Cytochrome c Oxidase dysfunction in cancer Exploring the molecular mechanisms

Ida Namslauer

©Ida Namslauer, Stockholm 2012 ISBN 978-91-7447-424-4 (pages 1-65) Printed in Sweden by Universitetsservice US-AB, Stockholm 2012 Distributor: Department of Biochemistry and Biophysics, Stockholm University

List of publications This thesis is based on the following publications, which will be referred to by their Roman numerals: I. Laura S Busenlehner, Gisela Brändén, Ida Namslauer, Peter

Brzezinski and Richard N Armstrong. Structural elements involved in

proton translocation by Cytochrome c Oxidase as revealed by

backbone amide hydrogen-deuterium exchange of the E286H mutant.

Biochemistry 2008, 47:73-83.

II. Ida Namslauer, Huyn-Ju Lee, Robert B Gennis and Peter Brzezinski.

A pathogenic mutation in Cytochrome c Oxidase results in impaired

proton pumping while retaining O2-reduction activity. Biochimica et

Biophysica Acta 2010, 1797:550-556.

III. Ida Namslauer and Peter Brzezinski. A mitochondrial DNA mutation

linked to colon cancer results in proton leaks in Cytochrome c Oxidase.

Proceedings of the National Academy of Sciences of the United States

of America 2009, 106:3402-3407.

IV. Ida Namslauer, Marina S Dietz and Peter Brzezinski. Functional

effects of mutations in Cytochrome c Oxidase related to prostate

cancer. Biochimica et Biophysica Acta 2011, 1807:1336-1341.

Additional publication

Suman Chakrabarty, Ida Namslauer, Peter Brzezinski and Arieh

Warshel. Exploration of the Cytochrome c Oxidase pathway puzzle

and examination of the origin of elusive mutational effects. Biochimica

et Biophysica Acta 2011, 1807:413-426.

Contents

Introduction ..................................................................................................9 Metabolism and oxidative phosphorylation ..........................................................9

Oxidative phosphorylation................................................................................10 The mitochondrial respiratory chain ...............................................................12

Cytochrome c Oxidase..............................................................................13 Structure and function of R. sphaeroides CytcO................................................14

General structural features ..............................................................................14 Function ...............................................................................................................15 Proton pumping in CytcO..................................................................................17 Methods for studying CytcO function .............................................................18

Functionally important structural features of CytcO .........................................19 The D pathway for proton uptake ...................................................................19 Glu286: an internal H+ donor/acceptor .........................................................20 Structural changes in the D pathway affect Glu286....................................21 The proton exit route beyond Glu286 ............................................................22

The mammalian CytcO ...........................................................................................24 Similarities between the mammalian and the R. sphaeroides CytcO .......24 The additional subunits of the mammalian CytcO .......................................26 An additional proton transfer pathway...........................................................27

The aim of this thesis ...............................................................................28

Modeling mtDNA mutations in the R. sphaeroides CytcO ................29 Mitochondrial DNA mutations ..........................................................................29

Model systems of mtDNA mutations....................................................................30 Transmitochondrial hybrid cell lines ...............................................................31 Yeast models.......................................................................................................31 Mouse models .....................................................................................................32 The relevance of the R. sphaeroides CytcO model ......................................33

CytcO amino-acid substitutions in cancer............................................35 Reports on CytcO defects in cancer................................................................35

Functional characterization of the pathogenic substitutions ...........................37 Effects on the CytcO catalytic activity............................................................38

Effects on proton pumping ...............................................................................40 The results correlate between mammalian and R. sphaeroides CytcO ....42 Comments on the results .................................................................................43

The relevance of CytcO substitutions found in cancer ......................44 Energy metabolism of cancer cells .................................................................45 Oxidative stress and apoptosis in cancer ......................................................46 The mitochondrial electrochemical gradient in cancer cells .......................47

Concluding remarks and future perspectives ......................................49

Sammanfattning på svenska ..................................................................50

Acknowledgements ...................................................................................52

References ..................................................................................................54

9

Introduction

Mutations in the mitochondrial DNA (mtDNA) have been found in

connection to various types of human cancer. Since the mtDNA encode

several polypeptides of the respiratory-chain enzymes, mtDNA mutations

often affect the function of oxidative phosphorylation. Some of the identified

mutations cause amino-acid substitutions in the enzyme Cytochrome c

Oxidase (CytcO). CytcO is an essential enzyme in oxidative phosphorylation

and it is found in almost all eukaryotic organisms as well as in many

bacterial and archaeal species. Although amino-acid substitutions in CytcO

have been found in human cancer cells, the impact of the substitutions on the

function of the enzyme has in most cases not been verified.

We have investigated, on a molecular level, the functional effects of

substitutions in CytcO related to colon- and prostate cancer. For these

studies the well-characterized bacterial CytcO from the model organism

Rhodobacter sphaeroides was used. The studies have provided detailed

insights into how the function of CytcO is affected by the mtDNA mutations

originally reported in connection to cancer.

In this thesis the relevance of using a bacterial model system when

investigating molecular mechanisms of diseases found in humans is

discussed. Further, possible links between a defectively functioning CytcO

and the development of cancer are proposed. First, however, there is some

background information on oxidative phosphorylation and on the structure

and function of both R. sphaeroides and mammalian CytcO.

Metabolism and oxidative phosphorylation

Every living cell needs energy. The source of energy that is used by a cell

varies depending on for example the cell type and its environment. Humans,

like all mammals and other animals, use chemical energy derived from food.

In every cell the energy needs to be converted in order to be accessible for

10

energy-requiring processes such as growth, organization, transport and

reproduction. Metabolism is a set of chemical reactions involved in the

uptake, conversion, digestion and excretion of energy-containing nutrients.

The key players of metabolism, catalyzing the chemical reactions, are

enzymes. In eukaryotic cells, mitochondria are the cellular compartments

(organelles) that play a key role in energy metabolism. Mitochondria contain

the enzyme complexes of metabolic processes such as the citric acid cycle,

the fatty acid oxidation and oxidative phosphorylation. Therefore, most of

the adenosine triphosphate (ATP), which is the molecule essential for many

energy-requiring reactions in the cell, is generated in mitochondria.

Mitochondria also contain their own genetic material; mitochondrial DNA

(mtDNA).

Oxidative phosphorylation

The process of producing ATP using energy released by oxidation of

nutrient molecules is called oxidative phosphorylation. In this process,

electrons (derived from nutrients) are transferred from electron donors to

electron acceptors in redox reactions. The reactions are catalyzed by a

number of membrane-bound enzyme complexes constituting the respiratory

chain (figure 1). The flow of electrons from an electron donor having a low

midpoint potential to an electron acceptor with a higher midpoint potential is

exergonic, releasing free energy. The energy is used by the enzyme

complexes to translocate protons across the inner mitochondrial membrane

in eukaryotes or the plasma membrane in bacteria and archaea. Thus, the

energy is transiently stored in the electrochemical proton gradient. The

exergonic flow of protons down the gradient is allowed through the enzyme

ATP synthase in which the proton flow is coupled to the phosphorylation of

adenosine diphosphate (ADP), producing ATP. The ATP is then transported

out of the mitochondrion to be used in various energy-requiring processes in

the cell.

Compared to glycolysis or lactic-acid fermentation, oxidative

phosphorylation is a highly efficient way of utilizing chemically stored

energy and is a process that is used by almost all aerobic organisms. In

mammalian cells with functional mitochondria it has been estimated that

approximately 90 % of the total cellular ATP is produced by oxidative

11

phosphorylation [1]. However, oxidative phosphorylation is also involved in

the production of potentially harmful reactive oxygen species (ROS). During

electron transfer between the complexes of the respiratory chain a small

amount of ROS is constantly being formed, mainly in respiratory chain

complexes I and III [2-4]. Normally, cells have ways of eliminating ROS.

Several enzymes, such as superoxide dismutase, peroxidases and catalase,

are involved in ROS detoxification. In addition, small antioxidant molecules

such as ascorbic acid and tocopherols are important in reducing the amount

of ROS in cells. Although ROS have been suggested to be important in

certain cell-signaling processes, an excess production of ROS (oxidative

stress) is harmful to cells, leading to damage of DNA, proteins and lipids.

Oxidative stress has been linked to the normal aging process as well as to

diseases such as cancer and Alzheimer’s disease [5-8]

Figure 1. The enzyme complexes of the mitochondrial respiratory chain; NADH dehydrogenase (I), succinate dehydrogenase (II), the bc1 complex (III) and CytcO (IV), together with ATP synthase. Electrons are provided through the oxidation of NADH and succinate. Quinone (Q) and cytochrome c (cyt c) are electron carriers. Complexes I, III and IV contribute to the establishment of the electrochemical pro-ton gradient by translocating protons from the negative side of the membrane to the positive. Protons flow down the gradient through ATP synthase, resulting in the production of ATP from ADP and inorganic phosphate (Pi ). Electron transfers are indicated by yellow arrows and protons by red arrows. The position of the mem-brane is indicated in yellow.

12

The mitochondrial respiratory chain

There is a general organization of the mitochondrial respiratory chain in

most eukaryotes. The mammalian respiratory chain consists of four enzyme

complexes; NADH dehydrogenase, succinate dehydrogenase, the bc1

complex and CytcO (figure 1). All four enzymes contain redox-active

cofactors involved in the transfer of electrons from an electron donor to an

acceptor. The electron donors of the respiratory chain are NADH and

FADH2. Quinone and cytochrome c act as electron carriers, transferring

electrons from one enzyme complex to another. The final electron acceptor

of the mammalian respiratory chain is molecular oxygen, which is reduced

to water at the catalytic site of CytcO. In mammalian cells the protein

subunits of NADH dehydrogenase, the bc1 complex, CytcO and ATP

synthase are encoded both in the nuclear genome and in the mtDNA.

Traditionally, it was believed that the enzyme complexes diffuse freely in

the membrane and that electron transfer would occur upon random collisions

[9]. A more recent suggestion is that the respiratory-chain complexes form

large supercomplexes. Such supercomplexes have been found in mammals

as well as in yeast and bacteria [10-12] (figure 2). An organization of the

enzymes into larger complexes has been suggested to allow greater structural

stability of the individual enzymes, efficient substrate channeling and a

minimum formation of ROS, as compared to a scenario where the individual

enzymes are located at some distance from each other.

Figure 2. Cryo-EM 3D map of a purified supercomplex from bovine heart mito-chondria, with the docked x-ray crystal structures of complex I (blue), complex III (red) and complex IV (green). The figure is modified from [10] and is reprinted with permission.

13

Cytochrome c Oxidase

Cytochrome c Oxidase (CytcO) is the terminal oxidase of the respiratory

chain in mitochondria and many bacteria and archaea. CytcO was discovered

in the 1930’s as a member of the respiratory chain. It was mainly the work of

Otto Warburg and David Keilin that established its role in metabolism [13,

14]. In the 1970’s and 1980’s the knowledge on CytcO was greatly increased

and in 1977 it was shown to be a proton pump [15].

Figure 3. The structure of the bacterial CytcO from R. sphaeroides (PDB 1M56). The four subunits are shown in different colours: Subunit I in yellow, subunit II in magenta, subunit III in orange and subunit IV in pink. The heme groups; heme a and heme a3, are shown in green and the cupper cofactors as orange spheres. The black line indicate the approximate position of the lipid bilayer. The figure was prepared using the PyMOL software.

14

About 15 years ago, the first x-ray crystal structures of CytcO were

published. These were the mammalian CytcO from bovine heart

mitochondria [16, 17] and the bacterial CytcO from Paracoccus

denitrificans [18]. A few years later, the R. sphaeroides CytcO x-ray crystal

structure was determined [19]. The following chapter describes features of

the bacterial CytcO from R. sphaeroides. Then, there is a comparison

between the bacterial and the mammalian enzyme.

Structure and function of R. sphaeroides CytcO

General structural features

The available 3D-structures of the R. sphaeroides CytcO have revealed an

enzyme complex containing four protein subunits [19] (figure 3). Subunit I

harbors three of the four redox-active cofactors; heme a and the heme a3-CuB

catalytic site. In subunit I, two proton transfer pathways leading from the

negatively charged (N) side of the membrane to the vicinity of the catalytic

site have been identified [19, 20]. Subunit II contains two transmembrane α-

helices and an extramembrane β-barrel domain where the additional copper

cofactor; CuA resides. The cytochrome c-binding site is presumably found in

subunit II [21, 22]. The interaction between reduced cytochrome c and

CytcO occurs through hydrophobic as well as electrostatic interactions

between positively charged Lysine residues on cytochrome c and negatively

charged residues on CytcO. Subunit III consists of seven membrane-

spanning α-helices. Although subunit III does not contain any redox-active

cofactors it is essential for the function of the enzyme. Removal of subunit

III results in defective proton pumping [23, 24]. There is also experimental

evidence showing that removal of subunit III leads to turnover-induced

inactivation of the enzyme [25]. Subunit IV of the R. sphaeroides CytcO

consists of one single transmembrane α-helix. It is connected to subunits I-

III and held in position only via lipid molecules [19].

15

Figure 4. Electron and proton transfers to the R. sphaeroides CytcO catalytic site. The heme and cupper cofactors are shown in green and orange, respectively. Elec-trons are transferred via CuA and heme a to the heme a3-CuB catalytic site. Oxygen binds to the reduced heme a3 iron. Protons are transferred via specific pathways to the catalytic site. For the reduction of one oxygen molecule, four protons and four electrons are required. In addition, one proton per electron transferred to the cata-lytic site is pumped through the enzyme. Subunit I of the CytcO is in yellow and subunit II in pink. The figure was prepared using the PDB entry 1M56 and the Py-MOL software.

Function

CytcO catalyses the reduction of oxygen to water. For this reaction to take

place electrons, protons and oxygen need to reach the catalytic site (figure

4). The electrons are donated one-by-one from reduced cytochrome c and are

transferred via the redox-active cofactors CuA and heme a to the CytcO

catalytic site [26]. Oxygen binds to the reduced heme a3 iron at the catalytic

site. The electron transfer is coupled to proton uptake to the catalytic site and

oxygen is reduced to water [27, 28]. Part of the energy available from the

exergonic reduction of oxygen is used by the enzyme to pump four protons

per oxygen molecule from the N side of the membrane to the positively

charged (P) side, contributing to the formation of the electrochemical

gradient [15].

Since the electrons needed for oxygen reduction are transferred one-by-

one to CytcO during turnover, the catalytic site cycles through a number of

16

partly reduced intermediate states, identified spectroscopically by their

different redox states and ligand-binding properties [29]. These are

illustrated schematically in the CytcO catalytic cycle (figure 5). In brief,

oxygen binds to the two-electron reduced (R2) catalytic site. Breakage of the

O-O bond forms the P2 intermediate. The next intermediate, F3, is formed

when an electron and a proton is transferred to the catalytic site. The F3

intermediate decays into the fully oxidized O 4(0) state when another electron

and a proton is transferred to the catalytic site. Subsequent transfer of two

protons and two electrons results in formation of water and reduction of the

catalytic site. Protons are pumped upon formation of the F3, O4(0) and R2

intermediate states. If the CytcO is preloaded with four electrons (all

cofactors are reduced) the oxidative part of the catalytic cycle (R2 to O4(0))

can be studied spectroscopically using the flow-flash technique [30]. In that

case, there is electron transfer from heme a to the catalytic site upon

formation of the P3 intermediate. Thus, the electron and proton transfer to the

catalytic site required for formation of the F3 intermediate are separated in

time and can be studied individually.

17

Figure 5. (a). A schematic illustration of the catalytic cycle of O2 reduction in CytcO. The redox intermediate states (reduced; R, peroxy; P, ferryl; F and oxidized; O) are shown as well as the events taking place in the transitions between the states. The digits in superscript indicate the number of electrons present at the catalytic site. (b). The sequence of events observed when starting with the fully reduced CytcO, as in the flow-flash absorption spectroscopy. The enzyme cofactors are shown as spheres and the colours indicate an oxidized (grey) or a reduced (red) cofactor. The catalytic cycle is further described in the text.

Proton pumping in CytcO

For proton pumping to be achieved, there are some basic requirements that

need to be fulfilled. First, there has to be a proton pathway all the way

through the enzyme. This pathway can, however, not be open to both sides

of the membrane simultaneously since that would result in proton leaks and

disruption of the electrochemical gradient. Thus, there has to be a gating

mechanism ensuring that the proton transfer is unidirectional. The gating

mechanism could, for example, be one or several amino-acid residues

alternating between two different positions. There are also suggestions for an

acceptor site for pumped protons in CytcO. This acceptor site (proton

loading site) is a yet unidentified protonatable group situated above the

18

hemes. Most probably, there is a connection between the proton gate and the

proton loading site, so that the proton loading site changes its protonation

state depending on the position of the gate. Such a mechanism implies that

the pKa of the proton loading site is high when there is a connection to the N

side and low when there is a connection to the P side [31, 32]. Experimental

evidence indicate that the proton loading site is located in the area near the

propionates of heme a and heme a3 [33, 34].

In CytcO, the energy needed for the pumping of protons against the

electrochemical gradient is provided by the exergonic reduction of oxygen.

As soon as the electrons and protons needed for the reaction reach the

catalytic site, the free energy is lost. Therefore, the events that drive proton

pumping must occur before that. There are several ways in which this could

be accomplished: The proton to be pumped could be transferred to the

proton loading site before the electron and substrate proton reach the

catalytic site or the transfer of the substrate proton to the catalytic site is

coupled to a reaction (e.g. a structural change) that conserves the free

energy. Then the transfer of the pumped proton would be triggered by the

structural change. The mechanism for proton pumping in CytcO is still not

fully revealed, but is reviewed in [31, 35].

Methods for studying CytcO function

The function of CytcO has been studied for several decades. Optical

absorption spectroscopy, together with other spectroscopic techniques, such

as Fourier Transform Infrared Spectroscopy (FTIR) and Electron

Paramagnetic Resonance Spectroscopy (EPR), are valuable tools for

functional studies of CytcO [32, 36-38]. Site-directed mutagenesis of the

bacterial CytcO allows substitution of specific amino-acid residues. In

combination with spectroscopic techniques, mutational studies have offered

insights into CytcO function and provided information on what parts of the

protein are involved in specific processes. The atomic resolution 3D

structures of CytcO are of great importance for the interpretation of results

from functional studies since they can be used to link the function of the

enzyme to the structure. Lately, computational tools such as electrostatic

calculations and molecular-dynamics simulations have been used, especially

for the mammalian CytcO, to investigate processes such as shifts in pKa

19

values of titratable groups and small structural changes taking place within

the enzyme [39-41]. Quantum chemistry tools are also used and are valuable

for evaluating whether a proposed mechanistic event is energetically

favorable or even possible [42]. In general, results from computer

simulations and calculations agree with experimental findings.

Functionally important structural features of CytcO Functional studies, using site-directed mutagenesis, as well as investigations

of the x-ray crystal structures have identified specific regions of CytcO that

are important for the function of the enzyme. Those are, for example, two

proton uptake pathways in subunit I, a highly conserved Glutamate residue

(Glu286) close to the catalytic site and a region close to the heme groups,

towards the positive side of the membrane, involved in the exit of pumped

protons [19, 20, 33, 34, 43-45]. Most of the amino-acid residue substitutions

investigated in the studies presented in this thesis are localized to the above-

mentioned areas, which are described in more detail below.

The D pathway for proton uptake

Since protons are charged they need specific transport pathways in a protein

interior. Such pathways consist of water molecules and polar amino-acid

residues, forming a hydrogen-bonded chain along which protons can travel.

In the R. sphaeroides CytcO, two input pathways for protons have been

identified [19, 20]. One of them is called the D pathway, named after an

Aspartic acid residue (Asp132) closest to the entrance. This pathway is used

for the transfer of protons needed for the reduction of oxygen as well as for

the pumped protons. The D pathway stretches along a distance of ~24 Å

from the N side of the membrane to the Glutamate residue (Glu286) near the

catalytic site (figure 6). According to the x-ray crystal structure ~10 water

molecules are found within the pathway. These molecules are hydrogen

bonded to a number of amino-acid residues along the path [19, 46]. Using

site-directed mutagenesis, it has been shown that these residues are

important for the efficient transfer of protons through the D pathway [20, 47-

50]. In addition to the D pathway, there is a second proton-uptake pathway

20

called the K pathway, which is suggested to be used for proton uptake during

reduction of the catalytic site [30].

Figure 6. The D pathway is used for delivery of protons to the catalytic site and for the uptake of pumped protons. It consists of approximately 10 water molecules (shown as red spheres) and several polar amino-acid residues forming a hydrogen-bonded chain from the N side of the membrane to the Glutamic acid residue (Glu286) near the catalytic site. The cluster of water molecules and the two Arginine residues (Arg481 and Arg482), located above the heme groups, are involved in the transfer of pumped protons. The heme groups are in green and the cupper cofactors as orange spheres. The CytcO subunit I is shown in yellow and subunit II in pink. The figure was prepared using the 1M56 PDB entry and the PymMOL software.

Glu286: an internal H+ donor/acceptor

The Glu286 residue is situated close to the catalytic site (figure 6). In wild-

type CytcO, proton pumping and proton transfer to the catalytic site takes

place upon formation of the F and O intermediates, as well as upon reduction

21

of the oxidized CytcO [32] (figure 5). Site-directed mutagenesis studies of

the R. sphaeroides CytcO in combination with spectroscopic techniques

have shown that the substitution of Glu286 to Glutamine, Alanine or

Histidine impairs proton transfer to the catalytic site as well as proton

pumping [43, 51] (paper I). The Glu286 residue is considered to be a

branching point, from where protons can be transferred either to the catalytic

site or to a proton loading site. In addition, Glu286 (or water molecules

around this residue) is suggested to function as a transient donor/acceptor for

protons. It is believed that protons can be transferred from Glu286 to the

catalytic site and that Glu286 is then quickly reprotonated from the N side.

Results from investigations of the internal proton transfer from Glu286 to the

catalytic site indicate that the pKa of Glu286 is 9.4 [52].

There are suggestions that the Glu286 residue can adopt two different

conformations and that the structural change of Glu286 is involved in

directing protons to the catalytic site and to the proton loading site [53]. This

mechanism was based on information obtained from CytcO x-ray crystal

structures. A comparison of the wild-type and the Glu286Gln CytcO crystal

structures showed that the Glu286 side chain adopted different

conformations in the two structures [19]. This conformational change of

Glu286 also conferred structural changes in the region around the hemes and

two conserved Arginine residues. A proton pumping mechanism involving

the conformational change of Glu286 was proposed [53]. More recently, a

conformational change of the equivalent Glutamate residue was observed in

the crystal structure of P. denitrificans CytcO harboring an amino-acid

substitution in the D pathway [54]. In this case, the proton pumping was

abolished and it was suggested that the reason for uncoupling was the altered

conformation of the Glutamate side chain resulting from the amino-acid

substitution in the D pathway.

Structural changes in the D pathway affect Glu286

Experimental evidence show that substitutions of specific amino-acid

residues in the D pathway result in both uncoupling of proton pumping and a

changed apparent pKa of the Glu286 residue [48, 50], but with retained

oxygen reduction activity. From these results, it was suggested that the

amino-acid substitutions change the hydrogen-bonded network around the

22

Glu286 residue. As a result, protons would still be transferred to the catalytic

site but not to the proton loading site. In addition, molecular-dynamics

studies based on x-ray crystal structures of CytcO have identified a cluster of

water molecules hydrogen-bonded to two Serine residues (Ser200 and

Ser201) in the D pathway [40, 55]. This region was suggested to be a

“proton trap” allowing fast protonation of Glu286, which is consistent with

the high pKa of Glu286. To further investigate the interactions between

amino-acid residues and water molecules in the D pathway and their role in

the proton-pumping mechanism, the Serine residues were substituted for

Valines. Absorption spectroscopy studies of the CytcO variants showed that

the substitutions resulted in a decreased steady-state activity, lowered

stoichiometry of proton pumping and slowed proton transfer from Glu286 to

the catalytic site [47]. The rate of the internal proton transfer to the catalytic

site was independent on the pH and the pKa of Glu286 was suggested to

increase to >12 in the Ser200Val/Ser201Val CytcO.

The proton exit route beyond Glu286

A specific pathway for pumped protons from Glu286 towards the P side of

the membrane has not been identified. However, there is a hydrogen-bonded

network observed in the area above the heme groups [19, 33, 34, 56] (figure

6). This region contains water molecules and two Arginine residues (Arg481

and Arg482) that interact electrostatically with the D-propionates of heme a

and heme a3. Spectroscopic studies have shown that the Arg481 and 482

residues are important for the function of CytcO [33, 34, 56]. Based on the

experimental data on CytcO in which the Arg481 was exchanged for a

Lysine, it was suggested that the proton loading site is the heme a3 D-

propionate [33]. In another study, though, the identity of the proton loading

site was suggested to be the heme a A-propionate [34]. Although a specific

group acting as a proton loading site has not been identified, the area around

the heme propionates and the Arginines are likely to be involved in the

proton pumping activity.

In paper I, structural elements involved in proton pumping were

investigated using amide hydrogen/deuterium exchange followed by mass

spectrometry. The study lead to the suggestion that a region of CytcO

involved in gating of pumped protons resides in a loop consisting of amino-

23

acid residues 169-175 (the gating loop) [57]. With the R. sphaeroides wild-

type CytcO, the gating loop showed a different amide H/D exchange pattern

in the F intermediate of the CytcO catalytic cycle compared to the P and O

intermediates, indicating that a conformational change takes place upon

formation of the F intermediate. Since proton pumping occurs before as well

as after formation of the F intermediate, a conformational change of the 169-

175 loop was suggested to be involved in proton gating, ensuring

unidirectional proton transfer. With the CytcO in which the Glu286 was

substituted for a Histidine, the amide H/D exchange pattern of the gating

loop was similar in all intermediate states of the catalytic cycle (paper I). In

addition, a peptide of residues 320-340, suggested to be part of a proton exit

pathway, showed dramatically decreased deuterium incorporation with the

Glu286His CytcO compared to the wild-type enzyme. The Glu286His

CytcO is unable to pump protons. Presumably, the Glu286His substitution

prevents conformational changes required for proton pumping. Since the

substitution of the Glu286 residue leads to conformational changes in the

fairly distant proton gating region of the enzyme, the results from paper I

indicate that there is a connection, probably formed by hydrogen bonds,

between Glu286 and the suggested proton gating region.

24

The mammalian CytcO

Now, the structure and function of the bacterial CytcO from R. sphaeroides

have been described. Since in these studies, the R. sphaeroides CytcO has

been used as a model system for investigations of amino-acid residue

substitutions originally identified in the human enzyme, the mammalian

CytcO from bovine heart mitochondria (figure 7) will now be described in

terms of structure and function. Overall, the mammalian and the R.

sphaeroides CytcOs are functionally as well as structurally similar.

Figure 7. The structure of the mammalian CytcO from bovine heart mitochondria (PDB entry 10CC). The mammalian CytcO consists of 13 subunits illustrated in different colours. The figure was prepared using the PyMOL software.

Similarities between the mammalian and the R. sphaeroides

CytcO

The amino-acid sequence identities of the R. sphaeroides CytcO compared

to the mammalian enzyme are 52 % (subunit I), 39 % (subunit II) and 49 %

(subunit III) [58], which is considered a high sequence identity. These

numbers are comparable to subunit I of P. denitrificans NADH

25

dehydrogenase, having a 64 % sequence identity to the mammalian

counterpart [59]. The ATP synthase from E. coli is, with a 70 % amino-acid

sequence identity, even slightly more similar to the mammalian enzyme

[60]. The structures of subunits I, II and III of the mammalian CytcO, which

are considered to be the catalytically active core of the enzyme, are very

similar to the equivalent subunits of R. sphaeroides CytcO [61]. A structural

alignment of subunits I and II of R. sphaeroides and mammalian CytcO is

shown in figure 8. The only clear differences between the structures are at

the N- and C-termini and in a loop consisting of residues 64-89 (R.

sphaeroides amino-acid numbering).

The functionally important parts of the R. sphaeroides CytcO discussed

previously are also found in the mammalian enzyme. The D pathway is

present in the mammalian CytcO and is believed to be used for transfer of

protons from the N side to the catalytic site [17]. The Glu286 residue and the

two Arginines (Arg481 and 481) close to heme a are conserved between the

R. sphaeroides and the mammalian CytcO. Due to the difficulties of using

site-directed mutagenesis to introduce amino-acid substitutions in the

mammalian CytcO, functional mechanisms are often investigated using

computational methods. Results from several in silico studies, including

molecular dynamics simulations and electrostatic calculations, suggest a

crucial role of Glu242 (Glu286 in the R. sphaeroides enzyme) in the

mammalian CytcO [39, 40, 62]. The area above the heme groups, important

for gating and exit of pumped protons, has also been investigated using

computational studies [41, 63]. Based on simulations showing dissociation

of a hydrogen-bonded network near the heme a3 propionate upon reduction

of heme a, it was suggested that the proton loading site is localized to the

area of the heme a3 propionates.

Experimental studies indicate that neither the spectral properties nor the

function of the mammalian CytcO differ considerably from the R.

sphaeroides enzyme [58]. Electron transfer that takes place during the single

turnover reaction of fully reduced to oxidized CytcO from R. sphaeroides

and bovine heart mitochondria have been evaluated and compared using

absorption spectroscopy techniques [64]. The rates of the electron-transfer

events in the bacterial and the mammalian CytcO are comparable and the

26

same intermediates of the catalytic cycle (figure 5) are formed with similar

rates [65].

Figure 8. An alignment of subunits I and II of the bovine (PDB 10CC) and the R. sphaeroides (PDB 1M56) CytcO. The R. sphaeroides enzyme is shown in magenta and the mammalian CytcO in blue. The figure was prepared using the PyMOL soft-ware.

The additional subunits of the mammalian CytcO

The mammalian CytcO is a large enzyme complex consiting of 13 subunits

(figure 7). The nuclear-encoded subunits (IV, Va, Vb, VIa, VIb, VIc, VIIa,

VIIb, VIIc and VIII) are translated in the cytosol and transported into the

mitochondrial inner membrane. These are not present in the bacterial CytcO.

Considering the functional similarities between the mammalian CytcO and

27

the four-subunit R. sphaeroides CytcO, the function of the additional 10

subunits is elusive. It has been suggested that these subunits stabilize and/or

regulate the mitochondrial CytcO. In a recent study, small interfering RNA

(siRNA) depletion of subunit Vb resulted in decreased activity and assembly

of CytcO as well as in a decreased membrane potential and increased ROS

formation [66]. Other studies have focused on the regulatory roles of the

nuclear encoded subunits and suggest that there are several ATP/ADP

binding sites in the mammalian CytcO, allowing allosteric regulation of the

enzymatic activity [67].

An additional proton transfer pathway

Apart from the two proton-conducting D and K pathways found in both the

bacterial and the mammalian CytcO, in the mitochondrial protein an

additional proton pathway, the H pathway, has been identified based on the

existence of a hydrogen-bonded network observed in the x-ray crystal

structure [17]. Mutational studies of the bovine CytcO confirmed the

importance of the H pathway for maintaining proton pumping [68, 69]. The

mutations were made using an expression system constructed in HeLa cells.

The system produces a bovine-human hybrid enzyme, which could

complicate the interpretation of the results. Moreover, there is no evidence

for an H pathway in the bacterial CytcO. Several of the amino-acid residues

found in the mitochondrial CytcO H pathway are not conserved in the R.

sphaeroides enzyme and substituting the corresponding H pathway amino-

acid residues in the bacterial CytcO had no effect on the proton pumping

activity [70, 71]. In the yeast (Saccharomyces cerevisiae) CytcO structure,

which is a homology model based on the bovine x-ray crystal structure, there

are hydrophilic amino-acid residues identified in the area of the H pathway.

However, amino-acid residues of the D and the K pathway are conserved to

a higher extent than residues of the H pathway [72].

28

The aim of this thesis

Despite the large number of reports on mtDNA mutations associated with

cancer, the functional significance of the amino-acid substitutions caused by

the mutations remains elusive in the majority of cases. In order to understand

disease mechanisms and develop therapeutic approaches it is important to

uncover the functional characteristics resulting from potentially pathogenic

mtDNA mutations. The aim of my studies has been to characterize, on a

molecular level, a number of phenotypes of CytcO found in connection to

cancer. On the basis of the results, the remaining part of this thesis addresses

the following issues:

1. The relevance of using CytcO from the bacterial model R. sphaeroides for

functional studies on amino-acid residue substitutions found in human

disease.

2. The possible consequences of the substitutions for the development or

progression of cancer.

29

Modeling mtDNA mutations in the R. sphaeroides CytcO

Mitochondrial DNA mutations

The human mtDNA is folded into a double-stranded, circular molecule of

approximately 16600 basepairs and contains 37 genes. 13 genes encode

polypeptides of the electron transport chain of which seven are subunits for

NADH dehydrogenase, one for the bc1 complex, three for CytcO and two for

ATP synthase. The other genes encode 22 tRNA molecules and two

ribosomal RNA molecules (figure 9).

Figure 9. A schematic illustration of the human mtDNA; a circular molecule of approximately 16600 basepairs. The mtDNA encodes mRNA for cyt b, NADH dehy-drogenase, CytcO and ATP synthase as well as tRNA and rRNA. The D loop (dis-placement loop) is a non-coding segment and is part of the control region of the mtDNA molecule.

30

The mammalian mtDNA molecule lacks introns and thus has a high

proportion of coding DNA. Moreover, the mtDNA has a higher mutation

rate compared to the nuclear genome [73, 74]. This is, at least partly,

probably due to its close proximity to the generation sites of ROS. Many

mtDNA alterations are neutral polymorphisms [75]. However, in the late

1980’s the first pathogenic mtDNA mutations were identified [76, 77]. Since

then, there has been extensive research on mtDNA and several hundreds of

potentially pathogenic mutations have been identified [78].

Mutations in the mtDNA have been found in a large number of different

diseases. The so-called mitochondrial diseases such as Leigh syndrome,

LHON (Leber’s hereditary optical neuropathy) and mitochondrial myopathy

are all severe disorders affecting the function of oxidative phosphorylation,

resulting in insufficient synthesis of ATP [78, 79]. More recently, mtDNA

mutations have been found in cancer, neurodegenerative diseases and normal

aging [5-8, 79, 80]. One of the first reports on mtDNA mutations in

connection to cancer is from 1998, when Polyak et al identified a number of

mtDNA mutations in human colorectal tumors [81]. Since then, there have

been more than 300 reports on human cancers containing mutations in

coding parts of the mtDNA [82].

Model systems of mtDNA mutations Although several mtDNA mutations are observed in cancer cells, there is a

lack of experimental data, investigating the phenotype resulting from the

mutations and how it could be linked to the disease. This has resulted in

discussions on the actual functional relevance of the mtDNA mutations [83-

85]. To determine the effects of a mutation and understand how the amino-

acid residue substitution resulting from a mutation could be involved in

disease mechanisms, functional studies have to be made. This requires some

kind of model system. There are various models available, enabling

observations of how different features and cellular processes are affected by

an amino-acid substitution. Three model systems that are frequently used to

investigate the functional effects of mtDNA mutations are

transmitochondrial hybrid cell lines, yeast cells and mouse models. I will

describe these briefly and discuss what information can be obtained using

31

the different systems. For my studies, I chose to model mtDNA mutations in

CytcO from the bacterium R. sphaeroides. The relevance of using this

bacterial model to investigate mtDNA mutations found in humans will also

be discussed.

A summary of some advantages and shortcomings of the different models

used for studies on mtDNA mutations are listed in Table 1.

Transmitochondrial hybrid cell lines

Transmitochondrial hybrid cell lines or cybrids, are obtained by fusing

enucleated cytoplasts, derived from patient cell lines, with immortalized

human cell lines devoid of mtDNA. The first transmitochondrial cybrids

were described in 1989 [86]. Since then, cybrids have been used for studies

of potentially pathogenic human mtDNA point mutations [87, 88].

Transmitochondrial cybrids is an in vitro system with the advantage that the

phenotype resulting from a mtDNA mutation can be studied against a fixed

nuclear background. The model allows a certain phenotype to be assayed

regarding for example oxygen consumption rates, activities of the

respiratory-chain enzymes, ROS production and apoptosis. Further, there are

studies where the transmitochondrial cybrids containing a mtDNA mutation

have been transferred to mice in order to investigate the impact of the

mutation on tumor development or metastasis [89, 90]. However, the use of

cybrids is not uncontroversial. There are several cases where in vivo

phenotypes could not be replicated in an in vitro cybrid system [91, 92].

Yeast models

Yeast models have been used in several cases to study the effects of disease-

related mtDNA mutations [93-95]. Introduction of mutations into the

mtDNA of yeast (such as S. cerevisiae) is possible [72]. The respiratory-

chain complexes of yeast show a high similarity in both structure and

function to the mammalian counterparts. The use of yeast cells enables

investigations of the impact of a mutation on for example cellular growth

rate, respiratory function, activities of the individual respiratory chain

enzymes and ROS production. In addition, the protein with the amino-acid

substitution resulting from the mutation can be purified. In that way a more

32

detailed analysis of the function of the protein can be obtained. The ability of

S. cerevisiae to survive without a functional respiratory chain makes it

possible to study the effects of mtDNA mutations that result in severe

respiratory defects. As S. cerevisiae lacks NADH dehydrogenase, studies on

mutations affecting complex I have been made using the obligate aerobic

yeast Yarrowia lipolytica, that contain a respiratory chain including NADH

dehydrogenase [96].

Mouse models

Using mouse models for studies on mtDNA mutations allows genetic

manipulations such as depletion or deletion of genes. Introduction of specific

point mutations in the mouse mtDNA is not easily accomplished. However,

the development of transmitochondrial mice models has enabled in vivo

studies of the impact of mtDNA point mutations [97]. Mammalian in vivo

models is the only alternative when studying certain aspects of mtDNA

mutation pathogenesis, such as how one specific mutation can result in

multiple different phenotypes, why a deficiency of a specific enzyme is

observed in several different diseases or why some tissues are more affected

by a mutation than others. Using mice it is also possible to address questions

concerning the onset and severity of a disease.

33

Advantages Shortcomings

Purified

bacterial enzyme

• Genetic modifications easily accomplished

• Well-defined system

• Relatively cheap & quick

• Molecular level studies

• Prokaryotic system

lacking several subunits

• No cellular context

Yeast cells • Eukaryotic system

• Genetic modifications possible

• Genetic modifications

more difficult compared to

bacteria

Cybrids • Allows comparisons of phenotypes against a

fixed nuclear background

• Mammalian system

• Experimentally difficult

Animals • Mammalian system

• Cellular context

• Organismal level

• Genetic modifications

difficult

• Expensive

• Ethical issues

Table 1. Some advantages and shortcomings of the different model systems that can be used when studying functional effects of mtDNA mutations.

The relevance of the R. sphaeroides CytcO model

Traditionally, CytcO has been purified from bovine heart mitochondria, S.

cerevisiae and the bacterial species P. denitrificans and R. sphaeroides.

Since the beginning of the 1990’s the R. sphaeroides CytcO has been used in

a large number of both structural and functional studies. Several

investigations have thoroughly evaluated the R. sphaeroides CytcO as a

model for the eukaryotic enzyme and found that it is functionally

homologous to the mammalian CytcO [58, 64, 65]. Functional studies using

R. sphaeroides CytcO have been of great importance for the elucidation of

for example the proton-transfer pathways and the proton-pumping

mechanism. Although some pieces of information are still missing, today we

have a good picture of the functional mechanisms of R. sphaeroides CytcO.

The increasing number of x-ray crystal structures, provide valuable

information for the interpretations of experimental results. Bacterial CytcO

has previously been used for studies on functional effects of pathogenic

mutations [93, 98]. These investigations provided detailed analyses of the

effects of mtDNA mutations found in mitochondrial diseases.

34

Similar to other bacterial models, R. sphaeroides is experimentally easy

and fairly cheap to work with. Methods for both expression and purification

of R. sphaeroides CytcO have been developed and work satisfactory.

Further, the use of R. sphaeroides CytcO enables site-directed mutagenesis

studies. As mentioned previously, S. cerevisiae CytcO has also been used to

study effects of pathogenic mtDNA mutations. However, genetic

modifications of S. cerevisiae are more difficult and time-consuming

compared to a bacterial organism. In spite of the fact that yeast is a

eukaryotic organism, the amino-acid sequence of subunits I-III of S.

cerevisiae and R. sphaeroides CytcO has in fact a similar degree of

homology to the mammalian enzyme [19, 99]. Further, the number of

functional studies made with the R. sphaeroides CytcO greatly outnumbers

the ones where yeast has been used. Naturally, a highly validated model is

more attractive to use. Experimental results are easier to interpret and

evaluate if there are already a large number of studies made using that

specific model.

The various models for studying mtDNA mutations enable investigation of

different functional characteristics of the phenotype. In vivo and also in vitro

models, such as purified mitochondria or transmitochondrial cybrids,

provide a cellular context that is absent when using purified proteins. On the

other hand, the use of purified proteins enables molecular-level studies and

can provide information that is impossible to obtain using other model

systems. Detailed information on the functional effects of amino-acid

substitutions found in pathogenic conditions is valuable since it leads to

elucidation of possible disease mechanisms.

35

CytcO amino-acid substitutions in cancer

Reports on CytcO defects in cancer

Amino-acid substitutions in CytcO have been found for example in prostate

cancer, pancreatic cancer and ovarian cancer [90, 100-102]. In addition,

substitutions in CytcO have been reported in normal colonic crypt stem cells

in colon cancer patients [103, 104]. Greaves et al identified the mutation

6277A>G in the CytcO subunit I gene. The mutation, resulting in the amino-

acid substitution Gly125Asp, was found in colonic crypt stem cells showing

a CytcO deficiency. The authors suggested that the substitution was highly

likely to be the cause for the CytcO deficiency, but no further studies were

made to confirm this. In another study, Pye et al identified a missense

mutation in the gene encoding subunit I of CytcO in respiratory-deficient

transmitochondrial cybrids [104]. The mutation resulted in a Tyr19His

amino-acid substitution. The transmitochondrial cybrids were produced

using normal human colon cells from patients treated for adenocarcinoma.

Studies on mitochondria isolated from the cybrids showed a decrease in the

CytcO activity. In 2005, there was a report on a number of mutations in the

gene encoding subunit I of CytcO found in prostate cancer cells [90]. The

amino-acid substitutions resulting from the mutations were located to

different regions of subunit I. Some of the substitutions affected highly

conserved amino-acid residues. The functional effects of the substitutions

were, however, not investigated.

The background to my studies is the reports on CytcO substitutions found

in patients with colon cancer and prostate cancer [90, 103, 104] and the aim

of the work was to elucidate if and/or how the functional mechanisms of the

enzyme was affected by the substitutions. All substitutions were found in

subunit I of CytcO (figure 10). The equivalent mutations were produced in

the subunit I gene of R. sphaeroides CytcO and the enzyme containing the

36

corresponding amino-acid substitutions was purified. The Gly125 and the

Tyr19 residues that were affected in the colon cancer patients, correspond to

Gly171 and Tyr33, respectively in the R. sphaeroides CytcO. Among the

amino-acid substitutions found in CytcO of prostate cancer cells, Asn11Ser,

Ala122Thr, Ala341Ser and Val380Ile were chosen for the study. In R.

sphaeroides CytcO these residues correspond to Asn25, Ser168, Ala384 and

Val423. For the functional studies, various spectroscopic techniques were

used. Thereby we could examine, at a molecular level, how the amino-acid

substitutions affected the individual electron- and proton transfer events of

the CytcO catalytic cycle as well as the proton pumping activity of the

enzyme.

37

Figure 10. Subunit I (yellow) and II (pink) of R. sphaeroides CytcO. The amino-acid residues that were substituted in colon- and prostate cancer are indicated in blue. The red spheres are the oxygen atoms of the water molecules in the D pathway. The Arginine residues Arg481 and Arg482 are shown in cyan. The arrows indicate the proton-transfer routes. The figure was prepared using the 1M56 PDB entry and the PyMOL software

Functional characterization of the pathogenic substitutions None of the amino-acid substitutions in subunit I of CytcO that were studied

resulted in a complete loss of enzymatic activity. However, with the

Gly171Asp and the Tyr33His CytcO, a decreased catalytic activity was

observed. Moreover, in these two cases specific steps of the catalytic cycle

were slowed compared to the wild-type enzyme. Investigations of the

38

proton-pumping activity showed that in four out of six of the CytcO variants

the proton pumping was lost or clearly defective. The results from papers I-

IV are summarized in Table 2.

Steady-

state

activity

(% of

wt)

Proton

pumping

(number

of H+/e-)

Proton

leak

Rate of

O2

binding

(µs)

Rate of P

formation

(µs)

Rate of F

formation

(µs)

Rate of O

formation

(ms)

Wild-type 100 0.6 No 10 50 100 1

Glu286His 1 0 No 10 80 10000 1200

Gly171Asp 30 0.35 Yes 10 50 100 1

Tyr33His 40 0 No 10 50 600 4

Asn25Ser 80 0.25 No 10 50 100 1

Ser168Thr 50 0.4 Slight 10 50 100 1

Ala384Ser 90 0.6 Slight 10 50 100 1

Val423Ile 100 0.6 Slight 10 50 100 1

Table 2. Summary of the results from papers I-IV. For details on the different cata-lytic intermediates (P, F and O) see figure 5 and page 16.

Effects on the CytcO catalytic activity

The Tyr33 residue is situated in the D pathway. Substitution of the Tyr33 for

a Histidine residue resulted in slowed proton uptake through the D pathway

(paper II) and a steady-state activity of approximately 40 % of that of the

wild-type CytcO. The rate of internal proton transfer from the Glu286

residue to the catalytic site was also slowed with the Tyr33His CytcO

compared to the wild-type enzyme. We suggest that the Tyr33His amino-

acid substitution results in a structural change of the hydrogen-bonded

network around the Glu286 residue and that this structural change affects the

proton delivery to the catalytic site as well as to the pump site. A similar

effect was observed with the CytcO in which the D pathway residues Ser200

and Ser201 were substituted to Valines [47]. The Tyr33 residue is located

next to the Ser200 and Ser201 residues (figure 11).

The Gly171 residue is situated close to heme a. It is part of the putative

proton gating loop above heme a (figure 11). With the Gly171Asp CytcO the

reduction of heme a was significantly slowed compared to the wild-type

enzyme, which caused a decrease of the steady-state catalytic activity to

39

around 30 % of the wild-type activity (paper III). Probably, the Gly171Asp

substitution results in a lowered midpoint potential of heme a. This effect

has been observed before upon the insertion of a negatively charged amino-

acid residue close to heme a [38].

The four amino-acid substitutions that were originally found in human

prostate cancer cells, showed steady-state turnover activities between 50 and

80 % of the wild-type (paper IV). At neutral pH, no effects on the rate of

heme oxidation were observed. However, when the heme oxidation was

monitored at pH 10, there was a decrease in the rate of the slowest phase of

the heme oxidation in all CytcO variants. Since the substituted amino acids

are located at different positions in subunit I of CytcO, we explain the slower

rate of heme oxidation as originating from a structural destabilization of the

protein, occurring only at high pH. Most likely there are small changes in the

electron equilibrium of the redox cofactors in the structurally modified

CytcOs at pH 10.

40

Figure 11. The structure around the heme groups of subunit I of R. sphaeroides CytcO. The two Arginine residues (Arg481 and 482) are shown in cyan and the Gly171 residue in blue. The loop consisting of residues 169-175, that is presumably involved in the gating mechanism, is shown in blue. The oxygen atoms of water molecules in the D pathway and in the area above the hemes are shown as red spheres. The polar contacts in the cluster of water molecules surrounding the Ser200, Ser201 and Tyr33 amino-acid residues is indicated as black dotted lines. The figure was prepared using PDB entry 1M56 and the PyMOL software.

Effects on proton pumping

Most of the disease-related substitutions in CytcO induced changes in the

proton-pumping activity. With the Tyr33His CytcO, there was a complete

loss of the proton-pumping activity (paper II). This could in part be due to

the slowed proton-transfer rate through the D pathway, as has been observed

before with the Asp132Asn, Asp132Ala, Ser197Asp and

Ser200Val/Ser201Val CytcOs [20, 47, 49]. There are, however, examples of

amino-acid residue substitutions in the D pathway that result in lost proton

pumping but a normal proton-transfer rate through the D pathway [48, 52,

105]. In these cases, changes in the structure around the Glu286 and

associated shifts in the pKa of the Glu286 side chain have been proposed to

41

lead to uncoupling of proton pumping. With the Tyr33His CytcO, there is

indeed a shift in the pKa of the Glu286 residue compared to the wild-type

CytcO. Presumably, also in this case, the amino-acid substitution results in a

change in the structure of water molecules around the Glu286 residue,

leading to loss of proton pumping.

With the Gly171Asp CytcO, there was a putative proton leak through the

enzyme (paper III). The residue is part of a peptide, consisting of residues

169-175, suggested to be involved in the gating of pumped protons [57]. A

suggested conformational change of the gating loop could control the

formation of a putative proton-exit path. Exchange of the Gly171 to an

Aspartic acid-residue could probably result in a structural change of the

gating loop allowing a proton back leak. Preliminary results from H/D

exchange studies indicate that with the Gly171Asp CytcO the loop

consisting of residues 158-172 is highly accessible to deuterium exchange at

all catalytic intermediate states (Busenlehner, Namslauer et al, unpublished).

This is different from the results obtained with the wild-type CytcO, with

which there was decreased deuterium incorporation in the F intermediate of

the catalytic cycle. Moreover, the Gly171 residue is situated close to the

Arg481 residue, that is suggested to be involved in the exit of pumped

protons [33, 56]. Taken together, data from paper III and from the

unpublished study by Busenlehner, Namslauer et al. indicate that the

structure and the function of the proton gating region is indeed affected by

the Gly171Asp substitution.

With the Asn25Ser, Ser168Thr, Ala384Ser and Val423Ile CytcOs proton

pumping was detected in all cases (paper IV), although the Asn25Ser CytcO

showed somewhat less proton pumping compared to the wild-type CytcO.

With the other three there were indications of a slight proton leak, although

not at all as prominent as with the Gly171Asp CytcO. As the Ser168 is part

of the putative gating loop and also close to Arg481 and Arg482, the effect

on proton pumping could in this case be explained in terms of a slight

structural change in the proton-gating region. Ala384 and Val423 are both

found in the interior of subunit I (figure 10). There were indications of a

slight proton leak also with these structurally modified CytcOs. The Val423

residue is close to heme a3; with a distance of about 7 Å to the heme a3 iron.

Ala384 is further away from both the catalytic site and Glu286.

42

Nevertheless, as evident from the slight tendency of a proton leak, both

Val423 and Ala384 are probably close enough to the structural parts of the

enzyme involved in gating of the pumped protons.

The results correlate between eukaryotic and R. sphaeroides

CytcO

With the Gly171Asp and the Tyr33His CytcO, found in colon cancer

patients, cellular and transmitochondrial cybrid studies suggested that the

amino-acid substitutions caused a CytcO deficiency and a lowered

respiratory function, respectively [103, 104]. My studies provided detailed

information on how the substitutions affected the CytcO functional

mechanisms (paper II, paper III). The observed effects confirmed the

pathogenic properties of the substitutions and lead to a deeper understanding

of the phenotypes.

In an unpublished study, we investigated the effects of a mutation in the

CytcO subunit I gene that was previously linked to a mitochondrial disease.

In the mammalian CytcO the mutation resulted in the substitution of a

Leucine residue to an Isoleucine (Leu240Ile) [106]. In the R. sphaeroides

CytcO Leu240 corresponds to Leu196, which is situated next to Ser197 in

the D pathway (figure 6). Previously published and unpublished cellular

studies indicated ambiguities concerning the relevance of the substitution for

the respiratory deficiency. [106](Taylor R W, personal communication). Our

studies showed that the Leu196Ile substitution had no effects on the CytcO

function (Namslauer I et al, unpublished). These results, using the R.

sphaeroides CytcO, correlated well with the results from previous

investigation of the equivalent substitution in S. cerevisiae CytcO [93]. In

that study, Bratton et al. investigated the functional properties of several

CytcO substitutions in yeast and bacteria. In the cases where the same

substitution was studied both in yeast and in R. sphaeroides, the results were

similar.

When the phenotypes of the mtDNA mutations found in prostate cancer

cells were characterized, in most cases no functional studies had been made

previously [90]. Thus, it is not possible to correlate the results to any

functional properties observed in mammalian cells. It should be mentioned,

though, that the Ser168Thr substitution (corresponding to the Ala122Thr

43

substitution in mammalian CytcO) has, apart from prostate cancer cells, also

been identified in breast cancer [100]. In this study, a 50 % decrease in the

CytcO steady-state activity was observed due to the substitution. This

corresponds well to the slight reduction in enzymatic activity that was

observed with the R. sphaeroides Ser168Thr CytcO (paper IV).

Comments on the results

To conclude, these studies have lead to a deep understanding of how the

cancer-related substitutions in CytcO affect the function of the enzyme on a

molecular level. This was accomplished using CytcO from the highly

validated model organism R. sphaeroides and various absorption

spectroscopy techniques. In several cases, the substitutions had large effects

on the enzymatic function. Considering the similarities between the bacterial

and the mammalian enzyme, the effects would most probably be similar in

the human CytcO. In cases where the phenotype resulting from a mtDNA

mutation has been investigated not only in R. sphaeroides but in CytcO from

other species as well, the results are similar.

In addition to elucidating possible disease mechanisms, this work has

contributed to the general knowledge of the function of CytcO, especially

concerning the D pathway and the putative gating loop. The results with the

Tyr33His CytcO agree with the observation that D pathway substitutions can

lead to hydrogen-bond reorganizations around Glu286. Further, the study

supports the suggestions that Glu286 is involved in directing protons to

different sites and that changes in the structure around Glu286 result in

uncoupling of proton pumping. The investigations of the Gly171Asp CytcO

propose that there is a proton-gating mechanism also in the vicinity of that

residue. Perhaps a conformational change in this region takes place at certain

steps of the CytcO catalytic cycle, making sure that protons can only be

transferred in one direction.

The rest of this thesis deals with the question whether the functional

impairments of CytcO that were observed could be involved in the

development or progression of cancer.

44

The relevance of CytcO substitutions found in cancer

As numerous mtDNA mutations have been observed in cancer, the effects of

the mutations might somehow be involved in disease mechanisms. By

modeling the mtDNA mutations found in colon and prostate cancer in the R.

sphaeroides CytcO and using spectroscopy techniques to study the

enzymatic activity, the effects of the amino-acid substitutions on the

molecular function of CytcO have been ascertained. In several cases, the

substitutions lead to a decreased catalytic activity and a defective proton-

pumping activity. Now; could these functional defects be involved in the

onset or progress of the disease in which they were detected? And in that

case, what could be the connection between the defects in CytcO and

cancer?

Analysis of tumor cells has revealed that mitochondrial functions,

including the respiratory chain activity, are impaired in cancer cells as

compared to normal cells [107-111]. Respiratory chain dysfunction could be

involved in many other abnormalities, that are frequently observed in cancer

cells, such as an increased dependency on glycolysis for ATP synthesis,

decreased apoptosis and increased ROS production [112, 113]. Some

examples of these and how they could be linked to each other are illustrated

in figure 12.

45

Figure 12. A schematic figure showing various effects of respiratory-chain dysfunc-tion and how these could be linked to each other The figure is based on information obtained from [112-117].

Energy metabolism of cancer cells

Cancer cells, to a much higher extent than normal cells, rely on glycolysis

for production of ATP [107]. This phenomenon is called “aerobic

glycolysis”, or the Warburg effect, and has been known for over 50 years to

be a common feature of tumor cells [118]. The mechanisms of metabolic

reprogramming are complex, involving activation or deactivation of a large

number of genes [108]. Even today the reason for the upregulation of

glycolysis is not fully understood and it remains controversial if the

metabolic shift is a cause or a consequence of cancer. Nevertheless, by

relying on glycolysis instead of oxidative phosphorylation, cancer cells are

able to survive in conditions of fluctuating oxygen tension. Although the

increased glycolytic activity is not necessarily accompanied by a decreased

respiratory chain activity, mtDNA mutations causing defects in oxidative

phosphorylation could contribute to the increased glycolysis observed in

cancer cells. Such a link has been observed for example in [116], where

inhibition of respiratory-chain activity by oligomycin triggered an increase

in glycolysis in cancer cells.

The amino-acid substitutions found in colon cancer resulted in a lowered

CytcO activity as compared to the wild-type enzyme (paper II and III). A

lowered CytcO catalytic activity could lead to a decreased rate of electron

46

flux through the entire respiratory chain and thus insufficient amounts of

ATP being produced. In addition, with the Tyr33His CytcO the oxygen

reduction was uncoupled from proton pumping (paper II). Thus, the energy

from the chemical reaction at the catalytic site is not conserved to the same

extent as in the wild-type CytcO, resulting in an even less energy-efficient

respiratory chain. Also in the Gly171Asp CytcO, the putative proton leak

(paper III) could dissipate the electrochemical gradient and thereby reduce

the energy-efficiency of the respiratory chain. A decreased synthesis of ATP

through oxidative phosphorylation might however not be as devastating for a

cancer cell as for a normal cell, due to the metabolic shift occurring in

tumors. In fact, a defective respiratory-chain activity could contribute to

initiating the metabolic reprogramming that occurs in cancer cells. If the

metabolic shift is a prerequisite for a normal cell to develop into a cancer

cell, then a decreased respiratory-chain activity could be involved in the

onset of the disease.

Oxidative stress and apoptosis in cancer

Oxidative stress has been one of the suggestions for how respiratory-chain

defects contribute to the development of cancer as well as to aging and

neurodegeneration. This was first explained in the so-called mitochondrial

free radical theory, proposing that mitochondria-derived ROS cause aging

due to oxidative damage to macromolecules [119]. This theory has been

further developed to suggest a ROS vicious cycle, where oxidative damage

to mitochondria themselves results in even faster rates of ROS production

[120]. Today, it is well known that ROS could have harmful effects on

DNA, lipids and amino acids. In addition, ROS can activate apoptosis [121].

An increased ROS production has been shown to result from decreased

respiratory-chain activity as well as increased mitochondrial membrane

potential [113, 115, 122].

There are several studies where ROS production, caused by mtDNA

mutations in cancer, has been investigated [89, 90, 117]. Petros et al., found

that an amino-acid substitution in ATP synthase enhanced cancer cell

growth. The same substitution caused a decreased ATP-synthase activity and

an increased ROS production [114]. Further, mutations in the NADH

dehydrogenase subunit 6 gene result in decreased complex I activity and

47

overproduction of ROS. These functional effects were indeed coupled to

tumor cell metastasis by Ishikawa et al [89]. An additional study on the role

of complex I substitutions in cancer showed that increased tumor growth was

linked to changes in ROS generation and apoptosis [117].

Normally, ROS are not formed from the reduction of oxygen at the

catalytic site of CytcO [123]. However, several studies have shown that

chemical inhibition of CytcO activity could in fact lead to increased ROS

formation from other respiratory-chain complexes. An increased ROS

production due to CytcO inhibition by azide has been observed in both

mammalian cells and submitochondrial particles [2, 124, 125]. Use of

cyanide and NO to inhibit CytcO also results in an increased ROS formation

[126, 127]. In these cases the suggested mechanism is that the inhibition of

CytcO causes an increased leak of electrons, and hence ROS formation, from

complexes I and/or III of the respiratory chain. Taking this into account, it is

possible that the decreased CytcO activity resulting from the Gly171Asp and

Tyr33His substitutions (paper II and III) could lead to an increased

mitochondrial ROS production. Moreover, the slowed rate of intramolecular

electron transfer observed in both the Gly171Asp and the Tyr33His CytcO,

result in accumulation of partly reduced intermediate states of the enzyme.

This could increase the probability of electron leakage and hence ROS

production also from CytcO itself.

The mitochondrial electrochemical gradient in cancer cells

The mitochondrial electrochemical gradient formed by the chemiosmotic

coupling is essential, not only for ATP synthesis, but also for other processes

such as ion, metabolite and protein trafficking. It constitutes an electrical as

well as a pH difference across the membrane. The electrical component (the

mitochondrial membrane potential) can be monitored in absolute scale and

has a value around 200 mV measured in isolated mitochondria [128, 129].

However, the value of the membrane potential varies depending on for

example cell type, metabolic state and age of the organism [130, 131].

Alterations of the mitochondrial membrane potential have been observed in

cancer cells [132, 133]. Heerdt et al found that an increased mitochondrial

membrane potential of mammalian cells was beneficial to tumor progression

48

as well as to the initiation of angiogenesis and escape from apoptosis

occurring in cancer cells.

Interestingly, mitochondrial uncoupling has recently been observed in

cancer cells [134]. In this case, an increase in expression of the

mitochondrial uncoupling protein-2 (UCP-2) in cancer cells resulted in a

decreased membrane potential. Further, overexpression of UCP-2 in cancer

cells results in a lowered production of ROS, inhibition of apoptosis and an

onset of the metabolic reprogramming [135-139]. Using the chemical

mitochondrial uncoupler FCCP, a similar effect on ROS production was

observed. Thus, it is suggested that a benefit of mitochondrial uncoupling in

cancer cells is a minimimal production of ROS [140, 141].

With the Gly171Asp and also to some extent with the Ser168Thr CytcO,

a proton leak was observed (paper III and IV). This would diminish the

energy efficiency of the respiratory chain, but could also influence other

mitochondrial processes such as ion transport. To keep the electrochemical

gradient unchanged upon the introduction of a specific proton leak, the

relative fraction of the membrane potential would presumably have to

increase. An increased membrane potential has been linked to increased

ROS production [115]. On the other hand, as pointed out above,

mitochondrial uncoupling has also been observed in cancer and is believed

to be beneficial for the tumor cells.

49

Concluding remarks and future perspectives

By using the R. sphaeroides bacterial model system, amino-acid

substitutions in CytcO originally found in cancer, have been investigated.

The studies have provided molecular insights into how the substitutions

affected the function of the enzyme. The results revealed functional

characteristics that could be involved in cancer mechanisms. However, since

cancer is caused by a number of different mutations, often affecting tumor

suppressor genes or oncogenes, it is unlikely that the mtDNA mutations

investigated here are the sole cause of the disease. It could be, though, that

the mtDNA mutations contribute to the carcinogenic process by impairing

the function of the respiratory chain. Respiratory chain dysfunction has

previously been observed in cancer cells and it is a feature also found in

aging and neurodegenerative disorders. Since both cancer and

neurodegenerative diseases are much more prevalent at a higher age, it is

possible that there are common mechanisms involved. On the basis of what

has been observed concerning cancer and mitochondria, it is possible that the

functional properties of the CytcO variants found in colon and prostate

cancer provide the conditions required for the development of a normal cell

into a cancer cell.

However, much is still uncertain regarding mtDNA mutations and the

metabolic changes occurring in cancer cells. To further investigate the

relevance of specific amino-acid substitutions for the development of a

disease, a combination of studies using various model systems would be

ideal. Studies on yeast cells or transmitochondrial cybrids would provide

information regarding certain parameters such as ROS formation and

apoptosis. By combining several different model systems a clear picture of

the phenotype of a specific pathogenic mutation could be obtained. This

would be useful information not only for the identification of possible

disease mechanisms, but also for the development of future alternative

therapeutic treatments.

50

Sammanfattning på svenska

Genom vår ämnesomsättning (metabolism), bryts maten vi äter ned i allt

mindre delar. Maten används dels till att bygga upp våra kroppar (muskler,

fettreserver o s v), men även till att ge oss den energi som behövs för att

driva alla processer som sker i kroppen, inte minst muskelarbete och

tankeverksamhet. De delar av maten som omsätts till energi är elektronerna i

näringsämnenas molekyler. Efter att maten har passerat ett antal metabola

processer hamnar elektronerna i ett protein som heter Cytokrom c Oxidas

(CytcO). CytcO är ett enzym, alltså ett protein som katalyserar en kemisk

reaktion. Den reaktion som CytcO katalyserar, där elektronerna behövs, är

omvandlingen av syre till vatten. Denna reaktion sker konstant i våra celler

och är en förutsättning för att vi ska kunna tillgodogöra oss energin i maten.

I de allra flesta av oss, fungerar CytcO som det ska. Men förändringar i

strukturen av CytcO har hittats i bland annat coloncancer och prostatacancer.

Vad som däremot inte har studerats är om, och i så fall hur, dessa

förändringar påverkar funktionen av CytcO. Detta är viktigt att veta för att

kunna utreda om förändringarna i CytcO är inblandade i uppkomsten av

cancer.

I vårt arbete har vi använt CytcO från en bakterie för att studera

effekterna av de förändringar i enzymet som hittats i colon- och

prostatacancer. Bakterierna har tillverkat det förändrade enzymet, som sedan

renats fram från bakteriecellerna. En biokemisk metod som kallas optisk

spektroskopi har sedan använts för att studera enzymets funktion på

molekylär nivå, vilket betyder att man kan följa till exempel hur elektroner

och protoner rör sig i enzymet.

Resultaten av dessa studier har publicerats i artiklarna II-IV som ligger till

grund för denna avhandling. Sex olika förändringar i CytcO har studerats

och resultaten visar att i två av fallen har förändringarna stora effekter på

CytcO’s funktion. I de andra fallen har endast små effekter kunnat

51

upptäckas. Artikel I handlar även den om effekterna av en förändring i

CytcO. I detta fall har däremot inte förändringen hittats i relation till någon

sjukdom, utan har konstruerats i syftet att studera den normala funktionen av

enzymet. Två viktiga aspekter av mitt arbete, som diskuteras i denna

avhandling, är:

1. Relevansen av att använda ett bakteriellt modellsystem för att studera

effekterna av förändringar i CytcO som hittats i mänskliga celler.

2. Om, och i så fall hur, de funktionella förändringarna i CytcO skulle kunna

leda till cancer.

Slutsatserna som dras är att det bakteriella CytcO vi använt för att studera

förändringar i enzymet är ett relevant modellsystem. Anledningarna till det

är att både strukturen och funktionen av det bakteriella CytcO i hög grad

stämmer överens med det mänskliga. Det bakteriella enzymet har använts i

snart 20 år i olika studier och är alltså ett mycket validerat system. Vidare

kan förändringar i CytcO produceras relativt enkelt i bakterieceller och

genom att isolera enzymet från bakterierna kan funktionen studeras på

detaljnivå. I några fall har de förändringar vi undersökt med hjälp av

bakteriellt CytcO även studerats i andra system, till exempel jästceller och

mänskliga celler och i dessa fall stämmer resultaten bra överens med våra.

Vad gäller kopplingarna mellan förändringarna i CytcO och cancer, kan jag

bara spekulera och jämföra resultaten av mina studier med vad som tidigare

visats. Jämfört med normala celler, har cancerceller en förändrad

metabolism. Ett sämre fungerande CytcO leder till en förändrad metabolism,

vilket skulle kunna vara en möjlig koppling till cancer. Jag hoppas och tror

att resultaten som presenteras i denna avhandling, kan bidra till att vi på

längre sikt får en tydligare bild av mekanismerna bakom uppkomsten av

cancer.

52

Acknowledgements

During my years (which are quite many) at this department, I have met so many great people. Thank you all for making this department such a nice place. Some people have been more involved in my work than others and I want to give them a special thanks. Peter, du delar alltid med dig av din kunskap och kan förklara alla små knepiga detaljer på ett så enkelt sätt. Men framför allt har du ett fantastiskt engegemang och en entusiasm som är väldigt smittsam. Tack för all hjälp och stöd. Pia, du har tid för vilka frågor man än kan tänkas ha och kan alltid svara på dem. Dessutom är du bra på att ställa frågor själv, vilket är väldigt värdefullt. Tack för all hjälp, både teoretisk och praktisk. I want to thank all my present and past co-workers for all the help, inspira-tion, nice chats, laughs, fun coffee-breaks and the good atmosphere that you: Ann-Louise, Linda, Rosa, Hyun-Ju, Christoph, Gustav, Irina, Emelie, Joachim, Maria, Davinia, Henrik, Josy, Lina, Håkan, Ullis, Peter L., Kristina, Gwen, Gisela, Magnus and Alessandro have provided during my years as a PhD student. It has been great fun working with you. Thanks also to my fantastic project worker Marina Dietz. Your work was very much appreciated. Ett tack också till Elzbieta, min biträdande handledare, för att du intresserat dig för mitt projekt och för att du alltid är glad och pratsam. This is a large department, run by many people. Thank you Stefan, Ann, Maria, Lotta, Håkan T., Peter N., Torbjörn, Haidi, Lina and all others for doing such a good job. Alla vänner och bekanta utanför denna betongkoloss som är min arbetsplats, tack för att ni ger mig andrum från naturvetenskapen. Ni är verkligen sköna att umgås med.

53

Mona och Rolf, mamma och pappa alltså; er hjälpsamhet och omsorg är helt otrolig och mycket uppskattad. Tro aldrig nåt annat. Jag hoppas att jag blir likadan mot mina barn och barnbarn. Tack för allt. Carl, Tina, Noah och Leo, det är alltid kul att träffa er. Tack för att ni är så mysiga och roliga att vara med. Claudia, Anna, Tommy, Anton och Elin, det är väldigt skönt att komma till er på Ekenäs. Tack för gästfrihet, trevligt sällskap och fina växter. Märta och Manne, nu när ni finns så vill jag inte leva en enda dag utan er. Ni gör mitt liv så roligt, intressant och livligt och jag älskar er helt otroligt mycket. Andreas, det liv vi har tillsammans är det bästa och skönaste jag kan tänka mig. Du är helt enkelt underbar.

54

References

1 Seyfrid, T. M. and Shelton, L. M. (2010) Cancer as a metabolic

diseaes. Nutrition and Metabolism 7 2 Chen, Q. and Lesnefsky, E. J. (2006) Depletion of cardiolipin and

cytochrome c during ischemia increases hydrogen peroxide produc-tion from the electron transport chain. Free Radical Biological Medicine 40, 976-982

3 Chen, Q., Vazquez, E. J., Moghaddas, S., Hoppel, C. L. and Lesnef-sky, E. J. (2003) Production of reactive oxygen species by mito-chondria. Journal of Biological Chemistry 278, 36027-36031

4 Liu, Y., Fiskum, G. and Schubert, D. (2002) Generation of reactive oxygen species by the mitochondrial electron transport chain. Jour-nal of Neurochemistry 80, 780-787

5 Brandon, M., Baldi, P. and Wallace, D. C. (2006) Mitochondrial mutations in cancer. Oncogene 25, 4647-4662

6 Chatterjee, A., Mambo, E. and Sidransky, D. (2006) Mitochondrial DNA mutations in human cancer. Oncogene 25, 4663-4674

7 Lin, M. T., Simon, D. K., Ahn, C. H., Kim, L. M. and Beal, M. F. (2002) High aggregate burden of somatic mtDNA point mutations in aging and Alzheimer's disease brain. Human Molecular Genetics 11, 133-145

8 Reddy, P. H. and Beal, M. F. (2005) Are mitochondria critical in the pathogenesis of Alzheimer's disease? Brain Research Reviews 49, 618-632

9 Hackenbrock, C. R., Chazotte, B. and Gupte, S. S. (1986) The ran-dom collision model and a critical assessment of diffusion and colli-sion in mitochondrial electron transport. Journal of Bioenergetics and Biomembranes 18, 331-368

10 Althoff, T., Mills, D. J., Popot, J. L. and Kühlbrandt, W. (2011) Ar-rangement of electron transport chain components in bovine mito-chondrial supercomplex I(1)III(2)IV(1). EMBO Journal 30, 4652-4664

11 Dudkina, N. V., Kudryashev, M., Stahlberg, H. and Boekema, E. J. (2011) Interaction of complexes I, III, and IV within the bovine res-pirasome by single particle cryoelectron tomography. Proceedings of the National Academy of Sciences of the United States of America 108, 15196-15200

55

12 Vonck, J. and Schäfer, E. (2009) Supramolecular organization of protein complexes in the mitochondrial inner membrane. Biochimica et Biophysica Acta 1793, 117-124

13 Koppenol, W. H., Bounds, P. L. and V, D. C. (2011) Otto Warburg's contributions to current concepts of cancer metabolism. Nature Re-views Cancer 11, 325-337

14 Warburg, O., Posener, K. and Negelein, E. (1924) Ueber den Stoffwechsel der Tumoren. Biochemische Zeitschrift 152, 319-344

15 Wikstrom, M. F. K. (1977) Proton pump coupled to cytochrome c oxidase in mitochondria. Nature 266, 271-273

16 Tsukihara, T., Aoyama, H., Yamashita, E., Tomizaki, T., Yamagu-chi, H., Shinzawaitoh, K., Nakashima, R., Yaono, R. and Yoshi-kawa, S. (1995) Structures of metal sites of oxidized bovine heart cytochrome-c-oxidase at 2.8 angstrom. Science 269, 1069-1074

17 Tsukihara, T., Aoyama, H., Yamashita, E., Tomizaki, T., Yamagu-chi, H., Shinzawa-Itoh, K., Nakashima, R., Yaono, R. and Yoshi-kawa, S. (1996) The whole strucure of the 13-subunit oxidized cyto-chrome c oxidase at 2.8 A. Science 272, 1136-1144

18 Iwata, S., Ostermeier, C., Ludwig, B. and Michel, H. (1995) Struc-ture at 2.8 Angstrom resolution of cytochrome c oxidase from Para-coccus denitrificans Nature 376, 660-669

19 Svensson-Ek, M., Abramson, J., Larsson, G., Tornroth, S., Brzezin-ski, P. and Iwata, S. (2002) The X-ray crystal structures of wild-type and EQ(I-286) mutant cytochrome c oxidases from Rhodobacter sphaeroides. Journal of Molecular Biology 321, 329-339

20 Fetter, J. R., Qian, J., Shapleigh, J., Thomas, J. W., Garcia-Horsman, A., Schmidt, E., Hosler, J. P., Babcock, G. T., Gennis, R. B. and Ferguson-Miller, S. (1995) Possible proton relay pathways in cyto-chrome c oxidase. Proceedings of the National Academy of Sciences of the United States of America 92, 1604-1608

21 Roberts, V. A. and Pique, M. E. (1999) Definition of the Interaction Domain for Cytochrome c on Cytochrome c Oxidase: III. Prediction of the docked complex by a complete systematic search. Journal of Biological Chemistry 274, 38051-38060

22 Zhen, Y., Hoganson, C. W. and Babcock, G. T. (1999) Definition of the Interaction Domain for Cytochrome c on Cytochrome c Oxidase: I. Biochemical, spectral and kinetic characterization of surface mu-tants in subunit II of Rhodobacter sphaeroides aa3 Journal of Bio-logical Chemistry 274, 38032-38041

23 Gilderson, G., Salomonsson, L., Aagaard, A., Gray, J., Brzezinski, P. and Hosler, J. (2003) Subunit III of cytochrome c oxidase of Rhodobacter sphaeroides is required to maintain rapid proton uptake through the D pathway at physiologic pH. Biochemistry 42, 7400-7409

24 Wilson, K. S. and Prochaska, L. J. (1990) Phospholipid vesicles containing bovine heart mitochondrial cytochrome c oxidase and subunit III deficient enzyme - analysis of respiratory control and

56

proton translocating activities. Archives of Biochemistry and Bio-physics 282, 413-420

25 Bratton, M. R., Pressler, M. A. and Hosler, J. P. (1999) Suicide Inac-tivation of Cytochrome c Oxidase: Catalytic Turnover in the Ab-sence of Subunit III Alters the Active Site. Biochemistry 38, 16236-16245

26 Hill, B. C. (1991) The reaction of the electrostatic cytochrome c-cytochrome oxidase complex with oxygen. Journal of Biological Chemistry 266, 2219-2226

27 Bloch, D., Belevich, I., Jasaitis, A., Ribacka, C., Puustinen, A., Verkhovsky, M. I. and Wikstrom, M. (2004) The catalytic cycle of cytochrome c oxidase is not the sum of its two halves. Proceedings of the National Academy of Sciences of the United States of Amer-ica 101, 529-533

28 Verkhovsky, M. I., Morgan, J. E. and Wikstrom, M. (1995) Control of electron delivery to the oxygen reduction site of cytochrome c oxidase: a role for protons. Biochemistry 34, 7483-7491

29 Han, S., Ching, Y. C. and Rousseau, D. L. (1990) Ferryl and hy-droxy intermediates in the reaction of oxygen with reduced cyto-chrome c oxidase. Nature 348, 89-90

30 Brändén, M., Sigurdson, H., Namslauer, A., Gennis, R. B., Ädelroth, P. and Brzezinski, P. (2001) On the role of the K-proton transfer pathway in cytochrome c oxidase. Proceedings of the National Academy of Sciences of the United States of America 98, 5013-5018

31 Brzezinski, P. and Johansson, A. L. (2010) Variable proton-pumping stoichiometry in structural variants of cytochrome c oxidase. Bio-chimica et Biophysica Acta 1797, 710-723

32 Faxén, K., Gilderson, G., Ädelroth, P. and Brzezinski, P. (2005) A mechanistic principle for proton pumping by cytochrome c oxidase. Nature 437, 286-289

33 Brändén, G., Brändén, M., Schmidt, B., Mills, D. A., Ferguson-Miller, S. and Brzezinski, P. (2005) The protonation state of a heme propionate controls electron transfer in cytochrome c oxidase. Bio-chemistry 44, 10466-10474

34 Lee, H. J., Ojemyr, L., Vakkasoglu, A., Brzezinski, P. and Gennis, R. B. (2009) Properties of Arg481 mutants of the aa3-type cyto-chrome c oxidase from Rhodobacter sphaeroides suggest that nei-ther R481 nor the nearby D-propionate of heme a3 is likely to be the proton loading site of the proton pump. Biochemistry 48, 7123-7131

35 Brzezinski, P. and Gennis, R. B. (2008) Cytochrome c oxidase: ex-citing progress and remaining mysteries. Journal of Bioenergetics and Biomembranes 40, 521-531

36 Belevich, I., Gorbikova, E., Belevich, N. P., Rauhamäki, V., Wik-strom, M. and Verkhovsky, M. I. (2010) Initiation of the proton pump of cytochrome c oxidase. Proceedings of the National Acad-emy of Sciences of the United States of America 107, 18469-18474

57

37 Maréchal, A. and Rich, P. R. (2011) Water molecule reorganization in cytochrome c oxidase revealed by FTIR spectroscopy. Proceed-ings of the National Academy of Sciences of the United States of America 108, 8634-8638

38 Mills, D. A., Xu, S., Geren, L., Hiser, C., Qin, L., Sharpe, M. A., McCracken, J., Durham, B., Millett, F. and Ferguson-Miller, S. (2008) Proton-dependent electron transfer from CuA to heme a and altered EPR spectra in mutants close to heme a of cytochrome oxi-dase. Biochemistry 47, 11499-11509

39 Kaila, V. R., Verkhovsky, M. I., Hummer, G. and Wikstrom, M. (2008) Glutamic acid 242 is a valve in the proton pump of cyto-chrome c oxidase. Proceedings of the National Academy of Sciences of the United States of America 105, 6255-6259

40 Xu, J. and Voth, G. A. (2005) Computer simulation of explicit pro-ton translocation in cytochrome c oxidase: the D-pathway. Proceed-ings of the National Academy of Sciences of the United States of America 102, 6795-6800

41 Xu, J. and Voth, G. A. (2008) Redox-coupled proton pumping in cytochrome c oxidase: further insights from computer simulation. Biochimica et Biophysica Acta 1777, 196-201

42 Blomberg, M. and Siegbahn, P. (2011) The mechanism for proton pumping in cytochrome c oxidase from an electrostatic and quantum chemical perspective. Biochimica et Biophysica Acta Sep 28

43 Ädelroth, P., Ek, M. S., Mitchell, D. M., Gennis, R. B. and Brzezin-ski, P. (1997) Glutamate 286 in cytochrome aa3 from Rhodobacter sphaeroides is involved in proton uptake during the reaction of the fully-reduced enzyme with dioxygen. Biochemistry 36, 13824-13829

44 Brändén, M., Tomson, F., Gennis, R. B. and Brzezinski, P. (2002) The entry point of the K-proton-transfer pathway in cytochrome c oxidase. Biochemistry 41, 10794-10798

45 Smirnova, I. A., Adelroth, P., Gennis, R. B. and Brzezinski, P. (1999) Aspartate-132 in cytochrome c oxidase from Rhodobacter sphaeroides is involved in a two-step proton transfer during oxo-ferryl formation. Biochemistry 38, 6826-6833

46 Qin, L., Hiser, C., Mulichak, A., Garavito, R. M. and Ferguson-Miller, S. (2006) Identification of conserved lipid/detergent-binding sites in a high-resolution structure of the membrane protein cyto-chrome c oxidase. Proceedings of the National Academy of Sciences of the United States of America 103, 16117-16122

47 Lee, H. J., Svahn, E., Swanson, J. M., Lepp, H., Voth, G. A., Brzez-inski, P. and Gennis, R. B. (2010) Intricate role of water in proton transport through cytochrome c oxidase. Journal of the American Chemical Society 132, 16225-16239

48 Lepp, H., Salomonsson, L., Zhu, J. P., Gennis, R. B. and Brzezinski, P. (2008) Impaired proton pumping in cytochrome c oxidase upon

58

structural alteration of the D pathway. Biochimica et Biophysica Acta 1777, 897-903

49 Namslauer, A., Lepp, H., Bränden, M., Jasaitis, A., Verkhovsky, M. I. and Brzezinski, P. (2007) Plasticity of proton pathway structure and water coordination in cytochrome c oxidase. Journal of Biologi-cal Chemistry 282, 15148-15158

50 Namslauer, A., Pawate, A. S., Gennis, R. B. and Brzezinski, P. (2003) Redox-coupled proton translocation in biological systems: proton shuttling in cytochrome c oxidase. Proceedings of the Na-tional Academy of Sciences of the United States of America 100, 15543-15547

51 Gilderson, G., Aagaard, A. and Brzezinski, P. (2002) Relocation of an internal proton donor in cytochrome c oxidase results in an al-tered pK(a) and a non-integer pumping stoichiometry. Biophysical Chemistry 98, 105-114

52 Namslauer, A., Aagaard, A., Katsonouri, A. and Brzezinski, P. (2003) Intramolecular proton-transfer reactions in a membrane-bound proton pump: the effect of pH on the peroxy to ferryl transi-tion in cytochrome c oxidase. Biochemistry 42, 1488-1498

53 Larsson, G. and Brzezinski, P. (2003) Redox-driven proton pumping by heme-copper oxidases. Biochimica et Biophysica Acta 1605, 1-13

54 Dürr, K. L., Koepke, J., Hellwig, P., Müller, H., Angerer, H., Peng, G., Olkhova, E., Richter, O. M. H., Ludwig, B. and Michel, H. (2008) A D-pathway mutation decouples the Paracoccus denitrifi-cans cytochrome c oxidase by altering the side-chain orientation of a distant conserved glutamate. Journal of Molecular Biology 384, 865-877

55 Xu, J., Sharpe, M. A., Qin, L., Ferguson-Miller, S. and Voth, G. A. (2007) Storage of an excess proton in the hydrogen-bonded network of the d-pathway of cytochrome c oxidase: identification of a proto-nated water cluster. Journal of the American Chemical Society 129, 2910-2913

56 Qian, J., Mills, D., Geren, L., Wang, K., Hoganson, C. W., Schmidt, B., Hiser, C., Babcock, G. T., Durham, B., Millett, F. and Ferguson-Miller, S. (2004) Role of the conserved arginine pair in proton and electron transfer in cytochrome c oxidase. Biochemistry 43, 5748-5756

57 Busenlehner, L. S., Salomonsson, L., Brzezinski, P. and Armstrong, R. N. (2006) Mapping protein dynamics in catalytic intermediates of the redox-driven proton pump cytochrome c oxidase. Proceedings of the National Academy of Sciences of the United States of America 103, 15398-15403

58 Hosler, J. P., Fetter, J., Tecklenburg, M., Espe, C., Lerma, C. and Ferguson-Miller, S. (1992) Cytochrome aa3 of Rhodobacter sphaeroides as a model for mitochondrial cytochrome c oxidase. Pu-

59

rification, kinetics, proton pumping, and spectral analysis. Journal of Biological Chemistry 267, 24264-24272

59 Yip, C. Y., Harbour, M. E., Jayawardena, K., Fearnley, I. M. and Sazanov, L. A. (2011) Evolution of respiratory complex I: "super-numerary" subunits are present in the alpha-proteobacterial enzyme. Journal of Biological Chemistry 286, 5023-5033

60 Runswick, M. J. and Walker, J. E. (1983) The amino acid sequence of the beta-subunit of ATP synthase from bovine heart mitochon-dria. Journal of Biological Chemistry 258, 3081-3089

61 Scott, R. A. (1995) Functional significance of cytochrome-c-oxidase structure. Structure 3, 981-986

62 Kaila, V. R., Verkhovsky, M. I., Hummer, G. and Wikstrom, M. (2009) Mechanism and energetics by which glutamic acid 242 pre-vents leaks in cytochrome c oxidase. Biochimica et Biophysica Acta 1787, 1205-1214

63 Kaila, V. R., Sharma, V. and Wikstrom, M. (2011) The identity of the transient proton loading site of the proton-pumping mechanism of cytochrome c oxidase. Biochimica et Biophysica Acta 1807, 80-84

64 Ädelroth, P., Brzezinski, P. and Malmstrom, B. G. (1995) Internal electron transfer in cytochrome c oxidase from Rhodobacter sphaer-oides. Biochemistry 34, 2844-2849

65 Adelroth, P., Ek, M. S. and Brzezinski, P. (1998) Factors determin-ing electron-transfer rates in cytochrome c oxidase: investigation of the oxygen reaction in the R. sphaeroides enzyme. Biochimica et Biophysica Acta 1367, 107-117

66 Galati, D., Srinivasan, S., Raza, H., Prabu, S. K., Hardy, M., Chandran, K., Lopez, M., Kalyanaraman, B. and Avadhani, N. G. (2009) Role of nuclear-encoded subunit Vb in the assembly and sta-bility of cytochrome c oxidase complex: implications in mitochon-drial dysfunction and ROS production. Biochemical Journal 420, 439-449

67 Kadenbach, B. and Arnold, S. (1999) A second mechanism of respi-ratory control. FEBS Letters 447, 131-134

68 Shimokata, K., Katayama, Y., Murayama, H., Suematsu, M., Tsuki-hara, T., Muramoto, K., Aoyama, H., Yoshikawa, S. and Shimada, H. (2007) The proton pumping pathway of bovine heart cytochrome c oxidase. Proceedings of the National Academy of Sciences of the United States of America 104, 4200-4205

69 Tsukihara, T., Shimokata, K., Katayama, Y., Shimada, H., Mu-ramoto, K., Aoyama, H., Mochizuki, M., Shinzawaitoh, K., Yama-shita, E., Yao, M., Ishimura, Y. and Yoshikawa, S. (2003) The low-spin heme of cytochrome c oxidase as the driving element of the proton-pumping process. Proceedings of the National Academy of Sciences of the United States of America 100, 15304

70 Lee, H. M., Das, T. K., Rousseau, D. L., Mills, D. A., Ferguson-Miller, S. and Gennis, R. B. (2000) Mutations in the putative H-

60

channel in the cytochrome c oxidase from Rhodobacter sphaeroides show that this channel is not important for proton conduction but re-veal modulation of the properties of heme a. Biochemistry 39, 2989-2996

71 Salje, J., Ludwig, B. and Richter, O. M. H. (2005) Is a third proton-conducting pathway operative in bacterial cytochrome c oxidase? Biochemical Society Transactions 33, 829-831

72 Maréchal, A., Meunier, B., Lee, D., Orengo, C. and Rich, P. R. (2011) Yeast cytochrome c oxidase: A model system to study mito-chondrial forms of the haem-copper oxidase superfamily. Bio-chimica et Biophysica Acta In press

73 Brown, W. M., George, M. and Wilson, A. C. (1979) Rapid evolu-tion of animal mitochondrial DNA. Proceedings of the National Academy of Sciences of the United States of America 76, 1967-1971

74 Nachman, M. W., Brown, W. M., Stoneking, M. and Aguadro, C. F. (1996) Nonneutral mitochondrial DNA variation in humans and chimpanzees. Genetics 142, 953-963

75 Ingman, M., Kaessmann, H., Pääbo, S. and Gyllensten, U. (2000) Mitochondrial genome variation and the origin of modern humans. Nature 408, 708-713

76 Holt, I. J., Harding, A. E. and Morgan-Hughes, J. A. (1988) Mito-chondrial DNA polymorphism in mitochondrial myopathy. Human Genetics 79, 53-57

77 Wallace, D. C., Singh, G., Lott, M. T., Hodge, J. A., Schurr, T. G., Lezza, A. M., Elsas, L. J. and Nikoskelainen, E. K. (1988) Mito-chondrial DNA mutation associated with Leber's hereditary optic neuropathy. Science 242, 1427-1430

78 Tuppen, H., Blakely, E. L., Turnbull, D. M. and Taylor, R. W. (2010) Mitochondrial DNA mutations and human disease. Bio-chimica et Biophysica Acta 1797, 113-128

79 Taylor, R. W. and Turnbull, D. M. (2005) Mitochondrial DNA mu-tations in human disease. Nature Reviews genetics 6, 389-402

80 Greaves, L. C. and Turnbull, D. M. (2009) Mitochondrial DNA mu-tations and ageing. Biochimica et Biophysica Acta 1790, 1015-1020

81 Polyak, K., Li, Y., Zhu, H., Lengauer, C., Willson, J. K., Markowitz, S. D., Trush, M. A., Kinzler, K. W. and Vogelstein, B. (1998) So-matic mutations of the mitochondrial genome in human colorectal tumours. Nature Genetics 20, 291-293

82 Lu, J., Sharma, L. K. and Bai, Y. (2009) Implications of mitochon-drial DNA mutations and mitochondrial dysfunction in tumorigene-sis. Cell Research 19, 802-815

83 Bandelt, H.-J., Salas, A., Taylor, R. W. and Yao, Y.-G. (2009) Ex-aggerated status of "novel" and "pathogenic" mtDNA sequence vari-ants due to inadequate database searches. Human Mutation 30, 191-196

61

84 Montoya, J., Gallardo-López, E., Díez-Sánchez, C., López-Pérez, M. J. and Ruiz-Pesini, E. (2009) 20 years of human mtDNA pathologic point mutations: carefully reading the pathogenicity criteria. Bio-chimica et Biophysica Acta 1787, 476-483

85 Salas, A., Yao, Y.-G., Macaulay, V., Vega, A., Carracedo, A. and Bandelt, H.-J. (2005) A critical reassessment of the role of mito-chondria in tumorigenesis. PLoS Medicine 2

86 King, M. P. and Attardi, G. (1989) Human cells lacking mtDNA: repopulation with exogenous mitochondria by complementation. Science 246, 500-503

87 Khan, S. M., Smigrodzki, R. M. and Swerdlow, R. H. (2007) Cell and animal models of mtDNA biology: progress and prospects. American Journal of Physiology and Cell Physiology 292, C658-669

88 Swerdlow, R. H. (2007) Mitochondria in cybrids containing mtDNA from persons with mitochondriopathies. Journal of Neuroscience Research 85, 3416-3428

89 Ishikawa, K., Takenaga, K., Akimoto, M., Koshikawa, N., Yamagu-chi, A., Imanishi, H., Nakada, K., Honma, Y. and Hayashi, J. (2008) ROS-generating mitochondrial DNA mutations can regulate tumor cell metastasis. Science 320, 661-664

90 Petros, J. A., Baumann, A. K., Ruiz-Pesini, E., Amin, M. B., Sun, C. Q., Hall, J., Lim, S., Issa, M. M., Flanders, W. D., Hosseini, S. H., Marshall, F. F. and Wallace, D. C. (2005) mtDNA mutations in-crease tumorigenicity in prostate cancer. Proceedings of the National Academy of Sciences of the United States of America 102, 719-724

91 Brown, M. D., Trounce, I. A., Jun, A. S., Allen, J. C. and Wallace, D. C. (2000) Functional analysis of lymphoblast and cybrid mito-chondria containing the 3460, 11778, or 14484 Leber's hereditary optic neuropathy mitochondrial DNA mutation. Journal of Biologi-cal Chemistry 275, 39831-39836

92 Rorbach, J., Yusoff, A. A., Tuppen, H., Abg-Kamaludin, D. P., Chrzanowska-Lightowlers, Z. M., Taylor, R. W., Turnbull, D. M., McFarland, R. and Lightowlers, R. N. (2008) Overexpression of human mitochondrial valyl tRNA synthetase can partially restore levels of cognate mt-tRNAVal carrying the pathogenic C25U muta-tion. Nucleic Acid Research 36, 3065-3074

93 Bratton, M., Mills, D., Castleden, C. K., Hosler, J. and Meunier, B. (2003) Disease-related mutations in cytochrome c oxidase studied in yeast and bacterial models. European Journal of Biochemistry 270, 1222-1230

94 Kucharczyk, R., Salin, B. and DiRago, J. P. (2009) Introducing the human Leigh syndrome mutation T9176G into Saccharomyces cere-visiae mitochondrial DNA leads to severe defects in the incorpora-tion of Atp6p into the ATP synthase and in the mitochondrial mor-phology. Human Molecular Genetics 18, 2889-2898

95 Rak, M., Tetaud, E., Duvezin-Caubert, S., Ezkurdia, N., Bieten-hader, M., Rytka, J. and diRago, J. P. (2007) A yeast model of the

62

neurogenic ataxia retinitis pigmentosa (NARP) T8993G mutation in the mitochondrial ATP synthase-6 gene. Journal of Biological Chemistry 282, 34039-34047

96 Kerscher, S., Grgic, L., Garofano, A. and Brandt, U. (2004) Applica-tion of the yeast Yarrowia lipolytica as a model to analyse human pathogenic mutations in mitochondrial complex I (NADH:ubiquinone oxidoreductase). Biochimica et Biophysica Acta 1659, 197-205

97 Fan, W., Waymire, K. G., Narula, N., Li, P., Rocher, C., Coskun, P. E., Vannan, M. A., Narula, J., Macgregor, G. R. and Wallace, D. C. (2008) A mouse model of mitochondrial disease reveals germline se-lection against severe mtDNA mutations. Science 319, 958-962

98 Lucioli, S., Hoffmeier, K., Carrozzo, R., Tessa, A., Ludwig, B. and Santorelli, F. M. (2006) Introducing a novel human mtDNA muta-tion into the Paracoccus denitrificans COX I gene explains func-tional deficits in a patient. Neurogenetics 7, 51-57

99 Capaldi, R. A., Gonzalez-Halphen, D. and Takamiya, S. (1986) Se-quence homologies and structural similarities between the polypep-tides of yeast and beef heart cytochrome c oxidase. FEBS Letters 207, 11-17

100 Gallardo, M. E., Moreno-Loshuertos, R., López, C., Casqueiro, M., Silva, J., Bonilla, F., Rodríguez de Córdoba, S. and Enríquez, J. A. (2006) m.6267G>A: A recurrent mutation in the humna mitochon-drial DNA that reduces cytochrome c oxidase activity and is associ-ated with tumors. Human Mutation 27, 575-582

101 Jerónimo, C., Nomoto, S., Caballero, O. L., Usadel, H., Henrique, R., Varzim, G., Oliveira, J., Lopes, C., Fliss, M. S. and Sidransky, D. (2001) Mitochondrial mutations in early stage prostate cancer and bodily fluids. Oncogene 20, 5195-5198

102 Liu, V. W., Shi, H. H., Cheung, A. N., Chiu, P. M., Leung, T. W., Nagley, P., Wong, L. C. and Ngan, H. Y. (2001) High incidence of somatic mitochondrial DNA mutations in human ovarian carcino-mas. Cancer Research 61, 5998-6001

103 Greaves, L. C., Preston, S. L., Tadrous, P. J., Taylor, R. W., Barron, M. J., Oukrif, D., Leedham, S. J., Deheragoda, M., Sasieni, P., Novelli, M. R., Jankowski, J. A. Z., Turnbull, D. M., Wright, N. A. and McDonald, S. A. C. (2006) Mitochondrial DNA mutations are established in human colonic stem cells, and mutated clones expand by crypt fission. Proceedings of the National Academy of sciences of the United States of America 103, 714-719

104 Pye, D., Kyriakouli, D. S., Taylor, G. A., Johnson, R., Elstner, M., Meunier, B., Chrzanowska-Lightowlers, Z. M., Taylor, R. W., Turnbull, D. M. and Lightowlers, R. N. (2006) Production of trans-mitochondrial cybrids containing naturally occurring pathogenic mtDNA variants. Nucleic Acid Research 34

105 Han, D., Namslauer, A., Pawate, A., Morgan, J. E., Nagy, S., Vak-kasoglu, A. S., Brzezinski, P. and Gennis, R. B. (2006) Replacing

63

Asn207 by aspartate at the neck of the D channel in the aa3-type cy-tochrome c oxidase from Rhodobacter sphaeroides results in de-coupling the proton pump. Biochemistry 45, 14064-14074

106 Varlamov, D. A., Kudin, A. P., Vielhaber, S., Schröder, R., Sassen, R., Becker, A., Kunz, D., Haug, K., Rebstock, J., Heils, A., Elger, C. E. and Kunz, W. S. (2002) Metabolic consequences of a novel mis-sense mutation of the mtDNA CO I gene. Human Molecular Genet-ics 11, 1797-1805

107 Gogvadze, V., Zhivotovsky, B. and Orrenius, S. (2010) The War-burg effect and mitochondrial stability in cancer cells. Molecular Aspects of Medicine 31, 60-74

108 Kroemer, G. and Pouyssegur, J. (2008) Tumor cell metabolism: cancer's Achilles' heel. Cancer Cell 13, 472-482

109 Pedersen, P. L. (1978) Tumor mitochondria and the bioenergetics of cancer cells. Progress in Experimental Tumor Research 22, 190-274

110 Seyfried, T. N. and Shelton, L. M. (2010) Cancer as a metabolic disease. Nutrition and Metabolism 7

111 Solaini, G., Sgarbi, G. and Baracca, A. (2011) Oxidative phosphory-lation in cancer cells. Biochimica et Biophysica Acta 1807, 534-542

112 Hanahan, D. and Weinberg, R. A. (2011) Hallmarks of cancer: the next generation. Cell 144, 646-674

113 Turrens, J. F. (2003) Mitochondrial formation of reactive oxygen species. Journal of Physiology 552, 335-344

114 Indo, H. P., Davidson, M., Yen, H. C., Suenaga, S., Tomita, K., Ni-shii, T., Higuchi, M., Koga, Y., Ozawa, T. and Majima, H. J. (2007) Evidence of ROS generation by mitochondria in cells with impaired electron transport chain and mitochondrial DNA damage. Mito-chondrion 7, 106-118

115 Korshunov, S. S., Skulachev, V. P. and Starkov, A. A. (1997) High protonic potential actuates a mechanism of production of reactive oxygen species in mitochondria. FEBS Letters 416, 15-18

116 López-Rios, F., Sánchez-Aragó, M., García-García, E., Ortega, A. D., Berrendero, J. R., Pozo-Rodríguez, F., López-Encuentra, A., Ballestín, C. and Cuezva, J. M. (2007) Loss of the mitochondrial bioenergetic capacity underlies the glucose avidity of carcinomas. Cancer Research 67, 9013-9017

117 Park, J. S., Sharma, L. K., Li, H. Z., Xiang, R. H., Holstein, D., Wu, J., Lechleiter, J., Naylor, S. L., Deng, J. J., Lu, J. X. and Bai, Y. D. (2009) A heteroplasmic, not homoplasmic, mitochondrial DNA mu-tation promotes tumorigenesis via alteration in reactive oxygen spe-cies generation and apoptosis. Human Molecular Genetics 18, 1578-1589

118 Warburg, O. (1956) On the origin of cancer cells. Science 123, 309-314

119 Harman, D. (1956) Aging: a theory based on free radical and radia-tion chemistry. Journal of Gerontology 11, 298-300

64

120 Ralph, S. J., Rodríguez-Enríquez, S., Neuzil, J., Saavedra, E. and Moreno-Sánchez, R. (2010) The causes of cancer revisited: "mito-chondrial malignancy" and ROS-induced oncogenic transformation - why mitochondria are targets for cancer therapy. Molecular Aspects of Medicine 31, 145-170

121 Indran, I. R., Tufo, G., Pervaiz, S. and Brenner, C. (2011) Recent advances in apoptosis, mitochondria and drug resistance in cancer cells. Biochimica et Biophysica Acta 1807, 735-745

122 Brand, M. D., Affourtit, C., Esteves, T. C., Green, K., Lambert, A. J., Miwa, S., Pakay, J. L. and Parker, N. (2004) Mitochondrial su-peroxide: production, biological effects, and activation of uncou-pling proteins. Free Radical Biology and Medicine 37, 755-767

123 Babcock, G. T. and Wikstrom, M. (1992) Oxygen activation and the conservation of energy in cell respiration. Nature 356, 301-309

124 Chen, Q., Camara, A. K., Stowe, D. F., Hoppel, C. L. and Lesnef-sky, E. J. (2007) Modulation of electron transport protects cardiac mitochondria and decreases myocardial injury during ischemia and reperfusion. American Journal of Physiology and Cell Physiology 292, C137-147

125 Dawson, T. L., Gores, G. J., Nieminen, A. L., Herman, B. and Le-masters, J. J. (1993) Mitochondria as a source of reactive oxygen species during reductive stress in rat hepatocytes. American Journal of Physiology 264, C961-967

126 Jacobson, J., Duchen, M. R., Hothersall, J., Clark, J. B. and Heales, S. J. (2005) Induction of mitochondrial oxidative stress in astrocytes by nitric oxide precedes disruption of energy metabolism. Journal of Neurochemistry 95, 388-395

127 Prabhakaran, K., Li, L., Borowitz, J. L. and Isom, G. E. (2002) Cya-nide induces different modes of death in cortical and mesencephalon cells. Journal of Pharmacology and Experimental Therapeutics 303, 510-519

128 Brand, M. D., Hafner, R. P. and Brown, G. C. (1988) Control of respiration in non-phosphorylating mitochondria is shared between the proton leak and the respiratory chain. Biochemical Journal 255, 535-539

129 Shears, S. B. and Kirk, C. J. (1984) Characterization of a rapid cellu-lar-fractionation technique for hepatocytes. Application in the meas-urement of mitochondrial membrane potential in situ. Biochemical Journal 219, 375-382

130 Hagen, T. M., Yowe, D. L., Bartholomew, J. C., Wehr, C. M., Do, K. L., Park, J. Y. and Ames, B. N. (1997) Mitochondrial decay in hepatocytes from old rats: membrane potential declines, heterogene-ity and oxidants increase. Proceedings of the National Academy of Sciences of the United States of America 94, 3064-3069

131 Hüttemann, M., Lee, I., Pecinova, A., Pecina, P., Przyklenk, K. and Doan, J. W. (2008) Regulation of oxidative phosphorylation, the mi-

65

tochondrial membrane potential, and their role in human disease. Journal of Bioenergetics and Biomembranes 40, 445-456

132 Heerdt, B. G., Houston, M. A. and Augenlicht, L. H. (2005) The intrinsic mitochondrial membrane potential of colonic carcinoma cells is linked to the probability of tumor progression. Cancer Re-search 65, 9861-9867

133 Heerdt, B. G., Houston, M. A. and Augenlicht, L. H. (2006) Growth properties of colonic tumor cells are a function of the intrinsic mito-chondrial membrane potential. Cancer Research 66, 1591-1596

134 Samudio, I., Fiegl, M., McQueen, T., Clise-Dwyer, K. and Andreeff, M. (2008) The warburg effect in leukemia-stroma cocultures is me-diated by mitochondrial uncoupling associated with uncoupling pro-tein 2 activation. Cancer Research 68, 5198-5205

135 Collins, P., Jones, C., Choudhury, S., Damelin, L. and Hodgson, H. (2005) Increased expression of uncoupling protein 2 in HepG2 cells attenuates oxidative damage and apoptosis. Liver International 25, 880-887

136 Derdak, Z., Mark, N. M., Beldi, G., Robson, S. C., Wands, J. R. and Baffy, G. (2008) The mitochondrial uncoupling protein-2 promotes chemoresistance in cancer cells. Cancer Research 68, 2813-2819

137 Harper, M. E., Antoniou, A., Villalobos-Menuey, E., Russo, A., Trauger, R., Vendemelio, M., George, A., Bartholomew, R., Carlo, D., Shaikh, A., Kupperman, J., Newell, E. W., Bespalov, I. A., Wal-lace, S. S., Liu, Y., Rogers, J. R., Gibbs, G. L., Leahy, J. L., Camley, R. E., Melamede, R. and Newell, M. K. (2002) Characterization of a novel metabolic strategy used by drug-resistant tumor cells. FASEB Journal 16, 1550-1557

138 Horimoto, M., Resnick, M. B., Konkin, T. A., Routhier, J., Wands, J. R. and Baffy, G. (2004) Expression of uncoupling protein-2 in human colon cancer. Clinical Cancer Research 10, 6203-6207

139 Lambert, A. J. and Brand, M. D. (2004) Superoxide production by NADH:ubiquinone oxidoreductase (complex I) depends on the pH gradient across the mitochondrial inner membrane. Biochemical Journal 282, 511-517

140 Divakaruni, A. S. and Brand, M. D. (2011) The regulation and physiology of mitochondrial proton leak. Physiology 26, 192-205

141 Baffy, G. (2010) Uncoupling protein-2 and cancer. Mitochondrion 10, 243-252