Analytical and Bioanalytical Chemistry

, Volume 406, Issue 4, pp 1011–1027 | Cite as

Reconstitution of supramolecular organization involved in energy metabolism at electrochemical interfaces for biosensing and bioenergy production

  • M. Roger
  • A. de Poulpiquet
  • A. Ciaccafava
  • M. Ilbert
  • M. Guiral
  • M. T. Giudici-Orticoni
  • E. Lojou
Review

Abstract

How the redox proteins and enzymes involved in bioenergetic pathways are organized is a relevant fundamental question, but our understanding of this is still incomplete. This review provides a critical examination of the electrochemical tools developed in recent years to obtain knowledge of the intramolecular and intermolecular electron transfer processes involved in metabolic pathways. Furthermore, better understanding of the electron transfer processes associated with energy metabolism will provide the basis for the rational design of biotechnological devices such as electrochemical biosensors, enzymatic and microbial fuel cells, and hydrogen production factories. Starting from the redox complexes involved in two relevant bacterial chains, i.e., from the hyperthermophile Aquifex aeolicus and the acidophile Acidithiobacillus ferrooxidans, examination of protein–protein interactions using electrochemistry is first reviewed, with a focus on the orientation of a protein on an electrochemical interface mimic of a physiological interaction between two partners. Special attention is paid to current research in the electrochemistry of essential membrane proteins, which is one mandatory step toward the understanding of energy metabolic pathways. The complex and challenging architectures built to reconstitute a membrane-like environment at an electrode are especially considered. The role played by electrochemistry in the attempt to consider full bacterial metabolism is finally emphasized through the study of whole cells immobilized at electrodes as suspensions or biofilms. Before the performances of biotechnological devices can be further improved to make them really attractive, questions remain to be addressed in this particular field of research. We discuss the bottlenecks that need to be overcome in the future.

Keywords

Protein orientation Membrane protein Metabolism Bioenergetics Electrochemistry 

Abbreviations

CcO

Cytochrome c oxidase

DET

Direct electron transfer

DvH

Desulfovibrio vulgaris Hildenborough

PMIRRAS

Polarization-modulation infrared reflection–adsorption spectroscopy

QCM

Quartz crystal microbalance

SAM

Self-assembled monolayer

SEIRA

Surface-enhanced infrared absorption

SHE

Standard hydrogen electrode

t-BLM

Tethered bilayer lipid membrane

Introduction

More and more biotechnological devices are nowadays expected to offer green and sustainable alternatives to chemical processes, including those in fields as varied as drug production, bioreactors, and biosensors, but also sustainable treatment of wastewater, carbon dioxide offsets, and bioremediation of recalcitrant chemicals. Proteins and enzymes and also whole microorganisms are preferred to chemical catalysts because they are often more specific and efficient and are more environmentally friendly. Among these devices, one class is based on electron exchange for substrate transformation and involves redox proteins or enzymes isolated or included in protein complexes or in whole microorganisms. The use of these redox biological entities allows us to define electrochemical biosensors, but also biofuel cells, which have undergone huge developments during the last 10 years [1, 2, 3, 4, 5, 6, 7]. Numerous biosensors have been designed for clinical, food, and environmental applications. Enzyme-based biofuel cells are promising energy sources operating under mild conditions and with no platinum-based catalysts. Microbial fuel cells have received much attention as alternatives to more or less stable enzymes because they rely on electroactive biofilms constituted by microbe networks. One key step in obtaining high-performances devices is the efficiency of the electron transfers all along the electron transfer chain, and also between the biological entities and the electrochemical interface which is required to give/receive the first/last electron to/from the biological electron transfer chain. A deep understanding and quantification of charge transport at the enzyme–electrode or microbe–electrode interface is ultimately central to defining fundamental limits and possibly further improving biotechnological devices such as biosensors or biofuel cells. Understanding the interactions and electron transfer kinetics between enzymes and proteins along the electron transfer chain is also crucial to take advantage of the natural substrate channeling expected to increase the efficiency.

On the other hand, the study of the electron transfer mechanisms of redox proteins and enzymes which are involved in many real biological processes will help us understand regulation of many life processes. This is a challenging task because the cellular redox environment includes all redox couples participating in the metabolic network. Metabolic transformations occur through sequential electron transfer steps across protein–protein interfaces or between specific domains of the protein assemblies. These intricate processes can be investigated by electrochemical tools which allow the study of electron transfer between an isolated enzyme and the electrochemical interface and also intramolecular and intermolecular electron transfers. To explore complex energy metabolisms, an attractive way is to reconstitute at the electrochemical interface all or part of the metabolic chain. This supposes rational modification of the electrochemical interface to ensure an electrical communication between the biological assembly and the electrode. Interesting reports have addressed this issue for carbon-based or gold-based electrodes [8, 9, 10, 11, 12, 13]. In this review, recent developments in the investigation of energy metabolic chains based on electrochemical tools are discussed. The redox complexes involved in two relevant bacterial chains, i.e., from the hyperthermophile Aquifex aeolicus and the acidophile Acidithiobacillus ferrooxidans, are first presented in order to clearly underline the organization of the redox entities for the efficiency of substrate transformation. The benefits offered by electrochemistry in the resolution of the electron transfer pathways are then discussed. Protein–protein interactions are first considered as the simplest electron transfer chain. Focus is placed on the orientation of a protein on an electrochemical interface mimic of a physiological interaction between two partners, and which is expected to enhance the electron transfer kinetics. Our understanding of this key issue has undergone a big jump during the last few years thanks to coupled electrochemical and spectroscopic methods. Then, special attention is paid to current research on the electrochemistry of membrane proteins, which are essential elements of energy metabolic chains. The complex architectures built to reconstitute a membrane-like environment at an electrode are especially considered and discussed. The role played by electrochemistry in understanding bacterial metabolisms is finally emphasized through the study of whole cells immobilized at electrodes as suspensions or biofilms. From the latest developments in electrochemically addressed protein–protein interactions, the main questions to be answered before the performances of biotechnological devices can be improved further are considered.

Structural organization of respiratory chain components

Energy is required for many cellular functions, such as for biosynthesis of molecules, transport of molecules across membranes, and cell motility. It is extracted by oxidation from nutrients present in the environment or from sunlight, and is conserved within cells as ATP, which is the universal energy-rich molecule used to drive metabolic reactions. ATP is mainly produced in cells by photosynthetic and respiratory processes involving electron transfer chains that couple electron transfer to the transfer of protons across membranes [14].

Many actors…

In bacteria, ATP synthesis through oxidation of nutrients involves a series of respiratory complexes. Bacteria are versatile organisms that are able to metabolize a wide range of energy substrates because they possess a large variety of redox enzymes. This diversity of prokaryotic respiratory chains is reflected by their ability to oxidize and reduce organic or inorganic, water-soluble, solid, or gaseous chemical compounds as electron donors (such as lactate, glucose, methanol, H2, and hydrogen sulfide) and electron acceptors (such as oxygen, sulfate, ferric iron, fumarate, and elemental sulfur). When the energy source is highly insoluble and thus remains outside the cells, some organisms have been able to develop sophisticated systems, which are not yet fully understood, to oxidize this energy source. In some cases, the secretion of redox molecules (such as cytochromes) used as shuttles to transfer extracted energy toward the cell has been described [15]. The existence of these varied respiratory systems allows adaptation of these microorganisms to distinct environments, some of them considered by humans as extreme.

…in a well-defined organization

The components involved in respiratory pathways are highly structured in the membrane through protein–protein interactions as well as lipid–protein interactions. These interactions, being transient or very stable, are highly specific and crucial for the optimal functioning of the chain. Among them, water-soluble or lipophilic electron carriers transfer electrons from one membrane complex to another by sophisticated interactions. It has been proposed that electrostatic interactions serve for preorientation of both partners and then hydrophobic interactions drive an orientational rearrangement allowing electron transfer [16].

Respiratory complexes can be organized into stable supramolecular assemblies called supercomplexes that have been extensively studied in mitochondria. They constitute functional units and are composed of protein complexes that can function both individually and inside a supercomplex to increase the yield and kinetic efficiency of catalytic reactions. In addition, association of complexes into supercomplexes is required for the stability and assembly of some individual complexes [17, 18]. Two extreme models for arrangement of complexes in the membrane are conceivable [19, 20]: the “fluid state” model proposes that respiratory complexes are independent and freely diffuse in the lipid bilayer, whereas the “solid state” model favors assembly of complexes into stable supercomplexes in vivo, thus making possible a channeling of electrons between physically associated complexes. It is now proposed that complexes are not entirely integrated into supramolecular structures and that individual complexes and supercomplexes can coexist in the membrane. A recent concept integrates the notion of dynamics with the idea that association/dissociation is related to the physiological state of the cell [21]. This novel concept will require further study to fully describe the molecular basis and the regulation of these processes. To date, few supercomplexes involved in respiration in bacteria have been evidenced and biochemically and functionally characterized, most of them containing only the bc-type complex and cytochrome c oxidase (CcO) [22, 23]. Why are bioenergetic supercomplexes so underrepresented among prokaryotes? We can hypothesize that (1) experimental conditions might not be suited to find them, (2) they might be more labile than eukaryotic ones, and (3) they might not be absolutely required for electron transfer in prokaryotes.

In addition, bacterial redox complexes tightly interact with cardiolipin, a minor lipid bearing two negative charges and four acyl chains, and which is present in energy-transducing membranes such as the bacterial inner membrane. This protein-bound lipid modulates the enzyme activity of some of the bacterial redox complexes and is essential for the stability of these protein structures [24, 25].

Aquifex aeolicus and Acidithiobacillus ferrooxidans: two relevant illustrations

The two examples of bacterial respiratory chains detailed below illustrate well the highly organized arrangement of proteins and complexes dedicated to electron transfer as well as the various strategies developed by microbes to recover energy from substrates.

The chemolithoautotrophic bacterium A. aeolicus, isolated from a shallow submarine hydrothermal system, grows at extreme temperatures (optimal growth at 85 °C) using H2, O2, and a sulfur compound. At least three bioenergetic pathways exist in this microorganism as it oxidizes H2 or H2S with O2, but it is also able to reduce elemental sulfur using H2 as an electron donor [26, 27]. The first enzyme of the H2/O2 pathway is a dimeric [NiFe] hydrogenase which interacts with a b-type integral membrane cytochrome and catalyzes the oxidation of H2 (Fig. 1a). This electron transport chain also involves a bc 1 complex as well as a ba 3 CcO catalyzing the reduction of O2 [29, 30]. These two respiratory complexes are strongly associated in the inner membrane of A. aeolicus and can hardly be separated from each other. Indeed, this structure is highly stable to various denaturing agents when purified. This extreme insensitivity to detergents is not a general feature of respiratory membrane-bound supercomplexes as interactions between supercomplex components are described to be rather weak [26, 27]. In the other two electron transfer chains (H2S/O2 and H2/S), the components are organized into stable supercomplexes in the inner membrane. The H2-oxidizing and sulfur-reducing supercomplex consists of a membrane-bound [NiFe] hydrogenase and a molybdenum–sulfur reductase [31]. The purified multienzyme supercomplex containing a sulfide–quinone reductase, a bc 1 complex, and a ba 3 CcO is also enzymatically functional, reducing O2 with H2S as an electron donor [22]. The association of respiratory enzymes in large assemblies in A. aeolicus might be a way, in addition to kinetic efficiency and structural stability, to regulate the electron flow toward the different pathways depending on the growth conditions.
Fig. 1

Models of the supramolecular organization of the respiratory chains of a Aquifex aeolicus and b Acidithiobacillus ferrooxidans. a The H2/O2 respiratory chain of A. aeolicus couples an O2-tolerant membrane-bound [Ni-Fe] hydrogenase (red) via a bc 1 complex (green) and a cytochrome c-555 (light blue) to a terminal cytochrome c oxidase (CcO; light blue), reducing O2 to water. The model of the membrane-bound hydrogenase is based on the closest homologous hydrogenase from Ralstonia eutropha [Protein Data Bank (PDB) ID 3RGW], and the cytochrome b model is based on the cytochrome b from Escherichia coli formate dehydrogenase (PDB ID 1KQF). The bc 1 complex is represented by the structure from Rhodobacter sphaeroides (PDB ID 2FYN), and CcO is represented by the structure from Thermus thermophilus (PDB ID 3S8F). b The respiratory chain of A. ferrooxidans forms a functional supercomplex that spans the outer membrane and the inner membrane. This chain couples the oxidation of Fe2+ to Fe3+ with the reduction of oxygen to water. It is composed of an outer-membrane monoheme c-type cytochrome (Cyc2) where Fe2+ oxidation occurs, the periplasmic blue copper protein rusticyanin (blue; PDB ID 1RCY), the diheme cytochrome c 4 (orange; PDB ID 1H1O), and an integral inner-membrane aa 3-type CcO (the four subunits of CcO (shades of green; PDB ID 1QLE) that catalyzes O2 reduction. An additional copper protein AcoP (blue-green; PDB ID 2BWI) plays a critical role for CcO activity at pH 2 [28]

A. ferrooxidans is a strictly acidophilic bacterium (optimal pH of 1.5–2.5) and an obligate chemolithoautotroph that uses for growth the energy produced by the oxidation of ferrous iron and/or reduced sulfur compounds and fixes atmospheric carbon dioxide as a carbon source. Most of the studies on this organism have been conducted to understand how it uses Fe2+ as an electron source under acidic conditions (for reviews, see [28, 32]). What has clearly been shown is that, instead of being organized “horizontally” with the redox proteins in the cytoplasmic membrane (as described in the previous example), the respiratory chain of A. ferrooxidans forms a “vertical” supramolecular structure involving various metalloproteins that span the outer and inner membranes [33] that conducts the electrons from Fe2+ to the terminal electron acceptor (O2) (Fig. 1b). Such vertical organization for a respiratory chain had never been shown before. In natural environments, the iron source is found mainly as insoluble forms of iron. This may explain the vertical organization of this chain, which moreover avoids the precipitation of Fe3+ at neutral pH of the cytoplasm as well as the formation of reactive oxygen species by reaction of Fe2+ with O2. This supramolecular organization is believed to optimize electron transfer, minimizing the effect of diffusion. In acidophile organisms, the pH of the periplasmic compartment is similar to that of the environment (around pH 1.5 – 2.5 for A. ferrooxidans). The supercomplex organization might consequently participate in the stabilization of the periplasmic components of the respiratory chain. Most of the metalloproteins involved in this supercomplex have been individually characterized and some functional interactions between them have been defined [28]. For example, the direct interaction between rusticyanin and diheme cytochrome c (cytochrome c 4) has been demonstrated. A specific glutamate residue of cytochrome c 4 has been shown to be involved in the interaction [34] that decreases the rusticyanin redox potential by more than 100 mV, facilitating electron transfer. The interaction between cytochrome c 4 and CcO has also been studied, and involves a critical tyrosine residue of cytochrome c 4 [34]. A recent study highlighted an additional partner in tight interaction with CcO [35]. This protein plays a key role in maintaining CcO activity at pH 2 [35]. A putative role as an electron carrier in the iron oxidation pathway is currently being investigated.

Electrochemistry as a tool to study the efficiency of energetic chains

From the two examples just described, we find that knowledge of both the structural organization in a metabolic chain and the ways biology controls the efficiency of electron transfer over long distances through proteins is mandatory to explain the global working of the energetic chain. Electrochemistry, which is able to provide the thermodynamic data for proteins and complexation processes, as well as kinetic data on the intermolecular and intramolecular electron transfer processes, thus appears to be an ideal tool to obtain insight into the energy pathways. For more than 30 years, bioelectrochemistry has been proved to be a powerful means to link electron transfer data to the structure and function of proteins. Early in the 1980s, direct electron transfer (DET) between metallic electrodes [36], then graphite electrodes [37] and mitochondrial cytochrome c, a small monoheme protein, was demonstrated. Nowadays, owing to the improvement in electronics, the diversity of electrode materials, and the concomitant knowledge of protein structures, many proteins and enzymes are addressable using electrochemistry. One key issue in that field is to consider the electrochemical interface as an artificial partner of one element of the energetic physiological chain. From this basic idea, bioelectrochemical interfaces allow the reconstitution of more or less fully reconstituted biological chains, and knowledge of bacterial energy metabolism complementary to that from biological studies. But one challenge remains: the electrochemical study of huge enzymes and especially of membrane proteins, membrane complexes, and finally whole organisms that are under far less control than soluble ones.

A key step: protein orientation at the electrochemical interface

Many proteins incorporate redox cofactors during maturation which are essential for biological processes. These includes metallic centers such as iron, molybdenum, and copper, and nonmetallic cofactors such as flavin. Conversion of the substrates occurs at these active sites, which are most often buried inside the protein moiety. If this position protects the protein against the environment, it also electrically isolates it. To ensure efficient intramolecular electron transfer, electron relays are positioned at distances of less than 14 Å so that electron tunneling can be achieved from the active site to the surface of the protein, where electrons are further transferred to the physiological partner in the energy chain [38].

Electrostatic forces are very often involved in the recognition process between partners before electron transfer, leading to a transitory complex in which the surface electron relay distances are minimized. However, for many biological chains, both electrostatic and hydrophobic interactions are involved. Electrostatic interactions play a key role in the specific recognition, whereas hydrophobic interactions may stabilize the protein complex. Such short distances are also required for intermolecular electron transfer or direct interfacial electron transfer (DET) (Fig. 2). Many proteins have a diameter of 5 nm or more. Consequently, their simple adsorption on an electrochemical interface yields a distribution of distances between the electrode and the electronic relay, and hence a distribution of the interfacial electron transfer rates. Knowledge of in vivo intermolecular electron transfer is one key to succeed in optimizing the distances between the surface electronic relays of a protein and an electrode. The mandatory condition to mimic physiological electron transfer processes is to achieve a specific orientation of the protein on the electrode as similar as possible to the physiological orientation.
Fig. 2

The direct or mediated interfacial electron transfer process depends on the distance between the surface electron relay on the protein and the electrode. When this relay is positioned at a distance of less than 15 Å, then direct electron transfer (DET) can be achieved between the protein and the electrode. For larger distances, a redox mediator is used to ensure a mediated electron transfer (MET) process

The first evidence for protein orientation was provided by the study of horse heart cytochrome c. In vivo, cytochrome c forms a complex with membrane CcO in which the binding domains involve a group of positively charged lysine residues surrounding the heme edge of cytochrome c and negatively charged glutamate residues on CcO [39]. Electrode chemical modifications are made on the basis of this structural feature to obtain a proper orientation for cytochrome c so that DET between the protein and the electrochemical interface occurs. Self-assembled-monolayers (SAM) on gold electrodes are especially targeted because they exhibit similarities with phospholipid membranes with respect to the amphiphilic character and the organization of the film. Determination of the heme redox potential and electron transfer kinetics as a function of various parameters such as temperature, ionic strength, and the presence of inhibitors is thus possible [40, 41, 42]. It was interestingly demonstrated using immobilization of a single cysteine mutant that a dynamic electrostatic arrangement of cytochrome c could be obtained [43]. In that study, switchable control of the interfacial electron transfer was induced by application of an appropriate potential. As cytochrome c is involved in many respiratory chains, it is still the subject of numerous electrochemical studies, either in view of its use in biosensors [44] or for the understanding of relevant chemical groups in biological processes [45]. In each case, the key step remains the oriented immobilization of the wild-type or mutant proteins. In a very recent report, comparison of the orientation of cytochrome c on hydrophilic or hydrophobic SAM gold electrodes proved that electron transfer was dependent on the orientation of the heme plane [46]. Coupling of electrochemistry with surface spectrometric methods is now needed to prove this assumption.

Immobilization of other relevant small proteins (less than 30 kDa) highlights the great fundamental contribution of oriented control at the interface for the understanding and quantification of intramolecular and intermolecular electron transfer processes. Furthermore, these proteins most often operate in membrane environments able to modify the protein conformation. Electron transfer processes within immobilized proteins are thus often compared with those in homogeneous conditions. This is the case for the bacterial diheme cytochrome c 4, which has been greatly studied in order to understand the effect of microscopic interactions in multicenter redox proteins on electron transfer (Fig. 3a). When immobilized in a thin-layer configuration at the electrode, the two heme groups of A. ferrooxidans cytochrome c 4 communicate independently with the electrode (Fig. 3b). The formation of an electron transfer complex between rusticyanin and cytochrome c 4 from the acidophilic bacterium A. ferrooxidans has been demonstrated by surface plasmon resonance, and residues involved in the interaction have been identified by site-directed mutagenesis [34]. As described above, the formation of this complex is proposed to play a key role in the electron transfer chain, by decreasing the redox potential of rusticyanin. However, no proof of such an interaction has been obtained at an electrochemical interface so far. Future work should aim to achieve orientation of this cytochrome c 4 on functionalized electrodes in order to reconstitute the respiratory chain either with CcO or with rusticyanin. Elegant works toward this goal have been performed in Ulstrup’s group on cytochrome c 4 from the neutrophilic Pseudomonas stutzeri [50]. This cytochrome c 4 is a highly dipolar protein with excess positive and negative charge on the C-terminal domain and the N-terminal domain, respectively [49]. The two hemes have redox potentials separated by 100 mV, and the highest redox potential heme is located at the C-terminal domain. The equilibrium distance between the two irons (19 Å) is not favorable for fast intramolecular electron transfer. In line with this structural feature, voltammetric signals of freely diffusing cytochrome c 4 exhibit two independent waves for the two hemes [51, 52]. Cytochrome c 4 from P. stutzeri has been specifically oriented at SAM electrodes, with the first heme able to exchange directly at the electrode and the second one only electrochemically addressable via intramolecular electron transfer through the first one. In contrast to the results obtained in homogeneous conditions, the voltammetric reversible waves turn into asymmetric waves, which is compatible only with a fast intramolecular electron transfer process (Fig. 3c). This suggests that alignment of the heme groups on the electrochemical interface is crucial for cytochrome c 4 function, and may illustrate the conformational triggering of electron transfer when binding to a biological membrane occurs.
Fig. 3

Electron transfer on diheme cytochrome c 4. a Structure of A. ferrooxidans cytochrome c 4. b Voltammogram for A. ferrooxidans cytochrome c 4 on a gold electrode in a thin layer between the gold electrode and a dialysis membrane (this configuration is described in [47]). c Oriented Pseudomonas stutzeri cytochrome c 4 at a carboxylic functionalized self-assembled monolayer gold electrode and the corresponding voltammograms for COOH(CH2)11SH; k 1 = 18 s-1 and k 2 = 18,000 s-1. SCE saturated calomel electrode. (a Adapted from [48]; c from [49])

Immobilization of enzymes, which are often around 100 kDa, is more challenging. The active sites are far less exposed to solvent than the monoheme or diheme cytochromes just reported, increasing the critical role of tunneling distances. Again the key issue is knowledge of the crystal structure of the enzyme associated with understanding of the electron and substrate pathways. This ensures guidance for proper modification of the electrochemical interface, allowing close approach of the surface electron relay and free access of the channels for the diffusion of substrates [53]. Multicopper oxidases are very efficient enzymes for O2 reduction. They are commonly used in glucose/oxygen biofuel cells. They contain four copper active sites, one copper ion at the blue T1 site, where electrons are accepted, and three copper ions at the trinuclear cluster T2/T3 site, where electrons are driven to reduce oxygen. Bilirubin oxidase belongs to this family. In vivo it catalyzes the oxidation of bilirubin to biliverdin, with the concomitant reduction of O2 to water. Three channels are identified running from the T2/T3 site to the surface. At the T1 site, a hydrophilic pocket for substrate binding has been identified [54]. An electrode modified directly by the substrate bilirubin has been demonstrated to enhance current densities for O2 reduction by taking advantage of this peculiar environment, with the enhancement most probably occurring through favored bilirubin oxidase orientation at the electrochemical interface. In contrast, laccase, another multicopper protein, has a hydrophobic pocket for oxidation of its substrates, such as products of lignin degradation. In accordance with the crystallographic data, a rational strategy of electrode modification based on modification of a graphite electrode with anthracene can be applied [55]. Accordingly, both current for O2 reduction and long-term stability are enhanced. SAM gold electrodes functionalized with phenyl groups ending with either amino or carboxyl functions have also been evaluated as efficient platforms for laccase immobilization [56]. Polarization modulation infrared reflection–adsorption spectroscopy (PMIRRAS), use of a quartz crystal microbalance (QCM), and electrochemistry were coupled and allowed the demonstration of a specific orientation linked to a different activity of the enzyme depending on the chemical function of the SAM. Although the study did not focus on the relation between the structure and the enzymatic activity, the work proves how coupling methods will be necessary in the future to increase knowledge of enzyme orientation and conformation in the immobilized state.

Another class of proteins, the hydrogenases, which are the key enzymes for H2 oxidation in many microorganisms, have been extensively studied in recent years, including in our group. This research is motivated by the unique efficiency of hydrogenases toward H2 oxidation and proton reduction which can be used to produce sustainable energy in biotechnological devices such as biofuel cells. As underlined at the beginning of this review, a [NiFe] hydrogenase has been identified in a metabolic chain coupling H2 oxidation to O2 reduction in the hyperthermophilic bacterium A. aeolicus (Fig. 1a). Besides, [NiFe] hydrogenases from mesophilic, anaerobic sulfate-reducing bacteria have been studied for many years [57]. That means that various hydrogenases displaying different properties against temperature and O2 sensitivity are available. In the search for enzymes able to replace platinum as biocatalysts in H2/O2 biofuel cells, the main question remains whether the enzymes can be efficiently immobilized on an electrochemical interface. Electrochemistry provides an outstanding tool in order to search for the physicochemical parameters that drive the interfacial electron transfer process. Once determined, these parameters will improve our understanding of the organization and efficiency of the biological chain.

All [NiFe] hydrogenases are composed of two subunits of around 40 and 60 kDa. The corresponding size of the enzyme is around 6 nm when considered as a globular protein. The electrons released on H2 oxidation at the active site buried inside the large subunit are conducted through a line of three FeS clusters separated by less than 15 Å. From this structural view, it is clear that fast DET between the hydrogenase and an electrode will only be achieved if the distal FeS cluster, i.e., the one exposed to the enzyme surface, is at a tunneling distance from the electrode. This has a physiological sense when considering the electrode as a mimic of the redox partner, which effectively binds the enzyme through this cluster. Consequently, the amino acid environment of the distal FeS cluster plays a key role in the search for an electrode interface that can match it for a proper orientation. This concept is becoming more and more attractive within the hydrogenase research domain, and in recent years several works have taken advantage of electrochemistry for hydrogenase-oriented immobilization studies [58, 59].

In the mesophilic [NiFe] hydrogenase from Desulfovibrio species, an anionic patch of amino acid residues, mainly glutamate residues, surrounds the FeS distal cluster [59]. Furthermore, a high dipole moment points toward the distal FeS cluster. This polarity allows electrostatic recognition with the positively charged tetraheme cytochrome c 3 in the physiological electron transfer pathway. Accordingly, H2 oxidation proceeds either directly or through a redox mediator according to the charge of the SAM on which the enzyme is immobilized (Fig. 4a) [60]. Through the coupling of surface-enhanced infrared absorption (SEIRA) spectroscopy with cyclic voltammetry, it was demonstrated that the active site of the enzyme was structurally intact and functional on a positively charged electrode, but the H2 catalytic oxidation currents were not stable with time because of progressive modification of the active site [61]. The recent combination of SEIRA spectroscopy with molecular dynamics opens a new avenue for the rational design of biocompatible interfaces that preserve hydrogenase integrity and functionality [62].
Fig. 4

Specific orientation of hydrogenases a from the mesophilic Desulfovibrio fructosovorans as a function of the charge of the interface and b from the hyperthermophilic A. aeolicus as a function of the hydrophobicity of the interface

A crystallographic model has been built from the sequence of the [NiFe] hydrogenase from A. aeolicus. From this model it appears that no charged amino acid residues surround the distal FeS cluster [53, 59]. Accordingly, whatever the charge of the electrode, two populations of A. aeolicus hydrogenases are identified by electrochemistry: the first one with the distal FeS cluster at a tunneling distance from the interface so that DET can occur for H2 oxidation, and the second one oriented with the distal FeS cluster far enough from the interface so that a redox mediator is needed for the catalytic process [63]. From the coupling of electrochemistry, atomic force microscopy, and PMIRRAS, it has been proposed that the hydrophobic transmembrane helix which serves as an anchor for the membrane-bound hydrogenase in the cell membrane is the key factor in the oriented immobilization [64]. This helix is surrounded by neutral detergent [64, 65], and defines a highly hydrophilic domain, less than 15 Å from the distal FeS cluster. As a consequence, hydrophobic interfaces are demonstrated to greatly favor an orientation of the enzyme with the FeS distal cluster far from the electrochemical interface, thus disfavoring a DET process for H2 oxidation (Fig. 4b). However, quantification of detergent molecules around the transmembrane helix and the position of this helix on the solubilized enzyme needs to be addressed to obtain a complete scheme for the enzyme orientation.

Photosystem II is an even bigger enzyme than multicopper oxidases or hydrogenases, having a molecular mass of around 700 kDa and a size of around 10 × 20 × 11 nm3. Its use for the sustainable production of solar fuels such as H2 requires an improvement in photocurrents once it has been immobilized. Nevertheless, early on, it was reported that an oriented monolayer of the bacterial photoreaction center could be obtained through specific modification of a graphite electrode by bifunctional reagents, able to be grafted on the graphite by ππ interactions, and to link a unique cysteine situated at the surface of the protein [66]. An enhancement in the photocurrent is observed because the cysteine engaged in the bonding to the electrode is situated near to redox sites QA and QB, providing electrical communication with the electrode interface. The crystal structure of photosystem II at a resolution of 1.9 Å [67] nowadays allows the rational design of efficient and stable photosystem II hybrid photoelectrodes [68]. Taking into account the positive dipole moment of photosystem II, which is located on the same side as the electron donors, in particular the quinones QA and QB, modification of electrodes by carboxylic functions serves for favorable orientation of photosystem II for efficient visible-light-driven water oxidation.

Finally, even if a specific orientation of the enzyme may be achieved at an electrode interface as a result of knowledge and control of both the enzyme and the interface surface, the true turnover frequency most often remains unknown. This is the case for most redox enzymes, because the low coverage on the electrode precludes any noncatalytic signals that would allow the quantification of the electroactive immobilized enzymes. The coupling of electrochemistry on SAM electrodes and electrochemical scanning tunneling microscopy has recently addressed this challenge. A high turnover frequency of 21,000 s-1 was calculated for hydrogen production by an [FeFe] hydrogenase [69]. Great efforts must, however, be made to develop such methods for other enzymes in the future.

Protein–protein interaction at the electrochemical interface

Specific orientation of enzymes at electrochemical interfaces is currently essentially targeted for enhanced DET processes. Besides, protein–protein binding is at the heart of biology, with affinity constants ranging over many orders of magnitude, and many diseases involving the disruption and/or elimination of vital interactions. Knowledge of the physical parameters that drive the electron transfer process for one protein will serve the reconstitution of more than two elements from a biological chain. But this is quite a challenging task. Although the orientation with a redox cofactor toward the electrode favors interfacial electron transfer, the same side is usually involved in docking of enzymes. As a consequence, only a few articles have reported protein–protein interaction studies via electrochemical tools.

The most reported studies in that domain concern enzymatic reactions involving the enzyme and its physiological partner and followed by electrochemistry [70, 71, 72, 73, 74, 75, 76, 77, 78]. Some examples of homogeneous electrocatalysis are listed in Table 1. Kinetic data are obtained from these studies as a function of the redox partner, highlighting that a so-called pseudospecificity can be sometimes obtained. Various parameters that may influence the catalysis are easily evaluated, such as ionic strength, presence of inhibitors, temperature, and pH. Electrochemistry also allows the determination of the role played by redundant proteins in physiological chains. A relevant example comes from the chain that couples H2 oxidation to sulfate reduction in Desulfovibrio species. In those chains, various multiheme cytochromes, with three to 16 hemes, have been identified. The electrochemical study of the interaction between some of these cytochromes permits the proposal that a high molecular weight cytochrome c would act as a shuttle to transfer the electrons across the membrane for sulfate reduction [31].
Table 1

Homogeneous enzymatic electrocatalysis involving two physiological partners

Microorganism

Enzyme

Partner

Substrate

k (M-1 s-1)

Reference

Desulfovibrio vulgaris Hildenborough

Hydrogenase

Cytochrome c 3

H2

30 × 107

[70]

Desulfovibrio fructosovorans

Hydrogenase

Cytochrome c 3

H2

4 × 107

[71]

Desulfovibrio africanus

Hydrogenase

Cytochrome c 3

H2

9 × 108

[72]

Pseudomonas aeruginosa

Nitrite reductase

Cytochrome c-551

NO2 -

2 × 106

[73]

Paracoccus pantotrophus

Cytochrome c peroxidase

Pseudoazurin

H2O2

1.4 × 105

[74]

Pseudomonas stutzeri

Cytochrome c peroxidase

Cytochrome c-551

H2O2

4 × 105

[75]

Achromobacter cycloclastes

Nitrous oxide reductase

Pseudoazurin

N2O

4 × 105

[76]

Sulfite oxidase

Cytochrome c

SO3 2-

2.5 × 106

[77]

Corynascus thermophilus

Cellobiose dehydrogenase

Cytochrome c

Cellobiose

2 × 105

[78]

The first evidence for the reconstitution of a physiological complex at the electrode was reported by Bagby et al. [79] and then Burrows et al. [80]. It was shown that negatively charged cytochrome b 5, a redox membrane protein involved in the electron transfer pathway from NADPH to CcO via cytochrome c, is electroactive only when in a complex with cytochrome c. Voltammetric signals revealed that the two proteins exchanged independently with the electrode. Unexpectedly, the same lysine residues are supposed to recognize the electrode and cytochrome b 5. The reasons were not clear at that time, but it was suggested that flexible binding domains were used in the binding recognition, leading to rapid interconversion of the different complexes.

A similar result was obtained in our group using one bacterial tetraheme cytochrome c 3 and bacterial ferredoxin [81]. An electrochemical signal for ferredoxin is obtained only in the presence of cytochrome c 3. A complex between the basic cytochrome c 3 (pI = 10.2) and ferredoxin (pI = 3.5) was demonstrated by coupling electrochemistry and use of a QCM. Although cytochrome c 3 and ferredoxin are nonphysiological partners, a pseudospecificity interaction occurs, since other positively charged proteins are not able to promote the electron transfer on ferredoxin. The four hemes exposed to the solvent in addition to the isotropy of charges on the surface in cytochrome c 3 may explain this pseudospecificity. In contrast, when cytochrome c 3 from another bacterium which has a high polarity was used, no promotion of electron transfer on ferredoxin was observed. NMR data demonstrate, however, the formation of a complex in which the positive charges around one heme of cytochrome c 3 constitute the binding domain for ferredoxin [82]. In this case, however, the preferred orientation of cytochrome c 3 at the interface competes with its interaction with ferredoxin. Figure 5 provides the rational explanation of the experimental results that puts forward the role of the protein/protein/electrode ternary complex in the efficiency of the electron transfers in a biological chain. Competition between the electrode and the physiological partners was also observed in layer-by-layer construction of cytochrome c 3 and hydrogenase [83].
Fig. 5

Model of the ferredoxin (Fd)/cytochrome c3/electrode ternary complex: a Fd alone; b Fd plus Desulfovibrio vulgaris Hildenborough (DvH) cytochrome c 3; c Fd plus Desulfovibrio Norway cytochrome c 3. (Adapted from [81])

Considering again the cytochrome b 5/cytochrome c complex, the simple modification of the electrode by polylysine allowed the observation of interfacial electron transfer on cytochrome b 5 followed by intermolecular electron transfer to cytochrome c [84]. Indeed, positive charges on the electrode exclude cytochrome c from the interface while allowing orientation of cytochrome b 5. The physiological complex is thus reconstituted at the electrochemical interface. Second-order rate constants were obtained in agreement with previous data obtained by flash photolysis techniques. Efficient reconstitution of the cytochrome c/cytochrome c peroxidase chain at the electrode was also achieved by taking advantage of a unique cysteine residue on iso-1-cytochrome c from Saccharomyces cerevisiae [85]. This residue is ideally located close to the surface and opposite the lysine-rich site around the heme edge, where cytochrome c peroxidase binds, at a tunneling distance of 16 Å. Reduction of cytochrome c before immobilization allows a favored orientation for both interfacial and intermolecular electron transfer. Chemisorption of cytochrome c on a gold electrode via the cysteine residue is compatible with fast interfacial electron transfer, and the lysine patch remains free for binding of cytochrome c peroxidase, and then for efficient electrocatalytic reduction of H2O2.

How can we reconstitute the membrane environment?

Energy metabolism occurs in the mitochondrial, chloroplast, or plasma membranes. Membrane-bound proteins thus represent around 40 % of the total genome. However, because of the difficulty in purifying and handling them, they are under far less control than soluble proteins, even from a biochemical point of view. This is undoubtedly the case when using electrochemistry, and only a few studies nowadays report examination of integral membrane proteins. Membrane-bound proteins are extracted from cell membranes thanks to the use of detergent, which must be at a concentration far above its critic micellar concentration to ensure stability of the extracted protein. A in-depth investigation of the influence of a neutral detergent on H2 catalytic oxidation by A. aeolicus hydrogenase allowed the determination of the conditions in which the detergent can be present with a beneficial effect on the electrochemical activity [64, 65]. The role of detergent has, however, been mentioned in only very few studies so far, although only this knowledge will open the route for the study of a large number of membrane proteins via electrochemistry.

Hydrophobic interactions are supposed to be predominant when looking for an efficient immobilization of membrane proteins. The simplest way to create a hydrophobic environment at an electrochemical interface is to modify electrodes with alkane thiol derivative SAMs [86]. SAMs furthermore reflect highly structured membranes. O2-resistant hydrogenases such as those from Ralstonia eutropha or A. aeolicus are membrane-bound hydrogenases, with a cytochrome b serving as an anchor in the membrane (see Fig. 1a). Because cytochrome b is dipolar, displaying a positively charged environment on the bottom side, the trimeric complex of the [NiFe] hydrogenase from the O2-tolerant bacterium R. eutropha with cytochrome b was immobilized on negatively charged SAM electrodes [87]. It was shown that electron transfer from the [NiFe] hydrogenase active site to cytochrome b was preserved in the immobilized state, in analogy with the electron pathway in physiological conditions. However, this pathway was demonstrated to be a minor pathway, in contrast to a major pathway which bypasses cytochrome b, most probably because of poor electron transfer on cytochrome b caused by detergent surrounding the membrane protein. Other proteins, such as blue copper proteins, were also proved to have a proper orientation on an alkane thiol SAM electrode [88]. In this latter case, favorable hydrophobic interactions between the copper redox center environment and the methyl-terminated alkane thiol SAM surface are responsible for the efficient immobilization.

Works aiming to develop biomimetic membranes have used CcO as a protein model. CcO is an integral membrane protein which couples the reduction of oxygen to water with the translocation of four protons across the membrane [89, 90, 91]. CcO enzymes of the aa 3 type are the most widely studied because they exhibit high activity for O2 reduction. A binuclear CuA redox site is the primary electron acceptor from cytochrome c. It is situated in a hydrophilic portion of the enzyme that extends out of the membrane into the periplasm. An electron is then transferred to heme a, and then the heme a 3/CuB binuclear site. Molecular O2 binds heme a 3. The location of the CuA site is 8 Å above the membrane surface, and the other three redox sites (heme a, heme a 3, and CuB) are 13 Å below the membrane surface in the 48-Å-thick transmembrane portion. Cytochrome c and CcO were immobilized on hydroxyl-terminated thiol SAM electrodes. No direct electrical connection between CcO and the SAM electrode was observed, supposedly linked to the long distance between the active site and the electrochemical interface. A preformed complex between CcO and cytochrome c promotes redox contact so that catalytic reduction of O2 is now efficient, probably because of orientation in a physiological way on the SAM layer [92].

To improve the membrane model, methyl-terminated alkane thiol can be exploited to assemble a second layer of phospholipids. A hybrid bilayer membrane is thus created at the electrochemical interface which mimics a true cell membrane, exhibiting stable organization and internal hydrophobicity [93]. CcO has been immobilized using a submonolayer of octadecyl mercaptan formed by self-assembly on gold, and then building a lipid bilayer membrane containing the enzyme. The electrode architecture mimics the native membrane structure of the oxidase, with the large hydrophilic portion of the enzyme protruding from the bilayer. This stable membrane-like assembly allowed the study of heterogeneous electron transfer between CcO and the gold electrode. Electrochemistry coupled with scanning force microscopy strongly suggested that CcO was in direct electrical connection with the gold electrode [94].

This architecture lacks, however, the lateral mobility of phospholipids. As hybrid layers are tightly attached to the SAM surface, they can hardly accommodate transmembrane proteins. An improved architecture was constructed on the basis of the anchoring of phospholipid membranes to the electrode by suitable tethers. A so-called tethered bilayer lipid membrane (t-BLM) is formed which is composed of a lipid head group, with hydrophilic spacer groups tethered covalently to the gold support via a thiol function [95]. It acts as a fluid lipid membrane possessing an aqueous phase between the electrode and the first lipid layer. When this surface is incubated with lipid vesicles, the vesicles spread on the first lipid layer (Fig. 6a). [97]. Characterization of these t-BLMs is nowadays undertaken by coupling electrochemistry and PMIRRAS [98].
Fig. 6

Electrochemistry of membrane proteins using hybrid layers. a Principle of the formation of the layer showing the vesicle fusion on the lipid layer. b Incorporation of CcO in a tethered bilayer lipid membrane through His-tag attachment. DMPE 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine. (a From [96]; b from [95])

Enzymes such as CcO were incorporated in these t-BLMs [96]. Improvement in the stability and activity of the incorporated enzyme was obtained by in situ reconstitution of the lipid bilayer around the enzyme, which was previously attached to the electrode using His-tag technology (Fig. 6b) [99, 100, 101, 102]. Orientation of CcO in this t-BLM allows one to study the intermolecular electron transfer with the freely diffusing cytochrome c physiological partner, the DET, and the possibility to catalyze O2 reduction. For these investigations, electrochemistry is coupled with surface enhanced Raman spectroscopy and SEIRA spectroscopy measurements. The His-tags are positioned on CcO on either the cytoplasm side or the periplasm side so that the CuA binding site for cytochrome c is either exposed to the solvent or facing the electrode. In the first orientation, which mimics the physiological orientation, a cathodic current is observed linked to the catalytic reduction of O2 via cytochrome c. However, cytochrome c is also demonstrated to be unable to cross the t-BLM, and tunneling across the membrane is unlikely considering the thickness of the t-BLM (10 nm). A conformational change of CcO induced by the binding of cytochrome c is evocated to be responsible for the long-range communication. In the second orientation, voltammetric waves in nonturnover conditions and direct catalytic current in the presence O2 are observed, suggesting that an electrical contact exists between CcO and the electrode. This opens an up attractive avenue toward O2 direct reduction in physiological-like conditions.

Nevertheless, the best solution to handle membrane proteins while maintaining their stability and activity is to reconstitute their membrane environment. Immobilization of proteoliposomes containing intact enzymes is an elegant way toward preservation of enzyme integrity. Insertion of the trimeric state composed of the two subunits of [NiFe] hydrogenase and cytochrome b into liposomes has been achieved [103]. H2 oxidation via the [NiFe] hydrogenase/cytochrome b complex, either free or inserted into liposomes, was performed using electrochemistry in a configuration where the protein complexes were entrapped in a thin layer between the electrode surface and a dialysis membrane. A five times faster rate constant was measured for H2 oxidation using the proteoliposomes instead of the free [NiFe] hydrogenase/cytochrome b complex, reflecting the higher stability of the enzyme when incorporated in a physiological-like environment.

Immobilization at an electrochemical interface of total membrane extracts instead of enzymes or proteoliposomes is a further step toward physiological-like conditions. Membrane extracts of Bacillus subtilis still containing all proteins and coenzymes were bound to the electrode using cholesterol tethers separated from the gold electrode by a ethyleneoxy chain [104]. It was shown that membrane vesicles remain intact on the surface and keep their physiological activity, being able to reduce fumarate via menaquinone as a mediator. This immobilization procedure is a very useful tool that provides essential information for the understanding of the role of quinones, the key hydrophobic redox mediators of respiratory organisms. By use of t-BLMs, the activity of ubiquinol oxidase and its inhibition by various known inhibitors has been quantified [105]. Very recently, it was clearly demonstrated using a QCM with dissipation and electrochemistry that incorporation of a menaquinol-7 dehydrogenase in solid-supported bilayer membranes resulted in catalysis of the reaction required to support anaerobic respiration of Shewanella oneidensis thanks to the formation of a long-lived complex with its native redox partner [106]. The formation of this complex is proposed to change the redox properties of the enzyme, as suggested earlier in this review for the metabolic chain in A. ferrooxidans. It also supports the concept of dynamic complexes involved in processes regulated by electron transfers. This work appears to be an excellent illustration of how electrochemistry can quantify protein–protein interactions which regulate the respiratory activity by reconstitution of part of a metabolic chain.

The electrochemical approach of energy metabolism

Relevant examples in this review have underlined that the redox metabolism of a bacterial cell must be understood in order to understand the physiology [107]. Although some of the redox couples can be quite easily identified and kinetically quantified, understanding the metabolism is still challenging because it implies integrating all the interacting redox couples. Spectrophotometry can provide data on the presence and function of enzymes in subcellular fractions, cell lysates, or cell extracts. Electrochemistry, in contrast to other methods such as spectrophotometry and mass spectrometry, provides information on signaling pathways, regulatory mechanisms, and the interplay of different subcellular compartments by measuring the intracellular redox changes in vivo. This is an area of increasing research, especially because of the development of microbial biofuel cells, which use microbial biofilm deposition on electrodes to generate electricity. Many extensive reviews have been published during the last few years on microbial fuel cells [108, 109]. The current developments in this domain are not in the scope of this review, unless they serve the understanding of bacterial metabolism by use of electrochemistry.

Hydrogen production from biomass fermentation is currently the subject of intense research for understanding the bacterial energy metabolism that could improve the rate of H2 production [110, 111]. A synergetic effect in a bacterial consortium—Desulfovibrio vulgaris Hildenborough (DvH), a sulfate-reducing bacterium, and Clostridium acetobutylicum, a hydrogen-producing bacterium—was highlighted that strongly suggested a change in energy metabolism in the mixed cultures compared with the pure cultures [112]. Caldicellulosiruptor saccharolyticus is also a targeted bacterium because it is able to produce hydrogen from plant biomass. However, its metabolism is known to be partly directed toward lactate production, which is deleterious for H2 production [113]. Moreover, it was also demonstrated that lactate formation was dependent on the carbon source. An amperometric method was recently proposed to follow the changes in the metabolism of the strictly anaerobic and thermophilic bacterium C. saccharolyticus [114]. The electrochemical signal is supposed to reflect the intracellular flow depicted in Fig. 7, which depends on the sugar uptake and oxidation, and the flux distribution at the pyruvate node. The use of ferricyanide as a redox mediator allows oxidation of the substrate to be monitored through regeneration of oxidized NAD+ and ferredoxin by the membrane-bound enzymes. Amperometric currents recorded at an activated platinum electrode confirmed the previous findings that metabolism is directed to lactate in the stationary phase, whereas lactate dehydrogenase activity is increased and pyruvate ferredoxin reductase activity remains constant during the growth phase.
Fig. 7

Amperometric measurement of energy metabolism in Caldicellulosiruptor saccharolyticus. a Sugar metabolism in C. saccharolyticus showing the pyruvate node. b The whole-cell-mediated bioelectrocatalysis. ADH alcohol dehydrogenase, CoA coenzyme A, Cof cofactor, DHAP dihydroxyacetone phosphate, FBP fructose 1,6-bisphosphate, Fd ferredoxin, F1P fructose 1-phosphate, F6P fructose 6-phosphate, GAP glyceraldehyde 3-phosphate, G6P glucose 6-phosphate, LDH lactate dehydrogenase, PEP phosphoenolpyruvate, PFOR pyruvate:ferredoxin oxidoreductase, 3PG 3-phosphoglycerate, S6P sucrose 6-phosphate, Xy5P xylose 5-phosphate. (From [114])

DvH is a sulfate-reducing bacterium that couples hydrogen oxidation to sulfate reduction in the cytoplasm. At least four different hydrogenases have been identified in this bacterium, which differ in the localization in the cell, the metal content of the active site (Fe–Fe, Ni–Fe, or Ni–Fe–Se), and the specific activity toward H2 oxidation or proton reduction. Suspensions of bacterial cells from DvH were demonstrated to catalyze both hydrogen oxidation and proton reduction via a suitable redox mediator [115]. Well-defined catalytic waves, both anodic and cathodic, occur in cyclic voltammetry and allow the determination of the kinetics of the enzymatic reactions. A very high value of the bimolecular rate constant of the reaction between the cell and the substrate was calculated (1012 M-1 s-1). This is a surprising result as mass transfer across the cell membrane is expected to limit the catalysis. Cells from DvH were otherwise immobilized on a carbon electrode thanks to a thin-layer configuration (the cells were entrapped in a thin layer between the electrode surface and a piece of dialysis membrane [41, 47, 116]) [71, 117, 118]. Electrochemistry performed in the presence of bacterial cells at the electrochemical interface revealed the occurrence of a mediated catalytic current which depends on the amount of cells, the growth phase, the substrate (H2 or H+), and the presence of inhibitors. Again, the current is a direct measure of enzymatic activity inside the bacterial cells. From the use of either wild-type or mutant cells at the electrode interface, a specific role for the different hydrogenases in H2 metabolism was proposed. In particular, electrochemistry confirms biochemical studies concerning the role of a periplasmic hydrogenase in H2 uptake.

In these works, however, artificial redox mediators are necessary to achieve the connection between the living cells and the electrode. Ten years ago it was discovered that some bacteria such as Geobacter metallireducens are able to directly exchange electrons with an electrochemical interface, with no need for a redox mediator [119]. Extracellular electron transfer has been widely investigated during the last 5 years. Different strategies have been proposed to be involved in this direct electron exchange: (1) mediated electron transfer through excreted soluble redox mediators, (2) DET from the outer cell membrane, and (3) filamentous conductive pili, or nanowires, whose molecular structure is still unknown [120, 121, 122, 123] (Fig. 8). In the first mechanism, the critical issue is the diffusion of the excreted proteins through the biofilm. The other two hypotheses are still a matter of debate. With use of an original setup of two electrodes separated by a nonconductive gap much larger than the size of a cell and on which Geobacter sulfureducens biofilm was grown, nanowires were shown to exhibit conductivities as high as those of synthetic metallic nanostructures, yielding conductive bacterial biolfilms [120]. Although these nanowires were decorated by various membrane cytochromes c, it seems that the hypothesis that electron transfer could occur through electrons hopping along these cytochromes c is nowadays ruled out, in particular because these cytochromes c are not close enough. Nevertheless, depending on the bacterium species, outer-membrane proteins such as cytochrome c play a key role in electron transfer between bacterial biofilms and electrodes [124, 125]. It remains to be demonstrated if nanowires must connect directly to the electrode or if electrons can be transferred from a cell to another cell inside the biofilm.
Fig. 8

DET between bacterial cells and an electrochemical interface. a Scanning electron micrograph showing Geobacter sulfurreducens biofilm formation on a graphite electrode poised at -500 mV versus Ag/AgCl. b Evidence for conductive pili in G. sulfurreducens biofilms (a From [119]; b from [120])

Direct H2 production by various microorganisms through the formation of biofilms on electrodes has been reported [126]. Hydrogenase enzymes were proposed to play a key role in the catalysis displayed by whole bacterial cells [127]. As an example, a graphite electrode incubated in a microbial suspension of Desulfovibrio paquesii displayed a reduction current when poised at a very negative potential of –900 mV versus the standard hydrogen electrode (SHE), which was proved to be linked to bioelectrocatalysis by the bacterial cells of H2 production [128]. In another study, the formation of a bacterial biofilm of sulfate-reducing bacteria was shown to decrease the overpotential for H2 evolution [129]. Although some discrepancies exist between the potential at which H2 can be produced, it appears that the bacterial cells are thus able to adapt their metabolism to the electron flow from the graphite electrode. It was furthermore clearly demonstrated that the electrocatalytic activity is linked to the cells attached to the electrode and not to the planktonic cells. Interestingly, H2 does not inhibit its own production, in contrast to the inhibition reported when using pure hydrogenase enzymes [130]. Cyclic voltammetry also reveals the presence of redox species other than hydrogenases in the biofilm that appear only after a period of polarization. The redox potential is compatible with that of cytochrome c, which is known to be abundant in Desulfovibrio species [131].

A. ferrooxidans cells have been evaluated as potential catalysts for oxygen reduction [132]. It was proved that this acidophilic bacterium is able to grow on a graphite felt electrode poised at 0 V versus SCE with no addition of redox mediators. Direct catalytic reduction of O2 is efficient at pH 2, with a decrease in the overpotential of 300 mV, a very attractive result since reduction of O2 to water is limited by the availability of protons [133]. Electrocatalysis in aerobic conditions is accompanied by the appearance of a cyclic voltammetry redox couple around 0 V versus SCE, which is linked to bacterial metabolism. This redox compound has been tentatively assigned to redox proteins located in the outer membrane. At the beginning of this review, it was shown that an outer-membrane-bound cytochrome was the first electron acceptor in the electron transfer chain that couples Fe2+ oxidation to O2 reduction. However, the redox potential of this outer-membrane-bound cytochrome is much higher (i.e., + 560 mV vs SHE [33]) than the potential observed when the biofilm is exposed to O2. This issue has thus to be investigated further.

Finally, one more question should be addressed and concerns the viability of bacterial cells under polarization. It is expected that under a stress signal such as an electric current the cell’s metabolism, shape, and movement would be impacted. Irreversible permeabilization of cell membranes can occur, with leakage of constituents. In a recent work [134], it was shown that currents below 6 A m-2 have only a small effect on the viability of bacterial cells, but for higher current densities, cells on the electrode surface exhibit a high death rate. This feature should be taken into account and analyzed further in microbial fuel cells, where reported current densities are in the range 5–10 A cm-2.

Remaining questions and future directions

Fundamental questions remain to be answered for us better understand the functioning of complexes and supercomplexes in physiological electron transfer chains. What drives the formation and maintenance of such complexes is still not fully understood. Lipidic accessory proteins or environmental conditions seem to play key roles in the regulation of the formation of such complexes and supercomplexes, so the next step will be to characterize them further. The identification of each partner involved in the diverse electron transfer chains and their respective binding site, a fundamental question to begin with, will be of great importance for bioelectrochemistry as illustrated in this review. Some of the current developments in bioelectrochemistry that serve the understanding of energy metabolism have been underlined. To gain insight into this fundamental problem, electrochemistry has still to overcome several bottlenecks. Among them, the quantification of the enzyme coverage which is electroactive at an interface is crucial to be able to access the kinetics of intermolecular electron transfer. Efficient immobilization at the electrode interface of membrane protein and membrane extracts must be improved so that more and more biological objects can be electrochemically addressed. The coupling of electrochemistry with spectroscopic methods, such as ultraviolet–visible spectroscopy, EPR spectroscopy, Raman spectroscopy, and infrared spectroscopy, and also with surface plasmon resonance and use of a QCM will no doubt allow considerable improvements in that domain. Understanding of electron transfer through bacterial biofilms associated with a change in bacterial metabolism also needs in-depth investigations. Concomitantly, research should focus on identifying the microbial species that are responsible for electroactivity, and on the communication between bacteria inside a consortium that has been proved to behave differently from the pure culture. Positive consequences are expected for the development of microbial fuel cells or the production of sustainable fuels such as H2. Of course, extensive research on electrode materials that can support proteins or bacteria while maintaining their electroactivity is needed. But this research cannot be conducted by electrochemists alone. Only strong collaboration between physical chemists, biochemists, and microbiologists will allow there to be an improvement in both the understanding of bacterial metabolism and the performances of biotechnological devices.

Notes

Acknowledgments

The authors thank Région Provence-Alpes-Côte d’Azur, CNRS, and ANR Bioénergie no. ANR-2010-BIOE-003-01 for financial support.

References

  1. 1.
    Leech D, Kavanagh P, Schuhmann W (2012) Electrochim Acta 84:223–234Google Scholar
  2. 2.
    Lovley D (2011) Energy Environ Sci 4:4896–4906Google Scholar
  3. 3.
    Moehlenbrock M, Toby T, Pelster L, Minteer S (2011) ChemCatChem 3:561–570Google Scholar
  4. 4.
    Borgmann S, Schulte A, Neugebaue S, Schuhmann W (2011) In: Alkire RC, Kolb DM, Lipkowski J (eds) Advances in electrochemical science and engineering: bioelectrochemistry, vol 13. Weinheim, Wiley-VCHGoogle Scholar
  5. 5.
    Wang J (2006) Biosens Bioelectron 21:1887–1892Google Scholar
  6. 6.
    Zayats M, Willner B, Willner I (2008) Electroanalysis 20:583–601Google Scholar
  7. 7.
    Amine A, Mohammadi H, Bourais I, Pallesschi G (2006) Biosens Bioelectron 21:1405–1423Google Scholar
  8. 8.
    Willner I, Katz E (2000) Angew Chem Int Ed 39:1180–1218Google Scholar
  9. 9.
    de Poulpiquet A, Ciaccafava A, Lojou E (2013) Electrochim Acta. doi: 10.1016/j.electacta.2013.07.133 Google Scholar
  10. 10.
    Zhou Z, Hartmann M (2013) Chem Soc Rev 42:3894–3912Google Scholar
  11. 11.
    Katz E, Minko S, Halamek J, MacVittie K, Yancey K (2013) Anal Bioanal Chem 405:3659–3672Google Scholar
  12. 12.
    Putzbach W, Ronkainen N (2013) Sensors 13:4811–4840Google Scholar
  13. 13.
    Prakash S, Chakrabarty T, Singh A, Shahi V (2013) Biosens Bioelectron 41:43–53Google Scholar
  14. 14.
    Willey JM, Sherwood LM, Woolverton CJ (2008) Prescott, Harley and Klein's microbiology, 7th edn. McGraw Hill, New YorkGoogle Scholar
  15. 15.
    Richardson DJ, Butt JN, Fredrickson JK, Zachara JM, Shi L, Edwards MJ, White G, Baiden N, Gates AJ, Marritt SJ, Clarke TA (2012) Mol Microbiol 85:201–212Google Scholar
  16. 16.
    Richter O, Ludwig B (2009) Biochim Biophys Acta 1787:626–634Google Scholar
  17. 17.
    Lenaz G, Genova ML (2012) Adv Exp Med Biol 748:107–144Google Scholar
  18. 18.
    Vonck J, Schäfer E (2009) Biochim Biophys Acta 1793:117–124Google Scholar
  19. 19.
    Boekema EJ, Braun HP (2007) J Biol Chem 282:1–4Google Scholar
  20. 20.
    Genova ML, Baracca A, Biondi A, Casalena G, Faccioli M, Falasca AI, Formiggini G, Sgarbi G, Solaini G, Lenaz G (2008) Biochim Biophys Acta 1777:740–746Google Scholar
  21. 21.
    Genova ML, Lenaz G (2013) Biol Chem. doi: 10.1515/hsz-2012-0317 Google Scholar
  22. 22.
    Prunetti L, Infossi P, Brugna M, Ebel C, Giudici-Orticoni MT, Guiral M (2010) J Biol Chem 285:41815–41826Google Scholar
  23. 23.
    Magalon A, Arias-Cartin R, Walburger A (2012) Adv Microb Physiol 61:217–266Google Scholar
  24. 24.
    Arias-Cartin R, Grimaldi S, Pommier J, Lanciano P, Schaefer C, Arnoux P, Giordano G, Guigliarelli B, Magalon A (2011) Proc Natl Acad Sci U S A 108:7781–7786Google Scholar
  25. 25.
    Bazán S, Mileykovskaya E, Mallampalli VK, Heacock P, Sparagna GC, Dowhan W (2013) J Biol Chem 288:401–411Google Scholar
  26. 26.
    Guiral M, Prunetti L, Lignon S, Lebrun R, Giudici-Orticoni MT (2009) J Proteome Res 8:1717–1730Google Scholar
  27. 27.
    Guiral M, Prunetti L, Aussignargues C, Ciaccafava A, Infossi P, Ilbert M, Lojou E, Giudici-Orticoni MT (2012) Adv Microb Physiol 61:125–194Google Scholar
  28. 28.
    Roger M, Castelle C, Guiral M, Infossi P, Lojou E, Giudici-Orticoni MT, Ilbert M (2012) Biochem Soc Trans 40:1324–1329Google Scholar
  29. 29.
    Brugna-Guiral M, Tron P, Nitschke W, Stetter K-O, Burlat B, Guigliarelli B, Bruschi M, Giudici-Orticoni MT (2003) Extremophiles 7:145–157Google Scholar
  30. 30.
    Prunetti L, Brugna M, Lebrun R, Giudici-Orticoni MT, Guiral M (2011) PLoS One 6(6):e21616Google Scholar
  31. 31.
    Guiral M, Tron P, Aubert C, Gloter A, Iobbi-Nivol C, Giudici-Orticoni MT (2005) J Biol Chem 280:42004–42015Google Scholar
  32. 32.
    Ilert M, Bonnefoy V (2013) Biochim Biophys Acta 1827:161–175Google Scholar
  33. 33.
    Castelle C, Guiral M, Malarte G, Ledgham F, Leroy G, Brugna M, Giudici-Orticoni MT (2008) J Biol Chem 283:25803–25811Google Scholar
  34. 34.
    Malarte G, Leroy G, Lojou E, Abergel C, Bruschi M, Giudici-Orticoni MT (2005) Biochemistry 44:6471–6481Google Scholar
  35. 35.
    Castelle C, Ilbert M, Infossi P, Leroy G, Giudici-Orticoni MT (2010) J Biol Chem 285:21519–21525Google Scholar
  36. 36.
    Frew JE, Hill H (1988) Eur J Biochem 172:261–269Google Scholar
  37. 37.
    Armstrong FA, Hill HAO, Walton NJ (1988) Acc Chem Res 21:407–413Google Scholar
  38. 38.
    Page C, Moser C, Chen X, Dutton P (1999) Nature 402:47–52Google Scholar
  39. 39.
    Witt H, Malatesta F, Nicoletti F, Brunori M, Ludwig B (1998) Eur J Biochem 251:367–373Google Scholar
  40. 40.
    Jin B, Wang GX, Millo D, Hildebrandt P, Xia XH (2012) J Phys Chem C 116:13038–13044Google Scholar
  41. 41.
    Lojou E, Luciano P, Nitsche S, Bianco P (1999) Electrochim Acta 44:3341–3352Google Scholar
  42. 42.
    Lojou E, Bianco P (2000) J Electroanal Chem 485:71–80Google Scholar
  43. 43.
    Pardo-Yissar V, Katz E, Willner I, Kotlyar A, Sanders C, Lill H (2000) Faraday Discuss 116:119–134Google Scholar
  44. 44.
    Wegerig F, Giachetti A, Allegrozzi M, Lisdat F, Turano P (2013) J Biol Inorg Chem 18:429–440Google Scholar
  45. 45.
    Chen Y, Yang XJ, Guo LR, Jin B, Xia XH, Zheng LM (2009) Talanta 78:248–252Google Scholar
  46. 46.
    Wang G, Bao W, Wang M, Xia H (2012) Chem Commun 48:10859–10861Google Scholar
  47. 47.
    Correira dos Santos M, Paes de Sousa P, Simoes Goncalves M, Krippahl L, Moura J, Lojou E, Bianco P (2003) J Electroanal Chem 541:153–162Google Scholar
  48. 48.
    Abergel C, Nitschke W, Malarte G, Bruschi M, Claverie JM, Giudici-Orticoni MT (2003) Structure 11:547–555Google Scholar
  49. 49.
    Chi Q, Zhang J, Arslan T, Borg L, Pedersen G, Christensen H, Nazmudtinov R, Ulsrup J (2010) J Phys Chem B 114:5617–5624Google Scholar
  50. 50.
    Zhang J, Chi Q, Hansen Q, Jensen P, Salvatore P, Ulstrup J (2012) FEBS Lett 586:526–535Google Scholar
  51. 51.
    Raffalt A, Schmidt L, Christensen H, Chi Q, Ulsrup J (2009) J Inorg Biochem 103:717–722Google Scholar
  52. 52.
    Monari S, Battistuzzi G, Borsari M, Di Rocco G, Martini L, Ranieri A, Sola M (2009) J Phys Chem 113:13645–13653Google Scholar
  53. 53.
    Ciaccafava A, Alberola M, Hameury S, Infossi P, Giudici-Orticoni MT, Lojou E (2011) Electrochim Acta 56:3359–3368Google Scholar
  54. 54.
    Cracknell J, McNamara T, Lowe E, Blanford C (2011) Dalton Trans 40:6668–6675Google Scholar
  55. 55.
    Blanford C, Heath R, Armstrong F (2007) Chem Commun 17:1710–1712Google Scholar
  56. 56.
    Olejnik P, Palys B, Kowalczyk A, Nowicka A (2012) J Phys Chem C 116:25911–25918Google Scholar
  57. 57.
    Vignais P, Billoud B (2007) Chem Rev 107:4206–4272Google Scholar
  58. 58.
    Rüdiger O, Gutierrez-Sanchez C, Olea D, Pereira I, Velez M, Fernandez V, de Lacey A (2010) Electroanalysis 22:776–783Google Scholar
  59. 59.
    Lojou E (2011) Electrochim Acta 56:10385–10397Google Scholar
  60. 60.
    Lojou E, Luo X, Brugna M, Candoni N, Dementin S, Giudici-Orticoni MT (2008) J Biol. Inorg Chem 13:1157–1167Google Scholar
  61. 61.
    Millo D, Pandelia ME, Utesch T, Wisitruangsakul N, Mroginski MA, Lubitz W, Hildebrandt P, Zebger I (2009) J Phys Chem 113:15344–15351Google Scholar
  62. 62.
    Utesch T, Millo D, Castro A, Hildebrandt P, Zebger I, Mroginski M (2013) Langmuir 29:673–682Google Scholar
  63. 63.
    Luo XJ, Brugna M, Infossi P, Giudici-Orticoni MT, Lojou E (2009) J Biol Inorg Chem 14:1275–1288Google Scholar
  64. 64.
    Ciaccafava A, Infossi P, Ilbert M, Guiral M, Lecomte S, Giudici-Orticoni MT, Lojou E (2012) Angew Chem Int Ed 51:953–956Google Scholar
  65. 65.
    Ciaccafava A, De Poulpiquet A, Infossi P, Robert S, Gadiou R, Giudici-Orticoni MT, Lecomte S, Lojou E (2012) Electrochim Acta 82:115–125Google Scholar
  66. 66.
    Katz E (1994) J Electroanal Chem 365:157–164Google Scholar
  67. 67.
    Umena Y, Kawakami K, Shen JR, Kamiya N (2011) Nature 473:55–60Google Scholar
  68. 68.
    Kato M, Cardona T, Rutherford W, Reisner E (2013) J Am Chem Soc 135:10610–10613Google Scholar
  69. 69.
    Madden C, Vaughn M, Diez-Perez I, Brown K, King P, Gust D, Moore A, Moore T (2012) J Am Chem Soc 134:1577–1582Google Scholar
  70. 70.
    Bianco P, Haladjian J, Bruschi M, Guerlesquin F (1992) Biochem Biophys Res Commun 189:633–639Google Scholar
  71. 71.
    Lojou E, Bianco P (2004) Electroanalysis 16:1093–1100Google Scholar
  72. 72.
    Pieulle L, Morelli X, Gallice P, Lojou E, Barbier P, Czjzek M, Bianco P, Guerlesquin F, Hatchikian C (2005) J Mol Biol 354:73–90Google Scholar
  73. 73.
    Lojou E, Cutruzzola F, Tegoni M, Bianco P (2003) Electrochim Acta 48:1055–1064Google Scholar
  74. 74.
    Paes de Sousa P, Pauleta S, Simoes Goncalves M, Pettigrew G, Correia dos Santos M, Moura J (2007) J Biol Inorg Chem 12:691–698Google Scholar
  75. 75.
    Paes de Sousa P, Rodrigues D, Timóteo C, Simões Gonçalves M, Pettigrew G, Moura I, Moura J, Correia dos Santos M (2011) J Biol Inorg Chem 16:881–888Google Scholar
  76. 76.
    Fujita K, Hirasawa-Fujita M, Brown D, Obara Y, Ijima F, Kohzuma T, Dooley D (2012) J Inorg Biochem 115:163–173Google Scholar
  77. 77.
    Ferapontova H, Ruzgas T, Gorton L (2003) Anal Chem 75:4841–4850Google Scholar
  78. 78.
    Sarauli D, Ludwig R, Haltrich D, Gorton L, Lisdat F (2012) Bioelectrochemistry 87:9–14Google Scholar
  79. 79.
    Bagby S, Barker P, Guo L, Hill H (1990) Bioelectrochemistry 29:3213–3219Google Scholar
  80. 80.
    Burrows A, Guo L, Hill H, McLendon G, Sherman F (1991) Eur J Biochem 202:543–549Google Scholar
  81. 81.
    Lojou E, Pieulle L, Guerlesquin F, Bianco P (2002) J Electroanal Chem 523:150–159Google Scholar
  82. 82.
    Cambillaud C, Frey M, Mossé J, Guerlesquin F, Bruschi M (1988) Proteins Struct Funct Genet 4:63–70Google Scholar
  83. 83.
    Lojou E, Bianco P (2004) J Electroanal Chem 573:159–167Google Scholar
  84. 84.
    Seetharaman R, White S, Rivera M (1996) Biochemistry 35:12455Google Scholar
  85. 85.
    Heering H, Wiertz F, Dekker C, de Vries S (2004) J Am Chem Soc 136:11103–11112Google Scholar
  86. 86.
    Grumelli D, Mendez de Leo L, Bonazzola C, Zamlynny V, Calvo E, Salvarezza R (2010) Langmuir 26:8226–8232Google Scholar
  87. 87.
    Sezer M, Frielingsdorf S, Millo D, Heidary N, Utesch T, Mroginski MA, Friedrich B, Hildebrandt P, Zebger I, Weidinger IM (2011) J Phys Chem B 115:10368–10374Google Scholar
  88. 88.
    Jensen P, Chi Q, Zhang J, Ulstrup J (2009) J Phys Chem C 113:13393–14000Google Scholar
  89. 89.
    Koepke J, Olkhova E, Angerer H, Müller H, Peng G, Michel H (2009) Biochim Biophys Acta 1787:635–645Google Scholar
  90. 90.
    Iwata S, Ostermeier C, Ludwig B, Michel H (1995) Nature 376:660–669Google Scholar
  91. 91.
    Qin L, Hiser C, Mulichak A, Garavito RM, Ferguson-Miller S (2006) Proc Natl Acad Sci U S A 103:16117–16122Google Scholar
  92. 92.
    Haas A, Pilloud D, Reddy K, Babcock G, Moser C, Blasie J, Dutton P (2001) J Phys Chem B 105:11351–11362Google Scholar
  93. 93.
    Devadoss A, Burgess J (2002) Langmuir 18:9617–9621Google Scholar
  94. 94.
    Su L, Kelly J, Hawkridge F, Rhoten M, Baskin S (2005) J Electroanal Chem 581:241–248Google Scholar
  95. 95.
    Knoll W, Morigaki K, Naumann R, Sacca B, Schiller S, Sinner E (2004) In: Mirsky VM (ed) Ultrathin electrochemical chemo- and biosensors, technology and performance. Springer, Berlin, pp 239–254Google Scholar
  96. 96.
    Naumann R, Schmidt E, Jonczyk A, Fendler K, Kadenbach B, Liebermann T, Offenhäusser A, Knoll W (1999) Biosens Bioelectron 14:651–662Google Scholar
  97. 97.
    Rossi C, Chopineau J (2007) Eur Biophys J 36:955–965Google Scholar
  98. 98.
    Leitch J, Kunze J, Goddard J, Qchwan A, Faragher R, Naumann R, Knoll W, Dutcher J, Lipkowski J (2009) J Am Chem Soc 25:10354–10363Google Scholar
  99. 99.
    Friedrich M, Plum M, Santonicola M, Kirste V, Knoll W, Ludwig B, Naumann R (2008) Biophys J 95:1500–1510Google Scholar
  100. 100.
    Ataka K, Richter B, Heberle J (2006) J Phys Chem B 110:9339–9347Google Scholar
  101. 101.
    Schach D, Nawak C, Gennis R, Ferguson-Miller S, Knoll W, Walz D, Naumann R (2010) J Electroanal Chem 649:268–276Google Scholar
  102. 102.
    Nowak C, Schach D, Gebert J, Grosserueschkamp M, Gennis R, Ferguson-Miller S, Knoll W, Walz D, Naumann R (2011) J Solid State Electrochem 15:105–114Google Scholar
  103. 103.
    Infossi P, Lojou E, Chauvin JP, Herbette G, Brugna M, Giudici-Orticoni MT (2010) Int J Hydrog Energy 35:10778–10789Google Scholar
  104. 104.
    Jeuken L, Connell S, Nurnabi M, O’Reilly J, Henderson P, Evans S, Bushby R (2005) Langmuir 21:1481–1488Google Scholar
  105. 105.
    Weiss S, Bushby R, Evans S, Jeuken L (2010) Biochim Biophys Acta 1797:1917–1923Google Scholar
  106. 106.
    McMillan D, Marritt S, Firer-Sherwood M, Shi L, Richardson D, Evans S, Elliot S, Butt J, Jeuken L (2013) 135:10550–10556Google Scholar
  107. 107.
    Yuan Y, Zhou S, Zhang J, Zhuang L, Yang G, Kim S (2012) Electrochem Commun 18:62–65Google Scholar
  108. 108.
    Kundu A, Sau J, Redzwan G, Hashim M (2013) Int J Hydrog Energy 38:1745–1757Google Scholar
  109. 109.
    Gnana kumar G, Sarathi V, Nahm K (2013) Biosens Bioelectron 43:461–475Google Scholar
  110. 110.
    Abo-Hashesh M, Wang R, Hallenbeck P (2011) Bioresour Technol 102:8414–8422Google Scholar
  111. 111.
    Eriksen S, Riis M, Holm N, Iversen N (2011) Biotechnol Lett 33:293–300Google Scholar
  112. 112.
    Quéméneur M, Hamelin J, Benomar S, Giudici-Orticoni MT, Latrille E, Steyer JP, Trably E (2011) Int J Hydrog Energy 36:11654–11665Google Scholar
  113. 113.
    Willquist K, van Niel E (2010) Metab Eng 12:282–290Google Scholar
  114. 114.
    Kostesha N, Willquist K, Emneus J, van Niel E (2011) Extremophiles 15:77–87Google Scholar
  115. 115.
    Ikeda T, Kano K (2001) J Biosci Bioeng 92:9–18Google Scholar
  116. 116.
    Lojou E, Bianco P (2004) Electroanalysis 16:1113–1121Google Scholar
  117. 117.
    Lojou E, Durand M, Dolla A, Bianco P (2002) Electroanalysis 14:913–922Google Scholar
  118. 118.
    Pohorelic B, Voordouw J, Lojou E, Dolla A, Harder J, Voordouw G (2002) J Bacteriol 184:679–686Google Scholar
  119. 119.
    Gregory K, Bond D, Lovley D (2004) Environ Microbiol 6:596–604Google Scholar
  120. 120.
    Malvankar N, Vargas M, Nevin K, Franks A, Leang C, Kim B, Inoue K, Mester T, Covalla S, Johnson J, Rotelllo V, Tuominen M, Lovley D (2011) Nat Nanotechnol 6:573–579Google Scholar
  121. 121.
    Snider R, Strycharz-Glaven S, Tsoi S, Erickson J, Tender L (2012) Proc Natl Acad Sci U S A 109:15467–15472Google Scholar
  122. 122.
    Jiang X, Hu J, Fitzgerald L, Biffinger J, Xie P, Ringeisen B, Lieber C (2010) Proc Natl Acad Sci U S A 107:16806–16810Google Scholar
  123. 123.
    Lovley D (2012) Biochem Soc Trans 40:1186–1190Google Scholar
  124. 124.
    Rabaey K, Rodriguez J, Blachkall L, Keller J, Gross P, Batstone D, Verstraete W, Nealson K (2007) ISME J 1:9–18Google Scholar
  125. 125.
    Smith J, Lovley D, Tremblay P (2013) Appl Environ Microbiol 79:901–907Google Scholar
  126. 126.
    Rosenbaum M, Aulenta F, Villano M, Angenent L (2011) Bioresour Technol 102:324–333Google Scholar
  127. 127.
    Rozendal R, Jeremiasse A, Hamelers H, Buisman C (2008) Environ Sci Technol 42:629–634Google Scholar
  128. 128.
    Aulenta F, Catapano L, Snip L, Villano M, Majone M (2012) ChemSusChem 5:1080–1085Google Scholar
  129. 129.
    Yu L, Duan J, Zhao W, Huang Y, Hou B (2011) Electrochim Acta 56:9041–9047Google Scholar
  130. 130.
    Armstrong F, Belsey N, Cracknell J, Goldet G, Parkin A, Reisner E, Vincent K, Wait A (2009) Chem Soc Rev 38:36–51Google Scholar
  131. 131.
    Lojou E, Bianco P (2006) Biogeosciences 69:237–247Google Scholar
  132. 132.
    Carbajosa S, Malki M, Caillard R, Lopez R, Palomares F, Martin-Gago J, Rodriguez N, Amils R, Fernandez V, De Lacey A (2010) Biosens Bioelectron 26:877–880Google Scholar
  133. 133.
    Erable B, Etcheverry L, Bergel A (2009) Electrochem Commun 11:619–622Google Scholar
  134. 134.
    Wei V, Elektorowicz M, Oleszkiewicz J (2011) Water Res 45:5058–5062Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • M. Roger
    • 1
  • A. de Poulpiquet
    • 1
  • A. Ciaccafava
    • 1
  • M. Ilbert
    • 1
  • M. Guiral
    • 1
  • M. T. Giudici-Orticoni
    • 1
  • E. Lojou
    • 1
  1. 1.Unité de Bioénergétique et Ingénierie des Protéines, UMR7281-FR3479, Centre National de la Recherche ScientifiqueAix Marseille UniversitéMarseilleFrance

Personalised recommendations