• 3D structure;
  • function;
  • multiheme protein;
  • octaheme nitrite reductase


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  8. Supporting Information

Octaheme nitrite reductase from the haloalkaliphilic bacterium Thioalkalivibrio paradoxus was isolated and characterized. A comparative structural and functional analysis of two homologous octaheme nitrite reductases from closely related Thioalkalivibrio species was performed. It was shown that both enzymes have similar catalytic properties, owing to high structural similarity. Both enzymes are characterized by specific structural features distinguishing them from pentaheme cytochrome c nitrite reductases, such as the Tyr-Cys bond in the active site, the hexameric structure resulting in the formation of a void space inside the hexamer, and the product channel that opens into the void interior space of the hexamer. It is suggested that these specific structural features are responsible for the higher nitrite reductase activity, the greater preference for nitrite than for sulfite as a substrate, and the wider pH range of the catalytic activity of octaheme nitrite reductases than of pentaheme homologs.


Nucleotide sequence data are available in the GenBank database under the accession number HQ665012.1. Structural data are available in the RCSB Protein Data Bank database under the accession numbers 3SXQ and 3TTB

Structured digital abstract


methyl viologen


pentaheme cytochrome c nitrite reductase


Protein Data Bank


octaheme nitrite reductase from Thioalkalivibrio nitratireducens


octaheme nitrite reductase from Thioalkalivibrio paradoxus


TvPaR complex with sulfite


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  8. Supporting Information

Pentaheme cytochrome c nitrite reductases (NrfAs, EC are found in various bacterial lineages [1-3]. They catalyze the reduction of nitrite, nitric oxide and hydroxylamine to ammonia and the reduction of sulfite to sulfide. It has been suggested that they play an important role either in respiratory nitrite ammonification or in nitrite and nitric oxide detoxification [1-5]. All structurally characterized NrfAs from different classes of proteobacteria [6-13] are homodimers with the conserved arrangement of five hemes c. Four of these hemes display a standard bis-histidine coordination, whereas the fifth one – the active site heme – is coordinated by lysine in the proximal position and water/hydroxide in the distal one, with a CXXCK amino acid signature. The active site of NrfAs comprises conserved histidine, tyrosine, and arginines, and is connected to the solvent by two channels. One of the channels is mainly positively charged and thought to serve for substrate transport, and the other is mainly negatively charged and is thought to be a product transport channel. NrfAs bind up to two calcium ions, one of which is located close to the enzyme active site and is thought to contribute to its proper conformation.

A comparative sequence analysis of multiheme cytochromes c from GenBank revealed a family of putative octaheme nitrite reductases with a common signature motif, CXXCK [14]. The first representative of this family, from the nitrate-reducing obligately chemolithoautotrophic haloalkaliphilic sulfur-oxidizing bacterium Thioalkalivibrio nitratireducens strain ALEN 2 (TvNiR) [15, 16], has been purified and characterized [17, 18].

Like NrfAs, TvNiR catalyzes the reduction of nitrite and hydroxylamine to ammonia [17] and of sulfite to sulfide [17, 19]. The high-resolution crystal structures of TvNiR and its complexes with substrates (nitrite and sulfite) and inhibitors (phosphate, azide, and cyanide) have been determined [19-21]. TvNiR exists as a highly stable homohexamer. The subunit of TvNiR consists of two domains. The N-terminal domain contains three hemes. The catalytic C-terminal domain is similar to the monomer of NrfAs, and hosts the remaining five hemes, their arrangement, including the catalytic heme, being identical to that found in NrfAs. TvNiR shows essentially the same architecture of the active site as NrfAs [20], comprising heme c ligated by Lys188 at the proximal side, and the catalytic residues Tyr303, His361 and Arg131 at the distal side. The subunit of the enzyme has two channels for substrate and product transport, which are analogous to the channels of NrfA, and contain two conserved Ca2+-binding sites.

However, several important features differentiate TvNiR from NrfAs: (a) there is a covalent bond between the CE2 atom of the catalytic residue Tyr303 and the sulfur atom of Cys305 – this bond is thought to be among the factors responsible for the enhanced nitrite reductase activity of TvNiR as compared with NrfAs, because the Tyr303 hydroxyl group should be a better proton donor than the hydroxyl group of the unmodified Tyr residue of NrfA [22]; (b) TvNiR forms a homohexamer with a large cavity in the center of the molecule – as a result of the hexameric organization, the product channel in the TvNiR subunit leads not to the protein surface but to the cavity in the hexamer; and (c) the TvNiR subunit contains a narrow channel connecting the active site to the protein surface, which was suggested to serve for the proton transport necessary for the catalytic reaction [22] (six-electron reduction of nitrite is accompanied by the binding of eight protons).

In the present article, we describe the second octaheme nitrite reductase (TvPaR), isolated from the haloalkaliphilic bacterium Thioalkalivibrio paradoxus, which is closely related to T. nitratireducens [23]. T. paradoxus can grow aerobically with thiocyanate, thiosulfate, sulfide, and polysulfide, and can also oxidize carbon disulfide. Unlike T. nitratireducens, T. paradoxus cannot grow under anaerobic conditions with nitrate as the electron acceptor, and is even unable to utilize nitrate or nitrite as a nitrogen source for growth [23].

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  8. Supporting Information

Comparison of the TvPaR and TvNiR structures

The translated nucleotide sequence of TvPaR (GenBank accession number HQ665012.1) contains 553 residues, including 28 residues of the signal peptide. The amino acid sequence alignment of TvPaR and TvNiR (GenBank accession number AJ880678.2) shows 81% identity.

The structures of the free form of TvPaR and the TvPaR complex with sulfite as the substrate (TvPaR–SO3) were refined to Rf 0.138 (1.9-Å resolution) and 0.134 (2.0-Å resolution), respectively. There are two TvPaR subunits per asymmetric unit (A and B). The subunits of TvPaR form a hexamer with point group symmetry 32 (Fig. S1), similar to the TvNiR hexamer [20]. In each subunit, 519 residues are visible in the electron density (residues 5–523).

The structures of the TvPaR and TvNiR subunits are very similar; 95 amino acid substitutions in the structure of the TvPaR subunit as compared with TvNiR are accompanied by no changes in the secondary structure. The subunits of TvPaR and TvNiR [Protein Data Bank (PDB) ID: 2OT4] can be superimposed by use of the Cα atoms of 519 residues with an rmsd of 0.49 Å. The sulfite binding in the active site of TvPaR is similar to that in TvNiR (PDB ID: 3FO3) [19] (Fig. 1). All features that distinguish TvNiR from NrfA were found in the TvPaR structures as well.


Figure 1. Binding of the sulfite ion in the active site of TvPaR. The 2Fo − Fc electron density map for Tyr303 and Cys305 (1σ) and the OMIT electron density map for the sulfite and cobalt ions (5σ) are shown. The OMIT map corresponds to the 2Fo − Fc map in these regions contoured at 1σ. Hydrogen bonds and coordination bonds are indicated by dashed lines. The protoporphyrin of heme 4 and the calcium ion are shown in gray.

Download figure to PowerPoint

Substantial differences between TvPaR and TvNiR are observed in the vicinity of heme 6 (Fig. 2). The replacement of Pro400 and Val225 in TvPaR by Gln400 and Leu225 in TvNiR results in rotation of the imidazole ring of His119 coordinated to heme 6 at the distal side about the CB–CG bond by ~ 30° (Fig. 2). Also, the Pro400[RIGHTWARDS ARROW]Gln substitution leads to a change in the side chain conformation of Phe77, which is, in turn, associated with the replacement of Pro63 in TvPaR by Leu63 in TvNiR. Because of the above-mentioned structural changes, the redox properties of heme 6 in TvPaR may differ from those in TvNiR.


Figure 2. Superposition of the structures of TvPaR (green) and TvNiR (pink), by use of the atoms of hemes 6. Ser399 and Gln400 in TvNiR have two conformations.

Download figure to PowerPoint

The only substantial difference concerning the substrate and product channels is, apparently, the replacement of Leu88 in TvNiR by Glu88 in TvPaR. This replacement leads to an increase in the negative charge at the exit from the product channel into the cavity in the hexamer, which may facilitate the directed transport of the reaction product (ammonium ion) from the active site.

There are two cobalt ions per subunit in the structures of TvPaR and TvPaR–SO3. The crystallization solution served as a source of cobalt ions. One cobalt ion (Fig. 1) is bound in the active site pocket of the enzyme and interacts with propionate D of heme 4 and the side chain of His113 located at the beginning of the product channel. In TvPaR–SO3, this ion is characterized by an octahedral environment and an occupancy of 0.5. The second cobalt ion is bound at the exit from the product channel, and is coordinated by the side chains of Glu88 and His85. The binding of both of these ions may interfere with ammonium ion transport from the active site, which accounts for the observed inhibition of the nitrite reductase activity by cobalt ions (Fig. S2).

In the structures of TvPaR and TvPaR–SO3, a sulfate ion is located in the vicinity of heme 1 of subunit A (Fig. 3). This ion is hydrogen-bonded to the ND1 atom of His18, ligating heme 1. In the structures of TvNiR complexes with sulfite (PDB IDs: 3LGQ and 3FO3) [19, 22], sulfite and sulfate ions are located in the same position.


Figure 3. Binding of the sulfate ion near heme 1 in the structure of TvPaR. The 2Fo − Fc electron density map for the sulfate ion is shown (1σ). The coordination bond and hydrogen bonds are indicated by dashed lines. The side chain of Thr19 has two conformations.

Download figure to PowerPoint

Catalytic properties of TvPaR and TvNiR

TvPaR and TvNiR exhibit high nitrite reductase activity, being the most active among cytochrome c nitrite reductases (Table 1). As in the case of NrfAs, ammonium is the final reaction product [17].

Table 1. Kinetic parameters of the nitrite and sulfite reduction catalyzed by TvPaR, TvNiR, and NrfAs
EnzymeNitrite reductionSulfite reduction
kcat (s−1)Km (mm)kcat/Km (m−1·s−1)kcat (s−1)Km (mm)kcat/Km (m−1·s−1)
  1. a

    Ki for competitive inhibition of nitrite reduction by sulfite was used instead of Km for sulfite.

TvPaR4160 ± 3200.10 ± 0.024.2 × 1070.06 ± 0.020.08 ± 0.04a7.5 × 102a
TvNiR3100 ± 3000.18 ± 0.051.7 × 1070.04 ± 0.01 [19]  
NrfA from D. desulfuricans [24-27]4151.1403.6 × 1050.630.759 × 102
NrfA from E. coli [24, 28]7690.022 ± 0.0073.5 × 107

0.030 ± 0.005


0.070 ± 0.0154.3 × 102
NrfA from Sulfospirillum deleyianum [8]962  0.34  
NrfA from Wolinella succinogenes [26]380  0.4  
NrfA from Desulfovibrio vulgaris [5]639  0.3  

In spite of the high structural similarity of TvPaR and TvNiR, there are some differences in their catalytic properties. For TvNiR, substantial inhibition by the substrate was observed at a nitrite concentration > 1 mm. The inhibition constant (Ks′) is 2.5 ± 0.5 mm. For TvPaR, no inhibition by the substrate was found up to a nitrite concentration of 3 mm (Fig. S3).

The catalytic constant (kcat) for nitrite reduction by TvPaR is 1000 units higher than that for by nitrite reduction TvNiR (Table 1). This difference was reproduced for different samples of both enzymes, and reflects the more efficient catalysis of the nitrite reductase reaction by TvPaR. This may be associated with the structural differences between the enzymes in the environment of heme 6 (Fig. 2), which is located at a distance of 9.6 Å (Fe–Fe) from the catalytic heme 4 and mediates the electron transfer to the catalytic heme. The redox properties of heme 6 can play a key role in the catalysis of the nitrite reductase reaction by influencing the rate of electron transfer to nitrite, as well as the rate of intramolecular electron transport. A higher negative charge at the exit of the product channel resulting from the replacement of Leu88 in TvNiR by Glu88 in TvPaR may contribute to the high nitrite reductase activity of TvPaR. The importance of the charge of the channel for efficient catalysis is confirmed by the inhibitory effect of cobalt(II) ions, which are bound at both ends of the product channel, thus decreasing the negative potentials in these regions.

The sulfite reductase activity of TvPaR, measured at a saturating concentration of sulfite (1 mm, 10 times larger than the Ki for sulfite; see below), is identical to that of TvNiR [19] (Table 1). This activity is similar to the sulfite reductase activity of NrfA from Escherichia coli, which, like Thioalkalivibrio species, belongs to the γ-proteobacteria, and is lower than the activities of NrfAs from δ-proteobacteria and ε-proteobacteria (Table 1). The nitrite activity/sulfite activity ratio for TvPaR and TvNiR (6–7 × 104) is much higher than that for NrfAs (0.6–5.5 × 103), which may be indicative of the finer tuning of the active sites of TvPaR and TvNiR to the catalysis of nitrite reduction.

For TvNiR [19, 20, 22], it was shown that sulfite and nitrite are bound to the catalytic heme, and hence sulfite should act as a competitive inhibitor in measurements of the nitrite reductase activity in the presence of sulfite. The concentration curves for TvPaR (v versus substrate concentration) measured at different sulfite concentrations, and represented as linear Lineweaver–Burk plots, intersect in the left upper quadrant (Fig. S4). This indicates the mixed type of inhibition, which is close to the competitive one but implies that there is another sulfite-binding site, where sulfite can be bound to the enzyme—substrate complex TvNiR—NO2-. The competitive inhibition constant is 0.08 ± 0.04 mм. Therefore, NO2 and HSO3 have almost equal affinities for the active site of TvPaR, and sulfite is an efficient inhibitor of the nitrite reductase reaction. This fact may be important in regulating the nitrite reductase activity of octaheme nitrite reductases in cells.

The Michaelis constant for nitrite reduction determined earlier for TvNiR (16.7 ± 4.0 mm) [17] is overestimated, owing to the competitive inhibition by sulfite that is formed in the course of oxidation of dithionite, which was used as an electron source. The fact that we obtained sulfite complexes of TvNiR [19] or TvPaR after soaking of crystals of these enzymes in the free form with dithionite strongly suggests that sulfite, rather than sulfate or another decomposition product of dithionite, is responsible for the observed inhibition of the enzyme.

The dissociation constant (Ki′) for the second sulfite-binding site is approximately 1.3 ± 0.3 mm. The additional sulfate- and sulfite-binding site found in the TvPaR and TvNiR structures in the vicinity of heme 1 may correspond to this second sulfite-binding site (Fig. 3). Heme 1 is one of the most solvent-exposed hemes [20], and can serve as one of the primary electron acceptors from such a donor as reduced methyl viologen (MV). The binding of negatively charged ions can lead to a change in the potential of this heme, thus influencing the rate of the nitrite reductase reaction.

Effect of pH and ionic strength on the nitrite reductase activity of TvPaR

Like TvNiR, TvPaR is a periplasmic protein from haloalkaliphilic bacteria, whose optimal growth conditions are pH 10–10.2 and a salt concentration of ~ 0.5 м (NaHCO3/Na2CO3) [23]. The examination of the influence of the pH value on the rate of the nitrite reductase reaction (Vm) catalyzed by TvPaR showed that the pH optimum is within a neutral and weakly alkaline pH range (7.0–7.5) (Fig. 4A). In alkaline media (pH 9.5–10.5), the activity is high enough and equal to 15–20% of the initial value.


Figure 4. (A) The pH dependence of the nitrite reductase activity of TvPaR. The reaction was performed in 50 mm potassium phosphate/Tris (pH 6.5–9.5) and sodium borate (pH 9.5–10.5) buffer systems; nitrite, 1 mm; TvPaR, 0.036 μg·mL−1. The activity at pH 7.0 was taken as 100%. (B) The influence of NaCl concentration on the nitrite reductase activity of TvPaR. The conditions were as follows: 50 mm potassium phosphate buffer, pH 7.0; TvPaR, 0.036 μg·mL−1; nitrite, 1 mm. The activity at 0 m NaCl was taken as 100%.

Download figure to PowerPoint

An increase in the NaCl concentration in solution to 1–2 m led to a decrease in the nitrite reductase activity of TvPaR (Fig. 4B). At 0.5 m NaCl, the activity was ~ 60% of the initial activity. This effect can be attributed to the fact that electrostatic interactions essential for the transport of charged substrate and product molecules play a smaller role in salt solutions. Therefore, the specific activating effect of the high-salinity and alkaline pH medium on the activity of TvPaR was not observed in the model nitrite reduction reaction with MV.

A decrease in the rate of the nitrite reductase reaction at alkaline pH values can be attributed to a combination of several factors: the stoichiometry of the reaction (the six-electron reduction of nitrite is accompanied by the binding of eight/seven protons, whose concentration decreases with increasing pH), the less efficient transport of an uncharged ammonia molecule (pKa 9.3) through the product channel, and the deprotonation of the active site residues essential for the catalysis.

The active sites of TvPaR and TvNiR contain two groups whose deprotonation occurs in the pH range where the enzymes exhibit catalytic activity (7.0–10.0), and this can influence the efficiency of the catalysis. One of these groups is the imidazole group of His361. It was suggested that the catalytic histidine is the most probable proton donor at the first stage of nitrite reduction by NrfAs [29]. Another group is the hydroxyl of Tyr303, which is hydrogen-bonded to the O1 atom of nitrite in the active site of TvNiR [22], and can also be involved in proton transfer to the substrate. Owing to the formation of a covalent bond between Tyr303 and Cys305, the pKa of the Tyr303 hydroxyl group can decrease by 0.5–1 [30-32]. Actually, the inflection point in the right branch of the pH dependence is equal to 8.5–8.6 (Fig. 4A), and is in the range of the possible pKa of Tyr303. This is indirect evidence in favor of the participation of Tyr303 in catalysis as a proton donor. The role of Tyr303 as the proton donor is also supported by the fact that the TvNiR and TvPaR structures have a channel through which protons from the protein surface could be transferred to the hydroxyl group of Tyr303 [22]. In the NrfA structures, this channel is absent. Therefore, the involvement of Tyr303 linked to Cys305 in the catalytic actions of TvPaR and TvNiR might enable the extension of the pH range in which the enzymes exhibit high catalytic activity, and this may be of essential importance for enzymes from haloalkaliphilic bacteria.

Sequence analysis of proteins homologous to TvPaR and TvNiR

A search for sequences homologous to TvNiR and TvPaR in the UniProt Knowledgebase (Swiss-Prot) revealed genes of 16 proteins with an identity of 40% and higher (Fig. S5). One of these proteins (G4DHD8) was discovered in the genome of another bacterium of the genus Thioalkalivibrio – Thioalkalivibrio thiocyanoxidans. According to blast, this protein has 100% identity with TvPaR. Representatives of the lineage T. thiocyanoxidans were isolated together with T. paradoxus from a medium containing thiocyanate as the only source of energy, nitrogen, and sulfur. The presence of octaheme nitrite reductase in yet another organism of the genus Thioalkalivibrio suggests the importance of this class of proteins for cell viability.

Among homologs from the genomes of bacteria belonging to other genera and classes of proteobacteria, the highest homology was found with octaheme cytochromes c from the haloalkaliphilic δ-proteobacterium Desulfurivibrio alkaliphilus (D6Z5C1; 53% identity) and the Desulfobulbaceae bacterium MLMS1 (Q1NQZ7; 54% identity) isolated from soda lake sediments, like the γ-proteobacteria T. nitratireducens and Tparadoxus. Homologous proteins were also found in the genomes of obligately anaerobic δ-proteobacteria of the genus Geobacter (Q74G90, D7AEA4, A5G8N0, C6E3S2, E8WT73, B5EIZ4, B9M3N3, and B3E660; 48–46% identity), Pelobacter propionicus (A1ATC6; 49% identity), aerobic β-proteobacteria of the genus Burkholderia (F3QGY9, E7H1H6, and D9Y3D5; 40–41% identity), and the anaerobic bacterium Calditerrivibrio nitroreducens belonging to the phylum Deferribacteres (E4TEZ1; 43% identity).

All translated amino acid sequences contain (Fig. S5) seven CXXCH heme-binding motifs and one CXXCK motif unique to cytochrome c nitrite reductases, the latter motif in all sequences being the fourth from the N-terminus of the molecule. The sequence alignment of the structures superimposed by use of the heme-binding motifs showed that all sequences contain: (a) the conserved catalytic residues corresponding to the active site residues His361, Arg131 and Tyr303 of TvNiR and TvPaR (hereinafter, the numbering of the sequence for TvNiR and TvPaR is used), as well as Phe109 and Lys358, which are involved in the active site pocket of TvNiR and TvPaR; (b) the conserved residues Glu302, Gln360 and the above-mentioned Lys358, which are responsible for the binding of the Ca2+ that is present in all pentaheme and octaheme nitrite reductases; and (c) Cys305, which is characteristic only of octaheme nitrite reductases. All distal bis-histidine heme-ligating histidines, with only one exception, are also conserved. The exception is the protein D6Z5C1, in which one of these histidines is replaced by methionine.

All of these data suggest that the above-mentioned proteins belong to a new family of octaheme nitrite reductases, whose representatives are found in different classes of proteobacteria. The physiological function of these homologous proteins in cells is unknown. The question is whether this function is associated with respiratory nitrite reduction under anaerobic growth conditions. The presence of this protein in cells of T. paradoxus, which cannot grow under anaerobic conditions with nitrate as the electron acceptor and is even unable to utilize nitrate or nitrite as a nitrogen source for growth [23], rules out rather than supports this possibility. Another possible function of octaheme nitrite reductases in cells is based on their ability to reduce compounds toxic to cells, such as nitrite, nitric oxide, hydroxylamine, and hydrogen peroxide [17], and is associated with their involvement in cell detoxification and oxidative and nitrosative stress defense. The key role of NrfAs in these processes is well documented [4].

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  8. Supporting Information

Culture and growth conditions

Cultures of the T. paradoxus strain ARh1 and the T. nitratireducens strain ALEN 2 were maintained as described previously [17, 33]. Cells were routinely grown at 30 °C in a mineral medium at pH 10.0 containing (g·L−1): Na2CO3, 23; NaHCO3, 7; NaCl, 5; K2HPO4, 1; and KNO3, 1. After sterilization, the medium was supplemented with 1 mL·L−1 of trace elements [34], 1 mm MgCl2, and 40 mm sodium thiosulfate (energy source). The cells were harvested by centrifugation (7000 g, 2 times for 30 min), and disrupted in a French press at 200–220 MPa. The membrane fraction was separated by ultracentrifugation at 150 000 g for 2 h, and the supernatant (soluble fraction) was used for enzyme purification.

Protein purification

TvPaR and TvNiR were purified with a two-step chromatography procedure described earlier in detail for TvNiR [18]. The procedure involves ion exchange chromatography on DEAE-Sepharose and size-exclusion chromatography on Superdex 200. The homogeneity of the enzymes was examined by SDS/PAGE according to Laemmli [35].

The protein concentration was determined with the Bradford assay [36], with BSA as a standard, or with ε410 nm = 870 mm−1·cm−1 and 935 mm−1·cm−1 for the oxidized forms of TvNiR and TvPaR, determined with the pyridine hemochrome method [37].

Activity assay

All activities were measured at 20 °C in an anaerobic glove box with a residual oxygen pressure of < 2 p.p.m. All reagents were previously flushed with argon and kept in the glove box for at least 10 h prior to the experiments to remove the residual oxygen. MV prereduced by EuCl2 was used as an electron donor [22].

The nitrite-reducing activity of TvPaR or TvNiR was measured by two methods: as the rate of enzymatic consumption of nitrite, and as the rate of MV oxidation. In the former method, the reaction solution (1 mL) contained 50 mm potassium phosphate buffer (pH 7.0), TvPaR or TvNiR (0.02–0.1 μg·mL−1), 1 mm sodium nitrite, and 10 mm MV reduced by EuCl2. At specified intervals of time, samples were taken from the reaction mixture, and the residual nitrite concentration in the samples was determined as described in [38]. The activity was calculated as μmol of nitrite reduced per min per mg of enzyme. In the pH-dependence study, potassium phosphate/Tris (50 mm each, pH 6.5–9.0) and sodium borate (50 mm, pH 8.5–10.5) buffer systems were used.

For the determination of the kinetic constants, the nitrite concentration was varied in the range of 0.05–3 mm. The kinetic constants of TvPaR were calculated with originpro75 by nonlinear least-squares data fitting to the Michaelis–Menten equation: v = Vm × [S]/(Km + [S]), where v is the measured rate, Vm is the apparent maximum rate, Km is the apparent Michaelis constant, and [S] is the initial concentration of the varied substrate. The apparent turnover number, kcat, is calculated from Vm by dividing by the enzyme concentration. The kinetic constants of TvNiR, including the substrate inhibition constant (Ks′), were calculated with the equation accounting for substrate inhibition: v = Vm × [S]/{Km + [S] × (1 + [S]/Ks′)}.

In the study of the sulfite reduction and the inhibition of nitrite reduction by sulfite and cobalt(II) ions, the rate of the reactions was measured as the rate of MV oxidation. Reduced MV was added to a substrate or substrate + inhibitor solution in 50 mm potassium phosphate buffer (pH 7.0), to an absorbance of 2.0–1.8 at 600 nm. The reaction was initiated by TvPaR. MV oxidation was followed spectrophotometrically at 600 nm (ε600 nm = 13.6 mm−1·cm−1) in the absorbance range of 1.8–10. The TvPaR concentration in the sulfite reduction was 2–10 μg·mL−1. To study the cobalt(II) inhibition, the reaction was carried out in 50 mm Mes buffer (pH 7.0).

A Lineweaver–Burk plot (1/v versus 1/[S] for different sulfite concentrations) was used for the estimation of the mode of inhibition by sulfite. The inhibition constants (Ki and Ki′) for the mixed type of inhibition by sulfite were determined from the equation v = Vm × [S]/{Km × (1 + [I]/Ki) + [S] × (1 + [I]/Ki′)}, where [I] is the inhibitor (sulfite) concentration.

Nucleotide sequencing of the TvPaR gene

The ORF corresponding to the 1393-bp fragment of the TvPaR gene was amplified by PCR with oligonucleotides 1 and 2 shown in Table S1, with T. paradoxus genomic DNA as a template. The PCR product was purified on an Ultrafree-DA microcolumn (Millipore, Bredford, MA, USA), and its nucleotide sequence was determined by dideoxy sequencing with oligonucleotides 1–4. To complete the TvPaR gene sequence, the inverse PCR with oligonucleotides 5 and 6 (Table S1) was performed [39]. The template for the inverse PCR amplification was prepared by ligation with T4 DNA ligase (Fermentas, Vilnius, Lithuania) of T. paradoxus genomic DNA digested with the restriction endonuclease NcoI. The 1800-bp PCR product was purified from agarose gel and sequenced with oligonucleotides 5 and 6.

A multiple alignment of the TvPaR and TvNiR sequences was performed with the t-coffee webserver [40].

Crystallization and complex preparation

Crystals of TvPaR were grown with the use of two methods. In the hanging-drop vapor-diffusion experiments (278 K), drops (4 μL) consisted of equal volumes of a protein solution (10 mg·mL−1 TvPaR, 0.05 m Tris/HCl, pH 8.0, 0.2 m NaCl) and a reservoir solution of the following composition: 0.02 m cobalt(II) chloride, 0.1 m Mes (pH 6.5), and 1.8 m ammonium sulfate. The crystals were also obtained by the free interface diffusion method in microgravity with the Modul′-1 protein crystallization apparatus [41]. In the latter case, the crystallization conditions were based on those used in the former method, and were as follows: 14 mg·mL−1 TvPaR, 0.02 m cobalt(II) chloride, 0.1 m Mes (pH 6.5), and 2.8 m ammonium sulfate. The TvPaR–SO3 complex was obtained by soaking a vapor-diffusion-grown crystal of TvPaR in a reservoir solution containing 0.05 m sodium dithionite for 10 min.

X-ray diffraction data collection, and processing and structure refinement

X-ray diffraction data were collected at the K4.4 beamline at the National Research Centre ‘Kurchatov Institute’ (Russia) and at the BL41XU beamline at SPring-8 (Japan) at 100 K. The X-ray data collection statistics are given in Table 2.

Table 2. Data collection and structure refinement statistics. The data for the last resolution shell are given in parentheses. Rmeas = Σhkl{[N/(N − 1)]1/2 Σi |Ii(hkl) − <I(hkl)>|}/Σhkli[Ii(hkl)], where N is the total number of times that a given reflection is measured. Cruickshank's DPI was calculated with refmac5 [44]. The small value for TvPaR–SO3 may be attributable to the twin refinement of the structure
PDB code 3SXQ 3TTB
X-ray source



Kurchatov Institute


Wavelength (Å)0.8000.978
Resolution (Å)1.90 (2.00–1.90)2.00 (2.20–2.00)
Unit cell parameter (Å)193.6191.0
No. of measured reflections590 8611 412 053
No. of unique reflections186 813155 516
Completeness (%)98.9 (99.7)100.0 (100.0)
R meas 0.116 (0.623)0.130 (0.672)
I/σ(I)9.5 (2.2)16.5 (3.9)
B-factor from Wilson plot (Å2)23.833.0
R f 0.1380.134
R free 0.1620.153
Bond lengths (Å)0.0200.019
Bond angles (°)1.4931.699
DPI (Å)0.0740.017
No. of protein atoms82478232
No. of water molecules1284736
Average B-factor (Å2)
Protein atoms14.322.7
Water molecules25.028.8
Ramachandran plot statistics, residues in (%)
Most favored regions89.088.7
Additionally allowed regions10.610.9
Generously allowed regions0.40.4
Disallowed regions00

The X-ray diffraction data were processed with xds and xscale [42]. Crystals of TvPaR belong to the space group P213 and are isomorphous with TvNiR. The structure of TvNiR (PDB ID: 2OT4) was used as the starting model for the structure refinement of TvPaR. The structure refinement and analysis were performed with the ccp4 program suite [43].

Refinement was performed with refmac5 [44]. For TvPaR–SO3, twin refinement was performed, with twin fractions of 0.49 and 0.51. Manual correction of the models was performed on the basis of analysis of difference Fourier maps with coot [45]. Organic molecules that were present in crystallization solutions were located in the crystal structures. The structure refinement statistics are given in Table 2.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  8. Supporting Information

The work was supported by the Federal Target Program ‘Scientific and Scientific-Pedagogical Personnel of the Innovative Russia in 2009–2013’ (Government Contracts 14.740.11.0632 and P1197) and the Russian Foundation for Basic Research (grants 10-04-01695 and 11-04-01613). We thank T. Safonova for translation of the article.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  8. Supporting Information
  • 1
    Richardson DJ & Watmough NJ (1999) Inorganic nitrogen metabolism in bacteria. Curr Opin Chem Biol 3, 207219.
  • 2
    Simon J (2002) Enzymology and bioenergetics of respiratory nitrite ammonification. FEMS Microbiol Rev 26, 285309.
  • 3
    Kern M & Simon J (2009) Electron transport chains and bioenergetics of respiratory nitrogen metabolism in Wolinella succinogenes and other Epsilonproteobacteria. Biochim Biophys Acta 1787, 646656.
  • 4
    Kern M, Volz J & Simon J (2011) The oxidative and nitrosative stress defence network of Wolinella succinogenes: cytochrome c nitrite reductase mediates the stress response to nitrite, nitric oxide, hydroxylamine and hydrogen peroxide. Environ Microbiol 13, 24782494.
  • 5
    Pereira IA, LeGall J, Xavier AV & Teixeira M (2000) Characterization of a heme c nitrite reductase from a non-ammonifying microorganism Desulfovibrio vulgaris Hildenborough. Biochim Biophys Acta, 1481, 119130.
  • 6
    Einsle O, Messerschmidt A, Stach P, Bourenkov GP, Bartunik HD, Huber R & Kroneck PM (1999) Structure of cytochrome c nitrite reductase. Nature 400, 476480.
  • 7
    Einsle O, Stach P, Messerschmidt A, Simon J, Kroger A, Huber R & Kroneck PM (2000) Cytochrome c nitrite reductase from Wolinella succinogenes. Structure at 1.6 A resolution, inhibitor binding, and heme-packing motifs. J Biol Chem 275, 3960839616.
  • 8
    Stach P, Einsle O, Schumacher W, Kurun E & Kroneck PM (2000) Bacterial cytochrome c nitrite reductase: new structural and functional aspects. J Inorg Biochem 79, 381385.
  • 9
    Bamford VA, Angove HC, Seward HE, Thomson AJ, Cole JA, Butt JN, Hemmings AM & Richardson DJ (2002) Structure and spectroscopy of the periplasmic cytochrome c nitrite reductase from Escherichia coli. Biochemistry 41, 29212931.
  • 10
    Cunha CA, Macieira S, Dias JM, Almeida G, Goncalves LL, Costa C, Lampreia J, Huber R, Moura JJ, Moura I et al. (2003) Cytochrome c nitrite reductase from Desulfovibrio desulfuricans ATCC 27774. The relevance of the two calcium sites in the structure of the catalytic subunit (NrfA). J Biol Chem 278, 1745517465.
  • 11
    Almeida MG, Macieira S, Goncalves LL, Huber R, Cunha CA, Romao MJ, Costa C, Lampreia J, Moura JJ & Moura I (2003) The isolation and characterization of cytochrome c nitrite reductase subunits (NrfA and NrfH) from Desulfovibrio desulfuricans ATCC 27774. Re-evaluation of the spectroscopic data and redox properties. Eur J Biochem 270, 39043915.
  • 12
    Rodrigues ML, Oliveira TF, Pereira IA & Archer M (2006) X-ray structure of the membrane-bound cytochrome c quinol dehydrogenase NrfH reveals novel haem coordination. EMBO J 25, 59515960.
  • 13
    Rodrigues ML, Scott KA, Sansom MS, Pereira IA & Archer M (2008) Quinol oxidation by c-type cytochromes: structural characterization of the menaquinol binding site of NrfHA. J Mol Biol 381, 341350.
  • 14
    Klotz MG, Schmid MC, Strous M, op den Camp HJ, Jetten MS & Hooper AB (2008) Evolution of an octahaem cytochrome c protein family that is key to aerobic and anaerobic ammonia oxidation by bacteria. Environ Microbiol 10, 31503163.
  • 15
    Sorokin DY, Antipov AN & Kuenen JG (2003) Complete denitrification in coculture of obligately chemolithoautotrophic haloalkaliphilic sulfur-oxidizing bacteria from a hypersaline soda lake. Arch Microbiol 180, 127133.
  • 16
    Sorokin DY, Touruva TP, Sjollema KA & Kuenen JG (2003) Thioalkalivibrio nitratireducens sp. nov., a nitrate-reducing member of an autotrophic denitrifying consortium from a soda lake. Int J Syst Evol Microbiol 53, 17791783.
  • 17
    Tikhonova TV, Slutsky A, Antipov AN, Boyko KM, Polyakov KM, Sorokin DY, Zvyagilskaya RA & Popov VO (2006) Molecular and catalytic properties of a novel cytochrome c nitrite reductase from nitrate-reducing haloalkaliphilic sulfur-oxidizing bacterium Thioalkalivibrio nitratireducens. Biochim Biophys Acta 1764, 715723.
  • 18
    Tikhonova TV, Slutskaya ES, Filimonenkov AA, Boyko KM, Kleimenov SY, Konarev PV, Polyakov KM, Svergun DI, Trofimov AA, Khomenkov VG et al. (2008) Isolation and oligomeric composition of cytochrome c nitrite reductase from the haloalkaliphilic bacterium Thioalkalivibrio nitratireducens. Biochemistry (Mosc) 73, 164170.
  • 19
    Trofimov AA, Polyakov KM, Boyko KM, Tikhonova TV, Safonova TN, Tikhonov AV, Popov AN & Popov VO (2010) Structures of complexes of octahaem cytochrome c nitrite reductase from Thioalkalivibrio nitratireducens with sulfite and cyanide. Acta Crystallogr D Biol Crystallogr 66, 10431047.
  • 20
    Polyakov KM, Boyko KM, Tikhonova TV, Slutsky A, Antipov AN, Zvyagilskaya RA, Popov AN, Bourenkov GP, Lamzin VS & Popov VO (2009) High-resolution structural analysis of a novel octaheme cytochrome c nitrite reductase from the haloalkaliphilic bacterium Thioalkalivibrio nitratireducens. J Mol Biol 389, 846862.
  • 21
    Trofimov AA, Polyakov KM, Boyko KM, Filimonenkov AA, Dorovatovskii PV, Tikhonova TV, Popov VO & Kovalchuk MV (2010) Structure of octaheme cytochrome c nitrite reductase from Thioalkalivibrio nitratireducens in a complex with phosphate. Crystallogr Rep 55, 5864.
  • 22
    Trofimov AA, Polyakov KM, Tikhonova TV, Tikhonov AV, Safonova TN, Boyko KM, Dorovatovsky PV & Popov VO (2012) Covalent modifications of catalytic tyrosine in octaheme cytochrome c nitrite reductase and their effect on the enzyme activity. Acta Crystallogr D Biol Crystallogr 68, 144153.
  • 23
    Sorokin DY, Tourova TP, Lysenko AM, Mityushina LL & Kuenen JG (2002) Thioalkalivibrio thiocyanoxidans sp. nov. and Thioalkalivibrio paradoxus sp. nov., novel alkaliphilic, obligately autotrophic, sulfur-oxidizing bacteria capable of growth on thiocyanate, from soda lakes. Int J Syst Evol Microbiol 52, 657664.
  • 24
    Clarke TA, Hemmings AM, Burlat B, Butt JN, Cole JA & Richardson DJ (2006) Comparison of the structural and kinetic properties of the cytochrome c nitrite reductases from Escherichia coli, Wolinella succinogenes, Sulfurospirillum deleyianum and Desulfovibrio desulfuricans. Biochem Soc Trans 34, 143145.
  • 25
    Pereira IC, Abreu IA, Xavier AV, LeGall J & Teixeira M (1996) Nitrite reductase from Desulfovibrio desulfuricans (ATCC 27774) a heterooligomer heme protein with sulfite reductase activity. Biochem Biophys Res Commun 224, 611618.
  • 26
    Lukat P, Rudolf M, Stach P, Messerschmidt A, Kroneck PMH, Simon J & Einsle O (2008) Binding and reduction of sulfite by cytochrome c nitrite reductase. Biochemistry 47, 20802086.
  • 27
    Liu MC & Peck HD Jr (1981) The isolation of a hexaheme cytochrome from Desulfovibrio desulfuricans and its identification as a new type of nitrite reductase. J Biol Chem 256, 1315913164.
  • 28
    Kemp GL, Clarke TA, Marritt SJ, Lockwood C, Poock SR, Hemmings AM, Richardson DJ, Cheesman MR & Butt JN (2010) Kinetic and thermodynamic resolution of the interactions between sulfite and the pentahaem cytochrome NrfA from Escherichia coli. Biochem J 431, 7380.
  • 29
    Bykov D & Neese F (2011) Substrate binding and activation in the active site of cytochrome c nitrite reductase: a density functional study. J Biol Inorg Chem 16, 417430.
  • 30
    Himo F, Noodleman L, Blomberg MRA & Siegbahn PEM (2002) Relative acidities of ortho-substituted phenols, as model for modified tyrosines in proteins. J Phys Chem 106, 87578761.
  • 31
    Whittaker MM, Chuang YY & Whittaker JW (1993) Models for the redox active site in galactose oxidase. J Am Chem Soc 115, 1002910035.
  • 32
    Itoh S, Takayama S, Arakawa R, Furuta A, Komatsu M, Ishida A, Takamuku S & Fukuzumi S (1997) Active site models for galactose oxidase. Electronic effect of the thioether group in the novel organic cofactor. Inorg Chem 36, 14071416.
  • 33
    Sorokin DY, Lysenko AM, Mityushina LL, Tourova TP, Jones BE, Rainey FA, Robertson LA & Kuenen JG (2001) Thioalkalimicrobium aerophilum gen. nov., sp. nov. and Thioalkalimicrobium sibericum sp. nov., and Thioalkalivibrio versutus gen. nov., sp. nov., Thioalkalivibrio nitratis sp. nov., novel and Thioalkalivibrio denitrificancs sp. nov., novel obligately alkaliphilic and obligately chemolithoautotrophic sulfur-oxidizing bacteria from soda lakes. Int J Syst Evol Microbiol 51, 565580.
  • 34
    Pfennig N & Lippert KD (1966) Utilisation of molecular hydrogen by Chlorobium thiosulfatophilum. Growth and CO2-fixation. Arch Microbiol 55, 245256.
  • 35
    Laemmli UK (1970) Cleavage of structural proteins during the assembly of the 533 head of bacteriophage T4. Nature, 227, 680685.
  • 36
    Bradford M (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248254.
  • 37
    Berry EA & Trumpower BL (1987) Simultaneous determination of hemes a, b, and c from pyridine hemochrome spectra. Anal Biochem 161, 115.
  • 38
    Nicolas DJD & Nason A (1957) Determination of nitrate and nitrite. Methods Enzymol 3, 981984.
  • 39
    Ochman H, Gerber AS & Hartl DL (1988) Genetic applications of an inverse polymerase chain reaction. Genetics 120, 621623.
  • 40
    Poirot O, O'Toole E & Notredame C (2003) Tcoffee@igs: a web server for computing, evaluating and combining multiple sequence alignments. Nucleic Acids Res 31, 35033506.
  • 41
    Smirnova EA, Kislitsyn YuA, Sosfenov NI, Lyashenko AV, Popov AN, Baidus AN, Timofeev I & Kuranova IP (2009) Protein crystal growth on the Russian segment of the International Space Station. Crystallogr Rep 54, 901911.
  • 42
    Kabsch W (1993) Automatic processing of rotation diffraction data from crystals of initially unknown symmetry and cell constants. J Appl Crystallogr 26, 795800.
  • 43
    Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P, Evans PR, Keegan RM, Krissinel EB, Leslie AG, McCoy A et al. (2011) Overview of the CCP4 suite and current developments. Acta Crystallogr D Biol Crystallogr 67, 235242.
  • 44
    Murshudov GN, Vagin AA & Dodson EJ (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr 53, 240255.
  • 45
    Emsley P & Cowtan K (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr 60, 21262132.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  8. Supporting Information

Fig. S1. Structure of the TvPaR hexamer.

Fig. S2. Inhibition of the nitrite reductase activity of TvPaR by cobalt(II) ions.

Fig. S3. Dependence of the rate of nitrite reduction by TvNiR and TvPaR on substrate concentration.

Fig. S4. Inhibition of the nitrite reductase activity of TvPaR by sulfite: determination of the type of inhibition from the Lineweaver–Burk plot.

Fig. S5. Alignment of the translated amino acid sequences of 17 homological octaheme c cytochromes.

Table S1. Oligonucleotides used for TvPaR gene sequencing.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.