Nitric oxide is an obligate bacterial nitrification intermediate produced by hydroxylamine oxidoreductase

Article abstract of DOI:10.1073/pnas.1704504114

Ammonia (NH₃)-oxidizing bacteria (AOB) emit substantial amounts of nitric oxide (NO) and nitrous oxide (N₂O), both of which contribute to the harmful environmental side effects of large-scale agriculture. The currently accepted model for AOB metabolism involves NH₃ oxidation to nitrite (NO₂^–) via a single obligate intermediate, hydroxylamine (NH₂OH). Within this model, the multiheme enzyme hydroxylamine oxidoreductase (HAO) catalyzes the four-electron oxidation of NH₂OH to NO₂^–. We provide evidence that HAO oxidizes NH₂OH by only three electrons to NO under both anaerobic and aerobic conditions. NO₂^– observed in HAO activity assays is a nonenzymatic product resulting from the oxidation of NO by O₂ under aerobic conditions. Our present study implies that aerobic NH₃ oxidation by AOB occurs via two obligate intermediates, NH₂OH and NO, necessitating a mediator of the third enzymatic step.

Full text of DOI:10.1073/pnas.1704504114

Nitric oxide is an obligate bacterial nitrification
intermediate produced by hydroxylamine oxidoreductase
a
a,1
Jonathan D. Caranto and Kyle M. Lancaster
a
Department of Chemistry and Chemical Biology, Cornell University, Ithaca, NY 14853
Edited by Brian M. Hoffman, Northwestern University, Evanston, IL, and approved June 16, 2017 (received for review March 17, 2017)
Ammonia (NH
of nitric oxide (NO) and nitrous oxide (N
tribute to the harmful environmental side effects of large-scale
agriculture. The currently accepted model for AOB metabolism
3
)-oxidizing bacteria (AOB) emit substantial amounts
O), both of which con-
suggests that alternate biochemical production routes exist. NO and
O have been proposed to arise from the incomplete oxidation of
NH OH by HAO, in which NO is an HAO by-product that is then
reduced to N O by the enzyme NO reductase (8). The results
reported herein show that NO is not merely a by-product––it is
2
N
2
2
2
–
2
involves NH
termediate, hydroxylamine (NH
heme enzyme hydroxylamine oxidoreductase (HAO) catalyzes the
3
oxidation to nitrite (NO
) via a single obligate in-
–
2
.
2
OH). Within this model, the multi-
an obligate intermediate of NH
3
oxidation to NO
Several previously proposed NH
3
oxidation pathways based
–
2
four-electron oxidation of NH
that HAO oxidizes NH OH by only three electrons to NO under
both anaerobic and aerobic conditions. NO
activity assays is a nonenzymatic product resulting from the oxi-
dation of NO by O under aerobic conditions. Our present study
implies that aerobic NH oxidation by AOB occurs via two obligate
intermediates, NH OH and NO, necessitating a mediator of the
2
OH to NO
. We provide evidence
on whole-cell studies included NO and nitroxyl (HNO) as
2
obligate intermediates (6, 9–11). Consistent with the intermediacy
–
–
2
observed in HAO
of NH OH, AOB produce NO when NH3 is replaced with
2
2
NH OH in aerobic cell culture media (12). Intriguingly, under
2
2
anaerobic conditions and in the presence of an oxidant, whole-cell
–
3
NH OH oxidation produces not NO but NO and N O (9, 13).
2
2
2
2
These observations of N O were interpreted as evidence that
2
third enzymatic step.
HNO is an obligate intermediate because it can dimerize and
2
decompose to N O (14) or be oxidized to NO. Together, these
nitrification
| nitric oxide | enzymology | bioinorganic chemistry
findings necessitate an O -dependent, final oxidation step of
2
–
either HNO or NO to produce NO2
.
ynthetic nitrogenous fertilizers are necessary in agriculture to
sustain the growing human population, but their use causes
significant imbalance in the biogeochemical nitrogen cycle (1).
The application of ammonia (NH )-based fertilizers increases
concentrations of nitrite (NO
Both NH OH oxidase activity and anaerobic NO and N O
2
2
S
production were pinpointed to the enzyme HAO by activity-
guided protein purification (15–17). Under anaerobic condi-
tions, purified HAO oxidizes NH OH to NO and N O, but
2 2
3
–
–
3
–
) and nitrate (NO
) in the water
again, NO
conditions, at most, 60% of the NH
(16, 17). The remaining NH OH is converted primarily to NO
product not observed during NH or NH
2
is not observed (17). Under aerobic turnover
2
–
table. These species pollute drinking water and drive the eu-
trophication of lakes and estuaries. Moreover, elevated NH3
concentrations in soil have been linked to nitrous oxide (N
and nitric oxide (NO) emissions. N O is an ozone-depleting
2
OH is converted to NO
2
–
2
3
, a
2
O)
3
2
OH oxidation by intact
–
2
cells (17, 18). However, NO yields can be increased to 70–95% by
2
greenhouse gas with a global warming potential ∼300× greater
than that of carbon dioxide (2), and NO contributes to the
production of ground-level ozone and acid rain (3, 4). Bal-
ancing human needs with environmental impact requires an
intimate understanding of the biological pathways that produce
these pollutants (5).
adding diethyldithiocarbamate or high concentrations of ferricya-
nide to the reaction mixture (18, 19). Together, these observations
–
were interpreted as evidence that HAO oxidizes NH OH to NO
2
2
via enzyme-bound HNO and NO intermediates (Fig. 1, NH OH
2
obligate intermediate model) (6). Under this model, the absence of
Biological sources of NO and N
2
O include NH
3
-oxidizing
Significance
–
bacteria (AOB), which mediate the oxidation of NH
3
2
to NO .
The prevailing view of NH oxidation, based largely on studies of
3
The enzymatic reactions that occur during nitrification are na-
ture’s means to use ammonia as cellular fuel. Complete un-
derstanding of nitrification and related processes are vital to
sustainable agriculture and renewable energy technologies.
The prevailing view of the first phase of nitrification is that
ammonia oxidizing bacteria use two enzymes, ammonia mon-
ooxygenase and hydroxylamine oxidoreductase, to oxidize
ammonia to nitrite via hydroxylamine as an obligate in-
termediate. Our work reveals nitric oxide as an additional
obligate intermediate. The presented findings necessitate re-
vision of a key biogeochemical process, identify a new bio-
energetic role for nitric oxide, predict participation of a third
enzyme in the biological oxidation of ammonia to nitrite, and
will inform models toward sustainable agriculture.
the model AOB Nitrosomonas europaea, is that it occurs via a
two-step enzymatic process (6):
−
+
NH3 + O2 + 2e + 2H → NH2OH + H2O
[1]
[2]
NH2OH + H2O → NO2− + 4e + 5H
−
+
An integral membrane metalloenzyme, NH
AMO), catalyzes the dioxygen (O )-dependent hydroxylation
of NH to hydroxylamine (NH OH; Eq. 1). Two electrons are
3
monooxygenase
(
2
3
2
required to turn over AMO. NH
2
OH is then oxidized by four
–
electrons to NO
2
(Eq. 2) by a multiheme enzyme, hydroxylamine
oxidoreductase (HAO). Two of these electrons return to AMO,
leaving two net electrons to enter the respiratory electron trans-
port chain using O
2
as the terminal electron acceptor. Under
Author contributions: K.M.L. designed research; J.D.C. performed research; J.D.C. and
K.M.L. analyzed data; and J.D.C. and K.M.L. wrote the paper.
anaerobic conditions, AOB carry out nitrifier denitrification, in
–
2
which O is substituted by NO as the terminal electron acceptor
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
2
and is reduced to N O or dinitrogen (7). The obligate intermediates
2
1
of nitrifier denitrification are NO and N
2
O, both of which can
To whom correspondence should be addressed. Email: kml236@cornell.edu.
escape from cells and into the atmosphere. Emissions of NO and
N O have also been linked to aerobic NH oxidation (4, 8), which
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
1073/pnas.1704504114/-/DCSupplemental.
2
3
www.pnas.org/cgi/doi/10.1073/pnas.1704504114
PNAS Early Edition
|
1 of 6

–
O
2
precludes the final oxidation step to form NO
and HNO to dissociate from HAO pathway intermediates, resulting to a mixture of NO
in the observed NO and N O.
S1). We treated 150 nM HAO with 5 μM phenazine methosulfate
Recent studies of additional enzymes with HAO-like active
(PMS) and various concentrations of NH OH at pH 8.0. The
samples were incubated at room temperature for 30 min. On the
time scale of the experiment, O rapidly oxidizes reduced PMS
(k = 180 M ·s ) (27) and therefore can be used in catalytic
concentrations. All of the NH OH was consumed within 30 min.
concentration was determined via the Griess di-
2
, allowing for NO
18). Under aerobic turnover conditions, HAO oxidizes NH
2
OH
–
–
2
and NO (Fig. 2 and SI Appendix, Table
3
2
2
–
sites support the conflicting view that NO
2
is not the enzymatic
product of HAO activity. Kartal and coworkers (20) purified
kustc1061, an octaheme oxidoreductase homologous to HAO,
from the anammox bacterium Kuenenia stuttgartiensis. Kustc1061
2
–1 –1
2
–
stoichiometrically oxidizes NH
not NO
2
OH by three electrons to NO,
The NO
azotization assay. The combined NO2 – and NO
were determined by quantifying NO in parallel samples pre-
2
–
2
–
–
. Additionally, an unrelated monoheme enzyme called
3
concentrations
cytochrome (cyt) P460, so named because it bears an active-site
2
–
3
–
2
cofactor similar to the heme P460 center found in HAO, was also
treated with NO
sistently accounted for ∼40% of the added NH
Fig. S1). Almost all of the remaining NH OH was accounted
reductase. Observed NO
concentrations con-
2
–
OH (SI Appendix,
reported to oxidize NH
2
OH to NO
2
(22–25). We recently re-
vised the cyt P460 activity, demonstrating that N O is the enzy-
2
2
–
for as NO3
.
matic product under anaerobic conditions (21). Mechanistic
Omitting PMS or HAO from the reaction mixture causes
NH OH oxidase activity to decrease substantially (Fig. 2 and SI
Appendix, Table S1). In the absence of HAO, less than 5% of the
NH OH is consumed over a prolonged 5-h incubation time. Fur-
thermore, final NO
studies showed that cyt P460 catalyzes the oxidation of iron-
6
bound NH
2
OH to a ferric-nitrosyl [{FeNO} in Enemark–
2
Feltham notation (26)]. Nucleophilic attack on the cyt P460
6
{
FeNO} by a second molecule of NH
2
OH results in the for-
2
–
–
2
3
and NO concentrations are negligible, which
mation of N O. Under aerobic cyt P460 turnover conditions,
2
–
2
2
is consistent with the necessity of HAO for NH OH oxidase activity.
at most, 70% of NH OH was converted to NO , with the
2
When PMS is omitted, only ∼15% of the NH OH was consumed
remaining NH
2
OH being converted to N
2
O. We proposed that
2
–
6
over 5 h, most of which was converted to NO . This slower
2
NO dissociation from the {FeNO} intermediate competes with
2
NH OH consumption rate is consistent with the previously repor-
attack by NH OH and that the nonenzymatic oxidation of disso-
2
–
2
ted HAO activity with O as the sole oxidant (18).
2
ciated NO by O accounts for the observed NO . The similar
2
–
–
–
2
Under anaerobic conditions, neither NO2 nor NO3 are pro-
duced. In an anaerobic glovebox, 150 nM HAO was treated with
substoichiometric NH
2
OH oxidation to NO
observed for both
cyt P460 and HAO prompted us to revisit the activity of HAO.
2
2
00 μM PMS and 100 μM NH OH at pH 8.0 for 30 min at room
The work described herein uses NO scavenging assays as evi-
–
2
temperature. Under these conditions, NH OH is completely con-
2
– –
dence that NO, not NO
, is the product of NH
2
OH oxidation by
sumed and negligible concentrations of NO2 or NO3 were ob-
N. europaea HAO under both aerobic and anaerobic conditions.
served (Fig. 2 and SI Appendix, Table S1), which is consistent with
–
–
previous reports (15). The absence of NO
their production is O dependen–t.
The low stoichiometry of NO
2 3
or NO indicated that
Results and Discussion
HAO Does Not Produce NO
–
2
2
in the Absence of O
2
. To firmly establish
conditions under which NO2 is produced by HAO, we repeated
several previously reported HAO turnover experiments (15, 17,
–
2
produced from NH
OH strongly
2
–
suggests that NO
more NO
2
is not the product of HAO activity. Indeed,
–
–
3 2
is produced than NO , which is inconsistent with the
assignment of the latter species as the product of HAO. The ob-
–
–
served mixture of NO
NH OH under aerobic conditions is consistent with previous work
17, 18). At low O concentration, the reaction mixture is more
complex; Hooper and coworkers (17) showed that under these
2 3
and NO formed during HAO oxidation of
2
(
2
–
–
2 3 2
conditions, four products are observed—NO , NO , N O, and
NO. These observed product mixtures could be the result of com-
peting reactions of a common reactive species, which is the true
HAO enzymatic product. The observation of NO among the four
products at low O concentration strongly suggests that NO is this
2
–
2 2
reactive molecule. Reactions of NO that produce N O, NO , or
–
NO3 are well characterized; the latter two products require O for
2
their formation (28). We propose that HAO oxidizes NH OH to
2
–
NO and that NO
2
is not a direct enzymatic product but is instead, a
result of nonenzymatic oxidation of NO by O
2
. Consistent with this
hypothesis, neither NO2 nor NO3 is observed as a product of
NH OH oxidation by HAO under anaerobic conditions (Fig. 2
and SI Appendix, Table S1).
In the presence of O , NO has a short lifetime and thus, would
be difficult to detect as a product of HAO. Aqueous NO reacts
–
–
2
2
–
6
–2 –1
2 2
with O to form NO (4k = 8 × 10 M ·s ) (29). Despite this
reaction being second order with respect to NO, the half-life of
aqueous NO is 5.6–56 s when the NO concentration is 10–
1
00 μM under air-saturated conditions (230 μM O ). O has been
2 2
–
reported to react with reduced PMS to form superoxide (O )
2
Fig. 1. Two models for the obligate intermediates of NH
by AOB. The HAO NH OH oxidation products in each model are those pro-
posed in previous literature (NH OH obligate intermediate model) or this
work (NH OH/NO obligate intermediate model). Solid boxes surround obli-
3
oxidation to NO −
2
–
–
(
27). The subsequent reaction of O with NO to form NO is
2
3
2
9 –1 –1
above the diffusion-controlled limit (k = 4.3 × 10 M ·s ) (30).
2
Even without O , NO could still be difficult to detect; Pacheco
2
2
and coworkers reported that NO reacts with reduced HAO (k =
gate intermediates and products. Square brackets enclose proposed enzyme-
bound intermediates. Pathways in red are proposed nonenzymatic pathways
1
0
–2 –1
1
2
.5 × 10 M ·s ) (31); this could account for the N O produced
–
–
leading to N
2
O, HNO, NO, NO
3
, and nonenzymatic NO
2
production. E is a
.
under anaerobic or low O2 conditions (17). In light of these
potential side reactions, we quantified NO produced during
–
proposed NO oxidoreductase necessary to oxidize NO to NO
2
2
of 6
|
www.pnas.org/cgi/doi/10.1073/pnas.1704504114
Caranto and Lancaster

in absorption at 433 nm (ΔA433) resulting from NO binding to cat-
–1
–1
alase (Δe433 = 47.9 mM ·cm ) (32).
Samples containing catalase and HAO in anaerobic turnover
conditions exhibit features consistent with NO binding to cata-
lase (SI Appendix, Fig. S2). Samples were prepared containing
1
1
50 nM HAO, 100 μM 2,6-dichlorophenolindophenol (DCPIP),
0 μM catalase, and 200 μM NH OH at pH 8.0 under anaerobic
2
conditions. UV-vis absorption spectra collected after incubating
these samples for 5 min exhibited a decrease in intensity of the
catalase 403 nm Soret band and appearance of absorption fea-
tures consistent with NO binding to catalase (SI Appendix, Fig.
S2A). These absorption features are absent when HAO is ex-
cluded from the reaction mixture (SI Appendix, Fig. S2B), in-
dicating that NO production is HAO dependent. Together, these
results indicate that NO is at least one product of NH
dation by HAO under anaerobic conditions.
2
OH oxi-
To quantify the amount of NO released under anaerobic
conditions, similar sample conditions were used and the NH OH
concentration was varied with excess catalase present. The re-
actions were monitored at 433 nm. Addition of NH OH initiated
2
–
–
Fig. 2. Percent NH
HAO under anaerobic (−O
Conditions after mixing for aerobic experiments were 150 nM HAO, 5 μM
PMS, and 100 μM NH OH at pH 8.0 at room temperature for 30 min. Controls
–PMS and –HAO) were under similar conditions except either PMS or HAO
2
OH converted to NO
2
(blue bar) or NO
3
(red bar) by
2
) or aerobic (100 μM NH
2
OH) turnover conditions.
2
an increase in the 433 nm absorbance (Fig. 3A). This absorbance
increase was not observed in the absence of HAO (SI Appendix,
Fig. S3A). Using ΔA to quantify the amount of NO released,
2
(
4
33
were omitted from the reaction mixture and the incubation time was 5 h.
Conditions after mixing for anaerobic experiments were 150 nM HAO,
00 μM PMS, and 100 μM NH OH at pH 8.0 at room temperature for 30 min.
2
we found that 88–100% of the NH OH was converted to NO
2
(
Table 1).
2
Parallel reactions were performed in the absence of catalase to
Error bars represent one SD of at least three replicates. For absolute values,
determine the oxidant stoichiometry. UV-vis absorption spec-
troscopy was used to monitor and quantify the reduction of DCPIP
see SI Appendix, Table S1.
–1
–1
or horse heart cyt c at 605 (Δe605 = 20.3 mM ·cm ) or 550 nm
–
1
–1
HAO turnover using NO sequestering or scavenging methods
that kinetically outcompete these NO reactions.
(Δe550 = 19.6 mM ·cm ), respectively (Fig. 3A). At several
2
NH OH concentrations, 1.5–1.7 mol DCPIP or 2.8–3.0 mol cyt
c were reduced per 1 mol NH OH (Table 1). Together, these
2
HAO Stoichiometrically Oxidizes NH
2
OH to NO Under Anaerobic
data are consistent with the quantitative three-electron oxida-
tion of NH OH to NO by HAO under anaerobic conditions.
Conditions. To test for NO as an HAO product, an NO seques-
2
tration system developed by Pacheco and coworkers was used
7
–1 –1
HAO Oxidizes NH
2
OH to NO Under Aerobic Conditions. Under the
(
32). In this assay, catalase rapidly binds NO (kon = 1.3 × 10 M ·s )
canonical model of HAO activity (Fig. 1, NH OH obligate in-
2
(33) with concomitant appearance of defined UV-visible (UV-vis)
termediate model), HAO requires O
2
to complete the oxidation
. Therefore, the observed quantitative oxidation
OH to NO could result from the lack of O . To demonstrate
absorption features (SI Appendix, Fig. S2A). Catalase exhibits a Soret
band at 403 nm and poorly resolved absorbance features between
–
of NH
2 2
OH to NO
of NH
2
2
5
00 and 700 nm (SI Appendix, Fig. S2A, solid black trace). Upon
that NO remains the product of HAO catalysis in the presence of
O , we used ferrous-O hemoglobin (oxyHb) to simultaneously
treatment of catalase with NO, the Soret band shifts to 433 nm and
two well-resolved features at 542 and 579 nm appear (SI Appendix,
Fig. S2A, black dashed trace). NO can be quantified from the change
2
2
scavenge and quantify NO under aerobic conditions. In this assay,
NO reacts with oxyHb to form ferric-hemoglobin (metHb) and
–
Fig. 3. Evidence that N. europaea HAO oxidizes NH
2
OH to NO instead of NO
2
. (A) Representative 433- and 605-nm absorption traces used to monitor NO
binding to catalase (Cat-NO) or reduction of DCPIP resulting from the addition of NH
2
OH to HAO under anaerobic turnover conditions. (B) Representative
01-, 421-, and 550-nm absorption traces used to monitor the conversion of oxyHb to metHb or cyt c reduction resulting from the addition of NH OH to HAO
production under aerobic HAO turnover conditions in the absence or presence of oxyHb. Samples in A and B
OH, and either (A) 8 μM catalase or (B) 20 μM oxyHb at pH 8.0. Catalase and oxyHb were excluded from reaction
mixtures to obtain DCPIP and cyt c traces in A and B, respectively. Asterisks in A and B indicate time point of NH OH addition. Samples in C contained 150 nM
HAO and 5 μM phenazine methosulfate in the absence (−oxyHb) or presence (+oxyHb) of 30 μM oxyHb at pH 8.0. NH OH concentrations are indicated on the
abscissa. Error bars in C represent one SD of the average of at least three trials. Numerical data are reported in Table 1, and see SI Appendix, Tables S2–S4.
4
2
–
2
under aerobic turnover conditions. (C) NO
contained 150 nM HAO, 25 μM DCPIP, 4 μM NH
2
2
2
Caranto and Lancaster
PNAS Early Edition
|
3 of 6

2
Table 1. Stoichiometries of NO production and oxidant consumption vs. NH OH concentration under anaerobic and aerobic HAO
turnover conditions
[
NH
2
OH], μM
[NO]final, μM % Conversion to NO [cyt c]reduced/[NH
2 2
OH] [DCPIP]reduced/[NH OH]
Anaerobic conditions: Catalase NO-sequestering method*
2
4
6
1.95 (0.04)
4.1 (0.1)
5.3 (0.2)
98 (2)
103 (3)
88 (3)
3.0 (0.3)
2.8 (0.1)
1.5 (0.1)
1.5 (0.1)
1.7 (0.1)
†
Aerobic conditions: OxyHb NO-scavenging method
3
4
2.2 (0.2)
2.9 (0.1)
75 (5)
74 (4)
2.91 (0.05)
Values are reported as averages of at least three replicates with SDs shown in parentheses. Expanded oxidant consumption data are presented in SI
Appendix, Tables S2 and S3.
*
Reactions contained 150 nM HAO, 8 μM catalase, and 25 μM DCPIP or 20 μM cyt c in degassed 250 mM Na
2
HPO
4
, pH 8.0.
, pH 8.0.
†
Reactions contained 150 nM HAO, 20 μM OxyHb, and 25 μM DCPIP or 20 μM cyt c in air-saturated 50 mM Na
2
HPO
4
–
3
NO . The oxyHb to metHb conversion can be monitored from the
NO Is an Obligate Ammonia Oxidation Intermediate. NO production
but
OH oxidation when oxidant
concentrations were insufficient (17, 18, 31, 35). As dis-
resulting UV-vis absorption increase at 421 nm and concomitant by HAO has been previously observed in the absence of O
decrease at 401 nm (ΔA421 and ΔA401, respectively). The differ- was attributed to incomplete NH
ence in ΔA421 and ΔA401 provides the concentration of oxyHb or O
2
2
2
–
1
–1
converted to metHb (Δe421–401 = 77.3 mM ·cm ) (34) and cussed above, the detection of NO as the HAO product was
indirectly, the concentration of NO released during HAO turnover.
The results of this assay showed that HAO produced NO under
aerobic turnover conditions. Samples were prepared containing
likely obscured by several rapid, nonenzymatic NO reaction
pathways, especially when O was present. The present work
quantified the NO produced under both anaerobic and aerobic
2
–
1
50 nM HAO, 20 μM oxyHb, 25 μM DCPIP, and varying NH
2
OH
conditions and demonstrates that NO
product of HAO.
2
is not an enzymatic
concentrations at pH 8.0. In the presence of HAO, the absorption
at 401 nm decreased over 30 s with a concomitant increase in ab-
sorption at 421 nm (Fig. 3B). These absorption changes were not
observed in the absence of HAO, once again indicating that NO
production is HAO dependent (SI Appendix, Fig. S3B). The
ΔA421–401 for samples containing HAO was consistent with the
The revised N. europaea HAO product is consistent with
that of the K. stuttgartiensis kustc1061 protein that Kartal and
coworkers (20) purified and characterized; in their paper, the
authors proposed that a tyrosine above the active site pocket
(Y358) present in N. europaea HAO but absent in K. stuttgartiensis
6
production of 2.4 and 2.9 μM NO when the reaction was initiated kustc1061 facilitates the {FeNO} reaction with H
O to form
2
–
with either 3 or 4 μM NH OH, respectively (Table 1). At each of
NO
2
. However, our results indicate that the N. europaea HAO
OH oxidation products are identical and,
2
these concentrations, NO accounted for ∼75% of the NH OH and kustc1061 NH
2
2
–
added to the reaction mixture.
therefore, Y358 does not promote NO
note that in this same paper, the authors reported that N. europaea
HAO oxidized NH OH by four electrons. However, no evidence
was presented demonstrating NO
2
formation. We briefly
Although the full complement of NO was not detected, the ox-
idant stoichiometry with respect to NH OH was consistent with its
2
2
–
full conversion to NO. Samples without oxyHb were prepared
production or the absence
2
containing 150 nM HAO, 20 μM cyt c, and 3.4 μM NH
2
OH at of NO formation by N. europaea HAO. We conclude based
OH oxidation products of N. europaea
HAO and K. stuttgartiensis kustc1061 enzymes are both NO. The
pH 8.0. By monitoring cyt c as described above, we found that 2.91 on our results that the NH
0.05) mol cyt c were reduced per 1 mol NH OH (Table 1); this is
2
(
2
consistent with the three-electron oxidation of NH
2
OH to NO and consolidation of the activities of these proteins suggests a similar
suggests that conversion to NO is quantitative. This result in turn
physiological function of heme P460-containing octaheme pro-
teins in aerobic and anaerobic ammonia-oxidizing bacteria.
suggests that the NH OH is completely converted to NO but not all
2
–
of it reacts with oxyHb. A fraction of the NO could react with O to
The oxidation of NH
2
OH to NO instead of NO2 –implies that
2
–
2
–
2
form NO
. In this scenario, the expected NO
concentrations for
AOB require three steps to oxidize NH
NH OH and NO acting as obligate intermediates (Fig. 1, NH
obligate intermediate model). One possibility accounting for the
3
to NO
2
, with both
these experiments were beneath the limit of detection of the Griess
2
OH/NO
2
–
2
assay (<1 μM). However, NO
similar conditions at higher NH
formation was observed under
OH concentrations (see below).
–
total conversion of NH
NO is the terminal product of the enzymatic NH
to NO
3 2
in aerobic cell culture is that
oxidation path-
2
The combined data provide evidence that HAO-driven NH OH
oxidation terminates at NO even under aerobic conditions.
3
2
–
way, and free NO is nonenzymatically oxidized to NO by O .
2 2
However, our results confirm those of previous studies showing
–
–
NO
2
Is a Nonenzymatic Product of NO Oxidation by O
thus far suggest that NO, not NO , is the product of enzymatic
2
. Our results
that NO3 is a significant by-product of in vitro NH OH oxidation
2
–
2
by HAO and appears only under aerobic conditions (Fig. 2).
–
–
–
NH
2
OH oxidation by HAO, implying that NO
. By this model, the
concentration should decrease in the presence of an NO
scavenger. To test this hypothesis, we reacted 150 nM HAO with
2
results from the
This outcome suggests that, much like NO , NO results from
2
3
nonenzymatic reaction of the NO with O
2
a side reaction of NO under aerobic conditions.
–
2
–
NO
There is no evidence for a NO3 reductase in the genome of
–
N. europaea (36). Therefore, the lack of observed NO3 forma-
5
μM PMS at various NH
2
OH concentrations under aerobic
tion by intact AOB (18) suggests that the NH oxidation pathway
3
conditions in the absence or presence of the NO scavenger
requires an unidentified component that rapidly transforms NO
–
–
oxyHb. Samples containing oxyHb showed a 66–75% decrease in
to NO2 to outcompete side reactions that produce NO3 or
–
NO2– formation (Fig. 3C and SI Appendix, Table S4). This loss of
N O. Such an agent would also avoid nonenzymatic oxidation of
2
–
–
NO
2
is consistent with our hypothesis. The remaining NO
2
in
2
NO to NO
2
, thereby permitting the full complement of four
the presence of oxyHb suggests that some NO reacts with O
.
electrons to be removed from NH OH for entry into cellular
2
This competition between the NO reactions with oxyHb and O
2
electron transport chains. One possibility for this third step is the
–
also likely accounts for the lower-than-expected conversion of
Cu NO reductase NirK. NirK generally functions to reduce
2
–
oxyHb to metHb discussed above (Fig. 3B and Table 1).
NO2 to NO in denitrification or nitrifier denitrification pathways.
4
of 6
|
www.pnas.org/cgi/doi/10.1073/pnas.1704504114
Caranto and Lancaster

–
–
NirK deletion variants of N. europaea show no disruption in N
production under anaerobic conditions, which suggests that NirK
is unnecessary for nitrifier denitrification in these organisms (37,
2
O
combined NO
2 3
and NO concentrations were determined by treating samples
with nitrate reductase and cofactor solution (Cayman Chemical) at room tem-
–
–1
–1
perature for 1 h followed by NO
2
quantitation. DCPIP (e605 = 20.3 mM ·cm ),
–
1
–1
–1
–1
oxyHb (e415 = 131 mM ·cm ) (34), catalase (e403 = 120 mM ·cm
)
3
8). However, these variant strains exhibit considerably decreased
–
1
–1
–
(32, 34), cyt c (e550 = 28.0 mM ·cm ), and phenazine methosulfate (PMS;
2
NO production. Given these observations, we hypothesize that
–1
–1
e
388 = 26.3 mM ·cm ) (43) were quantified by UV-vis absorption spectroscopy.
NirK catalyzes the oxidation of NO produced by HAO to form
NH OH stocks were quantified by the method of Frear and Burrell (44). NO
2
–
NO2 , for which there is precedent in the literature (39). The
was generated from disodium 1-(hydroxyl-NNO-azoxy)-L-proline (PROLI-
combined, emergent picture strongly implicates a three-step
II
NONOate; Cayman Chemical) where the NO was quantified by Fe EDTA
–
oxidation of NH
3
oxidation to NO
2
by AOB.
as previously reported (21).
Whether enzymatic NH OH oxidation terminates at NO or
2
–
–
–
proceeds via a third enzymatic step to NO , energy transduction
Stoichiometry of NO2 and NO3 by HAO Under Various Conditions. Solutions
2
– –
remains productive. In the currently accepted model, hydroxyl-
used for NO
tained 150 nM HAO and 5 μM PMS in 1 mL of air-saturated 50 mM Na
pH 8.0. The reaction was initiated by addition of an appropriate volume of
an NH OH stock solution in water. For samples treated with oxyHb, the
2
and NO
3
determination under aerobic HAO conditions con-
ation of NH
3
to NH
2
OH requires two electrons for turnover of
2
HPO
4
,
AMO (6). These reducing equivalents are believed to be fur-
nished from the pool of electrons liberated from NH OH. Ter-
mination of enzymatic NH OH oxidation at NO liberates three
2
2
conditions were identical except that the samples were incubated in the
presence of 30 μM oxyHb.
Anaerobic samples were prepared in an MBraun glovebox maintained
2
electrons; after cycling AMO, one net electron remains for cel-
lular respiration. Furthermore, we note termination at NO yields
more free energy per mol N, assuming cellular respiration in-
under an N
00 μM NH
pH 8.0. The reaction was initiated by addition of an appropriate volume of
an NH OH stock solution in water. All samples were incubated at room
2
atmosphere. The conditions after mixing were 150 nM HAO,
1
2
OH, and 200 μM PMS in 1 mL of degassed 250 mM Na HPO
2
4
,
volving O
2
as a terminal electron acceptor (40):
2
3
2
kcal
molꢀ N
–
–
°
′
temperature for 30 min and then analyzed for NO2 and NO3 as described
2
ꢀ NH2OH + ꢀ ꢀ O2ꢀ → 2ꢀ NO + 3ꢀ H2O; ΔG = ꢀ –117.4ꢀ
[3]
[4]
in Materials and Methods.
kcal
molꢀ N
2
Stoichiometry of NH OH Oxidation to NO by HAO Under Anaerobic Conditions.
NO production under anaerobic conditions was monitored by using a catalase
NO-binding assay previously described by Pacheco and coworkers (32). To
–
+
°′
NH2OH + O2ꢀ → NO + H2O + H ; ΔG = –70ꢀ
2
determine the stoichiometry of NH
containing 150 nM HAO, 8 μM catalase, and 25 μM DCPIP in 1 mL of degassed
50 mM Na HPO , pH 8.0, were prepared in an anaerobic glovebox. All re-
actions were performed at room temperature and initiated by addition of an
appropriate volume of an anaerobic stock of NH OH dissolved in water. The
2
OH oxidation to NO, anaerobic solutions
The identification of HAO as an NO source may provide in-
sights into the environmental impacts of AOB. N O emissions
2
2
2
4
from AOB are largely attributed to nitrifier denitrification (8).
However, this pathway activates under anaerobic conditions, and
2
AOB have been shown to emit N O under aerobic conditions.
2
reactions were monitored at 433 nm by a UV-vis absorption spectropho-
tometer. The amount of NO generated was determined from the change
Although we have recently identified a source of N O originating
2
2 2
from NH OH (21), N O is usually the result of NO reduction.
in absorption at 433 nm and using the extinction coefficient of Δe433
=
–
1
–1
Our results identify HAO as a major source of NO under aerobic
47.9 mM ·cm . The stoichiometry of NH OH to reduced DCPIP was
2
conditions. We propose that N O production under aerobic
monitored in parallel reactions under identical reaction conditions except
2
conditions arises when the rate of NO production outcompetes
that catalase was omitted. The concentration of DCPIP reduced was de-
termined from the change in absorption at 605 nm and using the extinction
–
the rate of its oxidation to NO
2
. This would result in higher
–1
–1
2
coefficient Δe605 = 20.3 mM ·cm . To determine the stoichiometry of NH OH
intracellular NO concentrations that lead to NO emission or its
to cyt c reduction, the reactions conditions were similar to the DCPIP re-
action, except the DCPIP was replaced with 20 μM cyt c. The reaction was
monitored at 550 nm and the concentration of cyt c reduced determined using
conversion to N O by the respiratory nitric oxide reductase,
2
NorBC. Alternatively, NO could be scavenged by cyt P460 or cyt
c
554 t–o form N2
O (35). A decrease in the rate of NO oxidation to
–1
–1
the extinction coefficient, Δe550 = 19.6 mM ·cm
.
NO
2 2
might be expected at low O concentrations, and this may
help explain the observed increase in N O emissions at low O or
2
2
Stoichiometry of NH
determine the stoichiometry of NH
2
OH Oxidation to NO by HAO Under Aerobic Conditions. To
OH oxidation to NO, solutions were
during oxic to anoxic transitions (41, 42). The presently afforded
2
revision of HAO’s activity should facilitate understanding of the
prepared containing 150 nM HAO, 20 μM oxyHb and 25 μM DCPIP in 1 mL of
sources and rates of NO and N O emissions from ammonia
2
air-saturated 50 mM Na HPO , pH 8.0. Reactions were initiated by addition
2
4
oxidizing bacteria.
of an appropriate volume of a NH OH stock solution. The reactions were
2
simultaneously monitored at 401 and 421 nm by a UV-vis absorption spec-
Materials and Methods
General Considerations. Purified N. europaea HAO was a generous gift from
Prof. Sean Elliott, Boston University. Milli-Q water (18.2 MΩ; Millipore) was
used in the preparation of all buffers and solutions. Hydroxylamine hy-
trometer. The amount of NO generated was determined via the change in
absorption at 421 nm minus the change in absorbance at 401 nm and using
–1
–1
the extinction coefficient of Δe421–401 = 77.3 mM ·cm . The stoichiometry
of NH OH to cyt c reduction was performed under identical reaction
2
drochloride (NH
2
OH•HCl) was purchased from Sigma Aldrich. Catalase
conditions except the oxyHb was omitted. Cyt c reduction was quan-
and ferrous hemoglobin were purchased from Sigma-Aldrich. All other chem-
icals were purchased from VWR International and used as obtained. UV-vis
tified as described above for the anaerobic experiments.
absorption measurements were performed using a Flame spectrophotometer
ACKNOWLEDGMENTS. We thank Sean Elliot for graciously providing us with
a sample of hydroxylamine oxidoreductase with which to carry out this
work. This work was funded by Department of Energy Office of Science Early
Career Award DE-SC0013997 (to K.M.L.).
–
(Ocean Optics) or a Cary 60 spectrophotometer (Agilent Technologies). NO
2
was quantified using the Griess diazotization assay (Cayman Chemical) and
–1
–1
measuring the resulting absorbance at 542 nm (e542 = 50 mM ·cm ). The
1
2
3
. Galloway JN, et al. (2004) Nitrogen cycles: Past, present, and future. Biogeochemistry
0:153–226.
. Ravishankara AR, Daniel JS, Portmann RW (2009) Nitrous oxide (N
ozone-depleting substance emitted in the 21st century. Science 326:123–125.
. Heil J, Vereecken H, Brüggemann N (2016) A review of chemical reactions of nitrifi-
cation intermediates and their role in nitrogen cycling and nitrogen trace gas for-
mation in soil. Eur J Soil Sci 67:23–39.
5. Lehnert N, Coruzzi G, Hegg E, Seefeldt L, Stein L (2016) NSF workshop report: Feeding
the world in the 21st century: Grand challenges in the nitrogen cycle. Available at https://
www.nsf.gov/mps/che/workshops/nsf_nitrogen_report_int.pdf. Accessed March 14, 2017.
6. Hooper A, Arciero D, Bergmann D, Hendrich M (2005) The oxidation of ammonia as
an energy source in bacteria in respiration. Respiration in Archaea and Bacteria,
Advances in Photosynthesis and Respiration, ed Zannoni D (Springer, Dordrecht,
The Netherlands), Vol 16, pp 121–147.
7
2
O): The dominant
4
. Pilegaard K (2013) Processes regulating nitric oxide emissions from soils. Philos Trans R
Soc Lond B Biol Sci 368:20130126.
7. Stein L (2011) Heterotrophic nitrification and nitrifier denitrification. Nitrification, eds
Ward BB, Arp DJ, Klotz MG (ASM, Washington, DC), pp 95–116.
Caranto and Lancaster
PNAS Early Edition
|
5 of 6

8
9
. Stein LY (2011) Surveying N
86:131–152.
. Anderson JH (1964) The metabolism of hydroxylamine to nitrite by Nitrosomonas.
Biochem J 91:8–17.
2
O-producing pathways in bacteria. Methods Enzymol
27. Halaka FG, Babcock GT, Dye JL (1982) Properties of 5-methylphenazinium methyl
sulfate. Reaction of the oxidized form with NADH and of the reduced form with
oxygen. J Biol Chem 257:1458–1461.
28. Hughes MN (2008) Chemistry of nitric oxide and related species. Methods Enzymol
436:3–19.
29. Ford PC, Wink DA, Stanbury DM (1993) Autoxidation kinetics of aqueous nitric oxide.
FEBS Lett 326:1–3.
30. Goldstein S, Czapski G (1995) The reaction of NO. with O2.- and HO2.: A pulse radi-
olysis study. Free Radic Biol Med 19:505–510.
4
1
1
0. Lees H (1960) Energy metabolism in chemolithotropic bacteria. Annu Rev Microbiol
4:83–98.
1
1. Hooper A (1978) Nitrogen oxidation and electron transport in ammonia-oxidizing
bacteria. Microbiology, ed Schlessinger D (ASM, Washington, DC), pp 299–304.
2. Lees H (1952) Hydroxylamine as an intermediate in nitrification. Nature 169:156–157.
3. Ritchie GA, Nicholas DJ (1972) Identification of the sources of nitrous oxide produced
by oxidative and reductive processes in Nitrosomonas europaea. Biochem J 126:
1
1
31. Kostera J, Youngblut MD, Slosarczyk JM, Pacheco AA (2008) Kinetic and product
distribution analysis of NO* reductase activity in Nitrosomonas europaea hydroxyl-
amine oxidoreductase. J Biol Inorg Chem 13:1073–1083.
1
181–1191.
1
4. Shafirovich V, Lymar SV (2002) Nitroxyl and its anion in aqueous solutions: Spin states,
protic equilibria, and reactivities toward oxygen and nitric oxide. Proc Natl Acad Sci
USA 99:7340–7345.
3
2. Pacheco AA, McGarry J, Kostera J, Corona A (2011) Techniques for investigating hy-
droxylamine disproportionation by hydroxylamine oxidoreductases. Methods
Enzymol 486:447–463.
1
5. Hooper AB, Nason A (1965) Characterization of hydroxylamine-cytochrome c re-
ductase from the chemoautotrophs Nitrosomonas europaea and Nitrosocystis oce-
anus. J Biol Chem 240:4044–4057.
33. Purwar N, McGarry JM, Kostera J, Pacheco AA, Schmidt M (2011) Interaction of nitric
oxide with catalase: Structural and kinetic analysis. Biochemistry 50:4491–4503.
34. Salter M, Knowles RG (1998) Assay of NOS activity by the measurement of conversion
of oxyhemoglobin to methemoglobin by NO. Methods Mol Biol 100:61–65.
35. Upadhyay AK, Hooper AB, Hendrich MP (2006) NO reductase activity of the tetra-
heme cytochrome C554 of Nitrosomonas europaea. J Am Chem Soc 128:4330–4337.
36. Chain P, et al. (2003) Complete genome sequence of the ammonia-oxidizing bacte-
rium and obligate chemolithoautotroph Nitrosomonas europaea. J Bacteriol 185:
1
1
1
6. Rees MK (1968) Studies of the hydroxylamine metabolism of Nitrosomonas europaea.
I. Purification of hydroxylamine oxidase. Biochemistry 7:353–366.
7. Hooper AB, Terry KR (1979) Hydroxylamine oxidoreductase of Nitrosomonas. Pro-
duction of nitric oxide from hydroxylamine. Biochim Biophys Acta 571:12–20.
8. Hooper AB, Terry KR, Maxwell PC (1977) Hydroxylamine oxidoreductase of Nitro-
somonas. Oxidation of diethyldithiocarbamate concomitant with stimulation of ni-
trite synthesis. Biochim Biophys Acta 462:141–152.
2
759–2773.
3
7. Kozlowski JA, Price J, Stein LY (2014) Revision of N O-producing pathways in the
2
19. Yamanaka T, Sakano Y (1980) Oxidation of hydroxylamine to nitrite catalyzed by
ammonia-oxidizing bacterium Nitrosomonas europaea ATCC 19718. Appl Environ
Microbiol 80:4930–4935.
hydroxylamine oxidoreductase purified from Nitrosomonas europaea. Curr Microbiol
4
:239–244.
3
8. Beaumont HJ, et al. (2002) Nitrite reductase of Nitrosomonas europaea is not es-
sential for production of gaseous nitrogen oxides and confers tolerance to nitrite.
J Bacteriol 184:2557–2560.
2
0. Maalcke WJ, et al. (2014) Structural basis of biological NO generation by octaheme
oxidoreductases. J Biol Chem 289:1228–1242.
2
1. Caranto JD, Vilbert AC, Lancaster KM (2016) Nitrosomonas europaea cytochrome
P460 is a direct link between nitrification and nitrous oxide emission. Proc Natl Acad
Sci USA 113:14704–14709.
3
4
4
9. Wijma HJ, Canters GW, de Vries S, Verbeet MP (2004) Bidirectional catalysis by
copper-containing nitrite reductase. Biochemistry 43:10467–10474.
0. Aleem M (1970) Oxidation of inorganic nitrogen compounds. Annu Rev Plant Physiol
2
2. Andersson KK, Kent TA, Lipscomb JD, Hooper AB, Münck E (1984) Mössbauer, EPR,
and optical studies of the P-460 center of hydroxylamine oxidoreductase from Ni-
trosomonas. A ferrous heme with an unusually large quadrupole splitting. J Biol
Chem 259:6833–6840.
2
1:67–90.
1. Khalil K, Mary B, Renault P (2004) Nitrous oxide production by nitrification and de-
nitrification in soil aggregates as affected by O concentration. Soil Biol Biochem 36:
87–699.
2. Perez-Garcia O, Villas-Boas SG, Swift S, Chandran K, Singhal N (2014) Clarifying the
regulation of NO/N production in Nitrosomonas europaea during anoxic-oxic
2
6
2
3. Erickson RH, Hooper AB (1972) Preliminary characterization of a variant co-binding
heme protein from Nitrosomonas. Biochim Biophys Acta 275:231–244.
4. Zahn JA, Duncan C, DiSpirito AA (1994) Oxidation of hydroxylamine by cytochrome
P-460 of the obligate methylotroph Methylococcus capsulatus Bath. J Bacteriol 176:
4
2
2
O
transition via flux balance analysis of a metabolic network model. Water Res 60:
267–277.
5
879–5887.
2
5. Pearson AR, et al. (2007) The crystal structure of cytochrome P460 of Nitrosomonas
europaea reveals a novel cytochrome fold and heme-protein cross-link. Biochemistry
43. Zaugg WS (1964) Spectroscopic characteristics and some chemical properties of
N-methylphenazinium methyl sulfate (phenazine methosulfate) and pyocyanine at
the semiquinoid oxidation level. J Biol Chem 239:3964–3970.
4
6:8340–8349.
2
6. Enemark J, Feltham R (1974) Principles of structure, bonding, and reactivity for metal
nitrosyl complexes. Coord Chem Rev 13:339–406.
44. Frear D, Burrell R (1955) Spectrophotometric method for determining hydroxylamine
reductase activity in higher plants. Anal Chem 27:1664–1665.
6
of 6
|
www.pnas.org/cgi/doi/10.1073/pnas.1704504114
Caranto and Lancaster