On the purification of nitrite reductase from Thiobacillus denitrificans and its reaction with nitrite under reducing conditions

Activation of the cytochrome cd1 nitrite reductase from Paracoccus pantotrophus. Reaction of oxidized enzyme with substrate drives a ligand switch at heme c

Cytochromes cd(1) are dimeric bacterial nitrite reductases, which contain two hemes per monomer. On reduction of both hemes, the distal ligand of heme d(1) dissociates, creating a vacant coordination site accessible to substrate. Heme c, which transfers electrons from donor proteins into the active site, has histidine/methionine ligands except in the oxidized enzyme from Paracoccus pantotrophus where both ligands are histidine. During reduction of this enzyme, Tyr(25) dissociates from the distal side of heme d(1), and one heme c ligand is replaced by methionine. Activity is associated with histidine/methionine coordination at heme c, and it is believed that P. pantotrophus cytochrome cd(1) is unreactive toward substrate without reductive activation. However, we report here that the oxidized enzyme will react with nitrite to yield a novel species in which heme d(1) is EPR-silent. Magnetic circular dichroism studies indicate that heme d(1) is low-spin Fe(III) but EPR-silent as a result of spin coupling to a radical species formed during the reaction with nitrite. This reaction drives the switch to histidine/methionine ligation at Fe(III) heme c. Thus the enzyme is activated by exposure to its physiological substrate without the necessity of passing through the reduced state. This reactivity toward nitrite is also observed for oxidized cytochrome cd(1) from Pseudomonas stutzeri suggesting a more general involvement of the EPR-silent Fe(III) heme d(1) species in nitrite reduction.

The purification and properties of a cd-cytochrome nitrite reductase from Paracoccus halodenitrificans

Paracoccus halodenitrificans, grown anaerobically in the presence of nitrite, contained membrane and cytoplasmic nitrite reductases. When assayed in the presence of phenazine methosulfate and ascorbate, the membrane-bound enzyme produced nitrous oxide whereas the cytoplasmic enzyme produced nitric oxide. When both enzymes were assayed in the presence of methyl viologen and dithionite, the cytoplasmic enzyme produced ammonia. Following solubilization, the membrane-bound enzyme behaved like the cytoplasmic enzyme, producing nitric oxide in the presence of phenazine methosulfate and ascorbate, and ammonia when assayed in the presence of methyl viologen and dithionite. The cytoplasmic and membrane-bound enzymes were purified to essentially the same specific activity. Only a single nitrite-reductase activity was detected on electrophoretic gels and the electrophoretic behavior of both enzymes suggested they were identical. The spectral properties of both enzymes suggested they were cd-type cytochromes. These data suggest that the products of nitrite reduction by the cd-cytochrome nitrite reductase are determined by the location of the enzyme and the redox potential of the electron donor.

From no-confidence to nitric oxide acknowledgement: a story of bacterial nitric-oxide reductase

The review briefly summarizes current knowledge of the bacterial nitric-oxide reductase (NOR). This membrane enzyme consists of two subunits, the smaller one contains haem C and the larger one two haems B and nonhaem iron. The protein sequence and structure of metal centres demonstrate the relationship of NOR to the family of terminal oxidases. The binuclear Fe-Fe reaction centre, consisting of antiferromagnetically coupled haem B and nonhaem iron, is analogous to Fe-Cu centre of terminal oxidases. The data on the structure and function of NOR and terminal oxidases suggest that all these enzymes are closely evolutionally related. The catalytic properties are determined most of all by the relatively high toxicity of nitric oxide as a substrate and the resulting strong need to maintain its concentration at nanomolar levels. A kinetic model of the action of the enzyme comprises substrate inhibition. NOR does not conserve the free energy of nitric oxide reduction because it does not work as a proton pump and, moreover, the protons coming into the reaction are taken from periplasm, i.e. they do not cross the membrane.

Bacterial nitric oxide synthesis

The structure-function relationships in nitrite reductases, key enzymes in the dissimilatory denitrification pathway which reduce nitrite to nitric oxide (NO), are reviewed in this paper. The mechanisms of NO production are discussed in detail and special attention is paid to new structural information, such as the high resolution structure of the copper- and heme-containing enzymes from different sources. Finally, some implications relevant to regulation of the steady state levels of NO in denitrifiers are presented.

Microbial and plant metabolism of NO

During microbial denitrification, NO is produced by reduction of nitrite by either the reduced high spin d1 hemes in a unique reductase (NIR) or at the expense of a blue copper protein that transfers electrons that move first to a type I copper and then to a type II copper in a unique trimeric NIR. This latter type of NIR is also produced by several denitrifying filamentous fungi. Reduction of NO is then carried out by either a specific cytochrome be complex NOR in denitrifying bacteria or a unique cytochrome P-450 in denitrifying filamentous fungi. NO is also produced by an anomalous reaction of a molybdoprotein, nitrate reductase (NAR), acting on an odd substrate, NO2-. NO is also reduced by a multiheme NIR that serves physiologically for reduction of NO2- to NH3. This type NIR reduces NO to either N2O, if only partially reduced, or NH3, if fully reduced, when it encounters NO. This multiheme NIR is very sensitive to cyanide. Transcription of the genes for NIR and NOR production in a denitrifier is activated by NO, a process that also requires the presence of the gene product, a transcriptional activator, NnrR.

Immunological identification and distribution of dissimilatory heme cd1 and nonheme copper nitrite reductases in denitrifying bacteria

Polyclonal antibodies were used to identify heme or copper nitrite reductases in the following groups: 23 taxonomically diverse denitrifiers from culture collections, 100 numerically dominant denitrifiers from geographically diverse environments, and 51 denitrifiers from a culture collection not selected for denitrification. Antisera were raised against heme nitrite reductases from Pseudomonas aeruginosa and Pseudomonas stutzeri and against copper nitrite reductase from Achromobacter cycloclastes. Nitrite reductases were identified by Western immunoblot. Diethyldithiocarbamate, which specifically inhibits copper nitrite reductases, was used to confirm the immunological characterization and determine which type was present in strains nonreactive with any antiserum. For groups in which the type of nitrite reductase has not been previously described, we found that Alcaligenes eutrophus, Bacillus azotoformans, Bradyrhizobium japonicum, Corynebacterium nephridii, and Rhizobium spp. contained copper nitrite reductase, while Aquaspirillum itersonii, Flavobacterium spp., and Pseudomonas fluorescens contained heme nitrite reductase. Heme nitrite reductases dominated, regardless of soil type or geographic origin. They occurred in 64 and 92%, respectively, of denitrifiers in the numerically dominant and nonselected collections. The two nitrite reductase types were mutually exclusive in individual bacteria, but both appeared in different strains from the Alcaligenes and Pseudomonas genera. The heme type predominated in Pseudomonas strains. The heme-type nitrite reductase appeared more conserved if judged by similarities in molecular weights and immunological reactions. The Cu type was found in more taxonomically unrelated strains and varied in molecular weight and antiserum recognition.

Two types of cytochrome cd1 in the aerobic photosynthetic bacterium, Erythrobacter sp. OCh 114

Components I and II of cytochrome cd1 which had different spectral features were purified from the aerobic photosynthetic bacterium, Erythrobacter sp. strain OCh 114. Component I showed an absorption maxima at 700 and 406 nm in the oxidized form, and at 621, 552.5, 548 and 416 nm in the reduced form. Component II showed an absorption maxima at 635 and 410 nm in the oxidized form and at 628, 552.5, 548 and 417 nm in the reduced form. The relative molecular mass, Mr, of both cytochromes was determined to be 135,000 with two identical subunits. Components I and II showed pI values of 7.6 and 6.8, respectively. The redox potential of hemes ranged from +234 mV to +242 mV, except for the heme d1 of component I (Em7 = +134 mV). Components I and II showed both cytochrome c oxidase and nitrite reductase activities. Cytochrome c oxidase activity was strongly inhibited by a low concentration of nitrite and cyanide. Erythrobacter cytochromes c-551 and c-552 were utilized as electron donors for the cytochrome c oxidase reaction. The high affinity of cytochrome c-552 to component II (Km = 1.27 microM) suggested a physiological significance for this cytochrome. Erythrobacter cytochromes cd1 are unique in their presence in cells grown under aerobic conditions as compared to other bacterial cytochromes cd1 which are formed only under denitrifying conditions.

Optical, EPR and Mössbauer spectroscopic studies on the NO derivatives of cytochrome cd1 from Thiobacillus denitrificans

We have used optical, EPR and Mössbauer spectroscopies to study the formation of heme-NO complex upon the addition of nitrite to reduced cytochrome cd1 from Thiobacillus denitrificans. The reduced d1 heme binds NO under both alkaline and acidic conditions, but the binding of NO to the reduced c heme was strongly pH-dependent. The Mössbauer data showed unambiguously that at pH 7.6 the c heme does not complex NO, whereas at pH 5.8 approximately half of the reduced c heme binds NO. This observation was confirmed by EPR studies, which showed that the spin concentration of the heme-NO EPR signal increased from 2 spins/molecule at pH 8.0 to approximately 3 spins/molecule at pH 5.8. Optical absorption study also showed strong pH dependence in the binding of NO to the reduced c heme. We have also analyzed the Mössbauer spectra of the ferrous d1 heme-NO complex using a spin-Hamiltonian formalism. The magnetic hyperfine coupling tensor was found to be consistent with the unpaired electron residing on a sigma orbital.

Comparative EPR studies on the nitrite reductases from Escherichia coli and Wolinella succinogenes

Hexaheme nitrite reductases purified to homogeneity from Escherichia coli K-12 and Wolinella succinogenes were studied by low-temperature EPR spectroscopy. In their isolated states, the two enzymes revealed nearly identical EPR spectra when measured at 12 K. Both high-spin and low-spin ferric heme EPR resonances with g values of 9.7, 3.7, 2.9, 2.3 and 1.5 were observed. These signals disappeared upon reduction by dithionite. Reaction of reduced enzyme with nitrite resulted in the formation of ferrous heme-NO complexes with distinct EPR spectral characteristics. The heme-NO complexes formed with the two enzymes differed, however, in g values and line-shapes. When reacted with hydroxylamine, reduced enzymes also showed the formation of ferrous heme-NO complexes. These results suggested the involvement of an enzyme-bound NO intermediate during the six-electron reduction of nitrite to ammonia catalyzed by these two hexaheme nitrite reductases. Heme proteins that can either expose bound NO to reduction or release it are significant components of both assimilatory and dissimilatory metabolisms of nitrate. The different ferrous heme-NO complexes detected for the two enzymes indicated, nevertheless, their subtle variation in heme reactivity during the reduction reaction.

An investigation of the ligand-binding properties of Pseudomonas aeruginosa nitrite reductase

The low-temperature e.p.r. and m.c.d. (magnetic-circular-dichroism) spectra of Pseudomonas aeruginosa nitrite reductase, together with those of its partially and fully cyanide-bound derivatives, were investigated. The m.c.d. spectra in the range 600-2000 nm indicate that the native axial ligands to haem c are histidine and methionine, and furthermore that it is the methionine ligand that must be displaced before cyanide binding at this haem. The m.c.d. spectra in the range 1000-2000 nm contain no charge-transfer bands arising from low-spin ferric haem d1, a chlorin. New optical transitions in the region 700-850 nm were found for the cyanide adduct of haem d1. The g-values of haem d1 in the native enzyme are 2.51, 2.43 and 1.71, suggesting co-ordination by two histidine ligands in the oxidized state. There is clear evidence in the e.p.r. data of an interaction between the c and d1 haem groups. This is not apparent in the optical spectra. The results are interpreted in terms of haem groups that are remote from each other, their interaction being mediated through protein conformational changes. The possible implications of this in relation to reduction processes catalysed by the enzyme are considered.

Correlations among parameters that describe low-spin ferric heme complexes

The g values from low-spin ferric hemes can be related through the t2g hole model to rhombic (V/lambda) and tetragonal (delta/lambda) ligand field components and to the lowest Kramer's doublet energy (E/lambda). The latter is also a measure of unpaired electron sharing among the iron 3d (t2g) orbitals. For a series of ligands (X), there is a monotonic increase in myoglobin complex (Mb . X) [E/lambda] values with nonheme hexacoordinate metal complex (M . X6) [eg-t2gPg] orbital separations. As the aqueous solution pKa values of the sulfurous or nitrogenous ligands in model heme complexes increase, values of V/lambda and delta/lambda increase linearly, but those of [E/lambda] decrease linearly. The greater the electron-acceptor ability of the ligand, as suggested by its position in the spectrochemical series or its pKa, the more the unpaired electron sharing among the heme t2g orbitals increases. The rate of change of [E/lambda] with V/lambda and the pKa is different with sulfurous and nitrogenous ligands, and the magnitude of both rates increases with two sulfurs less than sulfur and nitrogen less than two nitrogens bound to the heme. The maximum magnitude of this rate with V/lambda for cytochrome P-450 is four times less than that for myoglobin, which may explain, in part, the differences in ligand binding between these two hemeproteins. The perturbation of [E/lambda], V/lambda, and delta/lambda induced by strain of iron-ligand bonds is quantitated for several hemeproteins and heme models. In addition, energy level comparisons suggest that the largest-magnitude g value falls approximately along the iron-chlorin ring normal. This suggestion implies that the electron distribution of the iron at the catalytic sites of cytochrome P-450 and certain chlorin-containing enzymes is in some way similar, but distinct from that at the transport site of myoglobin.

Denitrification and nitrite reduction: Pseudomonas aeruginosa nitrite-reductase

Present knowledge of the different enzymatic steps of the denitrification chains in various bacteria, particularly Paracoccus denitrificans and Pseudomonas aeruginosa has been briefly reviewed. The question whether nitric oxide (NO), nitrous oxide (N2O) and other nitrogen derivatives are obligatory intermediates has been discussed. The second part is an extensive review of the structure and the function of a key enzyme in denitrification, cytochrome c551-nitrite-oxidoreductase from P. aeruginosa. Recent results on the stoichiometry of nitrite reduction have been discussed.

A dissimilatory nitrite reductase in Paracoccus halodenitrificans

Paracoccus halodenitrificans produced a membrane-associated nitrite reductase. Spectrophotometric analysis showed it to be associated with a cd-cytochrome and located on the inner side of the cytoplasmic membrane. When supplied with nitrite, membrane preparations produced nitrous oxide and nitric oxide in different ratios depending on the electron donor employed. The nitrite reductase was maximally active at relatively low concentrations of sodium chloride and remained attached to the membranes at 100 mM sodium chloride.

Nitrogen oxide reduction in Wolinella succinogenes and Campylobacter species

Wolinella succinogenes cells and extracts reduced nitric oxide, and cells, but not extracts, reduced nitrous oxide. Formate-reduced W. succinogenes extracts generated the 573-nm peak in difference spectra seen previously in response to nitric oxide in denitrifiers. The type strains of several Campylobacter species did not reduce either gaseous oxide. Cells, but not extracts, of C. fetus subspecies (fetus and venerealis) reduced nitrous oxide; acetylene inhibited reduction. Neither cells nor extracts reduced nitric oxide.

Discrimination of ascorbate-dependent nonenzymatic and enzymatic, membrane-bound reduction of nitric oxide in denitrifying Pseudomonas perfectomarinus

The marine nitrite-respiring (denitrifying) bacterium, Pseudomonas perfectomarinus, catalyzes by a membrane-bound enzyme the reduction of nitric oxide to nitrous oxide with ascorbic-reduced phenazine methosulfate as electron donor. The entire nitric oxide-reducing capability of a cell-free system was membrane bound and this process was studied with respect to pH and substrate dependency. The enzymatic process was perturbed by an identical nonenzymatic reduction by iron(II) ascorbate in neutral to alkaline aqueous solution. 2 mol nitric oxide and 1 mol ascorbate were consumed per mol nitrous oxide formed. Enzymatic and nonenzymatic processes were discriminated by their differential behavior towards pH and metal-chelating agents. The pH optimum for the enzymatic and nonenzymatic reaction was 5.2 and greater than 7.0, respectively. EDTA (10 mM) inhibited the nonenzymatic reduction completely without interfering with the membrane-bound activity. The nonenzymatic system mimics the reaction of nitric oxide reductase and could serve as a model to study the formation of the N-N bond in denitrification. Enzymatic generation of nitric oxide by cytochrome cd and subsequent nonenzymatic reduction to nitrous oxide simulate an overall quasi-enzymatic nitrous oxide formation by cytochrome cd. The nonenzymatic reduction of nitric oxide might have occurred in previous work due to the ubiquitous use of ascorbate in studies on nitrite respiration and the likelihood of adventitious iron in biological samples.

Electron transport-linked proton translocation at nitrite reduction in Campylobacter sputorum subspecies bubulus

Campylobacter sputorum subspecies bubulus contains a membrane-bound nitrite reductase which catalyses the six-electron reduction of nitrite to ammonia. Formate and L-lactate are used as hydrogen donors. Cells of C. sputorum grown with nitrate or nitrite contain cytochromes of the b- and c-type and a carbon monoxide-binding cytochrome c. In addition, a special membrane-bound carbon monoxide-binding pigment is found. Nitrite reduction with formate or L-lactate as a hydrogen donor is strongly inhibited by 2-n-heptyl-4-hydroxyquinoline-N-oxide (HQNO). Nitrite reduction by bacterial suspensions with lactate as a hydrogen donor is strongly inhibited by carbonylcyanide-m-chlorophenylhydrazone (CCCP) whereas nitrite reduction with formate as a hydrogen donor is not inhibited at all. Leads to H+/O values and leads to H+/NO-2 values were measured with ascorbate + N,N,N',N'-tetramethyl-p-phenylenediamine (TMPD), formate (in the absence and presence of carbonic anhydrase) and L-lactate as a hydrogen donor. The results are summarized in a scheme for electron transport from formate or lactate to oxygen or nitrite which shows a periplasmic orientation of formate dehydrogenase and nitrite reductase and a cytoplasmic orientation of lactate dehydrogenase and oxygen reduction, and which shows proton translocation with a leads to H+/2e value of 2.0. The leads to H+/O and leads to H+/NO-2 values predicted by this scheme are in good agreement with the experimental values.

Respiration-driven proton translocation with nitrite and nitrous oxide in Paracoccus denitrificans

(1)H+ leads to/electron acceptor ratios have been determined with the oxidant pulse method for cells of denitrifying Paracoccus denitrificans oxidizing endogenous substrates during reduction of O2, NO2- or N2O. Under optimal H+-translocation conditions, the ratios leads to H+/O, H+ leads to/N2O, H+ leads to/NO2- for reduction to N2 and H+ leads to/NO2- for reduction to N2O were 6.0-6.3, 4.02, 5.79 and 3.37, respectively. (2) With ascorbate/N,N,N,'N'-tetramethyl-p-phenylene-diamine as exogenous substrate, addition of NO2- or N2O to an anaerobic cell suspension resulted in rapid alkalinization of the outer bulk medium. H+/N2O, H+/NO2- for reduction to N2 and H+/NO2- for reduction to N2O were -0.84, -2.33 and -1.90, respectively. (3) The H+/oxidant ratios, mentioned in item 2, were not altered in the presence of valinomycin/K+ and the triphenylmethylphosphonium cation. (4) A simplified scheme of electron transport to O2, NO2- and N2O is presented which shows a periplasmic orientation of the nitrite reductase as well as the nitrous oxide reductase. Electrons destined for NO2-, N2O or O2 pass two H+-trans-locating sites. The H+ leads to/electron acceptor ratios predicted by this scheme are in good agreement with the experimental values.

Electron paramagnetic resonance studies on Pseudomonas nitrosyl nitrite reductase. Evidence for multiple species in the electron paramagnetic resonance spectra of nitrosyl haemoproteins

The e.p.r. spectra of reduced 14NO- and 15NO-bound Pseudomonas nitrite reductase have been investigated at pH 5.8 and 8.0 in four buffer systems. At pH 8.0, absorption spectra indicated that only the haem d1 was NO-bound, but, although quantification of the e.p.r. signals in all cases accounted for NO bound the the haem d1 in both subunits of the enzyme, the precise form of the signals varied with buffer and temperature. A rhombic species, with gx = 2.07, gz = 2.01 and gy = 1.96, represented in the low-temperature spectra seen in all the buffers was converted at high temperatures (approx. 200K) into a form showing a reduced anisotropy. Hyperfine splitting on the gz component of this rhombic signal indicated a nitrogen atom trans to NO and it is proposed that histidine provides the endogenous axial ligand for haem d1. At pH 5.8, absorption spectra indicated NO binding to both haems c and d1 and e.p.r. quantifications accounted for NO-bound haems c and d1 in both enzyme subunits. The e.p.r. spectra at pH 5.8 were generally similar to those at pH 8.0 with respect to g-values and hyperfine coupling constants, but were broader with less well defined hyperfine splittings. As at pH 8, rhombic signals present in spectra at low temperatures were converted to less anisotropic forms at high temperatures. The results are discussed in relation to work on model nitrosyl-protohaem complexes [Yoshimura, Ozaki, Shintani & Watanabe (1979) Arch. Biochem, Biophys. 193, 301-313]. No. e.p.r. signal was observed from oxidized NO-bound Pseudomonas nitrite reductase at pH 6.0, over the temperature range 6-100K.

Reduction of nitrite to nitrous oxide by a cytoplasmic membrane fraction from the marine denitrifier Pseudomonas perfectomarinus

A cytoplasmic membrane fraction from the marine denitrifier Pseudomonas perfectomarinus reduced nitrite to nitrous oxide in a stoichiometric reaction without nitric oxide as free intermediate. The membrane system had a specific requirement for FMN with NAD(P)H as electron donors. Other electron donors were ascorbate-reduced cytochrome c-551 or phenazine methosulfate. The membrane fraction contained tightly bound cytochrome cd which represented only a small portion of the total cytochrome cd of the cell. As further terminal oxidase cytochrome o was identified. The membrane fraction produced also nitrous oxide from nitric oxide, however, at a substantially lower rate than from nitrite when using ascorbate-reduced phenazine methosulfate as electron donor.