Acidithiobacillus ferrooxidans is an acidophilic chemolithoautotrophic Gram-negative bacterium that can derive energy from the oxidation of ferrous iron at pH 2 using oxygen as electron acceptor. The study of this bacterium has economic and fundamental biological interest because of its use in the industrial extraction of copper and uranium from ores. For this reason, its respiratory chain has been analysed in detail in recent years. Studies have shown the presence of a functional supercomplex that spans the outer and the inner membranes and allows a direct electron transfer from the extracellular Fe2+ ions to the inner membrane cytochrome c oxidase. Iron induces the expression of two operons encoding proteins implicated in this complex as well as in the regeneration of the reducing power. Most of these are metalloproteins that have been characterized biochemically, structurally and biophysically. For some of them, the molecular basis of their adaptation to the periplasmic acidic environment has been described. Modifications in the metal surroundings have been highlighted for cytochrome c and rusticyanin, whereas, for the cytochrome c oxidase, an additional partner that maintains its stability and activity has been demonstrated recently.
Micro-organisms are found in various ecosystems as they are adapted to their surroundings, in some of which life is sometimes hardly considered possible. Studying these organisms is challenging for researchers on account of the difficulty of reproducing their natural environments in the laboratory. Discovery of new concepts is the driving force that motivates researchers in this field beyond the difficulties linked to the unfriendly environments. In one of these environments, which combines acid and iron which are known as stress factors for ‘classic’ organisms, iron-oxidizing organisms have been discovered that drive their energy using ferrous iron (Fe2+) as electron donor. In the present mini-review, we first describe how life is possible in the presence of acid and iron, and we focus on one of these peculiar organisms, Acidithiobacillus ferrooxidans. This bacterium represents one of the most studied model organisms to understand life in acidic environments, adaptation to toxic metals and use of minerals containing iron as an energy source. As early as 1982, Ingledew  described A. ferrooxidans as “one of Nature's most unusual bioenergetic phenomena” owing to several challenges that it has to face. Various models of its electron transport pathways have been proposed and reviewed [2–6]. In the present paper, we summarize recent work on the iron oxidation pathway. We emphasize some critical points concerning (i) the organization of this pathway , and (ii) the adaptation to the low pH of periplasmic proteins involved in it.
Iron and acid: a challenge for living organisms
Acidophile organisms, the most suited for iron oxidation
Iron oxidizers have first to deal with the chemistry of iron. Spontaneous chemical oxidation of iron can be rapid at neutral pH under aerobic conditions, and oxidized iron precipitates rapidly as ferric hydroxides. Consequently, aerobic neutrophilic organisms able to oxidize iron (such as Mariprofundus ferrooxydans) are rare and are usually present under micro-oxic conditions to allow them to compete with the spontaneous oxidation of Fe2+. In acidic environments, however, Fe2+ is rather stable and available even in the presence of atmospheric oxygen. In these environments, several micro-organisms using Fe2+ as an energy source have been isolated (such as the bacterium A. ferrooxidans or the archaeon Ferroplasma acidarmanus).
High redox potential of the Fe2+/Fe3+ couple: a low energetic substrate
Owing to the high redox potential of the Fe2+/Fe3+ couple (+0.77 V at pH 2), iron can be used as electron donor only in presence of an electron acceptor of higher redox potential, such as the O2/H2O couple (+1.12 V at pH 2). On the basis of these values and the small amount of energy thus available from the oxidative reaction of Fe2+, Ingledew  commented that “growth on Fe2+ represents one of the narrowest thermodynamic limits for which growth is known to occur” and he proposed that 22.4 molecules of Fe2+ are required for the fixation of one molecule of CO2.
The electron acceptor for iron is not only the O2/H2O couple, but also Photosystem I (+0.45 V) and the NO3−/NO2− couple (+0.43 V). They can be used under anoxic and neutrophilic conditions because the redox potential of iron decreases to approximately 0.385 V [4,8]. Rhodobacter capsulatus (a phototroph) and Dechloromonas agitata (a nitrate reducer) are examples of known neutrophilic iron oxidizers. Even though the study of these organisms is in its infancy, they are of great interest, as they could help to decipher the evolution of iron oxidation. It was proposed that anoxygenic nitrate-dependent Fe2+ oxidation might have evolved in the primordial deep-sea water .
How to survive in acidic environments
Extremely acidic environments are also very challenging for life, to the point that acidophilic organisms have maintained their cytoplasm at neutral pH to prevent acid protein destabilization. Several strategies have been developed by acidophiles to maintain an internal pH of the cell close to neutrality  creating a natural PMF (protonmotive force) across the membrane, which is used, for example, by ATP synthase to produce ATP. Protons that have entered the cell have to be exported. Among other processes, active proton pumping seems to be important (four Na+/H+ antiporters and two proton P-type ATPases have been predicted) . To inhibit proton influx, the generation of a reverse membrane potential (Δψ) has been proposed to create a chemiosmotic barrier .
The periplasmic compartment of Gram-negative organisms, however, is close to pH 3, and periplasmic proteins had to adapt to function under such conditions. In the following sections, we first describe A. ferrooxidans, followed by characterization of its periplasmic iron-oxidation pathway, thoroughly described in recent years, illustrating how the pathway, as well as the periplasmic proteins involved in it, function under such extreme conditions (iron and acid).
A. ferrooxidans: main characteristics
A. ferrooxidans, for which the genome has been sequenced , is a Gram-negative rod-shaped bacterium that is commonly found in deep caves or acid mine drainage [12,13]. Long considered to be a member of the Gammaproteobacteria, Williams et al.  proposed that A. ferrooxidans arose before the split between Gammaproteobacteria and Betaproteobacteria. It is strictly acidophilic and grows at an optimal pH of 1.5–2.5. It is an obligate chemolithoautotroph that uses for growth the energy produced by the oxidation of reduced sulfur compounds and/or ferrous iron, using O2 as oxidant, and it fixes atmospheric CO2 as a carbon source. In its natural habitats, it derives Fe2+ from the insoluble pyrite (Figure 1). This capacity renders A. ferrooxidans the subject of considerable interest because of its use in industrial bioleaching to facilitate the extraction of precious metals such as copper and uranium from low-grade ores. Consequently, the understanding of A. ferrooxidans metabolism not only has a fundamental importance, but also has an impact on biotechnological processes.
Pyrite oxidation by micro-organisms
Iron-oxidation pathway: an unconventional respiratory pathway in A. ferrooxidans
Organization and function of the electron-transport pathways
The iron-respiratory chain of A. ferrooxidans needed some unusual characteristics owing to the non-conventional properties enumerated above. For several years, scientists have used distinct approaches, such as biochemistry, molecular genetics, bioenergetics, bioinformatics, proteomics and functional genomics, to understand how A. ferrooxidans bypasses all of the locks described above. They were able (i) to identify, characterize and test the functionality of the proteins involved in the respiratory chain that couples the oxidation of iron to the reduction of oxygen, as well as (ii) to understand how reducing power is produced. It has been shown that electrons from Fe2+ can either be transported along a ‘downhill’ or a ‘uphill’ (or reverse) electron pathway [15–17]. The majority of the electrons (estimated at 90%) are transported through the downhill pathway (thermodynamically favourable) to allow ATP synthesis. The remaining 10% of the electrons are transported through the uphill pathway (thermodynamically unfavourable in the standard state; the redox potential of the NAD+/NADH couple is approximately +0.32 V), to allow the production of reducing power (NAD) for anabolic activities such as CO2 and N2 fixation. The energy to push electrons against this thermodynamically unfavourable gradient is postulated to come from the natural PMF.
Uphill and downhill pathways have been proposed to be connected and to be regulated to balance ATP production with the reconstitution of reducing power. The cellular ATP/ADP ratio would regulate this balance . Potential regulators induced by Fe2+ have been identified in the genome sequence (CtaR and RegAB). These regulators could adjust the expression of proteins encoding the two electron pathways . Development of ‘-omics’ studies [3,18] have demonstrated that two operons are critical for iron oxidation: the petI and rus operons [19–21] (Figure 2). Both encode metalloproteins that had been proposed to belong to A. ferrooxidans electron-transport chains on the basis of bioinformatics, protein subcellular localization, spectroscopic properties or redox potentials characterization. The spatial organization of the proteins involved in the downhill pathway has been determined  (Figure 3). A supramolecular organization of the respiratory chain involving various metalloproteins that spans the outer and inner membranes has also been demonstrated . After purification under mild conditions, this supercomplex is functional, as it has iron oxidase and oxygen reductase activity. Proteins present in this complex have been identified and are mainly encoded by the rus operon. Most of them have been characterized individually, and some functional interactions have been demonstrated. First, the primary electron acceptor, the cytochrome Cyc2, is an outer-membrane monohaem c-type cytochrome able to catalyse iron oxidation [7,22]. Its high redox potential, unusual for a c-type cytochrome, matches the high redox potential of the entire pathway (Figure 2). Electrons are then transferred to the periplasmic rusticyanin, a blue cupredoxin , and to the dihaem cytochrome of the c4-type (Cyc1) [24,25] (Figures 2 and 3). It has been demonstrated that the direct interaction between these two proteins decreases the rusticyanin redox potential by more than 100 mV, facilitating electron transfer [26,27]. Next, Cyc1 transfers electrons to the terminal electron acceptor, the cytochrome c oxidase  via an essential tyrosine residue .
Metalloproteins and operons involved in ferrous iron oxidation in A. ferrooxidans
Model of the iron-oxidation pathway in A. ferrooxidans
Rusticyanin and cytochrome c can individually account for approximately 5–10% of the total cell protein. Owing to the low-energy substrate, they might constitute an important electron reservoir in A. ferrooxidans .
In the purified supercomplex, a substoichiometric amount of the bc1 complex (involved in the uphill pathway) has been identified, providing some evidence for a possible physical association of proteins involved in the downhill and uphill pathways. The genes encoding the bc1 complex, as well as a cytochrome of the c4-type (Cyc42), are encoded by the petI operon (Figure 2). It has been proposed that the branching point between the downhill pathway and the uphill pathway is at the level of the rusticyanin [7,30] (Figure 3).
After purification of the supercomplex, Omp40 has also been detected (Figure 3). A role for this protein in the interaction between the bacterium and the substrate has been proposed . However, the existence of excreted electron carriers, as found in neutrophilic iron-reducing organisms, cannot be ruled out and merits further attention. What is known is that the primary attachment of A. ferrooxidans to pyrite is mediated by exopolymer complexed with iron in an electrostatic interaction with the negatively charged pyrite surface [29,32].
This vertical spatial organization for a respiratory chain had never been shown before and confirms the fact that new concepts can be discovered by working on different model organisms with different constraints. Different hypotheses to explain such complex formation have been advanced: it could help to overcome the low energy of the substrate (optimization of the electron transfer) and/or to prevent intracellular accumulation of a substrate prone to precipitation (extracellular oxidation of the substrate) and/or to protect against the acidic periplasm (a stabilization of the system).
How periplasmic proteins adapt to low pH
In neutrophilic organisms such as enterobacteria, small changes in the environmental pH have a strong impact on the periplasmic proteins, as the outer membrane is permeable to protons. A chaperone protein, HdeA, from Escherichia coli prevents the aggregation of periplasmic proteins upon acid stress . The existence of such protection illustrates the vulnerability of periplasmic proteins to face a pH change. Periplasmic proteins from acidophile organisms had to adapt to this ‘permanent-stress’ condition. Proteomic analysis of 131 periplasmic proteins from A. ferrooxidans have shown that 70% are basic with a pI value of >7 , in contrast with neutrophiles for which the pI of proteins is more evenly distributed (40% are basic with a pI value of >7 in E. coli). This result illustrates a general adaptation of the periplasmic proteome of A. ferrooxidans.
The periplasmic metalloproteins described above have common characteristics related to the unconventional acid/iron pathway: a high redox potential (above 0.3 V), a basic isoelectric focusing point and high stability to pH variation. However, different strategies seem to have evolved to maintain protein activity as well as stability at low pH, especially for metalloproteins for which metal binding can be affected by pH. As for hyperthermophilic proteins, no general rules can be established, as the intrinsic stability can be enhanced by local structural changes. The structures of rusticyanin and of Cyc1 of A. ferrooxidans have been resolved [34–36] and compared with their neutrophilic counterparts to understand their high redox potentials and their pronounced acid stability. Aspartate and glutamate residues reported for c4-type cytochromes from neutrophilic species are absent in A. ferrooxidans proteins as they are protonated at pH 2, preventing the formation of salt bridges involved in structural stabilization in neutrophiles. An extensive internal hydrogen- and hydrophobic bonding network surrounding the haems has been proposed to be the stabilizing factor that could explain its acid stability [24,35].
For the rusticyanin, intrinsic structural changes in the immediate vicinity of the copper may explain its acid stability [34,36]. Most type I copper proteins are unstable under acidic conditions because of the protonation of one or both of the histidine ligands making copper fairly labile. An explanation for the acid stability of the copper site of the rusticyanin is that its histidine ligands are more constrained due to the close proximity of a number of hydrophobic groups (isoleucine and phenylalanine). A high density of hydrophobic groups, as well as the presence of aromatic and proline rings and reduction in the number of pH-labile charge residues in the immediate vicinity of the copper site, would explain its acid stability.
Another metalloprotein of the iron-respiratory chain is exposed to the low pH: the cytochrome c oxidase, a four-subunit complex (CoxA, CoxB, CoxC and CoxD) . CoxB receives the electrons from the cytochrome Cyc1 and transfers them via its dinuclear copper centre (CuA) to the catalytic subunit CoxA, containing two haem centres and a CuB centre where the reduction of water occurs. In contrast with CoxA, CoxB is exposed to the acidic periplasm, raising the question of the stability of its copper centre. On the basis of structural modelling of the CoxB subunit, no modification in the surroundings of CuA could be observed, in contrast with the rusticyanin and the cytochrome Cyc1. However, a protein of unknown function (AcoP) was found in tight interaction with the cytochrome c oxidase . An alternative strategy to protect the metal centre from acid stress via the presence of a supplementary protein (AcoP, described as ‘chaperone-like’) seems thus to have evolved for the cytochrome c oxidase . Indeed, experimental data show that AcoP maintains a stable and active conformation of CoxB and consequently an optimal cytochrome c oxidase activity in extreme acidic physiological conditions.
Even though the molecular basis of acidophilic adaptation remains obscure, mutative changes of amino acid sequences as well as extrinsic factors may be part of the extremophilic adaptation.
The studies of the iron oxidation pathway in A. ferrooxidans provide new perceptions of how life could adapt to an extremely acidic environment, as well as to the use of a low-energy substrate with poor solubility. Information emerging from other acidophiles such as Leptospirillum ferrooxidans, Metallosphaera sedula and Sulfolobus metallicus show some common features that are likely to be related to these strong constraints: (i) all seem to accomplish the preliminary oxidation of iron via an outer membrane cytochrome c, (ii) the final electron acceptor is in most cases the cytochrome c oxidase, as only the H2O/O2 couple has a thermodynamic potential superior to iron potential in acidic environment, and (iii) they have the capacity to push electrons uphill to reduce NAD via a bc complex working in reverse. Recent advances on sequence information from several genomes and metagenomes of acidophiles should help us to understand the diversity as well as the common features of the pathway involved in iron oxidation.
Electron Transfer at the Microbe–Mineral Interface: A Biochemical Society Focused Meeting held at University of East Anglia, Norwich, U.K., 2–4 April 2012. Organized and Edited by Jim Fredrickson (Pacific Northwest National Laboratory, U.S.A.), David Richardson (University of East Anglia, U.K.) and John Zachara (Pacific Northwest National Laboratory, U.S.A.).
This work was supported by research grants from the Centre National de la Recherche Scientifique (CNRS), Région Provence–Alpes–Côte d'Azur (Région PACA) and the Agence Nationale de la Recherche (ANR). M.R. has a fellowship of the French Ministry of Research (Ministère de l'Enseignement Supérieur et de la Recherche).