CHARACTERIZATION AND CRYSTALLIZATION OF TRANSLOCATOR PROTEIN 18 kDa
(TSPO) FROM RHODOBACTER SPHAEROIDES

By
Fei Li

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Biochemistry and Molecular Biology--Doctor of Philosophy
2014

CHARACTERIZATION AND CRYSTALLIZATION OF TRANSLOCATOR PROTEIN
18 kDa (TSPO) FROM RHODOBACTER SPHAEROIDES

By
Fei Li
Translocator protein 18 kDa (TSPO), previously known as the peripheral benzodiazepine
receptor (PBR), is a mitochondrial outer membrane protein that is a continuing research focus
due to its importance in cholesterol and porphyrin transport and apoptotic signaling. Dysfunction
of TSPO has been shown to relate to various diseases, including metastatic cancer, inflammation,
Alzheimer, and Parkinson disease. Functions of TSPO have been investigated but the molecular
details remain an enigma, due in part to two major bottlenecks: 1) the lack of atomic resolution
structural information; and 2) conflicting data from both in vivo and in vitro characterization.
The Rhodobacter sphaeroides homolog of TSPO (RsTSPO) was discovered as an oxygen
sensor regulating photosynthesis and respiration in this bacterium and shares 34 % sequence
identity with the human protein. In addition, the knock-out phenotype of R. sphaeroides can be
functionally complemented by the rat homolog, confirming the use of RsTSPO as a valid model
system for the mammalian protein. In this study, we have successfully established an expression,
purification, and characterization methodology for the RsTSPO that allowed us to investigate
RsTSPO with molecular detail. R. sphaeroides is also an highly successful system for producing
crystal structures of membrane proteins. A crystallization strategy was established and optimized
with RsTSPO that has resulted in the first high resolution crystal structure of a TSPO family
protein.
A sensitive tryptophan fluorescence quenching assay was used to characterize the binding
of purified RsTSPO with various ligands. Novel ligands of TSPO previously shown to affect

apoptosis were identified and support a role of TSPO in the regulation of the apoptosis pathway.
Detailed characterization of ligand binding with RsTSPO and mutants in vitro combined with
computational modeling of RsTSPO structure and analysis of previous mutagenesis data led to a
model of TSPO-ligand interaction.
A cholesterol binding enhancement motif was identified in TSPO that is highly conserved
within the mammalian proteins and is able to account for the 1000 fold different binding affinity
of RsTSPO compared to the mammalian homologs. This motif is also observed in other
membrane proteins interacting with cholesterol, highlighting the potential general importance of
this enhancement motif in the regulation of cholesterol binding in mammals. In addition,
residues within the same motif were identified by evolutionary covariance analysis as playing a
critical role in the stability of TSPO proteins, providing useful information for structural analysis
of TSPO.
Extensive screening and optimization of crystallization strategies for RsTSPO were carried
out with both the vapor diffusion and the lipidic cubic phase (LCP) methods. Successful
crystallization and structure determination of RsTSPO with the LCP method not only provided
the long-awaited structural information to facilitate interpretation of biochemical data on TSPO,
but it also provides unique and valuable knowledge to guide crystallization of other TSPO family
proteins, as well as similar small α helical membrane proteins.

ACKNOWLEDGEMENTS

First and foremost, I would like to express my sincerest gratitude to my adviser Dr. Shelagh
Ferguson-Miller. I thank her for the trust on this challenging project and her unconditional
support and encouragement throughout the years. Her enthusiasm for science, keen intelligence,
and determination has always inspired and motivated me. I am very fortunate to work with her,
to learn all the things about research and life from her, and to have her as my role model.
I am deeply grateful to my guidance committee members, Dr. Michael Garavito, Dr. Robert
Hausinger, Dr. John LaPres, and Dr. Kathleen Gallo, for their advice and support. I learned so
much from them in every conversation. Special thanks to Dr. Michael Garavito for all the
education and training on membrane protein crystallography and endless hours spent with me
before and after every synchrotron trip.
I would like to thank past and present members of the Ferguson-Miller lab for their help and
friendship. Working with them at MSU is such a wonderful memory I will always cherish: Dr.
Carrie Hiser who taught me all the molecular biology and reviewed all my manuscripts; Dr. Jian
Liu who froze all my tiny crystals and went with me on countless synchrotron trips; also Dr.
Leann Buhrow, Dr. Denise Mills, and Dr. Ling Qin. I also own many thanks to my talented
undergraduate assistant, Lance Valls, for getting all those nice binding curves.
I would like to thank Dr. John Wang, Dr. Jon Kaguni and Dr. Thomas Sharkey for taking me
into the BMB program; Dr. Laurie Kaguni for support and encouragement; Dr. Neil Bowlby for
teaching me biochemistry and trouble shooting wherever I need his help; Dr. Denis Proshlyakov,
Dr. Jian Hu, Dr. Kristin Parent and Dr. Michael Feig for suggestions and discussions; and Dr.
Kaillathe (Pappan) Padmanabhan for always being able to make my programs run smoothly.

iv

This project gave me great opportunities to work with various collaborators and learn new
techniques from them. We are indebted to all of them for their constant support and consultation
that made this project fruitful: Dr. Robert Stroud, Dr. Martin Caffrey, Dr. Brandon Ruotolo, Dr.
Philip Andrews, Dr. Jens Meiler, and their group members.
I thank staff scientists at LS-CAT and GM/CA-CAT at Advanced Photon Source, Argonne
National Laboratory, particularly Dr. Craig Ogata, Dr. Ruslan “Nukri” Sanishvili, Dr. Michael
Becker, Dr. Nagarajan Venugopalan, and Mr. Stephen Corcoran. Thank you all for being the
great hosts and teachers. I would also thank Dr. Janet Smith for her advice on the phasing
strategy and Ms. Sheila Trznadel for always getting us the precious beamtime.
Finally, I cannot thank my husband Dr. Yi Zheng, also great colleague from the Garavito lab,
enough. I appreciate him for taking me on my first synchrotron trip and patiently teaching me
protein crystallography ever since. His love, patience, understanding, and encouragements are
my strongest support to get through all the difficulties. Many thanks also go to my parents, other
family, and friends for their love and friendships.

v

TABLE OF CONTENTS

LIST OF TABLES…………………………………………………………………………….….ix
LIST OF FIGURES ……………………………………………………………………...……….x
KEY TO ABBREVIATIONS…………………...……………………………………...……….xiii
CHAPTER 1 ................................................................................................................................... 1
An introduction to Translocator Protein 18 kDa (TSPO) and membrane protein crystallography 1
Background on Translocator Protein 18 kDa (TSPO) .................................................................... 2
Discovery of the peripheral benzodiazepine receptor (PBR) .................................................. 2
Function of TSPO as a mitochondrial transporter ................................................................... 3
Function of TSPO in steroidogenic tissues .............................................................................. 3
Function of TSPO in porphyrin metabolic pathways .............................................................. 6
TSPO as a regulator of mitochondrial permeability transition pore (MPTP) and
mitochondrial homeostasis....................................................................................................... 9
TSPO in pathological conditions ............................................................................................11
TSPO in cancer .................................................................................................................. 11
TSPO in ischemia/reperfusion (I/R) injury ........................................................................ 12
Regulation of I/R injury by TSPO through MPTP ............................................................ 13
Targeting TSPO in diagnosis and treatment of neural diseases ......................................... 16
Application of TSPO ligands in neurological diseases ...................................................... 18
TSPO in Rhodobacter sphaeroides ........................................................................................ 19
Protein crystallography ................................................................................................................. 24
Importance of membrane protein structures .......................................................................... 24
Challenges and advances in membrane protein crystallography ........................................... 25
Production of membrane proteins ...................................................................................... 25
Purification of membrane proteins..................................................................................... 26
Crystallization of membrane proteins ................................................................................ 27
Importance of understanding the structure-function relationship of TSPO .................................. 28
REFERENCES ............................................................................................................................. 29
CHAPTER 2 ................................................................................................................................. 41
Characterization and modeling of the oligomeric state and ligand binding behavior of purified
Translocator Protein 18 kDa (TSPO) from Rhodobacter sphaeroides ......................................... 41
INTRODUCTION ........................................................................................................................ 42
MATERIALS AND METHODS .................................................................................................. 45
Materials ................................................................................................................................ 45
Cloning................................................................................................................................... 46
Protein expression and purification ....................................................................................... 46
Light scattering measurements of oligomeric state of purified RsTSPO............................... 48
Tryptophan fluorescence quenching binding assay ............................................................... 48
Modeling of the RsTSPO dimer ............................................................................................. 50
vi

RESULTS...................................................................................................................................... 51
RsTSPO forms a dimer in vitro .............................................................................................. 51
Tryptophan fluorescence quenching assay as a sensitive measure of Kd ............................... 58
Effect of detergent conditions on the binding assay .............................................................. 60
RsTSPO binds PK11195 and cholesterol at micro-molar affinity ......................................... 62
Competition binding assay of PpIX, PK11195 and cholesterol ............................................. 65
Compounds affecting apoptosis bind to RsTSPO at micro-molar affinity ............................ 70
Models of the RsTSPO dimer ................................................................................................ 72
DISCUSSION ............................................................................................................................... 78
Evidence relating to the role of the dimer .............................................................................. 78
Effect of detergent on the structure and binding of RsTSPO ................................................. 80
Using the TSPO dimer model to predict the binding sites for PpIX, cholesterol, and
PK11195 ................................................................................................................................. 81
TSPO as a mediator of the regulation of MPTP and apoptosis. ............................................ 86
CONCLUSIONS........................................................................................................................... 86
REFERENCES ............................................................................................................................. 88
CHAPTER 3 ................................................................................................................................. 97
Identification of a key cholesterol binding enhancement motif in Translocator Protein 18 kDa . 97
INTRODUCTION ........................................................................................................................ 98
MATERIAL AND METHODS ..................................................................................................... 99
Cloning, expression and purification of protein. ................................................................... 99
Tryptophan fluorescence quenching binding assay.............................................................. 100
TLC analysis of the purified RsTSPO_LAF sample............................................................ 100
Covariance analysis with PsiCOV. ...................................................................................... 100
Phylogenetic analysis of the enhancement motif. ................................................................ 101
Survey of the enhancement motif in human membrane protein database. .......................... 101
RESULTS AND DISCUSSION.................................................................................................. 101
Rationale and design of mutations in RsTSPO. ................................................................... 101
RsTSPO-LAF exhibits a tendency to aggregate when concentrated, but CHS can rescue it
from precipitation................................................................................................................. 103
RsTSPO-LAF binds cholesterol with nM affinity................................................................ 106
Binding behavior of RsTSPO-LAF with PK11195 and PpIX. ............................................ 108
Interactions between helix IV and V are critical for protein stability. ..................................110
Evolution of the cholesterol binding enhancement motif. ....................................................112
The cholesterol binding enhancement motif exists in other membrane proteins and may
indicate the presence of high affinity cholesterol binding sites. ...........................................115
SUMMARY .................................................................................................................................119
REFERENCES ........................................................................................................................... 120
CHAPTER 4 ............................................................................................................................... 127
Crystallization strategies of small α helical integral membrane proteins: a case study of
Translocator Protein 18 kDa from Rhodobacter sphaeroides (RsTSPO) ................................... 127
INTRODUCTION ...................................................................................................................... 128
MATERIAL AND METHODS ................................................................................................... 130
Materials .............................................................................................................................. 130
vii

Methods................................................................................................................................ 131
Protein expression and purification ................................................................................. 131
Determination of detergent content of purified RsTSPO by thin layer chromatography
(TLC) and a colorimetric method .................................................................................... 132
Detergent exchange .......................................................................................................... 132
Screening for solubilizing detergent ................................................................................ 133
Screening and optimization of the purification detergent with SEC ............................... 134
Optimization of SEC conditions ...................................................................................... 134
Crystallization screening of RsTSPO with the vapor diffusion method .......................... 135
Crystallization screening of RsTSPO with the LCP method ........................................... 135
Optimization of crystal leads with the LCP method ........................................................ 136
Preparation of heavy metal derivative of the RsTSPO A139T crystal ............................ 136
Data collection and processing ........................................................................................ 137
Structure determination and refinement........................................................................... 138
RESULTS.................................................................................................................................... 138
Initial crystallization screening with RsTSPO purified in DM indicated inhibition of
crystallization by detergent .................................................................................................. 138
DM level in purified RsTSPO was successfully reduced and exchanged for smaller
detergents ............................................................................................................................. 143
RsTSPO was successfully extracted by several detergents .................................................. 149
Identification of the optimal detergent by size exclusion chromatography ......................... 151
Identification of the optimum purification procedure by SEC ............................................ 153
Results of crystallization and optimization with the vapor diffusion method ..................... 157
Initial screening with the LCP method ................................................................................ 159
Optimization of LCP crystallization conditions................................................................... 161
Structure determination and refinement............................................................................... 166
Structure of RsTSPO-WT and A139T mutant ..................................................................... 166
DISCUSSION ............................................................................................................................. 169
Importance of promoting type I crystal packing for RsTSPO with the LCP method .......... 169
Role of detergent in crystallization of RsTSPO ................................................................... 173
Common additives in vapor diffusion and LCP crystallization optimization ...................... 175
Obtaining suitible heavy metal derivatives of RsTSPO ....................................................... 176
Insights from the RsTSPO crystal structures ....................................................................... 176
SUMMARY ................................................................................................................................ 177
Manuscript describing the crystal structures of RsTSPO .................................................... 177
APPENDIX ................................................................................................................................. 179
Crystal structures of translocator protein 18 kDa (TSPO) from Rhodobacter sphaeroides ....... 180
REFERENCES ........................................................................................................................... 181
CHAPTER 5 ............................................................................................................................... 186
Perspective and future direction on structure-function studies of TSPO .................................... 186
Crystallization of RsTSPO-ligand complexes ............................................................................ 187
Mutagenesis of RsTSPO based on the crystal structure and the co-evolution analysis .............. 188
Characterization of the oligomeric state of RsTSPO and the stoichiometry of the endogenous
ligand by mass spectrometry....................................................................................................... 188
Expression, purification and crystallization of human TSPO ..................................................... 189
viii

LIST OF TABLES

Table 2.1. Molecular weight calculation from static light scattering measurement ..................... 54
Table 2.2. Structure and Kd of ligands to RsTSPOWT and W38C mutant ................................... 64
Table 3.1 Classification of human membrane proteins with the cholesterol binding enhancement
motif .............................................................................................................................................117
Table 4.1 High throughput small scale crystallization screening results of RsTSPO ................. 148

ix

LIST OF FIGURES

Figure 1.1 Proposed mechanism of cholesterol transport through the cholesterol transporter
complex ........................................................................................................................................... 5
Figure 1.2 Functions of TSPO as a mitochondrial porphyrin and cholesterol transporter. ............ 8
Figure 1. 3 Proposed regulatory effect of TSPO and TSPO ligands on MPTP. ............................ 10
Figure 1.4 Proposed model of interaction of TSPO in the I/R injury process. ............................. 15
Figure 1.5 Effect of TSPO expression and TSPO ligands. ........................................................... 17
Figure 1.6 Sequence alignment of TSPO proteins. ....................................................................... 22
Figure 2.1. Purification and determination of the molecular weight of RsTSPO ......................... 53
Figure 2.2. Investigation of the oligomeric state of purified RsTSPO by SEC ............................ 54
Figure 2.3. Expression of TSPO fusion proteins .......................................................................... 57
Figure 2.4. Binding of PpIX, ALA, and ursodeoxycholate to RsTSPO and PK11195, PpIX to
HsTSPO. ....................................................................................................................................... 59
Figure 2.5. Purification of HsTSPO.............................................................................................. 59
Figure 2.6. Binding curves of PK11195 with RsTSPO in different detergent conditions as
described in Methods .................................................................................................................... 61
Figure 2.7. Binding of PpIX, PK11195 and cholesterol to purified RsTSPO (WT) and
RsTSPOW38C (W38C). ............................................................................................................... 63
Figure 2.8. Tryptophan fluorescence quenching behavior of PpIX and PK11195. ...................... 67
Figure 2.9. Competition binding studies between PpIX, PK11195 and cholesterol to RsTSPO .. 69
Figure 2.10. Binding curves of compounds affecting apoptosis: curcumin, retinoic acid, gossypol,
and bilirubin .................................................................................................................................. 71
Figure 2.11. Rosetta energy score shows that the top 7 models are significantly better than the
rest ................................................................................................................................................. 73
Figure 2.12. Top two models of a RsTSPO dimer ........................................................................ 74

x

Figure 2.13. Top 3-5 scoring models of RsTSPO dimer ............................................................... 75
Figure 2.14. Oligomerization of RsTSPO. .................................................................................... 77
Figure 2.15. Proposed model of binding sites of PpIX, PK11195 and cholesterol on RsTSPO ... 83
Figure 3.1. Location of the triple mutation of RsTSPO in relation to the CRAC sequence. ...... 102
Figure 3.2. Purification of RsTSPO-LAF and the rescue with cholesterol hemisuccinate. ........ 105
Figure 3.3 RsTSPO-LAF binds to cholesterol at nM affinity. .................................................... 107
Figure 3.4. Binding of RsTSPO-LAF with PK11195 and PpIX. ................................................ 109
Figure 3.5. Topology diagram of RsTSPO showing contact pairs predicted by covariance analysis.
......................................................................................................................................................111
Figure 3.6. Evolution of TSPO and correlation with LAF motif. ................................................114
Figure 4.1 RsTSPO was purified by metal affinity and size exclusion chromatography (SEC) and
analyzed by SDS PAGE. ............................................................................................................. 140
Figure 4.2 Determination of detergent concentration of purified RsTSPO samples. ................. 142
Figure 4.3 Detergent exchange results determined by TLC. ...................................................... 146
Figure 4.4 Detergent solubilization screening results. ................................................................ 150
Figure 4.5 SEC profiles of RsTSPO10ht purified in different detergents. ................................. 152
Figure 4.6. SEC profiles of RsTSPO10ht purified in (A) DM alone, (B) C10E5 alone, and (C)
DM/C10E5 mixture. ................................................................................................................... 154
Figure 4.7. SEC profiles for testing effects of glycerol and histag ............................................. 156
Figure 4.8 Representative RsTSPO crystals grown with the vapor diffusion method and their 158
Figure 4.9. Representative pictures of RsTSPO protein crystals grown in the LCP setup after 3
weeks........................................................................................................................................... 160
Figure 4.10. Schematic representation of optimization strategies for crystallization of RsTSPO by
the LCP method. ......................................................................................................................... 162
Figure 4.11. RsTSPO-PK11195 crystals ..................................................................................... 165
Figure 4.12. Crystal structure of RsTSPO-A139T. ..................................................................... 168
xi

Figure 4.13 Crystal packing of wildtype RsTSPO crystals grown in the LCP condition. .......... 170
Figure 4.14 Cartoon illustrations of different crystal types ........................................................ 172

xii

KEY TO ABBREVIATIONS

A. U.

asymmetric unit

βME

β-mercaptoethanol

BCA

bicinchroninic acid

BSA

bovine serum albumin

CV

column volume

CMC

critical micelle concentration

DDM

dodecyl maltoside

DM

decyl maltoside

EDTA

ethylenediaminetetraacetic acid

FPLC

fast protein liquid chromatography

kD

kilodalton

LCP

lipidic cubic phase

LDAO

lauryl dimethylamine n-oxide

LAPAO

3-dodecylamido-N,N′-dimethylpropyl amine oxide

MAG

monoacylglycerol

MES

2-(n-morpholino)ethanesulfonic acid

MALDI

matrix-assisted laser desorption ionization mass spectrometry

MW

molecular weight

MWCO

molecular weight cut-off

xiii

Ni-NTA

nickel-nitrilotriacetic acid

NMR

nuclear magnetic resonance

PCR

polymerase chain reaction

PEG

polyethylene glycol

Rs

Rhodobacter sphaeroides

SDS-PAGE

sodium dodecyl sulfate polyacrylamide gel electrophoresis

SeMet

selenomethionyl

TLC

thin layer chromatography

TM

transmembrane

Tris

tris (hydroxymethyl) aminomethane

TSPO

Translocator protein 18 kDa

WT

wild type

xiv

CHAPTER 1

An introduction to Translocator Protein 18 kDa (TSPO) and membrane protein
crystallography

1

Background on Translocator Protein 18 kDa (TSPO)
Discovery of the peripheral benzodiazepine receptor (PBR)
The peripheral benzodiazepine receptor (PBR) was first discovered by serendipity in 1977 in
an experiment searching for the high affinity binding site for the widely prescribed anxiolytic
and sedative hypnotic benzodiazepine drug Diazepam [1]. In addition to its target in the central
nervous system (CNS), another nanomolar affinity binding site was also found in a variety of
peripheral nervous system (PNS) [2]. Due to its abundance in peripheral tissues as well as to
distinguish it from the central benzodiazepine receptor (CBR), which is now known as the ligand
gated GABAA receptor, this peripheral binding protein was named the “peripheral
benzodiazepine receptor” (PBR). Later, another class of ligands, represented by the isoquinoline
carboxamide PK11195, was discovered to bind to PBR exclusively, with high affinity detected
consistently across species while benzodiazepines bind to PBR and CBR with varied affinities.
Therefore, PK11195 is now often used as the diagnostic ligand for PBR.
In 2006, PBR was renamed as Translocator Protein 18 kDa (TSPO) [3] to better represent
the emerging understanding of a family of highly conserved proteins found in Archaea to plants
and humans [4]. TSPO family proteins are integral membrane proteins with 5 predicted
transmembrane helices [3]. Functions of TSPO proteins span from regulation of steroid hormone
and heme biosynthesis, to regulation of apoptosis and a role in cardiovascular [4] and
neurological

disease

[5].

Over

the

past

30

years,

extensive

studies

of

this

developmentally-required protein have been carried out and resulted in significant advances in
our knowledge of TSPO. A brief review of the current understanding of the structure and
function of TSPO, primarily based on studies from mammalian and Rhodobacter sphaeroides
systems, is provided here.

2

Function of TSPO as a mitochondrial transporter
TSPO is located in the outer membrane of mitochondria and highly concentrated at contact
sites of the inner and outer mitochondrial membranes [6-8]. It has been purified in a complex
with the outer membrane-located voltage dependent anion channel (VDAC) and adenine
nucleotide translocator (ANT) in the inner membrane [9], but the significance of these
associations remains controversial [10]. Additionally, TSPO is reported to transport cholesterol
[11] and porphyrins [12] as endogenous ligands and to do so as part of a transport complex
involving proteins from both the inner and outer mitochondrial membranes. Although these
functions remain to be fully characterized, extensive investigations have provided considerable
understanding of the functions of TSPO under various physiological and pathological conditions.
Function of TSPO in steroidogenic tissues
TSPO is found in almost all tissues examined, among which it is most highly expressed in
steroid hormone producing tissues such as adrenal glands and kidney [13]. A major function of
TSPO in steroidogenic tissues is proposed to be transporting cholesterol across the mitochondrial
membrane to be further converted into steroid hormones [14].
Cholesterol is the sole precursor of steroid hormones and bile salts. Several pathways have
been suggested to traffic cholesterol to mitochondria for steroidogenesis [15,16]. The sources of
cholesterol in the cell can be divided into de novo synthesis in the endoplasmic reticulum (ER) or
transfer from external cholesterol sources, such as the cholesterol-containing low density
lipoproteins (LDL) or the high density lipoprotein (HDL). Free cholesterol is transported to
mitochondria through two distinct pathways: either via a cholesterol binding protein that
interacts with mitochondrial outer membrane proteins, such as TSPO, or directly from the ER
through membrane fusion. A receiver protein complex containing TSPO located in the outer

3

mitochondrial membrane (OMM) then transports cholesterol across the mitochondrial
membranes to be further processed by the cytochrome P450 cholesterol side chain cleavage
enzyme (CYP11A1). The transport complex was proposed to include several proteins related to
TSPO, including the steroidogenic acute regulatory (StAR) protein, TSPO-associated protein 7
(PAP7, ACBD3), and protein kinase A regulatory subunit 1a (PKAR1A). However, the details of
these interactions are not yet understood.
The first and the rate-limiting step of steroid biosynthesis is also the transport of cholesterol
from OMM to the inner mitochondrial membrane (IMM) [17,18]. TSPO has been proposed to be
the cholesterol transporter in this critical step [11,19-21], but other proteins are also suggested to
play a role in this process. A model of this translocation protein complex has been proposed by
Papadopoulos and colleagues [15] (Figure 1.1). Briefly, cholesterol is relayed to TSPO by
several proteins and eventually the StAR protein upon hormone stimulation. The mechanism of
transport of cholesterol by TSPO is not fully understood. One model proposes that cholesterol is
transported through the interface of TSPO and the VDAC [15,22]. The transfer complex is also
proposed to contain the ANT in the IMM although the exact function is unknown. Translocated
cholesterol is then cleaved and converted into pregnenolone by the CYP11A1 associated with the
matrix side of the IMM. Pregnenolone is further used to generate varieties of steroids in
downstream pathways.

4

Figure 1.1 Proposed mechanism of cholesterol transport through the cholesterol
transporter complex.
Cholesterol is shown as red pentagons while the converted pregnenolone is shown as a green
diamond. Proteins involved in this process are shown as cartoons at the proposed location in
mitochondria. Figure adapted from Rone et al. [15].

5

Function of TSPO in porphyrin metabolic pathways
In mammals, the biosynthesis of heme and porphyrins starts from the condensation of
glycine and succinyl-CoA into δ-aminolevulinic acid (ALA) in mitochondria. ALA is exported to
the cytosol and subsequently converted to coproporphyrinogen III after several enzymatic
reactions. Coproporphyrinogen III is then transported back into mitochondria to be further
converted to various porphyrin molecules including protoporphyrin IX (PpIX) and heme [23]
(Figure 1.2). Heme and hemin were originally found to inhibit the binding of PK11195 and
Ro5-4864 to TSPO in human blood and several rat organs [8]. Subsequent studies confirmed that
TSPO plays a role in the heme biosynthesis pathway [24] and a homolog of TSPO, TSPO from
Rhodobacter sphaeroides (RsTSPO), was identified to regulate the formation of photosynthetic
complexes [25]. Therefore TSPO was proposed to play a role in transporting porphyrin
precursors for both respiratory and photosynthetic functions across the membranes as well as
regulating the heme biosynthesis pathway [3].
TSPO has been reported to export porphyrins both from R. sphaeroides and from
mammalian mitochondria [25-27]. In both systems, TSPO is proposed to be part of a stress
response mechanism. In mammalian systems, both oxygen and light stress were reported to
induce polymerization and activate the transport function of TSPO [28,29] and therefore to
reduce the accumulation of porphyrin as well as the generation of reactive oxygen species (ROS)
[30]. In the R. sphaeroides system, the regulation by TSPO of the export of pigments produced in
the heme/bacteriochlorophyll/carotenoid biosynthesis pathways is proposed, thereby controlling
expression of related genes and the switch between photosynthesis and respiration upon
environmental cues such as light and oxygen stress. However, the exact function of TSPO in
relation to porphyrin transport, including both import and export under various stress conditions,

6

is still not well understood.

7

Figure 1.2 Functions of TSPO as a mitochondrial porphyrin and cholesterol transporter.
Cholesterol and porphyrin transport across mitochondrial membranes related to TSPO functions
is shown. In the cholesterol transport pathway, cholesterol is imported through the cholesterol
transport complex involving TSPO. Pregnenolone is produced by cleavage of the side-chain of
cholesterol immediately after transport and is exported to the ER for further processing. In the
porphyrin transport pathway, TSPO could potentially act both as the importer and/or exporter of
heme precursors or breakdown products, thus playing a role in the regulation of reactive oxygen
species (ROS).

8

TSPO as a regulator of mitochondrial permeability transition pore (MPTP) and
mitochondrial homeostasis
Mitochondria play important roles in the proper function of cells, not only by providing most
of the energy needed for various biological processes, but also by participating in the execution
of the cell by regulating cell death through apoptosis and/or necrosis [31,32]. Apoptosis refers to
a process of controlled cell death through the activation of the apoptosis pathway that leads to the
recycling of the cell with minimum damage to the tissue. Current understanding of apoptosis
puts mitochondria at center stage, although signaling pathways that do not involve mitochondria
have also been characterized [33]. The initiation of apoptosis is usually preceded by a loss of the
mitochondrial membrane potential (ΔΨm), which is proposed to be regulated by the MPTP, and
followed by the release of cytochrome c and apoptosis inducing factor (AIF) from the
mitochondrial intermembrane space [34]. MPTP is a multi-protein complex formed at the contact
sites of the inner and outer mitochondrial membranes. It was originally proposed to be mainly
composed of VDAC located in the outer membrane and ANT in the inner membrane [35,36].
However, an important role of ATP synthase has been identified lately [37]; thus, the exact
components of MPTP are still under debate. But together with other proteins, TSPO is proposed
to regulate the MPTP, since TSPO forms a 70 kDa complex with VDAC and ANT [9] and
various TSPO ligands have been shown to regulate apoptosis [38] (Figure 1.3). In addition, it has
been suggested that the diverse effects of TSPO ligands on multiple processes are mediated by
the fine interplay of TSPO with the MPTP complex and its interacting proteins, through which
the homeostasis of the mitochondria is altered, as discussed in a recent issue of Current
Molecular Medicine on the broad implications of TSPO in mitochondrial biology [39]. However,
it is currently still unclear how MPTP and apoptosis are regulated by TSPO.

9

Figure 1. 3 Proposed regulatory effect of TSPO and TSPO ligands on MPTP.
One model of the MPTP complex is shown in the cartoon with VDAC and ANT proposed to be
the major structural components of the pore. TSPO as well as other regulatory proteins are
associated with VDAC and ANT. TSPO ligands, including diazepines and PK11195, regulate the
function of MPTP, as well as the mitochondrial homeostasis, through binding to TSPO and
regulating the interaction of TSPO with the MPTP complex. (CK, creatine kinase; HK,
hexokinase; CypD, cyclosporine D; Bcl-2: B-cell lymphoma 2)

10

TSPO in pathological conditions
As mentioned above, TSPO is ubiquitously expressed in all tissues and has been proposed to
regulate the homeostasis of the whole cell through the essential MPTP pathway. Abnormal
expression and dysfunction of TSPO therefore would potentially cause severe diseases. Indeed,
high expression of TSPO is observed in various pathological conditions, including cancer, brain
injury, neuronal degeneration, and ischemia-reperfusion injury [13]. Efforts have been made to
understand the function of TSPO in disease states, to use TSPO as a diagnostic tool for early
identification, and to develop therapeutic treatments targeting TSPO. Although the mechanisms
of these diseases, as well as the involvement of TSPO and TSPO ligands during the disease
processes, are still not understood, a brief review is provided here to summarize current
understanding of the roles TSPO may play in these complex diseases and potential treatments
utilizing TSPO ligands.
TSPO in cancer
TSPO is highly expressed in cancerous tissues of the breast, ovary, colon, prostate, and brain
[40-45], and TSPO ligands PK11195 and Ro5-4864 have been shown to augment the apoptosis
of human tumor cells (MCF-7 [46], C6 glioma cells [47], and human prostate cancer cells [48])
when given along with various chemotherapeutic agents [49]. In addition, the level of TSPO
up-regulation is correlated with the aggressiveness of breast cancer [43]. Overexpression of
TSPO in MCF-7 cells increased proliferation, whereas silencing of TSPO in MDA-MB-231 cells
led to a decrease in proliferation [50]. On the other hand, accumulation of cholesterol has been
proposed to be a common phenotype of cancer cells since the original reports of cholesterol
crystals observed in solid tumors [51]. Therefore, Batarseh et al. [13] proposed that the higher
level of TSPO in tumor cells was the driving force for lipid transport to mitochondria and nuclei

11

and the accumulated cholesterol in the membranes of tumor cells might also influence the degree
of membrane fluidity, which in turn plays a role in activation of signaling pathways involved in
cancer cell proliferation [52]. Considering the role of TSPO in porphyrin transport, it was also
suggested that overexpression of TSPO in cancer cells increased their tolerance to ROS [53,54]
and enhanced their metastasis through ROS activated downstream pathways [55].
Taken together, it is reasonable to propose that overexpression of TSPO in cancer cells
supports their survival by both increasing the cell viability as well as suppressing cell death [10].
However, controversies still remain in the field and the mechanisms underlying
TSPO/mitochondria regulation on cancer metastasis remain unclear.
TSPO in ischemia/reperfusion (I/R) injury
Cardiovascular disease is the leading cause of death in the United States and a serious hazard
to public health worldwide. It refers to any disease that affects the cardiovascular system,
principally cardiac disease, vascular diseases of the brain and kidney, and peripheral arterial
disease [56]. Among all the acute and chronic cardiovascular diseases, ischemic heart
disease (IHD), or myocardial ischemia, is the most common cause of death in most western
countries and a major cause of hospital admissions. Ischemia refers to a state where the blood
supply is restricted, causing a shortage of oxygen and glucose needed for cellular metabolism
and resulting in damage or dysfunction of tissue. As a working heart requires a significant
amount of nutrients supplied by blood, restriction soon leads to severe consequences including
+

2+

decreases in ATP and pH, increased anaerobic glycolysis, and increases in Na and Ca

concentrations caused by the inhibition of oxidative phosphorylation. Reperfusion refers to the
restoration of blood flow to an organ or tissue after the ischemia. Ischemia/reperfusion injury
(I/R injury) is caused by augmented cell death when oxygen (O2) is reintroduced following
12

ischemia and frequently results in depressed myocardial function. The result may be lethal or
cause morbidities that prolong hospitalization and increase costs even after a successful
treatment of cardiovascular disease [57].
Induction of the opening of the MPTP is commonly accepted as the cause of cell death
during I/R injury, although the mechanism is still not well understood [58-60]. During normal
conditions, mitochondria in cardiac tissue mainly rely on the oxidative phosphorylation pathway
to generate ATP needed for muscle contraction. This process is highly efficient but also heavily
2+

O2 dependent. Mitochondrial calcium concentrations [Ca ]m play a critical role in regulating
oxidative phosphorylation during this process. During ischemia, when O2 supply is limited,
mitochondria in the myocytes can no longer produce enough ATP, which leads to a rapid
decrease in [ATP] and subsequent loss of [ADP], [AMP] and [Pi]. Metabolism will switch to
+

+

glycolysis, resulting in a decrease in pH and activation of the Na /H antiporter that eventually
2+

leads to an increase in mitochondrial [Ca ]m. When O2 is re-introduced, high levels of ROS are
generated when O2 meets the build up of a reducing in the electron transport chain (ETC)
2+

accumulated during ischemia. [Ca ]m is further increased as well. The overall effect is a
condition that favors opening of the MPTP. When uncontrolled, this massive opening of the
MPTP will lead to necrotic cell death.
Regulation of I/R injury by TSPO through MPTP
Much evidence suggests that the MPTP plays a critical role in regulating cell death through
apoptosis and/or necrosis. Low level opening of the MPTP facilitates transport of small
molecules into the mitochondrial matrix, leading to swollen mitochondrial inner membranes and
13

broken outer membranes. Release of apoptosis inducing factors, such as cytochrome c, from the
mitochondrial intermembrane space leads to activation of the apoptosis pathway and controlled
cell death through apoptosis. More dramatic opening of the MPTP further leads to complete
dissipation of the membrane potential, which impacts the bioenergetics of the cell and leads to
cell death through necrosis [61]. Conditions frequently experienced by the heart and brain during
reperfusion after a period of ischemia promote the MPTP opening by Ca

2+

overload (Figure. 1.4).

As indicated in Figure 1.3 and Figure 1.4, TSPO ligands have been proposed to reduce I/R injury
by inhibiting the MPTP opening and possibly by reducing the production of ROS [62] during
reperfusion.
The important role of TSPO in I/R injury is indicated by a variety of studies that show
effects of its ligands on the process of cell death through regulation of apoptosis and necrosis.
The high affinity TSPO ligands TRO40303, SSR180575 and Ro5-4864 have myocardial
protective effects against I/R injury in a mouse model by inhibiting necrosis [63,64] and improve
cardiac functional recovery during the post-I/R period [38]. Other drugs, such as the widely used
anesthetic Propofol

®

and related compounds, also show cardio-protection [65] and preliminary

data [66] suggest that these effects involve TSPO. A number of proteins including myxoma viral
protein M11L and HIV-1 protein Vpr appear to inhibit apoptotic cell death of the host cell
through interaction with TSPO [67,68]. It is clear from these many studies that TSPO ligands
may have strong regulatory effects on cell death related to I/R injury. In addition, all studies so
far regarding the regulatory effects of TSPO ligands on cell death suggest that these effects are
mediated through the regulation of the MPTP.

14

Figure 1.4 Proposed model of interaction of TSPO in the I/R injury process.
MPTP opening is the central regulator of I/R injury. Overexpression of TSPO inhibits the
opening of the MPTP by direct interaction with the MPTP complex and/or reducing the ROS
level during reperfusion [38]. TSPO ligands demonstrate additional regulation of the MPTP
opening [63,64,69], thus I/R injury, through positive/negative regulation of TSPO.

15

Targeting TSPO in diagnosis and treatment of neural diseases
TSPO expression levels in both CNS and PNS are very low under normal healthy conditions
[70]. In the CNS, TSPO is usually expressed in microglia [71], reactive astrocytes [72] and
neurons [73], while it has been reported to be expressed in Schwann cells, macrophages and
neurons in the PNS [74-76] (Figure 1.5). Interestingly, expression of TSPO is markedly
increased under disease conditions [5] including Alzheimer disease [77], Parkinson disease
[78,79], Huntington disease [80], stroke [81], brain damage [82], and multiple sclerosis [83]. The
up-regulation of TSPO expression mainly occurs in the activated microglia, microphages, and
astrocytes [84,85], which represent the site of damage, inflammation and sites of neuronal loss
[79,86] that indicate neurodegenerative disease. In addition, the level of TSPO up-regulation is
correlated with the severity of the disease [80]. Taken together, expression of TSPO has been
used as a sensitive in vivo biomarker for diagnosis of various neurological diseases [73,77].

16

Figure 1.5 Effect of TSPO expression and TSPO ligands.
Injury to brain and peripheral neurons will cause increased expression levels of TSPO in these
tissues and the activation of astrocytes and microglia. TSPO ligands were shown to have
neuron-protective effects by increasing the expression level of growth factor and neurosteroids,
thus reducing inflammation and gliosis. Figure adapted from Rupprecht et al. [5].

17

The mechanism of how TSPO is involved in various neurological diseases and whether the
overexpression of TSPO is the cause or the outcome of the pathology are still not well
understood. Several possible explanations have been suggested [5,87]. Since the rate-limiting
step of steroid biosynthesis, including all neurosteroids, is the transport of cholesterol through
TSPO, it was proposed that TSPO regulates the brain function through regulation of neurosteroid
production. Papadopoulos et al. [88] proposed that cholesterol and steroid levels regulated by
TSPO play a role in the degeneration and regeneration of neurons while others also proposed that
neurosteroids synthesized downstream of TSPO may act as signaling molecules to other
receptors in the brain. For instance, TSPO has been shown to potentiate the GABAA receptor
through up-regulation of allopregnanolone production after inflammation [89,90]. On the other
hand, TSPO could also regulate the function and state of the brain through its effects on
apoptosis and mitochondrial and whole cell homeostasis via regulation of the MPTP. In fact,
Jayakumar et al. [91] reported that the TSPO ligands PK11195, Ro5-4864, and PpIX stimulated
ROS production in the CNS in a MPTP dependent manner. It is also highly possible that the
combination and balancing of above mechanisms result in the overall effect of TSPO in
neuropathology. Further investigations are certainly needed to better understand how TSPO
functions in various neurological diseases.
Application of TSPO ligands in neurological diseases
Although the mechanism of how TSPO functions in the neural systems is still largely
unknown, TSPO ligands have played critical roles in facilitating the diagnosis and treatment of
various neurological diseases. Due to the highly site specific overexpression of TSPO in areas of
brain damage and inflammation, TSPO ligands have been widely used in imaging [92]. Various
derivatives of PK11195 were developed to allow better imaging of brain injury as well as to

18

provide tools for investigating the mechanisms behind these neurological diseases [82,83,92].
For instance, a human single polymorphism mutation with lower affinity to PK11195 derivatives
was identified in PET studies [93] and was shown to be related to reduced pregnenolone
production [94] and higher incidence of bipolar disorder [95]. Following up on these findings, a
cholesterol binding enhancement motif was identified and characterized in vitro by mutational
analysis of RsTSPO. The results of these studies are reported in Chapter 3.
In addition, the therapeutic potential of these compounds has been explored. A number of
ligands, including PK11195 [75,96] and Ro5-4864 [97,98], have shown neuroprotective effects
[5] and several drugs targeting TSPO are in phase II and phase III clinical trials. Overall, our
current knowledge suggests that TSPO plays an important role in maintaining the normal
functions of brain and it is very promising to target TSPO for the treatment of various severe
neurological diseases.
TSPO in Rhodobacter sphaeroides
R. sphaeroides is a metabolically versatile photosynthetic proteobacterium capable of
growth under a wide variety of environmental conditions. It is also one of the closest free-living
relatives of the mitochondrion [99] and provides a genetically tractable model system for
examining proteins in molecular detail as well as their in vivo functionality. Importantly, it is a
very successful system for expressing and crystallizing membrane proteins. Several homologs of
important membrane protein complexes, such as cytochrome c oxidase (CcO) and the
photosynthetic reaction center, have been crystallized from R. sphaeroides.
TSPO was independently discovered in R. sphaeroides and R. capsulatus as the CrtK gene in
the carotenoid synthesis operon [100] and was known for its ability to regulate the switch
between photosynthesis and respiration in response to oxygen and light conditions [26,27]. It

19

was subsequently recognized as homologous to the mitochondrial PBR [25,101] and similarly
proposed to be involved in the transport of small molecules, such as porphyrin intermediates of
the heme and chlorophyll biosynthesis/degradation pathways [26,102].
TSPO in R. sphaeroides (RsTSPO) is an 18 kDa transmembrane protein localized to the
outer membrane of R. sphaeroides. RsTSPO shares a high sequence identity (34 %) with human
TSPO (Figure 1.6) and rat TSPO has been shown to substitute for RsTSPO in R. sphaeroides,
suggesting a further functional link [27]. Along with overall high sequence identity, RsTSPO
shares high similarity to the human TSPO in the proposed porphyrin [102] and drug binding sites
[103] in the first extra-membrane loop (loop 1), and in the proposed cholesterol binding site, the
Cholesterol Recognition/interaction Amino acid Consensus (CRAC) sequence [104]. On the
other hand, RsTSPO displays interesting binding behavior compared to its mammalian
homologues: mammalian TSPO has been shown to bind to PK11195 and cholesterol at
nanomolar affinity and porphyrins at micromolar affinity, but we have shown that RsTSPO binds
to all three with micromolar affinity (Chapter 2). The similarities and differences therefore
provide us with a unique opportunity to better understand the critical determinants for the ligand
binding affinity and the binding mechanism by mutagenesis and systematic analysis in
Rhodobacter. For example, investigation of the large difference in cholesterol binding affinity
between the mammalian and the Rhodobacter TSPO discussed in Chapter 3 led to identification
of a novel cholesterol binding enhancement motif and better understanding of the molecular
determinants of high affinity cholesterol binding.
The bacterial protein RsTSPO is important to study in itself. Given its role in transport of
small metabolites in and out of bacteria, understanding of the mechanism of transport by
RsTSPO could provide important insights to facilitate the understanding of transport mechanisms

20

as well as the evolution of transporters. In addition, R. sphaeroides provides a potential in vivo
drug screen assay: a benzodiazepine drug inhibits the color change of R. sphaeroides cultures
containing the rat version of TSPO during the switch between photosynthesis and respiration
[27].

21

Figure 1.6 Sequence alignment of TSPO proteins.

22

Figure 1.6 (cont’d)
Shown in the alignment are TSPO proteins from R. sphaeroides 2.4.1 (accession number
AAF24291) R. capsulatus (Z11165), human (CAB55884), mouse (NP_033905), rat
(NP_036647), bovine (DAA29060), zebrafish (NP_001006032), Arabidopsis thaliana
(AAL16286), Schizosaccharomyces pombe (CAA22182.2), Bacillus subtilis (YP_004204918.1),
Archaeoglobus fulgidus (NP_070304.1), Pseudomonas fluorescens (YP_348542.1), Drosophila
melanogaster (AAF51482) and Eubacterium (EDS01113.1). Conserved amino acids are shown
in colors from deep red to yellow. Secondary structure is predicted based on the sequence of
RsTSPO by the server Toppred [105]. Transmembrane helices are labeled in blue on top of the
sequences. The cholesterol recognition/interaction amino acid consensus (CRAC) is highlighted
in a magenta rectangular box with the 3 critical amino acid labeled with cyan triangles. The
proposed drug/porphyrin binding loop (loop 1) is highlighted in a purple rectangular box. The
human polymorphism residue is labeled with star. This alignment was made using the
CLUSTALW server and the picture was created in the Aline program [106].

23

Protein crystallography
X-ray crystallography is a well established method of determining the arrangement of atoms
within a crystal, including for salts, metals, minerals, and semiconductors, as well as for
biological molecules. Macromolecular crystallography is a technique using X-rays to study the
arrangement of atoms within biological molecules such as proteins, viruses, and nucleic acids.
Since the determination of the first crystal structure of myoglobin in the 1950s [107], X-ray
crystallography has helped to elucidate the detailed structures of over thousands of proteins, and
advanced our understanding of their mechanisms and of numerous processes/diseases. These
atomic resolution structures have also served as platforms for the development of many
important classes of drugs.
X-rays are suitable for determining the structure of macromolecules because the wavelength
of X-rays (~ 1 Å) is similar to the bond length of atoms within the biological macromolecules.
Therefore, the diffraction of X-rays by the electrons around the atoms can be used to calculate
the position of the atoms and the structure of the protein to high accuracy. Despite the rapid
development of various other methods including nuclear magnetic resonance (NMR), electron
microscopy (EM), and atomic force microscopy (AFM), X-ray crystallography is still the most
powerful method to obtain the three-dimensional structures of macromolecules due to its
incomparable high resolution and high accuracy. Recent developments in protein production,
purification methods, automation in liquid handling, high energy X-ray sources, and structural
genomic pipelines have also sped up and popularized this technique. As a result, the Protein Data
Bank (PDB) recently reached the 100,000 structure deposit milestone in May 2014.
Importance of membrane protein structures
It is estimated that more than 30 % of proteins encoded in the human genome are membrane

24

proteins, including integral membrane proteins as well as peripheral membrane proteins. The
importance of membrane proteins cannot be overstated; they mediate the transport of almost
everything between and within cells, from nutrients to neuronal signals. Mutations of membrane
proteins have been identified in many diseases, including severe diseases such as Alzheimer
[108,109] and Huntington diseases [110]. Not surprisingly, more than 50 % of the drugs on the
market target membrane proteins [111]. Structures of membrane proteins are critical for
understanding the fundamental mechanisms of their functions, as well as for rationally designing
drugs targeting membrane proteins for better treatment. However, the structures of membrane
proteins are severely underrepresented in the PDB. Compared to the well-established pipeline
and high throughput production of structures of soluble proteins, solving the structures of
membrane proteins is still a formidable task.
Challenges and advances in membrane protein crystallography
Membrane proteins are notoriously hard to work with for several reasons: 1) they are usually
naturally expressed at low levels; 2) they are designed to be surrounded by lipids, which usually
are not compatible with the biochemical and biophysical techniques used for characterization; 3)
they are often unstable when extracted into solution even with the protection of detergent.
However, in order to determine a structure of a membrane protein by X-ray crystallography,
milligram quantities of purified, homogenous membrane protein are usually required. Despite the
challenges imposed by these requirements, significant advances have been made to facilitate the
crystallization and structure determination of membrane proteins in the past few years [112].
Production of membrane proteins
The first bottleneck towards solving a structure of a membrane protein is to obtain the
protein of interest in sufficient amount. Historically, most membrane proteins, particularly

25

eukaryotic membrane proteins, that have been crystallized are from rich natural sources, such as
the photosynthetic reaction center from R. sphaeroides [113], rhodopsin from the retina [114],
and cytochrome c oxidase from bovine heart [115]. A membrane protein that is synthesized by
ribosomes usually needs to be inserted into the biological membrane with the help of other
proteins, such as the translocon, which is often specific for the host organism. In addition,
membrane proteins are frequently post-translationally modified within the membrane. As a result,
overexpression of membrane proteins in bacterial systems has frequently led to inclusion bodies,
incorrect folding, and toxicity to the host cell.
Despite the fact that Escherichia coli is still the most popular system for protein expression,
several other systems have been developed to facilitate the expression of membrane proteins,
especially eukaryotic membrane proteins that are frequently found to be incorrectly folded when
expressed in bacteria. Among others, the mammalian expression system [116] and the insect cell
expression system [117] are two popular methods. Both systems have the insertion and
post-translation modification machinery to facilitate the correct folding and expression of
eukaryotic membrane proteins. Increasing numbers of membrane proteins have been successfully
expressed from these systems, but incorrect folding and modification have also been reported.
For each particular membrane protein, finding the right expression system still requires a
significant amount of testing and manipulations.
Purification of membrane proteins
The fundamental difficulty in purifying membrane proteins lies in their hydrophobic nature
and their membrane localization. A detergent, an amphiphilic molecule with a hydrophilic head
group and a hydrophobic tail, is needed to extract the membrane protein from the natural
membrane into solution. However, finding the right detergent for each membrane protein still

26

relies on trial and error. Even with the right detergent, a membrane protein is usually less stable
when it is extracted into solution. Lipid is another key factor that plays an important role in
membrane protein purification and crystallization. Experience in this lab as well as others has
suggested that natural lipids co-purified with the membrane protein may stabilize the protein and
facilitate crystallization. Unfortunately, it is very hard to quantify and control the lipid content in
each particular purification. Ligands also play an important role in stabilizing membrane proteins.
Recent

technical

developments,

such

as

the

fluorescence-detection

size-exclusion

chromatography (FSEC) [118] and high throughput thermal stability assays [119] have allowed
us to rapidly screen for the construct and ligand to optimize purification of membrane proteins.
Crystallization of membrane proteins
Crystallization itself still poses a significant difficulty towards obtaining a crystal structure
of a membrane protein. Over the past decade, major technical developments, including
automated liquid handing systems for high throughput crystallization experiments [120],
micro-beams at synchrotron X-ray sources to facilitate the data collection on small crystals [121],
and new software to process sub-optimal diffraction data collected for membrane protein crystals,
have greatly improved our ability to obtain structural information from membrane protein
crystals. In addition, major advances have been made in the theories and methods for
crystallization of membrane proteins, such as the bicelle [122] and the lipidic cubic phase (LCP)
[123] crystallization methods and the understanding of the influence of lipids during the process
of crystallization. In this thesis, various methods were investigated for the crystallization of
TSPO, providing a case study for evaluating and rationally designing crystallization strategies
for difficult membrane proteins. Details of the investigation and the successful optimization of
crystallization methods for TSPO from R. sphaeroides are discussed in Chapter 4.

27

Importance of understanding the structure-function relationship of TSPO
As mentioned above, TSPO has been shown to play critical roles in several aspects of
human health. Therefore, it has been a long-standing research focus in various fields from basic
science to the pharmaceutical industry. Despite years of study and hundreds of ligands developed
and investigated in various systems, it is still unclear how exactly TSPO functions in various
scenarios, and how its function is regulated by ligands. Additionally, there remain discrepancies
and controversies in the literature regarding the binding of TSPO ligands and the physiological
significance of their interactions [103,124,125]. For instance, TSPO ligands have been reported
to have pro-apoptotic effects in cancer cells [126] but to be anti-apoptotic in cardiac tissue after
ischemia and reperfusion injury [63,127,128]. Two major difficulties are significantly hindering
our ability to understand the structure-function relationship of TSPO: 1) difficulty related to the
interpretation of binding data measured in complex systems such as total membranes or whole
cells from different tissues; and 2) the lack of high resolution structures of TSPO to guide the
functional studies and rationalize the drug development process.
To address the aforementioned issues, the aims of this study are to characterize the binding
properties of various ligands to a homologue of human TSPO, RsTSPO, in its purified state and
to obtain high resolution structures of RsTSPO with and without ligand bound. Since RsTSPO is
a unique small integral transmembrane protein from the outer mitochondrial membrane, it
presents unusual challenges. It has very small cytoplasmic loop regions for mediating crystal
contacts and an odd number of α helices, representing a class of membrane proteins that are
difficult to crystallize. Therefore, the successful optimization of the purification and
crystallization conditions for RsTSPO that is reported here provides valuable knowledge to
facilitate the crystallization of similar membrane proteins.

28

REFERENCES

29

REFERENCES

1. Braestrup C, Squires RF: Specific benzodiazepine receptors in rat brain characterized by
high-affinity (3H)diazepam binding. Proc Natl Acad Sci U S A 1977, 74:3805-3809.
2. Verma A, Snyder SH: Peripheral type benzodiazepine receptors. Annu Rev Pharmacol
Toxicol 1989, 29:307-322.
3. Papadopoulos V, Baraldi M, Guilarte TR, Knudsen TB, Lacapère JJ, Lindemann P, Norenberg
MD, Nutt D, Weizman A, Zhang MR, et al.: Translocator protein (18kDa): new
nomenclature for the peripheral-type benzodiazepine receptor based on its structure
and molecular function. Trends Pharmacol Sci 2006, 27:402-409.
4. Fan J, Lindemann P, Feuilloley MG, Papadopoulos V: Structural and functional evolution of
the translocator protein (18 kDa). Curr Mol Med 2012, 12:369-386.
5. Rupprecht R, Papadopoulos V, Rammes G, Baghai TC, Fan J, Akula N, Groyer G, Adams D,
Schumacher M: Translocator protein (18 kDa) (TSPO) as a therapeutic target for
neurological and psychiatric disorders. Nat Rev Drug Discov 2010, 9:971-988.
6. Woods MJ, Zisterer DM, Williams DC: Two cellular and subcellular locations for the
peripheral-type benzodiazepine receptor in rat liver. Biochem Pharmacol 1996,
51:1283-1292.
7. Anholt RR, Pedersen PL, De Souza EB, Snyder SH: The peripheral-type benzodiazepine
receptor. Localization to the mitochondrial outer membrane. J Biol Chem 1986,
261:576-583.
8. Snyder SH, Verma A, Trifiletti RR: The peripheral-type benzodiazepine receptor: a protein
of mitochondrial outer membranes utilizing porphyrins as endogenous ligands.
FASEB J 1987, 1:282-288.
9. McEnery MW, Snowman AM, Trifiletti RR, Snyder SH: Isolation of the mitochondrial
benzodiazepine receptor: association with the voltage-dependent anion channel and
the adenine nucleotide carrier. Proc Natl Acad Sci U S A 1992, 89:3170-3174.
10. Veenman L, Papadopoulos V, Gavish M: Channel-like functions of the 18-kDa
translocator protein (TSPO): regulation of apoptosis and steroidogenesis as part of
the host-defense response. Curr Pharm Des 2007, 13:2385-2405.
11. Krueger KE, Papadopoulos V: Peripheral-type benzodiazepine receptors mediate
translocation of cholesterol from outer to inner mitochondrial membranes in
adrenocortical cells. J Biol Chem 1990, 265:15015-15022.
12. Verma A, Nye JS, Snyder SH: Porphyrins are endogenous ligands for the mitochondrial
(peripheral-type) benzodiazepine receptor. Proc Natl Acad Sci U S A 1987,
30

84:2256-2260.
13. Batarseh A, Papadopoulos V: Regulation of translocator protein 18 kDa (TSPO)
expression in health and disease states. Mol Cell Endocrinol 2010, 327:1-12.
14. Lacapère JJ, Papadopoulos V: Peripheral-type benzodiazepine receptor: structure and
function of a cholesterol-binding protein in steroid and bile acid biosynthesis.
Steroids 2003, 68:569-585.
15. Rone MB, Fan J, Papadopoulos V: Cholesterol transport in steroid biosynthesis: role of
protein-protein interactions and implications in disease states. Biochim Biophys Acta
2009, 1791:646-658.
16. Miller WL: Steroid hormone synthesis in mitochondria. Mol Cell Endocrinol 2013,
379:62-73.
17. Simpson ER, Waterman MR: Regulation by ACTH of steroid hormone biosynthesis in the
adrenal cortex. Can J Biochem Cell Biol 1983, 61:692-707.
18. Jefcoate C: High-flux mitochondrial cholesterol trafficking, a specialized function of the
adrenal cortex. J Clin Invest 2002, 110:881-890.
19. Mukhin AG, Papadopoulos V, Costa E, Krueger KE: Mitochondrial benzodiazepine
receptors regulate steroid biosynthesis. Proc Natl Acad Sci U S A 1989, 86:9813-9816.
20. Barnea ER, Fares F, Gavish M: Modulatory action of benzodiazepines on human term
placental steroidogenesis in vitro. Mol Cell Endocrinol 1989, 64:155-159.
21. Papadopoulos V, Mukhin AG, Costa E, Krueger KE: The peripheral-type benzodiazepine
receptor is functionally linked to Leydig cell steroidogenesis. J Biol Chem 1990,
265:3772-3779.
22. Rone MB, Liu J, Blonder J, Ye X, Veenstra TD, Young JC, Papadopoulos V: Targeting and
insertion of the cholesterol-binding translocator protein into the outer
mitochondrial membrane. Biochemistry 2009, 48:6909-6920.
23. Ajioka RS, Phillips JD, Kushner JP: Biosynthesis of heme in mammals. Biochim Biophys
Acta 2006, 1763:723-736.
24. Taketani S, Kohno H, Okuda M, Furukawa T, Tokunaga R: Induction of peripheral-type
benzodiazepine receptors during differentiation of mouse erythroleukemia cells. A
possible involvement of these receptors in heme biosynthesis. J Biol Chem 1994,
269:7527-7531.
25. Yeliseev AA, Kaplan S: A sensory transducer homologous to the mammalian
peripheral-type benzodiazepine receptor regulates photosynthetic membrane
complex formation in Rhodobacter sphaeroides 2.4.1. J Biol Chem 1995,
270:21167-21175.
31

26. Yeliseev AA, Kaplan S: A novel mechanism for the regulation of photosynthesis gene
expression by the TspO outer membrane protein of Rhodobacter sphaeroides 2.4.1. J
Biol Chem 1999, 274:21234-21243.
27. Yeliseev AA, Krueger KE, Kaplan S: A mammalian mitochondrial drug receptor
functions as a bacterial "oxygen" sensor. Proc Natl Acad Sci U S A 1997,
94:5101-5106.
28. Delavoie F, Li H, Hardwick M, Robert JC, Giatzakis C, Péranzi G, Yao ZX, Maccario J,
Lacapère JJ, Papadopoulos V: In vivo and in vitro peripheral-type benzodiazepine
receptor polymerization: functional significance in drug ligand and cholesterol
binding. Biochemistry 2003, 42:4506-4519.
29. Wendler G, Lindemann P, Lacapère JJ, Papadopoulos V: Protoporphyrin IX binding and
transport by recombinant mouse PBR. Biochem Biophys Res Commun 2003,
311:847-852.
30. Zeno S, Veenman L, Katz Y, Bode J, Gavish M, Zaaroor M: The 18 kDa mitochondrial
translocator protein (TSPO) prevents accumulation of protoporphyrin IX.
Involvement of reactive oxygen species (ROS). Curr Mol Med 2012, 12:494-501.
31. Kroemer G, Dallaporta B, Resche-Rigon M: The mitochondrial death/life regulator in
apoptosis and necrosis. Annu Rev Physiol 1998, 60:619-642.
32. Tait SW, Green DR: Mitochondria and cell death: outer membrane permeabilization and
beyond. Nat Rev Mol Cell Biol 2010, 11:621-632.
33. Wajant H: The Fas signaling pathway: more than a paradigm. Science 2002,
296:1635-1636.
34. Jiang X, Wang X: Cytochrome C-mediated apoptosis. Annu Rev Biochem 2004, 73:87-106.
35. Crompton M: The mitochondrial permeability transition pore and its role in cell death.
Biochem J 1999, 341 ( Pt 2):233-249.
36. Leung AW, Halestrap AP: Recent progress in elucidating the molecular mechanism of the
mitochondrial permeability transition pore. Biochim Biophys Acta 2008,
1777:946-952.
37. Giorgio V, von Stockum S, Antoniel M, Fabbro A, Fogolari F, Forte M, Glick GD, Petronilli
V, Zoratti M, Szabó I, et al.: Dimers of mitochondrial ATP synthase form the
permeability transition pore. Proc Natl Acad Sci U S A 2013.
38. Xiao JJ, Liang DD, Zhang H, Liu Y, Li FJ, Chen YH: 4 '-Chlorodiazepam, a translocator
protein (18 kDa) antagonist, improves cardiac functional recovery during
postischemia reperfusion in rats. Exp Biol Med 2010, 235:478-486.
39. Gatliff J, Campanella M: The 18 kDa translocator protein (TSPO): a new perspective in
32

mitochondrial biology. Curr Mol Med 2012, 12:356-368.
40. Batarseh A, Li J, Papadopoulos V: Protein kinase C epsilon regulation of translocator
protein (18 kDa) Tspo gene expression is mediated through a MAPK pathway
targeting STAT3 and c-Jun transcription factors. Biochemistry 2010, 49:4766-4778.
41. Katz Y, Eitan A, Amiri Z, Gavish M: Dramatic increase in peripheral benzodiazepine
binding sites in human colonic adenocarcinoma as compared to normal colon. Eur J
Pharmacol 1988, 148:483-484.
42. Katz Y, Ben-Baruch G, Kloog Y, Menczer J, Gavish M: Increased density of peripheral
benzodiazepine-binding sites in ovarian carcinomas as compared with benign
ovarian tumours and normal ovaries. Clin Sci (Lond) 1990, 78:155-158.
43. Hardwick M, Fertikh D, Culty M, Li H, Vidic B, Papadopoulos V: Peripheral-type
benzodiazepine receptor (PBR) in human breast cancer: correlation of breast cancer
cell aggressive phenotype with PBR expression, nuclear localization, and
PBR-mediated cell proliferation and nuclear transport of cholesterol. Cancer Res
1999, 59:831-842.
44. Brown RC, Degenhardt B, Kotoula M, Papadopoulous V: Location-dependent role of the
human glioma cell peripheral-type benzodiazepine receptor in proliferation and
steroid biosynthesis. Cancer Lett 2000, 156:125-132.
45. Corsi L, Geminiani E, Avallone R, Baraldi M: Nuclear location-dependent role of
peripheral benzodiazepine receptor (PBR) in hepatic tumoral cell lines proliferation.
Life Sci 2005, 76:2523-2533.
46. Carmel I, Fares FA, Leschiner S, Scherubl H, Weisinger G, Gavish M: Peripheral-type
benzodiazepine receptors in the regulation of proliferation of MCF-7 human breast
carcinoma cell line. Biochem Pharmacol 1999, 58:273-278.
47. Rechichi M, Salvetti A, Chelli B, Costa B, Da Pozzo E, Spinetti F, Lena A, Evangelista M,
Rainaldi G, Martini C, et al.: TSPO over-expression increases motility, transmigration
and proliferation properties of C6 rat glioma cells. Biochim Biophys Acta 2008,
1782:118-125.
48. Fafalios A, Akhavan A, Parwani AV, Bies RR, McHugh KJ, Pflug BR: Translocator protein
blockade reduces prostate tumor growth. Clin Cancer Res 2009, 15:6177-6184.
49. Sutter AP, Maaser K, Grabowski P, Bradacs G, Vormbrock K, Hopfner M, Krahn A, Heine B,
Stein H, Somasundaram R, et al.: Peripheral benzodiazepine receptor ligands induce
apoptosis and cell cycle arrest in human hepatocellular carcinoma cells and enhance
chemosensitivity to paclitaxel, docetaxel, doxorubicin and the Bcl-2 inhibitor
HA14-1. J Hepatol 2004, 41:799-807.
50. Li W, Hardwick MJ, Rosenthal D, Culty M, Papadopoulos V: Peripheral-type
benzodiazepine receptor overexpression and knockdown in human breast cancer
33

cells indicate its prominent role in tumor cell proliferation. Biochem Pharmacol 2007,
73:491-503.
51. White C: On the occurrence of crystals in tumours. J Pathol Bacteriol 1909, 13:3-10.
52. Singleton PA, Bourguignon LY: CD44 interaction with ankyrin and IP3 receptor in lipid
rafts promotes hyaluronan-mediated Ca2+ signaling leading to nitric oxide
production and endothelial cell adhesion and proliferation. Exp Cell Res 2004,
295:102-118.
53. Trachootham D, Alexandre J, Huang P: Targeting cancer cells by ROS-mediated
mechanisms: a radical therapeutic approach? Nat Rev Drug Discov 2009, 8:579-591.
54. Kumar B, Koul S, Khandrika L, Meacham RB, Koul HK: Oxidative stress is inherent in
prostate cancer cells and is required for aggressive phenotype. Cancer Res 2008,
68:1777-1785.
55. Wu WS: The signaling mechanism of ROS in tumor progression. Cancer Metastasis Rev
2006, 25:695-705.
56. Fuster V, Kelly B, Vedanthan R: Global Cardiovascular Health. J Am Coll Cardio 2011,
58:1208-1210.
57. Orhan G, Yapici N, Yuksel M, Sargin M, Senay S, Yalcin AS, Aykac Z, Aka SA: Effects of
N-acetylcysteine on myocardial ischemia-reperfusion injury in bypass surgery.
Heart and Vessels 2006, 21:42-47.
58. Halestrap AP: The mitochondrial permeability transition: its molecular mechanism and
role in reperfusion injury. Biochem Soc Symp 1999, 66:181-203.
59. Halestrap AP, Pasdois P: The role of the mitochondrial permeability transition pore in
heart disease. Biochim Biophys Acta 2009, 1787:1402-1415.
60. Halestrap AP: A pore way to die: the role of mitochondria in reperfusion injury and
cardioprotection. Biochem Soc Trans 2010, 38:841-860.
61. Kroemer G: Introduction: mitochondrial control of apoptosis. Biochimie 2002,
84:103-104.
62. Sileikyte J, Petronilli V, Zulian A, Dabbeni-Sala F, Tognon G, Nikolov P, Bernardi P,
Ricchelli F: Regulation of the inner membrane mitochondrial permeability
transition by the outer membrane translocator protein (peripheral benzodiazepine
receptor). J Biol Chem 286:1046-1053.
63. Obame FN, Zini R, Souktani R, Berdeaux A, Morin D: Peripheral benzodiazepine
receptor-induced myocardial protection is mediated by inhibition of mitochondrial
membrane permeabilization. J Pharmacol Exp Ther 2007, 323:336-345.

34

64. Schaller S, Paradis S, Ngoh GA, Assaly R, Buisson B, Drouot C, Ostuni MA, Lacapere JJ,
Bassissi F, Bordet T, et al.: TRO40303, a new cardioprotective compound, inhibits
mitochondrial permeability transition. J Pharmacol Exp Ther 333:696-706.
65. Roy N, Friehs I, Cowan DB, Zurakowski D, McGowan FX, del Nido PJ: Dopamine induces
postischemic cardiomyocyte apoptosis in vivo: an effect ameliorated by propofol.
Ann Thorac Surg 2006, 82:2192-2199.
66. Garib V, Lang K, Niggemann B, Zanker KS, Brandt L, Dittmar T: Propofol-induced
calcium signalling and actin reorganization within breast carcinoma cells. Eur J
Anaesthesiol 2005, 22:609-615.
67. Castedo M, Perfettini JL, Kroemer G: Mitochondrial apoptosis and the peripheral
benzodiazepine receptor: a novel target for viral and pharmacological manipulation.
J Exp Med 2002, 196:1121-1125.
68. Everett H, Barry M, Sun X, Lee SF, Frantz C, Berthiaume LG, McFadden G, Bleackley RC:
The myxoma poxvirus protein, M11L, prevents apoptosis by direct interaction with
the mitochondrial permeability transition pore. J Exp Med 2002, 196:1127-1139.
69. Halestrap AP, Clarke SJ, Javadov SA: Mitochondrial permeability transition pore opening
during myocardial reperfusion - a target for cardioprotection. Cardiovascular
Research 2004, 61:372-385.
70. Cosenza-Nashat M, Zhao ML, Suh HS, Morgan J, Natividad R, Morgello S, Lee SC:
Expression of the translocator protein of 18 kDa by microglia, macrophages and
astrocytes based on immunohistochemical localization in abnormal human brain.
Neuropathol Appl Neurobiol 2009, 35:306-328.
71. Gavish M, Bachman I, Shoukrun R, Katz Y, Veenman L, Weisinger G, Weizman A: Enigma
of the peripheral benzodiazepine receptor. Pharmacol Rev 1999, 51:629-650.
72. Maeda J, Higuchi M, Inaji M, Ji B, Haneda E, Okauchi T, Zhang MR, Suzuki K, Suhara T:
Phase-dependent roles of reactive microglia and astrocytes in nervous system injury
as delineated by imaging of peripheral benzodiazepine receptor. Brain Res 2007,
1157:100-111.
73. Chen MK, Guilarte TR: Translocator protein 18 kDa (TSPO): molecular sensor of brain
injury and repair. Pharmacol Ther 2008, 118:1-17.
74. Karchewski LA, Bloechlinger S, Woolf CJ: Axonal injury-dependent induction of the
peripheral benzodiazepine receptor in small-diameter adult rat primary sensory
neurons. Eur J Neurosci 2004, 20:671-683.
75. Mills CD, Bitler JL, Woolf CJ: Role of the peripheral benzodiazepine receptor in sensory
neuron regeneration. Mol Cell Neurosci 2005, 30:228-237.
76. Lacor P, Gandolfo P, Tonon MC, Brault E, Dalibert I, Schumacher M, Benavides J, Ferzaz B:
35

Regulation of the expression of peripheral benzodiazepine receptors and their
endogenous ligands during rat sciatic nerve degeneration and regeneration: a role
for PBR in neurosteroidogenesis. Brain Res 1999, 815:70-80.
77. Ji B, Maeda J, Sawada M, Ono M, Okauchi T, Inaji M, Zhang MR, Suzuki K, Ando K,
Staufenbiel M, et al.: Imaging of peripheral benzodiazepine receptor expression as
biomarkers of detrimental versus beneficial glial responses in mouse models of
Alzheimer's and other CNS pathologies. J Neurosci 2008, 28:12255-12267.
78. Gerhard A, Pavese N, Hotton G, Turkheimer F, Es M, Hammers A, Eggert K, Oertel W,
Banati RB, Brooks DJ: In vivo imaging of microglial activation with
[11C](R)-PK11195 PET in idiopathic Parkinson's disease. Neurobiol Dis 2006,
21:404-412.
79. Ouchi Y, Yoshikawa E, Sekine Y, Futatsubashi M, Kanno T, Ogusu T, Torizuka T: Microglial
activation and dopamine terminal loss in early Parkinson's disease. Ann Neurol 2005,
57:168-175.
80. Pavese N, Gerhard A, Tai YF, Ho AK, Turkheimer F, Barker RA, Brooks DJ, Piccini P:
Microglial activation correlates with severity in Huntington disease: a clinical and
PET study. Neurology 2006, 66:1638-1643.
81. Gerhard A, Schwarz J, Myers R, Wise R, Banati RB: Evolution of microglial activation in
patients after ischemic stroke: a [11C](R)-PK11195 PET study. Neuroimage 2005,
24:591-595.
82. Venneti S, Wagner AK, Wang G, Slagel SL, Chen X, Lopresti BJ, Mathis CA, Wiley CA:
The high affinity peripheral benzodiazepine receptor ligand DAA1106 binds
specifically to microglia in a rat model of traumatic brain injury: implications for
PET imaging. Exp Neurol 2007, 207:118-127.
83. Versijpt J, Debruyne JC, Van Laere KJ, De Vos F, Keppens J, Strijckmans K, Achten E,
Slegers G, Dierckx RA, Korf J, et al.: Microglial imaging with positron emission
tomography and atrophy measurements with magnetic resonance imaging in
multiple sclerosis: a correlative study. Mult Scler 2005, 11:127-134.
84. Vowinckel E, Reutens D, Becher B, Verge G, Evans A, Owens T, Antel JP: PK11195 binding
to the peripheral benzodiazepine receptor as a marker of microglia activation in
multiple sclerosis and experimental autoimmune encephalomyelitis. J Neurosci Res
1997, 50:345-353.
85. Bird JL, Izquierdo-Garcia D, Davies JR, Rudd JH, Probst KC, Figg N, Clark JC, Weissberg
PL, Davenport AP, Warburton EA: Evaluation of translocator protein quantification
as a tool for characterising macrophage burden in human carotid atherosclerosis.
Atherosclerosis 2010, 210:388-391.
86. Yasuno F, Ota M, Kosaka J, Ito H, Higuchi M, Doronbekov TK, Nozaki S, Fujimura Y,
Koeda M, Asada T, et al.: Increased binding of peripheral benzodiazepine receptor in
36

Alzheimer's disease measured by positron
[11C]DAA1106. Biol Psychiatry 2008, 64:835-841.

emission

tomography

with

87. Papadopoulos V, Lecanu L: Translocator protein (18 kDa) TSPO: an emerging
therapeutic target in neurotrauma. Exp Neurol 2009, 219:53-57.
88. Papadopoulos V, Lecanu L, Brown RC, Han Z, Yao ZX: Peripheral-type benzodiazepine
receptor in neurosteroid biosynthesis, neuropathology and neurological disorders.
Neuroscience 2006, 138:749-756.
89. Poisbeau P, Patte-Mensah C, Keller AF, Barrot M, Breton JD, Luis-Delgado OE,
Freund-Mercier MJ, Mensah-Nyagan AG, Schlichter R: Inflammatory pain upregulates
spinal inhibition via endogenous neurosteroid production. J Neurosci 2005,
25:11768-11776.
90. Inquimbert P, Rodeau JL, Schlichter R: Regional differences in the decay kinetics of
GABA(A) receptor-mediated miniature IPSCs in the dorsal horn of the rat spinal
cord are determined by mitochondrial transport of cholesterol. J Neurosci 2008,
28:3427-3437.
91. Jayakumar AR, Panickar KS, Norenberg MD: Effects on free radical generation by ligands
of the peripheral benzodiazepine receptor in cultured neural cells. J Neurochem 2002,
83:1226-1234.
92. Chauveau F, Boutin H, Van Camp N, Dollé F, Tavitian B: Nuclear imaging of
neuroinflammation: a comprehensive review of [11C]PK11195 challengers. Eur J
Nucl Med Mol Imaging 2008, 35:2304-2319.
93. Owen DR, Yeo AJ, Gunn RN, Song K, Wadsworth G, Lewis A, Rhodes C, Pulford DJ,
Bennacef I, Parker CA, et al.: An 18-kDa translocator protein (TSPO) polymorphism
explains differences in binding affinity of the PET radioligand PBR28. J Cereb Blood
Flow Metab 2012, 32:1-5.
94. Costa B, Pini S, Martini C, Abelli M, Gabelloni P, Landi S, Muti M, Gesi C, Lari L, Cardini
A, et al.: Ala147Thr substitution in translocator protein is associated with adult
separation anxiety in patients with depression. Psychiatr Genet 2009, 19:110-111.
95. Colasanti A, Owen DR, Grozeva D, Rabiner EA, Matthews PM, Craddock N, Young AH:
Bipolar Disorder is associated with the rs6971 polymorphism in the gene encoding
18 kDa Translocator Protein (TSPO). Psychoneuroendocrinology 2013, 38:2826-2829.
96. Ryu JK, Choi HB, McLarnon JG: Peripheral benzodiazepine receptor ligand PK11195
reduces microglial activation and neuronal death in quinolinic acid-injected rat
striatum. Neurobiol Dis 2005, 20:550-561.
97. Veiga S, Azcoitia I, Garcia-Segura LM: Ro5-4864, a peripheral benzodiazepine receptor
ligand, reduces reactive gliosis and protects hippocampal hilar neurons from kainic
acid excitotoxicity. J Neurosci Res 2005, 80:129-137.
37

98. Parker MA, Bazan HE, Marcheselli V, Rodriguez de Turco EB, Bazan NG:
Platelet-activating factor induces permeability transition and cytochrome c release
in isolated brain mitochondria. J Neurosci Res 2002, 69:39-50.
99. Bui ET, Bradley PJ, Johnson PJ: A common evolutionary origin for mitochondria and
hydrogenosomes. Proc Natl Acad Sci U S A 1996, 93:9651-9656.
100. Armstrong GA, Alberti M, Leach F, Hearst JE: Nucleotide-sequence, organization, and
nature of the protein products of the carotenoid biosynthesis gene-cluster of
Rhodobacter capsulatus. Mol Gen Gen 1989, 216.
101. Baker ME, Fanestil DD: Mammalian peripheral-type benzodiazepine receptor is
homologous to CrtK protein of Rhodobacter capsulatus, a photosynthetic bacterium.
Cell 1991, 65:721-722.
102. Yeliseev AA, Kaplan S: TspO of Rhodobacter sphaeroides. A structural and functional
model for the mammalian peripheral benzodiazepine receptor. J Biol Chem 2000,
275:5657-5667.
103. Scarf AM, Auman KM, Kassiou M: Is there any correlation between binding and
functional effects at the translocator protein (TSPO) (18 kDa)? Curr Mol Med 2012,
12:387-397.
104. Li H, Yao Z, Degenhardt B, Teper G, Papadopoulos V: Cholesterol binding at the
cholesterol recognition/ interaction amino acid consensus (CRAC) of the
peripheral-type benzodiazepine receptor and inhibition of steroidogenesis by an
HIV TAT-CRAC peptide. Proc Natl Acad Sci U S A 2001, 98:1267-1272.
105. Cserzö M, Wallin E, Simon I, von Heijne G, Elofsson A: Prediction of transmembrane
alpha-helices in prokaryotic membrane proteins: the dense alignment surface
method. Protein Eng 1997, 10:673-676.
106. Bond CS, Schüttelkopf AW: ALINE: a WYSIWYG protein-sequence alignment editor
for publication-quality alignments. Acta Crystallogr D Biol Crystallogr 2009,
65:510-512.
107. Kendrew JC, Dickerson RE, Strandberg BE, Hart RG, Davies DR, Phillips DC, Shore VC:
Structure of myoglobin: A three-dimensional Fourier synthesis at 2 A. resolution.
Nature 1960, 185:422-427.
108. Yan R, Bienkowski MJ, Shuck ME, Miao H, Tory MC, Pauley AM, Brashier JR, Stratman
NC, Mathews WR, Buhl AE, et al.: Membrane-anchored aspartyl protease with
Alzheimer's disease beta-secretase activity. Nature 1999, 402:533-537.
109. Li J, Xu M, Zhou H, Ma J, Potter H: Alzheimer presenilins in the nuclear membrane,
interphase kinetochores, and centrosomes suggest a role in chromosome segregation.
Cell 1997, 90:917-927.

38

110. Butterfield DA, Oeswein JQ, Markesbery WR: Electron spin resonance study of
membrane protein alterations in erythrocytes in Huntington's disease. Nature 1977,
267:453-455.
111. Yildirim MA, Goh KI, Cusick ME, Barabási AL, Vidal M: Drug-target network. Nat
Biotechnol 2007, 25:1119-1126.
112. Stroud RM: New tools in membrane protein determination. F1000 Biol Rep 2011, 3:8.
113. Allen JP, Feher G, Yeates TO, Komiya H, Rees DC: Structure of the reaction center from
Rhodobacter sphaeroides R-26: the cofactors. Proc Natl Acad Sci U S A 1987,
84:5730-5734.
114. Palczewski K, Kumasaka T, Hori T, Behnke CA, Motoshima H, Fox BA, Le Trong I, Teller
DC, Okada T, Stenkamp RE, et al.: Crystal structure of rhodopsin: A G
protein-coupled receptor. Science 2000, 289:739-745.
115. Tsukihara T, Aoyama H, Yamashita E, Tomizaki T, Yamaguchi H, Shinzawa-Itoh K,
Nakashima R, Yaono R, Yoshikawa S: The whole structure of the 13-subunit oxidized
cytochrome c oxidase at 2.8 A. Science 1996, 272:1136-1144.
116. Chaudhary S, Pak JE, Gruswitz F, Sharma V, Stroud RM: Overexpressing human
membrane proteins in stably transfected and clonal human embryonic kidney 293S
cells. Nat Protoc 2012, 7:453-466.
117. Kost TA, Condreay JP, Jarvis DL: Baculovirus as versatile vectors for protein expression
in insect and mammalian cells. Nat Biotechnol 2005, 23:567-575.
118. Kawate T, Gouaux E: Fluorescence-detection size-exclusion chromatography for
precrystallization screening of integral membrane proteins. Structure 2006,
14:673-681.
119. Alexandrov AI, Mileni M, Chien EY, Hanson MA, Stevens RC: Microscale fluorescent
thermal stability assay for membrane proteins. Structure 2008, 16:351-359.
120. Cherezov V, Peddi A, Muthusubramaniam L, Zheng YF, Caffrey M: A robotic system for
crystallizing membrane and soluble proteins in lipidic mesophases. Acta Crystallogr
D Biol Crystallogr 2004, 60:1795-1807.
121. Fischetti RF, Xu S, Yoder DW, Becker M, Nagarajan V, Sanishvili R, Hilgart MC, Stepanov
S, Makarov O, Smith JL: Mini-beam collimator enables microcrystallography
experiments on standard beamlines. J Synchrotron Radiat 2009, 16:217-225.
122. Faham S, Bowie JU: Bicelle crystallization: a new method for crystallizing membrane
proteins yields a monomeric bacteriorhodopsin structure. J Mol Biol 2002, 316:1-6.
123. Caffrey M, Cherezov V: Crystallizing membrane proteins using lipidic mesophases.
Nature Protocols 2009, 4:706-731.
39

124. Scarf AM, Luus C, Da Pozzo E, Selleri S, Guarino C, Martini C, Ittner LM, Kassiou M:
Evidence for complex binding profiles and species differences at the translocator
protein (TSPO) (18 kDa). Curr Mol Med 2012, 12:488-493.
125. Gonzalez-Polo RA, Carvalho G, Braun T, Decaudin D, Fabre C, Larochette N, Perfettini JL,
Djavaheri-Mergny M, Youlyouz-Marfak I, Codogno P, et al.: PK11195 potently
sensitizes to apoptosis induction independently from the peripheral benzodiazepin
receptor. Oncogene 2005, 24:7503-7513.
126. Decaudin D, Castedo M, Nemati F, Beurdeley-Thomas A, De Pinieux G, Caron A, Pouillart
P, Wijdenes J, Rouillard D, Kroemer G, et al.: Peripheral benzodiazepine receptor
ligands reverse apoptosis resistance of cancer cells in vitro and in vivo. Cancer Res
2002, 62:1388-1393.
127. Xiao J, Liang D, Zhang H, Liu Y, Li F, Chen YH: 4'-Chlorodiazepam, a translocator
protein (18 kDa) antagonist, improves cardiac functional recovery during
postischemia reperfusion in rats. Exp Biol Med (Maywood) 2010, 235:478-486.
128. Schaller S, Paradis S, Ngoh GA, Assaly R, Buisson B, Drouot C, Ostuni MA, Lacapere JJ,
Bassissi F, Bordet T, et al.: TRO40303, a new cardioprotective compound, inhibits
mitochondrial permeability transition. J Pharmacol Exp Ther 2010, 333:696-706.

40

CHAPTER 2
Characterization and modeling of the oligomeric state and ligand binding behavior of
purified Translocator Protein 18 kDa (TSPO) from Rhodobacter sphaeroides

Reprint (adapted) with permission from Fei Li, Yan Xia, Jens Meiler, and Shelagh
Ferguson-Miller. Characterization and modeling of the oligomeric state and ligand binding
behavior of purified Translocator Protein 18 kDa (TSPO) from Rhodobacter sphaeroides.
Biochemistry. 47(38): 9931-9933. Copyright 2013 American Chemical Society.

Computation modeling work was done in collaboration with Yan Xia and Dr. Jens Meiler at
Vanderbilt University.
41

INTRODUCTION
Translocator Protein 18 kDa (TSPO) was originally discovered as a secondary binding
site for the widely prescribed anti-anxiety drugs, benzodiazepines, in the mitochondria of rat
kidney [1]. For this reason, it became known as the peripheral-type benzodiazepine receptor
(PBR). In 2006, it was renamed as Translocator Protein 18 kDa (TSPO) [2] to better represent
the emerging understanding of a family of highly conserved integral membrane proteins (Figure
1.6) found in Archaea through to plants and human, with a broad spectrum of functions including
steroid and porphyrin transport and regulation of apoptosis, inflammation, metastatic cancer and
cardiovascular diseases [3,4].
In mammalian cells, TSPO is primarily located in the outer mitochondrial membrane and
concentrated at the outer/inner membrane contact sites [5]. It is expressed in all organs examined
so far and at particularly high levels in steroid hormone producing tissues, such as adrenal glands
and kidney. Knock-out of TSPO is developmentally lethal in the mouse [6]. The major function
of TSPO has been proposed to be transporting small molecules, including cholesterol and heme
biosynthesis intermediates, into or out of mitochondria to be further metabolized [7-10]. In
addition, TSPO has been found to be highly expressed in areas of brain injury and inflammation
[11], and aggressive cancers [12,13], as well as brains of Alzheimer and Huntington disease
patients [14]. In these situations, TSPO has been proposed to be involved in the regulation of the
mitochondrial permeability transition pore (MPTP) [15,16], thus exerting its effects on cellular
homeostasis (see reviews in Current Molecular Medicine [17,18]). TSPO ligands, including
benzodiazepines and more specific compounds represented by the diagnostic ligand PK11195,
have been shown to attenuate cancer cell proliferation [13], to have neuro-protective effects [19],
and to inhibit the MPTP [20]. The interaction of TSPO with ligands continues to be a strong
research focus aimed at the development of drugs targeting TSPO for imaging and treatment of
42

diverse disease states.
Mitochondria play important roles in the normal function of cells and also in regulating
cell death through apoptosis and/or necrosis [21,22]. The initiation of apoptosis is usually
preceded by a loss of the mitochondrial membrane potential (ΔΨm), which is proposed to be
regulated by the MPTP [23], a multi-protein complex whose structural components are still not
established. Together with other proteins, TSPO has been implicated as a regulator of the MPTP,
having been isolated originally in a complex with the voltage dependent anion channel (VDAC)
and the adenine nucleotide translocator (ANT) [24]. Both VDAC and ANT are candidate
components of the MPTP [15,25], but other possibilities remain [26]. Various TSPO ligands have
been shown to regulate apoptosis [27]; however, it is currently still unclear how MPTP or
apoptosis is regulated by TSPO.
In the purple non-sulfur bacteria that are suggested as ancestors of mitochondria [28],
TSPO was discovered in the carotenoid gene cluster as CrtK [29], first in Rhodobacter
capsulatus [29], and subsequently in R. sphaeroides (RsTSPO) [30]. Initially designated as the
tryptophan-rich sensory protein (TspO), the bacterial protein was recognized as homologous to
the mitochondrial PBR [30,31]. In Rhodobacter it is located in the outer membrane and involved
in regulating photosynthetic gene expression in response to oxygen and light conditions [32,33].
Similar to its mammalian ortholog, RsTSPO is proposed to be involved in the transport of small
molecules such as porphyrin intermediates of the heme and chlorophyll biosynthesis/degradation
pathways [33,34]. RsTSPO shares considerable sequence homology with human TSPO (HsTSPO)
(Figure 1.6) and rat TSPO has been shown to substitute for RsTSPO in R. sphaeroides,
suggesting a functional link [32]. Along with overall significant sequence identity (30 %),
RsTSPO shares noteworthy sequence similarity to the HsTSPO in the first extra-membrane loop

43

(loop 1) which is proposed to participate in porphyrin [34] and drug binding [35], as well as in
the cholesterol binding site located at the Cholesterol Recognition/interaction Amino acid
Consensus (CRAC) sequence at the C-terminus, including the end of trans-membrane helix V
[36].
Although R. sphaeroides does not make cholesterol, it does produce a related class of
steroid-like molecules, hopanoids [37], which may account for why the proposed critical residues
for cholesterol binding are well conserved between mammalian and RsTSPO. However, the
sequences before and after the CRAC region vary. A recent paper reported that the human single
polymorphism A147T immediately preceding the CRAC sequence in helix V resulted in 2 orders
of magnitude reduction in affinity for a PK11195 analog used for PET imaging [38]. This finding
highlights the adjacent variable region (residue 144-146 in the human sequence) that could play a
role in differences in ligand binding between Rhodobacter and human TSPO. In addition, loop 1,
implicated in drug and porphyrin binding by mutations in both Rhodobacter [34] and
mammalian [35] TSPO, is highly variable in the initial one third, while the rest is conserved.
Differences in this region may also play a role in the significantly higher affinity of PK11195
reported for human [39] and for rat TSPO expressed in R. sphaeroides compared to the
endogenous RsTSPO [32]. In addition, the mutation W38C in this loop 1 region was observed to
influence dimerization and stability of RsTSPO, suggesting a complex role in protein structure
[34]. With respect to cholesterol binding, information on the binding behavior of RsTSPO is still
not available and the possibility of direct or allosteric competition between PK11195, cholesterol
and porphyrin has been suggested [40] but not established.
There remain discrepancies and controversies in the literature regarding the binding of
TSPO ligands and the physiological significance of their interactions [35,41,42]. For instance,

44

TSPO ligands have been reported to have pro-apoptotic effects in cancer cells [43] but to be
anti-apoptotic in cardiac tissue after ischemia and reperfusion injury [20,44,45].

A major

difficulty comes from the interpretation of binding data measured in complex systems such as
total membranes or whole cells from different tissues where other proteins may affect the
interaction. To clarify some of these issues, we have expressed and purified to homogeneity the
RsTSPO wild-type and the reported dimer-stabilizing mutant W38C [34]. We describe studies of
their oligomerization state as well as their binding to ligands, including cholesterol, porphyrin,
the diagnostic drug PK11195, and a series of compounds with reported apoptosis-influencing
properties. Using a sensitive tryptophan fluorescence quenching assay, our results show that
these ligands have direct interactions with RsTSPO at micro-molar affinity. We provide an
atomic-detail model of the RsTSPO dimer, based on a 10 Å resolution cryo-EM density map [46]
and using EM-Fold [47], and discuss the relationship between ligand interactions in the context
of the dimer model. Some novel ligands characterized in this study provide candidates for the
optimization of crystallization of TSPO.
MATERIALS AND METHODS
Materials
Chemicals were purchased from Sigma-Aldrich (St. Louis, MO). Dodecyl maltoside (DDM),
decyl maltoside (DM) and n-dodecylphosphocholine (Fos-Choline-12) (Anagrade) were
purchased from Anatrace (Maumee, OH). The RsTSPO expression plasmid was a gift from Dr.
Samuel Kaplan from the University of Texas and the codon-optimized HsTSPO was synthesized
by DNA 2.0 ®. Pfu Turbo

®

polymerase was purchased from Agilent technologies (Santa Clara,

CA). Restriction enzymes and T4 ligase were purchased from New England Biolabs (Ipswich,
MA).
45

Cloning
The coding sequence of RsTSPO was sub-cloned into the pUC12 vector for mutagenesis.
The W38C mutant of RsTSPO was made by QuikChange Mutagenesis

©

(Stratagene). Fusion

protein constructs were made by the splicing-by-overlapping-extension method [48]. Constructs
were confirmed by sequencing and ligated into the PRK415 expression vector with KpnI and
HindIII restriction sites or the pJ411 (DNA 2.0

®

) expression vector with NdeI and EcoRI

restriction sites. The human TSPO gene was codon optimized for E. coli expression and
purchased from DNA 2.0 ® in the pJ411 vector. Expression vectors containing the desired TSPO
coding sequence were confirmed by sequencing and introduced into the E. coli expression strain
BL21 λDE3.
Protein expression and purification
The transformed bacteria were cultured at 30 °C in 100 mL LB medium containing
appropriate antibiotics overnight. Fifteen mL of the overnight culture were then transferred into 1
L of auto-induction media (ZYM-5025) [49] containing glycerol, glucose and lactose and the
appropriate antibiotics. Cultures were grown at 30 °C for RsTSPO, RsTSPOW38C and fusion
proteins and decreased to 12 °C for HsTSPO to reduce the inclusion bodies and GroEL. Cells
were harvested when the OD600 reached above 4.0. Harvested cells were resuspended in lysis
buffer (50 mM KH2PO4 pH 6.5, 1 mM EDTA) supplemented with Roche

®

protease inhibitor

cocktail and were lysed through two passes at 20,000 p.s.i in a French press homogenizer.
Broken cells were centrifuged at 200 x g for 5 mins, 10,000 x g for 30 mins, and 150,000 x g for
90 mins at 4 °C to isolate the membranes. Isolated membranes were resuspended with buffer A
(50 mM Tris-HCl, pH 8, 150 mM NaCl, 10% glycerol) and quick frozen with liquid nitrogen and
stored at -70 °C before purification.
46

Membranes were thawed on ice, resuspended to 10 mg/ml total protein in buffer A,
solubilized by adding 1 % DDM, 1 mM PMSF and Roche

®

protease inhibitor cocktail (no

EDTA) and stirred at 4 °C for 1 hr. Solubilized membranes were centrifuged at 150,000 x g for
30 mins to remove the unsolubilized fraction. The supernatant was loaded on a 10 mL Ni-NTA
(Qiagen) gravity column pre-equilibrated with buffer A. After extensive washing with buffer A
supplemented with 0.2 % DM and 50 mM imidazole, fractions containing the desired protein
were eluted using buffer A supplemented with 0.2 % DM and 300 mM imidazole over two
column volumes. Fractions containing the desired protein were pooled, concentrated with a 50
kDa cutoff Amicon filter (Millipore), washed once with buffer A supplemented with 0.2 % DM,
and diluted to 10 mL. Trypsin was added at 0.2 mg per 1 mg of RsTSPO10ht and the protein
solution was incubated at 4 °C overnight. Trypsin treated RsTSPO10ht was concentrated to 250
μL with a 30 kDa cutoff Amicon filter and loaded onto a Superose

®

12 300/10 gel filtration

column (GE Life Sciences) pre-equilibrated with buffer B (50 mM Tris-HCl, pH 8.0, 150 mM
NaCl, 0.2 % DM, 2 mM β-mercaptoethanol [βME]) and run at 0.3 mL/min. The homogeneous
peak fractions of RsTSPO were pooled and the concentration of the protein was determined by
UV absorbance at 280 nm. The purity of the protein was estimated on SDS-PAGE stained with
Coomassie blue. The average yield of wild-type RsTSPO was 5 mg per L of culture. RsTSPO
was also solublized and purified in Fos-Choline-12 (2 % and 0.2 %) using the same buffer
conditions but omitting the SEC step. This version of the enzyme was used for comparative
binding experiments (Figure 2.8).
Codon optimized HsTSPO was expressed at good levels in E. coli and was purified in the
same way as RsTSPO. However, the E. coli chaperone protein GroEL was found to co-purify
with HsTSPO as a persistent contaminant. As shown in Figure 2.3, a reasonably pure HsTSPO

47

monomer with a trace of dimer was present in the sample, but with a significant amount of
GroEL co-purified (Figure 2.3). The fortuitous absence of tryptophan in GroEL, combined with
the unusually high tryptophan content of TSPO (13 out of 169 amino acids), made it possible to
carry out some ligand binding studies with the human protein based on fluorescence quenching,
but not to characterize its oligomeric state.
Light scattering measurements of oligomeric state of purified RsTSPO
A static light scattering method combined with UV absorption and refractive index
measurement was used to investigate the oligomeric state of the purified RsTSPO [50]. Four uL
®

of a 30 mg/mL purified RsTSPO sample was loaded on a Superdex 200 300/10 column coupled
with the triple detector array (Malvern) and run in buffer B at 0.3 mL/min at 4 °C. Purified
ovalbumin and pure DM were run in the same buffer under the same conditions. Data analysis
was done in OmniSEC

®

software to calculate the molecular weight of the protein-detergent

complex and fraction of protein by comparing the light scattering, which represents the size of
the protein-detergent complex, and the amount of proteins in the complex as measured by UV280
(ε =71960 M-1 cm-1).
Tryptophan fluorescence quenching binding assay
A tryptophan fluorescence quenching assay was used to study the binding properties of
TSPO by utilizing the intrinsic tryptophan fluorescence [46]. For the RsTSPO and
RsTSPOW38C, 2.5 μM purified protein was titrated with increasing amounts of ligand in the
desired concentration range at room temperature. Each titration point was monitored by a
spectral scan from 290 nm to 400 nm (excitation at 285 nm) on a PTI QuantaMaster™
spectrofluorimeter. Control experiments were performed to evaluate the absorption/emission of
ligands as well as buffers and solvents. Due to the slight shift of the tryptophan fluorescence
48

peak during the experiment caused by differential quenching of buried tryptophans vs solvent
exposed ones, the area under each emission spectrum was integrated to account for all
tryptophans and a percentage quenching ratio for each concentration was calculated. The binding
curve was obtained by plotting the percentage quenching fQuen against total ligand concentration
[Lt] in Origin

©

. Equation (1) was derived based on a single binding site per monomer

assumption, taking into consideration ligand depletion. [Lt] is the total ligand concentration
plotted on the binding curves. The calculation of the Kd is corrected by Equation (1) for ligand
depletion which occurs at low ligand concentrations due the relatively high protein/ligand ratio.
fQuen  Offs  F max

( f [ Pt ]  [ Lt ]  Kd 

f ([ Pt ]  [ Lt ]  Kd ) 2  4 f [ Pt ][ Lt ]
2 f [ Pt ]

F max  Fi  Ff
Fi

(1)
(2)

fQuen represents the calculated percentage quenching at each concentration, Fmax is the
maximum amplitude of fluorescence quenching, which equals the difference between the initial
and final fluorescence signals divided by the initial fluorescence, [Lt] is the total ligand
concentration, [Pt] is the total concentration of purified RsTSPO protein (fixed at 2.5 μM in the
experiment and during the fitting procedure) and Kd is the apparent dissociation constant. Offs
(offset) allows a fit unconstrained to go through zero, while f is a parameter to account for the
effective fraction of protein involved in binding. Offs and f help to optimize the fitting by
accounting for binding modes involving different numbers of sites. For partially purified
HsTSPO with a total protein concentration at 2.5 μM, the binding curve was obtained the same
way as for RsTSPO but the apparent dissociation constant Kapp was estimated as the
concentration of the half-max quenching, since it is assumed that nonspecific binding to
49

contaminating GroEL may give a Kapp that is larger than the actual value. These HsTSPO
experiments were done with a mutant, C153E, that removed the possible complication of the
extra cysteine in the human protein and which showed no difference in terms of expression,
purification or reported function. All ligands are dissolved in DMSO as a stock solution and
diluted with DMSO or buffer for measurement except for cholesterol and PpIX, which were
dissolved in pure ethanol or 0.15 M NaCl, respectively. Control quenching curves with solvent
alone were carried out. One representative binding curve is shown in each figure while the Kd is
reported as the average of 3 replicate measurements fitted with Equation (1).
Modeling of the RsTSPO dimer
The experimental EM density map from the EM database (ID: EMDB-1698) was
processed in Chimera [51] to match published helix placement in a RsTSPO dimer density. Final
models were generated using a modified protocol from Lindert et al. [47] Twenty thousand
decoys were generated using a simultaneously combined BCL::EM-Fold and the membrane
protein protocol in BCL::MP-fold [52]. Briefly, the trans-membrane helices were first assembled
into the density rods as a monomer. The models were evaluated by their agreement with the
density map as well as a knowledge-based potential. The models were then clustered combining
all models with the same arrangement of trans-membrane helices (topology). The best scoring
150 topologies were then symmetrized and subjected to another round of EM-fold refinement
against the density map. The full-atom models were generated using Rosetta [53,54]; out of the
150 refined topological fold from previous step, 75 best scoring folds were used to generate 2000
all-atom models per fold. Then, within each topology, the models are clustered with respect to
loop conformation. For the best scoring 20 loop conformations from each of the 75 topologies,
three rounds of iterative loop modeling, side chain repacking, and relaxation were conducted

50

using the density map as a restraint in Rosetta [47,54]. Twenty-five models were created in each
iteration and the top-scoring models were moved forward to the next iteration. The final models
were ranked by the Rosetta full atom energy score [53-55]. All figures were generated in Pymol
v1.3. The simulation did not test topologies that include domain swaps.
RESULTS
RsTSPO forms a dimer in vitro
RsTSPO was successfully purified to homogeneity by nickel affinity followed by size
exclusion chromatography (SEC) (Figures 2.1, 2.2A). The molecular weight of purified RsTSPO
with a 10 histidine tag was determined to be 19371.76 ± 1.77 Da (17910.26 Da for un-tagged
RsTSPO) by mass spectrometry, correlating well with the calculated molecular weight from the
amino acid sequence (19347.58 Da). The SEC running profile in 0.2 % DM showed a peak that
suggested a dimer form based on elution profile and molecular weight standards. To obtain a
more accurate estimate of the size of RsTSPO, we characterized the purified RsTSPO in solution
by using a combination of light scattering, UV and refractive index measurements. A
representative run of the UV280 nm, refractive index and light scattering measurements is shown
in Figure 2.1A and a representative SDS-PAGE is shown in Figure 2.1B. The molecular weight
of the RsTSPO-detergent complex was determined to be ~100 kDa from the scattering peak, with
a protein fraction of ~36 % from comparison of scattering and 280 nm absorption; therefore, the
molecular weight of RsTSPO protein in the complex was calculated to be 37 kDa, indicating a
dimer of RsTSPO (given a monomer of 18 kDa) within the protein-detergent complex (Table
2.1). Purified ovalbumin and pure DM were also characterized under the same conditions.
Ovalbumin showed a calculated molecular weight of ~44 kDa as a monomer and DM showed a

51

micelle size of ~37 kDa; both agreed very well with literature values, and are consistent with the
equivalent of approximately 2 micelles of DM in the dimer/detergent complex.

52

Figure 2.1. Purification and determination of the molecular weight of RsTSPO.
(A) One representative trace of purified RsTSPO on the triple detector array (Malvern ®). The
peak of RsTSPO is labeled with a star and the molecular weight was calculated as described and
expressed on the Y axis. (B) SDS-PAGE for the purification of RsTSPO. The lane of
SEC-purified RsTSPO is labeled with a star corresponding with the light scattering profile. The
trypsin cutting mix represents the sample before SEC while a purified RsTSPO10ht sample was
use as the control.

53

Figure 2.2. Investigation of the oligomeric state of purified RsTSPO by SEC.
(A) SEC profile of RsTSPOWT, RsTSPOW38C and RsTSPOWT in the presence of 350 µM
PpIX on a prepacked Superose® 12 300/10 column. (B) SEC profile of RsTSPOW38C in the
presence and absence of βME on a homemade Superose® 12 column.

Table 2.1. Molecular weight calculation from static light scattering measurement
Sample

Mwcomplex

f protein

Mwprotein

1st
2nd
RsTSPO
3rd
4th
Ave.
Ovalbumin std
DM std

98430
104029
97328
111620
102852
42757
37243

0.347
0.365
0.372
0.362
0.3615
0.982
0

34155
37971
36206
40406
37181
41987
NA

Monomer
MwCal
17976
17976
17976
17976
17976
44287
NA

Nmonomer
1.9
2.1
2.0
2.2
2.1
0.95
NA

Note: Mwcomplex: molecular weight of the protein-detergent complex; fprotein: fraction of protein
in the protein-detergent complex; Mwprotein: molecular weight of the protein without detergent;
Mwcal: molecular weight calculated from sequence; Nmonomer: number of protein monomers in the
protein-detergent complex.

54

The question of whether ligands influence the oligomeric state was investigated by
incubating the RsTSPO with the ligand PpIX prior to SEC. The SEC trace in the presence of
PpIX showed the same size as that of the protein without ligand (Figure 2.2A). Since Yeliseev
and Kaplan reported the W38C mutant to stabilize a dimer form in R. sphaeroides [34], a further
question was whether a disulfide bond was involved in dimer formation, either through the
W38C mutant or the naturally existent cysteine, C15, in the first transmembrane helix. There was
in fact no apparent change in the amount or stability of the dimer form in the W38C mutant and
no effect of addition of ßME, as evidenced by no difference in the elution profile from SEC
(Figure 2.2A, B). Therefore a disulfide bond is unlikely to be important in dimer formation.
These observations and the lack of detectable monomer form in SEC suggest that the
dimer organization of RsTSPO in the DM micelle is a very strong association, but through
interactions other than disulfide bonds and independent of ligand binding. The persistence of a
small fraction of dimer even in the SDS gels run under strongly denaturing conditions, also
attests to the stability of the dimer.
We further explored the structure of the dimer in vivo by expressing in E. coli a dimer
created by a gene fusion of two monomers. Three fusion proteins of RsTSPO and HsTSPO were
constructed by making a direct connection between the C and N termini, resulting in a dimer
with an anti-parallel configuration (Figure 2.3) as suggested to occur in the case of EmrE [56], a
possible alternative to the parallel configuration supported by the cryo-EM and our modeling
studies of TSPO (see below). E. coli cells carrying the gene fusion constructs were characterized
by western blotting with antibodies against the histidine tag at the C-terminus of the fusion
proteins. Our results show that the full length fusion proteins with RsTSPO at the C-terminus and
human or Rhodobacter at the N-terminus (Hs-RsTSPO and Rs-RsTSPO) were expressed (Figure

55

2.4) and inserted in the membrane without formation of a significant amount of inclusion bodies,
implying a folded structure. In contrast, the dimer with the HsTSPO portion at the C-terminus
(Rs-HsTSPO) was not detected, suggesting that this version of the fusion protein may not be
correctly folded or inserted into the membrane.

56

Figure 2.3. Expression of TSPO fusion proteins.
(A) Topology illustration of fusion proteins. All three fusion proteins have the same anti-parallel
configuration. Shown in the figure is the Hs-RsTSPO fusion construct, in which a thrombin site
was placed in between the C-terminus of HsTSPO and the N-terminus of the RsTSPO. In the
cases of Rs-HsTSPO and Rs-RsTSPO, the fusion proteins are constructed by directly connecting
the C-terminus and the N-terminus. A 10-histag was constructed at the C-terminus of all the three
fusion proteins. (B) Western blot of the whole cells expressing fusion proteins with antibodies
against the C-terminal 10-histag. Lane 1: size marker; lane 2: Hs-RsTSPO10ht after 7 hrs growth;
lane 3: Hs-RsTSPO10ht after 22 hrs growth; lane 4: Rs-HsTSPO10ht after 7 hrs growth; lane 5:
Rs-HsTSPO10ht after 22 hrs growth; lane 6: Rs-RsTSPO10ht after 7 hrs growth; lane 7:
Rs-RsTSPO10ht after 22 hrs growth; lane 8: positive control of purified RsTSPO10ht. The fusion
proteins were also found to be mainly in the membrane fraction of the fractionated cells, based
on estimates from the yields of total membranes.

57

Although these results tell us little about the significance or existence of an anti-parallel
dimer of TSPO, the observation that expression and insertion of the dimer depends on whether
the Rs or Hs protein is at the C-terminal does suggest that membrane insertion of TSPO involves
a C-terminal mechanism that may require additional machinery in the case of the mammalian
protein, consistent with other evidence [57-59].
Overall, our studies are consistent with the dimer being the major structural unit of
RsTSPO as also implied by the cryo-EM studies of the purified RsTSPO and our modeling
analysis, which indicate a parallel dimer [46]. The possibility of an anti-parallel dimer and other
homo- or hetero-oligomeric forms in the native membrane remain to be explored.
Tryptophan fluorescence quenching assay as a sensitive measure of Kd
We tested the validity of the tryptophan fluorescence quenching assay by measuring the
binding of the known ligand, protoporphyrin IX (PpIX), to purified RsTSPO (Figure 2.4). We
determined a Kd for PpIX of 0.3 M, lower than reported by Korkhov et al. [46] in a previous
study (8.6 µM), possibly indicating the sensitivity of binding behavior to a variety of assay
conditions (see discussion). The non-ligand porphyrin precursor δ-aminolevulinic acid (ALA)
and the solvent DMSO were also measured and showed only a small quenching of the
tryptophan fluorescence compared to PpIX, as did the steroid ursodeoxycholate (see below). A
similar low micro-molar Kapp for PpIX (2.4 µM) was also obtained with HsTSPO (Figure 2.4), a
result within the range of previous reports of PpIX binding affinity for mammalian TSPO
[40,60]. (The presence of the contaminating protein GroEL, which has no tryptophan and does
not contribute to the fluorescence [see Methods], may, however, give a somewhat higher binding
constant for HsTSPO due to ligand depletion.)

58

Figure 2.4. Binding of PpIX, ALA, and ursodeoxycholate to RsTSPO and PK11195, PpIX
to HsTSPO.
(A) Binding curve of PpIX, ALA, and ursodeoxycholate to purified RsTSPO and PpIX to partial
purified HsTSPO measured with tryptophan fluorescence quenching assay. (B) Binding curve of
PK11195 to partially purified HsTSPO.

Figure 2.5. Purification of HsTSPO.
Lane 1: Coomassie stain of purified HsTSPO; lane 2: Coomassie stain of purified RsTSPO as
control; lane 3: western blot against histag of purified HsTSPO; lane 4: western blot against
histag of purified RsTSPO as control.

59

We also tested the sensitivity of the assay by measuring the binding of HsTSPO to the
diagnostic ligand PK11195 with expected binding affinity in the nano-molar range. Our
measurements show two Kapp of ~ 0.2 nM and 5 nM for PK11195 (Figure 2.4B), which agree
with previous reports [35] and two different nano-molar binding affinities for PK11195 in both
mouse liver [8] and a human Leydig cell line [61]. The results demonstrate the capability of the
assay to characterize high-affinity binding.
Effect of detergent conditions on the binding assay
It was also important to find detergent conditions that maintained a stable protein and
were compatible with the assay. The nano-molar binding affinity of PK11195 for HsTSPO
(Figure 2.5B) was measured in 0.2 % DM. However, a previous report suggested that binding of
PK11195 with mouse TSPO in the detergent SDS resulted in complete loss of affinity,
emphasizing the importance of detergent conditions [39]. We tested the binding of PK11195
with RsTSPO in two different concentrations of DM, as well as in the common detergent used in
NMR, Fos-Choline-12 (Figure 2.6). In 0.1 % DM, the binding affinity of purified RsTSPO for
PK11195 was slightly increased compared to 0.2 % DM (7 μM vs 10 μM, respectively), while
Fos-Choline-12 lowered the affinity to ~ 120 μM. Therefore, for standard assay conditions for
ligand comparisons, 0.2 % DM was selected to keep purified RsTSPO stable while exhibiting
good affinity for PK11195. (0.1 % DM is just below the critical micelle concentration (CMC) of
DM and has a tendency to allow protein precipitation over time).

60

Figure 2.6. Binding curves of PK11195 with RsTSPO in different detergent conditions as
described in Methods.

61

RsTSPO binds PK11195 and cholesterol at micro-molar affinity
In the pioneering study in 1997 [32], Yeliseev and colleagues reported that rat TSPO
overexpressed in Rhodobacter sphaeroides 2.4.1 displayed an affinity for PK11195 ranging from
4-12 nM, while the wild-type RsTSPO showed no binding affinity above background. However,
in our studies with purified RsTSPO, the binding affinity of PK11195 was found to be 10 ± 1
μM, suggesting a binding site in RsTSPO with lower but significant affinity compared to the
mammalian proteins (Figure 2.7 and Table 2.2). In addition, the RsTSPO form containing a
W38C mutation located in the middle of loop 1 where both PpIX and PK11195 are proposed to
bind [34,35] decreased the binding affinity of PK11195 by 3-fold. These results indicate that
loop 1 contributes to PK11195 binding. PpIX binding to RsTSPO was also influenced by the
W38C mutation (Figure 2.7 and Table 2.2), showing a 6-fold decrease in affinity (2 μM vs 0.3
μM), consistent with both binding sites involving W38C in the conserved region of loop 1.
Although it is suggested PK11195 and PpIX bind in the same general area (loop 1) in both
RsTSPO and HsTSPO, it should be noted that loop 1 is predicted to be flexible and shorter in
RsTSPO than in HsTSPO, and the corresponding residue of W38 in HsTSPO is S41. These
differences would be expected to impact the affinity and specificity of ligand binding.

62

Figure 2.7. Binding of PpIX, PK11195 and cholesterol to purified RsTSPO (WT) and
RsTSPOW38C (W38C).
All binding curves were obtained and plotted as described in the Methods. The inset shows a
blow-up of the 0-20 µM region to better show the binding curve of the PpIX.

63

Table 2.2. Structure and Kd of ligands to RsTSPOWTand W38C mutant
Kd to RsTSPO
(μM)

Kd to
RsTSPOW38C
(μM)

Bilirubin

4.7 ± 0.2

5.5 ± 0.3

Cholesterol

Kapp = 80 1

Kapp = 200 1

Curcumin

1.4 ± 0.1

2.6 ± 0.1

Gossypol

6.0 ± 0.2

3.9 ± 0.3

PK11195

10 ± 1

34 ± 2

Protoporphyrin IX
(PpIX)

0.31 ± 0.01

2.0 ± 0.07

Retinoic acid

0.50 ± 0.02

1.6 ± 0.1

ligands

Note:

Structure

1: Kapp is estimated from half-maximal quenching
2: Kd is obtained from fitting 3 independent titrations with equation (1) and standard
error is reported.
64

Since cholesterol does not naturally occur in Rhodobacter, no data has been reported on
its binding to RsTSPO. However, Rhodobacter is one of the closest ancestors of mitochondria
[28] and in fact has structural and functional analogs of cholesterol, hopanoids [37], in the
membrane where RsTSPO is located. In addition, the TSPO sequence alignment (Figure 1.6)
shows a remarkably high similarity at the proposed cholesterol binding site (CRAC:
-L/V-(X)(1-5)-Y-(X)(1-5)-R/K-) [36,62] across many species. In particular, 2 out of the 3 proposed
critical residues in the CRAC sequence (Leu and Arg) are identical between RsTSPO and
HsTSPO while the other is a conservative substitution of phenylalanine for tyrosine. Therefore,
we expected that RsTSPO might also bind cholesterol. Due to the low solubility of cholesterol in
aqueous solution, we were unable to determine an accurate Kd, but we estimate from the binding
curve a Kapp of ~ 80 μM to RsTSPO wild-type and ~ 200 μM for the W38C mutant (Figure 2.7),
significantly lower affinity than reported for mammalian TSPO [27], possibly due to the variable
region immediately preceding the CRAC sequence in helix V (residues 144-147) including the
site of a human polymorphism [38,63]. In contrast, the steroid derivative ursodeoxycholate
showed no detectable affinity for RsTSPO (Figure 2.5A), suggesting that the binding of
cholesterol, albeit low, is not just due to non-specific interactions.
Competition binding assay of PpIX, PK11195 and cholesterol
Loop 1 in the N terminal region has been proposed to be where drugs and porphyrin bind
[27,64], while the cholesterol CRAC site is located at the C-terminal region of TSPO [36].
However, the small size of TSPO and the fact that it may function as dimer or higher oligomer
introduces the possibility that these two sites are very close to each other and interacting.
Nevertheless, in the case of PpIX and PK11195, we observe some difference in spectroscopic
properties in the binding assay. The tryptophan fluorescence of RsTSPO and HsTSPO shows a
65

broad spectrum from 290 nm to 400 nm with a peak around 340 nm. PpIX quenches the
tryptophan fluorescence evenly at all wavelengths and shows a maximum quenching close to 100
% (Figure 2.8 A, B).

This indicates that the local environment of almost all tryptophans that

contribute to the total tryptophan fluorescence for RsTSPO is influenced by the binding of PpIX.
In contrast, PK11195 quenches the fluorescence of RsTSPO from 290 nm to 335 nm completely
but not in the region of 330-340 nm (Figure 2.8C). This behavior suggests that PK11195 is
interacting with a slightly different subset of the tryptophans compared to PpIX, implying some
differences in their binding sites in the case of RsTSPO.

66

Figure 2.8. Tryptophan fluorescence quenching behavior of PpIX and PK11195.
Tryptophan fluorescence spectra of (A) RsTSPO when binding with PpIX in 0.2 % DM; (B) HsTSPO when binding with PpIX in 0.2
% DM; (C) RsTSPO when binding with PK11195 in 0.2 % Fos-Choline-12. (PK11195 titration experiments done in 0.2 % DM show
the same spectral behavior as the experiment shown, done in 0.2% Fos-Choline-12).

67

To further investigate possible overlap between the binding sites, competition binding assays
were performed between PpIX, PK11195 and cholesterol (Figure 2.10). Purified RsTSPO was
pre-incubated with either 50 μM cholesterol or 20 μM PK11195 or both on ice before the
binding experiment.
The binding behavior of PpIX was somewhat affected by the presence of cholesterol,
giving a Kd increase of 2-3 fold, suggesting some influence of cholesterol on the PpIX binding
site. Interestingly, the presence of PK11195 caused a similar change in the binding of PpIX, ~ 3
fold increase in Kd, and PK11195 and cholesterol together had a greater effect of ~4 fold increase
in Kd. This behavior (Figure 2.9) suggests that PK11195 and PpIX binding sites are interacting
and that cholesterol is also altering the PpIX binding in manner that is somewhat additive with
PK11195, consistent with partial overlap or allosteric effects of ligand binding at different sites.

68

Figure 2.9. Competition binding studies between PpIX, PK11195 and cholesterol to
RsTSPO.
Experiments are done as described in Methods. (A) Competition between PpIX and cholesterol
(preincubated at 50 µM cholesterol) and PK11195 (preincubated at 20 µM PK11195) as well as
PK11195 plus cholesterol (preincubated with both). (B) Competition between PK11195 and
cholesterol (preincubated at 50 µM cholesterol).

69

Compounds affecting apoptosis bind to RsTSPO at micro-molar affinity
Besides the TSPO ligands PK11195, benzodiazepines, and PpIX, which all have been
reported to either induce or prevent apoptosis [27], other compounds have been reported to affect
apoptosis in human cell lines, but their targets are not well defined. In R. sphaeroides, TSPO is
reported

to

facilitate

the

transport

of

intermediates

of

heme

and

carotenoid

biosynthesis/breakdown pathways through the outer membrane [31,33,34]. Interestingly, in
mammals the porphyrin breakdown product bilirubin and some carotenoid homologs (retinoids)
have also been reported to induce apoptosis through the mitochondrial pathway [65-67]. These
and other compounds reported to have effects on MPTP and apoptosis, including curcumin [68]
and gossypol [69], were selected to test for their binding affinity to RsTSPO. Binding studies
with both wild-type and the W38C mutant show that several of these compounds bind to purified
RsTSPO at low micro-molar affinity similar to PpIX (Figure 2.10 and Table 2.2), which is
suggestive that some of these apoptosis-inducing agents could function through direct interaction
with TSPO.

70

Figure 2.10. Binding curves of compounds affecting apoptosis: curcumin, retinoic acid,
gossypol, and bilirubin

71

Models of the RsTSPO dimer
The structural model of the RsTSPO dimer was constructed using EM-Fold and Rosetta
guided by the 10Å cryo-EM density map of RsTSPO [46]. The two best-scoring models that
were energetically more favorable (Figure 2.11) are displayed in Figure 2.12. The five next
best-scoring topologies are provided in Figure 2.13 for comparison. At 10 Å resolution in the
cryo-EM, the connectivity between trans-membrane helices was not resolved. The objective of
the modeling study was to complete the model and use the Rosetta energy function to identify
the global energy minimum, thereby adding atomic detail not visible in the experimental data.
The seven topologies shown are energetically significantly more favorable than all alternative
topologies (Figure 2.11). The two best scoring topologies have a significant energy gap of 28 to
the best scoring model of an alternative topology. Therefore, we consider these two topologies as
most likely for RsTSPO.

72

Rosetta energy score of the top 20
topological fold
-1500
-1520
-1540
-1560
-1580
-1600
-1620

-1640

Figure 2.11. Rosetta energy score shows that the top 7 models are significantly better than
the rest.

73

Figure 2.12. Top two models of a RsTSPO dimer.
Model 1: (A) Side view, (B) top view, (C) bottom view. Model 2: (D) Side view, (E) top view, (F)
bottom view. Electron density from the EM structure was contoured at 1.5 σ. Each monomer is
colored with rainbow colors from blue to red corresponding to the N terminus to C terminus.

74

Figure 2.13. Top 3-5 scoring models of RsTSPO dimer.
Models are displayed as rainbow diagram in order of ranking score. (A) Model 3, (B) Model 4,
(C) Model 5, (D) Model 6, (E) Model 7.

75

Some common features are observed in the two top scoring models although the two
RsTSPO monomers are placed in opposite orientations within the EM density and have a
different dimer interface (Figure 2.12). Both models place helix I and helix V close to each other,
consistent with previous mutagenesis data [35]. In addition, helix I and helix II are on the same
face, providing the possibility of helices I and II playing a role in substrate binding and transport
together (see Discussion for details). The main differences are: 1) model 1 placed helix V in the
dimer interface while the dimer interface in model 2 is composed of helices I, II and III; 2)
model 1 has a relatively shorter loop 1 and a portion of predicted loop 1 forms a short helix that
contributes to part of the transmembrane region, as compared to model 2. However, ligand
binding at loop 1 is very likely to change the structure of this flexible region. Considering that
TSPO has been reported to form higher oligomers [61], we fit the dimer into the larger density of
the RsTSPO helical crystal (Figure 2.14) and found an additional (though less strong) dimer
interface mediated by helices I and II in model 1, as well as an additional dimer interface
mediated by helices I and V in model 2. We have focused our binding analysis on model 2,
which portrays helices I and II as being at the dimer interface while helix V is at the outside,
leaving open the possibility of helix V interacting with another monomer, as suggested in Figure
2.14, or a partner protein for cholesterol translocation, as suggested by Papadopoulos et al. [57].

76

Figure 2.14. Oligomerization of RsTSPO.
Models of RsTSPO dimers were placed into the extended cryo EM electron density of the helical
polymer crystal. (A) and (B): model 1; (C) and (D): model 2.

77

DISCUSSION
The ability of TSPO to bind to various ligands has been extensively investigated over the
past 30 years, providing important information that has led to our current understanding of the
function of TSPO. However, previous studies show significant variation in the binding properties
of TSPO when measured in different cell lines or tissues or purified forms. It has also been
suggested that TSPO functions in vivo as an oligomer [61], but details of the oligomerization and
conditions that foster it are difficult to establish. To address these and other issues, we purified
the homolog of human TSPO from R. sphaeroides and characterized the oligomeric state and
binding behavior in a well-defined system. The bacterial protein RsTSPO is important in itself,
given its role in photosynthetic gene regulation and membrane transport in bacteria. The
conserved aspects of sequence, along with some differences in ligand binding between the
mammalian and R. sphaeroides proteins, provide opportunities to identify the potential
determinants for the ligand interactions. In this study, we confirm that RsTSPO forms a dimer in
vitro and present models with atomic detail for the dimer form. Using one of these models,
options for PpIX, PK11195 and cholesterol interaction sites are proposed. The binding behavior
of other physiologically important ligands provides some new hints of the mechanism of
potential regulation of MPTP by TSPO.
Evidence relating to the role of the dimer
In this work, we characterized the minimum structural unit of purified RsTSPO as a
dimer. It is reasonable to postulate that the minimum functional unit of RsTSPO is also a dimer,
considering several lines of evidence.
Because TSPO was first discovered and is best known as a receptor, the drug binding
properties and the physiological effects of drug-related ligands were initially extensively

78

investigated. However, evidence has been obtained in both mammalian and Rhodobacter
systems that TSPO also transports molecules including both cholesterol and the metabolites of
heme and photosynthesis pathways [6,34,62]. It is reasonable to propose a monomer form of
TSPO functions as a receptor, but given the small size of the protein, it is more difficult to
postulate a mechanism for a monomer to transport either cholesterol or the heme/carotenoid
metabolic intermediates.

Yeliseev and Kaplan proposed that RsTSPO forms a dimer in vivo

and this dimer was stabilized by the W38C mutation [34]. Porphyrin has been reported to induce
dimerization [34], although in our hands the already dimeric state of RsTSPO was not influenced
by the addition of PpIX. In our studies, no monomer form was ever observed under
non-denaturing conditions during chromatography with or without βME, while a trace of dimer
remained even under the strong denaturing conditions of urea/SDS-PAGE. In addition, the
W38C mutant behaved very similarly to the wild-type in terms of dimer state, suggesting that
RsTSPO naturally forms a dimer by strong associations other than disulfide bonding. Similarly,
the E. coli drug transporter EmrE [70] forms a dimer in the purified state that is uninfluenced by
ligand binding. Biochemical and NMR data also show that EmrE can transport substrate as a
dimer in vivo [71] and in vitro [72], while structural studies show a ligand at the dimer interface.
In fact, a significant number of transporters function as dimers and/or show two-fold symmetry
with substrate binding at the dimer or domain interface in the crystal structures [73,74],
suggesting a common mechanism for substrate transport that could also apply to TSPO.
In view of the case of EmrE, which shows that a membrane protein can be inserted with
dual topology [56] and given the apparently successful expression of our forced anti-parallel
dimers, we considered the possibility that TSPO might also function as an anti-parallel structure
to create a “5+5 inverted repeat” topology as reported in several cases [73,75]. Studies (reviewed

79

by Bowie [76]) suggest that rearrangement of the orientation of helices is possible as well as
necessary in several cases for both large and small membrane proteins. However, the best
cryo-EM and modeling evidence available so far indicates a parallel dimer, as modeled in this
paper.
Taken together, our data as well as other studies indicate that RsTSPO functions as a
dimer, which is likely to be important for transporting substrates across the membrane. On the
other hand, current models regarding the effect of drugs generally agree with the 1:1 drug to
monomer ratio, suggesting that TSPO could perform receptor and transporter functions in
different oligomer and hetero–oligomer states.
Effect of detergent on the structure and binding of RsTSPO
Detergents play a critical role in the study of membrane proteins, allowing their
characterization in a purified state. However, these artificial lipid-mimicking molecules may
influence the binding of hydrophobic ligands, as observed in the case of a steroid binding site in
cytochrome c oxidase where a systematic analysis of detergent effects led to a clearer
understanding of competitive behavior between ligands and detergents and a more precise
definition of the binding domain [77,78].
In this study, we also observed the influence of detergent upon the binding of ligands to
purified RsTSPO (Figure 2.6).

Lacapere et al. [27] had previously reported that recombinant

mouse TSPO lost its binding to PK11195 when purified with SDS as detergent but regained the
binding ability when reconstituted into liposomes. In their hands, the binding of cholesterol was
less influenced than the binding of PK11195 by SDS. In our studies, when a non-ionic detergent
DM is used, a higher affinity PK11195 binding site appears to be maintained, as compared to
using the charged detergents Fos-Choline-12 and SDS. Our observations are consistent with the

80

model that the PK11195 binding site involves the flexible extra-membrane loop 1 and the
membrane interface where detergent/lipid head groups may have an impact, while cholesterol
bound at the more deeply buried CRAC site on helix V may be less impacted by the detergent.
Using the TSPO dimer model to predict the binding sites for PpIX, cholesterol, and
PK11195
Although we have demonstrated binding of both PK11195 and cholesterol to purified
RsTSPO, it is still a question as to what structural features result in an affinity for both ligands
that is several orders of magnitudes lower than in mammalian TSPO. To attempt to obtain new
insight into possible ligand interactions, models of RsTSPO in the dimer form were generated
(Figure 2.12). The top ranked candidate structures were compared and one (Model 2) was used
to facilitate an interpretation of the binding behaviors of PpIX, cholesterol, and PK11195 to
RsTSPO.
Our preliminary competition binding studies (Figure 2.9) give evidence of some level of
interaction between all three binding sites. The effects of PK11195 and cholesterol on the
affinity of RsTSPO for PpIX are almost additive, while the binding of PK11195 itself is less
influenced by cholesterol. These results are consistent with PK11195 and PpIX interacting at
somewhat different positions on the extra-membrane loop 1, and cholesterol acting at helix V of
the C-terminus, with some allosteric influence on both.
We propose that PpIX, as one of the potential endogenous transport substrates of TSPO,
binds initially at loop 1, en route to transport via the dimer interface, as illustrated in Figure 2.15.
Previous in vivo studies, in particular mutation of residues W30, K36, W38, and W39 in RsTSPO
[34], as well as our data, show that loop 1 plays a critical role in the binding and export of
porphyrins. However, loop 1 is not likely to be the only contributor to the binding of PpIX. The

81

observation that the tryptophan fluorescence was completely quenched by PpIX but not by
PK11195 and other ligands binding (Figure 2.8) suggests that either PpIX interacts with a rather
large area or induces a global conformational change, both of which are possible if PpIX is
binding and being transported through the dimer interface. Initial PpIX binding at loop 1 is
consistent with our observation that the W38C mutant reduced the affinity of PpIX for RsTSPO
by ~ 6 fold, while aromatic residues on transmembrane helix II, including highly conserved W44
and W50, also appear to play an important role in PpIX binding and transport, as evidenced by
lower porphyrin export activity in phenylalanine mutants of both residues [34]. Our model
(Figure 2.15) is consistent with helix II being a major part of the dimer interface where W44 and
W50 are in the region of potential close interaction with PpIX.

82

Figure 2.15. Proposed model of binding sites of PpIX, PK11195 and cholesterol on RsTSPO.
Structures of PK11195 and PpIX were generated in ChemDraw ® 3D and cholesterol was taken
from a published structure (PDB# 3D4S). Note that the computational model of RsTSPO (Model
2) is generated based on EM density observed in a sample without ligand. Therefore, these
models may represent a closed conformation, as indicated by close association of the monomers.
Ligands are manually placed into the model of RsTSPO on one monomer, for illustration of
proposed locations. Helices are labeled with Roman numerals. W38 (green sticks); PK11195
(orange sticks); PpIX (dark purple sticks); cholesterol (yellow sticks). Highly conserved residues
in the CRAC site are colored in light yellow and labeled according to the Rhodobacter sequence.
Corresponding residues in the human sequence are shown in parentheses. Human polymorphism
residue A139 (Rs number) is colored in dark blue, while the preceding Ala-Thr-Ala sequence is
colored in cyan. The sequence in loop 1, before W38, that is proposed to play a role in PK11195
binding is colored in light orange corresponding to the ligand. The proposed PpIX binding
sequence on loop 1 (including W44) is colored in light purple and the residue W50 on helix II is
colored in pink. A loading and transport route for PpIX is indicated by a magenta dotted line,
illustrating the potential role of the dimer in transport, which would require flexibility of the
external loops.

83

Although PK11195 was also proposed to bind to TSPO at loop 1 and has been reported to
compete with PpIX binding [40], the tryptophan quenching behavior suggests that PK11195 and
PpIX do not bind in precisely the same location (Figure 2.8). As mentioned above, the last 2/3 of
the loop 1 is highly conserved across all species, with conserved residues that play critical roles
for PpIX binding, while the sequence in the first 1/3 is not. Thus the much lower binding affinity
of PK11195 for RsTSPO compared to HsTSPO could be due in part to differences in the early
part of loop 1. However, some overlap between PK11195 and PpIX binding is suggested by
competition studies and the fact that the Kd of both PK11195 and PpIX were markedly reduced
by the W38C mutant. Based on these independent lines of evidence, here we propose that
PK11195 interacts mainly on the first half of loop 1 within the monomer, extending to residue
W38 (S41 in human), and also harnesses some of the structural features of the helix I/membrane
interface and the second half of loop 1 that binds the hydrophobic porphyrins. This interpretation
is consistent with the observation that RsTSPO binds both PK11195 and PpIX at micro-molar
affinity (Figure 2.6), while HsTSPO has nano-molar affinity for PK11195 but has similar affinity
as RsTSPO for PpIX (Figure 2.6) [40]. Nevertheless, given the flexible nature of the loop, a high
resolution structure with ligand bound will be essential to understand the details of the binding
mechanisms.
In this study, we show that cholesterol does bind to purified RsTSPO but only with low
affinity in the high micro-molar range. We note that the helix V and the C-terminus, where the
CRAC sequence is located, is perhaps the most conserved region in the whole protein. However,
the region right before the CRAC is much less conserved and shows some interesting features
(Figure 1.6 and Figure 2.15). In the human and mouse TSPO proteins, where nano-molar
affinities have been reported, a relatively hydrophobic sequence Leu-Ala-Phe (144-146)

84

precedes the CRAC, whereas in the R. sphaeroides sequence these residues are replaced by
Ala-Thr-Ala (136-138), a much more hydrophilic combination. In fact, the spontaneous human
polymorphism A147T, located right after the Leu-Ala-Phe, has been reported to reduce
pregnenolone production [63] and binding of an imaging ligand by 2 orders of magnitude [38].
Our model of RsTSPO (Figure 2.15) would allow the long hydrophobic tail of cholesterol to
extend to the location of Ala-Thr-Ala, while the ring structure is interacting mainly with the
CRAC site. The more hydrophobic character of the human helix V sequence may create a deeper
binding site or reposition helix V to favor binding interactions for the hydrophobic rings.
Another interesting question is whether this region of helix V and cholesterol binding
also affects other ligands. Besides our observations, extensive investigation of the binding of the
original benzodiazepine ligand Ro5-4864 has concluded that the C-terminal region of TSPO, in
particular residues 144-156 of helix V, which include both the CRAC site and the sequence
preceding, plays a role in its binding [35] although involvement of loop 1 is also observed. The
new PET ligand PBR28, a PK11195 related compound, shows a markedly reduced affinity for
the HsTSPO polymorphism A147T [38]. The same mutation in bacterial TSPO was recently
reported to disrupt the binding of PpIX [79]. All together, helix V and the cholesterol binding
region do appear to influence both PpIX and drug ligands. Our two top models (Figure 2.12 and
2.14) show that helix V and helix I are in close proximity within the dimer as well as the
higher-order oligomer, providing the possibility of direct interaction between binding sites.
From these considerations, a tentative model (Figure 2.15) for the ligand binding sites of
TSPO is proposed, suggesting that TSPO may perform its dual functions as transporter and
receptor by binding the substrate PpIX at the dimer interface for transporting, while binding
PK11195 and other ligands more superficially on loop 1. Cholesterol is proposed to bind at the

85

CRAC site located on helix V and we cite evidence that the 4 residues immediately preceding the
CRAC site are also important for the binding of cholesterol and other ligands. In Model 1, these
residues and the cholesterol binding site are located at the dimer interface, suggesting testable
differences between the two models.
TSPO as a mediator of the regulation of MPTP and apoptosis.
Although numerous studies have implicated TSPO as a regulator of MPTP and apoptosis,
the mechanism of the regulation is still unclear. In fact, Gonzalez-Polo et al. [42] reported that
the diagnostic TSPO ligand PK11195 shows apoptosis-inducing effects independent of TSPO.
Another important regulator of MPTP and apoptosis is the B-cell lymphoma 2 (Bcl-2) family of
proteins (see review by R. Youle [80]), which can be either anti-apoptotic, such as Bcl-2 and
Bcl-XL, or pro-apoptotic, as in the case of Bax, Bak and Bid. Bcl-2 family proteins are expressed
in the cytosol and recruited during the process of apoptosis to the MPTP on the outer
mitochondrial membrane, where TSPO is also localized. The fact that both TSPO and Bcl-2
family proteins regulate apoptosis and that the TSPO ligand PK11195 can reverse the
cytoprotective effect of Bcl-2 [27] suggests a possible interaction of TSPO and Bcl-2 in the
regulation of MPTP.
In this study we report binding of a number of apoptosis-inducing compounds to purified
RsTSPO with low micro-molar affinity, including PK11195, retinoic acid, curcumin, and
bilirubin. This is consistent with a direct interaction of TSPO with these ligands and supports a
role for TSPO during apoptosis. In addition, the binding of a known Bcl-2 inhibitor, gossypol, to
purified RsTSPO at low micro-molar affinity suggests a relationship between TSPO and Bcl-2
family proteins.
CONCLUSIONS

86

The current studies characterize the oligomeric state of purified TSPO from R. sphaeroides
as a dimer and report the binding properties of a variety of ligands. New atomic resolution
models of its dimer form are presented and used to predict TSPO’s interaction with various
ligands. Several novel ligands are reported, with implications for the role of TSPO in apoptosis
and for facilitation of the crystallization of TSPO.

87

REFERENCES

88

REFERENCES
1. Braestrup C, Squires RF: Specific benzodiazepine receptors in rat brain characterized by
high-affinity (3H)diazepam binding. Proc Natl Acad Sci U S A 1977, 74:3805-3809.
2. Papadopoulos V, Baraldi M, Guilarte TR, Knudsen TB, Lacapere JJ, Lindemann P, Norenberg
MD, Nutt D, Weizman A, Zhang MR, et al.: Translocator protein (18 kDa): new
nomenclature for the peripheral-type benzodiazepine receptor based on its structure
and molecular function. Trends Pharmacol Sci 2006, 27:402-409.
3. Fan J, Lindemann P, Feuilloley MG, Papadopoulos V: Structural and functional evolution of
the translocator protein (18 kDa). Curr Mol Med 2012, 12:369-386.
4. Krueger KE: Molecular and functional properties of mitochondrial benzodiazepine
receptors. Biochim Biophys Acta 1995, 1241:453-470.
5. Anholt RR, Pedersen PL, De Souza EB, Snyder SH: The peripheral-type benzodiazepine
receptor. Localization to the mitochondrial outer membrane. J Biol Chem 1986,
261:576-583.
6. Papadopoulos V, Amri H, Boujrad N, Cascio C, Culty M, Garnier M, Hardwick M, Li H, Vidic
B, Brown AS, et al.: Peripheral benzodiazepine receptor in cholesterol transport and
steroidogenesis. Steroids 1997, 62.
7. Li H, Papadopoulos V: Peripheral-type benzodiazepine receptor function in cholesterol
transport. Identification of a putative cholesterol recognition/interaction amino acid
sequence and consensus patterns. Endocrinology 1998, 139:4991-4997.
8. Taketani S, Kohno H, Furukawa T, Tokunaga R: Involvement of peripheral-type
benzodiazepine receptors in the intracellular transport of heme and porphyrins. J
Biochem 1995, 117:875-880.
9. Verma A, Nye JS, Snyder SH: Porphyrins Are Endogenous Ligands for the Mitochondrial
(Peripheral-Type) Benzodiazepine Receptor. Pro Natl Ac Sci U S A 1987,
84:2256-2260.
10. Krueger KE, Papadopoulos V: Mitochondrial benzodiazepine receptors and the
regulation of steroid biosynthesis. Annu Rev Pharmacol Toxicol 1992, 32:211-237.
11. Chen MK, Guilarte TR: Translocator protein 18 kDa (TSPO): molecular sensor of brain
injury and repair. Pharmacol Ther 2008, 118:1-17.
12. Brown RC, Degenhardt B, Kotoula M, Papadopoulous V: Location-dependent role of the
89

human glioma cell peripheral-type benzodiazepine receptor in proliferation and
steroid biosynthesis. Cancer Lett 2000, 156:125-132.
13. Hardwick M, Fertikh D, Culty M, Li H, Vidic B, Papadopoulos V: Peripheral-type
benzodiazepine receptor (PBR) in human breast cancer: correlation of breast cancer
cell aggressive phenotype with PBR expression, nuclear localization, and
PBR-mediated cell proliferation and nuclear transport of cholesterol. Cancer Res
1999, 59:831-842.
14. Ji B, Maeda J, Sawada M, Ono M, Okauchi T, Inaji M, Zhang MR, Suzuki K, Ando K,
Staufenbiel M, et al.: Imaging of peripheral benzodiazepine receptor expression as
biomarkers of detrimental versus beneficial glial responses in mouse models of
Alzheimer's and other CNS pathologies. J Neurosci 2008, 28:12255-12267.
15. Crompton M: The mitochondrial permeability transition pore and its role in cell death.
Biochem J 1999, 341 ( Pt 2):233-249.
16. Decaudin D, Castedo M, Nemati F, Beurdeley-Thomas A, De Pinieux G, Caron A, Pouillart P,
Wijdenes J, Rouillard D, Kroemer G, et al.: Peripheral benzodiazepine receptor
ligands reverse apoptosis resistance of cancer cells in vitro and in vivo. Cancer
Research 2002, 62:1388-1393.
17. Campanella M: Editorial: The physiology and pharmacology of the mitochondrial 18
kDa translocator protein (TSPO): an emerging molecular target for diagnosis and
therapy. Curr Mol Med 2012, 12:355.
18. Gatliff J, Campanella M: The 18 kDa translocator protein (TSPO): a new perspective in
mitochondrial biology. Curr Mol Med 2012, 12:356-368.
19. Rupprecht R, Papadopoulos V, Rammes G, Baghai TC, Fan J, Akula N, Groyer G, Adams D,
Schumacher M: Translocator protein (18 kDa) (TSPO) as a therapeutic target for
neurological and psychiatric disorders. Nat Rev Drug Discov 2010, 9:971-988.
20. Xiao J, Liang D, Zhang H, Liu Y, Li F, Chen YH: 4'-Chlorodiazepam, a translocator
protein (18 kDa) antagonist, improves cardiac functional recovery during
postischemia reperfusion in rats. Exp Biol Med (Maywood) 2010, 235:478-486.
21. Kroemer G, Dallaporta B, Resche-Rigon M: The mitochondrial death/life regulator in
apoptosis and necrosis. Annu Rev Physiol 1998, 60:619-642.
22. Tait SW, Green DR: Mitochondria and cell death: outer membrane permeabilization and
beyond. Nat Rev Mol Cell Biol 2010, 11:621-632.

90

23. Jiang X, Wang X: Cytochrome C-mediated apoptosis. Annu Rev Biochem 2004, 73:87-106.
24. McEnery MW, Snowman AM, Trifiletti RR, Snyder SH: Isolation of the mitochondrial
benzodiazepine receptor: association with the voltage-dependent anion channel and
the adenine nucleotide carrier. Proc Natl Acad Sci U S A 1992, 89:3170-3174.
25. Leung AW, Halestrap AP: Recent progress in elucidating the molecular mechanism of the
mitochondrial permeability transition pore. Biochim Biophys Acta 2008,
1777:946-952.
26. Giorgio V, von Stockum S, Antoniel M, Fabbro A, Fogolari F, Forte M, Glick GD, Petronilli
V, Zoratti M, Szabó I, et al.: Dimers of mitochondrial ATP synthase form the
permeability transition pore. Proc Natl Acad Sci U S A 2013.
27. Krestinina OV, Grachev DE, Odinokova IV, Reiser G, Evtodienko YV, Azarashvili TS:
Effect of peripheral benzodiazepine receptor (PBR/TSPO) ligands on opening of
Ca2+-induced pore and phosphorylation of 3.5-kDa polypeptide in rat brain
mitochondria. Biochemistry (Mosc) 2009, 74:421-429.
28. Bui ET, Bradley PJ, Johnson PJ: A common evolutionary origin for mitochondria and
hydrogenosomes. Proc Natl Acad Sci U S A 1996, 93:9651-9656.
29. Armstrong GA, Alberti M, Leach F, Hearst JE: Nucleotide sequence, organization, and
nature of the protein products of the carotenoid biosynthesis gene
cluster of Rhodobacter capsulatus. Mol Gen Gen 1989, 216, 254-268.
30. Baker ME, Fanestil DD: Mammalian peripheral-type benzodiazepine receptor is
homologous to CrtK protein of Rhodobacter capsulatus, a photosynthetic bacterium.
Cell 1991, 65:721-722.
31. Yeliseev AA, Kaplan S: A sensory transducer homologous to the mammalian
peripheral-type benzodiazepine receptor regulates photosynthetic membrane
complex formation in Rhodobacter sphaeroides 2.4.1. J Biol Chem 1995,
270:21167-21175.
32. Yeliseev AA, Krueger KE, Kaplan S: A mammalian mitochondrial drug receptor
functions as a bacterial "oxygen" sensor. Proc Natl Acad Sci U S A 1997,
94:5101-5106.
33. Yeliseev AA, Kaplan S: A novel mechanism for the regulation of photosynthesis gene
expression by the TspO outer membrane protein of Rhodobacter sphaeroides 2.4.1. J
Biol Chem 1999, 274:21234-21243.

91

34. Yeliseev AA, Kaplan S: TspO of Rhodobacter sphaeroides. A structural and functional
model for the mammalian peripheral benzodiazepine receptor. J Biol Chem 2000,
275:5657-5667.
35. Scarf AM, Auman KM, Kassiou M: Is there any correlation between binding and
functional effects at the translocator protein (TSPO) (18 kDa)? Curr Mol Med 2012,
12:387-397.
36. Li H, Yao Z, Degenhardt B, Teper G, Papadopoulos V: Cholesterol binding at the
cholesterol recognition/ interaction amino acid consensus (CRAC) of the
peripheral-type benzodiazepine receptor and inhibition of steroidogenesis by an
HIV TAT-CRAC peptide. Proc Natl Acad Sci U S A 2001, 98:1267-1272.
37. Rohmer M, Bouviernave P, Ourisson G: Distribution of Hapanoid triperpene in
Prokaryotes. Journal of General Microbiology 1984, 130:1137-1150.
38. Owen DR, Yeo AJ, Gunn RN, Song K, Wadsworth G, Lewis A, Rhodes C, Pulford DJ,
Bennacef I, Parker CA, et al.: An 18-kDa translocator protein (TSPO) polymorphism
explains differences in binding affinity of the PET radioligand PBR28. J Cereb Blood
Flow Metab 2012, 32:1-5.
39. Lacapère JJ, Papadopoulos V: Peripheral-type benzodiazepine receptor: structure and
function of a cholesterol-binding protein in steroid and bile acid biosynthesis.
Steroids 2003, 68:569-585.
40. Wendler G, Lindemann P, Lacapère JJ, Papadopoulos V: Protoporphyrin IX binding and
transport by recombinant mouse PBR. Biochem Biophys Res Commun 2003,
311:847-852.
41. Scarf AM, Luus C, Da Pozzo E, Selleri S, Guarino C, Martini C, Ittner LM, Kassiou M:
Evidence for complex binding profiles and species differences at the translocator
protein (TSPO) (18 kDa). Curr Mol Med 2012, 12:488-493.
42. Gonzalez-Polo RA, Carvalho G, Braun T, Decaudin D, Fabre C, Larochette N, Perfettini JL,
Djavaheri-Mergny M, Youlyouz-Marfak I, Codogno P, et al.: PK11195 potently
sensitizes to apoptosis induction independently from the peripheral benzodiazepin
receptor. Oncogene 2005, 24:7503-7513.
43. Decaudin D, Castedo M, Nemati F, Beurdeley-Thomas A, De Pinieux G, Caron A, Pouillart P,
Wijdenes J, Rouillard D, Kroemer G, et al.: Peripheral benzodiazepine receptor
ligands reverse apoptosis resistance of cancer cells in vitro and in vivo. Cancer Res
2002, 62:1388-1393.

92

44. Obame FN, Zini R, Souktani R, Berdeaux A, Morin D: Peripheral benzodiazepine
receptor-induced myocardial protection is mediated by inhibition of mitochondrial
membrane permeabilization. J Pharmacol Exp Ther 2007, 323:336-345.
45. Schaller S, Paradis S, Ngoh GA, Assaly R, Buisson B, Drouot C, Ostuni MA, Lacapere JJ,
Bassissi F, Bordet T, et al.: TRO40303, a new cardioprotective compound, inhibits
mitochondrial permeability transition. J Pharmacol Exp Ther 2010, 333:696-706.
46. Korkhov VM, Sachse C, Short JM, Tate CG: Three-dimensional structure of TspO by
electron cryomicroscopy of helical crystals. Structure 2010, 18:677-687.
47. Woetzel N, Lindert S, Stewart PL, Meiler J: BCL::EM-Fit: rigid body fitting of atomic
structures into density maps using geometric hashing and real space refinement. J
Struct Biol 2011, 175:264-276.
48. Higuchi R, Krummel B, Saiki RK: A general method of in vitro preparation and specific
mutagenesis of DNA fragments: study of protein and DNA interactions. Nucleic
Acids Res 1988, 16:7351-7367.
49. Studier FW: Protein production by auto-induction in high density shaking cultures.
Protein Expr Purif 2005, 41:207-234.
50. Slotboom DJ, Duurkens RH, Olieman K, Erkens GB: Static light scattering to characterize
membrane proteins in detergent solution. Methods 2008, 46:73-82.
51. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE:
UCSF Chimera--a visualization system for exploratory research and analysis. J
Comput Chem 2004, 25:1605-1612.
52. Karakaş M, Woetzel N, Staritzbichler R, Alexander N, Weiner BE, Meiler J: BCL::Fold--de
novo prediction of complex and large protein topologies by assembly of secondary
structure elements. PLoS One 2012, 7:e49240.
53. DiMaio F, Leaver-Fay A, Bradley P, Baker D, Andr?? I: Modeling symmetric
macromolecular structures in Rosetta3. PLoS One 2011, 6:e20450.
54. DiMaio F, Tyka MD, Baker ML, Chiu W, Baker D: Refinement of protein structures into
low-resolution density maps using rosetta. J Mol Biol 2009, 392:181-190.
55. Yarov-Yarovoy V, Schonbrun J, Baker D: Multipass membrane protein structure
prediction using Rosetta. Proteins 2006, 62:1010-1025.
56. Nasie I, Steiner-Mordoch S, Gold A, Schuldiner S: Topologically random insertion of
93

EmrE supports a pathway for evolution of inverted repeats in ion-coupled
transporters. J Biol Chem 2010, 285:15234-15244.
57. Rone MB, Liu J, Blonder J, Ye X, Veenstra TD, Young JC, Papadopoulos V: Targeting and
insertion of the cholesterol-binding translocator protein into the outer
mitochondrial membrane. Biochemistry 2009, 48:6909-6920.
58. Brambillasca S, Yabal M, Makarow M, Borgese N: Unassisted translocation of large
polypeptide domains across phospholipid bilayers. J Cell Biol 2006, 175:767-777.
59. Otera H, Taira Y, Horie C, Suzuki Y, Suzuki H, Setoguchi K, Kato H, Oka T, Mihara K: A
novel insertion pathway of mitochondrial outer membrane proteins with multiple
transmembrane segments. J Cell Biol 2007, 179:1355-1363.
60. Verma A, Facchina SL, Hirsch DJ, Song SY, Dillahey LF, Williams JR, Snyder SH:
Photodynamic tumor therapy: mitochondrial benzodiazepine receptors as a
therapeutic target. Mol Med 1998, 4:40-45.
61. Delavoie F, Li H, Hardwick M, Robert JC, Giatzakis C, Péranzi G, Yao ZX, Maccario J,
Lacapère JJ, Papadopoulos V: In vivo and in vitro peripheral-type benzodiazepine
receptor polymerization: functional significance in drug ligand and cholesterol
binding. Biochemistry 2003, 42:4506-4519.
62. Li H, Papadopoulos V: Peripheral-type benzodiazepine receptor function in cholesterol
transport. Identification of a putative cholesterol recognition/interaction amino acid
sequence and consensus pattern. Endocrinology 1998, 139:4991-4997.
63. Costa B, Pini S, Gabelloni P, Da Pozzo E, Abelli M, Lari L, Preve M, Lucacchini A, Cassano
GB, Martini C: The spontaneous Ala147Thr amino acid substitution within the
translocator protein influences pregnenolone production in lymphomonocytes of
healthy individuals. Endocrinology 2009, 150:5438-5445.
64. Taketani S, Kohno H, Furukawa T, Tokunaga R: Involvement of peripheral-type
benzodiazepine receptors in the intracellular-transport of heme and porphyrins. J
Biochem 1995, 117:875-880.
65. Keshavan P, Schwemberger SJ, Smith DL, Babcock GF, Zucker SD: Unconjugated
bilirubin induces apoptosis in colon cancer cells by triggering mitochondrial
depolarization. Int J Cancer 2004, 112:433-445.
66. Korichneva I, Waka J, Hammerling U: Regulation of the cardiac mitochondrial
membrane potential by retinoids. J Pharmacol Exp Ther 2003, 305:426-433.

94

67. Pfahl M, Piedrafita FJ: Retinoid targets for apoptosis induction. Oncogene 2003,
22:9058-9062.
68. Choudhuri T, Pal S, Agwarwal ML, Das T, Sa G: Curcumin induces apoptosis in human
breast cancer cells through p53-dependent Bax induction. FEBS Lett 2002,
512:334-340.
69. Oliver CL, Miranda MB, Shangary S, Land S, Wang S, Johnson DE: (-)-Gossypol acts
directly on the mitochondria to overcome Bcl-2- and Bcl-X(L)-mediated apoptosis
resistance. Mol Cancer Ther 2005, 4:23-31.
70. Butler PJ, Ubarretxena-Belandia I, Warne T, Tate CG: The Escherichia coli multidrug
transporter EmrE is a dimer in the detergent-solubilised state. J Mol Biol 2004,
340:797-808.
71. Lloris-Garcerá P, Bianchi F, Slusky JS, Seppälä S, Daley DO, von Heijne G: Antiparallel
dimers of the small multidrug resistance protein EmrE are more stable than parallel
dimers. J Biol Chem 2012, 287:26052-26059.
72. Morrison EA, DeKoster GT, Dutta S, Vafabakhsh R, Clarkson MW, Bahl A, Kern D, Ha T,
Henzler-Wildman KA: Antiparallel EmrE exports drugs by exchanging between
asymmetric structures. Nature 2012, 481:45-50.
73. Radestock S, Forrest LR: The alternating-access mechanism of MFS transporters arises
from inverted-topology repeats. J Mol Biol 2011, 407:698-715.
74. Forrest LR, Krämer R, Ziegler C: The structural basis of secondary active transport
mechanisms. Biochim Biophys Acta 2011, 1807:167-188.
75. Krishnamurthy H, Piscitelli CL, Gouaux E: Unlocking the molecular secrets of
sodium-coupled transporters. Nature 2009, 459:347-355.
76. Bowie JU: Structural biology. Membrane protein twists and turns. Science 2013,
339:398-399.
77. Qin L, Mills DA, Buhrow L, Hiser C, Ferguson-Miller S: A conserved steroid binding site
in cytochrome c oxidase. Biochemistry 2008, 47:9931-9933.
78. Hiser C, Buhrow L, Liu J, Kuhn LA, Ferguson-Miller SM: A conserved amphipathic
ligand binding region influences K-path dependent activity of cytochrome c oxidase.
Biochemistry 2013.
79. Ginter C, Kiburu I, Boudker O: Chemical catalysis by the translocator protein (18 kDa).
95

Biochemistry 2013.
80. Youle RJ, Strasser A: The BCL-2 protein family: opposing activities that mediate cell
death. Nat Rev Mol Cell Biol 2008, 9:47-59.

96

CHAPTER 3

Identification of a key cholesterol binding enhancement motif in Translocator Protein 18
kDa

This chapter will be submitted in part as “Fei Li, Jian Liu, Lance Valls, Shelagh Ferguson-Miller.
Identification of a key cholesterol binding enhancement motif in Translocator Protein 18 kDa”

Lance Valls (undergraduate student) performed all binding assays and Dr. Jian Liu did the protein
co-evolution analysis on RsTSPO.

97

INTRODUCTION
Translocator Protein 18 kDa (TSPO) is a widely distributed integral membrane protein
initially discovered as a peripheral tissue binding site for anxiolytic benzodiazepine drugs [1] and
therefore known as the peripheral benzodiazepine receptor (PBR). It was renamed in 2006 to
more accurately reflect its far-reaching phylogeny and increasing evidence of involvement in a
number of complex cellular processes, including cholesterol transport, inflammation, tumor
progression, Alzheimer disease and regulation of the mitochondrial permeability transition
associated with apoptosis [2-8]. Deletion of TSPO was found to be developmentally lethal in the
mouse [9].
In mammals, TSPO is expressed at highest levels in the outer membrane of mitochondria in
steroid hormone producing tissues, where it functions to facilitate cholesterol transport and is
proposed to be a rate-limiting step in steroidogenesis [10,11] (but see [12]). High expression is
also observed in the most metastatic cancers [13]

and in regions of inflammation [14,15],

possibly relating to a role in corticosteroid production in response to stress [16] and leading to
the use of TSPO ligands for sensitive positron emission tomography (PET) imaging of damage in
the brain [17]. A conserved cholesterol recognition/interaction amino acid consensus sequence
(CRAC: L/V-x(1-5)-Y-x(1-5)-R/K) near the C-terminus of the predicted transmembrane helix V
was identified as the cholesterol binding site in mammalian TSPO [18]. Despite sharing the
majority of the CRAC sequence (L-x(1)-F-x(3)-R), the homolog from Rhodobacter sphaeroides
(RsTSPO) shows more than 1000-fold lower affinity for cholesterol than mammalian TSPO. A
clue to this disparity was provided by recent publications reporting a spontaneous human
polymorphism A147T in the region immediately preceding the CRAC site. This change
correlated with reduced pregnenolone production [19] and bipolar disorder [20] as well as lower
98

imaging ligand binding [21], drawing attention to this region.
Comparing the sequences of bacterial and mammalian proteins, we noted that the helix V
region containing CRAC is generally well conserved, except that three residues preceding the
human polymorphism A147T are highly variable in bacteria but conserved in the mammalian
proteins and much more hydrophobic, suggesting that this region is a possible contributor to the
differing affinities. We therefore designed a mutant to test the functional implications of this
region preceding the human polymorphism site. Rhodobacter is a good model system for
investigating the cholesterol binding function of TSPO due to its dramatically different affinity
and availability in purified form [19]. Using a combination of biochemical and evolutionary
analysis, we identified a motif that correlates with nM affinity for cholesterol. The cholesterol
binding enhancement motif is highly conserved within mammalian TSPOs and is also observed
in other mammalian membrane proteins, indicating broad significance for cholesterol/protein
interactions and providing new evidence relevant to the recent debate on the role of TSPO in
cholesterol transport [12].
MATERIAL AND METHODS
Cloning, expression and purification of protein.
The coding sequence of RsTSPO was subcloned into the pUC12 vector for mutagenesis.
Mutant RsTSPO was made by QuikChange Mutagenesis

©

(Stratagene). The construct was

confirmed by sequencing and ligated into the pRK415 expression vector with KpnI and HindIII
restriction sites. The expression vector containing the desired TSPO coding sequence was
introduced into the E. coli expression strain BL21 λDE3. Purification procedures are as
previously described for the RsTSPO-WT protein [22] except for the modified procedure for the
RsTSPO-LAF, which is described in the text.
99

Tryptophan fluorescence quenching binding assay.
Binding of ligands to RsTSPO proteins is measured as previously described [22] on an
®

Agilent Eclipse fluorescence spectrometer. Kd was calculated by fitting the binding curve with
equations previously developed for single site binding mode [22] when possible. In the cases
where complicated binding behavior occurred or the precipitation of ligand at higher
concentrations prevented accurate fitting, an apparent Kd was estimated from the half-maximal
quenching. Each reported Kd is an average from three independent experiments.
TLC analysis of the purified RsTSPO_LAF sample.
One μl samples were spotted on a Si-60 thin layer chromatography (TLC) plate (EMD).
Samples contained either: 1) 0.1 % Cholesteryl hemisuccinate (CHS) standard in 50 mM Tris,
150 mM NaCl buffer pH 8; 2) rescue buffer (50 mM Tris, 150 mM NaCl, 10% glycerol, 0.2%
DM and 0.01% CHS, pH 8); or 3) purified RsTSPO-LAF, 21 µM in size exclusion
chromatography (SEC) buffer (50 mM Tris, 150 mM NaCl, 10% glycerol, 0.2% DM, pH 8). The
plate was run in chloroform:methanol:ammonia (63:35:5) and air dried before spraying with a
cholesterol specific staining solution (20% sulfuric acid) and baked in 110 ºC for ~5 mins.
Cholesterol and cholesterol esters are stained wine-red.
Covariance analysis with PsiCOV.
A total of 2332 sequences homologous to RsTSPO were obtained from the non-redundant
(nr) protein sequence database using the DELTA-Blast (Domain Enhanced Lookup Time
Accelerated BLAST) web server. Multiple sequence alignments of these sequences were done
with the Cobalt Constraint-based Multiple Protein Alignment Tool. PsiCOV was then used to
calculate the contact pair covariance of this alignment. The initial value of Glasso regulation

100

parameter ρ is set to 0.001.
Phylogenetic analysis of the enhancement motif.
Sequences of TSPO proteins were obtained from the NCBI database using the accession
number provided in [3]. Sequences of full length TSPO proteins were aligned with Clustal
Omega [23] and the picture was created in the Aline [24] program. Phylogenetic analysis was
done on the aligned full length TSPO sequences with the default NJ method in MEGA6 [25]
software.
Survey of the enhancement motif in human membrane protein database.
A database containing 5183 human membrane proteins was downloaded from the
UniProtKB/Swiss-Prot database with human, transmembrane, and helical as keyword filters. The
search pattern [L/V]-x1-5-[Y]-x1-5-[R] was used to search for the CRAC motif with the Fuzzpro
program within the EMBOSS suite. The pattern [L]-[A]-[F]-x2-5-[L/V]-x1-5-[Y]-x1-5-[R] was
used to search for the LAF-CRAC with enhancement motif.
RESULTS AND DISCUSSION
Rationale and design of mutations in RsTSPO.
TSPO from Rhodobacter shows more than 34 % amino acid identity compared to the
human protein [26] and rat TSPO is able to rescue the function of a RsTSPO knock-out mutant in
vivo [27], suggesting a conserved functional relationship. More importantly, helix V of RsTSPO
has a high degree of similarity to the mammalian sequence (Figure 3.1), but with some
distinctive differences. The CRAC site is highly conserved, but nevertheless, human TSPO
has >1000-fold higher affinity for cholesterol than RsTSPO [22]. The observation of a human
polymorphism (A147T in human and A139T in Rhodobacter) immediately preceding the CRAC
motif that appeared to affect ligand binding affinity, along with the fact that the three residues
101

prior to A139T were highly variable in bacteria, led us to propose a role for this region in
cholesterol binding.

Figure 3.1. Location of the triple mutation of RsTSPO in relation to the CRAC sequence.
Upper panel shows the sequence alignment of the C-terminal helix V region of TSPO from R.
sphaeroides (Rs), human, mouse, rat and bovine. The CRAC motif is boxed with magenta and
the three critical residues are labeled with dark magenta triangles. The location of the human
polymorphism A147 (A139 in RsTSPO) is labeled with a dark green star. Residues mutated in
this study are boxed green. Lower panel shows the location of aforementioned residues on a
model of RsTSPO colored coded: magenta for the CRAC region with dark magenta representing
the 3 critical residues, dark green for A139 and green for LAF.

102

To test our hypothesis, we created a mutant of RsTSPO, RsTSPO-A136L/T137A/A138F
(RsTSPO-LAF), to mimic human TSPO by replacing the Rhodobacter sequence Ala-Thr-Ala
(136-138) with the more hydrophobic human counterpart Leu-Ala-Phe (144-146). We predicted
this region to be deeply buried in the membrane, with the possibility of better interacting with the
alkyl tail of cholesterol and increasing cholesterol binding affinity. Since the C-terminus
including helix V is not suggested to play a major role in the binding of the specific diagnostic
ligand PK11195 [28] or protoporphyrin IX (PpIX), we also tested the binding of these ligands to
the mutant form.
RsTSPO-LAF exhibits a tendency to aggregate when concentrated, but CHS can rescue it
from precipitation.
The triple mutant RsTSPO-LAF was well expressed and initially purified following the same
procedure as RsTSPO-WT [22] but it was observed to precipitate during the concentration step
after Ni-NTA purification. Addition of buffer supplemented with 0.01% CHS rescued
RsTSPO-LAF during the concentration step and His-tag removal. The final purification by SEC
was performed in regular buffer without CHS and yielded considerably lower amounts of protein,
likely due to the loss from some precipitation before SEC.
Interestingly, we found that CHS was retained and enriched by RsTSPO-LAF during this
purification method, as shown in the TLC plate in Figure 3.2A. The RsTSPO-LAF to which CHS
was added during concentration and then further purified without any CHS by SEC retained 2-3
fold more CHS than the buffer itself used for rescue in the previous steps. RsTSPO-LAF was
also purified by replacing the concentrating step with a spin column buffer exchange step to
remove the imidazole after Ni-NTA elution and foregoing the concentration and SEC purification
step. Previous experience with RsTSPO-WT had shown that protein after Ni-NTA purification

103

has similar high purity compared to that treated with trypsin and purified by an additional SEC
step and has no observable difference in ligand binding. Since the protein concentration used in
the binding assay is very low, RsTSPO-LAF protein from the spin column could be used directly
in the binding assays and compared with samples purified after treatment with CHS. Given the
lower stability of the LAF mutant, we did a control experiment which confirmed that the
fluorescence reading was not significantly decreased during the time course of the binding assay
indicating little precipitation in the course of the experiments (Figure 3.2B). A SDS-PAGE of the
purified RsTSPO-LAF protein is shown in Figure 3.2C.

104

Figure 3. 2. Purification of RsTSPO-LAF and the rescue with cholesterol hemisuccinate.
(A) TLC of RsTSPO-LAF. 1: CHS standard (0.1%) in 50 mM Tris-150 mM NaCl pH 8 buffer; 2: rescue buffer containing CHS (50
mM Tris, 150 mM NaCl, 10% glycerol, 0.2% DM, pH 8, supplemented with 0.01% CHS); 3: RsTSPO-LAF after CHS rescue and
further purified in SEC buffer without CHS. (B) Fluorescence spectra of RsTSPO-LAF. T1 and T2 are two time points one hour apart,
representing the start and end of a typical fluorescence quenching experiment. The integration areas, which are used in calculating the
fraction of quenching of both curves, are noted in the figure. This experiment shows only ~5% of fluorescence decay during the
binding experiment for the RsTSPO-LAF protein, confirming that the slight decay, likely due to some tendency to aggregate, will not
influence the binding results. (C) SDS-PAGE of purified RsTSPO proteins. 1: size marker; 2: Ni-NTA purified RsTSPO-LAF; 3:
RsTSPO-WT
105

RsTSPO-LAF binds cholesterol with nM affinity.
Freshly purified RsTSPO-LAF protein not exposed to CHS was used in the binding assay
with cholesterol. Our results (Figure 3.3) show a high affinity cholesterol binding site in the LAF
mutant, with a Kd of ~ 20 nM, similar to the affinity of mouse and human TSPO previously
reported [18,29]. It is difficult to calculate an accurate Kd for the low affinity cholesterol binding
in WT since saturating concentrations could not be achieved due to the limited solubility of
cholesterol. Nevertheless, Figure 3.3 clearly shows that introducing the LAF motif causes the
affinity for cholesterol to greatly increase in the initial phase of the binding titration, from an
apparent Kd in the low μM range for WT to the low nM range for LAF. In addition, the samples
originally purified in the presence of CHS and retaining that ligand showed dramatically lower
apparent affinity for cholesterol similar to WT, suggesting CHS is binding at the same site as
cholesterol. This finding is consistent with the observation that CHS co-purifies with the mutant
form (Figure 3.2 and Figure 3.3). These results strongly support a role of the LAF motif in
determining the nM affinity of cholesterol for mammalian TSPO, since introduction of the three
amino acid changes are sufficient to account for the 1000-fold enhanced affinity.

106

Figure 3.3 RsTSPO-LAF binds to cholesterol at nM affinity.
Binding curve of RsTSPO-LAF with cholesterol compared to WT and RsTSPO-LAF purified in
the presence of CHS. Binding data for cholesterol with WT were previously reported [22] and
shown here for comparison with the mutant form. All binding assays were done with the same
protein concentration (2.5 μM) and the same buffer conditions as previously described [22] (50
mM Tris, 150 mM NaCl, 10% glycerol, 0.2% DM, pH 8). Error bars for each data point are not
provided because individual data points are not repeated; rather, the complete binding curve for
each ligand was repeated 3 times.

107

Two binding orientations of cholesterol to the CRAC site have been suggested [30,31], with
the hydroxyl group facing in or out of membrane. Considering our previous finding that
ursodeoxycholate, which has a similar ring structure as cholesterol but with a carboxyl attached
to the alkyl chain, does not bind to RsTSPO-WT [22], our results support a model with
cholesterol binding with the hydroxyl group facing out, the ring structure interacting with the
highly conserved CRAC site, and the alkyl chain interacting with the LAF motif towards the
center of the membrane. Consistent with this model, pregnenolone, testosterone and progesterone,
all with shorter and more hydrophilic tails, were not observed to be transported by mouse TSPO
expressed in E. coli [18].
Binding behavior of RsTSPO-LAF with PK11195 and PpIX.
We also tested the binding of PK11195 and PpIX with RsTSPO-LAF. Our results (Figure 3.4)
show that the RsTSPO-LAF mutant has some influence on the binding of PK11195 and PpIX but
not nearly as dramatic as the effects on cholesterol binding. Notably, the distinctive emission
spectrum previously seen in PK11195 quenching curves with the RsTSPO-WT protein [22] was
also seen in the mutant, suggesting that the binding site of PK11195 was not strongly influenced
by the mutation. Importantly, the modest effects of the three amino acid substitution in
RsTSPO-LAF on binding of ligands other than cholesterol indicate that it is not causing a
generalized change in structure, but rather a specific localized one.

108

Figure 3.4. Binding of RsTSPO-LAF with PK11195 and PpIX.
Upper panel shows the binding curves for RsTSPO-LAF and RsTSPO WT with PpIX and
PK11195. Lower panel reports the calculated Kd. Buffer conditions are as in Fig 3.3.

109

Interactions between helix IV and V are critical for protein stability.
Covariance analysis is a method to identify interacting residues within proteins based on the
concept that residues which interact in the structure have a higher probability of undergoing
parallel mutations during evolution to maintain protein function and stability. This method
provides independent information to identify critical interactions between residues as well as
powerful constraints for computational modeling [32]. We analyzed the sequences of TSPO with
the covariance method to investigate critical residue interactions as an independent test of our
previous models based on computational analysis and cryo-electron microscopy [22] as well as
to provide information regarding the source of observed instability of RsTSPO-LAF.
Several software packages [33-35] were used to analyze TSPO sequences and gave very similar
results. The top scoring interacting residue pairs predicted by PsiCOV on the basis of 2332
sequences are plotted on the predicted topology map of RsTSPO (Figure 3.5). Our analysis
identified a number of high scoring interactions between helix IV and helix V, while there were
very few between other helices, suggesting that these two helices may be critical for the stability
of TSPO. In fact, in an early effort to obtain NMR structures of TSPO fragments, Jamin and
colleagues showed that helix IV itself could not form a stable helix [31] but is stabilized by a
strong interaction with helix V and maintains this association when purified in various detergents
[36]. The recent NMR structure of a monomer form of mouse TSPO also shows helix IV–V
association [37].

110

Figure 3.5. Topology diagram of RsTSPO showing contact pairs predicted by covariance
analysis.
Helices are marked with Roman numerals. Predicted residue contact pairs are rendered in the
same color. Their covariances are calculated with PsiCOV [33]. Top 15 helical residue contact
pairs are shown (covariance > 0.77).This diagram is produced with the Protter web application
[38]. The sequence and helix information is from Toppred [11].

111

The two highest scoring interactions predicted by this covariance analysis are exactly within
the ATA/LAF region (Rs residues 136,137 shown in red in Figure 3.5), providing an explanation
for

the

observed

lower

stability

of

the

TSPO-LAF

mutant.

Independently,

the

hydrogen/deuterium (H/D) exchange analysis performed by an NMR group also shows that these
residues on helix IV and V are the most stable core in the whole protein [37]. In the wild type
RsTSPO, A136 and T137 on helix V are predicted to have strong interactions with V111 and
W108 on helix IV. In our RsTSPO-LAF mutant, we have created a better binding site for
cholesterol at the expense of weakening some of these critical interactions, accounting for the
observed lesser stability. For instance, in RsTSPO-WT, alanine136 on helix V is predicted to pair
with bulky residues, either tryptophan108 or valine111 on helix IV, while in the human/mouse
protein the larger leucine of LAF, the equivalent of 136, is paired with a smaller serine at the
equivalent of 108 or a smaller alanine at the equivalent of V111. By making double mutants that
take into account these pairings, a more stable form with high cholesterol affinity might be
achieved.
Evolution of the cholesterol binding enhancement motif.
The region immediately preceding the previously identified CRAC motif is suggested to be
important for the function of human TSPO on the basis of the characteristics of a human
polymorphism A147T [19,21]. In addition to A147, we further identify a motif that includes
residues L144-A145-F146 in human TSPO (136,137,138 in RsTSPO) as playing an important
role in determining the high affinity of mammalian TSPO for cholesterol, a property not found in
the bacterial proteins which lack this LAF sequence.
To investigate the evolutionary significance of the LAF motif, we analyzed the sequences of
TSPO proteins from a set of different species previously described [3,39]. Our analysis shows

112

that the cholesterol binding enhancement motif is very well conserved within mammals as
Leu-Ala-Phe (LAF, Figure 3.6 A&C), including human and mouse where extensive studies have
shown that TSPO plays a critical role in cholesterol transport into mitochondria and
steroidogenesis [40,41]. While bacteria and archaea have a similar CRAC sequence with the
central Y replaced by F, their LAF motifs are very diverse. Hopanoids, which have a very similar
ring structure to cholesterol, have been identified in bacteria [42,43] as serving a similar function
to cholesterol in the higher animals. However, they can have quite different tails as shown by two
representative hopanoids in Figure 3.6B, suggesting the tail could range from very long and
hydrophilic to very short and hydrophobic. The more variable motif in bacteria and other lower
orders may reflect the need to accommodate the distinctive tails of the hopanoids in different
bacteria, or to accommodate alternative ligands. Note that helix IV is also much more conserved
within mammals as compared to bacteria. Taken together with the covariance evidence that
helices IV and V interact and co-evolve, our results are consistent with the increasing functional
importance of cholesterol and its metabolic products in higher animals, supplying a selection
pressure for the evolution of high affinity cholesterol binding involving the enhancement motif.
Indeed, all mammals that metabolize cholesterol in mitochondria as the sole source of steroid
hormones seem to preserve the high affinity LAF sequence in this motif.

113

Figure 3.6. Evolution of TSPO and correlation with LAF motif.
(A) Phylogenetic tree of TSPO from selected species showing the presence/absence of the LAF
motif; (B) structure of cholesterol and of two representative hopanoids; (C) sequence alignment
of helix IV and helix V of TSPO from selected species.

114

The cholesterol binding enhancement motif exists in other membrane proteins and may
indicate the presence of high affinity cholesterol binding sites.
Cholesterol is a major component of the membranes of higher animals, accounting for
around 30% of the lipid bilayer [44], and it plays a critical role as the source of all steroid
hormones as well as regulating the chemical and physical properties of membranes and thus
membrane embedded proteins. Interest in this latter role of cholesterol has been stimulated by the
increasing number of high resolution membrane protein structures in which cholesterol binding
sites are observed or predicted [45-48]. The identification of the CRAC site in TSPO [18,29] led
to its recognition in other membrane proteins and suggested its general importance, but other
motifs have also been found [48,49]. In this study, we identify a motif adjacent to the CRAC site
in TSPO which enhances cholesterol binding and is highly conserved in mammals. Some
interesting questions are raised by our findings: 1) is this enhancement motif also present in other
membrane proteins that are regulated by cholesterol; 2) is this motif found in all mammalian
CRAC sites or only a subset; and 3) is the presence of this enhancement motif in other
membrane proteins associated with any common characteristic?
To address some of these questions, we surveyed 5183 currently identified human membrane
proteins from the UniProtKB/Swiss-Prot database. Both CRAC and LAF-x(2-5)-CRAC, which
allows 2-5 residues between the LAF and CRAC sites, were searched by the Fuzzpro program. A
total of 28627 CRAC sites were identified in 4684 human membrane proteins, about 6 sites per
protein on average, while only 66 LAF-x(2-5)-CRAC sites were identified in 66 membrane
proteins, including members from all major membrane protein families (Table 3.1). Although the
finding of 6 CRAC sites/protein in so many proteins is theoretically possible, the result is more
consistent with this broadly defined motif having relatively low predictive ability, as concluded

115

by Sanders et al. [50]. In contrast, the much smaller subset found with the extended motif
suggests more predictive value.

116

Table 3.1 Classification of human membrane proteins with the cholesterol binding
enhancement motif
Membrane protein family
Receptors
Transporters
Ion channels
Enzymes
Others
Total

Within transmembrane helix
16 (33%)
8 (16%)
4 (8.2%)
2 (4.1%)
19 (39%)
49 (100%)

117

In extracellular domains
2 (12 %)
0 (0%)
2 (12%)
8 (47%)
5 (29%)
17 (100%)

Total
18 (27%)
8 (12%)
6 (9.1%)
10 (15%)
24 (36%)
66 (100%)

Out of the 66 identified sites, 49 are in the later half of transmembrane helices, similar to the
location in TSPO and topologically consistent with the hydroxyl of cholesterol being close to the
membrane surface. The significance of sites identified by the LAF-CRAC motif remains to be
determined as more cholesterol-regulated membrane proteins are characterized biochemically
and structurally, but it seems likely that they represent higher affinity binding modes with
functional significance beyond that conferred by cholesterol as solvent. The location of TSPO in
mitochondria, which are low in cholesterol [44], and the function of TSPO as part of a
cholesterol transport system, are consistent with a requirement for the enhancement motif we
have identified in mammalian TSPO to provide significantly increased affinity for cholesterol.
Indeed a LAF-CRAC sequence is identified in the predicted transmembrane helix 2 of the
StAR-related lipid transfer protein 3 (STARD3). This protein has a steroidogenic acute
regulatory protein domain (StAR) in the extracellular domain and has been proposed to
translocate cholesterol between endosomes and mitochondria [51,52] similar to the proposed
TSPO-StAR transport system [53]. Also identified in our survey to have the enhanced cholesterol
binding motif is the sodium-dependent bile acid transporter (SLC10A2). The substrate, a
derivative of cholesterol, is seen in crystals of a bacterial homolog [54] to occupy a binding
pocket where the LAF-CRAC sequence is also found in the human homolog [55]. Since
cholesterol has been shown to be a regulator of this bile acid transporter [56] the proximity of
these sites could be significant.
Our survey suggests that cholesterol binding sites with the enhancement motif are fairly
widespread and could be critical for high affinity binding involved in transport or regulation,
while the CRAC sequence alone and other motifs may represent sites important for other
membrane-mediated cholesterol effects such as protein/protein association and domain formation

118

[50].
SUMMARY
Biochemical, mutational and evolutionary covariance studies of the purified bacterial
homolog of mitochondrial TSPO identify a cholesterol-binding enhancement motif that is highly
conserved in mammalian but not bacterial proteins. This motif in the C-terminal transmembrane
helix V of TSPO, adjacent to the previously identified CRAC sequence and a human single
polymorphism associated with bipolar disorder, significantly affects affinity for cholesterol. The
altered properties of a purified mutant form are consistent with genetic covariance analysis and
previous NMR studies that show this region to be of critical importance to the structure and
function of TSPO. The cholesterol binding enhancement motif associated with CRAC is
conserved in mammalian TSPO and also identified in many human membrane proteins. Our
survey suggests that the enhancement motif may be useful for identifying high affinity
cholesterol binding regions with functional and regulatory significance.

119

REFERENCES

120

REFERENCES

1. Braestrup C, Squires RF: Specific benzodiazepine receptors in rat brain characterized by
high-affinity (3H)diazepam binding. Proc Natl Acad Sci U S A 1977, 74:3805-3809.
2. Papadopoulos V, Baraldi M, Guilarte TR, Knudsen TB, Lacapère JJ, Lindemann P, Norenberg
MD, Nutt D, Weizman A, Zhang MR, et al.: Translocator protein (18kDa): new
nomenclature for the peripheral-type benzodiazepine receptor based on its structure
and molecular function. Trends Pharmacol Sci 2006, 27:402-409.
3. Fan J, Lindemann P, Feuilloley MG, Papadopoulos V: Structural and functional evolution of
the translocator protein (18 kDa). Curr Mol Med 2012, 12:369-386.
4. Krueger KE: Molecular and functional properties of mitochondrial benzodiazepine
receptors. Biochim Biophys Acta 1995, 1241:453-470.
5. Crompton M: The mitochondrial permeability transition pore and its role in cell death.
Biochem J 1999, 341 ( Pt 2):233-249.
6. Verma A, Snyder SH: Peripheral type benzodiazepine receptors. Annu Rev Pharmacol
Toxicol 1989, 29:307-322.
7. Barron AM, Garcia-Segura LM, Caruso D, Jayaraman A, Lee JW, Melcangi RC, Pike CJ:
Ligand for translocator protein reverses pathology in a mouse model of Alzheimer's
disease. J Neurosci 2013, 33:8891-8897.
8. Lin R, Angelin A, Settimo FD, Martini C, Taliani S, Zhu S, Wallace DC: Genetic analysis of
dTSPO, an outer mitochondrial membrane protein, reveals its functions in
apoptosis, longevity, and Aβ42-induced neurodegeneration. Aging Cell. 2014,
13:507-518.
9. Papadopoulos V, Amri H, Boujrad N, Cascio C, Culty M, Garnier M, Hardwick M, Li H, Vidic
B, Brown AS, et al.: Peripheral benzodiazepine receptor in cholesterol transport and
steroidogenesis. Steroids 1997, 62:21-28.
10. Anholt RR, Pedersen PL, De Souza EB, Snyder SH: The peripheral-type benzodiazepine
receptor. Localization to the mitochondrial outer membrane. J Biol Chem 1986,
261:576-583.
11. Cserzö M, Wallin E, Simon I, von Heijne G, Elofsson A: Prediction of transmembrane
alpha-helices in prokaryotic membrane proteins: the dense alignment surface
method. Protein Eng 1997, 10:673-676.

121

12. Morohaku K, Pelton SH, Daugherty DJ, Butler WR, Deng W, Selvaraj V: Translocator
protein/peripheral benzodiazepine receptor is not required for steroid hormone
biosynthesis. Endocrinology 2014, 155:89-97.
13. Hardwick M, Fertikh D, Culty M, Li H, Vidic B, Papadopoulos V: Peripheral-type
benzodiazepine receptor (PBR) in human breast cancer: correlation of breast cancer
cell aggressive phenotype with PBR expression, nuclear localization, and
PBR-mediated cell proliferation and nuclear transport of cholesterol. Cancer Res
1999, 59:831-842.
14. Chen MK, Guilarte TR: Translocator protein 18 kDa (TSPO): molecular sensor of brain
injury and repair. Pharmacol Ther 2008, 118:1-17.
15. Bird JL, Izquierdo-Garcia D, Davies JR, Rudd JH, Probst KC, Figg N, Clark JC, Weissberg
PL, Davenport AP, Warburton EA: Evaluation of translocator protein quantification
as a tool for characterising macrophage burden in human carotid atherosclerosis.
Atherosclerosis 2010, 210:388-391.
16. Lucki NC, Li D, Sewer MB: Sphingosine-1-phosphate rapidly increases cortisol
biosynthesis and the expression of genes involved in cholesterol uptake and
transport in H295R adrenocortical cells. Mol Cell Endocrinol 2012, 348:165-175.
17. Chauveau F, Boutin H, Van Camp N, Dollé F, Tavitian B: Nuclear imaging of
neuroinflammation: a comprehensive review of [11C]PK11195 challengers. Eur J
Nucl Med Mol Imaging 2008, 35:2304-2319.
18. Li H, Papadopoulos V: Peripheral-type benzodiazepine receptor function in cholesterol
transport. Identification of a putative cholesterol recognition/interaction amino acid
sequence and consensus pattern. Endocrinology 1998, 139:4991-4997.
19. Costa B, Pini S, Gabelloni P, Da Pozzo E, Abelli M, Lari L, Preve M, Lucacchini A, Cassano
GB, Martini C: The spontaneous Ala147Thr amino acid substitution within the
translocator protein influences pregnenolone production in lymphomonocytes of
healthy individuals. Endocrinology 2009, 150:5438-5445.
20. Colasanti A, Owen DR, Grozeva D, Rabiner EA, Matthews PM, Craddock N, Young AH:
Bipolar Disorder is associated with the rs6971 polymorphism in the gene encoding
18 kDa Translocator Protein (TSPO). Psychoneuroendocrinology 2013, 38:2826-2829.
21. Owen DR, Yeo AJ, Gunn RN, Song K, Wadsworth G, Lewis A, Rhodes C, Pulford DJ,
Bennacef I, Parker CA, et al.: An 18-kDa translocator protein (TSPO) polymorphism
explains differences in binding affinity of the PET radioligand PBR28. J Cereb Blood
Flow Metab 2012, 32:1-5.

122

22. Li F, Xia Y, Meiler J, Ferguson-Miller S: Characterization and modeling of the oligomeric
state and ligand binding behavior of purified translocator protein 18 kDa from
Rhodobacter sphaeroides. Biochemistry 2013, 52:5884-5899.
23. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H,
Remmert M, Söding J, et al.: Fast, scalable generation of high-quality protein
multiple sequence alignments using Clustal Omega. Mol Syst Biol 2011, 7:539.
24. Bond CS, Schüttelkopf AW: ALINE: a WYSIWYG protein-sequence alignment editor
for publication-quality alignments. Acta Crystallogr D Biol Crystallogr 2009,
65:510-512.
25. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S: MEGA6: Molecular Evolutionary
Genetics Analysis version 6.0. Mol Biol Evol 2013, 30:2725-2729.
26. Yeliseev AA, Kaplan S: TspO of Rhodobacter sphaeroides. A structural and functional
model for the mammalian peripheral benzodiazepine receptor. J Biol Chem 2000,
275:5657-5667.
27. Yeliseev AA, Krueger KE, Kaplan S: A mammalian mitochondrial drug receptor
functions as a bacterial "oxygen" sensor. Proc Natl Acad Sci U S A 1997,
94:5101-5106.
28. Farges R, Joseph-Liauzun E, Shire D, Caput D, Le Fur G, Ferrara P: Site-directed
mutagenesis of the peripheral benzodiazepine receptor: identification of amino acids
implicated in the binding site of Ro5-4864. Mol Pharmacol 1994, 46:1160-1167.
29. Li H, Yao Z, Degenhardt B, Teper G, Papadopoulos V: Cholesterol binding at the
cholesterol recognition/ interaction amino acid consensus (CRAC) of the
peripheral-type benzodiazepine receptor and inhibition of steroidogenesis by an
HIV TAT-CRAC peptide. Proc Natl Acad Sci U S A 2001, 98:1267-1272.
30. Rupprecht R, Papadopoulos V, Rammes G, Baghai TC, Fan J, Akula N, Groyer G, Adams D,
Schumacher M: Translocator protein (18 kDa) (TSPO) as a therapeutic target for
neurological and psychiatric disorders. Nat Rev Drug Discov 2010, 9:971-988.
31. Jamin N, Neumann JM, Ostuni MA, Vu TK, Yao ZX, Murail S, Robert JC, Giatzakis C,
Papadopoulos V, Lacapère JJ: Characterization of the cholesterol recognition amino
acid consensus sequence of the peripheral-type benzodiazepine receptor. Mol
Endocrinol 2005, 19:588-594.
32. de Juan D, Pazos F, Valencia A: Emerging methods in protein co-evolution. Nat Rev Genet
2013, 14:249-261.
123

33. Jones DT, Buchan DW, Cozzetto D, Pontil M: PSICOV: precise structural contact
prediction using sparse inverse covariance estimation on large multiple sequence
alignments. Bioinformatics 2012, 28:184-190.
34. Yang J, Jang R, Zhang Y, Shen HB: High-accuracy prediction of transmembrane
inter-helix contacts and application to GPCR 3D structure modeling. Bioinformatics
2013, 29:2579-2587.
35. Kamisetty H, Ovchinnikov S, Baker D: Assessing the utility of coevolution-based
residue-residue contact predictions in a sequence- and structure-rich era. Proc Natl
Acad Sci U S A 2013, 110:15674-15679.
36. Galvagnion C, Montaville P, Coïc YM, Jamin N: Production and initial structural
characterization of the TM4TM5 helix-loop-helix domain of the translocator protein.
J Pept Sci 2013, 19:102-109.
37. Jaremko L, Jaremko M, Giller K, Becker S, Zweckstetter M: Structure of the
mitochondrial translocator protein in complex with a diagnostic ligand. Science 2014,
343:1363-1366.
38. Omasits U, Ahrens CH, Müller S, Wollscheid B: Protter: interactive protein feature
visualization and integration with experimental proteomic data. Bioinformatics 2014,
30:884-886.
39. Fan J, Papadopoulos V: Evolutionary origin of the mitochondrial cholesterol transport
machinery reveals a universal mechanism of steroid hormone biosynthesis in
animals. PLoS One 2013, 8:e76701.
40. Papadopoulos V, Amri H, Boujrad N, Cascio C, Culty M, Garnier M, Hardwick M, Li H,
Vidic B, Brown AS, et al.: Peripheral benzodiazepine receptor in cholesterol
transport and steroidogenesis. Steroids 1997, 62.
41. Papadopoulos V, Lecanu L, Brown RC, Han Z, Yao ZX: Peripheral-type benzodiazepine
receptor in neurosteroid biosynthesis, neuropathology and neurological disorders.
Neuroscience 2006, 138:749-756.
42. Rohmer M, Bouviernave P, Ourisson G: Distribution of hapanoid triperpene in
prokaryotes. J Gen Micro1984, 130:1137-1150.
43. Poger D, Mark AE: The relative effect of sterols and hopanoids on lipid bilayers: when
comparable is not identical. J Phys Chem B 2013, 117:16129-16140.
44. van Meer G, Voelker DR, Feigenson GW: Membrane lipids: where they are and how they
124

behave. Nat Rev Mol Cell Biol 2008, 9:112-124.
45. Singh AK, McMillan J, Bukiya AN, Burton B, Parrill AL, Dopico AM: Multiple cholesterol
recognition/interaction amino acid consensus (CRAC) motifs in cytosolic C tail of
Slo1 subunit determine cholesterol sensitivity of Ca2+- and voltage-gated K+ (BK)
channels. J Biol Chem 2012, 287:20509-20521.
46. Barrett P, Song Y, Van Horn W, Hustedt E, Schafer J, Hadziselimovic A, Beel A, Sanders C:
The amyloid precursor protein has a flexible transmembrane domain and binds
cholesterol. Science 2012, 336:1168-1171.
47. Picazo-Juárez G, Romero-Suárez S, Nieto-Posadas A, Llorente I, Jara-Oseguera A, Briggs M,
McIntosh TJ, Simon SA, Ladrón-de-Guevara E, Islas LD, et al.: Identification of a
binding motif in the S5 helix that confers cholesterol sensitivity to the TRPV1 ion
channel. J Biol Chem 2011, 286:24966-24976.
48. Hanson MA, Cherezov V, Griffith MT, Roth CB, Jaakola VP, Chien EY, Velasquez J, Kuhn P,
Stevens RC: A specific cholesterol binding site is established by the 2.8 A structure of
the human beta2-adrenergic receptor. Structure 2008, 16:897-905.
49. Baier CJ, Fantini J, Barrantes FJ: Disclosure of cholesterol recognition motifs in
transmembrane domains of the human nicotinic acetylcholine receptor. Sci Rep 2011,
1:69.
50. Song Y, Kenworthy AK, Sanders CR: Cholesterol as a co-solvent and a ligand for
membrane proteins. Protein Sci 2014, 23:1-22.
51. Alpy F, Tomasetto C: MLN64 and MENTHO, two mediators of endosomal cholesterol
transport. Biochem Soc Trans 2006, 34:343-345.
52. Charman M, Kennedy BE, Osborne N, Karten B: MLN64 mediates egress of cholesterol
from endosomes to mitochondria in the absence of functional Niemann-Pick Type
C1 protein. J Lipid Res 2010, 51:1023-1034.
53. Rone MB, Fan J, Papadopoulos V: Cholesterol transport in steroid biosynthesis: role of
protein-protein interactions and implications in disease states. Biochim Biophys Acta
2009, 1791:646-658.
54. Hu NJ, Iwata S, Cameron AD, Drew D: Crystal structure of a bacterial homologue of the
bile acid sodium symporter ASBT. Nature 2011, 478:408-411.
55. Hussainzada N, Banerjee A, Swaan PW: Transmembrane domain VII of the human apical
sodium-dependent bile acid transporter ASBT (SLC10A2) lines the substrate
125

translocation pathway. Mol Pharmacol 2006, 70:1565-1574.
56. Alrefai WA, Sarwar Z, Tyagi S, Saksena S, Dudeja PK, Gill RK: Cholesterol modulates
human intestinal sodium-dependent bile acid transporter. Am J Physiol Gastrointest
Liver Physiol 2005, 288:G978-985.

126

CHAPTER 4
Crystallization strategies of small α helical integral membrane proteins: a case study of
Translocator Protein 18 kDa from Rhodobacter sphaeroides (RsTSPO)

127

INTRODUCTION
Membrane proteins play important roles in all living organisms, promoting a strong interest
in understanding their structures and functions in detail. Recent developments in biochemistry
and biophysics have greatly improved our ability to investigate membrane proteins. In particular,
the technical developments in the field of membrane protein crystallography, including
development of alternative crystallization methods for membrane proteins [1,2], high throughput
robotic machines [3], and the optimization of synchrotron beamlines for membrane protein
crystals [4] have allowed us to obtain high resolution structures of important membrane proteins
at an unprecedented rate (http://blanco.biomol.uci.edu/mpstruc/#Latest). As a result, our
understanding of various processes involving membrane proteins has tremendously improved
over the past decades, providing critical information for understanding the fundamental
mechanisms of membrane proteins as well as for rational design of drugs targeting membrane
proteins. Despite these developments, obtaining high resolution crystal structures of membrane
proteins remains a formidable task, and often requires significant experimentation and
optimization. Every successfully solved membrane protein crystal structure, especially those
with unique properties and/or those that are underrepresented in the current PDB, provides
valuable information to help us with mastering the art of membrane protein crystallization.
TSPO from Rhodobacter sphaeroides (RsTSPO) represents a type of small membrane
protein that is mainly composed of α-helices without big cytoplasmic domains for mediating
crystal contacts. The structure determination of RsTSPO was a challenging project in that
significant optimization was needed in almost every step to achieve the final high resolution
structures of this unique membrane protein. Investigation and optimization of the crystallization
strategies for RsTSPO was systematic and extensive, spanning from the traditional vapor

128

diffusion method to the newly developed lipidic cubic phase (LCP) method. From this
perspective, the optimization process to obtain crystals of RsTSPO is a valuable case study that
provides important lessons for optimizing crystallization conditions for similar types of small
and hydrophobic integral membrane proteins.
Over the past 20 years, alternative crystallization methods for membrane proteins have been
developed and optimized, and have shown great potential. In general, these methods aim to
provide a more natural environment for membrane proteins during crystallization. Such an
environment has a bilayer structure with more lipid content compared to the protein-detergent
complexes used in the regular vapor diffusion method. One example of these methods is the
bicelle method [2], proposed and popularized by Dr. James Bowie’s group, that reconstitutes the
detergent solubilized membrane protein into small bilayer-like discs before setting up a regular
vapor diffusion crystallization experiment. Other alternatives involve addition of detergent
and/or lipids during steps of purification and crystallization and have been successful in various
cases [5,6].
The LCP (also know as in meso) crystallization method first pioneered by Landau and
Rosenbusch [7] and further developed by Dr. Martin Caffrey’s group [1] aims to better mimic the
continuous bilayer environment during crystallization with the use of bilayer forming lipids to
host membrane proteins. The LCP method has shown great success with multiple classes of
membrane proteins including beta-barrels [8], alpha-helical transmembrane proteins [9], and
challenging membrane protein complexes [10]. The LCP method takes advantage of the phase
behavior of monoacylglycerol (MAG) lipids when mixed with aqueous solutions (i.e., protein
solutions). A bicontinuous three-dimensional highly curved lipid bilayer consisting of
interpenetrating but non-communicating aqueous channels forms spontaneously when the MAG

129

is mixed with a protein-water solution at a certain ratio. The specific structure formed is named
as the cubic phase or Pn3m phase [7]. This continuous bilayer serves as the stabilizing matrix for
membrane proteins incorporated in it, as well as providing the channels and portals allowing
diffusion and crystallogenesis [1].
Despite extensive screening with the vapor diffusion method, it failed to produce good
diffraction quality crystals of RsTSPO under all conditions tested. The choice of the LCP
crystallization method, which promotes type I crystal packing, was the key breakthough in
obtaining RsTSPO crystals. Additionally, additives such as ligands and lipids also play important
roles in stabilizing the purified membrane protein and are frequently screened during the
optimization [11,12]. In the case of RsTSPO, the addition of high affinity ligands significantly
improved the crystal size and quality, which eventually allowed structure determination. In
addition, a mutant form designed to mimic a human polymorphism, A147T (A139T in RsTSPO),
that is associated with anxiety related disease, was found to give the most stable and well
diffracting crystals.
MATERIAL AND METHODS
Materials
Chemicals were purchased from Sigma-Aldrich. All detergents were purchased from
Anatrace (Maumee, OH). Lipids were purchased from Avanti Polar Lipids. The RsTSPO
®

expression plasmid was a gift from Dr. Samuel Kaplan from the University of Texas. Pfu Turbo

polymerase was purchased from Agilent Technologies (Santa Clara, CA). Restriction enzymes
and T4 ligase were purchased from New England Biolabs (Ipswich, MA).

130

Methods
Protein expression and purification
The pRK415 plasmid containing the coding sequence of RsTSPO with a C-terminal 10
Histag (RsTSPO10ht) was transformed into the E. coli BL21 λDE3 strain. The transformed
bacteria were cultured at 30 °C in 100 mL LB medium containing appropriate antibiotics
overnight. Fifteen mL of the overnight culture were then transferred into 1 L of auto-induction
media (ZYM-5025) [13] containing glycerol, glucose and lactose and the appropriate antibiotics.
Cultures were grown at 30 °C. Cells were harvested when the OD600 reached above 4.0 (usually
after overnight growth). Harvested cells were resuspended in lysis buffer (50 mM KH2PO4 pH
6.5, 1 mM EDTA) supplemented with Roche

®

protease inhibitor cocktail and were lysed by two

passes through a French press homogenizer at 20,000 p.s.i. Broken cells were centrifuged at 200
x g for 5 mins, 10,000 x g for 30 mins and 150,000 x g for 90 mins at 4 °C to isolate the
membranes. Isolated membranes were resuspended with buffer A (50 mM Tris-HCl, pH 8, 150
mM NaCl, 10 % glycerol) and quick-frozen with liquid nitrogen and stored at -70 °C before
purification.
Membranes were thawed on ice, resuspended to 10 mg/ml total protein in buffer A,
solubilized by adding detergent, 1 mM PMSF and Roche

®

protease inhibitor cocktail (no EDTA)

and stirred at 4 °C for 1 hr. Solubilized membranes were centrifuged at 150,000 x g for 30 mins
®

to remove the unsolubilized fraction. The supernatant was loaded on a 10 mL Ni-NTA (Qiagen )
gravity column pre-equilibrated with 3 column volumes (CV) of buffer A followed by 1 CV of
buffer A with 10 mM imidazole. After extensive washing with buffer A supplemented with
detergent at the appropriate concentration and 50 mM imidazole, fractions containing the desired

131

protein were eluted using buffer A supplemented with detergent at appropriate concentration and
300 mM imidazole over two CV. Absorption at 280 nm was used to monitor the washing and
elution process. Fractions containing the desired protein were pooled and concentrated with a 50
®

kDa molecular weight cut off (MWCO) Amicon Filter (Millipore ). Protein purified in this way
was either loaded on SEC column directly to test whether the selected detergent will result in a
homogenous peak of RsTSPO10ht or washed with no imidazole buffer and treated with trypsin at
4 °C overnight before further purification by SEC. The purity of the protein was determined on
SDS-PAGE stained with Coomassie blue.
Determination of detergent content of purified RsTSPO by thin layer chromatography
(TLC) and a colorimetric method
The detergent species and levels during purification was monitored by both TLC [14] and a
colorimetric [15] method. Briefly, standard solutions containing known concentrations of
maltoside or glucoside detergents were spotted on TLC plates together with the purified protein
sample and run under the same conditions. The TLC plates were stained with iodine and alpha
naphthol. Detergent present in the sample was identified by comparing the Rf with standards,
while the density of the stained spot was quantified by using ImageJ [16] and used to calculate
the concentration of detergent in the sample based on a standard curve. In the colorimetric
method, standards as well as protein samples were mixed with reaction solution containing
phenol and sulfuric acid and the absorbances at 490 nm were measured with a PerkinElmer

®

spectrometer. Concentrations were similarly calculated from a standard curve based on A490.
Detergent exchange
Several schemes were used to exchange the detergent as well as to reduce the total detergent
132

concentration in the purified RsTSPO samples. One way to reduce the detergent concentration
was simply to wash the purified protein with no-detergent buffer during concentration. In this
method, purified protein solution was mixed with equivalent buffer with no detergent at a 1:1
ratio and concentrated with the Amicon concentrator containing an appropriate MWCO
membrane. The process was repeated several times to achieve the desired detergent
concentration. A cholate washing step was also tried because it has been shown that cholate can
help break the filament structure formed by DDM and reduce the detergent level [17]. In this
method, cholate was added into the protein solution to a final concentration of 6 % followed by
incubation on ice for 1 hr. An equal volume of buffer with no detergent was added and the
solution was concentrated with the appropriate Amicon concentrator. Alternatively, the Pierce
detergent-removal spin column was used to exchange and reduce the detergent concentration
with an in-house protocol. Briefly, the spin column was equilibrated 4 times with buffer
containing the desired detergent at 5x critical micelle concentration (CMC) by spinning at 1000 x
g. Three hundred µL of the buffer containing 2x CMC of the desired detergent was then added
into a clean collecting tube and 100 µL of protein sample in purification buffer containing DM
was loaded onto the column, followed by centrifugation for 2 mins at 1000 x g. The protein
sample in the desired final buffer was then removed from the collecting tube and concentrated
with a 0.5 mL 50 kDa MWCO Amicon concentrator. The final sample was quick-frozen in liquid
nitrogen and analyzed by the TLC methods previous described.
Screening for solubilizing detergent
Membranes of E. coli containing 10 mg/ml total protein (as determined by the bicinchoninic
acid assay (BCA) assay using bovine serum albumin (BSA) as the standard) were solubilized by
shaking with buffer containing 20x CMC of detergent for 2 h at 4 ºC and centrifuged at 200,000
133

x g for 30 min. The effectiveness of solubilization was judged by comparing the supernatant with
membranes solubilized by 1 % DDM by using a Western blot. Solubilized supernatant was also
kept at 4 ºC to test for long time stability.
Screening and optimization of the purification detergent with SEC
Detergents were also screened directly with SEC for their ability to produce a homogeneous
sample for crystallization. In this method, membranes containing RsTSPO were solubilized with
1% DM and purified as previously described on a Ni-NTA column with buffer containing 0.2 %
DM and 10 % glycerol. RsTSPO10ht was then directly loaded onto a Superose12

®

gel filtration

column equilibrated with buffer containing 50 mM NaCl, 150 mM NaCl, 10 % glycerol and the
desired detergent.
Optimization of SEC conditions
For any detergent that was able to give a homogeneous peak of RsTSPO10ht, the SEC
conditions were fine-tuned by testing if the Histag and glycerol influence the purification.
RsTSPO10ht samples were purified as described previously and run in buffer without glycerol.
Concentrated RsTSPO10ht samples were washed once and diluted with no-imidazole buffer to
10 ml and incubated with trypsin overnight to remove the Histag. Samples were then
concentrated with a 30 kDa MWCO Amicon filter unit down to 250 uL and loaded on the
Superose12

®

gel filtration column. Samples of good quality were concentrated and used in

crystallization screening directly.
The optimized purification procedure were to purify RsTSPO with buffer containing 0.2 %
DM and 10 % glycerol up to the setup of the SEC. Ni-NTA purified RsTSPO10ht was
concentrated to less than 1 mL with a 50 kDa MWCO Amicon filter unit and washed with 10 mL
134

buffer with no imidazole. This sample was concentrated again to less than 250 uL before being
diluted to 10 mL and adding trypsin for digestion overnight. The trypsin treated RsTSPO sample
was then concentrated with a 30 kDa MWCO Amicon filter unit down to 250 uL and run on a
Superdex 200 gel filtration column with buffer containing 0.2 % DM but no glycerol.
Crystallization screening of RsTSPO with the vapor diffusion method
The traditional vapor diffusion method [18] was first used to grow crystals with samples
purified by Ni-NTA and SEC. Both RsTSPO10ht and RsTSPO purified with various detergent
conditions were screened. High throughput screening plates were set up with the Crystal
Gryphon

®

(Art Robbins Instruments) robot with 200 nL protein:200 nL screen solution ratio

with a wide range of commercial crystallization screens. Different protein:screen solution ratios
at 1:2, 1:3 and 2:1 were also tested. Wells were examined under regular light as well as polarized
light with a microscope periodically. Scaled-up sitting drop or hanging drop trays were set up for
conditions that showed promising results from the initial screens. Initial leads were scaled up and
optimized with 1 µL protein:1 µL screen solution in both sitting drop and hanging drop formats.
All crystals were harvested with cryoloops and directly frozen in liquid nitrogen before being
tested for diffraction quality.
Crystallization screening of RsTSPO with the LCP method
Initial high throughput screening for crystallization conditions of RsTSPO with the LCP
method was done at the NIH-supported membrane protein center of Dr. Stroud at UCSF.
RsTSPO was purified in 0.2 % DM buffer without glycerol and concentrated to 30 mg/ml with a
30 kDa MWCO Amicon concentrator. Purified RsTSPO (30 mg/mL) was incorporated into
monoolein (MO) (v/v 2:3) by mechanical mixing. One hundred nL of the protein-lipid mixture
135

was dispensed onto the plastic plates by the LCP Mosquito

®

robot and 1 µL of precipitant

solution was laid on top of the protein-lipid bolus. Plates were sealed by covering the well with
another plastic cover and incubated at 20 ºC. Plates were monitored under bright light as well as
with the polarizer and UV light to confirm the growth of protein crystals.
Optimization of crystal leads with the LCP method
Further optimization of the lead conditions with both the LCP and sponge phase methods
was done at MSU. In the sponge phase method, 0.5 µL protein-MO mixture was covered with 2
µL precipitant solution and set up in a hanging drop setup with 125 µL reservoir solution. Initial
crystallization conditions were more finely screened by varying the pH, PEG concentration,
additive, and host lipid. Glass plates and coverslips were used in all optimization plates to enable
crystal harvesting.
Crystals were harvested with Mitigen

®

microloops or micromesh by using published

protocols [19,20]. Briefly, the glass coverslip was cut open with a capillary cutter and screen
solutions were laid on top before opening the well. Screen solution was also injected into the
well through holes opened at the corner of the cut coverslip. Broken glass was carefully removed
piece by piece with fine tweezers while screen solution was added to keep the LCP bolus from
drying. Crystals were harvested from the LCP bolus with a microloop or micromesh matching
the size of the crystal and immediately frozen with liquid nitrogen. The entire process was
monitored under the microscope with a polarizer to keep track of small crystals that were only
visible by this approach.
Preparation of heavy metal derivative of the RsTSPO A139T crystal
A tantalum bromide cluster (Ta6Br12)Br2 was soaked into A139T crystals at 1 mM final
136

concentration, while another potential heavy metal derivative was obtained by co-crystallizing
the A139T protein with mercury thiocyanide, Hg(SCN), under the same crystallization
conditions, but the concentrated heavy metal stock solution was pre-spotted and dried on the
crystallization plate before setting up crystals. Sixteen different heavy metal containing solutions
were used at two different concentrations each. The A139T crystal containing the Hg derivative
appeared after around two weeks.
Data collection and processing
Diffraction data on crystals from the vapor diffusion conditions were collected at 100 K at
LS-CAT (beamlines 21ID-G and 21ID-F) at the Advanced Photon Source (APS), Argonne
National Laboratory. Data on crystals from LCP conditions were collected at GM/CA-CAT
(beamline 23ID-B and 23ID-D) at APS. Crystals buried in the LCP bolus on the micromesh were
located by X-ray diffraction with an attenuated beam with proper rastering strategies before data
collection. Data were collected with a 5 µm micro-focusing beam at full beam intensity. Wedges
containing 10-30 degrees of data were individually processed with the XDS software (Version
March 20, 2013) [21] and merged and scaled with Aimless (CCP4 6.3.0) [22,23] to make the
complete dataset for structure determination. Data from the heavy metal derivative crystals were
collected using either the peak wavelength of the tantalum L3 edge (1.255 keV) or of the
mercury L3 edge (1.007 Å) at the 23ID-B beamline (Advanced Photon Source). Complete (360˚)
datasets for each derivative were acquired from single crystals by using a 10 μm minibeam at
20x attenuation with 1˚ oscillation and 1 s exposure per frame; the data were collected in 30˚
wedges with direct and inverse beams, combined with a helical rotation strategy to minimize the
radiation damage and maximize the anomalous signal. The Ta dataset was integrated and scaled
to 3.7 Å resolution and the Hg dataset was integrated and scaled to 3.2 Å by using HKL2000
137

(HKL Research) and XDS [21].
Structure determination and refinement
The

A139T

mutant

structure

was

solved

by

using

multiple

isomorphous

replacement-anomalous scattering phasing with the Ta and Hg datasets combined with the high
resolution native dataset method in SHARP (version 2.8.2) [24]. Complete models for the three
RsTSPO monomers in the A139T C2 asymmetric unit, except for loop 1 from one of the 3
monomers, were easily autotraced and built in SHARP after the initial rounds of density
modification. The structure of wild type RsTSPO was solved by molecular replacement with a
partially refined model of the A139T mutant. Initial refinement for all structures was performed
iteratively with Phenix.refine [25] followed by manual examination and model rebuilding with
the program Coot 7.2 [26] against 2Fo-Fc and Fo-Fc maps. Final refinement were done with
Phenix-1.9-1692 [25], followed by iterative rebuilding of the model with Coot 7.2 [26] against
2Fo-Fc, Fo-Fc, and feature-enhanced maps (Phenix-1.9-1692).
RESULTS
Initial crystallization screening with RsTSPO purified in DM indicated inhibition of
crystallization by detergent
The initial purification scheme developed by Dr. Ferguson-Miller in collaboration with Dr.
John Lee (UCSF) gave good yield of RsTSPO in high purity (Figure 4.1). A time course of
trypsin digestion showed that trypsin immediately cut around 50 % of RsTSPO10ht while the
SEC showed a major homogenous peak of RsTSPO at around 44 kDa. One percent DDM was
used for membrane solubilization in this method while all purification, including both Ni-NTA
and SEC, was done in buffer supplemented with 0.2 % DM and 10 % glycerol. The main peak
P3 (Figure 4.1A) was screened with several commercial high throughput crystallization kits at 6
138

mg/mL, 10 mg/mL and 15 mg/mL protein concentrations. Quasi-crystals started to appear in
multiple conditions after around two weeks while a large percentage of the conditions showed
phase separation. These behaviors usually indicate that the detergent concentration present in the
crystallization drop was too high and therefore inhibiting crystallization.

139

Figure 4.1 RsTSPO was purified by metal affinity and size exclusion chromatography
(SEC) and analyzed by SDS PAGE.
(A): SEC profile of RsTSPO. (B): Coomassie stained SDS-PAGE gel of SEC fractions of
purified RsTSPO showed that Peak 3 (SECP3) was fully cleaved by trypsin and highly purified.

140

Although DDM and DM are very good at stabilizing membrane proteins, they also tend to
inhibit crystal formation due to masking of potential crystal contacts by the detergents. Given
that RsTSPO is a very small protein, the use of DDM and DM in the purification may be
detrimental to crystallization. Additionally, the micelle size of DDM and DM is much bigger
than the monomer of RsTSPO. This will result in a very high DDM/DM concentration in the
purified RsTSPO sample because the detergent micelles will be concentrated together with
RsTSPO during the concentration step. This will further inhibit crystallization, as suggested by
the observed phase separation behavior.
Two methods were established and used to determine the detergent concentration in the
purified RsTSPO samples. Both the TLC method and colorimetric method showed good linearity
at the desired concentration range and had good agreement (Figure 4.2). Several purified and
concentrated RsTSPO and RsTSPO10ht samples used in crystallization were measured and one
representative TLC and one standard curve is shown in Figure 4.2A. All samples show > 5 %
DM/DDM concentration, which is more than 200x CMC for both detergents. Figure 4.2A also
shows that the DM concentration was significantly increased after concentrating the protein on a
50 kDa MWCO Amicon concentrator. These results confirmed that the detergent was indeed
significantly concentrated and was at a much too high a concentration for optimium crystal
growth.

141

Figure 4.2 Determination of detergent concentration of purified RsTSPO samples.
(A) TLC plate stained with iodine. (B) Standard curve of detergent concentration of samples in
(A) where the concentrations of detergent in the purified protein sample before (green diamond)
and after (green circle) the concentration step are indicated on the standard curve. (C)
Colorimetric assay of DM (see Methods) showing a strong peak at 492 nm that was used for
quantifying the sugar content (i.e., maltose detergent content). (D) Standard curve of detergent
concentration of samples in (C).
142

As a result, two strategies were explored to optimize the detergent in the purification
procedures to promote crystallization: 1) screening and optimization of purification conditions
with various detergents and buffer conditions to achieve a suitable detergent concentration in the
crystallization sample; and 2) screening for alternative detergents with smaller micelle size and
higher CMC for solubilization. Various methods and techniques were tested and optimized
iteratively with one final goal of maximizing the protein to detergent ratio for the purified
RsTSPO sample used in crystallization.
DM level in purified RsTSPO was successfully reduced and exchanged for smaller
detergents
Excessive amounts of detergent are known to inhibit the crystallization of membrane
proteins. This problem is usually dealt with by concentrating purified membrane proteins with a
concentrator that has large enough pores to let the detergent micelles go through and thus keep
the detergent concentration in the concentrated samples roughly the same as in the
unconcentrated samples at around 2x CMC. However, in the case of RsTSPO, the detergent
micelles of both DDM and DM are in fact larger than the protein, which makes it unavoidable to
concentrate the detergent while concentrating the protein. DDM is also known to bind to
membrane proteins strongly at very low concentrations, which makes it a favorable choice for
solubilization of larger membrane proteins but not a good choice for RsTSPO. To avoid this
complication, the solubilization detergent was changed to 1 % DM, which was shown in the
solubilization test to be equally good (see below).
Since DM was able to be used to successfully purify RsTSPO to homogeneity and produce
good yields for crystallization experiments, efforts were focused on reducing the DM
concentration in the concentrated RsTSPO sample. A simple dilution and wash method was tried

143

first based on the fact that the DM concentration in the concentrated RsTSPO sample was several
hundred fold higher than the CMC. Therefore, no-detergent buffer was used to dilute the sample
to reduce the DM concentration below 2x CMC and the sample was re-concentrated. This
procedure was repeated several times based on the rationale that the DM will be concentrated
again during protein concentration, but repeating the process should reduce the DM
concentration to just enough to keep the protein in solution. RsTSPO remained soluble after
dilution with the no-detergent buffer but precipitated during re-concentrating.
A cholate wash procedure was then tested since cholate has been reported to break the
filament structure formed by DDM [17]. We suspected that DM has a similar filament forming
property and therefore the cholate washing would also help DM to go through the membrane
during the concentrating step. As shown in Figure 4.3A, the cholate washing procedure
dramatically decreased the DM concentration in the final sample and the sample was then used
in crystallization screening. However, both RsTSPO and cholate formed heavy precipitation in
crystallization conditions, suggesting that the detergent concentration after this procedure was
too low.
A modified detergent removal procedure with the Pierce detergent-removal spin column was
also used to reduce the DM concentration or to exchange DM for other detergents. Figure 4.3
B&C clearly shows that this procedure was very effective in removing the DM in the sample and
replacing it with detergents with smaller micelles, such as octyl-glucoside (OG) or
nonyl-glucoside (NG). This procedure was also used to reduce the DM level by spinning the
samples into buffer containing 0.17 % DM in the collection tube. However, TLC results showed
that more than 5 % DM remained in the sample after the removal procedure, suggesting that OG
or NG used to equilibrate the column also played a role in facilitating the detergent removal,

144

possibly by breaking the detergent micelle formed by DM or forming mixed-micelles with
smaller sizes. Samples treated with these methods were used in the crystallization screening with
commercial screen kits.

145

Figure 4.3 Detergent exchange results determined by TLC.
(A) Cholate wash result, plate stained with iodine. Lane 1: RsTSPO sample washed with 6 %
cholate; lane 2: untreated RsTSPO sample. (B) Detergent exchange results with the Pierce
detergent-removal spin column, stained with iodine. Lane 1: mixed standard of 0.5 % DM and
0.5 % OG; lane 2: 1 % DM and 1 % OG; lane 3: 2 % DM and 2 % OG; lane 4: 0.4 % NG; lane 5:
0.8 % NG; lane 6: 1 % NG; lane 7: RsTSPO10ht directly from Ni-NTA; lane 8: RsTSPO10ht
exchanged into OG; lane 9: RsTSPO10ht sample exchanged into NG. (C) Same plate as in (B)
stained with alpha naphthol that detects the sugar moiety of detergents.

146

Samples treated by these various detergents and procedures were all monitored by TLC and
tested by small scale crystallization screening with a wide range of commerial screens by using
the vapor diffusion method. Experiments were carefully monitored and the results were
documented (Table 4.1) to guide the optimization of purification and crystallization. However,
no crystal were obtained from these treated samples.

147

Table 4.1 High throughput small scale crystallization screening results of RsTSPO

Screen Kit

Protein
concentratio
n
(mg/ ml)

Detergent used
in purification

Washing
with
cholate

Detergent level
quantified by
TLC

Screen set-up
Protein :Scre
en
nl: nl

6
Cryo I/II

30% clear, 30% phase separation
0.2 % DM

NO

N/A

200:200

10
Crystal screen
1/2

10

0.2 % DM

NO

N/A

200:200

Axygen I

5
7

0.3 % DM

YES

0.1- 0.5 % DM

200:200

Axygen IV

10

0.2 % DM

NO

N/A

200:200

0.3 % DM

YES

0.1- 0.5 % DM

200:200

5

0.15 % DM

YES

7

0.3 % DM

5
Axygen V

7

NO
YES

MemGold

MemStart/sys

Results

5

0.3 % DM

7

0.3 % DM

0.1- 0.5 % DM
~ 1% Cholate
0.5 % - 1 %
DM

YES

200:200

Most wells are clear

0.1- 0.5 % DM

YES

NO

200:200

30 % precipitation, usually oily
precipitation,
5 % phase separation
30% precipitation and phase
separation
Precipitation condensed in the
middle
> 50% phase separation, oily
spread precipitation
Most wells show precipitation,
some condensed in the middle of
well
Heavy precipitation possibly due
to cholate

N/A

200:100

0.5 % - 1 %
DM
0.1- 0.5 % DM

200:200

148

<1% phase amorphous
separation, 60% precipitation,
several wells show heavy
precipitation at first and changed
to clear sheet
30% clear, 30% amorphous
precipitation
30% clear, 30% precipitation

RsTSPO was successfully extracted by several detergents
Several representative detergents belonging to different chemical classes that have been used
successfully to facilitate membrane protein crystallization were tested (Figure 4.4A). Zwittergen
3-12, C12E8 and LDAO are all popular detergents for membrane protein crystallization with
higher CMCs and smaller micelles compared to DM and DDM. Experience shows that it is
easier to grow membrane protein crystals purified with these detergents if the protein is stable.
However, membrane proteins tend to precipitate in buffer with these detergents because of their
charge and because their micelles are not big enough to effectively protect the hydrophobic
regions of membrane proteins. LAPAO was tested because it was successfully used in
purification and crystallization of another related mitochondrial alpha-helical transmembrane
protein, the adenine nucleotide translocator (ANT) [27]. Cymal-type detergents were designed as
derivatives of DDM, acting as mild non-ionic detergents to keep the native properties of
membrane proteins but with higher CMCs and smaller micelles. Cymal-5 was chosen based on
the consideration of the size of RsTSPO as well as its success with several other membrane
proteins [28,29]. The protein yield using these various detergents to solubilize membranes
containing RsTSPO10ht was compared with 1 % DDM (Figure 4.4 B). All selected detergents
extracted RsTSPO from membranes equally well based on a western blot of supernatants reacted
with an antibody against the Histag. The sample treated with LDAO precipitated right after
solubilization. All other samples were kept at 4 °C to test for long time stability. The sample in
DDM buffer was stable for more than 2 months while others were able to keep the protein stable
for 3-5 weeks, roughly the time for a crystallization experiment.

149

Figure 4.4 Detergent solubilization screening results.
(A) Structures of selected detergents used in the solubilization screen. (B) Western blot of
supernatants reacted with antibodies against the Histag showing the results of the solubilization
screen.

150

Identification of the optimal detergent by size exclusion chromatography
A SEC method was also used to exchange and test detergents for their suitability to support
RsTSPO crystallization. The advantage of the SEC method is that the homogeneity of the sample,
which is critical for crystallization, was also monitored. Briefly, RsTSPO10ht was solubilized in
1 % DM and purified on Ni-NTA with buffer supplemented with 0.2 % DM and 10 % glycerol,
in which it has very good stability. Two times the CMC of the selected detergent was used in the
SEC step to replace the 0.2 % DM. The suitability of the detergent was judged by the SEC
profile. A minimal aggregation peak at the void volumn (~7.5 mL) and a tight homogeneous
dimer peak of RsTSPO10ht (~12.5-14 mL depending on the specific detergent) is a good
indication that the selected detergent was good.
Representative SEC profiles are shown in Figure 4.5. As expected, C10E5 and Zwittergent
3-12 were not able to stabilize RsTSPO, perhaps due to the small size of their micelles. On the
other hand, maltose detergents (DM, NM, OM, cymal-5) were much more effective in stabilizing
RsTSPO in solution with minimum void volumn and two homogeneous peaks as shown in the
SEC profiles. However, the two peaks representing the oligomer and dimer forms of RsTSPO
were not well separated in most cases. High throughput crystallization trials were set up with
both peaks purified in maltose detergents but no crystal hits were found with any of the
commercial screens.

151

Figure 4.5 SEC profiles of RsTSPO10ht purified in different detergents.
(A) DM, (B) NM, (C) Zwittergent 3-12, (D) C10E5, (E) OM, (F) cymal-5.

152

Identification of the optimum purification procedure by SEC
Despite extensive screening of detergents, DM proved to be the best detergent for
solubilization and purification of RsTSPO10ht in terms of good yield from extraction as well as
long time stability. However, samples purified in DM never succeeded in producing diffraction
quality crystals. Given the extremely small size of RsTSPO10ht, the thought was perhaps the
micelles of DM were too big and too stable such that no area of the protein was exposed to the
solution and therefore no crystal contacts could be formed. The SEC profile with DM also
showed that the separation of the oligomer and dimer peaks was not optimum (Figure 4.5A), a
potential factor that could inhibit crystallization. On the other hand, C10E5 type detergents
usually form smaller micelles but have shown to be de-stabilizing to RsTSPO10ht (Figure 4.5B).
To create the best balance of stability versus crystallizability, C10E5 was added to DM to
modulate and promote crystallization. A similar procedure was used successfully in the
optimization of crystallization conditions for the E. coli glycerol-3-phosphate transporter by
mixing DDM and C12E9 [30]. When RsTSPO10ht was purified with buffer containing 0.2 %
DM/0.06 % C10E5 (Figure 4.6 C), the aggregation peak was eliminated while the two peaks of
dimer and oligomer were better separated. The dimer peak was subsequently concentrated and
used in crystallization screening. Indeed, this sample successfully grew small crystals that
diffracted to around 16 Å (Figure 4.8A).

153

Figure 4.6. SEC profiles of RsTSPO10ht purified in (A) DM alone, (B) C10E5 alone, and
(C) DM/C10E5 mixture.
154

Ten percent glycerol was used in the original purification protocol as a general stabilizing
reagent but its effect was never tested. Therefore, a modified purification procedure was tested to
see if removing the glycerol is beneficial. Ten percent glycerol was used in all buffers in the
same way as in the original protocol up to the SEC step to maintain the maximum stability of the
protein. A buffer with the same detergent but no glycerol was used in the final SEC step. No
precipitation was observed during and after the SEC for proteins purified with the no glycerol
buffer, while the SEC profiles were improved slightly as suggested by a tighter major peak
(Figure 4.7 A, C, E, F).
The histidine tag has been shown to mediate crystal contacts in some cases [31] and can be
particularly useful if the target protein does not have big cytosolic regions for making crystal
contacts, for instance in the case of RsTSPO. However, it can also cause heterogeneity and is
usually removed during the final purification step. Removal of the 10-Histag was therefore tested
and the SEC profile showed that this step was extremely effective in suppressing the
oligomerization of RsTSPO (Figure 4.7 C, D).
Figure 4.7 shows that the combination of removing the glycerol and the 10-Histag produced
the best purification result in terms of having a tight, single major peak at the dimer size. All DM,
NM and DM/C10E5 mixtures were able to purify RsTSPO to good quality. RsTSPO proteins
purified in all three detergents were tested in crystallization screening. Protein purified with
DM/C10E5/no glycerol continued to produce crystals of similar quality as previously described
while protein purified with DM without glycerol was able to produce small crystals that
diffracted to around 20 Å for the first time. However, these conditions failed in further scale-up
and optimization. On the other hand, protein purified in NM buffer precipitated quickly after
setting up, suggesting that 10 carbons is the minimum alkyl tail length for a suitable detergent.

155

Figure 4.7. SEC profiles for testing effects of glycerol and histag.
RsTSPO10ht (B&D) runs at around 12.5 mL (around 66kDa) while RsTSPO (A,C,E&F) runs at
around 14-15 mL (around 44 kDa) on the Superose 12 column. The histag promotes
oligomerization and results in less well separated peaks between the oligomer and the dimer
while glycerol slightly increases the width of the peak

156

Results of crystallization and optimization with the vapor diffusion method
In addition to the extensive optimization of the purification procedure and detergents, more
than 1000 crystallization conditions with various protein ratios and temperatures were screened
to identify the lead for optimization. Although some conditions were able to produce small
crystals (Figure 4.8 A, B) occasionally, the results were not consistent and they all failed in
scale-up experiments. In addition, all crystals grown in the vapor diffusion method never
diffracted to better than 13 Å. Among all the potential lead conditions, one was worth
mentioning. The condition of 0.05 M NaCl, 0.02 M glycine pH 10, 33% PEG 1000 produced the
best crystal in the vapor diffusion method with a sample purified in 0.2% DM/ 0.06% C10E5,
and this crystal diffracted to 16 Å. Addition of 2-propanol slightly increased the resolution to 13
Å. Interestingly, the lead identified with the LCP method also contained ~32% PEG 1000 and
addition of 2-propanol significantly increased the size and the quality of the crystals (see
discussion below).

157

Figure 4.8 Representative RsTSPO crystals grown with the vapor diffusion method and
their diffraction patterns.
(A) RsTSPO purified in 0.2% DM/ 0.06% C10E5 buffer with no glycerol, and grown in 0.05 M
NaCl, 0.02 M glycine pH 10, 33% PEG1000, 20 °C; (B) RsTSPO purified in 0.2% DM buffer
with no glycerol, and grown in 26-33% PEG400, CaCl2, pH 6.5-8.5,4 °C; (C) diffraction pattern
of crystal in (A); (D) diffraction pattern of crystal in (B).

158

Initial screening with the LCP method
The failure of the extensive screening and optimization of crystallization conditions with
vapor diffusion confirmed our concern that RsTSPO is not suitable for crystallization by the
traditional vapor diffusion method and prompted us to consider the alternative methods. Given
that TSPO lacks large extra-membrane loops to mediate crystal contacts in type II crystals and
the fact that it successfully produced 2D crystals for cryo-EM structure determination [32], we
considered trying to promote the type I crystallization by using the LCP method.
RsTSPO was purified with the same optimized procedure, which uses 0.2 % DM
throughout the purification and no-glycerol buffer in the final SEC step and concentrated to 30
mg/mL with a 30 kDa MWCO filter. Initial high throughput screening was done as described in
the Methods. Small crystals started to appear after about 10 days of incubation at 20 ºC. More
crystals were observed after 2-3 weeks in conditions containing zinc acetate (Zn(OAC)2), MES
buffer pH 6.0 and ~20% PEG 400/ PEG 550 MME/ PEG 600 or PEG 1000 (Figure 4.9). Bigger
sheet-shaped crystals were observed after 4 weeks of incubation in conditions also containing
Zn(OAC)2. These lead conditions were manually repeated at MSU with both LCP setups and
sponge phase setups. The condition containing PEG 1000 was successfully reproduced and the
micro crystals were harvested with a micromesh and tested at GM/CA CAT with the micro beam.
These initial crystals diffracted to around 4 Å with the unattenuated X-ray beam.

159

Figure 4.9. Representative pictures of RsTSPO protein crystals grown in the LCP setup
after 3 weeks.
RsTSPO crystals grown in 100 nL+1 µL LCP setup. (A&B): 20 % PEG400, 0.1 M MES pH 6.0,
0.1 M Zn(OAC)2; (C&D) 20 % PEG550 MME, 0.1 M MES pH 6.0, 0.1 M Zn(OAC)2; (E&F):
20 % PEG600, 0.1 M MES pH 6.0, 0.1 M Zn(OAC)2; (G&H): 20 % PEG1000, 0.1 M MES pH
6.0, 0.2 M Zn(OAC)2; (I&J): crystals grown from protein-monoolein mixture incorporating 6 %
cholesterol in 20 % PEG1000, 0.1 M MES pH 6.0, 0.2 M Zn(OAC)2. Panels A/C/E/G/I are in
brightfield and panels B/D/F/H/J are under UV.
160

Optimization of LCP crystallization conditions
The original crystal lead identified in the high throughput screening was extremely small and
embedded in the LCP bolus. Data collection was impossible on such crystals other than to
comfirm that RsTSPO indeed grows 3D crystals in the LCP condition. Extensive optimization
was carried out to improve the size of the crystal to allow data collection as well as to improve
the quality of the crystal (Figure 4.10).

161

Figure 4.10. Schematic representation of optimization strategies for crystallization of
RsTSPO by the LCP method.

162

Addition of around 8 % cholesterol in the monoolein increased the size of the crystals to
around 10-20 μm in the longest dimension. Crystals harvested on the micro-mesh can be
separated when rastered with the 5 μm mini-beam. Several frames of data at 0 °, 45 ° and 90 °
were collected from a single crystal, which allowed the extraction of the unit cell parameter of
these RsTSPO crystals for the first time. Additional rounds of screening were carried out to
identify crystallization additives such as metals, salts, and organic compounds to further improve
the crystal size and quality. Acetate salts (including sodium acetate, potassium acetate and
ammonium acetate) and 2-propanol improved the crystal resolution to 2.8-3.3 Å. The size of the
crystals also further significantly increased, especially in the second and third dimensions, which
allowed rotation and continuous data collection on these crystals. Ten to twenty degrees of data
wedges could be collected on each single crystal before the crystal was destroyed by the X-ray
beam. Complete datasets for structure determination were obtained by merging 20-40 wedges to
achieve the required completeness and redundancy. A further improvement of the crystal quality
was realized by co-crystallization of RsTSPO with high affinity ligands previously identified
[33]. PK11195, retinoic acid, and Ro5-4864 all improved the crystal size and resolution to some
extent with PK11195 being the best. RsTSPO-PK11195 complex crystals (Figure 4.11A&B)
diffracted to around 2.3-2.5 Å (Figure 4.11C). The sizes of crystals were also much bigger,
which allowed 30-60 degrees up to 90 degrees of data to be collected on a single crystal. The
ability to obtain a complete dataset with much fewer wedges was also important in improving the
data quality and critical to obtaining a better structure solution. A further major improvement
that allowed full data collection on a single crystal was achieved using a mutant form of RsTSPO
that mimicked a human single polymorphism, A147T (A139T in RsTSPO), associated with

163

anxiety-related diseases. Crystals diffracting to 1.8 Å resolution were obtained, which allowed
structural and functional comparison with the wild-type (manuscript in preparation).

164

Figure 4.11. RsTSPO-PK11195 crystals.
(A) Crystals imbedded in the LCP bolus under bright light; (B) Same well as in (A) under
cross-polarizing light; (C) A representive diffraction image of crystals in (A) and (B).

165

Structure determination and refinement
Extensive attempts to solve the structure using various models of RsTSPO, the EM structure
of RsTSPO [32], and the NMR structure of mouse TSPO [34] failed. Despite the fact that some
of the RsTSPO models we previously constructed [33] (in collaboration with Dr. Jens Meiler)
were able to find a potential molecular replacement solution with the correct packing, no reliable
structures

were

obtained.

Concurrently,

experimental

phasing

was

performed.

Selenium-methionine labeled WT crystals were obtained but were not able to solve the phase due
to significant diminution of anomalous signal after merging datasets. Two suitable heavy metal
derivatives of the A139T crystals were obtained (Ta and Hg). The ability to collect complete
datasets on single crystals of the A139T crystal, and the fact that signals from the heavy-metal
derivatives were much stronger, provided significantly better phasing power for structure
determination. The structure of wild type RsTSPO was solved by molecular replacement with a
partially refined model of the A139T mutant.
Structure of RsTSPO-WT and A139T mutant
RsTSPO-A139T crystallized in 2 different space groups: C2 (at 1.8 Å) and P212121 (2.4 Å)
with identical overall structure except for the flexible C-terminus. The structure of the A139T
mutant resolves the nearly complete polypeptide chain for monomers A (residues 1-157), B
(residues 2-157), and C (residues 2-151), with no breaks in the main chain (Figure 4.12). The last
residue A158 is removed with the Histag by trypsin during purification. The wild-type structure,
determined at 2.5 Å resolution, resolved most of the polypeptide chain, except for disordered
portions of Loop 1 (LP1). WT crystallizes in a P21 space group with two monomers in the
asymmetric unit, the same as A139T in the P212121 space group, while A139T in the C2 space
group crystallized with three monomers in the asymmetric unit. In the latter case, the three
166

monomers form a dimer and a half arrangement, such that the monomer C and its symmetry
mate C’ form an identical dimer to the AB dimer. In all cases, RsTSPO is arranged in an identical
parallel dimer. WT and A139T adopt very similar overall structures. The C atom superposition
of the half of the molecule on the N-terminal side of the protein yields rms deviations less than
0.3 Å, while more deviation is observed on the C-terminal side containing LP1, which affects the
access to the central binding cavity. The arrangement of α-helices in the monomer is identical in
all three crystal structures: looking from the C-terminal side of the monomer (Fig. 4.12B), the
clockwise order of the helices is TM-I, TM-II, TM-V, TM-IV, and TM-III. TM-I is kinked at
residue G16 and TM5 is kinked at V130. TM2 and TM4 are curved, while TM3 is slightly
curved and kinked at W87. LP1 adopts three different conformations between the A139T and the
WT structures. A short helix is observed in the middle of LP1 in A139T chain A and chain B with
the LP1 extending up. The short helix is partially unwound in the chain C of the A139T structure
with the LP1 at an even higher up position.

167

Figure 4.12. Crystal structure of RsTSPO-A139T. (A): The overall structure of the A139T
dimer. Transmembrane helices are labeled as roman numerals and rainbow colored from N to C
terminus; (B): Top view from the C-terminuls; (C): bottom view from the N-terminus.

168

DISCUSSION
RsTSPO is unique in terms of being a 5-helix transmembrane α-helical protein as well as an
outer membrane helical protein. These unique properties and the fact that it is very small and
composed of almost exclusively transmembrane helices made it hard to crystallize with the
traditional vapor diffusion methods. By analyzing the crystal structure of RsTSPO and the
systematic optimization route retrospectively, a better understanding of the successful
methodologies can be obtained and potentially applied to other similar membrane proteins
Importance of promoting type I crystal packing for RsTSPO with the LCP method
Despite extensive optimization of crystallization conditions in the vapor diffusion method,
no good diffraction quality crystal was obtained. The lack of large soluble domains and loops to
mediate crystal packing in the third dimension may be detrimental to crystallizing RsTSPO by
the vapor diffusion method. It was clearly shown by the crystal packing of the LCP-grown
crystal (Figure 4.13) that RsTSPO indeed forms type I crystals and the crystal packing
interactions are almost exclusively through helical-helical interactions in the transmembrane
region. In the final crystal structure of wildtype RsTSPO, the only long loop (LP1) that could
mediate potential crystal packing was not completely resolved and was possibly at a downward
position into the transmembrane region, and thus not useful for crystal packing. Therefore,
promoting the type I crystal packing was the key, and perhaps the only way, to obtain the
diffraction quality crystals of RsTSPO. Given that RsTSPO had been previously crystallized into
a 2D crystal [32], this was also a very reasonable strategy.

169

Figure 4.13 Crystal packing of wildtype RsTSPO crystals grown in the LCP condition.
(A) Viewed perpendicular to the crystallography 2-fold axis; (B) Viewed along the
crystallography 2-fold axis; (C) Top view. Unit cells are shown in green boxes. Crystals were
grown in the condition of 32% (w/v) PEG 1000, 100 mM Zn(CH3CO2)2, 100 mM MES (pH 6),
100 mM NaCH3CO2, K or NH4CH3CO2, and 6-8 % (v/v) 2-propanol.

170

The difference between crystal packing is illustrated in Figure 4.14. Crystal packing in the
type II crystals (Figure 4.14C) relies heavily on the interaction of hydrophilic regions not
covered by detergent or lipid, while the transmembrane regions are protected by detergent
throughout the crystallization process and do not contribute to crystal packing. In contrast,
packing of the type I crystals, as well as for the 2D crystals, is heavily dependent on the
transmembrane regions as well as the lipids in between them (Figure 4.14 A&B). The
hydrophilic regions may also contribute to crystal packing, as in the cases of fusion modules
used in crystallization of G-protein coupled receptors [35], but this is not a requirement. In the
case of the RsTSPO-A139T crystals, the ability of the LP1 to form crystal contacts seems
important in improving the crystal quality but transmembrane regions still form the majority of
the crystal contacts. If a 2D crystal of a membrane protein can be induced to also pack in the
third dimension, a type I 3D crystal can be obtained, as in the case of RsTSPO, rhodopsin [1,36],
and the ca3 cytochrome oxidases [9].

171

Figure 4.14 Cartoon illustrations of different crystal types
Beige rectangles represent hydrophobic transmembrane regions of a membrane protein while
purple ovals represent the hydrophilic domains. Blue lines surrounding the transmembrane
domains represent detergents or lipids. (A) 2D crystal; (B) Type I 3D crystal (a head to head
interaction is shown but other types of interactions are also possible); (C) Type II 3D crystal.

172

In larger membrane proteins which have helices that are longer than the detergents and/or
big hydrophilic domains, type II crystals can be obtained through the interaction of the
hydrophilic domains or the transmembrane region not entirely covered by detergents. This is
predominantly the case for membrane protein crystals grown in the vapor diffusion method.
However, for small membrane proteins such as RsTSPO, 3D crystals can most likely be obtained
via type I crystal packing with the transmembrane domain or type II crystals with a fusion
partner. The latter strategy is reasonable for crystallization purposes but the expression of a
fusion protein can be challenging. RsTSPO was fused with three popular fusion modules (T4
lysozyme, rubredoxin, and cytochrome B562RIL) at various positions. All fusion constructs
disrupted the expression and purification of RsTSPO. As most of these fusion partners are even
bigger than RsTSPO itself, it is not surprising that the membrane insertion of RsTSPO was
severely influenced. Success of the fusion protein strategy for obtaining the GPCR structure [37]
suggests that this is a very useful strategy for bigger membrane proteins, but not small membrane
proteins like TSPO.
The LCP crystallization method is not only good for stabilizing membrane proteins in the
lipidic environment, but it is also extremely good for promoting type I crystal packing. To date,
all proteins crystallized with the LCP method formed type I crystal packing. Some of the third
dimension interactions were mediated by a fusion partner, as in the cases of the GPCRs [35]. In
other cases, the third dimension interaction was realized by interdigitation of the transmembrane
region (PDB 2F93 [38], 4H37 [39]). Indeed, type I crystal packing promoted by the LCP is the
only way that RsTSPO-WT could be crystallized to diffraction quality.
Role of detergent in crystallization of RsTSPO
Detergents play important roles in stabilizing membrane proteins, but unsuitable detergents

173

can also inhibit crystallization. In the early efforts of crystalling RsTSPO by the vapor diffusion
method, detergent was extensively optimized to find the detergent or detergent combination with
the minimum micelle size yet also able to stabilize RsTSPO during purification and
crystallization. The rationale behind such optimization is to expose enough of the hydrophilic
region of the membrane protein to mediate crystal packing in the type II crystal, but also to
protect the most hydrophobic region of the membrane protein during the crystallization process.
This strategy was successful for bigger membrane proteins [30], but finding the right balance
turned out to be extremely hard for RsTSPO. As a proof of concept, the combination of DM and
C10E5 indeed produced the only and the best diffraction quality crystals of RsTSPO by the
vapor diffusion method; however, they only diffracted to 16 Å.
The detergent does not seem to play a very important role during crystallization with the
LCP method as long as it can stabilize the membrane protein during purification. Once the
membrane protein is incorporated into the lipidic environment, the MAG used in making the
lipidic cubic phase plays a more important role. Experience with other membrane proteins shows
that the length of the alkyl chain of the MAG is critical in determining the crystallizability and
the crystal quality of membrane protein crystals [40,41]. In addition, experience from our lab and
others shows that detergent molecules are often observed to be bound to the membrane protein in
the crystal grown from the vapor diffusion method [31], but few are observed in crystals grown
from the LCP method. Instead, a large amount of MAG lipid is often observed in the crystals
grown from the LCP method [42], suggesting that detergent and native lipids are replaced by the
LCP lipid once the membrane protein is successfully incorporated into the lipidic cubic phase.
On the other hand, detergent does play a role in regulating the phase behavior of the MAG
lipid used in LCP crystallization. Studies show that increased detergent can act to induce a

174

sponge phase, a type of dissolved cubic phase [43,44], and some membrane proteins are in fact
crystallized in the sponge phase by increasing the detergent and precipitant concentrations [45].
If a proper balance can be achieved, crystals grown from the sponge phase can reach a much
bigger size compared to the regular LCP crystals. Additionally, crystal harvesting is much easier
since the lipidic cubic phase is dissolved under such conditions and the bigger crystals can be
easily harvested with a regular loop. In the case of RsTSPO, controlling the detergent
concentration was not easy because of the small size of the protein relative to the micelle of DM.
Excess detergent completely disrupted the phase behavior of the monoolein and inhibited crystal
growth when the protein concentration was increased in an effort to produce bigger crystals.
Common additives in vapor diffusion and LCP crystallization optimization
Additives to improve crystal quality were extensively screened with both the vapor diffusion
and LCP crystallization methods. One interesting observation is that small alcohols, 2-propanol
in the case of RsTSPO, improved the crystal quality of RsTSPO in both methods with different
crystallization conditions. Similarly, 1,4-butenediol also significantly increased the crystal size
of the β2-adrenergic receptor. Although far from conclusive, these experiences suggest that small
organic molecules may play a role in regulating the phase behavior as well as the interaction of
the membrane protein with lipid/detergent during crystallization.
Zinc acetate was identified in all the initial LCP hits regardless of the major precipitant.
Additional acetate salts (sodium, potassium or ammonium) were identified later as additives that
improved the diffraction quality of RsTSPO crystals. Experience from Dr. Caffrey’s group shows
that a common salt can be identified to improve crystal quality in lipidic cubic phase with
different types of MAGs [40,41]. These experiences show that crystallization and optimization
strategies may be transferred between different systems. Analysis of common additives and

175

results from vapor diffusion experiments may help save time and resources in optimization of
crystals grown by the LCP method.
Obtaining suitible heavy metal derivatives of RsTSPO
Despite the fact that the LCP method is increasingly popular for obtaining crystal structures
of important membrane proteins, few are novel structures. Part of the reason lies in the technical
difficulties of obtaining derivatives of membrane proteins for phase determination under the LCP
conditions. In the process of obtaining the phase information for RsTSPO, two popular methods
were tried: single-wavelength anomalous scattering (SAD), which involved incorporating
selenium into the protein, and multiple isomorphous replacement-anomalous scattering
(MIRAS) , which involved the addition of heavy metals. In our experience, the SAD method
failed due to low signal because many non-optimal datasets from small LCP crystals needed to
be merged. On the other hand, heavy metal derivatization provided better data to obtain the
initial phase information. However, adding the metal without disrupting the LCP system was a
technical challenge. Towards this end, we developed a novel high throughput scheme that uses a
commonly available crystallization robot to deliver the heavy metal compounds onto the plates
prior to the lipid/membrane protein sample in the LCP mixture. In the case of RsTSPO, the
ability to readily obtain a number of heavy metal derivatives was critical for solving the
structure.
Insights from the RsTSPO crystal structures
The crystal structures of RsTSPO WT and the A139T mutant provide some new insights
that facilitate the understanding of TSPO function. The tight interaction of the RsTSPO dimer
suggests that the dimer interface is not likely to transport substrates, in contrast to what was
previously proposed (see chapter 2). Instead, our structures suggest a possible external transport
176

pathway likely involving binding partners, which fits with previous observations of interactions
with other proteins and oligomerization of mouse TSPO and RsTSPO. In addition, significant
structural changes are observed between the A139T mutant and the WT in the area close to the
CRAC site, confirming that this region is playing an important role in cholesterol binding. The
three different conformations of LP1 observed between the WT and the mutant are consistent
with a regulatory role of LP1 in ligand binding and TSPO function. Further, a binding site was
identified with a bound porphyrin, but the drug ligand and cholesterol binding locations were not
observed in the current structure, despite their ability to improve the crystal quality.
The structural information at atomic resolution will enable more incisive experimental
and mutational approaches to fully understand TSPO function.
SUMMARY
Purification and crystallization conditions for RsTSPO were systematically and extensively
screened and optimized with both the vapor diffusion method and the LCP method. Analysis of
the optimization strategies and results for RsTSPO, in comparison with other membrane proteins,
shows that crystallization of small α-helical proteins requires special attention to strategies, in
particular regarding handling of detergents and selection of crystallization methods. A successful
strategy for crystallization of RsTSPO was presented here, providing a valuable example for
crystallization of similar membrane proteins. The structures for both WT and a mutant form of
TSPO were obtained, providing previously unavailable molecular insights into the controvercial
function of this unusual protein.
Manuscript describing the crystal structures of RsTSPO
A paper describing our new structures of WT RsTSPO and the A139T mutant is in final
stages of preparation to submit to Nature or Science. Due to the very restrictive policies of
177

Science, the data cannot be included in a thesis until after its publication. A preliminary abstract
is provided in the Appendix.

178

APPENDIX

179

Crystal structures of translocator protein 18 kDa (TSPO) from Rhodobacter sphaeroides
Fei Li, Jian Liu, Yi Zheng, Michael Garavito, and Shelagh Ferguson-Miller
Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing,
Michigan 48824, USA
Transport of cholesterol into mitochondria is the first and rate limiting step of steroid
hormone synthesis. Translocator protein 18 kDa (TSPO) is recognized as a key player in this
process and is highly expressed in steroidogenic tissues, as well as under conditions of
inflammation, metastatic cancer, Alzheimer, and Parkinson disease. A cholesterol recognition
amino acid consensus sequence (CRAC) was first identified in TSPO as a general cholesterol
binding motif. Various other ligands that bind to TSPO, including benzodiazepine drugs, have
been implicated in regulating apoptosis and extensively studied as diagnostic imaging agents and
treatment options. Conflicting data and the lack of high resolution structures has led to much
controversy concerning the precise role of this conserved, developmentally-required protein.
Here we present high resolution crystal structures, determined in a lipidic phase, of TSPO from
Rhodobacter sphaeroides (RsTSPO), a bacterial homolog of the mitochondrial protein. The
wild-type as well as a mutant equivalent to a human single polymorphism in three forms were
determined to 2.5 Å and 1.8 Å, 2.4 Å respectively. These structures belong to three different
space groups and different ligand binding states, but they show the same dimer arrangement,
consistent with previous results showing that RsTSPO is a dimer in solution. However, our
crystal structures differ significantly from a published NMR structure of the refolded monomeric
mouse TSPO, suggesting that the NMR structure may represent a ligand-trapped folding
intermediate. The crystal structures of RsTSPO reveal ligand interaction sites and provide the
basis for understanding the nature of cholesterol interactions, as well as serving as a platform for
structurally-guided drug design targeting of TSPO.
180

REFERENCES

181

REFERENCES

1. Caffrey M, Cherezov V: Crystallizing membrane proteins using lipidic mesophases. Nature
Protocols 2009, 4:706-731.
2. Faham S, Bowie JU: Bicelle crystallization: a new method for crystallizing membrane
proteins yields a monomeric bacteriorhodopsin structure. J Mol Biol 2002, 316:1-6.
3. Cherezov V, Peddi A, Muthusubramaniam L, Zheng YF, Caffrey M: A robotic system for
crystallizing membrane and soluble proteins in lipidic mesophases. Acta Crystallogr
D Biol Crystallogr 2004, 60:1795-1807.
4. Fischetti RF, Xu S, Yoder DW, Becker M, Nagarajan V, Sanishvili R, Hilgart MC, Stepanov S,
Makarov O, Smith JL: Mini-beam collimator enables microcrystallography
experiments on standard beamlines. J Synchrotron Radiat 2009, 16:217-225.
5. Gourdon P, Andersen J, Hein K, Bublitz M, Pedersen B, Liu X, Yatime L, Nyblom M, Nielsen
T, Olesen C, et al.: HiLiDe-systematic approach to membrane protein crystallization
in lipid and detergent. Crystal Growth & Design 2011, 11:2098-2106.
6. Yang J, Kulkarni K, Manolaridis I, Zhang Z, Dodd RB, Mas-Droux C, Barford D: Mechanism
of isoprenylcysteine carboxyl methylation from the crystal structure of the integral
membrane methyltransferase ICMT. Mol Cell 2011, 44:997-1004.
7. Landau EM, Rosenbusch JP: Lipidic cubic phases: a novel concept for the crystallization
of membrane proteins. Proc Natl Acad Sci U S A 1996, 93:14532-14535.
8. Cherezov V, Yamashita E, Liu W, Zhalnina M, Cramer WA, Caffrey M: In meso structure of
the cobalamin transporter, BtuB, at 1.95 A resolution. J Mol Biol 2006, 364:716-734.
9. Lyons JA, Aragão D, Slattery O, Pisliakov AV, Soulimane T, Caffrey M: Structural insights
into electron transfer in caa3-type cytochrome oxidase. Nature 2012, 487:514-518.
10. Rasmussen SG, DeVree BT, Zou Y, Kruse AC, Chung KY, Kobilka TS, Thian FS, Chae PS,
Pardon E, Calinski D, et al.: Crystal structure of the β2 adrenergic receptor-Gs
protein complex. Nature 2011, 477:549-555.
11. Alexandrov AI, Mileni M, Chien EY, Hanson MA, Stevens RC: Microscale fluorescent
thermal stability assay for membrane proteins. Structure 2008, 16:351-359.
12. Vedadi M, Niesen FH, Allali-Hassani A, Fedorov OY, Finerty PJ, Wasney GA, Yeung R,
Arrowsmith C, Ball LJ, Berglund H, et al.: Chemical screening methods to identify
ligands that promote protein stability, protein crystallization, and structure
determination. Proc Natl Acad Sci U S A 2006, 103:15835-15840.
13. Studier FW: Protein production by auto-induction in high density shaking cultures.
Protein Expr Purif 2005, 41:207-234.
182

14. Eriks LR, Mayor JA, Kaplan RS: A strategy for identification and quantification of
detergents frequently used in the purification of membrane proteins. Anal Biochem
2003, 323:234-241.
15. Urbani A, Warne T: A colorimetric determination for glycosidic and bile salt-based
detergents: applications in membrane protein research. Anal Biochem 2005,
336:117-124.
16. Schneider CA, Rasband WS, Eliceiri KW: NIH Image to ImageJ: 25 years of image
analysis. Nat Methods 2012, 9:671-675.
17. Lambert O, Levy D, Ranck JL, Leblanc G, Rigaud JL: A new "gel-like" phase in dodecyl
maltoside-lipid mixtures: implications in solubilization and reconstitution studies.
Biophys J 1998, 74:918-930.
18. Newby ZER, O'Connell JD, Gruswitz F, Hays FA, Harries WEC, Harwood IM, Ho JD, Lee
JK, Savage DF, Miercke LJW, et al.: A general protocol for the crystallization of
membrane proteins for X-ray structural investigation. Nature Protocols 2009,
4:619-637.
19. Li D, Boland C, Aragao D, Walsh K, Caffrey M: Harvesting and cryo-cooling crystals of
membrane proteins grown in lipidic mesophases for structure determination by
macromolecular crystallography. J Vis Exp 2012:e4001.
20. Liu W, Cherezov V: Crystallization of membrane proteins in lipidic mesophases. J. Vis.
Exp. (49), e2501, doi:10.3791/2501 (2011).
21. Kabsch W: XDS. Acta Crystallogr D Biol Crystallogr 2010, 66:125-132.
22. Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P, Evans PR, Keegan RM, Krissinel
EB, Leslie AG, McCoy A, et al.: Overview of the CCP4 suite and current
developments. Acta Crystallogr D Biol Crystallogr 2011, 67:235-242.
23. Evans P: Scaling and assessment of data quality. Acta Crystallogr D Biol Crystallogr 2006,
62:72-82.
24. Bricogne G, Vonrhein C, Flensburg C, Schiltz M, Paciorek W: Generation, representation
and flow of phase information in structure determination: recent developments in
and around SHARP 2.0. Acta Crystallogr D Biol Crystallogr 2003, 59:2023-2030.
25. Adams PD, Afonine PV, Bunkóczi G, Chen VB, Davis IW, Echols N, Headd JJ, Hung LW,
Kapral GJ, Grosse-Kunstleve RW, et al.: PHENIX: a comprehensive Python-based
system for macromolecular structure solution. Acta Crystallogr D Biol Crystallogr
2010, 66:213-221.
26. Emsley P, Lohkamp B, Scott WG, Cowtan K: Features and development of Coot. Acta
Crystallogr D Biol Crystallogr 2010, 66:486-501.

183

27. Pebay-Peyroula E, Dahout-Gonzalez C, Kahn R, Trézéguet V, Lauquin GJ, Brandolin G:
Structure of mitochondrial ADP/ATP carrier in complex with carboxyatractyloside.
Nature 2003, 426:39-44.
28. Schulze S, Kӧster S, Geldmacher U, Terwisscha van Scheltinga AC, Kühlbrandt W:
Structural basis of Na(+)-independent and cooperative substrate/product antiport in
CaiT. Nature 2010, 467:233-236.
29. Perez C, Koshy C, Yildiz O, Ziegler C: Alternating-access mechanism in
conformationally asymmetric trimers of the betaine transporter BetP. Nature 2012,
490:126-130.
30. Lemieux MJ, Song J, Kim MJ, Huang Y, Villa A, Auer M, Li XD, Wang DN:
Three-dimensional crystallization of the Escherichia coli glycerol-3-phosphate
transporter: a member of the major facilitator superfamily. Protein Sci 2003,
12:2748-2756.
31. Qin L, Hiser C, Mulichak A, Garavito RM, Ferguson-Miller S: Identification of conserved
lipid/detergent-binding sites in a high-resolution structure of the membrane protein
cytochrome c oxidase. Proc Natl Acad Sci U S A 2006, 103:16117-16122.
32. Korkhov VM, Sachse C, Short JM, Tate CG: Three-dimensional structure of TspO by
electron cryomicroscopy of helical crystals. Structure 2010, 18:677-687.
33. Li F, Xia Y, Meiler J, Ferguson-Miller S: Characterization and modeling of the oligomeric
state and ligand binding behavior of purified Translocator Protein 18 kDa from
Rhodobacter sphaeroides. Biochemistry 2013, 52:5884-5899.
34. Jaremko L, Jaremko M, Giller K, Becker S, Zweckstetter M: Structure of the
mitochondrial translocator protein in complex with a diagnostic ligand. Science 2014,
343:1363-1366.
35. Chun E, Thompson AA, Liu W, Roth CB, Griffith MT, Katritch V, Kunken J, Xu F, Cherezov
V, Hanson MA, et al.: Fusion partner toolchest for the stabilization and
crystallization of G protein-coupled receptors. Structure 2012, 20:967-976.
36. Fotiadis D, Liang Y, Filipek S, Saperstein DA, Engel A, Palczewski K: The G
protein-coupled receptor rhodopsin in the native membrane. FEBS Lett 2004,
564:281-288.
37. Katritch V, Cherezov V, Stevens RC: Structure-function of the G protein-coupled receptor
superfamily. Annu Rev Pharmacol Toxicol 2013, 53:531-556.
38. Moukhametzianov R, Klare JP, Efremov R, Baeken C, Göppner A, Labahn J, Engelhard M,
Büldt G, Gordeliy VI: Development of the signal in sensory rhodopsin and its
transfer to the cognate transducer. Nature 2006, 440:115-119.
39. Santos JS, Asmar-Rovira GA, Han GW, Liu W, Syeda R, Cherezov V, Baker KA, Stevens RC,
184

Montal M: Crystal structure of a voltage-gated K+ channel pore module in a closed
state in lipid membranes. J Biol Chem 2012, 287:43063-43070.
40. Li D, Shah ST, Caffrey M: Host lipid and temperature as important screening variables
for crystallizing integral membrane proteins in lipidic mesophases. Trials with
diacylglycerol kinase. Cryst Growth Des 2013, 13:2846-2857.
41. Li D, Howe N, Dukkipati A, Shah ST, Bax BD, Edge C, Bridges A, Hardwicke P, Singh OM,
Giblin G, et al.: Crystallizing membrane proteins in the lipidic mesophase.
Experience with human prostaglandin E2 synthase 1 and an evolving strategy. Cryst
Growth Des 2014, 14:2034-2047.
42. Li D, Lyons JA, Pye VE, Vogeley L, Aragão D, Kenyon CP, Shah ST, Doherty C, Aherne M,
Caffrey M: Crystal structure of the integral membrane diacylglycerol kinase. Nature
2013, 497:521-524.
43. Cherezov V, Clogston J, Papiz MZ, Caffrey M: Room to move: crystallizing membrane
proteins in swollen lipidic mesophases. J Mol Biol 2006, 357:1605-1618.
44. Wӧhri AB, Johansson LC, Wadsten-Hindrichsen P, Wahlgren WY, Fischer G, Horsefield R,
Katona G, Nyblom M, Oberg F, Young G, et al.: A lipidic-sponge phase screen for
membrane protein crystallization. Structure 2008, 16:1003-1009.
45. Waight AB, Pedersen BP, Schlessinger A, Bonomi M, Chau BH, Roe-Zurz Z, Risenmay AJ,
Sali A, Stroud RM: Structural basis for alternating access of a eukaryotic
calcium/proton exchanger. Nature 2013, 499:107-110.

185

CHAPTER 5
Perspective and future direction on structure-function studies of TSPO

186

TSPO proteins have been investigated for more than 30 years, resulting in a wealth of
information being available. Determination of the first high resolution crystal structure of
RsTSPO and the ability to purify RsTSPO in the functional state opens a new door for
investigation of the structure-function relationships of TSPO family proteins. Crystal structures
of RsTSPO, as well as the NMR structure of the mouse TSPO, provide a new opportunity to
interpret previous mutagenesis data from a structure-function point of view. Additional
mutagenesis experiments can also be designed with the help of the structures. On the other hand,
current data and interpretation are complicated by the fact that experiments are often performed
in different cell or tissue systems, which are hard for cross validation and often result in
inconsistant conclusions. The successful expression, purification and crystallization of RsTSPO,
as well as the established tryptophan fluorescence quenching binding assay, provides a platform
for testing the structure-function relationship of RsTSPO. Similar protocols can also be adapted
for expression, purification and charaterization of other TSPO family proteins. Much is still
unknown regarding the function of this important and conserved protein family and several lines
of investigation can be followed.
Crystallization of RsTSPO-ligand complexes
Although ligands were added into the current crystallization experiments and improved the
diffraction quality of RsTSPO crystals, only partial densities of ligands were observed in
potential ligand binding sites of RsTSPO. Several schemes will be used to improve the
incorporation of ligands into RsTSPO crystals: 1) ligand concentrations will be increased to a
higher level with the current method under conditions in which crystal quality will not be
significantly influenced by the addition of excessive ligand and solvent DMSO; 2) RsTSPO
crystals will be soaked with mother liquor supplemented with high concentrations of ligand

187

before crystal harvesting; 3) mutants of RsTSPO with higher ligand binding affinity will be
designed and crystallized. Of particular interest is the cholesterol binding site, so far not revealed
in our structures, mainly due to the fact that the wile-type (WT) Rhodobacter protein does not
bind cholesterol with high affinity. The ATA/LAF (136,137,138) mutant was shown to bind both
cholesterol and cholesterol hemisuccinate much more tightly than WT (Chapter 3) and well
diffracting crystals are being sought. The A139T mutant, predicted to bind cholesterol less tightly,
did indeed show this behavior, as well as producing our best crystals so far.
Preliminary data show that RsTSPO-LAF can be purified to good homogeneity and yield
similar to RsTSPOWT by using CHS-supplemented buffer. Crystallization conditions will be
screened and optimized with the robot to obtain diffraction quality crystals of RsTSPO-LAF.
Mutagenesis of RsTSPO based on the crystal structure and the co-evolution analysis
Mutants of RsTSPO can be designed based on the co-evolution analysis and the crystal
structure of RsTSPO to achieve good protein stability, as well to test hypotheses regarding the
function of TSPO in vitro with the established tryptophan fluorescence quenching assay. For
instance, residues on helix 4 that are prediced to interact with the LAF motif can be mutated to
complement the loss of stability of RsTSPO-LAF; residues at the dimer interface can be mutated
to test the function of the dimer; residues involved in the potential ligand binding site can be
mutated to study the ligand binding mechanism as well as to test the influence of ligand binding
in vitro.
Characterization of the oligomeric state of RsTSPO and the stoichiometry of the
endogenous ligand by mass spectrometry
Oligomerization of TSPO proteins have been reported but the interactions that maintain the
oligomer are not yet identified and its functional significance remains to be further investigated.

188

All our crystals so far show the same dimer form, but whether the solution form is the same, and
whether higher order associations exist, remains to be investigated. A collaboration has been
established with Dr. Brandon Ruotolo and Dr. Philip Andrews at the University of Michigan to
further characterize the oligomerization of purified RsTSPO with the ion-mobility mass
spectrometry (IM-MS) and cross-linking methodology. Preliminary data show that RsTSPO
monomer and dimer species are identified in the mass spectrum and purified RsTSPO was
cross-linked by the amine cross-linker DC4. Optimization of cross-linking experiments
combined with mutagenesis would allow us to identify residues critical for oligomerization.
Further investigation with IM-MS would also allow for identification of bound lipids and ligands
co-purified with RsTSPO in low abundance.
Expression, purification and crystallization of human TSPO
Despite the high sequence identity that RsTSPO shares with human TSPO, considerable
differences exist between RsTSPO and human TSPO, in particular at the loop 1 region where a
role was proposed in ligand binding. In addition, given the much higher affinity to drug ligands,
the chance of observing a bound ligand with the human TSPO is better. High resolution crystal
structures of the human TSPO with ligands bound will undoubtedly provide more direct
information for understanding of the regulation of TSPO by ligands and for facilitating drug
development targeting TSPO. The successful expression, purification and crystallization of
RsTSPO provides a good guide for other TSPO family proteins, including the human TSPO.
Human TSPO has been successfully expressed in E. coli with the current purification
procedure but was co-purified with GroEL. Further optimization of purification will be needed to
purify the human protein from E. coli. Meanwhile, preliminary data show that human TSPO can
be expressed and purified to good yield and homogeneity for crystallization with the sf9 insect

189

cell expression system (personal communication from Dr. Yi Zheng). The expression in
Rhodobacter is also a possibility, since R. sphaeroides can successfully express many membrane
proteins and a preliminary result shows that human TSPO was expressed in Rhodobacter.

190