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

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content

MEMBRANE CURVATURE SENSORS AND INDUCERS STUDIED BY
SITE-DIRECTED SPIN LABELING

by
Christine Chua Jao

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOCHEMISTRY AND MOLECULAR BIOLOGY)

May 2010

Copyright 2010 Christine Chua Jao

ii
Dedication
For my family
iii
Acknowledgements
I would like to thank my adviser, Dr. Ralf Langen, for his mentorship, advice and
support during my years in his laboratory. I would also like to thank Dr. Jeannie Chen
for letting me work in her lab, and for her advice and support. Additionally, I would like
to thank my committee members Drs. Ian Haworth, Tobias Ulmer, and Jonah Chan for
their comments, suggestions, and advice.
My time in the lab has been made more enjoyable by the friends I made. I would like
to thank J. Mario Isas for his mentorship and friendship, which started during our days at
UCI. I would like to thank Francis Concepcion, Thuzar Shin, Volker Luibl, Angela
Roca, Ana Mendez, Torsten Fischer, Yukihiro Koike, Brian Soreghan, and other current
and former lab members for their friendship. I would like to thank Ani der-Sarkissian for
her help with the alpha-synuclein project.
I would like to thank the Cellular, Biochemical, and Molecular Biology Training
Grant for funding.
Lastly, I would like to thank my family, especially my husband, for their unwavering
support.
iv
Table of Contents
Dedication ii
Acknowledgements iii
List of Tables viii
List of Figures ix
Abstract xii

Chapter 1: Introduction

1.1. Membrane Curvature in the Cell 1
1.1.1. Introduction 1
1.1.2. Lipid Composition and Membrane Curvature 4
1.1.3. Protein Structure and Membrane Curvature 6
1.1.3.1.Scaffolding 6
1.1.3.2.Helix Insertion 8
1.1.3.3.Effect of Membrane Curvature on Protein
Structure: Inside-Out Refolding 10
1.2. Curvature Sensors and Inducers 11
1.2.1. α-Synuclein, A Curvature Sensor 11
1.2.2. Epsin, A Curvature Inducer 17
1.2.3. Endophilin, A Curvature Inducer 20
1.3. Electron Paramagnetic Resonance 24
1.3.1. Introduction 24
1.3.2. Hyperfine Interaction 25
1.3.3. Site-Directed Spin Labeling 25
1.3.4. Mobility 27
1.3.5. Accessibility 29
1.3.6. Distances 31
1.3.6.1.Continuous Wave (CW) 31
1.3.6.2.Double Electron-Electron Resonance (DEER) 32
1.4. Structural Refinement 34
References 37

Chapter 2: Structure of Membrane-Bound α-Synuclein from Site-Directed
Spin labeling and Structural Refinement 47

Abstract 48
2.1. Introduction 49
2.2. Results and Discussion 51

v

2.2.1. Local Secondary Structure and Membrane Topology
Information from Continuous Wave EPR
Spectroscopy 51
2.2.2. Intramolecular Distances from Four-Pulse DEER
Experiments 55
2.2.3. Structural Refinement 64
2.3. Conclusions 81
2.4. Methods 82
2.4.1. Preparation of Spin-Labeled α-Synuclein Derivatives 82
2.4.2. Vesicle Preparation 82
2.4.3. Continuous Wave EPR 83
2.4.4. Pulsed EPR and Distance Analysis 83
2.4.5. Computational Structural Refinement 84
2.4.5.1.Structure Building 84
2.4.5.2.Simulated Annealing Molecular Dynamics
(SAMD) Calculations 85
2.4.5.3.Validation of Results 87
2.4.6. Interaction with SDS Micelles 89
2.4.7. Dye Leakage Assays 90
2.4.8. Gel Filtration 90
References 92

Chapter 3: Membrane Binding and Self-Association of Epsin ENTH
Domains 97

Abstract 98
3.1. Introduction 99
3.2. Materials and Methods 103
3.2.1. Generation of Epsin 1 ENTH (1-164) Cysteine
Mutants 103
3.2.2. Purification of Wildtype and Mutant Rat Epsin 1
ENTH 103
3.2.3. Preparation of Liposomes 104
3.2.4. Preparation of Pre-made Tubes 104
3.2.5. Spin Labeling of Single Cysteine Proteins 105
3.2.6. Sedimentation Assays and Electron Paramagnetic
Resonance Experiments 105
3.2.7. Depth Calibration 106
3.2.8. MD Simulation 107
3.2.9. MD System Preparation 108
3.3. Results and Discussion 109
3.3.1. The N-terminus of the Epsin 1 ENTH Domain Becomes
Structured upon Membrane Interaction 109

vi
3.3.2. Secondary Structure and Membrane Topology of
Liposome-Bound H
0
112
3.3.3. Structural Features of H
0
When Epsin is Bound to
Pre-formed Tubes 115
3.3.4. MD Simulation of ENTH Domains 121
3.3.5. ENTH Monomer and Dimer Molecular Dynamics
Simulation Results 126
3.4. Summary and Conclusions 131
References 134

Chapter 4: Mechanism of Endophilin N-BAR Domain-Mediated
Membrane Curvature 140

Abstract 141
4.1. Introduction 141
4.2. Results 147
4.2.1. An N-terminal Amphipathic Helix of Endophilin
Folds and Inserts into Membranes 147
4.2.2. BAR Domain Structure 150
4.2.3. Amphipathic Helices and the BAR Domain Collaborate
To Effect Membrane Curvature 163
4.2.4. Creation of High Curvature Membranes Promotes
Membrane Fusion 167
4.2.5. Model for Endophilin Membrane Binding and Curvature 171
4.3. Discussion 175
4.4. Experimental Procedures 179
4.4.1. Constructs and Liposome Assays 179
4.4.2. CD Spectroscopy 179
4.4.3. Analytical Ultracentrifugation 180
4.4.4. Crystallography of Endophilin N-BAR Domain 181
4.4.5. Electron Paramagnetic Resonance (EPR) 182
4.4.6. FRET Assay of Membrane Fusion 194
References 186

Chapter 5: The Role of Amphipathic Helices and the BAR domain in the
Membrane Curvature Generation of Endophilin 191

Abstract 192
5.1. Introduction 192
5.2. Materials and Methods 195
5.2.1. Preparation of Spin-Labeled Rat Endophilin
Derivatives 195
5.2.2. Liposome Preparation 195
5.2.3. Continuous Wave and EPR Experiments 195
5.2.4. Pulsed EPR and Distance Analysis 196
vii
5.3. Results 197
5.3.1 Rat Endophilin A1 BAR is a Dimer Upon Membrane
Interaction 197
5.3.2. The Concave Surface of the BAR Domain Does Not
Penetrate the Membrane 199
5.3.3. The Insert Region Becomes an Amphipathic Helix
Upon Membrane Interaction 204
5.3.4. The Dimeric Amphipathic Helices are Anti-Parallel,
and Perpendicular to the BAR domain in
Response to Membrane Interaction 208
5.4. Discussion 211
References 218

Concluding Remarks 220

Bibliography 223

viii
List of Tables
Table 2.1. Intramolecular Distances for 17 Membrane-Bound Doubly-
Labeled Derivatives from Four-Pulse DEER Experiments 60

Table 2.2. Label Depths Used in SAMD Calculations 71

Table 2.3. Restraints Used in SAMD Calculations 72

Table 2.4. Geometry of Structures Obtained in SAMD Calculations 73
Table 3.1. The RMSD (in Å) for the ENTH Monomer and Dimer 128
Table 3.2. Comparison Between MD and EPR Results 130
Table 4.1. Data Collection, Phasing and Refinement Statistics for Crystal
Structure of Rat Endophilin A1 BAR Domain 183
Table 5.1. EPR Distances 200
Table 5.2. Intramolecular Distances Between Same Sites in the Insert Region
of Membrane-Bound Endophilin 210
Table 5.3. Comparison of Data Obtained by EPR of Membrane-Bound and
Soluble Interdimer Distances with Distances Determined
by Crystallography 212

ix
List of Figures
Figure 1.1. Local Differences in Membrane Curvature are Observed in the Cell 2
Figure 1.2. Lipid Composition and Curvature 5
Figure 1.3. Mechanisms of Membrane Curvature Generation by Proteins 7
Figure 1.4. Structure of Annexin B12 under Different Conditions 12
Figure 1.5. α-Synuclein Structure 14
Figure 1.6. Epsin 18
Figure 1.7. Structure of Amphiphysin BAR Domain 21
Figure 1.8. Electron Paramagnetic Resonance 26
Figure 1.9. EPR Coupled with Site-Directed Spin Labeling 28
Figure 1.10. Four-Pulse DEER Pattern 33
Figure 2.1. α-Synuclein Sequence and the Spin Label 50
Figure 2.2. Accessibility Measurements for Membrane-Bound α-Synuclein 53
Figure 2.3. Continuous Wave EPR Analysis of Singly-Labeled α-Synuclein
Derivatives Indicates the Formation of an Ordered and
Continuous Helical Structure 56
Figure 2.4. Intramolecular Distance Analysis for Membrane-Bound
56R1/85R1 α-Synuclein Derivative 59
Figure 2.5. Intramolecular Distances from Four-Pulse DEER Experiments 61
Figure 2.6. α-Synuclein Interaction with SUVs of Different Lipid Compositions 65
Figure 2.7. Starting Structure for the α-Synuclein (9-89) Peptide with
26 Labels 69
Figure 2.8. Lysine Depth Parameter 74
Figure 2.9. Refinement of the Membrane-Bound Structure of α-Synuclein
Using Simulated Annealing Molecular Dynamics (SAMD) with
Restraints from EPR Data 75
x
Figure 2.10. Representations of the Interaction of α-Synuclein with a Curved
Lipid Surface 78
Figure 2.11. Comparison of the Structure of Membrane-Bound α-Synuclein
to that of a Right-Handed Coiled-Coil 80
Figure 3.1. EPR Spectra of R1-Labeled Epsin 1 Derivatives 110
Figure 3.2. Plot of O
2
and NiEDDA Accessibilities for Liposome-Bound
Epsin 1 113
Figure 3.3. Accessibility Measurements Indicate the Formation of an
Amphipathic Helical Structure in the N-terminus of Epsin 1 114
Figure 3.4. Dimerization of Epsin 1 117
Figure 3.5. EPR Spectra for Selected Epsin 1 Spin Labeled Derivatives
Bound to Pre-Formed Tubules 119
Figure 3.6. MD Simulations of the ENTH Monomer 123
Figure 3.7. MD Simulations of the ENTH Dimer 125
Figure 4.1. Ordering of N-Terminal Residues of Endophilin on Membrane
Binding 145
Figure 4.2. EPR Spectra of R1-Labeled Endophilin A1 N-BAR Derivatives 146
Figure 4.3. Membrane Insertion and Orientation of Endophilin N-Terminal
Amphipathic Helix 149
Figure 4.4. Sequence Alignments and 3D models of C. elegans Endophilin A
and Human Endophilin B2 N-BAR Domains 151
Figure 4.5. Endophilin N-BAR Crystal Structure and Alignments 152
Figure 4.6. Structural Alignment of Endophilin and Amphiphysin BAR
Domains and Alignment to Nadrin Showing Close
Homology to Endophilin 156
Figure 4.7. Endophilin N-BAR Dimerization on Membrane Binding 159
Figure 4.8. Circular Dichroism of Endophilin Constructs 160
Figure 4.9. Analytical Ultracentrifugation of Rat Endophilin A1 161
xi
Figure 4.10. Endophilin has Collaborative Membrane Binding and
Tubulation Regions 164
Figure 4.11. High Membrane Curvature Promotes Membrane Fusion 169
Figure 4.12. FRET Assay for Membrane Fusion 170
Figure 4.13. Endophilin N-BAR Domain Binding to Membrane is Initially
Driven by Electrostatics 172
Figure 5.1. The Crystal-Like Dimer Structure is Retained Upon Membrane
Interaction 198
Figure 5.2. DEER and CW-EPR Data for Membrane-Bound Endophilin
Derivatives 201
Figure 5.3. EPR Spectra and Depth Parameter Measurement of Selected
Sites on the Concave Face of the Endophilin BAR Domain
Indicate that the Concave Surface Does Not Penetrate
the Hydrocarbon Layer of the Membrane 202
Figure 5.4. EPR Spectra for Residues Located in the Helix 1 Insert Indicate
Ordering of the Region in the Presence of Membrane 205
Figure 5.5. Depth Parameter Ф Plotted as Function of Residue Number 207
Figure 5.6. Four-Pulse DEER Data Obtained for Membrane-Bound
Endophilin Derivatives 209
Figure 5.7. DEER and CW EPR data for Membrane-Bound Endophilin
Derivatives located in the Insert Region 213
Figure 5.8. Model of the Insert Region in Membrane-Bound Form 216
xii
ABSTRACT
Control and regulation of membrane curvature play important roles in membrane
trafficking and remodeling events. These processes are mediated by proteins that can
sense and/or induce membrane curvature. The focus of my thesis is to understand the
underlying molecular mechanisms that enable proteins to remodel membranes. Structural
and biophysical studies were performed on the curvature-inducing proteins epsin (an
ENTH protein) and endophilin (a BAR protein), both involved in membrane remodeling
during endocytosis. Membrane interaction of α-synuclein, a curvature sensor, was also
studied.
According to EPR and site-directed spin labeling, α-synuclein takes up an elongated
helix with a helical periodicity of 11 amino acids per 3 turns ( α11/3) in the membrane-
bound form. Combining EPR and structural refinement, we found that the extended
helical structure in the presence of membranes has a superhelical twist. This may be a
result of α11/3 and allows the protein to have an elongated helical structure. The
extended helix is at the level of the phosphate headgroup where it likely compensates for
the curvature strain that is present in highly curved vesicles.
BAR domains are found among proteins involved in endocytosis and represents a
membrane-binding and curvature-inducing module. The crystal structure of the BAR
domain-containing protein endophilin show that it is a banana-shaped dimer composed of
a 6-helical bundle arranged in a coiled-coil and suggests a scaffolding mechanism for the
sensing and generation of membrane curvature. We found that many structural features
of the crystal structure dimer are retained upon membrane interaction. The data also
suggest that the BAR domain is at a distance from phosphate level of the membrane and
xiii
is more likely to interact with the outermost region of the headgroup. The two regions not
resolved in the crystal structure undergo a conformational change to a helical structure in
the presence of membrane. The centers of both helices are at the level of the phosphate
headgroup, where they are likely to promote membrane curvature by wedging lipids
apart. Membrane-bound epsin was also analyzed.
Our work shows the importance of amphipathic helices. These studies will assist in
the elucidation of the mechanism of membrane curvature regulation.

1
CHAPTER 1
INTRODUCTION

1.1. Membrane Curvature in the Cell
1.1.1. Introduction
Cell membranes play important roles in the cell. In particular, they enclose the cell
and maintain the boundary differences between intracellular and extracellular
environments. Inside the eukaryotic cell, different membrane-enclosed organelles
partition the cell such as the golgi complex, endoplasmic reticulum, the lysosomal
compartment, and the nucleus.
Regulation of membrane curvature is essential for many vital cellular functions, such
as cell division and motility, and vesicular trafficking (1). Biological membranes are
subject to constant remodeling, and the regulated control of membrane shape and
curvature is essential for many cellular functions. Figure 1.1 is an illustration of
curvatures formed during endocytosis. In the course of generating a highly curved
vesicle during endocytosis, the curvature is positive during the formation of the pit,
meaning that the membrane curves in toward the cytoplasm. When the vesicle is ready to
bud off, there is positive curvature at the dome, but a negative curvature at the neck
region (1). This process involves transient changes in membrane curvature to generate
then stabilize, highly curved vesicles. All of these curvature changes require careful and
precise control.
The main focus of this thesis is to investigate the interplay between protein structure
and membrane curvature. How does protein structure affect membrane structure, and in
2

Figure 1.1. Local differences in membrane curvature are observed in the cell. (A)
Dynamic cell membrane remodeling occurs during processes such as movement,
division, and vesicle trafficking. Areas of high positive membrane curvature are outlined
in red. (B) A budding vesicle contains different types of curvatures. (A) and (B) were
adapted from McMahon & Gallop, 2005 (1). (C) Curvature can be negative (left), zero
(center), and positive (right). These different curvatures are schematically illustrated.
Figure is from Cambridge University Press Book Resources on-line.
A B
C
3
turn, how does membrane structure affect protein structure? Two types of proteins are
important in membrane remodeling: membrane curvature sensors and membrane
curvature inducers. Curvature sensors are proteins that can recognize and bind to highly
curved vesicles, while curvature inducers such as epsin, an ENTH domain-containing
protein, and endophilin, a BAR domain-containing protein, can transform large vesicles
into smaller ones, or into highly curved tubules (2-6). Studies using transgenic animals
have shown that BAR domain-containing proteins play a role in synapse formation and
are required for T-tubule formation in muscle (7-9). Alterations in BAR functions have
been linked to various cancers and mental retardation (10-15). Recently, it was reported
that naturally occurring point mutations in amphiphysin, another BAR domain-containing
protein, are linked to familial forms of centronuclear myopathy in humans, and leads to
its reduced ability to tubulate membranes (16). Understanding the basic molecular
mechanisms by which proteins can control membrane shape and curvature will, therefore,
not only advance our understanding of this fundamental process, but will be of
significance to human disease.
Relatively little is known about the interplay between membrane curvature and protein
structure. Although x-ray crystallography has begun to provide structural information on
a number of membrane proteins, it is challenging to use this tool to study the effects of
membrane curvature on proteins since it is difficult to incorporate lipid membranes of a
defined curvature in three-dimensional crystals. Similarly, high resolution nuclear
magnetic resonance (NMR) studies cannot readily be performed on intact vesicles or lipid
tubules of defined curvature. Instead, membrane mimetic systems such as detergent
4
micelles or bicelles are commonly used. Thus, it is important to apply alternative
methods to study the relationship between protein structure and membrane curvature.
Few biophysical tools are amenable to the use of membranes. One of the few best
suited for analysis of membrane-bound protein structures is site-directed spin labeling
(SDSL), coupled with electron paramagnetic resonance (EPR). This is a reporter
technique, whereby a paramagnetic species, or spin label, reports on the environment of a
specific amino acid residue. In addition, not only can it elucidate secondary structure, it
can also tell us about membrane topology. It has been used to decipher the protein
structure of globular proteins, membrane proteins, and amyloid proteins. Structural
studies on the membrane-bound form of the curvature inducers endophilin and epsin,
both endocytic proteins, as well as the curvature sensor α-synuclein, a protein involved in
Parkinson disease, will be discussed.
1.1.2. Lipid Composition and Membrane Curvature
Lipids and proteins both play important roles in the dynamic process of membrane
remodeling. The nature of the lipid headgroup and acyl chain can play a role in
membrane curvature. Different lipids have different molecular shapes, and different
packing preferences (17, 18). A mixture of bilayer and non-bilayer lipids exist in
biological membranes. Bilayer lipids, such as phosphatidylcholine, have a cylindrical
shape where the headgroup cross-sectional area is similar to the acyl chains’ area, and
prefer bilayer structures (Figure 1.2A). Type I nonbilayer lipids have headgroups larger
than the acyl chain cross-sectional areas and thus have inverted cone shapes. For
detergents and Type I nonbilayer lipids such as lysophospholipids, structures such as
micelles are favored, as well as the outer leaflet of a vesicle, which have positive
5

Figure 1.2. Lipid composition and curvature. (A) Curvature is positive in the outer
monolayer, while curvature is negative in the inner monolayer of a vesicle. Cylindrical
shaped lipids prefer bilayers. Inverted conical shaped lipids prefer micellar structures
(positive curvature). Conical shaped lipids prefer the hexagonal II phase (negative
curvature). Figure is from Zimmerberg, 2000 (18). (B) In a bilayer, lipids may be found
in non-preferred environments, resulting in stored curvature stress.
B
A
6
curvatures (Figure 1.2A). Type II nonbilayer lipids, for example, phosphatidyl-
ethanolamine, have a cone shape where the acyl chain area is greater than the headgroup
area, and prefer structures with negative curvatures, such as the hexagonal II phase, and
the inner leaflet of vesicles (Figure 1.2A).
As each lipid has an associated curvature, a mix of bilayer/non-bilayer lipids in a
bilayer will tend to curve because of the non-uniform distribution of lateral pressure or
stress, which varies with depth (19). Sources of pressure in the vesicle include headgroup
pressure, which could be repulsive or attractive, the cohesive interfacial tension, and the
repulsive chain pressure.
In a bilayer, each monolayer leaflet with its own intrinsic curvature is forced to flatten,
resulting in leaflets with stored curvature stress and thus, stored energy (Figure 1.2B).
This curvature stress is compensated for by the favorable hydrophobic interaction
between leaflets. Lipid composition most likely plays a role in the induction of
membrane curvature.
1.1.3. Protein Structure and Membrane Curvature
1.1.3.1. Scaffolding
Peripheral membrane proteins have been proposed to effect membrane curvature by
acting as a scaffold (1, 5, 20) (Figure 1.3A). These proteins must have an intrinsic
curvature which will interact with the lipid bilayer. Moreover, for proteins to induce
curvature by this mechanism, they must also have high rigidity to overcome the lipid
bilayer’s spontaneous curvature state. In addition, the protein must also have high
affinity for the lipid polar headgroup.
7

Figure 1.3. Mechanisms of membrane curvature generation by proteins. (A) In the
scaffold mechanism, a rigid protein, or protein domain, binds to the membrane surface by
electrostatic interaction, and bends the membrane. BAR domains are hypothesized to
induce membrane curvature by this mechanism. (B) In the local spontaneous curvature
mechanism, insertion of an amphipathic helix at the lipid headgroup acts like a wedge,
pushing apart headgroups and inducing membrane curvature. Figure was adapted from
Zimmerberg & Kozlov, 2006 (5).
A
B
8
Coat proteins such as clathrin polymerize into curved structures, thereby stabilizing
highly curved vesicles (21). Clathrin acts as a cage that is not in contact with
the vesicle. Dynamins, proteins which are involved in the release of transport vesicles
during endocytosis, are proposed to wrap around membranes. Dynamins are able to self-
assemble into higher order oligomers in the absence of lipids (22). In the presence of
lipids, dynamin forms cylindrical coats similar in size to dynamin oligomers (23-25).
Membrane remodeling by dynamin most probably acts by the scaffold mechanism due to
the rigidity of the dynamin coat and its lipid binding affinity.
The recently characterized BAR domain found in many endocytic proteins has been
proposed to bend membranes by this mechanism (4). It is a banana-shaped dimer with
positive charges along its concave face. Examples of BAR-containing proteins are
amphiphysin and endophilin. Endophilin will be discussed later.
1.1.3.2. Helix Insertion
There are at least two mechanisms by which wedging of the amphipathic helices in
curvature-inducing proteins promotes membrane curvature. In the local spontaneous
curvature mechanism, shallow insertion of an amphipathic helix into the upper part of the
membrane monolayer drives curvature generation (1, 5, 20) (Figure 1.3B). The helix acts
like a wedge, and results in reorientation of lipid headgroups and membrane deformation
to accommodate the presence of the helix. In the bilayer couple mechanism, the helices
are added to only on the outer leaflet (26), the additional material in just one leaflet
causes an imbalance between the leaflets, resulting in curvature.
Endophilin, as well as amphiphysin, belong to a class of BAR domain-containing
proteins which have N-terminal regions that are proposed to become amphipathic helices
9
in the presence of membranes. Both of these play a role in endocytosis. Epsin, a protein
with a conserved ENTH domain, is involved in clathrin-mediated endocytosis. The N-
terminal region of epsin is also proposed to become an amphipathic helix in the presence
of membrane, and will be discussed later (Chapter 3). In addition to endophilin and
amphiphysin, proteins involved in intracellular membrane trafficking such as Arf, Arl,
and Sar family of GTPases (1, 27-30), are proposed to act by the helix insertion
mechanism. Formation of an amphipathic helix may be a general mechanism for
membrane curvature sensing and induction.
In the recently published hydrophobic mechanism of membrane curvature (31), the
authors predict, by using computational methods, that insertion of amphipathic helices is
sufficient for the tubulation of membranes. For a range of insertion depths, less than 15%
of the tubule area is occupied by the inserted helices.
Studies on large unilamellar vesicles support the notion that differences in monolayer
surface areas can affect lipid morphologies. After transbilayer asymmetry was generated
by translocation of phosphatidylglycerol to the outer monolayer, tube formation was
observed (32). Addition of lysophosphatidylcholine, which partitions into the outer
monolayer thereby increasing outer monolayer area, results in tubules. Asymmetries in
the bilayer, due to lipid asymmetry or helix insertion, are likely mechanisms for
membrane curvature generation.
Helix insertion may also be a mechanism of curvature sensing. Arfs (ADP-
ribosylation factors) and related GTPases are important in membrane trafficking. Arf1
plays a role in the golgi-to-ER trafficking pathway, co-localizing with COPI vesicle coat
proteins (33, 34). Once bound with GTP, Arf1 becomes active, binds to lipid
10
membranes, and promotes COPI coat assembly, resulting in transport vesicles (35). Arf1
has also been found to be able to tubulate liposomes (36). ArfGAP1, a GTPase-
activating protein, catalyzes GTP hydrolysis, thereby disassembling the COPI coat (37).
GTP hydrolysis rate, as well as rate of COPI disassembly, were shown to increase, as
curvature of the bilayer increased (29). They later identified a lipid packing sensor motif,
which is an amphipathic α-helix rich in serines and threonines (28, 38). Helix insertion is
likely a mechanism for curvature sensing.
1.1.3.3. Effect of Membrane Curvature on Protein Structure: Inside-Out Refolding
Curvature can affect protein structure dramatically, as illustrated by work on annexins.
We have seen that proteins undergo major transformations during interactions with
membranes. These interactions can be modulated by pH, phospholipid content, and
vesicle size, among a few examples. As in the case of annexins, the structure found in
the crystal structure of the soluble form may not necessarily be the membrane-bound
form. It is important to test the hypothesis by other techniques.
Annexins are highly conserved soluble proteins that interact with membranes in a
Ca
+2
-dependent manner (39). Though their function has been implicated in membrane-
related events, their physiological role is thus far unknown. Structural studies have been
done to understand their physiological role. Studies in our laboratory have shown the
annexin B12 membrane-bound structure can be modulated by phospholipid and also pH.
At neutral pH in the presence of Ca
+2
, annexin forms a trimer that is peripherally bound
to large unilamellar vesicles (LUVs) (40-42). However, calcium-independent membrane-
bound forms of annexin have been observed. At low or mildly acidic pH, annexin
transforms into a transmembrane protein by an inside-out refolding of a helical hairpin
11
(42-44). This transmembrane form of annexin can also be seen by modulating
phospholipid content at an elevated pH, such as by using cardiolipin (45).
Recently, a peripherally-bound form of annexin B12 was observed using EPR (46).
The helical hairpin (residues 139-155) in the soluble form transformed to a peripherally
membrane-bound amphipathic helix in the presence of LUVs, and this form can be
modulated by pH and phospholipid content (Figure 1.4). This form is proposed to be a
kinetic intermediate during transmembrane insertion.
Membrane curvature was analyzed to study if this can also modulate protein-
membrane interactions of annexin. A membrane curvature dependent form of annexin
was recently identified in our lab. Annexin can also bind to small unilamellar vesicles
(SUVs), vesicles with higher curvatures, in a calcium-independent manner (47). Binding
is mediated by inside-out refolding of the protein where previously buried sites in the
soluble form become membrane-exposed.

1.2. Curvature Sensors and Inducers
1.2.1. α-Synuclein, A Curvature Sensor
α-Synuclein has been implicated in several neurodegenerative diseases, including
Parkinson and Alzheimer disease. It is a 140 amino acid protein enriched in presynaptic
nerve terminals (48, 49). It exists as a natively unstructured protein in solution (50) but
upon binding to acidic phospholipids, assumes an α-helical structure (51). α-Synuclein
has been shown to preferentially interact with small unilamellar vesicles (51, 52), similar
in size to synaptic vesicles. Thus, it is a curvature sensor. Although the precise
12

Figure 1.4. Structure of annexin B12 under different conditions. (A) Annexin B12
is a monomer in solution but forms a trimer in the presence of membranes and Ca
2+
,
binding to the membrane surface (B). The corresponding helix-loop-helix structure is
highlighted in red. (C) At intermediate pH values, the helix-loop-helix structure
becomes a surface-bound amphipathic helix. (D) At acidic pHs, the same amphipathic
helix becomes a transmembrane helix. Global refolding of the protein is seen when the
protein converts from one form into another. Figure is from Fischer et al., 2007 (47).
13
physiological role of α-synuclein is not fully understood, several lines of evidence
suggest that the curvature sensing property of α-synuclein is important, as α-synuclein
has been implicated in the modulation of the presynaptic vesicle pool size (53, 54),
synaptic plasticity, the modulation of neurotransmitter release, and synaptic vesicle
recycling (55-57). In vitro studies of lipid chain dynamics using small highly curved
vesicles showed that binding of α-synuclein resulted in an increase in the phase transition
temperature of the vesicles, suggesting stabilization of the highly curved vesicle (58, 59).
Curvature stabilization maybe important in preventing premature fusion of highly curved
vesicles.
Using sequence analysis, it was recognized that the N-terminal portion of α-synuclein
was likely to mediate protein-lipid interaction (56). Seven repeats, made up of eleven
residues, were identified (Figure 1.5A). These repeats are reminiscent of those found in
apolipoproteins, proteins involved in the transport of cholesterol, and it was proposed that
α-synuclein and apolipoproteins interact with membranes in a similar fashion (51, 56)
(Figure 1.5B).
In an effort to elucidate the physiological functions as well as pathological
mechanisms of α-synuclein folding and misfolding, many studies have been undertaken
to analyze structures of membrane-bound α-synuclein. The structure of α-synuclein in
the presence of the membrane mimetic SDS, which also induce a helical structure, has
been analyzed by NMR. Studies of micelle-bound α-synuclein using NMR (60-65) have
revealed the presence of two extended helices with a break in between. Combining
paramagnetic spin labels and NMR to study micelle-bound α-synuclein showed that the
14

Figure 1.5. α-Synuclein structure. (A) Seven repeats were indentified in α-synuclein
sequence based on similarities to apolipoprotein. (B) Helical wheel analysis of the α-
synuclein sequence resulted in the identification of 5 helices. Figure is from Davidson et
al., 1998 (51). (C) The micelle-bound structure of α-synuclein determined by NMR is
composed of 2 anti-parallel helices separated by a linker. Figure is from Ulmer et al.,
2005 (66).
A B
C
15
helical region of α-synuclein is in an α-11/3 arrangement (64). A recent structural study
using NMR of micelle-bound α-synuclein reported that the two-helix structure is in an
anti-parallel arrangement, with a well-ordered linker, and that these helices are curved
(66) (Figure 1.5C).
Studies using EPR techniques have been used to probe the structure of membrane-
bound α-synuclein. A pulsed EPR study measuring distances between different residues
to probe the structure of micelle-bound α-synuclein suggest that micelle size, and hence
membrane curvature, affected inter-helix distances (62). When the inter-helix distance
was measured using SDS micelles compared with lipid micelles, which have larger
diameters, distances measured between helices are greater. Synaptic vesicles, possessing
much larger diameters, maybe just the right size that the inter-helical break observed in
micelle-bound α-synuclein reorganizes into one long, elongated helix bound to the
surface of the membrane, in an α-11/3 periodicity.
In contrast, a broken helix structure was suggested by studies using negatively charged
small unilamellar vesicles. A small fragment of α-synuclein (residues 35-43) was
analyzed by CW-EPR in the presence of SDS micelles and POPS small unilamellar
vesicles (67). This region was chosen because it is the interhelical region identified in the
micelle-bound form analyzed by NMR (66). They found little difference between the
micelle and SUV-bound form, and conclude that the membrane-bound form is a broken
helix. A small number of double mutants were analyzed by pulsed EPR in the presence
of POPG SUVs (68). Their measurements agree with a broken helix structure when
bound to vesicles.
16
In our previous study using EPR coupled with SDSL, we proposed a structural model
of membrane-bound α-synuclein wherein 11 residues make up 3 turns (69). Such a
structure might not only be applicable to α-synuclein but also to other 11-amino acid-
repeat-containing proteins such as apolipoproteins (70, 71), which wrap around lipid
particles of defined size.
In our most recent EPR study (Chapter 2), we find that the membrane-bound form of
curvature sensor α-synuclein is an elongated helix with an unusual helical periodicity of
11 amino acids per 3 turns. This is in contrast to the broken helix that resulted from
analysis of SDS micelle-bound α-synuclein by NMR. The center of the helix is at the
level of the phosphate headgroup. We combined our EPR data for use in restraints for
structural refinement. From modeling, there is a superhelical twist. This superhelical
twist may be a result of the unusual helical periodicity and allows the protein to have an
elongated helical structure. This is the first time that a structure of a membrane-bound
protein has been solved using EPR.
Using pulsed EPR methods on a limited number of sites, the Eliezer group (72) found
that α-synuclein forms an extended helix in the presence of vesicles, bicelles, and rod-like
micelles. By Forster resonance energy transfer as a method to measure distances between
residues of α-synuclein in the presence of micelles as well as vesicles, the Rhoades group
(73) found a bent-helix on micelles, and an elongated helix on vesicles. These and other
studies are in agreement with our results.
Studies using SDS micelles as membrane mimetics are important, but there is a
caveat. SDS micelles are much smaller in size when compared to synaptic vesicles. The
17
small size of the micelles may prevent α-synuclein from adopting an elongated structure.
In the cell, both forms may coexist, depending on the membrane environment. The
structure of α-synuclein may interconvert between two helices to an elongated helix
based on the curvature of the binding surface (74). Conformational switching due to
membranes and other cell components may underlie the physiological and pathological
functions of α-synuclein.
1.2.2. Epsin, A Curvature Inducer
Epsin is a multi-domain protein shown to interact with clathrin and the clathrin
adaptor protein AP-2 in the early stages of clathrin-mediated endocytosis (75). Epsin was
first identified as a binding partner for Eps15 (75), one of the clathrin coat components.
Epsin is expressed in the brain, especially the nerve terminals (75). Other mammalian
epsins have been identified, along with homologs in Drosophila and yeast (76). The first
150 amino acids in the N-terminal region is highly conserved in evolution and has been
termed ENTH (epsin N-terminal homology) domain. Binding to α- and β-adaptin
subunits of AP-2 is mediated by multiple DPW repeats and the C-terminal region
contains NPF repeats for binding EH domain-containing proteins involved in endocytosis
(75).
In vitro experiments have shown that epsin can induce membrane curvature and
tubulate liposomes (Figure 1.6) (2). In addition, when added to lipid monolayers together
with clathrin, the epsin-clathrin mixture has been shown to generate structures that
resemble clathrin coated pits. This activity is mediated by the ENTH domain (2). The
crystal structure of soluble ENTH showed that it is composed of 8 α-helices connected
by loops (77), while the N-terminal region of ENTH could not be resolved. In an NMR
18

Figure 1.6. Epsin. (A) Tubulation of liposomes by epsin ENTH domain visualized
under electron microscopy. (B) Epsin recruits clathrin to lipid monolayers containing
10% PtdIns(4,5)P
2
and stimulates clathrin lattice assembly. (C) The N-terminal region
of epsin ENTH becomes ordered when crystallized in the presence of the headgroup
inositol 1,4,5 triphosphate. (D) Crystal structure of epsin ENTH domain without
headgroup. Figures are adapted from Ford et al., 2004 (2).
19
study of the epsin ENTH domain in the presence of the headgroup inositol-1,4,5-
triphosphate (IP
3
), the N-terminal region exhibited a large chemical shift in the presence
of IP
3
, but its conformation was unknown (78). When the ENTH domain was co-
crystallized with the headgroup IP
3
, a new helix (helix 0) became ordered at the N-
terminus (2). This helix is amphipathic, with residues coordinating the headgroup IP
3

oriented towards the inside of the protein, and the hydrophobic side facing outward. The
crystal structure presents an attractive model for the generation of membrane curvature,
with the N-terminal region inserting into the membrane.
In contrast to the banana-shaped BAR domain (which will be discussed in the next
section), the ENTH domain has a “block-like” shape. How can this compact structure
induce membrane curvature? It is possible that the N-terminal ENTH domain inserts into
the membrane, as in the helix insertion mechanism (5), and the ensemble behavior of
ENTH aggregates (79) play roles in membrane curvature induction. If low density
aggregates of ENTH domain form, a weak isotropic curvature may be generated, due to
helix insertion. On the other hand, if higher density aggregates form where ENTH
domains form into dimeric or higher oligomer structures, the combination of high density
and helix insertion may induce membrane curvature generation by ENTH domain.
Using EPR coupled with SDSL, we studied the structural changes that occur in the N-
terminal region of epsin1 ENTH (Chapter 3). We show that residues 4-14 of epsin1
ENTH is an amphipathic helix in the presence of liposomes, and residues 6, 10, and 14
are lipid-exposed. We show that in the presence of liposomes, helix 0 is oriented parallel
to the membrane, and the center of the helix is at the headgroup level, in support of an
earlier EPR study (80), and the helix insertion hypothesis. Insertion of the amphipathic
20
helix may be regulated by phophorylation in vivo. We also analyzed the structure of the
N-terminal region in the presence of pre-made tubes. Our results suggest that ENTH
monomers predominate in the presence of vesicles, while at least dimeric ENTH
predominate when bound to tubes.
1.2.2. Endophilin, A Curvature Inducer
The BAR (BIN/amphiphysin/Rvs) domain has been proposed to be a dimerization,
membrane-binding, and curvature-inducing module (1, 4, 5, 20). Crystal structures for
BAR-domain containing proteins have been solved (3, 4, 81-84). The BAR domain was
first identified as conserved modules in yeast proteins (Rvs, reduced viability upon
starvation) and the metazoan amphiphysin/BIN (bridging interactor) proteins (4, 85, 86).
Since then, the BAR domain superfamily has been identified in other proteins important
for membrane remodeling in cellular processes.
The BAR domain is a banana-shaped dimer composed of a 6-helical bundle arranged
in a coiled-coil. The concave surface of the dimer is proposed to fit a curved membrane
with a diameter of 22nm (4), similar in size to synaptic vesicles (Figure 1.7). The crystal
structure of the BAR domain suggests a scaffolding mechanism for the sensing and
generation of membrane curvature whereby the concave surface of the dimer is a rigid
scaffold that molds the membrane surface by electrostatic interaction (1, 4, 5). However,
the structure of the membrane-bound BAR domain is not yet known and there is no direct
structural evidence for this hypothesis yet.
Although the scaffolding shape of the BAR domain is thought to be important for the
generation of membrane curvature in endophilin also, it has become apparent that the
21

Figure 1.7. Structure of the amphiphysin BAR domain. (A) The BAR domain is a
banana-shaped homodimer (purple and green), 6-helical bundle arranged in a coiled-coil,
with a concave surface estimated to fit a curved membrane of 22nm diameter (4). (B)
The electrostatic equipotential surface is colored red for negative and blue for positive.
Positively charged clusters are located on the concave surface and are positioned to
interact with negatively charged membranes. Figure is from Zimmerberg & McLaughlin,
2004 (6).
22
BAR domain-dependent curvature generation is coupled to additional structural elements.
In addition to a BAR domain, an N-terminal amphipathic helix that inserts into the
membrane is proposed to drive curvature generation. The N-BAR domain, or the
combination of an α-helix and a scaffold, is one model for the structure of curvature
sensor/inducer modules. Examples of N-BAR containing proteins include the endocytic
proteins endophilin and amphiphysin. Endophilins are endocytic proteins implicated in
the process of synaptic vesicle retrieval (3, 87-89), correlating well with their ability to
sense and induce membrane curvature. Amphiphysins are very similar in structure to
endophilins. These are considered the ‘classical’ BAR domains.
BAR domain superfamily members are able to produce different degrees of membrane
curvature. F-BAR proteins have an FCH or EFC (Extended-FCH domain), examples of
which include members of the Toca family (transducer of Cdc42-dependent actin
assembly) (90). Crystal structures of F-BAR proteins show that they have an elongated
crescent shape (81). Tubes that are produced from liposomes have larger diameters than
from N-BAR proteins (4, 79). I-BAR (inverse-BAR) proteins have a nearly flat zeppelin
shape, and produce membrane protrusions, which are opposite in direction to the
curvature produced by N-BAR and F-BAR proteins (82). BAR superfamily members
share the characteristic of having coiled-coil dimers with positively charged clusters for
interaction with negatively charged lipid headgroups.
Computational methods have been performed by several groups to study the
mechanism of membrane remodeling by proteins. Using amphiphysin as the model N-
BAR, membrane binding of the BAR concave surface was observed by molecular
dynamics simulations (91, 92). Local membrane curvature matched that of the BAR
23
domain. Studies which only analyzed the N-terminal regions (helix 0) favor the
scaffolding mechanism (93) and the helix insertion mechanism (31, 94).
Our work on the N-terminal region of endophilin revealed that this region transforms
to an amphipathic helix that inserts into the membrane at the level of the headgroup (3)
(Chapter 4). This N-terminal amphipathic helix was not observed in the crystal for either
amphiphysin or endophilin as it is highly disordered. Our work also added molecular
detail to recent findings. An increase in α-helicity in the protein was observed in the
presence of membranes for both amphiphysin (4) and endophilin (3). Masuda and
colleagues (95) showed that removal of endophilin helix 0 abolished liposome binding
activity, and consequently, liposome tubulation activity, in support of a previous study
(86). Peter and colleagues (4) found that removal of the N-terminal region in
amphiphysin also disrupted liposome tubulation activity.
Analysis of the crystal structure of endophilin (3, 83) revealed an insert (helix 1 insert)
in the center of the concave surface. This insert is not found in other BAR domain
proteins such as amphiphysin. Modeling of the helix 1 insert (residues 62-79) on a
helical wheel produced an amphipathic helix. Hydrophobic residues are clustered on one
side flanked by charged residues. We analyzed the structure of this insert region by
EPR, and found that it is an amphipathic helix in the presence of membranes (Chapter 5).
Further, we analyzed depth of insertion of the helix into the membrane, and found that
the center of the helix is at the level of the phosphate headgroup. Fluorescence resonance
energy transfer (FRET) experiments suggested that this helix 1 insert is in contact with
the membrane (95), in support of our findings. By pulsed methods, we found that the
region in the middle of helix 1 is a helix arranged in an anti-parallel manner.
24
1.3. Electron Paramagnetic Resonance
1.3.1. Introduction
Electron paramagnetic resonance (EPR) or electron spin resonance (ESR) was first
observed in 1944 by Russian physicist Zavoisky (96). It is a spectroscopic technique
used to study systems with unpaired electrons, such as free radicals and transition metals.
The physical principles of EPR are analogous to those of the more widely used nuclear
magnetic resonance (NMR). In NMR, energy absorptions by nuclei are detected, while
in EPR, electronic energy absorptions are analyzed.
Electromagnetic radiation is absorbed if the change in energy ( ∆E) is equal to
Planck’s constant (h) multiplied by the radiation frequency ( ν). This is known as
Planck’s law. Energy absorption causes a transition from low to high energy state. In
EPR, the source of electromagnetic radiation is a laboratory magnet. When samples with
unpaired electrons are exposed to a magnetic field, energy is absorbed by the electron.
This is the Zeeman effect.
The electron is a spin ½ particle. Due to spin angular momentum, an electron in a
magnetic filed can take two orientations, corresponding to m
S
= ± ½ , each associated
with different energies. If the electron is parallel to the magnetic field, then m
s
= - ½ , the
minimum energy orientation. If the electron is anti-parallel to the magnetic field, then it
is at the maximum energy orientation, and m
s
= + ½ .
The energy difference is given by ∆E = h ν = g
e
µ
e
B
o
where g
e
is the g-factor, a
proportionality constant, B
o
the magnetic field of the laboratory magnet, and µ
e
is the
Bohr magneton. The Bohr magneton is given by the equation µ
e
= eh / (2mc) where e is
the charge, h is Planck’s constant, m is the electron mass, and c is the speed of light.
25
Combining the above equations, we obtain Larmor frequency ( ν = g
e
µ
e
B
o
/h ). From this
equation, we conclude that there is no energy difference in the absence of a magnetic
field, and that the energy difference is proportional to the magnetic field (Figure 1.8).
1.3.2. Hyperfine Interaction
The unpaired electron is sensitive to the local environment. In addition to the external
magnetic field, the nucleus also exerts a local magnetic field on the electron. This
interaction between the nucleus and the electron is termed the hyperfine interaction. The
nuclear magnetic field adds to or opposes the effect of the external magnetic field on the
electron, depending on the spin state of the nucleus.
The number of transitions observed is 2I+1. The nitroxide spin label contains an
unpaired electron, mainly located in the p-orbital of nitrogen. Because I=1, there are 3
nuclear spin states (-1, 0, +1), resulting in 3 spectral lines (Figure 1.8). For a nitroxide,
the spectrum of 3 lines is due to hyperfine interaction.
1.3.3. Site-Directed Spin Labeling
Electron Paramagnetic resonance is the study of systems with unpaired electrons. The
presence of an unpaired electron in biological systems is rare. To utilize this powerful
technique, the strategy of site-directed spin labeling (SDSL) is implemented. SDSL
requires the substitution of a native amino acid with a cysteine via site-directed
mutagenesis. In addition, removal of native cysteine residues and replacement with
serines, alanines or leucines is required. Site-directed mutagenesis involves the
replacement of the native triplet codon at the genomic level to a cysteine triplet codon
(TGT or TGC).
26

Figure 1.8. Electron paramagnetic resonance. (A) Energy level diagram for a free
electron as a function of applied magnetic field B. At resonance, energy is absorbed by
the electron, enabling it to jump to a higher energy level. (B) Hyperfine interaction
results in the splitting of the energy levels. In the case of a nitroxide, there are 3 spectral
lines. (A) is from page 4 and (B) from page 46 of Weil, Bolton & Wertz, 1994 (96).
27
The nitroxide reagent used is MTSL (1-oxyl-2,2,5,5-tetramethylpyrroline-3-methyl
methanethiosulfate) (Figure 1.9A). This is attached to the protein by modification of the
sulfhydryl group of the cysteine residue. At neutral pH, the sulfhydryl group found in
cysteines reacts with the methanethiosulfate group, creating a covalent bond.
Labeling of the protein with MTSL involves the reduction of the protein with
dithiothreitol (DTT) or other reducing agent, followed by its removal. MTSL is added to
the protein solution and allowed to incubate. After incubation, excess MTSL is removed
by size exclusion or by extensive dialysis.
Analysis of EPR spectral parameters, such as mobility, accessibility, and spin-spin
interaction, will allow us to generate three-dimensional models.
1.3.4. Mobility
The sidechain R1 is sensitive to its environment. Analysis of the resulting EPR
spectra provides important information with respect to the motion of the spin-labeled
sidechain R1 (97, 98) (Figure 1.9B). The R1 mobility is affected by backbone dynamics
and tertiary interactions (98). Crystal structures of spin labeled T4 lysozyme (99-101)
show that the S δ-C α interaction stabilizes the conformation that restricts the spin label
motion to X4 and X5 bonds of the liner. This reduced linker rotational freedom results in
a nitroxide highly sensitive to backbone motion.
The mobility parameter provides information on local structure and dynamics,
secondary structure, tertiary as well as quaternary packing. The spin label is moving fast
in an unconstrained environment, for example a loop site. Residues located in loop sites
have sharp lines, large amplitudes, and narrow central linewidths. As the motion of the
spin label decreases, the center linewidth becomes broader, the amplitude decreases, and
28

Figure 1.9. EPR coupled with site-directed spin labeling. (A) The reaction of the
spin label (1-Oxy-2,2,5,5-Tetramethyl-D-Pyrroline-3-Methyl)-methanethiosulfonate
(MTSL) with the sulfhydryl group of cysteine resulting in the generation of the side chain
R1. Figure is from Margittai & Langen, 2008 (104). (B) EPR spectral line shapes
correlate with the motion of the spin label. The central linewidth is the width from peak
to trough, and called ∆H
o
. (C) In accessibility experiments, nonpolar O
2
partitions in the
lipid, while polar NiEDDA partitions into the solvent.
29
the width of the outer lines increase. Less mobile sites, such as helix surface sites and
tertiary contact sites, have intermediate amplitudes and broad central lines. Buried or
rigid sites, which are immobile, have decreased amplitudes and very broad central lines
due to anisotropy. Mobility information by measurement of the central linewidth ( ∆H
o
)
can categorize side chain R1 as buried within the interior, at a tertiary contact site, on the
protein surface, or on a loop region (98, 102, 103) (Figure 1.9B).
1.3.5. Accessibility
To further characterize the structure and topology of the membrane-bound form, we
can measure the accessibilities ( Π) of the R1 sidechain to the paramagnetic colliders O
2

and NiEDDA. The accessibility of sidechain R1 to these colliders depends on its location
in the lipid bilayer. The nonpolar O
2
partitions in the hydrophobic environment of the
membrane while the polar NiEDDA partitions in the solvent (Figure 1.9C). Thus,
residues highly accessible to O
2
are membrane-exposed while residues highly accessible
to NiEDDA are solvent-exposed.
Previous studies using SDSL demonstrated that the contrast parameter Φ
( Φ=LN[ ΠO
2
/ ΠNiEDDA]) is proportional to depth of membrane insertion, and that this
depth can be calibrated using spin-labeled derivatives of phospholipids (105). Using a
similar calibration, we can estimate the average immersion depth of the lipid-facing R1
side chains ( Φ maxima).
Experimentally, three measurements are taken for the analysis of contrast parameter
Φ. Measurements are done in the presence of ambient air, nitrogen, and NiEDDA while
nitrogen is flowing. As air is made up of oxygen and nitrogen, the influence of nitrogen
30
can easily be removed from the measurements. Accessibilities to O
2
and NiEDDA are
calculated by removing the effect of nitrogen.
Collision frequency of the spin label to the paramagnetic collider is monitored by EPR
spectral signal amplitude changes as a function of microwave power. Because of the
Maxwell-Boltzmann distribution, there are more electrons at the lower energy level than
at the higher energy level. As microwave power is increased, more electrons absorb
energy and move to the higher energy level, until both energy levels are equally
populated. This is known as saturation. The presence of paramagnetic colliders will
increase the power at which saturation occurs as colliders will transfer energy away from
the spin system. Residues with high O
2
accessibility are membrane-exposed while
residues with high NiEDDA accessibility are solvent exposed.
Analysis of the accessibility parameter gives us secondary structure information (106).
There is a periodicity of 2 for ß-sheets and a periodicity between 3 to 4 for α-helices. By
plotting П O
2
and NiEDDA as a function of labeling position, a water soluble protein can
be distinguished from a transmembrane protein and from a surface helix. A soluble
protein has equal accessibilities to both O
2
and NiEDDA, as well as an in-phase
periodicity. Because of asymmetric solvation, transmembrane helices and surface-bound
helices have out-of-phase periodicities. However, surface-bound helices can be
distinguished from transmembrane helices because lipid-exposed residues in a surface
bound helix are located at approximately the same depth while the same residues in
transmembrane helices have variable depths, with the deepest insert at the center of the
bilayer.

31
1.3.6. Distances
1.3.6.1. Continuous Wave (CW)
Distances between spin labels can be determined by analysis of dipolar coupling
between electrons. Electrons, which have magnetic dipoles, interact with each other like
two bar magnets. It is this interaction that is measured by EPR.
By using continuous wave EPR (CW-EPR), distances up to 20 Å can be determined
(107, 108). When spin labels are close together, the resulting spectra have broad lines
and reduced signal amplitudes. In the slow motion regime (frozen sample), spectral
broadening is due to dipolar interactions and is proportional to 1/r
3
(109). The dipolar
interaction is directly related to distance determined by Fourier deconvolution of the
spectrum (108). However, freezing of samples may cause artifacts, and also loss of
dynamics information. To monitor conformational change, sample recording at
physiological temperature is ideal. In the fast motion regime (spectra recorded at
physiological temperature), the dipolar interaction averages out, the spectra are
broadened by relaxation effects, and the dipolar interaction is proportional to 1/r
6
(109).
Analysis of the Pake pattern can be used to determine distances (107). Spectra taken at
the fast motion regime were deconvoluted using rigid lattice methods, and resulted in
good agreement; dipolar interactions can be determined from samples taken at
physiological temperatures for slowly tumbling proteins (109). Samples with interspin
vector that tumble slowly, such as samples measured in the presence of membrane, can
be analyzed in this way.

32
1.3.6.2. Double electron-electron resonance (DEER)
Pulsed EPR methods can determine distances of 60-80 Å (110-114). Double electron-
electron resonance (DEER) technique is used in these experiments. In DEER
experiments, dipolar coupling frequency is determined. The spectrum that is obtained is
a plot of relative echo intensity versus time. This spectrum is Fourier transformed into
distance. A short oscillation means a short distance, while a long oscillation means a
long distance.
Imagine 2 magnets, a strong and a weak one, interacting with each other. The time
that the magnets align with each other can be directly converted to distance between
magnets. When magnets are close together, alignment time is short, hence the distance is
also short. However, when magnets are far from each other, alignment time is long, and
distance is also long.
Technically, two microwave frequencies are applied to the sample, the observer
frequency and the pump frequency. The observer frequency is applied to monitor a
specific EPR spectrum region, and the pump frequency is applied elsewhere in the
spectrum. A change due to the pump frequency is observed as a spin echo. In these
studies, 4-pulse DEER was used to determine distances between nitroxides as the pulse
sequence decreases dead free time thus enhancing signal to noise (115) (Figure 1.10).
Data were fit using Tikhonov regularization (116) using DEERAnalysis2006 and 2008
(117).

33

Figure 1.10. Four-pulse DEER pattern. Figure is from Pannier et al., 2000 (115).
34
1.4. Structural Refinement
Determination of three-dimensional structures of proteins is important in
understanding their functions. Membrane proteins, due to problems with protein
expression, and stability, may not be amenable to high resolution structural determination
by crystallography and NMR spectroscopy. Biophysical techniques, such as EPR, offer
an advantage in that samples can be analyzed in their physiologically relevant
environments. In combination with computational analysis, these techniques can be used
to determine structural models.
Constraints derived from EPR data were used for simulated annealing molecular
dynamics (SAMD) calculations. From accessibility experiments, secondary structure
information is derived. In addition, depth of insertion into the membrane from
accessibility experiments can be transformed into structural constraints. Distances
between spin labels acquired from both CW and DEER measurements provide short and
long range distances.
As the measured distances are between spin labels, the native amino acids are
converted into spin labels by PRONOX, a program developed by Dr. Ian Haworth. This
program generates sterically allowable conformers for the labels based on crystal
structures of spin labeled T4 lysozyme (99, 100).
SAMD calculations were implemented using AMBER8 (118). AMBER is the
acronym for ‘Assisted Model Building with Energy Refinement’. It is a set of molecular
mechanical force fields used in the simulation of biomolecules. The functional form is
(119):

35

The bond term represents energy between covalent bonds, and can be well approximated
by an ideal spring (harmonic force). The angle term accounts for the electron orbital
geometry in covalent bonds. The torsion term is the energy for twisting a bond. The last
term represents the van der Waals and electrostatic energies. For the electrostatic energy,
charges can be represented by a single point charge.
In SAMD, atoms and molecules are allowed to interact with each other over a period
of time defined by a potential function. This potential function, or force field, dictates the
conditions under which the atoms and molecules will interact. A square well potential is
employed to constrain the structure to satisfy the experimental data. For example, when
using distance constraints, distances are only allowed to differ +/- 3Å. There is no energy
penalty here. If the distances go beyond +/- 3Å, then the energy will increase rapidly,
and may stop the calculation prematurely.
During SAMD, the system is heated to very high temperature (1000K or more) to
unfold the protein. The system is maintained at high temperature to ensure complete
protein unfolding. The system is then cooled gradually to 0K. This slow cooling, under
36
influence of structural constraints, allows for the protein to arrive at its most stable
conformation. Several cycles of heating and cooling are performed, arriving at a family
of structures. Validation of results can be done by analysis of how well experimental
data are reproduced by the calculations.
37
CHAPTER 1
REFERENCES
1. McMahon HT & Gallop JL (2005) Membrane curvature and mechanisms of
dynamic cell membrane remodelling. Nature 438(7068):590-596.

2. Ford MG, et al. (2002) Curvature of clathrin-coated pits driven by epsin. Nature
419(6905):361-366.

3. Gallop JL, et al. (2006) Mechanism of endophilin N-BAR domain-mediated
membrane curvature. Embo J 25(12):2898-2910.

4. Peter BJ, et al. (2004) BAR domains as sensors of membrane curvature: the
amphiphysin BAR structure. Science 303(5657):495-499.

5. Zimmerberg J & Kozlov MM (2006) How proteins produce cellular membrane
curvature. Nat Rev Mol Cell Biol 7(1):9-19.

6. Zimmerberg J & McLaughlin S (2004) Membrane curvature: how BAR domains
bend bilayers. Curr Biol 14(6):R250-252.

7. Butler MH, et al. (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(6):1355-1367.

8. Lee E, et al. (2002) Amphiphysin 2 (Bin1) and T-tubule biogenesis in muscle.
Science 297(5584):1193-1196.

9. Razzaq A, et al. (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(22):2967-2979.

10. Bergmann C, et al. (2003) Oligophrenin 1 (OPHN1) gene mutation causes
syndromic X-linked mental retardation with epilepsy, rostral ventricular
enlargement and cerebellar hypoplasia. Brain 126(Pt 7):1537-1544.

11. Billuart P, et al. (2000) Determination of the gene structure of human
oligophrenin-1 and identification of three novel polymorphisms by screening of
DNA from 164 patients with non-specific X-linked mental retardation. Ann Genet
43(1):5-9.

38
12. David C, Solimena M, & 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(1):73-79.

13. De Camilli P, 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(6):2219-2223.

14. Sakamuro D, Elliott KJ, Wechsler-Reya R, & Prendergast GC (1996) BIN1 is a
novel MYC-interacting protein with features of a tumour suppressor. Nat Genet
14(1):69-77.

15. Wieland I, et al. (2007) Contiguous gene deletions involving EFNB1, OPHN1,
PJA1 and EDA in patients with craniofrontonasal syndrome. Clin Genet
72(6):506-516.

16. Nicot AS, et al. (2007) Mutations in amphiphysin 2 (BIN1) disrupt interaction
with dynamin 2 and cause autosomal recessive centronuclear myopathy. Nat
Genet 39(9):1134-1139.

17. Gennis RB (1989) Biomembranes: Molecular Structure and Function (Springer-
Verlag, New York).

18. Zimmerberg J (2000) Are the curves in all the right places? Traffic 1(4):366-368.

19. van den Brink-van der Laan E, Killian JA, & de Kruijff B (2004) Nonbilayer
lipids affect peripheral and integral membrane proteins via changes in the lateral
pressure profile. Biochim Biophys Acta 1666(1-2):275-288.

20. Farsad K & De Camilli P (2003) Mechanisms of membrane deformation. Curr
Opin Cell Biol 15(4):372-381.

21. Nossal R (2001) Energetics of clathrin basket assembly. Traffic 2(2):138-147.

22. Hinshaw JE & Schmid SL (1995) Dynamin self-assembles into rings suggesting a
mechanism for coated vesicle budding. Nature 374(6518):190-192.

23. Chen YJ, Zhang P, Egelman EH, & Hinshaw JE (2004) The stalk region of
dynamin drives the constriction of dynamin tubes. Nat Struct Mol Biol 11(6):574-
575.

24. Sweitzer SM & Hinshaw JE (1998) Dynamin undergoes a GTP-dependent
conformational change causing vesiculation. Cell 93(6):1021-1029.
39
25. Takei K, et al. (1998) Generation of coated intermediates of clathrin-mediated
endocytosis on protein-free liposomes. Cell 94(1):131-141.

26. Sheetz MP & Singer SJ (1974) Biological membranes as bilayer couples. A
molecular mechanism of drug-erythrocyte interactions. Proc Natl Acad Sci U S A
71(11):4457-4461.

27. Antonny B, Huber I, Paris S, Chabre M, & Cassel D (1997) Activation of ADP-
ribosylation factor 1 GTPase-activating protein by phosphatidylcholine-derived
diacylglycerols. J Biol Chem 272(49):30848-30851.

28. Bigay J, Casella JF, Drin G, Mesmin B, & Antonny B (2005) ArfGAP1 responds
to membrane curvature through the folding of a lipid packing sensor motif. Embo
J 24(13):2244-2253.

29. Bigay J, Gounon P, Robineau S, & Antonny B (2003) Lipid packing sensed by
ArfGAP1 couples COPI coat disassembly to membrane bilayer curvature. Nature
426(6966):563-566.

30. Lee MC, et al. (2005) Sar1p N-terminal helix initiates membrane curvature and
completes the fission of a COPII vesicle. Cell 122(4):605-617.

31. Campelo F, McMahon HT, & Kozlov MM (2008) The hydrophobic insertion
mechanism of membrane curvature generation by proteins. Biophys J 95(5):2325-
2339.

32. Mui BL, Dobereiner HG, Madden TD, & Cullis PR (1995) Influence of
transbilayer area asymmetry on the morphology of large unilamellar vesicles.
Biophys J 69(3):930-941.

33. D'Souza-Schorey C & Chavrier P (2006) ARF proteins: roles in membrane traffic
and beyond. Nat Rev Mol Cell Biol 7(5):347-358.

34. Robinson MS (2004) Adaptable adaptors for coated vesicles. Trends Cell Biol
14(4):167-174.

35. McMahon HT & Mills IG (2004) COP and clathrin-coated vesicle budding:
different pathways, common approaches. Curr Opin Cell Biol 16(4):379-391.

36. Krauss M, et al. (2008) Arf1-GTP-induced tubule formation suggests a function
of Arf family proteins in curvature acquisition at sites of vesicle budding. J Biol
Chem 283(41):27717-27723.

40
37. Tanigawa G, et al. (1993) Hydrolysis of bound GTP by ARF protein triggers
uncoating of Golgi-derived COP-coated vesicles. J Cell Biol 123(6 Pt 1):1365-
1371.

38. Drin G, et al. (2007) A general amphipathic alpha-helical motif for sensing
membrane curvature. Nat Struct Mol Biol 14(2):138-146.

39. Gerke V & Moss SE (2002) Annexins: from structure to function. Physiol Rev
82(2):331-371.

40. 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.

41. 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.

42. Langen R, Isas JM, Luecke H, Haigler HT, & Hubbell WL (1998) Membrane-
mediated assembly of annexins studied by site-directed spin labeling. J Biol Chem
273(35):22453-22457.

43. Isas JM, et al. (2000) Annexins V and XII insert into bilayers at mildly acidic pH
and form ion channels. Biochemistry 39(11):3015-3022.

44. Kim YE, Isas JM, Haigler HT, & 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(37):32398-32404.

45. Isas JM, et al. (2003) Global structural changes in annexin 12. The roles of
phospholipid, Ca2+, and pH. J Biol Chem 278(32):30227-30234.

46. Hegde BG, Isas JM, Zampighi G, Haigler HT, & Langen R (2006) A novel
calcium-independent peripheral membrane-bound form of annexin B12.
Biochemistry 45(3):934-942.

47. Fischer T, Lu L, Haigler HT, & Langen R (2007) Annexin B12 is a sensor of
membrane curvature and undergoes major curvature-dependent structural
changes. J Biol Chem 282(13):9996-10004.

48. Irizarry MC, et al. (1996) Characterization of the precursor protein of the non-A
beta component of senile plaques (NACP) in the human central nervous system. J
Neuropathol Exp Neurol 55(8):889-895.

41
49. Jakes R, Spillantini MG, & Goedert M (1994) Identification of two distinct
synucleins from human brain. FEBS Lett 345(1):27-32.

50. Weinreb PH, Zhen W, Poon AW, Conway KA, & Lansbury PT, Jr. (1996) NACP,
a protein implicated in Alzheimer's disease and learning, is natively unfolded.
Biochemistry 35(43):13709-13715.

51. Davidson WS, Jonas A, Clayton DF, & George JM (1998) Stabilization of alpha-
synuclein secondary structure upon binding to synthetic membranes. J Biol Chem
273(16):9443-9449.

52. Perrin RJ, Woods WS, Clayton DF, & George JM (2000) Interaction of human
alpha-Synuclein and Parkinson's disease variants with phospholipids. Structural
analysis using site-directed mutagenesis. J Biol Chem 275(44):34393-34398.

53. Cabin DE, et al. (2002) Synaptic vesicle depletion correlates with attenuated
synaptic responses to prolonged repetitive stimulation in mice lacking alpha-
synuclein. J Neurosci 22(20):8797-8807.

54. Murphy DD, Rueter SM, Trojanowski JQ, & Lee VM (2000) Synucleins are
developmentally expressed, and alpha-synuclein regulates the size of the
presynaptic vesicular pool in primary hippocampal neurons. J Neurosci
20(9):3214-3220.

55. Abeliovich A, et al. (2000) Mice lacking alpha-synuclein display functional
deficits in the nigrostriatal dopamine system. Neuron 25(1):239-252.

56. George JM, Jin H, Woods WS, & Clayton DF (1995) Characterization of a novel
protein regulated during the critical period for song learning in the zebra finch.
Neuron 15(2):361-372.

57. Lee FJ, Liu F, Pristupa ZB, & Niznik HB (2001) Direct binding and functional
coupling of alpha-synuclein to the dopamine transporters accelerate dopamine-
induced apoptosis. Faseb J 15(6):916-926.

58. Kamp F & Beyer K (2006) Binding of alpha-synuclein affects the lipid packing in
bilayers of small vesicles. J Biol Chem 281(14):9251-9259.

59. Nuscher B, et al. (2004) Alpha-synuclein has a high affinity for packing defects in
a bilayer membrane: a thermodynamics study. J Biol Chem 279(21):21966-21975.

60. Bisaglia M, et al. (2005) A topological model of the interaction between alpha-
synuclein and sodium dodecyl sulfate micelles. Biochemistry 44(1):329-339.
42
61. Bisaglia M, et al. (2006) Structure and topology of the non-amyloid-beta
component fragment of human alpha-synuclein bound to micelles: implications
for the aggregation process. Protein Sci 15(6):1408-1416.

62. Borbat P, Ramlall TF, Freed JH, & Eliezer D (2006) Inter-helix distances in
lysophospholipid micelle-bound alpha-synuclein from pulsed ESR measurements.
J Am Chem Soc 128(31):10004-10005.

63. Bussell R, Jr. & Eliezer D (2003) A structural and functional role for 11-mer
repeats in alpha-synuclein and other exchangeable lipid binding proteins. J Mol
Biol 329(4):763-778.

64. Bussell R, Jr., Ramlall TF, & Eliezer D (2005) Helix periodicity, topology, and
dynamics of membrane-associated alpha-synuclein. Protein Sci 14(4):862-872.

65. Chandra S, Chen X, Rizo J, Jahn R, & Sudhof TC (2003) A broken alpha -helix in
folded alpha -Synuclein. J Biol Chem 278(17):15313-15318.

66. Ulmer TS, Bax A, Cole NB, & Nussbaum RL (2005) Structure and dynamics of
micelle-bound human alpha-synuclein. J Biol Chem 280(10):9595-9603.

67. Bortolus M, et al. (2008) Broken helix in vesicle and micelle-bound alpha-
synuclein: insights from site-directed spin labeling-EPR experiments and MD
simulations. J Am Chem Soc 130(21):6690-6691.

68. Drescher M, et al. (2008) Antiparallel arrangement of the helices of vesicle-bound
alpha-synuclein. J Am Chem Soc 130(25):7796-7797.

69. Jao CC, Der-Sarkissian A, Chen J, & Langen R (2004) Structure of membrane-
bound alpha-synuclein studied by site-directed spin labeling. Proc Natl Acad Sci
U S A 101(22):8331-8336.

70. Oda MN, Forte TM, Ryan RO, & Voss JC (2003) The C-terminal domain of
apolipoprotein A-I contains a lipid-sensitive conformational trigger. Nat Struct
Biol 10(6):455-460.

71. Segrest JP, et al. (1999) A detailed molecular belt model for apolipoprotein A-I in
discoidal high density lipoprotein. J Biol Chem 274(45):31755-31758.

72. Georgieva ER, Ramlall TF, Borbat PP, Freed JH, & Eliezer D (2008) Membrane-
bound alpha-synuclein forms an extended helix: long-distance pulsed ESR
measurements using vesicles, bicelles, and rodlike micelles. J Am Chem Soc
130(39):12856-12857.

43
73. Trexler A & Rhoades E (2009) α-Synuclein binds large unilamellar vesicles as an
extended helix. Biochemistry.

74. Ferreon AC, Gambin Y, Lemke EA, & Deniz AA (2009) Interplay of alpha-
synuclein binding and conformational switching probed by single-molecule
fluorescence. Proc Natl Acad Sci U S A 106(14):5645-5650.

75. Chen H, et al. (1998) Epsin is an EH-domain-binding protein implicated in
clathrin-mediated endocytosis. Nature 394(6695):793-797.

76. De Camilli P, et al. (2002) The ENTH domain. FEBS Lett 513(1):11-18.

77. Hyman J, Chen H, Di Fiore PP, De Camilli P, & Brunger AT (2000) Epsin 1
undergoes nucleocytosolic shuttling and its eps15 interactor NH(2)-terminal
homology (ENTH) domain, structurally similar to Armadillo and HEAT repeats,
interacts with the transcription factor promyelocytic leukemia Zn(2)+ finger
protein (PLZF). J Cell Biol 149(3):537-546.

78. Itoh T, et al. (2001) Role of the ENTH domain in phosphatidylinositol-4,5-
bisphosphate binding and endocytosis. Science 291(5506):1047-1051.

79. Frost A, et al. (2008) Structural basis of membrane invagination by F-BAR
domains. Cell 132(5):807-817.

80. Kweon DH, et al. (2006) Membrane topology of helix 0 of the Epsin N-terminal
homology domain. Mol Cells 21(3):428-435.

81. Henne WM, et al. (2007) Structure and analysis of FCHo2 F-BAR domain: a
dimerizing and membrane recruitment module that effects membrane curvature.
Structure 15(7):839-852.

82. Saarikangas J, et al. (2009) Molecular mechanisms of membrane deformation by
I-BAR domain proteins. Curr Biol 19(2):95-107.

83. Weissenhorn W (2005) Crystal structure of the endophilin-A1 BAR domain. J
Mol Biol 351(3):653-661.

84. Tarricone C, et al. (2001) The structural basis of Arfaptin-mediated cross-talk
between Rac and Arf signalling pathways. Nature 411(6834):215-219.

85. Takei K, Slepnev VI, Haucke V, & De Camilli P (1999) Functional partnership
between amphiphysin and dynamin in clathrin-mediated endocytosis. Nat Cell
Biol 1(1):33-39.

44
86. Farsad K, et al. (2001) Generation of high curvature membranes mediated by
direct endophilin bilayer interactions. J Cell Biol 155(2):193-200.

87. Gad H, et al. (2000) Fission and uncoating of synaptic clathrin-coated vesicles are
perturbed by disruption of interactions with the SH3 domain of endophilin.
Neuron 27(2):301-312.

88. Ringstad N, et al. (1999) Endophilin/SH3p4 is required for the transition from
early to late stages in clathrin-mediated synaptic vesicle endocytosis. Neuron
24(1):143-154.

89. Verstreken P, et al. (2002) Endophilin mutations block clathrin-mediated
endocytosis but not neurotransmitter release. Cell 109(1):101-112.

90. Frost A, Unger VM, & De Camilli P (2009) The BAR domain superfamily:
membrane-molding macromolecules. Cell 137(2):191-196.

91. Ayton GS, Blood PD, & Voth GA (2007) Membrane remodeling from N-BAR
domain interactions: insights from multi-scale simulation. Biophys J 92(10):3595-
3602.

92. Blood PD & Voth GA (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(41):15068-15072.

93. Low C, et al. (2008) Structure and dynamics of helix-0 of the N-BAR domain in
lipid micelles and bilayers. Biophys J 95(9):4315-4323.

94. Blood PD, Swenson RD, & Voth GA (2008) Factors influencing local membrane
curvature induction by N-BAR domains as revealed by molecular dynamics
simulations. Biophys J 95(4):1866-1876.

95. Masuda M, et al. (2006) Endophilin BAR domain drives membrane curvature by
two newly identified structure-based mechanisms. Embo J 25(12):2889-2897.

96. Weil JA, Bolton JR, & Wertz JE (1994) Electron Paramagnetic Resonance:
Elementary Theory and Practical Applications (John Wiley & Sons, New York) p
568.

97. Altenbach C, Marti T, Khorana HG, & Hubbell WL (1990) Transmembrane
protein structure: spin labeling of bacteriorhodopsin mutants. Science
248(4959):1088-1092.

45
98. Mchaourab HS, Lietzow MA, Hideg K, & Hubbell WL (1996) Motion of spin-
labeled side chains in T4 lysozyme. Correlation with protein structure and
dynamics. Biochemistry 35(24):7692-7704.

99. Guo Z, Cascio D, Hideg K, & Hubbell WL (2008) Structural determinants of
nitroxide motion in spin-labeled proteins: solvent-exposed sites in helix B of T4
lysozyme. Protein Sci 17(2):228-239.

100. Langen R, Oh KJ, Cascio D, & Hubbell WL (2000) Crystal structures of spin
labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra
in terms of structure. Biochemistry 39(29):8396-8405.

101. Guo Z, Cascio D, Hideg K, Kalai T, & Hubbell WL (2007) Structural
determinants of nitroxide motion in spin-labeled proteins: tertiary contact and
solvent-inaccessible sites in helix G of T4 lysozyme. Protein Sci 16(6):1069-
1086.

102. Isas JM, Langen R, Haigler HT, & Hubbell WL (2002) Structure and dynamics of
a helical hairpin and loop region in annexin 12: a site-directed spin labeling study.
Biochemistry 41(5):1464-1473.

103. Margittai M, Fasshauer D, Pabst S, Jahn R, & Langen R (2001) Homo- and
heterooligomeric SNARE complexes studied by site-directed spin labeling. J Biol
Chem 276(16):13169-13177.

104. Margittai M & 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(3-4):265-297.

105. 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 U S
A 91(5):1667-1671.

106. Hubbell WL, Gross A, Langen R, & Lietzow MA (1998) Recent advances in site-
directed spin labeling of proteins. Curr Opin Struct Biol 8(5):649-656.

107. Altenbach C, Oh KJ, Trabanino RJ, Hideg K, & Hubbell WL (2001) Estimation
of inter-residue distances in spin labeled proteins at physiological temperatures:
experimental strategies and practical limitations. Biochemistry 40(51):15471-
15482.

108. Rabenstein MD & Shin YK (1995) Determination of the distance between two
spin labels attached to a macromolecule. Proc Natl Acad Sci U S A 92(18):8239-
8243.
46
109. Hubbell WL, Cafiso DS, & Altenbach C (2000) Identifying conformational
changes with site-directed spin labeling. Nat Struct Biol 7(9):735-739.

110. Altenbach C, Kusnetzow AK, Ernst OP, Hofmann KP, & Hubbell WL (2008)
High-resolution distance mapping in rhodopsin reveals the pattern of helix
movement due to activation. Proc Natl Acad Sci U S A 105(21):7439-7444.

111. Bhatnagar J, Freed JH, & Crane BR (2007) Rigid body refinement of protein
complexes with long-range distance restraints from pulsed dipolar ESR. Methods
Enzymol 423:117-133.

112. Borbat PP, Davis JH, Butcher SE, & Freed JH (2004) Measurement of large
distances in biomolecules using double-quantum filtered refocused electron spin-
echoes. J Am Chem Soc 126(25):7746-7747.

113. Jeschke G (2002) Distance measurements in the nanometer range by pulse EPR.
Chemphyschem 3(11):927-932.

114. Jeschke G & Polyhach Y (2007) Distance measurements on spin-labelled
biomacromolecules by pulsed electron paramagnetic resonance. Phys Chem Chem
Phys 9(16):1895-1910.

115. Pannier M, Veit S, Godt A, Jeschke G, & Spiess HW (2000) Dead-time free
measurement of dipole-dipole interactions between electron spins. J Magn Reson
142(2):331-340.

116. Chiang YW, Borbat PP, & Freed JH (2005) The determination of pair distance
distributions by pulsed ESR using Tikhonov regularization. J Magn Reson
172(2):279-295.

117. Jeschke G, et al. (2006) DeerAnalysis2006 - a Comprehensive Software Package
for Analyzing Pulsed ELDOR Data. Appl Magn Reson 30:473-498.

118. Case DA, et al. (2004) AMBER 8 (University of California, San Francisco).

119. Cornell WD, et al. (1995) A Second Generation Force Field for the Simulation of
Proteins, Nulceic Acids, and Organic Molecules. J Am Chem Soc 117(19):5179-
5197.

47
CHAPTER 2
Structure of membrane-bound α-synuclein from site-directed spin labeling and
computational refinement
Christine C. Jao
*†
, Balachandra G. Hegde
*†
, Jeannie Chen
†‡
, Ian S. Haworth
*§
, and Ralf
Langen
*†

*
Department of Biochemistry and Molecular Biology, University of Southern California
Keck School of Medicine, Los Angeles, CA 90033,
†
Zilkha Neurogenetic Institute and
Arnold and Mabel Beckman Macular Research Center, University of Southern California
Keck School of Medicine, Los Angeles, CA 90033,
‡
Department of Ophthalmology and
the Department of Cell and Neurobiology, University of Southern California Keck School
of Medicine, Los Angeles, CA 90033,
§
Department of Pharmacology and Pharmaceutical
Sciences, University of Southern California School of Pharmacy, Los Angeles, CA 90033

This work was originally published in PNAS, vol. 105, no. 50, pp. 19666-19671, Dec. 16,
2008, and formatted for use in this dissertation.

48
CHAPTER 2
ABSTRACT
α-Synuclein is known to play a causative role in Parkinson disease. While its
physiological functions are not fully understood, α-synuclein has been shown to interact
with synaptic vesicles and modulate neurotransmitter release. However, the structure of
its physiologically relevant membrane-bound state remains unknown. Here we developed
a site-directed spin labeling and EPR-based approach for determining the structure of α-
synuclein bound to a lipid bilayer. Continuous wave EPR was used to assign local
secondary structure and to determine the membrane immersion depth of lipid-exposed
residues, while pulsed EPR was used to map long-range distances. The structure of α-
synuclein was built and refined using simulated annealing molecular dynamics restrained
by the immersion depths and distances. We found that α-synuclein forms an extended,
curved α-helical structure that is over 90 amino acids in length. The monomeric helix has
a superhelical twist similar to that of right-handed coiled-coils which, like α-synuclein,
contain 11-amino-acid repeats, but which are soluble, oligomeric proteins
(RMSD=0.82Å). The α-synuclein helix extends parallel to the curved membrane in a
manner that allows conserved Lys and Glu residues to interact with the zwitterionic
headgroups, while uncharged residues penetrate into the acyl chain region. This structural
arrangement is significantly different from that of α-synuclein in the presence of the
commonly used membrane-mimetic detergent SDS, which induces the formation of two
antiparallel helices. Our structural analysis emphasizes the importance of studying
membrane protein structure in a bilayer environment.

49
2.1. Introduction
The interaction of α-synuclein with membranes is thought to be important in its
physiologic function in vivo, as well as in its misfolding and aggregation in the
pathogenesis of Parkinson disease (1-10). Although the function of α-synuclein in vivo is
not fully understood, it has been observed to localize to presynaptic nerve termini, where
it modulates presynaptic pool size and neurotransmitter release (11-16). These functions
are likely to be mediated by the interaction of α-synuclein with synaptic vesicles, and
in vitro studies have shown that α-synuclein interacts strongly with highly curved
vesicles that are similar in size to synaptic vesicles (17, 18). The structural
characterization of membrane-bound α-synuclein is significant, given the importance of
membrane interactions to the pathologic and physiologic roles of α-synuclein.
Previous studies have revealed that the interaction of monomeric α-synuclein with
negatively charged vesicles induces a predominantly α-helical structure located in the N-
terminal region of the protein (17, 19, 20). This region contains seven 11-amino-acid
repeat regions that share some sequence similarity with apolipoproteins (Fig. 2.1).
Sequence analysis using algorithms for apolipoproteins predicts the formation of five
separate helices (17). However, no high-resolution structure is available for α-synuclein
in its physiologically relevant membrane-bound form. High-resolution NMR (21)
indicates that α-synuclein bound to an SDS detergent micelle forms two antiparallel
helices (from Met-1 to Val-37 and Glu-46 to Thr-92) that wrap tightly around the
detergent micelle. Recently, it has been suggested (22, 23) that membrane-bound α-
synuclein may take up a similar structure, while other studies suggest an extended helical
structure (20).
50

Figure 2.1. α-Synuclein sequence and the spin label. (A) The 140 amino acid–
containing human α-synuclein has seven N-terminal 11 residue repeats whose starting
positions are highlighted above the vertical bars. The individual repeats are numbered in
roman numerals. Letters in bold denote sites where single substitution to R1 were made
and underlined letters denote those previously made (20). Numbers above the sequence
denote amino acid positions. In all, accessibility data for residues 25 through 90 are
reported. (B) The reaction of the spin label (1-Oxy-2,2,5,5-tetramethyl-D-pyrroline-3-
methyl)-methanethiosulfonate (MTSL) with the sulfhydryl group of cysteine resulting in
the generation of the side chain R1.
51
In structural biology, detergent micelles are commonly used as membrane-
mimetic environments since their small size facilitates high-resolution structural analysis
by NMR. However, it is often difficult to test whether the structure of proteins bound to
micelles is indeed the same as that of the respective membrane-bound form. We have
argued that the much smaller size of SDS micelles might be responsible for the break
between the α-synuclein helices, and that an extended helical structure might be formed
in the presence of membranes (20). Therefore, the main goals of the present study were to
develop an approach for refining membrane protein structure in the presence of lipid
bilayers and to apply this methodology to determine the three-dimensional structure of
membrane-bound α-synuclein.

2.2. Results and Discussion
A structural refinement process for membrane-bound proteins was developed
based upon site-directed spin labeling, EPR spectroscopy (24-26), and simulated
annealing molecular dynamics (SAMD). Continuous wave EPR of singly labeled α-
synuclein derivatives was used to generate local mobility, accessibility, and membrane
immersion depth, while pulsed EPR provided intramolecular distances. These data were
converted into restraints that were used to refine the structure of membrane-bound α-
synuclein.
2.2.1. Local secondary structure and membrane topology information from
continuous wave EPR spectroscopy.
In order to obtain secondary structure and topography information, we generated
26 singly labeled α-synuclein derivatives and investigated them by continuous wave EPR
52
spectroscopy. The experimental design for these studies was identical to that of our
previous work, which included a nitroxide scan from residue 59 to residue 90 (20). The
additional new sites were chosen to generate a contiguous nitroxide scan from residue 25
to 90. Collectively, this region contains most of the sequence that forms the two helices
and the intervening loop region in the presence of SDS (19, 21, 23, 27-29).
The continuous wave EPR spectra of all new sites were similar to those obtained
previously for residues 59 to 90 (20), and the mobility information from all spectra is
summarized using the inverse central line width. As illustrated in Fig. 2.2A, the
respective values for residues located within the repeat regions are consistent with the
formation of an ordered structure that is devoid of pronounced tertiary or quaternary
packing interactions. In contrast, the C-terminal residues exhibit elevated mobility and
are unstructured. In an effort to provide more detailed secondary structural information,
we determined the membrane topography for each of the labeled sites by measuring their
accessibilities to O
2
and NiEDDA (Π O
2
and Π NiEDDA, respectively). This
measurement is based upon the preferential partitioning of O
2
into the membrane and
NiEDDA into the aqueous environment. As a consequence, membrane-exposed sites
show enhanced accessibility to O
2
, while solvent-exposed sites are preferentially
accessible to NiEDDA. In agreement with the formation of an extended helical structure,
Π O
2
and Π NiEDDA exhibit continuous periodic oscillations wherein maxima (as well
as minima) are spaced three to four amino acids apart (Fig. 2.2B). An important feature
of the accessibilities is that they are precisely out-of-phase. Such behavior is typically
observed for asymmetrically solvated α-helices that are exposed to solvent on the
NiEDDA-accessible side and to the membrane on the O
2
-accessible side (20, 26).
53
Figure 2.2. Accessibility measurements for membrane-bound α-synuclein. (A) The
inverse of the central line width (in 1/Gauss) was obtained from the X-band EPR spectra
of singly labeled, membrane-bound α-synuclein derivatives, and is given as a function of
the labeling position. The respective values obtained for N-terminal sites indicate
mobilities that are typically observed for helix surface sites lacking significant tertiary or
quaternary interactions. In contrast, the C-terminal sites (gray-shaded area) exhibit
elevated mobility that is typically observed for disordered and unfolded regions. The
horizontal lines indicate the upper and lower limits of the values typically observed for
helix surface sites (30-32). Panels A and B contain data from sites examined in the
present study as well as those from our previous study (20) (see Fig. 2.1 for a complete
listing). (B) Secondary structure and topology of membrane-bound, R1-labeled α-
synuclein reveals an elongated helix. Nitroxide accessibility to O
2
( ПO
2
, black open
squares) and NiEDDA (ΠNiEDDA, blue filled circles) are plotted as a function of
residue numbers. ΠO
2
as well as ΠNiEDDA give rise to a periodic oscillation of 3 to 4
residues that is characteristic of an α-helical structure. Importantly, ΠO
2
and ΠNiEDDA
are out-of-phase indicating an asymmetrically-solvated helix, wherein one face is
exposed to the phospholipid and the other face is exposed to the solvent.

54
Figure 2.2. continued

Figure 2.2. Accessibility measurements for membrane-bound α-synuclein.
55
The accessibility data for both colliders can be conveniently summarized by the depth
parameter Φ (Φ= ln [Π O
2
/Π NiEDDA]), which increases linearly with increasing
immersion depth (see Methods for calibration). As shown in Fig. 2.3A, Φ also exhibits a
pronounced oscillation consistent with an extended α-helical structure. In our prior
analysis of residues 59 to 90, we noted the formation of a helical structure in which each
repeat (11 amino acids) takes up three turns, giving rise to a periodicity of 3.67 amino
acids/turn. This unique feature is retained throughout all other scanned regions (Fig.
2.3A), as further illustrated by the best fit to a cosine function, which results in a
periodicity of ~3.68 amino acids/turn (Fig. 2.3A, blue line). Moreover, when we use a
helical wheel representation that is based upon 11 amino acids taking up exactly 3 turns
(Fig. 2.3B), lipid-exposed (red) and solvent-exposed (green) sites fall cleanly onto
opposite sides of this helix. Thus, the data indicate the formation of a continuous helical
structure that extends parallel to the membrane and that has a slightly unusual helical
periodicity.
2.2.2. Intramolecular distances from four-pulse DEER experiments.
We measured inter-label distances in 17 doubly labeled α-synuclein derivatives to
obtain independent verification of an elongated helical structure and to collect additional
restraints for the structural refinement. Only sites directly facing the membrane were
spin-labeled to ensure that all sites had comparable orientations (chosen from the maxima
in Fig. 2.3A). This strategy minimizes the effect of spin-label orientation and the
measured distances are therefore expected to depend mainly on the respective backbone
distances.
56
Figure 2.3. Continuous wave EPR analysis of singly labeled α-synuclein derivatives
indicates the formation of an ordered and continuous helical structure.
(A) The ratios of the accessibilities to O
2
and NiEDDA for residues 25 through 90 are
summarized by the depth parameter Φ=ln(ΠO
2
/ΠNiEDDA), with increasing Φ values
indicating deeper membrane immersion depth. The blue line indicates the best fit to a
cosine function and the resulting periodicity is 3.68 amino acids per turn (r
2
=0.85), which
is close to the theoretically predicted periodicity of 3.67 amino acids/turn (11 amino acids
per 3 turns=3.67 amino acids/turn). (B) The repeat region residues are plotted onto a
helical wheel in which 11 amino acids make up three turns. Lipid-exposed sites (red) fall
onto one side, while solvent-exposed sites (green) lie on the opposite side. White circles
denote residues with Φ values that are neither maxima nor minima. Residues shown in
gray were not tested.
57
Figure 2.3. continued

Figure 2.3. Continuous wave EPR analysis of singly labeled α-synuclein derivatives
indicates the formation of an ordered and continuous helical structure.

58
Distances were obtained from a four-pulse DEER experiment, which monitors the
time evolution of a spin echo intensity (33). The frequency of the resulting signal is a
direct measurement of inter-residue distances, with longer distances giving rise to slower
oscillations. This method has been demonstrated to be effective for precise measurement
of distances of up to ~80 Å in model systems and up to ~ 60 Å in proteins (33-36). As
shown for the 56R1/85R1 derivative (Fig. 2.4A), the time domain signal exhibits
pronounced oscillations, indicating the presence of a specific and ordered structure with a
well-defined and relatively narrow distance distribution. The data were fit using
Tikhonov regularization (37) and the resulting distance distribution is given in Fig. 2.4B.
The experimentally determined distance (maximum at 42 Å) coincides remarkably well
with the backbone distance expected from a rise of 1.5 Å/amino acid in an idealized α-
helix (29 amino acids x 1.5 Å/amino acid = 43.5 Å). Similarly good agreement between
experimental distances and theoretical estimates for an idealized α-helix was obtained for
13 additional derivatives (Table 2.1 and Fig. 2.5). No distances could be determined for
the 11R1/70R1, 11R1/81R1, and 41R1/85R1 derivatives, all of which gave slowly
decaying signals with oscillations that were too long to fit unambiguously, indicating that
the underlying distances are too large (>60 Å as estimated in Fig. 2.5B) to be resolved
using the present method. This result is also consistent with an extended helical structure
in which these distances would exceed the upper limit of detection (Table 2.1).
Importantly, the above data are inconsistent with the anti-parallel helical structure
that forms in the presence of SDS micelles, inasmuch as shorter distances would be
expected for the 22R1/52R1, 26R1/56R1, 11R1/70R1 and 11R1/81R1 derivatives (Table
2.1). To test the structure of the micelle-bound α-synuclein, DEER data were obtained
59

Figure 2.4. Intramolecular distance analysis for membrane-bound 56R1/85R1 α-
synuclein derivative. (A) The baseline corrected time evolution data from a four-pulse
DEER experiment for the membrane-bound 56R1/85R1 α-synuclein derivative (black
line) were fit using Tikhonov regularization (37) (red line). The resulting distance
distribution is given in panel (B). All results, including those from 16 additional
membrane-bound doubly labeled derivatives, are summarized in Table 2.1 (for data, see
Fig. 2.5).

60
Table 2.1. Intramolecular distances for 17 membrane-bound doubly-labeled
derivatives from four-pulse DEER experiments

Derivative DEER (Å) Ideal helix (Å) SDS (NMR) (Å)
11R1/26R1 25 22.5 22.5
11R1/41R1 48 45 43.3
22R1/52R1 49 45 23.3
26R1/41R1 23 22.5 25.4
26R1/56R1 44 45 24.4
37R1/67R1 42 45 39.2
41R1/56R1 23 22.5 24.0
41R1/67R1 37 39 38.8
41R1/70R1 41 43.5 42.0
44R1/67R1 36 34.5 34.2
48R1/67R1 29 28.5 27.8
56R1/70R1 22 21 20.1
56R1/85R1 42 43.5 42.0
63R1/81R1 26 27 25.7
11R1/70R1 >60 88.5 22.7
11R1/81R1 >60 105 22.8
41R1/85R1 >60 66 62.6

The experimental distances are taken from the peaks of the Tikhonov regularization-
based fits (for data, see Fig. 2.5), but identical maximal distances (within 1 Å or less)
were obtained from Gaussian fits. The data are compared to the rise of an ideal helical
structure, which is taken to be 1.5 Å/residue. The final column shows the respective α-
carbon distances from the high-resolution NMR structure of SDS-bound α-synuclein. All
distances are given in Å.
61
Figure 2.5. Intramolecular distances from 4-pulse DEER experiments. (A) The left
panels show the dipolar evolution time for each of the indicated doubly labeled
derivatives bound to SUV containing POPS/POPC (molar ratio of 3:7). The black traces
are background-corrected experimental data while the red lines represent the results of
fits using Tikhonov regularization (37). The right panel show the resulting distance
distributions whose peaks are tabulated in Table 2.1. Gaussian fits gave rise to distance
maxima that were within 0.5 Å of those shown here. Two of the double mutants
containing a label at position 41 (26R1/41R1 and 41R1/56R1) show a subtle bimodal
distance distribution. The minor peak which contains about 20 to 30% of the signal in
both cases is shifted approximately by 5 Å compared to the main peak. The exact origin
of the bimodal distance distributions is not known, but it could arise from different, well-
resolved R1 rotameric states or minor heterogeneities in the backbone structure at
position 41. The effect is rather subtle, however, and no bimodal distance distribution
could be observed for two other double mutants containing a label at position 41
(11R1/41R1 and 41R1/70R1). Regardless, collectively all distances as well as the
continuous wave EPR data indicate an extended helical structure as the predominant
form. (B) The distances for some double mutants were too long to determine reliably.
The first three panels in (B) give the raw dipolar evolution time data for the 11R1/70R1,
11R1/81R1 and the 41R1/85R1, respectively. When compared to the corresponding raw
data for 41R1/70R1, which did result in a clear distance, it becomes apparent that the
signals for the former decay rather slowly and do not give a clear oscillation that can be
fitted. This is a typical feature for long range distances and, thus, only a lower limit for
the distances is given in Table 2.1. The estimate for the lower limit of ~60 Å in Table 2.1
62
Figure 2.5. continued
was estimated from the time scale of data acquisition with longer distances requiring
longer time scans. According to (36), the maximal distance that can clearly be resolved
from DEER data obtained for 3 to 3.5 µs is on the order of 60 Å. There is a hint of an
oscillation for the 41R1/85R1 derivative (corresponding to a distance of ~60 to 70 Å), but
a longer time scan would be required to determine the baseline unambiguously and obtain
reliable distance measurements. Such longer time scans, however, gave data with poor
signal to noise. (C) The left panels shows the dipolar evolution time for the indicated α-
synuclein derivatives in the presence of SDS micelles (see Methods). The black traces are
the background-corrected experimental data and the red lines indicate the fits using
Tikhonov regularization. The right panels give the resulting distance distributions.
63
Figure 2.5. continued

Figure 2.5. Intramolecular distances from 4-pulse DEER experiments.
64
for two of these derivatives (11R1/70R1 and 11R1/81R1) in their SDS micelle-bound
state. Indeed, in agreement with previous studies (21, 27), much shorter distances with
broad distance distributions were obtained (Fig. 2.5C). Thus, the membrane-bound
helical form of α-synuclein is different from the SDS-bound form in two respects: it
adopts a single helical structure and it is better defined.
Based on a small number of distance measurements, it has been suggested (23)
that the N- and C-terminal portions of the repeat region have comparable distances in the
membrane-bound and SDS-bound states. This conclusion was reached based upon use of
very highly charged small unilamellar vesicles (SUVs) containing 100% POPG.
Replicating these conditions, we obtained a DEER signal that corresponds to a broad
distance distribution for the 11R1/70R1 derivative centered around 26 Å (Fig. 2.6A-C).
When compared to the conditions of the present study, however, the signal exhibited a
very low modulation depth, suggesting that the short distance arises from a small fraction
of samples. Moreover, we noted that α-synuclein strongly disrupts the integrity of SUVs
containing 100% POPG, as judged by vesicle leakage experiments (Fig. 2.6D).
According to gel filtration, only a subset of α-synuclein binds to intact vesicles, while a
large fraction of α-synuclein induces formation of smaller, non-vesicular structures (Fig.
2.6E). Importantly, after purifying the vesicle-bound α-synuclein, the 11R1/70R1
derivative no longer gave any clearly detectable short distances (Fig. 2.6A), suggesting
that the shorter distances for this derivative did not arise from vesicle-bound protein.
2.2.3. Structural Refinement.
Collectively, our data indicate that α-synuclein forms an extended helical
structure when bound to phospholipid bilayer membranes, but that it can take up different
65
Figure 2.6. α-synuclein interaction with SUVs of different lipid compositions. (A)
The structural features of α-synuclein bound to SUV containing 100% POPG were
investigated by performing 4-pulse DEER experiments of the 11R1/70R1 derivative
before (blue trace) and after purification (black trace) of the vesicle peak shown in panel
(E). The dipolar evolution time data after purification (black line) are difficult to
distinguish from background (red dashed line), and cannot readily be fitted to obtain a
distance. Thus, these data indicate a long distance and are different from those obtained
for SDS micelles (Fig. 2.5C). Slight deviations from background (magenta dashed line)
can be observed for the time evolution data from the unpurified sample and the
corresponding basline-corrected time evolution data are shown as a black trace in panel
(B). The red line in (B) represents a fit using Tikhonov regularization and the resulting
distance distribution is given in (C). (D) Leakage of vesicle content was assayed by
dilution of the fluorophor-quencher pair into the extra-vesicular space and monitored
using the fluorescence at 520 nm (see Methods). No leakage was detected for SUV
containing 100% POPG (red trace) or 3POPS/7POPC (black trace) prior to the addition
of α-synuclein. Leakage is observed immediately after the addition of 2 µM α-synuclein
to SUV containing 100%POPG (indicated by arrow). In contrast, little or no membrane
disruption was detected for the 3POPS/7POPC-containing SUV indicating that those
SUVs remain intact upon α-synuclein binding (black). (E) To further investigate the
vesicle perturbation observed in panel (A), we performed gel filtration using Superdex
200. The top panel in (E) shows the elution profile for 100% POPG vesicles containing
0.1% rhodamine label which conveniently allowed for lipid detection by monitoring

66
Figure 2.6. continued
absorbance at 570 nm. A single peak corresponding to intact vesicles is obtained. Upon
addition of α-synuclein (lower panel), the peak corresponding to intact vesicles remains
(see arrow), but an additional lipid peak can be observed (solid line). This peak comes
later and is shifted to sizes that are smaller than those of the intact vesicles. α-Synuclein
was visualized using NBD labeling at position 131, and its dilution profile was monitored
at 478 nm (dashed line). While some of the protein co-eluted with the intact vesicles, the
majority of the protein eluted in the second peak. From the elution profiles and the
extinction coefficients for NBD and rhodamine, we were able to estimate a protein to
lipid ratio of ~1/50 with latter fractions having progressively more protein. These
fractions, which are α-helical by circular dichroism, contain significantly more α-
synuclein than is typically accommodated on SUVs consistent with the formation of
protein-lipid complexes that are no longer sealed vesicular structures. As predicted from
the lack of leakage in panel (A), this second peak is absent when 30% POPS vesicles
were used in the gel filtration assay (data not shown).

67
Figure 2.6. continued

Figure 2.6. α-synuclein interaction with SUVs of different lipid compositions.

68
structures when bound to non-vesicular lipids or detergents. To generate an atomistic
three-dimensional structure of α-synuclein bound to a vesicle, we developed a
computational approach for structural refinement based on our EPR data. An in-house
algorithm was used to generate the starting structure of spin-labeled α-synuclein as a
linear α-helix (see Methods and Fig. 2.7) and to convert the experimental data into three
types of structural restraints: (1) distance restraints between spin labels based on the
DEER data, (2) immersion depth restraints (Table 2.2) modeled as distance restraints
from each spin label to the center of the vesicle (Fig. 2.7), and (3) backbone dihedral and
hydrogen bonding restraints applied to regions found to be α-helical (Table 2.3). These
restraints were used in a SAMD-based refinement using AMBER8 (38). To
test the quality of the refinement, some experimental depth and distance data were
excluded from the actual refinement process and were used to determine the effectiveness
of the refinement.
The refinement process produced ten structures, nine of which reproduced the
omitted experimental data to within experimental error (Table 2.4 and Fig. 2.8). These
nine structures are overlaid in Fig. 2.9 and share strong similarities. To generate the final
structures, the labeled side chains (Fig. 2.9A) were replaced by the native side chains
(Fig. 2.9B). Some of the central features of the structure are as follows: α-synuclein
forms an extended, continuously curved helical structure (Figs. 2.9A-B) with a
superhelical twist (Fig. 2.9D). When viewed from the top, the Lys residues (colored blue
in Fig. 2.9C) are oriented approximately perpendicular to the helical axis, whereas the
Glu residues (green in Fig. 2.9C) are facing upwards. To illustrate the spatial relationship
between this helix and the membrane, we used an in-house algorithm to construct a
69
Figure 2.7. Starting structure for the synuclein (9-89) peptide with 26 labels. (A)
The 81-amino acid peptide fragment (9-89 of synuclein) with 26 spin labels (X) used in
the SAMD simulation. (B) The peptide was constructed as a linear alpha helix with its
center positioned 146Å from the center (pink dot) of an imaginary spherical lipid vesicle.
Some lipids of the outer leaflet of the vesicle are shown to give an improved sense of the
location of the peptide. The radius of the vesicle (the distance from the center of the
sphere to the P atom of the phosphate group of the lipid) was 150Å. (C) Rotation through
90° showing that the center of the alpha helix is 146Å from the lipid center. Amino acids
in the central region of the helix were initially positioned “below” the phosphate head
groups, whereas those at the N and C termini were “above” the head groups. (D) Close up
structure of the starting peptide, showing that the spin labels were generally oriented
towards the center of the imaginary vesicle. All lipids were removed prior to SAMD
calculations, but an “atom” representing the center of the sphere was used in the SAMD.

70
Figure 2.7. continued

Figure 2.7. Starting structure for the α-synuclein (9-89) peptide with 26 labels.
71
Table 2.2. Label depths used in SAMD calculations.

Label
Measured
Depth (Å)
1

Distance from
Center (Å)
2
Label
Measured
Depth (Å)
1

Distance from
Center (Å)
2

11 12.8 137.2 59 9.5 140.5
26 11.3 138.7 63 12.3 137.7
29 8.0 142.0 66 9.0 141.0
30 12.2 137.8 67 11.1 138.9
33 12.8 137.2 70 11.2 138.8
37 13.9 136.1 74 12.0 138.0
40 10.4 139.6 77 10.6 139.4
41 11.6 138.4 78 7.3 142.7
44 11.5 138.5 81 13.1 136.9
48 7.9 142.1 85 10.9 139.1
51 9.8 140.2 88 6.8 143.2
52 10.9 139.1 89 8.2 141.8
56 11.7 138.3

1
Depth of the label below the phosphate group of the lipid based on EPR accessibility
data and depth calibration using spin-labeled lipids (see methods).
2
Distance from the label (assumed to be from the N atom of the nitroxide group) to the
center of an imaginary lipid vesicle of diameter 300Å. This value is calculated as 150 Å
– N Å, where N is the measured depth and 150Å is the radius.
72
Table 2.3. Restraints used in SAMD calculations.

Geometry
Element
Structural
Element
Value
1
r1
2
r2
2
r3
2
r4
2

Inter-label
distance
Labels pairs in Table 2.1
3
Exptl. N (Å) N-3 N-2 N+2 N+3
Label-vesicle
center distance
Labels in Table 2.2 Exptl. N (Å) N-1 N-0.5 N+2.5 N+3
φ torsion
angle
All amino acids -57° -77° -72° -42° -37°
ψ torsion
angle
All amino acids -47° -67° -62° -32° -27°
ω torsion
angle
All amino acids 180° 177° 178° 182° 183°
Backbone H-
bond distance
All amino acids 2.15Å 1.3Å 1.8Å 2.5Å 3.0Å
H α – S δ
distance
All labels 2.8Å 1.8Å 2.3Å 3.3Å 3.8Å

1
The value is either the experimental data or an “idealized” value for the particular
geometry element.
2
In AMBER8, restraints for simulated annealing are included using a modified square-
well potential defined between r2 and r3, between which the variable can move freely
without incurring an energy penalty. Outer limits of r1 and r4 are also defined, with force
constants (rk2 and rk3) defining the energy penalty in the regions r1 to r2 and r3 to r4.
The maximum values of rk2 and rk3 were set at 10.0 kcal/mol/Å or 10.0 kcal/mol/° for
all restraints.
3
Twelve label pairs for which distances were obtained, excluding 26R1/41R1 and
56R1/70R1.
73
Table 2.4. Geometry of structures obtained in SAMD calculations.

Cycle
Distance
1

26-41 (Å)
Distance
1
56-70 (Å)
Average
2

Phi (°)
Average
2
Psi (°)
SD
3

Phi (°)
SD
3

Psi (°)
1 24.0 20.8 -65.0 -41.5 3.7 7.0
2 20.5 20.5 -65.0 -41.5 3.6 7.2
3 23.2 22.0 -65.4 -41.7 4.2 7.9
5 21.2 16.1 -65.3 -41.8 3.6 7.1
6 24.8 23.5 -64.9 -41.6 4.1 6.8
7 24.1 21.3 -64.3 -42.9 4.3 7.8
8 26.1 21.2 -64.7 -42.2 3.4 6.5
9 23.4 22.7 -64.2 -42.4 3.8 7.5
10 21.7 22.6 -65.4 -41.3 3.5 8.0

1
Distance between the N atoms of the nitroxide in the respective labels.
2
Backbone torsion angle averaged over 81 amino acids.
3
Standard deviation of the backbone torsion angle averaged over 81 amino acids.
74
0.5
1.0
1.5
2.0
2.5
3.0
3.5
012345678 9 10
cycle number
lysine (Nζ) depth parameter (Å)

Figure 2.8. Lysine Depth Parameter. Plot illustrating the behavior of seven lysine side
chains (K32, K34, K43, K45, K58, K60 and K80) during the 10 production cycles of
simulated annealing molecular dynamics. The depth for each lysine was calculated based
on the distance from the Nζ atom of each lysine to the center of the vesicle. The
difference in this depth from the experimental depth was evaluated for each lysine and
averaged over the seven residues to give the ‘lysine depth parameter’ for each structure in
the simulated annealing calculation (150 structures per cycle, collected at steps of 0.2 ps).
The plot shows this parameter at four points in each cycle reflecting averages of distances
from 0-4 ps (initial heating from ~0 to ~300K, ∆), 4-10 ps (rapid heating to >1000K, ○),
10-27 ps (slow cooling to ~300K, □) and 27-30 ps (cooling from ~ 300K to ~0K, ■).
75
Figure 2.9. Refinement of the membrane-bound structure of α-synuclein using
simulated annealing molecular dynamics (SAMD) with restraints from EPR data.
(A) Overlay of nine structures obtained from SAMD calculations using the 9-89 amino
acid fragment of α-synuclein, with spin labels added at 26 sites (Fig. 2.7). In the refined
structures, the labels (identifiable by the S-S bond depicted in yellow) are generally
positioned on the concave side of the curved protein structure (i.e., oriented into the
membrane). The N terminus is on the left of the figure. (B) The refined labeled structures
were converted to the respective α-synuclein structures by replacing each label with the
normal amino acid in the α-synuclein sequence, with retention of the C β position in the
label. (C) View of the overlaid structures from above the lipid surface. The 11 lysine
residues (colored in blue) are oriented approximately perpendicular to the helical axis,
permitting potential interactions with the lipid phosphates. In contrast, the eight glutamic
acid residues (colored in green) are oriented away from the membrane, on the top surface
of the α-helix. (D) Ribbon view of the nine overlaid structures (similar perspective to that
in (C)), indicating the general goodness of fit of the structures and, most importantly,
showing the subtle but reproducible superhelical twist in the α-helix that emerged in the
SAMD calculations. We speculate that the superhelicity may allow the helix to adopt a 3-
11 conformation that permits the helical axis to follow a curved surface over an extended
length.
76
Figure 2.9. continued

Figure 2.9. Refinement of the membrane-bound structure of α-synuclein using
simulated annealing molecular dynamics (SAMD) with restraints from EPR data.
77
vesicle of 300 Å in diameter (size of SUVs used in the present study (20) around the
vesicle center used in the SAMD simulations. The lipids were packed around the SAMD-
derived α-synuclein structures and the lipids displaced by α-synuclein were repacked
evenly. We emphasize that the protein position was directly obtained from the
experimental depth restraints of the SAMD calculations; thus, it is governed by the
experimental data. An example of the resulting protein-lipid interaction is shown in Fig.
2.10. The curved helical structure is aligned in such a way that it follows the curvature of
the vesicle surface (Fig. 2.10A), with the α-helix located just below the phosphate groups
of the lipids (Fig. 2.10B). This alignment permits the lysine residues to interact with the
phosphates, as shown for Lys-58 and Lys-60 in Fig. 2.10C. Negatively charged residues
are on the outside surface of the helix (Fig. 2.9C) and are located at the level of the
choline groups of the lipids (Fig. 2.10). In contrast to these charged residues, all of the
lipid-exposed residues are generally hydrophobic or Thr residues.
A particularly interesting feature of the α-synuclein helix is that it bends around
the membrane with a superhelical twist (Figs. 2.9D and 2.10C). It is likely that this twist
facilitates the formation of a continuously bent helical structure, with the slightly altered
periodicity of 3.67 amino acids/turn. In fact, continually curved helices with superhelical
twists have been observed in right-handed coiled-coils (39-41), which also contain 11-
amino-acid repeat regions, but which form soluble oligomers instead of interacting with
membranes. To investigate potential structural similarities, we overlaid the α-synuclein
structure with that of tetrabrachion, a naturally occurring, right-handed coiled-coil. This
overlay resulted in a remarkably good overlap with an RMSD of 0.82 Å (Fig. 2.11).
Thus, the 11-amino-acid repeat regions in both proteins encode for an α-helical fold that
78
Figure 2.10. Representations of the interaction of α-synuclein with a curved lipid
surface. (A) Space-filled model of α-synuclein (shown in green) binding to the surface of
a lipid vesicle 300 Å in diameter. Approximately 25% of the outer leaflet of the vesicle is
shown. The vesicle was fitted around one of the structures derived from the
experimentally restrained SAMD calculations. (B) A closer cross-sectional view of the α-
synuclein interaction with the lipid surface, with rotation through 90° from the image in
(A). The protein (green) follows the curved surface of the vesicle, with the helical axis
positioned just below the level of the phosphate groups of the lipids. This position of the
protein emerged from the SAMD calculations and is a reflection of the immersion depths
obtained from the continuous wave EPR data. (C) A more detailed image of the protein-
lipid interaction, viewed from the same angle as the image in (A). The N terminus of the
α-helix is in the foreground. Lysine residues 58 and 60 are shown in space-filled format
(K58 oriented to the right and K60 to the left of the helix). The image indicates the
proximity of the positively charged lysine side chains to the negatively charged
phosphate groups. (D) Cartoon representations of the structures of α-synuclein on
micelles and SUVs. The small and highly curved micelles cannot accommodate the
extended helical structure present on the membrane.

79
Figure 2.10. continued

Figure 2.10. Representations of the interaction of α-synuclein with a curved lipid
surface.
80

Figure 2.11. Comparison of the structure of membrane-bound α-synuclein to that
of a right-handed coiled-coil. Overlay of a single helix from tetrabrachion (red, PDB
1FE6) with that of α-synuclein (green) using Tm-align (42) results in a backbone RMSD
of 0.82 Å. The tetrabrachion helices contain 52 amino acids. Shown are residues 12-62
for α-synuclein (structure 3, which scored highest in the validation (Fig. 2.8) and residues
2-52 of tetrabrachion. Comparable RMSD values were obtained for all other structures.
Overlays with all structures resulted in Tm-scores larger than 0.5, which is an indication
of the same fold (42).
81
can either mediate interaction of a monomeric protein with curved membranes or mediate
protein-protein contacts. In one case the hydrophobic surface is exposed to the
membrane while in the other case it forms the core of a helical bundle. Considering the
sequence similarities between α-synuclein and apolipoproteins, it is likely that the latter
may use a similar fold to wrap around lipid particles.

2.3. Conclusions
Compared to the plethora of structural information available for soluble proteins,
relatively little is known about the structures of transmembrane- or membrane-associated
proteins in the physiologically important lipid bilayer environment. Here we have
presented an approach for refining the structure of membrane-bound α-synuclein based
solely upon experimental restraints from continuous wave and pulsed EPR. The resulting
structure of membrane-bound α-synuclein represents an extended α-helical structure and
provides detailed molecular insight into the mechanism by which α-synuclein interacts
with membranes. Importantly, this structure is different from that obtained in the
presence of detergent micelles whose small diameter may inhibit formation of an
extended helix (Fig. 2.10D). Our data underscore the importance of obtaining direct
structural information on membrane proteins in a lipid bilayer environment, and show
that it is important to consider the lipid composition of a given bilayer since this may
have pronounced effects on protein and bilayer structure. The approach presented here
should not only be applicable to testing the various modes of α-synuclein interactions
with lipids, but may also enable structural investigation of other membrane proteins.

82
2.4. Methods
2.4.1. Preparation of spin-labeled α-synuclein derivatives
Single and double cysteine mutants of α-synuclein were expressed and purified as
described previously (20). Briefly, α-synuclein mutants were expressed in
BL21(DE3)pLysS cells, and the cell pellet was resuspended in lysis buffer [100 mM Tris
(pH 8), 300 mM NaCl, 1 mM EDTA (pH 8)]. The cell lysate was boiled and then acid
precipitated. After precipitation, the supernatant was dialyzed against dialysis buffer
[20 mM Tris (pH 8), 1 mM EDTA (pH 8), 1 mM DTT]. Two rounds of anion exchange
chromatography were performed and proteins were eluted with a salt gradient of 0-1 M
NaCl. Samples were spin-labeled in 20 mM Hepes (pH 7.4), 100 mM NaCl buffer using
5x molar excess spin label, incubated for 1 hour at room temperature, and separated from
unreacted spin label by gel filtration using PD10 columns (GE Healthcare).
2.4.2. Vesicle preparation
The following synthetic lipids were used: 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). All lipids
were purchased from Avanti Polar Lipids (Alabaster, AL). Lipids were dried with
nitrogen and desiccated overnight. After desiccation, lipids were resuspended in buffer,
treated to bath sonication, then sonicated with a tip at 2 W power (Misonix Inc.,
Farmingdale, NY) for 30 minutes. Lipids were centrifuged for 1 hour at 55,000 rpm at
22
°
C and the supernatant was recovered and used for experiments.

83
2.4.3. Continuous wave EPR
Continuous wave EPR spectra were obtained from vesicle-bound α-synuclein
derivatives at a molar protein-to-lipid ratio of 1:250 in 20 mM Hepes (pH 7.4), 100 mM
NaCl buffer. Spectral scans were collected using a Bruker EMX X-band CW EPR
spectrometer and inverse central line width values were measured from the peak-to-peak
distance of the central line as described previously (20). Accessibilities to O
2
and
NiEDDA (ΠO
2
and ΠNiEDDA) were obtained from power saturation experiments using
a dielectric resonator (43). The oxygen accessibility was carried out in the presence of
ambient oxygen and the sample was equilibrated with 3 mM NiEDDA for NiEDDA
accessibility. The immersion depth of lipid-exposed sites was determined from the
relation d[Å]= a*Φ+b (39), where the values of a and b were obtained using calibration
with spin-labeled lipids (1-palmitoyl 2-stearoyl-(n-DOXYL)-sn-glycero-3-
phosphocholine). The values of a and b are 5.9 and -4.1, respectively (20).
2.4.4. Pulsed EPR and distance analysis
Samples were prepared at a protein-to-lipid ratio of 1:250 as described above. For all
experiments, 25% fully spin-labeled protein, containing two spin labels per protein, was
mixed with unlabeled wild-type protein prior to the addition of vesicles. Unbound
protein was washed using YM-100 concentrators (Amicon). DEER experiments were
performed using a Bruker Elexsys E580 X-band pulse EPR spectrometer fitted with a
3 mm split ring (MS-3) resonator, a continuous flow helium cryostat (CF935, Oxford
Instruments), and a temperature controller (ITC503S, Oxford Instruments). Samples
(20 µl) were flash-frozen in the presence of 30% sucrose and data were acquired at 78 K.
84
The ELDOR pump frequency was set to the maximum of the central absorption peak
of the nitroxide EPR spectrum, while the observer frequency was set to the maximum of
the low-field absorption peak. The observer pulse lengths used were 16 and 32 ns, while
the optimized pump pulse length was ~32 ns, and a two-step phase cycle was applied to
eliminate the unwanted echoes. Data were acquired for 5 to 12 hours with a repetition
rate of 500 Hz. DEERAnalysis2006 and DEERAnalysis2008 packages (37) were used to
obtain distance information from the dipolar time evolution data. The background
contribution from non-specific interaction was subtracted using a two-dimensional model
for vesicle-bound α-synuclein and a three-dimensional model for SDS-bound α-
synuclein. For α-synuclein bound to non-vesicular POPG the background dimensionality
was fit using DEERAnalysis2008. Tikhonov regularization was used with regularization
parameters of 100 or less obtained from the L-curve analysis to fit distances (37, 44). The
distances given in Table 2.1 correspond to the maximum in the Tikhonov distance
distributions. Importantly, fits using Gaussian distance distributions (using DEFit kindly
provided by Dr. Peter Fajer, Florida State University as well as custom programs kindly
provided by Dr. Christian Altenbach, UCLA) gave distance maxima that were within 0.5
Å of those obtained from Tikhonov regularization.
2.4.5. Computational Structural Refinement
2.4.5.1. Structure Building
A peptide of the sequence shown in Fig. 2.7 (where X = label) was built using an in-
house algorithm. The peptide was constructed as a linear alpha helix ( φ = -57.0°, ψ = -
47.0°, ω = 180.0°, χ1 (C γ-C β-C α-N) = 168.0°) with the center of the helical axis (at V49)
positioned 146Å from the center of an imaginary spherical lipid vesicle whose outer
85
diameter is ~ 300 Å (Fig. 2.7). Twenty-six labels (Fig. 2.7) were added with torsion
angles t1 (S γ-C β-C α-N) = -60° and t2 (S δ-S γ-C β-C α) = -60°, since these angles are
favorable for the label bound to an alpha helix (45).
2.4.5.2. Simulated Annealing Molecular Dynamics (SAMD) Calculations
Experimental data for 12 inter-label distances (11R1/26R1, 11R1/41R1, 22R1/52R1,
26R1/56R1, 37R1/67R1, 41R1/56R1, 41R1/67R1, 41R1/70R1, 44R1/67R1, 48R1/67R1,
56R1/86R1, 63R1/81R1) and 25 label “depths” (Table 2.2) were used to define restraints
for the SAMD calculations. These data were converted into an appropriate format for use
as restraints in AMBER8. The allowable ranges used for the inter-label distances and the
depths (Table 2.3) were established through a series of preliminary SAMD calculations in
which these ranges were varied. Similarly, restraints were defined for geometrical
elements of the alpha helix (Table 2.3). We also included a restraint for the H α – S δ
distance in each label, since it has been established that an interaction persists between
these atoms and reduces the mobility of the label (32, 45-47). No other restraints were
applied to the label. Parameters for the label backbone and side chain up to Sδ were
identical to those used in AMBER8 for a S-S cysteine-cysteine bridge. The parameters
for the remainder of the label were based on those reported for a similar label bound to
DNA (48).
The SAMD calculations were carried out in AMBER8. Lysine and glutamic acid
residues were included in uncharged forms: residues KNC and ENC in the AMBER
parm98 force field. These residues contain NH
3
and COO groups on the respective side
chains, but the overall charge on each residue is neutral. Preliminary calculations
indicated that inclusion of charged residues in gas phase SAMD calculations resulted in
86
formation of salt bridges between adjacent lysine and glutamic acid side chains. This is
not likely to occur in an aqueous or interfacial environment, and therefore we chose to
use the uncharged residues in the final calculation.
After a brief minimization of the starting structure (Fig. 2.7) SAMD was
performed in cycles of 30 ps, including a heating phase from 0K to 1200K in 4 ps using a
step of 0.002 ps, during which the force constants for the restraints were increased from
0.1 to 10.0; maintenance of the temperature at 1200K for a further 6 ps; and then cooling
to 0K over 20 ps, with stepwise adjustments of the TAUTP parameter. This approach was
based on the standard recommended protocol for simulated annealing calculations in the
AMBER8 manual (49) with some minor modifications. Eleven cycles were performed,
including an initial equilibration cycle, and 10 subsequent production cycles (numbered 1
to 10) from which structures were collected at 0K. The molecular dynamics simulation
was performed with a time step of 0.002 ps, a distance dependent dielectric of 4, and a
cut-off of 10.0 Å. The “atom” representing the center of the imaginary lipid vesicle was
constrained to the origin with a force constant of 1000.0 kcal/mol/Å. Full details of the
parameters and input files used to perform the SAMD calculations will be provided upon
request.
Since the calculations are the first report of using EPR data to refine a structure
using simulated annealing, some clarification of the method is required. First, the
molecule used in the calculation was the peptide with labels substituted at all positions
for which inter-label distance or depth data were used. Inter-label distances were defined
between the N atom of the nitroxide groups of each label pair. The only other element in
the calculation was an “atom” representing the center of an imaginary vesicle. This
87
“atom” was used to define the “depth” of the peptide in the lipid bilayer via distance
restraints from the N of the nitroxide to the center atom. As seen in the results, this
caused the peptide to bend in a manner consistent with its insertion into the lipid outer
leaflet. It should be noted that the present computational treatment of the experimentally
obtained immersion depths is based on the assumption that α-synuclein does not cause
significant local alterations in membrane shape. Although some proteins can affect the
curvature of larger vesicles (50, 51), α-synuclein does not appear to have this ability;
rather α-synuclein binds preferentially to curved vesicles (17) without disrupting the
integrity of the POPS/POPC containing vesicles used here (Fig. 2.6D) More importantly,
the present studies were performed using SUV. Such vesicles have an extreme curvature
strain and even curvature-inducing proteins are not known to generate smaller, more
highly curved vesicles. We therefore reason that substantial deformation of SUV would
further enhance the curvature strain and create a significant energetic barrier. The present
approach of converting depth measurements into distance restraints is likely to be
applicable to other proteins, but extra caution may need to be exercised when applying
this approach to proteins bound to larger vesicles. A central feature of the experimental
design was that the distances were chosen to be interlocking; that is, each position was
constrained by multiple inter-residue distances, as well as the distance to the vesicle
center. This strategy was chosen to counteract the intrinsic flexibility of the nitroxide side
chain.
2.4.5.3. Validation of Results
A family of 10 synuclein structures was obtained from the SAMD calculations.
To validate the results, we examined how well these structures reproduced experimental
88
data that were not included in the constraints. In particular, the depths of K32, K34, K43,
K45, K58, K60 and K80 in each structure were compared with the experimentally
determined depths of these residues by calculation of a ‘lysine depth parameter’. This
parameter was calculated based on the distance of the N ζ atoms to the center of the lipid
vesicle, as described in the legend to Fig. 2.8. The parameter is shown for four points in
each SAMD cycle (Fig. 2.8) reflecting the initial heating, rapid high temperature heating,
slow cooling and final cooling stages, and generally the lysine depth parameters reflect
the expected behavior in the simulation. For example, for cycle 1 (the first four data
points in Fig. 2.8) the lysine depth parameter increases from the first ( ∆) to the second ( ○)
point, reflecting heating of the system and greater disorder causing a larger difference
between the experimental and theoretical data, and then drops at the third ( □) point,
reflecting cooling of the system, and reaches a minimum at the end of the cycle ( ■).
The other cycles in Fig. 2.8 show largely similar behavior to that described for
cycle 1, in that the second point in each cycle ( ○) is generally the maximum and the
fourth point is the minimum ( ■). There is some fluctuation of the depth parameter, in part
due to the assumption the N ζ atom of the lysine reflects the position of the nitroxide of a
spin label at the same position. The two biggest deviations from the experimental depth
measurements were obtained for cycles 4 and 7. Inspection of the structure for cycle 7
indicated that this structure has the same fold as all other structures but that the specific
lysine side chain (but not backbone) orientations were responsible for the deviation. The
only other structure that did not give the expected result was that from cycle 4. This
structure had the largest value for the depth parameter (Fig. 2.8) and was the only
structure for which this value was > 3 Å. Visual inspection of the structure from cycle 4
89
indicated that it was substantially different from all other structures. Importantly, the
structure was tilted in a way that caused the lysine positions in the N-terminal region
point toward the acyl chain interior of the bilayer, an arrangement which is inconsistent
with the depth data and the helical wheel in Fig. 2.3. Therefore, we chose to eliminate
this structure from the final family of derived structures. It should be pointed out,
however, that the SAMD simulations “recovered” from this structure and reverted back
to giving structures that were consistent with experimental data in cycles 5 to 10.
We also determined the agreement between the theoretical and experimental data
for two inter-label distances (26R1/41R1 and 56R1/70R1) that were not included as
constraints in the SAMD calculation. The values for these distances are shown in Table
2.4 for the 9 structures (cycles 1-3 and 5-10). All distances are within the experimentally
observed distance ranges (Table 2.1) and most are within 1 or 2 Å of the respective
distance maxima.
The phi and psi values averaged over all amino acids in each structure are also
shown in Table 2.4 and illustrate that the majority of the angles did not approach the
edges of the respective restraint ranges for phi and psi (Table 2.3). Phi changed from -57°
in the linear helical starting structure (Fig. 2.7) to an overall average (across all amino
acids and all structures) of about -65 ± 4°, whereas psi moved from -47° (Table 2.4) to an
overall average of about -42 ± 7°. Most of the individual phi and psi angles were well
within the constraint range.
2.4.6. Interaction with SDS micelles
To study the interaction with SDS micelles, we adopted conditions previously used in
NMR studies (21). Samples were prepared in 20 mM Hepes pH 7.4, 100 mM NaCl buffer
90
at a protein to detergent molar ratio of 1:150 (final concentration of 250 uM α-synuclein
and 37.5 mM SDS). For all experiments, 25% spin-labeled protein was mixed with wild-
type protein prior to addition of SDS (63 uM spin-labeled protein and 187 uM wild-type
protein) to minimize intermolecular spin-spin interactions. 30% sucrose was present as
cryoprotectant for flash freezing.
2.4.7. Dye Leakage Assays
SUVs composed of 30% POPS / 70% POPC and 100% POPG loaded with 9 mM
ANTS (8-aminonaphthalene-1,3,6-trisulfonic acid, disodium salt) and 25 mM DPX (p-
xylene-bis-pyridinium bromide) (Invitrogen) were used in the dye leakage assay. SUVs
were made by first drying down lipids, then resuspending them in the ANTS/DPX
solution. The resuspended lipid solution was treated to five cycles of freeze/thaw, then
bath-sonicated for 20 minutes. Unencapsulated dye was removed by gel filtration using
G-100 resin. Assays were performed using a 1:250 molar ratio of protein to lipid.
Fluorescence measurements were made using a JASCO fluorometer (FP-6500), setting
excitation at 380 nm. Emitted fluorescence was monitored at 520 nm. Experiments with
30% POPS were conducted using 20 mM Hepes (pH 7.4), 100 mM NaCl buffer, while
those using 100% POPG were performed using 10 mM Tris (pH 7.4) to match the
conditions of the prior study (23).
2.4.8. Gel Filtration
The thiol-reactive probe N,N'-dimethyl-N-(iodoacetyl)-N'-(7-nitrobenz-2-oxa-1,3-
diazol-4-yl)ethylenediamine (IANBD amide from Invitrogen) was used to label a C-
terminal single cysteine mutant. SUVs composed of 30% POPS/70% POPC or 100%
POPG were made with 0.1% rhodamine-labeled phosphatidylethanolamine. After
91
desiccation, lipids were resuspended in 20 mM Hepes (pH 7.4), 100 mM NaCl for assays
with 30% POPS/70% POPC, or 10 mM Tris (pH 7.4) for assays with 100% POPG.
Protein and lipid were incubated at 1:250 molar ratio for 30 minutes before separation
using a Superdex 200 (GE Healthcare) gel filtration column. Protein elution was
monitored at 470 nm, while lipid elution was detected at 570 nm.

92
CHAPTER 2
REFERENCES
1. Aisenbrey C, et al. (2008) How is protein aggregation in amyloidogenic diseases
modulated by biological membranes? Eur Biophys J 37(3):247-255.

2. Cole NB, et al. (2002) Lipid droplet binding and oligomerization properties of the
Parkinson's disease protein alpha-synuclein. J Biol Chem 277(8):6344-6352.

3. Fortin DL, et al. (2004) Lipid rafts mediate the synaptic localization of alpha-
synuclein. J Neurosci 24(30):6715-6723.

4. Lee HJ, Choi C, & Lee SJ (2002) Membrane-bound alpha-synuclein has a high
aggregation propensity and the ability to seed the aggregation of the cytosolic
form. J Biol Chem 277(1):671-678.

5. Lotharius J & Brundin P (2002) Pathogenesis of Parkinson's disease: dopamine,
vesicles and alpha-synuclein. Nat Rev Neurosci 3(12):932-942.

6. Smith DP, et al. (2008) Formation of a high affinity lipid-binding intermediate
during the early aggregation phase of alpha-synuclein. Biochemistry 47(5):1425-
1434.

7. Soper JH, et al. (2008) α-Synuclein-induced Aggregation of Cytoplasmic Vesicles
in Saccharomyces cerevisiae. Mol Biol Cell 19(3):1093-1103.

8. Tofaris GK & Spillantini MG (2007) Physiological and pathological properties of
alpha-synuclein. Cell Mol Life Sci 64(17):2194-2201.

9. Vekrellis K, Rideout HJ, & Stefanis L (2004) Neurobiology of alpha-synuclein.
Mol Neurobiol 30(1):1-21.

10. Zhu M, Li J, & Fink AL (2003) The association of alpha-synuclein with
membranes affects bilayer structure, stability, and fibril formation. J Biol Chem
278(41):40186-40197.

11. Abeliovich A, et al. (2000) Mice lacking alpha-synuclein display functional
deficits in the nigrostriatal dopamine system. Neuron 25(1):239-252.

12. Cabin DE, et al. (2002) Synaptic vesicle depletion correlates with attenuated
synaptic responses to prolonged repetitive stimulation in mice lacking alpha-
synuclein. J Neurosci 22(20):8797-8807.

93
13. Gitler AD, et al. (2008) The Parkinson's disease protein alpha-synuclein disrupts
cellular Rab homeostasis. Proc Natl Acad Sci U S A 105(1):145-150.

14. Iwai A, et al. (1995) The precursor protein of non-A beta component of
Alzheimer's disease amyloid is a presynaptic protein of the central nervous
system. Neuron 14(2):467-475.

15. Masliah E, et al. (2005) Effects of alpha-synuclein immunization in a mouse
model of Parkinson's disease. Neuron 46(6):857-868.

16. Murphy DD, Rueter SM, Trojanowski JQ, & Lee VM (2000) Synucleins are
developmentally expressed, and alpha-synuclein regulates the size of the
presynaptic vesicular pool in primary hippocampal neurons. J Neurosci
20(9):3214-3220.

17. Davidson WS, Jonas A, Clayton DF, & George JM (1998) Stabilization of alpha-
synuclein secondary structure upon binding to synthetic membranes. J Biol Chem
273(16):9443-9449.

18. Kamp F & Beyer K (2006) Binding of alpha-synuclein affects the lipid packing in
bilayers of small vesicles. J Biol Chem 281(14):9251-9259.

19. 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.

20. Jao CC, Der-Sarkissian A, Chen J, & Langen R (2004) Structure of membrane-
bound alpha-synuclein studied by site-directed spin labeling. Proc Natl Acad Sci
U S A 101(22):8331-8336.

21. Ulmer TS, Bax A, Cole NB, & Nussbaum RL (2005) Structure and dynamics of
micelle-bound human alpha-synuclein. J Biol Chem 280(10):9595-9603.

22. Bortolus M, et al. (2008) Broken helix in vesicle and micelle-bound alpha-
synuclein: insights from site-directed spin labeling-EPR experiments and MD
simulations. J Am Chem Soc 130(21):6690-6691.

23. Drescher M, et al. (2008) Antiparallel arrangement of the helices of vesicle-bound
alpha-synuclein. J Am Chem Soc 130(25):7796-7797.

24. Fanucci GE & Cafiso DS (2006) Recent advances and applications of site-
directed spin labeling. Curr Opin Struct Biol 16(5):644-653.

25. Hubbell WL, Cafiso DS, & Altenbach C (2000) Identifying conformational
changes with site-directed spin labeling. Nat Struct Biol 7(9):735-739.
94
26. Hubbell WL, Gross A, Langen R, & Lietzow MA (1998) Recent advances in site-
directed spin labeling of proteins. Curr Opin Struct Biol 8(5):649-656.

27. Borbat P, Ramlall TF, Freed JH, & Eliezer D (2006) Inter-helix distances in
lysophospholipid micelle-bound alpha-synuclein from pulsed ESR measurements.
J Am Chem Soc 128(31):10004-10005.

28. Bussell R, Jr. & Eliezer D (2003) A structural and functional role for 11-mer
repeats in alpha-synuclein and other exchangeable lipid binding proteins. J Mol
Biol 329(4):763-778.

29. Chandra S, Chen X, Rizo J, Jahn R, & Sudhof TC (2003) A broken alpha -helix in
folded alpha -Synuclein. J Biol Chem 278(17):15313-15318.

30. Isas JM, Langen R, Haigler HT, & Hubbell WL (2002) Structure and dynamics of
a helical hairpin and loop region in annexin 12: a site-directed spin labeling study.
Biochemistry 41(5):1464-1473.

31. Margittai M, Fasshauer D, Pabst S, Jahn R, & Langen R (2001) Homo- and
heterooligomeric SNARE complexes studied by site-directed spin labeling. J Biol
Chem 276(16):13169-13177.

32. Mchaourab HS, Lietzow MA, Hideg K, & Hubbell WL (1996) Motion of spin-
labeled side chains in T4 lysozyme. Correlation with protein structure and
dynamics. Biochemistry 35(24):7692-7704.

33. Pannier M, Veit S, Godt A, Jeschke G, & Spiess HW (2000) Dead-time free
measurement of dipole-dipole interactions between electron spins. J Magn Reson
142(2):331-340.

34. Altenbach C, Kusnetzow AK, Ernst OP, Hofmann KP, & Hubbell WL (2008)
High-resolution distance mapping in rhodopsin reveals the pattern of helix
movement due to activation. Proc Natl Acad Sci U S A 105(21):7439-7444.

35. Bhatnagar J, Freed JH, & Crane BR (2007) Rigid body refinement of protein
complexes with long-range distance restraints from pulsed dipolar ESR. Methods
Enzymol 423:117-133.

36. Jeschke G & Polyhach Y (2007) Distance measurements on spin-labelled
biomacromolecules by pulsed electron paramagnetic resonance. Phys Chem Chem
Phys 9(16):1895-1910.

37. Jeschke G, et al. (2006) DeerAnalysis2006 - a Comprehensive Software Package
for Analyzing Pulsed ELDOR Data. Appl Magn Reson 30:473-498.
95
38. Pearlman DA, et al. (1995) AMBER, a package of computer programs for
applying molecular mechanics, normal mode analysis, molecular dynamics and
free energy calculations to simulate the structural and energetic properties of
molecules Comp Phys Comm 91(1-3):1-41.

39. Harbury PB, Plecs JJ, Tidor B, Alber T, & Kim PS (1998) High-resolution protein
design with backbone freedom. Science 282(5393):1462-1467.

40. Plecs JJ, Harbury PB, Kim PS, & Alber T (2004) Structural test of the
parameterized-backbone method for protein design. J Mol Biol 342(1):289-297.

41. Stetefeld J, et al. (2000) Crystal structure of a naturally occurring parallel right-
handed coiled coil tetramer. Nat Struct Biol 7(9):772-776.

42. Zhang Y & Skolnick J (2005) TM-align: a protein structure alignment algorithm
based on the TM-score. Nucleic Acids Res 33(7):2302-2309.

43. 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 U S
A 91(5):1667-1671.

44. Chiang YW, Borbat PP, & Freed JH (2005) The determination of pair distance
distributions by pulsed ESR using Tikhonov regularization. J Magn Reson
172(2):279-295.

45. Guo Z, Cascio D, Hideg K, & Hubbell WL (2008) Structural determinants of
nitroxide motion in spin-labeled proteins: solvent-exposed sites in helix B of T4
lysozyme. Protein Sci 17(2):228-239.

46. Guo Z, Cascio D, Hideg K, Kalai T, & Hubbell WL (2007) Structural
determinants of nitroxide motion in spin-labeled proteins: tertiary contact and
solvent-inaccessible sites in helix G of T4 lysozyme. Protein Sci 16(6):1069-
1086.

47. Langen R, Oh KJ, Cascio D, & Hubbell WL (2000) Crystal structures of spin
labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra
in terms of structure. Biochemistry 39(29):8396-8405.

48. Price EA, Sutch BT, Cai Q, Qin PZ, & Haworth IS (2007) Computation of
nitroxide-nitroxide distances in spin-labeled DNA duplexes. Biopolymers
87(1):40-50.

49. Case DA, et al. (2004) AMBER 8 (University of California, San Francisco).
96
50. Gallop JL, et al. (2006) Mechanism of endophilin N-BAR domain-mediated
membrane curvature. Embo J 25(12):2898-2910.

51. McMahon HT & Gallop JL (2005) Membrane curvature and mechanisms of
dynamic cell membrane remodelling. Nature 438(7068):590-596.

97
CHAPTER 3
Membrane Binding and Self-association of Epsin ENTH Domains
Christine C. Jao
§1
, Chun-Liang Lai
§2
, Gary S. Ayton
2
, Brian Peter
3
, Jennifer L. Gallop
3
,
Harvey T. McMahon
3
, Gregory A. V oth*
2
, and Ralf Langen*
1

1
Zilkha Neurogenetic Institute, University of Southern California, 1501 San Pablo Street,
Los Angeles, CA 90033,
2
Center of Biophysical Modeling and Simulation, University of
Utah, 315 S 1400 E, Room 2020, Salt Lake City, UT 84112-0850,
3
MRC Laboratory of
Molecular Biology, University of Cambridge, Hills Road, Cambridge, CB2 2QH, UK.
§
Authors contributed equally
*Corresponding Authors: Email, GA V voth@hec.utah.edu, RL langen@usc.edu

This work is a collaboration with the McMahon and V oth labs. The manuscript (to be
submitted) was formatted for use in this dissertation. EPR work was done by the author.
98
CHAPTER 3
ABSTRACT
Electron Paramagnetic Resonance (EPR) spectroscopy and Molecular Dynamics
(MD) simulations are used to investigate the structure and the stability of Epsin N-
Terminal Homology (ENTH) domain aggregates on lipid bilayers. EPR experiments
examined systems composed of liposomes as well as ones with pre-formed tubules. The
experimental results suggest that ENTH forms predominantly monomers on vesiculated
structures, while ENTH predominantly self-associates into at least dimeric structures on
the pre-formed tubules. The MD simulations were designed to complement the EPR
experiments and hence examined both ENTH monomers and ENTH dimers, and it was
observed that the penetration depth of the N-terminal amphipathic helix as measured
from MD simulation agreed well with the EPR measurements. The combined results of
the EPR experiments and MD simulation suggest that amphipathic helix insertion is
strong whether isolated monomers or more organized dimers are examined. However,
membrane remodeling requires a further collective organization of ENTH domains at a
density sufficient such that the wedge bending mechanism of the amphipathic helices can
trigger either vesiculation or tubulation.
99
3.1. Introduction
Clathrin-mediated endocytosis is essential for a number of cellular processes, and
requires the coordinated action of multiple membrane proteins functioning together (1,
2). The invagination of the clathrin-coated pit is an energetically costly process requiring
multiple proteins, including epsin (3-5), that comes into direct contact and binds to the
membrane surface. The epsin protein is believed to play a role in modelling membranes
in conjunction with clathrin polymerization, and contains multiple conserved binding
motifs which can interact with several accessory proteins associated with clathrin-
mediated endocytosis, for example AP2, Eps15, clathrin, and ubiqutinated proteins (6, 7).
An important feature of the epsin protein is its highly conserved amino-terminal domain,
known as the Epsin N-Terminal Homology (ENTH) domain, which is believed to both
play a role in targeting epsin to the membrane as well as membrane remodeling (8). The
targeting and remodeling ability of the ENTH domain relies on the specific interaction
with the phosphatidylinositol (4,5) bisphosphate (PIP
2
) lipid, in conjunction with the
insertion of an N-terminal amphipathic helix (denoted helix zero, H
0
) into the inner
leaflet of the plasma membrane (8-11). In Ford et al. (8), the full ENTH domain structure
was resolved by co-crystallizing with d-myo-inositol-1,4,5-triphosphate (IP
3
) where it
100
was observed that the three phosphate groups of IP
3
have a highly specific interaction
with up to eight correspondingly positively-charged residues in the ENTH domain. In this
study, H
0
was resolved in the crystal structure, and mutagenesis studies clearly
demonstrated H
0
’s critical role in membrane remodeling. In the absence of the head group
analogue, however, the H
0
region was disordered and could not be detected in the crystal
structure. These data suggest that ENTH membrane binding and bending is directly
related to the structuring of H
0
.
It is instructive to compare the structure of the ENTH domain with another class
of membrane remodeling proteins, namely Bin/Amphiphysin/Rvs-homology (N-BAR)
domains, which include both amphiphysin and endophilin (12-15). In contrast to the
“block-like” shape of the ENTH domain, the N-BAR dimers are crescent shaped with a
number of positively charged residues on the concave surface. Here, the α-helical
structure of the N-terminal amphipathic helix embeds in between the head and tail
regions of lipids of the contacting monolayer when the N-BAR domain comes in contact
with the membrane. Experiments have demonstrated that the concave shape of the BAR
domain, the positively charged residues on the concave surface, and the embedded N-
terminal amphipathic helices, are all critical factors for membrane remodeling (15-22).
101
How each of these structural features combine to drive membrane deformation has been a
topic of recent interest both experimentally (23-26), with atomic-level molecular
dynamics (MD) simulation (27-29), coarse-grain (CG) MD (30, 31) and mesoscopic
simulation (32).
In contrast to the “banana” shape of the N-BAR dimer, the ENTH domain has no
obvious crescent shaped structure and instead is more compact and “block-like”; how this
compact structure can readily bend membranes is not immediately apparent. Based on the
previous experience from N-BAR domain remodeling, it is plausible that the membrane
remodeling ability of the ENTH domain may reside in both the insertion of the H
0
helices
in the membrane outer leaflet (the so-called “wedge” mechanism (26)), as well as the
collective (ensemble (12)) behavior of aggregations of ENTH domains. A number of
scenarios can be imagined: First, randomly oriented, low density aggregations of ENTH
domains may bind onto the membrane surface. In this case, perhaps a weak isotropic
curvature could be generated due to the insertion of the H
0
helix. Another case could be
envisioned where higher density membrane bound aggregations exist where the ENTH
domains oligomerize into dimeric or higher order structures. In this case, the highly

102
correlated H
0
helix insertion, combined with the high density, may facilitate ENTH
domain tubulation.
In order to investigate the aforementioned mechanisms of ENTH membrane
binding and membrane bending, we used a combination of site-directed spin labeling and
Electron Paramagnetic Resonance (EPR) spectroscopy together with MD simulations.
EPR spectroscopy has become a powerful tool in elucidating structural features of
proteins (33-35). This approach has provided important details for the immersion depth
and membrane induced ordering of the N-terminal helix insertions of ENTH (10) and
BAR domain proteins (17). Moreover, EPR spectroscopy can be used to determine inter
residue distances (33, 36-38). In favorable cases, this information provides sufficient
constraints to generate detailed three dimensional structures of proteins with atomistic
resolution (35). Distance information can also provide detailed insights with respect to
the spatial arrangement of oligomerizing proteins (39). Similar to EPR spectroscopy,
atomic-level MD simulation can give quantitative information at atomic-level resolution
(e.g., averaged penetration depths of inserted helices), as well as examine the membrane
and ENTH domain structure at slightly longer length scales (e.g., probing the stability of
different ENTH domain aggregations). Thus, the simulations can be directly related to
103
quantities that are measurable by EPR spectroscopy. Using such a combination of
experiment and theory, this study sets out to investigate the structure and membrane
immersion of H
0
and investigate potential oligomerization upon membrane interaction.

3.2. Materials and Methods
3.2.1. Generation of epsin 1 ENTH (1-164) cysteine mutants.
Rat epsin 1 ENTH domain was cloned into plasmid pGEX4T2. The native
cysteine at position 96 was mutated to serine by site-directed mutagenesis. Specific
cysteine mutations were introduced individually into the cysteine-less background clone
by mutagenesis and confirmed by DNA sequencing.
3.2.2. Purification of wildtype and mutant rat epsin 1 ENTH.
The purification of rat epsin 1 ENTH domain has been described previously (see
Supplementary Information of (8)). Briefly, rat epsin 1 ENTH and its mutants were
generated as N-terminal glutathione-S-transferase fusion proteins. Proteins were
expressed using E. coli BL21 (DE3) cells and induced overnight. The bacterial lysate
was incubated with glutathione-sepharose beads, washed with buffer, then treated with
thrombin to release the protein of interest. Proteins were loaded onto an ion exchange
104
column, eluted with 20mM Hepes pH7.4, 75mM NaCl, 2mM DTT. The protein of
interest came out in the flow-through. Further purification was done by using a gel
filtration column.
3.2.3. Preparation of liposomes.
Liposome preparation has been described previously (8). Lipids were obtained
from Avanti Polar Lipids (Birmingham, AL). 10% cholesterol /40% phosphatidylcholine
/40% phosphatidylethanolamine /10% phosphatidylinositol-4,5-bisphosphate (wt/wt)
were mixed in a test tube and dried using nitrogen gas. The dried lipid was desiccated
overnight, hydrated with 20mM Hepes pH7.4, 150mM NaCl. Lipid samples were briefly
bath sonicated then extruded through 0.4um polycarbonate filters (Whatman) using a
mini-extruder (Avanti).
3.2.4. Preparation of pre-made tubes.
Tubes are composed of 10% cholesterol / 40% phosphatidylcholine / 40%
galactocerebrosides / 10% phosphatidylinositol-4,5-bisphosphate. Lipids were obtained
from Avanti Polar Lipids, while the galactocerebrosides were purchased from Sigma.
The procedure for tube preparation is the same as for liposome preparation (see above).

105
3.2.5. Spin-labeling of single cysteine proteins.
Dithiothreitol (DTT) was added to protein samples to a final concentration of
1mM. DTT was removed by size exclusion using PD-10 columns (GE Healthcare) in
20mM Hepes pH 7.4, 150mM NaCl buffer. 5x molar excess of the MTSL spin label (1-
oxyl-2,2,5,5-tetramethylpyrroline-3-methyl) was incubated with the protein samples for
30 min. at RT. Excess spin label was removed by size exclusion using PD-10 columns.
3.2.6. Sedimentation assays and electron paramagnetic resonance experiments.
10uM rat epsin 1 ENTH domain is incubated with 60ug liposomes or pre-made
tubes in 100ul total volume. After a 10-minute incubation at RT, samples were
centrifuged at 152,800xg for 20 minutes to separate liposome-bound or tube-bound from
unbound. EPR spectra were measured using a Bruker EMX spectrometer fitted with a
dielectric resonator. Spectra were measured at 4mW incident microwave power. The O
2

and NiEDDA П parameters were measured using the power saturation method previously
described (40, 41). The final concentration of NiEDDA was 10mM, while the final
concentration for O
2
was the concentration in equilibrium at RT.
To test for spin-spin interaction of singly labeled epsin derivatives, EPR spectra
were recorded at a scan width of 150 Gauss and spectra for fully labeled proteins were
106
compared with those from spin diluted proteins in which mixtures of 33% labeled and
67% unlabeled protein were used using the same protein to lipid ratios and concentrations
as described above. All spectra were then normalized to the same number of spins by
double integration. Distance estimates were obtained using the deconvolution methods
(36, 38) as implemented by Altenbach et al (36) using an automated fitting routine and a
Gaussian distribution (Shortdistances V21_041) developed by Dr. Christian Altenbach
using LabView (National Instruments) and available for download.
3.2.7. Depth calibration.
To calibrate the immersion depth of R1 at different sites, we used the previously
established relationship d[Å]= a* Φ+b, where Φ = LN( ΠO
2
/ ΠNiEDDA (41). To obtain
the parameters for a and b, we used liposomes containing 10% phosphatidylinositol-4,5-
bisphosphate as well as 1-palmitoyl-2-stearoyl(n-DOXYL)-sn-glycero-3-phosphocholine
(Avanti Polar Lipids) with the spin label attached at the 5, 7, 10 and 12 positions on the
acyl chains (41). For the spin-labeled phosphatidylcholines in liposomes containing
epsin 1 ENTH domain under the conditions described above, we found that a = 3.5 and b
= 0.37.

107
3.2.8. MD Simulation
The original coordinates for the ENTH domain bound to IP
3
were taken from
Ford et al. (Protein Data Bank ID code 1H0A) (8). MD simulations reported here consist
of an ENTH domain dimer and a lipid bilayer system solvated with an explicit TIP3P
water model (42), with sufficient counterions added to maintain overall charge neutrality.
The CHARMM22 (43) and CHARMM27 (44) force field parameters were used to
describe the protein and lipid-protein interactions respectively. The parameterization for
PIP
2
head groups was described previously (27). Except as noted below, the lipid bilayer
consists of 282 1-palmitoyl-2-oleoyl-phosphatidyl -choline (POPC) lipids and two PIP
2

lipids for the dimer systems. The solvated ENTH dimer/ lipid and ENTH monomer/lipid
system consists of ~100,000 atoms. The simulations were performed under isothermal,
isobaric conditions with fully anisotropic pressure coupling (zero surface tension) and
periodic boundary conditions. A Langevin thermostat with a damping coefficient of 0.5
ps
-1
was used to maintain the system temperature at 310 K. The system pressure was
maintained at 1 atm by using a Langevin piston barostat (45). Short-range non-bonded
interactions were truncated smoothly between 10 and 12 Å. The particle mesh Ewald
algorithm (46) was used to compute long-range electrostatic interactions at every time-
108
step. All covalent hydrogen bonds were constrained by the SHAKE algorithm (or
SETTLE for water) (47), permitting an integration time step of 2 fs. System
minimization, equilibration, and dynamics were performed using the NAMD 2.6 software
package (48) . System construction and image generation were done by using the VMD
software package (49).
3.2.9. MD System Preparation
The original coordinates for the pure POPC lipid bilayer were obtained from the
White group (50). In order to embed the ENTH dimer into the bilayer system, two copies
of the initial coordinates were rotated and translated to map over with the experimental
data. In building the starting structure of the dimer, we assumed that the overall crystal
structure geometry of the monomer is maintained. This assumption as well as the
experimental distance and membrane insertion data could only be satisfied with an anti-
parallel arrangement of the H
0
-helices in the dimer and this orientation was chosen as the
starting structure. The resulting dimer and accompanying IP
3
molecules were placed on
the surface of the bilayer such that the long axis of the H
0
helices were perpendicular to
the long axis of the membrane. The POPC lipids with head group phosphorus atoms
closest to the phosphate group on carbon 1 of each of the two IP
3
molecules were mutated
109
to create two PIP
2
lipids as was done previously (27). Each HIS residue in the ENTH
domain was assumed to be ε-protonated (HSE). The following solvation, minimization,
and MD simulation protocol for the protein-lipid system were performed essentially as
described in previous work (27).

3.3. Results and Discussion
3.3.1. The N-terminus of the epsin 1 ENTH domain becomes structured upon
membrane interaction
In order to test whether the N-terminus of epsin 1 undergoes a conformational change
upon membrane interaction, we generated spin labeled derivatives of the epsin 1 ENTH
domain and recorded their EPR spectra in solution and when bound to liposomes
containing 10% phosphatidylinositol-4,5-bisphosphate. As illustrated in Figure 3.1, all
EPR spectra for the singly labeled N-terminal sites (positions 4-14) give rise to sharp and
narrowly spaced lines of large amplitude (Figure 3.1, black dashed line). These spectra
indicate very high mobility and are in agreement with the highly dynamic structure of the
N-terminus shown by NMR and X-ray crystallography (8, 9).

110

Figure 3.1. EPR spectra of R1-labeled Epsin 1 derivatives . X-band EPR spectra of
R1-labeled Epsin 1 derivatives in buffer in absence (black dashed line) and presence of
lipid (red solid line). The green trace for 8R1 is the experimentally determined spectrum
prior to subtraction of the spectrum for the soluble form. All spectra were normalized to
same number of spins and the spectra for the membrane-bound forms are shown at five-
fold magnification. The scan width for all spectra is 100 Gauss.
111
Next, the spin labeled derivatives were bound to liposomes containing 10% PIP
2
.
All samples bound to membranes as assayed by co-sedimentation studies (8) and retained
their ability to induce membrane curvature. Under the present conditions, the epsin
derivatives caused tubulation as well as vesiculation with the latter being the predominant
species as judged by negative stain electron microscopy (data not shown). It should be
noted, that the 7R1 derivative was not generated because this residue is involved in
coordinating headgroup binding via two contacts (8). In the presence of liposomes
containing 10% PIP
2
(Figure 3.1, red trace), all spectra exhibit pronounced line
broadening and concomitant reduction in signal amplitude in agreement with a previous
study (10). To visualize the line shape changes of the R1 sidechain in the liposome-
bound form, all spectra measured in the presence of membrane are shown at five-fold
increased magnification. Most spectra display the characteristic line shapes observed for
surface sites in α-helical structures. The spectrum for 8R1 is more complex as it has an
additional sharp component (Figure 3.1, green trace) that could readily be subtracted
using the spectrum of the unbound protein and the resulting spectrum is given by the red
trace. According to the crystal structure of epsin 1 ENTH in the presence of the
headgroup IP
3
(8), R8 forms a direct contact with the headgroup and this mutation is
112
likely to reduce (but not abolish) the binding affinity. More strongly immobilized
components are observed at a few sites (5R1, 8R1, 9R1, and 12R1, Figure 3.1, see arrow)
suggesting that R1 motion is restricted at these sites, possibly due to contacts with nearby
residues or headgroups (see below). These spectral changes indicate structural ordering in
residues 4 through 14 in the presence of liposomes.
3.3.2. Secondary structure and membrane topology of liposome-bound H
0
To further characterize the structure and topology of the membrane-bound form of the
epsin ENTH domain, we measured the accessibilities of the R1 side chain to the
paramagnetic colliders, O
2
and NiEDDA. The nonpolar O
2
preferentially partitions in the
hydrophobic environment of the membrane while the more polar NiEDDA preferentially
partitions in the solvent. Thus, membrane-exposed residues are more O
2
accessible while
solvent-exposed sites are more accessible to NiEDDA (33, 34, 41). When plotted as
function of residue number, the accessibility parameters ( ΠO
2
and ΠNiEDDA; Figure
3.2) exhibit periodic and out-of-phase oscillations that are highly characteristic of regions
that are exposed to membrane on one side and to the aqueous environment on the other.
The accessibility data can be conveniently summarized by the depth parameter Φ (where
Φ=LN[ ΠO
2
/ ΠNiEDDA]) which increases proportionally with increasing immersion
113

Figure 3.2. Plot of O
2
and NiEDDA accessibilities for liposome-bound epsin 1. The
plot of R1 accessibilities to the paramagnetic colliders O
2
(top) and NiEDDA (bottom)
show a periodicity of between 3-4 amino acids, and an out-of-phase periodicity. This is
indicative of an asymmetrically solvated helix.
114

Figure 3.3. Accessibility measurements indicate the formation of an amphipathic
helical structure in the N-terminus of epsin 1. Panel A shows the contrast parameter Ф
as function of residue number. The period oscillation is indicative of helical structure as
indicated by the sinusoidal line drawn with a periodicity of 3.6 amino acids. Local
maxima (magenta circles) fall onto the hydrophobic face and local minima cluster on the
hydrophilic side of the helical wheel shown in panel B.
115
depth in the bilayer. As shown in Figure 3.3 A, Φ oscillates periodically with respect to
residue number and the period of this oscillation corresponds to that of an α-helix. In this
helix, residues 6, 10, 13 and 14 are lipid-exposed (magenta filled circles) while residues
5, 8, and 12 are the most solvent accessible (green filled circles). Moreover, when plotted
using a helical wheel representation (Figure 3.3B), the lipid-exposed residues (magenta)
cluster together and lie on the opposite side of the solvent accessible residues (green).
Previous studies using SDSL demonstrated that Φ values are proportional to depth of
membrane insertion, and that this depth can be calibrated using spin-labeled derivatives
of phospholipids (41). Based on the calibration described in Materials and Methods, the
average immersion depth of the lipid-facing R1 side chains ( Φ maxima) is on the order of
8 Å. Inasmuch as the nitroxide moiety of the R1 sidechain is typically within 7-10 Å of
the center of the helix (51), we can estimate that the center of the epsin 1 ENTH domain
N-terminal helix is, therefore, at the level of the headgroups where it is likely to exert
membrane curvature effects.
3.3.3. Structural features of H
0
when epsin is bound to pre-formed tubules
Having established that H
0
becomes ordered and forms an amphipathic α-helical
structure upon incubation with liposomes, we next tested whether similar conformational
116
changes might also occur when epsin is bound to pre-formed tubules. Toward this end,
pre-formed tubules were prepared according to previously published methods (52) and
the formation of predominantly tubular structures was confirmed by EM. As illustrated
with the example of 10R1 (Figure 3.4A, red trace), binding to pre-formed tubules causes
pronounced spectral changes indicating that the N-terminal region also becomes ordered
under these conditions. However, there are some clear differences, when compared to the
EPR spectra obtained in the presence of liposomes (Figure 3.4A, black trace). The most
pronounced difference is the overall line broadening and the concomitant loss in spectral
amplitude. Such changes are often indicative of strong dipolar interaction between spin
labels ( 20Å distance). In the present case, any spin-spin interaction would have to be
intermolecular in nature, as only one site (position 10) was labeled in each protein. To
investigate this possibility, we tested whether a dilution with unlabeled epsin (33%
labeled protein and 67% unlabeled protein) could alleviate the spectral broadening.
Indeed, such spin dilution causes significantly sharper lines (Figure 3.4B; green trace)
yielding a spectrum remarkably similar to that of 10R1 incubated with liposomes. It is
well established that the comparison of spectra with and without spin-spin interaction can
be used to evaluate an effective broadening function and determine an inter label distance
117

Figure 3.4. Dimerization of epsin 1. The EPR spectra of 10R1 incubated with pre-
formed tubules (red) or liposomes (black) are overlaid in A). Pre-formed tubule bound
10R1 exhibits pronounced line broadening indicative of spin-spin interaction. To
investigate the effect of spin-interaction on the line shape 10R1 was incubated with pre-
formed tubules in the presence of two- fold excess of unlabeled protein resulting in a
spectrum with reduced line broadening and increased amplitude (green trace in B). Panel
C shows the analogous dilution experiment as shown in B) with the exception that
liposomes were used. The scan width for all spectra is 150 Gauss and all spectra are
normalized to the same number of spins.
118
distribution (33-39, 53). Typically, distances are obtained under frozen conditions, but it
has been demonstrated that distances can also be estimated with reasonable accuracy
provided that the interspin vector does not tumble rapidly (36). This condition is met for
membrane-bound proteins and, using such an analysis (see Materials and Methods), we
obtain a distance distribution centered around 13 Å. Thus, under these conditions, epsin
predominantly forms dimers (or potentially even larger oligomers). Next, we performed
analogous spin-dilution experiments with 10R1 that had been incubated with liposomes
(Figure 3.4C). Also in this case an effect of spin-dilution could be observed, however, the
extent of the dilution-induced spectral change is much smaller. Spectral analysis suggests
that, under these conditions, only a subset of the 10R1 derivatives (~10 to 20%)
undergoes spin-spin interaction suggesting that the aforementioned dimerization is
strongly reduced.
To further investigate the structural features of epsin when bound to pre-formed
tubules, we recorded the EPR spectra of additional selected sites in the N-terminal region.
As shown in Figure 3.5, all of the sites tested revealed the presence of spin-spin
interaction. From analysis of the spectra in Figure 3.5, we find distances of 13 Å and 15
Å for the 6R1 and 13R1 derivatives, respectively. The distance for 4R1 is larger. It has a
119

Figure 3.5. EPR spectra for selected epsin 1 spin labeled derivatives bound to pre-
formed tubules. The black spectra are obtained from proteins labeled at the indicated
positions while the red spectra were obtained using the indicated spin labeled derivatives
diluted with two-fold excess of unlabeled protein.
120
broad distance distribution centered around 21 Å. In the crystal structure, this position
faces back toward the protein and the longer distance is likely due to the fact that this
residue is facing away from the dimerisation interface. In agreement with this notion we
find that the adjacent residue 5, which is projecting into the opposite direction, has the
shortest distance of 9 Å.
Very similar distances were obtained in the frozen state (data not shown).
Interestingly, the EPR spectrum of the spin-diluted form of the 5R1 derivative indicates
substantial immobilization. Such pronounced immobilization is indicative of significant
packing interactions further indicating that this residue is likely to be at or near the
contact surface between adjacent proteins. Collectively, all of these data support the
notion that epsin can dimerize in such a manner that brings the N-terminal regions from
two molecules into close proximity. To further validate the membrane insertion of H
0

under these conditions, we next determined the depth parameter Φ for each of these
derivatives. Positive Φ values were obtained for all sites located on the hydrophobic face
of helix 0 while negative Φ values were observed for positions 4 and 5. These data are
consistent with the notion that H
0
becomes helical with one membrane embedded and one
solvent exposed face.
121
3.3.4. MD simulation of ENTH domains
Both monomer and dimer structures were indicated from the EPR experiments,
thus the MD simulations set out to examine these structures in more detail. The time-
scales accessible to all-atom molecular dynamics (MD) simulation are typically on the
order of tens to perhaps hundreds of nano seconds (ns) and thus do not allow for a direct
examination of the aggregation of ENTH domains. However, an alternative approach that
works within MD accessible time-scales is to directly examine the stability (or instability)
of selected limiting cases and then measure key structural and dynamical quantities; for
example, structural variations and the average root-mean-square-deviation with respect to
the average structure (rmsd) of key residues. Two limiting cases based on previous EPR
experimental observations are thus proposed: 1) a single ENTH monomer, and 2) an
ENTH dimer.
Under the periodic boundary conditions of an MD simulation, both the monomer
and dimer are surrounded by periodic replicas of themselves. Thus, the monomer can be
viewed as a “low density” aggregate where ENTH domains are bound to a membrane but
cannot interact with each other. Likewise the dimer represents a structure with more
correlation. However, the periodic boundaries of the simulation can dampen out
122
membrane undulations and can inhibit membrane remodeling. It is easy to imagine a
scenario where a membrane is bent under periodic boundaries, but it must also “bend
back” to connect with its image. In order to examine ENTH domain membrane
remodeling with MD simulation, much larger systems, with multiple ENTH domains will
be required. The present studies are thus aimed at examining the local structural and
dynamical characteristics of membrane-bound ENTH domains.
We note the MD simulations of a single ENTH dimer bound to a large patch of
membrane (i.e., length scales where an amphiphysin N-BAR was able to sculpt a
membrane (27, 28)) did not result in any measurable membrane bending. This point will
be discussed in more detail below. Of course, real ENTH domain aggregates might span
the entire regime between low density monomers, intact dimers, and oligomers. The
present simulations reflect only the first two cases.
The structure of the monomer and dimer are shown in Figures 3.6 and 3.7,
respectively. In the case of the dimer, the ENTH domain on the left is denoted ENTH1,
while the domain on the right is ENTH2. Both the monomer and dimer were examined
over 45 ns of MD simulation. Details of how these systems were constructed are given in
the Methods section.
123
Figure 3.6. MD simulations of the ENTH monomer. (A) MD simulation snapshot for
the ENTH monomer. The dimer structure is embedded in a POPC lipid bilayer (not
shown) with a single PIP
2
lipid. The PIP
2
lipid head group is shown as red spheres. The
ribbon representation is employed for the ENTH monomer with the H
0
helix in dark
orange, and residues 150 to 158 in the C terminal region of the ENTH monomer shown in
yellow. The PIP
2
binding sites (R7, R8, K11, R25, N30, R63, K69, and H73 residues as
described elsewhere) are shown as blue points. The dashed line refers to the average
location of the lipid membrane phosphate groups. (B) Average position of the C α carbons
on the H
0
helix (residues 5 to 14) along the membrane normal with respect to the lipid
membrane phosphate group (dashed line) for the WT (black line). The S18 and V50 loops
are also shown. The horizontal axis gives residue number and the vertical axis gives H
0

helix penetration depth in Å.

124
Figure 3.6. continued

Figure 3.6. MD simulations of the ENTH monomer.
125

Figure 3.7. MD simulations of the ENTH dimer. (A) MD simulation snapshot for the
ENTH dimer (ENTH1: left; ENTH2: right). A snapshot of the dimer from an equilibrated
MD simulation is shown. The dimer structure is embedded in a POPC lipid bilayer (not
shown) with 2 PIP
2
lipids. The color scheme is the same as in Figure 3.6. The dashed line
refers to the average location of the lipid membrane phosphate groups. Panels (B,C)
Average position of the C
α
carbons on the H
0
helix (residues 5 to 14) along the membrane
normal with respect to the lipid membrane phosphate group (dashed line) for the WT
(black line). The S18 and V50 loops are also shown. The horizontal axis gives residue
number and the vertical axis gives H
0
helix penetration depth in Å.
126
3.3.5. ENTH Monomer and Dimer Molecular Dynamics Simulation Results
As stated earlier, it is believed that the amphipathic H
0
helix of the ENTH domain
is essential for inducing membrane remodeling and tubulation (8). Accordingly, Figure
3.6B shows the average positions (found from averaging over the last 30 ns of the
simulation) of the C
α
carbons corresponding to residues 5 to 14 of H
0
along the
membrane normal with respect to the plane designated by the lipid phosphorus atoms.
Comparing with the experimental results in Figure 3.3A, it can be seen that the overall
hydrophobic/hydrophilic nature of the amphipathic helix is preserved in the simulations.
Residues Lys6, Met10, Ile13 and Val14 all reside in the hydrophobic core of the
membrane, while Arg8 and Asn12 are solvent exposed. Similar behavior is observed with
the dimer in Figure 3.7B and C, although Lys6 is slightly less deeply docked in the dimer
structure. This could be due to the more slightly pronounced downward tilt of the helices
in the dimer.
As a test of the stability of the helices, as well as the duration of the MD
simulations, it should be noted that simulations of a K32E/E42K charge reversal mutant
(54) (results not shown) resulted in a rapid (i.e., over the first 20 ns of simulation)
relocation of the H
0
helices; in the case of the dimer, the helices actually rose out from
127
the membrane. Thus, the agreement of the helix penetration depths in Figures 3.6 and 3.7
with the experimental results in Figure 3.3 is not a consequence the starting ENTH
structures in the simulations, and is more a result of the complex interplay of interactions
involving both the H
0
helix, PIP
2
, and residues within the ENTH domain itself.
Furthermore, the simulation snapshots as shown in Figures 3.6 and 3.7, taken from the
final configurations of the simulations, should give a faithful representation of the actual
representative structures of the ENTH monomer and dimer bound to a lipid bilayer.
A rmsd analysis employing the RMSDTT v2 plugin in VMD (49) over the entire
45 ns trajectory is employed for both the monomer and dimer systems and the results are
shown in Table 3.1. The rmsd was found from aligning each structure in the simulation to
the initial starting structure, and then generating an averaged aligned structure from all
the data. Deviations off this average aligned structure were then reported. Since the
ENTH domain is in contact with the membrane, it was decided to use this averaged
structure rather than the crystal structure reported in Ref. (8) as the reference structure for
the rmsd analysis. Thus, in this context, the rmsd gives an indication of the magnitude of
the fluctuations from the average ENTH domain structure (whether it be the monomer or
the dimer) when bound to the membrane. Overall, interestingly, the dimer exhibits the
128
Table 3.1. The rmsd (in Å) for the ENTH monomer and dimer.

monomer Dimer
rmsd (Å) rmsd (Å) % change
ENTH monomer 1.08 1.61 49
Helix 0 0.95 1.05 10
PIP
2
binding* 0.68 1.11 29
K23 E42 0.66 1.31 98
K23 ( α1)
0.55 1.07 94
E42 ( α2)
0.72 1.50 51
S18 (Loop α0/ α1)
0.83 1.52 108
V50 (Loop α2/ α3)
1.15 1.90 65

*PIP
2
binding sites of the ENTH dimer are designated by residue number 7, 8, 11, 25, 30,
63, 69, and 73.
The rmsd values were calculated by using C
α
atoms relative to average position over all
45 ns simulation for both the monomer and dimer. All structures from the trajectory were
aligned with respect to the first structure of the trajectory. From that aligned point, an
average structure was obtained. The rmsd values were calculated from the deviation of
every structure in the trajectory with respect to average structure. The percent change is
defined as 100(rmsd
DIMER
– rmsd
MONOMER
)/ rmsd
MONOMER
.
129
largest rmsd, with the largest contributions originating from the regions around residues
K23 and E42 (shown in Figures 3.6 and 3.7A). This behavior could be the result of the
charged K23 and E42 residues in one monomer residing in close proximity with matching
residues in the other. Hardly any variation was observed for H
0
, suggesting that the
insertion of H
0
is maintained whether a monomer or dimer structure is considered. The
PIP
2
binding sites exhibited a 29% increase in rmsd in the dimer configuration, which is
probably tied to the large rmsd values of the K23 and E42 residues.
In order to compare the simulations to the experimental data, Table 3.2 shows
some key residue-residue distances between the two H
0
helices. It should be noted that
this is a comparison between the α-carbon distances from the simulations and the
experimental EPR data which represent the distances between different nitroxide moieties
on interacting spin labels. While these distances should be related they are not expected
to be identical. When taking the size of the nitroxide side chain into account, the
simulation data agree well with the experimental distances.

130
Table 3.2. Comparison between MD and EPR results.

Residue-Residue
Distance
MD Results*
α-carbon distances (Å)
EPR Results
Nitroxide distances (Å)
4-4 16.1 21
5-5 9 9
6-6 6.2 13
10-10 11.4 13
13-13 19.1 15
*The errors of the distance in MD results in this table fall within +/- 0.4 Å; the error was
estimated by calculating the standard deviation (SD) of the averaged distances for each
residue over 9 ns trajectory blocks over the entire 45ns simulation trajectory.

131
3.4. Summary and Conclusions
EPR spectroscopy of both vesicles and pre-made tubules indicates that the H
0

penetrates into the outer leaflet of the bilayer, and that in the case of pre-made tubules,
the existence of predominantly ENTH dimers is indicated. Conversely, vesiculated
structures contain mainly ENTH monomers under the conditions used. Immersion depth
measurements indicate that H
0
is located at the level of the head group phosphates, where
it is likely to induce curvature effects by a wedging mechanism.
MD simulations designed to complement the EPR spectroscopy measurements
were performed and considered two limiting cases for the ENTH domain aggregation, a
low density monomer and an ENTH dimer. The results of the MD simulations, in terms
of penetration depths of the H
0
helices, tie directly into the EPR measurements. In both
the monomer and dimer case, H
0
helix penetration into the outer leaflet of the bilayer was
indicated, suggesting that either motif could be capable of membrane remodeling, where
the exact nature of the remodeling (i.e., vesiculation or tubulation) would depend on the
local ENTH domain collective structural correlations (i.e., are the ENTH domains
arranged on average in an isotropic, or more oligomerized dimer-dimer structure). With
just single monomers or dimers, no substantial membrane bending was observed during
132
the time course of the simulations, despite the high degree of H
0
helix penetration. Thus,
the density of either monomers or dimers might need to be greater to allow for efficient
membrane bending. Future simulations at a coarse-grain (CG) level along these lines will
be the topic of future research. It should also be noted that the experimental readout only
distinguished between monomer and dimer formation. The formation of higher ordered
oligomers was not investigated in the present study and we cannot exclude the possibility
that the dimers further oligomerize into higher order oligomers.
From both the EPR measurements and simulation data, the H
0
helices are found to
be near the phosphate level and this location is likely to initiate curvature/wedging effect.
It is remarkable that the location of the H
0
helices in the membrane in the ENTH domain
system is very similar to that observed in BAR domains bound to membranes (29),
despite the fact that in the ENTH domain case, PIP
2
binding plays an additional role. The
amphipathic nature of the H
0
helix results in an ideal structural motif in which to generate
curvature via a wedge mechanism in that it does not penetrate too deeply into the bilayer
(55), yet does not completely dislodge. Given that the largest fluctuations seem to occur
within the ENTH domain itself, it may be possible that the curvature generating ability of

133
the ENTH domain may reside entirely in the H
0
helices, and that the ENTH domain
largely plays the role of a binding pocket for PIP
2
.
The origin of ENTH domain-dependent membrane bending must then clearly
reside in the collective organization of multiple ENTH dimers at sufficient density that
the collective effect of multiple nearly parallel H
0
helices results in a strong anisotropic
spontaneous curvature capable of tubulation, and is supported by the EPR measurements
on ENTH domains bound to pre-made tubules.

134
CHAPTER 3
REFERENCES
1. Doherty GJ & McMahon HT (2009) Mechanisms of endocytosis. Annu Rev
Biochem 78:857-902.

2. Schmid EM & McMahon HT (2007) Integrating molecular and network biology
to decode endocytosis. Nature 448(7156):883-888.

3. Chen H, et al. (1998) Epsin is an EH-domain-binding protein implicated in
clathrin-mediated endocytosis. Nature 394(6695):793-797.

4. McPherson PS, et al. (1998) EH domain-dependent interactions between Eps15
and clathrin-coated vesicle protein p95. Biochem Biophys Res Commun
244(3):701-705.

5. Salcini AE, Chen H, Iannolo G, De Camilli P, & Di Fiore PP (1999) Epidermal
growth factor pathway substrate 15, Eps15. Int J Biochem Cell Biol 31(8):805-
809.

6. Hurley JH & Wendland B (2002) Endocytosis: driving membranes around the
bend. Cell 111(2):143-146.

7. Wendland B (2002) Epsins: adaptors in endocytosis? Nat Rev Mol Cell Biol
3(12):971-977.

8. Ford MG, et al. (2002) Curvature of clathrin-coated pits driven by epsin. Nature
419(6905):361-366.

9. Itoh T, et al. (2001) Role of the ENTH domain in phosphatidylinositol-4,5-
bisphosphate binding and endocytosis. Science 291(5506):1047-1051.

10. Kweon DH, et al. (2006) Membrane topology of helix 0 of the Epsin N-terminal
homology domain. Mol Cells 21(3):428-435.
135
11. Stahelin RV, et al. (2003) Contrasting membrane interaction mechanisms of
AP180 N-terminal homology (ANTH) and epsin N-terminal homology (ENTH)
domains. J Biol Chem 278(31):28993-28999.

12. Frost A, et al. (2008) Structural basis of membrane invagination by F-BAR
domains. Cell 132(5):807-817.

13. Gad H, et al. (2000) Fission and uncoating of synaptic clathrin-coated vesicles are
perturbed by disruption of interactions with the SH3 domain of endophilin.
Neuron 27(2):301-312.

14. Ringstad N, et al. (1999) Endophilin/SH3p4 is required for the transition from
early to late stages in clathrin-mediated synaptic vesicle endocytosis. Neuron
24(1):143-154.

15. Takei K, Slepnev VI, Haucke V , & De Camilli P (1999) Functional partnership
between amphiphysin and dynamin in clathrin-mediated endocytosis. Nat Cell
Biol 1(1):33-39.

16. Farsad K, et al. (2001) Generation of high curvature membranes mediated by
direct endophilin bilayer interactions. J Cell Biol 155(2):193-200.

17. Gallop JL, et al. (2006) Mechanism of endophilin N-BAR domain-mediated
membrane curvature. Embo J 25(12):2898-2910.

18. Masuda M, et al. (2006) Endophilin BAR domain drives membrane curvature by
two newly identified structure-based mechanisms. Embo J 25(12):2889-2897.

19. Peter BJ, et al. (2004) BAR domains as sensors of membrane curvature: the
amphiphysin BAR structure. Science 303(5657):495-499.

20. Richnau N, Fransson A, Farsad K, & Aspenstrom P (2004) RICH-1 has a
BIN/Amphiphysin/Rvsp domain responsible for binding to membrane lipids and
tubulation of liposomes. Biochem Biophys Res Commun 320(3):1034-1042.

136
21. Weissenhorn W (2005) Crystal structure of the endophilin-A1 BAR domain. J
Mol Biol 351(3):653-661.

22. Zhang B & Zelhof AC (2002) Amphiphysins: raising the BAR for synaptic vesicle
recycling and membrane dynamics. Bin-Amphiphysin-Rvsp. Traffic 3(7):452-
460.

23. Farsad K & De Camilli P (2003) Mechanisms of membrane deformation. Curr
Opin Cell Biol 15(4):372-381.

24. Gallop JL & McMahon HT (2005) BAR domains and membrane curvature:
bringing your curves to the BAR. Biochem Soc Symp (72):223-231.

25. McMahon HT & Gallop JL (2005) Membrane curvature and mechanisms of
dynamic cell membrane remodelling. Nature 438(7068):590-596.

26. Zimmerberg J & Kozlov MM (2006) How proteins produce cellular membrane
curvature. Nat Rev Mol Cell Biol 7(1):9-19.

27. Blood PD, Swenson RD, & V oth GA (2008) Factors influencing local membrane
curvature induction by N-BAR domains as revealed by molecular dynamics
simulations. Biophys J 95(4):1866-1876.

28. Blood PD & V oth GA (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(41):15068-15072.

29. Cui H, Ayton GS, & V oth GA (accepted) Molecular dynamics simulation of
membrane remodeling by endophilin N-BAR domains. Biophys J.

30. Arkhipov A, Yin Y , & Schulten K (2008) Four-scale description of membrane
sculpting by BAR domains. Biophys J 95(6):2806-2821.

31. Ayton GS, Lyman E, & V oth GA (in press) Hierarchial coarse-graining strategy
for protein-membrane systems to access mesoscopic scales. Faraday Discussions.
137
32. Ayton GS, Blood PD, & V oth GA (2007) Membrane remodeling from N-BAR
domain interactions: insights from multi-scale simulation. Biophys J 92(10):3595-
3602.

33. Hubbell WL, Cafiso DS, & Altenbach C (2000) Identifying conformational
changes with site-directed spin labeling. Nat Struct Biol 7(9):735-739.

34. Hubbell WL, Gross A, Langen R, & Lietzow MA (1998) Recent advances in site-
directed spin labeling of proteins. Curr Opin Struct Biol 8(5):649-656.

35. 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 U S A 105(50):19666-19671.

36. Altenbach C, Oh KJ, Trabanino RJ, Hideg K, & Hubbell WL (2001) Estimation of
inter-residue distances in spin labeled proteins at physiological temperatures:
experimental strategies and practical limitations. Biochemistry 40(51):15471-
15482.

37. Jeschke G & Polyhach Y (2007) Distance measurements on spin-labelled
biomacromolecules by pulsed electron paramagnetic resonance. Phys Chem Chem
Phys 9(16):1895-1910.

38. Rabenstein MD & Shin YK (1995) Determination of the distance between two
spin labels attached to a macromolecule. Proc Natl Acad Sci U S A 92(18):8239-
8243.

39. Margittai M & 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(3-4):265-297.

40. Oh KJ, Altenbach C, Collier RJ, & Hubbell WL (2000) Site-directed spin labeling
of proteins. Applications to diphtheria toxin. Methods Mol Biol 145:147-169.

138
41. 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 U S
A 91(5):1667-1671.

42. Jorgensen WL, Chandrasekhar J, & Madura JD (1983) Comparison of simple
potential functions for simulating liquid water. J. Chem. Phys. 79:926-935.

43. MacKerell AD, et al. (1998) All-atom empirical potential for molecular modeling
and dynamics studies of proteins. J. Phys. Chem. B 102:3586-3616.

44. Feller SE & MacKerrell AD (2000) An improved empirical potential energy
function for molecular dynamics simulations of phospholipids. J. Phys. Chem. B
104:7510-7515.

45. Feller SE, Zhang Y , Pastor RW, & Brooks BR (1995) Constant pressure molecular
dynamics simulation: The Langevin piston. J. Chem. Phys. 103:4613-4621.

46. Darden T, York D, & Pedersen L (1993) Particle mesh Ewald: An N-log(N)
method for Ewald sums in large systems. J. Chem. Phys. 98:10089-10092.

47. Miyamoto S & Kollman PA (1992) Settle: An analytical version of the SHAKE
and RATTLE algorithm for rigid water models. J. Comput. Chem. 13:952-962.

48. Phillips JC, et al. (2005) Scalable molecular dynamics with NAMD. J Comput
Chem 26(16):1781-1802.

49. Humphrey W, Dalke A, & Schulten K (1996) VMD: visual molecular dynamics. J
Mol Graph 14(1):33-38, 27-38.

50. Jaud S, Tobias DJ, Falke JJ, & White SH (2007) Self-induced docking site of a
deeply embedded peripheral membrane protein. Biophys J 92(2):517-524.

139
51. Langen R, Oh KJ, Cascio D, & Hubbell WL (2000) Crystal structures of spin
labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra in
terms of structure. Biochemistry 39(29):8396-8405.

52. Stowell MH, Marks B, Wigge P, & McMahon HT (1999) Nucleotide-dependent
conformational changes in dynamin: evidence for a mechanochemical molecular
spring. Nat Cell Biol 1(1):27-32.

53. Bhatnagar J, Freed JH, & Crane BR (2007) Rigid body refinement of protein
complexes with long-range distance restraints from pulsed dipolar ESR. Methods
Enzymol 423:117-133.

54. Tong YY, et al. (submitted 2009) Molecular basis of the potent membrane
remodeling activity of the epsin 1 ENTH domain. Biol. Chem.

55. Campelo F, McMahon HT, & Kozlov MM (2008) The hydrophobic insertion
mechanism of membrane curvature generation by proteins. Biophys J 95(5):2325-
2339.

140
CHAPTER 4
Mechanism of endophilin N-BAR domain-mediated membrane curvature

Jennifer L. Gallop
1*
, Christine C. Jao
2*
, Helen M. Kent
1
, Yvonne Vallis, P. Jonathan G.
Butler
1
, Philip R. Evans
1
, Ralf Langen
2
& Harvey T. McMahon
1

1
MRC Laboratory of Molecular Biology, Hills Road, Cambridge, CB2 2QH, UK.
2
Department of Biochemistry and Molecular Biology, Zilkha Neurogenetic Institute,
University of Southern California, 1501 San Pablo Street, Los Angeles, CA 90033, USA.

* Equal contribution by authors

This work was originally published in The EMBO Journal 25, 2898-2910, 8 June 2006
and formatted according to dissertation guidelines. The EPR work was performed by the
author.

141
CHAPTER 4
ABSTRACT
Endophilin-A1 is a BAR domain-containing protein enriched at synapses and is
implicated in synaptic vesicle endocytosis. It binds to dynamin and synaptojanin via a C-
terminal SH3 domain. We examine the mechanism by which the BAR domain and an N-
terminal amphipathic helix, which folds upon membrane binding, work as a functional
unit (the N-BAR domain) to promote dimerisation and membrane curvature generation.
By electron paramagnetic resonance spectroscopy (EPR) we show that the amphipathic
helix is peripherally bound in the plane of the membrane with the midpoint of insertion
aligned with the phosphate level of headgroups. This places the helix in an optimal
position to effect membrane curvature generation. We solved the crystal structure of rat
endophilin-A1 BAR domain and examined the function of a distinctive loop which
protrudes from the membrane interaction face. This insert is predicted to form an
additional amphipathic helix and is important for curvature generation. Its presence
defines an endophilin/nadrin subclass of BAR domains. We propose that the N-BAR
domain functions as a low affinity dimers regulating binding partner recruitment to areas
of high membrane curvature.

4.1. Introduction
Endophilin A proteins have been implicated in membrane curvature generation in
synapses during clathrin-mediated endocytosis as they bind to the endocytic proteins
dynamin and synaptojanin. In Drosophila and in C.elegans, endophilin mutants have
defective synaptic vesicle recycling (1-4). During overexpression of endophilin SH3
142
domain in higher organisms, antibodies against endophilin and peptides that bind to the
SH3 domain all result in the inhibition of vesicle recycling and the accumulation of
clathrin-coated profiles, suggesting an involvement in clathrin coated vesicle formation
(5-7). There is still some vesicle endocytosis in endophilin-deficient flies and the slower
kinetics of this residual component is consistent with clathrin-mediated endocytosis (8).
Thus, endophilin must speed up a clathrin-mediated pathway in flies or be involved in a
separate, clathrin-independent endocytic pathway that is faster than clathrin-dependent
endocytosis.
The ability to effect membrane curvature may implicate endophilin in early stages
of vesicle formation where it could help to generate the initial membrane curvature, or in
late stages where it could aid in vesicle neck formation. The stage of action has been
examined in the context of clathrin-coated vesicle formation given that clathrin-coated
profiles are easily observed by electron microscopy. Evidence against an early stage
function for endophilin comes from studies on clathrin-coated vesicle formation in a cell-
free assay. Here depletion of endophilin did not affect the number and morphology of
clathrin-coated pits (6). A lipid modifying activity of endophilin to aid in membrane
curvature has also been excluded (9). Evidence for late stage involvement has been
obtained from this same cell-free coating assay where a significant reduction of dynamin
coated structures following endophilin depletion was observed. Thus endophilin could be
involved in late stages of endocytosis through either its recruitment of dynamin and/or
the lipid phosphatase, synaptojanin. PtdIns (4,5)P
2
is an important lipid in anchoring a
number of clathrin-coated vesicle components to the membrane, including the clathrin
recruitment and polymerising protein AP180 (10) and thus depletion of this lipid by
143
mobilising synaptojanin to coated vesicles would help release the coat components.
Indeed, endophilin is required in C.elegans for the recruitment of synaptojanin to nerve
terminals (3) and deletion of synaptojanin in mice leads to an accumulation of coated
vesicle profiles (11). In the lamprey reticulospinal synapse disruption of endophilin SH3
domain interactions perturbs uncoating of clathrin-coated vesicles (5). Defective vesicle
scission in this later study also points to a role in dynamin recruitment. We should note
however that there is no firm biochemical assignment of endophilin to clathrin-mediated
endocytosis, as endophilin does not enrich in clathrin-coated vesicles or bind to specific
components of the clathrin-coat machinery and the phenotypes observed could well be
indirect.
By sequence analysis, there are A and B subfamilies of endophilins. In the A
subfamily, there are endophilins A1 (also called endophilin1, SH3P4, SH3GL2 AND
EEN-B1), A2 (also called endophilin2, SH3P8 and SH3GL1) and A3 (also called
endophilin3, SH3P13 and SH3GL3) and in the B subfamily there are endophilins B1
(also called SH3GLB1) and B2 (also called Bif1, SH3GLB2 and EEN). Some of these
are synaptically enriched while others are more ubiquitously expressed (for review see
(12)). They all have the same overall domain structure, N-BAR domain (BAR domain
with an additional N-terminal amphipathic helix) coupled to an SH3 domain by a variable
linker region. The ubiquitous distribution of some endophilins, the interactions with
membranes and trafficking proteins and the role of endophilin-A1 in synaptic vesicle
endocytosis, supports the hypothesis that endophilins perform a general function in
forming transport carriers in different trafficking pathways. A homologous protein,
amphiphysin, has the same overall domain structure, (with an N-BAR domain followed
144
by an SH3 domain, see Figure 4.1A for scheme) and is implicated in T-tubule formation
in muscle and in clathrin-coated vesicle formation (13-15).
The deformation of membrane that is required to make small diameter transport
vesicles, as found at synapses, has a significant energetic requirement. When making
small liposomes in vitro, this energy is provided by intense sonication. In vivo high
curvature can be achieved using peripheral membrane binding proteins that effect and
stabilise curvature (for review see (16)). In particular, the insertion of amphipathic helices
into the hydrophobic phase of the bilayer is proposed to be a general biophysical
mechanism for curvature generation during vesicle budding, based on point mutagenesis
in amphiphysin (17), endophilin (18), epsin (19) and Sar1 GTPase (20). Until now, direct
insertion of amphipathic helices for vesicle budding proteins has not been shown. Here
we show using electron paramagnetic resonance (EPR) spectroscopy, that the N-terminal
amphipathic helix of endophilin inserts into membranes and elucidate the orientation and
depth of helix penetration. In the case of N-BAR domains, both the amphipathic helix
and BAR domain have been implicated in the promotion of membrane curvature and the
relative importance of these two modules has been unclear. In vitro, the N-BAR domain
of endophilin tubulates liposomes and a truncation that includes approximately half the
BAR domain is also effective, as are isolated BAR domains (17, 18). We now carry out a
thorough analysis of the endophilin N-BAR domain using crystallographic and
biophysical techniques. The principles uncovered (driving of curvature by an
amphipathic helix and selection or limiting of membrane curvature by a BAR domain,
and the feed-forward behaviour of N-BAR domain binding) will also apply to other
proteins where one finds this same combination of amphipathic helix followed by BAR
145

Figure 4.1. Ordering of N-terminal residues of endophilin on membrane binding.
(A) Domain structure of endophilin, nadrin and amphiphysin. The C-terminal region of
nadrin has been truncated. (B) In the CD spectrum (room temperature), there is additional
-helical structure in the N-BAR domain on incubation with 50 nm Folch liposomes. This
was not seen for the BAR alone (not shown). (C) EPR spectra of endophilin A1 N-BAR
domains in the absence (black dash) and presence (red) of liposomes. Sample traces for
residues 2, 4, 5 and 10 are shown. Other traces are shown in Figure 4.2. Asterisks point to
additional immobilisation compared to surrounding residues on membrane binding.
Protein (2 µM) was incubated with 1.4 mg/ml liposomes and centrifuged to separate
bound from unbound. Membrane-bound spectra are shown at a magnification of 2.5.
146

Figure 4.2. EPR spectra of R1-labeled endophilin A1 N-BAR derivatives. Samples
were recorded in the absence (black dashed line) and presence of membranes (solid red
line). Spectra were normalized to the same number of spins. Spectra of membrane-
bound samples were magnified by a factor of 2.5 and spectra of the samples in solution
were magnified by the amount indicated on the right. For labeling, a 5x molar excess of
spin label MTSL was incubated with each protein for 30 min. at room temperature and
excess removed by a PD-10 column. This results in near-quantitative labeling.
147

domain, including nadrin and BRAP1/Bin2 N-BAR domains. By homology screening
we also find that nadrin and endophilin are in the same structural subclass of BAR
domain due to a loop present on the concave face. This insert includes a further predicted
amphipathic helix that exhibits membrane interaction capability. Our measurements of
the behaviour of amphipathic helices on membranes are also likely true for the epsin
family of proteins and also for the Arf, Arl and Sar-family GTPases.

4.2. Results
4.2.1. An N-terminal amphipathic helix of endophilin folds and inserts into
membranes
Predicted N-terminal amphipathic helices have been proposed to fold on
membrane binding and anchor membrane-curvature generating proteins in the membrane
to cause displacement of lipids in one leaflet, promoting curvature (17, 19). We now use
direct biophysical methods to determine the structure of these residues on membrane
binding.
On liposome binding, there is an increase in α-helicity of the N-BAR domain of
endophilin from 36% to 48% (estimated from circular dichroism, see Figure 4.1B), which
is not observed without the N-terminal residues (not shown), implying the formation of
additional helical structures. To test how the predicted N-terminal amphipathic helix
folds, whether it inserts into membranes and its topology in relation to the membrane, we
used EPR together with site-directed spin labeling. A series of N-BAR domains were
made where cysteines were substituted for each residue from 2 to 16 by site-directed
148
mutagenesis. Spin labels were then attached to each cysteine mutant and the protein was
used for EPR analysis. The EPR spectra of the spin labelled derivatives in solution are
very sharp and on membrane binding these are broadened for each residue. These data
show that this N-terminal region is disordered in solution, and becomes ordered on
membrane binding (see example spectra in Figure 4.1C and the full range in Figure 4.2).
Residues 4, 5, 8 and 16, which are predicted to lie on a single plane of the amphipathic
helix also show additional immobilization (see asterisks in Figure 4.1C). This may
indicate some other interactions with the BAR or lipid headgroups.
The accessibility of the probe to oxygen (which preferentially partitions into
membranes) and to NiEDDA (preferentially in solution) is plotted in Figure 4.3A and
shows which residues penetrate the bilayer. Up to residue 16, the O
2
and N iEDDA
accessibilities ( П(O
2
) and П(NiEDDA)) exhibit a periodicityt of 3-4 amino acids,
consistent with the formation of a continuous α-helical structure. Importantly, the
periodicities of access to the respective coliiders are 180
o
out of phase. Such behavior is
typical for asymmentrically solvated α-helices in which one face is exposed to the
membrane where the accessibility to O
2
is high; in contrast, residues on the opposing
face are solvent-exposed and consequently more accessible to the hydrophilic NiEDDA.
The membrane exposure of a given site can be conveniently summarized by the constrast
parameter Ф=ln( П(O
2
/ П(NiEDDA)), which is proportional to the depth of membrane
immersion (21). As shown is Figure 4.3B and C, membrane-exposed residues,
corresponding to local maxima of Ф, cluster on one side of the helical wheel, whereas
solvent-exposed sites (local maxima of Ф) lie on the opposite face. The polar, more
149

Figure 4.3. Membrane insertion and orientation of endophilin N-terminal
amphipathic helix. (A) Oxygen (red circles) and NiEDDA (green squares) accessibilities
( П) of membrane-bound N-BAR domain as a function of label position. The graph below
shows a ln( П ratio) plot ( Ф) showing the differential access of colliders to the spin label
and the penetration of hydrophobic residues into the membrane. The periodic oscillation
is indicative of a helical structure. Equivalent maxima indicate that the helix lies planar to
the membrane. (B) Helical wheel representation showing hydrophobic and charged faces.
(C) Model of the amphipathic helix, residues 1–16 with hydrophobic residues coloured
green and surface charge potential also shown. (D) Model of the N-BAR amphipathic
helix to scale with PtdIns(4,5)P
2
and PtdSer lipids showing the depth of penetration of the
helix as calculated from data in (A) and penetration measurements, described in Materials
and methods.
150

solvent-exposed residues of the helix have considerable positive charge and will prefer
negatively charged lipid headgroups or a negative patch on an adjacent protein. The
immersion depths of the lipid facing residues were calibrated using labelled hydrocarbon
chains (21) (see Materials and Methods). Based on this calibration and the data in Figure
4.3A we can place the center of the helix near the phosphate level (Figure 4.3D). This is
the first direct demonstration of amphipathic helix membrane insertion for an endocytic
protein and we propose this will apply to all N-BAR proteins, epsin family members and
Arf/Arl/Sar proteins, thus providing a potential general mechanism by which membrane
curvature is generated by these classes of proteins.
4.2.2. BAR domain structure
The N-terminal amphipathic helix of endophilin is followed immediately by a
predicted BAR domain which is expected to sense or stabilise positive membrane
curvature (17). As the sequence homology between the endophilin and amphiphysin BAR
domains (where a structure was readily available) is low, we crystallized rat endophilin
BAR to elucidate the curvature of the domain (Figure 4.4 and Figure 4.5A). A similar
structure of mouse endophilin BAR domain has since been published (22) and when
these are overlaid the r.m.s. deviation on the dimer is 0.95 Å for 382/408 residues.
Structural details are marked on the endophilin sequence in Figure 4.6 and the initial
amphipathic helix is labeled as helix zero (after the epsin ENTH nomenclature (19))
given that it folds on membrane binding (Figure 4.3A) despite being invisible in the
unliganded crystal structure. The surface charge distribution of endophilin is similar to
that of amphiphysin but more negative charge is concentrated on the convex face (see
151

Figure 4.4. Sequence alignments and 3D models of C. elegans endophilin A and
human endophilin B2 N-BAR domains. Alignments between rat endophilin A1 and
other endophilins with highlighting of conserved charged and hydrophobic residues and
the QPNP sequence which breaks helix 1. The models are constructed according to the
alignments and indicate that the negative surface charge distribution is conserved
amongst the endophilins. The positive charge density on the concave face is widely
distributed rather than the clustering on the ends of the BAR as seen with endophilin A1.
152
Figure 4.5. Endophilin N-BAR crystal structure and alignments. (A) Ribbon
diagram of the banana-shaped rat endophilin-A1 BAR domain (Protein DataBank (PDB)
accession number 2c08) with a view of the concave surface below. Monomers are dark to
light from NH
2
to COOH-termini with one coloured in brown to yellow and the other in
dark blue to light blue. Lysine and arginine residues potentially important for membrane
binding are marked. (B) Superposition of endophilin BAR domain (orange) with
amphiphysin BAR (green). (C) Surface representation of the BAR domain of endophilin
coloured according to electrostatic potential and a mesh equipotential surface contoured
at 0.05V. (D) Similar representation for amphiphysin as in (C). Both molecules are
negatively charged (red) except on the concave face and the tips of the crescent which are
positively charged (blue). The overall shapes are very similar. (E) Hydrophobic residues
in the dimer interface of rat endophilin-A1 BAR are coloured green on the surface
represented monomer. Dimensions of the BAR domain are also indicated. (F) Calcium
does not bind to the endophilin N-BAR domain. Endophilin N-BAR domain was
decalcified by purification in the presence of 2 mM EDTA followed by incubation with
10 mM EDTA and then extensive dialysis against Ca-free buffer (prepared using
plastics). The absorbance of di-bromo BAPTA (affinity for Ca
2+
of 2 µM in the absence
of Mg
2+
) at 265 nm was followed on calcium titration in the presence and absence of
decalcified endophilin BAR domain. No difference was observed suggesting that
endophilin does not effectively compete with di-bromo BAPTA, even at double the
concentration, for Ca
2+
ions. (G) Isothermal titration calorimetry was used to test for heat
changes on CaCl
2
injection into decalcified endophilin N-BAR domain. These
experiments were performed with 6 µl injections of 1 mM CaCl2 into 45 µM full-length
153
Figure 4.5. continued
endophilin and 76 µM endophilin N-BAR domain at 10 or 25 °C. Example results for the
N-BAR domain are shown and no binding was observed. (H) Modelling of loop residues
62-79 as an α-helix. Hydrophobic residues are predominantly on one side and are flanked
by positively charged residues.
154
Figure 4.5. continued

Figure 4.5. Endophilin N-BAR crystal structure and alignments.
155
endophilin in Figure 4.5C and amphiphysin in Figure 4.5D). The high negative charge on
the convex surface is conserved among endophilins (Figure 4.4). In the structure solved
by Weissenhorn, 11 cadmium ions were bound to the surface and he posits that
these may mimic calcium-binding sites of endophilin in vivo. We looked for calcium
binding using diBromoBAPTA, which allow detection of low micromolar affinities and
by using isothermal titration calorimetry where we should detect nanomolar affinity
interactions (Figure 4.5F and G). We see no evidence of specific calcium interactions.
The curvature of endophilin BAR, formed by the angle of dimerisation and kinks
in its helices, is the same as amphiphysin BAR (see overlay Figure 4.5B and structural
alignment in Figure 4.6) and is also the same as arfaptin BAR (not shown), and thus these
proteins cannot be distinguished by a difference in their predicted membrane curvature
preference. A large hydrophobic patch is buried in the dimer interface (colored green on
the surface represented monomer in Figure 4.5E). The buried surface area is 2870Å
2

while in amphiphysin there is 2400Å
2
buried. The curvature of the domain is likely to be
rigid as the mouse and the rat endophilin BAR structures are derived from crystals with
different crystal packing, yet show the same radius of curvature (85 Å in the absence of
the helix 1 insert; Figure 4.5E; see next paragraph). In the present structure of the rat
endophilin BAR, the extremities are involved in crystal contacts, leading to ordering of
these flexible regions.
Endophilin BAR has an extra insert in the middle of its membrane binding face
compared to amphiphysin BAR. We call this the ‘helix 1 insert’ (H1I, residues 60-87).
The sequence of this insert differs considerably between endophilins and is also found in
nadrin N-BAR family members. The QPNP sequence, which follows the break in helix1,
156

Figure 4.6. Structural alignment of endophilin and amphiphysin BAR domains and
alignment to nadrin showing close homology to endophilin.
157
appears to be diagnostic for endophilin family members across different species (Figure
4.4). The H1I is mostly invisible (and this disordered) in the crystal structure (Figure 4.3)
apart from an initial short helix (see Figure 4.6). This is predicted to continue as an
amphipathic helix for several more turns. This would be particularly favored in the low
dielectric constant environment under the BAR domain and near the membrane. In a
helical wheel representation (see Figure 4.5H), the hydrophobic side of the predicted
helix is flanked on both sides by positively charged residues. This may indicate
penetration of the membrane by the hydrophobic residues (similar to the N-terminal
amphipathic helix) and accompanying electrostatic interactions with the lipid headgroups.
Although by circular dichroism we were unable to detect an increase in helicity upon
addition of liposomes to the BAR domain, this is not surprising as the BAR domain alone
does not bind well to membranes and at least part of the helix appears to be already
folded before binding. We chose one residue beyonf the initla helix (M70) to test the
possibility of folding on membrane interaction. The EPR spectrum shows some ordering
of this residue upon membrane interaction (Figure 4.5I). It should be noted, however,
that the spectral change upon membrane interaction is not as pronounced as in the N-
terminal regions. This is owing to the fact that, in solution, position 70 is less dynamic
than the N-terminal residues and consequently the observed mobility changes are less
dramatic. This is also not surprising given that this helix appears to start to fold in the
crystal structure. The O
2
and NiEDDA accessibilities for residue 70 resulted in a Ф value
of 1.5, demonstrating that this position is indeed membrane-exposed at an immersion
depth of approximately 6Å. A predicted amphipathic sequence is also found at the C-
terminal end of the nadrin insert.
158
We find no evidence in our structure for a lysophosphatidic acid acyl transferase
active site. We have previously tested extensively for biochemical evidence of this
activity and showed it was a contamination of protein preparation (9).
To test if the dimerisation seen in the crystal holds true in solution, we used
equilibrium ultracentrifugation of the N-BAR domain (Figure 4.7A and residuals plotted
in B) and full-length protein (Figure 4.8A or 4.9A). The dimerisation constant for the N-
BAR is 10 µM and fits very well to a monomer:dimer equilibrium. This means that the
protein could well be monomeric in cells – the concentration of endophilin in brain
extract was estimated by blotting to be ~0.1 µM, making the concentration at the synapse
at perhaps ~1 µM. For the full-length protein there is evidence of higher order oligomers
at high concentrations (Figure 4.8A or 4.9A). This could be due to the previously
proposed intramolecular interaction between the SH3 domain and the central proline-rich
linker (23).
As the model for BAR domain membrane binding is scaffolding via the concave
face (17), and the concave nature is only found in the dimeric form, it is surprising that
the K
D
for dimerisation in solution is as high as 10 µM (Figure 4.7A). Spin-coupling
between labeled site-directed cysteine mutants at position 227 was used to test if the
dimer is the predominant form of the protein on membranes. K227 is located near the
dimerisation interface and the α-carbon distance between the two K227 residues in the
crystal dimer is 8 Å (see residue marked in Figure 3A). Introduction of a spin label at
position 227 gave rise to a strong spin-spin interaction for this endophilin mutant and the
resulting spectrum of the membrane bound form exhibited strong dipolar broadening that
is characteristic for spin labels that are in close proximity (Figure 4.7C, red trace). This
159

Figure 4.7. Endophilin N-BAR dimerization on membrane binding. (A) Equilibrium
sedimentation data for rat endophilin-A1 N-BAR domain. Endophilin N-BAR domain
dimerizes with a Kd of ~10µM. The readings from 3 cells are fitted with an ideal
monomer/dimer equilibrium that fits with a 26kDa monomer. The calculated molecular
mass is 28.7kDa. Residual are plotted in B. (C) EPR spectra of membrane-bound
endophilin reveals dimer interactions. Spectra were obtained either from protein that was
fully labeled at position 227 (red trace) or from a mixture of 25% labelled protein and
75% unlabeled protein (black trace). The scan width was 200 Gauss. The amplitudes and
line shapes of the respective spectra were very different due to the strong dipolar spin-
spin interactions which were present in the fully labeled case. The difference between
the respective spectra was used to determine the distances using a Pake pattern analysis
(24) and the blue spectrum can be used to evaluate the quality of this distance analysis. It
is based on simulations in which the calculated distance distribution (shown in panel D)
was used to generate a dipolar broadened spectrum from the black spectrum. This
simulated spectrum closely corresponds to the observed spectrum for the fully labeled
case indicating a good quality fit.
160

Figure 4.8. Circular dichroism of endophilin constructs. CD of proteins obtained
using a Jasco J-810 spectrapolarimeter. Proteins were diluted to 0.2mg/ml in 5mM
Hepes, 30mM NaCl, 0.4mM DTT. Readings were performed at 20
0
C using an average of
5 scans with buffer subtracted. Mean residue ellipticity was calculated using
DICHROWEB (BBSRC), an online server for protein secondary structure analyses from
circular dichroism data (25).
161
Figure 4.9. Anaytical ultracentrifugation of rat endophilin A1. (A) Analytical
ultracentrifugation of wildtype rat endophilin A1. Equilibrium sedimentation of full-
length endophilin shows the formation of higher order oligomers. The data are fitted
with a monomer/dimer/tetramer equilibrium. For the fit, the monomer is predicted to be
26.2 kDa (but it should be 39.9 kDa). The errors in K
d
are also very large and thus we
conclude that there is a higher order ologomerization occurring in the cell. (B)
Analytical ultracentrifugation on endophilin N-BAR mutants. Velocity sedimentation on
endophilin N-BAR domain and mutants (the data are summarized in Figure 4.10). The
scans plot shows every 10
th
trace. The method used for the fitting has been previously
described (9). Except for N-BAR ∆H1I, the sedimentation fits as a dimer. For the helix1
insert excision, it can be seen that the peak in g(s
*
) is broader and this fits as a monomer.

162
Figure 4.9. continued

Figure 4.9. Anaytical ultracentrifugation of rat endophilin A1.
163
spectrum was very different from a control spectrum for the K227 derivative, in which
the dipolar interaction was strongly reduced by co-mixing of 25% labeled protein with
75% unlabeled protein (Figure 4.7C, black trace). Quantitative analysis of the spectra
with and without dipolar interaction allowed us to determine inter spin label distances
using Pake patterns (24, 26) and the resulting distances ranged between ~8 to 10 Å
(Figure 4.7D). These data are in excellent agreement with the crystal structure and clearly
demonstrate that membrane-bound endophilin forms the dimer interactions seen in the
crystal.
4.2.3. Amphipathic helices and the BAR domain collaborate to effect membrane
curvature
The ability of endophilin to generate and stabilize membrane curvature can be
assessed using liposomes with negatively charged lipids to which the N-BAR domain
binds. By electron microscopy, we can determine the shape changes of these liposomes.
The amphiphysin N-BAR domain and arfaptin2 BAR domain constrict vesicular
liposomes into tubules and higher concentrations of the BAR domain proteins lead to
vesiculation (17). Here, tubulation is the initial response of the liposome owing to an
increase in curvature in one direction being compensated by a relaxation in a
perpendicular direction.
The endophilin N-BAR domain has three components: the N-terminal
amphipathic helix, BAR domain and an internal amphipathic helix (H1I). We made a
series of mutants to examine their contribution to membrane binding and curvature
generation (Figure 4.10). Wild type N-BAR (and full-length protein) forms highly curved
164
Figure 4.10. Endophilin has collaborative membrane binding and tubulation
regions. (A) Table summarising constructs used their liposome binding and tubulation
abilities (see methods). N-BAR covers residues 1-247. F10E is a mutant of the
hydrophobic face of the amphipathic helix. BAR covers residues 33-247. KKK-EEE is a
mutant N-BAR with residues 171, 172 and 173 converted to glutamic acids. N-BAR loop
excision is a deletion of residues 59 to 87 and insertion of 2 glycines. The other mutants
are in the loop. BAR loop excision is residues 33-247 with the 59-87 deletion and 2
glycines inserted. Both the amphipathic helix and BAR domain are required for efficient
membrane binding of the N-BAR. Binding to 50 nm liposomes (p:pellet, s:supernatant)
and tubulation of 200 nm or 400 nm liposomes (monitored by electron microscopy). The
degree of tubulation is a reflection of the number of tubules. (B) Electron micrographs of
liposome tubulation by rat endophilin-A1 N-BAR and mutants. Insets: close-ups of the
tubules show their similar morphology. The M70S+I71S mutant gives both wide and
narrow tubules some with budding profiles. (C) Table showing velocity
ultracentrifugation results for mutants and WT protein. The N-BAR loop excision is a
monomer as the apparent molecular weight is close to that of monomeric N-BAR domain
(boxed). Both 40 and 150 µM protein were used and gave the same results.
1
Absorbance
optics were used and the experiments were carried out at 20 °C in 250 mM NaCl;
2

Interference optics were used at 5°C in 150 mM NaCl and the remaining conditions as in
Figure 4.9B.

165
Figure 4.10. continued

Figure 4.10. Endophilin has collaborative membrane binding and tubulation
regions.
166
tubules (of 35-50 nm diameter) from liposomes and at high protein concentrations there
are also many small vesicles (35-50 nm diameter).
Mutation of F10 in the endophilin N-BAR domain (on the hydrophobic face of
the N-terminal amphipathic helix) or deletion of the amphipathic helix leaving just the
BAR, reduces both liposome binding and tubulation (Figure 4.10A and B). Both of these
mutants lead to formation of non-uniform tubes and squashed liposomes (especially
F10E, see arrows). Mutation of the positively charged lysine residues at the tips of the
BAR domain to glutamates (KKK-EEE) also decreases binding and tubulation. Thus both
the amphipathic helix and the BAR contribute to stable tubule formation.
Three different mutants of helix1 insert (H1I) were made: a deletion (N-BAR
∆H1I), a double mutant of hydrophobic residues (M70S, I71S) and a double mutant of
positively charged residues (K76E, R78E). The N-BAR ∆H1I mutant protein has the
same overall secondary structure as WT protein according to CD spectroscopy (Figure
4.8). It also binds liposomes similarly to WT protein but tubulates less efficiently. This
was surprising given that amphiphysin and arfaptin BAR domains do not have this helix1
insert and are good tubulators (17). However, when we tested this mutant by
ultracentrifugation we found that it was monomeric (Figure 4.10C and Figure 4.8B).
Hence for efficient curvature generation the spontaneous curvature driven by the
amphipathic helix needs stabilization from the BAR domain dimer. It would also appear
from the position of this insert in the structure that it may help to stabilize the dimer.
We further examined the role of the H1I with various point mutants. Binding to
liposomes is decreased when the positive charges are made negative (K76E, R78E) and
167
this mutant does not tubulate well. This is likely owing to the repulsion of the Bar from
the similarly charged membrane. The M70S, I71S mutant is interesting because it binds
well but gives rise to both narrow (~20 nm diameter) and WT-diameter tubules (although
the narrow form are a minor component across the grids). We showed that M70 inserts
into the membrane and thus this mutant is unlikely to allow the BAR domain to sit down
on the membrane. The charge interaction of the BAR domain is still intact, as is the N-
terminal amphipathic helix, and these narrow tubules are very similar to those made by
the epsin ENTH domain, suggesting that this mutant results in an accumulation of helix
insertion in the absence of the curvature-stabilising BAR domain. Both the M70S, I71S
and K76E, R78E mutants form dimers (ultracentrifugation data) and thus these residues
do not contribute significantly to dimer formation in solution. The N-BAR ∆H1I shows
us that promoting membrane curvature requires the dimerised BAR domain, as predicted
from the structure. This is supported by the BAR ∆H1I binding weakly to all sizes of
liposomes (Figure 4.10A and data not shown) and suggests that the curvature sensing we
previously observed with the BAR domain is dependent on dimer formation. We propose
that the ∆H1I exhibits membrane interactions via insertion of an amphipathic helix. This
is consistent with a role in defining the precise membrane curvature and does not exclude
further interactions of ∆H1I residues with integral membrane proteins.
4.2.4. Creation of high curvature membranes promotes membrane fusion
The electron microscopy tubulation assay gives an accurate reflection of the
morphological consequences that curvature proteins have on liposomes, but it is not
quantitative, because tubules have a larger surface in contact with the grid than liposomes
and thus these are over-represented, and protein coated liposomes tend to bind better than
168
naked liposomes. Importantly, by electron microscopy, we observe that the tubules
formed are often longer than expected from tubulation of a single liposome (Figure
4.11A). We therefore tested whether the tubules fuse and if a fluorescence resonance
energy transfer (FRET)-based membrane fusion assay would provide a more quantitative
measurement (albeit indirect) of the extent of tubulation. Phosphatidylserine liposomes
were spiked with phosphatidylethnolamine lipids labelled on their polar headgroups with
a FRET pair, NBD and rhodamine, are subsequently mixed with unlabelled liposomes
(27). Concentrations of fluorophores are chosen such that FRET between them decreases
on fusion with unlabeled liposomes (27). As a control, fusion can be initiated by addition
of calcium (Figure 4.11B). Titration of the N-BAR domain into 50% labelled and 50%
unlabeled liposomes (heterotypic case) decreases the FRET efficiency showing that
fusion of the liposomes occurs (Figure 4.11C). There is a small background change
resulting from the addition of protein, which is revealed if only the fluorescent liposomes
are used (homotypic case, control). The BAR domain of endophilin tubulates
inefficiently and does not lead to a decrease in FRET signal. We analysed all our
mutants at 55 µM by calculating the ratio between the peaks in the emission spectrum
(530 nm of NBD and 585 nm of rhodamine) to decrease systematic errors (Figure 4.11D).
The background FRET change (likely due to binding and immobilization of the
fluorophores) in the homotypic case is subtracted from the heterotypic case. More
detailed results are shown in Figure 4.12. We observe membrane fusion for the wild type
N-BAR domain and M70S, I71S mutant but not for the other mutants or the BAR domain
on its own. This is consistent with the electron microscopy data and shows that tabulation
and fusion are coupled.
169

Figure 4.11. High membrane curvature promotes membrane fusion. (A) Example
of endophilin N-BAR tubules made from liposomes extruded to a size cut-off of 400nm.
(B) Emission spectra from mixed liposomes in the absence and presence of calcium
(which promotes fusion) and triton (to obtain total donor fluorescence). See methods for
details of the assay. (C) FRET assay of membrane fusion, showing dilution of the FRET
pair into unlabeled liposome in the presence of the N-BAR (see results). The N-BAR
control has no unlabeled liposome and thus there can be no dilution of the FRET pair.
BAR domain does not lead to membrane fusion. As highly curved membranes are more
fusogenic we believe that the fusion seen with the N-BAR domain is a readout of the
efficiency of curvature generation. (D) Comparison of mutants and WT N-BAR domains
at 55 µM in the fusion assay. Values displayed in the bar chart ± SEM are the difference
in ratios of emission maxima between experiment and controls (see results). Student t-test
* p0.001, ** p0.2.
170

Figure 4.12. FRET assay for membrane fusion. Bar chart showing mixed and
uniform liposome emission maxima intensity ratios. There is fusion of liposomes when
endophilin N-BAR and M70S, I71S mutant are added, shown by the decrease in FRET
on dilution of the two fluprophores and increase in the ratio of donor and acceptor
emission maxima (purple bars). In the control condition (uniform liposomes), there is
also a change on addition of protein, shown by the green bars with the proteins compared
with liposomes alone. The experimental condition (purple bars) is mixed with liposomes.
Errors are s.e.m. Values shown are for 55µM protein.

171
This in vitro observation raises the possibility that endophilin (and indeed other
N-BAR proteins) in vivo may be fusogenic, although there is no direct evidence for this.
4.2.5. Model for endophilin membrane binding and curvature
Given the rigidity of the positively-charged concave face of the BAR, this domain
will act like a scaffold in binding to negatively-charged membranes. We have previously
shown that the BAR domain of endophilin binds better to liposomes of higher curvature
(9), which supports the idea that the endophilin BAR domain can scaffold membrane
curvature using the crescent shape of the domain. We now confirm that the interaction of
the BAR domain with membranes is electrostatic, as binding is sensitive to the salt
concentration (Figure 4.13A), as expected from the scaffolding hypothesis. Whereas the
BAR domain alone does not bind to membranes in the presence of 250 mM NaCl, the N-
BAR binds tightly. This is consistent with salt-insensitive interactions being provided by
the amphipathic helix. Based on the observations with epsin1, where the amphipathic
helix decreases the off-rate of membrane binding (28), we would predict that the
amphipathic helix of endophilin would decrease the off rate, anchoring the domain in the
membrane. As electrostatic interactions act over a longer range than hydrophobic
interactions, electrostatics are likely to be primary determinants of the on-rate and most
of these charge interactions are contributed by the BAR domain (29). This hypothesis is
supported by the results in Figure 4.13B. If the N-BAR domain is pre-bound in
physiological concentrations of salt, then addition of high salt (500 mM) leaves more
protein bound than if the protein is added after high salt. (The liposomes remain intact
during this experiment and sucrose-filled liposomes were also used to control for osmotic
172
Figure 4.13. Endophilin N-BAR domain binds to membrane is initially driven by
electrostatics (A) Salt sensitivity of the BAR domain shows the predominance of
electrostatics in the interaction. Like the N-BAR domain of endophilin epsin ENTH
domain inserts an amphipathic helix giving a hydrophobic contribution to the interaction.
Liposomes were sedimented and pellets (p) and supernatants (s) were separated by SDS-
PAGE. (B) Recruitment of N-BAR domain to liposomes is driven by electrostatics, but
after the initial interaction the dissociation is not salt-sensitive. In the ‘salt before’
experiment protein was incubated with liposomes in the presence of 500 mM NaCl for 10
min. In the salt after experiment 500 mM NaCl was added to the protein liposome mix
after the 10 min incubation and further incubated for 3 min. (C) Model of endophilin N-
BAR domain binding to membranes. From experiments in Figure 4.5 and 4.7 we know
the BAR binds via its concave surface and from Figure 4.3 we know the amphipathic
helix lies flat in the plane of the membrane, but we do not have any information on the
orientation of the amphipathic helix in the membrane with respect to the BAR or the
position of the loops. If the amphipathic helices are parallel to the long axis of the BAR
then the domain will favor 2-dimensional membrane curvature. If the helices are
perpendicular, then the domain will favor one-dimensional curvature.

173
Figure 4.13. continued

Figure 4.13. Endophilin N-BAR domain binding to membrane is initially driven by
electrostatics.
174
shock.) The displacement of the N-BAR domain bound to membranes is less vulnerable
to the charge screening effect of ionic strength than is de novo binding of the domain.
The charge contribution is mostly from the BAR domain, which also binds better to small
liposomes (9). The preference for smaller liposomes is evident in this assay: when salt is
added before protein, more protein is bound to the 50 nm liposomes compared with 400
nm. This leads to the hypothesis that as well as acting to generate curvature via the
amphipathic helix and BAR domain scaffold, N-BAR domains can also respond to
membrane curvature as binding to these areas is kinetically favored. This gives a positive
feedback loop between curvature sensing and generation, leading to very rapid membrane
deformation.
To summarize, there are two synergistic ways by which endophilin modulates
membrane curvature (Figure 4.13C). The penetration of the amphipathic helix leads to an
asymmetry between the outer and inner leaflets of the bilayer causing an increase in
positive curvature while the concave face of the BAR domain acts to scaffold membrane
curvature. In Figure 4.13C, we show the amphipathic helices under the BAR domain,
whre it could shield the negatively charged patch on this surface and position the helices.
This would correlate well with the immobilisation of some of these residues in EPR
spectra. The internal amphipathic helices stabilizes the formed dimer as well as further
promoting membrane curvature generation. The principles described for amphipathic
helix insertion and BAR domain dimerisation and binding are expected to hold for all N-
BAR domains. In Figure 4.13D, we propose a temporal model by which these processes
occur.

175
4.3. Discussion
Here we have demonstrated directly the folding and insertion of an amphipathic
helix and its insertion into membranes for the N-terminus of endophilin N-BAR. We
determined its orientation in the membrane and penetration depth. The EPR technique
also allowed us to show that the BAR domain dimer is the presence on the membrane.
Using the monomeric ∆H1I protein, we can also conclude that dimerisation is essential
for the action of the BAR but that some binding and membrane deformation can be
provided by the amphipathic helix in the absence of the dimerised BAR domain scaffold.
A similar conclusion can be drawn from the N125 construct used in (18) where we would
posit that the tubulation observed is due to the insertion of the amphipathic helix rather
than the BAR. The importance of the N-terminal amphipathic helix to endophilin
function in vivo was demonstrated in C. elegans where F10E endophilin did not rescue
endophilin-null mutant (3).
We propose that a major component of any budding pathway is a mechanism for
bending membranes to the desired curvature. In clathrin-coated vesicle formation, epsin
molecules can bind to PtdIns(4,5)P
2
and insert an amphipathic helix promoting
membrane curvature while at the same time stimulating clathrin polymerisation and thus
stabilization of the generated curvature (19). It may well be that active insertion of an
amphipathic helix is a general mechanism in budding of many types of vesicles, and this
role can be executed for example by other ENTH domain proteins or by small GTPases
of the Arf and Arl families. Many BAR domain proteins also have associated
amphipathic helices (either in cis: N-BAR proteins, or in trans: e.g. Arf binding to
arfaptin) and thus we would predict that these are adept at effecting membrane curvature.
176
We have shown that electrostatics seem to govern the on-rate of the N-BAR domain on
membranes, whereas the interactions of amphipathic helices with membranes seem to
limit the off-rate. It is interesting that the endophilin dimer will have four amphipathic
helix insertions and thus this protein is likely to reside for a considerable period on highly
curved membranes. The N-terminal amphipathic helix has the midpoint of its insertion at
the phosphate level of headgroups, thus acting as a wedge in the membrane at the same
time as anchoring itself via positive charges to negatively charged lipid headgroups.
It is interesting to consider that N-BAR domains may lead to a positive feedback
loop whereby curvature generation leads to further curvature generation allowing very
rapid invagination once a critical concentration of endophilin is reached on the
membrane. Also, the minimal radius of curvature is likely to be defined by the
dimensions of the BAR domain, as shown by tubulation and vesiculation in our electron
microscopy observations. Thus, when the concave face is disrupted, as in the M70S,
I71S mutant, narrower tubule widths can be observed.
We observe many tubules in vitro after N-BAR domain binding to liposomes and
in vivo tubulation of membrane compartments is also observed on overexpression of
vaious N-BAR domains (15, 30); data not shown). In Drosophila muscle, the N-BAR-
containing protein amphiphysin does not bind to dynamin and thus these tubular
structures are not severed, whereas in vesicle budding the recruitment of dynamin by SH3
domains of N-BAR proteins will likely result in vesicle scission rather than extensive
tubulation. Thus, the balance between tubulation and vesiculation may depend largely on
the downstream interaction partners.
177
The binding of downstream partners may also be regulated by the dimerisation of
the proteins on membranes. We have measured the affinity of amphiphysin SH3 domain
for dynamin interaction peptides as 50-100 µM. This is a very weak interaction.
However, as dynamin is a dimer, multiple amphiphysin SH# domains bound to beads can
be used to purify dynamin (31). Likewise, it is very likely that endophilin membrane-
bound dimmers will have a much greater avidity for dynamin than endophilin monomers
in solution and thus recruit dynamin to areas of high membrane curvature.
We had previously observed that the tubules formed by epsin, amphiphysin and
others were longer than would be expected from the initial liposome size. We show here
that on addition of curvature-generating endophilin N-BAR domain that membrane
fusion takes place. This has provided a complementary approach to electron microscopy
as a more quantitative readout of the action of mutants on liposome deformation. More
interestingly it raises the possibility that exocytosis or other vesicle fusion events can be
enhanced by increased curvature generation.
Overexpression of endophilin A1 (without its SH3 domain) in fibroblasts does not
affect the endocytosis of transferring (a clathrin-mediated pathway), but overexpression
of a similar construct of amphiphysin does (data not shown). This is probably owing to
targeting sequences in amphiphysin, which allow it to bind to both clathrin and adaptors,
which are absent in endophilin. The absence of these interaction sequences makes
endophilin distinct from clathrin-endocytic proteins. It has previously been noted that
endophilin appears to be transported to synapses on vesicles (32) (and thus is likely to be
on synaptic vesicles before fusion), binds the membrane scission protein dynamin and is
missing the targeting sequences for recruitment into clathrin-coated pits. We therefore
178
speculate that the role of endophilin is distinct from that of amphiphysin, and that the role
of endophilin is to generate/stabilize curvature in a clathrin-independent mechanism in
synapses, where it is found enriched.
Given that endophilin is enriched in synapses, is capable of curvature-induced
curvature generation, binds to dynamin and may well be targeted to to synapses on
vesicles, this may suggest a possible involvement in fast, ‘kiss-and-run’ clathrin-
independent endocytosis at synapses. This is consistent with the observations of Schwarz
and colleagues in endophilin null mutants, where the kinetics of the remaining
endocytosis are slow (8). An inhibition of fast, kiss-and-run endocytosis would lead to a
kinetic slowing in synaptic vesicle endocytosis, as observed in multiples systems, as the
time constant for the clathrin-mediated pathway is slow (33). We note that the previous
assignment of endophilin to clathrin-mediated endocytosis (4, 6) is without firm
molecular basis and that the observed accumulation of clathrin-coated intermediates after
endophilin depletion could equally be due to a compensatory upregulation in this
pathway as opposed to its inhibition.
In this paper, we have presented a molecular mechanism of how the BAR and
amphipathic helices of N-BAR domains work as a unit to promote membrane curvature.
As membrane-bound protein is dimeric, it will lead to the presentation of two SH3
domains, favoring binding to multimeric ligands (like dynamin). Thus, a low affinity
dimerisation of the N-BAR domain may be a mechanism not only to regulate the
recruitment of endophilin to membranes but also to regulate binding partner recruitment
to areas of high membrane curvature.

179
4.4. Experimental procedures
4.4.1. Constructs and liposome assays. GST full-length rat endophilin, GST rat
endophilin N-BAR domain (residues 1-247) and GST-human arfaptin2 BAR domain
(residues 117-end) were cloned into ER1/Not1 sites of pGex4T2 and proteins were
thrombin cleaved before purification by anion exchange and gel filtration. Rat
synaptojanin 145 (gift from Peter Parker) was cloned into the Not1 site of pBac4x1 with
a hexa-histidine tag at the C-terminus. Protein was purified on Ni-NTA agarose (Qiagen)
followed by S200 gel filtration. Folch Fraction I (Sigma) liposomes in 150 mM NaCl, 20
mM HEPES pH 7.4 were extruded 11 times through polycarbonate membranes (Avanti)
to achieve desired diameter. For tubulation assays typically 1 mg/ml 200 nm liposomes
were incubated for 10 min with 10, 20 and 40 µM protein. Samples were spread on
electron microscopy grids and stained using 5% uranyl acetate. For details see
http://www.endocytosis.org/techniqs/techniqs.htm. For salt sensitivity experiments
endophilin N-BAR domain was bound at a concentration of 6 µM to Folch liposomes at a
concentration of 0.6 mg/ml. The ‘salt after’ samples were diluted to a concentration of
175 mM NaCl before ultracentrifugation which separates out the liposomes (which
pellet). For the ‘salt before’ samples dilution was performed after ultracentrifugation. For
the synaptojanin recruitment assays, synaptojanin and full-length endophilin were used at
a concentration of 3 µM with 0.2 mg/ml 400 nm Folch liposomes.
4.4.2. CD spectroscopy. Measurements were made using a CD6 Jobin Yvon CD
spectroscope. Proteins were diluted to a concentration of 0.2 mg/ml to give a buffer
concentration of 30 mM NaCl, 5 mM HEPES pH 7.4, 0.4 mM DTT. Readings were an
average of 5 scans from 190-260 nm. Buffer reading were subtracted and the data
180
smoothed using the moving average of 2.5 nm. Fractional helical content was calculated
as described previously (34). Ultracentrifugation methods have been described previously
(17).
4.4.3. Analytical Ultracentrifugation. Sedimentation equilibrium experiments were
performed in a Beckman Optima XL-A analytical ultracentrifuge with an An60-Ti rotor,
with 39, 78 and 156 µM in 150 mM NaCl, 20 mM HEPES pH 7.4, 1 mM TCEP.
Sedimentation was at 11,000 rev/min, 4.0 °C, with initial overspeeding at 18,000 rev/min
for 6 hr, to reduce the time to reach equilibrium (S9). Long sample columns were used.
Scans (averaging 10 readings) were taken at 280 nm at 24 hr intervals, until no movement
of the distribution was visible, when final scans (averaging 100 readings) were taken and
assumed to be operationally at equilibrium. The rotor was then accelerated to sediment
the macromolecule away from the meniscus, and further scans taken to provide initial
estimates of the baseline for each cell. Data were analyzed as described in detail in the
supplementary data of previous paper (35).
Sedimentation velocity runs were performed at the speeds indicated using a single
cell and taking scans as quickly as possible (~90sec intervals). All components were
tested at 40 and 150 µM protein in 150 mM NaCl, 20 mM HEPES pH 7.4, 2 mM DTT
unless otherwise stated. Data were analysed initially by plotting g(s*) against s* (where
g(s*) is the fraction of material sedimenting between s* and (s*+ds*)) using the DCDT+
software package (Version 1.05)(36). This software was also used for the direct fitting of
simple gaussian functions to dc/dt versus s curves to test for the number of components
and estimate their molecular mass. Partial specific volumes and solvent densities were
calculated as previously described in Peter et al. (17)(2004). For the plots above, every
181
tenth scan was taken to give visual separation of the traces. In these experiments we
calculate the partial specific volume but anything that broadens the boundary during
sedimentation (slight unfolding, microheterogeneity, a part of the protein that flips
around a bit etc) will give a larger estimate for the density (d) and as this is inversely
related to the apparent mass the result can be slightly smaller than the calculated
molecular weight.
4.4.4. Crystallography of endophilin N-BAR domain. Rat endophilin-A1 1-247 with
an N-terminal hexa-histidine tag was expressed in BL21 E.coli and purified on by Ni-
NTA (Qiagen) followed by Q-sepharose and gel filtration chromatography (Amersham
Biosciences). Selenomethionine rat endophilin-A1 1-247 crystals were obtained in
100mM Tris pH 8. 25% butane-1,4-diol and diffracted to 2.9Å on ID29, ESRF,
Grenoble, France. The structure was solved by 3-wavelength MAD (Table 4.1). Crystals
soaked in Ins(1,4,5)P3 and acetyl-CoA, because of binding to PtdIns(4,5)P2 liposomes
and a reported lysophosphatidic acid acyl transferase activity, gave no additional density.
Crystals belonged to spacegroup I4
1
with unit cell dimensions a = b = 126.5Å, c =
101.1Å. Three datasets were collected on beamline ID29 at ESRF on a single crystal
around the Se absorption edge, from the peak and inflection point of the fluorescence
curves, and at a "remote" high-energy point, in that order. At each wavelength, two
opposed 90° wedges were collected, and the initial wedge at the peak energy was
recollected at the end to show that there was no significant radiation damage. Intensities
were integrated to 2.9Å resolution using Mosflm (37) and scaled with Scala (38). Data
collection and refinement statistics are given below.
182
Of the eleven methionine residues in the sequence, seven Se sites were located
with the program Solve (39), and three more were located during phasing with Sharp
(40): the weakest of the ten sites was not on a Se-methionine residue, and two of the sites
represent alternative conformations of Met133. MAD phases were improved by solvent-
flattening with Solomon (41) using a solvent content of 77%. The model was built using
O (42), refined with Refmac (43) and the model was updated during refinement with
Coot (44). For refinement, data from the three wavelengths were merged together, to
improve the weak measurements at the high resolution limit. The final model included
residues 25 to 67 and 87 to 247, ie the hexahistidine tag and the N-terminal amphipathic
helix are disordered, as well as a loop in the middle of the first helix which contains two
methionine residues. Crystals soaked in Ins(1,4,5)P
3
and acetyl-CoA, because of binding
to PtdIns(4,5)P
2
liposomes and a reported lysophosphatidic acid acyl transferase activity,
gave no additional density.
4.4.5. Electron paramagnetic resonance (EPR). Spin labels were introduced onto each
residue from 2-16 of GST-tagged rat endophilin-A1 N-BAR domain. First we introduced
a cysteine residue at each position on a cysteine negative mutant (C108S). This native
cysteine is on the loop between helices 1 and 2 and the mutant functions normally on
liposome binding and tubulation assays (data not shown). Proteins were reacted in 150
mM NaCl, 20 mM HEPES pH 7.4 with an MTSL nitroxide spin label ([1-oxyl-2,2,5,5-
tetramethylpyrroline-3-methyl]-methanethiosulfonate) to generate the new side chain R1
using a previously described protocol (45). 2 µM protein was incubated with 1.4 mg/ml
183
Table 4.1. Data collection, phasing and refinement statistics for crystal structure
of rat endophilin A1 BAR domain.

Data collection statistics

Merged data
(outer shell)
Peak Inflection
Point
Remote
Space Group I4
1

Resolution (Å) 2.9 (3.06-2.9)
Completeness (%) 99.9 (100) 99.9(100) 99.9(100) 99.9(100)
Multiplicity 22.1 (22.3) 11.1 (11.2) 7.4 (7.4) 7.4 (7.4)
R
merge
0.155 (0.951) 0.130(0.942) 0.110(0.801) 0.124(0.891)
R
meas
(within I+/I-) 0.163 (0.966) 0.143(1.038) 0.128(0.934) 0.145(1.040)
R
p.i.m
(within I+/I-) 0.048 (0.209) 0.058 (0.436) 0.063 (0.482) 0.072 (0.537)
I/σ> outer shell
3.6 2.5 2.3
2.1
Anomalous completeness
99.7(99.8) 99.7(99.7) 99.7(99.8)
99.7(99.8)

Phasing

N Acentric

FOM Acentric

N Centric

FOM Centric

Overall

16913

0.474

596

0.368

Phasing powers Dispersive
Acentric

Centric
Anomalous
Acentric

Peak

1.047

1.021

1.725

Inflection point 1.237 1.251 0.69
Remote 0 0 0.85

Refinement statistics

Resolution Å (outer shell) 89 - 2.90 (2.97)
R
factor
(working set) 0.247 (0.36) Ramachandran plot:
R
free
0.266 (0.36) % in favoured region 95.5
R
free
test set size (%) 5.1 % outliers 0.5
B> (Å
2
) 69
N
reflections
16875
N
atoms
(non-hydrogen) 1653
R
msd
bond length (Å) 0.021
R
msd
bond angle (º) 0.940
R
merge
= ΣΣ
j
|I
hj
- I
h
>|/ΣΣ
j
I
hj

R
meas
= Σ√(n/(n-1))Σ
j
|I
hj
- I
h
>|/ΣΣ
j
I
hj
multiplicity-weighted R
merge

R
p.i.m
= Σ√(1/(n-1))Σ
j
|I
hj
- I
h
>|/ΣΣ
j
I
hj
precision indicating R
merge
(46-48)
184
liposomes and centrifuged to separate bound from unbound. Measurements were made
in the presence and absence of 400 nm Folch liposomes. EPR experiments were recorded
in a Bruker EMX spectrometer fitted with a dielectric resonator. The Bruker HS cavity
was used for endophilin labelled at position 227. Power saturation experiments to
determine the oxygen and NiEDDA accessibility (Π) were performed as previously
described (21, 45) with a 10 mM concentration of NiEDDA and oxygen as in air in
equilibrium with buffer. The Φ parameter was calculated by the relationship Φ = ln[Π
(O2)/ Π (NiEDDA)]. The immersion depth was calibrated using 1-palmitoyl-2-stearoyl-
(n-DOXYL)-sn-glycero-3-phosphocholine (Avanti Polar Lipids) as described previously
(21, 45). We obtained the following relationship between immersion depth (d) and Φ:
d[Å]= 6.3*Φ-3.9. This immersion depth represents the depth of the nitroxide moiety. For
peripherally bound helices, the lipid facing nitroxides are 7-10 Å deeper in the membrane
than the centre of the helix (45, 49). Based on Φ values of 1.5 to 2 obtained here, the
center of the helix is located near the level of the phosphates. Distances were determined
following previously established protocols (24) using a simulation program kindly
provided by Dr. Altenbach (UCLA). The simulations require a reference spectrum
without strong dipolar interactions, which was generated by adding three fold excess (6
µM) of unlabeled cysteine free protein. Very similar spin dilution results were obtained
using endophilin labelled at position 227 with a non-paramagnetic analogue of the spin
label (50) (data not shown).
4.4.6. FRET assay of membrane fusion. 250 µg/ml liposomes comprising 98:1:1
molar ratio of phosphatidylserine (PS), NBD-phosphatidylethanolamine (PE) and
rhodamine-PE were incubated for 5 min with a range of N-BAR and mutant domain
185
concentrations ± unlabeled liposomes (at 100% PS). The excitation wavelength was 450
nm and emission spectrum taken from 480-700 nm with slits of 5 nm using the
Fluoromax-2 fluorimeter (Jobin Yvon). 1% triton was added to obtain a value for donor
fluorescence in Figure 4.11C. In Figure 4.11D, 55 µM protein was used and the ratio of
emission peaks was taken within a single measurement to decrease the systematic error. 5
experiments and controls for each experiment were averaged separately and the
difference plotted ± SEM, except the KKK-EEE mutant (2 repeats). The major source of
error is differences between batches of liposomes.

186
CHAPTER 4
REFERENCES
1. Guichet A, et al. (2002) Essential role of endophilin A in synaptic vesicle budding
at the Drosophila neuromuscular junction. Embo J 21(7):1661-1672.

2. Rikhy R, Kumar V, Mittal R, & Krishnan KS (2002) Endophilin is critically
required for synapse formation and function in Drosophila melanogaster. J
Neurosci 22(17):7478-7484.

3. Schuske KR, et al. (2003) Endophilin is required for synaptic vesicle endocytosis
by localizing synaptojanin. Neuron 40(4):749-762.

4. Verstreken P, et al. (2002) Endophilin mutations block clathrin-mediated
endocytosis but not neurotransmitter release. Cell 109(1):101-112.

5. Gad H, et al. (2000) Fission and uncoating of synaptic clathrin-coated vesicles are
perturbed by disruption of interactions with the SH3 domain of endophilin.
Neuron 27(2):301-312.

6. Ringstad N, et al. (1999) Endophilin/SH3p4 is required for the transition from
early to late stages in clathrin-mediated synaptic vesicle endocytosis. Neuron
24(1):143-154.

7. Simpson F, et al. (1999) SH3-domain-containing proteins function at distinct
steps in clathrin-coated vesicle formation. Nat Cell Biol 1(2):119-124.

8. Dickman DK, Horne JA, Meinertzhagen IA, & Schwarz TL (2005) A slowed
classical pathway rather than kiss-and-run mediates endocytosis at synapses
lacking synaptojanin and endophilin. Cell 123(3):521-533.

9. Gallop JL, Butler PJ, & McMahon HT (2005) Endophilin and CtBP/BARS are
not acyl transferases in endocytosis or Golgi fission. Nature 438(7068):675-678.

10. Ford MG, et al. (2001) Simultaneous binding of PtdIns(4,5)P2 and clathrin by
AP180 in the nucleation of clathrin lattices on membranes. Science
291(5506):1051-1055.

11. Cremona O, et al. (1999) Essential role of phosphoinositide metabolism in
synaptic vesicle recycling. Cell 99(2):179-188.

187
12. Huttner WB & Schmidt A (2000) Lipids, lipid modification and lipid-protein
interaction in membrane budding and fission--insights from the roles of
endophilin A1 and synaptophysin in synaptic vesicle endocytosis. Curr Opin
Neurobiol 10(5):543-551.

13. Bauerfeind R, Takei K, & De Camilli P (1997) Amphiphysin I is associated with
coated endocytic intermediates and undergoes stimulation-dependent
dephosphorylation in nerve terminals. J Biol Chem 272(49):30984-30992.

14. Evergren E, et al. (2004) Amphiphysin is a component of clathrin coats formed
during synaptic vesicle recycling at the lamprey giant synapse. Traffic 5(7):514-
528.

15. Razzaq A, et al. (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(22):2967-2979.

16. McMahon HT & Gallop JL (2005) Membrane curvature and mechanisms of
dynamic cell membrane remodelling. Nature 438(7068):590-596.

17. Peter BJ, et al. (2004) BAR domains as sensors of membrane curvature: the
amphiphysin BAR structure. Science 303(5657):495-499.

18. Farsad K, et al. (2001) Generation of high curvature membranes mediated by
direct endophilin bilayer interactions. J Cell Biol 155(2):193-200.

19. Ford MG, et al. (2002) Curvature of clathrin-coated pits driven by epsin. Nature
419(6905):361-366.

20. Lee MC, et al. (2005) Sar1p N-terminal helix initiates membrane curvature and
completes the fission of a COPII vesicle. Cell 122(4):605-617.

21. 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 U S
A 91(5):1667-1671.

22. Weissenhorn W (2005) Crystal structure of the endophilin-A1 BAR domain. J
Mol Biol 351(3):653-661.

23. Chen Y, et al. (2003) Formation of an endophilin-Ca2+ channel complex is
critical for clathrin-mediated synaptic vesicle endocytosis. Cell 115(1):37-48.

188
24. Altenbach C, Oh KJ, Trabanino RJ, Hideg K, & Hubbell WL (2001) Estimation
of inter-residue distances in spin labeled proteins at physiological temperatures:
experimental strategies and practical limitations. Biochemistry 40(51):15471-
15482.

25. Whitmore L & Wallace BA (2004) DICHROWEB, an online server for protein
secondary structure analyses from circular dichroism spectroscopic data. Nucleic
Acids Res 32(Web Server issue):W668-673.

26. Rabenstein MD & Shin YK (1995) Determination of the distance between two
spin labels attached to a macromolecule. Proc Natl Acad Sci U S A 92(18):8239-
8243.

27. Struck DK, Hoekstra D, & Pagano RE (1981) Use of resonance energy transfer to
monitor membrane fusion. Biochemistry 20(14):4093-4099.

28. Stahelin RV, et al. (2003) Contrasting membrane interaction mechanisms of
AP180 N-terminal homology (ANTH) and epsin N-terminal homology (ENTH)
domains. J Biol Chem 278(31):28993-28999.

29. Zimmerberg J & McLaughlin S (2004) Membrane curvature: how BAR domains
bend bilayers. Curr Biol 14(6):R250-252.

30. Lee E, et al. (2002) Amphiphysin 2 (Bin1) and T-tubule biogenesis in muscle.
Science 297(5584):1193-1196.

31. Wigge P, et al. (1997) Amphiphysin heterodimers: potential role in clathrin-
mediated endocytosis. Mol Biol Cell 8(10):2003-2015.

32. Pierrat B, et al. (2001) SH3GLB, a new endophilin-related protein family
featuring an SH3 domain. Genomics 71(2):222-234.

33. Jockusch WJ, Praefcke GJ, McMahon HT, & Lagnado L (2005) Clathrin-
dependent and clathrin-independent retrieval of synaptic vesicles in retinal bipolar
cells. Neuron 46(6):869-878.

34. Chen YH, Yang JT, & Martinez HM (1972) Determination of the secondary
structures of proteins by circular dichroism and optical rotatory dispersion.
Biochemistry 11(22):4120-4131.

35. Mills IG, et al. (2003) EpsinR: an AP1/clathrin interacting protein involved in
vesicle trafficking. J Cell Biol 160(2):213-222.

189
36. Philo JS (2000) A method for directly fitting the time derivative of sedimentation
velocity data and an alternative algorithm for calculating sedimentation
coefficient distribution functions. Analyt. Biochem. 279(2):151-163.

37. Leslie AGW (1992) Recent changes to the MOSFLM package for processing film
and image plate data. in Joint CCP4 and ESF-EACMB Newsletter on Protein
Crystallography No. 26 (SERC, Daresbury Laboratory, Warrington, UK).

38. Collaborative Computational Project N (1994) The CCP4 suite: programs for
protein crystallography. Acta Crystallogr. D 50:760-763.

39. Terwilliger TC & Berendzen J (1999) Automated MAD and MIR structure
solution. Acta Crystallogr D 55(4):849-861.

40. de la Fortelle E & Bricogne G (1997) Maximum-likelihood heavy-atom parameter
refinement for multiple isomorphous replacement and multiwavelength
anomalous diffraction methods. Methods in Enzymology, eds Carter CW, Jr &
Sweet RM), Vol 276, pp 472-494.

41. Abrahams JP (1997) Bias reduction in phase refinement by modified interference
functions: introducing the γ function. Acta Cryst. D53:371-376.

42. Jones TA, Zou JY, Cowan SW, & Kjeldgaard M (1991) Improved methods for
building protein models in electron density maps and the location of errors in
these models. Acta crystallogr. A 47:110-119.

43. Murshudov GN, Vagin AA, & Dodson EJ (1997) Refinement of Macromolecular
Structures by the Maximum-Likelihood Method. Acta Cryst D53:240-255.

44. Emsley P & Cowtan K (2004) Coot: Model-Building Tools for Molecular
Graphics. Acta Crystallographica Section D - Biological Crystallography
60:2126-2132.

45. Jao CC, Der-Sarkissian A, Chen J, & Langen R (2004) Structure of membrane-
bound alpha-synuclein studied by site-directed spin labeling. Proc Natl Acad Sci
U S A 101(22):8331-8336.

46. Diederichs K & Karplus PA (1997) Improved R-factors for diffraction data
analysis in macromolecular crystallography. Nat Struct Biol 4(4):269-275.

47. Weiss MS & Hilgenfeld R (1997) On the use of the merging R factor as a quality
indicator for X-ray data. J.Appl.Cryst. 30:203-205.

48. Weiss MS (2001) Global indicators of X-ray data quality. J.Appl.Cryst. 34:130-
135.
190
49. Langen R, Oh KJ, Cascio D, & Hubbell WL (2000) Crystal structures of spin
labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra
in terms of structure. Biochemistry 39(29):8396-8405.

50. Gross A, Columbus L, Hideg K, Altenbach C, & Hubbell WL (1999) Structure of
the KcsA potassium channel from Streptomyces lividans: a site-directed spin
labeling study of the second transmembrane segment. Biochemistry
38(32):10324-10335.

191
CHAPTER 5
The role of amphipathic helices and the BAR domain in the membrane
curvature generation of endophilin

Christine C. Jao
1
, Balachandra G. Hegde
1
, Jennifer L. Gallop
2
, Prabhavati B. Hegde
1
,
Harvey T. McMahon
2
, Ian S. Haworth
3
and Ralf Langen*
1

1
Zilkha Neurogenetic Institute, University of Southern California, 1501 San Pablo Street,
Los Angeles, CA 90033
2
MRC Laboratory of Molecular Biology, Hills Road, Cambridge, CB2 2QH, UK.
3
Department of Pharmacology and Pharmaceutical Sciences, University of Southern
California School of Pharmacy, Los Angeles, CA 90033

*Corresponding Author

The manuscript draft (to be submitted) was formatted for use in this dissertation.
192
CHAPTER 5

ABSTRACT

Transient changes in membrane curvature are important in many cellular processes.
Previous studies suggest that the banana-shaped BAR domains, dimers composed of
coiled-coils with positively charged clusters located in the concave surface, drive the
generation of membrane curvature. Other studies indicate that amphipathic helices play a
role in membrane curvature generation by a ‘wedging’ mechanism. Earlier, we found
that the N-terminal region of rat endophilin A1 becomes an amphipathic helix in the
presence of membrane, and the center of the helix is at the level of the phosphate. We
now extend our analysis to the unstructured region found in the middle of helix 1 (helix 1
insert). This region, in a similar manner to the N-terminus, becomes an amphipathic
helix in the presence of membrane. Accessibility experiments show that the center of the
helix is at the level of the phosphate, while the concave surface of the BAR domain does
not penetrate the hydrocarbon layer of the bilayer. Continuous wave and 4-pulse DEER
experiments indicate that the helix 1 insert region is arranged in an anti-parallel manner,
and perpendicular to the BAR domain. A model for membrane curvature generation is
presented.

5.1. Introduction
Biological membranes are subject to constant remodeling, and the control of
membrane shape and curvature is essential for many vital cellular functions, such as cell
division and motility, endocytosis and vesicular trafficking (1). Recent work has
demonstrated that curvature-sensing and inducing proteins play important roles in these
193
membrane remodeling events and the molecular mechanisms by which they act have
become of significant interest.
Among the proteins thought to regulate membrane curvature in endocytosis are
epsin and N-BAR proteins, such as amphiphysin and endophilin (2-5). In vitro
experiments show that these proteins have a preference for highly curved membranes and
that they can induce the formation of highly curved vesicles (vesiculation) or narrow and
highly curved tubules. High resolution structural information has been obtained for the
aqueous forms of all of these proteins and it provides a convenient starting point for
studying the molecular mechanisms of membrane curvature induction. The crystal
structures of the BAR domains of endophilin and amphiphysin have the striking feature
that they represent curved, banana-shaped dimers (Figure 5.1A). Not only is the shape of
these dimers complementary to that of the curved membranes with which these proteins
interact, but their concave surface also has a high density of positively charged residues.
These charged residues are thought to favorably interact with negatively charged
membranes and simultaneous mutations of multiple such residues have been shown to
reduce membrane interaction (4, 5). Collectively, these data suggest that the concave
surface of the BAR domain plays an important role in membrane curvature generation
and that it might act as a rigid, positively charged scaffold (4-13).
Despite the importance of the BAR domain, it has become clear that additional
regions outside the BAR domain also play important roles in membrane curvature
induction. These regions, which include the N-terminal residues of amphiphysin and
endophilin, are unstructured in solution and, consequently are not resolved in the crystal
structure. Yet removal or mutations of these regions can abolish or inhibit liposome
194
binding activity and tubulation (4, 5). Using site-directed spin labeling and EPR
spectroscopy, we have recently shown that the endophilin N-terminus undergoes a
structural reorganization from an unfolded state in solution to an amphipathic helix (helix
0) that inserts into the membrane at the level of the headgroup (4). The inserted helix is
likely to act as a molecular wedge which creates a strain in the outer leaflet and thereby
promotes membrane curvature.
As in the case of endophilin and amphiphysin, the curvature generating epsin
ENTH domain has an unfolded N-terminus that, upon membrane interaction, undergoes a
conformational reorganization into an amphipathic helix. Again, mutagenesis data clearly
indicate that this region is essential for membrane binding and bending (3). In contrast to
BAR domains, however, the crystal structure of the epsin ENTH domain revealed a
block-like rather than a banana-like structure. Thus, it is not as clear whether scaffolding
occurs in the case of epsin; rather the insertion of the amphipathic helix 0 may be the
main driving force for membrane curvature induction. Another indication that protein
scaffolding may not be a necessary requirement for membrane tubule formation comes
from work on small peptides as well as studies on lipids. It has been shown that
tubulation and vesicle deformation can be induced by small helical peptides and even
occur in the absence of proteins, in response to induction of transbilayer area asymmetry
(14-16).
Here, we investigated the roles of helix insertion and scaffolding by examining
the structure of rat endophilin A1 upon vesiculation. Previous work on annexins had
shown that curvature dependent membrane interactions can lead to major conformational
reorganizations (17). Thus, our first goal was to test whether the structure of the
195
endophilin BAR domain is retained upon membrane interaction and, if so, whether its
concave surface interacts with the membrane. In addition, we investigated the structure of
a central loop region that is disordered in the crystal structure, but that had previously
been proposed to convert into an amphiphatic helix and insert into membranes. The
experimental work was performed using site-directed spin labeling and EPR
spectroscopy.

5.2. Materials and Methods
5.2.1 Preparation of spin-labeled rat endophilin derivatives. Single and double
cysteine mutants of rat endophilin were expressed and purified as previously described
(4). Briefly, the GST-tagged protein was expressed using BL21 (DE3) pLys S bacterial
cells. GST-tagged proteins were purified using glutathione beads. Endophilin was
released from GST by thrombin cleavage, followed by anion exchange chromatography.
Proteins were reacted with 5x molar excess spin label, and unreacted spin label was
separated using PD10 columns.
5.2.2. Liposome preparation. Folch fraction I, type I was purchased from Sigma-
Aldrich. Lipids were dried with nitrogen and dessicated overnight. They were
resuspended in 20mM Hepes pH7.4, 150mM NaCl, treated to brief bath sonication,
extruded through 400nm pore filters using the Avanti mini-extruder, and used.
5.2.3 Continuous wave EPR experiments. Sample preparation and measurement have
previously been described (4). Briefly, 2uM spin-labeled protein is combined with
1.4mg/ml Folch liposomes and incubated at RT for 20 minutes. After incubation,
unbound proteins are separated from membrane-bound by high speed centrifugation at
196
152,800 xg for 20 minutes at 22
O
C. The pellet, or membrane-bound portion, is measured
by EPR. EPR spectra are also obtained for samples in the absence of liposomes. EPR
spectra are collected from a Bruker EMX spectrometer fitted with a dielectric resonator at
1.59 mW incident microwave power using field modulation of 1.5 Gauss (for soluble
samples) or 3.0 Gauss (for membrane-bound samples) and scan width of 100 Gauss.
Power saturation experiments to determine O
2
and NiEDDA accessibility ( П) have
previously been described (18). Accessibility data were obtained using oxygen from air
in equilibrium with buffer, and 10mM NiEDDA. The Φ parameter was calculated by the
relationship Φ =ln[ П (O2)/ П(NiEDDA)]. The immersion depth was calibrated using 1-
palmitoyl-2-stearoyl-(n-DOXYL)-sn-glycero-3-phosphocholine (Avanti Polar Lipids) as
described previously (18). We obtained the following relation between immersion depth
(d) and Φ: d[Å]= 6.3*Φ-3.9. This immersion depth represents the depth of the nitroxide
moiety. For residue 63, 8uM protein (25% spin labeled) and 1.4mg/ml Folch liposomes
were used for nitroxide scans and power saturation experiments. Distance determination
was performed using deconvolution methods (19, 20) as implemented by Altenbach et al.
(19) using a program developed by Dr. Christian Altenbach.
5.2.4. Pulsed EPR and distance analysis. Samples were prepared by combining 4uM
protein and 1.4mg/ml Folch liposomes. After incubation for 20 minutes at RT, samples
were centrifuged at 152,800 xg for 20 minutes, and the pellet recovered for measurement.
The pellet was resuspended to a final volume of 10ul, and an equal volume of 60%
sucrose was added, mixed, and loaded onto quartz capillaries. Samples were flash
frozen, and data acquired at 78K. DEER experiments were performed using a Bruker
Elexsys E580 X-band pulse EPR spectrometer fitted with a 3-mm split ring (MS-3)
197
resonator, a continuous flow helium cryostat (CF935, Oxford Instruments), and a
temperature controller (ITC503S, Oxford Instruments). Data were fit using Gaussian
distribution, as implemented in DEERAnalysis2008 package (21).

5.3. Results
5.3.1. Rat endophilin A1 BAR is a dimer upon membrane interaction
In order to investigate whether the crystal-like dimer of endophilin is retained upon
membrane interaction, we generated a series of spin labeled endophilin derivatives
(Figure 5.1A) and incubated them with liposomes comprised of Folch lipids. The
experimental conditions were chosen such that the Folch liposomes were predominantly
vesiculated into highly curved vesicles with a diameter of ~250 Å as determined by
electron microscopy (data not shown). We then measured intra-subunit as well as intra-
dimer distances and compared them to the respective α-carbon distances of the crystal
structure. The distances measured by EPR are those between the nitroxide moieties of R1
and, therefore, they would be expected to be similar but not identical to the α-carbon
distances. Distances >20 Å were investigated using 4-pulse double electron-electron
resonance (DEER) while shorter distances were investigated using continuous wave EPR
(CW-EPR) (see Materials and Methods).
The DEER measurements are illustrated with the example of membrane-bound
endophilin spin labeled in helix 3 at position 216 (Figure 5.1B). A pronounced periodic
oscillation is observed for this 216R1 derivative and frequency analysis of this oscillation
yields a well defined distance of 37 Å (Figure 5.1C). This distance is in close agreement
198

Figure 5.1. The crystal-like dimer structure is retained upon membrane interaction.
(A) Crystal structure of rat endophilin A1 (PDB ID 2C08) indicating the positions of the
spin labeled sites. The individual subunits are shown in green and red, respectively. The
disordered insert in helix 1 not resolved in the crystal structure is schematically illustrated
by the dashed lines. (B) Example of 4-pulse DEER data obtained for the 216R1
derivative in membrane-bound form. The baseline subtracted data are shown in black,
while the Gaussian fit is shown in red. The distance distribution corresponding to the
observed oscillation is given in (C).
199
with the respective intra-dimer α-carbon distance in the crystal structure of endophilin
(Table 5.1). Similarly, the intra-dimer distances of membrane-bound 96R1 (in helix 1b),
63R1 (in helix 1 insert) as well as the previously reported (4) distance for 227R1 (in helix
3) are in good agreement with the crystal structure (Figures 5.1A, 5.2). As a control, we
also measured the intra-dimer distance for one of the derivatives (96R1) in solution and
obtained the same distance as for the membrane-bound form.
The tip–to-tip distance of the endophilin dimer in the crystal is nearly 130 Å. Since
this range is beyond the detection limit for intra-dimer contacts, we employed a different
approach and measured distance between 2 sites in the same subunit. As shown in Table
5.1 (and Figure 5.2), the distance of 31Å between 178R1 and 200R1 (in helix 2 and 3,
respectively) is in good agreement with the α-carbon distances of the crystal structure
(33Å). Collectively, this distance analysis supports the notion that the overall structure of
the BAR dimer is largely retained upon membrane interaction.
5.3.2. The concave surface of the BAR domain does not penetrate the membrane
Next, we used mobility and accessibility analysis of selected spin labeled derivatives
to investigate the mechanisms by which the BAR domain interacts with the membrane.
Spin labels were introduced at positions that were either located on the concave surface
of the BAR domain (position 96, 159, and 166) or that were located at positions expected
to face away from the membrane (positions 108, 172 and 247) (Figure 5.3A).
The EPR spectra for those derivatives in the soluble (black trace, Fig. 5.3B,C) and
membrane-bound forms (red trace, Fig. 5.3B,C) are qualitatively similar, further
indicating the lack of major conformational changes. For example, the EPR spectra for
247R1 (Fig. 5.3C) are characterized by three sharp and narrowly spaced lines in both
200
Table 5.1. EPR Distances. Distances obtained from continuous wave (CW) or 4-pulsed
DEER experiments (DEER) which were fit using Gaussian models (Figure 5.2). All
distances were obtained for indicated endophilin derivatives bound to membrane. The
experimental distances are given as the peak of the Gaussian distributions and are in
excellent agreement with distances expected from the crystal structure.

EPR crystal
mutant distance (Å) structure (Å)
227R1 9 (CW) 8
63R1 9 (CW) 11
96R1 38 (DEER) 36
216R1 37 (DEER) 34
178R1-200R1 31 DEER 33
201

Figure 5.2. DEER and CW-EPR data for membrane-bound endophilin derivatives.
(A) 4-pulse DEER data obtained for membrane-bound endophilin derivatives. The
baseline subtracted data are shown in black, while the Gaussian fit is shown in red. The
distance distribution corresponding to the observed oscillation is given in the right
column. (B) Continuous wave data for endophilin residue A63 bound to membrane. The
EPR spectrum of the fully labeled derivative (red) indicates spin-spin interaction. This
spectrum is seen when spin labels are in close proximity to each other. When the spin
label is diluted (black), spin-spin interaction is relieved. Deconvolution by Pake pattern
analysis (blue) of the spectra resulted in a distance of ~9Å between labels, and quality of
fit also assessed.
202
Figure 5.3. EPR spectra and depth parameter measurement of selected sites on the
concave face of the endophilin BAR domain indicate that the concave surface does
not penetrate the hydrocarbon layer of the membrane. (A) Locations of the spin
labeled sites tested by EPR in the crystal structure of endophilin. Continuous wave X-
band EPR spectra of soluble (black) and membrane-bound (red) endophilin derivatives
spin labeled positions on the concave surface (B) and facing away from the concave
surface (C). The similarity between the EPR spectra for the soluble and membrane-
bound forms suggests structural similarity. Only endophilin labeled at position 166
experiences more pronounced ordering. The depth parameter (Φ) values for all
derivatives in membrane-bound form are strongly negative (-1) indicating that the
respective spin labeled sites are exposed to an aqueous environment and do not
significantly penetrate into the hydrocarbon layer of the membrane. Scan width is 100
Gauss.
203
Figure 5.3. continued

Figure 5.3. EPR spectra and depth parameter measurement of selected sites on the
concave face of the endophilin BAR domain indicate that the concave surface does
not penetrate the hydrocarbon layer of the membrane.
204
conditions. The high motion indicated by these spectra is consistent with the location of
this site at the C-terminal end of a helix that is likely to be frayed in solution. Also, the
spectra for 96R1 and 172R1 exhibit rather similar spectra for the soluble and membrane-
bound forms (Figure 5.3B,C). We note, however, that some membrane-binding induced
immobilization can clearly be detected for 159R1 and 166R1. The immobilization
observed at those sites could be caused by a direct interaction with the lipids or other
regions of the proteins, such as the N-terminus that becomes ordered under these
conditions. In order to investigate the proximity to the membrane bilayer, we employed
accessibility to the paramagnetic colliders, O
2
and NiEDDA. While the concentration of
the nonpolar O
2
increases in the membrane, the more polar NiEDDA preferentially
partitions into the solvent. Thus, the deeper a spin labeled site penetrates into the acyl
chain region, the more it will become accessible to O
2
and, conversely, inaccessible to
NiEDDA. The accessibilities to O
2
and NiEDDA (ΠO
2
and ΠNiEDDA) are conveniently
summarized by the depth parameter Φ (Φ=ln(ΠO
2
/ΠNiEDDA). Interestingly, all Φ
values were in the range between -1.2 and -2.1 (Figure 5.3B,C) regardless of whether the
labeled sites are on the concave surface or not. Thus, analysis of the depth parameter Φ
does not provide a clear indication of membrane penetration of residues on the concave
surface of the BAR (also see below).
5.3.3. The insert region becomes an amphipathic helix upon membrane interaction
To investigate potential membrane-induced conformational changes in the helix insert
region, we conducted a nitroxide scanning experiment. The EPR spectra of the soluble
derivatives (Figure 5.4, black) are generally of higher amplitude with sharper spectral
lines than those from the respective membrane-bound forms (Figure 5.4, red). In
205

Figure 5.4. EPR spectra for residues located in the helix 1 insert indicate ordering of
the region in the presence of membrane. Spectra for soluble (black) and membrane-
bound (red) endophilin derivatives are shown. The line broadening and loss of spectral
amplitude of all sites reflects, to a significant extent, a reduction in mobility and ordering
that occurs upon membrane interaction. Residues 70 through 77 are highly disordered in
solution while, while residues 64 to 69 are more ordered. Immobile components can be
detected at position 64 and 68 in the soluble as well as the membrane-bound forms. This
immobilization is likely due to tertiary contacts with the BAR domain. Scan width is 100
Gauss.
206
particular, the line shapes of 70R1 to 77R1 indicate a largely disordered and unfolded
structure in solution while a much more ordered structure is indicated when bound to the
membrane. Only residues 64 to 69 are ordered in solution and pronounced
immobilization can be seen for 64R1, 65R1, and 68R1, and 69R1. Such immobilization is
likely due to tertiary contacts. Although this region was disordered in the crystal structure
of rat endophilin, it was resolved in that of human endophilin where residues 62 to 71
were found to form a short helical segment. Moreover, residues 64, 65, 68, and 69 are
pointing directly toward the BAR domain in this structure. Thus, the EPR spectra in the
absence of membranes are consistent with a structure akin to that in the crystal (also see
below). Residues 64-69 are also the most ordered region of the membrane-bound form,
but in the presence of membranes the ordered region becomes much more extended.
To obtain more detailed secondary structure information in the membrane-bound
form, we performed O
2
and NiEDDA accessibility measurements for 63R1 to 77R1. As
summarized with the depth parameter Φ (Figure 5.5A), the accessibility data give rise to
a periodic oscillation. The local Φ-maxima (residues 63, 66, 70, 73 and 74, magenta)
represent preferentially O
2
accessible, lipid-exposed sites and these positions cluster on
the hydrophobic portion of a helical wheel (Figure 5.5B). Conversely, the local minima
(green) correspond to solvent-exposed sites that fall onto the opposite face of the helical
wheel. The formation of an amphipathic α-helical structure is further supported by
comparison of the Φ-data with a cosine curve that has the standard periodicity of an α-
helix (3.6 amino acids/turn). This curve agrees well with the experimentally observed
periodicity from residues 63 to 75. The only deviation from this helical periodicity
occurs for the region containing residues 76 and 77. For these residues an increase in the
207

Figure 5.5. Depth parameter Φ plotted as function of residue number. The
accessibility of membrane-bound endophilin derivatives to NiEDDA and O
2
(ΠNiEDDA
and ΠO
2
) were determined using power saturation and Φ was calculated according to Φ
=ln(ΠO
2
/(ΠNiEDDA). When plotted as function of labeling position, Φ exhibits a
periodic oscillation corresponding to that of an alpha-helix as illustrated with the
sinusoidal line which has a periodicity of 3.6 amino acids/turn (A). Membrane exposed
sites (red maxima) fall onto one face of a helical wheel while the solvent-exposed sites
(green minima) fall onto the opposite face of a helical wheel (B). The immersion depth of
the lipid exposed sites 63, 66, and 70R1 is on average around 6 Å placing the center of
the helix close to the level of the phosphates (about 1 to 4 Å above). The immersion
depth for residues 73 and 74 is lower. This lowered immersion depth is likely to be
predominantly due to the fact that the latter positions are not facing directly downward
into the membrane but more sideways toward the headgroups. In addition, a local helix
tilt could further contribute to the reduced immersion.
208
Φ-values would be predicted, but the opposite is observed. Thus, these residues may no
longer be part of the helix.
Previous studies demonstrated that Φ values are proportional to the membrane
immersion depth of the R1 side chain, and that this depth can be calibrated using spin-
labeled derivatives of phospholipids (18). Based on such a calibration (Materials and
Methods), we obtain average immersion depths of the nitroxide moieties of 63R1, 66R1,
and 70R1 that are on the order of 5 to 6 Å. Previous work has shown that an R1
sidechain is typically at a distance of 7-10 Å from the center of an α-helix to which it is
attached (22). Thus, we can estimate that the center of the helix is located about 1 to 4 Å
above the phosphate layer. The Φ-values for residues 73 and 74 are somewhat lower. A
reduced immersion depth of those sites would be consistent with a slight helix tilt.
However, residues 73 and 74 are also located on the periphery of the hydrophobic face
(Fig. 5.5B) where the side chains might project laterally rather than directly toward the
bilayer interior.
5.3.4. The dimeric amphipathic helices are anti-parallel, and perpendicular to the
BAR domain in response to membrane interaction
To analyze the orientation of the amphipathic helices with respect to each other, we
measured the distances between same sites across the dimer using DEER (Figure 5.6). As
shown in Table 5.2, the distance between these sites generally increases with increasing
residue number, indicating that the helices must be facing away from each other. This
feature is best illustrated with the lipid-exposed sites 63R1, 66R1, and 70R1. For these
sites, we can assume that the nitroxide side chains are facing in the same direction (i.e.

209

Figure 5.6. Four-pulse DEER data obtained for membrane-bound endophilin
derivatives. The baseline subtracted data are shown in black, while the Gaussian fit is
shown in red. The distance distribution corresponding to the observed oscillation is also
given.

210
Table 5.2. Intramolecular distances between same sites in the insert region of
membrane-bound endophilin. Distances were obtained by 4-pulse DEER (Figure 5.6)
or continuous wave EPR (Figure 5.2). The experimental distances shown in the table
correspond to the intermolecular distances between same sites within the dimer. The
distances increase with increasing residue number indicating an anti-parallel arrangement
of the helices.

211
into the membrane) and the inter label distances are, therefore, likely to be close to those
of the respective α-carbons. An interesting feature of the respective distances is that they
increase by about 10Å for each helical turn. This increase is in good agreement with
anti-parallel helices considering that an individual helix increases by ~5Å/turn. Also the
other distances listed in Table 5.2 are in agreement with a helical structure.
Inasmuch as Masuda and colleagues (12) observed a short anti-parallel helical
structure for residues 63-71 in the crystal structure of human endophilin A1, we sought to
determine whether the distances in solution are similar to those in the membrane-bound
form. As shown in Table 5.3 (and Figure 5.7), the intra-dimer distances for endophilin
samples 63R1, 67R1, and 70R1 in the absence of membranes are significantly different
from those observed in the membrane-bound state. When compared to the α-carbon
distances from the crystal structure, the measured distances are in good agreement. These
data suggest that the structure of the insert taken up in the crystal of human endophilin
A1 is similar to that of rat endophilin A1 in solution and that the structure taken up in the
membrane-bound state is likely to be different.

5.4. Discussion
In the present study, we investigated the structure of endophilin when bound to
predominantly small and highly curved vesicles formed by endophilin-dependent
vesiculation. Under these conditions, the BAR domain of endophilin retains a dimeric
structure similar to that found in the crystal demonstrating that, unlike the case of annexin
B12 (17) , endophilin does not undergo major conformational refolding. Moreover, we
find that a central loop region (residues 63-75), becomes helical and inserts into the
212
Table 5.3. Comparison of data obtained by EPR of membrane-bound and soluble
intra-dimer distances with distances determined by crystallography.

EPR EPR
distance (Å) distance (Å) crystal
mutant membrane solution structure (Å)
63R1 9 (CW) 9 (CW) 12.1
66R1 19 (DEER) 24 (DEER) 22.99
70R1 28 (DEER) 35 (DEER) 35.23

213

Figure 5.7. DEER and CW EPR data for membrane-bound endophilin derivatives
located in the insert region. (A) 4-pulse DEER data obtained for soluble endophilin
derivatives. The baseline subtracted data are shown in black, while the Gaussian fit is
shown in red. The distance distribution corresponding to the observed oscillation is also
given. (B) Continuous wave data for endophilin residue 63R1 in the soluble form. The
EPR spectrum of the fully labeled derivative (red) indicates spin-spin interaction. This
spectrum is seen when spin labels are in close proximity to each other. When the spin
label is diluted (black), spin-spin interaction is relieved. Deconvolution by Pake pattern
analysis (blue) of the spectra resulted in a distance of ~9Å between labels, and quality of
fit also assessed.
214
membrane at an immersion depth that is comparable (albeit slightly shallower) than that
of the N- terminal helix (4). Two mechanisms have been suggested by which insertion of
amphipathic helices can induce curvature. By inserting helices into only the outer leaflet
of the bilayer, an imbalance is created between the two leaflets and the tighter packing in
the outer leaflet causes an expansion which, in turn, leads to a bending of the membrane.
In addition, insertion of helices at the level of the headgroups could induce additional
curvature strains by preferentially increasing the packing density in the headgroup rather
than the acyl chain region (23). In the case of membrane-inserted amphipathic helices,
these mechanisms are expected to promote membrane curvature in a direction
perpendicular to that of the helix axis. Thus, the anti-parallel insert helices are ideally
positioned to induce curvature along the convex surface of the BAR domain. While this
geometry is likely to be energetically favorable, it should be noted that the distance
distributions obtained for the cross-dimer distances in the insert region are relatively
broad when compared to those obtained from sites in the BAR domain. This could
indicate more structural flexibility in this region when compared to the much more rigid
BAR domain.
In fact, we observe a different structure in solution that is likely to be similar to
that observed in the crystals of human endophilin A1. In this crystal structure, the insert
helices are not perpendicular to the BAR domain but project at angle of about 60
0
. Thus,
it is likely that the helices might be able to take on a range of conformations and that the
precise structure and orientation of this helix might be able to adjust to different
membrane and curvature conditions.
215
Based on our EPR data, we place both the N-terminal and insert helices in a direction
perpendicular to the BAR domain (Figure 5.8). The rationale for such placement was
based on the idea that the “wedging effect” of the helix would create membrane curvature
along the direction of the curved BAR domain. Thus, the BAR domain and the helix
would induce curvature in the same direction. Moreover, if the helix is located beneath
the BAR domain, the BAR domain and the amphipathic helix could jointly coordinate the
same lipid headgroups. Future studies include the use of EPR data as constraints in
structural refinement to generate atomistic models.
Somewhat surprisingly, the BAR domain appears to be at a distance from the
membrane and, at least the specific sites tested, are not close enough to the bilayer
interior to allow significant membrane insertion of the spin labeled side chains. None of
the three sites tested on the concave surface of the BAR domain exhibited Φ values that
are significantly different from those at solvent-exposed control sites. Furthermore, the
overall values are similar to those obtained for sites on the solvent-exposed surface of the
insert helix. Thus, the nitroxide moiety of the spin labeled side chains must be well above
the level of the phosphate groups and we can conclude that the BAR domain is likely to
more peripherally interact with the distal regions of the headgroup. This finding is
unexpected given a recent cryo-electron microscopy study on F-BAR which provided
clear evidence for a much more direct membrane interaction and a more pronounced
scaffolding effect of that protein. However, it should be noted that the electron
microscopy study investigated tubules while the present study was performed under
predominantly vesiculating conditions. It may well be possible that endophilin as well as
other curvature-inducing proteins might be able to take up different structures and
216

Figure 5.8. Model of the insert region in membrane-bound form. Panel A shows the
orientation of the insert region with respect to the BAR domain. Panel B schematically
illustrates the location of the insert region with respect to the membrane.
217
orientations with respect to the membrane to induce different types of curvatures. Future
experiments are needed to further evaluate this question.

218
CHAPTER 5

REFERENCES

1. McMahon HT & Gallop JL (2005) Membrane curvature and mechanisms of
dynamic cell membrane remodelling. Nature 438(7068):590-596.

2. Farsad K, et al. (2001) Generation of high curvature membranes mediated by
direct endophilin bilayer interactions. J Cell Biol 155(2):193-200.

3. Ford MG, et al. (2002) Curvature of clathrin-coated pits driven by epsin. Nature
419(6905):361-366.

4. Gallop JL, et al. (2006) Mechanism of endophilin N-BAR domain-mediated
membrane curvature. Embo J 25(12):2898-2910.

5. Peter BJ, et al. (2004) BAR domains as sensors of membrane curvature: the
amphiphysin BAR structure. Science 303(5657):495-499.

6. Arkhipov A, Yin Y, & Schulten K (2008) Four-scale description of membrane
sculpting by BAR domains. Biophys J 95(6):2806-2821.

7. Ayton GS, Blood PD, & Voth GA (2007) Membrane remodeling from N-BAR
domain interactions: insights from multi-scale simulation. Biophys J 92(10):3595-
3602.

8. Ayton GS, et al. (2009) New insights into BAR domain-induced membrane
remodeling. Biophys J 97(6):1616-1625.

9. Blood PD & Voth GA (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(41):15068-15072.

10. Frost A, De Camilli P, & Unger VM (2007) F-BAR proteins join the BAR family
fold. Structure 15(7):751-753.

11. Henne WM, et al. (2007) Structure and analysis of FCHo2 F-BAR domain: a
dimerizing and membrane recruitment module that effects membrane curvature.
Structure 15(7):839-852.

12. Masuda M, et al. (2006) Endophilin BAR domain drives membrane curvature by
two newly identified structure-based mechanisms. Embo J 25(12):2889-2897.

13. Yin Y, Arkhipov A, & Schulten K (2009) Simulations of membrane tubulation by
lattices of amphiphysin N-BAR domains. Structure 17(6):882-892.
219
14. Mui BL, Dobereiner HG, Madden TD, & Cullis PR (1995) Influence of
transbilayer area asymmetry on the morphology of large unilamellar vesicles.
Biophys J 69(3):930-941.

15. Furuya T, et al. (2003) Nanotubules formed by highly hydrophobic amphiphilic
alpha-helical peptides and natural phospholipids. Biophys J 84(3):1950-1959.
16. Lee S, et al. (2001) De novo-designed peptide transforms Golgi-specific lipids
into Golgi-like nanotubules. J Biol Chem 276(44):41224-41228.

17. Fischer T, Lu L, Haigler HT, & Langen R (2007) Annexin B12 is a sensor of
membrane curvature and undergoes major curvature-dependent structural
changes. J Biol Chem 282(13):9996-10004.

18. 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 U S
A 91(5):1667-1671.

19. Altenbach C, Oh KJ, Trabanino RJ, Hideg K, & Hubbell WL (2001) Estimation
of inter-residue distances in spin labeled proteins at physiological temperatures:
experimental strategies and practical limitations. Biochemistry 40(51):15471-
15482.

20. Rabenstein MD & Shin YK (1995) Determination of the distance between two
spin labels attached to a macromolecule. Proc Natl Acad Sci U S A 92(18):8239-
8243.

21. Jeschke G, et al. (2006) DeerAnalysis2006 - a Comprehensive Software Package
for Analyzing Pulsed ELDOR Data. Appl Magn Reson 30:473-498.

22. Langen R, Oh KJ, Cascio D, & Hubbell WL (2000) Crystal structures of spin
labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra
in terms of structure. Biochemistry 39(29):8396-8405.

23. Campelo F, McMahon HT, & Kozlov MM (2008) The hydrophobic insertion
mechanism of membrane curvature generation by proteins. Biophys J 95(5):2325-
2339.

220
CONCLUDING REMARKS
The generation of membrane curvature is an important process in cell biology.
Curved membranes are evident during processes like organelle biogenesis and vesicle
trafficking. Proteins which orchestrate these processes belong to two groups: membrane
curvature sensors and membrane curvature inducers. The focus of my project is to
analyze the molecular mechanisms involved in membrane curvature sensing and
induction.
The amphipathic helix motif is important in membrane curvature regulation. We
found this motif in the membrane-bound forms of α-synuclein, epsin, and endophilin.
For all of them, unstructured regions become structured in the presence of membranes,
accompanied by the formation of amphipathic helices which insert into the membrane at
the level of the headgroup. For α-synuclein, a curvature sensor, this helix location is
likely energetically favorable to compensate for the curvature strain inherent in highly
curved vesicles, and provides support to the notion that the function of α-synuclein in the
cell is to stabilize synaptic vesicles.
The crystal structure of endophilin BAR domain suggested a scaffolding mechanism
in which the concave surface interacts electrostatically with the membrane to induce
curvature. We found that the concave surface does not deeply penetrate the membrane,
contrary to the expected result based on the crystal structure. We found that the
unstructured regions (the N-terminus and a region in the middle of helix 1), transform to
amphipathic helices in the presence of membrane and the centers of these helices are at
the level of the phosphate headgroup. These helices are well-positioned to act as wedges,
pushing apart the headgroups, and thereby inducing membrane curvature. Studies on
221
lipids as well as small peptides where transbilayer asymmetry was observed are in
support of our findings.
In a similar fashion, analysis of the membrane-bound form of epsin was performed.
The structure of the ENTH domain is ‘block-like’, unlike the BAR domain, and does not
immediately suggest a mechanisms for membrane curvature generation. We found that
the N-terminal region is an amphipathic helix in the presence of membranes, and the
center of the helix is at the level of the phosphate. The induction of membrane curvature
likely is mediated through the wedging mechanism.
Thus far, our work has been on the structure of vesicle-bound proteins. Is the
structure on the vesicle different from the structure bound to the tube? Simulation studies
on membrane curvature generation suggest that the density of protein on the membrane
plays a role in curvature generation. We compared the structure of epsin ENTH in the
presence of vesicles, and in the presence of pre-made tubes. We found that the vesicle-
bound form is largely monomeric, while the tube-bound form is likely at least a dimer.
The collective action of amphipathic helices likely mediate membrane curvature
generation. In future, it will be important to test this notion for endophilin and other
BAR-containing proteins by analyzing the structure of the vesicle-bound versus the tube-
bound form.
What are the characteristics of an amphipathic helix that is an inducer versus an
amphipathic helix that is a sensor? Studies on ArfGAP proteins reveal that a lipid
packing sensor is an amphipathic helix with a polar face rich in serines and threonines.
In future, experiments to transform an inducer to a sensor, and a sensor to an inducer by
222
changing the characteristics of the amphipathic helix will be undertaken. In this way, we
can define an inducer or sensor helix.
Recent work in the lab has shown that lipid composition is an important factor to
consider in studying proteins able to sense and/or induce curvature. Lipids are able to
effect structural changes in proteins, as was seen in work with the curvature sensitive
form of annexin. Structural determination of these proteins when bound to different
types of lipids, as well as different lipid compositions, will be important to perform.
Recent work on binding of amphipathic helices to membranes suggest that lipid packing
defects are important to consider. In addition, our EPR work will be combined with
structural refinement to generate atomistic models.
Our work shows the importance of amphipathic helices. In both sensors and inducers,
amphipathic helices insert at the level of the headgroup phosphates. Whereas curvature-
sensing proteins insert their helices only when headgroup packing densities are low,
curvature-inducing proteins have a much greater driving force for membrane interaction
and can push headgroups apart, in effect acting as wedges. Our studies on the
membrane-bound form of membrane curvature sensors and inducers will assist in the
elucidation of the mechanism of membrane curvature regulation.

223
BIBLIOGRAPHY

Abeliovich, A., Y. Schmitz, I. Farinas, D. Choi-Lundberg, W.H. Ho, P.E. Castillo, N.
Shinsky, J.M. Verdugo, M. Armanini, A. Ryan, M. Hynes, H. Phillips, D. Sulzer,
and A. Rosenthal. 2000. Mice lacking alpha-synuclein display functional deficits
in the nigrostriatal dopamine system. Neuron 25(1):239-252.

Abrahams, J.P. 1997. Bias reduction in phase refinement by modified interference
functions: introducing the gamma correction. Acta Crystallogr D Biol Crystallogr
53(Pt 4):371-376.

Aisenbrey, C., T. Borowik, R. Bystrom, M. Bokvist, F. Lindstrom, H. Misiak, M.A. Sani,
and G. Grobner. 2008. How is protein aggregation in amyloidogenic diseases
modulated by biological membranes? Eur Biophys J 37(3):247-255.

Altenbach, C., D.A. Greenhalgh, H.G. Khorana, and W.L. Hubbell. 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(5):1667-1671.

Altenbach, C., A.K. Kusnetzow, O.P. Ernst, K.P. Hofmann, and W.L. Hubbell. 2008.
High-resolution distance mapping in rhodopsin reveals the pattern of helix
movement due to activation. Proc Natl Acad Sci U S A 105(21):7439-7444.

Altenbach, C., T. Marti, H.G. Khorana, and W.L. Hubbell. 1990. Transmembrane protein
structure: spin labeling of bacteriorhodopsin mutants. Science 248(4959):1088-
1092.

Altenbach, C., K.J. Oh, R.J. Trabanino, K. Hideg, and W.L. Hubbell. 2001. Estimation of
inter-residue distances in spin labeled proteins at physiological temperatures:
experimental strategies and practical limitations. Biochemistry 40(51):15471-
15482.

Antonny, B., I. Huber, S. Paris, M. Chabre, and D. Cassel. 1997. Activation of ADP-
ribosylation factor 1 GTPase-activating protein by phosphatidylcholine-derived
diacylglycerols. J Biol Chem 272(49):30848-30851.

Arkhipov, A., Y. Yin, and K. Schulten. 2008. Four-scale description of membrane
sculpting by BAR domains. Biophys J 95(6):2806-2821.

Ayton, G.S., P.D. Blood, and G.A. Voth. 2007. Membrane remodeling from N-BAR
domain interactions: insights from multi-scale simulation. Biophys J 92(10):3595-
3602.
224
Ayton, G.S., E. Lyman, V. Krishna, R.D. Swenson, C. Mim, V.M. Unger, and G.A. Voth.
2009. New insights into BAR domain-induced membrane remodeling. Biophys J
97(6):1616-1625.

Ayton, G.S., E. Lyman, and G.A. Voth. in press. Hierarchial coarse-graining strategy for
protein-membrane systems to access mesoscopic scales. Faraday Discussions.

Bauerfeind, R., K. Takei, and P. De Camilli. 1997. Amphiphysin I is associated with
coated endocytic intermediates and undergoes stimulation-dependent
dephosphorylation in nerve terminals. J Biol Chem 272(49):30984-30992.

Bergmann, C., K. Zerres, J. Senderek, S. Rudnik-Schoneborn, T. Eggermann, M.
Hausler, M. Mull, and V.T. Ramaekers. 2003. Oligophrenin 1 (OPHN1) gene
mutation causes syndromic X-linked mental retardation with epilepsy, rostral
ventricular enlargement and cerebellar hypoplasia. Brain 126(Pt 7):1537-1544.

Bhatnagar, J., J.H. Freed, and B.R. Crane. 2007. Rigid body refinement of protein
complexes with long-range distance restraints from pulsed dipolar ESR. Methods
Enzymol 423:117-133.

Bigay, J., J.F. Casella, G. Drin, B. Mesmin, and B. Antonny. 2005. ArfGAP1 responds to
membrane curvature through the folding of a lipid packing sensor motif. Embo J
24(13):2244-2253.

Bigay, J., P. Gounon, S. Robineau, and B. Antonny. 2003. Lipid packing sensed by
ArfGAP1 couples COPI coat disassembly to membrane bilayer curvature. Nature
426(6966):563-566.

Billuart, P., J. Chelly, A. Carrie, M. Vinet, P. Couvert, N. McDonell, R. Zemni, A. Kahn,
C. Moraine, C. Beldjord, and T. Bienvenu. 2000. Determination of the gene
structure of human oligophrenin-1 and identification of three novel
polymorphisms by screening of DNA from 164 patients with non-specific X-
linked mental retardation. Ann Genet 43(1):5-9.

Bisaglia, M., I. Tessari, L. Pinato, M. Bellanda, S. Giraudo, M. Fasano, E. Bergantino, L.
Bubacco, and S. Mammi. 2005. A topological model of the interaction between
alpha-synuclein and sodium dodecyl sulfate micelles. Biochemistry 44(1):329-
339.

Bisaglia, M., A. Trolio, M. Bellanda, E. Bergantino, L. Bubacco, and S. Mammi. 2006.
Structure and topology of the non-amyloid-beta component fragment of human
alpha-synuclein bound to micelles: implications for the aggregation process.
Protein Sci 15(6):1408-1416.
225
Blood, P.D., R.D. Swenson, and G.A. Voth. 2008. Factors influencing local membrane
curvature induction by N-BAR domains as revealed by molecular dynamics
simulations. Biophys J 95(4):1866-1876.

Blood, P.D., and G.A. Voth. 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(41):15068-15072.

Borbat, P., T.F. Ramlall, J.H. Freed, and D. Eliezer. 2006. Inter-helix distances in
lysophospholipid micelle-bound alpha-synuclein from pulsed ESR measurements.
J Am Chem Soc 128(31):10004-10005.

Borbat, P.P., A.J. Costa-Filho, K.A. Earle, J.K. Moscicki, and J.H. Freed. 2001. Electron
spin resonance in studies of membranes and proteins. Science 291(5502):266-269.

Borbat, P.P., J.H. Davis, S.E. Butcher, and J.H. Freed. 2004. Measurement of large
distances in biomolecules using double-quantum filtered refocused electron spin-
echoes. J Am Chem Soc 126(25):7746-7747.

Bortolus, M., F. Tombolato, I. Tessari, M. Bisaglia, S. Mammi, L. Bubacco, A. Ferrarini,
and A.L. Maniero. 2008. Broken helix in vesicle and micelle-bound alpha-
synuclein: insights from site-directed spin labeling-EPR experiments and MD
simulations. J Am Chem Soc 130(21):6690-6691.

Bussell, R., Jr., and D. Eliezer. 2003. A structural and functional role for 11-mer repeats
in alpha-synuclein and other exchangeable lipid binding proteins. J Mol Biol
329(4):763-778.

Bussell, R., Jr., T.F. Ramlall, and D. Eliezer. 2005. Helix periodicity, topology, and
dynamics of membrane-associated alpha-synuclein. Protein Sci 14(4):862-872.

Butler, M.H., C. David, G.C. Ochoa, Z. Freyberg, L. Daniell, D. Grabs, O. Cremona, and
P. De Camilli. 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(6):1355-1367.

Cabin, D.E., K. Shimazu, D. Murphy, N.B. Cole, W. Gottschalk, K.L. McIlwain, B.
Orrison, A. Chen, C.E. Ellis, R. Paylor, B. Lu, and R.L. Nussbaum. 2002.
Synaptic vesicle depletion correlates with attenuated synaptic responses to
prolonged repetitive stimulation in mice lacking alpha-synuclein. J Neurosci
22(20):8797-8807.

226
Campelo, F., H.T. McMahon, and M.M. Kozlov. 2008. The hydrophobic insertion
mechanism of membrane curvature generation by proteins. Biophys J 95(5):2325-
2339.

Case, D.A., T.A. Darden, I. Cheatham, T.E., C.L. Simmerling, W. J., R.E. Duke, R. Luo,
K.M. Merz, B. Wang, D.A. Pearlman, M. Crowley, S. Brozell, V. Tsui, H.
Gohlke, J. Mongan, V. Hornak, G. Cui, P. Beroza, C. Shafmeister, J.W. Caldwell,
W.S. Ross, and P.A. Kollman. 2004. AMBER 8. University of California, San
Francisco.

Chandra, S., X. Chen, J. Rizo, R. Jahn, and T.C. Sudhof. 2003. A broken alpha -helix in
folded alpha-Synuclein. J Biol Chem 278(17):15313-15318.

Chen, H., S. Fre, V.I. Slepnev, M.R. Capua, K. Takei, M.H. Butler, P.P. Di Fiore, and P.
De Camilli. 1998. Epsin is an EH-domain-binding protein implicated in clathrin-
mediated endocytosis. Nature 394(6695):793-797.

Chen, Y., L. Deng, Y. Maeno-Hikichi, M. Lai, S. Chang, G. Chen, and J.F. Zhang. 2003.
Formation of an endophilin-Ca2+ channel complex is critical for clathrin-
mediated synaptic vesicle endocytosis. Cell 115(1):37-48.

Chen, Y.H., J.T. Yang, and H.M. Martinez. 1972. Determination of the secondary
structures of proteins by circular dichroism and optical rotatory dispersion.
Biochemistry 11(22):4120-4131.

Chen, Y.J., P. Zhang, E.H. Egelman, and J.E. Hinshaw. 2004. The stalk region of
dynamin drives the constriction of dynamin tubes. Nat Struct Mol Biol 11(6):574-
575.

Chiang, Y.W., P.P. Borbat, and J.H. Freed. 2005. The determination of pair distance
distributions by pulsed ESR using Tikhonov regularization. J Magn Reson
172(2):279-295.

Cole, N.B., D.D. Murphy, T. Grider, S. Rueter, D. Brasaemle, and R.L. Nussbaum. 2002.
Lipid droplet binding and oligomerization properties of the Parkinson's disease
protein alpha-synuclein. J Biol Chem 277(8):6344-6352.

Collaborative Computational Project, N. 1994. The CCP4 suite: programs for protein
crystallography. Acta Crystallogr D Biol Crystallogr 50(Pt 5):760-763.

Cornell, W.D., P. Cieplak, C.I. Bayly, I.R. Gould, K.M. Merz, D.M. Ferguson, D.C.
Spellmeyer, T. Fox, J.W. Caldwell, and P.A. Kollman. 1995. A Second
Generation Force Field for the Simulation of Proteins, Nulceic Acids, and
Organic Molecules. J Am Chem Soc 117(19):5179-5197.
227
Cremona, O., G. Di Paolo, M.R. Wenk, A. Luthi, W.T. Kim, K. Takei, L. Daniell, Y.
Nemoto, S.B. Shears, R.A. Flavell, D.A. McCormick, and P. De Camilli. 1999.
Essential role of phosphoinositide metabolism in synaptic vesicle recycling. Cell
99(2):179-188.

Cui, H., G.S. Ayton, and G.A. Voth. accepted. Molecular dynamics simulation of
membrane remodeling by endophilin N-BAR domains. Biophys J.

D'Souza-Schorey, C., and P. Chavrier. 2006. ARF proteins: roles in membrane traffic and
beyond. Nat Rev Mol Cell Biol 7(5):347-358.

Darden, T., D. York, and L. Pedersen. 1993. Particle mesh Ewald: An N-log(N) method
for Ewald sums in large systems. J. Chem. Phys. 98:10089-10092.

David, C., M. Solimena, and P. De Camilli. 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(1):73-
79.

Davidson, W.S., A. Jonas, D.F. Clayton, and J.M. George. 1998. Stabilization of alpha-
synuclein secondary structure upon binding to synthetic membranes. J Biol Chem
273(16):9443-9449.

De Camilli, P., H. Chen, J. Hyman, E. Panepucci, A. Bateman, and A.T. Brunger. 2002.
The ENTH domain. FEBS Lett 513(1):11-18.

De Camilli, P., A. Thomas, R. Cofiell, F. Folli, B. Lichte, G. Piccolo, H.M. Meinck, M.
Austoni, G. Fassetta, G. Bottazzo, and 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(6):2219-2223.

de la Fortelle, E., and G. Bricogne. 1997. Maximum-likelihood heavy-atom parameter
refinement for multiple isomorphous replacement and multiwavelength
anomalous diffraction methods. In Methods in Enzymology. Carter C, Jr., Sweet
R, editors. 472-494.

Dickman, D.K., J.A. Horne, I.A. Meinertzhagen, and T.L. Schwarz. 2005. A slowed
classical pathway rather than kiss-and-run mediates endocytosis at synapses
lacking synaptojanin and endophilin. Cell 123(3):521-533.

Diederichs, K., and P.A. Karplus. 1997. Improved R-factors for diffraction data analysis
in macromolecular crystallography. Nat Struct Biol 4(4):269-275.

Doherty, G.J., and H.T. McMahon. 2009. Mechanisms of endocytosis. Annu Rev
Biochem 78:857-902.
228
Drescher, M., G. Veldhuis, B.D. van Rooijen, S. Milikisyants, V. Subramaniam, and M.
Huber. 2008. Antiparallel arrangement of the helices of vesicle-bound alpha-
synuclein. J Am Chem Soc 130(25):7796-7797.

Drin, G., J.F. Casella, R. Gautier, T. Boehmer, T.U. Schwartz, and B. Antonny. 2007. A
general amphipathic alpha-helical motif for sensing membrane curvature. Nat
Struct Mol Biol 14(2):138-146.

Eliezer, D., E. Kutluay, R. Bussell, Jr., and G. Browne. 2001. Conformational properties
of alpha-synuclein in its free and lipid-associated states. J Mol Biol 307(4):1061-
1073.

Emsley, P., and K. Cowtan. 2004. Coot: model-building tools for molecular graphics.
Acta Crystallogr D Biol Crystallogr 60(Pt 12 Pt 1):2126-2132.

Evergren, E., M. Marcucci, N. Tomilin, P. Low, V. Slepnev, F. Andersson, H. Gad, L.
Brodin, P. De Camilli, and O. Shupliakov. 2004. Amphiphysin is a component of
clathrin coats formed during synaptic vesicle recycling at the lamprey giant
synapse. Traffic 5(7):514-528.

Fanucci, G.E., and D.S. Cafiso. 2006. Recent advances and applications of site-directed
spin labeling. Curr Opin Struct Biol 16(5):644-653.

Farsad, K., and P. De Camilli. 2003. Mechanisms of membrane deformation. Curr Opin
Cell Biol 15(4):372-381.

Farsad, K., N. Ringstad, K. Takei, S.R. Floyd, K. Rose, and P. De Camilli. 2001.
Generation of high curvature membranes mediated by direct endophilin bilayer
interactions. J Cell Biol 155(2):193-200.

Feller, S.E., and A.D. MacKerrell. 2000. An improved empirical potential energy
function for molecular dynamics simulations of phospholipids. J. Phys. Chem. B
104:7510-7515.

Feller, S.E., Y. Zhang, R.W. Pastor, and B.R. Brooks. 1995. Constant pressure molecular
dynamics simulation: The Langevin piston. J. Chem. Phys. 103:4613-4621.

Ferreon, A.C., Y. Gambin, E.A. Lemke, and A.A. Deniz. 2009. Interplay of alpha-
synuclein binding and conformational switching probed by single-molecule
fluorescence. Proc Natl Acad Sci U S A 106(14):5645-5650.

Fischer, T., L. Lu, H.T. Haigler, and R. Langen. 2007. Annexin B12 is a sensor of
membrane curvature and undergoes major curvature-dependent structural
changes. J Biol Chem 282(13):9996-10004.
229
Ford, M.G., I.G. Mills, B.J. Peter, Y. Vallis, G.J. Praefcke, P.R. Evans, and H.T.
McMahon. 2002. Curvature of clathrin-coated pits driven by epsin. Nature
419(6905):361-366.

Ford, M.G., B.M. Pearse, M.K. Higgins, Y. Vallis, D.J. Owen, A. Gibson, C.R. Hopkins,
P.R. Evans, and H.T. McMahon. 2001. Simultaneous binding of PtdIns(4,5)P2
and clathrin by AP180 in the nucleation of clathrin lattices on membranes.
Science 291(5506):1051-1055.

Fortin, D.L., M.D. Troyer, K. Nakamura, S. Kubo, M.D. Anthony, and R.H. Edwards.
2004. Lipid rafts mediate the synaptic localization of alpha-synuclein. J Neurosci
24(30):6715-6723.

Frost, A., P. De Camilli, and V.M. Unger. 2007. F-BAR proteins join the BAR family
fold. Structure 15(7):751-753.

Frost, A., R. Perera, A. Roux, K. Spasov, O. Destaing, E.H. Egelman, P. De Camilli, and
V.M. Unger. 2008. Structural basis of membrane invagination by F-BAR
domains. Cell 132(5):807-817.

Frost, A., V.M. Unger, andP. De Camilli. 2009. The BAR domain superfamily:
membrane-molding macromolecules. Cell 137(2):191-196.

Furuya, T., T. Kiyota, S. Lee, T. Inoue, G. Sugihara, A. Logvinova, P. Goldsmith, and
H.M. Ellerby. 2003. Nanotubules formed by highly hydrophobic amphiphilic
alpha-helical peptides and natural phospholipids. Biophys J 84(3):1950-1959.

Gad, H., N. Ringstad, P. Low, O. Kjaerulff, J. Gustafsson, M. Wenk, G. Di Paolo, Y.
Nemoto, J. Crun, M.H. Ellisman, P. De Camilli, O. Shupliakov, and L. Brodin.
2000. Fission and uncoating of synaptic clathrin-coated vesicles are perturbed by
disruption of interactions with the SH3 domain of endophilin. Neuron 27(2):301-
312.

Gallop, J.L., P.J. Butler, and H.T. McMahon. 2005. Endophilin and CtBP/BARS are not
acyl transferases in endocytosis or Golgi fission. Nature 438(7068):675-678.

Gallop, J.L., C.C. Jao, H.M. Kent, P.J. Butler, P.R. Evans, R. Langen, and H.T.
McMahon. 2006. Mechanism of endophilin N-BAR domain-mediated membrane
curvature. Embo J 25(12):2898-2910.

Gallop, J.L., and H.T. McMahon. 2005. BAR domains and membrane curvature: bringing
your curves to the BAR. Biochem Soc Symp(72):223-231.

Gennis, R.B. 1989. Biomembranes: Molecular Structure and Function. Springer-Verlag,
New York.
230
George, J.M., H. Jin, W.S. Woods, and D.F. Clayton. 1995. Characterization of a novel
protein regulated during the critical period for song learning in the zebra finch.
Neuron 15(2):361-372.

Georgieva, E.R., T.F. Ramlall, P.P. Borbat, J.H. Freed, and D. Eliezer. 2008. Membrane-
bound alpha-synuclein forms an extended helix: long-distance pulsed ESR
measurements using vesicles, bicelles, and rodlike micelles. J Am Chem Soc
130(39):12856-12857.

Gerke, V., and S.E. Moss. 2002. Annexins: from structure to function. Physiol Rev
82(2):331-371.

Gitler, A.D., B.J. Bevis, J. Shorter, K.E. Strathearn, S. Hamamichi, L.J. Su, K.A.
Caldwell, G.A. Caldwell, J.C. Rochet, J.M. McCaffery, C. Barlowe, and S.
Lindquist. 2008. The Parkinson's disease protein alpha-synuclein disrupts cellular
Rab homeostasis. Proc Natl Acad Sci U S A 105(1):145-150.

Gross, A., L. Columbus, K. Hideg, C. Altenbach, and W.L. Hubbell. 1999. Structure of
the KcsA potassium channel from Streptomyces lividans: a site-directed spin
labeling study of the second transmembrane segment. Biochemistry
38(32):10324-10335.

Guichet, A., T. Wucherpfennig, V. Dudu, S. Etter, M. Wilsch-Brauniger, A. Hellwig, M.
Gonzalez-Gaitan, W.B. Huttner, and A.A. Schmidt. 2002. Essential role of
endophilin A in synaptic vesicle budding at the Drosophila neuromuscular
junction. Embo J 21(7):1661-1672.

Guo, Z., D. Cascio, K. Hideg, and W.L. Hubbell. 2008. Structural determinants of
nitroxide motion in spin-labeled proteins: solvent-exposed sites in helix B of T4
lysozyme. Protein Sci 17(2):228-239.

Guo, Z., D. Cascio, K. Hideg, T. Kalai, and W.L. Hubbell. 2007. Structural determinants
of nitroxide motion in spin-labeled proteins: tertiary contact and solvent-
inaccessible sites in helix G of T4 lysozyme. Protein Sci 16(6):1069-1086.

Harbury, P.B., J.J. Plecs, B. Tidor, T. Alber, and P.S. Kim. 1998. High-resolution protein
design with backbone freedom. Science 282(5393):1462-1467.

Hegde, B.G., J.M. Isas, G. Zampighi, H.T. Haigler, and R. Langen. 2006. A novel
calcium-independent peripheral membrane-bound form of annexin B12.
Biochemistry 45(3):934-942.

231
Henne, W.M., H.M. Kent, M.G. Ford, B.G. Hegde, O. Daumke, P.J. Butler, R. Mittal, R.
Langen, P.R. Evans, and H.T. McMahon. 2007. Structure and analysis of FCHo2
F-BAR domain: a dimerizing and membrane recruitment module that effects
membrane curvature. Structure 15(7):839-852.

Hinshaw, J.E., and S.L. Schmid. 1995. Dynamin self-assembles into rings suggesting a
mechanism for coated vesicle budding. Nature 374(6518):190-192.

Hubbell, W.L., D.S. Cafiso, and C. Altenbach. 2000. Identifying conformational changes
with site-directed spin labeling. Nat Struct Biol 7(9):735-739.

Hubbell, W.L., A. Gross, R. Langen, and M.A. Lietzow. 1998. Recent advances in site-
directed spin labeling of proteins. Curr Opin Struct Biol 8(5):649-656.

Humphrey, W., A. Dalke, and K. Schulten. 1996. VMD: visual molecular dynamics. J
Mol Graph 14(1):33-38, 27-38.

Hurley, J.H., and B. Wendland. 2002. Endocytosis: driving membranes around the bend.
Cell 111(2):143-146.

Huttner, W.B., and A. Schmidt. 2000. Lipids, lipid modification and lipid-protein
interaction in membrane budding and fission--insights from the roles of
endophilin A1 and synaptophysin in synaptic vesicle endocytosis. Curr Opin
Neurobiol 10(5):543-551.

Hyman, J., H. Chen, P.P. Di Fiore, P. De Camilli, and A.T. Brunger. 2000. Epsin 1
undergoes nucleocytosolic shuttling and its eps15 interactor NH(2)-terminal
homology (ENTH) domain, structurally similar to Armadillo and HEAT repeats,
interacts with the transcription factor promyelocytic leukemia Zn(2)+ finger
protein (PLZF). J Cell Biol 149(3):537-546.

Irizarry, M.C., T.W. Kim, M. McNamara, R.E. Tanzi, J.M. George, D.F. Clayton, and
B.T. Hyman. 1996. Characterization of the precursor protein of the non-A beta
component of senile plaques (NACP) in the human central nervous system. J
Neuropathol Exp Neurol 55(8):889-895.

Isas, J.M., J.P. Cartailler, Y. Sokolov, D.R. Patel, R. Langen, H. Luecke, J.E. Hall, and
H.T. Haigler. 2000. Annexins V and XII insert into bilayers at mildly acidic pH
and form ion channels. Biochemistry 39(11):3015-3022.

Isas, J.M., Y.E. Kim, C.C. Jao, P.B. Hegde, H.T. Haigler, and R. Langen. 2005. Calcium-
and membrane-induced changes in the structure and dynamics of three helical
hairpins in annexin B12. Biochemistry 44(50):16435-16444.
232
Isas, J.M., R. Langen, H.T. Haigler, and W.L. Hubbell. 2002. Structure and dynamics of
a helical hairpin and loop region in annexin 12: a site-directed spin labeling study.
Biochemistry 41(5):1464-1473.

Isas, J.M., R. Langen, W.L. Hubbell, and H.T. Haigler. 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, J.M., D.R. Patel, C. Jao, S. Jayasinghe, J.P. Cartailler, H.T. Haigler, and R. Langen.
2003. Global structural changes in annexin 12. The roles of phospholipid, Ca2+,
and pH. J Biol Chem 278(32):30227-30234.

Itoh, T., S. Koshiba, T. Kigawa, A. Kikuchi, S. Yokoyama, and T. Takenawa. 2001. Role
of the ENTH domain in phosphatidylinositol-4,5-bisphosphate binding and
endocytosis. Science 291(5506):1047-1051.

Iwai, A., E. Masliah, M. Yoshimoto, N. Ge, L. Flanagan, H.A. de Silva, A. Kittel, and T.
Saitoh. 1995. The precursor protein of non-A beta component of Alzheimer's
disease amyloid is a presynaptic protein of the central nervous system. Neuron
14(2):467-475.

Jakes, R., M.G. Spillantini, and M. Goedert. 1994. Identification of two distinct
synucleins from human brain. FEBS Lett 345(1):27-32.

Jao, C.C., A. Der-Sarkissian, J. Chen, and R. Langen. 2004. Structure of membrane-
bound alpha-synuclein studied by site-directed spin labeling. Proc Natl Acad Sci
U S A 101(22):8331-8336.

Jao, C.C., B.G. Hegde, J. Chen, I.S. Haworth, and R. Langen. 2008. Structure of
membrane-bound alpha-synuclein from site-directed spin labeling and
computational refinement. Proc Natl Acad Sci U S A 105(50):19666-19671.

Jaud, S., D.J. Tobias, J.J. Falke, and S.H. White. 2007. Self-induced docking site of a
deeply embedded peripheral membrane protein. Biophys J 92(2):517-524.

Jeschke, G. 2002. Distance measurements in the nanometer range by pulse EPR.
Chemphyschem 3(11):927-932.

Jeschke, G., V. Chechik, P. Ionita, A. Godt, H. Zimmermann, J. Banham, C.R. Timmel,
D. Hilger, and H. Jung. 2006. DeerAnalysis2006 - a Comprehensive Software
Package for Analyzing Pulsed ELDOR Data. Appl Magn Reson 30:473-498.

Jeschke, G., and Y. Polyhach. 2007. Distance measurements on spin-labelled
biomacromolecules by pulsed electron paramagnetic resonance. Phys Chem Chem
Phys 9(16):1895-1910.
233
Jockusch, W.J., G.J. Praefcke, H.T. McMahon, and L. Lagnado. 2005. Clathrin-
dependent and clathrin-independent retrieval of synaptic vesicles in retinal bipolar
cells. Neuron 46(6):869-878.

Jones, T.A., J.Y. Zou, S.W. Cowan, and M. Kjeldgaard. 1991. Improved methods for
building protein models in electron density maps and the location of errors in
these models. Acta Crystallogr A 47 ( Pt 2):110-119.

Jorgensen, W.L., J. Chandrasekhar, and J.D. Madura. 1983. Comparison of simple
potential functions for simulating liquid water. J. Chem. Phys. 79:926-935.

Kamp, F., andK. Beyer. 2006. Binding of alpha-synuclein affects the lipid packing in
bilayers of small vesicles. J Biol Chem 281(14):9251-9259.

Kim, Y.E., J.M. Isas, H.T. Haigler, and R. Langen. 2005. A helical hairpin region of
soluble annexin B12 refolds and forms a continuous transmembrane helix at
mildly acidic pH. J Biol Chem 280(37):32398-32404.

Krauss, M., J.Y. Jia, A. Roux, R. Beck, F.T. Wieland, P. De Camilli, and V. Haucke.
2008. Arf1-GTP-induced tubule formation suggests a function of Arf family
proteins in curvature acquisition at sites of vesicle budding. J Biol Chem
283(41):27717-27723.

Kweon, D.H., Y.K. Shin, J.Y. Shin, J.H. Lee, J.B. Lee, J.H. Seo, and Y.S. Kim. 2006.
Membrane topology of helix 0 of the Epsin N-terminal homology domain. Mol
Cells 21(3):428-435.

Langen, R., J.M. Isas, H. Luecke, H.T. Haigler, and W.L. Hubbell. 1998. Membrane-
mediated assembly of annexins studied by site-directed spin labeling. J Biol Chem
273(35):22453-22457.

Langen, R., K.J. Oh, D. Cascio, and W.L. Hubbell. 2000. Crystal structures of spin
labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra
in terms of structure. Biochemistry 39(29):8396-8405.

Lee, E., M. Marcucci, L. Daniell, M. Pypaert, O.A. Weisz, G.C. Ochoa, K. Farsad, M.R.
Wenk, and P. De Camilli. 2002a. Amphiphysin 2 (Bin1) and T-tubule biogenesis
in muscle. Science 297(5584):1193-1196.

Lee, F.J., F. Liu, Z.B. Pristupa, and H.B. Niznik. 2001a. Direct binding and functional
coupling of alpha-synuclein to the dopamine transporters accelerate dopamine-
induced apoptosis. Faseb J 15(6):916-926.

234
Lee, H.J., C. Choi, and S.J. Lee. 2002b. Membrane-bound alpha-synuclein has a high
aggregation propensity and the ability to seed the aggregation of the cytosolic
form. J Biol Chem 277(1):671-678.

Lee, M.C., L. Orci, S. Hamamoto, E. Futai, M. Ravazzola, and R. Schekman. 2005.
Sar1p N-terminal helix initiates membrane curvature and completes the fission of
a COPII vesicle. Cell 122(4):605-617.

Lee, S., T. Furuya, T. Kiyota, N. Takami, K. Murata, Y. Niidome, D.E. Bredesen, H.M.
Ellerby, and G. Sugihara. 2001b. De novo-designed peptide transforms Golgi-
specific lipids into Golgi-like nanotubules. J Biol Chem 276(44):41224-41228.

Leslie, A. 1992. Recent changes to the MOSFLM package for processing film and image
plate data. Joint CCP4 and ESF-EACMB Newsletter on Protein
Crystallography(No. 26).

Lotharius, J., and P. Brundin. 2002. Pathogenesis of Parkinson's disease: dopamine,
vesicles and alpha-synuclein. Nat Rev Neurosci 3(12):932-942.

Low, C., U. Weininger, H. Lee, K. Schweimer, I. Neundorf, A.G. Beck-Sickinger, R.W.
Pastor, and J. Balbach. 2008. Structure and dynamics of helix-0 of the N-BAR
domain in lipid micelles and bilayers. Biophys J 95(9):4315-4323.

MacKerell, A.D., D. Dashford, M. Bellott, R.L. Dunbrack, J.D. Evanseck, M.J. Field, S.
Fischer, J. Gao, H. Guo, S. Ha, D. Joseph-McCarthy, L. Kuchnir, K. Kuczera,
F.T.K. Lau, C. Mattos, S. Michnick, T. Ngo, D.T. Nguyen, B. Prodhom, W.E.
Reiher, B. Roux, M. Schlenkrich, J.C. Smith, R. Stote, J. Straub, M. Watanabe, J.
Wiokiewicz-Kuczera, D. Yin, and M. Karplus. 1998. All-atom empirical potential
for molecular modeling and dynamics studies of proteins. J. Phys. Chem. B
102:3586-3616.

Margittai, M., D. Fasshauer, S. Pabst, R. Jahn, and R. Langen. 2001. Homo- and
heterooligomeric SNARE complexes studied by site-directed spin labeling. J Biol
Chem 276(16):13169-13177.

Margittai, M., and R. Langen. 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(3-4):265-297.

Masliah, E., E. Rockenstein, A. Adame, M. Alford, L. Crews, M. Hashimoto, P. Seubert,
M. Lee, J. Goldstein, T. Chilcote, D. Games, and D. Schenk. 2005. Effects of
alpha-synuclein immunization in a mouse model of Parkinson's disease. Neuron
46(6):857-868.
235
Masuda, M., S. Takeda, M. Sone, T. Ohki, H. Mori, Y. Kamioka, and N. Mochizuki.
2006. Endophilin BAR domain drives membrane curvature by two newly
identified structure-based mechanisms. Embo J 25(12):2889-2897.

Mchaourab, H.S., M.A. Lietzow, K. Hideg, and W.L. Hubbell. 1996. Motion of spin-
labeled side chains in T4 lysozyme. Correlation with protein structure and
dynamics. Biochemistry 35(24):7692-7704.

McMahon, H.T., and J.L. Gallop. 2005. Membrane curvature and mechanisms of
dynamic cell membrane remodelling. Nature 438(7068):590-596.

McMahon, H.T., and I.G. Mills. 2004. COP and clathrin-coated vesicle budding:
different pathways, common approaches. Curr Opin Cell Biol 16(4):379-391.

McPherson, P.S., E. de Heuvel, J. Phillie, W. Wang, A. Sengar, and S. Egan. 1998. EH
domain-dependent interactions between Eps15 and clathrin-coated vesicle protein
p95. Biochem Biophys Res Commun 244(3):701-705.

Mills, I.G., G.J. Praefcke, Y. Vallis, B.J. Peter, L.E. Olesen, J.L. Gallop, P.J. Butler, P.R.
Evans, and H.T. McMahon. 2003. EpsinR: an AP1/clathrin interacting protein
involved in vesicle trafficking. J Cell Biol 160(2):213-222.

Miyamoto, S., and P.A. Kollman. 1992. Settle: An analytical version of the SHAKE and
RATTLE algorithm for rigid water models. J. Comput. Chem. 13:952-962.

Mui, B.L., H.G. Dobereiner, T.D. Madden, and P.R. Cullis. 1995. Influence of
transbilayer area asymmetry on the morphology of large unilamellar vesicles.
Biophys J 69(3):930-941.

Murphy, D.D., S.M. Rueter, J.Q. Trojanowski, and V.M. Lee. 2000. Synucleins are
developmentally expressed, and alpha-synuclein regulates the size of the
presynaptic vesicular pool in primary hippocampal neurons. J Neurosci
20(9):3214-3220.

Murshudov, G.N., A.A. Vagin, and E.J. Dodson. 1997. Refinement of macromolecular
structures by the maximum-likelihood method. Acta Crystallogr D Biol
Crystallogr 53(Pt 3):240-255.

Nicot, A.S., A. Toussaint, V. Tosch, C. Kretz, C. Wallgren-Pettersson, E. Iwarsson, H.
Kingston, J.M. Garnier, V. Biancalana, A. Oldfors, J.L. Mandel, and J. Laporte.
2007. Mutations in amphiphysin 2 (BIN1) disrupt interaction with dynamin 2 and
cause autosomal recessive centronuclear myopathy. Nat Genet 39(9):1134-1139.

Nossal, R. 2001. Energetics of clathrin basket assembly. Traffic 2(2):138-147.
236
Nuscher, B., F. Kamp, T. Mehnert, S. Odoy, C. Haass, P.J. Kahle, and K. Beyer. 2004.
Alpha-synuclein has a high affinity for packing defects in a bilayer membrane: a
thermodynamics study. J Biol Chem 279(21):21966-21975.

Oda, M.N., T.M. Forte, R.O. Ryan, and J.C. Voss. 2003. The C-terminal domain of
apolipoprotein A-I contains a lipid-sensitive conformational trigger. Nat Struct
Biol 10(6):455-460.

Oh, K.J., C. Altenbach, R.J. Collier, and W.L. Hubbell. 2000. Site-directed spin labeling
of proteins. Applications to diphtheria toxin. Methods Mol Biol 145:147-169.

Pannier, M., S. Veit, A. Godt, G. Jeschke, and H.W. Spiess. 2000. Dead-time free
measurement of dipole-dipole interactions between electron spins. J Magn Reson
142(2):331-340.

Pearlman, D.A., D.A. Case, J.W. Caldwell, W.S. Ross, I. Cheatham, T.E., S. DeBolt, D.
Ferguson, G. Seibel, and P. Kollman. 1995. AMBER, a package of computer
programs for applying molecular mechanics, normal mode analysis, molecular
dynamics and free energy calculations to simulate the structural and energetic
properties of molecules Comp Phys Comm 91(1-3):1-41.

Perrin, R.J., W.S. Woods, D.F. Clayton, and J.M. George. 2000. Interaction of human
alpha-Synuclein and Parkinson's disease variants with phospholipids. Structural
analysis using site-directed mutagenesis. J Biol Chem 275(44):34393-34398.

Peter, B.J., H.M. Kent, I.G. Mills, Y. Vallis, P.J. Butler, P.R. Evans, and H.T. McMahon.
2004. BAR domains as sensors of membrane curvature: the amphiphysin BAR
structure. Science 303(5657):495-499.

Phillips, J.C., R. Braun, W. Wang, J. Gumbart, E. Tajkhorshid, E. Villa, C. Chipot, R.D.
Skeel, L. Kale, and K. Schulten. 2005. Scalable molecular dynamics with NAMD.
J Comput Chem 26(16):1781-1802.

Philo, J.S. 2000. A method for directly fitting the time derivative of sedimentation
velocity data and an alternative algorithm for calculating sedimentation
coefficient distribution functions. Anal Biochem 279(2):151-163.

Pierrat, B., M. Simonen, M. Cueto, J. Mestan, P. Ferrigno, and J. Heim. 2001. SH3GLB,
a new endophilin-related protein family featuring an SH3 domain. Genomics
71(2):222-234.

Plecs, J.J., P.B. Harbury, P.S. Kim, and T. Alber. 2004. Structural test of the
parameterized-backbone method for protein design. J Mol Biol 342(1):289-297.
237
Price, E.A., B.T. Sutch, Q. Cai, P.Z. Qin, and I.S. Haworth. 2007. Computation of
nitroxide-nitroxide distances in spin-labeled DNA duplexes. Biopolymers
87(1):40-50.

Rabenstein, M.D., and Y.K. Shin. 1995. Determination of the distance between two spin
labels attached to a macromolecule. Proc Natl Acad Sci U S A 92(18):8239-8243.

Ramakrishnan, M., P.H. Jensen, and D. Marsh. 2003. Alpha-synuclein association with
phosphatidylglycerol probed by lipid spin labels. Biochemistry 42(44):12919-
12926.

Razzaq, A., I.M. Robinson, H.T. McMahon, J.N. Skepper, Y. Su, A.C. Zelhof, A.P.
Jackson, N.J. Gay, and C.J. O'Kane. 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(22):2967-2979.

Richnau, N., A. Fransson, K. Farsad, and P. Aspenstrom. 2004. RICH-1 has a
BIN/Amphiphysin/Rvsp domain responsible for binding to membrane lipids and
tubulation of liposomes. Biochem Biophys Res Commun 320(3):1034-1042.

Rikhy, R., V. Kumar, R. Mittal, and K.S. Krishnan. 2002. Endophilin is critically
required for synapse formation and function in Drosophila melanogaster. J
Neurosci 22(17):7478-7484.

Ringstad, N., H. Gad, P. Low, G. Di Paolo, L. Brodin, O. Shupliakov, and P. De Camilli.
1999. Endophilin/SH3p4 is required for the transition from early to late stages in
clathrin-mediated synaptic vesicle endocytosis. Neuron 24(1):143-154.

Robinson, M.S. 2004. Adaptable adaptors for coated vesicles. Trends Cell Biol
14(4):167-174.

Saarikangas, J., H. Zhao, A. Pykalainen, P. Laurinmaki, P.K. Mattila, P.K. Kinnunen, S.J.
Butcher, and P. Lappalainen. 2009. Molecular mechanisms of membrane
deformation by I-BAR domain proteins. Curr Biol 19(2):95-107.

Sakamuro, D., K.J. Elliott, R. Wechsler-Reya, and G.C. Prendergast. 1996. BIN1 is a
novel MYC-interacting protein with features of a tumour suppressor. Nat Genet
14(1):69-77.

Salcini, A.E., H. Chen, G. Iannolo, P. De Camilli, and P.P. Di Fiore. 1999. Epidermal
growth factor pathway substrate 15, Eps15. Int J Biochem Cell Biol 31(8):805-
809.

Schmid, E.M., and H.T. McMahon. 2007. Integrating molecular and network biology to
decode endocytosis. Nature 448(7156):883-888.
238
Schuske, K.R., J.E. Richmond, D.S. Matthies, W.S. Davis, S. Runz, D.A. Rube, A.M.
van der Bliek, and E.M. Jorgensen. 2003. Endophilin is required for synaptic
vesicle endocytosis by localizing synaptojanin. Neuron 40(4):749-762.

Scoville, D., J.D. Stamm, C. Altenbach, A. Shvetsov, K. Kokabi, P.A. Rubenstein, W.L.
Hubbell, and E. Reisler. 2009. Effects of binding factors on structural elements in
F-actin. Biochemistry 48(2):370-378.

Segrest, J.P., M.K. Jones, A.E. Klon, C.J. Sheldahl, M. Hellinger, H. De Loof, and S.C.
Harvey. 1999. A detailed molecular belt model for apolipoprotein A-I in discoidal
high density lipoprotein. J Biol Chem 274(45):31755-31758.

Sheetz, M.P., and S.J. Singer. 1974. Biological membranes as bilayer couples. A
molecular mechanism of drug-erythrocyte interactions. Proc Natl Acad Sci U S A
71(11):4457-4461.

Simpson, F., N.K. Hussain, B. Qualmann, R.B. Kelly, B.K. Kay, P.S. McPherson, and
S.L. Schmid. 1999. SH3-domain-containing proteins function at distinct steps in
clathrin-coated vesicle formation. Nat Cell Biol 1(2):119-124.

Smith, D.P., D.J. Tew, A.F. Hill, S.P. Bottomley, C.L. Masters, K.J. Barnham, and R.
Cappai. 2008. Formation of a high affinity lipid-binding intermediate during the
early aggregation phase of alpha-synuclein. Biochemistry 47(5):1425-1434.

Soper, J.H., S. Roy, A. Stieber, E. Lee, R.B. Wilson, J.Q. Trojanowski, C.G. Burd, and
V.M. Lee. 2008. α-Synuclein-induced Aggregation of Cytoplasmic Vesicles in
Saccharomyces cerevisiae. Mol Biol Cell 19(3):1093-1103.

Stahelin, R.V., F. Long, B.J. Peter, D. Murray, P. De Camilli, H.T. McMahon, and W.
Cho. 2003. Contrasting membrane interaction mechanisms of AP180 N-terminal
homology (ANTH) and epsin N-terminal homology (ENTH) domains. J Biol
Chem 278(31):28993-28999.

Stetefeld, J., M. Jenny, T. Schulthess, R. Landwehr, J. Engel, and R.A. Kammerer. 2000.
Crystal structure of a naturally occurring parallel right-handed coiled coil
tetramer. Nat Struct Biol 7(9):772-776.

Stowell, M.H., B. Marks, P. Wigge, and H.T. McMahon. 1999. Nucleotide-dependent
conformational changes in dynamin: evidence for a mechanochemical molecular
spring. Nat Cell Biol 1(1):27-32.

Struck, D.K., D. Hoekstra, and R.E. Pagano. 1981. Use of resonance energy transfer to
monitor membrane fusion. Biochemistry 20(14):4093-4099.

239
Sweitzer, S.M., and J.E. Hinshaw. 1998. Dynamin undergoes a GTP-dependent
conformational change causing vesiculation. Cell 93(6):1021-1029.

Takei, K., V. Haucke, V. Slepnev, K. Farsad, M. Salazar, H. Chen, and P. De Camilli.
1998. Generation of coated intermediates of clathrin-mediated endocytosis on
protein-free liposomes. Cell 94(1):131-141.

Takei, K., V.I. Slepnev, V. Haucke, and P. De Camilli. 1999. Functional partnership
between amphiphysin and dynamin in clathrin-mediated endocytosis. Nat Cell
Biol 1(1):33-39.

Tanigawa, G., L. Orci, M. Amherdt, M. Ravazzola, J.B. Helms, and J.E. Rothman. 1993.
Hydrolysis of bound GTP by ARF protein triggers uncoating of Golgi-derived
COP-coated vesicles. J Cell Biol 123(6 Pt 1):1365-1371.

Tarricone, C., B. Xiao, N. Justin, P.A. Walker, K. Rittinger, S.J. Gamblin, and S.J.
Smerdon. 2001. The structural basis of Arfaptin-mediated cross-talk between Rac
and Arf signalling pathways. Nature 411(6834):215-219.

Terwilliger, T.C., and J. Berendzen. 1999. Automated MAD and MIR structure solution.
Acta Crystallogr D Biol Crystallogr 55(Pt 4):849-861.

Tofaris, G.K., and M.G. Spillantini. 2007. Physiological and pathological properties of
alpha-synuclein. Cell Mol Life Sci 64(17):2194-2201.

Tong, Y.Y., P.J. Lee, A. Albanese, N. Bhardwaj, M.A. Digman, H. Lu, E. Gratton, S. Y.-
K., and W. Cho. submitted 2009. Molecular basis of the potent membrane
remodeling activity of the epsin 1 ENTH domain. Biol. Chem.

Trexler, A., and E. Rhoades. 2009. -Synuclein binds large unilamellar vesicles as an
extended helix. Biochemistry.

Ulmer, T.S., A. Bax, N.B. Cole, and R.L. Nussbaum. 2005. Structure and dynamics of
micelle-bound human alpha-synuclein. J Biol Chem 280(10):9595-9603.

van den Brink-van der Laan, E., J.A. Killian, and B. de Kruijff. 2004. Nonbilayer lipids
affect peripheral and integral membrane proteins via changes in the lateral
pressure profile. Biochim Biophys Acta 1666(1-2):275-288.

Vekrellis, K., H.J. Rideout, and L. Stefanis. 2004. Neurobiology of alpha-synuclein. Mol
Neurobiol 30(1):1-21.

Verstreken, P., O. Kjaerulff, T.E. Lloyd, R. Atkinson, Y. Zhou, I.A. Meinertzhagen, and
H.J. Bellen. 2002. Endophilin mutations block clathrin-mediated endocytosis but
not neurotransmitter release. Cell 109(1):101-112.
240
Weil, J.A., J.R. Bolton, and J.E. Wertz. 1994. Electron Paramagnetic Resonance:
Elementary Theory and Practical Applications. John Wiley & Sons, New York.

Weinreb, P.H., W. Zhen, A.W. Poon, K.A. Conway, and P.T. Lansbury, Jr. 1996. NACP,
a protein implicated in Alzheimer's disease and learning, is natively unfolded.
Biochemistry 35(43):13709-13715.

Weiss, M. 2001. Global indicators of X-ray data quality. . J.Appl.Cryst. 34:130-135.

Weiss, M., and R. Hilgenfeld. 1997. On the use of the merging R factor as a quality
indicator for X-ray data. J.Appl.Cryst. 30:203-205.

Weissenhorn, W. 2005. Crystal structure of the endophilin-A1 BAR domain. J Mol Biol
351(3):653-661.

Wendland, B. 2002. Epsins: adaptors in endocytosis? Nat Rev Mol Cell Biol 3(12):971-
977.

Whitmore, L., and B.A. Wallace. 2004. DICHROWEB, an online server for protein
secondary structure analyses from circular dichroism spectroscopic data. Nucleic
Acids Res 32(Web Server issue):W668-673.

Wieland, I., C. Weidner, R. Ciccone, E. Lapi, D. McDonald-McGinn, W. Kress, S.
Jakubiczka, H. Collmann, O. Zuffardi, E. Zackai, and P. Wieacker. 2007.
Contiguous gene deletions involving EFNB1, OPHN1, PJA1 and EDA in patients
with craniofrontonasal syndrome. Clin Genet 72(6):506-516.

Wigge, P., K. Kohler, Y. Vallis, C.A. Doyle, D. Owen, S.P. Hunt, and H.T. McMahon.
1997. Amphiphysin heterodimers: potential role in clathrin-mediated endocytosis.
Mol Biol Cell 8(10):2003-2015.

Yin, Y., A. Arkhipov, and K. Schulten. 2009. Simulations of membrane tubulation by
lattices of amphiphysin N-BAR domains. Structure 17(6):882-892.

Zhang, B., and A.C. Zelhof. 2002. Amphiphysins: raising the BAR for synaptic vesicle
recycling and membrane dynamics. Bin-Amphiphysin-Rvsp. Traffic 3(7):452-
460.

Zhang, Y., and J. Skolnick. 2005. TM-align: a protein structure alignment algorithm
based on the TM-score. Nucleic Acids Res 33(7):2302-2309.

Zhu, M., J. Li, and A.L. Fink. 2003. The association of alpha-synuclein with membranes
affects bilayer structure, stability, and fibril formation. J Biol Chem
278(41):40186-40197.

241
Zimmerberg, J. 2000. Are the curves in all the right places? Traffic 1(4):366-368.

Zimmerberg, J., and M.M. Kozlov. 2006. How proteins produce cellular membrane
curvature. Nat Rev Mol Cell Biol 7(1):9-19.

Zimmerberg, J., and S. McLaughlin. 2004. Membrane curvature: how BAR domains
bend bilayers. Curr Biol 14(6):R250-252.

Asset Metadata

Creator Jao, Christine Chua (author)

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

Contributor Electronically uploaded by the author (provenance)

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Biochemistry and Molecular Biology

Degree Conferral Date 2010-05

Publication Date 02/12/2010

Defense Date 11/19/2009

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

Tag curvature,electron paramagnetic resonance,membrane,oai:digitallibrary.usc.edu:usctheses,OAI-PMH Harvest,site-directed spin labeling

Language English

Advisor Langen, Ralf (committee chair), Chan, Jonah R. (committee member), Chen, Jeannie (committee member), Haworth, Ian S. (committee member), UImer, Tobias (committee member)

Creator Email ccjao@yahoo.com,cjao@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2849

Unique identifier UC1178892

Identifier etd-Jao-3462 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-299240 (legacy record id),usctheses-m2849 (legacy record id)

Legacy Identifier etd-Jao-3462.pdf

Dmrecord 299240

Document Type Dissertation

Rights Jao, Christine Chua

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email uscdl@usc.edu

Abstract (if available)

Abstract Control and regulation of membrane curvature play important roles in membrane trafficking and remodeling events. These processes are mediated by proteins that can sense and/or induce membrane curvature. The focus of my thesis is to understand the underlying molecular mechanisms that enable proteins to remodel membranes. Structural and biophysical studies were performed on the curvature-inducing proteins epsin (an ENTH protein) and endophilin (a BAR protein), both involved in membrane remodeling during endocytosis. Membrane interaction of α-synuclein, a curvature sensor, was also studied.