• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. Feb 2007; 73(4): 1065–1072.
Published online Dec 15, 2006. doi:  10.1128/AEM.01978-06
PMCID: PMC1828659

Isolation of a Multiheme Protein with Features of a Hydrazine-Oxidizing Enzyme from an Anaerobic Ammonium-Oxidizing Enrichment Culture[down-pointing small open triangle]


A multiheme protein having hydrazine-oxidizing activity was purified from enriched culture from a reactor in which an anammox bacterium, strain KSU-1, was dominant. The enzyme has oxidizing activity toward hydrazine but not hydroxylamine and is a 130-kDa homodimer composed of a 62-kDa polypeptide containing eight hemes. It was therefore named hydrazine-oxidizing enzyme (HZO). With cytochrome c as an electron acceptor, the Vmax and Km for hydrazine are 6.2 ± 0.3 μmol/min · mg and 5.5 ± 0.6 μM, respectively. Hydrazine (25 μM) induced an increase in the proportion of reduced form in the spectrum, whereas hydroxylamine (500 μM) did not. Two genes coding for HZO, hzoA and hzoB, were identified within the metagenomic DNA from the culture. The genes encode the same amino acid sequence except for two residues. The sequences deduced from these genes showed low-level identities (<30%) to those of all of the hydroxylamine oxidoreductases reported but are highly homologous to two hao genes found by sequencing the genome of “Candidatus Kuenenia stuttgartiensis” (88% and 89% identities). The purified enzyme might therefore be a novel hydrazine-oxidizing enzyme having a critical role in anaerobic ammonium oxidation.

The anaerobic ammonium oxidation (anammox) process is a fascinating subject of study, not only because it is an innovative technological advance for the removal of nitrogenous contaminants from wastewater (14, 17, 27), but also because it includes interesting and novel biochemical reactions. In the anammox reaction, ammonium and nitrite contribute in equimolar amounts to the formation of dinitrogen gas (N2) according to the following equation (26): NH4+ + NO2 → N2 + 2H2O; G° = −358 kJ/mol ammonium.

This biological reaction is performed under anoxic conditions by novel autotrophic bacteria, which have not yet been isolated but which are identified as deeply branching planctomycetes, based on nucleotide sequences of 16S rRNA genes (17, 22, 26). The first anammox bacterium to be identified, discovered in The Netherlands, has been provisionally named “Candidatus Brocadia anammoxidans” (15). An increasing number of 16S rRNA gene sequences from other putative anaerobic ammonium-oxidizing bacteria have been reported to date from laboratories (6, 21, 29, 31) and from natural marine habitats such as the Black Sea (4, 5, 13, 16).

We designed a continuous-flow reactor shielded from light in which to enrich cultures for bacteria having the ability to catalyze the anammox reaction from a seed of denitrifying sludge originating in Kumamoto City, Japan. In the reactor, the culture turned a distinctly red color and yielded large amounts of sludge that could remove ammonium and nitrite with a relatively low production of nitrate at ratios strongly resembling that of the anammox reaction (20). To investigate the bacterial composition of the enriched culture community, 16S rRNA gene sequences were amplified by PCR and cloned. The dominant clones had an identical sequence, which has 92.2% identity with the 16S rRNA gene of “Candidatus Brocadia anammoxidans.” This strain was also confirmed to be dominant (about 70% of the total) in the culture community and was named KSU-1 (8).

It has been proposed that the anammox reaction consists of three enzyme reactions (16) which occur in an intracellular organelle, the anammoxosome (23, 24). Hydrazine, a unique intermediate of anammox, is formed from hydroxylamine and ammonium by a hydrazine-forming enzyme and sequentially oxidized to dinitrogen gas by a hydrazine-oxidizing enzyme generating four reducing equivalents. These electrons could conceivably be used to reduce nitrite to hydroxylamine by a nitrite-reducing enzyme according to the following equations: NO2 + 4e → NH2OH; NH2OH + NH4+ → N2H4; N2H4 → N2 + 4e, where e is an electron.

Schalk et al. purified and characterized a heme protein having both hydrazine- and hydroxylamine-oxidizing activities from an anammox enrichment culture under anaerobic conditions. They called it hydroxylamine oxidoreductase (HAO) because its hydroxylamine-oxidizing activity was the stronger. The HAO was demonstrated to be a 183-kDa homotrimer composed of a 58-kDa polypeptide containing approximately 26 hemes. Furthermore, the dithionite-reduced spectrum showed a unique peak at 468 nm (20). These features are similar to those of a well-studied HAO from an aerobic nitrifying bacterium, Nitrosomonas europaea (11). However, the HAO of the anammox bacteria was shown to catalyze the formation of dinitrogen gas from hydrazine and not to form hydrazine from hydroxylamine. Recently, an alternative model for the mechanism of anammox combined with an electron transport system has been proposed on the basis of relevant genes found in the genome of “Candidatus Kuenenia stuttgartiensis” (28). This model is as follows: NO2 + e → NO; NO + NH4+ + 3e → N2H4; N2H4 → N2 + 4e, where e is an electron.

In the new model of anammox catabolism, hydrazine is formed from nitric oxide and ammonium by a hydrazine hydrolase consuming three reducing equivalents. NirS, a nitrite-nitric oxide oxidoreductase, whose gene is found in the genome, catalyzes reduction of nitrite to nitric oxide by consuming one reducing equivalent. Eight HAO genes are found in the genome of “Candidatus Kuenenia stuttgartiensis.” In either model, HAOs are assumed to work by oxidizing hydrazine to dinitrogen gas, generating four reducing equivalents.

We have attempted to isolate and characterize an HAO that catalyzes the formation of dinitrogen gas from hydrazine from the enrichment culture where strain KSU-1 was dominant. The purified enzyme in this study catalyzed oxidation of hydrazine but was unable to use hydroxylamine as a substrate. Furthermore, its features were distinct compared to those of the HAOs reported by Schalk et al. (20) and of N. europaea (2, 11). Accordingly, we have tentatively named the enzyme hydrazine-oxidizing enzyme (HZO). In addition, two genes encoding HZO were identified, and this allowed clarification of the amino acid sequences.


Cultivation of anaerobic ammonia-oxidizing biomass.

A biomass capable of catalyzing anaerobic ammonia oxidation was enriched in a 15-liter reactor, using nonwoven porous polyester material with a pyridinium-type polymer as the support medium (9, 10). This culture was then used as a first seed for cultivation in a 50-liter reactor. The biomass was removed from the 50-liter reactor with the support materials every 4 months. About one-third of the support materials with the attached biomass (wet weight, 4 kg) were returned to the reactor equipped with new nonwoven materials as seeds for the next cultivation. The 50-liter reactor was continuously operated at 35°C in the upflow mode without recycling, as described previously (8). The influent was supplemented with NH4Cl and NaNO2 at several hundred mg of nitrogen liter−1, 125 mg liter−1 KHCO3, 54 mg liter−1 KH2PO4, and 9 mg liter−1 FeSO4 · 7H2O (with 5 mg liter−1 EDTA · 2Na as a chelating agent) mixed in groundwater containing approximately 1 mg N liter−1 of nitrate (19). At the time of sampling of the biomass, the ammonium and nitrite concentrations in the influent were 240 mg N liter−1, and the removal rate of total nitrogen from the medium was 1.0 g N liter−1day−1.

Preparation of cell-free extract and purification of HZO.

About 30 g (wet weight) of the biomass detached from the nonwoven materials was suspended in 20 mM potassium phosphate buffer (pH 7.5) containing 0.1 mM dithiothreitol and was subsequently disrupted by sonication and a Teflon homogenizer. To the homogenate was added sodium cholate to a concentration of 1% (wt/vol) and sodium deoxycholate to a concentration of 0.5% (wt/vol), and the sample was mixed, using a magnetic stirrer for 1.5 h at 4°C. The sample was then centrifuged at 15,000 × g for 15 min to remove cell debris and undisrupted cells. The prepared solution was used as a cell extract, which was further centrifuged at 160,000 × g for 1 h. The supernatant was diluted with the same volume of 20 mM potassium phosphate buffer (pH 7.5) lacking detergents and charged on a DE52 column (Whatman International Ltd., Kent, England) equilibrated with 20 mM potassium phosphate buffer (pH 7.5) lacking detergents, after which the gel was washed with the same buffer containing 0.2% sodium cholate and 0.1% sodium deoxycholate. A linear gradient of 0 to 0.5 M NaCl in buffer containing 0.2% sodium cholate and 0.1% sodium deoxycholate was used to elute HZO. The hydrazine- and hydroxylamine-oxidizing activities and the protein concentrations in each fraction were measured. The fractions showing relatively high levels of hydrazine-oxidizing activity were pooled, diluted, and charged on a second DE52 column equilibrated with 20 mM potassium phosphate buffer (pH 7.5) lacking detergents in order to completely remove any remaining HAO. The column was washed in the same way as the first DE52 column. A linear gradient of 0 to 0.35 M NaCl in buffer was used to elute HZO. Fractions having relatively high activities were pooled and concentrated by ultrafiltration and loaded on the top of a Superdex 200-pg gel filtration column (GE Healthcare Bio-Science Corp., Piscataway, NJ) equilibrated in 20 mM potassium phosphate buffer (pH 7.5) containing 0.5 M NaCl, 0.2% sodium cholate, and 0.1% sodium deoxycholate. The elution buffer was 20 mM potassium phosphate buffer (pH 7.5) containing 0.5 M NaCl, 0.2% sodium cholate, and 0.1% sodium deoxycholate. The fractions with relatively strong activity for the oxidation of hydrazine were pooled as Superdex eluates. The purity of the protein samples was evaluated by reverse-phase high-pressure liquid chromatography (HPLC) using a YMC-pack C4-AP column (AP-802; YMC Corp., Kyoto, Japan).

Determination of levels of oxidizing activities.

The standard reaction mixture for determination of hydrazine-oxidizing and hydroxylamine-oxidizing activities consisted of 100 mM potassium phosphate buffer (pH 8.0), 50 μM horse heart cytochrome c (oxidized form), an appropriate amount of enzyme solution, and 25 μM hydrazine or 500 μM hydroxylamine. The reactions were performed in 1-ml cuvettes sealed with butyl-rubber stoppers at 35°C under anoxic conditions, where argon was used to purge the oxygen. The reactions were followed by comparison of the increase in absorbance of cytochrome c at 550 nm in the standard mixture against that of a control without the enzyme, using a spectrophotometer (MPS-2400; Shimadzu Corp., Kyoto, Japan). The activities were expressed as μmol of cytochrome c reduced/min · ml enzyme solution.

Measurement of spectra of the purified HZO.

The absolute spectra of the purified proteins were recorded at 25°C, using a UV/visible spectrophotometer (Shimadzu MPS-2400) against the same buffer used for equilibration of the Superdex 200-pg column. The wavelength of the spectrophotometer was calibrated to within 0.2 nm using the emission lines of the deuterium lamp at 486.0 nm and 656.1 nm. The HZO was reduced with dithionite. The substrate-induced spectrum of the purified HZO was obtained by adding 25 μM hydrazine or 500 μM hydroxylamine.

Analysis of protein concentrations.

Protein concentrations were determined with the BCA protein assay kit (PIERCE, Rockford, IL), using bovine serum albumin as the standard (25).

Removal of heme from the purified enzyme.

Covalently bound c-type heme was removed using 2-nitrophenylsulfenyl chloride (Sigma-Aldrich, St. Louis, MO) according to the method of Fontana et al. (7). SDS-PAGE with a SH-reducing reagent was used to determine the molecular mass of the protein whose heme had been removed.

Amino acid sequence analysis.

Subsequent to the removal of hemes, the N-terminal amino acid sequence of HZO was determined. The protein was also digested with immobilized TPCK-trypsin according to the protocol provided by the manufacturer (PIERCE). The polypeptides generated were subjected to reverse-phase HPLC on a YMC-pack C4-AP column equilibrated with water containing 0.09% trifluoroacetic acid. The peptides were isolated with a linear gradient of the equilibration buffer and 80% acetonitrile containing 0.1% trifluoroacetic acid. The polypeptides eluted were detected by measuring the absorbance at 210 nm. The purified HZO and isolated polypeptides were sequenced using an amino acid sequencer (model 610 A; Applied Biosystems, Foster City, CA).

Cloning and sequencing of hzo genes and the 5′- and 3′-flanking regions.

Several degenerate oligonucleotide primers were designed based on the amino acid sequences of polypeptides generated by TPCK-trypsin digestion of purified HZO. Some combinations of primer pairs were used for PCR to amplify the internal regions of the hzo gene from metagenomic DNA extracted from the biomass. All PCR procedures were performed using a high-fidelity DNA polymerase, KOD-Plus (TOYOBO Co., Ltd., Osaka, Japan). The amplified fragment was cloned and sequenced. The fragment was labeled with digoxigenin and used as a probe in a Southern analysis of extracted metagenomic DNA to select restriction enzymes suitable for inverse PCR. The metagenomic DNA was digested with selected restriction enzymes, StuI or BanIII, and resolved by 0.7% agarose gel electrophoresis. The gel pieces containing the hzo gene were excised. DNA fragments extracted from the pieces were self-ligated and used as templates for inverse PCR. A forward primer, 5′-GTGTTCAAGCTTTGCTCA-3′ (the sequence within the amplified internal region of the hzo gene), and a reverse primer, 5′-CATTCACTGGTACCACAA-3′ (the complementary sequence of the locus upstream from the former), were used. The fragments amplified by inverse PCR were directly sequenced. On the basis of the sequences, an additional two pairs of PCR primers were designed to amplify StuI or BanIII fragments from the metagenomic DNA. Attempts were then made to clone the amplified fragments. The cloned StuI fragment was sequenced. Since a plasmid harboring the BanIII fragment yielded no transformants in repeated transformations, the BanIII fragment was determined by direct sequencing. These sequences were confirmed by repeated experiments.

Quantification of the relative amounts of the mRNAs of hzoA and hzoB.

A real-time PCR method with a fluorescence-quenching probe (Qprobe; J-Bio 21 Co., Tokyo, Japan) (32) was used for quantification of the mRNAs of hzoA and hzoB. cDNAs from the mRNA of hzoA and hzoB were synthesized using MultiScribe reverse transcriptase (Applied Biosystems), treated with RNase A and RNase H, extracted with phenol-chloroform, ethanol precipitated, and used as the templates for the real-time PCR. A reverse primer (5′-CTTCTTTCAGCATGAGAC-3′) was designed on the basis of a common sequence within the open reading frames (ORFs) of hzoA and hzoB, and the primer labeled at the 5′ terminus with a fluorescence reagent was purchased from J-Bio 21 Co. Forward primers, 5′-CTTTGTTTTATTGTTGCG-3′ and 5′-ATAATGCAATCGAGTGG-3′, which were designed on the basis of specific sequences upstream of the ORFs of hzoA and hzoB, respectively, were synthesized. PCR was performed in 50-μl reaction mixtures using KOD-Plus DNA polymerase with MicroAmp optical 96-well reaction plates, MicroAmp optical caps, and the PRISM 7700 Sequence Detection system (Applied Biosystems). To create a calibration curve for quantification of the transcripts, partial fragments of hzoA and hzoB (~150 bp) were amplified from a plasmid containing hzoA and a DNA fragment containing hzoB, respectively, using forward and nonlabeled reverse primers. The fragments were quantified and used to prepare DNA standards of known concentrations. The experimental procedure was performed according to the manufacturer's protocol supplied with the Qprobe PCR kit (J-Bio 21 Co.).

Peptide preparation for mass spectrometry (MS).

Lyophilized HZO (0.1 mg) was dissolved in 10 μl of 8 M urea and 10 mM dithiothreitol in 100 mM Tris-HCl buffer (pH 9.0) and incubated for 1 h at room temperature. To modify cysteine residues of HZO, the solution was incubated with 2 μl of 200 mM iodoacetoamide in 100 mM Tris-HCl buffer (pH 9.0) for 1 h at room temperature. To the reaction mixture was then added 4 μl of 200 mM dithiothreitol in 100 mM Tris-HCl buffer (pH 9.0), and the mixture was incubated for 1 h at room temperature. The mixture was then diluted with 85.5 μl of MilliQ water. The modified HZO was digested with 10 μl of 20 ng/μl Achromobacter protease I (ApI) (Wako Pure Chemicals, Ltd., Osaka, Japan), freshly prepared in 100 mM Tris-HCl buffer (pH 9.0) at 37°C overnight. After the digestion, formic acid was added to a final concentration of 0.1% (vol/vol). Prior to analysis by MS, the reaction mixture was desalted using a ZipTip C18 pipette tip (Millipore Corp., Billerica, MA). After being equilibrated with 0.1% (vol/vol) trifluoroacetic acid, the reaction mixture was loaded onto the tip and washed with an appropriate amount of equilibration solution. The bound peptides were eluted with 10 μl elution buffer consisting of 80% (vol/vol) acetonitrile and 0.1% (vol/vol) formic acid.

Peptide identification by mass spectrometry.

Peptide identification was achieved using electrospray ionization-quadrupole-time of flight-mass spectrometry. In brief, about 5 μl of the solution containing the ApI digests was loaded onto a HUNANO Tip (HUNANO Ltd., Hiroshima, Japan), which was placed on the spray head. Mass spectra were detected using a QSTAR XL system (Applied Biosystems) in reflection positive-ion mode and processed by Analyst QS 1.1 software. The signals detected were compared with the expected peptide mass, using the ProteinProspector program (http://prospector.ucsf.edu/).

Nucleotide sequence accession numbers.

The sequences of the 2.5-kb StuI and 3.4-kb BanIII fragments were registered in DDBJ with accession numbers AB257585 and AB255375, respectively.


Purification of a multiheme protein with hydrazine-oxidizing activity from an anaerobic ammonium-oxidizing enrichment culture.

A cell extract from a culture shown to contain an abundance of the KSU-1 strain (>70%) (8) was prepared in the presence of sodium cholate and sodium deoxycholate. The activity for the oxidation of hydrazine was detected at concentrations of hydrazine as low as 25 μM, while HAO activity was measured by adding hydroxylamine to a concentration of 500 μM in the reaction mixture. After centrifugation of the cell extract, the supernatant obtained was applied to a DEAE-cellulose (DE52) column. Figure Figure1A1A shows a typical elution profile of the first DE52 column that used sodium cholate and sodium deoxycholate, in which the peaks having hydroxylamine- and hydrazine-oxidizing activity were partially separated from each other. Fractions having relatively high levels of hydrazine-oxidizing activity were pooled and then charged on a second DE52 column. HZO was completely separated from HAO by the second DE52 column step (Fig. (Fig.1B).1B). These HZO fractions were pooled and loaded onto a gel filtration column (Superdex). The HZO eluted in fractions corresponding to a molecular mass of 130 ± 10 kDa and was collected and named Superdex eluate. The Superdex eluate exhibited only one peak by reverse-phase HPLC, indicating that it was homogeneous. The purity achieved by the purification process was assessed based on the specific activity of hydrazine oxidation and was shown to have increased 7.8-fold through the process (Table (Table1).1). This value suggests that up to 13% of the protein weight in the cell extract is HZO, assuming that HAO does not contribute to hydrazine oxidation under the analytical conditions.

FIG. 1.
Elution profiles of the DE52 column. (A) First column (2.2 × 33 cm; fraction volume, 8 ml). Fractions 26 to 31 were collected and applied to a second column. (B) Second column (1.8 × 30 cm; fraction volume, 8 ml). Fractions 34 to 40 were ...
Purification of hydrazine-oxidizing enzymea

Molecular mass of the purified enzyme.

The purified enzyme was treated with 2-nitrophenylsulfenyl chloride to remove covalently bound hemes and was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The gel shows a band with a molecular mass of 62 ± 2 kDa (Fig. (Fig.2).2). The enzyme is therefore likely to be a homodimer composed of an approximately 62-kDa polypeptide.

FIG. 2.
SDS-PAGE profile of HZO with heme removed. HZO with the heme removed was analyzed by 10% SDS-PAGE and stained with Coomassie brilliant blue G250. Lane 1, molecular mass markers; lane 2, 10 μg of HZO from which the hemes were removed, using 2-nitrobenzenesulfenyl ...

Activities of the purified protein.

The kinetic properties of the purified HZO were determined at pH 8.0 and 35°C with hydrazine or hydroxylamine as the electron donors. All of the electron acceptors used except for NAD+ could be used as electron acceptors for the oxidation of hydrazine, with horse heart cytochrome c showing the highest level of activity at 5.8 ± 0.3 μmol/min · mg. In addition to cytochrome c, alternative electron acceptors used were 0.5 mM phenzine methosulfate plus 4 mM methylthiazolyltetrazolium bromide (1.8 ± 0.1 μmol/min · mg); 1 mM ferricyanide (3.1 ± 0.1 μmol/min · mg); 75 μM 2,6-dichlorophenol-indophenol (2.5 ± 0.3 μmol/min · mg); and 1.0 mM NAD+ (<0.2 μmol/min · mg). No oxidation of hydroxylamine was observed with any of the acceptors. The use of higher concentrations of electron acceptors than those listed above did not affect the levels of activity. Km and Vmax were calculated using a nonlinear curve-fitting program, GraphPad Prism 4 (GraphPad Software, San Diego, CA), based on the Michaelis-Menten equation with cytochrome c as the electron acceptor. The Vmax and Km values for hydrazine were 6.2 ± 0.3 μmol of cytochrome c reduction/min · mg (1.6 ± 0.1 μmol of hydrazine oxidation/min · mg) and 5.5 ± 0.6 μM, respectively. Using an estimated molecular mass of 130,000 Da, the turnover rate, kcat, of HZO for oxidation of hydrazine was calculated as 1.7 × 102 min−1. This is comparable with that for HAO from another anammox bacterium (20). The Michaelis-Menten constant (Km) of the purified HZO for hydrazine is 5.5 μM, and thus the calculated value of kcat/Km is 5.1 × 105 M−1s−1.

Hydroxylamine was tested for an inhibitory effect on the activity of HZO. Several concentrations of hydroxylamine (1.0, 2.0, 3.0, and 5.0 μM), which is not a substrate of HZO, were added into the standard reaction mixture containing 5.0 or 10 μM hydrazine as a substrate. This resulted in 29%, 38%, 53%, and 66% inhibition when 5.0 μM hydrazine was used and 18%, 28%, 33%, and 43% inhibition when 10 μM hydrazine was used. These results demonstrate that hydroxylamine competes with hydrazine, and the Ki of HZO for hydroxylamine was estimated to be 2.4 ± 0.4 μM.

The enzyme was kept in 50 mM potassium phosphate buffer and Tris-HCl buffer at various pHs at 35°C for 10 min in order to study the effect of pH on the activity of the enzyme. These experiments demonstrated that the enzyme is stable between pH 6.0 and 9.0. Furthermore, oxidation of hydrazine occurred with relatively strong activity between pH 7.5 and 8.5 (data not shown).

Absorption spectra of the purified enzyme.

The air-oxidized spectrum of the enzyme gave absorption peaks at 408 nm and 531 nm. The Soret absorption peak of the oxidized form shifted to 419 nm, and additional peaks emerged at 552 nm (α-band), 523 nm (β-band), and 472 nm upon reduction with dithionite. Hydrazine (25 μM) induced the appearance of Soret peaks at the 419-nm, 524-nm, and 553-nm peaks, indicating reduction of the c hemes of the enzyme. In contrast, hydroxylamine at 500 μM induced only a shoulder around 419 nm and small peaks at 527 nm and 555 nm (Fig. (Fig.3).3). Hydroxylamine at 25 μM gave the same spectrum (data not shown). To determine the heme content of HZO, a pyridine hemochrome spectrum was measured according to the method reported by Berry and Trumpower (3). Using a millimolar extinction coefficient of 23.97 mM−1cm−1 at 550 nm, as described in the report (4), and an estimated molecular mass of 130 kDa (molecular mass of the homodimer), the heme c content of the enzyme was calculated to be 16.3 ± 0.1 (n = 2).

FIG. 3.
UV/visible absorption spectra of purified HZO. The concentration of HZO was 40 μg/ml. Symbols: , oxidized with air; - - - - , 25 μM N2H4 added; ...., 500 μM NH2OH added; An external file that holds a picture, illustration, etc.
Object name is zam0040775410007.jpg, reduced with dithionite. (A) Spectra in the range ...

The N-terminal sequence of the protein with heme removed was determined. The sequence obtained was VEIITHXVPH. Furthermore, amino acid sequences of polypeptides generated by TPCK-trypsin digestion were analyzed. The sequences of four polypeptides that were determined are as follows: AEPNPTGXDTXXGNN (peptide 1), (T or G) GEWLDQLTGPY (peptide 2), DXEAYDIGL (peptide 3), and SVXDDXHSPR (peptide 4). Degenerate oligonucleotide primers were designed based on these sequences, and PCR was performed using combinations of these primers. Only one primer pair, 5′-GCNGARCCNAAYCCNAC-3′ (based on the AEPNPT in peptide 1) and 5′-CCDATRTCRTANGCYTC-3′ (complementary nucleotide sequence designed based on the EAYDIG in peptide 3), where N = A+G+C+T, R = A+G, Y = C+T, and D = A+G+T, led to a successful amplification. The metagenomic DNA was digested with EcoRI, BanIII, SphI, NcoI, and StuI and subjected to Southern analysis to select restriction enzymes suitable for inverse PCR. Every sample gave two positive bands with almost the same intensities, implying that two hzo genes existed in the metagenomic DNA (data not shown). The individual shorter bands of StuI- or BanIII-digested fragments (about 2.5 kb or 3.5 kb, respectively) were separated from the other longer bands (more than 8 kb or 10 kb). Because our preliminary Southern analysis suggested that the shorter bands of StuI- or BanIII-digested fragments were contained within distinct regions in the metagenomic DNA, fragments produced by these enzymes were selected for inverse PCR. Approximately 2.4-kb and 3.2-kb fragments were amplified by inverse PCR from StuI and BanIII samples, respectively. The sequences differed significantly from each other except in the ORFs of the hzo genes. Two additional primer pairs were designed to recognize the 5′ and 3′ termini of the 2.5-kb StuI and 3.4-kb BanIII fragments in order to amplify and clone these fragments from the metagenomic DNA. Cloning was successful in the case of the 2.5-kb StuI fragment, while there were no transformants produced in repeated transformations of the 3.4-kb BanIII fragment. Thus, the 3.4-kb BanIII fragment was analyzed by direct sequencing. The hzo genes present in the StuI and BanIII fragments were named hzoA and hzoB, respectively.

The DNA sequences of the ORFs of hzoA and hzoB were identical except for four nucleotides. The sequences of the upstream regions of the ORFs were distinct and differed even at the first nucleotides preceding the translation initiation codons. The sequences of the downstream regions remained identical for 51 bp after the translation termination codons. With the exception of two residues, the amino acid sequences encoded by hzoA and hzoB are identical. As expected, the deduced amino acid sequences of hzoA and hzoB had sequences consistent with the sequences of the N termini and the four peptides generated by the digestion of HZO with TPCK-trypsin. Both hzoA and hzoB coded for polypeptides of 568 residues with the N-terminal-32 residues considered the leader peptides (Fig. (Fig.4).4). The molecular mass of mature HZO (536 residues) was calculated as 60,841 Da (HZOA) and 60,871 Da (HZOB), values which are consistent with the 62-kDa molecular mass of the monomeric polypeptide as determined by SDS-PAGE (Fig. (Fig.2).2). The amino acid sequences of the enzyme were compared with sequences translated from the nucleotide database and in the protein databases SwissProt, PDB, PIR, and PRF, using BLAST. HZO shares 88% and 89% sequence identity, respectively, with CAJ71439 and CAJ72085, which are deduced from ORFs (kustc0694 and kustd1340, respectively) in the genome of “Candidatus Kuenenia stuttgartiensis” (28), and less than 30% with HAOs from the Methylococcus capsulatus strain Bath (AAU92745) (29%), Silicibacter pomeroyi DSS-3 (YP 165030) (25%), Nitrosomonas sp. Strain ENI-11 (BAA82703) (25%), and N. europaea ATCC19718 (NP842336) (28%).

FIG. 4.
Alignment of the amino acid sequences of the HZO proteins and highly homologous HAO proteins of “Candidatus Kuenena stuttgartiensis.” Conserved positions are indicated by asterisks, and gaps are indicated by dashes. Peptides whose amino ...

Confirmation of the occurrence of transcription and translation of hzo genes.

To investigate which of the two genes is transcribed, transcripts were quantified by real-time PCR with specific primers for hzoA and hzoB. For hzoA, the cycle threshold (CT) was 21.9 ± 0.2, and the relative amount of mRNA was 16%; for hzoB, the CT was 19.2 ± 0.1, and the relative amount of mRNA was 84%. Each value includes the standard deviation for the results of three experiments. Standard curves were determined according to the formula CT = A × log(DNA amount) + B, where A and B, respectively, were −3.80 and 18.1 and −3.72 and 18.2 for hzoA and hzoB, respectively. Thus, the amount of the transcript of hzoB was five times higher than that of hzoA, indicating that, of the two genes, it is hzoB that is expressed at higher levels. These different levels of expression may be due to different promoter efficiencies. Because one residue near the C termini differs between the two gene products, the peptides generated from the digestion of HZO with a lysylendopeptidase, ApI, could be analyzed by MS to confirm the presence of the products of both hzo genes. The C-terminal 16-mer polypeptides with 4+ charged states, LEELGMRHESHGGAHH553-568 (calculated m/z, 449.96) and LEELGMRHESHGSAHH553-568 (calculated m/z, 457.46), were detected as shown in Fig. Fig.5.5. Fragments with other charged states (3+ and 5+) were also detected in the spectrum (data not shown). The peaks of HZOB were higher than those of HZOA in the 4+ state and in the other states.

FIG. 5.
Partial electrospray ionization-quadrupole-time of flight-mass spectrometry spectrum of peptides from the HZO digested with ApI. Peaks matching fragments of HZO digested with ApI are labeled with the charge state and their sequences. The m/z values of ...


Both hydrazine- and hydroxylamine-oxidizing activities were detected in cell extracts from the KSU-1 strain. These activities could be clearly separated by DEAE columns, implying that these activities arise from distinct enzymes (Fig. (Fig.1A).1A). The HZO activity could be measured at a low concentration of hydrazine (25 μM). If the hydrazine-oxidizing activity was a property of HAO, it would be at much lower levels than the hydroxylamine-oxidizing activity, when the levels of both activities of HAO reported in an anammox bacterium are taken into consideration (20). However, levels of hydrazine-oxidizing activity comparable to the levels of hydroxylamine-oxidizing activity were actually observed in the cell extract from the KSU-1 strain. Thus, most of the hydrazine-oxidizing activity seems to be from HZO, while the hydroxylamine-oxidizing activity is a result of HAO. These results indicate that HZO and HAO occur independently in the bacterium. There is a possibility that the HZO is derived from partially decomposed HAO or that the quaternary structure of the HAO molecule is altered by the detergents used in the preparation of the cell extract used for going from homotrimer to homodimer. To investigate the effects of detergents on both activities, the cell extract was prepared without detergents or with a more-potent detergent, namely 1.0% Triton X-100 (data not shown). In every case, the hydrazine- and hydroxylamine-oxidizing activities were at the same levels as observed when 1.0% sodium cholate and 0.5% sodium deoxycholate were used.

The purified HZO displays features which differ considerably from those of previously reported HAOs (11, 20) (Table (Table2).2). These differences include the fact that HZO is a homodimeric protein while HAOs are trimeric. Furthermore, HZO has a high affinity for hydrazine (Km, 5.5 μM) compared to that for an HAO from an anammox bacterium (Km, 18 μM) (20). It was previously reported that hydrazine accumulates in the culture broth after the addition of hydroxylamine to the anammox enrichment culture and gradually decreases after the disappearance of hydroxylamine (33). This phenomenon has been explained by the previous model in which hydroxylamine is a precursor of hydrazine. However, if the hypothesis that NO is a precursor of hydrazine but not hydroxylamine (28) is true, some other explanation for the phenomenon would be necessary. Such an explanation is offered by the low Ki value of HZO for hydroxylamine. Hydrazine would accumulate in the presence of hydroxylamine because hydroxylamine has a detrimental effect on hydrazine oxidation by HZO. The hydroxylamine added would be oxidized by HAO. When the concentration of hydroxylamine dropped down to near the Ki value of HZO, this enzyme's activity would be restored.

Features of HZO and HAO from the anammox biomass and N. europaea

The dithionite-reduced spectrum of HZO gave absorbance maxima at 419 nm, 552 nm, and 523 nm and a unique peak at 472 nm that is distinct from the peaks at 468 nm given by the HAO from an anammox enrichment culture (20) and at 460 nm given by the HAO of a nitrifying bacterium, N. europaea (1, 33). The addition of 25 μM hydrazine to the HZO preparation results in a shift of the maximum of the spectrum from 408 nm to 419 nm with the occurrence of α absorption (553 nm) and β peaks. In contrast, 500 μM of hydroxylamine gave a shoulder at 419 nm and small peaks of α and β absorption (Fig. (Fig.4).4). It is known that the HAO from N. europaea has four different classes of c-type cytochromes varying from −390 mV to +295 mV in redox potential (18). HZO contains 8.2 ± 0.1 of heme c per polypeptide, as determined by pyridine homochrome spectroscopy of HZO. This is supported by the presence of eight heme-binding motifs in the primary structure, as described below.

The deduced amino acid sequences from hzoA and hzoB show that neither the reported hydrophobic residues in the C-terminal domain in HAO of N. europaea (12) nor any transmembrane segment is present in HZO. This result suggests that HZO might interact indirectly with the membrane through some protein in the postulated electron transport system associated with the membrane. With DE52-column chromatography, if detergents were not used in the buffer, HZO adhered to the top of the DEAE-cellulose gel, giving it a deep-red color which could not be eluted with high-ionic-strength buffer (data not shown). HZO might aggregate physiologically with HAO and other proteins involved in electron transport through specific and hydrophobic interactions.

Eight heme-binding motifs, seven CXXCH, one CXXXXCH (30), and a tyrosine residue proposed to make a linkage to an adjacent P472 were conserved between the two HZO ORFs and two ORFs from “Candidatus Kuenenia stuttgartiensis” (Fig. (Fig.4).4). HZOA and HZOB exhibit about 90% identity to the deduced amino acid sequences encoded by these two ORFs from “Candidatus Kuenenia stuttgartiensis,” which are actually only two of eight presumed hao genes of this organism. This high level of identity is particularly interesting, given that the 16S rRNA gene has as low as 90% nucleotide identity with that of KSU-1 (AB057453). The presence of orthologous genes in both KSU-1 and “Candidatus Kuenenia stuttgartiensis” suggests that the gene products may generally be needed for anammox. HZOA and HZOB are presumed to exhibit the same activity, because only one residue near the C terminus is different between the mature proteins. In spite of a considerable difference in the amount of products, it was confirmed that both of the hzo genes were transcribed and translated (Fig. (Fig.5).5). The anammox bacterium might benefit from increasing the amount of HZO protein by expression of the two genes.

The physiological role of HAOs in “Candidatus Kuenenia stuttgartiensis” was postulated to be in catabolism, where they catalyze hydrazine oxidation to form dinitrogen gas, transferring four reducing equivalents to a specific electron acceptor, increasing transportation of protons across the anammoxosome membrane, and generating ATP through the resulting proton motive force when coupled to other enzymes and proteins involved in electron transfer (28). Only two of the eight HAOs of “Candidatus Kuenenia stuttgartiensis” (kustc0694 and kustd1340) show high identity to HZO from the KSU-1 strain. In the KSU-1 strain, HAO was found in amounts comparable to that for HZO, as estimated by their activities in the cell extract (Fig. (Fig.1A1A and Table Table1).1). Both the HZO and HAO appeared to be more than 10% of the total protein weight in the cells of the KSU-1 strain. It is highly likely that HAO and HZO play crucial roles in the catabolism of the anammox bacterium. Although HAO has not been completely purified from the KSU-1 strain, it has been shown that the protein is capable of using hydroxylamine and hydrazine as substrates (unpublished data). The purified HZO has distinctive features, particularly with regard to its ability to oxidize hydrazine at low substrate concentrations and its inability to oxidize hydroxylamine. These characteristics are different from those of HAOs purified from other origins (Table (Table2).2). HZO may be a more-specific molecule than HAO, which could be used for catalyzing denitrification at low concentrations of hydrazine.


This research was partially supported by the Ministry of Education, Culture, Sports, Science, and Technology of Japan [Grant-in-Aid for Basic Research (C) no. 16560687].


[down-pointing small open triangle]Published ahead of print on 15 December 2006.


1. Arciero, D. M., C. Balny, and A. B. Hooper. 1991. Spectroscopic and rapid kinetic studies of reduction of cytochrome c554 by hydroxylamine oxidoreductase from Nitrosomonas europaea. Biochemistry 30:11466-11472. [PubMed]
2. Arciero, D. M., and A. B. Hooper. 1993. Hydroxylamine oxidoreductase from Nitrosomonas europaea is a multimer of an octa-heme subunit. J. Biol. Chem. 268:14645-14654. [PubMed]
3. Berry, A. E., and B. L. Trumpower. 1987. Simultaneous determination of hemes a, b, and c from pyridine hemochrome spectra. Anal. Biochem. 161:1-15. [PubMed]
4. Bowman, J. P., and R. D. McCuaig. 2003. Biodiversity, community structural shifts, and biogeography of prokaryotes within Antarctic continental shelf sediment. Appl. Environ. Microbiol. 69:2463-2483. [PMC free article] [PubMed]
5. Dalsgaard, T., D. E. Canfield, J. Petersen, B. Thamdrup, and J. Acuna-Gonzalez. 2003. N2 production by the anammox reaction in the anoxic water column of Golfo Dulce, Costa Rica. Nature 422:606-608. [PubMed]
6. Egli, K., U. Fanger, P. J. Alvarez, H. Siegrist, J. R. van der Meer, and A. J. Zehnder. 2001. Enrichment and characterization of an anammox bacterium from a rotating biological contactor treating ammonium-rich leachate. Arch. Microbiol. 175:198-207. [PubMed]
7. Fontana, A., M. V. Francesco, and E. Boccu. 1973. Reaction of sulfenyl halides with cytochrome c. A novel method for heme cleavage. FEBS Lett. 32:135-138. [PubMed]
8. Fujii, T., H. Sugino, J. D. Rouse, and K. Furukawa. 2000. Characterization of the microbial community in an anaerobic ammonium-oxidizing biofilm cultured on a nonwoven biomass carrier. J. Biosci. Bioeng. 94:412-418. [PubMed]
9. Furukawa, K., Y. Ichimatsu, C. Harada, S. Shimozono, and M. Hazama. 2000. Nitrification of polluted urban river waters using zeolite-coated nonwovens. J. Environ. Sci. Health A 35:1267-1278.
10. Furukawa, K., J. D. Rouse, U. Imajo, K. Nakamura, and H. Ishida. 2002. Anaerobic oxidation of ammonium confirmed in continuous flow treatment using a non-woven biomass carrier. Japn. J. Wat. Treat. Biol. 38:87-94.
11. Hooper, A. B., P. C. Maxwell, and K. R. Terry. 1978. Hydroxylamine oxidoreductase from Nitrosomonas: absorption spectra and content of heme and metal. Biochemistry 17:2984-2989. [PubMed]
12. Igarashi, N., H. Moriyama, T. Fujiwara, Y. Fukumori, and N. Tanaka. 1997. The 2.8 Å structure of hydroxylamine oxidoreductase from a nitrifying chemoautotrophic bacterium, Nitrosomonas europaea. Nat. Struct. Biol. 4:276-284. [PubMed]
13. Jetten, M. S., O. Sliekers, M. Kuypers, T. Dalsgaard, L. van Niftrik, I. Cirpus, K. van de Pas-Schoonen, G. Lavik, B. Thamdrup, D. Le Paslier, H. J. Op den Camp, S. Hulth, L. P. Nielsen, W. Abma, K. Third, P. Engstrom, J. G. Kuenen, B. B. Jørgensen, D. E. Canfield, J. S. Sinninghe Damste, N. P. Revsbech, J. Fuerst, J. Weissenbach, M. Wagner, I. Schmidt, M. Schmid, and M. Strous. 2003. Anaerobic ammonium oxidation by marine and freshwater planctomycete-like bacteria. Appl. Microbiol. Biotechnol. 63:107-114. [PubMed]
14. Jetten, M. S., M. Wagner, J. Fuerst, M. van Loosdrecht, G. Kuenen, and M. Strous. 2001. Microbiology and application of the anaerobic ammonium oxidation (′anammox') process. Curr. Opin. Biotechnol. 12:283-288. [PubMed]
15. Kuenen, J. G., and M. S. Jetten. 2001. Extraordinary anaerobic ammonium-oxidizing bacteria. ASM News 67:456-463.
16. Kuypers, M. M., A. O. Sliekers, G. Lavik, M. Schmid, B. B. Jørgensen, J. G. Kuenen, J. S. Sinninghe Damste, M. Strous, and M. S. Jetten. 2003. Anaerobic ammonium oxidation by anammox bacteria in the Black Sea. Nature 422:608-611. [PubMed]
17. Op den Camp, H. J., B. Kartal, D. Guven, L. A. van Niftrik, S. C. Haaijer, W. R. van der Star, K. T. van de Pas-Schoonen, A. Cabezas, Z. Ying, M. C. Schmid, M. M. Kuypers, J. van de Vossenberg, H. R. Harhangi, C. Picioreanu, M. C. van Loosdrecht, J. G. Kuenen, M. Strous, and M. S. Jetten. 2006. Global impact and application of the anaerobic ammonium-oxidizing (anammox) bacteria. Biochem. Soc. Trans. 34:174-178. [PubMed]
18. Prince, R. C., C. Larroque, and A. B. Hooper. 1983. Resolution of the hemes of hydroxylamine oxidoreductase by redox potentiometry and optical spectroscopy. FEBS Lett. 163:25-27. [PubMed]
19. Rouse, J. D., K. Sumida, K. Kida, and K. Furukawa. 1999. Maintainability of denitrifying granular sludge in soft to marginally hard waters in an upflow sludge-blanket reactor. Environ. Technol. 20:219-225.
20. Schalk, J., S. de Vries, J. G. Kuenen, and M. S. Jetten. 2000. Involvement of a novel hydroxylamine oxidoreductase in anaerobic ammonium oxidation. Biochemistry 39:5405-5412. [PubMed]
21. Schmid, M., S. Schmitz-Esser, M. Jetten, and M. Wagner. 2001. 16S-23S rDNA intergenic spacer and 23S rDNA of anaerobic ammonium-oxidizing bacteria: implications for phylogeny and in situ detection. Environ. Microbiol. 3:450-459. [PubMed]
22. Schmid, M., U. Twachtmann, M. Klein, M. Strous, S. Juretschko, M. Jetten, J. W. Metzger, K. H. Schleifer, and M. Wagner. 2000. Molecular evidence for genus level diversity of bacteria capable of catalyzing anaerobic ammonium oxidation. Syst. Appl. Microbiol. 23:93-106. [PubMed]
23. Sinninghe Damste, J. S., W. I. Rijpstra, J. A. Geenevasen, M. Strous, and M. S. Jetten. 2005. Structural identification of ladderane and other membrane lipids of planctomycetes capable of anaerobic ammonium oxidation (anammox). FEBS J. 272:4270-4283. [PubMed]
24. Sinninghe Damste, J. S., M. Strous, W. I. Rijpstra, E. C. Hopmans, J. A. Geenevasen, A. C. van Duin, L. A. van Niftrik, and M. S. Jetten. 2002. Linearly concatenated cyclobutane lipids form a dense bacterial membrane. Nature 419:708-712. [PubMed]
25. Smith, P. K., R. I. Krohn, G. T. Hermanson, A. K. Mallia, M. D. Gartner, M. D. Provenzano, E. K. Fujimoto, N. M. Goeke, B. J. Olson, and D. C. Klenk. 1985. Measurement of protein using bicinchoninic acid. Anal. Biochem. 150:76-85. [PubMed]
26. Strous, M., J. A. Fuerst, E. H. Kramer, S. Logemann, G. Muyzer, K. T. van de Pas-Schoonen, R. Webb, J. G. Kuenen, and M. S. Jetten. 1999. Missing lithotroph identified as new planctomycete. Nature 400:446-449. [PubMed]
27. Strous, M., J. Heijinen, J. G. Kuenen, and M. S. Jetten. 1998. The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microorganisms. Appl. Microbiol. Biotechnol. 50:589-596.
28. Strous, M., E. Pelletier, S. Mangenot, T. Rattei, A. Lehner, M. W. Taylor, M. Horn, H. Daims, D. Bartol-Mavel, P. Wincker, V. Barbe, N. Fonknechten, D. Vallenet, B. Segurens, C. Schenowitz-Truong, C. Medigue, A. Collingro, B. Snel, B. E. Dutilh, H. J. Op den Camp, C. van der Drift, I. Cirpus, K. T. van de Pas-Schoonen, H. R. Harhangi, L. van Niftrik, M. Schmid, J. Keltjens, J. van de Vossenberg, B. Kartal, H. Meier, D. Frishman, M. A. Huynen, H. W. Mewes, J. Weissenbach, M. S. Jetten, M. Wagner, and D. Le Paslier. 2006. Deciphering the evolution and metabolism of an anammox bacterium from a community genome. Nature 440:790-794. [PubMed]
29. Strous, M., E. van Gerven, J. G. Kuenen, and M. S. Jetten. 1997. Effects of aerobic and microaerobic conditions on anaerobic ammonium-oxidizing (anammox) sludge. Appl. Environ. Microbiol. 63:2446-2448. [PMC free article] [PubMed]
30. Takayama, Y., E. Harada, R. Kobayashi, K. Ozawa, and H. Akutsu. 2004. Roles of noncoordinated aromatic residues in redox regulation of cytochrome c3 from Desulfovibrio vulgaris Miyazaki F. Biochemistry 43:10859-10866. [PubMed]
31. Tal, Y., J. E. Watts, and H. J. Schreier. 2006. Anaerobic ammonium-oxidizing (anammox) bacteria and associated activity in fixed-film biofilters of a marine recirculating aquaculture system. Appl. Environ. Microbiol. 72:2896-2904. [PMC free article] [PubMed]
32. Tani, H., N. Noda, K. Yamada, S. Kurata, S. Tsuneda, A. Hirata, and T. Kanagawa. 2005. Quantification of genetically modified soybean by quenching probe polymerase chain reaction. J. Agric. Food. Chem. 53:2535-2540. [PubMed]
33. van Dongen, L. G. J. M., M. S. M. Jetten, and M. C. M. van Loosdrecht. 2001. The combined Sharon/Anammox process, p. 37-38. In J. Hammett and L. Buzzard (ed.), STOWA report, 2001 ed. IWA Publishing, London, United Kingdom.

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...