insight into the mitochondrial apoptotic...
TRANSCRIPT
Insight into the mitochondrial apoptotic pathway:
The interplay of the pro-apoptotic Bax protein with oxidized phospholipids and its counterplayer, the pro-survival Bcl-2 protein
Marcus Wallgren
Department of Chemistry
Doctoral Thesis
Umeå 2012
© Marcus Wallgren, pp. i-iv, 1-62 ; Umeå, 2012
This work is protected by the Swedish Copyright Legislation (Act 1960:729)
ISBN: 978-91-7459-490-4
Cover picture: Model of OxPl-induced Bax activation at MOM, by Marcus Wallgren
Electronic version available at http://umu.diva-portal.org/
Printed by: VMC, KBC, Umeå University
Umeå, Sweden 2012
Paper I is reproduced by permission of The Royal Society of Chemistry
Papers II and III are reprinted with permission from the publishers
To both of my families
Table of contents
i
Table of contents
1 List of papers ii
2 Abbreviations iii
3 Abstract iv
4 Introduction 1
4.1 Biological background 2
4.1.1 Apoptosis 2
4.1.2 Membrane systems 5
4.1.2.1 Mitochondrion 5
4.1.2.2 Phospholipids 7
4.1.2.3 Oxidized phospholipids (OxPls) 10
4.1.3 Regulation of the mitochondrial apoptosis by the Bcl-2 protein family 12
4.1.3.1 The activation and pore-forming action of Bax 16
4.1.3.2 Bcl-2 – the guardian of cell integrity 18
4.2 Aim of this thesis 20
5 Methods 21
5.1 Differential scanning calorimetry (DSC) 21
5.2 Nuclear magnetic resonance (NMR) 22
5.2.1 Solid state NMR 24
5.2.1.1 31P NMR 25
5.2.1.2 2H NMR 26
5.2.1.3 MAS 26
5.3 Circular dichroism (CD) spectroscopy 27
5.4 Cell-free protein synthesis 28
6 Results and discussion 31
6.1 Influence of OxPls on membrane properties and Bax-membrane interactions 31
6.1.1 OxPls have a pronounced impact on membrane structure and dynamics 31
6.1.2 Modulation of Bax-membrane interactions by OxPls 34
6.1.3 OxPl-induced recruitment of Bax to the MOM during apoptosis 36
6.2 Cell-free protein synthesis of Bcl-2 38
6.2.1 Gene optimization 38
6.2.2 Batch-mode cell-free expression 39
6.3 Interplay between Bax and Bcl-2 41
6.4 Reconstitution of Bcl-2 into proteoliposomes 43
7 Main findings and outlook 46
8 Acknowledgements 48
9 References 51
List of papers
ii
1 List of papers
This thesis is based on the following papers:
I: Wallgren, M., Beranova, L., Pham, Q. D., Linh, K., Lidman, M., Procek, J.,
Cyprych, K., Kinnunen, K. J. P., Hof, M., Gröbner, G. “Impact of oxidized
phospholipids on the structural and dynamic organization of phospholipid
membranes: a combined DSC and solid state NMR study”, Faraday
Discussions, In press
II: Wallgren, M., Lidman, M., Pham, Q. D., Cyprych, K., Gröbner, G. “The
oxidized phospholipid PazePC modulates interactions between Bax and
mitochondrial membranes”, Biochimica et Biophysica Acta – Biomembranes,
1818, 2718-2724, 2012
III: Pedersen, A.‡, Wallgren, M.‡, Karlsson, B. G., Gröbner, G. “Expression and
purification of full-length anti-apoptotic Bcl-2 using cell-free protein
synthesis”, Protein Expression and Purification, 77, 220-223, 2011
IV: Wallgren, M., Lidman, M., Pedersen, A., Karlsson, B. G., Gröbner, G.
“Reconstitution of the anti-apoptotic Bcl-2 protein into lipid membranes and
biophysical evidence of its detergent-driven association with the pro-apoptotic
Bax protein”, Submitted to Biochemistry
Additional papers which are not included in the thesis:
Wallgren, M.‡, Ådén J.‡, Pylypenko, O., Mikaelsson, T., Johansson, L. B.,
Rak, A., Wolf-Watz, M. “Extreme temperature tolerance of a
hyperthermophilic protein governed by residual structure in the unfolded
state”, Journal of Molecular Biology, 379, 845-858, 2008
Rundqvist, L., Ådén, J., Sparrman, T., Wallgren, M., Olofsson, U., Wolf-
Watz, M. “Non-cooperative folding of subdomains in Adenylate Kinase”,
Biochemistry, 48, 1911-1927, 2009
Ådén, J., Wallgren, M., Storm, P., Weise, C., Christiansen, A., Schröder, W.,
Funk, C., Wolf-Watz, M. “Arabidopsis thaliana Peroxiredoxin Q is
extraordinary dynamic on the µs-ms timescale”, Biochimica et Biophysica
Acta – Proteins and Proteomics, 1814, 1880-1890, 2010
‡The authors contributed equally to this work
Abbreviations
iii
2 Abbreviations
Bax Bcl-2-associated X protein
Bcl-2 B-cell CLL/lymphoma 2
Brij-35 Polyoxyethylene-(23)-lauryl-ether
Brij-58 Polyoxyethylene-(20)-cetyl-ether
CD Circular dichroism
CL Cardiolipin
CMC Critical micelle concentration
CS Contact sites
CSA Chemical shift anisotropy
H Change in enthalpy
DMPC 1,2-dimyristoyl-sn-glycero-3-phosphocholine
DSC Differential scanning calorimetry
FID Free induction decay
hr Hot rod
IMM Inner mitochondrial membrane
IMS Intermembrane space
MAS Magic angle spinning
MLVs Multilamellar vesicles
MOM Mitochondrial outer membrane
MOMP Mitochondrial outer membrane permeabilization
NMR Nuclear magnetic resonance
OxPls Oxidized phospholipids
PazePC 1-palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine
PC Phosphatidylcholine
PE Phosphatidylethanolamine
POPC 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine
POPE 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine
PoxnoPC 1-palmitoyl-2-(9´-oxononanoyl)-sn-glycero-3-phosphocholine
PS Phosphatidylserine
ROS Reactive oxygen species
SUVs Small unilamellar vesicles
tBid Truncated Bid
Tm Melting temperature
TMCL 1,1',2,2'-tetra-myristoyl-cardiolipin
Abstract
iv
3 Abstract
Apoptosis plays a crucial role in multicellular organisms by preserving tissue
homeostasis and removing harmful cells. The anti-apoptotic B-cell CLL/lymphoma 2
(Bcl-2) and the pro-apoptotic Bcl-2-associated X protein (Bax) act as major
regulators of the mitochondrial apoptotic pathway. Activation of Bax via stress
signals causes its translocation to the mitochondrial outer membrane (MOM). There,
Bax forms homo-oligomeric pores, leading to the release of apoptogenic factors,
caspase activation and ultimately cell death. However, the underlying mechanism for
the recruitment and pore forming activity of Bax is still not elucidated. Nevertheless,
the mitochondrial membrane system seems to play an active and crucial role,
presumably being directly involved in the onset of the mitochondrial apoptosis. Since
the formation of reactive oxygen species (ROS) is a common stress signal and one of
the hallmarks of the mitochondrial apoptosis, direct damage can occur to these
membranes by the generation of oxidized phospholipids (OxPls), whose presence can
crucially influence the pro-apoptotic action of Bax there. To better understand the
impact of OxPls on membranes as well as their potential role in the mitochondrial
apoptotic process, defined OxPl species were incorporated into phospholipid vesicles
and studied with various biophysical techniques. Differential scanning calorimetry
(DSC) and solid state nuclear magnetic resonance (NMR) spectroscopy were used to
gain insight into changes in membrane properties in the presence of OxPls. In
addition to circular dichroism (CD) spectroscopy, DSC and solid state NMR were
furthermore performed to elucidate the impact of OxPls on Bax-membrane
interactions. The occurrence of OxPls gave rise to dramatic changes in membrane
organization and dynamics, manifested as lateral phase separation into OxPl-rich
and -poor domains and modified hydration at the membrane interface. The presence
of OxPls also had a great impact on the interaction between Bax and mitochondria-
mimicking vesicles, strongly promoting the association of the protein with the
membrane.
At the MOM, Bax is believed to be inhibited by Bcl-2. How this inhibition occurs is
still a mystery due to the lack of biophysical information on Bcl-2, in particular on
the full-length protein variant. Since Bcl-2 is also one of the main culprits in the
progression of various forms of cancer, knowledge of the structural and mechanistic
properties of the full-length protein is essential for a fundamental understanding of
its function at a molecular level. To this end, a method for the production of full-
length Bcl-2 was developed. By performing cell-free protein synthesis, preparative
amounts of the protein were obtained, which enabled a biophysical characterization
of the putative interaction between Bax and Bcl-2 using CD and fluorescence
spectroscopy. A protocol for the reconstitution of Bcl-2 into proteoliposomes was
also developed, promising for future studies of the full-length protein in its native
membrane environment; a prerequisite to fully understand its pro-survival functions
as well as providing crucial information for the design of novel anti-cancer drugs.
Introduction
1
4 Introduction
To sum up around five years of doctoral studies into one report is a challenging
task in many ways. One problem is the risk of blabbering too much about the basic
theory that “all” people already are well aware of, which inevitably leads to tedious
and boring reading. Or to the opposite, by writing too little one potentially leaves out
some of the relevant background information of the presented work. Another issue is
the consistency; all the papers in the thesis should ideally be boiled down into one
neat, concise story. Yet another concern is the outline of the report; should one take
the “easy” route and divide the thesis into distinct chapters and summarize each
included paper individually, or make an effort to construct an intricate piece of work
where all the papers are intertwined into one fascinating story?
When I was writing this thesis I tried to account for these above mentioned
parameters to provide a decent (and hopefully clear) illustration of my years spent in
the lab and in front of the computer. Perhaps the most difficult problem to solve
when writing a doctoral thesis – something that one may not always be able to
influence – is the consistency. To be able to write a nice, consistent story one needs to
have the material to achieve such a feat. For many fellow PhD students the famous
ketchup bottle effect seems to occur in the final year(s). The first years involve a lot of
struggle but little results – inevitably the project starts to occupy one’s mind more
and more, maybe leading to a slight (or severe) brooding over the situation. In most
cases the tides will turn after a couple of years and what in the past seemed to be a
Herculean feat may then turn out to be a still challenging, but doable task. Past and
current colleagues have told me about their experiences and how their projects were
wrapped up well in the end. For a long time I thought there would be an alternative
end for my thesis (a not so happy one; thesis is not finished, yet you still have to
survive), but thanks to a remarkable last year (especially May and June, which were
two amazing months), I can here proudly present my contribution to the scientific
community. So, without further delay, let’s embark on the quest of understanding
apoptosis, including the functions of the lipids and proteins involved in this
fundamental programmed cell death process.
Introduction
2
4.1 Biological background
4.1.1 Apoptosis
The turnover of cells in multicellular organisms is tightly regulated by the
evolutionary conserved cell death program called apoptosis. Depending on cell type
the life span ranges from a few days (e.g. intestine) to the whole life span of the
organism (e.g. neurons) 1. In average almost one million cells die each second and
become replaced by new ones via mitosis. Apart from regulating tissue homeostasis,
apoptosis also enables the controlled removal of damaged and harmful cells, crucial
for numerous processes such as tissue homeostasis, embryogenesis and defense
against pathogens. Thus apoptosis can be distinguished from necrosis, the latter
occurring as a response to non-controlled, acute tissue damage. Necrosis therefore
only takes place under non-physiological conditions whereas apoptosis occurs in both
healthy and disease states. Due to being involved in numerous crucial processes in
cell developmental biology and cytotoxic functions in the immune system, apoptosis
has during the last few decades gained a lot of research interest. In addition,
dysfunctional apoptosis can cause numerous fatal conditions; excess cell death can
lead to pathologies such as neurodegenerative disorders 2 and stroke 3, while
insufficient cell death rates can cause the formation of cancer tumors.
The term apoptosis was first introduced in 1972 by John Kerr, Andrew Wyllie and
Alistair Currie, which in turn had been suggested by James Cormack, at that time a
professor at the Department of Greek at University of Aberdeen 4. In the ancient
Greek language, apoptosis means “dropping off” petals or leaves from healthy plants,
reflecting the formation of the so-called apoptotic bodies that originate from a dying
cell. Initially, most of the apoptosis research was conducted using the nematode
Caenorhabditis elegans, an ideal model system to elucidate the underlying genetics
and biochemical pathways of apoptosis. For this work Sydney Brenner, Robert
Horvitz and John Sulston, were awarded the Noble Prize in Medicine 2002. Horvitz
made the important identification of key genes regulating the apoptotic machinery,
and also proved that the corresponding genes (and their functions) exist in higher
organisms such as humans 5. The nematode genes ced-3, ced-4 and ced-9 were found
to be homologous to the human caspase-9, apaf-1 and bcl-2 genes 6, which are all
deeply involved in the regulation of the mitochondrial apoptotic pathway (see below
and section 4.1.3).
The apoptotic degradation of a cell is accompanied by several characteristic
features (Fig. 1), and includes fragmentation of the nucleus, chromatin condensation
and cell shrinkage 7. The cell membrane starts to bulge outward, forming blebs which
are then budding off as apoptotic bodies 4. This blebbing is thought to occur due to
hydrolysis of sphingomyelin to ceramide, leading to a cholesterol efflux (the
interaction between cholesterol and sphingomyelins is lost) and alteration of the
bilayer 8. The translocation of phosphatidylserine (PS) to the outer leaflet of the
plasma membrane acts as a recognition ”eat me” signal for neighboring cells and
Introduction
3
macrophages, which will then phagocytize the apoptotic bodies 9. Since the content of
an apoptotic cell is not spilled out but remains enclosed in the apoptotic bodies
(ready for engulfment), there is no inflammatory response, in contrast to necrosis
where the cell membrane loses its integrity and the leaked content triggers an
inflammatory reaction 10.
Figure 1. Different characteristics of cells undergoing necrosis or apoptosis. The apoptotic cell preserves its
membrane integrity in contrast to the necrotic counterpart, and upon cell shrinkage it forms blebs that
subsequently become phagocytized (with permission from Saikumar et al., 1999 11).
Two main pathways have been proposed to convey apoptosis, namely the more
recently evolved extrinsic pathway and the evolutionary older intrinsic pathway (Fig.
2) 12; 13; 14. Both of these pathways initiate the activation of executioners, namely the
caspases (cysteine-aspartic acid proteases) -3 and -7, which in turn activate
downstream caspases, leading to the cleavage of intracellular key proteins and the
demise of the cell.
The extrinsic pathway is triggered by extracellular signals such as hormones and
toxins, and is mediated by death receptor proteins belonging to the tumor necrosis
factor receptor family 12; 15. These transmembrane proteins are located in the cell
membrane and are characterized by an 80 amino acid long intracellular death
domain. When the death receptors are engaged by their cognate ligands they become
activated and oligomerized. The oligomeric form of the death receptors binds then
(via their death domain) to the death domain of adaptor proteins (e.g. FADD). These
adaptor proteins carry another domain (the death effector domain) that can interact
with the corresponding domain of procaspase-8 (caspase-10 in humans), forming a
complex called DISC. This will lead to the activation of initiator caspase-8 16, which in
turn will cleave and activate the executioner caspases 17; 18.
Introduction
4
The intrinsic or mitochondrial pathway is mediated by the powerhouse of the cell,
the mitochondrion, and is triggered by various intracellular signals, which are
collectively termed stress signals 19; 20. These signals include e.g. DNA damage,
hypoxia, growth factor deprivation, cytoskeleton disruption and reactive oxygen
species (ROS). The B-cell CLL/lymphoma 2 (Bcl-2) protein family plays a major role
in the regulation of the intrinsic pathway and the opposing members of this family
interact at the mitochondrial level to dictate the fate of the cell (see section 4.1.3) 21.
During normal cell conditions anti-apoptotic Bcl-2 proteins sequester and inhibit the
activity of pro-apoptotic counterparts, ensuring the survival of the cell. But upon an
intrinsic apoptotic signal, pro-apoptotic family members become activated, thereby
overcoming the protective action of the pro-survival Bcl-2 proteins. Some pro-
apoptotic Bcl-2 family members can in their activated state oligomerize into pores in
the mitochondria, giving rise to mitochondrial outer membrane permeabilization
(MOMP) 22. These pores enable the release of apoptogenic factors such as cytochrome
c from the intermembrane space (IMS) out to the cytosol. Cytochrome c can engage
Apaf-1 and together they form the scaffold of a heptameric complex termed the
apoptosome 23; 24; 25. This apoptosome cleaves procaspase-9, forming the mature
initiator caspase-9, which in turn activates caspase-3 and caspase-7.
Figure 2. Schematic outlines of the extrinsic and intrinsic apoptotic pathways, arbitrated by the death
receptors and Bcl-2 family proteins, respectively. Both pathways converge in the triggering of a caspase
activating cascade, leading to apoptotic cell death (adapted with permission from Nys et al., 2011 26).
Intrinsic pathway
Extrinsic pathway
Introduction
5
The extrinsic and extrinsic pathways act separately in so-called Type I cells (e.g.
lymphocytes), but are interconnected in Type II cells (e.g. liver cells) 27; 28. In Type II
cells, death receptor-activated caspase-8 can in addition to activate executioner
caspases cleave and activate Bid. Bid is a pro-apoptotic Bcl-2 family protein and the
truncated variant, tBid, triggers the pore forming action of other pro-apoptotic Bcl-2
proteins in the mitochondrial outer membrane (MOM) 29; 30; 31. Hence, both the
extrinsic and intrinsic pathways converge on the degradation of the mitochondria to
facilitate the cell death process.
4.1.2 Membrane systems
4.1.2.1 Mitochondrion
One of the main responsibilities of the mitochondrion is the regulation of the
programmed cell death in the cell. Besides this important task, mitochondria are also
involved in numerous other crucial processes, ranging from the biosynthesis of lipids
and amino acids to the production of energy in the form of ATP molecules 32.
Originally evolved from a prokaryote presumably according to the endosymbiotic
theory (akin to e.g. chloroplasts) 33, the organelle still retains some bacterial features.
It stores its own DNA, though reduced to only encode a handful of the large number
of proteins (depending on tissue type) that are contained in the organelle, the
majority being encoded for by the nuclear genome 34. The mitochondrial DNA is
passed on to the next generation almost solely via the oocytes. The size of the
mitochondrion is similar to bacteria, ranging from 0.5 to 1 m in diameter, and its
number in a cell varies depending on the metabolic demand in that particular cell;
e.g. muscle cells contain several thousand mitochondria. The organelle is composed
of two membranes (Fig. 3), namely the MOM and the inner mitochondrial membrane
(IMM). These membrane systems are separated by the IMS, but are interconnected
via specific contact sites (CS). The apoptosis relevant cytochrome c is located in the
IMS, the majority being bound to the outer leaflet of the IMM in a complex with the
mitochondria-specific phospholipid cardiolipin (CL) 35. On the other side of the IMM
is the matrix, which harbors numerous proteins involved in metabolic processes such
as the citric acid cycle and -oxidation of fatty acids, granules which are believed to
store calcium 36, as well as the mitochondrial DNA. Both the MOM and the IMM have
a high abundance of proteins, especially the IMM 37. The MOM is highly permeable to
small molecules and proteins due to the action of channel forming porins and the
transporter protein translocase of the outer membrane, respectively. The IMM on the
other hand serves as an impregnable barrier to the matrix, and since there are no
porins located in the IMM most if not all molecules (including proteins) have to be
actively transported across the membrane. Specific proteins that are associated with
the IMM are crucially involved in the ATP production, via the redox reactions in the
electron transport chain underlying oxidative phosphorylation. The invaginations of
the IMM, forming regions called cristae, greatly increase the surface area of the
Introduction
6
membrane. This allows for a more efficient ATP production since the abundance of
the proteins involved in the oxidative phosphorylation, including ATP synthase,
becomes significantly larger. The oxygen-rich milieu in mitochondria, which is a
prerequisite for the ATP production, also gives rise to ROS such as oxygen radicals
and hydrogen peroxide 38; 39. These agents can damage cellular key components (e.g.
proteins, DNA and lipids) and have implications on the progress of apoptosis.
Figure 3. (A) Transmission electron micrograph of the mitochondrion, and (B) schematic picture of the
organelle, highlighting the two membranes (MOM and IMM), the space separating them (IMS), the many
invaginations (cristae), and the inner compartment (matrix) containing the granules (adapted with
permission from Medical Cell Biology, Academic Press, 2008, pp 101-148 40).
Mitochondria are highly dynamic organelles, for which the supply and transport of
lipids and proteins need to be tightly regulated in order to meet the demands of
different physiological processes 41; 42. Depending on their condition mitochondria
can alter shape, increase or decrease in number, exchange the genome and change
the subcellular distribution. The total number of mitochondria and their morphology
are dependent on their properties of undergoing fusion and fission events, and
multiple proteins have been shown to be involved in the remodeling of mitochondrial
membranes. Fusion activity leads to extended networks of mitochondria whereas
fission results in small spherically shaped organelles; crucial processes to maintain
the physiology of healthy cells. It has been suggested that fission is important for the
mitochondrial proliferation into different cells, whereas fusion is responsible for
communication between mitochondria, important e.g. for movements of the
organelle 42. For instance, lymphocytes have been shown to be dependent on
fusion/fission events for their migration 43, and in order to meet the high energy
demand in synapses mitochondria have been proposed to redistribute into these
regions 42.
Fission has recently been suggested to be a prerequisite step in apoptosis, based on
the observation that mitochondrial fragmentation occurs very close in time to the
apoptosis-induced release of cytochrome c from the mitochondria 44. Though, this
Introduction
7
hypothesis has been questioned by studies which showed an incomplete or no
induction at all of cytochrome c release or cell death upon mitochondrial
fragmentation 45; 46. Rather than being responsible for the execution of apoptosis,
mitochondria fragmentation may instead be a phenomenon that accompanies the
apoptotic process 42.
4.1.2.2 Phospholipids
Biological membranes, such as those enveloping the mitochondria, are composed
of a vast number of different lipids. In the past, membrane lipids were considered to
be static entities, building up a passive structural platform where various
physiological processes were executed. This view has changed drastically in recent
years, when numerous evidences support a highly dynamic and active role of these
entities 47; a concept not really surprising since lipids constitute the chemically most
diverse class of biomolecules, creating “Lipidomics”. Based on lipidomics analyses an
average eukaryotic cell can contain up to tens of thousands of different species, with
distinct cell type (or organelle)-dependent lipid compositions in order to carry out
specific functions 48. The unifying criterion for all lipids is their solubility in organic
solvents due to their hydrophobic molecular features, and they are classified into
numerous groups such as phospholipids, sterols and waxes, differing widely in
physicochemical behavior. Due to the sheer amount of data and the complexity of
lipids and their uncountable variety, a detailed description of them is out of the scope
of this thesis. Since the most commonly found membrane lipids belong to the class of
phospholipids, the focus here will therefore be on those species. The majority of
phospholipids, the glycerophospholipids, have a diglyceride component linked to a
phosphate-containing group. They can be neutral or negatively charged, depending
on the nature of the phosphate moiety. Glycerophospholipids are derived from sn-
glycero-3-phosphoric acid, usually having two fatty acids coupled to the glycerol
scaffold via ester bonds at the sn-1 and -2 positions. Due to their amphiphilic nature,
they can spontaneously form bilayers in water where the hydrophilic phosphate
moiety is pointed outwards to the aqueous environment and the hydrophobic fatty
acid chains in each leaflet are oriented inwards to face each other; the basic
organization of biological membranes (Fig. 4). Since the interactions between
individual lipids are non-covalent and weak, they can move rather freely along the
bilayer plane, which allows for rapid rotational and lateral diffusion in addition to
high conformational flexibility due to the trans-gauche bond rotations of the acyl
chains 49. This high degree of freedom of lipids does not necessarily mean they are
merely randomly distributed and aimlessly changing positions. Instead, membranes
exhibit function-related heterogeneity, manifested as lipid asymmetry and diversity
in their lateral organization. For instance, PS normally resides solely in the inner
leaflet of the plasma membrane, though this asymmetry between both leaflets is lost
during apoptosis when PS lipids become translocated to the outer leaflet to act as a
recognition signal for engulfment 9. Since it is now clear that biomembranes have a
Introduction
8
high abundance of proteins (the lipid-to-protein weight ratio of e.g. the plasma
membrane is approximately 1:1), the lateral heterogeneity is a consequence of a
charge or hydrophobicity related attraction between lipids as well as between lipids
and integral membrane proteins or peripheral proteins 50; 51; 52. The lateral
heterogeneity creates a very dynamic, fluctuating biological environment, where
clusters of preferred lipids and proteins are interplaying on different time- and length
scales, forming microdomains (or rafts) within the biomembranes where specific
processes take place.
Figure 4. Diagrammatic presentation of the characteristic features of biomembranes. The fluid-like bilayer
houses many integral and peripheral proteins (red) as well as cholesterol molecules (yellow chains), which are
highly mobile in the dynamic lipid environment (courtesy of Vanessa Kunkel).
The abundance of different phospholipids varies greatly between different
biomembranes. The neutral phosphatidylcholine (PC) is the main species in all
mammalian cells, comprising 40-50% of the total amount of phospholipids 53. In
mitochondrial membranes PC and also the uncharged phosphatidylethanolamine
(PE) are the most commonly found, comprising ~40% and ~30% of the total number
of phospholipids, respectively. CL is another main membrane constituent and has
been proposed to be crucial for the execution of the mitochondrial apoptotic pathway
(see 4.1.3) 35. PC is a bilayer-forming phospholipid due to its cylindrical shape with
almost the same diameter of the hydrophobic headgroup as for the hydrophobic fatty
acid chain domain, enabling a favorable packing into a leaflet. In contrast, PE and CL
do not themselves form bilayers (but can be part of them) as a consequence of their
conical shape. They have a smaller headgroup relative to their fatty acid part and can
therefore inflict a negative curvature on the bilayer , giving rise to the formation of
hexagonal phases that can create tensions in the membrane; presumably important
for various processes such as membrane fusion or protein transport 54.
Introduction
9
In the papers included in this thesis, biological membrane-mimicking liposomes
were used which were composed of different glycerophospholipids (Fig. 5); 1,2-
dimyristoyl-sn-glycero-3-phosphocholine (DMPC) (papers I and IV), 1-palmitoyl-2-
oleoyl-sn-glycero-3-phosphocholine (POPC) (paper II), 1-palmitoyl-2-oleoyl-sn-
glycero-3-phosphoethanolamine (POPE) (paper II) and 1,1',2,2'-tetra-myristoyl-
cardiolipin (TMCL) (paper II).
Figure 5. The chemical structures of DMPC, POPC, POPE and TMCL (Avanti Polar Lipids, Inc).
When these lipids are hydrated they form polymorphic structures that vary with
temperature, pH, ionic strength and constituting lipid species. At low temperatures,
membranes often exist in an ordered lamellar gel-phase (L’ for many PC lipids; Lfor
other species); see Fig. 6. As the temperature is increased L will convert into the
more disordered lamellar liquid crystalline-phase (L). The transition between L and
L occurs at specific temperatures for different lipids and lipid compositions. Also, for
many PC lipids (and some lipid mixtures containing PC) there is a distinct
pretransition occurring between the L’ and the so-called ripple-phase (P’), prior to
the main transition. Defined systems display characteristic melting event profiles
(thermograms), which can be analyzed to provide information on how the
organization of the membrane changes upon incorporation of specific lipids or in the
presence of e.g. proteins.
POPE
TMCL
POPC
DMPC
Introduction
10
Figure 6. Typical thermogram displaying the phase transitions for a simple membrane system. When heat is
transferred to the membrane sample it undergoes two phase transitions. For many membranes containing
pure PC lipids (as well as some PC membranes containing additional lipids species), a pretransition occurs at
low temperatures, shifting the phase of the membrane from the gel-phase (L’) to the ripple-phase (P’). At
higher temperatures the main transition takes place, shifting the phase from P’ to L. Though, for many
membrane compositions no ripple-phase occurs, hence giving rise to only a single transition (L to L).
4.1.2.3 Oxidized phospholipids (OxPls)
In recent years, interest has been growing in understanding the molecular impact
of oxidative stress on biological membranes and any subsequent physiological
consequences. During oxidative stress, (poly)unsaturated diacyl- and alk(en)ylacyl
glycerophospholipids are highly prone to react with ROS to form numerous OxPls,
including phospholipids that bear aldehyde or carboxyl moieties at their truncated
sn-2 acyl chain. The oxidation may be catalyzed by enzymes, though a considerable
amount is presumably formed by non-enzymatic reactions in vivo 53. Since PC is the
most commonly found phospholipid class in mammalian cells, the majority of OxPls
contain the choline moiety. When situated in biological membranes, the polar moiety
leads to the reversal of the sn-2 acyl chain due to the energy penalty of embedding a
Introduction
11
charged group in the hydrophobic membrane core 55; 56. As a consequence of this
extended conformation of oxidatively modified chains, OxPls have a profound effect
on the membranes’ structure and dynamics. It has been reported that the
reorientation of the sn-2 acyl chain leads to an increase in the average area per lipid
corresponding to a reduction in the bilayer thickness 57. Oxidative insult can also
drive self-association and expelling of cholesterol from OxPl-rich regions, due to the
reduction in the bilayer thickness, and induce the formation of liquid ordered phase
enriched domains 58. Furthermore, OxPls enhance the bilayer penetration of water
molecules 57; 59, ensuing an increased polarity of the interior of the bilayer. This effect
was demonstrated to greatly facilitate lipid transfer between both leaflets (flip-flop)
independent of any scramblase 59.
OxPls have been shown to possess important physiological functions under
oxidative stress conditions; functions which range from involvement in aging and
anti-inflammatory processes to disorders such as atherosclerosis, Alzheimer’s disease
and diabetes 60. Since oxidative stress is one of the hallmarks of apoptosis 61, 62, OxPls
may therefore play an important role in the onset and progression of programmed
cell death. For instance, NADPH oxidase-induced apoptosis was shown to be
associated with the oxidation of PC, PS and PE 63, and it has also been suggested that
oxidatively modified PS is involved in the recognition of apoptotic cells during
phagocytosis 64. In addition, it was found that OxPls could activate
sphingomyelinases, presumably leading to the formation of ceramides that
subsequently can induce MOMP and apoptosis 65, 66. As previously mentioned,
mitochondria are believed to be one of the major sources of ROS. Therefore, a
considerable oxidation of mitochondrial phospholipids can occur, causing locally
high concentrations of OxPls that potentially are able to trigger the onset of the
intrinsic apoptotic pathway. Indeed, OxPls have been shown to rapidly associate to
mitochondria, giving rise to release of cytochrome c and induction of the intrinsic
pathway 67. Also, overexpression of the anti-apoptotic Bcl-2 protein Bcl-XL was able
to protect cells and isolated mitochondria from the effect of OxPls 68, further
buttressing an important role of oxidatively truncated phospholipids in
mitochondrial apoptosis ( see 6.1.3). Nevertheless to understand how OxPls are able
to induce apoptosis and other cellular processes including pathologies, a molecular
insight into their modulation of membranes’ structure and dynamics, is essential.
In this thesis, the POPC-derived aldehyde-bearing 1-palmitoyl-2-(9´-oxononan-
oyl)-sn-glycero-3-phosphocholine (PoxnoPC) and the carboxyl group-containing 1-
palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine (PazePC) were used to study the
impact of OxPls on membrane dynamics and organization (paper I); see Fig. 7 for
structures. PazePC was further used for studying the effect of OxPls on the interplay
of pro-apoptotic Bcl-2-associated X protein (Bax) with mitochondrial-mimicking
membranes (paper II).
Introduction
12
Figure 7. Chemical structures of the POPC-derived OxPls PoxnoPC and PazePC (Avanti Polar Lipids, Inc).
4.1.3 Regulation of the mitochondrial apoptosis by the Bcl-2
protein family
Governed by the Bcl-2 family proteins (Fig. 8), the mitochondrial apoptosis occurs
via the execution of MOMP. This event enables the release of apoptogenic factors and
subsequently causes cellular death. The detailed mechanisms underlying the
progression of MOMP are still not fully understood, but it is believed that the main
culprits are the multidomain effector proteins, Bax and Bak. Both of them are capable
of forming homo-oligomers in the MOM that act as pores for the exportation of
apoptogenic proteins to the cytosol 22. They belong to the pro-apoptotic branch of the
Bcl-2 family and are activated by the action of their fellow pro-apoptotic members,
the BH3-only proteins. The pro-apoptotic proteins are restrained by the group of pro-
survival (anti-apoptotic) multidomain Bcl-2 proteins, including the founding member
Bcl-2, thereby inhibiting MOMP. So far more than 25 members of the bcl-2 gene
family have been discovered in humans 69; 70, most of them expressing membrane-
associated proteins.
The secondary structures of the anti-apoptotic Bcl-2 proteins and of the pro-
apoptotic proteins Bax and Bak are predicted to be very similar, being mainly -
helical 71. In fact, the 3D-structures that have been determined for seven members of
the Bcl-2 family (including the BH3-only protein Bid) share a striking similarity to
each other 72; 73; 74; 75; 76; 77; 78; 79; 80, although having opposing functions, as well as to
pore forming bacteria colicins and diphtheria toxin 81. They also share a high degree
Introduction
13
of sequence homology through their conserved BH1-4 domains. Most of the anti-
apoptotic Bcl-2 proteins contain all four conserved domains, whereas Bax and Bak
traditionally have been considered to harbor three of them. Though, a recent
structure-based alignment study has suggested the existence of a fourth conserved
motif for the two effector proteins 82. The more diverse group of BH3-only proteins,
with predicted structures unrelated to the effector and pro-survival Bcl-2 proteins
(with one exception, Bid) 71, only share homology by their single BH3-domain. This
domain is believed to be crucial for the binding interactions between the Bcl-2
proteins, protruding from one family member into the hydrophobic binding pocket of
another one 83; 84.
Figure 8. Overview of the Bcl-2 family, divided into the three subclasses of anti-apoptotic (multidomain) and
pro-apoptotic proteins (multidomains and BH3-only). The transmembrane (TM)- and conserved BH-regions
are denoted for each subclass (adapted with permission from Simpsons et al., 2008 85).
Upon a stress signal Bax and Bak become activated for executing the
permeabilization of the MOMP. In contrast to Bak, which is an integral membrane
protein located in the MOM, monomeric Bax translocates from the cytosol to the
mitochondria. Following insertion into the bilayer, the membrane-integrated Bax
(and Bak) forms oligomers that enable the leakage of apoptogenic factors from the
IMS. These include proteins such as cytochrome c 86, endonuclease G 87, AIF 88,
HtrA2/OMI 89 and SMAC (or DIABLO) 90; 91. Following its release into the cytosol
cytochrome c binds to Apaf-1 23, promoting dATP-dependent conformational changes
in the latter to oligomerize and finally form the heptameric wheel-like structure of the
apoptosome 24; 25; 92. When assembled, the apoptosome can recruit pro-caspase-9,
inducing its dimerization and activation 24, leading to downstream caspase activation.
Introduction
14
Endonuclease G is involved in the caspase-independent fragmentation of DNA 93, the
same as AIF, which in addition is assisting in chromatin condensation 88.
SMAC/DIABLO and HtrA2/OMI are both relieving the cell from the inhibitory action
of XIAP 89; 90; 91, a protein that neutralizes the activity of caspase-9, caspase-3 and
caspase-7 94.
The pro-apoptotic activities of Bax and Bak are clearly controlled by the anti-
apoptotic Bcl-2 proteins and the BH3-only proteins. Though, the precise mechanism
that underlies this regulation of the effector proteins by the other Bcl-2 family
members is still a mystery. Two competing models have been proposed suggesting
either a direct or indirect activation of Bax and Bak (see Fig. 9). In both of these
models, the BH3-only proteins clearly induce MOMP by acting upstream of the
effector proteins since cells lacking both Bax and Bak cannot undergo apoptosis 95; 96.
Figure 9. Different models of Bax (and Bak) activation. The direct activation model (A) suggests that Bim
and tBid (activators) can directly engage and activate Bax. Bcl-2-like proteins sequester tBid and Bim (not
Bax) but can in turn be inhibited by sensitizing BH3-only proteins. In the indirect activation model (B) Bcl-2-
like proteins are continuously sequestering the continuatively active Bax. BH3-only proteins displace Bax,
having different selectivity for different Bcl-2-like proteins (adapted with permission from Adams and Cory,
2007 97).
In the direct activation model, the BH3-only proteins can function either as
activators (tBid, Bim and maybe Puma) or sensitizers/derepressors (the remaining
BH3-only proteins). The activators can directly bind to and induce the pro-apoptotic
functions of Bax and Bak. Anti-apoptotic proteins can prevent this activation by
Sensitizer
(derepressor)
Activator PromiscuousSelective
Direct activation model Indirect activation modelA B
Introduction
15
sequestering the activators, but can in turn be bound by the sensitizers which will
then lead to the release of the activating BH3-only proteins, ready for engagement
with the effector proteins. According to the indirect activation or displacement model
Bax and Bak exist in an active state, independent of the activity of any BH3-only
protein. For securing the survival of the cell the pro-survival Bcl-2 members have to
continuatively bind to both of them. To trigger Bax and Bak-induced MOMP, the
BH3-only proteins engage the anti-apoptotic proteins, thereby displacing the already
activated effector proteins. Most BH3-only proteins display selectivity for a few
specific anti-apoptotic Bcl-2 proteins, whereas tBid and Bim are able to bind several
targets.
Plenty of studies have provided evidences for each of the models, leading to a lack
of consensus 97; 98. The reason for this discrepancy is that most of these studies have
some inherent flaws compared to the real in vivo situation. Biophysical
measurements to study interactions between different Bcl-2 family members have in
many cases involved peptides or truncated proteins. Also, when cell cultures or
tissues were used, the binding (or the lack thereof) was confirmed by co-
immunoprecipitation; a method that is not able to detect weak, transient protein-
protein interactions. Most importantly, the biophysical studies were carried out in
solution, due to the cumbersome task of working with full-length membrane proteins
in their native environment. Considering that monomeric Bax is more or less
cytosolic 99; 100 in contrast to its anti-apoptotic antagonist Bcl-2, which is an integral
membrane protein 101; 102, the real-life scenario cannot be achieved in the absence of
lipid bilayers. Andrew’s group therefore proposed an alternative regulative mode
concept, the embedded together model, which includes the mitochondrial membrane
as an active partner in the regulation of the Bcl-2 family proteins 98, 103. In addition to
the membrane part, this model combines elements from the direct and indirect
activation modes; Bax (and Bak) can be both directly activated or displaced by BH3-
only proteins, as well as sequestered by anti-apoptotic Bcl-2 proteins. According to
this model, membrane binding induces conformational changes in the Bcl-2 family
proteins, enabling protein-protein interactions. The embedded together model has
been supported by various reports, e.g. observations have shown the inability of tBid
to interact with Bax 104 (or anti-apoptotic Bcl-XL 105) in solution, while being able to
induce pronounced Bax permeabilization activity in the presence of membranes 104;
106; 107. These and several similar observations have led to the conclusion that the
mitochondrion itself is essential in the activation of the intrinsic apoptotic
machinery.
In this thesis the focus is on Bax and Bcl-2, whose activities function as a major
on/off-switch for the mitochondrial apoptosis. These two antagonizing proteins will
therefore be more closely investigated in the next two sections.
Introduction
16
4.1.3.1 The activation and pore-forming action of Bax
Under normal cellular conditions Bax is predominantly found in the cytosol 99; 100.
Following its activation upon apoptotic stress, Bax undergoes a series of
conformational changes that ultimately lead to pore formation in the MOM. The
conformational transformation steps necessary to generate Bax in its active state
have been thoroughly investigated, though their order of occurrence is still not clear.
An early event during Bax activation is the translocation of the protein to the MOM.
This action has been suggested to be dependent on the N-terminal region (including
helix 1), whose flexibility can expose an adjacent mitochondria-targeting sequence.
Residues 13-19 of the N-terminal domain seem to be involved in the mitochondria-
targeting 108, as these become exposed and thereby able to bind to a specific antibody,
6A7, only when Bax is activated 108; 109. In addition, the N-terminus has been shown
to associate with mitochondrial membrane 110; 111. The C-terminal hydrophobic,
transmembrane helix 9 has also been implicated in the targeting of Bax, though it
might primarily be important for the insertion of the protein into the MOM prior to
its oligomerization 112. Helix 9 is normally sequestered in a hydrophobic surface
groove but becomes exposed, presumably displaced by the BH3-only proteins tBid or
Bim, which transiently bind to the groove via their BH3-domain. Alternatively, tBid
or Bim may bind to a rear pocket of Bax, thereby inducing N-terminal exposure
followed by the release of helix 9 to facilitate insertion. When Bax is anchored in the
membrane it may interact with another Bax molecule to form homo-dimers and
subsequently homo-oligomers 113, which are crucial for pore formation 114. For some
time the homo-dimerization interface has been postulated to be constituted of the
hydrophobic surface groove of and the BH3-domain, the latter corresponding to helix
2. In fact, very recent studies have provided evidence for the formation of
symmetrical (reciprocal) BH3:groove Bax dimers both in liposomes and within the
MOM 107; 115. These homo-dimers are then linked via a second interface, helix 6:helix
6, to form the higher order oligomeric complexes 115. Helix 6 as well as helix 5 have
also been shown to insert into the MOM during apoptosis 116, and in addition be able
to form pores on their own 117. Whether they are inserted into the MOM before or
after Bax oligomerization remains to be clarified.
As already mentioned, increasing evidence has recently emerged supporting a
crucial role of the mitochondrial membranes and their lipids in the regulation of
MOMP 103; 118. In particular, the mitochondria-specific CL has been reported to be a
prerequisite for Bax pore-forming activation. It is fairly abundant in the IMM with a
small portion residing in the MOM, while the largest population is found in the CS
that interconnect the two membranes 37. CL may be crucial for the stabilization of
these CS due to its ability to form a hexagonal inverted HII membrane-phase 119. In
the IMM, most of CL is tightly associated to cytochrome c, modulating its binding to
the outer leaflet of the membrane 120; 121. During early stages of the intrinsic
apoptosis, cytochrome c has been shown to oxidize CL due to its peroxidase activity
and the abundance of ROS generated from the electron transport chain 122; 123. This
Introduction
17
oxidation of CL (forming e.g. hydroperoxide CL species) is believed to facilitate the
dissociation of cytochrome c from the IMM and in addition induce the transfer of CL
to the outer leaflet of the MOM 124; 125; 126. Apparently, CL does not per se activate Bax
but instead seems to act as a mitochondrial platform. There, Bid can dock into and
subsequently undergo cleavage by caspase-8 to yield activated tBid 127; 128, leading to
recruitment and activation of Bax. However, Bax has been shown to expose the 6A7-
epitope in the presence of both charged and uncharged liposomes independent of
tBid or CL, though activation was not complete since the protein was merely able to
bind weakly to the membranes 129. Several studies have investigated the impact of
tBid and CL on Bax activation. In one of these the absence or degradation of CL
precluded insertion of the protein into liposomes or isolated mitochondria
respectively 130, even in the presence of tBid. Several other studies support the
combined role of tBid and CL for triggering Bax membrane permeabilization activity 104; 107; 129; 131; 132. Hence, for Bax to become fully activated for the induction of MOMP,
the presence of both CL and tBid appears to be necessary. One has to be aware of that
tBid+CL-induced activation of Bax occurs as a consequence of cleavage of Bid by
caspases, which is coupled to death receptor-triggered apoptosis in type II cells. Since
cleavage of Bid has been reported to occur downstream of Apaf-1 by executioner
caspases 133, other BH3-only proteins may play a more dominant role, at least
initially, for the direct activation of Bax during stress conditions that strictly triggers
the intrinsic pathway. Bim has been demonstrated to directly engage and activate Bax 134; 135, whereas the role of Puma is still under debate 135; 136. Still, tBid may be crucial
for the instigation of the intrinsic apoptosis independent of the executioner caspases,
as shown in a previously 137. There, an undefined aspartate protease was suggested to
cleave and activate Bid upstream of the mitochondria in response to a DNA-
damaging anti-cancer drug, leading to the execution of MOMP.
It is now generally established that Bax alone is capable of forming membrane
pores that are sufficiently large to release protein-sized molecules from liposomes
and mitochondria 104; 132; 138; 139; 140; 141; 142. Though, the nature of the Bax pore, such as
the number of protein monomers assembled and whether it exists as a proteinaceous
or lipidic pore, is still unclear. To further increase the complexity of the MOMP
process, Bax has besides its own pore forming ability, also been implicated in the
regulation of other channels involved in membrane permeabilization. Ceramide, an
apoptosis-mediating lipid, has been reported to concentrate in the mitochondria
during apoptosis and being able to form large channels that allow proteins to cross
the MOM 143; 144. Bax appears to bind and stabilize these ceramide pores and,
although presumably being able to induce poration of membranes itself, may act
synergistically with these ceramide channels to amplify the release of apoptogenic
factors 65; 145.
Introduction
18
4.1.3.2 Bcl-2 – the guardian of cell integrity
Originally the bcl-2 gene was discovered at the t(14;18) translocation break point in
human B-cell follicular lymphomas, leading to its increased expression by the
immunoglobulin heavy chain enhancer 146; 147. Surprisingly, Bcl-2 was found to
promote cell survival rather than proliferation by inhibiting apoptosis 148; 149, and
therefore constituted the first identified member of a new category of oncogenes that
regulated programmed cell death 150. Since its discovery Bcl-2 has been established as
a key protein in the intrinsic apoptotic pathway, acting as a major repressor of the
pro-apoptotic proteins’ activity. Also, the overexpression of Bcl-2 is strongly
correlated to the development of many cancer tumors and their resistance to
chemotherapeutic drugs 151, therefore constituting a crucial target in novel anti-
cancer drug research. Hence, the protein has gained a lot of attention in the last
decades with many groups aiming to reveal its structural and mechanistic features;
information essential for a thorough understanding of its key role in apoptotic
regulation as well as in tumorigenesis and cancer treatment resistance.
The Bcl-2 itself is an integral membrane protein 101; 102, continuatively located at
intracellular membranes in the nucleus, the endoplasmic reticulum and the
mitochondrion through its C-terminal transmembrane domain 152. In the MOM, Bcl-
2 is believed to interact with pro-apoptotic family members, indicated from studies
using peptides from BH3-only proteins. Presumably, it also forms hetero-dimers with
Bax, as evident from previous work such as an early co-immunoprecipitation
experiment 153, and a more recent study using a pull-down assay where Bax bound to
truncated His6-Bcl-2 (devoid of its transmembrane part) in solution 154. These Bcl-
2/Bax hetero-dimers have been proposed to constitute dead ends for Bax
oligomerization, thereby hindering the progress of MOMP 155.
As already mentioned in 4.1.3.1, Bax activation depends on tBid and probably also
Bim for the induction of its oligomerization and pore formation. Bcl-2 has also been
shown to form pores in liposomes 156, though the size of those being much smaller
compared to the Bax pore 157. Intriguingly, upon an apoptotic stimulus Bcl-2
undergoes a conformational change, leading to a multispanning structure where
helices 5 and 6 are presumably inserted into the membrane 155; 158. A Bim BH3
derived-peptide was also able to induce a similar conformational alteration 155.
Furthermore, tBid triggered this conformation alteration in liposome-tethered Bcl-2,
resulting in pore formation of the latter 157. This pore forming action was triggered by
tBid-activated Bax as well and Bcl-2 could only inhibit the pro-apoptotic protein in
the pore forming state 157. The pore-forming ability of Bcl-2 appears to be important
for its anti-apoptotic function. Presumably, the protein has to change its
conformation to be able to form multi-spanning oligomers that are capable to create
pores, triggered by e.g. tBid 158; 159, in order to stably interact with (and inhibit) active
membrane-integrated Bax 158. Since tBid also has been suggested to induce the
translocation of Bcl-XL to MOM and its integration into the membrane 160, BH3-only
proteins are apparently able to not only inhibit pro-survival Bcl-2 proteins but also to
Introduction
19
activate their functions. Interestingly, Bcl-2 could be converted to a pro-apoptotic cell
killing protein by the action of caspase-3 or the nuclear receptor Nur77, leading to the
removal of the BH4-domain and exposure of the BH3-region 161; 162. These
observations, together with the pore forming ability, have led to the notion that Bcl-2
may be seen as a defective Bax 98.
Despite the efforts of providing biophysical and biochemical information on Bcl-2,
the data gathered so far are limited. Although some basic biophysical properties have
been elucidated, the structural information obtained so far is only based on a
truncated Bcl-2/Bcl-XL chimeric protein and Bcl-2 devoid of its transmembrane
region studied in solution 76; 163, respectively. Biophysical data regarding the full-
length protein’s regulation of apoptosis in its native membrane environment is yet to
be provided.
Introduction
20
4.2 Aim of this thesis
During the last few decades apoptosis has been the focus of numerous research
efforts leading to a better understanding of this cell-suicide program. Therefore it is
now established that the Bcl-2 family of proteins play a key role in the regulation of
the mitochondrial apoptotic pathway. The pro- and anti-apoptotic members
belonging to this family have been extensively studied to obtain insights into their
structural and functional properties, which determine the fate of the cell. Despite
these efforts, the mechanisms underlying the complex apoptotic process still remain
elusive. The problem lies in the multiple steps preceding the demise of the cell. In
addition, the process of mitochondrial apoptosis is triggered by a huge number of
events, and is tightly connected to proteins carrying out their action at membranes,
many of these being integrated into the lipid bilayer. The work done so far has mostly
focused on Bcl-2 family proteins in solution, in many cases forcing the researchers to
work with truncated and/or otherwise mutated variants to maintain protein
solubility. Hence, the significance of the membrane environment has been largely
neglected.
Therefore, the aim of this thesis was to provide comprehensive and in-depth
information on 1) the role of membranes in the regulation of the mitochondrial
apoptotic pathway, and 2) the characteristics and behavior of the pro-apoptotic Bax
and the pro-survival Bcl-2 as full-length human proteins in the presence of a native
mitochondrial membrane-mimicking environment. In particular, we were interested
in i) how the properties of membranes change upon the exposure to a potential
apoptosis-triggering factor (incorporation of OxPls); ii) whether the pro-apoptotic
Bax protein can interact with these membranes in the absence of any Bax-activating
protein/factor; and iii) if it is possible to probe the putative Bax-Bcl-2 interaction
using biophysical approaches.
Methods
21
5 Methods
5.1 Differential scanning calorimetry (DSC)
Studying the phase behavior and phase changes of a membrane system provides
valuable insights into the organization and dynamics of these lipid assemblies, and
how they are modulated by the presence of membrane effectors, e.g. proteins. DSC
has been used extensively over the years to investigate the thermotropic events that
occur during the melting of lyotropic systems 164. For pure DMPC-vesicles the
ordered L’-phase, existing at low temperatures, will convert into the P’-phase as the
temperature is raised. As the temperature is even further increased the P’-phase will
transfer into the disordered Lphase, which reflects the physiologically important
liquid crystalline state of a membrane. By using DSC these thermal events can easily
be visualized and quantified (Fig. 10). The idea of the method is simple; the sample
containing the membranes is injected into one of two identical cells residing in an
adiabatic jacket. In the other cell the sample buffer, serving as a reference, is added.
When the cells are individually heated up, the cell with the membrane sample
requires additional heat compared to the reference cell in order to maintain the same
temperature during an ongoing phase transition. The difference in heat capacity
between the sample and reference cells are then plotted as a function of temperature,
which yields a characteristic heat profile for that particular membrane system. The
pre- and main transitions are shown as distinct peaks at the respective melting
temperature (Tm). The area under the peaks corresponds to the change in enthalpy
(H) for each transition and the cooperativity of these melting events is reflected by
the sharpness of the peaks.
Figure 10. (A) The DSC apparatus and (B) DSC principle. (C) When the difference in heat capacity between
the sample and the reference is plotted as a function of temperature the thermogram displays distinct peaks,
reflecting pre- and main transitions. The area under the peaks corresponds to the H of the transitions.
A B
C
Methods
22
To study the impact of OxPls on membrane organization and protein-membrane
interactions, DSC experiments were carried out in papers I and II. The experimental
setup was basically identical in both papers; to ensure that thermal equilibrium had
been reached three thermograms between 5 C and 45 C (two fast up- and down-
scans, followed by a slow up-scan) were recorded for the samples containing 3 mM
multilamellar vesicles (MLVs) (9:1 DMPC:PoxnoPC/PazePC) only. For the melting of
protein-containing samples, only one fast up-scan was performed due to
denaturation of the protein at higher temperatures. The focus in paper I was to gain a
fundamental understanding of how OxPls can affect membrane systems. Therefore,
the multi-component thermograms were deconvoluted and quantified to thoroughly
characterize the different membrane melting events occurring in the presence of
OxPls. In paper II, the aim was to investigate whether the presence of the OxPl
PazePC could enhance the association between membrane systems and Bax. Visual
comparisons were made for Bax-containing MLVs samples with and without PazePC,
as well as Bax samples in the presence and absence of PazePC-containing
membranes. For all Bax-membrane samples a 1:200 protein-to-lipid molar ratio was
used.
5.2 Nuclear magnetic resonance (NMR) spectroscopy
In the last half century NMR spectroscopy has emerged as an extremely powerful
and versatile technique, now used routinely in many areas ranging from basic natural
science to medicine. NMR is one of the most useful methods for studying biological
samples due to the wealth of molecular information to be generated. Knowledge
obtained this way includes the determination of protein structure, fold, stability,
ligand binding as well as the organization and dynamics of membrane systems 165; 166.
A great advantage of the technique are the mild, noninvasive conditions used during
measurements (using radio frequency pulses), which maintains the integrity of the
samples. NMR relies on the nuclear spin properties of the nuclei of interest (in
biological sciences mainly 1H, 31P, 13C and 15N), as defined by a quantum number I.
When the nuclei, which possess a nuclear magnetic moment, are exposed to an
external magnetic field (as in an NMR magnet) they will populate discrete energy
levels whose number is 2I+1 according to Boltzmann’s law. At equilibrium, the
population of nuclei in the lower energy levels slightly outnumbers the nuclei in the
upper levels. This population difference will cause net magnetization of the sample
with a macroscopic magnetic moment (visualized as a vector) oriented along the z-
axis of the magnetic field. This population difference does not only depend on the size
of the external magnetic field but also on the nucleus-specific gyromagnetic ratio,
which is e.g. for protons ca. four times higher than for a 13C nucleus. Following
excitation (in a direction perpendicular to the outer magnetic field) with a resonance
frequency, the so-called Larmor frequency which corresponds to the energy
difference between the lower and upper level (for a spin ½ system such as the nuclei
mentioned above), the spin system can be perturbed. This leads to simultaneous
Methods
23
absorption (from the low energy nuclei) and emission (from the high energy nuclei)
processes. Due to the difference in population the macroscopic magnetization vector
of the sample will move away from the z-direction and instead undergo precession
around the z-axis with the nucleus specific Larmor frequency. The angle of the vector
in relation to the z-axis depends on the length of the pulse. Most commonly in NMR
experiments a 90 radio frequency pulse is applied, which causes the magnetization
vector to tilt into the x-y plane where also the NMR detection coil is situated (Fig. 11).
There, the rotation magnetization vector induces a current in the detection coil. This
time-dependent signal is called the free induction decay (FID) and its intensity is
reduced due to relaxation processes (see below). The NMR spectrum is being
obtained by processing the time-domain signal using a Fourier transformation. This
process converts the time-domain signal into a frequency-domain which is a
spectrum where intensities are plotted as a function of frequency. Since the
Boltzmann’s distribution is proportional to the strength of the applied external
magnetic field and the electric noise in NMR detection systems proportional to the
temperature, the use of stronger NMR magnets and cryoprobes leads to a large
enhancement of the NMR signal.
Figure 11. A radio frequency 90 pulse (matching the Larmor frequency) excites the nuclei in the sample,
leading to a tip down of the net magnetization vector from the z-axis to the x-y-plane. The vector precesses
around the z-axis at the Larmor frequency and induces an electric current in the detection coil (in the x-y-
plane), thereby giving rise to the NMR signal.
When the nuclei are perturbed and shifted towards the x-y-plane by the external
radio frequency pulse, the magnetization vector is then precessing around the z-axis
in the x-y-plane at the Larmor frequency until it is relaxed back to equilibrium. The
relaxation process is dependent on two individual parameters; the longitudinal and
transversal relaxation times, T1 and T2 respectively. T1 relaxation corresponds to the
Z
YX
90
pulse
External
magnetic
field
Methods
24
return of the net magnetization along the z-axis and is a reporter of fast vibrational
and translational motions on the ps-ns timescale, leading to the transfer of energy
from the nucleus to its surrounding (including the molecule containing the nuclei
itself). This process is dependent on the molecular structure and size as well as the
molecular environment, e.g. temperature and viscosity. T2 relaxation reflects the loss
of magnetization in the x-y-plane and arises due to energy transfer between nuclei in
a particular spin-system. It is related to slower dynamics, up to milliseconds, and is
affected by molecular interactions and inhomogeneities in the external magnetic
field. The linewidth of the NMR signal is inversely proportional to the T2. During
isotropic conditions T1 and T2 are more or less equal, whereas T2 drops dramatically
with increasing anisotropy (e.g. larger molecules, viscosity) thereby contributing to a
broadening of the signal.
The nuclei’s Larmor frequencies are dependent on their specific gyromagnetic ratio
and the applied external magnetic field. However, these frequencies vary also slightly
between nuclei of the same kind if these nuclei are located in different molecular
groups. Due to the differences in the electronic environment in a molecule (different
chemical bonds etc), these nuclei precess slightly different in a magnetic field –
essential for the success of NMR - and provide therefore information about molecular
structures. The electronic density around the nuclei generates locally induced
magnetic fields, which have a shielding (opposing) effect on the external magnetic
field. As a consequence, the external magnetic field becomes weaker and the
resonance frequency is reduced. In other words, for a static external magnetic field,
low electron density leads to a less weakened frequency to achieve a signal, whereas a
high density yields a signal at lower frequency. These specific resonance frequencies
of particular nuclei are called chemical shifts and they are normalized against a
reference (e.g. tetramethylsilane) and converted into ppm, thereby enabling easy
comparison between different external field strengths. It should be noted that the
locally induced magnetic fields are of several order of magnitudes lower than the
external field. Therefore, the difference between the same type of nuclei in different
chemical environment will only be a few Hz compared to the typical Larmor
frequency of 400-900 MHz. Chemical shifts provide neat fingerprints, enabling e.g.
the study of the structure but also the dynamics of biological macromolecules such as
membranes.
5.2.1 Solid state NMR
In general, the shielding of the external magnetic field is not constant in all
directions. Therefore, the chemical shift is dependent on the orientation of a molecule
in the magnetic field, hence it is anisotropic. Besides this chemical shift anisotropy
(CSA), dipolar interactions between two nuclei with spin ½ (e.g. 1H and 31P) also
contribute to anisotropic conditions. In liquid state NMR, the chemical shifts are
displayed as narrow and sharp peaks due to rapid molecular motion, leading to
isotropic conditions with no net molecular orientation and an effective averaging of
Methods
25
the CSA towards an isotropic value, while e.g. dipolar interactions are even averaged
to zero. This averaging is sufficient for molecules up to approximately 30 kDa, while
larger molecules will have slower tumbling rates and become more anisotropic. As a
consequence, the NMR signal will be significantly broader and resolution will be lost.
Therefore, to obtain information for large systems solid state NMR approaches have
to be used.
5.2.1.1 31P NMR
By using the inherent phosphorus nucleus as a natural NMR probe in
phospholipids, 31P NMR has shown to be a great method for studying membrane
structure and dynamics. The 31P signal of membranes is anisotropic due to slow
motions of the large systems. It is dominated by the CSA and reports on the
headgroup region of the phospholipids, reflecting the membrane surface properties,
and varies depending on the environment and behavior of the phospholipid species.
Since the phospholipids have rather fixed orientations depending on where they are
located in the system of interest (e.g. a liposome), the overlap of lineshapes arising
from different orientations results in a so-called NMR powder pattern (Fig. 12). Most
of the phospholipids are oriented perpendicular to the external magnetic field,
creating a low frequency peak at , while those that are aligned in parallel are shown
as a high frequency shoulder at . CSA is determined by the difference between
and (= - ), and can be used as an indicator of headgroup motions in the
membrane. When magic angle spinning (MAS) conditions (described in detail in
5.2.1.3) are applied to a membrane sample to resolve the anisotropy, the powder
pattern is replaced by the isotropic chemical shifts of each lipid species. The chemical
shift of the peaks can provide information regarding the electrostatic environment of
the membrane (see shielding in 5.2.1) whereas the linewidth reflects T2 relaxation
and slow motions of the lipids in the membrane.
Figure 12. Spectrum of a typical powder pattern of a lipid sample, arising due to anisotropic contributions.
CSA is determined from the difference between the high frequency σ and low frequency σ chemical shift
peaks.
020 -2040 -40
ppm
Methods
26
5.2.1.2 2H NMR
2H NMR is another valuable tool for assessing the state of membranes. Since the
2H nucleus has a spin of 1 it possesses an electric quadrupolar moment (holds true for
spins 1) that arises from an asymmetrical charge distribution in the nucleus. This
will lead to interactions with the electrons that generate the locally induced magnetic
field. In an isotropic sample these interactions will average out to zero, though during
anisotropic conditions the quadrupolar interactions will dominate the line shape of
the 2H spectrum, giving rise to the so-called (often symmetric-like) Pake pattern. The
shape and the size of the quadrupolar splitting in NMR spectra of deuterated water
molecules, which hydrate e.g. lipid membranes, are dependent on the local
anisotropic orientation, diffusion and the population difference between the various
water pools 167. The detected signal is the sum of the signals arising from these
different water fractions. The NMR spectra provide not only information on lipid
headgroup hydration but also on the phase of the lipid bilayer at a given temperature.
5.2.1.3 MAS
To reveal more information concealed in the anisotropic spectra of solid state
samples like membrane systems, the anisotropic conditions must be removed to
obtained well-resolved NMR spectra. In most cases the broad NMR lines are caused
by the CSA and dipolar interactions, as mentioned earlier. The size of both
interactions – and the NMR specific resonance - depend crucially on the angle () of
the molecule (or more precise, the CSA tensor and dipole-dipole interaction tensor,
respectively) with respect to the external magnetic field. At certain angles, these
anisotropic interactions will be cancelled out according to the second degree
Legendre polynomial (3cos2 -1)/2, and by orienting the molecules at such an angle
the contribution of the dipolar interactions to the anisotropy is eliminated. In MAS
NMR, the sample is tilted by an angle of 54.74, called the magic angle. To achieve
isotropic conditions, the sample is then spun at thousands of revolutions per second,
leading to a net orientation of the interaction tensors at the magic angle. In the case
of CSA the anisotropic chemical shifts are reduced to an isotropic value, whereas the
dipolar interactions are cancelled completely (in the case of protons often supported
by proton decoupling). This will greatly increase the resolution of the NMR spectra,
providing sharper, separated peaks which can be more easily assigned. Importantly,
the sample has to been spun at a frequency that corresponds at least to the chemical
shift anisotropy (specific for different nuclei) in order to generate the isotropic
signals. Both 31P and 1H can be used as probes for MAS NMR applications on
membrane samples. The resolved peaks in a 31P MAS spectrum contain information
on individual lipid species and their response on e.g. presence of proteins. The
change in linewidth of the NMR signals reflects restriction or promotion of molecular
movements in the headgroup region. 1H MAS NMR is useful for the investigation of
Methods
27
not only the dynamics of the lipid headgroups but also of the acyl chains, thereby
providing information on the hydrophobic core of the bilayer.
Solid state NMR was a crucial tool in papers I and II. In paper I 31P and 2H NMR
were used as complementary methods to the DSC measurements. MLVs consisting of
9:1 DMPC:PazePC/PoxnoPC, with a molar water to lipid ratio of 60:1, were prepared
for the 31P NMR study to monitor the impact of OxPls on the phases as a function of
temperature. 2H NMR samples, with a hydration level of 12-15:1 mol D2O/lipid molar
ratio, were investigated to deduce the modulation of the effect of OxPls on membrane
hydration and penetration of water molecules into the acyl region. In paper II, the
PazePC-containing MLVs samples were investigated using 1H and 31P MAS NMR, in
the presence and absence of Bax. The 31P MAS spectrum was studied to observe
specific interactions at the hydrophilic headgroup-protein interface, while 1H MAS
NMR was used to reveal interactions in both the headgroup and acyl chain regions,
reflecting binding and insertion, respectively.
5.3 Circular dichroism (CD) spectroscopy
CD spectroscopy is a very useful technique for probing the structural changes in
proteins following different stimuli or perturbations 168. By measuring the difference
in absorption between left- and right-handed circular polarized light, due to the
chiral nature of proteins, the structural elements in the molecules can be quantified.
The observed signal is proportional to the protein concentration according to
Lambert-Beer law. The protein chromophores that give rise to the CD signal are the
peptide bonds, which absorb below 250 nm (far-UV region), as well as aromatic side
chains and disulphide bonds, absorbing from 260 nm and above (near-UV region).
The far-UV region provides information about the secondary structure content in a
protein, where -helices, -sheets, turns and random coils exhibit characteristic
signals (see Fig. 13). Many valuable data can be extracted from the far-UV spectrum,
e.g. it is possible to gain a hint of whether recombinantly expressed proteins are
properly folded, probe conformational changes upon addition of ligands, and deduce
protein thermodynamics by chemical and thermal denaturation.
The signal in the near-UV spectral region, on the other hand, depends on the
number of aromatic residues (and S-S bonds) and their mobility, spatial distribution
and chemical environment, thereby providing a fingerprint of the tertiary structure of
the protein. This fingerprint spectrum contains important information regarding
structural differences upon a stimulus/perturbation which are not visible in the far-
UV spectra (showing the same secondary structure distribution).
Methods
28
Figure 13. Typical CD profiles of random coil and secondary structure elements as visible in the far-UV
region. The signal amplitude is proportional to the sample concentration (adapted with permission from
Czarnik-Matusewicz and Pilorz, 2006 169).
In papers II-IV far-UV CD was used as an important analytical tool. In paper II
the interplay between Bax and mitochondrial CS-mimicking small unilamellar
vesicles (SUVs) with and without PazePC was investigated by measuring the samples
between 200 and 260 nm, at 1:200 protein-to-lipid molar ratio. Spectra were
recorded at 10 C, 20 C and 37 C for Bax alone and in the presence of vesicles,
followed by deconvolution using the CDNN v2.1 software. For paper III CD was
utilized to confirm that the cell-free expressed Bcl-2 was folded properly in the
presence of 0.05% w/v Brij-58. To cover most of the -helical region the CD
spectrum was recorded between 190 and 260 nm. Bax and Bcl-2 were further
investigated in paper IV, where CD was used to study the putative interactions
between these proteins. Experiments were performed in the presence of Brij-35
detergent above (0.05% w/v) and below (0.0028% w/v) the critical micelle
concentration (CMC), to investigate whether the detergent micelles affected the
protein-protein interplay.
5.4 Cell-free protein synthesis
Although comprising approximately 25% of the total amount of proteins in both
eukaryotes and prokaryotes 170; 171, the study of membrane proteins has been greatly
hampered due to insufficient expression and purification strategies 172. Conventional
methods using host organisms (bacteria, yeast and insect cells) have proven to be too
expensive and/or have not been able to deliver any significant amounts of target
random coil
-turn-sheet-helix
Methods
29
protein. The difficulties of producing membrane proteins in significant amounts by
conventional overexpression are often caused by their inherent toxicity to the host
organisms, their limited solubility and their high aggregation propensity due to their
hydrophobic properties. These problems frequently obstruct their expression and/or
subsequent purification, the latter also true for the isolation of membrane proteins,
such as rhodopsin 173, from their natural sources. The host cell toxicity that arises
from heterologous expression, due to the overload of host cell translocon machinery
(preventing the production of crucial host cell membrane proteins), greatly reduces
the number of actively expressing cells and hence the amount of protein that can be
purified. Isolation of the initially modest amounts of a membrane protein in the cell
lysate is further obstructed by the aggregation into inclusion bodies, or cumbersome
extraction with detergents. Therefore, the yield of a target membrane protein is
mostly severely reduced 174. To solve the problems of expressing toxic and/or
hydrophobic proteins, cell-free protein synthesis is a suitable alternative 174; 175. This
method has recently emerged as a viable tool to successfully produce proteins for
which in vivo systems have failed. The strength with the cell-free system is that the
expression conditions can be optimized to a greater extent compared to living cells. A
range of different reagents and additives can be used, including detergents, isotopes,
lipids, cofactors and chaperones. In addition, the toxicity issue is more or less non-
existing since the cell-free system does not involve any living cells.
There are two types of cell-free protein synthesis, namely a continuous exchange-
mode and a batch-mode (Fig. 14). In continuous exchange cell-free expression of the
protein of interest is carried out in a chamber divided into two compartments,
denoted reaction and feeding chambers. The reaction chamber contains the DNA
source, extract (isolated e.g. from E. coli or wheat germ) and enzymes, while the
feeding chamber holds the nutrients (such as energy sources), amino acids and NTPs.
These two compartments are separated by a semi-permeable membrane which
enables the low molecular reagents to be transferred into the reaction chamber and,
simultaneously, shuttles inhibitory by-products produced from the expression
reactions into the feeding chamber. This continuous exchange is ensured by vigorous
shaking of the system. The expression reaction can proceed for a fairly long time, up
to 24 h due to the constant supply of nutrients, and may therefore result in rather
high protein yields. The second type, the batch expression mode, has a simpler setup.
Since all the reagents, DNA source, extract (providing the translational machinery)
and enzymes are added into the same vessel, there is no need for any advance system.
Also, the expression time is shorter (a few hours), the scaling easier, and the cost for
reagents lower compared to continuous exchange-mode, especially for isotope-
labeled amino acids for NMR purposes 176.
Methods
30
Figure 14. Principle of cell-free protein synthesis. Extract containing the translational machinery is isolated
form cell cultures (A), and is subsequently added to the reaction mixtures for either batch- (B) or continuous
exchange- (C) mode cell-free expression (adapted with permission from Carlson et al., 2012 175).
In papers III and IV an in-house cell-free expression system was used to produce
full-length human Bcl-2 for CD, fluorescence and reconstitution studies 177; 178. The
expression mixtures were prepared in 1.5 mL Eppendorf tubes and incubated at 800
rpm and 30 C for 2.5 h. To prevent protein aggregation, the hydrophobic Bcl-2 was
solubilized by adding 1% w/v Brij-58 (paper III) or 0.1% w/v Brij-35 (paper IV) into
the reaction mixtures.
Reaction chamber
Feeding chamber
by-products
nutrients
B
C
Extract preparation
2-3. Cell lysis and
centrifugation
1. Cell growth
4-5. Storage of extract after
removal of cell debris and
chromosomal DNA
A 1
2
3
4 5
- Energy substrates
- Amino acids
- Nucleotides
- Template
- Cofactors
- Salts
- Detergents
Milligrams of target protein
Results and discussion
31
6 Results and discussion
6.1 Influence of OxPls on membrane properties and Bax-membrane interactions
In the past, biophysical research on oxidatively stressed membranes has been
hampered due to the uncontrolled formation of numerous, badly characterized
oxidized lipid species. Only recently has commercially available OxPls enabled
thorough biophysical studies of defined OxPl-containing membrane systems 56; 59; 179.
Nevertheless, basic knowledge still remains scarce about the influence of these lipids
species on different membrane systems and their proteins, such as the mitochondrial
membranes and their Bcl-2 protein family. Therefore, to obtain a fundamental
understanding of the impact of OxPls on membranes as well as their potential role in
the mitochondrial apoptotic process, PazePC and PoxnoPC were incorporated into
phospholipid vesicles and studied by various biophysical techniques. Since both of
them are oxidation products of POPC, one of the most commonly occurring
phospholipids in human cells, they are presumably involved in physiologically
relevant processes.
A basic biophysical characterization of DMPC-based membranes was performed in
paper I, to study the impact of OxPls on membrane structural and dynamic
organization using DSC and solid state NMR. In paper II the study was extended to
involve Bax in order to elucidate the contribution of oxidatively stressed
mitochondrial membranes for the recruitment of the pro-apoptotic protein.
6.1.1 OxPls have a pronounced impact on membrane structure
and dynamics
It is now clear that OxPls exert a dramatic effect on membranes due to their polar
moieties at the end of their truncated sn-2 fatty acid chains. Since they are associated
with many pathologies as well as the apoptotic cell death program, a characterization
of their properties has been of great interest. Previous work has focused mainly on
simulation and fluorescence experiments, which have yielded plenty of valuable
information, e.g. the conclusion that OxPls reorient their sn-2 acyl chain towards the
aqueous surrounding. Nevertheless, in order to provide a deeper insight into the
change in membrane phase behavior and hydration properties additional techniques
have to be applied. Therefore, in paper I a combined DSC and solid state NMR
approach was carried out to elucidate the impact of OxPls on membrane properties.
There our findings might potentially explain their role in various biological events.
DSC provides a wealth of information on the phase behavior of lipid based
membranes, e.g. detection of the transition from the ordered gel-phase to the
disordered liquid crystalline-phase as the temperature increases. This information
provides valuable insights into the lipid bilayer organization and changes due to
Results and discussion
32
intrinsic (special lipids) or extrinsic (e.g. protein association). In paper I, pure
DMPC-membranes displayed a characteristic cooperative heat profile, with a
pretransition at 12.3 C and a main transition at 23.1 C. Incorporation of PoxnoPC or
PazePC into DMPC bilayers caused dramatic changes, even at low molar
concentrations; the cooperativity diminished, the pretransition vanished and the
main transition peak became considerably broader, showing a multi-component
profile (see Fig. 15). Deconvolution of the main transition peak revealed three
distinct thermal events. The general pattern of the main transition for both OxPls was
dominated by a broad component at lower temperatures, shifting to a much sharper
one upon increased temperature, followed by a broad component dominating at
elevated temperatures. This heterogeneous melting process was probably a
consequence of the presence of OxPl-poor and -rich populations, similar to what has
been shown for ergosterol-containing membranes, whose multi-component melting
features were attributed to ergosterol-rich and -poor regions 180. It was obvious that
the sharp, second component corresponded to an OxPl-poor domain since the Tm was
always close to the value of pure DMPC-vesicles, and in addition grew smaller as the
concentration of OxPls increased. The other two components displayed the opposite
behavior; both becoming larger and more prominent when increased amounts of
OxPls were present in the membrane, hence most likely reflecting OxPl-rich domains.
Figure 15. The change in heat profile as a consequence of OxPl-incorporation. Upper, pure DMPC-
liposomes; lower, incorporation of 10 mol% of PazePC. Solid and dotted lines correspond to thermograms of
samples in H2O- and D2O-buffer, respectively (adapted with permission from paper I ).
31P solid state NMR spectra of 10 mol% OxPls as a function of the temperature
mirrored the multi-domain transition as already visible in the corresponding DSC
traces. At lower temperatures the NMR spectra displayed very broad and not
completely axially symmetric powder lineshapes dominated by the CSA, typical for
membrane samples in the gel-phase. As the temperature approached levels
corresponding to the L’/L-transition, the NMR spectra showed a two-phase region,
reflecting the coexistence of two lamellar phases with axially symmetric powder
lineshape features. These two phases corresponded to the melting of the OxPl-poor
0% OxPls
10% PazePC
Results and discussion
33
domain and the second OxPl-rich region deduced from the DSC thermograms.
Analysis of the CSA-values for the OxPl-containing membranes provided some
surprising information. In contrast to pure DMPC-bilayers 181, the CSA increased with
temperature until the two-phase region was not visible anymore. Above this
temperature the lipid species seemed to be fully mixable in the L-phase. This is
surprising since the CSA would be expected to decrease upon a temperature increase
due to faster molecular motions and hence more averaging of the anisotropic signal 181. Hence, incorporation of PoxnoPC or PazePC clearly modulated the dynamics of
the headgroups in the membrane/water interface. The third thermal event observed
in the thermograms at low temperatures was not visible as an extra event in the
corresponding NMR-spectra. Although unexpected, this DSC visible event can
probably be attributed to a reversible solid-phase mixing/demixing process, as
reported earlier 180.
Since the presence of OxPls has previously been reported to increase water
permeability in membranes 179, the impact of OxPls on membrane hydration
properties was also addressed. This time DSC was used to study the phase behaviour
of PoxnoPC- and PazePC-vesicles upon hydration with D2O-buffer instead of ordinary
H2O-buffer. Compared to the previous thermograms where H2O-buffer was used, the
obtained thermograms looked very similar for both OxPls at all concentrations. The
main difference was a shift to higher temperatures, which can be accounted for by the
condensing effect of D2O on the bilayer, as indicated by MD simulations 182. This
phenomenon leads to a reduced surface area and more ordered acyl chain region,
thereby stabilizing the gel-phase (and ripple-phase). Since the effect was the same on
pure DMPC-vesicles as for those containing OxPls, the DSC results did not provide
any additional information on the change in membrane interface hydration as a
consequence of OxPl incorporation. Therefore, 2H NMR experiments were carried
out on mixed lipid vesicles with reduced hydration (12-15:1 water/lipid) to provide
any insight into the modulation of membrane hydration by the OxPl species. For pure
DMPC-vesicles two pools of water (one at the headgroup region, another in the
glycerol backbone) contributed to the lineshapes of the spectra 167. Due to the fast
exchange at elevated temperatures as a consequence of higher molecular tumbling
rate, these contributions were time-averaged, and gave rise to a single quadrupolar
splitting. Incorporation of PazePC resulted in a more isotropic, but still dynamically
broadened signal, apparently (at least visually) extinguishing the contributions from
the separate water pools. For PoxnoPC-membranes, the lineshapes looked quite
opposite; the peaks were broader at all temperatures (also compared to pure DMPC-
vesicles) and furthermore, at the Tm of the third melting event an additional
component appeared. Clearly, the presence of PazePC or PoxnoPC modulated the
hydration of the bilayer, either by eliminating or adding contributions of different
water pools.
Obviously, PazePC and PoxnoPC did not exhibit the same influence on the DMPC-
membrane. These dissimilarities arose due to the nature of the chain reversal of
respective OxPl; PazePC undergoes a complete sn-2 chain reversal, leading to a void
Results and discussion
34
space and compression of the bilayer, whereas the sn-2 chain of PoxnoPC intercalates
into neighboring lipid molecules. The lipid area of PoxnoPC is therefore larger
compared to the one of PazePC. For the DSC thermograms, the higher Tm for PazePC
compared to PoxnoPC was due to this smaller area per lipid, causing a more ordered
acyl chain region (similar to condensing effect due to D2O-hydration). In contrast, the
enthalpy was higher for PoxnoPC than PazePC. This Tm vs. enthalpy-discrepancy
appears to be puzzling at first, but can probably be explained in terms of structured
water. Since PoxnoPC has a larger area per lipid than PazePC this leads to a less
compressed membrane and therefore a total larger surface area. PoxnoPC-vesicles
can therefore accommodate more structured water, requiring more heat to complete
the melting events, analogous to what has been observed for insulin 183.
In summary, the presence of PazePC or PoxnoPC had a tremendous impact on both
the structural and the dynamical organization of membranes, causing lateral phase
separation into OxPl-rich and –poor microdomains and a complex hydration pattern.
These phenomena have significant influences on various biological processes, e.g.
apoptosis, as being discussed in the next two sections.
6.1.2 Modulation of Bax-membrane interactions by OxPls
Increasing amount of evidence points towards a significant and active role of the
mitochondrial membrane(s) in the activation of Bax to execute MOMP. Especially,
the presence of the mitochondria-specific lipid CL has been shown to be crucial for
the action of Bax at the MOM. CL, in combination with tBid, seems to promote the
recruitment of Bax to the MOM followed by subsequent pore formation.
Furthermore, oxidization of CL has been proposed to stimulate its translocation from
the IMM to the MOM, thereby enhancing the recruitment of tBid and hence the
activation of Bax 125. Under in vivo conditions, CL can presumably be oxidized by the
peroxidase activity of cytochrome c 122, or by ROS that arise in the highly oxidative
mitochondrial milieu. To explore if the presence of other oxidized lipid species can
influence the bilayer to accommodate Bax molecules, vesicles mimicking
mitochondrial CS (containing a lot of CL) with and without PazePC were prepared.
These lipid vesicles and their interplay with Bax were studied by DSC, CD and solid
state NMR spectroscopy. Importantly, this study was performed with Bax alone, i.e.
no other pro-apoptotic proteins such as tBid were present. Therefore, any occurring
membrane-protein interactions would be solely due to the dynamic mitochondrial
membrane environment. Since previous studies reported no or only weak association
of Bax in the absence of triggering factors 129; 131, the membranes without PazePC were
not expected to display any significant interactions with Bax. Evidently, little
happened to the membrane signal (DSC) when Bax was added and similarly the
protein signal (CD) was nearly unchanged upon addition of vesicles. Although
exposure of the 6A7-epitope (one of the hallmarks of Bax activation) occurred when
Bax was incubated with both charged and uncharged liposomes, this effect was
neither strong enough to alter the secondary structure in the protein nor to induce
Results and discussion
35
any persistent protein-membrane binding interaction 129. Also here in this study, the
DSC and CD data did not support any significant association of Bax to the
membranes in the absence of an external triggering event.
When PazePC-containing liposomes were used, a completely different scenario was
observed. The thermogram of pure membrane sample exhibited a dramatically
changed multi-component profile compared to the liposomes devoid of the OxPl. This
was not surprising considering the results in paper I, where PazePC induced similar
features on the DMPC-vesicles. Upon addition of Bax the melting event changed
considerably, displaying a more cooperative melting process which greatly abolished
the separated melting processes of individual microdomains (paper I). Furthermore,
the presence of PazePC-vesicles increased the thermostability of the protein, reflected
in a shift of Tm to a considerably higher temperature. Thus, the results from the DSC-
measurements indicated a significant association between Bax and vesicles, resulting
in both stabilization of the gel-phase of the membrane and higher Tm of the protein.
The question was then; did the PazePC-membranes inflict any detectable change in
the structure of the protein, potentially giving rise to the observed interaction? To
address this issue CD experiments were carried out at 20 °C. In contrast to the
previous measurements without the OxPl, there seemed to be an increase in -helical
content in the presence of PazePC-containing vesicles. Since the onset of the melting
event of these membranes was delayed upon addition of Bax (by ca. 3 C), compared
to the sample without PazePC, CD spectra were also recorded at 10 °C. Intriguingly,
the CD signal of Bax with PazePC-liposomes increased even further compared to the
sample at 20 °C, whereas the protein alone and in the presence of PazePC-devoid
vesicles exhibited almost no changes. This was in line with the DSC thermograms,
where vesicles without PazePC were largely unaffected by Bax, in contrast to PazePC-
vesicles which were stabilized. This stabilization could be explained by the structural
change of Bax, resulting in an increase in -helical content which potentially reflected
a penetration of the protein into the bilayer, thereby stabilizing the gel-phase (as
shown for cholesterol 184). To further examine the structural change of Bax, CD was
performed at 37 °C, the physiologically relevant temperature. The CD signal dropped
slightly for the protein alone and in the presence of vesicles without PazePC. For the
PazePC-containing sample the CD signal also decreased, though more interestingly,
the protein displayed a dramatic change in CD profile compared to the other two
samples. This could correspond to a physiologically relevant change in the
conformation of Bax, important for membrane-association and -insertion.
To gain further insight into the apparent interaction of Bax to PazePC-vesicles at
physiological temperature, the binding of the protein to the membranes was
investigated by using a simple assay. After incubation of Bax with PazePC-liposomes,
the sample was submitted to ultracentrifugation and the pellet and supernatant were
separated, followed by analysis with SDS-PAGE. By using densitometry, the intensity
of the bands corresponding to the pellet and supernatant could be quantified,
yielding a 70:30 ratio. Thus, it appeared that approximately 70% of the Bax
population was persistently bound to the PazePC-vesicles, in strong contrast to
Results and discussion
36
earlier published results that suggested a transient and weak association to
membranes 129; 131. Solid state NMR was then used to characterize the membrane-
bound population. 31P NMR spectra provided, as expected, perturbations in the
headgroup region upon the addition of Bax, leading to more narrow resonances. 1H
MAS NMR spectra supported this observation for the membrane interface, and
importantly, displayed perturbations in the acyl chain region. These results
confirmed that Bax was not only associated with the bilayer surface but also inserted
deeply into the hydrophobic core of the membrane. Hence, Bax was able to interact
extensively with the mit0chondrial CS-mimicking vesicles due to the change in
dynamics and organization, inflicted by the presence of PazePC.
6.1.3 OxPl-induced recruitment of Bax to the MOM during
apoptosis
The ability of OxPls to perturb the organization and hydration of membranes
seems to facilitate the binding and insertion of Bax into the lipid bilayer. Bax has
previously been reported to exist in at least two different conformations that can bind
to membranes. In healthy cells there is a minor population of weakly attached
inactive protein molecules, while in apoptotic cells a population of active, membrane-
inserted Bax exists 31; 113. The presence of OxPls clearly shifts the equilibrium of Bax
molecules from being predominantly cytosolic to largely membrane-bound,
potentially increasing the “primed” population. This population can then eventually
overcome the inhibition of pro-survival Bcl-2 proteins and further propagate to
induce MOMP 185.
Research addressing the question how individual, defined OxPl species can trigger
the intrinsic pathway has mainly focused on the peroxidation of CL, leading to its
dissociation from cytochrome c and translocation to the MOM 125; 186; 187.
Interestingly, McIntyre’s laboratory showed recently that another OxPl species
containing the azelaoyl residue (same as for PazePC), had a direct effect on
mitochondria and the onset of apoptosis 67. In that study, recapitulating exogenous
OxPls found in the circulation, addition of this OxPl species gave rise to both
cytochrome c release from isolated mitochondria and induction of apoptosis in intact
cells. Overexpression of Bcl-XL antagonized the effect of the OxPls, suggesting an
involvement of the Bcl-2 family in the regulation of OxPl-induced apoptosis; a finding
supported by further studies by the same group 68. In this later work they observed a
synergistic effect of full-length Bid (no effect for Bid alone) and yet another OxPl
species containing the azelaoyl residue. Bid can apart from regulating the activation
of Bax and Bak also act as a phospholipid transferase 188. This way it has presumably
the capability to promote the association of OxPl species to mitochondria. This was
evident as the presence of the OxPl had a significantly reduced effect on mitochondria
isolated from the liver of Bid knock-out mice. As in the previous study the addition of
OxPls had a great impact (facilitated by Bid), giving rise to membrane depolarization
in isolated mitochondria as well as initiation of apoptosis in intact cells. Again,
Results and discussion
37
overexpression of Bcl-XL provided protection against these two OxPl-induced effects,
reinforcing the idea of a Bcl-2 family-mediated regulation of the OxPl-induced
apoptosis.
It is tempting to speculate about the specific role of the PazePC-like OxPls in these
studies, especially when the results presented in paper II are taken into account. Bcl-
XL has been proposed to associate only with the truncated form of Bid (tBid), a
binding interaction strongly promoted by the membrane environment 105; 160. Intact
Bid is thus unable to bind to Bcl-XL and furthermore does not readily insert into the
membrane 189 . Hence, the protection of the mitochondrial integrity conferred by Bcl-
XL may not have been due to sequestering of the BH3-only protein, but potentially by
inhibiting membrane-associated Bax, primed by the presence of PazePC-like OxPls to
insert into the bilayer. Therefore, a possible scenario would reflect the recruitment of
exogenous OxPls to the mitochondrial bilayer by Bid (Fig. 16). This would be
followed by an increase in the population of Bax molecules associated to the MOM,
with Bax presumably also penetrating into the OxPl-containing membrane.
Alternatively, intracellular production of ROS (formed e.g. from mitochondrial
respiratory chain 190 or death receptor activity 191) could promote the generation of
OxPls in the MOM, again leading to an increased amount of primed Bax. Exceeding a
certain critical level (perhaps assisted by BH3-only proteins), Bax will overwhelm the
population of pro-survival Bcl-2 family members, homo-oligomerize and induce
MOMP. The first event would reflect an extracellular OxPl-induced onset of intrinsic
apoptosis, whereas the second would correspond to an intracellular ROS-triggered
initiation of the mitochondria-mediated apoptotic pathway and/or a later stage
(following MOMP), during which the formation of OxPls amplifies the action of Bax
and cytochrome c release.
Figure 16. Model of the recruitment of Bax to the MOM, triggered by OxPl-induced perturbation of the
bilayer. The presence of OxPls can either be due to internalization and mitochondria-association of
extracellular species mediated by Bid, or be a consequence of lipid peroxidation by intracellular ROS. The
OxPl-containing membrane will then trigger a significant increase in the MOM-associated population of Bax,
and also facilitate incorporation of the pro-apoptotic protein.
Bax
ROSBid
Cytoplasm
Extracellular site
MOM
OxPls
Bax Bax
Results and discussion
38
6.2 Cell-free protein synthesis of Bcl-2
Membrane proteins are extremely attractive targets, not only in basic life science
research but also in the hunt for novel pharmaceutical drugs. The reason for this
interest lies in the crucial roles they play in countless biochemical processes and
diseases. Even though a tremendous amount of effort has been made over the years
to characterize and understand their functions, mechanistic and structural data still
remain scarce. The reason for this is related to their inherent membrane-associated
properties, leading to cell toxicity as well as aggregation in aqueous solutions;
therefore reduced expression levels and very low yields following tedious extraction
and purification procedures. In many cases, these problems have severely restricted
the work on membrane proteins and researchers have used mutated and truncated
soluble variants. Although these modifications have enabled the collection of
biophysical and biochemical data of the proteins in solution, a complete insight into
the properties and functions of these membrane proteins cannot be obtained unless
working with the intact protein variants. The knowledge about how full-length
membrane proteins behave in their native membrane environment has therefore
remained very limited.
To this end a protocol to produce the full-length human integral membrane Bcl-2
protein for biophysical studies was elaborated. To overcome the toxicity and to some
extent hydrophobicity issues cell-free protein synthesis was employed. Until recently,
in vitro expression was seen as an expensive alternative method, but during the last
few years optimized systems and protocols have emerged, showing great promises for
good yields at lower cost 174; 176; 177.
6.2.1 Gene optimization
To ensure a good expression level of Bcl-2, an optimized artificial gene was used as
DNA template. The gene, ordered from Verdezyne (Carlsbad, CA, USA), was designed
to 1) prevent the formation of mRNA secondary structures, 2) be complementary to
the host cell tRNA pool, and 3) eliminate the number of redundant base-pairing 192.
All these three issues would normally lead to a major decrease in the efficiency of the
translation procedure. mRNA secondary structures, which arise due to pairing
between matching bases forming complementary double strands, prevent the reading
of the codons by the ribosomes. For non-complementary codon usage, the translation
will be significantly reduced due to non-matching amino acid-carrying tRNA-
molecules to the codons in the mRNA-strand. This is in particular problematic for
heterologous expression of eukaryotic genes in prokaryotic host cells, since the
prokaryotic tRNA-molecules recognize specific codons that are less abundant in
mRNA derived from eukaryotic genes. When ordering synthetic DNA, these two
parameters are generally accounted for in the design of the gene. Although having a
large negative effect on the expression level when ignored, optimization of the gene in
respect to mRNA secondary structure and codon usage does not automatically
Results and discussion
39
facilitate high expression levels. Verdezyne has therefore developed a benchmark for
the syntheses of genes that are engineered to not only resolve the first two mentioned
problems but also to be devoid of over-represented codon pairs.
The idea that elimination of over-represented codon pairs can enhance gene
expression comes from a number of observations. The discovery of high abundance of
certain codon pairs (over-represented codon pairs) led to the notion of highly biased
codon utilization patterns in organisms 193. Since different codons recruit different
isoacceptor tRNA molecules for translation of the same amino acid and that these
molecules can physically interact at the ribosomes 194, the codon usage bias was
assumed to affect the expression of genes. Indeed, this was later confirmed by a study
that showed the more over-represented a codon pair was the slower it was translated 195. The function of these abundant codon pairs has been proposed to be regulation of
the rate of translation by introducing pause sites that would lower the risk of protein
misfolding (providing time for correct positioning of different parts in the structure) 192. Although important in nature, this mechanism prevents overexpression of
proteins due to much slower protein elongation. This is especially true when using a
heterologous system due to the difference in tRNA isoacceptor structures, codon pair
bias and codon usage between different organisms, where over-represented pairs can
be introduced randomly. To account for this issue, Verdezyne has developed the
design of synthetic DNA sequences, called Hot Rod (hr) genes. These artificial genes
are devoid of redundant codon pairs and are in addition optimized regarding to
codon usage and mRNA secondary structure prevention; ensuring coupled
translation and transcription at high rates for heterologous gene expression.
6.2.2 Batch-mode cell-free expression
Very recently Pedersen et al developed a rapid and cost-efficient in-house in vitro
expression of challenging proteins, including membrane-associated species 177. In
papers III and IV this batch-mode cell-free expression system was utilized to
produce full-length human Bcl-2 for CD, fluorescence and reconstitution studies. The
scheme for the optimization of the in-house expression protocol is presented below in
Fig. 17, demonstrating the complexity of the protein synthesis process. Since no
living cells are involved in the synthesis of the protein, the risk of low protein levels
due to cell toxicity induced by heterologous expression is diminished. In addition, the
reaction conditions are easily controlled, allowing e.g. the addition of detergent
directly into the mixture prior to induction of the protein synthesis machinery (as in
the case of Bcl-2). Hence, following its formation the protein can immediately be
solubilized by detergent molecules, in contrast to cell systems where the membrane
protein is aggregated into inclusion bodies after cell lysis.
Results and discussion
40
Figure 17. Optimization scheme of the in-house cell-free protein synthesis protocol. GFPcyc3 and tdTomato
were used as reporter proteins of the expression yield for mixture preparations at 30 and 37 C, respectively.
All changes made compared to the initial protocol are highlighted in bold (reproduced with permission from
Pedersen et al. 177).
Different detergents were screened for the solubilization of Bcl-2, where Brij-78
and Brij-58 clearly constituted the most promising species, shifting almost the whole
population of recombinantly expressed protein to the soluble fraction. Brij-detergents
have been shown to be suitable for cell-free protein synthesis, resulting in good yield
of target proteins 176; 196. Since Bcl-2 started to precipitate after overnight storage in
the presence of Brij-78, Brij-58 was considered a better choice. But before any
upscaling of the protein synthesis could be performed, another issue had to be solved;
the expression level was still low despite the optimization of the conditions of the cell-
free system. Therefore an hr gene from Verdezyne was ordered, engineered as
outlined in 6.2.1, replacing the previously used DNA template (which was only
optimized in respect to codon usage and mRNA secondary structure avoidance).
What could be observed was a dramatic increase in the amount of synthesized
protein. Thus, the combination of Brij-58 and the hr gene showed good potential for a
decent yield of the full-length Bcl-2 protein. In the final protocol the expression
mixtures, containing the optimized gene (encoding His6-tagged protein), were
prepared in 1.5 mL Eppendorf tubes on ice and incubated at 800 rpm and 30 C for
2.5 h. To prevent protein aggregation, His6-Bcl-2 was solubilized by adding 1% w/v
Results and discussion
41
Brij-58 (paper III) or 0.1% w/v Brij-35 (paper IV) into the reaction mixture. The
reason for the switch of detergent in paper IV was because of detergent removal
issues (see 6.4). Following isolation using Ni-NTA affinity columns and cleavage of
the His6-tag, the yield of pure protein was 0.25-0.3 mg/mL reaction mixture.
Importantly, the replacement of Brij-58 with Brij-35 did neither occlude the solubility
nor the yield of Bcl-2. Also, the correct fold of the synthesized protein was confirmed
both by CD (predominantly -helical structure) and fluorescence spectroscopy
(proper ligand-binding), as described in 6.3. As recently reported for other
membrane proteins, cell-free protein synthesis proved to be a suitable method for the
low-cost production of intact Bcl-2, for the first time enabling biophysical studies on
the non-modified protein.
6.3 Interplay between Bax and Bcl-2
Termed the “holy grail” of apoptosis research 21, the regulation mechanism behind
the action of Bax (and Bak) still remains elusive. Due to the failure of producing full-
length Bcl-2 for biophysical studies, the putative association between Bax and Bcl-2
has previously been indicated mainly from co-immunoprecipitation studies 109; 161; 185.
In paper IV, thanks to the newly developed protocol (paper III) for the production of
full-length Bcl-2, a biophysical study could be carried out for the non-modified
proteins in the presence of 0.05% w/v Brij-35 (CMC = 0.011% w/v). In previous co-
immunoprecipitation experiments nonionic detergents were used to solubilize cell
lysates and, unexpectedly, were also able to activate Bax 197; 198, thereby triggering its
hetero-dimerization with Bcl-2 109. Recently it has been proposed that also Bcl-2 has
to be activated to directly engage and restrain activated Bax 157; 158. The
conformational rearrangement of membrane-bound Bcl-2, necessary for inhibition of
the pro-apoptotic protein, has been demonstrated to be induced by BH3-only
proteins, as both tBid and a peptide corresponding to the BH3-domain of Bim were
able to trigger this structural change 155; 157. Also, activated Bax (by tBid) managed to
activate Bcl-2 157, which explains why Bcl-2 and Bax can be immunoprecipitated in
the presence of certain detergents that trigger the activation of the latter. Although
classified as a nonionic detergent, Brij-35 has previously been reported to prevent
any conformational changes in Bax and its hetero-dimerization with Bcl-XL 198. It was
therefore surprising that Bax in the presence of Brij-35, above its CMC, displayed a
significantly increased CD signal in the far-UV region as well as a large reduction in
melting cooperativity (less narrow melting transition); characteristics of detergent- or
membrane-activated Bax 129; 199; 200. Hence, there existed a possibility that Bax in fact
was able to interact with Bcl-2 in the presence of 0.05% w/v Brij-35.
CD spectroscopy was used to detect any potential structural rearrangements or
changes in melting characteristics as a consequence of Bax-Bcl-2 interplay. To further
investigate the association between the two proteins, a ligand-binding assay was
performed, where the change in fluorescence signal of antimycin A2 (known ligand
for Bcl-2 163; 201) was monitored upon addition of the proteins. Unfortunately, all the
Results and discussion
42
CD and fluorescence experiments failed to prove any significant association between
Bax and Bcl-2. The creation of a theoretical CD spectrum consisting of the sum of the
individual protein spectra was next to identical to the experimental spectrum
corresponding to the sample with both proteins; this was true for both the far-UV
region and the thermal melting measurements. Fluorescence spectroscopy confirmed
the proper fold of Bcl-2 since a considerable amount of antimycin A2 bound to the
protein, manifested in lesser quenching of the ligand signal. But, Bax was unable to
displace antimycin A2 from the hydrophobic groove of Bcl-2, as the quenching of the
signal of the ligand remained unchanged compared to the case with only Bcl-2
present. The question was then; was the detergent-driven conformation of Bax not
recognized by Bcl-2? Or could the interactions between Bax and Bcl-2 not be probed
by CD, and if not be too weak and/or interfered by micelles to be able to compete out
the ligand-Bcl-2 interplay? Or, maybe Bax was not all activated by Brij-35, abolishing
any chance of protein-protein association.
To address the first two issues, the concentration of Brij-35 was lowered to
0.0028% w/v (well below its CMC) for all the samples. Bax still exhibited a slight CD
signal enhancement as well as a decrease in melting cooperativity (compared to the
sample without Brij-35), potentially reflecting a partial alteration/activation of the
protein. When the CD measurements were repeated for the new Brij-35
concentration, there were still no significant changes between the theoretical and
experimental far-UV CD spectra. Though, for the melting unfolding curves, there was
a clear reduction in cooperativity for the experimental spectrum, indicating that Bax
and Bcl-2 associated with each other. The ligand-binding assay strongly supported
this observation, showing firstly that Bcl-2 was still properly folded (large increase in
antimycin A2 signal) and secondly that Bax competitively inhibited the binding of the
ligand to the pro-survival protein.
Interestingly, Bax could only interact with Bcl-2 in the presence of low
concentration of Brij-35, below its CMC. It is difficult to give an unambiguous
explanation to this phenomenon – whether it was due to the absence of micelles or
prevention of a conformational change in Bax precluding its association with Bcl-2 –
but since Bcl-2 only can bind to activated Bax, the results here indicate that Bax was
indeed partially activated by the low amount of Brij-35. Hence, Bcl-2 may be able to
restrain Bax prior to the formation of the pro-apoptotic protein’s fully activated form.
This could reflect a scenario where a primed population in healthy cells is kept
detained at MOM 185. As discussed in 6.1.3, this inhibition of partially activated Bax
could then presumably be overcome by e.g. the occurrence of OxPls in the membrane
(Fig. 18), maybe further supported by BH3-only proteins. This would lead to an
increased number of membrane-associated Bax molecules, finally being able to
trigger homo-oligomerization and pore formation.
Results and discussion
43
Figure 18. (A) During normal conditions the few primed Bax molecules are inhibited by Bcl-2. (B) Upon
presence of OxPls the population of primed Bax increases. The large number of primed Bax can (maybe
further assisted by BH3-only proteins) eventually overcome the inhibition of Bcl-2, which can probably
sequester both membrane-associated and -inserted Bax.
6.4 Reconstitution of Bcl-2 into proteoliposomes
When working with membrane proteins the main aim is to study them in their
native membrane environment. Unfortunately, reconstitution of these proteins into
native membrane-mimicking lipid system is often a major obstacle. Even though the
use of detergent micelles can to some extent replace membrane-like conditions, only
the lipid bilayer itself provides a native-like environment for membrane proteins
where they can exert their functions under realistic physiological conditions.
Furthermore, the presence of detergent can potentially influence downstream
applications. When studying membrane proteins it is therefore desirable to restore
their native bilayer environment to gain complete structural and mechanistic
insights. To this end a protocol for the reconstitution of Bcl-2 into proteoliposomes
was developed in paper IV. The principle of reconstitution is rather simple (see Fig.
19). The protein of interest is solubilized by an appropriate detergent above the
CMCdetergent, followed by its addition to mixed micelles with both detergent and lipid
constituents. In order to transfer the protein from the detergent micelles to the
vesicles in the sample, the detergent molecules obviously have to be eliminated. This
can be achieved by different methods, e.g. dialysis or adding polystyrene beads (that
adsorb detergent molecules), leading to a gradual decrease in the detergent
concentration. Eventually, when sufficiently few detergent molecules are present in
the sample (at a concentration far below the CMCdetergent), the formation of vesicles
that can accommodate the protein is promoted. Although easy in theory, the choice of
detergent, method, vesicle composition and protein species heavily influence the
success rate of the protein incorporation.
Bax
Cytoplasm
Extracellular site
MOM
Bax BaxBax
Bax
Bcl-2
A B
Bcl-2
BH3-
only
Results and discussion
44
Figure 19. Reconstitution of a membrane protein into proteoliposomes (adapted from Rigaud, 2002 202).
In the first protocol for the expression and purification of Bcl-2, the protein was
solubilized with Brij-58 (paper III). Brij-58 was an excellent detergent to prevent
protein aggregation and also enabled a good expression level of Bcl-2. But, for the
reconstitution study it turned out to be inadequate due to its very low CMC that
severely impedes the removal of detergent molecules. To facilitate the incorporation
of Bcl-2 into the membrane bilayer, Brij-58 was replaced by Brij-35, a similar
detergent species that is possible to eliminate from protein samples according to
previous studies 203. The detergent exchange did neither affect the yield nor the
solubility of the protein and importantly, as shown in 6.3, Bcl-2 was correctly folded
in the presence of Brij-35 (inferred from CD and fluorescence spectroscopy
measurements).
To create mixed micelles, DMPC-vesicles were solubilized by detergent; the sample
changed its appearance from being cloudy to becoming a clear solution. Besides Brij-
35 also Triton X-100 was added to the mixture, due to the low capability of Brij-35
alone to dissolve the vesicles. To initiate the reconstitution of Bcl-2 into the vesicles,
the protein was added to the mixed micelles and incubated under gentle agitation in
the cold room. Since DMPC forms rather large liposomes which are easy to observe,
especially when they exist in the gel-phase, the reconstitution took place at low
temperature. Batches of SM-2 Bio-Beads were added at regular intervals for a several
days until the sample became turbid once again, looking identical to the reference
sample containing no detergent or protein. To isolate the potentially formed
proteoliposomes, the protein and reference samples underwent ultracentrifugation.
To be able to distinguish between protein aggregates and proteoliposomes, the
samples were layered on top of a 30% sucrose solution. Following centrifugation both
the protein and reference samples yielded flocculent liposome pellets in the
buffer/sucrose interface. Importantly, no precipitated protein could be seen in the
bottom of the centrifugation tube, strongly suggesting Bcl-2 was present in the
liposome fraction. To confirm this assumption, the interfacial region was carefully
+
Detergent
Lipid
Protein
Dialysis
Polystyrene
beads
+
Reconstitution
Solubilisation
of protein
Results and discussion
45
isolated. Following washing away any persistent sucrose by dilution with buffer and
centrifugation, the proteoliposomes were resuspended in buffer and analyzed with
SDS-PAGE. The gel showed only one, very clear band corresponding to a molecular
weight of roughly 26 kDa, the same as for Bcl-2. Thus, by changing the detergent
system from Brij-58 to a mixture of Brij-35 and Triton X-100, and stepwise removing
the detergent molecules with adsorbing Bio-Beads, the restoration of a lipid
membrane surrounding full-length Bcl-2 was feasible. This first attempt of
reconstituting this integral membrane protein into its native-like environment will
open up great possibilities for future studies, such as elucidating the impact of
different lipid compositions and protein-protein interactions (in particular with Bax)
on the function and structure of the Bcl-2 protein.
Main findings and outlook
46
7 Main findings and outlook
To conclude my work presented in this thesis, the following important findings are
highlighted here as a summary:
Presence of PoxnoPC or PazePC has a huge disordering effect on membrane
organization and dynamics, leading to distortion of the hydrophobic region and
lateral phase separation as well as complex hydration of the membrane interface
This disordering effect of OxPls appears to facilitate binding and insertion of Bax
to the MOM as seen for PazePC-containing membranes, thereby increasing the
membrane-associated population; reflecting an early step and or/death signal
amplification in the progression of apoptosis
Production of preparative amounts of full-length human Bcl-2 protein is feasible
using cell-free protein synthesis in the presence of Brij-detergents
The putative interaction between Bcl-2 and Bax can for the first time by
biophysical means be validated for the full-length proteins
Reconstitution of Bcl-2 into proteoliposomes can be performed by solubilizing the
protein in buffer containing Triton X-100 and Brij-35, followed by detergent
removal using Bio-Beads SM-2 adsorbents; an important tool for studying Bcl-2
in its native-like environment
The apoptotic machinery is incredibly complex and the depth of the details
regarding all the interacting proteins – some I mentioned here briefly, many others
not all – as well as the characteristics of the mitochondria and properties of
individual lipid species, is simply stunning. Still, the combined effort of many
researchers using various molecular and biophysical methods has shed light on many
of the reactions and biomolecules involved in the regulation of the apoptotic
processes. Through this thesis I can hopefully also contribute with a few, significant
pieces to the vast puzzle of mitochondria-mediated apoptosis. There are yet many
more gaps to fill; mechanistic information on how the Bcl-2 family proteins interact
with each other in the native membrane milieu; the structure and conformational
changes of the proteins as a consequence of protein-protein and protein-membrane
interactions; the mitochondrial response, not only for the whole organelle (e.g.
fission/fusion processes) but also for individual lipid species (e.g. OxPls); additional
interaction partners and functions of the Bcl-2 family. To be able to address some of
these questions, one has to study intact, functional mitochondria and their response
to different stimuli. Therefore, a future approach in this project is to isolate
Main findings and outlook
47
mitochondria from healthy, apoptotic as well as tumor tissues, and study their
responses by solid state NMR upon titration of pro- and anti-apoptotic proteins.
Furthermore, both solid state and solution NMR will be used for structural studies on
the Bcl-2 family proteins, in particular for the full-length Bcl-2 whose expression and
purification are feasible due to the protocol presented in this thesis. NMR provides
structural and dynamical information at atomic level, enabling a deeper insight into
how the proteins and membranes behave under different conditions. The data
obtained from these studies will ultimately not only be crucial to fully understand the
intricate pathways of reactions involved in apoptosis, but also for the engineering of
pharmaceutical drugs targeting various pathologies.
Acknowledgements
48
8 Acknowledgements
The final part of a thesis is probably the funniest but perhaps also the most difficult
one to write - who to include and who to exclude? Well, I decided to present the
people who 1) contributed to my scientific work and 2) provided good friendship
during the past 5 years. I would therefore like to acknowledge the following people:
My supervisor Gerhard, who has fulfilled both criteria mentioned above. You
have been very encouraging and supporting during the tough times of my project (e.g.
the protein production, a.k.a. banging-the-head-against-the-wall) and been giving me
plenty of valuable feedback during the good ones. Importantly, I have also got to
know you not only as your PhD student but also as your friend, and you have proved
numerous times that you genuinely care about your PhD students. Thank you for
everything, I hope we still can have plenty of opportunities to hang out over a beer or
two (or more) in the future.
Magnus, my co-supervisor, for introducing me to the world of research. Thank
you for the good years I spent in your group (which were very fruitful!), I also really
appreciate all the help and encouragement during the tough periods of my PhD
studies, as well as the access to your lab and all the equipment.
Anders Pedersen and Göran Karlsson at Swedish NMR Centre, for all the
effort you have made to produce the Bcl-2 protein for my projects, in addition to all
the valuable feedback on my data.
When the research is not going as well as you hope for and you feel more frustrated
than anything else, it is a great relief to have people around you whose company can
ease your mind. I have been very fortunate to have not only great group members but
also other colleagues/friends close by at work. Johan V, you have been an excellent
office mate and a good friend the past few years. Thanks for all the interesting
discussions about beer and life in general, hopefully I can have the chance to drop by
Sillviken in the future to enjoy fresh beer from your microbrewery. Martin, we
haven’t known each other for that long time, but we have so far shared a 120 cm wide
bed, been drunk together at an obscure salsa-place in the heart of Wroclaw and both
been served kebab by the arguably most eccentric kebab salesman in Switzerland.
Thanks for the friendship and the great times we have spent together; all our
discussions about everything under the sun (including apoptosis), the snus that you
have provided and the youtube-clips you have shared. Marco, Roberth and
Tofeeq, thanks for helping me feel at home when I started my work in Gerhard’s
power team.
Michael, you are probably the least selfish person I have got the privilege to know
Thanks for being a good friend and always willing to help with a smile, no matter
Acknowledgements
49
what. Patrik, thanks a lot for the patience answering all my “stupid” questions,
especially during the first period of my PhD studies. We have now become good
buddies and we will definitely get the chance to discuss more about existentialism,
philosophy of life etc over a beer or two in the future. Radek, I really appreciate the
time we have spent together; going to pubs in Poland and Czech Republic, racing
down the “easy” slopes in Tärnaby, and just sharing a glass of good wine in your
apartment. I look forward to see you again in Prague. Jörgen, we are diametrically
opposed to each other in many ways, for better or worse ;) Thanks for the things we
have shared and those that you have done for me; teaching me numerous lab
techniques, fixing the “perfect sound” for my parties, and bringing different (and
peculiar) perspectives of life. Barbara and Nina, thanks for being good friends and
trying to bend my eyes open to “see” the reality. Alexander, always a pleasure to
hang out with you, whether it is squash playing or drinking some beer. Ximena and
Lina, thanks for nice company during lunch and coffee breaks.
Dat, Konrad and Janek, thanks for the good times together, both in and outside
the lab. See you again in Lund and Wroclaw, next time I will for sure bring the correct
Mariestad (Old Ox)!
My brothers-in-arms of low-Klas Härnösandshumor, Ecke and Patrik, and the
rest of my jolly-good friends in the Härnta-med-omnejd-gänget, Wesa, Björn,
Danne, Tomas, Micke, Thomme and Johan Å. Well, we have had a lot of fun
together in the past and will for sure share more great experiences in the future.
Period.
Biomedicingänget, Sara, Åsa, Sonia, Kattis, Isabelle, Daniel and Micke, my
first friends in Umeå. We have known each other for a long time now and finally it’s
time for me to graduate and start my life for real. Thanks for everything we have
shared during the past >12 years. It has been a fun and inspiring ride, and has helped
forming me into the person I am today.
Biologgänget, especially Per, Danne, Rolle and Viktoria. I have known you
almost as long time as the people just mentioned above, and you have always been
there for me whenever I have needed it. Thanks a lot for all the fun we have
experienced together and the intimate discussions; including everything from anime
and training to man vs women topics. It has been a true pleasure to hang out with
you.
The Vietnamese community, especially Dat, Ha Giang, Linh, Tuyet, Hoa,
Minh Dang, Thong, Minh Nguyen and Tam, for broadening my horizons and
sharing good friendships with me.
Acknowledgements
50
Non-categorized friends who also have contributed to my well-being; Filip,
Rasmus, Johan H, Göran, Macke and Kalle.
My flat mates during this time – not enough space to name all of them here.
Min familj, som inte har haft en aning om vad jag pysslat med men som ändå alltid
varit stödjande och stolt över vad jag har gjort. Mamma, utan din kämpaglöd och
enastående förmåga att övervinna livets prövningar skulle denna avhandling aldrig
ha skrivits. Tack så mycket för allt som är mycket mer värdefullt än ord kan beskriva.
Lastly, I want to say thanks to Lan, my wonderful girlfriend and soon-to-be-wife.
You really have a great (and sometimes bad ;) sense of humor, I’m looking forward to
laugh together with you the rest of our lives, whether it is in Umeå, Saigon or
anywhere else in the world we settle down. Anh iu em!
References
51
References
1. Ulukaya, E., Acilan, C. & Yilmaz, Y. (2011). Apoptosis: why and how does it occur in biology? Cell
Biochem Funct 29, 468-80.
2. Ekshyyan, O. & Aw, T. Y. (2004). Apoptosis: a key in neurodegenerative disorders. Curr
Neurovasc Res 1, 355-71.
3. Mattson, M. P., Culmsee, C. & Yu, Z. F. (2000). Apoptotic and antiapoptotic mechanisms in
stroke. Cell Tissue Res 301, 173-87.
4. Kerr, J. F., Wyllie, A. H. & Currie, A. R. (1972). Apoptosis: a basic biological phenomenon with
wide-ranging implications in tissue kinetics. Br J Cancer 26, 239-57.
5. Horvitz, H. R. (1999). Genetic control of programmed cell death in the nematode Caenorhabditis
elegans. Cancer Res 59, 1701s-1706s.
6. Chinnaiyan, A. M. (1999). The apoptosome: heart and soul of the cell death machine. Neoplasia 1,
5-15.
7. Kroemer, G., El-Deiry, W. S., Golstein, P., Peter, M. E., Vaux, D., Vandenabeele, P., Zhivotovsky,
B., Blagosklonny, M. V., Malorni, W., Knight, R. A., Piacentini, M., Nagata, S. & Melino, G. (2005).
Classification of cell death: recommendations of the Nomenclature Committee on Cell Death. Cell
Death Differ 12, 1463-7.
8. Tepper, A. D., Ruurs, P., Wiedmer, T., Sims, P. J., Borst, J. & van Blitterswijk, W. J. (2000).
Sphingomyelin hydrolysis to ceramide during the execution phase of apoptosis results from
phospholipid scrambling and alters cell-surface morphology. J Cell Biol 150, 155-64.
9. Martin, S. J., Reutelingsperger, C. P., McGahon, A. J., Rader, J. A., van Schie, R. C., LaFace, D. M.
& Green, D. R. (1995). Early redistribution of plasma membrane phosphatidylserine is a general
feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and
Abl. J Exp Med 182, 1545-56.
10. Nathan, C. & Ding, A. (2010). Nonresolving inflammation. Cell 140, 871-82.
11. Saikumar, P., Dong, Z., Mikhailov, V., Denton, M., Weinberg, J. M. & Venkatachalam, M. A.
(1999). Apoptosis: definition, mechanisms, and relevance to disease. Am J Med 107, 489-506.
12. Krammer, P. H. (2000). CD95's deadly mission in the immune system. Nature 407, 789-95.
13. Green, D. R. & Reed, J. C. (1998). Mitochondria and apoptosis. Science 281, 1309-12.
14. Cory, S. & Adams, J. M. (2002). The Bcl2 family: regulators of the cellular life-or-death switch.
Nat Rev Cancer 2, 647-56.
15. Schulze-Osthoff, K., Ferrari, D., Los, M., Wesselborg, S. & Peter, M. E. (1998). Apoptosis signaling
by death receptors. Eur J Biochem 254, 439-59.
16. Medema, J. P., Scaffidi, C., Kischkel, F. C., Shevchenko, A., Mann, M., Krammer, P. H. & Peter, M.
E. (1997). FLICE is activated by association with the CD95 death-inducing signaling complex
(DISC). EMBO J 16, 2794-804.
17. Muzio, M., Salvesen, G. S. & Dixit, V. M. (1997). FLICE induced apoptosis in a cell-free system.
Cleavage of caspase zymogens. J Biol Chem 272, 2952-6.
18. Srinivasula, S. M., Ahmad, M., Fernandes-Alnemri, T., Litwack, G. & Alnemri, E. S. (1996).
Molecular ordering of the Fas-apoptotic pathway: the Fas/APO-1 protease Mch5 is a CrmA-
inhibitable protease that activates multiple Ced-3/ICE-like cysteine proteases. Proc Natl Acad Sci
U S A 93, 14486-91.
19. Hotchkiss, R. S., Strasser, A., McDunn, J. E. & Swanson, P. E. (2009). Cell death. N Engl J Med
361, 1570-83.
References
52
20. Pereira, W. O. & Amarante-Mendes, G. P. (2011). Apoptosis: a programme of cell death or cell
disposal? Scand J Immunol 73, 401-7.
21. Youle, R. J. & Strasser, A. (2008). The BCL-2 protein family: opposing activities that mediate cell
death. Nat Rev Mol Cell Biol 9, 47-59.
22. Westphal, D., Dewson, G., Czabotar, P. E. & Kluck, R. M. (2011). Molecular biology of Bax and Bak
activation and action. Biochim Biophys Acta 1813, 521-31.
23. Zou, H., Henzel, W. J., Liu, X., Lutschg, A. & Wang, X. (1997). Apaf-1, a human protein
homologous to C. elegans CED-4, participates in cytochrome c-dependent activation of caspase-3.
Cell 90, 405-13.
24. Zou, H., Li, Y., Liu, X. & Wang, X. (1999). An APAF-1.cytochrome c multimeric complex is a
functional apoptosome that activates procaspase-9. J Biol Chem 274, 11549-56.
25. Acehan, D., Jiang, X., Morgan, D. G., Heuser, J. E., Wang, X. & Akey, C. W. (2002). Three-
dimensional structure of the apoptosome: implications for assembly, procaspase-9 binding, and
activation. Mol Cell 9, 423-32.
26. Nys, K., Maes, H., Dudek, A. M. & Agostinis, P. (2011). Uncovering the role of hypoxia inducible
factor-1alpha in skin carcinogenesis. Biochim Biophys Acta 1816, 1-12.
27. Scaffidi, C., Fulda, S., Srinivasan, A., Friesen, C., Li, F., Tomaselli, K. J., Debatin, K. M., Krammer,
P. H. & Peter, M. E. (1998). Two CD95 (APO-1/Fas) signaling pathways. EMBO J 17, 1675-87.
28. Jost, P. J., Grabow, S., Gray, D., McKenzie, M. D., Nachbur, U., Huang, D. C., Bouillet, P., Thomas,
H. E., Borner, C., Silke, J., Strasser, A. & Kaufmann, T. (2009). XIAP discriminates between type I
and type II FAS-induced apoptosis. Nature 460, 1035-9.
29. Li, H., Zhu, H., Xu, C. J. & Yuan, J. (1998). Cleavage of BID by caspase 8 mediates the
mitochondrial damage in the Fas pathway of apoptosis. Cell 94, 491-501.
30. Korsmeyer, S. J., Wei, M. C., Saito, M., Weiler, S., Oh, K. J. & Schlesinger, P. H. (2000). Pro-
apoptotic cascade activates BID, which oligomerizes BAK or BAX into pores that result in the
release of cytochrome c. Cell Death Differ 7, 1166-73.
31. Eskes, R., Desagher, S., Antonsson, B. & Martinou, J. C. (2000). Bid induces the oligomerization
and insertion of Bax into the outer mitochondrial membrane. Mol Cell Biol 20, 929-35.
32. Dimmer, K. S. & Rapaport, D. (2008). Proteomic view of mitochondrial function. Genome Biol 9,
Article 209.
33. Thrash, J. C., Boyd, A., Huggett, M. J., Grote, J., Carini, P., Yoder, R. J., Robbertse, B., Spatafora,
J. W., Rappe, M. S. & Giovannoni, S. J. (2011). Phylogenomic evidence for a common ancestor of
mitochondria and the SAR11 clade. Sci Rep 1, 13; DOI:10.1038/srep00013.
34. Scheffler, I. E. (2001). Mitochondria make a come back. Adv Drug Deliv Rev 49, 3-26.
35. Gonzalvez, F. & Gottlieb, E. (2007). Cardiolipin: setting the beat of apoptosis. Apoptosis 12, 877-
85.
36. Kawahara, I., Koide, M., Tadokoro, O., Udagawa, N., Nakamura, H., Takahashi, N. & Ozawa, H.
(2009). The relationship between calcium accumulation in osteoclast mitochondrial granules and
bone resorption. Bone 45, 980-6.
37. Ardail, D., Privat, J. P., Egret-Charlier, M., Levrat, C., Lerme, F. & Louisot, P. (1990).
Mitochondrial contact sites. Lipid composition and dynamics. J Biol Chem 265, 18797-802.
38. Chakraborti, T., Das, S., Mondal, M., Roychoudhury, S. & Chakraborti, S. (1999). Oxidant,
mitochondria and calcium: an overview. Cell Signal 11, 77-85.
39. Murphy, M. P. (2009). How mitochondria produce reactive oxygen species. Biochem J 417, 1-13.
40. Goodman, S. R. (2008). Medical Cell Biology (Third Edition), Academic Press, pp 101-148.
References
53
41. Detmer, S. A. & Chan, D. C. (2007). Functions and dysfunctions of mitochondrial dynamics. Nat
Rev Mol Cell Biol 8, 870-9.
42. Autret, A. & Martin, S. J. (2010). Bcl-2 family proteins and mitochondrial fission/fusion
dynamics. Cell Mol Life Sci 67, 1599-1606.
43. Campello, S., Lacalle, R. A., Bettella, M., Manes, S., Scorrano, L. & Viola, A. (2006). Orchestration
of lymphocyte chemotaxis by mitochondrial dynamics. J Exp Med 203, 2879-86.
44. Suen, D. F., Norris, K. L. & Youle, R. J. (2008). Mitochondrial dynamics and apoptosis. Genes Dev
22, 1577-90.
45. Sheridan, C., Delivani, P., Cullen, S. P. & Martin, S. J. (2008). Bax- or Bak-induced mitochondrial
fission can be uncoupled from cytochrome C release. Mol Cell 31, 570-85.
46. Pletjushkina, O. Y., Lyamzaev, K. G., Popova, E. N., Nepryakhina, O. K., Ivanova, O. Y., Domnina,
L. V., Chernyak, B. V. & Skulachev, V. P. (2006). Effect of oxidative stress on dynamics of
mitochondrial reticulum. Biochim Biophys Acta 1757, 518-24.
47. Janmey, P. A. & Kinnunen, P. K. J. (2006). Biophysical properties of lipids and dynamic
membranes. Trends Cell Biol 16, 538-546.
48. Kinnunen, P. K. (2009). Amyloid formation on lipid membrane surfaces. The Open Biology
Journal 2, 163-175.
49. Weisz, K., Grobner, G., Mayer, C., Stohrer, J. & Kothe, G. (1992). Deuteron nuclear magnetic
resonance study of the dynamic organization of phospholipid/cholesterol bilayer membranes:
molecular properties and viscoelastic behavior. Biochemistry 31, 1100-12.
50. Lehtonen, J. Y., Holopainen, J. M. & Kinnunen, P. K. (1996). Evidence for the formation of
microdomains in liquid crystalline large unilamellar vesicles caused by hydrophobic mismatch of
the constituent phospholipids. Biophys J 70, 1753-60.
51. Lehtonen, J. Y. & Kinnunen, P. K. (1997). Evidence for phospholipid microdomain formation in
liquid crystalline liposomes reconstituted with Escherichia coli lactose permease. Biophys J 72,
1247-57.
52. Mustonen, P., Virtanen, J. A., Somerharju, P. J. & Kinnunen, P. K. (1987). Binding of cytochrome
c to liposomes as revealed by the quenching of fluorescence from pyrene-labeled phospholipids.
Biochemistry 26, 2991-7.
53. Fruhwirth, G. O., Loidl, A. & Hermetter, A. (2007). Oxidized phospholipids: from molecular
properties to disease. Biochim Biophys Acta 1772, 718-36.
54. van den Brink-van der Laan, E., Killian, J. A. & de Kruijff, B. (2004). Nonbilayer lipids affect
peripheral and integral membrane proteins via changes in the lateral pressure profile. Biochim
Biophys Acta 1666, 275-88.
55. Beranova, L., Cwiklik, L., Jurkiewicz, P., Hof, M. & Jungwirth, P. (2010). Oxidation changes
physical properties of phospholipid bilayers: fluorescence spectroscopy and molecular
simulations. Langmuir 26, 6140-4.
56. Khandelia, H. & Mouritsen, O. G. (2009). Lipid gymnastics: evidence of complete acyl chain
reversal in oxidized phospholipids from molecular simulations. Biophys J 96, 2734-43.
57. Wong-Ekkabut, J., Xu, Z., Triampo, W., Tang, I. M., Tieleman, D. P. & Monticelli, L. (2007). Effect
of lipid peroxidation on the properties of lipid bilayers: a molecular dynamics study. Biophys J 93,
4225-36.
58. Jacob, R. F. & Mason, R. P. (2005). Lipid peroxidation induces cholesterol domain formation in
model membranes. J Biol Chem 280, 39380-7.
59. Volinsky, R., Cwiklik, L., Jurkiewicz, P., Hof, M., Jungwirth, P. & Kinnunen, P. K. (2011). Oxidized
phosphatidylcholines facilitate phospholipid flip-flop in liposomes. Biophys J 101, 1376-84.
References
54
60. Leitinger, N. (2008). The role of phospholipid oxidation products in inflammatory and
autoimmune diseases: evidence from animal models and in humans. Subcell Biochem 49, 325-50.
61. Newmeyer, D. D. & Ferguson-Miller, S. (2003). Mitochondria: releasing power for life and
unleashing the machineries of death. Cell 112, 481-90.
62. Simon, H. U., Haj-Yehia, A. & Levi-Schaffer, F. (2000). Role of reactive oxygen species (ROS) in
apoptosis induction. Apoptosis 5, 415-8.
63. Arroyo, A., Modriansky, M., Serinkan, F. B., Bello, R. I., Matsura, T., Jiang, J., Tyurin, V. A.,
Tyurina, Y. Y., Fadeel, B. & Kagan, V. E. (2002). NADPH oxidase-dependent oxidation and
externalization of phosphatidylserine during apoptosis in Me2SO-differentiated HL-60 cells. Role
in phagocytic clearance. J Biol Chem 277, 49965-75.
64. Matsura, T., Togawa, A., Kai, M., Nishida, T., Nakada, J., Ishibe, Y., Kojo, S., Yamamoto, Y. &
Yamada, K. (2005). The presence of oxidized phosphatidylserine on Fas-mediated apoptotic cell
surface. Biochim Biophys Acta 1736, 181-8.
65. Ganesan, V., Perera, M. N., Colombini, D., Datskovskiy, D., Chadha, K. & Colombini, M. (2010).
Ceramide and activated Bax act synergistically to permeabilize the mitochondrial outer
membrane. Apoptosis 15, 553-62.
66. Loidl, A., Sevcsik, E., Riesenhuber, G., Deigner, H. P. & Hermetter, A. (2003). Oxidized
phospholipids in minimally modified low density lipoprotein induce apoptotic signaling via
activation of acid sphingomyelinase in arterial smooth muscle cells. J Biol Chem 278, 32921-8.
67. Chen, R., Yang, L. & McIntyre, T. M. (2007). Cytotoxic phospholipid oxidation products. Cell
death from mitochondrial damage and the intrinsic caspase cascade. J Biol Chem 282, 24842-50.
68. Chen, R., Feldstein, A. E. & McIntyre, T. M. (2009). Suppression of mitochondrial function by
oxidatively truncated phospholipids is reversible, aided by bid, and suppressed by Bcl-XL. J Biol
Chem 284, 26297-308.
69. Koonin, E. V. & Aravind, L. (2002). Origin and evolution of eukaryotic apoptosis: the bacterial
connection. Cell Death Differ 9, 394-404.
70. Reed, J. C., Doctor, K. S. & Godzik, A. (2004). The domains of apoptosis: a genomics perspective.
Sci STKE 2004, re9.
71. Aouacheria, A., Brunet, F. & Gouy, M. (2005). Phylogenomics of life-or-death switches in
multicellular animals: Bcl-2, BH3-Only, and BNip families of apoptotic regulators. Mol Biol Evol
22, 2395-416.
72. Moldoveanu, T., Liu, Q., Tocilj, A., Watson, M., Shore, G. & Gehring, K. (2006). The X-ray
structure of a BAK homodimer reveals an inhibitory zinc binding site. Mol Cell 24, 677-88.
73. Day, C. L., Chen, L., Richardson, S. J., Harrison, P. J., Huang, D. C. & Hinds, M. G. (2005).
Solution structure of prosurvival Mcl-1 and characterization of its binding by proapoptotic BH3-
only ligands. J Biol Chem 280, 4738-44.
74. Hinds, M. G., Lackmann, M., Skea, G. L., Harrison, P. J., Huang, D. C. & Day, C. L. (2003). The
structure of Bcl-w reveals a role for the C-terminal residues in modulating biological activity.
EMBO J 22, 1497-507.
75. Denisov, A. Y., Madiraju, M. S., Chen, G., Khadir, A., Beauparlant, P., Attardo, G., Shore, G. C. &
Gehring, K. (2003). Solution structure of human BCL-w: modulation of ligand binding by the C-
terminal helix. J Biol Chem 278, 21124-8.
76. Petros, A. M., Medek, A., Nettesheim, D. G., Kim, D. H., Yoon, H. S., Swift, K., Matayoshi, E. D.,
Oltersdorf, T. & Fesik, S. W. (2001). Solution structure of the antiapoptotic protein bcl-2. Proc
Natl Acad Sci U S A 98, 3012-7.
References
55
77. Suzuki, M., Youle, R. J. & Tjandra, N. (2000). Structure of Bax: coregulation of dimer formation
and intracellular localization. Cell 103, 645-54.
78. McDonnell, J. M., Fushman, D., Milliman, C. L., Korsmeyer, S. J. & Cowburn, D. (1999). Solution
structure of the proapoptotic molecule BID: a structural basis for apoptotic agonists and
antagonists. Cell 96, 625-34.
79. Chou, J. J., Li, H., Salvesen, G. S., Yuan, J. & Wagner, G. (1999). Solution structure of BID, an
intracellular amplifier of apoptotic signaling. Cell 96, 615-24.
80. Muchmore, S. W., Sattler, M., Liang, H., Meadows, R. P., Harlan, J. E., Yoon, H. S., Nettesheim,
D., Chang, B. S., Thompson, C. B., Wong, S. L., Ng, S. L. & Fesik, S. W. (1996). X-ray and NMR
structure of human Bcl-xL, an inhibitor of programmed cell death. Nature 381, 335-41.
81. Petros, A. M., Olejniczak, E. T. & Fesik, S. W. (2004). Structural biology of the Bcl-2 family of
proteins. Biochim Biophys Acta 1644, 83-94.
82. Kvansakul, M., Yang, H., Fairlie, W. D., Czabotar, P. E., Fischer, S. F., Perugini, M. A., Huang, D.
C. & Colman, P. M. (2008). Vaccinia virus anti-apoptotic F1L is a novel Bcl-2-like domain-
swapped dimer that binds a highly selective subset of BH3-containing death ligands. Cell Death
Differ 15, 1564-71.
83. Sattler, M., Liang, H., Nettesheim, D., Meadows, R. P., Harlan, J. E., Eberstadt, M., Yoon, H. S.,
Shuker, S. B., Chang, B. S., Minn, A. J., Thompson, C. B. & Fesik, S. W. (1997). Structure of Bcl-
xL-Bak peptide complex: recognition between regulators of apoptosis. Science 275, 983-6.
84. Petros, A. M., Nettesheim, D. G., Wang, Y., Olejniczak, E. T., Meadows, R. P., Mack, J., Swift, K.,
Matayoshi, E. D., Zhang, H., Thompson, C. B. & Fesik, S. W. (2000). Rationale for Bcl-xL/Bad
peptide complex formation from structure, mutagenesis, and biophysical studies. Protein Sci 9,
2528-34.
85. Simpson, C. D., Anyiwe, K. & Schimmer, A. D. (2008). Anoikis resistance and tumor metastasis.
Cancer Lett 272, 177-85.
86. Liu, X., Kim, C. N., Yang, J., Jemmerson, R. & Wang, X. (1996). Induction of apoptotic program in
cell-free extracts: requirement for dATP and cytochrome c. Cell 86, 147-57.
87. Li, L. Y., Luo, X. & Wang, X. (2001). Endonuclease G is an apoptotic DNase when released from
mitochondria. Nature 412, 95-9.
88. Susin, S. A., Lorenzo, H. K., Zamzami, N., Marzo, I., Snow, B. E., Brothers, G. M., Mangion, J.,
Jacotot, E., Costantini, P., Loeffler, M., Larochette, N., Goodlett, D. R., Aebersold, R., Siderovski,
D. P., Penninger, J. M. & Kroemer, G. (1999). Molecular characterization of mitochondrial
apoptosis-inducing factor. Nature 397, 441-6.
89. Suzuki, Y., Imai, Y., Nakayama, H., Takahashi, K., Takio, K. & Takahashi, R. (2001). A serine
protease, HtrA2, is released from the mitochondria and interacts with XIAP, inducing cell death.
Mol Cell 8, 613-21.
90. Du, C., Fang, M., Li, Y., Li, L. & Wang, X. (2000). Smac, a mitochondrial protein that promotes
cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102, 33-42.
91. Verhagen, A. M., Ekert, P. G., Pakusch, M., Silke, J., Connolly, L. M., Reid, G. E., Moritz, R. L.,
Simpson, R. J. & Vaux, D. L. (2000). Identification of DIABLO, a mammalian protein that
promotes apoptosis by binding to and antagonizing IAP proteins. Cell 102, 43-53.
92. Kim, H. E., Du, F., Fang, M. & Wang, X. (2005). Formation of apoptosome is initiated by
cytochrome c-induced dATP hydrolysis and subsequent nucleotide exchange on Apaf-1. Proc Natl
Acad Sci U S A 102, 17545-50.
93. van Loo, G., Schotte, P., van Gurp, M., Demol, H., Hoorelbeke, B., Gevaert, K., Rodriguez, I., Ruiz-
Carrillo, A., Vandekerckhove, J., Declercq, W., Beyaert, R. & Vandenabeele, P. (2001).
References
56
Endonuclease G: a mitochondrial protein released in apoptosis and involved in caspase-
independent DNA degradation. Cell Death Differ 8, 1136-42.
94. Eckelman, B. P., Salvesen, G. S. & Scott, F. L. (2006). Human inhibitor of apoptosis proteins: why
XIAP is the black sheep of the family. EMBO Rep 7, 988-94.
95. Cheng, E. H., Wei, M. C., Weiler, S., Flavell, R. A., Mak, T. W., Lindsten, T. & Korsmeyer, S. J.
(2001). BCL-2, BCL-X(L) sequester BH3 domain-only molecules preventing BAX- and BAK-
mediated mitochondrial apoptosis. Mol Cell 8, 705-11.
96. Zong, W. X., Lindsten, T., Ross, A. J., MacGregor, G. R. & Thompson, C. B. (2001). BH3-only
proteins that bind pro-survival Bcl-2 family members fail to induce apoptosis in the absence of
Bax and Bak. Genes Dev 15, 1481-6.
97. Adams, J. M. & Cory, S. (2007). The Bcl-2 apoptotic switch in cancer development and therapy.
Oncogene 26, 1324-37.
98. Leber, B., Lin, J. & Andrews, D. W. (2007). Embedded together: the life and death consequences
of interaction of the Bcl-2 family with membranes. Apoptosis 12, 897-911.
99. Wolter, K. G., Hsu, Y. T., Smith, C. L., Nechushtan, A., Xi, X. G. & Youle, R. J. (1997). Movement
of Bax from the cytosol to mitochondria during apoptosis. J Cell Biol 139, 1281-92.
100. Hsu, Y. T., Wolter, K. G. & Youle, R. J. (1997). Cytosol-to-membrane redistribution of Bax and Bcl-
X(L) during apoptosis. Proc Natl Acad Sci U S A 94, 3668-72.
101. Chen-Levy, Z., Nourse, J. & Cleary, M. L. (1989). The bcl-2 candidate proto-oncogene product is a
24-kilodalton integral-membrane protein highly expressed in lymphoid cell lines and lymphomas
carrying the t(14;18) translocation. Mol Cell Biol 9, 701-10.
102. Tsujimoto, Y., Ikegaki, N. & Croce, C. M. (1987). Characterization of the protein product of bcl-2,
the gene involved in human follicular lymphoma. Oncogene 2, 3-7.
103. Leber, B., Lin, J. & Andrews, D. W. (2010). Still embedded together binding to membranes
regulates Bcl-2 protein interactions. Oncogene 29, 5221-30.
104. Lovell, J. F., Billen, L. P., Bindner, S., Shamas-Din, A., Fradin, C., Leber, B. & Andrews, D. W.
(2008). Membrane binding by tBid initiates an ordered series of events culminating in membrane
permeabilization by Bax. Cell 135, 1074-84.
105. Billen, L. P., Kokoski, C. L., Lovell, J. F., Leber, B. & Andrews, D. W. (2008). Bcl-XL inhibits
membrane permeabilization by competing with Bax. PLoS Biol 6, e147.
106. Kudla, G., Montessuit, S., Eskes, R., Berrier, C., Martinou, J. C., Ghazi, A. & Antonsson, B. (2000).
The destabilization of lipid membranes induced by the C-terminal fragment of caspase 8-cleaved
bid is inhibited by the N-terminal fragment. J Biol Chem 275, 22713-8.
107. Bleicken, S., Classen, M., Padmavathi, P. V., Ishikawa, T., Zeth, K., Steinhoff, H. J. & Bordignon,
E. (2010). Molecular details of Bax activation, oligomerization, and membrane insertion. J Biol
Chem 285, 6636-47.
108. Peyerl, F. W., Dai, S., Murphy, G. A., Crawford, F., White, J., Marrack, P. & Kappler, J. W. (2007).
Elucidation of some Bax conformational changes through crystallization of an antibody-peptide
complex. Cell Death Differ 14, 447-52.
109. Hsu, Y. T. & Youle, R. J. (1997). Nonionic detergents induce dimerization among members of the
Bcl-2 family. J Biol Chem 272, 13829-34.
110. Sani, M. A., Dufourc, E. J. & Grobner, G. (2009). How does the Bax-alpha1 targeting sequence
interact with mitochondrial membranes? The role of cardiolipin. Biochim Biophys Acta 1788,
623-31.
111. Garcia-Saez, A. J., Mingarro, I., Perez-Paya, E. & Salgado, J. (2004). Membrane-insertion
fragments of Bcl-xL, Bax, and Bid. Biochemistry 43, 10930-43.
References
57
112. Brock, S. E., Li, C. & Wattenberg, B. W. (2010). The Bax carboxy-terminal hydrophobic helix does
not determine organelle-specific targeting but is essential for maintaining Bax in an inactive state
and for stable mitochondrial membrane insertion. Apoptosis 15, 14-27.
113. Antonsson, B., Montessuit, S., Sanchez, B. & Martinou, J. C. (2001). Bax is present as a high
molecular weight oligomer/complex in the mitochondrial membrane of apoptotic cells. J Biol
Chem 276, 11615-23.
114. George, N. M., Evans, J. J. & Luo, X. (2007). A three-helix homo-oligomerization domain
containing BH3 and BH1 is responsible for the apoptotic activity of Bax. Genes Dev 21, 1937-48.
115. Dewson, G., Ma, S., Frederick, P., Hockings, C., Tan, I., Kratina, T. & Kluck, R. M. (2012). Bax
dimerizes via a symmetric BH3:groove interface during apoptosis. Cell Death Differ 19, 661-70.
116. Annis, M. G., Soucie, E. L., Dlugosz, P. J., Cruz-Aguado, J. A., Penn, L. Z., Leber, B. & Andrews, D.
W. (2005). Bax forms multispanning monomers that oligomerize to permeabilize membranes
during apoptosis. EMBO J 24, 2096-103.
117. Valero, J. G., Sancey, L., Kucharczak, J., Guillemin, Y., Gimenez, D., Prudent, J., Gillet, G.,
Salgado, J., Coll, J. L. & Aouacheria, A. (2011). Bax-derived membrane-active peptides act as
potent and direct inducers of apoptosis in cancer cells. J Cell Sci 124, 556-64.
118. Crimi, M. & Esposti, M. D. (2011). Apoptosis-induced changes in mitochondrial lipids. Biochim
Biophys Acta 1813, 551-7.
119. Renault, T. T. & Manon, S. (2011). Bax: Addressed to kill. Biochimie 93, 1379-91.
120. Rytomaa, M., Mustonen, P. & Kinnunen, P. K. (1992). Reversible, nonionic, and pH-dependent
association of cytochrome c with cardiolipin-phosphatidylcholine liposomes. J Biol Chem 267,
22243-8.
121. Nicholls, P. (1974). Cytochrome c binding to enzymes and membranes. Biochim Biophys Acta
346, 261-310.
122. Kagan, V. E., Tyurin, V. A., Jiang, J., Tyurina, Y. Y., Ritov, V. B., Amoscato, A. A., Osipov, A. N.,
Belikova, N. A., Kapralov, A. A., Kini, V., Vlasova, II, Zhao, Q., Zou, M., Di, P., Svistunenko, D. A.,
Kurnikov, I. V. & Borisenko, G. G. (2005). Cytochrome c acts as a cardiolipin oxygenase required
for release of proapoptotic factors. Nat Chem Biol 1, 223-32.
123. Belikova, N. A., Vladimirov, Y. A., Osipov, A. N., Kapralov, A. A., Tyurin, V. A., Potapovich, M. V.,
Basova, L. V., Peterson, J., Kurnikov, I. V. & Kagan, V. E. (2006). Peroxidase activity and
structural transitions of cytochrome c bound to cardiolipin-containing membranes. Biochemistry
45, 4998-5009.
124. Shidoji, Y., Hayashi, K., Komura, S., Ohishi, N. & Yagi, K. (1999). Loss of molecular interaction
between cytochrome c and cardiolipin due to lipid peroxidation. Biochem Biophys Res Commun
264, 343-7.
125. Korytowski, W., Basova, L. V., Pilat, A., Kernstock, R. M. & Girotti, A. W. (2011). Permeabilization
of the mitochondrial outer membrane by Bax/truncated Bid (tBid) proteins as sensitized by
cardiolipin hydroperoxide translocation: mechanistic implications for the intrinsic pathway of
oxidative apoptosis. J Biol Chem 286, 26334-43.
126. Ott, M., Robertson, J. D., Gogvadze, V., Zhivotovsky, B. & Orrenius, S. (2002). Cytochrome c
release from mitochondria proceeds by a two-step process. Proc Natl Acad Sci U S A 99, 1259-63.
127. Lutter, M., Fang, M., Luo, X., Nishijima, M., Xie, X. & Wang, X. (2000). Cardiolipin provides
specificity for targeting of tBid to mitochondria. Nat Cell Biol 2, 754-61.
128. Gonzalvez, F., Schug, Z. T., Houtkooper, R. H., MacKenzie, E. D., Brooks, D. G., Wanders, R. J.,
Petit, P. X., Vaz, F. M. & Gottlieb, E. (2008). Cardiolipin provides an essential activating platform
for caspase-8 on mitochondria. J Cell Biol 183, 681-96.
References
58
129. Yethon, J. A., Epand, R. F., Leber, B., Epand, R. M. & Andrews, D. W. (2003). Interaction with a
membrane surface triggers a reversible conformational change in Bax normally associated with
induction of apoptosis. J Biol Chem 278, 48935-41.
130. Lucken-Ardjomande, S., Montessuit, S. & Martinou, J. C. (2008). Contributions to Bax insertion
and oligomerization of lipids of the mitochondrial outer membrane. Cell Death Differ 15, 929-37.
131. Satsoura, D., Kucerka, N., Shivakumar, S., Pencer, J., Griffiths, C., Leber, B., Andrews, D. W.,
Katsaras, J. & Fradin, C. (2012). Interaction of the full-length Bax protein with biomimetic
mitochondrial liposomes: A small-angle neutron scattering and fluorescence study. Biochim
Biophys Acta 1818, 384-401.
132. Kuwana, T., Mackey, M. R., Perkins, G., Ellisman, M. H., Latterich, M., Schneiter, R., Green, D. R.
& Newmeyer, D. D. (2002). Bid, Bax, and lipids cooperate to form supramolecular openings in the
outer mitochondrial membrane. Cell 111, 331-42.
133. Shelton, S. N., Dillard, C. D. & Robertson, J. D. (2010). Activation of caspase-9, but not caspase-2
or caspase-8, is essential for heat-induced apoptosis in Jurkat cells. J Biol Chem 285, 40525-33.
134. Gavathiotis, E., Suzuki, M., Davis, M. L., Pitter, K., Bird, G. H., Katz, S. G., Tu, H. C., Kim, H.,
Cheng, E. H., Tjandra, N. & Walensky, L. D. (2008). BAX activation is initiated at a novel
interaction site. Nature 455, 1076-81.
135. Kim, H., Rafiuddin-Shah, M., Tu, H. C., Jeffers, J. R., Zambetti, G. P., Hsieh, J. J. & Cheng, E. H.
(2006). Hierarchical regulation of mitochondrion-dependent apoptosis by BCL-2 subfamilies. Nat
Cell Biol 8, 1348-58.
136. Jabbour, A. M., Heraud, J. E., Daunt, C. P., Kaufmann, T., Sandow, J., O'Reilly, L. A., Callus, B. A.,
Lopez, A., Strasser, A., Vaux, D. L. & Ekert, P. G. (2009). Puma indirectly activates Bax to cause
apoptosis in the absence of Bid or Bim. Cell Death Differ 16, 555-63.
137. Werner, A. B., Tait, S. W., de Vries, E., Eldering, E. & Borst, J. (2004). Requirement for aspartate-
cleaved bid in apoptosis signaling by DNA-damaging anti-cancer regimens. J Biol Chem 279,
28771-80.
138. Bleicken, S., Classen, M., Padmavathi, P. V., Ishikawa, T., Zeth, K., Steinhoff, H. J. & Bordignon,
E. (2009). Molecular details of Bax activation, oligomerization, and membrane insertion. J Biol
Chem 285, 6636-47.
139. Epand, R. F., Martinou, J. C., Montessuit, S., Epand, R. M. & Yip, C. M. (2002). Direct evidence
for membrane pore formation by the apoptotic protein Bax. Biochem Biophys Res Commun 298,
744-9.
140. Saito, M., Korsmeyer, S. J. & Schlesinger, P. H. (2000). BAX-dependent transport of cytochrome c
reconstituted in pure liposomes. Nat Cell Biol 2, 553-5.
141. Dejean, L. M., Martinez-Caballero, S., Guo, L., Hughes, C., Teijido, O., Ducret, T., Ichas, F.,
Korsmeyer, S. J., Antonsson, B., Jonas, E. A. & Kinnally, K. W. (2005). Oligomeric Bax is a
component of the putative cytochrome c release channel MAC, mitochondrial apoptosis-induced
channel. Mol Biol Cell 16, 2424-32.
142. Pavlov, E. V., Priault, M., Pietkiewicz, D., Cheng, E. H., Antonsson, B., Manon, S., Korsmeyer, S.
J., Mannella, C. A. & Kinnally, K. W. (2001). A novel, high conductance channel of mitochondria
linked to apoptosis in mammalian cells and Bax expression in yeast. J Cell Biol 155, 725-31.
143. Dai, Q., Liu, J., Chen, J., Durrant, D., McIntyre, T. M. & Lee, R. M. (2004). Mitochondrial
ceramide increases in UV-irradiated HeLa cells and is mainly derived from hydrolysis of
sphingomyelin. Oncogene 23, 3650-8.
References
59
144. Siskind, L. J., Kolesnick, R. N. & Colombini, M. (2002). Ceramide channels increase the
permeability of the mitochondrial outer membrane to small proteins. J Biol Chem 277, 26796-
803.
145. Perera, M. N., Lin, S. H., Peterson, Y. K., Bielawska, A., Szulc, Z. M., Bittman, R. & Colombini, M.
(2012). Bax and Bcl-xL exert their regulation on different sites of the ceramide channel. Biochem J
445, 81-91.
146. Tsujimoto, Y., Finger, L. R., Yunis, J., Nowell, P. C. & Croce, C. M. (1984). Cloning of the
chromosome breakpoint of neoplastic B cells with the t(14;18) chromosome translocation. Science
226, 1097-9.
147. Tsujimoto, Y., Cossman, J., Jaffe, E. & Croce, C. M. (1985). Involvement of the bcl-2 gene in
human follicular lymphoma. Science 228, 1440-3.
148. Vaux, D. L., Cory, S. & Adams, J. M. (1988). Bcl-2 gene promotes haemopoietic cell survival and
cooperates with c-myc to immortalize pre-B cells. Nature 335, 440-2.
149. Nunez, G., London, L., Hockenbery, D., Alexander, M., McKearn, J. P. & Korsmeyer, S. J. (1990).
Deregulated Bcl-2 gene expression selectively prolongs survival of growth factor-deprived
hemopoietic cell lines. J Immunol 144, 3602-10.
150. Korsmeyer, S. J. (1992). Bcl-2 initiates a new category of oncogenes: regulators of cell death.
Blood 80, 879-86.
151. Reed, J. C. (2008). Bcl-2-family proteins and hematologic malignancies: history and future
prospects. Blood 111, 3322-30.
152. Akao, Y., Otsuki, Y., Kataoka, S., Ito, Y. & Tsujimoto, Y. (1994). Multiple subcellular localization of
bcl-2: detection in nuclear outer membrane, endoplasmic reticulum membrane, and
mitochondrial membranes. Cancer Res 54, 2468-71.
153. Oltvai, Z. N., Milliman, C. L. & Korsmeyer, S. J. (1993). Bcl-2 heterodimerizes in vivo with a
conserved homolog, Bax, that accelerates programmed cell death. Cell 74, 609-19.
154. Tan, C., Dlugosz, P. J., Peng, J., Zhang, Z., Lapolla, S. M., Plafker, S. M., Andrews, D. W. & Lin, J.
(2006). Auto-activation of the apoptosis protein Bax increases mitochondrial membrane
permeability and is inhibited by Bcl-2. J Biol Chem 281, 14764-75.
155. Kim, P. K., Annis, M. G., Dlugosz, P. J., Leber, B. & Andrews, D. W. (2004). During apoptosis bcl-
2 changes membrane topology at both the endoplasmic reticulum and mitochondria. Mol Cell 14,
523-9.
156. Schendel, S. L., Xie, Z., Montal, M. O., Matsuyama, S., Montal, M. & Reed, J. C. (1997). Channel
formation by antiapoptotic protein Bcl-2. Proc Natl Acad Sci U S A 94, 5113-8.
157. Peng, J., Tan, C., Roberts, G. J., Nikolaeva, O., Zhang, Z., Lapolla, S. M., Primorac, S., Andrews, D.
W. & Lin, J. (2006). tBid elicits a conformational alteration in membrane-bound Bcl-2 such that it
inhibits Bax pore formation. J Biol Chem 281, 35802-11.
158. Dlugosz, P. J., Billen, L. P., Annis, M. G., Zhu, W., Zhang, Z., Lin, J., Leber, B. & Andrews, D. W.
(2006). Bcl-2 changes conformation to inhibit Bax oligomerization. EMBO J 25, 2287-96.
159. Peng, J., Ding, J., Tan, C., Baggenstoss, B., Zhang, Z., Lapolla, S. M. & Lin, J. (2009).
Oligomerization of membrane-bound Bcl-2 is involved in its pore formation induced by tBid.
Apoptosis 14, 1145-53.
160. Garcia-Saez, A. J., Ries, J., Orzaez, M., Perez-Paya, E. & Schwille, P. (2009). Membrane promotes
tBID interaction with BCL(XL). Nat Struct Mol Biol 16, 1178-85.
161. Lin, B., Kolluri, S. K., Lin, F., Liu, W., Han, Y. H., Cao, X., Dawson, M. I., Reed, J. C. & Zhang, X.
K. (2004). Conversion of Bcl-2 from protector to killer by interaction with nuclear orphan receptor
Nur77/TR3. Cell 116, 527-40.
References
60
162. Grandgirard, D., Studer, E., Monney, L., Belser, T., Fellay, I., Borner, C. & Michel, M. R. (1998).
Alphaviruses induce apoptosis in Bcl-2-overexpressing cells: evidence for a caspase-mediated,
proteolytic inactivation of Bcl-2. EMBO J 17, 1268-78.
163. Kim, K. M., Giedt, C. D., Basanez, G., O'Neill, J. W., Hill, J. J., Han, Y. H., Tzung, S. P.,
Zimmerberg, J., Hockenbery, D. M. & Zhang, K. Y. (2001). Biophysical characterization of
recombinant human Bcl-2 and its interactions with an inhibitory ligand, antimycin A.
Biochemistry 40, 4911-22.
164. Demetzos, C. (2008). Differential Scanning Calorimetry (DSC): a tool to study the thermal
behavior of lipid bilayers and liposomal stability. J Liposome Res 18, 159-73.
165. Wagner, G. (1997). An account of NMR in structural biology. Nat Struct Biol 4 Suppl, 841-4.
166. Auger, M. (1997). Membrane structure and dynamics as viewed by solid-state NMR spectroscopy.
Biophys Chem 68, 233-41.
167. Sparrman, T. & Westlund, P. O. (2003). An NMR line shape and relaxation analysis of heavy water
powder spectra of the L-alpha, L-beta ' and P-beta ' phases in the DPPC/water system. Phys Chem
Chem Phys 5, 2114-21.
168. Kelly, S. M., Jess, T. J. & Price, N. C. (2005). How to study proteins by circular dichroism. Biochim
Biophys Acta 1751, 119-39.
169. Czarnik-Matusewicz, B. & Pilorz, S. (2006). 2DCOS and MCR-ALS as a combined tool of analysis
of beta-lactoglobulin CD spectra. J Mol Struct 799, 211-220.
170. Kihara, D. & Kanehisa, M. (2000). Tandem clusters of membrane proteins in complete genome
sequences. Genome Res 10, 731-43.
171. Wallin, E. & von Heijne, G. (1998). Genome-wide analysis of integral membrane proteins from
eubacterial, archaean, and eukaryotic organisms. Protein Sci 7, 1029-38.
172. Grisshammer, R. (2006). Understanding recombinant expression of membrane proteins. Curr
Opin Biotechnol 17, 337-40.
173. Grobner, G., Taylor, A., Williamson, P. T., Choi, G., Glaubitz, C., Watts, J. A., de Grip, W. J. &
Watts, A. (1997). Macroscopic orientation of natural and model membranes for structural studies.
Anal Biochem 254, 132-8.
174. Schwarz, D., Dotsch, V. & Bernhard, F. (2008). Production of membrane proteins using cell-free
expression systems. Proteomics 8, 3933-46.
175. Carlson, E. D., Gan, R., Hodgman, C. E. & Jewett, M. C. (2012). Cell-free protein synthesis:
Applications come of age. Biotechnol Adv 30, 1185-94.
176. Isaksson, L., Enberg, J., Neutze, R., Goran Karlsson, B. & Pedersen, A. (2012). Expression
screening of membrane proteins with cell-free protein synthesis. Protein Expr Purif 82, 218-25.
177. Pedersen, A., Hellberg, K., Enberg, J. & Karlsson, B. G. (2010). Rational improvement of cell-free
protein synthesis. N Biotechnol 28, 218-24.
178. Pedersen, A., Wallgren, M., Karlsson, B. G. & Grobner, G. (2011). Expression and purification of
full-length anti-apoptotic Bcl-2 using cell-free protein synthesis. Protein Expr Purif 77, 220-3.
179. Lis, M., Wizert, A., Przybylo, M., Langner, M., Swiatek, J., Jungwirth, P. & Cwiklik, L. (2011). The
effect of lipid oxidation on the water permeability of phospholipids bilayers. Phys Chem Chem
Phys 13, 17555-63.
180. Mannock, D. A., Lewis, R. N. & McElhaney, R. N. (2010). A calorimetric and spectroscopic
comparison of the effects of ergosterol and cholesterol on the thermotropic phase behavior and
organization of dipalmitoylphosphatidylcholine bilayer membranes. Biochim Biophys Acta 1798,
376-88.
References
61
181. Dufourc, E. J., Mayer, C., Stohrer, J., Althoff, G. & Kothe, G. (1992). Dynamics of phosphate head
groups in biomembranes. Comprehensive analysis using phosphorus-31 nuclear magnetic
resonance lineshape and relaxation time measurements. Biophys J 61, 42-57.
182. Rog, T., Murzyn, K., Milhaud, J., Karttunen, M. & Pasenkiewicz-Gierula, M. (2009). Water Isotope
Effect on the Phosphatidylcholine Bilayer Properties: A Molecular Dynamics Simulation Study. J
Phys Chem B 113, 2378-87.
183. Dzwolak, W., Ravindra, R., Lendermann, J. & Winter, R. (2003). Aggregation of bovine insulin
probed by DSC/PPC calorimetry and FTIR spectroscopy. Biochemistry 42, 11347-55.
184. Veiga, M. P., Arrondo, J. L., Goni, F. M., Alonso, A. & Marsh, D. (2001). Interaction of cholesterol
with sphingomyelin in mixed membranes containing phosphatidylcholine, studied by spin-label
ESR and IR spectroscopies. A possible stabilization of gel-phase sphingolipid domains by
cholesterol. Biochemistry 40, 2614-22.
185. Fletcher, J. I., Meusburger, S., Hawkins, C. J., Riglar, D. T., Lee, E. F., Fairlie, W. D., Huang, D. C.
& Adams, J. M. (2008). Apoptosis is triggered when prosurvival Bcl-2 proteins cannot restrain
Bax. Proc Natl Acad Sci U S A 105, 18081-7.
186. Tyurin, V. A., Tyurina, Y. Y., Kochanek, P. M., Hamilton, R., DeKosky, S. T., Greenberger, J. S.,
Bayir, H. & Kagan, V. E. (2008). Oxidative lipidomics of programmed cell death. Methods
Enzymol 442, 375-93.
187. Jiang, J., Huang, Z., Zhao, Q., Feng, W., Belikova, N. A. & Kagan, V. E. (2008). Interplay between
bax, reactive oxygen species production, and cardiolipin oxidation during apoptosis. Biochem
Biophys Res Commun 368, 145-50.
188. Esposti, M. D., Erler, J. T., Hickman, J. A. & Dive, C. (2001). Bid, a widely expressed proapoptotic
protein of the Bcl-2 family, displays lipid transfer activity. Mol Cell Biol 21, 7268-76.
189. Luo, X., Budihardjo, I., Zou, H., Slaughter, C. & Wang, X. (1998). Bid, a Bcl2 interacting protein,
mediates cytochrome c release from mitochondria in response to activation of cell surface death
receptors. Cell 94, 481-90.
190. Brand, M. D. (2010). The sites and topology of mitochondrial superoxide production. Exp
Gerontol 45, 466-72.
191. Latchoumycandane, C., Marathe, G. K., Zhang, R. & McIntyre, T. M. (2012). Oxidatively truncated
phospholipids are required agents of tumor necrosis factor alpha (TNFalpha)-induced apoptosis. J
Biol Chem 287, 17693-705.
192. Hatfield, G. W. & Roth, D. A. (2007). Optimizing scaleup yield for protein production:
Computationally Optimized DNA Assembly (CODA) and Translation Engineering. Biotechnol
Annu Rev 13, 27-42.
193. Gutman, G. A. & Hatfield, G. W. (1989). Nonrandom utilization of codon pairs in Escherichia coli.
Proc Natl Acad Sci U S A 86, 3699-703.
194. Smith, D. & Yarus, M. (1989). tRNA-tRNA interactions within cellular ribosomes. Proc Natl Acad
Sci U S A 86, 4397-401.
195. Irwin, B., Heck, J. D. & Hatfield, G. W. (1995). Codon pair utilization biases influence translational
elongation step times. J Biol Chem 270, 22801-6.
196. Klammt, C., Schwarz, D., Eifler, N., Engel, A., Piehler, J., Haase, W., Hahn, S., Dotsch, V. &
Bernhard, F. (2007). Reprint of "Cell-free production of G protein-coupled receptors for
functional and structural studies" [J. Struct. Biol. 158 (2007) 482-493]. J Struct Biol 159, 194-
205.
References
62
197. Antonsson, B., Montessuit, S., Lauper, S., Eskes, R. & Martinou, J. C. (2000). Bax oligomerization
is required for channel-forming activity in liposomes and to trigger cytochrome c release from
mitochondria. Biochem J 345, 271-8.
198. Hsu, Y. T. & Youle, R. J. (1998). Bax in murine thymus is a soluble monomeric protein that
displays differential detergent-induced conformations. J Biol Chem 273, 10777-83.
199. Ivashyna, O., Garcia-Saez, A. J., Ries, J., Christenson, E. T., Schwille, P. & Schlesinger, P. H.
(2009). Detergent-activated BAX protein is a monomer. J Biol Chem 284, 23935-46.
200. Bleicken, S. & Zeth, K. (2009). Conformational changes and protein stability of the pro-apoptotic
protein Bax. J Bioenerg Biomembr 41, 29-40.
201. Tzung, S. P., Kim, K. M., Basanez, G., Giedt, C. D., Simon, J., Zimmerberg, J., Zhang, K. Y. &
Hockenbery, D. M. (2001). Antimycin A mimics a cell-death-inducing Bcl-2 homology domain 3.
Nat Cell Biol 3, 183-91.
202. Rigaud, J. L. (2002). Membrane proteins: functional and structural studies using reconstituted
proteoliposomes and 2-D crystals. Braz J Med Biol Res 35, 753-66.
203. Antharavally, B. S., Mallia, K. A., Rosenblatt, M. M., Salunkhe, A. M., Rogers, J. C., Haney, P. &
Haghdoost, N. (2011). Efficient removal of detergents from proteins and peptides in a spin column
format. Anal Biochem 416, 39-44.