The periplasmic triheme cytochrome PpcA from Geobacter sulfurreducens is highly abundant; it is the likely reservoir of electrons to the outer surface to assist the reduction of extracellular terminal acceptors; these include insoluble metal oxides in natural habitats and electrode surfaces from which electricity can be harvested. A detailed thermodynamic characterization of PpcA showed that it has an important redox-Bohr effect that might implicate the protein in e/H+ coupling mechanisms to sustain cellular growth. This functional mechanism requires control of both the redox state and the protonation state. In the present study, isotope-labeled PpcA was produced and the three-dimensional structure of PpcA in the oxidized form was determined by NMR. This is the first solution structure of a G. sulfurreducens cytochrome in the oxidized state. The comparison of oxidized and reduced structures revealed that the heme I axial ligand geometry changed and there were other significant changes in the segments near heme I. The pH-linked conformational rearrangements observed in the vicinity of the redox-Bohr center, both in the oxidized and reduced structures, constitute the structural basis for the differences observed in the pKa values of the redox-Bohr center, providing insights into the e/H+ coupling molecular mechanisms driven by PpcA in G. sulfurreducens.

Introduction

Multiheme cytochromes are key players in extracellular electron transfer (EET) pathways of the dissimilatory metal-reducing bacterium Geobacter sulfurreducens. These cytochromes form highly complex networks that extend from the inner membrane (IM) to the outer membrane (OM) of the cell, allowing electrons to flow to the exterior, where insoluble minerals in natural habitats or anodes in microbial fuel cells (MFCs) are reduced [1]. Compared with other dissimilatory metal-reducing bacteria, G. sulfurreducens is capable of completely oxidizing a variety of organic compounds to CO2 and is among the highest density current producers in MFCs [2,3]. For these reasons G. sulfurreducens has been targeted to develop improved bioelectrochemical systems.

Gene knockout and proteomic studies identified several c-type multiheme cytochromes that participate in G. sulfurreducens EET pathways. These include the IM-associated cytochromes ImcH, CbcL and MacA, the five periplasmic triheme cytochromes from the PpcA family (named PpcA–E) and several OM cytochromes [416]. Despite the considerable gaps in the current knowledge about G. sulfurreducens EET pathways, it is established that oxidation of organic molecules releases electrons to the quinone pool at the cytoplasmic membrane via NADH dehydrogenases. From this point, the IM-associated cytochromes are proposed to accept electrons from the reduced quinone pool and transfer them to periplasmic cytochromes, which then bridge the electron transfer between the cytoplasmic and OM electron transfer components [17]. Periplasmic cytochromes are the reservoir of electrons available to the cell exterior [8,18]. Their strategic position in the periplasm allows them to regulate the redox state of this compartment and control the electron flow toward OM components [19]. Therefore, they are in the front line of potential targets to develop strains carrying mutated cytochromes rationally designed to increase the respiratory capabilities of Geobacter species.

The PpcA family of cytochromes has been studied in detail by biochemical and biophysical methods [8,2032]. The members of this family contain ∼70 amino acids and three low-spin heme groups with bis-histidinyl axial co-ordination, which are diamagnetic (S = 0) and paramagnetic (S = 1/2) in the reduced and oxidized forms, respectively [33,34]. The crystal structures of the PpcA family cytochromes have been determined in the oxidized form [20,23,24]. However, in contrast with the structures obtained for the other four members of the PpcA family (PpcB–E), PpcA crystals could only be obtained in the presence of the additive deoxycholate [23]. The crystal structure of PpcA showed that parts of its structure are altered compared with the other members due to the binding of deoxycholate to PpcA in the crystal. The solution structure of PpcA in the reduced form obtained in the absence of deoxycholate confirmed that the structure in the crystal was altered in order to accommodate the bound deoxycholate molecule [21]. The solution structure of reduced PpcA is very similar to the crystal structures obtained in the oxidized state of the other members of the family [21].

PpcA is the most abundant member of the G. sulfurreducens PpcA family [8,18] and its functional mechanism has been characterized in detail [22,25]. The heme reduction potentials of PpcA are negative, different from each other, and are modulated by heme–heme interactions and by interactions with protonated groups (redox-Bohr effect) that allow the protein to couple e/H+ transfer within the physiological pH range for G. sulfurreducens. Studies on G. sulfurreducens biofilms grown on electrodes showed that the cells have a Nernstian response around −150 mV [3537], which can be driven by PpcA, as the protein is functionally active at this redox potential [25].

Future investigation of EET G. sulfurreducens respiratory mechanisms should involve protein–protein interaction studies to survey the electron transfer between redox partners and establish foundations to engineer redox proteins for various applications. To achieve this, information on the structure and functional mechanisms of key proteins is crucial. Since the functional mechanism of PpcA is well established, the absence of a PpcA solution structure in the oxidized form prevents the identification of the precise interaction regions with its redox partner(s) and, consequently, slows the rational design of improved electron transfer pathways that can provide important advances to enhance the efficiency of MFCs and other G. sulfurreducens-based biotechnological applications. In this work, we describe the solution structure of PpcA in the oxidized form, which is the first oxidized solution structure reported for a cytochrome from G. sulfurreducens. The structural information obtained was used to explore the redox- and pH-linked conformational changes that sustain the functional mechanism of PpcA.

Materials and methods

Bacterial growth and protein purification

Natural abundance and isotopic labeled (15N and 13C,15N) PpcA samples were produced in Escherichia coli and purified as previously described [38]. Briefly, E. coli BL21 (DE3) cells containing plasmids pEC86 (encoding cytochrome c maturation gene cluster ccmABCDEFH) and pCK32 (encoding PpcA [39]) were grown in 2× YT medium to an OD600 of ∼1.5. This culture was then processed as follows: (a) to produce natural abundance PpcA, 10 µM isopropyl β-d-thiogalactoside (IPTG) was added, with overnight culture growth at 30 °C and harvested by centrifugation; (b) to produce 15N-PpcA, the cells were collected by centrifugation after reaching an OD600 of 0.6–1.0, washed twice with 250 ml of a salt solution containing 3 g/l KH2PO4, 6 g/l Na2HPO4 and 0.5 g/l NaCl, resuspended in minimal medium (in a ratio of 250 ml of minimal medium for each liter of 2× YT medium) supplied with 1 g/l 15NH4Cl as the nitrogen source (together with 1 mM of the heme precursor α-aminolevulinic acid, trace amounts of metal salts, biotin and thiamine), grown overnight at 30°C in the presence of 100 µM IPTG and harvested by centrifugation; (c) to produce 13C,15N-PpcA, as described in (b) but also supplied with 2 g/l 13C6-glucose as the carbon source. In each case, the periplasmic fractions were isolated using lysis buffer [100 mM Tris–HCl (pH 8.0), 0.5 mM EDTA and 20% sucrose] containing 0.5 mg/ml lysozyme. The proteins were then purified by ion-exchange and gel filtration chromatography. Protein purity was evaluated by SDS gel electrophoresis and Coomassie blue staining.

Sample preparation and NMR experiments

Samples for NMR experiments with 1.2 mM concentrations were prepared in 92% H2O/8% 2H2O or in 2H2O (99% atom) containing 45 mM sodium phosphate buffer of pH 5.5 (with NaCl to 100 mM final ionic strength).

The NMR experiments were carried out on a Bruker Avance 600 or 800 MHz spectrometers equipped with a z-gradient cryoprobe at 298 K. For backbone and side chain assignments, the following spectra were acquired: (1) 2D 1H,15N-HSQC, 3D 1H,15N-TOCSY (45 ms) and 3D 1H,15N-NOESY (80 ms) for 15N-PpcA sample; (2) 2D 1H,13C-HSQC, 3D CBCANH, 3D CBCA(CO)NH, 3D HNCA, 3D HN(CO)CA, 3D HCC(H)-TOCSY and 3D HC(C)H-TOCSY for 13C,15N-PpcA sample and (3) 2D 1H-COSY, 1H-TOCSY (25 and 70 ms) and 1H-NOESY (30 and 80 ms) for the natural abundance PpcA sample, all samples in 92% H2O/8% 2H2O. For the assignment of the heme signals, 2D 1H-COSY, 1H-TOCSY (25 and 70 ms) and 1H-NOESY (30 and 80 ms) experiments were acquired for the natural abundance PpcA sample prepared in 2H2O. 1H chemical shifts were calibrated using the water signal as the internal reference, and both 13C and 15N chemical shifts were calibrated through indirect referencing.

The effect of the pH on the chemical shifts was determined by analyzing a series of 2D 1H,15N-HSQC NMR spectra acquired in the pH range of 5.5–9.5. The pH of the sample was adjusted by the addition of small amounts of NaO2H or 2HCl. The weighted average chemical shift (Δδavg) of each backbone and side chain NH signal was calculated as: Δδavg = √[(Δδ2N/25 + Δδ2H)/2], where ΔδH and ΔδN are the differences in the 1H and 15N chemical shifts, respectively [40].

Structure calculations and analysis

The structure was determined by NMR according to the methodology described by Paixão et al. [41]. Nuclear overhauser effect (NOE) signal volumes were measured from 2D 1H-NOESY NMR spectra obtained with a mixing time of 80 ms, except for the cross peaks most affected by paramagnetic leakage, which were measured with a mixing time of 30 ms. The peak volumes and chemical shifts were used as input for the program PARADYANA [42], a version of DYANA [43], that optimizes the NOE calibration and fits the magnetic properties to paramagnetic shifts simultaneously with the structure calculation. The preliminary structures were then used to calculate corrections to the NOE volumes for paramagnetic relaxation and to refine the structure. In addition to the usual tools for validating solution structures, the relations between magnetic properties [44] and the Fermi contact shifts of heme substituents [45,46] on the one hand and the geometry of the axial ligands to the iron on the other were also used to validate the structure.

Results

Assignment of the heme substituents and polypeptide signals

In contrast with the diamagnetic (S = 0) reduced form of PpcA, in which the 1H chemical shifts of the heme substituents are dominated by the porphyrin ring-current shifts and, therefore, appear in well-defined regions of the NMR spectra, the unpaired electron of each heme iron in the paramagnetic (S = 1/2) oxidized state spreads the signals over the entire NMR spectral window, making their assignment more complex (for a review, see ref. [47]). To establish unequivocally the starting points to assign the heme substituents in the oxidized state, 2D 1H,13C-HSQC NMR spectra were acquired for a natural abundance sample and for a 13C/15N sample labeled exclusively in its polypeptide chain. The comparison of these spectra allowed the straightforward discrimination of 90% of the heme substituents' signals, as reported by Morgado et al. [47]. The specific assignment of the heme signals was then achieved as follows. In the first step, the heme methyls (numbered according to the IUPAC-IUB nomenclature for the heme substituents [48]) were assigned by following their signals from the fully reduced into their final position in the fully oxidized protein using 2D 1H-EXSY NMR spectra according to a procedure well established for low-spin multiheme cytochromes [49]. In the second step, the 1H and 13C signals of the remaining heme methyls, as well as those of the other nuclei directly attached to the heme porphyrin rings, were specifically assigned through the analysis of 2D 1H-NOESY, 2D 1H-TOCSY and 1H,13C-HSQC NMR spectra, following the strategy first described for tetraheme cytochromes c3 [5052]. In the end, 57 of 60 heme proton signals were assigned and their chemical shifts are indicated in Table 1.

Table 1
1H chemical shifts (ppm) of the heme substituents in the oxidized triheme PpcA

The hemes are numbered I, III and IV, a scheme that derives from the superimposition of the hemes in triheme cytochromes with those of the structurally homologous tetraheme cytochromes c3. Abbreviations: n.d., not defined.

Heme substituent Heme I Heme III Heme IV 
21 17.79 12.18 14.46 
31 0.91 −2.56 0.72 
32 1.15 −2.24 2.05 
n.d. −3.90 n.d. 
71 10.43 18.00 10.99 
81 −4.34 −0.86 −0.04 
82 −4.00 −1.01 1.63 
10 −1.54 n.d. 1.13 
121 20.65 13.18 17.38 
131 2.63 16.09 2.43 
6.72 19.94 6.45 
132 −1.48 −1.74 0.10 
−0.33 −0.68 0.44 
15 3.71 −1.67 0.75 
171 0.89 3.67 2.87 
3.01 5.55 4.18 
172 −0.87 −2.07 −0.86 
−0.43 −2.00 −0.26 
181 15.71 0.64 14.58 
20 −0.70 8.07 −1.60 
Heme substituent Heme I Heme III Heme IV 
21 17.79 12.18 14.46 
31 0.91 −2.56 0.72 
32 1.15 −2.24 2.05 
n.d. −3.90 n.d. 
71 10.43 18.00 10.99 
81 −4.34 −0.86 −0.04 
82 −4.00 −1.01 1.63 
10 −1.54 n.d. 1.13 
121 20.65 13.18 17.38 
131 2.63 16.09 2.43 
6.72 19.94 6.45 
132 −1.48 −1.74 0.10 
−0.33 −0.68 0.44 
15 3.71 −1.67 0.75 
171 0.89 3.67 2.87 
3.01 5.55 4.18 
172 −0.87 −2.07 −0.86 
−0.43 −2.00 −0.26 
181 15.71 0.64 14.58 
20 −0.70 8.07 −1.60 

After assigning the heme proton signals, the combined analysis of 2D 1H,15N-HSQC and a series of 3D NMR spectra [CBCANH, CBCA(CO)NH, HNCA and HNCACO] allowed the assignment of all backbone amide signals except for the N-terminal residue (Ala1), which is usually not observable due to fast exchange, and those of Gly36 and Gly42, most probably due to signal broadening caused by the nearby paramagnetic irons of hemes I and IV. Figure 1 shows the 2D 1H,15N-HSQC NMR spectrum of PpcA with the backbone NH signals labeled. To extend the assignment to aliphatic protons such as 1Hα, 1Hβ, 1Hγ, 1Hδ and 1Hε and respective aliphatic carbons, we analyzed the 2D 1H,13C-HSQC and a series of 3D spectra that included HCC(H)-TOCSY, HC(C)H-TOCSY, 1H,15N-TOCSY and 1H,15N-NOESY. This analysis yielded an essentially complete assignment of the side chains for most residues. The aromatic ring protons were assigned on the basis of 2D 1H-COSY, 1H-TOCSY and 1H-NOESY NMR spectra, except for the axial His ring protons, which experience large paramagnetic shifts and broadness due to their proximity to the heme irons. In total, 79% of the proton resonances were assigned after excluding exchangeable protons other than the backbone HN atoms. A summary of the sequential connectivities between 1HN, 1Hα and 1Hβ protons is shown in Supplementary Figure S1.

Region of the backbone NH signals in the 2D 1H,15N-HSQC NMR spectra of a labeled PpcA sample in the oxidized state (298 K and pH 5.5).

Figure 1.
Region of the backbone NH signals in the 2D 1H,15N-HSQC NMR spectra of a labeled PpcA sample in the oxidized state (298 K and pH 5.5).

The square indicates the position of the NH signal from residue Cys51, which has small intensity due to its close proximity to the water signal and is not visible at the spectra level represented in the figure.

Figure 1.
Region of the backbone NH signals in the 2D 1H,15N-HSQC NMR spectra of a labeled PpcA sample in the oxidized state (298 K and pH 5.5).

The square indicates the position of the NH signal from residue Cys51, which has small intensity due to its close proximity to the water signal and is not visible at the spectra level represented in the figure.

Structure determination

To avoid fractional protonation of the redox-Bohr center and the concomitant broadening of the signals, the PpcA solution structure in the oxidized state was determined at a pH of 5.5, which is one pH unit lower than the of the redox-Bohr center previously estimated by the detailed thermodynamic characterization of the protein [22]. A total of 1498 upper volume limits (upv) and 1181 lower volume limits (lov) were obtained from the 80 and 30 ms 2D 1H-NOESY NMR spectra (Table 2). Additional restraints used in the final structure calculation included a set of 96 fixed upper limit distances and 274 dipolar shifts, as described by Paixão et al. [41]. The NOE volume limits, together with dipolar shifts and the fixed upper limit distances, were used as input for the program PARADYANA [42]. During the solution structure calculation process, the preliminary structures were analyzed using the program GLOMSA, as previously described [41], and 21 stereo-specific assignments were made for diastereotopic pairs of protons or methyl groups. A summary of the restraints used is presented in Table 2 and the number of restraints per residue is shown in Figure 2. The number of constraints is fairly uniform along the protein sequence. The higher number of constraints shown by residues 30, 54 and 68 includes the many long-distance contacts of hemes I, III and IV, which are, respectively, attached to these positions.

Number of constraints per residue used for the calculation of the ferricytochrome PpcA solution structure.

Figure 2.
Number of constraints per residue used for the calculation of the ferricytochrome PpcA solution structure.

Bars are white, light gray, dark gray and black for intraresidue, sequential, medium- and long-range restraints, respectively. Residues 30, 54 and 68 also include restraints to hemes I, III and IV, respectively.

Figure 2.
Number of constraints per residue used for the calculation of the ferricytochrome PpcA solution structure.

Bars are white, light gray, dark gray and black for intraresidue, sequential, medium- and long-range restraints, respectively. Residues 30, 54 and 68 also include restraints to hemes I, III and IV, respectively.

Table 2
Summary of volume constraints, scaling factors, restraint violations and quality analysis for the final families of structures for ferricytochrome PpcA

‘lov’ and ‘upv’ stands for lower and upper NOE volume limits, respectively.

Parameter Value 
Distance constraints (upv/lov) 
 Intraresidue 687/515 
 Sequential (|i − j| = 1) 320/262 
 Medium range (2 ≤ |i − j| < 5) 228/185 
 Long range (|i − j| ≥ 5) 263/219 
Dipolar shift constraints 274 
Scaling factors 
 Proton–proton 13.6 (0.1) 
 Proton–methyl 16.1 (0.2) 
 Methyl–methyl 18.1 (0.3) 
 Backbone proton–proton 12.3 (0.1) 
 Factor relating 30 and 80 ms datasets 1.001 (0.007) 
DYANA target function (type 1) 
 Average total (Å24.12 (0.32) 
Upper distance limit violations 
 Average maximum (Å) 0.55 (0.15) 
 Number of consistent violations (>0.2 Å) 
Lower distance limit violations 
 Average maximum (Å) 0.51 (0.24) 
 Number of consistent violations (>0.2 Å) 
Van der Waals violations 
 Average maximum (Å) 0.23 (0.07) 
 Number of consistent violations (>0.2 Å) 
Dipolar shift violations 
 Average maximum (ppm) 0.43 (0.09) 
 rms violation per shift (ppm) 0.009 
 Number of consistent violations (>0.5 ppm) 
Ramachandran plot (%) 
 Core 61.8 
 Allowed 23.6 
 Generously allowed 12.7 
 Disallowed 1.8 
Precision 
 Backbone rmsd (Å) 0.69 (0.15) 
 Heavy atom rmsd (Å) 1.46 (0.12) 
Parameter Value 
Distance constraints (upv/lov) 
 Intraresidue 687/515 
 Sequential (|i − j| = 1) 320/262 
 Medium range (2 ≤ |i − j| < 5) 228/185 
 Long range (|i − j| ≥ 5) 263/219 
Dipolar shift constraints 274 
Scaling factors 
 Proton–proton 13.6 (0.1) 
 Proton–methyl 16.1 (0.2) 
 Methyl–methyl 18.1 (0.3) 
 Backbone proton–proton 12.3 (0.1) 
 Factor relating 30 and 80 ms datasets 1.001 (0.007) 
DYANA target function (type 1) 
 Average total (Å24.12 (0.32) 
Upper distance limit violations 
 Average maximum (Å) 0.55 (0.15) 
 Number of consistent violations (>0.2 Å) 
Lower distance limit violations 
 Average maximum (Å) 0.51 (0.24) 
 Number of consistent violations (>0.2 Å) 
Van der Waals violations 
 Average maximum (Å) 0.23 (0.07) 
 Number of consistent violations (>0.2 Å) 
Dipolar shift violations 
 Average maximum (ppm) 0.43 (0.09) 
 rms violation per shift (ppm) 0.009 
 Number of consistent violations (>0.5 ppm) 
Ramachandran plot (%) 
 Core 61.8 
 Allowed 23.6 
 Generously allowed 12.7 
 Disallowed 1.8 
Precision 
 Backbone rmsd (Å) 0.69 (0.15) 
 Heavy atom rmsd (Å) 1.46 (0.12) 

Quality and analysis of the structures

The final family of 20 structures with the lowest target function values (from 3.42 to 4.53 Å2, average value 4.12 Å2) was selected as being representative of the solution structure of the ferricytochrome PpcA. There were no consistent violations of NOEs or dipolar shifts. The structures superimpose with an average pairwise backbone (N-Cα-C′) rmsd (root mean square deviation) of 0.69 Å and a heavy atom rmsd of 1.46 Å with respect to the mean structure (Supplementary Figure S2). The statistics for the family of structures are presented in Table 2. A total of 71 hydrogen bonds were identified in the family of 20 structures. The Ramachandran plot shows 61.8% of the residues in the most favored regions, 23.6% in the additionally allowed, 12.7% in the generously allowed and 1.8% in the disallowed regions. Residues in the disallowed regions include Asp2 (in 6 of 20 structures), Lys33 (in 1 structure) and Glu39 (in 3 structures), which show a low number of restraints.

Solution structure of PpcA in the fully oxidized state

The oxidized structure of the triheme cytochrome PpcA in solution (Figure 3) revealed a two-strand β-sheet at the N-terminus formed by Asp3–Leu6 and Asp12–Pro16 strands, followed by four α-helices between residues His17 and Gln21, Lys43 and Ala46, Lys52 and Met58, and Cys65 and Cys68. At the C-terminal, a small β-strand is formed by residues His69 and Lys70. The three heme groups are arranged in a triangular pattern, with hemes I and IV almost parallel to each other and both nearly perpendicular to heme III.

PpcA solution structure in the oxidized form (pH 5.5).

Figure 3.
PpcA solution structure in the oxidized form (pH 5.5).

(A) Overlay of the 20 lowest energy NMR structures of PpcA. Superimposition was performed using all of the heavy atoms. The peptide chain and the hemes are colored gray and red, respectively. (B) Ribbon diagram of PpcA structure. Residues forming secondary structure elements are labeled. Figures were produced using MOLMOL [57].

Figure 3.
PpcA solution structure in the oxidized form (pH 5.5).

(A) Overlay of the 20 lowest energy NMR structures of PpcA. Superimposition was performed using all of the heavy atoms. The peptide chain and the hemes are colored gray and red, respectively. (B) Ribbon diagram of PpcA structure. Residues forming secondary structure elements are labeled. Figures were produced using MOLMOL [57].

Analysis of the magnetic susceptibility tensors and geometry of the axial heme ligands

The average magnetic susceptibility tensors and the orientation of the heme axial ligands obtained for the family of structures are given in Table 3. The magnetic susceptibility tensors are purely empirical, obtained by fitting proton dipolar shifts, and are independent of the heme geometry in the calculated structures or of the chemical shifts of the heme substituents, which are dominated by Fermi contact interaction. Therefore, analysis of the magnetic susceptibility tensors provides independent tests for the reliability of the chemical shifts used in the calculations, proton structural co-ordinates and geometry of the heme axial ligands. A further validation for the structure obtained is provided by the comparison of the observed and calculated 1H paramagnetic shifts of the heme methyl groups (Supplementary Table S1).

Table 3
Magnetic properties of the three hemes in ferricytochrome PpcA

The averaged properties of the magnetic susceptibility tensors and geometry of the axial ligands obtained for the lowest energy solution structure of PpcA in the reduced form (PDB code 2LDO [21]) are also included. Standard deviations are given in parentheses.

 Hemes 
III IV 
NMR oxidized structure (this work) 
Magnetic anisotropy 
 Δχax × 1032 (m34.35 (0.23) 3.07 (0.11) 4.43 (0.10) 
 Δχeq × 1032 (m3−0.68 (0.29) −1.69 (0.14) −0.16 (0.12) 
 Tilt angle of χzz (°) 3.0 (2.0) 5.5 (2.0) 10.0 (1.5) 
 In-plane rotation of χyy (°)1 81.1 (8.3) −31.7(2.6) 63.4 (9.8) 
Fermi contact shifts2: 1H [46]/13C [33
 Rotation of molecular orbitals (°) −52.8/−70.0 56.1/66.9 −54.8/−63.3 
 Implied angle between imidazole planes (°) 69.9/74.0 25.1/42.0 81.2/84.0 
Solution structure 
 Orientation of average His normal θ (°) −91.2 (10.5) 52.2 (5.2) −62.7 (2.6) 
 Angle between imidazole planes β (°) 56.7 (13.4) 73.9 (11.3) 58.8 (11.5) 
NMR reduced structure [21
 Orientation of average His normal θ −2.5 62.3 −58.5 
 Angle between imidazole planes β (°) 49.9 22.7 76.8 
 Hemes 
III IV 
NMR oxidized structure (this work) 
Magnetic anisotropy 
 Δχax × 1032 (m34.35 (0.23) 3.07 (0.11) 4.43 (0.10) 
 Δχeq × 1032 (m3−0.68 (0.29) −1.69 (0.14) −0.16 (0.12) 
 Tilt angle of χzz (°) 3.0 (2.0) 5.5 (2.0) 10.0 (1.5) 
 In-plane rotation of χyy (°)1 81.1 (8.3) −31.7(2.6) 63.4 (9.8) 
Fermi contact shifts2: 1H [46]/13C [33
 Rotation of molecular orbitals (°) −52.8/−70.0 56.1/66.9 −54.8/−63.3 
 Implied angle between imidazole planes (°) 69.9/74.0 25.1/42.0 81.2/84.0 
Solution structure 
 Orientation of average His normal θ (°) −91.2 (10.5) 52.2 (5.2) −62.7 (2.6) 
 Angle between imidazole planes β (°) 56.7 (13.4) 73.9 (11.3) 58.8 (11.5) 
NMR reduced structure [21
 Orientation of average His normal θ −2.5 62.3 −58.5 
 Angle between imidazole planes β (°) 49.9 22.7 76.8 
1

The in-plane rotation of χy correlates with the angle θ but with the opposite sign (see text for details).

2

Results for 1H chemical shifts were obtained using the empirical equation δi(ppm) = cosβ[38.0sin2(θi − ϕ) − 4.1cos2(θi − ϕ) − 15.9] + 13.8 described in ref. [46] and the values obtained from the analysis of 13C Fermi contact shifts are from ref. [33].

pH titration

The pH titration of PpcA was carried out by 2D 1H,15N-HSQC NMR experiments in the pH range of 5.5–9.5, and all 1H and 15N chemical shifts of the polypeptide backbone were measured. To estimate the effects of pH changes on the oxidized PpcA solution structure, the average chemical shift differences (Δδavg) of each amide were calculated as described by Garret et al. [40]. The residues whose backbone NH signals indicated larger differences (Δδavg > 0.15 ppm) were from Lys7, Ala8, Phe15, Lys18, Ile38 and Ala46 (Figure 4A). These residues were mapped on the structure indicated in Figure 4B. The pH titration of the most affected amide signals is indicated in Figure 4C. The NH signals of Lys7, Ala8, Phe15, Lys18 and Ala46 have more basic pKa values (6.9, 6.9, 7.9, 7.7 and 6.9, respectively) compared with that of Ile38 (5.8).

pH-linked conformational changes in ferricytochrome PpcA.

Figure 4.
pH-linked conformational changes in ferricytochrome PpcA.

(A) Weighted average of 1H and 15N chemical shifts (Δδavg) between pH 5.5 and 9.5. (B) Mapping of the residues showing large pH-dependent shifts on PpcA solution structure. Residues that are close to heme IV are indicated in green and Ile38 that forms a hydrogen bond with carboxyl oxygen is indicated in blue. Heme groups are colored red. (C) pH titration data of the most affected PpcA amide signals. In the expansion of each 1H–15N-HSQC NMR spectrum, the pH increases from red to green. Ala46 backbone amide signal is not shown since it appears in a very crowded region of the spectrum.

Figure 4.
pH-linked conformational changes in ferricytochrome PpcA.

(A) Weighted average of 1H and 15N chemical shifts (Δδavg) between pH 5.5 and 9.5. (B) Mapping of the residues showing large pH-dependent shifts on PpcA solution structure. Residues that are close to heme IV are indicated in green and Ile38 that forms a hydrogen bond with carboxyl oxygen is indicated in blue. Heme groups are colored red. (C) pH titration data of the most affected PpcA amide signals. In the expansion of each 1H–15N-HSQC NMR spectrum, the pH increases from red to green. Ala46 backbone amide signal is not shown since it appears in a very crowded region of the spectrum.

Discussion

The detailed thermodynamic characterization of the redox centers of PpcA, using NMR and visible spectroscopy, showed that each heme reduction potential is modulated by interactions with the neighboring ones and by the pH of the solution [22]. The modulation of the heme reduction potentials by the pH is designated the redox-Bohr effect, which in the case of PpcA gives the protein the necessary properties to establish preferred e/H+ transfer pathways in the physiological pH range [25]. The redox-Bohr center was assigned to the heme IV propionate [21], and the pKa value of this center is different in the oxidized (pKox = 6.5) and in the reduced (pKred = 8.6) forms [22]. The solution structure of PpcA in the reduced form was previously determined at a pH of 7.1 to avoid the presence of different levels of protonation of the redox-Bohr center and the concomitant broadening of the NMR signals [21]. Likewise, in the present work, the structure of PpcA in the oxidized form was determined at a pH of 5.5 to ensure that the redox-Bohr center is protonated and to investigate possible redox-linked conformational changes.

Heme electronic structure

In addition to the usual tests for validating structures in solution obtained by NMR, low-spin paramagnetic heme proteins can make use of dipolar shifts and Fermi contact shifts. Dipolar shifts are linked to the heme axial ligand geometry through the magnetic susceptibility tensor, and Fermi contact shifts are linked via the heme molecular orbitals [4446]. This is fortunate because NOEs to the His rings are bleached by paramagnetic relaxation. In each of the three hemes, the z-axis of the magnetic susceptibility tensor is found only slightly tilted from the normal to the heme plane, so that the in-plane rotation of the axes is well defined. The anisotropy of the tensors is correlated with the angle between the imidazole planes, β [44], with the values ordered Heme IV > Heme I > Heme III. Approximate heme frontier molecular orbitals are obtained from the 13C or 1H shifts of the heme substituents (Tables 3 and 4). They provide an independent estimate of the orientation of the average of the normals to the planes of the imidazole groups and the angle between them. Note that, as the axis of the molecular orbitals rotates, the in-plane axes of the magnetic susceptibility tensor rotate in the opposite sense, so that the values of θ obtained from the structure or from the methyl shifts have the opposite sign to those from the magnetic susceptibility. In fact, the agreement is only approximate, as might be expected because the uncertainties increase rapidly as β approaches 90°. Compared with the values taken from the oxidized structure, the heme axial geometries are in broad agreement, except for the ligands of heme III, in which case the overall orientation agrees but the angle between the His planes in the structure is probably too large.

Comparison of PpcA solution and crystal structures of PpcA family members in the oxidized state

As noted earlier, the crystal structure of PpcA is altered due to binding of deoxycholate [21]. Therefore, the solution structure of PpcA in the oxidized state was compared with the crystal structures available for the other four PpcA family members (PpcB–E) [20,23,24]. The sequence identities between PpcA and the other family members are 77%, 63%, 60% and 56% for PpcB, PpcC, PpcD and PpcE, respectively.

The general fold and the relative positions of the three hemes are similar, with a good agreement between the consensus secondary structure elements identified in the NMR and crystal structures. The parameters describing the heme geometry of the oxidized structure of PpcA in solution along with equivalent values taken from the crystal structures of the other family members are presented in Table 4. An overlap of the PpcA oxidized solution structure and PpcB crystal structure is shown in Figure 5. The average Fe–Fe distances found among the crystal structures of PpcB, PpcC, PpcD and PpcE differ by <3.5% compared with those of PpcA solution structure (Table 4). When the structures are overlapped using all Cα atoms, the rms deviations between the PpcA oxidized solution structure and the PpcB–E crystal structures are: PpcB (1.2 Å; 1.3 Å) < PpcC (1.4 Å) < PpcE (1.5 Å) < PpcD (1.9 Å; 1.8 Å). When the structures were overlapped using the three heme ring atoms including Fe atoms, the rms deviations are: PpcB (0.49 Å; 0.45 Å), PpcC (0.46 Å), PpcD (0.57 Å; 0.58 Å) and PpcE (0.47 Å). Interestingly, when the hemes were overlapped, the highest deviation was observed between the corresponding hemes III in each case. In all comparisons, apart from the terminal residues, which are generally less well defined, the largest deviations were consistently in segments around heme I with highest deviations observed for Pro25 and Lys33. It is remarkable that these are the same regions which differed between all the oxidized crystal structures of the family [24]. This region was previously proposed to interact with the electron acceptors; the lower structural degree of conservation among the five cytochromes might result from the structural specificity required to assure efficient electron transfer between each cytochrome and their electron acceptor. It is not clear whether all the changes observed in protein structure around heme I between the reduced and oxidized PpcA solution structures are fully redox-linked or if this region of the protein structure is less rigid and conducive to different conformations, especially in the oxidized state, as all the crystal structures were determined in this oxidation state.

Superimposition of the ferricytochrome PpcA solution structure (this work) and PpcB crystal structure (PDB code 3BXU).

Figure 5.
Superimposition of the ferricytochrome PpcA solution structure (this work) and PpcB crystal structure (PDB code 3BXU).

The PpcA and PpcB structures are indicated in gray and light green, respectively.

Figure 5.
Superimposition of the ferricytochrome PpcA solution structure (this work) and PpcB crystal structure (PDB code 3BXU).

The PpcA and PpcB structures are indicated in gray and light green, respectively.

Table 4
Heme geometry for G. sulfurreducens triheme cytochromes

The values for PpcA NMR solution structures in the oxidized (this work) and reduced [21] forms were obtained from the lowest energy structure.

 NMR Crystal structures3 
PpcAox PpcAred PpcB4 PpcC PpcD4 PpcE 
Heme Fe–Fe distance (Å) 
 I–III 11.6 11.7 11.7 (11.5) 11.2 11.2 (11.3) 11.8 
 I–IV 18.5 19.0 18.2 (18.8) 18.5 18.2 (18.2) 18.2 
 III–IV 13.0 12.6 12.6 (12.8) 12.6 12.6 (12.6) 12.4 
Angle between heme planes (°) 
 I–III 75 82 73 (72) 81 78 (76) 71 
 I–IV 27 27 20 (16) 18 25 (24) 22 
 III–IV 64 74 70 (71) 75 72 (70) 72 
 NMR Crystal structures3 
PpcAox PpcAred PpcB4 PpcC PpcD4 PpcE 
Heme Fe–Fe distance (Å) 
 I–III 11.6 11.7 11.7 (11.5) 11.2 11.2 (11.3) 11.8 
 I–IV 18.5 19.0 18.2 (18.8) 18.5 18.2 (18.2) 18.2 
 III–IV 13.0 12.6 12.6 (12.8) 12.6 12.6 (12.6) 12.4 
Angle between heme planes (°) 
 I–III 75 82 73 (72) 81 78 (76) 71 
 I–IV 27 27 20 (16) 18 25 (24) 22 
 III–IV 64 74 70 (71) 75 72 (70) 72 
1

The values were taken from ref. [24].

2

PpcB and PpcD have two independent molecules in the crystal. The values corresponding to the second molecule in each case are given in parentheses.

Redox-linked conformational changes

In solution, the reduced structure of the triheme cytochrome PpcA folds in a two-strand β-sheet at the N-terminus, formed by Asp3–Leu6 and Val13–Pro16 strands, followed by three α-helices between residues Ala19 and Lys22, Lys43 and His47 and Lys52 and Met58. The secondary structure elements of the reduced structure are similar to those in the oxidized structure except for an extra α-helix (Cys65–Cys68) and a β-strand (His69–Lys70) at the C-terminal end of the oxidized structure. Despite this, the general fold and the relative positions of the three hemes are similar (cf. Figure 2 and Table 4). It is worth noting that the extra helix and the β-strand observed at the C-terminus of the PpcA oxidized structure in solution are also observed in the oxidized crystal structure of PpcB [24].

Differences are observed in the geometry of the heme axial ligands, namely in the orientation of the average His normal (θ), relative to the NC–NA axis of the porphyrin, and in the angle between the imidazole planes (β) (Table 3). The principal difference between the oxidized and reduced structures is a rotation of ∼60° in the imidazole of the distal ligand to heme I, His31. The geometry of the structure in solution is fully consistent with the analysis of paramagnetic shifts of heme I (Table 3); hence, the change of orientation in the His31 imidazole ring between the oxidized and reduced structures is significant and is likely to be a redox-linked conformational change.

The comparison between the solution structures of PpcA in the oxidized and reduced forms (Figure 6) showed that the global rmsd value between them is 1.38 Å. The two loop regions around heme I that include the peptide segments between His17–Pro25 and the residues Gln32–Lys33 clearly have greater deviations, in agreement with the rotation of the heme I axial ligand. Interestingly, 15N NMR relaxation experiments have shown that the polypeptide segment between His17 and Cys27 that surrounds heme I is the most dynamic part of the PpcA-reduced structure [21], suggesting that this could form an interaction surface of PpcA with other proteins or molecules [21]. In addition, the detailed thermodynamic characterization of PpcA also indicates that hemes I and IV might have a role in recognition of PpcA redox partners [22]. No significant backbone dynamic effects were detected around heme IV, which has been suggested to be a requisite for selective sequence recognition [53]. Structurally, heme IV is stabilized by hydrogen bonds to both propionate groups, restricting the backbone dynamics around the heme. This is in line with the fact that the region around heme IV is highly conserved in the crystal structures of the entire family of triheme cytochromes, whereas that around heme I varies significantly, suggesting a functional significance [21,24]. In fact, it was proposed that heme IV interacts with the electron donor, probably located at the IM, whereas heme I was proposed to interact with the electron acceptors [21]. There are several examples in the literature showing that the heme redox potentials can be fine-tuned by subtle changes in the geometry of the axial ligands [5456]. Therefore, the observed changes in the orientation of the geometry of heme I axial ligands between the oxidized and reduced structures, as well as the neighboring loop regions, could fine-tune the reduction potential of heme I for electron transfer effectiveness when the protein meets its redox partners.

Comparison of the lowest energy solution structure of PpcA in the oxidized (this work) and reduced state (PDB code 2LDO [21]).

Figure 6.
Comparison of the lowest energy solution structure of PpcA in the oxidized (this work) and reduced state (PDB code 2LDO [21]).

Structures were superimposed in MOLMOL [57] using backbone atoms. (A) Average rmsd between the structures. (B) Superimposition of the PpcA solution structures in the oxidized (gray) and reduced (green) states.

Figure 6.
Comparison of the lowest energy solution structure of PpcA in the oxidized (this work) and reduced state (PDB code 2LDO [21]).

Structures were superimposed in MOLMOL [57] using backbone atoms. (A) Average rmsd between the structures. (B) Superimposition of the PpcA solution structures in the oxidized (gray) and reduced (green) states.

pH-linked conformational changes and their impact on the redox-Bohr center properties

The redox-Bohr effect in PpcA allows the protein to couple electron and proton transfer in the physiological pH range [22]. The detailed thermodynamic characterization of PpcA indicated that this coupling involves two microstates: the microstate with heme I oxidized and the redox-Bohr center protonated (P1H), and a deprotonated microstate with hemes I and IV both oxidized (P14) [22]. This involves a change in the pKa value of the redox-Bohr center between the oxidized and reduced forms. The redox-Bohr interactions determined in this study, which modulate the heme redox potentials upon deprotonation/protonation of the redox-Bohr center, were −32, −31 and −58 mV for hemes I, III and IV, respectively. These values suggest that the redox-Bohr center is located in the vicinity of heme IV, and it was assigned to heme IV propionate 13 [21]. The lower pKa of the redox-Bohr center in the oxidized state is expected based on electrostatic terms, since the removal of an electron from the heme favors deprotonation of the propionate. In addition, the conformation of the side chain of Lys43 in the oxidized state is significantly different from that in the reduced state (Figure 7), coming within hydrogen bonding distance to the propionate (NZ — O1D is 3.1 Å in the lowest energy structure). The interaction with Lys43 also promotes deprotonation of this propionic acid group, as reflected in the lower pKa.

Relative position of heme IV and residue 43 for the PpcA lowest energy solution structure in the oxidized (this work) and reduced state (PDB code 2LDO [21]).

Figure 7.
Relative position of heme IV and residue 43 for the PpcA lowest energy solution structure in the oxidized (this work) and reduced state (PDB code 2LDO [21]).

PpcA solution structures in the oxidized and reduced forms are represented in gray and green, respectively. The structures were superimposed in PyMOL [58] using backbone atoms.

Figure 7.
Relative position of heme IV and residue 43 for the PpcA lowest energy solution structure in the oxidized (this work) and reduced state (PDB code 2LDO [21]).

PpcA solution structures in the oxidized and reduced forms are represented in gray and green, respectively. The structures were superimposed in PyMOL [58] using backbone atoms.

The analysis of the chemical shift variation of the backbone amide signals with pH showed that those of Lys7, Ala8, Phe15, Lys18, Ile38 and Ala46 were the most affected (Figure 4A and Table 5). Lys7, Ala8 and Ala46 are located close to (redox-Bohr center), whereas the other residues are not (Figure 4B). A similar analysis was previously carried out for the fully reduced protein [21], which showed that NH signals of Lys7, Ala8, Asn10 and Ile38 were the most affected ones (Table 5). That the most affected signals are not exactly the same for oxidized and reduced protein suggests that localized conformational rearrangements occur in the H-bond network, particularly in the vicinity of the redox-Bohr center. In fact, the backbone NH signals of Phe15, Lys18 and Ala46 are affected by pH only in the oxidized form, and Asn10 is only affected in the reduced form. Analysis of the reduced structure showed that the backbone NH groups of Phe15, Lys18 and Ala46 establish H-bonds with the main chain oxygens of Ile4, Pro16 and Gly42, respectively, and are therefore not affected by the pH. However, the redox-linked rearrangements in the vicinity of these residues prevent them from forming H-bonds in the oxidized protein. In the case of Asn10, the reverse is true and it forms a new hydrogen bond with the O1A atom of in the oxidized structure.

Table 5
The most strongly pH-dependent amide signals in the reduced and oxidized forms of PpcA and corresponding pKa values

The signals that are not affected by the pH are labeled as n.a. The pKa values for the oxidized form were determined in this work and those for the reduced form were taken from ref. [21].

Residue Lys7 Ala8 Asn10 Phe15 Lys18 Ile38 Ala46 
 7.0 7.9 8.5 n.a. n.a. 5.5 n.a. 
 6.9 6.9 n.a. 7.9 7.7 5.8 6.9 
Residue Lys7 Ala8 Asn10 Phe15 Lys18 Ile38 Ala46 
 7.0 7.9 8.5 n.a. n.a. 5.5 n.a. 
 6.9 6.9 n.a. 7.9 7.7 5.8 6.9 

Residues 7, 8 and 38 are significantly affected in both states. The first two are part of the β-turn segment connecting the two-strand β-sheet near heme IV, whereas Ile38 forms a conserved hydrogen bond with the carboxyl oxygen of . As observed in the reduced protein, the chemical shift of the Ile38 amide signal decreases significantly at lower pH values, suggesting the disruption of a hydrogen bond formed between the amide proton of Ile38 and the carboxyl oxygen of upon protonation of the latter. The pKa value for Ile38 in the oxidized (5.8) and reduced (5.5) states is substantially lower than that of the other amide signals and clearly has a different origin. The same applies to residues Phe15 and Lys18, which show more basic pKa values (∼8). However, the amide signals of the residues located in the neighborhood of heme IV (Lys7, Ala8 and Ala46; see Figure 4B) have an identical pKa value (6.9), which correlates quite well with the pKa value of the redox-Bohr center in the fully oxidized protein (6.5) [22].

Overall, the study of the pH-linked conformational changes provided structural insights into the modulation of the pKa value of the redox-Bohr center in the reduced and oxidized states. Such modulation is crucial to allowing the protein to display a concerted e and H+ transfer at physiological pH and therefore to its function. Indeed, as shown previously [22], PpcA can take up strongly reducing electrons (−167 mV) and a weakly acidic proton (pKa 8) from the donor associated with the cytoplasmic membrane. When meeting the physiological downstream redox partner, PpcA donates less reducing electrons (−109 mV) and a more acidic proton (pKa 7.2). Such directional events can only be achieved by a redox-dependent modulation of the properties of the redox-Bohr center, whose structural determinants were identified for the first time in the present work.

Conclusions

The structure of triheme cytochrome PpcA presented here is the first solution structure of a cytochrome from G. sulfurreducens in the oxidized state. The pH value chosen for the structure determination ensured that the redox-Bohr center is protonated, which in turn assures comparability with the protonation state of the structure in the fully reduced state. Therefore, it was possible to probe redox- and pH-linked conformational changes in the most abundant cytochrome from G. sulfurreducens with its crucial role in the EET pathways of the bacterium. The structure of the oxidized PpcA was calculated together with its magnetic properties, which provided an independent test for the heme ligand geometry. Comparing the oxidized and reduced structures in solution showed concerted motion in the region of heme I involving two loop regions, as well as modification of the heme I axial ligand geometry. Although the changes in heme I axial ligand are very likely to be redox-linked, it is less clear whether all the other changes observed near heme I are redox-linked or not. However, the changes near heme I are in line with the previous suggestion that heme I is the most likely region of PpcA interacting with its electron acceptor [21]. Furthermore, the analysis of the chemical shift variation of the backbone amide signals with pH allowed us to identify structural rearrangements of the H-bond network in the vicinity of the redox-Bohr center (heme propionate ), map pH-linked conformational changes caused by protonation/deprotonation of this center and identify the structural origin for the difference observed in the and values of the redox-Bohr center. The pKa values of the closest residues to the redox-Bohr center matched the pKa values obtained independently from the global thermodynamic analysis of the properties of the redox centers in PpcA. The solution structure of PpcA together with chemical shift perturbation experiments on the NMR assigned signals can be used to identify and make a structural map of the interaction interfaces with potential interacting partners of this versatile electron transfer protein, with eventual realization of improved variants of G. sulfurreducens for applications in biotechnology.

Data bank accession number

The 1H and 15N chemical shifts have been deposited in the BioMagResBank (http://www.bmrm.wisc.edu) under BMRB accession number 25477. The solution structure of PpcA in the fully oxidized state has been deposited in the Protein Data Bank with RCSB ID code rcsb104222 and PDB ID code 2mz9.

Abbreviations

     
  • EET

    extracellular electron transfer

  •  
  • IM

    inner membrane

  •  
  • IPTG

    isopropyl β-d-thiogalactoside

  •  
  • MFCs

    microbial fuel cells

  •  
  • NOE

    Nuclear Overhauser effect

  •  
  • OM

    outer membrane

  •  
  • rmsd

    root mean square deviation.

Author contribution

L.M., M.B. and C.A.S. designed and conducted the NMR experiments. L.M. and D.L.T. calculated the solution structure. L.M., P.R.P., C.A.S. and D.L.T. analyzed the data and wrote the manuscript.

Funding

This work was supported by L'Oréal Portugal Medal of Honor for Women in Science, L'Oréal-UNESCO Awards (to L.M.), PTDC/BBB-BQB/3554/2014 (to C.A.S.) and UID/Multi/04378/2013 from Fundação para a Ciência e a Tecnologia, Portugal. The NMR spectrometers are part of The National NMR Facility, supported by Fundação para a Ciência e a Tecnologia [RECI/BBB-BQB/0230/2012]. P.R.P. is partially supported by the division of Chemical Sciences, Geosciences and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy program under contract no. DE-AC02-06CH11357.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Lovley
,
D.R.
,
Ueki
,
T.
,
Zhang
,
T.
,
Malvankar
,
N.S.
,
Shrestha
,
P.M.
,
Flanagan
,
K.A.
et al. 
(
2011
)
Geobacter: the microbe electric's physiology, ecology, and practical applications
.
Adv. Microb. Physiol.
59
,
1
100
doi:
2
Wrighton
,
K.C.
,
Agbo
,
P.
,
Warnecke
,
F.
,
Weber
,
K.A.
,
Brodie
,
E.L.
,
DeSantis
,
T.Z.
et al. 
(
2008
)
A novel ecological role of the Firmicutes identified in thermophilic microbial fuel cells
.
ISME J.
2
,
1146
1156
doi:
3
Bond
,
D.R.
and
Lovley
,
D.R.
(
2003
)
Electricity production by Geobacter sulfurreducens attached to electrodes
.
Appl. Environ. Microbiol.
69
,
1548
1555
doi:
4
Methé
,
B.A.
,
Nelson
,
K.E.
,
Eisen
,
J.A.
,
Paulsen
,
I.T.
,
Nelson
,
W.
,
Heidelberg
,
J.F.
et al. 
(
2003
)
Genome of Geobacter sulfurreducens: metal reduction in subsurface environments
.
Science
302
,
1967
1969
doi:
5
Kim
,
B.-C.
and
Lovley
,
D.R.
(
2008
)
Investigation of direct vs. indirect involvement of the c-type cytochrome MacA in Fe(III) reduction by Geobacter sulfurreducens
.
FEMS Microbiol. Lett.
286
,
39
44
doi:
6
Shelobolina
,
E.S.
,
Coppi
,
M.V.
,
Korenevsky
,
A.A.
,
Didonato
,
L.N.
,
Sullivan
,
S.A.
,
Konishi
,
H.
et al. 
(
2007
)
Importance of c-type cytochromes for U(VI) reduction by Geobacter sulfurreducens
.
BMC Microbiol.
7
,
16
doi:
7
Ding
,
Y.-H.R.
,
Hixson
,
K.K.
,
Aklujkar
,
M.A.
,
Lipton
,
M.S.
,
Smith
,
R.D.
,
Lovley
,
D.R.
et al. 
(
2008
)
Proteome of Geobacter sulfurreducens grown with Fe(III) oxide or Fe(III) citrate as the electron acceptor
.
Biochim. Biophys. Acta
1784
,
1935
1941
doi:
8
Lloyd
,
J.R.
,
Leang
,
C.
,
Hodges Myerson
,
A.L.
,
Coppi
,
M.V.
,
Ciufo
,
S.
,
Methe
,
B.
et al. 
(
2003
)
Biochemical and genetic characterization of PpcA, a periplasmic c-type cytochrome in Geobacter sulfurreducens
.
Biochem. J.
369
,
153
161
doi:
9
Mehta
,
T.
,
Coppi
,
M.V.
,
Childers
,
S.E.
and
Lovley
,
D.R.
(
2005
)
Outer membrane c-type cytochromes required for Fe(III) and Mn(IV) oxide reduction in Geobacter sulfurreducens
.
Appl. Environ. Microbiol.
71
,
8634
8641
doi:
10
Kim
,
B.-C.
,
Postier
,
B.L.
,
DiDonato
,
R.J.
,
Chaudhuri
,
S.K.
,
Nevin
,
K.P.
and
Lovley
,
D.R.
(
2008
)
Insights into genes involved in electricity generation in Geobacter sulfurreducens via whole genome microarray analysis of the OmcF-deficient mutant
.
Bioelectrochemistry
73
,
70
75
doi:
11
Nevin
,
K.P.
,
Kim
,
B.-C.
,
Glaven
,
R.H.
,
Johnson
,
J.P.
,
Woodard
,
T.L.
,
Methe
,
B.A.
et al. 
(
2009
)
Anode biofilm transcriptomics reveals outer surface components essential for high density current production in Geobacter sulfurreducens fuel cells
.
PLoS ONE
4
,
e5628
doi:
12
Holmes
,
D.E.
,
Chaudhuri
,
S.K.
,
Nevin
,
K.P.
,
Mehta
,
T.
,
Methe
,
B.A.
,
Liu
,
A.
et al. 
(
2006
)
Microarray and genetic analysis of electron transfer to electrodes in Geobacter sulfurreducens
.
Environ. Microbiol.
8
,
1805
1815
doi:
13
Leang
,
C.
,
Coppi
,
M.V.
and
Lovley
,
D.R.
(
2003
)
OmcB, a c-type polyheme cytochrome, involved in Fe(III) reduction in Geobacter sulfurreducens
.
J. Bacteriol.
185
,
2096
2103
doi:
14
Levar
,
C.E.
,
Chan
,
C.H.
,
Mehta-Kolte
,
M.G.
and
Bond
,
D.R.
(
2014
)
An inner membrane cytochrome required only for reduction of high redox potential extracellular electron acceptors
.
mBio
5
,
e02034
-
14
doi:
15
Zacharoff
,
L.
,
Chan
,
C.H.
and
Bond
,
D.R.
(
2016
)
Reduction of low potential electron acceptors requires the CbcL inner membrane cytochrome of Geobacter sulfurreducens
.
Bioelectrochemistry
107
,
7
13
doi:
16
Seidel
,
J.
,
Hoffmann
,
M.
,
Ellis
,
K.E.
,
Seidel
,
A.
,
Spatzal
,
T.
,
Gerhardt
,
S.
et al. 
(
2012
)
MacA is a second cytochrome c peroxidase of Geobacter sulfurreducens
.
Biochemistry
51
,
2747
2756
doi:
17
Morgado
,
L.
,
Fernandes
,
A.P.
,
Dantas
,
J.M.
,
Silva
,
M.A.
and
Salgueiro
,
C.A.
(
2012
)
On the road to improve the bioremediation and electricity-harvesting skills of Geobacter sulfurreducens: functional and structural characterization of multihaem cytochromes
.
Biochem. Soc. Trans.
40
,
1295
1301
doi:
18
Ding
,
Y.-H.R.
,
Hixson
,
K.K.
,
Giometti
,
C.S.
,
Stanley
,
A.
,
Esteve-Núñez
,
A.
,
Khare
,
T.
et al. 
(
2006
)
The proteome of dissimilatory metal-reducing microorganism Geobacter sulfurreducens under various growth conditions
.
Biochim. Biophys. Acta
1764
,
1198
1206
doi:
19
Esteve-Núñez
,
A.
,
Sosnik
,
J.
,
Visconti
,
P.
and
Lovley
,
D.R.
(
2008
)
Fluorescent properties of c-type cytochromes reveal their potential role as an extracytoplasmic electron sink in Geobacter sulfurreducens
.
Environ. Microbiol.
10
,
497
505
doi:
20
Morgado
,
L.
,
Bruix
,
M.
,
Orshonsky
,
V.
,
Londer
,
Y.Y.
,
Duke
,
N.E.C.
,
Yang
,
X.
et al. 
(
2008
)
Structural insights into the modulation of the redox properties of two Geobacter sulfurreducens homologous triheme cytochromes
.
Biochim. Biophys. Acta
1777
,
1157
1165
doi:
21
Morgado
,
L.
,
Paixão
,
V.B.
,
Schiffer
,
M.
,
Pokkuluri
,
P.R.
,
Bruix
,
M.
and
Salgueiro
,
C.A.
(
2012
)
Revealing the structural origin of the redox-Bohr effect: the first solution structure of a cytochrome from Geobacter sulfurreducens
.
Biochem. J.
441
,
179
187
doi:
22
Morgado
,
L.
,
Bruix
,
M.
,
Pessanha
,
M.
,
Londer
,
Y.Y.
and
Salgueiro
,
C.A.
(
2010
)
Thermodynamic characterization of a triheme cytochrome family from Geobacter sulfurreducens reveals mechanistic and functional diversity
.
Biophys. J.
99
,
293
301
doi:
23
Pokkuluri
,
P.R.
,
Londer
,
Y.Y.
,
Duke
,
N.E.C.
,
Long
,
W.C.
and
Schiffer
,
M.
(
2004
)
Family of cytochrome c7-type proteins from Geobacter sulfurreducens: structure of one cytochrome c7 at 1.45 Å resolution
.
Biochemistry
43
,
849
859
doi:
24
Pokkuluri
,
P.R.
,
Londer
,
Y.Y.
,
Yang
,
X.
,
Duke
,
N.E.C.
,
Erickson
,
J.
,
Orshonsky
,
V.
et al. 
(
2010
)
Structural characterization of a family of cytochromes c7 involved in Fe(III) respiration by Geobacter sulfurreducens
.
Biochim. Biophys. Acta
1797
,
222
232
doi:
25
Morgado
,
L.
,
Dantas
,
J.M.
,
Bruix
,
M.
,
Londer
,
Y.Y.
and
Salgueiro
,
C.A.
(
2012
)
Fine tuning of redox networks on multiheme cytochromes from Geobacter sulfurreducens drives physiological electron/proton energy transduction
.
Bioinorg. Chem. Appl.
2012
,
1
9
doi:
26
Pessanha
,
M.
,
Londer
,
Y.Y.
,
Long
,
W.C.
,
Erickson
,
J.
,
Pokkuluri
,
P.R.
,
Schiffer
,
M.
et al. 
(
2004
)
Redox characterization of Geobacter sulfurreducens cytochrome c7: physiological relevance of the conserved residue F15 probed by site-specific mutagenesis
.
Biochemistry
43
,
9909
9917
doi:
27
Dantas
,
J.M.
,
Morgado
,
L.
,
Londer
,
Y.Y.
,
Fernandes
,
A.P.
,
Louro
,
R.O.
,
Pokkuluri
,
P.R.
et al. 
(
2012
)
Pivotal role of the strictly conserved aromatic residue F15 in the cytochrome c7 family
.
J. Biol. Inorg. Chem.
17
,
11
24
doi:
28
Pokkuluri
,
P.R.
,
Yang
,
X.
,
Londer
,
Y.Y.
and
Schiffer
,
M.
(
2012
)
Pitfalls in the interpretation of structural changes in mutant proteins from crystal structures
.
J. Struct. Funct. Genomics
13
,
227
232
doi:
29
Morgado
,
L.
,
Dantas
,
J.M.
,
Simões
,
T.
,
Londer
,
Y.Y.
,
Pokkuluri
,
P.R.
and
Salgueiro
,
C.A.
(
2013
)
Role of Met58 in the regulation of electron/proton transfer in trihaem cytochrome PpcA from Geobacter sulfurreducens
.
Biosci. Rep.
33
,
11
22
doi:
30
Dantas
,
J.M.
,
Morgado
,
L.
,
Pokkuluri
,
P.R.
,
Turner
,
D.L.
and
Salgueiro
,
C.A.
(
2013
)
Solution structure of a mutant of the triheme cytochrome PpcA from Geobacter sulfurreducens sheds light on the role of the conserved aromatic residue F15
.
Biochim. Biophys. Acta
1827
,
484
492
doi:
31
Morgado
,
L.
,
Lourenço
,
S.
,
Londer
,
Y.Y.
,
Schiffer
,
M.
,
Pokkuluri
,
P.R.
and
Salgueiro
,
C.A.
(
2014
)
Dissecting the functional role of key residues in triheme cytochrome PpcA: a path to rational design of G. sulfurreducens strains with enhanced electron transfer capabilities
. PLoS ONE
9
,
e105566
doi:
32
Santos
,
T.C.
,
Silva
,
M.A.
,
Morgado
,
L.
,
Dantas
,
J.M.
and
Salgueiro
,
C.A.
(
2015
)
Diving into the redox properties of Geobacter sulfurreducens cytochromes: a model for extracellular electron transfer
.
Dalton Trans.
44
,
9335
9344
doi:
33
Morgado
,
L.
,
Saraiva
,
I.H.
,
Louro
,
R.O.
and
Salgueiro
,
C.A.
(
2010
)
Orientation of the axial ligands and magnetic properties of the hemes in the triheme ferricytochrome PpcA from G. sulfurreducens determined by paramagnetic NMR
. FEBS Lett.
584
,
3442
3445
doi:
34
Dantas
,
J.M.
,
Saraiva
,
I.H.
,
Morgado
,
L.
,
Silva
,
M.A.
,
Schiffer
,
M.
,
Salgueiro
,
C.A.
et al. 
(
2011
)
Orientation of the axial ligands and magnetic properties of the hemes in the cytochrome c7 family from Geobacter sulfurreducens determined by paramagnetic NMR
.
Dalton Trans.
40
,
12713
12718
doi:
35
Srikanth
,
S.
,
Marsili
,
E.
,
Flickinger
,
M.C.
and
Bond
,
D.R.
(
2008
)
Electrochemical characterization of Geobacter sulfurreducens cells immobilized on graphite paper electrodes
.
Biotechnol. Bioeng.
99
,
1065
1073
doi:
36
Liu
,
Y.
,
Kim
,
H.
,
Franklin
,
R.R.
and
Bond
,
D.R.
(
2011
)
Linking spectral and electrochemical analysis to monitor c-type cytochrome redox status in living Geobacter sulfurreducens biofilms
.
ChemPhysChem
12
,
2235
2241
doi:
37
Marsili
,
E.
,
Rollefson
,
J.B.
,
Baron
,
D.B.
,
Hozalski
,
R.M.
and
Bond
,
D.R.
(
2008
)
Microbial biofilm voltammetry: direct electrochemical characterization of catalytic electrode-attached biofilms
.
Appl. Environ. Microbiol.
74
,
7329
7337
doi:
38
Fernandes
,
A.P.
,
Couto
,
I.
,
Morgado
,
L.
,
Londer
,
Y.Y.
and
Salgueiro
,
C.A.
(
2008
)
Isotopic labeling of c-type multiheme cytochromes overexpressed in E. coli
.
Prot. Expr. Purif.
59
,
182
188
doi:
39
Londer
,
Y.Y.
,
Pokkuluri
,
P.R.
,
Tiede
,
D.M.
and
Schiffer
,
M.
(
2002
)
Production and preliminary characterization of a recombinant triheme cytochrome c7 from Geobacter sulfurreducens in Escherichia coli
.
Biochim. Biophys. Acta
1554
,
202
211
doi:
40
Garrett
,
D.S.
,
Seok
,
Y.-J.
,
Peterkofsky
,
A.
,
Clore
,
G.M.
and
Gronenborn
,
A.M.
(
1997
)
Identification by NMR of the binding surface for the histidine-containing phosphocarrier protein HPr on the N-terminal domain of enzyme I of the Escherichia coli phosphotransferase system
.
Biochemistry
36
,
4393
4398
doi:
41
Paixão
,
V.B.
,
Vis
,
H.
and
Turner
,
D.L.
(
2010
)
Redox linked conformational changes in cytochrome c3 from Desulfovibrio desulfuricans ATCC 27774
.
Biochemistry
49
,
9620
9629
doi:
42
Turner
,
D.L.
,
Brennan
,
L.
,
Chamberlin
,
S.G.
,
Louro
,
R.O.
and
Xavier
,
A.V.
(
1998
)
Determination of solution structures of paramagnetic proteins by NMR
.
Eur. Biophys. J.
27
,
367
375
doi:
43
Güntert
,
P.
,
Mumenthaler
,
C.
and
Wüthrich
,
K.
(
1997
)
Torsion angle dynamics for NMR structure calculation with the new program DYANA
.
J. Mol. Biol.
273
,
283
298
doi:
44
Turner
,
D.L.
,
Brennan
,
L.
,
Messias
,
A.C.
,
Teodoro
,
M.L.
and
Xavier
,
A.V.
(
2000
)
Correlation of empirical magnetic susceptibility tensors and structure in low-spin haem proteins
.
Eur. Biophys. J.
29
,
104
112
doi:
45
Turner
,
D.L.
(
1995
)
Determination of haem electronic structure in His-Met cytochromes c by 13C-NMR. The effect of the axial ligands
. Eur. J. Biochem.
227
,
829
837
doi:
46
Turner
,
D.L.
(
2000
)
Obtaining ligand geometries from paramagnetic shifts in low-spin haem proteins
.
J. Biol. Inorg. Chem.
5
,
328
332
doi:
47
Morgado
,
L.
,
Fernandes
,
A.P.
,
Londer
,
Y.Y.
,
Bruix
,
M.
and
Salgueiro
,
C.A.
(
2010
)
One simple step in the identification of the cofactors signals, one giant leap for the solution structure determination of multiheme proteins
.
Biochem. Biophys. Res. Commun.
393
,
466
470
doi:
48
Moss
,
G.P.
(
1988
)
Nomenclature of tetrapyrroles: recommendations 1986 IUPAC-IUB joint commission on biochemical nomenclature (JCBN)
.
Eur. J. Biochem.
178
,
277
328
doi:
49
Salgueiro
,
C.A.
,
Turner
,
D.L.
,
Santos
,
H.
,
LeGall
,
J.
and
Xavier
,
A.V.
(
1992
)
Assignment of the redox potentials to the four haems in Desulfovibrio vulgaris cytochrome c3 by 2D-NMR
.
FEBS Lett.
314
,
155
158
doi:
50
Turner
,
D.L.
,
Salgueiro
,
C.A.
,
Schenkels
,
P.
,
LeGall
,
J.
and
Xavier
,
A.V.
(
1995
)
Carbon-13 NMR studies of the influence of axial ligand orientation on haem electronic structure
.
Biochim. Biophys. Acta
1246
,
24
28
doi:
51
Salgueiro
,
C.A.
,
Turner
,
D.L.
and
Xavier
,
A.V.
(
1997
)
Use of paramagnetic NMR probes for structural analysis in cytochrome c3 from Desulfovibrio vulgaris
.
Eur. J. Biochem.
244
,
721
734
doi:
52
Turner
,
D.L.
,
Salgueiro
,
C.A.
,
LeGall
,
J.
and
Xavier
,
A.V.
(
1992
)
Structural studies of Desulfovibrio vulgaris ferrocytochrome c3 by two-dimensional NMR
.
Eur. J. Biochem.
210
,
931
936
doi:
53
Mandel
,
A.M.
,
Akke
,
M.
and
Palmer
, III,
A.G.
(
1995
)
Backbone dynamics of Escherichia coli ribonuclease HI: correlations with structure and function in an active enzyme
.
J. Mol. Biol.
246
,
144
163
doi:
54
Xue
,
L.-L.
,
Wang
,
Y.-H.
,
Xie
,
Y.
,
Yao
,
P.
,
Wang
,
W.-H.
,
Qian
,
W.
et al. 
(
1999
)
Effect of mutation at valine 61 on the three-dimensional structure, stability, and redox potential of cytochrome b5
.
Biochemistry
38
,
11961
11972
doi:
55
Sarma
,
S.
,
Dangi
,
B.
,
Yan
,
C.
,
DiGate
,
R.J.
,
Banville
,
D.L.
and
Guiles
,
R.D.
(
1997
)
Characterization of a site-directed mutant of cytochrome b5 designed to alter axial imidazole ligand plane orientation
.
Biochemistry
36
,
5645
5657
doi:
56
Sarma
,
S.
,
DiGate
,
R.J.
,
Goodin
,
D.B.
,
Miller
,
C.J.
and
Guiles
,
R.D.
(
1997
)
Effect of axial ligand plane reorientation on electronic and electrochemical properties observed in the A67V mutant of rat cytochrome b5
.
Biochemistry
36
,
5658
doi:
57
Koradi
,
R.
,
Billeter
,
M.
and
Wüthrich
,
K.
(
1996
)
MOLMOL: a program for display and analysis of macromolecular structures
.
J. Mol. Graph.
14
,
51
55
doi:
58
DeLano
,
W.L.
(
2002
)
The PyMOL Molecular Graphics System
. http://www.pymol.org

Author notes

*

Present address: Biozentrum, University of Basel, Basel, Switzerland.