Genomic Insights Into Microbial Iron Oxidation and Iron Uptake Strategies in Extremely Acidic Environments

Environmental Microbiology (2011) doi:10.1111/j.1462-2920.2011.02626.x

Minireview

Genomic insights into microbial iron oxidation and iron uptake strategies in extremely acidic

environmentsemi_2626 1..15

Violaine Bonnefoy1,2* and David S. Holmes3 tiple iron uptake systems. This could be an adaption 1Laboratoire de Chimie Bactérienne, UPR-CNRS 9043, allowing them to respond to different iron con- Institut de Microbiologie de la Méditerranée, Marseille, centrations via the use of a multiplicity of different France. siderophores. Both Leptospirillum spp. and Acidithio- 2Aix-Marseille Université, Marseille, France. bacillus spp. are predicted to synthesize the acid 3Center for Bioinformatics and Genome Biology, stable citrate siderophore for Fe(III) uptake. In addi- Fundación Ciencia para la Vida and Depto. de Ciencias tion, both groups have predicted receptors for Biologicas, Facultad de Ciencias Biologicas, siderophores produced by other microorganisms, Universidad Andres Bello, Santiago, Chile. suggesting that competition for iron occurs influenc- ing the ecophysiology of acidic environments. Little is known about the genetic regulation of iron oxidation Summary and iron uptake in acidophiles, especially how the use This minireview presents recent advances in our of iron as an energy source is balanced with its need to understanding of iron oxidation and homeostasis in take up iron for metabolism. It is anticipated that inte- acidophilic Bacteria and Archaea. These processes grated and complex regulatory networks sensing dif- influence the flux of metals and nutrients in pristine ferent environmental signals, such as the energy and man-made acidic environments such as acid mine source and/or the redox state of the cell as well as the drainage and industrial bioleaching operations. Acido- oxygen availability, are involved. philes are also being studied to understand life in extreme conditions and their role in the generation of Introduction biomarkers used in the search for evidence of existing or past extra-terrestrial life. Iron oxidation in acido- Scope of this minireview philes is best understood in the model organism This minireview focuses on iron-oxidizing microorganisms Acidithiobacillus ferrooxidans. However, recent func- that inhabit acidophilic environments. There are a number tional genomic analysis of acidophiles is leading to a of other reviews and papers that discuss issues related to deeper appreciation of the diversity of acidophilic iron- life in extremely acidic conditions that will not be covered oxidizing pathways. Although it is too early to paint a here, including the occurrence and composition of acido- detailed picture of the role played by lateral gene philic communities (Gonzalez-Toril et al., 2003; Rawlings transfer in the evolution of iron oxidation, emerging and Johnson, 2007; Johnson, 2008; Demergasso et al., evidence tends to support the view that iron oxidation 2010), resistance to low pH and metal homeostasis arose independently more than once in evolution. (Baker-Austin and Dopson, 2007; Franke and Rensing, Acidic environments are generally rich in soluble iron 2007; Dopson, 2010), reduced inorganic sulfur com- and extreme acidophiles (e.g. the Leptospirillum pounds (RISCs) energetics (Holmes and Bonnefoy, 2007; genus) have considerably fewer iron uptake systems Quatrini et al., 2009; Bonnefoy, 2010) and genomics and compared with neutrophiles. However, some acido- metabolic reconstruction (Tyson et al., 2004; Valenzuela philes have been shown to grow as high as pH 6 and, in et al., 2006; Holmes and Bonnefoy, 2007; Jerez, 2007; the case of the Acidithiobacillus genus, to have mul- 2008; Quatrini et al., 2007a; 2009; Valdes et al., 2008; Siezen and Wilson, 2009; Bonnefoy, 2010; Denef et al., Received 17 June, 2011; accepted 16 September, 2011. *For corre- spondence. E-mail [email protected]; Tel. (+33)4911641 2010) including genomics of over 50 acidophiles (Carde- 46; Fax (+33)491718914. nas et al., 2010).

Acidophiles and their environments hydroxides (as low as 10-18 M at pH 7.0), rendering it difficult to access by biological systems. In contrast, under We define ‘extreme acidophiles’ as those organisms acidic conditions Fe(II) is stable even in the presence of whose growth optimum is lower than pH 3. Most acido- atmospheric oxygen. This provides an opportunity for philes belong to the eubacterial or archaeal kingdoms microorganisms to use Fe(II) as an electron donor and as (prokaryotes) (Johnson, 1998; 2007; 2009; Hallberg and a source of energy that is not readily available to micro- Johnson, 2001; Johnson and Hallberg, 2003), although organisms living in circum-neutral environments. unicellular eukaryotes have also been detected in some However, extremely acidic environments can also chal- acidic environments (e.g. Lopez-Archilla et al., 2001; lenge microorganisms with the highest levels of soluble Amaral Zettler et al., 2002; Baker et al., 2009; Johnson, iron and the threat that this imposes on life via the reaction 2009; Cid et al., 2010). of Fe(II) with oxygen generating free radicals (Fenton Acidophilic microorganisms inhabit pristine environ- reaction) that damage macromolecules and cause cell ments such as volcanic, geothermal areas and acid rock death (Touati, 2000). drainage (ARD) (Lopez-Archilla et al., 2001; Gonzalez- Toril et al., 2003). They are also found in acidic environments of anthropogenic origin such as bioleaching heaps Phylogeny of iron oxidizers and acid mine drainage (AMD) (Rawlings, 2002; 2007; Baker and Banfield, 2003; Johnson and Hallberg, 2003; The ability to oxidize iron is widely distributed in acido- Johnson, 2007; 2009; Norris, 2007; Schippers, 2007; philic Bacteria and Archaea (Fig. 1) (Weber et al., 2006; Schippers et al., 2010). Acidophiles are found over a wide Emerson et al., 2010; Hedrich et al., 2011). Until now, the range of temperatures and include psychrotolerants (but majority of known acidophilic iron oxidizers are found in not psychrophiles) that can grow down to 4°C (Harrison, the Nitrospira class as well as in Gram-positive (Firmicute 1982; Johnson et al., 2001; Dopson et al., 2007; Hallberg and Actinobacteria) and archaeal (Crenarchaeota et al., 2010), mesophiles (up to 40°C), moderate thermo- and Euryarchaeota) phyla. In Proteobacteria, only five philes (between 40°C and 55°C) and extreme thermo- acidophiles are encountered (‘Ferrovum myxofaciens’, philes (above 55°C) (Hallberg and Johnson, 2001; Norris, Acidiferrobacter thiooxydans,‘Thiobacillus prosperus’, 2007; Schippers, 2007; Johnson, 2009). Many acidophilic Acidithiobacillus ferrivorans and At. ferrooxidans) prokaryotes are obligate chemolithoautotrophs obtaining (Hedrich et al., 2011). It is not yet known whether this their energy and electrons from the oxidation of ferrous distribution reflects real evolutionary trends or merely iron [Fe(II)] and RISCs (Hallberg and Johnson, 2001; results from an incomplete analysis due to biases of Rawlings, 2005; 2007; Johnson and Hallberg, 2008; sampling that have relied mainly on culture-based approaches. Johnson, 2009) and their carbon by fixing CO2 from the atmosphere (Rawlings, 2007; Johnson and Hallberg, Widespread distribution of iron oxidation capabilities has also been observed in microorganisms from circum- 2008). Some also fix atmospheric N2 (Johnson, 2007; Rawlings, 2007). Heterotrophs have also been detected neutral pH environments. However, as opposed to the that scavenge organic carbon compounds produced by acidophiles, there seems to be an enrichment of iron the autotrophs (Hallberg and Johnson, 2001; Rawlings, oxidizers in the Proteobacteria (Emerson et al., 2010). 2002; Johnson and Hallberg, 2003; Johnson, 2007; 2009; This minireview will focus mainly on iron oxidation in Norris, 2007; Schippers, 2007; Nancucheo and Johnson, acidic environments. Microbial iron oxidation in circum- 2010; Schippers et al., 2010). neutral environments has been reviewed recently (Emerson et al., 2010; Bird et al., 2011) and will not be discussed here, except for comparison with acidophilic Problems of high iron load in acidic conditions pathways. Acidic environments provide a special opportunity and at the same time an unusual challenge for life. Natural Fe(II) iron oxidation geomicrobiological processes and industrial operations, Bioenergetic problems of iron oxidation such as bioleaching, convert insoluble metal sulfides into water-soluble metal sulfates that include extraordinarily When growing on Fe(II), autotrophs are confronted with high concentrations of soluble iron that can reach values a crucial bioenergetic problem. The reduction potential of as high as 160 g l-1, a concentration about 1016 higher the Fe(II)/Fe(III) couple is +0.78 V which is just below than typically found in circum-neutral environments. In that of O2/H2O (1.12 V) at the pH of the external medium oxygen (O2)-saturated environments at neutral pH, Fe(II) (pH 2), consequently, there is little energy available from is rapidly oxidized to ferric iron [Fe(III)]. Thus, iron pre- the oxidation of Fe(II) (Ingledew, 1982; Bird et al., 2011). dominantly occurs in the ferric form as poorly soluble iron Also, since the reduction potential of NAD+ (-0.32 V at

Fig. 1. An inferred phylogenetic tree of acidophilic iron-oxidizing Bacteria and Archaea. The tree was produced by a neighbour-joining method based on 16S rRNA gene sequences (Kumar et al., 2008). Numbers at nodes represent bootstrap values. The scale bars represent the average number of substitutions per site. Whereas At. ferrooxidans clusters with the Gammaproteobacteria in this tree, it has recently been proposed to have arisen before the split between Gamma- and Betaproteobacteria based upon a phylogeny using protein concatenation (Williams et al., 2010). Strain designations (when speciﬁed) and 16 rRNA gene accession numbers are: 1 = embl|HM044161|, 2 = DSM 2392 and embl |AF387301|, 3 = DSM 5130 and ENA|EU653291|, 4 = DSM 22755 and ENA|AF376020|, 5 = ATCC 23270 and gi|28194033|, 6 = ATCC BAA-1645 and ENA|AY140237|, 7 = ENA|GQ225721|, 8 = ENA|AF251436|, 9 = DSM 10331 and ENA|CP001631|, 10 = gi|157649846|, 11 = gi|10998845|, 12 = ATCC 51911 and ENA|AB089843|, 13 = DSM 16297 and ENA|AB222265|, 14 = ATCC 700253 and ENA|EF088287|, 15 = DSM 9293 and ENA|AB089844|, 16 = DSM 17362 and ENA |AM502928|, 17 = gi|225380152|, 18 = DSM 17363T and ENA|AY079150|, 19 = ATCC BAA-1181 and ENA|ACNP01000064|, 20 = DSM 16297/ENA and AF356832|, 21 = ENA|AAWO01000056|, 22 = DSM 14647 and ENA|AF356830|, 23 = DSM 18409T and gi|261599790|, 24 = DSM 16651 and gi|254971292|, 25 = ENA|AABC05000010|, 26 = DSM 6482 and ENA|AJ224936|, 27 = DSM 16993 and ENA|BA000023|, 28 = DSM 6482 and ENA|D85519|, 29 = DSM 3191 and ENA|D85505|, 30 = DSM 1651 and ENA|D26489|, 31 = DSM 5348 and ENA|CP00682|.

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology 4 V. Bonnefoy and D. S. Holmes pH 6.5, the pH of the cytoplasm) is much more negative evolved in prokaryotes. This hypothesis has since than that of the Fe(II)/Fe(III) couple, electrons must be received additional bioinformatic and proteomic/ pushed ‘uphill’ or ‘in reverse’ from Fe(II) to NAD+ against transcriptomic support (see reviews Holmes and Bonne- the redox potential gradient. The energy to accomplish foy, 2007; Bonnefoy, 2010). this comes from the proton motive force across the cell membrane that results from the high concentration of Acidithiobacillus ferrivorans. Acidithiobacillus ferrivorans protons outside the cell (pH 2) compared with inside (pH has been recently characterized as an Fe(II)- and RISCs- 6.5). On the other hand, electrons extracted from Fe(II) oxidizing acidophile (Hallberg et al., 2010; Amouric et al., can pass ‘downhill’ via a thermodynamically favourable 2011). Although it is phylogenetically closely related to gradient to reduce O2 to water, thus neutralizing protons At. ferrooxidans at the 16S rRNA gene level (see Fig. 1), that have entered the cell via the ATPase complex. unlike this microorganism it is motile and psychrotolerant. These up- and downhill pathways are intimately con- It exhibits iro (Hallberg et al., 2009; Amouric et al., 2011) nected at the level of electron and proton fluxes and are encoding a high potential iron-sulfur protein (Table 1) that likely to be homeostatically regulated in order to balance has been proposed to be the iron oxidase (Fukumori ATP production with the reconstitution of reducing power et al., 1988; Kusano et al., 1992; Cavazza et al., 1995). In (NADH) as been demonstrated for Acidithiobacillus addition, it has not only the archetypal rusA encoding the caldus (Dopson et al., 2002). blue copper protein rusticyanin (Liljeqvist et al., 2011) as found in the At. ferrooxidans type strain and others (Ben- grine et al., 1998; Valdes et al., 2008) but also rusB (Hall- Best-studied case: At. ferrooxidans berg et al., 2010; Liljeqvist et al., 2011; Amouric et al., The most detailed account of electron transport pathways 2011) as described in some other At. ferrooxidans strains and the molecular complexes used during Fe(II) oxidation (Ida et al., 2003; Sasaki et al., 2003). is available for the Gram-negative bacterium At. ferrooxi- The iron-oxidizing Acidithiobacillus spp. have recently dans that was considered until recently to be a member of been shown to be composed of at least four groups likely the Gammaproteobacteria but now is thought to have representing four species (Amouric et al., 2011). In two of arisen before the split between Gamma- and Betaproteo- these groups, including the At. ferrooxidans type strain, bacteria (Williams et al., 2010). These studies were initi- only rusA was detected, while the two other taxons, which ated almost 30 years ago in a pioneering paper by include At. ferrivorans, encode iro and have either rusA Ingledew who not only provided a theoretical framework and rusB or no rus (Liljeqvist et al., 2011; Amouric et al., for the electrochemistry of electron and proton fluxes but 2011). This suggests that at least two different pathways also proposed the bifurcating (uphill and downhill) elec- for Fe(II) oxidation exist in the Acidithiobacillus spp., the tron transport pathways and suggested which molecular first via rusA and the second possibly through the HiPIP complexes might be involved (Ingledew, 1982). The Ingle- encoded by iro. dew model has proved to be essentially correct and, in the intervening years, has received validation from numerous ‘Thiobacillus prosperus’. This Gammaproteobacterium is experimental and bioinformatics studies (see Fig. 2) an Fe(II) and S°-oxidizing, salt-tolerant acidophile. A (reviewed in Holmes and Bonnefoy, 2007; Quatrini et al., cluster of genes with similarity to a substantial part of the 2009; Bonnefoy, 2010; Bird et al., 2011 and additional rus operon of At. ferrooxidans was identified including: a references therein). predicted outer membrane cytochrome c (37% identity with Cyc2); a multicopper oxidase whose function is subject to controversy (46% identity with Cup) (Quatrini Iron oxidation in other acidophiles et al., 2009; Castelle et al., 2010); four subunits of a puta-

In addition to At. ferrooxidans, other known acidophilic tive aa3-type cytochrome oxidase (49%, 54%, 31% and prokaryotes whose iron oxidation pathways have been 32% identity with CoxB, A, C and D respectively) and a studied, include the bacteria At. ferrivorans,‘T. prospe- predicted blue copper protein showing 50% identity with rus’, Leptospirillum ferrooxidans, ‘L. rubarum’ and ‘L. fer- rusticyanin A and 49% with rusticyanin B (Table 1) (Nicolle rodiazotrophum’ and the archaea Metallosphaera sedula, et al., 2009). However, cyc1 encoding the membrane- ‘Ferroplasma acidarmanus’, Ferroplasma Types I and II, bound cytochrome c proposed to transfer the electrons

Sulfolobus metallicus and S. tokodaii. during uphill ﬂow from rusticyanin to the terminal aa3 It has been known since the 1990s that At. ferrooxi- oxidase in At. ferrooxidans was not detected in this locus dans, L. ferrooxidans, M. sedula and S. metallicus (see Fig. 2). In ‘T. prosperus’, it is possible that this func- contain spectrally distinct redox proteins when grown on tion is assumed by another electron transfer protein, such Fe(II) (Barr et al., 1990; Blake et al., 1992; 1993) suggest- as the multicopper oxidase Cup. Alternatively, electron ing that several different Fe(II) oxidation pathways have transfer between rusticyanin and the terminal oxidase is

Fig. 2. Model of the oxidation of Fe(II) in At. ferrooxidans based on biochemical, molecular genetics, bioenergetic, bioinformatic and functional genomics evidence. A redox tower has been placed below the model to facilitate comparison of the redox potentials of some of the reactions. Electrons extracted from the oxidation of Fe(II) by the outer membrane embedded cytochrome c Cyc2 are passed to the periplasmic copper protein rusticyanin (R). From rusticyanin, the electrons can take a download pathway to reduce O2 to water (solid arrow) passing through the cytochrome c4 Cyc1 and the aa3 type cytochrome oxidase complex or an uphill pathway (dotted arrow) to the NADH1 complex via the cytochrome c4 CycA1, the bc1 complex and membrane-associated quinones. The energy to push electrons uphill against this thermodynamically unfavourable gradient is postulated to come from the influx of protons (solid arrows) generated by the proton motive force across the inner membrane resulting from the difference in proton concentration inside the cell (pH 6.5) and outside (pH 2). This uphill pathway is similar in many respects to the pathways taken by electrons and protons in mitochondrial and prokaryotic oxidative phosphorylation, only in reverse. The figure also shows the influx of protons through the ATP synthetase complex (ATPase) driving the biosynthesis of ATP. These protons and also those that enter the cell to drive electrons uphill are postulated to be consumed, at least in part, by the reduction of O2 to water using electrons derived from the oxidation of Fe(II) via the downhill pathway. Abbreviations used: R, • 1 rusticyanin; OM, outer membrane; IM, inner membrane. * = values of E 0 for the Fe(III)/Fe(II) and /2 O2/H2O couples for pH 2 (Ferguson and Ingledew, 2008). Figure reproduced with modifications from Quatrini and colleagues (2009). made directly without the need for an intermediary 2009). However, based on the sequence similarity protein. Also, it cannot be excluded that cyc1 exists in between the redox proteins and the organization of the some other part of the genome and is not associated with corresponding genes, it has been proposed that the com- the rus operon as in At. ferrooxidans. position of the electron transfer chain from Fe(II) to O2 in Another difference is that rus is transcribed indepen- ‘T. prosperus’ might be similar to that of At. ferrooxidans dently from the upstream genes in ‘T. prosperus’ whereas (Nicolle et al., 2009). it is either co-transcribed with them or transcribed from an internal promoter in At. ferrooxidans (Bengrine et al., Leptospirillum spp. Leptospirillum spp. are bacteria 1998). In addition, it has been reported that in ‘T. prospe- belonging to the deep branching class Nitrospira. Based rus’ transcription of cyc2 and rus is higher in S° than on metagenomic and metaproteomic information derived in Fe(II)-medium while the coxBACD cluster is more from a study of biofilms in the acid mine drainage of Iron expressed during growth on Fe(II) (Nicolle et al., 2009). Mountain (USA), a model has been proposed for Fe(II) This contrasts with the situation in At. ferrooxidans where oxidation for Leptospirillum ferriphilum,‘L. rubarum’ and rus, cyc2 and coxBACD are upregulated in medium ‘L. ferrodiazotrophum’ (Table 1) (Tyson et al., 2004; Ram containing Fe(II) (Yarzabal et al., 2004; Quatrini et al., et al., 2005; Lo et al., 2007; Simmons et al., 2008; Golts-

Table 1. Comparison of the redox proteins demonstrated, or proposed, to be involved in Fe(II) oxidation in acidophiles.

Bacteria Outer membrane Periplasm Inner membrane peripheral Inner membrane

Acidithiobacillus ferrooxidans Cyc2 Rus A Cyc1 aa3 oxidase CycA1 bc1 complex Acidithiobacillus ferrivorans Iro RusAorB?

‘Thiobacillus prosperus’ Cyc2-like Rus-like aa3 oxidase Leptospirillum spp. Cyc572 Cyc579 Cytc? cbb3 oxidase Cytc? bc1 complex

Archaea Membrane peripheral Inner membrane

Ferroplasma spp. Sulfocyanin cbb3 oxidase Sulfolobus metallicus and S. tokodaii cyt b, [Fe-S], haem-copper terminal oxidase Metallosphaera sedula cyt b, [Fe-S], haem-copper terminal oxidase man et al., 2009; Singer et al., 2010). In this model, elec- Sulfolobus spp. These archaea are acidophilic and trons are extracted from Fe(II) by an outer membrane extremely thermophilic. Among them, two species have cytochrome c termed Cyc572.Cyc572 passes these elec- been shown to oxidize Fe(II): S. metallicus and S. toko- trons to a periplasmic cytochrome c termed Cyt579, which daii. As for all thermoacidophilic archaea studied so far, subsequently transfers them via periplasmic cytochromes there is no evidence for genes encoding cytochromes c.A c either downhill to a cytoplasmic membrane embedded CbsA-like cytochrome b and a haem-copper terminal cbb3 terminal oxidase that reduces O2 or uphill ﬁrst to a oxidase encoded by the fox [for Fe(II) oxidation] cluster bc1 complex and then to the NADH dehydrogenase have been proposed to be involved in Fe(II) oxidation complex via the quinone pool. The postulated route and in S. metallicus based on a subtractive hybridization protein complexes for both downhill and uphill ﬂow in approach (Table 1) (Bathe and Norris, 2007). In this same Leptospirillum has many features similar to those pro- cluster, are found genes encoding electron transporters, posed for At. ferrooxidans with the notable difference that such as putative ferredoxins and other iron-sulfur pro- Leptospirillum has no equivalent to rusticyanin and the teins. In agreement with the hypothesis that this cluster is branch point at which electrons from Fe(II) oxidation are involved in Fe(II) oxidation, these genes have also been channelled either uphill or downhill was proposed to be detected in the genome sequence of S. tokodaii, but not in

Cyt579. other Sulfolobus species that do not oxidize Fe(II) (Bathe and Norris, 2007). Ferroplasma spp. These acidophilic archaea are aerobic The fox cluster may encode components involved in Fe(II) oxidizers that grow mixotrophically or chemohet- both downhill and uphill electron ﬂow since it has genes erotrophically. An analysis of the metagenome of the Iron potentially encoding a haem terminal oxidase, presum-

Mountain biofilm (Tyson et al., 2004; Allen et al., 2007) ably involved in O2 reduction, and also cytochrome b and and biochemical, genomic and proteomic analysis iron-sulfur proteins that could be components of a bc1 (Dopson et al., 2005), allowed a model to be built for complex. Fe(II) oxidation in three Ferroplasma spp. (Table 1). The presence of the blue copper sulfocyanin, a member of a Metallosphaera sedula. This obligate thermoacidophilic family of proteins to which rusticyanin also belongs, is archaeon is a facultative chemolithoautotroph, being able proposed to transfer electrons to a cbb3 terminal oxidase, to oxidize Fe(II) or RISCs in aerobic conditions, and to but little is known about the protein(s) involved in the grow on complex organic substrates. Genomic and tran- initial extraction of electrons from Fe(II) nor in other scriptomic analyses have allowed the prediction of potential intermediate steps. Unlike At. ferrooxidans, components involved in Fe(II) oxidation (Auernik and cytochrome c has not been detected in Ferroplasma spp. Kelly, 2008; Auernik et al., 2008). A fox cluster has been The presence of sulfocyanin raises the possibility that it detected in M. sedula and has been determined to be serves as branch point like rusticyanin between uphill more expressed in the presence of Fe(II) than S°. It has and downhill electron flow. been suggested that this cluster could be involved in the No electron carriers for an uphill pathway of Fe(II) oxi- ‘uphill’ pathway (Auernik and Kelly, 2008). However, dation have been identified in Ferroplasma spp., but it is given the possible connection between the fox cluster possible that this pathway is not required since the and terminal electron transfer in S. metallicus,we organisms may be able to generate NADH using elec- suggest that it is more likely that the fox cluster also trons and energy derived from the oxidation of carbon. plays a similar role in downhill electron flow in M. sedula.

In addition, a cytochrome ba complex encoded by the A comparison of Fe(II) oxidation between acidophiles soxN-soxL-cbsABA locus and proposed recently to be and neutrophiles analogous to the bc1 complex of bacteria and mitochon- dria (Bandeiras et al., 2009) seems to be also involved in Whereas information regarding iron oxidation for several the oxidation of Fe(II) and could therefore represent the acidophiles is beginning to emerge after its ﬁrst charac- uphill pathways (Table 1). However, these genes are also terization in At. ferrooxidans, data are scarce for neutro- present in Acidianus ambivalens, Caldivirga maquilin- philes with only two operons described (Bird et al., 2011). gensis, Sulfolobus acidocaldarius and Sulfolobus solfa- In Rhodobacter capsulatus SB1003, a cytochrome c,a taricus (Bandeiras et al., 2009). pyrroloquinoline quinone containing protein and an inner membrane protein encoded by the foxEYZ operon (Croal et al., 2007) while a decahaem periplasmic cytochrome c, Summary of Fe(II) oxidation pathways in acidophiles an outer membrane protein and a high potential iron sulfur A pathway for Fe(II) oxidation is known in detail only protein (HiPIP) encoded by the pioABC operon in in At. ferrooxidans. However, information is emerging Rhodopseudomonas palustris TIE-1 (Jiao and Newman, from other acidophiles and Table 1 summarizes our 2007) have been suggested to be involved in iron oxida- current knowledge. A comparison of known Fe(II) tion. Interestingly, PioA and PioB present some similarities oxidation pathways in acidophilic bacteria suggests to the soluble decahaem cytochrome c MtrA and to the that they all accomplish the preliminary removal of elec- outer membrane MtrB of Shewanella spp. and Geobacter trons from their Fe(II) substrates using an outer mem- spp. In these latter Fe(III)-reducing neutrophilic bacteria, brane cytochrome c. Most likely this is to avoid the MtrB was proposed to serve as a sheath within which the production of damaging free radicals and precipitates of periplasmic MtrA receives the electrons from the outer ferric oxyhydroxides within the cell in the case of soluble membrane MtrC cytochrome c which reduces Fe(III) substrates such FeSO4. Cyc2, which accomplishes this (Hartshorne et al., 2009). By homology, PioB could pos- role in At. ferrooxidans, has 47% and 48% sequence sibly allow the transfer of the electrons from Fe(II) to the similarity within a stretch of 489 and 65 amino acids cytochrome c PioA or to the HiPIP PioC which are pre- including the haem C binding site with the predicted dicted to be periplasmic. Therefore, the Fe(II) oxidation outer membrane cytochromes of ‘T. prosperus’ and Lep- system of R. palustris TIE-1 shares more similarities to tospirillum spp., respectively, suggesting that they may the ferric-reducing system of Shewanella spp. and Geo- have arisen from a common ancestor. However, since bacter spp. than to the Fe(II) oxidation system of R. cap- the identity is very low they either diverged very earlier sulatus SB1003, which is phylogenetically closer. A in evolution or else have been subjected to rapid pioABC operon was also detected in the genome of the selection. Fe(II) oxidizer Gallionella capsiferriformans suggesting A commonality between acidophilic Bacteria and that this Betaproteobacterium oxidizes Fe(II) in a similar Archaea is the use of a terminal oxidase located in the way to R. palustris TIE-1. In most acidophilic bacteria, inner membrane that uses electrons derived from the while an outer membrane cytochrome c has been pre- oxidation of Fe(II) (downhill pathway) to consume or dicted, and in some cases characterized (see above), no extrude protons that enter the cell via the ATP synthetase equivalent to pioB/mtrB was detected. Even if a HiPIP was in order to maintain intracellular neutrality. proposed to be the iron oxidase in the iron-oxidizing A third feature that appears to be shared is the ability Acidithiobacillus spp. of Groups III and IV (Kusano et al., to push electrons uphill against a thermodynamically 1992; Amouric et al., 2011) (see above), the percentage of unfavourable gradient using PMF in order to reduce identity/similarity with PioC is rather low (about 36/50%) + NAD . This appears to be accomplished by a bc1 and the genetic context completely different (Kusano complex. et al., 1992; Jiao and Newman, 2007; Amouric et al., However, what is striking, is the dissimilarity in the com- 2011) suggesting that the model proposed for R. palustris ponents involved in Fe(II) oxidation in many acidophiles, TIE-1 is unlikely in these acidophiles. Therefore, from and we speculate that the ability to oxidize Fe(II) arose what we know nowadays on the iron oxidation pathways, independently more than once in evolution. For example, no consensus model in neutrophiles or acidophiles can be whereas At. ferrooxidans and Ferroplasma spp. use deduced and each prokaryote seems to have evolved its members of the small copper protein family as electron own system. carriers (rusticyanin and sulfocyanin respectively), this However, there are six species that do not follow the role may be accomplished by Fe-S proteins in Sulfolobus split between acidophilic and neutrophilic iron oxidizers spp. and in M. sedula. In addition, the terminal oxidase according to the 16S rRNA gene phylogenetic tree used can be of the aa3 (e.g. At. ferrooxidans)orcbb3 type (Fig. 1): namely the neutrophile Crenarchaeota Ferro- (e.g. Leptospirillum sp.). globus placidus and the acidophilic Proteobacteria

‘Ferrovum myxofaciens’, Acidiferrobacter thiooxydans, cular adaptations that underpin the ability of these micro- ‘T. prosperus’, At. ferrivorans and At. ferrooxidans.At organisms to cope with high concentrations of soluble least four of these acidophiles, the Groups I (At. ferrooxi- iron. Attention has focused to date on Bacteria of the dans), II and III (At. ferrivorans) of the iron-oxidizing Leptospirillum and Acidithiobacillus genera and Archaea Acidithiobacillus spp. as well as the ‘T. prosperus’, share a of the Ferroplasma genera. rus locus presenting similarities (see above) while the iron-oxidizing Acidithiobacillus spp. from Groups III and IV, which are phylogenetically and physiologically close to Iron homeostasis in the Leptospirillum spp. and Groups I and II, have in addition the rusB gene or no Acidithiobacillus spp. detectable rus gene at all (Liljeqvist et al., 2011; Amouric The apparently obligatory Fe(II)-oxidizing Leptospirillum et al., 2011). One possibility is that lateral gene transfers spp. have only a few predicted iron uptake mechanisms could have occurred. that include EfeU (FtrI-Fet3P-like permease) for Fe(II) Neutrophilic Fe(III)-reducing and Fe(II)-oxidizing micro- and only three TonB-dependent ferri-citrate siderophore organisms have been shown to use external electron receptor systems for Fe(III) (Tyson et al., 2004; Parro carriers (reviewed in Gralnick and Newman, 2007) such et al., 2007; Osorio et al., 2008). Citrate siderophore as redox-active antibiotics (Hernandez et al., 2004; Wangt uptake systems are among the most stable siderophore– and Newman, 2008), humic substances (Lovley et al., Fe(III) complexes at low pH (Harrington and Crumbliss, 1999) and quinones (Newman and Kolter, 2000) to shuttle 2009; Hider and Kong, 2010) and therefore would be best electrons from insoluble iron (such as ferrihydrite at suited to the extremely low and restricted pH environment neutral pH) to cellular electron carriers such as multihaem of the Leptospirillum spp. The paucity of Fe(III) uptake cytochromes c and conductive pili (nanowires). systems in the Leptospirillum spp. might also be However, few investigations have explored the possible explained by their occupation of this low-pH environment use of external electron carriers in acidophilic Fe(II) oxi- where high concentrations of soluble Fe(II) would nearly dation. It was suggested that extracellular polymers with always be available, rendering it unnecessary to have an embedded ferric iron molecules could play a role in metal extensive suite of Fe(III) uptake systems as is typically sulfide ores oxidation in biofilms of At. ferrooxidans (Ingle- found in neutrophiles (Andrews et al., 2003). dew, 1986; Sand and Gehrke, 2006). The aporusticyanin In contrast, Acidithiobacillus spp. exhibit a surprisingly of At. ferrooxidans has been claimed to be excreted and large number of predicted iron transporters (Quatrini to potentially shuttle electrons from mineral surfaces to et al., 2005a; Osorio et al., 2008; Valdes et al., 2008) the cell (Ohmura and Blake, 1997); however these data including predicted FeoB-like Fe(II) and Nramp-like Fe(II)- are controversial. Mn(II) transporters and 14 different TonB-dependent ferri-siderophore transporters of diverse siderophore Iron uptake strategies specificity. The latter include siderophores of the citrate, linear and cyclic catecholate and hydroxamate families Introduction compared with just the citrate family of the Leptospirillum A bioinformatic analysis of the conservation, organization spp. (Fig. 3). and distribution of iron homeostasis functions in acido- It was hypothesized that the abundance of different philic microorganisms is beginning to identify the mole- types of siderophore receptors might provide versatility

Fig. 3. Predicted Fe(III) siderophore transporters in Acidithiobacillus spp. and Leptospirillum spp. The Venn diagram shows species-speciﬁc and shared TonB-dependent outer membrane receptors: Fec family = citrate; Cir family = linear catecholate; Fep family = cyclic catecholate; Fhu family = hydroxamate. Figure after Osorio and colleagues (2008).

Fig. 4. Predicted iron acquisition modules in Acidithiobacillus spp. Fe(II) uptake probably occurs primarily through an FeoABCP system, although a secondary MtnH system has also been implicated. Fourteen different siderophore systems for the uptake of Fe(III) have been predicted based on the TonB/ABC outer membrane receptor system, although only one potential siderophore Fe(III) biosynthesis system has been detected that has similarity to the citrate system. A potential iron/phosphonate biosynthesis and uptake system may also be involved in Fe(III) acquisition. Figure after Osorio and colleagues (2008). for growth of Acidithiobacillus spp. in higher-pH environ- spp. may be capable of scavenging siderophores pro- ments with less abundant sources of iron and where iron duced by other microorganisms, inﬂuencing the ecology co-precipitation with phosphates or sulfates during iron of the community. and sulfur biooxidation might compromise their ability to Acidithiobacillus ferrooxidans, and the acidophilic scavenge iron (Osorio et al., 2008). On the other hand, sulfur-oxidizers At. thiooxidans and At. caldus, have Leptospirillum spp. live in more acidic and narrow pH candidate genes for the iron storage protein bacterio- range habitats than Acidithiobacillus spp. and exhibit only ferritin, although its predicted amino acid sequence citrate siderophore uptake systems which are the most in all three cases suggests that it could exhibit differ- stable siderophore–Fe(III) complexes at low pH (Har- ences in its ability to bind haem groups compared with rington and Crumbliss, 2009; Hider and Kong, 2010). the classical bacterioferritins found in neutrophiles Neither the Acidithiobacillus spp. nor the Leptospiril- (Osorio et al., 2008). The Acidithiobacillus spp. may also lum spp. have predicted genes that encode classical sid- store iron in polyphosphate inclusion bodies (Osorio erophore biosynthesis pathways, including an absence et al., 2008). No genes potentially encoding bacteriofer- of pathways for the linear and cyclic catecholates and ritin or polyphosphate inclusion bodies have been hydroxamate siderophores for which the Acidithiobacillus detected in the Leptospirillum spp. Perhaps these spp. have receptors. Instead, they have predicted path- extreme acidophiles, living in an environment with abun- ways for citrate synthesis and possibly for phosphate- dant soluble Fe(II), do not have a requirement to store chelation-mediated iron uptake (see Fig. 4) (Osorio iron. et al., 2008). The presence of the citrate biosynthesis Iron storage proteins have also been implicated as a pathways is consistent with the aforementioned acid way of relieving oxidative stress by serving as a sink of stability of citrate siderophores and the presence of potentially dangerous excess iron. Thus, the apparent cognate receptors in both the Acidithiobacillus spp. and absence of conventional iron storage capabilities in Lep- Leptospirillum spp. The absence of siderophore biosyn- tospirillum spp. might present it with a special challenge thesis pathways for the three other types of siderophore for avoiding oxidative stress. One possibility is that Fe(II) receptor complexes suggests that the Acidithiobacillus oxidation might serve not only as an energy source but

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology 10 V. Bonnefoy and D. S. Holmes also as a way to evade stress by converting the potentially Regulation of iron oxidation in At. ferrooxidans dangerous soluble Fe(II) to the less soluble Fe(III) (Osorio et al., 2008). Acidithiobacillus ferrooxidans is confronted with the challenge of regulating the electron flux derived from Fe(II) oxidation either uphill towards the reduction of NAD+ for Iron homeostasis in Ferroplasma spp. anabolic activities such as CO2 and N2 fixation or down- Ferroplasma acidiphilum, a member of the Euryarchaea, hill towards the reduction of O2 to water to aid in the lives at an extremely acidic pH (around pH 1) with high neutralization of protons that entered the cell via the ATP concentrations of ferrous iron. A proteomic analysis has synthetase complex. It was suggested about a decade shown that it exhibits the unusual property of having a ago that the cellular ATP/ADP ratio regulates the large number of iron containing proteins. It has been balance of reducing equivalents from Fe(II), favouring suggested that the presence of iron aids in the stabili- either the activation of the aa3 cytochrome oxidase and zation of these proteins at the relatively low pH (pH 5.6) thus promote the downhill pathway or, conversely, the of the cytoplasm (Ferrer et al., 2007; 2008). The high repression of the aa3 cytochrome oxidase promoting the iron load per cell could increase the potential for oxida- use of the uphill pathway (Elbehti et al., 2000). In addi- tive damage. A further combined bioinformatics, tran- tion to regulatory decisions regarding the flux of elec- script profiling and proteomic study of iron uptake and trons uphill or downhill, At. ferrooxidans also regulates homeostasis in ‘F. acidarmanus’ Fer1 and F. acidiphilum enzymes and electron carriers depending on whether its YT has been carried out (Potrykus et al., 2011). Potential energetic substrate is Fe(II) or RISCs (Yarzabal et al., importers for Fe(II) via an MtnH-like NRAMP and 2004; Bruscella et al., 2007; Amouric et al., 2009; for Fe(III) via a Fhu-hydroxamate-like family sidero- Quatrini et al., 2009). phore receptor were detected and their genes seem to Regulator circuits potentially involved in making deci- be regulated in an Fe(II)/Fe(III)-dependent manner. The sions about electron flux might therefore exist. Recently, Ferroplasma spp. also exhibit three candidate genes candidate regulator systems have emerged that might be for isochorismatase potentially involved in the produc- involved in these functions. Adjacent to the rus operon tion of siderophores of the enterobactin family that encoding components of the downhill pathway is a six are upregulated under iron-limiting conditions. The gene operon (ctaABTRUS) that contains genes predicted enterobactin-like siderophores have a high affinity for to be involved in the biogenesis of the aa3 cytochrome Fe(III) but are less stable at low pH than the oxidase. The expression of this operon is upregulated in hydroxamate-like siderophores (Hider and Kong, 2010). the presence of Fe(II) as an energy source in a manner The presence of enterobactin-like siderophores is unex- similar to the rus operon (Amouric et al., 2009; Quatrini pected since Ferroplasma spp. lives at very low pH and et al., 2009). CtaR is predicted to encode an iron respon- might be expected to use the more acid stable citrate- sive regulator of the Rrf2 family of transcriptional regula- type siderophores. tors that, in many organisms, contains an [Fe-S] cluster and responds indirectly to iron concentration and oxidative stress as measured by the availability of intracellular Regulation of Fe(II) oxidation and iron homeostasis [Fe-S] clusters (Johnston et al., 2007). CtaR may function as a regulator of its cognate operon as observed in Introduction other organisms. Since it may also be a regulator of the All microorganisms use iron as a cofactor of many rus operon, this suggests a model in which the expres- enzymes involved in essential biological systems and sion of the cta and rus operons could be coordinately they must balance the uptake and storage of this metabo- regulated. lite with its potential for exacerbating oxidative stress Genes potentially encoding a sensor/regulator two- through the Fenton reaction. Microorganisms that also component signal transducing system of the RegB/RegA use Fe(II) as a source of energy and electrons face the family involved in redox sensing in other microorganisms additional problem of balancing iron uptake for metabo- (Wu and Bauer, 2008; Bauer et al., 2009) have also been lism versus its use for energy and electron transfer. Very detected further downstream of the rus operon (Amouric little is known about how this balance is achieved and we et al., 2009; Quatrini et al., 2009). This system has been focus on initial investigations that are beginning to shed shown to measure the redox state of the quinone pool in light on some of the regulatory processes that At. ferrooxi- many microorganisms via a quinone bound to a trans- dans may be using to modulate its use of iron. Hopefully, membrane domain and also by the redox state of a cys- some of this information will subsequently prove to be teine located in the cytoplasmic transmitter domain (Wu relevant for understanding iron function regulation in other and Bauer, 2008; Bauer et al., 2009). Both redox-sensing iron oxidizers. domains are conserved in RegB in At. ferrooxidans

(Quatrini et al., 2009). The regBA gene pair is upregu- several different transcriptional regulators (Osorio et al., lated in the presence of Fe(II) similarly to the rus operon 2008). (Amouric et al., 2009; Quatrini et al., 2009) and could potentially activate/repress other genes involved in Fe(II) Concluding remarks oxidation. Other candidate genes potentially involved in the regu- • Whereas substantial progress has been made in our lation of Fe(II) oxidation are Fur (Lefimil et al., 2009) and understanding of Fe(II) oxidation and iron homeostasis FNR (Osorio et al., 2009), master regulators of iron homeo- in the model acidophile At. ferrooxidans, knowledge of stasis and anaerobiosis respectively in At. ferrooxidans. these processes in other acidophiles lags far behind. The recent advent of sequence information for over 50 genomes and metagenomes of acidophiles will Regulation of iron uptake and homeostasis advance our understanding of the diversity of pathways Genomic and experimental evidence demonstrate that involved in Fe(II) and iron homeostasis and provide At. ferrooxidans has a functional transcriptional regulator insight into their evolution. of the Fur family involved in the regulation of iron respon- There are a number of different pathways for oxidizing sive functions such as Fe(II) and Fe(III) uptake and iron iron that are phylogenetically widespread in Bacteria and storage (Quatrini et al., 2005b; 2007b; Osorio et al., Archaea and it seems plausible that the capacity to 2008). Also, bioinformatic predictions suggest that oxidize iron was invented independently more than once At. thiooxidans and At. caldus have a suite of Fur regu- during evolution perhaps followed by later diversification lated genes similar to those identified in At. ferrooxidans and lateral gene transfer. Major diversification and (Osorio et al., 2008). It is predicted that these three exploitation of new opportunities could have taken place Acidithiobacillus spp. have two other members of the Fur during the ‘great oxygenation event’ about 2.2–2.4 billion family: a haem responsive (Irr-type) regulator responsible years when acidophiles would have been presented with for the control of haem biosynthesis in response to iron novel opportunities to use Fe(II) as an energy source availability and a peroxide responsive (PerR-type) regu- using oxygen as a terminal electron acceptor. However, lator responsible for the control of a variety of basic physi- they would also have been challenged by the significant ological processes in response to peroxide stress (Osorio increase in soluble Fe(III) in their environment, poten- et al., 2008). tially exacerbating problems of oxidative stress. From a combined metagenomic and proteomic studies, ‘L. rubarum’, L. ferriphilum and ‘L. ferrodiazotrophum’ are • Most likely, neither redox nor Fe(II) sensors and their predicted to encode three members of the Fur family: an signals operate in isolation to regulate the expression of iron responsive Fur-type regulator, a peroxide-sensitive Fe(II) oxidation. It is more probable that integrated and PerR-type regulator and a zinc responsive Zur type complex regulatory networks are involved, providing regulator. Gene context analysis supports a role for the optimal responses to changing levels of oxygen and as PerR-like regulator in alkylperoxide stress response in different energy sources become available. It is hoped Leptospirillum sp. group II, where a cytochrome c peroxi- that the availability of sequence information from many dase and a peroxiredoxin of the AhpC/Tsa family are new genomes will provide improved predictive models divergently transcribed. Partial conservation of this not only for At. ferrooxidans but for other Fe(II) oxidizers context also occurs in the Acidithiobacillus spp. suggest- as well. ing a similar function of the PerR-like regulator in this • Although preliminary data suggest that genes involved group (Osorio et al., 2008). in iron metabolism in At. ferrooxidans are regulated by Several of the iron responsive pathways predicted to transcription factors that sense iron and/or the redox be under the regulation of Fur in the Acidithiobacillus state as well as oxygen availability, much remains to be spp. and Leptospirillum spp. exhibit functional redun- done to understand how acidophiles discriminate dancy that could provide an advantage in changing envi- between iron as a micronutrient and as an energy ronmental conditions such as might be found in naturally source and how the coupling of electron flow between + acidic conditions and in industrial copper bioleaching the NAD uphill pathway and the O2 reduction downhill heaps in which differences in iron availability could pathway reduction is regulated. The molecular mecha- arise due to variations in the environmental pH, directly nisms and network connections that underlie these key affecting iron solubility (Osorio et al., 2008). Understand- regulatory decisions are most likely far more complex ing how regulatory factors other than Fur control expres- than the preliminary evidence suggests and a major sion of iron uptake genes in acidophiles is still limited, effort will be required to unravel the suspected dynamic yet the picture is growing increasingly complex with the connections that underlie the proposed versatility of recent findings of superimposed positive regulation by pathway connections.

• The paucity of investigations into the possible use of drainage biofilm communities. Appl Environ Microbiol 75: extracellular electron carriers in acidophilic Fe(II) oxida- 2192–2199. tion represents a substantial lacuna in our knowledge Baker-Austin, C., and Dopson, M. (2007) Life in acid: pH homeostasis in acidophiles. Trends Microbiol 15: 165– and merits further attention. Since many naturally 171. occurring substrates such as pyrite and chalcopyrite are Bandeiras, T.M., Refojo, P.N., Todorovic, S., Murgida, D.H., insoluble, the more widespread use of excreted electron Hildebrandt, P., Bauer, C., et al. (2009) The cytochrome ba carriers as aids in initial oxidation of such substrates complex from the thermoacidophilic crenarchaeote Acidi- might be expected, as has been observed in cases of anus ambivalens is an analog of bc(1) complexes. Biochim neutrophilic iron oxidation and iron reduction. Biophys Acta 1787: 37–45. Barr, D.W., Ingledew, W.J., and Norris, P.R. (1990) Respira- tory chain components of iron-oxidizing, acidophilic bacte- Acknowledgements ria. FEMS Microbiol Lett 70: 85–90. Bathe, S., and Norris, P.R. (2007) Ferrous iron- and sulfur- Fondecyt 1090451, DI-UNAB 34-06, Conicyt Basal CCTE induced genes in Sulfolobus metallicus. Appl Environ PFB16 and an ECOS-Conicyt award. Part of this work was Microbiol 73: 2491–2497. financed by the EU framework 6 project ‘BioMinE’ (N°NM2.ct, Bauer, C., Setterdahl, A., Wu, J., and Robinson, B.R. (2009) 2005.500329), and we thank our various partners on the Regulation of gene expression in response to oxygen project for their contributions to the work reported in this tension. In The Purple Phototrophic Bacteria. Hunter, C.N., article. We thank René Sepúlvida for his help in constructing Daldal, F., Thurnauer, M.C., and Beatty, J.T. (eds). Dor- the phylogenetic tree. The G. capsiferriformans data were drecht, The Netherlands: Springer Science, pp. 707–725. produced by the US Department of Energy Joint Genome Bengrine, A., Guiliani, N., Appia-Ayme, C., Jedlicki, E., Institute (http://www.jgi.doe.gov/) in collaboration with the Holmes, D.S., Chippaux, M., and Bonnefoy, V. (1998) user community. Sequence and expression of the rusticyanin structural gene from Thiobacillus ferrooxidans ATCC33020 strain. References Biochim Biophys Acta 1443: 99–112. Bird, L.J., Bonnefoy, V., and Newman, D.K. (2011) Bioener- Allen, E.E., Tyson, G.W., Whitaker, R.J., Detter, J.C., Rich- getic challenges of microbial iron metabolism. Trends ardson, P.M., and Banfield, J.F. (2007) Genome dynamics Microbiol 19: 330–340. in a natural archaeal population. Proc Natl Acad Sci USA Blake, R.C., 2nd, Shute, E.A., Waskovsky, J., and Harrison, 104: 1883–1888. A.P., Jr (1992) Respiratory components in acidophilic Amaral Zettler, L.A., Gomez, F., Zettler, E., Keenan, B.G., bacteria that respire on iron. Geomicrobiol J 10: 173– Amils, R., and Sogin, M.L. (2002) Microbiology: eukaryotic 192. diversity in Spain’s River of Fire. Nature 417: 137. Blake, R.C., 2nd, Shute, E.A., Greenwood, M.M., Spencer, Amouric, A., Appia-Ayme, C., Yarzabal, A., and Bonnefoy, V. G.H., and Ingledew, W.J. (1993) Enzymes of aerobic res- (2009) Regulation of the iron and sulfur oxidation pathways piration on iron. FEMS Microbiol Rev 11: 9–18. in the acidophilic Acidithiobacillus ferrooxidans. Adv Mat Bonnefoy, V. (2010) Bioinformatics and genomics of iron- and Res 71–73: 163–166. sulfur-oxidizing acidophiles. In Geomicrobiology: Molecular Amouric, A., Brochier-Armanet, C., Johnson, D.B., Bonnefoy, and Environmental Perspective. Barton, L.L., Mandl, M., V., and Hallberg, K.B. (2011) Phylogenetic and genetic and Loy, A. (eds). Dordrecht, Heidelberg, London, New variation among Fe(II)-oxidizing acidithiobacilli supports York: Springer, pp. 169–192. the view that these comprise multiple species with different Bruscella, P., Appia-Ayme, C., Levican, G., Ratouchniak, J., ferrous iron oxidation pathways. Microbiology 157: 111– Jedlicki, E., Holmes, D.S., and Bonnefoy, V. (2007) Differ- 122. ential expression of two bc(1) complexes in the strict aci- Andrews, S.C., Robinson, A.K., and Rodriguez-Quinones, F. dophilic chemolithoautotrophic bacterium Acidithiobacillus (2003) Bacterial iron homeostasis. FEMS Microbiol Rev ferrooxidans suggests a model for their respective roles in 27: 215–237. iron or sulfur oxidation. Microbiology 153: 102–110. Auernik, K.S., and Kelly, R.M. (2008) Identification of Cardenas, J.P., Valdes, J., Quatrini, R., Duarte, F., and components of electron transport chains in the extremely Holmes, D.S. (2010) Lessons from the genomes of thermoacidophilic crenarchaeon Metallosphaera sedula extremely acidophilic bacteria and archaea with special through iron and sulfur compound oxidation transcrip- emphasis on bioleaching microorganisms. Appl Microbiol tomes. Appl Environ Microbiol 74: 7723–7732. Biotechnol 88: 605–620. Auernik, K.S., Maezato, Y., Blum, P.H., and Kelly, R.M. (2008) Castelle, C., Ilbert, M., Infossi, P., Leroy, G., and Giudici- The genome sequence of the metal-mobilizing, extremely Orticoni, M.T. (2010) An unconventional copper protein thermoacidophilic archaeon Metallosphaera sedula pro- required for cytochrome c oxidase respiratory function vides insights into bioleaching-associated metabolism. under extreme acidic conditions. J Biol Chem 285: 21519– Appl Environ Microbiol 74: 682–692. 21525. Baker, B.J., and Banfield, J.F. (2003) Microbial communities Cavazza, C., Guigliarelli, B., Bertrand, P., and Bruschi, M. in acid mine drainage. FEMS Microbiol Ecol 44: 139–152. (1995) Biochemical and EPR characterization of a high Baker, B.J., Tyson, G.W., Goosherst, L., and Banfield, J.F. potential iron-sulfur protein in Thiobacillus ferrooxidans. (2009) Insights into the diversity of eukaryotes in acid mine FEMS Microbiol Lett 130: 193–200.

Cid, C., Garcia-Descalzo, L., Casado-Lafuente, V., Amils, R., Fukumori, Y., Yano, T., Sato, A., and Yamanaka, T. (1988) and Aguilera, A. (2010) Proteomic analysis of the response Fe(II) oxidizing enzyme purified from Thiobacillus ferrooxi- of an acidophilic strain of Chlamydomonas sp. (Chloro- dans. FEMS Microbiol Lett 50: 169–172. phyta) to natural metal-rich water. Proteomics 10: 2026– Goltsman, D.S., Denef, V.J., Singer, S.W., VerBerkmoes, 2036. N.C., Lefsrud, M., Mueller, R.S., et al. (2009) Community Croal, L.R., Jiao, Y., and Newman, D.K. (2007) The fox genomic and proteomic analyses of chemoautotrophic operon from Rhodobacter strain SW2 promotes pho- iron-oxidizing ‘Leptospirillum rubarum’ (Group II) and ‘Lep- totrophic Fe(II) oxidation in Rhodobacter capsulatus tospirillum ferrodiazotrophum’ (Group III) bacteria in acid SB1003. J Bacteriol 189: 1774–1782. mine drainage biofilms. Appl Environ Microbiol 75: 4599– Demergasso, C., Galleguillos, F., Soto, P., Seron, M., and 4615. Iturriaga, V. (2010) Microbial succession during a heap Gonzalez-Toril, E., Llobet-Brossa, E., Casamayor, E.O., bioleaching cycle of low grade copper sulfides. Does this Amann, R., and Amils, R. (2003) Microbial ecology of an knowledge mean a real input for industrial process design extreme acidic environment, the Tinto River. Appl Environ and control? Hydrometallurgy 104: 382–390. Microbiol 69: 4853–4865. Denef, V.J., Mueller, R.S., and Banfield, J.F. (2010) AMD Gralnick, J.A., and Newman, D.K. (2007) Extracellular respi- biofilms: using model communities to study microbial evo- ration. Mol Microbiol 65: 1–11. lution and ecological complexity in nature. ISME J 4: 599– Hallberg, K.B., and Johnson, D.B. (2001) Biodiversity of aci- 610. dophilic prokaryotes. Adv Appl Microbiol 49: 37–84. Dopson, M. (2010) Physiological adaptations and biotechni- Hallberg, K.B., Amouric, A., Brochier-Armanet, C., Bonnefoy, cal applications of acidophiles. In Extremophiles: Microbi- V., and Johnson, D.B. (2009) Physiological and phyloge- ology and Biotechnology. Anitori, R. (ed.). Norwich, UK: netic heterogeneity among iron-oxidizing Acidithiobacillus Caister Academic Press, pp. 265–294. spp., and characteristics of the novel species Acidithioba- Dopson, M., Lindstrom, E.B., and Hallberg, K.B. (2002) ATP cillus ferrivorans. Adv Mat Res 71–73: 167–170. generation during reduced inorganic sulfur compound Hallberg, K.B., Gonzalez-Toril, E., and Johnson, D.B. (2010) oxidation by Acidithiobacillus caldus is exclusively due to Acidithiobacillus ferrivorans, sp. nov.; facultatively anaero- electron transport phosphorylation. Extremophiles 6: 123– bic, psychrotolerant iron-, and sulfur-oxidizing acidophiles 129. isolated from metal mine-impacted environments. Extre- Dopson, M., Baker-Austin, C., and Bond, P.L. (2005) Analysis mophiles 14: 9–19. of differential protein expression during growth states of Harrington, J.M., and Crumbliss, A.L. (2009) The redox Ferroplasma strains and insights into electron transport for hypothesis in siderophore-mediated iron uptake. Biometals iron oxidation. Microbiology 151: 4127–4137. 22: 679–689. Dopson, M., Halinen, A.K., Rahunen, N., Ozkaya, B., Harrison, A.P., Jr (1982) Genomic and physiological diversity Sahinkaya, E., Kaksonen, A.H., et al. (2007) Mineral and amongst strains of Thiobacillus ferrooxidans, and genomic iron oxidation at low temperatures by pure and mixed cul- comparison with Thiobacillus thiooxidans. Arch Microbiol tures of acidophilic microorganisms. Biotechnol Bioeng 97: 131: 68–76. 1205–1215. Hartshorne, R.S., Reardon, C.L., Ross, D., Nuester, J., Elbehti, A., Brasseur, G., and Lemesle-Meunier, D. (2000) Clarke, T.A., Gates, A.J., et al. (2009) Characterization of First evidence for existence of an uphill electron transfer an electron conduit between bacteria and the extracellular through the bc(1) and NADH-Q oxidoreductase complexes environment. Proc Natl Acad Sci USA 106: 22169–22174. of the acidophilic obligate chemolithotrophic ferrous ion- Hedrich, S., Schlomann, M., and Johnson, D.B. (2011) The oxidizing bacterium Thiobacillus ferrooxidans. J Bacteriol iron-oxidizing proteobacteria. Microbiology 157: 1551– 182: 3602–3606. 1664. Emerson, D., Fleming, E.J., and McBeth, J.M. (2010) Iron- Hernandez, M.E., Kappler, A., and Newman, D.K. (2004) oxidizing bacteria: an environmental and genomic perspec- Phenazines and other redox-active antibiotics promote tive. Annu Rev Microbiol 64: 561–583. microbial mineral reduction. Appl Environ Microbiol 70: Ferguson, S.J., and Ingledew, W.J. (2008) Energetic prob- 921–928. lems faced by micro-organisms growing or surviving on Hider, R.C., and Kong, X. (2010) Chemistry and biology of parsimonious energy sources and at acidic pH: I. Acidithio- siderophores. Nat Prod Rep 27: 637–657. bacillus ferrooxidans as a paradigm. Biochim Biophys Acta Holmes, D., and Bonnefoy, V. (2007) Genetic and bioinfor- 1777: 1471–1479. matic insights into iron and sulfur oxidation mechanisms of Ferrer, M., Golyshina, O.V., Beloqui, A., Golyshin, P.N., and bioleaching organisms. In Biomining. Rawlings, D.E., and Timmis, K.N. (2007) The cellular machinery of Ferroplasma Johnson, D.B. (eds). Berlin Heidelberg: Springer-Verlag, acidiphilum is iron-protein-dominated. Nature 445: 91–94. pp. 281–307. Ferrer, M., Golyshina, O.V., Beloqui, A., Bottger, L.H., Ida, C., Sasaki, K., Ando, K., Blake, R.C., 2nd, Saiki, H., and Andreu, J.M., Polaina, J., et al. (2008) A purple acidophilic Ohmura, N. (2003) Kinetic rate constant for electron trans- di-ferric DNA ligase from Ferroplasma. Proc Natl Acad Sci fer between ferrous ions and novel rusticyanin isoform in USA 105: 8878–8883. Acidithiobacillus ferrooxidans. J Biosci Bioeng 95: 534– Franke, S., and Rensing, C. (2007) Acidophiles: mechanisms 537. to tolerate metal and acid toxicity. In Physiology and Bio- Ingledew, W.J. (1982) Thiobacillus ferrooxidans. The bioen- chemistry of Extremophiles. Gerday, C., and Glansdorff, N. ergetics of an acidophilic chemolithotroph. Biochim (eds). Washington, DC, USA: ASM press, pp. 271–277. Biophys Acta 683: 89–117.

Ingledew, W.J. (1986) Ferrous iron oxidation by Thiobacillus Lo, I., Denef, V.J., Verberkmoes, N.C., Shah, M.B., Goltsman, ferrooxidans. Biotechnol Bioeng Symp 16: 23–33. D., DiBartolo, G., et al. (2007) Strain-resolved community Jerez, C.A. (2007) Proteomics and metaproteomics applied proteomics reveals recombining genomes of acidophilic to biomining microorganisms. In Microbial Processing of bacteria. Nature 446: 537–541. Metal Sulfides. Donati, E.R., and Sand, W. (eds). Dor- Lopez-Archilla, A.I., Marin, I., and Amils, R. (2001) Microbial drecht, the Netherlands: Springer, pp. 241–251. community composition and ecology of an acidic aquatic Jerez, C.A. (2008) The use of genomics, proteomics and environment: the Tinto river, Spain. Microb Ecol 41: 20–35. other OMICS technologies for the global understanding Lovley, D.R., Fraga, J.L., Coates, J.D., and Blunt-Harris, E.L. of biomining microorganisms. Hydrometallurgy 94: 162– (1999) Humics as an electron donor for anaerobic respira- 169. tion. Environ Microbiol 1: 89–98. Jiao, Y., and Newman, D.K. (2007) The pio operon is essen- Nancucheo, I., and Johnson, D.B. (2010) Production of gly- tial for phototrophic Fe(II) oxidation in Rhodopseudomonas colic acid by chemolithotrophic iron- and sulfur-oxidizing palustris TIE-1. J Bacteriol 189: 1765–1773. bacteria and its role in delineating and sustaining acido- Johnson, D.B. (1998) Biodiversity and ecology of acidophilic philic sulfide mineral-oxidizing consortia. Appl Environ microorganisms. FEMS Microbiol Ecol 27: 307–317. Microbiol 76: 461–467. Johnson, D.B. (2007) Physiology and ecology of acidophilic Newman, D.K., and Kolter, R. (2000) A role for excreted microorganisms. In Physiology and Biochemistry of Extre- quinones in extracellular electron transfer. Nature 405: mophiles. Gerday, C., and Glansdorff, N. (eds). Washing- 94–97. ton, DC, USA: ASM press, pp. 257–269. Nicolle, J.L.C., Simmons, S., Bathe, S., and Norris, P.R. Johnson, D.B. (2008) Biodiversity and interactions of acido- (2009) Ferrous iron oxidation and rusticyanin in halotoler- philes: key to understanding and optimizing microbial pro- ant, acidophilic ‘Thiobacillus prosperus’. Microbiology 155: cessing of ores and concentrates. T Nonferr Metal Soc 18: 1302–1309. 1367–1373. Norris, P.R. (2007) Acidophile diversity in mineral sulfide oxi- Johnson, D.B. (2009) Extremophiles: Acidic Environments. In dation. In Biomining. Rawlings, D.E., and Johnson, D.B. Encyclopaedia of Microbiology. Schaechter, M. (ed.). (eds). Berlin Heidelberg: Springer-Verlag, pp. 199–216. Oxford: Elsevier, pp. 107–126. Ohmura, N., and Blake, R.C. (1997) Aporusticyanin mediates Johnson, D.B., and Hallberg, K.B. (2003) The microbiology of the adhesion of Thiobacillus ferrooxidans to pyrite. In IBS acidic mine waters. Res Microbiol 154: 466–473. Biomine. Foundation, A.M. (ed.). Sydney, Australia: Aus- Johnson, D.B., and Hallberg, K.B. (2008) Carbon, iron and tralian Mineral Foundation, pp. PB1.1–PB1.10. sulfur metabolism in acidophilic micro-organisms. Adv Osorio, H., Martinez, V., Nieto, P.A., Holmes, D.S., and Qua- Microb Physiol 54: 201–255. trini, R. (2008) Microbial iron management mechanisms in Johnson, D.B., Rolfe, S., Hallberg, K.B., and Iversen, E. extremely acidic environments: comparative genomics (2001) Isolation and phylogenetic characterization of aci- evidence for diversity and versatility. BMC Microbiol 8: dophilic microorganisms indigenous to acidic drainage 203. waters at an abandoned Norwegian copper mine. Environ Osorio, H., Cárdenas, J.P., Valdés, J., and Holmes, D.S. Microbiol 3: 630–637. (2009) Prediction of Fnr regulated genes and metabolic Johnston, A.W., Todd, J.D., Curson, A.R., Lei, S., Nikolaidou- pathways potentially involved in anaerobic growth of Katsaridou, N., Gelfand, M.S., and Rodionov, D.A. (2007) Acidithiobacillus ferrooxidans. Adv Mat Res 71–73: 195– Living without Fur: the subtlety and complexity of iron- 198. responsive gene regulation in the symbiotic bacterium Parro, V., Moreno-Paz, M., and Gonzalez-Toril, E. (2007) Rhizobium and other alpha-proteobacteria. Biometals 20: Analysis of environmental transcriptomes by DNA microar- 501–511. rays. Environ Microbiol 9: 453–464. Kumar, S., Nei, M., Dudley, J., and Tamura, K. (2008) MEGA: Potrykus, J., Jonna, V.R., and Dopson, M. (2011) Iron homeo- a biologist-centric software for evolutionary analysis of stasis and responses to iron limitation in extreme acido- DNA and protein sequences. Brief Bioinform 9: 299– philes from the Ferroplasma genus. Proteomics 11: 52–63. 306. Quatrini, R., Jedlicki, E., and Holmes, D.S. (2005a) Genomic Kusano, T., Takeshima, T., Sugawara, K., Inoue, C., Shiratori, insights into the iron uptake mechanisms of the biomining T., Yano, T., et al. (1992) Molecular cloning of the gene microorganism Acidithiobacillus ferrooxidans. J Ind Micro- encoding Thiobacillus ferrooxidans Fe(II) oxidase. High biol Biotechnol 32: 606–614. homology of the gene product with HiPIP. J Biol Chem 267: Quatrini, R., Lefimil, C., Holmes, D.S., and Jedlicki, E. 11242–11247. (2005b) The ferric iron uptake regulator (Fur) from the Lefimil, C., Osorio, H., Quatrini, R., Holmes, D., and Jedlicki, extreme acidophile Acidithiobacillus ferrooxidans. Microbi- E. (2009) Regulation of expression of the petI operon ology 151: 2005–2015. involved in iron oxidation in the biomining bacterium Quatrini, R., Valdes, J., Jedlicki, E., and Holmes, D.S. Acidithiobacillus ferrooxidans. Adv Mat Res 71–73: 199– (2007a) The use of bioinformatics and genomic biology to 202. advance our understanding of bioleaching microorgan- Liljeqvist, M., Valdes, J., Holmes, D.S., and Dopson, M. isms. In Microbial Processing of Metal Sulfides. Donati, (2011) Draft genome of the psychrotolerant acidophile E.R., and Sand, W. (eds). Dordrecht, the Netherlands: Acidithiobacillus ferrivorans SS3 provides insights into Springer, pp. 221–239. growth at low temperature by the Acidithiobacilli. Quatrini, R., Lefimil, C., Veloso, F.A., Pedroso, I., Holmes, J Bacteriol 193: 4304–4305. D.S., and Jedlicki, E. (2007b) Bioinformatic prediction and

experimental verification of Fur-regulated genes in the Siezen, R.J., and Wilson, G. (2009) Bioleaching genomics. extreme acidophile Acidithiobacillus ferrooxidans. Nucleic Microb Biotechnol 2: 297–303. Acids Res 35: 2153–2166. Simmons, S.L., Dibartolo, G., Denef, V.J., Goltsman, D.S., Quatrini, R., Appia-Ayme, C., Denis, Y., Jedlicki, E., Holmes, Thelen, M.P., and Banfield, J.F. (2008) Population genomic D.S., and Bonnefoy, V. (2009) Extending the models for analysis of strain variation in Leptospirillum group II bacte- iron and sulfur oxidation in the extreme acidophile ria involved in acid mine drainage formation. PLoS Biol 6: Acidithiobacillus ferrooxidans. BMC Genomics 10: 394. e177. Ram, R.J., Verberkmoes, N.C., Thelen, M.P., Tyson, G.W., Singer, S.W., Erickson, B.K., VerBerkmoes, N.C., Hwang, M., Baker, B.J., Blake, R.C., 2nd, et al. (2005) Community Shah, M.B., Hettich, R.L., et al. (2010) Posttranslational proteomics of a natural microbial biofilm. Science 308: modification and sequence variation of redox-active pro- 1915–1920. teins correlate with biofilm life cycle in natural microbial Rawlings, D.E. (2002) Heavy metal mining using microbes. communities. ISME J 4: 1398–1409. Annu Rev Microbiol 56: 65–91. Touati, D. (2000) Iron and oxidative stress in bacteria. Arch Rawlings, D.E. (2005) Characteristics and adaptability of Biochem Biophys 373: 1–6. iron- and sulfur-oxidizing microorganisms used for the Tyson, G.W., Chapman, J., Hugenholtz, P., Allen, E.E., Ram, recovery of metals from minerals and their concentrates. R.J., Richardson, P.M., et al. (2004) Community structure Microb Cell Fact 4: 13. and metabolism through reconstruction of microbial Rawlings, D.E. (2007) Relevance of cell physiology and genomes from the environment. Nature 428: 37–43. genetic adaptability of biomining microorganisms to indus- Valdes, J., Pedroso, I., Quatrini, R., Dodson, R.J., Tettelin, H., trial processes. In Biomining. Rawlings, D.E., and Johnson, Blake, R., 2nd, et al. (2008) Acidithiobacillus ferrooxidans D.B. (eds). Berlin Heidelberg: Springer-Verlag, pp. 177– metabolism: from genome sequence to industrial applica- 198. tions. BMC Genomics 9: 597. Rawlings, D.E., and Johnson, D.B. (2007) The microbiology Valenzuela, L., Chi, A., Beard, S., Orell, A., Guiliani, N., of biomining: development and optimization of mineral- Shabanowitz, J., et al. (2006) Genomics, metagenomics oxidizing microbial consortia. Microbiology 153: 315– and proteomics in biomining microorganisms. Biotechnol 324. Adv 24: 197–211. Sand, W., and Gehrke, T. (2006) Extracellular polymeric sub- Wangt, Y., and Newman, D.K. (2008) Redox reactions of stances mediate bioleaching/biocorrosion via interfacial phenazine antibiotics with ferric (hydr)oxides and molecu- processes involving iron(III) ions and acidophilic bacteria. lar oxygen. Environ Sci Technol 42: 2380–2386. Res Microbiol 157: 49–56. Weber, K.A., Achenbach, L.A., and Coates, J.D. (2006) Sasaki, K., Ida, C., Ando, A., Matsumoto, N., Saiki, H., and Microorganisms pumping iron: anaerobic microbial iron Ohmura, N. (2003) Respiratory isozyme, two types of rus- oxidation and reduction. Nat Rev Microbiol 4: 752–764. ticyanin of Acidithiobacillus ferrooxidans. Biosci Biotechnol Williams, K.P., Gillespie, J.J., Sobral, B.W., Nordberg, E.K., Biochem 67: 1039–1047. Snyder, E.E., Shallom, J.M., and Dickerman, A.W. (2010) Schippers, A. (2007) Microorganisms involved in bioleaching Phylogeny of gammaproteobacteria. J Bacteriol 192: and nucleic acid-based molecular methods for their identi- 2305–2314. fication and quantification. In Microbial Processing of Metal Wu, J., and Bauer, C.E. (2008) RegB/RegA, a global redox- Sulfides. Donati, E.R., and Sand, W. (eds). Dordrecht, the responding two-component system. Adv Exp Med Biol Netherlands: Springer, pp. 3–33. 631: 131–148. Schippers, A., Breuker, A., Blazejak, A., Bosecker, K., Kock, Yarzabal, A., Appia-Ayme, C., Ratouchniak, J., and D., and Wright, T.L. (2010) The biochemistry and microbi- Bonnefoy, V. (2004) Regulation of the expression of the ology of sulfidic mine waste and bioleaching dumps and Acidithiobacillus ferrooxidans rus operon encoding two heaps, and novel Fe(II)-oxidizing bacteria. Hydrometal- cytochromes c, a cytochrome oxidase and rusticyanin. lurgy 104: 342–350. Microbiology 150: 2113–2123.