Nitric Oxide Production from Nitrite Reduction and Hydroxylamine Oxidation by Copper-containing Dissimilatory Nitrite Reductase (NirK) from the Aerobic Ammonia-oxidizing Archaeon, Nitrososphaera viennensis

Abstract

Aerobic ammonia-oxidizing archaea (AOA) play a crucial role in the global nitrogen cycle by oxidizing ammonia to nitrite, and nitric oxide (NO) is a key intermediate in AOA for sustaining aerobic ammonia oxidation activity. We herein heterologously expressed the NO-forming, copper-containing, dissimilatory nitrite reductase (NirK) from Nitrososphaera viennensis and investigated its enzymatic properties. The recombinant protein catalyzed the reduction of ¹⁵NO₂⁻ to ¹⁵NO, the oxidation of hydroxylamine (¹⁵NH₂OH) to ¹⁵NO, and the production of ^14–15N₂O from ¹⁵NH₂OH and ¹⁴NO₂⁻. To the best of our knowledge, the present study is the first to document the enzymatic properties of AOA NirK.

Aerobic ammonia oxidation, a rate-limiting step of nitrification, drives the global nitrogen cycle (24, 40), which involves aerobic ammonia-oxidizing archaea and bacteria (AOA and AOB, respectively) and complete ammonia oxidizers (comammox) (9, 44). Of these, AOA primarily contribute to aerobic ammonia oxidation in natural environments including soil and open ocean (19, 31, 46). AOA are affiliated with the phylum Thaumarchaeota, which includes phylogenetically and physiologically diverse members (6) and the soil-inhabiting archaeon Nitrososphaera viennensis (41). The biochemistry of aerobic ammonia oxidation by AOA has received a great deal of interest because ammonia oxidation to nitrite (NO₂⁻) proceeds in a different manner to that of AOB. AOA oxidize ammonia to hydroxylamine by ammonia monooxygenase (Amo) as well as AOB (43), while hydroxylamine is further oxidized to NO₂⁻ by an unidentified enzyme (17). All known AOA genomes lack the gene encoding hydroxylamine dehydrogenase (Hao), and the involvement of a copper-protein complex has been proposed (40, 45). In parallel with the oxidation of ammonia to NO₂⁻, AOA produce nitric oxide (NO) (22). NO is a key intermediate in AOA cells because this highly reactive molecule is essential for sustaining aerobic ammonia oxidation activity (17, 33, 36, 47). To date, the following 2 pathways have been reported as a source of prokaryotic NO formation: NO₂⁻ reduction to NO by copper-containing and cytochrome cd₁-type dissimilatory nitrite reductases (NirK and NirS, respectively) (38) and NH₂OH oxidation to NO by hydroxylamine oxidoreductase (Hao) (4, 21). Although neither nirS nor hao are found in AOA genomes (6), AOA commonly possess nirK, which is transcribed and expressed during aerobic ammonia oxidation (8, 15, 20, 37). These findings suggest that NirK are involved in NO formation in AOA cells. However, NO₂⁻ reduction to NO by AOA NirK has never been demonstrated.

Bacterial NirK have been characterized as homotrimeric enzymes, and each subunit has 2 Cu-binding sites (Type 1 and 2 Cu-binding sites). Type 1 Cu-binding sites receive an electron from an electron donor, such as type 1 Cu proteins (single-domain cupredoxins) and/or cytochrome c, and the electron is then further transferred to a type 2 Cu-binding site that is the catalytic center of NirK (14, 25). Bacterial NirK have been classified into 2 phylogenetically distinct groups (class 1 and class 2 groups) based on sequence similarities, and the NirK of the class 1 group contains linker loop and tower loop regions in the amino acid sequence (3). AOA NirK, including Ns. viennensis NirK, are affiliated with a distinct clade of bacterial class 1 and 2 groups (Fig. 1A). Lund et al. (20) reported that AOA NirK may be further classified into several phylogenetic clades showing specific geographic distributions. Ns. viennensis NirK has amino acid residues consistent with those of type 1 and 2 Cu-binding sites (His₁₀₆, His₁₄₀, and His₃₁₆ for type 1 Cu-binding sites and His₁₀₁, Cys₁₄₁, His₁₅₂, and Met₁₅₇ for type 2 Cu-binding sites) as well as the linker and tower loop regions, whereas the C terminus has unusual extensions of ~26 residues (Fig. 1B). These phylogenetic affiliations of and structural variations in Ns. viennensis NirK raise concerns regarding its enzymatic properties, such as specific enzymatic activity, affinity for NO₂⁻, and products of NO₂⁻ reduction.

Fig. 1.

Phylogeny (A) and sequence alignments (B) of prokaryotic NirK. A) A phylogenetic tree of prokaryotic NirK was constructed by the maximum likelihood method with the Jones-Taylor-Thornton model using the protein sequence of multicopper oxidase type 3 of Nitrososphaera viennensis (accession number; AIC14243.1) as an outgroup. Branching points that support a probability >80% in bootstrap analyses (based on 500 replicates) are shown as filled circles. The scale bar represents 10% sequence divergence. Sequence accession numbers are indicated in parentheses. B) Protein sequence alignment of nirK. NirK sequences were aligned using ClustalW software. Circles and triangles correspond to the amino acid residues of type 1 and 2 Cu-binding sites, respectively. Linker, Tower loop (3), and C-terminal extension regions are underlined. Abbreviations of microorganisms are as follows: Nitrosomonas europaea is N. europaea, A. xylosoxidans is Achromobacter xylosoxidans, A. faecalis is Alcaligenes faecalis, N. gonorrhoeae is Neisseria gonorrhoeae, S. thermophilus is Sphaerobacter thermophilus, and Ns. viennensis is Nitrososphaera viennensis.

Based on its unique sequence and lack of biochemical information, the purpose of the present study was to characterize Ns. viennensis NirK. Prior to the present study, we aimed to isolate Ns. viennensis NirK from a batch culture of Ns. viennensis as a native enzyme. However, the activity of aerobic ammonia oxidation often disappeared when we scaled up the cultures (data not shown). Additionally, a slow growth rate (μ_max 0.024 h⁻¹) (41) and low biomass concentration in the culture (ca. 10^7~8 cells mL⁻¹) further precluded the preparation of the biomass required for protein purification. Since recombinant NirK proteins have been successfully used to previously examine several enzymatic properties (7, 16, 32), the authors decided to heterologously express Ns. viennensis NirK in Escherichia coli, and investigate its enzymatic properties. The nirK gene located in the Ns. viennensis genome (accession number; CP007536.1) was cloned into the expression vector pCold I (Takara Bio, Shiga, Japan) with the 6×His tag using the Mighty cloning reagent set (Takara Bio), and transformed into E. coli strain BL21(DE3) (Takara Bio). The N-terminal region of Ns. viennensis NirK was predicted to be the signal peptide sequence (Met₁ to Ala₂₄), and nirK without the signal peptide sequence was amplified by PCR using ExTaq polymerase (Takara Bio) and specific forward (5′-GGCATATGGCCCCGACTGGTGTCACTAGACACTAT-3′) and reverse (5′-GGAAGCTTAACCAGAGGTGGTGTTGC CACCGGAGG-3′) oligonucleotide primers. The restriction sites of NdeI and HindIII in the forward and reverse primers above are underlined. Genomic DNA extracted from Ns. viennensis cells (JCM19564) was used as the DNA template for PCR. The constructed plasmid was subjected to Sanger sequencing, and no mutations were found in the sequence. Regarding the expression of the recombinant protein in E. coli cells, the expression culture was aerobically cultivated at 37°C in Luria-Bertani media containing 100 ng μL⁻¹ ampicillin. When the OD₆₀₀ of the culture increased to 0.4, the culture was transferred to 15°C and held for 30 min, and protein expression was then induced by adding isopropyl β-D-1-thiogalactopyranoside (IPTG) at a final concentration of 0.1 mM. After being incubated at 15°C for 24 h, cells were harvested by centrifugation at 8,500×g at 4°C for 10 min. The harvested cells were suspended in buffer containing 20 mM Tris HCl (pH 8), 200 mM NaCl, and 10% glycerol. The cells were disrupted using a sonifier 250 (Branson) (output 20, duty 20% for 60 s, 6 cycles), and centrifuged at 13,000×g at 4°C for 1 h. The supernatant was recovered as a soluble protein fraction, and the recombinant protein was purified using His-tag affinity chromatography. The recombinant protein was bound to His60 Ni Superflow resin (Takara Bio), and washed with washing buffer containing 20 mM Tris HCl (pH 8), 200 mM NaCl, 10% glycerol, and 20 mM imidazole. The bound recombinant protein was eluted with elution buffer containing 20 mM Tris HCl (pH 8), 200 mM NaCl, 10% glycerol, and 300 mM imidazole. Protein concentrations were measured using the DC-protein assay kit (Bio-Rad, Hercules, CA, USA) with bovine serum albumin as previously described (26), and purity was evaluated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on a 10% polyacrylamide gel as previously described (28). As shown in Fig. 2A, a single protein band appeared at a molecular mass of 40 kDa, which closely matched the molecular mass deduced from amino acid sequences of the recombinant protein (i.e., 39.7 kDa). The protein band was excised from the gel, and subjected to a matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) analysis after in-gel tryptic digestion for protein identification (The detailed methodology is described in the Supplementary text). The MALDI-TOF MS analysis confirmed that the protein band corresponded to Ns. viennensis NirK (Fig. S1). Regarding the reconstitution of Cu-binding sites of the recombinant protein, the purified recombinant protein was dialyzed against buffer containing 20 mM Tris HCl (pH 8), 300 mM NaCl, and 0.5 mM CuSO₄ at 4°C for 57 h. The protein solution was dialyzed again using the above Tris buffer without CuSO₄ at 4°C for 6 h. The dialyzed recombinant protein was concentrated using a Vivaspin column (MWCO; 30 kDa) (GE Healthcare Japan, Tokyo, Japan). The recombinant protein was loaded onto a gel-filtration HiLoad 16/600 Superdex 200 pg column (GE Healthcare) to assess the molecular mass of the recombinant protein, which was 105±1.3 kDa (Fig. 2B). Since the deduced molecular mass of Ns. viennensis NirK was 39.7 kDa, the molecular mass obtained by gel filtration indicated that the recombinant protein forms a homotrimeric structure, similar to canonical NirK.

Fig. 2.

Characterization of recombinant Nitrososphaera viennensis NirK. A) SDS-PAGE of the recombinant protein purified by His-tag affinity chromatography. B) Assessment of the molecular mass of the recombinant protein by gel filtration chromatography. Catalase from bovine liver (232 kDa), lactate dehydrogenase (140 kDa), and bovine serum albumin (66 kDa) were used to prepare a standard calibration curve. C) UV-VIS absorption spectra. The measurement was performed in a 20 mM Tris buffer (pH 8) containing 300 mM NaCl at 25°C. The solid line indicates the recombinant protein (1 mL mL⁻¹) oxidized with air. A 5×enlarged spectrum is also shown as a dashed line. D) ESR spectra. The measurement was performed using the recombinant protein (4.9 mg mL⁻¹) at −253°C.

NirK have been characterized as metalloproteins showing a blue or green color spectrum, and exhibit absorption peaks at approximately 450 and/or 600 nm (3). Bacterial NirK, which belong to the class 1 group, often show a maximum absorption peak at approximately 450 nm, although an exception (Achromobacter xylosoxidans NirK) that shows a peak at 593 nm has been previously reported (16). The purified recombinant protein was pale blue in color, and showed an absorption peak at 590 nm (Fig. 2C). This feature indicated that Ns. viennensis NirK is affiliated with the subgroup of NirK showing a blue color spectrum. The blue or green color spectrum of NirK is derived from a copper atom in the type 1 Cu-binding site (14), while the type 2 Cu-binding site does not contribute to the UV or visible spectrum. The type 2 Cu-binding site shows a characteristic electron spin resonance (ESR) spectrum (7, 16); therefore, an ESR analysis was performed using a JES-FA200 spectrometer (JEOL, Tokyo, Japan) to test for the presence of the type 2 Cu-binding site in the recombinant protein. An axial type 2 Cu signal (g_//=2.24, A_//=18.31 mT, and g_⊥=2.06) was found in the ESR measurement (Fig. 2D), indicating that the recombinant protein has a type 2 Cu-binding site coordinating with a copper atom. Additionally, we assessed the copper content of the recombinant protein by inductively coupled plasma mass spectrometry (ICP-MS). The copper content was found to be 2.9 atoms per subunit of the recombinant protein, indicating that Cu was fully incorporated into the recombinant protein. Overall, the recombinant protein shared the structural and spectroscopic features of class 1 and 2 bacterial NirK, which is consistent with sequencing information.

The kinetics of NO₂⁻ reduction were examined by anoxically incubating the recombinant protein at 25°C and pH 6.5 with ¹⁵NO₂⁻ and artificial electron donors as previously described (7). All of the buffers and stock solutions were prepared anoxically as previously described (27). Two milliliters of reaction buffer (20 mM phosphate buffer, 0.1 to 1.6 mM Na¹⁵NO₂⁻, 0.5 mM benzyl viologen (BV), and 0.24 mM sodium dithionite) was dispensed into a 1-cm sealable quartz cuvette and placed in an anaerobic chamber in which the O₂ concentration was maintained at lower than 1 ppm. BV was used as an artificial electron donor because it has been employed to examine the kinetics of the NO₂⁻ reduction of bacterial NirK (7, 13). The cuvette was set in a UV-VIS spectrometer UV-2700 (Shimadzu, Kyoto, Japan), and the initial absorbance of the prepared reaction mixture at a wavelength of 550 nm was approximately 2.0. The reaction was initiated by adding the recombinant protein (50 μL containing 250 μg of protein) using a gastight syringe, and the oxidation rate of reduced BV (molecular extinction coefficient, 10.4 mM⁻¹ cm⁻¹) (13) was monitored at 550 nm. The recombinant protein reduced NO₂⁻ by oxidizing BV, whereas no significant BV oxidation was found in the cuvette without the recombinant protein. The turnover number and K_m value for NO₂⁻ reduction by the recombinant protein were 3.1 s⁻¹ and 287 μM, respectively (Table 1), and the turnover number and affinity constant were markedly lower and higher, respectively, than those of other canonical NirK proteins, including those from AOB. The product of NO₂⁻ reduction by the recombinant protein was examined using phenazine methosulfate (PMS) as the electron donor instead of BV. When BV was used as the electron donor, NO₂⁻ was reduced to NO, and further reduced to ammonia (approx. 60% of consumed ¹⁵NO₂⁻) as observed in a previous study in which the NO₂⁻ reduction activity of A. xylosoxidans NirK was examined using methyl viologen (MV) as the electron donor (1). BV and MV have low redox potentials (−350 and −440 mV, respectively) (23), resulting in the reduction of NO to NH₃; therefore, PMS with a higher redox potential (+80 mV) was used in the present study. The recombinant protein was incubated as described above in a 1.8-mL gas-tight vial with the addition of 0.5 mM PMS and 5 mM ascorbic acid instead of BV and dithionite, and the production of ¹⁵N-labeled gaseous compounds (i.e., N₂, NO, and N₂O) in the headspace was examined by gas chromatography mass spectrometry (GC/MS) as previously described (27). The diluted gases of ^15-15N₂ (Cambridge Isotope Laboratories, Tewksbury, MA, USA), ¹⁴NO, and ^14-14N₂O (GL Science, Tokyo, Japan) were also analyzed to prepare standard curves for quantification. The recombinant protein reduced ¹⁵NO₂⁻ with the oxidation of PMS, and 38 and 48% of consumed ¹⁵NO₂⁻ were converted to ¹⁵NO and ^15-15N₂O, respectively. This is direct evidence to show that the recombinant protein is a NO-forming nitrite reductase. We found that the production of ^15-15N₂O was equal to the production of ¹⁵NO, which likely results from the reduction of ¹⁵NO₂⁻ to H¹⁵NO (i.e., NO₂⁻+2e⁻+3H⁺ → HNO+H₂O) and the chemical formation of ^15-15N₂O from the formed H¹⁵NO (i.e., 2HNO → N₂O+H₂O) (35), as previously observed for a sulfide-linked nitrite reductase (34).

Table 1.

Enzymatic properties of archaeal and bacterial copper-containing nitrite reductase (NirK). ND; not determined.

Organisms	MW* (kDa)	Cu content† (atom per subunit)	Absorption (nm)	Activity‡ Turnover (s−1)	Km (μM)	Reference
Archaeal NirK
Nitrososphaera viennensis	105±1.5	2.9	590
NO2− reduction				3.1	287	This study
NH2OH oxidation				0.039	97	This study
Bacterial NirK (NO2− reduction)
Nitrosomonas europaea	96	ND	450, 597	288	ND	18
Nitrosococcus oceani	114	1.67	455, 575	1,600	52	16
Achromobacter xylosoxidans	110	1.99	595	172	35	14, 32
Candidatus Jettenia caeni	101	ND	449, 598	319	250	7

^*Molecular weight (MW) of a trimeric NirK. The MW of Ca. Jettenia caeni NirK was calculated from amino acid sequences without a signal peptide sequence.

^†Copper contents previously assessed by chemical analyses were shown.

^‡The following electron donors were used to evaluate the turnover number of NO₂⁻ reduction; methyl viologen for N. europaea and Nc. oceani, pseudoazurine for A. xylosoxidans, and benzyl viologen for Ns. viennensis and Ca. Jettenia caeni NirK.

Aside from NO₂⁻ reduction, NH₂OH oxidation was also investigated using the recombinant protein because NH₂OH is produced as an intermediate during aerobic ammonia oxidation by AOA. The kinetics of NH₂OH oxidation were examined by aerobically incubating the recombinant protein (245 μg mL⁻¹) at 30°C and pH 7.5 with 0.5 mM NH₂OH, with dissolved oxygen being available as an oxidant. The reaction was initiated by the addition of NH₂OH solution, and the concentration of NH₂OH was assessed colorimetrically (5). The concentration of H₂O₂, which may be produced by the oxidase activity of NirK (12), was also evaluated colorimetrically using horseradish peroxidase (Wako, Osaka, Japan) and 3,3′,5,5′-tetramethylbenzidine (TMBZ) (Dojindo, Kumamoto, Japan) (2). As shown in Fig. S2, the recombinant protein oxidized NH₂OH with the production of H₂O₂. No NH₂OH oxidation or H₂O₂ production was observed when the incubation was repeated without the addition of the recombinant protein. The values for the turnover number and affinity constant for NH₂OH oxidation were 0.039 s⁻¹ and 97 μM (Table 1), respectively, and the value for the turnover number was two orders of magnitude lower than that observed for NO₂⁻ reduction; therefore, the recombinant protein catalyzed NO₂⁻ reduction more efficiently. The addition of cytochrome c from equine heart (1 mg mL⁻¹) or BV (0.5 mM) did not result in an increase in the reaction rate or affinity for NH₂OH oxidation. The product of NH₂OH oxidation by the recombinant protein was examined in a ¹⁵NH₂OH tracer experiment (29). The recombinant protein was incubated in a 1.8-mL gas-tight vial with the addition of 0.5 mM ¹⁵NH₂OH (Cambridge Isotope Laboratories) instead of ¹⁴NH₂OH. After a 2-h incubation, the concentrations of the ¹⁵N-labeled gaseous products were assessed by GC/MS. The recombinant protein oxidized ¹⁵NH₂OH and produced ¹⁵NO, ^15-15N₂O, and ^15-15N₂ gases quantitatively (Fig. 3), whereas the production of NO₂⁻ and NH₃ was not detectable (detection limits: 50 and 100 μM, respectively). The oxidation of NH₂OH to NO has been described in bacterial Hao (21); however, to the best of our knowledge, this is the first description of NH₂OH oxidation by NirK. We also observed ^15-15N₂O production from ¹⁵NH₂OH oxidation, which likely resulted from the oxidation of ¹⁵NH₂OH to H¹⁵NO and abiotic coupling of H¹⁵NO, as previously described. Notably, ^15-15N₂ was the major product of ¹⁵NH₂OH oxidation by the recombinant protein. Hydroxylamine disproportionation (30) may not be responsible for ^15-15N₂ production because NH₃ production was not detectable in the liquid phase. The molecular mechanisms underlying the oxidation of ¹⁵NH₂OH to ^15-15N₂ by the recombinant protein warrant further studies.

Fig. 3.

NH₂OH oxidation by recombinant Nitrososphaera viennensis NirK. The recombinant protein was incubated at 30°C and pH 7.5 in 1.8-mL vials (volume of the headspace: 1.5 mL), with i) 0.5 mM ¹⁵NH₂OH, ii) ¹⁵NH₂OH and ¹⁴NO₂⁻ (each 0.5 mM), or iii) ¹⁴NH₂OH and ¹⁵NO₂⁻. The production of N₂, NO, and N₂O in the headspace was examined by gas chromatography mass spectrometry (GC/MS). NH₃ and NO₂⁻ concentrations were also measured; however, they were not detectable during the incubation. During a 2-h incubation, i) 63±35 (mean±SD), ii) 149±1, and iii) 120±1 nmol N of NH₂OH were consumed in the liquid phase, resulting in 75–137% of the ¹⁵N-labeled nitrogen mass balance in the vials. Error bars represent the SD derived from triplicate incubations, and the graph bars represent the mean values. NH₂OH oxidation was not found in the vials without the addition of the recombinant protein.

We repeated the above incubation with the addition of NH₂OH and NO₂⁻ because both compounds are available in AOA cells during aerobic ammonia oxidation. Therefore, the above incubation was repeated with the addition of ¹⁵NH₂OH and ¹⁴NO₂⁻ (each 0.5 mM) or ¹⁴NH₂OH and ¹⁵NO₂⁻ (Cambridge Isotope Laboratories) (each 0.5 mM). In both cases, ^14–15N₂O was the major product (Fig. 3), indicating that the recombinant protein produces N₂O by oxidizing NH₂OH using NO₂⁻ as an electron acceptor. N₂O production by the denitrifier NirK from NH₂OH and NO₂⁻ has been previously described (10), and the N-nitrosation reaction is involved in N₂O production (39). Notably, Ns. viennensis cells produce N₂O when they are incubated aerobically with NH₃ and NO₂⁻ (42), although the Ns. viennensis genome lacks the gene encoding nitric oxide reductase (nor) that is involved in N₂O production from nitrifier-denitrification. Stieglmeier et al. (42) suggested the involvement of Ns. viennensis NirK in the production of N₂O in an Ns. viennensis culture, and our results support this hypothesis. Although the catalytic efficiency of Ns. viennensis NirK for NH₂OH oxidation was markedly lower than that of NO₂⁻ reduction (Table 1), Ns. viennensis NirK may act as an NH₂OH oxidase in Ns. viennensis cells and produce N₂O under oxic growth conditions. Aside from ^14–15N₂O production, the production of ¹⁵NO and ^15-15N₂O was also observed when the recombinant protein was incubated with ¹⁴NH₂OH and ¹⁵NO₂⁻ (Fig. 3).

Although the recombinant protein catalyzes NO₂⁻ reduction and NH₂OH oxidation, the catalytic efficiency of both reactions was low, as shown in Table 1. AOA nirK transcripts are abundant in the transcriptome (8, 11, 20, 37), suggesting the strong expression of AOA NirK in cells. NirK was the 225^th most abundant protein of the 1,503 proteins detected in the proteome of the late exponential phase of Ns. viennensis cells aerobically oxidizing ammonia (15). The strong expression of NirK appears to support the activity of NO₂⁻ reduction to NO as well as NH₂OH oxidation to NO by the low efficiency catalytic enzyme. Ns. viennensis NirK may function as a bifunctional enzyme that supplies NO molecules from 2 different sources (i.e., NH₂OH and NO₂⁻), which provides Ns. viennensis cells with a competitive advantage. In the present study, the enzymatic kinetics of recombinant Ns. viennensis NirK for NO₂⁻ reduction were examined using artificial electron donors; further studies are needed to identify physiological electron donors in Ns. viennensis cells. Bacterial NirK may accept electrons supplied from single-domain cupredoxin and cytochrome c (14, 25). A number of genes encoding single-domain cupredoxin were found in the Ns. viennensis genome (Table S1), whereas the ortholog of the gene encoding cytochrome c was not. To date, the biochemistry of AOA cupredoxin has not been investigated using natural enzymes and recombinant proteins, and our study provides basic information that furthers our understanding of the biochemistry of AOA.