Electrocatalytic Reductions of Nitrite, Nitric Oxide, and Nitrous Oxide by Thermophilic Cytochrome P450 CYP119 in Film-Modified Electrodes and an Analytical Comparison of Its Catalytic Activities with Myoglobin

Article abstract of DOI:10.1021/ja038925c

Previous investigations of nitrite and nitric oxide reduction by myoglobin in surfactant film modified electrodes characterized several distinct steps in the denitrification pathway, including isolation of a nitroxyl adduct similar to that proposed in the P450nor catalytic cycle. To investigate the effect of the axial ligand on these biomimetic reductions, we report here a comparison of the electrocatalytic activity of myoglobin (Mb) with a thermophilic cytochrome P450 CYP119. Electrocatalytic nitrite reduction by CYP119 is very similar to that by Mb: two catalytic waves at analogous potentials are observed, the first corresponding to the reduction of nitric oxide, the second to the production of ammonia. CYP119 is a much more selective catalyst, giving almost exclusively ammonia during the initial half-hour of reductive electrolysis of nitrite. More careful investigations of specific steps in the catalytic cycle show comparable rates of nitrite dehydration and almost identical potentials and lifetimes for ferrous nitroxyl intermediate (Fe^II-NO^-) in CYP119 and Mb. The catalytic efficiency of nitric oxide reduction is reduced for CYP119 as compared to Mb, attributable to both a lower affinity of the protein for NO and a decreased rate of N-N coupling. Isotopic labeling studies show ammonia incorporation into nitrous oxide produced during nitrite reduction, as has been termed co-denitrification for certain bacterial and fungal nitrite reductases. Mb has a much higher co-denitrification activity than CYP119. Conversely, CYP119 is shown to be slightly more efficient at the two-electron reduction of N₂O to N₂. These results suggest that thiolate ligation does not significantly alter the catalytic reactivity, but the dramatic difference in product distribution may suggest an important role for protein stability in the selectivity of biocatalysts.

Full text of DOI:10.1021/ja038925c

Published on Web 03/26/2004
Electrocatalytic Reductions of Nitrite, Nitric Oxide, and
Nitrous Oxide by Thermophilic Cytochrome P450 CYP119 in
Film-Modified Electrodes and an Analytical Comparison of Its
Catalytic Activities with Myoglobin
†
‡
§
Chad E. Immoos, Ju Chou, Mekki Bayachou, Emek Blair, John Greaves, and
Patrick J. Farmer*
Contribution from the Department of Chemistry, UniVersity of California,
IrVine, California 92697-2025
Received October 7, 2003; E-mail: pfarmer@uci.edu
Abstract: Previous investigations of nitrite and nitric oxide reduction by myoglobin in surfactant film modified
electrodes characterized several distinct steps in the denitrification pathway, including isolation of a nitroxyl
adduct similar to that proposed in the P450nor catalytic cycle. To investigate the effect of the axial ligand
on these biomimetic reductions, we report here a comparison of the electrocatalytic activity of myoglobin
(Mb) with a thermophilic cytochrome P450 CYP119. Electrocatalytic nitrite reduction by CYP119 is very
similar to that by Mb: two catalytic waves at analogous potentials are observed, the first corresponding to
the reduction of nitric oxide, the second to the production of ammonia. CYP119 is a much more selective
catalyst, giving almost exclusively ammonia during the initial half-hour of reductive electrolysis of nitrite.
More careful investigations of specific steps in the catalytic cycle show comparable rates of nitrite dehydration
II
-
and almost identical potentials and lifetimes for ferrous nitroxyl intermediate (Fe -NO ) in CYP119 and
Mb. The catalytic efficiency of nitric oxide reduction is reduced for CYP119 as compared to Mb, attributable
to both a lower affinity of the protein for NO and a decreased rate of N-N coupling. Isotopic labeling
studies show ammonia incorporation into nitrous oxide produced during nitrite reduction, as has been termed
co-denitrification for certain bacterial and fungal nitrite reductases. Mb has a much higher co-denitrification
activity than CYP119. Conversely, CYP119 is shown to be slightly more efficient at the two-electron reduction
2 2
of N O to N . These results suggest that thiolate ligation does not significantly alter the catalytic reactivity,
but the dramatic difference in product distribution may suggest an important role for protein stability in the
selectivity of biocatalysts.
5
Introduction
thiolate and the N OR from Wolinella contains an Fe-heme
2
6
adjacent to a binuclear Cu-heme catalytic center.
Heme proteins play many catalytic roles in the biocycling of
nitrogen. For instance, assimilatory nitrite reductases (as NiR)
catalyze the multielectron reduction of nitrite to ammonia, which
Scheme 1
as NiR
NOR
1
NO2-
is then utilized, or assimilated, by the plants. The dissimilatory
8NH3
nitrite reductases (ds NiR) are part of a multienzyme denitri-
fication pathway that, in combination with nitric oxide reduc-
tases (NoR) and nitrous oxide reductases (N2OR), transforms
nitrite to dinitrogen, Scheme 1. Examples of heme-containing
enzymes are known for each of these steps; for instance, several
assimilatory NiR contain multiple type c hemes, while the
dissimilatory NiR contain heme cd1, a fungal NOR, cytochrome
P450nor, contains an Fe-heme coordinated by a cysteinyl
N OR
ds NiR
2
NO2-
8 NO
8 N O
8 N2
2
2
Cytochrome P450nor, a fungal enzyme which catalyzes the
reduction of NO to N2O, has been the focus of much recent
interest, including spectroscopic, mechanistic, and synthetic
3
7
8
4
9
model studies. The ferric-nitrosyl complex of P450nor reacts
(
3) (a) Einsle, O.; Messerschmidt, A.; Huber, R.; Kroneck, P. M. H.; Neese,
F. J. Am. Chem. Soc. 2002, 124, 11737. (b) Rudolf, M.; Einsle, O.; Neese,
F.; Kroneck, P. M. H. Biochem. Soc. Trans. 2002, 30, 649.
*
Corresponding author.
Present address: Department of Chemistry, Duke University, Durham,
†
(
4) F u¨ l o¨ p, V.; Moir, J. W. B.; Ferguson, S. J.; Hajdu, J. Cell 1995, 81, 369.
NC 27710.
‡
(5) Shoun, H.; Tanimoto, T. J. Biol. Chem. 1991, 266, 11078.
Present address: Department of Chemical Engineering, University of
(
6) (a) Liang, J.; Burris, R. H. J. Bacteriol. 1989, 171, 3176. (b) Drummond,
J. T.; Matthews, R. G. Biochemistry 1994, 33, 3732. (c) Drummond, J. T.;
Matthews, R. G. Biochemistry 1994, 33, 3742.
California, Santa Barbara, CA 93106.
§
Present address: Department of Chemistry, Cleveland State University,
Cleveland, OH 44115.
(7) Obayashi, E.; Tsukamoto, K.; Adachi, S.; Takahashi, S.; Nomura, M.;
Iizuka, T.; Shoun, H.; Shiro, Y. J. Am. Chem. Soc. 1997, 119, 7807.
(8) Daiber, A.; Nauser, T.; Takaya, N.; Kudo, T.; Weber, P.; Hultschig, C.;
Shoun, H.; Ullrich, V. J. Inorg. Biochem. 2002, 88, 343.
(
1) Averill, B. A. Chem. ReV. 1996, 96, 2951.
(
2) (a) Payne, W. J. Denitrification; John Wiley & Sons, Inc.: New York,
1
981. (b) Zumft, W. G. Microbiol. Mol. Biol. ReV. 1997, 61, 553.
4934
9
J. AM. CHEM. SOC. 2004, 126, 4934-4942
10.1021/ja038925c CCC: $27.50 © 2004 American Chemical Society

Electrocatalytic Nitrite Reduction by CYP119
A R T I C L E S
bined with its thermo- and barostability,21 make CYP119 an
excellent candidate for protein film voltammetry.
Scheme 2
2
2
Experimental Section
Materials. Expressed thermophilic cytochrome P450 CYP119 was
obtained from the laboratories of Thomas Poulos at University of
California, Irvine, and Paul Ortiz de Montellano at University of
California, San Francisco. The expressed enzyme samples were used
as received without further purification unless otherwise indicated.
1
5
Horse skeletal muscle myoglobin was obtained from Sigma. All N-
labeled compounds were obtained from Isotec. Nitric oxide obtained
from Air Gas was further purified by bubbling through a 1 M NaOH
solution. [ N}Nitric oxide was generated from L-ascorbic acid and [ N]-
sodium nitrite and purified by the same procedure. Angeli’s salt,
with NADH to generate an identifiable ferrous-nitroxyl inter-
II
-
mediate (Fe -NO ) which reacts rapidly with nitric oxide to
15
15
8
give nitrous oxide, Scheme 2. Although its spectroscopic and
structural properties are nearly identical to those of other P450
enzymes, P450nor does not exhibit the usual monooxygenation
activity of other P450 enzymes. Theoretical interest has revolved
around the effect of the thiolate coordination on the reduced
nitroxyl intermediate and its subsequent reactivity.1
Na
lized yield, 89%; ꢀ ) 8550 M cm at 250 nm). Sodium hyponitrite,
Na , was prepared by modification of methods of Scott and
2 2 3
N O , was prepared and purified by the method of Hunt (recrystal-
-1
-1
23
2
2 2
N O
-1
-1
Polydoropoulos, (recrystallized yield, 11%; ꢀ ) 6980 M cm at 248
0
24
nm). Dimethyldidodecylammonium bromide (DDAB) was purchased
A number of iron-porphyrin and other transition metal
complexes demonstrate nitrite or nitric oxide reductase activ-
from Acros. All other chemicals were reagent grade and used without
further purification. Water was purified with a Barnstead Nanopure
11,12
-2
ity.
Likewise, simple heme proteins such as myoglobin (Mb)
system to a specific resistance of >18 MΩ/cm .
or hemoglobin function such as NOx reductases when electro-
Electrochemical Apparatus and Procedures. A BAS 100B/W
electrochemical system was used for cyclic voltammetry (CV). A three
electrode cells consisting of a Pt wire counter electrode, silver/silver
chloride reference electrode, and basal plane pyrolytic graphite (PG)
electrodes were used for all experiments. The reference compartment
was separated from the working compartment by a modified Luggin
capillary. Basal plane pyrolytic electrodes were made by sealing a PG
cylinder into a glass tube with epoxy and connected to a wire with
silver epoxy paste. Electrodes were prepared by roughening the surface
with 400-grit SiC paper and cleaning in an ultrasonic bath for 10 min
prior to surface modification. Experiments were performed at room
temperature in 50 mM sodium phosphate buffer, pH 7.0, containing
chemically reduced in surface-modified electrodes.1
3-15
This
method has led to key findings on several distinct steps in the
denitrification pathway, including isolation of a stable nitroxyl
adduct of myoglobin analogous to the intermediate proposed
in the P450nor catalytic cycle,16 a kinetic model of the N-N
coupling reaction producing N2O,17 and the first example of a
heme-catalyzed reduction of nitrous oxide producing dinitrogen
and water,18 which is the ultimate transformation in denitrifi-
cation.
The proximal histidine ligation of the heme in myoglobin is
a poor mimic for the cysteinate coordination found in cyto-
chrome P450nor. To investigate the effect of the axial ligand
on the catalysis of nitrite and nitric oxide reductions, we report
here the electrochemical NOx activity of cytochrome P450
CYP119, a thermostable P450 enzyme. CYP119 is isolated from
Sulfolobus solfataricus, a hyperthermophilic, acidophilic ar-
1
00 mM supporting electrolyte, either KCl or NaBr. The solution was
purged with nitrogen for at least 20 min to remove oxygen before each
experiment. A nitrogen atmosphere was maintained over the solution
during all the experiments. Digital simulations of voltammograms at
pH 7 were performed using the BAS Digisim program; each simulation
was repeated until the simulated peak current and peak potential
matched that of experimental results.
Protein/DDAB Film Preparation. A 10 µL portion of 0.01 M
DDAB solution and another 10 µL portion of 8 mg/mL protein solution
were cast onto the surface of a basal plane pyrolytic graphite (PG)
electrode. The wet film electrode was covered with a test tube and
allowed to stand overnight before being uncovered and allowed to dry
in air. Film electrodes were allowed to dry for at least 24 h prior to
use. Protein film electrodes of the NO adducts of enzymes were also
prepared in an anaerobic glovebox under nitrogen for single-turnover
experiments. Ferrous nitrosyl adducts were prepared by the addition
of sodium dithionite and sodium nitrite to a solution of the enzyme
followed by purification by size-exclusion chromatography.
1
9
chaebacteria found in sulfurous volcanic hot springs. The
crystal structure of CYP119 reveals that the enzyme is smaller
and more compact than typical P450 enzymes,20 which com-
(
9) Suzuki, N.; Higuchi, T.; Urano, Y.; Kikuchi, K.; Uchida, T.; Mukai, M.;
Kitagawa, T.; Nagano, T. J. Am. Chem. Soc. 2000, 122, 12059.
(
10) (a) Harris, D. L. Int. J. Quantum Chem. 2002, 88, 183. (b) Silaghi-
Dumitrescu, R. Eur. J. Inorg. Chem. 2003, 1048.
(
11) (a) Murphy, W. R., Jr.; Takeuchi, K.; Barley, M. H.; Meyer, T. J. J. Am.
Chem. Soc. 1982, 104, 5817. (b) Barley, M. H.; Takeuchi, K.; Meyer, T.
J. J. Am. Chem. Soc. 1986, 108, 5876. (c) Barley, M. H.; Rhodes, M. R.;
Meyer, T. J. Inorg. Chem. 1987, 26, 1746. (d) Younathan, J. N.; Wood, K.
S.; Meyer, T. J. Inorg. Chem. 1992, 31, 3280. (e) Bedioui, F.; Trevin, S.;
Albin, V.; Villegas, M. G. G.; Devynck, J. Anal. Chim. Acta 1997, 341,
1
77.
Rotating Disk Electrode Voltammetry. Rotating disk electrode
experiments utilized a basal plane pyrolytic graphite disk as the working
(
12) (a) Toth, J. E.; Anson, F. C. J. Am. Chem. Soc. 1989, 111, 2444. (b) Rhodes,
M. R.; Barley, M. H.; Meyer, T. J. Inorg. Chem. 1991, 30, 629. (c)
Sunohara, S.; Nishimura, K.; Yahikozawa, K.; Ueno, M.; Eny, M.; Takasu,
Y. J. Electroanal. Chem. 1993, 354, 161. (d) Keita, B.; Belhouri, A.; Nadjo,
L.; Constant, R. J. Electroanal. Chem. 1995, 381, 243.
(19) (a) Wright, R. L.; Harris, K.; Solow, B.; White, R. H.; Kennelly, P. J.
FEBS Lett. 1996, 384, 235. (b) McLean, M. A.; Mavew, S. A.; Weiss, M.
K.; Krepich, S.; Sligar, S. G. Biochem. Biophys. Res. Commun. 1998, 252,
166.
(20) (a) Yano, J. K.; Koo, L. S.; Schuller, D. J.; Li, H.; Ortiz de Montellano, P.
R.; Poulos, T. L. J. Biol. Chem. 2000, 275, 31086. (b) Park, S.; Yamane,
K.; Adachi, S.; Shiro, Y.; Weiss, K. E.; Sligar, S. G. Acta Crystallogr.
2000, D56, 1173.
(21) Koo, L. S.; Tschirret-Guth, R. A.; Straub, W. E.; Moenne-Loccoz, P.; Loehr,
T. M.; Ortiz de Montellano, P. R. J. Biol. Chem. 2000, 275, 14112.
(22) Koo, L. S.; Immoos, C. E.; Cohen, M. S.; Farmer, P. J.; Ortiz de Montellano,
P. R. J. Am. Chem. Soc. 2002, 124, 5684.
(23) Hunt, H. R., Jr.; Cox, J. R., Jr.; Ray, J. D. Inorg. Chem. 1962, 1, 938.
(24) (a) Scott, A. W. J. Am. Chem. Soc. 1927, 49, 986. (b) Polydoropoulos, C.
N. Chem. Chron. 1959, 24, 147.
(13) (a) Lin, R.; Bayachou, M.; Greaves, J.; Farmer, P. J. J. Am. Chem. Soc.
1
997, 119, 12689. (b) Farmer, P. J.; Lin, R.; Bayachou, M. Comments Inorg.
Chem. 1998, 20, 101.
(14) Mimica, D.; Zagal, J. H.; Bedioui, F. J. Electroanal. Chem. 2001, 497,
1
06.
(15) (a) Zhou, Y. L.; Hu, N. F.; Zeng, Y. H.; Rusling, J. F. Langmuir 2002, 18,
2
11. (b) Huang, R.; Hu, N. F. Biophys. Chem. 2003, 104, 199. (c) Huang,
R.; Hu, N. F. Bioelectrochemistry 2001, 54, 75. (d) Liu, H. Y.; Wang, L.
W.; Hu, N. F. Electrochim. Acta 2002, 47, 2515.
(
16) Lin, R.; Farmer, P. J. J. Am. Chem. Soc. 2000, 122, 2393.
17) Bayachou, M.; Lin, R.; Cho, W.; Farmer, P. J. J. Am. Chem. Soc. 1998,
(
1
20, 9888.
(18) Bayachou, M.; Elkbir, L.; Farmer, P. J. Inorg. Chem. 2000, 39, 289.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 15, 2004 4935

A R T I C L E S
Immoos et al.
Figure 1. Cyclic voltammograms of CYP119/DDAB in the absence
Figure 2. Cyclic voltammograms of the pH dependence of CYP119/DDAB
in the presence of 34 mM sodium nitrite at (a) pH 5.2, (b) pH 6.0, (c) pH
(dotted) and presence (solid) of 34 mM sodium nitrite. Conditions: 50 mM
phosphate; pH 7.0; 100 mM NaBr; scan rate ) 50 mV/s. Control
experiments are shown as a dashed line, under the same conditions on a
DDAB-modified electrode without CYP119.
6.5, (d) pH 7.0, (e) pH 7.5, (f) pH 8.2, (g) pH 8.9, (h) pH 9.6, and (i) no
nitrite. Conditions: 50 mM buffer; 100 mM NaBr; scan rate ) 100 mV/s.
displays current controlled by linear diffusion at scan rates above
electrode. The working electrode was prepared in a manner similar to
that mentioned previously for the preparation of protein film electrodes.
Experiments were carried out with a conventional three-electrode
configuration as described above. Rotating disk electrode experiments
were conducted using a Radiometer rotating disk electrode. Only
rotation rates up to 400 rpm were used to prevent the protein film from
separating from the electrode surface at higher rotation speeds.
Bulk Electrolysis. For bulk electrolysis reactions, pyrolitic graphite
plates (1.0 cm × 1.0 cm × 0.2 cm) coated with Mb or CYP119 and
5
00 mV/s, where the current is proportional to the square root
of the scan rate (Supporting Information S2). At rates below
5
00 mV/s, the current is proportional to the scan rate, typical
17
of the thin-film behavior seen for other DDAB/protein systems.
As seen previously for Mb/DDAB films, the R1 couple varies
linearly ca. -48 mV/pH unit in the range from pH 5.5 to 10,
but the R2 couple remains constant over this range (Supporting
Information S3).
1
7
DDAB were used as working electrodes. Controlled potential was
applied by using a BAS 100 W and a Power Module potentiostat, an
Ag/AgCl reference, and a Pt grid counter electrode separated by a fine
frit from the main compartment. The headgas over the electrolysis
solution was sampled in a loop and transferred directly into a Micromass
Autospec mass spectrometer via a GC interface through a Plot GC
Nitrite Reduction. Addition of sodium nitrite to the elec-
trochemical cell changes the electrochemical response of
III
II
CYP119/DDAB, Figure 1. The current at the Fe /Fe couple
decreases upon addition of nitrite but does not shift in potential.
At more negative potentials two new catalytic waves are seen
which are dependent on nitrite concentration, with the half-wave
potential of R3 at ca. -0.86 V and that for R4 at ca. -1.11 V.
column (fused silica 50 m × 0.32 mmsAl
2 3
O , KCl coating). Data were
acquired in EI mode at 1000 resolution. NO was generated by
1
5
1
5
The electrocatalytic current at R exhibits saturation kinetics,
treatment of Na NO
2
with Fe(SO
4
2
)/H SO
4
, with a He carrier gas used
4
1
5
to transfer the NO into the electrolysis cell through a fused silica
increasing to a limiting value at ∼34 mM. The kcat. for nitrite
1
7
line.
reduction by CYP119 can be calculated from a Lineweaver-
Burke plot, yielding a Km value of 11.5 mM (Supporting
Information S4 and S5).
Nitrite Reduction. For the concentration dependence of nitrite
reduction, aliquots of concentrated nitrite solution (1 M) were added
via a gastight syringe. At each concentration, the solution was stirred
vigorously upon addition and degassed for at least 5 min before each
experiment. Colorimetric determinations of solution-based nitrite,
hydroxylamine, and ammonia were conducted as previously described.13
Nitric Oxide and Nitrous Oxide Reduction. An excess of pre-
purified nitric oxide or nitrous oxide was bubbled through a thoroughly
degassed solution of 50 mM phosphate, pH 7, with 100 mM NaBr as
supporting electrolyte within the electrochemical cell. Special care was
taken to ensure that the electrochemical cell was airtight and free from
oxygen before the addition of nitric oxide. After addition, the solution
was kept under static nitrogen pressure throughout the experiments.
The electrocatalytic currents are strongly pH dependent, as
shown in Figure 2. At pH 5.2, the maximum peak current ratios
II
I
compared to the noncatalytic current for the Fe /Fe peak are
∼
4 for R3 and ∼40 for R4. The catalytic efficiency for R3 is
more strongly affected by pH, decreasing significantly as the
pH is increased, as expected for a proton-dependent process.
II
I
At pH > 9, the current does not exceed that of the Fe /Fe
reduction process in the absence of nitrite. The solution pH
affects the two catalytic reduction peaks differently. The
potential of R shifts ∼-40 mV/pH unit with a small decrease
4
in catalytic current with increasing pH. The potential of R3 has
Results
smaller pH dependence, shifting ∼-20 mV/pH unit with the
catalytic current decreasing significantly with increasing pH.
Voltammograms of CYP119/DDAB are unaffected by the
presence of ammonia at 50 mM (Supporting Information S6);
hydroxylamine at 34 mM does induce a small catalytic response
beginning at the R1 potential (Supporting Information S7).
Under the conditions used during analytical voltammetric
experiments, neither of these products is expected to influence
the current response observed.
Deposition of CYP119/DDAB films was analogous to those
of Mb/DDAB;13,16 films deposited on quartz displayed a ferric
P450 Soret band at 418 nm, with distinctive Q-band splitting
between 500 and 600 nm, both of which suggest retention of
the native thiolate-ligated heme environment (Supplemental S1).
A typical cyclic voltammogram for a CYP119/DDAB film on
a basal plane pyrolytic graphite electrode in pH 7 buffered
solution (100 mM NaBr electrolyte) is shown in Figure 1. Two
III
well-defined reduction couples are evident, assigned to the Fe /
In nitrite solutions, limiting the potential window from +0.20
II
II
I
III
II
Fe redox couple, R1, at -0.21 V vs Ag/AgCl, and the Fe /Fe
redox couple, R2 at -1.04 V vs Ag/AgCl. The R1 couple
to -0.60 V results in loss of the Fe /Fe redox couple after
successive scans, Figure 3. Holding the electrode potential at
4936 J. AM. CHEM. SOC.
9
VOL. 126, NO. 15, 2004

Electrocatalytic Nitrite Reduction by CYP119
A R T I C L E S
Figure 3. Successive scans of CYP119/DDAB in the presence of 34 mM
3
+/2+
nitrite, showing loss of the Fe
couple. Conditions: 50 mM phosphate;
Figure 6. Rotating disk electrode voltammetry of Mb/DDAB under varying
concentrations of nitrite: (a) none; (b) 2 mM; (c) 4 mM; (d) 8 mM; (e) 12
mM. Conditions: 50 mM phosphate buffer; pH 7; 100 mM NaBr; scan
rate ) 20 mV/s; rotation rate, 400 rpm. Enlarged plot shows the loss of
current at -250 mV with increasing nitrite concentration.
pH 7.0; 100 mM NaBr; scan rate ) 100 mV/s.
Figure 4. Rapid scan of CYP119/DDAB in the presence of 34 mM nitrite
after holding the potential at -0.4 V for 20 s. Conditions: 50 mM sodium
phosphate; pH 7.0; 100 mM NaBr; scan rate ) 5 V/s.
Figure 7. Analysis of ammonia (striped) and hydroxylamine (solid) at
different time points during bulk electrolysis of a 10 mM NaNO2 solution,
pH 6.0 (100 mM phosphate buffer) using (A) Mb/DDAB and (B) CYP119/
DDAB electrodes at -1.2V vs Ag/AgCl. Yields given as percentage of
nitrite consumed at equivalent times; data in B average of four trials. Note
the difference in y-axis scales.
tions; in agreement with previous reports. In contrast, the
cathodic current associated with the Fe /Fe redox couple
decreases with increasing nitrite concentration.
Figure 5. Regeneration of the Fe3+/Fe2+ redox couple after holding the
potential at -0.4 V for 60 s. First two scan segments are in solid lines; the
third scan segment is dashed. Conditions: 50 mM sodium phosphate; pH
III
II
7
.0; 100 mM NaBr; scan rate ) 100 mV/s.
Product Analysis. Controlled-potential electrolysis experi-
ments at the potential of -1.2 V of CYP119/DDAB in nitrite
solutions at pH 6 were used to determine the products of the
catalysis. The aqueous products ammonia and hydroxylamine
were determined by the standard colorimetric methods. A
comparison of the product distributions under identical experi-
ments using CYP119 and Mb as electrocatalyst is shown in
Figure 7.
-
0.40 V for 20 s prior to scanning results in a reversible redox
couple at -0.85 V, Figure 4. Equivalent cathodic and anodic
currents for this couple, with ipa/ipc ∼ 1, is only observed at
scan rates >1 V/s; at slower scan rates, the reduction is
irreversible and the Fe /Fe redox couple reappears, Figure 5.
Rotating disk voltammetry of Mb/DDAB were performed in
the presence of varying concentrations of nitrite, Figure 6. The
steady-state catalytic wave at -1.1 V vs Ag/AgCl shows an
increase of the limiting current with increasing nitrite concentra-
III
II
To identify the gaseous products of the catalytic reduction
1
5
of nitrite, N-labeled sodium nitrite solution at pH 6 was
electrolyzed at -1.2 V; separate experiments were performed
J. AM. CHEM. SOC.
9
VOL. 126, NO. 15, 2004 4937

A R T I C L E S
Immoos et al.
Figure 10. Gas chromatograms of sampled headgas above the reaction of
natural abundance Na2N2O2 with 15NH4Cl in pH 10 buffered solution,
14
Figure 8. Relative yields of 15N2O from GC/MS analysis of sampled
headgas over bulk electrolysis at -1.2 V of a 10 mM Na NO2, pH 6.0
using Mb/DDAB (squares) or CYP119/DDAB (circles) electrocatalysts of
comparable loadings.
2
demonstrating the formation of both N O, m/z ) 44, and mixed-label
4,15
1
1
5
2
N O, m/z ) 45. Minutes mark time from the mixing of reagents before
sampling of the headgas.
Figure 11. Cyclic voltammograms of CYP119/DDAB in the absence
(bottom, solid line) and presence of a saturated solution nitric oxide (top,
solid line). Conditions: 50 mM sodium phosphate; pH 7.0; 100 mM NaBr;
scan rate ) 100 mV/s. Control experiments are shown as a dashed line,
under the same conditions without CYP119.
dependently during the bulk electrolysis, demonstrated for Mb/
DDAB in Supporting Information S10.
Ammonia Incorporation. To probe incorporation of am-
15
monium into formation of N2O, a mixed solution of N-labeled
1
4
-
NH4Cl and N-labeled NO2 was subjected to bulk electrolysis
by CYP119/DDAB and Mb/DDAB. Analysis of the headgas
Figure 9. Time-based analysis of relative yields of 14N NO and N NO
14
15 14
from GC/MS analysis of sampled headgas over 3 h of bulk electrolysis at
14
14
1
4
15
by GC-MS shows growing signals identified as both N NO
and N NO, in which one N from 15N-labeled NH4Cl is
-
1.2 V of a 10 mM Na NO2 with 10 mM NH4Cl solution, pH 6.0: (A)
1
5
14
using the Mb/DDAB electrode; (B) using the CYP119/DDAB electrode.
The y-axis is the relative yield as a percentage of the maximum 14N NO
counts in MS.
14
incorporated, Figure 9.
To test for possible chemical species responsible for the
incorporation, experiments utilizing the two possible intermedi-
ates, hyponitrite and trioxodinitrate, were undertaken. Each was
then dissolved in pH 13 buffer in an airtight reaction cell and
using CYP119/DDAB and Mb/DDAB as catalysts using similar
catalyst loadings. The time-based formation of 15N2O from both
experiments is shown in Figure 8. The headgas over the reaction
vessel was sampled into a loop and then transferred into the
mass spectrometer via the GC interface using a fused silica
column; a 106 ppm concentration of Ar in the He mixture carrier
gas was used as a mass standard (Supporting Information S8).
After a few minutes of electrolysis, signals at m/z ) 46,
1
5
+
15
+
NH4 or NH3OH added during their decomposition to probe
15
the incorporation of N into the gas product N2O. The reaction
cell was connected to a mass spectrometer with fused silica line
1
5
+
through a GC interface. Addition of NH3OH to both solutions
1
4,15
generated mixed label
N2O, but only hyponitrite generated
15
+
the mixed label gaseous product upon addition of NH4 , Figure
10 and Supporting Information S11 and S12.
1
5
corresponding to N2O, were seen and grew with time (Sup-
porting Information S9) along with small peaks corresponding
NO Reduction. Addition of NO(g) to a thoroughly degassed
solution of 50 mM sodium phosphate buffer, pH 7, changes
the voltammetric response as seen in Figure 11. An irreversible
catalytic wave at -0.81 V, close to the R3 couple seen in the
15
+
15
+
to the fragments [ NO] and [ N2] . Column-separated peaks
1
5
+
15
+
of [ NO] (m/z 31) and [ N2] (m/z 30) are also readily
identified, indicating that both compounds are produced in-
4938 J. AM. CHEM. SOC.
9
VOL. 126, NO. 15, 2004

Electrocatalytic Nitrite Reduction by CYP119
A R T I C L E S
myoglobin and P450cam film electrodes, we assign the reduction
III
II
II
I
couples to the Fe /Fe and Fe /Fe redox couples, eqs 1 and 2.
III
-
II
Fe -+ e h Fe
(1)
(2)
II
-
I
Fe + e h Fe
NO2- + ne + mH f
-
+
4
+
+
NH + NH OH + N O + NO + N (3)
3
2
2
Like Mb, CYP119 in DDAB films acts as a highly active
catalyst for the reduction of nitrite in aqueous solutions, and a
similar range of products are produced, eq 3. By analogy, we
can identify two catalytic waves, the first, R3 corresponds to
the reduction of the nitric oxide adduct and the second R4 to
the production of ammonia. The catalytic efficiency of CYP119,
measured at 20 mV/s scan rate as current per mole substrate
per second, is 140 as opposed to over 500 for Mb. Thus thiolate
ligation does not enhance or significantly alter the catalytic
behavior.
But, time-based analysis of the products of CYP119 electro-
catalysis yielded unexpected results, Figure 7. First, ammonia
is the initial product formed with high apparent yield; hydroxy-
lamine is not detectable in the first hour, and only a small yield
of hydroxylamine is obtained after 2 h. In the first 20 min,
almost all the consumed nitrite is being used to produce
ammonia, but the percent of ammonia from consumed nitrite
decreased dramatically as the reaction progressed. For example,
Figure 12. Cyclic voltammograms of CYP119/DDAB in the absence
(dotted) and presence (solid) of a saturated solution of N2O. Conditions:
5
0 mM sodium phosphate; pH 7.0; 100 mM NaBr; scan rate ) 20 mV/s.
electrocatalytic reduction of nitrite by CYP119/DDAB, that
demonstrates the catalytic reduction of nitric oxide. The catalysis
exhibits a strong dependence on scan rate. An increase in the
scan rate results in attenuation of the catalytic peak and the
III
II
reappearance of the Fe /Fe redox couple. Attempts to observe
a reversible redox couple for the nitrosyl to nitroxyl transforma-
tion in the presence of excess nitric oxide were unsuccessful,
15
even at scan rates as high as 20 V/s. In experiments using NO
generated in a separate vessel from 15N-labled sodium nitrite,
mass spectral analysis of the headgas showed a large buildup
9
9% of consumed nitrite is being used to produce ammonia
15
after 20 min, but only about 20% after 4 h. In comparison, using
Mb/DDAB as a catalyst generates considerable gaseous products
immediately; after 20 min the aqueous products (ammonia and
hydroxylamine) account for less than 20% of the nitrite
of N2O (data not shown).
N2O Reduction. The CVs of CYP119/DDAB electrode in
the absence and presence of N2O are shown in Figure 12. The
III
II
addition of N2O causes no change in the Fe /Fe current but a
significant increase in current along with a loss of reversibility
1
3
consumed.
II
I
Time-based analysis of the gaseous products are also instruc-
tive; for Mb/DDAB formation of the main product, N2O, begins
with the initial electrolysis, but for CYP119/DDAB gas produc-
tion begins slowly and never reaches comparable levels, Figure
of Fe /Fe . The large current is assigned as the catalytic
reduction of N2O.
Discussion
8
. Thus, despite the similarity in electrochemical response, there
Electrochemical and Catalytic Activity of CYP119/DDAB.
Surfactant films have proven useful in obtaining direct elec-
trochemistry from a number of heme proteins, including
is a large difference in product distribution between the two
heme protein catalysts. To more fully understand this reactivity,
several specific steps in the complex reduction of nitrite were
probed by electrochemical methods, as described below.
Binding and Dehydration of Nitrite. The apparent similarity
of Mb and CYP119 in the electrocatalytic reduction of nitrite
is not found in solution binding studies of their ferric states.
For met Mb, the presence of 30 mM nitrite shifts the Soret peak
absorbance to 412 nm, and the Fe /Fe couple of Mb/DDAB
by ca. -50 mV, both consistent with the binding of nitrite to
the ferric heme, eq 4. The catalytic current dependence on nitrite
concentration yields an apparent Km ) 3.3 mM, very close to
the known Kd of nitrite to met Mb in solution. For CYP119,
a similar analysis yields an apparent Km ) 11.5 mM, but there
are no corresponding shifts in Soret absorbance or Fe /Fe
couple up to 40 mM nitrite (Supporting Information S13).
Although there are previous reports of spectral changes for
myoglobin,25 hemoglobin, and peroxidases. Rusling first
reported the electrochemistry of P450cam on film-modified
electrodes28 and we have previously obtained similar results for
26
27
22
P450 CYP119 in DDAB films. A typical cyclic voltammogram
for CYP119/DDAB film on a basal plane pyrolytic graphite
electrode in pH 7 buffered solution (100 mM NaBr electrolyte)
is shown in Figure 1A. In analogy to previous reports on
III
II
(
25) (a) Rusling, J. F.; Nassar, A. E. F. J. Am. Chem. Soc. 1993, 115, 11891.
(
b) Nassar, A. E. F.; Bobbit, J. M.; Stuart, J. D.; Rusling, J. F. J. Am.
Chem. Soc. 1995, 117, 10986. (c) Nassar, A. E. F.; Zhang, Z.; Chynwat,
V.; Frank, H. A.; Rusling, J. F.; Suga, K. J. Phys. Chem. 1995, 99, 11013.
1
3
(
d) Nassar, A. E. F.; Narikiyo, Y.; Sagara, T.; Nakashima, N.; Rusling, J.
F. J. Chem Soc., Faraday Trans. 1995, 91, 1775. (e) Nassar, A. E. F.;
Willis, W. S.; Rusling, J. F. Anal. Chem. 1995, 67, 2386.
III
II
(26) Lu, Z. Q.; Huang, Q. D.; Rusling, J. F. J. Electroanal. Chem. 1997, 423,
5
9.
(
27) (a) Zhang, Z.; Chouchane, S.; Magliozzo, R. S.; Rusling, J. F. Chem.
Commun. 2001, 177. (b) Zhou, Y. L.; Hu, N. F.; Zeng, Y. H.; Rusling, J.
F. Langmuir 2002, 18, 211.
29
P450cam in the presence of nitrite, these are irreproducible
for CYP119. Thus the initial step in the catalysis is likely not
reduction of the ferric-nitrite adduct, as in eq 5. Instead, the
(
28) (a) Zu, X. L.; Lu, Z. Q.; Zhang, Z.; Schenkman, J. B.; Rusling, J. F.
Langmuir 1999, 15, 7372. (b) Lvov, Y. M.; Lu, Z. Q.; Schenkman, J. B.;
Zu, X. L.; Rusling, J. F. J. Am. Chem. Soc. 1998, 120, 4073. (c) Zhang,
Z.; Nassar, A. E. F.; Lu, Z. Q.; Schenkman, J. B.; Rusling, J. F. J. Chem.
Soc., Faraday Trans. 1997, 93, 1769.
(29) Sono, M.; Dawson, J. H. J. Biol. Chem. 1982, 257, 5496.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 15, 2004 4939

A R T I C L E S
Immoos et al.
III
II
lack of a shift in the Fe /Fe redox couple in the presence of
electrode potential at -0.40 V for 20 s prior to rapidly scanning
to negative potentials resulted in a reversible redox couple at
E° ) -0.85 V vs Ag/AgCl, Figure 4, corresponding to the
reduction of the ferrous nitrosyl to nitroxyl, eq 8. This behavior
suggests that generation of the ferrous enzyme in the presence
of nitrite results in formation of the nitrosyl enzyme and allows
observation of the corresponding nitrosyl/nitroxyl redox couple.
Under similar conditions and scan rates, Mb/DDAB also exhibits
a redox couple corresponding to nitrosyl/nitroxyl reduction, at
potential identical to that previously reported. As was seen for
nitrite, in addition to the lack of a change in the UV-visible
III
II
spectra, suggests a simple reduction of Fe to Fe followed by
a bimolecular reductive dehydration of nitrite yielding a ferric-
NO complex, as was previously suggested for nitrite catalysis
by water-soluble iron-porphyrins, eq 6.1
1b
Fe + NO2 f Fe -NO2-
III
-
III
(4)
(5)
(6)
(7)
Fe -NO2- + e f Fe -NO
III
-
II
-
2
1
7
the preformed Mb-NO adduct, the reversibility of this redox
couple for CYP119 is strongly dependent on the scan rate,
becoming more reversible as the scan rate is increased; a
reversible signal, where ipa/ipc ) 1, is only observed at scan
rates > 1 V/s. At slower scan rates, the reduction is irreversible
II
-
+
III
-
Fe + NO2 + H f Fe -NO + OH
III
-
II
Fe -NO + e f Fe -NO
III
II
Using rotating disk electrochemistry, Bedioui and co-workers
reported that Mb/DDAB in the presence of nitrite does show a
current enhancement at Fe /Fe potentials. These reductive
waves, assigned to the single electron reduction of the [Fe -
NO)] adduct, had very slow charge-transfer characteristics
and were not observable by cyclic voltammetry. Our attempts
to repeat these experiments have given significantly different
results, Figure 6. The steady-state wave at -1.1 V vs Ag/AgCl
shows an increase of the limiting current with increasing nitrite
concentrations, in agreement with a previous report. In contrast,
the cathodic current associated with the Fe /Fe redox couple
decreases with increasing nitrite concentration. We attribute loss
of the Fe /Fe cathodic wave during rotating disk voltammetry
and the Fe /Fe redox couple reappears, Figure 5.
The heterogeneous rate constant for the reduction of the
ferrous CYP119 NO adduct was obtained by simulation of
voltammograms of ferrous NO generated in nitrite solutions,
III
II
14
III
+
0
-2
(
yielding K ) 7.3 × 10 cm/s (Supporting Information S14).
II -
As observed for Mb/DDAB, the reduced Fe -NO in CYP119/
DDAB film undergoes a pH dependent following reaction.
Digital simulation of a single turnover reduction at pH 7 yields
a rate of 45 s for the irreversible loss of the Fe-coordinated
NO; the comparable rate for Mb/DDAB is 22 s . It is our
current supposition that this following reaction is the formation
-
1
-
1 17
III
II
II
of the nitrosyl hydride adduct, Fe -HNO in eq 9, which we
III
II
have shown to be unusually stable in aqueous solutions from
II
16,30,31
experiments to a buildup of Fe -NO during the course of the
pH 7 to 10.
steady-state experiment.
The formation of Fe -NO can be observed indirectly by cyclic
voltammetry. Limiting the potential window to a range from
II
II
-
+
II
Fe -NO + H f Fe -HNO
(9)
III
II
+
0.20 to -0.40 V results in loss of the Fe /Fe redox couple
Catalytic NO Reductions. To investigate the NoR activity
of CYP119/DDAB, electrochemical studies were also performed
under an atmosphere of nitric oxide, resulting the voltammetric
response seen in Figure 11. As seen with Mb/DDAB, the redox
couple associated with the Fe /Fe redox couple disappears
upon addition of nitric oxide, and a catalytic wave develops at
-0.82 V, close to the catalytic wave R3 in Figure 1, and
approximately the same potential as the single electron reduction
of the ferrous NO adduct, eq 8. Mass spectral analysis of the
III
after successive scans, Figure 3. Plotting the loss of the Fe /
II
-3
Fe couple with time gives a rate of dehydration of 9.0 × 10
-
1
s
for CYP119/DDAB in 34 mM nitrite. Under similar
III
II
III
II
conditions, Mb/DDAB also exhibits a loss of the Fe /Fe redox
-
3
-1
couple under identical conditions at a rate of 9.2 × 10 s ,
well within the experimental error of that for CYP119. This
implies a similar mechanism is followed by both proteins.
Indeed, we have recently reported that the solution reactivity
of deoxy Mb with nitrite is zero order in Mb and first order in
both nitrite and protons,30 which again suggests that the initial
1
5
products of electrocatalytic reduction of NO gives strong
signals corresponding to N2O (46 m/z). The observed catalysis
15
II
step is dependent on the reaction of Fe with HONO, or its
is substrate-limited, as indicated by the lack of steady-state
current in Figure 11; at lower scan rates the current drop is
lessened (Supporting Information S14). Increasing the scan rate
results in the catalytic peak decreasing and the Fe /Fe redox
couple reappearing (Supporting Information S15), but no
reversibility was observed at the catalytic wave, i.e., no anodic
current similar to that seen in Figure 4, even at scan rates as
high as 20 V/s.
The irreversibility of the nitric oxide reduction couple is due
to reaction of the ferrous-bound nitroxyl with nitric oxide from
bulk solution. This models a key step in the enzymatic activity
of cytochrome P450nor, as the last step in the catalytic cycle
shown in Scheme 1. Digital simulations of catalytic voltam-
mograms of the catalytic reduction of NO on a CYP119/DDAB
+
equivalent as NO .
II
Single Turnover Reduction of Fe -NO. For Mb/DDAB it
III
II
was shown that, the R3 wave corresponds to the reduction of
the ferrous NO adduct.17 Using preformed NO-Mb/DDAB
films, a reversible single-electron reduction was observed to a
reduced nitroxyl species, Fe -NO in eq 8, whose lifetime was
pH-dependent. The intermediacy of nitric oxide can also be
suggested in the electrochemical reduction of nitrite by CYP119
in surfactant film, but similar characterization of the single-
electron reduction of the NO adduct is precluded because for
CYP119 the ferrous NO adduct is unstable over the time needed
to form a modified electrode.
II
-
II
-
II
-
Fe -NO + e f Fe -NO
(8)
(
30) Sulc, F.; Immoos, C.; Pervitsky, D.; Farmer, P. J. J. Am. Chem. Soc. 2004,
126, 1096.
A simple method was derived to generate the NO adduct in
situ via the reductive dehydration of nitrite. Holding the
(
31) Sulc, F.; Fleischer, E.; Farmer, P. J.; Ma, D.; La Mar, G. J. Biol. Inorg.
Chem. 2003, 8, 348.
4940 J. AM. CHEM. SOC.
9
VOL. 126, NO. 15, 2004

Electrocatalytic Nitrite Reduction by CYP119
A R T I C L E S
natural abundance nitrite solution during bulk electrolysis by
both Mb/DDAB and CYP119/DDAB, and indeed there was
1
5
14
evidence by mass spectrometry of formation of N NO (m/z
4
5) in both systems, Figure 9. For CYP119, the signal of
14
14
N NO is initially observed after 8 min, but a quite small signal
of 15N NO is only seen after 30 min and grew slowly during
the electrolysis; the final ratio of two gases is 3.3:1 after 3 h.
Under the same experimental conditions with Mb/DDAB as the
electrocatalyst, the formation of N NO is much higher
initially, and the incorporation of the labeled ammonia, seen as
14
1
4
14
15
14
the signal for mixed-labeled N NO, is greatly enhanced. After
h of electrolysis the mixed-label product becomes dominant,
3
at a ratio of 1:1.3.
In both cases, a lag time is evident before label incorporation,
which suggests that incorporation may be due to buildup of a
transient, solution-based intermediate. The increased incorpora-
tion during catalysis by Mb/DDAB perhaps reflects the multi-
tude of side products as seen in the product analysis. We had
Figure 13. Experimental (dotted) and simulated (solid) cyclic voltammo-
grams for CYP119/DDAB film electrode in saturated NO solutions.
Conditions: 50 mM sodium phosphate; pH 7.0; 10 mM NaCl; scan rate )
3
V/s.
-
film were performed to obtain rate constants for the catalytic
coupling step, Figure 13. Because of the similarity in catalytic
voltammograms of NO reduction by Mb/DDAB,17 the same
mechanistic steps were used in the simulation, which encompass
in sequence eqs 1, 10, 8, and 11.
previously suggested that free NO , or HNO, was an initial
17
byproduct of nitrite reduction by Mb/DDAB; recent work has
shown that HNO is electrophilic, for instance in its reaction
with thiols that ultimately generates hydroxylamine and disul-
36
fide, eqs 12 and 13. In this regard, a similar reactivity between
HNO and labeled ammonia may lead to the mixed-labeled N2O
product. With this in mind, two possible HNO intermediates,
trioxodinitrate and hyponitrite, were tested for incorporation
labeled ammonia into the gaseous N2O product. By this logic
Angeli’s salt decomposition, which generates HNO as in eq 14,
would incorporate the labeled ammonia into the N2O product;
II
II
Fe + NO f Fe -NO
(10)
(11)
II
-
+
III
-
Fe -NO + NO + H f Fe + N O + HO
2
The Kf of ferrous NO binding to CYP119, eq 10, is 1.1 ×
0 M s , 2 orders of magnitude smaller than that of Mb,
which is generally typical of P450 enzymes including P450nor.
7
-1 -1 32
1
-
hyponitrite, which is the product of NO coupling and a direct
33
precursor to N2O, eqs 15 and 16, would not. As seen in the
mass spectral chromatograms in Figure 10, the decomposition
Using these values, the digital simulation of a voltammogram
of NO reduction in a saturated NO solution at pH 7 was
obtained, Figure 13. Kcat value of the catalytic reaction of NO
15
+
of hyponitrite upon addition of NH4 indeed does generate
the mixed-label
14,15
N2O but, contrary to our expectations, that
II
-
6
-1 -1
with Fe -NO , eq 11, was determined as 3 × 10 M s ,
of Angeli’s salt does not (Supporting Information S10). For both
8
-1 -1
significantly slower than that of Mb, at 1 × 10 M s . Thus,
the thiolate ligation of CYP119 does not increase its effective-
ness in N-N coupling reaction central to the P450nor reactivity.
Ammonia Incorporation during N2O Formation. Certain
denitrifying bacteria such as P. stutzeri and denitrifying fungi
such as Fusarium oxysporum, can use a secondary nitrogen atom
source such as azide, ammonium ion, and hydroxylamine to
form N2O under denitrifying conditions, a process termed “co-
denitrification”. These secondary nitrogen sources are inca-
pable by themselves of inducing N2O production, but their
incorporation into the gas during nitrite turnover is demonstrated
by N-labeling studies. Our results show that both Mb and
CYP119 can incorporate ammonia into the nitrous oxide
produced during the catalytic reduction of nitrite.
15
HNO precursors, addition of NH2OH led to generation of
mixed-label 14,15N2O. Thus attempts to identify the role of
denitrification intermediates suggest that the hyponitrite anion
2-
N2O2 may be involved in this co-denitrification, resulting in
the formation of N2O from ammonia.
HNO + RSH f RS-NHOH
(12)
(13)
RS-NHOH + RSH f NH2OH + RS-SR
3
4
+
-
Na N O + H f NO + HNO
(14)
(15)
(16)
2
2
3
2
1
5
35
2
HNO f H N O
2 2 2
H N O f N O + H O
2
2
2
2
2
In the quantification of ammonia production, Figure 7, we
found that almost all the consumed nitrite is initially transformed
into ammonia, but the efficiency significantly decreases as the
electrolysis progresses. To test if solution-based ammonia is
Catalytic Reduction of N2O. As previously demonstrated
1
8
I
for Mb/DDAB, the Fe state of CYP119 is capable of
catalytically reducing N2O to N2, Figure 12, formally a two-
electron reduction. Evidence for such reactivity is that the
addition of N2O causes a significant increase in current at the
15
incorporated into N2O, N-labeled ammonia was added to
(
32) Scheele, J. S.; Bruner, E.; Kharitonov, V. G.; Martasek, P.; Roman, L. J.;
II
I
Fe /Fe couple, along with a loss of reversibility, and that the
irreversible current increases with increased N2O concentrations.
Masters, B. S. S.; Sharma, V. S.; Magde, D. J. Biol. Chem. 1999, 274,
1
3105.
(
33) Shiro, Y.; Fujii, M.; Iizuka, T.; Adachi, S.; Tsukamoto, K.; Nakahara, K.;
Shoun, H. J. Biol. Chem. 1995, 270, 1617.
III
II
The Fe /Fe couple is not affected by the addition of substrate
(
34) (a) Tanimoto, T.; Hatano, K.; Kim, D.; Uchiyama, H.; Shoun, H. FEMS
Microbiol. Lett. 1992, 93, 177. (b) Kumon, Y.; Sasaki, Y.; Kato, I.; Takaya,
N.; Shoun, H.; Beppu, T. J. Bacteriol. 2002, 184, 2963.
(36) (a) DeMaster, E. G.; Redfern, B.; Nagasawa, H. T. Biochem. Pharmacol.
1998, 55, 2007. (b) Demaster, E. G.; Redfern, B.; Quast, B. J.; Dahlseid,
T.; Nagasawa, H. T. Alcohol 1997, 14, 181.
(35) Laughlin, R. J.; Stevens, R. J. Soil Sci. Soc. Am. 2002, 66, 1540.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 15, 2004 4941

A R T I C L E S
Immoos et al.
N2O, indicating that N2O does not bind to either oxidation state
in CYP119/DDAB; as with nitrite dehydration, this reactivity
is of the EC mechanism, electron transfer to produce the Fe
fication, resulting in the formation of mixed-label N2O. Con-
versely, CYP119 is shown to be more efficient at the two-
electron reduction of N2O to N2.
I
state, followed by reaction with nitrous oxide, generating
dinitrogen, eq 17. Here the catalytic efficiency of CYP119, at
Our results suggest that thiolate ligation does not enhance or
significantly alter the catalytic reactivity needed for P450nor
activity. But the dramatic difference in product distribution
suggests a major role of the protein itself: the rigidity and
stability of the thermophilic CYP119 may be the controlling
factor inhibiting side reactions, as demonstrated by the high
yields of ammonia and low gaseous product formation.
107 turnovers/s under saturated solution, far exceeds that of Mb
at 29 turnovers/s.
I
+
III
-
Fe + N O + H f Fe + N + HO
(17)
2
2
Conclusions
Acknowledgment. We are deeply indebted to Thomas Poulos
at University of California, Irvine, and Paul Ortiz de Montellano
at University of California, San Francisco, for the several
samples of CYP119 used in this work. This research was
supported by the National Science Foundation (P.J.F.; Grant
CHE-0100774). J.C. acknowledges funding by the Committee
on Research, University of California, Irvine. C.E.I. and E.B.
acknowledge graduate fellowships from the UC Toxic Sub-
stances Research and Teaching Program.
Cyclic voltammograms of electrocatalytic nitrite reduction
by CYP119/DDAB are very similar to those previously reported
for the analogous catalysis by Mb/DDAB: two catalytic waves
at similar potentials are observed, the first corresponding to the
reduction of nitric oxide, the second to the production of
ammonia. The catalytic waves are highly pH-sensitive, and
variation of nitrite concentrations demonstrates saturation kinet-
ics. But analysis of the products shows that CYP119 is a much
more selective catalyst, giving almost exclusively ammonia
during the initial half-hour of reductive electrolysis. More careful
investigations of specific steps in the catalytic cycle show
comparable rates of nitrite dehydration, and almost identical
Supporting Information Available: Electronic absorbance
spectra of CYP119/DDAB, plots of anodic current versus scan
rate, and concentration dependence of nitrite reduction by
CYP119/DDAB, a Lineweaver-Burke plot for electrocatalytic
reduction of nitrite by CYP119/DDAB, a schematic describing
the GC/MS setup, examples of mass spectral analysis and time-
II
potentials and lifetimes for ferrous nitroxyl intermediate (Fe -
-
NO ) in CYP119 and Mb. The catalytic efficiency of NO
reduction is much reduced for CYP119 as compared to Mb;
modeling suggests that the lower efficiency results from both a
lower affinity of the protein for NO, as well as a decreased rate
of N-N coupling. Isotopic studies show that the heme-protein
modified electrodes perform biomimetic co-denitrification, and
here again Mb/DDAB is much more active. Solution-based tests
suggest that the hyponitrite anion N2O22-, rather than its
immediate precursor, HNO, may be responsible for co-denitri-
15
based GC-MS plots of N-labeled gases, examples of chro-
matograms of headgas over HNO precursor solutions, and digital
II
simulation of reversible reduction of Fe -NO species in
CYP119. This material is available free of charge via the Internet
at http://pubs.acs.org.
JA038925C
4942 J. AM. CHEM. SOC.
9
VOL. 126, NO. 15, 2004