Is there a pathway for N₂O production from hydroxylamine oxidoreductase in ammonia-oxidizing bacteria?

The nitrogen cycle is a key biogeochemical cycle on Earth, as nitrogen is an essential nutrient required for all life-forms. The largest source of bioavailable nitrogen originates from biological and synthetic nitrogen (N₂) fixation, whereby N₂ is converted to ammonia (NH₃), which can then be incorporated into biomass by plants. In the developing world, soils often do not contain enough nitrogen to give large crop yields, so fertilizers, produced by synthetic nitrogen fixation, are critically important to supplement this lack of nitrogen. On the other hand, in the developed world, overfertilization is a common problem (1). In the soil, NH₃-oxidizing bacteria (AOB) and archaea compete for the uptake of NH₃ with plants, aerobically oxidizing it to nitrite (${\text{NO}}_2^{-}$) or nitrate (${\text{NO}}_3^{-}$). The first step of this process is catalyzed by ammonia monooxygenase, producing hydroxylamine (NH₂OH). Hydroxylamine oxidoreductase (HAO) then further oxidizes NH₂OH to ${\text{NO}}_2^{-}$ (2). These oxidation reactions provide soil microbes with reducing equivalents for ATP synthesis. Complementary to this process is denitrification, which ultimately reduces ${\text{NO}}_2^{-}$ back down to N₂ during anaerobic respiration, producing nitric oxide (NO) and nitrous oxide (N₂O) in intermediate steps (3).

Release of both NO and N₂O during these transformations is of global consequence, as NO participates in the depletion of the ozone layer, whereas N₂O has become the third most significant greenhouse gas, with a global warming potential that is 300 times that of CO₂ (4). With the increase in the use of fertilizer since the preindustrial era, global N₂O emissions have increased substantially. Understanding the processes that generate N₂O from fertilizer in soil and seawater is therefore essential in devising practical strategies to minimize its production.

AOBs are of particular interest in this regard, as they are capable not only of nitrification but also of denitrification, termed nitrifier denitrification. These AOBs have been implicated as significant contributors in increased N₂O emissions (5, 6). Previous work has found that AOBs express nitrite and nitrous oxide reductases under anaerobic conditions, protecting against ${\text{NO}}_2^{-}$ accumulation, and producing N₂O as a direct product (7). Other studies have implicated aerobic NH₂OH oxidation as another source of trace amounts of NO and N₂O (8). Typically, NH₂OH is oxidized to ${\text{NO}}_2^{-}$ by HAOs, as mentioned above. These enzymes contain a catalytic heme P460 cofactor in the active site. It has been postulated that HAOs could incompletely oxidize NH₂OH to HNO, which, after release from the active site, could then dimerize and dehydrate to form N₂O and H₂O. NO is another potential incomplete oxidation product of NH₂OH, which could serve as a substrate for denitrification to produce N₂O, or escape into the atmosphere (8). Now, work by Lancaster and coworkers (9) has revealed a potential pathway for the direct production of N₂O by HAOs.

To better understand NH₂OH oxidation, the Lancaster group studied the enzyme cyt P460 from the AOB Nitrosomonas europaea (10). N. europaea cyt P460 exists as a 36-kDa homodimer with a c-type heme in the active site that has an unusual N−C cross-link from the heme 13′ mesocarbon to the amine of Lys70 as shown in Fig. 1, Left (11). In contrast, the heme P460 cofactor in the active site of HAOs is doubly cross-linked by Tyr491 at the 5′ mesocarbon and the pyrrole α-carbon positions (Fig. 1, Right) (12). Despite this structural difference, studies on cyt P460s from other AOBs, Methylococcus capsulatus and Kuenenia stuttgartiensis, have broadly implicated cyt P460s in NH₂OH oxidation (13, 14). Nevertheless, it still remains an open question whether the results for the N. europaea cyt P460 discussed herein can simply be transferred to HAOs. Further studies on HAOs will be necessary to confirm this.

Fig. 1.

PyMOL depiction of the N. europaea cyt P460 active site [Protein Data Bank (PDB) ID code 2JE2, Left] and of the N. europaea HAO heme P460 active site (PDB ID code 4N4N, Right).

In their PNAS paper, Lancaster and coworkers (9) report that the aerobic oxidation of NH₂OH by cyt P460 does not result in stoichiometric conversion to ${\text{NO}}_2^{-}$ but, instead, leads to the production of 70% ${\text{NO}}_2^{-}$ and 30% N₂O. This finding prompted further studies under anaerobic conditions that point toward a direct enzymatic pathway for stoichiometric N₂O production from the cyt P460 catalyzed oxidation of NH₂OH. N₂O analysis of solutions of the enzyme by GC, with varying concentrations of either NH₂OH or the two-electron oxidant phenazine methosulfate (PMS), reveals the stoichiometry for this reaction to be 2:1 of NH₂OH and PMS to N₂O, respectively, suggesting that the reaction requires two equivalents of NH₂OH and four oxidizing equivalents to generate one molecule of N₂O,

$$2\hspace{0.17em}{\text{NH}}_2\text{OH}\to {\text{N}}_2\text{O}+{\text{H}}_2\text{O}+{4\text{e}}^{-}+{4\text{H}}^{+}.$$ [1]

Mechanistic details of this reaction were obtained by UV/visible and EPR studies (see Fig. 2). Spectra obtained for the as-isolated cyt P460 show the heme in the high-spin ferric state with a Soret band at 440 nm and g values of 6.57, 5.09, and 1.97. It is proposed that the active site is six-coordinate, with a likely solvent H₂O loosely bound at the active site (Fe^III−H₂O). Upon addition of NH₂OH to cyt P460, the 440-nm Soret band associated with the Fe^III−H₂O resting state shifts to 445 nm, and Q bands broaden with the formation of an isosbestic point at 438 nm, suggesting a single-step conversion. EPR of the product revealed a 95% conversion to a low-spin species, consistent with the displacement of the bound water molecule with the stronger field NH₂OH ligand. The remaining 5% are converted to an off-path and nonproductive species, a ferrous NO complex, or {FeNO}⁷ in Enemark−Feltham notation (15). Because the electronic structures of transition metal nitrosyls can be ambiguous (3), this notation is useful in counting electrons of transition metal NO complexes by treating the metal−NO unit as a single entity. Here, the superscript “7” corresponds to the sum of iron d- and NO π*-electrons.

Fig. 2.

Proposed mechanism of NH₂OH oxidation by N. europaea cyt P460. The square brackets indicate that the ferrous product has not been directly observed so far. Reproduced from ref. 9.

Addition of the oxidant [Ru(NH₃)₆]Cl₃ and excess NH₂OH to the Fe^III−NH₂OH complex results in the single-step formation of an intermediate, which exhibits a Soret band at 455 nm and is EPR-silent. This intermediate accumulates, suggesting that its decay is the rate-limiting step and, thereby, allowing it to be studied in detail. This species reacts with hydroxylamine in the next step of the reaction with a second-order rate constant, k_obs(2) = 0.07 mM⁻¹⋅min⁻¹.

UV/visible data indicate that this intermediate is oxidized by 3e⁻ relative to the Fe^III−NH₂OH adduct. Lancaster and coworkers propose that this species is a diamagnetic ferric NO complex, or {FeNO}⁶ in the Enemark−Feltham notation (see Fig. 2). This idea was confirmed by independently producing the {FeNO}⁶ species by reacting the resting Fe^III−H₂O complex with the NO donor PROLI-NONOate [1-(hydroxy-NNO-azoxy)-L-proline] in an NO-shunted pathway. The resulting complex has an identical UV/visible signature to the observed intermediate and was shown to be stable both in solution and in the presence of excess oxidant. Importantly, addition of varying concentrations of NH₂OH to the {FeNO}⁶ complex, generated by reaction of the enzyme with NO, shows the same second-order decay (0.07 ± 0.01 mM⁻¹⋅min⁻¹), confirming that the intermediate that accumulates in the rate-determining step of cyt P460 catalysis is, in fact, the {FeNO}⁶ complex.

To confirm that N₂O is the product of the reaction between the {FeNO}⁶ intermediate and NH₂OH, the Fe^III−H₂O complex was reacted with varying concentrations of NO (provided by PROLI-NONOate) and NH₂OH, and the N₂O yield was determined. A clear 1:1 stoichiometry is observed between the NO added (and, correspondingly, the {FeNO}⁶ generated) and N₂O produced. When using isotopically labeled ¹⁵NH₂OH, ^14/15N₂O is produced only in the presence of cyt P460, confirming that the N−N coupling reaction occurs between the {Fe¹⁴NO}⁶ complex and ¹⁵NH₂OH. The product of this coupling reaction should be a ferrous complex (see Fig. 2); however, this species was never observed directly. Under the reaction conditions used in this study, this species is rapidly oxidized to Fe^III due to the presence of excess oxidant (k_obs > 1,100 s⁻¹), which prevents its characterization.

Finally, Lancaster and coworkers examined the reversibility of NO binding in the NO shunt by generating the {FeNO}⁶ complex with one equivalent PROLI-NONOate. Upon exposure to O₂, the UV/visible features of the {FeNO}⁶ complex disappeared while the Fe^III−H₂O signal appeared. Additionally, ${\text{NO}}_2^{-}$ is produced as the oxidized product.

In summary, the paper by the Lancaster group (9) describes a new, direct pathway of N₂O production from cyt P460s via oxidation of NH₂OH that has thus far been overlooked in the field. This result implies that N₂O production by AOBs does not require nitrifier denitrification. In fact, these findings provide a rationale for previous reports of increased N₂O production under anaerobic conditions and high concentrations of NH₃, where nitrifier denitrification is not promoted (16). The second-order decay of the {FeNO}⁶ intermediate suggests that the N. europaea cyt P460 may be responsible for detoxifying high levels of NH₂OH. Taking these ideas a step further, it seems possible that ${\text{NO}}_2^{-}$ production by HAOs may not actually be the catalytic function of HAOs under high NH₂OH concentrations but, instead, that N₂O is the main product and that the observed nitrite is a byproduct of NO oxidation to ${\text{NO}}_2^{-}$ in the presence of O₂—a potentially game-changing result. These findings prompt further studies on HAOs to better understand their catalytic function and to determine how these enzymes regulate NO and ${\text{NO}}_2^{-}$ production. All of these results have dramatic implications for the function of HAOs in nitrification, the potentially underestimated role of nitrification in N₂O production, and the significance of NH₂OH in the nitrogen cycle. The mechanistic understanding of NH₂OH oxidation seems now essential in designing ways to minimize N₂O release from agricultural soils and wastewater treatment plants.