Mitochondrial translation and its impact onprotein homeostasis and agingTamara Suhm

Academic dissertation for the Degree of Doctor of Philosophy in Biochemistry at StockholmUniversity to be publicly defended on Friday 15 February 2019 at 09.00 in Magnélisalen,Kemiska övningslaboratoriet, Svante Arrhenius väg 16 B.

AbstractBesides their famous role as powerhouse of the cell, mitochondria are also involved in many signaling processes andmetabolism. Therefore, it is unsurprising that mitochondria are no isolated organelles but are in constant crosstalk withother parts of the cell. Due to the endosymbiotic origin of mitochondria, they still contain their own genome and geneexpression machinery. The mitochondrial genome of yeast encodes eight proteins whereof seven are core subunits of therespiratory chain and ATP synthase. These subunits need to be assembled with subunits imported from the cytosol to ensureenergy supply of the cell. Hence, coordination, timing and accuracy of mitochondrial gene expression is crucial for cellularenergy production and homeostasis. Despite the central role of mitochondrial translation surprisingly little is known aboutthe molecular mechanisms.

In this work, I used baker’s yeast Saccharomyces cerevisiae to study different aspects of mitochondrial translation.Exploiting the unique possibility to make directed modifications in the mitochondrial genome of yeast, I established amitochondrial encoded GFP reporter. This reporter allows monitoring of mitochondrial translation with different detectionmethods and enables more detailed studies focusing on timing and regulation of mitochondrial translation. Furthermore,employing insights gained from bacterial translation, we showed that mitochondrial translation efficiency directly impactson protein homeostasis of the cytoplasm and lifespan by affecting stress handling. Lastly, we provided first evidencethat mitochondrial protein quality control happens at a very early stage directly after or during protein synthesis at theribosome. Surveillance of protein synthesis and assembly into complexes is important to avoid accumulation of misfoldedor unassembled respiratory chain subunits which would disturb mitochondrial function.

Keywords: mitochondrial ribosome, mitochondrial translation accuracy, mitochondrial communication, interorganellarcommunication, stress signaling, proteostasis, aging, yeast genetics, mitochondrial protein quality control,mitochondrial membrane protein insertion.

Stockholm 2019http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-163149

ISBN 978-91-7797-542-7ISBN 978-91-7797-543-4

Department of Biochemistry and Biophysics

Stockholm University, 106 91 Stockholm

MITOCHONDRIAL TRANSLATION AND ITS IMPACT ON PROTEINHOMEOSTASIS AND AGING 

Tamara Suhm

Mitochondrial translation andits impact on proteinhomeostasis and aging 

Tamara Suhm

©Tamara Suhm, Stockholm University 2019 ISBN print 978-91-7797-542-7ISBN PDF 978-91-7797-543-4 Cover image by Natalie Schwenteck  Printed in Sweden by Universitetsservice US-AB, Stockholm 2019

Success consists of goingfrom failure to failurewithout loss ofenthusiasm.  - Winston Churchill -

9

List of publications

I. Suhm T, Habernig L, Rzepka M, Kaimal JM, Andréasson C,

Büttner S, Ott M. (2018) A novel system to monitor mitochon-

drial translation in yeast. Microb Cell. 5(3):158-164

II. Suhm T and Ott M. Different genetic approaches to mutate the

mitochondrial ribosomal protein S12. Manuscript

III. Suhm T, Kaimal JM, Dawitz H, Peselj C, Masser AE, Hanzén

S, Ambrožič M, Smialowska A, Björck ML, Brzezinski P,

Nyström T, Büttner S, Andréasson C, Ott M. (2018)

Mitochondrial translation efficiency controls cytoplasmic

protein homeostasis. Cell Metab. 27 (6):1309-1322

IV. Vargas Möller-Hergt B, Carlström A, Suhm T, Ott M. (2018)

Insertion defects of mitochondrially encoded proteins burden the

mitochondrial quality control system. Cells 7 (10), 172

Additional publications

I. Suhm T and Ott M. (2016) Mitochondrial translation and cellu-

lar stress response. Cell Tissue Res. 367(1):21-31

10

11

Abbreviations

AAA ATPase associated with diverse cellular activities

CLS chronological lifespan

EF elongation factor

eIF2α eukaryotic initiation factor 2α

HSE heat shock element

HSR heat shock response

IF initiation factor

IMM inner mitochondrial membrane

IPOD insoluble protein deposit

IMS intermembrane space

INQ intracellular quality control compartment

IMiQ intramitochondrial quality control compartment

IPTP interorganellar proteostasis transcription program

MAGIC mitochondria as guardian of cytosol

mitoCPR mitochondrial comprised protein import response

mtDNA mitochondrial DNA

MFRAT mitochondrial free radical theory of aging

mPOS mitochondrial precursor overaccumulation stress

MTS mitochondrial targeting signal

NLS nuclear localization signal

OMP orotidine-5'-phosphate

OMM outer mitochondrial membrane

OXPHOS oxidative phosphorylation

PN proteostasis network

Prx peroxiredoxin

ROS reactive oxygen species

RF release factor

RLS replicative lifespan

RTG retrograde response

RRF ribosome recycling factor

STRE stress element

UPRmt mitochondrial unfolded protein response

12

UPS ubiquitin proteasome system

UTR untranslated region

UPRam UPR activated by mistargeted proteins

13

Contents

Introduction ................................................................................................... 15 Yeast as a model organism ......................................................................................... 15

Basic yeast genetics .............................................................................................. 15 Yeast in mitochondria research ............................................................................. 16 Yeast as model for higher eukaryotes .................................................................... 17

Mitochondria ............................................................................................................... 19 General functions................................................................................................... 19 Mitochondria and aging – what is the connection?................................................. 21

Translation .................................................................................................................. 23 Bacterial translation ............................................................................................... 23 Mitochondrial translation ........................................................................................ 26

Protein Quality Control ................................................................................................ 30 Protein quality control in the cytosol ....................................................................... 30 Protein quality control inside mitochondria ............................................................. 35

Interorganellar communication .................................................................................... 39 Mitochondria - nucleus communication .................................................................. 39 Mitochondrial stress and cytoplasmic proteostasis ................................................ 43

Aims .............................................................................................................. 47

Summaries of papers .................................................................................... 49

Future perspectives ....................................................................................... 53

Sammanfattning på svenska ......................................................................... 57

Acknowledgments ......................................................................................... 59

References .................................................................................................... 61

14

15

Introduction

Yeast as a model organism

In my work I used baker’s yeast Saccharomyces cerevisiae as a model or-

ganism. Yeast has many advantages for being used as a eukaryotic model

organism including the number of tools accessible to manipulate its genome.

Basic yeast genetics

The complete genome of yeast was first sequenced and published in 1996 [1]

and today it is the best annotated genome. It is relatively easy to create

knock-outs or knock-ins by making use of the yeast’s own homologous re-

combination system [2]. There is even a commercially available deletion

library in which each gene has been replaced by a kanamycin selection cas-

sette. Using the kanamycin cassette to make knock-outs has the advantage

that all auxotrophic markers are still available, e.g. to select for plasmids.

However, knock-outs can also be made in the same way using auxotrophic

selection cassettes. Laboratory yeast strains are modified in a way that they

cannot synthesize all amino acids they need for growth and are therefore

auxotroph for certain amino acids. Reintroducing the gene missing for bio-

synthesis of this amino acid concomitant with another protein modification

of interest allows screening for positive transformation events by growth in

the absence of this amino acid. Additional to a number of auxotrophic mark-

ers for positive selection there is the auxotrophic marker URA3 which is a

counter-selectable cassette, meaning that it can be selected for but also

against the cassette [3]. In more detail, URA3 encodes orotidine-5'-phosphate

(OMP) decarboxylase which catalyzes a step in the de novo synthesis of

uracil [4]. Therefore, cells expressing OMP decarboxylase can grow without

uracil in the medium. However, in the presence of 5-fluoroorotic acid, yeast

cells harboring the URA3 cassette are not viable, as OMP decarboxylase will

metabolize it to the toxic compound 5-fluorouracil [3]. 5-fluorouracil acts as

thymidylate synthase inhibitor, thereby preventing thymidylate synthesis and

eventually DNA replication.

A number of yeast plasmids are available containing different selection

markers, promotors with various strength, as well as different multiple clon-

16

ing sites to make the cloning as easy and straight forward as possible. Addi-

tionally, there is the possibility to use integrative plasmids which lack an

origin of replication and are therefore only propagated if integrated into the

genome.

Yeast in mitochondria research

Being a facultative anaerobe organism makes yeast a very attractive model

organism when studying mitochondrial processes. Facultative anaerobe de-

scribes the ability to survive without a functioning respiratory chain, as long

as kept on the fermentative carbon source glucose. This ability of yeast can

be used to study deletions of proteins of the respiratory chain, disease related

mutations of respiratory chain proteins or mitochondrial gene expression

proteins which often is not possible in mammalian cells [5]. Even the loss of

mitochondrial DNA (mtDNA) (rho0) is not lethal to yeast as long as kept on

a fermentative carbon source.

A rho0 strain can be repopulated with mtDNA using cytoduction. This meth-

od allows the transfer of the mitochondrial genome between two strains [6,

7], while a mutation in one of the genes involved in karyogamy (kar1-1)

slows down karyogamy [8] and prevents recombination of nuclear DNA.

Ultimately, we can make directed alterations in the mitochondrial genome

using biolistic transformation [9]. Besides plants, yeast is the only organism,

so far, there this is possible. In order to translate a mRNA successfully on

mitochondrial ribosomes it needs to be under control of 5’ and 3’ untranslat-

ed regions (UTR) [10-12]. The mitochondrial plasmid used in my studies

encodes a part of endogenous COX2 as well as superfold GFP under control

of 5’ and 3’ UTR of COX2 (Figure 1). A strain with an unfunctional COX2

(cox2-62) [13] is used as final recipient. By homologous recombination the

plasmid is integrated at the COX2 locus giving rise to a functional COX2.

Therefore, respiration can be used to screen for positive mitochondrial trans-

formants. Thanks to the high mitochondrial recombination rate, yeast be-

comes homoplasmic (all mitochondrial DNA copies are identical) within a

few generations [14, 15]. So far it is not possible to make directed alterations

of the mtDNA in mammalian cells which is not due to the lack of trial [16,

17]. It might, however, become possible in the future allowing for more de-

tailed studies of mitochondrial translation in higher eukaryotes.

17

Figure 1: Biolistic transformation. A plasmid encoding superfold GFP (sfGFPm)

flanked by 5’ and 3’UTR of COX2 as well as COX2 is transformed into cells lacking

a functional COX2 (cox2-62). Eventually, mitochondrial homologous recombination

gives rise to a new, fully functional, mitochondrial genome encoding sfGFPm as a

ninth protein (UTR, untranslated region; COX2, cytochrome c oxidase subunit 2;

mtDNA, mitochondrial DNA) (figure copied from [18]).

Yeast as model for higher eukaryotes

Already in 1988 Botstein and Fink published an article in the journal Science

about the promising future of yeast as an eukaryotic model organism and in

2011 they renewed this statement based on advances made in yeast research

and the plethora of databases available [19, 20]. Many signaling pathways

are conserved from the single cellular eukaryote yeast over Caenorhabditis

elegans (C. elegans), Drosophila melanogaster and mice to humans. The

first but not only example of functional conservation between yeast and

mammals were the proto-oncogene proteins Ras1 and 2 [21]. Today it is

clear that many pathways are conserved between humans and yeast and that

without the great work in yeast many of these pathways would still be poorly

characterized.

These pathways also include the ones which are implicated in aging. Studies

performed in yeast gave insights into the aging process which was instru-

mental to uncover aging modulators in higher eukaryotes up to mice [22].

There are mainly two methods in yeast to study aging: chronological and

replicative lifespan (CLS and RLS, respectively) (Figure 2) [22, 23]. In

chronological aging, survival of non-dividing cells is monitored. In detail,

cells are grown to stationary phase where they enter a quiescent-like state.

By plating on rich media, the ability of the cells to escape this quiescent-like

state is studied. In replicative aging, the amount of cell divisions of a single

mother cell is observed. It was argued that RLS is a model which replicates

mitotically active cells as stem cells in humans, while CLS more relates to

18

post-mitotic cells. However, it was shown that genes involved in yeast RLS

show a significant overlap with worm aging (post-mitotic organism). Fur-

thermore, both methods played an important role in discovering the role of

the conserved Ras-PKA and TOR-Sch9 signaling pathways in aging [23].

Figure 2: Replicative and chronological lifespan. In replicative lifespan the num-

ber of cell divisions of a single mother cell is followed (A). In chronological lifespan

a liquid culture is grown to stationary phase and the ability to escape a quiescent-like

state is studied (B) (d, daughter cell; m, mother cell).

19

Mitochondria

General functions

Mitochondria are surrounded by a double membrane subdividing the orga-

nelle into four compartments: the outer mitochondrial membrane (OMM),

the intermembrane space (IMS), the inner mitochondrial membrane (IMM)

and the matrix. Mitochondria are most famous for being the powerhouse of

the cell, producing most of the cell’s ATP.

Figure 3: Mitochondrial functions. Besides their well-known role as powerhouse

of the cell, mitochondria are involved in other processes including metabolic reac-

tions as TCA cycle, fatty acid oxidation, amino acid metabolism, cell signaling by

maintaining calcium homeostasis, ATP/ADP ratio and apoptosis as well as iron

sulfur cluster biogenesis and heme biosynthesis (TCA, tricarboxylic acid; ANT,

adenine nucleotide translocase; OXPHOS, oxidative phosphorylation; ROS, reactive

oxygen species).

The oxidative phosphorylation (OXPHOS) system, which is located in the

IMM, consists of three respiratory complexes in yeast (complex II, III and

IV) as well as the ATP synthase (complex V). Electrons are transferred from

NADH and FADH2, which were reduced in the TCA cycle, through the res-

piratory complexes to eventually reduce oxygen to water. Complex III and

IV use the energy released by the transport of electrons from a donor to a

more electronegative acceptor to translocate protons from the matrix site to

the IMS. The resulting electrochemical gradient is used by the ATP synthase

20

to phosphorylate ADP to ATP. Besides this crucial role, mitochondria also

fulfill a number of other important tasks like heme biosynthesis, iron-sulfur

cluster biogenesis, calcium storage and signaling, harboring many metabolic

pathways at least partly and playing an important role in apoptosis (Figure 3)

[24, 25].

About 1.5 billion years ago, mitochondria originated by endosymbiosis of an

α-proteobacterium with an ancestral eukaryotic cell [26]. During evolution,

mitochondria transferred most of its DNA to the nucleus [27]. Nevertheless,

the organelle still contains its own small genome and gene expression ma-

chinery (Figure 4). In yeast, mtDNA encodes for two rRNAs, 24 tRNAs and

eight proteins, whereof seven are membrane proteins and subunits of the

OXPHOS system [28]. However, most of the about 1000 mitochondrial pro-

teins are encoded in the nucleus, transcribed there, translated on cytosolic

ribosomes and imported into mitochondria (Figure 4) [29, 30].

Figure 4: Dual genetic origin of OXPHOS system. Mitochondrial DNA encodes

seven subunits of the OXPHOS, while the remaining subunits are encoded in the

nuclear genome, translated in the cytosol and imported into mitochondria

(OXPHOS, oxidative phosphorylation; mtDNA, mitochondrial DNA; TOM, trans-

locase of outer membrane; TIM, translocase of inner membrane).

The high hydrophobicity of mitochondrial encoded proteins could be one

reason why mitochondria kept these genes. It was shown that a nuclear en-

coded version of cytochrome b could not be imported into mitochondria

[31], while Atp6 expressed from the nuclear genome could be imported

though not very efficiently [32]. Another plausible reason is the different

codon usage of mitochondria. Lastly, it might be an advantage to directly

21

couple assembly to synthesis in the organelle. Even though there are several

nonexclusive explanations why mitochondria kept this small number of

genes, it is still questioned if this is worth the costs as there are roughly 250

proteins involved in maintaining and expressing the mitochondrial genome

[33].

As the OXPHOS system, as well as the mitochondrial ribosome, consist of

subunits of dual genetic origin, the two gene expression systems need to be

coordinated to allow for proper assembly of these macromolecular complex-

es ensuring ATP production. This coordination will be discussed in the ‘Mi-

tochondrial translation’ chapter.

Mitochondria and aging – what is the connection?

Mitochondrial dysfunction is a hallmark of aging [34]. In many model or-

ganisms as well as in humans an increase in mtDNA mutations was observed

during aging [35-37] as well as an overall decline in mitochondrial function

[38] and respiratory chain function [36, 37, 39].

Already in 1956, Harman postulated the free radical theory of aging. This

theory describes how an increased production of reactive oxygen species

(ROS) over time attacks DNA, proteins and lipids in the cell [40]. As the

respiratory chain, located in the IMM, is the major site of ROS production, it

was later extended to the mitochondrial free radical theory of aging

(MFRAT) (Figure 5) [41]. Being the main site of ROS production, mito-

chondria, especially mtDNA, were thought to be the primary site of ROS

damage. Damage to mtDNA gives rise to mutated versions of respiratory

chain proteins which in turn result in an even higher ROS leakage thereby

entering a vicious cycle of ROS production and mtDNA mutations.

In favor of the MFRAT, many studies show a clear involvement of ROS in

aging phenotypes. In contrast, others show that an increase in ROS can even

prolong lifespan [37, 42, 43] and that an increased oxidative defense system

alone is not sufficient to prolong lifespan [37, 44-49]. The involvement of

ROS in aging is still a highly debated topic and most likely the dosage

makes the poison.

One of the main studies challenging the MFRAT is the mutator mouse,

which accumulates mtDNA mutations due to a mutation in the proofreading

function of the mitochondrial polymerase Polg. However, this increase in

mtDNA mutations which was accompanied by an early onset of aging phe-

notypes, did not correlate with an increase in oxidative stress even though

there was mitochondrial dysfunction reported [50, 51]. Additionally, the

increase in mtDNA mutations observed in the mutator mouse showed a ra-

ther linear increase in ROS during adulthood and not an exponential increase

as it would be expected in the postulated vicious cycle of the MFRAT [51].

22

Even though the mutator mouse nicely shows the first causative link between

mtDNA mutations and early onset of aging some limitations have to be kept

in mind. In normal aging tissue, the number of mtDNA mutations is much

lower than what was observed in the mutator mouse and it is not clear how

they account for aging [52]. Even in the heterozygous mutator mouse, the

number of mutations acquired is higher than in humans, yet this mouse did

not show an aging phenotype [53]. Furthermore, the mutations in the mutator

mouse are point mutations, while in elderly humans mtDNA deletions domi-

nate [54, 55]. Such deletions were replicated by the deletor mouse which

contains patient-mutations in the mtDNA helicase TWINKEL [56]. The

deletor mouse, however, did not show an early onset of aging despite

mtDNA deletions and respiratory chain dysfunction.

In contrast to MFRAT, the error-catastrophe theory of aging postulated that

a decrease in the accuracy of translation drives aging [57]. Experiments test-

ing this hypothesis were not able to show a relationship between translation

accuracy and aging [58]. However, these experiments focused on cytosolic

ribosomes while a study focusing on mitochondrial ribosomes in yeast

showed that increasing mitochondrial translation accuracy with the drug

erythromycin increased RLS [59]. Also in bacteria, the ancestor of mito-

chondria, a decrease in translation fidelity during cell cycle arrest was re-

ported [60, 61].

Making a long story short, the involvement of ROS in aging is still debated

and the MFRAT has been largely refuted. Nowadays it is rather widely as-

sumed that mtDNA mutations caused by replication errors correlate with the

aging phenotype in humans [62]. The involvement of mitochondrial dys-

function in aging, however, is clear but the complex nature of mitochondria

and its involvement in so many crucial functions makes it difficult to disen-

tangle cause and consequence [55].

Figure 5: Mitochondrial free radical theory of aging. ROS leakage from

OXPHOS damages mtDNA over time. mtDNA mutations give rise to defective

OXPHOS proteins causing more ROS leakage and entering a vicious cycle which

drives aging (mtDNA, mitochondrial DNA; OXPHOS, oxidative phosphorylation;

ROS, reactive oxygen species).

23

Translation

Bacterial translation

An overview

In bacteria, translation was and is extensively studied and a detailed picture

of the single steps was built over the last decades. Translation can be divided

into four steps: initiation, elongation, termination and recycling. These steps

will be briefly described in the following paragraph. For a more detailed

review see Rodnina 2018 [63].

In a non-translating state, initiation factor 3 (IF3) binds to the small subunit

of the ribosome to prevent its premature association with the large subunit.

During initiation the small ribosomal subunit together with IF1, 2 and the

fMet-tRNAfMet binds the mRNA. The tRNA binding displaces IF3 which in

turn allows the large subunit of the ribosome to join. Subsequent GTP hy-

drolysis by IF2 releases the initiation factors. The ribosome is now ready for

translation.

The ternary complex, consisting of aminoacyl-tRNA together with elonga-

tion factor Tu and GTP (EF-Tu-GTP), first makes codon independent inter-

actions with the ribosome, followed by codon recognition to start the first

round of elongation. In most cases, cognate and near-cognate tRNAs only

differ in one base. Therefore, the difference in free energy released by codon

recognition between cognate and near-cognate tRNA is not enough to reach

the high accuracy of 10-3 to 10-4 [64, 65] which is observed in living cells.

Instead, two independent kinetic proofreading steps, which are separated by

GTP hydrolysis, are required [66-68]. In a first step, initial selection can lead

to rejection of non-cognate tRNAs in complex with EF-Tu from the ribo-

some. Non-cognate tRNAs are rejected with higher probability than cognate

ones. After codon-anticodon interaction and GTP hydrolysis by EF-Tu, the

tRNA can either be rejected in a second proofreading step or accommodated

in the A site (Figure 6). The recognition of the cognate tRNA induces a con-

formational change in the small subunit referred to as domain closure (see

section ‘Domain closure model during decoding’). Once the cognate tRNA

is accommodated in the A site, peptide bond formation is catalyzed by the

peptidyl transferase center of the large ribosomal subunit leaving a dipep-

tidyl tRNA in the A site and a deacetylated tRNA in the P site. Thereafter,

the mRNA and tRNA translocate through the ribosome with help of elonga-

tion factor G (EF-G), another GTPase. This places the dipeptidyl tRNA in

the P site, while leaving the A site with the new codon.

Once the ribosome encounters a stop codon translation is terminated. There

are three different stop codons in bacteria: UAA, UGA and UAG. Depend-

24

ing on the stop codon release factor 1 or 2 (RF1 or 2) bind. They hydrolyze

the esterbond, by which the peptide is bound to the tRNA, to release the

peptide. RF3 then catalyzes the dissociation of RF leaving the ribosome with

bound tRNA and mRNA.

In order to initiate a new round of translation, ribosome recycling is neces-

sary. The ribosome is split by ribosome recycling factor (RRF) and EF-G.

IF3 binding releases the tRNA while the mRNA dissociates spontaneously

[63, 69].

Figure 6: Two kinetic proofreading steps to select cognate tRNA. During initial

selection the aminoacyl-tRNA interacts codon independent with the ribosome. Sub-

sequent codon-anticodon interaction activates GTP hydrolysis and induces domain

closure. In a second proofreading step the aminoacyl tRNA is rejected or accommo-

dated in the A-site followed by peptide bond formation (figure copied from [70]).

Domain closure model during decoding

The decoding center of the ribosome is highly conserved and is located in

the small subunit of the ribosome [71, 72]. It consists of the bases G530 (in

helix18) as well as A1492 and A1493 (in helix 44) of the 16S rRNA in bac-

teria [73]. The domain closure model, established by the Ramakrishnan

group [74, 75], describes how the ribosome discriminates cognate tRNAs

from near-cognate tRNAs. This model suggests that only if the cognate

tRNA is placed in the A site, a conformational change of the ribosome is

triggered.

In detail, A1492 and A1493 flip outwards towards the codon anticodon helix

establishing interactions with the first and second position of the codon. Ad-

ditionally, G530 switches from syn to anti conformation and contacts the

second and third position of the codon. These small conformational changes

trigger a larger change in the small subunit by rotation of the head towards

the subunit interface and the shoulder towards the intersubunit space.

During this conformational change hydrogen bonds between lysine residues

in S12 and the phosphate backbone in helix 27 and 44 of the 16S rRNA are

formed establishing new contacts between the 16S rRNA and the ribosomal

25

protein S12. Furthermore, contacts between the ribosomal proteins S4 and

S5 are broken.

Summed up, this model suggests that the ribosome is in an open form and

undergoes rearrangements to a closed form only when the cognate tRNA has

bound. This model was based on structures of the small subunit with bound

A site tRNA anticodon stem loop (ASL) and oligonucleotide for mRNA.

Today, this model for discrimination between cognate and near-cognate

tRNA is debated in the ribosome field. There are structures of the 70S ribo-

some with long mRNA and natural tRNA available. These structures showed

identical rearrangements of the small subunit irrespective whether the bound

tRNA was cognate or near-cognate [76-78].

Changed translation accuracy by aminoglycosides and mutations

Aminoglycosides are a class of antibiotics which decrease translation accu-

racy by binding to the small subunit of the ribosome. Streptomycin for ex-

ample binds to the small subunit at the decoding center [79]. It was proposed

to stabilize the closed form of the small subunit resulting in an error-prone

translation. Paromomycin, another aminoglycoside, binds to helix 44 of the

16S rRNA. It was shown to induce domain closure which normally requires

the presence of cognate tRNA and proper codon-anticodon interaction [74,

80]. It leads to non-specific binding of tRNA to the A site, increases GTPase

activation rate of near-cognate codon-anticodon interactions, thereby de-

creasing dissociation rate and increasing peptide bond formation [80].

Mutations affecting translation fidelity were first found by hypersensitivity

of strains towards the aminoglycoside streptomycin. These strains contained

mutations in the ribosomal proteins S4 and S5 [81]. It was suggested that

these mutations facilitate the breakage of polar interactions between the two

proteins which in turn induces domain closure and causes error-prone trans-

lation [74, 79, 82]. This model was, however, questioned and criticized as

oversimplified due to the observation that not all mutations affecting fidelity

were necessarily located at the interface of S4 and S5 or destabilize their

interaction [83, 84].

The ribosomal protein S12 has a unique position in the bacterial ribosome. It

contacts helices 18, 27 and 44 [85] of the 16S rRNA which makes it the only

protein in such close vicinity to the decoding center. S12 helps to orient sev-

eral 16S rRNA residues for codon-anticodon base pairing. Mutations in the

ribosomal protein S12 showed resistance or hypersensitivity towards strep-

tomycin [86-89]. Also here it is assumed that the phenotypes are caused by

the interference of the mutations with domain closure [74]. Mutations which

cause a streptomycin resistant phenotype are often in lysine residues which

are involved in salt bridge formation during domain closure. It is likely that

26

by mutating these residues and thereby destabilizing the salt bridges, domain

closure is hindered resulting in a hyperaccurate translation.

Mitochondrial translation

An overview

Because of the endosymbiosis theory for mitochondrial origin it was as-

sumed that the translation machinery is highly conserved between bacteria

and mitochondria. As known today, yeast mitochondrial ribosomes have a

higher content of proteins than rRNA compared to their bacterial counter-

parts and this difference is even more pronounced in mammalian mitochon-

drial ribosomes [90]. Moreover, recent advances in cryo-electron microsco-

py showed that the structure of the ribosome did change remarkably during

evolution [71, 91]. Besides the increased protein content, mitochondrial ri-

bosomes also do not contain a 5S rRNA [92]. However, certain parts of the

ribosome are highly conserved, like the decoding center (in the small ribo-

somal subunit) and the peptidyl transferase center (in the large ribosomal

subunit) [71, 72, 90, 92]. Also the susceptibility towards aminoglycosides

has not changed, which explains the side effects caused by certain antibiotics

when used as drugs [93].

Another key component of translation is the mRNA, which also differs be-

tween bacteria and mitochondria, as mitochondrial mRNAs do not contain a

Shine-Dalgarno sequence. In bacteria, this sequence in the 5’ untranslated

region (UTR) assures proper initiation by pairing with bases in the 16S

rRNA and positioning of the start codon in the ribosomal P site. Although in

yeast mitochondria, the mRNA contains a 5’UTR, it does not contain a

Shine-Dalgarno sequence, while mammalian mitochondrial mRNAs do not

contain a 5’UTR at all [92].

In yeast, instead, a set of proteins called translational activators has been

described to interact with the 5’UTR and the ribosome to initiate and coordi-

nate translation in mitochondria [94, 95]. So far only one translational acti-

vator has been identified in mammalian mitochondria (TACO1) [96].

Nevertheless, the general steps of translation (initiation, elongation, termina-

tion and recycling) are conserved. There are homologs for IF2 and IF3. IF1

is missing, but its role may be played by an extra domain in IF2. Also the

delivery of the fMet-tRNAfMet (in mammalian mitochondria the Met is not

formylated) and joining of the large subunit are similar as in bacteria [92].

The most conserved step is elongation where homologs of EF-Tu, EF-G and

EF-Ts (not in yeast) are present. Due to a reduced number of codons encod-

ing a stop in mitochondria (only UAA and UAG, UGA encodes tryptophan),

there is only a homolog for the release factor RF1. A homolog of the recy-

27

cling factor RRF eventually splits the ribosome into subunits. In contrast to

bacteria, there are two EF-G homologs, one involved in translocation during

elongation while the other one assists RRF1 during recycling [92].

Accuracy of translation

As described above, bacterial translation is highly accurate with an error

frequency of 10-3-10-4. To date, not much is known about accuracy of mito-

chondrial translation, as there is neither a robust in vitro translation system

nor in vivo reporters. One fact that is however known, is that mitochondrial

translation accuracy is important for life. First of all, proteins encoded by the

mitochondrial genome are core subunits of the respiratory chain. An inaccu-

rate mitochondrial translation, therefore, gives rise to malfunctioning or un-

assembled respiratory chain complexes and eventually low levels of ATP as

well as increased rates of ROS production [97].

Secondly, severe mitochondrial diseases are caused by mutations in mito-

chondrial tRNAs. The disease MELAS, for example, is caused by a mutation

in Leu-tRNA which causes a decrease in translation accuracy [98]. Further-

more, the A1555G mutation in 12S rRNA of human mitochondrial ribosome

causes aminoglycoside hypersensitivity and stress-induced deafness. In a

study with mitochondria-bacteria-hybrid ribosomes, this mutation was also

shown to decrease translation accuracy [99]. Another study performed in

yeast, showed an increased lifespan when translation accuracy was increased

by the antibiotic erythromycin [59].

As described in the chapter ‘Bacterial translation’, there are mainly three

ribosomal proteins involved in the decoding step in bacteria, namely S4, S5

and S12. These proteins have homologs in mitochondria (Nam9, Mrps5 and

Mrps12 respectively) (Table 1). Also the recent high resolution ribosome

structures showed conservation of the decoding center [71, 92]. Therefore,

the assumption that mutations in Nam9, Mrps5 or Mrsp12 would affect

translation in a similar way as their bacterial counterparts is eligible. Indeed,

a recent study showed that a mutation in Mrps5 decreases mitochondrial

translation accuracy and alters stress- and age-related behavior in mice

[100]. An older study discovered a mutation in Nam9 which suppresses an

ochre mutation in COX2. The authors speculate that the premature stop co-

don in the mutated version of COX2 is read through by decreased translation

accuracy in mitochondria due to the Nam9 mutation [101].

Table 1: Orthologues of ribosomal proteins in bacteria, yeast and mammals.

Bacteria Yeast Mammals

S4 Nam9 Mrps4

S5 Mrps5 Mrps5

S12 Mrps12 Mrps12

28

Monitoring mitochondrial translation

Mitochondrial translation can be followed by incorporation of radioactive

methionine (35S) into newly translated proteins, followed by separation of

the translation products on SDS-PAGE and subsequent visualization by au-

toradiography [102-104]. One hurdle, however, is to circumvent labeling of

cytosolic proteins. There are two ways to address this problem. Cytosolic

translation can be inhibited by cycloheximide, an antibiotic which blocks the

translocation step of elongation. This prevents labeling of cytosolic proteins

but also comes with a number of side effects. Inhibition of protein synthesis

leads to an increase in free amino acids which activates TOR signaling [105-

107], thereby causing changes in cell growth, nutrient signaling and life span

[108-110]. Furthermore, a recent study showed that cycloheximide directly

alters mitochondrial protein synthesis within less than 15 minutes [111].

Lastly, not only protein synthesis but also protein degradation is altered by

cycloheximide [112]. Taken together, these effects severely impact on cellu-

lar homeostasis and do not reflect a physiological scenario.

Alternatively, the labeling can be performed in isolated organelles. This sys-

tem certainly enables visualization of merely mitochondrial proteins but also

lacks all the signals coming from the cytosol or nucleus which were shown

to impact on mitochondrial translation [95, 110, 111].

Taken together, both methods enable visualization of protein synthesis in

mitochondria but do not reflect a physiological scenario and, therefore, are

not appropriate to study activation, timing or regulation of translation in the

context of cell physiology. Another approach by Fox and coworkers was

integration of GFP into the mitochondrial genome. Following GFP expres-

sion allows readout of mitochondrial translation in vivo but as the GFP re-

placed the open reading frame of COX3 in the mitochondrial genome it led

to non-respiring cells which prevented physiological studies [113].

Co-translational membrane insertion of mitochondrial encoded proteins

Most proteins produced by the mitochondrial ribosome are highly hydropho-

bic membrane proteins and need to be inserted into the membrane to prevent

aggregation. This task is facilitated by membrane attachment of the mito-

chondrial ribosome [92, 94, 114]. In mammalian mitochondria, the ribosome

is anchored to the membrane via Mrpl45, a ribosomal protein of the large

subunit [92]. Mba1, the yeast homolog of Mrpl45, is a peripheral membrane

protein which was shown to interact with the ribosome near the exit tunnel in

close proximity to the nascent chain [115, 116]. Another membrane protein

interacting with the tunnel exit as well as the nascent chain is the insertase

Oxa1. Oxa1 is part of the Oxa1/YidC/Alb3 protein family and inserts mito-

chondrial proteins into the IMM even though it is not absolutely essential for

membrane protein insertion [115, 117, 118]. A double deletion of Oxa1 and

Mba1, however, showed impaired membrane insertion [115]. A third mem-

29

brane protein interacting with the large subunit is Mdm38. Its mammalian

homolog is involved in neurodegeneration. A recent study identified Mrx15

as a new ribosome receptor and showed overlapping functions of Mrx15 and

Mba1 in co-translational protein membrane insertion [119]. Together these

factors mediate mitochondrial ribosome membrane attachment and co-

translational membrane insertion of proteins [94].

Coordination of mitochondrial and nuclear translation

Assembly of a functioning OXPHOS system is crucial for aerobic life and

requires coordination of mitochondrial and nuclear gene expression. In a

recent study, Couvillion et al applied a ribosome profiling approach to study

mitochondrial as well as cytosolic translation upon a metabolic switch from

fermentation to respiration in yeast [111]. They reported that mitochondrial

as well as cytosolic translation of OXPHOS subunits is regulated in a fast

and coordinated fashion.

Furthermore, translational activators are known to play an important role in

coordination of translation and assembly of OXPHOS complexes in yeast

mitochondria [95]. It was shown that translational activators can also act as

chaperones which can be sequestered in assembly intermediates and only

upon proper assembly of the complex are released to start a new round of

translation activation. Thus, mitochondrial translation is leveled to the num-

ber of nuclear-encoded subunits available.

In mammalian mitochondria only one translational activator has been dis-

covered so far. A different concept, however, was described for translational

plasticity in mammalian mitochondria. It was shown that for complex IV

assembly, mitochondrial translation is adjusted to fit the amount of nuclear-

encoded subunits [120]. In Cox4 depleted cells, COX1 translation is halted

and the stalled ribosomes, containing the Cox1 peptide partially inserted into

the membrane, interact with assembly intermediates.

This shows that mitochondrial translation is regulated in one or another way

to match the availability of nuclear-encoded subunits in yeast as well as

mammals, albeit with distinct mechanisms.

30

Protein Quality Control

Protein quality control in the cytosol

The happiness of every cell depends on proper functioning proteins. This is

achieved by the constant surveillance of de novo protein folding as well as

protein refolding and degradation of damaged or unfolded proteins to main-

tain protein homeostasis (proteostasis). The system responsible is called

proteostasis network (PN) and consists of molecular chaperones and co-

chaperones as well as the degradation pathways autophagy and ubiquitin-

proteasome system (UPS). Also the active sequestration of aggregates into

specific compartments inside the cell is part of the PN. In parallel to the PN,

adaptive transcription programs are turned on to restore proteostasis during

stress.

Molecular chaperones

There are five classes of ATP-dependent chaperones: Hsp70, Hsp40, Hsp90,

Hsp60 (chaperonin) and Hsp100. Moreover, there are the ATP-independent

small heat shock proteins (sHsp) (Figure 7) [121, 122].

Hsp70 can act on the nascent polypeptide emerging from the ribosome to

ensure proper folding as well as on misfolded proteins. It targets the proteins

for degradation or reactivation by recognizing and binding exposed hydro-

phobic segments. Hsp40 (J-proteins) interacts with Hsp70 to accelerate its

ATPase activity and is responsible for Hsp70 substrate specificity. Addition-

ally, Hsp70 interacts with nucleotide exchange factors (NEF) [123, 124].

Hsp90 functions downstream of Hsp70, where it mainly assist in final fold-

ing and assembly of higher order complexes [123, 125].

Hsp60 also acts downstream of Hsp70 and facilitates final folding of pro-

teins. Hsp60 forms multimeric complexes of cylindric shape building a fold-

ing chamber for its client proteins [122, 126].

Despite all these surveillance mechanisms for protein folding and refolding,

the proteostatic capacity can become overwhelmed by stress or aging, caus-

ing aggregate formation [125, 127]. For a long time, it was assumed that

proteins accumulated in aggregates cannot be reactivated. It was only in the

1990s when Lindquist and co-workers showed that Hsp104, a protein of the

Clp/Hsp100 protein family, can disaggregate proteins in yeast [128]. As

many other chaperones and proteases, Hsp104 and its bacterial homolog

ClpB are members of the AAA+ (ATPase associated with diverse cellular

activities) superfamily [123, 129]. Homologs of the AAA+ superfamily are

found in fungi, bacteria, plants and mitochondria of eukaryotes. As the name

suggests, members of the AAA+ superfamily fulfill many different cellular

functions but share a highly conserved 200 to 250 amino acid long ATPase

module. This module contains the Walker A and B motifs, which harbor the

31

nucleotide binding and hydrolysis domain. Another similarity amongst the

members is the assembly into hexameric ring structures [130].

Hsp104/ClpB mostly act in a bi-chaperone system together with

Hsp70/DnaK [129, 131]. First, Hsp40/DnaJ binds to the aggregated proteins

ensuring substrate specificity of Hsp70/DnaK. Once Hsp70/DnaK bound to

the hydrophobic patches of the aggregates, it recruits Hsp104/ClpB.

Hsp104/ClpB subsequently unfolds the aggregates [127, 132, 133].

Metazoans lack a homolog of Hsp104/ClpB in the cytosol but they still show

disaggregase activity [121, 125]. It was recently reported that this disaggre-

gase activity comes from a concerted action of Hsp70, Hsp40 and Hsp110

(NEF of Hsp70) [134-136].

Following unfolding, the protein is either refolded assisted by Hsp70/DnaK

or channeled to the degradation pathway. It is still not exactly clear how the

triage decision between refolding, degradation and sequestration is achieved

on a molecular level [125].

Figure 7: Proteostasis network. Hsp70 assists proteins emerging from the ribo-

some in folding. Interaction with Hsp40 confers substrate specificity. It also interacts

with misfolded proteins helping them to refold or channel them to the UPS for deg-

radation. Hsp90 and Hsp60 act downstream of Hsp70 and assist in final folding as

well as complex assembly. In a bi-chaperone system with Hsp104, Hsp70 helps

disaggregating proteins (NEF, nucleotide exchange factor; UPS, ubiquitin pro-

teasome system).

The ubiquitin-proteasome system

The ubiquitin-proteasome system (UPS) mediates degradation of misfolded

or unfunctional proteins as well as proteins which are no longer needed. The

first step to mark a protein for degradation is ubiquitination. Ubiquitin is

transferred to the protein via the concerted actions of three enzymes, E1, E2

32

and E3. The ubiquitin-activating enzyme E1 forms an energy rich thioester

with ubiquitin. Following activation, ubiquitin is transferred to the ubiquitin-

conjugating enzyme E2 and, subsequently, the ubiquitin-ligase E3 transfers

ubiquitin to a lysine residue within the target protein. E3 is also the enzyme

which confers substrate specificity. This process can be reversed by deubiq-

uitinating enzymes. Ubiquitinated proteins are recognized and degraded by

the proteasome.

The 26S proteasome consist of the 20S core particle and 19S regulatory par-

ticles. The 19S particle, a member of the AAA+ superfamily, recognizes the

substrate and uses energy derived from ATP hydrolysis to transfer it into the

proteolytic chamber of the 20S particle [137, 138]. Degradation of proteins

via the UPS not only determines the proteome of the cytosol but also of oth-

er organelles.

Sequestration compartments

Another arm of the PN is the sequestration of aggregated proteins into sepa-

rated compartments. Besides stress-induced aggregation, aggregates can also

form during the normal life of a cell. Aggregation under non-stress condi-

tions can be caused by mutated versions of proteins, inefficient protein bio-

genesis or accumulation of unassembled complex subunits [125, 139]. At

first aggregates form stochastically within the cytosol but are then actively

collected at specific sequestration sites [122, 140].

In yeast, there are three different aggregate structures described: the intracel-

lular quality control compartment (INQ, former JUNQ), the insoluble protein

deposit (IPOD) and CytoQ (also stress foci/Q-bodies).

INQ contains ubiquitinated proteins and was shown, by cryo-electron mi-

croscopy, to be localized inside the nucleus [141]. It only builds up during

stress and disappears after stress release. The clearance of INQ probably

happens by either refolding with the help of the bi-chaperone system or via

degradation through the colocalized proteasome [142].

IPOD, which in contrast to INQ, is a more rigid structure of immobile, not

ubiquitinated proteins, can also build up in unstressed cells. It colocalizes

with the autophagy marker Atg8 [142].

The third compartment, CytoQ, is the equivalent to INQ but is localized in

the cytosol. In yeast, for example, Hsp42, a small Hsp, colocalizes with sev-

eral small stress foci which then merge to a single aggregate [125, 140, 142,

143].

In mammals, the formation of similar compartments was reported, pointing

towards an evolutionary conservation of this process [142, 144]. What might

be the advantage to sequester aggregates in specific compartments? Seques-

tration of aggregates separates the aggregates from cellular processes, lowers

the load on the protein quality control system, increases refold-

ing/degradation rates and allows asymmetric inheritance of aggregates. All

this converges in the reduction of cytotoxicity [125, 140].

33

Hsf1 and Msn2/4 - adaptive responses

In parallel to the PN handling the misfolded proteins during stress, the heat

shock response (HSR) is turned on [122, 126]. This adaptive response results

in a decreased overall translation rate to lower the burden on the PN, while

chaperones and the proteolytic system are upregulated.

Figure 8: Transcriptional programs upon stress in yeast. In yeast, Hsf1 is always

in the nucleus and bound to HSE in promotors of target genes but upon stress its

activity is increased by hyperphosphorylation. The transcription factors Msn2/4 are

phosphorylated by the nutrient sensitive kinases TOR and PKA and anchored in the

cytosol via interaction with Bmh2. Upon stress, PKA and TOR are inhibited and

Msn2/4 migrate to the nucleus where they bind to STRE elements in promotors of

target genes (Hsf1, heat shock factor 1; PKA, protein kinase A; TOR, target of ra-

pamycin; HSE, heat shock element; STRE, stress responsive element).

In unstressed mammalian cells the transcription factor heat shock factor 1

(Hsf1) is bound by chaperones as an inactive monomer in the cytosol. Upon

stress, the chaperones are titrated away to help in protein refolding leaving a

free form of Hsf1 [145, 146]. Free Hsf1 then forms trimers and relocates to

the nucleus where it induces transcription of PN components by binding to

heat shock elements (HSE) in their promotor [145, 146]. Additionally, Hsf1

is regulated by post-translational modifications [147, 148]. In contrast to

mammals, Hsf1 in yeast is always localized in the nucleus and active, but its

activity is increased by hyperphosphorylation during stress (Figure 8) [123].

Another adaptive response is orchestrated by the transcription factors

Msn2/4. Msn2/4 are zinc-finger transcription factors activating the general

stress response in yeast [149, 150]. Under non-stress conditions, the nutrient

sensitive signaling pathways TOR and cAMP/PKA inhibit Msn2/4 activity

by phosphorylation and anchoring it in the cytosol via interaction with the

14-3-3 protein Bmh2 [151]. Upon stress, Msn2/4 translocate to the nucleus

34

where they bind to stress elements (STRE) in promotors of target genes,

activating transcription and thereby regulating protein synthesis, cell growth,

stress resistance and metabolism (Figure 8) [110, 152, 153].

Often, but not always, STRE and HSE are found in promotors of the same

target genes. A few years back, for example, it was shown that they can act

together as well as independently to activate HSP104 transcription [154].

Peroxiredoxins

Peroxiredoxins (Prx) are a conserved class of antioxidative enzymes, con-

taining a peroxidatic cysteine residue in their catalytic site. This cysteine

(-SH) gets oxidized to sulfenic acid (-SOH) during the catalytic cycle, fol-

lowed by disulfide formation (S-S) with a second cysteine, resulting in a

dimer. Reduction of the disulfide by thioredoxin (Trx) completes the catalyt-

ic cycle of Prx. Alternatively, the Prx-SOH can become hyperoxidized to

sulfonate (-SOOH) (Figure 9). This represents an inactive, locked form of

Prx which can be reduced to Prx-SOH by sulfiredoxins (Srx) [155-157].

Figure 9: Peroxiredoxins in yeast. Tsa1-SH, the major peroxiredoxin in yeast, gets

hyperoxidized to Tsa1-SOOH. The enzymes Srx1 and Trx1 reduce it again to Tsa1-

SH (figure copied from [158]).

As the route of Prx hyperoxidation and inactivation is evolutionary con-

served, it was concluded that it must be accompanied with a gain-of-

function. Indeed, it was shown that this form of Prx not only forms high

molecular-weight complexes but also has chaperone function. Furthermore,

the floodgate model was proposed. In this model, Prx is inactivated in order

to allow a local increase in peroxide levels which enables peroxide to act as

a local signaling molecule without disturbing the cells redox state [156, 157].

The function of Prx in aging is currently under investigation. It was shown

that during aging, yeast as well as rat Prx become hyperoxidized [159, 160]

and that an extra copy of Srx1 prolongs lifespan in yeast [159]. Additionally,

Hanzén et al showed that the hyperoxidized form of Tsa1 (the major Prx in

yeast) is needed for H2O2- and aging-induced aggregate recognition by the

bi-chaperone system Hsp70/Hsp104 and that the clearance of these aggre-

gates is dependent on Tsa1 as well as Srx1 [161].

35

Proteostasis during aging

Loss of proteostasis is a hallmark of aging [34]. It was shown in model or-

ganisms ranging from yeast to mice as well as in humans, that the proteostat-

ic capacity declines with age and aggregates accumulate [139, 162, 163].

The significance of keeping proteostasis is further shown by the number of

age-related diseases caused by aggregates, such as Parkinson’s or Alz-

heimer’s disease [126, 139]. Several studies also reported that deletion of

components of the PN resulted in a shortened RLS in yeast accompanied by

accumulation of aggregates [159, 164, 165]. It is assumed that during aging

the capacity of the PN declines due to errors in translation or splicing and

accumulation of oxidatively damaged proteins. This, of course, would also

affect proteins of the PN, especially oxidation-prone chaperones [139].

The sequestration of aging-induced aggregates is similar as during acute

stress but the sequestration sites do neither correspond to INQ nor IPOD.

With increasing age, however, the ability of the cell to sequester aggregates

decreases, probably due to a breakdown of different cellular functions [139].

This leaves the cell with multiple cytotoxic aggregates.

Protein quality control inside mitochondria

Mitochondria harbor many crucial processes for cell survival (see chapter

‘Mitochondria’). Therefore, it is not surprising that mitochondrial protein

homeostasis is involved in cellular fitness and mitochondrial dysfunction is

associated with many, but especially neurological, diseases as well as aging.

Hence, it is an important task of the organelle to keep its proteins in a func-

tional state. In order to do so, mitochondria contain their own proteostasis

network which closely resembles the one of their bacterial ancestor [166,

167]. They contain a complete set of chaperones as well as proteases (Figure

10) [167, 168]. In a second line of defense, damaged mitochondria can be

eliminated by mitophagy [168].

Mitochondrial chaperones

Mitochondria contain the same classes of chaperones as are found in bacteria

and the cytoplasm of eukaryotes. Members of the Hsp70 family, Hsp40 fam-

ily, Hsp60 family as well as Hsp100/ClpB family have been found and stud-

ied inside the organelle [167, 168].

The major mitochondrial Hsp70 in yeast is Ssc1. Ssc1 assists in two im-

portant events: import of proteins into mitochondria as well as folding of

polypeptides [169]. It was shown to interact with Tim44 [170], a protein of

the import machinery, and is necessary for proper import of proteins into the

matrix. After import, Ssc1 is also the first chaperone to interact with the

newly imported, unfolded protein. As other Hsp70s, Ssc1 binds to hydro-

phobic patches in the protein and assists in folding. It also helps mitochon-

36

drial encoded proteins to fold once they emerge from the mitochondrial ribo-

some [171]. Furthermore, Ssc1 interacts with different J-proteins which con-

fer substrate specificity [172], as well as Mge1, the major mitochondrial

nucleotide exchange factor [173]. As its cytosolic counterpart, it is not only

needed during physiological conditions but also during stress for refolding of

misfolded proteins [174].

After release by Hsp70, some proteins need further assistants with folding

and therefore interact with the chaperonin Hsp60. As cytosolic Hsp60,

mtHsp60 forms a folding chamber and Hsp10 builds the lid [167, 175].

Some proteins, however, can be folded by Hsp60 in the absence of Hsp10

[176]. Upregulation of mtHsp60 transcription is a major indicator for induc-

tion of the mitochondrial unfolded protein response (UPRmt) [177]. The im-

portance of mitochondrial proteostasis is also underlined by the finding that

deletion of mtHsp60, Hsp10 or Ssc1 in yeast are lethal and that mutations in

mtHsp60 are associated with neurological disorders in patients [168].

Hsp78 is the mitochondrial disaggregase in yeast and a member of the

ClpB/Hsp100 family [129]. It shows a high sequence homology to the cyto-

solic disaggregase Hsp104 and can even replace Hsp104 function in hsp104

deletion strains [178]. Hsp78 itself was reported to be required for restoring

mitochondrial function after, rather than during, stress [179]. As its cytosolic

counterpart, Hsp78 was also shown to act in a bi-chaperone system together

with Ssc1 and Mge1 [180]. Besides the interaction with the Hsp70 chaper-

one system, Hsp78 also cooperates with the proteolysis system [166, 168].

Mitochondrial proteases

Mitochondria are surrounded by two membranes. Thus, proteins residing

inside the organelle cannot be targeted and degraded by the cytosolic ubiqui-

tin proteasome system. In order to maintain mitochondrial proteostasis, and

thereby cellular fitness and survival, the organelle contains its own proteases

to degrade unassembled or misfolded proteins. As many other chaperones

and proteases, mitochondrial proteases belong to the AAA+ superfamily of

proteins (see section ‘Molecular chaperones’) [167, 181]. Here, I will focus

on the three main classes of proteases.

The soluble serine protease Lon (Pim1 in yeast) resides in the matrix and

resembles most closely the UPS in the cytoplasm [181, 182]. Main targets of

Lon are oxidatively damaged proteins [167, 183]. Lon lacks an intrinsic un-

folding activity probably to restrict degradation to unfolded and damaged

proteins. Lon cooperates with the Hsp70 chaperone system [167].

Another soluble matrix protease, ClpXP, is found in mammalian mitochon-

dria but its function is not well characterized yet [168]. The protease ClpP

cooperates with the hexameric chaperone ClpX which unfolds the substrate

and ensures substrate specificity of ClpP [181]. ClpP was reported to play a

role in the UPRmt (see chapter ‘Interorganellar communication’). Surprising-

37

ly, only a homolog for ClpX is found in yeast mitochondria (Mcx1) but not

for ClpP [184].

The IMM represents the protein richest membrane in the cell as it harbors

the OXPHOS complexes, many different carriers as well as the import ma-

chinery [167, 181]. For cell survival it is critical to keep these systems intact,

explaining the need for a protease system in the IMM. Besides oxidatively

damaged proteins, unassembled OXPHOS subunits are major targets of

these proteases. There are two main metalloproteases in the IMM, i-AAA

and m-AAA, both members of the FtsH/AAA family. Both proteases assem-

ble in hexameric ring structures forming a proteolytic chamber but expose

their catalytic sites on opposite sides of the IMM allowing surveillance of

proteostasis on both sides of the membrane [168, 181, 182, 185]. Having a

rather degenerate substrate specificity, they recognize membrane topology

and folding state of proteins and 10 to 20 amino acids of solvent exposed

stretches are sufficient for recognition and complete degradation of sub-

strates [186].

In yeast, the m-AAA protease consists of alternating subunits of Yta10 and

Yta12 and the catalytic site faces the matrix [187]. Not many endogenous

substrates of these proteases are known yet and their proteolytic activity was

mainly studied on imported model substrates. However, it was shown that

unassembled OXPHOS subunits are substrates of the m-AAA protease [188,

189]. Besides complete degradation of proteins, the m-AAA protease was

also shown to have non-proteolytic functions [190]. The prohibitins Phb1

and 2 are negative regulators of m-AAA protease. They form large assem-

blies in the IMM and interact with m-AAA protease [191]. The exact molec-

ular basis of inhibition of m-AAA activity by Phb1/2 is not understood but it

was shown that deletion of PHB1 increases degradation of unassembled

complex IV subunits [192]. It was suggested that prohibitins might interact

with and stabilize newly synthesized OXPHOS subunits to protect them

from degradation [193]. Another possibility is that they affect protease activ-

ity by changing the lipid environment, as prohibitins have a scaffolding

function and can compartmentalize the membrane [191].

The i-AAA protease is built by six Yme1 subunits and the catalytic domains

face the IMS [194]. The substrates for i-AAA protease are still more elusive

than for m-AAA protease, but unassembled Cox2 was shown to be a sub-

strate [195, 196]. In contrast to the negative regulation of m-AAA protease

by prohibitins, i-AAA protease is positively regulated by Mgr1 and Mgr3.

Mgr1 and 3 form a complex and interact with substrates as well as with i-

AAA protease, serving as an adaptor between them. Deletion of either pro-

tein decreases protein turnover by i-AAA protease [197, 198]. In addition to

their proteolytic activities both proteases also have chaperone functions and

play a role in OXPHOS assembly [167].

38

Figure 10: Mitochondrial proteostasis network. Mitochondria harbor a full set of

chaperones as well as proteases to maintain organellar proteostasis. Hsp70 interacts

with imported, newly synthesized as well as misfolded proteins to assist in folding.

Hsp60, in concert with Hsp10, acts downstream of Hsp70. Hsp78 is the mitochon-

drial disaggregase and cooperates with the Hsp70 chaperone system. The Lon prote-

ase degrades soluble proteins in the matrix. The i- and m-AAA proteases are mem-

brane anchored proteases (NEF, nucleotide exchange factor; IMM, inner mitochon-

drial membrane; IMS, intermembrane space; TIM, translocase of inner membrane).

Sequestration compartments

A recent study reported a new sequestration compartment called intramito-

chondrial quality control compartment (IMiQ) [199]. It forms upon heat

stress as well as import of aggregation-prone model proteins into mitochon-

dria and, as the name suggests, is located inside mitochondria. However, it is

sequestered to the end of the mitochondrial network and colocalizes with

mitochondria devoid of membrane potential. In contrast to cytosolic INQ

which disappears a few hours after the stress release [142], a clearance of

IMiQ was not observed within 24 hours. Further studies are needed to de-

termine the contents of this aggregation site and follow its fate which might

be clearance by mitophagy.

39

Interorganellar communication

Mitochondria are deeply integrated into the cells energy metabolism and

signaling events. Furthermore, the central component of energy production,

the OXPHOS system, is encoded by two separate genomes and expression

needs to be coordinated. Having this in mind, it is not surprising that mito-

chondria constantly sent and receive signals from the nucleus as well as from

the cytosol.

Mitochondria - nucleus communication

Anterograde signaling

Mitochondrial metabolism has to be adapted to cellular needs. Therefore,

mitochondrial biogenesis needs to be tightly regulated and coordinated

through changes in nuclear gene expression. The pathways involved are the

main nutrient sensitive signaling pathways PKA, TOR and AMPK [110,

200]. Under conditions of high glucose, representing fermentative growth

conditions in yeast, PKA and TOR are activated and repress mitochondrial

biogenesis and OXPHOS components on a transcriptional level [151, 201].

In the presence of non-fermentable carbon sources as glycerol PKA and

TOR are inhibited, allowing for mitochondrial biogenesis via activation of

the transcription factors Msn2/4 [200]. As described in the section ‘Protein

quality control in the cytosol’, PKA and TOR also respond to stress, result-

ing in a coordination of aerobe metabolism and stress signaling as shown by

the similarities in gene expression in stressed and diauxic shift cells [200].

Retrograde signaling

The retrograde response in yeast

The retrograde response (RTG) describes induction of a transcription pro-

gram in the nucleus induced by mitochondrial dysfunction in yeast [202].

Rtg1 and 3 are basic helix-loop-helix transcription factors which form a het-

erodimer and bind to R-box elements in the promotor of target genes [203,

204]. Partial dephosphorylation of Rtg3, which is mediated by Rtg2 pro-

motes translocation of the heterodimer Rtg1/3 to the nucleus and its binding

to target genes (Figure 11) [205]. The kinase Mks1, in contrast, phosphory-

lates Rtg3 thereby preventing its nuclear localization [206, 207]. Over 400

genes are regulated by these transcription factors including genes of the TCA

cycle, mitochondria and peroxisome biogenesis as well as redox defense,

ultimately leading to a remodeling of metabolism [202, 208]. Such a remod-

eling is not only activated by dysfunctional mitochondria when energy sup-

ply is low, but also during the diauxic shift, where the cells switch from fer-

mentation to respiration. The RTG is interconnected with multiple other

40

signaling pathways, one of them being TOR signaling [200]. Reflecting a

state of high energy, active TOR signaling inhibits Rtg1/3 gene expression

[209, 210].

Even though the signaling sequence of the RTG is well studied, the signal

which activates it remains more elusive. Originally the RTG was described

in cells lacking mtDNA, but today also defects in OXPHOS assembly, mito-

chondrial proteostasis or mitochondrial energy metabolism are known to

induce RTG signaling [202]. Likely a drop in ATP levels [211] or a drop in

mitochondrial membrane potential [212], both resulting from mitochondrial

dysfunction, are the signals.

Most mitochondria-to-nucleus signaling pathways were reported to influence

lifespan in various model organisms. Along these lines, it was shown that

activation of RTG signaling increases RLS [213].

Although there are no homologs for the Rtg proteins in metazoans, there are

similar pathways which enable a signaling from mitochondria to the nucleus

regulated by calcium concentration, ROS levels and ATP concentration

[200, 214].

Figure 11: Retrograde response in yeast. Dysfunctional mitochondria activate

RTG signaling. Likely a drop in ATP concentration or mitochondrial membrane

potential activates the transcription factors Rtg1 and 3 to translocate to the nucleus

where they activate transcription of target gene (TCA cycle, tricarboxylic acid cycle;

mt biogenesis, mitochondrial biogenesis) (figure modified from [110]).

The mitochondrial unfolded protein response in C. elegans and mammals

Because maintaining mitochondrial proteostasis is crucial for cell survival,

mitochondria contain their own set of chaperones and proteases to ensure

proteostasis [167] as already described in the chapter ‘Mitochondrial protein

quality control’. However, if the stress is too severe and the mitochondrial

protein quality control system becomes overwhelmed, an adaptive response

is activated. Such stress responses can be counted as a branch of protein

quality control and they were described for bacteria, the cytosol (HSR) and

for the endoplasmic reticulum (UPRER) of eukaryotes.

The first description of a mitochondrial unfolded protein response (UPRmt)

was already over 20 years ago [215]. Martinus et al observed an upregula-

tion of mitochondrial hsp60 and hsp10 in rho0 hepatozytes, which was dis-

tinctive from the general HSR. Today, the UPRmt is best studied in C. ele-

41

gans where a well described sequence of events is established (Figure 12)

[110, 216]. Accumulation of misfolded proteins inside mitochondria acti-

vates the protease ClpP which degrades these proteins [217]. The generated

peptides are then exported by the peptide transporter HAF-1 [218] resulting

in relocation of the transcription factor ATFS-1 from the cytosol to the nu-

cleus. In detail, ATFS-1 contains a mitochondrial targeting signal (MTS) as

well as a nuclear localization signal (NLS). Under non-stress conditions

ATFS-1 is imported to mitochondria and degraded by the protease Lon. Dur-

ing mitochondrial stress, however, ATFS-1 is imported to the nucleus where

it, together with the transcription factors Dve1 and Ubl5, induces expression

of target genes including genes of mitochondrial biogenesis, antioxidative

defense as well as mitochondrial proteases and chaperones [177, 218].

How do peptides activate translocation of ATFS-1 to the nucleus? Might it

rather be a general inhibition of mitochondrial protein import by a decrease

in membrane potential that causes the nuclear localization of ATFS-1? This

statement is supported by the observation that HAF-1 is not vital for activa-

tion of UPRmt [216]. On the other hand, also in yeast the release of peptides

from mitochondria was suggested to play a role in mitochondria to nucleus

signaling [219].

Figure 12: Mitochondrial unfolded protein response in C. elegans. Misfolded

proteins inside mitochondria are degraded by the protease ClpP. The generated pep-

tides are exported to the cytosol via HAF-1. By a yet unknown mechanism the pep-

tides activate translocation of the transcription factor ATFS-1 into the nucleus where

it activates transcription of target genes to restore mitochondrial proteostasis. In a

second branch of the UPRmt ROS released from mitochondria inhibit cytosolic trans-

lation via activation of the kinase Gcn2 and subsequent phosphorylation of the elon-

gation factor 2α (ROS, reactive oxygen species; OXPHOS, oxidative phosphoryla-

tion; MTS, mitochondrial targeting signal; NLS, nuclear localization signal) (figure

modified from [110]).

42

To lower the burden on the PN, downregulation of general cytosolic transla-

tion is a common response to stress. Also the UPRmt results in a reduction of

cytosolic protein synthesis. Gcn2, which is activated by an increase in ROS,

phosphorylates the eukaryotic initiation factor 2α (eIF2α) [220]. P-eIF2α

inhibits general cytosolic translation but promotes translation of stress-

induced proteins [221].

In mammals, the picture is not as clear as in C. elegans. ClpP degrades mis-

folded proteins inside mitochondria which are exported by a yet unknown

mechanism, resulting in activation of the kinase JNK and subsequent phos-

phorylation of the transcription factor cJun. cJun then induces expression of

CHOP and C/EBPβ [222-224], which are two leucine zipper transcription

factors involved in induction of gene expression of mitochondrial chaper-

ones and proteases in response to mitochondrial stress [222, 225]. Similar to

the response in worm, a reduction in cytosolic translation was reported

[226].

The mammalian transcription factor ATF5 is likely a homolog of ATFS-1,

as it can replace ATFS-1 function in worms. As ATFS-1, ATF5, is a leucine

zipper transcription factor containing a MTS as well as a NLS and is in-

volved in protective gene expression during mitochondrial stress [227].

However, it remains to be studied in the future if ATF5 localization is regu-

lated by peptide export and how CHOP, C/EBPβ and ATF5 play together.

More recently another protein, GPS2, with dual targeting sequence was re-

ported to mediate mitochondrial retrograde signaling in mouse adipocytes

during mitochondrial depolarization [228]. Taken the dual localization of

ATFS-1, ATF5 and GPS2 together this might be a common mechanism of

regulation of gene expression in response to mitochondrial stress.

Different perturbations inside mitochondria were shown to activate the

UPRmt. Besides mtDNA depletion, impaired mitochondrial protein quality

control and defects in OXPHOS [177, 229], also changes in mitochondrial

translation were reported to change nuclear gene expression in model organ-

isms from yeast to mice [110].

Stress signaling evoked by dysfunctional mitochondrial translation

Several studies in yeast reported that deletion of certain mitochondrial ribo-

somal proteins activated retrograde signaling resulting in a changed nuclear

gene expression and an increase in RLS [219, 230-232]. The exact nature of

this response, however, seems to depend on the gene deleted and is not

caused by a general defect in mitochondrial translation. This is also support-

ed by the fact that deletion of mitochondrial ribosomal proteins in general

did not induce such a response or change lifespan [230-232]. The following

examples show how diverse the signals are when different components of

the mitochondrial translation machinery are deleted.

Deletion of Sov1, the translational activator of the ribosomal protein Var1,

increased RLS. This increase in RLS was dependent on the transcription

43

factors Msn2/4, which were activated by decreased PKA signaling, and on

the deacetylase Sir2 [230].

An increase in RLS was also observed when the ribosomal protein Mrpl32

was deleted or its assembly into the ribosome was blocked. The increase in

RLS was dependent on the transcription factor Gcn4 [231, 233]. In contrast,

deletion of another ribosomal protein of the large subunit, Mrpl25, increased

RLS through activation of Tor signaling. Activated Tor signaling resulted in

nuclear localization of the transcription factor Sfp1 [232].

The increase in lifespan caused by deletion of mitochondrial ribosomal pro-

teins was not only observed in yeast but is conserved in higher eukaryotes.

In an RNAi screen, 22 mitochondrial ribosomal proteins or translation fac-

tors were identified to activate UPRmt and change lifespan in C. elegans

[234]. In another study, knock down of Mrps5 in C. elegans and mice in-

creased lifespan by specifically activating the UPRmt, but not UPRER or HSR.

In contrast to the observations in yeast, this was accompanied by a decrease

in mitochondrial translation causing a mitonuclear protein imbalance of

OXPHOS subunits. Also inhibition of mitochondrial translation with

doxycycline activated the UPRmt. In line with this, deletions of other mito-

chondrial ribosomal proteins increased lifespan, pointing towards a more

general mechanism [235]. This is different to the observations in yeast,

where only deletion of certain mitochondrial ribosomal proteins increased

lifespan but not others [230-232].

Mitochondrial stress and cytoplasmic proteostasis

Besides retrograde signaling changing nuclear gene expression in response

to mitochondrial stress, more and more evidence accumulates for a mito-

chondria-cytosol communication axis. Specifically work in yeast has identi-

fied pathways where mitochondrial function impacts on cytoplasmic proteo-

stasis [200, 214, 236-240].

In two studies published back to back in 2015 cytosolic accumulation of

mitochondrial precursor proteins was shown to induce a response in the cy-

tosol (Figure 13). By simultaneously decreasing cytosolic translation and

increasing protein degradation, this response reduces the burden on the PN

[236, 237]. In the first study, Chen et al screened for multisuppressor genes

of cell death induced by mitochondrial precursor overaccumulation stress

(mPOS). Surprisingly, the suppressor genes identified were not mitochondri-

al proteins but instead involved in cytosolic proteostasis. Amongst others,

the identified proteins are known to reduce cytosolic protein synthesis and

induce protein degradation [236]. The identification of proteasome assembly

factors was especially interesting as they were also found in the second study

to be induced during mPOS. In that study, Wrobel et al used a mutant which

is impaired in mitochondrial protein import into the IMS to study the effects

of mistargeted proteins (which activate mPOS) on cell physiology. An in-

44

crease in proteasome activity via proteasome assembly was observed, con-

comitant with a reduction in protein translation in the cytosol [237]. It was

suggested that this UPR activated by mistargeted proteins (UPRam) is a

protective response triggered by mPOS to lower the load on the PN and re-

store cytosolic proteostasis during import stress. Both studies show a previ-

ously undervalued communication between mitochondrial function and cyto-

solic proteostasis.

In line with these studies, IMS stress in mammalian cells also activated the

proteasome [241] and was linked to longevity [242, 243]. The observed in-

crease in proteasome activity caused by mPOS and IMS stress, however,

stands in contrast to the previously shown proteasome disassembly caused

by an increase in ROS during mitochondrial dysfunction [244, 245]. The

opposite effects on the proteasome might stem from the different signals

released from mitochondria (ROS versus mPOS).

A more recent study reported on another stress response activated by mPOS.

The mitochondrial comprised protein import response (mitoCPR) surveils

mitochondrial protein import by targeting proteins accumulating in the trans-

locase channel or at the surface of mitochondria for degradation [238].

Figure 13: Responses caused by mitochondrial dysfunction. Cytosolic accumula-

tion of mitochondrial precursor proteins due to dysfunctional mitochondria disturbs

cytosolic proteostasis. UPRam reduces the burden on the PN by simultaneously

decreasing translation and increasing protein degradation. mitoCPR also increases

protein degradation. In parallel adaptive transcription programs are turned on by

retrograde signaling (mPOS, mitochondrial precursor overaccumulation stress;

UPRam, UPR activated by mistargeted proteins; mitoCPR, mitochondrial comprised

protein import response; TOM, translocase of outer membrane; TIM, translocase of

inner membrane).

45

In addition, an active role of mitochondria in clearance of aggregated pro-

teins was proposed. The import of aggregated proteins from the cytosol into

mitochondria was reported, where the proteins are degraded by the Lon pro-

tease Pim1. This mitochondria as guardian in the cytosol (MAGIC) response

required the disaggregase Hsp104 as well as active mitochondrial protein

import [246]. Though it remains open how cytosolic proteins should be im-

ported by the highly selective mitochondrial import machinery in the ab-

sence of a MTS.

A boost in the research field of mitochondrial signaling and its connection to

homeostasis was observed in 1996 after the discovery that apoptosis is trig-

gered by the release of cytochrome c from mitochondria [247]. After 20

years of research in this field, many important findings have been made and

it is well established that mitochondrial dysfunction activates protective re-

sponses to restore mitochondrial (retrograde signaling, UPRmt) as well as

cytosolic proteostasis (UPRam, mitoCPR). Besides a well-defined RTG

signaling in yeast and UPRmt in C. elegans, other signaling pathways remain

more elusive and clearly need further investigation.

46

47

Aims

Mitochondria harbor the OXPHOS system which produces most of the cellu-

lar energy in form of ATP. The OXPHOS system is encoded by two ge-

nomes residing in different organelles. Consequently, to supply the cell with

ATP, mitochondrial and nuclear gene expression need to be coordinated to

allow for proper OXPHOS assembly and eventually ATP production. In

contrast to bacterial translation, where all the steps are mechanistically and

kinetically well characterized, mitochondrial translation is a process which is

not yet defined on a molecular level. This lacking behind is mainly due to a

missing in vitro translation system. Also the accuracy of mitochondrial trans-

lation remains elusive, as directed modification and, therefore, the use of

mitochondrial encoded reporter proteins is not possible in mammals. Even

though this is possible in yeast, it is not straight forward.

The work presented in this thesis focuses on mitochondrial translation in the

model organism S. cerevisiae. Establishing methods to study timing, coordi-

nation and accuracy of mitochondrial translation was a major goal. Since the

cryoEM structure of the mitochondrial ribosome from yeast as well as from

mammals was solved, it is clear that the mitochondrial ribosome changed

during evolution compared to its bacterial ancestor. However, the decoding

center is highly conserved. Therefore, mutations in this region of the ribo-

some, which were shown to change bacterial translation accuracy, were used

to study mitochondrial translation accuracy and its effect on cellular homeo-

stasis and aging.

48

49

Summaries of papers

Paper I – A novel system to monitor mitochondrial transla-

tion in yeast Mitochondria arose from the engulfment of an α-proteobacterium by an an-

cestral eukaryotic cell in a process called endosymbiosis. Even though, most

of the mitochondrial genome was transferred to the nucleus during evolution,

mitochondria still contain a small organellar genome. In yeast, this genome

encodes eight proteins. To date, studying mitochondrial translation in vivo

requires inhibition of cytosolic translation with the antibiotic cycloheximide.

By inhibiting translation, however, levels of free amino acids increase,

which activates TOR signaling and eventually changes cell growth, nutrient

signaling and life span. Not only protein synthesis but also protein degrada-

tion is affected by cycloheximide and even a direct effect on mitochondrial

translation was reported.

We established a mitochondrial encoded superfold GFP (GFPm) which can

be used as a reporter to follow mitochondrial protein synthesis. Using biolis-

tic transformation we modified the mitochondrial genome in a way that

GFPm is expressed as a ninth protein and does not interfere with mitochon-

drial respiration. Using superfold GFP we aimed at improving the folding as

it has faster folding kinetics and is more stable compared to normal GFP. We

show that expression of GFPm is regulated in the same way as of other mito-

chondrial encoded proteins and that expression can be followed via western

blot, flow cytometry as well as fluorescent microscopy. Furthermore, after

inhibition of mitochondrial translation or during glucose repression of mito-

chondrial biogenesis, the GFPm signal decreased, making it a good reporter

to detect changes in mitochondrial translation. With this system mitochon-

drial translation can be studied in vivo by visualizing GFP fluorescence

without interfering with cellular homeostasis. This can be a pivotal tool to

study timing, coordination and regulation of mitochondrial translation.

50

Paper II - Different genetic approaches to mutate the mito-

chondrial ribosomal protein S12 Over the last decades, an ever-growing number of methods and tools to mod-

ify the genome of the model organism S. cerevisiae became available. When

the genome of yeast was first sequenced in 1996 in an international project,

it became clear that many genes are conserved between humans and yeast.

Therefore, and for many other reasons, yeast is a great eukaryotic model

organism for basic research.

The most common method to study a protein and its implications in a certain

pathway or process is deleting the gene by replacing it with a selection cas-

sette using the yeast’s own homologous recombination system. The selection

cassette can confer resistance to an antibiotic. Consequently, growth in the

presence of this antibiotic can be used as selection to screen for successful

recombination events. Alternatively, selection cassettes can be auxotrophic

markers providing the cell with the ability to grow in the absence of a certain

amino acid.

A mutated version of the protein can then be expressed from a plasmid or

from the genome. Even though expression from a plasmid is standardly used,

in some cases it is necessary to express the mutated protein from the ge-

nome.

Here, we aimed at using the URA3 cassette, a counter selectable auxotrophic

marker, to introduce a mutated version of the mitochondrial ribosomal pro-

tein S12 (Mrps12) into the genome of yeast. URA3 encodes for Orotidine-5'-

phosphate (OMP) decarboxylase, an enzyme required for uracil biosynthesis.

Therefore, growth in the absence of uracil can be used as a first selection

step. In the following step the URA3 cassette is replaced by the mutated gene

using 5-fluoroorotic acid (5-FOA) for selection. 5-FOA is only toxic for

cells still harboring the URA3 cassette as it will be metabolized to the toxic

compound 5-fluorouracil. In our hands, this method only gave false-positive

results in the second selection step, most likely caused by loss of mtDNA

during the first selection step. Loss of mtDNA is frequently observed when

deleting mitochondrial ribosomal proteins and can cause a nuclear genome

instability. This would explain why cells still harboring the URA3 cassette,

but a mutated version, can grow in the presence of 5-FOA. Moving forward

with the deletion strain, we used an integrative plasmid instead to express the

mutated MRPS12 from the genome. After repopulating the cells with

mtDNA, these mutants can be used to study translation accuracy in mito-

chondria as described in paper III.

51

Paper III – Mitochondrial translation efficiency controls cy-

toplasmic protein homeostasis Mitochondria are mostly known for their role in production of the cells ener-

gy in form of ATP. Besides this crucial role, they also fulfill many other

important tasks like harboring many metabolic processes, regulating calcium

homeostasis and cell death. Disturbances in mitochondrial function are im-

plicated in aging and disease. Therefore, it is not surprising that mitochon-

dria are no isolated organelles but are in a constant crosstalk with other or-

ganelles to receive and sent signals.

Due to their endosymbiotic origin, mitochondria still contain a small orga-

nellar genome and the associated gene expression machinery. While the

mitochondrial ribosome changed remarkably during evolution compared to

its bacterial ancestor, the decoding site remained highly conserved. The de-

coding step during elongation determines the accuracy of the protein and is a

trade-off between speed of peptide bond formation and accuracy of amino-

acyl-tRNA selection. In bacteria, this step involves the ribosomal protein

S12. Mutations in this protein were shown to change translation accuracy

and were characterized by their resistance or hypersensitivity towards ami-

noglycosides. Aminoglycosides are a class of antibiotics binding to the small

subunit of the ribosome and interfering with the decoding step by accelerat-

ing peptide bond formation and thereby decreasing translation accuracy.

Consequently, error-prone S12 mutants were hypersensitive towards amino-

glycosides while hyperaccurate strains were resistant.

Here, we introduced mutations in Mrps12, the mitochondrial orthologue of

S12 in yeast. The mutations recapitulate the bacterial phenotype in regard to

sensitivity towards aminoglycosides, the hyperaccurate mutant being re-

sistant and the error-prone mutant being hypersensitive, indicating a con-

served role in the decoding step. Furthermore, the error-prone mutant

showed a shortened CLS, an increase in ROS levels during aging and a fail-

ure in handling cytosolic aggregates. During stress, it activated an interorga-

nellar proteostasis transcription program (IPTP) upregulating mitochondrial

biogenesis, proteases and chaperones besides the general heat shock re-

sponse. Activation of IPTP was dependent on the stress transcription factors

Msn2/4 downstream of TOR. In contrast, the hyperaccurate mutant repressed

IPTP and an inappropriate activation was even disadvantageous. In addition,

the hyperaccurate mutant showed increased CLS and increased proteostatic

capacity in the cytosol. Together, these results demonstrate a direct crosstalk

between mitochondrial translation output and nuclear gene expression as

well as cytosolic proteostasis.

52

Paper IV - Insertion defects of mitochondrially encoded pro-

teins burden the mitochondrial quality control system Mitochondria are the powerhouse of the cell and produce most of the cells

ATP via the OXPHOS system. Some subunits of the OXPHOS system are

encoded in the mitochondrial genome, translated on mitochondrial ribo-

somes and co-translationally inserted into the IMM. Mitochondrial ribo-

somes are membrane anchored by different proteins, including Mrx15 and

Mba1. Correct membrane insertion and assembly of mitochondrial transla-

tion products is pivotal for cellular ATP production. To ensure proper bio-

genesis and assembly of OXPHOS subunits, mitochondria also contain a

whole set of chaperones and proteases building the mitochondrial proteosta-

sis network. There are two IMM proteases, i-AAA and m-AAA, responsible

for degradation of IMM proteins including unassembled or misfolded

OXPHOS subunits. The i-AAA protease is regulated by Mgr1 and Mgr3,

while the m-AAA protease is under control of Phb1 and Phb2.

Here, we showed that deletion of Mba1 increased sensitivity towards accu-

mulation of mitochondrial misfolded proteins, while deletion of Mrx15 ren-

dered the cells resistant towards mitochondrial proteotoxic stress. This sug-

gests interaction of the ribosome receptors Mba1 and Mrx15 with compo-

nents of the mitochondrial proteostasis network. Indeed, Mba1 showed ge-

netic interaction with regulators of both IMM proteases. In contrast, Mrx15

only showed genetic interaction with the i-AAA protease regulators Mgr1

and Mgr3. In summary, these results are a first indication for a very early

protein quality control step during OXPHOS biogenesis.

53

Future perspectives

How accurate is mitochondrial translation?

In this work I established a mitochondrial encoded reporter to follow orga-

nellar translation in respiring cells under physiological conditions (paper I).

Timing, regulation and coordination of translation can be studied with this

reporter but not accuracy. Therefore, it is a major task for future work to

establish a mitochondrial encoded reporter resembling the one used to meas-

ure error frequency in bacteria. There, a dual reporter construct consisting of

two reporter proteins connected by a short linker region is used to study

translation accuracy [248, 249]. Following expression of the first protein

serves as an internal standard. By introducing mutations in the linker region

or in the active site of the second protein error frequency can be measured

(Figure 14). An example for such a mutation would be a stop codon in the

linker region. If translation is accurate only the first protein is expressed. If

translation, however, is inaccurate both proteins are expressed due to a

readthrough of the stop codon. By calculating the ratio between the expres-

sion levels of the two proteins a number for error frequency can be estab-

lished.

Figure 14: Dual reporter system. A reporter consisting of two proteins connected

by a short linker region can be used to measure translation accuracy by calculating

the ratio between expression of both proteins. By introducing mutations in the linker

region readthrough events can be monitored.

54

Additionally, error frequency of bacterial translation is measured using an in

vitro translation system where the exact incorporation of non-cognate or

near-cognate tRNAs over cognate tRNAs can be followed. Having a similar

system for mitochondrial translation would not only enable us to determine

accuracy of translation but open up a whole new research field gaining

mechanistic as well as kinetic insights into the single steps of mitochondrial

translation.

We set out to get first insights into the role of mitochondrial translation accu-

racy for cell survival by relaying the knowledge gained from decoding in

bacteria to mitochondria. Therefore, we introduced mutations into the highly

conserved decoding center of the mitochondrial ribosome (paper II). Alt-

hough we have good evidence suggesting a changed translation accuracy in

our mutants based on of their resistance and hypersensitivity towards the

aminoglycoside paromomycin (paper III), this is only an indirect measure.

Being able to determine translation accuracy in the mutants, either with a

dual reporter system or an in vitro translation system, would be direct proof.

Furthermore, albeit all our efforts to get estimates about the speed of transla-

tion in the mutants we were not able to detect any differences. In bacteria, it

was shown that translation accuracy is a trade-off between speed of peptide

bond formation and accuracy of aminoacyl-tRNA selection [250] and evolu-

tion most likely selected for the right balance between these two factors. In

analogy, the hyperaccurate mutant should have a slower translation and the

error-prone mutant should have a faster translation compared to wild type.

Therefore, it would be interesting to be able to use an in vitro translation

system to measure the kinetics of amino acid incorporation in the mutants.

What is the signal released from mitochondria?

During stress in form of nutrient deprivation, we observed activation of the

protective IPTP in the wild type and error-prone mutant, which was re-

pressed in the hyperaccurate mutant. This response was dependent on the

transcription factors Msn2/4 (paper III). However, it remained elusive how

changes in mitochondrial translation activate Msn2/4-dependent gene ex-

pression. Is it the drop in ATP due to nutrient deprivation which activates the

stress program? This is unlikely as all three strains should experience the

same drop in ATP levels but still show different responses. The error-prone

mutant showed the strongest activation of IPTP and it could be argued that

this is due to the decreased respiratory capacity of this mutant. On the other

hand, also the hyperaccurate mutant showed a respiratory deficient pheno-

type but did not activate IPTP. For now it will remain an open question how

changes in mitochondrial translation affects nuclear gene expression in our

strains.

Furthermore, we showed that mitochondrial translation impacts on cytosolic

homeostasis (paper III). However, the way of communication remained

undiscovered. This is not the first time a crosstalk between the two orga-

55

nelles was observed and different signals were described to be part of this

communication. It was reported that accumulation of mitochondrial precur-

sor proteins in the cytosol would increase proteasome activity via activation

of UPRam [237]. Yet, we did neither observe an accumulation of precursor

proteins in the cytosol nor an increase in proteasome activity which would

argue against an activation of UPRam in our mutants. However, we cannot

exclude small changes in import kinetics from our experiments. Therefore,

we would need to follow protein import into mitochondria or measure mem-

brane potential which is crucial for protein import. Another obvious candi-

date would be ROS, especially because we observed changed sensitivity

towards oxidative stress in our mutants. But even under strict anaerobic con-

ditions we still observed a different proteostatic capacity excluding ROS as

signaling molecule in our pathway. Furthermore, an additional copy of the

mitochondrial superoxide dismutase (SOD2) could not prolong lifespan in

the error-prone mutant. Being able to detect the signal sent from mitochon-

dria to the cytosol will be a challenging task for future investigations.

Is mitochondrial protein quality control happening at the ribosome?

We showed a genetic interaction between proteins located at the tunnel exit

of the mitochondrial ribosome which are responsible for membrane protein

insertion (Mba1 and Mrx15) and regulators of the i-AAA and m-AAA prote-

ases of the IMM (paper IV). This is first evidence that protein quality con-

trol might happen directly after protein synthesis to avoid accumulation of

aggregation-prone hydrophobic proteins. A recent publication from our

group also supports this idea, as both AAA proteases were copurified with

the ribosome indicating colocalization [251].

To proof a direct involvement of Mba1 and Mrx15 in mitochondrial proteo-

stasis, clearly further experiments are needed. A good starting point would

be to monitor mitochondrial proteostasis in the absence of either of these

proteins.

In bacteria, a similar scenario is observed where the homologs of m-AAA

protease, Oxa1 and Phb1/Phb2 interact and form a complex. Here, degrada-

tion of misassembled membrane proteins is mediated by a chaperone func-

tion of YidC, the Oxa1 homolog, at a very early step during or directly after

membrane insertion [252]. If Mba1and Mrx15 also form a complex with i-

AAA protease, m-AAA protease and their regulators remains to be investi-

gated.

56

57

Sammanfattning på svenska

Cellers överlevnad är beroende av energiförsörjningen för att kunna utföra

livets alla processer. Denna energi ges av maten vi äter. En kedja av

proteinkomplex (andningskedjan och ATP-syntas), som finns i det inre

mitokondriella membranet, använder denna energi för att bilda ATP.

Mitokondrier, som vi känner dem idag, härstammar från bakterier som tagits

upp av en eukaryot förfadercell. Därför är mitokondrier den enda organellen,

förutom cellkärnan, som innehåller DNA (mtDNA). Kärnsubenheterna i

andningskedjan kodas i mtDNA och produceras i mitokondrier. För att

säkerställa energiproduktionen måste dessa subenheter monteras med

subenheter som kommer från cytosolen. Därför ger felaktig eller

okoordinerad mitokondriell proteinsyntes upphov till felaktiga

andningskedjekomplex, vilket i sin tur leder till minskad energiförsörjning

till cellen. Vidare orsakas många sjukdomar av mutationer i mitokondriens

proteinsyntesmaskin. Trots denna centrala roll för mitokondriell

proteinsyntes är överraskande lite känt om de molekylära mekanismerna.

Förutom rollen som cellens kraftverk är mitokondrier även inblandade i

många viktiga signalprocesser som bestämmer cellens välmående. Det är

därför inte överraskande att mitokondrier inte är isolerade organeller utan är

istället i konstant kommunikation med andra delar av cellen.

I detta arbete använde jag bakjästen Saccharomyces cerevisiae som en

modellorganism för att studera olika aspekter av proteinsyntes i

mitokondrier. I den första artikeln skapade jag en reporter för att övervaka

mitokondriell proteinsyntes. Denna reporter tillåter mer detaljerade studier

av timing och reglering av den mitokondriella proteinsyntesen. I den andra

och tredje artikeln använde jag tidigare kunskap om proteinsyntesen i

bakterier, mitokondriernas förfäder, för att studera noggrannhet i

mitokondriell proteinsyntes och dess konsekvenser för cellen. Vi visade att

noggrannhet i mitokondriell proteinsyntes bestämmer cellens stresshantering

och livslängd. I den sista artikeln fann vi några första bevis för en mycket

tidig kvalitetskontroll under eller direkt efter den mitokondriella

proteinsyntesen. Denna övervakningsmekanism kan undvika skadlig

ackumulering av oveckade eller ej sammansatta proteiner såväl som

felaktiga andningskedjekomplex, varigenom mitokondriell funktion kan

säkerställas.

58

59

Acknowledgments

Many people contributed to the success of this thesis. First of all, Martin,

you gave me the opportunity to do my PhD in your lab. I learned so much

from you, especially to always keep a positive attitude and never give up, no

matter how frustrating it is. You never doubted that it will pay out eventually

– and it did!

I would like to thank my co-supervisor Elzbieta Glaser and my mentor Pia

Ädelroth for their support.

Dina Petranovic and Thomas Nyström, you gave me the great opportunity to

spent time in your labs in Gothenburg. You and your groups helped me sig-

nificantly with my experiments.

Claes Andréasson, Sabrina Büttner, Per Ljungdahl, thank you for all the

intense discussions during our Friday yeast meetings and beyond. I enjoyed

getting new ideas from a “non-mitochondria-centered” point of view. Claes,

without your help and great input my paper would not have become such a

great story.

Thanks to all the past and present members of the Ott group. When I started

my PhD we were a German group of six people, which changed dramatically

over the last years to a group of 12 people from six different countries.

Kirsten, without you I would not have survived the first year of my PhD.

You cheered me up when something did not work out as planned (which

happened constantly) and you shared your endless knowledge with me.

Braulio, we started our PhD at the same time and it was fun sharing this in-

credible experience with you. Even if we had our doubts from time to time –

we did it! Hannah, you were my first student and I am glad you decided to

stay in our group. I will miss our coffee breaks during which we, of course,

only discussed about science. Kathi, it was fun working on the bench next to

you and chatting about scientific and non-scientific topics.

Povilas and Aziz, I enjoyed sharing an office with you for almost five years.

We laughed so much together, making the days full of failed experiments

bearable.

60

Anna and Nandu, I don’t know how many times I came over to your lab

asking for help and you were never tired of sharing all your knowledge about

chaperones with me. Since the paper is accepted, I sometimes catch myself

looking for excuses to be able to stop by your lab.

Lukas, you made it possible that the GFP manuscript was submitted within

no time. It was fun working with you and exchanging our knowledge about

the best model organism of all.

Agata, I am deeply grateful for the bioinformatics support you gave us.

All the people who made the last six years in Sweden such a great experi-

ence. Hannah, Jacob, Braulio, Caro, Kirsten, Kathi, Jörg, I enjoyed all our

get-togethers, especially the barbecues.

Last but not least, I could not have achieved all this without the constant

support of my family. Papa, Andrea, Melissa und Romina, ich weiss ich

kann immer auf euch zählen, was auch passiert. Dieses Wissen gibt mir die

Sicherheit und den Mut meine Träume zu verfolgen. Mama, ich weiss, du

bist stolz auf mich.

Melissa, Romina und Hannes, eure vielen Besuche waren immer eine

willkommene Abwechslung zu unserem Alltag.

Andi, ich schätze mich überglücklich dich an meiner Seite zu wissen. Du

hast mich immer wieder aufgebaut wenn ich am Boden war und hast mit mir

jeden noch so kleinen Meilenstein gefeiert. Ich hatte sechs unglaubliche

Jahre mit dir in Stockholm und sehe voller Vorfreude unserer gemeinsamen

Zukunft in Freiburg entgegen. Leonie och Andi, jag älskar er.

61

References

1. Goffeau, A., et al., Life with 6000 genes. Science, 1996. 274(5287): p.

546, 563-7.

2. Sherman, F., Getting started with yeast. Methods Enzymol, 2002. 350: p.

3-41.

3. Boeke, J.D., F. LaCroute, and G.R. Fink, A positive selection for mutants

lacking orotidine-5'-phosphate decarboxylase activity in yeast: 5-fluoro-

orotic acid resistance. Mol Gen Genet, 1984. 197(2): p. 345-6.

4. Umezu, K., et al., Purification and properties of orotidine-5'-phosphate

pyrophosphorylase and orotidine-5'-phosphate decarboxylase from baker's

yeast. J Biochem, 1971. 70(2): p. 249-62.

5. Rinaldi, T., et al., Mitochondrial diseases and the role of the yeast

models. FEMS Yeast Res, 2010. 10(8): p. 1006-22.

6. Zakharov, I.A. and B.P. Yarovoy, Cytoduction as a new tool in studying

the cytoplasmic heredity in yeast. Mol Cell Biochem, 1977. 14(1-3): p. 15-8.

7. Lancashire, W.E. and J.R. Mattoon, Cytoduction: a tool for mitochondrial

genetic studies in yeast. Utilization of the nuclear-fusion mutation kar 1-1

for transfer of drug r and mit genomes in Saccharomyces cerevisiae. Mol

Gen Genet, 1979. 170(3): p. 333-44.

8. Conde, J. and G.R. Fink, A mutant of Saccharomyces cerevisiae defective

for nuclear fusion. Proc Natl Acad Sci U S A, 1976. 73(10): p. 3651-5.

9. Bonnefoy, N. and T.D. Fox, Directed alteration of Saccharomyces

cerevisiae mitochondrial DNA by biolistic transformation and homologous

recombination. Methods Mol Biol, 2007. 372: p. 153-66.

10. Costanzo, M.C. and T.D. Fox, Control of mitochondrial gene

expression in Saccharomyces cerevisiae. Annu Rev Genet, 1990. 24: p. 91-

113.

11. Ding, M.G., et al., Introduction of cytochrome b mutations in

Saccharomyces cerevisiae by a method that allows selection for both

functional and non-functional cytochrome b proteins. Biochim Biophys

Acta, 2008. 1777(9): p. 1147-56.

12. Gruschke, S., et al., The Cbp3-Cbp6 complex coordinates cytochrome

b synthesis with bc(1) complex assembly in yeast mitochondria. J Cell Biol,

2012. 199(1): p. 137-50.

13. Bonnefoy, N. and T.D. Fox, In vivo analysis of mutated initiation

codons in the mitochondrial COX2 gene of Saccharomyces cerevisiae fused

62

to the reporter gene ARG8m reveals lack of downstream reinitiation. Mol

Gen Genet, 2000. 262(6): p. 1036-46.

14. Birky, C.W., The inheritance of genes in mitochondria and

chloroplasts: laws, mechanisms, and models. Annu Rev Genet, 2001. 35: p.

125-48.

15. Shibata, T. and F. Ling, DNA recombination protein-dependent

mechanism of homoplasmy and its proposed functions. Mitochondrion, 2007.

7(1-2): p. 17-23.

16. Patananan, A.N., et al., Modifying the Mitochondrial Genome. Cell

Metab, 2016. 23(5): p. 785-96.

17. Gammage, P.A., C.T. Moraes, and M. Minczuk, Mitochondrial

Genome Engineering: The Revolution May Not Be CRISPR-Ized. Trends

Genet, 2018. 34(2): p. 101-110.

18. Suhm, T., et al., A novel system to monitor mitochondrial translation

in yeast. Microb Cell, 2018. 5(3): p. 158-164.

19. Botstein, D. and G.R. Fink, Yeast: an experimental organism for

modern biology. Science, 1988. 240(4858): p. 1439-43.

20. Botstein, D. and G.R. Fink, Yeast: an experimental organism for 21st

Century biology. Genetics, 2011. 189(3): p. 695-704.

21. Kataoka, T., et al., Functional homology of mammalian and yeast RAS

genes. Cell, 1985. 40(1): p. 19-26.

22. Carmona-Gutierrez, D. and S. Büttner, The many ways to age for a

single yeast cell. Yeast, 2014. 31(8): p. 289-98.

23. Longo, V.D., et al., Replicative and chronological aging in

Saccharomyces cerevisiae. Cell Metab, 2012. 16(1): p. 18-31.

24. Friedman, J.R. and J. Nunnari, Mitochondrial form and function.

Nature, 2014. 505(7483): p. 335-43.

25. McBride, H.M., M. Neuspiel, and S. Wasiak, Mitochondria: more than

just a powerhouse. Curr Biol, 2006. 16(14): p. R551-60.

26. Sagan, L., On the origin of mitosing cells. J Theor Biol, 1967. 14(3): p.

255-74.

27. Archibald, J.M., Endosymbiosis and Eukaryotic Cell Evolution. Curr

Biol, 2015. 25(19): p. R911-21.

28. Borst, P. and L.A. Grivell, The mitochondrial genome of yeast. Cell,

1978. 15(3): p. 705-23.

29. Chacinska, A., et al., Importing mitochondrial proteins: machineries

and mechanisms. Cell, 2009. 138(4): p. 628-44.

30. Neupert, W. and J.M. Herrmann, Translocation of proteins into

mitochondria. Annu Rev Biochem, 2007. 76: p. 723-49.

31. Claros, M.G., et al., Limitations to in vivo import of hydrophobic

proteins into yeast mitochondria. The case of a cytoplasmically synthesized

apocytochrome b. Eur J Biochem, 1995. 228(3): p. 762-71.

63

32. Manfredi, G., et al., Rescue of a deficiency in ATP synthesis by transfer

of MTATP6, a mitochondrial DNA-encoded gene, to the nucleus. Nat Genet,

2002. 30(4): p. 394-9.

33. Sickmann, A., et al., The proteome of Saccharomyces cerevisiae

mitochondria. Proc Natl Acad Sci U S A, 2003. 100(23): p. 13207-12.

34. López-Otín, C., et al., The hallmarks of aging. Cell, 2013. 153(6): p.

1194-217.

35. Pikó, L., A.J. Hougham, and K.J. Bulpitt, Studies of sequence

heterogeneity of mitochondrial DNA from rat and mouse tissues: evidence

for an increased frequency of deletions/additions with aging. Mech Ageing

Dev, 1988. 43(3): p. 279-93.

36. Trifunovic, A. and N.G. Larsson, Mitochondrial dysfunction as a cause

of ageing. J Intern Med, 2008. 263(2): p. 167-78.

37. Bratic, A. and N.G. Larsson, The role of mitochondria in aging. J Clin

Invest, 2013. 123(3): p. 951-7.

38. Seo, A.Y., et al., New insights into the role of mitochondria in aging:

mitochondrial dynamics and more. J Cell Sci, 2010. 123(Pt 15): p. 2533-42.

39. Kukat, A. and A. Trifunovic, Somatic mtDNA mutations and aging--

facts and fancies. Exp Gerontol, 2009. 44(1-2): p. 101-5.

40. HARMAN, D., Aging: a theory based on free radical and radiation

chemistry. J Gerontol, 1956. 11(3): p. 298-300.

41. Harman, D., Free radical theory of aging: dietary implications. Am J

Clin Nutr, 1972. 25(8): p. 839-43.

42. Yang, W. and S. Hekimi, A mitochondrial superoxide signal triggers

increased longevity in Caenorhabditis elegans. PLoS Biol, 2010. 8(12): p.

e1000556.

43. Copeland, J.M., et al., Extension of Drosophila life span by RNAi of the

mitochondrial respiratory chain. Curr Biol, 2009. 19(19): p. 1591-8.

44. Mockett, R.J., et al., Ectopic expression of catalase in Drosophila

mitochondria increases stress resistance but not longevity. Free Radic Biol

Med, 2003. 34(2): p. 207-17.

45. Orr, W.C., et al., Effects of overexpression of copper-zinc and

manganese superoxide dismutases, catalase, and thioredoxin reductase

genes on longevity in Drosophila melanogaster. J Biol Chem, 2003. 278(29):

p. 26418-22.

46. Zhang, Y., et al., Mice deficient in both Mn superoxide dismutase and

glutathione peroxidase-1 have increased oxidative damage and a greater

incidence of pathology but no reduction in longevity. J Gerontol A Biol Sci

Med Sci, 2009. 64(12): p. 1212-20.

47. Van Raamsdonk, J.M. and S. Hekimi, Deletion of the mitochondrial

superoxide dismutase sod-2 extends lifespan in Caenorhabditis elegans.

PLoS Genet, 2009. 5(2): p. e1000361.

48. Doonan, R., et al., Against the oxidative damage theory of aging:

superoxide dismutases protect against oxidative stress but have little or no

64

effect on life span in Caenorhabditis elegans. Genes Dev, 2008. 22(23): p.

3236-41.

49. Mockett, R.J., B.H. Sohal, and R.S. Sohal, Expression of multiple

copies of mitochondrially targeted catalase or genomic Mn superoxide

dismutase transgenes does not extend the life span of Drosophila

melanogaster. Free Radic Biol Med, 2010. 49(12): p. 2028-31.

50. Trifunovic, A., et al., Premature ageing in mice expressing defective

mitochondrial DNA polymerase. Nature, 2004. 429(6990): p. 417-23.

51. Trifunovic, A., et al., Somatic mtDNA mutations cause aging

phenotypes without affecting reactive oxygen species production. Proc Natl

Acad Sci U S A, 2005. 102(50): p. 17993-8.

52. Khrapko, K. and J. Vijg, Mitochondrial DNA mutations and aging: a

case closed? Nat Genet, 2007. 39(4): p. 445-6.

53. Vermulst, M., et al., Mitochondrial point mutations do not limit the

natural lifespan of mice. Nat Genet, 2007. 39(4): p. 540-3.

54. Williams, S.L., et al., The mtDNA mutation spectrum of the progeroid

Polg mutator mouse includes abundant control region multimers. Cell

Metab, 2010. 12(6): p. 675-82.

55. Theurey, P. and P. Pizzo, The Aging Mitochondria. Genes (Basel),

2018. 9(1).

56. Tyynismaa, H., et al., Mutant mitochondrial helicase Twinkle causes

multiple mtDNA deletions and a late-onset mitochondrial disease in mice.

Proc Natl Acad Sci U S A, 2005. 102(49): p. 17687-92.

57. ORGEL, L.E., The maintenance of the accuracy of protein synthesis

and its relevance to ageing. Proc Natl Acad Sci U S A, 1963. 49: p. 517-21.

58. Hipkiss, A.R., Errors, mitochondrial dysfunction and ageing.

Biogerontology, 2003. 4(6): p. 397-400.

59. Holbrook, M.A. and J.R. Menninger, Erythromycin slows aging of

Saccharomyces cerevisiae. J Gerontol A Biol Sci Med Sci, 2002. 57(1): p.

B29-36.

60. Ballesteros, M., et al., Bacterial senescence: protein oxidation in non-

proliferating cells is dictated by the accuracy of the ribosomes. EMBO J,

2001. 20(18): p. 5280-9.

61. Dukan, S., et al., Protein oxidation in response to increased

transcriptional or translational errors. Proc Natl Acad Sci U S A, 2000.

97(11): p. 5746-9.

62. Larsson, N.G., Somatic mitochondrial DNA mutations in mammalian

aging. Annu Rev Biochem, 2010. 79: p. 683-706.

63. Rodnina, M.V., Translation in Prokaryotes. Cold Spring Harb Perspect

Biol, 2018. 10(9).

64. Grosjean, H., D.G. Söll, and D.M. Crothers, Studies of the complex

between transfer RNAs with complementary anticodons. I. Origins of

enhanced affinity between complementary triplets. J Mol Biol, 1976. 103(3):

p. 499-519.

65

65. Grosjean, H.J., S. de Henau, and D.M. Crothers, On the physical basis

for ambiguity in genetic coding interactions. Proc Natl Acad Sci U S A,

1978. 75(2): p. 610-4.

66. Ninio, J., Kinetic amplification of enzyme discrimination. Biochimie,

1975. 57(5): p. 587-95.

67. Hopfield, J.J., Kinetic proofreading: a new mechanism for reducing

errors in biosynthetic processes requiring high specificity. Proc Natl Acad

Sci U S A, 1974. 71(10): p. 4135-9.

68. Rodnina, M.V. and W. Wintermeyer, Fidelity of aminoacyl-tRNA

selection on the ribosome: kinetic and structural mechanisms. Annu Rev

Biochem, 2001. 70: p. 415-35.

69. Ramakrishnan, V., Ribosome structure and the mechanism of

translation. Cell, 2002. 108(4): p. 557-72.

70. Zaher, H.S. and R. Green, Hyperaccurate and error-prone ribosomes

exploit distinct mechanisms during tRNA selection. Mol Cell, 2010. 39(1): p.

110-20.

71. Amunts, A., et al., Ribosome. The structure of the human

mitochondrial ribosome. Science, 2015. 348(6230): p. 95-98.

72. Alksne, L.E., et al., An accuracy center in the ribosome conserved over

2 billion years. Proc Natl Acad Sci U S A, 1993. 90(20): p. 9538-41.

73. Noller, H.F., Biochemical characterization of the ribosomal decoding

site. Biochimie, 2006. 88(8): p. 935-41.

74. Ogle, J.M., et al., Selection of tRNA by the ribosome requires a

transition from an open to a closed form. Cell, 2002. 111(5): p. 721-32.

75. Ogle, J.M., A.P. Carter, and V. Ramakrishnan, Insights into the

decoding mechanism from recent ribosome structures. Trends Biochem Sci,

2003. 28(5): p. 259-66.

76. Demeshkina, N., et al., A new understanding of the decoding principle

on the ribosome. Nature, 2012. 484(7393): p. 256-9.

77. Jenner, L., et al., Structural rearrangements of the ribosome at the

tRNA proofreading step. Nat Struct Mol Biol, 2010. 17(9): p. 1072-8.

78. Rozov, A., et al., New Structural Insights into Translational

Miscoding. Trends Biochem Sci, 2016. 41(9): p. 798-814.

79. Carter, A.P., et al., Functional insights from the structure of the 30S

ribosomal subunit and its interactions with antibiotics. Nature, 2000.

407(6802): p. 340-8.

80. Pape, T., W. Wintermeyer, and M.V. Rodnina, Conformational switch

in the decoding region of 16S rRNA during aminoacyl-tRNA selection on the

ribosome. Nat Struct Biol, 2000. 7(2): p. 104-7.

81. Rosset, R. and L. Gorini, A ribosomal ambiguity mutation. J Mol Biol,

1969. 39(1): p. 95-112.

82. Agarwal, D., et al., Modulation of decoding fidelity by ribosomal

proteins S4 and S5. J Bacteriol, 2015. 197(6): p. 1017-25.

66

83. Vallabhaneni, H. and P.J. Farabaugh, Accuracy modulating mutations

of the ribosomal protein S4-S5 interface do not necessarily destabilize the

rps4-rps5 protein-protein interaction. RNA, 2009. 15(6): p. 1100-9.

84. Roy-Chaudhuri, B., N. Kirthi, and G.M. Culver, Appropriate

maturation and folding of 16S rRNA during 30S subunit biogenesis are

critical for translational fidelity. Proc Natl Acad Sci U S A, 2010. 107(10):

p. 4567-72.

85. Gregory, S.T., J.H. Cate, and A.E. Dahlberg, Streptomycin-resistant

and streptomycin-dependent mutants of the extreme thermophile Thermus

thermophilus. J Mol Biol, 2001. 309(2): p. 333-8.

86. Ozaki, M., S. Mizushima, and M. Nomura, Identification and

functional characterization of the protein controlled by the streptomycin-

resistant locus in E. coli. Nature, 1969. 222(5191): p. 333-9.

87. GORINI, L. and E. KATAJA, STREPTOMYCIN-INDUCED

OVERSUPPRESSION IN E. COLI. Proc Natl Acad Sci U S A, 1964. 51: p.

995-1001.

88. Sharma, D., et al., Mutational analysis of S12 protein and implications

for the accuracy of decoding by the ribosome. J Mol Biol, 2007. 374(4): p.

1065-76.

89. Agarwal, D., S.T. Gregory, and M. O'Connor, Error-prone and error-

restrictive mutations affecting ribosomal protein S12. J Mol Biol, 2011.

410(1): p. 1-9.

90. Kehrein, K., N. Bonnefoy, and M. Ott, Mitochondrial protein

synthesis: efficiency and accuracy. Antioxid Redox Signal, 2013. 19(16): p.

1928-39.

91. Amunts, A., et al., Structure of the yeast mitochondrial large

ribosomal subunit. Science, 2014. 343(6178): p. 1485-1489.

92. Ott, M., A. Amunts, and A. Brown, Organization and Regulation of

Mitochondrial Protein Synthesis. Annu Rev Biochem, 2016. 85: p. 77-101.

93. Zhang, L., et al., Antibiotic susceptibility of mammalian mitochondrial

translation. FEBS Lett, 2005. 579(28): p. 6423-7.

94. Fox, T.D., Mitochondrial protein synthesis, import, and assembly.

Genetics, 2012. 192(4): p. 1203-34.

95. Herrmann, J.M., M.W. Woellhaf, and N. Bonnefoy, Control of protein

synthesis in yeast mitochondria: the concept of translational activators.

Biochim Biophys Acta, 2013. 1833(2): p. 286-94.

96. Weraarpachai, W., et al., Mutation in TACO1, encoding a translational

activator of COX I, results in cytochrome c oxidase deficiency and late-onset

Leigh syndrome. Nat Genet, 2009. 41(7): p. 833-7.

97. Priesnitz, C. and T. Becker, Pathways to balance mitochondrial

translation and protein import. Genes Dev, 2018. 32(19-20): p. 1285-1296.

98. Sasarman, F., H. Antonicka, and E.A. Shoubridge, The A3243G

tRNALeu(UUR) MELAS mutation causes amino acid misincorporation and a

combined respiratory chain assembly defect partially suppressed by

67

overexpression of EFTu and EFG2. Hum Mol Genet, 2008. 17(23): p. 3697-

707.

99. Hobbie, S.N., et al., Mitochondrial deafness alleles confer misreading

of the genetic code. Proc Natl Acad Sci U S A, 2008. 105(9): p. 3244-9.

100. Akbergenov, R., et al., Mutant MRPS5 affects mitoribosomal accuracy

and confers stress-related behavioral alterations. EMBO Rep, 2018. 19(11).

101. Boguta, M., et al., NAM9 nuclear suppressor of mitochondrial ochre

mutations in Saccharomyces cerevisiae codes for a protein homologous to

S4 ribosomal proteins from chloroplasts, bacteria, and eucaryotes. Mol Cell

Biol, 1992. 12(1): p. 402-12.

102. Westermann, B., J.M. Herrmann, and W. Neupert, Analysis of

mitochondrial translation products in vivo and in organello in yeast.

Methods Cell Biol, 2001. 65: p. 429-38.

103. Fox, T.D., et al., Analysis and manipulation of yeast mitochondrial

genes. Methods Enzymol, 1991. 194: p. 149-65.

104. Langer, T., et al., Proteolytic breakdown of membrane-associated

polypeptides in mitochondria of Saccharomyces cerevisiae. Methods

Enzymol, 1995. 260: p. 495-503.

105. Beugnet, A., et al., Regulation of targets of mTOR (mammalian target

of rapamycin) signalling by intracellular amino acid availability. Biochem J,

2003. 372(Pt 2): p. 555-66.

106. Binda, M., et al., The Vam6 GEF controls TORC1 by activating the

EGO complex. Mol Cell, 2009. 35(5): p. 563-73.

107. Urban, J., et al., Sch9 is a major target of TORC1 in Saccharomyces

cerevisiae. Mol Cell, 2007. 26(5): p. 663-74.

108. Loewith, R. and M.N. Hall, Target of rapamycin (TOR) in nutrient

signaling and growth control. Genetics, 2011. 189(4): p. 1177-201.

109. Wullschleger, S., R. Loewith, and M.N. Hall, TOR signaling in growth

and metabolism. Cell, 2006. 124(3): p. 471-84.

110. Suhm, T. and M. Ott, Mitochondrial translation and cellular stress

response. Cell Tissue Res, 2017. 367(1): p. 21-31.

111. Couvillion, M.T., et al., Synchronized mitochondrial and cytosolic

translation programs. Nature, 2016. 533(7604): p. 499-503.

112. Dai, C.L., et al., Inhibition of protein synthesis alters protein

degradation through activation of protein kinase B (AKT). J Biol Chem,

2013. 288(33): p. 23875-83.

113. Cohen, J.S. and T.D. Fox, Expression of green fluorescent protein from

a recoded gene inserted into Saccharomyces cerevisiae mitochondrial DNA.

Mitochondrion, 2001. 1(2): p. 181-9.

114. Ott, M. and J.M. Herrmann, Co-translational membrane insertion of

mitochondrially encoded proteins. Biochim Biophys Acta, 2010. 1803(6): p.

767-75.

115. Ott, M., et al., Mba1, a membrane-associated ribosome receptor in

mitochondria. EMBO J, 2006. 25(8): p. 1603-10.

68

116. Preuss, M., et al., Mba1, a novel component of the mitochondrial

protein export machinery of the yeast Saccharomyces cerevisiae. J Cell Biol,

2001. 153(5): p. 1085-96.

117. Hell, K., W. Neupert, and R.A. Stuart, Oxa1p acts as a general

membrane insertion machinery for proteins encoded by mitochondrial DNA.

EMBO J, 2001. 20(6): p. 1281-8.

118. He, S. and T.D. Fox, Membrane translocation of mitochondrially

coded Cox2p: distinct requirements for export of N and C termini and

dependence on the conserved protein Oxa1p. Mol Biol Cell, 1997. 8(8): p.

1449-60.

119. Möller-Hergt, B.V., et al., The ribosome receptors Mrx15 and Mba1

jointly organize cotranslational insertion and protein biogenesis in

mitochondria. Mol Biol Cell, 2018. 29(20): p. 2386-2396.

120. Richter-Dennerlein, R., et al., Mitochondrial Protein Synthesis Adapts

to Influx of Nuclear-Encoded Protein. Cell, 2016. 167(2): p. 471-483.e10.

121. Aguado, A., et al., Chaperone-assisted protein aggregate reactivation:

Different solutions for the same problem. Arch Biochem Biophys, 2015.

580: p. 121-34.

122. Buchberger, A., B. Bukau, and T. Sommer, Protein quality control in

the cytosol and the endoplasmic reticulum: brothers in arms. Mol Cell,

2010. 40(2): p. 238-52.

123. Verghese, J., et al., Biology of the heat shock response and protein

chaperones: budding yeast (Saccharomyces cerevisiae) as a model system.

Microbiol Mol Biol Rev, 2012. 76(2): p. 115-58.

124. Amm, I., T. Sommer, and D.H. Wolf, Protein quality control and

elimination of protein waste: the role of the ubiquitin-proteasome system.

Biochim Biophys Acta, 2014. 1843(1): p. 182-96.

125. Chen, B., et al., Cellular strategies of protein quality control. Cold

Spring Harb Perspect Biol, 2011. 3(8): p. a004374.

126. Klaips, C.L., G.G. Jayaraj, and F.U. Hartl, Pathways of cellular

proteostasis in aging and disease. J Cell Biol, 2018. 217(1): p. 51-63.

127. Mogk, A., E. Kummer, and B. Bukau, Cooperation of Hsp70 and

Hsp100 chaperone machines in protein disaggregation. Front Mol Biosci,

2015. 2: p. 22.

128. Parsell, D.A., et al., Protein disaggregation mediated by heat-shock

protein Hsp104. Nature, 1994. 372(6505): p. 475-8.

129. Doyle, S.M. and S. Wickner, Hsp104 and ClpB: protein

disaggregating machines. Trends Biochem Sci, 2009. 34(1): p. 40-8.

130. Snider, J., G. Thibault, and W.A. Houry, The AAA+ superfamily of

functionally diverse proteins. Genome Biol, 2008. 9(4): p. 216.

131. Glover, J.R. and S. Lindquist, Hsp104, Hsp70, and Hsp40: a novel

chaperone system that rescues previously aggregated proteins. Cell, 1998.

94(1): p. 73-82.

69

132. Winkler, J., et al., Chaperone networks in protein disaggregation and

prion propagation. J Struct Biol, 2012. 179(2): p. 152-60.

133. Lee, J., et al., Heat shock protein (Hsp) 70 is an activator of the

Hsp104 motor. Proc Natl Acad Sci U S A, 2013. 110(21): p. 8513-8.

134. Shorter, J., The mammalian disaggregase machinery: Hsp110

synergizes with Hsp70 and Hsp40 to catalyze protein disaggregation and

reactivation in a cell-free system. PLoS One, 2011. 6(10): p. e26319.

135. Rampelt, H., et al., Metazoan Hsp70 machines use Hsp110 to power

protein disaggregation. EMBO J, 2012. 31(21): p. 4221-35.

136. Kaimal, J.M., et al., Coordinated Hsp110 and Hsp104 Activities Power

Protein Disaggregation in Saccharomyces cerevisiae. Mol Cell Biol, 2017.

37(11).

137. Finley, D., et al., The ubiquitin-proteasome system of Saccharomyces

cerevisiae. Genetics, 2012. 192(2): p. 319-60.

138. Kleiger, G. and T. Mayor, Perilous journey: a tour of the ubiquitin-

proteasome system. Trends Cell Biol, 2014. 24(6): p. 352-9.

139. Hill, S.M., S. Hanzén, and T. Nyström, Restricted access: spatial

sequestration of damaged proteins during stress and aging. EMBO Rep,

2017. 18(3): p. 377-391.

140. Miller, S.B., A. Mogk, and B. Bukau, Spatially organized aggregation

of misfolded proteins as cellular stress defense strategy. J Mol Biol, 2015.

427(7): p. 1564-74.

141. Miller, S.B., et al., Compartment-specific aggregases direct distinct

nuclear and cytoplasmic aggregate deposition. EMBO J, 2015. 34(6): p.

778-97.

142. Kaganovich, D., R. Kopito, and J. Frydman, Misfolded proteins

partition between two distinct quality control compartments. Nature, 2008.

454(7208): p. 1088-95.

143. Specht, S., et al., Hsp42 is required for sequestration of protein

aggregates into deposition sites in Saccharomyces cerevisiae. J Cell Biol,

2011. 195(4): p. 617-29.

144. Kopito, R.R., Aggresomes, inclusion bodies and protein aggregation.

Trends Cell Biol, 2000. 10(12): p. 524-30.

145. Zheng, X., et al., Dynamic control of Hsf1 during heat shock by a

chaperone switch and phosphorylation. Elife, 2016. 5.

146. Zou, J., et al., Repression of heat shock transcription factor HSF1

activation by HSP90 (HSP90 complex) that forms a stress-sensitive complex

with HSF1. Cell, 1998. 94(4): p. 471-80.

147. Akerfelt, M., R.I. Morimoto, and L. Sistonen, Heat shock factors:

integrators of cell stress, development and lifespan. Nat Rev Mol Cell Biol,

2010. 11(8): p. 545-55.

148. Anckar, J. and L. Sistonen, Regulation of HSF1 function in the heat

stress response: implications in aging and disease. Annu Rev Biochem,

2011. 80: p. 1089-115.

70

149. Martínez-Pastor, M.T., et al., The Saccharomyces cerevisiae zinc

finger proteins Msn2p and Msn4p are required for transcriptional induction

through the stress response element (STRE). EMBO J, 1996. 15(9): p. 2227-

35.

150. Schmitt, A.P. and K. McEntee, Msn2p, a zinc finger DNA-binding

protein, is the transcriptional activator of the multistress response in

Saccharomyces cerevisiae. Proc Natl Acad Sci U S A, 1996. 93(12): p.

5777-82.

151. Beck, T. and M.N. Hall, The TOR signalling pathway controls nuclear

localization of nutrient-regulated transcription factors. Nature, 1999.

402(6762): p. 689-92.

152. Görner, W., et al., Nuclear localization of the C2H2 zinc finger protein

Msn2p is regulated by stress and protein kinase A activity. Genes Dev, 1998.

12(4): p. 586-97.

153. Estruch, F., Stress-controlled transcription factors, stress-induced

genes and stress tolerance in budding yeast. FEMS Microbiol Rev, 2000.

24(4): p. 469-86.

154. Grably, M.R., et al., HSF and Msn2/4p can exclusively or

cooperatively activate the yeast HSP104 gene. Mol Microbiol, 2002. 44(1):

p. 21-35.

155. Hall, A., P.A. Karplus, and L.B. Poole, Typical 2-Cys peroxiredoxins--

structures, mechanisms and functions. FEBS J, 2009. 276(9): p. 2469-77.

156. Perkins, A., et al., Peroxiredoxins: guardians against oxidative stress

and modulators of peroxide signaling. Trends Biochem Sci, 2015. 40(8): p.

435-45.

157. Nyström, T., J. Yang, and M. Molin, Peroxiredoxins, gerontogenes

linking aging to genome instability and cancer. Genes Dev, 2012. 26(18): p.

2001-8.

158. Suhm, T., et al., Mitochondrial Translation Efficiency Controls

Cytoplasmic Protein Homeostasis. Cell Metab, 2018. 27(6): p. 1309-

1322.e6.

159. Molin, M., et al., Life span extension and H(2)O(2) resistance elicited

by caloric restriction require the peroxiredoxin Tsa1 in Saccharomyces

cerevisiae. Mol Cell, 2011. 43(5): p. 823-33.

160. Musicco, C., et al., Accumulation of overoxidized Peroxiredoxin III in

aged rat liver mitochondria. Biochim Biophys Acta, 2009. 1787(7): p. 890-

6.

161. Hanzén, S., et al., Lifespan Control by Redox-Dependent Recruitment

of Chaperones to Misfolded Proteins. Cell, 2016. 166(1): p. 140-51.

162. Labbadia, J. and R.I. Morimoto, The biology of proteostasis in aging

and disease. Annu Rev Biochem, 2015. 84: p. 435-64.

163. Dillin, A. and E. Cohen, Ageing and protein aggregation-mediated

disorders: from invertebrates to mammals. Philos Trans R Soc Lond B Biol

Sci, 2011. 366(1561): p. 94-8.

71

164. Erjavec, N., et al., Accelerated aging and failure to segregate damaged

proteins in Sir2 mutants can be suppressed by overproducing the protein

aggregation-remodeling factor Hsp104p. Genes Dev, 2007. 21(19): p. 2410-

21.

165. Kruegel, U., et al., Elevated proteasome capacity extends replicative

lifespan in Saccharomyces cerevisiae. PLoS Genet, 2011. 7(9): p. e1002253.

166. Voos, W., Mitochondrial protein homeostasis: the cooperative roles of

chaperones and proteases. Res Microbiol, 2009. 160(9): p. 718-25.

167. Voos, W., Chaperone-protease networks in mitochondrial protein

homeostasis. Biochim Biophys Acta, 2013. 1833(2): p. 388-99.

168. Baker, M.J., T. Tatsuta, and T. Langer, Quality control of

mitochondrial proteostasis. Cold Spring Harb Perspect Biol, 2011. 3(7).

169. Kang, P.J., et al., Requirement for hsp70 in the mitochondrial matrix

for translocation and folding of precursor proteins. Nature, 1990.

348(6297): p. 137-43.

170. Rassow, J. and N. Pfanner, Molecular chaperones and intracellular

protein translocation. Rev Physiol Biochem Pharmacol, 1995. 126: p. 199-

264.

171. Herrmann, J.M., et al., Mitochondrial heat shock protein 70, a

molecular chaperone for proteins encoded by mitochondrial DNA. J Cell

Biol, 1994. 127(4): p. 893-902.

172. Rassow, J., W. Voos, and N. Pfanner, Partner proteins determine

multiple functions of Hsp70. Trends Cell Biol, 1995. 5(5): p. 207-12.

173. Miao, B., J.E. Davis, and E.A. Craig, Mge1 functions as a nucleotide

release factor for Ssc1, a mitochondrial Hsp70 of Saccharomyces cerevisiae.

J Mol Biol, 1997. 265(5): p. 541-52.

174. Kubo, Y., et al., Two distinct mechanisms operate in the reactivation of

heat-denatured proteins by the mitochondrial Hsp70/Mdj1p/Yge1p

chaperone system. J Mol Biol, 1999. 286(2): p. 447-64.

175. Horwich, A.L., E.U. Weber-Ban, and D. Finley, Chaperone rings in

protein folding and degradation. Proc Natl Acad Sci U S A, 1999. 96(20): p.

11033-40.

176. Dubaquié, Y., et al., Identification of in vivo substrates of the yeast

mitochondrial chaperonins reveals overlapping but non-identical

requirement for hsp60 and hsp10. EMBO J, 1998. 17(20): p. 5868-76.

177. Nargund, A.M., et al., Mitochondrial import efficiency of ATFS-1

regulates mitochondrial UPR activation. Science, 2012. 337(6094): p. 587-

90.

178. Schmitt, M., W. Neupert, and T. Langer, The molecular chaperone

Hsp78 confers compartment-specific thermotolerance to mitochondria. J

Cell Biol, 1996. 134(6): p. 1375-86.

179. Lewandowska, A., et al., Hsp78 chaperone functions in restoration of

mitochondrial network following heat stress. Biochim Biophys Acta, 2006.

1763(2): p. 141-51.

72

180. Moczko, M., et al., The mitochondrial ClpB homolog Hsp78

cooperates with matrix Hsp70 in maintenance of mitochondrial function. J

Mol Biol, 1995. 254(4): p. 538-43.

181. Koppen, M. and T. Langer, Protein degradation within mitochondria:

versatile activities of AAA proteases and other peptidases. Crit Rev Biochem

Mol Biol, 2007. 42(3): p. 221-42.

182. Glynn, S.E., Multifunctional Mitochondrial AAA Proteases. Front Mol

Biosci, 2017. 4: p. 34.

183. Bota, D.A. and K.J. Davies, Lon protease preferentially degrades

oxidized mitochondrial aconitase by an ATP-stimulated mechanism. Nat Cell

Biol, 2002. 4(9): p. 674-80.

184. van Dyck, L., et al., Mcx1p, a ClpX homologue in mitochondria of

Saccharomyces cerevisiae. FEBS Lett, 1998. 438(3): p. 250-4.

185. Lee, S., et al., Electron cryomicroscopy structure of a membrane-

anchored mitochondrial AAA protease. J Biol Chem, 2011. 286(6): p. 4404-

11.

186. Leonhard, K., et al., Membrane protein degradation by AAA proteases

in mitochondria: extraction of substrates from either membrane surface.

Mol Cell, 2000. 5(4): p. 629-38.

187. Arlt, H., et al., The YTA10-12 complex, an AAA protease with

chaperone-like activity in the inner membrane of mitochondria. Cell, 1996.

85(6): p. 875-85.

188. Arlt, H., et al., The formation of respiratory chain complexes in

mitochondria is under the proteolytic control of the m-AAA protease. EMBO

J, 1998. 17(16): p. 4837-47.

189. Augustin, S., et al., Characterization of peptides released from

mitochondria: evidence for constant proteolysis and peptide efflux. J Biol

Chem, 2005. 280(4): p. 2691-9.

190. Gerdes, F., T. Tatsuta, and T. Langer, Mitochondrial AAA proteases--

towards a molecular understanding of membrane-bound proteolytic

machines. Biochim Biophys Acta, 2012. 1823(1): p. 49-55.

191. Merkwirth, C. and T. Langer, Prohibitin function within mitochondria:

essential roles for cell proliferation and cristae morphogenesis. Biochim

Biophys Acta, 2009. 1793(1): p. 27-32.

192. Steglich, G., W. Neupert, and T. Langer, Prohibitins regulate

membrane protein degradation by the m-AAA protease in mitochondria. Mol

Cell Biol, 1999. 19(5): p. 3435-42.

193. Nijtmans, L.G., et al., Prohibitins act as a membrane-bound chaperone

for the stabilization of mitochondrial proteins. EMBO J, 2000. 19(11): p.

2444-51.

194. Leonhard, K., et al., AAA proteases with catalytic sites on opposite

membrane surfaces comprise a proteolytic system for the ATP-dependent

degradation of inner membrane proteins in mitochondria. EMBO J, 1996.

15(16): p. 4218-29.

73

195. Nakai, T., et al., Multiple genes, including a member of the AAA

family, are essential for degradation of unassembled subunit 2 of

cytochrome c oxidase in yeast mitochondria. Mol Cell Biol, 1995. 15(8): p.

4441-52.

196. Pearce, D.A. and F. Sherman, Degradation of cytochrome oxidase

subunits in mutants of yeast lacking cytochrome c and suppression of the

degradation by mutation of yme1. J Biol Chem, 1995. 270(36): p. 20879-82.

197. Dunn, C.D., et al., A genomewide screen for petite-negative yeast

strains yields a new subunit of the i-AAA protease complex. Mol Biol Cell,

2006. 17(1): p. 213-26.

198. Dunn, C.D., et al., Mgr3p and Mgr1p are adaptors for the

mitochondrial i-AAA protease complex. Mol Biol Cell, 2008. 19(12): p.

5387-97.

199. Bruderek, M., et al., IMiQ: a novel protein quality control

compartment protecting mitochondrial functional integrity. Mol Biol Cell,

2018. 29(3): p. 256-269.

200. Guaragnella, N., et al., Mitochondria-cytosol-nucleus crosstalk:

learning from Saccharomyces cerevisiae. FEMS Yeast Res, 2018. 18(8).

201. Leadsham, J.E. and C.W. Gourlay, cAMP/PKA signaling balances

respiratory activity with mitochondria dependent apoptosis via

transcriptional regulation. BMC Cell Biol, 2010. 11: p. 92.

202. Liu, Z. and R.A. Butow, Mitochondrial retrograde signaling. Annu

Rev Genet, 2006. 40: p. 159-85.

203. Rothermel, B.A., J.L. Thornton, and R.A. Butow, Rtg3p, a basic helix-

loop-helix/leucine zipper protein that functions in mitochondrial-induced

changes in gene expression, contains independent activation domains. J Biol

Chem, 1997. 272(32): p. 19801-7.

204. Jia, Y., et al., A basic helix-loop-helix-leucine zipper transcription

complex in yeast functions in a signaling pathway from mitochondria to the

nucleus. Mol Cell Biol, 1997. 17(3): p. 1110-7.

205. Sekito, T., J. Thornton, and R.A. Butow, Mitochondria-to-nuclear

signaling is regulated by the subcellular localization of the transcription

factors Rtg1p and Rtg3p. Mol Biol Cell, 2000. 11(6): p. 2103-15.

206. Liu, Z., et al., Retrograde signaling is regulated by the dynamic

interaction between Rtg2p and Mks1p. Mol Cell, 2003. 12(2): p. 401-11.

207. Sekito, T., et al., RTG-dependent mitochondria-to-nucleus signaling is

regulated by MKS1 and is linked to formation of yeast prion [URE3]. Mol

Biol Cell, 2002. 13(3): p. 795-804.

208. Epstein, C.B., et al., Genome-wide responses to mitochondrial

dysfunction. Mol Biol Cell, 2001. 12(2): p. 297-308.

209. Dilova, I., C.Y. Chen, and T. Powers, Mks1 in concert with TOR

signaling negatively regulates RTG target gene expression in S. cerevisiae.

Curr Biol, 2002. 12(5): p. 389-95.

74

210. Dilova, I., et al., Tor signaling and nutrient-based signals converge on

Mks1p phosphorylation to regulate expression of Rtg1.Rtg3p-dependent

target genes. J Biol Chem, 2004. 279(45): p. 46527-35.

211. Zhang, F., et al., Adenosine Triphosphate (ATP) Is a Candidate

Signaling Molecule in the Mitochondria-to-Nucleus Retrograde Response

Pathway. Genes (Basel), 2013. 4(1): p. 86-100.

212. Miceli, M.V., et al., Loss of mitochondrial membrane potential

triggers the retrograde response extending yeast replicative lifespan. Front

Genet, 2011. 2: p. 102.

213. Kirchman, P.A., et al., Interorganelle signaling is a determinant of

longevity in Saccharomyces cerevisiae. Genetics, 1999. 152(1): p. 179-90.

214. Topf, U., L. Wrobel, and A. Chacinska, Chatty Mitochondria: Keeping

Balance in Cellular Protein Homeostasis. Trends Cell Biol, 2016. 26(8): p.

577-586.

215. Martinus, R.D., et al., Selective induction of mitochondrial chaperones

in response to loss of the mitochondrial genome. Eur J Biochem, 1996.

240(1): p. 98-103.

216. Qureshi, M.A., C.M. Haynes, and M.W. Pellegrino, The mitochondrial

unfolded protein response: Signaling from the powerhouse. J Biol Chem,

2017. 292(33): p. 13500-13506.

217. Haynes, C.M., et al., ClpP mediates activation of a mitochondrial

unfolded protein response in C. elegans. Dev Cell, 2007. 13(4): p. 467-80.

218. Haynes, C.M., et al., The matrix peptide exporter HAF-1 signals a

mitochondrial UPR by activating the transcription factor ZC376.7 in C.

elegans. Mol Cell, 2010. 37(4): p. 529-40.

219. Arnold, I., et al., Evidence for a novel mitochondria-to-nucleus

signalling pathway in respiring cells lacking i-AAA protease and the ABC-

transporter Mdl1. Gene, 2006. 367: p. 74-88.

220. Baker, B.M., et al., Protective coupling of mitochondrial function and

protein synthesis via the eIF2α kinase GCN-2. PLoS Genet, 2012. 8(6): p.

e1002760.

221. Sonenberg, N. and A.G. Hinnebusch, Regulation of translation

initiation in eukaryotes: mechanisms and biological targets. Cell, 2009.

136(4): p. 731-45.

222. Zhao, Q., et al., A mitochondrial specific stress response in

mammalian cells. EMBO J, 2002. 21(17): p. 4411-9.

223. Horibe, T. and N.J. Hoogenraad, The chop gene contains an element

for the positive regulation of the mitochondrial unfolded protein response.

PLoS One, 2007. 2(9): p. e835.

224. Al-Furoukh, N., et al., ClpX stimulates the mitochondrial unfolded

protein response (UPRmt) in mammalian cells. Biochim Biophys Acta,

2015. 1853(10 Pt A): p. 2580-91.

75

225. Aldridge, J.E., T. Horibe, and N.J. Hoogenraad, Discovery of genes

activated by the mitochondrial unfolded protein response (mtUPR) and

cognate promoter elements. PLoS One, 2007. 2(9): p. e874.

226. Rath, E., et al., Induction of dsRNA-activated protein kinase links

mitochondrial unfolded protein response to the pathogenesis of intestinal

inflammation. Gut, 2012. 61(9): p. 1269-1278.

227. Fiorese, C.J., et al., The Transcription Factor ATF5 Mediates a

Mammalian Mitochondrial UPR. Curr Biol, 2016. 26(15): p. 2037-2043.

228. Cardamone, M.D., et al., Mitochondrial Retrograde Signaling in

Mammals Is Mediated by the Transcriptional Cofactor GPS2 via Direct

Mitochondria-to-Nucleus Translocation. Mol Cell, 2018. 69(5): p. 757-

772.e7.

229. Yoneda, T., et al., Compartment-specific perturbation of protein

handling activates genes encoding mitochondrial chaperones. J Cell Sci,

2004. 117(Pt 18): p. 4055-66.

230. Caballero, A., et al., Absence of mitochondrial translation control

proteins extends life span by activating sirtuin-dependent silencing. Mol

Cell, 2011. 42(3): p. 390-400.

231. Delaney, J.R., et al., Stress profiling of longevity mutants identifies

Afg3 as a mitochondrial determinant of cytoplasmic mRNA translation and

aging. Aging Cell, 2013. 12(1): p. 156-66.

232. Heeren, G., et al., The mitochondrial ribosomal protein of the large

subunit, Afo1p, determines cellular longevity through mitochondrial back-

signaling via TOR1. Aging (Albany NY), 2009. 1(7): p. 622-36.

233. Nolden, M., et al., The m-AAA protease defective in hereditary spastic

paraplegia controls ribosome assembly in mitochondria. Cell, 2005. 123(2):

p. 277-89.

234. Bennett, C.F., et al., Activation of the mitochondrial unfolded protein

response does not predict longevity in Caenorhabditis elegans. Nat

Commun, 2014. 5: p. 3483.

235. Houtkooper, R.H., et al., Mitonuclear protein imbalance as a

conserved longevity mechanism. Nature, 2013. 497(7450): p. 451-7.

236. Wang, X. and X.J. Chen, A cytosolic network suppressing

mitochondria-mediated proteostatic stress and cell death. Nature, 2015.

524(7566): p. 481-4.

237. Wrobel, L., et al., Mistargeted mitochondrial proteins activate a

proteostatic response in the cytosol. Nature, 2015. 524(7566): p. 485-8.

238. Weidberg, H. and A. Amon, MitoCPR-A surveillance pathway that

protects mitochondria in response to protein import stress. Science, 2018.

360(6385).

239. D'Amico, D., V. Sorrentino, and J. Auwerx, Cytosolic Proteostasis

Networks of the Mitochondrial Stress Response. Trends Biochem Sci, 2017.

42(9): p. 712-725.

76

240. Coyne, L.P. and X.J. Chen, mPOS is a novel mitochondrial trigger of

cell death - implications for neurodegeneration. FEBS Lett, 2018. 592(5): p.

759-775.

241. Papa, L. and D. Germain, Estrogen receptor mediates a distinct

mitochondrial unfolded protein response. J Cell Sci, 2011. 124(Pt 9): p.

1396-402.

242. Vilchez, D., et al., Increased proteasome activity in human embryonic

stem cells is regulated by PSMD11. Nature, 2012. 489(7415): p. 304-8.

243. Vilchez, D., et al., RPN-6 determines C. elegans longevity under

proteotoxic stress conditions. Nature, 2012. 489(7415): p. 263-8.

244. Livnat-Levanon, N., et al., Reversible 26S proteasome disassembly

upon mitochondrial stress. Cell Rep, 2014. 7(5): p. 1371-1380.

245. Segref, A., et al., Pathogenesis of human mitochondrial diseases is

modulated by reduced activity of the ubiquitin/proteasome system. Cell

Metab, 2014. 19(4): p. 642-52.

246. Ruan, L., et al., Cytosolic proteostasis through importing of misfolded

proteins into mitochondria. Nature, 2017. 543(7645): p. 443-446.

247. Liu, X., et al., Induction of apoptotic program in cell-free extracts:

requirement for dATP and cytochrome c. Cell, 1996. 86(1): p. 147-57.

248. Kramer, E.B. and P.J. Farabaugh, The frequency of translational

misreading errors in E. coli is largely determined by tRNA competition.

RNA, 2007. 13(1): p. 87-96.

249. Grentzmann, G., et al., A dual-luciferase reporter system for studying

recoding signals. RNA, 1998. 4(4): p. 479-86.

250. Wohlgemuth, I., et al., Evolutionary optimization of speed and

accuracy of decoding on the ribosome. Philos Trans R Soc Lond B Biol Sci,

2011. 366(1580): p. 2979-86.

251. Kehrein, K., et al., Organization of Mitochondrial Gene Expression in

Two Distinct Ribosome-Containing Assemblies. Cell Rep, 2015.

252. van Bloois, E., et al., Detection of cross-links between FtsH, YidC,

HflK/C suggests a linked role for these proteins in quality control upon

insertion of bacterial inner membrane proteins. FEBS Lett, 2008. 582(10):

p. 1419-24.