The contribution of decameric vanadate species to vanadate toxic effects in cardiac muscle was studied following an intravenous administration of a decavanadate solution (1mM total vanadium) in Sparus aurata. Although decameric vanadate is unstable in the assay medium, it decomposes with a half-life time of 16 allowing studying its effects not only in vitro but also in vivo. After 1, 6 and 12h upon decavanadate administration the increase of vanadium in blood plasma, red blood cells and in cardiac mitochondria and cytosol is not affected in comparison to the administration of a metavanadate solution containing labile oxovanadates. Cardiac tissue lipid peroxidation increases up to 20%, 1, 6 and 12h after metavanadate administration, whilst for decavanadate no effects were observed except 1h after treatment (+20%). Metavanadate administration clearly differs from decavanadate by enhancing, 12h after exposure, mitochondrial superoxide dismutase (SOD) activity (+115%) and not affecting catalase (CAT) activity whereas decavanadate increases SOD activity by 20% and decreases (-55%) mitochondrial CAT activity. At early times of exposure, 1 and 6h, the only effect observed upon decavanadate administration was the increase by 20% of SOD activity. In conclusion, decavanadate has a different response pattern of lipid peroxidation and oxidative stress markers, in spite of the same vanadium distribution in cardiac cells observed after decavanadate and metavanadate administration. It is suggested that once formed decameric vanadate species has a different reactivity than vanadate, thus, pointing out that the differential contribution of vanadium oligomers should be taken into account to rationalize in vivo vanadate toxicity.

Decameric vanadate species (V10) inhibit the rate and the extent of G-actin polymerization with an IC50 of 68+/-22 microM and 17+/-2 microM, respectively, whilst they induce F-actin depolymerization at a lower extent. On contrary, no effect on actin polymerization and depolymerization was detected for 2mM concentration of "metavanadate" solution that contains ortho and metavanadate species, as observed by combining kinetic with (51)V NMR spectroscopy studies. Although at 25 degrees C, decameric vanadate (10 microM) is unstable in the assay medium, and decomposes following a first-order kinetic, in the presence of G-actin (up to 8 microM), the half-life increases 5-fold (from 5 to 27 h). However, the addition of ATP (0.2mM) in the medium not only prevents the inhibition of G-actin polymerization by V10 but it also decreases the half-life of decomposition of decameric vanadate species from 27 to 10h. Decameric vanadate is also stabilized by the sarcoplasmic reticulum vesicles, which raise the half-life time from 5 to 18h whereas no effects were observed in the presence of phosphatidylcholine liposomes, myosin or G-actin alone. It is proposed that the "decavanadate" interaction with G-actin, favored by the G-actin polymerization, stabilizes decameric vanadate species and induces inhibition of G-actin polymerization. Decameric vanadate stabilization by cytoskeletal and transmembrane proteins can account, at least in part, for decavanadate toxicity reported in the evaluation of vanadium (V) effects in biological systems.

In this work we present a kinetic study of the superoxide-mediated electron transfer reactions between rubredoxin-type proteins and members of the three different classes of superoxide reductases (SORs). SORs from the sulfate-reducing bacteria Desulfovibrio vulgaris (Dv) and D. gigas (Dg) were chosen as prototypes of classes I and II, respectively, while SOR from the syphilis spirochete Treponema pallidum (Tp) was representative of class III. Our results show evidence for different behaviors of SORs toward electron acceptance, with a trend to specificity for the electron donor and acceptor from the same organism. Comparison of the different kapp values, 176.9+/-25.0 min(-1) in the case of the Tp/Tp electron transfer, 31.8+/-3.6 min(-1) for the Dg/Dg electron transfer, and 6.9+/-1.3 min(-1) for Dv/Dv, could suggest an adaptation of the superoxide-mediated electron transfer efficiency to various environmental conditions. We also demonstrate that, in Dg, another iron-sulfur protein, a desulforedoxin, is able to transfer electrons to SOR more efficiently than rubredoxin, with a kapp value of 108.8+/-12.0 min(-1), and was then assigned as the potential physiological electron donor in this organism.

Superoxide reductase (SOR) is a metalloprotein containing a non-heme iron centre, responsible for the scavenging of superoxide radicals in the cell. The crystal structure of Treponema pallidum (Tp) SOR was determined using soft X-rays and synchrotron radiation. Crystals of the oxidized form were obtained using poly(ethylene glycol) and MgCl2 and diffracted beyond 1.55 A resolution. The overall architecture is very similar to that of other known SORs but TpSOR contains an N-terminal domain in which the desulforedoxin-type Fe centre, found in other SORs, is absent. This domain conserves the beta-barrel topology with an overall arrangement very similar to that of other SOR proteins where the centre is present. The absence of the iron ion and its ligands, however, causes a decrease in the cohesion of the domain and some disorder is observed, particularly in the region where the metal would be harboured. The C-terminal domain exhibits the characteristic immunoglobulin-like fold and harbours the Fe(His)4(Cys) active site. The five ligands of the iron centre are well conserved despite some disorder observed for one of the four molecules in the asymmetric unit. The participation of a glutamate as the sixth ligand of some of the iron centres in Pyrococcus furiosus SOR was not observed in TpSOR. A possible explanation is that either X-ray photoreduction occurred or there was a mixture of redox states at the start of data collection. In agreement with earlier proposals, details in the TpSOR structure also suggest that Lys49 might be involved in attraction of superoxide to the active site.

Superoxide reductase is a 14 kDa metalloprotein containing a catalytic non-haem iron centre [Fe(His)4Cys]. It is involved in defence mechanisms against oxygen toxicity, scavenging superoxide radicals from the cell. The oxidized form of Treponema pallidum superoxide reductase was crystallized in the presence of polyethylene glycol and magnesium chloride. Two crystal forms were obtained depending on the oxidizing agents used after purification: crystals grown in the presence of K3Fe(CN)6 belonged to space group P2(1) (unit-cell parameters a = 60.3, b = 59.9, c = 64.8 A, beta = 106.9 degrees) and diffracted beyond 1.60 A resolution, while crystals grown in the presence of Na2IrCl6 belonged to space group C2 (a = 119.4, b = 60.1, c = 65.6 A, beta = 104.9 degrees) and diffracted beyond 1.55 A. A highly redundant X-ray diffraction data set from the C2 crystal form collected on a copper rotating-anode generator (lambda = 1.542 A) clearly defined the positions of the four Fe atoms present in the asymmetric unit by SAD methods. A MAD experiment at the iron absorption edge confirmed the positions of the previously determined iron sites and provided better phases for model building and refinement. Molecular replacement using the P2(1) data set was successful using a preliminary trace as a search model. A similar arrangement of the four protein molecules could be observed.

The authors report the case of a 39-year-old male patient who had an ischemic stroke (complete infarction of right anterior cerebral circulation) and an acute myocardial infarction during the same year. Molecular study revealed he was homozygous for the 677C-->T mutation in the gene coding for methylenetetrahydrofolate reductase, a key enzyme of folate metabolism; deficiency of this enzyme is associated with increased cardiovascular risk and neurological lesions. Some considerations are put forward about hyperhomocysteinemia and the MTHFR 677C-->T mutation as cardiovascular risk factors.

Among the biotargets interacting with vanadium is the calcium pump from the sarcoplasmic reticulum (SR). To this end, initial research efforts were launched with two vanadium(V)-citrate complexes, namely (NH(4))(6)[V(2)O(4)(C(6)H(4)O(7))(2)].6H(2)O and (NH(4))(6)[V(2)O(2)(O(2))(2)(C(6)H(4)O(7))(2)].4H(2)O, potentially capable of interacting with the SR calcium pump by combining kinetic studies with (51)V NMR spectroscopy. Upon dissolution in the reaction medium (concentration range: 4-0.5mM), both vanadium(V):citrate (VC) and peroxovanadium(V):citrate (PVC) complexes are partially converted into vanadate oligomers. A 1mM solution of the PVC complex, containing 184microM of the PVC complex, 94microM oxoperoxovanadium(V) (PV) species, 222microM monomeric (V1), 43microM dimeric (V2) and 53microM tetrameric (V4) species, inhibits Ca(2+) accumulation by 75 %, whereas a solution of the VC complex of the same vanadium concentration, containing 98microM of the VC complex, 263microM monomeric (V1), 64microM dimeric (V2) and 92microM tetrameric (V4) species inhibits the calcium pump activity by 33 %. In contrast, a 1 mM metavanadate solution, containing 460microM monomeric (V1), 90.2microM dimeric (V2) and 80microM tetrameric (V4) species, has no effect on Ca(2+) accumulation. The NMR signals from the VC complex (-548.0ppm), PVC complex (-551.5ppm) and PV (-611.1ppm) are broadened upon SR vesicle addition (2.5mg/ml total protein). The relative order for the half width line broadening of the NMR signals, which reflect the interaction with the protein, was found to be V4>PVC>VC>PV>V2=V1=1, with no effect observed for the V1 and V2 signals. Putting it all together the effects of two vanadium(V)-citrate complexes on the modulation of calcium accumulation and ATP hydrolysis by the SR calcium pump reflected the observed variable reactivity into the nature of key species forming upon dissolution of the title complexes in the reaction media.

The aldehyde oxidoreductase from Desulfovibrio gigas belongs to the family of molybdenum hydroxylases. Besides a molybdenum cofactor which constitutes their active site, these enzymes contain two [2Fe-2S](2+,1+) clusters which are believed to transfer the electrons provided by the substrate to an acceptor which is either a FAD group or an electron-transferring protein. When the three metal centers of D. gigas AOR are simultaneously paramagnetic, splittings due to intercenter spin-spin interactions are visible when the EPR spectra are recorded at low temperatures. By studying quantitatively these interactions with a model based on the X-ray crystal structure, which takes into consideration the interactions between the magnetic moments carried by all the metal sites of the system, it is possible to determine the location of the reducible sites of the [2Fe-2S] clusters. When combined with the electron-transfer pathways proposed on the basis of the X-ray crystal structure, the results provide a detailed description of the electron-transfer system of D. gigas AOR.

n/a

Superoxide reductases are a class of non-haem iron enzymes which catalyse the monovalent reduction of the superoxide anion O2- into hydrogen peroxide and water. Treponema pallidum (Tp), the syphilis spirochete, expresses the gene for a superoxide reductase called neelaredoxin, having the iron protein rubredoxin as the putative electron donor necessary to complete the catalytic cycle. In this work, we present the first cloning, overexpression in Escherichia coli and purification of the Tp rubredoxin. Spectroscopic characterization of this 6 kDa protein allowed us to calculate the molar absorption coefficient of the 490 nm feature of ferric iron, epsilon=6.9+/-0.4 mM(-1) cm(-1). Moreover, the midpoint potential of Tp rubredoxin, determined using a glassy carbon electrode, was -76+/-5 mV. Reduced rubredoxin can be efficiently reoxidized upon addition of Na(2)IrCl(6)-oxidized neelaredoxin, in agreement with a direct electron transfer between the two proteins, with a stoichiometry of the electron transfer reaction of one molecule of oxidized rubredoxin per one molecule of neelaredoxin. In addition, in presence of a steady-state concentration of superoxide anion, the physiological substrate of neelaredoxin, reoxidation of rubredoxin was also observed in presence of catalytic amounts of superoxide reductase, and the rate of rubredoxin reoxidation was shown to be proportional to the concentration of neelaredoxin, in agreement with a bimolecular reaction, with a calculated k(app)=180 min(-1). Interestingly, similar experiments performed with a rubredoxin from the sulfate-reducing bacteria Desulfovibrio vulgaris resulted in a much lower value of k(app)=4.5 min(-1). Altogether, these results demonstrated the existence for a superoxide-mediated electron transfer between rubredoxin and neelaredoxin and confirmed the physiological character of this electron transfer reaction.

Recently reported decameric vanadate (V(10)) high affinity binding site in myosin S1, suggests that it can be used as a tool in the muscle contraction regulation. In the present article, it is shown that V(10) species induces myosin S1 cleavage, upon irradiation, at the 23 and 74 kDa sites, the latter being prevented by actin and the former blocked by the presence of ATP. Identical cleavage patterns were found for meta- and decavanadate solutions, indicating that V(10) and tetrameric vanadate (V(4)) have the same binding sites in myosin S1. Concentrations as low as 50 muM decavanadate (5 muM V(10) species) induces 30% of protein cleavage, whereas 500 muM metavanadate is needed to attain the same extent of cleavage. After irradiation, V(10) species is rapidly decomposed, upon protein addition, forming vanadyl (V(4+)) species during the process. It was also observed by NMR line broadening experiments that, V(10) competes with V(4) for the myosin S1 binding sites, having a higher affinity. In addition, V(4) interaction with myosin S1 is highly affected by the products release during ATP hydrolysis in the presence or absence of actin, whereas V(10) appears to be affected at a much lower extent. From these results it is proposed that the binding of vanadate oligomers to myosin S1 at the phosphate loop (23 kDa site) is probably the cause of the actin stimulated myosin ATPase inhibition by the prevention of ATP/ADP exchange, and that this interaction is favoured for higher vanadate anions, such as V(10).

We report the characterization of the molecular properties and EPR studies of a new formate dehydrogenase (FDH) from the sulfate-reducing organism Desulfovibrio alaskensis NCIMB 13491. FDHs are enzymes that catalyze the two-electron oxidation of formate to carbon dioxide in several aerobic and anaerobic organisms. D. alaskensis FDH is a heterodimeric protein with a molecular weight of 126+/-2 kDa composed of two subunits, alpha=93+/-3 kDa and beta=32+/-2 kDa, which contains 6+/-1 Fe/molecule, 0.4+/-0.1 Mo/molecule, 0.3+/-0.1 W/molecule, and 1.3+/-0.1 guanine monophosphate nucleotides. The UV-vis absorption spectrum of D. alaskensis FDH is typical of an iron-sulfur protein with a broad band around 400 nm. Variable-temperature EPR studies performed on reduced samples of D. alaskensis FDH showed the presence of signals associated with the different paramagnetic centers of D. alaskensis FDH. Three rhombic signals having g-values and relaxation behavior characteristic of [4Fe-4S] clusters were observed in the 5-40 K temperature range. Two EPR signals with all the g-values less than two, which accounted for less than 0.1 spin/protein, typical of mononuclear Mo(V) and W(V), respectively, were observed. The signal associated with the W(V) ion has a larger deviation from the free electron g-value, as expected for tungsten in a d(1) configuration, albeit with an unusual relaxation behavior. The EPR parameters of the Mo(V) signal are within the range of values typically found for the slow-type signal observed in several Mo-containing proteins belonging to the xanthine oxidase family of enzymes. Mo(V) resonances are split at temperatures below 50 K by magnetic coupling with one of the Fe/S clusters. The analysis of the inter-center magnetic interaction allowed us to assign the EPR-distinguishable iron-sulfur clusters with those seen in the crystal structure of a homologous enzyme.

Cytochrome c peroxidase (CCP) catalyses the reduction of H2O2 to H2O, an important step in the cellular detoxification process. The crystal structure of the di-heme CCP from Pseudomonas nautica 617 was obtained in two different conformations in a redox state with the electron transfer heme reduced. Form IN, obtained at pH 4.0, does not contain Ca2+ and was refined at 2.2 Angstrom resolution. This inactive form presents a closed conformation where the peroxidatic heme adopts a six-ligand coordination, hindering the peroxidatic reaction from taking place. Form OUT is Ca2+ dependent and was crystallized at pH 5.3 and refined at 2.4 Angstrom resolution. This active form shows an open conformation, with release of the distal histidine (His71) ligand, providing peroxide access to the active site. This is the first time that the active and inactive states are reported for a di-heme peroxidase.

A biosensor for nitrite analytical determination was developed using a cytochrome c nitrite reductase (ccNiR) from Desulfovibrio desufuricans ATCC 27774 immobilized and electrically connected on a glassy carbon electrode by entrapment in an electrogencrated poly(pyrrole-viologen) matrix. The modified bioelectrode was studied by cyclic voltammetry and a catalytic current was observed in presence of nitrite. The linear range of the electrode response was 5.4-43.4 muM. The detection limit and the sensitivity were 5.4 muM and 1721 mA M-1 cm(-2), respectively. The K-M(app) value determined from the Lineweaver-Burk plot was 86 muM. The biosensor fully maintained its electroenzymatic activity towards nitrite after four days.. No catalytic response was observed in the presence of nitrate ions while interference from sulfites was considered negligible. Finally, the biosensor composition was optimized in term of monomer-enzyme ratio. (C) 2004 Elsevier B.V. All rights reserved.

The electronic-vibrational couplings of the two-centre non-heme iron protein Desulfovibrio desulfuricans desulfoferrodoxin (DFx) in three oxidation states, i.e. fully oxidised (grey), half-oxidised (pink), and fully reduced (colourless), have been investigated by variable temperature (VT) UV/VIS, MCD, CD, and EPR spectroscopy. The UV/VIS spectra of grey DFx at room temperature is characterised by broad charge transfer (CT) transitions associated with oxidised centre 1 (495 and 368 nm) and II (335 and 635 nm). The transitions are resolved at 78 K, substantiated by VT-MCD and -CD. The data offer novel information about the electronic-vibrational couplings of the transitions. Multiphonon bandshape analysis discloses strong contributions from both local Fe-S and S-C stretching and solvent/protein modes. A number of transitions are blue- or red-shifted compared with monomeric desulforedoxin, superoxide reductase or dismutase, and cloned Desulfovibrio vulgaris DFx fragments. Conversion from grey to pink DFx is accompanied by drastic electronic-vibrational changes of both centres. The data suggest that electron transfer and optical CT-transitions of DFx are controlled by environmental reorganization in the whole region between the metal centres.

The cytochrome c nitrite reductase is isolated from the membranes of the sulfate-reducing bacterium Desulfovibrio desulfuricans ATCC 27774 as a heterooligomeric complex composed by two subunits (61 kDa and 19 kDa) containing c-type hemes, encoded by the genes nrfA and nrfH, respectively. The extracted complex has in average a 2NrfA:1NrfH composition. The separation of ccNiR subunits from one another is accomplished by gel filtration chromatography in the presence of SDS. The amino-acid sequence and biochemical subunits characterization show that NrfA contains five hemes and NrfH four hemes. These considerations enabled the revision of a vast amount of existing spectroscopic data on the NrfHA complex that was not originally well interpreted due to the lack of knowledge on the heme content and the oligomeric enzyme status. Based on EPR and Mossbauer parameters and their correlation to structural information recently obtained from X-ray crystallography on the NrfA structure [Cunha, C.A., Macieira, S., Dias, J.M., Almeida, M.G., Goncalves, L.M.L., Costa, C., Lampreia, J., Huber, R., Moura, J.J.G., Moura, I. & Romao, M. (2003) J. Biol. Chem. 278, 17455-17465], we propose the full assignment of midpoint reduction potentials values to the individual hemes. NrfA contains the high-spin catalytic site (-80 mV) as well as a quite unusual high reduction potential (+150 mV)/low-spin bis-His coordinated heme, considered to be the site where electrons enter. In addition, the reassessment of the spectroscopic data allowed the first partial spectroscopic characterization of the NrfH subunit. The four NrfH hemes are all in a low-spin state (S = 1/2). One of them has a gmax at 3.55, characteristic of bis-histidinyl iron ligands in a noncoplanar arrangement, and has a positive reduction potential.

The gene encoding cytochrome c nitrite reductase (NrfA) from Desulfovibrio desulfuricans ATCC 27774 was sequenced and the crystal structure of the enzyme was determined to 2.3-A resolution. In comparison with homologous structures, it presents structural differences mainly located at the regions surrounding the putative substrate inlet and product outlet, and includes a well defined second calcium site with octahedral geometry, coordinated to propionates of hemes 3 and 4, and caged by a loop non-existent in the previous structures. The highly negative electrostatic potential in the environment around hemes 3 and 4 suggests that the main role of this calcium ion may not be electrostatic but structural, namely in the stabilization of the conformation of the additional loop that cages it and influences the solvent accessibility of heme 4. The NrfA active site is similar to that of peroxidases with a nearby calcium site at the heme distal side nearly in the same location as occurs in the class II and class III peroxidases. This fact suggests that the calcium ion at the distal side of the active site in the NrfA enzymes may have a similar physiological role to that reported for the peroxidases.

The differences in geometry at the metal centres in the two known [Fe-4S] proteins rubredoxin (Rd) and desulforedoxin (Dx) are postulated to be a result of the different spacing of the C-terminal cysteine pair in the two proteins. In order to address this question, two mutants of Desulfovibrio gigas Dx with modified cysteinyl spacing were prepared and their solution structures have been determined by NMR. Mutant 1 of Dx (DxM1) has a single glycine inserted between the adjacent cysteines (C28 and C29) found in the wild type Dx sequence. Mutant 3 (DxM3) has two amino acid residues, -P-V-, inserted between C28 and C29 in order to mimic the primary sequence found in Rd from Desulfovibrio gigas. The solution structure of DxM1 exists, like wild type Dx, as a dimer in solution although the single glycine inserted between the adjacent cysteines disrupts the stability of the dimer resulting in exchange between a dimer state and a small population of another, probably monomeric, state. For DxM3 the two amino acid residues inserted between the adjacent cysteines results in a monomeric protein that has a global fold near the metal centre very similar to that found in Rd.

Superoxide reductases catalyze the monovalent reduction of superoxide anion to hydrogen peroxide. Spectroscopic evidence for the formation of a dinuclear cyano-bridged adduct after K(3)Fe(CN)(6) oxidation of the superoxide reductases neelaredoxin from Treponema pallidum and desulfoferrodoxin from Desulfovibrio vulgaris was reported. Oxidation with K(3)Fe(CN)(6) reveals a band in the near-IR with lambda(max) at 1020 nm, coupled with an increase of the iron content by almost 2-fold. Fourier transform infrared spectroscopy provided additional evidence with CN-stretching vibrations at 2095, 2025-2030, and 2047 cm(-)(1), assigned to a ferrocyanide adduct of the enzyme. Interestingly, the low-temperature electronic paramagnetic resonance (EPR) spectra of oxidized TpNlr reveal at least three different species indicating structural heterogeneity in the coordination environment of the active site Fe ion. Given the likely 6-coordinate geometry of the active site Fe(3+) ion in the ferrocyanide adduct, we propose that the rhombic EPR species can serve as a model of a hexacoordinate form of the active site.

We report the kinetic behavior of the enzyme aldehyde oxidoreductase (AOR) from the sulfate reducing bacterium Desulfovibrio gigas (Dg) encapsulated in reverse micelles of sodium bis-(2-ethylhexyl) sulfosuccinate in isooctane using benzaldehyde, octaldehyde, and decylaldehyde as substrates. Dg AOR is a 200-kDa homodimeric protein that catalyzes the conversion of aldehydes to carboxylic acids. Ultrasedimentation analysis of Dg AOR-containing micelles showed the presence of 100-kDa molecular weight species, confirming that the Dg AOR subunits can be dissociated. UV-visible spectra of encapsulated Dg AOR are indistinguishable from the enzyme spectrum in solution, suggesting that both protein fold and metal cofactor are kept intact upon encapsulation. The catalytic constant (k(cat)) profile as a function of the micelle size W(0) (W(0)=[H(2)O]/[AOT]) using benzaldehyde as substrate showed two bell-shaped activity peaks at W(0)=20 and 26. Furthermore, enzymatic activity for octaldehyde and decylaldehyde was detected only in reverse micelles. Like for the benzaldehyde kinetics, two peaks with both similar k(cat) values and W(0) positions were obtained. EPR studies using spin-labeled reverse micelles indicated that octaldehyde and benzaldehyde are intercalated in the micelle membrane. This suggests that, though Dg AOR is found in the cytoplasm of bacterial cells, the enzyme may catalyze the reaction of substrates incorporated into a cell membrane.

The enzyme hydrogenase isolated from the sulphate reducing anaerobic bacterium Desulfovibrio gigas was encapsulated in reverse micelles of AOT-water-isooctane. The enzyme ability to consume molecular hydrogen was studied as a function of the micelle size (given by W-o = [H2O]/[organic solvent]). A peak of catalytic activity was obtained for W-o = 18, a micelle size theoretically fitting the heterodimeric hydrogenase molecule. At this W-o value, the recorded catalytic activity was slightly higher than in a buffer system (K-cat = 169.43 s(-1) against the buffer value of 151 s(-1)). The optimal buffer used to encapsulate the enzyme was found to be imidazole 50 mM, pH 9.0, The molecular hydrogen production activity was also tested in this reverse micelle medium. (C) 2002 Elsevier Science Inc. All lights reserved.

Desulfovibrio gigas desulforedoxin (Dx) consists of two identical peptides, each containing one [Fe-4S] center per monomer. Variants with different iron and zinc metal compositions arise when desulforedoxin is produced recombinantly from Escherichia coli. The three forms of the protein, the two homodimers [Fe(III)/Fe(III)]Dx and [Zn(II)/Zn(II)]Dx, and the heterodimer [Fe(III)/Zn(II)]Dx, can be separated by ion exchange chromatography on the basis of their charge differences. Once separated, the desulforedoxins containing iron can be reduced with added dithionite. For NMR studies, different protein samples were prepared labeled with (15)N or (15)N + (13)C. Spectral assignments were determined for [Fe(II)/Fe(II)]Dx and [Fe(II)/Zn(II)]Dx from 3D (15)N TOCSY-HSQC and NOESY-HSQC data, and compared with those reported previously for [Zn(II)/Zn(II)]Dx. Assignments for the (13)C(alpha) shifts were obtained from an HNCA experiment. Comparison of (1)H-(15)N HSQC spectra of [Zn(II)/Zn(II)]Dx, [Fe(II)/Fe(II)]Dx and [Fe(II)/Zn(II)]Dx revealed that the pseudocontact shifts in [Fe(II)/Zn(II)]Dx can be decomposed into inter- and intramonomer components, which, when summed, accurately predict the observed pseudocontact shifts observed for [Fe(II)/Fe(II)]Dx. The degree of linearity observed in the pseudocontact shifts for residues >/=8.5 A from the metal center indicates that the replacement of Fe(II) by Zn(II) produces little or no change in the structure of Dx. The results suggest a general strategy for the analysis of NMR spectra of homo-oligomeric proteins in which a paramagnetic center introduced into a single subunit is used to break the magnetic symmetry and make it possible to obtain distance constraints (both pseudocontact and NOE) between subunits.