|Home | About | Journals | Submit | Contact Us | Français|
Previous studies have demonstrated that Geobacter sulfurreducens requires the c-type cytochrome OmcZ, which is present in large (OmcZL; 50-kDa) and small (OmcZS; 30-kDa) forms, for optimal current production in microbial fuel cells. This protein was further characterized to aid in understanding its role in current production. Subcellular-localization studies suggested that OmcZS was the predominant extracellular form of OmcZ. N- and C-terminal amino acid sequence analysis of purified OmcZS and molecular weight measurements indicated that OmcZS is a cleaved product of OmcZL retaining all 8 hemes, including 1 heme with the unusual c-type heme-binding motif CX14CH. The purified OmcZS was remarkably thermally stable (thermal-denaturing temperature, 94.2°C). Redox titration analysis revealed that the midpoint reduction potential of OmcZS is approximately −220 mV (versus the standard hydrogen electrode [SHE]) with nonequivalent heme groups that cover a large reduction potential range (−420 to −60 mV). OmcZS transferred electrons in vitro to a diversity of potential extracellular electron acceptors, such as Fe(III) citrate, U(VI), Cr(VI), Au(III), Mn(IV) oxide, and the humic substance analogue anthraquinone-2,6-disulfonate, but not Fe(III) oxide. The biochemical properties and extracellular localization of OmcZ suggest that it is well suited for promoting electron transfer in current-producing biofilms of G. sulfurreducens.
The properties of proteins involved in the production of current in microbial fuel cells are of interest because it seems likely that the applications of microbial fuel cells could be further optimized and expanded if the mechanisms for current production were better understood. Geobacter sulfurreducens produces current densities that are among the highest of any known microorganism (40, 59) and is closely related to the microorganisms that most effectively colonize the anodes of microbial fuel cells from a diversity of complex microbial communities (4, 18-20, 24-27, 53).
A comparison of gene expression in current-producing biofilms of G. sulfurreducens growing on the graphite anodes of microbial fuel cells versus gene expression in biofilms growing on the same material but with fumarate as the electron acceptor demonstrated that transcript abundance for the gene for the outer-surface c-type cytochrome OmcZ was much higher in the current-producing cells (39). Deletion of omcZ greatly inhibited current production (39) and increased the resistance in electron transfer between the biofilm and the anode surface (48). G. sulfurreducens has an abundance of other outer-surface c-type cytochromes that are important in other forms of extracellular electron transfer, such as the reduction of Fe(III) (25, 37), U(VI) (49), or humic substances (58), and some may play an important electron storage role (13), but OmcZ is the only c-type cytochrome that was found to be essential for optimal current production (39).
Previous proteome analyses had revealed that most of the more than 100 putative c-type cytochrome genes in G. sulfurreducens are expressed (11, 12). However, only a few of them have been purified and characterized. They include PpcA, an abundant triheme periplasmic c-type cytochrome that plays an important role in Fe(III) reduction (28, 38, 43), as well as several monoheme cytochromes that may play a sensory role (22, 44, 45). None of the outer-surface c-type cytochromes has been purified to homogeneity. A c-type cytochrome preparation, designated FerA, was found to contain components of the NADH dehydrogenase Fe(III)-reducing protein complex (34, 35) and had the capacity to reduce Fe(III). It was later concluded that FerA was probably a mixture of the multiheme cytochrome homologues OmcB and OmcC (25). Gene deletion studies demonstrated that OmcB is required for optimal Fe(III) reduction but OmcC is not (24, 25). Thus, there is a dearth of information on the biochemistry of outer-surface cytochromes of G. sulfurreducens. Studies with Shewanella oneidensis have demonstrated the importance of purifying and characterizing outer-surface cytochromes in order to understand the mechanisms for extracellular electron transfer (50, 51). Here, we report on the isolation and characterization of OmcZ from G. sulfurreducens.
Geobacter sulfurreducens strain ZKI was produced from G. sulfurreducens strain DL-1 as described by B.-C. Kim (unpublished data). Briefly, the omcZ gene, combined with 500 bp containing the ompJ promoter sequence, was integrated in strain DL-1. Strain ZKI, which overproduces the omcZ gene product, was used for OmcZ small-form (OmcZS; 30-kDa) purification. Strains DL-1 and ZKI and an omcZ-deficient mutant (39) were cultured anaerobically in acetate-fumarate medium as previously described (8).
Subcellular fractions were prepared using modifications of previously described methods (9, 21, 41). Wild-type and strain ZKI cells in their mid-log and stationary growth phases in 100 ml of acetate-fumarate medium were harvested by centrifugation for 15 min at 6,000 × g at 4°C. The supernatants were reserved as the culture supernatant fraction. The harvested cells were washed with 50 ml of spheroplast wash medium consisting of 0.42 g/liter KH2PO4, 0.22 g/liter K2HPO4, 0.38 g/liter KCl, 4.96 g/liter NaCl, 1.8 g/liter NaHCO3, and 0.5 g/liter Na2CO3. The washed cells were pelleted by centrifugation for 6 min at 6,000 × g at 4°C and resuspended in 30 ml of spheroplast wash medium containing 350 mM sucrose. Following another centrifugation for 6 min at 6,000 × g at 4°C, the cells were resuspended in 10 ml of 250 mM Tris-HCl (pH 7.5). After 1 min of incubation at 30°C, 1 ml of 500 mM EDTA (pH 7.5) was added, followed by the addition of 10 ml of 700 mM sucrose at 2 min, 150 mg lysozyme at 3.5 min, and 20 ml of water at 4 min. Spheroplast formation in the suspension was confirmed by transmission electron microscopy (negatively stained whole mounts). After centrifugation for 10 min at 20,000 × g at 4°C to pellet the spheroplasts, the supernatant was reserved as the periplasmic fraction. The pellet was resuspended in 20 ml of 100 mM Tris-HCl buffer (pH 7.5). A few crystals of DNase I were added, and the spheroplasts were disrupted by sonication for 5 min on ice (Sonic Dismembrator F550; Fisher Scientific). The resulting crude extract was centrifuged for 30 min at 3,000 × g at 4°C. The pellet was reserved as the cell debris fraction. The supernatant was subsequently centrifuged for 30 min at 20,000 × g at 4°C. The supernatant was reserved as the cytoplasmic fraction. The pellet was resuspended in 4 ml of 100 mM Tris-HCl buffer (pH 7.5). Following the addition of 4 ml of 100 mM Tris-HCl (pH 7.5) containing 2% (wt/vol) lauroylsarcosine, the suspension was stirred for 15 min at room temperature and then centrifuged for 30 min at 125,000 × g at 4°C. The supernatant was reserved as the inner membrane fraction. The pellet was resuspended in 200 μl of 100 mM Tris-HCl buffer (pH 7.5) and reserved as the outer membrane fraction. All the fraction samples were concentrated, and their buffer was replaced with 100 mM Tris-HCl (pH 7.5) by ultrafiltration with Amicon Ultra-15 Centrifugal Filter Units (10,000 MW; Millipore) or Nanosep 10k Omega (Pall).
All purification steps were performed at room temperature. G. sulfurreducens strain ZKI was anaerobically grown in six 1.5-liter volumes of acetate-fumarate medium (8) to stationary phase at 30°C. The cells were harvested and resuspended in Tris-HCl buffer (50 mM Tris-HCl, pH 7.0). After disruption by freeze-thaw 3 times, the pellet was collected as an insoluble fraction by centrifugation at 6,000 × g for 15 min at room temperature and suspended in 200 ml of Tris-HCl buffer, followed by centrifugation at 12,000 × g for 10 min. The pellet was washed twice with 120 ml of 50 mM Tris-HCl containing 1% SDS and suspended in 60 ml of 50 mM Tris-HCl buffer, and 60 ml of 10% 3-(N,N-dimethylmyristyl-ammonio)propanesulfonate (Zwittergent 3-14) solution was added. The insoluble fraction was collected by centrifugation and suspended in 10 ml of 10% Zwittergent 3-14 solution. After centrifugation at 12,000 × g for 10 min, the red supernatant was collected and diluted to 1:4 with 50 mM Tris-HCl buffer. The red material containing OmcZS was precipitated by centrifugation at 12,000 × g for 10 min because of its insolubility in highly concentrated Tris-HCl buffer. At this step, OmcZS was the only protein in the precipitant identified on the SDS-PAGE gel. The OmcZ large form (OmcZL; 50 kDa) was not detected in the precipitant because the ZKI stationary-phase cells contain very little OmcZL, which is more soluble than OmcZS according to the subcellular fractioning. The pellet was resuspended in 10 ml of 1% SDS solution and incubated at 90°C for 5 min, followed by filtration with Nanosep 300k Omega (Pall). To remove SDS, the solvent of the filtered solution was replaced in 5 mM Tris-HCl buffer by ultrafiltration using Nanosep 10k Omega (Pall). OmcZS was purified by gel filtration chromatography using a fast protein liquid chromatography (FPLC) system (Pharmacia Biotech), a HiPrep Sephacryl S-200 HR column (GE Healthcare), and buffer containing 50 mM Tris-HCl (pH 7.0) and 0.1 mM NaCl. Fractions containing OmcZS were collected and concentrated with a Vivaspin 20 (10-kDa cutoff; Sartorius).
Protein concentrations were determined by the Bradford method with bovine serum albumin (BSA) as a standard (6), using Quick Start Bradford Dye Reagent (Bio-Rad). For outer membrane and inner membrane fractions, cell debris protein concentrations were measured by the bicinchoninic acid method (52). For outer membrane and cell debris fractions, protein concentrations were determined after the proteins were boiled with SDS (0.5% final concentration) for 10 min to solubilize them as much as possible. SDS-PAGE analyses were performed using 12.5% (wt/vol) polyacrylamide gels. Proteins were stained with Coomassie brilliant blue R-250 and were heme stained as previously described (14, 56) for identification of c-type cytochromes. SeeBlue Plus2 Pre-Stained Standard (Invitrogen) was used as a protein molecular mass standard. Western blotting was performed using polyvinylidene difluoride (PVDF) membranes and a One-Step Western Blot Kit (GenScript), according to the manufacturer's instructions. Polyclonal antibodies to OmcZ were produced against purified OmcZS in New Zealand White rabbits (New England Peptide). The antibody against OmcZS was purified by using blotted antigen. The purified OmcZS was separated by SDS-PAGE and blotted to a PVDF membrane using a semidry blotter. A portion of the blot containing OmcZS was cut and washed with TBST buffer (20 mM Tris-HCl [pH 7.5], 0.5 M NaCl, and 0.05% Tween 20). The portion was blocked with 3% (wt/vol) BSA in TBST buffer for 30 min, followed by washing with TBST buffer, and was soaked in crude serum with antibody to OmcZS. After incubation for 4 h, the portion was washed with TBST buffer 3 times. The antibody bound to the portion was eluted with 0.1 M glycine-HCl (pH 2.7) buffer. The eluted antibody was neutralized with 1/5 the volume of 1 M Tris-HCl (pH 7.5). The purified antibody was used at 1:1,000 dilution. The Western blots were scanned and digitized with NIH Image J software for the subcellular-localization analysis of OmcZL and OmcZS. The integrated pixel densities (the pixel density of a band after background subtraction) of OmcZL and OmcZS were used as the intensities for calculation. The percentage of OmcZL or OmcZS molecules that was present in each subcellular fraction was calculated. The total relative amount of OmcZL or OmcZS in each subcellular fraction was calculated by multiplying the ratio of the total protein in each fraction to the amount of the protein loaded in each well of the gel. This multiplied intensity gave the relative amount of OmcZL or OmcZS in that fraction. The multiplied intensity of a fraction—for example, the cytoplasmic fraction—was divided by the total of the multiplied intensities of all the fractions (cytoplasmic, inner membrane, periplasmic, outer membrane, and culture supernatant fractions), which gave the relative localization ratio of the cytoplasmic fraction. The localization ratios of the other subcellular fractions were calculated in the same manner. The localization ratios provided qualitative rather than quantitative relative amounts in the different subcellular fractions. The loosely bound outer membrane-enriched protein fractions (LBOP) of the cells were prepared as previously described (37).
The N terminus sequence data were obtained by Edman sequencing at the Rockefeller University Protein/DNA Technology Center. The C terminus sequence data of OmcZS were obtained at Commonwealth Biotechnologies, Inc., by a method using carboxypeptidase. Carboxypeptidase Y and the Amino Quant high-performance liquid chromatography (HPLC) system (Agilent) were used for enzymatic degradation and amino acid quantification, respectively. To obtain internal sequence data, the purified OmcZS bands in SDS-PAGE gels were excised and sent to the Laboratory for Proteomic Mass Spectrometry at the University of Massachusetts Medical School for tryptic digestion, followed by liquid chromatography-tandem mass spectrometry (LC-MS-MS) analysis.
The relative molecular mass of OmcZS was estimated by gel filtration chromatography using an MW-GF-200 kit (Sigma). Mass spectra from electrospray ionization (ESI)-MS were acquired on a QStar-XL hybrid quadrupole time-of-flight mass spectrometer (ABI/MDS-Sciex), equipped with an ESI source. Purified OmcZS was dissolved in a mixture of water, ethanol, and acetic acid (50:50:3 [vol/vol/vol]) and was analyzed by ESI-MS. Analysis was also conducted on OmcZS dissolved in 10 mM ammonium bicarbonate. Voltages were adjusted to obtain the best ion transmission of m/z 3,000 corresponding to the protein. An MS scan (m/z 500 to 2,200) was performed in the positive ion acquisition mode.
UV/visible-light absorption spectrum measurements were performed using a Cary 50 Bio UV-visible spectrophotometer (Varian) in the range of 275 to 650 nm. The reduced OmcZS samples were obtained by adding 5 μl of 100 mM dithiothreitol (DTT) to the oxidized samples. CD spectra were recorded in the far-UV region using a Jasco J-715 spectropholarimeter at room temperature. A 2-mg/ml purified OmcZS sample in water was measured using a cuvette with a path length of 0.01 cm. Cytochrome c from horse heart (Sigma) was used as a spectrum standard for c-type cytochrome analysis. The proportions of the helices, sheets, and turns were calculated using CDNN version 2.1 (3).
Pyridine hemochrome analysis was performed as previously described (2). After the addition of NaOH and pyridine (final concentrations, 75 mM and 2.1 M, respectively), the oxidized form of the protein was obtained by the addition of 10 μl of 150 mM potassium ferricyanide solution. The spectrum was recorded, and the protein was then reduced by the addition of 1 M DTT until no further increase was observed at 550 nm. The heme c content of the purified OmcZS was calculated using the difference millimolar extinction coefficient of 19.1 mM−1 cm−1 at 550 nm for the pyridine ferrohemochrome minus the pyridine ferrihemochrome (1).
The thermal stability of OmcZS was investigated by differential scanning calorimetry (DSC) using a VP-DSC/ETR (Microcal). A sample (0.5 mg/ml) in 10 mM sodium phosphate buffer (pH 7.0) was placed in the sample cell, and the same buffer was placed in the reference cell. The DSC scan was run at a rate of 1°C/min.
Redox titrations of the isolated form of OmcZS were followed by visible spectroscopy inside an anaerobic glove box kept at <1 ppm oxygen. As previously described (29), the protein solutions (~0.16 mg/ml) were in 32 mM phosphate buffer with NaCl (100 mM final ionic strength) at pH 7 and 298 K. Each redox titration was performed in both the oxidative and reductive directions, using sodium dithionite and potassium ferricyanide solutions as the reductant and oxidant, respectively. To ensure a good equilibrium between the redox centers and the working electrode, a mixture of the following redox mediators was added to the solution, all at approximately 1.5 μM final concentration: phenazine methosulfate, phenazine ethosulfate, gallocyanine, methylene blue, indigo tetrasulfonate, indigo trisulfonate, indigo disulfonate, 2-hydroxy-1,4-naphtoquinone, antraquinone-2,6-disulfonate, antraquinone-2-sulfonate, safranine 0, neutral red, benzyl viologen, diquat, and methyl viologen. These mediators covered the potential range of −440 to +80 mV. The OmcZS reduced fraction was determined by integrating the area of the α-band above the line connecting the flanking isobestic points (544 and 561 nm) to subtract the optical contribution of the redox mediators, as previously described (42).
The range of electron acceptors reduced by OmcZS was determined with a spectrophotometric assay. The purified OmcZS (55.4 μg) was dissolved in 0.9 ml of Tris-HCl (pH 7.0) and anaerobically reduced with 5 μl of 100 mM DTT. The spectrum of the DTT-reduced OmcZS was recorded between 275 nm and 650 nm. Ten microliters of the potential electron acceptor was added from 100 mM anoxic stock solutions of Fe(III) citrate, potassium chromate, Au(III) chloride trihydrate, or anthraquinone-2,6-disulfonate (AQDS); a 20 mM stock solution of U(VI) acetate; or a suspension of Mn(IV) oxide or Fe(III) oxide at 100 mM per liter (7, 33).
The PSORT algorithm (http://psort.ims.u-tokyo.ac.jp) and SignalP 3.0 Server (http://www.cbs.dtu.dk/services/SignalP) were used to predict the cell localization and signal peptide cleavage sites of the OmcZ protein. The SOSUI engine ver. 1.11 (http://bp.nuap.nagoya-u.ac.jp/sosui) and TMHMM server ver. 2.0 (http://www.cbs.dtu.dk/services/TMHMM-2.0) were used to predict the transmembrane region in the OmcZ protein.
The omcZ gene (GSU2076; GenBank accession no. AAR35452) encodes a protein of 473 amino acids with 7 typical heme c-binding motifs (CXXCH), as well as the possible heme-binding sequence CX14CH (Fig. (Fig.1).1). The program PSORT predicted OmcZ to be localized in the outer membrane (score, 0.39; certainty, 0.743). The topology prediction programs SOSUI and TMHMM suggested that the N-terminal region (amino acids 5 to 27) forms a transmembrane helix. According to the SignalP server 3.0, OmcZ has a signal sequence with a predicted cleavage site between amino acids 24 and 25 in the transmembrane helix. The cleaved polypeptide, which consists of 449 residues, contains no transmembrane site.
No close OmcZ homologues from outside the genus Geobacter were found in the GenBank database. The amino acid sequence of OmcZ had the following percent identities to Geobacter proteins: 50.3 to GSU1334 from G. sulfurreducens (GenBank accession no. AAR34710), 46.2 to Gbem_3056 from Geobacter bemidjiensis strain Bem (ACH40058), 45.6 to GM21_1194 from Geobacter sp. strain M21 (EDV70209), and 44.5 to Gmet_0930 from Geobacter metallireducens strain GS-15 (ABB31172). The unusual heme-binding motif CX14CH was completely conserved in GM21_1194 and Gbem_3056 and was also conserved with shorter distances between the cysteine residues in GSU1334 (CX13CH) and Gmet_0930 (CX11CH) (Fig. (Fig.1).1). From amino acid 350 to the C terminus, OmcZ showed homology with the D domain of cyclodextrin glycosyltransferases from Bacillus spp. The D domain is well conserved among the cyclodextrin glycosyltransferases, but the function of the domain is still unknown (46).
It was previously reported that the loosely bound outer membrane fraction from G. sulfurreducens contained two forms of OmcZ with masses of approximately 50 and 30 kDa on SDS-PAGE gels (39). The large and the small forms were designated OmcZL and OmcZS, respectively.
OmcZS was more abundant than OmcZL in wild-type cells and in strain ZKI, which had been engineered to overexpress OmcZ (Fig. (Fig.2).2). In wild-type cells, both forms of OmcZ were detected in the outer membrane fraction and culture supernatant, but only OmcZS was detected in the cell debris fraction, which primarily contained insoluble extracellular matrix material (Fig. (Fig.2A).2A). More OmcZS was found in the supernatant and cell debris fractions of stationary-phase cells than in mid-log-phase cells. To evaluate the relative amounts of protein in the fractions, the signal intensities of the bands of the Western blots were multiplied by the total amount of protein in each fraction. This analysis indicated that OmcZS was most abundant in the cell debris fraction regardless of the growth phase (data not shown). When OmcZ was overexpressed in strain ZKI, some OmcZL was also detected in the periplasm of mid-log-phase cells, but OmcZS was not (Fig. 2A and B). In both mid-log- and stationary-phase cells, OmcZS was mainly distributed in the cell debris fraction and partially in the outer membrane fraction (Fig. 2B and C).
These results demonstrated that the mature protein OmcZS can be found in the outer membrane and cell debris fractions but not in the periplasmic or cytoplasmic fraction. The recovery of some OmcZL, but not OmcZS, in the periplasm suggests that OmcZL is cleaved during secretion across the outer membrane, as has been reported for the secretion of other outer-surface proteins in other microorganisms (55).
OmcZS was purified from the stationary-phase cells of strain ZKI because it was much more abundant than OmcZL in these cells (Fig. (Fig.3A).3A). It was purified by detergent extraction followed by gel filtration chromatography. After detergent extraction, the protein sample was already purified to near homogeneity (Fig. (Fig.3B).3B). The detergent-extracted OmcZS, originally solubilized in 10% Zwittergent 3-14 solution, was poorly soluble in biochemical buffers, such as 50 mM Tris-HCl (pH 7.0), after removal of the detergent. However, detergent-extracted OmcZS was readily soluble in pure water. Almost all of the protein stacked on top of the gel when the sample was loaded without heating (see Fig. S1A in the supplemental material). A clear OmcZS band appeared when the sample was heated at 100°C for 5 min with SDS sample buffer (Fig. (Fig.3B).3B). Moreover, the detergent-extracted OmcZS was also retained on a 300-kDa-cutoff filter after centrifugation (see Fig. S1B in the supplemental material). These observations indicate that OmcZS polymerizes or assembles with other cell constituents at this stage. To disassemble the OmcZS into monomers, the detergent-extracted OmcZS was heated (90°C) with 1% SDS for 5 min. After this procedure, OmcZS passed through 300-kDa-cutoff filters (see Fig. S1B in the supplemental material).
The heat-treated sample was further purified by gel filtration chromatography. The single peak for OmcZS was the only peak observed in the chromatograph (data not shown). The relative molecular mass of OmcZS was calculated as 32 kDa by gel filtration chromatography. Typical yields were approximately 2 mg of protein per liter of culture. After gel filtration chromatography, the purified OmcZS was poorly soluble in 50 mM Tris-HCl buffer but highly soluble in pure water. The purified OmcZS in water remained on the 300-kDa-cutoff filters after centrifugation again (see Fig. S1B in the supplemental material), indicating that OmcZS reassembled after removal of the detergents. SDS-PAGE analysis also showed self-assembling characteristics of OmcZS. The apparently dimerized and trimerized bands of the purified OmcZS were observed when the protein sample was not heated before being loaded on the SDS-PAGE gel (see Fig. S1A in the supplemental material). SDS-PAGE and Western blotting confirmed that the final product of the purification was pure and contained a single protein band with a mass of ca. 30 kDa (Fig. 3B and C).
ESI-MS analysis indicated an average molecular mass of 32,582 Da for the purified OmcZS when measured in 50% methanol and 3% acetic acid and 32,578 Da when measured under milder conditions in 10 mM ammonium bicarbonate buffer. Edman sequencing analysis of the purified OmcZS revealed that the N terminus sequence is AVPPP, which is located 25 to 29 residues from the N terminus (Fig. (Fig.1).1). This indicates that the N terminus signal peptide of OmcZ is cleaved between amino acids 24 and 25, which corresponds with the cleavage site predicted by SignalP.
Digestion of the OmcZS with carboxypeptidase Y yielded 3 amino acids: Gly (44% of the total amino acids detected [mol/mol]), Phe (33%), and Asn (19%). Among all possible sequence combinations of these 3 amino acids, GNF (136 to 138 from the N terminus) and FGN (280 to 282) were found in the OmcZ amino acid sequence. A protein with the latter C terminus gave the expected molecular mass of ca. 30 kDa.
The estimated molecular mass of OmcZS is consistent with the predicted amino acid content (Fig. (Fig.1)1) and eight heme groups with a molecular mass of 616 Da. The predicted peptide sequence of OmcZS (Fig. (Fig.1)1) contains all the seven predicted OmcZ heme-binding sites represented by the CXXCH motif, as well as the unusual (CX14CH) heme-binding site. Pyridine hemochrome analysis revealed that OmcZS contains 7.7 hemes per molecule, indicating that hemes bind to all the possible heme-binding sites.
In general, c-type cytochrome biogenesis in bacteria has been thought to be strictly dependent on the presence of two cysteine residues arranged in a CX2-4CH/K motif (16). However, the octaheme c-type cytochrome MccA in Wolinella suggincogenes contains a covalent heme attached to an unusual heme-biding CX15CH motif, which requires a specialized cytochrome c heme lyase, CcsA1, for heme attachment (16, 23). Six CcsA1-type heme lyase homologues are in the G. sulfurreducens genome (17), one or more of which could account for the heme incorporation into the unusual heme-binding site in OmcZ.
OmcZS was exposed to heat (90°C) and harsh detergents, such as SDS and Zwittergent 3-14, in the purification procedure. DSC analysis of the thermal stability of OmcZS indicated that the denaturation temperature was 94.2°C at pH 7.0.
CD spectra of purified OmcZS demonstrated the presence of significant secondary structures. The far-UV CD spectrum indicated that OmcZS is comprised of 13% α-helix, 18% antiparallel β-sheet, 5% parallel β-sheet, and 28% β-turn.
The UV/visible redox spectrum was characteristic of c-type cytochromes (Fig. (Fig.4).4). The maxima of the spectrum of the oxidized OmcZS were at 408 nm (910,200 M−1 cm−1; γ Soret band) and 530 nm (103,130 M−1 cm−1) (Fig. (Fig.4).4). After reduction with DTT for 1 h, OmcZS had absorption maxima at 419 nm (1,084,400 M−1 cm−1; shifted γ Soret band), 523 nm (126,000 M−1 cm−1; β Soret band), and 552 nm (171,000 M−1 cm−1; α Soret band). These spectra are typical for c-type cytochromes with six coordinated low-spin hemes (2).
The redox behavior of OmcZs was investigated with redox titrations, followed by visible spectroscopy (Fig. (Fig.5).5). Both the oxidative and reductive curves spanned a large range of reduction potentials (−420 to −60 mV). The curves exhibited some hysteresis, which indicates that under these experimental conditions the protein can cycle between the fully reduced and fully oxidized states in a nonreversible way. This suggests that slowly relaxing modifications in the protein structure are associated with the redox transition. The Eapp (i.e., the point at which the oxidized and reduced fractions are equal) values for the reductive and oxidative curve were −206 and −234 mV (versus the standard hydrogen electrode [SHE]), respectively. The shapes of the experimental curves deviate significantly from one that considers identical reduction potential values for the 8 heme groups (dashed line in Fig. Fig.5).5). This observation points to nonequivalence of the redox centers, which is expected for a multiredox center protein with eight heme groups (47, 57), as is the case for OmcZS.
The large potential range of OmcZ (−420 to −60 mV versus SHE) can most probably be attributed to the wide range of the redox potentials for the 8 hemes in the molecule. This is similar to the decaheme c-type outer-surface cytochrome MtrC of S. oneidensis, which has a potential range of −500 to +100 mV (15). The potential range of OmcZ covers the lowest anode potential observed in microbial fuel cells of G. sulfurreducens (−420 mV versus Ag/AgCl, approximately equal to −220 mV versus SHE) (5), suggesting that OmcZ has a low enough potential to directly transfer electrons to the anode.
Spectrophotometric analysis revealed that OmcZS is rapidly (less than 5 min) oxidized with known electron acceptors for G. sulfurreducens, such as Fe(III) citrate, Mn(IV) oxide, U(VI), Cr(VI), Au(III), and the humic acid analogue AQDS (Fig. (Fig.6).6). OmcZS was only partially reoxidized when it was incubated with Fe(III) oxide (Fig. (Fig.6H).6H). Even after 90 min of incubation, the 419-nm γ-band did not completely shift to 408 nm (Fig. (Fig.6H,6H, inset), indicating that OmcZS has little activity toward Fe (III) oxide. After the addition of Fe(III) oxide, all the peaks decreased with time, probably because OmcZS attached to the insoluble Fe(III) oxide.
Previous studies have suggested that c-type cytochromes are involved in the reduction of a variety of metals and the humic acid analogue AQDS by Geobacter spp. (25, 30-32, 37, 49, 58). The ability of OmcZ to transfer electrons to metals and AQDS, coupled with its extracellular location, suggests that OmcZ could play a role in electron transfer to these extracellular electron acceptors. The finding that OmcZ did not readily transfer electrons to insoluble Fe(III) oxide is consistent with the fact that deleting the gene for OmcZ did not have any impact on the capacity for Fe(III) oxide reduction by G. sulfurreducens (39). OmcZ could reduce Mn(IV) oxide quickly, but not Fe(III) oxide. This might be explained by the fact that the midpoint potential of Mn(IV) oxide is much higher (from 500 to 600 mV at 25°C, pH 7 ) than that of OmcZ (−220 mV), whereas the midpoint potential of Fe(III) oxide (−300 to 0 mV ) is comparable to that of OmcZ.
Several of the characteristics of OmcZ reported here are consistent with the requirement for OmcZ for optimal current production by G. sulfurreducens (39) and electrochemical results (48) that suggest that OmcZ aids in electron conduction from G. sulfurreducens biofilms to the anodes of microbial fuel cells. The results demonstrate that OmcZ is primarily localized in the extracellular matrix, as would be expected for a protein contributing to electron conduction in biofilms. Precedents for extracellular localization of c-type cytochromes include the presence of the cytochromes MtrC and OmcA in the polymeric substance surrounding cells of S. oneidensis (36) and recovery of a c-type cytochrome in the extracellular matrix purified from Myxococcus xanthus (10). The tendency for OmcZ to self-assemble, coupled with its poor solubility in buffers, suggests that OmcZ would be retained within the biofilm matrix rather than lost to the external medium. Therefore, the energy commitment to produce OmcZ to promote electron transfer through the biofilm is not likely to be dissipated in loss of OmcZ to the external medium. The multiple redox potentials of the hemes in OmcZ seem well suited to promoting electron transfer to anode biofilms, which may function at different potentials, depending on the resistance to electron flow and the rates of metabolism in microbial fuel cells. Studies are now under way to purify OmcS, another abundant outer-surface c-type cytochrome of G. sulfurreducens that is required for Fe(III) oxide reduction but not for high-density current production, in order to compare biochemical features that differentiate the functions of outer-surface cytochromes.
This work was supported by the Office of Science (BER), U.S. Department of Energy, Cooperative Agreement no. DE-FC02-02ER63446, by Office of Naval Research award no. N00014-10-1-0084, and by Fundação para a Ciência e a Tecnologia (Portugal) research grant PTDC/BIA-PRO/74498/2006. K. Inoue is a recipient of a Japan Society for the Promotion of Science Postdoctoral Fellowship for Research Abroad. L. Morgado is supported by Fundação para a Ciência e a Tecnologia (Portugal) doctoral grant SFRH/BD/37415/2007.
We thank J. Fernandez (Rockefeller University Protein/DNA Technology Center) for the Edman sequencing analysis; S. J. Eyles (University of Massachusetts Core Mass Spectrometry Facility) for ESI-MS analysis; J. Leszyk (University of Massachusetts Medical School) for LC-MS-MS analysis; J. Nakamura, M. Uozaki, and Y. Sakaguchi (DKSH, Japan) and E. Katoh (National Institute of Agrobiological Sciences, Japan) for the DSC measurements; and T. Arakawa (Alliance Protein Laboratories) for far-UV CD analysis. We are grateful for technical support from B. Blunt and J. Ward.
Published ahead of print on 16 April 2010.
†Supplemental material for this article may be found at http://aem.asm.org/.