...

pdf

by user

on
Category: Documents
1

views

Report

Comments

Description

Transcript

pdf
Limnol. Oceanogr., 57(1), 2012, 272–280
2012, by the Association for the Sciences of Limnology and Oceanography, Inc.
doi:10.4319/lo.2012.57.1.0272
E
Chelator-induced inhibition of copper metalloenzymes in denitrifying bacteria
James W. Moffett,a,* Caroline B. Tuit,b,1 and B. B. Ward b
a Department
b Department
of Biological Sciences, University of Southern California, Los Angeles, California
of Geosciences, Princeton University, Princeton, New Jersey
Abstract
Copper (Cu) is required by the enzyme nitrous oxide reductase (N2OR), which catalyzes the last step of the
complete denitrification pathway in denitrifying bacteria. Some denitrifiers also require copper for nitrite
reductase (NiRK), whereas others use the iron nitrite reductase (NiRS). We report the inhibition of the activity of
these enzymes in three strains of denitrifiers (two containing NiRK, the other NiRS), by forming nonbioavailable
complexes with 1,4,8,11-tetraazacyclotetradecane1,4,8,11-tetraacetic acid hydrochoride hydrate (TETA), a strong
Cu(II) chelator, and tetrathiomolybdate (TTMo), a strong Cu(I) chelator. Both ligands complex Cu with stability
constants comparable to naturally occurring ligands and much more strongly than other widely used chelators,
such as ethylenediaminetetraacetic acid. Addition of TETA to growth media lowered free Cu2+ concentrations
below 10216 mol L21 and induced Cu limitation in all organisms. While Cu is strongly complexed in seawater,
10216 mol L21 free Cu2+ is lower than most reported values, suggesting that the organisms have evolved highaffinity Cu transport systems. TTMo had different effects, and inhibited NiRK more effectively than N2OR. It is
likely that TTMo inhibits NiRK through direct, noncompetitive inhibition, as reported for other reduced sulfur
compounds, rather than by inducing Cu limitation. The activity of NiRK may be sensitive to trace levels of
reduced sulfur, which could account for its scarcity in marine systems. While Cu limitation of denitrification is
probably uncommon in aquatic systems, the presence of reduced sulfur compounds may induce Cu limitation or
enzyme inhibition leading to the accumulation of nitrite and nitrous oxide.
Copper (Cu) is required by metalloenzymes that catalyze
two key steps in denitrification: nitrite reduction and
nitrous oxide reduction (Stiefel 2007). Some denitrifiers
utilize a copper nitrite reductase (NiRK), while others have
an iron nitrite reductase (NiRS). Copper is absolutely
required for the enzyme nitrous oxide reductase (N2OR),
which catalyzes the last step of the complete denitrification
pathway in denitrifying bacteria. Thus, N2O reduction has
an absolute Cu requirement, whereas nitrite reduction can
always be mediated by denitrifiers with NiRS, in oxygenminimum zones where soluble iron is relatively abundant
(Moffett et al. 2007). Granger and Ward (2003) showed
that nitrous oxide reductase activity was inhibited in
cultures of the NiRS-type denitrifier, Pseudomonas stutzeri
using high levels of the chelator ethylenediaminetetraacetic
acid (EDTA). They argued that Cu formed complexes with
EDTA that were not biologically available, leading to a
direct decline in N2OR activity and accumulation of excess
N2O during denitrification. They suggested that this could
be important in the ocean, where Cu is strongly complexed
by organic ligands (Moffett and Dupont 2007). This is a
compelling hypothesis with important implications since
N2O is a radiatively important trace gas, and Cu may well
be complexed by reduced sulfur in the reducing regimes
where denitrification occurs. Recently, Buick (2007) argued
that Cu limitation of denitrification in the sulfidic
Archaean ocean might have led to the accumulation of
N2O in the atmosphere and a long-term warming scenario.
The link between copper availability and denitrification
rates or end products has not been substantiated in the
field. Ward et al. (2008) evaluated the control of
denitrification in incubations of water from the three major
oxygen-minimum zones of the world ocean and concluded
that organic carbon availability, rather than copper, limited
denitrification at all sites. Twining et al. (2007) induced Cu
limitation of N2O reduction by a natural freshwater
denitrifying assemblage by adding a strong Cu chelator
but saw no evidence for Cu limitation in natural samples
without added chelators, even those containing natural
ligands. They also saw no inhibition of N2O reduction of
natural assemblages at the same EDTA additions that were
effective in marine cultures (Granger and Ward 2003).
EDTA is widely used to lower the bioavailability of metals
to marine phytoplankton in culture experiments (Brand et
al. 1983). Generally, results from cultures containing
EDTA are extrapolated to real systems using the free ion
model (Morel and Hering 1993) since free ion activities can
be readily calculated from the EDTA stability constant
data. However, the discrepancy between the results of
Granger and Ward (2003) and Twining et al. (2007)
suggests that different ligands may have specific effects
on different organisms.
EDTA is not always an appropriate ligand for metal
limitation studies. While a strong Cu chelator, its high
stability constants with Ca and Mg result in a much weaker
conditional stability constant in seawater (log K ,10),
several orders of magnitude weaker than naturally occurring Cu-organic complexes (Martell and Smith 2001). As a
result, the very high concentrations of EDTA that are
typically used to mimic Cu complexation in the environment (Mann et al. 2002) may have secondary negative
* Corresponding author: [email protected]
1 Present address: Gradient Corporation, Cambridge, Massachusetts
272
Copper limitation of denitrification
effects on the organisms (Muggli and Harrison 1996).
Furthermore, a significant fraction of Cu in seawater may
be present as Cu(I) (Moffett and Zika 1988), particularly in
reducing conditions, and EDTA interacts only weakly with
Cu(I). Calculations using EDTA stability constants and
assuming that Cu(II) predominates will not be accurate if a
significant fraction of the Cu is present as Cu(I).
In this study, two strong copper ligands, which might be
more representative of ligands found in seawater, including
oxygen-minimum zones, were investigated for their ability
to induce copper limitation with three denitrifying bacterial
strains containing NiRK and NiRS. The objective was to
probe the mechanism of Cu acquisition using ligands with
selective affinities for Cu(I) and Cu(II) comparable in
strength to natural ligands. Very strong ligands enable us to
study the relationship between Cu limitation and free Cu2+
concentrations at the very lowest ranges of free Cu2+
reported in the literature without resorting to millimolar
levels of EDTA, which would presumably interfere with
many other metals. The inhibition of enzyme activity was
investigated using two chelators with very different
chemistries: 1,4,8,11-tetraazacyclotetradecane1,4,8,11-tetraacetic acid hydrochloride hydrate (TETA) and tetrathiomolybdate (TTMo). TETA binds Cu2+ with a stability
constant of 1021.74 (Clarke and Martell 1991) but has a
conditional stability constant in seawater of about 1016
because of side reactions with Ca2+ and Mg2+. This is
comparable to the highest value reported for natural Cu
organic complexes in seawater of 1016.6, which Kogut and
Voelker (2003) determined using cathodic stripping voltammetry and argued was technically indistinguishable
from an inert complex.
TTMo is widely used as a copper chelator, including for
medicinal purposes, but its chemistry is surprisingly poorly
characterized. While TTMo binds Cu(I), it will also reduce
Cu(II) to Cu(I) (Clarke and Laurie 1982). The 1 : 1 reaction
of TTMo and Cu(I) produces Cu2MoS4, but other ratios
yield a wide variety of Cu : S clusters (Clarke et al. 1987).
This makes it difficult to determine speciation coefficients
for the reaction of TTMo and Cu and precludes the exact
calculation of the free metal concentrations in these
cultures. Given that Cu(I) and Cu(II) undergo dynamic
redox cycling in seawater even under oxic conditions
(Moffett and Zika 1988), it seems likely that Cu would
form Cu(I) TTMo complexes in cultures where it was
added, TTMo can also inhibit many Cu metalloenzymes by
formation of ternary complexes (Bissig et al. 2001). This
property makes TTMo a powerful tool to probe the
stepwise transformations of nitrogen in denitrification, but
it also complicates the interpretation of the data.
Methods
Organisms—Three strains of facultative aerobic heterotrophic denitrifying bacteria were examined: P. stutzeri
(ATCC 14405), generally associated with eutrophic marine
environments; Ochrobactrum sp., isolated from the Arthur
Kill (Song and Ward 2003); and WLB20, a psychrophilic
isolate from a hypersaline Antarctic lake (Ward and Priscu
1997). All three perform complete denitrification, in which
273
RNO {
R NO R N2O R N2) is
each step (NO {
3
2
catalyzed by a different reductase enzyme. WLB20 and
Ochrobactrum have two copper enzymes in their denitrification pathway: NiRK, the Cu form of nitrite reductase,
and N2OR, the multicopper nitrous oxide reductase. P.
stutzeri, which has the Fe form of nitrite reductase, NiRS,
utilizes copper only for nitrous oxide reductase.
Culture conditions—Cells were cultured as described in
Granger and Ward (2003). Briefly, cells were grown in
batch culture using trace metal clean–techniques in the
artificial seawater medium Aquil (Price et al. 1988–1989)
modified for heterotrophic cultures (Granger and Price
1999). Synthetic ocean water and phosphate and nutrient
stocks were stripped of trace metals using Chelex 100 ion
exchange resin (Bio-Rad Laboratories) following the
procedure of Price et al. (1988–1989). Phosphate and
nitrate were added to a final concentration of 10 mmol L21
and 250–300 mmol L21, respectively. The nitrogen and
carbon rich organic stock, a 1 : 1 mixture of bactopeptone
and casamino acids, was added to a final concentration
0.8 g L21 total. Chelexed nutrient stocks were sterilized by
heating to just below boiling in the microwave, several
times, before storage. On nutrient addition, the medium
was sterilized in the same way in acid-cleaned polycarbonate bottles and then enriched with filter-sterilized trace
metals and vitamins (B12, thiamin, and biotin). The trace
metal additions, Fe 1025.08, Mn 1026.92, Zn 1027.10, Co
1027.3, Mo 1027, and Se 1028 M, were buffered with filtersterilized 100 mmol L21 EDTA. Premixed Cu-EDTA (1 : 1)
was added to a final concentration of 11 nmol L21 in Cureplete cultures. Cu-limited cultures contained no added
Cu. The background Cu concentration was estimated to be
3 nmol L21 (Granger and Ward 2003).
Experimental design—Cultures were initiated from frozen stocks. They were initially grown aerobically on a rich
medium containing 5 g of peptone, 0.5 g of yeast extract
stock per liter of seawater, and the trace metal mixture
above. Cultures were then transferred to Cu-free Aquil and
allowed to acclimate to low Cu culture conditions. They
were transferred twice under trace metal–clean aerobic
conditions to limit background Cu contamination. Cultures were finally inoculated into trace metal–clean gastight
trilaminate bags (Pollution Management Corporation;
Granger and Ward 2003) without headspace. The medium
(, 400 mL) was not purged of oxygen before inoculations.
Initial culture growth was supported by oxygen, and
denitrification set in as the initial oxygen was consumed.
The bags were sampled aseptically without air contamination by expelling the medium out through narrow tubing.
Growth rate on each alternative electron acceptor was
determined from the log-linear plot of cell density over time
as the cultures sequentially respired oxygen, nitrate, nitrite,
and nitrous oxide. Cell density was determined by flow
cytometry counts of samples fixed with 2% glutaraldehyde
using Syto 13 dye (Molecular Probes) and a BectonDickinson FACSort Cell Sorter Flow Cytometer with a
488-nm laser. P. stutzeri and Ochrobactrum sp. were grown
at 20uC, while the psychrophilic WLB20 was grown at
274
Moffett et al.
Table 1.
Estimates of free Cu2+ and free Fe3+ in the growth media using MINEQL.
Cu added 3
TETA (mmol L21)
0
0
0
11 nmol L21
11 nmol L21
11 nmol L21
1.5 mmol L21 CuEDTA
1.5 mmol L21 CuEDTA
1.5 mmol L21 CuEDTA
0
1
4
0
1
4
0
1
4
12uC. Single cell counts from each of the duplicate bags
were averaged for each time point, and the growth rate was
estimated from a linear fit to the log-transformed data
during the interval corresponding to growth on the
different respiratory substrates.
Nitrite was measured colorimetrically (Parsons 1984).
Nitrous oxide was measured on a Shimadzu mini-2 gas
chromatograph with an electron capture detector and a 2-m
3 2.2-mm–inner-diameter Haysep D column (80–100
mesh). The column was set at ambient temperature, while
the injection port and detector were maintained at 50uC
and 300uC, respectively. The carrier gas was ultrahighpurity N2. Cultures were subsampled for N2O by collecting
25 mL of culture in a 60-mL syringe. N2O concentration in
the sample was calculated from three serial equilibration
samples using the multiphase equilibrium approach (Byrne
and Nicholas 1986). The method was standardized by using
both air and a nitrous oxide standard.
Speciation of Cu and other metals in the medium was
determined using MINEQL (Environmental Research
Software) with stability constants for TETA input from
Martell and Smith (2001). Estimates of free Cu2+ and free
Fe3+ are shown in Table 1. The carbon source used in these
cultures was undefined and not considered in these
calculations. The bactopeptone casamino acids mixture
likely has some weak metal binding capacity; therefore, all
estimates of free ion concentrations are likely to be upper
limits.
Copper chelators—A TETA stock was prepared in
aqueous solution and added to each bag for final
concentrations of 1 mmol L21 and 4 mmol L21 and allowed
to equilibrate with Cu in the medium for at least 12 h
before bacteria were added to ensure that all of the Cu was
in the Cu(II)TETA22 form. Aqueous TTMo stocks
(100 mmol L21) were prepared aerobically and buffered at
pH 8.2 with 1 mol L21 Na2CO3 and trace metal–grade
Sigma HCl. TTMo will degrade in the presence of oxygen
to tri-, di-, and monothiomolybdate and release free sulfide
(Erickson and Helz 2000). The abiotic oxidation is slow,
but since the effect of microbial catalysis is unknown,
TTMo was made fresh and added to trilaminate bags only
after nitrite was already detectable, indicating the onset of
denitrification and lowered oxygen concentration in the
culture. Final concentrations in each bag ranged from
40 nmol L21 to 1 mmol L21.
Free Fe3+ (mmol L21) Free Cu2+ (mmol L21)
1.43310218
1.28310218
8.29310219
1.43310218
1.28310218
8.29310219
1.46310218
1.43310218
1.05310218
2.5310215
3.81310217
6.28310218
9.15310215
1.41310216
2.31310217
1.27310212
4.45310213
6.29310215
Results
P. stutzeri (NiRS strain)
No added chelators: The Cu-replete (Cu 11) P. stutzeri
cultures evolved only small (, 10 nmol L21) and transient
amounts of N2O because N2O was rapidly reduced to N2
via N2OR (Fig. 1). In the treatments with no added Cu (Cu
0), N2O concentrations accumulated to 80–90 mmol L21
and remained stable for over 60 h (Fig. 1a), indicating
significant inhibition of N2OR activity. The conversion of
NO {
3 to N2O was not always stoichiometric, indicating
outgassing of N2O during sampling or residual microbial
reduction.
Added chelators: TETA clearly inhibited N2O reduction,
even in the Cu 11 treatment (Fig. 1a). At 4 mmol L21
TETA, inhibition was observed for the entire experiment.
In contrast, in Cu 11 with 1 mmol L21 TETA, N2O
increased to the same level as in the Cu 0 and 4 mmol L21
TETA Cu 11 treatment but then decreased, suggesting only
partial or transient Cu limitation. The error bars in this
treatment probably arise from slight differences in initial
growth rates that lead to larger differences in the timing of
exponential growth.
TETA did not affect growth rates when the organisms
{
were utilizing NO {
3 , but after NO 3 was depleted, growth
21) was approximately
on NO {
(m
5
0.0023
6
0.0004
h
2
half that in the TETA-free cultures (m 5 0.005 6 0.0002),
regardless of the Cu content (Table 2).
TTMo inhibited N2O reduction in all cases, but the
effect was temporary, and reduction of N2O eventually
occurred in all treatments (Fig. 1b). Larger TTMo additions resulted in larger N2O peaks, up to 117 nmol L21 in
the 1 mmol L21 TTMo treatment. But inevitably, N2O
concentration decreased again.
WLB20 (NiRK strain)—This Antarctic isolate was first
investigated by Granger and Ward (2003), who found that
it responded to Cu limitation by accumulating NO {
2 in the
medium and postulated that the organism contained the
copper form of nitrite reductase.
No added chelator: Under Cu-replete conditions,
{
WLB20 stoichiometrically reduced NO {
3 to NO 2 , and,
2
after a lag of 24 h or less, the NO2 was removed (Fig. 2a).
Under Cu-limited conditions, NO {
concentration de2
creased much more slowly than under replete conditions.
Large differences between replicates, indicated by the error
Copper limitation of denitrification
275
Fig. 1. Pseudomonas stutzeri experiment with (a) TETA treatments and (b) TTMo treatments. Nitrous oxide concentrations are
plotted on the right axis with dashed lines, and nitrite concentrations are plotted on the left axis have solid lines. Error bars indicate the
range of duplicate treatments.
bars, arise from slight differences in timing of growth in
replicate bags.
At 4 mmol L21 TETA, nitrite reduction was completely
inhibited for 250 h, even at high Cu (Cu 11), until extra
copper was added (see below). At 1 mmol L21 TETA,
nitrite decreased slowly, indicating partial inhibition, very
similar to the Cu 0 treatment (Fig. 2a).
concentrations persisted in all TTMo
High NO {
2
treatments with WLB20 (Fig. 2b) and were stable for over
200 h. This indicated complete inhibition of nitrite
reduction by TTMo.
Addition of extra copper: The high Cu incubations
containing 4 mmol L21 TETA and 1 mmol L21 TTMo were
treated with 1.55 mmol L21 CuEDTA in the middle of the
experiment (Fig. 2a,b). These additional inputs of Cu were
estimated to raise the free pCu2+ (2log[Cu2+]) to the level of
the Cu-replete cultures (Table 1). In these treatments,
NO {
concentrations decreased within 90 min of Cu
2
addition, and NO {
2 reduction rates equal to those observed
in the Cu-replete cultures were rapidly attained.
Growth rates on alternate electron acceptors—Growth
rates were measured in most incubations in WLB20 and are
shown along with denitrification rates in Table 2. Not
surprisingly, inhibition of nitrite reduction essentially
stopped growth when the cells were growing on nitrite
but did not affect growth rates when nitrate was present at
the beginning of each incubation. Addition of extra Cu
276
Moffett et al.
21 h21) under different
Table 2. WLB20 Growth (m h21) and denitrification rates (NO {
2 production or reduction; mmol L
concentrations of Cu (nmol L21) and TETA or TTMo.
Cu added
Chelator
m (NO {
3 )
m (NO {
2 )
NO {
2 production
NO {
2 reduction
0
0
0
11
11
1500
11
11
11
11
1500
4 mmol L21 TETA
1 mmol L21 TTMo
None
None
4 mmol L21 TETA
4 mmol L21 TETA
1 mmol L21 TETA
0.25 mmol L21 TTMo
0.5 mmol L21 TTMo
1 mmol L21 TTMo
1 mmol L21 TTMo
0.009
0.010
0.010
0.010
nm*
nm
0.013
nm
nm
nm
nm
0.0002
0.0000
0.0002
0.0011
nm
0.0005
0.0007
nm
nm
nm
0.0013
9.761.4
9.160.4
6.860.4
9.560.4
8.061.2
nm
8.761.7
8.560.8
6.860.5
7.160.9
nm
060.0
060.0
0.860.3
2.460.0
0.060.1
2.260.0
0.960.1
0.160.0
0.160.1
0.160.0
nm
* nm, parameter not measured in this treatment.
during the experiment restored the growth rate in the
TTMo-treated incubation but increased the growth rate to
only 50% of its initial value in the TETA treatment despite
recovery of the Cu-replete rate of NO {
2 reduction.
Ochrobactrum sp. (NiRK strain): Ochrobactrum exhibited different denitrification activity compared to P. stutzeri
and WLB20. Ochrobactrum was observed to support
simultaneous nitrate reductase and nitrite reductase activity
in this study since NO {
did not accumulate as an
2
intermediate. The onset of NO {
2 reduction was apparently
induced by ambient NO {
concentrations of between
2
10 mmol L21 and 50 mmol L21 rather than being delayed
until NO {
3 was depleted.
No added chelator: Under Cu-replete conditions, the
NO {
2 peak was a small and transient feature (Fig. 3a).
Under Cu-limiting conditions, NO {
concentrations in
2
Ochrobactrum approached the initial NO {
3 concentration
(250–300 mmol L21; Fig. 3a).
Results for Ochrobactrum using TETA are summarized
in Fig. 3a. Addition of 4 mmol L21 TETA to a Cu 0
treatment induced more severe Cu limitation than Cu 0
alone; this treatment was the only one that accumulated
{
NO {
2 equal to the initial NO 3 concentration. Presumably,
the TETA scavenged traces of Cu still remaining in the
control. In the presence of added Cu (Cu 11), TETA was
unable to completely eliminate Cu uptake. Nitrite was
removed in both the 1 mmol L21 and the 4 mmol L21 TETA
treatments; in the 1 mmol L21 treatment, the pattern of
N2O removal was indistinguishable from the Cu-replete
treatment. The 4 mmol L21 TETA treatment produced up
to 200 mmol L21 NO {
2 before nitrite reductase activity
began and removed all the NO {
2 by the fourth day.
Therefore, despite the addition of TETA, there was
sufficient Cu available to form active NiRK.
The addition of only 40 nmol L21 TTMo to Cu-replete
cultures of Ochrobactrum at the onset of NO {
3 reduction
resulted in total cessation of NO {
2 reduction (Fig. 3b).
Nitrite concentrations were more stable in both of the
TTMo additions (average concentration of 248 6
10 mmol L21 maintained for over 70 h) than in the Cufree culture (nitrite concentration decreased 75 mmol L21 in
100 h), suggesting that there was residual Cu in the Cu-free
culture that allowed the formation of some small amount of
active NiRK.
Effects of TETA and TTMo unrelated to copper—TETA
did not affect growth rates of P. stutzeri when the cultures
(m 5 0.010 6 0.002 for TETA
were reducing NO {
3
treatments vs. 0.009 6 0.0006 for controls), but growth on
NO {
2 (m 5 0.0023 6 0.0004) was approximately half that in
the TETA-free cultures (m 5 0.005 6 0.0002) regardless of
the Cu content. Nitrite reduction requires Fe in P. stutzeri
and TETA is an Fe chelator, but Fe limitation is unlikely,
as Fe concentration exceeded that of TETA in the medium.
By contrast, there was no difference in growth rates during
NO {
2 reduction among the TTMo treatments, suggesting
no effects of TTMo on P. stutzeri unrelated to N2O
reduction (data not shown). There was no evidence that
TETA had any side effects on Ochrobactum unrelated to
Cu. For instance, 0–2 mmol L21 TETA had no effect on
Ochrobactrum growing aerobically (m 5 0.078 6 0.003; n 5
6). For WLB20, the Cu addition experiments revealed a
possible effect of TETA unrelated to copper. In the Cu11
treatment with 4 mmol L21 TETA, growth rate remained
suppressed after Cu addition despite recovery of the Cureplete rate of NO {
2 reduction, whereas for TTMo, both
growth and NO {
remained
low.
2
Discussion
The experimental data show that both ligands inhibit
N2OR and NirK activities in bacterial denitrifiers. Both
chelators are highly specific for these processes. Effects not
related to copper are modest and associated only with
TETA. These effects probably arise from the ability of
TETA to bind other bioactive metals besides Fe (always in
excess of TETA) and Cu.
This study builds on previous work utilizing chelators to
inhibit Cu metalloenzymes, from which two mechanisms of
inhibition have emerged. The first mechanism is chelation
of the Cu in the medium, resulting in low Cu bioavailability and, ultimately, Cu limitation of the enzyme synthesis,
as outlined in our introduction. Chelation of Cu in solution
to form nonbioavailable complexes is the underlying
assumption behind most limitation studies utilizing metal
Copper limitation of denitrification
277
Fig. 2. WLB20 experiment with (a) TETA treatments and (b) TTMo treatments. Error bars indicate the range of duplicate
treatments. Arrow indicates point at which Cu was added during the experiment.
ion buffers such as EDTA. The second mechanism is the
direct coordination of the chelator to Cu within the active
site—a form of noncompetitive inhibition. This mechanism
is feasible because both N2OR and NiRK are located in the
periplasm (Zumft 1997) and are exposed to small molecules
in the medium. Distinguishing between these two mechanisms is important in order to develop a relationship
between inhibition and free Cu2+ in the medium that can be
applied to Cu speciation measurements from the water
column. An excellent example of noncompetitive inhibition
is the inhibition of ammonium monooxygenase by
allylthiourea (ATU), which is highly effective in natural
marine assemblages (Molina and Farias 2009). ATU is a
weak Cu chelator, and at the concentrations required for
inhibition, it does not lower free Cu significantly in
seawater media. In such a system, if we attributed
inhibition to Cu limitation in the media, we would
calculate an unreasonably high free Cu2+ threshold for
limitation and conclude the organisms were Cu limited
everywhere. TTMo is well known to act as a noncompetitive inhibitor of many Cu metalloenzymes, such as
mushroom tyrosinase (Park et al. 2005). In that system,
inhibition was observed in vitro, demonstrating noncompetitive inhibition unequivocally. While TTMo has never
been shown to inhibit NirK or N2OR, these enzymes are
inhibited by other monodentate sulfur ligands, such as
diethyldithiocarbamate (DDC) and inorganic sulfide
(Zumft 1997). In the marine environment, TTMo has been
shown to inhibit the activity of copper oxidases associated
with Fe uptake in diatoms (Maldonado et al. 2006).
278
Moffett et al.
Fig. 3. Ochrobactrum sp. experiment with (a) TETA treatments and (b) TTMo treatments. Error bars indicate the range of
duplicate treatments.
The two mechanisms can be distinguished as follows. Cu
limitation is a function of the concentration of bioavailable
Cu in the medium (which is typically proportional to the
free Cu in the medium), whereas noncompetitive inhibition
by the ligand is proportional to the free ligand concentration. In most experiments, these parameters covary,
making resolution difficult. An elegant solution to this
quandary was utilized by Bartacek et al. (2010). These
authors determined that sulfide inhibits N2OR in Pseudomonas aeruginosa through formation of a complex with the
enzyme (i.e., noncompetitive inhibition) rather than Cu
starvation because the activity was restored by adding high
levels of zinc to the system. Zinc addition does not increase
bioavailable Cu since its interaction with sulfide is weaker
than that of copper, but it does precipitate excess free
sulfide from solution. Noncompetitive inhibition by sulfide
was partially reversible, so depleting its concentration in
the medium resulted in partial or complete restoration of
activity. Reversibility was also observed in other systems
using TTMo and ATU, where dilution of the ligands in the
growth medium restored activity.
In our experiments, it was not possible to distinguish
unequivocally between these alternatives for TTMo. The
rapid recovery of NiRK activity seen when very high Cu
was added in the middle of the WLB20 experiments is not
conclusive because such effects have been observed for both
mechanisms.
TETA is unlikely to be a noncompetitive inhibitor for
NiRK or N2OR. There are no previous reports of TETA
acting as a noncompetitive ligand for any metalloenzymes.
Copper limitation of denitrification
This could reflect steric considerations; as a large, cyclic
chelator with four coplanar nitrogen atoms, it is hard to
envision how TETA could insert itself into the active site in
contrast to a small ligand like TTMo. Moreover, it is hard
to envision why noncompetitive inhibition would go away
with time, as observed for some TETA experiments with
Ochrobactrum and WLB20, since TETA, unlike TTMo, is
not subject to oxidation. Therefore, the relationship
between free Cu2+ and inhibition identified in the TETA
experiments can reasonably be extrapolated to other ligand
systems, including the natural environment. Our data
indicate that the ‘‘threshold’’ at which limitation occurs
must lie between 10216 mol L21 and 10217 mol L21 free
Cu2+, about 100–1000-fold lower than in the EDTA
experiments of Ward and Granger (2003). That suggests
that EDTA at high levels may be exerting an effect on Cu
uptake that cannot be simply modeled by its thermodynamic data, perhaps through a noncompetitive interaction.
The threshold in this study is in agreement with Twining
et al. (2007), who studied N2O production in a natural
assemblage of denitrifiers. They found that even a
temporary suppression of N2OR activity required addition
of 100 mmol L21 8-hydroxyquinoline (oxine), with a
corresponding free Cu2+ concentration of less than
10216.5 M. Estimates of free Cu2+ reported in the literature,
available only from oxygenated waters, are higher, ranging
from 10215 M in some organic-rich coastal waters (Shank
et al. 2004) to 10214–10212 M in open ocean waters
(Moffett and Dupont 2007). Evidently, a significant
alteration of Cu chemistry in a suboxic system, such as
the accumulation of reduced sulfur ligands or a shift in
valence state to Cu(I), must occur before Cu limitation will
become important.
TTMo may function as a noncompetitive inhibitor in
our experiments. Such a mechanism may explain the
difference in the sensitivity of NiRK and N2OR to TTMo.
Nitrite reductase activity in both NiRK strains (Ochrobactrum and WLB20) was readily inhibited by TTMo
compared to N2OR activity in P. stutzeri, the NiRS
organism. This may be a result of differences in the two
reductase enzymes, NiRK and N2OR, which have strikingly different metal centers. NiRK metal centers consist of
simple metal-histidine bonds (Zumft 1997), which may be
readily susceptible to insertion of TTMo. N2OR contains
both a simple protein-bound copper center and a more
complex CuS cluster (Honisch and Zumft 2003). Interestingly, the concentration of TTMo required to inhibit NiRK
in Ochrobactum (40 nmol L21) was comparable to the
concentration required to inhibit the purified Enterococcus
hirae CopB copper ATPase in vitro (34 nmol L21; Bissig et
al. 2001). This suggests that direct inhibition of NiRK by
TTMo for Ochrobactrum is at least plausible.
It is also possible, however, that TTMo interferes with
Cu uptake in these organisms. If uptake is occurring via a
Cu(I) transport system, then TTMo would also be more
effective than TETA. Clearly, more work is required,
including the determination of stability constants for Cu
MoS4 complexes and innovative experiments like those
carried out for sulfide by Bartacek et al. (2010). We need to
determine the oxidation rates of TTMo under the
279
experimental conditions and determine the inhibitory
effects of the oxidation products MoS3O22, MoS2O 2{
2 ,
.
Bissig
et
al.
(2001)
demonstrated
that
these
and MoSO 2{
3
molecules are also powerful inhibitors.
TTMo can be an extraordinarily useful inhibitor that
provides insight into how these systems behave in the
presence of traces of reduced sulfur compounds. Our
experiments reveal that NiRK is very sensitive to inhibition
by reduced S, consistent with studies using glutathione,
DDC, and sulfide (Zumft 1997). The presence of traces of
such compounds in suboxic marine environments where
denitrification occurs is likely since denitrification often
occurs in close proximity to regions where there is active
sulfate reduction and, in some cases, coincidentally
(Canfield et al. 2010). Dupont et al. (2006) reported
glutathione in oxygenated waters in the North Pacific.
Reduced sulfur has many abundant sources in seawater,
and this may account for the observation that NiRK is
relatively less abundant in marine systems than the Fe
form, NiRS (Ward et al. 2009; A. Jayakumar and B. B.
Ward, unpubl.). A similar case was made by Joye and
Hollibaugh (1995), who argued that reduced sulfur inhibits
nitrification in sedimentary regimes in close proximity to
sources of reduced sulfur because of the vulnerability of
ammonium monooxygenase to sulfide inhibition. TTMo
itself has been hypothesized to be produced from molybdate in seawater under sulfidic conditions (Erickson and
Helz 2000) with a half-life of TTMo in oxygenated seawater
greater than six months, so it may well diffuse from
reducing sulfidic waters into suboxic zones where denitrification is occurring.
In summary, the activity of key Cu-containing enzymes
in the marine nitrogen cycle can be inhibited by Cu
complexation and by direct inhibition of the enzymes
involved. Some compounds, particularly those containing
reduced sulfur, may function in both capacities. Accumulation of these compounds in anoxic regimes and their
subsequent diffusion into areas where active denitrification
is occurring may contribute significantly to nitrous oxide
production.
Acknowledgments
Supported by the National Science Foundation Chemical
Oceanography Program through grants to J.W.M. and B.B.W.
References
BARTACEK, J., I. MANCONI, G. SANSONE, R. MURGIA, AND P. N. L.
LENS. 2010. Divalent metal addition restores sulfide-inhibited
N2O reduction in Pseudomonas aeruginosa. Nitric Oxide—
Biol. Chem. 23: 101–105, doi:10.1016/j.niox.2010.04.005
BISSIG, K. D., T. C. VOEGELIN, AND M. SOLIOZ. 2001. Tetrathiomolybdate inhibition of the Enterococcus hirae CopB
copper ATPase. FEBS Lett. 507: 367–370, doi:10.1016/S00145793(01)03009-5
BRAND, L. E., W. G. SUNDA, AND R. R. L. GUILLARD. 1983.
Limitation of marine-phytoplankton reproductive rates by
zinc, manganese, iron. Limnol. Oceanogr. 28: 1182–1198,
doi:10.4319/lo.1983.28.6.1182
BUICK, R. 2007. Did the proterozoic ‘‘Canfield Ocean’’ cause a
laughing gas greenhouse? Geobiology 5: 97–100, doi:10.1111/
j.1472-4669.2007.00110.x
280
Moffett et al.
BYRNE, M. D., AND D. J. D. NICHOLAS. 1986. Multiple-phase
equilibration headspace analysis for the determination of
N2O and N-2 during bacterial denitrification. Anal. Biochem.
154: 470–475.
CANFIELD, D. E., AND OTHERS. 2010. A cryptic sulfur cycle in
oxygen-minimum–zone waters off the Chilean coast. Science
330: 1375–1378, doi:10.1126/science.1196889
CLARKE, E. T., AND A. E. MARTELL. 1991. Stabilities of trivalent
metal-ion complexes of the tertraacetate derivatives of 12membered, 13-membered and 14-membered tetraazamacrocycles. Inorg. Chim. Acta 190: 37–46, doi:10.1016/S0020-1693
(00)80229-7
CLARKE, N. J., AND S. H. LAURIE. 1982. The copper-molydenum
antagonism in ruminants 2. Interactions of thiomolybdates
with copper(II) in aqueous-media. Inorg. Chim. Acta
Bioinorg. Chem. 66: L35–L38, doi:10.1016/S0020-1693(00)
85767-9
———, ———, AND D. E. PRATT. 1987. The copper molybdenum
antagonism in ruminants 4. Reactions of thiomolybdate with
proteins and metalloenzymes. Inorg. Chim. Acta 108:
103–113, doi:10.1016/S0020-1693(00)81190-1
DUPONT, C. L., J. W. MOFFETT, R. R. BIDIGARE, AND B. A. AHNER.
2006. Distributions of dissolved and particulate biogenic
thiols in the subarctic Pacific Ocean. Deep Sea Res. I 53:
1961–1974, doi:10.1016/j.dsr.2006.09.003
ERICKSON, B. E., AND G. R. HELZ. 2000. Molybdenum(VI)
speciation in sulfidic waters: stability and lability of thiomolybdates. Geochim. Cosmochim. Acta 64: 1149–1158,
doi:10.1016/S0016-7037(99)00423-8
GRANGER, J., AND N. M. PRICE. 1999. The importance of
siderophores in iron nutrition of heterotrophic marine
bacteria. Limnol. Oceanogr. 44: 541–555, doi:10.4319/
lo.1999.44.3.0541
———, AND B. B. WARD. 2003. Accumulation of nitrogen oxides
in copper-limited cultures of denitrifying bacteria. Limnol.
Oceanogr. 48: 313–318, doi:10.4319/lo.2003.48.1.0313
HONISCH, U., AND W. G. ZUMFT. 2003. Operon structure and
regulation of the nos gene region of Pseudomonas stutzeri,
encoding an ABC-type ATPase for maturation of nitrous
oxide reductase. J. Bacteriol. 185: 1895–1902, doi:10.1128/
JB.185.6.1895-1902.2003
JOYE, S. B., AND J. T. HOLLIBAUGH. 1995. Influence of sulfide
inhibition of nitrification on nitrogen regeneration in sediments. Science 270: 623–625, doi:10.1126/science.270.
5236.623
KOGUT, M. B., AND B. M. VOELKER. 2003. Kinetically inert Cu in
coastal waters. Environ. Sci. Technol. 37: 509–518,
doi:10.1021/es020723d
MALDONADO, M. T., A. E. ALLEN, J. S. CHONG, K. LIN, D. LEUS,
N. KARPENKO, AND S. L. HARRIS. 2006. Copper-dependent
iron transport in coastal and oceanic diatoms. Limnol.
Oceanogr. 51: 1729–1743, doi:10.4319/lo.2006.51.4.1729
MANN, E. L., N. AHLGREN, J. W. MOFFETT, AND S. W. CHISHOLM.
2002. Copper toxicity and cyanobacteria ecology in the
Sargasso Sea. Limnol. Oceanogr. 47: 976–988, doi:10.4319/
lo.2002.47.4.0976
MARTELL, A. E., AND R. M. SMITH. 2001. NIST critically selected
stability constants of metal complexes. National Institute of
Science and Technology (NIST) Standard Reference Database 46 Version 6.0.
MOFFETT, J. W., AND C. L. DUPONT. 2007. Copper complexation
by organic ligands in the subarctic NW Pacific and Bering
Sea. Deep-Sea Res. I 54: 586–595, doi:10.1016/
j.dsr.2006.12.013
———, T. J. GOEPFERT, AND S. W. A. NAQVI. 2007. Reduced iron
associated with secondary nitrite maxima in the Arabian Sea.
Deep-Sea Res. I 54: 1341–1349, doi:10.1016/j.dsr.2007.04.004
———, AND R. G. ZIKA. 1988. Measurement of copper(II) in
surface waters of the sub-tropical Atlantic and Gulf of
Mexico. Geochim. Cosmochim. Acta 52: 1849–1857,
doi:10.1016/0016-7037(88)90008-7
MOLINA, V., AND L. FARIAS. 2009. Aerobic ammonium oxidation
in the oxycline and oxygen minimum zone of the eastern
tropical South Pacific off northern Chile (,20uS). Deep Sea
Res. I 56: 1032–1041.
MOREL, F. M. M., AND J. G. HERING. 1993. Principles and
applications of aquatic chemistry. Wiley Interscience.
MUGGLI, D. L., AND P. J. HARRISON. 1996. EDTA suppresses the
growth of oceanic phytoplankton from the northeast subarctic Pacific. J. Exp. Mar. Biol. Ecol. 205: 221–227, doi:10.1016/
S0022-0981(96)02611-1
PARK, K. H., Y. PARK, J. LEE, H. HAHN, AND S. LEE. 2005.
Inhibition kinetics of mushroom tyrosinase by copperchelating ammonium tetrathiomolybdate. Biochim. Biophys.
Acta 1726: 115–120, doi:10.1016/j.bbagen.2005.06.010
PARSONS, T. R. 1984. A manual of chemical and biological
methods for seawater analysis. Pergamon Press.
PRICE, N. M., G. I. HARRISON, J. G. HERING, R. J. HUDSON, P. M.
V. NIREL, B. PALENIK, AND F. M. M. MOREL. 1988–1989.
Preparation and chemistry of the artificial algal culture
medium Aquil. Biol. Oceanogr. 6: 443–461.
SHANK, G. C., S. A. SKRABAL, R. F. WHITEHEAD, AND R. J. KIEBER.
2004. Strong copper complexation in an organic-rich estuary:
the importance of allochthonous dissolved organic matter.
Mar. Chem. 88: 21–39, doi:10.1016/j.marchem.2004.03.001
SONG, B. W., AND B. B. WARD. 2003. Nitrite reductase genes in
halobenzoate degrading denitrifying bacteria. FEMS Microbiol. Ecol. 42: 349–357, doi:10.1111/j.1574-6941.2003.
tb01075.x
STIEFEL, E. I. 2007. Bioinorganic chemistry and the biogeochemical cycles, p. 7–30. In I. Bertini, H. Gray, E. Stiefel, and J.
Valentine [eds.], Biological inorganic chemistry. University
Science Books. p. 7–30.
TWINING, B. S., S. E. MYLON, AND G. BENOIT. 2007. Potential role
of copper availability in nitrous oxide accumulation in a
temperate lake. Limnol. Oceanogr. 52: 1354–1366,
doi:10.4319/lo.2007.52.4.1354
WARD, B. B., AND J. C. PRISCU. 1997. Detection and characterization of denitrifying bacteria from a permanently icecovered Antarctic lake. Hydrobiologia 347: 57–68,
doi:10.1023/A:1003087532137
———, C. B. TUIT, A. JAYAKUMAR, J. J. RICH, J. MOFFETT, AND S.
W. A. NAQVI. 2008. Organic carbon, and not copper, controls
denitrification in oxygen minimum zones of the ocean. DeepSea Res. I 55: 1672–1683, doi:10.1016/j.dsr.2008.07.005
ZUMFT, W. G. 1997. Cell biology and molecular basis of
denitrification. Microbiol. Mol. Biol. Rev. 61: 533–570.
Associate Editor: Mary I. Scranton
Received: 29 June 2011
Accepted: 26 October 2011
Amended: 27 October 2011
Fly UP