Bioelectrochemical systems: fundaments and classification

Bioelectrochemical systems (BES) are based on the ability of some microorganisms to catalyze different electrochemical reactions, specifically, reactions that involve electron transfer, such as redox (^{Rabaey et al., 2007}; ^{Rozendal et al., 2008}). In the literature, they are also called microbial electrochemical technologies (METs) (^{Logan and Rabaey, 2012}). Depending on their mode of operation and their application, they are classified into four large groups: microbial fuel cells (MFC) for generating electricity, microbial electrolysis cells (MEC) to produce mainly inorganic compounds, such as hydrogen, microbial electrosynthesis cells (MES) to synthesize organic chemical compounds, and microbial desalinization cells (MDC) to desalinize water in combination with other functions (^{Wang and Red, 2013}). Among these systems, MFC are the most studied and are the model for all others (microbial X cells, MXC), the X representing different applications (^{Harnisch and Schröder, 2010}).

Microbial fuel cells

Microbial fuel cells (MFC) are devices that use microorganisms to oxidize organic and inorganic matter to generate electric energy (^{Logan et al., 2006}). Like all bioelectrochemical systems, they consist of two electrodes, an anode and a cathode, which are joined by an external wire to complete an electric circuit. Moreover, the compartments, or chambers, containing the electrodes are separated by a proton exchange membrane (^{Bond and Lovley, 2003}).

In the anode chamber, the microorganisms grow and oxidize the substrate available under anaerobic conditions; they release electrons, protons and CO₂ into the medium. The protons move into the cathode chamber passing through the proton exchange membrane and, at the same time, the electrons are pumped to the anode because some microorganisms are able to transfer electrons outside the cell (known as extracellular electron transfer, EET) (^{Rabaey, 2010}). Once in the anode, they travel to the cathode through the external wire that connects the electrodes. When the electrons arrive at the cathode under aerobic conditions, they combine with the protons to reduce oxygen molecules to form water. This creates an electron flow (electric current) (^{Rabaey and Verstraete, 2005}; ^{Logan and Regan, 2006a}; ^{Lovley, 2006}). When oxygen or other receptor molecules, such as nitrates or sulfates, are present in the cathode chamber, an electric current is produced, but if they are not present, generation is not spontaneous. The electrons that reach the cathode reduce a molecule to complete the redox process (that begins in the anode) and to stimulate generation of an electric current (^{Pant et al., 2012}). The microorganisms that generate electric energy are referred to as exoelectrogenic because of their capacity to transfer electrons out of the cell (^{Logan and Regan, 2006b}).

When, instead of using cells to produce electricity, the process is reversed and an electric current from an external source is applied, the second type of bioelectrochemical systems develops; these are known as microbial electrolysis cells.

Microbial electrolysis cells

Microbial electrolysis cells (MEC) are a type of bioelectrochemical system to which electric energy is provided to achieve a given process or formation of chemical products, mainly inorganic, such as hydrogen, hydrogen peroxide and sodium hydroxide, among others (^{Logan, 2008}). They function like MFC since MEC are modified MFC and were designed primarily to produce H₂ (^{Liu et al., 2005a}).

The system consists of using the electrons that arrive at the cathode, as occurs in an MFC, but the aim is to combine them with the protons to produce H₂. Therefore, the system must be under anaerobic conditions so that the electrons and protons do not combine with oxygen. This reaction is not spontaneous; it needs some external energy as well as that generated by the bacteria to carry out the reaction (^{Logan and Grot, 2006}; ^{Rozendal et al., 2006}). This process is known as electrohydrogenesis or microbial electrolysis (^{Cheng and Logan, 2007}).

The voltage, also denominated potential difference, is the pressure or force with which the electrons are driven toward the cathode. It is supplied through a direct current power source or a potentiostat. For the production of hydrogen, the voltage generated by the microorganisms in the anode is not enough to drive the reaction. For this reason, external voltage must be applied to the system. The calculations to determine the necessary voltage are based on Gibbs free energy for redox reaction (^{Call and Logan, 2008}; ^{Logan et al., 2008}; ^{Rozendal et al., 2008}).

Biodegradable organic matter that can be used in these systems is variable, from simple molecules, such as acetate, glucose, starch or cellulose, and even complex mixtures like those present in wastewater from different industries (^{Pant et al., 2010}).

The most studied bioelectrochemical systems are MFC and MEC. One important characteristic is the absence of microorganisms in the cathode. Nevertheless, ^{Hamelers et al. (2010)} defined bioelectrochemical systems specifically as emerging technologies that use microorganisms in the anode to catalyze oxidizing reactions and in the cathode for reducing reactions. That is, there may be microorganisms in the two electrodes at the same time. When microorganisms are present in the anode chamber, they transfer electrons to the anode. When microorganisms are in the cathode chamber, transfer occurs in the opposite direction; the cathode gives up electrons to the microorganisms (^{Gregory et al., 2004}). Consumption and use of these electrons is focused on stimulating and modifying microbial metabolism (^{Thrash and Coates, 2008}). This process gives rise to the third type of bioelectrochemical system.

Microbial electrosynthesis cells (MES)

Microbial electrosynthesis cells (MES) are a type of bioelectrochemical system based on microbial electrosynthesis, a term coined by ^{Nevin et al. (2010)} to refer to the process in which electricity is the energy source for some microorganisms to synthesize organic compounds from CO₂. Some examples of compounds obtained with this technique are acetate (^{Nevin et al., 2010}), butyrate (^{Ganigué et al., 2015}) and methane (^{Cheng et al., 2009}).

Later on, the term microbial electrosynthesis was also used to refer to the synthesis of organic compounds from substrates different from CO₂, such as glucose or glycerol (^{Rabaey and Rozendal, 2010}). This technique was already known also as electrofermentation. Now, it is the term that several authors continue to use as more adequate to differentiate this process from CO₂ microbial electrosynthesis (^{Kracke and Krömer, 2014}; ^{Rosenbaum and Franks, 2014}; ^{Harnisch et al., 2015}).

Electrofermentation

This technique manipulates the patterns of a fermentation process by applying electric current to the culture medium to modify end products (^{Rabaey and Rozendal, 2010}). Each microorganism has a specific metabolism based on an electron flow that determines how fermentation is carried out. If this flow is altered by the modulating action of another electron flow (in the form of an electric current), the existing metabolic pathways are redirected causing an increase or decrease in fermentation products. Moreover, when the electron flow changes, the carbon flow also changes (^{Logan and Rabaey, 2012}).

The energy needed to accelerate or divert a fermentation pathway depends on the substrate and the product of interest. It should be low enough to not cause cell death, but high enough to stimulate changes in microbial metabolism. Some microorganisms have the capacity to accept electrons, directly or indirectly. This fact can be used for different ends, such as wastewater treatment, carbon fixation, formation of chemical compounds, or bioremediation (^{Rosenbaum et al., 2011}).

Electric current can act as a source of reducing power, or reducing equivalents, which rapidly regenerates the coenzymes NAD and NADP to form NADH and NADPH, respectively. The abundance of these molecules in the cells represents an effective way to increase the fermentation yield (^{Kim and Kim, 1988}; ^{Park and Zeikus, 1999}). Another manner in which electric current can contribute to changing metabolic pathways is by forming H₂, as occurs in microbial electrolysis cells. In the process of fermentation, H₂ is an electron donor, and there is no direct interaction between electrode and microorganisms (^{Steinbusch et al., 2010}; ^{Harnisch et al., 2015}). Accumulation of H₂ increases partial pressure in a fermenting bacterial culture. Consequently, there will be a change in the electron balance and, therefore, in microbial metabolism (^{Yerushalmi et al., 1985}).

Mechanisms of electron transfer

Mechanisms by which microorganisms transfer or receive electrons toward or from an electrode are:

Manipulation of microbial metabolism by electrofermentation

Bioelectrochemical systems have been used to increase synthesis of a variety of fermentation products. The first study was conducted by ^{Hongo and Iwahara (1979a)}. These authors developed a method they called electro-energizing fermentation (EEF) and today is known as electrofermentation. This methodology consists of applying 1.5 V to a culture of Brevibacterium flavum through a platinum electrode with the objective of accelerating its reductive metabolism. In electro-energizing fermentation, as well as in the control, glucose was used as the substrate and neutral red as the redox mediator. The result was production of L-glutamic acid (51 mg mL^-1), with an increase of 15 % in yield, relative to the control fermentation (44.3 mg mL^-1). The authors suggested that the increase in production seemed to be stimulated by the reducing action of the cathode that transferred electrons to the microorganisms (^{Hongo and Iwahara, 1979b}).

^{Kim and Kim (1988)} also used the electro-energizing method to manipulate fermentation of the bacterium Clostridium acetobutylicum and increase butanol production from glucose. They inoculated the microorganisms in the cathode and applied -2.5 V and methyl viologen to the culture as the redox mediator. They did not observe a difference in substrate consumption or cell growth relative to the control, but butanol production did increase 26 % (93.7 mmol), relative to the control (74.6 mmol). Simultaneously, there was a decrease of 25 % in acetone production (37.8 mmol in the control and 28.4 mmol in the bioelectrochemical system).

^{Park and Zeikus (1999)} showed that the reducing power of an electric current can be used to manipulate fermentation of Actinobacillus succinogenes with glucose as substrate and neutral red as the redox mediator. In the cathode chamber, they inoculated the microorganisms with a voltage of 2 V through graphite cloth electrodes. They observed that the reducing power increased glucose consumption, cell growth and succinate production by 20 %, and reduced acetate production (50 %, relative to the controls). The authors showed that the redox mediator neutral red bonds with the enzyme fumarate reductase and transfers the electrons from the electrode to the cell and, thus, the enzyme reduces fumarate to succinate.

Ethanol production by Clostridium thermocellum and Saccharomyces cerevisiae was also manipulated by a bioelectrochemical system with cellulose and glucose, as substrates, and neutral red (^{Shin et al., 2002}). In that study, ethanol increased 61 % with C. thermocellum. The control culture produced 1.04 g L^-1, and fermentation with -1.5 V yielded 1.68 g L^-1. The increase using S. cerevisiae was less but was also significant, 46.7 g L^-1 to 52.5 g L^-1 ethanol, equivalent to 12 %. In contrast, acetate production decreased with both microorganisms relative to the controls (^{Shin et al., 2002}).

Change in lactate production with Corynebacterium glutamicum was accomplished in a bioelectrochemical reactor with a cathode regulated to -0.6 V and anthraquinone-2,6-disulfonate as redox mediator. Lactate concentration increased from 1.10 mol of product per mol of glucose to 1.62 mol of product per mol of glucose (^{Sasaki et al., 2014}). Another research group was able to modify fermentation of Clostridium pasteurianum using a potential difference of 0.045 V across the system, but without adding redox mediators to the culture medium. Butanol production increased to 13.5 mmol relative to 5.4 mmol when fermentation was without electricity (^{Choi et al., 2014}).

^{Harrington et al. (2015)} observed the effect of electric current on Klebsiella pneumoniae. They reported an increase of 93 % in ethanol production simultaneous to the 76 % increase in lactate, with neutral red as redox mediator and applying -0.65 V through the system. Ethanol concentration in the control was 9.61 mmol and that of lactate was 2.66 mmol, contrasting with the concentration in electrofermentation, which was 22.34 mmol ethanol and 5.64 mmol lactate. In theory, any fermentation metabolism can be manipulated by electrochemical supply of reduced equivalents, which change the NAD/NADH molar ratio, whether in the presence or absence of an exogenous redox mediator (^{Peguin et al., 1994}).

Bioelectrochemical methods for increasing propionate production

Anaerobic bacteria of the genus Propionibacterium produce propionate through fermentation of glucose or lactate. As secondary metabolites they also generate acetate and CO₂ in a molar ratio of 2:1:1. However, the concentration of the final products may vary because of diverse factors of the culture, the bacterial strain and the substrate (^{Piveteau, 1999}).

The dairy product industry has been greatly interested in the study of manipulating fermentation by propionogenic bacteria since the quantity of metabolites of these bacteria affect the flavor of Swiss cheese (^{Hettinga and Reinbold, 1972}). It is also of great importance in animal production because propionate is the most important quantitative gluconeogenic precursor in ruminant metabolism (^{Huntington et al., 2006}).

The first study to enhance propionate concentration in a bacterial culture medium administrated hydrogen to increase the partial pressure of the Propionispira arboris growing medium. P. arboris is a Gram-negative, nitrogen-fixing bacterium that releases propionate, acetate and CO₂ as major end products of carbohydrate fermentation. Two atmospheres of hydrogen in the medium drastically changed the propionate:acetate molar ratio from 2:1 to 16:1. Thus, propionate increased in concentration as practically the sole end product (12.6 mmol propionate and 0.8 mmol acetate), while the control produced 11.2 mmol propionate and 5.8 mmol acetate. Excess hydrogen altered the carbon and electron flow, preventing pyruvate from transforming into acetate and CO₂ (^{Thompson et al., 1984}).

^{Emde and Schink (1990)} produced the first bioelectrochemical system as such for increasing propionate concentrations in a bacterial culture. They tested a system of electrodes to change fermentation patterns of Propionibacterium freudenreichii with glucose as substrate using a three-electrode amperometric culture system: one working electrode connected to a potentiostat that applies a potential difference, a reference electrode and a counter electrode through which the current flows and can be recorded. In this culture system, four different exogenous mediators were tested with the bacteria: anthraquinone-2,6-disulfonic acid (AQ), cobalt sepulcrate (CoS), benzyl viologen and methyl viologen, with anodic potential 40 mV more negative than the standard redox potential of each mediator.

During the first trials, ^{Emde and Schink (1990)} observed that both benzyl and methyl viologen inhibited bacterial growth and no electron transfer or propionate production was registered. This was likely due to depolarization of the cell membrane caused by these non-polar compounds. In contrast, with AQ and CoS, large quantities of electrons were transferred when propionate production increased. In the presence of AQ, 90 % of all products formed were propionate with a concentration of 475 μM and 53 μM acetate, while with CoS, 624 μM propionate and 17 μM acetate were obtained. These quantities indicated that propionate represented 97.3 % of the end products.

^{Schuppert et al. (1992)} continued the studies of ^{Emde and Schink (1990)}, focusing on increasing propionate production during fermentation of a medium with whey permeate and Propionibacterium acidipropionici. Based on previous findings, they used CoS as the redox mediator and a platinum electrode regulated to a potential of -0.47 V. By applying electric potential in fed-batch culture, the lactose of the whey permeate fermented to produce 70 mmol propionate with no acetate production. This means obtaining 100 % propionate in the culture medium, relative to the control, which produced 49 mmol propionate and 23 mmol acetate. Nevertheless, in experiments even without adding redox mediators in continuous cultures, there was electron transfer from the electrode to the microorganisms. This indicated that the redox mediator may not be necessary, and if so, costs of the continuous mode scaling process would decrease.

^{Wang et al. (2008)} tested whether Propionibacterium freudenreichii ET-3 was able to use, as a redox mediator, a bifidobacterium growth stimulator, 1,4-dihydroxy-2-naphthoic acid, it secretes into the culture medium. This molecule has the redox activity to regenerate NAD molecules. Their objective was to develop a bioelectrochemical system by applying 0.4 V without adding a redox mediator and observe its effects on propionate production in glucose fermentation. Their most important finding was a change in the molar ratio of acetate:propionate from 2:3 (11.5 mmol acetate and 17.7 mmol propionate) to 1:1 (13.6 mmol acetate and 13.7 mmol propionate). The decrease in the propionate production was probably caused by a decrease in electrons available for its formation and the increase in oxidation of the substrates to acetate and CO₂. The results indicated that those conditions are not adequate for increasing propionate.

Bioelectrochemical methods for the study of ruminal fermentation

The first study on bioelectrochemical processes in ruminal fermentation focused on the possibility of generating electricity in MFC with ruminal microorganisms as biocatalyzers. ^{Rismani-Yazdi et al. (2007)} reported the use of ruminal microbial communities to convert cellulose into electric energy. With two-chamber graphite electrode bioelectrochemical systems, they used a mineral medium supplemented with clarified ruminal liquid to stimulate the ruminal habitat and provide the microorganisms with growth factors. They inoculated the medium with ruminal microorganisms in an anaerobic environment and maintained the cells in optimum ruminal conditions. The researchers showed that, ruminal microbiota can in fact be used to generate electricity from cellulose since it was capable of transferring electrons to an electrode producing a constant electric current for at least 60 days without adding a redox mediator. Interested in determining the composition of the microbiota present in the cells, they conducted a phylogenetic analysis of the microorganisms based on 16s rRNA gene sequence comparisons. Among the main genera identified were Firmicutes, Clostridium, Sedimentibacter, Desulfotomaculum and Ruminococcus, bacteria that hydrolyze lignocellulosic biomass via a complex cellulase system known as cellulosome. The researchers suggest that this system can generate electricity from a diversity of substrates rich in cellulosic wastes (^{Rismani-Yazdi et al., 2007}).

MFC can also be used as a tool in the study of ruminal microorganisms and their physiological functions. One of the most important aspects is how to manipulate the relation between methane and VFAs. Methane production accounts for 12 % of ruminant energy loss, with a direct impact on fermentation efficiency. Moreover, it is a greenhouse gas and livestock produces 44 % of anthropogenic methane emissions (Pinos-Rodríguez et al., 2012; ^{Gerber et al., 2013}). Among the ruminal microbiota are methanogenic archaea, microorganisms that produce methane from H₂ and CO₂. Their function is very important for adequate ruminal fermentation. If H₂ accumulates in the rumen, partial pressure increases, the function of the proteins involved in electron transport is inhibited, feed digestibility decreases, and fermentation becomes unbalanced (^{Wolin et al., 1997}). For this reason, methane production is necessary in spite of the disadvantages of its emission.

MFC can act positively as an alternative for using the hydrogen produced in the rumen as an electron donor for the electrogenic microbial communities while favoring the conditions to increase acetate and propionate production. They also compete with methanogenic archaea for hydrogen (^{Bretschger et al., 2009}). ^{Ishii et al. (2008)} showed that in an MFC, methanogenesis is inhibited, from 0.128 mmol d^-1 methane in a control fermentation to 0.009 mmol d^-1 inside MFC during the first 30 h. Moreover, acetate and propionate concentrations increased temporarily in soil samples that originally produced high methane concentrations.

^{Wang et al. (2012)} studied the effect of degrading straw, as the substrate, by ruminal microorganisms on production of VFAs and electricity. For these studies, the MFC also consisted of two-chamber cells and the electrodes used were graphite plates. They observed that the total VFA concentration increased rapidly after inoculation with ruminal microorganisms. After reaching a certain high concentration, however, it began to decrease. The authors explained that the microorganisms, after a certain concentration, oxidized VFAs and continued to generate electricity because there was no other substrate available.

Our group is currently developing a bioelectrochemical system to assess changes in in vitro fermentation by ruminal microorganisms with direct electric current. We have observed modifications in the fermentation patterns to increased production of the volatile fatty acids acetate, propionate and butyrate (unpublished data). Our group is also studying how to increase ruminal propionic acid at the expense of decreasing acetic acid. However, we have found that electric energy affects production of the three VFAs and not just of propionate, as was expected at the beginning. We have also tested the effect of a redox mediator on ruminal fermentation and various levels of potential difference, with which we determined that our system did not require external mediators. Our data show evidence that it is possible to increase in vitro production of VFAs, and the systems can be modified for their in vivo application with more research and technology (^{Aguilar-Glez et al. Unpublished data}).

In vivo application of bioelectrochemical systems

The gap between in vitro and in vivo studies is one of the greatest challenges for scientific research. Bioelectrochemical systems should be optimized in their design to be administrated to animals orally so that they reach the rumen and remain there without passing to other organs of the digestive system. The factors to consider are the following:

The new challenges in this area consist of creating effective systems that allow the microorganisms to perform optimally as biocatalyzers and to produce compounds of interest in significant concentrations. Another challenge is to reach full understanding of how they use the reducing power administrated by electric energy to modify their metabolic pathways and how they interact with electrodes surfaces.