Anaerobic Microbial Iron Oxidation and Reduction

Home , Bolt (web browser), Strain 121

University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

Faculty Publications in the Biological Sciences Papers in the Biological Sciences

2006

Microorganisms pumping iron: Anaerobic microbial iron oxidation and reduction

Karrie A. Weber University of Nebraska-Lincoln, [email protected]

Laurie A. Achenbach Southern Illinois University Carbondale, [email protected]

John D. Coates University of California, Berkeley, [email protected]

Follow this and additional works at: https://digitalcommons.unl.edu/bioscifacpub

Part of the Life Sciences Commons

Weber, Karrie A.; Achenbach, Laurie A.; and Coates, John D., "Microorganisms pumping iron: Anaerobic microbial iron oxidation and reduction" (2006). Faculty Publications in the Biological Sciences. 204. https://digitalcommons.unl.edu/bioscifacpub/204

This Article is brought to you for free and open access by the Papers in the Biological Sciences at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Faculty Publications in the Biological Sciences by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. Published in Nature Reviews Microbiology 4 (October 2006), pp. 752-764; doi: 10.1038/nrmicro1490 Copyright © 2006 Nature Publishing Group. Used by permission.

Microorganisms pumping iron: Anaerobic microbial iron oxidation and reduction Karrie A. Weber,1 Laurie A. Achenbach,2 & John D. Coates 1

1. Department of Plant and Microbial Biology, 271 Koshland Hall, University of California, Berkeley, Berkeley, California, 94720 USA. 2. Department of Microbiology, Southern Illinois University, Carbondale, Illinois, 62901 USA. Corresponding author — John D. Coates, email [email protected]

Abstract Iron (Fe) has long been a recognized physiological requirement for life, yet for many microorganisms that persist in water, soils and sediments, its role extends well beyond that of a nutritional necessity. Fe(II) can function as an electron source for iron-oxidizing microorganisms under both oxic and anoxic conditions and Fe(III) can function as a terminal electron acceptor under anoxic conditions for iron-reducing microorganisms. Given that iron is the fourth most abundant element in the Earth’s crust, iron redox reactions have the potential to support substantial microbial populations in soil and sedimentary environments. As such, biological iron apportionment has been described as one of the most ancient forms of microbial metabolism on Earth, and as a conceivable extraterrestrial metabolism on other iron-mineral-rich planets such as Mars. Furthermore, the metabolic versatility of the microorganisms involved in these reactions has resulted in the development of biotechnological applications to remediate contaminated environments and harvest energy.

At pH values at or above a circumneutral pH (~pH both the Archaea and Bacteria domains are capable 7), iron (Fe) exists primarily as insoluble, solid- of metabolically exploiting the favourable redox po- phase minerals in the divalent ferrous (Fe(II)) or tri- tential between the Fe(III)/Fe(II) couple and vari- valent ferric (Fe(III)) oxidation states1. The solubil- ous electron donors or acceptors. In this way, Fe(II) is ity of Fe(III) increases with decreasing pH values2. used as an electron donor to provide reducing equiv- Decreasing pH values also enhance the stability of alents for the assimilation of carbon into biomass by Fe(II), and below pH 4.0, Fe(II) primarily exists as lithotrophic FOM under both oxic and anoxic condi- an aqueous species, even in the presence of oxygen. tions, and Fe(III) can be used as a terminal electron The biogeochemical role of Fe(II)-oxidizing micro- acceptor under anaerobic conditions for lithotrophic organisms (FOM) in acidic environments has been and heterotrophic Fe(III)-reducing microorganisms well established (reviewed in Reference 3). At a cir- (FRM) (Figure 1). This Review discusses the intricate cumneutral pH, microbial iron redox cycling can biogeochemical role of microbially mediated iron ox- significantly affect the geochemistry of hydromor- idation and reduction in suboxic environments, de- phic soils (that is, soils showing poor drainage) and tailing the metabolisms, the microorganisms and the Lithotrophic sediments, leading to the degradation of organic mechanisms described so far. A lithotrophic matter, mineral dissolution and weathering, the for- organism uses mation of geologically significant minerals, and the Iron cycling: the microbial redox workout an inorganic mobilization or immobilization of various anions In the anoxic zone of neoteric environments at pH substrate and cations, including contaminants.1, 4, 5 >4.0, Fe(III) oxides are readily reduced and pro- (usually of The redox transition between the Fe(II) and vide an important electron sink for both chemical mineral origin) Fe(III) valence states has a fundamental role in mod- and biological processes. It is now established that to obtain energy ern environmental biogeochemistry and was prob- microbial reduction of Fe(III) oxide minerals by for growth. ably an important biogeochemical process on early FRM primarily controls the Fe(III)-reductive pro- Heterotrophic Earth. Before microbially mediated iron redox re- cess in non-sulphidogenic sedimentary environ- 6 A heterotrophic actions were discovered, abiotic mechanisms were ments (Figure 1). Even in some sulphidogenic en- organism thought to dominate environmental iron redox vironments (for example, some marine sediments) requires organic chemistry. However, it is now accepted that micro- where Fe(III) reduction is traditionally thought to compounds as a bial metabolism primarily controls iron redox chem- result from an abiotic reaction with biogenic hydrocarbon source. istry in most environments. Microorganisms from gen sulphide (H2S), direct enzymatic Fe(III) reduc- 752 A n a e r o b i c m i c r o b i a l i r o n o x i d at i o n a n d r e d u c t i o n 753 tion by FRM has been shown to be substantial, and the activity of these organisms might account for as much as 90% of the oxidation of organic matter7. Organic carbon compounds are not the only electron donors that FRM are capable of utilizing. These microorganisms are also capable of utilizing inor- Suboxic ganic electron donors such as hydrogen (H2) (reviewed in Reference 6). Ammonium might have a An environment role in Fe(III) reduction as an electron donor8, how- with a partial ever, direct pure-culture evidence is still required pressure of oxygen that is to support this observation. The activity of FRM re- substantially sults in the generation of aqueous Fe(II) (Fe(II)aq), lower than the and solid-phase Fe(II)-bearing minerals (Fe(II)s) in- atmospheric cluding siderite, vivianite and geologically signifi- oxygen content. cant mixed-valence Fe(II)–Fe(III) minerals, such as magnetite and green rust.9, 10 The ubiquity and phy- Anoxic logenetic diversity of FRM, combined with the el- An environment emental abundance of iron in the earth’s crust, es- lacking oxygen. tablishes the global significance of this metabolic process.6 Figure 1. The microbially mediated iron redox cycle. Neoteric environments Similarly, microbially mediated Fe(II) oxidation Microorganisms have a significant role mediating iron oxidation and reduction reactions in soils and sedimentary Modern is also known to contribute to the dynamic iron bio- environments. geochemical cycle at circumneutral pH in both oxic environments. The reduction of Fe(III) oxides occurs in the absence of oxygen. The re-oxidation of the biogenic Fe(II) and anoxic environments (Figure 1). The recent can occur through several biological mechanisms and is Electron sink identification of FOM in various aquatic and- sedi not simply limited to abiotic reactions with molecular oxy- A compound mentary systems — both freshwater and marine — gen. The regeneration of Fe(III) in the anoxic environment that receives indicates that Fe(II) undergoes both biotic and abi- promotes a dynamic iron redox cycle. electrons as otic oxidation in the environment. Abiotic oxidation an endpoint of an oxidative of Fe(II)aq can be mediated by reaction with oxidized manganese (Mn(IV)) species or by the diffusion of reaction. Fe(II)aq into an oxic environment where it subse- Fe(III) oxides produced through the activity of these Bioturbation quently reacts with molecular oxygen (O2). Disrup- organisms potentially function as terminal electron The disturbance tion of sediments by macrophytes and macrofauna acceptors for FRM, thereby perpetuating a dynamic of sediment can induce particle mixing and aeration, result- microbially mediated iron redox cycle (Figure 1). layers by ing in the subsequent oxidation of both Fe(II)aq and biological Fe(II)s (References 11, 12). However, so far, microbi- Microbial Fe(II) oxidation activity. ally mediated oxidative processes in direct associa- The microbial oxidation of Fe(II) coupled to the re- tion with bioturbation have not been studied to any duction of oxygen in environments at acidic and Microaerophilic significant extent, with such studies being limited circumneutral pH values has been recognized for An organism to the rhizosphere.13, 14, 15 Microaerophilic FOM ca- more than a century (References 16, 28 and refer- that is an obligate pable of competing with the abiotic oxidation kinet- ences therein). Yet, despite this historical knowl- anaerobe but ics between oxygen and Fe(II) have been shown to edge, the geological significance of aerobic Fe(II) ox- can survive in contribute to iron cycling in oxic environments, cou- idation at circumneutral pH had been discounted environments 16, 17, 18, 19 pling this metabolism to growth. based on the rapid rate of abiotic Fe(II) oxidation where the Although the aerobic microbial oxidation of coupled to oxygenreduction29. It is now known partial pressure Fe(II) has been recognized for decades, the re- that a broad diversity of microorganisms exist that of oxygen is cent identification of anaerobic Fe(II) bio-oxida- are capable of aerobic, neutrophilic fe(II) oxida- substantially tion closed a missing gap in the iron redox cycle.20, tion and, although only three genera (Gallionella, lower than in the 21 Recent evidence indicates that anaerobic Fe(II) Leptothrix and Marinobacter) have been described atmosphere. oxidation can contribute to a dynamic anaerobic so far, several microaerophilic neutrophilic FOM 22, 23, 24 Phototrophic iron redox cycle , in addition to soil and sed- were recently identified belonging to the α-, β-, and A phototrophic iment biogeochemistry, mineralogy, and heavy- γ-proteobacteria19, 30. 4, 5, 24, 25 organism metal and radionuclide immobilization. In The environments that could potentially sup- obtains energy anoxic environments, microbial Fe(II) oxidation port aerobic, neutrophilic Fe(II) oxidation are for growth from has been demonstrated to be coupled to the reduc- stream sediments, groundwater iron seeps, wetland sunlight; carbon tion of nitrate, perchlorate and chlorate21, 24, 26 (Fig- surface sediments, sediments associated with the is derived from inorganic carbon ure 1). These FOM can oxidize Fe(II)s (References rhizosphere, cave walls, irrigation ditches, subsur- 4, 25), and Fe(II) associated with structural iron face bore holes, municipal and industrial water dis- (carbon dioxide) in minerals, such as the iron aluminum silicate al- tribution systems, deep-ocean basalt and hydrother- or organic 30 carbon. mandine (Fe3Al2(SiO4)3) or staurolite ((Fe,Mg,Zn)2 mal vents . In these environments, microaerophilic Al9γ(Si,Al)4O22(OH)2) (References 4, 27). In zones of FOM seem to successfully compete with the kinetics Neutrophilic sufficient light penetration, Fe(III) can also be pro- of abiotic Fe(II) oxidation. Although the quantita- Fe(II) oxidation duced through the activity of Fe(II)-oxidizing pho- tive significance of this microbial metabolic process Microbial Fe(II) totrophic bacteria that utilize Fe(II) as a source of in terms of accelerating Fe(II) oxidation rates is sub- oxidation electrons to produce reducing equivalents for the ject to interpretation, there is unequivocal evidence that occurs at assimilation of inorganic caβrbon (Figure 1). FOM to demonstrate that FOM can conserve energy circumneutral are ubiquitous and have been identified in numer- from this process and convert inorganic carbon, in pH values (~pH 17 7). ous environments. Anaerobic, biogenically formed the form of carbon dioxide (CO2), into biomass . 754 W e b e r , A c h e n b a c h , & C o at e s i n N at u r e R e v i e w s M icrobiology 4 (2006)

Chlorobium ferrooxidans, Rhodovulum robiginosum, Rhodomicrobium vannielii, Thiodictyon sp., Rhodopseu- domonas palustris and Rhodovulum spp. (Figure 3), so far, an archaeon capable of this metabolism has not been identified. Several purple and green anaerobic, anoxygenic, photosynthetic FOM have been isolated from freshwater and marine environments and described in pure culture20, 32, 33, 34, 35, 36. With the ex- ception of R. vannielii, these phototrophic FOM can completely oxidize Fe(II)aq to Fe(III). Incomplete Fe(II) oxidation by R. vannielii has been attributed to encrustation of the bacterial cell with biogenic Fe(III) oxides, inhibiting further metabolic activ- ity20, 34. The production of low-molecular-weight compounds that can solubilize biogenic Fe(III) has been suggested as a mechanism for preventing cell encrustation in cultures of other phototrophic FOM, including R. robiginosum and C. ferrooxidans33, 37. In contrast to R. vannielii37, the concentrations of soluble Fe(II) and Fe(III) in spent culture media of R. robiginosum and C. ferrooxidans far exceeded the amount predicted by solubility constants and an uninoculated control37. However, a subsequent study failed to find significant evidence to support a direct role for organic compounds or chela- Figure 2. Potential electron donors and acceptors: a redox tower. Theoreti- tors complexing iron when R. robiginosum, C. ferro- cal Eh (volts) values for reduction–oxidation couples that are significant to micro- oxidans and Thiodictyon sp. strain F4 were grown on bially mediated iron redox cycling, calculated at circumneutral pH, are shown. The 38 Fe(II)aq or solid-phase ferrous sulphide (FeS) . As redox tower is an effective way to visualize the potential electron donors and ac- such, it was proposed that Fe(III) was released from ceptors utilizable by microorganisms measuring the tendency of a compound to the active cell as an inorganic aqueous complex or interact as an electron donor or acceptor. An electron donor will have a greater colloidal aggregate38. Nonetheless, it remains to be negative potential than the electron acceptor will. The directional flow of electrons explained why R. vannielii is subject to cell encrus- (e-) is denoted by the orange (negative) to yellow (positive) bolt. Reduction po- tentials were obtained from the following sources: *Reference 149, ‡Reference 101, tation whereas the other phototrophic FOM are not §Reference 150 and ||Reference 151. ¶Calculated based on data collected from Ref- affected. The production of compounds capable of erences 2 & 48. solubilizing Fe(III) has profound implications not only for FOM cellular metabolism but also for the dissolution of solid-phase minerals and the release Aerobic Fe(II) oxidation not only has a role in iron of soluble iron as a terminal electron acceptor or a redox reactions at the oxic–anoxic interface, it also micronutrient for other aquatic and terrestrial liv- influences mineral weathering in the environment19. ing organisms. Before oxygen was available in the Fe(II)-rich Phototrophic Fe(II) oxidation results in the for- Precambrian environment, anaerobic microbial ox- mation of poorly crystalline Fe(III) oxides, which idation of Fe(II) potentially provided an early respi- subsequently transform into the more crystalline ratory metabolism, contributing to the precipitation Fe(III) oxide minerals goethite and lepidocrocite in of iron oxide minerals, including magnetite4, 20, 31, the presence of metabolically active FOM38. How- 32. The reduction potential of the Fe(III)/Fe(II) cou- ever, the significance of phototrophic Fe(II) -oxida ple is sufficient to provide reducing power between tion processes in natural terrestrial environments bacterial photosystems or alternative terminal elec- is limited by the maximum penetration of light tron acceptors involved in respiratory processes to a depth of 200 μm in soil and sediments39. Fur- (Figure 2) to sustain microbial growth. In the past thermore, recent studies indicate that phototrophic decade the true extent of this metabolism has been FOM are unable to promote Fe(II) mineral disso- realized and anaerobic Fe(II) oxidation mediated by lution and are limited by the mineral solubility38. members of both the Archaea and Bacteria domains Therefore, the impact of this microbial process on has been described in anoxic environments at cir- iron redox cycling and mineral weathering might cumneutral pH20, 21, 31. be significant locally but of only minor importance in global iron biogeochemical cycling in terrestrial Seeing the light — anaerobic, photoautotrophic ox- environments. idation of Fe(II). The discovery of photoautotrophic, anaerobic Fe(II) oxidation was the first dem- Working in the dark — anaerobic, nitrate-depen- onstration of microbially mediated oxidation of dent Fe(II) oxidation. Anaerobic Fe(II) oxidation is Fe(II) in anoxic environments20. The FOM involved not limited to environments exposed to light. At cir- in this process oxidize Fe(II), utilizing light energy cumneutral pH, light-independent microbially me- to fix CO2 into biomass as shown below. diated oxidation of both soluble and insoluble Fe(II) – + coupled to nitrate reduction has been demonstrated HCO3 + 4Fe(II) + 10H2O → CH2O + 4Fe(OH)3 + 7H in various freshwater and saline environmental sys- Although the phototrophic FOM in the Bacte- tems, including paddy soil, pond, stream, ditch, ria domain are phylogenetically diverse, including brackish lagoon, lake, wetland, aquifer, hydrother- A n a e r o b i c m i c r o b i a l i r o n o x i d at i o n a n d r e d u c t i o n 755

Psychrophilic An organism that grows Figure 3. Phylogenetic affiliation of microorganisms contributing to iron redox cycling. Unrooted phylogenetic tree optimally based on nearly complete 16S ribosomal DNA sequences from representative iron cycling prokaryotes. The names of iron in a cold reducers are given in red text, iron oxidizers are given in black text. environment (<15°C). 4, 19, 21, 25, 27, 31, 40, 41, 42, mal, and deep-sea sediments on a global scale, provided that adequate concen- Mesophilic 43. These environmental systems support abundant trations of a suitable electron acceptor are readily An organism nitrate-dependent Fe(II)-oxidizing microbial com- available. In environments where zones of nitrate that grows munities in the order of 1×103 to 5×108 cells g-1 sed- reduction and Fe(III) reduction merge, this metab- optimally in iment (Reference 44 and references therein), poten- olism not only influences the iron cycle but can also a moderate tially contributing to iron redox cycling. influence the nitrogen cycle. However, the quantita- environment Various archaeal and bacterial genera (Figure tive significance of this metabolism to global nitro- (~25–45°C). 3), representing a range of optimal thermal growth gen cycling is currently unknown. Hyperthermo- conditions (from psychrophilic through mesophilic So far, FOM have been demonstrated to exploit philic to hyperthermophilic), have been identified. The the favourable thermodynamics between Fe(OH3)/ An organism ubiquity and diversity of these anaerobic FOM sug- Fe(II) and nitrate reduction redox pairs (NO –/½N , – – – 3 2 that grows gests that metabolic, light-independent reactions NO /NO and NO /NH +)21, 24, 26, 31, and be- optimally in hot 3 2 3 4 – – such as nitrate-dependent Fe(II) oxidation have the tween Fe(OH3)/Fe(II) and perchlorate (ClO4 /Cl ) environments – – 26 (>80°C). potential to contribute to anoxic Fe(II) oxidation and chlorate (ClO3 /Cl ) (Figure 2). Several phylo- 756 W e b e r , A c h e n b a c h , & C o at e s i n N at u r e R e v i e w s M icrobiology 4 (2006)

Autotrophic gentically diverse mesophiles have been described strated in two pure-culture isolates; a hyperther- An autotrophic as being capable of nitrate-dependent Fe(II) oxi- mophilic archaeon, Ferroglobus placidus31, and a organism uses dation (Figure 3). However, in most cases growth mesophilic -proteobacterium, strain 2002 (Refer- inorganic carbon was either not associated with this metabolism or ence 45). Even in the case of a known Fe(II)-oxidiz- (carbon dioxide) as a carbon was not demonstrated in the absence of an addi- ing chemolithoautotrophic bacterium, Thiobacillus source. tional electron donor or organic carbon as an en- denitrificans, energy conservation directly coupled ergy source at circumneutral pH4, 21, 26, 27. The oxi- to the nitrate-dependent oxidation of Fe(II) could Chemolithoau- dation of Fe(II), including Fe(II)s, coupled to nitrate not be demonstrated without an additional electron 46 - totrophic reduction is energetically favourable at neutral pH donor . Nitrite (NO2 ) and nitrogen gas (N2) were An organism and should yield enough energy to support car- thought to be the sole end products of nitrate reduc- that obtains bon fixation and microbial growth. However, so tion4, 21, 31 until the recent demonstration of nitrate- energy from far autotrophic growth under nitrate-dependent dependent Fe(II) oxidation by Geobacter metalliredu- inorganic Fe(II)-oxidizing conditions has only been demon- cens resulting in the production of ammonium24. compounds and carbon from carbon dioxide. Box 1. Banded iron formations The deposition of alternating iron-rich and silica-rich mineral layers in the late Fe(II)-rich Archean to early Protero- zoic periods of the Earth’s history resulted in the genesis of the Precambrian banded iron formations (BIFs)125 (see figure). Early models suggested that BIFs were a consequence of abiotic reactions involving Fe(II) photo-oxidation51, 126 and/or Fe(II) oxidation by the metabolic end product of oxygenic photosynthesis — oxygen (Reference 127). This led to the formation of the iron-rich laminae, containing hematite (Fe2O3) and the mixed-valence phase Fe(II)– Fe(III)-bearing mineral magnetite (Fe3O4). In recent years, microbial reductive and oxidative respiratory metabolisms have been implicated in the deposition of iron-bearing minerals in the iron-rich laminae of BIFs4, 20, 128, 129, 130. Microaerophilic Fe(II) oxidizers, such as Gallionella ferruginea, might have metabolically oxidized Fe(II) coupled to the oxygen generated by oxygenic photosynthesis129, 131. However, this model assumes that oxygenic photosynthesis evolved enough oxygen to account for the precipitation of Fe(III)-rich minerals. There has been much debate concerning the evolution of oxygen in the early Proterozoic and therefore the role of oxygen in the deposition of BIFs132, 133. Other microbially mediated mechanisms influencing the precipitation of Fe(II)- and Fe(III)-rich minerals in anoxic environments have been proposed as plausible alternatives, including Fe(III) reduction, phototrophic Fe(II) oxidation, and nitrate-dependent Fe(II) oxidation (see figure). The precipitation of magnetite associated with microbial Fe(III) reduction linked the deposition of BIFs to FRM9, 53. Yet the generation of Fe(III) as a terminal electron acceptor from available Fe(II) would have to precede metabolic Fe(III) reduction. In addition to photo-oxidation, the anoxic deposition of Fe(III) oxides by photoautotrophic20, 130 and nitrate-dependent Fe(II)-oxidizing microorganisms4 (FOM) during the late Archean to the early Proterozoic has been proposed (see figure). However, the bio-oxidation of Fe(II) not only results in the precipitation of Fe(III) oxides but also results in the direct precipitation of magnetite4, 32, 36 and hematite4, potentially linking FOM directly to the formation of BIFs. The availability of light in the Precambrian is inarguable, as such phototrophic Fe(II) oxidation has received much attention as a model for BIF deposition130. By contrast, light-independent Fe(II) oxidation coupled to nitrate reduction has received much less attention as a model; however, it does provide an additional mechanism leading to BIF formation in the Precambrian and, in contrast to known phototrophic FOM, extant nitrate-dependent FOM have been shown to produce significant quantities of extracellular magnetite and hematite4, 32, 36. Lightning discharge contrib- 12 uted to the fixation of nitrogen (N2) into nitric oxide (NO (~10 g per yr) further forming nitrous oxide (N2O), the ni- - - 134, 135 trite ion (NO2 ) and the nitrate ion (NO3 ) through abiotic disproportion reactions . Such oxidized nitrogen species could have functioned as an electron acceptor for early FOM, leading to BIFs in anoxic environments. In modern environmental systems, complex microbial communities contribute to a dynamic anoxic iron redox cycle. Similarly complex communities consisting of phototrophic FOM, nitrate-dependent FOM and Fe(II)-reducing microorganisms (FRM) might have existed in the Precambrian era, each of which contributed to the formation of the BIFs rather these being the result of one individual metabolism. The precipitation of biomass with biogenic Fe(III) oxides would contribute to diagensis and mineralization of organic matter, resulting in heterotrophic reduction of Fe(III) by FRM and the formation of additional magnetite in BIFs53, 130, which is further supported by the iso- topically light carbon associated with BIFs136. The figure shows the models proposed for the microbial mediation of anoxic deposition of BIF. The deposition of Fe-rich laminae, hematite and magnetite, could be a result of anoxic microbial metabolisms. The direct oxidation of Fe(II) by photoautotrophic bacteria (a) and nitrate-dependent Fe(II) oxidizing bacteria (b) results in the formation of magnetite and hematite, and solid-phase Fe(III) oxides. The biogenic Fe(III) oxides are subsequently reduced by FRM (c) forming magnetite. A n a e r o b i c m i c r o b i a l i r o n o x i d at i o n a n d r e d u c t i o n 757

Growth of FOM that are capable of autotrophy FOM D. agitata and A. suillum demonstrated the in- or mixotrophy associated with perchlorate/chlo- volvement of c-type cytochrome(s) when grown un- rate-dependent Fe(II) oxidation has not yet been der Fe(II)-oxidizing conditions with nitrate or chlo- identified. However, perchlorate/chlorate-depen- rate, respectively4, 26. dent Fe(II)-oxidizing metabolic activity is observed The demonstrated capability of some FOM to in stationary-phase cultures of Dechloromonas agi- use CO2 as the sole carbon source requires a CO2- tata and Azospira suillum strain PS (previously called fixation pathway. In the case of the archaeon Fer- Dechlorosoma suillum) containing acetate, perchlo- roglobus placidus grown on CO2, the reductive ace- rate and Fe(II) (References 4, 47). Although per- tyl coenzyme A pathway is expressed, implicating chlorate and chlorate are not naturally abundant its involvement in carbon assimilation49. Interest- compounds, their potential to serve as electron ac- ingly, genes associated with the reductive pen- ceptors in environmental systems cannot be dis- tose phosphate cycle, RuBisCo, were identified in counted48. The legal discharge of perchlorate into the finished genome sequence of Dechloromonas ar- natural waters has led to widespread anthropogenic omatica, an Fe(II)-oxidizing bacterium capable of contamination throughout the United States. Given utilizing nitrate, chlorate or perchlorate as alter- the ubiquity of perchlorate-reducing bacteria48 and native electron acceptors45; however, autotrophic the ability of some of these microorganisms, specif- growth associated with Fe(II) oxidation could not ically A. suillum and D. agitata, to oxidize Fe(II), an- be demonstrated and the conditions under which aerobic perchlorate/chlorate-dependent Fe(II) ox- these genes are expressed remain unidentified idation could affect iron biogeochemical cycling (J.D.C., unpublished data). By contrast, PCR am- in environments exposed to contaminated waters. plification using degenerative RuBisCo primers Further studies are needed to quantify the potential of the genomic DNA from strain 2002, the meso- influence of perchlorate/chlorate-dependent Fe(II) philic autotrophic nitrate-dependent FOM, did not 45 oxidation. yield a PCR product . The CO2-fixation pathway expressed in strain 2002 during growth under ni- Doing hard time: solid-phase microbial Fe(II) ox- trate-dependent Fe(II)-oxidizing conditions is cur- idation. In contrast to the reaction carried out rently unknown. Nonetheless, the availability of by phototrophic FOM, Fe(II)s including surface- genomic sequence information for the FOM D. aro- bound Fe(II) (References 24, 25), crystalline Fe(II) matica, Marinobacter aquaeolei, G. metallireducens, R. minerals (siderite, magnetite, pyrite, arsenopyrite palustris, and T. denitrificans provides the first op- and chromite)4, 25, and structural Fe(II) in neosili- portunity to comprehensively investigate the func- cate (almandine and staurolite)4 and phyllosilicate tional genes and regulatory pathways associated (nontronite)27 are known to be subject to direct ni- with this metabolism. However, the genomes of trate-dependent microbial oxidation. Although we other phylogenetically and physiologically distinct know that nitrate-dependent FOM have a role in microorganisms such as F. placidus, strain 2002 and the oxidation of Fe(II) that is structurally incor- A. suillum, which also potentially have a key role porated into silicate minerals, and contribute to in anaerobic nitrate-dependent Fe(II) oxidation, Fe(II) mineral dissolution, little is known about have yet to be sequenced. the mineral structure and stability of the resid- ual material. The oxidative dissolution of Fe(II) s in anoxic environments presents an additional The other half — microbial iron reduction mechanism for rock weathering and the precipi- Antiquity of iron reduction. It has been proposed tation of Fe(III) oxide minerals in anoxic soils and that life emerged on a hot (possibly as high as 140– sediments. So far, bio-oxidation products of Fe(II) 150°C), Fe(II)-rich early Earth 3.8 billion years ago50. aq and amorphous Fe(II)s have been characterized. The abiotic photochemical generation of Fe(III) Various biogenic Fe(III) oxide minerals, including and H would have provided an electron accep- 2-line ferrihydrite40, 47, goethite24, lepidocrocite (J. 2 4 tor and energy source, respectively, for ancient life, D.C. and K.W., unpublished data) and hematite , as shown below (References 6, 51 and references and mixed-phase Fe(II)–Fe(III) minerals, magne- therein). tite, maghemite and green rust4 (J.D.C. and K.W., hv unpublished data), were identified as oxidation 2 Fe(II) + 2 H+ → 2 Fe(III) + H products. As a result of this biogenic formation of 2 magnetite and hematite, nitrate-dependent Fe(II) As such, iron respiration has been proposed as oxidation has been implicated as having a direct one of the first forms of microbial metabolism to role in the genesis of banded iron formations in 4, 31 have evolved, preceding the development of oxy- Precambrian Earth (Box 1). gen, nitrate and sulphate respiration52, 53. In support of this, Fe(III) respiration has been identified Function over form: physiology of anaerobic Fe(II) in a diversity of extant microorganisms includ- oxidation at circumneutral pH. Little is known ing those most closely related to the last common about the biochemistry or genetic regulation of an- ancestor52, 54. Extracellular electron transfer to in- Mixotrophy aerobic Fe(II) oxidation at circumneutral pH and as soluble Fe(III) oxide minerals has been conserved A mixotrophic such this can only be discussed in a general sense. in the hyperthermophilic Archaea52, 54, 55, 56 and is organism uses The reduction potential of the possible Fe(III)/Fe(II) 6, 53 an inorganic widely distributed among the Bacteria (Figure chemical redox pairs ranges from –0.314V to +0.014V, indi- 3), further suggesting that this is an early metab- cating that electrons can readily be donated to the energy source olism that has spread through the microbial do- and organic more electropositive type b, c, or a cytochrome com- mains throughout evolution. compounds as a ponents of an electron transport chain (Figure 2). In carbon source. support of this, previous studies using the known 758 W e b e r , A c h e n b a c h , & C o at e s i n N at u r e R e v i e w s M icrobiology 4 (2006)

Figure 4. Microbial strategies mediating electron transfer to insoluble Fe(III) oxides. Three primary strategies have been proposed to facilitate the electron transfer between microorganisms and solid Fe(III) oxide surfaces. a) In Geobacter spp. direct contact with the oxide surface is required. The production of ‘nanowires’, conductive extracellular appendages, facilitates electron transfer by functioning as an electrical conduit to the Fe(III) oxide surface. b) An endogenously or exog- enously produced electron shuttle mediates electron transfer to solid-phase Fe(III) oxides. c) The production of complexing ligands as in the case of Geothrix sp. aids in the dissolution of the solid-phase Fe(III) oxide providing a soluble Fe(III) form more readily available to the microorganism. Although these strategies have only been demonstrated for Fe(III)-reducing microorganisms, similar strategies might be used by Fe(II)-oxidizing microorganisms that are utilizing solid-phase Fe(II) electron donors. e– = electrons; L = ligand

A hard rock chord: crystalline Fe(III) oxide reduc- ganisms utilizing Fe(III) oxides as a respiratory ter- tion. Poorly crystalline Fe(III) oxide minerals, such minal electron acceptor. Various mechanisms have as ferrihydrite, readily function as electron accep- been proposed as possible strategies that microor- tors for FRM6. This is not surprising, given the ganisms might use to transfer electrons to extracel- relative energetic favourability of the reduction lular Fe(III) oxide minerals. Direct contact between of amorphous ferric iron oxides (Figure 2). How- the microorganism and the solid-phase Fe(III) oxide ever, iron oxides predominantly exist in a crystal- mineral was shown to be a requirement for the re- line phase or as a structural component of clays in duction of insoluble Fe(III) oxides in Geobacter spp.61 modern soils and sediments. Although the ther- (Figure 4A). The molecular scale interaction(s) oc- modynamic favourability of crystalline Fe(III) ox- curring between the cell surface and Fe(III) oxide is ide mineral reduction (goethite, hematite and mag- currently unknown. netite) is decreased compared with amorphous The formation of flagella and pili had been pro- Fe(III) oxides (Figure 2), previous studies have posed as the mechanism by which Geobacter spp. di- suggested that FRM are capable of living ‘on the rectly attached to the Fe(III) oxide surface62. Recent energetic edge,’ utilizing structural57 or crystal- evidence, however, indicates that pili are not re- line solid-phase Fe(III) as an electron acceptor58, 59. quired for attachment of Geobacter spp. to the solid- There has been considerable debate regarding the phase Fe(III) oxide surface. Instead, the pili function direct environmental relevance of the laboratory as an electrical conduit for the transfer of electrons studies that have examined the reduction of crys- to insoluble Fe(III) oxides and, potentially, other talline iron minerals under artificial organic and solid-phase terminal electron acceptors63. The for- nutrient-rich conditions to optimize FRM growth6, mation of these conductive cellular ‘nanowires’ ex- 60. Glasauer and colleagues did not observe reduc- pands the accessible spatial area available beyond tion of goethite and hematite in defined minimal the cell membrane, allowing the penetration of media by Shewanella putrefaciens strain CN32, in nanometer pore spaces in soils and sediments pre- direct contrast to previous reduction studies con- viously thought to be physically unavailable to the ducted with this organism in nutrient-rich media10, cell. The potential for these ‘nanowires’ to create a 59. However the inability of FRM to reduce crys- bridge between individual cells introduces the pos- talline Fe(III) oxide minerals under nutrient-lim- sibility of cell-to-cell communication63, and for the ited conditions is not a universal observation. Un- cell attached to the Fe(III) oxide or other electron ac- der oligotrophic culture conditions, both a culture ceptor to function as an electron shuttle. of G. metallireducens and a freshwater enrichment Conductive ‘nanowires’ were thought to be ex- culture (1% vol:vol inoculum) reduced goethite be- clusive to Geobacter spp. as pili produced by other yond the amorphous Fe(III) oxide content (deter- metal-reducing organisms tested, including She- mined by 0.5 N HCl extraction)24, indicating the wanella oneidensis, were not conductive63. However, possibility for microbial reduction of crystalline recent evidence indicates that S. oneidensis, among Fe(III) oxides in the environment. other microorganisms, also produce conductive appendages under conditions in which the electron ac- Between a rock and a hard place: microbial strate- ceptor is limited64. The direct involvement of these gies for reduction of an insoluble electron acceptor. appendages in electron transport or reduction of The insoluble nature of Fe(III) oxide minerals at val- Fe(III) remains unproven and their functional role is ues >pH 4 creates a metabolic dilemma for microor- currently unknown. A n a e r o b i c m i c r o b i a l i r o n o x i d at i o n a n d r e d u c t i o n 759

FRM might not necessarily need to establish di- in dissimilatory Fe(III) reduction and the oxida- rect contact with the solid-phase surface to reduce tion of organic matter in soils and sediments. Out- Fe(III) oxides, as other alternative active mecha- side the δ-proteobacteria, the genus Shewanella in nisms could be used. Exogenous65 and endoge- the γ-proteobacteria is another well-character- nously produced66, 67, 68 soluble external electron ized group of FRM. Although Shewanella spp. have shuttles can be exploited as mediators to complete been isolated from diverse metal-reducing sedi- the transfer of electrons to the solid-phase terminal ments80, 81, 82, several studies focused on recover- electron acceptor (Figure 4B). The electron shuttle ing 16S ribosomal DNA (rDNA) gene sequences alleviates the need for the FRM to directly contact representing these FRM from natural Fe(III)-reduc- the Fe(III) oxide and functions in a combination ing environments have not yielded a positive re- of a microbially catalysed and abiotic process; the sult83, 84, 85. As a result, these data indicated that FRM oxidizes an electron donor coupled to the re- these microorganisms might only have a minor duction of the soluble electron shuttle and the re- role in the reduction of Fe(III) oxides in situ. How- duced electron shuttle diffuses and subsequently ever, this result might not be universal; the recent donates electrons to the solid-phase Fe(III) oxide identification of 16S rDNA gene sequences closely abiotically65 (Figure 4B). The abiotic regeneration related to Shewanella spp. in a minerotrophic wet- of the oxidized electron acceptor restarts the cycle land indicates the potential for this group of FRM (Figure 4B). to contribute to iron cycling in situ86. Redox-reactive organic compounds common in Recent evidence indicates that other organisms soils and sediments, such as humic acids65, plant ex- belonging to the β-proteobacteria might also have udates69 and antibiotics70, have been identified as a role in the reduction of Fe(III) in sediments, al- electron shuttles. Although the significance of these though the extent of this role has yet to be quanti- compounds as electron shuttles for microbially me- fied24, 85, 87, 88 (J.D.C., unpublished data). Addition- diated Fe(III) reduction in eutrophic environments ally, enzymatic Fe(III) reduction is not limited to is still unknown, their utility in oligotrophic envi- the Proteobacteria. The potential of the poorly de- ronments might be limited owing to the low avail- scribed Acidobacteria to contribute to Fe(III) reduc- ability of suitable redox-reactive refractory organic tion was demonstrated with the isolation of Geothrix substances. However, the endogenous produc- fermentans89 and supported by the identification of tion of an electron shuttle in two FRM genera, She- 16S rDNA gene sequences most similar to this mi- wanella66, 68, 71, 72 and Geothrix67, and the production croorganism in an analysis of subsurface sedi- of a chelating ligand by Geothrix sp. (Fig 4c), could ments.90 So far, the exact metabolic role of members further mitigate the reliance of these FRM on exoge- of this phylum in the environment remains elusive nous electron shuttles. because of the limited availability of pure cultures. Although the energetic expense of producing However, given that Acidobacteria are ubiquitous and excreting redox-reactive compounds into the in the environment,91, 92 the contribution of these Eutrophic environment would not yield a competitive advan- microorganisms to iron biogeochemical cycling in Eutrophic tage in situations of low cell mass, it has been spec- soils and sediments could be globally significant. waters are rich ulated that the release of electron shuttles in a bio- In addition to the FRM described above, which in minerals film community would facilitate electron transfer rely on energy conservation for growth from the and organic by cells distant from the substrate surface.72 In ad- dissimilatory reduction of Fe(III), various fermen- nutrients. dition, the recently identified ubiquitous biologi- tative microorganisms have also been recognized Oligotrophic cal re-oxidation of these diffusing electron shuttles, to catalyse the enzymatic reduction of Fe(III) (Ref- environment coupled to carbon assimilation in the presence of a erences 93 & 94 and references therein). These fer- An environment suitable electron acceptor such as nitrate, offers the mentative microorganisms transfer only a minor that is relatively potential for a previously unidentified symbiotic re- fraction, approximately 5%, of the available reduc- low in nutrients lationship at the microbial level.73, 74 ing equivalents to Fe(III) (Reference 94), and Fe(III) and cannot is not required for growth. The diversion of reduc- support much The iron maidens: diversity of FRM. A wide phy- ing equivalents to Fe(III) might provide an ener- plant life. logenetic diversity of microorganisms capable of getic advantage, utilizing the oxidation of NAD(P) conserving energy coupled to growth by the dis- H coupled to Fe(III) reduction to yield ATP.95 As Thermophilic similatory reduction of Fe(III) have been identified such, Fe(III) reduction coupled to fermentative me- An organism that grows throughout the Archaea and Bacteria (Figure 3) and tabolism is considered to have a minor role in iron optimally at across a range of chemical and physical conditions, geochemical cycling relative to respiratory Fe(III) temperatures demonstrating the ubiquity of this type of microbial reduction. ranging from metabolism. Among the isolated microorganisms, In addition to the fermentative microorganisms 45–80°C. Fe(III)-reducing extremophiles including hyperther- that have been identified that reduce Fe(III), some mophilic, thermophilic, psychrophilic, acidophilic sulphate-reducing96, 97 and methanogenic98 mi- Acidophilic and alkaliphilic Archaea and Bacteria have been de- croorganisms are also capable of Fe(III) reduction An organism scribed in pure culture54, 75, 76, 77, 78, 79 (see Lovley et without the demonstrated benefit of growth. Com- that grows al.6 for a review). One such isolate surviving in hy- petition for available electron donors between mi- in an acid drothermal vents has pushed the upper tempera- crobial communities supporting these metabolisms environment (

Figure 5. Physiological model of the biochemistry involved in microbial Fe(III) reduction by Shewanella and Geo- bacter spp. Full details of the model are given in the text. CM = cytoplasmic membrane; CymA = cytoplasmic-membrane-bound tetraheme c-type cytochrome; Cytc3 = periplasmic c-type cytochrome; e– = electrons; MacA = cytoplasmic-membrane-bound cytochrome; MQ = menaquinone; MtrA = periplasmic decaheme c-type cytochrome; OM = outer membrane; Omc = outer-membrane-bound cytochromes B, E, and S partially exposed on the cell surface; PpcA = a tri- heme periplasmic c-type cytochrome; Q = quinone.

cellular level. The physiological advantage created ilarity between the electron transport mechanisms by coupling hydrogen oxidization to Fe(III) reduc- involves the transfer of electrons from the dehydro- tion is unknown. In fact, this metabolism would genase to a quinone pool consisting of ubiquinones scavenge reducing equivalents, impairing the abil- and menaquinones in the cytoplasmic membrane ity of these microorganisms to grow with the trans- to c-type cytochromes and finally to a reductase in fer of electrons to sulphate or carbon dioxide. The the outer membrane66, 103, 104, 105, 106. Whereas a ter- true ecological implications of this alternative life- minal iron reductase has so far eluded identifica- style still remain a mystery. tion, significant developments have advanced our knowledge of Fe(III) reduction biochemistry in both Go with the flow: electron transport to Fe(III) ox- model FRM. ides. Iron(III) oxide minerals are insoluble and so The development of electron transport models is are unable to diffuse inside microbial cells. There- still in the formative stages as the number of c-type fore electron transport to the solid-phase termi- cytochromes directly implicated in Fe(III) reduction nal electron acceptor cannot occur in the periplasm is far less than the number of c-type cytochromes as it does for soluble electron acceptors such as ni- that have been identified in completed genome se- trate101 or fumarate in the case of Shewanella frigidi- quences — the S. oneidensis and G. sulfurreducens marina.102 Instead, it would require the reduction of genomes revealed 39 and 111 putative c-type cyto- Fe(III) to occur outside the cell with a protein local- chromes, respectively.107, 108 The high number of c- ized in the outer membrane, presumably a termi- type cytochromes in G. sulfurreducens has led to the nal iron reductase. Just as the strategies to utilize suggestion of the existence of multiple electron- insoluble Fe(III) oxides as a terminal electron accep- transport pathways.109 The models depicted in Fig- tor differ between the two model FRM, Shewanella ure 5 illustrate the components known to be directly spp. and Geobacter spp., the proteins involved in involved in electron transport to Fe(III) in both She- electron transport also differ. The only general sim- wanella spp. and Geobacter spp. so far. A n a e r o b i c m i c r o b i a l i r o n o x i d at i o n a n d r e d u c t i o n 761

In Shewanella spp. the electrons are transferred Box 2. Biotechnological applications from the menaquinone to a tetraheme c-type cytochrome, CymA, located in the cytoplasmic mem- Heavy-metal and radionuclide immobilization brane.104, 110, 111 The electrons are then passed to The end products of anaerobic microbial oxidation of Fe(II) can also affect the electron carriers in the periplasm. Two c-type cy- geochemistry of contaminants through the formation of various environmentally tochromes in the periplasm have been identi- relevant Fe(III)-bearing minerals that can regulate contaminant solubility in natural environments,5, 22, 43 such as ferric oxyhydroxide, goethite, hematite, iron hy- fied so far as potential electron carriers necessary 4, 24, 40, 47 for Fe(III) reduction in Shewanella spp., MtrA and drogen carbonate or maghemite. The precipitation of these biogenic Fe(III) oxides provides a mechanism for the immobilization of heavy metals and Cytc3 (References 112–114). Cytc3, a small tetra- 115, 116 metalloids through co-precipitation or physical envelopment, and provides a re- heme c-type cytochrome in the periplasm 3- active surface with an adsorptive affinity for anions (that is, PO4 ) and cations might function as an electron shuttle between elec- (that is, Zn(II), As(V), Co(II) and U(vi)) (References 5, 22). Heavy metals and radio- tron carriers117. MtrA is a decaheme c-type cyto- nuclides including U(VI) are rapidly removed from solution during anaerobic ni- chrome that is similarly thought to accept elec- trate-dependent microbial Fe(II) oxidation in association with the biogenic Fe(III) trons from the cytoplasmic membrane electron oxides.5 As such, the anaerobic formation of biogenic Fe(III)-oxide-containing carrier CymA, and transfer these electrons to an minerals has been identified as a plausible bioremediative strategy for perma- 5, 137 outer membrane protein (OMP)113. It has also been nently immobilizing heavy metals and radionuclides. proposed that MtrA alternatively functions as a Energy generation and contaminant immobilization or degradation terminal reductase for soluble Fe(III) in the peri- An electrode is another solid-phase electron acceptor used by Fe(II)-reducing plasm114 (Figure 5). An outer membrane cyto- microorganisms (FRM) that we can exploit to our energetic advantage. The har- chrome partially exposed on the cell surface113, 118, vest of electrical energy mediated by FRM in sediments and microbial fuel cells is OmcB (formerly denoted as MtrC113), accepts elec- a technological reality.4, 138, 139, 140, 141, 142 The utilization of electrodes as an electrons from periplasmic proteins and has the poten- tron acceptor has not only been proposed to harvest energy from sediments tial to reduce extracellular Fe(III) directly118 (Fig- but also to remediate environments contaminated with biodegradable organic compounds, such as mono- and polycyclic-aromatic hydrocarbons141, poten- ure 5). This is supported by studies involving a tially harvesting energy in the process. FRM are not only capable of generat- mutation in omcB that resulted in a decreased abil- ing electricity, previous studies have indicated an incredible metabolic versatility to reduce Fe(III). However, some Fe(III) reduc- ity and these organisms have been demonstrated to transform various organic 119 tion did still occur , indicating that Fe(III) reduc- (that is, benzene, toluene, phenol and chlorinated compounds)90, 143, 144, 145, 146 tion is not absolutely dependent on a functioning and heavy metal or radionuclide contaminants (that is, uranium, technetium and omcB (Figure 5). chromium).55, 147, 148 Given the limited information available so far, the relationship between the periplasmic cy- Concluding remarks tochromes and OMPs has not been firmly estab- Over the past two decades, recognition of the mi- lished in Geobacter spp.120 A key periplasmic cyto- crobial reduction and oxidation of the fourth most chrome, MacA, was identified in the cytoplasmic abundant element in the Earth’s crust has identified membrane and shown to have a central role in the a globally significant biogeochemical process. Al- transfer of electrons to Fe(III). MacA is predicted though it is recognized that the microorganisms in- to function as an intermediate carrier similarly to volved in these competing metabolisms are ubiqui- MtrA in Shewanella spp.120 and therefore might tous, the physiology of microbial Fe(III) reduction pass electrons to other periplasmic proteins such and Fe(II) oxidation remains an enigma, as a ter- as PpcA, a tri-heme periplasmic c-type cytochrome minal Fe(III) reductase has yet to be identified and involved in electron transport to OMPs121 (Figure nothing apart from the implicated role of c-type 5). One such OMP, OmcB, was determined to have cytochrome(s) is known of Fe(II) oxidation at cir- a significant role in Fe(III) reduction109, 122. Disrup- cumneutral pH. Genome sequencing and the subse- tions in omcB by gene replacement impaired the quent development of in silico physiological mod- ability of G. sulfurreducens to reduce Fe(III) by ap- els can be used to predict microbial metabolism in proximately 94–97% (Reference 122). However, the particular environmental conditions123, 124, and can omcB-deficient mutant adapted to growth on sol- potentially provide greater insight into microbially uble Fe(III) over time with similar reduction rates mediated iron oxidation and reduction reactions in to the wild type, although growth was only ap- environmental systems. These advances can also be proximately 60% of the wild type109. Interestingly, applied to enhance and predict the behaviour of mi- the adapted mutant was unable to reduce insolu- croorganisms exploited for their metabolism associ- ble Fe(III) oxides, indicating that different electron ated with biotechnological applications (Box 2). Al- transport mechanisms are used to reduce insoluble though significant advances have been made, we and soluble Fe(III) sources109. The recent identifi- are continuously learning of previously unknown cation of conductive ‘nanowires’ implicated the in- microbially mediated iron redox reactions in di- volvement of other OMPs in electron transport to verse environments and searching for microorgan- Fe(III) oxides. Reguera et al.63 have proposed that isms responsible for catalysing these unique me- pili directly accept electrons from intermediary tabolisms. A topical example of this is the recently electron-transfer proteins located in the periplasm identified anaerobic oxidation of ammonium- cou and/or outer membrane for transfer to the solid- pled to dissimilatory Fe(III) reduction8 for which phase Fe(III) oxide surface (Figure 5). In contrast to no organism has yet been identified. As these sorts the appendages identified in S. oneidensis64, redox- of discoveries are made, the true ubiquity and di- reactive structures have not been identified on the versity of the organisms involved in iron biogeo- conductive pili observed in Geobacter spp.63 The chemistry will be uncovered. Because of the geolog- conductive nature of these pilin structures remains ical and geochemical importance of iron on a global a mystery and introduces an exciting new twist to scale, the activity of these organisms shapes the cell biology. world we live in today. 762 W e b e r , A c h e n b a c h , & C o at e s i n N at u r e R e v i e w s M icrobiology 4 (2006)

References 19. Edwards, K. J., Rogers, D. R., Wirsen, C. O. & Mc- lum iodosum sp. nov, and Rhodovulum robigi- Collom, T. M. Isolation and characterization of nosum sp. nov., two new marine phototrophic 1. Cornell, R. M. & Schwertmann, U. The Iron Ox- novel psychrophilic, neutrophilic, Fe-oxidizing, ferrous-iron-oxidizing purple bacteria. InT. J. ides: Structure, Properties, Reactions, Occur- chemolithoautotrophic α- and γ-Proteobacteria SysT. Bacteriol. 49, 729–735 (1999). rences and Uses (Wiley-VCH, Weinheim, 2003). from the deep sea. Appl. Environ. Microbiol. 69, 37. Straub, K. L., Benz, M. & Schink, B. Iron metabo- 2. Stumm, W. & Morgan, J. J. Aquatic Chemistry: 2906–2913 (2003). lism in anoxic environments at near neutral pH. Chemical Equilibria and Rates in Natural Waters 20. Widdel, F. et al. Ferrous iron oxidation by anox- FEMS Microbiol. Ecol. 34, 181–186 (2001). (John Wiley & Sons, New York, 1996). ygenic phototrophic bacteria. Nature 362, 834– 38. Kappler, A. & Newman, D. K. Formation of 3. Baker, B. J. & Banfield, J. F. Microbial commu- 836 (1993). Fe(III)-minerals by Fe(II)-oxidizing photoauto- nities in acid mine drainage. FEMS Microbiol. This paper gives the first demonstration of trophic bacteria. Geochim. Cosmochim. Acta Ecol. 44, 139–152 (2003). anaerobic, phototrophic Fe(II) oxidation and 68, 1217–1226 (2004). 4. Chaudhuri, S. K., Lack, J. G. & Coates, J. D. Bio- suggests the importance of this metabolism 39. Ciania, A., Gossa, K.-U. & Schwarzenbach, R. P. genic magnetite formation through anaerobic in the Earth’s early history. Light penetration in soil and particulate miner- biooxidation of Fe(II). Appl. Environ. Microbiol. 21. Straub, K. L., Benz, M., Schink, B. & Widdel, F. als. Eur. J. Soil Sci. 53, 561–574 (2005). 67, 2844–2848 (2001). Anaerobic, nitrate-dependent microbial oxida- 40. Straub, K. L., Hanzlik, M. & Buchholz-Cleven, This paper gives the first demonstration of tion of ferrous iron. Appl. Environ. Microbiol. B. E. E. The use of biologically produced ferri- magnetite formation through anaerobic 62, 1458–1460 (1996). hydrite for the isolation of novel iron-reduc- iron bio-oxidation and the first demonstra- This paper gives the first clear demonstra- ing bacteria. SysT. Appl. Microbiol. 21, 442–449 tion of the bioavailability of Fe(II) in silica- tion of anaerobic, mesophilic nitrate-depen- (1998). ceous minerals. dent Fe(II) oxidation. 41. Kluber, H. D. & Conrad, R. Effects of nitrate, ni-

5. Lack, J. G. et al. Immobilization of radionuclides 22. Senn, D. B. & Hemond, H. F. Nitrate controls on trite, NO and N2O on methanogenesis and and heavy metals through anaerobic bio-oxida- iron and arsenic in an urban lake. Science 296, other redox processes in anoxic rice field soil. tion of Fe(II). Appl. Environ. Microbiol. 68, 2704– 2373–2376 (2002). FEMS Microbiol. Ecol. 25, 301–318 (1998). 2710 (2002). 23. Straub, K. L., Schonhuber, W., Buchholz-Cleven, 42. Ratering, S. & Schnell, S. Nitrate-dependent 6. Lovley, D. R., Holmes, D. E. & Nevin, K. P. Dis- B. & Schink, B. Diversity of ferrous iron-oxidiz- iron(II) oxidation in paddy soil. Environ. Micro- similatory Fe(III) and Mn(IV) reduction. Adv. Mi- ing, nitrate-reducing bacteria and their involve- biol. 3, 100–109 (2001). crob. Physiol. 49, 219–286 (2004). ment in oxygen-independent iron cyclinG. Geo- 43. Finneran, K. T., Housewright, M. E. & Lovley, D. 7. Canfield, D. E.et al. Pathways of organic car- microbiol. J. 21, 371–378 (2004). R. Multiple influences of nitrate on uranium sol- bon oxidation in three continental margin sedi- 24. Weber, K. A., Urrutia, M. M., Churchill, P. F., Kuk- ubility during bioremediation of uranium-con- ments. Mar. Geol. 113, 27–40 (1993). kadapu, R. K. & Roden, E. E. Anaerobic redox taminated subsurface sediments. Environ. Mi- This paper gives the first demonstration of cycling of iron by freshwater sediment microor- crobiol. 4, 510–516 (2002). the importance of microbial Fe(III) reduction ganisms. Environ. Microbiol. 8, 100–113 (2006). 44. Weber, K. A. & Coates, J. D. in Manual of Envi- for the oxidation of organic matter in ma- 25. Weber, K. A., Picardal, F. W. & Roden, E. E. Mi- ronmental Microbiology, 3rd edn (eds Hurst, C. rine sediments. crobially catalyzed nitrate-dependent oxidation J., Crawford, R. L., Knudsen, G. R., McInerney, M. 8. Clement, J., Shrestha, J., Ehrenfeld, J. & Jaffe, P. of biogenic solid-phase Fe(II) compounds. Envi- J. & Stetzenbach, L. D.) in the press (ASM Press). Ammonium oxidation coupled to dissimilatory ron. Sci. Technol. 35, 1644–1650 (2001). 45. Weber, K. A. et al. Anaerobic nitrate-dependent reduction of iron under anaerobic conditions in 26. Bruce, R. A., Achenbach, L. A. & Coates, J. D. Re- iron(II) bio-oxidation by a novel, lithoautotro- wetland soils. Soil Biol. Biochem. 37, 2323–2328 duction of (per)chlorate by a novel organism phic, β-proteobacterium, strain 2002. Appl. En- (2005). isolated from paper mill waste. Environ. Micro- viron. Microbiol. 72, 686–694 (2006). 9. Lovley, D. R. Dissimilatory Fe(III) and Mn(IV) re- biol. 1, 319–329 (1999). The first description of the only organism duction. Microbiol. Rev. 55, 259–287 (1991). 27. Shelobolina, E. S., VanPraagy, C. G. & Lovley, D. that has been clearly demonstrated to grow 10. Kukkadapu, R. K., Zachara, J. M., Smith, S. C., R. Use of ferric and ferrous iron containing min- by mesophilic autotrophic nitrate-depen- Fredrickson, J. K. & Liu, C. X. Dissimilatory bac- erals for respiration by Desulfitobacterium frap- dent Fe(II) oxidation. terial reduction of Al-substituted goethite in pieri. Geomicrobiol. J. 20, 143–156 (2003). 46. Beller, H. R. Anaerobic, nitrate-dependent ox- subsurface sediments. Geochim. Cosmochim. 28. Harrison Jr, A. P. The acidophilic Thiobacilli and idation of U(IV) oxide minerals by the chemo- Acta 65, 2913–2924 (2001). other acidophilic bacteria that share their habi- lithoautotrophic bacterium Thiobacillus denitri- 11. Mendelssohn, I. A., Kleiss, B. A. & Wakeley, J. S. taT. Ann. Rev. Microbiol. 38, 265–292 (1984). ficans. Appl. Environ. Microbiol. 71, 2170–2174 Factors controlling the formation of oxidized 29. Davison, W. & Seed, G. The kinetics of the ox- (2005). root channels. Wetlands 15, 37–46 (1995). idation of ferrous iron in synthetic and natural 47. Lack, J. G., Chaudhuri, S. K., Chakraborty, R., 12. Furukawa, Y., Smith, A. C., Kostka, J. E., Wat- waters. Geochim. Cosmochim. Acta 47, 67–79 Achenbach, L. A. & Coates, J. D. Anaerobic bio- kins, J. & Alexander, C. R. Quantification of mac- (1983). oxidation of Fe(II) by Dechlorosoma suillum. Mi- robenthic effects on diagenesis using a mul- 30. Emerson, D. & Weiss, J. V. Bacterial iron oxida- crob. Ecol. 43, 424–431 (2002). ticomponent inverse model in salt marsh tion in circumneutral freshwater habitats: find- 48. Coates, J. D. & Achenbach, L. A. Microbial per- sediments. Limnol. Oceanogr. 49, 2058–2072 ings from the field and the laboratory. Geomi- chlorate reduction: rocket-fuelled metabolism. (2004). crobiol. J. 21, 405–414 (2004). Nature Rev. Microbiol. 2, 569–580 (2004). 13. Emerson, D., Weiss, J. V. & Megonigal, J. P. Iron- 31. Hafenbradl, D. et al. Ferroglobus placidus gen. 49. Vorholt, J. A., Hafenbradl, D., Stetter, K. O. &

oxidizing bacteria are associated with ferric hy- nov., sp. nov. a novel hyperthermophilic ar- Thauer, R. K. Pathways of autotrophic CO2 fixa- droxide precipitates (Fe-plaque) on the roots chaeum that oxidizes Fe(II) at neutral pH under tion and dissimilatory nitrate reduction to N2O of wetland plants. Appl. Environ. Microbiol. 65, anoxic conditions. Arch. Microbiol. 166, 308– in Ferroglobus placidus. Arch. Microbiol. 167, 2758–2761 (1999). 314 (1996). 19–23 (1997). 14. Weiss, J. V., Emerson, D., Backer, S. M. & Mego- This paper describes the first isolation of a 50. Gold, T. The deep, hot biosphere. Proc. Natl nigal, J. P. Enumeration of Fe(II)-oxidizing and hyperthermophilic anaerobic, nitrate-de- Acad. Sci. USA 89, 6045–6049 (1992). Fe(III)-reducing bacteria in the root zone of pendent Fe(II)-oxidizing organism and the 51. Cairns-Smith, A. G. Precambrian solution pho- wetland plants: Implications for a rhizosphere only archaeum known to be capable of this tochemistry, inverse segregation, and banded iron cycle. Biogeochemistry 64, 77–96 (2003). metabolism. iron formations. Nature 276, 807–808 (1978). 15. Weiss, J. V., Emerson, D. & Megonigal, J. P. Rhi- 32. Jiao, Y., Kappler, A., Croal, L. R. & Newman, D. 52. Vargas, M., Kashefi, K., Blunt-Harris, E. L. & Lov- zosphere iron(III) deposition and reduction in a K. Isolation and characterization of a geneti- ley, D. R. Microbiological evidence for Fe(III) Juncus effusus L.-dominated wetland. Soil Sci. cally tractable photoautotrophic Fe(II)-oxidizing reduction on early Earth. Nature 395, 65–67 Soc. Am. J. 69, 1861–1870 (2005). bacterium, Rhodopseudomonas palustris strain (1998). 16. Ghiorse, W. C. Biology of iron-depositing and TIE-1. Appl. Environ. Microbiol. 71 4487–4496 Describes the first demonstration of dissim- manganese-depositing bacteria. Ann. Rev. Mi- (2005). ilatory Fe(III) reduction by hyperthermo- crobiol. 38, 515–550 (1984). 33. Ehrenreich, A. & Widdel, F. Anaerobic oxidation philic Archaea. 17. Emerson, D. & Moyer, C. L. Isolation and char- of ferrous iron by purple bacteria, a new type of 53. Lovley, D. R. in Origins: Genesis, Evolution and acterization of novel iron-oxidizing bacteria phototrophic metabolism. Appl. Environ. Micro- Diversity of Life (ed. Seckbach, J.) 707 (Kluwer that grow at circumneutral pH. Appl. Environ. biol. 60, 4517–4526 (1994). Dordrecht, Boston, 2004). Microbiol. 63, 4784–4792 (1997). 34. Heising, S. & Schink, B. Phototrophic oxidation 54. Kashefi, K. & Lovley, D. R. Extending the up- This paper demonstrates the unsuspected of ferrous iron by a Rhodomicrobium vannielii per temperature limit for life. Science 301, 934 ubiquity and diversity of organisms capable strain. Microbiology 144, 2263–2269 (1998). (2003). of microaerophilic Fe(II) bio-oxidation at cir- 35. Heising, S., Richter, L., Ludwig, W. & Schink, Provides a description of the most heat-tol- cumneutral pH. B. Chlorobium ferrooxidans sp. nov., a photo- erant organism known, strain 121, which 18. Sobolev, D. & Roden, E. E. Suboxic deposition trophic green sulfur bacterium that oxidizes grows at 121oC. of ferric iron by bacteria in opposing gradients iron in coculture with a ‘Geospirillum’ sp. strain. 55. Kashefi, K. & Lovley, D. Reduction of Fe(III), of Fe(II) and oxygen at circumneutral pH. Appl. Arch. Microbiol. 172, 116–124 (1999). Mn(IV), and toxic metals at 100 degrees C by Environ. Microbiol. 1328–1334 (2001). 36. Straub, K. L., Rainey, F. A. & Widdel, F. Rhodovu- Pyrobaculum islandicum. Appl. Environ. Micro- A n a e r o b i c m i c r o b i a l i r o n o x i d at i o n a n d r e d u c t i o n 763

biol. 66, 1050–1056 (2000). 74. Coates, J. D., Cole, K. A., Chakraborty, R., the Fe(III) reduction zone of petroleum-con- 56. Tor, J. M., Kashefi, K. & Lovley, D. R. Acetate ox- O’Connor, S. M. & Achenbach, L. A. The diver- taminated aquifers. Environ. Sci. Technol. 32, idation coupled to Fe(III) reduction in hyper- sity and ubiquity of bacteria utilizing humic 1222–1229 (1998). thermophilic microorganisms. Appl. Environ. substances as an electron donor for anaerobic 91. Hugenholtz, P., Goebel, B. M. & Pace, N. R. Im- Microbiol. 67, 1363–1365 (2001). respiration. Appl. Environ. Microbiol. 68, 2445– pact of culture-independent studies on the 57. Kostka, J. E., Stucki, J. W., Nealson, K. H. & Wu, 2452 (2002). emerging phylogenetic view of bacterial diver- J. Reduction of structural Fe(III) in smectite by 75. Roh, Y. et al. Isolation and characterization of sity. J. Bacteriol. 180, 4765–4774 (1998). a pure cultrue of Shewanella putrefaciens strain metal-reducing Thermoanaerobacter strains 92. Barns, S. M., Takala, S. L. & Kuske, C. R. Wide MR-1. Clays Clay Miner. 44, 522–529 (1996). from deep subsurface environments of the distribution and diversity of members of the This paper gives the first demonstration Piceance Basin, Colorado. Appl. Environ. Micro- bacterial kingdom Acidobacterium in the en- that structural Fe(III) in clay minerals is bio- biol. 68, 6013–6020 (2002). vironmenT. Appl. Environ. Microbiol. 65, 1731– available for dissimilatory microbial Fe(III) 76. Bowman, J. P. et al. Shewanella gelidimarina 1737 (1999). reduction. sp. nov. and Shewanella frigidimarina sp. nov., 93. Roberts, J. L. Reduction of ferric hydroxide by 58. Kostka, J. E. & Nealson, K. H. Dissolution and novel Antarctic species with the ability to pro- strains of Bacillus polymyxa. Soil Sci. 63, 135– reduction of magnetite by bacteria. Environ. Sci. duce eicosapentaenoic acid (20:5 omega 3) and 140 (1947). Technol. 29, 2535–2540 (1995). grow anaerobically by dissimilatory Fe(III) re- 94. Lovley, D. R. & Phillips, E. J. P. Organic matter 59. Fredrickson, J. K. et al. Biogenic iron mineraliza- duction. InT. J. SysT. Bacteriol. 4, 1040–1047 mineralization with reduction of ferric iron in tion accompanying the dissimilatory reduction (1997). anaerobic sediments. Appl. Environ. Microbiol. of hydrous ferric oxide by a groundwater bac- 77. Kusel, K., Dorsch, T., Acker, G. & Stackebrandt, 51, 683–689 (1986). terium. Geochim. Cosmochim. Acta 62, 3239– E. Microbial reduction of Fe(III) in acidic sedi- 95. Dobbin, P. S. et al. Dissimilatory Fe(III) reduction 3257 (1998). ments: isolation of Acidiphilium cryptum JF-5 by Clostridium beijerinckii isolated from fresh- 60. Glasauer, S., Weidler, P. G., Langley, S. & Bev- capable of coupling the reduction of Fe(III) to water sediment using Fe(III) maltol enrichmenT. eridge, T. J. Controls on Fe reduction and min- the oxidation of glucose. Appl. Environ. Micro- FEMS Microbiol. LetT. 176, 131–138 (1999). eral formation by a subsurface bacterium. Geo- biol. 65, 3633–3640 (1999). 96. Coleman, M. L., Hedrick, D. B., Lovley, D. R., chim. Cosmochim. Acta 67, 1277–1288 (2003). 78. Ye, Q. et al. Alkaline anaerobic respiration: Isola- White, D. C. & Pye, K. Reduction of Fe(III) in 61. Nevin, K. P. & Lovley, D. R. Lack of production of tion and characterization of a novel alkaliphilic sediments by sulphate-reducing bacteria. Na- electron-shuttling compounds or solubilization and metal-reducing bacterium. Appl. Environ. ture 361, 436–438 (1993). of Fe(III) during reduction of insoluble Fe(III) Microbiol. 70, 5595–5602 (2004). 97. Lovley, D. R., Roden, E. E., Phillips, E. J. P. & oxide by Geobacter metallireducens. Appl. Envi- 79. Gorlenko, V. et al. Anaerobranca californien- Woodward, J. C. Enzymatic iron and uranium ron. Microbiol. 66, 2248–2251 (2000). sis sp nov., an anaerobic, alkalithermophilic, fer- reduction by sulfate-reducing bacteria. Mar. 62. Childers, S. E., Ciufo, S. & Lovley, D. R. Geo- mentative bacterium isolated from a hot spring Geol. 113, 41–53 (1993). bacter metallireducens accesses insoluble Fe(III) on Mono Lake. InT. J. SysT. Evol. Microbiol. 54, 98. Bond, D. R. & Lovley, D. Reduction of Fe(III) ox- oxide by chemotaxis. Nature 416, 767–769 739–743 (2004). ide by methanogens in the presence and ab- (2002). 80. Myers, C. R. & Nealson, K. H. Bacterial manga- sence of extracellular quinones. Environ. Micro- 63. Reguera, G. et al. Extracellular electron trans- nese reduction and growth with manganese biol. 4, 115–124 (2002). fer via microbial nanowires. Nature 435, 1098– oxide as the sole electron-acceptor. Science 99. Lovley, D. & Phillips, E. Competitive mecha- 1101 (2005). 240, 1319–1321 (1988). nisms for inhibition of sulfate reduction and Provides the first description of the nanow- 81. Caccavo F. Jr, Blakemore, R. P. & Lovely, D. R. methane production in the zone of ferric iron ire concept involved in the transfer of elec- A hydrogen-oxidizing, Fe(III)-reducing micro- reduction in sediments. Appl. Environ. Micro- trons onto insoluble electron acceptors dur- organism from the Great Bay Estuary, New biol. 53, 2636–2641 (1987). ing microbial respiration. Hampshire. Appl. Environ. Microbiol. 58, 3211– 100. Roden, E. E. & Wetzel, R. G. Organic carbon ox- 64. Gorby, Y. A. et al. Electrically conductive bacte- 3216 (1992). idation and supression of methane produc- rial nanowires produced by Shewanella onei- 82. Zachara, J. M. et al. Bacterial reduction of crys- tion by microbial Fe(III) oxide reduction in veg- densis strain MR-1 and other microorganisms. talline Fe(III) oxides in single phase suspensions etated and unvegetated freshwater wetland Proc. Natl. Acad. Sci. USA 103, 11358–11363 and subsurface materials. Am. Mineral. 83, sediments. Limnol. Oceanogr. 41, 1733–1748 (2006). 1426–1443 (1998). (1996). 65. Lovley, D. R., Coates, J. D., Blunt-Harris, E. L., 83. Stein, L., La Duc, M., Grundl, T. & Nealson, K. 101. Madigan, M. T., Martinko, J. M. & Parker, J. Phillips, E. J. P. & Woodward, J. C. Humic sub- Bacterial and archaeal populations associated Brock Biology of Microorganisms (Pearson Edu- stances as electron acceptors for microbial res- with freshwater ferromanganous micronodules cation, New Jersey, 2002). piration. Nature 382, 445–448 (1996). and sediments. Environ. Microbiol. 3, 10–18 102. Reid, G. A. et al. Structure and function of flav- This paper provides the first demonstration (2001). ocytochrome c(3), the soluble fumarate reduc- that bacteria can mediate the reduction of 84. Snoeyenbos-West, O. L., Nevin, K. P., Anderson, tase from Shewanella NCIMB400. Biochem. Soc. insoluble electron acceptors through the use R. T. & Lovley, D. R. Enrichment of Geobacter Trans. 26, 418–421 (1998). of redox-active natural organic matter. species in response to stimulation of Fe(III) re- 103. Lovley, D. R. et al. Geobacter metallireducens 66. Newman, D. K. & Kolter, R. A role for excreted duction in sandy aquifer sediments. Microb. gen. nov. sp. nov., a microorganism capable of quinones in extracellular electron transfer. Na- Ecol. 39, 153–167 (2000). coupling the complete oxidation of organic- ture 405, 94–97 (2000). 85. Roling, W., van Breukelen, B., Braster, M., Lin, B. compounds to the reduction of iron and other 67. Nevin, K. P. & Lovley, D. R. Mechanisms for ac- & van Verseveld, H. Relationships between mi- metals. Arch. Microbiol. 159, 336–344 (1993). cessing insoluble Fe(III) oxide during dissimi- crobial community structure and hydrochemis- 104. Myers, C. R. & Myers, J. M. Role of menaqui- latory Fe(III) reduction by Geothrix fermentans. try in a landfill leachate-polluted aquifer. Appl. none in the reduction of fumarate, nitrate, Appl. Environ. Microbiol. 68, 2294–2299 (2002). Envir. Microbiol. 67, 4619–4629 (2001). iron(III) and manganese(IV) by Shewanella pu- 68. Turick, C. E., Tisa, L. S. & Caccavo, F. Melanin 86. Todorova, S. G. & Costello, A. M. Design of She- trefaciens Mr-1. FEMS Microbiol. LetT. 114, 215– production and use as a soluble electron shut- wanella specific 16S rRNA primers and appli- 222 (1993). tle for Fe(III) oxide reduction and as a termi- cation to analysis of Shewanella in a minero- 105. Saffarini, D. A., Blumerman, S. L. & Man- nal electron acceptor by Shewanella algae BrY. trophic wetland. Environ. Microbiol. 8, 426–432 soorabadi, K. J. Role of menaquinones in Fe(III) Appl. Environ. Microbiol. 68, 2436–2444 (2002). (2006). reduction by membrane fractions of Shewanella 69. Nevin, K. P. & Lovley, D. R. Potential for nonen- 87. Cummings, D., Caccavo, F., Spring, S. & Rosen- putrefaciens. J. Bacteriol. 184, 846–848 (2002). zymatic reduction of Fe(III) via electron shut- zweig, R. Ferribacterium limneticum, gen. 106. Myers, C. R. & Myers, J. A. Shewanella oneiden- tling in subsurface sediments. Environ. Sci. nov., sp. nov., an Fe(III)-reducing microorgan- sis MR-1 restores menaquinone synthesis to a Technol. 34, 2472–2478 (2000). ism isolated from mining-impacted freshwater menaquinone-negative mutanT. Appl. Environ. 70. Hernandez, M. E., Kappler, A. & Newman, D. K. lake sediments. Arch. Microbiol. 171, 183–188 Microbiol. 70, 5415–5425 (2004). Phenazines and other redox-active antibiotics (1999). 107. Heidelberg, J. F. et al. Genome sequence of the promote microbial mineral reduction. Appl. En- 88. Finneran, K., Johnsen, C. & Lovley, D. Rho- dissimilatory metal ion-reducing bacterium viron. Microbiol. 70, 921–928 (2004). doferax ferrireducens sp nov., a psychrotolerant, Shewanella oneidensis. Nature Biotechnol. 20, 71. Hernandez, M. E. & Newman, D. K. Extracellular facultatively anaerobic bacterium that oxidizes 1118–1123 (2002). electron transfer. Cell. Mol. Life Sci. 58, 1562– acetate with the reduction of Fe(III). InT. J. SysT. 108. Methe, B. A. et al. Genome of Geobacter sul- 1571 (2001). Evol. Microbiol. 53, 669–673 (2003). furreducens: metal reduction in subsurface en- 72. Nevin, K. P. & Lovley, D. R. Mechanisms for 89. Coates, J. D., Ellis, D. J., Gaw, C. V. & Lovley, D. R. vironments. Science 302, 1967–1969 (2003). Fe(III) oxide reduction in sedimentary environ- Geothrix fermentans gen. nov. sp. nov. a novel 109. Leang, C. et al. Adaptation to disruption of the ments. Geomicrobiol. J. 19, 141–159 (2002). Fe(III)-reducing bacterium from a hydrocarbon electron transfer pathway for Fe(III) reduction 73. Lovley, D. R., Fraga, J. L., Coates, J. D. & Blunt- contaminated aquifer. InT. J. SysT. Bacteriol. 49, in Geobacter sulfurreducens. J. Bacteriol. 187, Harris, E. L. Humics as an electron donor for an- 1615–1622 (1999). 5918–5926 (2005). aerobic respiration. Environ. Microbiol. 1, 89–98 90. Anderson, R. T., Rooney-Varga, J. N., Gaw, C. V. 110. Myers, C. R. & Myers, J. M. Cloning and se- (1999). & Lovely, D. R. Anaerobic benzene oxidation in quence of cymA a gene encoding a tetra- 764 W e b e r , A c h e n b a c h , & C o at e s i n N at u r e R e v i e w s M icrobiology 4 (2006)

heme cytochrome c required for reduction of 126. Braterman, P. S., Cairns-Smith, A. G. & Sloper, R. ron. Sci. Technol. 30, 2784–2789 (1996). iron(III), fumarate, and nitrate by Shewanella W. Photo-oxidation of hydrated Fe(II) — signif- 145. Coates, J., Anderson, R. & Lovley, D. Oxidation putrefaciens MR-1. J. Bacteriol. 179, 1143–1152 icance for banded iron formations. Nature 303, of polycyclic aromatic hydrocarbons under sul- (1997). 163–164 (1983). fate-reducing conditions. Appl. Environ. Micro- 111. Myers, J. M. & Myers, C. R. Role of the tetra- 127. Cloud, P. E. Significance of the Gunflint (Pre- biol. 62, 1099–1101 (1996). heme cytochrome CymA in anaerobic elec- cambrian) microflora. Science 148, 27–35 146. Anderson, R. T. & Lovley, D. Naphthalene and tron transport in cells of Shewanella putrefa- (1965). benzene degradation under Fe(III)-reducing ciens MR-1 with normal levels of menaquinone. 128. Lovley, D. R., Stolz, J. F., Nord, G. L. & Phillips, conditions in petroleum-contaminated aquifers. J. Bacteriol. 182, 67–75 (2000). E. J. P. Anaerobic production of magnetite by a Bioremediation J. 3, 121–135 (1999). 112. Beliaev, A. S. & Saffarini, D. A. Shewanella putre- dissimilatory iron-reducing microorganism. Na- 147. Lovley, D. R., Phillips, E. J. P., Gorby, Y. A. & faciens mtrB encodes an outer membrane pro- ture 330, 252–254 (1987). Landa, E. R. Microbial reduction of uranium. tein required for Fe(III) and Mn(IV) reduction. J. This paper gives the first demonstration of Nature 350, 413–416 (1991). Bacteriol. 180, 6292–6297 (1998). extracellular biogenic magnetite formation. This paper gives the first description of the 113. Beliaev, A. S., Saffarini, D. A., McLaughlin, J. L. & 129. Kohnauser, K. O. et al. Could bacteria have microbially mediated reduction of hexava- Hunnicutt, D. MtrC, an outer membrane deca- formed the Precambrian banded iron forma- lent uranium. haem c cytochrome required for metal reduc- tions? Geology 20, 1079–1082 (2002). 148. Lovley, D. R. & Coates, J. D. Novel forms of an- tion in Shewanella putrefaciens MR-1. Mol. Mi- 130. Kappler, A., Pasquero, C., Konhauser, K. O. & aerobic respiration of environmental relevance. crobiol. 39, 722–730 (2001). Newman, D. K. Deposition of banded iron for- Curr. Opin. Microbiol. 3, 252–256 (2000). 114. Pitts, K. E. et al. Characterization of the She- mations by anoxygenic phototrophic Fe(II)-oxi- 149. Langmuir, D. Aqueous Environmental Geo- wanella oneidensis MR-1 decaheme cyto- dizing bacteria. Geology 33, 865–868 (2005). chemistry (Prentice-Hall, New Jersey, 1997). chrome MtrA. J. Biol. Chem. 278, 27758–27765 131. Holm, N. G. The 13C/12C ratios of siderite and 150. Thamdrup, B. in Advanced Microbial Ecology (2003). organic matter from a modern metalliferrous (ed. Schink, B.) 41–84 (Kluwer Academic/Ple- 115. Dobbin, P. S., Butt, J. N., Powell, A. K., Reid, G. A. hydrothermal sediment and their implications num Publishers, New York, 2000). & Richardson, D. J. Characterization of a flavo- for banded iron formations. Chem. Geol. 77, 151. Dutton, P. L. & Prince, R. C. in The Photosyn- cytochrome that is induced during the anaer- 41–45 (1989). thetic Bacteria (eds Clayton, R. A. & Sistrom, W. obic respiration of Fe(III) by Shewanella frigi- 132. Towe, K. M. Early Precambrian oxygen: a case R.) 525–570 (Plenum, New York, 1978). dimarina NCIMB400. Biochem. J. 342, 439–448 against photosynthesis. Nature 274, 657–661 (1999). (1978). 116. Gordon, E. H. J. et al. Identification and charac- 133. Canfield, D. E. & Teske, A. Late Proterozoic rise Databases terization of a novel cytochrome c(3) from She- in atmospheric oxygen concentration inferred wanella frigidimarina that is involved in Fe(III) from phylogenetic and sulphur-isotope studies. respiration. Biochem. J. 349, 153–158 (2000). Nature 382, 127–132 (1996). Entrez Genome Project: http://www. 117. Leys, D. et al. Crystal structures at atomic res- 134. Yung, Y. L. & McElroy, M. B. Fixation of nitro- ncbi.nlm.nih.gov/entrez/query. olution reveal the novel concept of ‘electron- gen in the prebiotic atmosphere. Science 203, fcgi?db=genomeprj harvesting’ as a role for the small tetraheme cy- 1002–1004 (1979). Chlorobium ferrooxidans tochrome c. J. Biol. Chem. 277, 35703–35711 135. Mancinelli, R. L. & McKay, C. P. The evolution of Dechloromonas aromatica (2002). nitrogen cycle. OriG. Life Evol. Biosph. 18, 311– Geobacter metallireducens 118. Myers, C. R. & Myers, J. M. Cell surface expo- 325 (1988). Geothrix fermentans sure of the outer membrane cytochromes of 136. Walker, J. C. G. Suboxic diagenesis in banded Marinobacter aquaeolei Shewanella oneidensis MR-1. LetT. Appl. Micro- iron formations. Nature 309, 340–342 (1984). biol. 37, 254–258 (2003). 137. Coates, J. D. & Chakraborty, R. in Bioremedia- Rhodopseudomonas palustris 119. Myers, J. M. & Myers, C. R. Overlapping role tion: A Critical Review (eds Head, I. M., Single- Shewanella frigidimarina of the outer membrane cytochromes of She- ton, I. & Milner, M. G.) 227–257 (Horizon Scien- Shewanella oneidensis wanella oneidensis MR-1 in the reduction of tific, Wymondham UK, 2003). Shewanella putrefaciens manganese(IV) oxide. LetT. Appl. Microbiol. 37, 138. Kim, H. J. et al. A mediator-less microbial fuel Thiobacillus denitrificans 21–25 (2003). cell using a metal reducing bacterium. Enzyme 120. Butler, J. E., Kaufmann, F., Coppi, M. V., Nunez, Microbiol. Technol. 30, 145–152 (2002). UniProtKB: http://ca.expasy.org/sprot C. & Lovley, D. R. MacA a diheme c-type cyto- 139. Bond, D., Holmes, D., Tender, L. & Lovely, D.R. CymA chrorne involved in Fe(III) reduction by Geo- Electrode-reducing microorganisms that har- bacter sulfurreducens. J. Bacteriol. 186, 4042– vest energy from marine sediments. Science MtrA 4045 (2004). 295, 483–485 (2002). Cytc3 121. Lloyd, J. R. et al. Biochemical and genetic char- 140. Tender, L. et al. Harnessing microbially gener- PpcA acterization of PpcA, a periplasmic c-type cyto- ated power on the seafloor. Nature Biotechnol. chrome in Geobacter sulfurreducens. Biochem. J. 20, 821–825 (2002). Further Information 369, 153–161 (2003). 141. Bond, D. & Lovley, D. Electricity production 122. Leang, C., Coppi, M. V. & Lovley, D. R. OmcB, a by Geobacter sulfurreducens attached to elec- John D. Coates’ homepage: c-type polyheme cytochrome, involved in Fe(III) trodes. Appl. Environ. Microbiol. 69, 1548–1555 http://pmb.berkeley.edu/~coates reduction in Geobacter sulfurreducens. J. Bacte- (2003). riol. 185, 2096–2103 (2003). 142. Bond, D. & Lovley, D. Evidence for involvement 123. Lovely, D. Cleaning up with genomics: applying of an electron shuttle in electricity generation Acknowledgments — Research on microbial molecular biology to bioremediation. Nature by Geothrix fermentans. Appl. Envir. Microbiol. Rev. Microbiol. 1, 35–44 (2003). 71, 2186–2189 (2005). redox cycling of iron in the laboratories 124. Mahadevan, R. et al. Characterization of me- 143. Lovley, D. et al. Oxidation of aromatic contami- of J.D.C. and L.A.A. is supported by grants tabolism in the Fe(III)-reducing organism Geo- nants coupled to microbial iron reduction. Na- from the US Department of Energy bacter sulfurreducens by constraint-based mod- ture 339, 297–299 (1989). Environmental Remediation Sciences elinG. Appl. Environ. Microbiol. 72, 1558–1568 144. Coates, J., Anderson, R., Woodward, J., Phil- Program. (2006). lips, E. & Lovely, D. R. Anaerobic hydrocarbon 125. Cloud, P. E. Paleoecological significance of the degradation in petroleum-contaminated har- banded iron-formation. Econ. Geol. 68, 1135– bor sediments under sulfate-reducing and arti- 1143 (1973). ficially imposed iron-reducing conditions. Envi-