Cytochrome OmcS Is Not Essential for Extracellular Electron Transport via Conductive Pili in Geobacter sulfurreducens Strain KN400

ABSTRACT

The multi-heme c-type cytochrome OmcS is one of the central components used for extracellular electron transport in the Geobacter sulfurreducens strain DL-1, but its role in other microbes, including other strains of G. sulfurreducens, is currently a matter of debate. Therefore, we investigated the function of OmcS in the G. sulfurreducens strain KN400, which is even more effective in extracellular electron transfer than the DL-1 strain. We found that deleting omcS from strain KN400 did not negatively impact the rate of Fe(III) oxide reduction and that the cells expressed conductive filaments. Replacing the wild-type pilin gene with the aro-5 pilin gene eliminated the OmcS-deficient strain’s ability to transport electrons to insoluble electron acceptors and diminished filament conductivity. These results are consistent with the concept that electrically conductive pili are the primary conduit for long-range electron transfer in G. sulfurreducens and closely related species. These findings, coupled with the lack of OmcS homologs in other microbes capable of extracellular electron transfer, suggest that OmcS is not a common critical component for extracellular electron transfer.

IMPORTANCE OmcS has been widely studied and noted to be one of the key components for extracellular electron exchange by the Geobacter sulfurreducens strain DL-1. However, the true importance of OmcS warrants further investigation because it is well known that few bacteria, even within the Geobacteraceae family, contain OmcS homologs, and many bacteria that are capable of extracellular electron transfer lack an abundance of any type of outer surface c-type cytochrome. In addition, there is debate about the importance of OmcS filaments in the mechanism of extracellular electron transport to insoluble electron acceptors by G. sulfurreducens. It has been suggested that filaments comprised of OmcS rather than e-pili are the predominant conductive filaments expressed by G. sulfurreducens. However, the results presented here, along with multiple other sources of evidence, indicate that OmcS filaments cannot be the primary, conductive, protein nanowires expressed by G. sulfurreducens.

INTRODUCTION

The Geobacter sulfurreducens strain DL-1 is one of the most widely studied microorganisms that is capable of extracellular electron transfer. Some studies have indicated that extracellular electron transfer by DL-1 is complex and involves several components, most notably the electrically conductive pili (e-pili) (1, 2) and numerous outer cell surface c-type cytochromes (3). One of the most important and thoroughly studied c-type cytochromes involved in extracellular electron transfer in the strain DL-1 is the six-heme protein, OmcS (4, 5). Strain DL-1 requires OmcS for growth with insoluble Fe(III) oxide but not soluble Fe(III) citrate at the electron acceptor (6). Although early studies found that deletion of the gene for OmcS inhibited growth of DL-1 on the anodes of microbial fuel cells (7), more detailed studies using electrochemically controlled anode potentials demonstrated within the same mutant that OmcS is not required for high density current production or biofilm conductivity (8–10).

Under some conditions, OmcS assembles into micrometer-long filaments (11–12). Based on these findings, it was suggested that the formation of OmcS filaments “explains the remarkable capacity of soil bacteria to transport electrons to remote electron acceptors for respiration and energy sharing” (12). However, OmcS homologs are found in only a few soil bacteria that are capable of extracellular electron exchange (2, 13). Furthermore, many bacteria that are capable of extracellular electron exchange do not produce any outer surface cytochromes (14–17). OmcS homologs are not prevalent within the genus Geobacter (13). For example, G. metallireducens, a close relative of G. sulfurreducens, does not contain any OmcS family proteins yet its rate of Fe(III) oxide reduction is significantly faster than G. sulfurreducens (18). Geobacter uraniireducens, one of the few microbes that has an OmcS family protein (13), lacks phenotypes that are associated with long-range electron transport through conductive filaments, such as producing high current densities on anodes (18). G. uraniireducens transfers electrons to Fe(III) oxides with a soluble electron shuttle and not through filament-mediated, direct electron transport (19).

To further evaluate whether OmcS is an important component for extracellular electron transfer in any microbes other than the G. sulfurreducens strain DL-1, we studied the G. sulfurreducens strain KN400, which is a strain that is more effective at extracellular electron transfer than the DL-1 strain (20, 21). We found that deleting the gene for OmcS had no impact on extracellular electron transfer or the production of electrically conductive filaments, suggesting a limited significance for OmcS beyond the G. sulfurreducens strain DL-1.

RESULTS AND DISCUSSION

An OmcS-deficient mutant of G. sulfurreducens effectively reduces Fe(III) oxide.

A strain of G. sulfurreducens KN400 in which omcS was deleted reduced Fe(III) oxide as fast as the wild-type strain during the active phase of Fe(III) oxide reduction (Fig. 1). In the initial transfer from fumarate medium, the lag period that occurred prior to rapid Fe(III) oxide reduction was longer in the OmcS-deficient strain (12 days) than for the wild-type strain (5 days). However, it is difficult to attribute physiological significance to this longer initial lag period because construction of the mutant required selection and recovery on medium with the soluble electron acceptor fumarate, a growth condition that can deadapt cells for growth by extracellular electron transfer (10). After several transfers, the lag phase in the OmcS-deficient mutant was comparable to that of the wild-type, and both strains continued to reduce Fe(III) oxide at similar rates (Fig. 1B). These results demonstrated that OmcS was not required for effective Fe(III) oxide reduction.

FIG 1.

(A) Fe(II) production from Fe(III) oxide reduction over time in the KN400 wild-type and OmcS-deficient strains after the first transfer into Fe(III) oxide medium from fumarate medium and (B) Fe(II) production from Fe(III) oxide reduction over time in the KN400 wild-type and KN400 OmcS-deficient strain after five consecutive transfers in Fe(III) oxide medium.

Deletion of omcS does not increase expression of the electron shuttle PgcA.

A previous study of a KN400 pilA-deletion mutant demonstrated that a strain could be adapted to effectively reduce Fe(III) oxide after hundreds of days (22). This adaptation was attributed to a mutation that substantially increased expression of the soluble extracellular c-type cytochrome PgcA (KN400_1784), which can function as an extracellular electron shuttle to facilitate electron transfer to Fe(III) oxides (22, 23). To evaluate the possibility of increased PgcA expression in the OmcS-deficient strain, outer cell surface proteins were sheared from Fe(III) oxide-grown cells and combined with cell-free filtrate. The proteins were analyzed on SDS-polyacrylamide gels stained for heme (Fig. 2A). PgcA, a 41 kDa protein, was expressed at low levels in the wild-type and OmcS-deficient strains, but not at levels that were high enough to account for the Fe(III) reduction phenotype that was previously observed in the adapted PilA-deficient strain (22). In fact, transcript abundance of the gene for PgcA was slightly lower in the OmcS-deletion strain than the wild-type strain, but these differences in transcript abundance were not significant (P > 0.05) (Fig. 2B).

FIG 2.

Lack of increased cytochrome expression in the OmcS-deficient mutant. (A) c-type cytochrome content of outer surface protein and cell-free filtrates from the wild-type and OmcS-deficient cultures of KN400 grown with Fe(III) oxide as the electron acceptor. Proteins were separated by SDS-PAGE and stained for heme. (B) Transcript abundance of pgcA in the wild-type and OmcS-deficient cultures of KN400 grown with Fe(III) oxide as the electron acceptor. Relative expression ratios were determined by the 2^-ΔΔCt method (42), using the housekeeping gene proC for data normalization. Differences in pgcA transcript abundance were not significant (P > 0.05). Technical triplicates and biological replicates were done for all samples.

e-Pili are required for long-range electron transport.

While it is possible that other outer cell surface c-type cytochromes may be able to compensate for the lack of OmcS, there were very few notable differences in the heme-stained gel (Fig. 2A). One intensely stained band was found in both the wild-type and OmcS-deletion strain below the 37 kDa marker. This band is likely OmcZ, which has previously been shown to be highly expressed by strains of KN400 that are adapted for Fe(III) oxide reduction (22). However, multiple studies have shown that deletion of omcZ has no effect on Fe(III) oxide reduction in the G. sulfurreducens strains DL-1 or KN400 (10, 22). A larger band was observed above the 75 kDa marker and is likely to be OmcB, an outer cell membrane c-type cytochrome (24) that is thought to be a key component in the transfer of electrons across the outer membrane (25).

To evaluate the potential role of c-type cytochrome filaments and e-pili in long-range extracellular electron transfer, the wild-type pilin gene in the OmcS-deficient mutant was replaced with the previously described (26) synthetic pilin gene, which codes for a pilin monomer in which five of the aromatic amino acids in the wild-type pilin are replaced with alanine (KN400 ΔomcS-Aro5) (Fig. S1). This ‘Aro-5′ pilin gene yields a pilin with greatly reduced conductivity (26–28). A control strain (KN400 ΔomcS-pilin control) was constructed in the same manner but with the wild-type pilin gene sequence (Fig. S1).

Atomic force microscopy (AFM) of cell culture aliquots drop-cast on a conductive 35 nm platinum-coated silicon wafer, as previously described (29), revealed abundant filaments emanating from each cell (Fig. 3A and C). As expected, none of the filaments emanating from the KN400 ΔomcS-Aro5 strain or the ΔomcS-pilin control strain had a morphology characteristic of OmcS filaments (diameter 4 nm, axial pitch 20 nm) (11, 12). Only filaments with a diameter of approximately 3 nm, which corresponded to e-pili (29), and 12 to 14 nm, which are presumably flagella, were observed (Fig. 3B). The filaments in the ΔomcS-pilin control strain were conductive with a conductance of 3.22 ± 0.18 nS, which was comparable to the conductance of 3.01 ± 0.20 nS for individual e-pili in the G. sulfurreducens strain DL-1 (29). Filaments with a diameter of ∼3 nm emanating from wild-type KN400 cells were also comparable with a conductance of 3.14 ± 0.15 nS (Fig. S3).

FIG 3.

Characterization of filaments emanating from KN400 ΔomcS-pilin control and KN400 ΔomcS-Aro5. (A) AFM amplitude image of KN400 ΔomcS-pilin control. (B) Typical height profiles for pili (designated by yellow lines) and flagella (designated by red lines) shown in (A), as determined by cross sections from corresponding height images (Fig. S2A). (C) AFM amplitude image of KN400 ΔomcS-Aro5. (D) Comparison of point-mode current response (I-V) spectroscopy for pili from KN400 ΔomcS-pilin control (blue) and KN400 ΔomcS-Aro5 (green). The responses shown are representative of three different measurements on each of three individual filaments (Fig. S2B).

In contrast, the filaments emanating from KN400 ΔomcS-Aro5 were poorly conductive (0.023 ± 0.004 nS) (Fig. 3D and Fig. S5), which is consistent with previous results in which the Aro-5 pilin gene was expressed in the DL-1 strain of G. sulfurreducens (26, 27, 29).

Along with e-pili, OmcZ has been shown to be a key component for electron transfer to electrodes in G. sulfurreducens current-producing biofilms (10). OmcZ was produced in high abundance in KN400 ΔomcS-Aro5 cells grown on an anode surface (Fig. 4A). Regardless of the abundance of OmcZ, the KN400 ΔomcS-Aro5 strain was deficient in both Fe(III) oxide reduction (Fig. 4B) and current production (Fig. 4C). These results demonstrate that pili conductivity rather than cytochrome-based filaments is the essential feature for long-range extracellular electron transport in the G. sulfurreducens strain KN400.

FIG 4.

Characterization of G. sulfurreducens strain KN400 ΔomcS-pilin control and KN400 ΔomcS-Aro5. (A) c-type cytochrome content from biofilms grown on an anode (+300 mV versus Ag/AgCl) surface. (B) Fe(II) production from Fe(III) oxide reduction over time after transferring into Fe(III) oxide medium from ferric citrate medium. (C) Current production on the anode of microbial fuel cells (+300 mV versus Ag/AgCl). Results shown are representative results from triplicate experiments.

Implications.

The results presented here demonstrated that the G. sulfurreducens strain KN400 does not require OmcS for effective extracellular electron transfer or production of electrically conductive filaments. These results suggest that the importance of OmcS in extracellular electron transfer is limited, even among the sulfurreducens species of Geobacter.

It has been suggested (12) that filaments comprised of OmcS rather than e-pili are the predominant conductive filaments expressed by G. sulfurreducens, but multiple lines of evidence have refuted this hypothesis (2, 17, 29). For example, direct examination of filaments emanating from G. sulfurreducens found that only 10% of the filaments were comprised of OmcS and that the remaining 90% were e-pili (29). When a short peptide tag was included in a synthetic gene for the pilin monomer, all filaments observed emanating from the cell contained the peptide tag (30). In addition, the conductivity of filaments harvested from G. sulfurreducens can be increased or decreased by modifying the abundance of aromatic amino acids in the pilin monomer (19, 26, 27, 31–33), a modification that did not influence the conductivity or abundance of OmcS filaments (29). Also, construction of strains that produce an abundance of OmcS, but express synthetic pilins that yield poorly conductive pili, are defective in long-range electron transfer (19, 26, 27, 33). The findings reported here and in another recent study (34) show that deleting the gene for OmcS has no impact on the conductivity of the filaments harvested from G. sulfurreducens, further demonstrating that OmcS filaments cannot be the primary conductive protein nanowires expressed by G. sulfurreducens. Thus, the actual role of OmcS in G. sulfurreducens warrants further study.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

Geobacter sulfurreducens strain KN400 (21), the ΔomcS strain of G. sulfurreducens DL-1 (6), and the ‘Aro-5′ and ‘Gent’ strains of G. sulfurreducens DL-1 (26) were obtained from our laboratory culture collection. All G. sulfurreducens strains were routinely cultured under strict anaerobic conditions (80% N₂ and 20% CO₂) with 10 mM acetate provided as the sole electron donor and either 100 mM Fe(III) oxide, 56 mM ferric citrate (FC), or fumarate (40 mM) provided as the sole terminal electron acceptor (35, 36). Cultures were incubated at 30°C. The medium composition per liter of deionized water with either Fe(III) oxide or fumarate as the sole electron acceptor was prepared as follows: 0.42 g KH₂PO₄, 0.22 g K₂HPO₄, 0.2 g NH₄Cl, 0.38 g KCl, 0.36 g NaCl, 0.04 g CaCl₂ 2H₂0, 0.1 g MgSO₄ 7H₂O, 1.8 g NaHCO₃, 0.5 g Na₂CO₃, 1.0 mL of 100 mM Na₂SeO₄, 10.0 ml of DL vitamin solution listed in (37), and 10.0 mL of NB trace mineral solution listed in (35). To prepare cultures for genetic manipulation and create KN400 ΔomcS, KN400 △omcS-Aro5, and KN400 ΔomcS-pilin control, cells were grown in either liquid or agar-solidified acetate-fumarate medium (35) then placed in an anaerobic chamber with a N₂-CO₂-H₂ (83%, 10%, and 7%, respectively) atmosphere at a temperature of 30°C.

KN400 mutant construction.

To construct the KN400 omcS deletion mutant, primers OmcS_fwd (GGTCGTGATGCTCGATCCGGAAG) and OmcS_rev (GGTTGGCGGTAAGGAGGTGCCG) were used to amplify the mutated region from the DL-1 omcS-deleted strain. For construction of G. sulfurreducens strains KN400 ΔomcS-Aro5 and KN400 ΔomcS-pilin control, linear DNA fragments, including the gentamicin resistant cassette, and either the wild-type pilA allele (for KN400 ΔomcS-pilin control) or mutant allele pilA^{F53A, Y56A, Y61A, F80A, Y86A} (for KN400 ΔomcS-Aro5) were amplified from G. sulfurreducens strain Gent or Aro-5, respectively, using primers GsuAro5_fwd (CAGGAAGCTTACGTTCCGTTCTTTCC) and GsuAro5_rev (CAGATGACTACTGCGACTTCCACTCG).

Polymerase chain reaction products were purified using the QIAquick gel extraction kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions, and verified by Sanger sequencing. Electrocompetent G. sulfurreducens strain KN400 wild-type cells (for the ΔomcS strain) and KN400 ΔomcS cells (for the ΔomcS-Aro5 and ΔomcS-pilin control strains) were prepared from cultures maintained on acetate-fumarate medium, and all manipulations were carried out on ice in an anaerobic chamber with a N₂-CO₂-H₂ (83%, 10%, and 7%, respectively) atmosphere at a temperature of 30°C. Two hundred milliliters of cells were harvested (when they reached an optical density at 600 nm of 0.2) by centrifugation for 8 min at 4300 × g and at 4°C. Cells were washed twice with electroporation buffer (1 mM HEPES pH 7.0, 1 mM MgCl₂, and 175 mM sucrose). Cell pellets were resuspended in electroporation buffer to achieve a final concentration of 10¹¹ cells/mL. One hundred nanograms of the respective purified PCR product was electroporated into the appropriate competent cells at 14.7 kV/cm for 6 ms (35). The cells were then inoculated into fumarate-acetate medium and allowed to recover for 8 h at 30°C prior to plating on fumarate-acetate agar. Spectinomycin (25 μg/mL) was added to the medium for selective purposes for the KN400 ΔomcS strain, and gentamicin (20 μg/mL) was added for selection of ΔomcS-Aro5 and ΔomcS-pilin control strains. Replacement of wild-type alleles by mutant alleles was verified by PCR and Sanger sequencing.

Cell-free filtrate SDS-PAGE and protein identification.

After five consecutive transfers in Fe(III) oxide medium, outer cell surface and cell-free proteins in the medium were collected from KN400 wild-type and KN400 ΔomcS cells when Fe(II) concentrations reached approximately 40 mM during the exponential growth phase. Loosely bound outer cell surface proteins from 100 mL of culture were sheared using a Waring blender operated at room temperature at low speed for 2 min (6, 38). Blended samples were then centrifuged at 10,000 × g for 20 min to remove cells and insoluble Fe(III) oxide, and the supernatant was collected and concentrated with an Amicon Ultra-15 centrifugal filter unit (Millipore, Billerica, MA).

Current-producing biofilms were grown until the current from KN400 ΔomcS-pilin control and KN400 ΔomcS-Aro5 were stable (8 d). The biofilms were gently harvested by scraping and washing with isotonic wash buffer (10). Proteins were isolated as previously described (6).

Proteins were quantified with a Micro BCA protein assay kit (Thermo Scientific, Rockford, IL) according to the manufacturer’s instructions. Equal amounts of protein (5 μg) were separated by SDS-PAGE in glycine-buffered 12.5% polyacrylamide gels. Heme groups were identified using the heme stain to detect peroxidase activity with H₂O₂ and 3,3′,5,5′-tetramethylbenzidine (TMBZ) (39).

RT-qPCR.

Total RNA was extracted from triplicate cultures during exponential growth in acetate (10 mM)-Fe(III) oxide (100 mM) medium using the HG method (40). For this extraction, 50 mL of cells were pelleted by centrifugation at 3000 × g for 20 min at 4°C and immediately resuspended in 10 mL of preheated (65°C) HG buffer (100 mM Tris-HCl pH 8, 100 mM NaCl, 10 mM EDTA, 2.5% β-mercaptoethanol, 1% sodium dodecyl sulfate, 2% Plant RNA Isolation Aid [Ambion, Woodward, TX, USA], 5 mM ascorbic acid, 0.6 mg/mL Proteinase K, and 5 mg/mL lysozyme). The mixture was incubated at 65°C for 10 min and then 2 μL Superase-In (Ambion) and 0.025 mM CaCl₂ were added. After centrifugation for 10 min at 16,100 × g, the supernatant was transferred to new 2 mL screw cap tubes. Fifty microliters of Plant RNA Isolation Aid (Ambion), 4 μL of 5 mg/mL linear acrylamide (Ambion), 600 μL of preheated (65°C) acidic phenol pH 4.5 (Ambion), and 400 μL of chloroform-isoamyl alcohol (24:1; Sigma, St. Louis, MO, USA) were added, and rotated for 10 min. The samples were then centrifuged for 5 min at 16100 × g at which point the aqueous layer was removed and transferred to a new tube containing 100 μL of 5 M ammonium acetate (Ambion), 20 μL of 5 mg/mL glycogen (Ambion), and 1 mL of cold isopropanol. The samples were left to precipitate at −30°C for 1 h.

After precipitation, the nucleic acids were pelleted by centrifugation at 16100 × g for 30 min. The pellet was washed twice with 70% ethanol, dried, and resuspended in sterile diethylpyrocarbonate- treated (DEPC) water (Ambion). The RNA was purified with a RNeasy MiniElute cleanup kit (Qiagen) according to the manufacturer's instructions and treated with Turbo DNA-free DNase (Ambion). The RNA samples were tested for genomic DNA (gDNA) contamination by PCR amplification of the 16S rRNA gene. The concentration and quality of the RNA samples were determined with a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE) and by agarose gel electrophoresis. All RNA samples had A260/A280 ratios of 1.8 to 2.0, indicating high purity. The cDNA was generated with a TransPlex whole-transcriptome amplification kit (Sigma-Aldrich, St. Louis, MO) according to the manufacturer's instructions.

Primer pairs with amplicon sizes of 100 to 200 bp were designed for the G. sulfurreducens pgcA (pgcAF, ACGGGAATAACTGTCAGGCA; pgcAR, CGTGAGGGTTGAAAGGAACC) gene. Expression of this gene was normalized with expression of proC (proCF, ACCGATGACGATCTGTTCTTT; proCR, ATGAGCTTTTCCTCCACCAC), a housekeeping gene constitutively expressed in Geobacter species (41). Relative levels of expression of the studied genes were calculated by the 2−^ΔΔCT threshold cycle (CT) method (42).

Power SYBR green PCR master mix (Applied Biosystems [ABI], Foster City, CA) and an ABI 7500 real-time PCR system were used to amplify and to quantify the PCR products. Each reaction mixture consisted of forward and reverse primers at a final concentration of 200 nM, 5 ng of cDNA or gDNA, and 12.5 μL of Power SYBR green PCR master mix (Applied Biosystems).

Analytical techniques.

Organic acids were monitored with high performance liquid chromatography (HPLC) as previously described (43). Fe(II) concentrations were measured with the ferrozine assay (44). For this assay, samples of Fe(III) oxide cultures were dissolved at a 1:10 ratio in 0.5 N HCl for 2 h at room temperature. Next, 0.1 mL of sample extract was added to 4.9 mL of ferrozine (1 g/L in 50 mM HEPES buffer at pH 7) and mixed. Concentrations of Fe(II) were determined by measuring absorbance at 562 nm in a split-beam dual-detector spectrophotometer (Spectronic Genosys 2; Thermo Electron Corp., Mountain View, CA).

Microbial fuel cells.

Strains were grown in H-cell culturing systems to test the capacity to produce current as previously described (10). Graphite block electrodes (65 cm²) were in the two chambers and separated by Nafion 117 cation-exchange membranes (Electrolytica). The anode was poised at +300 mV versus Ag/AgCl by a potentiostat (Gamry Instruments). Cells were grown in freshwater medium (10) with 10 mM acetate and 40 mM fumarate. Once cultures reached an A₆₀₀ of ∼0.3, the system was switched to a continuous flow through mode in which the anode chamber was continuously supplied with freshwater medium containing 10 mM acetate at a dilution rate of 0.15/h. Both the anode and cathode chambers were continuously bubbled with N₂-CO₂ (80:20).

Atomic force microscopy analysis.

Biological samples were prepared by growing strains to mid to late log phase in NBF medium at 25°C. The filaments associated with cells were directly examined under atomic force microscopy (Asylum Research, Oxford Instrument). Fifty microliters of the culture was drop-cast onto a conductive 35 nm platinum-coated silicon wafer as previously described (29). After air drying for 15 min, samples were rinsed twice with 50 μL of deionized water. Excess water was removed with filter paper, and samples were then equilibrated at 40% humidity for 1 h at 25°C in a scanning chamber.

The filaments were first observed under tapping mode (AC-air topography) with a Pt/Ir-coated tip (PtSi-FM, NanoWorld AG) at ∼2.0 N/m spring force constant and ∼70 kHz resonance frequency. Contact mode was then applied and the tip was lightly (30 nN force) placed on top of the filaments as the translatable top electrode. The conductance of individual filaments was evaluated as previously described (29). Quadruplicate amplitude of ±0.4 V voltage sweeping at a frequency of 0.99 Hz was applied for current voltage (I-V) response. A linear slope between −0.2 V and +0.2 V in I-V response was used to calculate the conductance (G = I/V). Three independent points from three individual filaments were analyzed.