Juglone, a plant-derived 1,4-naphthoquinone, binds to hydroxylamine oxidoreductase and inhibits the electron transfer to cytochrome c₅₅₄

ABSTRACT

Nitrification is an essential biological process that converts ammonia into nitrate via nitrite. In the initial phase of nitrification, ammonia-oxidizing bacteria convert ammonia to hydroxylamine and subsequently to nitrite. The electrons generated in this process provide energy for cellular growth. In this study, we observed the inhibitory effects of plant-derived 1,4-naphthoquinones, including juglone, plumbagin, and 1,4-naphthoquinone, on nitrification in Nitrosomonas europaea, a typical nitrifying bacterium. These compounds disrupted the transfer of electrons produced during hydroxylamine oxidation to cytochrome c₅₅₄, a physiological electron acceptor. In vitro electron transfer experiments confirmed the function of these three 1,4-naphthoquinones as electron acceptors in hydroxylamine oxidation. The results indicate that these compounds interfere with electron transfer to cytochrome c₅₅₄ during the ammonia oxidation process leading to nitrification inhibition. Remarkably, lawsone, an isomer of juglone, exhibited no effect on the activity of hydroxylamine oxidoreductase, implying a correlation between the redox potential of each 1,4-naphthoquinone and their ability to hinder canonical electron transfer. Furthermore, X-ray crystallographic analysis revealed that juglone resides in close proximity to the catalytic heme P460 in hydroxylamine oxidoreductase. The hydrophobic environment surrounding the binding cavity appears critical for the selectivity of 1,4-naphthoquinones.

IMPORTANCE

Nitrification, the microbial conversion of ammonia to nitrate via nitrite, plays a pivotal role in the global nitrogen cycle. However, the excessive use of ammonium-based fertilizers in agriculture has disrupted this cycle, leading to groundwater pollution and greenhouse gas emissions. In this study, we have demonstrated the inhibitory effects of plant-derived juglone and related 1,4-naphthoquinones on the nitrification process in Nitrosomonas europaea. Notably, the inhibition mechanism is elucidated in which 1,4-naphthoquinones interact with hydroxylamine oxidoreductase, disrupting the electron transfer to cytochrome c₅₅₄, a physiological electron acceptor. These findings support the notion that phytochemicals can impede nitrification by interfering with the essential electron transfer process in ammonia oxidation. The findings presented in this article offer valuable insights for the development of strategies aimed at the management of nitrification, reduction of fertilizer utilization, and mitigation of greenhouse gas emissions.

INTRODUCTION

Nitrification, the microbial process that converts ammonia to nitrate via nitrite, plays a critical role in the nitrogen cycle within the environment. However, the extensive use of ammonium-based fertilizers in agriculture has been associated with adverse consequences related to nitrification. These fertilizers promote the leaching of nitrogen compounds, primarily in the forms of nitrite and nitrate, which then contaminates groundwater (1, 2). This reduces the availability of nitrogen for plants and causes the eutrophication of freshwater resources (3). Furthermore, nitrogenous fertilizers contribute to nitrous oxide emission, a greenhouse gas with approximately 300 times greater warming potential than carbon dioxide (4). Therefore, the mitigation of nitrification assumes an important role in reducing fertilizer usage in agricultural systems and preserving the global environment.

The oxidation of ammonia to nitrite is the primary and rate-limiting step in nitrification, which is essential for the biological nitrogen cycle (5). This process is facilitated by three distinct groups of microorganisms, ammonia-oxidizing bacteria (AOB), ammonia-oxidizing archaea (AOA), and complete ammonia-oxidizing (comammox) bacteria (6–8). Ammonia oxidation within AOB involves two enzymatic reactions. Initially, membrane-bound ammonia monooxygenase (AMO) catalyzes the conversion of ammonia to hydroxylamine (Fig. 1). Subsequently, hydroxylamine oxidoreductase (HAO), a periplasmic enzyme, further oxidizes hydroxylamine to nitrite (9). Cytochrome c₅₅₄ (Cyt c₅₅₄) functions as an electron acceptor in this process, receiving four electrons generated in each turnover of HAO (10). The electrons are then transferred to a quinone pool (QH₂) via membrane cytochrome c₅₅₂ (Cyt c_m552). Among these electrons, two are recycled back to AMO to sustain its constitutive activity, while the remaining two are transferred to the terminal oxidase via other cytochromes, generating NAD(P)H and ATP (11, 12). This redox cycle plays a crucial role in the growth of AOB by providing the necessary energy (13). Therefore, modulating AOB growth through the inhibition of electron flow presents a promising strategy to address the agronomic and environmental issues mentioned above.

Fig 1.

Schematic representation of the ammonia oxidation modules in ammonia-oxidizing bacteria. AMO, ammonia monooxygenase; HAO, hydroxylamine oxidoreductase; Cyt c₅₅₄; cytochrome c₅₅₄; Cyt c_m552, membrane cytochrome c₅₅₂; QH₂; quinol.

The molecular mechanism of HAO has been extensively elucidated through structural investigations (14–17). HAO from Nitrosomonas europaea cells (NeHAO) forms a heterocomplex with an associated protein, NE1300 (14, 15). Three copies of NeHAO are associated by a three-fold symmetry, with each NeHAO protomer containing seven canonical c-type hemes and an unusual c-type heme P460 modified by a covalent linkage to Tyr491 from the adjacent protomer. The electrons generated by the oxidation of hydroxylamine are recruited to the free octahedral coordination site of the heme P460 iron and then delivered to Cyt c₅₅₄ through the arrays of c-type hemes in HAO (16). While the structural characteristics of HAO have been extensively studied, the inhibitory mechanism of nitrification inhibitors has not been well understood. Only the structure of HAO-related oxidoreductase from Kuenenia stuttgartiensis bound to phenylhydrazine provides information on the inhibitory mechanism (15).

N. europaea has been used as a model bacterium in studies investigating nitrification, and several bioactive chemicals that inhibit the nitrification process in this microorganism have been identified within the root exudates of various plants. These secondary metabolites are termed biological nitrification inhibitors (BNIs) (18). Methyl-p-coumarate and methyl ferulate have been isolated from koronivia grass (Brachiaria humidicola), and sorgoleone and sakuranetin have been isolated from sorghum (Sorghum bicolor) (19, 20). Recently, Otaka et al. identified a 1,4-naphthoquinone derivative named zeanone from maize (Zea mays) as a BNI (21). 1,4-Naphthoquinones, secondary plant metabolites, exhibit versatile bioactivity in cellular processes such as protease, topoisomerase, and photosynthetic reactions (22–25). The interference of quinone-dependent electron transfer reactions is primarily attributed to their redox-active bicyclic structures. They accept one or two electrons leading to the formation of semiquinone radicals, which contribute to their physiological activities against various clinical and biological targets. Despite the definite inhibitory effect of these BNIs on nitrification, the specific interacting target and underlying molecular mechanisms of inhibition remain unclear.

In this study, we focused on plant-derived 1,4-naphthoquinones (Fig. 2) to investigate their effects on the activity of NeHAO. Additionally, we explored the potential interference of these compounds with canonical electron flow pathways during hydroxylamine oxidation. This observation suggests that the disruption of redox equilibrium and electron flow caused by 1,4-naphthoquinones could be associated with the nitrification inhibition in N. europaea. Furthermore, the crystal structure of NeHAO in complex with juglone was also determined to reveal the precise binding mode of juglone. The impact of juglone on NeHAO activity was discussed, with a specific emphasis on electron transfer.

Fig 2.

Chemical structures of 5-hydroxy-1,4-naphthoquinone (juglone), 2-hydroxy-1,4-naphthoquinone (lawsone), 5-hydroxy-2-methyl-1,4-naphthoquinone (plumbagin), and 1,4-naphthoquinone (1,4-NQ). Numbering of the carbon atoms of naphthoquinone moiety was displayed.

RESULTS

Effect of 1,4-naphthoquinones on the nitrification activity of N. europaea cells

To examine the effect of 1,4-naphthoquinone and its derivatives on nitrification activity, we quantified the consumption of ammonia and the production of nitrite in the medium with varying concentrations of 1,4-naphthoquinones. Fig. 3E shows the complete consumption of ammonium during a 3-day cultivation period, accompanied by a concomitant increase in nitrite concentration. By the third day of cultivation, 32.8 ± 0.5 mM nitrite, equivalent to the initial ammonium concentration in the medium, had been detected. This observation established a balanced mass equilibrium between ammonium within the medium and nitrite produced through the ammonium transformation. When N. europaea was cultivated in the presence of varying concentrations of 1,4-naphthoquinones, it was observed that 100 µM juglone exhibited complete inhibition on nitrite production, and 10 µM juglone resulted in an 85% reduction compared to the control. However, 1 µM juglone had no inhibitory effect on nitrite production (Fig. 3A). Similarly, 100 µM plumbagin and 100 µM 1,4-NQ inhibited nitrite production by 88% and 94%, respectively, whereas concentrations of 10 µM and 1 µM had no inhibitory effect (Fig. 3C and D). Lawsone displayed no inhibitory effect on nitrification at any tested concentrations (Fig. 3B).

Fig 3.

Cultivation of N. europaea in medium containing 1,4-naphthoquinones. Ammonium and nitrite concentrations in the culture broth were measured at various cultivation periods. Each culture medium was supplemented with different concentrations (1 µM, 10 µM, and 100 µM) of juglone, lawsone, plumbagin, and 1,4-naphthoquinone (1,4-NQ), respectively (A–D). Cultivation with 1% DMSO as control was also carried out (E).

Purification and properties of Cyt c₅₅₄ from N. europaea

Cyt c₅₅₄ was purified to homogeneity using a simplified three-step column chromatography procedure described in Materials and Methods, resulting in a single band on SDS-PAGE analysis (Fig. 4A). The purified Cyt c₅₅₄ exhibited an A406/A280 spectral absorbance ratio of 6.4, consistent with the previously reported ratio of 6.6 by Arciero et al. (26). Molecular mass determination by SDS-PAGE and gel filtration provided values of 28 kDa (Fig. 4A) and 22 kDa (data not shown), respectively, in close agreement with the calculated molecular weight of 23.6 kDa for the mature form.

Fig 4.

Characteristics of the purified Cyt c₅₅₄. (A) SDS-PAGE analysis of purified Cyt c₅₅₄ (0.4 µg per well). (B) Optical spectra of purified Cyt c₅₅₄. HAO (−): Oxidized form; HAO (+): Reduced form obtained through NeHAO reaction with hydroxylamine; DT (+): Fully reduced form obtained by adding 5 mM sodium dithionite; DT buffer: HEPES buffer containing 5 mM sodium dithionite. The inset shows a magnified view of the wavelength range from 500 nm to 600 nm.

Cyt c₅₅₄ was prepared in oxidized form under aerobic conditions. The UV-visible spectrum of purified Cyt c₅₅₄ displayed a prominent peak at 406 nm in its oxidized form [HAO (−)] and three peaks at 419.6 nm, 523.8 nm, and 553.3 nm in its reduced form [HAO (+)] (Fig. 4B). Notably, the fully reduced Cyt c₅₅₄ spectrum using sodium dithionite exhibited a distinctive shoulder at 430 nm on the Soret peak, a characteristic feature of reduced Cyt c₅₅₄ in N. europaea (27). Upon reduction of Cyt c₅₅₄, coupled with hydroxylamine oxidation in the presence of HAO (Fig. 4B), the intensity of the signal at 406 nm decreased, and new peaks emerged at 419.5 nm, 524.7 nm, and 555.1 nm with lower intensities compared to the fully reduced form. A shoulder on the Soret peak at 430 nm was also observed, consistent with the spectrum previously reported for the half-reduced state of Cyt c₅₅₄ (28).

Effect of 1,4-naphthoquinones on nitrite production by NeHAO

To examine the effect of 1,4-naphthoquinones on NeHAO activity, we quantified the nitrite production using hydroxylamine as a substrate (Fig. 5). In the absence of an electron acceptor, no conversion of hydroxylamine to nitrite was observed within a 60-min reaction period when 24 nM NeHAO was used (Fig. 5F). However, with the addition of an established artificial electron acceptor, 1-methoxy-5-methylphenazinium methylsulfate (PMS), into the reaction mixture (29), nitrite production exhibited a significant and dose-dependent increase (Fig. 5E). The addition of 1 µM PMS led to the enhanced activity, producing 137 ± 8.1 µM nitrite during the 60-min reaction. Among the evaluated 1,4-naphthoquinones, juglone, plumbagin, and 1,4-NQ demonstrated the ability to promote the nitrite production. Specifically, at a concentration of 100 µM, juglone, plumbagin, and 1,4-NQ yielded 88.0 ± 2.8 µM, 34.0 ± 0.8 µM, and 25.4 ± 0.8 µM nitrite, respectively, during the 60-min reaction (Fig. 5A, C, and D). However, lawsone did not exhibit any promotion of nitrite production (Fig. 5B).

Fig 5.

Time courses of nitrite production by the enzymatic reaction of NeHAO. The reaction mixture consisting of 1 mM hydroxylamine and 24 nM NeHAO was incubated with various concentrations of 1,4-naphthoquinones: 25 µM, 50 µM, and 100 µM juglone, lawsone, plumbagin, and 1,4-naphthoquinone (1,4-NQ), respectively. Additionally, a reaction with 1-methoxy-5-methylphenazinium methylsulfate (PMS) was carried out at concentrations of 0.25 µM, 0.5 µM, and 1 µM. Negative controls included a reaction without NeHAO [HAO (−)] and a reaction supplemented with NeHAO and 1% DMSO [HAO (+)]. Every data point and standard deviation were estimated from three technical repeats.

1,4-Naphthoquinones inhibited the enzymatic reduction of Cyt c₅₅₄

We investigated the effects of 1,4-naphthoquinones on the electron transfer efficiency from NeHAO to Cyt c₅₅₄, a physiological electron acceptor. Fig. 6 shows the differential absorbance of Cyt c₅₅₄ obtained by subtracting the spectra of oxidized Cyt c₅₅₄ from those of Cyt c₅₅₄ reduced by NeHAO. The magnitude of the differential absorbance value represents the extent of Cyt c₅₅₄ reduction facilitated by electrons supplied from NeHAO. Upon reduction of Cyt c₅₅₄ concurrent with hydroxylamine oxidation, the spectral difference displayed a negative peak at 390nm and positive peaks at 422nm and 554nm. The addition of 10µM juglone, plumbagin, and 1,4-NQ to the reaction solution diminished the spectral changes, indicating the inhibition of Cyt c₅₅₄ reduction by NeHAO. These three compounds displayed the dose-dependent inhibition of Cyt c₅₅₄ reduction; in contrast, lawsone did not inhibit the reaction at any tested concentrations (Fig. 6B).

Fig 6.

Inhibition of Cyt c₅₅₄ reduction by 1,4-naphthoquinones. Differential spectra were obtained by subtracting the spectra of oxidized Cyt c₅₅₄ from those of Cyt c₅₅₄ reduced by NeHAO. The reactions were carried out with a range of 0–10 µM concentrations of 1,4-naphthoquinones, (A) juglone, (B) lawsone, (C) plumbagin, and (D) 1,4-naphthoquinone (1,4-NQ).

1,4-Naphthoquinones inhibited electron transfer to resazurin in vitro

NeHAO activity was evaluated by monitoring the fluorescence intensity of resorufin, which was produced through the concurrent reduction of resazurin and oxidation of hydroxylamine catalyzed by NeHAO (Fig. 7). Juglone, plumbagin, and 1,4-NQ exhibited inhibitory effects on this reaction, showing a dose-dependent manner with IC₅₀ values of 5.5 nM, 23.7 nM, and 9.8 nM, respectively. In contrast, lawsone exhibited substantially lower inhibition compared to the other 1,4-naphthoquinones.

Fig 7.

Inhibition of electron transfer to resazurin by 1,4-naphthoquinones. The enzymatic reduction of resazurin, coupled with the oxidation of hydroxylamine by NeHAO, was measured in the presence of different concentrations of 1,4-naphthoquinones, juglone, lawsone, plumbagin, and 1,4-naphthoquinone (1,4-NQ). Each data point and standard deviation were estimated from four technical repeats.

The structure of juglone-bound NeHAO

We determined the structures of juglone-bound NeHAO and apo-NeHAO at resolutions of 2.78Å and 1.99Å, respectively. The backbone structure of both proteins was identical, with a root mean square deviation (RMSD) of 0.260Å for 1,522 Cα atoms. The structure of juglone-bound NeHAO provided insights into the binding manner of juglone to NeHAO (Fig. 8A). Similar to the naphthoquinones found in close proximity to b-type hemes in oxidoreductases CYP158A2 and quinol-fumarate reductase (30, 31), juglone molecules resided in the vicinity of the catalytic hemes P460 in NeHAO at two of the three binding pockets on trimer interfaces. The third site was occupied by polyethylene glycol derived from crystallization reagent. There were no significant rearrangements of the residues in the catalytic pocket upon juglone binding, compared to apo-NeHAO. The catalytic site of NeHAO was primarily composed of hydrophobic and aromatic residues. Heme P460 and Thr238 were located within 4.0Å distance from juglone mediating van der Waals interactions (Fig. 8A). Additionally, Tyr358 and Trp221 interacted with the benzoquinone moiety of juglone by T-shaped π-stacking, and Phe448 was involved in π-π interaction with juglone. Hydrophobic and aromatic interactions with juglone have been observed in the structures of acetylcholine esterase AChE and β-hydroxyacyl-ACP dehydratase FabZ (32, 33). The apo-form NeHAO structure clearly exhibited the electron density of the NE1300 C-terminal seven residues (Asp85´ to Tyr91´). This C-terminal loop interacted with NeHAO (Fig. 8B and C) and resided at a significant distance from NeHAO’s catalytic center (Fig. 8A), implying that NE1300’s C-terminal loop does not impede substrate access.

Fig 8.

The structure of NeHAO. (A) The binding manner of juglone shown in yellow. The residues involved in the recognition of juglone were displayed by sticks. The polder map (contoured at 3.5 σ) (34) of juglone was represented by mesh (blue). (B) Surface representation of each protomer of NeHAO (green) and NE1300 (orange). Eight hemes of NeHAO and six C-terminal residues of NE1300 were designated as sticks. (C) The interaction network of the C-terminal part of NE1300. The residues of NeHAO and NE1300 were represented by sticks colored blue and orange, respectively. Polar interaction network is indicated by dashed lines.

DISCUSSION

Oxidation of hydroxylamine constitutes a vital reaction supplying the energy required by ammonia-oxidizing bacteria for their growth. In this study, we have discovered that 1,4-naphthoquinones, namely juglone, plumbagin, and 1,4-naphthoquinone (1,4-NQ), exhibit inhibitory effects on nitrification in N. europaea (Fig. 3) and disrupt the reduction of Cyt c₅₅₄, a physiological electron acceptor of NeHAO (Fig. 6). Additionally, these naphthoquinones can act as electron acceptors, facilitating the turnover of NeHAO and promoting the conversion of hydroxylamine to nitrite (Fig. 5). Similarly, inhibition of resazurin reduction suggests that these compounds interfere with electron transfer from NeHAO to resazurin (Fig. 7). In conclusion, juglone, plumbagin, and 1,4-NQ attract electrons from the inherent electron transfer process between NeHAO and Cyt c₅₅₄. Consequently, the diminished electron transfer to Cyt c₅₅₄ probably attenuates the energy production system, thereby impeding the growth of N. europaea.

Naturally occurring 1,4-naphthoquinones are known for their diverse bioactivities and significance in ecology and pharmacology, e.g., cytotoxic, anticancer, antibacterial, and antiparasitic activities (22). The bioactive functions of 1,4-naphthoquinones are attributed to their electrophilic properties. These compounds undergo radicalization and transition to semiquinones upon one-electron reduction, contributing to their biological activities (35, 36). The redox potentials of the 1,4-naphthoquinones used in this study were ranked as follows: juglone (−95 mV) >1,4-NQ (−140 mV) and plumbagin (−156 mV) >lawsone (−415 mV) (35). Notably, the redox potentials of these compounds were correlated with the extent of nitrification inhibition (Fig. 3). Due to its higher redox potential, juglone is deduced to be a functional electrophilic compound, effectively inhibiting electron transfer to Cyt c₅₅₄. In contrast, lawsone showed no significant effect on NeHAO activity in any of the conducted assays. Electron transfer from the catalytic heme P460 with a redox potential of −260 mV (37) to juglone, 1,4-NQ, and plumbagin is electrostatically feasible. However, only lawsone exhibits a redox potential lower than that of heme P460, suggesting that spontaneous electron transfer is electrostatically improbable.

To understand the electron flows, we created a superposed model of NeHAO complexed with juglone (PDB ID 6M0P) and hydroxylamine (PDB ID 4N4O) (Fig. 9A). The carbonyl oxygen of juglone was positioned at a distance of 3.9 Å from the nitrogen atom of hydroxylamine. Previous studies have proposed that hydroxylamine undergoes oxidation through the {FeNO}⁷ and {FeNO}⁶ nitrosyl states in the catalytic cycle of HAO (16). Subsequently, the electrons released from hydroxylamine then pass through the array of internal c-type hemes, eventually reaching Cyt c₅₅₄ (17) (Fig. 9B). However, in the presence of 1,4-naphthoquinones, the electrons generated by the NeHAO reaction would be directly transferred to the 1,4-naphthoquinones instead of being transferred to the physiological electron acceptor Cyt c₅₅₄ through an intermolecular electron transfer pathway (Fig. 9B).

Fig 9.

The structure of NeHAO in complex with juglone. (A) The superposed model of NeHAO in complex with juglone (PDB ID 6M0P) and hydroxylamine (PDB ID 4N4O). (B) The predicted electron transfer pathway involving the array of hemes and the direct path of electron to juglone. Each was shown by a red arrow. (C) Surface hydrophobicity (38) of NeHAO and NE1300. Hydrophobic (red) to non-hydrophobic (white) gradient was illustrated. (D) The hydrophobic residues in the vicinity of the juglone were designated as sticks. The C2 atom of the juglone was represented as a ball.

Our experimental data showed that lawsone did not affect NeHAO activity or electron transfer (Fig. 5B and 7). The structure of juglone-bound NeHAO provides valuable insights into the selectivity of 1,4-naphthoquinones. The juglone binding site exhibits a highly hydrophobic nature (Fig. 9C), with the C2 and C3 atoms of juglone directed toward a hydrophobic patch formed by Trp221, Tyr358, Phe360, and Val361, creating a T-shaped π-stacking with Tyr358. This hydrophobic patch could also accommodate the 2-methyl group of plumbagin enabling a CH-π interaction with the aromatic ring. In contrast, the 2-hydroxy group of lawsone may prevent its binding due to its direction toward the hydrophobic region (Fig. 9D). The hydrophobic environment at the entrance of the catalytic site appears essential in the selectivity of 1,4-naphthoquinones. The structural incompatibility and lower electrophilicity of lawsone provide a clear explanation for its inability to inhibit NeHAO.

This study elucidates the mechanisms by which plant secondary metabolites inhibit nitrification and provides crucial possibilities for developing strategies to control the biological nitrification process. These inhibitors are expected to contribute to the efficient utilization of nitrogen fertilizers and the mitigation of greenhouse gas emissions.

MATERIALS AND METHODS

Chemicals

5-Hydroxy-1,4-naphthoquinone (juglone) (Merck, Germany), 2-hydroxy-1,4-naphthoquinone (lawsone) (Tokyo Chemical Industry; TCI, Japan), 5-hydroxy-2-methyl-1,4-naphthoquinone (plumbagin) (TCI), and 1,4-naphthoquinone (1,4-NQ) (TCI) were dissolved in dimethyl sulfoxide (DMSO) (FUJIFILM Wako Chemicals, Japan) immediately prior to use. 1-Methoxy-5-methylphenazinium methylsulfate (PMS) (Dojindo, Japan) was dissolved in ultrapure water. Sodium dithionite (Na₂S₂O₄) (TCI) was dissolved in ultrapure water immediately prior to use.

Culture conditions of N. europaea

The culture conditions of N. europaea NBRC 14298 were described previously (39). A portion of the original culture was stored in the dark at 4°C and used as a seed culture within 2 weeks. For protein purification, the cultivation was carried out in a 10-L tank with air ventilation at 26°C. The cells were stored at −80°C until use.

Cultivation of N. europaea with 1,4-naphthoquinones

An aliquot of N. europaea seed cultures stored at 4°C was inoculated into 100 mL of media in a 250-mL Erlenmeyer flask with an air-vented cap and incubated for 3 days at 26°C with shaking at 160 rpm. Cells were harvested by centrifugation at 2,600g for 20 min at 26°C and resuspended in fresh medium at OD₆₀₀ of 0.05. Each 1,4-naphthoquinone dissolved in sterilized DMSO was added to 3 mL of cell suspension in a 14-mL culture tube with an air-vented cap at concentrations of 1 µM, 10 µM, and 100 µM. The final concentration of DMSO was 1%. The cells were incubated at 26°C with reciprocal shaking at 160 rpm, and an aliquot of the culture broth was tested for the quantitation of ammonium and nitrite concentrations, respectively. As the culture broth became acidic due to ammonium consumption, the pH was adjusted between 7.0 and 8.0 by adding 2M NaOH.

Analytical methods

The concentration of nitrite was determined colorimetrically using Griess reagents (40). In a 96-well microplate, 50 µL of Griess reagent A (58 mM sulfanilamide and 1.2 M HCl) was added to 100 µL of appropriately diluted samples, followed by the addition of 50 µL of Griess reagent B [0.96 mM N-(1-naphthyl)ethylenediamine]. After a 10-min incubation at 25°C, the absorbance at 540 nm was measured using the Infinite M1000 PRO microplate reader (TECAN, Switzerland). The nitrite concentration was calculated using sodium nitrite as standard. The concentration of ammonium in the culture broth was determined using the LabAssay Ammonia test kit (FUJIFILM Wako Chemicals) following the manufacturer’s instructions.

Purification of NeHAO

The purification of NeHAO was performed following the previously described procedure (39). For crystallization, the purified enzyme was concentrated to an absorbance of approximately 100 at 409 nm using Amicon Ultra 50K MWCO (Merck) to exchange the buffer with 25 mM HEPES-KOH (pH 7.5) and 25 mM KCl. The samples were collected and stored at –80°C until use.

Purification of Cyt c₅₅₄

All chromatography procedures were performed using the ÄKTA pure system (Cytiva, Japan) at 4°C. For the purification of Cyt c₅₅₄ from N. europaea cells, a simplified method was implemented based on a previously described procedure (26, 27). The frozen cell pellet from 20 L of cultivation was suspended in 40mL of buffer A (50mM Tris-HCl, pH 7.5) and disrupted by sonication. The suspension was then centrifuged at 40,000g for 40min at 4°C. The supernatant was applied to a HiTrap Q HP 5-mL column (Cytiva) equilibrated with buffer A. The flowthrough fraction was collected and applied to a HiTrap SP HP 5-mL column (Cytiva) equilibrated with buffer A. The column was washed with 10 column volumes of buffer A, and subsequently, Cyt c₅₅₄ was eluted with a linear NaCl gradient from 0 mM to 500mM. The fractions containing Cyt c₅₅₄ were concentrated to 0.5mL by ultrafiltration and subsequently applied to a Superdex 200 Increase 10/300 GL gel filtration column (Cytiva) equilibrated with buffer (10mM Tris-HCl, pH 7.5, 150mM NaCl). Fractions were pooled and concentrated to an absorbance of approximately 10 at 406nm in 50mM Tris-HCl and 150mM NaCl (pH 7.5) by ultrafiltration using Amicon Ultra 10 K MWCO (Merck). The purified protein was stored at –80°C until use. The purity of proteins was evaluated by SDS-PAGE and UV-visible spectroscopy at each chromatography step. The absorbance of Cyt c₅₅₄ was monitored at 280 and 406nm wavelengths during the purification process.

Characterization of proteins

The protein concentration was determined using the Bradford method with bovine serum albumin as a standard (41). The molecular mass of purified Cyt c₅₅₄ was estimated by gel filtration chromatography. The column (Superdex 200 Increase 10/300 GL, Cytiva) was equilibrated with 20 mM Tris-HCl and 150 mM NaCl (pH 7.5) and was calibrated with conalbumin (75 kDa), ovalbumin (43 kDa), and ribonuclease A (13.7 kDa) as molecular weight standards. Isocratic elution was performed at a flow rate of 0.4 mL/min with detections at 280 and 406 nm. The molecular mass of the subunits was estimated by SDS-PAGE according to the method of Laemmli (42).

Spectroscopic analysis of Cyt c₅₅₄

UV-visible spectra were measured using a spectrophotometer (Cary 400 Bio, Agilent Technologies, USA) in the wavelength range of 320–600 nm with a bandwidth of 0.1 nm. The measurements were performed on a 50-µL solution consisting of 6 µM Cyt c₅₅₄, 50 mM HEPES-NaOH (pH 7.0), and 50 mM NaCl. To obtain the fully reduced Cyt c₅₅₄ spectrum, we added 50 mM sodium dithionite to the solution.

Reduction of Cyt c₅₅₄ by NeHAO with 1,4-naphthoquinones

For the enzymatic reduction of Cyt c₅₅₄ with NeHAO, a 50-µL reaction mixture was prepared, containing 50 mM HEPES-NaOH (pH 7.0), 50 mM NaCl, 6 µM Cyt c₅₅₄, 3 nM NeHAO, and 50 µM hydroxylamine. Additionally, 0.5 µL of 1,4-naphthoquinones dissolved in DMSO was added to the solution. The reactions were initiated by adding hydroxylamine and incubated for 5 min at 25°C. Then the spectrum of Cyt c₅₅₄ was measured in the wavelength range of 320–600 nm with a bandwidth of 1 nm using a spectrophotometer (DS-11 FX+, DeNovix, USA).

NeHAO activity assay

Nitrite concentrations were determined using a colorimetric assay in a 96-well plate format by diazotizing and coupling with Griess reagent (40). The reaction mixture consisting of 50 mM HEPES-NaOH (pH 7.0), 50 mM NaCl, and 24 nM NeHAO was prepared in a total volume of 80 µL. The mixture was then incubated with 1,4-naphthoquinones or PMS for 5 min at 25°C, with a final DMSO concentration of 1%. The reaction was initiated by adding 20 µL of 4 mM hydroxylamine and incubated for 60 min at 25°C. The concentration of produced nitrite was quantitatively measured using Griess reagent. The reaction was stopped at each interval by adding 50 µL of Griess reagent A followed by 50 µL of Griess reagent B. After a 10-min incubation at 25°C, the absorbance at 540 nm was measured using the Infinite M1000 PRO microplate reader (TECAN).

Electron transfer assay

The reduction of resazurin coupled with hydroxylamine oxidation was measured following the described procedure (43). Serial dilutions of each 1,4-naphthoquinone in DMSO were added to a reaction mixture consisting of 3 nM NeHAO, 50 mM sodium phosphate buffer (pH 7.0), 150 mM NaCl, 100 µM resazurin (FUJIFILM Wako chemicals), and 1 mM hydroxylamine in a total volume of 200 µL. The reaction was initiated by adding hydroxylamine and incubated at 25°C. The final DMSO concentration was set to 1%. The fluorescence intensity of the produced resorufin resulting from the reduction of resazurin was measured with an excitation wavelength at 562 nm and an emission wavelength at 592 nm using an Infinite M1000 PRO microplate reader (TECAN). The IC₅₀ value of each compound was estimated by fitting the inhibition data to a dose-dependent sigmoidal curve using GraphPad Prism 6 (GraphPad Software).

Crystallization and data collection

The purified NeHAO was used for crystallization by the hanging-drop vapor-diffusion method, in which 1.0µL of protein solution was mixed with an equal volume of the reservoir solution. The reservoir condition previously reported (44) was optimized to 0.1 M MES-NaOH (pH 6.3), 30% (v/v) PEG400, and 50mM potassium nitrate. The platelet crystals colored red-brown were grown to approximate dimensions of 0.3 × 0.3 × 0.2mm within 1 week at 20°C. To obtain the crystals of NeHAO in complex with juglone, we added a saturating concentration of juglone to the drop containing NeHAO crystals. The drop was then incubated for 1 day. A single crystal with soaked juglone was flash-cooled under a stream of liquid nitrogen at 100 K. The diffraction data from the apo- and juglone-bound forms of NeHAO were collected on beamlines NE3A and NW12A at PF-AR (Tsukuba, Japan), respectively. All diffraction data sets were processed using the XDS program suite (45), and data statistics are summarized in Table 1.

TABLE 1.

Data collection and refinement statistics

	apo-NeHAO	Juglone-bound NeHAO
Data collection
Wavelength (Å)	1.0000	1.0000
Space group	P212121	P21212
Unit cell parameters [a, b, c (Å)]	141.2, 141.9, 213.1	139.3, 141.1, 106.1
Resolution (Å)	50.0–1.99 (2.10–1.99)	50.0–2.78 (2.96–2.78)
Rmergea	0.198 (1.374)	0.368 (1.848)
Redundancy	7.0 (6.7)	7.8 (7.7)
Completeness (%)	99.8 (99.2)	99.9 (99.8)
No. of unique reflections	567,408 (91,130)	101,458 (16,383)
Mean I/σ (I)	8.3 (1.3)	6.7 (1.3)
CC1/2	0.995 (0.544)	0.980 (0.622)
Refinement
Resolution (Å)	46.18–1.99	46.77–2.78
Rwork / Rfree	0.163/0.200	0.256/0.292
No. of atoms
Protein	27,024	13,300
Water	2,501	114
Ligand	2,064 (c-type HEM)	1,032 (c-type HEM)
	174 (polyethylene glycol)	49 (polyethylene glycol)
		26 (juglone)
RMSDs
Bond length (Å)	0.011	0.003
Bond angle (°)	1.445	1.018
Ramachandran (%)
Favored	97.28	95.86
Allowed	2.69	3.96
Outliers	0.03	0.18
Mean B factors (Å2)
Protein	33.49	65.06
Water	38.90	28.32
Ligands	24.52 (c-type HEM)	45.70 (c-type HEM)
	56.08 (polyethylene glycol)	62.83 (polyethylene glycol)
		61.95 (juglone)
PDB ID	6M0Q	6M0P

^a R_merge = Σ_hklΣ_i|I_i(hkl) – <I_i(hkl)>| / Σ_hklΣ_iI_i(hkl), where i is the number of observations of a given reflection and I(hkl) is the average intensity of the i observations. R_free was calculated with a 5% fraction of randomly selected reflections evaluated from refinement. The highest resolution shell is shown in parentheses.

Structure determination and refinement

The structures of the apo- and juglone-bound NeHAO were determined by the molecular replacement method using the AutoMR program in the Phenix program package (46) with NeHAO (PDB ID 4N4N) as a search model. Rotation and translation functions were calculated using data of 45.0–3.5 Å resolution. Several rounds of refinement were performed using the Phenix program package and Refmac5 in the CCP4 program package (47, 48), with iterative manual fitting and rebuilding based on 2F_o – F_c and F_o – F_c electron densities using COOT (49). Subsequently, waters, ligands, and polyethylene glycols were built based on 2F_o – F_c and F_o – F_c electron densities. The final refinement statistics and geometry defined by MolProbity (50) are summarized in Table 1.

Structural characterization and preparation of figures

The structural figures were all prepared with PyMOL (https://www.pymol.org).

DATA AVAILABILITY

The coordinates of apo-NeHAO and juglone-bound NeHAO have been deposited in the Protein Data Bank under PDB ID 6M0Q and 6M0P, respectively.