Soluble Iron Enhances Extracellular Electron Uptake by Shewanella oneidensis MR-1

Abstract

Extracellular electron transfer (EET) is a process that microorganisms use to reduce or oxidize external insoluble electron acceptors or donors. Much of our mechanistic understanding of this process is derived from studies of transmembrane cytochrome complexes and extracellular redox shuttles that mediate outward EET to anodes and external electron acceptors. In contrast, there are knowledge gaps concerning the reverse process of inward EET from external electron donors to cells. Here, we describe a role for soluble iron (exogenous FeCl₂) in enhancing EET from cathodes to the model EET bacterium Shewanella oneidensis MR-1, with fumarate serving as the intracellular electron acceptor. This iron concentration-dependent electron uptake was eradicated upon addition of an iron chelator and occurred only in the presence of fumarate reductase, confirming an electron pathway from cathodes to this periplasmic enzyme. Moreover, S. oneidensis mutants lacking specific outer membrane and periplasmic cytochromes exhibited significantly decreased current levels relative to wild-type. These results indicate that soluble iron can function as an electron carrier to the EET machinery of S. oneidensis.

Graphical Abstract

Driving electron into cells Soluble iron is found to enhance electron uptake by Shewanella oniedensis MR-1. This enhancement is attributed to freely diffusing iron ions where Fe³⁺/Fe²⁺ is continuously oxidized/reduced by the cells/electrodes.

Introduction

Some microorganisms, such as the metal-reducing bacterium Shewanella oneidensis MR-1, are capable of performing extracellular electron transfer (EET), a respiratory strategy that allows them to access external electron acceptors such as redox-active elements (e.g. Fe, Mn) in solid minerals.^[1,2] Importantly, this process can also be exploited in bioelectrochemical technologies, such as microbial fuel cells, and living electronics that ‘wire’ microbes to anodes.^[3,4] In the case of S. oneidensis, the challenge of outward EET is overcome by a network of cytochromes (collectively referred to as the Mtr pathway) that bridge the otherwise electrically insulating cell envelope. This network includes the inner membrane tetraheme cytochrome CymA, periplasmic cytochromes including the small tetraheme cytochrome (CctA) and flavocytochrome fumarate reductase FccA, the transmembrane porin-cytochrome complex MtrAB, and the cell surface decaheme cytochromes MtrC and OmcA that function as electron conduits to the outside world.^[5-7] When these cell surface conduits transfer electrons to external surfaces (e.g. minerals or electrodes) through direct contact, this process is referred to as direct EET. EET can also proceed to external surfaces indirectly via soluble diffusing redox-active shuttles, such as flavins.^[8,9] Flavins can also play a role as cytochrome-bound cofactors that enhance EET rates.^[10] Interestingly, Oram and Jeuken recently provided evidence that soluble iron may also serve as an extracellular mediator linking the cellular EET machinery to anodes by redox cycling.^[11]

EET is not only an outward (cell-to-surface) phenomenon. A number of microorganisms, such as Fe(II) oxidizing bacteria, can acquire electrons from the exterior of the cell and derive energy from this inward EET.^[12-14] Other microbes are proposed to utilize direct electron uptake, such as corrosive sulfate reducing bacteria. ^[15] For example, the sulfate reducing bacteria Desulfovibrio ferrophilus IS5 and Desulfopila corrodens IS4 have been shown to directly interact with electrodes and demonstrated electron uptake from cathodes.^[16-18] This ability is crucial for microbial electrosynthesis technologies, where EET from cathodes drives reduction of CO₂ and production of biofuels or other chemicals.^[19,20] The underlying mechanisms of this cathodic extracellular electron uptake remain less understood than its anodic counterpart, but also include mediated EET via small molecules (e.g. H₂) and direct EET via porin-cytochrome complexes analogous to those described in Shewanella.^[21,22] Interestingly, some bacteria appear to be capable of bi-directional EET. S. oneidensis, for example, is capable of directing electrons from a cathode to intracellular electron acceptors such as fumarate or O₂, allowing regeneration of ATP and NADH.^[7,23] This process appears to rely, at least partially, on some of the same molecules that facilitate outward EET; outer membrane multiheme cytochromes play a role in mediating inward EET into S. oneidensis,^[23] and flavins have likewise been observed to play a role.^[7,8] However, previous studies suggest that additional enigmatic components, beyond the canonical inward EET pathway, may play a role in directing electrons from cathodes into S. oneidensis.^[24]

Recent evidence of oxidation of dissolved Fe²⁺ by an outer membrane cytochrome common to iron oxidizing bacteria,^[25] and the recent report that redox cycling of dissolved iron (Fe³⁺/Fe²⁺) may play a role in mediating EET from S. oneidensis to anodes,^[11] motivated us to consider whether dissolved iron can also play a role in inward EET from cathodes to S. oneidensis. Here we present electrochemical measurements of enhanced electron uptake by S. oneidensis in the presence of exogenously added FeCl₂ and the terminal electron acceptor fumarate. Electron uptake from the cathode is coupled to fumarate reduction by the fumarate reductase FccA, and outer membrane cytochromes play an important role in the inward EET pathway, as indicated by the significantly reduced current levels in S. oneidensis mutants deficient in cytochromes.

Results and Discussion

Iron (II) chloride enhances extracellular electron uptake by S. oneidensis.

In order to investigate the effect of soluble iron on extracellular electron uptake by S. oneidensis, we performed electrochemical analyses in single chamber bioelectrochemical reactors (Figure 1A) in the presence and absence of 1 mM FeCl₂ with the electrode serving as the sole electron source. In brief, following an anodic cultivation step in anaerobic reactors (with lactate as the electron donor and the electrode as electron acceptor), to promote biofilm formation, the electrodes were rinsed and transferred to new reactors with fresh media for cathodic measurements. The cathodic reactors included fumarate as the electron acceptor but no lactate (or other exogenous carbon source) to ensure that the cathodes serve as the sole electron donor. From S. oneidensis chronoamperometry measurements, a sustained and increasing cathodic current was immediately observed upon addition of FeCl₂ into reactors with working graphite electrodes poised at −305 mV (vs SHE) and in the presence of 30 mM fumarate as a cellular electron acceptor (Figure 1B). This cathodic current, which reached as high as −271.5 ± 41.8 μA in less than an hour, was absent in a control including cells but no added FeCl₂, and an abiotic control with FeCl₂ (with current levels ~ −0.3 μA for both). It is important to note that S. oneidensis has been previously shown to be capable of direct and flavin-mediated electron uptake from electrodes.^[7,8,23] Our observation of low current densities (in the absence of exogenous FeCl₂) reflects different experimental conditions, as previous studies were performed either in aerobic conditions^[23] or with higher cell densities and experimenting with added flavins.^[7]

Figure 1:

Addition of FeCl₂ enhances cathodic electron uptake by Shewanella oneidensis MR-1. (A) Schematic of the bioelectrochemical reactor. The reactor consisted of a stoppered glass bottle with a graphite working electrode (WE), platinum counter electrode (CE), Ag/AgCl reference electrode (RE), and N₂ ports. (B) Chronoamperometry in anaerobic cathodic conditions with 30 mM fumarate as the electron acceptor and the working electrode poised at −305 mV (vs SHE) in the presence or absence of 1 mM FeCl₂. 1 mM FeCl₂ was added at the time point indicated by the black arrow. Error bars indicate standard error of triplicate measurements. (C) Representative cyclic voltammetry (1 mV/s scan rate) of S. oneidensis with 30 mM fumarate as the electron acceptor in the presence or absence of 1 mM FeCl₂.

Following this period of chronoamperometry in the presence of FeCl₂, cyclic voltammetry scan was performed to assess the electrochemical interaction between cathodes and S. oneidensis cells. In the presence of FeCl₂ and fumarate, cyclic voltammetry revealed a clear catalytic wave with an onset potential of ~0 mV vs. SHE, a feature absent from a control without FeCl₂ (Figure 1C). This onset potential is consistent with the redox characteristics of FeCl₂ in the abiotic (cell-free) control, which revealed quasi-reversible electron transfer behavior (Figure S1). Fluorescence microscopy of the electrode-bound biofilms showed (Figure S2) similar cell density both in the presence and absence of added FeCl₂, which reflects the anodic biofilm formation step done prior to cathodic measurements (see Methods) and indicating that the observed enhancement in cathodic current is the result of improved electron transfer rather than differences in biomass.

The amperometry and voltammetry measurements described above demonstrate that addition of soluble ferrous iron significantly enhances (and can account for almost all the) extracellular electron transfer from graphite electrodes to S. oneidensis under the experimental conditions tested here. Although cathodic current started increasing immediately upon iron addition and saturated in ~30 minutes, we considered the possibility that soluble iron addition may have contributed to increased electron uptake indirectly by impacting the expression of cytochromes. However, TMBZ heme-staining SDS-PAGE gels of protein content, from cells harvested after the anodic cultivation step and re-suspended in the same media as the cathodic reactors, revealed no discernible difference in cytochrome production between FeCl₂ addition and 30 minutes after FeCl₂ addition (Figure S3). Previous studies under comparable conditions (anaerobic with fumarate as the electron acceptor) highlighted the role that added flavins can play as soluble electron shuttles to enhance cathodic electron uptake.^[7,8] Our results suggested that redox cycling of dissolved iron could play a similar role, which next motivated us to examine the effect of added FeCl₂ concentration and specific interactions with cellular redox components.

Fe-enhanced cathodic electron uptake in S. oneidensis is concentration dependent and due to free Fe ions.

Increases in EET mediated by soluble redox shuttles are expected to depend on the concentration of the shuttle, as previously observed for riboflavin-enhanced EET in S. oneidensis.^[26] To that end, we investigated the concentration dependence of Fe-enhanced electron uptake by S. oneidensis in the presence of 0.2, 0.6, and 1 mM FeCl₂ and with 30 mM fumarate serving as electron acceptor. Since Fe²⁺ is toxic to cells at high concentrations, 1 mM FeCl₂ was chosen as an upper bound that is tolerated by S. oneidensis.^[27] Once again, a period of chronoamperometry (Figure 2A) was followed up with cyclic voltammetry (Figure 2B) to assess the impact of each FeCl₂ concentration on sustained inward EET and the resulting catalytic waves, in the presence of fumarate as a cellular electron acceptor. F

Figure 2:

The FeCl₂-enhanced cathodic current into S. oneidensis is concentration dependent. (A) Chronoamperometry (−305 mV vs SHE) of S. oneidensis with increasing concentration of FeCl₂: 0 mM, 0.2 mM, 0.6 mM and 1 mM. 30 mM fumarate served as the electron acceptor. Error bars indicate standard error of triplicate measurements. (B) Representative cyclic voltammetry scan (at 1 mV/s) of S. oneidensis with increasing concentration of FeCl₂.

From chronoamperometry, the observed cathodic current was found to depend very strongly on FeCl₂ concentration. While the response to 0.2 mM FeCl₂ was similar to a no-FeCl₂ control, modest current was detected at 0.6 mM FeCl₂, and an order of magnitude enhancement was seen at the 1 mM FeCl₂ level (Figure 2A). Corresponding cyclic voltammograms are shown in Figure 2B. While catalytic activity was not observed in the absence of FeCl₂, the catalytic waves were better defined and increased sharply in magnitude as a function of increasing FeCl₂ concentration (Figure 2B). The strong dependence of catalytic activity on FeCl₂ concentration, and the requirement for millimolar concentration to achieve appreciable inward EET are consistent with a redox shuttling mechanism where Fe³⁺/Fe²⁺ is continuously oxidized/reduced by the cells and electrode respectively. The inherently small diffusion coefficients of redox shuttles are indeed known to limit outward EET, which necessitates high shuttle concentrations to maximize the concentration gradient that drives diffusion.^[28,29] It is worth noting however that iron is an essential trace nutrient and its homeostasis is regulated by S. oneidensis using both import^[30] and export^[27] systems which could potentially influence the concentration available for EET mediation.

To further confirm that the Fe-dependent inward EET to S. oneidensis stems specifically from freely diffusing Fe ions in solution, we performed experiments with added free iron chelator, deferoxamine myselate salt (DFO). Following the approach suggested by Oram and Jeuken,^[11] we hypothesized that if iron acts as a free soluble shuttle, DFO will abolish the enhanced electron uptake phenotype. Indeed, the observed cathodic current by S. oneidensis (at −305 mV vs SHE in the presence of FeCl₂) was immediately eradicated upon the addition of 1.5 mM DFO (Figure 3A). Likewise, the catalytic wave detected by cyclic voltammetry was diminished in DFO containing reactors (Figure 3B). While DFO is known to chelate ferric iron specifically, it can also influence the oxidation of iron prior to its complexation.^[31] Such a change in the redox background of the experiment may underlie the appearance of a transient anodic response after DFO addition, which occurred even in abiotic control reactors containing 1 mM FeCl₂ without cellular cathodic activity (Figure 3A). Regardless of this effect, it is clear that chelating iron (specifically free extracellular iron, since DFO cannot chelate heme iron from protein) had a significant effect in eradicating the observed iron-mediated electron uptake. To investigate whether DFO itself has a toxic impact on cells, which may contribute to the eradication of cellular cathodic activity, we performed colony forming units (CFU) counts before and 30 minutes after DFO addition to cells in the same media conditions used for cathodic measurements. This comparison revealed no difference in cell viability as a result of DFO exposure (Figure S4), a finding consistent with previous studies showing that DFO exhibits low membrane permeability, is produced by certain bacterial species for iron acquisition, and can alleviate the toxic effects of iron overload.^[32,33] The DFO addition experiments provide further evidence that the enhanced inward EET into S. oneidensis is due specifically to freely diffusing iron ions in solution.

Figure 3:

Fe-enhanced electron uptake by S. oneidensis MR-1 is diminished by the addition of the iron chelator deferoxamine (A) Representative chronoamperometry curves of S. oneidensis on graphite electrodes poised at −305 mV (vs SHE) with 30 mM fumarate as the electron acceptor. Arrow indicates times of FeCl₂ and DFO addition. (B) Representative cyclic voltammogram profiles of cell-containing and control bare electrodes with 1 mM FeCl₂ with and without deferoxamine.

It is interesting to consider the possible contribution of soluble iron in mediating extracellular electron uptake, in light of previous reports that flavins may also function as mediators for extracellular electron uptake by S. oneidensis.^[7] The enhancement in cathodic current resulting from FeCl₂ addition, observed in this study, appears to be higher than the enhancement previously observed from flavin addition,^[7] although under different experimental conditions and with much higher concentrations of added Fe (1mM) relative to flavins (1 μM). Also interesting is the situation in a natural environment where both flavins and soluble Fe may be present and contributing to inward or outward EET. In certain environments, such as anoxic marine sediments, flavins were detected at low nM concentrations and showed an increasing trend with depth; this trend was inversely correlated with dissolved Fe, which was detected in the 20-200 μM range.^[34] The relative contributions of each mediator to EET would ultimately depend on their respective concentrations in specific environments, possible redox interaction,^[35] and the specific redox reactions occurring in that environment. Outside the natural environment, we suggest that exogenously added soluble Fe may serve as an inexpensive mediator to enhance EET in microbial electrochemical technologies.

Fe-mediated electron uptake in S. oneidensis is linked to fumarate reduction.

Our electrochemical experiments include fumarate as a terminal electron acceptor for S. oneidensis. Under anaerobic conditions, the periplasmic flavocytochrome FccA serves as the only fumarate reductase in S. oneidensis,^[36] catalyzing the reduction of fumarate to succinate. Since previous studies of extracellular electron uptake by S. oneidensis confirmed that this process can be coupled to fumarate reduction,^[7] we tested whether FccA is required for the Fe-mediated electron uptake mechanism reported here. In contrast to the significant cathodic current observed with wild-type S. oneidensis, no current was detected from a ΔfccA mutant (Figure 4A) in chronoamperometry under similar experimental conditions (1 mM FeCl₂, 30 mM fumarate, and the working electrode poised at −305 mV vs. SHE). Likewise, the cyclic voltammetry measurements of ΔfccA were similar to the abiotic (cell-free) control, rather than the catalytic wave exhibited by wild-type S. oneidensis (Figure 4B). In the absence of fumarate with 1 mM FeCl₂, the abiotic control and cell containing reactors also demonstrated similar electrochemical signatures (Figure S5). This requirement for the fumarate reductase FccA and the presence of fumarate confirm that the electrode-sourced and Fe-mediated inward EET can enter the cellular periplasmic space to drive reduction of fumarate.

Figure 4:

Fe-enhanced electron uptake in S. oneidensis MR-1 is linked to fumarate reduction. (A) Chronoamperometry measurement at −305 mV (vs SHE) of S. oneidensis MR-1 WT and fumarate reductase mutant (ΔfccA) on graphite electrode surface in the presence of 1 mM FeCl₂ and 30 mM fumarate as an electron acceptor. Error bars indicate standard error of triplicate measurements. (B) Representative cyclic voltammogram of S. oneidensis MR-1 and ΔfccA on graphite electrode with 30 mM fumarate.

Fe-mediated electron uptake is dependent on outer membrane and periplasmic cytochromes.

Previous studies of cathodic electron uptake by S. oneidensis implicated the reversibility of the same Mtr pathway responsible for catalyzing outward EET to minerals and anodes.^[7,23] To examine whether the Fe-mediated process interacts with the outer membrane cytochromes, which in turn pass electrons to the periplasmic cytochromes and ultimately to FccA, we performed electrochemical experiments comparing Fe-mediated electron uptake by wild-type S. oneidensis and mutant strains lacking key Mtr pathway components (Table 1): ΔmtrC, ΔomcA, ΔmtrC/omcA, and ΔOMC (ΔomcA/ΔmtrA/ΔmtrF/ΔdmsE/ΔSO4360/ΔcctA/ΔrecA).

Table 1.

Strains used in this study.

Strain	Characteristics	Source
MR-1	S. oneidensis wild-type
JG731	S. oneidensis MR-1, ΔmtrC	[6]
JG719	S. oneidensis MR-1, ΔomcA	[6]
JG749	S. oneidensis MR-1, ΔmtrC/omcA	[6]
JG686	S. oneidensis MR-1, ΔfccA	[7]
JG1486	S. oneidensis MR-1, ΔomcA/ΔmtrA/ΔmtrF/ΔdmsE/ΔSO4360/ΔcctA/ΔrecA (ΔOMC)	[5]

Figure 5A displays the Fe-mediated inward EET currents from the tested strains with the working electrode poised at −305 mV vs. SHE. The mutant strains lacking either of the outer membrane decaheme cytochromes MtrC and OmcA (ΔmtrC and ΔomcA) demonstrated cathodic current levels 37% and 42% lower than wild-type S. oneidensis, respectively. A double deletion strain lacking both MtrC and OmcA (ΔmtrC/omcA) exhibited an even lower cathodic current – about 73% less than wild-type levels. Electron uptake was severely diminished in a strain lacking genes encoding eight functional periplasmic and outer membrane cytochromes (ΔOMC), with current levels about 96% lower than wild-type S. oneidensis.

Figure 5:

Fe-enhanced cathodic electron uptake is catalyzed by outer membrane cytochromes in S. oneidensis MR-1 (A) Chronoamperometry experiments (−305 mV vs SHE) of S. oneidensis, mtrC mutant (ΔmtrC), omcA mutant (ΔomcA), mtrC and omcA double mutant (ΔmtrC/omcA), and outer membrane cytochrome (OMC) mutant (ΔomcA/ΔmtrA/ΔmtrF/ΔdmsE/ΔSO4360/ΔcctA/ΔrecA) with 30 mM fumarate serving as an electron acceptor. Addition of 1mM iron is indicated by the arrowhead. Error bars indicate standard error of triplicate measurements. (B) Representative cyclic voltammograms of S. oneidensis WT and outer membrane cytochrome mutants (−465 to 435 mV at 1 mV/s) under nitrogen atmosphere.

Cyclic voltammetry (Figure 5B) echoed the chronoamperometry measurements, showing the highest catalytic activity in wild-type S. oneidensis and proportional reduction in the magnitude of the catalytic wave resulting from cytochrome deletions. Additionally, fluorescence microscopy (Figure S5) of the electrode-bound biofilms did not reveal significant differences between the strains, confirming the differences in cathodic electron uptake cannot be simply attributed to different cellular coverage. Taken collectively, these results indicate that MtrC and OmcA are the primary cell surface conduits through which electrons are passed from the dissolved Fe into the cells, and that the reversible Mtr pathway is the primary route for Fe-mediated inward EET.

It is interesting to consider the energetics of the interaction between soluble iron and the cell surface multiheme cytochromes. Taking MtrC as the representative cell surface conduit, previous protein film voltammetry (PFV) measurements show some variation depending on experimental conditions,^[37] but at pH 7 it can be reduced over a potential window from +100 mV to −400 mV vs SHE,^[38] with this broad range reflecting the presence of the 10 heme centers with overlapping potentials. Moreover, calculations of the individual heme reduction potentials show that this window can shift depending on the overall reduction state of the other hemes in the protein.^[39,40] The reduction potential of the iron mediator depends on the its interactions with other ligands (e.g. fumarate in our experiments) and the particular electrode material,^[11] but our measurements show a catalytic onset potential for the Fe-mediated mechanism of ~ 0 mV vs. SHE (Figure 1). This value is at the higher end of the potential window from PFV measurements of purified MtrC (and OmcA), suggesting a weak driving force for electron transfer from Fe²⁺ to cytochromes. However, it is also important to consider that the reduction potentials measured with direct live cell voltammetry of Shewanella are shifted higher than those obtained from purified cytochromes,^[10,41] which would result in a higher driving force for Fe-mediated electron injection into cells.

Conclusions

We have demonstrated that soluble iron (exogenously added as FeCl₂) significantly enhances extracellular electron uptake from graphite electrodes by the model EET organism Shewanella oneidensis MR-1. Experiments with the iron chelator deferoxamine indicate that the enhancement is due to freely diffusing ions acting as redox shuttles, where Fe³⁺/Fe²⁺ is continuously oxidized/reduced by the cells/electrodes. While extracellular electron uptake was previously demonstrated in S. oneidensis, and a role for soluble iron in mediating anodic EET was previously suggested, our study highlights how soluble iron can also facilitate cathodic EET into S. oneidensis. In addition, we demonstrated that the Fe-mediated inward EET is largely routed into the cells through the same cytochrome network responsible for outward EET, and can be coupled to fumarate reduction by S. oneidensis through the activity of the periplasmic fumarate reductase. These findings have implications for microbial electrochemical technologies, including electrosynthesis where electrode-sourced currents drive microbial biosynthetic pathways for production of high value chemicals and fuels.

Experimental Section

Cell cultivation

A preculture culture was cultivated from a frozen stock in 5 mL lysogeny broth (LB) aerobically at 30°C up to an optical density (OD 600 nm) of 2.4-2.6. Defined Media (DM) was inoculated with the washed preculture culture to a final OD of 0.05. DM contained (per liter): 0.2 g MgCl₂ 6H₂O, 1.0 g NH₄Cl, 0.2 g CaCl₂ 2H₂O, 0.9 g NaCl, 2.5 g NaHCO₃, 7.2 g HEPES buffer. For this culturing stage we added 0.5 g of yeast extract, 3.316 g Sodium lactate 60%(w/w) syrup. Trace minerals, amino acids and vitamin solution were added to supplement the medium as described previously.^[23] The pH of the medium was adjusted to 7.0 using NaOH. Cells were incubated at 30°C to an OD of 1-1.5. Cells were then washed 3X with fresh DM and were resuspended in 25 mL DM-anode (OD 0.1) with 30 mM lactate, trace minerals and amino acids for anodic cultivation (no yeast extract or vitamins).

Reactor setup and electrochemical measurements

A custom-built single chamber bioreactor was used to investigate the effect of FeCl₂ on extracellular electron uptake rate by S. oneidensis MR-1. The reactor contained a graphite working electrode (POCO AXF-5Q 0.059” x 0.225” x 0.83”) (Tri-Gemini LLC, CAT-num: XM15839C), platinum wire as a counter electrode, and an Ag/AgCl in 1M KCl reference electrode (CH Instruments, Inc.). The working electrode was polished using a 600-grit sandpaper, rinsed, and sonicated twice, first in distilled de-ionized water, then in ethanol. The working electrodes were stored in 1 M HCl until further processing. For anodic conditions (+435 mV vs SHE) 25 mL DM contained 30 mM sodium lactate, trace minerals and amino acids (DM-anode). For cathodic conditions (−305 mV vs SHE) 25 mL DM contained 30 mM sodium fumarate (DM-cathode). Cyclic voltammetry (CV) was performed by scanning the potential range of −565 mV to 635 mV (vs SHE) at a scan rate of 1 mV/s. All electrochemical analysis were performed under nitrogen atmosphere, at 30°C using a CHI1000 8-channel potentiostat (CH Instruments, Inc.).

Biofilm formation

Biofilm formation was facilitated in anaerobic conditions by poising the working electrode at an anodic potential (+435 mV vs SHE) with lactate serving as the electron source for 24 h. After 24 h the reactors were taken into an anaerobic chamber and electrodes were gently rinsed and transferred into a fresh reactor containing fresh DM-cathode. The rinsing of the electrodes ensures that only the firmly attached cells are present on the electrode for downstream experiments. The same culturing, biofilm formation, and electrochemical preparation was done for all S. oneidensis strains: wild-type, ΔmtrC, ΔomcA, ΔmtrC/omcA, ΔOMC and ΔfccA (Table 1).

FeCl₂ addition

To understand the effect of soluble ferrous chloride addition on the magnitude of cathodic electron uptake by S. oneidensis, we added anaerobic FeCl₂ to a final reactor concentration of 1 mM during chronoamperometry in the presence of fumarate. With working electrodes poised at −305 mV (vs SHE) current was monitored for ~30 min until a steady baseline was reached before the 1 mM FeCl₂ addition. FeCl₂ stock was made as previously described.^[42] To determine if the enhanced electron uptake phenotype is dependent on free iron ions in solution we added sterile anaerobic stock of the iron chelator deferoxamine (1.5 mM per 1 mM FeCl₂). To investigate the FeCl₂ concentration dependence of the enhanced electron uptake in S. oneidensis, we performed chronoamperometry and cyclic voltammetry measurements with increasing concentrations: 0.2 mM, 0.6 mM and 1 mM FeCl₂. To determine the link between Fe-enhanced electron uptake and fumarate reduction, we performed chronoamperometry experiments with the working electrode poised at −305 mV vs SHE using a fumarate reductase mutant strain (ΔfccA) in the presence of fumarate. Similar conditions were utilized for investigating the role of outer membrane and periplasmic cytochromes in Fe-enhanced cathodic EET (ΔmtrC, ΔomcA, ΔmtrC/omcA, ΔOMC).

Iron chelator addition and toxicity test

During chronoamperometry 1.5 mM iron chelator deferoxamine myselate salt (DFO) (Sigma-Aldrich Co.) was added 30-40 minutes after FeCl₂ addition. Cyclic voltammetry was performed 30 minutes after DFO addition to investigate the electrochemical interaction between the cathode and S. oneidensis after iron chelation. Voltammetry was performed by scanning the potential range of −565 mV to 635 mV (vs SHE) at a scan rate of 1 mV/s. To investigate any possible toxic effects on cells due to DFO, CFU counts were performed via serial dilutions of cells in DM-cathode media before and 30 minutes after DFO addition. Plates were incubated in anaerobic conditions at 30°C with fumarate serving as the electron acceptor. Colonies were quantified 24 h after incubation.

Fluorescence microscopy

After running electrochemical experiments, electrodes were processed for fluorescence microscopy imaging. Electrodes were gently rinsed with fresh media and then fixed and stained in 1 mL fresh DM containing 25% glutaraldehyde, and 0.25 μg/mL FM4-64FX membrane stain solution (Thermo Fisher Scientific). Samples were stored in 4°C for ~15 h prior to imaging. Sample images were taken using the 100X objective of a Nikon Eclipse Ti-E inverted fluorescent microscope. Images were taken from 10 fields of view per sample.

Tetramethylbenzidine heme stain SDS-PAGE protein gel

After anodic biofilm formation, planktonic cells were harvested and centrifuged for 10 minutes at 6,500 RCF. The pellet was resuspended and incubated in anaerobic serum bottles containing anaerobic DM-cathode media (with fumarate) for 30 minutes before FeCl₂ addition to reach 1 mM concentration. 5 mL aliquots were taken immediately and then again 30 minutes after addition FeCl₂, centrifuged for 10 minutes at 6,500 RCF, and resuspended in 1 mL DM-cathode media. These samples were centrifuged again for 10 minutes at 20,000 RCF and resuspended in 50 μL Laemmli sample buffer (Bio-Rad). Subsequently, the sample was boiled for 5 minutes. 10 μL of sample was loaded for running a 12% SDS-PAGE protein gel. The gel was rinsed with DI water for 5 minutes followed by immersion in a solution containing 15 mL 6.3 mM TMBZ in methanol and 35 mL 0.25 M sodium acetate (pH 5.0). Gels were left in the dark shaking at 40 RPM for 2 h followed by addition of 3 mL 30% hydrogen peroxide, and bands were visible within 3 minutes.