Magnetic circular dichroism evidence for a weakly coupled heme-radical pair at the active site of cytochrome cd1, a nitrite reductase

In nitrite-treated cytochrome cd1 nitrite reductase, heme d1 is electron paramagnetic resonance silent but paramagnetic. Analysis of the unusual temperature dependence of the magnetic circular dichroism spectra unambiguously demonstrates that the heme d1 is not in the oxoferryl (FeIV=O) state but is low-spin FeIII weakly coupled to a radical species. This species could be either a protein-bound radical generated by a nitrite ion reacting with a heme group resulting in a one-electron oxidation of an amino acid residue, possibly tyrosine or tryptophan, adjacent to heme d1, or a heme d1 FeIIINO complex.     

NO reductase activity of the tetraheme cytochrome C554 of Nitrosomonas europaea

The tetraheme cytochrome c(554) (cyt c(554)) from Nitrosomonas europaea is believed to function as an electron-transfer protein from hydroxylamine oxidoreductase (HAO). We show here that cyt c(554) also has significant NO reductase activity. The protein contains one high-spin and three low-spin c-type hemes. HAO catalyzed reduction of the cyt c(554), ligand binding, intermolecular electron transfer, and kinetics of NO reduction by cyt c(554) have been investigated. We detect the formation of a NO-bound ferrous heme species in cyt c(554) by EPR and M?ssbauer spectroscopies during the HAO catalyzed oxidation of hydroxylamine, indicating that N-oxide intermediates produced from HAO readily bind to cyt c(554). In the half-reduced state of cyt c(554), we detect a spin interaction between the [FeNO](7) state of heme 2 and the low-spin ferric state of heme 4. We find that ferrous cyt c(554) will reduce NO at a rate greater than 16 s(-1), which is comparable to rates of other known NO reductases. Carbon monoxide or nitrite are shown not to bind to the reduced protein, and previous results indicate the reactions with O(2) are slow and that a variety of ligands will not bind in the oxidized state. Thus, the enzymatic site is highly selective for NO. The NO reductase activity of cyt c(554) may be important during ammonia oxidation in N. europaea at low oxygen concentrations to detoxify NO produced by reduction of nitrite or incomplete oxidation of hydroxylamine.     

In Rhodobacter sphaeroides respiratory nitrate reductase, the kinetics of substrate binding favors intramolecular electron transfer

The respiratory nitrate reductase (NapAB) from Rb. sphaeroides is a periplasmic molybdenum-containing enzyme which belongs to the DMSO reductase family. We report a study of NapAB by protein film voltammetry (PFV), and we present the first quantitative interpretation of the complex redox-state dependence of activity that has also been observed with other related enzymes. The model we use to fit the data assumes that binding of substrate partly limits turnover and is faster and weaker when the Mo ion is in the V oxidation state than when it is fully reduced. We explain how the presence in the catalytic cycle of such slow chemical steps coupled to electron transfer to the active site decreases the driving force required to reduce the MoV ion and makes exergonic the last intramolecular electron-transfer step (between the proximal cubane and the Mo cofactor). Importantly, comparison is made with all Mo enzymes for which PFV data are available, and we emphasize general features of the energetics of the catalytic cycles in enzymes of the DMSO reductase family.     

Spectroelectrochemical investigation of intramolecular and interfacial electron-transfer rates reveals differences between nitrite reductase at rest and during turnover

A combined fluorescence and electrochemical method is described that is used to simultaneously monitor the type-1 copper oxidation state and the nitrite turnover rate of a nitrite reductase (NiR) from Alcaligenes faecalis S-6. The catalytic activity of NiR is measured electrochemically by exploiting a direct electron transfer to fluorescently labeled enzyme molecules immobilized on modified gold electrodes, whereas the redox state of the type-1 copper site is determined from fluorescence intensity changes caused by Fo?rster resonance energy transfer (FRET) between a fluorophore attached to NiR and its type-1 copper site. The homotrimeric structure of the enzyme is reflected in heterogeneous interfacial electron-transfer kinetics with two monomers having a 25-fold slower kinetics than the third monomer. The intramolecular electron-transfer rate between the type-1 and type-2 copper site changes at high nitrite concentration (≥520 μM), resulting in an inhibition effect at low pH and a catalytic gain in enzyme activity at high pH. We propose that the intramolecular rate is significantly reduced in turnover conditions compared to the enzyme at rest, with an exception at low pH/nitrite conditions. This effect is attributed to slower reduction rate of type-2 copper center due to a rate-limiting protonation step of residues in the enzyme's active site, gating the intramolecular electron transfer.     

Why isn't 'standard' heme good enough for c-type and d1-type cytochromes?

This perspective seeks to discuss why biology often modifies the fundamental iron-protoporphyrin IX moiety that is the very versatile cofactor of many heme proteins. A very common modification is the attachment of this cofactor via covalent bonds to two (or rarely one) sulfur atoms of cysteine residue side chains. This modification results in c-type cytochromes, which have diverse structures and functions. The covalent bonds are made in different ways depending on the cell type. There is little understanding of the reasons for this complexity in assembly routes but proposals for the rationale behind the covalent modification are presented. In contrast to the widespread c-type cytochromes, the d1 heme is restricted to a single enzyme, the cytochrome cd1 nitrite reductase that catalyses the one-electron reduction of nitrite to nitric oxide. This is an extensively derivatised heme; a comparison is drawn with another type of respiratory nitrite reductase in which the active site is a c-type heme, but the product ammonia.     

Direct measurement by laser flash photolysis of intramolecular electron transfer in a two-domain construct of murine inducible nitric oxide synthase

Intersubunit intramolecular electron transfer (IET) from FMN to heme is essential in the delivery of electrons required for O2 activation in the heme domain and the subsequent nitric oxide (NO) synthesis by NO synthase (NOS). Previous crystal structures and functional studies primarily concerned an enzyme conformation that serves as the input state for reduction of FMN by electrons from NADPH and FAD in the reductase domain. To favor formation of the output state for the subsequent IET from FMN to heme in the oxygenase domain, a novel truncated two-domain oxyFMN construct murine inducible nitric oxide synthase (iNOS), in which only the FMN and heme domains were present, was designed and expressed. The kinetics of the IET between the FMN and heme domains in this construct was directly determined using laser flash photolysis of CO dissociation in comparative studies on partially reduced oxyFMN and single domain heme oxygenase constructs.     

Mechanism of the six-electron reduction of nitrite to ammonia by cytochrome c nitrite reductase

Cytochrome c nitrite reductase catalyzes the six-electron reduction of nitrite to ammonia without the release of potential reaction intermediates, such as NO or hydroxylamine. On the basis of the crystallographic observation of reaction intermediates and of density functional calculations, we present a working hypothesis for the reaction mechanism of this multiheme enzyme which carries a novel lysine-coordinated heme group (Fe-Lys). It is proposed that nitrite reduction starts with a heterolytic cleavage of the N-O bond which is facilitated by a pronounced back-bonding interaction of nitrite coordinated through nitrogen to the reduced (Fe(II)) but not the oxidized (Fe(III)) active site iron. This step leads to the formation of an [FeNO](6) species and a water molecule and is further facilitated by a hydrogen bonding network that induces an electronic asymmetry in the nitrite molecule that weakens one N-O bond and strengthens the other. Subsequently, two rapid one-electron reductions lead to an [FeNO](8) form and, by protonation, to an Fe(II)-HNO adduct. Hereafter, hydroxylamine will be formed by a consecutive two-electron two-proton step which is dehydrated in the final two-electron reduction step to give ammonia and an additional water molecule. A single electron reduction of the active site closes the catalytic cycle.     

The ferrous verdoheme-heme oxygenase complex is six-coordinate and low-spin

A biosynthetic and enzymatic method was developed for the preparation of 13C-labeled verdoheme, which permits the 13C NMR spectroscopic characterization of this elusive intermediate in the heme oxidation path catalyzed by the enzyme heme oxygenase. The 13C NMR data indicate that the ferrous verdoheme complex of Neisseria meningitides heme oxygenase is hexacoordinate and diamagnetic, with a proximal histidine and likely a distal hydroxide as axial ligands. The coordination number and spin state of the ferrous verdoheme-heme oxygenase complex is in stark contrast to the pentacoordinate and paramagnetic nature of the heme-heme oxygenase complex and heme centers in general.     

Redox-state-dependent complex formation between pseudoazurin and nitrite reductase

Bacterial copper-containing nitrite reductase catalyzes the reduction of nitrite to nitric oxide as part of the denitrification process. Pseudoazurin interacts with nitrite reductase in a transient fashion to supply the necessary electrons. The redox-state dependence of complex formation between pseudoazurin and nitrite reductase was studied by nuclear magnetic resonance spectroscopy and isothermal titration calorimetry. Binding of pseudoazurin in the reduced state is characterized by the presence of two binding modes, a slow and a fast exchange mode, with a K(d)(app) of 100 microM. In the oxidized state of pseudoazurin, binding occurs in a single fast exchange mode with a similar affinity. Metal-substituted proteins have been used to show that the mode of binding of pseudoazurin is independent of the metal charge of nitrite reductase. Contrary to what was found for other cupredoxins, protonation of the exposed His ligand to the copper of pseudoazurin, His81, does not appear to be involved directly in the dual binding mode of the reduced form. A model assuming the presence of a minor form of pseudoazurin is proposed to explain the behavior of the complex in the reduced state.     

Protein film voltammetry of copper-containing nitrite reductase reveals reversible inactivation

The Cu-containing nitrite reductase from Alcaligenes faecalis S-6 catalyzes the one-electron reduction of nitrite to nitric oxide (NO). Electrons enter the enzyme at the so-called type-1 Cu site and are then transferred internally to the catalytic type-2 Cu site. Protein film voltammetry experiments were carried out to obtain detailed information about the catalytic cycle. The homotrimeric structure of the enzyme is reflected in a distribution of the heterogeneous electron-transfer rates around three main values. Otherwise, the properties and the mode of operation of the enzyme when it is adsorbed as a film on a pyrolytic graphite electrode are essentially unchanged compared to those of the free enzyme in solution. It was established that the reduced type-2 site exists in either an active or an inactive conformation with an interconversion rate of approximately 0.1 s(-1). The random sequential mechanism comprises two routes, one in which the type-2 site is reduced first and subsequently binds nitrite, which is then converted into NO, and another in which the oxidized type-2 site binds nitrite and then accepts an electron to produce NO. At high nitrite concentration, the second route prevails and internal electron transfer is rate-limiting. The midpoint potentials of both sites could be established under catalytic conditions. Binding of nitrite to the type-2 site does not affect the midpoint potential of the type-1 site, thereby excluding cooperativity between the two sites.     

PHEOPHORBIDE a-INDUCED PHOTO-OXIDATION OF CYTOCHROME c: IMPLICATION FOR PHOTODYNAMIC THERAPY

Pheophorbide a-induced photo-oxidation, in vitro, of cytochrome c oxidase and cytochrome c results in irreversible modifications to both protein components. Photo-oxidation of cytochrome c, as exhibited by change in its heme oxidation state, displays exponential kinetics and is detected with a lag period. Both the photo-induced inactivation of the enzyme, and destruction of the substrate ability of cytochrome c occur as complex multi-process events. Under similar experimental conditions, the loss of the substrate capability of cytochrome c develops approximately three times faster than inactivation of the enzyme. The slight lag in the photo-oxidation of cytochrome c is due to pheophorbide a-induced superoxide production. However, the relative amount of photo-oxidant produced is considerably more effective than the cytochrome c reducing capacity of the superoxide. Neither hydroxyl radical nor hydrogen peroxide are involved in the photo-oxidation of the heme function. The possibilities of heme oxidation by a singlet oxygen mediated pathway or direct electron abstraction involving the heme or apoprotein are not excluded. It is proposed that a multi-site oxidation of numerous reduced energy cofactors within cells may augment collateral enzyme inactivation in maximizing photosensitizer-induced cytotoxicity. Accordingly, amphipathic photosensitizers, capable of accessing both lipid and aqueous compartments containing reduced cofactors, may be more effective agents for photodynamic therapy than those which exhibit a high specificity of subcellular localization.     

Reductively activated nitrous oxide reductase reacts directly with substrate

In the terminal step of bacterial denitrification, N2O is converted to N2 at the mu4-sulfide bridged tetranuclear CuZ center of nitrous oxide reductase. The enzyme can be activated by reduced methyl viologen, with up to a 15-fold increase in specific activity. The reductively activated nitrous oxide reductase from Achromobacter cycloclastes was isolated and characterized by visible absorption and EPR spectroscopy, and both methods showed that the CuZ center can attain a [4Cu(I)] oxidation state. When N2O was added to the activated, reductant-free enzyme, distinct spectral changes were observed, indicating that this state of the enzyme interacts with substrate. This was further supported by the detection of 15N-labeled product in the absence of steady-state turnover conditions. A new absorption band around 970 nm appeared following reaction of activated nitrous oxide reductase with N2O, which may represent a catalytic intermediate state of the enzyme.     

Spin Crossover in Nitrito‐Myoglobin as Revealed by Resonance Raman Spectroscopy

The myoglobin (Mb) heme Fe‐O‐N=O and heme Fe‐O‐N=O/2‐nitrovinyl species have been characterized by resonance Raman spectroscopy. In the heme Fe‐O‐N=O species, the bound nitrite ligand is removed by solvent exchange, thus reforming metmyoglobin (metMb). The high‐spin heme Fe‐O‐N=O unit is converted into a low‐spin heme Fe‐O‐N=O/2‐nitrovinyl species that can be reversibly switched between a low‐ and a high‐spin state without removing the bound nitrite ligand, as observed in the case of the heme Fe‐O‐N=O species. This spin‐state change is likely to be accompanied by a general structural rearrangement in the protein‐binding pocket. This example is the first of a globin protein that can reversibly change its metal spin state through an internal perturbation. These findings provide a basis for understanding the structure–function relationship of the spin cross found in other metalloenzymes and Fe^III–porphyrin complexes.     

Oxidation of heme to beta- and delta-biliverdin by Pseudomonas aeruginosa heme oxygenase as a consequence of an unusual seating of the heme

The origin of the unusual regioselectivity of heme oxygenation, i.e. the oxidation of heme to delta-biliverdin (70%) and beta-biliverdin (30%), that is exhibited by heme oxygenase from Pseudomonas aeruginosa (pa-HO) has been studied by (1)H NMR, (13)C NMR, and resonance Raman spectroscopies. Whereas resonance Raman indicates that the heme-iron ligation in pa-HO is homologous to that observed in previously studied alpha-hydroxylating heme oxygenases, the NMR spectroscopic studies suggest that the heme in this enzyme is seated in a manner that is distinct from that observed for all other alpha-hydroxylating heme oxygenase enzymes for which a structure is known. In pa-HO, the heme is rotated in-plane approximately 110 degrees, so the delta-meso-carbon of the major orientational isomer is located within the HO-fold in the place where the alpha-hydroxylating enzymes typically place the alpha-meso-carbon. The unusual heme seating displayed by pa-HO places the heme propionates so that these groups point in the direction of the solvent-exposed heme edge and appears to originate in large part from the absence of stabilizing interactions between the polypeptide and the heme propionates, which are typically found in alpha-hydroxylating heme oxygenase enzymes. These interactions typically involve Lys-16 and Tyr-112, in Neisseriae meningitidis HO, and Lys-16 and Tyr-134, in human and rat HO-1. The corresponding residues in pa-HO are Asn-19 and Phe-117, respectively. In agreement with this hypothesis, we found that the Asn-19 Lys/Phe-117 Tyr double mutant of pa-HO exists as a mixture of molecules exhibiting two distinct heme seatings; one seating is identical to that exhibited by wild-type pa-HO, whereas the alternative seating is very similar to that typical of alpha-hydroxylating heme oxygenase enzymes and is related to the wild-type seating by approximately 110 degrees in-plane rotation of the heme. Furthermore, each of these heme seatings in the pa-HO double mutant gives rise to a subset of two heme isomeric orientations that are related to each other by 180 degrees rotation about the alpha-gamma-meso-axis. The coexistence of these molecules in solution, in the proportions suggested by the corresponding area under the peaks in the (1)H NMR spectrum, explains the unusual regioselectivity of heme oxygenation observed with the double mutant, which we found produces alpha- (55%), delta- (35%), and beta-biliverdin (10%). Alpha-biliverdin is obtained by oxidation of the heme seated similar to that of alpha-hydroxylating enzymes, whereas beta- and delta-biliverdin are formed from the oxidation of heme seated as in wild-type pa-HO.     

Redox properties of wild-type and heme-binding loop mutants of bacterial cytochromes C measured by direct electrochemistry

We have used protein film voltammetry (PFV) to determine the midpoint potentials of the Pseudomonas aeruginosa, Hydrogenobacter thermophilus, and Nitrosomonas europaea wild-type monoheme cytochromes c (cyts c; PA, HT, and NE, respectively), as well as PA N64Q, HT Q64N, and NE V65delta mutants, as a function of pH, and buffer conditions. Recent studies have suggested that the identity of the 64 position of the heme-binding loop (either Asn or Gln) strongly influences the conformation of the Met ligand that binds the heme iron. The PFV studies reveal that HT and NE possess significantly lower potentials (wild-type cyts c having E(m) values of +227 and +250 mV vs SHE) than PA (+290 mV) in 50 mM phosphate buffer, pH 7 at 3 degrees C. The HT Q64N mutant rises in potential compared to wild-type, and the PA N64Q mutant has a lower potential, indicating relationships between Met ligand fluxion, hydrogen bonding to the Met ligand, and redox chemistry. Surprisingly, NE V65delta, possessing a heme binding loop nearly identical to that of the PA protein, displayed an E(m) of +232 mV, even lower than wild-type NE. These data are discussed in terms of models of Met ligand properties and proton dependence.     

Direct immobilization of native yeast iso-1 cytochrome C on bare gold: fast electron relay to redox enzymes and zeptomole protein-film voltammetry

Cyclic voltammetry shows that yeast iso-1-cytochrome c (YCC), chemisorbed on a bare gold electrode via Cys102, exhibits fast, reversible interfacial electron transfer (k(0) = 1.8 x 10(3) s(-1)) and retains its native functionality. Vectorially immobilized YCC relays electrons to yeast cytochrome c peroxidase, and to both cytochrome cd(1) nitrite reductase (NIR) and nitric oxide reductase from Paracoccus denitrificans, thereby revealing the mechanistic properties of these enzymes. On a microelectrode, we measured nitrite turnover by approximately 80 zmol (49 000 molecules) of NIR, coadsorbed on 0.65 amol (390 000 molecules) of YCC.     

Midpoint potentials of hemes a and a3 in the quinol oxidase from Acidianus ambivalens are inverted

The aa3 type B oxygen reductase from the thermophilic archaeon Acidianus ambivalens (QO) was immobilized on silver electrodes and studied by potential-dependent surface-enhanced resonance Raman (SERR) spectroscopy. The immobilized enzyme retains the native structure at the level of the heme pockets and exhibits reversible electrochemistry. From the potential dependence of specific spectral marker bands, the midpoint potentials of hemes a and a3 were unambiguously determined for the first time, being 320 +/- 20 mV for the former and 390 +/- 20 mV for the latter. Both hemes could be treated as independent one-electron Nernstian redox couples, indicating that the interaction potential is smaller than 50 mV. The reversed order of the midpoint potentials compared to those of type A (mitochondrial-like) oxidases, as well as the lack of substantial Coulombic interactions, suggests a different mechanism of electroprotonic energy transduction. In contrast to type A enzymes, a-a3 intraprotein electron transfer in QO is already guaranteed by the order of the midpoint potentials at the onset of enzyme reduction and, therefore, does not require a complex network of cooperativities to ensure exergonicity. In the immobilized state, conformational transitions of the QO a3-CuB active site, which are believed to be essential for proton translocation, are drastically slowed compared to those in solution. We ascribe this finding to the effect of the interfacial electric field, which is of the same order of magnitude as in biological membranes. These results suggest that the membrane potential may play an active role in the regulation of the enzymatic activity of QO.     

Methionine ligand lability of type I cytochromes c: detection of ligand loss using protein film voltammetry

Protein film voltammetry (PFV) is used to interrogate the behavior of a variety of bacterial and mitochondrial His/Met-ligated cytochromes c. While analogous studies upon alkanethiol-modified gold electrodes reveal the anticipated Fe(II/III) couple only, PFV using pyrolytic graphite edge (PGE) electrodes demonstrates the presence of a lower-potential form of each of the cyts c studied, with a potential of approximately -100 mV (vs hydrogen). The generation of the novel, lower-potential state is shown to arise specifically from the interaction with the PGE electrode. Simultaneously, the typical Fe(II/III) couple can be observed. PFV of a series of wild-type cytochromes and mutants in the Met-donating loop show that the lower-potential state is highly similar between proteins from Pseudomonas aeruginosa (PA), Hydrogenobacter thermophilus (HT), and horse heart. The generation of the lower-potential form correlates inversely with the stability of the Met-Fe interaction for each of the cytochromes. Comparison with chemically unfolded cyts c indicates that the lower-potential forms detected here are unique, and this distinct state is ascribed to the loss of the Met ligand. Thus, PGE is demonstrated to be a non-innocent electrode surface in PFV studies of His/Met-ligated cytochromes c.     

Conversion of ammonia to dinitrogen in wastewater by Nitrosomonas europaea

Because Nitrosomonas europaea contains ammonia-oxidizing enzyme, nitrite reductase, and nitrous oxide reductase, the conversion of ammonia to dinitrogen was tried with different reaction conditions. In aerobic reaction conditions, ammonium was converted to nitrite (NO ₂ ⁻ ), while under oxygen-limiting or oxygen-free conditions, NO ₂ ⁻ -N formed from ammonia oxidation by N. europaea was reduced to N₂O and dinitrogen with 22% conversion. During denitrification, optimal pH for the production of N₂O and dinitrogen was found to be 7.0–8.0. Dinitrogen was not produced in acidic pH<7.0. A low partial oxygen pressure as well as oxygen-free conditions are favorable for high production of dinitrogen.     

Alkylamine-ligated H93G myoglobin cavity mutant: a model system for endogenous lysine and terminal amine ligation in heme proteins such as nitrite reductase and cytochrome f

His93Gly sperm whale myoglobin (H93G Mb) has the proximal histidine ligand removed to create a cavity for exogenous ligand binding, providing a remarkably versatile template for the preparation of model heme complexes. The investigation of model heme adducts is an important way to probe the relationship between coordination structure and catalytic function in heme enzymes. In this study, we have successfully generated and spectroscopically characterized the H93G Mb cavity mutant ligated with less common alkylamine ligands (models for Lys or the amine group of N-terminal amino acids) in numerous heme iron states. All complexes have been characterized by electronic absorption and magnetic circular dichroism spectroscopy in comparison with data for parallel imidazole-ligated H93G heme iron moieties. This is the first systematic spectral study of models for alkylamine- or terminal amine-ligated heme centers in proteins. High-spin mono- and low-spin bis-amine-ligated ferrous and ferric H93G Mb adducts have been prepared together with mixed-ligand ferric heme complexes with alkylamine trans to nitrite or imidazole as heme coordination models for cytochrome c nitrite reductase or cytochrome f, respectively. Six-coordinate ferrous H93G Mb derivatives with CO, NO, and O(2) trans to the alkylamine have also been successfully formed, the latter for the first time. Finally, a novel high-valent ferryl species has been generated. The data in this study represent the first thorough investigation of the spectroscopic properties of alkylamine-ligated heme iron systems as models for naturally occurring heme proteins ligated by Lys or terminal amines.