The Role of Intermembrane Space Redox Factors in Glutathione Metabolism and Intracellular Redox Equilibrium Hatice Kubra Ozer University of South Carolina

University of South Carolina Scholar Commons

Theses and Dissertations

2015 The Role of Intermembrane Space Redox Factors In Glutathione Metabolism And Intracellular Redox Equilibrium Hatice Kubra Ozer University of South carolina

Follow this and additional works at: https://scholarcommons.sc.edu/etd Part of the Chemistry Commons

Recommended Citation Ozer, H. K.(2015). The Role of Intermembrane Space Redox Factors In Glutathione Metabolism And Intracellular Redox Equilibrium. (Doctoral dissertation). Retrieved from https://scholarcommons.sc.edu/etd/3702

This Open Access Dissertation is brought to you by Scholar Commons. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected]. THE ROLE OF INTERMEMBRANE SPACE REDOX FACTORS IN GLUTATHIONE METABOLISM AND INTRACELLULAR REDOX EQUILIBRIUM

Hatice Kubra Ozer

Bachelor of Science Uludag University, 2004

Master of Food Science and Nutrition Clemson University, 2010

Submitted in Partial Fulfillment of the Requirements

For the Degree of Doctor of Philosophy in

Chemistry

College of Arts and Sciences

University of South Carolina

2015

Accepted by:

Caryn E. Outten, Major Professor

F. Wayne Outten, Committee Chair

Erin Connolly, Committee Member

Andrew B. Greytak, Committee Member

ACKNOWLEDGEMENTS

First, I would like to thank my advisor, Dr. Caryn E. Outten for her patience, guidance, expertise, and confidence in me to complete the work contained herein. She has been an excellent mentor during my graduate program. She is also the only person beside myself who is guaranteed to have read every word of this manuscript and to review every presentations in the past and her insight was irreplaceable.

I would like to thank my committee members: Dr. F. Wayne Outten, Dr. Erin

Connolly, and Dr. Qian Wang for their valuable input on my plan, and proposal, for helping me to fulfill the requirements for a Ph.D. in Chemistry. Also, a special thanks to

Dr. Andrew Greytak for joining my dissertation defense committee at the last minute to replace Dr. Wang.

Thanks to all the members of the Outten lab present and past (Dr. Vidyadhar

Daithankar, Dr. Angela-Nadia Albetel, Crystal Conaway, Adrienne Dlouhy, Maxwell

Darch, John S Hepburn, Malini Gupta, Kirsten R Collins, Naimah Bolaji, Matthew

Blahut, Rabindra Behera, Yuyuan Dai, Suning Wang, Haoran (Henry) Li, and Zuqin

Xie,) for their support, discussions, and advice. I owe to thanks to Henry and Rabi for all he taught me with huge patience and for helpful discussions. I would especially like to thank Vidy, Angela, Crystal, Naimah, Yuyuan, and Suning for their friendship and appreciate the contributions they have made toward my research.

iii

I would like to thank to my husband, Ufuk Ozer for his love, patience, great support, and for being there for me whenever I needed somebody to listen the complaints about my failures in research. My research would not have been possible without the support of him.

I would like to thank to my parents, Mevlut and Hacer Tokpunar, have loved and stood with me my entire life, and for encouraging me to pursue a career in Biochemistry.

Thanks to my mother in-law, Yeter Ozer for their love, support, taking good care of my children during my school years.

Finally, I would like to acknowledge the Ministry of Education, Turkey for awarding me the Graduate Study Abroad Scholarship in USA.

ABSTRACT

The mitochondrial intermembrane space (IMS) is a unique subcellular compartment that houses key thiol-dependent redox pathways such as protein transport, mitochondrial respiration, and detoxification of ROS (reactive oxygen species). These pathways are all dependent on cysteine-rich proteins, thus maintaining thiol-disulfide balance in this organelle is crucial for cellular functions. An IMS protein import pathway called the Mia40-Erv1 disulfide relay system uses disulfide bond formation for the import and retention of substrate proteins in the IMS. Erv1 is also suggested to be involved in maturation of cytosolic Fe-S cluster proteins and regulation of iron homeostasis in S. cerevisiae. However, these studies were performed on one particular erv1 mutant strain

(named as erv1-1) that we discovered has additional defects in glutathione (GSH) metabolism due to a secondary mutation in the gene encoding the GSH biosynthesis enzyme, Gsh1. Since the tripeptide GSH is also required for iron homeostasis and cytosolic Fe-S protein biogenesis, the Erv1-dependent connection between mitochondrial protein import, GSH metabolism, and iron homeostasis was investigated in several erv1 mutants. The GSH depletion phenotype was only detected in the erv1-1 strain and could be rescued by expressing GSH1 or adding GSH to the growth media. Additionally, expression of the iron uptake gene, FET3 and enzyme activities of Fe-S cluster proteins in several erv1 mutants were tested. Only the erv1-1 mutant has an iron misregulation defect, which could be rescued with GSH addition, and no significant effects on Fe-S cluster protein activities were detected. Our data suggests that the defects of cytosolic Fe-

S maturation and iron regulation first reported in the erv1-1 strain is a direct consequence of GSH depletion rather than indicating a direct role for Erv1 in iron metabolism and cytosolic Fe-S cluster biogenesis.

We also characterized how mutations of Mia40 influence GSH metabolism and

GSH:GSSG pools in the cytosol, mitochondrial matrix and intermembrane space. We have found that defects in Mia40 only influence the IMS redox state and do not alter cellular GSH levels. Additionally, we determined that defects in Mia40 do not impact iron homeostasis or Fe-S cluster biogenesis.

Furthermore, utilizing the roGFP2 in vivo sensors, we demonstrated how the deletion of manganese cofactor of superoxide dismutase 2 transporter Mtm1 and the citrate-oxoglutarate carrier Yhm2 affect the redox status of the mitochondrial matrix and

IMS and cellular GSH levels. Deletion of MTM1 only leads to a large oxidative shift in the IMS GSH:GSSG redox state. In the contrary, deletion of YHM2 shows a smaller effect on the mitochondrial GSH:GSSG redox state even though, both mtm1∆ and yhm2∆ mutants were shown to have reduced mitochondrial GSH levels. Overall, we successfully characterized the roles of Erv1 and Mia40 in GSH metabolism, mitochondrial import and subcellular redox state which hereby helps to reveal their roles in Fe-S cluster biogenesis and iron regulation.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... iii

ABSTRACT ...... V

LIST OF TABLES ...... iX

LIST OF FIGURES ...... X

LIST OF ABBREVIATIONS ...... Xiii

CHAPTER 1. INTRODUCTION ...... 1

CELLULAR DISTRIBUTION OF THIOL REDOX SYSTEMS IN SACCHAROMYCES CEREVISIAE ...... 2 IN VIVO GENETICALLY ENCODED REDOX SENSORS ...... 8

GSH , ERV 1-MIA 40 IMS DISULFIDE RELAY SYSTEM AND OTHER REDOX CONTROL SYSTEMS IN THE MITOCHONDRIAL IMS ...... 9 FE-S BIOGENESIS , IRON HOMEOSTASIS AND THEIR LINK WITH ERV 1 AND GSH ...... 15

SCOPE OF THESIS ...... 20

CHAPTER 2. THE ROLE OF ERV 1 IN GLUTATHIONE METABOLISM , CYTOSOLIC FE-S CLUSTER MATURATION AND IRON REGULATION ...... 24 ABSTRACT ...... 25

INTRODUCTION ...... 26

EXPERIMENTAL PROCEDURES ...... 29

RESULTS ...... 37

DISCUSSION ...... 60

CHAPTER 3. MUTATIONS IN MIA 40 AFFECTS THE IMS REDOX STATE AND HAVE NO EFFECTS ON IRON REGULATION ...... 65 ABSTRACT ...... 66

INTRODUCTION ...... 67

EXPERIMENTAL PROCEDURES ...... 71

vii

RESULTS ...... 74

DISCUSSION ...... 83

CHAPTER 4. DELETION OF MITOCHONDRIAL CARRIER PROTEINS , MTM 1 AND YHM 2 AFFECTS REDOX STATUS OF MITOCHONDRIA ...... 87 ABSTRACT ...... 88

INTRODUCTION ...... 89

EXPERIMENTAL PROCEDURES ...... 92

RESULTS ...... 93

DISCUSSION ...... 99

CHAPTER 5. SUPPLEMENTARY METHODS ...... 104

INTRODUCTION ...... 105

PREPARE TEMPERATURE -SENSITIVE MUTANT STRAINS FOR WHOLE CELL GSH MEASUREMENTS BY USING THE GLO -KIT (PROMEGA GSH /GSSG -GLO ™ ASSAY ) ....105 BLOCKING CONDITIONS FOR GSH 1 AND ERV 1 PROTEINS ...... 107

RT-PCR PROTOCOL ...... 107

REFERENCES ...... 116

viii

LIST OF TABLES

Table 2.1 Strains used in this study ...... 35

Table 2.2 Primers used for sequencing and qRT-PCR ...... 36

LIST OF FIGURES

Figure 1.1 GSH-Glutaredoxin and Thioredoxin systems in the cytosol ...... 4

Figure 1.2 Erv1-Mia40 import pathway and glutaredoxins redox systems in the IMS ....11

Figure 1.3 Erv1 C-Terminal Core Domain Structure which is from Asp 86 to Asp 188 ...... 13

Figure 1.4 Putative Roles for Erv1, GSH, and Atm1 in Cytosolic Fe-S Cluster Biogenesis and Iron Regulation...... 19

Figure 2.1 In vivo Mia40 redox state measurements in erv1 mutants ...... 38

Figure 2.2 Total GSH values in cytosolic, mitochondrial and whole cell extracts for erv1 strains ...... 39

Figure 2.3 In vivo Mia40 redox state and whole cell GSH measurements in erv1-(F124S) strain ...... 42

Figure 2.4 In vivo Mia40 redox state, immunoblot for Erv1 protein and whole cell GSH measurements in GAL-ERV1 strain ...... 43

Figure 2.5 Mitochondrial Erv1 protein levels in erv1 strains ...... 44

Figure 2.6 Total GSH values in cytosolic and mitochondrial extracts of WT and erv1 ts mutant carrying F124S mutation with high mitochondrial Fe levels ...... 45

Figure 2.7 Subcellular GSH:GSSG redox state in erv1 mutants ...... 47

Figure 2.8 erv1-1 strain has dramatically decreased Gsh1 protein levels ...... 51

Figure 2.9 Mutation found in GSH1 gene of erv1-1 strain likely compromises Gsh1 protein folding and/or stability...... 52

Figure 2.10 Real time RT-PCR analysis of mRNA levels of WT and erv1-1 strains ...... 53

Figure 2.11 Expression of the Aft1/2-regulated gene FET3 ...... 54

Figure 2.12 Human cytosolic aconitase activity (IRP1) in WT and erv1 strains ...... 55

Figure 2.13 Confirmation assays for complementation of erv1-1 strain ...... 58

Figure 2.14 Spot testing of cell growth with wild-type (JRY675), and erv1-1 mutant cells expressing pRS415 (empty plasmid), ERV1 , and GSH1 ...... 59

Figure 3.1 Schematic diagram of Mia40 structure ...... 68

Figure 3.2 In vivo Mia40 redox state measurements in mia40 temperature sensitive mutants ...... 77

Figure 3.3 Mitochondrial Erv1 protein levels in mia40 strains ...... 78

Figure 3.4 Subcellular GSH:GSSG redox state in mia40 mutants ...... 79

Figure 3.5 Total GSH values in cytosolic, mitochondrial and whole cell extracts for mia40 strains ...... 80

Figure 3.6 Gsh1 protein levels in mia40 mutant cells...... 81

Figure 3.7 Expression of the Aft1/2-regulated gene FET3 ...... 82

Figure 4.1 Total GSH values in cytosolic and mitochondrial extracts of mtm1 ∆ and yhm2 ∆ strains ...... 96

Figure 4.2 roGFP2 redox response in WT, mtm1 ∆, and yhm2 ∆ strains ...... 97

Figure 4.3 Grx1-roGFP2 redox response in WT, mtm1 ∆, and yhm2 ∆ strains ...... 98

Figure 4.4 Proposed model pathways for mtm1∆ and yhm2∆ cells ...... 103

xii

LIST OF ABBREVIATIONS

ADH1 ...... Alcohol dehydrogenase 1

Aco1 ...... Aconitase

Aft1/2 ...... Activator of Ferrous Transport

ALR...... Augmenter of Liver Regeneration

ATP ...... Adenosine triphosphate

Atm1 ...... ATP-binding cassette (ABC) transporter

Bcl-2 ...... B-cell lymphoma 2

Bio2 ...... Biotin synthase cDNA ...... Complementary DNA

CIA ...... Cytosolic Fe-S protein assembly machinery

CT ...... Cycle threshold

DIC ...... Dicarboxylate carrier dNTP ...... Deoxyribonucleotide triphosphate

DNase I ...... Deoxyribonuclease I

xiii

DNA ...... Deoxyribonucleic acid

DPEC ...... Diethylpyrocarbonate

DTT ...... Dithiolthreitol

DTNB ...... 5,5’-dithiobis (2-nitrobenzoic acid)

EDTA ...... Ethylenediaminetetraacetic acid

ER ...... Endoplasmic reticulum

Erv1 ...... Essential for Respiration and Viability

FAD...... Flavin Adenine Dinucleotide

FET3 ...... Ferrous Transport

FXN...... Frataxin

Gfer1 ...... Growth factor Erv1-like

GLR/GR ...... Glutathione reductase

GPD/TDH3 ...... Triose-phosphate dehydrogenase

Gpx ...... Glutathione peroxidase

GRX ...... Glutaredoxin

Grx1-roGFP2 ...... Glutaredoxin attached redox sensitive green fluorescent protein

GSH...... Glutathione

xiv

Gsh1 ...... Gamma-glutamylcysteine synthetase

Gsh2 ...... Glutathione synthase

GSSG ...... Glutathine disulfide

GSH:GSSG ...... Reduced:oxidized glutathione

GST ...... Glutathione-S-transferase

HGT1 ...... Oligopeptide Transporter

H2O2 ...... Hydrogen peroxide

Hsp ...... Heat shock protein

Iba57 ...... Iron-sulfur cluster assembly factor for Biotin synthase

IMS ...... Intermembrane space

IRP ...... Iron regulatory protein

IMM ...... Inner mitochondrial membrane

ISC ...... Iron sulfur cluster assembly machinery

Isd11 ...... Desulfurase-interacting protein

Isu1 ...... Iron-sulfur cluster assembly protein

Leu1 ...... Isopropylmalate isomerase

MCs ...... Mitochondrial carriers

Mia40 ...... Mitochondrial intermembrane space Import and Assembly mRNA ...... Messenger ribonucleic acid

Mrx ...... Methionine sulfoxide reductase

Msn2/4 ...... Stress-responsive transcriptional activator

Mtm1 ...... Manganese trafficking factor for mitochondrial SOD2 mtDNA ...... Mitochondrial DNA

MTS ...... Mitochondrial targeting sequences

NAC ...... N-acetyl cysteine

NADP ...... Nicotinamide Adenine Dinucleotide Phosphate

NEM ...... N-ethylmaleimide

Nfs1 ...... Cysteine desulfurase

OD ...... Optical density

OGC ...... 2-oxoglutarate carrier

OMM...... Outer mitochondrial membrane

PAPS ...... 3’-phosphoadenylsulfate reductase

PBS-T ...... Phosphate buffered saline with tween 20

PCR ...... Polymerase chain reaction

xvi

Pgk1 ...... 3-phosphoglycerate kinase

PMS...... Post-mitochondrial supernatant

Prx1 ...... Peroxiredoxin

Rli1 ...... RNase L Inhibitor

RNase ...... Ribonuclease

ROS ...... Reactive Oxygen Species

RNR ...... RiboNucleotide Reductase roGFP2 ...... Redox Sensitive Green Fluorescent Protein

RT-PCR...... Reverse transcription polymerase chain reaction rxYFP ...... Redox sensitive yellow fluorescent protein

SC ...... Synthetic media

SD ...... Synthetic defined medium

SDS ...... Sodium dodecyl sulfate

SDS-PAGE ...... Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

Sod2 ...... Superoxide dismutase 2

SSA ...... 5-sulfosalicylic acid

TCA...... Trichloroacetic acid

xvii

TIM23 ...... Inner membrane translocase

TOM ...... Outer membrane translocase tRNA ...... Transfer ribonucleic acid

TRX...... Thioredoxin

TRR ...... Thioredoxin reductase

Tsa ...... Thioredoxin peroxidase

VDAC ...... Voltage-dependent anion channel

WT ...... Wild type

Yap1/5 ...... Basic leucine zipper (bZIP) transcription factor

Ycf1...... Yeast cadmium factor

Yhm2...... Yeast suppressor of HM mutant

YPD...... Yeast extract peptone media supplement

xviii

CHAPTER 1

INTRODUCTION

Cellular distribution of thiol redox systems in Saccharomyces cerevisiae . Thiol- disulfide homeostasis is critical for eukaryotic cells since the redox state of redox-active cysteine residues influences the structure and the function of enzymes, transcriptional factors and receptors. In addition, thiol-disulfide balance is regulated by reactive oxygen species (ROS) and unbalanced production of ROS can lead to damage to DNA, proteins, and lipids. In order to protect themselves against oxidative damage, cells have developed a series of response systems to maintain thiol-disulfide redox homeostasis. These defense systems include antioxidant enzymes and free radical scavengers. Two primary antioxidant systems, glutathione/glutaredoxin (GRX) and thioredoxin (TRX) systems are universally conserved thiol-reductase systems. Thioredoxin and glutathione pathways in the model eukaryote Saccharomyces cerevisiae are indispensable for survival. Both pathways exist in the cytosol of Saccharomyces cerevisiae , of which the thioredoxin pathway has the primary dominant thiol redox function, while the glutathione pathway functions as back up of the former (Kumar C et al 2011). The mitochondrial matrix possesses thioredoxin and glutathione pathways in which the glutathione pathway rules over the thioredoxin pathway. In addition, the endoplasmic reticulum (ER) and mitochondrial intermembrane space (IMS) harbor essential thiol oxidation reactions.

Unlike the IMS, the ER lacks thiol reductase pathways except glutathione but the IMS houses glutaredoxin 1/2 as a reducing pathway (Kojer K et al 2015).

The cytosolic thiol redox system in the cytosol of S. cerevisiae is composed of two thioredoxin isoforms (Trx1 and Trx2) and a thioredoxin reductase (Trr1) (Toledano

MB et al 2003) and two dithiol glutaredoxins Grx1 and Grx2 (Luikenhuis S et al 1998) and a glutathione reductase Glr1 (Collinson LP et al 1995). TRXs are thiol

2 oxidoreductases and reduce disulfide bonds by thiol-disulfide exchange reactions via two conserved cysteine residues at the active site. Oxidized thioredoxin is reduced to the active thiol form by NADPH-dependent thioredoxin reductase (Figure 1.1-A) (Holmgren

A 1989). TRXs are required for the numerous pathways, including reduction of ribonucleotide reductase (RNR), which is essential for producing dNTP pools for DNA synthesis (Camier S et al 2007), for reduction of 3’-phosphoadenylsulfate (PAPS) reductase participating in inorganic sulfate assimilation (Muller EG 1991), or for thioredoxin peroxides. Moreover, thioredoxins reduce cytosolic peroxiredoxins Tsa1,

Tsa2, and Ahp1, glutathione peroxidase-like enzymes Gpx1, Gpx2, and Gpx3 which are crucial to detoxify of hydrogen peroxide, and methionine sulfoxide reductases Mrx1 and

Mrx2. Thioredoxins also negatively regulate the peroxide-stress regulator Yap1 by reduction (Carmel-Harel O et al 2001) and activates the general stress regulators Msn2/4 in response to hydrogen peroxide (H 2O2) (Boisnard S et al 2009). Generally, thioredoxins are important for controlling the redox state of protein sulfhydryls, especially in antioxidant proteins.

In addition to TRXs, the yeast glutathione pathway is another major thiol redox system in the cytosol. Similar to TRXs, glutaredoxins (GRXs) are small thiol oxidoreductases required for the reduction of protein disulfides or glutathione-protein mixed disulfides. Glutathione (GSH) acts as a hydrogen donor by reducing oxidized glutaredoxin, and glutathione reductase (Glr1) provides electrons from NADPH to regenerate GSH from glutathione disulfide (GSSG) (Figure 1.1-A) (Toledano MB et al

2013).

Figure 1.1 GSH -Glutaredoxin and Thioredoxin systems in the cytosol. A) Yeast possesses cytoplasmic glutaredoxins (GRX1 and GRX2) and thioredoxins (TRX1 TRX2). The oxidized disulfide form of thioredoxin is reduced by thioredoxin reductase (Trr1) and NADPH. GSH reduces oxidized GRX and GSSG, in turn, is reduced by GSSG reductase (Glr1) in an NADPH-dependent manner. In addition, TRX systems (Trx1, Trx2 and Trr1) can reduce oxidized GSSG (Morgan B et al 2013 and Tan SX et al 2010). B) GSH-GRX and TRX systems in the mitochondrial matrix. Yeast also includes a mitochondrial TRX system consisting of TRX3 and a thioredoxin reductase (TRR2). Grx2 and Glr1 co-localize to the cytosol and mitochondrial matrix. Adapted from ref. (Trotter EW et al 2005).

Cells lacking GSH are inviable both aerobically and anaerobically without exogenously added GSH, while ∆grx1∆grx2 and ∆glr1 cells are viable. Thus, GSH has functions not shared with other pathways. The reason of the requirement of GSH for viability is confusing since this requirement is not related to oxidative stress, or to inadequate DNA synthesis or defective reduction of the essential RNR disulfide. In addition, alternative thiol-reducing agents such as cysteine, DTT or N-acetyl cysteine (NAC) do not rescue

GSH auxotrophy of the ∆gsh1 mutant (Lee JC et al 2001). Kumar C et al showed that

GSH depletion and toxic levels of GSH conditions using the ∆gsh1 and HGT1- overexpressing strains resulted in major changes in iron metabolism, but did not significantly affect thiol-redox control. This study demonstrated that the essential role of

GSH is related to its function in iron metabolism, which requires minute amounts of the tripeptide, whereas its thiol-redox function only performs as a backup of the TRX system which requires high levels of the tripeptide (Kumar C et al 2011). RNR reduction is an important target for thiol redox control carried out by the TRX pathway since RNR provides the dNTPs essential for DNA synthesis and repair. However, ∆trx1∆trx2 knockout cells are viable which indicates that the GSH pathway can back up the TRX pathway. The lethality arising from the several deletions of both pathways such as

∆glr1∆trx1∆trx2 , ∆trx1∆trx2∆grx1∆grx2 , ∆glr1∆trr1 , ∆gsh1∆trx1∆trx2 , or ∆gsh1∆trr1 might be result from deficient RNR reduction (Trotter EW et al 2003 and Draculic T et al

2000). This observation also favors GSH acting as a backup system in RNR reduction in the absence of TRXs. Another clue of incapacity of GSH pathway is redox control of phosphoadenylyl-sulfate reductase (PAPS reductase) (Muller EG 1994). Grx1 and Grx2 could replace thioredoxins in PAPS reduction only when low aeration is present since

∆trx1∆trx2 cells are able to grow on sulfate-containing media under the decreased reactive oxygen species conditions (Draculic T et al 2000). Cells depleted of GSSG reductase (Glr1) are viable, indicating that accumulated GSSG in the ∆glr1 strain is not deleterious or stored in any subcellular compartment (Collinson et al 1995). With regard to this, a recent study demonstrated that the ATP-binding cassette-C (ABC-C) transporter, Ycf1 (yeast cadmium factor), mediates removal of cytosolic GSSG that is not immediately reduced in the cytosol via transport into the vacuole. This study also showed that cytosolic GSH:GSSG homeostasis is strictly regulated and remains in the reduced state after the deletion of GLR1 (the major GSSG reductive system) or treatment with

H2O2 and is completely independent of changes in whole-cell GSSG. Furthermore, Trx2 and Grx2 were identified as backup systems to GSSG reductase for cytosolic GSSG reduction (Morgan B et al 2013). Consistent with these observations, it was shown that the TRX/TRX reductase system (Trx1, Trx2, and Trr1) can function as an alternate

GSSG reducing system (Figure 1.1-A) (Tan SX et al 2010).

Saccharomyces cerevisiae possesses two dithiol GRXs, Grx1 and Grx2. Both of them are present at the cytosol and a fraction of Grx2 is also located in mitochondria.

Grx1 and Grx2 protect cells against the superoxide anion and hydroperoxides

(Luikenhuis et al 1998). Their protective role might be explained by the fact that they are active in vitro as glutathione-S-transferases (GSTs) and they show peroxidase and GST activities in the presence of GSH in vitro (Collinson EJ et al 2002). Yeast cells have two

GSTs (Gtt1 and Gtt2) and Grx1/2 can function cooperatively with both GSTs to protect against hydroperoxides and toxic xenobiotics (Collinson EJ et al 2002).

In conclusion, the TRX pathway has the major role in cytosolic thiol redox control in Saccharomyces cerevisiae . In contrast, the GRX/GSH pathway has only a limited role which serves as a backup system in the absence of the TRX pathway or in the presence of severe redox stresses (Kumar C et al 2011). In contrast to their E. coli partners, these pathways cooperate with each other to achieve thiol-redox homeostasis in the cell.

The yeast mitochondrial matrix has a complete TRX pathway consisting of a TRX

(Trx3) and a TRX reductase (Trr2) and GRX/GSH pathway consisting of GSSG reductase (Glr1) and dithiol GRX (Grx2) which are not functionally separated as observed in their cytosolic partners (Figure 1.1-B). In addition, the yeast matrix has its own NADPH source produced by the NADH kinase Pos5 (Outten CE et al 2003). The known function of the mitochondrial TRX and GRX/GSH pathways is to reduce respiratory and exogenous ROS through Prx1, which is a mitochondrial peroxiredoxin with TRX peroxidase activity and through possibly also GSH peroxidase 2 (Gpx2)

(Perkins A et al 2014). Furthermore, the mitochondrial matrix redox homeostasis is maintained separately and isolated from redox changes from the other cell compartments

In addition, using Grx1-roGFP2 sensors, J. Riemer and coworkers showed that the inner mitochondrial membrane (IMM) prevents the equilibration of GSH:GSSG pools in the matrix and IMS in vivo and in vitro (Riemer J et al 2012). Nonetheless, the matrix has to import GSH or GSSG from the cytosol to replenish its GSH:GSSG pool but the import mechanism is still unknown. The mitochondrial matrix also harbors a monothiol GRX

Grx5. It is not known to have a function in redox homeostasis but it is involved in Fe-S cluster biosynthesis (Rodriguez MT et al 2002).

In vivo genetically encoded redox sensors . As in vivo , assessing the subcellular redox state of GSH:GSSG pools in different compartments is crucial since different cell organelles have distinct redox environments because of their unique functions.

Østergaard and coworkers have engineered a genetically-encoded, cytosolic redox sensor based on the yellow variant of green fluorescent protein (GFP) called redox-sensitive

YFP (rxYFP) which is used for intracellular redox state measurements (Ostergaard H et al 2004). GFP and its derivatives are ideal in vivo sensors because of their protease resistance and high stability under the different range of pH and buffer conditions (Cubitt

AB et al 1995). Measurement of the intracellular redox state by rxYFP in cells is performed via generation of an engineered disulfide bond that perturbs the local chromophore environment without significantly changing the overall β-can fold

(Ostergaard H et al 2001). The relative ratio of oxidized to reduced rxYFP can be easily determined via non-reducing SDS-PAGE in which the two protein forms show different electrophoretic mobilities. In addition, Østergaard and coworkers have demonstrated both in vivo and in vitro that the cysteines in rxYFP specifically equilibrate with GSH and

GSSG via rapid disulfide exchange reactions with the cytosolic glutaredoxins (GRXs)

(Ostergaard H et al 2004). Another redox-sensitive fluorescent probe is roGFP2 developed by inserting an artificial dithiol-disulfide pair into GFP with two cysteines at sites 147 and 202. Oxidation causes an increase in the 400 nm excitation peak and a decrease in the 475 nm peak for roGFP2. roGFP2 is more advantageous over rxYFP since it is a ratiometric sensor and therefore does not include errors obtained from variations in light intensity, sensor concentration and photobleaching problems (Hanson et al 2004 and Dooley et al 2004). roGFP2 was also fused with human glutaredoxin-1

(Grx1) creating Grx1-roGFP2. Recently Grx1-roGFP2 became more widely used than the rxYFP because Grx1-roGFP2 facilitates rapid equilibration between the sensor protein and the GSH/GSSG redox couple (Gutscher M et al 2008).

GSH, Erv1-Mia40 IMS disulfide relay system and other redox control systems in the mitochondrial IMS . Mitochondria consist of two compartments, the matrix and the intermembrane space. Most mitochondrial proteins are nuclear-encoded and synthesized in the cytosol. They are imported into mitochondria as precursors with a mitochondrial targeting signal (MTS) or internal targeting sequences (ITS). The import of matrix- targeted proteins is carried out by the outer membrane translocase (TOM complex) and inner membrane translocase (TIM23 complex) in an ATP and membrane potential- dependent manner (Herrmann JM et al 2012). The mitochondrial IMS is a small subcellular compartment that houses several proteins and proteins in the IMS show distinct functions. A recently published yeast proteome study identified ~50 different

IMS proteins (Vögtle FN et al 2012). Several apoptotic components are stored in the IMS and they are released into the cytosol when cell death program is triggered (Martinou JC et al 2011). In addition, IMS proteins are involved in the detoxification of ROS (reactive oxygen species) (Godon C et al 1998), mitochondrial respiration (Smith PM et al 2012), and transporting of proteins, metabolites, lipids and metal ions between mitochondrial membranes (Leary SC et al 2007 and Osman C et al 2009). IMS proteins are synthesized by cytosolic ribosomes similar to matrix proteins but only a few of them possess bipartite presequences including an N-terminal MTS and a hydrophobic sorting region. IMS proteins lacking N-terminal targeting signals are usually smaller than 20 kDa and their proper folding in the IMS prevents their translocation to the cytosol (Petrungaro C et al

2014). Folding of some of those proteins can be initiated by the binding of a cofactor. For instance, folding and retention of cytochrome c in the IMS depends on heme cofactor binding (Dumont ME et al 1988).

An IMS protein import pathway called the Mia40-Erv1 disulfide relay system uses disulfide bond formation for the import and retention of proteins in the IMS (Sideris

DP et al 2010 and Riemer J et al 2011). The substrate proteins of this machinery pass the outer membrane through the TOM complex in an unfolded and reduced state. Most of the substrates of this pathway possess twin CX 3C or twin CX 9C motifs (C, cysteine; X, any amino acid). They have a common structure in which cysteine motifs are aligned in two antiparallel helices constituting a helix-loop-helix motif. These substrate proteins are maintained in the reduced state in the cytosol by cytosolic GRXs in human cells and the

TRX system in yeast (Banci L et al 2013 and Durigon R et al 2012). In addition, the cytosolic proteasome system controls the degradation of newly synthesized substrate proteins of the Mia40-Erv1 import pathway before they are imported (Bragoszewski P et al 2013). Newly imported substrate proteins become oxidized by the oxidoreductase, import receptor, and chaperone Mia40 (mitochondrial IMS import and assembly) and

Mia40 becomes reduced. The N-terminal domain of the sulfhydryl oxidase Erv1

(Essential for Respiration and Viability) reoxidizes reduced Mia40, which forms reduced

Erv1. Electrons are then transported to cytochrome c via the FAD domain of Erv1, which additionally transfers the electrons through cytochrome c oxidase to the final electron acceptor oxygen (Figure 1.2) (Sideris et al 2010).

Grx1,2 -S-S- - S-S- 26S HS Proteasome SH Trx1, Trx2 GSH HS and Trr1 GSSG SH Mia40-Erv1 Cytosol Substrate

Porin

HS S SH S FAD ox HS S SH Cyt c S SH GSH GSSG HS -SS- HS S FAD GSSG SH -SS- S Mia40 Erv1 Grx1,2 HS S HS FAD -SS- S S -S-S- - S S S S Cyt c red GSH S-S- -SS- FAD S IMS

Cyt c Oxidase Matrix O H O 2 2

Figure 1.2 Erv1-Mia40 import pathway and GRX redox systems in the IMS. Substrates of this import pathway are synthesized in the cytosol and are kept in the reduced state by the local GSH:GSSG pool and GRXs in humans and TRXs in yeast. The cytosolic proteasome regulates the degradation of substrate proteins. The cytosolic GSH:GSSG pool is connected to the IMS GSH:GSSG pool by porins in the outer mitochondrial membrane. Limiting amounts of the IMS-localized Grx1 and 2 impedes the full equilibration of Mia40 with the local GSH:GSSG pool.

The oxidoreductase Mia40 possesses a helix-loop-helix domain that is conserved in all Mia40 homologs. This domain contains six cysteine residues present in the pattern

CPC-..-CX 9C-..CX 9C and is stabilized by two parallel structural disulfide bonds. The redox active CPC motif is positioned in a short helix and connected to a helix-loop-helix region that includes the twin CX 9C motif. The CPC motif has to be in an oxidized state to oxidize incoming substrates. The two long helices constitute a hydrophobic binding groove in which substrate proteins containing a MISS signal (mitochondrial IMS-sorting signal or ITS for IMS-targeting signal) are able to bind (Herrmann JM and Riemer J

2012).

The sulfhydryl oxidase Erv1 was initially found to serve as a compound that stimulates liver regeneration and in mammalian cells is referred to as augmenter of liver regeneration (ALR), growth factor Erv1-like (Gfer1) or hepatopoietin (Hagiya M et al

1994). Erv1 is a homodimeric flavoprotein in which each subunit is composed of two domains. The N-terminal shuttle domain containing a redox-active CXXC motif forms an intermolecular disulfide with Mia40 during reoxidation of Mia40 and shuttles electrons to the CXXC motif in the C-terminal (FAD) domain of Erv1. The C-terminal FAD-binding domain shuttles electrons via the FAD cofactor to cytochrome c, which channels electrons to the cytochrome c oxidase complex. Additionally, Erv1 includes a structural disulfide bond stabilizing the four-helix bundle structure (Guo P et al 2012) (Figure 1.3).

Another factor facilitating oxidative folding is Hot13, a small cysteine-rich IMS protein.

Hot13 is able to chelate zinc ions from reduced Mia40, which improves the reoxidation of Mia40 by Erv1, thereby accelerating mitochondrial import (Curran SP et al 2004 and

Mesecke N et al 2008).

Figure 1.3 Erv1 C-Terminal Core Domain Structure which is from Asp 86 to Asp 188 . A) The active sites are shown with ball and stick models, and the bound FAD molecules are shown as green sticks. The core redox center (Cys 130 –Cys 133 ) is in proximity to the isoalloxazine ring of FAD. B) Hydrophobic residues at the dimeric interface are shown as sticks. Phe124 is at the center of the dimeric interface of Erv1 protein. Reproduced with permission from (Guo P et al 2012).

In addition to Mia40/Erv1, there are other factors influencing the disulfide relay system in the IMS: the local GSH:GSSG pool and Grx1 and 2 (Kojer K et al 2012, Bien

MLS et al 2010 and Kojer K et al 2015). By the use of rxYFP probes, our lab showed that the redox state of the IMS is maintained separately and considerably more oxidizing than the cytosol and matrix (Hu J et al 2008). However, by the use of GRX-fused redox- sensitive fluorescence proteins (Grx1-roGFP2), it was suggested that IMS GSH:GSSG pools equilibrate with the cytosol , resulting in a reducing redox state in the IMS.

Moreover, this study showed that the IMS GSH:GSSG pool is dependent upon cytosolic

GSSG reductase (GLR) activity, cytosolic NADPH generation and porins in the OMM

(outer mitochondrial membrane) (Kojer K et al 2012). In addition, the cytosolic GSSG reductase affects the Mia40 redox state in intact wild-type cells. In terms of this study, the Mia40 redox state is balanced between Erv1 with oxidizing impact and GSH:GSSG and substrates with reducing impact and appears to be in a semioxidized state (Kojer K et al 2012).

In in vitro studies, during the substrate oxidation, GSH serves in a vital proofreading role by hindering or removing long-lived complexes between Mia40 and its substrates to ensure sufficient substrate oxidation. These accumulated intermediates are likely composed of non-natively semioxidized substrates and stable off-pathway intermediates that prevent Mia40 from becoming oxidized and postpone substrate oxidation. Previously, it was shown that GSH concentrations at physiological levels stimulate the Erv1-Mia40 import pathway. Addition of GSH frees Mia40 from trapped complexes and leads to acceleration of protein import into the IMS (Bien MLS et al

2010). However, it is still unclear how the GSH:GSSG pool influences the Mia40 redox state and the Erv1-Mia40 oxidative folding pathway in IMS in vivo.

A recently published study showed the presence of GRXs (Grx1 and Grx2), predominantly Grx2, in the IMS and proposed that limiting amounts of these GRXs provide kinetic control to hamper reduction of Mia40 substrates by local GSH:GSSG pool (Figure 2). Overexpression of an IMS-targeted Grx2 variant leads to Mia40 becoming more reduced and grx deletion prevents Mia40 reduction by GSH.

Furthermore, overexpression of Grx2 results in delay in oxidative folding for Mia40 substrates whose IMS import relies on oxidation by Mia40 (Kojer K et al 2015).

Fe-S biogenesis, iron homeostasis and their link with Erv1 and GSH . Fe-S proteins perform diverse functions in mitochondria, cytosol and nucleus. They function in energy production (e.g. aconitase in the TCA cycle), electron transfer in respiratory chain complexes I and III, fatty acid oxidation, and lipoate and biotin biosynthesis in mitochondria. In the cytosol, they are involved in the amino acid synthesis

(isopropylmalate isomerase), tRNA modification, post-transcriptional regulation of iron metabolism (IRP1) or regulation of ribosomal translation (ABC protein Rli1). Fe-S clusters are also crucial for nuclear proteins such as DNA polymerases and DNA helicases. Overall, mitochondrial Fe-S biogenesis plays a significant role in nuclear DNA maintenance, protein synthesis, translation, and nucleotide metabolism (Stehling O et al

2013).

The maturation of cytosolic and nuclear Fe-S proteins in eukaryotic cells depends on the function of the mitochondrial iron-sulfur cluster (ISC) assembly machinery. This

15 process is assisted by the cytosolic Fe-S protein assembly (CIA) machinery and the ISC export machinery. The mitochondrial ISC assembly system produces a sulfur-containing compound that is exported to the cytosol for biogenesis of Fe-S cluster proteins in the cytosol and nucleus.

The ISC assembly and export systems are highly conserved from yeast to humans and eukaryotic cells are not viable without these proteins. Several disorders are associated with mutations in mitochondrial ISC assembly proteins and many of these diseases are fatal in humans. One of these Fe-S diseases is the neurodegenerative disorder

Friedreich’s ataxia in which mutations have been found in the gene frataxin, FXN

(homologue of yeast Yfh1) which is one of the core proteins of the ISC assembly pathway (Schmucker S et al 2010). As a result of decreased frataxin protein levels, impaired respiratory complexes I-III and mitochondrial aconitase activities were observed. In addition, defects in frataxin cause impaired cytosolic IRP1 maturation and mitochondrial iron accumulation. Another Fe-S protein whose mutation results in a clinical phenotype is mitochondrial GLRX5 (Grx5 in yeast). Mutation in GLRX5 leads to sideroblastic anemia characterized by mitochondrial iron accumulation (Camaschella et al 2007). Similarly, defects in the ISC assembly scaffold protein ISCU cause a myopathy with exercise intolerance because of impairments in mitochondrial and cytosolic Fe-S protein maturation and altered iron homeostasis (Mochel F et al 2008 and Crooks DR et al 2012).

The biogenesis of extramitochondrial Fe-S proteins consists of three steps. First of all, in mitochondria, [2Fe-2S] cluster is synthesized on the scaffold protein Isu1 and this process requires the cysteine desulfurase complex Nfs1–Isd11 (Muhlenhoff U et al 2003

16 and Pandey A et al 2013). In addition, this reaction utilizes frataxin for sulfur transfer and iron supply, and NADPH-ferredoxin reductase–ferredoxin redox chain for sulfur reduction to sulfide (Parent A et al 2015 and Webert H et al 2014). Fe-S cluster is then released from the scaffold protein Isu1 by the Hsp70-Hsp40 chaperone system, followed by Fe-S cluster transfer to Grx5 in a GSH-dependent manner. These ISC components constitute the core ISC system. The final biogenesis step includes the transfer of a [4Fe-

4S] cluster by ISC-targeting factors Isa1-Isa2 and Iba57 to specific apoproteins and insertion into the polypeptide chains by Nfu1 and Ind1 (Netz DJ et al 2014).

It is well known that Fe-S protein biogenesis regulates cellular iron metabolism.

The mitochondrial ISC system has a major role in iron metabolism since depletion of the

ISC assembly core proteins or components of the export system such as GSH/Erv1 or

Atm1 is related to increased iron uptake and iron accumulation in mitochondria. In yeast, under iron-deficient conditions, iron regulation is controlled by the low iron-sensing transcriptional factor Aft1 and Aft2. Under the Fe-replete conditions, iron homeostasis is regulated by the transcriptional activator Yap5 (Outten CE et al 2013). In human cells, the major regulatory factors that influence cellular iron supply and distribution are iron regulatory proteins (IRP1) and (IRP2) via several post-transcriptional mechanisms

(Anderson CP et al 2012).

As mentioned above, the core mitochondrial ISC proteins generate a sulfur- containing compound required for maturation of cytosolic and nuclear Fe/S clusters

(Kispal G et al 1999 and Gerber J et al 2004). It was proposed that the ISC export system includes the ATP binding cassette (ABC) transporter (Atm1) of mitochondria, the IMS sulfhydryl oxidase Erv1, and GSH. This export system is crucial to both the maturation

17 of cytosolic Fe-S proteins and the regulation of cellular iron uptake (Kispal G et al 1999).

It was also suggested that ATMs transfer the intermediate of Fe-S cluster because of high iron accumulation in mitochondria of a yeast ∆atm1 mutant (Kispal G et al 1997). In order to emphasize Atm1’s role in this export process, a recent study demonstrated that

Arabidopsis ATM3 and yeast Atm1 transport glutathione trisulfide (GS-S-SG) for Fe-S cluster assembly in the cytosol (Schaedler TA et al 2014).

Another member of the ISC export machinery is GSH. Yeast ∆atm1 mutant cells showed increased GSH in mitochondria. In addition, GSH stimulates Atm1 ATPase activity suggesting that GSH may function in the export process (Kispal G et al 1997 and

Kuhnke G et al 2006). In addition, Kispal and coworkers showed that depletion of GSH resulted in defects in maturation of cytosolic Fe-S proteins and promoted mitochondrial iron accumulation, but did not affect maturation of mitochondrial Fe-S proteins, implying that GSH performs a novel and specific function in this pathway (Sipos K et al 2002).

GSH depletion or any defects in the ISC assembly machinery or ISC export machinery

(Erv1 or Atm1) result in constitutive activation of Aft1/2 and induction of iron regulon genes that encode proteins involved in cellular iron uptake and intracellular iron distribution (Rutherford JC et al 2005) (Figure 1.4). Nevertheless, as mentioned above, according to a recent study, GSH is part of the exported molecule as a GSH trisulfide

(GS-S-SG) carried out by Arabidopsis ATM3 and yeast Atm1 (Schaedler TA et al 2014).

Consistent with these results, the recently solved crystal structures of yeast Atm1 and bacterial Atm1-type ABC transporter include bound GSH in a positively charged binding pocket (Srinivasan V et al 2014). In addition, our lab pointed out that the importance of

GSH’s function might result from its role as a [2Fe-2S] cluster ligand in Grx3/4 which is

Cytosolic & nuclear Fe-S proteins IMS apo Cfd1 MATRIX Nbp35 Erv1 Isu1 GSH Atm1 GS-S-SG GSH G -S S- -S S- G apo Grx3/4 Mitochondrial Fe-S proteins Fe regulon Aft1

CYTOSOL

Figure 1.4 Putative Roles for Erv1, GSH, and Atm1 in Cytosolic Fe-S Cluster Biogenesis and Iron Regulation. Glutathione tripeptide (GS-S-SG) is generated by the mitochondrial ISC assembly machinery and exported by the mitochondrial ABC transporter Atm1. Erv1 and GSH are proposed to play roles in exporting this sulfur- containing compound needed for assembling cytosolic Fe-S clusters and regulating Aft1. GSH-ligated, [2Fe-2S]-bridged Grx3/4 homodimers are also required for both cytosolic Fe-S cluster biogenesis and iron regulation.

19 required for both cytosolic Fe-S cluster biogenesis and iron regulation (Li H et al 2012).

Similar to GSH depletion, GSH overaccumulation led to significant impacts on iron metabolism. Toledano and coworkers indicated that the fundamental function of GSH is linked to iron metabolism, which requires very low amounts of the tripeptide (Kumar C et al 2011).

The last partner in the export process is the sulfhydryl oxidase Erv1 located in the intermembrane space. In one study, a temperature-sensitive erv1 mutant ( erv1 ts ) exhibited severe defects in the maturation of cytosolic Fe-S proteins (Lange H et al 2001) similar to

∆atm1 and ∆gsh1 mutants. When Erv1 function is compromised at the non-permissive temperature (37 °C) in this strain, maturation of the cytosolic Fe-S proteins Leu1

(isopropylmalate isomerase) and Rli1 (Rnase L inhibitor) is strongly impaired. Lill and coworkers also analyzed activities of several mitochondrial Fe-S cluster-containing enzymes such as aconitase, succinate dehydrogenase and cytochrome c reductase, and reported that Erv1 did not impact the activities of these enzymes. Additionally, the free iron levels of mitochondria in the erv1 ts strain was reported to be 20 times higher iron than that found in wild-type mitochondria. However, the molecular mechanism of Erv1 and its function in this process is unknown. Furthermore, it is not known whether Mia40 is also required for this pathway.

Scope of Thesis . This thesis outlines the role of redox pathways in the IMS on GSH metabolism and subcellular thiol-disulfide redox homeostasis in different compartments thus helping characterize their role in cytosolic Fe-S biogenesis and iron homeostasis.

Chapter 2 focuses on how mutations and depletion of Erv1 influence GSH metabolism and GSH:GSSG pools in the cytosol, mitochondrial matrix and intermembrane space.

Chapter 3 investigates whether mutations on Mia40 affect the GSH:GSSG redox status of the cytosol, matrix, and intermembrane space and cellular GSH:GSSG pools. Chapter 4 focuses on determining how the deletion of the inner membrane transporters Mtm1 and

Yhm2 influence the nearby compartments. Finally, Chapter 5 provides detailed experimental protocols developed in the lab to complete these experiments.

Thiol-disulfide redox balance is crucial for the Erv1-Mia40 import pathway since it plays a role in importing cysteine-rich proteins from the cytosol into the IMS. Mutations in Erv1 create several distinct phenotypes, such as decreased import of small IMS proteins, mitochondrial morphology changes, defects in Fe-S cluster biogenesis and maturation, and defects in iron metabolism. The role of Erv1 dysfunction on GSH metabolism was studied by using different erv1 mutants in Chapter 2. We used different strains including temperature-sensitive erv1 mutants and a GAL-regulated ERV1 strain and tested their GSH levels and Mia40 oxidation. Only one erv1 mutant ( erv1-1) exhibited defects in GSH biosynthesis for all strains tested. In addition, we expressed cytosol-, IMS- or matrix-rxYFP in erv1 temperature-sensitive mutants to determine the effects of Erv1 dysfunction on the redox status of subcellular compartments and we found that only the erv1-1 mutant shows more oxidizing redox state in cytosol, matrix, and IMS due to severely decreased levels of total GSH. We further investigated the reason of GSH depletion in erv1-1 mutant and discovered that this mutant has a point mutation in the gene encoding the GSH biosynthesis enzyme, Gsh1. Furthermore, we did complementation studies for erv1-1 strain and showed that GSH1 does not substitute for

Erv1 function.

The oxidoreductase and import receptor Mia40 facilitates IMS import of a broad range of substrate proteins for the Erv1-Mia40 import pathway. Mia40 is also shown to bind a Fe-S cluster in vitro and in vivo . The impact of mia40 mutations on GSH metabolism and subcellular GSH:GSSG redox state were examined in Chapter 3. We expressed cytosol-, IMS- or matrix-rxYFP in mia40 temperature-sensitive mutants to detect the effects of Mia40 dysfunction on the redox changes. In addition, along with

GSH assay, we found that defects on Mia40 influence only IMS redox state and does not alter cellular GSH levels. Finally, we determined that defects in mia40 do not impact Fe-

S cluster biogenesis or iron regulation in yeast.

Mitochondrial inner membrane carrier proteins are involved in the translocation of various metabolites, nucleotides, and coenzymes between mitochondrial matrix and cytosol. The impact of the deletion of two mitochondrial carrier proteins, Mtm1 and

Yhm2 on total GSH levels and subcellular GSH:GSSG redox state was determined in

Chapter 4. Studies utilizing the redox sensors roGFP2 and Grx1-roGFP2 were performed to investigate the GSH:GSSG redox state of intermembrane space, matrix, and cytosol in mtm1∆ and yhm2∆ mutants. Based on the results, the mtm1∆ has the highest oxidized roGFP2 and Grx1-roGFP2 ratio in the IMS. In order to confirm the roGFP2 data, mitochondrial and cytosolic GSH levels were measured by Dr. Daithankar and he found that mtm1∆ possesses comparatively low mitochondrial GSH levels. According to these results, Mtm1 might transport a solute that directly or indirectly influences GSH metabolism.

In Chapter 5, detailed methods for the protocols developed or modified during the course of study are described. These methods include quantitative real-time PCR, whole

22 cell GSH measurement for temperature-sensitive mutants by using the Promega GSH-

GSSG Glo Kit, and blotting conditions for Erv1 and Gsh1 proteins.

In summary, we successfully characterized the role of Erv1 and Mia40 in GSH metabolism, mitochondrial import and subcellular redox state which hereby helps to reveal their roles in Fe-S cluster biogenesis and iron regulation. Discovering a secondary mutation in erv1-1 mutant causing GSH depletion and testing GSH levels in several erv1 and mia40 mutants aimed to rule out a significant role for Erv1 or Mia40 in cytosolic Fe-

S assembly and iron homeostasis. In addition, in Chapter 4, the role of inner membrane transporters on GSH:GSSG redox balance was studied. Further studies will provide new insights about whether they play a role in GSH transport or redox homeostasis in the mitochondrial matrix or IMS.

CHAPTER 2

THE ROLE OF ERV 1 IN GLUTATHIONE METABOLISM , CYTOSOLIC FE-S CLUSTER

MATURATION , AND IRON REGULATION 1

1 Ozer, H.K., Dlouhy, A.C., Thornton, J.D., Hu, J., Liu, Y., Barycki, J.J., Balk, J., and Outten, C.E. (2015). To be submitted to EMBO Rep.

ABSTRACT

Erv1 is a FAD-dependent sulfhydryl oxidase that partners with the import receptor Mia40 to import small cysteine-rich proteins into the mitochondrial intermembrane space. A previous study suggested that Erv1 has an additional role in maturation of cytosolic Fe-S cluster proteins and regulation of iron homeostasis in S. cerevisiae . However, these studies were performed on one particular erv1 mutant strain

(named as erv1-1) that we discovered has additional defects in glutathione (GSH) metabolism. In this study, we found that the erv1-1 strain possesses extremely low total

GSH levels. Since GSH is also required for iron regulation and cytosolic Fe-S cluster assembly, we investigated the Erv1-dependent connection between mitochondrial protein import, GSH metabolism, and the subcellular GSH:GSSG redox state. We tested Mia40 oxidation and GSH levels in a variety of erv1 mutants. While all the erv1 mutants tested show defects in Mia40 oxidation, only the erv1-1 strain has significantly lower GSH levels. We determined that the cause of GSH depletion in the erv1-1 strain is an additional mutation in the gene encoding the GSH biosynthesis enzyme gamma- glutamylcysteine synthetase (Gsh1) that compromises Gsh1 protein folding and/or stability. We also showed that the GSH depletion phenotype seen in erv1-1 strain could be rescued by expressing GSH1 or adding GSH to the growth media. Additionally, we measured mRNA and proteins levels of Gsh1 to confirm the instability or unfolding of

Gsh1 protein in the erv1-1 strain. We also investigated the influence of erv1 mutations on the subcellular GSH:GSSG redox state and showed that erv1 mutations led to a more reduced IMS redox state. Nevertheless, complementation studies demonstrated that

GSH1 could not replace ERV1 in cell growth assays. On the other hand, erv1-1 cells

25 expressing ERV1 were able to restore the growth of erv1-1 cells. Together, this study suggests that the defects of cytosolic Fe-S maturation and iron regulation first reported in the erv1-1 strain is a direct consequence of GSH depletion rather than indicating a direct role for Erv1 in iron metabolism and cytosolic Fe-S cluster biogenesis.

INTRODUCTION

The mitochondrial intermembrane space (IMS) is a unique subcellular compartment that houses key thiol-dependent redox pathways. The mitochondrial Mia40-

Erv1 disulfide relay system facilitates protein import and protein folding in the IMS

(Sideris DP et al 2010 and Riemer J et al 2011). Reduced and unfolded substrate proteins translocate across the outer mitochondrial membrane (TOM), are recognized and become oxidized by Mia40 redox active center following the generation of an intramolecular disulfide bond, which results in reduction of the Mia40 CPC motif. In order to initiate a new reaction cycle, Mia40 is reoxidized by the N-terminal domain of Erv1 forming reduced Erv1. Erv1 can then transfer electrons to cytochrome c via the FAD domain of

Erv1 or directly to oxygen which forms H 2O2 (Bihlmaier K et al 2007 and Tienson HL et al 2009). In vitro , GSH has been shown to impact substrate oxidation by removing stable off-pathway complexes between Mia40 and its substrates (Bien M et al 2010).

The ERV1 gene is essential for cell viability and vital for a functional respiratory chain in yeast. Mutations in the yeast ERV1 gene result in morphological changes in mitochondria, and irreversible damage of mitochondria (Becher D et al 1999). Erv1 is also required for the cell division cycle and maintenance of the mitochondrial genome in yeast (Lisowsky T et al 1992). Moreover, cells with impaired Erv1 have defects in the

26 import of small Tim proteins, decreased amounts of IMS proteins, and deficient respiratory complex function (Mesecke N et al 2005 and Di Fonzo A et al 2009). Mia40 is an essential IMS import receptor possessing a redox-active CPC motif that is oxidized by a disulfide relay system consisting of Erv1, cytochrome c and cytochrome c oxidase.

Defects in Mia40 function through mutations or depletion of Mia40 prevents the import of cysteine-rich substrate proteins (Chacinska A et al 2008 and Chacinska A et al 2004).

Mia40 is partially oxidized in intact wild-type yeast cells and reacts slowly with GSH in vitro despite a very reducing surrounding GSH pool. Its redox state is influenced by cytosolic glutathione reductase levels, Erv1, reducing substrates and GSH (Bien M et al

2010).

GSH is the primary intracellular low molecular weight thiol and serves as an antioxidant to scavenge reactive oxygen species (ROS) and electrophiles. GSH is synthesized in two steps via γ-glutamylcysteine synthetase, encoded by GSH1 and GSH synthase, encoded by GSH2 . First of all, Gsh1 conjugates glutamate and cysteine yielding

γ- glutamylcysteine. Second, Gsh2 adds glycine to γ-glutamylcysteine and produces the tripeptide GSH. The rate-limiting step in GSH biosynthesis is Gsh1 that is feedback inhibited at the enzyme level by GSH (Meister A et al 1983). Therefore, Gsh1 expression rate is very crucial to determine the rate of GSH biosynthesis. In addition to its biochemical and antioxidant functions, GSH is also important for cytosolic Fe-S biogenesis. GSH depletion disrupts the maturation of cytosolic Fe-S proteins and leads to drastic accumulation of iron in mitochondria (Sipos K et al 2002). The activity of the Fe-

S protein Leu1 (Isopropylmalate isomerase) and de novo Fe-S cluster assembly into both

Leu1 and Rli1 (RNase L Inhibitor) proteins were impaired upon GSH depletion.

However, enzyme activities of mitochondrial Fe-S proteins and de novo incorporation of

55 Fe into the mitochondrial Fe-S protein Bio2 (Biotin synthase) were unaffected (Sipos K et al 2002). Nonetheless, depletion of GSH also caused constitutive activation of the

Aft1/2 iron regulon (Rutherford JC et al 2005).

Besides involvement in the Erv1-Mia40 import pathway, the IMS sulfhydryl oxidase Erv1 is also implicated in the biogenesis of cytosolic Fe-S proteins, together with the mitochondrial transporter Atm1 and GSH. The mitochondrial ISC assembly machinery is required for the biogenesis of both mitochondrial and cytosolic/nuclear Fe-S proteins and affects cellular iron metabolism. The ATP binding cassette (ABC) transporter Atm1 in yeast mitochondria exports glutathione persulfide (GS-S-SG) to the cytosol that is crucial for both the maturation of cytosolic Fe-S proteins and the regulation of cellular iron uptake (Stehling O et al 2013 and Schaedler TA et al 2014).

Erv1 has been implicated in this process since inactivation of Erv1 (by using erv1-1 temperature sensitive mutant) led to defects in the maturation of cytosolic Fe-S proteins

(Lange H et al 2001). Upon inactivation of Erv1 at the non-permissive temperature

(37 °C), cytosolic Fe-S proteins maturation (Leu1 and Rli) was strongly decreased in comparison with wild-type cells or erv1-1 cells grown at 24 °C. However, Erv1 impairment did not impact the activities of mitochondrial Fe-S enzymes but resulted in the accumulation of iron in mitochondria (Lange H et al 2001).

Since GSH and Erv1 are each required for cytosolic Fe-S cluster assembly, we wanted to assess whether there is any connection between GSH and Erv1 function and how the mutations of Erv1 influence the subcellular GSH:GSSG redox state in the IMS.

We transformed cytosol-, matrix-, and IMS-rxYFP into erv1 temperature-sensitive

28 mutants and found that only erv1-1 strain possesses more oxidized redox state in all compartments. Since both the GSH:GSSG ratio and total GSH affect the rxYFP redox state, additionally, we measured total GSH levels in cytosolic and mitochondrial extracts in erv1-1 mutant and additional erv1 mutants and found that erv1-1 strain has severely low GSH levels in cytosolic and mitochondrial fractions. To reveal the reason of the GSH depletion in erv1-1 strain, we measured total GSH levels in several different erv1 mutants. We found that of all the erv1 mutants tested, only the erv1-1 strain has dramatically lowered GSH levels. We determined that the cause of GSH depletion in the erv1-1 strain is an additional mutation in the gene encoding the glutathione biosynthesis enzyme Gsh1 which likely compromises Gsh1 protein folding and/or stability. Iron misregulation was only found in the erv1-1 strain and is rescued with GSH addition to the growth media. In addition, erv1 mutants did not show any significant changes in activity of the human cytosolic Fe-S assembly reporter IRP1. This suggests that the defects of cytosolic Fe-S maturation and iron homeostasis seen in erv1-1 mutant might be a consequence of GSH depletion caused by a mutation in the GSH1 gene of this strain, rather than indicating a direct role for Erv1 in iron metabolism.

EXPERIMENTAL PROCEDURES

Yeast Strains, Media, and Growth Conditions —S. cerevisiae strains used in this study are listed in Table 2.1. erv1 and mia40 strains were grown on synthetic complete medium

(SC) supplemented with 2% glucose and the appropriate amino acids overnight at 24 ºC, and one set was shifted to 37 ºC and grown for 5 more hours until the cells reach mid-log phase (OD between 1-2). The GAL-ERV1 strain was maintained at 30 ºC on synthetic complete medium (SC) supplemented with either 2% raffinose and 0.5% galactose and

29 the appropriate amino acids for 24h to overexpress Erv1. In order to deplete Erv1, cells were grown on SC media with 2% raffinose for 64h. erv1-1 mutant cells expressing pRS415 (empty plasmid), ERV1, and GSH1 expressing plasmids were grown overnight on synthetic medium SC (-Leu) supplemented with 2% glucose. erv1 (F124S) mutant was created by Rabindra Behera by using a plasmid shuffling technique. For plasmid shuffling, an erv1 deletion strain with wild-type Erv1 expressed on a URA3 plasmid

(Spore 2A) was transformed with a TRP1 plasmid carrying Erv1(F124S) (pYX232- erv1 (F124S). Transformants were selected on SC-Trp plates and shuffling of the URA3 - marked plasmid expressing WT Erv1 was carried out with 5-fluoroorotic acid. Spore 2A strain and pYX232-ERV1 plasmid were obtained from Jan Riemer (Universität

Kaiserslautern, Germany). Yeast transformations were performed by the lithium acetate procedure (Gietz RD et al 1995). For spot assays, overnight cultures were adjusted to

OD 600 =1 and 5 µl of serial dilutions (10-fold each) were spotted on enriched yeast extract-peptone-based medium supplemented with 3% glycerol at 37 ºC for 2 days. For cloning of Erv1 into erv1-1 strain, the ERV1 gene was cloned from pYX232 (2µ TRP1 plasmid), which harbors wild-type ERV1 under the TPI1 promoter, into the LEU - containing plasmid pYX142 (Bien M et al 2010).

Plasmids - rxYFP sensors used in this study, including pHOJ150 (cytosol-rxYFP LEU2-

YIp ), pJH200 (IMS-rxYFP LEU2-YIp ), and pLD206 (matrix-rxYFP LEU2-YIp ), were described previously (Hu, 2010). The 2µ TRP1 plasmid pYX232-erv1 (F124S) was generated by Rabindra Behera by site-directed mutagenesis of pYX232-ERV1 (Bien et al,

2010)) using the QuikChange II Mutagenesis kit (Agilent). The CEN LEU2 plasmid p415-ADH-GSH1 was constructed by insertion of the GSH1 open reading frame in the

XbaI and XhoI sites of p415-ADH (Mumberg D et al 1995). All plasmid sequences were verified by DNA sequencing .

Subcellular Fractionation —Yeast cells were grown aerobically to mid-log phase in selecting SC medium with 2% glucose. Mitochondrial and post-mitochondrial supernatant (PMS) fractions were obtained as previously described by converting cells to spheroplasts followed by gentle lysis by Dounce homogenization and differential centrifugation (Daum G et al 1982).

Glutathione Assay - Total glutathione was measured by the DTNB-GR (Glutathione reductase) recycling assay as described previously (Strain J et al 1998). Freshly made

PMS and mitochondria extracts were deproteinated by the addition of 10% 5- sulfosalicylic acid (SSA) to a final concentration of 1% to prevent spontaneous oxidation of reduced glutathione and incubated on ice for 30 min. Precipitated protein was then removed by centrifugation at 13,000 × g for 5 min. Total glutathione was measured by adding 20 μL acidified lysate to 1 mL assay mixture (100 mM sodium phosphate, pH 7.5,

1 mM EDTA, 0.2 mM NADPH, 0.2 mM DTNB) that had been pre-warmed to 30 °C. The assay mixture was transferred into a cuvette by adding 1 unit of GSSG reductase. Assay mixture is mixed widely and then the change in absorbance at 412 nm was measured for

2 min. Glutathione solubilized in 1% SSA was used as standard curves and the rate of increase in absorbance was proportional to the amount of glutathione over this range.

Glutathione Assay by using the GSH/GSSG-Glo TM Kit - Total glutathione was measured by GSH/GSSG-Glo TM Kit (Promega, USA) following the manufacturer’s protocol with slight modifications. 4-5×10 5 cells were harvested by centrifugation and suspended in

31 lysis buffer supplied by the kit. Whole cell extracts were performed through mechanical disruption using glass beads. Based on manufacturers’ indications, desired reactions were performed and stable luciferin luminescent signals were detected using Synergy H1

Hybrid Multi-Mode Microplate Reader (Biotek, USA). The luminescent signals were proportional to the amount of GSH and correlated with the total cell amount. The results are expressed in nM of GSH per 10 7 cell. See Chapter 5 for the details.

Redox Western - Redox Western blot analysis of rxYFP and Mia40 was measured by as previously described (Hu J et al 2008). Cells were grown in SC media to mid-log phase and then acid-quenched with trichloroacetic acid (Sigma) (15% (w/v) final concentration) at 4 °C for 20 min. Five A600 units of cells were harvested by centrifugation and resuspended in 1 ml of 10% trichloroacetic acid. Following glass bead lysis, the lysed cells were transferred to a new tube and pelleted by centrifugation. The pellet was resuspended in 500 μl of 1X non-reducing SDS sample buffer containing 40 mm N- ethylmaleimide (Sigma). Following a 10-min incubation at room temperature, the proteins were separated on a 16% Tris-glycine gel (Invitrogen). Reduced and oxidized forms of rxYFP and Mia40 were analyzed by quantitative immunoblot using an Odyssey

Infrared Imaging System (LI-COR).

Immunoblotting Techniques - Cytosolic and mitochondrial fractions were subjected to electrophoresis on Tris-glycine gels (Invitrogen) and analyzed by Western blotting using an anti-Erv1 (kind gift of J.Riemer) or anti-Gsh1 (kind gift of J. Barycki) and a secondary anti-rabbit IgG (IRDye, LI-COR Lincoln, NE). Cytosolic fractions were monitored by anti-3-phosphoglycerate kinase (Pgk1) antibody (Invitrogen). Mitochondrial fractions were monitored by anti-Porin (Invitrogen). Proteins were analyzed by Western blot using

32 an Odyssey Infrared Imaging System (LI-COR). Protein concentrations were determined using the Bradford method (Bio-Rad) with bovine serum albumin as the calibration standard.

GSH1 Gene Sequencing - Genomic DNA for JRY-675 and erv1-1 strains was isolated by using Promega Wizard Genomic DNA Purification Kit and the GSH1 gene amplified from –720 to +2380 by PCR using primers listed in Table 2. Resulting DNA fragment was purified by Gel/PCR DNA Fragments Extraction Kit from IBI. The sequence integrity of PCR fragments were verified by double-stranded DNA sequencing (USC

Selah Clinical Genomics Center at Innovista (Columbia, SC).

RNA Isolation and qRT-PCR - Total RNA was isolated from JRY-675 and erv1-1 strains with the PureLink RNA Mini kit (Invitrogen, California) following the manufacturer’s protocol by using lysis buffer supplied by the kit. Yeast RNAs were reverse transcribed into cDNAs with SuperScript III (Invitrogen) following the manufacturer’s recommendations. The Real-time reverse-transcription PCR (qPCR) was carried out on

IQ ICycler (Biorad) with RT 2 SYBR Green Fast Mastermix kit (Qiagen) by using the primers shown in Table 2. ). The relative quantification of GSH1 expression was calculated by the 2 -ΔΔCT method (Livak KJ et al 2001) using the ACT1 gene (actin) as a control. See Chapter 5 for the details. bb bb --galactosidase Assays - This assay was performed by C. Outten group member

Adrienne Dlouhy. All strains tested were transformed with the FET3-lacZ reporter construct pFC-W that contains the FET3 iron-response element in a minimal promoter

(Yamaguchi-Iwai et al, 1996). Temperature-sensitive erv1, and corresponding parent

33 strains were grown in selecting SC glucose media at 24 °C to an OD 600 of 1 and then divided into 3-ml aliquots for induction. The cultures were incubated at 24 and 37 °C with either 50 μM FeCl3 (high Fe), 100 μM bathophenanthroline disulfonate (BPS) (low

Fe), or no addition (normal Fe) for 5 hours. GAL-ERV1 and the parent W303A strain were grown for 64 hours at 30 °C to an OD 600 of 1 in inducing (SC-raffinose/galactose) or repressing (SC-raffinose) media and were divided similarly into high, low, and normal

Fe aliquots and grown for an additional 4 hours at 30 °C. Cells were then harvested and assayed for β-galactosidase activity as previously described (Thorvaldsen et al, 1993).

In-Gel Aconitase Activity Assay - This assay was performed by C. Outten group member

Adrienne Dlouhy. Cytosolic and mitochondrial aconitase activities were monitored using an in-gel activity assay described previously (Tong & Rouault, 2006). Yeast cells expressing human cytosolic aconitase IRP1 (pRS425-IRP1 kindly provided by A.

Dancis) were grown to mid-log phase in selecting SC media, harvested and washed with sterile water. The pellets were then resuspended in lysis buffer (50 mM Tris-HCl, pH

8.0, 10% glycerol, 50 mM NaCl, 2.5% Triton X-100, 0.5 mM PMSF, 1 mM DTT, 2 mM citrate, protease inhibitor cocktail, 200 U/ml catalase) and subjected to glass bead lysis.

Extracts were centrifuged and the supernatants were collected for assaying. A chilled 8%

Tris-borate-citrate polyacrylamide gel was pre-electrophoresed in Tris-glycine-citrate running buffer at 140 V for 40 min. Subsequently, 100 μg of protein was loaded to the gel and electrophoresed at 140 V for 3.5 hours on ice. Gels were incubated at 37 °C for

30 min in the dark with aconitase activity assay stain (100 mM Tris-HCl, pH 8.0, 1 mM

NADP+, 2.5 mM cis-aconitate, 5 mM MgCl2, 1.2 mM MTT, 0.3 mM phenazine methosulfate, and 5 U/ml isocitrate dehydrogenase) and scanned

Table 2.1 Strains used in this study Strain Genotype Reference

JRY-675 MAT a, ∆leu2, ura3-52, his4-519 (Lisowsky T 1992)

erv1-1 MAT a, ∆leu2, ura3-52, pet492 ts (Lisowsky T (pet492 -6A ) 1992)

YPH499 MAT a, ade2-101 , his3-∆200 , leu2-∆1, ura3-52 , trp1- (Sikorski RS et ∆63 , lys2-801 al 1989)

erv1-2 MAT a, ade2-101 , his3-∆200 , leu2-∆1, ura3-52 , trp1- (Rissler M et al, ∆63 , lys2-801 , erv1 ∷ADE2 [pFL39-erv1-2] 2005)

erv1-5 MAT a, ade2-101 , his3-∆200 , leu2-∆1, ura3-52 , trp1- (Muller JM et al ∆63 , lys2-801 , erv1 ∷ADE2 [pFL39-erv1-C159S] 2008)

W303A MAT a, ade2-1, his3-11,15, leu2-3,112, trp1-1, ura3-1, can1-100

GAL-ERV1 MAT a, ade2-1, his3-11,15, leu2-3,112, trp1-1, ura3-1, R.Lill, can1-100, pERV1 ∷GALL-natNT2 unpublished

Spore 2A MAT a, ade2-1, his3-11,15, leu2-3,112, trp1-1, ura3-1, (Bien M et al (W303) can1-100 erv1::HIS3 [p ERV1 (URA3)] 2010)

W303-ERV1 MAT a, ade2-1, his3-11,15, leu2-3,112, trp1-1, ura3-1, This study can1-100 erv1::HIS3 [pYX232-ERV1 ]

W303- MAT a, ade2-1, his3-11,15, leu2-3,112, trp1-1, ura3-1, This study erv1(F124S) can1-100 erv1::HIS3 [pYX232-erv1(F124S) ]

erv1 ts trp1-1, ura3-1, leu2-3,112, his3-11,15, ade2-1, can1- (Aloria K et al 100, GAL2+, met2-Δ1, lys2- Δ2, erv1::TRP 2004)

W303 MATa ura3-1 ade2-1 trp1-1 his3-11, 15 leu2-3, (Aloria K et al 112[rho+] 2004)

Table 2.2 Primers used for sequencing and qRT-PCR.

Primer Name Primer Sequence GSH1 Sequencing Primers GSH1 -700 5¢ -GTTTGGTTCAGCGCGACGAG-3¢ GSH1 term 5¢ -CAGGAGATTCCAGGTTCGAG-3¢ GSH1 seq forward 5¢ -GCTAGCCGACCAAGATGTTC-3¢ GSH1 seq reverse 5¢ -CAATGCCTCCAAATCCGTTC-3¢ qRT-PCR Primers GSH1 forward 5¢ - TCTAGGACGTACAATGAACAC-3¢ GSH1 reverse 5¢ - CCTCCATATTAAGCTCAGTG -3¢ ACT1 forward 5¢ - GGATTCCGGTGATGGTGTTACT -3¢ ACT1 reverse 5¢ - TGGCGTGAGGTAGAGAGAAACC-3¢

RESULTS

Only erv1-1 shows a GSH depletion phenotype. To uncover whether there is any linkage between Erv1 and GSH metabolism, we measured total GSH levels in the erv1-1 strain.

First, the redox state of Mia40 was analyzed to verify the defect of Erv1 in this strain. erv1-1 temperature sensitive mutant shows normal function when grown at 24 °C, but exhibits defects in IMS protein import when grown at 37 °C (Becher D et al 1999 and

Rissler M et al 2005). Since Mia40 includes a redox-active disulfide bond that is oxidized by Erv1, malfunctions in Erv1 will cause reduction of this disulfide in Mia40 (Figure 2.1-

A). We grew the erv1-1 strain overnight at the permissive temperature (24 °C) and then switched them to the restrictive temperature (37 °C) for 5 hours to impair Erv1 function.

As expected, erv1-1 mutant showed defects in Mia40 oxidation at 37 °C (reduced Mia40,

~35%) (Figure 2.1-B). We measured the Mia40 redox state by performing redox western blot technique (non-reducing SDS-PAGE). Then, we measured total GSH levels in cytosolic and mitochondrial fractions in erv1-1. We noticed a 16-fold decrease of total cytosolic GSH levels in the erv1-1 strain compared to WT at 24 °C and 13-fold decrease of total cytosolic GSH levels in the erv1-1 strain compared to WT at 37 °C (Figure 2.2-

A). In addition, mitochondrial total GSH levels decreased at both temperatures, however, not as much as cytosolic GSH pools (Figure 2.2-B). In order to test whether the GSH depletion phenotype discovered in the erv1-1 strain is directly related to the Erv1-Mia40 import system or unique to that specific mutant, we measured total GSH levels in cytosol and mitochondria of additional erv1 strains (Figure 2.2 A-B). Similarly, we determined

Mia40 redox state for these additional mutants and found that Mia40 was mainly in reduced state at 37 °C in all erv1 strains (Figure 2.1-C).

A 24°C 37°C

WT erv1-1 WT erv1-1 red ox

-Mia40 Mia40 Oxidized %

C 120 24°C 37°C 100

40 % Oxidized Mia40 Oxidized % 20

0 WT(2) erv1-2 erv1-5

Figure 2.1. In vivo Mia40 redox state measurements in erv1 mutan ts. WT and erv1 strains were grown to mid-log phase in SD media overnight at 24 ºC, and one set was shifted to 37 ºC and grown for 5 more hours. Redox western blots were performed on these strains and were immunoblotted with α-Mia40 antibody. Reduced and oxidized forms of Mia40 were quantified using an Odyssey Infrared Imaging System. The reported values are the mean of 3-4 independent experiments. Error bars are the means + standard deviations. The WT strain used for erv1-1 is JRY675 (WT(1)). The isogenic WT strain for erv1-2 and erv1-5 is YPH499 (WT(2)).

Figure 2.2 Total GSH values in cytosolic, mitochondrial and whole cell extracts for erv1 strains. WT and erv1 strains were grown in SD media overnight at 24 ºC, and one set was shifted to 37 ºC and grown for 5 more hours. The reported values are the mean of 4-5 independent experiments. Error bars are the means + standard deviations. The WT strain used for erv1-1 is JRY675 (WT(1)). The isogenic WT strain for erv1 -2 and erv1 -5 is YPH499 (WT(2)).

Total GSH levels are only slightly changed in these erv1 strains in both cytosolic and mitochondrial fractions (Figure 2.2 A-B). Furthermore, we tested whole cell GSH levels in all erv1 mutants and demonstrated that only erv1-1 mutant possesses severely decreased GSH levels in whole cell extracts (Figure 2.2-C). These results showed that only the erv1-1 mutant exhibited a GSH depletion phenotype while all the other erv1 mutants did not. Nevertheless, we questioned whether the specific mutation site of the erv1-1 strain is the reason for the GSH depletion. Since the erv1-1 mutant has mutation on the site of phenylalanine 124 to serine, an erv1 mutant carrying the same mutation

(F124S) was recreated in a different strain background by using a plasmid shuffling technique. Reduced Mia40 at 37˚C (33% oxidized) in the erv1(F12S) mutant confirmed its temperature sensitivity and mutation (Figure 2.3-A). Nevertheless, no significant changes were observed in whole cell GSH values between WT and erv1(F124S) mutant at both temperatures (Figure 2.3-B). Furthermore, in order to see the effects of Erv1 depletion on GSH values without any temperature effect, a GAL-ERV1 strain in which the

ERV1 gene is controlled under the regulatable promoter was used to deplete Erv1 protein.

GAL-ERV1 cells were cultivated on media with raffinose-galactose (inducing condition) or only raffinose (non-inducing condition) and the Mia40 redox state, Erv1 protein levels, and whole cell GSH levels were examined. As a control study of this strain as in others, we tested the Mia40 redox state. Reduced Mia40 (~%35 oxidized) was detected as expected 64h after shifting to non-inducing conditions, confirming depletion of Erv1 in that strain (Figure 2.4-A). Moreover, as shown in Figure 2.4-B, western blot assay demonstrated that under the repressive conditions, Erv1 was not expressed in mitochondria. More importantly, we did not observe any reduction in whole cell GSH

40 values between the WT and GAL-ERV1 strains (Figure 2.4 C). On the contrary, we found higher GSH values when Erv1 is overexpressed and non-expressed. In addition, in order to determine if Erv1 protein expression level influences GSH metabolism, Erv1 levels were demonstrated by immunostaining analysis for erv1 strains. Erv1 expression levels in mitochondrial fractions mostly decreased at 37 °C on erv1 mutants including erv1-1 and showed a similar pattern at 37 °C (Figure 2.5). This result was expected since cells lacking functional Erv1 have impaired import of substrate proteins of the Erv1-Mia40 pathway into the IMS, and Erv1 is a substrate of this pathway. (Rissler M et al 2005 and

Mesecke N et al 2005). As an additional control to rule out the effects of Erv1 on GSH metabolism, an erv1 mutant carrying the same F124S mutation but reported to accumulate high mitochondrial iron levels at the non-permissive temperature of 37 °C

(Aloria K et al 2004) was tested for total GSH measurement (Figure 2.6). We did not observe any decrease in total GSH levels in cytosolic and mitochondrial fractions at both temperatures and GSH values of this erv1 mutant are very similar to that of WT. Taken together, these results demonstrate that only the erv1-1 mutant displayed a severe GSH deficiency phenotype independent of the F124S mutation, defective Erv1 levels, and depletion of Erv1 protein.

red ox

Figure 2.3 in vivo Mia40 redox state and whole cell GSH measurements in erv1- (F124S) strain A) In vivo Mia40 redox state measurements of WT (W303A) and mutant strain erv1(F124S). B) Total GSH values in whole cell extracts for erv1(F124S) mutant and WT strain. WT and mutant strains were grown in SD (-Trp) media overnight at 24 ºC, and one set was shifted to 37 ºC and grown for 5 more hrs. Total cell extracts were prepared by using glass-beads lysis. Error bars represent the standard deviation for three independent experiments.

red ox

Figure 2.4. In vivo Mia40 redox state, immunoblot for Erv1 protein and whole cell GSH measurements in GAL-ERV1 strain. A) In vivo Mia40 redox state measurements of WT (W303A) strain and GAL-ERV1 strain. Cells were grown in the presence of defined medium with both 2% raffinose and 0.5% galactose, centrifuged, and resuspended in 2% raffinose to deplete Erv1. B) Immunoblot for Erv1 protein in mitochondrial extracts. 20 µg mitochondrial protein were used for western blot analysis, respectively. C) Total GSH measurements in whole cell extracts for GAL- ERV1 and WT strain. Cells were collected following shift to non-inducing condition

(0 h) and 64 h later. Error bars represent the standard deviation for 3-4 independent experiments.

Figure 2.5 Mitochondrial Erv1 protein levels in erv 1 strains . WT and erv1 strains were grown in SD media at 24 ºC and 37 ºC and were analyzed by SDS-PAGE and immunoblotted with α-Erv1 and α-Porin antibodies. 20 µg mitochondrial protein was used. The WT strain used for erv1-1 is JRY675 (WT(1)). The isogenic WT strain for erv1-2 and erv1-5 is YPH499 (WT(2)).

Cytosolic GSH Mitochondrial GSH 40 8 24 °C 24 °C 37 °C 37 °C 30 6

20 4

10 2

0 0 ts WT erv1 WT erv1 ts

Figure 2.6 Total GSH values in cytosolic and mitochondrial extracts of WT and ts erv1 mutant carrying F124S mutation with high mitochondrial Fe levels. WT(W303A) and erv1 ts (F124S) strains were grown to mid-log phase in SD (-Trp) media overnight at 24 ºC, and one set was shifted to 37 ºC and grown for 5 more hrs and were subjected to GSH assay.

Defects in Erv1 result in a more oxidizing GSH:GSSG redox state in the cytosol, matrix and IMS. We also wanted to test how GSH depletion in erv1-1 strain and how mutations in erv1 affect the GSH: GSSG environment in the cell. We expressed cytosol-,

IMS-, or matrix-rxYFP in erv1 mutants and monitored the redox state of rxYFP by redox western technique. rxYFP was used as an in vivo sensor since the cysteines in rxYFP specifically equilibrate with GSH and GSSG via rapid disulfide exchange reactions catalyzed by endogenous glutaredoxins (GRXs). In addition, reduced and oxidized forms of rxYFP protein show different mobilities by non-reducing SDS-PAGE providing facile measurement of the protein’s redox state. The redox western results utilizing the targeted rxYFP redox sensors indicated a more oxidized GSH:GSSG redox state in all compartments tested (cytosol, matrix, and IMS) in the erv1-1 strain in comparison to the

WT strain (Figure 2.7). In addition, using the differently-targeted versions of rxYFP, we tested the redox state in the cytosol, mitochondrial matrix, and mitochondrial IMS in additional erv1 mutants. As shown in Figure 2.7-A, the cytosolic redox state of erv1 mutants is highly similar to the cytosolic redox state of WT. This finding is logical since we did not observe any significant difference in total cytosolic GSH levels in additional erv1 mutants. The only difference was seen in erv1-1, which was highly oxidized compared to WT (Figure 2.7-A). The matrix redox state in erv1-1 is also compatible with total mitochondrial GSH levels tested. erv1-1 strain has dramatically lowered mitochondrial GSH levels and showed more oxidized matrix redox state than that of WT

(Figure 2.7-B). However, other erv1 mutants have no significant impact on matrix GSH:

GSSG compared to WT (Figure 2.7-B). Additionally, the IMS redox state in the erv1-1 strain is more oxidized than WT at both temperatures. However, a more reduced

Figure 2. 7 Subcellular GSH :GSSG redox state in erv1 mutants. WT and erv1 strains transformed with an rxYFP expression plasmid pHOJ150 (cytosol-rxYFP), pLD206 (matrix-rxYFP), or pJH200 (IMS-rxYFP) and cells were grown in SD (-Leu) selection media. Redox western blots were performed on the indicated strains and immunoblotted with α-GFP. Reduced and oxidized forms of rxYFP were quantified using an Odyssey Infrared Imaging System. Error bars represent the standard deviation for 3-4 independent experiments.

47 redox state was observed in other erv1 mutants at 37 ˚C as expected since malfunction of

ERV1 causes a defect in the oxidation of substrate proteins and leads to accumulation of reduced substrate proteins and reduced IMS GSH:GSSG (Figure 2.7-C). Thus, this data indicate that GSH depletion might not be directly related to Erv1 function in this erv1-1 strain.

The cause of the GSH depletion in the erv1-1 strain is an additional mutation in the gene encoding the glutathione biosynthesis enzyme Gsh1. To uncover the mechanism for GSH depletion in this erv1-1 strain, we tested whether this strain exhibited decreased

GSH synthesis in comparison to WT cells as well as other erv1 and mia40 mutant strains.

The tripeptide glutathione (GSH) is synthesized in two steps via γ-glutamylcysteine synthetase (Gsh1) and GSH synthase (Gsh2). We measured the protein levels of the first

GSH biosynthesis enzyme (Gsh1). Cytosolic extract was isolated from erv1 mutant cells grown in SD medium and the cytosolic fractions were analyzed by immunostaining. The erv1-1 strain shows dramatically decreased Gsh1 protein levels (10-fold at 24 ˚C, 15-fold at 37 ˚C) compared to WT at both temperatures (Figure 2.8). However, Erv1 expression levels in other erv1 strains were similar to WT. For this western blot assay, Pgk1 (3- phosphoglycerate kinase) was used as a loading control and the levels of Pgk1 were not altered. Additionally, we observed decreased Gsh1 expression at 37 °C for both WT and erv1 strains. Next, we questioned if the strong decrease in Gsh1 levels in erv1-1 strain is due to a secondary mutation of GSH1 gene. To determine this, we sequenced the GSH1 gene of erv1-1 and discovered an arginine to histidine mutation at position 280. Arg-280 is located in a well-conserved region of the Gsh1 protein sequence and is likely important for protein folding and stability (Figure 2.9 A-B). This mutation may be the direct cause

48 of decreased Gsh1 expression levels in this erv1-1 strain. Additionally, we measured mRNA levels of the GSH1 gene in erv1-1 to determine whether there are translational or post-translational defects in GSH1 expression. We used ACT1 (Actin1) that does not vary in expression between different samples as a reference gene and endogenous control and showed the results as relative fold change of GSH1 /ACT1 (Figure 2.10). As shown in

Figure 2.10, we found that mRNA levels of GSH1 were increased in erv1-1 at both temperatures as compared to WT. This upregulation is likely due to decreased Gsh1 stability and activity, which leads to lower intracellular GSH levels which triggers oxidative stress sensing by the transcription factor Yap1 (Wu AL et al 1994 and Wheeler

GL et al 2003). Oxidized Yap1 in turn activates expression of the GSH1 gene (Stephen

DW et al 1997 and Dormer UH et al 2002). In addition, we noticed that WT GSH1 mRNA levels at 37 ˚C increased as compared to WT mRNA levels at 24˚C. This result is consistent with the fact that heat shock causes the activation of Yap1 (Sugiyama K et al

2000), leading to activation of the GSH1 gene. Taken together, these results demonstrate that GSH deficiency in erv1-1 is caused by a previously unknown secondary mutation in the Gsh1 protein sequence that likely leads to protein instability.

Using the erv1-1 strain, a previous study suggested that Erv1 plays a role in cytosolic Fe-S cluster biogenesis since this strain showed defects in cytosolic Fe-S cluster assembly (Lange H et al 2001). Another study also demonstrated that GSH is essential for the maturation of cytosolic Fe-S proteins (Sipos K et al 2002). Therefore, our hypothesis was that the cytosolic Fe-S assembly defect reported for the erv1-1 mutant strain results from GSH depletion in that specific mutant, rather than indicating a direct role for Erv1 in cytosolic Fe-S cluster assembly. Our lab member Adrienne Dlouhy

49 measured iron-dependent expression of the Aft1-regulated gene ( FET3 ) in erv1 strains.

The FET3 gene, which encodes a multicopper ferroxidase, is required for ferrous iron uptake as part of the high affinity iron uptake system at the cell surface and is regulated by the transcription factors Aft1 and Aft2 (Philpott CC et al 2008). Induction of FET3 via

Aft1/2 is linked to Fe-S cluster biogenesis and hereby the activity of the FET3 promoter is connected with Fe-S biogenesis (Rutherford JC et al 2005 and Chen OS et al 2004).

FET3 is normally repressed when cells are iron-replete and becomes activated under conditions of iron depletion. However, defects in Fe-S cluster synthesis leads to constitutive expression of this gene in the presence of both low and high iron (Outten CE et al 2013). Adrienne measured expression of this gene via β-galactosidase assays by transforming a FET3-LacZ reporter vector into the erv1 mutants with and without the addition of GSH in iron-replete and iron-limited medium. Of all the erv1 mutants tested, the only strain to exhibit iron misregulation is the GSH-deficient erv1-1 strain, which could be rescued with GSH addition (Figure 2.11). In addition, she expressed human cytosolic aconitase (IRP1), which functions as a cytosolic Fe-S assembly reporter, by transforming erv1 mutants with a plasmid expressing IRP1 under a strong promoter

(Zhang Y et al 2008) and found that human cytosolic aconitase (IRP1) activity is similar in WT and erv1 mutants (Figure 2.12). Based on the measurements of GSH levels in erv1 mutants, the measurements of activity of a cytosolic Fe-S protein IRP1 and the measurement of expression of iron regulon gene FET3 , we suggest that the Fe-S cluster defect found in the erv1-1 strain is a direct result of GSH depletion in that particular strain.

Figure 2.8 erv1-1 strain has dramatically decreased Gsh1 protein levels . WT and erv1 mutant cells were grown in SD media at 24 °C and 37 °C. Cytosolic fractions were isolated and 75 µg cytosolic protein was used for the western blot. Gsh1 and Pgk1 bands were quantified using an Odyssey Infrared Imaging System and relative to control ratio was calculated. Error bars are standard deviation for 3 independent experiments.

Figure 2.9 Mutation found in GSH1 gene of erv1-1 strain likely compromises

Gsh1 protein folding and/or stability. A) Residues in the conserved region of gamma glutamylcysteine synthetase (Gsh1) in S. cerevisiae , S. pombe , mouse, rat, and human. The position of Arg-280 is indicated with a green arrow. B) Crystal structure of yeast Gsh1 enzyme (Biterova E et al 2009). Arg-280 is shown in green and forms a salt bridge with Asp-403 (in orange) in a nearby helix. Mg 2+ ions in active site are shown as blue spheres with glutamate (in orange) and ADP (in purple).

Figure 2.10 Real time RT-PCR analysis of mRNA levels of WT and erv1-1 strains. WT and erv1-1 cells were grown in YPD media at 24 °C and 37 °C and total RNA was isolated. Complementary DNA (cDNA) was produced from total RNA by using reverse transcriptase. cDNA was used as template for RT-PCR analysis with the fluorescent DNA probe SYBR green.

Figure 2. 11 Expression of the Aft1/2 -regulated gene FET3. WT, erv1 , and GAL- ERV1 strains were transformed with a FET3-LacZ reporter plasmid and grown at 24 or 37˚C. β-galactosidase activity was measured for cells grown in medium supplemented with iron (high Fe), no iron (normal Fe), or the iron chelator BPS (low Fe) in the presence or absence of 1 mM GSH. A) β-galactosidase activity measurement of WT (JRY675) and erv1-1 strains. B) β-galactosidase activity measurement of WT(YPH499) and erv1 strains. C) β-galactosidase activity measurement of WT(W303-A) and GAL-ERV1 strains . Error bars indicate standard deviation for 3-5 independent experiments. This assay was performed by Adrienne Dlouhy.

Figure 2.12 Human cytosolic aconitase activity (IRP1) activity in WT and erv1 strains. Cytosolic and mitochondrial Fe-S cluster assembly was monitored using an in-gel aconitase activity assay in cells expressing human cytosolic aconitase IRP1. Aconitase activities for 100 μg of cell lysates were separated by native gel electrophoresis. Positions of the endogenous mitochondrial aconitase activity (Aco1) and plasmid-expressed cytoplasmic aconitase activity (IRP1) are indicated in erv1-1 strain (A) and erv1-2 and erv1-5 strains (B). In addition, immunoblots for IRP1 protein was performed. Whole cell lysates were analyzed by SDS-PAGE and immunoblotted with anti-IRP1 (100 μg of protein for each). This assay was performed by Adrienne Dlouhy.

Complementation of the growth defects of erv1-1 cells. To address whether GSH replaces the function of Erv1 or rescues the GSH depletion phenotype, GSH1 and ERV1 genes were expressed in erv1-1 mutant cells. We also expressed the GSH1 gene in erv1-1 by using a plasmid with GSH1 under the control of the TEF1 promoter to see whether

GSH1 overexpression rescues the GSH depletion phenotype seen in erv1-1. For the confirmation of ERV1 complementation, we measured the Mia40 redox state in these erv1-1/ERV1 transformants. Expression of WT Erv1 complemented the Mia40 oxidation phenotype since Mia40 was primarily oxidized at both 24 °C and 37 °C for erv1-1 cells expressing wild-type Erv1 (Figure 2.13-A). For the confirmation of GSH1 complementation, we measured whole cell GSH in WT, erv1-1 cells with empty plasmid

(pRS415), and erv1-1 cells expressing GSH1 (Figure 2.13-B). We found that total GSH levels in erv1-1 cells expressing GSH1 are very high compared to WT and erv1-1 with empty plasmid. Nevertheless, we measured whole cell GSH levels in erv1-1 strain with the addition of 1 mM GSH to the growth media since our lab member, Adrienne found the recovery in Aft1-regulated gene expression in erv1-1 strain when adding 1 mM GSH to the media. Whole cell GSH levels in erv1-1 are increased with 1 mM GSH addition which confirms that erv1-1 does not have any defects in GSH uptake (Figure 2.13-B).

Additionally, we performed western blot assay to check Gsh1 protein levels in erv1-1 cells expressing GSH1 and found that Gsh1 is highly expressed in these transformants

(Figure 2.14-C). Thus, immunostaining and GSH measurements show that the GSH depletion phenotype is rescued by adding back GSH1 into erv1-1.

For the complementation of growth defects of erv1-1 strain, we did spot assays on

YPG (glycerol) plates. It has been shown that oxidative phosphorylation and the presence

56 of mitochondrial genome are dispensable as long as fermentable carbon sources such as glucose or fructose are present in the growth media. In addition, temperature-sensitive erv1 strains are not able to grow on media containing non-fermentable carbon sources such as glycerol due to defects in respiratory chain assembly (Lisowsky T 1992 and

Lisowsky T et al 1994). At the permissive temperature (24 °C) when Erv1 functions normally, we did not observe any growth defect in erv1-1 with WT ERV1 and erv1-1 cells expressing GSH1 as expected (Figure 2.14-A). On the other hand, GSH1 could not substitute for ERV1 in cell growth assays even though cells have sufficient GSH concentrations (Figure 2.14-B). We also used YPG plate with 1 mM GSH for spot tests and adding GSH to the growth media did not help to rescue Erv1 defects in erv1-1.

However, erv1-1 cells expressing ERV1 were able to fully restore the growth of erv1-1 cells at the restrictive temperature (37 °C) (Figure 2.14-B).

red ox

Figure 2.13 Confirmation assays for complementation of erv1-1 strain. A) Mia40 redox state for WT (JRY 675), erv1-1, erv1-1 with empty plasmid (pRS 415) and erv1-1 with WT Erv1. B) Total GSH levels in whole cell extracts. C) Western blotting against Gsh1 and Pgk1. 75 µg cytosolic extracts were used for western blotting and Pgk1 was used for loading control.

Figure 2.14 Spot testing of cell growth with wild-type (JRY675), and erv1-1 mutant cells expressing pRS415 (empty plasmid), ERV1 , and GSH1 . Cells were analyzed by spotting 5 µl of cells, at OD values of 1.0, 0.1, and 0.01 onto YPG and 600 YPG + 1 mM GSH plates under aerobic conditions at 24 °C and 37 °C for 48 h.

DISCUSSION

The sulfhydryl oxidase Erv1, one of the components of the Erv1-Mia40 disulfide relay system in the IMS, was originally characterized as a new gene essential for cell viability and for functional respiratory chain activity in yeast (Lisowsky T et al 1994).

Besides its role as a redox partner with Mia40, it was proposed that Erv1 also contributes to the biogenesis of cytosolic Fe-S cluster proteins and cellular iron homeostasis. In a previous study published by Roland Lill’s group, cytosolic Fe-S protein defects were found in the erv1 ts mutant strain (named as erv1-1 in this study) at the non-permissive temperature (37 °C), while no mitochondrial Fe-S cluster defects were found (Lange H et al 2001). In addition to Erv1, GSH has been reported to impact cytosolic Fe-S proteins maturation since both depletion and overaccumulation of GSH causes impaired cytosolic

Fe-S cluster biogenesis (Sipos K et al 2002, Kumar C et al 2011). Our research demonstrated that the erv1-1 mutant strain has much lower GSH levels than the corresponding WT strain. Since GSH is required for cytosolic Fe-S cluster maturation

(Sipos K et al 2002), it is important to determine whether the GSH depletion in this mutant is the underlying cause of the cytosolic Fe-S cluster biogenesis defects rather than loss of Erv1 function. In order to investigate this, we tested if other erv1 strains show

GSH decrease or this phenotype is unique to this erv1-1 strain. Our results indicate that the erv1-1 mutant strain exhibits GSH depletion at both temperatures while the other erv1 mutant strains did not. Moreover, Mia40 was still significantly more reduced in these strains at 37 °C as compared to WT. Therefore, these findings suggest that severe GSH depletion upon Erv1 dysfunction is unique to the erv1-1 strain and is not the direct reason of the Mia40 oxidation defect seen in all the strains tested.

Our next step was to question the reason for the GSH depletion in the erv1-1 mutant. We know that this mutant has a F124S mutation in the ERV1 gene. In order to test if this single mutation is the cause of the GSH depletion phenotype seen in erv1-1, we tested the same mutation in a different genetic background. Interestingly, we did not observe any GSH decrease in this newly created erv1-(F124S) mutant even though it exhibits Mia40 oxidation defect at the restrictive temperature, thus suggesting that the

F124S single point mutation is not the cause for GSH depletion in erv1-1.

Moreover, in order to examine how changes in Erv1 expression affects GSH levels which is independent of temperature effects, we tested a GAL-ERV1 strain. Erv1 depletion did not lead to reduction in GSH levels. On the other hand, both Erv1 overexpression and depletion led to somewhat increased total GSH levels in whole cell extracts. Increased total GSH levels in the absence of Erv1 function might be due to an oxidative stress generated by the accumulation of H2O2. Dabir et al demonstrated that oxygen serves as the final electron acceptor for Erv1-mediated reactions causing the release of H 2O2, which is subsequently reduced to water by cytochrome c peroxidase. In addition, they showed that H2O2 levels increase in erv1 mutants at the restrictive temperature, suggesting that one of the Erv1 electron acceptor pathways relies on cytochrome c (Dabir DV et al 2007). Consequently, depletion of Erv1 leading to increased H2O2 production might activate the Yap1 transcriptional factor. Yap1 induces the expression of its target antioxidant genes such as GSH1 , GLR1 and GPX1 upon oxidative stress such as H2O2 generation. Therefore, upon depletion of Erv1 causing H2O2 accumulation, Yap1 might induce GSH1 expression leading to the elevated levels of

GSH. In the future, it would be interesting to test if depletion or overexpression of Erv1 leads to increase in H2O2 levels and also leads to increase in Gsh1 expression levels.

We also tested another erv1 ts strain that has the same point mutation as erv1-1 but in a different strain background (W303). This strain was reported to accumulate 10-fold higher iron than normal in its mitochondria, presumably due to defects in cytosolic Fe-S cluster assembly (Aloria K et al 2004). However, we did not observe any decrease in total

GSH levels in this erv1 ts strain. Furthermore, inconsistent with this previous result, our group did not detect any increase in mitochondrial Fe in this erv1 strain (Dlouhy A 2015).

Additionally, in order to confirm the GSH assay results, we measured the

GSH:GSSG redox state in the cytosol, IMS, and matrix of erv1 mutant strains using our targeted rxYFP sensors. Cytosol- and matrix-rxYFP are significantly more oxidized in the erv1-1 strain as compared to the other strains which is consistent with the severe GSH decrease found in this strain. However, IMS-rxYFP is slightly more oxidized in the erv1-

1 strain compared to WT, implying that GSH depletion seen in erv1-1 does not further impact the IMS since the IMS redox state is already mostly oxidized. We also found that

IMS-rxYFP is consistently more reduced in other erv1 mutants than corresponding wild- type strains at 37 °C. This result is consistent with the fact that the Mia40-Erv1 system assures the oxidation of the substrate proteins in the IMS. Disruption of this pathway might cause a more reduced GSH:GSSG redox state in this compartment because mutations on Erv1 lead to defects on precursor release, formation of the oxidized monomers, and eventually inhibition in substrate oxidation (Müller JM et al 2008 and

Rissler M et al 2005) which causes the accumulation of reduced precursors in the IMS.

Because the F124S point mutation and defective Erv1 are not the cause of the GSH depletion phenotype in erv1-1, we decided to check the expression level of the first GSH biosynthesis enzyme, Gsh1. Surprisingly, we found dramatically low Gsh1 expression and with further studies by sequencing, discovered a mutation in the GSH1 gene corresponding to residue 280. Arg 280 is located in a well-conserved region of Gsh1 and it forms a salt bridge with Asp 402 (Figure 2.9). It also forms backbone hydrogen bonds with Leu 369 and Gly 370 (Biterova EI et al 2009). Disruption of these interaction may impact the placement of Tyr 362 in a loop region which helps coordinate the alpha- carboxylate of the glutamate substrate. Thus, the R280H mutation might indirectly impact substrate binding and reduce GSH levels or it might impact protein stability.

Further structural and enzymatic studies on the Gsh1 R280H mutant are required to confirm these effects.

In order to rescue the GSH depletion phenotype seen in erv1-1, we added 1 mM

GSH to the growth media and also expressed the WT GSH1 gene in erv1-1. Either adding

GSH to the media or ectopic expression of WT GSH1 rescued the GSH depletion phenotype in this mutant. We also performed real-time PCR and found that mRNA levels of erv1-1 are 4-fold higher than that of wild-type when Erv1 is functional and 2-fold higher than that of wild-type when Erv1 is nonfunctional. We also observed that GSH1 mRNA levels are elevated in wild-type cells at the restrictive temperature. Recently, it has been shown that post-translational regulation leads to low correlation between mRNA and protein concentrations. Therefore, the processes of translation and protein degradation are at least as important as mRNA transcription and stability to steady-state protein abundance concentrations (Vogel C et al 2010). Increased mRNA level in erv1-1

63 strain at both temperatures is explained by the fact that mutation in the gene might lead to instability, degradation and eventually lowered half-life in the protein triggering the resynthesis of mRNA. Moreover, mutated or misfolded proteins are recognized and removed by ubiquitin-mediated proteolysis. Increase in the protein degradation rate might cause resynthesis of the protein thus more transcription of mRNA since GSH1 is expressed ubiquitously in the cell and gsh1∆ cells are not viable (Penninckx M et al

2000). These findings are also consistent with the increased mRNA levels found in wild- type strain at restrictive temperature (37˚C). Unstable protein in the presence of high temperature degrades and this degradation leads to increased turnover rate of mRNAs which is supported by decreased level of Gsh1 expression (Figure 2.8). In addition, GSH depletion and heat-shock stress result in activation of the Yap1 transcription factor causing increased GSH1 expression leading to increased mRNA levels (Wheeler GL et al

2003, Sugiyama K et al 2000).

To investigate Erv1’s role in iron regulation, our lab member Adrienne Dlouhy found that FET3 expression was only misregulated in the GSH-deficient erv1-1 strain which is rescued with GSH addition. Furthermore, she established that activity of the Fe-

S cluster enzymes mitochondrial aconitase and cytosolic human IRP1 were unaffected in erv1 mutants. Additionally, our collaborator Janneke Balk’s group determined that only the erv1-1 strain showed a small reduction in activity of the cytosolic Fe-S cluster protein

Leu1. These results demonstrate that the reason of the defects in cytosolic Fe-S protein biogenesis and iron homeostasis in the erv1-1 strain is GSH depletion in that strain which is confirmed by the additional mutation found in the gene encoding the glutathione biosynthesis enzyme Gsh1.

CHAPTER 3

MUTATIONS IN MIA 40 AFFECTS THE IMS REDOX STATE AND HAVE NO EFFECTS ON

IRON REGULATION 1

1 Ozer, H.K., Dlouhy, A.C., Thornton, J.D., Hu, J., Liu, Y., Barycki, J.J., Balk, J., and Outten, C.E. (2015). To be submitted to EMBO Rep.

ABSTRACT

A recently identified disulfide relay system in the mitochondrial intermembrane space (IMS) is crucial for importing small, cysteine-rich proteins by facilitating disulfide bond formation. The sulfhydryl oxidase Erv1 and the import receptor/oxidoreductase

Mia40 constituting this system are affected by the redox state of GSH:GSSG in this compartment since mitochondrial protein import is stimulated by GSH. Mia40 has been shown to bind a Fe-S cluster in vitro and when overexpressed in vivo ; however, its potential role in Fe-S cluster assembly and iron metabolism has not previously been addressed. In order to uncover how the Erv1-Mia40 system impacts GSH metabolism, we determined the total GSH in temperature-sensitive mia40 mutants. Our data suggest that defective Mia40 leads to no significant difference in total GSH levels in cytosolic, mitochondrial, and whole cell fractions. We also measured the subcellular GSH:GSSG redox state by utilizing targeted in vivo redox sensors in mia40 mutants and found that

Mia40 mutations solely impact the IMS GSH:GSSG redox state. Additionally, we found a decrease in Gsh1 expression level in mia40 mutants, when Mia40 is not functional.

Nonetheless, we wanted to reveal whether Mia40 plays a role in iron homeostasis, which is dependent on iron-sulfur (Fe-S) cluster assembly. We monitored iron homeostasis by transforming a FET3-LacZ reporter vector into the mia40 mutants. Expression of the iron-uptake gene FET3 is not altered with and without GSH addition in mia40 mutants suggesting that Mia40 has no direct effect on iron regulation.

INTRODUCTION

Proteins in the IMS have diverse functions such as mitochondrial respiration, transport of proteins, metal ions, metabolites, and lipids between membranes, and apoptosis. Most IMS proteins lack N-terminal targeting signals and are nuclear-encoded, synthesized in the cytosol, and directed into the IMS on a unique import pathway differing in several aspects from the matrix-targeting pathway. IMS proteins containing conserved twin CX 3C and CX 9C motifs play important roles in protein transport of hydrophobic membrane proteins, assembly of cytochrome c and copper loading (Hell K

2008 and Kawamata H et al 2008). Import of these proteins is performed by a mitochondrial disulfide relay system consisting of the import receptor Mia40 and the sulfhydryl oxidase Erv1 (Herrmann JM et al 2012). Both proteins are highly conserved in eukaryotes and essential for viability in yeast. Conditional Mia40 and Erv1 temperature- sensitive mutants exhibit severe defects in mitochondrial biogenesis (Becher D et al

1999, Lisowsky T 1994, and Chacinska A et al 2004). In yeast, Mia40 is synthesized as a precursor with a mitochondrial targeting signal and a sorting domain, imported into mitochondria via the TOM and TIM23 complexes and gets anchored at the inner membrane via its N-terminal targeting signal (Chacinska A et al 2004). In mammalian cells, Mia40 is not anchored to the inner membrane due to lack of an N-terminal extension and is soluble in the IMS (Hofmann S et al 2005).

Mia40 can bind directly to imported substrate proteins in the IMS and acts as an import receptor. It has six conserved cysteine residues including a redox-active CPC motif, and two structural disulfides consisting of twin CX 9C motifs (Figure 3.1) (Banci L et al 2009 and Kawano S et al 2009). The import of cysteine-rich precursor proteins into

Figure 3.1 Schem atic diagram of Mia40 structure . Mia40 contains a helix-loop- helix domain stabilized by two structural disulfide bonds. It has a short helical region including the redox-active CPC motif that has to be in an oxidized state to oxidize its substrates. In addition, the substrate proteins contain a MISS motif able to bind to the hydrophobic groove formed by the helix-loop-helix domain of Mia40 (Herrmann JM et al 2012).

68 the IMS occurs through the TOM complex. Mia40 promotes disulfide bond formation and oxidative folding of the imported precursors by forming transient intermolecular disulfides with its substrate proteins. Release of substrate proteins from Mia40 is performed and oxidative folding traps the substrates in the IMS. Moreover, reduced

Mia40 is reoxidized by Erv1 transferring electrons to cytochrome c and finally to oxygen

(Böttinger L et al 2012). The Mia40-Erv1 import machinery cooperates with the TOM complex and MICOS (MItochondrial contact site and Cristae Organizing System), which brings Mia40 close to the TOM complex in order to accelerate protein folding (von der

Malsburg K et al 2011 and Zerbes RM et al 2012). Mia40 also mediates the oxidation of proteins that are further imported into the matrix via the TIM23 complex or into the inner membrane (Wrobel L et al 2013 and Longen S et al 2014), as well as the oxidation of larger and more complex substrate proteins such as Atp23 and Tim22 (Weckbecker D et al 2012 and Wrobel L et al 2013). In addition, glutathione (GSH) affects the Mia40 redox state and hence protein import by assisting Mia40-mediated substrate oxidation (Bien M et al 2010, Kojer K et al 2012, and Kojer K et al 2015). However, it is not clear how

Mia40-Erv1 function impacts GSH metabolism and the GSH:GSSG redox state in the

IMS and other compartments.

In addition to its role in IMS protein import, Mia40 is also suggested to influence

Fe-S cluster assembly in vitro . Banci L et al showed Mia40-dependent import of the human version of Dre2 (anamorsin) into the intermembrane space (Banci L et al 2011).

Dre2 is a Fe-S protein localized in the IMS and was proposed to possess two Fe-S clusters in its C-terminal domain containing two conserved cysteine motifs (Zhang Y et al 2008). It cooperates with the diflavin oxidoreductase, Tah18 by forming an electron

69 transfer chain that is essential for the assembly of cytosolic and nuclear Fe-S proteins

(Netz DJ et al 2010). Mia40 oxidizes a twin CX 2C motif of the C-terminal domain of anamorsin and interacts with anamorsin through an intermolecular disulfide-bonded intermediate (Banci L et al 2011). In addition, overexpressed Mia40 was shown to bind iron in vivo , while Fe–S-containing Mia40 is not an electron donor for Erv1 (Spiller MP et al 2013). Because recombinant human Mia40 might bind iron or an ISC in vitro

(Daithankar VN et al 2009), it was speculated that Mia40 might coordinate an Fe-S cluster in the intermembrane space. On the contrary, a recent study indicated that Dre2 is not localized to IMS, but associates with the cytosolic surface of mitochondria and contains only reduced thiol residues and no disulfide bonds. Additionally, yeast Dre2 does not interact with Mia40 in vivo or in vitro , and when it is mixed with purified Mia40 in vitro , it is not oxidized (Peleh V et al 2014). Th fiesendings question the role of Mia40 as a Fe-S protein important for cytosolic Fe-S cluster assembly.

Two mia40 temperature sensitive mutant ( mia40-3 and mia40-4) yeast cells used in this study grow like wild type cells at the permissive temperature (24 ˚C), but were strongly impaired at the non-permissive temperature (37 ˚C) on fermentable media

(glucose) and were unable to grow on nonfermentable media (glycerol) at 37 ˚C

(Chacinska A et al 2004). The mutant protein Mia40-3 is defective in its interaction with the precursors of small Tim proteins, which results in decreased amounts of Tim proteins.

The mutant protein Mia40-4, however, is able to bind the precursors of small Tim proteins, and the precursor proteins are translocated into the IMS. However, as precursor proteins accumulate in this mia40-4 mutant, assembly of the small Tim proteins into the mature TIM complexes is postponed (Chacinska A et al 2004).

In this chapter, we wanted to demonstrate the effects of Mia40 on GSH metabolism to reveal their roles in thiol-disulfide equilibrium. We were also interested in whether mutations in Mia40 have an effect on the redox homeostasis in the IMS using the fluorescent protein-based redox sensor, rxYFP. Moreover, we wanted to show whether

Mia40 plays a role in iron homeostasis. Based on our findings, disruption of Mia40 specifically impacts the IMS GSH:GSSG pool, but not matrix and cytosol GSH:GSSG redox balance. Furthermore, Mia40 mutations do not alter total GSH levels in the cell and do not impact iron regulation.

EXPERIMENTAL PROCEDURES

Yeast Strains, Media, and Growth Conditions — S. cerevisiae strains used in this study were WT strain YPH499 ( MATa, ade2-101, his3-Δ200, leu2-Δ1, ura3-52, trp1-Δ63, lys2-

801 ) and the mutant strains mia40-3 (MATa, ade2-101, his3- Δ200, leu2-Δ1, ura3-52, trp1-Δ63, lys2-801, mia40::ADE2 [pFL39-FOMP2-8ts/mia40-3] ), mia40-4 (MATa, ade2-101, his3-Δ200, leu2-Δ1, ura3-52, trp1-Δ63, lys2-801, mia40::ADE2 [pFL39-

FOMP2-7ts/mia40-4]) obtained from Dr. Agnieszka Chacinska (Institut fur Biochemie und Molekularbiologie, ZBMZ and Centre for Biological Signalling Studies (BIOSS),

Universitat Freiburg, 79104 Freiburg, Germany). Yeast transformations were performed by the lithium acetate pro cedure (Gietz RD et al 1995). Strains were grown on synthetic complete medium (SC) supplemented with 2% glucose and the appropriate amino acids overnight at 24 ºC, and one set was shifted to 37 ºC and grown for 5 more hours.

Plasmids - rxYFP sensors used in this study included pHOJ150 (cytosol-rxYFP LEU2-

YIp ) pJH200 (IMS-rxYFP LEU2-YIp ) and pLD206 (matrix-rxYFP LEU2-YIp ) are

71 described before (Hu, 2010). The FET3-lacZ reporter construct pFC-W (pDW840) that contains the FET3 iron-response element in a minimal promoter is described before

(Yamaguchi-Iwai et al, 1996).

Subcellular Fractionation - Yeast cells were grown aerobically to mid-log phase in selecting SC medium with 2% glucose. Mitochondrial and post-mitochondrial supernatant (PMS) fractions were obtained as previously described by converting cells to spheroplasts followed by gentle lysis by Dounce homogenization and differential centrifugation (Daum G et al 1982).

Glutathione Assay - Total glutathione was measured by the DTNB-GR (glutathione reductase) recycling assay as described previously (Outten CE 2004).

GSH Assay by using the GSH/GSSG-Glo TM Kit - Whole cell GSH was measured by

GSH/GSSG-Glo TM Kit (Promega, USA) following the manufacturer’s protocol with slight modifications. 4-5×10 5 cells were harvested by centrifugation and suspended in lysis buffer, supplied by the kit. Whole cell extracts were prepared through mechanical disruption using glass beads. Based on the manufacturers’ indications, desired reactions were performed and stable luciferin luminescent signals were detected using a Synergy

H1 Hybrid Multi-Mode Microplate Reader (Biotek, USA). The luminescent signals were proportional to the amount of GSH and correlated with the total cell amount. The results are expressed in nmol GSH per 10 7 cell.

Redox Western - Redox western blot analysis of rxYFP and Mia40 was measured as previously described (Hu J et al 2008). Cells were grown in SC media to mid-log phase and then acid-quenched with trichloroacetic acid (Sigma) (15% (w/v) final concentration)

72 at 4 °C for 20 min. Five A 600 units of cells were harvested by centrifugation and resuspended in 1 ml of 10% trichloroacetic acid. Following glass bead lysis, the lysed cells were transferred to a new tube and pelleted by centrifugation. The pellet was resuspended in 500 μl of 1X non-reducing SDS sample buffer containing 40 mm N- ethylmaleimide (Sigma). Following a 10-min incubation at room temperature, the proteins were separated on a 16% Tris-glycine gel (Invitrogen). Reduced and oxidized forms of rxYFP and Mia40 were analyzed by quantitative immunoblot using an Odyssey

Infrared Imaging System (LI-COR).

Immunoblotting Techniques - Cytosolic and mitochondrial fractions were subjected to electrophoresis on Tris-glycine gels (Invitrogen) and analyzed by western blotting using an anti-Erv1 (kind gift of J.Riemer) (1:10000 dilution) or anti-Gsh1 (kind gift of J.

Barycki) (1:10000 dilution) antibody, and a secondary anti-rabbit IgG (IRDye, LI-COR

Lincoln, NE) (1:25000 dilution). Cytosolic fractions were monitored by an anti-3- phosphoglycerate kinase (Pgk1) (1:10000 dilution) antibody (Invitrogen). Mitochondrial fractions were monitored using an anti-Porin antibody (Invitrogen) (1:12500 dilution).

Proteins were analyzed by western blot using an Odyssey Infrared Imaging System (LI-

COR). Protein concentrations were determined using the Bradford method (Bio-Rad) with bovine serum albumin as the calibration standard. bb bb -galactosidase Assays —mia40 strains tested were transformed with the FET3-lacZ reporter construct pFC-W that contains the FET3 iron-response element in a minimal promoter (Yamaguchi-Iwai et al, 1996). Temperature-sensitive mia40 and corresponding parent strain were grown in selecting SC glucose media at 24 °C to an OD 600 of 1 and then divided into 3-ml aliquots for induction. The cultures were incubated at 24 and 37

°C with either 50 μM FeCl3 (high Fe), 100 μM bathophenanthroline disulfonate (BPS)

(low Fe), or no addition (normal Fe) for 5 hours. GAL-ERV1 and the parent W303A strain were grown for 64 hours at 30 °C to an OD 600 of 1 in inducing (SC raffinose/galactose) or repressing (SC-raffinose) media and were divided similarly into high, low, and normal Fe aliquots and grown for an additional 4 hours at 30 °C. Cells were then harvested and assayed for β-galactosidase activity as previously described

(Thorvaldsen et al, 1993).

RESULTS

Changes in Mia40 redox state and Erv1 protein levels in mia40 mutants. A recently identified disulfide relay system in the mitochondrial intermembrane space, including the sulfhydryl oxidase Erv1 and import receptor Mia40, is affected by GSH, which accelerates the oxidation between Mia40 and its substrates in vivo and in vitro . On the other hand, it is not known how the disruption of this pathway affects the redox state of the IMS and nearby compartments. In order to investigate this, we used two different mia40 temperature sensitive mutants ( mia40-3 and mia40-4) that contain several point mutations generated by error-prone PCR and are isolated by screening for a temperature- sensitive growth phenotype. The mutations led to the following amino-acid changes for mia40-3: K100R, E142G, E144A, D178G, M361V, E370Stop and for mia40-4: T313A,

I316M, F342L. First of all, similar to what we did in Chapter 2, the redox state of Mia40 was measured in mia40 strains to confirm the defect of Mia40 oxidation at 37 °C. Mia40 has a redox-active cysteine motif that has to be in an oxidized state to oxidize its substrates. Mutations in Mia40 result in reduction of this disulfide bond. We grew the mia40 strains overnight at the permissive temperature (24 °C) as described in Chapter 2

74 and then switched them to the restrictive temperature (37 °C) for 5 hours to impair Mia40 function. When we measured Mia40 redox state in these mia40 mutants, we observed the reduced Mia40 at 37 °C in these strains which confirms the Mia40 defect (Figure 3.2).

We also observed the faster mobility on SDS-PAGE for mia40-3 strain. The reason is that a stop codon in the mia40-3 allele caused a C-terminal truncation of Mia40-3 and a faster mobility on SDS–PAGE than the wild-type protein (Chacinska A et al 2004).

Second, we tested Erv1 protein expression levels in mia40 strains as an additional confirmation study. Disruption of Mia40 function will lead to defects in Erv1 import and consequently, reduced amount of Erv1 in the IMS. We performed reducing SDS-PAGE to test the expression level of Erv1 protein in mitochondrial fractions of mia40 strains and found that Erv1 expression is mostly reduced in mitochondria at the restrictive temperature (37 °C) as expected (Figure 3.3). Based on these findings, we conclude that mia40 temperature-sensitive mutants possess defective Mia40 when grown at 37 °C.

Disruption of Mia40 leads to a more reducing IMS redox state and no significant changes on iron homeostasis . In order to uncover the mechanism of the connection between Mia40 and GSH metabolism and their impact on subcellular compartments, we measured total GSH levels and the GSH:GSSG redox state in mia40 strains. We expressed cytosol-, IMS-, or matrix-rxYFP in mia40 temperature-sensitive mutants.

Redox western results utilizing the targeted rxYFP redox sensors demonstrated that mutations in Mia40 function cause a more reduced GSH:GSSG redox state in the IMS compared to the WT strain (Figure 3.4-A). The matrix-rxYFP and cytosol-rxYFP redox states of the mia40 strains are similar to the WT strain at both the permissive and restrictive temperatures (Figure 3.4B-C). Thus, we observed that mutations in Mia40

75 impact the GSH:GSSG redox state in IMS; however, the effect varies depending on the specific mutation. We then performed total GSH assays to confirm the redox western data. Total GSH levels of cytosolic and mitochondrial fractions in mia40 strains are similar to the WT strain at both temperatures (Figure 3.5A-B). In addition, we measured whole cell GSH in these mia40 strains and found no significant changes in total GSH levels between WT and mia40 strains (Figure 3.5-C). We also checked Gsh1 protein expression levels in mia40 strains and found that when Mia40 is defective at the restrictive temperature (37 °C), Gsh1 expression levels are decreased by ~2 fold compared to WT in both mia40 mutants (Figure 3.6). It would be interesting to check mRNA levels of GSH1 in mia40 strains to examine whether Mia40 dysfunction also impacts GSH1 gene expression. Therefore, we conclude that Mia40 malfunction has no effect on cellular GSH levels and only influences the IMS GSH:GSSG redox state.

As we mentioned in the introduction of this Chapter, Mia40 was proposed to serve as a Fe-S protein and thus may be involved in cytosolic Fe-S assembly. Adrienne

Dlouhy from the C. Outten group measured expression of the FET3 gene via β- galactosidase assays of mia40 mutants with and without the addition of GSH in iron- replete and iron-limited medium. She found no significant changes in FET3 expression in low iron media with and without the addition of GSH (Figure 3.7). Since mia40 mutants did not show FET3 activation under high and low iron conditions, we conclude that

Mia40 is not involved in iron homeostasis.

red ox

Figure 3.2 In vivo Mia40 redox state measurements in mia40 temperature sensitive mutants. Redox western blots were performed on mia40 strains and blots were immunoblotted with α-Mia40 antibody. Reduced and oxidized forms of Mia40 were quantified using an Odyssey Infrared Imaging System. The reported values are the mean of 3-4 independent experiments. Error bars are the means + standard deviations. The WT strain used for mia40-3 and mia40-4 is YPH499.

Figure 3.3 Mitochondrial Erv1 protein levels in mia40 strains . WT and mia40 strains were grown in SD glucose media at 24 ºC and 37 ºC and were analyzed by SDS-PAGE and immunoblotted with α-Erv1 and α-Porin antibodies. 15 µg mitochondrial protein was used. WT strain used for mia40-3 and mia40-4 is YPH499.

Figure 3.4 Subcellular GSH :GSSG redox state in mia40 mutants. WT and mia40 strains were transformed with an rxYFP expression plasmid pHOJ150 (cytosol- rxYFP), pLD206 (matrix-rxYFP), or pJH200 (IMS-rxYFP) and grown in SD (-Leu) glucose selection media. Redox western blots were performed on the indicated strains. Error bars represent the standard deviation for 3-4 independent experiments. WT(2) strain used for mia40-3 and mia40-4 is YPH499.

Figure 3.5 Total GSH values in cytosolic, mitochondrial and whole cell extracts for mia40 strains. WT and mia40 strains were grown in SD glucose media overnight at 24 ºC, and one set was shifted to 37 ºC and grown for 5 more hours. The reported values are the mean of 4-5 independent experiments. Error bars are the means + standard deviations. WT strain used for mia40-3 and mia40-4 is YPH499.

Fig ure 3.6 Gsh1 protein levels in mia40 mutant cells. WT and mia40 strains were grown in SD glucose media at 24 °C overnight and one set was shifted to 37 °C for 5 hours. Cytosolic fractions were isolated and used for the western blot. Pgk1 was used as a loading control with 75 µg cytosolic protein. Gsh1 and Pgk1 bands were quantified using an Odyssey Infrared Imaging System and relative to control ratio was calculated. The reported values are the mean of at least 3 independent experiments. Error bars are the means + standard deviations.

24 C 37 C high Fe normal Fe WT(2) low Fe

mia40-3

mia40-4 SC

WT(2)

mia40-3

mia40-4

SC + GSH SC + GSH 0 100 200 300 400 0 100 200 300 400 -galactosidase activity -galactosidase activity

Figure 3.7 Expression of the Aft1/2 -regulated gene FET3. WT and mia40 strains were transformed with a FET3-LacZ reporter plasmid and grown at 24 or 37 °C in SC glucose media. β- galactosidase activity was measured for cells grown in medium supplemented with iron (high Fe), no iron (normal Fe), or the iron chelator BPS (low Fe) ± 1 mM GSH. Error bars indicate standard deviation for 3-5 independent experiments. This assay was performed by Adrienne Dlouhy.

DISCUSSION

In this study, the redox state of Mia40 in vivo was analyzed for mia40 strains to confirm their defects. Mia40 exists in a mixture of oxidized ( ∼80%) and reduced ( ∼

20%) forms in vivo for WT and mia40 strains when Mia40 is functional. Our result is consistent with a recent study by Kojer et al 2012 in which approximately 70% of Mia40

CPC was oxidized in vivo in WT strain. However, mutations on Mia40 lead to reduced

Mia40 (~25-45%) at the restrictive temperature (37 °C) as expected. Since Mia40 is required for the import of Erv1 into the IMS, as another confirmation study, we analyzed

Erv1 expression in mitochondrial fractions of mia40 strains. The steady-state level of

Erv1 was significantly reduced when mia40 strains were grown at elevated temperature which is consistent with earlier findings that mia40 mutants contain reduced levels of substrate proteins such as small Tim proteins (Chacinska A et al 2004 and Bragoszewski

P et al 2013).

We have also established that Mia40 dysfunction does not influence the redox state of matrix or cytosol, but impacts the IMS redox state demonstrated by utilizing organelle-targeting rxYFP sensors. As we have shown in Chapter 2, mutations on Erv1 caused the reduction in oxidized rxYFP ratio in the IMS. Similarly, defects on Mia40 led to a more reduced IMS redox state. The mutant protein Mia40-3 is defective in its interaction with Mia40 due to impaired formation of Mia40 intermediate with substrate proteins, as a consequence, impaired import of the precursors into the IMS. The mutant protein Mia40-4, however, is able to efficiently bind the precursor proteins, but is impaired in their release (Chacinska A et al 2004), hereby mia40 mutations inhibit the oxidation of substrates in the IMS.

In this study, we maintained reduced IMS GSH:GSSG redox state by using rxYFP, which relies on the availability of endogenous dithiol glutaredoxins for the rapid and complete equilibration with the surrounding glutathione pool in mia40 mutants. Previously, our lab found that an IMS ‐targeted version of rxYFP was more oxidized than its respective cytosolic and matrix variants (Hu J et al 2008). Recently published study pointed out the presence of limiting availability of glutaredoxins in the IMS (Kojer K et al 2015). When endogenous glutaredoxin availability or activity is limiting, then rxYFP may become oxidized by other sources such as ROS from the respiratory chain and thus lose its ability to serve as a specific reporter for the glutathione redox couple. Therefore, unfused probes rxYFP sense the redox environment the same way proteins thiols do. In the IMS, rxYFP behaves similar to a substrate protein that is influenced primarily by the Mia40-Erv1 system and not by GSH:GSSG since there is limiting Grx available to allow equilibration with GSH:GSSG. Thus, Mia40 dysfunction leads to reduced rxYFP response in the IMS.

In addition, in this study, we demonstrated that any defects on Mia40 did not impact total

GSH levels in mitochondria, cytosol, or whole cell.

We also observed decreased Gsh1 expression, but no change in total GSH levels when Mia40 is not functional. So far, there is no study to show the link between Mia40 and Gsh1. The reason of the decrease in Gsh1 expression might be changes at the transcriptional or post-transcriptional level. In addition, post-translational changes such as degradation or ubiquination might lead to reduced protein turnover and, eventually reduced protein synthesis (Lu SC 2009). However, Mia40 dysfunction does not result in lower total GSH levels. The reasons for the normal level of total GSH are following: First of all, decreased Gsh1 expression might activate Yap1 transcription factor that

84 upregulates Gsh2 catalyzing the second step of GSH synthesis which hereby normalizes cellular GSH levels (Lu SC 2009). Second, decreased Gsh1 expression might increase the rate of Gsh1 enzymatic reaction by increasing the substrate cysteine uptake whose availability regulates Gsh1 activity (Tateishi N et al 1974).

Mia40 has been also shown to bind iron when overexpressed in vivo and to assemble a Fe-S cluster in vitro (Spiller MP et al 2013). Moreover, it was speculated that

Mia40 might coordinate a Fe/S cluster in the IMS, but the existence of such a cofactor for this function in vivo is still missing. In Chapter 1, we showed that Erv1 is not involved in extramitochondrial Fe-S protein maturation and iron regulation. Similarly, measurements of the expression of the Aft1-regulated iron-uptake gene, FET3 , performed by Adrienne

Dlouhy demonstrated that FET3 expression does not change under low iron conditions in mia40 strains suggesting that Mia40 plays no significant role in iron homeostasis since impairment of mitochondrial Fe-S cluster biogenesis and export activates the FET3 expression (Outten CE et al 2013). Previously, yeast Dre2 and its human homolog,

Anamorsin, components of the cytosolic iron sulfur (Fe-S) assembly machinery and their interaction with Mia40 in the IMS were documented (Banci L et al 2011). Purified Mia40 was able to oxidize anamorsin in vitro (Banci L et al 2011), by introducing two disulfide bonds into a domain suggested to serve as a Fe-S binding region (Zhang Y et al 2008).

Based on this, they conclude that anamorsin/Dre2 was the first identified Fe-S protein imported into the IMS, and might play a role in cytosolic Fe-S biogenesis. On the other hand, recently published study indicates that Dre2 does not interact with Mia40 in vivo or in vitro , contains only reduced thiol residues and no disulfide bonds and is not oxidized when mixed with purified Mia40 in vitro (Peleh V et al 2014). Interestingly, they showed

85 that Dre2 is not localized to IMS, but bounds to the cytosolic side of the outer membrane of mitochondria excluding the role of Mia40 in this process (Peleh V et al 2014). This study suggests that there is no redox modification of the cysteine residues of Dre2 in vivo by Mia40 or no coordination of a Fe/S cluster in the mitochondrial intermembrane space.

Our data also supports these findings. No Aft1 activation seen in mia40 mutants suggests that Mia40 is not involved in iron regulation. Moreover, the Fe-S cluster in Mia40 may still be physiologically relevant as shown before, but not involved in Fe-S cluster biogenesis pathways as previously suggested.

CHAPTER 4

DELETION OF MITOCHONDRIAL CARRIER PROTEINS , MTM 1 AND YHM 2 AFFECTS

REDOX STATUS OF MITOCHONDRIA

ABSTRACT

The mitochondrion has several distinct functions such as oxidative phosphorylation, Fe-S clusters and heme biosynthesis, and biosynthesis of lipids, amino acids, and nucleotides. These metabolic pathways are all dependent on cysteine- containing proteins. Therefore, maintaining thiol-disulfide redox balance in this organelle is critical for cellular functions. Intracellular redox homeostasis is mainly controlled by the reduced and oxidized forms of glutathione (GSH and GSSG), which serve as an intracellular redox buffer. To investigate the factors that affect the mitochondrial

GSH:GSSG redox state, we used genetically modified in vivo redox sensors roGFP2 and

Grx1-roGFP2 (green fluorescence protein and glutaredoxin-fused green fluorescence protein). Both sensors equilibrate with GSH:GSSG pools via disulfide bond formation.

Utilizing the roGFP2 in vivo sensors, we demonstrate how the deletion of manganese cofactor of superoxide dismutase 2, (Mtm1) and citrate-oxoglutarate carrier (Yhm2) affect the redox status of the mitochondrial matrix and intermembrane space (IMS). We have found that deletion of the MTM1 leads to a large oxidative shift in the IMS

GSH:GSSG redox state, with little change in the matrix GSH:GSSG redox state.

Moreover, deletion of YHM2 shows a smaller influence on the mitochondrial GSH:GSSG redox state under the same conditions. However, both mtm1∆ and yhm2∆ mutants were shown to have decreased mitochondrial GSH levels. Overall, these results indicate that reduction in mitochondrial GSH levels in mtm1∆ and yhm2∆ mutants impact mitochondrial compartments distinctively due to their different transport properties.

INTRODUCTION

Mitochondria perform diverse functions such as respiration, energy metabolism, synthesis of amino acids, membrane phospholipids, Fe-S clusters and heme.

Mitochondria also provides a network for cell signaling and regulation. On the other hand, metabolic exchange and chemical trafficking is severely restricted between the mitochondrial matrix and the cytoplasm. The mitochondrial outer membrane possesses non-selective pores (porins or voltage-dependent anion channel (VDAC)) and is freely permeable to molecules below ~ 5 kDa such as sucrose, adenine nucleotides, Coenzyme

A, and tRNA. However, the mitochondrial inner membrane transports metabolites based on driving forces and substrate specificity (Wohlrab H 2009). A nuclear-encoded superfamily of secondary transport proteins called mitochondrial carriers (MCs) catalyze the translocation of several metabolites, nucleotides and coenzymes across the inner membrane of mitochondria and provide a connection between metabolic reactions of the cytosol and mitochondrial matrix. The members of this carrier family are widespread in eukaryotes and share common sequence features such as a tripartite structure, six transmembrane α-helices and a 3-fold repeated signature motifs (Kunji ERS et al 2004).

Mitochondrial carrier proteins function in several distinct metabolic pathways, such as oxidative phosphorylation, citric acid cycle, fatty acid oxidation, amino acid degradation, gluconeogenesis, transfer of reducing equivalents, synthesis and breakdown of mitochondrial DNA, RNA, and proteins, heme biosynthesis, modulation of the nucleotide and deoxynucleotide pools in the mitochondrial matrix, and apoptotic cell death (Palmieri

F et al 2013). In yeast, 35 mitochondrial carrier proteins (MCPs) have been identified by genome analysis (Palmieri L et al 2000).

Mtm1 is a one of the mitochondrial carrier proteins localized in the mitochondrial inner membrane. Initially, it was identified as a Manganese Trafficking factor for

Mitochondrial Superoxide dismutase (Sod2) and proposed that to function as a manganese chaperone (Luk E et al 2003). However, later on, this function was revised since Mtm1 plays a role in mitochondrial Fe–S cluster biosynthesis and iron homeostasis.

First of all, it was shown that mtm1 deletion results in the inactivation of Sod2, while

Sod2 activity is restored in mtm1Δ mutants by adding back manganese into the growth media. However, mtm1∆ mutants have no manganese depletion phenotype and conversely, they show a slight increase in the amount of mitochondrial manganese (Luk

E et al 2003). Furthermore, interestingly, elevated iron levels were found in mtm1∆ mitochondria. Another paper published by the same group also demonstrated that decreased Sod2 protein activity in mtm1∆ mutants is because of misincorporation of iron into Sod2 protein rather than manganese. Therefore, as iron homeostasis is impaired in the case of inactivation of Mtm1, this leads to iron accumulation in a reactive form in mitochondria which likely competes with manganese for binding to Sod2, and inactivates the enzyme in yeast cells (Yang M et al 2006). In addition, Mtm1 depletion caused the activation of the iron regulator Aft1 but no major defects on cytosolic Fe-S protein Leu1

(isopropylmalate isomerase) and mitochondrial Fe-S protein Aco1 (Aconitase). This study suggested that deactivation of Mtm1 may change the levels of a particular metabolite within the mitochondrial matrix that helps to stabilize iron in a bioavailable form or Mtm1 may exchange an element that directly or indirectly influences iron homeostasis.

Another mitochondrial inner membrane carrier protein, Yhm2 was identified as a multicopy suppressor of temperature-sensitive growth defect displayed at 37 °C by cells lacking ABF2, which is involved in mitochondrial DNA replication and recombination. It was shown that YHM2 is associated with mitochondrial DNA in vivo and shows general

DNA-binding activity in vitro (Cho JH et al 1998). Later on, the Palmieri lab characterized Yhm2 as a citrate-oxoglutarate carrier protein. In addition to transporting citrate and oxoglutarate, Yhm2 can also transport oxaloacetate, succinate, and fumarate to a lesser extent, but not malate and isocitrate (Castegna A et al 2010). Furthermore, they showed that yhm2∆ cells have a growth defect upon exposure to hydrogen peroxide, decreased cytosolic NADPH/NADP + and GSH/GSSG ratios as well as an increased

NADPH/NADP + ratio in mitochondria. Based on these findings, it was proposed that a primary physiological function of Yhm2 is to increase the cytosolic NADPH reducing power that is essential for thiol-dependent peroxidases and to serve as an important component of the citrate-oxoglutarate NADPH redox shuttle between the cytosol and mitochondria (Castegna A et al 2010).

In mammalian cells, two members of the superfamily of anion carriers in the mitochondrial inner membrane, dicarboxylate carrier (DIC) and 2-oxoglutarate carrier

(OGC) that is the mammalian homolog of Yhm2 have been shown to catalyze uptake of

GSH into the matrix in rat renal cortex cells (Chen Z et al 1998). Because GSH is a negatively charged molecule at physiological pH, anion carriers in the mitochondrial inner membrane, DIC and OGC might be good candidates for transporting GSH. The same group also showed that DIC and OGC carriers can transport cytoplasmic GSH into the mitochondrial matrix in rat kidney cortex cells (Chen Z et al 2000). Moreover, OGC

91 has shown to interact with the central regulator of the intrinsic apoptotic cascade Bcl-2 to increase the mitochondrial GSH pool in primary cerebellar granule neurons (Wilkins HM et al 2012). The Linseman group also pointed out that overexpression of OGC in motor neuronal cells resulted in an increase in mitochondrial GSH and increased resistance to oxidative stress (Wilkins HM et al 2014). On the contrary, the Murphy group showed that overexpression of the dicarboxylate and 2-oxoglutarate carriers in Lactococcus lactis cells did not lead to GSH uptake or transport since excess GSH had no effect on the transport rate of substrates (Booty LM et al 2015).

Here we show that the deletion of MTM1 leads to lower mitochondrial GSH levels and large shift in the IMS redox state. However, deletion of YHM2 influences mitochondrial redox status in a smaller extent and leads to decrease in mitochondrial

GSH levels. Dysfunction of these two proteins impact GSH:GSSG pools in mitochondria, and this impact varies in mtm1∆/yhm2∆ mutants depending on their specific transporter functions.

EXPERIMENTAL PROCEDURES

Yeast Strains, Media, and Growth Conditions — Saccharomyces cerevisiae strains used were BY4741 ( MATa his3∆1 leu2∆0 met15∆0 ura3∆0 ), yhm2∆::kanMX4, and mtm1∆::kanMX4 obtained from Open Biosystems. Yeast transformations were performed using the lithium acetate procedure. Strains were maintained at 30 °C on synthetic media

(SC) supplemented with 2% glucose with the appropriate amino acids.

Plasmids — roGFP2 sensors used in this study included pZX201 (cytosol-roGFP2 LEU2 -

YIp) (Xue, 2011), pJH602 (IMS-roGFP2 LEU2 -YIp) (Hu, 2010) and pJH601 (matrix-

92 roGFP2 LEU2 -YIp) (Hu, 2010). IMS Grx1-roGFP2 (p416-b2-Grx1-roGFP2, URA3 -

CEN ), matrix Grx1-roGFP2 (p416-Su9-Grx1-roGFP2, URA3 -CEN ), and cytosol Grx1- roGFP2 (p416-Grx1-roGFP2, URA3 -CEN ) were obtained from Jan Riemer (Universität

Kaiserslautern, Germany).

Cytosolic and mitochondrial GSH measurements — Yeast cells were grown aerobically to mid-log phase in YP medium with 2% galactose. Mitochondrial and post- mitochondrial supernatant (PMS) fractions were obtained as previously described by converting cells to spheroplasts followed by gentle lysis by Dounce homogenization and differential centrifugation (Daum G et al 1982). Total GSH was measured by

GSH/GSSG-Glo TM Kit (Promega, USA) following the manufacturer’s protocol with slight modifications. 1:20 dilution for PMS extracts (5µl extract for the assay) and 1:5 dilution for mitochondrial extracts (10µl extract for the assay) were used. See Chapter 5 for the detailed protocol. This assay was performed by C. Outten group member Dr.

Vidyadhar Daithankar.

Redox Western Technique - See Chapter 1.

RESULTS

Deletion of Mtm1 affects the redox state of the IMS. In order to identify membrane transporters that influence GSH levels in mitochondria, our lab member, Dr. Vidyadhar

Daithankar, screened strains containing deletions of mitochondrial membrane transport genes from our collection of yeast deletion strains (Open Biosystems). These mutants include the VDAC homologues, outer membrane TOM proteins and inner membrane

TIM proteins, members of the mitochondrial carrier family (including the yeast

93 homologues of OGC and DIC), ATP-binding cassette transporters, and P-type ATPases.

Based on this screen, he found that mtm1∆ and yhm2∆ mutants have decreased levels of

GSH in mitochondria (Figure 4.1-B). Since the human homolog of Yhm2 has been shown to transport GSH into the mitochondrion, we wanted to determine how mtm1∆ and yhm2∆ mutants containing significantly low GSH in their mitochondria influence the subcellular GSH:GSSG redox state in the cytosol, mitochondrial matrix, and IMS using the modified fluorescence protein roGFP2 and glutaredoxin (Grx)-fused roGFP2 sensors.

In order to observe the redox state of roGFP2, the free thiols were capped with N- ethylmaleimide (NEM) to prevent oxidation during sample preparation. The reduced and oxidized proteins have different mobilities by non-reducing SDS-PAGE, thus, the relative ratio of oxidized to reduced roGFP2 was determined. Even though mtm1∆ and yhm2∆ cells have considerably low GSH levels in mitochondria, the matrix GSH:GSSG redox state measured in these mutants by using roGFP2 and Grx1-roGFP2 sensors was not more oxidized. It is reasonable to expect a more oxidized GSH:GSSG redox state in the matrix due to low mitochondrial GSH seen in mtm1∆ and yhm2∆ cells since the mitochondrial matrix constitutes the major portion of the mitochondrial volume. We observed a large difference in the IMS redox state when MTM1 was deleted. IMS- roGFP2 is more oxidized in mtm1∆ cells (41% oxidized compared to 6% oxidized for

WT), but no significant differences were measured in matrix and cytosol roGFP2 (Figure

4.2-B). However, it is interesting to note that roGFP2 is not well as expressed in the matrix and IMS in mtm1∆, making it difficult to get a reliable reading (Figure 4.2-A). In contrast, there is a little change in the IMS, matrix and cytosol roGFP2 responses upon deletion of the YHM2 gene (Figure 4.2-B).

We also performed redox measurements using Grx1-roGFP2 sensors since these sensors equilibrate more rapidly with GSH:GSSG pools in subcellular compartments such as the IMS that do not contain GRXs (Meyer A et al 2010). Consistent with the roGFP2 response in mtm1∆ and yhm2∆ cells, we found that only mtm1∆ possesses a highly oxidized IMS GSH:GSSG redox state (20.6% oxidized Grx1-roGFP2 compared to

7.5% oxidized for WT) versus yhm2∆ strain and its corresponding WT (Figure 4.3-B).

However, when the YHM2 gene was deleted, we did not observe a difference in the % oxidized IMS and cytosol Grx1-roGFP2, but found a slight increase in the matrix

GSH:GSSG redox state (Figure 4.3-B).

Cytosolic GSH 40

0 WT yhm2 mtm1

Mitochondrial GSH

0 WT yhm2 mtm1

Figure 4.1 Total GSH values in cytosolic and mitochondrial extracts of mtm1∆ and yhm2∆ strains. WT, mtm1∆ and yhm2∆ strains were grown in 5 ml YPGalactose media overnight at 30 ºC until mid-log phase and this overnight culture was reinoculated into 50 or 100 mL YPGalactose media until they reach mid-log phase. The reported values are the mean + standard deviation of 2-5 independent experiments. These experiments were performed by Vidyadhar Daithankar.

Figure 4. 2 roGFP2 redox response in WT, mtm1∆, and yhm2∆ strains. WT, mtm1∆, and yhm2∆ yeast cells expressing cytosol (pZX201), IMS (pJH602), and matrix (pJH601)-roGFP2 were grown to mid-log phase in selective –Leu SC glucose media. (A) Redox western blot of the samples separated by non-reducing SDS-PAGE and immunoblotted with anti-GFP antibodies. (B) Percent oxidized roGFP2 was quantified using an Odyssey Infrared Imaging System. Error bars represent standard deviation for 2-3 independent measurements.

Figure 4.3 Grx1 -roGFP2 redox response in W T, mtm1∆, and yhm2∆ strains. WT, mtm1∆, and yhm2∆ yeast cells expressing cytosol p416-Grx1-roGFP2, matrix p416-

Su9-Grx1-roGFP2, and IMS p416-b2-Grx1-roGFP2 were grown to mid-log phase in selective –Ura SC glucose media. (A) Redox western blot of the samples separated by non-reducing SDS-PAGE and immunoblotted with anti-GFP antibodies. (B) Percent oxidized roGFP2 quantified using an Odyssey Infrared Imaging System. Error bars represent the standard deviation for 2 independent experiments.

DISCUSSION

As described before, the S. cerevisiae Mtm1 protein was identified as a member of the mitochondrial carrier family of transporters (Luk E et al 2003). mtm1∆ cells possess inactive Sod2 protein in the mitochondrial matrix due to misincorporation of Fe instead of Mn in the active site of Sod2 caused by disruption of mitochondrial iron homeostasis (Luk E et al 2003 and Yang M et al 2006). Dysregulation of iron homeostasis, in turn, is due to accumulation of mitochondrial iron by activated of the iron regulator Aft1 (Yang M et al 2006). In addition, it was proposed that mtm1 mutations affect the Fe-S cluster biogenesis pathway since this strain has similar phenotypes to grx5∆ and ssq1∆ mutants, including defects in Fe-S cluster synthesis leading to induction of the Aft1 iron regulon and high mitochondrial iron (Yang M et al

2006), loss of Sod2 activity (Yang M et al 2006) and induction of Isu1 expression

(Naranuntarat A et al 2009). In addition, misincorporated Fe was shown to originate from the Fe-S cluster donated by the Fe-S scaffold protein Isu1 (Nilsson R et al 2009). The

Culotta group suggested that mtm1∆ cells may counteract Fe-S cluster defects by increasing Isu1 protein levels (Naranuntarat A et al 2009) (Figure 4.4-A).

We found that there is an increase in % oxidized roGFP2 in the IMS in mtm1∆ cells grown in the fermentable carbon source glucose. In these conditions, since cells rely on anaerobic glycolysis for energy, mitochondrial function is repressed. However, within the matrix, only slight decrease in oxidized roGFP2 was observed in mtm1∆ cells with no change in the cytosolic roGFP2 response. Recent studies showed that glutaredoxin

(GRX)-fused roGFP2 sensors more efficiently equilibrate with local GSH:GSSG pools.

Fusion of GRX to roGFP overcomes kinetic limitations due to the absence of GRXs in

99 subcellular compartments (Gutscher M et al 2008 and Meyer A et al 2010). When we used Grx1-roGFP2 sensors, mtm1∆ cells showed mostly oxidized IMS redox state (21% oxidized comparing to 7% oxidized for WT) parallel to roGFP2 results suggesting that mtm1 deletion only impacts the IMS GSH:GSSG pool. Further studies might include measuring subcellular redox state under respiratory growth conditions such as growing the cells in media supplemented with galactose, a partially fermentable carbon source in which the cells are more dependent on mitochondrial function. Nevertheless, mitochondrial GSH levels are severely reduced in mtm1∆ cells, when the cells are grown in galactose-supplemented media. Contrary to mitochondrial GSH levels, mtm1∆ cells does not possess greater % oxidized roGFP2 in the matrix even though the redox state of roGFP2 is reflected by total GSH and the GSH/GSSG ratio. Therefore, GSH measurement studies in mtm1∆ cells suggest that the reason of the iron accumulation may be lower mitochondrial GSH levels in mtm1∆ cells (Figure 4.1) since mutants known to accumulate mitochondrial iron such as gsh1∆ or erv1-1 cells have decreased mitochondrial GSH with the induction of Aft1. As a mitochondrial transport carrier,

Mtm1 exchanges a solute between the matrix and the IMS which may affect iron homeostasis or redox equilibrium, which may strongly influence the IMS redox state rather than the matrix GSH:GSSG pool. In addition, a recently published study demonstrated that mtm1 deletion diminishes Sod2 activity in a manner that does not involve Fe misincorporation. Most of the accumulated Fe in mtm1∆ is present as Fe III oxyhydroxide nanoparticles, which is unlikely to misincorporate into apo-Sod2, suggesting an alternative cellular function for Mtm1 (Park J et al 2013).

100

Another protein that studied in this Chapter is Yhm2 serving as a citrate- oxoglutarate carrier (Castegna A et al 2010). We found that YHM2 deletion causes a slight increase in the % oxidized matrix-roGFP2 sensors and no change in the response of cytosolic roGFP2 sensors. This might be explained by the fact that decreased level of mitochondrial GSH in yhm2∆ cells (Figure 4.1) causes an increase in oxidized roGFP2 response in the matrix, but still the impact is not as significant as we expected.

Furthermore, the mmmalian homolog of Yhm2 (OGC) was shown to transport GSH into the matrix and over-expression of the OGC in neuronal cells increased mitochondrial

GSH levels and oxidative stress resistance. This is consistent with the GSH measurements of yhm2∆ cells suggesting Yhm2 may transport GSH into the matrix

(Figure 4.4-B). A previous study using acetate-supplemented media for yhm2∆ cells showed the decreased level of cytosolic NADPH/NADP + and GSH/GSSG ratios suggesting elevated levels of GSSG and decreased levels of NADPH in the cytosol.

These effects are more apparent when cells are exposed to hydrogen peroxide (Castegna

A et al 2010). However, we did not observe any changes in the cytosolic GSH:GSSG redox state or cytosolic GSH levels. Since we used glucose containing media for the measurements, it would be interesting to utilize acetate containing media because cells lacking both IDP2 and ZWF1 (i.e. major sources of cytosolic NADPH) display a dramatic growth defect due to increased levels of intracellular oxidants produced during acetate metabolism (Minard KI et al 2001 and Minard KI et al 1999).

Taken together, our studies indicate that mtm1∆ and yhm2∆ cells affect GSH metabolism in mitochondria and the redox status of the IMS or mitochondrial matrix depending on the carbon source in the media. On the other hand, mtm1∆ mutants have the

101 largest change in IMS redox state than the other mutant yhm2∆. This effect might be due to fact that Mtm1 exchanges a glutathione-conjugated metabolite between the matrix and the IMS affecting cellular redox equilibrium.

102

Figure 4.4 Proposed m odel pathways for mtm1∆ and yhm2∆ cells. A) Working model of mitochondrial Fe-S cluster biogenesis in mtm1∆ cells. Mtm1 is involved in activation of mitochondrial Sod2 by facilitating insertion of essential manganese cofactor. Therefore, deletion of Mtm1 causes inactive Sod2. mtm1∆ cells accumulate high iron which is activated by iron sensing transcriptional factor Aft1. In addition, Fe binds Sod2 blocking late stages of iron-sulfur cluster biogenesis which is assisted by the upregulation of Isu1 protein level. In this study, it was shown that mtm1∆ cells have severely decreased level of mitochondrial GSH. Concerning its low level of GSH in its mitochondria, it might provide glutathione-c onjugated Fe–S clusters or other glutathione persulfide forms to the Atm1 which helps to activate Aft1. B) Alterations in mitochondria of yhm2∆ cells. Yhm2 transports citrate and oxoglutarate between mitochondrial matrix and cytosol. yhm2∆ cells show decreased cytosolic

NADPH/NADP + and GSH/GSSG ratios as well as an increased NADPH/NADP + ratio in mitochondria. Our study showed that yhm2∆ cells have decreased level of mitochondrial GSH suggesting that it might transport GSH between mitochondria and cytosol since its human homologs have been shown to transport cytoplasmic GSH into the mitochondria l matrix. 103

CHAPTER 5

SUPPLEMENTARY METHODS

104

INTRODUCTION

This chapter includes detailed descriptions of three methods developed during the research project. The whole cell GSH measurement is a combination of the temperature sensitive strains growth protocol and GSH assay and is modified from Promega

GSH/GSSG-Glo TM Assay. Blocking protocol for Gsh1 protein is provided by Dr. Joseph

J. Barycki (Department of Biochemistry and the Redox Biology Center, University of

Nebraska, Lincoln, Nebraska) and modified for Erv1 and Gsh1 proteins. RT-PCR protocol was derived from the protocol of Dr. Erin Connolly’s lab (Department of

Biological Sciences, University of South Carolina, Columbia) and the protocol of Dr.

Hexin Chen’s lab (Department of Biological Sciences, University of South Carolina,

Columbia).

Preparing temperature-sensitive mutant strains for whole cell GSH measurements by using the Glo-kit (Promega GSH/GSSG-Glo™ Assay)

1- Grow fresh cells on SC plates (along with the WT strain) anaerobically for 2 or 3

days at 24 °C room.

2- Start the WT strain and temperature-sensitive mutant strains at OD 0.02 in 15 ml

media in 50-ml tube except erv1-1. Start erv1-1 strain at OD 0.04. Grow them at

24 °C overnight, shaking at 250 rpm.

3- Next morning, check OD and make 1:2 or 1:3 dilution with fresh media for 24 °C

and 37 °C set of cells and grow them for 5 hours.

4- Five hours later, check OD and transfer 4-5×10 5 cells into a microcentrifuge tube.

5- Spin down the pellet at 12k rpm for 20 sec.

105

6- Remove supernatant (media) carefully with the pipette and add 50 µl 1X Lysis

buffer supplied by the kit. Suspend the cell pellet with lysis buffer several times

and transfer it to a new microcentrifuge tube.

7- Add 20 µl glass beads and lyse them in bead-beater machine (cold room) for 4

min.

8- Pipette 10 or 20 µl lysate into the well of the 96-well plate (white-solid plates)

and add 15 or 5 µl sterile water up to 25 µl (Total extract volume should be max

25 µl).

9- Add 25 µl total glutathione lysis reagent and shake 5 min in plate reader.

10- Add 50 µl luciferin generation reagent and incubate 30 min.

11- Add 100 µl luciferin detection reagent and incubate another 15 min and record the

luminescence values immediately.

Total G lutath ion e L ys is Bu ffe r µL per reaction reaction number (n) µL per n reaction Luciferin-NT 1 n 1 × n Passive lysis B 10 n 10 × n Water 14 n 14 × n

Luciferin Generation Reagent µL per reaction reaction number (n) µL per n reaction DTT(100 mM) 1.25 n 1.25 × n Glutathione-S-transf 3 n 3 × n Glutathione Reaction B 45.75 n 45.75 × n

Luciferin Detection Reagent (100 µl /well)

106

Blocking Conditions for Gsh1 and Erv1 proteins

1- Transferred blot is soaked in Odyssey blocking buffer or 5% milk in PBS-T,

O/N at 4°C

2- Blot is washed 4 times × 5 min with PBS-T

3- Primary antibody is diluted to 1/10000 (1µl) in 1% non-fat milk PBS-T for Gsh1

protein and in 5% non-fat milk PBS-T for Erv1 protein and incubated at RT for

1h.

4- Blot is washed 3 times × 15 min with PBS-T

5- Secondary Ab is diluted to 1/25000 (0.4 µl) in 5% non-fat milk PBS-T and

incubated at RT for 1 h.

6- Blot is washed for 10 min × 3 times

7- Blot is rinsed with PBS buffer and ready to scan.

RT-PCR Protocol

Preparation of total RNA by using PureLink® RNA Mini Kit from Invitrogen

1) Grow fresh cells on an SC plate (along with the WT strain) anaerobically for 2 or

3 days at 24 °C room.

2) Start the WT strain and temperature-sensitive mutant strains at OD 0.02 in 15 ml

media in 50-ml tube except erv1-1. Start erv1-1 strain at OD 0.04. Grow them at

24 °C overnight, shaking at 250 rpm.

3) Next morning, check OD and make 1:2 or 1:3 dilution with fresh media for 24 °C

and 37 °C set of cells and grow them for 5 hours.

107

4) Five hours later, check OD and transfer 4×10 8 cells to centrifuge tubes and

centrifuge 500 × g for 5 min at 4 °C. Discard the supernatant.

5) Add up to 100 μL of Zymolase digestion buffer (20 mg/ml) to the yeast cell

pellet.

6) Incubate for 30–60 minutes at 24 °C.

7) Add 200 μL Lysis Buffer prepared with DTT (Prepare fresh amount of lysis

buffer containing 40 mM DTT for each purification procedure. Add 20 µl of 2 M

DTT for each 1 mL of lysis buffer. Use RNAse-free water to prepare DTT)

8) Vortex to mix thoroughly.

9) Centrifuge at 12,000 × g for 2 minutes at room temperature to pellet cells.

10) Transfer the supernatant to a clean RNase-free microcentrifuge tube.

11) Proceed to Binding, Washing, and Elution (Follow the manufacturer’s protocol,

page 40)

12) Store your purified RNA at -80 °C or proceed to DNase I treatment.

DNase I treatment (Applied Biosystems Turbo DNase; AM1907)

1- Mix the following in a 0.5 ml Eppendorf tube and incubate for 30 min at 37 °C:

RNA (10µg total RNA)

10 × TURBO DNase buffer 2 µl

TURBO DNase I 1 µl

Volume up to 20 µl with RNase-free water

2- Add 2 µl of DNase Inactivation Reagent that has been resuspended and mix well.

Incubate at room temperature for 5 min

108

3- Spin at 11,000 rpm for 1.5 min. Transfer RNA to a new tube, making sure not to

transfer any of the DNase Inactivation Reagent. Use directly in cDNA reaction. cDNA synthesis using M-MLV RT kit from Invitrogen (28025-013)

1- Add the following components to nuclease-free microcentrifuge tube (always

prepare one more than sample’s #) as below;

3 µl oligo (dT) 12-18

1 µg total RNA (Calculate how many µl is needed for 1 µg RNA in each sample)

1 µl 10 mM dNTP Mix,

Volume up to 12 µl with DEPC-treated water

2- Heat mixture to 65 °C for 5 min and quick chill on ice. Collect the contents of the

tube by brief centrifugation and add:

4 µl 5X first-strand buffer

2 µl 0.1 M DTT

1 µl RNaseOut Recombinant ribonuclease inhibitor (40 units/ µl)

3- Mix contents of tube by pipetting up and down gently and incubate at 37 °C for 2

min

4- Add 1 µl (200 units) of M-MLV RT and mix by pipetting gently up and down

5- Incubate tubes at 37 °C for 50 min

6- Terminate the reaction by incubating tubes at 70 °C for 15 min

7- Dilute each cDNA to 200 µl by adding 180 µl DEPC water (Total cDNA volume

is 20 µl)

8- Use 2 µl of cDNA for RT-PCR experiment

109

Setting up real time reactions

1. Determine how many reactions you are doing (n) and make up a master mix

for n+1. Perform reactions in triplicate. Make up SYBR green master mix and

aliquot it out before touching your cDNA samples.

2. Include the following controls: Negative controls

A. No-template control. (use 2 µl DEPC water instead of cDNA)

B. No-reverse-transcriptase control.

For SYBR Green Master Mix (1 rxn: 25 µl)

2× SYBR Green Supermix 12.5 µl

Forward primer (10 µM) 1 µl

Reverse primer (10 µM) 1 µl

DEPC-treated H 2O 8.5 µl

cDNA 2 µl

3. Aliquot 23 µl of master mix into each well of 96 well plate.

4. Aliquot 2 µl cDNA/2 µl of H 2O or no Reverse Transcriptase controls into

each well.

5. Put sealing tape on plate and secure with roller. Tear off the white strips that

remain on the sides of the tape. Avoid touching the surface of the sealing tape

with gloved fingers.

6. Spin plate for 1 min at 800 rpm in centrifuge.

110

Thermal Protocol for RT-PCR

95 °C 2 min

95 °C 15 sec X 50 cycles 60 °C 30 sec

7. Place the plate in the thermal cycler and start the program.

8. Run the PCR reaction to determine C T values for each sample. (It is important

to have C T values for the target sequence and the housekeeping gene for each

sample.)

9. Calculate the relative quantification of target gene by the 2-ΔΔCT method

(Livak & Schmittgen, 2001) using the ACT1 gene (actin) as a control.

111

REFERENCES

Aloria, K., Schilke, B., Andrew, A., and Craig, E. A. (2004) Iron-induced oligomerization of yeast frataxin homologue Yfh1 is dispensable in vivo. EMBO Rep. 5, 1096–101.

Anderson, C. P., Shen, M., Eisenstein, R. S., and Leibold, E. A. (2012) Mammalian iron metabolism and its control by iron regulatory proteins. Biochim. Biophys. Acta . 1823 , 1468–83.

Banci, L., Bertini, I., Cefaro, C., Ciofi-Baffoni, S., Gallo, A., Martinelli, M., Sideris, D. P., Katrakili, N., and Tokatlidis, K. (2009) MIA40 is an oxidoreductase that catalyzes oxidative protein folding in mitochondria. Nat. Struct. Mol. Biol. 16 , 198–206.

Banci, L., Bertini, I., Ciofi-Baffoni, S., Boscaro, F., Chatzi, A., Mikolajczyk, M., Tokatlidis, K., and Winkelmann, J. (2011) Anamorsin is a [2Fe-2S] cluster-containing substrate of the Mia40-dependent mitochondrial protein trapping machinery. Chem. Biol. 18 , 794–804.

Banci, L., Barbieri, L., Luchinat, E., and Secci, E. (2013) Visualization of redox- controlled protein fold in living cells. Chem. Biol. 20 , 747–52.

Becher, D., Kricke, J., Stein, G., and Lisowsky, T. (1999) A mutant for the yeast scERV1 gene displays a new defect in mitochondrial morphology and distribution. Yeast . 15 , 1171–81.

Bien, M., Longen, S., Wagener, N., Chwalla, I., Herrmann, J. M., and Riemer, J. (2010) Mitochondrial disulfide bond formation is driven by intersubunit electron transfer in Erv1 and proofread by glutathione. Mol. Cell . 37 , 516–28.

Bihlmaier, K., Mesecke, N., Terziyska, N., Bien, M., Hell, K., and Herrmann, J. M. (2007) The disulfide relay system of mitochondria is connected to the respiratory chain. J. Cell Biol. 179 , 389–95.

116

Biterova, E. I., and Barycki, J. J. (2009) Mechanistic details of glutathione biosynthesis revealed by crystal structures of Saccharomyces cerevisiae glutamate cysteine ligase. J. Biol. Chem. 284 , 32700–8.

Boisnard, S., Lagniel, G., Garmendia-Torres, C., Molin, M., Boy-Marcotte, E., Jacquet, M., Toledano, M. B., Labarre, J., and Chédin, S. (2009) H2O2 activates the nuclear localization of Msn2 and Maf1 through thioredoxins in Saccharomyces cerevisiae. Eukaryot. Cell . 8, 1429–38.

Booty, L. M., King, M. S., Thangaratnarajah, C., Majd, H., James, A. M., Kunji, E. R. S., and Murphy, M. P. (2015) The mitochondrial dicarboxylate and 2-oxoglutarate carriers do not transport glutathione. FEBS Lett. 589 , 621–8.

Böttinger, L., Gornicka, A., Czerwik, T., Bragoszewski, P., Loniewska-Lwowska, A., Schulze-Specking, A., Truscott, K. N., Guiard, B., Milenkovic, D., and Chacinska, A. (2012) In vivo evidence for cooperation of Mia40 and Erv1 in the oxidation of mitochondrial proteins. Mol. Biol. Cell . 23 , 3957–69.

Bragoszewski, P., Gornicka, A., Sztolsztener, M. E., and Chacinska, A. (2013) The ubiquitin-proteasome system regulates mitochondrial intermembrane space proteins. Mol. Cell. Biol. 33 , 2136–48.

Camaschella, C., Campanella, A., De Falco, L., Boschetto, L., Merlini, R., Silvestri, L., Levi, S., and Iolascon, A. (2007) The human counterpart of zebrafish shiraz shows sideroblastic-like microcytic anemia and iron overload. Blood . 110, 1353–8.

Camier, S., Ma, E., Leroy, C., Pruvost, A., Toledano, M., and Marsolier-Kergoat, M.-C. (2007) Visualization of ribonucleotide reductase catalytic oxidation establishes thioredoxins as its major reductants in yeast. Free Radic. Biol. Med. 42 , 1008–16.

Carmel-Harel, O., Stearman, R., Gasch, A. P., Botstein, D., Brown, P. O., and Storz, G. (2001) Role of thioredoxin reductase in the Yap1p-dependent response to oxidative stress in Saccharomyces cerevisiae. Mol. Microbiol. 39 , 595–605.

Castegna, A., Scarcia, P., Agrimi, G., Palmieri, L., Rottensteiner, H., Spera, I., Germinario, L., and Palmieri, F. (2010) Identification and functional characterization of a novel mitochondrial carrier for citrate and oxoglutarate in Saccharomyces cerevisiae. J. Biol. Chem. 285 , 17359–70.

117

Chacinska, A., Pfannschmidt, S., Wiedemann, N., Kozjak, V., Sanjuán Szklarz, L. K., Schulze-Specking, A., Truscott, K. N., Guiard, B., Meisinger, C., and Pfanner, N. (2004) Essential role of Mia40 in import and assembly of mitochondrial intermembrane space proteins. EMBO J. 23 , 3735–46.

Chacinska, A., Guiard, B., Müller, J. M., Schulze-Specking, A., Gabriel, K., Kutik, S., and Pfanner, N. (2008) Mitochondrial biogenesis, switching the sorting pathway of the intermembrane space receptor Mia40. J. Biol. Chem. 283 , 29723–9.

Chen, O. S., Crisp, R. J., Valachovic, M., Bard, M., Winge, D. R., and Kaplan, J. (2004) Transcription of the yeast iron regulon does not respond directly to iron but rather to iron- sulfur cluster biosynthesis. J. Biol. Chem. 279 , 29513–8.

Chen, Z., and Lash, L. H. (1998) Evidence for mitochondrial uptake of glutathione by dicarboxylate and 2-oxoglutarate carriers. J. Pharmacol. Exp. Ther. 285 , 608–18.

Chen, Z., Putt, D. A., and Lash, L. H. (2000) Enrichment and functional reconstitution of glutathione transport activity from rabbit kidney mitochondria: further evidence for the role of the dicarboxylate and 2-oxoglutarate carriers in mitochondrial glutathione transport. Arch. Biochem. Biophys. 373 , 193–202.

Cho, J. H., Ha, S. J., Kao, L. R., Megraw, T. L., and Chae, C. B. (1998) A novel DNA- binding protein bound to the mitochondrial inner membrane restores the null mutation of mitochondrial histone Abf2p in Saccharomyces cerevisiae. Mol. Cell. Biol. 18 , 5712–23.

Collinson, E. J., Wheeler, G. L., Garrido, E. O., Avery, A. M., Avery, S. V, and Grant, C. M. (2002) The yeast glutaredoxins are active as glutathione peroxidases. J. Biol. Chem. 277 , 16712–16717.

Collinson, L. P., and Dawes, I. W. (1995) Isolation, characterization and overexpression of the yeast gene, GLR1, encoding glutathione reductase. Gene . 156 , 123–7.

Crooks, D. R., Jeong, S. Y., Tong, W.-H., Ghosh, M. C., Olivierre, H., Haller, R. G., and Rouault, T. A. (2012) Tissue specificity of a human mitochondrial disease: differentiation-enhanced mis-splicing of the Fe-S scaffold gene ISCU renders patient cells more sensitive to oxidative stress in ISCU myopathy. J. Biol. Chem. 287 , 40119–30.

118

Cubitt, A. B., Heim, R., Adams, S. R., Boyd, A. E., Gross, L. A., and Tsien, R. Y. (1995) Understanding, improving and using green fluorescent proteins. Trends Biochem. Sci. 20 , 448–55.

Curran, S. P., Leuenberger, D., Leverich, E. P., Hwang, D. K., Beverly, K. N., and Koehler, C. M. (2004) The role of Hot13p and redox chemistry in the mitochondrial TIM22 import pathway. J. Biol. Chem. 279 , 43744–51.

Dabir, D. V, Leverich, E. P., Kim, S.-K., Tsai, F. D., Hirasawa, M., Knaff, D. B., and Koehler, C. M. (2007) A role for cytochrome c and cytochrome c peroxidase in electron shuttling from Erv1. EMBO J. 26 , 4801–11.

Daithankar, V. N., Farrell, S. R., and Thorpe, C. (2009) Augmenter of liver regeneration: substrate specificity of a flavin-dependent oxidoreductase from the mitochondrial intermembrane space. Biochemistry . 48 , 4828–37.

Daum, G., Böhni, P. C., and Schatz, G. (1982) Import of proteins into mitochondria. Cytochrome b2 and cytochrome c peroxidase are located in the intermembrane space of yeast mitochondria. J. Biol. Chem. 257 , 13028–33.

Di Fonzo, A., Ronchi, D., Lodi, T., Fassone, E., Tigano, M., Lamperti, C., Corti, S., Bordoni, A., Fortunato, F., Nizzardo, M., Napoli, L., Donadoni, C., Salani, S., Saladino, F., Moggio, M., Bresolin, N., Ferrero, I., and Comi, G. P. (2009) The Mitochondrial Disulfide Relay System Protein GFER Is Mutated in Autosomal-Recessive Myopathy with Cataract and Combined Respiratory-Chain Deficiency. Am. J. Hum. Genet. 84 , 594– 604.

Dlouhy, A. (2015) Illuminating the Interactions and Functions of Glutaredoxins, BolA Proteins, and Erv1 in Iron Homeostasis. Ph.D. thesis, University of South Carolina.

Dooley, C. T., Dore, T. M., Hanson, G. T., Jackson, W. C., Remington, S. J., and Tsien, R. Y. (2004) Imaging Dynamic Redox Changes in Mammalian Cells with Green Fluorescent Protein Indicators. J. Biol. Chem. 279 , 22284–22293.

Dormer, U. H., Westwater, J., Stephen, D. W. S., and Jamieson, D. J. (2002) Oxidant regulation of the Saccharomyces cerevisiae GSH1 gene. Biochim. Biophys. Acta . 1576 , 23–9.

119

Draculic, T., Dawes, I. W., and Grant, C. M. (2000) A single glutaredoxin or thioredoxin gene is essential for viability in the yeast Saccharomyces cerevisiae. Mol. Microbiol. 36 , 1167–74.

Dumont, M. E., Ernst, J. F., and Sherman, F. (1988) Coupling of heme attachment to import of cytochrome c into yeast mitochondria. Studies with heme lyase-deficient mitochondria and altered apocytochromes c. J. Biol. Chem. 263 , 15928–37.

Durigon, R., Wang, Q., Ceh Pavia, E., Grant, C. M., and Lu, H. (2012) Cytosolic thioredoxin system facilitates the import of mitochondrial small Tim proteins. EMBO Rep. 13 , 916–22.

Gerber, J., Neumann, K., Prohl, C., Mühlenhoff, U., and Lill, R. (2004) The yeast scaffold proteins Isu1p and Isu2p are required inside mitochondria for maturation of cytosolic Fe/S proteins. Mol. Cell. Biol. 24 , 4848–57.

Gietz, R. D., Schiestl, R. H., Willems, A. R., and Woods, R. A. (1995) Studies on the transformation of intact yeast cells by the LiAc/SS-DNA/PEG procedure. Yeast . 11 , 355– 60.

Godon, C., Lagniel, G., Lee, J., Buhler, J. M., Kieffer, S., Perrot, M., Boucherie, H., Toledano, M. B., and Labarre, J. (1998) The H2O2 stimulon in Saccharomyces cerevisiae. J. Biol. Chem. 273 , 22480–9.

Guo, P.-C., Ma, J.-D., Jiang, Y.-L., Wang, S.-J., Bao, Z.-Z., Yu, X.-J., Chen, Y., and Zhou, C.-Z. (2012) Structure of yeast sulfhydryl oxidase erv1 reveals electron transfer of the disulfide relay system in the mitochondrial intermembrane space. J. Biol. Chem. 287 , 34961–9.

Gutscher, M., Pauleau, A.-L., Marty, L., Brach, T., Wabnitz, G. H., Samstag, Y., Meyer, A. J., and Dick, T. P. (2008) Real-time imaging of the intracellular glutathione redox potential. Nat. Methods . 5, 553–9.

120

Hagiya, M., Francavilla, A., Polimeno, L., Ihara, I., Sakai, H., Seki, T., Shimonishi, M., Porter, K. A., and Starzl, T. E. (1994) Cloning and sequence analysis of the rat augmenter of liver regeneration (ALR) gene: expression of biologically active recombinant ALR and demonstration of tissue distribution. Proc. Natl. Acad. Sci. U. S. A. 91 , 8142–6.

Hanson, G. T., Aggeler, R., Oglesbee, D., Cannon, M., Capaldi, R. A., Tsien, R. Y., and Remington, S. J. (2004) Investigating mitochondrial redox potential with redox-sensitive green fluorescent protein indicators. J. Biol. Chem. 279 , 13044–53.

Hell, K. (2008) The Erv1-Mia40 disulfide relay system in the intermembrane space of mitochondria. Biochim. Biophys. Acta . 1783 , 601–9.

Herrmann, J. M., and Riemer, J. (2012) Mitochondrial disulfide relay: redox-regulated protein import into the intermembrane space. J. Biol. Chem. 287 , 4426–33.

Hofmann, S., Rothbauer, U., Mühlenbein, N., Baiker, K., and Hell, K. (2005) Functional and mutational characterization of human MIA40 acting during import into the mitochondrial intermembrane space. J. Mol. Biol. 353 (3), 517-28.

Holmgren, A. (1989) Thioredoxin and glutaredoxin systems. J. Biol. Chem. 264 , 13963– 6.

Hu, J., Dong, L., and Outten, C. E. (2008) The redox environment in the mitochondrial intermembrane space is maintained separately from the cytosol and matrix. J. Biol. Chem. 283 , 29126–34.

Hu, J. (2010) Investigating subcellular thiol redox chemistry with GFP-based redox sensors. Ph.D. thesis, University of South Carolina.

Kawamata, H., and Manfredi, G. (2008) Different regulation of wild-type and mutant Cu,Zn superoxide dismutase localization in mammalian mitochondria. Hum. Mol. Genet. 17 , 3303–17.

Kawano, S., Yamano, K., Naoé, M., Momose, T., Terao, K., Nishikawa, S., Watanabe, N., and Endo, T. (2009) Structural basis of yeast Tim40/Mia40 as an oxidative translocator in the mitochondrial intermembrane space. Proc. Natl. Acad. Sci. U. S. A. 106 , 14403–7.

121

Kispal, G., Csere, P., Guiard, B., and Lill, R. (1997) The ABC transporter Atm1p is required for mitochondrial iron homeostasis. FEBS Lett. 418 , 346–50.

Kispal, G., Csere, P., Prohl, C., and Lill, R. (1999) The mitochondrial proteins Atm1p and Nfs1p are essential for biogenesis of cytosolic Fe/S proteins. EMBO J. 18 , 3981–9.

Kojer, K., Bien, M., Gangel, H., Morgan, B., Dick, T. P., and Riemer, J. (2012) Glutathione redox potential in the mitochondrial intermembrane space is linked to the cytosol and impacts the Mia40 redox state. EMBO J. 31 , 3169–82.

Kojer, K., Peleh, V., Calabrese, G., Herrmann, J. M., and Riemer, J. (2015) Kinetic control by limiting glutaredoxin amounts enables thiol oxidation in the reducing mitochondrial intermembrane space. Mol. Biol. Cell . 26 , 195–204.

Kuhnke, G., Neumann, K., Mühlenhoff, U., and Lill, R. (2006) Stimulation of the ATPase activity of the yeast mitochondrial ABC transporter Atm1p by thiol compounds. Mol. Membr. Biol. 23 , 173–8.

Kumar, C., Igbaria, A., D’Autreaux, B., Planson, A.-G., Junot, C., Godat, E., Bachhawat, A. K., Delaunay-Moisan, A., and Toledano, M. B. (2011) Glutathione revisited: a vital function in iron metabolism and ancillary role in thiol-redox control. EMBO J. 30 , 2044– 56.

Kunji, E. R. S. (2004) The role and structure of mitochondrial carriers. FEBS Lett. 564 , 239–44.

Lange, H., Lisowsky, T., Gerber, J., Mühlenhoff, U., Kispal, G., and Lill, R. (2001) An essential function of the mitochondrial sulfhydryl oxidase Erv1p/ALR in the maturation of cytosolic Fe/S proteins. EMBO Rep. 2, 715–20.

Leary, S. C., Cobine, P. A., Kaufman, B. A., Guercin, G.-H., Mattman, A., Palaty, J., Lockitch, G., Winge, D. R., Rustin, P., Horvath, R., and Shoubridge, E. A. (2007) The human cytochrome c oxidase assembly factors SCO1 and SCO2 have regulatory roles in the maintenance of cellular copper homeostasis. Cell Metab. 5, 9–20.

122

Lee, J. C., Straffon, M. J., Jang, T. Y., Higgins, V. J., Grant, C. M., and Dawes, I. W. (2001) The essential and ancillary role of glutathione in Saccharomyces cerevisiae analysed using a grande gsh1 disruptant strain. FEMS Yeast Res. 1, 57–65.

Li, H., and Outten, C. E. (2012) Monothiol CGFS glutaredoxins and BolA-like proteins: [2Fe-2S] binding partners in iron homeostasis. Biochemistry . 51 , 4377–89.

Lisowsky, T. (1992) Dual function of a new nuclear gene for oxidative phosphorylation and vegetative growth in yeast. MGG Mol. Gen. Genet. 232 , 58–64.

Lisowsky, T. (1994) ERV1 is involved in the cell-division cycle and the maintenance of mitochondrial genomes in Saccharomyces cerevisiae. Curr. Genet. 26 , 15–20.

Livak, K. J., and Schmittgen, T. D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods . 25 , 402–8.

Longen, S., Woellhaf, M. W., Petrungaro, C., Riemer, J., and Herrmann, J. M. (2014) The disulfide relay of the intermembrane space oxidizes the ribosomal subunit mrp10 on its transit into the mitochondrial matrix. Dev. Cell . 28 , 30–42.

Lu, S. C. (2009) Regulation of glutathione synthesis. Mol. Aspects Med. 30 , 42–59.

Luikenhuis, S., Perrone, G., Dawes, I. W., and Grant, C. M. (1998) The yeast Saccharomyces cerevisiae contains two glutaredoxin genes that are required for protection against reactive oxygen species. Mol. Biol. Cell . 9, 1081–91.

Luk, E., Carroll, M., Baker, M., and Culotta, V. C. (2003) Manganese activation of superoxide dismutase 2 in Saccharomyces cerevisiae requires MTM1, a member of the mitochondrial carrier family. Proc. Natl. Acad. Sci. U. S. A. 100 , 10353–7.

Martinou, J.C., and Youle, R. J. (2011) Mitochondria in apoptosis: Bcl-2 family members and mitochondrial dynamics. Dev. Cell . 21 , 92–101.

123

Meister, A., and Anderson, M. E. (1983) Glutathione. Annu. Rev. Biochem. 52 , 711–60.

Mesecke, N., Terziyska, N., Kozany, C., Baumann, F., Neupert, W., Hell, K., and Herrmann, J. M. (2005) A disulfide relay system in the intermembrane space of mitochondria that mediates protein import. Cell . 121 , 1059–69.

Mesecke, N., Bihlmaier, K., Grumbt, B., Longen, S., Terziyska, N., Hell, K., and Herrmann, J. M. (2008) The zinc-binding protein Hot13 promotes oxidation of the mitochondrial import receptor Mia40. EMBO Rep. 9, 1107–13.

Meyer, A. J., and Dick, T. P. (2010) Fluorescent protein-based redox probes. Antioxid. Redox Signal. 13 , 621–50.

Minard, K. I., and McAlister-Henn, L. (1999) Dependence of peroxisomal beta-oxidation on cytosolic sources of NADPH. J. Biol. Chem. 274 , 3402–3406.

Minard, K. I., and McAlister-Henn, L. (2001) Antioxidant function of cytosolic sources of NADPH in yeast. Free Radic. Biol. Med. 31 , 832–43.

Mochel, F., Knight, M. A., Tong, W.-H., Hernandez, D., Ayyad, K., Taivassalo, T., Andersen, P. M., Singleton, A., Rouault, T. A., Fischbeck, K. H., and Haller, R. G. (2008) Splice mutation in the iron-sulfur cluster scaffold protein ISCU causes myopathy with exercise intolerance. Am. J. Hum. Genet. 82 , 652–60.

Morgan, B., Ezeriņa, D., Amoako, T. N. E., Riemer, J., Seedorf, M., and Dick, T. P. (2013) Multiple glutathione disulfide removal pathways mediate cytosolic redox homeostasis. Nat. Chem. Biol. 9, 119–25.

Muller, E. G. (1991) Thioredoxin deficiency in yeast prolongs S phase and shortens the G1 interval of the cell cycle. J. Biol. Chem. 266 , 9194–202.

Muller, E. G. (1994) Deoxyribonucleotides are maintained at normal levels in a yeast thioredoxin mutant defective in DNA synthesis. J. Biol. Chem. 269 , 24466–71.

124

Mühlenhoff, U., Gerber, J., Richhardt, N., and Lill, R. (2003) Components involved in assembly and dislocation of iron-sulfur clusters on the scaffold protein Isu1p. EMBO J. 22 , 4815–25.

Netz, D. J. A., Mascarenhas, J., Stehling, O., Pierik, A. J., and Lill, R. (2014) Maturation of cytosolic and nuclear iron-sulfur proteins. Trends Cell Biol. 24 , 303–12.

Müller, J. M., Milenkovic, D., Guiard, B., Pfanner, N., and Chacinska, A. (2008) Precursor oxidation by Mia40 and Erv1 promotes vectorial transport of proteins into the mitochondrial intermembrane space. Mol. Biol. Cell . 19 , 226–36.

Naranuntarat, A., Jensen, L. T., Pazicni, S., Penner-Hahn, J. E., and Culotta, V. C. (2009) The interaction of mitochondrial iron with manganese superoxide dismutase. J. Biol. Chem. 284 , 22633–40.

Netz, D. J. A., Stümpfig, M., Doré, C., Mühlenhoff, U., Pierik, A. J., and Lill, R. (2010) Tah18 transfers electrons to Dre2 in cytosolic iron-sulfur protein biogenesis. Nat. Chem. Biol. 6, 758–65.

Netz, D. J. A., Mascarenhas, J., Stehling, O., Pierik, A. J., and Lill, R. (2014) Maturation of cytosolic and nuclear iron-sulfur proteins. Trends Cell Biol. 24 , 303–12.

Nilsson, R., Schultz, I. J., Pierce, E. L., Soltis, K. A., Naranuntarat, A., Ward, D. M., Baughman, J. M., Paradkar, P. N., Kingsley, P. D., Culotta, V. C., Kaplan, J., Palis, J., Paw, B. H., and Mootha, V. K. (2009) Discovery of genes essential for heme biosynthesis through large-scale gene expression analysis. Cell Metab. 10 , 119–30.

Osman, C., Haag, M., Potting, C., Rodenfels, J., Dip, P. V., Wieland, F. T., Brügger, B., Westermann, B., and Langer, T. (2009) The genetic interactome of prohibitins: coordinated control of cardiolipin and phosphatidylethanolamine by conserved regulators in mitochondria. J. Cell Biol. 184 , 583–96.

Østergaard, H., Henriksen, A., Hansen, F. G., and Winther, J. R. (2001) Shedding light on disulfide bond formation: engineering a redox switch in green fluorescent protein. EMBO J. 20 , 5853–62.

125

Østergaard, H., Tachibana, C., and Winther, J. R. (2004) Monitoring disulfide bond formation in the eukaryotic cytosol. J. Cell Biol. 166 , 337–45.

Outten, C. E., and Culotta, V. C. (2003) A novel NADH kinase is the mitochondrial source of NADPH in Saccharomyces cerevisiae. EMBO J. 22 , 2015–24.

Outten, C. E., and Culotta, V. C. (2004) Alternative start sites in the Saccharomyces cerevisiae GLR1 gene are responsible for mitochondrial and cytosolic isoforms of glutathione reductase. J. Biol. Chem. 279 , 7785–91.

Outten, C. E., and Albetel, A.-N. (2013) Iron sensing and regulation in Saccharomyces cerevisiae: Ironing out the mechanistic details. Curr. Opin. Microbiol. 16 , 662–8.

Palmieri, F. (2013) The mitochondrial transporter family SLC25: identification, properties and physiopathology. Mol. Aspects Med. 34 , 465–84.

Palmieri, L., Runswick, M. J., Fiermonte, G., Walker, J. E., and Palmieri, F. (2000) Yeast Mitochondrial Carriers: Bacterial Expression, Biochemical Identification and Metabolic Significance. J. Bioenerg. Biomembr. 32 , 67–77.

Pandey, A., Gordon, D. M., Pain, J., Stemmler, T. L., Dancis, A., and Pain, D. (2013) Frataxin directly stimulates mitochondrial cysteine desulfurase by exposing substrate- binding sites, and a mutant Fe-S cluster scaffold protein with frataxin-bypassing ability acts similarly. J. Biol. Chem. 288 , 36773–86.

Parent, A., Elduque, X., Cornu, D., Belot, L., Le Caer, J.-P., Grandas, A., Toledano, M. B., and D’Autréaux, B. (2015) Mammalian frataxin directly enhances sulfur transfer of NFS1 persulfide to both ISCU and free thiols. Nat. Commun. 6, 5686.

Park, J., McCormick, S. P., Chakrabarti, M., and Lindahl, P. A. (2013) Insights into the iron-ome and manganese-ome of Δmtm1 Saccharomyces cerevisiae mitochondria. Metallomics . 5, 656–72.

126

Peleh, , V., Riemer, J., Dancis, A., and Herrmann, J.M. (2014) Protein oxidation in the intermembrane space of mitochondria is substrate-specific rather than general. Microb. Cell . 1, 81–93.

Penninckx, M. (2000) A short review on the role of glutathione in the response of yeasts to nutritional, environmental, and oxidative stresses. Enzyme Microb. Technol. 26 , 737– 742.

Perkins, A., Poole, L. B., and Karplus, P. A. (2014) Tuning of peroxiredoxin catalysis for various physiological roles. Biochemistry . 53 , 7693–705.

Petrungaro, C., and Riemer, J. (2014) Mechanisms and physiological impact of the dual localization of mitochondrial intermembrane space proteins. Biochem. Soc. Trans. 42 , 952–8.

Philpott, C. C., and Protchenko, O. (2008) Response to iron deprivation in Saccharomyces cerevisiae. Eukaryot. Cell . 7, 20–7.

Riemer, J., Fischer, M., and Herrmann, J. M. (2011) Oxidation-driven protein import into mitochondria: Insights and blind spots. Biochim. Biophys. Acta . 1808 , 981–9.

Rissler, M., Wiedemann, N., Pfannschmidt, S., Gabriel, K., Guiard, B., Pfanner, N., and Chacinska, A. (2005) The essential mitochondrial protein Erv1 cooperates with Mia40 in biogenesis of intermembrane space proteins. J. Mol. Biol. 353 , 485–92.

Rodríguez-Manzaneque, M. T., Tamarit, J., Bellí, G., Ros, J., and Herrero, E. (2002) Grx5 is a mitochondrial glutaredoxin required for the activity of iron/sulfur enzymes. Mol. Biol. Cell . 13 , 1109–21. Rutherford, J. C., Ojeda, L., Balk, J., Mühlenhoff, U., Lill, R., and Winge, D. R. (2005) Activation of the iron regulon by the yeast Aft1/Aft2 transcription factors depends on mitochondrial but not cytosolic iron-sulfur protein biogenesis. J. Biol. Chem. 280 , 10135–40.

127

Schaedler, T. A., Thornton, J. D., Kruse, I., Schwarzländer, M., Meyer, A. J., van Veen, H. W., and Balk, J. (2014) A conserved mitochondrial ATP-binding cassette transporter exports glutathione polysulfide for cytosolic metal cofactor assembly. J. Biol. Chem. 289 , 23264–74.

Schmucker, S., and Puccio, H. (2010) Understanding the molecular mechanisms of Friedreich’s ataxia to develop therapeutic approaches. Hum. Mol. Genet. 19 , R103–10.

Sideris, D. P., and Tokatlidis, K. (2010) Oxidative protein folding in the mitochondrial intermembrane space. Antioxid. Redox Signal. 13 , 1189–204.

Sikorski, R. S., and Hieter, P. (1989) A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics . 122 , 19–27.

Sipos, K., Lange, H., Fekete, Z., Ullmann, P., Lill, R., and Kispal, G. (2002) Maturation of cytosolic iron-sulfur proteins requires glutathione. J. Biol. Chem. 277 , 26944–9.

Smith, P. M., Fox, J. L., and Winge, D. R. (2012) Biogenesis of the cytochrome bc(1) complex and role of assembly factors. Biochim. Biophys. Acta . 1817 , 276–86.

Spiller, M. P., Ang, S. K., Ceh-Pavia, E., Fisher, K., Wang, Q., Rigby, S. E. J., and Lu, H. (2013) Identification and characterization of mitochondrial Mia40 as an iron-sulfur protein. Biochem. J. 455 , 27–35.

Srinivasan, V., Pierik, A. J., and Lill, R. (2014) Crystal structures of nucleotide-free and glutathione-bound mitochondrial ABC transporter Atm1. Science . 343 , 1137–40.

Stehling, O., and Lill, R. (2013) The role of mitochondria in cellular iron–sulfur protein biogenesis: mechanisms, connected processes, and diseases. Cold Spring Harb. Perspect. Med. 3, 1–17.

128

Stephen, D. W., and Jamieson, D. J. (1997) Amino acid-dependent regulation of the Saccharomyces cerevisiae GSH1 gene by hydrogen peroxide. Mol. Microbiol. 23 , 203– 10. Strain, J., Lorenz, C. R., Bode, J., Garland, S., Smolen, G. A., Ta, D. T., Vickery, L. E., and Culotta, V. C. (1998) Suppressors of superoxide dismutase (SOD1) deficiency in Saccharomyces cerevisiae. Identification of proteins predicted to mediate iron-sulfur cluster assembly. J. Biol. Chem. 273 , 31138–44.

Sugiyama, K. (2000) The Yap1p-dependent Induction of Glutathione Synthesis in Heat Shock Response of Saccharomyces cerevisiae. J. Biol. Chem. 275 , 15535–15540.

Tan, S.-X., Greetham, D., Raeth, S., Grant, C. M., Dawes, I. W., and Perrone, G. G. (2010) The thioredoxin-thioredoxin reductase system can function in vivo as an alternative system to reduce oxidized glutathione in Saccharomyces cerevisiae. J. Biol. Chem. 285 , 6118–26.

Tateishi, N., Higashi, T., Shinya, S., Naruse, A., and Sakamoto, Y. (1974) Studies on the regulation of glutathione level in rat liver. J. Biochem. 75 , 93–103.

Terziyska, N., Grumbt, B., Kozany, C., and Hell, K. (2009) Structural and functional roles of the conserved cysteine residues of the redox-regulated import receptor Mia40 in the intermembrane space of mitochondria. J. Biol. Chem. 284 , 1353–63.

Thorvaldsen, J. L., Sewell, A. K., McCowen, C. L., and Winge, D. R. (1993) Regulation of metallothionein genes by the ACE1 and AMT1 transcription factors. J. Biol. Chem. 268 , 12512–8. Tienson, H. L., Dabir, D. V., Neal, S. E., Loo, R., Hasson, S. A., Boontheung, P., Kim, S.-K., Loo, J. A., and Koehler, C. M. (2009) Reconstitution of the Mia40-Erv1 Oxidative Folding Pathway for the Small Tim Proteins. Mol. Biol. Cell . 20 , 3481–3490.

Toledano, M. B., Kumar, C., Le Moan, N., Spector, D., and Tacnet, F. (2007) The system biology of thiol redox system in Escherichia coli and yeast: differential functions in oxidative stress, iron metabolism and DNA synthesis. FEBS Lett. 581 , 3598–607.

129

Toledano, M. B., Delaunay-Moisan, A., Outten, C. E., and Igbaria, A. (2013) Functions and cellular compartmentation of the thioredoxin and glutathione pathways in yeast. Antioxid. Redox Signal. 18 , 1699–711.

Tong, W.-H., and Rouault, T. A. (2006) Functions of mitochondrial ISCU and cytosolic ISCU in mammalian iron-sulfur cluster biogenesis and iron homeostasis. Cell Metab. 3, 199–210.

Trotter, E. W., and Grant, C. M. (2003) Non-reciprocal regulation of the redox state of the glutathione-glutaredoxin and thioredoxin systems. EMBO Rep. 4, 184–8.

Trotter, E. W., and Grant, C. M. (2005) Overlapping roles of the cytoplasmic and mitochondrial redox regulatory systems in the yeast Saccharomyces cerevisiae. Eukaryot. Cell . 4, 392–400.

Vogel, C., Abreu, R. de S., Ko, D., Le, S.-Y., Shapiro, B. A., Burns, S. C., Sandhu, D., Boutz, D. R., Marcotte, E. M., and Penalva, L. O. (2010) Sequence signatures and mRNA concentration can explain two-thirds of protein abundance variation in a human cell line. Mol. Syst. Biol. 6, 400. von der Malsburg, K., Müller, J. M., Bohnert, M., Oeljeklaus, S., Kwiatkowska, P., Becker, T., Loniewska-Lwowska, A., Wiese, S., Rao, S., Milenkovic, D., Hutu, D. P., Zerbes, R. M., Schulze-Specking, A., Meyer, H. E., Martinou, J.-C., Rospert, S., Rehling, P., Meisinger, C., Veenhuis, M., Warscheid, B., van der Klei, I. J., Pfanner, N., Chacinska, A., and van der Laan, M. (2011) Dual role of mitofilin in mitochondrial membrane organization and protein biogenesis. Dev. Cell . 21 , 694–707.

Vögtle, F.-N., Burkhart, J. M., Rao, S., Gerbeth, C., Hinrichs, J., Martinou, J.-C., Chacinska, A., Sickmann, A., Zahedi, R. P., and Meisinger, C. (2012) Intermembrane space proteome of yeast mitochondria. Mol. Cell. Proteomics . 11 , 1840–52.

Webert, H., Freibert, S.-A., Gallo, A., Heidenreich, T., Linne, U., Amlacher, S., Hurt, E., Mühlenhoff, U., Banci, L., and Lill, R. (2014) Functional reconstitution of mitochondrial Fe/S cluster synthesis on Isu1 reveals the involvement of ferredoxin. Nat. Commun. 5, 5013.

130

Weckbecker, D., Longen, S., Riemer, J., and Herrmann, J.M. (2012) Atp23 biogenesis reveals a chaperone-like folding activity of Mia40 in the IMS of mitochondria. EMBO J. 31 , 4348–4358.

Wheeler, G. L., Trotter, E. W., Dawes, I. W., and Grant, C. M. (2003) Coupling of the transcriptional regulation of glutathione biosynthesis to the availability of glutathione and methionine via the Met4 and Yap1 transcription factors. J. Biol. Chem. 278 , 49920–8.

Wilkins, H. M., Marquardt, K., Lash, L. H., and Linseman, D. A. (2012) Bcl-2 is a novel interacting partner for the 2-oxoglutarate carrier and a key regulator of mitochondrial glutathione. Free Radic. Biol. Med. 52 , 410–9.

Wilkins, H. M., Brock, S., Gray, J. J., and Linseman, D. A. (2014) Stable over-expression of the 2-oxoglutarate carrier enhances neuronal cell resistance to oxidative stress via Bcl- 2-dependent mitochondrial GSH transport. J. Neurochem. 130 , 75–86.

Wohlrab, H. (2009) Transport proteins (carriers) of mitochondria. IUBMB Life . 61 , 40–6.

Wrobel, L., Trojanowska, A., Sztolsztener, M. E., and Chacinska, A. (2013) Mitochondrial protein import: Mia40 facilitates Tim22 translocation into the inner membrane of mitochondria. Mol. Biol. Cell . 24 , 543–54.

Wu, A. L., and Moye-Rowley, W. S. (1994) GSH1, which encodes gamma- glutamylcysteine synthetase, is a target gene for yAP-1 transcriptional regulation. Mol. Cell. Biol. 14 , 5832–5839.

Xue, Z. (2011) Using GFP-Based Redox and pH Sensors to Monitor Glutathione Metabolism In the Mitochondrial Intermembrane Space. MS. Thesis, University of South Carolina.

Yamaguchi-Iwai, Y., Stearman, R., Dancis, A., and Klausner, R. D. (1996) Iron-regulated DNA binding by the AFT1 protein controls the iron regulon in yeast. EMBO J. 15 , 3377– 84.

131

Yang, M., Cobine, P. A., Molik, S., Naranuntarat, A., Lill, R., Winge, D. R., and Culotta, V. C. (2006) The effects of mitochondrial iron homeostasis on cofactor specificity of superoxide dismutase 2. EMBO J. 25 , 1775–83.

Zerbes, R. M., Bohnert, M., Stroud, D. A., von der Malsburg, K., Kram, A., Oeljeklaus, S., Warscheid, B., Becker, T., Wiedemann, N., Veenhuis, M., van der Klei, I. J., Pfanner, N., and van der Laan, M. (2012) Role of MINOS in mitochondrial membrane architecture: cristae morphology and outer membrane interactions differentially depend on mitofilin domains. J. Mol. Biol. 422 , 183–91.

Zhang, Y., Lyver, E. R., Nakamaru-Ogiso, E., Yoon, H., Amutha, B., Lee, D.-W., Bi, E., Ohnishi, T., Daldal, F., Pain, D., and Dancis, A. (2008) Dre2, a conserved eukaryotic Fe/S cluster protein, functions in cytosolic Fe/S protein biogenesis. Mol. Cell. Biol. 28 , 5569–82.

132