Eur J Biochem 269, 1650–1661 (2002) Ó FEBS 2002 The Fe-only nitrogenase and the Mo nitrogenase from Rhodobacter capsulatus A comparative study on the redox properties of the metal clusters present in the dinitrogenase components Stefan Siemann*, Klaus Schneider, Melanie Drottboom and Achim Muller ă ă Lehrstuhl fuăr Anorganische Chemie I, Fakultaăt fuăr Chemie der Universitaăt Bielefeld, Bielefeld, Germany The dinitrogenase component proteins of the conventional Mo nitrogenase (MoFe protein) and of the alternative Fe-only nitrogenase (FeFe protein) were both isolated and purified from Rhodobacter capsulatus, redox-titrated according to the same procedures and subjected to an EPR spectroscopic comparison In the course of an oxidative titration of the MoFe protein (Rc1Mo) three significant S ¼ 1/2 EPR signals deriving from oxidized states of the P-cluster were detected: (1) a rhombic signal (g ¼ 2.07, 1.96 and 1.83), which showed a bell-shaped redox curve with midpoint potentials (Em) of )195 mV (appearance) and )30 mV (disappearance), (2) an axial signal (g|| ¼ 2.00, g^ ¼ 1.90) with almost identical redox properties and (3) a second rhombic signal (g ¼ 2.03, 2.00, 1.90) at higher redox potentials (> 100 mV) While the Ôlow-potentialÕ rhombic signal and the axial signal have been both attributed to the one-electron-oxidized P-cluster (P1+) present in two conformationally different proteins, the Ôhigh-potentialÕ rhombic signal has been suggested rather to derive from the P3+ state Upon oxidation, the FeFe protein (Rc1Fe) exibited three significant S ¼ 1/2 EPR signals as well However, the Rc1Fe signals strongly deviated from the MoFe protein signals, suggesting that they cannot simply be assigned to different P-cluster states (a) The most prominent feature is an unusually broad signal at g ¼ 2.27 and 2.06, which proved Correspondence to A Muller, Lehrstuhl fur Anorganische Chemie I, ă ă Fakultat fur Chemie, Universitat Bielefeld, Postfach 100131, 33501 ă ă ă Bielefeld, Germany Fax: + 49 521 1066003, Tel.: + 49 521 1066153, E-mail: a.mueller@uni-bielefeld.de Abbreviations: nif, nitrogen fixation; vnf, vanadium dependent nitrogen fixation; anf, alternative nitrogen fixation; FeMoco, iron–molybdenum cofactor; FeFeco, iron–iron cofactor; Rc1Mo, MoFe protein of R capsulatus; Rc1Fe, FeFe protein of R capsulatus; Rc2Mo, Fe protein of the Mo nitrogenase of R capsulatus; Rc2Fe, Fe protein of the Fe-only nitrogenase of R capsulatus; EXAFS, extended X-ray absorption fine structure Enzyme: nitrogenase (EC 1.18.6.1) *Present address: Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada Present address: Transferstelle Umweltbiotechnology, Ruhr-Universitat Bochum, 44780 Bochum, Germany ă (Received 19 September 2001, revised 28 December 2001, accepted 22 January 2002) to be fully reversible and to correlate with catalytic activity The cluster giving rise to this signal appears to be involved in the transfer of two electrons The midpoint potentials determined were: )80 mV (appearance) and 70 mV (disappearance) (b) Under weakly acidic conditions (pH 6.4) a slightly altered EPR signal occurred It was characterized by a shift of the g values to 2.22 and 2.05 and by the appearance of an additional negative absorption-shaped peak at g ¼ 1.86 (c) A very narrow rhombic EPR signal at g ¼ 2.00, 1.98 and 1.96 appeared at positive redox potentials (Em ¼ 80 mV, intensity maximum at 160 mV) Another novel S ¼ 1/2 signal at g ¼ 1.96, 1.92 and 1.77 was observed on further, enzymatic reduction of the dithionite-reduced state of Rc1Fe with the dinitrogenase reductase component (Rc2Fe) of the same enzyme system (turnover conditions in the presence of N2 and ATP) When the Rc1Mo protein was treated analogously, neither this Ôturnover signalÕ nor any other S ¼ 1/2 signal were detectable All Rc1Fe-specific EPR signals detected are discussed and tentatively assigned with special consideration of the reference spectra obtained from Rc1Mo preparations Keywords: Fe nitrogenase; FeFe cofactor; FeMo cofactor; P-cluster; EPR spectroscopy Four types of nitrogenase systems have been demonstrated to exist in bacteria and archea so far They have been clearly shown to be genetically as well as biochemically distinct The first genetic nitrogen fixation (nif ) system discovered is responsible for encoding the conventional molybdenum (Mo)-containing nitrogenase Two nitrogenase systems are closely related to the Mo nitrogenase, but Mo-independent One is the vanadium (V)-dependent nitrogen fixation (vnf ) system encoding a nitrogenase which contains V instead of Mo in the cofactor (vanadium nitrogenase) [1–4], whereas the other, represented by the alternative nitrogen fixation (anf ) gene system, encodes a nitrogenase containing neither Mo, V nor any other heterometal atom [4–9], and has therefore been designated as the Fe nitrogenase or Fe-only nitrogenase Recently, a heterotrimeric and completely nif/vnf/anf-independent nitrogenase system has been reported to occur in Streptomyces thermoautotrophicus, in which electrons for N2 reduction are derived from superoxide oxidation coupled to CO oxidation [10] Ó FEBS 2002 Redox properties of the FeFe protein (Eur J Biochem 269) 1651 The characteristics of Mo, V and Fe nitrogenases have been reviewed recently by Eady [3] and Smith [4] All three nitrogenase systems consist of two-component proteins, the dinitrogenase component (MoFe protein, VFe protein, FeFe protein) and the dinitrogenase-reductase component (also termed Fe protein with respect to all three types of nitrogenases) While the MoFe protein consists of four subunits forming an a2b2 tetramer, the dinitrogenase proteins of the Mo-independent, alternative nitrogenases, contain an additional small 13–15 kDa subunit to form an a2b2d2 hexameric structure The dinitrogenase component of nitrogenases contains two types of unique metal clusters, the so-called M-cluster (FeMo cofactor, FeV cofactor, FeFe cofactor), which represents the site of substrate reduction [11], and the P-cluster, whose function is likely to transfer electrons as well as protons to the cofactor [12] Based on X-ray crystal structure analysis of MoFe proteins, the structures of the FeMo cofactor (Fe7MoS9/homocitrate) and the P-cluster (Fe8S7) have been elucidated [12,13], the specific site(s) of substrate binding and reduction within the cofactor, however, still remain a matter of controversial discussion [14–17] So far, only three Fe-only nitrogenases have been genetically (as anf systems) as well as biochemically identified and characterized These are the enzymes of Azotobacter vinelandii [5,6], Rhodospirillum rubrum [9] and Rhodobacter capsulatus [8,18,19], the heterometal-free N2-fixation system from the latter organism being the most intensively studied During the early years of Fe nitrogenase research, doubts were widespread as to whether an Fe-only nitrogenase can be isolated as an intact, functioning enzyme These doubts primarily arose due to the fact that preparations of the type of anf-dependent nitrogenase were, regardless of their origin, generally characterized by either extremely low catalytic activity [5,6,9,18] or the wrong cofactor (namely the FeMo cofactor) incorporated into the alternative dinitrogenase component [6,19] However, a comprehensive characterization of the Fe-only nitrogenase isolated from R capsulatus, which included a detailed comparison with the Mo-containing nitrogenase from the same organism, showed that: (a) the Fe nitrogenase components can indeed be isolated and purified as intact and catalytically active proteins, and (b) that the FeFe protein definitely does not contain an iron–molybdenum cofactor (FeMoco), but a clearly wellfunctioning Fe-only cofactor [8] Relatively high specific activities have been reported for N2 reduction (350 nmol of NH3 formed per per mg protein), acetylene reduction as well as very high activities (1300 nmol H2Ỉmin)1Ỉmg)1 in an N2 atmosphere) for the evolution of molecular hydrogen [8] It is interesting to note that, particularly in the simultaneous presence of a second substrate (N2 or C2H2 in addition to H+), the H2 production rates were distinctly higher than the respective activities of the Mo nitrogenase ( sixfold) Samples of such highly active FeFe protein preparations contained 26 ± Fe atoms per protein molecule, but neither molybdenum nor vanadium [8] A recent 57Fe-Mossbauer-/Fe-EXAFS study on the FeFe ă protein from R capsulatus provided strong evidence that: (a) the FeFe cofactor is diamagnetic in the Na2S2O4reduced state containing 4FeII and 4FeIII centers, and (b) the main structural feature of the FeMoco, the central trigonal prismatic arrangement of Fe atoms, is also present in the FeFe cofactor, thus indicating a structural homology between both cofactor types [20,21] A definite identification of the Fe-only cofactor by EPR is still lacking Nevertheless, based on the results of preceding investigations [8], the FeFe protein exhibited several interesting and, with respect to the MoFe protein, deviating EPR spectroscopic properties (a) Highly active FeFe protein samples (reduced with Na2S2O4) neither showed a FeMocotypical S ¼ 3/2 EPR signal nor any other signal indicative of a S ¼ 3/2 spin system Instead they were, in agreement with the analysis of Mossbauer spectra [21], EPR silent ă (b) A novel S ẳ 1/2 signal (g ¼ 1.96, 1.92, 1.77) appeared on dinitrogenase reductase-mediated reduction of the FeFe protein (turnover conditions) (c) Two further significant EPR signals were observed when the FeFe protein was partially oxidized with K3[Fe(CN)6] or thionine: an unusually broad signal centered at g ¼ 2.27 and 2.06 and a very narrow rhombic signal at g ¼ 2.00, 1.98 and 1.96 A conclusive assignment of these novel EPR signals to either the cofactor or the P-cluster has proven elusive due to the fact that both of these metal clusters present in the Fe-only nitrogenase are diamagnetic in the dithionitereduced state, but probably become EPR-active upon oxidation In the present work we focused on the identification or tentative assignment of the most significant EPR signals detected with FeFe protein samples, by pursuing the following approach: the FeFe and the MoFe proteins were isolated from the same organism, samples were prepared according to the same procedures and subsequently characterized and compared by EPR spectroscopy, particularly with respect to their redox properties MATERIALS AND METHODS Bacterial strains The organisms used were the R capsulatus wild-type strain B10S and the Mo-resistant double mutant with a nifHDK deletion as well as an additional deletion in the modABCD region [19,22] The products of the latter genes are involved in high-affinity molybdenum transport [22] Growth medium and culture conditions The growth medium and culture conditions applied were as described previously [8] Purification of nitrogenase proteins Preparation of cell-free extracts (cell disruption by lysozyme followed by ultracentrifugation) were performed as described by Schneider et al [8] In view of the difficulty in separating the dinitrogenase (Rc1Mo) and dinitrogenase reductase component (Rc2Mo) of the Mo nitrogenase from R capsulatus by DEAE chromatography, we used Q-Sepharose (from Pharmacia), a stronger and more effective anion exchanger, for the purification of both the Fe-only and the Mo nitrogenase components The column (internal diameter: 2.5 cm) containing approximately 60 mL gel, was cooled to °C with a cryostat and equilibrated with Ar-gassed Tris buffer (50 mM, pH 7.8) containing NaCl (150 mM) and sodium dithionite (2 mM) Ó FEBS 2002 1652 S Siemann et al (Eur J Biochem 269) The cell-free extract was loaded onto the Q-Sepharose column, followed by the stepwise elution with approximately 50–60 mL of NaCl solutions (in equilibration buffer) of increasing concentrations (200/250/300/350/400 mM in the case of the Mo nitrogenase and 200/250/280/310/340/ 400 mM in the case of the Fe nitrogenase) The Rc1Mo component was eluted with 300 mM NaCl, whereas Rc2Mo was recovered in the 350 mM NaCl fraction In the case of the Fe nitrogenase the Rc2Fe component was eluted with 280 mM NaCl prior to the recovery of Rc1Fe with 330 mM NaCl All nitrogenase component proteins were concentrated to approximately mL by anaerobic ultrafiltration in a 50-mL chamber equipped with a PM30 Amicon membrane, and subsequently further concentrated to a final volume of  mL in a B15 Amicon chamber Both dinitrogenase components, which were of relevance for the present comparative EPR study (Rc1Mo, Rc1Fe), were, based on SDS/PAGE analysis, 90–95% pure The protocol previously employed to purify the MoFe protein (DEAE chromatography, Sephadex G-150 gelfiltration) [8] led to a homogeneous preparation with significantly lower protein yield Because the EPR spectra of samples obtained from both the Q-Sepharose and the DEAE/gel-filtration procedures were indistinguishable, we preferred the use of the rapid and high-yield one-column method (Q-Sepharose) also for the purification of the MoFe protein in the present study Determination of nitrogenase activity and protein content For the determination of nitrogenase activity the routine assay (C2H2 reduction) was employed [8] Protein was determined according to Beisenherz et al [23] Metal and acid-labile sulfide determinations The quantitative determination of Fe and Mo was achieved by inductively coupled plasma mass spectrometry as reported previously [24] Fe was additionally determined by the bathophenanthrolin method [2] Acid-labile sulfide analysis was performed according to Chen & Mortenson [25] Redox titrations Redox titrations were performed in a modified redox titration cell similar to that described by Dutton [26] The redox potential was measured with a combined platinumAg/AgCl electrode (PT 4800-M5-S7/80; Mettler Toledo, Steinbach, Germany) and the achieved potentials were quoted relative to the standard hydrogen electrode The method involved titrating the protein in Hepes buffer (50 mM, pH 7.4) at 25 °C in the presence of the following mediators (each at 43 lM): 2,6-dichlorophenolindophenol, phenazine methosulfate, thionine, methylene blue, indigo trisulfonate, indigo carmine, resorufin, anthraquinone2-sulfonate, safranin O, benzyl viologen, methyl viologen Prior to the redox titration, the protein sample was subjected to buffer exchange by gel filtration on Sephadex G25 equilibrated with 50 mM Hepes (pH 7.4) containing mM Na2S2O4 (sodium dithionite) It is pertinent to note that the reducing agent was not entirely removed from FeFe protein preparations in view of the lability of the protein even in the presence of only trace amounts of oxygen [8] For the sake of direct comparison, MoFe protein samples were treated under analogous conditions The final sample solution (3 mL) containing 12–14 mg of protein per mL was adjusted to different redox potentials by the stepwise addition (0.5 lL) of K3[Fe(CN)6] (ferricyanide) as oxidant and Na2S2O4 as reductant After equilibration, which was usually achieved after 1–2 min, 170-lL samples were withdrawn from the solution with a gas-tight syringe, placed in an EPR tube and immediately frozen in liquid N2 for EPR spectroscopic measurements EPR measurements EPR (X band) spectra were recorded on a Bruker ECS 106 spectrometer equipped with an ECS 041 MR Bruker microwave bridge and an Oxford Instruments EPR 900 helium flow cryostat All spectra were recorded at a microwave frequency of 9.44 GHz and a field modulation of 1.0 mT at 100 kHz Spin quantification was performed using 10 mM CuSO4/10 mM HCl/2 M NaClO4 as an external standard for integration RESULTS EPR signals from oxidized states of the MoFe protein In recent years EPR spectroscopic properties have been reported for several MoFe proteins, mainly focusing on P-cluster-type signals [27–31] Based on the notion, however, that, dependent on the origin, the purification procedure and the sample quality (specific activity), considerable differences within one class of enzyme may occur, we did not rely on literature data, but attempted the direct experimental comparison of the MoFe and the FeFe protein We therefore isolated and prepared both proteins not only from the same organism (R capsulatus) but also under the same conditions (lysozymatic cell disruption, Q-Sepharose chromatography, EPR sample preparation) For EPR experiments, protein samples were used which displayed approximately maximal specific activities, i.e  250 U (nmol acetylene reducedỈmin)1) per mg of FeFe protein and 1000–1200 mg)1 of MoFe protein (compare [8]) In the course of these studies two experimental routes to obtain different redox states of the dinitrogenase protein were pursued: (a) a rough, stepwise oxidation with K3[Fe(CN)6] and (b) a redox titration, adjusting the protein solution to defined potentials in the presence of redox mediators Stepwise oxidation of the MoFe protein In its Na2S2O4reduced state the R capsulatus MoFe protein (Rc1Mo) only exhibited the characteristic S ¼ 3/2 EPR signal at g ¼ 4.29, 3.67 and 2.01, arising from the cofactor (compare Fig 6B, which presents a signal-comparison of the dithionitereduced and the turnover state of Rc1Mo) In the same redox state the P-cluster was EPR-silent (PN state) Upon oxidation two significant types of P-cluster signals appeared When samples (pH 7.4), reduced with mM dithionite, were supplemented with successively increasing amounts of K3[Fe(CN)6], a rhombic S ¼ 1/2 EPR signal at g ¼ 2.07, 1.96 and 1.83 appeared (Fig 1, spectrum 1) This signal was Ó FEBS 2002 Redox properties of the FeFe protein (Eur J Biochem 269) 1653 Fig Redox titration of the cofactor and P-cluster EPR signals deriving from the MoFe protein of R capsulatus The redox titration was performed as described in Materials and methods The sample contained 12 mg MoFe protein per mL (d) Redox titration curve of the FeMo cofactor signal For the determination of relative signal intensity the resonance at g ¼ 3.67 was used Spectra were measured at K and 20 mW (j) Redox titration curve of the rhombic P-cluster (P1+ state) signal Intensity determination was performed using the g ¼ 1.96 resonance Spectra were recorded at 16 K and 100 mW rise to this signal cannot be oxidized further under the conditions applied In studies with Av1Mo a similar signal, although much broader and shifted to distinctly higher fields (g ¼ 1.97, 1.88, 1.68), has been observed and attributed to the P-cluster in its 3e– oxidized state [27] Fig P-cluster EPR signals of the MoFe protein compared to the EPR signals detected with the oxidized FeFe protein The MoFe protein sample contained 21 mg protein per mL, 1.9 (± 0.2) Mo atoms and 27 (± 3) Fe atoms per molecule; the FeFe protein sample contained 18 mg protein per mL and 29 (± 3) Fe atoms per molecule Both samples were prepared in 50 mM Tris (pH 7.4) containing mM Na2S2O4 Spectrum 1, MoFe protein, oxidation with mM K3[Fe(CN)6], measured at 16 K; spectrum 2, MoFe protein, oxidation with mM K3[Fe(CN)6], measured at 16 K; spectrum 3, FeFe protein, oxidation with 2.5 mM K3[Fe(CN)6], measured at 10 K; spectrum 4, FeFe protein, oxidation with 2.5 mM K3[Fe(CN)6], measured at 23 K All spectra were recorded at 100 mW most prominent with mM K3[Fe(CN)6] and decreased again above this concentration With respect to its shape and position of the g values, this signal appears to correspond to the S ¼ 1/2 signal that has been reported for the partially oxidized MoFe proteins from Klebsiella pneumoniae (Kp1) and A vinelandii (Av1Mo) [28,30,32] This type of signal has been interpreted to arise from the 1e– oxidized P-cluster (P1+) [28] After the occurrence of an almost EPR-silent intermediate redox state (spectrum not shown), a second rhombic, but much more narrow EPR signal at g ¼ 2.03, 2.00 and 1.90 appeared upon further oxidation (Fig 1, spectrum 2) This signal reached an intensity maximum with mM K3[Fe(CN)6] and remained unchanged with higher oxidant concentrations This result suggests that the cluster giving Equilibrium-mediated redox titration of the MoFe protein The EPR spectroscopic investigation of Rc1Mo samples (in 50 mM Hepes buffer, pH 7.4), which were subjected to a redox titration in the presence of mediators (see Materials and methods), yielded in parts agreeing, in other parts, however, somewhat differing spectral data In accordance with studies on MoFe proteins from other organisms (e.g [27]), a midpoint potential (Em) of )50 mV was determined for the S ¼ 3/2 FeMoco signal of Rc1Mo (Fig 2) Above +100 mV the FeMoco signal disappeared completely The EPR signal originating from the 1e– oxidized P-cluster with the central g value at 1.96 (in the following designated as ÔP1+ signalÕ) appeared at  )250 mV, reached an intensity maximum at )120 mV and decreased again with increasing potentials The bell-shaped redox curve of the P1+ signal thus confirms the involvement of the P-cluster in the transfer of at least two electrons (compare [27,30]) The midpoint potentials determined were: )195 mV (Em for appearance of the signal representing the PN/1+ transition) and )30 mV (Em for disappearance; P1+/2+ transition) In contrast to the pronounced pH dependence of the P1+ signal caused by the partially oxidized Av1Mo protein [30] (see Discussion), the Rc1Mo-induced P1+ signal was not significantly influenced by the pH value The intensity was almost identical at pH 6.4 and 7.4 and was still 60% (with respect to peak height) at pH 8.4 It was, however, a surprise that, in the course of the redox titration and pH dependence studies, a new axial S ¼ 1/2 signal in the g ¼ region 1654 S Siemann et al (Eur J Biochem 269) Ó FEBS 2002 mediators (at pH 7.4), although with much lower intensity (data not shown) The rhombic signal at g ¼ 2.03, 2.00 and 1.90, which appeared prominently after oxidation with K3[Fe(CN)6] (> mM) and was proposed to represent the P3+ state (Fig 1, spectrum 4), was only noticeable as a very weak signal during redox titration (at potentials > 100 mV) Even excessive amounts of K3[Fe(CN)6] did not cause a significant increase in signal intensity It is interesting to note that S ¼ 5/2 signals, observed in the case of Av1Mo and attributed to the P1+ state [28], as well as S ¼ 7/2 signals (P3+ state) [27] both simultaneously present with S ¼ 1/2 signals (forming so-called spin mixtures), were not detected in the case of the Rhodobacter enzyme At potentials > mV an additional weak signal near g ¼ 12 was detected (spectrum not shown) In the case of Av1Mo this low field signal has been attributed to the 2e–-oxidized P-cluster (S ¼ 3) [27,30] An exact determination of the midpoint potential was, however, not possible due to the low intensity of this signal (integer spin system) under standard EPR conditions (perpendicular mode) The two characteristic S ¼ 1/2 signals of the partially oxidized FeFe protein Fig pH-dependent occurrence of the axial EPR signal (g|| ¼ 2.00, g^ ¼ 1.90) resulting from the partially oxidized MoFe protein Two samples of the redox titration, both of the potential region where the rhombic P1+ signal shows maximal intensity ()120 to )90 mV), were thawed and adjusted to pH 6.4 and 8.4, respectively, with a concentrated three-component buffer system (0.87 M Bistris, 0.44 M Hepps, 0.44 M Ches) according to [37] The spectrum of the pH 7.4 sample represents the original spectrum All spectra were recorded at 16 K and 100 mW Inset: difference spectrum (spectrum pH 8.4 ) spectrum pH 6.4) depicting the axial signal (g|| ¼ 2.00, g^ ¼ 1.90) was detected, which showed a distinctly stronger but, referred to the P1+ signal of Av1Mo, opposite pH dependence (Fig 3) The intensity of this signal was maximal at pH 8.4, with no significant change up to pH 9.0 At pH 7.4 the signal intensity accounted for approximately 40% and at pH < 6.5 the signal was absent The profile of the entire signal, without interference of the rhombic signal, was obtained by subtracting the pH 6.4spectrum from the pH 8.4-spectrum (see the inset of Fig 3) The axial signal and the rhombic P1+ signal differed significantly with respect to temperature and microwavepower dependency The P1+ signal was most pronounced around 18 K, the axial signal around 13 K While the P1+ signal appeared to be slightly power saturated already above 25 mW, the axial signal remained unsaturated even at 200 mW However, both signals behaved similarly with respect to their dependence on the redox potential This observation indicates that the axial signal might arise from the P1+ cluster as well, possibly in a slightly modified environment (protein conformation) It is pertinent to note that this axial signal is also detectable in the spectrum obtained after partial oxidation with K3[Fe(CN)6] without Stepwise oxidation of the FeFe protein The protein preparations used in this study contained 29 (± 3) Fe and 31 (± 4) acid-labile sulfur atoms The high Fe/S content indicates that these FeFe protein (Rc1Fe) preparations were virtually devoid of any significant amounts of inactive (oxidatively damaged clusters) or incompletely assembled (vacant cofactor sites) enzyme It is interesting to note that in the case of dithionite-reduced VFe proteins [3,33] and also in some instances with MoFe proteins [27,34] both such protein forms gave rise to S ¼ 1/2 signals In sharp contrast, the Rc1Fe protein is, in agreement with the preceding report [8], apparently EPR silent in the presence of excess dithionite Neither an S ¼ 3/2 nor a significant S ¼ 1/2 signal in the g ¼ region (< 0.05 spins/Rc1Fe molecule) was detectable Recent Mossbauer studies conă rmed that both the FeFe cofactor and the P-cluster are diamagnetic in the dithionite-reduced state and that the cofactor contains four FeII- and four FeIII-centers [21] For the analogous, dithionite-reduced state of the FeMocofactor, the presence of four FeII but only three FeIII centers in addition to the MoIV center has been postulated [35] Thus, the FeFe-cofactor may be (formally) regarded as a FeMo-cofactor molecule in which molybdenum has been replaced by an FeIII center [21] When the FeFe protein was oxidized with K3[Fe(CN)6], in a stepwise fashion similar to that described for the MoFeprotein, several novel EPR signals were detected The two most prominent signals (both S ¼ 1/2) have already been partially characterized [8] One of these is a very narrow rhombic signal at g ¼ 2.00, 1.98 and 1.96 (in the following designated as g ¼ 1.98 signal) and the other, a characteristic broad signal with an absorption-shaped peak at g ¼ 2.27 and a derivative-shaped feature at g ¼ 2.06 (in the following termed g ¼ 2.27 signal) The two signals are depicted in spectra and of Fig and directly compared to the most characteristic S ¼ 1/2 signals of the reference system (the oxidized MoFe protein), that have been attributed to P1+ Ó FEBS 2002 Redox properties of the FeFe protein (Eur J Biochem 269) 1655 and P3+ (Fig 1, spectra and 2) It is interesting to note that the two proteins exhibited a quite different signal pattern When the FeFe protein solution, which contained mM dithionite, was oxidized with 2.5 mM ferricyanide, both types of signals were detectable They could, however, easily be differentiated by their temperature dependence At 10 K, the g ¼ 2.27 signal showed maximal intensity but was partially overlapped by the narrow g ¼ 1.98 signal and a further sharp resonance peak at g ¼ 2.01 (Fig 1, spectrum 3) With increasing temperature this latter peak as well as the broad g ¼ 2.27 feature disappeared, whereas the narrow signal became more prominent At 23 K, the g ¼ 1.98 signal was observed as an undisturbed resonance of a single paramagnetic species (Fig 1, spectrum 4) The most interesting feature is undoubtedly the unusually broad g ¼ 2.27 signal, which was reproducibly detectable and correlated with catalytic activity The samples with highest activities displayed the most intense signals Furthermore, partially oxidized samples exhibiting this signal were found to be catalytically intact (no loss of activity upon oxidation) These results provide conclusive evidence that the g ¼ 2.27 signal is not an artifact As regards the nature of the signal, the lack of a visible negative absorptionshaped peak at higher magnetic fields appears to be, at first glance, indicative of an axial signal However, a closer Fig pH-dependent shift of the broad g ¼ 2.27 EPR signal under slightly acidic conditions A FeFe protein sample (9 mgỈmL)1), freshly prepared in the presence of mM Na2S2O4, was oxidized with 10 mM thionine (Serva, Heidelberg, Germany) by adding 0.6 mL of the protein solution to 1.5 mg solid thionine, which had been pre-exposed to O2-free argon for 30 After thorough mixing, the resulting solution was aliquoted into three samples of equal volume These were then adjusted to pH 6.4, 7.4 and 8.4, respectively, with a three-component buffer system (see legend of Fig 3) All spectra were recorded at 10 K and 100 mW inspection of the spectrum in Fig reveals that the derivative-shaped resonance at g ¼ 2.06 has approximately the same intensity above and below the baseline, suggesting that the g ¼ 2.27 signal is rhombic The inability to observe the negative absorption-shaped peak may be the consequence of inhomogeneous line broadening (g strain), a phenomenon frequently observed in EPR spectra of metalloproteins [36] In the case of Av1Mo, the typical P1+-cluster signal was only detectable at neutral and weakly acidic pH, but was absent at pH values near 8.0 [30] Because Rc1Fe samples were routinely prepared at pH 7.8, it was of considerable interest to determine the EPR properties also under weakly acidic conditions In fact, EPR spectra of thionine-oxidized samples prepared at pH 8.4, 7.4 and 6.4 (Fig 4) revealed a new signal, which was most pronounced at pH 6.4, but proved to be very similar to the g ¼ 2.27 feature The signal was slightly less broad corresponding to a shift of the g values to 2.22 and 2.05 and displayed an additional broad negative absorptionshaped peak centered at g  1.86 (Fig 4, bottom spectrum) Both the g ¼ 2.27 and the narrower rhombic signal (in the following termed g ¼ 2.22 signal) exhibited an identical behaviour with respect to redox treatment as well as to temperature and microwave power dependency (data not shown) The g ¼ 2.27 signal was most prominent at pH 8.4, whereas at pH 7.4, a superimposition of both, the predominant g ¼ 2.27 and the minor g ¼ 2.22 signal was observed (Fig 4, middle spectrum) While the resonance peaks at g ¼ 2.05 and 2.06 fused to form a common, broad, unresolved peak at that pH value, the broad, negative absorption-shaped resonance was weak, but easily detectable The similarity between the two, pH-differentiated signals indicates that the g ¼ 2.22 signal is not novel, but rather arises from the same cluster as the g ¼ 2.27 signal It is important to note that the occurrence of the negative absorption-shaped peak near g ¼ 1.86 in the pH 6.4 spectrum (gz component of the g ¼ 2.22 signal) provides further support to our view that the broader g ¼ 2.27 signal is rhombic but that the resonance in the g ¼ 1.8 region might be too broad to be detectable Because the g ¼ 2.22 signal is narrower than the g ¼ 2.27 signal, the g strain may not completely obscure the negative absorption-shaped peak, thus rendering it detectable as a broad resonance in the spectrum The pH-dependent occurrence of the two EPR signals is due either to subtle conformational differences in the cluster environment or to protonation/deprotonation effects in the cluster itself pH-based signal shifts and the occurrence of additional signals have been reported for the cofactor of the MoFe proteins from K pneumoniae (Kp1) [38], Xanthobacter autotrophicus (Xa1) [39] and even for the isolated FeMoco from Av1Mo [40] The strong resonance near g ¼ 2.00, present in all spectra of Fig 4, originates from the thionine radical signal Under the oxidation conditions (10 mM thionine) applied in these experiments, the narrow g ¼ 1.98 signal was absent or of such low intensity that it was completely obscured by the radical signal Analogous experiments on pH-dependence with samples oxidized with K3[Fe(CN)6] revealed that the narrow signal was not significantly influenced by the pHvalue (data not shown) For the clarity of presentation of the broad-type signals, we chose the spectra of the Ó FEBS 2002 1656 S Siemann et al (Eur J Biochem 269) thionine-oxidized samples, because these showed more intense signals with high reproducibility Quantification of the g ¼ 2.27 signal yielded  0.6 spins per protein molecule, corresponding to 0.3 spins per P-cluster and per FeFe cofactor, respectively This relatively low spin content may be due to the parallel existence of the same cluster species in different oxidation states, of which only one is EPR-detectable Thus, the theoretical maximum of signal intensity (accounting for two spins per protein molecule) may not be obtained It is interesting to note that this interpretation is in full accordance with the results obtained from redox titration experiments, which will be discussed in a later section In addition, the quantification of the presumably rhombic g ¼ 2.27 signal is solely based on the g ¼ 2.27, 2.06 peaks and did not include the putative third peak in the g ¼ 1.8 region, thus leading to an underestimation of the spin content The rhombic signal at g ¼ 2.22, measured at pH 6.4, was not quantified because of its generally lower intensity as well as the decreased stability of the protein at this pH value The narrow g ¼ 1.98 signal integrated to only 0.25 spins per protein molecule, indicating that the corresponding cluster, at least in this specific redox state, is not of catalytic relevance Equilibrium-mediated redoxtitration of the FeFe protein A redox titration (at pH 7.4) of the broad g ¼ 2.27 signal in the presence of mediators resulted in a bell-shaped titration curve with midpoint potentials of )80 mV (appearance) and +70 mV (disappearance) Maximal signal intensity was achieved by adjusting the potential to )5 mV (Fig 5) These results imply that the cluster giving rise to the g ¼ 2.27 signal is, in analogy to the P1+ cluster of the MoFe protein, involved in a 2e– transfer process It could be reversibly converted into an EPR-silent state either by reduction or by further oxidation Fig Redox titration of EPR signals deriving from oxidized states of the FeFe protein The redox titration was performed as described in Materials and methods The sample contained 14 mg FeFe protein per mL (j) Redox titration curve of the signal at g ¼ 2.27 and 2.06 For determination of the relative signal intensity the peak at g ¼ 2.27 was used Spectra were recorded at 10 K and 100 mW (d) Redox titration curve of the signal at g ¼ 2.00,1.98 and 1.96 Signal determination was performed with the resonance at g ¼ 1.96 Spectra were measured at 23 K and 100 mW The narrow g ¼ 1.98 signal was observed in a region shifted about 150 mV to more positive redox potentials For this signal a midpoint potential of 80 mV (appearance) was determined (Fig 5) Maximal signal intensity was reached at approximately 160 mV To avoid oxidative damage of the protein higher potentials than 220 mV were not adjusted Nevertheless, the cluster giving rise to this type of signal could only be re-reduced to 20–30% by dithionite Nature of the g ¼ 2.01 signal Upon oxidation with ferricyanide, a signal at g ¼ 2.01, located between the broad g ¼ 2.27 and the narrow g ¼ 1.98 signal (Figure 1, spectrum 3), was detected at low temperatures (10–14 K) and potentials above )50 mV, and was therefore never observed as a complete and undisturbed signal The intensity of this signal strongly varied from preparation to preparation A decrease in specific activity was always accompanied by an increase in signal intensity, suggesting that the g ¼ 2.01 signal is an artifact arising from an oxidatively damaged cluster This view was substantiated by the following observations (a) The increase in intensity of the g ¼ 2.01 signal during oxidation resulted in a concomitant increase of the g ¼ 4.3 feature (data not shown) The g ¼ 4.3 signal, an accompanying signal found with most FeS proteins, has been characterized as an S ¼ 5/2 system caused by nonfunctional ÔadventitiousÕ FeIII [41], which often occurs as the result of destruction of FeS clusters In the case of Rc1Fe, the g ¼ 4.3 signal significantly increased towards the end of the redox titration (+220 mV) (b) Preliminary studies on the isolation and purification of the FeFe apoprotein (from a nifBB’strain) revealed that the cofactorless protein, when prepared according to the procedure approved for the native enzyme [8], cannot be obtained in an intact hexameric, but only in a tetrameric a2b2 form The small d subunit could be isolated by DEAE chromatography as a separate peptide (D Tiemann, S Fuchs, K Schneider & A Muller, ă unpublished results) Dissociation of the d subunit from the apodinitrogenase under certain conditions (e.g during gel filtration) has also been reported in the case of the vanadium nitrogenase (VFe protein) from A vinelandii [42] The tetrameric FeFe apoprotein from R capsulatus did not show any EPR signal typical of a P-cluster signal Only a signal at g ¼ 2.01, rather reminiscent of an [Fe3S4]1+ cluster, was detected This signal increased dramatically during oxidation (data not shown) Simultaneously, the signal intensity of nonfunctional ferric ions increased as well Because the apoprotein only contains P-clusters, it is evident that the g ¼ 2.01 signal stems from a P-cluster fragment as the result of oxidative damage The phenomenon of oxidative conversion of [Fe4S4] clusters to [Fe3S4] clusters and thus, the occurrence of g ¼ 2.01 signals, is widespread among iron–sulfur proteins (e.g [43]) It is conceivable that the apoprotein tetramer, lacking both the FeFe cofactor and the d subunit, is a highly unstable protein form, in which the P-clusters, if not becoming completely destroyed, tend to convert into three-iron clusters Ó FEBS 2002 Redox properties of the FeFe protein (Eur J Biochem 269) 1657 Fig EPR signals of the FeFe protein and the MoFe protein under turnover conditions (A) EPR spectra of both the FeFe protein (spectrum 1) and the MoFe protein (spectrum 2) measured at 16 K under turnover conditions The samples were prepared anaerobically (under N2) directly in the EPR tube They contained 24 mg Rc1Fe per mL and 0.6 mg Rc2Fe per mL in the case of the Fe nitrogenase and 28 mg Rc1Mo per mL and 0.7 mg Rc2Mo per mL in the case of the Mo nitrogenase The other constituents were: 100 mM Hepes (pH 7.8), mM ATP, 10 mM MgCl2, mM Na2S2O4, 20 mM creatine phosphate and 0.2 mg creatine kinase The enzymatic reduction was started by the addition of ATP After incubation at room temperature, the samples were immediately frozen in isopentane cooled by liquid nitrogen The spectra were recorded at 100 mW (B) EPR signals of the MoFe protein measured at K, either Na2S2O4-reduced (spectrum 3) or enzyme(Rc2Mo)-reduced, i.e under turnover conditions (spectrum 4) The Na2S2O4-reduced sample contained 28 mg Rc1Mo per mL but no Rc2Mo All other conditions were equal to those described for the turnover samples Both spectra were recorded at 20 mW Comparison of the turnover signals from enzymereduced states of the MoFe protein and the FeFe protein The characteristic EPR signal of the FeFe protein (Rc1Fe) under turnover conditions at g ¼ 1.96, 1.92 and 1.77 has already been documented [8] The occurrence of this S ¼ 1/2 type signal with samples containing ATP, a substrate (N2, C2H2, H+) and the dinitrogenase reductase component for enzymatic reduction of the FeFe protein, was confirmed in the present study (Fig 6A, spectrum 1) The signal was most prominent in the presence of the natural substrate, N2, when measured at 16 K and 100 mW In order to avoid (a) P-cluster oxidation (see Discussion section on this topic) and (b) an interference of the turnover signal and the dinitrogenase reductase (Rc2Fe) signal, a molar Rc2Fe:Rc1Fe ratio of : 10 was chosen Under these catalytically suboptimal conditions (low electron flux) a spin content of  0.4 spins per Rc1Fe molecule has been determined for the turnover signal A control spectrum of a sample containing the same amount of Rc2Fe ( 0.6 mgỈmL)1) in the absence of Rc1Fe protein, confirmed that the intensity of the Rc2Fe signal was marginal under the measuring conditions The intention of the present turnover experiments was to include the Mo nitrogenase of the same organism in order to minimize the possibility that the described turnover signal results from a Rhodobacter-specific contamination or from an artifact caused by the preparation conditions applied In fact, when we isolated and prepared the protein components of the Mo nitrogenase (Rc1Mo, Rc2Mo) according to the same procedures as the Fe nitrogenase components (Rc1Fe, Rc2Fe) and finally applied exactly identical turnover and EPR conditions, the EPR signal detected with the Fe-only nitrogenase (g ¼ 1.96, 1.92, 1.77) at 16 K, was absent (Fig 6A, spectrum 2) The minimal resonance (g ¼ 1.9–2.0) visible in spectrum resulted from Rc2Mo It was identical with the control spectrum in the absence of Rc1Mo Furthermore, with the exception of the classical S ¼ 3/2 signal of the MoFe protein at lower temperatures (< 12 K), no other signal was detectable In full accordance with literature data (e.g [44]), the enzyme(Rc2Mo)-reduced state of the MoFe protein showed, compared to the dithionite-reduced state (Fig 6B, spectrum 3), a drastic decrease ( 70%) in signal intensity (Fig 6B, spectrum 4) In the case of Mo nitrogenases from other organisms, this behaviour has been interpreted to be due to one-electron reduction of the semireduced to the reduced and EPR silent state of the FeMo cofactor [45] DISCUSSION The FeMo cofactor, one of the two unique metal clusters present in the MoFe protein component of the conventional nitrogenase, has been subjected to extensive EPR spectroscopic investigations since the early 1970s (reviewed in [45]) On the other hand, studies on EPR and redox properties of the P-cluster have long been neglected Only in recent years have EPR investigations on MoFe proteins focused on this second unusual type of nitrogenase cluster [27–31] In view of the lack of analogous studies on the cofactor and the P-cluster of the MoFe protein from R capsulatus (Rc1Mo), as well as our intention to use the Rc1Mo protein as a reference system for the characterization and identification of the EPR signals displayed by the FeFe protein component of the Fe-only nitrogenase (Rc1Fe), investigations on the MoFe protein were included in this comparative work Several results obtained with the Rc1Mo component are in excellent agreement with important EPR and redox properties previously reported for the MoFe proteins of other bacteria These include: demonstration of the classical Ó FEBS 2002 1658 S Siemann et al (Eur J Biochem 269) Table A comparative overview on the EPR signals of the MoFe- and FeFe protein from R capsulatus The enzyme (dinitrogenase reductase)reduced MoFe protein molecules (E1 state) are EPR-silent However, under the conditions used, i.e at low electron flux (Rc2Fe:Rc1Fe ¼ : 10), the sample also contained  30% of dithionite-reduced MoFe protein molecules (Ôresting stateÕ E0) showing the typical S ¼ 3/2 FeMoco signal EPR signals (g values) Redox state FeFe protein MoFe protein I Enzyme-reduced II Na2S2O4-reduced III Oxidized Emax  )125 mV Emax  )150 mV Emax  )5 mV 1.96, 1.92, 1.77 EPR-silent EPR-silent 4.29, 3.67, 2.01 E > 100 mV (1) 2.07, 1.96, 1.83 (pH 7.4) (2) 2.00, 1.90 (pH 8.4) (1) 2.27, 2.06 (pH 8.4) (2) 2.22, 2.05, 1.86 (pH 6.4) 2.00, 1.98, 1.96 S ¼ 3/2 FeMoco signal (Em  )50 mV), of a characteristic rhombic S ¼ 1/2 EPR signal at g ¼ 2.07, 1.96, 1.83, apparently deriving from the one-electron-oxidized P-cluster (P1+) and of a weak signal near g ¼ 12 (propably S ¼ 3), reminiscent of the 2e – oxidized P-cluster (P2+) signal [27,30] However, we have found that Rc1Mo significantly deviates from other MoFe proteins with respect to a number of relevant characteristics (1) Although the lineshape and g value positions (2.07, 1.96, 1.83) of the EPR signals from Rc1Mo and Av1Mo, which have been interpreted to represent the P1+ redox state, are almost identical [30], some features associated with these P1+ signals exhibit remarkable differences (a) The midpoint potential of the PN/1+ redox couple of Av1Mo is about 100 mV more negative (Em ¼ )290 mV) (b) The midpoint potential of the P1+/2+ redox couple of Av1Mo displays a pronounced pH dependence, being distinctly more positive at lower pH values This pH dependence has been interpreted as an indication of a coupled electron and proton transfer [30] Despite increased midpoint potentials in an acidic environment, at pH 6.0 Em of P1+/2+ is still drastically lower ( )150 mV) than Em for the analogous state of the Rc1Mo P-cluster (c) The intensity of the rhombic S ¼ 1/2 P1+ signal of Av1Mo is strongly pH-dependent as well, being maximal at pH 6.0 At pH 7.5 this signal is of only very low intensity and at pH 8.0 it is even absent This observation might explain why the characteristic P1+ cluster signal has not been detected by some research groups [27] The absence of the P1+ state in a weakly alkaline medium is likely to be caused by the simultaneous transfer of two electrons, thereby resulting in a transition from PN directly into the P2+ state In the case of the rhombic P1+ signal of Rc1Mo (g ¼ 2.07, 1.96, 1.83), which corresponds to the Av1Mo P1+ signal, neither the intensity nor the dependence on the redox potential was significantly influenced by the pH value (2) In the course of the redox titration of Rc1Mo, a novel axial S ¼ 1/2 signal (g|| ¼ 2.00, g^ ¼ 1.90) was detected, which has not yet been described for other MoFe proteins This signal, which appears to be associated with the P1+ redox state as well, was, in contrast to the rhombic P1+ signal, influenced by the pH value However, compared to Av1Mo, opposite pH-dependence was observed (maximal intensity of this axial signal at pH 8.4) (3) At positive redox potentials, the Av1Mo protein showed a rhombic S ¼ 1/2 EPR signal (g ¼ 1.97, 1.88, 2.03, 2.00, 1.90 1.68; maximal intensity at 150–300 mV) which has been attributed to the 3e– oxidized P-cluster (P3+) [27] A rhombic signal in the same potential range has also been observed in the case of Rc1Mo, however, this signal was less broad and located in a lower magnetic field region (g ¼ 2.03, 2.00, 1.90) Moreover, the Rc1Mo signal showed significant intensity only after nonmediated oxidation and did not, in contrast to the Av1Mo signal, disappear upon further oxidation with K3[Fe(CN)6] (> 300 mV) (4) The phenomenon of Ôspin mixturesÕ, i.e the simultaneous occurrence of S ¼ 1/2, 5/2 signals (P1+ state) [28] and of S ¼ 1/2, 7/2 signals (P3+ state) [27], was not observed with the Rhodobacter enzyme The observation that Pclusters of one and the same redox state may be present in different spin states within one protein sample or even within one MoFe protein molecule, has been discussed to be due to an artifact caused by temperature and solvent influence [46] With respect to the assignment of specific EPR spectroscopic characteristics of the FeFe protein (overview in Table 1) to certain redox states of the P-cluster and the FeFe cofactor, only two conclusions can be drawn at the present time (1) The native, dithionite-reduced Rc1Fe protein was proven to be EPR-silent This is in accordance with the results of recent Mossbauer studies [21], which indicaă ted that both the iron–iron cofactor (FeFeco) and the P-cluster are diamagnetic (S ¼ 0) in the dithionite-reduced state (2) The S ¼ 1/2 signal at g ¼ 1.96, 1.92, 1.77, obtained after dinitrogenase reductase (Rc2Fe)-mediated reduction of Rc1Fe (turnover conditions), represents a reduced state of the FeFe cofactor This conclusion is based on the following arguments: (a) The fact that the g ¼ 1.96, 1.92, 1.77 signal is absent in the turnover sample of the Mo nitrogenase provides evidence that this signal is not an artifact or caused by a Rhodobacter-specific paramagnetic impurity (b) Spectroscopic studies (Mossbauer, Integer-spină EPR) on MoFe proteins have revealed that all iron atoms in the dithionite-reduced P-cluster are most likely in the ferrous state [47,48] This excludes the possibility that the diamagnetic, fully reduced P-cluster becomes further reduced during enzyme turnover Thus, the turnover signal of the FeFe protein cannot arise from a ‘super-reduced’ P-cluster Ó FEBS 2002 Redox properties of the FeFe protein (Eur J Biochem 269) 1659 (c) The P-cluster has been shown to be intermediately oxidized during enzyme turnover under conditions of high electron flux, i.e at a dinitrogenase reductase:dinitrogenase ratio of : or higher [49] However, under conditions of low electron flux, as employed in our study, the P-cluster remains reduced and the predominant states of the cofactor are E0 (resting state, equivalent with the semireduced or dithionite-reduced state) and E1 (one-electronreduced state) [50] Further reduced states, such as E3 and E4, postulated to play a role in N2-binding/reduction and suggested to be connected to P-cluster oxidation, are presumably present only in very small proportions, undetectable by EPR under these conditions These theoretical expectations are in full accordance with the results obtained with the reference system used in this study, the Mo nitrogenase from R capsulatus Upon enzymatic reduction of the MoFe protein at a Rc2Mo/Rc1Mo ratio of : 10 the FeMoco EPR signal (representing the E0 state) maintained 30% of its intensity In addition, no signal indicative of an oxidized P-cluster was detected Under analogous conditions, the Fe-only nitrogenase failed to exhibit any of the signals appearing upon oxidative titration Consequently, the turnover signal does not arise from an oxidized P-cluster (d) As concluded from the results of the preceding Mossbauer study [21], the FeFe cofactor of the Fe-only ă nitrogenase contains an equal number (four each) of ferric and ferrous iron centers in the dithionite-reduced state, resulting in a diamagnetic (S ¼ 0) state Hence, it appears reasonable to assume that the FeFeco becomes converted into an EPR detectable state by 1e– oxidation as well as by further 1e– reduction According to the mechanism postulated by Lowe and coworkers [49], the turnover signal can therefore be assigned to the one-electron-reduced state (E1) of the FeFe cofactor The two relevant S ¼ 1/2 signals displayed by the oxidized FeFe protein (the narrow g ¼ 2.00, 1.98, 1.96 and the broad g ¼ 2.27, 2.06 signal) are more difficult to assign At first glance, there appears to be no similarity between P-cluster signals of the MoFe protein and the EPR signals of the FeFe protein However, some significant EPR spectroscopic data (compare Table 1) indicate that the narrow g ¼ 98 signal represents the P3+-cluster state: (a) Although this signal differs from the P3+ signal (MoFe protein) with respect to lineshape (signal broadness), the g region at which the two signals are detectable, is in principle the same (1.90–2.03) (b) The redox potentials at which the narrow FeFe protein signal occurs (Em ¼ 80 mV) and reaches maximal intensity ( 160 mV), are in excellent agreement with the values reported for the P3+-cluster signal of the MoFe protein from A vinelandii [27] (c) After being induced by oxidation with ferricyanide, both clusters giving rise to this type of signal can only partially (20–30%) be re-reduced by the addition of excessive amounts of dithionite Although a three-electron-oxidized P-cluster can apparently be produced by chemical oxidation, the irreversibility of this in vitro process as well as the low spin content of the corresponding EPR signal indicate that the P3+ state is not of physiological/catalytical relevance The assignment of the novel broad g ¼ 2.27 feature of the partially oxidized FeFe protein appears to be even more challenging In fact, such a signal has never been observed for any type of FeS cluster The characteristic g values of all known S ¼ 1/2 systems arising from homonuclear FeS clusters are situated between 1.8 and 2.15 [51] Several fundamental considerations oppose the attribution of this signal to an oxidized P-cluster: (a) The signal profile and the position of g values fundamentally deviate from that of the P1+ signal (g ¼ 2.06, 1.96, 1.83) of the Rhodobacter MoFe protein Furthermore, the redox potential of the cluster responsible for the g ¼ 2.27 feature is, compared to the P1+ signal of Rc1Mo, shifted by 100–115 mV to more positive potentials (compare Table and the course of both bell-shaped redox curves in Figs and 5) A P1+ cluster as the species responsible for a signal with such strongly deviating characteristics would imply either a significant alteration in the protein environment (conformation, interaction with amino acids) compared to the P-cluster environment in the MoFe protein, or a structural modification of the cluster itself At least the latter possibility seems highly unlikely, since the six cysteine residues coordinating the P-cluster in the MoFe protein, are also conserved in the FeFe protein, as judged by sequence comparisons [7] In addition, Mossbauer spectra did not yield any indication ă for a structural difference between the P-clusters of the two proteins [21] (b) The possibility that the g ¼ 2.27 signal represents P2+ appears to be highly unlikely as well The 2e–-oxidized P-cluster of MoFe proteins has been reported not to reveal an S ¼ 1/2 signal, but to exhibit a signal at g  12, resulting from an integer spin state (presumably S ¼ [27]) Although this feature has been demonstrated in the case of Rc1Mo as well, a corresponding signal for the Rc1Fe protein was not detected Under the EPR spectroscopic conditions employed in this study (perpendicular mode), the 2e– oxidized P-cluster of Rc1Fe is EPR-silent (c) If our conclusion is correct that the narrow g ¼ 1.98 signal arises indeed from the 3e–oxidized P-cluster, the possibility that the broad g ¼ 2.27 signal represents P1+ (or P2+) can automatically be excluded in view of the following interrelations: redox re-titrations were performed starting from a potential of  200 mV At this redox potential, the broad low potential g ¼ 2.27 signal was absent and the narrow high potential g ¼ 1.98 signal reached an intensity maximum If both signals were to originate from the P-cluster, the majority of P-cluster molecules would be present in the P3+ state at this redox potential Because the P-cluster in this state cannot be reversibly reduced to lower oxidation states, the broad signal would, provided it represents P1+, necessarily not reappear with significant intensity during the re-titration procedure However, as demonstrated in our experiments, the broad signal did reappear upon reductive titration, even with maximal intensity This controversial behaviour of the two signals with respect to redox treatment and reversibility shows that both the narrow and the broad signal originate from different paramagnetic species In conclusion, if the g ¼ 1.98 signal stems from the P3+ cluster, then the g ¼ 2.27 feature derives from the FeFe cofactor On the other hand, if the broad signal, despite its unusual properties, represents the P1+ cluster, the narrow signal, in turn, would arise from the FeFeco However, in view of Ó FEBS 2002 1660 S Siemann et al (Eur J Biochem 269) the high potential region at which this narrow signal occurs, this possibility appears very unrealistic In summary, if the g ¼ 2.27 signal does not arise from P1+ or P2+, the only alternative is that it results from the FeFe cofactor Because evidence has been provided by EPR and Mossbauer spectroscopy that the FeFeco is diamagă netic in the dithionite-reduced state, it appears plausible that the g ¼ 2.27 signal might represent the one-electron-oxidized state of the cofactor The observation of a bell-shaped redox curve (Fig 4) implies that the cluster giving rise to this signal can undergo further oxidation This seems to be somewhat surprising, since the EPR-silent 1e–-oxidized FeMoco cannot be further oxidized to another EPR detectable state Although such a redox state appears unlikely to participate in the catalytic events leading to substrate reduction, its formation in the case of the Fe-only nitrogenase might be explained by the involvement of the additional iron atom (which replaces the heterometal atom) in that redox process As regards the lineshape and width of the g ¼ 2.27 signal, its unusual broadness may arise from Heisenberg broadening as a consequence of an unusually short relaxation time Another possible cause for the signal broadness may be due to the transition of one (or more) of the Fe centers in the FeFe cofactor from high-spin to lowspin Such transition, although truly speculative at this time, would lead to an increased g-anisotropy and thus, increased line width It is pertinent to note that the molecular/structural basis underlying the broadening of the g ¼ 2.27 signal is unknown at present Future Mossbauer ă studies on the oxidized FeFe protein may prove valuable in solving this question Provided our assignment of the g ¼ 2.27 signal to the partially oxidized FeFeco and the g ¼ 1.98 signal to the 3e–oxidized P3+ cluster is correct, the question arises as to why an additional signal corresponding to P1+ is missing It is conceivable that the P-cluster transfers two electrons simultaneously (direct PN to P2+ transition), thereby obscuring the P1+ transition As already outlined above, such a 2e– transfer, responsible for the absence of a characteristic P1+ EPR-signal, has been postulated to occur in the case of the Av1Mo protein as well, however, only at weakly alkaline pH [27,30] In conclusion, the assignment of the characteristic EPR signals of the FeFe protein is tentative Future studies will therefore aim at a conclusive identification of the EPR signals, particularly of those arising from the oxidized states of the P-cluster and the FeFeco Such investigations will include the development of a procedure for the isolation and stabilization of a hexameric FeFe apoprotein, as well as a detailed EPR spectroscopic characterization of this cofactorless protein system 10 11 12 13 14 15 16 17 ACKNOWLEDGEMENTS The authors are indebted to Mrs S Selsemeier-Voigt for technical assistance This work was supported by the Deutsche Forschungsgemeinschaft (DFG) 18 REFERENCES 19 Hales, B.J., Case, E.E., Morningstar, J.E., Dzeda, M.F & Mauterer, L.A (1986) Isolation of a new vanadium-containing nitrogenase from Azotobacter vinelandii Biochemistry 25, 7251– 7255 Eady, R.R., Robson, R.L., Richardson, T.H., Miller, R.W & Hawkins, M (1987) The vanadium nitrogenase of Azotobacter chroococcum Biochem J 244, 197–207 Eady, R.R (1996) Structure–function relationships of alternative nitrogenases Chem Rev 96, 3013–3030 Smith, B.E (1999) Structure, function, and biosynthesis of the metallosulfur clusters in nitrogenases Adv Inorg Chem 47, 159–218 Chisnell, J.R., Premakumar, R & Bishop, P.E (1988) Purification of a second alternative nitrogenase from a nifHDK deletion strain of Azotobacter vinelandii J Bacteriol 170, 27–33 Pau, R.N., Eldridge, M.E., Lowe, D.J., Mitchenall, L.A & Eady, R.R (1993) Molybdenum-independent nitrogenases of Azotobacter vinelandii: a functional species of alternative nitrogenase-3 isolated from a molybdenum-tolerant strain contains an ironmolybdenum cofactor Biochem J 293, 101–107 Schuddekopf, K., Hennecke, S., Liese, U., Kutsche, M & Klipp, ă W (1993) Characterization of anf genes specific for the alternative nitrogenase and identification of nif genes required for both nitrogenases in Rhodobacter capsulatus Mol Microbiol 8, 673– 684 Schneider, K., Gollan, U., Drottboom, M., Selsemeier-Voigt, S & ă Muller, A (1997) Comparative biochemical characterization of ă the iron-only nitrogenase and the molybdenum nitrogenase from Rhodobacter capsulatus Eur J Biochem 244, 789–800 Davis, R., Lehmann, L., Petrovich, R., Shah, V.K., Roberts, G.P & Ludden, P.W (1996) Purification and characterization of the alternative nitrogenase from the photosynthetic bacterium Rhodospirillum rubrum J Bacteriol 178, 1445–1450 Ribbe, M., Gadkari, D & Meyer, O (1997) N2 fixation by Streptomyces thermoautotrophicus involves a molybdenum-dinitrogenase and a manganese-superoxide oxidoreductase that couples N2 reduction to the oxidation of superoxide produced from O2 by a molybdenum-CO dehydrogenase J Biol Chem 272, 26627–26633 Hawks, T.R., McLean, P.A & Smith, B.E (1984) Nitrogenase from nifV mutants of Klebsiella pneumoniae contains an altered form of the iron-molybdenum cofactor Biochem J 217, 317–321 Peters, J.W., Stowell, M.H.B., Soltis, S.M., Finnegan, M.G., Johnson, M.K & Rees, D.C (1997) Redox-dependent structural changes in the nitrogenase P-cluster Biochemistry 36, 1181–1187 Howard, J.B & Rees, D.C (1994) Nitrogenase: a nucleotidedependent molecular switch Annu Rev Biochem 63, 235–264 Deng, H & Hoffmann, R (1993) How N2 might be activated by the FeMo-cofactor in nitrogenase Angew Chem Int Ed 32, 1026–1029 Sellmann, D & Sutter, J (1997) Elementary reactions, structurefunction relationships, and the potential relevance of low molecular weight metal-sulfur ligand complexes to biological N2 fixation J Biol Inorg Chem 1, 587–593 Coucouvanis, D (1996) Functional analogs for the reduction of certain nitrogenase substrates Are multiple sites within the Fe/Mo/S center involved in the 6e– reduction of N2? J Biol Inorg Chem 1, 594–600 Gronberg, K.L.C., Gormal, C.A., Durrant, M.C., Smith, B.E & ¨ Henderson, R.A (1998) Why R-homocitrate is essential to the reactivity of FeMo-cofactor of nitrogenase Studies on NifV– extracted FeMo-cofactor J Am Chem Soc 120, 10613–10621 Schneider, K., Muller, A., Schramm, U & Klipp, W (1991) ă Demonsration of a molybdenum- and vanadium-independent nitrogenase in a nifHDK-deletion mutant of Rhodobacter capsulatus Eur J Biochem 195, 653–661 Gollan, U., Schneider, K., Muller, A., Schuddekopf, K & Klipp, ă ă W (1993) Detection of the in vivo incorporation of a metal cluster into a protein The FeMo cofactor is inserted into the FeFe protein Ó FEBS 2002 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Redox properties of the FeFe protein (Eur J Biochem 269) 1661 of the alternative nitrogenase of Rhodobacter capsulatus Eur J Biochem 215, 25–35 Krahn, E., Weiss, B.J.R., Krockel, M., Cramer, S.P., Trautwein, ă A.X., Schneider, K & Muller, A (1998) The Fe-only nitrogenase ă from Rhodobacter capsulatus: The FeFe protein metal centers probed by EXAFS and Mossbauer spectroscopy In Biological ă Nitrogen Fixation for the 21st Century (Elmerich, C., Kondorosi, A & Newton, W.E., eds), pp 59–60 Kluwer Academic Publishers, Dordrecht, the Netherlands Krahn, E., Weiss, B.J.R., Krockel, M., Groppe, J., Henkel, G., ă Cramer, S.P., Trautwein, A.X., Schneider, K & Muller, A (2002) ă The Fe-only nitrogenase from Rhodobacter capsulatus: identification of the cofactor, an unusual, high-nuclearity iron-sulfur cluster, by Fe K-edge EXAFS and 57Fe Mossbauer spectroscopy ¨ J Biol Inorg Chem 7, 37–45 Wang, G., Angermuller, S & Klipp, W (1993) Characterization ă of Rhodobacter capsulatus genes encoding a molybdenum transport system and putative molybdenum-pterin-binding proteins J Bacteriol 175, 3031–3042 Beisenherz, G., Bolze, H.J., Bucher, T., Czok, R., Garbade, K.H., ă Meyer-Arendt, E & Peiderer, G (1953) Diphosphofructosealdolase, Phosphoglyceraldehyd-dehydrogenase, Milchsaureă dehydrogenase, Glycerophosphat-dehydrogenase and Pyruvatkinase aus Kaninchenmuskulatur in einem Arbeitsgang Z Naturforsch 8b, 555–577 Siemann, S., Schneider, K., Behrens, K., Knochel, A., Klipp, W & ă Muller, A (2001) FeMo cofactor biosynthesis in a nifE– mutant of ¨ Rhodobacter capsulatus Eur J Biochem 268, 1940–1952 Chen, J.S & Mortenson, L.E (1977) Inhibition of methylene blue formation during determination of the acid-labile sulfide of ironsulfur protein samples containing dithionite Anal Biochem 79, 157–165 Dutton, P.L (1978) Redox potentiometry: determination of midpoint potentials of oxidation-reduction components of biological electron-transfer systems Methods Enzymol 54, 411–435 Pierik, A.J., Wassink, H., Haaker, H & Hagen, W.R (1993) Redox properties and EPR spectroscopy of the P clusters of Azotobacter vinelandii MoFe protein Eur J Biochem 212, 51–61 Tittsworth, R.C & Hales, B.J (1993) Detection of EPR signals assigned to the 1-equivalent-oxidized P-clusters of the nitrogenase MoFe protein from Azotobacter vinelandii J Am Chem Soc 115, 9763–9767 Tittsworth, R.C & Hales, B.J (1996) Oxidative titration of the nitrogenase VFe protein from Azozobacter vinelandii: an example of redox-gated electron flow Biochemistry 35, 479–487 Lanzilotta, W.N., Christiansen, J., Dean, D.R & Seefeldt, L.C (1998) Evidence for coupled electron and proton transfer in the [8Fe)7S] cluster of nitrogenase Biochemistry 37, 11376–11384 Chan, J.M., Dean, D.R & Seefeldt, L.C (1999) Spectroscopic evidence for changes in the redox state of the nitrogenase P cluster during turnover Biochemistry 38, 5779–5785 Smith, B.E., Lowe, D.J., Chen, G.-X., O’Donell, M.J & Hawkes, T.R (1983) Evidence on intramolecular electron transfer in the MoFe protein of nitrogenase from Klebsiella pneumoniae from rapid-freeze electron-paramagnic-resonance studies of its oxidation by ferricyanide Biochem J 209, 207–213 Blanchard, C.Z & Hales, B.J (1996) Isolation of two forms of the nitrogenase VFe protein from Azotobacter vinelandii Biochemistry 35, 472–478 Zumft, W.G & Mortenson, L.E (1973) Evidence for a catalyticcentre heterogeneity of molybdoferredoxin from Clostridium pasteurianum Eur J Biochem 35, 401–409 Yoo, S.J., Angove, H.C., Papaefthymiou, V., Burgess, B.K & Munck, E (2000) Mossbauer study of the MoFe protein of ¨ ¨ 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 nitrogenase from Azotobacter vinelandii using selective 57Fe enrichment of the M centers J Am Chem Soc 122, 4926–4936 Hagen, W.R (1989) g-strain: inhomogeneous broadening in metalloprotein EPR In Advanced EPR Applications in Biology and Biochemistry (Hoff, A.J., ed.), pp 785–812 Elsevier, Amsterdam, the Netherlands Pham, D.N & Burgess, B.K (1993) Nitrogenase activity Effects of pH on substrate reduction and CO inhibition Biochemistry 32, 13725–13731 Lowe, D.J & Smith, B.E (1985) Electron-paramagneticresonance spectroscopy and related techniques in the study of nitrogenase Biochem Soc Trans 13, 579–581 Schneider, K., Muller, A., Krahn, E., Hagen, W.R., Wassink, H ă & Knuttel, K.-H (1995) The molybdenum nitrogenase from ă wild-type Xanthobacter autotrophicus exhibits properties reminiscent of alternative nitrogenases Eur J Biochem 230, 666–675 Newton, W.E., Gheller, S.F., Feldman, B.J., Dunham, W.R & Schultz, F.A (1989) Isolated iron-molybdenum cofactor of nitrogenase exists in multiple forms in its oxidized and semireduced states J Biol Chem 264, 1924–1927 Palmer, G (1985) The electron paramagnetic resonance of metalloproteins Biochem Soc Trans 13, 548–560 Chatterjee, R., Ludden, P.W & Shah, V.K (1997) Characterization of VNFG, the d subunit of the vnf-encoded apodinitrogenase from Azotobacter vinelandii Implications for its role in the formation of functional dinitrogenase J Biol Chem 272, 3758–3765 Zaborosch, C., Koster, M., Bill, E., Schneider, K & Schlegel, ¨ H.G & Trautwein, A X (1995) EPR and Mossbauer spectroă scopic studies on the tetrameric, NAD-linked hydrogenase of Nocardia opaca 1b and its two dimers: The bd-dimer – a prototype of a simple hydrogenase Biometals 8, 149–162 Smith, B.E., Lowe, D.J & Bray, R.C (1973) Studies by electron paramagnetic resonance on the catalytic mechanism of nitrogenase of Klebsiella pneumoniae Biochem J 135, 331–341 Burgess, B.K (1990) The iron–molybdenum cofactor of nitrogenase Chem Rev 90, 1377–1406 Onate, Y.A., Finnegan, M.G., Hales, B.J & Johnson, M.K (1993) Variable temperature magnetic circular dichroism studies of reduced nitrogenase iron proteins and [4Fe)4S]+ synthetic analog clusters Biochim Biophys Acta 1164, 113–123 Surerus, K.K., Hendrich, M.P., Christie, P.D., Rottgardt, D., Orme-Johnson, W.H & Munck, E (1992) Mossbauer and ă ă integer-spin EPR of the oxidized P-clusters of nitrogenase: Pox is a non-Kramers system with a nearly degenerate ground doublet J Am Chem Soc 114, 8579–8590 Orme-Johnson, W.H (1993) The molybdenum-iron protein of nitrogenase In Molybdenum Enzymes, Cofactors, and Model Systems (Stiefel, E.I., Coucouvanis, D & Newton, W.E., eds), pp 257–270 American Chemical Society, Washington, DC Lowe, D.J., Fisher, K & Thornley, R.N.F (1993) Klebsiella pneumoniae nitrogenase: pre-steady-state absorbance changes show that redox changes occur in the MoFe protein that depend on substrate and component protein ratio; a role for P-centres in reducing dinitrogen? Biochem J 292, 93–98 Fisher, K., Lowe, D.J & Thorneley, R.N.F (1991) Klebsiella pneumoniae nitrogenase The pre-steady kinetics of MoFe protein reduction and hydrogen evolution under conditions of limiting electron flux show that the rates of association with the Fe-protein and electron transfer are independent of the oxidation level of the MoFe protein Biochem J 279, 81–85 Bertini, I., Ciurli, S & Luchinat, C (1995) The electronic structure of FeS centres in proteins and models A contribution to the understanding of their electron transfer properties Struct Bonding 83, 1–54  ... signal The intensity of this signal strongly varied from preparation to preparation A decrease in specific activity was always accompanied by an increase in signal intensity, suggesting that the. .. characterization of the Fe-only nitrogenase isolated from R capsulatus, which included a detailed comparison with the Mo- containing nitrogenase from the same organism, showed that: (a) the Fe nitrogenase. .. performed as described by Schneider et al [8] In view of the difficulty in separating the dinitrogenase (Rc 1Mo) and dinitrogenase reductase component (Rc 2Mo) of the Mo nitrogenase from R capsulatus by