Characterization of Axial and Proximal Histidine Mutations of the Decaheme Cytochrome MtrA from Shewanella sp. Strain ANA-3 and Implications for the Electron Transport System

Abstract

Extracellular respiration of solid-phase electron acceptors in some microorganisms requires a complex chain of multiheme c-type cytochromes that span the inner and outer membranes. In Shewanella species, MtrA, an ∼35-kDa periplasmic decaheme c-type cytochrome, is an essential component for extracellular respiration of iron(III). The exact mechanism of electron transport has not yet been resolved, but the arrangement of the polypeptide chain may have a strong influence on the capability of the MtrA cytochrome to transport electrons. The iron hemes of MtrA are bound to its polypeptide chain via proximal (CXXCH) and distal histidine residues. In this study, we show the effects of mutating histidine residues of MtrA to arginine on protein expression and extracellular respiration using Shewanella sp. strain ANA-3 as a model organism. Individual mutations to six out of nine proximal histidines in CXXCH of MtrA led to decreased protein expression. However, distal histidine mutations resulted in various degrees of protein expression. In addition, the effects of histidine mutations on extracellular respiration were tested using ferrihydrite and current production in microbial fuel cells. These results show that proximal histidine mutants were unable to reduce ferrihydrite. Mutations to the distal histidine residues resulted in various degrees of ferrihydrite reduction. These findings indicate that mutations to the proximal histidine residues affect MtrA expression, leading to loss of extracellular respiration ability. In contrast, mutations to the distal histidine residues are less detrimental to protein expression, and extracellular respiration can proceed.

INTRODUCTION

c-type cytochromes are proteins that occur in almost all organisms, are essential for cellular respiration (1), and were most likely present in the earliest life forms (13, 17). In contrast to animals and plants, which possess only single-heme c-type cytochromes (a heme group consists of a porphyrin ring with a central iron atom), certain microorganisms possess c-type cytochromes with multiple hemes (13, 17). These proteins are known to play a fundamental role as electron transport proteins, allowing the organism to respire a variety of terminal electron acceptors (16, 17). However, because of the complex molecular structures of multiheme cytochromes (15) and difficulty in obtaining sufficient amounts of protein for detailed biochemical characterization, their electron transfer function, including how electrons are passed between or within multiheme proteins, remains poorly understood.

Several methods have been developed in Escherichia coli and Shewanella oneidensis strain MR-1 to express multiheme cytochromes, such as the decaheme-containing protein MtrA (18, 28). In previous studies, MtrA was expressed in the periplasm as a soluble protein and in association with membrane proteins MtrB and MtrC (11, 22, 26, 28). Previous studies have shown that these cytochromes are part of an electron transport chain that allows the organism to extracellularly respire a number of electron acceptors, including poorly soluble electron acceptors such as iron(III). A recent X-ray-scattering study (10) revealed that MtrA is a monomeric rod-shaped protein that appears to be soluble and membrane associated, supporting its role as a protein shuttle between cytoplasmic and outer membrane cytochromes (10). Nevertheless, the exact mechanism by which MtrA transfers electrons from the cytoplasm to the outer membrane cytochromes and how electrons are transported within MtrA remain poorly understood.

Currently, the only crystal structure for an outer membrane decaheme cytochrome in Shewanella is that of the MtrF decaheme cytochrome from MR-1 (6). The hemes of MtrF are coordinated with two histidine residues (bis-histidine coordination), and these residues facilitate electron transport by orienting the hemes in proximity to other heme groups (6). Although we do not have a crystal structure for MtrA, we hypothesize that certain histidine residues are necessary for electron transport to occur within the hemes of MtrA.

In an effort to understand how electrons are transported within multiheme cytochromes, the goal of our study was to determine what physiological effects mutating the histidine residues of MtrA would have on their capability to transfer electrons. For this purpose, we performed site-directed mutagenesis to generate histidine mutants of MtrA in Shewanella sp. strain ANA-3. This organism is known to use MtrA in the reduction of ferric iron as a terminal electron acceptor (18). Growing the organism using ferrihydrite, Fe(III) citrate, or an electrode in microbial fuel cells (MFCs) as the only electron acceptor allowed electron transport to be easily screened by the amount of iron turned over or current generated. For the site-directed mutagenesis, we mutated the His of the Cys-Xaa-Xaa-Cys-His (CXXCH) (Xaa = any amino acid) motif of each heme of MtrA. A feature used to identify c-type cytochromes, the CXXCH sequence motif refers to the covalent coupling of a heme porphyrin ring to two cysteine residues and a noncovalently bound histidine side chain. Since the heme iron can be coordinated with a distal amino acid found elsewhere in the protein backbone, often a histidine or methionine (2, 17), we also mutated His residues of MtrA that were not part of the CXXCH motif but are conserved among Shewanella strains. Understanding which residues of MtrA are important for its ability to function in iron reduction metabolism is one step forward in understanding the nature of bis-histidine coordination in multiheme cytochromes and provides a foundation for future biochemical studies.

MATERIALS AND METHODS

Strains and plasmids.

To determine which amino acid residues of MtrA are important for its ability to function in iron reduction metabolism, the first step of our study was to generate a ΔmtrA deletion strain deficient in iron reduction and a complementation plasmid that could restore its iron reduction ability. Overexpression of mtrA was achieved by cloning the gene into pBBR1MCS2 (12), a medium-copy-number vector in which the expression of mtrA is driven by the constitutive lac promoter and expressed in the ΔmtrA deletion strain. The complementation plasmid pBBR1MCS2mtrA served as a template for site-directed mutagenesis steps of the various histidines of MtrA. The strains and plasmids used in this study are listed in Table 1. The forward and reverse primers listed in Table S1 in the supplemental material were used to generate an mtrA PCR product with six histidine codons inserted before the stop codon. The PCR product was then cloned into the XbaI and KpnI sites of pBBR1MCS2 to generate pBBR1MCS2mtrA. The M13 forward and reverse primers listed in Table S1 were used to amplify the mtrA gene using pBBR1MCS2mtrA as the template.

Table 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Genotype or markers, characteristics, and usesa	Source or reference
E. coli
UQ950	E. coli DH5α λ(pir) host for cloning; F-Δ(argF-lac)169ϕ80dlacZ58(ΔM15)glnV44(AS)rfbD1gyrA96(NalR) recA1 endA1 spoT1 thi-1 hsdR17 deoR λpir+	23
WM3064	Donor strain for conjugation; thrB1004 pro thi rpsL hsdS lacZΔM15 RP4-1360 Δ(araBAD)567 ΔdapA1341::[erm pir(wt)]	23
Shewanella
Shewanella sp. ANA-3	Isolated from an As-treated wooden pier piling in a brackish estuary (Eel Pond, Woods Hole, MA)	23
ΔmtrA mutant	Shewanella sp. ANA-3; ΔmtrA(Shewana3_2677)	This study
ΔmtrAp	Shewanella sp. ANA-3; ΔmtrA(Shewana3_2677) carrying the pBBR1MCS2 plasmid	This study
pmtrA	Shewanella sp. ANA-3; ΔmtrA(Shewana3_2677) carrying the pBBR1MCS2mtrA plasmid	This study
Plasmids
pSMV10	9.1-kb mobilizable suicide vector; oriR6K mobR4 sacB Kmr Gmr	23
pBBR1MCS2	5.1-kb broad-host-range, low-copy-number plasmid; Kmr lacZ	12
pBBR1MCS2mtrA	ANA-3 mtrA PCR fragment, including 21 bp upstream of the start site and 6-histidine tags attached to its 3′ end followed by a stop codon, cloned into the XbaI/KpnI sites of pBBR1MCSC2	This study
pDH1	pBBR1MCS2mtrA with mtrA H81R mutation	This study
pDH2	pBBR1MCS2mtrA with mtrA H106R mutation	This study
pDH4	pBBR1MCS2mtrA with mtrA H163R mutation	This study
pDH5	pBBR1MCS2mtrA with mtrA H196R mutation	This study
pDH6	pBBR1MCS2mtrA with mtrA H212R mutation	This study
pDH7	pBBR1MCS2mtrA with mtrA H244R mutation	This study
pDH8	pBBR1MCS2mtrA with mtrA H259R mutation	This study
pDH9	pBBR1MCS2mtrA with mtrA H283R mutation	This study
pDH10	pBBR1MCS2mtrA with mtrA H317R mutation	This study
pPH1	pBBR1MCS2mtrA with mtrA H67R mutation	This study
pPH2	pBBR1MCS2mtrA with mtrA H100R mutation	This study
pPH3	pBBR1MCS2mtrA with mtrA H136R mutation	This study
pPH4	pBBR1MCS2mtrA with mtrA H160R mutation	This study
pPH6	pBBR1MCS2mtrA with mtrA H209R mutation	This study
pPH7	pBBR1MCS2mtrA with mtrA H233R mutation	This study
pPH8	pBBR1MCS2mtrA with mtrA H255R mutation	This study
pPH9	pBBR1MCS2mtrA with mtrA H278R mutation	This study
pH 10	pBBR1MCS2mtrA with mtrA H313R mutation	This study

^aDH, distal histidine; PH, proximal histidine.

Mutagenesis of mtrA.

In-frame nonpolar deletions of mtrA(Shewana3_2677) were generated using previously published methods (23). The primers used to generate the ΔmtrA mutant are listed in Table S1 in the supplemental material as X-mtrA-A, X-mtrA-B, X-mtrA-C, and X-mtrA-D.

Site-directed mutagenesis.

Mutations to proximal (PH) and distal (DH) histidines of ANA-3 mtrA were generated using the QuikChange Primer Design Program (Agilent Technologies). The primers listed in Table S1 in the supplemental material were designed with a histidine-to-arginine substitution. The mtrA mutagenesis reaction mixture consisted of a 25-μl mixture containing 2.5 μl 10× Pfu-Turbo Hotstart Buffer (Agilent Technologies), 0.2 mM deoxynucleoside triphosphate mixture, 125 ng primer, 25 ng pmtrA plasmid, 1 μl Pfu-Turbo Hotstart DNA polymerase (Agilent Technologies), and nuclease-free water. Samples were incubated with the following thermocycle profile: 95°C for 30 s and 18 cycles of 95°C for 30 s, 55°C for 1 min, and 68°C for 11 min. The reaction mixtures were cooled to room temperature, and 1 μl DpnI (10 units) was added to each reaction mixture and incubated for 1 h at 37°C. Each reaction mixture was transformed into E. coli strain DH5α. Plasmids were extracted and sequenced to confirm the correct mutation. The resulting histidine-to-arginine-mutated pBBR1MCS2mtrA vectors were then transformed into ΔmtrA by conjugation, as previously described (23). The strains were designated pmtrA for ANA-3 harboring pBBR1MCS2mtrA, PH1 to PH10 for strains harboring proximal histidine mutations, and DH1 to DH10 for strains harboring distal histidine mutations.

Microbial fuel cell assembly.

Laboratory-built dual-chamber MFC reactors were used to characterize the bioelectricity generated from the mutants. An individual MFC (see Fig. 7A and B) consists of an anode chamber and a cathode chamber separated by a cation-exchange membrane (CMI7000; Membranes International, Inc., Ringwood, NJ). Each chamber has a working volume of 25 ml. The electrodes were made using graphite felt or carbon cloth (projected area, 4 cm²) connected to titanium wires (0.3-mm diameter). Before strain inoculation, the MFC assembly was autoclaved at 120°C for 30 min. The catholyte (50 mM ferricyanide in 100 mM KH₂PO₄/K₂HPO₄, pH 7.4, buffer solution) and anolyte (cell culture) were transferred into the cathode and anode chambers, respectively, under sterile conditions. ANA-3 strains were cultured aerobically at 30°C for 12 h on tryptic soy broth (TSB) medium or 24 h on a basal salts medium (TME) (24) with shaking at 200 rpm. Before they were further transferred to the anode chambers of individual MFCs, the cultures were adjusted to an optical density at 600 nm (OD₆₀₀) of 1.4 (with TSB) or 0.6 (with TME). Anthraquinone-2,6-disulfonate (AQDS) (100 mM) was added to the anolyte as an artificial electron mediator.

MFC data collection and calculations.

After strain inoculation and filling with ferricyanide catholyte, the electrodes of each individual MFC were connected by a 2-kΩ external resistor to complete the circuit. The potential drop (V) across the external resistor (R) was sampled every other minute with a high-impedance digital multimeter (model 2700; Keithley Instruments), and the current in the circuit (I) was calculated via Ohm's law: I = V/R. The output power (P) was calculated as follows: P = V × I. Polarization and power curves were measured by systematically changing the external resistors from 1 MΩ to 200 Ω and recording each V at the equilibrium. Current and power densities were calculated based on the projected anode area (19, 21). Bacterial strains were tested in parallel and in triplicate tests.

Expression of MtrA in Shewanella ANA-3.

To check for expression of MtrA in the soluble fraction and whole cells of the ΔmtrA strain expressing the mutated plasmids, these cells were first grown aerobically overnight in 5 ml of Luria-Bertani (LB) medium at 30°C. Kanamycin (Km) was included in the medium for strains carrying the pBBR1MCS2 plasmid at 50-μg/μl final concentration. For the soluble fraction, the next day, 40 μl of the overnight culture was used to inoculate 80 ml of LB medium containing 50 mM fumarate and 20 mM lactate, and the culture was grown anaerobically at 30°C overnight. Cells were harvested by centrifugation at 2,988 × g for 10 min at 4°C with an SS-34 rotor (Thermo Scientific) and a Sorvall RC-5 centrifuge. The next day, the pellet was resuspended in 30 ml of 0.1 M HEPES buffer at pH 7.6 containing 0.2 mM 4-(2-aminoethyl) benzenesulfonyl fluoride (AEBSF) protease inhibitor. The samples were sonicated to lyse cells at 60% amplitude in increments of 30 s with 2-min breaks on ice at 4°C. Following lysis, the cells were centrifuged for 20 min at 12,000 × g to remove cell debris. The supernatant was next centrifuged at 206,000 × g for 1 h at 4°C with a T-865 rotor (Thermo Scientific) and a Sorvall Discovery ultracentrifuge. The soluble fraction was collected and concentrated using Vivaspin 20 (GE Healthcare) centrifuge devices according to the manufacturer's instructions. Samples were then stored at −70°C until they were analyzed by Western blotting and heme staining. For the whole-cell fraction, 5 ml of overnight culture grown on LB medium with Km was centrifuged at 12,000 × g, and the cells were resuspended in 1 ml of 0.1% Triton X-100 (Sigma-Aldrich). The cells were sonicated and centrifuged as described above, and the whole-cell lysate was analyzed by heme staining.

Protein quantification.

The Bradford assay was used to determine total protein concentrations.

SDS-PAGE and heme staining.

NuPAGE Novex Bis Tris 12% gels were used to visualize proteins, using a 1× MOPS (morpholinepropanesulfonic acid) SDS running buffer (50 mM MOPS, 50 mM Tris base, 0.1% SDS, 1 mM EDTA, pH 7.7). The gels were run at 75 V for 2 h. Heme staining was done according to previously published methods (9).

Western blots.

For Western blotting, gels were first incubated in 1× transfer buffer (9.1 g Tris base, 43.3 g glycine, 450 ml methanol, 2 liters of water) for 20 min and then transferred to Amersham Hybond-P membranes that had been allowed to soak in methanol for 10 s. Blot transfer was carried out overnight at 4°C at 30 V. The next day, the blots were incubated in Tris-buffered saline-0.1% Tween (TBS-Tween) and 5% nonfat dry milk and allowed to block overnight at 4°C. Western blot detection was performed by first incubating gels with anti-His (C-terminal) alkaline phosphatase-conjugated antibody (Invitrogen). The antibody was diluted 1:1,000 in TBS-0.1% Tween–1% nonfat dry milk, and the overnight blot was incubated with the antibody for 2 h at room temperature. Afterward, the gels were washed 3 times with TBS-0.1% Tween for 5 min, and then BCIP/NBT Blue Liquid Substrate Reagent (Sigma-Aldrich) was added to the blots. The blots were allowed to react with the 5-bromo-4-chloro-3-indolylphosphate/nitroblue tetrazolium (BCIP/NBT) reagent for up to 1 h before purple bands appeared and the reaction was stopped by adding deionized (DI) water to the blots.

Growth experiments.

E. coli was cultured in LB medium. ANA-3 was cultured in TME medium (24). Lactate was included in the medium as the electron donor at 20 mM final concentration. Anaerobic media and anaerobic culturing were prepared as previously described (24). The following electron acceptors were used at 10 mM unless otherwise indicated: fumarate, nitrate, arsenate, Fe(III) citrate, and ferrihydrite.

Iron measurements.

Ferrihydrite iron reduction was measured according to previously published methods (20) with the following modification. Cells were grown overnight in LB medium (with or without Km [50 μg/μl]) at 30°C. The next day, the cells were washed twice in 1× PBS and resuspended in TME to 10⁸ cells ml⁻¹.

RESULTS

Reduction of terminal electron acceptors.

Mutant and complementation strains reduced nitrate, fumarate, and As(V) to wild-type levels (Table 2). However, the mutants were deficient for Fe(III) citrate reduction, and the complementation strain pmtrA restored their iron reduction ability (Fig. 1).

Table 2.

Growth rates of ΔmtrA, pmtrA complementation, and control strains for known terminal electron acceptors

Strain	Growth ratea
	Nitrate (h−1)	Fumarate (h−1)	Arsenate (h−1)	Fe(III) citrate (μM min−1 ml−1)
ANA-3	0.7 ± 0.08	0.7 ± 0.09	1.0 ± 0.03	2.9 ± 0.15
ΔmtrA mutant	0.7 ± 0.06	0.9 ± 0.08	0.9 ± 0.10	ND
pmtrA	0.7 ± 0.05	0.7 ± 0.08	0.9 ± 0.08	2.8 ± 0.13
ΔmtrAp	0.7 ± 0.05	0.8 ± 0.05	0.9 ± 0.08	NA
ANA-3p	0.7 ± 0.05	0.8 ± 0.09	1.0 ± 0.05	NA

^aThe data for nitrate, fumarate, and arsenate represent averages and standard deviations of quadruplicate cultures. The data for Fe(III) citrate represent averages and standard deviations of triplicate cultures. ΔmtrAp and ANA-3p are mutant and wild-type strains with the empty vector. ND, levels were near the detection limit; NA, no data available.

Fig 1.

Measured concentrations of total Fe(II) in incubations with wild-type ANA-3 and pmtrA and ΔmtrA mutant strains at 10⁶ cells ml⁻¹. The cells were incubated with Fe(III) citrate. The data represent averages and standard deviations of triplicate cultures. ●, wild type; ■, pmtrA; ▼, ΔmtrA; ×, no-cell control.

Site-directed mutagenesis.

Amino acid alignment of MtrA from Shewanella sp. strain ANA-3 with other Shewanella MtrA sequences indicated that each of the 10 CXXCH heme motif sequences is strictly conserved among various Shewanella strains (Fig. 2). Sequencing results showed that 9 out of 10 histidines of MtrA from ANA-3 were successfully mutated by replacing the proximal His of the CXXCH motif with Arg or by replacing the distal His with Arg (Table 1; see Table S2 in the supplemental material). We were unsuccessful in generating mutants for histidine residues H202, H167, and H168 (Fig. 2) using the PCR mutagenesis method. Mutants were designated proximal histidine mutants (PH) or distal histidine mutants (DH).

Fig 2.

Amino acid alignment of MtrA sequences from various Shewanella species: MR-4 (gi:113970846), MR-7 (gi:114048071), ANA-3 (gi:117921118), HN41 (gi:336311068), Shewanella baltica OS223 (gi:217973926), MR-1 (gi:24373343), Shewanella putrefaciens 200 (gi:319425877), W13-81-1 (gi:120599427), Shewanella putrefaciens CN-32 (gi:146292577), Shewanella halifaxensis HAW-EB4 (gi:167624703), Shewanella pealeana (gi:157963670), Shewanella piezotolerans WP3 (gi:212636045), Shewanella sediminis HAW-EB3 (gi:157374663), Shewanella loihica PV-4 (gi:127513454), and Shewanella frigidimarina NCIMB 400 (gi:28375578). Histidines conserved among species are marked with a red dot above the column. CXXCH motifs are underlined and marked as CXXCH motifs 1 to 10. The degree of conservation is noted as a percentage below the alignment. A higher bar represents a higher degree of conservation, and a lower bar represents a lower degree of conservation. Amino acid numbers from the N terminus to C terminus are shown. The alignment was made with CLC Sequence Viewer 6.5.3.

MtrA expression in soluble fractions of histidine mutant cells.

To determine if MtrA was expressed in each mutant, we attached a His₆ tag at the C terminus of MtrA to facilitate protein detection (see Table S1 in the supplemental material). Previous studies have successfully expressed correctly folded and functional multiheme cytochromes MtrC and MtrA carrying a C-terminal His₆ tag in E. coli and Shewanella MR-1 (14, 27, 28). MtrA could be detected in soluble preparations from cells grown anaerobically with fumarate and lactate. Analysis by SDS-PAGE and immunoblotting using anti-His₆ antibody, which targets the C-terminal His₆ tag of each mutant, confirmed the presence of an ∼35-kDa band in the pmtrA control and distal histidine mutants DH6, DH7, DH9, and DH10 and PH9 (Fig. 3). Low expression levels were observed in the DH5, DH8, PH6, PH8, and PH10 preparations, and expression was not visible in the DH1, DH2, DH4, PH1, PH2, PH3, PH4, and PH7 preparations (Fig. 3).

Fig 3.

Western blots of ANA-3 (WT) and ΔmtrA, ΔmtrAp, pmtrA, DH, and PH mutant soluble-fraction samples. Cells were grown anaerobically with fumarate and Km. All samples were loaded at 8 μg/well.

In addition to studying the expression of MtrA in our samples using Western blotting, we also used a chemiluminescence technique (9) as a way to visualize all heme-containing proteins in our samples, including those expressed at low levels. This method is capable of detecting cytochromes in the subpicomole range. Multiple heme-containing proteins, including one present in the same size range as MtrA, were present in the majority of the distal histidine samples, including DH2 (Fig. 4A), and the PH6 and PH8 to PH10 samples (Fig. 4B and C). This ∼35-kDa heme band was missing from ANA-3 (wild type [WT]), ΔmtrA, ΔmtrAp, and the remaining proximal histidine mutant samples (Fig. 4C). These results imply that the DH1 to DH4 mutants that appear to be absent in the Western blot may in fact be expressed at low levels.

Fig 4.

Heme stains of ANA-3 (WT) and ΔmtrA, ΔmtrAp, pmtrA, DH, and PH mutant soluble-fraction samples. (A) DH mutants. (B) PH mutants. (C) Control samples and PH3. Cells were grown anaerobically with fumarate. All samples were loaded at 8 μg/well.

We next assessed whether MtrA in the distal histidine mutant samples was also associated with whole-cell lysates. When ANA-3 whole-cell lysates of distal histidine mutants were stained for heme-containing proteins, we saw expression of proteins of MtrA size in pmtrA, DH5, DH6, DH7, DH8, and DH9 samples (Fig. 5). However, these proteins were not detected in DH1 and DH4 preparations (Fig. 5). Interestingly, the level of MtrA in the whole-cell lysates for DH10 was lower than in the soluble fraction.

Fig 5.

Heme stains of pmtrA and DH whole-cell lysate samples. Cells were grown aerobically with Luria-Bertani medium and kanamycin overnight. All samples were loaded at 30 μg/well.

Ferrihydrite reduction by histidine mutants.

At the initial time point, none of the mutants or controls demonstrated iron reduction (data not shown). Twenty-five hours postincubation, ferrihydrite reduction was observed in all distal histidine mutant cultures, excluding the DH8 culture (Fig. 6). In particular, DH1 reduced ferrihydrite comparably to pmtrA. Mutant strains showing high ferrihydrite reduction compared to pmtrA included DH5 (∼70%) and DH4 and DH6 (∼50%). The strains with the lowest ferrihydrite reduction compared to pmtrA were DH9 (∼40%) and DH2 and DH7 (∼20%). No ferrihydrite reduction was observed in all proximal mutant strain incubations and the ΔmtrA and ΔmtrAp control strains (Fig. 6). These results demonstrate that, in general, disrupting particular histidine residues of MtrA impaired its ability to function in the iron reduction pathway.

Fig 6.

Measured concentrations of total Fe(II) in incubations with wild-type ANA-3 and pmtrA and ΔmtrA mutant strains at high cell densities (10⁸ cells ml⁻¹) were incubated with ferrihydrite for up to 25 h. The data represent averages and standard deviations of triplicate cultures. pmtrA, complementation strain; ΔmtrAp, mutant strain with empty plasmid.

Measurements of extracellular reduction in microbial fuel cells.

In addition to characterizing the histidine mutants using ferrihydrite, representative pmtrA, DH1, DH4, and DH7 samples were also tested in MFCs to validate the effects of this distal histidine mutation on extracellular electron transfer. These laboratory-built MFCs were configured as dual-chamber electrochemical cells (Fig. 7A and B).

Fig 7.

(A) Cartoon representation of the MFC device; R, resistor; PEM, proton exchange membrane. (B) MFC device; R, resistor. (C) Current generated by pmtrA and DH7 mutants after the start of the experiment with TSB medium.

In the MFC experiments, all the tested strains generated stable currents until the substrates were depleted (data not shown). Notably, the four different strains exhibited distinct current generation of 300, 130, 70, and 40 μA (current densities of 750, 325, 175, and 100 mA/m²) for pmtrA, DH1, DH4, and DH7, respectively, each producing about 50% of the current produced by the previous strain (Fig. 7C). These results are consistent with the suspension ferrihydrite results, where DH1, DH4, and DH7 achieved ∼100, 49, and 15% iron reduction compared to the pmtrA strain. Furthermore, analysis of the polarization yielded increasing internal resistance of pmtrA (1.0 kΩ), DH1 (5.7 kΩ), DH4 (9.4 kΩ), and DH7 (25 kΩ), measured by the changes in slopes. Additionally, power curves confirmed the observations from the polarization curves, in that pmtrA, DH1, DH4, and DH7 produced maximal power densities of 210, 88, 70, and 45 mW/m², respectively (Fig. 8A and B). In other words, DH1, DH4, and DH7 generated electrical power at 42, 33, and 21% of that of pmtrA. Considering that the MFC devices and operation conditions were nearly identical between devices, the larger internal resistance observed in the histidine deletion mutants DH4 and DH7 can be attributed to their intrinsic inefficient extracellular electron transfer capabilities.

Fig 8.

(A) Voltage versus current density for MFC incubations with TSB medium. (B) Power density versus current density for MFC incubations in TSB medium.

DISCUSSION

In this study, we conducted site-directed mutagenesis experiments to examine the roles of proximal and distal histidine residues of the decaheme c-type cytochrome MtrA from Shewanella sp. strain ANA-3 in iron(III) and extracellular reduction. Multiple amino acid sequence alignment of Shewanella MtrA proteins identified 12 possible histidine residues that could serve as the distal Fe-heme ligand, as well as the proximal histidines in the 10 CXXCH motifs (Fig. 2). The N-terminal segment harboring the first two CXXCH motifs contains a single histidine between the two motifs (Fig. 2). Similar histidine residues were identified between CXXCH motifs 4 and 5, 6 and 7, 8 and 9, and 9 and 10, and two histidines (His-167 and His-168) could potentially serve as the distal histidine ligands for other heme groups (Fig. 2).

Comparison of iron-heme coordinating mutants expressed at wild-type levels provides evidence that disruption of specific histidine residues can affect the reduction of solid-phase electron acceptors (e.g., ferrihydrite and graphite electrodes in MFCs). In our study, mutating specific distal and proximal histidine residues (i.e., DH6, DH7, DH9, DH10, and PH9) did not affect the expression of MtrA. However, these mutants were deficient in ferrihydrite reduction abilities. Because these proteins are expressed at levels comparable to those of the wild-type pmtrA protein, these results strongly suggest that these particular mutations impede the function of MtrA after assembly, resulting in decreased levels of ferrihydrite reduction. Additionally, mutating histidines in DH2, DH4, DH5, DH8, PH1 to PH8, and PH10 strains resulted in decreased expression of MtrA and low levels of ferrihydrite reduction. We can speculate that these particular mutations could make MtrA more unstable and prone to degradation based on similar observations in other studies involving multiheme proteins (1, 5, 8). Together, these results demonstrate that mutations to specific residues of certain iron heme binding sites of MtrA can negatively affect extracellular electron transfer. However, in our investigation, we also observed a histidine mutant strain (DH1) that was expressed at low levels (Fig. 4 and 5) but was still able to reduce ferrihydrite to wild-type levels (Fig. 6). This mutant will require further study to determine why it retains iron reduction ability.

Microbial fuel cell devices are also useful in characterizing electron transfer and have been used in previous studies of the Mtr pathway in Shewanella in real time (4). In our studies, the use of an MFC allowed us to determine how an MtrA heme mutation affects the efficiency of electron transfer to a solid-phase electron acceptor, such as a graphite anode (Fig. 7A). The DH1(H81R), DH4(H163R), and DH7(H244R) mutants tested in our MFC mirrored our observations for ferrihydrite reduction. Although not tested in an MFC, we predict that strains expressing MtrA with proximal histidine mutations will exhibit little to no electrical current production, which would be consistent with our observations for ferrihydrite reduction. Overall, the results from these experiments provided another level of confirmation that mutations to specific residues of a certain iron heme binding site of MtrA can negatively effect bioelectricity generation and thus extracellular electron transfer.

We also observed variability in expression of DH1 and DH2 to DH9 proteins in the cell lysates and soluble fractions of mutants. Decreased expression levels may be due to protein instability and degradation. In studies of the tetraheme c-type cytochrome NapC in Paracoccus denitrificans and Paracoccus pantotrophus, mutations to two of its predicted distal histidine residues eliminated the expression of NapC. This observation was attributed to protein instability and possible degradation in the Paracoccus strains (5). In another study with the tetraheme c-type cytochrome c₃ from Desulfovibrio vulgaris Hildenborough, mutations to the distal histidine residues of each of its hemes also resulted in varying degrees of expression attributed to differences in protein stability (8). Although MtrA was present in the soluble fraction of DH10(H317R), it was expressed at low levels in whole-cell lysates. These results suggest that mutating H317R may inhibit the association of MtrA with membrane components, leading to high expression in the soluble fraction but lower expression in the membrane. Meanwhile, expression of DH2 and DH5 occurred at wild-type levels in the cell lysate, but not the soluble fraction. This result implies that mutating residue H196R may lead to tighter association with membrane components. In the case of DH1, this mutant was present at low levels in the soluble fraction, based on chemiluminescence results, but absent from the cell lysate. Despite its decreased level of expression in the soluble fraction and absence from the cell lysate, DH1 is still capable of reducing ferrihydrite at wild-type levels. The mechanism involved in DH1 mutant strains' ferrihydrite capability is unclear given our results and will have to be further investigated.

In the case of the proximal histidine mutations, it is likely that the majority of these mutations led to loss of heme incorporation in the MtrA protein. Previous studies involving monoheme c-type cytochromes showed that heme attachment is a posttranslational modification that occurs in the periplasm and that replacing the proximal histidine of CXXCH motifs with a methionine, lysine, or arginine prevents heme attachment to a maturing c-type cytochrome (3). Combining what we know about monoheme c-type cytochrome maturation with our results, mutating PH1(H67R), PH2(H81R), PH3(H136R), PH4(H160R), and PH7(H233R) could have interfered with heme incorporation, resulting in structural instability of MtrA and increased susceptibility to degradation. Expression of MtrA was not detected in the Western blots in the soluble fractions of strains expressing these heme variants. However, chemiluminescent stains capable of detecting subpicomole levels of protein showed that a protein similar in size to MtrA is expressed in these mutants at low levels. In contrast, MtrA protein expression was observed in both Western and chemiluminescence strains in several other mtrA alleles with mutations to the proximal histidines in CXXCH, PH6(H209R), PH8(H255R), PH9(H278R), and PH10(H313R). We can speculate that these forms of MtrA may be more stable than MtrAs with mutations to PH1, PH2, PH3, PH4, and PH7.

For proximal and distal histidine mutants in which expression was low relative to the pmtrA strain, the effects of these histidine mutations could also have been due to protein instability resulting from increased degradation of MtrB. It appears that MtrB plays a major role in the formation of a stable MtrCAB complex in the outer membrane, which is necessary for iron(III) reduction (25). Another study showed that MtrA is mislocalized in Shewanella strains lacking mtrB (7). Degradation of MtrA due to decreased protein stability has also been shown to render MtrB more susceptible to degradation by the DegP protease (25).

The iron(III) reduction model and current generation in MFC devices involves the transfer of electrons from a quinol pool in the cytoplasmic membrane via CymA to periplasmic MtrA, which then donates electrons to MtrC/OmcA for insoluble iron(III) reduction. Recent studies of the outer membrane decaheme MtrF, a homolog of MtrC, has shown the protein to be composed of four domains and the arrangement of the hemes to have a cross pattern (6). The study also shows the hemes of MtrF to be bis-His coordinated, influencing the orientation distances between heme groups. Based on the crystal structure of MtrF, there is a heme that appears to be oriented in such a way as to accept electrons from MtrD, a homolog of MtrA, and to donate electrons through another heme to solid and soluble iron(III) or electron shuttles. Alternatively, it has been proposed that two additional hemes of MtrF, which are buried in the protein structure, could also deliver electrons to soluble iron(III) or electron shuttles when other hemes are occluded by interaction with a solid surface (6). Although both are decaheme cytochromes, MtrA has a different physiological function than a terminal reductase like MtrF (or MtrC). By analogy to other electron transport chains, such as NrfB and NrfA nitrite reductase, electron flow through intermediate subunits, such as MtrA, may have a particular arrangement of hemes, influenced by bis-His coordination, that facilitate a linear movement of electrons to the outer membrane terminal reductase, MtrF (and possibly MtrC) (6). Although no crystal structure for MtrA is available, structural studies involving small-angle X-ray scattering (SAXS) and analytical ultracentrifugation (AUC) show that MtrA is elongated like a wire and could span 130 to 250 Å of periplasmic space (10). Perhaps certain hemes of MtrA may also be solvent exposed, which could permit rapid electron exchange between hemes of neighboring redox partners, such as CymA and MtrC (or MtrF). In contrast, it appears that an outer membrane decaheme cytochrome, such as MtrF, may have more flexibility in which heme could be used for solid-phase iron(III) reduction or soluble substrates, like flavins.

Based on past studies with other multiheme c-type cytochromes, mutations in iron-heme coordinating amino acids, like the distal and proximal histidines of the CXXCH motif, generally result in secondary effects, such as protein instability and degradation. Untangling the actual redox effects of iron-heme coordination mutations with secondary effects on protein structure may be difficult to achieve experimentally. Until a crystal structure for MtrA becomes available, future studies will have to address the biochemical nature of each mutation in terms of protein integrity and protein-protein interaction, as well as enzyme kinetics with soluble and insoluble substrates.