APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 2011, p. 76337639 0099-2240/11/$12.00 doi:10.1128/AEM.05365-11 Copyright 2011, American Society for Microbiology.

. All Rights Reserved.

Vol. 77, No. 21

Evidence for Direct Electron Transfer by a Gram-Positive Bacterium Isolated from a Microbial Fuel Cell
K. C. Wrighton,1,2 J. C. Thrash,1,3 R. A. Melnyk,1 J. P. Bigi,4,5 K. G. Byrne-Bailey,1 J. P. Remis,6 D. Schichnes,7 M. Auer,6 C. J. Chang,4,5 and J. D. Coates1*
Department of Plant and Microbial Biology, University of California, Berkeley, California 947201; Department of Earth and Planetary Science, University of California, Berkeley, California 947202; Department of Microbiology, Oregon State University, Corvallis, Oregon 973313; Department of Chemistry, University of California, Berkeley, California 947204; Howard Hughes Medical Institute, University of California, Berkeley, California 947205; Life Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California 947206; and College of Natural Resources Biological Imaging Facility, University of California, Berkeley, California 947207
Received 4 May 2011/Accepted 1 September 2011

Despite their importance in iron redox cycles and bioenergy production, the underlying physiological, genetic, and biochemical mechanisms of extracellular electron transfer by Gram-positive bacteria remain insufciently understood. In this work, we investigated respiration by Thermincola potens strain JR, a Grampositive isolate obtained from the anode surface of a microbial fuel cell, using insoluble electron acceptors. We found no evidence that soluble redox-active components were secreted into the surrounding medium on the basis of physiological experiments and cyclic voltammetry measurements. Confocal microscopy revealed highly stratied biolms in which cells contacting the electrode surface were disproportionately viable relative to the rest of the biolm. Furthermore, there was no correlation between biolm thickness and power production, suggesting that cells in contact with the electrode were primarily responsible for current generation. These data, along with cryo-electron microscopy experiments, support contact-dependent electron transfer by T. potens strain JR from the cell membrane across the 37-nm cell envelope to the cell surface. Furthermore, we present physiological and genomic evidence that c-type cytochromes play a role in charge transfer across the Gram-positive bacterial cell envelope during metal reduction. Microbial metal metabolism has signicant medical, environmental, and industrial importance. Assimilatory Fe(III) reduction impacts pathogen physiology and virulence in phylogenetically diverse bacteria (6, 51), while dissimilatory metal reduction is central to biogeochemical processes. Specically, mesophilic dissimilatory metal reduction impacts the bioattenuation of various contaminants (12), while reduction at elevated temperatures has been associated with the formation of corrosive products in heated delivery systems and organic carbon mineralization in deep petroleum reservoirs (54). Additionally, the mechanisms involved in dissimilatory metal reduction are critical to power production in microbial fuel cells (MFCs) (32, 47). Fe(III) oxides and Mn(IV) oxides are insoluble at circumneutral pH and, like anodic electrodes, must be reduced extracellularly. As such, dissimilatory metal-reducing bacteria (DMRB) must complete the circuit between the respiratory electron transport chain located in the plasma membrane and the insoluble electron acceptor located outside the cell (58). Despite the phylogenetic diversity of DMRB, knowledge of electron transfer mechanisms is largely conned to two model Gram-negative bacterial genera, Geobacter and Shewanella. The extracellular electron transfer mechanisms utilized by these bacteria involve either contact-dependent or contactindependent strategies. Contact-dependent (direct) mechanisms transfer electrons via multiheme cytochrome c proteins (MHCs) (53, 58) or nanowires (22, 48). In contrast, contactindependent mechanisms involve soluble redox-active mediators that shuttle electrons from the electron transport chain to the insoluble electron acceptor (23, 57). Compared to Gram-negative bacteria, Gram-positive bacteria lack an outer membrane, have a thicker cell wall (10 to 80 nm), and may be encased in a glycoprotein S layer. Due to these structural differences, direct electron transfer to insoluble electron acceptors was once considered incompatible with a Gram-positive bacterial envelope (45). Here we present a study on the use of insoluble electron acceptors by a Grampositive bacterium, Thermincola potens strain JR. This bacterium was isolated from a current-producing anodic biolm and in pure culture was demonstrated to respire 91% of the electrons in acetate to current (59). Using electrochemical, physiological, and imaging methods, we conrm that T. potens participates in direct extracellular electron transfer. Furthermore, we present genomic evidence that c-type cytochromes are abundant in T. potens and physiological verication that c-type cytochromes are capable of charge transfer to extracellular electron acceptors. Taken together, our ndings have broad implications for understanding the phylogenetic and metabolic diversity fundamental to environmental redox cycles, pathogenesis, and sustainable bioenergy applications.
7633

* Corresponding author. Mailing address: Department of Plant and Microbial Biology, University of California, Berkeley, 271 Koshland Hall, Berkeley, CA 94720. Phone: (510) 643-8455. Fax: (510) 642-4995. E-mail: jdcoates@berkeley.edu. Supplemental material for this article may be found at http://aem .asm.org/. Published ahead of print on 9 September 2011.

7634

WRIGHTON ET AL.
MATERIALS AND METHODS

APPL. ENVIRON. MICROBIOL.

microscope with an underfocus value of 10 m. The microscope was equipped with an FEG electron source operating at 300 kV, an Omega energy lter, a Gatan 795 2,000- by 2,000-pixel charge-coupled-device camera, and a cryotransfer stage. The stage and sample were cooled to 80 K for the duration of data collection. Cells were placed onto lacey carbon grids (01881; Ted Pella) that were pretreated by glow discharge. The grids were blotted and plunged into liquid C2H6 using a Vitrobot FEI automated vitrication device before being transferred and stored in liquid N2 until imaged. To visually examine strain JR biolms using confocal laser scanning microscopy (CSLM), three separate portions (2 cm2) located on spatially distinct regions of the carbon cloth anode were aseptically and anoxically removed from two separate current-producing MFCs. Anode portions were dipped in freshwater medium to remove debris loosely associated with the biolm and were uorescently stained with 1 SYTO BC and 2 g/ml propidium iodide. Biolm structures were immediately examined using an inverted LSM710 (Carl Zeiss, Inc.) microscope equipped with a 63 magnication, 1.2-numerical-aperture water immersion lens. To ensure that the biolm structure was not compressed, the sample was placed biolm side down into a MatTek dish with a 1.5 coverslip bottom. Image stacks were acquired with the pinhole set to 1 Airy unit, allowing in 1.1 m of light in the z dimension. Images were acquired at every 0.55 m in the z dimension, in accordance with Nyquist sampling. Two- and three-dimensional images were obtained, and green- and red-stained cells were quantied using Imaris (version 6.2) software (Bitplane, AG). Search for redox-active mediators. The role played by biolm-independent processes was evaluated using a medium-replacement experiment. Here, the medium in the anode was removed daily (n 5; to minimize concerns of nutrient carryover, only the ndings for the last 3 time points are reported) and the anode biolm and chamber were rinsed with 30 mM bicarbonate buffer three times at each replacement. Chambers were relled with anoxic sterile medium (10 mM acetate) lacking vitamins and minerals to remove any growth and putative redoxactive components. Current production was recorded every 10 s over the course of the experiment. The accumulation of soluble mediators in the medium surrounding the anode was also evaluated using spent-medium experiments. Culture broth was obtained from duplicate current-producing anodes (operated for 8 weeks in batch format to concentrate secreted metabolites), centrifuged (8,000 g, 20 min) to remove cell debris, and made anoxic. Filters were not used to avoid losing organic molecules by adsorption. Conditioned medium was added (nal concentration, 80%) to 100 mM HFO and 10 mM acetate-amended bicarbonate fresh water medium. Three conditioned medium-free controls were also run in parallel: 10 mM acetate, 100 mM HFO, and 10 m AQDS (positive control), no donor and 100 mM HFO (negative control), and 10 mM acetate and 100 mM HFO (spentmedium control). A 10% inoculum was used, with Fe(II) production monitored in triplicate for each treatment (33). Cyclic voltammetry (CV), using BASis Epsilon potentiostat and C-3 cell stand (West Lafayette, IN), was performed on anode culture medium and rinsed biolm samples (scan rate, 2 mV/s), with voltammograms for each included (see Fig. S1 in the supplemental material). Thirty milliliters anode medium was removed, and washed biolms were collected (times of 0, 4, and 12 weeks from the time of current stabilization). Samples were kept anoxic in tinfoil-wrapped bottles at 60C. The working electrode was a glassy carbon disk (diameter, 3.0 mm), the counterelectrode was a platinum wire, and the solutions were referenced with an aqueous Ag/AgCl electrode purchased from BASi. The working electrode was cleaned prior to use with polishing alumina (particle size, 0.05 m). Electrochemical cells were dried at 160C and cooled to room temperature under a ow of N2. Aqueous samples (5 ml) were kept anoxic (N2) during CV, with medium temperature and pH quantied immediately before and after each voltammogram to ensure measurement under biologically relevant conditions. As a positive control, CV was performed on cell-free samples spiked with stock solutions of riboavin and AQDS (scan rates, 2, 100, and 1,000 mV/s). The increased scan rate was necessary for riboavin samples to account for adsorbed redox species (35). Cytochrome c analysis. The presence, predicted protein length, presence of signal peptide/transmembrane, and homology of MHCs were conrmed by genome sequencing of T. potens strain JR (8). Analysis using the BLASTP program was performed against the entire nr database on 2 November 2009 (1). Domain structures of the gene models were assembled by searching against the InterPro database (26). Transmembrane domains and signal peptides were identied using the web-based programs TMHMM (version 2.0) and SignalP (version 3.0), respectively (2, 27). The c-type cytochrome complement of strain JR was identied with the aid of the HMMER (version 3.0b3) software package (13), using the doubled-heme-domain Pfam prole PF09699 (17) as the input prole. To evaluate capacity for electron transfer to iron(III) oxides by c-type cyto-

Phylogenetic analysis. Genomic DNA of strain JR was isolated from pure cultures using a PowerSoil kit (MoBio) according to the manufacturers protocol, and the 16S rRNA gene was amplied as described previously (59). Sequences were aligned with the MUSCLE (version 3.6) program (14) and analyzed with MrBayes (version 3.2) software (25, 50) using 4 chains until the standard deviation of the split frequencies was stabilized below 0.01, in this case, for 1,087,000 generations, with a sample frequency of 1,000. The rst 25% of the samples were discarded for accurate estimation of the posterior probability distribution of the summary tree. Geothrix fermentans (GenBank accession number U41563) and Acidobacteria capsulatum (GenBank accession number D26171) were selected as representative acidobacteria; Geobacter metallireducens strain GS-15 (GenBank accession number L07834) and Shewanella oneidensis strain MR-1 (GenBank accession number AF005251) were chosen as representative proteobacteria. The 16S rRNA GenBank accession number for Thermincola potens strain JR is GU815244. Accession numbers for additional Firmicutes used for the analysis were EU016427, EU016431, EU016410, EU016421, EU016442, EU016420, EU016424, EU016416, EU016422, EU016418, EU016409, EU016435, EU016425, EU016448, EU194832, EU194831, EU194830, EU638403, EU638851, EU638690, EU638872, EU638393, EU638902, EU638687, EU639376, EU638700, EU638698, EU638838, EU638794, EU638786, EU638783, EU638810, EU638644, AB159558, AB154390, AB232785, AB091323, X91169, AY631277, AY603000, EF542810, U76363, AJ621886, Y11575, DQ148942, AF516177, AJ276351, X61138, AJ575812, AB075768, and X81021. MFC construction and operation. The MFC was constructed with two 50-mmdiameter glass chambers with three sample ports connected with a 30-mm pinch clamp assembly (Laboratory Glass Apparatus, Berkeley, CA). Chambers were separated with a cation-exchange membrane (Naon 117). The electrodes were either graphite blocks or graphite carbon ber. For graphite blocks, unpolished graphite 2.5 by 1.25 by 7.5 cm (G-10 Graphite Engineering and Sales, Greenville, MI) was connected with watertight threaded ttings (Impulse, San Diego, CA) to wires and sealed with conductive silver epoxy (Epoxy Technology, Billerica, MA). For graphite ber electrodes, graphite ber (G-10 Graphite Engineering and Sales, Greenville, MI) was trimmed into rectangles (2.5 by 7.5 cm). For both electrode versions, electrodes were connected to resistors (470 ) through stoppers in the top of each chamber. Voltage (Ecell or cell potential) was measured across a xed external resistor (Rext) and logged using Chart (version 4.0) software (ADInstruments, CA). With the exception of the medium-swap experiments, data were recorded every 10 min. Current (I) was calculated from measured voltages using Ohms law (I Ecell/Rext). Maximum overall power was calculated as Ecell2 maximum/Rext. Coulombic efciency, or the electrons in acetate that are recovered as current, was calculated in duplicate, and the mean value is reported. Additional images and details are included elsewhere (59). Prior to inoculation with strain JR, the chambers, electrodes, and all stoppers were autoclaved to ensure sterility. The anodic chamber was aseptically lled with 250 ml of presterilized, anoxic, 30 mM bicarbonate buffer (pH 6.8) amended with 10 mM acetate as the sole electron donor, while the cathodic chamber was aseptically lled with 250 ml of 30 mM Tris buffer (pH 6.8). The anode chamber was bubbled continuously with ltered (pore size, 0.22 m) N2-CO2 (80:20, vol/vol) gas to ensure anoxic operation and maintenance of pH throughout operation, while the cathode chamber was bubbled continuously with ltered air (pore size, 0.22 m). Anode compartments were maintained at 60C over the course of the experiments using heated water jackets surrounding the anode. Microbiological methods. Unless otherwise noted, cultures were maintained at 60C on 30 mM bicarbonate buffer (pH 6.8) freshwater medium with the N2-CO2 (80:20, vol/vol) headspace amended with 10 mM acetate and 100 mM amorphous hydrous Fe(III) oxide (HFO) (33). To evaluate the electrochemical activity of Thermincola potens strain JR, pure cultures were washed and inoculated into the MFC. Cells were grown anaerobically at 60C in 1-liter volumes using acetate (10 mM) as the electron donor and anthraquinone-2,6-disulfonate (AQDS; 10 mM) as the electron acceptor, standard anoxic techniques, and basal bicarbonate medium. Bacterial cells were harvested upon reduction of AQDS by centrifugation under an anoxic atmosphere, washed once, and resuspended in 10 ml basal bicarbonate medium in an anaerobic glove bag. The cell suspension was anoxically and aseptically added to the anode compartment of MFCs. Microscopy. For cryo-electron microscopy (cryo-EM) and cytochrome scans, cells grown in 1 liter medium containing AQDS and acetate (10 mM each) were centrifuged (8,000 g, 10 min) and resuspended in 1 ml piperazine-N,N-bis(2ethanesulfonic acid) (PIPES) buffer (pH 7.0) with N2 headspace. Cryo-EM images were acquired at a 20,000 magnication on a JEOL-3100 electron

VOL. 77, 2011

BACTERIAL DIRECT EXTRACELLULAR ELECTRON TRANSFER

7635

chromes, spectrophotometric analyses were performed in triplicate in an anoxic glove bag using a Varian Cary 50 MPR UV/visible microplate spectrophotometer (Varian Inc., CA). Whole cells were diluted to approximately 0.11 mg ml1 total protein in 30 mM phosphate buffer and titrated with small amounts of 0.5 M sodium dithionite until reduction of the c-type cytochromes was observed (1 to 3 l was required). Subsequently, aliquots of 0.15 M HFO were added (3 to 6 l), and the oxidation spectrum was rerecorded.

RESULTS AND DISCUSSION Thermincola potens strain JR is phylogenetically and morphologically Gram positive. Strain JR is a Firmicutes belonging to the Peptococcaceae in the Clostridiales by 16S rRNA gene sequence analysis (1,228 nucleotides; Fig. 1A). It shared 99% 16S rRNA gene sequence identity with T. carboxydophila and T. ferriacetica, the two previously characterized members of the Thermincola genus (55, 62). Outside of cultured representatives, 16S rRNA gene sequence analyses from MFC anodes (operated at 55C) inoculated with anaerobic digester sludge (59) or marine sediment (36) were dominated by sequences 93 to 99% similar to the rRNA gene sequence of strain JR. In addition, stable isotope probing and 16S rRNA gene clone library studies revealed the prevalence and activity of sequences related to strain JR (92% identity) in mesophilic benzene-oxidizing Fe(III)-reducing mesocosms (28). Together, these results suggest a selective advantage for bacteria related to Thermincola in systems requiring external electron transfer. Cryo-EM conrmed that strain JR was morphologically Gram positive with a 33-nm-thick bipartite envelope that includes a 16-nm low-density periplasm surrounded by a 17-nm high-density cell wall (Fig. 1B). These results are consistent with those of previous studies of four mesophilic Gram-positive bacteria: Bacillus subtilis, Staphylococcus aureus, Enterococcus gallinarum, and Streptococcus gordonii (37). This cellular architecture has important ramications for extracellular electron transfer by strain JR. Evidence for direct electron transfer. As a rst method for detecting secreted redox-active components by strain JR, we assessed the contribution of anodic biolm-independent processes to MFC current production. The spent culture broth surrounding the anode was removed and replaced with fresh acetate-amended medium. Previously, this method demonstrated that Shewanella oneidensis MR-1 and Geothrix fermentans produced soluble mediators, as described by a notable reduction in current (greater than 50%) and a lag (3 to 10 days) for current to return to original levels (5, 35). In contrast, with strain JR current recovered after medium replacement to original levels within 29 15 min (n 3) (Fig. 2A). Moreover, the rinsing and transfer of electrode biolms to sterile anoxic MFCs had no impact on current production. As an alternative method for detecting the presence of endogenous shuttling compounds, we supplemented HFO-reducing cultures with cell-free spent anodic culture broth. Similar experiments with Shewanella algae strain BRY, Geothrix fermentans, and Pyrobaculum aerophilum demonstrate that endogenous mediators accumulate in conditioned culture broth such that addition of spent medium minimizes the lag phase associated with HFO reduction, increasing the initial ferric iron reduction rate relative to that for a non-spent-medium control (16, 41, 42). In strain JR, spent medium failed to stimulate the reduction of HFO (Fig. 2B). In contrast, when

cells were supplemented with an exogenous electron shuttle (AQDS, 10 M), initial HFO reduction occurred signicantly faster than the rate for controls not amended with spent medium, indicating that strain JR could use soluble electron shuttles for transferring electrons onto iron hydroxides but that a soluble mediator does not accumulate in strain JR medium to a level physiologically detectable by the organism under these assay conditions. For more sensitive detection of soluble redox-active mediators, we performed cyclic voltammetry (CV), previously used to characterize bacterial electron transfer mechanisms to electrodes (20, 24, 31, 35, 44, 46, 61). No electrochemical differences were evident between the spent MFC culture broth and an uninoculated control, indicating that redox-active components either were not present or were present at levels below the limit of detection (Fig. 2C; see Fig. S2 in the supplemental material). Likewise, CV performed on washed biolms did not differ from that performed on sterile-medium controls. To determine the detection limits of the known redox-active mediators AQDS and riboavin in this electrochemical system, we titrated these compounds into spent MFC culture medium prior to CV analysis (Fig. 2C). Both compounds were detectable at the appropriate midpoint potential, with conservative detection limits (100 and 200 nM, respectively) being below those reported for endogenous mediators from Shewanella in current-producing anodes (35). The results from experiments presented here are in agreement with those for a Geobacter sp., a model organism for direct extracellular electron transfer, where medium exchange failed to impact current production (4), the addition of spent medium did not stimulate iron reduction (4, 9, 40), and CV failed to detect soluble redox-active components (56). Lack of support for long-range electron transfer through anodic biolms. CSLM with live-dead stain has previously been used to implicate long-range electron transfer by Geobacter sulfurreducens (19, 43, 49). We used CSLM to monitor temporal changes in strain JR biolm thickness and viability on an anode surface. Once stable power production was achieved (7 to 10 days), CSLM revealed a 2-m-high monolayer of 92% viable cells on the electrode surface (see Fig. S2 in the supplemental material). One month later, the electrode biolm had grown 10-fold in height (22 m 5 m, n 6), without a corresponding increase in current production (Fig. 3; see also Fig. S2C in the supplemental material). Quantied cell density and viability on the aged biolm revealed a dense layer of viable cells in contact with the electrode surface, with the middle and top biolm portions showing decreasing cell densities (34% and 37%, respectively) and decreasing percentages of viable cells (44% and 41%, respectively) of the biolm relative to the cells on the anode surface (Fig. 3). The lack of correspondence between current production and biolm height coupled to the decreased cell density and viability beyond the anode surface monolayer signies that cells in contact with the electrode primarily contribute to current production. These results are opposed to the CSLM results obtained with G. sulfurreducens, where cells up to 50 m from the electrode surface remain equally dense and viable, with current production correlated to biolm height. It was suggested that while Geobacter cells in intimate contact with the anode surface may rely on electron transfer via outer membrane c-type

7636

WRIGHTON ET AL.

APPL. ENVIRON. MICROBIOL.

FIG. 1. (A) A 16S rRNA gene tree constructed using Bayesian analysis reveals that T. potens is a member of the Gram-positive bacterial phylum Firmicutes within the family Peptococcaceae. Closed circles at nodes indicate posterior probabilities of 0.97; the open circle indicates a posterior probability of 0.97. (B) Cryo-EM shows that the plasma membrane (PM) is enclosed by a 16-nm-thick low-density periplasmic space (PS) that is bound by a 17-nm-thick high-density cell wall (CW), indicative of a Gram-positive cell wall.

VOL. 77, 2011

BACTERIAL DIRECT EXTRACELLULAR ELECTRON TRANSFER

7637

FIG. 3. Confocal laser scanning microscopy of T. potens anodic biolm. The electrode surface is denoted by white arrows, with the topographic variation associated with carbon ber weave being apparent. Three-dimensional rendering revealed a biolm an average of 22 m thick with a monolayer of viable (green) cells in contact with the electrode surface. Bar, 10 m. Quantication of biolm cross-section reveals a greater percentage of viable cells and a higher cell density associated with electrode surface.

FIG. 2. Evidence for contact-dependent electron transfer by T. potens. (A) Current production by strain JR returned to original levels 30 min after replacement of anode medium (n 3). d, number of days. (B) Fe(III) oxide reduction coupled to acetate oxidation was not stimulated by amendment with 12-week-old cell-free spent medium (80%, vol/vol) from microbial fuel cells. Effects on the rate of iron reduction over time with the addition of 8 ml of either supernatant (E), basal medium (f), or 10 M AQDS and basal medium (). No signicant reduction occurred in the absence of acetate (). Mean results of triplicate cultures and standard errors are reported. (C) No detectable current production was detected in the subtracted difference between cell-free supernatant from a 12-week old current-generating MFC and control medium (blue line). MFC cell-free supernatant was spiked with 250 nM riboavin (black line). SHE, standard hydrogen electrode.

membrane electron transport chain to the electron acceptor on the cell surface. The physiological signicance of MHCs is evident in genomic data, with 23 and 69 MHCs annotated in S. oneidensis and G. sulfurreducens, respectively. Furthermore, the average numbers of predicted heme-binding motifs per MHC in genome-sequenced Shewanella and Geobacter species are 6.42 and 9.31, respectively, indicating their importance in charge transfer events (7, 38, 39, 43). Analogous to the c-type cytochrome repertoire in Gramnegative organisms, the genome of strain JR contains 35 putative MHCs with an average of 11.9 hemes per cytochrome. We used a spectrophotometric approach to conrm the expression and inferred function for c-type cytochromes in strain JR. Reduced spectra minus HFO-oxidized spectra of whole cells showed absorption peaks at 420, 526, and 552 nm, corresponding to the gamma, beta, and alpha bands of c-type cytochromes, respectively (Fig. 4). The oxidation of c-type cytochromes by insoluble iron signies that electrons from strain JR c-type cytochromes can be transferred to external electron acceptors in intact cells.

cytochromes, nanowires were responsible for long-range transfer across the multilayer biolm (43, 49). Keeping with this interpretation, the confocal microscopy results presented here suggest that T. potens strain JR does not have similar machinery for long-range electron transfer through the anodic biolm. Expression and metal reduction activity of c-type cytochromes. In direct extracellular electron transfer by S. oneidensis and G. sulfurreducens, MHCs located in the periplasm and outer membrane transport electrons generated from the inner

FIG. 4. T. potens cells grown on 10 mM AQDS express c-type cytochromes, which are capable of transferring electrons to extracellular electron acceptors, as revealed by the difference spectrum of strain JR whole cells of dithionite reduced and HFO oxidized with absorbance maxima at 426, 522, and 553 nm. Values on the x axis indicate absorbance wavelengths.

7638

WRIGHTON ET AL.

APPL. ENVIRON. MICROBIOL.

Direct extracellular electron transfer in Gram-positive bacteria. In addition to the work reported here, contact-dependent dissimilatory metal reduction has been identied in two Gram-positive DMRB. The inability of culture ltrate to augment HFO reduction was evidence that Thermoanaerobacter sp. strain BSB-33 fails to release extracellular redox mediators (3). Likewise, CV demonstrated that Thermincola ferriacetica employs a direct mechanism of electron transfer to MFC anodes (34). The shared mechanism of electron transfer to anodes between Thermincola spp. (Fig. 1A) is worth noting. Consistent with results obtained here, within the genus Geobacter, G. metallireducens and G. sulfurreducens both utilize a direct mechanism of extracellular electron transfer (4, 40). On the other hand, different strategies for insoluble Fe(III) reduction within a single genus exist for both Pyrobaculum and Shewanella genera (10). T. potens strain JR is similar to Gram-negative DMRB capable of direct electron transfer in the number and heavy loading of heme motifs on MHCs; however, the relationship between MHCs and direct electron transfer has yet to be established in other Gram-positive DMRB. Diagnostic cytochrome spectra performed on two Gram-positive DMRB identied c-type cytochromes in Carboxydothermus ferrireducens (21) but not in Desultobacterium metallireducens (18). Comparative genomics of 594 prokaryote genomes revealed that Carboxydothermus hydrogenoformans and Desultobacterium hafniense were the only Gram-positive DMRB to contain more than 10 MHCs, with 13 and 16, respectively (52). However, the ability of these bacteria to transfer electrons extracellularly has not been characterized at a mechanistic level. Future directions. A key challenge facing researchers investigating external electron transfer is elucidating the molecular details of the reduction process. Although contact-dependent and -independent mechanisms of reduction are often presented to be distinct from one another, the extent to which they are exclusive or overlap is unclear (30, 35). Furthermore, it was also shown that Geobacter sulfurreducens uses a combination of direct mechanisms by aligning MHC proteins along pili to promote electron transfer to Fe(III) oxides (29). In T. potens, there is the additional unique problem of conducting electrons across the 17-nm-thick cell wall. We propose three molecular models to rationalize contact-dependent electron transfer in strain JR (60): (i) conductive cell appendages (nanowires), (ii) conductive cell walls (15), or (iii) a redox-active protein conduit linking the inner membrane to the cell surface. Integrating MHCs into these models, we consider that in the conductive cell wall model, MHCs localized to the 16-nm-thick periplasmic space could function as terminal reductases. These MHCs would shuttle electrons from the plasma membrane to nonpeptide components of the cell wall, e.g., teichoic and teichuronic acids, which bind reversible metal ions or yet to be recognized components to convey electrons from the periplasmic MHCs to electron acceptors on the cell surface (15). An alternative or, possibly, complementary hypothesis considers that the cell wall could be hardwired with cytochromes, as Gram-positive bacteria have several methods for covalently and noncovalently attaching proteins to the peptidoglycan layer (11). Beyond the cell wall, the S layer is a protein matrix that could also be studded with MHCs and proteins that me-

diate attachment and electron transfer to insoluble electron acceptors. With Thermincola potens JR as a model organism, we used a combination of physiological, electrochemical, and genomic observations to expand the knowledge of Gram-positive bacterial extracellular respiration. Our ndings demonstrate that strain JR is capable of contact-dependent electron transfer to these electron acceptors, and while the molecular basis for this mechanism is presently not understood, implementation of additional technologies, such as biolm CV, proteomics, and transcriptomics, will reveal how MHCs are functionally integrated into the physiology of strain JR. Ultimately, this research will elucidate the phylogenetic and physiological variation in bacterial electron transfer mechanisms and improve our understanding of the metabolism of Gram-positive bacteria with applications to pathogen physiology, bioremediation, and energy generation.
ACKNOWLEDGMENTS Funding for this work was provided to J.D.C. through the DOE LDRD and the UCB SPS programs. K.C.W. was supported by a Tien Scholars Biodiversity Graduate Fellowship, and J.P.B. was supported by an NSF Graduate Fellowship. C.J.C. is a Howard Hughes Medical Institute investigator. We thank Steven Ruzin of the UCB Biological Imaging Facility.
REFERENCES 1. Altschul, S. F., et al. 1997. Gapped blast and psi-blast: a new generation of protein database search programs. Nucleic Acids Res. 25:33893402. 2. Bendtsen, J. D., G. H. N. von Heijne, and S. Brunak. 2004. Improved prediction of signal peptides: Signalp 3.0. J. Mol. Biol. 340:783795. 3. Bhowmick, D. C., B. Bal, N. S. Chatterjee, A. N. Ghosh, and S. Pal. 2009. A low-GC Gram-positive Thermoanaerobacter-like bacterium isolated from an Indian hot spring contains Cr(VI) reduction activity both in the membrane and cytoplasm. J. Appl. Microbiol. 106:20062016. 4. Bond, D. R., and D. R. Lovley. 2003. Electricity production by Geobacter sulfurreducens attached to electrodes. Appl. Environ. Microbiol. 69:1548 1555. 5. Bond, D. R., and D. R. Lovley. 2005. Evidence for involvement of an electron shuttle in electricity generation by Geothrix fermentans. Appl. Environ. Microbiol. 71:21862189. 6. Brown, J. S., and D. W. Holden. 2002. Iron acquisition by Gram-positive bacterial pathogens. Microbes Infect. 4:11491156. 7. Butler, J. E., D. Nelson, and D. R. Lovley. 2010. Evolution of electron transfer out of the cell: comparative genomics of six Geobacter genomes. BMC Genomics 11:4052. 8. Byrne-Bailey, K. G., et al. 2010. Complete genome sequence of the electricity producing Thermincola potens strain JR. J. Bacteriol. 15:40784079. 9. Chaudhuri, S. K., and D. R. Lovley. 2003. Electricity generation by direct oxidation of glucose in mediatorless microbial fuel cells. Nat. Biotechnol. 21:12291232. 10. Childers, S. E., and D. R. Lovley. 2001. Differences in Fe(III) reduction in the hyperthermophilic archaeon, Pyrobaculum islandicum, versus mesophilic Fe(III)-reducing bacteria. FEMS Microbiol. Lett. 195:253258. 11. Desvaux, M., E. Dumas, I. Chafsey, and M. Hebraud. 2006. Protein cell surface display in Gram-positive bacteria: from single protein to macromolecular protein structure. FEMS Microbiol. Lett. 256:115. 12. Dubiel, M., C. H. Hsu, C. C. Chien, F. Mansfeld, and D. K. Newman. 2002. Microbial iron respiration can protect steel from corrosion. Appl. Environ. Microbiol. 68:14401445. 13. Eddy, S. R. 1998. Prole hidden Markov models. Bioinformatics 14:755763. 14. Edgar, R. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32:17921797. 15. Ehrlich, H. L. 2008. Are gram-positive bacteria capable of electron transfer across their cell wall without an externally available electron shuttle? Geobiology 6:220224. 16. Feinberg, L. F., and J. F. Holden. 2006. Characterization of dissimilatory Fe(III) versus NO3 reduction in the hyperthermophilic archaeon Pyrobaculum aerophilum. J. Bacteriol. 188:525531. 17. Finn, R. D., et al. 2008. The Pfam protein families database. Nucleic Acids Res. 36:D281D288. 18. Finneran, K., H. Forbush, C. Gaw VanPraagh, and D. Lovley. 2002. Desultobacterium metallireducens sp. nov., an anaerobic bacterium that couples

VOL. 77, 2011

BACTERIAL DIRECT EXTRACELLULAR ELECTRON TRANSFER

7639

19.

20.

21.

22.

23. 24.

25. 26. 27.

28.

29.

30.

31.

32. 33.

34.

35. 36. 37.

38.

39. 40.

growth to the reduction of metals and humic acids as well as chlorinated compounds. Int. J. Syst. Evol. Microbiol. 52:1929. Franks, A. E., K. P. Nevin, R. H. Glaven, and D. R. Lovley. 2010. Microtoming coupled to microarray analysis to evaluate the spatial metabolic status of Geobacter sulfurreducens biolms. ISME J. 4:509519. Freguia, S., M. Masuda, S. Tsujimura, and K. Kano. 2009. Lactococcus lactis catalyses electricity generation at microbial fuel cell anodes via excretion of a soluble quinone. Bioelectrochemistry 76:1418. Gavrilov, S., A. Slobodkin, F. Robb, and S. de Vries. 2007. Characterization of membrane-bound Fe (III)-EDTA reductase activities of the thermophilic gram-positive dissimilatory iron-reducing bacterium Thermoterrabacterium ferrireducens. Microbiology 76:139146. Gorby, Y. A., et al. 2006. Electrically conductive bacterial nanowires produced by Shewanella oneidensis strain MR-1 and other microorganisms. Proc. Natl. Acad. Sci. U. S. A. 103:1135811363. Gralnick, J. A., and D. K. Newman. 2007. Extracellular respiration. Mol. Microbiol. 65:111. Hartshorne, R. S., et al. 2009. Characterization of an electron conduit between bacteria and the extracellular environment. Proc. Natl. Acad. Sci. U. S. A. 106:2216922174. Huelsenbeck, J. P., and F. Ronquist. 2001. Mrbayes: Bayesian inference of phylogenetic trees. Bioinformatics 17:754755. Hunter, S., et al. 2009. InterPro: the integrative protein signature database. Nucleic Acids Res. 37:D211D215. Krogh, A., B. Larsson, G. von Heijne, and E. L. L. Sonnhammer. 2001. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J. Mol. Biol. 305:567580. Kunapuli, U., T. Lueders, and R. U. Meckenstock. 2007. The use of stable isotope probing to identify key iron-reducing microorganisms involved in anaerobic benzene degradation. ISME J. 1:643653. Leang, C., X. Qian, T. Mester, and D. R. Lovley. 2010. Alignment of the c-type cytochrome OmcS along pili of Geobacter sulfurreducens. Appl. Environ. Microbiol. 12:40804084. Lies, D. P., et al. 2005. Shewanella oneidensis MR-1 uses overlapping pathways for iron reduction at a distance and by direct contact under conditions relevant for biolms. Appl. Environ. Microbiol. 71:44144426. Liu, M., et al. 2010. Bioelectricity generation by a Gram-positive Corynebacterium sp. strain MFC03 under alkaline condition in microbial fuel cells. Bioresource Technol. 101:18071811. Lovley, D. R. 2006. Bug juice: harvesting electricity with microorganisms. Nat. Rev. Microbiol. 4:497508. Lovley, D. R., and E. J. P. Phillips. 1988. Novel mode of microbial energy metabolism: organic carbon oxidation coupled to dissimilatory reduction of iron or manganese. Appl. Environ. Microbiol. 54:14721480. Marshall, C., and H. May. 2009. Electrochemical evidence of direct electrode reduction by a thermophilic Gram-positive bacterium, Thermincola ferriacetica. Energy Environ. Sci. 2:699705. Marsili, E., et al. 2008. Shewanella secretes avins that mediate extracellular electron transfer. Proc. Natl. Acad. Sci. U. S. A. 105:39683973. Mathis, B. J., et al. 2008. Electricity generation by thermophilic microorganisms from marine sediment. Appl. Microbiol. Biotechnol. 78:147155. Matias, V. R. F., and T. J. Beveridge. 2005. Cryo-electron microscopy reveals native polymeric cell wall structure in Bacillus subtilis 168 and the existence of a periplasmic space. Mol. Microbiol. 56:240251. Mehta, T., M. V. Coppi, S. E. Childers, and D. R. Lovley. 2005. Outer membrane c-type cytochromes required for Fe(III) and Mn(IV) oxide reduction in Geobacter sulfurreducens. Appl. Environ. Microbiol. 71:8634 8641. Methe, B., et al. 2003. Genome of Geobacter sulfurreducens: metal reduction in subsurface environments. Science 302:19671969. Nevin, K. P., and D. R. Lovley. 2000. Lack of production of electron-shuttling compounds or solubilization of Fe(III) during reduction of insoluble Fe(III)

41.

42. 43.

44. 45.

46.

47. 48. 49.

50. 51. 52.

53.

54. 55.

56.

57.

58.

59. 60.

61. 62.

oxide by Geobacter metallireducens. Appl. Environ. Microbiol. 66:2248 2251. Nevin, K. P., and D. R. Lovley. 2002. Mechanisms for accessing insoluble Fe(III) oxide during dissimilatory Fe(III) reduction by Geothrix fermentans. Appl. Environ. Microbiol. 68:22942299. Nevin, K. P., and D. R. Lovley. 2002. Mechanisms for Fe(III) oxide reduction in sedimentary environments. Geomicrobiol. J. 19:141159. Nevin, K. P., et al. 2008. Power output and columbic efciencies from biolms of Geobacter sulfurreducens comparable to mixed community microbial fuel cells. Environ. Microbiol. 10:25052514. Park, D., et al. 1997. Electrode reaction of Desulfovibrio desulfuricans modied with organic conductive compounds. Biotechnol. Tech. 11:145148. Pham, T. H., et al. 2008. Metabolites produced by Pseudomonas sp. enable a Gram-positive bacterium to achieve extracellular electron transfer. Appl. Microbiol. Biotechnol. 77:11191129. Rabaey, K., N. Boon, S. D. Siciliano, M. Verhaege, and W. Verstraete. 2004. Biofuel cells select for microbial consortia that self-mediate electron transfer. Appl. Environ. Microbiol. 70:53735382. Rabaey, K., et al. 2007. Microbial ecology meets electrochemistry: electricitydriven and driving communities. ISME J. 1:918. Reguera, G., et al. 2005. Extracellular electron transfer via microbial nanowires. Nature 435:10981101. Reguera, G., R. B. Pollina, J. S. Nicoll, and D. R. Lovley. 2007. Possible nonconductive role of Geobacter sulfurreducens pilus nanowires in biolm formation. J. Bacteriol. 189:21252127. Ronquist, F., and J. P. Huelsenbeck. 2003. Mrbayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:15721574. Schro der, I., E. Johnson, and S. de Vries. 2003. Microbial ferric iron reductases. FEMS Microbiol. Rev. 27:427447. Sharma, S., G. Cavallaro, and A. Rosato. 2010. A systematic investigation of multiheme c-type cytochromes in prokaryotes. J. Biol. Inorg. Chem. 15:559 571. Shi, L., T. C. Squier, J. M. Zachara, and J. K. Fredrickson. 2007. Respiration of metal (hydr)oxides by Shewanella and Geobacter: a key role for multihaem c-type cytochromes. Mol. Microbiol. 65:1220. Slobodkin, A. I. 2005. Thermophilic microbial metal reduction. Mikrobiologiia 74:581595. Sokolova, T. G., et al. 2005. Thermincola carboxydiphila gen. nov., sp. nov., a novel anaerobic, carboxydotrophic, hydrogenogenic bacterium from a hot spring of the Lake Baikal area. Int. J. Syst. Evol. Microbiol. 55:20692073. Srikanth, S., E. Marsili, M. C. Flickinger, and D. R. Bond. 2008. Electrochemical characterization of Geobacter sulfurreducens cells immobilized on graphite paper electrodes. Biotechnol. Bioengineer. 99:10651073. Von Canstein, H., J. Ogawa, S. Shimizu, and J. Lloyd. 2008. Secretion of avins by Shewanella species and their role in extracellular electron transfer. Appl. Environ. Microbiol. 74:615. Weber, K. A., L. A. Achenbach, and J. D. Coates. 2006. Microorganisms pumping iron: anaerobic microbial iron oxidation and reduction. Nat. Rev. Microbiol. 4:752764. Wrighton, K. C., et al. 2008. A novel ecological role of the Firmicutes identied in thermophilic microbial fuel cells. ISME J. 2:11461156. Wrighton, K. C., A. E. Engelbrektson, I. C. Clark, R. A. Melnyk, and J. D. Coates. 2011. Accentuate the positive: dissimilatory iron reduction by grampositive bacteria. In J. Stolz and R. Oremland (ed.), Microbial metal and metalloid metabolism: advances and applications. ASM Press, Washington, DC. Xing, D., Y. Zuo, S. Cheng, J. Regan, and B. Logan. 2008. Electricity generation by Rhodopseudomonas palustris DX-1. 42:4146-4151. Zavarzina, D. G., et al. 2007. Thermincola ferriacetica sp. nov., a new anaerobic, thermophilic, facultatively chemolithoautotrophic bacterium capable of dissimilatory Fe(III) reduction. Extremophiles 11:17.