From the Department of Physiology, The Wenner-Gren Institute,

Stockholm University, Stockholm, Sweden

Limiting factors in ATP synthesis

Tatiana V. Kramarova

Stockholm 2006

Abstract The main function of mitochondria is to provide cells with ATP, synthesized by the

mitochondrial ATP synthase with use of the proton gradient generated by the respiratory chain.

In different tissues, however, the efficiency of mitochondrial energy transduction may vary, and

ATP synthesis can be limited depending upon the physiological context. Several mitochondrial

proteins, some ubiquitous, some present only in specific tissues, could function as uncouplers,

i.e. divert the use of the proton gradient from the ATP synthase and lower the amount of ATP

molecules synthesized by the respiratory chain. Also, little is known about the mechanisms that

underlie the regulation of the biosynthesis of the mitochondrial ATP synthase in a tissue-specific

and physiologically relevant context in higher eukaryotes. Brown adipose tissue of rodents,

where the amount of the ATP synthase is innately low relative to the respiratory chain

components, provides an ideal model system for such an investigation.

The aim of the present study was to clarify some details of the biosynthesis of the ATP

synthase in brown adipose tissue mitochondria, as well as in other tissues, and to test hypotheses

about possible models of activation of several mitochondrial proteins, the ATP/ADP translocase

and UCPs, that could actively or passively utilize the proton gradient, thus bypassing the ATP

synthase.

We have examined the role of the expression of the P1 isoform of the c-Fo subunit in the

biogenesis of ATP synthase in brown adipose tissue. We show that in transgenic mice that

overexpress the P1 isoform, increased levels of c-subunit protein lead to an increase in protein

levels of the β subunit F1-ATPase, and that there is a parallel increase in functional activity of the

ATP synthase in the transgenic mice. Our findings point to a role for the c-Fo subunit in defining

the final content of the ATP synthase in brown adipose tissue (and possibly in other tissues).

We have analyzed sequence regions in the 3’UTR of the β subunit F1-ATPase mRNA

that are important for formation of RNA-protein complexes. We could detect two distinct protein

complexes (RPC1 and RPC2) that bind to two different sequence regions of the 3’UTR, one

being the poly(A) tail and an adjacent region (RPC2), and the other being a sequence stretch at

the 3’ end of the 3’UTR able to form a stem-loop structure (RPC1), which is evolutionarily

conserved throughout mammalian species.

We investigated a possible participation of the ATP/ADP translocase in fatty acid-

induced uncoupling in brown-fat mitochondria. We found an increased content of ANT in

brown-fat mitochondria of UCP1(-/-) mice. However, we found low ANT-mediated fatty acid

uncoupling in these mice despite a higher content of this protein. We conclude that the ANT

cannot substitute for UCP1 in fatty acid uncoupling in brown-fat mitochondria from mice

lacking UCP1. We propose that the two ANT isoforms mediate proton translocation under

2

different conditions.

We evaluated possible uncoupling activity of UCP3 in skeletal muscle from warm- and

cold-acclimated UCP1(+/+) and UCP1(-/-) mice. We found that the UCP3 protein level was

elevated in cold-acclimated UCP1(-/-) mice. However, the effect of GDP (a suggested UCP3

inhibitor) on respiration under various conditions was independent of the level of UCP3

expression. We conclude that no evidence exists for a higher UCP3-mediated uncoupling

activity in mitochondria from UCP1(-/-) mice; a high UCP3 content in cold-acclimated

UCP1(-/-) mice could possibly be linked to improved fatty acid oxidative capacity.

We have investigated a role of UCP1 in defence against oxidative stress. We found that

products of oxidative stress such as 4-hydroxynonenal (HNE) could neither reactivate purine

nucleotide-inhibited UCP1, nor induce additional activation of innately active UCP1 in brown-

fat mitochondria from UCP1(+/+) and UCP1(-/-) mice. We conclude that HNE does not affect

UCP1 activity, and UCP1 would appear not to be physiologically involved in defence against

oxidative stress.

© Tatiana Kramarova, 2006

ISBN 91-7155-246-4

Akademitryck AB, Edsbruk, Stockholm, 2006

3

List of publications This thesis is based on the following publications, which will be referred to in the text by their

Roman numerals.

I

Mitochondrial ATP synthase amount is governed by the c-Fo subunit P1 isoform:

overexpression in brown adipose tissue increases F1Fo-ATPase content

Tatiana V. Kramarova, Irina G. Shabalina, Ulf Andersson, Rolf Westerberg, Josef

Houstek, Jan Nedergaard and Barbara Cannon

To be re-submitted after revision to JBC

II

A conserved sequence in the mammalian 3’ UTR of β-subunit F1Fo-ATPase mRNA

forms a stem-loop necessary for the formation of RNA-protein complexes

Tatiana V. Kramarova, Hana Antonicka, Josef Houstek, Barbara Cannon and Jan

Nedergaard

Under revision for Biochim. Biophys. Acta: Gene Structure and Expression

III

Carboxyactractyloside effects on brown-fat mitochondria implies that the ATP/ADP-

translocator isoforms Ant1 and Ant2 may be responsible for basal and fatty acid-

induced uncoupling, respectively

Irina G. Shabalina, Tatiana V. Kramarova, Jan Nedergaard, and Barbara Cannon

Submitted for publication

IV

UCP1 and defence against oxidative stress: 4-Hydroxy-2-nonenal effects on brown-fat

mitochondria are uncoupling protein 1-independent

Irina G. Shabalina, Natasa Petrovic, Tatiana V. Kramarova, Joris Hoeks, Barbara Cannon,

Jan Nedergaard

JBC 2006, in press

V Cold-induced increase in UCP3 level in skeletal muscle mitochondria of UCP1-

knockout mice does not mediate increased uncoupling

Irina G. Shabalina, Joris Hoeks, Tatiana V. Kramarova, Patrick Schrauwen, Nils-Göran

Larsson, Barbara Cannon, Jan Nedergaard

Manuscript The original publications were reproduced with permission from the publishers.

4

Table of Contents Abstract 2

List of publications 4

List of abbreviations 6

Introduction 7

I. Mammalian mitochondria, their proteins and functions 8

I.a Uncoupling proteins 9

II. F1Fo-ATPase: structure and catalytic activity 13

III. Transcriptional regulation of F1Fo-ATPase biosynthesis 14

III.a General transcription factors regulating OXPHOS gene expression 15

III.b Transcription factors regulating gene expression of ATP synthase subunits 17

III.c Expression of gene isoforms of the c-Fo-subunit of the ATP synthase 21

IV. β-subunit F1-ATPase: post-transcriptional regulation of synthesis 25

V. Assembly of F1Fo-ATPase 29

V.a Proteins, assisting assembly of the ATP synthase 29

V.b Stepwise assembly of the ATP synthase 30

VI. ADP/ATP translocase, structure and functions 33

VI.a ADP/ATP translocase and its role in permeability transition 36

VI.b Uncoupling effect of fatty acids, basal proton leak and the ADP/ATP translocase 37

VII. Mitochondria and ROS production 39

VII.a Role of 4-Hydroxy-2-nonenal in activation of uncoupling proteins 40

VIII. Comments on methods 44

Conclusions 46

Acknowledgements 47

References 48

5

List of abbreviations

ANT ATP/ADP translocase

ATP Adenosine 5’-triphosphate

ATR Atractyloside

BA Bongkrekic acid

BAT Brown adipose tissue

CATR Carboxyactractyloside

COX Cytochrome c oxidase

ERRα Estrogen-related receptor α

4-HNE 4-hydroxy-2-nonenal

MAPK Mitogen-activated protein kinase

NRF 1, 2 Nuclear respiratory factors 1, 2

SCL25 Mitochondrial solute carrier gene family 25

SOD Superoxide dismutase

OSCP Oligomycin-sensitivity conferring protein

OXPHOS Oxidative phosphorylation protein complexes

PCR Polymerase chain reaction

PGC-1 PPARγ coactivator-1

PPAR Peroxisomal proliferator-activated receptor

mtPTP Mitochondrial permeability transition pore

ROS Reactive oxygen species

UCP Uncoupling protein(s)

UTR Untranslated region

6

Introduction The main function of mitochondria is to provide cells with ATP, which is synthesized by the

mitochondrial ATP synthase through the use of the proton gradient generated by the

mitochondrial respiratory chain. ATP synthesis in eukaryotic cells is tightly controlled through a

number of pathways to match the cellular energetic requirements. In different tissues, however,

the efficiency of mitochondrial energy transduction may vary, and ATP synthesis can be limited

depending upon the physiological context. Thus, in thermogenic brown adipose tissue, the

energy from oxidative processes is only partially utilized by the mitochondrial ATP synthase to

generate ATP, and is released instead as heat through the activity of the brown adipose tissue-

specific uncoupling protein UCP1. It has also been proposed that several other mitochondrial

proteins, some ubiquitous, some present only in specific tissues, could function as uncouplers,

i.e. divert the use of the proton gradient from the ATP synthase and lower the amount of ATP

molecules synthesized per oxygen consumed by the respiratory chain (ATP/O ratio).

The aim of the present study was to clarify some details of the biosynthesis of the ATP

synthase in brown adipose tissue mitochondria, as well as in other tissues, and to test hypotheses

about possible models of activation of several mitochondrial proteins, the ATP/ADP translocase

and UCPs, that could actively or passively utilize the proton gradient, thus bypassing the ATP

synthase. For these studies, we have used several murine knockout and transgenic models,

generated either previously in other groups (e.g. UCP1-deficient mice) or by us (P1-isoform

transgenic mice).

In the following review of the literature, I will present how the biosynthesis of the ATP

synthase is regulated at the transcriptional and post-transcriptional levels in brown adipose tissue

and in other tissues, so the ATP synthase content in each tissue could correlate with its

bioenergetic needs; also I will summarize what is known about the ATP synthase structure and

assembly in eukaryotes. In relation to this major focus of the present studies, I will also describe

other mitochondrial proteins, the ATP/ADP translocase, UCP1 and UCP3, their structure, and

their main and ascribed functions in different tissues. Contributions from the present

investigations to the topics described are discussed directly throughout the relevant chapters.

7

I. Mammalian mitochondria, their proteins and functions

Discovered as a separate cellular organelle as early as 1841 (for historical review see (Ernster

and Schatz 1981), the mitochondrion’s central role in cellular physiology has been appreciated

ever since. Apart from providing around 90% of the ATP used in various reactions of cellular

metabolism, mitochondria directly affect the cell in many other ways, participating in apoptosis

and Ca2+ metabolism, producing reactive oxygen species and also protecting from them;

mitochondrial disorders have been implicated in various human diseases, such as

neurodegeneration, aging, cancer, obesity and diabetes. In the following sections, I will briefly

discuss some of these aspects of mitochondrial functioning with respect to ATP synthesis.

Mitochondrial respiration, ultimately leading to ATP synthesis, consists of substrate

oxidation by oxygen, where electrons from the reducing equivalents NADH and FADH2 are

transported via the electron transport chain complexes I-IV to reduce O2 to H2O and

concomitantly pump protons (H+) out of the mitochondrial matrix (Fig. 1). The electrochemical

gradient of protons (Δp) generated by the respiratory chain consists of a membrane potential

(Δψ) and a pH gradient (ΔpH). The major portion of the proton gradient built up by the

respiratory chain is used for the synthesis of ATP by complex V of respiratory chain, or the ATP

synthase (F1Fo-ATPase), as well as for the transport of various metabolites and ions by numerous

mitochondrial carriers. A small part of the proton gradient returns to the mitochondrial matrix

through a so-called basal proton leak. The proton leak does not lead to ATP synthesis and occurs

naturally in mitochondria in all tissues, although the exact mechanisms are not fully understood

at present. Recent studies of basal proton conductance in ATP/ADP translocase-KO mice

implicated the ATP/ADP translocase in this process, at least in part (discussed in Section VI).

In brown adipose tissue, mitochondria have a specialized protein, uncoupling protein 1 (UCP1),

which actively mediates a proton leak and thus produces heat in this tissue.

8

ATP

ADP+Pi

O2 H2O

γ ε

Fo

F1

Figure 1 Mammalian mitochondrion In the mitochondrial inner membrane, the respiratory chain complexes (indicated here by numbers I-IV) drive proton translocation into the intermembrane space, thus creating a proton gradient (Δp). Protons can then return back into the matrix through several alternative routes: primarily through the ATP synthase (complex V) to drive ATP synthesis (see Section II for the mechanism), or, as in brown adipose tissue, through uncoupling protein 1 (UCP1). Also some fraction of the protons return to the matrix through the basal proton leak (discussed in the text), and through the transport of metabolites and ions by carriers, e.g. through the phosphate carrier (Pi) with translocation of phosphate ions. The phosphate carrier (Pi) together with the ATP/ADP translocase (ANT), provide the ATP synthase with ADP3- and Pi. The ATP/ADP translocase has also been implicated in the basal proton leak (discussed in Section VI).

I.a Uncoupling proteins

Since the discovery in 1997 of genes termed UCP2 and UCP3, great interest has been

focused on investigating the subfamily of mitochondrial uncoupling proteins and characterizing

their gene expression, tissue distribution and physiological functions, and a large number of

papers and reviews have been published on the subject (for recent reviews see (Ricquier and

9

Bouillaud 2000; Nedergaard and Cannon 2003; Cannon and Nedergaard 2004; Andrews, Diano

et al. 2005; Brand and Esteves 2005; Krauss, Zhang et al. 2005; Nedergaard, Ricquier et al.

2005)). Here I will only briefly outline a few issues concerning uncoupling proteins and their

established, as well as possible, functions that are of relevance to the current investigation.

Uncoupling protein 1 (UCP1, or thermogenin) is found in the mitochondrial inner

membrane of brown adipose tissue (for review see (Cannon and Nedergaard 2004)). Brown

adipose tissue is a thermogenic organ, which facilitates survival of mammals in a cold

environment. UCP1 was first identified in 1978 (Heaton, Wagenvoord et al. 1978), being the

first protein of the subfamily to be discovered. UCP1 has now an established function in

transporting proton equivalents, pumped out by the respiratory chain, back into the

mitochondrial matrix, bypassing the ATP synthase, thus mediating heat production by the tissue.

The crucial role of UCP1 in heat production was confirmed by studies in UCP1-KO mice

(Enerback, Jacobsson et al. 1997; Golozoubova, Hohtola et al. 2001; Nedergaard, Golozoubova

et al. 2001), where it was shown that mice lacking this protein have difficulties surviving in cold

and establishing the UCP1 as the mediator of non-shivering thermogenesis. The proton

conductance of UCP1 is regulated by purine nucleotides, which are believed to inhibit the UCP1

activity at normal temperatures in vivo. Free fatty-acids, released from triacylglycerol stores in

brown adipose tissue in response to adrenergic stimulation, play a role as re-activators of

inhibited UCP1 (Shabalina, Jacobsson et al. 2004). A number of alternative potential regulators

of UCP1 activity have been proposed, such as superoxide and 4-hydroxynonenal (Echtay,

Roussel et al. 2002; Echtay, Esteves et al. 2003), which are discussed later in more detail in

(Section VII.a and in Paper III).

UCP1 has several structurally related protein homologs, recently identified in different

tissues in mammals (named UCP2 and UCP3) and other species (birds, fish, plants) (reviewed in

(Ricquier and Bouillaud 2000)), united into the subfamily of uncoupling proteins, which together

are a part of the larger mitochondrial transporter family SLC25 (Palmieri 2004). UCP2 protein is

ubiquitously expressed in many tissues, although in some tissues (like BAT), UCP2 mRNA has

been detected, but protein has not; UCP3 is expressed and present in brown adipose tissue,

skeletal muscle and heart (Nedergaard and Cannon 2003). These proteins are structurally related

to the UCP1 (between UCP1 and its closest homologs, UCP2 and UCP3, there is a 58%

similarity in amino acid sequence), but in contrast to the UCP1, do not possess proven

uncoupling thermogenic activity (and perhaps no uncoupling activity at all), have not been

proven to be regulated by purine nucleotides and their physiological function(s) are still under

active investigation (for review, see (Nedergaard and Cannon 2003)). Alternative functions that

have been ascribed to UCP3 include defence against reactive oxygen species (ROS) production

10

(this function has also been proposed for UCP2) (discussed in Section VII) (Echtay, Roussel et

al. 2002), transport of fatty-acids from the mitochondrial matrix, facilitation of fatty-acid

oxidation or detoxification of excess of fatty-acids in mitochondria (Himms-Hagen and Harper

2001; Schrauwen, Saris et al. 2001; Schrauwen and Hesselink 2004). The absence of these UCPs

(2 and 3) in some tissues, however, raises reasonable doubts about the universality of such

mitochondrial rescue functions of UCPs and prompts a still continuing search for other functions

in mitochondria.

In order to investigate a possible contribution of UCP3 in cold accliamtion of mammals,

we have studied mitochondria isolated from skeletal muscle of several mouse models (Paper V).

UCP1-KO mice, double mutant (SOD2-overexpressing/UCP1-KO) mice (Enerback, Jacobsson

et al. 1997; Silva, Shabalina et al. 2005) and UCP3-overexpressing mice (Clapham, Arch et al.

2000) were used in this study to be able to establish if UCP3 in skeletal mucle could be activated

or inhibited in the same manner as described for UCP1.

Due to the lack of UCP1 in the brown adipose tissue of UCP1-KO mice, the mice

maintain body temperature when placed in cold (4˚C) only by heat production from constant

muscle shivering (Golozoubova, Hohtola et al. 2001). Remarkably, these mice require

preacclimation at 18˚C before they are placed at lower temperatures; otherwise they do not

survive (Golozoubova, Hohtola et al. 2001) (see also Fig. 6 in Paper V). UCP1-KO mice,

acclimated to cold, appear to be an appropriate model to investigate if UCP3 can participate in

non-shivering thermogenesis.

We have found (Paper V) that in skeletal muscle mitochondria of the UCP1-KO mice,

the UCP3 protein content was significantly increased in KO-mice acclimated to cold. We then

investigated if the observed high UCP3 content in cold-acclimated UCP1-KO mice

correspondingely affects respiration in muscle mitochondria, by the use of both various

established (free-fatty acids) and proposed (superoxide, produced here by reverse electron flow)

UCP activators, and without them (as could be observed for UCP1-mediated uncouling). We

could not see that in mitochondria with the increased content of UCP3, activation of UCP3-

mediating uncoupling occurs. Also, with the natural inhibitor of UCP1, GDP, no effect was

observed which would correspond to the increased UCP3 content, and concurrent results were

obtained for the UCP3-overexpressing mice. Thus, we could not conclude that UCP3, although

its content is increased in UCP1-KO mice acclimated to cold, plays a role as an uncoupler that

would translocate protons and thus mediate non-shivering thermogenesis.

Another possible function of the UCP3 related to its proposed involvement in fatty-acid

metabolism, has been investigated in (Paper V). We could observe an improved fatty-acid

oxidation in UCP1-KO mice acclimated to cold, and a parallel increase in the content of

11

carnitine-palmitoyl transferase-1 (CPT1) in these mice. We conclude that improved fatty-acid

oxidation in KO-mice, paralleled with an increase in UCP3 content, in skeletal muscle

mitochondria, suggests a facilitative role for this protein in some aspects of the process of fatty-

acid oxidation.

12

II. F1Fo-ATPase: structure and catalytic activity

The F1Fo-ATPase, or ATP synthase is a membrane-bound enzymatic complex found in

mitochondria, chloroplasts and bacteria. The most extensively studied mitochondrial F1Fo-

ATPase, isolated from bovine heart, contains 15 different subunits (16 including endogenous

inhibitor IF1) and has a molecular mass of about 600 kDa (Walker, Lutter et al. 1991). The

unique property of the F1Fo-ATPase is that it produces ATP at the expense of an ion (H+)-motive

force (or vice-versa, although structurally and kinetically synthesis is not the exact reverse of

hydrolysis), by a rotary catalytic mechanism (for reviews, see (Boyer 1997; Vinogradov 1999;

Yoshida, Muneyuki et al. 2001; Boyer 2002).

The F1Fo-ATPase has been separated into two molecular units (Fig. 1). The first unit, the

catalytic F1 oligomer, extends to the mitochondrial matrix and is composed of five subunit types,

forming the “headpiece” (α- and β-subunits in a ratio 3:3) and the central stalk (γ-, δ- and ε-

subunits in ratio 1:1:1) (Walker, Fearnley et al. 1985; Abrahams, Leslie et al. 1994). The central

stalk (γ, δ, and ε subunits) provides the structural link between the catalytic sites in F1 and the Fo.

The second unit, the Fo oligomer, is membrane-bound and composed of 10 different

subunits in mammals (a, b, c10-14, d, e, f, g, (F6), A6L, OSCP), in which OSCP-, F6-, b- and d-

subunits form a peripheral stalk (Walker, Lutter et al. 1991; Collinson, Runswick et al. 1994;

Carbajo, Kellas et al. 2005). Two subunits of the Fo part of the ATP synthase are responsible for

proton translocation (subunits a and c), and the peripheral stalk is thought to serve as a

stabilization structure for the (αβ)3 complex . The structure of the whole ATP synthase from

bovine heart has recently been demonstrated at 32Å resolution (Rubinstein, Walker et al. 2003).

ATP synthase functioning is dependent on conformational changes of the catalytic β-

subunits, which are achieved by rotation of the ring of c-subunits (caused by proton movement

from the intermembrane space into the mitochondrial matrix through the interface between the

ring of c-subunits and the a-subunit) that transfers the rotational motion to the central stalk.

During production of ATP (as well as its hydrolysis), all three F1-αβ subunit pairs sequentially

adopt conformations that favor binding of the substrate nucleotide ADP and Pi/ATP, then the

catalysis reaction and then release of the product (Abrahams, Leslie et al. 1994; Yasuda, Noji et

al. 1998; Itoh, Takahashi et al. 2004; Rondelez, Tresset et al. 2005). It is now generally accepted

that the transition between the different conformations/catalytic states of the β-subunit is the

result of physical rotation of the γε-subunits relative to the (αβ)3 sector. Direct rotation of the γ-

13

subunit has been observed during ATP hydrolysis (this can be executed by the F1 component

when it is separated from the intact enzyme) when fluorescently labeled actin filaments were

attached to the γ-subunit and α/ β subunits were attached to a nickel-coated glass surface through

incorporated histidine residues (Noji, Yasuda et al. 1997). Additional visualization of ATP-

dependent rotation of the F1 complex relative to the c-subunit ring and of the a-subunit relative to

the c ring was obtained using a similar approach (Nishio, Iwamoto-Kihara et al. 2002).

To date, considerable progress has been made in elucidating details of the ATP synthase

structure and the mechanisms of rotary catalysis, both synthesis and, to a greater extent,

hydrolysis. Naturally, a number of unanswered questions remain e.g. clarifying the relationship

between the enzymatic mechanism and rotation; accepting a bi- or tri-site model of

hydrolysis/synthesis; explaining the diversity of stoichiometries of the c-subunit forming the ring

in different organisms and clarifying the roles of different subunits of the Fo and stalk complexes

in rotation and catalysis, that define the focus of ongoing investigations (Boyer 1993; Dou,

Fortes et al. 1998; Ferguson 2000; Papa, Zanotti et al. 2000; Boyer 2001; Menz, Walker et al.

2001; Boyer 2002; Weber and Senior 2003; Gao, Yang et al. 2005).

III. Transcriptional regulation of F1Fo-ATPase biosynthesis

During cell proliferation, tissue development and differentiation, as well as in response to

various external stimuli, the process of mitochondrial biogenesis depends on the mechanisms

that coordinate expression of the nuclear and mitochondrial genomes, leading to changes in

mitochondrial protein levels and in the total number of mitochondria per cell (Garesse and

Vallejo 2001; Goffart and Wiesner 2003). Mitochondrial DNA encodes 13 genes that are

components of the respiratory chain (out of a total of 37 genes found in animal mtDNAs; other

24 genes encode components of the mitochondrial translational apparatus), and the rest of the

genes encoding mitochondrial proteins are located in the nucleus.

Transcriptional and post-transcriptional mechanisms that underlie the concerted

intergenomic expression of genes that belong to the group of oxidative phosphorylation

complexes (OXPHOS), depending on cell-type and energy demands, have been under active

investigation for over two decades (Attardi and Schatz 1988; Goffart and Wiesner 2003).

Transcriptional regulation of ATP synthase genes encoding different subunits is considered to

play a key role in the overall regulation of F1Fo-ATPase synthesis. In brown adipose tissue, the

transcriptional pattern of the majority of the ATP synthase subunits appears to follow the general

transcriptional activation found for other OXPHOS genes. The specificity of the biosynthesis of

the ATP synthase in brown adipose tissue, in contrast to other high energy demanding tissues, is

14

emphasized by the fact that the protein content of the ATP synthase in BAT is very low (Fig. 2)

(Lindberg, de Pierre et al. 1967; Cannon and Vogel 1977). A general overview of transcriptional

factors that might contribute to regulation of the biosynthesis of the ATP synthase in brown

adipose tissue, and also in other tissues, is given below. Post-transcriptional regulation, which is

known to also be important for the regulation of the expression of ATP synthase genes is

discussed further in (Section IV)

0

50

100

150 Liver mitochondria

0 2 4 6 8 10

nmol

O2 ·

min

-1

· m

g -1

min

Glut

ADP

FCCP

M

ADPolig

A B

Figure 2 ADP-stimulated respiration in brown fat-mitochondria (A) and liver

mitochondria (B), measured by the oxygen consumption method (described in Experimental Procedures in Paper I). Induction of respiration in response to ADP addition seen in (B), in comparison to the minimal induction in brown fat-mitochondria (A), illustrates the low activity of the ATP synthase in brown adipose tissue compared to other tissues (The data on ADP-stimulated respiration in liver mitochondria are courteously provided for this review by Dr. Irina Shabalina).

III.a General transcription factors regulating OXPHOS gene expression

Analysis of the OXPHOS genes revealed a number of common features in their promoters that

would allow a coordinated response under specific activation. Among the major transcription

factors involved in mitochondrial biogenesis in mammals, nuclear respiratory factors NRF1 and

NRF2/GABP were found to be involved in transcriptional regulation of a significant number of

the OXPHOS genes (Scarpulla 2002; Kelly and Scarpulla 2004; Scarpulla 2006). NRF1 is not

related to any known gene families; however, homologous regulatory proteins exist in

Drosophila, birds and fish (Virbasius, Virbasius et al. 1993; Scarpulla 2002). NRF1 was initially

15

found to activate cytochrome c expression (Evans and Scarpulla 1989) and putative NRF1

recognition sites have been later found in many promoters of nuclear OXPHOS genes.

Chromatin immunoprecipitation studies (ChiP) identified as many as 691 human promoters out

of 13000 investigated, that were bound with NRF1 in vivo, with the majority of these 691 genes

being involved in mitochondrial biogenesis and metabolism (Cam, Balciunaite et al. 2004). It

was suggested that the presence of the NRF1 site could be associated with ubiquitously

expressed OXPHOS isogenes (Virbasius, Virbasius et al. 1993; Kelly and Scarpulla 2004),

whereas tissue-specific expression could be mediated by other transcriptional factors (discussed

below). Posttranslational modifications of NRF1, such as phosphorylation and glycosylation,

were found to affect target gene transcription either in an activating or an inhibiting mode

(Scarpulla 2006).

NRF2 (the mouse homolog of NRF2 is called GABP) belongs to the family of EST-

domain proteins, sequence-specific transcriptional activators (Virbasius, Virbasius et al. 1993).

NRF2 was initially found to activate COX IV transcription (Virbasius and Scarpulla 1991) and

also other COX subunits (Scarpulla 2002). The presence of NRF1 or/and NRF2 recognition sites

in a promoter of the same OXPHOS gene in rodents and humans is usually evolutionarily

conserved, but in some cases could be distributed differentially between species (Kelly and

Scarpulla 2004).

Although found in the majority of OXPHOS promoters, binding sites for NRF1 and

NFR2/GABP are not present in all OXPHOS genes; therefore, additional factors must contribute

to transcriptional regulation. The role of general transcription factors such as CREB, Sp-1

(positive and negative regulator), USF2, YY1 (positive and negative regulator), nuclear hormone

receptors PPARs has been described in the induction of nuclear-encoded OXPHOS genes

(Nelson, Luciakova et al. 1995; Basu, Lenka et al. 1997; Zaid, Li et al. 1999; Scarpulla 2002;

Safe and Kim 2004). It was noted that the presence of the GC-box (GGGCGG), which is a

consensus promoter sequence element influencing the frequency of transcription initiation, is a

characteristic of many promoters of the (OXPHOS) genes and thus may regulate the relative

rates of transcription of such genes.

An additional mode of regulation in the scheme of the transcription of mitochondrial

protein genes involves upstream action of coactivator proteins (such as PGC-1, PRC) and/or

hormones (glucocorticoid and thyroid), which might help to integrate the action of the

transcriptional factors mentioned above in a coordinated manner (Scheller and Sekeris 2003; Lin,

Handschin et al. 2005).

The role of PGC-1 (PPARγ coactivator-1, α and β) in mitochondrial biogenesis has

been actively studied, since its tissue-specific (abundant in energy-demanding tissues, such as

16

heart, brain, slow-twitch skeletal muscle, and brown adipocytes) and inducible (cold- and

exercise-) expression correlated with transcriptional induction of several OXPHOS genes, both

nuclear and mitochondria-encoded, and other mitochondrial proteins, including UCP1

(Puigserver, Wu et al. 1998; Finck and Kelly 2006). Recruitment by PGC-1 through protein-

protein interactions with other transcription factors activates transcription of various target

genes. Multiple PGC-1-interacting proteins have been identified, among those are PPARγ, NRF1

and NRF2, ERRα, thyroid hormone receptor, liver X receptor (LXR) and many others (reviewed

in (Finck and Kelly 2006)). PGC-1α co-transfection induces mitochondrial biogenesis through

the NRF1 and NRF2 transcriptional factors in brown adipose tissue and muscle (Wu, Puigserver

et al. 1999). PGC1α was able to induce expression of several OXPHOS genes in several tissues

including BAT, which was analyzed by microarray studies (Mootha, Lindgren et al. 2003).

Among those genes were three ATP synthase subunit genes: F6, g and OSCP, also several COX

subunits. The specificity of target gene transcriptional activation in a tissue-specific context by

PGC-1 and other coactivators is a subject of ongoing investigations. Also very little is known

about factors that could modulate PGC1 activity itself. One such factor has been recently found,

p160MBP, which represses PGC-1α-mediated transcription of target genes. The repression can be

overcome by phosphorylation of PGC-1α by p38 MAPK, a kinase downstream β-adrenergic

signaling in brown adipose tissue (Fan, Rhee et al. 2004).

Thyroid hormone-dependent activation of several OXPHOS genes (cytochome c, subunit

c (total) and the β-subunit of the ATP synthase, ADP/ATP translocase) has been shown to

operate in liver (Luciakova and Nelson 1992; Izquierdo and Cuezva 1993; Nelson, Luciakova et

al. 1995; Andersson, Houstek et al. 1997).

III.b Transcription factors regulating gene expression of ATP synthase subunits

Elements controlling α-subunit F1-ATPase gene expression have been characterized, one of

which is the so-called cis-acting regulatory element 1, which is necessary for activity of the

ATPA gene in HeLa cells (Vander Zee, Jordan et al. 1994; Breen and Jordan 1997). Cis-acting

regulatory element 1 includes the E-box (CACGTG) region that is recognized by the USF2

transcription factor, which together with co-factor p300 activated reporter gene transcription

(Breen and Jordan 1999). Later it was shown that the cis-acting regulatory element 1 is also

recognized by another transcription factor COUP-TFII/ARP-1, acting in a competitive manner

with USF2 and decreasing transcription of the α-subunit gene, and by the YY1 transcription

factor, acting in a similar manner (Breen, Vander Zee et al. 1996; Jordan, Worley et al. 2003).

17

COUP-TFs are known to be able to bind to a variety of dispositions of the basic AGGTCA

motif, including those recognized by the receptors of retinoic acid (RAR), thyroid hormone, and

vitamin D (Safe and Kim 2004). These receptors are also known to interact with the PGC1

coactivator (Wallberg, Yamamura et al. 2003).

Hybridization experiments suggested that the β-subunit of the F1-ATPase is encoded in

the nucleus by a single copy gene (Izquierdo, Jimenez et al. 1995). The basal promoter of the

human β-subunit gene is characterized by the absence of a TATA-box and the presence of a

CCAAT-box (in the Drosophila gene, the CCAAT–box is also absent) (Ohta, Tomura et al.

1988; Neckelmann, Warner et al. 1989; Pena, Ugalde et al. 1995); proximal and distal promoter

regions contain multiple binding sites for the Sp-1 factor. Sp1 is a transcriptional factor, the

function of which depends on promoter context, cell type, and interaction with other nuclear

coregulatory proteins (Safe and Kim 2004). Sp1-dependent transactivation from CG-rich

promoters with TATA boxes or initiator sites (TATA-less) involves complex interactions with

several TATA-binding protein (TBPs)-associated factors (TAFs) that form part of the

transcriptional machinery.

Deletion of one of the core Sp-1-sites in the proximal β-subunit promoter increased the

transcription of a reporter gene, implicating a negative effect of as least some of the Sp-1 regions

on the β-subunit gene expression. (Zaid, Hodny et al. 2001). Putative NRF-1 and EST-binding

sites were found in the human and Drosophila promoter regions of the β-subunit;

OXBOX/REBOX elements that were found both in β-subunit and ADP/ATP translocase,

isoform 1 (ANT1, see Section VI) (Neckelmann, Warner et al. 1989), are involved in regulation

of tissue-specific (OXBOX) and metabolic-responsive (REBOX) gene expression of the β-

subunit (Tomura, Endo et al. 1990; Chung, Stepien et al. 1992; Haraguchi, Chung et al. 1994;

Villena, Martin et al. 1994; Martin, Villena et al. 1996). The β-F1-ATPase gene contains a

putative nuclear respiratory factor-2 (NRF-2) responsive element in its first intron, able to

interact with NRF-2 in HeLa cells (Virbasius and Scarpulla 1991). In BAT, by deletion analysis

of the β-subunit gene promoter region, a regulatory element between –306 and –266, including

the EST-binding site has been found, which binds two transcriptional factors, NRF-2 and Sp1.

During brown adipocyte differentiation, these two factors alter their affinity to the regulatory

element, Sp1 acting in non-differentiated cells and NRF-2 being a major factor at later stages

(Villena, Vinas et al. 1998; Villena, Carmona et al. 2002). It has also been shown that thyroid

hormones regulate the basal expression of the β-F1-ATPase gene (Izquierdo, Luis et al. 1990;

Izquierdo and Cuezva 1993). Administration of thyroid hormones to hypothyroid neonatal rats

induced a rapid increase in liver β-F1-ATPase mRNA and protein content that is thought to be

18

mainly due to the stimulation of transcription of the β-F1-ATPase gene. ERRα (estrogen-related

receptor α), is a nuclear receptor able to interact with PGC-1α and thus mediate transcriptional

activation of many OXPHOS genes, including β-F1-ATPase (Schreiber, Emter et al. 2004). In

experiments, where ERRα expression was inhibited by siRNA, even in the presence of PGC-1α,

the β-F1-ATPase mRNA levels were decreased; gel-shift experiments confirmed ERRα

interaction with the β-F1-ATPase gene promoter (Schreiber, Emter et al. 2004).

The transcription of the rotating γ-subunit of the F1-ATPase complex has been shown to

be activated by the transcription factor NRF-1 (Chau, Evans et al. 1992). Human γ subunits exist

in two different isoforms generated by alternative splicing, that are expressed in a tissue-specific

manner (Matsuda, Endo et al. 1993). The γ-subunit isoform 1 contains NRF-1, Sp-1 and AP1

(coactivator of Sp-1) putative recognition sites in its proximal promoter region (Table 1).

Database analysis of other ATP synthase subunit promoter regions revealed the presence

of putative binding sites for NRF1 and/or NRF2 (Table 1) and other possibly significant factors:

f-subunit: NRF1; d-subunit: NRF1 and NRF2, Sp-1, p300, PPAR γ; b-subunit (isoform 1):

YY1 (coactivator of Sp-1), PPAR γ; g-subunit: NRF-1, PPAR γ; F6: NRF-1, Sp3, p300; OSCP:

NRF2.

OXPHOS enzyme

Promoter regulatory regions

Transcription factors

Co-activators

Tissue Reference

E-box (CANNTG) AGGGTCA basic promoter: lack of TATA-box

YY1 USF2 COUP-TFII/ARP1

p300

HeLa cells

(Breen, Vander Zee et al. 1996; Breen and Jordan 1998; Breen and Jordan 1999; Jordan, Worley et al. 2003)

GC-box (GGGCGG) GGAA OXBOX/REBOX basic promoter: lack of TATA-box

Sp1 NRF2 (human/GABP (mouse) PPAR γ (putative)

PGC1, PPAR γ

Drosophila cells lines;

ATP synthase α-subunit

Myoblasts; HeLa HeLa, Myoblasts

(Ohta, Tomura et al. 1988; Neckelmann, Warner et al. 1989; Chung, Stepien et al. 1992; Virbasius, Virbasius et al. 1993; Haraguchi, Chung et al. 1994; Wu, Puigserver et al. 1999; Zaid, Li et al. 1999; Safe and Kim 2004)

TGCGCATGCGCA NRF-1 Sp1, AP2

(Chau, Evans et al. 1992; Virbasius, Virbasius et al. 1993)

β-subunit γ-subunit

19

OXPHOS enzyme

Promoter regulatory regions

Transcription factors

Co-activators

Tissue Reference

T3

liver

(Luciakova and Nelson 1992)

ATP synthase c (P1)-subunit c (P2)-subunit

lack of TATA-box

NRF-1

(Dyer and Walker 1993)

b putative YY1 PPARγ

MAPPER database

d putative NRF1 NRF2 Sp1, p300, PPARγ

MAPPER database

F

putative NRF1 MAPPER database

g putative NRF1 PPARγ

MAPPER database

F6 putative NRF1 Sp3, p300

MAPPER database

OSCP

putative NRF2 MAPPER database

e putative NRF1

MAPPER database

Cytochrome c oxidase

GGAA

NRF 1 NRF2 /GABP

PGC1 Myoblasts HeLA neurons

(Virbasius, Virbasius et al. 1993; Virbasius, Virbasius et al. 1993; Wu, Puigserver et al. 1999; Ongwijitwat, Liang et al. 2006)

UCP1 CRE, PPRE CREB, PPARγ, α

PGC1 (α, β), RXR

Brown adipocytes

(Cassard-Doulcier, Larose et al. 1994; Puigserver, Wu et al. 1998; Wu, Puigserver et al. 1999; Nedergaard, Petrovic et al. 2005)

ADP/ATP translocase

OXBOX/REBOX (absent in mouse gene) TATA-box

Heart, skeletal muscle,

ANT1 (Chung, Stepien et al. 1992)

ANT2 GC-box GRBOX (CATTGTT) Lack of TATA-box

Sp1 Drosophila cells lines, mammalian cells Proliferative human cells, tumors

(Giraud, Bonod-Bidaud et al. 1998; Zaid, Li et al. 1999; Zaid, Hodny et al. 2001)

ANT3 TATA box GC-box

Sp1 (Fiore, Trezeguet et al. 1998)

20

Table 1 Transcriptional factors involved in regulation of expression of nuclear-encoded

genes of several mitochondrial proteins

AP2-activator protein 2; COUP-TF-chicken ovalbumin upstream promoter-transcription factor;

CRE-CREB response element; CREB-cAMP responsive element; GRBOX-glycolysis-regulated

box; NRF1, 2- nuclear respiratory factor 1, 2; PGC1-PPARγ coactivator 1; PPARγ-peroxisomal

proliferator-activated receptor; PPRE-PPAR response element; RXR-retinoid X receptor; Sp1, 3-

specificity protein 1, 3; T3-triiodothyronine; USF2-upstream stimulatory factor 2; YY1-Ying

Yang 1.

III.c Expression of gene isoforms of the c-Fo-subunit of the ATP synthase

In the vertebrate genome, the c-subunit (Fig. 1) of the F1Fo-ATPase is encoded by three separate

genes, named ATP5G1 (P1 isoform), ATP5G2 (P2 isoform) and ATP3G3 (P3 isoform), located

on different chromosomes (De Grassi, Lanave et al. 2006). The P3 gene is located on

chromosome 2 (while the P1 and P2 genes in humans are located on chromosomes 17 and 12,

respectively). The P1 and P2 genes in the mammalian genome were found and described first

(Dyer and Walker 1993), followed by the discovery of a third member of a c-subunit genes

family, P3, in a human liver cDNA library (Yan, Lerner et al. 1994). The presence of the P3

isoform in other mammals was later confirmed by analysis of genomic databases (De Grassi,

Lanave et al. 2006).

All three genes consist of 5 exons and code for homologous mRNAs, each of ~500 bp,

that are translated into the same mature c-subunit protein, with different mitochondrial import

pre-sequences. The mature protein part of the c-subunit is highly conserved in all vertebrates, the

pre-sequences, however, are not (Gay and Walker 1985; Farrell and Nagley 1987; Dyer and

Walker 1993; De Grassi, Lanave et al. 2006). P3 cDNA is 80% homologous to P1 and P2

cDNAs, also translating to the same mature protein with different import pre-sequences.

Phylogenetic analysis of the P1-P3 distribution in selected Metazoan classes revealed that the P1

gene is the most evolutionarily divergent isoform and P3 is the most conserved (De Grassi,

Lanave et al. 2006). In the mammalian genome, multiple processed pseudogenes exist, as has

been shown for the P2 isoform (Dyer and Walker 1993) and possibly for the P3 isoform (De

Grassi, Lanave et al. 2006), that are not transcribed. When performing genome analysis of P1-

21

overexpressing transgenic mice (Paper I) by PCR, we could show that the mouse genome also

contains processed pseudogenes for P1 isoform (T. Kramarova, unpublished results).

Analysis of genomic DNA sequences of human P1 and P2 isoforms revealed that

promoters of both P1 and P2 contain CG-boxes. Notably, the P1 gene was the only one that

contained a TATA-box, whereas the P2 did not (Dyer and Walker 1993).

Analysis of tissue distribution of the P1 and P2 isoforms revealed that the P2 isoform is

constitutively expressed at the same level in all tissues tested. The P1 isoform, on the other hand,

is only expressed in tissues with high-energy demands (heart, brain, to the lesser extent, kidney)

(Gay and Walker 1985). The P1 and P2 isoforms are also differentially regulated in vivo; the P1

isoform being the one the mRNA levels of which changed in response to various physiological

stimuli (such as ontogenic development and cold acclimation), whereas the P2 expression is not

affected (Andersson, Houstek et al. 1997). It has been proposed that the P2 isoform expression

maintains the basal levels of the c-Fo subunit for a cell and the P1 isoform is regulated when it is

necessary to modulate the relative content of the ATP synthase in a tissue-specific or

physiologically relevant context (Andersson, Houstek et al. 1997).

In BAT, remarkably low expression of the P1 (and not of the P2) gene, in comparison to

other ATP synthase subunits and to other tissues, correlated with the low amount of the ATP

synthase in this tissue (Houstek, Andersson et al. 1995; Andersson, Houstek et al. 1997). The

possibility of the P1 isoform being the target of transcriptional regulation, and thus, defining the

final (low) ATP synthase amount in BAT was thus proposed. This hypothesis was strengthened

by several lines of observation; first, the rather high amount of ATP synthase found in cultured

brown adipocytes (Kopecky, Baudysova et al. 1990), coincided with levels of the P1 mRNA that

were increased, whereas the P2 mRNA stayed at the same level as in BAT in situ (Houstek J.,

Andersson U. and Cannon B., unpublished results). Second, in brown adipose tissue of newborn

lamb, the content of the ATP synthase is high, compared to the BAT of other species (Cannon,

Romert et al. 1977). When the c subunit mRNA was analyzed in brown adipose tissue of

newborn lamb, it was also high (Houstek, Andersson et al. 1995).

As described above, the majority of the ATP synthase subunits genes in BAT are

expressed coordinately with other OXPHOS enzymes, the mRNA levels of which increase in

response to cold exposure (and in development). Apparently, the P1 isoform of the c subunit

eludes this concerted upregulation. Which factors mediate the specific suppression of

transcription of the c-subunit P1 gene that occurs in BAT remains to be investigated. In order to

examine in vivo the significance of the level of P1 expression in the biogenesis of the F1Fo-

ATPase, we have generated a transgenic mouse which overexpresses the c-Fo subunit P1 isoform

specifically in brown adipose tissue (Paper I). However, to maintain brown-adipose specific

22

23

expression and to be able to cold-induce the transgenic P1 gene expression, we placed the P1

cDNA under the control of the UCP1 promoter; therefore we could not estimate the relevant

innate transcriptional regulation of the P1 gene in these mice during the course of the present

investigation.

Analysis of putative promoter regions of P1-P3 indicate the presence of several

recognition sites of transcription regulation factors that are involved in regulation of other

OXPHOS genes, several of which were found to directly interact with PGC-1 (Fig. 3).

Recognition sequence sites of interest include several NRF1 sequence regions in proximity of the

P2 basal promoter that are absent in P1. Other PGC-1-interacting factors include FOXO1

(forkhead box O1), found at –600 position of P2 promoter, SREBP (-800) and thyroid hormone

receptor T3R (-600). The transcription of the c subunit P1 gene has been shown to be induced to

a certain extent in liver by thyroid hormone treatment and there are several potential sites in the

P1 promoter that can bind this hormone (Luciakova and Nelson 1992; Izquierdo and Cuezva

1993; Martin, Villena et al. 1996). However, in brown adipose tissue no such activation of either

P1 or P2 has been found (Andersson, Houstek et al. 1997).

Notably, whereas both P2 and P3 promoters resemble each other in containing putative

NRF1 recognition sites, the P1 promoter appears to be quite distinct and does not contain NRF

recognition sites. Surprisingly, the P1 promoter contains a lot of putative sites for transcription

factors that are involved in regulation of genes involved in fatty-acid metabolism, such as ERRα

(estrogen-related receptor α), PXR (pregnane X receptor), LXR (liver X receptor), RXR (retinoid

X receptor), FXR (farnesyl X receptor); all these nuclear receptors are known to interact with

PGC-1 (Finck and Kelly 2006). ERRα is expressed in BAT, heart and kidney, its interaction with

PGC-1α activates or inhibits its action on the target genes (Ichida, Nemoto et al. 2002; Mootha,

Handschin et al. 2004).

As we proposed in (Paper I), the levels of the c-Fo subunit P1 isoform are crucial for

defining the final amounts of the F1Fo-ATPase in brown adipose tissue. We suggest that this may

also be a generally applied factor in regulation of the assembly of the F1Fo-ATPase and of the

tissue-specific content of the enzyme in higher eukaryotes. Thorough analysis of P1-3 gene

expression is necessary to elucidate which transcription factors could be the possible tissue-

specific regulators of the c-subunit gene expression, thus affecting the content of the ATP

synthase to maintain bioenergetic balance of a given tissue.

Figure 3 Promoters of the P1-P3 isoforms of the c-subunit Fo-ATPase were analyzed by the MAPPER (http://bio.chip.org/mapper) database, which uses a combination of TRANSFAC and JASPAR recourses to describe putative DNA sequences recognized by a number of transcription factors (see text).

Atp5g1

ter region

P2 promoter region

Atp5g3

P3 promoter region

Atp5g2

P1 promo

IV. β-subunit F1-ATPase: post-transcriptional regulation of

synthesis

Being part of the catalytic domain of the mitochondrial F1Fo-ATPase, the β-subunit is

one of the proteins, whose structure, function and regulation of synthesis have been studied the

most extensively amongst the other subunits of the mitochondrial ATP synthase. The β-subunit

mRNA and protein levels are often used as a marker to assess the OXPHOS nuclear gene

activation in different tissues and during development. One of the early observations that the β-

F1-ATPase mRNA steady-state levels showed no correlation with protein levels in different

tissues from rat (Izquierdo and Cuezva 1993) or mouse (Houstek, Tvrdik et al. 1991), prompted

the idea of the involvement of post-transcriptional regulation in β-subunit synthesis. As it was

described in Section III, transcriptional regulation has been studied extensively for the β-subunit

(and other ATP synthase subunits), but the significance of post-transcriptional regulation has

also been acknowledged after a series of studies performed in various tissues and organisms. It

was shown that the post-transcriptional regulation of the β-subunit gene expression could be

executed at different stages of mRNA turnover, affecting its half-life, localization and/or

translational efficiency. The latter process has been studied most extensively during liver

development in rat. Depending on the liver developmental stage, a change in efficiency of the β-

F1-ATPase mRNA translation was observed (Luis, Izquierdo et al. 1993). In studies of the

development of mitochondria in liver tissues in rats, an observation was made that the amount of

β-F1-ATPase protein increases during the course of development from the fetal stage to adult,

with a parallel decrease in mRNA levels and stability (Izquierdo, Ricart et al. 1995).

It has later been shown that changes in translational efficiency during development are

not the result of intrinsic changes in the 3’ or 5’ ends of the mRNA (Izquierdo and Cuezva

1997), so a search of other cis- and trans-acting factors was performed. The 3’UTR (untranslated

region) of the β-F1-ATPase mRNA had been proved to be an essential cis-acting element

involved in translation. It has been reported that the translation of a 3’UTR β-F1-ATPase mRNA

deletion mutant was negligible when completed to the full-length transcript (Izquierdo and

Cuezva 1997). When placed downstream of a reporter gene, the 3’UTR of the β-F1-ATPase

mRNA increased the translation efficiency significantly (Izquierdo and Cuezva 1997; Izquierdo

and Cuezva 2000). However, if β-F1-ATPase mRNA and a reporter, carrying the β-F1-ATPase

3’-UTR, were in vitro translated in fetal liver extracts, inhibition of synthesis was observed.

These results suggested that the translational control of β-F1-ATPase mRNA in liver may be

25

exerted by trans-acting inhibitory protein(s) that block the regulatory elements in the 3’UTR.

Several proteins called 3’βFBPs (~50 kDa) that bind to the 3’ UTR of the β-F1-ATPase mRNA

have been identified in fetal rat liver extracts (Izquierdo and Cuezva 1997; Ricart, Izquierdo et

al. 2002). It has been shown that translation inhibition of β-F1-ATPase mRNA and the presence

of 3’βFBPs are correlated in fetal, as well as in adult, liver tissues (Izquierdo and Cuezva 1997).

The activity of 3’βFBPs was shown to decline from the fetal to the adult stage of liver

development and has been found to be increased in human hepatoma cell lines (de Heredia,

Izquierdo et al. 2000). The mechanism by which 3’βFBPs may inhibit translation of β-F1-

ATPase mRNA remains unclear; however, it has been suggested that upon binding with the

3’UTR, 3’βFBPs repress re-initiation events, displacing other 3’UTR- and cap structure-binding

proteins that are known to play a significant role in initiation of translation.

A regulatory sequence element at the 3’UTR of the β-F1-ATPase mRNA has been found

to functionally resemble an IRES (internal ribosome entry site), which recruits the 40S ribosomal

subunit to mRNA in viruses (Vagner, Galy et al. 2001). When the 3’UTR of the β-F1-ATPase

mRNA was placed between dicistronic RNA, it was able to promote translation of a second

cistron (Izquierdo and Cuezva 2000). This enhancing activity has been shown to be β-F1-

ATPase-specific, since the 3’ UTR from the α-F1-ATPase subunit did not exhibit the same

activity.

Another post-transcriptional mechanism, localization of mRNA, is thought to facilitate

translation and post- or co-translational import of a number of mitochondrial proteins. In S.

cerevisiae, several OXPHOS nuclear genes, among those also ATP synthase subunits (namely,

α, β and γ), were found to be preferentially translated in mitochondria-bound polysomes rather

than in free polysomes (Ades and Butow 1980; Karlberg and Andersson 2003). Also in yeast,

targeting of several mRNAs encoding mitochondrial proteins to mitochondria-bound polysomes

has been shown to be preceded by localization of these transcripts to the mitochondria periphery;

this localization was dependent on sequences within the 3’UTR of the transcripts, as well as the

mitochondrial targeting sequences (Corral-Debrinski, Blugeon et al. 2000). The β-F1-ATPase

mRNA in S. cerevisiae had been shown to contain a specific region in the 3’UTR that, together

with import signal, mediated the proper localization of the reporter transcript. This specific

region was 100 bp-long and supposedly could fold into a long stem-loop structure. Moreover, the

ability of successful import of mutated β-subunit protein, translated from the β-subunit mRNA

construct, deprived of its 3’UTR, was strongly reduced. The resultant respiratory deficiency of

the yeast strain could only be rescued by expressing a β-subunit protein construct, where the 250

26

bp-sequence region (included the 100 bp, necessary for localization) of the 3’UTR was added to

the template (Margeot, Blugeon et al. 2002). In rat liver hepatocytes, β-F1-ATPase mRNA was

found to be localized in clusters in the close vicinity of mitochondria (Egea, Izquierdo et al.

1997); these clusters were translationally active, indicating the importance of the sorting

mechanism to translation and successive import into mitochondria (Ricart, Egea et al. 1997). The

role of the 3’UTR of the β-F1-ATPase mRNA in assembly of the ribonucleoprotein complexes

was demonstrated (Ricart, Izquierdo et al. 2002).

Several lines of evidence confirm that the stability of transcripts of the ATP synthase

subunit genes, along with localization and translational efficiency, contributes to a set of other

mechanisms, defining final levels of subunit proteins, incorporated into an active enzyme.

Estimation of the half-life of the β-F1-ATPase mRNA was performed after inhibition of

transcription using actinomycin D, and the neonatal β-F1-ATPase mRNA was 4-5 times more

stable than the adult transcript. The transcription rates of the β-F1-ATPase gene did not parallel

the sharp reduction in steady-state β-F1-ATPase mRNA levels. The transcription rate in adult

nuclei was found to be at least 3-fold higher than at any stage of liver development (Izquierdo,

Ricart et al. 1995).

In brown adipose tissue, post-transcriptional regulation seems to play an important role

for majority of the ATP synthase subunits transcripts. High oxidative capacity of this tissue is

paralleled by very low amounts of ATP synthase (Lindberg, de Pierre et al. 1967; Cannon and

Vogel 1977; Houstek and Drahota 1977) (Fig. 2). However, the level of transcripts of several

subunits has been shown to correlate with other OXPHOS genes, with the exception of the c-

subunit (discussed earlier in Section III.c). No observable accumulation of intermediate

subcomplexes of the ATP synthase was found in brown adipose tissue (Andersson U., Houstek

J., Cannon B., unpublished results), which, among other things, could also point to the existence

of an additional post-transcriptional mechanism to prevent the occurrence of excessive

translation. The β-F1-ATPase mRNA is an example of a transcript, the levels of which in brown

adipose tissue were comparable to or even exceeded the β-subunit mRNA levels present in ATP-

synthase-rich heart and muscle tissues (Houstek, Tvrdik et al. 1991). However, these high

amounts of the β-F1-ATPase mRNA were not translated to a corresponding amount of protein

(Houstek, Tvrdik et al. 1991; Tvrdik, Kuzela et al. 1992). Two mechanisms, mRNA stability and

possibly also a decrease in translational efficiency could account for the low β-F1-ATPase

protein levels in BAT (Tvrdik, Kuzela et al. 1992). Following an injection of actinomycin D, the

half-lives of the β-F1-ATPase mRNA in BAT and heart tissue were estimated. In BAT, the half-

27

life of the β-F1-ATPase transcript was approximately 6-fold shorter than in heart (Tvrdik, Kuzela

et al. 1992). Also in this study, an indication of lower translational efficiency of the β-F1-ATPase

mRNA extracted from BAT was found.

In our group, a further investigation was conducted in order to elucidate the role of the

3’UTR of the β-F1-ATPase mRNA-mediated post-transcriptional regulation in mammalian

tissues (Paper II). In a previously published study from our group, the specific nature of the full-

length 3’UTR (and not of other regions of the β-F1-ATPase mRNA) interaction with proteins

extracted from brain tissue of mice had been demonstrated (Andersson, Antonicka et al. 2000).

In the present study, we have used mobility shift assays with radioactively labeled 3’UTR of the

β-F1-ATPase mRNA of different lengths to find out if there are proteins that could bind to

different sequences in this region. We could detect two distinct protein complexes (RPC1 and

RPC2) that bind to two different sequence regions of the 3’UTR, one being the poly(A) tail and

an adjacent region (RPC2), and another is a sequence stretch at the 3’ end of the 3’UTR able to

form a stem-loop structure (RPC1). Although the presence of a stem-loop (see Fig.1 in Paper II)

has only been confirmed by RNA folding prediction computer programs and not experimentally,

a proof of the significance of this particular stem-loop structure might come from the following

observations: first, this stem-loop for this region invariably appeared in all possible folding

variants for the whole β-F1-ATPase mRNA as well for its 3’UTR only, predicted by several

RNA folding programs, whereas other regions of the β-F1-ATPase mRNA could structurally

vary. More importantly, this region forming the stem-loop was found to be evolutionarily

conserved throughout mammalian species. In fact, the stem-loop region and the region

responsible for the RPC2 formation, were the only regions in the whole 3’UTR of the β-F1-

ATPase mRNA, which had the highest similarity rate between mammalian species (see Fig. 7 in

Paper II). Remarkably, in more evolutionarily advanced species (chimpanzee and human),

disruption of the stem-loop occurs due to the loss of three nucleotides.

Thus, our findings highlighted specific mRNA sequence stretches in the β-F1-ATPase

mRNA that are able to form complexes with proteins. The next logical step in this line of

investigation would be to clarify the role of the observed RNA-protein complexes in the post-

transcriptional regulation of the β-F1-ATPase mRNA in mammalian tissues, and whether the

observed stem-loop structure protein complex mediates a type of regulation which could then be

lost in evolution to the advantage of primates.

28

V. Assembly of F1Fo-ATPase The assembly process of such a multisubunit protein complex as the ATP synthase occurs in a

number of steps and requires the assistance of various chaperones. The process starts with the

transport of nuclear-encoded ATP synthase subunits, synthesized as precursors on cytosolic

ribosomes, into the mitochondria, where the F1- and Fo-oligomers and the stalk subsequently

form a functional protein complex that resides both in the inner mitochondrial membrane and the

matrix.

The general pathway that works for all proteins imported into mitochondria includes

transport in an unfolded state through the TOM/TIM complexes (translocases of the outer and

inner membrane, respectively)(for recent review see (Mokranjac and Neupert 2005)). Matrix-

directed proteins have a corresponding target presequence at their N-terminus, whereas inner-

membrane directed proteins most often do not have cleavable presequences, but instead contain

several internal targeting signals. Targeting signals are first recognized by the TOM complex and

then direct the import through to TIM23 (matrix-directed precursors) or to TIM22 (inner

membrane-directed precursors) (Bauer, Hofmann et al. 2000; Yamamoto, Esaki et al. 2002;

Rehling, Model et al. 2003; Pfanner, Wiedemann et al. 2004; Wiedemann, Frazier et al. 2004). In

the mitochondrial matrix, the mitochondrial processing peptidase (MPP) removes the

presequences. Proteins destined for the inner membrane are then exported from the matrix into

the membrane with the help of the oxidase assembly complex (OXA1)(Stuart 2002).

V.a Proteins assisting assembly of the ATP synthase

For proper assembly of subunits within the F1 complex, involvement of chaperone-like proteins

is required that act downstream of the translocases. In S. cerevisiae, several such proteins have

been identified (for review, see (Ackerman and Tzagoloff 2005)).

For F1 formation, proteins named Atp11p and Atp12p were found to be necessary to

mediate the formation of the α3 β3 complex (Ackerman and Tzagoloff 1990). Mutant atp11 or

atp12 yeast strains accumulate unsolubilized, i.e. aggregated α and β subunits inside their

mitochondria. It had also been found that aggregation of β-subunits occurs in yeast strains that

have a deletion of the α-subunit and vice versa, whereas the deletion of other subunits of the F1-

oligomer does not produce the same effect (Ackerman and Tzagoloff 1990; Paul, Ackerman et

al. 1994; Ackerman 2002). Two-hybrid screens and affinity interaction studies have shown that

Atp12p binds selectively with the adenine nucleotide binding domain (NBD) of the α-subunit

and Atp11p binds with the β-subunit, interacting with its NBD (Wang and Ackerman 2000;

29

Wang, Sheluho et al. 2000). Atp11p/12p are presumably exchanged in the complete α3 β3

complex through a complementary-structure exchange mechanism, meaning that each chaperone

temporarily provides a contact surface mimicking the α- or β-subunit NBD and thus protecting

β-and α-subunits from aggregation and then is replaced from interaction by the corresponding

subunit (Ackerman and Tzagoloff 2005). In human cells, proteins analogous to Atp11p and

Atp12p were found (Wang, White et al. 2001), and human Atp12 can complement this protein in

a yeast strain with a disrupted atp12 gene. In mouse, only the mRNA and not protein levels of

Atp12 and Atp11 were analyzed (Pickova, Paul et al. 2003). Remarkably, in brown adipose

tissue the Atp12 mRNA levels were found to be the highest among the tissues analyzed and

correlated with the α-subunit mRNA levels; however, the Atp11 mRNA levels were similar in

different tissues and did not parallel the β-subunit mRNA levels (Pickova, Paul et al. 2003).

Another yeast protein, Fmc1p, has been shown to be required for F1 assembly in cells grown at

elevated temperatures (Lefebvre-Legendre, Vaillier et al. 2001).

All these chaperone proteins do not belong to the any of the major chaperone families

(Hsp/hsc) that have a broad range of substrates, but appear to act exclusively on the ATP

synthase biogenesis pathway and are now separated into two distinct chaperone families, Atp11p

and Atp12p, the members of which are found throughout different species (Pickova, Potocky et

al. 2005).

Formation of the Fo oligomer is less studied than that of F1. An additional yeast protein,

Atp10p, has been shown to assist in Fo assembly through binding with subunit 6 (subunit a) that

is encoded in the mitochondrial genome (Ackerman and Tzagoloff 1990; Paul, Barrientos et al.

2000). Another membrane-integrated protein, Atp22p, appears to be involved in Fo-complex

formation in yeast, although its precise role has not yet been described (Helfenbein, Ellis et al.

2003).

V.b Stepwise assembly of the ATP synthase

After the ATP synthase subunits are imported into the mitochondria, independent pre-

formation of the F1-and Fo-oligomers of the ATP synthase occurs. Studies in S. cerevisiae strains

partially or completely depleted of mitochondrial DNA were still able form an active, but

oligomycin-insensitive, F1-ATPase (Schatz 1968; Hadikusumo, Meltzer et al. 1988).

It is believed that primary assembly of a c-subunit ring in the membrane is a crucial step

for formation of an Fo structure (Hadikusumo, Meltzer et al. 1988). In E. coli, recombinant c

subunit is able to assemble itself into ring structures in the absence of any other subunits of the

complex (Arechaga, Butler et al. 2002). In mammals, c subunit appears to play an important role

30

in the formation of the whole ATP synthase. Low amounts of the c subunit correlated with low

amount of the enzyme in the mitochondria of brown adipose tissue (Houstek, Andersson et al.

1995). The presence of several gene isoforms of the c subunit (see also Section III.c) that are

differentially regulated in different tissues and in response to various physiological and

environmental stimuli (Andersson, Houstek et al. 1997) would apparently allow to adjustment of

the ATP synthase content according to the specific bioenergetic needs of a particular tissue. In

the course of the present study, we have attempted to validate this hypothesis in vivo (Paper I)

through generation of a transgenic mouse which overexpresses the c-Fo subunit specifically in

BAT. As the increased levels of the c subunit protein do lead to an increase in the final ATP

synthase amount in this tissue in the transgenic mice, we were able to confirm a significant and

important role of the c subunit structure in a general assembly process. However, how exactly

the F1/Fo-oligomer fates are defined in conditions when the c subunit amount is decreased still

remains to be investigated. In brown adipose tissue, since a very low content of the ATP

synthase is found in the mitochondria, analysis of accumulation of ATP synthase intermediates

by conventional methods, such as 2D electrophoresis in native conditions, or pulse-chase, could

be beyond the resolution provided by these methods. Accumulation of assembly intermediates

was not detected in brown-fat mitochondria, when analyzed (Houstek J., Andersson U. and

Cannon B., unpublished results), therefore, another experimental approach could be used to

investigate the assembly of the ATP synthase in BAT even further.

The stalk complex of the ATP synthase apparently assembles after Fo and F1 are already

associated with each other. From single-subunit depletion studies in yeast, it is known that

association of subunit b with the Fo-oligomer is important for the subsequent assembly of OSCP

and subunit d to proceed (Straffon, Prescott et al. 1998). Subunits e and g probably join the

assembly at a late stages, since their disruption in yeast did not significantly impede oxidative

phosphorylation-mediated growth (reviewed in (Devenish, Prescott et al. 2000)). Assembly of

subunit a (subunit 6 in yeast), which is a part of the Fo-oligomer, was found to depend on the

presence of the OSCP subunit in the pre-assembled F1Fo-ATPase, as it is not found in F1-

immunoprecipitates in OSCP-depleted strains (Prescott, Bush et al. 1994) and joins the assembly

after the c subunit (subunit 9) ring and A6L (subunit 8) association (Hadikusumo, Meltzer et al.

1988).

Studies in human cells, taken from patients with illnesses caused by mutations in

mitochondrial or nuclear DNA coding for ATP synthase subunits, provide a specific model,

where some assembly steps operating in higher eukaryotes could be evaluated (Nijtmans,

Klement et al. 1995; Houstek, Mracek et al. 2004; Jesina, Tesarova et al. 2004).

31

Mutations in mtDNA affecting ATP6 (subunit a) gene are known to occur in a majority of

cases connected to the ATP synthase defects (Holt, Harding et al. 1990). These mutations usually

change protonophoric function of subunit a, which leads to impaired function of the ATP

synthase, thus affecting such energy-demanding tissues as heart, brain, skeletal muscle, and

resulting in severe or lethal illnesses. In mitochondria of these tissues, abnormal accumulation of

subcomplexes was found, substantial depletion of subunit a and of the stalk subunit (subunit b)

protein amounts but an increase in the c subunit protein content was observed (Houstek, Klement

et al. 1995; Jesina, Tesarova et al. 2004).

Mutations of nuclear origin result in a decreased content of an otherwise normal ATP

synthase and were found to affect mostly the genes involved in assembly of the ATP synthase

(such as ATP12) ((Houstek, Mracek et al. 2004) and references therein). As these studies provide

valuable insights into the process of ATP synthase assembly, more substantial data accumulation

of patient cases, however, is required to be able to generalize these observations and incorporate

them in the present ATP synthase assembly model.

Easily accessible genetic manipulations of bacterial and yeast cells provide an accurate

stepwise description of ATP synthase assembly, which could readily be extrapolated to higher

eukaryotes. Nevertheless, different subunit composition of bacterial, yeast and mammalian ATP

synthases and occurrence of the same subunits encoded either in the nuclear or the mitochondrial

genomes, along with various other species-specific factors, indicate steps relevant only to one or

the other group of species and thus more detailed investigations into the assembly process in

higher eukaryotes would undoubtedly be required.

32

VI. ADP/ATP translocase, structure and functions

The ADP/ATP translocase (ANT) is a mitochondrial carrier, the function of which consists of

transporting ADP from the cytosol to the mitochondrial matrix and in exchange to deliver ATP,

synthesized by the mitochondrial ATP synthase, to the cytosol. ANT is generally the most

abundant protein of the inner mitochondrial membrane, constituting up to 10% of the total

mitochondrial protein amount (in mammalian muscle mitochondria, the ratio of ANT to ATP

synthase is approximately 5 to 1) (Klingenberg 1989; Nury, Dahout-Gonzalez et al. 2006). The

ADP/ATP translocase belongs to the large mitochondrial transporter family (SLC25), another

member of which is UCP1 and other uncoupling proteins (Palmieri 2004).

The ADP/ATP translocase exists in three (possibly four, see below) different isoforms

(ANT1, ANT2 and ANT3) in human and in two isoforms (Ant1 and Ant2) in mouse tissues

(Houldsworth and Attardi 1988; Cozens, Runswick et al. 1989; Levy, Chen et al. 2000), all

encoded by separate genes. A fourth member of the human ANT subfamily was described

recently (Dolce, Scarcia et al. 2005). Reported findings of an Ant3 isoform in mouse tissue by

(Stepien, Torroni et al. 1992) have apparently not been reproduced and the same group has

recently confirmed that only two isoforms are present in mouse (Levy, Chen et al. 2000).

Two of the human genes (ANT2 and ANT3) are located on chromosome X and ANT1 is

located on chromosome 4; in mouse, Ant1 is located on chromosome 8 and Ant2 on

chromosome X. The human ANT3 gene is located on the X chromosome major

pseudoautosomal region (PAR1), which is identical in X and Y chromosomes, and therefore

ANT3 escapes X-inactivation. ANT2, on the other hand, is present in one active genomic copy

in both mouse and human, although it maps to different arms of the X-chromosome in these two

organisms (Fiore, Trezeguet et al. 1998). All ADP/ATP translocase genes in human and mouse

consist of four exons and three introns and span from 3.1 to 5.9 kb. There are also seven

ADP/ATP translocase intronless pseudogenes which map to chromosome X in human (Chen,

Chang et al. 1990).

Nucleotide sequence identities between human and mouse and also between isoforms are

high, being within the 80%-90% range (Levy, Chen et al. 2000). The recently described ANT4

shares 66-68% identity with the other isoforms of the human ADP/ATP translocase (Dolce,

Scarcia et al. 2005). All isoforms of the ADP/ATP translocase lack an N-terminal mitochondrial

targeting signal, as do the majority of other SLC25 members (with the exception of the

phosphate and citrate carriers) (Neckelmann, Li et al. 1987; Palmieri 1994).

33

Expression of the ADP/ATP translocase genes depends on the cell physiological

conditions and energy demands (Lunardi and Attardi 1991). Differences observed in the

promoter regions of all three genes account for selective expression of the ADP/ATP translocase

in various tissues (Cozens, Runswick et al. 1989; Fiore, Trezeguet et al. 1998).

ANT1 is highly expressed in heart and skeletal muscle (Stepien, Torroni et al. 1992) in

all organisms tested. In muscle, the activation of ANT1 gene transcription has been found to

depend on the presence of the specific regulatory element OXBOX in its promoter (Li, Hodge et

al. 1990), which has also been later found in the β-subunit F1-ATPase promoter (Chung, Stepien

et al. 1992; Haraguchi, Chung et al. 1994). Studies of Ant1 deficient mice showed that these

mice are viable, although they develop heart hypertrophy and mitochondrial myopathy (Graham,

Waymire et al. 1997). More interestingly, the Ant1 (-/-) mice showed an increased production of

reactive oxygen species (ROS, see more on this subject in Section VII) in heart and skeletal

muscle mitochondria and also an increased production of ROS detoxification enzymes (Esposito,

Melov et al. 1999).

ANT2 is modestly expressed in the majority of tissues of human (Stepien, Torroni et al.

1992) but has been shown to be induced in highly proliferative cells and tumors (Giraud, Bonod-

Bidaud et al. 1998). In mouse tissues, Ant2 is expressed in all tissues, is particularly high in

kidney, and low in skeletal muscle and spleen (Levy, Chen et al. 2000).

ANT3 expression has been investigated in brain, liver, heart and skeletal muscle, where it

was found to be moderately expressed (Stepien, Torroni et al. 1992). ANT4 is expressed only in

liver, brain and testis (Dolce, Scarcia et al. 2005).

In brown adipose tissue the level of expression of different ANT isoforms has been

investigated in the present study (Paper III). We have shown that in BAT, both ANT isoforms

are expressed at approximately similar levels. Protein levels of the ADP/ATP translocase in

brown adipose tissue were found to be similar to the protein levels of the ANT in liver, which is

remarkable, since the ATP synthase amounts in liver are much higher than in BAT (Fig. 2).

Attempts to resolve an atomic structure of the ANT and describe a possible mechanism

of transport began in the mid-70’s. The amino acid sequence of the ADP/ATP translocase was

the first of the SLC25 family to be resolved (Aquila, Misra et al. 1982). It was shown to contain

motifs that were later discovered as characteristic of the whole family: a sequence of three

tandem repeats of ~100 amino acids, each of which is able to form two α-helixes, and a

PxD/ERRK/R sequence, present three times in each carrier, which is used to identify new

members of the SCL25 family (Walker 1992; Pebay-Peyroula and Brandolin 2004). Predictions

from the amino acid sequence were recently confirmed by crystallography studies of the beef

34

heart ADP/ATP translocase at 2.2Å resolution that showed that the enzyme indeed consists of

six α-helixes that form a transmembrane domain (Pebay-Peyroula, Dahout-Gonzalez et al. 2003).

The ADP/ATP carrier is thought to operate as an antiporter, possibly in a dimer

conformation. The antiport model describing the ANT function postulates that each monomer

within the dimer binds either ATP at the mitochondrial matrix side or ADP at the cytosolic side

and for the transport to occur this binding has to be simultaneous (Duyckaerts, Sluse-Goffart et

al. 1980; Pebay-Peyroula, Dahout-Gonzalez et al. 2003), although whether the dimerization of

the ADP/ATP translocase occurs in vivo is still an open question (Pebay-Peyroula and Brandolin

2004). From biochemical data, it was deduced that the conformation of the translocase shifts

from either open only towards the intermembrane space (and closed towards the mitochondrial

matrix) or vice versa. How exactly the transport is performed by the ADP/ATP translocase

remains to be elucidated (Pebay-Peyroula and Brandolin 2004; Nury, Dahout-Gonzalez et al.

2006). Consensus nucleotide binding sites are absent in the ANT sequence, but it is known that

the specific inhibitor of the ADP/ATP translocase, carboxyatractyloside (CATR, a lethal poison,

found in a number of plants, but originally extracted from Atractylis gummifera), competes for

binding with ADP, therefore the sites for the inhibitor and the nucleotide could coincide (Pebay-

Peyroula, Dahout-Gonzalez et al. 2003). Exactly where the ATP might bind has not been shown

with certainty, possibly in the matrix loop M3 (Fig. 4).

Along with its function as a transporter, the ADP/ATP translocase is thought to take part

in several other mitochondrial processes, such as basal proton leak and fatty-acid induced

uncoupling, as well as to participate in formation of the mitochondrial permeability transition

pore (discussed below).

C Intermemb Figure 4 A schematic diagram of the ATP/ADP carrier secondary structure. The carrier is shaped similarly to a basket, which is closed towards the matrix and opened towards intermembrane space. Each pair of transmembrane α-helices consists of tandem repeats (see

N

Matrix

M1 M2 M3

rane space

IMM

35

text), and a proline residue in the PxD/ERRK/R sequence mediates sharp kinks in odd-numbered helices. Transmembrane and surface helices are drawn as cylinders. M1-M3-matrix loops; IMM-inner mitochondrial membrane. Adapted from (Nury, Dahout-Gonzalez et al. 2006).

VI.a ADP/ATP translocase and its role in permeability transition

The role of the ADP/ATP translocase in formation of the mitochondrial permeability transition

pore (mtPTP) is now highly debated in the literature. The mtPTP is thought to mediate a process

of mitochondrial permeability transition (MPT), when, in response to a variety of triggers, the

mitochondria lose transmembrane potential, the mitochondrial matrix swells and subsequently,

the mitochondrial membrane(s) breaks, which in its turn leads to the release of a number of pro-

apoptotic factors, such as cytochrome c (Green and Kroemer 2004; Green 2005).

The role of mtPTP, as well as the process of the permeability transition itself in apoptosis

and/or necrosis pathways of cell death is considered significant. However, this question is

beyond the scope of the present review and will therefore not be discussed here in detail (for

further reading see (Li, Johnson et al. 2004) .

One of the models of mtPTP suggests the formation of the pore in contact sites between

the outer and inner mitochondrial membranes. The complete molecular composition of the

mtPTP is still unclear, and the implication in mtPTP formation of each of the proteins mentioned

below is under ongoing debate in the literature. MtPTP formation is thought to be related to the

voltage-dependent anion channel (VDAC)/porin interaction with the ADP/ATP translocase,

which binds together the outer (VDAC) and the inner (ANT) membranes and builds the

intermembrane pore, and with cyclophilin D (interacting with cyclosporin A, an inhibitor of

mtPTP), found in the matrix, and possibly with other proteins (Crompton, Virji et al. 1998;

Marzo, Brenner et al. 1998; Tsujimoto and Shimizu 2000; Vyssokikh and Brdiczka 2003;

Antonsson 2004; Green and Kroemer 2004; Sharpe, Arnoult et al. 2004; Rostovtseva, Tan et al.

2005).

Originally the ADP/ATP translocase was connected to mtPTP formation based on

experiments that showed that atractyloside (ATR) and bongkrekic acid (BA), two specific

ligands of the ANT, enhance or reduce, respectively, the probability of permeability transition in

purified mitochondria and intact cells (Bucheler, Adams et al. 1991; Zamzami, Susin et al. 1996;

Halestrap, Woodfield et al. 1997; Haworth and Hunter 2000). ATR is known to bind the

ADP/ATP translocase from the cytosolic-side of the inner membrane (external) and BA interacts

from the matrix side; upon binding either of these ligands the ANT is thought to get locked in a

cytosol-facing state (ATR) or matrix-facing state (BA).

36

Other studies further elaborated the hypothesis of ANT involvement in mtPTP and its

role in apoptosis. Immunoprecipitation experiments showed that the pro-apoptotic Bax protein

could interact with ANT in a yeast two hybrid system (Marzo, Brenner et al. 1998);

overexpression studies in human cells showed that ANT1, but not ANT2, induced apoptosis and

this could be repressed by cyclophilin D (Bauer, Schubert et al. 1999). Later, it was suggested

that different ANT isoforms could contribute differentially to pore formation and to cyclophilin

D interaction, due to uneven distribution in the inner membrane; ANT1 was mostly found in

contact sites, and ANT2 was mainly located in cristae (Vyssokikh, Katz et al. 2001)

Recently, reports began to appear that questioned the possible involvement of ANT in the

formation of PTP. In yeast mutants, none of the three isoforms were found to be essential for

Bax-dependent cytochrome c release from isolated mitochondria (Shimizu, Shinohara et al.

2000). Most importantly, in knockout mice lacking both isoforms of ANT (Kokoszka, Waymire

et al. 2004), it was found that mitochondria could still undergo permeability transition, resulting

in the release of cytochrome c and the cells were competent to respond to various initiators of

apoptosis (Kokoszka, Waymire et al. 2004). Thus the essential role of the ADP/ATP translocase

in the formation of the mtPTP could be now questioned.

VI.b Uncoupling effect of fatty acids, basal proton leak and the ADP/ATP translocase

Basal proton leak through the inner mitochondria membrane, bypassing the ATP synthase (or

uncoupling proteins), is known to contribute considerably to the respiration rate in isolated

mitochondria and in intact cells, sometimes being as high as 50% of the respiration, measured in

vitro (Rolfe and Brand 1996; Brand 2005). Such leaks lead to lowering of the efficiency of

respiration in terms of decreasing the amount of ATP synthesized per oxygen atom consumed.

The molecular basis of the leak is still largely unknown, although it was shown not to be

dependable on phospholipid composition of the inner mitochondrial membrane (Brookes, Rolfe

et al. 1997). Recently, in mitochondria isolated from skeletal muscle of mice lacking the Ant1

isoform of the ADP/ATP translocase, decreased basal proton leak was observed (Brand, Pakay et

al. 2005). Overexpressed ADP/ATP translocase in Drosophila mitochondria in the same study

showed an increase in proton conductance, and disruption of one of the two genes of Drosophila

ADP/ATP translocase (underexpression) resulted in a decrease in basal proton leak (Brand,

Pakay et al. 2005). A role for the ADP/ATP translocase in mediating basal proton conductance,

which is independent of fatty-acid induced proton conductance, was thus proposed.

37

We have shown (Paper III) that in brown-fat mitochondria, a large fraction of basal

respiration was sensitive to CATR and suggested a role for different ADP/ATP translocase

isoforms in proton translocation under different conditions.

As was noted above, the ADP/ATP translocase belongs to the same carrier family

(SLC25), as UCP1 and is known to share a significant similarity in protein primary, and

possibly, secondary structures. The ADP/ATP translocase was thought for a long time to be a

possible contributor (along with several other mitochondrial transporters) to fatty-acid induced

uncoupling that was proposed to mediate thermoregulatory heat production in addition to UCP1

in BAT, or instead of UCP1 in other tissues (Brustovetsky, Dedukhova et al. 1990; Brustovetsky

and Klingenberg 1994; Skulachev 1999; Mokhova and Khailova 2005). Reported fatty-acid

induced uncoupling found in liver and skeletal muscle mitochondria was proposed to be

mediated by the ADP/ATP translocase (Andreyev, Bondareva et al. 1989). All three inhibitors of

the ANT (CATR, ATR and BA) were able to reverse the respiration activated by palmitate, each

to a various degree (Andreyev, Bondareva et al. 1989; Brustovetsky, Dedukhova et al. 1990). In

heart mitochondria of rats exposed to cold for 48h, a more pronounced CATR effect on

membrane potential was observed in comparison to non-cold exposed rats (Simonyan and

Skulachev 1998).

In brown adipose tissue mitochondria, UCP1-independent fatty-acid proton conductance

was observed in UCP1-KO mice (Shabalina, Jacobsson et al. 2004). In the present work, we

have investigated if the ADP/ATP translocase could be involved in this process, substituting the

lack of UCP1 in mitochondria of UCP1-KO mice (Paper III). The activity of the ANT in

mitochondria of wild-type and KO-mice was inhibited by CATR and the inhibitory effect on

fatty acid-induced uncoupling was low in brown-fat mitochondria from wild type and UCP1-KO

mice, compared to liver mitochondria, implicating minimal ANT involvement in this process.

Moreover, in UCP1-KO mice brown-fat mitochondria, CATR effect was lower than in wild type

mice, although the ANT protein content was higher in UCP1-KO mice. Thus, a role of ANT as a

possible substitute of UCP1 in non-shivering thermogenesis seems, in the light of our results,

implausible.

38

VII. Mitochondria and ROS production

In the process of electron transfer performed by the mitochondrial respiratory chain, the

final step is accompanied by utilization of the O2 as an acceptor of four electrons and, with the

use of four protons, its reduction to two molecules of H2O. However, along the respiratory chain,

several sites are known to be potential contributors to the generation of free radicals, when

electrons are “prematurely” transferred to O2, (giving rise to the superoxide anion, O2-) or other

electron acceptors. Respiratory complexes I and III are believed to be the major contributors to

the O2- production. Generated O2

- can be converted to hydrogen peroxide H2O2, which, in the

presence of transition metals, can be converted to the highly toxic hydroxyl radical (OH-). All

three so called reactive oxygen species (ROS), H2O2, OH-and O2-, are able to damage lipids,

proteins and nucleic acid molecules within mitochondria (and the whole cell) directly or

indirectly. A decline in mitochondrial energy production and an increase in oxidative stress can

lead to the mitochondrial permeability transition pore (mtPTP) opening and initiation of

apoptosis. A variety of degenerative diseases have now been shown to be caused by mutations in

mitochondrial genes encoded by the mitochondrial DNA (mtDNA) or the nuclear DNA and

generation of free radicals was proposed to be a probable cause of aging (free-radical theory of

ageing) (reviewed in (Lieber and Karanjawala 2004; Balaban, Nemoto et al. 2005), although

studies in aging mice models that have appeared recently do not fully support this theory

(Kujoth, Hiona et al. 2005; Trifunovic, Hansson et al. 2005).

Inhibition of the mitochondrial electron transport chain could also be one of the possible

triggers to increase mitochondrial ROS production, thus increasing oxidative stress in the

mitochondria. In one of the initial investigations of causes of ROS production, it was shown that

inhibition of ATP synthesis by depleting ADP in mitochondria or inhibition of respiratory chain

complexes by specific inhibitors could increase accumulation of hydrogen peroxide or

superoxide molecules (Loschen, Flohe et al. 1971; Turrens and Boveris 1980). Already

mentioned in a previous section, mice lacking the ANT1 gene exhibit an increased production of

hydrogen peroxide (H2O2), increased expression of two enzymes that degrade ROS (glutathione

peroxidase-1 and SOD2 (MnSOD)) in skeletal muscle mitochondria, and show increased damage

in mtDNA, supposedly a direct effect of blocking ADP/ATP transport and thus affecting the

respiratory chain function (Esposito, Melov et al. 1999). Interestingly, mitochondria isolated

from heart (another tissue with prevalent expression of ANT1) of knockout mice produced

greater amounts of H2O2 than the knockout skeletal muscle mitochondria, but did not show an

increase in SOD2 (compared to wild-type controls), and only a moderate increase in glutathione

39

peroxidase-1, therefore the oxidative damage was greater, which resulted in greater damage in

mtDNA, compared to skeletal muscle (Esposito, Melov et al. 1999).

Animal cells have developed several mechanisms to defend themselves against the

harmful effects of ROS production. Several enzymes were described that are active in different

cellular compartments and are able to utilize different substrates. Superoxide anion (O2-) is

reduced by superoxide dismutases (SOD) to hydrogen peroxide (H2O2). The SOD has three

isoforms, SOD1 functioning in the cytosol, SOD2 (or MnSOD) in mitochondria and SOD3 in the

extracellular space. After generation of H2O2 by the SODs, the H2O2 can be further reduced to

water by glutathione peroxidase, located in mitochondria and cytosol, or by catalase, located

primarily in peroxisomes. A vast amount of literature, describing also non-enzymatic defense

mechanisms that include tocopherol (vitamin E), carotenoids, CoQH2, ascorbic acid and several

others, i.e. anti-oxidants, exist on the subject but will not be discussed here (for review, see

(Andreyev, Kushnareva et al. 2005).

It was proposed that in addition to the scavenging of ROS, mitochondria could also

possess some mechanisms for prevention of ROS production. One of the many ideas of how

ROS production could be decreased turned investigators’ attention to study possible activation

by ROS or ROS products of proteins that could mediate proton leak through the inner

mitochondrial membrane (thus decreasing membrane potential), i.e. uncoupling protein(s) and

the ATP/ADP translocase. In the sections below, the possible role of these proteins in protection

against oxidative damage is discussed in more detail.

VII.a Role of 4-Hydroxy-2-nonenal in activation of uncoupling proteins

As a result of the mitochondrial production of ROS, polyunsaturated fatty acids in

mitochondrial membranes can undergo the process of lipid peroxidation, whereby various

aldehydes and alkenals are formed. The latter can, in their turn, modify membrane proteins and

are also able to diffuse from the membranes and react with proteins and nucleic acids outside the

mitochondria. One of the most studied groups of products of lipid peroxidation are 4-

hydroxyalkenals, in particular, 4-hydroxynonenal (4-HNE, Fig. 5)(Esterbauer, Schaur et al.

1991).

40

Figure 5 4-Hydroxynonenal (4-HNE).

Primarily discovered as biologically active water-soluble products of auto-oxidation of

linoleic and arachidonic acids in the 60’s, 4-hydroxyalkenals were then studied intensely for

their potential inhibitory effects on the growth of tumor cells and, later, for cytotoxic effects

naturally occurring as a result of lipid peroxidation in biological systems (Schauenstein 1967;

Benedetti, Comporti et al. 1980).

4-Hydroxynonenal is highly reactive toward proteins with SH-groups, and under certain

conditions (slightly alkaline pH and an increased concentration of 4-hydroxynonenal), it was

shown to be able to form bonds with nearly all amino acids (Esterbauer, Schaur et al. 1991). It

was found that inactivation of various enzymes by 4-HNE binding with cysteine (or to a lesser

extent, lysine) is nevertheless reversible, if an excess of external SH-groups is provided (cysteine

or glutathione) (Schauenstein 1967). DNA (and RNA) can be modified by 4-HNE by cyclization

of deoxyguanosine or deoxyadenosine (Sodum and Chung 1988; Sodum and Chung 1991). 4-

Hydroxynonenal in concentrations 1-20 μM (possible concentrations under oxidative stress) can

inhibit DNA and protein synthesis, and above that level is considered cytotoxic and causes cell

death (for review see (Esterbauer, Schaur et al. 1991). It has also been observed that 4-HNE can

activate JNK and p38 MAP kinases and induce transcription of several genes involved in cellular

response to a stress in a number of cell types (Uchida, Shiraishi et al. 1999; Song, Soh et al.

2001; Zhang, Dickinson et al. 2005). A group of enzymes that are resposible for the

detoxification and removal of 4-HNE from cells include glutathione S-transferases (found in the

cytosol), aldehyde dehydrogenases (mitochondrial enzymes found in many tissues) and alcohol

dehydrogenases (found in the cytosol of hepatocytes).

Apart from the inhibitory action of 4-HNE on mitochondrial proteins, another function

for it was proposed recently, namely as a signaling molecule, which should provide a negative

feedback response to mitochondria and mediate a reduction in ROS production (Echtay, Esteves

et al. 2003). This function of alkenals was first proposed after the observation that superoxide

could activate proton conductance in mitochondria from various tissues (Echtay, Roussel et al.

41

2002). This effect was GDP-inhibitable and correlated with the presence or absence of

uncoupling proteins in mitochondria from tissues studied. This work prompted a series of

follow-up publications that attempted to explain observed effects in connection with uncoupling

protein activities stimulated by various products of oxidative stress.

4-HNE and other structurally related chemical compounds were proposed to activate

mitochondrial anion carriers (uncoupling proteins and ATP/ADP translocase) that could provide

a mild uncoupling, reducing membrane potential and thus decreasing oxidative damage (Echtay,

Esteves et al. 2003). Activation of oxygen consumption induced by alkenals could be inhibited

by GDP (UCP1 inhibitor) or CATR/BA (ATP/ADP translocase inhibitors) in mitochondria from

tissues expressing different uncoupling proteins (kidney (UCP2), skeletal muscle (UCP3)), in a

non-additive manner. In tissues without UCPs (liver, skeletal muscle from UCP3-KO mice)

CATR-sensitive activation of respiration with 4-HNE was visible, but no inhibitory effect of

GDP was observed (Echtay, Esteves et al. 2003). Activation of UCP1 was investigated in brown

adipose tissue from warm-acclimated mice or in UCP1-overexpressing yeast and again a GDP-

inhibitable effect on activation of respiration was observed. In this article, the 4-HNE-mediated

effect on proton conductance was not dependent on the presence of free fatty acids. Recent

investigations of the role of 4-HNE and related alkenals on proton conductance in yeast

mitochondria expressing UCP1 and in rat brown adipose tissue mitochondria, however, stressed

a defining role of the presence of palmitate in mediating the observed 4-HNE effect on UCP1

activity (Esteves, Parker et al. 2006). As a result of these studies, an effect of 4-HNE on

activation of uncoupling proteins and ANT was proposed through modification of particular

residues that would aid proton translocation.

We have studied the possible 4-HNE-mediated activation of UCP1 in brown adipose

tissue mitochondria (Paper IV). We have used UCP1-knockout mice (Enerback, Jacobsson et al.

1997) as a control that allowed us to distinguish between UCP1-dependent and independent

effects of the 4-HNE on respiration. In contrast to the results described above, we were unable to

observe a re-activation of GDP-inhibited UCP1 or activation of uninhibited UCP1 by 4-HNE.

We were also unable to detect any significant differences in accumulation of nonenal-modified

proteins (adducts) between wild-type mice and knockouts, as well as no covalent modification of

UCP1 (a proposed feedback signal to decrease subsequent ROS production) was observed in

warm- or cold-acclimated animals. Thus, in the course of the present study, we could not

conclude that ROS or ROS products, such as 4-HNE, are involved in regulation of UCP1 activity

and it is unlikely that UCP1 itself can function as an antioxidant.

42

In the (Paper V), we have investigated the possible activation of UCP3 protein by 4-

HNE; however, we could not observe any significant differences in accumulation of nonenal-

modified proteins (adducts) between wild-type mice and knockouts.

Thus, in the light of results obtained in our group, including (Silva, Shabalina et al. 2005), the

role of UCP1 and UCP3 in preventing ROS production by uncoupling (or any other means)

seems unfeasible.

43

VIII. Comments on methods All methods used in this study have been described in the original publications included at the

end of the thesis, or in the references therein. However, there are several methodological issues

that I would like to emphasize specifically in this section

REMSA

To characterize cis-acting elements within RNA, which could possibly bind to regulatory

proteins, several techniques could be employed. One of the first choice techniques is usually the

RNA electrophoretic mobility shift assay (REMSA, Paper II), where radioactively labeled RNA

is incubated with protein extract and then electrophoretically separated in a polyacrylamide gel

under native conditions to detect RNA-protein complexes formed (Thomson, Rogers et al. 1999).

REMSA is a relatively straightforward method; however, it requires a number of optimization

steps. To reduce unspecific RNA-protein interactions, we have included heparin in the reactions,

since heparin, being a negatively charged anion, serves to mimic the RNA phosphate backbone.

Another component, RNase T1, which is usually added to the reaction after the presumed RNA-

protein complexes are formed, attacks the 3’ phosphate group of guanosine residues of the

unprotected RNA, thus only the RNA sequences that are protein-bound stay intact. Low- and

high-affinity RNA-protein complexes differ in susceptibility to RNase T1 attack, therefore an

optimization of RNase T1 concentrations is required to be able to fully characterize all possible

RNA-protein interactions. Of general consideration is also the fact that the double-stranded RNA

regions are less susceptible to RNase T1 attack. To resolve this, another RNase (RNase A), in

conjunction with RNase T1, could be used. However, in our experiments the use of RNase A

negatively affected formation of RNA-protein complexes and we therefore confined the use of

RNases to only the T1.

The protein extract S130 has been previously described (Brewer and Ross 1990), and had been

proven useful to assay mRNA turnover in different cells.

Relative Quantitative RT-PCR

To characterize gene expression, reverse transcription (RT) coupled with PCR has proved to be

an indispensable methodological tool and is widely used in modern research. Alternative and

more traditional non-PCR-based methods, used for measuring steady-state levels of RNA, such

as Northern blotting, nuclease protection assays and in situ hybridization, have several

advantages but have one main disadvantage, a limited sensitivity. Methods based on RT-PCR

allow estimation of RNA levels independently of the amount of starting material or transcript

44

abundance, but also have several drawbacks, that include a requirement of extensive

optimization, sensitivity to contaminations and cost-inefficiency compared to non-PCR-based

methods.

In my experiments, I have used relative-quantitative RT-PCR (Paper I) to estimate mRNA

levels, which is based on measuring the levels of the mRNA of interest (copied into cDNA),

amplified in a “multiplex” PCR together with an endogenous control (18S rRNA). The levels of

product from the gene are then normalized against the product from the control. The

optimization procedure in this method includes determination of the linear range of amplification

(when amplification proceeds at a constant exponential rate) for every primer pair, so that the

reliable quantification of PCR products could be done. To achieve amplification of the 18S

rRNA endogenous control to be in the same linear range as the mRNA (cDNA) of interest,

optimization of the ratio of 18S primers to modified 18S primers (provided in a kit by Ambion)

is required. Detection of the PCR product was done in agarose gels stained with ethidium

bromide (EtBr), although this is not a preferable detection agent due to it relative insensitivity.

An alternative and better option is to include in the PCR traces of P32-dATP or end-label PCR

primers and resolve the PCR products on a polyacryamide gel. The major advantages of relative-

quantitative RT-PCR are its simplicity (no need for extensive optimization in comparison to real-

time RT-PCR); use of amplification data within the linear range also allows accurate

quantification of PCR products.

45

Conclusions In the course of the present investigation, we have attempted to clarify several aspects of

mitochondrial metabolism that ultimately affect ATP/O ratio in various tissues. In brown

adipose tissue mitochondria, the main focus of the present studies, ATP synthesis is not a

preferred pathway to utilize the energy from oxidative processes, which is instead released as

heat through the activity of the brown adipose tissue-specific uncoupling protein UCP1. By the

use of different murine models, we were able to verify key player(s) in the regulation of the final

content of the ATP synthase in brown-fat mitochondria and to test hypotheses that have been

proposed over the years for the roles of several mitochondrial carriers (ATP/ADP carrier, UCP3

and UCP1) to mediate regulated proton leaks (uncoupling) as a result of induced activation by

various agents.

Based on the results obtained and presented in the current study, it is possible to draw several

conclusions:

1) The final (low) content of the ATP synthase in brown adipose tissue is dependent on the

availability of the c-Fo subunit.

2) The post-transcriptional regulation of the ATP synthase β-F1 subunit the gene expression

of which is induced in parallel with other oxidative phosphorylation (OXPHOS) genes in

brown adipose tissue, could depend on the presence of two evolutionarily conserved

regions in its 3’UTR.

3) UCP1 activity in brown adipose tissue cannot be induced by reactive oxygen species

(ROS) products (4-hydroxynonenal).

4) The ATP/ADP carrier in brown adipose tissue is not involved in fatty-acid induced

uncoupling.

5) UCP3 is not an uncoupling protein per se, and no such proposed activity of UCP3 could

be induced by fatty-acids or ROS, which points to the existence of other functions for this

protein in vivo.

46

Acknowledgements

This study was carried out at the Department of Physiology, The Wenner-Gren Institute, Stockholm University. I would like to thank everyone who has helped me along the way, and especially following people: Barbara Cannon, for scientific freedom and for letting me run things at my own pace, and for financial and other support I received throughout these years. Jan Nedergaard, who taught me the basics of scientific writing and how to critically assess data, and for ever so many ideas and suggestions. Irina Shabalin, for getting me interested in mitochondria, for ideas, for motivation, for encouragement and support. Lars Wieslander (and all of his group members at the time I worked there) - for the opportunity to learn so much about the molecular biology in practice. My collaborators, Hana Antonicka and Josef Houstek, for intellectual and practical input to the ATP synthase-related research; Ulf Andersson and Rolf Westerberg, for starting up the P1-transgenic project. I also would like to thank all of my friends and colleagues in the department: Elisabeth Bergner - for handling all of the organizational problems that otherwise would be nearly insuperable to me. Britta Alsterman and Helene Allbrink, for their helpfulness and keeping things running smoothly and nicely. Stig Sundelin, Richard Ekström and Robert Persson, for helping me with computers, programs, broken freezers, etc. Dayana Alonso, Eva Nygren and Solveig Sundberg in the Animal house - for competent help in handling my precious mice. Bengt Håkansson, for taking care of numerous matters in the lab. Our cheerful Russian team: Irina Sabanova - for conversations on PCR- and non-PCR-related topics, Katja Chernogubova and Julia Nevzorova - for great company in- and outside of the laboratory, Viktor Sabanov - for subtle discussions on linguistics and politics. My colleagues in the lab, each and everyone of you, I’d like to thank for so many things that made our department such a great place to be in: Emma Backlund and Katarina Lennström, Annelie Brolinson, Tore Bengtsson, Olof Dallner, Helena Feldmann, Therese Holmström, Dana Hutchinson, Anders Jacobsson, Andreas Jakobsson, Johanna Jörgensen, Tsutomu Kobayashi, Birgitta Leksell, Charlotta Mattson, Natasa Petrovic, Tomas Walden, Daniel Yamamoto and Damir Zadravec - it’s been fun to work alongside with all of you! My friends outside of the department, who brightened up my everyday life, especially my dear friends Petra and Tor - can’t thank you enough, guys. Also I would specially like to thank Gennady Bronnikov, for scientific advice and ideas on ATP synthase research and help in so many other respects. And finally, I’d like to thank my parents, for always being there for me, believing in me and for encouraging me to become a scientist. …and Maxim - without your love and support I would never make it through.

47

References Abrahams, J. P., A. G. Leslie, et al. (1994). "Structure at 2.8 A resolution of F1-ATPase from

bovine heart mitochondria." Nature 370(6491): 621-8. Ackerman, S. H. (2002). "Atp11p and Atp12p are chaperones for F(1)-ATPase biogenesis in

mitochondria." Biochim Biophys Acta 1555(1-3): 101-5. Ackerman, S. H. and A. Tzagoloff (1990). "ATP 10, a yeast nuclear gene required for the

assembly of the mitochondrial F1-F0 complex." J Biol Chem 265(17): 9952-9. Ackerman, S. H. and A. Tzagoloff (1990). "Identification of two nuclear genes (ATP11, ATP12)

required for assembly of the yeast F1-ATPase." Proc Natl Acad Sci U S A 87(13): 4986-90.

Ackerman, S. H. and A. Tzagoloff (2005). "Function, structure, and biogenesis of mitochondrial ATP synthase." Prog Nucleic Acid Res Mol Biol 80: 95-133.

Ades, I. Z. and R. A. Butow (1980). "The products of mitochondria-bound cytoplasmic polysomes in yeast." J Biol Chem 255(20): 9918-24.

Andersson, U., H. Antonicka, et al. (2000). "A novel principle for conferring selectivity to poly(A)-binding proteins: interdependence of two ATP synthase beta-subunit mRNA-binding proteins." Biochem J 346 Pt 1: 33-9.

Andersson, U., J. Houstek, et al. (1997). "ATP synthase subunit c expression: physiological regulation of the P1 and P2 genes." Biochem J 323 ( Pt 2): 379-85.

Andrews, Z. B., S. Diano, et al. (2005). "Mitochondrial uncoupling proteins in the CNS: in support of function and survival." Nat Rev Neurosci 6(11): 829-40.

Andreyev, A., T. O. Bondareva, et al. (1989). "The ATP/ADP-antiporter is involved in the uncoupling effect of fatty acids on mitochondria." Eur J Biochem 182(3): 585-92.

Andreyev, A. Y., Y. E. Kushnareva, et al. (2005). "Mitochondrial metabolism of reactive oxygen species." Biochemistry (Mosc) 70(2): 200-14.

Antonsson, B. (2004). "Mitochondria and the Bcl-2 family proteins in apoptosis signaling pathways." Mol Cell Biochem 256-257(1-2): 141-55.

Aquila, H., D. Misra, et al. (1982). "Complete amino acid sequence of the ADP/ATP carrier from beef heart mitochondria." Hoppe Seylers Z Physiol Chem 363(3): 345-9.

Arechaga, I., P. J. Butler, et al. (2002). "Self-assembly of ATP synthase subunit c rings." FEBS Lett 515(1-3): 189-93.

Attardi, G. and G. Schatz (1988). "Biogenesis of mitochondria." Annu Rev Cell Biol 4: 289-333. Balaban, R. S., S. Nemoto, et al. (2005). "Mitochondria, oxidants, and aging." Cell 120(4): 483-

95. Basu, A., N. Lenka, et al. (1997). "Regulation of murine cytochrome oxidase Vb gene expression

in different tissues and during myogenesis. Role of a YY-1 factor-binding negative enhancer." J Biol Chem 272(9): 5899-908.

Bauer, M. F., S. Hofmann, et al. (2000). "Protein translocation into mitochondria: the role of TIM complexes." Trends Cell Biol 10(1): 25-31.

Bauer, M. K. A., A. Schubert, et al. (1999). "Adenine Nucleotide Translocase-1, a Component of the Permeability Transition Pore, Can Dominantly Induce Apoptosis." J. Cell Biol. 147(7): 1493-1502.

Benedetti, A., M. Comporti, et al. (1980). "Identification of 4-hydroxynonenal as a cytotoxic product originating from the peroxidation of liver microsomal lipids." Biochim Biophys Acta 620(2): 281-96.

Boyer, P. D. (1993). "The binding change mechanism for ATP synthase--some probabilities and possibilities." Biochim Biophys Acta 1140(3): 215-50.

Boyer, P. D. (1997). "The ATP synthase--a splendid molecular machine." Annu Rev Biochem 66: 717-49.

48

Boyer, P. D. (2001). "Toward an adequate scheme for the ATP synthase catalysis." Biochemistry (Mosc) 66(10): 1058-66.

Boyer, P. D. (2002). "Catalytic site occupancy during ATP synthase catalysis." FEBS Lett 512(1-3): 29-32.

Boyer, P. D. (2002). "A research journey with ATP synthase." J Biol Chem 277(42): 39045-61. Brand, M. D. (2005). "The efficiency and plasticity of mitochondrial energy transduction."

Biochem Soc Trans 33(Pt 5): 897-904. Brand, M. D. and T. C. Esteves (2005). "Physiological functions of the mitochondrial

uncoupling proteins UCP2 and UCP3." Cell Metab 2(2): 85-93. Brand, M. D., J. L. Pakay, et al. (2005). "The basal proton conductance of mitochondria depends

on adenine nucleotide translocase content." Biochem J 392(Pt 2): 353-62. Breen, G. A. and E. M. Jordan (1997). "Regulation of the nuclear gene that encodes the alpha-

subunit of the mitochondrial F0F1-ATP synthase complex. Activation by upstream stimulatory factor 2." J Biol Chem 272(16): 10538-42.

Breen, G. A. and E. M. Jordan (1998). "Upstream stimulatory factor 2 activates the mammalian F1F0 ATP synthase alpha-subunit gene through an initiator element." Gene Expr 7(3): 163-70.

Breen, G. A. and E. M. Jordan (1999). "Transcriptional activation of the F(1)F(0) ATP synthase alpha-subunit initiator element by USF2 is mediated by p300." Biochim Biophys Acta 1428(2-3): 169-76.

Breen, G. A., C. A. Vander Zee, et al. (1996). "Nuclear factor YY1 activates the mammalian F0F1 ATP synthase alpha-subunit gene." Gene Expr 5(3): 181-91.

Brewer, G. and J. Ross (1990). "Messenger RNA turnover in cell-free extracts." Methods Enzymol 181: 202-9.

Brookes, P. S., D. F. Rolfe, et al. (1997). "The proton permeability of liposomes made from mitochondrial inner membrane phospholipids: comparison with isolated mitochondria." J Membr Biol 155(2): 167-74.

Brustovetsky, N. and M. Klingenberg (1994). "The reconstituted ADP/ATP carrier can mediate H+ transport by free fatty acids, which is further stimulated by mersalyl." J Biol Chem 269(44): 27329-36.

Brustovetsky, N. N., V. I. Dedukhova, et al. (1990). "Inhibitors of the ATP/ADP antiporter suppress stimulation of mitochondrial respiration and H+ permeability by palmitate and anionic detergents." FEBS Lett 272(1-2): 187-9.

Bucheler, K., V. Adams, et al. (1991). "Localization of the ATP/ADP translocator in the inner membrane and regulation of contact sites between mitochondrial envelope membranes by ADP. A study on freeze-fractured isolated liver mitochondria." Biochim Biophys Acta 1056(3): 233-42.

Cam, H., E. Balciunaite, et al. (2004). "A common set of gene regulatory networks links metabolism and growth inhibition." Mol Cell 16(3): 399-411.

Cannon, B. and J. Nedergaard (2004). "Brown adipose tissue: function and physiological significance." Physiol Rev 84(1): 277-359.

Cannon, B., L. Romert, et al. (1977). "Morphology and biochemical properties of perirenal adipose tissue from lamb (Ovis aries). A comparison with brown adipose tissue." Comp Biochem Physiol B 56(1): 87-99.

Cannon, B. and G. Vogel (1977). "The mitochondrial ATPase of brown adipose tissue. Purification and comparison with the mitochondrial ATPase from beef heart." FEBS Lett 76(2): 284-9.

Carbajo, R. J., F. A. Kellas, et al. (2005). "Structure of the F1-binding domain of the stator of bovine F1Fo-ATPase and how it binds an alpha-subunit." J Mol Biol 351(4): 824-38.

Cassard-Doulcier, A. M., M. Larose, et al. (1994). "In vitro interactions between nuclear proteins and uncoupling protein gene promoter reveal several putative transactivating factors

49

including Ets1, retinoid X receptor, thyroid hormone receptor, and a CACCC box-binding protein." J Biol Chem 269(39): 24335-42.

Chau, C. M., M. J. Evans, et al. (1992). "Nuclear respiratory factor 1 activation sites in genes encoding the gamma-subunit of ATP synthase, eukaryotic initiation factor 2 alpha, and tyrosine aminotransferase. Specific interaction of purified NRF-1 with multiple target genes." J Biol Chem 267(10): 6999-7006.

Chen, S. T., C. D. Chang, et al. (1990). "A human ADP/ATP translocase gene has seven pseudogenes and localizes to chromosome X." Somat Cell Mol Genet 16(2): 143-9.

Chung, A. B., G. Stepien, et al. (1992). "Transcriptional control of nuclear genes for the mitochondrial muscle ADP/ATP translocator and the ATP synthase beta subunit. Multiple factors interact with the OXBOX/REBOX promoter sequences." J Biol Chem 267(29): 21154-61.

Clapham, J. C., J. R. Arch, et al. (2000). "Mice overexpressing human uncoupling protein-3 in skeletal muscle are hyperphagic and lean." Nature 406(6794): 415-8.

Collinson, I. R., M. J. Runswick, et al. (1994). "Fo membrane domain of ATP synthase from bovine heart mitochondria: purification, subunit composition, and reconstitution with F1-ATPase." Biochemistry 33(25): 7971-8.

Corral-Debrinski, M., C. Blugeon, et al. (2000). "In yeast, the 3' untranslated region or the presequence of ATM1 is required for the exclusive localization of its mRNA to the vicinity of mitochondria." Mol Cell Biol 20(21): 7881-92.

Cozens, A. L., M. J. Runswick, et al. (1989). "DNA sequences of two expressed nuclear genes for human mitochondrial ADP/ATP translocase." J Mol Biol 206(2): 261-80.

Crompton, M., S. Virji, et al. (1998). "Cyclophilin-D binds strongly to complexes of the voltage-dependent anion channel and the adenine nucleotide translocase to form the permeability transition pore." Eur J Biochem 258(2): 729-35.

De Grassi, A., C. Lanave, et al. (2006). "Evolution of ATP synthase subunit c and cytochrome c gene families in selected Metazoan classes." Gene.

de Heredia, M. L., J. M. Izquierdo, et al. (2000). "A conserved mechanism for controlling the translation of beta-F1-ATPase mRNA between the fetal liver and cancer cells." J Biol Chem 275(10): 7430-7.

Devenish, R. J., M. Prescott, et al. (2000). "Insights into ATP synthase assembly and function through the molecular genetic manipulation of subunits of the yeast mitochondrial enzyme complex." Biochim Biophys Acta 1458(2-3): 428-42.

Dolce, V., P. Scarcia, et al. (2005). "A fourth ADP/ATP carrier isoform in man: identification, bacterial expression, functional characterization and tissue distribution." FEBS Lett 579(3): 633-7.

Dou, C., P. A. Fortes, et al. (1998). "The alpha 3(beta Y341W)3 gamma subcomplex of the F1-ATPase from the thermophilic Bacillus PS3 fails to dissociate ADP when MgATP is hydrolyzed at a single catalytic site and attains maximal velocity when three catalytic sites are saturated with MgATP." Biochemistry 37(47): 16757-64.

Duyckaerts, C., C. M. Sluse-Goffart, et al. (1980). "Kinetic mechanism of the exchanges catalysed by the adenine-nucleotide carrier." Eur J Biochem 106(1): 1-6.

Dyer, M. R. and J. E. Walker (1993). "Sequences of members of the human gene family for the c subunit of mitochondrial ATP synthase." Biochem J 293 ( Pt 1): 51-64.

Echtay, K. S., T. C. Esteves, et al. (2003). "A signalling role for 4-hydroxy-2-nonenal in regulation of mitochondrial uncoupling." Embo J 22(16): 4103-10.

Echtay, K. S., D. Roussel, et al. (2002). "Superoxide activates mitochondrial uncoupling proteins." Nature 415(6867): 96-9.

Egea, G., J. M. Izquierdo, et al. (1997). "mRNA encoding the beta-subunit of the mitochondrial F1-ATPase complex is a localized mRNA in rat hepatocytes." Biochem J 322 ( Pt 2): 557-65.

50

Enerback, S., A. Jacobsson, et al. (1997). "Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese." Nature 387(6628): 90-4.

Ernster, L. and G. Schatz (1981). "Mitochondria: a historical review 10.1083/jcb.91.3.227s." J. Cell Biol. 91(3): 227s-255. Esposito, L. A., S. Melov, et al. (1999). "Mitochondrial disease in mouse results in increased

oxidative stress." Proc Natl Acad Sci U S A 96(9): 4820-5. Esterbauer, H., R. J. Schaur, et al. (1991). "Chemistry and biochemistry of 4-hydroxynonenal,

malonaldehyde and related aldehydes." Free Radic Biol Med 11(1): 81-128. Esteves, T. C., N. Parker, et al. (2006). "Synergy of fatty acid and reactive alkenal activation of

proton conductance through uncoupling protein 1 in mitochondria." Biochem J. Evans, M. J. and R. C. Scarpulla (1989). "Interaction of nuclear factors with multiple sites in the

somatic cytochrome c promoter. Characterization of upstream NRF-1, ATF, and intron Sp1 recognition sequences." J Biol Chem 264(24): 14361-8.

Fan, M., J. Rhee, et al. (2004). "Suppression of mitochondrial respiration through recruitment of p160 myb binding protein to PGC-1alpha: modulation by p38 MAPK." Genes Dev 18(3): 278-89.

Farrell, L. B. and P. Nagley (1987). "Human liver cDNA clones encoding proteolipid subunit 9 of the mitochondrial ATPase complex." Biochem Biophys Res Commun 144(3): 1257-64.

Ferguson, S. J. (2000). "ATP synthase: what dictates the size of a ring?" Curr Biol 10(21): R804-8.

Finck, B. N. and D. P. Kelly (2006). "PGC-1 coactivators: inducible regulators of energy metabolism in health and disease." J Clin Invest 116(3): 615-22.

Fiore, C., V. Trezeguet, et al. (1998). "The mitochondrial ADP/ATP carrier: structural, physiological and pathological aspects." Biochimie 80(2): 137-50.

Gao, Y. Q., W. Yang, et al. (2005). "A Structure-Based Model for the Synthesis and Hydrolysis of ATP by F(1)-ATPase." Cell 123(2): 195-205.

Garesse, R. and C. G. Vallejo (2001). "Animal mitochondrial biogenesis and function: a regulatory cross-talk between two genomes." Gene 263(1-2): 1-16.

Gay, N. J. and J. E. Walker (1985). "Two genes encoding the bovine mitochondrial ATP synthase proteolipid specify precursors with different import sequences and are expressed in a tissue-specific manner." Embo J 4(13A): 3519-24.

Giraud, S., C. Bonod-Bidaud, et al. (1998). "Expression of human ANT2 gene in highly proliferative cells: GRBOX, a new transcriptional element, is involved in the regulation of glycolytic ATP import into mitochondria." J Mol Biol 281(3): 409-18.

Goffart, S. and R. J. Wiesner (2003). "Regulation and co-ordination of nuclear gene expression during mitochondrial biogenesis." Exp Physiol 88(1): 33-40.

Golozoubova, V., E. Hohtola, et al. (2001). "Only UCP1 can mediate adaptive nonshivering thermogenesis in the cold." Faseb J 15(11): 2048-50.

Graham, B. H., K. G. Waymire, et al. (1997). "A mouse model for mitochondrial myopathy and cardiomyopathy resulting from a deficiency in the heart/muscle isoform of the adenine nucleotide translocator." Nat Genet 16(3): 226-34.

Green, D. R. (2005). "Apoptotic Pathways: Ten Minutes to Dead." Cell 121(5): 671-674. Green, D. R. and G. Kroemer (2004). "The Pathophysiology of Mitochondrial Cell Death."

Science 305(5684): 626-629. Hadikusumo, R. G., S. Meltzer, et al. (1988). "The definition of mitochondrial H+ ATPase

assembly defects in mit- mutants of Saccharomyces cerevisiae with a monoclonal antibody to the enzyme complex as an assembly probe." Biochim Biophys Acta 933(1): 212-22.

Halestrap, A. P., K. Y. Woodfield, et al. (1997). "Oxidative stress, thiol reagents, and membrane potential modulate the mitochondrial permeability transition by affecting nucleotide binding to the adenine nucleotide translocase." J Biol Chem 272(6): 3346-54.

51

Haraguchi, Y., A. B. Chung, et al. (1994). "OXBOX and REBOX, overlapping promoter elements of the mitochondrial F0F1-ATP synthase beta subunit gene. OXBOX/REBOX in the ATPsyn beta promoter." J Biol Chem 269(12): 9330-4.

Haworth, R. A. and D. R. Hunter (2000). "Control of the mitochondrial permeability transition pore by high-affinity ADP binding at the ADP/ATP translocase in permeabilized mitochondria." J Bioenerg Biomembr 32(1): 91-6.

Heaton, G. M., R. J. Wagenvoord, et al. (1978). "Brown-adipose-tissue mitochondria: photoaffinity labelling of the regulatory site of energy dissipation." Eur J Biochem 82(2): 515-21.

Helfenbein, K. G., T. P. Ellis, et al. (2003). "ATP22, a nuclear gene required for expression of the F0 sector of mitochondrial ATPase in Saccharomyces cerevisiae." J Biol Chem 278(22): 19751-6.

Himms-Hagen, J. and M. E. Harper (2001). "Physiological role of UCP3 may be export of fatty acids from mitochondria when fatty acid oxidation predominates: an hypothesis." Exp Biol Med (Maywood) 226(2): 78-84.

Holt, I. J., A. E. Harding, et al. (1990). "A new mitochondrial disease associated with mitochondrial DNA heteroplasmy." Am J Hum Genet 46(3): 428-33.

Houldsworth, J. and G. Attardi (1988). "Two distinct genes for ADP/ATP translocase are expressed at the mRNA level in adult human liver." Proc Natl Acad Sci U S A 85(2): 377-81.

Houstek, J., U. Andersson, et al. (1995). "The expression of subunit c correlates with and thus may limit the biosynthesis of the mitochondrial F0F1-ATPase in brown adipose tissue." J Biol Chem 270(13): 7689-94.

Houstek, J. and Z. Drahota (1977). "Purification and properties of mitochondrial adenosine triphosphatase of hamster brown adipose tissue." Biochim Biophys Acta 484(1): 127-39.

Houstek, J., P. Klement, et al. (1995). "Altered properties of mitochondrial ATP-synthase in patients with a T-->G mutation in the ATPase 6 (subunit a) gene at position 8993 of mtDNA." Biochim Biophys Acta 1271(2-3): 349-57.

Houstek, J., T. Mracek, et al. (2004). "Mitochondrial diseases and ATPase defects of nuclear origin." Biochim Biophys Acta 1658(1-2): 115-21.

Houstek, J., P. Tvrdik, et al. (1991). "Low content of mitochondrial ATPase in brown adipose tissue is the result of post-transcriptional regulation." FEBS Lett 294(3): 191-4.

Ichida, M., S. Nemoto, et al. (2002). "Identification of a specific molecular repressor of the peroxisome proliferator-activated receptor gamma Coactivator-1 alpha (PGC-1alpha)." J Biol Chem 277(52): 50991-5.

Itoh, H., A. Takahashi, et al. (2004). "Mechanically driven ATP synthesis by F1-ATPase." Nature 427(6973): 465-8.

Izquierdo, J. M. and J. M. Cuezva (1993). "Evidence of post-transcriptional regulation in mammalian mitochondrial biogenesis." Biochem Biophys Res Commun 196(1): 55-60.

Izquierdo, J. M. and J. M. Cuezva (1993). "Thyroid hormones promote transcriptional activation of the nuclear gene coding for mitochondrial beta-F1-ATPase in rat liver." FEBS Lett 323(1-2): 109-12.

Izquierdo, J. M. and J. M. Cuezva (1997). "Control of the translational efficiency of beta-F1-ATPase mRNA depends on the regulation of a protein that binds the 3' untranslated region of the mRNA." Mol Cell Biol 17(9): 5255-68.

Izquierdo, J. M. and J. M. Cuezva (2000). "Internal-ribosome-entry-site functional activity of the 3'-untranslated region of the mRNA for the beta subunit of mitochondrial H+-ATP synthase." Biochem J 346 Pt 3: 849-55.

Izquierdo, J. M., E. Jimenez, et al. (1995). "Hypothyroidism affects the expression of the beta-F1-ATPase gene and limits mitochondrial proliferation in rat liver at all stages of development." Eur J Biochem 232(2): 344-50.

52

Izquierdo, J. M., A. M. Luis, et al. (1990). "Postnatal mitochondrial differentiation in rat liver. Regulation by thyroid hormones of the beta-subunit of the mitochondrial F1-ATPase complex." J Biol Chem 265(16): 9090-7.

Izquierdo, J. M., J. Ricart, et al. (1995). "Changing patterns of transcriptional and post-transcriptional control of beta-F1-ATPase gene expression during mitochondrial biogenesis in liver." J Biol Chem 270(17): 10342-50.

Jesina, P., M. Tesarova, et al. (2004). "Diminished synthesis of subunit a (ATP6) and altered function of ATP synthase and cytochrome c oxidase due to the mtDNA 2 bp microdeletion of TA at positions 9205 and 9206." Biochem J 383(Pt. 3): 561-71.

Jordan, E. M., T. Worley, et al. (2003). "Transcriptional Regulation of the Nuclear Gene Encoding the alpha-Subunit of the Mammalian Mitochondrial F(1)F(0) ATP Synthase Complex: Role for the Orphan Nuclear Receptor, COUP-TFII/ARP-1." Biochemistry 42(9): 2656-63.

Karlberg, E. O. and S. G. Andersson (2003). "Mitochondrial gene history and mRNA localization: is there a correlation?" Nat Rev Genet 4(5): 391-7.

Kelly, D. P. and R. C. Scarpulla (2004). "Transcriptional regulatory circuits controlling mitochondrial biogenesis and function." Genes Dev 18(4): 357-68.

Klingenberg, M. (1989). "Molecular aspects of the adenine nucleotide carrier from mitochondria." Arch Biochem Biophys 270(1): 1-14.

Kokoszka, J. E., K. G. Waymire, et al. (2004). "The ADP/ATP translocator is not essential for the mitochondrial permeability transition pore." Nature 427(6973): 461-5.

Kopecky, J., M. Baudysova, et al. (1990). "Synthesis of mitochondrial uncoupling protein in brown adipocytes differentiated in cell culture." J Biol Chem 265(36): 22204-9.

Krauss, S., C. Y. Zhang, et al. (2005). "The mitochondrial uncoupling-protein homologues." Nat Rev Mol Cell Biol 6(3): 248-61.

Kujoth, G. C., A. Hiona, et al. (2005). "Mitochondrial DNA mutations, oxidative stress, and apoptosis in mammalian aging." Science 309(5733): 481-4.

Lefebvre-Legendre, L., J. Vaillier, et al. (2001). "Identification of a nuclear gene (FMC1) required for the assembly/stability of yeast mitochondrial F(1)-ATPase in heat stress conditions." J Biol Chem 276(9): 6789-96.

Levy, S. E., Y. S. Chen, et al. (2000). "Expression and sequence analysis of the mouse adenine nucleotide translocase 1 and 2 genes." Gene 254(1-2): 57-66.

Li, K., J. A. Hodge, et al. (1990). "OXBOX, a positive transcriptional element of the heart-skeletal muscle ADP/ATP translocator gene." J Biol Chem 265(33): 20585-8.

Li, Y., N. Johnson, et al. (2004). "Cyclophilin-D promotes the mitochondrial permeability transition but has opposite effects on apoptosis and necrosis." Biochem J 383(Pt 1): 101-9.

Lieber, M. R. and Z. E. Karanjawala (2004). "Ageing, repetitive genomes and DNA damage." Nat Rev Mol Cell Biol 5(1): 69-75.

Lin, J., C. Handschin, et al. (2005). "Metabolic control through the PGC-1 family of transcription coactivators." Cell Metab 1(6): 361-70.

Lindberg, O., J. de Pierre, et al. (1967). "Studies of the mitochondrial energy-transfer system of brown adipose tissue." J Cell Biol 34(1): 293-310.

Loschen, G., L. Flohe, et al. (1971). "Respiratory chain linked H(2)O(2) production in pigeon heart mitochondria." FEBS Lett 18(2): 261-264.

Luciakova, K. and B. D. Nelson (1992). "Transcript levels for nuclear-encoded mammalian mitochondrial respiratory-chain components are regulated by thyroid hormone in an uncoordinated fashion." Eur J Biochem 207(1): 247-51.

Luis, A. M., J. M. Izquierdo, et al. (1993). "Translational regulation of mitochondrial differentiation in neonatal rat liver. Specific increase in the translational efficiency of the nuclear-encoded mitochondrial beta-F1-ATPase mRNA." J Biol Chem 268(3): 1868-75.

53

Lunardi, J. and G. Attardi (1991). "Differential regulation of expression of the multiple ADP/ATP translocase genes in human cells." J Biol Chem 266(25): 16534-40.

Margeot, A., C. Blugeon, et al. (2002). "In Saccharomyces cerevisiae, ATP2 mRNA sorting to the vicinity of mitochondria is essential for respiratory function." Embo J 21(24): 6893-904.

Martin, I., J. A. Villena, et al. (1996). "Influence of thyroid hormones on the human ATP synthase beta-subunit gene promoter." Mol Cell Biochem 154(2): 107-11.

Marzo, I., C. Brenner, et al. (1998). "Bax and adenine nucleotide translocator cooperate in the mitochondrial control of apoptosis." Science 281(5385): 2027-31.

Matsuda, C., H. Endo, et al. (1993). "Gene structure of human mitochondrial ATP synthase gamma-subunit. Tissue specificity produced by alternative RNA splicing." J Biol Chem 268(33): 24950-8.

Menz, R. I., J. E. Walker, et al. (2001). "Structure of bovine mitochondrial F(1)-ATPase with nucleotide bound to all three catalytic sites: implications for the mechanism of rotary catalysis." Cell 106(3): 331-41.

Mokhova, E. N. and L. S. Khailova (2005). "Involvement of mitochondrial inner membrane anion carriers in the uncoupling effect of fatty acids." Biochemistry (Mosc) 70(2): 159-63.

Mokranjac, D. and W. Neupert (2005). "Protein import into mitochondria." Biochem Soc Trans 33(Pt 5): 1019-23.

Mootha, V. K., C. Handschin, et al. (2004). "Erralpha and Gabpa/b specify PGC-1alpha-dependent oxidative phosphorylation gene expression that is altered in diabetic muscle." Proc Natl Acad Sci U S A 101(17): 6570-5.

Mootha, V. K., C. M. Lindgren, et al. (2003). "PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes." Nat Genet 34(3): 267-73.

Neckelmann, N., K. Li, et al. (1987). "cDNA sequence of a human skeletal muscle ADP/ATP translocator: lack of a leader peptide, divergence from a fibroblast translocator cDNA, and coevolution with mitochondrial DNA genes." Proc Natl Acad Sci U S A 84(21): 7580-4.

Neckelmann, N., C. K. Warner, et al. (1989). "The human ATP synthase beta subunit gene: sequence analysis, chromosome assignment, and differential expression." Genomics 5(4): 829-43.

Nedergaard, J. and B. Cannon (2003). "The 'novel' 'uncoupling' proteins UCP2 and UCP3: what do they really do? Pros and cons for suggested functions." Exp Physiol 88(1): 65-84.

Nedergaard, J., V. Golozoubova, et al. (2001). "Life without UCP1: mitochondrial, cellular and organismal characteristics of the UCP1-ablated mice." Biochem Soc Trans 29(Pt 6): 756-63.

Nedergaard, J., N. Petrovic, et al. (2005). "PPARgamma in the control of brown adipocyte differentiation." Biochim Biophys Acta 1740(2): 293-304.

Nedergaard, J., D. Ricquier, et al. (2005). "Uncoupling proteins: current status and therapeutic prospects." EMBO Rep 6(10): 917-21.

Nelson, B. D., K. Luciakova, et al. (1995). "The role of thyroid hormone and promoter diversity in the regulation of nuclear encoded mitochondrial proteins." Biochim Biophys Acta 1271(1): 85-91.

Nijtmans, L. G., P. Klement, et al. (1995). "Assembly of mitochondrial ATP synthase in cultured human cells: implications for mitochondrial diseases." Biochim Biophys Acta 1272(3): 190-8.

Nishio, K., A. Iwamoto-Kihara, et al. (2002). "Subunit rotation of ATP synthase embedded in membranes: a or beta subunit rotation relative to the c subunit ring." Proc Natl Acad Sci U S A 99(21): 13448-52.

54

Noji, H., R. Yasuda, et al. (1997). "Direct observation of the rotation of F1-ATPase." Nature 386(6622): 299-302.

Nury, H., C. Dahout-Gonzalez, et al. (2006). "Relations Between Structure and Function of the Mitochondrial ADP/ATP Carrier." Annu Rev Biochem.

Ohta, S., H. Tomura, et al. (1988). "Gene structure of the human mitochondrial adenosine triphosphate synthase beta subunit." J Biol Chem 263(23): 11257-62.

Ongwijitwat, S., H. L. Liang, et al. (2006). "Nuclear respiratory factor 2 senses changing cellular energy demands and its silencing down-regulates cytochrome oxidase and other target gene mRNAs." Gene.

Palmieri, F. (1994). "Mitochondrial carrier proteins." FEBS Letters 346(1): 48-54. Palmieri, F. (2004). "The mitochondrial transporter family (SLC25): physiological and

pathological implications." Pflugers Arch 447(5): 689-709. Papa, S., F. Zanotti, et al. (2000). "The structural and functional connection between the catalytic

and proton translocating sectors of the mitochondrial F1F0-ATP synthase." J Bioenerg Biomembr 32(4): 401-11.

Paul, M. F., S. Ackerman, et al. (1994). "Cloning of the yeast ATP3 gene coding for the gamma-subunit of F1 and characterization of atp3 mutants." J Biol Chem 269(42): 26158-64.

Paul, M. F., A. Barrientos, et al. (2000). "A single amino acid change in subunit 6 of the yeast mitochondrial ATPase suppresses a null mutation in ATP10." J Biol Chem 275(38): 29238-43.

Pebay-Peyroula, E. and G. Brandolin (2004). "Nucleotide exchange in mitochondria: insight at a molecular level." Curr Opin Struct Biol 14(4): 420-5.

Pebay-Peyroula, E., C. Dahout-Gonzalez, et al. (2003). "Structure of mitochondrial ADP/ATP carrier in complex with carboxyatractyloside." Nature 426(6962): 39-44.

Pena, P., C. Ugalde, et al. (1995). "Analysis of the mitochondrial ATP synthase beta-subunit gene in Drosophilidae: structure, transcriptional regulatory features and developmental pattern of expression in Drosophila melanogaster." Biochem J 312 ( Pt 3): 887-97.

Pfanner, N., N. Wiedemann, et al. (2004). "Assembling the mitochondrial outer membrane." Nat Struct Mol Biol 11(11): 1044-8.

Pickova, A., J. Paul, et al. (2003). "Differential expression of ATPAF1 and ATPAF2 genes encoding F(1)-ATPase assembly proteins in mouse tissues." FEBS Lett 551(1-3): 42-6.

Pickova, A., M. Potocky, et al. (2005). "Assembly factors of F1FO-ATP synthase across genomes." Proteins 59(3): 393-402.

Prescott, M., N. C. Bush, et al. (1994). "Properties of yeast cells depleted of the OSCP subunit of mitochondrial ATP synthase by regulated expression of the ATP5 gene." Biochem Mol Biol Int 34(4): 789-99.

Puigserver, P., Z. Wu, et al. (1998). "A cold-inducible coactivator of nuclear receptors linked to adaptive thermogenesis." Cell 92(6): 829-39.

Rehling, P., K. Model, et al. (2003). "Protein insertion into the mitochondrial inner membrane by a twin-pore translocase." Science 299(5613): 1747-51.

Ricart, J., G. Egea, et al. (1997). "Subcellular structure containing mRNA for beta subunit of mitochondrial H+-ATP synthase in rat hepatocytes is translationally active." Biochem J 324 ( Pt 2): 635-43.

Ricart, J., J. M. Izquierdo, et al. (2002). "Assembly of the ribonucleoprotein complex containing the mRNA of the beta-subunit of the mitochondrial H+-ATP synthase requires the participation of two distal cis-acting elements and a complex set of cellular trans-acting proteins." Biochem J 365(Pt 2): 417-28.

Ricquier, D. and F. Bouillaud (2000). "The uncoupling protein homologues: UCP1, UCP2, UCP3, StUCP and AtUCP." Biochem J 345 Pt 2: 161-79.

Rolfe, D. F. S. and M. D. Brand (1996). "Proton leak and control of oxidative phosphorylation in perfused, resting rat skeletal muscle." Biochimica et Biophysica Acta (BBA) - Bioenergetics 1276(1): 45-50.

55

Rondelez, Y., G. Tresset, et al. (2005). "Highly coupled ATP synthesis by F1-ATPase single molecules." Nature 433(7027): 773-7.

Rostovtseva, T. K., W. Tan, et al. (2005). "On the Role of VDAC in Apoptosis: Fact and Fiction." J Bioenerg Biomembr 37(3): 129-42.

Rubinstein, J. L., J. E. Walker, et al. (2003). "Structure of the mitochondrial ATP synthase by electron cryomicroscopy." Embo J 22(23): 6182-92.

Safe, S. and K. Kim (2004). "Nuclear receptor-mediated transactivation through interaction with Sp proteins." Prog Nucleic Acid Res Mol Biol 77: 1-36.

Scarpulla, R. C. (2002). "Transcriptional activators and coactivators in the nuclear control of mitochondrial function in mammalian cells." Gene 286(1): 81-9.

Scarpulla, R. C. (2006). "Nuclear control of respiratory gene expression in mammalian cells." J Cell Biochem 97(4): 673-83.

Schatz, G. (1968). "Impaired binding of mitochondrial adenosine triphosphatase in the cytoplasmic "petite" mutant of Saccharomyces cerevisiae." J Biol Chem 243(9): 2192-9.

Schauenstein, E. (1967). "Autoxidation of polyunsaturated esters in water: chemical structure and biological activity of the products." J Lipid Res 8(5): 417-28.

Scheller, K. and C. E. Sekeris (2003). "The effects of steroid hormones on the transcription of genes encoding enzymes of oxidative phosphorylation." Exp Physiol 88(1): 129-40.

Schrauwen, P. and M. K. Hesselink (2004). "The role of uncoupling protein 3 in fatty acid metabolism: protection against lipotoxicity?" Proc Nutr Soc 63(2): 287-92.

Schrauwen, P., W. H. Saris, et al. (2001). "An alternative function for human uncoupling protein 3: protection of mitochondria against accumulation of nonesterified fatty acids inside the mitochondrial matrix." Faseb J 15(13): 2497-502.

Schreiber, S. N., R. Emter, et al. (2004). "The estrogen-related receptor alpha (ERRalpha) functions in PPARgamma coactivator 1alpha (PGC-1alpha)-induced mitochondrial biogenesis." Proc Natl Acad Sci U S A 101(17): 6472-7.

Shabalina, I. G., A. Jacobsson, et al. (2004). "Native UCP1 displays simple competitive kinetics between the regulators purine nucleotides and fatty acids." J Biol Chem 279(37): 38236-48.

Sharpe, J. C., D. Arnoult, et al. (2004). "Control of mitochondrial permeability by Bcl-2 family members." Biochim Biophys Acta 1644(2-3): 107-13.

Shimizu, S., Y. Shinohara, et al. (2000). "Bax and Bcl-xL independently regulate apoptotic changes of yeast mitochondria that require VDAC but not adenine nucleotide translocator." Oncogene 19(38): 4309-18.

Silva, J. P., I. G. Shabalina, et al. (2005). "SOD2 overexpression: enhanced mitochondrial tolerance but absence of effect on UCP activity." Embo J 24(23): 4061-70.

Simonyan, R. A. and V. P. Skulachev (1998). "Thermoregulatory uncoupling in heart muscle mitochondria: involvement of the ATP/ADP antiporter and uncoupling protein." FEBS Lett 436(1): 81-4.

Skulachev, V. P. (1999). "Anion carriers in fatty acid-mediated physiological uncoupling." J Bioenerg Biomembr 31(5): 431-45.

Sodum, R. S. and F. L. Chung (1988). "1,N2-ethenodeoxyguanosine as a potential marker for DNA adduct formation by trans-4-hydroxy-2-nonenal." Cancer Res 48(2): 320-3.

Sodum, R. S. and F. L. Chung (1991). "Stereoselective formation of in vitro nucleic acid adducts by 2,3-epoxy-4-hydroxynonanal." Cancer Res 51(1): 137-43.

Song, B. J., Y. Soh, et al. (2001). "Apoptosis of PC12 cells by 4-hydroxy-2-nonenal is mediated through selective activation of the c-Jun N-terminal protein kinase pathway." Chem Biol Interact 130-132(1-3): 943-54.

Stepien, G., A. Torroni, et al. (1992). "Differential expression of adenine nucleotide translocator isoforms in mammalian tissues and during muscle cell differentiation." J Biol Chem 267(21): 14592-7.

56

Straffon, A. F., M. Prescott, et al. (1998). "The assembly of yeast mitochondrial ATP synthase: subunit depletion in vivo suggests ordered assembly of the stalk subunits b, OSCP and d." Biochim Biophys Acta 1371(2): 157-62.

Stuart, R. A. (2002). "Insertion of proteins into the inner membrane of mitochondria: the role of the Oxa1 complex." Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1592(1): 79-87.

Thomson, A. M., J. T. Rogers, et al. (1999). "Optimized RNA gel-shift and UV cross-linking assays for characterization of cytoplasmic RNA-protein interactions." Biotechniques 27(5): 1032-9, 1042.

Tomura, H., H. Endo, et al. (1990). "Novel regulatory enhancer in the nuclear gene of the human mitochondrial ATP synthase beta-subunit." J Biol Chem 265(12): 6525-7.

Trifunovic, A., A. Hansson, et al. (2005). "Somatic mtDNA mutations cause aging phenotypes without affecting reactive oxygen species production." Proc Natl Acad Sci U S A 102(50): 17993-8.

Tsujimoto, Y. and S. Shimizu (2000). "VDAC regulation by the Bcl-2 family of proteins." Cell Death Differ 7(12): 1174-81.

Turrens, J. F. and A. Boveris (1980). "Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria." Biochem J 191(2): 421-7.

Tvrdik, P., S. Kuzela, et al. (1992). "Low translational efficiency of the F1-ATPase beta-subunit mRNA largely accounts for the decreased ATPase content in brown adipose tissue mitochondria." FEBS Lett 313(1): 23-6.

Uchida, K., M. Shiraishi, et al. (1999). "Activation of stress signaling pathways by the end product of lipid peroxidation. 4-hydroxy-2-nonenal is a potential inducer of intracellular peroxide production." J Biol Chem 274(4): 2234-42.

Vagner, S., B. Galy, et al. (2001). "Irresistible IRES. Attracting the translation machinery to internal ribosome entry sites." EMBO Rep 2(10): 893-8.

Vander Zee, C. A., E. M. Jordan, et al. (1994). "ATPF1 binding site, a positive cis-acting regulatory element of the mammalian ATP synthase alpha-subunit gene." J Biol Chem 269(9): 6972-7.

Villena, J. A., M. C. Carmona, et al. (2002). "Mitochondrial biogenesis in brown adipose tissue is associated with differential expression of transcription regulatory factors." Cell Mol Life Sci 59(11): 1934-44.

Villena, J. A., I. Martin, et al. (1994). "ETS transcription factors regulate the expression of the gene for the human mitochondrial ATP synthase beta-subunit." J Biol Chem 269(51): 32649-54.

Villena, J. A., O. Vinas, et al. (1998). "Regulation of mitochondrial biogenesis in brown adipose tissue: nuclear respiratory factor-2/GA-binding protein is responsible for the transcriptional regulation of the gene for the mitochondrial ATP synthase beta subunit." Biochem J 331 ( Pt 1): 121-7.

Vinogradov, A. D. (1999). "Mitochondrial ATP synthase: fifteen years later." Biochemistry (Mosc) 64(11): 1219-29.

Virbasius, C. A., J. V. Virbasius, et al. (1993). "NRF-1, an activator involved in nuclear-mitochondrial interactions, utilizes a new DNA-binding domain conserved in a family of developmental regulators." Genes Dev 7(12A): 2431-45.

Virbasius, J. V. and R. C. Scarpulla (1991). "Transcriptional activation through ETS domain binding sites in the cytochrome c oxidase subunit IV gene." Mol Cell Biol 11(11): 5631-8.

Virbasius, J. V., C. A. Virbasius, et al. (1993). "Identity of GABP with NRF-2, a multisubunit activator of cytochrome oxidase expression, reveals a cellular role for an ETS domain activator of viral promoters." Genes Dev 7(3): 380-92.

57

Vyssokikh, M. Y. and D. Brdiczka (2003). "The function of complexes between the outer mitochondrial membrane pore (VDAC) and the adenine nucleotide translocase in regulation of energy metabolism and apoptosis." Acta Biochim Pol 50(2): 389-404.

Vyssokikh, M. Y., A. Katz, et al. (2001). "Adenine nucleotide translocator isoforms 1 and 2 are differently distributed in the mitochondrial inner membrane and have distinct affinities to cyclophilin D." Biochem J 358(Pt 2): 349-58.

Walker, J. E. (1992). "The mitochondrial transporter family." Curr Biol 2: 519-526. Walker, J. E., I. M. Fearnley, et al. (1985). "Primary structure and subunit stoichiometry of F1-

ATPase from bovine mitochondria." J Mol Biol 184(4): 677-701. Walker, J. E., R. Lutter, et al. (1991). "Identification of the subunits of F1F0-ATPase from

bovine heart mitochondria." Biochemistry 30(22): 5369-78. Wallberg, A. E., S. Yamamura, et al. (2003). "Coordination of p300-mediated chromatin

remodeling and TRAP/mediator function through coactivator PGC-1alpha." Mol Cell 12(5): 1137-49.

Wang, Z. G. and S. H. Ackerman (2000). "The assembly factor Atp11p binds to the beta-subunit of the mitochondrial F(1)-ATPase." J Biol Chem 275(8): 5767-72.

Wang, Z. G., D. Sheluho, et al. (2000). "The alpha-subunit of the mitochondrial F(1) ATPase interacts directly with the assembly factor Atp12p." EMBO J 19(7): 1486-93.

Wang, Z. G., P. S. White, et al. (2001). "Atp11p and Atp12p are assembly factors for the F(1)-ATPase in human mitochondria." J Biol Chem 276(33): 30773-8.

Weber, J. and A. E. Senior (2003). "ATP synthesis driven by proton transport in F1F0-ATP synthase." FEBS Lett 545(1): 61-70.

Wiedemann, N., A. E. Frazier, et al. (2004). "The protein import machinery of mitochondria." J Biol Chem 279(15): 14473-6.

Wu, Z., P. Puigserver, et al. (1999). "Mechanisms controlling mitochondrial biogenesis and respiration through the thermogenic coactivator PGC-1." Cell 98(1): 115-24.

Yamamoto, H., M. Esaki, et al. (2002). "Tim50 is a subunit of the TIM23 complex that links protein translocation across the outer and inner mitochondrial membranes." Cell 111(4): 519-28.

Yan, W. L., T. J. Lerner, et al. (1994). "Sequence analysis and mapping of a novel human mitochondrial ATP synthase subunit 9 cDNA (ATP5G3)." Genomics 24(2): 375-7.

Yasuda, R., H. Noji, et al. (1998). "F1-ATPase is a highly efficient molecular motor that rotates with discrete 120 degree steps." Cell 93(7): 1117-24.

Yoshida, M., E. Muneyuki, et al. (2001). "ATP synthase--a marvellous rotary engine of the cell." Nat Rev Mol Cell Biol 2(9): 669-77.

Zaid, A., Z. Hodny, et al. (2001). "Sp1 acts as a repressor of the human adenine nucleotide translocase-2 (ANT2) promoter." Eur J Biochem 268(21): 5497-503.

Zaid, A., R. Li, et al. (1999). "On the role of the general transcription factor Sp1 in the activation and repression of diverse mammalian oxidative phosphorylation genes." J Bioenerg Biomembr 31(2): 129-35.

Zamzami, N., S. A. Susin, et al. (1996). "Mitochondrial control of nuclear apoptosis." J Exp Med 183(4): 1533-44.

Zhang, H., D. A. Dickinson, et al. (2005). "4-Hydroxynonenal increases gamma-glutamyl transpeptidase gene expression through mitogen-activated protein kinase pathways." Free Radic Biol Med 38(4): 463-71.

58

59