Sei sulla pagina 1di 358

SERIES EDITOR

D. ROLLINSON

Life Sciences Department The Natural History Museum, London, UK d.rollinson@nhm.ac.uk

EDITORIAL BOARD

M. G. BASÁÑEZ

Professor in Parasite Epidemiology, Department of Infectious Disease Epidemiology Faculty of Medicine (St Mary s Campus), Imperial College, London, London, UK

S. BROOKER

Wellcome Trust Research Fellow and Professor, London School of Hygiene and Tropical Medicine, Faculty of Infectious and Tropical, Diseases, London, UK

R. B. GASSER

Department of Veterinary Science, The University of Melbourne, Parkville, Victoria, Australia

N. HALL

School of Biological Sciences, Biosciences Building, University of Liverpool, Liverpool, UK

R. C. OLIVEIRA

Centro de Pesquisas Rene Rachou/ CPqRR - A FIOCRUZ em Minas Gerais, Rene Rachou Research Center/CPqRR - The Oswaldo Cruz Foundation in the State of Minas Gerais-Brazil, Brazil

R. E. SINDEN

Immunology and Infection Section, Department of Biological Sciences, Sir Alexander Fleming Building, Imperial College of Science, Technology and Medicine, London, UK

D. L. SMITH

Johns Hopkins Malaria Research Institute & Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA

R. C. A. THOMPSON

Head, WHO Collaborating Centre for the Molecular Epidemiology of Parasitic Infections, Principal Investigator, Environmental Biotechnology CRC (EBCRC), School of Veterinary and Biomedical Sciences, Murdoch University, Murdoch, WA, Australia

X. N. ZHOU

Professor, Director, National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention, Shanghai, People s Republic of China

Academic Press is an imprint of Elsevier

32 Jamestown Road, London, NW1 7BY, UK

525

B Street, Suite 1800, San Diego, CA 92101-4495, USA

225

Wyman Street, Waltham, MA 02451, USA

Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands

First edition 2013

Copyright © 2013 Elsevier Ltd. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher.

Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (þ 44) (0) 1865 843830; fax (þ44) (0) 1865 853333; email: permissions@elsevier.com. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/ permissions, and selecting Obtaining permission to use Elsevier material.

Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas con- tained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.

ISBN: 978-0-12-407705-8 ISSN: 0065-308X

For information on all Academic Press publications visit our website at store.elsevier.com

Printed and bound in UK

13 14 15 16 12 11 10 9 8 7 6
13
14 15 16
12
11
10
9
8
7
6

5

4

3

2

1

CONTRIBUTORS

Vahab Ali Department of Biochemistry, Rajendra Memorial Research Institute of Medical Sciences, Indian Council of Medical Research, Agam-kuan, Patna, India

J. Kevin Baird Eijkman-Oxford Clinical Research Unit, Jakarta, Indonesia, and Centre for Tropical Medicine, Nuffield Department of Medicine, University of Oxford, Oxford, United Kingdom

Michael J. Bangs Public Health and Malaria Control Department, International SOS, PT Freeport Indonesia, Kuala Kencana, Indonesia

John R. Barta Department of Pathobiology, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada

Damer Blake Royal Veterinary College, Hatfield, United Kingdom

H. David Chapman Department of Poultry Science, University of Arkansas, Fayetteville, Arkansas, United States

Iqbal R.F. Elyazar Eijkman-Oxford Clinical Research Unit, Jakarta, Indonesia

Robin B. Gasser Faculty of Veterinary Science, The University of Melbourne, Parkville, Victoria, Australia

Peter W. Gething Spatial Ecology and Epidemiology Group, Department of Zoology, University of Oxford, Oxford, United Kingdom

Arthur Gruber Department of Parasitology, Institute of Biomedical Sciences, University of Sa˜ o Paulo, Sa˜ o Paulo, Brazil

Simon I. Hay Spatial Ecology and Epidemiology Group, Department of Zoology, University of Oxford, Oxford, United Kingdom

Mark Jenkins Animal Parasitic Diseases Laboratory, Agricultural Research Service, USDA, Beltsville, Maryland, United States

Aaron R. Jex Faculty of Veterinary Science, The University of Melbourne, Parkville, Victoria, Australia

ix

x

Contributors

Rita Kusriastuti Directorate of Vector-Borne Diseases, Indonesian Ministry of Health, Jakarta, Indonesia

Tomoyoshi Nozaki Department of Parasitology, National Institute of Infectious Diseases, Tokyo, and Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Japan

Florian Roeber Faculty of Veterinary Science, The University of Melbourne, Parkville, Victoria, Australia

Marianne E. Sinka Spatial Ecology and Epidemiology Group, Department of Zoology, University of Oxford, Oxford, United Kingdom

Nicholas C. Smith Queensland Tropical Health Alliance Laboratory, Faculty of Medicine, Health and Molecular Sciences, James Cook University, Cairns, Queensland, Australia

Xun Suo National Animal Protozoa Laboratory & College of Veterinary Medicine, China Agricultural University, Beijing, China

Asik Surya Directorate of Vector-Borne Diseases, Indonesian Ministry of Health, Jakarta, Indonesia

Siti N. Tarmidzi Directorate of Vector-Borne Diseases, Indonesian Ministry of Health, Jakarta, Indonesia

Fiona M. Tomley Royal Veterinary College, Hatfield, United Kingdom

Winarno Directorate of Vector-Borne Diseases, Indonesian Ministry of Health, Jakarta, Indonesia

CHAPTER ONE Iron – Sulphur Clusters, Their Biosynthesis, and Biological Functions in Protozoan Parasites Vahab

CHAPTER ONE

Iron Sulphur Clusters, Their Biosynthesis, and Biological Functions in Protozoan Parasites

Vahab Ali * , Tomoyoshi Nozaki , { ,1

* Department of Biochemistry, Rajendra Memorial Research Institute of Medical Sciences, Indian Council of Medical Research, Agam-kuan, Patna, India Department of Parasitology, National Institute of Infectious Diseases, Tokyo, Japan

{ Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Japan 1 Corresponding author: e-mail address: nozaki@nih.go.jp

Contents

1. Introduction

2. Variation and Features of FeS Clusters

2.1 Discovery of FeS clusters

2.2 Heterogeneity of FeS clusters

2.3 Physicochemical features and analytical methods of FeS clusters

2.4 Biochemical features of Fe S clusters

3

4

4

4

10

11

3. General Biological and Physiological Roles of FeS Proteins in Prokaryotes

and Eukaryotes

3.1 Roles of FeS proteins in bacteria

3.2 Roles of FeS proteins in eukaryotes

4. General Source of Iron and Sulphur for Fe S Cluster Biosynthesis

4.1 Acquisition and transport of iron

4.2 Acquisition, biosynthesis, and transport of L-cysteine

5. Four Systems for Fe S Cluster Biogenesis in Prokaryotes and Eukaryotes

5.1 Catalytic reaction in Fe S cluster biosynthesis

5.2

5.3

5.4

5.5

5.6 Mechanism of repair of Fe S clusters

ISC machinery

SUF machinery

NIF machinery

CIA machinery

6. Genetic Disorders by a Defect of Fe S Cluster Biogenesis

6.1 Friedreich s ataxia

6.2 Sideroblastic anaemia

6.3 X-linked sideroblastic anaemia and ataxia (XLSA/A)

6.4 Other genetic disorders

Advances in Parasitology , Volume 83 ISSN 0065-308X

# 2013 Elsevier Ltd All rights reserved.

12

12

15

17

17

19

20

21

21

29

32

34

38

38

39

40

40

41

1

2

Vahab Ali and Tomoyoshi Nozaki

7.

Outline of Conservation, Unique Distribution, and Diversity of Fe S Cluster Biogenesis Machineries in Protozoan Parasites

41

8.

Fe S Cluster Biogenesis in Protozoan Parasites

42

8.1 Entamoeba

42

8.2 Giardia

47

8.3 Trichomonas

50

8.4 Leishmania

52

8.5 Trypanosoma

55

8.6 Plasmodium

57

8.7 Cryptosporidia

60

8.8 Microsporidia

61

8.9 Blastocystis

63

9.

Regulation of FeS Protein Biosynthesis under Stress Conditions

65

9.1 Factors that affect FeS clusters

65

9.2 Stress-dependent regulation of ISC and SUF systems

66

9.3 Stress-dependent response and regulation of FeS cluster biosynthesis in parasitic protozoa

67

10.

Unsolved Questions and Future Perspectives

69

10.1 Acquisition and secondary loss of the machineries for Fe S cluster biosynthesis

69

10.2 Origins of individual machineries and crosstalk between the organellar and cytosolic compartments

70

10.3 Cooperation of two cytosolic machineries in Entamoeba and Blastocystis :

SUF/NIF and CIA systems

71

10.4 Significance of compartmentalization of Fe S cluster biosynthesis in mitochondrion-related organelles (MROs)

71

10.5 New strategy for the identification of Fe S cluster-containing proteins

72

10.6 NIF and SUF systems as drug target

72

Acknowledgements

73

 

References

74

Abstract

FeS clusters are ensembles of sulphide-linked di-, tri-, and tetra-iron centres of a variety of metalloproteins that play important roles in reduction and oxidation of mitochondrial electron transport, energy metabolism, regulation of gene expression, cell survival, nitro- gen fixation, and numerous other metabolic pathways. The Fe S clusters are assembled by one of four distinct systems: NIF, SUF, ISC, and CIA machineries. The ISC machinery is a house-keeping system conserved widely from prokaryotes to higher eukaryotes, while the other systems are present in a limited range of organisms and play supplementary roles under certain conditions such as stress. Fe S cluster-containing proteins and the components required for FeS cluster biosynthesis are modulated under stress condi- tions, drug resistance, and developmental stages. It is also known that a defect in FeS proteins and FeS cluster biogenesis leads to many genetic disorders in humans, which indicates the importance of the systems. In this review, we describe the biological and

IronSulphur Cluster Biogenesis in Protozoan Parasites

3

physiological significance of FeS cluster-containing proteins and their biosynthesis in parasitic protozoa including Plasmodium , Trypanosoma, Leishmania , Giardia , Trichomo- nas , Entamoeba , Cryptosporidium, Blastocystis , and microsporidia. We also discuss the roles of FeS cluster biosynthesis in proliferation, differentiation, and stress response in protozoan parasites. The heterogeneity of the systems and the compartmentalization of FeS cluster biogenesis in the protozoan parasites likely reflect divergent evolution under highly diverse environmental niches, and influence their parasitic lifestyle and pathogenesis. Finally, both FeS cluster-containing proteins and their biosynthetic machinery in protozoan parasites are remarkably different from those in their mamma- lian hosts. Thus, they represent a rational target for the development of novel chemo- therapeutic and prophylactic agents against protozoan infections.

and prophylactic agents against protozoan infections. 1. INTRODUCTION Iron–sulphur (Fe–S) proteins are

1. INTRODUCTION

Iron–sulphur (Fe–S) proteins are involved in many central biological functions such as enzymatic catalysis, electron transport, photosynthesis, nitrogen fixation (NIF), and the regulation of gene expression ( Beinert et al., 1997; Lill and Muhlenhoff, 2006). They are found in all domains of life: Archaea, Bacteria and Eukarya. The number of proteins containing Fe–S clusters that are present in eukaryotes is generally much higher than in prokaryotes due to the complexity of the eukaryotic lifestyle such as envi- ronmental response and development ( Py and Barras, 2010). The assembly of Fe–S clusters in vitro occurs spontaneously under favourable conditions when sufficient amounts of free iron and sulphide are available. However, as these substances are toxic to cells in vivo (Balk and Lobreaux, 2005; Johnson et al., 2005; Lill and Muhlenhoff, 2006, 2008; Rouault and Tong, 2008; Vickery and Cupp-Vickery, 2007; Xu and Moller, 2008 ), their concentrations have to be tightly regulated. Thus, Fe–S cluster synthesis does not occur chemically and requires enzymes and cofactors. The assembly of Fe–S clusters is a complex process involving many different systems made up of numerous specific proteins (e.g. > 100 in Escherichia coli) that are wide- spread across the life (Lill, 2009 ). In eukaryotes, Fe–S clusters biosynthesis involves three major systems: the ISC (iron–sulphur cluster), SUF (sulphur utilization factors), and CIA (cytosolic iron–sulphur cluster assembly) machineries. The ISC and SUF machineries are found only in the mito- chondria and the plastids, respectively. The ISC machinery is considered to be a house-keeping system and is widely distributed from prokaryotes to eukaryotes. In contrast, the SUF system plays a role particularly under stress conditions such as iron deprivation and oxidative conditions.

4

Vahab Ali and Tomoyoshi Nozaki

Furthermore, the maturation of Fe–S cluster proteins in the cytoplasm and the nucleus depends on the CIA machinery, which is essential and ubiqui- tous in all eukaryotes (Lillig and Lill, 2009). The NIF machinery is a unique system present in limited lineages of microorganisms such as microaerophilic bacteria, cyanobacteria, nitrogen-fixing bacteria, and unicellular protozoa such as Entamoeba histolytica (Ali et al., 2004 ). In this chapter, we review our current understanding of Fe–S cluster biogenesis in general, and the conservation and/or unique acquisition and secondary loss of four biosyn- thetic systems in the representative parasitic protozoa. Finally, we discuss perspectives and possible exploitations of the research outcomes of Fe–S cluster biogenesis in parasitic protozoa.

outcomes of Fe–S cluster biogenesis in parasitic protozoa. 2. VARIATION AND FEATURES OF Fe – S

2. VARIATION AND FEATURES OF Fe S CLUSTERS 2.1. Discovery of Fe S clusters

The Fe–S clusters were first discovered in the early 1960s, when enzymes with characteristic electron paramagnetic resonance (EPR) signals were purified. Some of the first Fe–S proteins that were discovered include plant and bacterial ferredoxins and respiratory complexes I–III from bacteria and the mitochondria. In the late 1960s, chemical reconstitution was devised to assemble Fe–S clusters into apo-proteins in vitro , which led to the view that the Fe–S clusters can assemble spontaneously on proteins (Malkin and Rabinowitz, 1966). However, genetic, biochemical, and cell biological studies in the 1990s provided ample evidence that the maturation of Fe–S proteins in living cells in vivo is catalyzed by enzymes, but does not occur chemically.

2.2. Heterogeneity of Fe S clusters

Many proteins depend on covalently or non-covalently bound cofactors for their function. Organic cofactors include nucleotides (e.g. FMN and FAD), vitamins (biotin, pantothenate and folate), and metal–organic compounds (haem and molybdenum cofactors). The common inorganic cofactors include metal ions (Mg 2 þ , Zn 2 þ , Mn 2 þ , Cu 1 þ /2 þ , and Fe 2 þ /3 þ ). Among them, Fe–S clusters are considered to be the oldest and most versatile inorganic cofactors. The chemically simple Fe–S clusters are the rhombic [2Fe–2S] and the cubane [4Fe–4S] types, which contain iron (Fe 2 þ /3 þ ) and sulphide (S 2 ). Representative types of Fe–S clusters and their functions are summarised in Table 1.1 .

Table 1.1 Representative Fe S cluster-containing proteins Type and name

Group of protein

Source

Function (reaction)

Type of FeS cluster

1 Simple Fe–S proteins

[2Fe–2S] Ferredoxin, plant type

Cyanobacteria, Clostridia, Protozoa, Chloroplasts

Photosynthetic reduction of NADP, nitrite, and thioredoxin

[2Fe–2S]

[3Fe–4S] Ferredoxin

Bacteria, e.g.,

Electron transfer

[3Fe–4S]

Desulfovibrio gigas

[4Fe–4S] Ferredoxin

Bacteria, e.g., Bacillus , Electron transfer Desulfovibrio spp.

[4Fe–4S]

High-potential Fe–S protein (HiPIP)

Photosynthetic

Anaerobic electron transport

[4Fe–4S]

bacteria, e.g.,

 

Chromatium vinosum

7Fe Ferredoxin

Azotobacter vinelandii

Storage of Fe

[4Fe–4S] þ

 

[3Fe–4S]

8Fe Ferredoxin

Anaerobic bacteria,

Electron transfer

2[4Fe–4S]

e.g., Clostridium

pasteurianum

Continued

Table 1.1 Representative Fe S cluster-containing proteins cont'd Type and name

Group of protein

Source

Function (reaction)

Type of Fe S cluster

2 Membrane-bound electron transfer proteins

NADH ubiquinone

oxidoreductase

Aerobic bacteria,

mitochondria

þ

NADH þ H þ þ UQ þ 4H in ! NAD þ þUQH 2 þ 4H

þ

out

2[2Fe–2S] þ 6

[4Fe–4S]

(complex I)

Succinate

Aerobic bacteria,

Succinate þ UQ ! Fumarate þ UQH 2

[2Fe–2S] þ [3Fe– 4S] þ [4Fe–4S]

dehydrogenase

mitochondria

(complex II)

UQH 2 : cytochrome c reductase (complex III)

Aerobic bacteria,

UQH 2 þ CytC ox ! UQ þ CytC red

[2Fe–2S]

mitochondria

 

Cytochrome b 6 / f complex

Cyanobacteria,

PQH 2 þ PC ox ! PQ þ PC red

[2Fe–2S]

chloroplasts

 

[NiFe] hydrogenase

Bacteria, e.g. E. coli ,

H 2 þ Menaquinone ! Menaquinol

[4Fe–4S] þ

(respiratory)

C.

vinosum ;

[3Fe–4S]

A.

vinelandii

Gylcerol phosphate

E.

coli

Glycerol phosphate þ Menaquinone ! Glyceraldehyde phosphate þ Menaquinol

2[4Fe–4S] þ [2Fe–

dehydrogenase

2S]

(anaerobic)

Fumarate reductase

Saccharomyces cerevisiae Fumarate þ Menaquinol ! Succinate þ Menaquinone [2Fe–2S] þ [3Fe– 4S] þ [4Fe–4S]

3 Soluble Fe–S enzymes

NAD(P)H-glutamate

E.

coli , plants

Glutamine þ 2-Oxoglutarate þ NAD(P)H ! 2Glutamate þ NAD(P) þ

2[4Fe–4S]

synthase

Ferredoxin-glutamate

Plants

Glutamate þ 2-Oxoglutarate þ 2Fd ox ! 2Glutamate þ 2 Fd red

[4Fe–4S] þ

synthase

[2Fe–2S]

Pyruvate:

Euglena gracilis

Pyruvate þ CoA þ NADP þ ! Acetyl CoA þ CO 2 þ NADPH

[4Fe–4S]

NADP þ oxidoreductase

Pyruvate:ferredoxin

Cyanobacteria,

Pyruvate þ CoA þ Fd ox ! Acetyl CoA þ CO 2 þ Fd red 2–3[4Fe–4S]

oxidoreductase

Clostridia, Protozoa

[NiFe] hydrogenase (cytochrome c reducing)

Desulfovibro spp.

H 2 þ CytC ox ! 2H þ þ CytC red

2[4Fe–4S] þ [3Fe–

 

4S]

[NiFe] hydrogenase

Hydrogen bacteria, e.g. A. eutrophus , Nocardia opaca

H 2 þ NAD þ ! H þ þ NADH

3–4[4Fe–4S] þ

(NAD reducing)

[2Fe–2S]

4 Hydroxylases and dioxygenases

 

Ferredoxin (P-450

Bacteria, e.g. Pseudomonas putida ; mitochondria

Electron transfer from flavoprotein reductase to cytochrome P-450

[2Fe–2S]

reducing)

Continued

Table 1.1 Representative Fe S cluster-containing proteins cont'd Type and name

Group of protein

Source

Function (reaction)

Type of Fe S cluster

5 Enzymes with the molybdopterin cofactor

Xanthine oxidase

Milk

Xanthine þ O 2 > Urate þ H 2 O 2 þ O 2

2[2Fe–2S]

Ferredoxin-nitrate

Cyanobacteria, e.g.

NO 3 þ Fd red þ 2H þ ! NO 2 þ Fd ox þ H 2 O

2[2Fe–2S]

reductase

Plectonema boryanum

6 Enzymes containing sirohaem

 

Ferredoxin: sulphite

Bacteria, plants

SO 3 2 þ 6Fd red ! S 2 þ 6Fd ox

2[4Fe–4S]

reductase

7 Proteins with catalytic Fe–S or mixed-metal clusters

 

[Fe] hydrogenase

Anaerobic bacteria, e.g. C. pasteurianum

2H þ þ 2e ! H 2

H cluster þ 2–4

[4Fe–4S]

Carbon monoxide

Photosynthetic bacteria, e.g. Rhodospirillum rubrum

CO þ H 2 O > CO 2 þ 2e þ 2H þ

7[4Fe–4S] þ Ni– [Ni–3Fe–4S] þ [4Fe–4S] C-cluster

dehydrogenase

Carbon monoxide dehydrogenase (acetyl CoA synthase)

Acetogenic and methanogenic bacteria, e.g. C. thermoaceticum , Methanothrix soehngenii

CH 3 –[CP] þ CO þ CoA ! CH 3 CO–CoA þ [CP]

7[4Fe–4S] þ Ni– [Ni–4Fe–4S] þ [4Fe–4S] A-cluster

Mo nitrogenase

Nitrogen-fixing bacteria, e.g. Rhizobium , Azotobacter

N 2 þ 8e þ 10H þ ! 2NH 4 þ þ H 2

P clusters, Fe–Mo

[8Fe–7S]

8 Enzymes with nonredox Fe–S clusters

 

Aconitase (aconitate

Bacteria, cytoplasm,

Citrate ! Isocitrate

[4Fe–4S]

hydratase)

mitochondria

DNA endonuclease III E. coli

Apurinic and apyridimic endonuclease

[4Fe–4S]

L -Serine dehydratase

Peptostreptococcus

Serine > Pyruvate þ NH 4 þ

[4Fe–4S]

asaccharolyticus

9 Regulatory proteins

Ferredoxin: thioredoxin Cyanobacteria,

Thioredoxin þ Fd red ! Thioredoxin red þ Fd ox

[4Fe–4S]

reductase

chloroplasts

Names of proteins, their sources (organisms or intracellular localizations), functions (reactions) and types of Fe–S clusters are shown.

10

Vahab Ali and Tomoyoshi Nozaki

Nitrogenase, carbon monoxide dehydrogenase (CODH), and hydroge- nase are good examples to show the heterogeneity of Fe–S clusters. Nitro- genase, which fixes atmospheric nitrogen gas (N 2 ) as ammonia, is composed of the heterotetrameric iron–molybdenum (FeMo)-cofactor-associated pro- tein ([8Fe–7S], 7Fe:Mo:9S:homocitrate:X cluster), that is transiently associ- ated with the homodimeric Fe protein ([4Fe–4S]). The FeMo protein binds to a substrate and reduces H þ and N 2 to H 2 and ammonia, while the Fe pro- tein receives electrons from ferredoxin, hydrolyses ATP, and reduces the FeMo protein. The Fe protein has a P-cluster, which is a putative electron transfer centre (Burgess and Lowe, 1996; Howard and Rees, 1996) and undergoes dramatic conformational changes between at least two oxidative states (P N -cluster to P OX -cluster) (Peters et al., 1997 ). The metabolism of CO by CODH/acetyl CoA synthetase is mediated by two different clusters responsible for different reaction mechanisms (Ragsdale and Kumar, 1996). The reversible oxidation of CO to CO 2 is catalysed by the C-cluster, which is an Ni–[3Fe–4S] cluster of CODH, while the reversible reaction of CO with coenzyme A (CoA) is catalysed by the A-cluster (Ni– [4Fe–4S]) on acetyl CoA synthetase. Hydrogenases catalyse the reversible reduction of protons to hydrogen. Two types of clusters that differ in metal binding are known in hydrogenases. Iron-only hydrogenases (Fe- hydrogenase) have the active site H-cluster composed of a [4Fe–4S] cluster bridged to a binuclear iron centre. The second type of cluster in Ni–Fe- hydrogenases contains a binuclear cluster of iron and nickel in part bridged by sulphides (Volbeda et al., 1995). The existence of various Fe–S clusters in nature, from the simplest cluster with broad non-specific properties to sophis- ticated clusters with highly specific and efficient catalytic properties, suggests that Fe–S clusters have evolved into various transitional states under selective pressures such as changes in atmospheric conditions (Rees and Howard, 2003).

2.3. Physicochemical features and analytical methods of Fe S clusters

Iron is usually coordinated in a tetrahedron by sulphurs from inorganic sul- phide and cysteine thiol groups of the proteins. Side chains of other amino acids (His, Asp, Arg, Ser) are also known to coordinate iron. Sulphides gen- erally bridge two or three irons, while in some cases more irons are bound to sulphides. The two simplest Fe–S clusters are [2Fe–2S] and [4Fe–4S]. They have versatile electrochemical properties with reduction potentials ranging from over 400 to below 400 mV ( Beinert, 2000 ), which is a range larger than any other simple redox cofactors. Moreover, these simple clusters are

IronSulphur Cluster Biogenesis in Protozoan Parasites

11

sometimes chemically integrated as part of non-redox catalysis, such as in aconitase (Beinert et al., 1996 ) ( Table 1.1; also see Section 8.4.1 ), where the cluster serves as a Lewis acid. These physicochemical properties are the basis of a wide range of distribution of the Fe–S clusters in a multitude of functions (Rees and Howard, 2003 ). Another example of the adaptation of [4Fe–4S] clusters is also often found in some proteins where one iron can be lost with ease to form a [3Fe–4S] cluster under the influence of protein structure or oxygen. The [4Fe–4S] cluster is oxygen sensitive and labile, while the [3Fe–4S] cluster is more stable against oxygen than [4Fe–4S]. Depending on the overall oxidation state of the Fe–S cluster, iron may har- bour unpaired electrons, and the resulting electron spin can be detected by EPR spectroscopy. This biophysical technique provides valuable informa- tion on the type, the oxidation state, and the electronic environment of the clusters (Moulis et al., 1996 ). Other biophysical methods to study the structure and properties of the Fe–S clusters include Mo¨ ssbauer, extended X-ray absorption fine structure, magnetic circular dichroism, electron nuclear double resonance spectroscopy, nuclear magnetic resonance, and resonance Raman spectroscopy. They provide specific information on the electronic and magnetic properties of the Fe–S clusters in their particular environment. In UV–Vis spectrophotometric measurements, Fe–S proteins exhibit typical absorption maxima around 420 nm ([4Fe–4S] clusters) and 320, 410 and 560 nm ([2Fe–2S] clusters). These spectral properties give rise to the dark brown colour of purified Fe–S holo-proteins.

2.4. Biochemical features of Fe S clusters

The robust and unique physicochemical properties of Fe–S clusters confer diverse biochemical abilities on Fe–S cluster proteins. First, Fe–S clusters mediate electron transfer due to the properties of accessing various redox states. Second, Fe–S clusters mediate redox catalysis. Fe–S clusters can reach very low redox potentials and thereby reduce redox-resistant substances. Third, Fe–S clusters are involved in non-redox catalysis. Small compounds are allowed to bind to accessible ferric sites with extensive Lewis acid prop- erties. Fourth, Fe–S clusters are able to regulate gene expression. This is due to the reversible inter-conversional properties of Fe–S clusters to be exqui- site sensors of several redox- or iron-related stresses ( Beinert, 2000; Fontecave, 2006; Kiley and Beinert, 2003 ) (see Section 3.1.5). Although the robustness of Fe–S clusters is highly valuable to life, oxygen poses a threat to Fe–S proteins and, consequently, to the organisms relying

12

Vahab Ali and Tomoyoshi Nozaki

on them (Imlay, 2006). Fe–S-containing dehydratase from E. coli is a good example to show the high sensitivity of Fe–S clusters against oxygen. Fe–S clusters in E. coli dehydratase directly react with univalent oxidants such as hydrogen peroxide and peroxynitrite, leading to inactivation of dehydratase with concomitant loss of iron. The oxidised and liberated iron can cause a Fenton reaction, which produces highly toxic reactive oxygen species (ROS) in the cell. Furthermore, ROS leads to deleterious effects on DNA and other macromolecules. Therefore, while Fe–S clusters play essen- tial roles in various biological processes (see Section 3), they make proteins and organisms that harbour them susceptible to oxidative stress.

organisms that harbour them susceptible to oxidative stress. 3. GENERAL BIOLOGICAL AND PHYSIOLOGICAL ROLES OF Fe

3. GENERAL BIOLOGICAL AND PHYSIOLOGICAL ROLES OF Fe S PROTEINS IN PROKARYOTES AND EUKARYOTES

The biological functions of Fe–S proteins are diverse in all domains of life. Fe–S proteins play major roles in electron transfer, respiration, photosyn- thesis, substrate binding and activation, iron and cluster storage, regulation of gene expression, and enzymatic activities, sulphur donation, disulphide reduc- tion, structural modification, and regulation of metabolic pathways.

3.1. Roles of Fe S proteins in bacteria

3.1.1 Electron transfer The electron transfer by Fe–S clusters employs the property of the iron ions of the Fe–S clusters to switch between reduced (ferrous Fe 2 þ ) and oxidised (ferric Fe 3 þ ) states. The ability to delocalise electron density over iron atoms makes Fe–S clusters ideal for their role in mediating biological electron transport. Fe–S clusters are major components in the photosynthetic and respiratory electron transport chains. They define the electron transport pathways in numerous membrane-bound and soluble redox enzymes, and constitute the redox-active centres in ferredoxins, one of the largest classes of mobile electron carriers in biology. The clusters involved in electron transfer contain a [2Fe–2S], [3Fe–4S], [4Fe–4S], or [8Fe–7S] core unit with cysteinate generally completing tetrahedral S coordination at each iron site. Aspartate, histidine, serine, or backbone amide ligation at a unique iron site is occasionally encountered in clusters that function in electron transport. These ligands are likely to play a role in modifying redox potential, gating electron transport, or coupling proton and electron transport (Johnson et al., 2005). Although the vast majority of electron transferring Fe–S clusters are

IronSulphur Cluster Biogenesis in Protozoan Parasites

13

one-electron carriers, the double-cubane [8Fe–7S] cluster that is found only in nitrogenase has the potential to act as a two-electron carrier.

3.1.2 Substrate binding and activation

Fe–S clusters serve as a substrate-binding site for a wide variety of redox and non-redox enzymes in which Lewis acid-assisted enzyme catalysis is used for performing the catalytic reactions. A site for substrate binding and activation can be established in three different ways. First, a non-cysteinyl residue of the protein side chain binds to a unique iron site of an Fe–S cluster. For instance, the amino and carboxylate groups of methionine in S -adenosyl methionine (SAM) are used to facilitate reductive cleavage and generate the 5 0 -deoxyadenosyl radical in case of the radical-SAM family of Fe–S enzymes ( Cheek and Broderick, 2001 ). The radical-SAM super-family

comprises more than 60 enzymes that catalyse radical reactions in DNA pre- cursors, vitamins, cofactors, antibiotics, herbicide biosynthesis and degrada- tion pathways. Second, a heterometal can be incorporated in the Fe–S cluster for achieving the active configuration of the enzymes, for example, the Ni–[Ni–4Fe–4S] cluster in CODH (also see Section 2.2) (Dobbek et al., 2001). Third, a substrate-binding metal site is bound to an iron of a [4Fe–4S] cluster via a bridging cysteinyl residue. For instance, the dinickel centre is attached to form the functional form of the acetyl CoA synthase active site and a di-iron centre is attached to form the Fe-hydrogenase active site (Peters et al., 1998 ).

3.1.3 Iron and cluster storage

Ferredoxins containing two [4Fe–4S] clusters were shown to play an essen- tial role in iron homeostasis and in iron storage in Clostridium. This is further supported by the presence of polyferredoxins containing up to 12 [4Fe–4S] clusters in tandemly repeated [8Fe] ferredoxin-like domains in methanogenic archaea (Johnson et al., 2005). They are presumed to serve

as a store of Fe–S clusters.

3.1.4 Structural integrity

Fe–S clusters control protein structure. This was demonstrated by an Fe–S cluster-driven protein reorganization in response to solvent effects and cys- teine substitutions, as well as by the ability of designed and unstructured minimal synthetic peptides to correctly assemble Fe–S clusters. For instance,

[4Fe–4S] clusters of DNA repair enzymes, endonuclease III and MutY, are redox inactive and control the structure of a protein loop essential for

14

Vahab Ali and Tomoyoshi Nozaki

recognition and repair of damaged DNA, similar to zinc in zinc-finger pro- teins (Guan et al., 1998; Kuo et al., 1992).

3.1.5 Regulation of gene expression

The sensitivity of Fe–S clusters to oxidative damage is exploited to sense intracellular or environmental conditions, and hence regulates gene expres- sion in prokaryotes. For instance, transcription factors (gene or operon repressors) such as IscR, SoxR, and FNR are regulated at the transcriptional level in oxidative or iron-deficient conditions. It was shown that a single promoter upstream of the iscR gene directs expression of the four iscRSUA genes in an operon ( Schwartz et al., 2001 ), and IscR directly represses the iscR promoter (Schwartz et al., 2001). Analysis of IscR by EPR showed that the anaerobically isolated protein indeed contains a [2Fe–2S] cluster that is able to undergo reversible oxidation and reduction ( Schwartz et al., 2001). The Fe–S cluster is important for the repressor function of IscR because iscS or hscA mutations, which abolish Fe–S cluster biogenesis (see Section 5.2), led to nearly constitutive expression of downstream genes in the operon (Schwartz et al., 2001 ). In addition, the expression of the iscRSUA operon was induced by hydrogen peroxide and by 2,2 0 -dipyridyl-mediated iron starvation, and the induction was dependent upon the presence of an intact proximal iscR gene (Outten et al., 2004 ). It was also demonstrated that in E. coli , hydrogen peroxide induces the expression of a set of genes via the transcriptional activator, OxyR (Zheng et al., 2001). It was further shown that OxyR directly regulates expression of genes in the SufABCDSE operon (see also Sections 3.2.1 and 9.2) (Lee et al., 2004b; Outten et al., 2004). In addition, OxyR and Fur were also shown to mediate up-regulation of the Suf operon induced by nitrosylated glutathione (GSH) or nitric oxide known as nitrosative stress in bacteria ( Kim et al., 2002; Mukhopadhyay et al., 2004).

3.1.6 Regulation of enzymatic activity

Fe–S clusters are implicated in the regulation of enzymatic activity in response to external stimuli, with the restoration of activity requiring cluster

assembly or repair. For instance, in the case of DNA glycosylase, involved in DNA repair, and aconitase, it was shown that when the [4Fe–4S] cluster in the holo-form of the enzyme is misassembled, oxidised, damaged, or removed, its enzymatic activity is lost ( Tang and Guest, 1999 ). A similar observation was also made for the [2Fe–2S] cluster in mammalian ferrochelatases, the terminal enzyme of haem biosynthesis (Wu et al., 2001 ).

IronSulphur Cluster Biogenesis in Protozoan Parasites

15

3.1.7 Disulphide reduction

[4Fe–4S] clusters in disulphide reductases, such as ferredoxin:thioredoxin reductase in chloroplasts and heterodisulphide reductase in methanogenic archaea, were shown to be involved in disulphide reduction in two sequen- tial steps, in which one-electron transfer is involved in a formation of an intermediate with two thiolate ligands at a unique iron site (Dai et al.,

2000; Duin et al., 2002 ).

3.1.8 Sulphur donor

Fe–S clusters may also serve as a sulphur donor by its reductive cleavage. For example, a [2Fe–2S] cluster of b iotin synthase is reduced in each catalytic cycle to provide sulphur for the conversion of dethio-biotin to biotin. After each catalytic cycle, the cluster needs to be reconstituted to regain the enzymatic activity ( Berkovitch et al., 2004; Jameson et al., 2 004; Ugulava et al., 2001 ).

3.2. Roles of Fe S proteins in eukaryotes

Besides the functions demonstrated in bacteria as mentioned above, addi- tional functions were reported in eukaryotes.

3.2.1 Cellular iron homeostasis and gene expression regulation

It was reported that the mitochondrial ISC machinery (see Section 5.2) and the export machinery (see Section 5.2.6 ) have a strong influence on the uptake, intracellular distribution, and utilization of iron in yeasts ( Kaplan et al., 2006; Lill and Kispal, 2000 ). The roles of frataxin (FXN) (see Section 4.1.4 ) and Atm1 (see Section 5.2.6 ) in iron homeostasis had been implicated before their specific functions in Fe–S clusters metabolism were identified. In yeast, the major regulatory factors controlling iron acquisition and intracellular iron distribution in response to different iron levels in the environment are the transcription factors Aft1 and Aft2 (Blaiseau et al., 2001; Rutherford et al., 2001; Yamaguchi-Iwai et al., 1996). Aft1 was identified as a major iron-responsive transcription factor, while Aft2 was identified when its deletion was found to be associated with mild disturbance of iron homeo- stasis. Upon iron depletion, Aft1 is translocated from the cytosol to the nucleus, via interaction with importin, and transcriptionally activates iron regulon ( Ueta et al., 2003). Importantly, Aft1 and Aft2 interact with

Fe–S clusters or their precursors that are synthesised in the mitochondria and transported to the cytosol via Atm1 (see Section 5.2.6).

16

Vahab Ali and Tomoyoshi Nozaki

In contrast to prokaryotes, in which expression of Fe–S proteins is reg- ulated at the transcriptional level via transcription factors such as SoxR, IscR, FNR, and OxyR (see Section 3.1.5), in eukaryotes, gene expression is regulated in a more complex manner. For instance, in mammals, iron reg- ulatory protein 1 (IRP1) is known to be regulated at a post-transcriptional level (Lill and Muhlenhoff, 2006; Lill et al., 2006 ). IRP1 appears to have a dual role in the cell. As an Fe–S holo-protein, it functions as a cytosolic aconitase. When the Fe–S cluster is lost, apo-IRP1 functions as an RNA- binding protein, which binds to the RNA stem-loop structures, called iron regulatory elements (IRE), of specific mRNAs encoding proteins involved in iron uptake, utilization, storage, and export. As a consequence of the IRP1–IRE binding, target mRNAs are stabilised and thus protected from degradation, which results in enhanced protein synthesis, whereas for non-target mRNAs, the stem-loop structures at the 5 0 -untranslated end block efficient translation by the ribosomes and lead to a decrease in protein synthesis (Walden et al., 2006 ).

3.2.2 Photosynthesis Fe–S clusters are essential components for photosynthesis, which is the pro- cess unique to algae and plants. Iron of Fe–S clusters is engaged in the transfer of an electron from water to NADPH, which is used to reduce carbon diox- ide to form sugars (Raven and Falkowski, 1999). It is also involved in nitro- gen and sulphur/sulphate reduction and assimilation, for example, by nitrite reductase and sulphite reductase (Krueger and Siegel, 1982; Lancaster et al., 1979). Both mitochondrial and photosynthetic electron transport chains contain a number of proteins that require Fe–S clusters as a cofactor for their activity. The photosynthetic electron transport chain contains three major complexes known as photosystem I (PSI), photosystem II (PSII) and cyto- chrome b 6 / f complex. Iron is present in all of these complexes and is the most important redox-active metal ion for the photosynthetic electron transport chain, both quantitatively and qualitatively (Bar-Ness et al., 1992; Yehuda et al., 1996 ). Iron is present in both haem and non-haem forms in PSII. In contrast, in the cytochrome b 6 complex, iron is present as haem and as [2Fe–2S] Rieske-type clusters. PSI contains one [4Fe–4S] cluster each in PsaA and PsaB and one [2Fe–2S] cluster in ferredoxin (Raven and Falkowski, 1999 ). Other plastidial proteins that contain Fe–S clusters include ferredoxin–thioredoxin reductase (4Fe–4S), sulphite reduc- tase (4Fe–4S), nitrite reductase (4Fe–4S), and glutamate synthase (3Fe–4S) (Balk and Lobreaux, 2005).

IronSulphur Cluster Biogenesis in Protozoan Parasites

17

3.2.3 tRNA biosynthesis

It is known that bacterial IscS plays a crucial role in providing sulphur for thio modification of tRNA (Kambampati and Lauhon, 1999; Mueller, 2006) (see Section 5.2.4 and 8.5.3). Recently in yeast and human, thio mod- ification was also demonstrated in both mitochondrial and cytosolic tRNAs with two uridine nucleosides in positions 34 and 35. The modification of uridine in the wobble position 34 seems to be present in all living organisms and essential for viability as it increases the decoding efficiency of tRNAs (Bjork et al., 2007 ). Mitochondrial and cytosolic Nfs1 are responsible for thio modification of mitochondrial and cytoplasmic tRNAs in yeast (Muhlenhoff et al., 2004; Nakai et al., 2004 ). Similarily, thiouridine mod- ification of tRNA (tRNA S ) by Nfs1 also occurs in the nucleus (Nakai et al., 2004). Despite the previous notion that thio modification of tRNA can occur in both the mitochondria and the cytosol, thio modification of tRNA was shown to be dependent on mitochondrial Nfs1. In addition, IscU1 and IscU2, the scaffold components of the mitochondrial ISC assembly (see Section 5.2), and the components of cytosolic CIA machinery, cluster- deficient protein 1 (Cfd1), nucleotide-binding protein 35 (Nbp35) and CIA machinery protein 1 (Cia1) (see Section 5.5) were shown to be needed for cytosolic tRNA thio modification (Nakai et al., 2007 ). Thus, thio mod- ification of cytosolic tRNA requires the entire Fe–S protein assembly apparatus.

tRNA requires the entire Fe–S protein assembly apparatus. 4. GENERAL SOURCE OF IRON AND SULPHUR FOR

4. GENERAL SOURCE OF IRON AND SULPHUR FOR Fe S CLUSTER BIOSYNTHESIS

4.1. Acquisition and transport of iron

4.1.1 Iron acquisition via the plasma membrane

Iron is indispensable for all living cells, and thus, organisms have multiple pathways to scavenge iron. The ferric iron (Fe 3 þ ) is largely insoluble and can react with hydrogen peroxide, a by-product of respiration, generates highly toxic hydroxyl radicals, and thus is toxic to the cell at physiological pH under aerobic conditions. Hence, iron is usually present in a protein- bound form in the cell. Some bacteria and eukaryotes possess surface receptors that specifically bind transferrin or lactoferrin and facilitate the internalization of iron–protein complexes to scavenge iron (Blanton et al., 1990; Steverding et al., 1995). Subsequent acidification of endosomes

induces conformational changes in the complex with a consequent release of iron (Sipe and Murphy, 1991). Yeast and other bacteria possess reductases

18

Vahab Ali and Tomoyoshi Nozaki

that reduce transferrin- or lactoferrin-bound Fe 3 þ to ferrous (Fe 2 þ ) iron, a soluble ion that is more readily internalised (Deneer et al., 1995 ).

4.1.2 Siderophores

Bacteria and fungi acquire iron by utilization of secreted siderophores, which are low-molecular weight iron-chelating compounds and can effec- tively compete with the host proteins for iron. Bacteria incorporate iron– siderophore complexes through specific membrane receptors (Nikaido, 1993; Schryvers and Stojiljkovic, 1999; Visca et al., 2002). Bacteria lacking siderophores also utilise specific receptors for transferrin or lactoferrin for the iron acquisition (Braun and Braun, 2002; Braun and Hantke, 2011; Kishore et al., 1991).

4.1.3 Iron transport to the mitochondria and the plastids

Major metabolites can cross the mitochondrial outer membrane easily. However, the inner membrane can be crossed only with specialised carriers, specially known as mitochondrial carrier family proteins ( Nury et al., 2006). The high affinity mitochondrial carriers, Mrs3 and Mrs4, have been pro- posed to mediate iron transport in yeast ( Muhlenhoff et al., 2003; Zhang et al., 2005, 2006 ). However, as these two transporters were demonstrated not to be essential for iron transfer to the mitochondrial matrix, other trans- porter is likely present. In vertebrates, two isoforms of mitoferrin were reported for iron transport ( Li et al., 2002 ). In human, a homologue of yeast Mrs3/4 is ubiquitously expressed and responsible for the iron transport to the mitochondria (Li et al., 2002 ). In contrast, iron uptake andtransport acrossthe plastid (chloroplast) mem- branes is not well understood. In Arabidopsis thaliana, a membrane-spanning iron transporter, permease in chloroplasts 1 (PIC1), was identified. PIC1 con- tains four predicted a-helices, is targeted to the inner envelope membrane, and displays homology with cyanobacterial permease-like proteins. Knock- out mutants of PIC1 grew only heterotrophically and were characterised by a chlorotic and dwarfish phenotype reminiscent of iron-deficient plants. The mutants revealed severely impaired chloroplast development and a striking increase in ferritin clusters. In addition, pic1 mutants showed differential reg- ulation of genes and proteins related to iron stress or transport, photosynthe- sis, and Fe–S cluster biogenesis. Furthermore, PIC1 and its cyanobacterial homologue mediated iron accumulation in an iron uptake-defective yeast mutant. These data suggest that PIC1 functions in iron transport across the inner envelope of chloroplasts (Duy et al., 2007, 2011 ).

IronSulphur Cluster Biogenesis in Protozoan Parasites

19

4.1.4 Frataxin In yeast, FXN is essential for providing iron to the mitochondrial ISC sys- tem, while in the bacterial system, the SUF proteins were found to be nec- essary for ferric siderophore acquisition (Barras et al., 2005; Gerber et al., 2003; Nachin et al., 2003; Ramazzotti et al., 2004 ). Human FXN is present in multiple forms in the cell after mitochondrial import and processing. The size variation of FXN was documented in haematopoietic, muscle, cardiac, and non-haematopoietic cells, suggesting tissue-specific processing during mitochondrial transport ( Condo et al., 2007 ). The suggested major functions of FXN include, in general, iron and haem metabolism, ISC assembly, oxi- dative phosphorylation, and protection against oxidative stress (Bulteau et al., 2004; Gakh et al., 2002; Kumanovics et al., 2008; Martelli et al., 2007; Radisky et al., 1999). FXN functions as a metallochaperone during Fe–S cluster biosynthesis as shown by its in vitro ability to bind ferrous iron with micromolar-binding affinity. FXN also directly binds Isu in an iron-dependent manner with a submicromolar binding affinity and stimulates [2Fe–2S] cluster production (Stemmler et al., 2010). In yeast, it was shown that FXN interacts with Isu (and Nfs1) in vitro and that FXN depletion was associated with a defect in the de novo Fe–S cluster formation on Isu (Gerber et al., 2003 ). The strong sequence conservation of FXN and Isu orthologues across species indicates a conserved mechanism for Fe(II) delivery and Fe–S cluster production in eukaryotes. Cellular distribution of FXN changes upon stress response. In adult human tissues, FXN is localised to the mitochondria under normal non-stress conditions, while under oxidative stress, cytosolic FXN accumu- lates, which is caused by damage on the Fe–S cluster by ROS (Martelli et al., 2007). A donor of iron for Fe–S cluster biogenesis in the mitochondria is still debated and direct in vivo demonstration of interaction of IscU/IscA with FXN is required. In humans, mutations affecting FXN expression or func- tion resulted in Friedreich’s ataxia (FRDA), an autosomal recessive neuro- and cardio-degenerative disorder that represents the most common inherited ataxia in humans, affecting 1 in 30,000–50,000 individuals (Bencze et al., 2006) (see Section 6.1 ).

4.2. Acquisition, biosynthesis, and transport of L-cysteine

L -Cysteine is obtained via several independent routes. First, L-cysteine is de novo synthesised from sulphide and L -serine in bacteria, fungi, protozoa (i.e. Entamoeba, Trichomonas , Trypanosoma , and Leishmania ), and plants.

20

Vahab Ali and Tomoyoshi Nozaki

Second, L-cysteine is synthesised from L -methionine by reverse trans- sulphuration in mammals. Third, L -cysteine can be directly obtained by scavenging via amino acid transporters from the environmental milieu or after degradation of ingested proteins in lysosomes and phgosomes. Cysteine biosynthesis occurs in bacteria, yeast, plants and some protozoa, in which L -serine and sulphide are converted into L -cysteine via O- acetylserine or O-phosphoserine. O-Acetylserine or O-phosphoserine reacts with sulphide, which is provided directly or by reduction of sulphate, to form L-cysteine. The cysteine biosynthetic pathway was previously described in details in the previous volume (Nozaki et al., 2005). In contrast, animals require L -homocysteine, which is derived from L-methionine via the reverse trans-sulphuration pathways as a source of sulphur to produce L -cysteine via cystathionine ( Ali and Nozaki, 2007 ). Cysteine permease/transporters have been identified in bacteria and eukaryotes (During-Olsen et al., 1999; Kosugi et al., 2001), and their role in cysteine transport was confirmed in yeast. L-Cysteine is apparently trans- ported not by a specific permease, but by multiple permeases with a broad specificity, and all of these permeases are active under different sets of growth conditions. However, a specific cysteine transporter (Yct1p, YLL05wp) with a high affinity, was identified and characterised by genetic and bio- chemical analysis (Kaur and Bachhawat, 2007 ). A novel plasma membrane cystine transporter from Candida glabrata ( CgCYN1), which belongs to the amino acid permease family, but has no similarity to known plasma mem- brane or lysosomal cystine transporters from human and yeast was reported recently (Yadav and Bachhawat, 2011). L -Cystine uptake by CgCYN1 was energy dependent and inhibited by L -cystine, but not by other amino acids including L -cysteine. However, structurally similar amino acids, for exam- ple, cystathionine, lanthionine, and selenocystine, also inhibited transport, confirming the cystine specificity. Cysteine or cystine transporter has not been demonstrated in protozoan parasites, while it was demonstrated that L -cysteine is specifically transported in Trypanosoma cruzi epimastigotes (Canepa et al., 2009).

Trypanosoma cruzi epimastigotes ( Canepa et al., 2009 ). 5. FOUR SYSTEMS FOR Fe – S

5. FOUR SYSTEMS FOR Fe S CLUSTER BIOGENESIS IN PROKARYOTES AND EUKARYOTES

Fe–S cluster assembly has been discovered in all living organisms examined so far. Four major systems, ISC, SUF, NIF, and CIA, are known to be involved in the process, and two to three of these systems are present in

IronSulphur Cluster Biogenesis in Protozoan Parasites

21

each organism. Saccharomyces cerevisiae is the best characterised organism to study Fe–S clusters machinery among eukaryotes (Lill and Muhlenhoff, 2006; Lill et al., 2006, 2012; Muhlenhoff and Lill, 2000).

5.1. Catalytic reaction in Fe S cluster biosynthesis

In the ISC, SUF, and NIF systems from bacteria, lower eukaryotes, plants, and humans, Nif S- and Nif S-like proteins, namely, IscS (Nfs1) and Suf S (Csd, Nfs2), are responsible for the catalysis of initial donation of sulphur/ sulphide for the assembly of Fe–S clusters. These enzymes catalyse a cysteine desulphurase reaction: the mobilization of sulphane sulphur (S ) or sulphide (S 2 ) from L -cysteine with concomitant production of L -alanine and in the presence of reducing agents ( Flint, 1996; Flint et al., 1996; Mihara and Esaki, 2002; Mihara et al., 1997, 1999). The cysteine desulphurase activity of SufS and IscS is enhanced by SufE in bacteria and plants, or Isd11 in yeast and other eukaryotes, respectively. The reaction catalysed by cysteine des- ulphurase produces a substrate-ketimine intermediate from L -cysteine and pyridoxal phosphate. The sulphur atom of this intermediate is then attacked by the catalytic cysteine residue, resulting in the formation of a cysteine per- sulphide residue at the active side of the enzyme, with a concomitant release of L -alanine. The substrate sulphur atom of the cysteine persulphide residue can be subsequently transferred to different biomolecules, for building of Fe–S clusters (Mihara et al., 2000 ). These catalytic enzymes can also act on L-cysteine sulphinate and L-selenocysteine as substrates, releasing SO 2 and selenium, respectively (Mihara et al., 1999, 2000).The mechanisms of degradation of L-selenocysteine and L-cysteine sulphinate differ from that of L-cysteine desulphuration (Mihara et al., 2000). Furthermore, genetic studies revealed that the IscS mutant con- tained reduced activities of several soluble and membrane-bound Fe–S proteins (Djaman et al., 2004; Schwartz et al., 2000). The assembly of Fe–S clusters and transfer to apo-proteins was achieved in vitro using IscS, cysteine, dithiothreitol and ferrous ammonium sulphate (Kurihara et al., 2003).

5.2. ISC machinery

The first system for Fe–S cluster assembly, designated as ISC, is required for the generation of the majority of cellular Fe–S proteins and thus performs a general house-keeping biosynthetic function in all ISC-possessing organisms from bacteria to higher eukaryotes (Fig. 1.1A).

22

Vahab Ali and Tomoyoshi Nozaki

A Bacterial ISC machinery Fe 2+ CyaY -S-SH IscS IscU HS-S- Sulphur - e Fdx
A Bacterial ISC machinery
Fe 2+
CyaY
-S-SH
IscS
IscU
HS-S-
Sulphur
-
e
Fdx
Fdx
Cys
Ala
IscA
IscR
–

iscR

iscS

iscU

iscA

Cys Ala IscA IscR – iscR iscS iscU iscA ISC operon Apo HscA HscB Holo Bacterial

ISC operon

Apo HscA HscB Holo Bacterial
Apo
HscA
HscB
Holo
Bacterial
iscU iscA ISC operon Apo HscA HscB Holo Bacterial Fe–S proteins Fe–S cluster B Bacterial SUF

Fe–S

proteins

Fe–S cluster

B Bacterial SUF machinery SufA Apo Fe 2+ -S-SH sulphur HS-S- SufE SufU SufS -S-SH
B Bacterial SUF machinery
SufA
Apo
Fe 2+
-S-SH sulphur
HS-S-
SufE
SufU
SufS
-S-SH
HS-S-
(CsdB)
-
e
ATP
Holo
Cys
Ala
SufC
Bacterial
Fe–S
SufD
proteins
SufB
ATP

Fe–S clusterCys Ala SufC Bacterial Fe–S SufD proteins SufB ATP Figure 1.1 Schematic diagrams of bacterial ISC

Figure 1.1 Schematic diagrams of bacterial ISC (A) and SUF (B) systems. Grey and black circles indicate sulphur and iron, respectively. Only representative [2Fe 2S] clusters are shown. An electron is depicted by e . Apo and Holo indicate the apo- and holo-forms of Fe S cluster-containing proteins. Note that IscR is the transcriptional repressor of the ISC operon.

5.2.1 Genes encoded by ISC operon in bacteria In prokaryotes, the proteins involved in the ISC system are encoded by the ISC operon, which consists of iscRiscSiscUiscAhscB hscA fdx orf3 genes. The two central proteins of the ISC machinery are IscS, cysteine des- ulphurase, and IscU, a scaffold, which are homologous in function to NifS

IronSulphur Cluster Biogenesis in Protozoan Parasites

23

and NifU, respectively (see Section 5.4). Genes homologous to iscSUA , hscB /hscA and ddx are widely distributed in nature and can be found in all bacterial and archae-bacterial genomes that have been sequenced so far, often in the same arrangement as in E. coli . The operon has been shown to be essential for their viability (Tokumoto et al., 2004 ).

5.2.2 Mechanism of Fe S cluster synthesis by ISC system The conserved Cys328 of IscS participates in a nucleophilic attack on the sulphur atom of a PLP-bound cysteine substrate to form a cysteine per- sulphide residue, as revealed by structural studies (Cupp-Vickery et al., 2003; Kaiser et al., 2003). Furthermore, mutational analysis of residues pre- sent in a loop (Ser323Ala, Ser326Ala, Leu333Ala and Ser336Ala) close to the catalytic residue Cys328 demonstrated that these residues are necessary for in vivo enzymatic activities of Fe–S cluster-containing proteins synthesised by the ISC system (Lauhon et al., 2004 ). Sulphur transfer from IscS to IscU is initiated by the attack of the con- served cysteine residue, Cys63 of IscU on the S g atom of the cysteine per- sulphide residue transiently produced on IscS (Agar et al., 2000a). IscU and SufU (Isu in eukaryotes) also possess three conserved cysteine residues, which have similar functions, in a C-X 2426 -C-X 4243 -C arrangement. However, IscU and SufU show distinguishable structural properties. IscU lacks an 18–21-amino acid insertion between the second and third con- served cysteine residues of SufU. IscU has a conserved histidine in place of lysine in SufU, and possesses, unlike SufU, the LPPVK motif that is required for interaction with HscA chaperone (Cupp-Vickery et al., 2004a; Ramelot et al., 2004). Hence, dissimilar to SufU, IscU requires HscA/HscB cochaperones for Fe–S cluster assembly or transfer. IscR is a Fe–S protein in itself and a transcription factor to regulate the transcription of the ISC operon. Hence, IscR regulates ISC assembly machinery by a feedback mechanism (see Sections 3.1.5 and 3.2.1). IscA serves as an alternative scaffold. In Synechocystis , a [2Fe–2S] cluster was assembled in vitro between the two protomers of the IscA dimer and ligated by two conserved cysteine residues, Cys110 and Cys112, of both protomers (Wollenberg et al., 2003 ). In E. coli , thioredoxin reductase system promotes iron binding on IscA, but prevents it on IscU. On the other hand, Fe–S clus- ter assembly on IscU was promoted in the presence of iron-loaded IscA, IscS, and L-cysteine ( Ding et al., 2005). These data reinforce the view that IscA is capable of binding to the intracellular free iron and providing it for the IscS-mediated ISC assembly on IscU in cells in which the accessible free

24

Vahab Ali and Tomoyoshi Nozaki

iron content is probably restricted. HscB and HscA showed high degrees of homology to the molecular chaperones DnaJ/Hsp40 and DnaK/Hsp70, respectively. Therefore, HscA and HscB likely assist the maturation of Fe–S proteins. By site-directed mutational analysis, two cysteine residues in IscU were shown to be essential for binding with HscB, but not with IscS. The ferredoxin-containing [2Fe–2S] cluster encoded by the fdx gene is assumed to serve as a reductant at some point during the maturation of Fe–S clusters (Jung et al., 1999; Kakuta et al., 2001). The role of orf3 remains to be demonstrated. A general working model of the ISC system in bacteria is shown in Fig. 1.1A.

5.2.3 ISC system in the mitochondria In eukaryotes, the ISC assembly machinery in the mitochondria is required for the maturation of virtually all cellular Fe–S proteins, that is, mitochondrial as well as cytosolic proteins. The proteins involved in the ISC system are produced in the cytosol and transported to the mitochondria in yeasts and human (Gerber et al., 2004; Li et al., 2006; Lill and Muhlenhoff, 2006, 2008; Muhlenhoff and Lill, 2000; Muhlenhoff et al., 2003; Netz et al., 2007; Tong and Rouault, 2000, 2006). A current working model of the ISC assembly machinery in mitochondria is shown in Fig. 1.2. The biogen- esis of Fe–S clusters in the mitochondria involves the synthesis of a transient Fe–S cluster on the scaffold proteins Isu1 and Isu2 (Nif U/Suf U homo- logue), and the transfer and incorporation of Fe–S clusters into target apo-proteins, as seen in the bacterial ISC system. However, the process of Fe–S cluster biogenesis in the mitochondria is more complex than that in bacteria, requiring mitochondria-specific components. The process starts with the membrane potential-dependent import of ferrous iron into the mitochondria, which is facilitated by the carrier proteins Mrs3 and Mrs4 and unknown factors. Sulphur is released from L-cysteine by the cysteine des- ulphurase complex, Nfs1/Isd11 complex, in the form of sulphane sulphur. The reduction of sulphur needs electron transfer chain from NADH to fer- redoxin (Yah1) via ferredoxin reductase (Arh1). The Nfs1/Isd11 complex further interacts with IscU (Isu1/2) through a direct interaction between Nfs1 and Isu1/2. Iron binding to Isu1 is accom- plished by FXN (Yfh1, see Section 4.1.4; homologous to CyaY in bacteria), and the iron binding is enhanced with the Nfs1–Isu1 complex (Bencze et al., 2006; Gerber et al., 2003; Layer et al., 2006; Wang and Craig, 2008). It was also shown in bacteria that CyaY regulates Fe–S cluster biosynthesis by inhibiting IscS in an iron-dependent manner (Adinolfi et al., 2009).

IronSulphur Cluster Biogenesis in Protozoan Parasites

25

Mitochondrial ISC machinery Mitochondrion IscA1 /A2 Apo Isd11 ATP Nfs1 Sulphur Isu1/2 Ssq1(HscA, Hsp70) (IscS)
Mitochondrial ISC machinery
Mitochondrion
IscA1
/A2
Apo
Isd11
ATP
Nfs1
Sulphur
Isu1/2
Ssq1(HscA, Hsp70)
(IscS)
(IscU )
Jac1 (HscB), Grx5
-
Mge1(GrpE)
Cys
Ala
e
GSH
Fdx
-
(Yah1)
Holo
e
Frataxin
Fdx
Nfu
(Yfh1/Cya
reductase
Y)
Mitochondrial
(Arh1)
e - NADH
Fe–S proteins
Mrs3/4
?
Cytosol
Fe–S cluster
Fe 2+

Figure 1.2 A schematic diagram of mitochondrial ISC system. A transfer of sulphur/ sulphane is depicted by small arrows. Synonyms (in bacteria or other eukaryotes) are shown in parentheses. ? depicts an unknown iron transporter. See the legend of Fig. 1.1 for other symbols.

5.2.4 Isd11 Isd11 is unique to eukaryotes, absent in prokaryotes, and required for the biogenesis of Fe–S proteins in both the mitochondria and the cytosol. Isd11 functions closely with the a -proteobacterium-derived IscS. It was demonstrated that Isd11 is present in all five eukaryotic supergroups, includ- ing some of the hydrogenosomal and mitosomal lineages (Richards and van der Giezen, 2006) (see Section 8 ). These data indicate that the eukaryotic ancestor invented Isd11 as a functional partner to IscS, which supports the premise of a single shared a -proteobacterial endosymbiotic ancestry for all eukaryotes. Isd11 is also localised to the nucleus as well as the mitochondria in mam- malian cells. Thus, Isd11 is considered to be a unique subunit of mitochondrial ISC machinery (Adam et al., 2006; Wiedemann et al., 2006). It forms a com- plex with cysteine desulphurase (Nfs1) (Wiedemann et al., 2006). Isd11 seems to stabilise Nfs1 in the mitochondria and mediate the interaction of Nfs1 with other proteins (Adam et al., 2006). The Isd11 mutant yeast showed severe defect inthe Fe–S protein activities, and accumulation of iron in the mitochon- dria, indicating that Isd11 is essential for cell viability (Adam et al., 2006).

26

Vahab Ali and Tomoyoshi Nozaki

It was also shown that suppression of Isd11 enhanced the binding activity of IRP1 (see Section 3.2.1) to the iron-responsive element, increased the protein levels of iron regulatory protein 2 and resulted in abnormal punctate ferric iron accumulations in the cell. Isd11 of the Nfs1/IscU complex also interacts with FXN, as shown in co-immunoprecipitation and immunoflu- orescence assays (Shan et al., 2007). In addition, Isd11 was also shown to be involved in thiolation of both cytoplasmic and mitochondrial tRNA (Paris et al., 2010).

5.2.5 Additional proteins required for the cluster transfer in the mitochondria Several additional proteins are involved in the later stage of the Fe–S cluster transfer: the Hsp70 (homologous to bacterial HscA), Ssq1 (the co-chaperone of Hsp70), Jac1 (homologous to bacterial HscB), Mge1 (the nucleotide exchange factor, homogolous to bacterial GrpE) and glutaredoxin, Grx5. Jac1 facilitates the ATP-dependent binding of Ssq1 to Isu1/2 and stimulates the ATPase activity of Ssq1. Ssq1p interacts with a specific conserved PVK sequence on Isu1p (Hoff et al., 2002, 2003), which is located on an exposed loop between two a-helices. The carboxyl-terminal region of Jac1 is suffi- cient for interaction with Isu1. Isu1/Jac1 complex is targeted to Ssq1 (Andrew et al., 2006) and this targeting is suggested to be important under the conditions where need for Fe–S cluster biogenesis is high or in organisms lacking specialised Hsp70 for Fe–S cluster biogenesis. Glutaredoxins (Grx) are a group of thioltransferases that reduce disulphide bonds or catalyse reversible protein glutathionylation or deglutathionylation ( Herrero and de la Torre-Ruiz, 2007 ). A mitochondrial monothiol glutaredoxin, Grx5, also catalyses the reduction of disulphide bonds in proteins with a concomitant conversion of GSH to GSH- disulphide. It has been recently demonstrated that dimeric monothiol Grx can coordinate a [2Fe–2S] cluster via the cysteine residue of the active side of each monomer and the cysteines of two GSH molecules ( Picciocchi et al., 2007). Grx5 was shown to be required for Fe–S cluster biogenesis in yeast and zebrafish ( Rodriguez-Manzaneque et al., 2002; Wingert et al., 2005). Two chloroplastic monothiol Grx in plants were also shown to ligate a [2Fe–2S] cluster in vitro (Bandyopadhyay et al., 2008), suggesting that Grx may function in general as a scaffold in the organellar Fe–S cluster assembly and transfer. Grx5 assists Fe–S protein maturation by serving as a transient Fe–S cluster binding site before the cluster is inserted into apo-proteins (Bandyopadhyay et al., 2008). Recently, in the nitrogen-fixing bacterium

IronSulphur Cluster Biogenesis in Protozoan Parasites

27

Azotobacter vinelandii, it has been shown in vitro that HscA and HscB are required for efficient ATP-dependent [2Fe–2S] cluster transfer from IscU to Grx5, reinforcing the general role of Grx5 in the storage and transport of the [2Fe–2S] cluster assembled on IscU with the help of HscA and HscB (Shakamuri et al., 2012 ). However, this hypothesis has not been confirmed in vivo ( Lill et al., 2012). Mge1 is the nucleotide exchange factor involved in the ATP/ADP exchange, which triggers a conformational change of the binding domain of Ssq1 from a closed to an open state, thus leading to dis- assembly of the Ssq1–Isu1 complex.

5.2.6 The ISC export apparatus

5.2.6.1 Transport of Fe S clusters

Fe–S clusters synthesised in the mitochondria are also exported to the cytosol for the activation of cytosolic Fe–S apo-proteins. The currently known components of the specific export apparatus are: ATP-binding cassette (ABC) transporter, Atm1, which is localised on the mitochondrial inner membrane, Erv1, that is localised in the mitochondrial intermembrane space, and GSH, which is in the mitochondria and the cytoplasm (Fig. 1.3 ). Atm1 exports Fe–S clusters per se or an unknown form of iron from the mitochondria to the cytosol and is a key component of the Fe–S cluster export machinery (Csere et al., 1998; Pondarre et al., 2006; Savary et al., 1997). As shown for the yeast ABCB7-like homologue (Pondarre et al., 2006), Atm1 exposes its carboxyl-terminal ATP-binding domain to the mitochondrial matrix and spans the inner membrane with six trans-membrane segments. An analysis of ATM3 (plant Atm1 homo- logue) mutant plants indicated that the export of iron requires not only ATM3 but also one or two additional factors for the assembly of the Fe–S clusters as well as molybdenum cofactors (Moco) in the cytosol (Bernard et al., 2009 ). It was also shown that Atm1 plays a crucial role in the regulation of cellular iron uptake because Fe–S clusters or precursors transported by Atm1 interact with the iron-responsive transcription factors, Aft1 and Aft2 (see Section 3.2.1 ). Disruption of the ISC assembly in the mitochondria or the ISC export machinery caused the translocation of Aft1 and Aft2 into the nucleus and the induction of the iron regulon (Ueta et al., 2003 ). As described above, in A. thaliana, biosynthesis of Moco also depends on ATM3 (Teschner et al., 2010). The first step of Moco assembly takes place in the mitochondria, in which a pterin precursor is formed. This precursor is exported from the mitochondria via ATM3, and subsequent steps occur in

28

Vahab Ali and Tomoyoshi Nozaki

Mitochondrial ISC export machinery

Apo Holo ISC assembly Mitochondrial Fe–S proteins GSH ATP Atm1 Mia40 Erv1 - e e
Apo
Holo
ISC assembly
Mitochondrial
Fe–S proteins
GSH
ATP
Atm1
Mia40
Erv1
-
e
e -
NADPH
GSH FMN - e - e Tah18 Dre2 FAD Cytosol ? CIA machinery Apo Nar1
GSH
FMN
-
e -
e
Tah18
Dre2
FAD
Cytosol
?
CIA machinery
Apo
Nar1
-
e
Cfd1
Cfd
Nbp35
Nbp35
Cia1
Fe 2+
Nar1
Holo
Nar1 - e Cfd1 Cfd Nbp35 Nbp35 Cia1 Fe 2+ Nar1 Holo Fe–S cluster Cytosolic/ nuclear

Fe–S cluster

Cytosolic/

nuclear

Fe–S proteins

Figure 1.3 A schematic diagram of ISC export machinery and CIA system. See the legends of Figs. 1.1 and 1.2 for other symbols.

the cytosol, including a sulphur insertion. Thus, analysis of ATM3 in plants and protozoa may help the understanding of the precise role of the mito- chondrial ABC transporters in Fe–S clusters and Moco biosynthesis as the Moco biosynthesis and Moco-containing enzymes are lacking in yeast (Balk and Pilon, 2010 ).

5.2.6.2 Reduction and oxidation of Fe S clusters in the ISC export system

Erv1, FAD-binding sulphhydryl oxidase, catalyses the formation of dis- ulphide bridges (Lee et al., 2000) and is involved in the disulphide

IronSulphur Cluster Biogenesis in Protozoan Parasites

29

formation-driven transport of proteins into the inter-membrane space by reoxidizing the import component Mia40 (Allen et al., 2005; Rissler et al., 2005; Terziyska et al., 2005). Mia40 is an oxidoreductase that catalyses oxidative protein folding in the mitochondria. Mia40 has a characteristic conserved CPC motif, which can donate the disulphide bond to the sub- strate ( Banci et al., 2009 ). Mia40, in cooperation with Erv1, promotes the formation of two disulphide bonds in the substrate protein. Thus, the disulphide bond formation pathway is based on a relay of reactions involving disulphide transfer from Erv1 to Mia40 and from Mia40 to the substrate pro- teins, ensuring the efficiency of oxidative folding in the inter-membrane space (Bottinger and Becker, 2012 ). An electron derived from the sul- phhydryl oxidation is presumed to be passed by Erv1 onto either molecular oxygen or cytochrome c (Thorpe and Coppock, 2007). Thus, Erv1 appears to play two key roles: oxidation of Fe–S clusters and drainage of electrons produced via Fe–S cluster transport. Although GSH is necessary to coordi- nate Fe–S cluster binding, it remains unknown how GSH helps a cluster transfer from the mitochondria to the cytoplasm.

5.3. SUF machinery

5.3.1 Catalytic components of SUF machinery in bacteria

The second machinery involved in Fe–S cluster biogenesis is known as the SUF system (Fig. 1.1B). In bacteria, the SUF gene cluster comprises sufA, B , C, D , S , and E genes (Takahashi and Tokumoto, 2002 ), which are regu- lated by the Fe 2 þ -dependent repressor, Fur. SufS (also called CsdB) is a cysteine desulphurase, whereas SufE is an activator of cysteine desulphurase. Suf S and Suf E form a complex, in which SufE accepts sulphane sulphur from Suf S.

5.3.2 Scaffold components of SUF machinery in bacteria

There are three apparently redundant scaffold components: SufA, SufBCD and SufU. SufA serves as a scaffold component, and binds to the sulphurated form of SufE exclusively via the three conserved cysteine residues, Cys50, Cys114 and Cys116, as demonstrated by site-directed mutagenesis. The reaction consists of a transfer of S atoms from persulphide/polysulphide spe- cies on SufE to the three conserved cysteines of SufA. In E. coli , it has been shown that the SufBC2D complex can assemble a [4Fe–4S] cluster and transfer it to apo-proteins in vitro . This complex assem- bles an Fe–S cluster through the mobilization of sulphur from the SufSE complex and the FADH 2 -dependent reductive mobilization of iron

30

Vahab Ali and Tomoyoshi Nozaki

(Wollers et al., 2010 ). SufC is an ABC ATPase and forms a complex with SufB and SufD. In E. coli , the carboxyl-terminal a -helices of SufD interact with SufC (Wada et al., 2009). SufU, (Isu in eukaryotes) dissimilar to IscU, serves as an alternative scaf- fold protein. SufU possesses three conserved cysteine residues in a C-X 2426 - C-X 4243 -C arrangement. However, SufU has an 18–21-amino acid inser- tion between the second and third conserved cysteine residues, which is missing in IscU. In addition, SufU lacks the conserved histidine (replaced with lysine), and the LPPVK motif that is required for interaction of IscU with HscA chaperone ( Cupp-Vickery et al., 2004b; Ramelot et al., 2004), suggesting lack of interaction with HscA/HscB chaperones of SufU for Fe–S cluster assembly. SufU can assemble the [2Fe–2S] cluster and trans- fer it to the apo-proteins. In contrast, SufA can hold both [2Fe–2S] and [4Fe–4S] clusters and transfer them to the apo-proteins (Gupta et al., 2009; Tan et al., 2009 ).

5.3.3 Plastidial (chloroplastic) SUF system In organisms other than well-studied bacteria such as E. coli (Takahashi and Tokumoto, 2002) and Erwinia chrysanthemi ( Nachin et al., 2001 ), the SUF system is also found in the chloroplasts of plants. The components and reac- tions involved in the Fe–S cluster assembly in the plastids are shown in Fig. 1.4. The plastids including the apicoplasts, found in the phylum Apicomplexa, have evolutionarily arisen via one to a few rounds of endo- symbiosis of an ancestor of the extant cyanobacteria, which possess the SUF system and lack the ISC system (Takahashi and Tokumoto, 2002). Plastids are thus expected to possess the SUF proteins for the house-keeping role of Fe–S clusters assembly in the organelle. The plastids generate Fe–S clusters for their own requirements, whereas the mitochondria provide the clusters for the cytosolic and nuclear Fe–S proteins (Balk and Pilon, 2010 ). In A. thaliana, all the components of SufABCDSE are conserved and involved in Fe–S cluster biogenesis in the plastids along with plant-specific compo- nents such as cytochrome b 6 /f complex, [2Fe–2S] cluster-containing ferre- doxin, PSI, [4Fe–4S] cluster-containing ferredoxin–thioredoxin reductase and proteins involved in nitrogen assimilation pathways (nitrite reductase, glutamate synthatase, and sulphite reductase) (Ye et al., 2006 ). These plant-specific components in the SUF system may explain the ability of plants to carry out Fe–S cluster assembly in the presence of oxygen. In the chloroplast stroma, chloroplastic (Cp) NifS (also known as

AtNFS2 or CpSuf S),

CpSuf E, and CpIscA form a 600-kDa complex

IronSulphur Cluster Biogenesis in Protozoan Parasites

31

Plastidial SUF machinery Cytoplasm Fe 2+ Plastid Photo e - e - Fdx NADPH e
Plastidial SUF machinery
Cytoplasm
Fe 2+
Plastid
Photo
e -
e -
Fdx
NADPH
e -
? system I
Nfu
Apo
e -
ATP
ADP
SufA
Sulphur
SufD
HS-S- SufE
IscA?
SufB
Grx
HS-S- NFS2
SufC
SufC
(CpNifS/CpSufS)
Holo
FADH 2
ATP
ADP
Ala
Cys
FAD +

Figure 1.4 A schematic diagram of plastidial SUF system. ?depicts an unknown iron transporter. See the legends of Figs. 1.1 and 1.2 for other symbols. Interaction of IscA in vivo not confirmed.

(Ye et al., 2006). CpNif S is closest to Suf S among three Nif S-like proteins in E. coli . CpSufE (At4g26500) is an essential protein and forms a hetero- tetrameric subcomplex with CpNif S. CpSufE1 stimulates cysteine des- ulphurase activity 40-fold and increases the substrate affinity of CpNif S towards cysteine. An essential cysteine residue, Cys65, in CpSufE, which is composed of the SufE- and BolA-like domains, is likely the acceptor site for the intermediate sulphur, but is not required for binding to CpNif S. The BolA-like domain of CpSuf E is also presumed to interact with Grx in the chloroplasts ( Huynen et al., 2005 ). The components, their localiza- tion, and function of the SUF system in plants were reviewed (Balk and Lobreaux, 2005). Similar to bacteria, in the Arabidopsis chloroplasts, SufB was shown to interact with SufC (Xu et al., 2005), which in turn interacts with Suf D, as demonstrated by a yeast two-hybrid system and bimolecular fluorescence studies (Xu and Moller, 2004 ). It was shown using isolated chloroplasts that the Suf BCD complex requires ATP for Fe–S assembly, as in E. coli ( Wollers et al., 2010). Both Arabidopsis Suf B and SufC showed ATPase activity (Xu and Moller, 2004; Xu et al., 2005). In addition, plant SufB and SufC partially complemented the respective E. coli mutants.

32

Vahab Ali and Tomoyoshi Nozaki

It is important to note that the mitochondrial ISC and plastidial SUF machineries synthesise their Fe–S clusters independently. The inhibition of Suf S (NFS2) by RNA interference in A. thaliana caused defects in PSI and decreased the levels of all tested chloroplast Fe–S proteins. However, no defect was seen in the amount and functions of the mitochondrial Fe–S proteins (Balk and Pilon, 2010; Ye et al., 2006 ).

5.4. NIF machinery

The NIF system was discovered first in the nitrogen-fixing bacterium A. vinelandii and is thought to be responsible for the assembly of the Fe–S clusters as well as Fe–Mo clusters, required for activation of nitrogenase, which is the enzyme responsible for the conversion of nitrogen gas to ammonia (see review by Rees and Howard, 2000). The main components of the NIF system are Nif S and Nif U. The operon of the NIF gene cluster was identified ( Jacobson et al., 1989 ) and NifS was identified as a cysteine desulphurase in A. vinelandii ( Zheng et al., 1993 ). The NIF machinery is present in bacteria such as nitrogen-fixing bacteria, non-diazotropic e -proteobacteria such as Campylobacter jejuni and Helicobacter pylori (Olson et al., 2000 ), and a protozoan E. histolytica (Ali et al., 2004). The character- ization of the NIF system in non-diazotrophs enforced a view on the premise that the NIF system is required for Fe–S cluster assembly of nitro- genase. All the organisms that possess the NIF system are microaerophilic or anaerobic and restricted to limited lineages.

5.4.1 NifS and NifU NifS is the homodimeric pyridoxal phosphate-containing catalytic compo- nent with cysteine desulphurase activity. Nif S releases the sulphur atom from L -cysteine, which involves a formation of cysteine-PLP ketimine adducts with a subsequent nucleophilic attack by the thiolate anion of Cys325 on the sulphur of the substrate cysteine. An intermediate persulphide is formed in the reaction to be incorporated as a transient [2Fe–2S] cluster on the scaffold component, Nif U (Yuvaniyama et al., 2000). The specific role of Nif S in the maturation of nitrogenases is supported by the fact that NifS is co-transcribed with a CysE homologue that encodes serine O -acetyltransferase, which catalyses the rate-limiting step in cysteine biosynthesis. Nif U is a homodimeric scaffold protein on which the Fe–S clusters are preassembled before insertion into nitrogenase. Nif U holds a [2Fe–2S] clus- ter in each protomer and possesses three domains, that is, the amino-terminal,

IronSulphur Cluster Biogenesis in Protozoan Parasites

33

central, and carboxyl-terminal domains. The amino-terminal domain is sim- ilar to IscU of the ISC system and contains three conserved cysteine residues, Cys35, Cys62 and Cys106, which provide a labile rubredoxin-like ferric- binding site or a transient [2Fe–2S] cluster. This domain also possesses the LPPVK (or LPPEK in A. vinelandii Nif U) motif responsible for the IscU–cochaperone interaction (see below) ( Hoff et al., 2002; Johnson et al., 2005). However, the pentapeptide motif is not conserved in other organisms. The central domain of A. vinelandii Nif U contains a permanent, redox-active [2Fe–2S] cluster ligated by Cys137, Cys139, Cys172 and Cys175 residues with sequence homology to bacterioferritin-associated [2Fe–2S] ferredoxin from E. coli ( Agar et al., 2000b; Yuvaniyama et al., 2000). It was presumed that the domain helps in the formation and/or release of the transient Fe–S clusters formed on the amino- and carboxyl-terminal domains by serving as an iron donor for the Fe–S cluster assembly ( Agar et al., 2000b). The carboxyl-terminal domain of Nif U as well as Nif U-like proteins in eukaryotes, termed as Nfu or Cnf U, contains two conserved cys- teine residues, Cys272 and Cys275, and can assemble a labile [2Fe–2S] cluster for the transfer to nitrogenases (Agar et al., 2000b). The [2Fe–2S] cluster was demonstrated in the purified Nfu protein from Synechocystis (Tong et al., 2003; Yabe et al., 2004). The [4Fe–4S] cluster was also verified on the human Nfu protein by absorption, analytical, and Mo¨ ssbauer studies (Tong et al., 2003). Furthermore, genetic and biochemical studies indicate that both the amino- and carboxyl-terminal domains are involved in the assembly of a [4Fe–4S] cluster on apo-nitrogenase (Dos Santos et al., 2004). A recent study further explained that Nif U sequentially assembles a labile [2Fe–2S] cluster in the amino-terminal IscU-like domain, and labile [4Fe–4S] clusters in the carboxyl-terminal Nfu-like domain, and that these labile [4Fe–4S] clusters are rapidly transferred to apo-nitrogenase (Smith et al., 2005 ). Thus, Nif U can serve as a scaffold for Fe–S cluster assembly per se . Nif S and Nif U are uniquely present in E. histolytica among parasitic protozoa ( Ali et al., 2004; van der Giezen et al., 2004 ) (see Section 8.1).

5.4.2 Alternative scaffold Nif IscA Nif IscA shows considerable sequence similarity to IscA, and its encoding gene is located immediately upstream of nifU in A. vinelandii (Johnson et al., 2005). Nif IscA is an alternative scaffold protein that is involved in the assembly of transient Fe–S clusters (Krebs et al., 2001; Morimoto et al., 2006 ) or the delivery of iron to U-type scaffold proteins as a metal- lochaperone ( Ding et al., 2004). Spectroscopic studies showed that

34

Vahab Ali and Tomoyoshi Nozaki

A. vinelandii Nif IscA is a homodimeric protein and can assemble [2Fe–2S] and [4Fe–4S] clusters between the two protomers, mediated by Nif S (Krebs et al., 2001 ). The possibility that Nif IscA is exclusively involved in the Fe–S cluster formation of nitrogenase was not appreciated until the dis- covery of IscA because there was no phenotype associated with the inacti- vation of Nif iscA . Nevertheless, the conservation of three cysteines in both IscA and Nif IscA was interpreted to indicate that this family of proteins play an analogous role in Fe–S cluster biosynthesis.

5.5. CIA machinery

5.5.1 Four identified components in CIA machinery The additional cytosolic machinery for the Fe–S cluster assembly was first identified in yeast and named as the cytosolic iron-sulphur cluster assembly (CIA) machinery (Lill and Muhlenhoff, 2005; Roy et al., 2003). The CIA machinery was shown to be required for the maturation of cytosolic and nuclear Fe–S proteins (Hausmann et al., 2005 ). The CIA system is present in most eukaryotes sequenced to date ( Hausmann et al., 2005), located in both the cytosol and nucleus, and expressed at comparatively low levels (Balk et al., 2004; Roy et al., 2003). To date, functional evidence for the assembly and maturation of four cytosolic and nuclear Fe–S proteins have been demonstrated in yeast ( Netz et al., 2007; Sharma et al., 2010). The main components elucidated till date are the soluble P-loop NTPases (ATPases or GTPases), Cfd1 and Nbp35 (homologous to bacterial ApbC and yeast Ind1), the iron-only hydrogenase-like protein, nuclear architecture-related protein 1 (Nar1) and the seven-bladed WD40 repeat protein, Cia1, all of which work in a coordinated fashion for the transfer of a Fe–S cluster to apo-proteins ( Fig. 1.3). First, an Fe–S cluster is transiently assembled on the P-loop NTPases Cfd1 and Nbp35, which serve as a scaf- fold. In vitro reconstitution studies showed that an oligomeric complex con- sisted of Nbp35 and Cfd1 assembles labile Fe–S clusters on Nbp35. Thus, these cysteine residues likely serve as a transient scaffold for the Fe–S cluster before transfer to apo-proteins (Broderick, 2007; Netz et al., 2007 ). This step essentially requires mitochondrial Nfs protein. Unlike the mitochon- drial Isu1 scaffold, Cfd1 and Nbp35 do not directly interact with a sulphur-donating protein such as Nfs1 (Biederbick et al., 2006; Muhlenhoff et al., 2004; Nakai et al., 2001). Instead, the Cfd1 and Nbp35 complex interacts with Nar1 for the transfer of its clusters. The Fe–S clusters on the Cfd1 and Nbp35 scaffold are subsequently transferred to an apo-protein by Nar1 and Cia1. Nar1, in contrast to the

IronSulphur Cluster Biogenesis in Protozoan Parasites

35

Nbp35 and Cfd1 complex, possesses two interacting Fe–S clusters, and interacts with the Nbp35–Cfd1 complex. Cia1 also makes a complex with

Nar1, holds the Fe–S cluster and is proposed to be involved in the final trans-

fer of the cluster to apo-proteins. Important aspects of the CIA machinery

such as the source of sulphide and iron, and detailed mechanisms of the Fe–S cluster assembly remain unclear. The four components of the CIA machin-

ery are described in the following section.

5.5.2 Cytosolic Fe S cluster-deficient protein 1

Cfd1 (Cfd1p in yeast) was the first protein shown to be involved in the cyto- solic Fe–S cluster assembly (Roy et al., 2003 ). Cfd1p shares similarities with, and falls into, the subfamily of P-loop ATPases, which include Nif H of nitrogenase ( Dean et al., 1993; Leipe et al., 2002). Nif H itself is a Fe–S pro-

tein; however, the cluster-coordinating residues of Nif H are not conserved

in Cfd1p. Cfd is more closely related to Mrp from bacteria, which is also

involved in the Fe–S cluster assembly (Skovran and Downs, 2003 ). Cfd1p functions to make an iron available for the cytosolic Fe–S cluster assembly.

However, it remains unknown whether iron is provided from the mito-

chondria or acquired from other cellular pools (Roy et al., 2003). By analogy

to known metallochaperones, the metal-binding element of an iron chap-

erone is likely at, or near, the protein surface (Arnesano et al., 2002). Cfd1p

possesses the CX2CX2C motif exposed on the surface and this motif is predicted to be involved in metal binding (Roy et al., 2003 ). Cellular frac- tionation of yeast showed that > 90% of Cfd1 is present in the cytoplasm while about 10% is in the membrane-rich fraction. It was shown that cyto- solic aconitase in IRP1-transformed cells contained 80% [4Fe–4S] and 20% [3Fe–4S] clusters (Brown et al., 2002 ), while Cfd1p-deficient IRP1- transformed strain showed no or little detectable [3Fe–4S]- or [4Fe–4S]- containing cytosolic aconitase activity, suggesting that Cfd1p is essential for de novo assembly of the Fe–S clusters on aconitase in the cytosol ( Roy et al., 2003).

5.5.3 Nucleotide-binding protein 35

Nbp35p is a soluble protein located in both the cytosol and the nucleus, but not the mitochondria (Hausmann et al., 2005 ). Nbp35 belongs to the Mrp/ Nbp35 subgroup of the P-loop NTPases that share the highest sequence similarity with members of other subgroups involved in NIF, metal inser-

tion, and bacterial cell division (Lutkenhaus and Sundaramoorthy, 2003).

A feature of Nbp35 is an approximately 50-a.a.-long amino-terminal

36

Vahab Ali and Tomoyoshi Nozaki

extension containing four conserved cysteine residues, which Mrp lacks, and is likely a binding site for either a metal (e.g. zinc) or the Fe–S cluster ( Leipe et al., 2002). Nbp35 indeed carries the Fe–S cluster at the amino-terminal domain. Cluster integration of Nbp35p depends strictly on the mitochon- drial ISC assembly and export machineries. Thus, maturation of Nbp35p occurs in a similar manner to other cytosolic Fe–S proteins. However, Nbp35p differs from similar NTPases yeast Cfd1p and bacterial ApbC in that Nbp35p, but not the latter, stably binds to iron or the Fe–S cluster ( Roy et al., 2003; Skovran and Downs, 2003). As predicted by the striking structural similarities between Nbp35p and Cfd1p, the two proteins are, at least in part, functionally and genet- ically interchangeable. Cfd1 and N bp35 have a genetic interaction. Nbp35p overproduction exacerbat ed the growth defect of Cfd1p- deficient cells ( Hausmann et al., 2005 ). It was shown that Nbp35p stably binds iron in vivo . As described above, Nbp35 does not directly interact with sulphur-donating protein, cytosolic Nfs1. Instead, mitochondrial Nfs1, together with other mitochondrial ISC and ISC exports compo- nents, is required for the cytosolic Fe–S cluster formation by the CIA sys- tem ( Netz et al., 2007 ). Hence, Cfd1 and Nbp35 appear to receive the sulphur moiety from mito chondrial Nfs1. Cfd1 and Nbp35 form a stable complex in vivo , which suggests that their cooperation is of functional importance. The Cfd1–Nbp35 complex binds up to three [4Fe–4S] clus- ters, one at the amino terminus of Nbp35 and one each in the carboxyl- terminal regions of both Nbp35 and Cfd1. These clusters are labile and can be rapidly transferred to target [Fe–S] apo-proteins. A recent muta- tional study of Cfd1 and Nbp35 has shown that two central cysteine res- idues (CPXC) of the carboxyl-terminal motif are crucial for the coordination of the labile [4Fe–4S] clusters and the formation of the Cfd1–Nbp35 heterotetramer complex, as well as their functions as scaf- fold proteins and the viability of the cells. In contrast, the proximal cys- teine residues are dispensable, despite their evolutionary conservation ( Netz et al., 2012 ).

5.5.4 Nuclear architecture-related protein 1 Nar1 (or Nar1p in yeast) is an Fe–S protein containing two [4Fe–4S] clusters (Horner et al., 2002; Nicolet et al., 2002) and acts at the late stage of the cytosolic Fe–S biogenesis, after the assembly of the Fe–S cluster on the Cfd1–Nbp35 scaffold complex. The Cfd1- and Nbp35-bound Fe–S clusters are transferred to Nar1, which binds two Fe–S clusters and thereby is

IronSulphur Cluster Biogenesis in Protozoan Parasites

37

converted into a functional CIA component. Nar1 has high sequence sim- ilarity with bacterial iron-only hydrogenase found in virtually all organisms including those that are not known to produce or consume hydrogen such as yeast and humans (Balk et al., 2004; Horner et al., 2002). Both the [4Fe–4S] clusters on Nar1 are essential for function and cell viability (Urzica et al., 2009). At the amino terminus, Nar1p carries a ferredoxin-like domain with four conserved cysteine residues that may bind a [4Fe–4S] cluster (Balk et al., 2004), while the carboxyl-terminal part contains four conserved cysteine residues that, in iron-only hydrogenases, hold a unique H-cluster. This moi- ety consists of two sub-clusters, a cubane [4Fe–4S] cluster and a binuclear [2Fe] centre bridged by a cysteine sulphur ( Nicolet et al., 2000 ). This H-cluster forms the catalytic centre of iron-only hydrogenases (Nicolet et al., 2000). The closest human homologue of Nar1p is NARF, which is a nuclear protein that binds to prenylated lamin A in the nucleus (Barton and Worman, 1999). IOP1 is the second closest mammalian homologue of Nar1p ( Song and Lee, 2007 ), which regulates the expression of the hypoxia- inducible factor that represents the global mediator of the mammalian response to hypoxia. Neither IOP1 nor NARF possesses hydrogenase activ- ities (Huang et al., 2007 ). Nar1 is involved in the transfer of the Fe–S cluster from the scaffold complex to the target apo-proteins. Nar1p is predomi- nantly localised in the cytosol, where it interacts specifically with Cia1 (see Section 5.5.5), which is mainly located in the nucleus ( Balk et al., 2004). Depletion of Nar1p in the cytosol caused defects in the enzymatic activity of the cytosolic Fe–S proteins. Mitochondrial Fe–S proteins, on the other hand, exhibited normal activity and showed wild-type efficiencies in the Fe–S cluster assembly (Balk et al., 2004 ).

5.5.5 CIA machinery protein 1 Cia1 acts after the assembly of the Fe–S clusters on Nbp35 and Nar1, and thus, likely plays a role in the final incorporation of the Fe–S clusters into target proteins. Cia1 belongs to the large family of WD40 proteins and exhibits a seven-bladed propeller structure. These proteins function in rather diverse processes and generally act as protein-interaction devices (Smith et al., 1999 ). It was shown that Cia1 interacts with Nar1 in vivo (Balk and Lobreaux, 2005). Depletion of Cia1 impaired maturation of cytosolic and nuclear Fe–S proteins, but did not reduce iron binding to Nbp35 or Nar1, suggesting that Cia1 acts in the transfer of Fe–S cluster from Nbp35 to apo-proteins (Balk and Lobreaux, 2005 ).

38

Vahab Ali and Tomoyoshi Nozaki

5.5.6 Dre2 and Tah18 Dre2 contains a [4Fe–4S] cluster and a [2Fe–2S] cluster ( Zhang et al., 2008 ). These clusters are stable even after prolonged exposure to air, suggesting that they play structural and catalytic roles in the protein. Depletion of Dre2 impaired cytosolic, but not mitochondrial, Fe–S cluster biogenesis, although Dre2 was found partially localised in the mitochondrial inter-membrane space (Zhang et al., 2008 ). It is possible that Dre2 is likely involved in an early step in the cytosolic Fe–S cluster biogenesis, possibly worked with association of the ISC export system to deliver a substrate necessary for the Fe–S cluster formation on Cfd1 and Nbp35. Recently, a new essential flavoprotein Tah18 was discovered in yeast. Tah18 and Dre2 form a com- plex and is part of an electron transfer chain functioning in an early step of the cytosolic Fe–S protein biogenesis. Electrons are transferred from NADPH via the FAD- and FMN-containing Tah18 to the Fe–S clusters of Dre2. This electron transfer chain is required for the assembly of the tar- get, but not the scaffold Fe–S proteins, suggesting a need for the reducing power in the generation of the stably inserted Fe–S clusters. It was also shown that human Ndor1–Ciapin1 proteins can functionally replace yeast Tah18–Dre2 ( Netz et al., 2010 ).

5.6. Mechanism of repair of Fe S clusters

In general, Fe–S clusters are sensitive to oxidative stress. For instance, [4Fe–4S] 2 þ clusters of dehydratases are rapidly damaged by univalent oxi- dants, including hydrogen peroxide, superoxide, and peroxynitrite. The loss of an electron destabilises the cluster, causing it to release its catalytic iron atom and converting the cluster initially to an inactive [3Fe–4S] 1 þ form (Djaman et al., 2004 ). Chaperones such as HscA and HscB (Hsp70- and Jac1-type chaperones, respectively), encoded in the ISC gene cluster in E. coli , which are necessary for the Fe–S cluster biogenesis, may also be involved in the repair of damaged Fe–S clusters (Schilke et al., 1999 ).

repair of damaged Fe–S clusters ( Schilke et al., 1999 ). 6. GENETIC DISORDERS BY A

6. GENETIC DISORDERS BY A DEFECT OF FeS CLUSTER BIOGENESIS

Several genetic diseases have been attributed to mutations in genes involved in the Fe–S clusters assembly. Although many genes have been proven responsible for diseases in humans, such as FRDA, microcytic anae- mia, X-linked sideroblastic anaemia, cellular ataxia, xeroderma pigmenta- tion, Fanconi anaemia, colon cancer, erythropoietic anaemia, and diseases

IronSulphur Cluster Biogenesis in Protozoan Parasites

39

due to the disturbance of the complexes I and II (Lill and Muhlenhoff, 2008; Ye and Rouault, 2010), only a few examples are described here. For other genetic disorders caused by Fe–S cluster biogenesis, one may refer to the above-mentioned reviews.

6.1. Friedreich s ataxia

FRDA is a neurodegenerative disease, which is caused by a mutation in the FXN gene that is involved in the transport of iron. Numerous reports have indicated the roles of FXN, such as that of an iron chaperone (Bulteau et al., 2004), an iron donor for haem biosynthesis (Yoon and Cowan, 2004 ), an iron storage gene ( Gakh et al., 2002), an iron donor forthe Fe–S cluster repair and a regulator of the iron export (Bulteau et al., 2004; Radisky et al., 1999). How- ever, the main role of FXN is currently believed to be supplying iron in a bioavailable form for mitochondrial Fe–S cluster biosynthesis. FRDA is the most common form of inherited ataxia in humans. FRDA involves the central and peripheral nervous systems and the heart, and is characterised by a progressive gait and limb ataxia, a lack of tendon reflexes in the legs, loss of position sense, dysarthria, and leg weakness ( Campuzano et al., 1996 ). Neurons in the dorsal root ganglia are adversely affected, and hypertrophic cardiomyopathy is found in almost all patients (De Biase et al., 2007). FRDA is common before the age of 25, and most commonly in puberty. So far, little success has been achieved in the treatment of FRDA (Ye and Rouault, 2010 ). The excess iron accumulation is observed in the mitochondria of cardiac myocytes and neurons of affected tissues (Rotig et al., 2002; Seznec et al., 2005). The activities of Fe–S proteins such as aconitase and succinate dehydrogenase are reduced in FRDA patient cells. Excess iron is imported to and accumulated in the mitochondria, which leads to iron deficiency in the cytosol, activation of the iron regulatory pro- teins and the impairment of cellular iron homeostasis (Huang et al., 2009; Li et al., 2008). It was observed that about 2% of FRDA patients have point mutations in the FXN gene, whereas 98% have alterations in the FXN gene such as a mul- tiple insertion of unstable GAA trinucleotides in the first intron ( Campuzano et al., 1996). Normal alleles contain 6–34 of the GAA repeats, whereas the alleles of patients contain 66–1700 repeats. The expansion of the GAA repeat appears to cause gene silencing by allowing heterochromatin forma- tion to silence transcription or to inhibit transcription of the FXN gene due to the formation of a persistent DNA–RNA hybrid ( Wells, 2008).

40

Vahab Ali and Tomoyoshi Nozaki

6.2. Sideroblastic anaemia

A mutation in Grx (GLRX5) was discovered to be responsible for micro-

cytic sideroblastic anaemia (SA) in one human male ( Camaschella et al., 2007). The GLRX5 mutation was associated with severe SA, insulin- dependent diabetes and cirrhosis with high levels of blood transferrin satu- ration, serum ferritin, bilirubin, liver transaminases, urinary iron and liver iron concentrations. The patient was successfully treated with iron chelation (Camaschella et al., 2007). SA is characterised by the presence of ring sideroblasts, which represent erythroid precursor cells due to the iron over- load in the mitochondria. The SA patient has a homozygous A294G mutation in the third nucle- otide of the last codon of GLRX5 exon 1. This substitution does not change the encoded amino acid (glutamine), but interferes with the correct splicing and removal of intron 1 and drastically reduces the amount of the grx5 tran- script ( Camaschella et al., 2007). GLRX5 is involved in the Fe–S cluster bio-

synthesis in the mitochondria as a scaffold protein, the GLRX5 mutation likely results in a defect in the Fe–S cluster biogenesis in multiple organs and tissues. Interestingly, however, pathologic phenotypes are not devel- oped in all tissues, but restricted to erythrocytes and haematopoietic cells (Ye and Rouault, 2010 ).

6.3. X-linked sideroblastic anaemia and ataxia (XLSA/A)

Mutations of the ABCB7 gene in humans cause hereditary X-linked sideroblastic anaemia and ataxia (XLSA/A) ( Lill and Kispal, 2001 ). The ABCB7 gene encodes the human homologue of Atm1, which is involved

in the export of the Fe–S clusters or their precursors synthesised in the mito-

chondria to the cytosol (see Section 5.2.6 ). The mutations in the human ABCB7 gene generally occur in trans-membrane segments, indicating the importance of the integrity of the membrane-spanning regions. In plants, cyclic pyranopterin monophosphate (cPMP), a Moco synthesis intermedi- ate, was shown to be accumulated in the mitochondria of an ATM3 mutant (Teschner et al., 2010 ). ATM3 was proposed to be a cPMP transporter. Similar SA phenotypes are attributable to mutations of the related genes glycine (Solute carrier family 25, member 38; SLC25A38) and thiamine (SLC19A2 ) transport genes ( Guernsey et al., 2009), a gene encoding an ABC transporter for the export of Fe–S clusters to the cytoplasm (ABCB7) and a gene involved in haem synthesis (ALAS2) (Bergmann et al., 2010). It is predicted that SLC25A38 provides glycine for haem biosynthesis, while

IronSulphur Cluster Biogenesis in Protozoan Parasites

41

SLC19A2 provides thiamine for pyruvate dehydrogenase. ABCB7 exports a haem or haem precursor to the cytosol that conveys a signal regarding the status of the mitochondria to the cytosolic/nuclear compartment (Ye and Rouault, 2010).

6.4. Other genetic disorders

In Sjo¨ gren’s syndrome, IscA, an alternative scaffold protein, was identified as an autoantigen ( Cozar-Castellano et al., 2004). However, it remains unknown how IscA is associated with the disease. It was also demonstrated that mitochondrial respiratory chain enzymes, particularly succinate dehy- drogenase (complex II), function as a tumour suppressor (Gottlieb and Tomlinson, 2005). The impaired function of complex II leads to an accu- mulation of succinate, which inhibits propyl hydroxylase in the cytosol. Defects in complex I (NADH-ubiquinone oxidoreductase) are associated with a number of mitochondrial diseases, including mitochondrial encephalomyopathy, lactic acidosis, stroke-like episodes, Leigh syndrome and Leber’s hereditary optic atrophy (see DiMauro, 2004 for details). Inherited mutations in ATP-dependent DNA helicase XPD are found in cluster-coordinated residues in patients and shown to be associated with the disease known as Xeroderma pigmentation, Cockayne syndrome, and tri- chothiodystrophy ( Andressoo et al., 2006 ). XPD and related proteins, such as FancJ, show a striking similarity to archae-bacterial helicase, an Fe–S pro- tein. Similarly, FancJ mutations are associated with Fanconi anaemia patients and defects in DNA glycosylase (MutY homologue), and are perhaps asso- ciated with colon cancer ( Lukianova and David, 2005; Peterlongo et al., 2006; Rouault, 2012; Ye and Rouault, 2010).

et al., 2006; Rouault, 2012; Ye and Rouault, 2010 ). 7. OUTLINE OF CONSERVATION, UNIQUE DISTRIBUTION,

7. OUTLINE OF CONSERVATION, UNIQUE DISTRIBUTION, AND DIVERSITY OF Fe S CLUSTER BIOGENESIS MACHINERIES IN PROTOZOAN PARASITES

All protozoan parasites analysed so far are known to possess one to three independent machineries for the biosynthesis of Fe–S clusters described in Section 5. The majority of protozoan parasites are similar in that they possess the ISC system including the ISC export and cytosolic CIA machineries. However, there are a few exceptions: E. histolytica lacks the ISC and SUF systems, and instead possesses the NIF system along with lim- ited components of the CIA system. Plasmodium and Blastocystis possess both ISC and SUF systems, as well as the CIA machinery. In other parasitic

42

Vahab Ali and Tomoyoshi Nozaki

protozoa reviewed in this chapter, the ISC and CIA systems are conserved as seen in the higher vertebrates. Although Plasmodium and Cryptosporidium belong to the same phylum Apicomplexa, they are remarkably different in Fe–S cluster biogenesis. Plas- modium species conserve the apicoplast-compartmentalized SUF system and the mitochondrion-compartmentalized ISC system, while Cryptosporidium lacks the SUF system and retains only a limited number of proteins for the ISC system. In contrast to Apicomplexa, at least two species of trypano- somes and four species of Leishmania that belong to the order Kinetoplastida and are discussed in this review are similar in the Fe–S cluster biogenesis

machineries. Blastocystis is also very unique in that the SUF system is uniquely localised in the cytoplasm, while the ISC system is localised in the mitochondrion-related organelle [MRO or mitochondrion (mitochondrial)-like organelle], similar to the case in Giardia and Cryptospo- ridium , and in Trichomonas (the hydrogenosomes, instead of the mitosomes).

It is of importance that among the ISC, SUF, and NIF systems, the lowest

number of components is required in the NIF system which is sensitive to oxygen and retained only in some prokaryotes, Entamoeba, and Mastigamoeba (Gill et al., 2007), the latter two of which are exposed to very low oxygen pressure. In contrast, it is presumed that the SUF system is required and retained in the organisms that encounter high oxygen pressure as in the case for bacteria, plants, and Plasmodium.

as in the case for bacteria, plants, and Plasmodium . 8. Fe – S CLUSTER BIOGENESIS

8. Fe S CLUSTER BIOGENESIS IN PROTOZOAN PARASITES

8.1. Entamoeba

Entamoeba is the anaerobic/microaerophilic intestinal protozoan that lacks the aerobic mitochondria and, instead, possesses the highly divergent MRO, called mitosomes. The mitosomes is incapable of oxidative phos- phorylation and energy generation is solely dependent on glycolysis and fer- mentation in the cytoplasm (Ali and Nozaki, 2007 ).

8.1.1 Fe S clusters proteins in Entamoeba

A majority, if not all, of the characterised Fe–S proteins as well as the ones

predicted in silico from the genome database apparently contain [4Fe–4S] 2 þ clusters except Nif U, which contains permanent [2Fe–2S]. [4Fe–4S] pro- teins include ferredoxins (Reeves et al., 1977 ), pyruvate:ferredoxin oxido- reductase (Pineda et al., 2010), and NADPH-dependent oxidoreductase

IronSulphur Cluster Biogenesis in Protozoan Parasites

43

( Jeelani et al., 2010). For instance, each of two isotypes of NADPH- dependent oxidoreductase (EhNO1 and 2) has the CX 2 CX 4 CX 3 CP and

CX

3 CX 3 CX 3 C motifs, suggestive of two [4Fe–4S] clusters are present

per

subunit ( Jeelani et al., 2010 ). Among four ferredoxins in the database,

EhFd1 appears to possess one each of [4Fe–4S] and [3Fe–4S], while three other ferredoxin contain two [4Fe–4S] clusters. Although only a limited number of Fe–S proteins containing [4Fe–4S] or [3Fe–4S] clusters have been characterised so far, Fe–S proteins containing [2Fe–2S] clusters remain largely uncharacterised. In other anaerobic/microaerobic protozoa, Giardia intestinalis and Trichomonas vaginalis, ferredoxins are known to hold [2Fe–2S] clusters (Ali and Nozaki, 2007 ).

8.1.2 NIF system

E. histolytica is unique among parasitic protozoa in that it possesses only the

NIF (or NIF-like) system among the NIF, ISC, and SUF systems as a sole and non-redundant system for the biosynthesis of all Fe–S proteins. No orthologous proteins involved in the ISC and SUF systems in other organ-

isms are present in E. histolytica (Tables 1.2 and 1.3). Amino acid comparison

and phylogenetic analyses indicate that both a catalytic component, NifS and

a scaffold component, NifU (EhNifS and EhNifU, respectively) showed a close kinship to orthologues from e-proteobacteria, suggesting that both of these genes were likely acquired by lateral gene transfer from an ancestor

of e-proteobacteria ( Ali et al., 2004; Van Der Giezen et al., 2004). EhNifS expressed in E. coli is present as a homodimer, and shows cysteine des- ulphurase activity with a very basic optimum pH, compared with NifS from other organisms. EhNifU protein is a tetramer and contains one stable [2Fe–

2S] 2 þ cluster per monomer, revealed by spectroscopic and iron analyses ( Ali et al., 2004 ). Since E. histolytica does not possess nitrogenase activity and is incapable of NIF, the presence of the NIF system and the lack of other sys- tems in this organism reinforce the premise that the NIF system is not spe- cific for the Fe–S cluster formation of nitrogenase, but is also involved in the Fe–S cluster assembly of both [2Fe–2S] and [4Fe–4S] clusters for non- nitrogenase proteins, as proposed for the NIF-like system in H. pylori (Olson et al., 2000). In vivo complementation of a temperature-dependent growth defect of the E. coli isc / suf mutant strain by expression of the

E.

histolytica NIF-like system, that is, EhNifS and EhNifU, indicates that

the

NIF-like system plays an interchangeable role in the Fe–S cluster assem-

bly,

but only under anaerobic conditions (Ali et al., 2004). It is worth noting

Table 1.2 Genome wide analysis of the components of ISC system

Organisms

ISC system

 

IscR

IscS

IscU

IscA

Nfu

HscA

HscB

Fdx

Mge1

Isd11

Frataxin

ATM1

Erv

Grx5

E.

coli

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

S.

cerevisiae

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

þþ

H.

sapiens

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

þþ

E.

histolytica

þ

þ

þ

G.

intestinalis

þ

þ

þ

þ

þ

þ

þ

þ

þ

T.

vaginalis

þþ

þ

þþ

þ

þþ

þ

þ

þ a

Trypanosoma spp.

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

þþ

Leishmania spp.

þþ a

þ a

þþ

þ a

þþ

þ a

þ

þþ a

þ

Plasmodium spp.

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

þ

þþ

C.

parvum

þ

þ

þ

þ

þ

þ

þ

þ

þ

Microsporidia

þ

þ

þ

þ

þ

þ

þ

þ

þþ

B.

hominis

þþ

þþ a

þ

þ

þ

þ

þ

þ

þþ

Note: ‘ þ’ denotes that a putative gene is annotated in the database or the gene was previously described. ‘ þ a ’ denotes that the corresponding gene is annotated as ‘hypo- thetical protein’ in the database, and we manually annotated the gene based on our blast search, which showed a significant similarity to the ortholog from S. cerevisiae, E. coli, H. pylori, or A. thaliana ( >30% amino acid identity). ‘ ’ denotes that the organism does not have an orthologous gene based on previously published data, or our survey, which shows no significant similarity to the ortholog from S. cerevisiae , E. coli , H. sapiens, or A. thaliana (<30% a.a. identity). ‘ ’ denotes that the organism does not have a homologous gene, but that either a partial sequence or a conserved domain is present in the database.

Table 1.3 Genome wide analysis of the components of SUF and CIA systems

Systems

SUF system

 

CIA system

 

Components

 

Organisms

Fur

SufS

SufA

SufB

SufC

SufD

SufE

Nar1

Nbp

Cfd

Cia1

Dre2

E. coli

þþ

þ

þ

þ

þ

þ

þ

þ

þþ

S. cerevisiae

þ

þ

þþ

þ

A. thaliana

þ

þ

þ

þ

þ

þ

þ

þ

þþ

þ

H. sapiens

þ

þ

þþ

þ

E. histolytica

þ

þ

þ

G. intestinalis

þ

þ

þ

T. vaginalis

þ

þ

þ

Trypanosoma spp.

þ

þ

þ a

þ a

þ

Leishmanias spp.

þ

þ

þþ a

Plasmodium spp.

þ

þ

þ

þ

þ

þ

þ

þ