Laser photoinitiated nitrosylation of 3-electron reduced Nm europaea hydroxylamine oxidoreductase: Kinetic and thermodynamic properties of the nitrosylated enzyme

Article abstract of DOI:10.1021/ic048822a

Hydroxylamine-cytochrome c554 oxidoreductase (HAO) catalyzes the 4-e ^- oxidation of NH₂OH to NO₂^- by cytochrome c554. The electrons are transferred from NH₂OH to a 5-coordinate heme known as P₄₆₀, the active site of HAO. From P ₄₆₀, c-type hemes transport the electrons through the enzyme to a remote solvent-exposed c-heme, where cyt c554 reduction occurs, When 3-60 μM NO^? are photogenerated by laser flash photolysis of N,N′-bis-(carboxymethyl)-N,N′-dinitroso-1,4-phenylenediamine, in a solution containing ～1 μM HAO prereduced by 3 e^-/subunit, the HAO c-heme pool is subsequently oxidized by up to 1 e^-/HAO subunit. The reaction rate for HAO oxidation shows first-order dependence on [HAO], and zero-order dependence on [NO^?] (k_obs = 1250 ± 150 s^-1), However, the total HAO oxidized shows hyperbolic dependence on [NO^?]. We suggest that NO^? first binds reversibly to P₄₆₀ giving a {Fe(NO)}⁶ moiety. Intramolecular electron transfer (IET) from the c-heme pool then reduces P ₄₆₀ to {Fe(NO)}.⁷ The overall binding constant (K) for formation of {Fe(NO)}⁷ from free NO^? and 3-e ^- reduced HAO was measured at (7.7 ± 0.6) × 10 ⁴ M^-1. This value is larger than that for typical ferriheme proteins (～10⁴ M^-1), but much smaller than that for the corresponding ferroheme proteins (～10¹¹ M ^-1). The final product generated by nitrosylating 3-e^- reduced HAO is believed to be the same species obtained by adding NH ₂OH to the fully oxidized enzyme. The experiments described herein suggest that when NH₂OH and HAO first react, only two of the NH ₂OH electrons end up in the c-heme pool. The other two remain at P₄₆₀ as part of an {Fe(NO)}⁷ moiety. These results are discussed in relation to earlier studies that investigated the effect of putting fully oxidized and fully reduced HAO under 1 atm of NO^?.

Full text of DOI:10.1021/ic048822a

Inorg. Chem. 2005, 44, 225−231
Laser Photoinitiated Nitrosylation of 3-Electron Reduced Nm europaea
Hydroxylamine Oxidoreductase: Kinetic and Thermodynamic Properties
of the Nitrosylated Enzyme
Maria Zulema Cabail, Joshua Kostera, and A. Andrew Pacheco*
Department of Chemistry, UniVersity of WisconsinsMilwaukee, Milwaukee, Wisconsin 53211
Received August 25, 2004
Hydroxylamine-cytochrome c554 oxidoreductase (HAO) catalyzes the 4-e- oxidation of NH2OH to NO2- by cytochrome
c554. The electrons are transferred from NH2OH to a 5-coordinate heme known as P460, the active site of HAO.
From P460, c-type hemes transport the electrons through the enzyme to a remote solvent-exposed c-heme, where
cyt c554 reduction occurs. When 3
(carboxymethyl)-N,N -dinitroso-1,4-phenylenediamine, in a solution containing
e-/subunit, the HAO c-heme pool is subsequently oxidized by up to 1 e-/HAO subunit. The reaction rate for HAO
−60
µ
M NO• are photogenerated by laser flash photolysis of N,N
M HAO prereduced by 3
′-bis-
′
∼
1 µ
oxidation shows first-order dependence on [HAO], and zero-order dependence on [NO•] (kobs
) 1250 ±
150 s-1).
However, the total HAO oxidized shows hyperbolic dependence on [NO•]. We suggest that NO• first binds reversibly
6
to P460 giving a
{
Fe(NO)
}
moiety. Intramolecular electron transfer (IET) from the c-heme pool then reduces P460
7
7
to
{Fe(NO)
}
.
The overall binding constant (K) for formation of
0.6)
{Fe(NO)
}
from free NO• and 3-e- reduced HAO
was measured at (7.7
±
×
104 M-1. This value is larger than that for typical ferriheme proteins (
∼
104 M-1),
but much smaller than that for the corresponding ferroheme proteins (
∼
1011 M-1). The final product generated by
nitrosylating 3-e- reduced HAO is believed to be the same species obtained by adding NH2OH to the fully oxidized
enzyme. The experiments described herein suggest that when NH2OH and HAO first react, only two of the NH2OH
7
electrons end up in the c-heme pool. The other two remain at P460 as part of an
{
Fe(NO)
}
moiety. These results
are discussed in relation to earlier studies that investigated the effect of putting fully oxidized and fully reduced
HAO under 1 atm of NO•.
Introduction
oxidation of NH2OH to NO2- by cytochrome c554 (cyt c554),
which is a critical step in Nm europaea’s respiratory
process.2-5 HAO is chemically intriguing because of its
complexity: it is a homotrimer with a molecular weight of
204 kDa that contains a total of 24 hemes (Figure 1).10 Our
long-term goal is to shed light on how the complex design
features of HAO relate to its function in the Nm europaea
energy transduction system. One approach to this problem
has been to develop a series of novel pulsed-laser techniques
for photogenerating putative intermediates of HAO’s catalytic
Nm europaea is a chemolithotrophic bacterium that obtains
all its energy from the aerobic oxidation of NH4+ to NO2 .
This net oxidation of NH4 to NO2 forms a part of the
biosphere’s nitrogen cycle, and as such has enormous
economic and ecological importance.2-9 Hydroxylamine
cytochrome c554 oxidoreductase (HAO) catalyzes the 4-e-
- 1,2
+
-
* To whom correspondence should be addressed. E-mail: apacheco@
uwm.edu.
(1) Ferguson, S. J. Curr. Opin. Chem. Biol. 1998, 2, 182-193.
(2) Hooper, A. B. In Autotrophic Bacteria; Schlegel, H. G., Bowien, B.,
Eds.; Science Tech Publishers: Madison, WI, 1989; pp 239-265.
(3) Kroneck, P. M. H.; Beuerle, J.; Schumacher, W. In Metal Ions in
Biological Systems; Sigel, H., Sigel, A., Eds.; Marcel Dekker Inc.:
New York, 1992; pp 455-505.
(4) Wood, P. M. In The Nitrogen and Sulfur Cycles; Cole, J. A., Ferguson,
S. J., Eds.; Cambridge University Press: New York, 1988; pp 219-
243.
(6) Schlegel, H. G. In Biology of Inorganic Nitrogen and Sulfur;
Bothe, H., Trebst, A., Eds.; Springer-Verlag: New York, 1981; pp
3-12.
(7) Biological Nitrogen Fixation; National Academy Press: Washington,
DC, 1994; pp 6-32.
(8) Burns, R. C.; Hardy, R. W. F. Nitrogen Fixation in Bacteria and
Higher Plants; Springer-Verlag: New York, 1975.
(9) Prince, R. C.; George, G. N. Nat. Struct. Biol. 1997, 4, 247-250.
(10) Igarashi, N.; Moriyama, H.; Fujiwara, T.; Fukumori, Y.; Tanaka, N.
Nat. Struct. Biol. 1997, 4, 276-284.
(5) Bock, E.; Koops, H.-P.; Harms, H.; Ahlers, B. In Variations in
Autotrophic Life; Shively, J. M., Barton, L. L., Eds.; Academic
Press: San Diego, CA, 1991; pp 171-200.
10.1021/ic048822a CCC: $30.25
Published on Web 12/24/2004
© 2005 American Chemical Society
Inorganic Chemistry, Vol. 44, No. 2, 2005 225

Cabail et al.
Figure 2. Putative N-containing intermediates generated at the P460 active
site during catalytic NH2OH oxidation to NO2-. The electrons leaving the
P460 fragment will be passed on to the HAO c-hemes (see Figure 1). N2O
and free NO• are possible side products, as described in the text.18
Intermediates containing Fe-NO fragments are each represented in terms
of two resonance structures, and of the corresponding Enemark-Feltham
description {Fe(NO)}n. In this notation, the superscript n is the sum of the
d electrons that would be counted on Fe if the ligand were actually NO•,
and the π* electron from the NO•.26 Note that “P460” is used in place of
“Fe” in the notation, to emphasize that it is the P460 heme that is nitrosylated.
periplasm of Nm europaea,18 and is now generally recognized
to be the acceptor of the electrons from HAO in vivo.
An important goal of our research is to characterize the
nitrogen-containing intermediates generated at the P460 active
Figure 1. (a) Arrangement of the eight hemes within one subunit of HAO.
(b) Arrangement of all the hemes within the HAO trimer; primes are used
to distinguish subunits. The hemes are labeled according to the format used
by Igarashi et al.,10 except for P460 which was labeled heme 4 in the original
article.
-
site during catalytic NH2OH oxidation to NO2 . On the basis
of extensive literature precedent,19-25 Figure 2 outlines some
of the N-containing fragments that are predicted. The figure
also introduces a slight modification of the convenient
Enemark-Feltham notation, {P460(NO)}n, to describe inter-
mediates containing Fe-NO fragments.26 The likelihood that
{P460(NO)}n species are intermediates in the catalytic cycle
led us to adopt and extend a methodology for generating
such species from free NO• using a photochemical trigger.27-30
This method exploits the fact that the water-soluble com-
pathway. Herein we describe the results obtained using one
such technique.
NH2OH oxidation by HAO takes place on three novel
hemes known as P460’s, that have vacant coordination sites
at which NH2OH can bind and be oxidized (Figure 1).10-14
The remaining 21 HAO hemes are of c-type and 6-coordi-
nate, with two axial His ligands each.10,15 These hemes are
low-spin ferric in the resting enzyme, and closely spaced.
Hence, they can act as electron transfer agents, accepting
the electrons that P460 abstracts from NH2OH.10,15 Moreover,
heme 8 in each subunit lies sufficiently near heme 2 of an
adjacent subunit to possibly allow e- transfer between
subunits. Indeed, the 18-heme circle seen in the HAO trimer
(Figure 1b) appears to be superbly designed to allow e-
entering at a single P460 to be rapidly distributed throughout
the trimer. Heme 1 (Figure 1) is the only heme apart from
(16) Iverson, T. M.; Arciero, D. M.; Hsu, B. T.; Logan, M. S. P.; Hooper,
A. B.; Rees, D. C. Nat. Struct. Biol. 1998, 5, 1005-1012.
(17) Andersson, K., K.; Lipscomb, J. D.; Valentine, M.; Munck, E.; Hooper,
A. B. J. Biol. Chem. 1986, 261, 1126-1138.
(18) Whittaker, M.; Bergman, D.; Arciero, D.; Hooper, A. B. Biochim.
Biophys. Acta 2000, 1459, 346-355.
(19) Hoshino, M.; Maeda, M.; Konishi, R.; Seki, H.; Ford, P. C. J. Am.
Chem. Soc. 1996, 118, 5702-5707.
(20) Ford, P. C.; Lorkovic, I. M. Chem. ReV. 2002, 102, 993-1017.
(21) Wolak, M.; van Eldik, R. Coord. Chem. ReV. 2002, 230, 263-282.
(22) Fernandez, B. O.; Ford, P. C. J. Am. Chem. Soc. 2003, 125, 10510-
10511.
P460 that is solvent exposed, and it appears to be the in vivo
exit point for the electrons extracted from NH2OH oxidation.
The electrons go from heme 1 to cyt c554, a soluble 26.1
kDa tetra-heme protein16,17 that is quite abundant in the
(23) Choi, I.-K.; Liu, Y.; Wei, Z.; Ryan, M. D. Inorg. Chem. 1997, 36,
3113-3118.
(24) Feng, D.; Ryan, M. D. Inorg. Chem. 1987, 26, 2480-2483.
(25) Einsle, O.; Messerschmidt, A.; Huber, R.; Kroneck, P. M. H.; Neese,
F. J. Am. Chem. Soc. 2002, 124, 11737-11745.
(11) Hooper, A. B.; Terry, K. R. Biochemistry 1977, 16 (3), 455-459.
(12) Hendrich, M. P.; Logan, M.; Andersson, K. K.; Arciero, D. M.;
Lipscomb, J. D.; Hooper, A. B. J. Am. Chem. Soc. 1994, 116, 11961-
11968.
(13) Logan, M. S. P.; Balny, C.; Hooper, A. B. Biochemistry 1995, 34,
9028-9037.
(14) Logan, M. S. P.; Hooper, A. B. Biochemistry 1995, 34, 9257-9264.
(15) Arciero, D. M.; Hooper, A. B. J. Biol. Chem. 1993, 268, 14645-
14654.
(26) Enemark, J. H.; Feltham, R. D. Coord. Chem. ReV. 1974, 13, 339-406.
(27) Namiki, S.; Arai, T.; Fujimori, K. J. Am. Chem. Soc. 1997, 119, 3840-
3841.
(28) Cabail, M. Z.; Lace, P. J.; Uselding, J.; Pacheco, A. A. J. Photochem.
Photobiol., A 2002, 152, 109-121.
(29) Bodemer, G.; Ellis, L. M.; Lace, P. J.; Mooren, P. E.; Patel, N. K.;
Ver Haag, M.; Pacheco, A. A. J. Photochem. Photobiol., A 2004, 53-
60.
(30) Moua, V.; Bae, E.; Pacheco, A. A. Manuscript in preparation.
226 Inorganic Chemistry, Vol. 44, No. 2, 2005

Nitrosylation of Reduced Hydroxylamine Oxidoreductase
Scheme 1 a
to generate the putative 4-e- reduced intermediate that
would be obtained by adding 1 equiv per HAO monomer of
NH2OH to the fully oxidized enzyme (Figure 2).
Experimental Section
Materials. HAO was purified as described in ref 31, the
photoactive NO releasing species 4 was synthesized according to
the procedure of ref 28, and TiIII(citrate) was prepared as described
in ref 32. [Ru(NH3)6]Cl2 was obtained from Aldrich. Myoglobin
(crystallized and lyophilized horse skeletal muscle) and ferricyto-
chrome c (cyt c, lyophilized horse heart prepared using TCA) were
from Sigma. All other chemicals were from Fisher. Unless otherwise
specified, all solutions were made up in 100 mM phosphate buffer,
pH ) 7.4.
Stock Solution Preparations. Stock solutions of 4, TiIII(citrate)
(TiIII), and [Ru(NH3)6]Cl2 (RuII) were prepared and stored in a
nitrogen-filled glovebox. The RuII stocks were made fresh daily,
and stored in a 4 °C fridge until needed. The concentrations of
RuII and TiIII solutions were obtained by monitoring their ability
to reduce cyt c, using UV-vis spectroscopy.33 The myoglobin
obtained from Sigma was in the fully oxidized “met” state (met-
Mb) and was reduced in the glovebox with TiIII prior to use. The
reduced myoglobin (Mb) stock solutions used for laser-induced
nitrosylation experiments contained 39.2 µM of the protein, and
also 146 µM of RuII. The concentrations of Mb, met-Mb, and cyt
c solutions were determined directly by UV-vis spectroscopy.33-35
The concentration of HAO solutions is reported throughout in
terms of subunit concentration, determined from the extinction
coefficient at the c-heme soret peak of the oxidized enzyme (700
mM-1 cm-1 at 408 nm).13 Aliquots (100 µL) of HAO (typically
∼20-40 µM in subunit concentration) were stored in 1 mL
cryogenic tubes, in a liquid nitrogen storage vessel. When needed,
an uncapped cryogenic tube containing one aliquot of still-frozen
HAO was placed in the mini-antechamber of the glovebox,
submitted to 4 rapid pump-purge cycles, then brought inside the
box still frozen. The tube was recapped and placed in a 4 °C fridge
inside the box for ∼60 min. Unless otherwise stated, the following
procedure was then used. The HAO aliquot was diluted to 1.5 mL
with phosphate buffer, and enough RuII stock to make the final
solution 140 µM in RuII. This solution was transferred to a cuvette
and titrated with TiIII(citrate). The titration was monitored by
UV-vis spectroscopy and stopped upon obtaining the absorbance
ratios A410/A420 ) 0.882, and A552/A420 ) 0.153. Previously,
spectropotentiometric titrations revealed that these ratios correspond
to HAO in which 3/8 c-hemes in each monomer have been
reduced.36
a “Red” refers to an arbitrary electron donor.
pound N,N′-bis(carboxymethyl)-N,N′-dinitroso-p-phenylene-
diamine (species 4, Scheme 1) will efficiently generate free
NO• upon irradiation with 308 nm light.27-30 If a XeCl
excimer laser is used to initiate the photoreaction, the NO•
becomes available on the microsecond to millisecond time
scales, and can then be used to nitrosylate the P460 active
site (dashed arrows, Figure 2).
Scheme 1 describes the NO• generation process. The laser
pulse promotes the fragmentation of 4 into 5 and NO•
(Scheme 1, eq 1). Species 5 is very reactive and will rapidly
recombine with NO•, or fragment further to give the relatively
stable species 6 and a second equivalent of NO• (Scheme 1,
eq 2).27,28 Species 5 is also a powerful oxidant, which will
be irreversibly reduced to 7 in the presence of a suitable
electron donor (Scheme 1, eq 3).29,30 Thus, species 4 must
be used with one of two experimental protocols, depending
on the situation. When nitrosylation of a nonreducing species
is desired, solutions containing only 4 and the species to be
nitrosylated can be irradiated. However, if the species to be
nitrosylated is also a reducing agent, then a sacrificial electron
donor must be added to the mixture in order to avoid
oxidizing the species of interest with 5.29,30 We previously
reported on the reaction of fully oxidized HAO with pho-
togenerated NO•, which resulted in formation of {Fe(NO)}6
(Figure 2) and allowed us to simulate the final reductive
nitrosylation steps of the proposed mechanism.31 These
experiments did not require addition of a co-reducing agent.31
Herein we present the results of exposing HAO reduced by
3-e-/monomer to NO•, photogenerated in the presence of
RuII as co-reductant. The goal of these experiments was
Laser-Initiated Nitrosylation Experiments. In preparation for
most experiments, the following procedure was used. HAO or Mb
stock solution (75 µL) was mixed with varying volumes of stock
4 and phosphate buffer, to give solutions with fixed total volumes
of 150 µL. These solutions were transferred to 3 mm × 3 mm
quartz cuvettes that were then sealed with greased ground glass
stoppers. The amount of NO• generated in any given experiment
was controlled either by varying [4] in the cuvette, or by using
(31) Cabail, M. Z.; Pacheco, A. A. Inorg. Chem. 2003, 42, 270-272.
(32) Codd, R.; Astashkin, A. V.; Pacheco, A.; Raitsimring, A. M.; Enemark,
J. H. J. Biol. Inorg. Chem. 2002, 7, 338-350.
(33) Margoliash, E.; Frohwirt, N. Biochem. J. 1959, 71, 570-573.
(34) James, B. R. Physical Chemistry, Part C; Dolphin, D., Ed.; The
Porphyrins, Vol. V; Academic Press, Inc.: New York, 1978; pp 207-
302.
(35) Hoshino, M.; Ozawa, K.; Seki, H.; Ford, P. C. J. Am. Chem. Soc.
1993, 115, 9568-9575.
(36) Cabail, M. Z.; Pacheco, A. A. Unpublished.
Inorganic Chemistry, Vol. 44, No. 2, 2005 227

Cabail et al.
width was 0.6 mm, and the exit slit width was 0.12 mm. The 418.7
nm band of holmium oxide (IBM Standards 9420) was used as a
reference for calibrating the spectrophotometer wavelength in each
mode. The data were analyzed using the commercially available
software packages Specfit/32, Version 3.0 (Spectrum Software
Associates), Microcal Origin, Version 6.0 (Microcal Software, Inc.),
and Mathcad11 (Mathsoft Engineering and Education, Inc.).
Results
An investigation that will be presented elsewhere revealed
that [Ru(NH3)6]2+ (RuII) can reduce the radical species 5
(Scheme 1, eq 3) at diffusion-limited rates (k ) 2 × 109
M-1 s-1).30 This makes RuII an ideal candidate for the role
of sacrificial electron donor in laser experiments involving
photolysis of 4, because relatively low [RuII] will be needed
to ensure that NO• and the stable species 7 (Scheme 1, eq 3)
are obtained quantitatively from 5. In all the experiments
described below, [RuII] was high enough to remove all traces
of species 5 in less than ∼20 µs. Figure 3 shows the
difference spectrum obtained 20 ms after irradiating a
thoroughly anaerobic solution containing 4, RuII, and 3-e-
reduced HAO. The Figure 3 difference spectrum is charac-
teristic of c-heme oxidation.37 The prominent negative signal
at 424 nm would be seen for any of the HAO c-hemes. On
the other hand, the signal at 560 nm is unique to heme 3,
which is one of two c-hemes that have reduction potentials
of ∼0 mV. This signal has a prominent shoulder at ∼552
nm, which is where the remaining c-hemes show difference
absorbance minima during oxidation.37 The difference spec-
trum shows no evidence of spectral changes involving
nitrosylation of the P460 center, which results in visible
spectroscopic characteristics very different from those due
to c-hemes.38 However, these spectral changes are often
roughly 15 times weaker than corresponding c-heme sig-
nals,13,38 so {P460(NO)}n species may be present and not
readily detectable (see below).
Figure 4 shows a typical ∆A424 versus time trace obtained
after irradiating mixtures of 4, 3-e- reduced HAO, and RuII
(further examples are given as Supporting Information). The
negative signal at 424 nm grows in exponentially from zero,
with a rate constant kobs ) 1250 ( 150 s-1 that is independent
of [NO•] generated by the laser pulse (Figure 5), and an
amplitude that has hyperbolic dependence on [NO•] (Figure
6). For experiments in which the laser was used at full
intensity and [4] was <100 µM, an additional very fast drop
in A424 took place in the dead-time of the instrumentation.
This c-heme oxidation phase is believed to be initiated by
photoexcitation of HAO itself, and is discussed more fully
in Supporting Information.
Figure 3. Spectral changes observed 20 ms after photoexcitation of a
solution initially containing 68 µM 4, 70 µM RuII, and 650 nM HAO, in
which 3/8 hemes of each monomer had been prereduced with TiIII citrate.
For this experiment, the solution was irradiated in a 2 mm × 10 mm
fluorescence cuvette, with the laser pulse traversing the 2 mm width of the
cuvette, as described in ref 28.
filters to attenuate the laser pulse and keeping [4] constant at 900
µM.28 In all cases, the [NO•] generated exceeded the amount of
HAO nitrosylated by a sufficient amount that, for kinetic analysis,
pseudo-first-order conditions prevailed. The amount of NO• pro-
duced under a given set of conditions was quantified in separate
experiments using excess Mb as a spectroscopically detectable NO•
scavenger.29,35 Full details of these experiments will be presented
elsewhere.30 For HAO nitrosylation experiments in which only small
amounts of NO• were generated using attenuated laser pulses, each
run was carried out in triplicate, and the results were averaged. In
addition, a run with Mb was carried out before and after the three
runs with HAO. The [NO•] for the HAO experiments was then
taken to be the average of the [NO•] generated in the Mb runs. For
HAO nitrosylation experiments in which relatively large amounts
of NO• were produced by using the laser at full intensity, the [NO•]
generated in each experiment was taken as equal to 20% of the
concentration of 4 used in that experiment.30 In all cases, the average
variation in [NO•] generated from one run to the next was ∼10%,
presumably attributable to variations in the laser pulse intensity.
For the experiment that resulted in Figure 3, the sample prep-
aration method was somewhat different in detail than described
above, and laser photolysis was carried out in a 10 mm × 2 mm
quartz cuvette, as described in ref 28.
Photochemical fragmentation of species 4 was initiated with a
10 ns, 308 nm pulse from a XeCl excimer laser (TUI, Existar 200).
An OLIS RSM-1000 spectrophotometer was used to monitor the
absorbance changes induced by the laser pulse. The configuration
of the laser and spectrophotometric equipment has been described
in general terms elsewhere.28 Preliminary data were collected with
the OLIS RSM-1000 in rapid-scanning mode (scanning slit width
) 0.2 mm), which allows complete spectra to be obtained in 1 ms.
The preliminary data showed that the reactions of interest had half-
lives of less than 0.6 ms, so subsequent experiments were performed
with the RSM-1000 in fixed-wavelength mode (fixed middle slit
width ) 0.6 mm). In this mode, changes in absorbance at a single
wavelength could be obtained for time intervals as short as 4 µs.
The preliminary multiwavelength data were used to choose an
appropriate wavelength to monitor. For both the fixed-wavelength
and rapid-scanning experiments, the monochromator entrance slit
In 3-e- reduced HAO, the extra electrons reside on two
∼0 mV hemes (heme 3 and heme 8, Figure 1), and a high
potential heme 2 (E° ) 288 mV).37,39,40 The molar absorp-
(37) Collins, M. J.; Arciero, D. M.; Hooper, A. B. J. Biol. Chem. 1993,
268, 14655-14662.
(38) Hendrich, M. P.; Upadhyay, A. K.; Riga, J.; Arciero, D. M.; Hooper,
A. B. Biochemistry 2002, 41, 4603-4611.
(39) Hendrich, M. P.; Petasis, D.; Arciero, D. M.; Hooper, A. B. J. Am.
Chem. Soc. 2001, 123, 2997-3005.
(40) Kurnikov, I.; Ratner, M. A.; Pacheco, A. A. Biochemistry , accepted.
228 Inorganic Chemistry, Vol. 44, No. 2, 2005

Nitrosylation of Reduced Hydroxylamine Oxidoreductase
Figure 4. ∆A vs time trace obtained at 424 nm, after photoexcitation of
a solution initially containing 905 µM 4, 73 µM RuII, and 1.04 µM HAO,
in which 3/8 hemes of each monomer had been prereduced with TiIII citrate.
The solution was irradiated in a 3 mm × 3 mm cuvette. For this experiment,
a filter with 44% transmittance was used to attenuate the laser pulse, and
[NO•] generated ∼10 µM. The data were fitted with an exponential function,
y ) A0 + b[1 - exp(-kobst)]. Other data sets are provided as Supporting
Information.
Figure 6. Dependence of the amplitude of the exponential fitting function
on [NO•] generated, for a series of experiments analogous to that depicted
in Figure 4. The data are fitted to a rectangular hyperbola, b ) bmax[NO•]/
(K-1 + [NO•]), where the significance of K is shown in Scheme 2. The
calculated parameters are the following: bmax ) -0.035 ( 0.001; K-1
13 ( 1 µM. The raw data are given as Supporting Information.
)
species 7 and [Ru(NH3)6]3+ (RuIII). Therefore, an obvious
explanation for the observed oxidation of HAO is that RuIII
serves as the oxidant. However, the data are not consistent
with this explanation. The standard reduction potential for
RuIII is 100 mV,41 and calculations show that even for the
lowest amount of RuIII generated (3 µM) the fraction of 0
mV hemes oxidized should have been about 70%. At this
[RuIII], the fractional oxidation obtained from Figure 6 is
roughly 20%. At the highest concentrations of RuIII generated
(60 µM, Figure 6), >99% of the 0 mV hemes should have
been oxidized, whereas experimentally the value is less than
45%. This suggests that oxidation of HAO by RuIII, while
thermodynamically favorable, is slow and not the reaction
observed herein. The independence of the kobs value on [RuIII]
()[NO•], Figure 5) is also inconsistent with [RuIII] being
the oxidizing agent in the observed reactions. Oxidation by
RuIII would almost certainly take place at the solvent-exposed
c-heme 1 (Figure 1) via an outer-sphere process, and should
have first-order dependence on [RuIII]. Saturation is not
expected in an interaction between HAO and a nonphysi-
ological, small oxidant.42,43
Figure 5. Dependence of the kobs values on [NO•] generated, for a series
of experiments analogous to that depicted in Figure 4. The raw data are
given as Supporting Information. The horizontal line represents the average
kobs (1250 s-1), while the vertical error bars give the standard deviation
((150 s-1).
Scheme 2 presents an alternative hypothesis to explain the
observed HAO oxidation, which is that oxidation of the
c-heme pool is coupled to nitrosylation of the P460 active
site. According to this hypothesis, NO• first binds to the
P460 ferric heme to give {P460(NO)}6 (Scheme 2, eq 1).
This is followed by intramolecular electron transfer (IET)
from the 0 mV hemes to P460, which reduces the active site
to {P460(NO)}7 (Scheme 2, eq 2). Formation of a stable
{P460(NO)}7 species following an initial nitrosylation step
tivity differences at 424 nm (∆ꢀ424) associated with oxidation
of these hemes are 101 mM-1 cm-1 for two ∼0 mV hemes
3 and 8, and 114 mM-1cm-1 for heme 2.37 Application of
Beer’s law to the data in Figure 6, using the published ∆ꢀ424
values and the cell path length of 3 mm, shows that roughly
one out of three electrons is removed from the c-heme pool
at sufficiently high [NO•]. Because heme 3 and heme 8 have
approximately the same potential, in the resulting 2-e-
c-heme system, one of the electron is expected to equilibrate
between these two hemes. The fact that the signal with
maximal intensity at 560 nm also has a significant shoulder
at 552 nm (Figure 3) supports this assumption.
(41) Bard, A. J.; Faulkner, L. R. Electrochemical Methods, Fundamentals
and Applications, 2nd ed.; John Wiley and Sons: New York, 2001; p
809. This value was verified in our buffer solution.
(42) Liang, Z.-X.; Nocek, J. M.; Huang, K.; Hayes, R. T.; Kurnikov, I. V.;
Beratan, D. M.; Hoffman, B. M. J. Am. Chem. Soc. 2002, 124, 6849-
6859.
The same reaction that traps the photogenerated NO•
(Scheme 1, eq 3) also generates equivalent amounts of
(43) Liang, Z. X.; Kurnikov, I. V.; Nocek, J. M.; Mauk, A. G.; Beratan,
D. N.; Hoffman, B. M. J. Am. Chem. Soc. 2004, 126, 2785-2798.
Inorganic Chemistry, Vol. 44, No. 2, 2005 229

Cabail et al.
a
Scheme 2
Hendrich et al. recently reported that when 0.48 mM of
fully reduced HAO was put under 1 atm NO•, most of the
hemes were reoxidized.38 The lowest potential HAO hemes
(E° < -150 mV)37,39 might be capable of reducing NO• back
to NH2OH; however, much of the heme oxidation in these
experiments was probably coupled to the reduction of 2 equiv
of NO• to N2O, via the shunt pathway shown in Figure 2.
N2O is the most common product of nonenzymatic NH2OH
oxidation by metals,46 and significant amounts of N2O are
released in vitro when HAO reacts with NH2OH.18,47 N2O
is also produced by Nm europaea in vivo, though in this
case HAO appears to be one of several sources of the
gas.18,48,49 In our experiments, the limited oxidation of the
c-heme pool following nitrosylation shows that, with [NO•]
of up to 60 µM and on the time scale investigated herein,
N2O is not released from HAO. Such a release would require
the concomitant removal of another electron from the c-heme
pool.
a The HAO c hemes are labeled with their redox potentials subscripted
in brackets.
is consistent with the fact that at most one electron is lost
from the HAO c-heme pool, irrespective of the amount of
Ru(III) or NO• photogenerated (Figure 6).
Discussion
Figure 2 suggests that NH2OH binds to P460 to give species
2, and is then oxidized and deprotonated to give species 3.
If the reaction begins with HAO fully oxidized, and is carried
out in the absence of an external electron acceptor such as
cyt c554, then one of the electrons initially removed from 2
will end up on the high potential heme 2, and the other will
equilibrate between the two 0 mV hemes (3 and 8).37,39,40
The subsequent deprotonation step leaves the active site in
the {P460(NO)}7 state, and HAO overall in the state suggested
by Scheme 2 to result from nitrosylating the 3-e- reduced
enzyme. The complete dashed pathway in Figure 2 corre-
sponds to Scheme 2, with a protonation step added. Our
experiments do not reveal whether the product of Scheme
2, eq 2, is subsequently protonated. However, they clearly
demonstrate that in 4-e- reduced HAO (as would be obtained
by adding 1 equiv of NH2OH to a fully oxidized enzyme
subunit) only two electrons will be transferred to the c-heme
pool, while the other two will remain on the P460-NO
moiety. If Scheme 2, eq 2, did not lie far to the right, then
in the nitrosylation experiments the c-heme pool should be
oxidized by less than 1 equiv, even at saturating [NO•].
The fact that the c-heme pool was not oxidized by more
than 1 e-/subunit in the nitrosylation experiments provides
excellent evidence that, upon addition of 1 equiv NH2OH to
the fully oxidized enzyme, the first two electrons would be
readily transferred to the c-heme pool, as proposed in Figure
2. Presumably, these intramolecular electron transfer steps
would be coupled to concomitant H+ transfers. The HAO
crystal structure shows that Asp 267 and His 268 are oriented
toward the P460 active site, and it has been suggested that
these residues may contribute to coupled e--H+ transfer.10
Recent papers report the isolation of stable Fe and Ru
porphyrin species best formulated as the protonated forms
of {M(NO)}8 ({M(HNO)}8).44,45 For HAO, such a species
would sit between 2 and 3 in Figure 2. In our experi-
ments, the generation of a more highly reduced P460 nitrosy-
lation product, such as {P460(HNO)},8 would require the
removal of two electrons from the c-heme pool. Again, the
results show that the c-heme pool is reduced by only one
electron following nitrosylation, so {P460(HNO)}8 is not being
generated.
From the hyperbolic fit in Figure 6, one can obtain an
equilibrium binding constant K ) (7.7 ( 0.6) × 104 M-1
for the net nitrosylation reaction of Scheme 2. This value is
∼5 times higher than that for typical ferriheme proteins such
as Mb and Hb (∼1.4 × 104 M-1),19,35 but orders of magnitude
smaller than that for the corresponding ferroheme proteins
(∼1011 M-1).19 Presumably, the stability of {P460(NO)}7
relative to oxidation by the 0 mV c-hemes helps to minimize
NO• release from the enzyme during turnover, which should
be higher from a {P460(NO)}6 moiety (but see below).
However, excessive stabilization of {P460(NO)}7 would trap
this moiety and prevent catalysis.
Making the assumption that K2 > 10 (on the basis of the
fact that Scheme 2, eq 2, lies far to the right) provides an
upper estimate of ∼8 × 103 M-1 for the value K1, which
governs the formation of {P460(NO)}6 from NO• and 1
(Scheme 2, eq 1). This low value suggests that NO• should
bind only weakly to fully oxidized HAO, in which stabiliza-
tion of {P460(NO)}7 by IET from the c-heme pool is not an
option. As mentioned above, NO• normally binds weakly in
{Fe(NO)}6 hemoproteins,19,35 and it binds even more weakly
in nonprotein {Fe(NO)}6 moieties.50 However, in their recent
report Hendrich and co-workers reported obtaining an
extremely tenacious {P460(NO)}6 species after putting fully
oxidized HAO under 1 atm NO•.38 Indeed, after removing
all the free NO• from solution, these workers could only
remove the bound NO• by irradiating the solution with
intense white light. These results suggest that on a time scale
longer than was monitored in our experiments, a protein
conformational change can trap NO• in the active site, even
(45) Lee, J. Y.; Richter-Addo, G. B. J. Biol. Inorg. Chem. 2004, 98,
1247-1250.
(46) Wieghardt, K. AdV. Inorg. Bioinorg. Mech. 1984, 3, 213-274.
(47) Hooper, A. B.; Terry, K. R. Biochim. Biophys. Acta 1979, 571,
12-20.
(48) Beaumont, H. J. E.; Hommes, N. G.; Sayavedra-Soto, L. A.; Arp, D.
J.; Arciero, D. M.; Hooper, A. B.; Westerhoff, H. V.; van Spanning,
R. J. M. J. Bacteriol. 2002, 184, 2557-2560.
(49) Beaumont, H. J. E.; van Schooten, B.; Lens, S. I.; Westerhoff, H. V.;
van Spanning, R. J. M. J. Bacteriol. 2004, 186, 4417-4421.
(50) Ellison, M. K.; Scheidt, W. R. J. Am. Chem. Soc. 1999, 121, 5210-
5219.
(44) Sulc, F.; Immoos, C. E.; Pervitsky, D.; Farmer, P. J. J. Am. Chem.
Soc. 2004, 126, 1096-1101 and references therein.
230 Inorganic Chemistry, Vol. 44, No. 2, 2005

Nitrosylation of Reduced Hydroxylamine Oxidoreductase
as a {P460(NO)}6 moiety. Such behavior has precedent in
the NO•-transporting nitrophorins,51,52 in which {P460(NO)}6
is stabilized by a major protein conformational change that
effectively seals the NO• within the protein. This conforma-
tional change is slow relative to initial NO• binding.51,52
The fact that the apparent rate constant for nitrosylation
(kobs, Figure 5) is independent of [NO•] suggests that even
the rapid binding observed in the current study is preceded
by a comparatively slow conformational change, which
determines the attainable reaction rate. The IET step (Scheme
2, eq 2) could only become rate-limiting at a [NO•] where
all of the P460 was in the {P460(NO)}6 form. On the basis of
the arguments of the previous paragraph, such a [NO•] should
be fairly high. At present, it is unclear what kind of
conformational change would be necessary in order to allow
NO• to bind, since the crystal structure appears to show a
relatively open P460 active site. More studies will be required
to address this question. Future experiments are also planned
in which higher [NO•] and longer time scales will be
monitored, so that the N2O shunt pathway can be character-
ized (Figure 2). The fact that HAO nitrosylation and
subsequent N2O formation appear to proceed on very
different time scales will greatly simplify investigations of
the two processes.
Supporting Information Available: Conditions for all photo-
initiated reactions of NO• with 3-e- reduced HAO (Table S1); all
∆A424 vs time traces obtained after irradiating mixtures of 4, 3-e-
reduced HAO, and RuII (Figure S1a-h); dependence of A0 (∆A424
at t ) 0 from Figure S1a-h) on [NO•] generated (Figure S2);
interpretation of this dependence. This material is available free of
charge via the Internet at http://pubs.acs.org.
(51) Andersen, J. F.; Ding, X. D.; Balfour, C.; Shokhireva, T. K.;
Champagne, D. E.; Walker, F. A.; Montfort, W. R. Biochemistry 2000,
39, 10118-10131.
(52) Ding, X. D.; Weichsel, A.; Andersen, J. F.; Shokhireva, T. K.; Balfour,
C.; Pierik, A. J.; Averill, B. A.; Montfort, W. R.; Walker, F. A. J.
Am. Chem. Soc. 1999, 121, 128-138.
IC048822A
Inorganic Chemistry, Vol. 44, No. 2, 2005 231