Dissecting structural and electronic effects in inducible nitric oxide synthase

Article abstract of DOI:10.1042/BJ20141319

Nitric oxide synthases (NOSs) are haem-thiolate enzymes that catalyse the conversion of L-arginine (L-Arg) into NO and citrulline. Inducible NOS (iNOS) is responsible for delivery of NO in response to stressors during inflammation. The catalytic performance of iNOS is proposed to rely mainly on the haem midpoint potential and the ability of the substrate L-Arg to provide a hydrogen bond for oxygen activation (O-O scission). We present a study of native iNOS compared with iNOS-mesohaem, and investigate the formation of a low-spin ferric haem-aquo or -hydroxo species (P) in iNOS mutant W188H substituted with mesohaem. iNOS-mesohaem and W188H-mesohaem were stable and dimeric, and presented substrate-binding affinities comparable to those of their native counterparts. Single turnover reactions catalysed by iNOSoxy with L-Arg (first reaction step) or N-hydroxy-L-arginine (second reaction step) showed that mesohaem substitution triggered higher rates of Fe^IIO₂ conversion and altered other key kinetic parameters. We elucidated the first crystal structure of a NOS substituted with mesohaem and found essentially identical features compared with the structure of iNOS carrying native haem. This facilitated the dissection of structural and electronic effects. Mesohaem substitution substantially reduced the build-up of species P in W188H iNOS during catalysis, thus increasing its proficiency towards NO synthesis. The marked structural similarities of iNOSoxy containing native haem or mesohaem indicate that the kinetic behaviour observed in mesohaem-substituted iNOS is most heavily influenced by electronic effects rather than structural alterations.

Full text of DOI:10.1042/BJ20141319

Biochem. J. (2015) 467, 153–165 (Printed in Great Britain) doi:10.1042/BJ20141319
153
Dissecting structural and electronic effects in inducible nitric oxide synthase
1
Luciana Hannibal*† , Richard C. Page‡§, Mohammad Mahfuzul Haque*, Karthik Bolisetty*, Zhihao Yu*, Saurav Misra‡ and
Dennis J. Stuehr*1
*Department of Pathobiology, Lerner Research Institute, Cleveland Clinic, Cleveland, OH 44195, U.S.A.
†Lehrstuhl fu¨r Bioanorganische Chemie, Department Chemie und Pharmazie, Universita¨t Erlangen-Nu¨rnberg, Egerlandstraße 1, D-91058, Erlangen, Germany
‡Department of Molecular Cardiology, Lerner Research Institute, Cleveland Clinic, Cleveland, OH 44195, U.S.A.
§Department of Chemistry and Biochemistry, Miami University, 651 E. High Street, Oxford, OH 45056, U.S.A.
FeIIO2 conversion and altered other key kinetic parameters. We
elucidated the first crystal structure of a NOS substituted with
mesohaem and found essentially identical features compared
with the structure of iNOS carrying native haem. This facilitated
the dissection of structural and electronic effects. Mesohaem
substitution substantially reduced the build-up of species P in
W188H iNOS during catalysis, thus increasing its proficiency
towards NO synthesis. The marked structural similarities of
iNOSoxy containing native haem or mesohaem indicate that
the kinetic behaviour observed in mesohaem-substituted iNOS is
most heavily influenced by electronic effects rather than structural
alterations.
Nitric oxide synthases (NOSs) are haem-thiolate enzymes that
catalyse the conversion of L-arginine (L-Arg) into NO and
citrulline. Inducible NOS (iNOS) is responsible for delivery of
NO in response to stressors during inflammation. The catalytic
performance of iNOS is proposed to rely mainly on the haem
midpoint potential and the ability of the substrate L-Arg to
provide a hydrogen bond for oxygen activation (O-O scission).
We present a study of native iNOS compared with iNOS-
mesohaem, and investigate the formation of a low-spin ferric
haem-aquo or -hydroxo species (P) in iNOS mutant W188H
substituted with mesohaem. iNOS-mesohaem and W188H-
mesohaem were stable and dimeric, and presented substrate-
binding affinities comparable to those of their native counterparts.
Single turnover reactions catalysed by iNOSoxy with L-Arg (first
reaction step) or N-hydroxy-L-arginine (second reaction step)
showed that mesohaem substitution triggered higher rates of
Key words: catalysis, haem, mesohaem, nitric oxide synthase,
reaction mechanism, steady-state kinetics.
reduced the rate of NO synthesis compared with wild-type
(wt) iNOS. In addition, the W188H mutation stabilized an
inert enzyme species, P, that formed downstream of the FeO2
species, which reacted slowly with L-Arg to form NOHA [3].
Surprisingly, the W188F and W188A mutants of iNOS exhibit
defective haem binding, impeding further characterization [4].
Substitution of Trp409 by phenylalanine in rat nueronal NOS
(nNOS) reduced the haem midpoint potential of the protein, led
to a faster formation of the FeO2 species and to greater rates
of Fe(II)NO oxidation compared with the wt protein [5–7]. A
side-by-side characterization of Bacillus subtilis NOS and its
proximal tryptophan variants W66H and W66F revealed distinct
thermodynamic and kinetic behaviours depending on the nature of
the amino acid residue replacing the native tryptophan (histidine
or phenylalanine) [8–10]. Thus, the proximal tryptophan residue
may play a role in controlling the enzyme’s reactivity by tuning the
properties and reactivity of its haem. Further, a study performed
with Staphylococcus aureus NOS revealed that NOSs display a
steep dependence between the haem midpoint potential and the
rates of oxygen activation under single turnover conditions for
the hydroxylation of L-Arg to NOHA [8]. Lang et al. [8] showed
that mutations on the tryptophan residue proximal to the haem-
thiolate bond altered the haem midpoint potential and the rates of
disappearance of the FeO2 speciesin trendsthatcorrelated with the
donating or withdrawing electron density from the haem-thiolate
bond that NOS enzymes display.
INTRODUCTION
Nitric oxide synthases (NOSs) are a group of homodimeric
enzymes (EC 1.14.13.39) that catalyse the conversion of L-
arginine (L-Arg) and dioxygen into citrulline and nitric oxide.
Each monomer comprises an N-terminal oxygenase domain
(NOSoxy) incorporating binding sites for its substrate L-Arg,
(6R)-5,6,7,8-tetrahydro-L-biopterin (H4B) and a Cys-co-ordinated
haem, and a C-terminal reductase domain (NOSred) that hosts
sites for NADPH, FAD and FMN. A calmodulin (CaM)-binding
sequence bridges the NOSoxy and NOSred domains [1,2].
The sequence of reactions that leads to NO synthesis by NOS
involves several steps of reduction and oxidation. Resting NOS is
first reduced by electrons coming from the reductase domain.
Ferrous haem then reacts with molecular oxygen to form a
transient FeO2 species (Figure 1). This FeO2 intermediate is
further reduced by the cofactor H4B to form a ferric-peroxo
complex. Scission of the O-O bond follows to form the highly
valent compound I species, which ultimately oxidizes L-Arg to
form N-hydroxy-L-arginine (NOHA). Conversion of NOHA into
citrulline and NO requires a second event of oxygen activation, in
which H4B is proposed to serve as the electron donor (Figure 1).
All NOSs sequenced and crystallized to date possess a
conserved tryptophan residue that forms stacking interactions
with the porphyrin ring and hydrogen bonds with the haem-
thiolate bond. Replacement of this proximal tryptophan residue
by histidine in murine inducible NOS (iNOS) (W188H) increased
the midpoint potential of the haem group by + 88 mV and
We have employed mesohaem-substituted iNOS to study the
effect of the haem midpoint potential in catalysis. Mesohaem
Abbreviations: H4B, (6R)-5,6,7,8-tetrahydro-L-biopterin; NOHA, N-hydroxy-L-arginine; NOS, nitric oxide synthase; iNOS, inducible nitric oxide synthase;
nNOS, neuronal nitric oxide synthase; wt, wild-type.
1
Correspondence may be addressed to either of these authors (email luciana.hannibal@fau.de or stuehrd@ccf.org).
The atomic co-ordinates for iNOSoxy in complex with mesohaem and H4B have been deposited in the PDB under code 4JS9.
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

154
L. Hannibal and others
Figure 1 Model for catalysis by mammalian NOS
Ferric haem is first reduced by NADPH via intramolecular electron transfer from the reductase domain. Ferrous haem binds O2 to form FeIIO2. This haem–oxy complex rearranges to form a short-lived
FeIII-O-O– species. A second electron derived from H4B generates a ferric-peroxo species and the corresponding pterin radical, H4B+ •. The second semi-reaction that converts NOHA into citrulline
and NO is less well understood; however, some of the reaction intermediates are common to the L-Arg hydroxylation pathway.
incorporates 2,4-ethyl groups instead of the vinyl substituents
found in the naturally occurring haem, leading to an increase
in electron density. We utilized murine iNOS variant W188H to
further examine the thermodynamic component of catalysis and
attempted the reversal of the W188H phenotype by substitution
with mesohaem. We present a comprehensive examination of the
effect of porphyrin replacement on the following features: struc-
ture of wt iNOSoxy-mesohaem, haem midpoint potential, enzyme
oligomeric state, substrate binding, FeO2 stability, kinetics of
haem transitions during catalysis, extent of product formation,
reactivity of the Fe–NO complex and NO synthesis. This is the first
study to: (i) provide a crystal structure of iNOS substituted with
mesohaem; (ii) establish the differential influence of structural
and electronic components on the kinetic behaviour of iNOS; and
(iii) demonstrate that the formation of inert enzyme species P
formed by iNOS mutant W188H [3,11] is heavily influenced by
an increased midpoint potential of the haem centre.
induction, along with 0.375 mM IPTG. Cells were harvested 24–
30 h post-induction and the cell pellets were frozen at − 80◦C
until further use. Protein concentration was determined from
the absorbance at 444 nm of the ferrous haem–CO complex
in proteins carrying native haem (Fe-protoporphyrin IX), using
an molar absorption coefficient of 76 mM− 1·cm− 1 [25]. Protein
concentration was determined from the absorbance at 432 nm
of the ferrous mesohaem–CO complex, using an adjusted molar
absorption coefficient of 55 mM− 1·cm− 1 (obtained by comparison
of the respective absorption coefficient for free haem and
mesohaem). All proteins were purified to homogeneity (ꢀ95%)
as assessed by SDS/PAGE. Expression yields were low compared
with their corresponding proteins containing native haem under
the same culture conditions (15–20 mg/l of culture): iNOSoxy-
mesohaem, ∼8 mg/l of culture; W188Hoxy-mesohaem, 2 mg/l
of culture; iNOS FL-mesohaem, 1 mg/l of culture and W188H
FL-mesohaem, 0.5 mg/l of culture. Mesohaem to haem ratios
were determined by HPLC according to a published procedure
[12]. Mesohaem content ranged from 83% to 94% in both
wt and W188H iNOS. Attempts to express these proteins in
rich medium (LB broth or Terrific broth) resulted in little
incorporation of mesohaem (less than 5%). The oligomeric
state of wt and W188H iNOS reconstituted with H4B and
L-Arg was examined by size-exclusion chromatography. Protein
samples (∼150 μM) in 40 mM EPPS, pH 7.6, containing 150 mM
NaCl were incubated with 2 mM L-Arg, 400 μM H4B and 1.2 mM
DTT for 15 min. Samples were injected on a Superdex 200 resin
pre-equilibrated with 40 mM EPPS, pH 7.6, containing 150 mM
NaCl supplemented with 100 μM L-Arg, 40 μM H4B and 120 μM
DTT. Under these conditions, all proteins existed predominantly
in the dimeric state (80–90%).
EXPERIMENTAL
Reagents
H4B was purchased from Schircks Laboratories. CO gas was
obtained from Praxair, Inc. N OHA and L-[14C]Arg were
purchased from MP Biomedicals. EPPS was purchased from
Fisher Scientific. DTT was purchased from RPI. All other reagents
were purchased from Sigma.
Protein expression and purification
Wt and mutant W188H iNOS ꢀ65 (deletion of the first 65 amino
acids in the N-terminus) proteins containing a His6 tag were
overexpressed in Escherichia coli strain BL21(DE3) and purified
using Ni2+-nitrilotriacetate (Ni-NTA) affinity chromatography as
previously described [3]. Haem replacement studies required
expression in M9 minimal medium supplemented with twice
the amount of glucose and casamino acids suggested by the
manufacturer in order to achieve acceptable expression yields.
Mesohaem (5 μM final concentration) was added at the time of
Imidazole and L-Arg binding
Binding affinities of imidazole and L-Arg were studied at 25◦C
by perturbation difference spectroscopy according to methods
described previously [3,10]. NOS samples (around 5 μM) in
40 mM EPPS, pH 7.6, with 10% glycerol, 0.6 mM DTT and
0.2 mM H4B were titrated by stepwise addition of imidazole, to
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

Characterization of iNOS-mesohaem
155
a final concentration of 10 mM. The Kd of imidazole (Kdimid)
was calculated by fitting the data to a simple saturation binding
equation. The Kd of L-Arg molar absorption was determined
under the same conditions, in the presence of 10 mM imidazole.
The data were fitted to a simple saturation binding equation,
and Kd was calculated according to the following equation:
Ferrous haem–NO complex oxidation (kox
)
Native and mesohaem-substituted iNOSoxy wt or W188Hoxy
(∼5 μM) in 100 mM EPPS, pH 7.6, containing 150 mM NaCl
and 10% glycerol containing 2 mM L-Arg, 0.2 mM H4B and
1 mM DTT were carefully titrated with dithionite under anaerobic
conditions. The Fe(II)–NO complexes were generated by
stepwise addition of anaerobic NO-saturated buffer. Both proteins
displayed stable Fe(II)–NO complexes at pH 7.6 (100 mM EPPS,
pH 7.6, containing 150 mM NaCl and 10% glycerol) and at
pH 9.5 (100 mM CHES, pH 9.5, containing 150 mM NaCl and
10% glycerol). Fe(II)–NO protein samples were then transferred
to an anaerobic stopped-flow instrument using a gastight syringe,
and the reactions were initiated by mixing with air-saturated buffer
at 10◦C. Spectral data were fitted to a single exponential model,
A→B, using Specfit global analysis software.
[imid]
appKd = Kd(1 + imid ) , where appKd is the apparent Kd of the
K
enzyme for L-Arg idn the presence of saturating concentrations of
imidazole.
Single turnover reactions
L-Arg hydroxylation and NOHA oxidation experiments were
carried out in an SF61-DX2 stopped-flow instrument (Hi-Tech
Scientific) coupled to a diode array detector, as previously
described [10,13]. An anaerobic solution of 20 μM ferrous NOS
(obtained by titration of ferric NOS with a solution of dithionite),
2 mM L-Arg (or 1 mM NOHA), 0.2 mM H4B and 1 mM DTT
in 40 mM EPPS, pH 7.6, containing 10% glycerol and 150 mM
NaCl was mixed at 10◦C with a solution containing air-saturated
buffer, 2 mM L-Arg, 0.2 mM H4B and 1 mM DTT. Sequential
spectral data were fitted to a two-exponential, A → B → C,
model using the Specfit/32 global analysis software, version
3.0 (Spectrum Software Associates). Specfit is a multivariate
data analysis program for modelling and fitting 3D chemical
kinetics data by singular value decomposition (SVD). Rates are
NO synthesis
NO synthesis by native and mesohaem-substituted full-length
proteins was assessed by the oxyhaemoglobin NO capture assay
using an molar absorption coefficient of 38 mM− 1·cm− 1 for
methaemoglobin minus oxyhaemoglobin as described elsewhere
[12]. Total nitrite produced by these proteins in the presence of
NOHA was determined by the Griess assay [14] according to
published protocols [10].
+
the average for at least five reactions S.D.
−
Crystallization and data collection
Mesohaem-substituted iNOS ꢀ65 was expressed in minimal
medium and purified as described above. Crystals of murine
iNOSoxy (residues 66–498) in complex with mesohaem and
H4B were grown by hanging drop vapour diffusion at 293
K in 1 μl drops (0.5 μl of protein with 0.5 μl of screen).
Hanging drops consisted of a 1:1 ratio mixture of 20 mg/ml
iNOSoxy with 10 mM H4B and a reservoir solution. Crystals grew
within 3–6 days in a reservoir solution composed of 700 mM
ammonium sulfate, 100 mM MES, pH 5.3, and 3.5% octyl-
glucoside. Crystals were cryoprotected by brief transfer through
reservoir solution supplemented with 30% (w/v) ethylene
glycol and flash-frozen in liquid nitrogen. Diffraction data
for a crystal in space group P6122 was collected using a
Rigaku MicroMax-007HF generator and a Rigaku Saturn 044 +
CCD detector. Data integration, reduction and scaling were
performed with d*TREK [15] and processed to a cut-off of
2.78 Å. The resolution cut-off was based on significant drops in
non-averaged I/σ (<2.0) and significant increases in Rmerge
in shells at higher resolutions.
Determination of reaction products by HPLC
End-point conversion of L-[14C]Arg to [14C]NOHA under single
turnover conditions were carried out according to a previously
published procedure [10]. L-[14C]Arg and [14C]NOHA were
extracted from the reaction mixtures and the conversion of
substrate to product was monitored by HPLC. The radioactivity
of each fraction was counted using a Scintillation counter [10].
Midpoint potential determinations
Spectroelectrochemical titrations were conducted in a glove-
box (Belle Technology) under an N2 atmosphere, as previously
described [3,10]. NOS proteins were made air-free by gel filtration
in a Sephadex G-25 column (PD 10, GE Healthcare) pre-
equilibrated with anaerobic buffer (100 mM phosphate buffer,
pH 7.0 and 125 mM NaCl). Protein samples were diluted to
a 3.5 ml final volume (final concentration ∼10 μM) and L-
Arg (2 mM) and H4B (100 μM) were added. The following
electron carrier dyes (0.5–1 μM) were utilized: phenosafranine
(Em = − 252 mV), Benzyl Viologen (Em = − 358 mV),
Methyl Viologen ( − 450 mV) and anthraquinone-2-sulfonate
(Em = − 225 mV). The titration was performed at 15◦C by
stepwise addition of a sodium dithionite solution. Oxidative
titrations were carried out under the same conditions using
potassium ferricyanide to oxidize the haem centre. Reductive
and oxidative titrations yielded midpoint potential values within
Structure solution, refinement, validation and deposition
An existing structure of iNOS (PDB code 3DWJ: chain A) [3]
was used as a molecular replacement search model. Phases were
calculated by molecular replacement using the PHASER [16]
component of PHENIX [17]. The best molecular replacement
solution was subjected to iterative rounds of model building in
COOT [18] and refinement in PHENIX. The refinement protocol
used isotropic atomic displacement parameters for all atoms.
Stereochemical analysis of the iNOSoxy structure was completed
with MolProbity [19]. All residues are within the allowed regions
of the Ramachandran plot and no poor rotamers or Cβ deviations
were found. MolProbity statistics are listed in Supplementary
Table S1. The atomic co-ordinates for iNOSoxy in complex with
mesohaem and H4B in space group P6122 has been deposited in
the PDB (accession code 4JS9).
+
2 mV, suggesting that the process involves one-electron transfer
in either direction. Absorption spectra were collected with a
−
Cary 50 spectrophotometer equipped with a dip-probe detector,
and the potentials were measured using a silver/silver chloride
microelectrode saturated with 4 M KCl (Fisher Scientific).
Midpoint potential values are expressed as midpoint potential
+
error of the fit for a one-electron process.
−
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

156
L. Hannibal and others
Steady-state kinetics of the reaction of full-length W188H
containing native haem
Steady-state reactions were carried out according to the conditions
described by Abu-Soud et al. [20]. Briefly, full length W188H
(∼2 μM) was mixed with 1 mM L-Arg and 4 μM H4B/DTT
in 40 mM EPPS, pH 7.6 containing 125 mM NaCl and 10%
glycerol. This solution was mixed with 140 μM NADPH in the
same buffer, and the reaction was monitored at different time
points by rapid UV–visible scanning.
Simulations of iNOS distribution during steady-state
NO synthesis
The NO synthesis kinetics of wt iNOS were simulated using the
global model described earlier [21] and implemented in Gepasi
software version 3.30. For mutant W188H, the global model
was adjusted to account for the formation and disappearance
of inert product P. The simulations assume constant values for
[O2] = 180 μM and [NADPH] = 40 μM. Experimental kox values
for wt iNOS ferrous haem–NO complexes obtained under half
air-saturation conditions were multiplied by a factor of 1.5 (air-
saturation) [22]. Likewise, oxygen-binding rates k2 and k6 (for
wt enzymes) and k2 and k7 (for W188H mutant enzymes) were
multiplied by a factor of 2.0 (air-saturated condition) assuming
that this rate is proportional to [O2]. The selected concentrations of
NADPH and dioxygen were kept constant during each simulation,
to avoid secondary effects due to their exhaustion.
RESULTS AND DISCUSSION
Expression of mesohaem-containing iNOS proteins
The oxygenase and full-length versions of wt iNOS and
mutant W188H were expressed in E. coli grown in minimal
medium supplemented with 5 μM mesohaem. The expression
yield was very low, particularly for the full-length proteins
(yield ∼0.5 mg of protein/l of culture). Despite this practical
hurdle, the purified proteins presented the expected spectral
characteristics, suggesting the haem environment was suitable
for further characterization. Substitution of iNOS and W188H
with Fe(III)-deuterohemin and Fe(III)-2,4-diacetyl-deuterohemin
proved impossible under our experimental conditions. Gel
filtration analysis showed that both iNOSoxy and W188Hoxy
containing mesohaem exist predominantly as a dimer (results
not shown). Fractions of highly purified proteins were examined
for their haem content (Reinheitszahl values, Rz, ASoret/A280).
The haem content of both proteins substituted with mesohaem
was low compared with their corresponding native counterparts
(Rz = 1.5–2.5). wt iNOSoxy-mesohaem had an Rz of 0.45,
whereas W188Hoxy showed ∼2.5-fold decreased haem content
compared with iNOSoxy-mesohaem (Rz = 0.20). This poorer
haem load has been observed in W188Hoxy carrying native haem,
and the effect was exacerbated upon incorporation of the foreign
mesohaem. The lower mesohaem content observed in W188Hoxy
seems to be an effect of the tryptophan to histidine mutation.
This could be related to the recently noted misalignment between
the imidazole ring of His188 and the porphyrin ring as opposed
to the native arrangement of the indole group of Trp188 [11].
Conceivably, this structural restriction may affect cellular haem
insertion during protein maturation. In fact, iNOSoxy mutants
W188A and W188F are isolated as apo-proteins (unpublished
work, Zhiqiang Wang, Jesus Tejero and Dennis J. Stuehr).
Figure 2 X-ray crystal structure of iNOSoxy ꢀ65 with mesohaem (PDB code
4JS9)
(A)ThestructuresofnativeiNOSoxyꢀ65(PDBcode1NOD,grey)andiNOSoxyꢀ65-mesohaem
(magenta) are superimposed and with backbones shown as ribbons. Native haem (orange),
mesohaem (cyan) and H4B groups from native iNOSoxy ꢀ65 (blue) and iNOSoxy
ꢀ65-mesohaem (yellow) are shown as sticks. (B) A close-up of the iNOSoxy ꢀ65-mesohaem
active site highlights the spatial arrangement of haem and H4B groups (coloured as in A). (C)
Electron density for ligands in the X-ray crystal structure of iNOSoxy ꢀ65 with mesohaem
(PDB code 4JS9). The structure of iNOSoxy ꢀ65-mesohaem (magenta) is shown as ribbons
while mesohaem (cyan) and H4B groups (yellow) are shown as sticks. Electron density (2Fo-Fc,
drawn at 1σ) is drawn as a mesh (grey). A proposed location for L-Arg (not included in the
deposited co-ordinates) places the backbone and guanidinum groups into two areas of weak
electron density situated above the haem iron and is consistent with the position of L-Arg in
native iNOSoxy structure (PDB code 1NSI).
X-ray crystal structure of iNOSoxy-mesohaem
Crystals of iNOSoxy ꢀ65 substituted with mesohaem were
analysed and compared with the reported crystal structures of
iNOSoxy ꢀ65 carrying native haem (PDB code 1NOD [2]), as
well as with the structure of iNOS mutant W188Hoxy (PDB
code 3DWJ [3]). X-ray crystallographic analysis revealed a
highly conserved iNOSoxy structure upon replacement with
mesohaem. The structures of iNOSoxy-mesohaem and iNOSoxy
superimpose within 0.3 Å rmsd (Supplementary Table S1),
suggesting an almost identical structure (Figure 2A). The haem
centre displays a similar hydrogen-bonding network compared
with native iNOSoxy, which is suitable for catalysis. H4B is also
oriented within its pocket in the right configuration for electron
transfer (Figures 2B and 2C). Weak electron density was observed
for the guanidinum moiety and backbone atoms of the natural
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

Characterization of iNOS-mesohaem
157
Table 1 Spectral properties of iNOS wt and mutants W188H substituted
with mesohaem
UV–visible absorption maxima in the presence of H4B and L-Arg or imidazole (Imid) of wt
iNOSoxy and W188Hoxy substituted with mesohaem in EPPS (40 mM, pH 7.6) containing
150 mM NaCl and 10% glycerol.
Soret (nm)
Visible (nm)
iNOS-mesohaem*
Fe(III)
Fe(III)-Imid
Fe(II)
Fe(II)-CO
Fe(II)-NO
W188H-mesohaem
Fe(III)
381
417
404
431
506, 536, 642
540
549
545
559
422 (broad)
377
507, 536, 638
Fe(III)-Imid
Fe(II)
Fe(II)-CO
Fe(II)-NO
414
400
540
547
546
556
428†, 413
420 (broad)
*The values for iNOS-mesohaem are in agreement with a previous study [24].
†The species at 428 nm shifts slowly to an uncharacterized species with absorption maximum
at 413 nm.
midpoint potential compared with the corresponding proteins har-
+
bouring native Fe-protoporphyrin IX (iNOSoxy = − 261 2 mV;
−
+
−
W188Hoxy = − 173
2 mV [3]). As expected, substitution
of W188Hoxy with mesohaem did shift its midpoint potential
to a value that is closer to that of wt iNOSoxy, reversing the
effect of the mutation, thus enabling the dissection of electronic
compared with structural effects. This provided us with a unique
opportunity to interrogate the role of the midpoint potential on
the formation of inert enzyme species P that forms downstream
of the FeO2 species.
Spectroscopic properties and substrate binding of
mesohaem-containing iNOS proteins
UV–visible data for wt iNOSoxy and W188Hoxy-containing
mesohaem are given in Table 1. The UV–visible spectral features
of iNOS proteins substituted with mesohaem are essentially
identical to those reported for nNOSoxy-mesohaem [12,23].
Substitution with mesohaem led to a 12 nm blue-shift in the UV–
visible spectra of the Fe(III), Fe(III)–imidazole, Fe(II), Fe(II)–
CO and Fe(II)–NO complexes of iNOS proteins, consistent
with spectral changes observed in another variant of iNOS
[24] and in other proteins substituted with this haem analogue
[25–30]. In the presence of L-Arg and H4B both wt iNOSoxy
and W188Hoxy substituted with mesohaem exist in the typical
Fe(III) high-spin configuration, suggesting that the integrity
of the haem electronic environment remained comparable to
that of their native counterparts (Figure 4). Reduction with
dithionite and exposure to CO resulted in the formation of the
characteristic Fe(II)–CO complex with a Soret band appearing at
428–431 nm. UV–visible spectra of W188H proteins displayed
the previously observed 2–5 nm blue shift compared with
wt iNOS [3], regardless of the haem centre used. Binding
dissociation constants, Kd, for imidazole and L-Arg in the
presence of H4B were measured spectrophotometrically. Both wt
iNOSoxy and mutant W188Hoxy substituted with mesohaem dis-
Figure 3 Determination of the midpoint potentials of wt iNOSoxy and
mutant W188Hoxy substituted with mesohaem
The fraction of oxidized protein at each redox potential upon titration with dithionite was
fitted to the Nernst equation. The calculated values for the midpoint potentials are (A)
+
+
iNOSoxy-mesohaem − 295 3 mV and (B) W188Hoxy-mesohaem − 239 6 mV.
−
−
substrate L-Arg, although the density was deemed too weak for
inclusion in the final co-ordinates. The electron density for L-Arg
is consistent with the position observed in wt iNOS, with the
guanidinum moiety sitting in close proximity to the iron centre
(Figure 2C).
Redox potentiometry
An anticipated consequence of replacing native Fe-
protoporphyrin IX with Fe-mesoporphyrin IX is a lowering
of the haem midpoint potential. Redox titrations in the presence
of L-Arg and H4B were performed for mesohaem-containing
iNOSoxy and W188Hoxy by stepwise addition of dithionite
(reduction) or potassium ferrycyanide (oxidation). Nernst plots
for the reductive curves are shown in Figure 3 and the raw
data are provided in Supplementary Figure S1. The calculated
+
played competent binding for imidazole (Kd,iNOSoxy = 56 9 μM;
−
+
+
Kd,W188Hoxy = 39 2 μM) and for L-Arg (Kd,iNOSoxy = 2.6 0.4 μM;
−
−
+
+
midpoint potentials were − 295 3 mV for iNOSoxy and
Kd,W188Hoxy = 5.6 0.3 μM). The lower affinity of W188Hoxy-
−
−
+
− 239 6 mV for W188Hoxy. Thus, substitution of native
mesohaem for L-Arg compared with its wt counterpart is
not unprecedented [3]. However, substitution with mesohaem
−
haem by mesohaem led to a significant decrease in the haem
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

158
L. Hannibal and others
P with absorption maximum at 411 nm and Fe(III) as the only
detectable species (Figure 6). A summary of the measured rate
constants is given in Table 2. Fitting of the spectral data to a two-
exponential model for reaction times 2 and 0.6 s confirmed the
formation of an enzyme species with a marked shoulder at 411 nm
(results not shown). For both mesohaem-containing proteins,
disappearance of the Fe(II)–O2 species was faster compared with
their native counterparts. The effect of the midpoint potential
on the stability of the FeIIO2 species has been studied in-depth
in several enzyme systems [10,31–35]. The most widely used
approach involves the introduction of mutations to modify the
strength of the haem-thiolate bond in both NOS and cytochrome
P450 enzymes, which leads to a selective tuneable increase or
decrease in the haem midpoint potential. The FeIIO2 species is
stabilized in proteins with higher haem midpoint potentials due
to a decrease in the driving force for its reduction to FeIIIO2 .
–
Likewise, a lower midpoint potential favours the ferric form of
the protein, with the concomitant faster disappearance of the Fe–
O2 species with respect to native haem-containing proteins. This
general relationship between the stability of the FeO2 species and
the haem midpoint potential has been well documented both in
mammalian and bacterial NOSs and in related enzymes [5–7,10].
A good illustration of this phenomenon is given by the haem
transition rates for the conversion of FeO2 to its downstream
species: k2 W188Hoxy-mesohaem = 10.5 s− 1 ≈ k2 iNOSoxy
native haem = 12.5 s− 1; i.e. restoring the midpoint potential
of mutant W188H to resemble that of wt iNOSoxy regenerated
the wt-like behaviour of the FeO2 species in W188Hoxy. This
suggests that changes in haem electronics posed by the W188H
mutation can be at least partly reversed by substitution with
mesohaem, making electron density the primary factor for the
longer-lived FeO2 species seen in W188Hoxy harbouring native
haem (k2 = 2.03 s− 1 [3]). Reversal of the midpoint potential
in W188Hoxy did not result in the abolition of inert species
P that forms downstream of the FeO2 species (Figure 6),
arguing in favour of a structural effect of the mutation [11].
However, a decrease in the midpoint potential in W188Hoxy
significantly diminished the stability of species P compared
Figure 4 UV–visible spectral properties of wt iNOSoxy and mutant
W188Hoxy substituted with mesohaem, in the presence of H4B and L-Arg
with native W188Hoxy (k3 W188Hoxy mesohaem = 0.198 s− 1
;
Substitution with mesohaem led to a blue shift of ∼12 nm in the UV–visible spectra of both wt
and W188Hoxy proteins. (A) wt iNOSoxy-mesohaem. (B) W188Hoxy-mesohaem. Absorption
maxima are provided in Table 2.
k3 W188Hoxy = 0.104 s− 1), highlighting the contribution of
electronic effects.
produced a stronger association with the natural substrate
Stopped-flow analysis of a single turnover NOHA oxidation reaction
in the presence of H4B
compared with W188Hoxy possessing native haem (Kd
∼
+
29 3 μM) [3], thus partially reversing the effects of the structural
−
The fact that mutant W188H displays altered behaviour in terms
of NO synthesis compared with wt iNOS [3] prompted us
to examine NOHA oxidation reactions under single turnover
conditions in proteins substituted with mesohaem. Reaction of
wt iNOSoxy-mesohaem with NOHA and H4B could be best fitted
to a sequential model with three well-defined haem transitions
(Table 2), similar to the reaction pattern observed for this
protein carrying native haem. The observed reaction rates (k1,
k2 and k3) suggested that the reaction is slower in iNOSoxy-
mesohaem compared with the native haem-containing protein.
A stable FeIII–NO complex was identified downstream of the
FeO2 species, with a broad absorption maximum at 431 nm
(Figure 7). Single turnover reactions of W188Hoxy carrying
native haem in the presence of NOHA and H4B do also occur
via the formation of three distinct species as observed with wt
iNOSoxy, although formation and disappearance of the FeO2
species is substantially slower in the mutant (Table 5). A
remarkable effect of mesohaem substitution in mutant W188Hoxy
was the change to an extremely short-lived FeO2 species and the
replacement of Trp188 by a histidine residue.
Stopped-flow analysis of a single turnover L-Arg hydroxylation
reaction in the presence of H4B
We next investigated the enzyme haem species that occur
in iNOSoxy and W188Hoxy substituted with mesohaem. Of
particular interest was to examine the formation (or absence) of
the inert enzyme species P whose exact nature remains elusive
[3,11]. L-Arg hydroxylation reactions of mesohaem-containing
iNOSoxy could be best fitted to a two-exponential model A →B
→C, with Fe(II), Fe(II)–O2 and Fe(III) as the only detectable
species (Figure 5). We investigated whether reversal of the
midpoint potential in W188Hoxy-mesohaem affects the formation
and stability of species P, which forms downstream of the Fe–
O2 species [3,11]. L-Arg hydroxylation reactions of mesohaem-
containing W188Hoxy were fitted to a three-exponential model
A →B →C→D, with Fe(II), Fe(II)–O2, a presumptive species
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

Characterization of iNOS-mesohaem
159
Figure 5 Single turnover reactions in iNOSoxy-mesohaem with L-Arg and H4B
Global analysis with Specfit 3.0 software identified three distinct species: Fe(II) (Soret maximum ∼402 nm), Fe-O2 (Soret maximum ∼415 nm) and Fe(III) (Soret maximum ∼378 nm). Haem
transitions occurred faster than in native iNOSoxy (Table 2), with a return to the resting state within 0.4 s. (A) Spectral changes observed during single turnover reactions. (B) Haem enzyme species
identified by global analysis. (C) Time courses for each haem enzyme species calculated by global analysis.
Table 2 Single turnover reactions
Observed rate constants (s− 1) for haem transitions during single turnover reactions in wt iNOSoxy and W188Hoxy substituted with mesohaem. Values for native and mesohaem-containing nNOS
are provided for comparative purposes.
L-Arg/H4B
NOHA/H4B
Reference
k
k
k
k
k
3
1
2
1
2
iNOS
iNOSoxy
iNOSoxy-mesohaem
Fe(II)−→ FeO2−→ Fe(III)
Fe(II)−→ FeO2−→ FeNO−→ Fe(III)
+
+
+
+
+
52.7 2.2; 12.5 0.2*
47.3 3.1; 36.7 1.8; 2.30 0.09
This work
This work
−
+
−
+
−
+
−
+
−
+
37.8 2.3; 22.9 0.8
35.6 0.6; 18.1 0.2; 1.70 0.12
−
−
−
−
−
k
k
k
k
k
k
3
1
2
3
1
2
Fe(II)−→ FeO2−→ P−→ Fe(III)
Fe(II)−→ FeO2−→ FeNO−→ Fe(III)
k
k
2
1
Or Fe(II)−→ FeNO−→ Fe(III)
+
+
+
−
+
+
+
W188Hoxy
W188Hoxy-mesohaem
nNOS
nNOSoxy
46.0 1.1; 2.03 0.04; 0.104 0.003*
43.1 2.1; 2.97 0.27; 0.239 0.010
This work
This work
−
+
−
+
−
−
+
−
+
−
†
+
45.3 8.6; 10.5 0.9; 0.198 0.06
37.7 1.2; 5.75 0.08
−
−
−
−
− 1
− 1
− 1
− 1
− 1
− 1
− 1
− 1
+
+
−
+
+
−
+
183 11 s ; 17.5 0.2 s
133 6 s ; 23.8 0.4 s ; 5.1 0.1 s
[12]
[12]
−
−
−
−
−
− 1
− 1
+
+
+
+
+
nNOSoxy-meso
168 7 s ; 52.1 0.7 s
153 13 s ; 117 6 s ; 8.2 0.1 s
−
−
−
*Values taken from reference [3].
†The FeO2 species could not be observed during single turnover reactions of W188Hoxy with NOHA and H4B.
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

160
L. Hannibal and others
Figure 6 Single turnover reactions in W188Hoxy-mesohaem with L-Arg and H4B
(A) Spectral changes observed for a 2 s reaction time. (B) Global analysis of the reaction kinetics identified an additional species (P + FeIII) in addition to those observed in the single turnover
reactions of wt iNOSoxy-mesohaem. Species P previously identified during hydroxylation of L-Arg to NOHA appeared downstream of the Fe–O2 complex, with an absorption maximum at 411 nm
(shoulder). This inert enzyme species P has a shorter half-life (C and D) compared with the same reaction performed with native haem [3] and, as a result, its net build-up is diminished. The enzyme
returns to the resting ferric state with an observed rate of 0.198 s− 1
.
stabilization, for the first time, of a FeIII–NO complex (Figure 8).
The observed rates of formation and disappearance of FeIII–NO in
iNOSoxy-mesohaem were proportionally lower (∼1.5-fold) than
in the native enzyme. The FeIII–NO complexes identified during
single turnover reactions of iNOSoxy and W188Hoxy containing
mesohaem are practically identical to the same species generated
by addition of NO to an anaerobic solution of the corresponding
FeIII–NOS enzyme (insets to Figure 7A, and Figure 8B). Global
analysis of the single turnover reactions of mutant W188Hoxy-
mesohaem was compatible with a sequential two-exponential
reaction. Previous studies provided strong evidence that species P
is a hydroxide-bound haem whose formation is accompanied by
H4B• + radical build-up, rather than it being a ferryl or a peroxo-
hame species [11]. It was proposed that the haem-hydroxide
species was stabilized by the W188H mutation, and that the
hydroxide ligand prevents NO binding to the haem before it
escapes the active site channel [11]. Our findings argue against
a pure structural effect of the W188H mutation, by showing that
the stability of this presumptive hydroxide-bound haem species is
challenged by sole reversal of the midpoint potential in W188H.
This in turns rescues the enzyme’s ability to synthesize NO by
shifting oxygen activation toward the productive cycle of NOS.
Nonetheless, our data support the proposal that species P could
be indeed a hydroxide-bound haem derived from NOHA (product
complex). Binding of substrate and pterin-free W188Hoxy to the
NOHA analogues phenylguanidine and methoxyphenylguanidine
(4 mM) led to perturbation of the UV–visible spectra, as expected.
Binding of W188Hoxy to phenylguanidine produced a complex
with absorption maximum at 424 nm, suggestive of Fe–N co-
ordination. Binding to methoxyphenylguanidine, which possesses
mechanism, with the reaction being best described as: FeII
→
P + FeIII-NO →FeIII, where P defines an enzyme species
with a Soret absorption maximum at ∼420 nm that forms
in the reactions of W188H carrying native haem [3,11]. We
originally referred to this species as a ‘reaction intermediate’,
but recent work [11] and the experimental findings presented
herein strongly suggest that species P is a non-native low-spin
enzyme species that forms right after NO is made, and thus
masks or antagonizes the formation of the enzyme FeIII–NO
product complex that otherwise transiently builds up in the wt
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

Characterization of iNOS-mesohaem
161
Figure 7 Single turnover reactions in iNOSoxy-mesohaem with NOHA and H4B
Global analysis of the spectral changes identified four species during the course of this reaction: Fe(II) (Soret maximum ∼402 nm), Fe–O2 (Soret maximum ∼415 nm), Fe(III)–NO (Soret maximum
∼431 nm) and Fe(III) (Soret maximum ∼378 nm). (A) Species identified by global analysis. The inset to (A) shows authentic FeIII–NO complex generated by addition of NO to anaerobic FeIII
iNOSoxy-mesohaem. (B) Time courses for the haem enzyme species calculated by global analysis. (C) Overlay of the experimental and global analysis curves for the time course of Fe(III)–NO
species during single turnover reactions with NOHA and H4B.
FeIII–NO complex during single turnover reactions with NOHA
and the improved NO synthesis rate by this mutant (see next
section).
an available oxygen atom for axial ligation, resulted in a complex
with absorption maximum at 419 nm reminiscent of species P
(Supplementary Figure S2). This finding is in line with binding
studies carried out with NOHA [11] and points to the hydroxo
group as the axial ligand of species P.
NO synthesis
NO synthesis by full-length iNOS and W188H substituted with
Extent of L-Arg hydroxylation
mesohaem was examined by the oxyhaemoglobin capture ass−ay1 .
An examination of [14C]NOHA formation from L-[14C]Arg under
single turnover conditions at infinite time (10 min) showed that
the yield of product formation for both mesohaem-substituted
proteins (iNOSoxy = 0.40 NOHA per haem; W188Hoxy = 0.48
NOHA per haem) is in accord with previously reported data
for iNOSoxy (NOHA/haem 0.4–0.6 [36–39]) and W188Hoxy
(0.37 NOHA/haem [3]) carrying native haem. Substitution of
mesohaem in W188H led to a slightly greater yield of L-Arg
hydroxylation. This correlates well with the formation of a stable
The rate of NO synthesis by wt iNOS-mesohaem (34.9 3 min
)
+
−
was lower than that reported for the native enzyme (51–66 min− 1
[3,40]). In contrast, the NO synthesis rate of W188H-mesohaem
was 22 5.4 min− 1, which is approximately 4-fold higher than
the documented value for native W188H [3]. Although wt iNOS
+
−
has a higher NO synthesis capacity than mutant W188H with
mesohaem, the difference is less than in their counterparts with
native haem. Thus, replacement of native haem by mesohaem
improved the NO synthesis yield of iNOS mutant W188H.
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

162
L. Hannibal and others
Figure 8 Single turnover reactions in W188Hoxy-mesohaem with NOHA and H4B
(A) Spectral changes observed during a 0.6 s reaction time. (B) Global analysis of the spectral changes showed an extremely short-lived FeIIO2 species, with three major observable haem transitions:
Fe(II) (Soret maximum ∼402 nm), Fe(III)–NO (Soret maximum ∼418–422 nm) and Fe(III) (Soret maximum ∼376–380 nm). Build-up of a FeIII–NO complex was observed for the first time in
the reactions of W188Hoxy, and it is thought to be a result of changes in the electronic environment of the haem. The inset to (B) shows authentic FeIII–NO complex generated by addition of NO
to anaerobic FeIII W188Hoxy-mesohaem. (C) Time courses for each species calculated by global analysis. (D) Overlay of the experimental trace for the formation and disappearance of Fe(III)–NO
complex and its calculated fit obtained by global analysis.
into FeIII compared with the native-haem containing versions
(iNOSoxy kox = 3.11 s− 1 at pH 7.6) [43]. Overall, the FeII–NO
oxidation rates by dioxygen in mutant W188H-mesohaem are ∼2-
fold slower compared with wt iNOS-mesohaem at both pH 7.6
and pH 9.5. This decreased kox rate is nonetheless relatively high
compared with the predicted haem reduction rate for W188H-
mesohaem (0.3 s− 1), thus of predictably little effect on the
enzyme’s steady-state behaviour and NO synthesis activity.
Oxidation rate of the ferrous–NO complex (kox
)
During NO synthesis, an FeIII–NO complex forms and can be
reduced by NOS reductase at rates comparable to that of FeIII–
NO dissociation [41,42]. Reaction of the ferrous–NO complex
with O2 resumes the catalytic cycle, ultimately leading to NO
release. We investigated the oxidation of anaerobic FeII–NO
complexes by O2 ([O2] ∼ 180 μM) by stopped-flow spectroscopy
at pH 7.6 and pH 9.5 (Supplementary Figure S3). Oxidation of
FeII–NO by O2 occurred as a single step. Observed rate constants
were derived by fitting the data to a first-order exponential
model. FeII–NO complexes were stable at both pHs, with rates
+
Haem reduction rates of iNOS and W188H containing native haem
(kr)
in the following rank order: iNOS-mesohaem pH 7.6 = 8.89
−
0.43 s− 1 > iNOS-mesohaempH 9.5 = 5.08 0.09 s− 1 >W188H-
+
−
The measured haem reduction rates for iNOS and W188H
+
mesohaem pH 7.6 = 4.68
0.05 s− 1 > W188H-mesohaem
−
+
under our experimental conditions were kr = 0.5
0.03 and
pH 9.5 = 2.89 0.16 s− 1. At both pHs, the FeII–NO complex
−
+
0.03 s− 1, respectively (Supplementary Figure S4). The
−
+
1.79
−
of W188Hoxy-mesohaem reacted more slowly with O2 compared
with wt iNOSoxy-mesohaem, and more generally, the mesohaem-
containing proteins displayed a faster conversion of FeII–NO
kr values reported for iNOS under similar conditions are 0.6–1.2
s− 1 [20,42,43]. Poor expression yields of the corresponding full-
length mesohaem-containing proteins prevented our measuring
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

Characterization of iNOS-mesohaem
163
Figure 9 Partition of haem-enzyme species during the reactions of wt and W188H iNOS simulated with Gepasi software version 3.3
The reactions of wt iNOS are characterized by FeIII as the predominant species, with a shorter-lived FeO2 species in iNOS containing mesohaem compared with its native counterpart. In contrast,
mutant W188H features the formation and build-up of an inert species, P, which competes with both the productive NO–synthesis pathway and the futile cycle of iNOS. Incorporation of mesohaem
into W188H reduces the formation of and accelerates the decay of enzyme species P enabling a larger fraction of the oxygen-activated substrate to be channelled into the productive cycle of NO
synthesis.
their haem kr, therefore estimated values were calculated by
extrapolation of kr values from a previous study performed
with nNOS, where kr for nNOS containing native or mesohaem
were: nNOS native = 6.6 s− 1, nNOS mesohaem = 1.1 s− 1 [12].
Assuming a similar trend would apply to iNOS, we estimated
values of kr would be 0.1 and 0.3 s− 1 for iNOS-mesohaem
and W188H-mesohaem, respectively. While assumptions carry an
intrinsic limitation, it is unlikely that haem reduction rates in iNOS
will deviate substantially from the trend observed in nNOS for two
fundamental reasons: (i) the haem electronic environment and
the surface amino acid residues involved in reductase-oxygenase
interactions necessary for electron transfer are highly conserved
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

164
L. Hannibal and others
in both enzymes [44], and (ii) X-ray structural analysis presented
herein indicates that substitution of native haem with mesohaem
in iNOSoxy did not alter the bonding interactions around the
porphyrin moiety where reduction by the reductase domain would
occur. We utilized these estimates of kr to model steady-state
reactions that otherwise could not be studied experimentally.
reaction with L-Arg has been noted in the case of nNOS substituted
with mesohaem [12]. Build-up of the FeIII–NO complex was
faster in W188Hoxy-mesohaem compared with wt iNOSoxy-
mesohaem, and the same was true for its transition to FeIII. This is
favourable for NO synthesis in W188H, by diminishing diversion
of the available NO into the unproductive NO-bound form of the
enzyme [7,41], but not sufficient to account for its greater NO
synthesis performance compared with its native counterpart. A
similar relationship has been identified in nNOS mutant W409F
[6,7], B. subtilis mutant W66F [10] and in nNOS harbouring
Fe-mesoporphyrin IX [12], all of which display reduced haem
midpoint potentials compared with their respective native or
wt counterparts. While our results support the recent proposal
that structural changes are partly responsible for the build-up
of species P in iNOS W188H [11], the remarkable effect of
mesohaem insertion in W188H leading to partial restoration
of catalytic activity points to haem electronics as a key factor
in controlling the formation and lifetime of species P. The present
study advanced our understanding of the nature of species P
by demonstrating that: (i) the haem midpoint potential is a
fundamental contributor to the formation of species P, (ii) species
P forms during steady-state conditions, and (iii) decreasing the
build-up of species P leads to an increase in NO synthesis by
W188H. From a mechanistic standpoint, we conclude that species
P is a hydroxo-haem ligated state that halts the reaction at a new
point, competing with both the productive and futile cycles of
NOS.
Steady-state analysis of W188H carrying native haem
The enzyme species formed under steady-state conditions were
examined in full-length W188H in the presence of L-Arg,
H4B and NADPH. Spectra collected during 2 min (300 scans)
(Supplementary Figure S5A) were subjected to global analysis
using Specfit software as described earlier. The reaction could be
best fitted to a three-exponential sequential model, with FeIII as
the initial and final species, and FeII and inert product P as the
other two detectable haem species (Supplementary Figure S5B).
This is the first study to demonstrate the formation of species P
in a NOS enzyme under steady-state conditions. The long-life
(Supplementary Figure S5C) and spectral features of species P
(Supplementary Figures S5B and S5D) are consistent with those
observed during single turnover reactions with L-Arg and NOHA,
thus confirming that product P forms downstream of the FeIIO2
intermediate in both phases of oxygen activation in NOS catalysis.
Steady-state simulations of wt and W188H iNOS
To examine the predicted partition of enzyme species under
our experimental conditions, the experimental kinetic parameters
obtained for native and mesohaem-containing wt and W188H
iNOS (Supplementary Tables S2 and S3) were submitted to
Gepasi software version 3.30 pre-loaded with our global model for
NOS catalysis as described above. The results of these simulations
are presented in Figure 9. Substitution of mesohaem significantly
increased the partition of the respective enzymes towards the ferric
state, FeIII, and reduced the build-up of the FeIIO2 intermediate.
Replacement of native haem with mesohaem in W188H restored
the relative partition of enzymes species to a pattern that resembles
that of wt iNOS, i.e. it led to a substantial decrease in build-up of
species P, with the concomitant stabilization of ferric enzyme.
While we observe FeIII–NO complex formation during single
turnover reactions with NOHA in W188H, this species is unlikely
to build up under steady-state conditions where product P is the
predominant species when L-Arg is used as the substrate and
NADPH as the source of electrons (Supplementary Figure S5).
Our global model for W188H was built to depict this experimental
finding (Supplementary Figure S7). We next ran simulations of the
steady-state reactions for all four proteins utilized in the present
study (Supplementary Figures S6 and S7). We found that the
relative distribution of each enzyme species is consistent with
their different kinetic parameters and is sensitive to mesohaem
replacement both in wt and in W188H iNOS (Figure 9).
AUTHOR CONTRIBUTION
Luciana Hannibal designed the study, performed the experiments and prepared the paper;
Richard Page and Saurav Misra solved the crystal structure of iNOSoxy-mesohaem and
wrote the corresponding Experimental and Results sections of the paper; Mohammad
Mahfuzul Haque carried out the simulation studies under steady-state conditions; Karthik
Bolisetty purified proteins and performed experiments; Zhihao Yu purified proteins and
obtained high-quality crystals of iNOSoxy-mesohaem; Dennis Stuehr designed the study
and prepared the paper. All authors performed a critical reading of the paper prior to
submission and approved the final version of the paper.
ACKNOWLEDGEMENTS
We thank Deborah Durra for excellent technical support for the preparation of iNOSoxy-
mesohaem and Dr D. Mansuy and Dr J.L. Boucher for the gift of NOHA analogues
phenylguanidine and methoxyphenylguanidine.
FUNDING
The present work was supported by the National Institutes of Health [grant numbers
CA53914, GM51491, HL76491] (to D.J.S.), a National Institutes of Health Postdoctoral
Fellowship [grant number T32 HL007914 (to R.C.P.)] and the American Heart Association
Postdoctoral Fellowship [grant number 11POST650034 (to L.H.)].
REFERENCES
Conclusions
1
2
3
Crane, B.R., Arvai, A.S., Gachhui, R., Wu, C., Ghosh, D.K., Getzoff, E.D., Stuehr, D.J. and
Tainer, J.A. (1997) The structure of nitric oxide synthase oxygenase domain and inhibitor
complexes. Science 278, 425–431 CrossRef PubMed
Crane, B.R., Arvai, A.S., Ghosh, D.K., Wu, C., Getzoff, E.D., Stuehr, D.J. and Tainer, J.A.
(1998) Structure of nitric oxide synthase oxygenase dimer with pterin and substrate.
Science 279, 2121–2126 CrossRef PubMed
Tejero, J., Biswas, A., Wang, Z.Q., Page, R.C., Haque, M.M., Hemann, C., Zweier, J.L.,
Misra, S. and Stuehr, D.J. (2008) Stabilization and characterization of a heme-oxy
reaction intermediate in inducible nitric-oxide synthase. J. Biol. Chem. 283,
33498–33507 CrossRef PubMed
We employed two different strategies to investigate the kinetic
behaviour of iNOS in depth: (i) mutation of the proximal
tryptophan residue Trp188 to histidine, and (ii) replacement
of native protohaem with the electron- rich mesohaem. The
kinetic behaviour of W188Hoxy-mesohaem with NOHA and H4B
differed from its native counterpart in two ways: (i) a FeIIO2
species could not be detected during single turnover reactions,
and (ii) a stable FeIII–NO was observed, for the first time. The
reduced stability of the FeIIO2 species, also observed in the half-
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society

Characterization of iNOS-mesohaem
165
4
5
Wilson, D.J. and Rafferty, S.P. (2001) A structural role for tryptophan 188 of inducible
nitric oxide synthase. Biochem. Biophys. Res. Commun. 287, 126–129 PubMed
Adak, S., Crooks, C., Wang, Q., Crane, B.R., Tainer, J.A., Getzoff, E.D. and Stuehr, D.J.
(1999) Tryptophan 409 controls the activity of neuronal nitric-oxide synthase by
regulating nitric oxide feedback inhibition. J. Biol. Chem. 274, 26907–26911
CrossRef PubMed
Adak, S. and Stuehr, D.J. (2001) A proximal tryptophan in NO synthase controls activity
by a novel mechanism. J. Inorg. Biochem. 83, 301–308 CrossRef PubMed
Adak, S., Wang, Q. and Stuehr, D.J. (2000) Molecular basis for hyperactivity in tryptophan
409 mutants of neuronal NO synthase. J. Biol. Chem. 275, 17434–17439
CrossRef PubMed
25 Stuehr, D.J. and Ikeda-Saito, M. (1992) Spectral characterization of brain and
macrophage nitric oxide synthases. Cytochrome P-450-like hemeproteins that contain a
flavin semiquinone radical. J. Biol. Chem. 267, 20547–20550
26 Jeyarajah, S. and Kincaid, J.R. (1990) Resonance Raman studies of hemoglobins
reconstituted with mesoheme. Unperturbed iron-histidine stretching frequencies in a
functionally altered hemoglobin. Biochemistry 29, 5087–5094 PubMed
27 Kincaid, J.R., Zheng, Y., Al-Mustafa, J. and Czarnecki, K. (1996) Resonance Raman
spectra of native and mesoheme-reconstituted horseradish peroxidase and their catalytic
intermediates. J. Biol. Chem. 271, 28805–28811 CrossRef PubMed
6
7
28 Seybert, D.W. and Moffat, K. (1977) Structure of hemoglobin reconstituted with
mesoheme. J. Mol. Biol. 113, 419–430 CrossRef PubMed
8
9
Lang, J., Santolini, J. and Couture, M. (2011) The conserved Trp-Cys hydrogen bond
dampens the “push effect” of the heme cysteinate proximal ligand during the first catalytic
cycle of nitric oxide synthase. Biochemistry 50, 10069–10081
29 Seybert, D.W., Moffat, K. and Gibson, Q.H. (1976) Ligand binding properties of horse
hemoglobins containing deutero- and mesoheme. J. Biol. Chem. 251, 45–52 PubMed
30 Wojaczynski, J., Wojtowicz, H., Bielecki, M., Olczak, M., Smalley, J.W., Latos-Grazynski,
L. and Olczak, T. (2011) Iron(III) mesoporphyrin IX and iron(III) deuteroporphyrin IX bind
to the Porphyromonas gingivalis HmuY hemophore. Biochem. Biophys. Res. Commun.
411, 299–304 CrossRef PubMed
31 Adak, S., Aulak, K.S. and Stuehr, D.J. (2001) Chimeras of nitric-oxide synthase types I
and III establish fundamental correlates between heme reduction, heme-NO complex
formation, and catalytic activity. J. Biol. Chem. 276, 23246–23252 CrossRef PubMed
32 Ost, T.W., Miles, C.S., Munro, A.W., Murdoch, J., Reid, G.A. and Chapman, S.K. (2001)
Phenylalanine 393 exerts thermodynamic control over the heme of flavocytochrome P450
BM3. Biochemistry 40, 13421–13429 CrossRef PubMed
33 Ost, T.W., Munro, A.W., Mowat, C.G., Taylor, P.R., Pesseguiero, A., Fulco, A.J., Cho, A.K.,
Cheesman, M.A., Walkinshaw, M.D. and Chapman, S.K. (2001) Structural and
spectroscopic analysis of the F393H mutant of flavocytochrome P450 BM3. Biochemistry
40, 13430–13438 CrossRef PubMed
34 Matsumura, H., Wakatabi, M., Omi, S., Ohtaki, A., Nakamura, N., Yohda, M. and Ohno, H.
(2008) Modulation of redox potential and alteration in reactivity via the peroxide shunt
pathway by mutation of cytochrome P450 around the proximal heme ligand. Biochemistry
47, 4834–4842 CrossRef PubMed
35 Yoshioka, S., Takahashi, S., Ishimori, K. and Morishima, I. (2000) Roles of the axial push
effect in cytochrome P450cam studied with the site-directed mutagenesis at the heme
proximal site. J. Inorg. Biochem. 81, 141–151 CrossRef PubMed
36 Wei, C.C., Wang, Z.Q., Wang, Q., Meade, A.L., Hemann, C., Hille, R. and Stuehr, D.J.
(2001) Rapid kinetic studies link tetrahydrobiopterin radical formation to heme-dioxy
reduction and arginine hydroxylation in inducible nitric-oxide synthase. J. Biol. Chem.
276, 315–319 CrossRef PubMed
37 Boggs, S., Huang, L. and Stuehr, D.J. (2000) Formation and reactions of the
heme-dioxygen intermediate in the first and second steps of nitric oxide synthesis as
studied by stopped-flow spectroscopy under single-turnover conditions. Biochemistry
39, 2332–2339 CrossRef PubMed
38 Hurshman, A.R., Krebs, C., Edmondson, D.E., Huynh, B.H. and Marletta, M.A. (1999)
Formation of a pterin radical in the reaction of the heme domain of inducible nitric oxide
synthase with oxygen. Biochemistry 38, 15689–15696 CrossRef PubMed
39 Bec, N., Gorren, A.C., Voelker, C., Mayer, B. and Lange, R. (1998) Reaction of neuronal
nitric-oxide synthase with oxygen at low temperature. Evidence for reductive activation of
the oxy-ferrous complex by tetrahydrobiopterin. J. Biol. Chem. 273, 13502–13508
PubMed
CrossRef PubMed
Brunel, A., Wilson, A., Henry, L., Dorlet, P. and Santolini, J. (2011) The proximal
hydrogen bond network modulates Bacillus subtilis nitric-oxide synthase electronic and
structural properties. J. Biol. Chem. 286, 11997–12005 CrossRef PubMed
10 Hannibal, L., Somasundaram, R., Tejero, J., Wilson, A. and Stuehr, D.J. (2011) Influence
of heme-thiolate in shaping the catalytic properties of a bacterial nitric-oxide synthase. J.
Biol. Chem. 286, 39224–39235 CrossRef PubMed
11 Sabat, J., Egawa, T., Lu, C., Stuehr, D.J., Gerfen, G.J., Rousseau, D.L. and Yeh, S.R.
(2013) Catalytic intermediates of inducible nitric-oxide synthase stabilized by the W188H
mutation. J. Biol. Chem. 288, 6095–6106 CrossRef PubMed
12 Tejero, J., Biswas, A., Haque, M.M., Wang, Z.Q., Hemann, C., Varnado, C.L., Novince, Z.,
Hille, R., Goodwin, D.C. and Stuehr, D.J. (2010) Mesohaem substitution reveals how
haem electronic properties can influence the kinetic and catalytic parameters of neuronal
NO synthase. Biochem. J. 433, 163–174 CrossRef PubMed
13 Wei, C.C., Wang, Z.Q. and Stuehr, D.J. (2002) Nitric oxide synthase: use of stopped-flow
spectroscopy and rapid-quench methods in single-turnover conditions to examine
formation and reactions of heme-O2 intermediate in early catalysis. Methods Enzymol.
354, 320–338 CrossRef PubMed
14 Griess, P. (1879) Bemerkungen zu der abhandlung der H.H. Weselsky und Benedikt
“Ueber einige azoverbindungen.”. Chem. Ber. 12, 426–428 CrossRef
15 Pflugrath, J.W. (1999) The finer things in X-ray diffraction data collection. Acta
Crystallogr. D Biol. Crystallogr. 55, 1718–1725 CrossRef PubMed
16 McCoy, A.J., Grosse-Kunstleve, R.W., Adams, P.D., Winn, M.D., Storoni, L.C. and Read,
R.J. (2007) Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674
CrossRef PubMed
17 Adams, P.D., Afonine, P.V., Bunkoczi, G., Chen, V.B., Davis, I.W., Echols, N., Headd, J.J.,
Hung, L.W., Kapral, G.J., Grosse-Kunstleve, R.W. et al. (2010) PHENIX: a comprehensive
Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol.
Crystallogr. 66, 213–221 CrossRef PubMed
18 Emsley, P., Lohkamp, B., Scott, W.G. and Cowtan, K. (2010) Features and development of
Coot. Acta Crystallogr. D Biol. Crystallogr. 66, 486–501
CrossRef PubMed
19 Davis, I.W., Leaver-Fay, A., Chen, V.B., Block, J.N., Kapral, G.J., Wang, X., Murray, L.W.,
Arendall, W.B. 3rd, Snoeyink, J., Richardson, J.S. and Richardson, D.C. (2007)
MolProbity: all-atom contacts and structure validation for proteins and nucleic acids.
Nucleic Acids Res. 35, W375–W383 CrossRef PubMed
20 Abu-Soud, H.M., Ichimori, K., Nakazawa, H. and Stuehr, D.J. (2001) Regulation of
inducible nitric oxide synthase by self-generated NO. Biochemistry 40, 6876–6881
CrossRef PubMed
21 Haque, M.M., Tejero, J., Bayachou, M., Wang, Z.Q., Fadlalla, M. and Stuehr, D.J. (2013)
Thermodynamic characterization of five key kinetic parameters that define neuronal nitric
oxide synthase catalysis. FEBS J. 280, 4439–4453 CrossRef PubMed
22 Tejero, J., Santolini, J. and Stuehr, D.J. (2009) Fast ferrous heme-NO oxidation in nitric
oxide synthases. FEBS J. 276, 4505–4514 CrossRef PubMed
40 Wang, Z.Q., Tejero, J., Wei, C.C., Haque, M.M., Santolini, J., Fadlalla, M., Biswas, A. and
Stuehr, D.J. (2012) Arg375 tunes tetrahydrobiopterin functions and modulates catalysis
by inducible nitric oxide synthase. J. Inorg. Biochem. 108, 203–215 CrossRef PubMed
41 Santolini, J., Adak, S., Curran, C.M. and Stuehr, D.J. (2001) A kinetic simulation model
that describes catalysis and regulation in nitric-oxide synthase. J. Biol. Chem. 276,
1233–1243 CrossRef PubMed
42 Santolini, J., Meade, A.L. and Stuehr, D.J. (2001) Differences in three kinetic parameters
underpin the unique catalytic profiles of nitric-oxide synthases I, II, and III. J. Biol. Chem.
276, 48887–48898 CrossRef PubMed
23 Bender, A.T., Kamada, Y., Kleaveland, P.A. and Osawa, Y. (2002) Assembly and activation
of heme-deficient neuronal NO synthase with various porphyrins. J. Inorg. Biochem. 91,
625–634 CrossRef PubMed
43 Wang, Z.Q., Wei, C.C. and Stuehr, D.J. (2010) How does a valine residue that modulates
heme-NO binding kinetics in inducible NO synthase regulate enzyme catalysis? J. Inorg.
Biochem. 104, 349–356 CrossRef PubMed
24 Woodward, J.J., Martin, N.I. and Marletta, M.A. (2007) An Escherichia coli
expression-based method for heme substitution. Nat. Methods 4, 43–45
CrossRef PubMed
44 Panda, K., Haque, M.M., Garcin-Hosfield, E.D., Durra, D., Getzoff, E.D. and Stuehr, D.J.
(2006) Surface charge interactions of the FMN module govern catalysis by nitric-oxide
synthase. J. Biol. Chem. 281, 36819–36827 CrossRef PubMed
Received 29 October 2014/23 December 2014; accepted 22 January 2015
Published as BJ Immediate Publication 22 January 2015, doi:10.1042/BJ20141319
ꢀ
c
ꢀ
c
The Authors Journal compilation 2015 Biochemical Society