Photo-control of nitric oxide synthase activity using a caged isoform specific inhibitor.

Article abstract of DOI:10.1016/S0968-0896(02)00050-0

Nitric oxide (NO) plays a critical role in a number of physiological processes and is produced in mammalian cells by nitric oxide synthase (NOS) isozymes. Because of the diverse functions of NO, pharmaceutical interventions which seek to abrogate adverse

Full text of DOI:10.1016/S0968-0896(02)00050-0

Bioorganic & Medicinal Chemistry 10 (2002) 1919–1927
Photo-Control of Nitric Oxide Synthase Activity Using
a Caged Isoform Specific Inhibitor
Heather J. Montgomery,a,y Basil Perdicakis,b,y Dan Fishlock,a,y Gilles A. Lajoie,a
Eric Jervisb and J. Guy Guillemettea,*
aDepartment of Chemistry, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1
bDepartment of Chemical Engineering, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1
Received 9 October 2001; accepted 18 December 2001
Abstract—Nitric oxide (NO) plays a critical role in a number of physiological processes and is produced in mammalian cells by
nitric oxide synthase (NOS) isozymes. Because of the diverse functions of NO, pharmaceutical interventions which seek to abrogate
adverse effects of excess NOS activity must not interfere with the normal regulation of NO levels in the body. A method has been
developed for the control of NOS enzyme activity using the localized photochemical release of a caged isoform-specific NOS inhi-
bitor. The caged form of an iNOS inhibitor has been synthesized and tested for photosensitivity and potency. UV and multiphoton
uncaging were verified using a hemoglobin-based assay. IC50 values were determined for the inhibitor (70Æ11 nM), the caged
inhibitor (1098Æ172 nM), the UV uncaged inhibitor (67Æ26 nM) and the multiphoton uncaged inhibitor (73Æ11 nM). UV irra-
diation of the caged inhibitor resulted in a 86% reduction in iNOS activity after 5 min. Multiphoton uncaging had an apparent first
order time constant of 0.007Æ0.001 minÀ1. A therapeutic range exists, with molar excess of inhibitor to enzyme from 3- to 7-fold,
over which the full dynamic range of the inhibition can be exploited. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
smooth muscle cells lining the cardiovascular system
causing relaxation. eNOS also inhibits the adhesion of
leukocytes to the arterial and venous walls. The nNOS
enzyme is active in the central and peripheral nervous
system and produces NO that may be involved in neu-
rotransmission and memory formation.2 The unregu-
lated Ca2+ influx occurring in cerebral ischemia causes
nNOS to overproduce cytotoxic levels of NO.4
Nitric oxide is an inorganic free radical gas that is
implicated in a number of biological functions. NO
plays a crucial regulatory role in the cardiovascular sys-
tem1 and is involved in neurotransmission, non-specific
immunity, vasodilation, platelet inhibition, bronchodi-
lation, and contractions of heart and skeletal muscle, as
well as perpetuating the inflammatory process upon tissue
injury.2 NO is produced in different mammalian tissues
by three distinct nitric oxide synthase isoforms (NOS;
EC 1.14.13.39). These enzymes are homodimeric pro-
teins that catalyse a five-electron oxidation of l-arginine
to yield NO and l-citrulline.
The transcriptionally regulated third isoform, inducible
NOS (iNOS), was first identified in macrophages but is
now recognized to be produced by most eukaryotic
cells.5 The production of iNOS is absent until induced
by hypoxia/ischemia or immunostimulants such as bac-
terial lipopolysaccharide and cytokines.2 Once expressed,
iNOS is active, essentially unregulated, and produces
NO at potentially cytotoxic levels, resulting in local tissue
damage.6 The control of NO production is therefore
important in the regulation of a large number of physi-
ological processes. During cerebral ischemia the role of
NO is protective or destructive depending on the stage
of evolution of the ischemic process and the cellular com-
partment producing NO.7 In the acute phase of patho-
genesis during ischemia, parenchymal and microvascular
NO overproduction is due to activation of nNOS and
eNOS. The NO produced by eNOS is likely beneficial
by acting as a vasodilator and improving blood flow to
the affected area by inhibiting platelet aggregation and
Two of the three NOS isoforms, endothelial (eNOS) and
neuronal (nNOS), are constitutively expressed (cNOS)
enzymes. The cNOS enzymes are inactive at basal cellular
levels of calcium (Ca2+). An influx of Ca2+ is necessary
for calmodulin (CaM) to bind and activate these two
NOS isozymes. The eNOS enzyme is membrane bound
and is located in the endothelial lining of the vascular
system.3 The NO produced by eNOS diffuses into the
*Corresponding author. Tel.: +1-519-888-4567 x5954; fax: +1-519-
746-0435; e-mail: jguillem@sciborg.uwaterloo.ca (J. G. Guillemette).
yThese authors made equal contributions to the research.
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00050-0

1920
H. J. Montgomery et al. / Bioorg. Med. Chem. 10 (2002) 1919–1927
leukocyte adhesion.8 In contrast, studies using nNOS
knockout mice and selective inhibitors for nNOS indicate
that NO produced by nNOS contributes to the initial
metabolic deterioration of the infarct penumbra.9À11
NO produced by iNOS contributes to progressive tissue
damage leading to larger infarcts during the later stages
of the ischemic process.12 Thus, the development of
methods for the proper management of the spatial and
temporal production of NO in the post-ischemic brain or
heart may ultimately lead to treatments that will decrease
cell damage from stroke and other pathophysiological
processes associated with NO over-production. In addi-
tion, the controlled release of NOS isoform-specific
inhibitors should allow for a more robust treatment
program. The spatial and temporal control of inhibition
would prevent undesirable side effects arising from
inhibition of NO production by all three isoforms in
other parts of the patient, (i.e., evidence of constitutive
iNOS expression in the human retina, cerebellum, and
skeletal muscle tissue has been reported).13 To date
there are no reported treatments that would permit
selective inhibition of specific NOS isoforms with spatial
and temporal control.
Figure 1. Molecular structures of 1400W (1) and Bhc-1400W (3).
cross section to a therapeutic agent. This technology
could be used to minimise NO related damage inflicted
during ischemia or inflammation by selectively inhibit-
ing NO production in the vicinity of affected tissues,
while not interfering with iNOS activity in other regions
of the body. The use of MP photolysis to selectively
activate pharmaceuticals confers significant advantages
over current therapies including minimisation of drug
side effects in non-target tissue, real time control of drug
concentration, and the delivery of active drugs to spe-
cific regions of tissue. Our current research is aimed at
the creation of a new class of therapy where MP tech-
niques are utilised in conjunction with caged thera-
peutics to effect spatial and temporal control of drug
activation.
Multiphoton (MP) microscopy has recently been intro-
duced as an alternative to standard confocal microscopy
methods.14,15 With MP excitation, molecular fluores-
cence is induced by absorbing two or more infrared (IR)
photons simultaneously. MP excitation is achieved with
femtosecond pulsed IR laser irradiation. The require-
ment of near simultaneous interaction of multiple pho-
tons with the target fluorophore produces a restricted
region of fluorescence excitation about the focal volume
(i.e., sufficient photon flux density only occurs at the
focal point of the objective lens). For therapeutic pur-
poses, MP absorption confers the advantages of intrin-
sic confinement of excitation, greater tissue penetration
depth, and reduced tissue damage over visible and UV
radiation sources.
Results
Properties of the inhibitor 1400W
The NOS inhibitor 1400W (1), was readily synthesised
as previously described.22 The amount of inhibition by
1400W was determined using a modified hemoglobin
capture assay for the NOS enzyme.23,24 An IC50 value of
70Æ11 nM was measured for 1400W inhibition of the
human iNOS enzyme. The human iNOS enzyme used in
the investigation carries a deletion of the first 70 amino
acids and an amino terminal polyhistidine tail.
MP techniques have been employed for the release of
‘caged’ compounds in biological samples.16,17 Less than
1 mm3 of sample is uncaged during irradiation because
uncaging occurs only in the focal volume, thus permit-
ting the spatially confined release of a molecule from a
MP-labile protecting group. There is considerable
ongoing research directed toward developing molecules
with increased two-photon cross section.18,19 Of parti-
cular significance to this report is the recent synthesis of
7-hydroxycoumarin-4-ylmethyl esters and carbamates
with uncaging cross sections 100 times higher than pre-
viously reported for MP-induced uncaging.20
While some NOS inhibitors covalently interact with the
heme iron, others only bind near the heme iron. The
type of substrate or inhibitor interaction can be mon-
itored by optical difference spectroscopy.25 Substrate
perturbation spectroscopy was used to investigate the
interactions of iNOS with l-arginine and 1400W.
Increasing concentrations of 1400W added to iNOS
cause a type-I difference spectrum similar to that
obtained with l-arginine (results not shown). This is
consistent with the recently published crystal structure
of the eNOS–1400W complex.26 The amidine group of
1400W was found to anchor the compound in the heme
active site through hydrogen bonds to Glu 363 and the
Trp 358 carbonyl oxygen.26 Our Ks values for l-arginine
and 1400W are 2.6 and 4.2 mM, respectively, in close
agreement with reported kinetic constants for l-arginine
(Ks=2.5 mM)25 and 1400W (K1=2.0 mM).21
Here, we report the protection of the iNOS selective inhi-
bitor 1400W21 (1) with the MP-labile coumarin-derived
protecting group developed by Furuta and co-workers20
to produce the caged inhibitor (3) (Fig. 1). 1400W binds to
human iNOS competitively with l-arginine, resulting in a
time-dependent loss of enzyme activity.21
Properties of the caged inhibitor Bhc-1400W
To our knowledge, this is the first reported conjugation
of a caging molecule with a large two photon uncaging
Coupling of the 6-bromo-7-hydroxy-4-hydroxymethyl-
coumarin (Bhc) (2) group to 1400W via a carbamate

H. J. Montgomery et al. / Bioorg. Med. Chem. 10 (2002) 1919–1927
1921
Scheme 1. (a) DMAP, 4-nitrophenylchloroformate, CH3CN, rt, 9 h;
(b) 1, DMAP, 24 h.
linker produced the caged compound Bhc-1400W (3)
(Scheme 1).
Figure 2. Time course of photolysis of Bhc-1400W. The kinetics of
photolysis following ultra-violet (*, inset) or two-photon IR (&)
irradiation compared to uncaging in the dark (~). Fractional conver-
sion of Bhc-1400W to 1400W versus uncaging time data was fit to a
decaying exponential with a time constant of 0.007Æ0.001 minÀ1 for
two-photon IR uncaging (R2=0.98) and 0.39Æ0.28 minÀ1 for UV
uncaging (R2=0.95). Error bars indicate 95% confidence intervals and
are representative of collected data.
Although Bhc-1400W was readily soluble in 10%
methanol, ethanol, or DMSO, the presence of these
solutions was found to interfere with enzyme kinetics.
Further investigation determined that Bhc-1400W was
soluble in KMops buffer (pH 7.2) with a maximum
solubility of approximately 900 mM. Stock solutions in
KMops at 600 mM were routinely prepared. No dark
uncaging was observed for a 300 mM sample after 6.5 h
storage at 4ꢀ in the dark (Fig. 2).
MP showed similar kinetics to that of Bhc-glutamate as
determined by Furuta and co-workers (first order rate
constant of 0.026 minÀ1).20 Discrepancies between the
estimated rate constants are likely due to the higher
power (530 mW exiting the cuvette) and different optics
utilised by Furuta et al.20 Exposure of Bhc-1400W to a
non-mode-locked infrared beam exiting the Ti/Sapphire
laser for 1 h did not result in fluorescence or uncaging,
verifying that the multiphoton effect is required to pho-
toactivate Bhc-1400W. The in vitro enzyme assay was
used to ensure that MP photolysis of Bhc-1400W pro-
duced the desired biological effect.
The type of inhibitor interaction, if any, occurring
between Bhc-1400W and the heme iron of NOS could
not be determined using the substrate perturbation
spectroscopy method described for 1400W due to the
strong absorbance maximum of coumarin at 370 nm.
Although the coumarin is coupled via the amine of
1400W so that the amidine remains free to associate
with the guanidine-binding pocket of the enzyme, it is
unknown whether Bhc-1400W still binds in the vicinity
of the heme prior to electron transfer.
IC50 values of 1400W and Bhc-1400W
IC50 curves were obtained by normalising the initial
enzymatic rates of inhibitor titrations with the un-
inhibited enzyme’s initial rate. All analyses were carried
out in quadruplicate in 96-well plates. To simplify the
comparison of regression results, the maximal and
minimal reaction rates were normalized to 1 and 0,
respectively, to reduce the four-parameter to the two-
parameter logistic eq 1. This simplification did not sig-
nificantly affect the calculated parameters and resulted
in a good fit to experimental data (Fig. 3). Results for
1400W, Bhc-1400W, UV uncaged 1400W, and MP
uncaged 1400W were fit by nonlinear regression (Statis-
tics Toolbox, MATLAB version 5.3, Mathworks, Bos-
ton, MA) to eq 1:
Photolysis of Bhc-1400W
The uncaging kinetics of Bhc-1400W by one-photon
excitation was determined by irradiating Bhc-1400W
with 366 nm UV light for given time periods. Fluores-
cence excitation at 340 nm and emission detection at
465 nm revealed that coumarin photo-bleaching was
complete in less than 15 min (data not shown). The
release of 1400W was characterized by apparent first
order kinetics with a time constant of 0.39Æ0.28 minÀ1
(R2=0.95). UV irradiation of Bhc-1400W resulted in a
86% reduction in iNOS activity after 5 min (Fig. 2).
TLC and ESI-MS were also used to monitor the unca-
ging of Bhc-1400W. NMR studies on the purified cou-
marin by-product revealed the coumarin ring to be
intact.
VI
1
ꢀ
ꢁ
¼
ð1Þ
n
VWT
½IŠ
IC50
1 þ
MP uncaging of Bhc-1400W was performed at 740 nm20
and the laser power exiting the cuvette was approxi-
mately 350 mW. MP irradiation of Bhc-1400W resulted
in a 60% reduction in iNOS activity after 80 min (Fig.
2). MP uncaging data was fit to a decaying exponential
yielding an apparent first order rate constant of
0.007Æ0.001 minÀ1. The uncaging of Bhc-1400W by
where VI is the initial rate of l-arginine consumption at
inhibitor concentration I (nM l-arginine minÀ1 mgÀ1
enzyme), VWT is the initial rate of l-arginine consump-
tion for inhibitor-free reactions (nM l-arginine minÀ1

1922
H. J. Montgomery et al. / Bioorg. Med. Chem. 10 (2002) 1919–1927
could not be obtained for the nNOS and eNOS
enzymes. Nonetheless the UV and MP uncaged Bhc-
1400W also maintained the iNOS isoform specificity of
1400W. The large ratio of the 1400W IC50 to the Bhc-
1400W IC50 value demonstrates the therapeutic poten-
tial of this caged inhibitor. For example, Figure 3 indi-
cates that a Bhc-1400W dose of 200 nM results in a 7%
reduction of iNOS activity, while MP uncaging of this
sample results in an 89% reduction of iNOS activity.
Time-dependent inhibition of iNOS by 1400W and Bhc-
1400W
1400W is a slow tight-binding inhibitor of iNOS.21 Time
dependent inhibition occurs when the attainment of
equilibrium between enzyme and inhibitor is slow rela-
tive to the rate of enzymatic turnover.27 Two potential
mechanisms for time-dependent inhibition are:28
Figure 3. IC50 curves of 1400W, Bhc-1400W, MP uncaged Bhc-
1400W, and UV uncaged Bhc-1400W. The IC50 values for 1400W (*,
70 nM), Bhc-1400W (-, – 1098 nM), MP uncaged Bhc-1400W (&, – –,
73 nM), and UV uncaged Bhc-1400W (~,- - -, 67 nM) were deter-
mined. Samples were pre-incubated for 15 min at 30 ꢀC in the presence
of 500 mM NADPH, 287 nM iNOS and varying concentrations of
inhibitor. Reactions were initiated by the addition of 10 mM l-argi-
nine. Error bars indicate 95% confidence intervals and are repre-
sentative of collected data.
k3
*
k4
E þ I ) EI
ðAÞ
ðBÞ
k3
k5
*
k6
ꢀ
*
k4
E þ I ) EI ) EI
mgÀ1 enzyme), [I] is the inhibitor concentration (nM),
and n is the logistic sensitivity (Hill coefficient) (Table 1).
In mechanism A, the time dependence of inhibition is a
result of the slow formation of the EI complex, whereas
in mechanism B it is due to the slow rate of a secondary
conformational change of the EI complex to an EI*
complex. Equilibrium concentrations of enzyme, inhi-
bitor, and the EI complex are related by the dissociation
constant, which is defined as:27
The IC50/[E] values, following 15 min of enzyme-inhibitor
pre-incubation, for 1400W were 22 and 382 times
greater for nNOS and eNOS, respectively, when com-
pared to iNOS. The caging of 1400W with Bhc pro-
duced a compound with significantly reduced potency.
The caged Bhc-1400W inhibited 50% of iNOS activity
at a concentration of 1.1Æ0.2 mM (Table 1)—15.7 times
the amount of 1400W required to obtain the same
degree of inhibition. Coupling of the Bhc molecule did
not affect isoform selectivity of 1400W as no nNOS or
eNOS inhibition by Bhc-1400W was observed (Table 1).
Uncaging of Bhc-1400W by either MP (150 min) or UV
(5 min) irradiation quantitatively released 1400W with
iNOS IC50 values of 73Æ11 and 67Æ26 nM, respec-
tively. There was no statistically significant variation in
the calculated Hill coefficients.
½EŠ½IŠ
Ki ¼
ð2Þ
½EIŠ
For mechanism B the overall dissociation constant for a
slow binding inhibitor is:
k6
Kꢀi ¼ Ki
ð3Þ
k5 þ k6
1400W is a competitive inhibitor of iNOS, and in the
presence of l-arginine, Ki is related to an apparent Ki by:
The calculated IC50 values for 1400W and the UV and
MP uncaged Bhc-1400W are in excellent agreement,
indicating that fully uncaged samples possess the same
potency for iNOS as native 1400W. Due to the limited
solubility of Bhc-1400W (ꢁ900 mM), IC50/[E] values
ꢀ
ꢁ
½SŠ
Kai pp ¼ Ki 1 þ
ð4Þ
KM
Table 1. Inhibition kinetics of iNOS, nNOS and eNOS by 1400W, Bhc-1400W and uncaged Bhc-1400W
Inhibitor
iNOS
nNOS
eNOS
IC50/[E]
n
IC50/[E]
n
IC50/[E]
n
1400W
Bhc-1400W
UV uncaged Bhc-1400W
MP uncaged Bhc-1400W
2.4Æ0.4
38Æ6
1.8Æ0.6
1.6Æ0.3
1.6Æ1.0
2.1Æ0.7
53Æ9
NAI
NCIS
NCIS
1.8Æ0.6
916Æ170
NAI
NAI
1.3Æ0.3
2.3Æ0.9
2.6Æ0.4
NAI
Enzyme concentrations were 28.7 nM for iNOS and nNOS, and 1170 nM for eNOS.
NAI, no apparent inhibition.
NCIS, no change in isoform specificity.
95% Confidence intervals are shown.
n, logistic sensitivity (Hill coefficient) from eq 1.

H. J. Montgomery et al. / Bioorg. Med. Chem. 10 (2002) 1919–1927
1923
ꢀ
ꢁ
Time-dependent inhibition of iNOS by free and caged
inhibitor was studied by the addition of iNOS to reac-
tion solutions containing l-arginine and 1400W or Bhc-
1400W. The resulting progress curve data was fit to eq 5
by nonlinear regression analysis:28,29
k5½IŠ
kobs ¼ k6 þ
ð6Þ
Kai pp þ ½IŠ
The kobs versus inhibitor concentration data for 1400W
and Bhc-1400W are shown in Figure 4B. For 1400W,
A401ðtÞ ꢁ A401ðt ¼ 0Þ
values of Kiapp of 30Æ7.3 mM, k5 of (3.5Æ0.5)Â10À2 sÀ1
,
and a maximum k6 value of 1.9Â10À6 sÀ1 characterizing
slow tight binding inhibition of iNOS were obtained by
nonlinear regression of eq 6 to the 1400W data (Fig.
4B). Accurate kobs values could not be obtained at
higher 1400W concentrations due to rapid inhibition
kinetics. This problem has previously been encountered
by others.21 Substituting a Kai pp of 30Æ7.3 mM, a [S] of
20 mM, and a KM of 9Æ0.8 mM into eq 4 results in a Ki
of 9.3Æ2.4 mM for 1400W. Substituting a Ki of 9.3Æ2.4
ðꢀ0 ꢁ ꢀsÞð1 ꢁ expðꢁkobstÞÞ
¼ ꢀst þ
ð5Þ
kobs
where A401(t) is the absorbance at 401 nm at time t (OD),
n0 is the initial velocity (OD/s), ns is the steady state
velocity (OD/s), and kobs is the apparent first order rate
constant for the conversion of n0 to ns (sÀ1).28 Figure 4A
shows typical progress curves generated in the presence
of 1400W and Bhc-1400W. Experimental data were fit
to eq 5 while the uninhibited control reaction was linear.
For all experiments the inhibitor concentration was
appreciably greater than the total enzyme concentra-
tion.28 iNOS, l-arginine, and oxyhemoglobin con-
centrations of 10 nM, 20 mM, and 15 mM, respectively,
were selected to allow sufficient time to observe the
curvilinear nature of the inhibited progress curves and
maintain a linear OD versus methemoglobin production
relation with an adequate signal to noise ratio.
mM, a k6 max of 1.9Æ10À6
s
À1, and a k5 of (3.5Æ0.5)
Â10À2 sÀ1 into eq 3 results in a Ki*max of 508Æ152 pM
for 1400W. The kinetic parameters calculated char-
acterizing the slow tight binding inhibition of iNOS by
1400W are in good agreement with the values of 2Æ0.5
mM for Ki, 0.028Æ0.003 sÀ1 for k5, and (7.3Æ0.6)Â10À5
s
À1 for k6max reported by Garvey and co-workers.21
Bhc-1400W is a slow binding inhibitor of iNOS when
examined under the given experimental conditions; the
Bhc-1400W IC50 value, after 15 min of pre-incubation
with iNOS, is one order of magnitude greater than the
total iNOS concentration.27 Thus the coumarin moiety
substantially reduces the tight binding of 1400W to
iNOS. The linear kobs versus Bhc-1400W relation shown
in Figure 4B is likely due to the degeneration of eq 6
into a linear relation when [I] < <Kai pp as indicated by
eqs 7 and 8:
The inhibition kinetics of 1400W conformed to
mechanism B, agreeing with the results of Garvey and
co-workers,21 and fit to the following rectangular
hyperbolic relation:28,29
ꢀ
ꢁ
k5½IŠ
kobs ¼ k6 þ
¼ k6 þ kaopnp½IŠ
ð7Þ
ð8Þ
Kai pp
ꢀ
ꢁ
½SŠ
k5
Ki
kon ¼ kaopnp 1 þ
¼
KM
app
on
where k
(mMÀ1 À1) is the slope of the kobs versus [I]
s
plot. An alternate, but unlikely, explanation for the lin-
ear kobs versus [Bhc-1400W] relation is that coupling of
the Bhc moiety to 1400W alters the mode of inhibition
from mechanism B to A. For Bhc-1400W, a k6 value of
(2.1Æ0.7)Â10À4 sÀ1 and a kon value of (3.3Æ0.3)Â10À4
s
À1 characterizes the slow binding inhibition of iNOS as
obtained by linear regression of the Bhc-1400W data
(Fig. 4B).
PC12 cell viability in the presence of 1400W and Bhc-
1400W
Figure 4. Inhibition kinetics of iNOS with 1400W and Bhc-1400W.
Panel A, iNOS activity for representative 1400W (*, kobs=6.0Â10À3
s
À1) and Bhc-1400W (&, kobs=1.4Â10À3
s
À1) at concentrations of 6
The viability of PC12 cells incubated with 1400W, Bhc-
1400W and 7-amino-4-methly-coumarin for 48 h was
determined by exclusion of erythrosine B dye. The
highly soluble 1400W was toxic at concentrations above
5 mM and gave an ED50 value of 1 mM. In the presence
mM and 12 mM, respectively. Inhibitor-free iNOS activity (~) is also
shown. Panel B, kobs versus 1400W (*, k5=0.035 sÀ1, k6 <1.9Â10À6
s
À1, Ki=9.3 mM, maximum Ki*=508 pM) and Bhc-1400W (&,
koapnp=3.3Â10À4 mM-1
s
À1, k6=2.1Â10À4
s
À1) concentrations. Error
bars indicate Æ1 standard deviation.

1924
H. J. Montgomery et al. / Bioorg. Med. Chem. 10 (2002) 1919–1927
of 1 mM Bhc-1400W the viability of the PC12 cells was
39%. Lowering the concentration of Bhc-1400W 10-fold
to 100 mM increased the viability of the PC12 cells to
80%. Only a low concentration of 7-amino-4-methly-
coumarin (27 mM) was soluble in the 25 mM HEPES/
McCoy’s media used. This concentration of 7-amino-
4-methly-coumarin was non-toxic to PC12 cells.
This report provides the first example of a caged NOS
inhibitor that undergoes multiphoton photoactivation.
A similar approach can be used for a variety of appli-
cations including photodynamic therapies and non-
invasive mapping of cellular responses to uncaged
molecules. The synthesis of a drug covalently bound to
a fluorophore with a large two-photon cross section is a
novel class of compounds with the potential to be
photoactivated allowing for the spatial and temporal
control of drug activity in tissue. This approach should
greatly expand the available range of therapeutics that
may otherwise not be used because of side effects in
non-target tissues.
Discussion
N-(3-(Aminomethyl)benzyl)acetamidine (1400W) was
selected for investigation because it has the highest
reported potency and selectivity for iNOS inhibition.21
The inhibition of eNOS and nNOS by 1400W is rapidly
reversible and inefficient. The extremely slow off-rate of
1400W bound to iNOS makes it a potent inhibitor. The
caging of 1400W did not affect isoform selectivity. A
large decrease in the level of iNOS inhibition by the
caged compound yields a wide therapeutic range over
which the full action of the inhibitor can be exploited.
The poor inhibitory properties of Bhc-1400W are likely
due to the loss of a positive charge and steric effects by
conversion of the amine to a carbamate.
Experimental
Materials
All reagents were purchased from Sigma-Aldrich
Canada, Ltd. (Oakville, ON, Canada) and used without
further purification. Electrospray ionization-mass spec-
trometry (ESI-MS) was performed in positive ion mode
on a Micromass Quattro II triple quadrupole mass
spectrometer equipped with an electrospray source
(Micromass, Manchester, UK). Samples were intro-
duced by flow injection analysis using acetonitrile/water
(1:1) with 0.1% formic acid as solvent. Cone voltage
was set to 10 V for all samples. High-resolution mass
spectrometry was performed using a Quadrupole Time-
Of-Flight (QTOF) (Micromass, Manchester, UK) using
an appropriate internal standard.
Coupling Bhc to 1400W may significantly increase the
off-rate (k6) of the inhibitor when bound to iNOS,
resulting in a substantial reduction in potency. This is
consistent with the observation that the Bhc-1400W and
1400W show poor inhibition of the constitutive NOS
isoforms. The calculated k6 value for Bhc-1400W is
approximately 100-fold greater than that of the max-
imum 1400W value. The Ki value for Bhc-1400W is also
substantially larger than that of 1400W, as no curvature
for the Bhc-1400W data is apparent in Figure 4B. The
reduction in potency of iNOS inhibition by Bhc-1400W,
as compared to 1400W, is a result of higher k6 and Ki
values and a reduced k5 value. These effects would sub-
stantially increase the overall slow binding inhibitor
dissociation constant (Ki*) for Bhc-1400W. This result
is supported by the 15-fold increase in the IC50 value of
Bhc-1400W compared to 1400W as shown in Table 1.
Synthesis of the free and caged forms of 1400W
N-(3-(Aminomethyl)benzyl)acetamidine dihydrochloride
(1400W) 1.22 A solution of m-xylyenediamine (5 g,
36.71 mmol) and triethylamine (2.55 g, 18.34 mmol) in
dry methanol (100 mL) was cooled to 0 ꢀC. To this
solution, di-t-butyldicarbonate (4 g, 18.34 mmol) in
THF (30 mL) was added drop-wise over 1 h. The reac-
tion was stirred an additional 2 h at 0 ꢀC under
argon, and then filtered. The filtrate was con-
centrated, then purified by silica column chromato-
graphy by elution with 5:95:0.5 MeOH/CH2Cl2/
NH4OH. t-Butyl-N-(3-(aminomethyl)benzyl) carbamate
(2.44 g, 28% yield) was obtained as a viscous yellow oil.
t-Butyl-N-(3-(aminomethyl)benzyl) carbamate (2 g, 8.4
mmol) was dissolved in ethanol (25 mL) at 0 ꢀC under
argon. S-2-Naphthylmethyl thioacetimidate hydro-
bromide32 (2.5 g, 8.4 mmol) was added, and the cloudy
mixture was stirred for 4 h as it warmed to room tem-
perature. The mixture was concentrated in vacuo and
the resultant off-white solid was taken up in 1:1 H2O/
Et2O (150 mL), and washed three times with Et2O (50
mL). The aqueous fraction was lyophilized to provide
t-butyl-N-(3-((acetamidoyl)-aminomethyl)benzyl carba-
mate hydrobromide (N-BOC-1400W) (1.97 g, 65%
yield) as a white waxy solid. N-BOC-1400W (1 g, 2.8
mmol) was dissolved in anhydrous dioxane (20 mL) and
4 N HCl in dioxane (5 mL) under argon and stirred for
24 h at room temperature. The reaction mixture was
diluted with Et2O (30 mL) and the product collected by
MP irradiation has been recently employed for the
release of caged compounds in biological samples.8,19,30
The spatially confined release of caged molecules is
possible because uncaging occurs only in the focal
volume—less than 1 mm3 of sample. With IR irradia-
tion, molecules can be uncaged without tissue damage
at significantly greater depths than possible with UV
irradiation.
Over the past decade MP microscopy has progressed
from a novel technology to an important tool in medical
research. Neural imaging of freely moving rats by
means of a head mounted MP microscope and a single-
mode optical fiber has recently been demonstrated.31
A
similar device could be utilized to investigate photo-
activation of a caged therapeutic in the cortex following
surgically induced focal ischemia. Continued advance-
ments in MP microscopy will enable researchers to
explore the use of this technology in an increasing
number of therapeutic applications.

H. J. Montgomery et al. / Bioorg. Med. Chem. 10 (2002) 1919–1927
1925
filtration and dried under vacuum with gentle heating to
1
Multiphoton uncaging was performed by placing a
micro-cuvette in the focal plane of an objective lens
(Axoplan 10Â/0.25 Ph 1) located at the exit aperture of
a femtosecond mode-locked Ti/Sapphire laser (Mira
Model 900-B laser, Coherent, Inc., Santa Clara, CA,
USA) (76 MHz repetition rate, ꢂ200 fs pulse width).
The Mira laser was pumped with 5 W at 532 nm
(Coherent Verdi laser, Coherent, Inc., Santa Clara, CA,
USA). Ten-mL aliquots of Bhc-1400W were used to
obtain blank and multiphoton uncaged samples at var-
ious times. Uncaging was performed at 740 nm.20 The
laser power exiting the cuvette was 350 mW. Dark
uncaging of Bhc-1400W was monitored with a sample
wrapped in foil and kept at room temperature. The
amount of uncaging that occurred in the samples was
determined by the kinetic assay described below.
yield 1 as off-white flakes (0.68 g, 96% yield). H NMR
(300 MHz, D2O) d 7.27–7.40 (4H, m), 4.4 (2H, s), 4.06
(2H, s), 2.15 (3H, s); 13C NMR (75 MHz, D2O) d 165.0,
135.2, 133.4, 129.9, 128.7, 128.4, 127.9, 45.5, 42.9, 18.5.
ES-MS (MH+) 177.9.
6 - Bromo - 7 - hydroxy - 4 - hydroxymethylcoumarin 2.20
4-Bromoresorcinol (922.5 mg, 4.881 mmol) was stirred
in concentrated H2SO4 (5 mL) with ethyl 4-chloro-
acetoacetate (1 mL, 7.4 mmol) for 6 days at room tem-
perature. This mixture was poured into 50 mL of ice
cold H2O and stirred vigorously for 2 h. The precipitate
was filtered, washed with ice-cold H2O, and dried over
P2O5, to provide the brown, granular 6-bromo-
7-hydroxy-4-chloromethyl-coumarin (0.547 g, 39%
yield). This chloride (450 mg, 1.55 mmol) was sus-
pended in 75 mL H2O and refluxed for 4 days, then
lyophilized to provide 6 as a brown powder (370 mg, 88%
Protein expression and purification
1
yield). H NMR (250MHz, CD3OD) d 7.80 (1H, s), 6.83
Human iNOS enzyme carrying a deletion of the first 70
amino acids and an amino terminal polyhistidine tail
was co-expressed with calmodulin in Escherichia coli
and purified using metal chelating chromatography fol-
lowed by 20,50-ADP column chromatography. The
expression and protein purification of human eNOS and
rat nNOS enzymes in E. coli was performed as pre-
viously reported.33
(1H, s), 6.37 (1H, s), 4.76 (2H, s); 13C NMR (75MHz,
(CD3)2CO) d 160.1, 156.9, 155.1, 154.4, 128.2, 111.9, 108.6,
105.8, 103.4, 59.7. ES-MS (MH+) 270.7, 272.7 (1:1).
N-(6-Bromo-7-hydroxycoumarin-4-yl)methoxy-carbonyl-
N-(3-(aminomethyl)benzyl)acetamidine (Bhc-1400W) 3.
2 was suspended in dry acetonitrile (20 mL), to which
was added 4-nitrophenylchloroformate (111 mg, 1.2 mol
equiv) and DMAP (112 mg, 2 mol equiv). This was
stirred for 9 h under argon, after which time a solution
of 1 (114 mg, 0.99 mol equiv) and DMAP (112 mg, 2
mol equiv) in DMF (10 mL) was added. After 24 h of
stirring under argon in the dark, the reaction mixture
was concentrated in vacuo to a dark yellow oil. This was
suspended in 1:1 CHCl3/H2O (50 mL), and washed
three times with CHCl3 (25 mL). The aqueous fraction
(containing the product 3) was lyophilized, then 3 was
purified by silica chromatography eluting with 7:2.5:0.5
CHCl3/MeOH/NH4OH (Rf 0.23). The caged compound
3 (40 mg, 20% yield) was obtained as a pale-yellow
Enzyme kinetics
The activity of NOS was determined using the hemo-
globin capture assay23,24 that yields results consistent
with the direct radioactive assay that monitors the for-
mation of l-[3H]-citrulline from l-[3H]-arginine.32
Quadruplicate reactions were initiated by the addition
of l-arginine and monitored on a 96-well plate reader
(Spectramax 190, Molecular Devices, Sunnyvale, CA,
USA) at 30 ꢀC. Nitric oxide-mediated oxidation of
oxyhemoglobin was monitored at 401 nm (e=0.1036
OD/nanomol).23 The concentration of Bhc-1400W in
the wells, as well as for UV and MP uncaged samples,
was 178 nM. For the determination of IC50 values, 100
mL reaction mixtures contained 50 mM Tris–HCl (pH
7.5), 500 mM NADPH, 5.0 mM H4B, 1.0 mM CaM, 1.0
mM CaCl2, 1.0 mM DTT, 1.0 mM FAD, 1.0 mM FMN,
100 units/mL superoxide dismutase (SOD), 50 units/mL
catalase, 5.0 mM bovine oxyhemoglobin, 0.2 mg/mL
bovine serum albumin, 10 mM l-arginine and 28.7 nM
iNOS (Vmax=376Æ105 nM minÀ1 mgÀ1 iNOS, KM
(l-ARG=9Æ0.8 mM), determined over a l-arginine
concentration of 5–25 mM). Enzyme in the presence or
absence of varying amounts of 1400W, Bhc-1400W, or
uncaged Bhc-1400W, and in the presence of 500 mM
NADPH, 10.0 mM FAD, 10.0 mM FMN, 50.0 mM H4B,
10.0 mM DTT, 100 units/mL SOD and 50 units/mL
catalase were incubated at 30 ꢀC for 15 min.21 For these
experiments the Bhc-1400W was exposed to UV light
for 5 min or to MP irradiation for 150 min to obtain the
uncaged samples.
1
powder. H NMR (300 MHz, CD3OD) d 7.52 (1H, s),
7.17–7.47 (4H, m), 6.08 (1H, s), 5.63 (1H, s), 5.11 (2H,
s), 4.38 (2H, s), 4.20 (2H, s), 2.16 (3H, s); 13C NMR
(75 MHz, CD3OD) d 170.9, 166.1, 165.0, 158.2, 157.4,
153.8 141.6, 136.0, 130.3, 128.4, 127.9, 127.6, 115.1,
106.6, 106.0, 104.0, 63.2, 47.3, 45.2, 18.4. ES-MS
(MH+) 473.78, 475.71 (1:1). HR-QTOF MH+
(C21H20BrN3O5) expected 474.0665, found 474.0652.
Fmoc-Tyr(OtBu)(MH+ 460.2124) used as internal
standard.
Uncaging Bhc-1400W
Dried Bhc-1400W was dissolved in 250 to 1000 mL of
100 mM KCl/10 mM Mops buffer (pH 7.2) (KMops) by
shaking the solution for 30 min at 4 ꢀC in the dark. The
concentration in solution was measured using UV–vis
spectrophotometry (lmax=368 nm, e=17,470 MÀ1
cmÀ1).20 UV light at 366 nm was used to uncage Bhc-
1400W. Ten 10-mL aliquots were taken after various
exposure times and the amount of uncaging in each
sample was determined by the kinetic assay described
below.
Kinetic parameters characterizing slow binding inhibi-
tion of iNOS by 1400W and Bhc-1400W were calculated
from progress curve data obtained using the hemoglobin

1926
H. J. Montgomery et al. / Bioorg. Med. Chem. 10 (2002) 1919–1927
assay.34 Reaction conditions were similar to those given
above except that 15 mM bovine oxyhemoglobin, 20 mM
l-arginine and 10 nM NOS were used. Under these
conditions inhibitor-free reactions were linear for
approximately 10 min and did not exceed the linear
range of the spectrophotometer. Blank well data was
subtracted from progress curve data. Blank wells con-
taining only 200 mM 1400W and iNOS produced similar
values to wells containing no iNOS. The addition of
Bhc-1400W did not significantly contribute to back-
ground signal at 401 nm at the concentrations tested.
The removal of FMN, FAD, DTT, and H4B from the
reaction mixture did not reduce iNOS activity, and
improved signal to noise ratios. Reactions were initiated
by the addition of iNOS.
of 1Â105 cells/mL. In all studies, the medium was
changed to 25 mM HEPES/supplemented McCoy’s
media before plating. The viability of the cells was
ꢃ90% as determined by exclusion of the dye ery-
throsine B. 1400W (500 to 5Â10À5 mM), 7-amino-
4-methly-coumarin (29 uM) and Bhc-1400W (1 to
1Â10À9 mM) were taken up into buffered McCoy’s
media, filter sterilized and added to the cells in decreas-
ing concentrations. Cells were incubated at 37 ꢀC for 48
h. The viability of the cells in the presence of these
compounds was determined as described above.
Acknowledgements
This research was supported by a grant from the Nat-
ural Sciences and Engineering Research Council of
Canada and the Heart and Stroke Foundation. We
would like to thank Dr. E. Meiering for the use of the
spectrofluorometer, Dr. J. Lepock for use of the mam-
malian cell culture facilities, and Elizabeth Jones for
technical assistance.
Optical difference spectroscopy
Optical spectra were recorded on a Cary 1 Bio UV–vis
spectrophotometer (Varian Instruments, Sugar Land,
TX, USA), with a Peltier 1Â1 temperature control
accessory (Varian Instruments, Sugar Land, TX, USA).
Sample and reference cuvettes initially contained 750 mL
of 50 mM Tris–HCl, pH 7.5, 10% glycerol, and 3 mM
iNOS at 15 C. The base line spectrum was adjusted to
zero. l-Arginine or 1400W were titrated into sample
cuvettes, while buffer was used to adjust the volume of
the reference cuvette. Difference spectra were recorded
after each addition from 350 to 500 nm at 0.2-nm
intervals using a 2-nm slit width.25 Final volume chan-
ges were less than 2%. The x-intercept of double-reci-
procal plots of the difference in the peak to trough
absorbances versus the concentration of l-arginine or
1400W were used to determine the spectral binding
constants (Ks).
References and Notes
1. Moncada, S.; Higgs, A. N. Engl. J. Med. 1993, 329, 2002.
2. Kleinert, H.; Boissel, J. P.; Schwarz, P. M.; Forestmann, U.
In Nitric Oxide, Biology and Pathobiology; Ignarro, L. J., Ed.;
Academic: New York, 2000; p 105.
3. Dawson, T. D.; Dawson, V. L. Methods Enzymol. 1996,
268, 349.
4. Bredt, D. S.; Snyder, S. H. Annu. Rev. Biochem. 1994, 63,
175.
5. Marletta, M. A. J. Biol. Chem. 1993, 268, 12231.
6. Estrada, C.; Gomez, C.; Martin, C.; Moncada, S.; Gonza-
lez, C. Biochem. Biophys. Res. Commun. 1992, 186, 475.
7. Iadecola, C. Trends Neurosci. 1997, 20, 132.
8. Lutz, P. L.; Nilsson, G. E. The Brain Without Oxygen, 2nd
ed.; Chapman and Hall: New York, 1997.
9. Zhang, Z. G.; Reif, D.; Macdonald, J.; Tang, W. X.;
Kamp, D. K.; Gentile, R. J.; Shakespeare, W. C.; Murray,
R. J.; Chopp, M. J. Cereb. Blood Flow Metab. 1996, 16, 599.
10. Huang, Z.; Huang, P. L.; Panahian, N.; Dalkara, T.;
Fishman, M. C.; Moskowitz, M. A. Science 1994, 265, 1883.
11. Hara, H.; Huang, P. L.; Panahian, N.; Fishman, M. C.;
Moskowitz, M. A. J. Cereb. Blood Flow Metab. 1996, 16,
605.
12. Iadecola, C.; Zhang, F. Y.; Casey, R.; Nagayama, M.;
Ross, M. E. J. Neurosci. 1997, 17, 9157.
13. Park, C. S.; Park, R.; Krishna, G. Life Sci. 1996, 59, 219.
14. Denk, W.; Strickler, J. H.; Webb, W. Science 1990, 248, 73.
15. Williams, R. M.; Piston, D. W.; Webb, W. FASEB J.
1994, 8, 804.
Fluorescence spectroscopy
Fluorescence analysis of 1400W, Bhc-1400W and
uncaged Bhc-1400W was performed on a Spex Fluoro-
log-22 spectrofluorometer (John Yvon Horiba Group,
Edison, NJ, USA). An excitation maximum of 340 nm
and an emission maximum of 465 nm were determined
for Bhc-1400W. No fluorescence was observed for
1400W over these wavelengths. For the analysis of
uncaging by UV light, 10-mL aliquots were taken every
15–20 min over a 180-min time period. Samples were
diluted in deionised, double distilled, filtered water to a
final concentration of 8.3 mM and their fluorescence
measured.
Viability of PC12 cells in the presence of inhibitor
16. Denk, W. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 6629.
17. Adams, S. R.; Lev-Ram, V.; Tsien, R. Chem. Biol. 1997,
867, 1256.
18. Reinhardt, B. A.; Brott, L. L.; Clarson, S. J.; Dillard,
A. G.; Bhatt, J. C.; Kannan, R.; Yuan, L.; He, G. S.; Prasad,
P. N. Chem. Mater. 1998, 10, 1863.
19. Albota, M.; Beljonne, D.; Bredas, J. L.; Ehrlich, J. E.; Fu,
J. Y.; Heikal, A. A.; Hess, S. E.; Kogej, T.; Levin, M. D.;
Marder, S. R.; McCord-Maughon, D.; Perry, J. W.; Rockel,
H.; Rumi, M.; Subramaniam, G.; Webb, W. W.; Wu, X. L.;
Xu, C. Science 1998, 281, 1653.
The rat pheochromocytoma cell line PC12 was pur-
chased from American Type Culture Collection
(CATCC) and maintained in McCoy’s media supple-
mented with 10% heat-inactivated horse serum and 5%
heat-inactivated bovine calf serum at 37 ꢀC, in a humi-
dified atmosphere of 5% CO2. Cells were passaged
every 3 days. The cells were harvested, triturated with a
18-gauge needle, concentrated to 5Â105 cells/mL and
then plated in 96-well cell culture chambers (VWR
Canlab, Mississauga, ON, Canada) at a concentration

H. J. Montgomery et al. / Bioorg. Med. Chem. 10 (2002) 1919–1927
1927
20. Furuta, T.; Wang, S. S.-H.; Dantzker, J. L.; Dore, T. M.;
Bybee, W. J.; Callaway, E. M.; Denk, W.; Tsien, R. Y. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 1193.
27. Copeland, R. A. Enzymes: A Practical Introduction to
Structure, Mechanism and Data Analysis; Wiley-VCH: New
York, 1996.
21. Garvey, E. P.; Oplinger, J. A.; Furfine, E. S.; Kiff, R. J.;
Laszio, F.; Whittle, B. J. R.; Knowles, R. G. J. Biol. Chem.
1997, 272, 4959.
22. Oplinger, J. A.; Garvey, E. P.; Furfine, E. S.; Shearer, B.
G.; Collins, J. L. US Patent 5 866 612, 1999.
23. Gross, S. S. Methods Enzymol. 1996, 267, 159.
24. Hevel, J. M.; Marletta, M. A. Methods Enzymol. 1994,
233, 250.
25. McMillan, K.; Masters, B. S. S. Biochemistry 1993, 32,
9875.
26. Li, H.; Raman, C. S.; Martasek, P.; Masters, B. S. S.;
Poulos, T. L. Biochemistry 2001, 40, 5399.
28. Morrison, J. F.; Walsh, C. T. Adv. Enzymol. Relat. Areas
Mol. Biol. 1988, 61, 201.
29. Szedlacsek, S. E.; Duggleby, R. G. Methods Enzymol.
1995, 249, 144.
30. Brown, E. B.; Webb, W. Methods Enzymol. 1998, 291, 356.
31. Helmchen, F.; Fee, M. S.; Tank, D. W.; Denk, W. Neuron
2001, 31, 903.
32. Shearer, B. G.; Oplinger, J. A.; Lee, S. Tetrahedron Lett.
1997, 38, 179.
33. Montgomery, H. J.; Romanov, V.; Guillemette, J. G.
J. Biol. Chem. 2000, 275, 5052.
34. Dawson, J.; Knowles, R. G. Mol. Biotechnol 1999, 12, 275.