- Email: [email protected]
One simple step in the identification of the cofactors signals, one giant leap for the solution structure determination of multiheme proteins
Biochemical and Biophysical Research Communications 393 (2010) 466–470
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
One simple step in the identification of the cofactors signals, one giant leap for the solution structure determination of multiheme proteins Leonor Morgado a, Ana P. Fernandes a, Yuri Y. Londer b,1, Marta Bruix c, Carlos A. Salgueiro a,* a
Requimte-CQFB, Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, Campus Caparica, 2829-516 Caparica, Portugal Biosciences Division, Argonne National Laboratory, Argonne, IL 60439, USA c Departamento de Espectroscopía y Estructura Molecular, Instituto de Química-Física ‘‘Rocasolano”, CSIC, Serrano 119, 28006 Madrid, Spain b
a r t i c l e
i n f o
Article history: Received 25 January 2010 Available online 10 February 2010 Keywords: Multiheme proteins Cytochromes NMR Isotopic labeling Protein structure
a b s t r a c t Multiheme proteins play major roles in various biological systems. Structural information on these systems in solution is crucial to understand their functional mechanisms. However, the presence of numerous proton-containing groups in the heme cofactors and the magnetic properties of the heme iron, in particular in the oxidised state, complicates significantly the assignment of the NMR signals. Consequently, the multiheme proteins superfamily is extremely under-represented in structural databases, which constitutes a severe bottleneck in the elucidation of their structural–functional relationships. In this work, we present a strategy that simplifies the assignment of the NMR signals in multiheme proteins and, concomitantly, their solution structure determination, using the triheme cytochrome PpcA from the bacterium Geobacter sulfurreducens as a model. Cost-effective isotopic labeling was used to double label (13C/15N) the protein in its polypeptide chain, with the correct folding and heme post-translational modifications. The combined analysis of 1H–13C HSQC NMR spectra obtained for labeled and unlabeled samples of PpcA allowed a straight discrimination between the heme cofactors and the polypeptide chain signals and their confident assignment. The results presented here will be the foundations to assist solution structure determination of multiheme proteins, which are still very scarce in the literature. Ó 2010 Elsevier Inc. All rights reserved.
Introduction Multiheme proteins drive essential cellular events in all biological systems, such as electron transfer (e.g., cytochromes), catalysis (e.g., cytochrome c oxidase), oxygen transport (e.g., hemoglobin), ligand binding and signal transduction (e.g., heme-based sensors). The vast functional variety of the heme-containing proteins is a consequence of the chemical versatility of the heme groups, whose properties are modulated by particular features of the heme iron and by the protein environment surrounding the heme (for a review see Bertini et al. [1]). Although functional and structural studies have provided in the past decade important data for heme proteins in general, the solution structural information is still scarce for multiheme proteins. The abundance of these proteins in several microorganisms has been revealed by the genome sequences recently obtained for several members of the c- and d-proteobacteria. The genomes of the bacteria Desulfovibrio vulgaris (Hildenborough) [2], Shewanella
* Corresponding author. Fax: +351 212 9485 50. E-mail address: [email protected] (C.A. Salgueiro). 1 Present address: New England Biolabs, 240 County Road, Ipswich, MA 01938, USA. 0006-291X/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2010.02.024
oneidensis (MR1) [3] and Geobacter sulfurreducens [4] encode large numbers of multiheme cytochromes. The total number of multiheme cytochromes produced by these microorganisms can represent up to 66% (73 out of 111) of the total number of the c-type cytochromes, as it is the case for G. sulfurreducens [4]. Phenotypic studies, using mutant bacteria with genes encoding multiheme proteins knocked-out, showed that many of these proteins are involved in crucial metabolic cellular pathways [5]. Important contributions for the solution structure determination of multiheme cytochromes have been made by the research groups of Xavier [6–9], Bertini [10,11] and Akutsu [12]. Although their molecular weights are appropriate for structural studies in solution, only a few structures have been determined for this superfamily. Indeed, reports of solution structures are still limited to four tetraheme cytochromes c3 [6–9,12] and to one triheme cytochrome [10,11]. In total, eight structures have been reported, five for fully reduced proteins [6,8–10,12] and three for fully oxidised [6,7,11]. This clearly contrasts with the number of structures determined by Xray crystallography [13]. The smaller number of solution structures obtained in the oxidised form is undoubtedly associated with the inherent complexity of the analysis of their NMR spectra. Indeed, in the oxidised form, the paramagnetic effect of the iron unpaired electrons, causes the spread of the signals of the heme cofactors, as
L. Morgado et al. / Biochemical and Biophysical Research Communications 393 (2010) 466–470
well as those of the amino acid residues located in their neighborhoods, all over large NMR spectral widths. Additionally, these resonances are generally broader, which makes the complete assignment of the heme and polypeptide resonances a laborious and very time consuming task. A second reason underlying the solution structural scarcity relates to the very poor expression yields obtained for mature multiheme proteins (<1 mg per liter of culture) [14–16], which makes their isotopic labeling inefficient. For all these reasons, the multiheme cytochromes superfamily is extremely under-represented in structural terms, which mitigates a deeper understanding of their functional mechanisms. An efficient expression system to produce multiheme c-type cytochromes, using Escherichia coli as host, was recently described and successfully applied to the expression of multiheme cytochromes containing up to 12 heme groups [17,18]. We showed that this system could also be used to achieve cost-effective labeling of multiheme cytochromes [19]. In the present work, we produced for the first time a 13C/15N double-labeled multiheme protein, the triheme cytochrome PpcA from G. sulfurreducens. By comparing the 1H–13C HSQC NMR spectra of labeled and unlabeled oxidised forms of PpcA, we could discriminate NMR signals originating from the heme cofactors from those of the polypeptide chain in a straightforward manner, which is a major breakthrough for gathering essential structural data.
Material and methods Bacterial strains, plasmids and growth conditions. Competent cells of E. coli strain BL21(DE3) (Novagen) containing plasmid pEC86 were co-transformed with plasmid pCK32. This plasmid is a pUCderivative containing the lac promoter, OmpA leader and the gene for mature PpcA. Construction of pCK32 was previously described [17]. Plasmid pEC86, a derivative of pACYC184, containing the cytochrome c maturation gene cluster, ccmABCDEFGH [20], was a kind gift from Dr. Thöny-Meyer (Zürich, Switzerland). E. coli cells harbouring both pEC86 and pCK32 plasmids were grown on 2xYT medium containing 34 lg/mL chloramphenicol (CLO) and 100 lg/ mL ampicillin (AMP), both from Sigma. Production and purification of unlabeled protein. PpcA was produced as previously described [17]. Briefly, E. coli strain BL21(DE3) cells harbouring both pEC86 and pCK32 plasmids were aerobically grown to mid-exponential phase at 30 °C in 2xYT media supplemented with CLO and AMP. At this point, cytochrome expression was induced with 10 lM IPTG. After overnight incubation at 30 °C, cells were harvested by centrifugation at 6400g for 20 min and resuspended in 30 mL of lysis buffer (100 mM Tris–HCl, pH 8.0, 0.5 mM EDTA, 20% sucrose) per liter of cell culture, containing 0.5 mg/mL lysozyme. The periplasmic fraction was recovered by centrifugation at 15,000g at 4 °C for 20 min. The supernatant constituting the periplasmic fraction was dialysed against 2 5 L of 20 mM Tris–HCl pH 8.5 and loaded onto 2 5 ml Econo-Pac High S cartridges (Bio-Rad), equilibrated with the same buffer. The protein was eluted with a sodium chloride gradient (0–300 mM). Protein purity was evaluated by Coomassie stained SDS-PAGE Purified PpcA was concentrated and the buffer was exchanged to 20 mM NaCl by ultrafiltration methods. Production and purification of 13C/15N labeled protein. Expression and purification of 13C/15N double-labeled PpcA was carried out according to the procedure previously described [19], adapted in the present study for double labeling of the protein. Briefly, cells of E. coli strain BL21(DE3) co-transformed with plasmids pEC86 and pCK32 were first grown in 2xYT media supplemented with CLO and AMP as described above. At mid-exponential growth phase, cells were harvested and resuspended in minimal media containing 1 g/L 15NH4Cl and 2 g/L (13C6)D-glucose both from CIL
467
isotopes. The minimal media was supplemented with 34 lg/mL CLO, 100 lg/mL AMP and 1 mM of unlabeled d-aminolevulinic acid (Sigma). Purification of the double-labeled PpcA was performed as described for the unlabeled protein. The concentrations of labeled and unlabeled PpcA samples were determined by measuring the absorbance of the reduced form at 552 nm, using an extinction coefficient of 97.5 mM 1 cm 1 [21]. NMR experiments. PpcA was lyophilised twice and 1.2 mM samples were prepared in 45 mM phosphate buffer pH 5.5 with NaCl (100 mM final ionic strength) in 92%H2O/8%D2O. All the NMR experiments were performed on a Bruker Avance-800 spectrometer at 25 °C using 3 mm diameter NMR tubes. 1D-1H NMR spectra were recorded before and after protein lyophilization to check for protein integrity. To discriminate between the polypeptide chain and heme signals 2D-1H–13C HSQC NMR spectra were acquired on both labeled (80 scans) and unlabeled (640 scans) PpcA oxidised samples. To assist the specific assignment of the heme cofactors signals, 2D-NOESY (80 ms mixing time) and 2D-TOCSY (25 and 75 ms) NMR spectra were acquired with the unlabeled sample. To specifically assign the polypeptide chain signals, 2D 1H–15NHSQC, 1H–13C HSQC and 3D HNCA, HNCOCA, HNCACB, HNCOCACB, HCCH-TOCSY spectra were acquired using the 13C/15N double-labeled sample. 1H chemical shifts were calibrated using the water signal as internal reference and the 13C/15N chemical shifts calibrated through indirect referencing [22]. The program Sparky – NMR Assignment and Integration Software was used for NMR spectra inspection and assignment of the signals. The specific assignment of the hemes signals to the respective heme was obtained following the strategy previously described [23,24]. Results and discussion Protein solution structure determination relies on distance restraints that are obtained for the measurement of the intensity of the NOEs signals between atoms in close spatial proximity. These restraints are obtained from the analysis of 2D-NOESY NMR spectra, for which an unambiguous assignment of the protein signals is essential. However, for multiheme proteins these assignments are not easily achievable, in particular for the paramagnetic ones, due to large spread and broadening of the signals. In this work, we used
Fig. 1. Crystal structure of PpcA in the oxidised form [39]. Roman numerals indicate the hemes (blue) by the order of attachment to the CXXCH motif in the polypeptide chain (gray). To maintain the consistency with the literature on structurally related tetraheme cytochromes c3 [28], the PpcA heme groups are numbered I, III and IV. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
468
L. Morgado et al. / Biochemical and Biophysical Research Communications 393 (2010) 466–470
the triheme cytochrome PpcA from G. sulfurreducens as a representative member of the multiheme proteins superfamily to illustrate how these difficulties can be overcome. PpcA is composed by 71 amino acids and three c-type heme groups (Fig. 1) with bis-His axial coordination. The heme groups are low-spin both in the reduced (S = 0) and in the oxidised (S = 1/2) forms [25]. Each heme includes several proton-containing groups (four methyl groups, four meso protons, two thioether protons, two thioether methyl and two propionate groups – see inset in Fig. 2) that contribute to the NMR resonances ensemble. Due to the proximity of these groups and those of the polypeptide chain (Fig. 1), a large number of NOE connectivities involving both types of signals are also present in multidimensional NMR spectra, which must be unambiguously identified and assigned so that adequate structural restraints can be established and used to assist the determination of the structure in solution. Thus, in addition to the conventional structural work related to the assignment of the polypeptide chain signals, in a multiheme protein it is also necessary to identify all the heme signals, their connectivities with the polypeptide chain and specifically assign them to their respective heme group. For all these reasons, the complete assignment of the 2D-NMR signals in a multiheme pro-
tein, in comparison with those of non-hemic proteins, is extremely laborious. For diamagnetic multiheme proteins, as it is the case of the reduced PpcA, the assignment of the heme substituents is facilitated since they are dominated by the porphyrin ring-current shifts and, therefore, appear in well defined regions of the 1H NMR spectra (Fig. 2A). The only exception is observed for the heme propionate protons as they are structurally more variable. The strategy to assign the heme proton signals in the reduced form of a low-spin protein containing c-type hemes was described by Keller and Wüthrich [26] for the monoheme ferrocytochrome c and this procedure has been used to assign the heme proton signals in reduced multiheme proteins containing up to four low-spin heme groups [12,27–34]. On the contrary, in the oxidised form of PpcA the unpaired electron of each heme iron exerts significant paramagnetic shifts on the heme signals and nearby residues. Consequently, the same type of signals is differently affected by the paramagnetic centers, showing different levels of broadness and being spread all over the entire NMR spectral width (Fig. 2B). For these reasons, the complete and unambiguous assignment of these signals in the oxidised
Fig. 2. 1D-1H NMR spectra of the reduced (A) and oxidised (B) unlabeled triheme cytochrome PpcA (1.2 mM) obtained at 25 °C. The typical regions of the heme substituents are indicated. The inset indicates a diagram of a heme c numbered according to the IUPAC-IUB nomenclature.
469
L. Morgado et al. / Biochemical and Biophysical Research Communications 393 (2010) 466–470
Fig. 3. 2D-1H–13C HSQC NMR spectra of the unlabeled (blue contours) and labeled (black contours) PpcA (1.2 mM) obtained at 25 °C, with 640 and 80 scans, respectively. In order to not overcrowd the figure only the resonances separated from the main signal envelope are indicated. Blue and black labels indicate the heme substituents and the polypeptide resonances, respectively. The peaks of the protons connected to the same carbon atom (CH2 groups) are linked by a straight line. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
form is arduous at best. In this case, the acquisition of 1H–13C HSQC NMR spectra may be used to assist the assignment of the heme substituents signals, since the dipolar shifts of the carbon nuclei attached to the pyrrole b-carbons are very small and, unlike the protons bonded to them, their Fermi contact shifts are directly proportional to the spin density on the pyrrole b-carbons [35]. A few 1H–13C HSQC NMR studies carried out on natural abundance samples have been reported for tetraheme cytochromes c3 [24,36–38]. However, the complete assignment of both proton and carbon resonances of the heme substituents could not be accomplished solely from the analysis of these spectra. This difficulty arises from the overlap between the heme signals with those of the polypeptide chain, which are also displaced from their typical positions in paramagnetic proteins, making their assignment extremely complex. This is well illustrated by the 2D 1H–13C HSQC NMR spectrum (blue contours in Fig. 3) obtained for the unlabeled oxidised PpcA. In this spectrum, both the heme cofactors and polypeptide chain 1H–13C connectivities are observable and cannot be distinguished. To overcome this difficulty, we double labeled (13C/15N) PpcA, exclusively on its polypeptide chain. The yield of double-labeled protein obtained after purification was 3.6 mg per litre of culture, sufficient for all the experiments. The labeled PpcA was used to reveal the connectivities arising entirely from the polypeptide chain in the 2D 1H–13C HSQC NMR spectrum (black contours in Fig. 3). This is possible because the heme precursor (d-aminolevulinic acid) added into the minimal media was not labeled and, consequently, the heme signals are not observable in a short (80 scans) 1H–13C HSQC NMR spectrum. Therefore, by simple comparison of the two NMR spectra, the heme
cofactors signals and those of the polypeptide chain can be easily discriminated and subsequently assigned. This comparison allowed identifying more than 90% of the heme substituents signals (Table 1), the exception being the signals that are severely broad or appear in extremely overcrowded spectral regions.
Table 1 1 H and 13C chemical shifts (ppm) of the heme substituents in the oxidised triheme PpcA. The resonances not detected by the direct comparison of the two spectra in Fig. 3 are indicated by ‘n.d.’. Heme substituent
Heme I 1
H
1
2 31 32 5 71 81 82 10 121 131 132 15 171 172 181 20
17.79 0.91 1.15 n.d. 10.43 4.34 4.00 1.54 20.65 2.63 6.72 1.48 0.33 3.71 0.89 3.01 0.87 0.43 15.71 0.70
Heme III 13
C
1
36.72 13.55 61.34 n.d. 23.09 8.80 53.13 23.42 49.63 18.00
12.18 2.56 2.24 3.90 18.00 0.86 1.01 n.d. 13.18 16.09 19.94 1.74 0.68 1.67 3.67 5.55 2.07 2.00 0.64 8.07
95.68 0.62 6.89 73.87 37.62 n.d.
H
Heme IV 13
C
1
23.86 21.36 75.98 46.67 41.15 26.54 16.02 n.d. 22.61 58.02
14.46 0.72 2.05 n.d. 10.99 0.04 1.63 1.13 17.38 2.43 6.45 0.10 0.44 0.75 2.87 4.18 0.86 0.26 14.58 1.60
n.d. 46.87 16.05 87.29 1.63 46.71
H
13
C
32.40 6.69 62.28 n.d. 23.26 7.40 62.33 n.d. 36.95 18.59 98.50 13.50 14.37 89.76 36.32 32.42
470
L. Morgado et al. / Biochemical and Biophysical Research Communications 393 (2010) 466–470
As mentioned above, the amino acid residues located in the close proximity of the paramagnetic hemes are strongly affected by the dipolar field of the heme centers, being displaced from their typical positions. This is the case of the aCH and bCH2 connectivities of the six heme axial histidines (H17, H20, H31, H47, H55 and H69), aCH of G11, K43, C51, G61, C65; bCH3 of A46 and cCH3 of V13. The displacement of these signals over the entire spectrum, as illustrated in Fig. 3, further complicates the assignment. Again, the specific labeling of the polypeptide chain allows identifying these amino acid signals. Besides signal assignment, this identification is of major importance for structure determination as it identifies the residues closer to the heme groups. In conclusion, in this work the polypeptide chain of a multiheme protein was double labeled (13C/15N) for the first time. This allowed the straightforward and unambiguous discrimination between the heme and the polypeptide chain signals in the 1H–13C HSQC NMR spectra. The clear identification of these signals is of major importance in multiheme proteins, since the displacement of the resonance by paramagnetic effects overcomplicates their assignment and the concomitant high-quality structural restraints, a crucial step in protein solution structure determination. The difficulties in obtaining such data are the main reasons for the scarcity of solution structures of multiheme proteins. As the number of heme cofactors increases, as for example in dodecaheme cytochromes, for which a total of 240 cofactor protons need to be identified, this step is even more crucial in structure determination. Overall, we envisage that the application of the strategy described here will have an important impact, opening new routes in the understanding of the structural and functional mechanisms in the superfamily of multiheme proteins. Acknowledgments This work was supported by project Grants PTDC/QUI/70182/ 2006 and PTDC/BIA-PRO/74498/2006 from FCT (Portugal) and Acção Integrada E-69/07 from Fundação das Universidades Portuguesas (Portugal); CTQ2008-0080/BQU and Hispanic-Portuguese Project HP2006-0047 from Ministerio de Educación y Ciencia (Spain). L.M. is supported by Fundação para a Ciência e a Tecnologia Grant SFRH/BD/37415/2007. The authors are grateful to Dr. M. Schiffer and Dr. P.R. Pokkuluri for critical reading of the manuscript. References [1] I. Bertini, A. Sigel, H. Sigel, Handbook on Metalloproteins, CRC Press, 2001. [2] J.F. Heidelberg, R. Seshadri, S.A. Haveman, et al., The genome sequence of the anaerobic, sulfate-reducing bacterium Desulfovibrio vulgaris Hildenborough, Nat. Biotechnol. 22 (2004) 554–559. [3] J.F. Heidelberg, I.T. Paulsen, K.E. Nelson, et al., Genome sequence of the dissimilatory metal ion-reducing bacterium Shewanella oneidensis, Nat. Biotechnol. 20 (2002) 1118–1123. [4] B.A. Methé, K.E. Nelson, J.A. Eisen, et al., Genome of Geobacter sulfurreducens: metal reduction in subsurface environments, Science 302 (2003) 1967–1969. [5] L. Shi, T.C. Squier, J.M. Zachara, J.K. Fredrickson, Respiration of metal (hydr)oxides by Shewanella and Geobacter: a key role for multihaem c-type cytochromes, Mol. Microbiol. 65 (2007) 12–20. [6] L. Brennan, D.L. Turner, A.C. Messias, et al., Structural basis for the network of functional cooperativities in cytochrome c3 from Desulfovibrio gigas: solution structures of the oxidised and reduced states, J. Mol. Biol. 298 (2000) 61–82. [7] A.C. Messias, A.P. Aguiar, L. Brennan, et al., Solution structures of tetrahaem ferricytochrome c3 from Desulfovibrio vulgaris (Hildenborough) and its K45Q mutant: the molecular basis of cooperativity, Biochim. Biophys. Acta 1757 (2006) 143–153. [8] A.C. Messias, D.H. Kastrau, H.S. Costa, et al., Solution structure of Desulfovibrio vulgaris (Hildenborough) ferrocytochrome c3: structural basis for functional cooperativity, J. Mol. Biol. 281 (1998) 719–739. [9] V.B. Paixão, C.A. Salgueiro, L. Brennan, et al., The solution structure of a tetraheme cytochrome from Shewanella frigidimarina reveals a novel family structural motif, Biochemistry 47 (2008) 11973–11980. [10] M. Assfalg, L. Banci, I. Bertini, et al., A proton-NMR investigation of the fully reduced cytochrome c7 from Desulfuromonas acetoxidans. Comparison between the reduced and the oxidized forms, Eur. J. Biochem. 266 (1999) 634–643.
[11] M. Assfalg, L. Banci, I. Bertini, et al., 800 MHz 1H NMR solution structure refinement of oxidized cytochrome c7 from Desulfuromonas acetoxidans, Eur. J. Biochem. 256 (1998) 261–270. [12] E. Harada, Y. Fukuoka, T. Ohmura, et al., Redox-coupled conformational alternations in cytochrome c3 from D. vulgaris Miyazaki F on the basis of its reduced solution structure, J. Mol. Biol. 319 (2002) 767–778. [13] C.G. Mowat, S.K. Chapman, Multi-heme cytochromes – new structures, new chemistry, Dalton Trans. (2005) 3381–3389. [14] A. Brige, J.A. Cole, W.R. Hagen, et al., Overproduction, purification and novel redox properties of the dihaem cytochrome c, NapB, from Haemophilus influenzae, Biochem. J. 356 (2001) 851–858. [15] P.N. da Costa, C. Conte, L.M. Saraiva, Expression of a Desulfovibrio tetraheme cytochrome c in Escherichia coli, Biochem. Biophys. Res. Commun. 268 (2000) 688–691. [16] M.L. Herbaud, C. Aubert, M.C. Durand, et al., Escherichia coli is able to produce heterologous tetraheme cytochrome c3 when the ccm genes are co-expressed, Biochim. Biophys. Acta 1481 (2000) 18–24. [17] Y.Y. Londer, P.R. Pokkuluri, D.M. Tiede, M. Schiffer, Production and preliminary characterization of a recombinant triheme cytochrome c7 from Geobacter sulfurreducens in Escherichia coli, Biochim. Biophys. Acta 1554 (2002) 202–211. [18] Y.Y. Londer, P.R. Pokkuluri, V. Orshonsky, et al., Heterologous expression of dodecaheme ‘‘nanowire” cytochromes c from Geobacter sulfurreducens, Protein Expr. Purif. 47 (2006) 241–248. [19] A.P. Fernandes, I. Couto, L. Morgado, et al., Isotopic labeling of c-type multiheme cytochromes overexpressed in E. coli, Protein Expr. Purif. 59 (2008) 182–188. [20] E. Arslan, H. Schulz, R. Zufferey, et al., Overproduction of the Bradyrhizobium japonicum c-type cytochrome subunits of the cbb3 oxidase in Escherichia coli, Biochem. Biophys. Res. Commun. 251 (1998) 744–747. [21] S. Seeliger, R. Cord-Ruwisch, B. Schink, A periplasmic and extracellular c-type cytochrome of Geobacter sulfurreducens acts as a ferric iron reductase and as an electron carrier to other acceptors or to partner bacteria, J. Bacteriol. 180 (1998) 3686–3691. [22] D.S. Wishart, C.G. Bigam, J. Yao, et al., 1H, 13C and 15N chemical shift referencing in biomolecular NMR, J. Biomol. NMR 6 (1995) 135–140. [23] C.A. Salgueiro, D.L. Turner, A.V. Xavier, Use of paramagnetic NMR probes for structural analysis in cytochrome c3 from Desulfovibrio vulgaris, Eur. J. Biochem. 244 (1997) 721–734. [24] D.L. Turner, C.A. Salgueiro, P. Schenkels, et al., Carbon-13 NMR studies of the influence of axial ligand orientation on haem electronic structure, Biochim. Biophys. Acta 1246 (1995) 24–28. [25] M. Pessanha, Y.Y. Londer, W.C. Long, et al., Redox characterization of Geobacter sulfurreducens cytochrome c7: physiological relevance of the conserved residue F15 probed by site-specific mutagenesis, Biochemistry 43 (2004) 9909–9917. [26] R.M. Keller, K. Wuthrich, Assignment of the heme c resonances in the 360 MHz H NMR spectra of cytochrome c, Biochim. Biophys. Acta 533 (1978) 195–208. [27] L. Morgado, M. Bruix, V. Orshonsky, et al., Structural insights into the modulation of the redox properties of two Geobacter sulfurreducens homologous triheme cytochromes, Biochim. Biophys. Acta 1777 (2008) 1157–1165. [28] M. Pessanha, L. Morgado, R.O. Louro, et al., Thermodynamic characterization of triheme cytochrome PpcA from Geobacter sulfurreducens: evidence for a role played in e /H+ energy transduction, Biochemistry 45 (2006) 13910–13917. [29] D.L. Turner, C.A. Salgueiro, J. LeGall, A.V. Xavier, Structural studies of Desulfovibrio vulgaris ferrocytochrome c3 by two-dimensional NMR, Eur. J. Biochem. 210 (1992) 931–936. [30] P.M. Pereira, I. Pacheco, D.L. Turner, R.O. Louro, Structure-function relationship in type II cytochrome c3 from Desulfovibrio africanus: a novel function in a familiar heme core, J. Biol. Inorg. Chem. 7 (2002) 815–822. [31] E. Harada, J. Kumagai, K. Ozawa, et al., A directional electron transfer regulator based on heme-chain architecture in the small tetraheme cytochrome c from Shewanella oneidensis, FEBS Lett. 532 (2002) 333–337. [32] I.B. Coutinho, D.L. Turner, M.Y. Liu, et al., Structure of the three-haem core of cytochrome c551.5 determined by 1H NMR, J. Biol. Inorg. Chem. 1 (1996) 305– 311. [33] M. Pessanha, L. Brennan, A.V. Xavier, et al., NMR structure of the haem core of a novel tetrahaem cytochrome isolated from Shewanella frigidimarina: identification of the haem-specific axial ligands and order of oxidation, FEBS Lett. 489 (2001) 8–13. [34] R.O. Louro, I. Pacheco, D.L. Turner, et al., Structural and functional characterization of cytochrome c3 from D. desulfuricans ATCC 27774 by 1H NMR, FEBS Lett. 390 (1996) 59–62. [35] J.R. Bolton, G.K. Fraenkel, Electron spin resonance study of the pairing theorem for alternant hydrocarbons: 13C splittings in the anthracene positive and negative ions, J. Chem. Phys. 40 (1964) 3307–3320. [36] R.O. Louro, M. Pessanha, G.A. Reid, et al., Determination of the orientation of the axial ligands and of the magnetic properties of the haems in the tetrahaem ferricytochrome from Shewanella frigidimarina, FEBS Lett. 531 (2002) 520–524. [37] R.O. Louro, I.J. Correia, L. Brennan, et al., Electronic structure of low-spin ferric porphyrins: 13C NMR studies of the influence of axial ligand orientation, J. Am. Chem. Soc. 120 (1998) 13240–13247. [38] D.L. Turner, H.S. Costa, I.B. Coutinho, et al., Assignment of the ligand geometry and redox potentials of the trihaem ferricytochrome c3 from Desulfuromonas acetoxidans, Eur. J. Biochem. 243 (1997) 474–481. [39] P.R. Pokkuluri, Y.Y. Londer, N.E. Duke, et al., Family of cytochrome c7-type proteins from Geobacter sulfurreducens: structure of one cytochrome c7 at 1.45 Å resolution, Biochemistry 43 (2004) 849–859.
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
One simple step in the identification of the cofactors signals, one giant leap for the solution structure determination of multiheme proteins Leonor Morgado a, Ana P. Fernandes a, Yuri Y. Londer b,1, Marta Bruix c, Carlos A. Salgueiro a,* a
Requimte-CQFB, Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, Campus Caparica, 2829-516 Caparica, Portugal Biosciences Division, Argonne National Laboratory, Argonne, IL 60439, USA c Departamento de Espectroscopía y Estructura Molecular, Instituto de Química-Física ‘‘Rocasolano”, CSIC, Serrano 119, 28006 Madrid, Spain b
a r t i c l e
i n f o
Article history: Received 25 January 2010 Available online 10 February 2010 Keywords: Multiheme proteins Cytochromes NMR Isotopic labeling Protein structure
a b s t r a c t Multiheme proteins play major roles in various biological systems. Structural information on these systems in solution is crucial to understand their functional mechanisms. However, the presence of numerous proton-containing groups in the heme cofactors and the magnetic properties of the heme iron, in particular in the oxidised state, complicates significantly the assignment of the NMR signals. Consequently, the multiheme proteins superfamily is extremely under-represented in structural databases, which constitutes a severe bottleneck in the elucidation of their structural–functional relationships. In this work, we present a strategy that simplifies the assignment of the NMR signals in multiheme proteins and, concomitantly, their solution structure determination, using the triheme cytochrome PpcA from the bacterium Geobacter sulfurreducens as a model. Cost-effective isotopic labeling was used to double label (13C/15N) the protein in its polypeptide chain, with the correct folding and heme post-translational modifications. The combined analysis of 1H–13C HSQC NMR spectra obtained for labeled and unlabeled samples of PpcA allowed a straight discrimination between the heme cofactors and the polypeptide chain signals and their confident assignment. The results presented here will be the foundations to assist solution structure determination of multiheme proteins, which are still very scarce in the literature. Ó 2010 Elsevier Inc. All rights reserved.
Introduction Multiheme proteins drive essential cellular events in all biological systems, such as electron transfer (e.g., cytochromes), catalysis (e.g., cytochrome c oxidase), oxygen transport (e.g., hemoglobin), ligand binding and signal transduction (e.g., heme-based sensors). The vast functional variety of the heme-containing proteins is a consequence of the chemical versatility of the heme groups, whose properties are modulated by particular features of the heme iron and by the protein environment surrounding the heme (for a review see Bertini et al. [1]). Although functional and structural studies have provided in the past decade important data for heme proteins in general, the solution structural information is still scarce for multiheme proteins. The abundance of these proteins in several microorganisms has been revealed by the genome sequences recently obtained for several members of the c- and d-proteobacteria. The genomes of the bacteria Desulfovibrio vulgaris (Hildenborough) [2], Shewanella
* Corresponding author. Fax: +351 212 9485 50. E-mail address: [email protected] (C.A. Salgueiro). 1 Present address: New England Biolabs, 240 County Road, Ipswich, MA 01938, USA. 0006-291X/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2010.02.024
oneidensis (MR1) [3] and Geobacter sulfurreducens [4] encode large numbers of multiheme cytochromes. The total number of multiheme cytochromes produced by these microorganisms can represent up to 66% (73 out of 111) of the total number of the c-type cytochromes, as it is the case for G. sulfurreducens [4]. Phenotypic studies, using mutant bacteria with genes encoding multiheme proteins knocked-out, showed that many of these proteins are involved in crucial metabolic cellular pathways [5]. Important contributions for the solution structure determination of multiheme cytochromes have been made by the research groups of Xavier [6–9], Bertini [10,11] and Akutsu [12]. Although their molecular weights are appropriate for structural studies in solution, only a few structures have been determined for this superfamily. Indeed, reports of solution structures are still limited to four tetraheme cytochromes c3 [6–9,12] and to one triheme cytochrome [10,11]. In total, eight structures have been reported, five for fully reduced proteins [6,8–10,12] and three for fully oxidised [6,7,11]. This clearly contrasts with the number of structures determined by Xray crystallography [13]. The smaller number of solution structures obtained in the oxidised form is undoubtedly associated with the inherent complexity of the analysis of their NMR spectra. Indeed, in the oxidised form, the paramagnetic effect of the iron unpaired electrons, causes the spread of the signals of the heme cofactors, as
L. Morgado et al. / Biochemical and Biophysical Research Communications 393 (2010) 466–470
well as those of the amino acid residues located in their neighborhoods, all over large NMR spectral widths. Additionally, these resonances are generally broader, which makes the complete assignment of the heme and polypeptide resonances a laborious and very time consuming task. A second reason underlying the solution structural scarcity relates to the very poor expression yields obtained for mature multiheme proteins (<1 mg per liter of culture) [14–16], which makes their isotopic labeling inefficient. For all these reasons, the multiheme cytochromes superfamily is extremely under-represented in structural terms, which mitigates a deeper understanding of their functional mechanisms. An efficient expression system to produce multiheme c-type cytochromes, using Escherichia coli as host, was recently described and successfully applied to the expression of multiheme cytochromes containing up to 12 heme groups [17,18]. We showed that this system could also be used to achieve cost-effective labeling of multiheme cytochromes [19]. In the present work, we produced for the first time a 13C/15N double-labeled multiheme protein, the triheme cytochrome PpcA from G. sulfurreducens. By comparing the 1H–13C HSQC NMR spectra of labeled and unlabeled oxidised forms of PpcA, we could discriminate NMR signals originating from the heme cofactors from those of the polypeptide chain in a straightforward manner, which is a major breakthrough for gathering essential structural data.
Material and methods Bacterial strains, plasmids and growth conditions. Competent cells of E. coli strain BL21(DE3) (Novagen) containing plasmid pEC86 were co-transformed with plasmid pCK32. This plasmid is a pUCderivative containing the lac promoter, OmpA leader and the gene for mature PpcA. Construction of pCK32 was previously described [17]. Plasmid pEC86, a derivative of pACYC184, containing the cytochrome c maturation gene cluster, ccmABCDEFGH [20], was a kind gift from Dr. Thöny-Meyer (Zürich, Switzerland). E. coli cells harbouring both pEC86 and pCK32 plasmids were grown on 2xYT medium containing 34 lg/mL chloramphenicol (CLO) and 100 lg/ mL ampicillin (AMP), both from Sigma. Production and purification of unlabeled protein. PpcA was produced as previously described [17]. Briefly, E. coli strain BL21(DE3) cells harbouring both pEC86 and pCK32 plasmids were aerobically grown to mid-exponential phase at 30 °C in 2xYT media supplemented with CLO and AMP. At this point, cytochrome expression was induced with 10 lM IPTG. After overnight incubation at 30 °C, cells were harvested by centrifugation at 6400g for 20 min and resuspended in 30 mL of lysis buffer (100 mM Tris–HCl, pH 8.0, 0.5 mM EDTA, 20% sucrose) per liter of cell culture, containing 0.5 mg/mL lysozyme. The periplasmic fraction was recovered by centrifugation at 15,000g at 4 °C for 20 min. The supernatant constituting the periplasmic fraction was dialysed against 2 5 L of 20 mM Tris–HCl pH 8.5 and loaded onto 2 5 ml Econo-Pac High S cartridges (Bio-Rad), equilibrated with the same buffer. The protein was eluted with a sodium chloride gradient (0–300 mM). Protein purity was evaluated by Coomassie stained SDS-PAGE Purified PpcA was concentrated and the buffer was exchanged to 20 mM NaCl by ultrafiltration methods. Production and purification of 13C/15N labeled protein. Expression and purification of 13C/15N double-labeled PpcA was carried out according to the procedure previously described [19], adapted in the present study for double labeling of the protein. Briefly, cells of E. coli strain BL21(DE3) co-transformed with plasmids pEC86 and pCK32 were first grown in 2xYT media supplemented with CLO and AMP as described above. At mid-exponential growth phase, cells were harvested and resuspended in minimal media containing 1 g/L 15NH4Cl and 2 g/L (13C6)D-glucose both from CIL
467
isotopes. The minimal media was supplemented with 34 lg/mL CLO, 100 lg/mL AMP and 1 mM of unlabeled d-aminolevulinic acid (Sigma). Purification of the double-labeled PpcA was performed as described for the unlabeled protein. The concentrations of labeled and unlabeled PpcA samples were determined by measuring the absorbance of the reduced form at 552 nm, using an extinction coefficient of 97.5 mM 1 cm 1 [21]. NMR experiments. PpcA was lyophilised twice and 1.2 mM samples were prepared in 45 mM phosphate buffer pH 5.5 with NaCl (100 mM final ionic strength) in 92%H2O/8%D2O. All the NMR experiments were performed on a Bruker Avance-800 spectrometer at 25 °C using 3 mm diameter NMR tubes. 1D-1H NMR spectra were recorded before and after protein lyophilization to check for protein integrity. To discriminate between the polypeptide chain and heme signals 2D-1H–13C HSQC NMR spectra were acquired on both labeled (80 scans) and unlabeled (640 scans) PpcA oxidised samples. To assist the specific assignment of the heme cofactors signals, 2D-NOESY (80 ms mixing time) and 2D-TOCSY (25 and 75 ms) NMR spectra were acquired with the unlabeled sample. To specifically assign the polypeptide chain signals, 2D 1H–15NHSQC, 1H–13C HSQC and 3D HNCA, HNCOCA, HNCACB, HNCOCACB, HCCH-TOCSY spectra were acquired using the 13C/15N double-labeled sample. 1H chemical shifts were calibrated using the water signal as internal reference and the 13C/15N chemical shifts calibrated through indirect referencing [22]. The program Sparky – NMR Assignment and Integration Software was used for NMR spectra inspection and assignment of the signals. The specific assignment of the hemes signals to the respective heme was obtained following the strategy previously described [23,24]. Results and discussion Protein solution structure determination relies on distance restraints that are obtained for the measurement of the intensity of the NOEs signals between atoms in close spatial proximity. These restraints are obtained from the analysis of 2D-NOESY NMR spectra, for which an unambiguous assignment of the protein signals is essential. However, for multiheme proteins these assignments are not easily achievable, in particular for the paramagnetic ones, due to large spread and broadening of the signals. In this work, we used
Fig. 1. Crystal structure of PpcA in the oxidised form [39]. Roman numerals indicate the hemes (blue) by the order of attachment to the CXXCH motif in the polypeptide chain (gray). To maintain the consistency with the literature on structurally related tetraheme cytochromes c3 [28], the PpcA heme groups are numbered I, III and IV. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
468
L. Morgado et al. / Biochemical and Biophysical Research Communications 393 (2010) 466–470
the triheme cytochrome PpcA from G. sulfurreducens as a representative member of the multiheme proteins superfamily to illustrate how these difficulties can be overcome. PpcA is composed by 71 amino acids and three c-type heme groups (Fig. 1) with bis-His axial coordination. The heme groups are low-spin both in the reduced (S = 0) and in the oxidised (S = 1/2) forms [25]. Each heme includes several proton-containing groups (four methyl groups, four meso protons, two thioether protons, two thioether methyl and two propionate groups – see inset in Fig. 2) that contribute to the NMR resonances ensemble. Due to the proximity of these groups and those of the polypeptide chain (Fig. 1), a large number of NOE connectivities involving both types of signals are also present in multidimensional NMR spectra, which must be unambiguously identified and assigned so that adequate structural restraints can be established and used to assist the determination of the structure in solution. Thus, in addition to the conventional structural work related to the assignment of the polypeptide chain signals, in a multiheme protein it is also necessary to identify all the heme signals, their connectivities with the polypeptide chain and specifically assign them to their respective heme group. For all these reasons, the complete assignment of the 2D-NMR signals in a multiheme pro-
tein, in comparison with those of non-hemic proteins, is extremely laborious. For diamagnetic multiheme proteins, as it is the case of the reduced PpcA, the assignment of the heme substituents is facilitated since they are dominated by the porphyrin ring-current shifts and, therefore, appear in well defined regions of the 1H NMR spectra (Fig. 2A). The only exception is observed for the heme propionate protons as they are structurally more variable. The strategy to assign the heme proton signals in the reduced form of a low-spin protein containing c-type hemes was described by Keller and Wüthrich [26] for the monoheme ferrocytochrome c and this procedure has been used to assign the heme proton signals in reduced multiheme proteins containing up to four low-spin heme groups [12,27–34]. On the contrary, in the oxidised form of PpcA the unpaired electron of each heme iron exerts significant paramagnetic shifts on the heme signals and nearby residues. Consequently, the same type of signals is differently affected by the paramagnetic centers, showing different levels of broadness and being spread all over the entire NMR spectral width (Fig. 2B). For these reasons, the complete and unambiguous assignment of these signals in the oxidised
Fig. 2. 1D-1H NMR spectra of the reduced (A) and oxidised (B) unlabeled triheme cytochrome PpcA (1.2 mM) obtained at 25 °C. The typical regions of the heme substituents are indicated. The inset indicates a diagram of a heme c numbered according to the IUPAC-IUB nomenclature.
469
L. Morgado et al. / Biochemical and Biophysical Research Communications 393 (2010) 466–470
Fig. 3. 2D-1H–13C HSQC NMR spectra of the unlabeled (blue contours) and labeled (black contours) PpcA (1.2 mM) obtained at 25 °C, with 640 and 80 scans, respectively. In order to not overcrowd the figure only the resonances separated from the main signal envelope are indicated. Blue and black labels indicate the heme substituents and the polypeptide resonances, respectively. The peaks of the protons connected to the same carbon atom (CH2 groups) are linked by a straight line. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
form is arduous at best. In this case, the acquisition of 1H–13C HSQC NMR spectra may be used to assist the assignment of the heme substituents signals, since the dipolar shifts of the carbon nuclei attached to the pyrrole b-carbons are very small and, unlike the protons bonded to them, their Fermi contact shifts are directly proportional to the spin density on the pyrrole b-carbons [35]. A few 1H–13C HSQC NMR studies carried out on natural abundance samples have been reported for tetraheme cytochromes c3 [24,36–38]. However, the complete assignment of both proton and carbon resonances of the heme substituents could not be accomplished solely from the analysis of these spectra. This difficulty arises from the overlap between the heme signals with those of the polypeptide chain, which are also displaced from their typical positions in paramagnetic proteins, making their assignment extremely complex. This is well illustrated by the 2D 1H–13C HSQC NMR spectrum (blue contours in Fig. 3) obtained for the unlabeled oxidised PpcA. In this spectrum, both the heme cofactors and polypeptide chain 1H–13C connectivities are observable and cannot be distinguished. To overcome this difficulty, we double labeled (13C/15N) PpcA, exclusively on its polypeptide chain. The yield of double-labeled protein obtained after purification was 3.6 mg per litre of culture, sufficient for all the experiments. The labeled PpcA was used to reveal the connectivities arising entirely from the polypeptide chain in the 2D 1H–13C HSQC NMR spectrum (black contours in Fig. 3). This is possible because the heme precursor (d-aminolevulinic acid) added into the minimal media was not labeled and, consequently, the heme signals are not observable in a short (80 scans) 1H–13C HSQC NMR spectrum. Therefore, by simple comparison of the two NMR spectra, the heme
cofactors signals and those of the polypeptide chain can be easily discriminated and subsequently assigned. This comparison allowed identifying more than 90% of the heme substituents signals (Table 1), the exception being the signals that are severely broad or appear in extremely overcrowded spectral regions.
Table 1 1 H and 13C chemical shifts (ppm) of the heme substituents in the oxidised triheme PpcA. The resonances not detected by the direct comparison of the two spectra in Fig. 3 are indicated by ‘n.d.’. Heme substituent
Heme I 1
H
1
2 31 32 5 71 81 82 10 121 131 132 15 171 172 181 20
17.79 0.91 1.15 n.d. 10.43 4.34 4.00 1.54 20.65 2.63 6.72 1.48 0.33 3.71 0.89 3.01 0.87 0.43 15.71 0.70
Heme III 13
C
1
36.72 13.55 61.34 n.d. 23.09 8.80 53.13 23.42 49.63 18.00
12.18 2.56 2.24 3.90 18.00 0.86 1.01 n.d. 13.18 16.09 19.94 1.74 0.68 1.67 3.67 5.55 2.07 2.00 0.64 8.07
95.68 0.62 6.89 73.87 37.62 n.d.
H
Heme IV 13
C
1
23.86 21.36 75.98 46.67 41.15 26.54 16.02 n.d. 22.61 58.02
14.46 0.72 2.05 n.d. 10.99 0.04 1.63 1.13 17.38 2.43 6.45 0.10 0.44 0.75 2.87 4.18 0.86 0.26 14.58 1.60
n.d. 46.87 16.05 87.29 1.63 46.71
H
13
C
32.40 6.69 62.28 n.d. 23.26 7.40 62.33 n.d. 36.95 18.59 98.50 13.50 14.37 89.76 36.32 32.42
470
L. Morgado et al. / Biochemical and Biophysical Research Communications 393 (2010) 466–470
As mentioned above, the amino acid residues located in the close proximity of the paramagnetic hemes are strongly affected by the dipolar field of the heme centers, being displaced from their typical positions. This is the case of the aCH and bCH2 connectivities of the six heme axial histidines (H17, H20, H31, H47, H55 and H69), aCH of G11, K43, C51, G61, C65; bCH3 of A46 and cCH3 of V13. The displacement of these signals over the entire spectrum, as illustrated in Fig. 3, further complicates the assignment. Again, the specific labeling of the polypeptide chain allows identifying these amino acid signals. Besides signal assignment, this identification is of major importance for structure determination as it identifies the residues closer to the heme groups. In conclusion, in this work the polypeptide chain of a multiheme protein was double labeled (13C/15N) for the first time. This allowed the straightforward and unambiguous discrimination between the heme and the polypeptide chain signals in the 1H–13C HSQC NMR spectra. The clear identification of these signals is of major importance in multiheme proteins, since the displacement of the resonance by paramagnetic effects overcomplicates their assignment and the concomitant high-quality structural restraints, a crucial step in protein solution structure determination. The difficulties in obtaining such data are the main reasons for the scarcity of solution structures of multiheme proteins. As the number of heme cofactors increases, as for example in dodecaheme cytochromes, for which a total of 240 cofactor protons need to be identified, this step is even more crucial in structure determination. Overall, we envisage that the application of the strategy described here will have an important impact, opening new routes in the understanding of the structural and functional mechanisms in the superfamily of multiheme proteins. Acknowledgments This work was supported by project Grants PTDC/QUI/70182/ 2006 and PTDC/BIA-PRO/74498/2006 from FCT (Portugal) and Acção Integrada E-69/07 from Fundação das Universidades Portuguesas (Portugal); CTQ2008-0080/BQU and Hispanic-Portuguese Project HP2006-0047 from Ministerio de Educación y Ciencia (Spain). L.M. is supported by Fundação para a Ciência e a Tecnologia Grant SFRH/BD/37415/2007. The authors are grateful to Dr. M. Schiffer and Dr. P.R. Pokkuluri for critical reading of the manuscript. References [1] I. Bertini, A. Sigel, H. Sigel, Handbook on Metalloproteins, CRC Press, 2001. [2] J.F. Heidelberg, R. Seshadri, S.A. Haveman, et al., The genome sequence of the anaerobic, sulfate-reducing bacterium Desulfovibrio vulgaris Hildenborough, Nat. Biotechnol. 22 (2004) 554–559. [3] J.F. Heidelberg, I.T. Paulsen, K.E. Nelson, et al., Genome sequence of the dissimilatory metal ion-reducing bacterium Shewanella oneidensis, Nat. Biotechnol. 20 (2002) 1118–1123. [4] B.A. Methé, K.E. Nelson, J.A. Eisen, et al., Genome of Geobacter sulfurreducens: metal reduction in subsurface environments, Science 302 (2003) 1967–1969. [5] L. Shi, T.C. Squier, J.M. Zachara, J.K. Fredrickson, Respiration of metal (hydr)oxides by Shewanella and Geobacter: a key role for multihaem c-type cytochromes, Mol. Microbiol. 65 (2007) 12–20. [6] L. Brennan, D.L. Turner, A.C. Messias, et al., Structural basis for the network of functional cooperativities in cytochrome c3 from Desulfovibrio gigas: solution structures of the oxidised and reduced states, J. Mol. Biol. 298 (2000) 61–82. [7] A.C. Messias, A.P. Aguiar, L. Brennan, et al., Solution structures of tetrahaem ferricytochrome c3 from Desulfovibrio vulgaris (Hildenborough) and its K45Q mutant: the molecular basis of cooperativity, Biochim. Biophys. Acta 1757 (2006) 143–153. [8] A.C. Messias, D.H. Kastrau, H.S. Costa, et al., Solution structure of Desulfovibrio vulgaris (Hildenborough) ferrocytochrome c3: structural basis for functional cooperativity, J. Mol. Biol. 281 (1998) 719–739. [9] V.B. Paixão, C.A. Salgueiro, L. Brennan, et al., The solution structure of a tetraheme cytochrome from Shewanella frigidimarina reveals a novel family structural motif, Biochemistry 47 (2008) 11973–11980. [10] M. Assfalg, L. Banci, I. Bertini, et al., A proton-NMR investigation of the fully reduced cytochrome c7 from Desulfuromonas acetoxidans. Comparison between the reduced and the oxidized forms, Eur. J. Biochem. 266 (1999) 634–643.
[11] M. Assfalg, L. Banci, I. Bertini, et al., 800 MHz 1H NMR solution structure refinement of oxidized cytochrome c7 from Desulfuromonas acetoxidans, Eur. J. Biochem. 256 (1998) 261–270. [12] E. Harada, Y. Fukuoka, T. Ohmura, et al., Redox-coupled conformational alternations in cytochrome c3 from D. vulgaris Miyazaki F on the basis of its reduced solution structure, J. Mol. Biol. 319 (2002) 767–778. [13] C.G. Mowat, S.K. Chapman, Multi-heme cytochromes – new structures, new chemistry, Dalton Trans. (2005) 3381–3389. [14] A. Brige, J.A. Cole, W.R. Hagen, et al., Overproduction, purification and novel redox properties of the dihaem cytochrome c, NapB, from Haemophilus influenzae, Biochem. J. 356 (2001) 851–858. [15] P.N. da Costa, C. Conte, L.M. Saraiva, Expression of a Desulfovibrio tetraheme cytochrome c in Escherichia coli, Biochem. Biophys. Res. Commun. 268 (2000) 688–691. [16] M.L. Herbaud, C. Aubert, M.C. Durand, et al., Escherichia coli is able to produce heterologous tetraheme cytochrome c3 when the ccm genes are co-expressed, Biochim. Biophys. Acta 1481 (2000) 18–24. [17] Y.Y. Londer, P.R. Pokkuluri, D.M. Tiede, M. Schiffer, Production and preliminary characterization of a recombinant triheme cytochrome c7 from Geobacter sulfurreducens in Escherichia coli, Biochim. Biophys. Acta 1554 (2002) 202–211. [18] Y.Y. Londer, P.R. Pokkuluri, V. Orshonsky, et al., Heterologous expression of dodecaheme ‘‘nanowire” cytochromes c from Geobacter sulfurreducens, Protein Expr. Purif. 47 (2006) 241–248. [19] A.P. Fernandes, I. Couto, L. Morgado, et al., Isotopic labeling of c-type multiheme cytochromes overexpressed in E. coli, Protein Expr. Purif. 59 (2008) 182–188. [20] E. Arslan, H. Schulz, R. Zufferey, et al., Overproduction of the Bradyrhizobium japonicum c-type cytochrome subunits of the cbb3 oxidase in Escherichia coli, Biochem. Biophys. Res. Commun. 251 (1998) 744–747. [21] S. Seeliger, R. Cord-Ruwisch, B. Schink, A periplasmic and extracellular c-type cytochrome of Geobacter sulfurreducens acts as a ferric iron reductase and as an electron carrier to other acceptors or to partner bacteria, J. Bacteriol. 180 (1998) 3686–3691. [22] D.S. Wishart, C.G. Bigam, J. Yao, et al., 1H, 13C and 15N chemical shift referencing in biomolecular NMR, J. Biomol. NMR 6 (1995) 135–140. [23] C.A. Salgueiro, D.L. Turner, A.V. Xavier, Use of paramagnetic NMR probes for structural analysis in cytochrome c3 from Desulfovibrio vulgaris, Eur. J. Biochem. 244 (1997) 721–734. [24] D.L. Turner, C.A. Salgueiro, P. Schenkels, et al., Carbon-13 NMR studies of the influence of axial ligand orientation on haem electronic structure, Biochim. Biophys. Acta 1246 (1995) 24–28. [25] M. Pessanha, Y.Y. Londer, W.C. Long, et al., Redox characterization of Geobacter sulfurreducens cytochrome c7: physiological relevance of the conserved residue F15 probed by site-specific mutagenesis, Biochemistry 43 (2004) 9909–9917. [26] R.M. Keller, K. Wuthrich, Assignment of the heme c resonances in the 360 MHz H NMR spectra of cytochrome c, Biochim. Biophys. Acta 533 (1978) 195–208. [27] L. Morgado, M. Bruix, V. Orshonsky, et al., Structural insights into the modulation of the redox properties of two Geobacter sulfurreducens homologous triheme cytochromes, Biochim. Biophys. Acta 1777 (2008) 1157–1165. [28] M. Pessanha, L. Morgado, R.O. Louro, et al., Thermodynamic characterization of triheme cytochrome PpcA from Geobacter sulfurreducens: evidence for a role played in e /H+ energy transduction, Biochemistry 45 (2006) 13910–13917. [29] D.L. Turner, C.A. Salgueiro, J. LeGall, A.V. Xavier, Structural studies of Desulfovibrio vulgaris ferrocytochrome c3 by two-dimensional NMR, Eur. J. Biochem. 210 (1992) 931–936. [30] P.M. Pereira, I. Pacheco, D.L. Turner, R.O. Louro, Structure-function relationship in type II cytochrome c3 from Desulfovibrio africanus: a novel function in a familiar heme core, J. Biol. Inorg. Chem. 7 (2002) 815–822. [31] E. Harada, J. Kumagai, K. Ozawa, et al., A directional electron transfer regulator based on heme-chain architecture in the small tetraheme cytochrome c from Shewanella oneidensis, FEBS Lett. 532 (2002) 333–337. [32] I.B. Coutinho, D.L. Turner, M.Y. Liu, et al., Structure of the three-haem core of cytochrome c551.5 determined by 1H NMR, J. Biol. Inorg. Chem. 1 (1996) 305– 311. [33] M. Pessanha, L. Brennan, A.V. Xavier, et al., NMR structure of the haem core of a novel tetrahaem cytochrome isolated from Shewanella frigidimarina: identification of the haem-specific axial ligands and order of oxidation, FEBS Lett. 489 (2001) 8–13. [34] R.O. Louro, I. Pacheco, D.L. Turner, et al., Structural and functional characterization of cytochrome c3 from D. desulfuricans ATCC 27774 by 1H NMR, FEBS Lett. 390 (1996) 59–62. [35] J.R. Bolton, G.K. Fraenkel, Electron spin resonance study of the pairing theorem for alternant hydrocarbons: 13C splittings in the anthracene positive and negative ions, J. Chem. Phys. 40 (1964) 3307–3320. [36] R.O. Louro, M. Pessanha, G.A. Reid, et al., Determination of the orientation of the axial ligands and of the magnetic properties of the haems in the tetrahaem ferricytochrome from Shewanella frigidimarina, FEBS Lett. 531 (2002) 520–524. [37] R.O. Louro, I.J. Correia, L. Brennan, et al., Electronic structure of low-spin ferric porphyrins: 13C NMR studies of the influence of axial ligand orientation, J. Am. Chem. Soc. 120 (1998) 13240–13247. [38] D.L. Turner, H.S. Costa, I.B. Coutinho, et al., Assignment of the ligand geometry and redox potentials of the trihaem ferricytochrome c3 from Desulfuromonas acetoxidans, Eur. J. Biochem. 243 (1997) 474–481. [39] P.R. Pokkuluri, Y.Y. Londer, N.E. Duke, et al., Family of cytochrome c7-type proteins from Geobacter sulfurreducens: structure of one cytochrome c7 at 1.45 Å resolution, Biochemistry 43 (2004) 849–859.