Microbial ferric iron reductases

FEMS Microbiology Reviews 27 (2003) 427^447

www.fems-microbiology.org

Microbial ferric iron reductases Imke Schro«der a

a;


, Eric Johnson a , Simon de Vries

b

Department of Microbiology, Immunology and Molecular Genetics, University of California, Los Angeles, 1602 Molecular Sciences Bldg., Los Angeles, CA 90095-1489, USA b Department of Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands Received 11 December 2002; received in revised form 20 March 2003; accepted 7 April 2003 First published online 26 April 2003

Abstract Almost all organisms require iron for enzymes involved in essential cellular reactions. Aerobic microbes living at neutral or alkaline pH encounter poor iron availability due to the insolubility of ferric iron. Assimilatory ferric reductases are essential components of the iron assimilatory pathway that generate the more soluble ferrous iron, which is then incorporated into cellular proteins. Dissimilatory ferric reductases are essential terminal reductases of the iron respiratory pathway in iron-reducing bacteria. While our understanding of dissimilatory ferric reductases is still limited, it is clear that these enzymes are distinct from the assimilatory-type ferric reductases. Research over the last 10 years has revealed that most bacterial assimilatory ferric reductases are flavin reductases, which can serve several physiological roles. This article reviews the physiological function and structure of assimilatory and dissimilatory ferric reductases present in the Bacteria, Archaea and Yeast. Ferric reductases do not form a single family, but appear to be distinct enzymes suggesting that several independent strategies for iron reduction may have evolved. 3 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Archaeoglobus fulgidus; Saccharomyces cerevisiae ; Ferric iron reductase ; Iron assimilation; Dissimilatory iron reductase ; Iron cycle ; Flavin reductase; Fre

Contents 1.

2.

3.

4.

5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 1.1. Assimilatory iron reduction . . . . . . . . . . . 1.2. Dissimilatory iron reduction . . . . . . . . . . Assimilatory ferric iron reductases . . . . . . . . . 2.1. Bacterial enzymes . . . . . . . . . . . . . . . . . . 2.2. Ferric reductase or £avin reductase? . . . . 2.3. Are there ferric iron-speci¢c reductases? . 2.4. Flavin reductases with dual functions . . . 2.5. Regulation of ferric reductases . . . . . . . . Archaeal ferric iron reductases . . . . . . . . . . . . 3.1. The A. fulgidus ferric reductase (FeR) . . . 3.2. Crystal structure of the A. fulgidus FeR . 3.3. Reaction mechanism . . . . . . . . . . . . . . . . Yeast ferric reductases . . . . . . . . . . . . . . . . . . 4.1. The Fre1 enzyme . . . . . . . . . . . . . . . . . . 4.2. Other Fre enzymes . . . . . . . . . . . . . . . . . 4.3. Regulation by iron and copper . . . . . . . . Dissimilatory ferric iron reductases . . . . . . . . 5.1. Strategies for dissimilatory iron reduction

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

428 428 429 429 430 431 432 432 433 433 433 434 435 435 435 437 438 438 439

* Corresponding author. Tel. : +31 (0) 310 825 8085; Fax: +31 (0) 310 206 5231. E-mail address : [email protected] (I. Schro«der). 0168-6445 / 03 / $22.00 3 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0168-6445(03)00043-3

FEMSRE 776 4-6-03

Cyaan Magenta Geel Zwart

428

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

5.2. Proteins involved in respiration of insoluble ferric iron in Shewanella species . . . . . . . 5.3. Ferric iron reduction in Geobacter species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

439 441 441

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

442

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

442

6.

1. Introduction Ferric reductases catalyze the reduction of ferric iron (Fe3þ ) to ferrous iron (Fe2þ ). Although iron is the fourth most abundant element on earth, it is not readily available in the environment. While soils and sedimentary environments are rich in iron, iron is considered to be a trace element in aquatic habitats. We will provide a brief overview of the global iron cycle to illustrate the physiological roles of ferric reductases (Fig. 1). A more detailed description of the iron cycle can be found in Ehrlich’s book on geomicrobiology [1]. Ferrous iron (Fe2þ ) is released into the environment from corroding iron-containing minerals that are present in rocks, soils or sediments (Fig. 1). While stable under anaerobic or acidic conditions, Fe2þ readily oxidizes chemically to ferric iron aerobically at a pH greater than 5. Fe3þ is also generated by the action of a diverse microbial community that can use Fe2þ as the electron donor to respire with oxygen or nitrate, or as the electron source for photosynthesis under anaerobic conditions [1]. In each case Fe3þ is formed and readily reacts in moist environments to form precipitates as hydroxide, oxide, phosphate or sulfate with very low solubility (10318 M).

Consequently, many environments that are seemingly rich in iron actually contain a free iron concentration of less than 1 WM, which is considered to be the threshold to sustain life [2]. The availability of Fe3þ is increased in environments that are rich in organic compounds including humic substances. Humic substances result from incomplete degradation of complex organic compounds such as lignin and can chelate a variety of metals including Fe3þ . In addition, many microbes secrete metabolic intermediates such as citrate or specially synthesized siderophores that chelate Fe3þ and make it accessible for reduction and/or cellular uptake [3]. Siderophores are lowmolecular mass molecules that are synthesized by bacteria, fungi and plants [4^6]. Whether archaea also produce some type of siderophore is unclear, but it will be interesting to discover whether this strategy of iron solubilization is conserved through all three kingdoms of life. The reduction of Fe3þ for the purpose of intracellular incorporation into protein is called assimilatory iron reduction. In contrast, dissimilatory iron reduction serves the generation of energy to fuel cell propagation. Both pathways are essential to the global iron cycle and will be brie£y introduced below. 1.1. Assimilatory iron reduction

Fig. 1. The role of ferric iron reductases in the iron cycle. See text for further explanation.

FEMSRE 776 4-6-03

Assimilatory ferric reductases are key enzymes of the iron assimilatory pathway. This pathway is found in all living organisms with the exception of a small group of homofermentative lactic acid bacteria [7]. Dependent on the organism, chelated Fe3þ is reduced by a ferric reductase either before or after transport into the cell. Reduction occurs by a mechanism that involves a £avin cofactor in prokaryotes and a b-type cytochrome in yeast (see Section 2). Whether the latter enzymes also contain a £avin cofactor is unclear. In general, the reduction of complexed Fe3þ results in a weak Fe2þ -chelate complex allowing for dissociation and release of Fe2þ for transport or cellular incorporation. This chelation mechanism occurs for citrate and also most of the siderophores that are produced by the diverse microbes. All of these chelators have redox midpoint potentials much lower (Em o P = 0.0 V to 30.7 V) than that of free iron (Em o P = +0.77V) [8^20]. The negative redox potential re£ects the much higher stability (1012 ^1025 ) of the Fe3þ -chelate complex relative to the Fe2þ -chelate complex. Following reduction, the resulting Fe2þ becomes avail-

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

Fig. 2. Reactions proposed to be catalyzed by bacterial £avin reductases and the A. fulgidus ferric reductase (FeR). A: Flavin reductases are proposed to be present as the apoenzymes and act by forming transiently a ternary complex between the reductant (NAD(P)H or glutathione) and free £avin. Following reduction by NAD(P)H the reduced £avin dissociates from the enzyme and may serve as the direct chemical reductant of complexed ferric iron (C-Fe3þ ). In the reaction catalyzed by E. coli £avin reductase, Fre, a sequential ordered reaction mechanism was demonstrated [94,95]. B: The A. fulgidus FeR contains bound £avin as a cofactor, which accepts hydride ions from NAD(P)H. The solvent-exposed £avin can either reduce complexed ferric iron (C-Fe3þ ) or free £avin (£avinox ). We propose that the complete catalytic cycle proceeds according to a ping-pong mechanism [85].

able for intracellular incorporation into heme- and nonheme Fe-containing proteins. Excess iron is sequestered by iron storage proteins such as ferritin and bacterioferritin [21,22]. Iron assimilation at neutral pH and the role of siderophores have been reviewed previously [4^6,23,24] and are also discussed by Simon Andrews in this issue. In acidic environments, ferric iron is soluble and can, therefore, be directly transported into the cell. Reduction to ferrous iron could occur either prior to transport or immediately upon entry into the neutral cytoplasm. The ferric reductase involved in this process may distinguish itself by using free rather than complexed Fe3þ as the substrate. However, iron assimilation in acidophiles has not yet been investigated. This review will summarize the state of knowledge of prokaryotic (bacterial and archaeal) and yeast assimilatory ferric reductases. It is interesting to note that bacteria use £avin reductases as their physiological assimilatory ferric iron reductases (Fig. 2A). This may also hold true for the archaea since the only archaeal ferric iron reductase isolated to date also exhibits £avin reductase activity. In contrast, yeast ferric reductases appear to be dedicated ferric iron and copper reductases. 1.2. Dissimilatory iron reduction

429

clusively in bacterial and archaeal specialists, are an essential part of the iron cycle [25^28] (Fig. 1). Both, inorganic Fe3þ precipitates and a variety of complexed Fe3þ species can be used as terminal electron acceptor in dissimilatory iron reduction [26,27,29^31]. As a result Fe2þ is generated and becomes available to microbes that may live in syntrophy with ferric iron-respiring microbes. In dissimilatory iron reduction, the ferric iron reductase acts as the terminal reductase of an electron transport chain that is somehow linked to the cytoplasmic membrane. The reduction of ferric iron is coupled to the generation of the proton motive force across the cytoplasmic membrane, which is then used to generate ATP by a membrane-bound ATP synthase, to fuel active transport of nutrients, or to drive motility. The proton motive force can also be used to reduce NAD(P) in chemolithotrophs for various biosynthetic reactions. Like sulfur respiration, dissimilatory iron reduction is thought to be one of the oldest energy-generating processes that may have evolved on earth. It has been speculated that iron respiration predates the evolution of nitrate and oxygen respiration [25,32]. Thus far, little is known about dissimilatory ferric iron reductases, and the mechanism by which iron reduction is coupled to the generation of energy. However, it appears that the dissimilatory ferric reductases are not related to any assimilatory ferric reductase characterized thus far. The recent progress on the characterization of dissimilatory ferric reductases will be summarized in Section 5. Dissimilatory ferric reductases are predicted to have evolved about 3.5 billion years ago. At this time, it is thought that ferric iron had formed by intense ultra violet radiation and as a result of anaerobic photosynthesis that utilized Fe2þ as the electron donor [33,34]. Assimilatorytype ferric reductases became physiologically important only when the amount of ferrous iron decreased signi¢cantly around 2.4 billion years ago as a result of oxygen generation by oxygenic phototrophs. With only one recently characterized archaeal ferric reductase (see Section 3), we must be cautious when generalizing about how assimilatory ferric reductases may have evolved. Based on their three-dimensional structure the Archaeoglobus fulgidus ferric reductase and the bacterial £avin reductases belong to the same class of enzymes. This type of enzyme may have evolved in Bacteria, and A. fulgidus may have acquired the gene via horizontal gene transfer when it came into contact with more oxygenated environments. Thus far, it is di⁄cult to recognize ferric reductase encoding genes in the various sequenced genomes. More archaeal ferric reductases need to be identi¢ed in order to trace the evolutionary roots of this important class of enzymes.

2. Assimilatory ferric iron reductases

The dissimilatory ferric reductases, which are found ex-

FEMSRE 776 4-6-03

Assimilatory ferric iron reductases play an essential role

Cyaan Magenta Geel Zwart

430

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

in all organisms that have to acquire iron in the Fe3þ form. Therefore, this type of enzyme activity is present in many prokaryotes and eukaryotes that live in aerobic and neutral environments. Despite the importance of ferric reductase for aerobic life, it appears that this enzyme may have evolved divergently in prokaryotes and eukaryotes since the enzymes of both groups di¡er in their primary amino acid sequences and biochemical properties.

iron uptake, ferric reductases have been identi¢ed in the cell’s cytoplasm, periplasm and cytoplasmic membrane. Alternatively, certain pathogenic bacteria have been found to secrete their ferric reductase enzymes into the culture medium or expose them on their cell surface. Many bacteria produce several ferric iron reductases that can be localized in more than one cellular compartment (described below).

2.1. Bacterial enzymes

2.1.1.1. Periplasmic and cytoplasmic ferric reductases. The majority of the ferric reductases described thus far are either localized in the bacterial cytoplasm or in the periplasm of Gram-negative bacteria. Avirulent L. pneumophila cells produce at least two enzymes that are localized in the periplasm and cytoplasm [44]. While the periplasmic enzyme utilizes reduced glutathione (GSH) as the electron donor, NADPH is the preferred electron donor of the cytoplasmic enzyme. In contrast, virulent cells also use NADH as the electron donor suggesting a virulence-dependent induction of a third cytoplasmic ferric reductase. P. aeruginosa was reported to contain a periplasmic ferripyochelin reductase activity that uses both NAD(P)H and GSH as electron donor [52]. Pyochelin is a siderophore produced and secreted by P. aeruginosa [59]. In addition, a P. aeruginosa cytoplasmic ferric citrate reductase activity was determined that was strictly NAD(P)H-dependent [52]. In other bacteria, periplasmic and cytoplasmic ferric reductases could not be distinguished on the basis of their electron donor and Fe3þ substrate usage. In general, all of these ferric reductases exhibit a very broad substrate speci¢city towards complexed Fe3þ compounds and may even reduce free Fe3þ [36,37,40,51,53]. It is interesting to note that several ferric reductases appear to be in loose association with the cytoplasmic membrane suggesting a con-

Many bacterial ferric iron reductases have been described over the past 30 years from a variety of di¡erent bacteria including Agrobacterium tumefaciens [35], Azotobacter vinelandii [36], Bacillus megaterium [37], Bacillus subtilis [38,39], Escherichia coli [40^42], Legionella pneumophila [43,44], Listeria monocytogenes [45^47], Magnetospirillum magnetotacticum [48], Mycobacterium paratuberculosis [49], Mycobacterium smegmatis [50], Neisseria gonorrhoeae [51], Pseudomonas aeruginosa [52^54], Pseudomonas £uorescens [55], Rhodopseudomonas sphaeroides [56], Treponema denticola [57], and Vibrio vulni¢cus [58]. Only a few of these enzymes have been puri¢ed to homogeneity and characterized in detail. These are summarized in Table 1. In essence, all soluble ferric reductases appear to be very similar biochemically and may be classi¢ed as £avin reductases (see below). The majority use NADH or NADPH as the electron donor for ferric iron reduction. In general, bacterial ferric reductases lack bound £avin. Instead, £avin in the form of FMN, FAD or ribo£avin serves as a di¡usible cofactor. 2.1.1. Cellular location Dependent on the bacterial species and the strategy for Table 1 Summary of £avin reductases and assimilatory-type ferric iron reductases Prokaryote

Mr

e-donora

Location

e-acceptora

A. fulgidus

36 000 dimer

Cytoplasm

NADH (61 WM), NADPH (80 WM)

A. vinelandii

Soluble fraction Cytoplasm Cytoplasm

NADH (15.8 WM)

B. subtilis E. coli

44 600 and 69 000 n.d.b 27 000

L. pneumophila

38 000

Cytoplasm

25 000

Periplasm

36 000 17 000 25 000 27 500 32 000

Cytoplasm Extracellular Cytoplasm Cytoplasm Cytoplasm

M. magnetotacticum M. paratuberculosis N. gonorrhoeae P. aeruginosa R. sphaeroides

NADH, NADPH (19^170 WM) NADH (9 WM), NADPH (30 WM) NADPH (8.85 WM), NADH (11.35 WM) Glutathione (159.2 WM), NADH (114.8 WM) NADH (4.3 WM) NADH NADH NADH NADH (18.2 WM)

FAD, FMN (0.3 WM), Fe -citrate, Fe3þ -EDTA Fe3þ -citrate, Fe3þ -azotobactin, Fe3þ -azotochelin Fe3þ -dihydroxybenzoic acid, FMN FMN (1.5 WM)c , FAD, ribo£avin (1.9 WM)c , broad for Fe3þ substratesd Fe3þ -citrate

[82] [36] [38,70] [40,41] [44]

Fe3þ -citrate Fe3þ -citrate, Fe3þ -quinate Fe3þ NHþ 4 -citrate, transferrin, ferritin Fe3þ -citrate, FMN, broad for Fe3þ Fe3þ -citrate

a

The number in brackets indicates the Km determined for the electron donor. n.d. for not determined. c Values were determined with NADPH. The values are very similar to those determined with NADH [94]. d A variety of organic and inorganic compounds are reduced non-speci¢cally. b

FEMSRE 776 4-6-03

Reference 3þ

Cyaan Magenta Geel Zwart

[48] [49] [51] [54] [56]

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

certed function with a transporter to control iron uptake into the cell [48,51,52]. 2.1.1.2. Extracellular ferric reductases. Certain pathogens, such as L. monocytogenes, produce a surface-bound ferric reductase that may also be secreted into the culture medium [45,47]. L. monocytogenes is a food-borne, Grampositive pathogen that escapes into the cytoplasm of host cells, and from there rapidly spreads to other cells [60]. Ferric transferrin reductase activity was detected in spent culture medium and was suggested to serve as the pathogen’s means for acquiring iron from the host [61]. The activity was dependent on the presence of NADH, catalytic amounts of FMN and Mg2þ [45]. Besides ferric transferrin, this ferric reductase could also reduce Fe3þ -pyrophosphate, Fe3þ -citrate, Fe3þ -EDTA and ferritin. This extracellular enzyme resembles the periplasmic and cytoplasmic ferric reductases found in other bacteria that exhibit very broad substrate speci¢city and require a £avin cofactor for optimal activity. The L. monocytogenes ferric reductase could be enriched from spent culture medium, however, further puri¢cation failed due to apparent oxygen sensitivity of the enzyme [45]. In an independent study, ferric reductase activity was identi¢ed to be associated with the cell surface of L. monocytogenes [47]. With respect to stimulation of this activity by NADH, FMN, and Mg2þ the surface-bound enzyme appeared to be similar and possibly identical to the secreted enzyme. Recently, secreted ferric reductase activities were also reported for E. coli, P. aeruginosa and Salmonella typhimurium suggesting that extracellular iron reduction may be a more common strategy of obligate and opportunistic pathogens to mobilize iron for cellular homeostasis, growth and virulence [62]. The only extracellular ferric reductase isolated thus far is from the culture supernatant of M. paratuberculosis, a Gram-positive pathogen capable of surviving ingestion by macrophages [49,63]. This enzyme is dependent on NADH as the electron donor and its activity is stimulated by Mg2þ ions (Table 1). Apparently, £avin is not required for its activity. It is possible that this enzyme may contain bound £avin, which would truly distinguish this enzyme from other soluble bacterial ferric reductases. Unfortunately, the £avin content of this enzyme was not reported. Using anti-reductase antibody and immunoelectron microscopy, the ferric reductase was located to the M. paratuberculosis cell wall surface and also in the cytoplasm of infected host cells [49]. This suggests that the released and surface-bound enzymes are identical or at least very similar in their amino acid composition. Importantly, the extracellular ferric reductase could potentially be useful in vaccine development against their pathogenic producers [49]. In general, extracellular and periplasmic ferric reductases are dependent on the environmental supply of NAD(P)H, GSH and even £avin unless the enzyme ¢rmly

FEMSRE 776 4-6-03

431

binds the latter. Neither NAD(P)H, GSH or £avin is membrane permeable or known to be secreted by any cell. Therefore, we hypothesize that extracellular and periplasmic ferric reductases may only be physiologically functional if these cofactors are provided to the bacteria, i.e. to pathogens that are internalized by a host cell. Whether a specialized protein or any of the substrate-non-speci¢c porins present in the outer membrane facilitates the uptake of the cofactors into the periplasmic space is currently unknown. We consider it highly unlikely that extracellular and periplasmic ferric reductases are utilized to mobilize iron from soils and aqueous environments as was suggested by Barchini and Cowart [45]. Thus far, extracellular and periplasmic ferric reductases are only known to be produced by obligate or opportunistic intracellular pathogens and may be considered as one of several virulence factors [45]. In general, pathogens are thought to be limited by iron when colonizing a host, however, this was recently contradicted. Eriksson et al. infected murine macrophage-like J774-A.1 cells with S. enterica sv. Typhimurium and demonstrated that Fur-dependent genes are highly down-regulated suggesting that intracellular Salmonella are not starved for Fe2þ [179]. Since J774-A.1 cells are derived from macrophages defective in Nramp1, which transports Fe2þ into acidic late endosomes and lysosomes, it would be interesting to repeat this experiment with a macrophage cell line that is wild-type for Nramp1. In M. paratuberculosis, extracellular ferric reductases appear to be mainly cell surface-associated or in the close vicinity of the pathogen. The advantage of periplasmic or extracellular ferric reductases for pathogens could be several-fold: ¢rst they may deliver ferric iron to a speci¢c transport system to satisfy their own nutritional requirement. Extracellular ferric reductases can also mobilize Fe3þ from proteins such as ferritin [49,61]. It is questionable whether ferric reductase molecules that di¡use too far from the pathogen are still bene¢cial for iron assimilation but rather serve as a defense mechanism. They may compete with phagocytes’ free iron, which is directly involved in the generation of OHc radicals to kill the pathogenic invaders [49]. 2.1.1.3. Membrane-bound ferric reductases. Membranebound ferric reductase activities were reported from Spirillum itersonii [64], E. coli [41], and Staphylococcus aureus [65]. These ferric reductases were suggested to be part of a membrane-bound electron transport chain that uses NADH, succinate, glycerol-3 phosphate, or L-lactate as the donor to reduce ferric iron. They were implied to function in iron assimilation. Thus far, none of the enzymes responsible for these activities has been puri¢ed and characterized. 2.2. Ferric reductase or £avin reductase ? The apparent lack of substrate speci¢city for most ferric

Cyaan Magenta Geel Zwart

432

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

reductases had been a puzzle to many. It was recognized that it was not the enzyme that reduced the Fe3þ compound directly but free reduced £avin produced by the so called ferric reductases (Fig. 2A). The ferripyoverdine reductase puri¢ed from P. aeruginosa provided the ¢rst example for a NADH :FMN oxidoreductase activity when the Fe3þ substrate was absent [53,54]. Halle and Meyer then demonstrated that FMNH2 can chemically reduce a variety of Fe3þ complexes anaerobically, which were previously thought to be reduced by the enzyme directly. This observation is in agreement with an earlier ¢nding, in which dihydro£avins generated by the £avin reductase from Beneckea harveyi could be used to reduce and release iron from ferritin [66]. The best characterized £avin reductase that physiologically also serves as the major ferric reductase is the E. coli Fre enzyme [42]. A £avin reductase was independently puri¢ed by Fisher et al. [41] and by Coves and Fontecave [40,42] and it has been suggested that this enzyme is most likely the same Fre enzyme. Fre was shown to catalyze the reduction of free £avins (FMN, FAD, ribo£avin), which in turn transferred electrons to a variety of ferric siderophores including some that could not be used by E. coli for iron assimilation [40]. Therefore, this enzyme is regarded a £avin reductase and not a ferric reductase. Based on the studies on the P. aeruginosa ferripyoverdine reductase by Halle and Meyer and their own work on the E. coli Fre, Fontecave, Coves and Pierre suggested that the seemingly broad substrate speci¢city of ferric reductases for ferric siderophores (hydroxamates and catecholates) and ferric iron-containing proteins such as ferritins could be explained by a reaction in which reduced £avin non-speci¢cally reduces ferric iron [40, 53,67,68]. The rate of ferric iron reduction is governed by the redox potential of the Fe3þ complex [40,53,67,68]. As a result, Fe3þ complexes with redox potentials more negative than that of the £avin/dihydro£avin couple (3216 mV) can be chemically reduced only in the presence of very strong Fe2þ chelators such as ferrozine or phenanthroline, which are typically used to determine ferric reductase activity in vitro [68]. Under physiological conditions, the reduction of ferric siderophores with low redox potentials such as the hydroxamate siderophores and the reduction of iron-containing proteins such as transferrin remains problematic. The reduction of these compounds has to be coupled to a very high a⁄nity transport system, to the utilization of the membrane potential or to an intracellular high a⁄nity iron-binding protein in vivo. Since reduced free £avin readily reacts with oxygen, a low rate, complete inhibition or lag phase of iron reduction was noted for almost all ferric reductases that were assayed aerobically and monitored by following Fe2þ production. The question arises as to how e¡ectively a £avin reductase may function as ferric reductase in aerobic environments or inside aerobically grown cells. However, given that the bacterial cytoplasm several thiol:disul¢de oxidoreductases, involving thioredoxin and glutaredoxin,

FEMSRE 776 4-6-03

which maintain reducing conditions, it is unlikely that reduced £avins will be oxidized by oxygen [69]. 2.3. Are there ferric iron-speci¢c reductases ? Few ferric reductases are apparently very speci¢c for certain Fe3þ -chelates and are unable to reduce the broad spectrum of ferric compounds reduced by dihydro£avins. This is di⁄cult to explain if these enzymes function as £avin reductases and the reduced £avin is solely responsible for the reduction. The ferric reductase from R. sphaeroides is pro¢cient in ferric citrate reduction but cannot reduce ferric siderophores [56]. In P. aeruginosa two ferric reductases were reported, a ferric siderophore reductase that was later described as a £avin reductase [54] and a distinct ferric citrate reductase [52]. Since the R. sphaeroides and the P. aeruginosa ferric citrate reductases have not been puri¢ed, it is di⁄cult to ascertain their nature as £avin or ferric ion-speci¢c reductase. The only FMN-independent ferric reductases reported thus far are from L. pneumophila [44]. Both the periplasmic glutathionedependent and the cytoplasmic NAD(P)H-dependent enzymes speci¢cally reduce ferric citrate, the only known iron source to satisfy the large iron requirement of this pathogen for growth. Unfortunately, it was not shown whether these enzymes contain any bound £avin. 2.4. Flavin reductases with dual functions If ferric reductases are indeed £avin reductases, these enzymes could assume several independent functions within the cell, where the reduced £avin is used both in enzymatic reactions, as a cofactor, and in non-enzymatic reactions (Fig. 2). The B. subtilis £avin reductase that acts also as a ferrisiderophore reductase forms a complex with chorismate synthase and dehydroquinate synthase, both functioning in aromatic amino acid biosynthesis [38,70]. The reduced £avin produced by the reductase is implicated to play a catalytic role as a cofactor during the synthesis of chorismate [71,72]. Interestingly, the siderophore, 2,3,-dihydroxybenzoic acid, produced by B. subtilis is a product of the aromatic amino acid biosynthesis pathway [73]. It is, therefore, possible that the B. subtilis £avin reductase may also have a regulatory role that somehow interfaces the need for iron with increased production of siderophores [38]. The E. coli £avin reductase (Fre) was discovered as part of a multicomponent complex formed with aerobically expressed ribonucleotide reductase, a key enzyme in DNA biosynthesis [42]. The £avin reductase is required for the activation of ribonucleotide reductase by reducing its non-heme diferric center directly or indirectly via a ferrous ion intermediate [74]. Whether the £avin reductase may play a regulatory role by regulating ribonucleotide reductase activity and thus DNA biosynthesis is not clear [75]. More recently, the E. coli £avin reductase was invoked as azo reductase and as chromate reductase

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

in vitro, both important activities for bioremediation processes [76,77]. 2.5. Regulation of ferric reductases Consistent with the versatile physiological function as £avin reductase, ferric reductases are generally constitutively produced. In almost all publications on ferric reductases it was noted that the ferric reductase activity levels did not change in low versus high iron concentrations in the culture medium. This is in contrast to the di¡erential expression of other genes involved in iron assimilation, such as the siderophore biosynthesis and iron transporter genes. The expression of these genes is repressed by the Fur regulatory protein in response to high iron availability allowing for expression only at low iron concentration [78^80]. A noticeable exception is the ferric reductase from M. magnetotacticum [48]. In this organism ferric reductase activity increased with increasing Fe3þ -quinate concentration in the culture medium up to a 5 WM concentration. A higher iron concentration did not e¡ect the enzyme levels. Since in M. magnetotacticum ferric reductase activity could be correlated with the number of magnetosomes, Noguchi et al. suggested that this enzyme might play a role in magnetite biosynthesis. Magnetosomes are small, intracellular particles made of magnetite (F3 O4 ) and greigite (F3 S4 ). These intracellular magnets allow the alignment of magnetotactic bacteria such as M. magnetotacticum along the geomagnetic ¢eld lines and have been implicated to orient the highly motile bacteria towards their optimal microaerophilic environment [81].

3. Archaeal ferric iron reductases Our knowledge about dissimilatory and assimilatory iron reduction in the Archaea is scarce. To date, only one archaeal ferric reductase enzyme has been isolated and characterized extensively (Table 1) [82]. This enzyme, the ferric reductase (FeR) from the hyperthermophilic archaeon A. fulgidus, will be the focus of the following section. It still remains unclear whether FeR serves an assimilatory or dissimilatory role in A. fulgidus. Previously, Vargas et al. reported iron-reducing activities with H2 as electron donor in cell suspensions of selected hyperthermophilic Archaea including A. fulgidus [25]. The authors suggested that these activities may have a dissimilatory function. They demonstrated that the archaeon, Pyrobaculum islandicum, is capable of utilizing Fe3þ -citrate as electron acceptor for growth. However, our attempts to culture A. fulgidus with Fe3þ -citrate and H2 have thus far not been successful. Magnetite, which is usually taken as an indication for Fe3þ respiration, is also formed in the culture medium in the absence of A. fulgidus (unpublished observation by I. Schro«der and A. Slobodkin).

FEMSRE 776 4-6-03

433

More recently, Kashe¢ et al. have reported the isolation of a novel hyperthermophilic archaeon, Geoglobus ahangari, which can grow autotrophically on hydrogen with Fe3þ serving as the sole electron acceptor [83]. These studies suggest the existence of at least one archaeal ferric reductase involved in dissimilatory iron reduction. Future isolation and characterization of the enzymes involved promises to provide new insight into the process of iron respiration in the Archaea. 3.1. The A. fulgidus ferric reductase (FeR) A. fulgidus is a strictly anaerobic, hyperthermophilic, sulfate-reducing archaeon and was found to contain ferric reductase activity exclusively in its soluble fraction [25,82]. A. fulgidus may encounter ferric iron at the interface where hot anaerobic deep sea vent water mixes with the surrounding oxygenated water resulting in the oxidation and precipitation of many metals. As isolated FeR was highly active and very abundant comprising close to 1% of the soluble A. fulgidus protein [25,82]. Whether this enzyme is involved in the citrate-complexed Fe3þ reduction with H2 as electron donor is not clear and, because of the absence of a genetic system for A. fulgidus, this issue is di⁄cult to address. FeR catalyzes the reduction of EDTA or citrate-complexed Fe3þ with either NADH or NADPH as the electron donor. Puri¢ed FeR lacked any prosthetic group and required the addition of either FMN or FAD for Fe3þ reduction. In that respect, the A. fulgidus FeR is very similar to the bacterial £avin reductases, which act as ferric reductases. In fact, the A. fulgidus FeR is a powerful £avin reductase that can reduce either FAD or FMN in the absence of a Fe3þ electron acceptor [82]. Its £avin reductase activity is 750-fold higher than that of the E. coli Fre [84]. In contrast to the E. coli Fre enzyme, ribo£avin cannot serve as electron acceptor or cofactor in Fe3þ reduction suggesting a di¡erent £avin coordination in the A. fulgidus FeR. While all bacterial ferric and £avin reductases comprise monomeric enzymes in solution, the A. fulgidus FeR forms a homodimer. At elevated, but not at room temperature, the £avin-less FeR can be reconstituted with FMN or FAD, suggesting that the A. fulgidus FeR is a £avoprotein [82]. Furthermore, overexpression of the fer gene in E. coli resulted in a £avinated protein that contained one FMN per dimer [85]. We speculate that this £avin is loosely bound to the enzyme and that it can be removed during more extensive puri¢cation procedures as this was done when FeR was puri¢ed from A. fulgidus [82]. Interestingly, the A. fulgidus FeR primary sequence lacks any signature sequence for £avin and NAD(P) binding, both of which are found in the primary sequence of the E. coli Fre [86]. In fact, the A. fulgidus FeR shares no homology to the E. coli Fre or to any other ferric reductase [87,88]. Instead, weak but signi¢cant homology of the A. fulgidus ferric reductase exists to a family of

Cyaan Magenta Geel Zwart

434

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

NAD(P)H:£avin oxidoreductases that are members of a two-enzyme system. Here the £avin reductase generates reduced FMN for a monooxygenase [25,82]. These twoenzyme systems are involved in the oxidative degradation of hydrocarbons, aromatic compounds, and in polyketide biosynthesis. It is interesting to note that thus far no archaeal homologue was found suggesting that A. fulgidus may have acquired the fer gene via lateral gene transfer. 3.2. Crystal structure of the A. fulgidus FeR Very recently the recombinant FeR was crystallized and its structure was solved with bound FMN and as a complex with NADPþ (Fig. 3) [85]. Structural analysis revealed that a disul¢de bridge covalently links the FeR dimer. Disul¢de bridges are suggested to enhance thermo-

stability in proteins [89]. It is therefore possible that dimerization contributes to the thermostable nature rather than being important for the reaction mechanism of FeR. Each monomer consists of a single domain that is comprised mainly of a L barrel structure. This type of L barrel structure is also present in the FMN-binding protein of Desulfovibrio vulgarus [90,91], and has been shown to be a version of the £avin-binding domain of proteins belonging to the ferredoxin reductase superfamily (Fig. 3) [92,93]. Although the FeR dimer is comprised of two identical subunits, only one subunit (subunit A) was found to bind FMN [85]. Soaking of the crystals with NADPH introduced binding only to the subunit that was preoccupied with £avin. Subtle structural di¡erences between the £avin-binding subunit A and the £avin-less subunit B appear to prevent NADPþ coordination at subunit B.

Fig. 3. The three-dimensional structures of the A. fulgidus ferric reductase (FeR) shown as a monomer [85], and selected members of the ferredoxin reductase family including the E. coli £avin reductase (Fre) [86], the NADH:cytochrome b5 reductase from porcine liver [177], and the E. coli £avodoxin reductase [178]. The £avin is shown in red and NADPH in yellow. The A. fulgidus ferric reductase is the ¢rst representative of a new sub-family that binds NADPH without a Rossmann fold domain.

FEMSRE 776 4-6-03

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

Each atom of the FMN isoalloxazine ring is stabilized mostly via hydrogen bonds to the backbone main chain of the L barrel structure. It is not too surprising that these backbone amino acid residues are not conserved among homologues. What matters is the ability of the contributing amino acids to form the L barrel providing the structural signature for £avin binding typical for the ferredoxin reductase superfamily. Additional bonding to the phosphate ¢rmly anchors FMN to the L barrel. The E. coli Fre enzyme, the crystal structure of which was also recently solved (Fig. 3), binds FAD solely via hydrophobic interactions to the isoalloxazine ring and hydrogen-bonding to the ribose [86]. This allows for acceptance of ribo£avin as a cofactor, which cannot be stabilized by the A. fulgidus FeR. The most signi¢cant di¡erence between FeR and other members of the ferredoxin reductase family is the absence of a Rossmann fold domain, which has been implicated in NAD(P)þ binding (Fig. 3). Instead, FeR binds NADPþ via two K helices that comprise the N- and C-terminal extension of each monomer. Thus, the A. fulgidus FeR constitutes the ¢rst member in a new subclass of NAD(P)H-binding enzymes within the ferredoxin reductase superfamily (Fig. 3). 3.3. Reaction mechanism The crystal structure of the A. fulgidus FeR revealed that NADPH can bind only in the presence of FMN [85]. Since the structure cannot accommodate both NADPH and complexed Fe3þ , and since a speci¢c ironbinding site could not be identi¢ed, we suggest a pingpong reaction mechanism in which NADPþ leaves the enzyme prior to the binding of Fe3þ complex. This is in contrast to the E. coli Fre, where an ordered sequential mechanism for £avin reduction was proposed [94,95]. Using inhibitor studies it was demonstrated that the E. coli Fre enzyme binds NADPH ¢rst followed by ribo£avin [94,95]. While the reduced £avin is the ¢rst product to be released during the E. coli Fre reaction cycle [94], we suggest that the reduced £avin remains with the A. fulgidus FeR as this is the case for £avoenzymes. In most £avoproteins the £avin is shielded from solvent [96]. However, it is exposed in the A. fulgidus FeR, which makes it readily accessible to other substrates such as £avins in the £avin reductase reaction and complexed Fe3þ in the ferric iron reduction reaction without necessarily accommodating the latter substrates at a high a⁄nity binding site (Fig. 2B).

4. Yeast ferric reductases Eukaryotic microorganisms have evolved two basic iron-scavenging strategies: in the budding yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe the major iron uptake route consists of two consecutive steps:

FEMSRE 776 4-6-03

435

reduction of complexed ferric iron to the ferrous state by a plasma membrane-bound ferric reductase (Fre) followed by transport of the uncomplexed, relatively soluble, ferrous iron by two independent transport systems. In yeasts, uptake is mediated through the high a⁄nity Fet3/Ftr1 ferroxidase/permease complex, or by the low a⁄nity iron transporter Fet4 (Fig. 4, Table 2) [8^20]. While Fet4 is a Fe2þ transporter, the multicopper oxidase Fet3 oxidizes Fe2þ to Fe3þ prior to transport by Ftr1. Therefore, iron can enter the yeast cell in either the ferric or the ferrous state (Fig. 4). It appears puzzling that ferric iron is reduced via Fre only to be reoxidized for high a⁄nity uptake by Fet3/Ftr1. However, it is possible that ferric iron reduction is required to remove iron from chelators (siderophores). Subsequent reoxidation to Fe3þ may serve to provide increased substrate speci¢city to iron import since Ftr1 is highly speci¢c only for free Fe3þ [18^20]. In contrast, the Fet4 transporter can import a variety of divalent metals including Fe2þ [15,97]. In contrast to yeast, fungi seem to take up iron in a single step by transporting the Fe3þ -siderophore complex as a whole into the cell [98]. While fungi and bacteria are capable of synthesizing a great variety of siderophores, budding and ¢ssion yeasts are not [99]. Nevertheless, yeasts are capable of reducing the ferric siderophores produced by other organisms and subsequently transporting the free ferrous iron via the Fet3/Ftr1 complex. In addition, S. cerevisiae contains, like fungi, a variety of ferric siderophore permeases of the Arn family (Arn1^4) [100^ 104]. At present, the immediate intracellular acceptor for either Fe2þ or Fe3þ is not known, nor is the fate of the siderophores upon cell entry known. The best-described eukaryotic ferric reductases are the Fre enzymes from S. cerevisiae. We will, therefore, focus mainly on these enzymes. 4.1. The Fre1 enzyme NAD(P)H-dependent Fe3þ -citrate reductase activity was localized to the plasma membrane of S. cerevisiae [17, 99,105,106]. Interestingly, this activity was inhibited in the intact cell by CCCP, an uncoupler of the proton motif force, by Hþ -ATPase inhibitors, and by plasma membrane permeabilizing agents suggesting that the activity is linked to the generation of a membrane potential [99]. Two membrane-bound ferric reductases, Fre1 and Fre2, account for more than 99% of the in vivo activity in S. cerevisiae. Of these, Fre1 is the major ferric reductase. In spite of many e¡orts, none of the ferric reductases has been obtained in a pure form since preparation of the plasma membrane results in complete loss of Fe3þ -citrate reductase activity [107]. Consequently, previously puri¢ed NADPH :ferrireductase activities do not represent the physiologically relevant assimilatory ferric reductase identi¢ed by genetic analyses [107^109]. Fre1 was discovered by mutational analysis of S. cere-

Cyaan Magenta Geel Zwart

436

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

Fig. 4. Iron reduction and uptake systems in S. cerevisiae. Fe3þ -chelate is reduced by six Fre homologues, of which Fre1 and Fre2 are the major ferric iron reductases. The resulting Fe2þ is either reoxidized and transported by the high a⁄nity Fet3/Ftr1 system, or transported directly by the low a⁄nity Fet4 transporter. The Arn3 permease is located in post-Golgi vesicle membranes and transports chelated Fe3þ directly. Also shown is Fre7, which is implicated in Cu2þ reduction rather than in Fe3þ reduction. Cu transport systems (e.g. Ctr1) and copper chaperones are not included in the ¢gure. The expression of the genes for all iron reductases and iron transporters is under control of the Aft1 and Mac1 activators in response to low iron and copper availability.

visiae [105]. A mutation of the FRE1 gene greatly reduced ferric reduction but not ferric uptake. Subsequently, Dancis et al. cloned and sequenced the FRE1 gene [87]. The predicted 79-kDa Fre1 protein contains seven putative transmembrane helices and is homologous to the large L subunit, gp91phox , of NADPH oxidase from human phagocytes [87,110^112]. NADPH oxidase consists of two subunits, gp91phox and p22phox , that form a £avo-cytochrome b558 . NADPH oxidase is located in the plasma membrane of phagocytotic cells such as neutrophile granules, where it becomes part of the phagocytotic vacuole. Here, NADPH oxidase functions to transfer electrons from NADPH at the cytoplasmic site across the vacuole membrane to oxygen inside the vacuole to generate superoxide radicals as part of an antibactericidal defense system [113]. The enzyme’s FAD and cytochrome b558 are critical parts of the transmembrane electron £ow. Fre1 shares the FAD- and NADPH-binding motifs with gp91phox . Since the biochemical analysis of Fre1 has proven to be di⁄cult, the well-investigated NADPH oxidase serves as a model for Fre1. While it has been established that Fre1 contains cytochrome b it is not clear whether FAD is also present in this enzyme or provided by an additional enzyme that may form a complex with Fre1 [111]. Using site-directed mutagenesis four histidine residues were identi¢ed and suggested to coordinate two b-type hemes [110]. Finegold et al. suggested that these hemes are sandwiched between two transmembrane domains similar to the two heme b groups

FEMSRE 776 4-6-03

of the mitochondrial cytochrome bc1 complex [114]. It is the transmembrane electron transfer between the two heme b groups which is responsible for energy conservation and the generation of the proton motif force by the bc1 complex. The four histidine residues and the spacing between them are also conserved in gp91phox predicting a similar heme arrangement in the NADPH oxidase. Henderson et al. have demonstrated that the gp91phox subunit alone generates a proton motive force in accordance with the proposed transmembrane heme b arrangement [115]. Fre1 has been implicated in acidifying the cell’s environment consistent with transmembrane electron transfer [116], although this has to be veri¢ed by more direct methods. The midpoint potential of heme b in Fre1 was determined at approximately 3250 mV, similar to that of the NADPH oxidase, where two slightly di¡erent midpoint potentials of 3225 and 3265 mV have been measured [117]. Such a low redox midpoint potential is required to reduce oxygen to superoxide. Reduction of oxygen is presumed to occur via the heme edge, without the formation of a stable ferrous oxygen intermediate. The ability of the heme b to form a low a⁄nity ferrous CO complex is taken as support for this mechanism. Surprisingly, the Fre1 heme b as present in the plasma membrane was also shown to bind CO [111]. The signi¢cance of this is not clear, but it might indicate that the reduction of ferric chelates also occurs via the heme edge without formation of a Michaelis complex with the substrate. A low redox

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

midpoint potential of the Fre1 heme b is in principle not needed to reduce Fe3þ -citrate (Em is about 0 mV at pH 7 and approximately 200 mV at pH 4). However, it may be required for the reduction of the various ferric siderophore complexes (Table 2), which have midpoint potentials ranging between 3350 mV and 3468 mV at pH 7, or even lower in the case of the Fe3þ -enterobactin (3750 mV at pH 7) [102]. Considering that the physiological pH for growth of yeast is at pH 4, the midpoint potentials of several ferric siderophores will increase by 180 mV relative to the values quoted above, bringing them in the same range as the Fre1 heme b. Furthermore, because electron transfer presumably proceeds via the heme b from the inside to the outside of the yeast cell, additional power to reduce ferric chelates is obtained by an amount corresponding to the magnitude of the plasma membrane potential [118]. 4.2. Other Fre enzymes The remaining 10% of ferric reductase activity present in a FRE1 deletion strain was mainly attributed to Fre2, which shares 25% amino acid sequence identity with Fre1 [87,105]. Recently, the genome sequences of both S. cerevisiae and S. pombe became available [119,120]. Analysis of the S. cerevisiae genome revealed the presence of ¢ve additional FRE genes (FRE3^7) with sequence similarity to FRE1 and FRE2. The genome of S. pombe harbors two orthologues of FRE1 and FRE2 (FRP1 and FRP2). In Candida albicans a FRE1 orthologue (CFL1) has been identi¢ed [121]. The S. cerevisiae FRE genes encode membrane-bound proteins with molecular masses between 70 and 80 kDa. FRE3^6 are 76%, 57%, 38% and 36% identical to FRE2, respectively. FRE7, encoding the shortest protein, has 21% identity with either FRE1 or FRE2. Structural features such as the membrane topology, putative NADPH- and

437

FAD-binding sites are conserved among all Fre enzymes. The physiological role of the Fre3^7 enzymes is thus far not known. However, they could contribute to utilization of ferric substrates that are only poorly reduced by Fre1 and Fre2 (Table 2) [102,103]. Both Fre1 and Fre2 can reduce Fe3þ -citrate and a variety of ferric siderophores (Table 2). Surprisingly, both ferric reductases are also capable of reducing free Cu2þ to Cu1þ [122,123]. Therefore, Fre1 and Fre2 signi¢cantly contribute to the copper homeostasis in yeast. Cu1þ is transported by the plasma membrane transporters Ctr1 and Ctr3, and subsequently bound to one of the copper chaperones in the cytoplasm [124,125]. It is interesting to note that in yeast metabolism of iron and copper is linked through the Fre enzymes. This connection is also re£ected in the regulation of FRE1 and FRE2 expression by both iron and copper (see below). An additional Fe/Cu connection exists through the high a⁄nity Fet3/Ftr1 ferroxidase/ permease complex (Fig. 4). The copper-containing Fet3 is a homologue of the human ceruloplasmin, a ferroxidase that oxidizes ferrous iron to the ferric state, which is then bound to the iron storage protein apotransferrin [12]. Thus, copper and iron metabolism also appear to be closely linked in higher eukaryotes. As a result, copper starvation in humans a¡ects iron metabolism causing iron de¢ciencies such as anemia [12]. A similar e¡ect is observed in Menkes or Wilson’s disease, in which copper loading of ceruloplasmin is disturbed [12,20]. Thus, iron metabolism in yeast has been used as a model system to study iron and copper metabolism in humans. The lack of successful puri¢cation of any Fre enzyme and the loss of ferric reductase activity upon cell disruption have severely hampered a detailed biochemical analysis of this important class of enzymes. As a consequence, it is uncertain whether NADPH is the direct physiological electron donor and ferric iron the physiological direct electron acceptor. In analogy to the phagocytotic NADPH

Table 2 Properties of proteins involved in iron reduction and iron transport in S. cerevisiae Protein

Function/substrate

Gene regulation

Reference

Fre1

Fe3þ -citrate, Cu2þ , ferrioaxamine B, ferrichrome, triacetylfusarinine, rhodotoluric acid, enterobactin Fe3þ -citrate, Cu2þ , ferrioaxamine B, ferrichrome, triacetylfusarinine, rhodotoluric acid Ferrioaxamine B, ferrichrome, triacetylfusarinine, rhodotoluric acid Fe3þ -rhodotoluric acid (low a⁄nity) Unknown Unknown Unknown Multicopper ferroxidase Low a⁄nity iron transporter High a⁄nity iron transporter Hydroxamate/catecholate permease Transcriptional activator Transcriptional activator Transcriptional activator

Activated by Aft1, Aft2 and Mac1

[15,87,103,105,123,126^129,131,132]

Activated by Aft1; down-regulated by Cu1þ by an unknown mechanism Activated by Aft1

[127,129]

Activated Activated Activated Activated Activated Activated Activated Activated Activated Activated Activated

[127,129] [127] [127] [127] [18^20,131,132] [9,10,15] [12,20] [100^102,104] [128^130] [131,132] [122,123,126,127,133]

Fre2 Fre3 Fre4 Fre5 Fre6 Fre7 Fet3 Fet4 Ftr1 Arn3 Aft1 Aft2 Mac1

FEMSRE 776 4-6-03

by Aft1 by Aft1 by Aft1 by Mac1 by Aft1 and Aft2 at high Fe2þ by Aft1 by Aft1 at low Fe2þ at low Fe2þ at low Cu1þ

Cyaan Magenta Geel Zwart

[127,129]

438

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

oxidase, Fre might generate O3 2 as a transient intermediate, which in turn reduces Fe3þ . The loss of ferric citrate reductase activity upon cell disruption is di⁄cult to understand from a purely thermodynamic point of view, i.e. the plasma membrane potential is theoretically not needed as an extra driving force to reduce Fe3þ -citrate with NAD(P)H as electron donor. It is possible that the Fre enzymes associate with other proteins forming a complex. Upon disruption of the cell, the integrity of such a complex, and hence, the reductase activity is lost [107]. Alternatively, the catalytic mechanism might require a plasma membrane potential for presently unknown reasons. If this is correct, closed plasma membrane vesicles should regain Fe3þ -citrate reductase activity. Investigation of the biochemistry of ferric reductases, elucidation of their structure, and reaction mechanism are important issues for future research. 4.3. Regulation by iron and copper In S. cerevisiae all components of the high a⁄nity iron uptake system are regulated in response to the environmental iron concentration. At low iron concentration ( 6 5 WM) the genes for FRE1, FRE2 and for the FRE3^ 6 homologues as well as the Fet3/Ftr1 transporter genes are highly expressed [13,18,20,97,105,126]. Expression of all genes is greatly reduced when iron becomes available at elevated concentrations. A noticeable exception is the FRE7 gene, which is not regulated in response to iron but to copper [127]. Iron regulation is mainly mediated by the Aft1 transcriptional activator, which promotes transcription of the FRE1^6, FET3 and FTR1 genes at low iron concentration (Table 2). In the presence of elevated iron, Aft1 no longer binds to its regulatory target elements and expression of the iron homeostasis genes is abolished [128,129]. Yamaguchi-Iwai et al. recently demonstrated that the Aft1 protein was localized to the cytoplasm when iron was abundant, while the protein could be detected in the nucleus under iron-deplete conditions [130]. The authors suggested that the nuclear import/export systems are involved in iron sensing by Aft1. The current model for iron regulation is that the nuclear retention of Aft1 results in gene expression under iron starvation, while Aft1 is exported from the nucleus when iron is available. Thus far, it is not clear whether Aft1 can bind iron directly or whether Aft1 undergoes a post-translational modi¢cation in response to changes of the cell’s iron status. Recently, a second iron regulatory protein, Aft2, was identi¢ed in S. cerevisiae [131,132]. Aft1 and Aft2 share 39% identical amino acid residues. Rutherford et al. established that FRE1 and FET3 transcription could be activated by Aft2 in a AFT1 deletion strain suggesting that Aft1 and Aft2 may overlap in their physiological function [132]. This result was independently con¢rmed by Blaiseau et al. [131]. Since both Fre1 and Fre2 also act as cupric metal re-

FEMSRE 776 4-6-03

ductases, it is not surprising that their genes are regulated in response to copper availability [123,127,133]. Under copper-de¢cient conditions, the FRE1 gene and the copper transporter genes CTR1 and CTR3 are expressed highly. Also, FRE7 appears to be induced by copper de¢ciency. The expression of all of these genes is low when copper is abundant in the environment. The transcriptional activator protein Mac1 is the main mediator of this copper-dependent regulation (Table 2). The three-dimensional structure of Mac1 was recently solved by Brown et al. showing that Mac1 binds Cu1þ as a [Cu4 ^S6 ] cage [134]. The authors suggested that copper binding results in a conformational change that promotes interaction between the N-terminal DNA-binding domain and the C-terminal copper-binding domain of Mac1 occluding the DNA target element. Like FRE1, expression of FRE2 is repressed by copper. However, FRE2 expression is not regulated by Mac1 since normal expression occurs in a MAC1 deletion strain [123,127]. Also MAC1 expression is maximal in the absence of copper [123]. The mechanism of FRE2 and MAC1 gene regulation is currently not known. The presence of so many di¡erent Fre enzymes in S. cerevisiae undoubtedly contributes to its successful competition with bacteria and other fungi under a great variety of natural environments to acquire scarcely available and essential iron and copper. Fre1 and Fre2 are the main enzymes involved in iron and copper reduction. Fre3 and Fre4 appear to be specialized iron siderophore reductases exclusively regulated by iron. Functions for Fre5 and Fre6 have so far not been established, but they may widen the substrate repertoire in obtaining transition metal ions that have yet to be identi¢ed. Alternatively, Fre5 and Fre6 might act as metal sensors. However, their regulation by iron suggests a role of these enzymes for iron metabolism. Based on its regulation in response to copper, one might speculate that Fre7 acts as a copper reductase. In addition, weak Cu2þ reductase activity was measured in a strain deleted for FRE1 and FRE2 indicating the presence of an additional Cu2þ -reducing enzyme. However, the additional deletion of FRE7 did not a¡ect this residual Cu2þ reductase activity suggesting that Fre7 is most likely not a Cu2þ reductase [123]. Since homology of Fre7 to Fre1^6 is very poor it becomes questionable whether this protein is a metal ion reductase at all. It is possible that Fre7 somehow functions in copper metabolism but its physiological role has yet to be established.

5. Dissimilatory ferric iron reductases Iron respiration is utilized by a very diverse group of bacteria and archaea as the means to support growth (reviewed in [27,135]). While other respiratory pathways have been studied extensively, iron respiration has received little attention. Only recently, signi¢cant progress has been made by several research groups on unraveling the strat-

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

egies used for the dissimilatory reduction of insoluble Fe3þ oxides. In general, three strategies for iron respiration have been identi¢ed thus far and will be reviewed below. 5.1. Strategies for dissimilatory iron reduction 5.1.1. Reduction of insoluble iron via an electron shuttle Lovley and coworkers have shown that humic substances, which are naturally abundant in many environments, can serve as electron shuttle to Fe3þ oxides [26,30,31,136]. Humic substances contain quinone-like moieties that can undergo redox cycling. In their oxidized state humics can act as the terminal electron acceptor for a respiratory chain. Once reduced, the extracellular redox compound is chemically reoxidized using Fe3þ oxides as electron sink. This type of respiratory chain could represent a primitive electron transport chain that is not iron-speci¢c but could also use a variety of other compounds present in the environment as an electron sink. It is possible that iron respiratory chains involving dedicated iron-reducing enzymes may have evolved from such primitive electron transport chains. However, it is also possible that several strategies for iron respiration evolved independently. Recently, Newman and Kolter discovered that some bacteria reduce Fe3þ oxides by producing and secreting small, di¡usible redox compounds that can serve as an electron shuttle between the microbe and the insoluble iron substrate [137,138]. As a consequence, the microbe does not need to directly contact the Fe3þ oxide. In an elegant approach, Nevin and Lovley embedded Fe3þ oxides into microporous alginate beads that were inaccessible to bacteria [137,138]. Nevertheless, the iron-reducing bacteria Geothrix fermentans and Shewanella alga strain BrY were able to reduce the Fe3þ oxide within the beads via compounds that these bacteria had secreted [137,139]. Examples for secreted, di¡usible electron shuttles used for iron reduction include hydrophilic quinones of yet unknown nature by Shewanella oneidensis strain MR-1 (formerly Shewanella putrefaciens strain MR-1) and G. fermentans [137,138], as well as melanin by S. alga BrY [140]. The role of excreted compounds in extracellular electron transfer was recently reviewed by Hernandez and Newman [141]. 5.1.2. Secretion of Fe3+ chelators It had been noted by Nevin and Lovley that in addition to quinones G. fermentans and S. alga BrY secreted ironchelating compounds that resulted in signi¢cant solubilization of Fe3þ oxide [137,139]. The nature of these siderophores has yet to be established. It is also not yet clear whether the Fe3þ -chelates can serve as terminal electron acceptors for dissimilatory iron reduction. In fact, Hernandez and Newman argue that most siderophores are unsuitable as electron acceptors due to their negative redox potentials [141]. However, one could envision that siderophores used for iron respiration di¡er from those

FEMSRE 776 4-6-03

439

that might function in iron assimilation. It is noteworthy that many iron-respiring microorganisms are cultured with Fe3þ -citrate (Em at pH 7 is about 0 mV) as a terminal electron acceptor. It would be interesting to know whether any of the iron-respiring microbes speci¢cally secrete citrate for the purpose of chelating iron for respiration. 5.1.3. Iron respiration via direct contact to insoluble iron A fundamentally di¡erent mode of iron respiration is used by Geobacter metallireducens [142]. This anaerobic iron-reducing bacterium must directly contact the Fe3þ oxide for reduction unless a soluble electron acceptor such as Fe3þ -citrate is provided in the environment. When presented with insoluble Fe3þ or Mn4þ G. metallireducens was shown to produce £agella and pili to allow for chemotaxis and attachment to the metal oxides [143]. Neither appendage is formed when G. metallireducens is cultured in the presence of Fe3þ -citrate. In contrast to previous reports [144], Shewanella species can respire Fe3þ without attaching themselves to Fe3þ oxide particles [138], although this has been recently disputed by Caccavo and Das, who presented evidence that S. alga BrY may use its £agella for attaching to Fe3þ oxide [145,146]. Additional proteins also appear to function in adhesion to the metal oxide, since de£agellated S. alga BrY still attached to Fe3þ oxides, albeit with reduced e⁄ciency [145,147]. Although £agellum-mediated adhesion was not found to be a prerequisite for Fe3þ reduction [145], there appears to be a correlation between adhesion and Fe3þ oxide reduction [148]. Adherence to the Fe3þ oxide particle may present a distinct advantage for iron-respiring bacteria to facilitate Fe3þ oxide respiration. Whether Fe3þ oxide adherence in S. alga BrY is regulated by the presence of soluble Fe3þ electron acceptors is currently unknown. 5.2. Proteins involved in respiration of insoluble ferric iron in Shewanella species In the past years, many studies have focussed on S. oneidensis MR-1, a versatile Gram-negative bacterium that can respire with a variety of di¡erent electron acceptors including Mn4þ and Fe3þ supplied as Fe3þ -citrate or Fe3þ oxide [149,150]. It has become evident that c-type cytochromes are major components in iron and manganese respiration, either in electron transfer or as a possible terminal reductase [151,152]. These cytochromes are mainly located in the periplasmic space [153^155] or associated with the membrane [151] and are expressed anaerobically regardless of the electron acceptor [156]. Many e¡orts have focussed on identifying and purifying the ferric reductase enzyme(s) in S. oneidensis MR-1. These efforts have led to the elucidation of functions for proteins that are involved in iron respiration and are summarized below. In S. oneidensis MR-1, more than 50% of the total ferric iron reductase activity was localized to the outer mem-

Cyaan Magenta Geel Zwart

440

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

brane [166], an unusual location for a respiratory enzyme that is typically found in the cytoplasmic membrane or periplasm of an organism. However, the location of this enzyme is perfectly suitable for the utilization of an insoluble substrate. To date, this ferric reductase could not be isolated directly from the outer membrane of S. oneidensis MR-1. Biochemical analysis of the outer membrane from S. oneidensis MR-1 by Myers and Myers revealed the existence of four cytochrome c-containing proteins that could be reduced with formate and reoxidized by Fe3þ and Mn4þ suggesting that one or more of these proteins could constitute the ferric reductase [151,152]. However, based on deletion mutants, the same authors recently suggested that at least two of these cytochrome c proteins, OmcA and OmcB, do not participate in Fe3þ reduction but more likely are involved in the reduction of Mn4þ [158,159]. OmcA was also puri¢ed and characterized from Shewanella frigidimarina NCIMB400 by Field et al. [160]. This protein contains 10 c-type hemes with redox potentials of 3243 and 3324 mV. Field et al. proposed that this protein might function as a putative terminal ferric iron reductase since the reduced protein could be reoxidized by Fe3þ -EDTA. However, given the low redox potential of the hemes, oxidation by Fe3þ may not be too surprising. In addition, the protein was found to easily detach from cells and may thus contact insoluble Fe3þ substrates [160]. Whether this S. frigidimarina NCIMB400 protein is indeed Fe3þ - and not Mn4þ -speci¢c has yet to be con¢rmed. The outer membrane protein MtrB has also been proposed to be a component of the ferric and manganese reductase in S. oneidensis MR-1 [161,162]. This was based on the observation that a transposon mutation inactivating the mtrB gene resulted in the loss of Fe3þ and Mn4þ reduction. However, Myers and Myers recently established that a mtrB replacement mutant was defective in localizing proteins including OmcA and OmcB to the outer membrane and, therefore, suggested that MtrB is required for protein tra⁄cking to the outer membrane [163]. Coincidentally, MtrB of S. oneidensis MR-1 was also found to be required for the reduction of anthraquinone-2,6-disulfonate (AQDS), a compound commonly used as extracellular electron shuttle [164]. This ¢nding suggests that the reductase(s) responsible for transferring electrons to humic substances and AQDS is also located in the outer membrane. That the ferric and manganese reductase enzyme(s) is indeed associated with the outer membrane was also supported by DiChristina et al. [165]. These authors showed that the inactivation of a type II protein secretion apparatus that targets proteins to the outer membrane abolished both Fe3þ and Mn4þ reduction. What other components are suggested to be components of a respiratory chain to Fe3þ ? Myers and Myers demonstrated that iron respiration is dependent on the presence of menaquinone, since a menaquinone-de¢cient mutant of S. oneidensis MR-1 lost the ability to respire

FEMSRE 776 4-6-03

with Fe3þ as electron acceptor [166]. A screen for S. oneidensis MR-1 mutants defective in iron respiration revealed the involvement of several genes including cymA, mtrA and mtrC [154,155,162]. The latter two genes form an operon with mtrB described above [162]. All three genes encode heme c-containing proteins that are predicted to be located either in the outer membrane (MtrC) or in the periplasmic space (CymA and MtrA). While a deletion of mtrA had abolished electron transfer activity from formate to Fe3þ and Mn4þ by whole cells, ferric reductase activity assayed in the membrane fraction was not impaired [162] indicating that MtrA is not a component of the ferric reductase enzyme, but may serve as periplasmic electron shuttle. Whether MtrA is a speci¢c component of the iron and manganese respiratory pathway or also functions in shuttling electrons to nitrate and fumarate is not clear. In contrast, a mtrC deletion mutant not only had reduced electron transfer activity to Fe3þ , but was also impaired in ferric reductase activity catalyzed by the membrane fraction. It is noteworthy that membranes prepared from a mtrC mutant still retained about 30% of the wildtype ferric reductase activity. Therefore, MtrC may be associated with, or is, like MtrB, essential for the maturation of the ferric reductase enzyme in the outer membrane [162]. A cymA mutant is abolished in iron, nitrate and fumarate respiration, and diminished in manganese respiration [154,155]. CymA was puri¢ed from S. frigidimarina NCIMB400 [160]. The 20-kDa protein contains four lowspin hemes with midpoint reduction potentials of +10, 3108, 3136, and 3229 mV. Based on its homology to the periplasmic electron transfer proteins TorC, NapC and NirT from other bacteria, it was postulated that CymA in both Shewanella species functions to transfer electrons from menaquinol to several terminal reductases that are located in the periplasm or outer membrane [155,160]. Yet another c3 -type cytochrome encoded by cctA was identi¢ed in S. frigidimarina NCIMB400 as a periplasmic electron shuttle, speci¢cally involved in iron respiration [167]. A disruption of cctA almost completely abolished iron respiration, however, did not a¡ect growth with nitrate, fumarate, TMAO, DMSO and several other electron acceptors. A cctA homologue is also present in S. oneidensis MR-1 suggesting that it may have a similar function in this bacterium. A 150-kDa outer membrane protein was predicted by Lower et al. to contact and thus facilitate electron transfer to insoluble iron in S. oneidensis MR-1 [168]. This group used an innovative approach to monitor the interaction of live S. oneidensis MR-1 cells with goethite (K-FeOOH) on a nanoscale level by biological force microscopy. The bacteria were tethered via small beads to the end of a cantilever. When presented with the goethite, under anaerobic conditions that allowed iron respiration the energy for adhesion and the energy for retraction from the metal

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

were monitored. From the size of the retraction force the molecular mass of the outer membrane interacting protein was calculated. Whether the 150-kDa cytochrome c described earlier by Myers and Myers is the participating protein that contacts the metal [151] or whether this protein comprises a complex of several unknown subunits has yet to be established. To date, the identity of the dissimilatory ferric reductase and its relationship to the manganese reductase have yet to be established. The availability of the S. oneidensis MR-1 genome sequence will predictably facilitate future genetic and biochemical e¡orts to elucidate the nature of this important enzyme. 5.3. Ferric iron reduction in Geobacter species Also in Geobacter species, c-type cytochromes have been implicated to be involved in iron respiration [157,169]. In G. sulfurreducens, Gaspard et al. reported ferric reductase activity that was measured with NADH or reduced horse heart cytochrome c as electron donor and nitrilotriacetic acid-complexed Fe3þ (Fe3þ -NTA) as electron acceptor [157]. The NADH-dependent and cytochrome c-dependent activities were about 75^79% associated with the outer membrane fraction. Because the activity could be removed from whole cells by EDTA/KCl washes, Gaspard et al. suggested that the ferric reductase is probably peripheral to the outer membrane. The authors speculated that NADH is not the direct electron donor but most likely the electron donor to an electron transport chain originating in the cytoplasm. A c-type cytochrome was proposed to be part of the electron transfer chain to the extracellular ferric iron reductase in G. sulfurreducens [157]. Membraneassociated and soluble NADH-dependent ferric reductase activities were also reported for G. metallireducens and S. oneidensis MR-1, however, these enzymes have thus far not been further characterized [166,170,171]. Recently, both soluble and membrane-bound ferric iron-reducing enzymes were puri¢ed from G. sulfurreducens ([172,176]). The membrane-bound enzyme was detergent-solubilized and puri¢ed [172]. The enzyme complex has a native molecular mass of about 300 kDa and consists of at least ¢ve major polypeptides, one of which is a 90-kDa c-type cytochrome. Also FAD was present as a cofactor. The 90-kDa cytochrome c was determined to have a redox potential of about 3100 mV, suitable for electron transfer to the Fe3þ compounds utilized by G. sulfurreducens. Magnuson et al. demonstrated that the cytochrome c could be reduced with dithionite and reoxidized by Fe3þ -NTA or Fe3þ -oxyhydroxide. Based on this observation and inhibitor studies, the authors suggested that the ferric reductase enzyme complex consists of a NADH dehydrogenase and a cytochrome c terminal ferric reductase. This ¢nding is in concert with the peripheral ferric reductase activity that has been described by Gaspard et al. earlier, except that NADH oxidation was

FEMSRE 776 4-6-03

441

not postulated to be part of the ferric reductase enzyme complex itself [157]. Thus far, it is not clear whether the enzyme complex puri¢ed by Magnuson et al. is a membrane-intrinsic protein or whether it is associated with the outer membrane as was suggested by Gaspard et al. [157,172]. Whether the enzyme puri¢ed by Magnuson et al. is unequivocally a terminal ferric iron reductase in G. sulfurreducens has yet to be con¢rmed. Subsequently, Magnuson et al. puri¢ed the 90-kDa c-type cytochrome of the ferric reductase enzyme complex from the membrane fraction of G. sulfurreducens and con¢rmed its function in ferric iron reduction [173]. One gross redox potential of about 3190 mV was determined for the decaheme cytochrome c, similar to what had been determined earlier [172]. The cytochrome c is encoded by the ferA gene and displays weak homology to other known c cytochromes such as MtrA from S. oneidensis MR-1 [173]. A conserved lipid-attachment motif (LxxC) is present at the N-terminus of FerA. This motif has been shown to anchor the S. oneidensis OmcA protein to the outer membrane [152]. Therefore, Magnuson et al. suggest that FerA may also be attached to the outer membrane via its hydrophobic N-terminus while the remainder of the protein extends into the cell’s environment [173]. A 9.6-kDa periplasmic triheme cytochrome c7 , PpcA, in G. sulfurreducens appears to serve as electron mediator between acetate and Fe3þ , humic substances and uranium (VI) electron acceptors, but not to fumarate [174]. Lloyd et al. demonstrated that a ppcA deletion mutant was signi¢cantly reduced in its ability to grow under the selective conditions. This study clari¢es the disputed function of this protein in earlier reports [169,175]. A second ferric reductase of G. sulfurreducens was puri¢ed from the soluble fraction [176]. This enzyme is a NADPH-dependent Fe3þ -NTA reductase consisting of two subunits with molecular masses of 87 and 78 kDa. The protein contains FAD, iron and acid-labile sulfur suggesting the presence of several Fe^S centers. The amino acid sequences of both subunits were about 30% identical to the respective subunits of formate dehydrogenase from Moorella thermoacetica, however, the enzyme lacked any formate dehydrogenase activity. Whether this enzyme acts physiologically as ferric reductase or as ferredoxin:NADP oxidoreductase, as was suggested by Kaufmann and Lovley based on its high methyl viologen :NADP oxidoreductase activity, has yet to be established.

6. Conclusion This review provides an overview of the structurally very diverse enzymes involved in assimilatory and dissimilatory iron reduction. Ferric iron reductases serve an important role in iron assimilation of most organisms and in iron respiration of certain prokaryotic specialists. While the assimilatory-type bacterial and archaeal ferric reduc-

Cyaan Magenta Geel Zwart

442

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

tases either enzymatically or chemically reduce iron via a £avin cofactor, the eukaryal assimilatory-type enzymes involve b-type cytochromes and possibly also FAD. The only bacterial ferric reductase isolated thus far that could have a dissimilatory function is from G. sulfurreducens and contains a c-type cytochrome and FAD. The speci¢city of ferric reductases for various Fe3þ compounds is not clear. Several of the assimilatory yeast Fre enzymes can also reduce copper, while there appears to be a tight relationship between dissimilatory ferric and manganese reduction in certain bacteria. In assimilatory iron reduction, microorganisms must deal with iron insolubility and transport, and with the toxicity of the resulting Fe2þ under aerobic conditions. Therefore, many of these organisms have evolved speci¢c enzymes to synthesize siderophores for iron transport and proteins for intracellular storage. Chaperones as identi¢ed for copper assimilation [124,125] that might aid in tra⁄cking and targeting of intracellular Fe2þ have thus far not been found. In many organisms the assimilatory ferric reductases are membrane-bound or membrane-associated pointing to a role of these enzymes in coupling ferric reduction to iron transport. Since bacterial assimilatory ferric reductases are most often £avin reductases, it is not surprising that ferric reductase gene expression is, in general, not regulated by iron. This is in contrast to S. cerevisiae, where the FRE genes are tightly regulated in response to the environmental iron and copper availability. In the last 10 years considerable progress has been made to determine the identity and biochemical properties of various assimilatory ferric reductases. This includes the solution of two three-dimensional structures, those of the E. coli Fre and the A. fulgidus FeR enzymes. In S. cerevisiae, at least six putative ferric reductase enzymes have been identi¢ed by genetic methods. Due to their apparent instability, the biochemical characterization of these enzymes has been exceedingly di⁄cult. In the forthcoming years signi¢cant progress is expected in the elucidation of the structure, function and physiological role of the S. cerevisiae Fre proteins. Dissimilatory ferric iron reductases face the di⁄cult task of utilizing an insoluble electron acceptor. Di¡erent strategies have evolved that allow for the usage of ferric iron as electron acceptor for growth. Not all of these strategies are ferric iron-speci¢c but may also allow for the reduction of other organic or inorganic compounds present in the microbe’s environment. It is striking that the predominant proteins in electron transfer to Fe3þ are c-type cytochromes. Dependent on its coordination, heme c can assume a range of redox potentials suitable to bridge the redox span between a variety of electron donors to Fe3þ , and possibly also to other electron acceptors in branched electron transfer chains. In addition, c-type cytochromes are ideal for electron transfer between proteins, to and from quinones, and to the terminal insoluble ferric iron. In light of only one puri¢ed putative dissimilatory ferric

FEMSRE 776 4-6-03

reductase, very little is known about the possible diversity of these types of enzymes in the variety of iron-reducing bacteria. The recent completion of the S. oneidensis MR-1 and the G. sulfurreducens genomes will greatly facilitate the future identi¢cation of components including the terminal reductases involved in this important energy-generating process.

Acknowledgements I.S. was supported by NSF award MCB 0091351, E.J. by USPHS National Research Service Award GM07185, and S.d.V. by NOVEM (375001/0060). I.S. and S.d.V. were also supported by NATO (CLG 978616). We thank Robert Gunsalus for critically reading this manuscript.

References [1] Ehrlich, H.L. (1996). Geomicrobiology, 3rd edn. Marcel Dekker, New York. [2] Neilands, J.B. (1974) Iron and its Role in Microbial Physiology. Academic Press, London. [3] Guerinot, M.L. (1994) Microbial iron transport. Annu. Rev. Microbiol. 48, 743^772. [4] Braun, V. and Braun, M. (2002) Active transport of iron and siderophore antibiotics. Curr. Opin. Microbiol. 5, 194^201. [5] Crosa, J.H. and Walsh, C.T. (2002) Genetics and assembly line enzymology of siderophore biosynthesis in bacteria. Microbiol. Mol. Biol. Rev. 66, 223^249. [6] Visca, P., Leoni, L., Wilson, M.J. and Lamont, I.L. (2002) Iron transport and regulation, cell signaling and genomics: lessons from Escherichia coli and Pseudomonas. Mol. Microbiol. 45, 1177^1190. [7] Archibald, F.S. (1983) Lactobacillus plantarum, an organism not requiring iron. FEMS Microbiol. Lett. 19, 29^32. [8] Aisen, P., Enns, C. and Wessling-Resnick, M. (2001) Chemistry and biology of eukaryotic iron metabolism. Int. J. Biochem. Cell Biol. 33, 940^959. [9] Dix, D., Bridgham, J., Broderius, M. and Eide, D. (1997) Characterization of the FET4 protein of yeast. Evidence for a direct role in the transport of iron. J. Biol. Chem. 272, 11770^11777. [10] Dix, D.R., Bridgham, J.T., Broderius, M.A., Byersdorfer, C.A. and Eide, D.J. (1994) The FET4 gene encodes the low a⁄nity Fe(II) transport protein of Saccharomyces cerevisiae. J. Biol. Chem. 269, 26092^26099. [11] Askwith, C. and Kaplan, J. (1997) An oxidase-permease-based iron transport system in Schizosaccharomyces pombe and its expression in Saccharomyces cerevisiae. J. Biol. Chem. 272, 401^405. [12] Askwith, C. and Kaplan, J. (1998) Iron and copper transport in yeast and its relevance to human disease. Trends Biochem. Sci. 23, 135^ 138. [13] Hassett, R.F., Romeo, A.M. and Kosman, D.J. (1998) Regulation of high a⁄nity iron uptake in the yeast Saccharomyces cerevisiae. Role of dioxygen and Fe. J. Biol. Chem. 273, 7628^7636. [14] Hassett, R.F., Yuan, D.S. and Kosman, D.J. (1998) Spectral and kinetic properties of the Fet3 protein from Saccharomyces cerevisiae, a multinuclear copper ferroxidase enzyme. J. Biol. Chem. 273, 23274^ 23282. [15] Hassett, R., Dix, D.R., Eide, D.J. and Kosman, D.J. (2000) The Fe(II) permease Fet4p functions as a low a⁄nity copper transporter and supports normal copper tra⁄cking in Saccharomyces cerevisiae. Biochem. J. 351, 477^484.

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447 [16] Jordan, I. and Kaplan, J. (1994) The mammalian transferrin-independent iron transport system may involve a surface ferrireductase activity. Biochem. J. 302, 875^879. [17] Lesuisse, E., Raguzzi, F. and Crichton, R.R. (1987) Iron uptake by the yeast Saccharomyces cerevisiae : involvement of a reduction step. J. Gen. Microbiol. 133, 3229^3236. [18] de Silva, D.M., Askwith, C.C., Eide, D. and Kaplan, J. (1995) The FET3 gene product required for high a⁄nity iron transport in yeast is a cell surface ferroxidase. J. Biol. Chem. 270, 1098^1101. [19] de Silva, D., Davis-Kaplan, S., Fergestad, J. and Kaplan, J. (1997) Puri¢cation and characterization of Fet3 protein, a yeast homologue of ceruloplasmin. J. Biol. Chem. 272, 14208^14213. [20] Stearman, R., Yuan, D.S., Yamaguchi-Iwai, Y., Klausner, R.D. and Dancis, A. (1996) A permease-oxidase complex involved in high-af¢nity iron uptake in yeast. Science 271, 1552^1557. [21] Arosio, P. and Levi, S. (2002) Ferritin, iron homeostasis, and oxidative damage. Free Radic. Biol. Med. 33, 457^463. [22] Andrews, S.C. (1998) Iron storage in bacteria. Adv. Microb. Physiol. 40, 281^351. [23] Braun, V. and Killmann, H. (1999) Bacterial solutions to the iron supply problem. Trends Biochem. Sci. 24, 104^109. [24] Neilands, J.B. (1995) Siderophores ^ structure and function of microbial iron transport compounds. J. Biol. Chem. 270, 26723^ 26726. [25] Vargas, M., Kashe¢, K., Blunt-Harris, E.L. and Lovley, D.R. (1998) Microbiological evidence for Fe(III) reduction on early earth. Nature 395, 65^67. [26] Lovley, D.R., Fraga, J.L., BluntHarris, E.L., Hayes, L.A., Phillips, E.J.P. and Coates, J.D. (1998) Humic substances as a mediator for microbially catalyzed metal reduction. Acta Hydrochim. Hydrobiol. 26, 152^157. [27] Lovley, D.R. and Coates, J.D. (2000) Novel forms of anaerobic respiration of environmental relevance. Curr. Opin. Microbiol. 3, 252^ 256. [28] Straub, K.L., Benz, M. and Schink, B. (2001) Iron metabolism in anoxic environments at near neutral pH. FEMS Microbiol. Ecol. 34, 181^186. [29] Lovley, D.R. (1993) Dissimilatory metal reduction. Annu. Rev. Microbiol. 47, 263^290. [30] Lovley, D.R., Coates, J.D., Bluntharris, E.L., Phillips, E.J.P. and Woodward, J.C. (1996) Humic substances as electron acceptors for microbial respiration. Nature 382, 445^448. [31] Lovley, D.R. and Blunt-Harris, E.L. (1999) Role of humic-bound iron as an electron transfer agent in dissimilatory Fe(III) reduction. Appl. Environ. Microbiol. 65, 4252^4254. [32] Richardson, D.J. (2000) Bacterial respiration : a £exible process for a changing environment. Microbiology 146, 551^571. [33] Widdel, F., Schnell, S., Heising, S., Ehrenreich, A., Assmus, B. and Schink, B. (1993) Ferrous iron oxidation by anoxygenic phototrophic bacteria. Nature 362, 834^836. [34] Ehrenreich, A. and Widdel, F. (1994) Anaerobic oxidation of ferrous iron by purple bacteria, a new type of phototrophic metabolism. Appl. Environ. Microbiol. 60, 4517^4526. [35] Lodge, J.S., Gaines, C.G., Arceneaux, J.E. and Byers, B.R. (1982) Ferrisiderophore reductase in Agrobacterium tumefaciens. J. Bacteriol. 149, 771^774. [36] Huyer, M. and Page, W.J. (1989) Ferric reductase activity in Azotobacter vinelandii and its inhibition by Zn2þ . J. Bacteriol. 171, 4031^ 4037. [37] Arceneaux, J.E. and Byers, B.R. (1980) Ferrisiderophore reductase activity in Bacillus megaterium. J. Bacteriol. 141, 715^721. [38] Gaines, C.G., Lodge, J.S., Arceneaux, J.E. and Byers, B.R. (1981) Ferrisiderophore reductase activity associated with an aromatic biosynthetic enzyme complex in Bacillus subtilis. J. Bacteriol. 148, 527^ 533. [39] Lodge, J.S., Gaines, C.G., Arceneaux, J.E. and Byers, B.R. (1980) Non-hydrolytic release of iron from ferrienterobactin analogs by ex-

FEMSRE 776 4-6-03

[40]

[41]

[42]

[43]

[44] [45]

[46]

[47]

[48]

[49]

[50]

[51] [52] [53]

[54]

[55] [56] [57]

[58]

[59] [60]

[61]

443

tracts of Bacillus subtilis. Biochem. Biophys. Res. Commun. 97, 1291^1295. Coves, J. and Fontecave, M. (1993) Reduction and mobilization of iron by a NAD(P)H:£avin oxidoreductase from Escherichia coli. Eur. J. Biochem. 211, 635^641. Fischer, E., Strehlow, B., Hartz, D. and Braun, V. (1990) Soluble and membrane-bound ferrisiderophore reductases of Escherichia coli K-12. Arch. Microbiol. 153, 329^336. Fontecave, M., Eliasson, R. and Reichard, P. (1987) NAD(P)H :£avin oxidoreductase of Escherichia coli. A ferric iron reductase participating in the generation of the free radical of ribonucleotide reductase. J. Biol. Chem. 262, 12325^12331. Johnson, W., Varner, L. and Poch, M. (1991) Acquisition of iron by Legionella pneumophila ^ role of iron reductase. Infect. Immun. 59, 2376^2381. Poch, M.T. and Johnson, W. (1993) Ferric reductases of Legionella pneumophila. Biometals 6, 107^114. Barchini, E. and Cowart, R.E. (1996) Extracellular iron reductase activity produced by Listeria monocytogenes. Arch. Microbiol. 166, 51^57. Coulanges, V., Andre, P., Ziegler, O., Buchheit, L. and Vidon, D.J.M. (1997) Utilization of iron-catecholamine complexes involving ferric reductase activity in Listeria monocytogenes. Infect. Immun. 65, 2778^2785. Deneer, H.G., Healey, V. and Boychuk, I. (1995) Reduction of exogenous ferric iron by a surface-associated ferric reductase of Listeria spp.. Microbiology 141, 1985^1992. Noguchi, Y., Fujiwara, T., Yoshimatsu, K. and Fukumori, Y. (1999) Iron reductase for magnetite synthesis in the magnetotactic bacterium Magnetospirillum magnetotacticum. J. Bacteriol. 181, 2142^2147. Hohmuth, M., Valentin-Weiganz, P., Rohde, M. and Gerlach, G.-F. (1998) Identi¢cation and characterization of a novel extracellular ferric reductase from Mycobacterium paratuberculosis. Infect. Immun. 66, 710^716. MacCready, K.A. and Ratledge, C. (1979) Ferrimycobactin reductase activity from Mycobacterium smegmatis. J. Gen. Microbiol. 113, 67^ 72. Le Faou, A. and Morse, S. (1991) Characterization of a soluble ferric reductase from Neisseria gonorrhoeae. Biol. Met. 4, 126^131. Cox, C.D. (1980) Iron reductases from Pseudomonas aeruginosa. J. Bacteriol. 141, 199^204. Halle, F. and Meyer, J.M. (1992) Iron release from ferrisiderophores ^ a multi-step mechanism involving a NADH:FMN oxidoreductase and a chemical reduction by FMNH2. Eur. J. Biochem. 209, 621^ 627. Halle, F. and Meyer, J.M. (1992) Ferrisiderophore reductases of Pseudomonas ^ puri¢cation, properties and cellular location of the Pseudomonas aeruginosa ferripyoverdine reductase. Eur. J. Biochem. 209, 613^620. Halle, F. and Meyer, J.M. (1989) Ferripyoverdine-reductase activity in Pseudomonas £uorescens. Biol. Met. 2, 18^24. Moody, M.D. and Dailey, H.A. (1985) Ferric iron reductase of Rhodopseudomonas sphaeroides. J. Bacteriol. 163, 1120^1125. Scott, D., Chan, E.C. and Siboo, R. (1996) Iron acquisition by oral hemolytic spirochetes : isolation of a hemin-binding protein and identi¢cation of iron reductase activity. Can. J. Microbiol. 42, 1072^1079. Mazoy, R., Lopez, E.M., Fouz, B., Amaro, C. and Lemos, M.L. (1999) Ferric-reductase activities in Vibrio vulni¢cus biotypes 1 and 2. FEMS Microbiol. Lett. 172, 205^211. Cox, C.D. and Graham, R. (1979) Isolation of an iron-binding compound from Pseudomonas aeruginosa. J. Bacteriol. 137, 357^364. Portnoy, D.A., Chakraborty, T., Goebel, W. and Cossart, P. (1992) Molecular determinants of Listeria monocytogenes pathogenesis. Infect. Immun. 60, 1263^1267. Cowart, R.E. and Foster, B.G. (1985) Di¡erential e¡ects of iron on the growth of Listeria monocytogenes: minimum requirements and mechanism of acquisition. J. Infect. Dis. 151, 721^730.

Cyaan Magenta Geel Zwart

444

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

[62] Vartivarian, S.E. and Cowart, R.E. (1999) Extracellular iron reductases: Identi¢cation of a new class of enzymes by siderophore-producing microorganisms. Arch. Biochem. Biophys. 364, 75^82. [63] Fenton, M.J. and Vermeulen, M.W. (1996) Immunopathology of tuberculosis ^ roles of macrophages and monocytes. Infect. Immun. 64, 683^690. [64] Dailey, H.A. and Lascelles, J. (1977) Reduction of iron and synthesis of protoheme by Spirillum itersonii and other organisms. J. Bacteriol. 129, 815^820. [65] Lascelles, J. and Burke, K.A. (1978) Reduction of ferric iron by L-lactate and DL-glycerol-3-phosphate in membrane preparations from Staphylococcus aureus and interaction with the nitrate reductase system. J. Bacteriol. 134, 585^589. [66] Jones, T., Spencer, R. and Walsh, C. (1978) Mechanism and kinetics of iron release from ferritin by dihydro£avins and dihydro£avin analogues. Biochemistry 17, 4011^4017. [67] Fonteclave, M., Coves, J. and Pierre, J.-L. (1994) Ferric reductases or £avin reductases ? Biometals 7, 3^8. [68] Pierre, J.L., Fontecave, M. and Crichton, R.R. (2002) Chemistry for an essential biological process : the reduction of ferric iron. Biometals 15, 341^346. [69] Prinz, W.A., Aslund, F., Holmgren, A. and Beckwith, J. (1997) The role of the thioredoxin and glutaredoxin pathways in reducing protein disul¢de bonds in the Escherichia coli cytoplasm. J. Biol. Chem. 272, 15661^15667. [70] Hasan, N. and Nester, E.W. (1978) Puri¢cation and characterization of NADPH-dependent £avin reductase: an enzyme required for the activation of chorismate synthase in Bacillus subtilis. J. Biol. Chem. 253, 4987^4992. [71] Bornemann, S., Lowe, D.J. and Thorneley, R.N.F. (1996) The transient kinetics of Escherichia coli chorismate synthase ^ substrate, consumption, product formation, phosphate dissociation, and characterization of a £avin intermediate. Biochemistry 35, 9907^ 9916. [72] Macheroux, P., Schmid, J., Amrhein, N. and Schaller, A. (1999) A unique reaction in a common pathway : mechanism and function of chorismate synthase in the shikimate pathway. Planta 207, 325^334. [73] Downer, D.N., Davis, W.B. and Byers, B.R. (1970) Repression of phenolic acid synthesizing enzymes and its relation to iron uptake in Bacillus subtilis. J. Bacteriol. 101, 181^187. [74] Coves, J., Laulhere, J.P. and Fontecave, M. (1997) The role of exogenous iron in the activation of ribonucleotide reductase from Escherichia coli. J. Biol. Inorg. Chem. 2, 418^426. [75] Coves, J., Delon, B., Climent, I., Sjoberg, B.M. and Fontecave, M. (1995) Enzymic and chemical reduction of the iron center of the Escherichia coli ribonucleotide reductase protein R2 the role of the C-terminus. Eur. J. Biochem. 233, 357^363. [76] Puzon, G.J., Petersen, J.N., Roberts, A.G., Kramer, D.M. and Xun, L.Y. (2002) A bacterial £avin reductase system reduces chromate to a soluble chromium(III)-NAD(+) complex. Biochem. Biophys. Res. Commun. 294, 76^81. [77] Russ, R., Rau, J. and Stolz, A. (2000) The function of cytoplasmic £avin reductases in the reduction of azo dyes by bacteria. Appl. Environ. Microbiol. 66, 1429^1434. [78] Hantke, K. (2001) Iron and metal regulation in bacteria. Curr. Opin. Microbiol. 4, 172^177. [79] Panina, E.M., Mironov, A.A. and Gelfand, M.S. (2001) Comparative analysis of FUR regulons in gamma-proteobacteria. Nucleic Acid Res. 29, 5195^5206. [80] Ratledge, C. and Dover, L.G. (2000) Iron metabolism in pathogenic bacteria. Annu. Rev. Microbiol. 54, 881^941. [81] Schuler, D. (1999) Formation of magnetosomes in magnetotactic bacteria. J. Microbiol. Biotechnol. 1, 79^86. [82] Vadas, A., Monbouquette, H.G. and Schro«der, I. (1999) Identi¢cation and characterization of a novel ferric reductase from the hyperthermophilic Archaeon Archaeoglobus fulgidus. J. Biol. Chem. 274, 36715^36721.

FEMSRE 776 4-6-03

[83] Kashe¢, K., Tor, J.M., Holmes, D.E., Gaw Van Praagh, C.V., Reysenbach, A.L. and Lovley, D.R. (2002) Geoglobus ahangari gen. nov., sp. nov., a novel hyperthermophilic archaeon capable of oxidizing organic acids and growing autotrophically on hydrogen with Fe(III) serving as the sole electron acceptor. Int. J. Syst. Evol. Microbiol. 52, 719^728. [84] Eschenbrenner, M., Coves, J. and Fontecave, M. (1994) Ferric reductases in Escherichia coli: the contribution of the haemoglobinlike protein. Biochem. Biophys. Res. Commun. 198, 127^131. [85] Chiu, H.J., Johnson, E., Schro«der, I. and Rees, D.C. (2001) Crystal structures of a novel ferric reductase from the hyperthermophilic archaeon Archaeoglobus fulgidus and its complex with NADP(+). Structure 9, 311^319. [86] Ingelman, M., Ramaswamy, S., Niviere, V., Fontecave, M. and Eklund, H. (1999) Crystal structure of NAD(P)H:£avin oxidoreductase from Escherichia coli. Biochemistry 38, 7040^7049. [87] Dancis, A., Roman, D., Anderson, G., Hinnebusch, A. and Klausner, R. (1992) Ferric reductase of Saccharomyces cerevisiae : Molecular characterization, role in iron uptake, and transcriptional control by iron. Proc. Natl. Acad. Sci. USA 89, 3869^3873. [88] Spyrou, G., Haggard-Ljungquist, E., Krook, M., Jornval, H., Nilsson, E. and Reichard, P. (1991) Characterization of the £avin reductase gene (fre) of Escherichia coli and construction of a plasmid for overproduction of the enzyme. J. Bacteriol. 173, 3673^3679. [89] Matsumura, M., Signor, G. and Matthews, B.W. (1989) Substantial increase of protein stability by multiple disulphide bonds. Nature 342, 291^293. [90] Liepinsh, E., Kitamura, M., Murakami, T., Nakaya, T. and Otting, G. (1997) Pathway of chymotrypsin evolution suggested by the structure of the FMN-binding protein from Desulfovibrio vulgaris (Miyazaki F). Nat. Struct. Biol. 4, 975^979. [91] Suto, K., Kawagoe, K., Shibata, N., Morimoto, Y., Higuchi, Y., Kitamura, M., Nakaya, T. and Yasuoka, N. (2000) How do the X-ray structure and the NMR structure of FMN-binding protein di¡er? Acta Crystallogr. D-Biol. Cryst. 56, 368^371. [92] Murzin, A.G. (1998) Probable circular permutation in the £avinbinding domain. Nat. Struct. Biol. 5, 101. [93] Liepinsh, E., Kitamura, M., Murakami, T., Nakaya, T. and Otting, G. (1998) Common ancestor of serine proteases and £avin-binding domains. Nat. Struct. Biol. 5, 102^103. [94] Fieschi, F., Niviere, V., Frier, C., Decout, J.-L. and Fontecave, M. (1995) The mechanisms and substrate speci¢city of the NADPH:£avin oxidoreductase from Escherichia coli. J. Biol. Chem. 270, 30392^30400. [95] Niviere, V., Fieschi, F., Decout, J. and Fonteclave, M. (1999) The NAD(P)H :£avin oxidoreductase from Escherichia coli. Evidence for a new mode of binding for reduced pyridine nucleotides. J. Biol. Chem. 274, 18252^18260. [96] Fraaije, M.W. and Mattevi, A. (2000) Flavoenzymes : diverse catalysts with recurrent features. Trends Biochem. Sci. 25, 126^132. [97] Li, L. and Kaplan, J. (1998) Defects in the yeast high a⁄nity iron transport system result in increased metal sensitivity because of the increased expression of transporters with a broad transition metal speci¢city. J. Biol. Chem. 273, 22181^22187. [98] Kosman, D.J. (1994) Reductive iron assimilation in Saccharomyces cerevisiae. In: Metal Ions in Fungi (Winkelman, G. and Winge, D.R., Eds.), pp. 1^38. Marcel Dekker, New York. [99] Lesuisse, E. and Labbe, P. (1994) Reductive iron assimilation in Saccharomyces cerevisiae. In: Metal Ions in Fungi (Winkelman, G. and Winge, D.R., Eds.), pp. 149^178. Marcel Dekker, New York. [100] Yun, C.W., Ferea, T., Rashford, J., Ardon, O., Brown, P.O., Botstein, D., Kaplan, J. and Philpott, C.C. (2000) Desferrioxaminemediated iron uptake in Saccharomyces cerevisiae. Evidence for two pathways of iron uptake. J. Biol. Chem. 275, 10709^10715. [101] Yun, C.W., Tiedeman, J.S., Moore, R.E. and Philpott, C.C. (2000) Siderophore-iron uptake in Saccharomyces cerevisiae. Identi¢cation

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

[102]

[103]

[104]

[105]

[106]

[107]

[108]

[109]

[110]

[111]

[112]

[113] [114]

[115]

[116]

[117]

[118]

of ferrichrome and fusarinine transporters. J. Biol. Chem. 275, 16354^16359. Yun, C.W., Bauler, M., Moore, R.E., Klebba, P.E. and Philpott, C.C. (2001) The role of the FRE family of plasma membrane reductases in the uptake of siderophore-iron in Saccharomyces cerevisiae. J. Biol. Chem. 276, 10218^10223. Lesuisse, E., Blaiseau, P.L., Dancis, A. and Camadro, J.M. (2001) Siderophore uptake and use by the yeast Saccharomyces cerevisiae. Microbiology 147, 289^298. Lesuisse, E., Simon-Casteras, M. and Labbe, P. (1998) Siderophoremediated iron uptake in Saccharomyces cerevisiae : the SIT1 gene encodes a ferrioxamine B permease that belongs to the major facilitator superfamily. Microbiology 144, 3455^3462. Dancis, A., Klausner, R.D., Hinnebusch, A.G. and Barriocanal, J.G. (1990) Genetic evidence that ferric reductase is required for iron uptake in Saccharomyces cerevisiae. Mol. Cell. Biol. 10, 2294^2301. Anderson, G.J., Lesuisse, E., Dancis, A., Roman, D.G., Labbe, P. and Klausner, R.D. (1992) Ferric iron reduction and iron assimilation in Saccharomyces cerevisiae. J. Inorg. Biochem. 47, 249^ 255. Lesuisse, E., Casterassimon, M. and Labbe, P. (1996) Evidence for the Saccharomyces cerevisiae ferrireductase system being a multicomponent electron transport chain. J. Biol. Chem. 271, 13578^ 13583. Lesuisse, E., Crichton, R.R. and Labbe, P. (1990) Iron-reductases in the yeast Saccharomyces cerevisiae. Biochim. Biophys. Acta 1038, 253^259. Lesuisse, E., Casteras-Simon, M. and Labbe, P. (1997) Cytochrome P-450 reductase is responsible for the ferrireductase activity associated with isolated plasma membranes of Saccharomyces cerevisiae. FEMS Microbiol. Lett. 156, 147^152. Finegold, A.A., Shatwell, K.P., Segal, A.W., Klausner, R.D. and Dancis, A. (1996) Intramembrane bis-heme motif for transmembrane electron transport conserved in a yeast iron reductase and the human NADPH oxidase. J. Biol. Chem. 271, 31021^31024. Shatwell, K.P., Dancis, A., Cross, A.R., Klausner, R.D. and Segal, A.W. (1996) The FRE1 ferric reductase of Saccharomyces cerevisiae is a cytochrome b similar to that of NADPH oxidase. J. Biol. Chem. 271, 14240^14244. Roman, D.G., Dancis, A., Anderson, G.J. and Klausner, R.D. (1993) The ¢ssion yeast ferric reductase gene frp1+ is required for ferric iron uptake and encodes a protein that is homologous to the gp91-phox subunit of the human NADPH phagocyte oxidoreductase. Mol. Cell. Biol. 13, 4342^4350. Wientjes, F.B. and Segal, A.W. (1995) NADPH oxidase and the respiratory burst. Semin. Cell. Biol. 6, 357^365. Zhang, Z., Huang, L., Shulmeister, V.M., Chi, Y.I., Kim, K.K., Hung, L.W., Crofts, A.R., Berry, E.A. and Kim, S.H. (1998) Electron transfer by domain movement in cytochrome bc1 . Nature 392, 677^684. Henderson, L.M., Banting, G. and Chappell, J.B. (1995) The arachidonate-activable, NADPH oxidase-associated Hþ channel. Evidence that gp91-phox functions as an essential part of the channel. J. Biol. Chem. 270, 5909^5916. Lesuisse, E., Casteras-Simon, M. and Labbe, P. (1995) Ferrireductase activity in Saccharomyces cerevisiae and other fungi: colorimetric assays on agar plates. Anal. Biochem. 226, 375^377. Cross, A.R., Rae, J. and Curnutte, J.T. (1995) Cytochrome b-245 of the neutrophil superoxide-generating system contains two nonidentical hemes. Potentiometric studies of a mutant form of gp91phox. J. Biol. Chem. 270, 17075^17077. Boxman, A.W., Dobbelmann, J. and Borst-Pauwels, G.W. (1984) Possible energization of Kþ accumulation into metabolizing yeast by the protonmotive force. Binding correction to be applied in the calculation of the yeast membrane potential from tetraphenylphosphonium distribution. Biochim. Biophys. Acta 772, 51^57.

FEMSRE 776 4-6-03

445

[119] Cherry, J.M., Ball, C., Weng, S., Juvik, G., Schmidt, R., Adler, C., Dunn, B., Dwight, S., Riles, L., Mortimer, R.K. and Botstein, D. (1997) Genetic and physical maps of Saccharomyces cerevisiae. Nature 387, 67^73. [120] Wood, V., Gwilliam, R., Rajandream, M.A., Lyne, M., Lyne, R., Stewart, A., Sgouros, J., Peat, N., Hayles, J., Baker, S., Basham, D., Bowman, S., Brooks, K., Brown, D., Brown, S., Chillingworth, T., Churcher, C., Collins, M., Connor, R., Cronin, A., Davis, P., Feltwell, T., Fraser, A., Gentles, S., Goble, A., Hamlin, N., Harris, D., Hidalgo, J., Hodgson, G., Holroyd, S., Hornsby, T., Howarth, S., Huckle, E.J., Hunt, S., Jagels, K., James, K., Jones, L., Jones, M., Leather, S., McDonald, S., McLean, J., Mooney, P., Moule, S., Mungall, K., Murphy, L., Niblett, D., Odell, C., Oliver, K., O’Neil, S., Pearson, D., Quail, M.A., Rabbinowitsch, E., Rutherford, K., Rutter, S., Saunders, D., Seeger, K., Sharp, S., Skelton, J., Simmonds, M., Squares, R., Squares, S., Stevens, K., Taylor, K., Taylor, R.G., Tivey, A., Walsh, S., Warren, T., Whitehead, S., Woodward, J., Volckaert, G., Aert, R., Robben, J., Grymonprez, B., Weltjens, I., Vanstreels, E., Rieger, M., Schafer, M., Muller-Auer, S., Gabel, C., Fuchs, M., Fritzc, C., Holzer, E., Moestl, D., Hilbert, H., Borzym, K., Langer, I., Beck, A., Lehrach, H., Reinhardt, R., Pohl, T.M., Eger, P., Zimmermann, W., Wedler, H., Wambutt, R., Purnelle, B., Go¡eau, A., Cadieu, E., Dreano, S., Gloux, S., Lelaure, V., Mottier, S., Galibert, F., Aves, S.J., Xiang, Z., Hunt, C., Moore, K., Hurst, S.M., Lucas, M., Rochet, M., Gaillardin, C., Tallada, V.A., Garzon, A., Thode, G., Daga, R.R., Cruzado, L., Jimenez, J., Sanchez, M., del Rey, F., Benito, J., Dominguez, A., Revuelta, J.L., Moreno, S., Armstrong, J., Forsburg, S.L., Cerrutti, L., Lowe, T., McCombie, W.R., Paulsen, I., Potashkin, J., Shpakovski, G.V., Ussery, D., Barrell, B.G. and Nurse, P. (2002) The genome sequence of Schizosaccharomyces pombe. Nature 415, 871^ 880. [121] Hammacott, J.E., Williams, P.H. and Cashmore, A.M. (2000) Candida albicans CFL1 encodes a functional ferric reductase activity that can rescue a Saccharomyces cerevisiae fre1 mutant. Microbiology 146, 869^876. [122] Hassett, R. and Kosman, D.J. (1995) Evidence for Cu(II) reduction as a component of copper uptake by Saccharomyces cerevisiae. J. Biol. Chem. 270, 128^134. [123] Georgatsou, E., Mavrogiannis, L.A., Fragiadakis, G.S. and Alexandraki, D. (1997) The yeast Fre1p/Fre2p cupric reductases facilitate copper uptake and are regulated by the copper-modulated Mac1p activator. J. Biol. Chem. 272, 13786^13792. [124] Rae, T.D., Schmidt, P.J., Pufahl, R.A., Culotta, V.C. and O’Halloran, T.V. (1999) Undetectable intracellular free copper: the requirement of a copper chaperone for superoxide dismutase. Science 284, 805^808. [125] Valentine, J.S. and Gralla, E.B. (1997) Delivering copper inside yeast and human cells. Science 278, 817^818. [126] Jungmann, J., Reins, H.A., Lee, J., Romeo, A., Hassett, R., Kosman, D. and Jentsch, S. (1993) MAC1, a nuclear regulatory protein related to Cu-dependent transcription factors is involved in Cu/Fe utilization and stress resistance in yeast. EMBO J. 12, 5051^ 5056. [127] Martins, L.J., Jensen, L.T., Simon, J.R., Keller, G.L., Winge, D.R. and Simons, J.R. (1998) Metalloregulation of FRE1 and FRE2 homologs in Saccharomyces cerevisiae. J. Biol. Chem. 273, 23716^ 23721. [128] Yamaguchi-Iwai, Y., Dancis, A. and Klausner, R.D. (1995) AFT1: a mediator of iron regulated transcriptional control in Saccharomyces cerevisiae. EMBO J. 14, 1231^1239. [129] Yamaguchi-Iwai, Y., Stearman, R., Dancis, A. and Klausner, R.D. (1996) Iron-regulated DNA binding by the AFT1 protein controls the iron regulon in yeast. EMBO J. 15, 3377^3384. [130] Yamaguchi-Iwai, Y., Ueta, R., Fukunaka, A. and Sasaki, R. (2002) Subcellular localization of Aft1 transcription factor responds to iron status in Saccharomyces cerevisiae. J. Biol. Chem. 277, 18914^18918.

Cyaan Magenta Geel Zwart

446

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447

[131] Blaiseau, P.L., Lesuisse, E. and Camadro, J.M. (2001) Aft2p, a novel iron-regulated transcription activator that modulates, with Aft1p, intracellular iron use and resistance to oxidative stress in yeast. J. Biol. Chem. 276, 34221^34226. [132] Rutherford, J.C., Jaron, S., Ray, E., Brown, P.O. and Winge, D.R. (2001) A second iron-regulatory system in yeast independent of Aft1p. Proc. Natl. Acad. Sci. USA 98, 14322^14327. [133] Yamaguchi-Iwai, Y., Serpe, M., Haile, D., Yang, W., Kosman, D.J., Klausner, R.D. and Dancis, A. (1997) Homeostatic regulation of copper uptake in yeast via direct binding of MAC1 protein to upstream regulatory sequences of FRE1 and CTR1. J. Biol. Chem. 272, 17711^17718. [134] Brown, K.R., Keller, G.L., Pickering, I.J., Harris, H.H., George, G.N. and Winge, D.R. (2002) Structures of the cuprous-thiolate clusters of the Mac1 and Ace1 transcriptional activators. Biochemistry 41, 6469^6476. [135] Lovley, D.R. (2002) Dissimilatory metal reduction: from early life to bioremediation. ASM News 68, 231^237. [136] Lovley, D.R., Kashe¢, K., Vargas, M., Tor, J.M. and Blunt-Harris, E.L. (2000) Reduction of humic substances and Fe(III) by hyperthermophilic microorganism. Chem. Geol. 169, 289^298. [137] Nevin, K.P. and Lovley, D.R. (2002) Mechanisms for accessing insoluble Fe(III) oxide during dissimilatory Fe(III) reduction by Geothrix fermentans. Appl. Environ. Microbiol. 68, 2294^2299. [138] Newman, D.K. and Kolter, R. (2000) A role for excreted quinones in extracellular electron transfer. Nature 405, 94^97. [139] Nevin, K.P. and Lovley, D.R. (2002) Mechanism for Fe(III) reduction in sedimentary environments. Geomicrobiol. J. 19, 141^159. [140] Turick, C.E., Tisa, L.S. and Caccavo, F. (2002) Melanin production and use as a soluble electron shuttle for Fe(III) oxide reduction and as a terminal electron acceptor by Shewanella algae BrY. Appl. Environ. Microbiol. 68, 2436^2444. [141] Hernandez, M.E. and Newman, D.K. (2001) Extracellular electron transfer. Cell. Mol. Life Sci. 58, 1562^1571. [142] Nevin, K.P. and Lovley, D.R. (2000) Lack of production of electron-shuttling compounds or solubilization of Fe(III) during reduction of insoluble Fe(III) oxide by Geobacter metallireducens. Appl. Environ. Microbiol. 66, 2248^2251. [143] Childers, S.E., Ciufo, S. and Lovley, D.R. (2002) Geobacter metallireducens accesses insoluble Fe(III) oxide by chemotaxis. Nature 416, 767^769. [144] Arnold, R.G., Ho¡mann, M.R., Dichristina, T.J. and Picardal, F.W. (1990) Regulation of dissimilatory Fe(III) reduction activity in Shewanella putrefaciens. Appl. Environ. Microbiol. 56, 2811^ 2817. [145] Caccavo, F. and Das, A. (2002) Adhesion of dissimilatory Fe(III)reducing bacteria to Fe(III) minerals. Geomicrobiol. J. 19, 161^177. [146] Das, A. and Caccavo, F. (2001) Adhesion of the dissimilatory Fe(III)-reducing bacterium Shewanella alga BrY to crystalline Fe(III) oxides. Curr. Microbiol. 42, 151^154. [147] Caccavo Jr., F. (1999) Protein-mediated adhesion of the dissimilatory Fe(III)-reducing bacterium Shewanella alga BrY to hydrous ferric oxide. Appl. Environ. Microbiol. 65, 5017^5022. [148] Das, A. and Caccavo Jr., F. (2000) Dissimilatory Fe(III) oxide reduction by Shewanella alga BrY requires adhesion. Curr. Microbiol. 40, 344^347. [149] Venkateswaran, K., Moser, D.P., Dollhopf, M.E., Lies, D.P., Saffarini, D.A., MacGregor, B.J., Ringelberg, D.B., White, D.C., Nishijima, M., Sano, H., Burghardt, J., Stackebrandt, E. and Nealson, K.H. (1999) Polyphasic taxonomy of the genus Shewanella and description of Shewanella oneidensis sp. nov.. Int. J. Syst. Bacteriol. 49, 705^724. [150] Myers, C.R. and Myers, J.M. (1994) Ferric iron reduction-linked growth yields of Shewanella putrefaciens MR-1. J. Appl. Bacteriol. 76, 253^258. [151] Myers, C.R. and Myers, J.M. (1997) Outer membrane cytochromes of Shewanella putrefaciens MR-1: spectral analysis, and puri¢cation

FEMSRE 776 4-6-03

[152]

[153]

[154]

[155]

[156]

[157]

[158]

[159]

[160]

[161]

[162]

[163]

[164]

[165]

[166]

[167]

[168]

of the 83-kDa c-type cytochrome. Biochim. Biophys. Acta 1326, 307^318. Myers, J.M. and Myers, C.R. (1998) Isolation and sequence of omcA, a gene encoding a decaheme outer membrane cytochrome c of Shewanella putrefaciens MR-1, and detection of omcA homologs in other strains of S. putrefaciens. Biochim. Biophys. Acta 1373, 237^251. Tsapin, A.I., Nealson, K.H., Meyers, T., Cusanovich, M.A., Van Beuumen, J., Crosby, L.D., Feinberg, B.A. and Zhang, C. (1996) Puri¢cation and properties of a low-redox-potential tetraheme cytochrome c3 from Shewanella putrefaciens. J. Bacteriol. 178, 6386^ 6388. Myers, C.R. and Myers, J.M. (1997) Cloning and sequence of cymA a gene encoding a tetraheme cytochrome c required for reduction of iron(III), fumarate, and nitrate by Shewanella putrefaciens MR-1. J. Bacteriol. 179, 1143^1152. Myers, J.M. and Myers, C.R. (2000) Role of the tetraheme cytochrome CymA in anaerobic electron transport in cells of Shewanella putrefaciens MR-1 with normal levels of menaquinone. J. Bacteriol. 182, 67^75. Myers, C.R. and Myers, J.M. (1992) Localization of cytochromes to the outer membrane of anaerobically grown Shewanella putrefaciens MR-1. J. Bacteriol. 174, 3429^3438. Gaspard, S., Vazquez, F. and Holliger, C. (1998) Localization and solubilization of the iron(III) reductase of Geobacter sulfurreducens. Appl. Environ. Microbiol. 64, 3188^3194. Myers, J.M. and Myers, C.R. (2001) Role for outer membrane cytochromes OmcA and OmcB of Shewanella putrefaciens MR-1 in reduction of manganese dioxide. Appl. Environ. Microbiol. 67, 260^269. Myers, J.M. and Myers, C.R. (2002) Genetic complementation of an outer membrane cytochrome omcB mutant of Shewanella putrefaciens MR-1 requires omcB plus downstream DNA. Appl. Environ. Microbiol. 68, 2781^2793. Field, S.J., Dobbin, P.S., Cheesman, M.R., Watmough, N.J., Thomson, A.J. and Richardson, D.J. (2000) Puri¢cation and magnetooptical spectroscopic characterization of cytoplasmic membrane and outer membrane multiheme c-type cytochromes from Shewanella frigidimarina NCIMB400. J. Biol. Chem. 275, 8515^8522. Beliaev, A.S. and Sa¡arini, D.A. (1998) Shewanella putrefaciens mtrB encodes an outer membrane protein required for Fe(III) and Mn(IV) reduction. J. Bacteriol. 180, 6292^6297. Beliaev, A.S., Sa¡arini, D.A., McLaughlin, J.L. and Hunnicutt, D. (2001) MtrC, an outer membrane decahaem c cytochrome required for metal reduction in Shewanella putrefaciens MR-1. Mol. Microbiol. 39, 722^730. Myers, C.R. and Myers, J.M. (2002) MtrB is required for proper incorporation of the cytochromes OmcA and OmcB into the outer membrane of Shewanella putrefaciens MR-1. Appl. Environ. Microbiol. 68, 5585^5594. Shyu, J.B., Lies, D.P. and Newman, D.K. (2002) Protective role of tolC in e¥ux of the electron shuttle anthraquinone-2,6-disulfonate. J. Bacteriol. 184, 1806^1810. DiChristina, T.J., Moore, C.M. and Haller, C.A. (2002) Dissimilatory Fe(III) and Mn(IV) reduction by Shewanella putrefaciens requires ferE, a homolog of the pulE (gspE) type II protein secretion gene. J. Bacteriol. 184, 142^151. Myers, C.R. and Myers, J.M. (1993) Ferric reductase is associated with the membranes of anaerobically grown Shewanella putrefaciens MR-1. FEMS Microbiol. Lett. 108, 15^22. Gordon, E.H.J., Pike, A.D., Hill, A.E., Cuthbertson, P.M., Chapman, S.K. and Reid, G.A. (2000) Identi¢cation and characterization of a novel cytochrome c(3) from Shewanella frigidimarina that is involved in Fe(III) respiration. Biochem. J. 349, 153^158. Lower, S.K., Hochella Jr., M.F. and Beveridge, T.J. (2001) Bacterial recognition of mineral surfaces : nanoscale interactions between Shewanella and alpha-FeOOH. Science 292, 1360^1363.

Cyaan Magenta Geel Zwart

I. Schro«der et al. / FEMS Microbiology Reviews 27 (2003) 427^447 [169] Seeliger, S., Cord-Ruwisch, R. and Schink, B. (1998) A periplasmic and extracellular c-type cytochrome of Geobacter sulfurreducens acts as a ferric iron reductase and as an electron carrier to other acceptors or to partner bacteria. J. Bacteriol. 180, 3686^3691. [170] Dobbin, P.S., Butt, J.N., Powell, A.K., Reid, G.A. and Richardson, D.J. (1999) Characterization of a £avocytochrome that is induced during the anaerobic respiration of Fe3þ by Shewanella frigidimarina NCIMB400. Biochem. J. 342, 439^448. [171] Gorby, Y.A. and Lovley, D.R. (1991) Electron transport in the dissimilatory iron reducer, GS-15. Appl. Environ. Microbiol. 1991, 867^870. [172] Magnuson, T.S., Hodges-Myerson, A.L. and Lovley, D.R. (2000) Characterization of a membrane-bound NADH-dependent Fe3þ reductase from the dissimilatory Fe3þ -reducing bacterium Geobacter sulfurreducens. FEMS Microbiol. Lett. 185, 205^211. [173] Magnuson, T.S., Isoyama, N., Hodges-Myerson, A.L., Davidson, G., Maroney, M.J., Geesey, G.G. and Lovley, D.R. (2001) Isolation, characterization and gene sequence analysis of a membraneassociated 89 kDa Fe(III) reducing cytochrome c from Geobacter sulfurreducens. Biochem. J. 359, 147^152. [174] Lloyd, J.R., Leang, C., Hodges Myerson, A.L., Coppi, M.V., Cuifo,

FEMSRE 776 4-6-03

[175]

[176]

[177]

[178]

[179]

447

S., Methe, B., Sandler, S.J. and Lovley, D.R. (2003) Biochemical and genetic characterization of PpcA, a periplasmic c-type cytochrome in Geobacter sulfurreducens. Biochem. J. 369, 153^161. Lloyd, J.R., Blunt-Harris, E.L. and Lovley, D.R. (1999) The periplasmic 9.6-kilodalton c-type cytochrome of Geobacter sulfurreducens is not an electron shuttle to Fe(III). J. Bacteriol. 181, 7647^ 7649. Kaufmann, F. and Lovley, D.R. (2001) Isolation and characterization of a soluble NADPH-dependent Fe(III) reductase from Geobacter sulfurreducens. J. Bacteriol. 183, 4468^4476. Nishida, H., Inaka, K., Yamanaka, M., Kaida, S., Kobayashi, K. and Miki, K. (1995) Crystal structure of NADH-cytochrome b5 reductase from pig liver at 2.4 Angstrom resolution. Biochemistry 34, 2763^2767. Ingelman, M., Bianchi, V. and Eklund, H. (1997) The three-dimensional structure of £avodoxin reductase from Escherichia coli at 1.7A resolution. J. Mol. Biol. 268, 147^157. Eriksson, S., Lucchini, S., Thompson, A., Rhen, M. and Hinton, J.C.D. (2003) Unraveling the biology of macrophage infection by gene expression pro¢ling of intracellular Salmonella enterica. Mol. Microbiol. 47, 103^118.

Cyaan Magenta Geel Zwart