View of the Pichia Expression System

Home , Frataxin, Transfer RNA

STRUCTURE-FUNCTION STUDY OF CELLULAR IRON CHEMISTRY

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

JIA HUANG

*****

The Ohio State University 2009

Dissertation Committee: Approved by Professor James A. Cowan, Advisor

Professor Karin Musier-Forsyth ______Advisor Professor Ross E. Dalbey Graduate Program in Chemistry

ABSTRACT

Iron-sulfur clusters are small inorganic cofactors found in all kingdoms of life.

The iron-sulfur cluster biogenesis is a complex system which involves a surprisingly large group of proteins. In human cells, frataxin is believed to recruit ferrous iron from the labile iron pool and subsequently deliver it to the scaffold protein human ISU where the iron-sulfur clusters will form. Deficiency in cellular frataxin production results in a human disease, Friedreich's ataxia which affects 1 in 50,000 humans. By using mutagenesis, the activities of frataxin mutants were investigated. We found that frataxin may have a pool of potential sites that can stably bind an iron center when bridged to a variety of physiological targets; there may not be unique binding loci, but rather a number of locations that provide flexibility in the binding of physiological partner proteins.

Frataxin has been found to repair the damaged cluster on mitochondrial aconitase.

We herein investigated its activity on cytosolic aconitase, IRP1, which can register the iron level and is important for the iron homeostasis of human cells. Frataxin can bind to cytosolic aconitase and repair the damaged clusters ([3Fe-4S]) into [4Fe-4S] clusters.

Recently, we reported a potential role for the N-terminus (residues 56-77) as a structural switch to control access to the acidic surface. We also characterized a high affinity iron-binding site in the full-length frataxin that promotes an autocatalytic ii cleavage of the N-terminal domain. Herein we reported the cloning and overexpression of human mitochondrial HscB, a J-type co-chaperone, and demonstrated an interaction between human frataxin and human HscB by cross-linking experiments and ITC. HscB was also shown to promote cleavage of the N-terminal domain of full-length human frataxin.

Human ISU protein contains three highly conserved cysteine residues (C44, C70, and C113) that mutagenesis studies suggested to be directly coordinated to the cluster.

The fourth ligand is still unclear. H112 was mutated into alanine or aspartate, and the effects of the mutation were evaluated. We found that H112 might be the fourth ligand that helps to stabilize the cluster, however, it was not absolutely needed for the formation of cluster; it might be the ligand that helps to deliver iron.

Different lines of results from other labs suggested that ependymin might be involved in long-term memory formation and neurodegeneration. Human ependymin was cloned for expression in E coli, yeast and insect cell. The protein from overexpression using pET-32 was extracted, purified and refolded. The biochemical characteristics were evaluated.

iii

To my father and mother, Xinyao Huang and Zhihua Dou

ACKNOWLEDGMENTS

I thank my adviser, Dr. J. A. Cowan, for providing a stimulating and flexible

environment that helps me to develop independent thinking and research skills. I thank

him for providing intellectual guidance and being continuously patient and understanding.

I am also thankful for his effort on assisting me with scientific writing. I would also like

to thank those that have served on my dissertation committee.

I would like to thank all members of Cowan research group, including Taejin

Yoon, Yan Jin, Wenbin Qi, Shu-pao Wu, Yushi Liu, Chun-An Chen, Seth Bradford,

Lalintip Hocharoen, Shu Ding , Jeff Joyner, Jaye Murdoch, and Monica Luo, for all the

help that you offered.

I would like to thank my parents. They have been a constant source of support. I

am very grateful to my husband, Dr. Jin Wang, for his patience and effort of being a

computer support and formatting this dissertation.

I greatly appreciate the funding that I received from the Department of Chemistry.

VITA

February 5, 1981 ...... Born, Yichang, Hubei, China

2003...... B. S. Chemistry Wuhan University

2005 – 2009...... Teaching Assistant The Ohio State University

2003 – 2009...... Research Assistant The Ohio State University

PUBLICATIONS

Research Publications

1. Huang J, Dizin E, Cowan JA (2008) Mapping iron binding sites on human frataxin: implications for cluster assembly on the ISU Fe-S cluster scaffold protein. J. Biol Inorg Chem; 13, 825-36.

2. Huang J, Cowan JA. (2009) Iron-sulfur cluster biosynthesis: role of a semi- conserved histidine. Chem Commun. 2009, 3071-3.

FIELDS OF STUDY

Major Field: Chemistry vi

TABLE OF CONTENTS

Page

Abstract ...... ii

Dedication ...... iv

Acknowledgments...... v

Vita ...... vi

List of Tables ...... xiv

List of Figures ...... xv

List of Abbreviations ...... xviii

CHAPTERS:

1. Introduction ...... 1

1.1. Importance of Iron-Sulfur Clusters ...... 1

1.2. Pathways of Iron-Sulfur Cluster Biogenesis ...... 3

1.3. The ISC Assembly System ...... 3

1.3.1. Role of the Scaffold Protein IscU ...... 4

1.3.2. Iron Binding and Delivery Protein Frataxin ...... 6

1.3.3. Role of Cysteine Desulfurase ...... 6

1.3.4. Role of Chaperones ...... 7

1.3.5. Other Important Components of ISC Assembly System ...... 8

vii

1.4. Human Diseases Caused by Defects in the Iron-Sulfur Cluster Biosynthesis . 9

1.5. References for Chapter 1 ...... 9 2. Mapping Iron Binding Sites on Human Frataxin: Implications for Cluster Assembly on The ISU Fe–S Cluster Scaffold Protein ...... 14

2.1. Introduction ...... 14

2.2. Materials and Methods ...... 20

2.2.1. General chemicals ...... 20 2.2.2. Cloning, Expression, and Purification of His-tagged Truncated Native Frataxin ...... 20 2.2.3. Cloning, Expression, and Purification of Frataxin Derivatives A, B, and C ...... 22

2.2.4. Expression and Purification of Apo Human ISU and D37A ISU ...... 23

2.2.5. Mass-Spectrometric Determination of Protein Molecular Mass ...... 23

2.2.6. Quantitation of Iron Binding to Frataxin and Derivatives by ITC ...... 24

2.2.7. Quantitation of Zinc Binding to Frataxin and Derivatives by ITC ...... 25

2.2.8. Quantitation of Iron Required for Frataxin–ISU Binding...... 25

2.2.9. Fe–S Cluster Assembly on ISU Mediated by Frataxin and Derivatives . 26

2.3. Results ...... 27 2.3.1. Cloning, Expression, and Mass-Spectrometric Characterization of Frataxin and Derivatives ...... 27

2.3.2. Iron Binding to Native Frataxin and Derivatives ...... 32

2.3.3. Zinc Binding to Native Frataxin ...... 36 2.3.4. Evaluation of Iron-Promoted Frataxin-ISU Complex Formation by ITC ...... 38 2.3.5. [2Fe–2S] Cluster Reconstitution Mediated by Native and Derivative Frataxins ...... 42

2.4. Discussion ...... 44

2.4.1. Iron Binding Properties of Native Frataxin and its Derivatives ...... 44

2.4.2. Frataxin–ISU Complex Formation ...... 47 viii

2.4.3. Function of the Acidic Residues Implicated by the Fe–S Reconstitution Assays ...... 48

2.4.4. Comparison of Frataxin Homologues ...... 50

2.5. Summary ...... 51

2.6. References for Chapter 2 ...... 53

3. Iron-Sulfur Cluster Biosynthesis: Role of a Conserved Histidine ...... 58

3.1. Introduction ...... 58

3.2. Materials and Methods ...... 61

3.2.1. General Chemicals ...... 61

3.2.2. Mutagenesis, Expression and Purification of Proteins ...... 61

3.2.3. Chemical Reconstitution with ISU and ISU Derivatives ...... 62

3.2.4. UV-Vis Spectroscopy for ISU and ISU Derivatives ...... 62

3.2.5. Quantification of Iron Content and Extinction Coefficient of Cluster ... 63 3.2.6. Studies on Iron Binding Capability of ISU and ISU Derivatives via ITC ...... 63 3.2.7. Studies on the Interaction between Frataxin and ISU and ISU Derivatives via ITC ...... 64

3.2.8. Tm Nifs-assisted Reconstitution ...... 64

3.3. Results ...... 65

3.3.1. Expression and Purification of ISU Derivatives ...... 65

3.3.2. UV-Vis Spectroscopy of ISU and its Derivatives ...... 67

3.3.3. Chemical Reconstitution with ISU and Derivatives ...... 69

3.3.4. Quantification of Iron Content and Extinction Coefficient of Cluster ... 71

3.3.5. Studies on Iron Binding Capability of ISU and derivatives via ITC ...... 72

3.3.6. The Interaction between Frataxin and ISU or ISU derivatives via ITC . 75

3.3.7. Tm Nifs-assisted Reconstitution ...... 77

3.4. Discussion ...... 79

3.5. References for Chapter 3 ...... 81

4. Human Frataxin Promotes Cytosolic Repair of IRP1 ...... 83

4.1. Introduction ...... 83

4.2. Materials and Methods ...... 86

4.2.1. General Chemicals ...... 86

4.2.2. Expression and Purification of Human IRP1 ...... 86

4.2.3. Aconitase Assay ...... 87 4.2.4. Reconstitution of Partially Active IRP1 into Holo-IRP1 with Iron and Sulfur...... 87

4.2.5. Converting [4Fe-4S]-IRP1 to [3Fe-4S]-IRP1 ...... 88

4.2.6. Converting [4Fe-4S]-IRP1 to Apo Form ...... 88

4.2.7. Measuring the Iron Content ...... 89

4.2.8. Binding Study of IRP1 with Frataxin via ITC ...... 89

4.2.9. Activation of the Aconitase Activity for [3Fe-4S]-IRP1 ...... 90

4.3. Results ...... 90

4.3.1. Preparation of Apo-, [3Fe-4S]- and [4Fe-4S]-IRP1 ...... 90

4.3.2. Binding between [3Fe-4S]-IRP1 and Truncated Frataxin via ITC ...... 91

4.3.3. Binding of [3Fe-4S]-IRP1 to Full-length Frataxin via ITC ...... 95

4.3.4. Binding between Apo IRP1 and Frataxin via ITC ...... 97

4.3.5. Reactivation of [3Fe-4S]-IRP1 ...... 101

4.4.1. The Interaction between [3Fe-4S]-IRP1 and Truncated Frataxin ...... 105 4.4.2. The Anionic Patch on Frataxin Is Not Involved in Binding to [3Fe-4S]- IRP1 ...... 105

4.4.3. Interaction between Apo-IRP1 and Frataxin ...... 106

4.4.4. Frataxin Enhances the Rate of Reactivation of [3Fe-4S]-IRP1 ...... 107

4.5. References for Chapter 4 ...... 110

5. Cloning And Expression of Recombinant Human Ependymin and Initial Characterization ...... 114

5.1. Introduction ...... 114

5.2. Materials and Methods ...... 119

5.2.1. General Chemicals ...... 119 5.2.2. Cloning of Mature Ependymin into pET-32 for Expression in E. coli with a Thioredoxin Fusion Protein in the N-terminal End ...... 119

5.2.3. Expression of pET-32-epen ...... 120 5.2.4. Purification and On-Column Refolding of Thioredoxin-Ependymin Fusion Protein ...... 121

5.2.5. Removing Thioredoxin from Thioredoxin-Ependymin Fusion Protein 121

5.2.6. Circular Dichroism of Thioredoxin-Ependymin Fusion Protein ...... 122

5.2.7. Probing the Calcium Binding Sites by Terbium Luminescence ...... 122

5.2.8. Cloning of Human Ependymin into pPIC9 for Expression in Pichia ... 123 5.2.9. Transformation of pPIC9-ependymin into Pichia strains GS115 or KM71 ...... 124

5.2.10. Screening for Mut+ and MutS Transformants ...... 125

5.2.11. PCR Analysis of Pichia Integrants ...... 125

5.2.12. Expression of MutS KM71-pPIC9-epen ...... 126

5.2.13. Expression of Mut+ GS115-pPIC9-epen ...... 127 5.2.14. Cloning of Mature Ependymin into pRmHa3 for Expression in Insect Cell Schneider's Drosophila Line-2 ...... 127 5.2.15. Expression, Identification and Purification of Ependymin in Insect Cell Schneider's Drosophila Line-2 (S2) ...... 128

5.2.16. Cloning of Mature Human Ependymin into pET 21 ...... 129

5.3. Results and Discussion ...... 129

5.3.1. Cloning and Expression of Ependymin Gene with pET-32...... 129 5.3.2. Purification and On-column Refolding of Thioredoxin-Ependymin Fusion ...... 135 xi

5.3.3. Removing the Tags from the Fusion Protein ...... 137

5.3.4. Circular Dichroism of Thioredoxin-Ependymin Fusion Protein ...... 140

5.3.5. Probing the Calcium Binding Sites by Terbium Luminescence ...... 142

5.3.6. Overview of the Pichia Expression System ...... 145

5.3.7. Cloning of Human Ependymin into pPIC9 for Expression in Pichia ... 146

5.3.8. Expression of MutS KM71-pPIC9-epen and Mut+ GS115-pPIC9-epen 151 5.3.9. Cloning, Expression and Purification of Ependymin in Insect Cell Schneider's Drosophila Line-2 ...... 151

5.4. References for Chapter 5 ...... 157 6. Human Mitochondrial HscB Chaperone Mediate Maturation of Human Frataxin ...... 160

6.1. Introduction ...... 160

6.2. Materials and Methods ...... 165

6.2.1. General Chemicals ...... 165 6.2.2. Cloning, Expression and Purification of Human HscB, J-type Co- chaperone ...... 165 6.2.3. Quantitation of HscB Protein Binding with Frataxin by Isothermal Titration Calorimetry ...... 167 6.2.4. Quantitation of Metal Binding to HscB by Isothermal Titration Calorimetry ...... 168 6.2.5. Self-cleavage of N-terminal His-tagged full-length Frataxin under Different Conditions ...... 169 6.2.6. Self-cleavage of C-terminal His-tagged Full-length Frataxin with the Aid of HscB ...... 169

6.2.7. Cross-linking Experiment ...... 170

6.2.8. Mass Spectroscopy...... 170

6.3. Results ...... 170

6.3.1. Cloning, Expression, and Purification of the Human HscB Chaperone 170

6.3.2. Interaction between HscB and Frataxin ...... 173 xii

6.3.3. Metal Binding to HscB ...... 181

6.3.4. Zinc does not Promote Binding between Frataxin and HscB ...... 183

6.3.5. HscB Accelerates the Maturation of Human Frataxin ...... 185

6.4. Discussion ...... 193

6.5. References for Chapter 6 ...... 200

Bibliography ...... 205

xiii

LIST OF TABLES

Table Page

2.1. Ferrous ion titration to wild-type and derivative frataxin monitored by isothermal titration calorimetry (ITC) ...... 33

2.2. Summary of data for interaction between frataxin derivatives and ISU in the presence of 1 equiv of ferrous ion ...... 41

3.1. Extinction coefficient of human ISU and derivatives...... 71

3.2. Summary of iron binding study on ISU and ISU derivatives via ITC...... 73

4.1. A summary of the binding affinities between IRP1 and frataxin...... 100

4.2. Rates of reactivation of [3Fe-4S]-IRP1 under different conditions...... 102

6.1. Summary of the rates of the cleavage under different conditions ...... 192

xiv

LIST OF FIGURES

Figure Page

1.1. Structures of [2Fe-2S] cluster, [4Fe-4S] cluster and [3Fe-4S] cluster...... 2

1.2. A model for the mechanism of Fe-S cluster biogenesis in yeast mitochondria...... 4

1.3. Proposed mechanism of cysteine desulfurase...... 7

2.1. Crystal structure of human frataxin...... 17

2.2. The crystal structure of the ISU-type protein from Streptococcus Pyrogenes ...... 19

2.3. Purification profile of mutant A, B and C...... 28

2.4. Molecular weights of derivative A, B and C determined by mass spectroscopy. 31

2.5. Iron binding to frataxin derivatives monitored by isothermal titration calorimetry (ITC) ...... 34

2.6. Zinc binding to native frataxin via ITC ...... 37

2.7. ITC measurement of the interaction between human ISU and frataxin derivatives in the presence of 1 equiv of ferrous ion ...... 39

2.8. Fe–S reconstitution assay with native frataxin and the three derivatives ...... 43

2.9. Multiple sequence alignment generated by ClustalW ...... 45

3.1. C-terminal sequence alignment of bacterial and human ISU-type proteins ...... 60

3.2. Expression and purification of H112D/D37A and H112A/D37A ISU ...... 66

3.3. H105D/D37A and H112A/D37A ISU right after elution from Ni-NTA column. 68

3.4. UV spectra of native, Asp37Ala, His103Asp/Asp37Ala and His103Ala/Asp37Ala ISU after chemical reconstitution ...... 70

3.5. Studies on iron binding capability of ISU or ISU derivatives via ITC ...... 74

3.6. Interaction between D37A ISU or derivatives and frataxin in the presence of ferrous ion ...... 76

3.7. Tm Nifs-assisted reconstitution of D37A ISU, H112D/D37A ISU, H112A/D37A ISU and a control lacking ISU ...... 78

4.1. Multiple sequence alignment generated by CLUSTALW ...... 85

4.2. Study of binding between [3Fe-4S]-IRP1 and truncated frataxin with varying concentration of citrate in present via ITC ...... 93

4.3. Study of binding between [3Fe-4S]-IRP1 or apo IRP1 and full-length frataxin with 2 mM citrate in present via ITC ...... 96

4.4. The conformational change from closed position ([3Fe-4S]- or [4Fe-4S]-IRP1) to open position (apo IRP1) ...... 98

4.5. Study of the binding between apo IRP1 and truncated frataxin with 0 mM or 2 mM citrate via ITC ...... 99

4.6. Reactivation of [3Fe-4S]-IRP1 ...... 103

5.1. The sequence of full-length human ependymin ...... 116

5.2. The polymerization hypothesis of ependymin ...... 118

5.3. The domains of the ependymin fusion protein overexpressed with a pET-32b (+) and pET-32a-c (+) cloning/expression region ...... 131

5.4. Overexpression of human ependymin-thioredoxin fusion protein ...... 132

5.5. Purification profile of ependymin-thioredoxin fusion protein ...... 134

5.6. Purification and refolding of ependymin-thioredoxin fusion protein ...... 136

5.7. Treating ependymin-thioredoxin fusion protein with enterokinase ...... 138

5.8. Circular Dichroism characterization of ependymin-thioredoxin fusion ...... 141

5.9. Terbium may bind to ependymin-thioredoxin fusion ...... 144

5.10. Multiple cloning site of pPIC 9 and surrounding sequences...... 147

5.11. Colony PCR of transfromants with primers for subcloning or AOX primers .... 148

xvi

5.12. Screening for phenotype of Pichia strains ...... 150

5.13. Cloning of mature ependymin into pRmHa3 for expression ...... 153

5.14. Immunoprecipitation with anti-His antibody conjugated PGS beads...... 155

5.15. Purification of ependymin and Western Blot ...... 156

6.1. Ribbons diagram of Hs frataxin showing surface carboxylates that are putative binding sites for iron ions...... 162

6.2. Amino acid sequence alignment of HscB proteins from E. coli (Hsc20), Jac1p from Saccharomyces cerevisiae and residues # 72 ~ 235 of Homo sapiens ...... 172

6.3. Crosslinking between HscB and frataxin ...... 174

6.4. Isothermal titration calorimetry (ITC) measurements of 0.32 ~ 0.38 mM HscB with 0.02 ~ 0.03 mM truncated frataxin or full-length frataxin in the absence of iron ...... 176

6.5. Isothermal titration calorimetry (ITC) measurements of 10 mM ferrous ion with 0.1 mM truncated Frataxin with or without 0.1 mM HscB in present ...... 180

6.6. Titrating zinc ion into HscB by ITC ...... 182

6.7. Electrostatic potential maps of HscB ...... 184

6.8. Effects of HscB in the cleavage chemistry of full-length frataxin ...... 187

6.9. Time-dependent cleavage of full-length frataxin in the absence or the presence of HscB ...... 188

6.10. Time-dependent cleavage of N-terminal His-tagged full-length frataxin under different conditions as noted in the figure...... 189

6.11. Self-cleavage of C-terminal His-tagged full-length frataxin ...... 191

6.12. Model of HscB mediated cleavage of frataxin ...... 197

xvii

LIST OF ABBREVIATIONS

ATP Adenosine 5'-triphosphate bp Base Pair

BSA Bovine Serum Albumin

CD Circular Dichroism dH2O Distilled Water

DMSO dimethylsulfoxide

DNA Deoxyribonucleic Acid dNTP Deoxynucleotide 5'-triphosphate

DTT Dithiothreitol

EDTA Ethylenediaminetetraacetic Acid

EDC 1-ethyl-3-(dimethylaminopropyl) carbodiimide hydrochloride

EPR Electron Paramagnetic Resonance

ESI Electrospray Ionization

ExPASy Expert Protein Analysis System

Fd Ferredoxin

FPLC Fast Protein Liquid Chromatography

Gdn-HCl Guanidinium Hydrochloride

HEPES (N-[2-hydroxyethyl] piperazine-N'-[2-ethanesulfonic acid]) xviii

IPTG Isopropyl Thiogalactoside

ISC Iron-Sulfur Cluster

LB Luria-Bertani

MALDI Matrix Assisted Laser Desorption/Ionization

MS mass spectrometry

MW Molecular Weight

Nif Nitrogen Fixation

NMR Nuclear Magnetic Resonance

NOE Nuclear Overhauser Effect

NOESY Nuclear Overhauser Effect Spectroscopy

NTA Nitrilotriacetic Acid

PAGE Polyacrylamide Gel Electrophoresis

PCR Polymerase Chain Reaction

PMSF Phenylmethylsulfonyl Fluoride

RNA Ribonucleic Acid

SDS Sodium Dodecyl Sulfate

TOF Time of Flight

Tris Trihydroxyethylene Amine tRNA Transfer RNA

UV Ultraviolet

WT Wild Type

xix

CHAPTER 1

INTRODUCTION

1.1. Importance of Iron-Sulfur Clusters

Iron-sulfur clusters are small inorganic cofactors found in all kingdoms of life. They serve a variety of roles, such as catalysts in the Complex I and Complex II of oxidative phosphorylation (1), electron carriers in ferredoxin (2), regulatory sensors as in the iron regulatory protein 1 (IRP1) (3), structural stabilizers as in endo III/MutY subfamily (4), and so on. The most common iron-sulfur clusters include [2Fe-2S], [3Fe-4S], and [4Fe-

4S] (Figure 1.1). The iron ions are tetrahedrally coordinated by thiolate ligands of cysteine side chains and additional non-protein sulfides (5). Non-cysteinyl ligands to iron-sulfur clusters were also found in naturally occurring iron-sulfur clusters, as well as introduced by mutagenesis studies where cysteine residues were replaced by a histidine or serine (6-8).

[2Fe-2S]

[4Fe-4S]

[3Fe-4S]

Figure 1. 1. Structures of [2Fe-2S] cluster, [4Fe-4S] cluster and [3Fe-4S] cluster.

1.2. Pathways of Iron-Sulfur Cluster Biogenesis

Iron-sulfur cluster biogenesis is a complex process which involves a surprisingly large group of proteins, the function of many of which is still not clear. Three iron-sulfur clusters biosynthesis systems have been discovered so far: the Nif (nitrogen fixation) system which generates clusters for nitrogenase, the ISC (iron-sulfur cluster) assembly system for general housekeeping, and the SUF (sulfur mobilization) system which may function under oxidative stress or iron limitation. Certain organisms may have more than one system, and all systems have both cysteine desulfurases and the scaffold proteins (9).

Some important components are highly conserved. Similar observations are often seen in the counterparts in other systems. Of these systems, the ISC assembly system will be discussed most herein.

1.3. The ISC Assembly System

The key proteins in human mitochondrial Fe-S cluster assembly and trafficking include the scaffold protein human ISU (Isu1), the iron binding and delivery protein frataxin (Yfh1), cysteine desulfurase IscS (Nfs1), ISA(Isa1 and Isa2), molecular chaperones HscA and HscB (Ssq1, Jac1), electron carrier ferredoxin (Yah1), and cluster export proteins ABC7 (Atm1) (10). The names in parenthesis are the notations for yeast proteins. Yeast was extensively used as a model system to investigate the ISC assembly system. There are other proteins that work collectively in the ISC assembly, as well.

Figure 1.2 shows a model of iron-sulfur cluster biogenesis.

Figure 1.2. A model for the mechanism of Fe-S cluster biogenesis in yeast mitochondria

(11).

1.3.1. Role of the Scaffold Protein IscU

Genetic and cell biology studies support the presence of IscU in the ISC assembly system, and IscU may be among the most highly conserved genes in evolution (12-14).

By using a combination of absorption, resonance Raman, Mossbauer, analytical and biochemical methods, it was revealed that the iron-sulfur clusters could be formed by an

IscS-mediated assembly on IscU and that [4Fe-4S] cluster could be formed sequentially from [2Fe-2S] clusters. The assembly of the [2Fe-2S] cluster on IscU proteins can be

achieved by providing sufficient inorganic ferrous/ferric ion and sulfide under anaerobic

condition in vitro, as well (15). These results pointed out the critical role of IscU in

initiating iron-sulfur clusters formation. The results also suggested the presence of partial

noncysteinyl ligation in all species (16). The [2Fe-2S] cluster is oxidatively and

reductively labile. The cluster can be stabilized by substitution of a highly conserved

aspartate which regulates a solvent-accessible channel to the cluster (D37 for Hs ISU and

Sp ISU, and D40 for Tm IscU) (5).

IscU can also transfer the loaded iron-sulfur cluster to a targeted apo protein

exemplified by the reaction between holo human, S. pombe, or T. martima IscU and the

apo form of ferredoxin (Fdx) (15, 17). The transfer was a direct protein-to-protein

transfer without the disassembly and reassembly of the cluster (17).

Several solved structures of the zinc bound monomeric apo IscU proteins share a

highly conserved α + β globular core architecture. The three highly conserved cysteine

residues which are thought to be the iron-sulfur cluster formation site are located on a

solvent-accessible surface, and a zinc ion is coordinated by the three cysteine residues.

The fourth ligand for this zinc ion is an aspartate or a histidine residue (18). Recently, the

crystal structure of IscU from the hyperthermophilic bacterium Aquifex aeolicus with a

bound [2Fe–2S] cluster was solved. The protein is an asymmetric trimer, and each trimer

coordinates one iron-sulfur cluster. Each monomer in the trimer complex shares a high

structural similarity to the structures of Haemophilus influenzae (Hi) and Mus musculus

(Mm) IscU (18). Another important finding is that the fourth ligand of the [2Fe-2S] cluster is a histidine (18).

1.3.2. Iron Binding and Delivery Protein Frataxin

Frataxin is believed to be the iron donor to IscU for several reasons. First of all, although iron is an essential element in normal metabolism, it has a high toxicity if the concentration in the cell is not properly regulated. Iron ions can generate reactive oxygen species which can damage cells by attacking cellular membranes, proteins, and DNA (19).

Therefore, to recruit iron from the labile iron pool, there needs to be a protein that can bind to and subsequently deliver iron to IscU. Second, the gene of frataxin is found in the isc operon of many organisms. Its distribution indicates that frataxin is involved in iron- sulfur cluster biogenesis (20). Finally, a deficiency of frataxin is the cause of a neurodegenerative disease, Friedreich's ataxia. It was found that cells from such patients showed a loss of activities for enzymes requiring iron-sulfur clusters to function, which may be the primary effect of frataxin deficiency (21). Biochemical studies also suggested that frataxin can interact with IscU in an iron-dependent manner and that it can assist iron-sulfur cluster formation on IscU in vitro (22). Frataxin may play other roles such as an iron chaperone during cellular heme production (23) or helping to repair oxidatively damaged aconitase clusters (24).

1.3.3. Role of Cysteine Desulfurase

Cysteine desulfurases are found throughout the three iron-sulfur assembly systems.

They have been found to assist in vitro iron-sulfur cluster formation in a number of apo scaffold proteins from a variety of organisms (25-28). The desulfurases are pyridoxal phosphate (PLP)-dependent enzymes and the reaction catalyzed was found to produce alanine and either sulfane (S0) or, in the presence of a reducing agent, sulfide (S2 ) 6

(Figure 1.3) (29). IscS from E coli can interact with IscU and transfer sulfur directly through disulfide bond formation between the cysteine residues from IscS and IscU (30).

However, whether this is the initial step of iron-sulfur cluster formation is still under debate (31). Cysteine desulfurase also plays important roles in biological sulfur mobilization such as thiolation of certain thionucleotides (9).

Figure 1.3. Proposed mechanism of cysteine desulfurase.

1.3.4. Role of Chaperones

HscA and HscB were also encoded in a conserved region of the isc operon, which revealed that they may have specific roles in iron-sulfur biogenesis (32, 33). E coli HscA shares around 40% sequence identity with E coli Hsp70, DnaK. HscA has been shown to

specifically bind to a peptide sequence, LPPVK in the scaffold protein IscU in E coli (34).

The crystal structure of a complex of a peptide containing LPPVK bound to the HscA

substrate binding domain has been determined (35). HscB is homologous to the J-type co-chaperone. HscB contains an N-terminal J-domain and a distinct C-terminal domain

(36). Consistent with the facts of the Hsp70 family and its co-chaperone, the interaction of IscU with HscA is regulated by HscB, which binds the scaffold protein and assists directing it to the HscA. At the same time, HscB help to stabilize the whole complex

(36-38). The ATPase activity of HscA is enhanced during this process (39).

In vivo and in vitro studies of yeast Ssq1 and Jac1 indicate that the chaperones are not required for iron sulfur cluster assembly on IscU (40). Deletion of either Ssq1 or Jac1 resulted in iron accumulation on Isu1 and a reduction in the amount of iron present in iron-sulfur enzymes (40). Additional genetic and biochemical studies are needed to establish the exact functions of chaperone in the iron-sulfur protein biosynthesis.

1.3.5. Other Important Components of ISC Assembly System

Isa proteins were proposed to be an alternative scaffold for de novo Fe-S cluster assembly or an iron donor for IscU in bacteria. In yeast, though, Isa proteins might be important for the maturation of a specific type of iron-sulfur proteins, namely aconitase- type Fe-S proteins and SAM proteins inside mitochondria (10).

In yeast, the ABC transporter Atm1 of the mitochondrial inner membrane, together with sulfhydryl oxidase Erv1 and glutathione, forms the export machinery which is essential for exporting clusters out of mitochondria in yeast (10, 24, 41).

1.4. Human Diseases Caused by Defects in the Iron-Sulfur Cluster Biosynthesis

Since iron-sulfur clusters are the essential prosthetic centers of a number of

human proteins, defects in iron-sulfur clusters will affect a wide range of cellular

activities. Some important proteins that contain iron-sulfur clusters include complexes I–

III in the mitochondrial respiratory chain, mitochondrial aconitase in the tricarboxylic acid cycle, cytosolic aconitase, which is important for human iron homeostasis, and

ferrochelatase for heme biosynthesis. Disruptions in iron-sulfur cluster biogenesis and repair will affect numerous basic cellular processes, which is the reason for several recognized human diseases. One example is Friedreich’s ataxia, affecting 1 out of 50,000 humans, which is caused by a deficiency in cellular frataxin production, and the inherited human sideroblastic anemia caused by mutations in glutaredoxin 5 (21). Glutaredoxin 5 might be involved in transferring iron-sulfur clusters to target proteins. The exact function of glutaredoxin is still not clear (21).

1.5. References for Chapter 1

1. Sazanov, L. A., and Hinchliffe, P. (2006) Structure of the hydrophilic domain of respiratory complex I from Thermus thermophilus, Science 311, 1430-1436.

2. Ciurli, S., and Musiani, F. (2005) High potential iron-sulfur proteins and their role as soluble electron carriers in bacterial photosynthesis: tale of a discovery, Photosynth Res 85, 115-131.

3. Volz, K. (2008) The functional duality of iron regulatory protein 1, Curr Opin Struct Biol 18, 106-111.

4. Lukianova, O. A., and David, S. S. (2005) A role for iron-sulfur clusters in DNA repair, Curr Opin Chem Biol 9, 145-151.

5. Mansy, S. S., and Cowan, J. A. (2004) Iron-sulfur cluster biosynthesis: toward an understanding of cellular machinery and molecular mechanism, Acc Chem Res 37, 719-725.

6. Iwata, S., Saynovits, M., Link, T. A., and Michel, H. (1996) Structure of a water soluble fragment of the 'Rieske' iron-sulfur protein of the bovine heart mitochondrial cytochrome bc1 complex determined by MAD phasing at 1.5 A resolution, Structure 4, 567-579.

7. Kounosu, A., Li, Z., Cosper, N. J., Shokes, J. E., Scott, R. A., Imai, T., Urushiyama, A., and Iwasaki, T. (2004) Engineering a three-cysteine, one- histidine ligand environment into a new hyperthermophilic archaeal Rieske-type [2Fe-2S] ferredoxin from Sulfolobus solfataricus, J Biol Chem 279, 12519-12528.

8. Hurley, J. K., Weber-Main, A. M., Hodges, A. E., Stankovich, M. T., Benning, M. M., Holden, H. M., Cheng, H., Xia, B., Markley, J. L., Genzor, C., Gomez- Moreno, C., Hafezi, R., and Tollin, G. (1997) Iron-sulfur cluster cysteine-to- serine mutants of Anabaena -2Fe-2S- ferredoxin exhibit unexpected redox properties and are competent in electron transfer to ferredoxin:NADP+ reductase, Biochemistry 36, 15109-15117.

9. Johnson, D. C., Dean, D. R., Smith, A. D., and Johnson, M. K. (2005) Structure, function, and formation of biological iron-sulfur clusters, Annu Rev Biochem 74, 247-281.

10. Lill, R., and Muhlenhoff, U. (2008) Maturation of iron-sulfur proteins in eukaryotes: mechanisms, connected processes, and diseases, Annu Rev Biochem 77, 669-700.

11. Lill, R., Dutkiewicz, R., Elsasser, H. P., Hausmann, A., Netz, D. J., Pierik, A. J., Stehling, O., Urzica, E., and Muhlenhoff, U. (2006) Mechanisms of iron-sulfur protein maturation in mitochondria, cytosol and nucleus of eukaryotes, Biochim Biophys Acta 1763, 652-667.

12. Dean, D. R., Bolin, J. T., and Zheng, L. (1993) Nitrogenase metalloclusters: structures, organization, and synthesis, J Bacteriol 175, 6737-6744.

13. Hwang, D. M., Dempsey, A., Tan, K. T., and Liew, C. C. (1996) A modular domain of NifU, a nitrogen fixation cluster protein, is highly conserved in evolution, J Mol Evol 43, 536-540.

14. Garland, S. A., Hoff, K., Vickery, L. E., and Culotta, V. C. (1999) Saccharomyces cerevisiae ISU1 and ISU2: members of a well-conserved gene family for iron- sulfur cluster assembly, J Mol Biol 294, 897-907.

15. Mansy, S. S., Wu, G., Surerus, K. K., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis. Thermatoga maritima IscU is a structured iron-sulfur cluster assembly protein, J Biol Chem 277, 21397-21404.

16. Agar, J. N., Krebs, C., Frazzon, J., Huynh, B. H., Dean, D. R., and Johnson, M. K. (2000) IscU as a scaffold for iron-sulfur cluster biosynthesis: sequential assembly of [2Fe-2S] and [4Fe-4S] clusters in IscU, Biochemistry 39, 7856-7862.

17. Wu, S. P., Wu, G., Surerus, K. K., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis. Kinetic analysis of [2Fe-2S] cluster transfer from holo ISU to apo Fd: role of redox chemistry and a conserved aspartate, Biochemistry 41, 8876- 8885.

18. Shimomura, Y., Wada, K., Fukuyama, K., and Takahashi, Y. (2008) The Asymmetric Trimeric Architecture of [2Fe-2S] IscU: Implications for Its Scaffolding during Iron-Sulfur Cluster Biosynthesis, J Mol Biol.

19. Aisen, P., Enns, C., and Wessling-Resnick, M. (2001) Chemistry and biology of eukaryotic iron metabolism, Int J Biochem Cell Biol 33, 940-959.

20. Huynen, M. A., Snel, B., Bork, P., and Gibson, T. J. (2001) The phylogenetic distribution of frataxin indicates a role in iron-sulfur cluster protein assembly, Hum Mol Genet 10, 2463-2468.

21. Rouault, T. A., and Tong, W. H. (2008) Iron-sulfur cluster biogenesis and human disease, Trends Genet 24, 398-407.

22. Yoon, T., and Cowan, J. A. (2003) Iron-sulfur cluster biosynthesis. Characterization of frataxin as an iron donor for assembly of [2Fe-2S] clusters in ISU-type proteins, J Am Chem Soc 125, 6078-6084.

23. Yoon, T., and Cowan, J. A. (2004) Frataxin-mediated iron delivery to ferrochelatase in the final step of heme biosynthesis, J Biol Chem 279, 25943- 25946.

24. Bulteau, A. L., O'Neill, H. A., Kennedy, M. C., Ikeda-Saito, M., Isaya, G., and Szweda, L. I. (2004) Frataxin acts as an iron chaperone protein to modulate mitochondrial aconitase activity, Science 305, 242-245.

25. Zheng, L., and Dean, D. R. (1994) Catalytic formation of a nitrogenase iron-sulfur cluster, J Biol Chem 269, 18723-18726.

26. Khoroshilova, N., Beinert, H., and Kiley, P. J. (1995) Association of a polynuclear iron-sulfur center with a mutant FNR protein enhances DNA binding, Proc Natl Acad Sci U S A 92, 2499-2503.

27. Hidalgo, E., and Demple, B. (1996) Activation of SoxR-dependent transcription in vitro by noncatalytic or NifS-mediated assembly of [2Fe-2S] clusters into apo- SoxR, J Biol Chem 271, 7269-7272.

28. Yuvaniyama, P., Agar, J. N., Cash, V. L., Johnson, M. K., and Dean, D. R. (2000) NifS-directed assembly of a transient [2Fe-2S] cluster within the NifU protein, Proc Natl Acad Sci U S A 97, 599-604.

29. Urbina, H. D., Silberg, J. J., Hoff, K. G., and Vickery, L. E. (2001) Transfer of sulfur from IscS to IscU during Fe/S cluster assembly, J Biol Chem 276, 44521- 44526.

30. Kato, S., Mihara, H., Kurihara, T., Takahashi, Y., Tokumoto, U., Yoshimura, T., and Esaki, N. (2002) Cys-328 of IscS and Cys-63 of IscU are the sites of disulfide bridge formation in a covalently bound IscS/IscU complex: implications for the mechanism of iron-sulfur cluster assembly, Proc Natl Acad Sci U S A 99, 5948- 5952.

31. Nuth, M., Yoon, T., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis: characterization of iron nucleation sites for assembly of the [2Fe-2S]2+ cluster core in IscU proteins, J Am Chem Soc 124, 8774-8775.

32. Blattner, F. R., Plunkett, G., 3rd, Bloch, C. A., Perna, N. T., Burland, V., Riley, M., Collado-Vides, J., Glasner, J. D., Rode, C. K., Mayhew, G. F., Gregor, J., Davis, N. W., Kirkpatrick, H. A., Goeden, M. A., Rose, D. J., Mau, B., and Shao, Y. (1997) The complete genome sequence of Escherichia coli K-12, Science 277, 1453-1474.

33. Fleischmann, R. D., Adams, M. D., White, O., Clayton, R. A., Kirkness, E. F., Kerlavage, A. R., Bult, C. J., Tomb, J. F., Dougherty, B. A., Merrick, J. M., and et al. (1995) Whole-genome random sequencing and assembly of Haemophilus influenzae Rd, Science 269, 496-512.

34. Hoff, K. G., Cupp-Vickery, J. R., and Vickery, L. E. (2003) Contributions of the LPPVK motif of the iron-sulfur template protein IscU to interactions with the Hsc66-Hsc20 chaperone system, J Biol Chem 278, 37582-37589.

35. Cupp-Vickery, J. R., Peterson, J. C., Ta, D. T., and Vickery, L. E. (2004) Crystal structure of the molecular chaperone HscA substrate binding domain complexed with the IscU recognition peptide ELPPVKIHC, J Mol Biol 342, 1265-1278.

36. Vickery, L. E., and Cupp-Vickery, J. R. (2007) Molecular chaperones HscA/Ssq1 and HscB/Jac1 and their roles in iron-sulfur protein maturation, Crit Rev Biochem Mol Biol 42, 95-111.

37. Cupp-Vickery, J. R., and Vickery, L. E. (2000) Crystal structure of Hsc20, a J- type Co-chaperone from Escherichia coli, J Mol Biol 304, 835-845.

38. Hoff, K. G., Silberg, J. J., and Vickery, L. E. (2000) Interaction of the iron-sulfur cluster assembly protein IscU with the Hsc66/Hsc20 molecular chaperone system of Escherichia coli, Proc Natl Acad Sci U S A 97, 7790-7795.

39. Silberg, J. J., Tapley, T. L., Hoff, K. G., and Vickery, L. E. (2004) Regulation of the HscA ATPase reaction cycle by the co-chaperone HscB and the iron-sulfur cluster assembly protein IscU, J Biol Chem 279, 53924-53931.

40. Muhlenhoff, 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, 4815-4825.

41. Pondarre, C., Antiochos, B. B., Campagna, D. R., Clarke, S. L., Greer, E. L., Deck, K. M., McDonald, A., Han, A. P., Medlock, A., Kutok, J. L., Anderson, S. A., Eisenstein, R. S., and Fleming, M. D. (2006) The mitochondrial ATP-binding cassette transporter Abcb7 is essential in mice and participates in cytosolic iron- sulfur cluster biogenesis, Hum Mol Genet 15, 953-964.

CHAPTER 2

MAPPING IRON BINDING SITES ON HUMAN FRATAXIN: IMPLICATIONS FOR CLUSTER ASSEMBLY ON THE ISU FE–S CLUSTER SCAFFOLD PROTEIN

2.1. Introduction

Friedreich’s ataxia (FRDA) is among the most common progressive neurodegenerative diseases. The molecular basis stems from defective expression of the mitochondrial protein frataxin that often results from the presence of the unstable homozygous GAA triplet repeat expansion within the first intron of the FRDA gene (1-

3). Cells of FRDA patients show a deficiency of active Fe–S cluster containing enzymes, an accumulation of iron ion, and an increase in oxidative stress (4). Understanding the functional role of frataxin in mitochondrial iron homeostasis is a fast-growing area of investigation, and active participation in mitochondrial energy conversion and oxidative phosphorylation has been suggested (5). Several lines of evidence suggest that frataxin is involved in the regulation of Fe–S cluster biosynthesis. Deletion of the yeast frataxin homologue, Yfh1, results in a decrease in activity of Fe–S cluster containing enzymes, including aconitase and succinate dehydrogenase (4). The conserved core region of 14

frataxin/CyaY proteins shares sequence similarity with the N-terminal region of small

proteins conferring resistance to tellurium, which is closely related to sulfur in the

periodic table and has been proposed to interfere with sulfur metabolism (6).

Furthermore, a phenotypic response indicative of Fe–S enzyme deficiency is observed

earlier than that of iron accumulation in mouse mitochondria, where frataxin has been

deleted (7).

ISU is a scaffold protein for assembly of Fe–S clusters prior to delivery to cluster- dependent apo target proteins. Inasmuch as frataxin has also demonstrated iron binding properties (8), a role as a chaperone for cellular iron and as an iron donor for assembly of

Fe–S clusters in ISU-type proteins seems plausible, and is supported by biochemical studies in our laboratory (9), as well as cellular studies by other groups (10-12). Frataxin has also been demonstrated to play an important role in regulating heme biosynthesis, another major iron-dependent metabolic pathway, by providing ferrous ion for ferrochelatase-mediated insertion to protoporphyrin IX (13, 14). A role for frataxin in the restoration of aconitase activity has also been suggested (15). More recently it has been shown that overexpression of frataxin in the cytoplasm is able to rescue FRDA cells (16,

17), suggesting important cytosolic roles for frataxin (18).

While frataxin seems likely to serve diverse cellular roles, several of these functions appear to involve delivery of iron from a negatively charged patch on the protein surface (Figure 2.1). The anionic surface of human frataxin is defined by the 12 acidic residues located in the α1 helix and β1 sheet (Figure 2.1) (19, 20), with direct binding of partner proteins to the same surface (21). Five acidic residues (E100, E108,

E111, D112, and D124) are completely conserved in other animals and in plants, yeast, and eubacteria. All frataxin proteins show very similar structures, with the anionic patch located at the boundary of the same α1 and β1 secondary structural domains, suggestive of an essential role in frataxin function (22).

Figure 2.1. Crystal structure of human frataxin (Protein Data Bank ID 1EKG) (19). The substituted residues are highlighted with stick structures. Residues highlighted in light blue and green were substituted by alanine in derivative A (E100A/E101A/D104A/

E108A/E111A/D112A); residues highlighted in magenta were substi-tuted by alanine in derivative B (D115A/E121A/D122A/D124A); residues highlighted in yellow and green were substituted by alanine in derivative C (E92A/E96A/E100A/E101A).

We have earlier reported that a form of human frataxin (residues 78–210) that is typically investigated in biochemical studies, but which is truncated beyond the point defined by mitochondrial protease sites, can bind up to seven iron ions (9). NMR studies of iron binding to bacterial, yeast, and truncated human frataxin show that those residues demonstrating the most significant resonance shifts lie within the anionic patch (21, 23).

The truncated human frataxin is the result of a self-cleavage reaction of the full-length frataxin product that is formed following the proteolytic loss of the mitochondrial targeting sequence (24). Recent cellular studies also suggest that this truncated form of human frataxin is the principal active form within the cell (25), and that it also is formed by cleavage of a longer precursor protein in vivo.

Frataxin can bind to human ISU in the presence of iron and support Fe–S cluster reconstitution; however, frataxin does not show significant binding to ISU in the absence of iron (9). This has been interpreted as reflecting a need for iron to bridge the two proteins (9). Consistent with this model, both X-ray and NMR solution structures of ISU- type proteins from several organisms show five conserved acidic residues near the Fe–S cluster assembly site (26) (Figure 2.2) that may be responsible for binding to a partner iron delivery protein, with iron serving as a bridge to promote binding between frataxin and ISU through these two anionic surfaces.

Figure 2.2. The crystal structure of the ISU-type protein from Streptococcus Pyrogenes

(Protein Data Bank ID 1SU0). Residues colored yellow are those cysteines required for

Fe–S cluster formation. Residues that are highlighted in magenta are highly conserved acidic residues.

To better define the location and roles of potential iron binding sites, we have used the frataxin–ISU interaction as a model system, and investigated three multiply substituted frataxin derivatives (frataxin A, frataxin B, and frataxin C; Figure 2.1) to explore the functional roles of these acidic residues, including their role in binding iron ions and partner proteins, and/or iron delivery. Figure 2.1 shows the position of the amino acid substitutions in the three derivatives. Herein, we report the results of thermodynamic and kinetic experiments designed to evaluate the likely binding sites for frataxin-bound iron on the anionic patch, as well as the protein binding interface and residues that are likely responsible for iron delivery. Both iron and ISU binding properties for each derivative were characterized by isothermal titration calorimetry (ITC), and their efficiency in a Fe–S cluster reconstitution assay has also been assessed.

2.2. Materials and Methods

2.2.1. General chemicals

Inorganic salts were obtained from Aldrich (Milwaukee, WI, USA).

Nitrilotriacetic acid (NTA) resin was purchased from QIAGEN (Valencia, CA, USA).

2.2.2. Cloning, Expression, and Purification of His-tagged Truncated Native

Frataxin

Truncated frataxin (residues 78–210) was subcloned from a pET-28b (+) containing the full-length frataxin (9). PCR amplification of the gene was accomplished by use of DeepVent polymerase, 5’-GGA AGG CCA TAT GTT GAG GAA ATC TGG

AAC TTT G as 5’ primer, and 5’-AGA ATT CAA GCA TCT TTT CCG GAA TAG

GCC AA as 3’ primer. The underlined motifs correspond to NdeI and EcoRI cleavage sequences, respectively. The PCR product was subsequently purified by use of a

QIAGEN PCR purification kit and analyzed by agarose gel electrophoresis. The PCR product and pET-21b (+) expression vector were prepared separately by double digestion with NdeI/EcoR1 at 37 ºC. The vector was further treated with calf intestinal alkaline phosphatase and then the digested insert and vector were mixed in an equimolar ratio and ligated with T4 DNA ligase by overnight reaction at 14 ºC. Chemically competent

Escherichia coli DH5α cells were transformed with the ligation product and, following standard selection procedures, the correct insert in the plasmid product was confirmed by

DNA sequencing.

Addition of a His-tag to the N-terminus of this truncated frataxin protein (residues

78–210) was accomplished by subcloning the gene previously inserted into pET-21b (+) and cloning it back into pET-28b (+). Following PCR amplification using Pfu Turbo and the previously described primers, the double digestion, ligation, and transformation procedures were repeated and the DNA was sequence confirmed. This plasmid is designated pET-28b (+) trFtx.

E. coli BL21 (DE3) cells transformed with pET-28b (+) trFtx and an overnight culture (40 mL) was used as an inoculum for 4 L of Luria–Bertani medium containing 35

µg/mL kanamycin. Cells were grown at 37 ºC to an optical density of 0.6–0.8, and expression was induced by addition of isopropyl β-D-thiogalactopyranoside to a final concentration of 0.1 mM. The culture was incubated for an additional 4 h at 30 ºC, and

then the cells were harvested by centrifugation. Cell pellets were lysed by sonication and

centrifuged at 15,000 rpm, 4 ºC, for 15 min. Following centrifugation the supernatant was loaded onto a Ni-NTA column that had been equilibrated with binding buffer [20 mM tris

(hydroxymethyl) aminomethane hydrochloride (Tris–HCl), pH 7.9, 5 mM imidazole, 500

mM NaCl]. The column was then washed with 5 column vol of binding buffer and 5 vol

of binding buffer plus 25 mM imidazole, and the protein was subsequently eluted with

binding buffer plus 795 mM imidazole. Purity was confirmed by sodium dodecyl sulfate

polyacrylamide gel electrophoresis (SDS-PAGE).

2.2.3. Cloning, Expression, and Purification of Frataxin Derivatives A, B, and C

The original expression strains for three multiply substituted frataxin derivatives

(residues 56–210) were a gift from Grazia Isaya of the Mayo Clinic. To improve the ease

of purification and increase yield, each protein derivative was cloned into a pET-28b (+)

expression vector with an N-terminal His-tag. To allow comparison with truncated native

protein, the derivatives were subcloned to include residues 78–210. Moreover, it was

found that the original protein derivatives lost residues 56–77 by a self-cleavage reaction

as previously identified for full-length native frataxin (24). The following primers were

used to prepare the His-tagged derivatives: 5’ primer, 5’-GGA AGG CCA TAT GTT

GAG GAA ATC TGG AAC T-3’;3’ primer, 5’-ACC GGA ATT CAA GCA TCT TTT

CCG GAA TAG G-3’. These primers were based on the human frataxin sequence and incorporated NdeI and EcoRI restriction sites (underlined) at the 5’ and 3’ ends, respectively. The sequence of the cloned frataxin mutant gene product was confirmed by

DNA sequencing. Plasmid pET-28b (+)-(frataxin mutants) were transformed into E. coli 22

BL21 (DE3). Expression and purification procedures were similar to those described for truncated native frataxin.

2.2.4. Expression and Purification of Apo Human ISU and D37A ISU

Expression and purification of human ISU and D37A ISU proteins was performed as previously described (27, 28). Cell mass was thawed on ice and resuspended in binding buffer (5 mM imidazole, 500 mM NaCl, 20 mM Tris–HCl, pH 7.9; approximately 2.5 mL/g of cells) containing 20 µg/ mL phenylmethylsulfonyl fluoride,

2.5 µg/mL leupeptin, and lysed by sonication. Ultrapure urea was added to the cell lysate to a final concentration of 6 M and the mixture was stirred on ice for 1 h. The cellular debris was removed by centrifugation at 15,000 rpm and 4 ºC for 30 min, and the supernatant was loaded onto a Ni-NTA column (QIAGEN) previously equilibrated with binding buffer and 6 M urea. After loading, the column was washed with 10 column volume of wash buffer (binding buffer with 20 mM imidazole and 6 M urea) and eluted with elution buffer (binding buffer with 200 mM imidazole and 6 M urea). Fractions containing native ISU or D37A ISU, as judged by SDS-PAGE, were pooled, desalted, and concentrated by ultrafiltration using an Amicon stirred-cell concentrator. The protein was stored at -80 ºC until use. Protein purity at each stage of purification was monitored by SDS-PAGE.

2.2.5. Mass-Spectrometric Determination of Protein Molecular Mass

Mass-spectral data was obtained through the Campus Chemical Instrument Center at The Ohio State University. Mass determination was performed by electrospray 23

ionization time-of-flight measurements using a Micromass Q-TOF II (Micromass,

Wythenshawe, UK) mass spectrometer equipped with an orthogonal electrospray source

(Z-spray) operated in positive ion mode. Background salt was removed from holo frataxin by extensive dialysis against NANOpure water.

2.2.6. Quantitation of Iron Binding to Frataxin and Derivatives by ITC

ITC measurements of ferrous ion binding to frataxin were carried out at 25 ºC

using a MicroCal OMEGA ultrasen-sitive titration calorimeter. All ITC experiments were carried out using the same instrument at 25 ºC, unless otherwise specified. To a solution of each truncated frataxin protein (100 µM native, 441 µM derivative A, 311 µM derivative B, or 287 µM derivative C) was added the ferrous ammonium sulfate titrant.

The concentration of the titrant solution varied according to the frataxin used (10 mM for native, 30 mM for derivative A, and 20 mM for derivatives B and C). In each injection,

10 µL titrant was delivered over a period of 24 s, with an adequate interval (5–10 min) between injections to allow complete equilibration. The buffer consisted of 50 mM N-(2- hydroxyethyl) piperazine-N’-ethanesulfonic acid (Hepes) and 100 mM NaCl at pH 7.5. In the case of derivative A, buffers were prepared with 0.05% Tween. To maintain the ferrous form, 5–10 mM sodium dithionite was added both to the titrant and to the solution in the sample cell, and all solutions were thoroughly argon-purged and degassed.

Retention of the ferrous state was confirmed spectrophotometrically after each experiment. The data for the background titration, consisting of an identical titrant solution, but containing only the buffer solution in the sample cell, were subtracted from the data for each experimental titration to account for heat of dilution. The data were 24

collected automatically and subsequently ana-lyzed with a one-site binding model by the

Windows-based Origin software package supplied by MicroCal. The Origin software uses a nonlinear least-squares algorithm (minimization of χ2) and the concentrations of

the titrant and the sample to fit the heat flow per injection to an equilibrium binding

equation, providing best-fit values of the stoichiometry (n), change in enthalpy (ΔH),

change in entropy (ΔS), and dissociation constant (KD).

2.2.7. Quantitation of Zinc Binding to Frataxin and Derivatives by ITC

ITC measurements of zinc ion binding to frataxin were carried out at 25 ºC using

a MicroCal OMEGA ultrasen-sitive titration calorimeter. To a solution of 20 µM

truncated frataxin protein was added the titrant containing 200 µM zinc sulfate. Each

injection contained 10 µL titrant and was delivered over a period of 24 s, with an

adequate interval (5–10 min) between injections to allow complete equilibration. The

buffer consisted of 50 mM N-(2-hydroxyethyl) piperazine-N’-ethanesulfonic acid (Hepes)

and 100 mM NaCl at pH 7.5. The data for the background titration, consisting of an

identical titrant solution, but containing only the buffer solution in the sample cell, were

subtracted from the data for each experimental titration to account for heat of dilution.

The data were collected and analyzed as mentioned earlier.

2.2.8. Quantitation of Iron Required for Frataxin–ISU Binding

All solutions were thoroughly argon-purged and degassed prior to use. Typically,

10 µL of titrant containing 200– 250 µM human ISU was delivered to a solution of 20–

30 µM native or derivative frataxin in the presence of 0, 1, 2, 4, or 7 equiv of iron over a 25 period of 24 s with an adequate interval (3–5 min) between injections to allow complete equilibration. To maintain the ferrous state, 5 mM sodium dithionite was used both in solution and in the sample cell, and retention of the ferrous state was confirmed spectrophotometrically after each experiment. The data for the background titration, consisting of an identical titrant solution, but containing only the buffer solution in the sample cell, were subtracted from the data for each experimental titration to account for heat of dilution. The experiment was carried out in the same buffer as described earlier and the data were collected and analyzed similarly. A similar experiment with 1 equiv of ferric chloride was performed as a control, which also provides information on the effect of the charge of the bridging species. Frataxin binding to native and D37A ISU is essentially undistinguishable. The latter protein was used more extensively in studies of cluster assembly on ISU since the resulting cluster is more stable and the rate of formation more readily determined.

2.2.9. Fe–S Cluster Assembly on ISU Mediated by Frataxin and Derivatives

To a solution containing 40 µM frataxin protein or derivative, excess (280 µM) ferrous ion, and 5 mM dithiothreitol were added (to final concentration) 2.4 mM Na2S and 100 µM D37A ISU in 50 mM Hepes buffer at pH 7.5. D37A ISU was used for assays of cluster assembly since the resulting cluster is more stable; however, as noted earlier the affinity for frataxin is similar to that of native ISU. All components for D37A ISU reconstitution were prepared under anaerobic conditions and cluster formation was monitored by measuring the UV–vis absorbance at 456 nm, a characteristic absorption band from the [2Fe–2S]2+ cluster. The time course was followed by measuring the 26

absorption of the reaction mixture directly. A control experiment was carried out in the

absence of ISU. UV–vis spectra were recorded with a Hewlett-Packard 8425A diode-

array spectrophotometer using On-Line Instrument Systems 4300S operating system

software.

2.3. Results

2.3.1. Cloning, Expression, and Mass-Spectrometric Characterization of Frataxin

and Derivatives

To facilitate purification and improve isolated yields, the truncated form of recombinant native frataxin (residues 78–210) and the derivative proteins defined

in Figure 2.1 were subcloned into a pET-28b(+) vector that yields an N-terminal His-tag.

This provided 30–60 mg of each protein from a 4 L culture. Figure 2.3 shows the

purification profile of mutant A, B and C.

Figure 2.3. Purification profile of mutant A, B and C. Top: lane 1, low molecular marker;

lane 2, lysate of mutant A; lane 3, flowthrough of mutant A; lane 4, wash with 30 mM

imidazole in the binding buffer for mutant A; lane 5 and 6, elution with 200 mM imidazole for mutant A; lane 7, lysate of mutant B; lane 8, flowthrough of mutant B; lane

9, wash with 30 mM imidazole for mutant B; lane 10, elution with 200 mM imidazole for mutant B. Bottom, purification of mutant C: lane 1, low molecular marker; lane 2, lysate; lane 3, flowthrough, lane 4, 5 and 6, wash with 5, 20 and 40 mM imidazole in the binding buffer; lane 7, elution with 200 mM in the binding buffer.

A 1 2 3 4 5 6 7 8 9 10

30.0 kDa

B kDa 1 2 3 4 5 6 7 97.0 66.0 45.0

30.0

20.1

14.4

The three mutants overexpressed and purified using the constructs from Grazia

Isaya of the Mayo Clinic were sent to determine their mass by mass spectroscopy (Figure

2.4). The mass of mutant A, B and C are 14.345 kDa, 14.475 kDa and 14.433 kDa, respectively. The mass of all three proteins correspond to residue 78-210 of human frataxin which is shorter than the original constructs. This indicates that the purified proteins were truncated. Similar observation was found by many other research groups that human frataxin undergoes self-cleavage both in vitro and in vivo (25, 29).

Figure 2.4. Molecular weights of derivative A, B and C determined by mass spectroscopy.

The highest peak in each graph represents each mutant protein (A, 14345 Da, B, 14476

Da, and C, 14433 Da). 31

2.3.2. Iron Binding to Native Frataxin and Derivatives

Ferrous ion binding to the three frataxin derivatives was evaluated by use of ITC and binding stoichiometries and other thermodynamic parameters are summarized in Table 2.1. The modest decrease in affinity observed for the truncated native protein, relative to our prior measurement, reflects the presence of the N-terminal His-tag. Direct coordination by the His-tag appears unlikely, since the affinity decreases rather than increases, while prior studies have demonstrated that the His-tag does not have an intrinsic affinity for iron (30, 31). However, as noted in recent work (24), the N-terminal domain of full-length frataxin, or His-tagged versions of truncated frataxin, appears to lie over the anionic patch. Consequently, iron and partner proteins must compete for binding sites and the observed KD is larger than expected.

-1 -1 -1 n ΔH (kcal·mol ) ΔS (cal·mol K ) KD (mM)

WT frataxin 6.8 ± 0.3 -1.2 ± 0.2 11.1 0.46 ± 0.02

Derivative A 1.9 ± 0.1 -10.7 ± 0.5 -19.2 0.21 ± 0.03

Derivative B 3.9 ± 0.1 -1.6 ± 0.1 10.5 0.32 ± 0.02

Derivative C 4.8 ± 0.1 -1.0 ±0.1 13.8 0.18 ± 0.02 a. Control experiments indicate that the enthalpy change for derivative A does not reflect Tween contributions. Tween was added to aid the solubility of derivative A at the higher concentrations required for ITC experiments. The thermodynamic parameters for derivative A appear to reflect disaggregation of derivative A (which shows lower solubility relative to derivatives B and C) following dilution and more pronounced changes in solvation.

Table 2.1. Ferrous ion titration to wild-type and derivative frataxin monitored by isothermal titration calorimetry (ITC)

Figure 2.5. Iron binding to frataxin derivatives monitored by isothermal titration calorimetry (ITC); a–d) Data obtained for derivatives A, B, and C and native frataxin, respectively. As described in ‘‘Materials and methods’’, data were collected by delivering 10 µL of titrant, containing stock ferrous ion, to a solution of truncated frataxin, or derivatives, in 50 mM N-(2-hydroxyethyl)piperazine-N’-ethanesulfonic acid

(Hepes), 100 mM NaCl at pH 7.5 in the presence of 5 mM dithionite, at 25 ºC. All solutions were thoroughly argon-purged. The upper portion of each graph is the raw data, while the lower portion shows the integrated heat for each peak.

Relative to the native control, all of the derivative proteins exhibit KD values in a

similar range (approximately 0.2–0.4 mM), suggesting the absence of substantial long-

range electrostatic influences on binding, although the stoichiometry varies from

approximately 7 for native frataxin, to 2 for derivative A, 4 for derivative B, and 5 for

derivative C.

2.3.3. Zinc Binding to Native Frataxin

Zinc is similar to iron in terms of geometry, ionic radii and ability to coordinate to hydroxyl ligand. Zinc usually provides tighter binding due to smaller ionic radii. We found that zinc can bind to native frataxin with a KD of 7.6 ± 0.8 μM and the

stoichiometry is 1.22 ± 0.08. ΔH and ΔS were determined to be -5601 ± 513 kcal·mol-1

and 4.65 cal·mol-1K-1, respectively (Figure 2.8). In our case, zinc indeed binds tighter to

frataxin and the stoichiometry decreased to around 1. This is probably due to the ionic

radii of zinc is smaller, and lower stoichiometry result in lower repulsion among zinc ions;

unlike iron binding, the stoichiometry of which is around 7. Frataxin may not bind to 7

iron ions under physiological conditions. In the ITC experiment, titrating excess iron to

saturate iron binding sites gave a measurement of the maximum iron bind ability, and the

fitting may reflect an average of all iron binding sites.

Time (min) -10 0 10 20 30 40 50 60 70 80 90 100110120130 0.2

0.1

0.0

-0.1

-0.2

cal/sec -0.3 μ -0.4

-0.5

-0.6

-2

-4 kcal/mole of injectant

0246810 Molar Ratio

Figure 2.6. Zinc binding to native frataxin via ITC. To a solution of 20 µM truncated frataxin protein was added the titrant containing 200 µM zinc sulfate. Each injection contained 10 µL titrant and was delivered over a period of 24 s, with an adequate interval

(5–10 min) between injections to allow complete equilibration. The buffer consisted of

50 mM Hepes and 100 mM NaCl at pH 7.5.

2.3.4. Evaluation of Iron-Promoted Frataxin-ISU Complex Formation by ITC

Prior work from this laboratory (9), and others (18, 21, 23, 32), has suggested a role for iron in promoting binding of the partner protein ISU to native frataxin, possibly through iron-bridged carboxylates. To determine the number of iron ions required to promote binding between frataxin and human ISU, ITC binding experiments were performed with various equivalents of iron in the presence of ISU. Table 2.2 shows that the iron-dependent binding affinity of each frataxin derivative to the Fe–S cluster scaffold protein ISU is similar to that of native frataxin, as defined by ITC experiments, requiring only one iron center to promote nanomolar binding. A control experiment with 1 equiv of ferric ion was performed and KD was found to be modestly lower (KD ~ 21 ± 2 nM) than that obtained in the presence of 1 equiv of ferrous ion, consistent with the higher charge and stronger electrostatic bridging interaction to surface carboxylates.

Figure 2.7. ITC measurement of the interaction between human ISU and frataxin derivatives in the presence of 1 equiv of ferrous ion. A solution of human ISU (200–250

μM) was titrated into a sample containing 0.02 mM frataxin derivative A (a), frataxin derivative B (b), or frataxin derivative C (c), in the presence of 1 equivalent of ferrous ion under anaerobic conditions at 25 ºC. A control experiment showing titration of ISU into a

0.02 mM solution of native truncated frataxin in the absence of iron is shown in d.

ISU Frataxin derivatives Stoichiometry (n) KD (nM)

native a 0.96 ± 0.01 40 ± 6

derivative A 0.91 ± 0.01 50 ± 9

derivative B 0.95 ± 0.01 86 ± 7

derivative C 0.90 ± 0.01 114 ± 12

a. Relative to a KD ~ 12 nM in the presence of 7 equiv of iron

Table 2.2. Summary of data for interaction between frataxin derivatives and ISU in the

presence of 1 equiv of ferrous ion

To elaborate the binding interface between frataxin and ISU, and potentially localize the principal iron binding site(s), the binding of multiply substituted frataxin derivatives and ISU, in the presence of 1 equiv of iron, was evaluated by ITC. The expectation was for a decrease in binding affinity if the frataxin derivative lacked essential metal binding residues for iron-promoted association with ISU. However, since only one iron is needed to promote binding between frataxin and ISU, and all the derivatives bind two or more iron ions, there exists the potential for compensatory changes in binding by use of more than one iron binding location to promote association

with ISU. Table 2.2 documents that, in the presence of 1 equiv of ferrous ion, derivatives

A, B, and C show binding to ISU with KD values of 50 ± 9, 86 ± 7, and 114 ± 12 nM,

respectively, relative to KD ~ 30 nM for native truncated frataxin. Additional iron

equivalents have a minimal impact on partner affinity. In the presence of 2 equiv of iron,

KD for ISU binding to native truncated frataxin is approximately 16 ± 5 nM, with

essentially no improvement with increasing iron stoichiometry.

2.3.5. [2Fe–2S] Cluster Reconstitution Mediated by Native and Derivative Frataxins

Fe–S cluster assembly on ISU is a relatively slow reaction in the absence of

frataxin as an iron donor. Frataxin mediates cluster assembly by binding to ISU and

subsequently delivering the iron centers required for assembly of the Fe–S cluster core

(9). To better identify carboxylate residues that favor iron delivery, relative to complex

formation, reconstitution assays were performed under similar reaction conditions, using

native frataxin and its derivatives (Figure 2.8), with each experiment performed in

triplicate at least. Native frataxin was observed to reconstitute ISU with kobs = 0.126 ±

0.009 min-1 (relative to 0.15 min-1 for non-His-tagged native(9)), the rate of reconstitution for frataxin A was 0.080 ± 0.005 min-1, for frataxin B it was 0.086 ± 0.006 min-1, and for

frataxin C it was determined as 0.058 ± 0.008 min-1.

0.14

0.12

0.10

0.08

abs 0.06 Δ 0.04 WT 0.02 Mutant B Mutant A 0.00 Mutant C 0 102030405060

Time (min)

Figure 2.8. Fe–S reconstitution assay with native frataxin (filled squares) and the three

derivatives (A circles; B triangles; C open squares). Reaction solutions contained 40 µM

frataxin, or its derivatives, 280 µM ferrous ions, 5 mM dithiothreitol, 2.4 mM Na2S, and

100 µM D37A ISU in 50 mM Hepes buffer (pH 7.5). Each solution component required for the reconstitution assay was prepared under anaerobic conditions and cluster formation was monitored by UV–vis absorption spectroscopy. Experiments were

performed at least in triplicate and data points reflect average values with corresponding errors.

2.4. Discussion

Human frataxin displays a continuous anionic surface that is formed by 12 acidic

residues on the α1 helix and β1 sheet (Figure 2.1) (19). The yeast homologues exhibit a

similar charged surface patch (21). In the case of the human protein this surface appears capable of binding up to seven ferrous ions with KD = 55 µM (for non-His-tagged

frataxin), with the His-tagged protein showing an approximately ten-fold decrease in

binding affinity (9). While relatively weak by physiological standards, the affinity for

iron increases significantly when frataxin forms complexes with natural partner proteins,

and KD values on the order of nanomolar affinity are found (9, 33). Accordingly, the iron

binding sites revealed for frataxin alone are most likely best viewed as a pool of sites that are available for iron binding in complexes with partner proteins. A question to be

addressed in this paper is how these iron centers are distributed over the patch, and their

potential role(s) in promoting binding to specific partner proteins as well as mediating

delivery of iron to the Fe–S cluster assembly protein ISU.

2.4.1. Iron Binding Properties of Native Frataxin and its Derivatives

Frataxin has been implicated in mitochondrial iron homeostasis and many of its

functions are exerted through the anionic patch (9, 14, 15, 19). However, certain

derivatives of yeast Yfh1 that carry substitutions within the anionic patch show only

modest change in cellular oxidative stress and no significant defect in Fe–S cluster

formation (34, 35). Accordingly, we set out to determine the role of the negatively

charged surface by comparing the binding behavior of the native protein with that of

three derivative proteins carrying multiple amino acid substitutions. Derivative A binds 44 two irons, derivative B binds four irons, and derivative C binds five irons, consistent with prior NMR iron titration studies with human frataxin that suggested the iron binding sites to be distributed over the anionic patch (Figure 2.1 and Figure 2.9) (23).

Frataxin-human LRKSGTLGHPGSGSHMGSLDETTYERLAEETLDSLAEFFEDLADKPYTFEDYDVSFGSGV 131 Yfh1-yeast --ESSTDGQVVPQEVLN-LPLEKYHEEADDYLDHLLDSLEELSEAHPDCIP-DVELSHGV 108 CyaY-Ecoli ------MNDSEFHRLADQLWLTIEERLDDWDGDS----DIDCEINGGV 38 : :.. *:: : : ::: * .:. **

Frataxin-human LTVKLGGDLGTYVINKQTPNKQIWLSSPSSGPKRYDWTGKNWVYSHDGVSLHELLAAELT 191 Yfh1-yeast MTLEIP-AFGTYVINKQPPNKQIWLASPLSGPNRFDLLNGEWVSLRNGTKLTDILTEEVE 167 CyaY-Ecoli LTITFE-NGSKIIINRQEPLHQVWLATKQGG-YHFDLKGDEWICDRSGETFWDLLEQAAT 96 :*: : .. :**:* * :*:**:: .* ::* . :*: :.* .: ::*

Frataxin-human KALKTKLDLSSLAYSGKDA 210 Yfh1-yeast KAISKSQ------174 CyaY-Ecoli QQAGETVSFR------106 : .

Figure 2.9. Multiple sequence alignment generated by ClustalW; colons indicate positions where conserved substitutions are observed; dots indicate residues where semi- conserved substitutions are observed, and asterisks indicate identical residues.

Recent NMR studies of iron binding to human and yeast frataxin show that the

amide resonances for yeast frataxin residues D82 and D86, and the corresponding

residues D104 and E108 in the human protein (Figure 2.1 and Figure 2.9), are significantly broadened when 1 equiv of iron is added (23), as are the corresponding bacterial residues L15 and E19 (33). Another iron binding site on Yfh1 was assigned in the vicinity of D101 in Yfh1 on a neighboring surface (23), which corresponds to the conserved residue D124 in human frataxin (Figure 2.1), although broadening was not

observed in the human and bacterial proteins (23). For both bacterial and human proteins

additional broadening of cross peaks from the amide backbone for residues on the C-

terminal side of E108 (human numbering) was observed, consistent with the possibility

of additional iron binding sites for human frataxin, as previously documented (9). The

two iron centers characterized for Yfh1 are most likely sufficient to promote its

functional chemistry. The similarity in binding affinities that we have determined (Table

2.1) suggests that the multiple amino acid substitutions that have been introduced do not

significantly influence the binding of the remaining iron centers, either as a result of

compensatory binding by neighboring carboxylates or as a result of the sites being

sufficiently separated that long-range electrostatic effects have negligible influence.

While the iron binding affinities for native frataxin alone (including yeast and bacterial)

tend to be nonphysiological, physiologically relevant iron binding must be viewed in the

context of partner binding (such as to ISU or ferrochelatase) where the affinity increases

to the nanomolar range and iron plays a vital role in promoting complex formation. The

exception would be full-length human frataxin that has a unique high-affinity site and is

discussed elsewhere (24, 25). In other words, iron binding to truncated frataxin, which

has been demonstrated in cell studies to be the principal active form (25), is

physiologically relevant in the context of binding to partner proteins. The iron-only

binding studies presented herein essentially map out latent iron binding sites on the

protein surface that are of potential physiological relevance.

2.4.2. Frataxin–ISU Complex Formation

Inasmuch as the proposed binding mechanism requires bridging iron ions, ISU

should possess an appropriate constellation of iron binding residues to mediate the

bridging contact. A recent crystal structure of an ISU-type protein from Streptococcus

pyrogenes (Figure 2.2) shows that there is a highly conserved negatively charged surface

patch formed by five conserved residues that is located adjacent to the Fe–S cluster

binding site (26). This structural evidence is consistent with a recognition model where

iron serves to bridge ISU and the iron donor protein frataxin (9, 36). Adventitious binding

to nonpartner proteins would have little consequence if the resulting complex were not also functionally active.

Relative to yeast and bacterial frataxins, the human homologue apparently has a

greater number of possible iron binding sites, and so the recognition chemistry is

potentially more complex. To address the question of how many iron centers are required

to promote binding of ISU to frataxin, and to identify any preferred locations on the

anionic surface, the ISU–frataxin binding interaction was studied in the presence of 1–7

equiv of iron by ITC. The goal was to define the minimum iron requirements to promote

complex formation with ISU. It was found that the binding affinity of frataxin for ISU 47

with 1 equiv of iron present was similar to the affinity determined in the presence of 7 equivalents of iron (KD ~ 30 and 12 nM, respectively). This suggests that only one iron is

needed to promote the interaction between frataxin and ISU. Similar experiments with

the frataxin derivatives (Figure 2.1), in the presence of various equivalents of iron,

indicated that all three derivatives bind to ISU with similar affinity, even with only 1

equiv of iron added. Apparently, the ISU-binding site on frataxin can be on any area of

the anionic patch, so long as there is iron binding to that surface that can promote binding

of ISU. A consequence of this idea is that the orientation of native frataxin and its

derivatives, relative to ISU, may vary.

2.4.3. Function of the Acidic Residues Implicated by the Fe–S Reconstitution Assays

To investigate the functional chemistry of the acidic residues, Fe–S cluster

reconstitution experiments were performed for native and derivative frataxins. The

concentrations of ISU and frataxin used in the assay were high enough to ensure

complete complex formation between ISU and frataxin, while inorganic sulfide was

added in excess. Under physiological conditions, sulfide is delivered enzymatically

through an IscS-type protein (37), with assistance from NFU that appears to mediate

persulfide bond cleavage on IscS to yield inorganic sulfide (38).

Frataxin derivative B binds to ISU with similar affinity as native protein and can

-1 support reconstitution at a similar reaction rate (knat ~ 0.126 min compared with kderB ~

0.086 min-1, Figure 2.8), suggesting that the carboxylates responsible for binding and

delivery of iron are intact. Frataxin derivative A also supports Fe–S cluster reconstitution,

-1 albeit with a slightly slower rate (kderA ~ 0.080 min ) than that observed for native 48

- frataxin. Derivative C shows a significantly lower reconstitution rate (kderC ~ 0.058 min

1). One possible explanation is that the residues knocked out in derivative C can facilitate

iron delivery. However, this conclusion assumes that the derivatives bind to ISU in the

same orientation and relative position. Such an explanation is consistent with the higher

reconstitution rates of derivatives A and B, relative to derivative C. In the case of

derivative B, residues E92 and E96 that may facilitate iron transfer are retained, with

E108, E111, D112, and D115 (which are relatively closely positioned and include

conserved residues) presumably forming an acidic cluster that promotes both iron and

partner binding. For derivative A, residues E92 and E96 may again serve to promote iron

delivery, while residues D115, E121, D122, and D124 form an acidic cluster that

promotes both iron and partner binding.

Relative to the position of acidic patches, the overall structural features of both

frataxin and ISU appear to serve a minor role in dictating the relative binding affinity and

positioning of each protein. Consequently, the reaction sequence is likely to involve

initial binding of iron to a cluster of acidic residues that drives frataxin binding to ISU,

with subsequent transfer to the conserved cysteine residues in the Fe–S cluster binding

pocket of ISU. The latter transfer step may be facilitated by additional carboxylate

residues. If the acidic residues are not positioned to facilitate iron binding and positioning

in the neighborhood of the ISU cysteines, then the reconstitution rate will be

compromised.

2.4.4. Comparison of Frataxin Homologues

The structures of frataxin homologues are found to be extremely similar (19, 21,

39-42). All of them have a negatively charged surface that appears to promote the binding of iron ions and partner proteins. In turn this is reflected in their similarity in function, and specifically their role in support of Fe–S cluster biosynthesis (9, 10, 43).

Human frataxin has been observed to functionally complement frataxin deletion mutants in the yeast Saccharomyces cerevisiae (36). However, some variations in iron binding properties have been observed within this family of frataxin. Compared with Yfh1 in yeast (23), and the bacterial orthologue of frataxin, CyaY (44), human frataxin binds iron with slightly lower affinity (in the partner-free state), but with higher stoichiometry. The distances between some carboxylates on the anionic surface of Yfh1 and CyaY are generally farther than are observed for the human protein, which may eliminate stabilization of additional irons through formation of binuclear iron centers as suggested for human frataxin (24). The modest decrease in binding affinity for iron in the His- tagged version of the truncated protein most likely reflects a weak competing electrostatic interaction with the anionic surface. However, the thrombin-cleaved versions of frataxin proteins do demonstrate iron-binding results that are consistent with previous reports from this laboratory (9).

Cellular studies of frataxin derivatives that carry multiple substitutions within the anionic patch of Yfh1 show little change in cell viability, while immunoprecipitation results were comparable to those obtained with the native protein (34). In a related cellular study, only complete elimination of the putative iron binding sites, or changing 50

the negative acidic patch to a positively charged basic patch, had any significant effect

(32). This indicates that frataxin has the ability to accommodate the loss of many acidic residues; a result that is consistent with the biochemical results reported herein that frataxin can adapt to such changes by binding to partner proteins in possibly distinct

orientations by making use of other available acidic residues. That is, there is flexibility

in the mode of partner binding. Such a model helps to rationalize prior observations from the yeast homologue that alanine substitutions in the acidic patch impact iron binding in vitro, but not Fe–S cluster assembly in vivo (32).

2.5. Summary

In summary, iron binds to human frataxin in an extended fashion. Among the seven iron binding sites on human frataxin, one iron may associate with E92 and E96, another iron may bind to D115, E121, D122, and D124, and the remaining four or five irons may bind to E100, E101, E108, D112, and D115. The binding is localized; deletion of some acidic residues results in a decrease in the stoichiometry, while the relative affinity remains the same. Only one iron is needed to promote binding between frataxin and ISU, and frataxin derivatives bind to ISU with an affinity that is similar to that of native frataxin, even with only 1 equiv of iron present. In the Fe–S reconstitution assay, only derivative C shows a significant (though modest) decrease in activity. If the derivatives bind to ISU in the same orientation and relative position, E92 and E96 are likely to facilitate iron delivery; however, these two residues are not conserved, and so the possibility that the derivatives may bind in other ways cannot be ruled out.

While the overall number of potential iron binding ligands (carboxylate residues

and histidine) are similar for human and yeast frataxin, the number of iron binding sites

determined by ITC is distinct, being 7 and 2, respectively, with the yeast protein showing

modestly higher affinity. In both cases, all sites are not necessarily populated during

functional activity. The variation in iron binding stoichiometry most likely reflects the relative binding affinities at each potential site, where minor changes in ligand orientation

(replacing glutamate for aspartate or histidine) can reduce the affinity beyond limits

detectable by ITC. The overall consensus viewpoint from these and other data is to

consider a pool of potential sites that can stably bind an iron center when bridged to a

variety of physiological targets. In that sense, there may not be unique binding loci, but

rather a number of locations that provide flexibility in the binding of physiological

partner proteins.

Structurally, frataxin is defined by two major motifs. The anionic surface defined

by acidic residues on the a1 helix and b1 sheet, and a surface defined by 21 conserved

hydrophobic residues from strands β1–β5 that is nearly neutral. The presence of such two

patches creates a large dipole that is essential to the function of frataxin (19). With the

evidence in hand, it seems plausible that the general recognition of partner proteins is

driven both by the need for a bridging iron and also by a general dipole interaction that

accommodates flexibility in binding interactions.

2.6. References for Chapter 2

1. Warren, S. T. (1996) The expanding world of trinucleotide repeats, Science 271, 1374-1375.

2. Campuzano, V., Montermini, L., Molto, M. D., Pianese, L., Cossee, M., Cavalcanti, F., Monros, E., Rodius, F., Duclos, F., Monticelli, A., Zara, F., Canizares, J., Koutnikova, H., Bidichandani, S. I., Gellera, C., Brice, A., Trouillas, P., De Michele, G., Filla, A., De Frutos, R., Palau, F., Patel, P. I., Di Donato, S., Mandel, J. L., Cocozza, S., Koenig, M., and Pandolfo, M. (1996) Friedreich's ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion, Science 271, 1423-1427.

3. Durr, A., Cossee, M., Agid, Y., Campuzano, V., Mignard, C., Penet, C., Mandel, J. L., Brice, A., and Koenig, M. (1996) Clinical and genetic abnormalities in patients with Friedreich's ataxia, N Engl J Med 335, 1169-1175.

4. Rotig, A., de Lonlay, P., Chretien, D., Foury, F., Koenig, M., Sidi, D., Munnich, A., and Rustin, P. (1997) Aconitase and mitochondrial iron-sulphur protein deficiency in Friedreich ataxia, Nat Genet 17, 215-217.

5. Ristow, M., Pfister, M. F., Yee, A. J., Schubert, M., Michael, L., Zhang, C. Y., Ueki, K., Michael, M. D., 2nd, Lowell, B. B., and Kahn, C. R. (2000) Frataxin activates mitochondrial energy conversion and oxidative phosphorylation, Proc Natl Acad Sci U S A 97, 12239-12243.

6. Foury, F. (1999) Low iron concentration and aconitase deficiency in a yeast frataxin homologue deficient strain, FEBS Lett 456, 281-284.

7. Puccio, H., Simon, D., Cossee, M., Criqui-Filipe, P., Tiziano, F., Melki, J., Hindelang, C., Matyas, R., Rustin, P., and Koenig, M. (2001) Mouse models for Friedreich ataxia exhibit cardiomyopathy, sensory nerve defect and Fe-S enzyme deficiency followed by intramitochondrial iron deposits, Nat Genet 27, 181-186.

8. Cavadini, P., O'Neill, H. A., Benada, O., and Isaya, G. (2002) Assembly and iron- binding properties of human frataxin, the protein deficient in Friedreich ataxia, Hum Mol Genet 11, 217-227.

9. Yoon, T., and Cowan, J. A. (2003) Iron-sulfur cluster biosynthesis. Characterization of frataxin as an iron donor for assembly of [2Fe-2S] clusters in ISU-type proteins, J Am Chem Soc 125, 6078-6084.

10. Gerber, J., Muhlenhoff, U., and Lill, R. (2003) An interaction between frataxin and Isu1/Nfs1 that is crucial for Fe/S cluster synthesis on Isu1, EMBO Rep 4, 906- 911.

11. Muhlenhoff, U., Richhardt, N., Gerber, J., and Lill, R. (2002) Characterization of iron-sulfur protein assembly in isolated mitochondria. A requirement for ATP, NADH, and reduced iron, J Biol Chem 277, 29810-29816.

12. Muhlenhoff, U., Richhardt, N., Ristow, M., Kispal, G., and Lill, R. (2002) The yeast frataxin homolog Yfh1p plays a specific role in the maturation of cellular Fe/S proteins, Hum Mol Genet 11, 2025-2036.

13. Bencze, K. Z., Yoon, T., Millan-Pacheco, C., Bradley, P. B., Pastor, N., Cowan, J. A., and Stemmler, T. L. (2007) Human frataxin: iron and ferrochelatase binding surface, Chem Commun (Camb), 1798-1800.

14. Yoon, T., and Cowan, J. A. (2004) Frataxin-mediated iron delivery to ferrochelatase in the final step of heme biosynthesis, J Biol Chem 279, 25943- 25946.

15. Bulteau, A. L., O'Neill, H. A., Kennedy, M. C., Ikeda-Saito, M., Isaya, G., and Szweda, L. I. (2004) Frataxin acts as an iron chaperone protein to modulate mitochondrial aconitase activity, Science 305, 242-245.

16. Condo, I., Ventura, N., Malisan, F., Tomassini, B., and Testi, R. (2006) A pool of extramitochondrial frataxin that promotes cell survival, J Biol Chem 281, 16750- 16756.

17. Ventura, N., Rea, S. L., Handerson, S. T., Condo, I., Testi, R., and Johnson, T. E. (2006) C. elegans as a model for Friedreich Ataxia, Faseb J 20, 1029-1030.

18. Acquaviva, F., De Biase, I., Nezi, L., Ruggiero, G., Tatangelo, F., Pisano, C., Monticelli, A., Garbi, C., Acquaviva, A. M., and Cocozza, S. (2005) Extra- mitochondrial localisation of frataxin and its association with IscU1 during enterocyte-like differentiation of the human colon adenocarcinoma cell line Caco- 2, J Cell Sci 118, 3917-3924.

19. Dhe-Paganon, S., Shigeta, R., Chi, Y. I., Ristow, M., and Shoelson, S. E. (2000) Crystal structure of human frataxin, J Biol Chem 275, 30753-30756.

20. Baker, N. A., Sept, D., Joseph, S., Holst, M. J., and McCammon, J. A. (2001) Electrostatics of nanosystems: application to microtubules and the ribosome, Proc Natl Acad Sci U S A 98, 10037-10041.

21. He, Y., Alam, S. L., Proteasa, S. V., Zhang, Y., Lesuisse, E., Dancis, A., and Stemmler, T. L. (2004) Yeast frataxin solution structure, iron binding, and ferrochelatase interaction, Biochemistry 43, 16254-16262.

22. Bencze, K. Z., Kondapalli, K. C., Cook, J. D., McMahon, S., Millan-Pacheco, C., Pastor, N., and Stemmler, T. L. (2006) The structure and function of frataxin, Crit Rev Biochem Mol Biol 41, 269-291.

23. Cook, J. D., Bencze, K. Z., Jankovic, A. D., Crater, A. K., Busch, C. N., Bradley, P. B., Stemmler, A. J., Spaller, M. R., and Stemmler, T. L. (2006) Monomeric yeast frataxin is an iron-binding protein, Biochemistry 45, 7767-7777.

24. Yoon, T., Dizin, E., and Cowan, J. A. (2007) N-terminal iron-mediated self- cleavage of human frataxin: regulation of iron binding and complex formation with target proteins, J Biol Inorg Chem 12, 535-542.

25. Condo, I., Ventura, N., Malisan, F., Rufini, A., Tomassini, B., and Testi, R. (2007) In vivo maturation of human frataxin, Hum Mol Genet 16, 1534-1540.

26. Liu, J., Oganesyan, N., Shin, D. H., Jancarik, J., Yokota, H., Kim, R., and Kim, S. H. (2005) Structural characterization of an iron-sulfur cluster assembly protein IscU in a zinc-bound form, Proteins 59, 875-881.

27. Wu, S. P., Wu, G., Surerus, K. K., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis. Kinetic analysis of [2Fe-2S] cluster transfer from holo ISU to apo Fd: role of redox chemistry and a conserved aspartate, Biochemistry 41, 8876- 8885.

28. Foster, M. W., Mansy, S. S., Hwang, J., Penner-Hahn, J. E., Surerus, K. K., and Cowan, J. A. (2000) A mutant human IscU protein contains a stable [2Fe-2S](2+) center of possible functional significance, J Am Chem Soc 122, 6805-6806.

29. Schmucker, S., Argentini, M., Carelle-Calmels, N., Martelli, A., and Puccio, H. (2008) The in vivo mitochondrial two-step maturation of human frataxin, Hum Mol Genet 17, 3521-3531.

30. Ding, H. G., and Clark, R. J. (2004) Characterization of iron binding in IscA, an ancient iron-sulphur cluster assembly protein, Biochem J 379, 433-440.

31. Ishikawa, T., Mizunoe, Y., Kawabata, S., Takade, A., Harada, M., Wai, S. N., and Yoshida, S. (2003) The iron-binding protein Dps confers hydrogen peroxide stress resistance to Campylobacter jejuni, J Bacteriol 185, 1010-1017.

32. Foury, F., Pastore, A., and Trincal, M. (2007) Acidic residues of yeast frataxin have an essential role in Fe-S cluster assembly, EMBO Rep 8, 194-199.

33. Lesuisse, E., Santos, R., Matzanke, B. F., Knight, S. A., Camadro, J. M., and Dancis, A. (2003) Iron use for haeme synthesis is under control of the yeast frataxin homologue (Yfh1), Hum Mol Genet 12, 879-889.

34. 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, 1096-1101.

35. Gakh, O., Park, S., Liu, G., Macomber, L., Imlay, J. A., Ferreira, G. C., and Isaya, G. (2006) Mitochondrial iron detoxification is a primary function of frataxin that limits oxidative damage and preserves cell longevity, Hum Mol Genet 15, 467- 479.

36. Cavadini, P., Gellera, C., Patel, P. I., and Isaya, G. (2000) Human frataxin maintains mitochondrial iron homeostasis in Saccharomyces cerevisiae, Hum Mol Genet 9, 2523-2530.

37. Kaiser, J. T., Clausen, T., Bourenkow, G. P., Bartunik, H. D., Steinbacher, S., and Huber, R. (2000) Crystal structure of a NifS-like protein from Thermotoga maritima: Implications for iron sulphur cluster assembly, J Mol Biol 297, 451-464.

38. Liu, Y., and Cowan, J. A. (2007) Iron sulfur cluster biosynthesis. Human NFU mediates sulfide delivery to ISU in the final step of [2Fe-2S] cluster assembly, Chem Commun (Camb), 3192-3194.

39. Cho, S. J., Lee, M. G., Yang, J. K., Lee, J. Y., Song, H. K., and Suh, S. W. (2000) Crystal structure of Escherichia coli CyaY protein reveals a previously unidentified fold for the evolutionarily conserved frataxin family, Proc Natl Acad Sci U S A 97, 8932-8937.

40. Lee, M. G., Cho, S. J., Yang, J. K., Song, H. K., and Suh, S. W. (2000) Crystallization and preliminary X-ray crystallographic analysis of Escherichia coli CyaY, a structural homologue of human frataxin, Acta Crystallogr D Biol Crystallogr 56, 920-921.

41. Musco, G., Stier, G., Kolmerer, B., Adinolfi, S., Martin, S., Frenkiel, T., Gibson, T., and Pastore, A. (2000) Towards a structural understanding of Friedreich's ataxia: the solution structure of frataxin, Structure 8, 695-707.

42. Nair, M., Adinolfi, S., Pastore, C., Kelly, G., Temussi, P., and Pastore, A. (2004) Solution structure of the bacterial frataxin ortholog, CyaY: mapping the iron binding sites, Structure 12, 2037-2048.

43. Layer, G., Ollagnier-de Choudens, S., Sanakis, Y., and Fontecave, M. (2006) Iron-sulfur cluster biosynthesis: characterization of Escherichia coli CYaY as an

iron donor for the assembly of [2Fe-2S] clusters in the scaffold IscU, J Biol Chem 281, 16256-16263.

44. Bou-Abdallah, F., Adinolfi, S., Pastore, A., Laue, T. M., and Dennis Chasteen, N. (2004) Iron binding and oxidation kinetics in frataxin CyaY of Escherichia coli, J Mol Biol 341, 605-615.

CHAPTER 3

IRON-SULFUR CLUSTER BIOSYNTHESIS: ROLE OF A CONSERVED HISTIDINE

3.1. Introduction

Human ISU is an iron-sulfur cluster scaffold protein that assembles [2Fe-2S] clusters prior to transfer to a target protein (1, 2). The ISU/IscU family of proteins are highly conserved, both structurally and functionally, and represent a key component of the iron-sulfur cluster (ISC) biosynthetic apparatus (3). All ISU-type proteins contain three highly conserved cysteine residues (Cys44, Cys70, and Cys113 in the case of human ISU) that mutagenesis studies suggest to be directly coordinated to the cluster (4). The identity of the fourth ligand has remained elusive, although circular dichroism studies support an oxygen (possibly H2O) or nitrogen donor ligand (5).

Several structures of zinc-bound IscU’s have been determined, and all share a highly conserved α + β globular core with the three conserved cysteines that form the iron- sulfur cluster binding pocket located on a solvent-accessible surface (6). In the zinc- bound forms the non-cysteinyl ligand is either an aspartate or a histidine residue (6),

where aspartate is used in the case of both gram-positive bacteria and thermophiles

that lack the conserved His. A recent structural study of the [2Fe-2S]-loaded A.

aeolicus IscU shows His as the non-cysteinyl ligand, and studies with Azotobacter

vinelandii (Av) also suggest the presence of a coordinated histidine (7, 8). While a

His residue is retained in human ISU, both gram-positivie bacteria and thermophiles show a positively-charged lysine at the corresponding position (Figure 3.1), and both crystallographic and NMR structures of the zinc-bound IscU’s from gram-positive

Bacillus subtilis and Streptococcus pyogenes show the lysine turned away from the zinc with a highly conserved carboxylate residue as a coordinating ligand. While this conserved aspartate is used as a ligand in the zinc-bound state, this is unlikley to coordinate to the cluster, since we have previously shown that substitution of this Asp with Ala results in an overall increase in cluster stability. In these cases the most likely candidate for non-cysteinyl ligation to the [2Fe-2S] cluster remains a water molecule.

SA ARIKCATLAWKALEKGTVAKEGKAEGTTEEE------154 BA ARIKCATLAWKALEKG--LNEDK------143 SP QRIKCSTLAWNALKEAIKRSANAQHLTDQNVKEGKNV 159 TM ARVKCFILAWKTLKEALKKISRP------142 EC VKIHCSILAEDAIKAAIADYKSKREAK------128 HS VKLHCSMLAEDAIKAALADYKLKQEPKKGEAEKK--- 142

Figure 3.1. C-terminal sequence alignment of bacterial and human ISU-type proteins. S.

aureus (SA), B. anthracis (BA), Streptococcus pyogenes (SP), T. maritima (TM), E. coli

(EC), Homo sapiens (HS). IscU proteins from TM and the gram positive pathogens are

highly homologous and the insert (residues 92-109, Ba numbering) is clear. A conserved

cluster binding Cys residues are highlighted in yellow, the semi-conserved His in green,

and Lys from gram positive and thermophiles in red.

Figure 3.1 shows His112 neighboring the cluster coordinating ligand Cys113

(human numbering), and the Lys that is found in gram-positive bacteria and thermophiles. To explore the specific role of His112, derivatives of ISU, H112A and

H112D, were generated with wild type human ISU or D37A ISU, which can form a more stable [2Fe-2S] cluster, by site-directed mutagenesis (4). The influence of these mutations was investigated. Herein are reported the results of experiments designed to elucidate the role of this semi-conserved His in defining cluster stability and/or mediating assembly of the [2Fe-2S] cluster, and how these roles are accommodated in gram-positive and thermophilic bacteria that lack the His.

3.2. Materials and Methods

3.2.1. General Chemicals

HEPES, Tris, DTT, Na2S·10H2O, Fe(NH4)2(SO4)2· 6H2O, NaCl and G-25 were purchased from Sigma-Aldrich (St. Louis, MO).

3.2.2. Mutagenesis, Expression and Purification of Proteins

Two human ISU derivatives, H112A and H112D were prepared using site- directed mutagenesis with D37A ISU following the procedure provided by Stratagene’s

QuickChange Mutagenesis kit. The sequences were confirmed by DNA sequencing. The plasmids with the correct sequences were transformed to BL21DE3. A 10 ml culture was grown overnight at 37 ºC to inoculate a 1 L culture which was first allowed to shake for

3.5 h before adding IPTG to 1 mM. After 4 h, the cell was harvested via centrifuging at

5,000 rpm for 10 minutes. The harvested cell was resuspended with binding buffer (20 mM Tris, 500 mM NaCl, pH 7.9) and 5 mM imidazole, and subsequently broken by sonication. The crude lysate was cleared by centrifuging at 15000 rpm for 45 minutes.

After that, the supernatant was loaded onto a Ni-NTA column previously equilibrated with the same binding buffer. The column was washed with binding buffer plus 25 mM imidazole. The protein was finally eluted with binding buffer with 200 mM imidazole.

The proteins were dialyzed against a buffer containing 50 mM Hepe, 50 mM NaCl,

pH7.5, and stored at -80 ºC until use.

3.2.3. Chemical Reconstitution with ISU and ISU Derivatives

ISU proteins were dissolved in buffer (50 mM Hepes, 100 mM NaCl, pH 7.5)

containing 50 mM DTT. Subsequently, FeCl3 and Na2S were added from stock

solutions, with continuous stirring, to a final concentration of 1 mM. All solutions were Ar-purged prior to use, and the reaction mixture allowed stirred under Ar for another 30 min. The mixture was then centrifuged at 12,000 rpm for 1 min and concentrated before desalting on a G-25 column equilibrated with 50 mM Hepes, 100 mM NaCl, pH 7.5.

3.2.4. UV-Vis Spectroscopy for ISU and ISU Derivatives

Apo or holo ISU proteins were subject to dialysis after they were purified by

Ni-NTA column or G-25 column chromatographies. Buffer that contains the same

component as in the dialysis was used as the reference. Subsequently the UV

absorbance from 250 nm to 700 nm of the protein solutions were recorded.

3.2.5. Quantification of Iron Content and Extinction Coefficient of Cluster

One milliliter protein solution was mixed with 0.3 mL of 12 M HCl in Ependorff

tubes, followed by heating to 100 C for 15 min. To the solution was then added 0.4 ml

distilled water. The precipitate was removed by centrifuging at 15,000 rpm for 4 min and

aliquots from the supernatant was transferred and diluted to 1.5 mL with 0.5 M Tris-HCl

buffer, pH 8.5. Sodium dithionite (0.1 mL, 5% in water) and 0.4 mL of 2,2'-bipyridine

(0.1% in water) were added to the solution. After 1 h, the absorbance was measured at

520 nm against a control containing the buffer and the reagents. The molar extinction

coefficient of iron-2,2'-bipyridine complex was taken as 8 400 M-1cm-1. The extinction

coefficient of cluster was calculated by dividing the UV absorbance of cluster at 456 nm

by the concentration of cluster which was calculated by assuming the iron concentration was double the concentration of cluster.

3.2.6. Studies on Iron Binding Capability of ISU and ISU Derivatives via ITC

The proteins used were dialyzed against 50 mM Hepes, and 100 mM NaCl at pH 7.5. All experiments were carried out at 25°C. To a solution of around 50 µM ISU or ISU derivatives was added 10 µL of titrant containing 2 mM Fe(NH4)2(SO4)2 over a period of 24 s, with an adequate interval (4 or 5 min) between injections to allow complete equilibration. Both titrant and solution in sample cell were supplemented

with 2 mM sodium dithionite. The data were collected automatically and subsequently

analyzed with a one-site binding model by the Windows-based Origin software package supplied by MicroCal. ITC experiments were performed on a MicroCal

OMEGA ultrasensitive titration calorimeter.

3.2.7. Studies on the Interaction between Frataxin and ISU and ISU Derivatives via

ITC

To a solution of 20 µM frataxin was added 10 µL of titrant containing ~ 200

µM D37A ISU or its derivatives over a period of 24 s, with an adequate interval (4 or

5 min) between injections to allow complete equilibration. Both titrant and solution in

sample cell were supplemented with 200 µM Fe(NH4)2(SO4)2 and 2 mM sodium dithionite. The background titration, consisting of the identical titrant solution, omitting frataxin in the solution for the sample cell , was subtracted from each experimental titration to account for heat of dilution.

3.2.8. Tm Nifs-assisted Reconstitution

Reaction mixtures were prepared under anearonic condition and contained 1

µM Tm Nifs, 20 µM human frataxin, 50 µM human ISU or derivatives, 100 µM

Fe(NH4)2(SO4)2, 2 mM DTT. The reaction was initiated by adding 2 mM cysteine.

The UV absorbance at 456 nm was monitored every 5 min for an hour. UV-Vis spectra were obtained on a Hewlett-Packard HP 8452A Diode Array UV-Vis

Spectrophotometer.

3.3. Results

3.3.1. Expression and Purification of ISU Derivatives

The purification procedures for the mutant proteins were found to be similar to

D37A ISU (4). Figure 3.2 shows the whole cell lysate before and after adding IPTG on SDS-PAGE, along with the aliquots taken after each step of the purification for

H112D/D37A and H112A/D37A ISU with Ni-NTA column. Different from the fact that D37A ISU appeared to be red, H112D/D37A ISU appeared to be brown and

H112A/D37A ISU was colorless when it was eluted from the Ni-NTA column native

ISU and derivatives did not stably bind cluster to any significant extent.

(a) 1 2 3 4 5

20.1 kDa

14.4 kDa

(b) 1 2 3 4 5

20.1 kDa

14.4 kDa

Figure 3.2. Expression and purification of H112D/D37A (a) and H112A/D37A ISU (b); a) lane 1, low molecular marker (the molecular weights are 97.0, 66.0, 45.0, 30.0 and 20.1 kDa from high to low molecular weight); lane 2, before adding IPTG; lane 3, after adding

IPTG; lane 4, flowthrough from Ni-NTA column; lane 5, wash with 25 mM imidazole in the binding buffer; lane 6, elution from Ni-NTA column with 200 mM imidazole. b) lane

1, low molecular marker; lane 2, lysate; lane 3, flowthrough; lane 4, wash with 25 mM imidazole in the binding buffer; lane 5, elution from Ni-NTA column with 200 mM imidazole.

3.3.2. UV-Vis Spectroscopy of ISU and its Derivatives

The UV spectra of H112D/D37A or H112A/D37A ISU following Ni-NTA

column and G-25 purification are shown in Figure 3.3. H112D/D37A ISU showed a

broad band around 420 nm typical of [2Fe-2S] clusters, while H112A/D37A did not

show any appreciable bands above 280 nm and was colorless. The yields of the two

derivatives were approximately two third of that for D37A ISU. Under this

circumstance, the change of color could mean that the cluster binding environment in

H112D/D37A ISU could have been modified, and the ability of H112A/D37A ISU to

form iron-sulfur cluster had been undermined since the rate of iron-sulfur cluster

biosynthesis in E coli can catch up with that of the protein production. Purified native

ISU was in apo form. Because of the presence of the Asp37 which controls the solvent accesibility of the cluster, it is not suprising that the mutants generated with native ISU did not bind to a cluster, since in most cases, the derivatives are not the optimal state.

0.5 0.40

0.35 0.4

0.30

0.3 0.25

0.20 0.2

0.15 Absorbance Absorbance

0.10 0.1

0.05 0.0 0.00

200 300 400 500 600 700 200 300 400 500 600 700 Wavelength (nm) Wavelength (nm)

Figure 3.3. H105D/D37A (left, two different concentrations) and H112A/D37A ISU

(right) right after elution from Ni-NTA column.

3.3.3. Chemical Reconstitution with ISU and Derivatives

Iron-sulfur cluster can also be formed in vitro by supplying inorganic iron and sulfide under anaerobic conditions (9-11). H112D/D37A and H112A/D37A ISU were both brown in color, similar to H112D/D37A ISU produced by overexpressing in E

coli. The UV spectra of chemically reconstituted proteins are shown in Figure 3.4.

The peak maxima for each protein were generally found to be similar, allowing for

the underlying protein absorbance. The peak maxima for H112D/D37A ISU was less clearly defined; most likely reflecting the replacement of a protein ligand with water

and the relative instability of the cluster. The spectra of other derivatives were

similar to native, with clearly defined bands at ~ 330 nm, 420 nm and 456 nm, with a

broad band evident above 500 nm. While less stable, the cluster could also be formed

for H112D and H112A derivatives of native ISU. This implies that the presence of

H112 is not absolutely necessary to stably bind the cluster to ISU; however, the difference in color compared to holo D37A or native ISU indicates that H112 does play roles in the formation of the clusters, and most likely, in the process of forming clusters in vivo.

1.5

1.0

0.5 Absorbance

0.0 300 400 500 600 700

Wavelength (nm)

Figure 3.4. UV spectra of native (blue), Asp37Ala (red), His103Asp/Asp37Ala (green) and His103Ala/Asp37Ala (black) ISU after chemical reconstitution. The spectra were scaled according to relative cluster content to allow for comparsion of cluster absorbance features. Collected data did not extend beyond the regular Beer limit. Relative iron concentration was evaluated by use of a standard bipyridine test, and the data represent an effective [2Fe-2S]2+ concentration of 50 μM. Relative extinction coefficients at 456

nm were determined as 9.8 × 103 M-1cm-1 (native); 9.4 × 103 M-1cm-1 (Asp37A); 1.0 ×

104 M-1cm-1 (Asp37Ala/His103Ala); 8.1 × 103 M-1cm-1 (Asp37Ala/His103Asp). Spectra

were obtained in 50 mM Hepes, 100 mM NaCl, pH 7.5.

3.3.4. Quantification of Iron Content and Extinction Coefficient of Cluster

Extinction coefficients were evaluated following quantitation of bound iron to

evaluate cluster concentration in the holo protein. Minor variations in extinction coeffcient were observed (Table 3.1). The slight variations in relative extinction coefficients suggest that His112 does influence the cluster either directly through coordination, or indirectly by influencing solvent access and the polarity of the cluster

binding pocket.

Extinction coefficient (M-1cm-1) Native ISU 9.8 × 103 D37A ISU 9.4 × 103 H112D/D37A ISU 8.1 × 103 H112A/D37A ISU 1.0 × 104

Table 3.1. Extinction coefficient of human ISU and derivatives.

3.3.5. Studies on Iron Binding Capability of ISU and derivatives via ITC

It is still not clear where the iron ions bind to ISU, however, several structures

of bacterial IscU proteins showed a zinc ion coordinated by three cysteines and a

histidine at the active sites(12, 13) or three cysteines and an aspartate or glutamate for

IscU from Bacillus subtilis (PDB ID: 1xjs). Given the similarity between zinc and

ferrous ion, the iron binding site could also be the zinc binding site. Iron binding to native, D37A, H112A/D37A, and H112A/D37A ISU’s was evaluated by ITC (Figure

3.5). A modest increase in affinity following H112D substitution, and a lower affinity for iron binding with H112A were observed (Table 3.2). The magnitude of these

changes does not support a major requirement for His to coordinate the cluster to the

ISU scaffold, or to stably bind iron in the cluster assembly pocket. The three cysteines

can still provide tight enough binding. Substitution of His did, however, result in a

more favorable entropy change (and lower enthalpic component) for iron binding,

consistent with coordination by a harder oxygen ligand (carboxylate or water) (14).

Protein KD (μM) ΔH (kcal/mol) ΔS (cal/K·mol) native 1.9 ± 0.2 -2.9 ± 0.3 17 ± 0.2 D37A 2.0 ± 0.2 -2.9 ± 0.3 16 ± 0.2 H103D/D37A 0.9 ± 1.0 -0.4 ± 0.1 26 ± 0.3 H103A/D37A 1.3 ± 0.1 -0.9 ± 0.1 24 ± 0.2

Table 3.2. Summary of iron binding study on ISU and ISU derivatives via ITC.

Time (min) Time (min) -15 0 15 30 45 60 75 90 105 120 135 150 165 -15 0 15 30 45 60 75 90 105 120 135 150 165

0.0 0.0

-0.2 -0.5 -0.4

-0.6 -1.0 cal/sec cal/sec μ μ -0.8 -1.5 -1.0 D37A WT -1.2 -2.0

0 0

-2 kcal/mole ofinjectant kcal/mole injectant of -2 0123456789 0123456789 Molar Ratio Molar Ratio

Time (min) Time (min) -15 0 15 30 45 60 75 90 105 120 135 150 165 -15 0 15 30 45 60 75 90 105 120 135 150 165 0.05 0.05 0.00 0.00 -0.05 -0.05 -0.10 -0.10 -0.15 -0.15 -0.20 -0.20 -0.25 cal/sec cal/sec

-0.25 μ μ -0.30 -0.30 -0.35 -0.35 H103D/D37A -0.40 H103A/D37A -0.40 0.1 0.0 0.0 -0.1 -0.1 -0.2 -0.2 -0.3 -0.4 -0.3 -0.5

-0.4 -0.6 -0.7 kcal/mole of injectant of kcal/mole kcal/mole of injectant of kcal/mole -0.5 -0.8

0123456789 0123456789 Molar Ratio Molar Ratio

Figure 3.5. Studies on iron binding capability of ISU or ISU derivatives via ITC.

3.3.6. The Interaction between Frataxin and ISU or ISU derivatives via ITC

Prior studies have suggested that delivery of iron from frataxin to ISU is most

likely rate limiting (15). To determine if the variation in rate constants for cluster

assembly reflected non-saturating binding of frataxin to the derivative ISU proteins

the affinity of the ISU derivatives and frataxin were determined by ITC. Binding stoichometries were all close to one, with relative KD’s of 10.4 ± 0.3 µM and 4.7 ±

0.9 µM for H112A/D37A ISU and H112D/D37A, respectively, compared to 1.2 ± 0.1

µM for D37A ISU (Figure 3.6). Accordingly, the concentration of frataxin used in

reconstitution studies was saturating, and the variations in assembly rate constants do

not reflect the extent of bound frataxin.

Time (min) Time (min) -10 0 10 20 30 40 50 60 70 80 90 100 -10 0 10 20 30 40 50 60 70 80 90 100 0.05 0.05

0.00 0.00

-0.05 -0.05

-0.10 -0.10

-0.15 -0.15 cal/sec cal/sec μ μ -0.20 -0.20 -0.25 -0.25 D37A ISU H112D/D37A ISU -0.300

-2 -2 kcal/mole of injectant kcal/mole of injectant

0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Molar Ratio Molar Ratio

Time (min) 0 20 40 60 80 100 120 140 160

0.00

-0.05

-0.10 cal/sec

μ -0.15

-0.20 H112A/D37A ISU

-2 kcal/mole of injectant

0.0 0.5 1.0 1.5 2.0 2.5 Molar Ratio

Figure 3.6. Interaction between D37A ISU or derivatives and frataxin in the presence of ferrous ion.

3.3.7. Tm Nifs-assisted Reconstitution

In vitro reconstitution with Tm Nifs, a cysteine desulfurase, cysteine, frataxin and ferrous ion was used to mimic the situation in vivo to see if the derivatives can form iron-sulfur cluster, where Nifs/cysteine mediates enzymatic production of sulfide and frataxin promotes iron delivery (16). Previous work from our lab has shown that the use of Tm Nifs can result in successful reconstitution of human ISU

(15). Derivative H112A/D37A showed a negligible rate of reconstitution, relative to the control, while H112D/D37A showed a slower reconstitution rate compared to

D37A ISU (Figure 3.7 and Table 3.3).

0.055

0.050

0.045 0.040

0.035

0.030

0.025

0.020

Absorbance 0.015 0.010

0.005

0.000

-0.005 0 102030405060 Time (min)

Figure 3.7. Tm Nifs-assisted reconstitution of D37A ISU ( ), H112D/D37A ISU ( ),

H112A/D37A ISU ( ) and a control lacking ISU ( ).The reaction was followed by monitoring the UV absorbance at 456 nm over time.

ISU derivatives ΔA/Δt Native 1.20 × 10-3 D37A 1.24 × 10-3 H112D/D37A 7.92 × 10-4 H112A/D37A 2.64 × 10-4 No ISU 1.88 × 10-4

Table 3.3. Initial velocities of rate of reconstitution with ISU and derivatives. 78

3.4. Discussion

His112 may have an important functional role in the process of initial formation of cluster. H112A/D37A ISU cannot form the iron-sulfur cluster when it was overexpressed in E. coli, neither does it in Nifs-assisted reconstitution. However, it can form the iron-sulfur cluster via chemical reconstitution. This implies that H112A/D37A

ISU has the ability to bind to cluster, and the high concentration of ferrous ion and sulfide may overcome the absence of His112, although, lacking the ability to form cluster under physiological conditions.

His112 could be the residue involved in binding to frataxin and chelating iron.

This is evidenced by the weaker interaction between the two proteins and a significantly impaired rate in the Tm Nifs-assisted reconstitution. This is also consistent with the result of H112D/D37A ISU, which showed intermediate behavior. Aspartic acid, although not an ideal fit in the active site, can still interact with iron, and so the binding affinity is not significantly lowered. The modest decrease in the binding affinity is also consistent with the iron binding studies via ITC in which both derivatives showed similar binding affinity to ferrous ion compared with D37A ISU. H112D/D37A ISU forms clusters in E coli less efficiently than D37A ISU, in contrast to H112A/D37A ISU, which does not form clusters under the same conditions.

The different colors of the purified D37A ISU and H112D/D37A ISU from E coli suggests that His112 could interact with or coordinate the cluster as a ligand.

Holo D37A ISU is more stable than holo H112D/D37A and H112A/D37A ISU suggesting that His112 does have a stabilizing role.

A structural study of Aquifex aeolicus IscU is in support of our analysis (6).

His106 of Aquifex aeolicus IscU is the fourth ligand coordinating the [2Fe-2S] cluster

in the protein overexpressed and purified in holo form. The corresponding ligand of

His106 in Human ISU is His112 and the other three cysteinal ligands in Aquifex aeolicus IscU correspond to the conserved cysteines mentioned earlier. Other studies

with Azotobacter vinelandii (Av) also suggested the presence of the histidine ligand

(17, 18). In some gram-positive bacteria, there is a positively charged lysine at the

corresponding position. This could argue against the role of His112. However, in the

crystal or NMR structures of the gram-positive Bacillus subtilis and Streptococcus

pyogenes, the lysine turned away and a highly conserved aspartic acid or glutamic

acid becomes the ligand instead. Therefore, gram-positive bacteria could use either an

aspartic acid or glutamic acid to deliver iron and form a cluster which is possible

according to the results of H112D/D37A ISU.

In summary, substitution of H112 with Asp and Ala residues result in ISU

derivatives that have the capacity to coordinate a [2Fe-2S] cluster. Relatively stable

cluster formation for H112D/D37A and H112A/D37A substituted ISU derivatives is

consistent with a non-essential coordinating role for His, with either carboxylate or

water as suitable ligand replacements. His112 conveys a modest stabilizing influence

on the cluster, but is not essential. Water coordination appears the most likely

possibility for IscU’s from gram-positivie bacteria and thermophiles. The semi-

conserved His would appear to be principally involved in mediating iron delivery to

the cluster pocket, presumably following transfer from frataxin. This role appears to

be taken up by a conserved asparate residue in the case of gram-positivie bacteria and

thermophiles.

3.5. References for Chapter 3

1. Agar, J. N., Krebs, C., Frazzon, J., Huynh, B. H., Dean, D. R., and Johnson, M. K. (2000) IscU as a scaffold for iron-sulfur cluster biosynthesis: Sequential assembly of [2Fe-2S] and [4Fe-4S] clusters in IscU, Biochemistry 39, 7856-7862.

2. Mansy, S. S., Wu, G., Surerus, K. K., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis. Thermatoga maritima IscU is a structured iron-sulfur cluster assembly protein, J Biol Chem 277, 21397-21404.

3. Johnson, D. C., Dean, D. R., Smith, A. D., and Johnson, M. K. (2005) Structure, function, and formation of biological iron-sulfur clusters, Annu Rev Biochem 74, 247-281.

4. Foster, M. W., Mansy, S. S., Hwang, J., Penner-Hahn, J. E., Surerus, K. K., and Cowan, J. A. (2000) A mutant human IscU protein contains a stable [2Fe-2S](2+) center of possible functional significance, J. Am. Chem. Soc 122, 6805-6806.

5. Mansy, S. S., and Cowan, J. A. (2004) Iron-sulfur cluster biosynthesis: toward an understanding of cellular machinery and molecular mechanism, Acc Chem Res 37, 719-725.

6. Shimomura, Y., Wada, K., Fukuyama, K., and Takahashi, Y. (2008) The Asymmetric Trimeric Architecture of [2Fe-2S] IscU: Implications for Its Scaffolding during Iron-Sulfur Cluster Biosynthesis, J Mol Biol. 383, 133-143.

7. Huang, J., Dizin, E., and Cowan, J. A. (2008) Mapping iron binding sites on human frataxin: implications for cluster assembly on the ISUFe-S cluster scaffold protein, J. Biol. Inorg. Chem. 13, 825-836.

8. Pastore, C., Franzese, M., Sica, F., Temussi, P., and Pastore, A. (2007) Understanding the binding properties of an unusual metal-binding protein--a study of bacterial frataxin, FEBS J 274, 4199-4210.

9. Wu, G., Mansy, S. S., Wu, S. P., Surerus, K. K., Foster, M. W., and Cowan, J. A. (2002) Characterization of an iron-sulfur cluster assembly protein (ISU1) from Schizosaccharomyces pombe, Biochemistry 41, 5024-5032.

10. Mansy, S. S., Wu, G., Surerus, K. K., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis - Thermatoga maritima IscU is a structured iron-sulfur cluster assembly protein, Journal of Biological Chemistry 277, 21397-21404.

11. Hoff, K. G., Silberg, J. J., and Vickery, L. E. (2000) Interaction of the iron-sulfur cluster assembly protein IscU with the Hsc66/Hsc20 molecular chaperone system of Escherichia coli, Proceedings of the National Academy of Sciences of the United States of America 97, 7790-7795.

12. Ramelot, T. A., Cort, J. R., Goldsmith-Fischman, S., Kornhaber, G. J., Xiao, R., Shastry, R., Acton, T. B., Honig, B., Montelione, G. T., and Kennedy, M. A. (2004) Solution NMR structure of the iron-sulfur cluster assembly protein U (IscU) with zinc bound at the active site, J Mol Biol 344, 567-583.

13. Liu, J., Oganesyan, N., Shin, D. H., Jancarik, J., Yokota, H., Kim, R., and Kim, S. H. (2005) Structural characterization of an iron-sulfur cluster assembly protein IscU in a zinc-bound form, Proteins 59, 875-881.

14. Muhlenhoff, 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, 4815-4825.

15. Yoon, T., and Cowan, J. A. (2003) Iron-sulfur cluster biosynthesis. Characterization of frataxin as an iron donor for assembly of [2Fe-2S] clusters in ISU-type proteins, J Am Chem Soc 125, 6078-6084.

16. Yuvaniyama, P., Agar, J. N., Cash, V. L., Johnson, M. K., and Dean, D. R. (2000) NifS-directed assembly of a transient [2Fe-2S] cluster within the NifU protein, Proc Natl Acad Sci U S A 97, 599-604.

17. Johnson, D. C., Unciuleac, M. C., and Dean, D. R. (2006) Controlled expression and functional analysis of iron-sulfur cluster biosynthetic components within Azotobacter vinelandii, J Bacteriol 188, 7551-7561.

18. Agar, J. N., Krebs, C., Frazzon, J., Huynh, B. H., Dean, D. R., and Johnson, M. K. (2000) IscU as a scaffold for iron-sulfur cluster biosynthesis: sequential assembly of [2Fe-2S] and [4Fe-4S] clusters in IscU, Biochemistry 39, 7856-7862.

CHAPTER 4

HUMAN FRATAXIN PROMOTES CYTOSOLIC REPAIR OF IRP1

4.1. Introduction

Iron regulatory protein 1 (IRP1), a member of the aconitase gene family (1), is a

bifunctional cytosolic mammalian protein that underpins the cellular sensory apparatus

controlling cytosolic iron concentration (2). Under iron-replete condition, IRP1 exists in

an [4Fe-4S] holo state that displays aconitase activity, but lacks the ability to bind iron

regulatory elements (IRE’s) in mRNA’s; while under iron depleted conditions the [4Fe-

4S] cluster is disassembled to form an apo-state that does bind to IRE’s (3). The loss of

the cluster is accompanied by extensive conformational change (4) and apo IRP1 displays

high binding affinity towards the IRE’s of mRNA’s involved in iron uptake, transport and

storage proteins (5-9). Depending on where the apo IRP1 binds, the translation of coded

protein is either enhanced or depressed (2). And so IRP1 regulates the optimal

composition of proteins involved in iron homeostasis by switching between [4Fe-4S]-

and apo-form, while also maintaining the labile iron pool at a safe level for cellular

chemistry (3). 83

Recent solution and cellular studies suggest human frataxin to promote

reactivation of a [3Fe-4S] mitochondrial aconitase following loss of iron from oxidative

damage (10). A role for citrate was suggested, however, these data provided little insight on the molecular pathways for frataxin-mediated reactivation of the aconitase cluster and potential roles for citrate in promoting frataxin binding and/ or iron delivery chemistry. In humans, a deficiency of functional frataxin is an underlying cause of Friedreich's ataxia

(FRDA), and is characterized by severe disruption of iron homeostasis in a patient’s cells.

The activities of Fe-S cluster containing enzymes are significantly diminished, and iron accumulation is observed in the mitochondrion (11-14). Following mitochondrial import in human cells, the mitochondrial targeting sequence (residues 1-55) is removed in a two- step process by mitochondrial processing peptidases located in the matrix (15). Recent studies (16-18) have demonstrated human frataxin to be truncated beyond the expected residue site following removal of the mitochondrial targeting sequence through the action of two mitochondrial processing peptidases (MPP’s). Cellular studies by Testi and coworkers have shown that this truncated form of human frataxin is also formed in vivo by cleavage of a longer precursor protein, and is a principal active form within the cell

(16). The truncated form is, in fact, similar in overall size to that of the yeast and other bacterial frataxin homologues (Figure 4.1). Prior research in our laboratory has suggested that this truncation is the result of a catalytic self-cleavage reaction rather than a result of protease activity. Moreover, the self-cleavage activity appears to be stimulated by iron ion (18).

human 78 LRKSGTLGHPG-----SLDETTYERLAEETLDSLAEFFEDLADKPYTFEDYDVSFGSGVL 132 yeast 52 -VESSTDGQVVPQEVLNLPLEKYHEEADDYLDHLLDSLEELSEAHPDCIP-DVELSHGVM 109 E. coli 1 ------MNDSEFHRLADQLWLTIEERLDDWDGDS----DIDCEINGGVL 39 : :.. *:: : : ::: * .:. **:

human TVKLGGDLGTYVINKQTPNKQIWLSSPSSGPKRYDWTGKNWVYSHDGVSLHELLAAELTK 192 yeast TLEIP-AFGTYVINKQPPNKQIWLASPLSGPNRFDLLNGEWVSLRNGTKLTDILTEEVEK 118 E. coli TITFE-NGSKIIINRQEPLHQVWLATKQGG-YHFDLKGDEWICDRSGETFWDLLEQAATQ 97 *: : .. :**:* * :*:**:: .* ::* . :*: :.* .: ::*

human ALKTKLDLSSLAYSGKDA 210 yeast AISKSQ------174 E. coli QAGETVSFR------106

Figure 4.1. Multiple sequence alignment generated by CLUSTALW, where “*” indicates residues that are identical in all sequences in the alignment; “:" indicates positions where conserved substitutions are observed; and "." indicates positions where semi-conserved substitutions are observed.

We have previously suggested that full-length frataxin could serve primarily in cytosolic roles, while the truncated form serves in mitochondrial iron cofactor biosynthesis. In fact, overexpression of frataxin in the cytosol promotes FRDA cell survival (16), while it has also been reported that there is more cellular cytoplasmic frataxin than mitochondrial, and that cytoplasmic frataxin turns over threefold faster than the mitochondrial frataxin (20). Moreover, the possibility of distinct forms of frataxin

(full-length and truncated in the cytosol and mitochondrion, respectively) offers the potential for unique cytosolic roles and targeting opportunities for the full-length protein.

It is clearly of interest to distinguish between the activities of full-length and truncated frataxin toward the cytosolic IRP1, since a cytoplasmic pool of full-length frataxin has been demonstrated in several different cell types (17) and functional differentiation could provide insights on potential regulatory mechanisms underlying the cellular role for each.

4.2. Materials and Methods

4.2.1. General Chemicals

Inorganic salts were obtained from Aldrich (Milwaukee, WI), NTA resin was purchased from QIAGEN (Valencia, CA), and ICDH was obtained Sigama-Aldrich.

4.2.2. Expression and Purification of Human IRP1

A 40 ml culture containing 50 mg/L ampicillin was incubated at 37 ºC with continuous shaking for 10 h before use as an inoculum for 4 L LB with ampicillin at the

same concentration for 16 h. Cells were harvested by centrifugation at 4 ºC. Cell pellets

were resuspended in binding buffer (20 mM Tris, 500 mM NaCl, pH 7.9), lysed by

sonication, and cleared by centrifugation at 15,000 rpm at 4 ºC for 30 min. The

supernatant was loaded onto a Ni-NTA column that had been pre-equilibrated with the

binding buffer. The column was washed with 5 column volumes of binding buffer

containing 5 mM imidazole, and IRP1 was subsequently eluted by 5 column volumes of

binding buffer containing 100 mM imidazole. The protein was found to be brownish in

color and purity was verified by SDS-PAGE.

4.2.3. Aconitase Assay

The aconitase activity of holo IRP1 was measured spectrophotometrically by

following the rate of NADPH formation at 340 nm using the coupled assay outlined by

Gardner (21). An 8 μL aliquot of the aconitase activation reaction was added to 400 μL of assay buffer (50 mM Tris–HCl pH 7.4, 0.6 mM MnCl2, 5 mM sodium citrate, 0.2 mM

NADP+, 0.1% v/v Triton X-100 and 3.0 units/ml isocitrate dehydrogenase). The

absorbance was measured every second for a period of 2 min and the slope of the

changing absorbance was regarded to be directly proportional to the active aconitase in

the sample within the range of concentrations used.

4.2.4. Reconstitution of Partially Active IRP1 into Holo-IRP1 with Iron and Sulfur

IRP was incubated with 150 mM potassium acetate, 1.5 mM MgCl2, and 50 mM

DTT under vacuum for 30 min. Subsequently, FeCl3 and Na2S, degassed and Ar-purged,

were added to the reaction mixture, which was then allowed to react for an additional 1.5

h under Ar before centrifuging at 4,000 rpm for 10 min, and concentrating by Amicon

ultrafiltration. The resulting solution was passed through a G-25 column and eluted with

degassed buffer (50 mM Tris, 100 mM NaCl, pH 7.4) to remove unwanted small

molecules. The flow-through, which showed a brownish color, was collected and stored

at –80 °C (22).

4.2.5. Converting [4Fe-4S]-IRP1 to [3Fe-4S]-IRP1

The [3Fe-4S] form of IRP was generated by following the method of Hentze and

coworkers (22). A 20 μM solution of reactivated [4Fe-4S] IRP was treated with 30%

H2O2. The aconitase activity was checked periodically until there was no further change in absorbance at 340 nm. The reaction mixture was concentrated and passed through a G-

25 column. Subsequently, the iron content of the sample was measured by taking the

-1 -1 absorbance of the iron-bipyridine complex at 520 nm (ε520= 8400 cm /M ) after incubating the protein with excess bipyridine and dithionite. The molar ratio of iron to

IRP was found to be 3 to 1.

4.2.6. Converting [4Fe-4S]-IRP1 to Apo Form

[4Fe-4S]-IRP1 was incubated with 2% mercaptoethanol. The solution was passed onto G-25 column to remove low molecular weight molecules.

4.2.7. Measuring the Iron Content

The iron content of the sample was measured by taking the absorbance of the

-1 -1 iron-bipyridine complex at 520 nm (ε520 = 8400 cm ·M ) after incubating the protein

with excess bipyridine and sodium dithionite.

4.2.8. Binding Study of IRP1 with Frataxin via ITC

ITC measurements of apo- and [3Fe-4S]-IRP binding to frataxin, both in the

presence and absence of citrate, were carried out at 25 C (or the desired temperature) on a

MicroCal OMEGA ultrasensitive titration calorimeter. To a solution of 20 μM of each

IRP protein, with or without citrate, was added 10 μL of titrant containing frataxin over a period of 24 s, with an adequate interval (4 or 5 min) between injections to allow for equilibration. The proteins were dialyzed against 50 mM Hepes, and 100 mM NaCl at pH

7.5. Background control experiments included an identical titrant solution, but containing only the buffer solution or the buffer with citrate in the sample cell, which was subtracted from each experimental titration to account for the heat of dilution. Data were collected automatically and subsequently analyzed with a one-site binding model by the Windows- based Origin software package supplied by MicroCal. The Origin software uses a nonlinear least-squares algorithm (minimization of c2) and the concentrations of the titrant and the sample to fit the heat flow per injection to an equilibrium binding equation, providing best fit values of the stoichiometry (n), change in enthalpy (ΔH), change in entropy (ΔS), and association constant (KA).

4.2.9. Activation of the Aconitase Activity for [3Fe-4S]-IRP1

A solution containing 5 mM DTT with or without frataxin or citrate was Ar-

purged and degassed for 30 min. To this solution was added the degassed [3Fe-4S]-IRP1 to 10 μM. Subsequently, ferrous ion was added to a final concentration of 10 or 30 μM.

Every a few minutes, a 12 μL reaction mixture was drawn with a syringe and added to an

+ 800 μL solution containing 0.6 mM MnCl2, 5 mM sodium citrate, 0.2 mM NADP , 0.1% v/v Triton X-100 and 3.0 units/ml isocitrate dehydrogenase to evaluate the aconitase activity. The slope obtained from each assay was plotted against time. Control experiments were conducted in a similar fashion, but lacked IRP. The change of slope was found to be negligible. The difference between different batches of purified [3Fe-

4S]-IRP1was considered. And the same batch of [3Fe-4S]-IRP1 was used in one set of experiments to minimize variation.

4.3. Results

4.3.1. Preparation of Apo-, [3Fe-4S]- and [4Fe-4S]-IRP1

Purification of IRP1 from E. coli overexpression followed the general procedure of Ni-NTA column. IRP1 did not stick to Ni-NTA column tightly; therefore, 5 mM imidazole was used in the wash step. The impurity was removed by subjecting the eluent to a size exclusion column. The eluent was yellowish brown in color, and may consist of apo-, [3Fe-4S]- and [4Fe-4S]- IRP1. Since apo- and [3Fe-4S]-IRP1 are prepared from

[4Fe-4S]-IRP1, the eluent was directly used to reconstitute the [4Fe-4S] cluster. After

reconstitution and removing the inorganic salts, the eluent from G-25 column became

darker in color. To determine if most of the IRP1 was in the [4Fe-4S] form, the eluent

was subjected to an aconitase assay and the result was compared with a standard obtained

by using holo aconitase purchased from Aldrich. The standard was prepared by

measuring out several samples with different weights, and their aconitase activities

measured. Then the activities of the samples were plotted against the concentration of

holo aconitase calculated by using the published molecular weight of aconitase. A

standard curve was obtained by making a straight line fit. H2O2 was added to [4Fe-4S]-

IRP1 to generate [3Fe-4S]-IRP1. To check if all of the holo IRP1 had been converted to

[3Fe-4S]-IRP1, the aconitase assay was conducted periodically to determine if the

activity was lost; the concentration of IRP1 and the iron content were also measured. If

the sample contains all [3Fe-4S]-IRP1, it should have no aconitase activity, and the ratio

between iron and protein should be equal to 3. Similarly, if the sample contains only apo

IRP1, it should have no aconitase activity and the iron content should be zero.

4.3.2. Binding between [3Fe-4S]-IRP1 and Truncated Frataxin via ITC

[3Fe-4S]-containing IRP1 can bind to truncated frataxin when 2 mM citrate is present, with a Kd of 3.4 ± 0.4 μM (Figure 4.2 b). When the same experiment was carried

out without citrate, there was only a slight change in the heat released by each titration

which made it difficult to determine the binding curve. With higher concentrations of

citrate, there is higher ΔH (Figure 4.2 a and c). The Kd values are 15.0 ± 2.7, 3.4 ± 0.4,

and 5.7 ± 0.1 μM for experiments conducted with 0.5 mM, 2 mM and 5 mM citrate, respectively. The ΔH is -2662 ± 538, -5008 ± 407, and -5838 ± 546 kJ/mol.

Figure 4.2. Study of binding between [3Fe-4S]-IRP1 and truncated frataxin with varying concentration of citrate in present via ITC: (a), 0.5 mM citrate; (b), 2 mM citrate; and (c),

5 mM citrate. The data were collected by injecting small aliquots of around 0.2 mM frataxin solution into a 20 µM [3Fe-4S]-IRP1 solution in 50 mM HEPES (pH 7.5), 100 mM NaCl, with citrate in both the sample cell and the syringe. Upper panels: raw data for sequential injections of frataxin into the protein solution; Lower panels: integrated heats as a function of the frataxin/IRP1 ratio.

Time (min) Time (min) -10 0 10 20 30 40 50 60 70 80 90 100 110 -10 0 10 20 30 40 50 60 70 80 90 0.05 0.0 0.00 -0.1 -0.05 -0.10 -0.2 -0.15

-0.20 cal/sec -0.3 μ cal/sec

μ -0.25 -0.4 -0.30 b -0.35 a -0.5 -0.40 0.0 0

-0.2

-0.4 -2

-0.6

-0.8 -4 -1.0 kcal/mole of injectant -1.2 kcal/mole of injectant of kcal/mole -6 -1.4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Molar Ratio Molar Ratio

Time (min) -10 0 10 20 30 40 50 60 70 80 90 100110120130 0.02

0.00

-0.02

-0.04

-0.06 cal/sec μ -0.08

-0.10 c -2500 -0.12 -3000

-3500 0 -4000

-4500 H Cal/mol Δ -5000

-2 -5500

kcal/mole of injectant of kcal/mole -6000

012345 0246810 Concentration of citrate (mM) Molar Ratio

4.3.3. Binding of [3Fe-4S]-IRP1 to Full-length Frataxin via ITC

To determine whether the anionic patch of frataxin is where IRP1 binds, an ITC experiment was carried out between [3Fe-4S]-IRP1 and full-length frataxin with 2 mM citrate in present. The rationale is that the N- terminus of full-length frataxin interacts with the anionic patch through electrostatic interaction, which results in less accessibility of the anionic patch. And full-length frataxin has distinct properties from the truncated form, such as much lower solubility in the buffer (18). We found that [3Fe-4S]-IRP1 binds to full-length frataxin equally well as to the truncated frataxin (Figure 4.2 b and

Figure 4.3 b). The Kd is 7.2 ± 0.1 μM. This might imply that the anionic patch of frataxin is not the IRP1 binding site.

Time (min) Time (min) -10 0 10 20 30 40 50 60 70 80 90 -10 0 10 20 30 40 50 60 70 80 90 100 0.05 0.0 0.00 -0.05 -0.2 -0.10 -0.4 -0.15 -0.20 cal/sec cal/sec -0.6 μ μ -0.25 -0.8 a -0.30 b -1.0 -0.35 0 0

-2 -2 -4

-6 -4 kcal/mole ofinjectant

-8 kcal/mole of injectant -6 0123456789 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Molar Ratio Molar Ratio

Figure 4.3. Study of binding between [3Fe-4S]-IRP1 (a) or apo IRP1 (b) and full-length

frataxin with 2 mM citrate in present via ITC, the Kd of full-length frataxin binding to

[3Fe-4S]-IRP1 was 7.2 ± 0.1 μM; and the Kd of full-length frataxin binding to apo IRP1

was 0.39 ± 0.03 μM. The data were collected by injecting small aliquots of approximately

0.2 mM full-length frataxin solution into a 20 µM [3Fe-4S]- or apo-IRP1 solution in 50 mM Hepes (pH 7.5), 100 mM NaCl.

4.3.4. Binding between Apo IRP1 and Frataxin via ITC

It has been shown that the structure of apo-IRP1 differs from the [4Fe-4S]-IRP1 by moving domains 3 and 4 by angles of about 52 and 32 degrees, respectively (4, 23).

This movement causes part of the covered surface to be exposed within domain 3 and 4 while the rest of surfaces remain largely unchanged (Figure 4.4). Therefore, the ITC experiments on the interaction between apo IRP1 and frataxin could give some information on where the binding interface could be. Apo IRP1 can bind to truncated frataxin with a Kd of 3.1 ± 0.3 μM (Figure 4.5), while for the binding to full-length frataxin with 2 mM citrate present, Kd is 0.39 ± 0.03 µM (Figure 4.3 b).

Figure 4.4. The conformational change from closed position ([3Fe-4S]- or [4Fe-4S]-IRP1, left) to open position (apo IRP1, right). The circled region is positively charged and might be where citrate binds.

Time (min) Time (min) -10 0 10 20 30 40 50 60 70 80 90 100110120130 -10 0 10 20 30 40 50 60 70 80 0.05 0.02

0.00 0.00 -0.02 -0.04 -0.05

-0.06 -0.10 -0.08 -0.10 -0.15 cal/sec cal/sec μ μ -0.12 -0.20 -0.14 -0.16 a -0.25 b -0.18

0 0

-2 -2 kcal/mole of injectant kcal/mole kcal/mole of injectant of kcal/mole

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Molar Ratio Molar Ratio

Figure 4.5. Study of the binding between apo IRP1 and truncated frataxin with 0 mM (a) or 2 mM (b) citrate via ITC. The data were collected by injecting small aliquots of around

0.2 mM frataxin solution into a 20 µM apo IRP1 solution in 50 mM HEPES (pH 7.5),

100 mM NaCl. Both the sample and titrant were supplemented with 2 mM citrate if citrate was used. Kd was determined to be 8.4 ± 0.6 μM or 3.1 ± 0.3 μM for experiments done in 2 or 0 mM citrate, respectively.

Dissociation constant of Concentration of Apo IRP1 [3Fe-4S] IRP1 corresponding interaction citrate (mM) Frataxin 5 - 5.7 ± 0.1 μM 2 8.4 ± 0.6 μM 3.4 ± 0.4 μM 0.5 - 15.0 ± 2.7 μM 0 3.1 ± 0.3 μM - Full-length frataxin 2 0.39 ± 0.03 μM 7.2 ± 0.1 μM 0 - -

Table 4.1. A summary of the binding affinities between IRP1 and frataxin. The dash lines indicate the experiments that have not been done.

100

4.3.5. Reactivation of [3Fe-4S]-IRP1

A set of experiments was performed to explore the rate of frataxin in mediating

repair of the [3Fe-4S]-IRP1 into the active form. With 10 μM ferrous ion in the reaction

mixture, the reactivation rate was 0.014 ± 0.003 min-1; when the concentration of ferrous

ion was increased to 30 μM, the reactivation rate increased to 0.064 ± 0.020 min-1(Figure

4.6, A). When keeping the concentration of ferrous ion constant, and varying the concentration of citrate, the reactivation rates were found to be 0.064 ± 0.020 min-1 and

0.014 ± 0.002 min-1 for citrate concentrations of 0 and 0.5 mM, respectively.

Reactivation was extremely slow when the concentration of citrate was 2 mM. If the

concentrations of ferrous ion and citrate were kept at 30 μM and 0.5 mM, respectively,

increasing the concentration of frataxin increased the rates and yield of the reactivation

accordingly. Without frataxin, the reactivation rate was determined to be 0.014 ±0.002 min-1; with 10 μM frataxin in the reaction mixture, the reactivation rate became 0.016 ±

0.002 min-1; and when 60 μM frataxin was present, the rate was increased to 0.027 ±

0.004 min-1. Comparing the rate of reactivation when 30 μM ferrous ion alone was

present in the reaction mixture with that when 30 μM ferrous ion, 0.5 mM citrate and 60

μM frataxin were present, the latter was two folds slower but still achieved near complete

reactivation. The data is summarized in Table 4.2.

101

Conditions of reactivation Rate ( min-1) 10 µM Fe2+ 0.014 ± 0.003 30 µM Fe2+ 0.064 ± 0.020 30 µM Fe2+ and 0.5 mM citrate 0.014 ±0.002 30 µM Fe2+ and 2 mM citrate < 0.001 30 µM Fe2+ , 0.5 mM citrate, and 10 µM frataxin 0.016 ± 0.002 30 µM Fe2+ , 0.5 mM citrate, and 60 µM frataxin 0.027 ±0.004

Table 4.2. Rates of reactivation of [3Fe-4S]-IRP1 under different conditions.

102

Figure 4.6. Reactivation of [3Fe-4S]-IRP1. The reaction mixture contains 10 μM [3Fe-

4S]-IRP1, 5 mM DTT, and (A) Δ, 10 μM ferrous ion, or ■,30 μM ferrous ion; (B) 30 μM ferrous ion, 0.5 mM citrate, and increasing concentration of frataxin: ■, 0 μM, ○, 10 μM, or ▲, 60 μM frataxin; (C) 30 μM ferrous ion and increasing concentration of citrate: ■, no citrate, Δ, 0.5 mM or ●, 2 mM citrate; (D) ■, 30 μM ferrous ion or ○, 30 μM ferrous ion, 0.5 mM citrate and 60 μM frataxin. All solutions were previously Ar purged and degassed repetitively. Ferrous ion was added last to initiate the reaction. UV absorbance at 456 nm was monitored over time.

103

1.0 2+ 30 μM Fe 1.0 0 μM frataxin 2+ 10 μM frataxin 0.8 10 μM Fe 60 μM frataxin 0.8 0.6 0.6

0.4 0.4

0.2 0.2 Fraction of IRPreactivated 0.0 Fraction of IRP reactivated A 0.0 B -10 0 10 20 30 40 50 60 70 80 0 20406080100 Time (min) Time (min)

1.2 2+ 30 μM Fe 1.0 30 μM Fe2+ and 2 mM citrate 1.0 2+ 30 μM Fe and 0.5 mM citrate 0.8 0.8 0.6 0.6

0.4 0.4 2+ 0.2 30 μM Fe 0.2 2+ 30 μM Fe , 0.5 mM citrate Fraction of IRP reactivated IRP of Fraction Fraction ofFraction IRP reactivated and 60 μM frataxin 0.0 C 0.0 D -10 0 10 20 30 40 50 60 70 80 90 -10 0 10 20 30 40 50 60 70 80 90 100 110 Time (min) Time (min)

104

4.4. Discussion

4.4.1. The Interaction between [3Fe-4S]-IRP1 and Truncated Frataxin

We demonstrated via ITC that [3Fe-4S]-IRP1 can bind to human truncated

frataxin in the presence of citrate (Table 1). However, it has previously been shown that

[3Fe-4S]-IRP1 does not bind to citrate (24). In addition, we found that citrate does not bind to frataxin by ITC (data not shown here). These results most likely reflect the relatively high concentration of cellular citrate (25) with either a resulting requirement for only modest binding affinity by citrate, which would be difficult to detect by most routine methods, or a binding environment that is only formed by the co-complex of IRP1 and frataxin. The [3Fe-4S]-IRP1 frataxin binding experiments were performed using 5 mM and 0.5 mM citrate, and the results were consistent with a role for citrate in promoting

binding between these two proteins. The enthalpy change was found to decrease in

magnitude with decreasing concentration of citrate (Figure 4.2).

4.4.2. The Anionic Patch on Frataxin Is Not Involved in Binding to [3Fe-4S]-IRP1

To bind to partner proteins, frataxin often uses iron to bridge between the acidic

residues of partner proteins and itself. This was found to be true for both IscU proteins

and ferrochelatase (19, 26). However, our results indicated that [3Fe-4S]-IRP1 binds to

truncated frataxin in an iron-independent manner. Therefore, the way frataxin binds to

IRP1 species could be different from its binding to other partner proteins, and the anionic

patch may not be involved. Consistent with this result, additional experiments have also

demonstrated full-length frataxin to bind to apo cytosolic aconitase (c-aconitase) in an 105

iron-independent manner. Prior literature provides no indication that the [3Fe-4S]-bound

form of aconitase can bind citrate in the absence of other binding partners; therefore any

direct influence must be the result of the association of IRP1 and frataxin. The demonstration of binding by full-length frataxin (Figure 4.3) suggests that the most likely binding site is the hydrophobic domain rather than the charged anionic domain noted earlier. Aconitase shows a positive patch around the cluster binding pocket, which is the

most likely location for frataxin to bind if it is to promote iron delivery. A possible role

for citrate in promoting binding of frataxin through its hydrophobic patch is charge

neutralization of the positively-charged domain. Indeed, our studies find a significantly

weaker response in the absence of citrate.

4.4.3. Interaction between Apo-IRP1 and Frataxin

In the presence of citrate, apo IRP1 binds to truncated frataxin with a slightly

higher KD, compared with the interaction between [3Fe-4S] IRP1 (closed conformation,

essentially the same as the [4Fe-4S] containing species (27)) and truncated frataxin, this

implies that the conformational change of domain 3 and domain 4 (Figure 4.4) does not affect the interaction significantly, and the interaction may not involve domain 3 and 4 at

the same time. In the presence of citrate, the positively charged channel where citrate

enters the active site may be occupied by citrate; therefore, the surface should be neutral

or acidic, since the concentration of citrate is 100 fold higher than IRP1. The anionic

patch of frataxin needs to face towards the active site in order to deliver iron, so repulsion

between the anionic patch and the citrate-occupied surface may exist, if this is true, this

106

could explain why full-length frataxin binds tighter to apo IRP1 than to [3Fe-4S] IRP1, since the anionic patch of full-length frataxin is covered. Without citrate, apo IRP1 binds to truncated frataxin with a similar affinity as the binding in the presence of citrate, this could imply that citrate does not directly involved in binding between truncated frataxin and apo IRP1, unlike the interaction when [3Fe-4S] IRP1 is involved.

4.4.4. Frataxin Enhances the Rate of Reactivation of [3Fe-4S]-IRP1

Under anaerobic conditions, the [3Fe-4S] bound IRP1 can be reactivated with ferrous ion alone. Higher concentration of ferrous ion gives higher reactivation rate.

When the concentration of ferrous ion was kept constant, the rate of reactivation decreased as the increase of the citrate concentration. This presumably reflects complexation of iron ion by citrate, thereby limiting the availability of ferrous ion in solution. If the concentrations of ferrous ion and citrate were kept at 30 μM and 0.5 mM, respectively, when increasing the concentration of frataxin, the rates and yield of reactivation increased accordingly. This is consistent with a role for frataxin in mediating the repair of the [3Fe-4S]-IRP1.

Comparing the rate of reactivation when only 30 µM ferrous ion was present with that when 30 µM ferrous ions, 0.5 mM citrate and 60 µM frataxin were present, the latter was observed to be two-fold slower, but still achieved near complete reactivation (Figure

4.6, D). This most likely reflects equilibration among ferrous ion, citrate and frataxin, where reactivation is controlled by the delivery of ferrous ion by frataxin. The binding constant of citrate to ferrous ion is ~ 104 M-1 (28), while the binding constant for frataxin

107

to ferrous ion is ~105 M-1. When citrate, ferrous ion and frataxin were added to the reaction mixture, equilibrium among them will establish. A portion of ferrous ion will bind to frataxin. Since frataxin can bind to [3Fe-4S]-IRP1, frataxin might bring ferrous ion to the vicinity of the cluster and thus facilitate the formation of the [4Fe-4S] cluster.

The higher the concentration of frataxin, the larger the portion of ferrous ion binds to

frataxin and the faster the reactivation.

It has been shown that the reactivation rate of mitochondrial aconitase was greatly

enhanced by frataxin; 30 µM ferrous ions alone could not reactivate purified [3Fe-4S]

bound bovine mitochondrial aconitase significantly in a buffer containing 10 mM Hepes

and 1 mM DTT, pH 7.3, at 30°C; and frataxin helped to fully reactivate it within 30 min

(10). Herein we show that 30 µM ferrous ions alone can reactivate human [3Fe-4S] IRP1

with a rate of 0.064 ± 0.020 min-1, consistent with a report in which 30 µM ferrous ions

alone was found to reactivate the [3Fe-4S] form of beef heart mitochondrial aconitase

with a good rate (29). Although [3Fe-4S]-IRP1 does not require frataxin to be reactivated

in vitro, the physiological concentration of iron is around 20 nM in yeast cells, under

such condition, the rate of reactivation will be negligible. The experiment with 30 µM

ferrous ion and 2 mM citrate showed a negligible rate, in which the estimated ferrous

concentration in solution was 1 µM. In addition, there is little free iron in cell, and iron is

always bound to chelators or chaperone proteins for proper delivery and storage. Citrate

which is 0.1 to 0.4 mM in cytosol of human cell is one of the major iron chelators (25).

Therefore, the effect of frataxin is best shown when citrate is present. And the rate of

reactivation indeed showed a positive correspondence with the concentration of frataxin.

108

Extra-mitochondrial frataxin (56-210) has been discovered in some human cells

(16). The overexpression of this frataxin in the cytosol promotes survival of FRDA cells.

However, it has been indicated that the functional frataxin in mitochondria corresponds to the construct starting from the 81st residue of frataxin (30). The reason underlying this difference is still not clear, but it could be associated with how frataxin is translocated to and processed in mitochondria. It is further suggested that selective cleavage within the mitochondrion reflects the distinct concentrations of available iron within the cytosol and mitochondrion. While there is considerable disparity and lack of consensus on the concentrations of available (chelatable) cellular iron and variations in discrete organelles within the cell, it is likely that the concentration of available iron in the mitochondrion will exceed that in the cytosol and the nucleus. This is consistent with the focus of iron cofactor dependent chemistry (especially redox chemistry) within the mitochondrion. We have earlier demonstrated full-length frataxin to possess a unique high affinity iron binding site that is approximately an order of magnitude greater than those found for the truncated version (12, 31). We hypothesize that higher concentrations of mitochondrial iron allow this site to be populated, with subsequent cleavage, resulting in the truncated form that appears to be the active form in the mitochondrion. Such a model is consistent with prior cellular studies of the full-length and truncated forms (20), and leads to the

prediction that the truncated form is the dominant active form in the mitochondrion,

while full-length frataxin serves a more significant role in the cytosol.

In conclusion, it appears that cytosolic frataxin can promote the repair of [3Fe-

4S]-containing IRP1, and that a cytosolic unprocessed form of frataxin targets IRP1 in a

109

citrate-dependent manner. These results support a cellular model where iron-promoted self-cleavage of MPP-processed frataxin allows for distinct targeting of mitochondrial versus cytosolic partners. Overall, the outcomes from these experiments advance our understanding of the regulation of intracellular iron chemistry, and provide further support for frataxin truncation as a novel metal-regulated protein self-cleavage reaction of importance for modulating cellular chemistry.

4.5. References for Chapter 4

1. Gruer, M. J., Artymiuk, P. J., and Guest, J. R. (1997) The aconitase family: three structural variations on a common theme, Trends Biochem Sci 22, 3-6.

2. Rouault, T. A. (2006) The role of iron regulatory proteins in mammalian iron homeostasis and disease, Nat Chem Biol 2, 406-414.

3. Beinert, H., Kennedy, M. C., and Stout, C. D. (1996) Aconitase as Ironminus signSulfur Protein, Enzyme, and Iron-Regulatory Protein, Chem Rev 96, 2335- 2374.

4. Dupuy, J., Volbeda, A., Carpentier, P., Darnault, C., Moulis, J. M., and Fontecilla-Camps, J. C. (2006) Crystal structure of human iron regulatory protein 1 as cytosolic aconitase, Structure 14, 129-139.

5. Casey, J. L., Hentze, M. W., Koeller, D. M., Caughman, S. W., Rouault, T. A., Klausner, R. D., and Harford, J. B. (1988) Iron-responsive elements: regulatory RNA sequences that control mRNA levels and translation, Science 240, 924-928.

6. Rouault, T. A., Hentze, M. W., Caughman, S. W., Harford, J. B., and Klausner, R. D. (1988) Binding of a cytosolic protein to the iron-responsive element of human ferritin messenger RNA, Science 241, 1207-1210.

7. Koeller, D. M., Casey, J. L., Hentze, M. W., Gerhardt, E. M., Chan, L. N., Klausner, R. D., and Harford, J. B. (1989) A cytosolic protein binds to structural elements within the iron regulatory region of the transferrin receptor mRNA, Proc Natl Acad Sci U S A 86, 3574-3578.

8. Dandekar, T., Stripecke, R., Gray, N. K., Goossen, B., Constable, A., Johansson, H. E., and Hentze, M. W. (1991) Identification of a novel iron-responsive element 110

in murine and human erythroid delta-aminolevulinic acid synthase mRNA, Embo J 10, 1903-1909.

9. Gray, N. K., Pantopoulos, K., Dandekar, T., Ackrell, B. A., and Hentze, M. W. (1996) Translational regulation of mammalian and Drosophila citric acid cycle enzymes via iron-responsive elements, Proc Natl Acad Sci U S A 93, 4925-4930.

10. Bulteau, A. L., O'Neill, H. A., Kennedy, M. C., Ikeda-Saito, M., Isaya, G., and Szweda, L. I. (2004) Frataxin acts as an iron chaperone protein to modulate mitochondrial aconitase activity, Science 305, 242-245.

11. Stehling, O., Elsasser, H. P., Bruckel, B., Muhlenhoff, U., and Lill, R. (2004) Iron-sulfur protein maturation in human cells: evidence for a function of frataxin, Hum Mol Genet 13, 3007-3015.

12. Rotig, A., de Lonlay, P., Chretien, D., Foury, F., Koenig, M., Sidi, D., Munnich, A., and Rustin, P. (1997) Aconitase and mitochondrial iron-sulphur protein deficiency in Friedreich ataxia, Nat Genet 17, 215-217.

13. Babcock, M., de Silva, D., Oaks, R., Davis-Kaplan, S., Jiralerspong, S., Montermini, L., Pandolfo, M., and Kaplan, J. (1997) Regulation of mitochondrial iron accumulation by Yfh1p, a putative homolog of frataxin, Science 276, 1709- 1712.

14. Foury, F., and Cazzalini, O. (1997) Deletion of the yeast homologue of the human gene associated with Friedreich's ataxia elicits iron accumulation in mitochondria, FEBS Lett 411, 373-377.

15. Cavadini, P., Adamec, J., Taroni, F., Gakh, O., and Isaya, G. (2000) Two-step processing of human frataxin by mitochondrial processing peptidase. Precursor and intermediate forms are cleaved at different rates, J Biol Chem 275, 41469- 41475.

16. Condo, I., Ventura, N., Malisan, F., Tomassini, B., and Testi, R. (2006) A pool of extramitochondrial frataxin that promotes cell survival, J Biol Chem 281, 16750- 16756.

17. Acquaviva, F., De Biase, I., Nezi, L., Ruggiero, G., Tatangelo, F., Pisano, C., Monticelli, A., Garbi, C., Acquaviva, A. M., and Cocozza, S. (2005) Extra- mitochondrial localisation of frataxin and its association with IscU1 during enterocyte-like differentiation of the human colon adenocarcinoma cell line Caco- 2, J Cell Sci 118, 3917-3924.

111

18. Yoon, T., Dizin, E., and Cowan, J. A. (2007) N-terminal iron-mediated self- cleavage of human frataxin: regulation of iron binding and complex formation with target proteins, J Biol Inorg Chem 12, 535-542.

19. Yoon, T., and Cowan, J. A. (2003) Iron-sulfur cluster biosynthesis. Characterization of frataxin as an iron donor for assembly of [2Fe-2S] clusters in ISU-type proteins, J Am Chem Soc 125, 6078-6084.

20. Condo, I., Ventura, N., Malisan, F., Rufini, A., Tomassini, B., and Testi, R. (2007) In vivo maturation of human frataxin, Hum Mol Genet 16, 1534-1540.

21. Gardner, P. R. (2002) Aconitase: sensitive target and measure of superoxide, Methods Enzymol 349, 9-23.

22. Brazzolotto, X., Gaillard, J., Pantopoulos, K., Hentze, M. W., and Moulis, J. M. (1999) Human cytoplasmic aconitase (Iron regulatory protein 1) is converted into its [3Fe-4S] form by hydrogen peroxide in vitro but is not activated for iron- responsive element binding, J Biol Chem 274, 21625-21630.

23. Walden, W. E., Selezneva, A. I., Dupuy, J., Volbeda, A., Fontecilla-Camps, J. C., Theil, E. C., and Volz, K. (2006) Structure of dual function iron regulatory protein 1 complexed with ferritin IRE-RNA, Science 314, 1903-1908.

24. Zheng, W., Ren, S., and Graziano, J. H. (1998) Manganese inhibits mitochondrial aconitase: a mechanism of manganese neurotoxicity, Brain Res 799, 334-342.

25. Naz, R. K. (1997) Prostate : basic and clinical aspects, CRC Press, Boca Raton.

26. Yoon, T., and Cowan, J. A. (2004) Frataxin-mediated iron delivery to ferrochelatase in the final step of heme biosynthesis, J Biol Chem 279, 25943- 25946.

27. Robbins, A. H., and Stout, C. D. (1989) Structure of activated aconitase: formation of the [4Fe-4S] cluster in the crystal, Proc Natl Acad Sci U S A 86, 3639-3643.

28. Park, S., Gakh, O., O'Neill, H. A., Mangravita, A., Nichol, H., Ferreira, G. C., and Isaya, G. (2003) Yeast frataxin sequentially chaperones and stores iron by coupling protein assembly with iron oxidation, J Biol Chem 278, 31340-31351.

29. Kennedy, M. C., Emptage, M. H., Dreyer, J. L., and Beinert, H. (1983) The role of iron in the activation-inactivation of aconitase, J Biol Chem 258, 11098-11105.

112

30. Schmucker, S., Argentini, M., Carelle-Calmels, N., Martelli, A., and Puccio, H. (2008) The in vivo mitochondrial two-step maturation of human frataxin, Hum Mol Genet 17, 3521-3531.

31. Bencze, K. Z., Yoon, T., Millan-Pacheco, C., Bradley, P. B., Pastor, N., Cowan, J. A., and Stemmler, T. L. (2007) Human frataxin: iron and ferrochelatase binding surface, Chem Commun (Camb), 1798-1800.

113

CHAPTER 5

CLONING AND EXPRESSION OF RECOMBINANT HUMAN EPENDYMIN AND INITIAL CHARACTERIZATION

5.1. Introduction

Ependymins are secretary glycoproteins that were initially found predominantly in the cerebrospinal fluid of some teleost fish (1, 2). Ependymins were also found later in mammalian brains (3). The level of ependymin in the central neuron system underwent

changes following various types of behavior training with goldfish and mice (3). Thus,

ependymins are thought to be implicated in fundamental processes involved in the

formation of long-term memory and neuronal regeneration (4).

Ependymins have an N-terminal leader sequence (Figure 5.1). The sequence of

ependymin shows little homology to any known polypeptides (3). The quaternary

structure of goldfish ependymin is formed by forming disulfide bonds between two

constructs of ependymin (M.W. 37K and 31K) (2, 4). Ependymin remains soluble at

milimolar calcium concentration; however, it starts to polymerize into a fibrous insoluble

polymer (FIP) when the calcium concentration is decreased (5). Once formed, the

114

polymer could not be dissolved by restoring the calcium concentration, or by using other

reagents, such as 2% sodium dodecylsulfate (SDS) in 5 M urea, 100% acetic acid,

chloroform/methanol (2:1) and even 100% trifluoroacetic acid. It could be partially

soluble in 80% formic acid (3). There might be an ATP-dependent phosphorylation step

which converts ependymin into a polymerizable form (3). Some evidence also suggests

that aggregation of ependymin might occur in vivo during learning processes (6).

However, some research groups were able to isolate ependymin that did not form a polymer under low calcium concentration.

115

MPGRAPLRTV PGALGAWLLG GLWAWTLCGL CSLGAVGAPR PCQAPQQWEG RQVMYQQSSG

RNSRALLSYD GLNQRVRVLD ERKALIPCKR LFEYILLYKD GVMFQIDQAT KQCSKMTLTQ

PWDPLDIPQN STFEDQYSIG GPQEQITVQE WSDRKSARSY ETWIGIYTVK DCYPVQETFT

INYSVILSTR FFDIQLGIKD PSVFTPPSTC QMAQLEKMSE DCSW

Figure 5.1. The sequence of full-length human ependymin. The first residue after the signal peptide is highlighted in yellow, as predicted by SignalP V1.1. The two Asp residues predicted to be N-glycosylated are highlighted in blue, as predicted by

NetNGlyc 1.0. There are multiples phosphorylation sites predicted by the NetPhos 2.0

Server.

116

According to Hebbian theory, memories are represented by vastly interconnected

networks of synapses in the brain (7). Synaptic plasticity, the ability to connect between

two neurons to change in strength, is one of the important neurochemical foundations of

learning and memory (8, 9). The biochemical mechanism of synaptic plasticity is that

the second messenger neurotransmitters which regulate the gene transcription of certain

proteins are released. This subsequently results in changes in the levels of key proteins at

synapses (10). This mechanism can be triggered by protein phosphorylation which takes

longer and lasts longer, and provides the foundation for long-term memory formation

(10).

Based on these mechanisms, a molecular hypothesis relating the aggregation

properties of ependymins to neuroplasticity and learning was proposed. Figure 5.2

illustrates the ependymin polymerization hypothesis (3). In scheme (A), one strong

synapse and one weak synapse converge onto a dendrite. During associative learning, the

electric impulses from the synapses results in opening and closing of the calcium

channels, causing calcium ion depletion in the local synaptic cleft and an increase of Ca2+ level within the dendrite (B). Ependymin in the synaptic cleft polymerizes to form a FIP matrix. The increased Ca2+ level in the dendrites activates proteases that break down

intracellular cytoskeletons. A spine forms where ependymin can enter. In scheme (C),

the polymers newly formed by ependymin help to form a new cytoskeleton and are

eventually converted to strong synapses that can independently fire the cell. In other

words, a new connection is formed (3).

117

Figure 5.2. The polymerization hypothesis of ependymin (adopted from reference (3))

118

Human ependymin has some similar properties to fish ependymin. Human ependymin was predicted to contain a signal peptide (residue 1 to 37) by using the

SignalP V1.1 server (Figure 5.1). This is consistent with it being a secretary protein. It has two possible glycosylation sites and multiple phosphorylation sites (Figure 5.1).

However, it shares little sequence homology with those from lower organisms. In this work, human ependymin was cloned to overexpress in E coli, yeast and insect cells. A variety of methods were used to investigate its biochemical properties.

5.2. Materials and Methods

5.2.1. General Chemicals

Inorganic salts were obtained from Sigma-Aldrich (Milwaukee, WI, USA).

Nitrilotriacetic acid (NTA) resin was purchased from QIAGEN (Valencia, CA, USA). All restriction enzymes and buffers were from Invitrogen (Carlsbad, CA). The Pichia expression kit was from Invitrogen (Carlsbad, CA), Pfu polymerase was obtained from

Stratagene (La Jolla, CA), and Expression vectors pET-28 and pET-32 were from

Novagen.

5.2.2. Cloning of Mature Ependymin into pET-32 for Expression in E. coli with a

Thioredoxin Fusion Protein in the N-terminal End

The mature form of human ependymin (residue 38 to 224) was PCR amplified using forward primer, 5’-GGGCCCGGGAATTCGCCCCGCGCCCGTGC-3’ and

119

reverse primer, 5’-CCGGCCCTCGAGCCAGGAGCAGTCTTCGCTCA-3’. Underlined

are the EcoRI and XhoI restriction sites. The PCR product was subsequently inserted into a pET-32 plasmid between restriction site EcoRI and XhoI. The ligation product was transformed into TOP10F’, which was allowed to grow on LB plates containing 100

µg/ml ampicillin overnight at 37 ºC. Potential colonies were screened and further confirmed by nucleotide sequencing.

5.2.3. Expression of pET-32-epen

The plasmid with the correct sequence was transferred into BL21 (DE3) for expression. Four colonies were picked from a fresh grown plate and transferred into 10 ml LB medium containing 50 mg/L ampicillin. They were allowed to grow till the OD600 reached 0.6-0.8. A 1 ml aliquot was taken and spun down. The cells were put at -80 ºC until used. Subsequently IPTG was added to the remaining medium to 1 mM concentration and the cells grew for another 4 h. After that, another 1 ml medium was taken out and spun down. To these samples obtained before and after adding IPTG, were added 30 µL SDS loading buffer. The mixture was boiled for 5 minutes and subsequently clarified by centrifugation. 5 ml of the supernatant was taken out and subject to SDS-

PAGE. Colonies that showed expression were used to do large scale expression. These colonies were grown and induced with IPTG. The cells were broken by sonication. The lysate was centrifuged for 1 minute at 15,000 rpm. The supernatant and pellet was added

SDS loading buffer and subjected to SDS-PAGE to check if the produced protein is soluble.

120

5.2.4. Purification and On-Column Refolding of Thioredoxin-Ependymin Fusion

Protein

Cells grown for 4 h after addition of IPTG were harvested by centrifuging for 10

min at 4 ºC. The harvested cell mass was resuspended with binding buffer (20 mM Tris,

500 mM NaCl, pH 7.9), which was then broken by sonication. The lysate was cleared by

centrifuging for 30 minutes at 15,000 rpm. The pellet was resuspended with binding

buffer with addition of 8 M urea, 1 mM PMSF and 10 mM beta-mercaptoethanol, with

stirring for 3 h at 4 ºC. The mixture was cleared by centrifuging for 45 min and the

supernatant incubated with Ni-NTA resin that had been pre-equilibrated with the same

buffer, with mild shaking at 4 ºC for 4 h to ensure proper binding. The resin bound with

the protein was then washed with 5 column volume of the same buffer containing 25 mM

imidazole, and the protein refolded by stepwise lowering the concentration of urea and

beta-mercaptoethanol, (for example, using 6, 4, 2, 1, 0.5 and 0 M urea and 8, 6, 4, 2, and 0

mM beta-mercaptoethanol in the binding buffer). Finally, 200 mM imidazole was added to

the binding buffer to elute the protein from the column. The protein solution was dialyzed

against buffer containing 20 mM Tris, 100 mM NaCl, pH 7.9 and put at 4 ºC till used.

Pure protein was sent to determine the mass by Mass Spectroscopy. Aliquots from each

purification steps were collected and subjected to SDS-PAGE.

5.2.5. Removing Thioredoxin from Thioredoxin-Ependymin Fusion Protein

In pET-32, the gene for thioredoxin is followed by a His-tag, an S-tag and an

enterokinase cleavage site before the multiple cloning sites. In an initial experiment, 10

121

µg of ependymin was mixed with 0.2 unit enterokinase, 1 µl 10X cleavage buffer and

H2O to make a 10 µL reaction mixture. Aliquots were taken to check the progress of the reaction.

5.2.6. Circular Dichroism of Thioredoxin-Ependymin Fusion Protein

Circular Dichroism spectra of the thioredoxin-ependymin fusion protein were measured on an AVIV model 202 circular dichroism spectrometer. The CD spectra were acquired with a 3 mm path length cuvette. The proteins were dialyzed against 2 mM

Hepes, pH 7.5, and then passed through a 0.22-μm filter to remove aggregated protein.

Absorbance were taken every 0.2 nm in triplicate and averaged. A control spectrum of

buffer alone was collected under the same conditions and was subtracted before the

secondary structure analysis. Secondary structure quantization was determined via the

CDPro data analysis package, which utilizes the self-consistent method of Sreerama and

Woody (11) and the ridge regression procedure of Provencher and Glöckner (12). To

estimate the effect of calcium on the conformation of ependymin, 2 mM calcium ion was

added to the ependymin solution and the spectra were taken. The spectra were compared

with and without calcium present, and dilution effects were considered.

5.2.7. Probing the Calcium Binding Sites by Terbium Luminescence

The terbium emission intensity was measured on a Perkin-Elmer LS50B

luminescence spectrometer, using a cuvette with a 0.5 cm path length at room

temperature. To the cuvette was added 5 µl aliquots of terbium chloride from a stock

122

solution made by dissolving terbium chloride to buffer containing 50 mM Hepes, 100

mM NaCl, pH 7.5. The cuvette contained a total of 800 µl solution with 5 µM ependymin

in the same buffer. Terbium was excited by energy transfer from the aromatic residues of

ependymin following excitation by light at a wavelength of 280 nm. The emission

spectrum from 450 to 530 nm was recorded and the maximum emission intensity

observed at 488 nm. Peak height was used as an indicator of peak area. The emission

intensity at 530 nm was taken as the baseline. The data were all corrected for dilution and control experiments with the same amount of terbium added to buffer were recorded over the same range. In all experiments, solutions were allowed to equilibrate after additions of terbium.

5.2.8. Cloning of Human Ependymin into pPIC9 for Expression in Pichia

Human ependymin was PCR amplified with a pET-21-epen vector, using forward primer, 5’-GGGCCCGGGAATTCGCCCCGCGCCCGTGC-3’, and reverse primer, 5’-

CCGGCCGCGGCCGCTCAGTGGTGGTGGTGG-3’. The GTG sequence in the reverse

primer is due to the intent to add a C-terminal His-tag that was already located in the

pET-21-ependymin vector. The product was subsequently inserted into pPIC9 between

EcoR I and Not I sites. Sequence was confirmed by double digestion with EcoR I and Not

I, and nucleotide sequencing by using 5’AOX1, 3’AOX1 and α-factor sequencing primers.

123

5.2.9. Transformation of pPIC9-ependymin into Pichia strains GS115 or KM71

Vector pPIC9-epen was linearized with Sal I in order to isolate His+ Mut+ transformants of GS115. Transfromation of GS115 with an intact AOX1 gene with Sal I favor the generation of Mut+. And to generate His+ MutS transformants of KM71, which

is deficient in AOX1 gene, the same vector was linearized with Sac I. The parent vector

pPIC9 was digested in the same way as a control. The products were purified with the

Qiagen PCR purification kit.

Pichia pastoris was grown in YPD medium (1% yeast extract, 2% peptone and

2% dextrose) at 30 ºC overnight for transformation. To a 500 ml of fresh YPD medium

was transferred 0.5 ml of the overnight culture. The next morning, the cell mass was

harvested at 4 ºC when the OD600 had reached 1.3. The pellet was resuspended with 500

ml ice-cold, sterile water and then centrifuged again. The pellet was washed three more

times with 250 ml ice-cold, sterile water, and 20 ml 1 M sorbitol. Finally, the pellet was

resuspended in 1 ml 1 M sorbitol. 10 μg linearized DNA in water was mixed with 80 μL cell solution. The mixture was transferred to a 0.2 cm electroporation cuvette and allowed to incubate on ice for 5 minutes. The electroporation was done following the parameters provided by the manufacturer (BioRad). Then, the mixture was resuspened with 1 ml 1 M sorbitol in the cuvette, 200 μL aliquots of which was spread on MD plates (1.34% YNB,

4×10-5% biotin and 2% dextrose). The plates were allowed to incubate at 30 ºC until colonies appeared.

124

5.2.10. Screening for Mut+ and MutS Transformants

The colonies of KM71 did not need to be screened because the AOX1 gene was

disrupted. All recombinants will be MutS. For GS115, since our construct was linearized

with Sal I, most of the transformants should be Mut+; however, there is a slight chance

that MutS may appear. Therefore, screening was done for GS115 colonies. Colonies of

GS115 on MD plates were picked and streaked onto a MM plate (1.34% YNB, 4×10-5%

biotin and 0.5 % methanol) and a MD plate in a regular pattern. An MM plate was

patched first. This process was continued until 50 colonies were picked. The plates were

incubated at 30 ºC for 2 days. Mut+ transformants will grow equally well on both MM

and MD plates, while MutS transformants should grow well on MD plates but show little or no growth on the MM plates, because MutS transformants do not produce alcohol

oxidase and cannot efficiently metabolize methanol.

5.2.11. PCR Analysis of Pichia Integrants

It is recommended by the manual that further analysis is needed to confirm that

the gene has been integrated into the genome of Pichia. This was done by using a colony

PCR method from Hahn lab at the Fred Hutchinson Cancer Research Center (FHCRC)

(13). Some medium-size colonies with the right phenotype were transferred to 30 µL

0.2% SDS solution. The cells were resuspended by vortexing for 15 seconds followed by

heating in a hot block for 4 minutes at 90 ºC. The mix was then spun in a micro

centrifuge at 13,000 rpm for 1 min. The supernatant was transferred to a new tube. The

PCR reaction was conducted by following the standard procedure of Pfu turbo, with the

125

exception that the reaction mixture was supplemented with 1% Triton X-100. 5’ AOX1

and 3’AOX1 were used. In order to see the 2.2 kb band of AOX1 gene, the extension time was increased to 3 min. Positive results for Mut+ integrants should show two bands, one

corresponds to the size of the target gene, the other to the AOX1 gene at approximately

2.2 kb. For MutS, only the gene of interest should appear. The same procedure was used with the primers for subcloning, and the cell with the gene of interest inserted should show a single band corresponding to the size of the inserted gene.

5.2.12. Expression of MutS KM71-pPIC9-epen

A single colony of verified recombinant is inoculated in 100 ml BMGY (1% yeast

extract, 2% peptone, 100 mM potassium phosphate, pH 6.0, 1.34% YNB, 4×10-5% biotin and 1% glycerol) in a 1 liter flask. It was kept in an incubator overnight with shaking (275 rpm) at 28 ºC. OD600 should be between 2 and 6 before it was taken out.

The cell was harvested by centrifugation at 5,000 rpm for 10 min. The cell was

resuspended in 20 ml BMMY ( 1% yeast extract, 2% peptone, 100 mM potassium phosphate, pH 6.0, 1.34% YNB, 4×10-5% biotin and 0.5 % methanol) in a 100 ml liter

flask to induce expression at 28 ºC. 100% methanol was added to a final concentration of

0.5-1.0% every 24 hours. At 0, 24, 48, 72, 84, 96, 120 and 144 h, 1 ml of culture was taken out and centrifuged. The supernatant and the cell pellet were stored under -80 ºC until ready to assay. GS115/His+ MutS albumin was grown in the same way as a test for

proper handling.

126

5.2.13. Expression of Mut+ GS115-pPIC9-epen

A single colony of verified recombinant was inoculated in 25 ml BMGY in a 500

ml flask. It was kept in an incubator overnight with shaking (275 rpm) at 28 ºC. OD600 should be between 2 and 6 before it was taken out. The cell was harvested by centrifugation at 5,000 rpm for 10 min. The cell was resuspended to an OD600 of 1.0 in

BMMY in a 2 liter flask to induce expression at 28 ºC. 100% methanol was added to a final concentration of 0.5-1.0% every 24 hours. At 0, 6, 12, 24, 36, 48, 60, 72, 84, and 96 h, 1 ml of culture was taken out, and centrifuged. The supernatant and the cell pellet were stored under -80 ºC until ready to assay. GS115/His+ Mut+ β-Gal was grown in the same

way as a test for proper handling.

5.2.14. Cloning of Mature Ependymin into pRmHa3 for Expression in Insect Cell

Schneider's Drosophila Line-2

In order to insert the mature ependymin gene into pRmHa3, two PCR reactions

were conducted. The first reaction used 5’-GAAGAGCTCATGGCC CCG CGC CCG-3’ as the forward primer. Underlined is the sequence for SstI. And the reverse primer is 5’-

GTG GTG CTC CTC CCA GGA GCA GTC TTC GCT CAT C-3’. The letters in bold are the sequence of two glutamates added to make the overall pI acidic for purification purpose. The second PCR reaction was performed using the product from the first PCR

reaction, the same forward primer, and another reverse primer, the sequence of which is

5’- ATG TCG ACT CAG TGG TGG TGG TGG TGG TGC TCC TCC CAG GAG-3’.

The underlined sequence is SalI. The product of the second PCR reaction was digested

127

with SstI and SalI in Invitrogen reaction buffer 2. The product was purified and ligated

with the pRmHa3 vector treated with the same double digestion procedure. The product was transformed by electrophoresis and selected with LB agar plates containing 50 mg/l ampicillin. The good colonies were picked, extracted plasmids and were then subjected to

DNA sequencing.

5.2.15. Expression, Identification and Purification of Ependymin in Insect Cell

Schneider's Drosophila Line-2 (S2)

S2 cells were transfected with purified plasmid and pUCHsneo (for neomycin selection) using the calcium-phosphate precipitation method. After 6-10 weeks’ selection, selected cells were incubated in Schneider’ Drosophila medium containing 10 % FBS, 2 mM L-glutamine, 50 µg/mL gentamycin for a few weeks to increase the volume. Then it was transferred to a larger volume of Pharmingen BaculoGold protein-free insect medium containing 2 mM glutamine and 50 µg/mL gentamycin. When the cell number reached 10 million/mL, 100 mM CuSO4 was added to induce it. The cells were harvested one week

after induction.

For small scale induction, a small aliquot of transfected S2 cell (10 million

cells/mL, 1 mL volume) was taken out and added 1mM CuSO4 (final concentration) to

induce the target protein. After one week of induction, the supernatant was harvested

(target protein is secreted to medium) and immunoprecipitated with anti-his antibody. In

a 4 ºC water bath, 40 mL of PGS beads were incubated with 4 mL of antibodies in lysis

buffer (50 mM Tris-HCl, 150 mM NaCl, 2 mM EDTA, 1% Igepal CA-630) for 1 hour.

128

Then, the beads were washed with cold PBS buffer 3 times. Collected supernatants or

lysed cell pellets were incubated with anti-His PGS beads overnight. After that, the beads

were washed with lysis buffer 5 times, and subsequently boiled with reducing loading

buffer at 100 ºC for 10 min before loaded on a SDS-PAGE gel. Proteins from the gel

were transferred to a membrane for western blot with anti-His tag antibody (Qiagen), and

Goat-anti mouse IgG1 HRPO conjugate (Caltag).

5.2.16. Cloning of Mature Human Ependymin into pET 21

The mature form of human ependymin (residue 38 to 224) was PCR amplified

using forward primer, 5’-GGGCCCGGCATATGGCCCCGCGCCCGTGC-3’ and

reverse primer, 5’-CCGGCCCTCGAGCCAGGAGCAGTCTTCGCTCA-3’. Underlined

are the NdeI and XhoI restriction sites. The PCR product was subsequently inserted into a

pET-21 vector between restriction site Nde I and XhoI. Potential clones were screened

and further confirmed by nucleotide sequencing.

5.3. Results and Discussion

5.3.1. Cloning and Expression of Ependymin Gene with pET-32

Since pET-32 is for intracellular expression of recombinant genes, the inserted gene encodes the mature ependymin, without the targeting sequence. The construct has a thioredoxin tag, followed by a His-tag, a thrombin cleavage site, an S-tag, an enterokinase cleavage site, ependymin gene and another His-tag (Figure 5.3). A whole

129

cell SDS-PAGE showed that after 4 hours’ induction with IPTG, the cells have produced had a significant amount of recombinant protein (Figure 5.4). Further examination of the supernatant and the pellet indicated that the fusion protein exists mostly in the inclusion body.

130

Figure 5.3. Top, the domains of the ependymin fusion protein overexpressed with a pET-

32b (+). Bottom, pET-32a-c (+) cloning/expression region (adopted from Novagen pET-

32 vector map)

131

1 2 3 4 5

30.0 kDa

20.1 kDa

Figure 5.4. Overexpression of human ependymin-thioredoxin fusion protein. Lane 1, low molecular marker (the molecular weights are 97.0, 66.0, 45.0, 30.0, 20.1 and 14.4 kDa from top to bottom); lane 2, cell extract before adding IPTG; lane 3, one hour after adding IPTG; lane 4, two hours after adding IPTG; lane 5, four hours after adding IPTG.

The lanes between lane 1 and 2 are the control without adding IPTG. The samples were prepared by adding SDS loading buffer to 1 ml cells followed by heating in a boiling water bath for 5 minutes. The supernatant of the extract was used for SDS-PAGE.

132

Human ependymin may be among the most difficult proteins that could possibly be overexpressed. Ependymin from fish was found to precipitate under low calcium concentration (< 1 mM) (5) which is the case in the cytoplasm of E. coli. Ependymin was found to be prone to degradation by an unknown protease in the cytoplasm of E coli.

Breaking the cell with a binding buffer containing 8 M urea followed by centrifugation mainly yields truncated protein (Figure 5.5). In addition, it is a secretary protein from higher eukaryotes.

It has been recognized that linking a target protein to thioredoxin can dramatically increase the solubility and the stability of the overexpressed heterologous protein in E. coli cytoplasm. The yield is usually high. This is because of the intrinsic thermal stability of thioredoxin, which can be retained by the fusion proteins and thus provide convenient purification (14). Therefore, we linked thioredoxin with ependymin to improve the yield, and this indeed produced a sufficient amount of protein for further characterization.

133

1 2 3 4 5 6 7

Figure 5.5. Purification profile of ependymin-thioredoxin fusion protein. Lane 1, low molecular marker (the molecular weights are 97.0, 66.0, 45.0, 30.0, 20.1 and 14.4 kDa from top to bottom); lane 2 and 3, cell lysate extracted by adding 8 M urea to cell pellet before sonication; lane 4, inclusion body solublized by 8 M urea; lane 5, flowthrough, collected when loading onto Ni-NTA column; lane 6, wash with 25 mM imidazole; lane

7, elution of ependymin-thioredoxin fusion with 200 mM imidazole.

134

5.3.2. Purification and On-column Refolding of Thioredoxin-Ependymin Fusion

The overexpressed protein runs at around 40 kDa on SDS-PAGE gel, which corresponds well with the calculated molecular weight of the fusion protein (thioredoxin tags ~ 17 kDa and ependymin ~ 23 kDa). Figure 5.5 shows the purification profile of the fusion protein under denaturing conditions. To get pure and refolded protein, after loading the Ni-NTA column 25 mM imidazole was used to remove non-specifically bound proteins. Then, the column was washed with a total of 20 column volume of binding buffer with a stepwise lowering concentration of urea. The elution from the column contained a single band at around 40 kDa on SDS-PAGE (Figure 5.6).

135

1 2 3 4 5

Figure 5.6. Purification and refolding of ependymin-thioredoxin fusion protein. Lane 1, low molecular marker (the molecular weights are 97.0, 66.0, 45.0, 30.0, 20.1 and 14.4 kDa from top to bottom); lane 2, supernatant of cell lysate; lane 3, inclusion body; lane 4, flowthrough from Ni-NTA column still in 8 M urea; lane 5, elution from Ni-NTA column after on-column refolding.

136

5.3.3. Removing the Tags from the Fusion Protein

In pET-32, the gene for thioredoxin is followed by a His-tag, an S-tag and an enterokinase cleavage site before the multiple cloning sites (Figure 5.3). Enterokinase was used to remove the extraneous peptide.

137

(a) 1 2 3 4 5 6 7 8 9 10

(b) 1 2 3 4

Figure 5.7. Treating ependymin-thioredoxin fusion protein with enterokinase. (a) Lane 1, low molecular marker (the molecular weights are 97.0, 66.0, 45.0, 30.0, 20.1 and 14.4 kDa from top to bottom); lane 2, ependymin-thioredoxin fusion, before adding enterokinase; lane 3, 80 minutes after adding enterokinase; lane 4, 18 hours after adding enterokinase. (b) Lane 1, low molecular marker (the molecular weights are 97.0, 66.0,

45.0, 30.0, 20.1 and 14.4 kDa from top to bottom); lane 2 and 6, cell lysate extracted by adding 8 M urea to cell pellet before sonication; lane 3, 4 and 5, 1, 3 and 4.5 hours after adding enterokinase; lane 7, 8, 9 and 10, 1, 3, 4.5, and 6 hours after adding thrombin.

138

Figure 5.7 (a) shows the result from an initial experiment of enterokinase cleavage. After 80 minutes, the amount of fusion protein decreased and two proteins bands with lower molecular weights appeared. After 18 hours, the two bands that run lower on the SDS-PAGE were further digested. According to the calculated molecular weights (thioredoxin tags ~ 17 kDa and ependymin ~ 23 kDa), the product that runs higher should be ependymin and the one that runs below it should be the thioredoxin tag.

However, both appeared at higher molecular weight on the gel. To further ascertain the identity of the two bands, the ependymin-thioredoxin fusion protein was treated with thrombin and the products were compared with that from enterokinase reaction (Figure

5.7(b)). The fusion proteins used in Figure 5.7 (b) contains both full-length and truncated fusion protein. If comparing lane 5 and 9, the band around 30 kDa in lane 9 ran higher than that in lane 5; if the bands are ependymin, this is consistent with the fact that the thrombin cleavage site is before the enterokinase cleavage site in Figure 5.7. For the other product bands between 20 and 14.4 kDa, since the truncation site is within ependymin, the band ran lower in lane 10 than that in lane 5.

Enterokinase is a protease that cleaves after the lysine in the Asp-Asp-Asp-Asp-

Lys cleavage site in trypsinogen to produce trypsin (15). However, it lacks stringent specificity and may sometimes cleave at other basic residues, and the conformation of the protein substrate is an important factor affecting its specificity (16). In our attempt to get ependymin from the fusion protein, non-specific cleavage occurred and the yield of ependymin was lowered significantly.

139

5.3.4. Circular Dichroism of Thioredoxin-Ependymin Fusion Protein

Since it was difficult to obtain high quality ependymin due to the reasons mentioned earlier, the ependymin fusion protein was used to access if the ependymin was refolded. The predicted secondary structure was compared with the published data on thioredoxin. By using CDPro, ependymin-thioredoxin fusion was predicted to have 7% helices and 41.2% strands; while thioredoxin has 33% helices and 27% strands. The composition of the secondary structures changed considerably, which could mean that the ependymin portion has folded into an arrangement of secondary structures and many of them could be strands. This is consistent with the fact that fish ependymin consists of a significant portion of strands (17). Conformational changes occurred after adding calcium to fish ependymin, as indicated in a CD experiment (17). Lowering the calcium concentration could initiate a change in the conformation of ependymin. As the change increases to certain extents, ependymin may fold into a conformation that is prone to aggregate. If human ependymin has similar characters, this could be used as evidence whether the refolded human ependymin is active or not. As shown in Figure 5.8, after adding calcium, there were minor changes in the near-UV and far-UV region which could indicate that conformational changes occurred in response to the change of calcium concentration. To further confirm that the change truly occurs, a titration experiment could be useful titrating calcium into protein solution, and the conformational changes can be followed by CD. Some control experiments remain to be done, such as adding calcium to thioredoxin fusion to see if this part changes conformation, and adding calcium to the buffer alone and using it as the reference.

140

0.0 1500 Ependymin fusion ExpCD 1000 Ependymin fusion with calcium -0.2 CalcCD 500 )

-1 0 -0.4 -500 dmol 2

) -0.6 -1000 -1 cm

-1500

-1 -0.8

(M -2000 δε (degrees.cm

M -2500 -1.0 [θ] -3000 -1.2 a -3500 b -4000 -1.4 200 210 220 230 240 200 250 300 350 400 Wavelength (nm) Wavelength (nm)

Figure 5.8. Circular Dichroism characterization of ependymin-thioredoxin fusion. (a),

Far-UV region of ependymin-thioredoxin fusion; the black line is the experimental data and the red line is the calculated CD spectrum plotted using predicted secondary structure data; (b), the CD spectra of ependymin-thioredoxin fusion before (black) and after (red) adding calcium to 2 mM.

141

5.3.5. Probing the Calcium Binding Sites by Terbium Luminescence

It is reported that ependymin undergoes a conformational change and could aggregated into a polymer form when the concentration of calcium changes (17) (5).

Therefore, ependymin might be a calcium binding protein. It was found that calcium binding was mostly attributed to the N-linked oligosaccharides in fish ependymin (17), and the protein backbone possesses limited calcium binding affinity (17). Ependymin does not belong to any of the well characterized calcium binding protein families. A similarity search didn’t detect any consensus motifs or homologous structures. Moreover, in the cerebrospinal fluid where ependymin is primarily found, the calcium concentration is in the milimolar range. Therefore, we speculate that if the human ependymin backbone can bind calcium, the binding affinity would be very low. Some initial experiments were performed with ITC or titration fluoremetry to check if human ependymin binds calcium.

None of these gave a conclusive answer.

Terbium has long been used to probe calcium binding sites due to the ionic and spectroscopic characteristics of lanthanides (18-20). The lanthanides and calcium are similar in terms of the variable coordination number, ionic radii, and preference of oxygen as the donor groups over nitrogen or sulfur, and lack of strong directionality in binding donor groups (21). Lanthanides usually have a higher binding affinity to proteins than calcium (22). A very useful lanthanide is terbium, which is luminescent upon Fórster energy transfer from tryptophan or tyrosine residues (21). If terbium can bind to a protein, the excitation of the aromatic residues of the protein may excite terbium, and thus the emission spectrum of terbium can be observed. In our case, terbium would be particularly

142

useful because it might enhance the binding affinity of ependymin to a detectable level.

When terbium chloride solution was titrated into the ependymin-thioredoxin fusion, and

excited with a light at wavelength of 280 nm, the characteristic emission spectra of

terbium appeared with the increase of calcium concentration (Figure 5.9). The intensity at

488 nm increased to a greater extent than the baseline. In the control experiment, the changes of intensity at 488 and 530 nm almost overlap. This indicates that terbium might bind to the ependymin-thioredoxin fusion. The concentration of terbium reached 350 µM at the end of the experiment. The bind affinity of terbium towards ependymin-thioredoxin fusion is also low. There is no indication that thioredoxin bind to calcium. This could suggest that ependymin-thioredoxin fusion might have some calcium binding affinity.

143

1400

1200

1000

800

600 Intensity 400

200

-200 440 460 480 500 520 540 Wavelength (nm)

900 Emission at 488 nm Emission at 530 nm 800 Emission difference Control, emission at 488 nm 700 Control, emission at 530 nm 600

500

400 Intensity 300

200

100

-50 0 50 100 150 200 250 300 350 400 Concentration of TbCl (μM) 3

Figure 5.9. Terbium may bind to ependymin-thioredoxin fusion. Top, the emission spectra of terbium as terbium was added to ependymin-thioredoxin fusion protein;

Bottom, comparison of the increase of fluorescence intensity at 488 and 530 between the experiment and control.

144

5.3.6. Overview of the Pichia Expression System

Pichia Pastoris is methylotrophic yeast. The first step in its metabolism is the

oxidation of methanol as its sole carbon source by the enzyme alcohol oxidase. Alcohol

oxidase has a poor affinity towards O2; therefore, Pichia produces a large amount of this enzyme to compensate. Two genes code for alcohol oxidase, AOX1 and AOX2. The

AOX1 gene product accounts for the majority of alcohol oxidase activity in the cell. The expression of the alcohol oxidase genes are regulated and induced by methanol. The promoter of AOX1 and AOX2 genes is used to drive heterologous protein expression in

Pichia. As a result, Pichia has the advantage of 10- to 100-fold higher heterologous protein expression level, compared to other yeasts (23). Moreover, Pichia has much shorter oligosaccharide chains compared to Saccharomyces cerevisiae (24) and the linkage of the oligosaccharide chains may be more compatible for therapeutic use on humans (25). Pichia also has many advantages of higher eukaryotic expression systems.

All of these make Pichia an increasingly more popular expression system.

The process to create a Pichia strain containing the desired gene includes cloning the gene of interest into selected vectors, transforming into E. coli and selecting with plates containing proper antibiotics, converting the plasmid containing the gene into a linear form, transforming it into yeast strains, screening for phenotypes of the strains, and confirming the integration of the desired gene. After all those steps, the positive strains can be used to express the recombinant gene.

145

5.3.7. Cloning of Human Ependymin into pPIC9 for Expression in Pichia

Since ependymin is a secretory protein, to facilitate exporting protein into the medium, the ependymin gene was cloned into pPIC9, which contains a α-factor signal sequence. Figure 5.10 shows the multiple cloning sites of pPIC9. After confirming the

DNA sequence and cutting it into a linear form using proper enzymes according to the manual, the linear DNA was transformed into Pichia to generate stable transfromants via homologous recombination between the regions of homology within the genome and the transforming DNA. Some colonies were selected for colony PCR with the primers for subcloning into pPIC9. In the top graph of Figure 5.11, all four colonies shows bands around 600 bp, consistent with the size of the inserted gene. In the bottom of the same graph, the colony PCR was done with AOX1 primers and the size of these DNA bands increased to around 1.0 kbp. This is because in between the two primers in Figure 5.11, there is the α-factor sequence and some additional sequences.

146

Figure 5.10. Multiple cloning site of pPIC 9 and surrounding sequences.

147

1 2 3 4 5

1.0 kbp

0.5 kbp

1 2 3 4 5

1.0 kbp

0.5 kbp

Figure 5.11. Colony PCR of transfromants with primers for subcloning (top) or AOX primers (bottom). For top and bottom, lane 1, 1 kb DNA ladder; and lane 2-5, PCR product.

148

The positive colonies need to be screened for phenotypes. The colonies can be

either Mut+ (both AOX1 and AOX2 are intact) or MutS (defect AOX1 gene, intact AOX2

gene). The MutS strain grows much more slowly than the Mut+ strain. Mut+

transformants will grow equally well on both MM and MD plates, while MutS

transformants should grow well on MD plates but show little or no growth on the MM

plates. 50 of the colonies of GS115 picked up grew well on the MD plates and 48 of them

grew well on the MM plates. The colonies that grew well on both plates were picked for

further confirmation of construct integration. There is no need to screen for KM71 since

the strain itself lacks an intact AOX1 gene. Figure 5.12 shows the colony PCR of pPIC9-

epen in GS115 priming with the 5´ and 3´AOX1 primers. Lane 2 and 3 do not have the

inserted gene; therefore, the size of the DNA band would be equal to the size of the

sequence between two AOX1 primer sites, which is 0.5 kbp. Lanes 4-8 are the GS115

strain, so they can be either MutS or Mut+. The bands around 2.2 kbp indicate that they may have the intact AOX1 gene. The product around 1.0 kbp should be the same as that shown in Figure 11. Lanes 9 and 10 are the KM71 strain so they should only have the product corresponding to the inserted gene and surrounding sequences.

149

1 2 3 4 5 6 7 8 9 10

2.0 kbp 1.0 kbp

0.5 kbp

Figure 5.12. Screening for phenotypes of Pichia strains. Lane 1, 1 kb marker (NEB); lane

2-3, Colonies with no insert in pPIC9; lane 4-8, Colonies of GS115/Mut+-pPIC9-epen; lane 9-10, Colonies of KM71/MutS-pPIC9-epen.

150

5.3.8. Expression of MutS KM71-pPIC9-epen and Mut+ GS115-pPIC9-epen

Aliquots of growth medium were collected and subject to SDS-PAGE. No

obvious increase in intensity of protein bands was observed. The medium was

concentrated, exchanged to the binding buffer for Ni-NTA column and loaded onto the

column; all proteins didn’t seem to bind.

Various expression conditions were used, including different growth medium,

supplements in medium, pH, and temperature. None of them give significant

improvement. The cells were harvested and lysed. The lysate was checked for protein expression, because the signal peptide may not be properly processed, the overexpressed protein may not be able to export to the medium. No expression was detected.

5.3.9. Cloning, Expression and Purification of Ependymin in Insect Cell Schneider's

Drosophila Line-2

Insect cells are a higher eukaryotic system than yeast, and are able to carry out more complex post-translational modifications that are closer to what is seen in mammalian cells (26). They also have better machinery for the folding of mammalian proteins which gives a better chance of obtaining soluble proteins (27). There are

advantages in the transcription and translation step, as well as in folding, post-

translational modification, and oligomerization, which are often found to be identical to

those occur in mammalian cells (28). In addition, the non-reducing environment of insect

cytoplasmic environment allows proper folding and S-S bond formation. Moreover,

insect cell expression is optimal for glycosylated protein expression in a cost-effective

151

manner (29). Post-translational processing including N-glycosylation identical to that of

mammalian cells has been reported for many proteins (26).

We have tried to express ependymin without thioredoxin fusion in E coli, and two different types of yeast. None of them produce enough protein for further characterization.

This is probably because the mammalian origin of ependymin, the potential to aggregate, being prone to protease digestion (30), multiple disulfide bonds needed to fold correctly and sialic acid rich type glycosylation. Herein we introduced it into an insect cell expression system for overexpression. Figure 5.13 shows the process of cloning mature ependymin into pRmHa3 for expression. The second PCR was to add a His-tag at the C- terminal of ependymin to facilitate protein detection and purification.

152

LL P1 P2 RER V C1 C2 C3 HL

2000 bp 1200 bp 800 bp

400 bp

200 bp

Figure 5.13. Cloning of mature ependymin into pRmHa3 for expression. LL, Low DNA ladder; P1, Product from the first PCR reaction; P2, Product from the second PCR reaction; RER, P2 after restriction enzyme reaction with SstI & SalI; V, pRmHa3 vector after restriction enzyme reaction with SstI & SalI; C1, C2, and C3: colonies that are

believed to have insert cloned in; HL, high DNA ladder.

153

The mature ependymin (ependymin without the original signal peptide) is targeted to the medium. Both the medium and cell pellets were immunoprecipitated with anti-His antibody conjugated PGS beads. A band around 30 kDa appeared in the secreted proteins which is a few kilo-Daltons higher than the size of mature ependymin (Figure 5.14). This could be interpreted as glycosidated ependymin but needs further confirmation.

The medium from a large scale expression was loaded onto Ni-NTA for purification (Figure 5.15). The red arrows indicate where the ependymin might be. It seemed that the protein was eluted out by 40 mM imidazole. The anti-His western blot was not highly selective and many other protein bands were picked up.

154

Figure 5.14. Immunoprecipitation with anti-His antibody conjugated PGS beads.

155

Figure 5.15. Purification of ependymin and Western Blot. Gel A: Coomassie blue staining; Gel B: anti-His tag Western Blot. M, Molecular Marker; 1, S2 cell medium; 2,

Flow through (Ni-NTA column); 3, Wash step with 5 mM imidazole in binding buffer; 4,

Wash step with 40 mM imidazole in binding buffer; 5, Wash step with 40 mM imidazole in binding buffer; 6-11, Elution with 200 mM imidazole, 12-14, Elution with 400 mM imidazole.

156

5.4. References for Chapter 5

1. Sterrer, S., Konigstorfer, A., and Hoffmann, W. (1990) Biosynthesis and expression of ependymin homologous sequences in zebrafish brain, Neuroscience 37, 277-284.

2. Konigstorfer, A., Sterrer, S., and Hoffmann, W. (1989) Biosynthesis of ependymins from goldfish brain, J Biol Chem 264, 13689-13692.

3. Shashoua, V. E. (1991) Ependymin, a brain extracellular glycoprotein, and CNS plasticity, Ann N Y Acad Sci 627, 94-114.

4. Shashoua, V. E. (1985) The role of brain extracellular proteins in neuroplasticity and learning, Cell Mol Neurobiol 5, 183-207.

5. Shashoua, V. E. (1988) Monomeric and polymeric forms of ependymin: a brain extracellular glycoprotein implicated in memory consolidation processes, Neurochem Res 13, 649-655.

6. Shashoua, V. E. (1988) The role of ependymin in the development of long lasting synaptic changes, J Physiol (Paris) 83, 232-239.

7. Hebb, D. O. (2002) The organization of behavior : a neuropsychological theory, L. Erlbaum Associates, Mahwah, N.J.

8. Paulsen, O., and Sejnowski, T. J. (2000) Natural patterns of activity and long-term synaptic plasticity, Curr Opin Neurobiol 10, 172-179.

9. Fessard, A., Delafresnaye, J. F., and Council for International Organizations of Medical Sciences. (1961) Brain mechanisms and learning : a symposium, Blackwell Scientific Publications, London.

10. Gaiarsa, J. L., Caillard, O., and Ben-Ari, Y. (2002) Long-term plasticity at GABAergic and glycinergic synapses: mechanisms and functional significance, Trends Neurosci 25, 564-570.

11. Sreerama, N., and Woody, R. W. (2000) Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set, Anal Biochem 287, 252-260.

12. Provencher, S. W., and Glockner, J. (1981) Estimation of globular protein secondary structure from circular dichroism, Biochemistry 20, 33-37.

13. Fishburn, J., Mohibullah, N., and Hahn, S. (2005) Function of a eukaryotic transcription activator during the transcription cycle, Mol Cell 18, 369-378. 157

14. LaVallie, E. R., DiBlasio, E. A., Kovacic, S., Grant, K. L., Schendel, P. F., and McCoy, J. M. (1993) A thioredoxin gene fusion expression system that circumvents inclusion body formation in the E. coli cytoplasm, Biotechnology (N Y) 11, 187-193.

15. Light, A., and Janska, H. (1989) Enterokinase (enteropeptidase): comparative aspects, Trends Biochem Sci 14, 110-112.

16. Shahravan, S. H., Qu, X., Chan, I. S., and Shin, J. A. (2008) Enhancing the specificity of the enterokinase cleavage reaction to promote efficient cleavage of a fusion tag, Protein Expr Purif 59, 314-319.

17. Ganss, B., and Hoffmann, W. (1993) Calcium binding to sialic acids and its effect on the conformation of ependymins, Eur J Biochem 217, 275-280.

18. Wallace, R. W., Tallant, E. A., Dockter, M. E., and Cheung, W. Y. (1982) Calcium binding domains of calmodulin. Sequence of fill as determined with terbium luminescence, J Biol Chem 257, 1845-1854.

19. Szebenyi, D. M., Obendorf, S. K., and Moffat, K. (1981) Structure of vitamin D- dependent calcium-binding protein from bovine intestine, Nature 294, 327-332.

20. Sundaralingam, M., Bergstrom, R., Strasburg, G., Rao, S. T., Roychowdhury, P., Greaser, M., and Wang, B. C. (1985) Molecular structure of troponin C from chicken skeletal muscle at 3-angstrom resolution, Science 227, 945-948.

21. Snyder, A. P., Sudnick, D. R., Arkle, V. K., and Horrocks, W. D., Jr. (1981) Lanthanide ion luminescence probes. Characterization of metal ion binding sites and intermetal energy transfer distance measurements in calcium-binding proteins. 2. Thermolysin, Biochemistry 20, 3334-3339.

22. Rhee, M. J., Horrocks, W. D., Jr., and Kosow, D. P. (1982) Laser-induced europium(III) luminescence as a probe of the metal ion mediated association of human prothrombin with phospholipid, Biochemistry 21, 4524-4528.

23. Cregg, J. M., Cereghino, J. L., Shi, J., and Higgins, D. R. (2000) Recombinant protein expression in Pichia pastoris, Mol Biotechnol 16, 23-52.

24. Grinna, L. S., and Tschopp, J. F. (1989) Size distribution and general structural features of N-linked oligosaccharides from the methylotrophic yeast, Pichia pastoris, Yeast 5, 107-115.

25. Cregg, J. M., Vedvick, T. S., and Raschke, W. C. (1993) Recent advances in the expression of foreign genes in Pichia pastoris, Biotechnology (N Y) 11, 905-910.

158

26. Schmidt, F. R. (2004) Recombinant expression systems in the pharmaceutical industry, Appl Microbiol Biotechnol 65, 363-372.

27. Murhammer, D. W. (2007) Baculovirus and insect cell expression protocols, 2nd ed., Humana Press, Totowa, N.J.

28. Murhammer, D. W. (2007) Baculovirus and insect cell expression protocols, 2nd ed., pp xiv, 470 p., Humana Press, Totowa, N.J.

29. Harrison, R. L., and Jarvis, D. L. (2006) Protein N-glycosylation in the baculovirus-insect cell expression system and engineering of insect cells to produce "mammalianized" recombinant glycoproteins, Adv Virus Res 68, 159-191.

30. Adams, D. S., Hasson, B., Boyer-Boiteau, A., El-Khishin, A., and Shashoua, V. E. (2003) A peptide fragment of ependymin neurotrophic factor uses protein kinase C and the mitogen-activated protein kinase pathway to activate c-Jun N-terminal kinase and a functional AP-1 containing c-Jun and c-Fos proteins in mouse NB2a cells, J Neurosci Res 72, 405-416.

159

CHAPTER 6

HUMAN MITOCHONDRIAL HSCB CHAPERONE MEDIATE MATURATION OF HUMAN FRATAXIN

6.1. Introduction

Frataxin is a mitochondrial matrix protein that is involved in iron metabolism (1-

4). A genetic defect in the gene (FRDA) that encodes frataxin results in a neurological disorder termed Friedreich’s ataxia. The accumulation of mitochondrial iron has been observed in nerve and cardiac tissues of Friedreich’s ataxia patients. Frataxin is a 210 amino acid protein that includes N-terminal mitochondrial targeting sequences (residues

1-55) that are cleaved by the mitochondrial processing peptidase (5-8). The crystallographically determined structure of human frataxin (9, 10) shows that several conserved acidic residues are mainly located in the α1 helix and β1 sheet, and that these residues define an anionic surface (Figure 6.1). Recently, several groups have investigated the iron-delivery functions of frataxin, including assisting iron-sulfur cluster biosynthesis (11-14), heme formation (15-18), and aconitase activation (19). We and others have suggested that the putative iron binding sites might be located on the anionic

160

surface. The iron binding sites of bacterial, yeast and human frataxin have been evaluated

by NMR spectroscopy (17, 20) and the results are consistent with such a model. Several

groups (10, 13, 16, 21) have observed the cleavage of N-terminal residues in human

frataxin both in vivo and in vitro but the reason and role of the cleavage have not been

studied. Recently, we characterized an iron-mediated auto-cleavage pathway involving

the N-terminal domain of full-length frataxin from human and suggested a role for this N-

terminal domain as a structural switch to mask the multiple iron-binding sites to prevent

non-specific interactions with other protein partners or charged species, or to open the anionic surface for interaction (22). Although the in vitro cleavage of full-length human frataxin happened by iron-mediated self-cleavage, it is necessary to further study the N- terminus cleavage event in vivo.

161

Figure 6.1. Ribbons diagram of Hs frataxin (PDB ID: 1EKG) showing surface carboxylates that are putative binding sites for iron ions.

162

In E. coli the isc operon encodes gene products involved in the biosynthesis of iron sulfur cluster assembly (23, 24). The genes hscA and hscB encode an HscA, (Hsp70 family) and HscB, (a J-type co-chaperone), respectively. Because of the existence of the molecular chaperone genes in isc genes, many believed that the molecular chaperones might be involved in the iron-sulfur cluster biosynthesis or maintenance. Vickery and coworkers (25-29) have demonstrated interactions between the scaffold protein IscU and chaperones Hsc66 and Hsc20 in E. coli. They clearly demonstrated that IscU is a partner of Hsc66 and Hsc20. Ssq1 and Jac1 chaperones (30-34) are yeast homologues of HscA

(or Hsc66) and HscB (or Hsc20) chaperones, respectively, which have been implicated in

iron-sulfur cluster biosynthesis and iron homeostasis. Interestingly, several groups (30,

31) have reported the roles of mitochondrial Ssq1 (Hsp70 family in yeast) and Jac1 (J-

type co-chaperone in yeast) chaperones in the maturation of Yfh1 (yeast frataxin

homologue). The two-step removal of a mitochondrial targeting sequence for frataxin is

carried out by the mitochondrial processing peptidase (MPP). Dancis and co-workers (30,

34) have observed that the secondary processing cleavage step of the yeast frataxin

homologue for the mature form was delayed in certain mutants; including Δssq1, Δjac1, and the Δssq1Δjac1, double mutant. Although the functional significance of this observation was uncertain, it was clear that ssq1 and jac1 mutant mitochondria exhibit

similar defects in Yfh1p maturation. They also suggested that frataxin could be another

substrate for the HscA/HscB chaperone system. The same phenomenon was also

observed by Craig and co-workers (31). They mentioned that there are two Hsp70 type

molecular chaperones, Ssc1 and Ssq1 in the mitochondrial matrix of yeast 163

Saccharomyces cerevisiae. Ssc1 is 1000-fold excess over Ssq1 and is required for Yfh1

import into mitochondrial matrix. Ssq1 is required for the efficient processing of Yfh1 to

the mature mitochondrial form. However, they were unable to observe any stimulation of

Ssq1 ATPase activity by Yfh1 in the presence of Jac1, or any interaction between Yfh1

(matured form) and Jac1 (35). While lacking evidence for any possible interaction between the mature Yfh1 and Jac1, the interaction between the premature form of Yfh1 and Jac1 remained open.

According to Huynen and co-workers (36), the phylogenetic distribution of frataxin is identical with that of the HscA/HscB chaperones. They also predicted a functional connection between frataxin and the HscB chaperone based on the co- occurrence of the genes and the similar distribution of their encoded proteins in the eukaryotic cell. Taken together, several interesting questions remain to be answered. In particular “Is frataxin another substrate for the HscA/HscB chaperones? If so, what is the function of the HscA/HscB chaperone system?” Herein we demonstrated human HscB to mediate N-terminal cleavage of human frataxin, and suggested a role of HscB chaperone as a trigger to change frataxin into a functional form, in which the anionic surface becomes accessible.

164

6.2. Materials and Methods

6.2.1. General Chemicals

NTA resin was purchased from QIAGEN (Valencia, CA). Homogenous-20 precast

polyacrylamide gels were purchased from Pharmacia (NJ). EDTA was purchased from

Aldrich.

6.2.2. Cloning, Expression and Purification of Human HscB, J-type Co-chaperone

The expression and purification of human frataxin was performed as previously

described (13, 16). The DNA sequence corresponding to the mature HscB protein,

beginning with an aspartic acid which immediately follows the mitochondrial targeting

sequence as predicted by an sequence alignment of E. coli and human HscB, was amplified by the polymerase chain reaction (PCR) from the Marathon-ReadyTM Human

Heart cDNA library (Clontech). The following primers were used: 5’ primer: 5’ – GGA

AGG CCA TAT GGA CTA CTT CAG CCT TAT GGA CTG CAA CC – 3’; 3’ primer:

5’ – ACC GCT CGA GCC ACA ATT AAA GGG GAA TCT TCT TTA ACT T – 3’.

These primers were based on the human HscB sequence and incorporated NdeI and XhoI

restriction sites (underlined) at the 5’ and 3’ ends, respectively. The reaction mixture

contained 1 µL of cDNA template (48 ng/µL), 3 µL of primers (10 µM each primer), 25

µL of IQ supermix (Bio-Rad) and 18 µL of sterile ddH2O. PCR amplification conditions

followed a protocol with denaturation at 95 oC for 2 min 30 sec followed by an

165

amplification sequence of 94 oC (45 s), 55 oC (45 s), and 68 oC (45 s) over 33 cycles. A

final extension was achieved at 72 oC for 2 min. The amplified HscB gene was digested

with restriction enzymes NdeI and XhoI and cloned into a similarly treated expression

vector pET-28b (+) to produce plasmid pET-28b (+)-(HscB). The DNA sequence of the

cloned HscB gene product was confirmed by DNA sequencing. Plasmid pET-28b (+)-

(HscB) was transformed into E. coli BL21(DE3) (Novagen). For protein hyper

production, an overnight culture (40 mL) was used as inoculums for a 4 L LB medium

containing 40 µg/mL kanamycin. Cells were grown at 37 oC to an OD of 0.6 ~ 0.8 and

expression was induced by the addition of IPTG to a final concentration of 0.3 mM. The culture was incubated for an additional 5 h, and then the cells were harvested by centrifugation. The cell pellet was washed with 20 mM Tris-HCl, 500 mM NaCl pH 7.9,

0.4 % sodium cholate buffer and stored at - 80 oC until used.

Cells were thawed on ice and re-suspended in binding buffer (5 mM imidazole, 20

mM Tris-HCl, 500 mM NaCl pH 7.9) containing 1 mM PMSF and lysed by sonication.

The cellular debris was removed by centrifugation at 15000 rpm, 4 oC for 30 min, and the

supernatant was loaded onto a Ni-NTA column (Qiagen) previously equilibrated with the

binding buffer. After loading, the column was washed with 10 volumes of the wash

buffer (40 mM imidazole, 20 mM Tris-HCl, and 500 mM NaCl pH 7.9) and eluted with the elution buffer (200 mM imidazole, 20 mM Tris-HCl, 500 mM NaCl pH 7.9).

Fractions containing HscB as judged by SDS-PAGE were pooled and concentrated by ultrafiltration using an Amicon stirred cell concentrator. Finally, the buffer was exchanged by dialysis. Protein purity was monitored by SDS-PAGE and the final

166

- concentration was determined from the calculated extinction coefficient (ε278 = 8,400 M

1cm-1).

6.2.3. Quantitation of HscB Protein Binding with Frataxin by Isothermal Titration

Calorimetry

ITC measurements of human HscB to full-length human frataxin and truncated

frataxin were carried out at 25 °C on a MicroCal VP-ITC. The titrant and sample

solutions were made with the same buffer solution (50 mM HEPES buffer, pH 7.50, 100

mM NaCl), and both experimental solutions were thoroughly Ar-purged and degassed

before each titration. The solution in the cell was stirred at 200 rpm by syringe to ensure

rapid mixing. Typically, 5 - 10 µL of titrant (HscB protein) was delivered to a solution of

20 ~ 50 µM truncated or full-length frataxin in the absence or in the presence of 10 fold

excess ferrous iron over a period of 20 s with an adequate interval (5 - 10 min) between

injections to allow complete equilibration. To keep the reducing environment, 5 – 10 mM

dithionite were applied to the buffer in the presence of ferrous iron. Titrations continued

until 3-5 equivalents had been added to ensure no further complex formation following

addition of excess titrant. A background titration, done with the identical titrant, and the

buffer solution alone in the sample cell, was subtracted from each experimental titration

to account for heat of dilution.

The data were collected automatically and subsequently analyzed with a one-site

binding model by the Windows-based Origin software package supplied by MicroCal.

The Origin software uses a nonlinear least-squares algorithm (minimization of χ2) and the

167

concentrations of the titrant and the sample to fit the heat flow per injection to an equilibrium binding equation, providing best fit values of the stoichiometry (n), change in enthalpy (ΔH), change in entropy (ΔS), and association constant (KA).

6.2.4. Quantitation of Metal Binding to HscB by Isothermal Titration Calorimetry

ITC measurements for the interaction between human HscB and zinc or ferrous ions were carried out at 25 °C on a MicroCal VP-ITC. The titrant and sample solutions were made in the same stock buffer solution (50 mM HEPES buffer, pH 7.50, 100 mM

NaCl), and both experimental solutions were thoroughly Ar-purged and degassed before each titration. When using zinc, extensive Ar-purge was not necessary. The solution in the cell was stirred at 300 rpm by the syringe to ensure rapid mixing. Typically, 24 µL titrant containing metal ions was delivered to a solution of 20 ~ 50 µM HscB in the presence of 2 mM TCEP and 2 mM dithionite (when using ferrous ion) over a period of

20 s with an adequate interval (5-10 min) between injections to allow complete equilibration. Titrations continued until 3-5 equivalents had been added to ensure no further complex formation following addition of excess titrant. To monitor a background titration, consisting of the identical titrant solution, but containing only the buffer solution in the sample cell, was subtracted from each experimental titration to account for heat of dilution. Data analysis followed the same procedure noted earlier.

168

6.2.5. Self-cleavage of N-terminal His-tagged full-length Frataxin under Different

Conditions

To a solution containing 30 µM full-length frataxin was added 200 µM ferric ions,

or 2 mM EDTA, or 30 µM HscB depending on the desired condition. The reaction

mixture was incubated at 4 or 25 ºC for several days and the cleavage process was check periodically by SDS-PAGE. Aliquots were taken from the sample and mixed with SDS loading buffer and stored at -20 ºC until use. Samples at different time points under different conditions were run SDS-PAGE and the degradation was quantitated by gel-doc.

The control experiments were done by incubating HscB or full-length frataxin alone at 4 or 25 ºC. The percent cleavage was calculated by dividing the intensity of the cleavage product at different time points by that of final product. The percent cleavage was plotted against time which was fitted with a first-order-decay equation.

6.2.6. Self-cleavage of C-terminal His-tagged Full-length Frataxin with the Aid of

HscB

To a solution containing 25 µM C-terminal His-tagged full-length frataxin and 2 mM TCEP was added 25 µM ferric ions, or 25 µM HscB or both ferric ion and HscB.

The reaction mixture was incubated at 25 ºC for several days and the cleavage process was check periodically by SDS-PAGE. The degradation was quantitated by gel-doc. The percent cleavage was calculated by dividing the intensity of the full-length frataxin at different time points by the total intensity of the final product and the full-length frataxin

169

that has not been cleaved. The percent cleavage was plotted against time which was fitted with a first-order-decay equation.

6.2.7. Cross-linking Experiment

Intermolecular cross-linking experiments using 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) between HscB protein and frataxin were performed on ice for 2 h. All proteins were at 50 µM in 50 mM HEPES, pH 7.5, 100 mM NaCl buffer and a 10-fold molar excess of cross-linker (EDC) was applied for the reactions. All the products were visualized by SDS-PAGE.

6.2.8. Mass Spectroscopy

All mass spectra were acquired from CCIC (campus chemical instrument center) at The Ohio State University. Mass determination was performed by MALDI-TOF

(Bruker Reflex III MALDI-TOF) mass spectrometer. HscB was dialyzed against nano-

pure water prior to mass spectroscopic examination.

6.3. Results

6.3.1. Cloning, Expression, and Purification of the Human HscB Chaperone

Vickery and coworkers had identified the gene sequence of human HscB (37), its

genomic locus, and its expression pattern in normal human tissues. The highest levels of

mRNA for HscB are mainly found in the liver, muscle and heart. Such an expression pattern of HscB gene in mitochondria-rich tissues is similar to that of human frataxin. 170

The sequence of the human HscB gene, lacking the mitochondrial targeting

sequence (residues 1-71), was targeted for cloning. The mature HscB was cloned from

human cDNA library and was subsequently over-expressed in E. coli. The protein, which

has an N-terminal His-tag, was initially purified by Ni-NTA affinity chromatography.

SDS-PAGE showed a molecular weight of HscB at a little more than 20 kDa, which agreed with the predicted molecular weight of 21.7 kDa for the His6-tagged HscB protein.

The mass was accurately defined by MALDI-TOF (Mr = 21,699), which corresponds to

the His-tagged HscB protein within 0.14 %, smaller than a normal error (< 0.5 %) from

MALDI-TOF. The purified HscB protein is a monomeric, soluble protein, similar to

Hsc20 (E. coli) or Jac1p (S. cerevisiae). Sequence alignments (Figure 6.2) show that

these proteins commonly have the signature sequence of J-type chaperone, His-Pro-Asp.

171

Hsc20_Ec ------MDYFTLFGLPAR-----YQLDTQALSLRFQDLQRQYHPDKFASGSQAEQ 44 Hsc B_Hs ------DYFSLMDCNRS-----FRVDTAKLQHRYQQLQRLVHPDFFSQRSQTEK 114 Jac1p_Sc MLKYLVQRRFTSTFYELFPKTFPKKLPIWTIDQSRLRKEYRQLQAQHHPDMAQQGS---- 56 :: *: : :* * .:::** *** . *

Hsc20_Ec LAAVQQSATINQAWQTLRHPLMRAEYLLSLHG-FDLASEQHT----VRDTAFLMEQLELR 99 Hsc B_Hs DFSEKHSTLVNDAYKTLLAPLSRGLYLLKLHG-IEIPERTDY----EMDRQFLIEIMEIN 169 Jac1p_Sc ----EQSSTLNQAYHTLKDPLRRSQYMLKLLRNIDLTQEQTSNEVTTSDPQLLLKVLDIH 112 ::*: :*:*::** ** *. *:*.* :::... * :*:: :::.

Hsc20_Ec EELDEIEQAKDEARLESFIKRVKKMFDTRHQLMVEQLDNETWDAAADTVRKLRFLDKLRS 159 Hsc B_Hs EKLAEAESEAAMKEIESIVKAKQKEFTDNVSSAFEQDD---FEEAKEILTKMRYFSNIEE 226 Jac1p_Sc DELSQMDDEAGVKLLEKQNKERIQDIEAQLGQCYNDKD---YAAAVKLTVELKYWYNLAK 169 ::* : :. :*. * : : . :: * : * . :::: :: .

Hsc20_Ec SAEQLEEKLLDF--- 171 Hsc B_Hs KIKLKKIPL------235 Jac1p_Sc AFKDWAPGKQLEMNH 184 :

Figure 6.2. Amino acid sequence alignment of HscB proteins from E. coli (Hsc20), Jac1p from Saccharomyces cerevisiae and residues # 72 ~ 235 of Homo sapiens (mitochondrial targeting sequence (residues 1 ~ 71) is not included). Sequence identities are indicated by asterisk (*) and similarities are presented by a dot (.) or a colon (:). The highlighted sequence, His-Pro-Asp, shows the J-protein signature sequence. The alignment of the sequences is the result of ClustalW, which can be accessed at the European

Bioinformatics Institute http://www.ebi.ac.uk/clustalw.

172

6.3.2. Interaction between HscB and Frataxin

To understand the functional relationship between HscB and frataxin, we first looked for interactions between these two proteins by cross-linking experiments.

Incubation of HscB with the cross-linking reagent EDC in the presence of human frataxin

(truncated form) showed five bands corresponding to the monomeric forms of both HscB

and frataxin, the dimeric forms of both HscB (a) and frataxin (b) and a cross-linking

product (*) (Figure 6.3). Although the band intensity of the cross-linked product is weak,

it is clearly visible in lane 6, but not in lane 5.

173

kDa M 1 2 3 4 5 6 M 45.0

30.0

20.1

14.4

Figure 6.3. Crosslinking between HscB and frataxin. 50 µM of HscB and 50 µM of

truncated human frataxin in 50 mM HEPES, 100 mM NaCl pH 7.4 buffer were incubated

on ice for 2 hour with or without 10-fold molar excess EDC. Lane M: low molecular

weight marker; lane 1: HscB; lane 2: HscB with EDC; lane 3: full-length frataxin; lane 4:

full-length frataxin with EDC; lane 5: HscB and full-length frataxin; lane 6: HscB and

full-length frataxin with EDC; a and b are cross-linking products of HscB and truncated frataxin, respectively. Asterisk (*) presents the cross-linking product between truncated

frataxin and HscB.

174

The interactions between HscB and frataxin were also characterized by isothermal titration calorimetry (ITC), for both the truncated and full-length forms of human frataxin, in the presence or absence of iron. In the absence of iron the truncated frataxin binds to

HscB protein with a stoichiometry n ~ 0.89 ± 0.014, and a dissociation constant (KD) ~

3.26 ± 0.28 µM, while full-length frataxin shows a similar stoichiometry n ~ 0.94 ± 0.050, but a weaker binding affinity (KD ~ 11.9 ± 0.13 µM) relative to the truncated form. In the presence of iron ion, the truncated frataxin binds to HscB protein with a stoichiometry n

~ 1.04 ± 0.019, and a dissociation constant (KD) ~ 2.60 ± 0.29 µM, while full-length frataxin shows a similar stoichiometry n ~ 0.93 ± 0.009, but in this case a higher binding affinity (KD ~ 0.61 ± 0.10 µM) relative to that of the truncated form. ITC data are summarized in Figure 6.4 and Table 6.1.

175

Figure 6.4. Isothermal titration calorimetry (ITC) measurements of 0.32 ~ 0.38 mM HscB

with 0.02 ~ 0.03 mM truncated frataxin (a) or full-length frataxin (b) in the absence of

iron. ITC measurements of 0.32 ~ 0.38 mM HscB with 0.02 ~ 0.03 mM truncated

frataxin (c) or full-length frataxin (d) in the presence of 10-fold molar excess ferrous iron.

For the measurements in the presence of ferrous iron, 5 mM dithionite was used for

preventing oxidation of ferrous ion. Stoichiometry (n) and dissociation constants (KD) were summarized in Table 6.1.

176

177

HscB

Stoichiometry (n) KD (µM)

apo truncated Hs frataxin 0.89 ± 0.014 3.26 ± 0.28 apo full-length Hs frataxin 0.94 ± 0.050 11.9 ± 0.13 holo truncated Hs frataxina 1.04 ± 0.019 2.60 ± 0.29 holo full-length Hs frataxina 0.93 ± 0.009 0.61 ± 0.10

a 10-fold excess ferrous iron and 5 mM dithionite were pre-incubated with frataxin protein under Ar-purged conditions for 20 min.

Table 6.1. Summary of ITC measurements of binding parameters for HscB and frataxin

Ferrous ion

Stoichiometry (n) KD (mM)

truncated Hs frataxin 6.79 ± 0.26 0.46 truncated Hs Frataxin and HscB 5.07 ± 0.12 0.65

Table 6.2. Summary of ITC measurements of binding parameters for iron titration into frataxin with or without HscB

178

To probe the binding site of HscB on frataxin, we looked at the iron titration into solutions containing truncated frataxin alone or truncated frataxin with HscB present.

Ferrous iron titration into truncated frataxin solution yields a stoichiometry of 6.79 ± 0.26,

-1 - with dissociation constant (KD) ~ 0.46 mM, ΔH ~ -1.24 Kcal mol and ΔS ~11.1 cal mol

1 k-1; ferrous iron titration into truncated frataxin in presence of HscB yields a

stoichiometry of 5.06 ± 0.12, with dissociation constant (KD) ~ 0.65 mM, ΔH ~ -2.58

Kcal mol-1 and ΔS ~5.93 cal mol-1 k-1. The data are summarized in Table 6.2.

179

Time (min) Time (min) -15 0 15 30 45 60 75 90 105 120 135 150 -15 0 15 30 45 60 75 90 105 120 135 150

0.0 0

-0.5 -1 -1.0 -2

cal/sec -1.5 cal/sec μ μ

-2.0 -3

-2.50.1 -4 0.0 0.0 -0.1 -0.2 -0.2 -0.3 -0.4 -0.4 -0.6 -0.5 -0.6 -0.8 -0.7 -1.0

kcal/mole of injectant -0.8 kcal/mole injectant of -0.9 -1.2 0 1020304050 0 1020304050 Molar Ratio Molar Ratio

Figure 6.5. Isothermal titration calorimetry (ITC) measurements of 10 mM ferrous ion with 0.1 mM truncated Frataxin with (a) or without 0.1 mM HscB (b) in present. 5 mM dithionite was used for preventing oxidation of ferrous iron. Stoichiometry (n) and dissociation constants (KD) were summarized in Table 6.2. The raw data only shows the

first part of the titration, the second part where the syringe was reloaded with the same

titrant and the signal saturated was not shown here. The integrated heat was fit with

complete data.

180

6.3.3. Metal Binding to HscB

HscB was found to be able to bind to zinc. There appears to be a secondary binding site for zinc. The fitting gives a KD of 0.077 μM for N1 site with a stoichiometry of 0.708 and the KD of N2 site is 7.7 μM with the stoichiometry of 1.01

(Figure 6.6). There are several solvent exposed acidic residues that could coordinate to the zinc ions (Figure 6.7). However, it did not bind to ferrous ion.

181

Time (min) -10 0 10 20 30 40 50 60 70 80 90 100110120130

0.0

-0.5 cal/sec μ -1.0

-2

-4

-6

kcal/mole of injectant -8

0.00.51.01.52.02.53.03.5 Molar Ratio

Figure 6.6. Titrating zinc ions into HscB by ITC. Typically, a 24 µL solution (metal ions) was delivered to a solution of 20 ~ 50 μM HscB in the presence of 2 mM TCEP and 2 mM dithionite (when using ferrous ion) over a period of 20 s with an adequate interval (5

- 10 min) between injections to allow complete equilibration. Titrations continued until 3-

5 equivalents had been added to ensure no further complex formation following addition of excess titrant.

182

6.3.4. Zinc does not Promote Binding between Frataxin and HscB

Since both full-length and truncated frataxin can bind to zinc, and HscB can bind to zinc, as well, we set up experiments to test if zinc can promote the interaction between frataxin and HscB, similar to what we saw in the experiments involving iron. The ITC data did not give significant change, with or without zinc, for both full-length and truncated frataxin.

183

A B

C D

Figure 6.7. Electrostatic potential maps of HscB reviewed from A top, B left, C right, and

D bottom.

184

6.3.5. HscB Accelerates the Maturation of Human Frataxin

We recently characterized (22) the N-terminal cleavage of frataxin and noted an involvement by iron in the cleavage chemistry. Several groups (30, 31) have suggested that HscB chaperone could be involved in the maturation of frataxin. To prove the hypothesis, we set up experiments to define the influence of HscB in the cleavage chemistry of frataxin. Purified full-length frataxin was placed in two vials. EDTA was added to one vial (lane 4, Figure 6.8), but not to the other (lane 3). On an SDS-PAGE gel, the EDTA treated aliquot (lane 4) does not show the cleaved product (~ 14.6 kDa), since

EDTA inhibits the cleavage chemistry as we described before. However, lane 3 shows the cleaved product in the same reaction conditions. When purified HscB protein was added to the tubes containing full-length frataxin, EDTA was unable to inhibit the cleavage chemistry (Figure 6.8). Whether the aliquot contained EDTA (lane 2) or not

(lane 1), the cleaved products (~ 14.6 kDa) were observed to run similar to those in lanes

3 or 4 on an SDS-PAGE gel. To further explore the role of HscB in promoting maturation of frataxin, we set up a series of time-dependent experiments. Three distinct aliquots

(full-length frataxin only, HscB only, and the mixture of both proteins) were placed in several vials and maintained at 4 oC for five days. The cleavage chemistry in all three

aliquots was stopped by addition of SDS loading buffer and finally run an SDS-PAGE

gel (Figure 6.9). Quantitation of the cleaved products (~ 14.6 kDa) was achieved by use

of a Gel-Doc 1000 (Bio-Rad), and the data was quantitated by use of the Multianalysis

software (version 1.1) provided by the manufacturer. In all cases, the background was

subtracted through use of a reference lane in which the proteins were not loaded. Figure

185

6.9 clearly shows that the cleavage of frataxin occurs more rapidly in the presence of

HscB.

The cleavage of N-terminal His-tagged full-length frataxin was further characterized in more detail at 25 ºC (Figure 6.10). The intensity of the full-length frataxin was plotted against time, and fitted with the equation for the first order decay.

Comparing the first (t = 0) and second lane (t = 23.5), the intensity of the protein bends are very different; this could be due to protein precipitation. The total intensity of the proteins bands in the rest of the lanes are similar, therefore, the first lane was omitted.

The rates of cleavage are summarized in Table 6.3. Full-length frataxin alone cleaved at

39.18 hr-1. With EDTA present, it cleaved slower with a rate of 70.15 hr-1. Full-length frataxin and HscB together cleaved at 42.68 hr-1. With the help of iron, it cleaved at a rate

of 15.19 hr-1. If incubating full-length frataxin, HscB and iron together, it cleaved at a rate of 25.77 hr-1.

The cleavage of C-terminal His-tagged full-length frataxin was mixed with

reagents as shown in Table 6.4, and the rate of cleavage was summaries in the same table.

186

kDa M 1 2 3 4 30.0

20.1

14.4

EDTA - + - + HscB + + - -

Figure 6.8. Effects of HscB in the cleavage chemistry of full-length frataxin. Because 1 mM EDTA prevents the cleavage chemistry of full-length frataxin in vitro, 1 mM EDTA was added to both 25 µM full-length frataxin and the mixture of 25 µM full-length frataxin and 25 µM HscB in 50 mM HEPES, pH 7.5 buffer, 100 mM NaCl. Comparing the 1 mM EDTA treated aliquots (lane 4), and the aliquots without EDTA (lane 3),

EDTA was seen to severely reduce the cleavage products of frataxin-only aliquots. But in the presence of HscB (lane 2), the cleavage chemistry was not affected by EDTA.

Reactions were performed at 4 oC for 3 days.

187

30.0 20.1 14.4

30.0 20.1 14.4 kDa M 0 1 2 3 4 5 days

Figure 6.9. Time-dependent cleavage of full-length frataxin in the absence or the presence of HscB. Reactions was held at 4 oC for 5 days, as indicated above. After each reaction, the cleavage chemistry was stopped by adding SDS-loading buffer. The numbers of each lane presented the reaction time (0 day ~ 5 days).

188

Figure 6.10. Time-dependent cleavage of N-terminal His-tagged full-length frataxin under different conditions as noted in the figure. Top, SDS-PAGE gels showing the cleavage of full-length frataxin; bottom, the intensity of full-length frataxin was plotted against time and was fitted with the equation of first order decay (panel A). Reactions were held at 25 oC for 6 days, as indicated above. After each reaction, the cleavage chemistry was stopped by adding SDS-loading buffer.

189

kDa 0 23.5 53.5 79.5 98 118.5 143.5 h 30.0 A) Full-length frataxin: 20.1

14.4

B) Full-length frataxin and 20.1 EDTA: 14.4 30.0

C) Full-length frataxin and 20.1 HscB: 14.4 30.0

D) Full-length frataxin, HscB 20.1 and Fe: 14.4 30.0

E) Full-length frataxin and Fe: 20.1

14.4 30.0 F) HscB: 20.1 14.4

70000

60000

50000

40000

30000 Intensity 20000

10000

20 40 60 80 100 120 140 160 Time (h)

190

kDa M 0 25 44 69 0 25 44 69 M 30.0 20.1 14.4

M 0 25 44 69 0 25 44 69 30.0

20.1

14.4

Figure 6.11. Self-cleavage of C-terminal His-tagged full-length frataxin. To a solution containing 25 µM C-terminal His-tagged full-length frataxin and 2 mM TCEP was added

25 µM ferric ions, or 25 µM HscB or both ferric ion and HscB. The reaction mixture was incubated at 25 ºC for several days and the cleavage process was check periodically by

SDS-PAGE. The degradation was quantitated by gel-doc. The percent cleavage was calculated by dividing the intensity of the full-length frataxin at different time points by the total intensity of the final product and the full-length frataxin that has not been cleaved. The percent cleavage was plotted against time which was fitted with a first- order-decay equation.

191

Disappearance of full- length frataxin (h-1)

Full-length frataxin 0.026 Full-length frataxin and ferrous ion 0.066 Full-length frataxin and EDTA 0.014 Full-length frataxin and HscB 0.023 Full-length frataxin, HscB and ferrous ion 0.039

Table 6.3. Summary of the rates of the cleavage of N-terminal His-tagged full-length frataxin under different conditions.

192

C-terminal His-tagged full- HscB Ferric ion Cleavage Rate (h-1) length frataxin (µM) (µM) (µM)

25 25 25 0.041 25 0 0 0.019 25 0 25 0.024 25 25 0 0.023

Table 6.4. Summary of the rates of the cleavage of C-terminal His-tagged full-length frataxin under different conditions

6.4. Discussion

Several groups (25, 27, 28, 31-35, 38, 39) have suggested that the HscA/HscB chaperone system might be involved in iron-sulfur cluster biosynthesis or delivery.

While several theories have been put forward, the exact function of the chaperone system

in iron-sulfur cluster biosynthesis remains unclear. The possibility that frataxin could be

a substrate of the HscA/HscB chaperone system has also been suggested, but has not

been experimentally addressed.

In this paper, we reported for the first time the cloning and overexpression of the

human HscB chaperone protein and characterized its interaction with human frataxin.

Binding of the two proteins is confirmed by chemical cross-linking and ITC experiments. 193

According to the ITC results, the binding affinities of HscB for truncated or full-length

frataxin were distinguishable. Truncated frataxin (residues 78 ~ 210) showed similar

binding affinities with HscB protein whether in the absence of iron (apo frataxin, KD =

3.26 ± 0.28 µM) or in the presence of iron (holo frataxin, KD = 2.60 ± 0.29 µM). It is an

interesting result compared with the binding affinities of the iron-delivery target proteins

(ISU and ferrochelatase). As we reported, the binding affinities of truncated frataxin with

the iron-delivery target proteins in the absence of iron were too weak to measure by ITC

(22). However, in the presence of iron ion frataxin showed a tight binding affinity with

both ISU (KD = 0.15 µM) (13) and ferrochelatase (KD = 17 nM) (16). These results suggested that iron might promote binding to partner proteins either by stabilizing the structure of frataxin, and/or cross-linking frataxin with ISU or ferrochelatase. However, the interaction of truncated frataxin with HscB does not affect whether it binds iron or not.

Full-length frataxin, however, showed much tighter binding with HscB in the presence of iron (KD = 0.61 ± 0.10 µM) compared to the binding affinity in the absence of iron (KD =

11.9 ± 0.13 µM). Importantly, full-length frataxin interacts with HscB in the absence of

iron, although the affinity is relatively lower than that in the presence of iron. In contrast

to the binding of frataxin to ISU and ferrochelatase, the interaction between frataxin and

HscB chaperone is not promoted by iron ion, presumably because the binding domain of

frataxin for HscB chaperone is different from that for the iron-delivery target proteins

ISU and ferrochelatase. Recently we reported (22) that the N-terminal domain of full-

length frataxin has a secondary structure (an additional α-helix and β-sheet) that forms a

stable contact with, and covers the anionic surface of frataxin. The secondary structure is

194

removed in the presence of iron ion after the iron ions bind to a unique high affinity iron-

binding site in full-length frataxin. Consequently, the anionic surface of frataxin becomes

solvent exposed allowing iron ion to occupy additional iron binding sites that promote

binding and iron delivery to partner proteins. These observations suggest that the affinity difference between full-length frataxin and HscB in the presence or absence of iron might be influenced by the N-terminal secondary structure of full-length frataxin. The HscB recognition domain in full-length frataxin may partially overlap with the domain covered by the N-terminal residues, and so the affinity between full-length frataxin and HscB in the absence of iron is relatively low. In the presence of iron, the secondary structure of

the N-terminal domain in full-length frataxin will change and be released from the

anionic patch. Such a conformation of full-length frataxin might be favorable to

recognition by HscB. After removal of the N-terminal domain (residues 55 ~ 77), and

forming the truncated protein, HscB still binds to frataxin. The ability of HscB to

recognize frataxin does not depend on whether iron occupies the low affinity iron-binding sites or not. The different affinities of HscB to full-length frataixn in the presence of, or absence of iron will be an important characteristic for fully understanding the function of

HscB.

We also found that HscB cannot bind to ferrous ion. This eliminates the possibility that the interaction between HscB and frataxin with ferrous ion present is due to iron bridging. HscB can bind to zinc with a KD of 0.077 and 7.7 μM for two binding

sites, respectively. HscB has several solvent exposed acidic residues which might

contribute to the binding of zinc. A conserved patch of residues in the C-terminal domain

195

of E coli HscB might be of a direct interaction with IscU, as indicated by NMR

(40).Whether this is physiologically relevant is not clear.

A recently published work revealed a tetracysteine metal-binding domain in the

N-terminus of human HscB which is similar to some C-4 Zinc Finger Domains and

rubredoxin, and the identity of the physiological bound metal remained an open question

(41). With a zinc ion bound, the random coiled N-terminal sequence folds to give a

relatively separate domain. The N-terminus of this construct is in fact in front of the cleavage site of the mitochondrial targeting sequence as predicted by TargetP 1.1, which we used. However, another program called the MitoProt II predicted the cleavage site to be at the 34th residue, and so whether this sequence is present in mitochondria might need to be clarified. It has been suggested that proteins of the intermembrane space (IMS) of mitochondria, such as Tim13 may share with the mitochondria matrix proteins a translocation route through the general insertion pore in the TOM complex. The import into mitochondria depends on a metal cofactor bound with four cysteines in the N- terminal sequence which may stabilize folding on the trans-side of the outer membrane

and traps Tim13 in the IMS, and this drives unidirectional movement of the protein

across the outer membrane of mitochondria (42). Whether the cysteine-coordinated metal

cofactor in the N-terminus of HscB is related to iron-sulfur biosynthesis needs further

evidence. The construct used in our work starts right after the cleavage site predicted by

TargetP 1.1.

196

Figure 6.12. Model of HscB mediated cleavage of frataxin. Under the environment of low iron concentration, the interaction between HscB and full-length human frataxin has a dissociation constant of 11.9 µM. We hypothesize that the N-terminal residues of frataxin

may block its anionic surface and therefore HscB cannot tightly bind. In the presence of

iron, a high affinity iron-binding site is occupied by iron, followed by structural change

of N-terminus in frataxin. Unstructured N-terminal residues of frataxin presumably make

better interaction with HscB (KD = 0.61) and HscB cleaved the N-terminal residues to form maturate frataxin for the purpose of iron-delivery to target proteins. It is unclear whether the complex of truncated holo frataxin and HscB (KD = 2.6) directly moves to

HscA (or iron delivery target proteins) or the movement occurs after disassembly of the complex.

197

We designed iron titration experiments to see if the iron binding properties of

frataxin are influenced by HscB. The ferrous iron titration into truncated frataxin gives a

stoichiometry of 6.79 ± 0.26, consistent with what we have reported, and the dissociation

constant (KD) is 0.46 mM. If the iron binding site on frataxin overlaps with the binding

site on HscB, ferrous ion will displace HscB that is already bound to frataxin. In the

experiment with both HscB and frataxin in the sample cell, the iron binding stoichiometry

is 5.07 ± 0.12, with dissociation constant (KD) 0.65 mM. The stoichiometry decreases,

suggesting that there might be an overlap of the binding site. The dissociation constant

did not vary too much, which may be because the iron binding is much weaker than HscB

binding to truncated frataxin, as the result, at the end of the experiment, the concentration

of iron is still not high enough to displace the bind of HscB. There is a slight increase of

Kd. This can be because the iron binding is localized, and eliminating out some acidic

residues does not affect the binding on other acidic residues (Data not shown here). The

binding affinity of iron to frataxin is almost the lowest that ITC can be determined with

accuracy; binding affinity lower then that can not be detected. The ΔH of the experiment

in the presence of HscB is twice that with frataxin alone, indicating more chemistry

occured and it could be the results of electrostatic interaction.

For the cleavage studies at 25 ºC, EDTA lowered the rate of cleavage by two

folds compared with that of full-length frataxin alone. This could be due to that the

removal of the residual iron was not complete. Iron accelerated the rate of cleavage by

2.6 folds consistent with the result at 4 ºC. Without iron, HscB did not enhance the rate of

cleavage. This may be due to the degradation of HscB at 25 ºC, as shown in Figure 6.10.

198

Comparing the rate of cleavage of full-length frataxin assisted by iron and HscB with that assisted by ferrous ion alone, the former was actually at a slower rate.

HscB is not stable at room temperature. The soluble HscB that remained in the solution decreased over time, which might explain why the activity of HscB seemed lower at room temperature; however, we noted that the degradation of HscB was slower when frataxin was added (Figure 6.10, C, D and F). Full-length frataxin may have a stabilizing role. This implies indirectly that an interaction exists between full-length frataxin and HscB with either iron or no iron in present.

The C-terminal His-tagged full-length frataxin by itself cleaved with a rate of

0.0186 h-1. With 25 μM HscB or ferric ion in present, the rate increased to 0.0239 and

0.0226 h-1, respectively. With both HscB and ferric ion in present, the cleavage rate was the fastest, 0.0409 h-1 (Figure 6.11). This could suggest that both HscB and ferric ion facilitate the cleavage of the C-terminal His-tagged full-length frataxin. This construct of full-length frataxin does not have additional sequence at the N-terminus, which eliminate the possible interference as we saw in the N-terminal His-tagged full-length frataxin.

Several workers have suggested that HscB (with HscA) might be involved in the maturation of frataxin. We have recently reported (22) that the N-terminal cleavage of full-length frataxin (maturation of frataxin) is accelerated by iron and inhibited by EDTA.

To further test this hypothesis, we examined the cleavage chemistry in the absence or

presence of EDTA, and/or HscB (Figure 6.8). As we expected, EDTA (lane 4 compare to

lane 3) inhibits the cleavage chemistry in the absence of HscB chaperone. However, the

cleavage chemistry was not influenced by EDTA in the presence of the HscB chaperone

199

(lanes 1 and 2). The time-dependent cleavage chemistry of frataxin in the presence of

HscB chaperone (Figure 6.9) clearly shows that the amount of cleaved products is greater with the HscB chaperone. These results support involvement of the HscB chaperone in the maturation of frataxin, by accelerating the N-terminal cleavage. It is unclear whether

HscB acts like Lon (43), Afg3p and Rca1p (44), the mitochondrial proteins that have both a chaperone activity and a proteolytic activity. We cannot rule out the possible way that

HscB only facilitates or stabilizes the structure of full-length frataxin, which may have effects on the self-cleavage activity. Although the function of the overall chaperone system (HscA and HscB) in iron-sulfur cluster pathway is still unclear, here we show that

HscB interacts with frataxin and is involved in the maturation of frataxin. In addition, our observations suggest that the relationship among frataxin and chaperones would be critical to understanding the exact role of the HscA/HscB chaperone system in promoting iron-sulfur cluster assembly.

6.5. References for Chapter 6

1. Babcock, M., de Silva, D., Oaks, R., Davis-Kaplan, S., Jiralerspong, S., Montermini, L., Pandolfo, M., and Kaplan, J. (1997) Regulation of mitochondrial iron accumulation by Yfh1p, a putative homolog of frataxin, Science 276, 1709- 1712.

2. Campuzano, V., Montermini, L., Lutz, Y., Cova, L., Hindelang, C., Jiralerspong, S., Trottier, Y., Kish, S. J., Faucheux, B., Trouillas, P., Authier, F. J., Durr, A., Mandel, J. L., Vescovi, A., Pandolfo, M., and Koenig, M. (1997) Frataxin is reduced in Friedreich ataxia patients and is associated with mitochondrial membranes., Hum Mol Genet. 6, 1771-1780.

3. Foury, F., and Cazzalini, O. (1997) Deletion of the yeast homologue of the human gene associated with Friedreich's ataxia elicits iron accumulation in mitochondria, FEBS Lett. 411, 373-377. 200

4. Koutnikova, H., Campuzano, V., Foury, F., Dolle, P., Cazzalini, O., and Koenig, M. (1997) Studies of human, mouse and yeast homologues indicate a mitochondrial function for frataxin, Nat Genet. 16, 345-351.

5. Branda, S. S., Yang, Z.-y., Chew, A., and Isaya, G. (1999) Mitochondrial intermediate peptidase and the yeast frataxin homolog together maintain mitochondrial iron homeostasis in Saccharomyces cerevisiae, Hum. Mol. Genet. 8, 1099-1110.

6. Branda, S. S., Cavadini, P., Adamec, J., Kalousek, F., Taroni, F., and Isaya, G. (1999) Yeast and human frataxin are processed to mature form in two sequential steps by the mitochondrial processing peptidase, J Biol Chem. 274, 22763-22769.

7. Gordon, D. M., Shi, Q., Dancis, A., and Pain, D. (1999) Maturation of frataxin within mammalian and yeast mitochondria: one-step processing by matrix processing peptidase, Hum Mol Genet. 8, 2255-2262.

8. Cavadini, P., Adamec, J., Taroni, F., Gakh, O., and Isaya, G. (2000) Two-step processing of human frataxin by mitochondrial processing peptidase. Precursor and intermediate forms are cleaved at different rates, J Biol Chem. 275, 41469- 41475.

9. Dhe-Paganon, S., Shigeta, R., Chi, Y.-I., Ristow, M., and Shoelson, S. E. (2000) Crystal structure of human frataxin, J Biol Chem. 275, 30753-30756.

10. Musco, G., Stier, G., Kolmerer, B., Adinolfi, S., Martin, S., Frenkiel, T., Gibson, T., and Pastore, A. (2000) Towards a structural understanding of Friedreich's ataxia: the solution structure of frataxin, Structure 8, 695-707.

11. Chen, O. S., Hemenway, S., and Kaplan, J. (2002) Inhibition of Fe-S cluster biosynthesis decreases mitochondrial iron export: evidence that Yfh1p affects Fe- S cluster synthesis, Proc. Natl. Acad. Sci.U. S. A. 99, 12321-12326.

13. Yoon, T., and Cowan, J. A. (2003) Iron-sulfur cluster biosynthesis. Characterization of frataxin as an iron donor for assembly of [2Fe-2S] clusters in ISU-type proteins, J. Am. Chem. Soc. 125, 6078-6084.

14. Gerber, J., Muhlenhoff, U., and Lill, R. (2003) An interaction between frataxin and Isu1/Nfs1 that is crucial for Fe/S cluster synthesis on Isu1, EMBO Rep. 4, 906-911.

201

15. Lesuisse, E., Santos, R., Matzanke, B. F., Knight, S. A., Camadro, J. M., and Dancis, A. (2003) Iron use for haeme synthesis is under control of the yeast frataxin homologue (Yfh1). Hum Mol Genet. 12, 879-889.

16. Yoon, T., and Cowan, J. A. (2004) Frataxin-mediated Iron Delivery to Ferrochelatase in the Final Step of Heme Biosynthesis., J Biol Chem. 279, 25943- 25946.

17. He, Y., Alam, S. L., Proteasa, S. V., Zhang, Y., Lesuisse, E., Dancis, A., and Stemmler, T. L. (2004) Yeast frataxin solution structure, iron binding, and ferrochelatase interaction., Biochemistry 43, 16254-16262.

18. Park, S., Gakh, O., O'Neill, H. A., Mangravita, A., Nichol, H., Ferreira, G. C., and Isaya, G. (2003) Yeast frataxin sequentially chaperones and stores iron by coupling protein assembly with iron oxidation, J Biol Chem. 278, 31340-31351.

19. Bulteau, A., O'Neill, H., Kennedy, M., Ikeda-Saito, M., Isaya, G., and Szweda, L. (2004) Frataxin acts as an iron chaperone protein to modulate mitochondrial aconitase activity., Science 305, 242-245.

20. Nair, M., Adinolfi, S., Pastore, C., Kelly, G., Temussi, P., and Pastore, A. (2004) Solution structure of the bacterial frataxin ortholog, CyaY: mapping the iron binding sites., Structure 12, 2037-2048.

21. Cavadini, P., O'Neill, H. A., Benada, O., and Isaya, G. (2002) Assembly and iron- binding properties of human frataxin, the protein deficient in Friedreich ataxia, Hum Mol Genet. 11, 217-227.

22. Yoon, T., and Cowan, J. A. (submitted) N-terminal iron-mediated self-cleavage of human frataxin: Regulation of iron binding and complex formation with target proteins.

23. Takahashi, Y., and Nakamura, M. (1999) Functional assignment of the ORF2- iscS-iscU-iscA-hscB-hscA-fdx-ORF3 gene cluster involved in the assembly of Fe-S clusters in Escherichia coli, J. Biochem. (Tokyo). 126, 917-926.

24. Tokumoto, U., and Takahashi, Y. (2001) Genetic analysis of the isc operon in Escherichia coli involved in the biogenesis of cellular iron-sulfur proteins, J. Biochem. (Tokyo). 130, 63-71.

25. Hoff, K. G., Silberg, J. J., and Vickery, L. E. (2000) Interaction of the iron-sulfur cluster assembly protein IscU with the Hsc66/Hsc20 molecular chaperone system of Escherichia coli, Proc Natl Acad Sci U S A. 97, 7790-7795.

202

26. Silberg, J. J., Hoff, K. G., Tapley, T. L., and Vickery, L. E. (2001) The Fe/S assembly protein IscU behaves as a substrate for the molecular chaperone Hsc66 from Escherichia coli, J Biol Chem. 276, 1696-1700.

27. Hoff, K. G., Ta, D. T., Tapley, T. L., Silberg, J. J., and Vickery, L. E. (2002) Hsc66 substrate specificity is directed toward a discrete region of the iron-sulfur cluster template protein IscU., J Biol Chem. 277, 27353-27359.

28. Hoff, K. G., Cupp-Vickery, J. R., and Vickery, L. E. (2003) Contributions of the LPPVK motif of the iron-sulfur template protein IscU to interactions with the Hsc66-Hsc20 chaperone system, J Biol Chem. 278, 37582-37589.

29. Cupp-Vickery, J. R., Peterson, J. C., Ta, D. T., and Vickery, L. E. (2004) Crystal structure of the molecular chaperone HscA substrate binding domain complexed with the IscU recognition peptide ELPPVKIHC., J. Mol. Biol. 342, 1265-1278.

30. Knight, S. A. B., Sepuri, N. B. V., Pain, D., and Dancis, A. (1998) Mt-Hsp70 homolog, Ssc2p, required for maturation of yeast frataxin and mitochondrial iron homeostasis, J. Biol. Chem. 273, 18389-18393.

31. Voisine, C., Schilke, B., Ohlson, M., Beinert, H., Marszalek, J., and Craig, E. A. (2000) Role of the mitochondrial Hsp70s, Ssc1 and Ssq1, in the maturation of Yfh1., Mol Cell Biol. 20, 3677-3684.

32. Lutz, T., Westermann, B., Neupert, W., and Herrmann, J. M. (2001) The Mitochondrial Proteins Ssq1 and Jac1 are Required for the Assembly of Iron Sulfur Clusters in Mitochondria, J Mole Biol 307, 815-825.

33. Voisine, C., Cheng, Y. C., Ohlson, M., Schilke, B., Hoff, K., Beinert, H., Marszalek, J., and Craig, E. A. (2001) Jac1, a mitochondrial J-type chaperone, is involved in the biogenesis of Fe/S clusters in Saccharomyces cerevisiae., Proc Natl Acad Sci U S A. 98, 1483-1488.

34. Kim, R., Saxena, S., Gordon, D. M., Pain, D., and Dancis, A. (2001) J-domain protein, Jac1p, of yeast mitochondria required for iron homeostasis and activity of Fe-S cluster proteins., J Biol Chem. 276, 17524-17532.

35. Dutkiewicz, R., Schilke, B., Knieszner, H., Walter, W., Craig, E. A., and Marszalek, J. (2003) Ssq1, a mitochondrial Hsp70 involved in iron-sulfur (Fe/S) center biogenesis. Similarities to and differences from its bacterial counterpart, J Biol Chem. 278, 29719-29727.

36. Huynen, M. A., Snel, B., Bork, P., and Gibson, T. J. (2001) The phylogenetic distribution of frataxin indicates a role in iron-sulfur cluster protein assembly, Hum Mol Genet. 10, 2463-2468. 203

37. Sun, G., Gargus, J. J., Ta, D. T., and Vickery, L. E. (2003) Identification of a novel candidate gene in the iron-sulfur pathway implicated in ataxia-susceptibility: human gene encoding HscB, a J-type co-chaperone., J Hum Genet. 48, 415-419.

38. Dutkiewicz, R., Schilke, B., Cheng, S., Knieszner, H., Craig, E. A., and Marszalek, J. (2004) Sequence-specific interaction between mitochondrial Fe-S scaffold protein Isu and Hsp70 Ssq1 is essential for their in vivo function., J Biol Chem. 279, 29167-29174.

39. Knieszner, H., Schilke, B., Dutkiewicz, R., D'Silva, P., Cheng, S., Ohlson, M., Craig, E. A., and Marszalek, J. (2005) Compensation for a defective interaction of the hsp70 ssq1 with the mitochondrial fe-s cluster scaffold isu., J Biol Chem. 280, 28966-28972.

40. Fuzery, A. K., Tonelli, M., Ta, D. T., Cornilescu, G., Vickery, L. E., and Markley, J. L. (2008) Solution structure of the iron-sulfur cluster cochaperone HscB and its binding surface for the iron-sulfur assembly scaffold protein IscU, Biochemistry 47, 9394-9404.

41. Bitto, E., Bingman, C. A., Bittova, L., Kondrashov, D. A., Bannen, R. M., Fox, B. G., Markley, J. L., and Phillips, G. N., Jr. (2008) Structure of human J-type co- chaperone HscB reveals a tetracysteine metal-binding domain, J Biol Chem 283, 30184-30192.

42. Lutz, T., Neupert, W., and Herrmann, J. M. (2003) Import of small Tim proteins into the mitochondrial intermembrane space, EMBO J 22, 4400-4408.

43. Rep, M., van Dijl, J. M., Suda, K., Schatz, G., Grivell, L. A., and Suzuki, C. K. (1996) Promotion of mitochondrial membrane complex assembly by a proteolytically inactive yeast Lon., Science 274, 103-106.

44. Langer, T., and Neupert, W. (1996) Regulated protein degradation in mitochondria, Experientia 52, 1069-1076.

204

BIBLIOGRAPHY

1. Zheng, W., Ren, S., and Graziano, J. H. (1998) Manganese inhibits mitochondrial aconitase: a mechanism of manganese neurotoxicity, Brain Res 799, 334-342.

2. Zheng, L., and Dean, D. R. (1994) Catalytic formation of a nitrogenase iron-sulfur cluster, J Biol Chem 269, 18723-18726.

3. Yuvaniyama, P., Agar, J. N., Cash, V. L., Johnson, M. K., and Dean, D. R. (2000) NifS-directed assembly of a transient [2Fe-2S] cluster within the NifU protein, Proc Natl Acad Sci U S A 97, 599-604.

4. Yoon, T., Dizin, E., and Cowan, J. A. (2007) N-terminal iron-mediated self- cleavage of human frataxin: regulation of iron binding and complex formation with target proteins, J Biol Inorg Chem 12, 535-542.

5. Yoon, T., and Cowan, J. A. (2004) Frataxin-mediated iron delivery to ferrochelatase in the final step of heme biosynthesis, J Biol Chem 279, 25943- 25946.

6. Yoon, T., and Cowan, J. A. (2003) Iron-sulfur cluster biosynthesis. Characterization of frataxin as an iron donor for assembly of [2Fe-2S] clusters in ISU-type proteins, J Am Chem Soc 125, 6078-6084.

7. Wu, S. P., Wu, G., Surerus, K. K., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis. Kinetic analysis of [2Fe-2S] cluster transfer from holo ISU to apo Fd: role of redox chemistry and a conserved aspartate, Biochemistry 41, 8876- 8885.

8. Wu, G., Mansy, S. S., Wu, S. P., Surerus, K. K., Foster, M. W., and Cowan, J. A. (2002) Characterization of an iron-sulfur cluster assembly protein (ISU1) from Schizosaccharomyces pombe, Biochemistry 41, 5024-5032.

9. Warren, S. T. (1996) The expanding world of trinucleotide repeats, Science 271, 1374-1375.

205

10. Wallace, R. W., Tallant, E. A., Dockter, M. E., and Cheung, W. Y. (1982) Calcium binding domains of calmodulin. Sequence of fill as determined with terbium luminescence, J Biol Chem 257, 1845-1854.

11. Walden, W. E., Selezneva, A. I., Dupuy, J., Volbeda, A., Fontecilla-Camps, J. C., Theil, E. C., and Volz, K. (2006) Structure of dual function iron regulatory protein 1 complexed with ferritin IRE-RNA, Science 314, 1903-1908.

12. Volz, K. (2008) The functional duality of iron regulatory protein 1, Curr Opin Struct Biol 18, 106-111.

13. Voisine, C., Schilke, B., Ohlson, M., Beinert, H., Marszalek, J., and Craig, E. A. (2000) Role of the mitochondrial Hsp70s, Ssc1 and Ssq1, in the maturation of Yfh1., Mol Cell Biol. 20, 3677-3684.

14. Voisine, C., Cheng, Y. C., Ohlson, M., Schilke, B., Hoff, K., Beinert, H., Marszalek, J., and Craig, E. A. (2001) Jac1, a mitochondrial J-type chaperone, is involved in the biogenesis of Fe/S clusters in Saccharomyces cerevisiae., Proc Natl Acad Sci U S A. 98, 1483-1488.

15. Vickery, L. E., and Cupp-Vickery, J. R. (2007) Molecular chaperones HscA/Ssq1 and HscB/Jac1 and their roles in iron-sulfur protein maturation, Crit Rev Biochem Mol Biol 42, 95-111.

16. Ventura, N., Rea, S. L., Handerson, S. T., Condo, I., Testi, R., and Johnson, T. E. (2006) C. elegans as a model for Friedreich Ataxia, Faseb J 20, 1029-1030.

17. Urbina, H. D., Silberg, J. J., Hoff, K. G., and Vickery, L. E. (2001) Transfer of sulfur from IscS to IscU during Fe/S cluster assembly, J Biol Chem 276, 44521- 44526.

18. Tokumoto, U., and Takahashi, Y. (2001) Genetic analysis of the isc operon in Escherichia coli involved in the biogenesis of cellular iron-sulfur proteins., J. Biochem. (Tokyo). 130, 63-71.

19. Takahashi, Y., and Nakamura, M. (1999) Functional assignment of the ORF2- iscS-iscU-iscA-hscB-hscA-fdx-ORF3 gene cluster involved in the assembly of Fe-S clusters in Escherichia coli., J. Biochem. (Tokyo). 126, 917-926.

20. Szebenyi, D. M., Obendorf, S. K., and Moffat, K. (1981) Structure of vitamin D- dependent calcium-binding protein from bovine intestine, Nature 294, 327-332.

21. Sundaralingam, M., Bergstrom, R., Strasburg, G., Rao, S. T., Roychowdhury, P., Greaser, M., and Wang, B. C. (1985) Molecular structure of troponin C from chicken skeletal muscle at 3-angstrom resolution, Science 227, 945-948. 206

22. Sun, G., Gargus, J. J., Ta, D. T., and Vickery, L. E. (2003) Identification of a novel candidate gene in the iron-sulfur pathway implicated in ataxia-susceptibility: human gene encoding HscB, a J-type co-chaperone., J Hum Genet. 48, 415-419.

23. Sterrer, S., Konigstorfer, A., and Hoffmann, W. (1990) Biosynthesis and expression of ependymin homologous sequences in zebrafish brain, Neuroscience 37, 277-284.

24. Stehling, O., Elsasser, H. P., Bruckel, B., Muhlenhoff, U., and Lill, R. (2004) Iron-sulfur protein maturation in human cells: evidence for a function of frataxin, Hum Mol Genet 13, 3007-3015.

25. Sreerama, N., and Woody, R. W. (2000) Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set, Anal Biochem 287, 252-260.

26. Snyder, A. P., Sudnick, D. R., Arkle, V. K., and Horrocks, W. D., Jr. (1981) Lanthanide ion luminescence probes. Characterization of metal ion binding sites and intermetal energy transfer distance measurements in calcium-binding proteins. 2. Thermolysin, Biochemistry 20, 3334-3339.

27. Silberg, J. J., Tapley, T. L., Hoff, K. G., and Vickery, L. E. (2004) Regulation of the HscA ATPase reaction cycle by the co-chaperone HscB and the iron-sulfur cluster assembly protein IscU, J Biol Chem 279, 53924-53931.

28. Silberg, J. J., Hoff, K. G., Tapley, T. L., and Vickery, L. E. (2001) The Fe/S assembly protein IscU behaves as a substrate for the molecular chaperone Hsc66 from Escherichia coli., J Biol Chem. 276, 1696-1700.

29. Shimomura, Y., Wada, K., Fukuyama, K., and Takahashi, Y. (2008) The asymmetric trimeric architecture of [2Fe-2S] IscU: implications for its scaffolding during iron-sulfur cluster biosynthesis, J Mol Biol 383, 133-143.

30. Shimomura, Y., Wada, K., Fukuyama, K., and Takahashi, Y. (2008) The Asymmetric Trimeric Architecture of [2Fe-2S] IscU: Implications for Its Scaffolding during Iron-Sulfur Cluster Biosynthesis, J Mol Biol.

31. Shashoua, V. E. (1991) Ependymin, a brain extracellular glycoprotein, and CNS plasticity, Ann N Y Acad Sci 627, 94-114.

32. Shashoua, V. E. (1988) The role of ependymin in the development of long lasting synaptic changes, J Physiol (Paris) 83, 232-239.

207

33. Shashoua, V. E. (1988) Monomeric and polymeric forms of ependymin: a brain extracellular glycoprotein implicated in memory consolidation processes, Neurochem Res 13, 649-655.

34. Shashoua, V. E. (1985) The role of brain extracellular proteins in neuroplasticity and learning, Cell Mol Neurobiol 5, 183-207.

35. Shahravan, S. H., Qu, X., Chan, I. S., and Shin, J. A. (2008) Enhancing the specificity of the enterokinase cleavage reaction to promote efficient cleavage of a fusion tag, Protein Expr Purif 59, 314-319.

36. Schmucker, S., Argentini, M., Carelle-Calmels, N., Martelli, A., and Puccio, H. (2008) The in vivo mitochondrial two-step maturation of human frataxin, Hum Mol Genet 17, 3521-3531.

37. Schmidt, F. R. (2004) Recombinant expression systems in the pharmaceutical industry, Appl Microbiol Biotechnol 65, 363-372.

38. Sazanov, L. A., and Hinchliffe, P. (2006) Structure of the hydrophilic domain of respiratory complex I from Thermus thermophilus, Science 311, 1430-1436.

39. Rouault, T. A., and Tong, W. H. (2008) Iron-sulfur cluster biogenesis and human disease, Trends Genet 24, 398-407.

40. Rouault, T. A., Hentze, M. W., Caughman, S. W., Harford, J. B., and Klausner, R. D. (1988) Binding of a cytosolic protein to the iron-responsive element of human ferritin messenger RNA, Science 241, 1207-1210.

41. Rouault, T. A. (2006) The role of iron regulatory proteins in mammalian iron homeostasis and disease, Nat Chem Biol 2, 406-414.

42. Rotig, A., de Lonlay, P., Chretien, D., Foury, F., Koenig, M., Sidi, D., Munnich, A., and Rustin, P. (1997) Aconitase and mitochondrial iron-sulphur protein deficiency in Friedreich ataxia, Nat Genet 17, 215-217.

43. Robbins, A. H., and Stout, C. D. (1989) Structure of activated aconitase: formation of the [4Fe-4S] cluster in the crystal, Proc Natl Acad Sci U S A 86, 3639-3643.

44. Ristow, M., Pfister, M. F., Yee, A. J., Schubert, M., Michael, L., Zhang, C. Y., Ueki, K., Michael, M. D., 2nd, Lowell, B. B., and Kahn, C. R. (2000) Frataxin activates mitochondrial energy conversion and oxidative phosphorylation, Proc Natl Acad Sci U S A 97, 12239-12243.

208

45. Rhee, M. J., Horrocks, W. D., Jr., and Kosow, D. P. (1982) Laser-induced europium(III) luminescence as a probe of the metal ion mediated association of human prothrombin with phospholipid, Biochemistry 21, 4524-4528.

46. Rep, M., van Dijl, J. M., Suda, K., Schatz, G., Grivell, L. A., and Suzuki, C. K. (1996) Promotion of mitochondrial membrane complex assembly by a proteolytically inactive yeast Lon., Science 274, 103-106.

47. Ramelot, T. A., Cort, J. R., Goldsmith-Fischman, S., Kornhaber, G. J., Xiao, R., Shastry, R., Acton, T. B., Honig, B., Montelione, G. T., and Kennedy, M. A. (2004) Solution NMR structure of the iron-sulfur cluster assembly protein U (IscU) with zinc bound at the active site, J Mol Biol 344, 567-583.

48. Puccio, H., Simon, D., Cossee, M., Criqui-Filipe, P., Tiziano, F., Melki, J., Hindelang, C., Matyas, R., Rustin, P., and Koenig, M. (2001) Mouse models for Friedreich ataxia exhibit cardiomyopathy, sensory nerve defect and Fe-S enzyme deficiency followed by intramitochondrial iron deposits, Nat Genet 27, 181-186.

49. Provencher, S. W., and Glockner, J. (1981) Estimation of globular protein secondary structure from circular dichroism, Biochemistry 20, 33-37.

50. Pondarre, C., Antiochos, B. B., Campagna, D. R., Clarke, S. L., Greer, E. L., Deck, K. M., McDonald, A., Han, A. P., Medlock, A., Kutok, J. L., Anderson, S. A., Eisenstein, R. S., and Fleming, M. D. (2006) The mitochondrial ATP-binding cassette transporter Abcb7 is essential in mice and participates in cytosolic iron- sulfur cluster biogenesis, Hum Mol Genet 15, 953-964.

51. Paulsen, O., and Sejnowski, T. J. (2000) Natural patterns of activity and long-term synaptic plasticity, Curr Opin Neurobiol 10, 172-179.

52. Pastore, C., Franzese, M., Sica, F., Temussi, P., and Pastore, A. (2007) Understanding the binding properties of an unusual metal-binding protein--a study of bacterial frataxin, FEBS J 274, 4199-4210.

53. Park, S., Gakh, O., O'Neill, H. A., Mangravita, A., Nichol, H., Ferreira, G. C., and Isaya, G. (2003) Yeast frataxin sequentially chaperones and stores iron by coupling protein assembly with iron oxidation, J Biol Chem 278, 31340-31351.

54. Nuth, M., Yoon, T., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis: characterization of iron nucleation sites for assembly of the [2Fe-2S]2+ cluster core in IscU proteins, J Am Chem Soc 124, 8774-8775.

55. Naz, R. K. (1997) Prostate: basic and clinical aspects, CRC Press, Boca Raton.

209

56. Nair, M., Adinolfi, S., Pastore, C., Kelly, G., Temussi, P., and Pastore, A. (2004) Solution structure of the bacterial frataxin ortholog, CyaY: mapping the iron binding sites, Structure 12, 2037-2048.

57. Musco, G., Stier, G., Kolmerer, B., Adinolfi, S., Martin, S., Frenkiel, T., Gibson, T., and Pastore, A. (2000) Towards a structural understanding of Friedreich's ataxia: the solution structure of frataxin, Structure 8, 695-707.

58. Murhammer, D. W. (2007) Baculovirus and insect cell expression protocols, 2nd ed., Humana Press, Totowa, N.J.

59. Muhlenhoff, U., Richhardt, N., Ristow, M., Kispal, G., and Lill, R. (2002) The yeast frataxin homolog Yfh1p plays a specific role in the maturation of cellular Fe/S proteins, Hum Mol Genet 11, 2025-2036.

60. Muhlenhoff, U., Richhardt, N., Gerber, J., and Lill, R. (2002) Characterization of iron-sulfur protein assembly in isolated mitochondria. A requirement for ATP, NADH, and reduced iron, J Biol Chem 277, 29810-29816.

61. Muhlenhoff, 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, 4815-4825.

62. Mansy, S. S., Wu, G., Surerus, K. K., and Cowan, J. A. (2002) Iron-sulfur cluster biosynthesis. Thermatoga maritima IscU is a structured iron-sulfur cluster assembly protein, J Biol Chem 277, 21397-21404.

63. Mansy, S. S., and Cowan, J. A. (2004) Iron-sulfur cluster biosynthesis: toward an understanding of cellular machinery and molecular mechanism, Acc Chem Res 37, 719-725.

64. Lutz, T., Westermann, B., Neupert, W., and Herrmann, J. M. (2001) The Mitochondrial Proteins Ssq1 and Jac1 are Required for the Assembly of Iron Sulfur Clusters in Mitochondria, J Mole Biol 307, 815-825.

65. Lutz, T., Neupert, W., and Herrmann, J. M. (2003) Import of small Tim proteins into the mitochondrial intermembrane space, EMBO J 22, 4400-4408.

66. Lukianova, O. A., and David, S. S. (2005) A role for iron-sulfur clusters in DNA repair, Curr Opin Chem Biol 9, 145-151.

67. Liu, Y., and Cowan, J. A. (2007) Iron sulfur cluster biosynthesis. Human NFU mediates sulfide delivery to ISU in the final step of [2Fe-2S] cluster assembly, Chem Commun (Camb), 3192-3194.

210

68. Liu, J., Oganesyan, N., Shin, D. H., Jancarik, J., Yokota, H., Kim, R., and Kim, S. H. (2005) Structural characterization of an iron-sulfur cluster assembly protein IscU in a zinc-bound form, Proteins 59, 875-881.

69. Lill, R., and Muhlenhoff, U. (2008) Maturation of iron-sulfur proteins in eukaryotes: mechanisms, connected processes, and diseases, Annu Rev Biochem 77, 669-700.

70. Lill, R., Dutkiewicz, R., Elsasser, H. P., Hausmann, A., Netz, D. J., Pierik, A. J., Stehling, O., Urzica, E., and Muhlenhoff, U. (2006) Mechanisms of iron-sulfur protein maturation in mitochondria, cytosol and nucleus of eukaryotes, Biochim Biophys Acta 1763, 652-667.

71. Light, A., and Janska, H. (1989) Enterokinase (enteropeptidase): comparative aspects, Trends Biochem Sci 14, 110-112.

72. Lesuisse, E., Santos, R., Matzanke, B. F., Knight, S. A., Camadro, J. M., and Dancis, A. (2003) Iron use for haeme synthesis is under control of the yeast frataxin homologue (Yfh1), Hum Mol Genet 12, 879-889.

73. Lee, M. G., Cho, S. J., Yang, J. K., Song, H. K., and Suh, S. W. (2000) Crystallization and preliminary X-ray crystallographic analysis of Escherichia coli CyaY, a structural homologue of human frataxin, Acta Crystallogr D Biol Crystallogr 56, 920-921.

74. Layer, G., Ollagnier-de Choudens, S., Sanakis, Y., and Fontecave, M. (2006) Iron-sulfur cluster biosynthesis: characterization of Escherichia coli CYaY as an iron donor for the assembly of [2Fe-2S] clusters in the scaffold IscU, J Biol Chem 281, 16256-16263.

75. LaVallie, E. R., DiBlasio, E. A., Kovacic, S., Grant, K. L., Schendel, P. F., and McCoy, J. M. (1993) A thioredoxin gene fusion expression system that circumvents inclusion body formation in the E. coli cytoplasm, Biotechnology (N Y) 11, 187-193.

76. Langer, T., and Neupert, W. (1996) Regulated protein degradation in mitochondria., Experientia 52, 1069-1076.

77. Koutnikova, H., Campuzano, V., Foury, F., Dolle, P., Cazzalini, O., and Koenig, M. (1997) Studies of human, mouse and yeast homologues indicate a mitochondrial function for frataxin., Nat Genet. 16, 345-351.

78. Kounosu, A., Li, Z., Cosper, N. J., Shokes, J. E., Scott, R. A., Imai, T., Urushiyama, A., and Iwasaki, T. (2004) Engineering a three-cysteine, one-

211

histidine ligand environment into a new hyperthermophilic archaeal Rieske-type [2Fe-2S] ferredoxin from Sulfolobus solfataricus, J Biol Chem 279, 12519-12528.

79. Konigstorfer, A., Sterrer, S., and Hoffmann, W. (1989) Biosynthesis of ependymins from goldfish brain, J Biol Chem 264, 13689-13692.

80. Koeller, D. M., Casey, J. L., Hentze, M. W., Gerhardt, E. M., Chan, L. N., Klausner, R. D., and Harford, J. B. (1989) A cytosolic protein binds to structural elements within the iron regulatory region of the transferrin receptor mRNA, Proc Natl Acad Sci U S A 86, 3574-3578.

81. Knight, S. A. B., Sepuri, N. B. V., Pain, D., and Dancis, A. (1998) Mt-Hsp70 homolog, Ssc2p, required for maturation of yeast frataxin and mitochondrial iron homeostasis, J. Biol. Chem. 273, 18389-18393.

82. Knieszner, H., Schilke, B., Dutkiewicz, R., D'Silva, P., Cheng, S., Ohlson, M., Craig, E. A., and Marszalek, J. (2005) Compensation for a defective interaction of the hsp70 ssq1 with the mitochondrial fe-s cluster scaffold isu., J Biol Chem. 280, 28966-28972.

83. Kim, R., Saxena, S., Gordon, D. M., Pain, D., and Dancis, A. (2001) J-domain protein, Jac1p, of yeast mitochondria required for iron homeostasis and activity of Fe-S cluster proteins., J Biol Chem. 276, 17524-17532.

84. Khoroshilova, N., Beinert, H., and Kiley, P. J. (1995) Association of a polynuclear iron-sulfur center with a mutant FNR protein enhances DNA binding, Proc Natl Acad Sci U S A 92, 2499-2503.

85. Kennedy, M. C., Emptage, M. H., Dreyer, J. L., and Beinert, H. (1983) The role of iron in the activation-inactivation of aconitase, J Biol Chem 258, 11098-11105.

86. Kato, S., Mihara, H., Kurihara, T., Takahashi, Y., Tokumoto, U., Yoshimura, T., and Esaki, N. (2002) Cys-328 of IscS and Cys-63 of IscU are the sites of disulfide bridge formation in a covalently bound IscS/IscU complex: implications for the mechanism of iron-sulfur cluster assembly, Proc Natl Acad Sci U S A 99, 5948- 5952.

87. Kaiser, J. T., Clausen, T., Bourenkow, G. P., Bartunik, H. D., Steinbacher, S., and Huber, R. (2000) Crystal structure of a NifS-like protein from Thermotoga maritima: Implications for iron sulphur cluster assembly, Journal of Molecular Biology 297, 451-464.

88. Johnson, D. C., Unciuleac, M. C., and Dean, D. R. (2006) Controlled expression and functional analysis of iron-sulfur cluster biosynthetic components within Azotobacter vinelandii, J Bacteriol 188, 7551-7561. 212

89. Johnson, D. C., Dean, D. R., Smith, A. D., and Johnson, M. K. (2005) Structure, function, and formation of biological iron-sulfur clusters, Annu Rev Biochem 74, 247-281.

90. Iwata, S., Saynovits, M., Link, T. A., and Michel, H. (1996) Structure of a water soluble fragment of the 'Rieske' iron-sulfur protein of the bovine heart mitochondrial cytochrome bc1 complex determined by MAD phasing at 1.5 A resolution, Structure 4, 567-579.

91. Ishikawa, T., Mizunoe, Y., Kawabata, S., Takade, A., Harada, M., Wai, S. N., and Yoshida, S. (2003) The iron-binding protein Dps confers hydrogen peroxide stress resistance to Campylobacter jejuni, Journal of Bacteriology 185, 1010-1017.

92. Hwang, D. M., Dempsey, A., Tan, K. T., and Liew, C. C. (1996) A modular domain of NifU, a nitrogen fixation cluster protein, is highly conserved in evolution, J Mol Evol 43, 536-540.

93. Huynen, M. A., Snel, B., Bork, P., and Gibson, T. J. (2001) The phylogenetic distribution of frataxin indicates a role in iron-sulfur cluster protein assembly, Hum Mol Genet 10, 2463-2468.

94. Hurley, J. K., Weber-Main, A. M., Hodges, A. E., Stankovich, M. T., Benning, M. M., Holden, H. M., Cheng, H., Xia, B., Markley, J. L., Genzor, C., Gomez- Moreno, C., Hafezi, R., and Tollin, G. (1997) Iron-sulfur cluster cysteine-to- serine mutants of Anabaena -2Fe-2S- ferredoxin exhibit unexpected redox properties and are competent in electron transfer to ferredoxin:NADP+ reductase, Biochemistry 36, 15109-15117.

95. Huang, J., Dizin, E., and Cowan, J. A. (2008) Mapping iron binding sites on human frataxin: implications for cluster assembly on the ISUFe-S cluster scaffold protein, J. Biol. Inorg. Chem. 13, 825-836.

96. Hoff, K. G., Ta, D. T., Tapley, T. L., Silberg, J. J., and Vickery, L. E. (2002) Hsc66 substrate specificity is directed toward a discrete region of the iron-sulfur cluster template protein IscU., J Biol Chem. 277, 27353-27359.

97. Hoff, K. G., Silberg, J. J., and Vickery, L. E. (2000) Interaction of the iron-sulfur cluster assembly protein IscU with the Hsc66/Hsc20 molecular chaperone system of Escherichia coli, Proc Natl Acad Sci U S A 97, 7790-7795.

98. Hoff, K. G., Cupp-Vickery, J. R., and Vickery, L. E. (2003) Contributions of the LPPVK motif of the iron-sulfur template protein IscU to interactions with the Hsc66-Hsc20 chaperone system, J Biol Chem 278, 37582-37589.

213

99. Hidalgo, E., and Demple, B. (1996) Activation of SoxR-dependent transcription in vitro by noncatalytic or NifS-mediated assembly of [2Fe-2S] clusters into apo- SoxR, J Biol Chem 271, 7269-7272.

100. Hebb, D. O. (2002) The organization of behavior: a neuropsychological theory, L. Erlbaum Associates, Mahwah, N.J.

101. He, Y., Alam, S. L., Proteasa, S. V., Zhang, Y., Lesuisse, E., Dancis, A., and Stemmler, T. L. (2004) Yeast frataxin solution structure, iron binding, and ferrochelatase interaction, Biochemistry 43, 16254-16262.

102. Harrison, R. L., and Jarvis, D. L. (2006) Protein N-glycosylation in the baculovirus-insect cell expression system and engineering of insect cells to produce "mammalianized" recombinant glycoproteins, Adv Virus Res 68, 159-191.

103. Gruer, M. J., Artymiuk, P. J., and Guest, J. R. (1997) The aconitase family: three structural variations on a common theme, Trends Biochem Sci 22, 3-6.

104. Grinna, L. S., and Tschopp, J. F. (1989) Size distribution and general structural features of N-linked oligosaccharides from the methylotrophic yeast, Pichia pastoris, Yeast 5, 107-115.

105. Gray, N. K., Pantopoulos, K., Dandekar, T., Ackrell, B. A., and Hentze, M. W. (1996) Translational regulation of mammalian and Drosophila citric acid cycle enzymes via iron-responsive elements, Proc Natl Acad Sci U S A 93, 4925-4930.

106. Gordon, D. M., Shi, Q., Dancis, A., and Pain, D. (1999) Maturation of frataxin within mammalian and yeast mitochondria: one-step processing by matrix processing peptidase, Hum Mol Genet. 8, 2255-2262.

107. Gerber, J., Muhlenhoff, U., and Lill, R. (2003) An interaction between frataxin and Isu1/Nfs1 that is crucial for Fe/S cluster synthesis on Isu1, EMBO Rep 4, 906- 911.

108. Garland, S. A., Hoff, K., Vickery, L. E., and Culotta, V. C. (1999) Saccharomyces cerevisiae ISU1 and ISU2: members of a well-conserved gene family for iron- sulfur cluster assembly, J Mol Biol 294, 897-907.

109. Gardner, P. R. (2002) Aconitase: sensitive target and measure of superoxide, Methods Enzymol 349, 9-23.

110. Ganss, B., and Hoffmann, W. (1993) Calcium binding to sialic acids and its effect on the conformation of ependymins, Eur J Biochem 217, 275-280.

214

111. Gakh, O., Park, S., Liu, G., Macomber, L., Imlay, J. A., Ferreira, G. C., and Isaya, G. (2006) Mitochondrial iron detoxification is a primary function of frataxin that limits oxidative damage and preserves cell longevity, Hum Mol Genet 15, 467- 479.

112. Gaiarsa, J. L., Caillard, O., and Ben-Ari, Y. (2002) Long-term plasticity at GABAergic and glycinergic synapses: mechanisms and functional significance, Trends Neurosci 25, 564-570.

113. Fuzery, A. K., Tonelli, M., Ta, D. T., Cornilescu, G., Vickery, L. E., and Markley, J. L. (2008) Solution structure of the iron-sulfur cluster cochaperone HscB and its binding surface for the iron-sulfur assembly scaffold protein IscU, Biochemistry 47, 9394-9404.

114. Foury, F., Pastore, A., and Trincal, M. (2007) Acidic residues of yeast frataxin have an essential role in Fe-S cluster assembly, EMBO Rep 8, 194-199.

115. Foury, F., and Cazzalini, O. (1997) Deletion of the yeast homologue of the human gene associated with Friedreich's ataxia elicits iron accumulation in mitochondria, FEBS Lett 411, 373-377.

116. Foury, F. (1999) Low iron concentration and aconitase deficiency in a yeast frataxin homologue deficient strain, FEBS Lett 456, 281-284.

117. Foster, M. W., Mansy, S. S., Hwang, J., Penner-Hahn, J. E., Surerus, K. K., and Cowan, J. A. (2000) A mutant human IscU protein contains a stable [2Fe-2S](2+) center of possible functional significance, Journal of the American Chemical Society 122, 6805-6806.

118. Fleischmann, R. D., Adams, M. D., White, O., Clayton, R. A., Kirkness, E. F., Kerlavage, A. R., Bult, C. J., Tomb, J. F., Dougherty, B. A., Merrick, J. M., and et al. (1995) Whole-genome random sequencing and assembly of Haemophilus influenzae Rd, Science 269, 496-512.

119. Fishburn, J., Mohibullah, N., and Hahn, S. (2005) Function of a eukaryotic transcription activator during the transcription cycle, Mol Cell 18, 369-378.

120. Fessard, A., Delafresnaye, J. F., and Council for International Organizations of Medical Sciences. (1961) Brain mechanisms and learning: a symposium, Blackwell Scientific Publications, London.

121. Dutkiewicz, R., Schilke, B., Knieszner, H., Walter, W., Craig, E. A., and Marszalek, J. (2003) Ssq1, a mitochondrial Hsp70 involved in iron-sulfur (Fe/S) center biogenesis. Similarities to and differences from its bacterial counterpart, J Biol Chem. 278, 29719-29727. 215

122. Dutkiewicz, R., Schilke, B., Cheng, S., Knieszner, H., Craig, E. A., and Marszalek, J. (2004) Sequence-specific interaction between mitochondrial Fe-S scaffold protein Isu and Hsp70 Ssq1 is essential for their in vivo function., J Biol Chem. 279, 29167-29174.

123. Durr, A., Cossee, M., Agid, Y., Campuzano, V., Mignard, C., Penet, C., Mandel, J. L., Brice, A., and Koenig, M. (1996) Clinical and genetic abnormalities in patients with Friedreich's ataxia, N Engl J Med 335, 1169-1175.

124. Dupuy, J., Volbeda, A., Carpentier, P., Darnault, C., Moulis, J. M., and Fontecilla-Camps, J. C. (2006) Crystal structure of human iron regulatory protein 1 as cytosolic aconitase, Structure 14, 129-139.

125. Ding, H. G., and Clark, R. J. (2004) Characterization of iron binding in IscA, an ancient iron-sulphur cluster assembly protein, Biochem J 379, 433-440.

126. Dhe-Paganon, S., Shigeta, R., Chi, Y. I., Ristow, M., and Shoelson, S. E. (2000) Crystal structure of human frataxin, J Biol Chem 275, 30753-30756.

127. Dean, D. R., Bolin, J. T., and Zheng, L. (1993) Nitrogenase metalloclusters: structures, organization, and synthesis, J Bacteriol 175, 6737-6744.

128. Dandekar, T., Stripecke, R., Gray, N. K., Goossen, B., Constable, A., Johansson, H. E., and Hentze, M. W. (1991) Identification of a novel iron-responsive element in murine and human erythroid delta-aminolevulinic acid synthase mRNA, Embo J 10, 1903-1909.

129. Cupp-Vickery, J. R., and Vickery, L. E. (2000) Crystal structure of Hsc20, a J- type Co-chaperone from Escherichia coli, J Mol Biol 304, 835-845.

130. Cupp-Vickery, J. R., Peterson, J. C., Ta, D. T., and Vickery, L. E. (2004) Crystal structure of the molecular chaperone HscA substrate binding domain complexed with the IscU recognition peptide ELPPVKIHC, J Mol Biol 342, 1265-1278.

131. Cregg, J. M., Vedvick, T. S., and Raschke, W. C. (1993) Recent advances in the expression of foreign genes in Pichia pastoris, Biotechnology (N Y) 11, 905-910.

132. Cregg, J. M., Cereghino, J. L., Shi, J., and Higgins, D. R. (2000) Recombinant protein expression in Pichia pastoris, Mol Biotechnol 16, 23-52.

133. Cook, J. D., Bencze, K. Z., Jankovic, A. D., Crater, A. K., Busch, C. N., Bradley, P. B., Stemmler, A. J., Spaller, M. R., and Stemmler, T. L. (2006) Monomeric yeast frataxin is an iron-binding protein, Biochemistry 45, 7767-7777.

216

134. Condo, I., Ventura, N., Malisan, F., Tomassini, B., and Testi, R. (2006) A pool of extramitochondrial frataxin that promotes cell survival, J Biol Chem 281, 16750- 16756.

135. Condo, I., Ventura, N., Malisan, F., Rufini, A., Tomassini, B., and Testi, R. (2007) In vivo maturation of human frataxin, Hum Mol Genet 16, 1534-1540.

136. Ciurli, S., and Musiani, F. (2005) High potential iron-sulfur proteins and their role as soluble electron carriers in bacterial photosynthesis: tale of a discovery, Photosynth Res 85, 115-131.

137. Cho, S. J., Lee, M. G., Yang, J. K., Lee, J. Y., Song, H. K., and Suh, S. W. (2000) Crystal structure of Escherichia coli CyaY protein reveals a previously unidentified fold for the evolutionarily conserved frataxin family, Proc Natl Acad Sci U S A 97, 8932-8937.

138. Chen, O. S., Hemenway, S., and Kaplan, J. (2002) Inhibition of Fe-S cluster biosynthesis decreases mitochondrial iron export: evidence that Yfh1p affects Fe- S cluster synthesis, Proc. Natl. Acad. Sci.U. S. A. 99, 12321-12326.

139. Cavadini, P., O'Neill, H. A., Benada, O., and Isaya, G. (2002) Assembly and iron- binding properties of human frataxin, the protein deficient in Friedreich ataxia, Hum Mol Genet 11, 217-227.

140. Cavadini, P., Gellera, C., Patel, P. I., and Isaya, G. (2000) Human frataxin maintains mitochondrial iron homeostasis in Saccharomyces cerevisiae, Hum Mol Genet 9, 2523-2530.

141. Cavadini, P., Adamec, J., Taroni, F., Gakh, O., and Isaya, G. (2000) Two-step processing of human frataxin by mitochondrial processing peptidase. Precursor and intermediate forms are cleaved at different rates, J Biol Chem 275, 41469- 41475.

142. Casey, J. L., Hentze, M. W., Koeller, D. M., Caughman, S. W., Rouault, T. A., Klausner, R. D., and Harford, J. B. (1988) Iron-responsive elements: regulatory RNA sequences that control mRNA levels and translation, Science 240, 924-928.

143. Campuzano, V., Montermini, L., Molto, M. D., Pianese, L., Cossee, M., Cavalcanti, F., Monros, E., Rodius, F., Duclos, F., Monticelli, A., Zara, F., Canizares, J., Koutnikova, H., Bidichandani, S. I., Gellera, C., Brice, A., Trouillas, P., De Michele, G., Filla, A., De Frutos, R., Palau, F., Patel, P. I., Di Donato, S., Mandel, J. L., Cocozza, S., Koenig, M., and Pandolfo, M. (1996) Friedreich's ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion, Science 271, 1423-1427.

217

144. Campuzano, V., Montermini, L., Lutz, Y., Cova, L., Hindelang, C., Jiralerspong, S., Trottier, Y., Kish, S. J., Faucheux, B., Trouillas, P., Authier, F. J., Durr, A., Mandel, J. L., Vescovi, A., Pandolfo, M., and Koenig, M. (1997) Frataxin is reduced in Friedreich ataxia patients and is associated with mitochondrial membranes., Hum Mol Genet. 6, 1771-1780.

145. Bulteau, A. L., O'Neill, H. A., Kennedy, M. C., Ikeda-Saito, M., Isaya, G., and Szweda, L. I. (2004) Frataxin acts as an iron chaperone protein to modulate mitochondrial aconitase activity, Science 305, 242-245.

146. Brazzolotto, X., Gaillard, J., Pantopoulos, K., Hentze, M. W., and Moulis, J. M. (1999) Human cytoplasmic aconitase (Iron regulatory protein 1) is converted into its [3Fe-4S] form by hydrogen peroxide in vitro but is not activated for iron- responsive element binding, J Biol Chem 274, 21625-21630.

147. Branda, S. S., Yang, Z.-y., Chew, A., and Isaya, G. (1999) Mitochondrial intermediate peptidase and the yeast frataxin homolog together maintain mitochondrial iron homeostasis in Saccharomyces cerevisiae, Hum. Mol. Genet. 8, 1099-1110.

148. Branda, S. S., Cavadini, P., Adamec, J., Kalousek, F., Taroni, F., and Isaya, G. (1999) Yeast and human frataxin are processed to mature form in two sequential steps by the mitochondrial processing peptidase, J Biol Chem. 274, 22763-22769.

149. Bou-Abdallah, F., Adinolfi, S., Pastore, A., Laue, T. M., and Dennis Chasteen, N. (2004) Iron binding and oxidation kinetics in frataxin CyaY of Escherichia coli, J Mol Biol 341, 605-615.

150. Blattner, F. R., Plunkett, G., 3rd, Bloch, C. A., Perna, N. T., Burland, V., Riley, M., Collado-Vides, J., Glasner, J. D., Rode, C. K., Mayhew, G. F., Gregor, J., Davis, N. W., Kirkpatrick, H. A., Goeden, M. A., Rose, D. J., Mau, B., and Shao, Y. (1997) The complete genome sequence of Escherichia coli K-12, Science 277, 1453-1474.

151. Bitto, E., Bingman, C. A., Bittova, L., Kondrashov, D. A., Bannen, R. M., Fox, B. G., Markley, J. L., and Phillips, G. N., Jr. (2008) Structure of human J-type co- chaperone HscB reveals a tetracysteine metal-binding domain, J Biol Chem 283, 30184-30192.

152. Bencze, K. Z., Yoon, T., Millan-Pacheco, C., Bradley, P. B., Pastor, N., Cowan, J. A., and Stemmler, T. L. (2007) Human frataxin: iron and ferrochelatase binding surface, Chem Commun (Camb), 1798-1800.

218

153. Bencze, K. Z., Kondapalli, K. C., Cook, J. D., McMahon, S., Millan-Pacheco, C., Pastor, N., and Stemmler, T. L. (2006) The structure and function of frataxin, Crit Rev Biochem Mol Biol 41, 269-291.

154. Beinert, H., Kennedy, M. C., and Stout, C. D. (1996) Aconitase as Ironminus signSulfur Protein, Enzyme, and Iron-Regulatory Protein, Chem Rev 96, 2335- 2374.

155. Baker, N. A., Sept, D., Joseph, S., Holst, M. J., and McCammon, J. A. (2001) Electrostatics of nanosystems: application to microtubules and the ribosome, Proc Natl Acad Sci U S A 98, 10037-10041.

156. Babcock, M., de Silva, D., Oaks, R., Davis-Kaplan, S., Jiralerspong, S., Montermini, L., Pandolfo, M., and Kaplan, J. (1997) Regulation of mitochondrial iron accumulation by Yfh1p, a putative homolog of frataxin, Science 276, 1709- 1712.

157. 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, 1096-1101.

158. Aisen, P., Enns, C., and Wessling-Resnick, M. (2001) Chemistry and biology of eukaryotic iron metabolism, Int J Biochem Cell Biol 33, 940-959.

159. Agar, J. N., Krebs, C., Frazzon, J., Huynh, B. H., Dean, D. R., and Johnson, M. K. (2000) IscU as a scaffold for iron-sulfur cluster biosynthesis: sequential assembly of [2Fe-2S] and [4Fe-4S] clusters in IscU, Biochemistry 39, 7856-7862.

160. Adams, D. S., Hasson, B., Boyer-Boiteau, A., El-Khishin, A., and Shashoua, V. E. (2003) A peptide fragment of ependymin neurotrophic factor uses protein kinase C and the mitogen-activated protein kinase pathway to activate c-Jun N-terminal kinase and a functional AP-1 containing c-Jun and c-Fos proteins in mouse NB2a cells, J Neurosci Res 72, 405-416.

161. Acquaviva, F., De Biase, I., Nezi, L., Ruggiero, G., Tatangelo, F., Pisano, C., Monticelli, A., Garbi, C., Acquaviva, A. M., and Cocozza, S. (2005) Extra- mitochondrial localisation of frataxin and its association with IscU1 during enterocyte-like differentiation of the human colon adenocarcinoma cell line Caco- 2, J Cell Sci 118, 3917-3924.

219