for future work. Using the same analysis we used in chapter 3 to determine the influence of 216 phosphorylation on helical depth in the membrane, the structural consequences of lysine acetylation could be determined. Future work would also require the elucidation of the acetylases, deacetylases, and signaling pathways that are responsible for regulating membrane curving proteins in vivo. 217 REFERENCES Bates, I.R., Libich, D.S., Wood, D.D., Moscarello, M.A., and Harauz, G. (2002). An Arg/Lys-- >Gln mutant of recombinant murine myelin basic protein as a mimic of the deiminated form implicated in multiple sclerosis. Protein Expr. Purif. 25, 330–341. Campelo, F., McMahon, H.T., and Kozlov, M.M. (2008). The hydrophobic insertion mechanism of membrane curvature generation by proteins. Biophys. J. 95, 2325–2339. Drin, G., and Antonny, B. (2010). Amphipathic helices and membrane curvature. FEBS Lett. 584, 1840–1847. Gallop, J.L., Jao, C.C., Kent, H.M., Butler, P.J.G., Evans, P.R., Langen, R., and McMahon, H.T. (2006). Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J. 25, 2898–2910. Hornbeck, P.V., Kornhauser, J.M., Tkachev, S., Zhang, B., Skrzypek, E., Murray, B., Latham, V., and Sullivan, M. (2012). PhosphoSitePlus: a comprehensive resource for investigating the structure and function of experimentally determined post-translational modifications in man and mouse. Nucleic Acids Res. 40, D261–D270. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., De Bock, P.-J., Morais, V.A., Vilain, S., Haddad, D., et al. (2012). LRRK2 controls an EndoA phosphorylation cycle in synaptic endocytosis. Neuron 75, 1008–1021. McMahon, H.T., and Gallop, J.L. (2005). Membrane curvature and mechanisms of dynamic cell membrane remodelling. Nature 438, 590–596. Mizuno, N., Varkey, J., Kegulian, N.C., Hegde, B.G., Cheng, N., Langen, R., and Steven, A.C. (2012). Remodeling of lipid vesicles into cylindrical micelles by α-synuclein in an extended α-helical conformation. J. Biol. Chem. 287, 29301–29311. Peter, B.J., Kent, H.M., Mills, I.G., Vallis, Y., Butler, P.J.G., Evans, P.R., and McMahon, H.T. (2004). BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 303, 495–499. Qualmann, B., Koch, D., and Kessels, M.M. (2011). Let‘s go bananas: revisiting the endocytic BAR code. EMBO J. 30, 3501–3515. Varkey, J., Isas, J.M., Mizuno, N., Jensen, M.B., Bhatia, V.K., Jao, C.C., Petrlova, J., Voss, J.C., Stamou, D.G., Steven, A.C., et al. (2010). Membrane curvature induction and tubulation are common features of synucleins and apolipoproteins. J. Biol. Chem. 285, 32486– 32493.

Abstract (if available)

Abstract Membrane curvature is an essential biophysical property utilized within cells to execute their most basic functions. Recent advances in this field have shown proteins capable of shaping membranes to be the major driving force behind these remodeling events. These proteins are highly regulated by cells and are recruited to remodeling events in a location, time, and function specific manner. The Bin/Amphiphysin/Rvs (BAR) superfamily of proteins are banana-shaped homodimers which are involved in remodeling membranes during endocytosis. BAR proteins endophilin A1 and amphiphysin have been shown to recruit important endocytosis-specific co-factors to the membrane as well as directly act in membrane remodeling during this process. In vitro, these proteins are observed to bind large vesicles and convert them into highly curved structures such as small vesicles or cylindrical tubes. While previous studies have shown these proteins to be able to curve membranes, the underlying mechanisms of this process are not fully understood. Furthermore, how cells regulate the shape or curvature a protein induces is unclear. ❧ The central thesis of this work is using biophysical tools to elucidate the structures and mechanisms protein use to bend membranes as well as to evaluate how these mechanisms may be regulated in the cell. Our initial studies evaluated the shape-dependent structures that membrane curving proteins endophilin A1 and amphiphysin have on lipid vesicles or tubes using site-directed spin labeling and electron paramagnetic resonance. These studies revealed that these proteins use different mechanisms and structures to generate distinct membrane shapes. We additionally investigated the affect of known post-translation modifications of these proteins. Phosphorylation of endophilin A1 was found to regulate the membrane shape this protein generates. Using a proteomics approach, lysine acetylation, a modification previously not considered to be important for protein-membrane interactions, is observed to be localized in membrane-binding regions of peripheral membrane proteins. Biophysical characterization of acetylation-mimicking mutations in proteins amphiphysin, EHD2, and synaptotagmin show that this modification is capable of regulating the type of curvature these proteins generate in cells and in vitro by reducing their overall affinity for the membrane. ❧ While the field is rapidly growing, new membrane curving proteins and processes are still being discovered. Myelin, defined as concentric membrane wraps enveloping neuronal axons, are an essential component of the mammalian nervous system. How these membranes wrap and remain tightly curved around an axon is not completely understood. Loosely compacted myelin is a hallmark of the neurodegenerative disease Multiple Sclerosis. Myelin is mostly composed of lipids and two proteins, myelin basic protein and the proteolipid protein. Myelin basic protein is known as the executive protein in myelination and knockout results in loosely compacted and defective myelin wraps. Using a biophysical characterization of myelin basic protein, we find this protein to be able to induce curvature similar to that of other known membrane curving proteins. Myelin basic protein-induced tubes exhibit intermembrane adhesion and the ability to wrap neighboring vesicles, or form extensive multi-layered tube networks. Finally, we find that the Multiple-sclerosis related aberrant post-translational modification of myelin basic protein results in a decreased ability of this protein to generate curvature. We therefore conclude that this protein is a membrane curving protein and that this function may be inhibited in disease.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Membrane curvature sensors and inducers studied by site-directed spin labeling

PDF

Before they were amyloid: understanding the toxicity of disease-associated monomers and oligomers prior to their aggregation

PDF

The folding, misfolding and refolding of membrane proteins and the design of a phosphatidylserine-specific membrane sensor

PDF

Structural studies on functional amyloids and the mechanism of aggregation and disaggregation of Huntingtin Exon 1

PDF

Molecular basis of CD33 receptor function, and effects of a destabilizing transmembrane motif on αXβ2 integrin

PDF

Protein-protein interaction and protein-lipid interaction of membrane proteins

PDF

Uncovering the influence of N-terminal phosphorylation on conformational dynamics of huntingtin exon 1 monomer

PDF

Dual effects of transmembrane proline residues on integrin function

PDF

Oligomer formation of functional amyloid protein - Orb2A

PDF

Structural studies of the IAPP membrane-mediated aggregation pathway

PDF

Regulating functional amyloid formation: the promiscuous behavior of the Orb2A N-terminal amphipathic region

PDF

Structure and kinetics of the Orb2 functional amyloid

PDF

Expression and purification of extracellular domain of human CD33 in Escherichia coli and Pichia pastoris

PDF

The effect of familial mutants of Parkinson's disease on membrane remodeling

PDF

Role of CD33 structure and function in Alzheimer’s disease

PDF

Protein phosphatase 2A and annexin A5: modulators of cellular functions

PDF

Structural characterization of the functional amyloid Orb2A using EPR spectroscopy

PDF

Role of mitochondrially derived peptides in the inhibition of IAPP misfolding

PDF

Functional role of chromatin remodeler proteins in cancer biology

PDF

Study of integrin αIIbβ3 transmembrane and cytosolic domain interaction and the expression of extracellular domain by Pichia Pastoris

Asset Metadata

Creator Ambroso, Mark Robert (author)

Core Title The structure and function of membrane curving proteins on different membrane shapes and their regulation by post-translational modifications

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Genetic, Molecular and Cellular Biology

Publication Date 07/29/2016

Defense Date 04/27/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag BAR proteins,electron paramagnetic resonance,lysine acetylation,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Haworth, Ian S. (committee chair), Langen, Ralf (committee member), Okamoto, Curtis Toshio (committee member), Siemer, Ansgar (committee member), Ulmer, Tobias (committee member)

Creator Email ambroso@usc.edu,markambroso@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-622126

Unique identifier UC11303059

Identifier etd-AmbrosoMar-3766.pdf (filename),usctheses-c3-622126 (legacy record id)

Legacy Identifier etd-AmbrosoMar-3766.pdf

Dmrecord 622126

Document Type Dissertation

Format application/pdf (imt)

Rights Ambroso, Mark Robert

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA