Inhibitory effects of a series of 7-substituted-indazoles toward nitric oxide synthases: Particular potency of 1H-indazole-7-carbonitrile

Article abstract of DOI:10.1016/j.bmc.2008.04.056

A series of new 7-monosubstituted and 3,7-disubstituted indazoles have been prepared and evaluated as inhibitors of nitric oxide synthases (NOS). 1H-Indazole-7-carbonitrile (6) was found equipotent to 7-nitro-1H-indazole (1) and demonstrated preference for constitutive NOS over inducible NOS. By contrast, 1H-indazole-7-carboxamide (8) was slightly less potent but demonstrated a surprising selectivity for the neuronal NOS. Further substitution of 6 by a Br-atom at carbon-3 of the heterocycle enhanced 10-fold the inhibitory effects. Inhibition of NO formation by 6 appeared to be competitive versus both substrate and the cofactor (6R)-5,6,7,8-tetrahydro-l-biopterin (H₄B). In close analogies with 1, compound 6 strongly inhibited the NADPH oxidase activity of nNOS and induced a spin state transition of the heme-Fe^III. Our results are explained with the help of the X-ray structures that identified key-features for binding of 1 at the active site of NOS.

Full text of DOI:10.1016/j.bmc.2008.04.056

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 5962–5973
Inhibitory effects of a series of 7-substituted-indazoles
toward nitric oxide synthases: Particular potency
of 1H-indazole-7-carbonitrile
Betty Cottyn,a Francine Acher,b Booma Ramassamy,b Luke Alvey,a Michel Lepoivre,c
Yves Frapart,b Dennis Stuehr,d Daniel Mansuy,b Jean-Luc Boucherb and
Dominique Vicharda,*
aInstitut Lavoisier, CNRS UMR 8180, Universite´ de Versailles, 45 Avenue des Etats Unis, 78035 Versailles cedex, France
bLaboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques, CNRS UMR 8601, Universite´ Paris Descartes,
`
45 rue des Saints-Peres, 75270 Paris, France
cCNRS UMR 8619, IBBMC, Universite´ Paris Sud-Orsay, Bat 430, 91105 Orsay, France
dDepartment of Immunology, Lerner Research Fundation, Cleveland Clinic, Cleveland, OH 44195, USA
Received 25 January 2008; revised 15 April 2008; accepted 23 April 2008
Available online 26 April 2008
Abstract—A series of new 7-monosubstituted and 3,7-disubstituted indazoles have been prepared and evaluated as inhibitors of
nitric oxide synthases (NOS). 1H-Indazole-7-carbonitrile (6) was found equipotent to 7-nitro-1H-indazole (1) and demonstrated
preference for constitutive NOS over inducible NOS. By contrast, 1H-indazole-7-carboxamide (8) was slightly less potent but dem-
onstrated a surprising selectivity for the neuronal NOS. Further substitution of 6 by a Br-atom at carbon-3 of the heterocycle
enhanced 10-fold the inhibitory effects. Inhibition of NO formation by 6 appeared to be competitive versus both substrate and
the cofactor (6R)-5,6,7,8-tetrahydro-L-biopterin (H4B). In close analogies with 1, compound 6 strongly inhibited the NADPH oxi-
dase activity of nNOS and induced a spin state transition of the heme–FeIII. Our results are explained with the help of the X-ray
structures that identified key-features for binding of 1 at the active site of NOS.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
aggregation, and leukocyte adhesion on the endothelial
surface. In the central nervous system and in the periph-
eral non-adrenergic non-cholinergic nerves, NO is a po-
Nitric oxide (NO) is an endogenously produced free rad-
ical, which has important biological functions in mam-
mals.1–3 NO acts as a second messenger molecule
through the activation of its main target, soluble guany-
late cyclase. In the cardiovascular system, it is impli-
cated in the regulation of vascular tone, platelet
tent
neurotransmitter
involved
in
long-term
potentiation, migraine, and gastric motility. Elsewhere,
NO is a cytotoxic and cytostatic agent associated with
phagocytic cells in the immune system.1–3
NO is synthesized by oxidation of the amino acid L-argi-
nine (L-arg) catalyzed by three distinct isoforms of heme
proteins called NO synthases (NOS).4–6 The neuronal
(nNOS) and endothelial NOS (eNOS) are constitutively
expressed and are Ca++- and calmodulin (CaM)-depen-
dent enzymes, whereas inducible NOS (iNOS) is ex-
pressed in response to an immune challenge.4–6 All
three isoforms contain an N-terminal domain (oxygen-
ase domain) with binding sites for substrate, heme
(FeIII-protoporphyrin IX), and cofactor (6R)-5,6,7,8-tet-
rahydrobiopterin (H4B), and a linked C-terminal do-
main (reductase domain) containing NADPH, FAD,
Abbreviations: L-arg, L-arginine; CaM, calmodulin; DTT, dithiothre-
itol; H4B, (6R)-5,6,7,8-tetrahydro-L-biopterin; Hepes, N-(2-hydroxy-
ethyl)piperazine-N0-2-ethane sulfonic acid; HS, high spin; ImH,
imidazole; LS, low spin; NOS, nitric oxide synthase; n-, i- and eNOS,
neuronal, inducible and endothelial NOS, respectively; nNOSoxy, ox-
ygenase domain of neuronal NOS; NO2-arg, Nx-nitro-L-arginine;
SOD, superoxide dismutase.
Keywords: 7-Nitroindazole; Substituted indazoles; Nitric oxide syn-
thase; Inhibitors.
*
Corresponding author. Tel.: +33 1 39 25 44 71; fax: +33 1 39 25 44 52;
e-mail: vichard@chimie.uvsq.fr
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.04.056

B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
5963
FMN, and CaM binding sites.4–6 The observed biologi-
cal activities of NO are closely linked to the NOS iso-
form involved in its synthesis and overproduction of
NO leads to toxic effects.1–3 Overproduction of NO by
iNOS results in severe deleterious effects in septic shock
and tissue degradation during chronic inflammation
while overproduction of NO by nNOS appears to be in-
volved in neurodegenerative disorders. Therefore, selec-
tive NOS inhibitors would be interesting therapeutic
agents for the treatment of the above-mentioned dis-
eases. However, to have therapeutic value, NOS inhibi-
tors must be selective of one of the three isoforms. In
particular, iNOS and nNOS inhibitors must not inter-
fere with eNOS that plays key roles in vasodilation
and blood pressure regulation.1–3
bond between one oxygen-atom of the nitro group of
1 and a residue of the binding site than between the
O-atom of the methoxy group and the same residue.25,26
Introduction of a Br-atom at the 3-position of 1, in 3-
bromo-7-nitro-1H-indazole 12, is known to strongly en-
hance its inhibitory effects.18 This has encouraged us to
synthesize some new 3-bromo-7-substituted indazole
derivatives. In addition to the nitro group, it has been
proposed that the N(1)-H group of 1 is also able to
establish an hydrogen bond with a residue of the active
site.27,28 In order to get new information about the
importance of this N(1)-H group, we have synthesized
N(1)-methyl analogues of 1 and 12 that should disrupt
this H-bond.
Several classes of reversible and irreversible NOS inhib-
itors have been described.7–9 Substrate analogues such
as Nx-nitro-L-arginine (NO2-arg), Nx-methyl-L-argi-
nine, thio-L-citrulline, NO2-arg containing dipeptides,
and heterocyclic compounds are widely used as NOS
inhibitors in in vitro studies.7–15 Even though highly in
vitro isoform-selective inhibitors are now available, only
a few compounds have shown to be selective toward
nNOS in vivo. Moore and co-workers first reported
the potent inhibitory effects of the heterocyclic com-
pound 7-nitro-1H-indazole, 1.16,17 This compound
exhibits a clear anti-nociceptive activity without altering
blood pressure in vivo and is extensively used as a selec-
tive inhibitor of nNOS that evidences a 10-fold selectiv-
ity for nNOS over iNOS.16–18 However, 1 inhibits nNOS
by competing with both L-arg and H4B and it is note-
worthy that, despite its selectivity for nNOS in vivo, 1
inhibits the three NOS isoforms to a similar extent under
in vitro conditions.12,13
This series of indazole derivatives was evaluated as
inhibitors of purified recombinant n-, i,- and eNOS.
Two of them, 1H-indazole-7-carbonitrile 6, and its 3-
bromo derivative, 3-bromo-1H-indazole-7-carbonitrile
14, were found to be potent inhibitors for nNOS and
slightly more isoform-selective than 1 and its 3-bromo
derivative 12.
2. Chemistry
Our first target molecule was 1H-indazole-7-carbonitrile
6. Moore and co-workers have suggested that this com-
pound could be a NOS inhibitor but its synthesis had
not been reported when we started this work.17 In order
to design a general and efficient approach that would al-
low an easy preparation of 6 and new 7-monosubsti-
tuted indazole derivatives, we first envisioned that they
could be practically obtained via a palladium-mediated
cyanation reaction from 1H-indazol-7-yl trifluorometh-
anesulfonate 3. New triflate 3 was obtained in an almost
quantitative yield by trifluoromethanesulfonation of
1H-indazol-7-ol 2 by N-(2-pyridyl)triflimide29 in the
presence of Cs2CO3 and K2CO3 without competitive
N-triflation. However, triflate 3 failed to participate in
the palladium-catalyzed cyanation and was recovered
unchanged. We found an efficient alternative by prepar-
ing 7-iodo-1H-indazole 5 using a diazotization/iodin-
ation sequence starting from 1H-indazol-7-amine 4
(Scheme 1). Amine 4 was quantitatively obtained by
Furthermore, potential future uses of 1 could be limited
by its nitroaromatic nature. Actually, nitroaromatic
compounds often lead to toxic effects in vivo that are
linked either to their ability to start futile cycles of O2
reduction leading to superoxide anion and related spe-
cies, or to their metabolic reduction into reactive electro-
philic metabolites such as aromatic hydroxylamines or
nitroso compounds.19 Thus, secondary toxic effects of
nitrofurantoin or nitroimidazole drugs have been related
to reactive metabolite formation upon reductive metab-
olism of the nitro-function of these drugs.20
In order to avoid such possible toxic effects of 1, we have
synthesized a series of new 7-substituted indazoles bear-
ing an electron-withdrawing substituent, distinct from a
nitro group. The choice of keeping an electron-with-
drawing substituent on position 7 was made on the basis
of previous studies that showed that substitution of the
heterocycle by a NO2 group at the 4-, 5-, or 6-position
leads to less active compounds than at the 7-posi-
tion.12,16,18,21 Substitution of the indazole nucleus by
the electron-donating methoxy group at the 4-, 5-, 6-,
or 7-position leads to a decrease of the inhibitory activ-
ity but the 7-methoxy derivative remains the most po-
tent isomer in this series.22–24 The higher affinity of 1
for NOSs when compared to 7-methoxy-1H-indazole
has been attributed to the existence of a stronger H-
a
N
NH
N
NH
NO2
1
NH2
4
b
c
N
NH
N
NH
I
CN
6
5
Scheme 1. Synthesis of 7-substituted indazoles 4–6. Reagents and
conditions: (a) Pd/C, HCO2NH4, EtOH, reflux; (b) NaNO2, KI, HCl,
0 °C then rt; (c) dppf, Pd2(dba)3, Zn, Zn(CN)2, DMAC, 120 °C.

5964
B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
reduction of 1 by ammonium formate in the presence of
Pd/C.30 Further diazotization of 4 and reaction with KI
led to 7-iodo-1H-indazole 5 in 53 % overall yield from 1.
We were delighted to observe that our recently described
methodology afforded 6 in a reproducible 63% yield
from 5 with no need for protection of any nitrogen
atom.31 1H-Indazole-7-carbonitrile 6 was a versatile pre-
cursor for compounds 7-9 and the 3-bromo-derivatives
14-16 (Scheme 2). Acidic hydrolysis of 6 quantitatively
provided 1H-indazole-7-carboxylic acid 7. Interestingly,
only two syntheses of 7 have been recently patented. The
first one is based on lithiation of 7-bromo-1H-indazole
followed by addition of CO2 while the second one in-
volves the cyclization of 2-amino-3-methylbenzoic
acid.32,33 Using the mixed anhydride method, 7 was
quantitatively converted to 1H-indazole-7-carboxamide
8.34 Finally, 7-(1H-tetrazol-5-yl)-1H-indazole 9, which
can be considered as a bioisoster of 7, was obtained
using a Huysgens cycloaddition between 6 and Me3SiN3
in the presence of AlMe3.35 We next moved to the
synthesis of the 3-bromo derivatives 13–16. 3-Bromo-
7-iodo-1H-indazole 13 and 3-bromo-1H-indazole-7-car-
bonitrile 14 (Scheme 2) were easily obtained from 5 and
6, respectively, after regioselective bromination by N-
bromosuccinimide in the presence of KOH following a
previously described method.36 Acidic hydrolysis of 14
afforded 3-bromo-1H-indazole-7-carboxylic acid 15 in
an almost quantitative yield whereas 3-bromo-1H-inda-
zole-7-carboxamide 16 was easily obtained by activation
as a mixed anhydride and reaction with ammonia as de-
scribed above for 8.
a
N
NH
N
NH
I
5
19
SiMe3
b
c
N
NH
N
NH
NH
N
N
11
10
Scheme 3. Synthesis of triazole 11. Reagents and conditions: (a)
Me3Si–C„CH, Pd(PPh3)4, CuI, NEt3, CH3CN, rt; (b) K2CO3,
MeOH, rt; (c) Me3SiN3, CuI, MeOH, DMF, 120 °C.
Selective methylation of the N(1)-H group of 1 and 12
was performed by treatment with CH3I in the presence
of KOH as previously described.36 1-Methyl-7-nitro-
1H-indazole 17 and 3-bromo-1-methyl-7-nitro-1H-
indazole 18 were obtained in almost quantitative yields
without any formation of the N(2)-methyl isomers.36
3. Results and discussion
3.1. Effects of the indazole derivatives on NO formation
catalyzed by n-, i-, and eNOS
In order to extend the structure-activity relationship
profile, we next focussed on the preparation of 7-(1H-
1,2,3-triazol-5-yl)-1H-indazole 11. This compound was
obtained from 5 using a Sonogashira coupling with tri-
methylsilylacetylene in the presence of tetrakis(triphen-
ylphosphine)palladium followed by deprotection of the
terminal alkyne 10. Cycloaddition of compound 10 with
trimethylsilylazide afforded triazole 11 in 55 % overall
yield from 5 (Scheme 3).37,38
Compounds 1–18 were tested for their inhibitory effects
on the oxidation of L-arg to NO catalyzed by the three
recombinant isoforms of NOS containing all their cofac-
tors. The rates of NO formation by NOS were measured
using the oxidation of oxyhemoglobin to methemoglo-
bin by NO following a usual protocol.39 For compara-
tive purposes, 1 and 12, two already known inhibitors
of NOSs, were included in this study.12,13,16–18 Under
our conditions, 1 appeared to be a good but almost
non-selective inhibitor of the three NOSs, with IC50 val-
ues of 25 5, 50 10, and 35 5 lM towards n-, i-, and
eNOS, respectively (Table 1). Compound 12 was also a
non-selective inhibitor about 10-fold more potent than 1
with IC50 values around 2 lM, in agreement with previ-
ous literature data.18 From this series of 7-substituted
indazoles, compound 6 gave the best results with IC50
values similar to those observed with 1 towards nNOS
and eNOS. Its selectivity for nNOS versus iNOS was
two times better than that of 1 (determined at a simple
ratio of the IC50 values). As previously observed with
1, introduction of a 3-bromo substituent into compound
6 led to a strong increase of the inhibitory potency, and
the IC50 value of compound 14 towards nNOS was 10
times lower than that of 6 (2.5 0.5 lM). Interestingly,
14 was more selective of nNOS versus iNOS and eNOS
than 12 (Table 1). Furthermore, another new indazole
derivative, compound 8, was even more selective to-
wards nNOS versus iNOS and eNOS. It was half as po-
tent as 6 towards nNOS (IC50 = 55 10 lM) but was
R
R
a
N
N
NH
NH
CN
COOH
7 : R=H
15 : R=Br
6 : R=H
14 : R=Br
d
c
b
R
R
N
N
NH
NH
CONH2
N
NH
N
N
8 : R=H
16 : R=Br
9 : R=H
Scheme 2. Synthesis of 7-substituted indazoles 6–9 and 3-bromo
derivatives 14–16. Reagents and conditions: (a) 6 N HCl, reflux; (b)
NEt3, ClCO2Et, CH2Cl2, ꢀ20 °C, then NH3, ꢀ20 °C; (c) Me3SiN3,
AlMe3, toluene, 0 °C then 80 °C; (d) NBS, NaOH, DMF, rt.

B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
5965
Table 1. Effects of indazole derivatives 1–18 as inhibitors of purified recombinant NOS and on the visible spectrum of nNOSoxy
Compound
Substituent
Inhibitory effects (IC50 values, lM)a
Visible spectra
(kmax in nm)b
nNOS
25
iNOS
eNOS
35
1
2
7-NO2
7-OH
5
50 10
650 125
5
398
700 100
500 100
850 100
200 50
700 100
>1 mM(45%)c
400 80
n.d.
n.d.
n.d.d
n.d.
398
c
3
7-OSO2CF3
7-NH2
7-I
>1 mM (35%)
110 20
4
5
370 80
110 30
>1 mM (20%)c
850 150
>1 mM (5%)c
>1 mM (15%)c
>1 mM (20%)c
2.2 0.6
125 20
30
6
7-CN
7-CO2H
25
8
8
7
>1 mM (35%)c
55 10
>1 mM (40%)c
500 100
n.d.
n.d.
n.d.
n.d.
n.d.
396e
n.d.
398
8
7-CONH2
7-(Tetrazol-5-yl)
7-C„CH
7-(Triazol-5-yl)
3-Br-7-NO2
3-Br-7-I
9
>1 mM (15%)c
>1 mM (30%)c
>1 mM (45%)c
1.8 0.5
>1 mM (30%)c
225 50
10
11
12
13
14
15
16
17
18
>1 mM (45%)c
2.0 0.5
500 100
750 50
900 100
5.7 1.5
3-Br-7-CN
3-Br-7-CO2H
3-Br-7-CONH2
1-CH3-7-NO2
1-CH3-3-Br-7-NO2
2.5 0.5
13
3
>1 mM (30%)c
>1 mM (35%)c
700 50
700 150
>1 mM (15%)c
750 50
>500 (15%)c
>1 mM (10%)c
>1 mM (45%)c
250 50
n.d.
n.d.
n.d.
n.d.
>500 (40%)c
>500 (45%)c
a IC50 values for the inhibition of purified recombinant n-, i-, or eNOS by compounds 1–18 were determined by monitoring the oxidation by NO of
HbFeIIO2 to MetHbFeIII in incubations containing 10 lM L-arg and 10 lM H4B (final concentrations), as described inSection 5. Specific activities
for n-, i-, and eNOS were 350 50, 800 100, and 80 30 nmol minꢀ1 mg proteinꢀ1, respectively. IC50 values are means SD from 4 to 6
experiments.
b Position of the maximum appearing in the difference spectrum obtained after addition of 100 lM compounds 1–18 to a solution of 1 lM nNOSoxy
containing 1 mM ImH. This position is mentioned only for compounds leading to a significant difference spectrum. n.d., no appearance of a
significant difference spectrum.
c Percent inhibitions observed at the highest concentration tested (500 lM or 1 mM).
d When added to native nNOSoxy (without ImH), this compound gave rise to a peak at 429 nm and a valley at 420 nm.
e Appearance of a broad difference spectrum obscured by the absorption of 12 itself.
inactive against iNOS and eNOS (IC50 = 850 and
500 lM, respectively).
duces conformational changes with a displacement of a
glutamate residue involved in L-arg binding from its usual
position (Glu363, numbering from bovine eNOS).27,28
All the other tested compounds, which bear an OH,
OSO2CF3, NH2, I, CO2H, ethynyl, tetrazol-5-yl, or tria-
zol-5-yl substituent at position 7 of the indazole ring,
were very poor inhibitors of purified NOSs, with IC50
values between 200 lM and >1 mM. Introduction of a
methyl group at the N(1) position of 1 and 12 led to a
strong loss of the inhibitory effects (see compounds 17
and 18 in Table 1).
In order to compare the possible interactions of com-
pounds 1–18 with the nNOS active site, we have studied
their effects on the oxygenase domain of nNOS
(nNOSoxy) by differential UV–visible spectroscopy.
The UV–visible spectrum of nNOSoxy in Hepes buffer,
pH 7.4, displayed a broad Soret peak with a maximum
at 415 nm indicating that nNOSoxy exists as a mixture
of hexacoordinated heme–FeIII–H2O (low spin, LS)
and pentacoordinated heme–FeIII states (high spin,
HS), the former state being predominant, in agreement
with the literature data.42 Addition of 1 mM imidazole
(ImH) was done to completely convert nNOSoxy into a
LS hexacoordinated heme–FeIII–ImH complex absorb-
ing at 430 nm. This method was very often used to more
easily follow the binding of substrates to the NOS active
site.43 As previously described,24 the stepwise addition
of 1 to the nNOSoxy–FeIII–ImH complex gave rise to a
difference spectrum characterized by a trough at
425 nm and a peak at 398 nm, showing that the binding
of 1 close to the heme shifted the nNOSoxy spin state
equilibrium to the HS pentacoordinated heme–FeIII
state (called Type 1 interaction) corresponding to a dis-
placement of bound ImH.43 Under identical conditions,
the stepwise addition of 12 led to the appearance of a
broad difference spectrum, which was difficult to analyze
because of the absorption of the compound itself in the
380–420 nm region. By contrast, stepwise additions of 6
or 14 led to intense difference spectra almost identical to
The inhibitory effects of compounds 1–18 were tested to-
wards an inducible NOS expressed in a murine macro-
phage cell line.40 Using the nitrite accumulation in the
supernatant of LPS + c-IFN-stimulated RAW 264.7
cells as a measure of iNOS activity,41 compounds 1–
11, 13, 15–18 were almost inactive (IC50 > 200 lM, data
not shown). Only compound 12 and new compound 14
displayed significant inhibitory effects (IC50 = 80
25 lM for both compounds) without noticeable cell
toxicity. Further experiments using cell line expressing
nNOS are under investigation.
3.2. Spectroscopic studies of the binding of the indazole
derivatives 1–18 to nNOSoxy
High resolution crystal structures of the complexes be-
tween 1 and 12 with the oxygenase domains of i- and
eNOS (i- and eNOSoxy) have shown that these inhibitors
are bound at the active site and stacked parallel to the
heme plane within van der Waals contact. This binding in-

5966
B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
that obtained with 1 (Table 1). All the other indazole
derivatives tested led to very weak, not significant, dif-
ference spectra after addition either to native nNOSoxy
or to nNOSoxy–ImH complex (Table 1).
studies of the inhibition of recombinant nNOS in the pres-
ence of 10 lM H4B and various concentrations of sub-
strate L-arg and 6 were carried out. Double-reciprocal
plots of NO formation as a function of [L-arg] showed a
change of the apparent Km for nNOS-catalyzed oxidation
of L-arg in the presence of 6 without significant effect on
the apparent Vm (Fig. 2). These results indicated that
nNOS was inhibited by 6 in a competitive manner versus
L-arg. Replots of the apparent Km for L-arg oxidation as a
function of [6] led to an apparent Ki value of 13 lM for
this inhibition under the conditions used in these experi-
ments (Inset of Fig. 2).
In order to confirm the strong binding of 6 to the nNOSoxy
active site suggested by UV–visible spectroscopy, its ef-
fects on nNOSoxy spin state were followed by EPR spec-
troscopy at cryogenic temperatures. The X-band EPR
spectrum of 25 lM nNOSoxy at 12 K (Fig. 1A) showed
signals at g = 7.60, 4.10, and 1.870 corresponding to the
HS heme–FeIII complex, and broad signals which could
correspond to mixtures of LS heme–FeIII complexes
(g = 2.557, 2.512, 2.306 and 1.843).44 Addition of
100 lM of 1 led to a dramatic change of the EPR spectrum
with the appearance of an intense signal at g = 7.49, 4.62,
and 1.867, corresponding to a HS heme–FeIII complex,
clearly distinct from the native HS heme–FeIII complex.
Moreover, the features around g = 2 corresponding to
LS heme–FeIII complexes were almost undetectable
(Fig. 1B). Addition of 100 lM of 6 to nNOSoxy led to very
similar results with the appearance of intense signals at
g = 7.45, 4.34 and 1.862 corresponding to a HS heme–
FeIII complex characterized by g-values close to those of
the nNOSoxy–1 complex (Fig. 1C). As expected for HS
heme–FeIII complexes, an important decrease in intensity
of the HS features was observed at 35 K (data not shown).
By contrast, addition of compound 8 (200 lM) to nNO-
Soxy under identical conditions led to different EPR char-
acteristics, with much less intense signals at g-values (7.57,
4.17, and 1.861) that were closer to those of native nNO-
Soxy (data not shown).
The effects of H4B concentration on the inhibition of
nNOS-catalyzed oxidation of L-arg by 6 were studied
0.5
0.4
0.3
0.2
0.1
0
-0.4
0
0.4
0.8
)
1.2
3.3. Mode of inhibition of nNOS by 6
1 / L-arg (μΜ-1
On the basis of these results, we more deeply investigated
the inhibition of nNOS by the new compound 6. Kinetic
15
10
7.45
4.34
C
7.49
1.862
5
0
4.62
B
1.867
2.557
250
2.512
4.10
7.60
2.306
A
-20
0
20
40
60
80
Concentration of 6 (μM)
1.870
350 400
Figure 2. Effect of compound 6 on the apparent Km value of nNOS-
catalyzed oxidation of L-arg. Incubations were performed in the
presence of 10 lM H4B and the various concentrations of L-arg,
without (h) or in the presence of 5 lM (m), 25 lM (n), and 50 lM (d)
of 6. Incubations were initiated with the introduction of 20–40 nM
nNOS and the oxidation of oxyhemoglobin to methemoglobin by NO
formation was measured at 37 °C, as indicated in Section 5. Data are
plotted in double-reciprocal format and are representative of a typical
experiment. The inset shows the replot of apparent Km values versus
inhibitor concentrations.
0
50
100
150
200
300
Field (mT)
Figure 1. Effects of the addition of 1 or 6 on the EPR spectrum of
purified nNOSoxy. Line A, EPR spectrum of 25 lM native nNOSoxy
.
Line B, EPR spectrum after the addition of 100 lM 1. Line C, EPR
spectrum after the addition of 100 lM 6. Experimental conditions were
as follows: sample temperature, 12 K; microwave power, 20 mW;
modulation frequency, 100 kHz; modulation amplitude, 0.5 mT; aver-
aged scans, 4; scan width, 0.5 T; center field, 0.255 T.

B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
5967
from incubations of nNOS containing 10 lM L-arg and
various concentrations of 6 and H4B. Figure 3 shows
that the IC50 for 6 was reduced 10-fold when [H4B] de-
creased from 10 to 0.1 lM. These data suggested that 6
could compete with both the L-arg substrate and the
H4B cofactor in the nNOS active site, as previously
found for 1, 12, and 7-methoxy-1H-indazole.12,13,24,45,46
Finally, in order to check for the reversibility of the inhi-
bition of nNOS by 6, nNOS containing 10 lM H4B,
100 lM NADPH, and CaM was preincubated for
5 min at 25 °C in the presence of 2.5 mM of 6, a concen-
tration 100-fold higher than its IC50 value. Then, the
mixture was diluted 1000-fold and 10 lM L-arg was
added. NO formation was measured and the activity
was found to be not significantly different from that of
a control experiment performed without 6 (data not
shown).
plexes of 1 and 12 bound to i- and eNOS.27,28 Three key
features have been identified (Fig. 4A). First, the
indazole ring is positioned above the heme, making a
p-stacking between the aromatic rings. Second, two H-
bonds secure the position of the indazole ring: one
between the indazole N(1)-H and the carbonyl O-atom
of Trp358, and the second one between one O-atom of
the nitro group and the peptide NH group of Met360.
Finally, some hydrophobic stabilization of the 3-bro-
mine atom fitted in an hydrophobic pocket delimited
by Phe355, Gly357, Val 338, and the heme provides a
large increase in affinity (Fig. 4A).27,28 To analyze our
data at a molecular level, we have superimposed the
indazole ring of derivatives 2–11, and 13–18, to the inda-
zole ring of 12 bound to eNOS. We could then evaluate
if any of the three features listed above were still present
for each derivative. Obviously, the p-stacking between
the aromatic rings and the heme is always conserved.
Comparison of IC50’s of 1, 5–7 with those of 12-15,
respectively, reveals a correlation between values of 7-
monosubstituted and 3,7-disubstituted compounds.
Binding of substrates or inhibitors to the NOS active site
can also alter the electron-transfer from the reductase to
the heme. NADPH oxidase activity measures the effects
of compounds on electron-transfer between the two do-
mains of NOS, a process strongly dependent upon CaM
binding.47–49 As already described for this isoform,47,49
NADPH oxidase activity of nNOS was half-reduced in
the presence of 100 lM L-arg, and completely abolished
in the presence of 10 lM NO2-arg (data not shown).
The addition of 10 lM of 6 or 14 to nNOS containing
CaM completely reduced its NADPH oxidase activity
(data not shown).
3.4. Molecular modeling analysis
Comparison of NOS inhibitory potencies between our
new indazole derivatives and 12 can be rationalized on
the basis of the X-ray structures published for the com-
100
80
60
40
20
0
1
10
100
1000
Figure 4. (A) Crystallographic structure of 12 bound at the active site
of the oxygenase domain of eNOS (PDB access 1D0C). Heme and
inhibitor 12 are displayed in boldsticks and hydrogen bonds as green
dotted lines. A van der Waals sphere around the bromine atom shows
how it is fitted into the hydrophobic pocket. Atom color code: carbon,
gray; oxygen, red; nitrogen, blue; bromine, brown; sulfur, yellow. (B)
Superimposition of 14 to the crystal structure of 12 bound to the
oxygenase domain of eNOS. Visualization of crystallographic struc-
tures and superimposition were carried out using Discovery Studio 1.6
(Accelrys, San Diego).
Concentration of 6 (μM)
Figure 3. Effect of different H4B concentrations on the inhibition of
nNOS by 6. Incubations were performed in the presence of 10 lM
L-arg and 0.1 lM (d), 1 lM (m) or 10 lM (j) H4B. Incubations were
initiated with the introduction of 20–40 nM nNOS and formation of
NO was measured at 37 °C using the oxyhemoglobin assay, as
indicated in Section 5. Activities of nNOS in the presence of 0.1, 1.0,
and 10 lM H4B were 65 15, 310 80, and 330 50 nmol minꢀ1 mg
proteinꢀ1, respectively.

5968
B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
For compounds 12 and 14, a 10-fold decrease in IC50
values is observed by comparison to 1 and 6. This sug-
gests that the bromine atom of 13–15 is positioned in
the same hydrophobic pocket as for 12 and can provide
a similar stabilization. While the 3-bromo substitution is
not essential, the two H-bonds between the inhibitor
and Trp358 and Met360 are required for activity. In-
deed, the N(1)-methyl substitution, in 17 and 18, elimi-
nates the binding to Trp358 and dramatically reduces
the inhibition (Table 1). In compounds 2, 4, 5, and 13,
the hydrogen bond to Met360 may not occur and inhi-
bition is weakened although the interaction between
the N(1)-H of indazoles and Trp358 is still present.
The lack of activity of 9 and 11 could be due to the bulk
of their substituents at the 7-position that could not be
accomodated in that part of the binding site. On the
other hand, in 6 and 14, the cyano group could establish
H-bonds with the Trp358 and Met360 residues very
close to those observed with the nitro group of 1 and
12 (Fig. 4B). Unable to establish such H-bonds, com-
pound 10 is inactive. Interestingly, when a carboxylic
acid group is introduced in the 7-position, in compounds
7 and 15, the activity is abolished (Table 1). This could
be explained by the close proximity of the side chain of
Glu363, a critical residue for NOS activity able to adopt
various conformations, with a carboxylate function that
should be predominantly in an anionic form. The dis-
tance between the negative charge of 7 or 15 and of
between 7-substituted-1H-indazoles requires further
crystallographic studies and will help in the search of
more potent and selective inhibitors of nNOS with low
toxicity.
4. Conclusion
In order to find substitutes for 7-nitro-1H-indazole as
inhibitors of NOSs, we have synthesized six new inda-
zole derivatives bearing different substituents at position
7 and four new 3,7-disubstituted derivatives. Among our
series of fifteen indazole derivatives that were studied for
the first time as inhibitors of purified n-, i-, and eNOS
(compounds 3–11 and 13–18), only three compounds,
6, 8, and 14, acted as good inhibitors of the formation
of NO (IC50 < 100 lM). Compounds 6 and 14 exhibit
IC50 values similar to those of 1 and 12 (25 and 2 lM,
respectively, Table 1) and compound 14 seems slightly
more selective than 12 towards nNOS compared to
iNOS and eNOS.
Even though compound 8 is interesting, as it is selective
towards nNOS by comparison to i- and eNOS, it exhib-
its a very different behavior toward nNOS by compari-
son to 1. On the contrary, compounds 1 and 6 exhibit
very similar properties. Both are reversible inhibitors
of NOS-catalyzed NO formation in a competitive man-
ner versus both L-arg and H4B (Figs. 2 and 3),12,13,45,46
and they inhibit electron-transfer from the reductase to
the heme.24 These results showed that 1 and 6 inhibit
the catalytic activities that are catalyzed by the heme do-
main of nNOS. This was confirmed by our UV–visible
and EPR studies using nNOSoxy that clearly indicated
that binding of 1 and 6 to the nNOSoxy active site led
to HS heme–FeIII complexes exhibiting highly similar
UV–visible and EPR characteristics (Fig. 1). Finally, an-
other common feature between 1 and 6 resides in the
strong enhancement of their inhibitory effects by further
introduction of a bromine-atom at the 3-position of the
heterocycle. Although bearing distinct substituents, 1
and 6 thus display close analogies between their mecha-
nisms of inhibition and their binding modes.
˚
Glu363 varies from 1.2 to 4.0 A according to the differ-
ent conformations. This results in an electrostatic repul-
sion that may be the cause of the loss of inhibitory effect.
Such an argument would also explain the poor inhibi-
tory effects of compound 9 that should be negatively
charged at physiological pH. Such a repulsion should
not occur in the case of compound 8 that bears a
CONH2 7-substituent. This could explain the inhibitory
effects of this compound. However, several data show
that compounds 1 and 8 exhibit different behaviors to-
ward nNOS: (i) contrary to 1, 8 failed to produce any
UV–visible difference spectrum after addition to nNO-
Soxy, (ii) 1 and 8 led to clearly different EPR spectra with
nNOSoxy, and (iii) introduction of a 3-Br-substituent in
1 or 8 resulted in opposite effects on their IC50 values for
inhibition of nNOS (i.e., a clear decrease in the case of 1
instead of a marked increase in the case of 8, see Table
1). This suggests that the CONH2 group of 8 may still
interfere with the H-bonding network between the inhib-
itor and nNOS but with a somewhat distinct orienta-
tion. This could lead to a different positioning of the
indazole heterocycle in the active site resulting in the
inability of the bromine-atom to fit in the hydrophobic
pocket described above. A factor that could be involved
to explain these facts is the distinct electron-withdrawing
ability of the NO2 and CN groups compared to the
CONH2 group that might modify the p-stacking interac-
tions involved between the indazole ring and the heme.
New compounds 6 and 14 thus appear as good substi-
tutes for 1 and 12. They are as potent inhibitors of
nNOS as 1 and 12, and one may expect that they are less
toxic than 1 and 12 since they do not have any nitro sub-
stituent. Further experiments are required to determine
whether they keep the in vivo selectivity towards nNOS
reported for 1 and to confirm that they are non-toxic.
5. Experimental
5.1. Chemistry—general procedures
Altogether, our experimental results are in agreement
with this simple molecular analysis that demonstrates
that the three structural features identified for 12 must
be simultaneously present for high NOS inhibition po-
tency, as it is the case with the new compound 14. Better
understanding of these molecular structural differences
7-Nitro-1H-indazole 1, 3-bromo-7-nitro-1H-indazole
12, and N-(2-pyridyl) bis(trifluoromethanesulfonimide)
were purchased from Aldrich. All other commercially
available chemicals and reagents were purchased
from Sigma or Acros and were used without further
purification. 1H-Indazol-7-ol 2 was obtained from

B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
5969
7-methoxy-1H-indazole by treatment with BBr3 in
CH2Cl2 following a previously described protocol.22 Or-
ganic solvents were distilled from a drying agent prior to
use following usual methods. Chemical reactions were
monitored by analytical TLC using Merck precoated sil-
ica gel 60F254 plates. Flash chromatography was per-
formed on silica gel 60 (particle size 35–70 lm)
supplied by SDS. Melting points were determined on a
and 98% H2SO4 (2 mL) was added dropwise a solution
of NaNO2 (154.0 mg, 2.23 mmol) in 1 mL of water.
The mixture was stirred vigorously at 0 °C for 40 min.
Then, a solution of KI (747.0 mg, 4.500 mmol) in
0.9 mL of water was added dropwise and the mixture
was stirred at room temperature for 2 h. The product
was extracted with ethyl acetate, washed with a satu-
rated sodium carbonate solution, water and brine. After
drying on MgSO4, the solution was concentrated in va-
cuo. The crude product was purified by column chroma-
tography on silica gel, eluting with diethyl ether/pentane
(2:3) to afford 394.0 mg (71 %) of 5 as a brown powder.
Mp: 114 °C. 1H NMR (200 MHz, CDCl3) d 6.97 (t, 1 H,
J = 7.5), 7.55 (br s, 1H, NH), 7.73 and 7.77 (2d, 2H,
J = 7.5), 8.25 (s, 1H). 13C NMR (75 MHz, CDCl3) d
115.7, 121.0, 122.7, 123.2, 135.6, 136.2, 142.4. MS (EI)
m/z 117 (M-I)+, 127 (I+.), 244 (M+). Anal. Calcd for
C7H5N2I: C 34.43, H 2.05, N 11.47, Found: C 34.46,
H 1.93, N 11.38.
Buchi B-445 melting point apparatus and are uncor-
¨
1
rected. H, 13C, and 19F NMR spectra were recorded
in commercial deuterated solvents on Bruker AC 200
or Advance 300 spectrometers. Chemical shifts are re-
ported in parts per million (d) relative to tetramethylsil-
ane with peak multiplicities abbreviated as follows:
singlet, s; broad singlet, br s; doublet, d; triplet, t; mul-
tiplet, m. Coupling constants, J, are reported in Hz.
Chemical shifts for 19F NMR spectra are relative to
CFCl3. Infra-red spectra were recorded on a Nicolet
Opus-IR spectrometer with samples prepared as pellets
mixed with KBr. Mass spectra were recorded on a
GC-MS HP-5989B spectrometer (ILV, Versailles,
France). High resolution mass spectra were recorded
on a Jeol MS700 spectrometer (ENSCP, Paris, France).
Elemental analyses were performed at Service de Micro-
analyse ICSN-CNRS (Gif sur Yvette, France).
5.1.4. 1H-Indazole-7-carbonitrile (6). A flask flushed
with argon was charged with 5 (500.0 mg, 2.05 mmol),
1,10-bis(diphenylphosphino)ferrocene (dppf, 90.0 mg,
0.16 mmol, 7.8 mol %), tris(dibenzylideneacetone)dipal-
ladium (Pd2(dba)3, 75.0 mg, 0.082 mmol, 4.0 mol %),
zinc powder (32.0 mg, 0.490 mmol), and Zn(CN)2
(144.0 mg, 1.23 mmol). Degassed N,N-dimethylacet-
amide (40 mL) was added, and the resulting mixture
was heated at 120 °C with vigorous stirring for 4 h.
The reaction mixture was concentrated under vacuum
and diluted with ethyl acetate. The organic layer was
washed with 2 N NH4OH and brine, dried on MgSO4,
and concentrated in vacuo. The crude product was puri-
fied by chromatography on silica gel, eluting with
diethyl ether/pentane (2:3) to afford 190.0 mg (63%) of
5.1.1. 1H-Indazol-7-yl trifluoromethanesulfonate (3). To
a solution of 1H-indazol-7-ol 2 (20.0 mg, 0.149 mmol)
in 15 mL of THF containing Cs2CO3 (29.0 mg,
0.089 mmol) and K2CO3 (20.6 mg, 0.149 mmol), N-(2-
pyridyl)bis(trifluoromethane sulfonimide) (53.4 mg,
0.149 mmol) was added at ꢀ10 °C.29The resulting mix-
ture was stirred at room temperature for 2 h and the sol-
vent was evaporated. The product was extracted with
CH2Cl2, washed with water, 5% HCl, and brine. After
drying on MgSO4, the solvent was evaporated in vacuo
to afford 3 (38.0 mg, 96%) as a brown powder. Mp:
6
as
a
white powder. Mp: 160 °C. IR (KBr)
2218 cmꢀ1. H NMR (200 MHz, acetone-d6) d 7.32 (t,
1 H, J = 13.3), 7.87 (d, 1H, J = 13.3), 8.18 (d, 1H,
J = 13.3), 8.28 (s, 1H), 13.14 (br s, NH). 13C NMR
(75 MHz, acetone-d6) d 94.9, 116.9, 121.5, 125.2, 127.5,
132.8, 135.5, 136.2. MS (EI) m/z 143 (M+). Anal. Calcd
for C8H5N3: C 67.13, H 3.50, N 29.37. Found: C 67.14,
H 3.71, N 29.28.
1
1
94 °C. H NMR (200 MHz, acetone-d6) d 7.27 (t, 1H,
J = 9.2), 7.50 (d, 1H, J = 9.2), 7.93 (d, 1H, J = 9.2),
8.27 (s, 1H). 19F NMR (188 MHz, acetone-d6) d
ꢀ68.92 (s, 3F). 13C NMR (75 MHz, acetone-d6) d
115.8, 116.5, 119.0, 119.4, 121.8, 122.5, 122.9, 128.3.
MS (CICH4) m/z 267 (M+H+). HRMS Calcd for
C8H5N2F3O3S ðM þ NHþ4 Þ: 267.0051. Found: 267.0056.
5.1.5. 1H-Indazole-7-carboxylic acid (7). A flask was
charged with 6 (100 mg, 0.698 mmol) in 6 N HCl
(8 mL) and the resulting mixture was refluxed for 12 h.
The white precipitate was filtered, washed with cold
water and dried in vacuo to afford 7 (111 mg, 98%) as
a white powder. Mp: 225 °C (230–233 °C).33 1H NMR
(200 MHz, DMSO-d6) d 7.23 (t, 1H, J = 6.6), 7.97 (d,
1H, J = 6.6), 8.06 (d, 1H, J = 6.6), 8.20 (s, 1H), 9.85
and 13.12 (2 br s, NH and COOH). 13C NMR
(75 MHz, DMSO-d6) d 113.6, 119.9, 124.5, 126.3,
128.7, 134.1, 137.8, 166.8. MS (ESI negative mode) m/
z 161 (M-Hꢀ). HRMS Calcd for C8H7N2O2 (M+H)+:
163.0508. Found: 163.0504.
5.1.2. 1H-Indazol-7-amine (4). To a solution of 1 (5.00 g,
30.6 mmol) in EtOH (200 mL) containing 98 mg of Pd
on activated carbon was added ammonium formate
(25.1 g, 398 mmol). The solution was refluxed with vig-
orous stirring for 1 h. The catalyst was removed by fil-
tration and the solvent evaporated in vacuo. The
product was extracted with ethyl acetate, washed with
water and brine. After drying on MgSO4, the solution
was evaporated in vacuo to afford 3.06 g (75 %) of 4 as
a brown powder. Mp: 159 °C (162 °C).50 1H NMR
(200 MHz, CDCl3) d 5.80 (br s, 2H, NH2), 6.67 (d,
1H, J = 7.6), 7.01 (t, 1H, J = 7.6), 7.24 (2d, 2H,
J = 7.6), 8.08 (s, 1H). 13C NMR (75 MHz, CDCl3) d
110.8, 111.6, 122.2, 124.2, 130.6, 133.0, 134.8. MS (EI)
m/z 133 (M+).
5.1.6. 1H-Indazole-7-carboxamide (8). To a solution of 7
(140 mg, 0.863 mmol) and NEt3 (0.260 mL, 1.87 mmol)
dissolved in CH2Cl2 (13 mL), ethyl chloroformate
(95 lL, 0.99 mmol; 1.1 equiv) was added at ꢀ20 °C.
The mixture was stirred for 40 min and NH3 was then
5.1.3. 7-Iodo-1H-indazole (5). To a cooled solution of 4
(300.0 mg, 2.25 mmol) in concentrated HCl (0.56 mL)

5970
B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
slowly bubbled for 10 min at ꢀ20 °C. The reaction mix-
ture was poured into water and the product extracted
with CH2Cl2. The organic layer was washed with water,
1 N HCl, brine, dried over MgSO4 and evaporated to
dryness in vacuo. The product was purified by column
chromatography on silica gel, eluting with CH2Cl2/
EtOH (98:2), to afford 8 (128 mg, 92%) as a white pow-
der. Mp: 136 °C. 1H NMR (200 MHz, acetone-d6) d 5.70
(br s, 2H), 7.21 (t, 1H, J = 7.2), 7.97 (2d, 2H, J = 7.2),
8.11 (s, 1H), 12.25 (br s, 1H). 13C NMR (75 MHz, ace-
tone-d6) d 115.0, 120.7, 124.1, 124.6, 125.6, 125.9, 135.9,
168.1. MS (EI) m/z 161 (M+). HRMS Calcd for
C8H8N3O (M+H+): 162.0663. Found: 162.0667.
7.56 (d, 1H, J = 7.7), 7.78 (d, 1H, J = 7.7), 8.15 (s,
1H), 10.70 (br s, 1H). 13C NMR (75 MHz, CDCl3) d
79.2, 82.5, 104.4, 121.1, 122.1, 122.8, 130.8, 134.0,
143.5. MS (EI) m/z 142 (M+). HRMS Calcd for
C9H7N2 (M+H+): 143.0609. Found: 143.0620.
5.1.9. 7-(1H-[1,2,3]Triazol-5-yl)-1H-indazole (11).
A
sealed tube was charged with 10 (100 mg, 0.703 mmol),
CuI (6.7 mg, 0.035 mmol; 0.050 eq), MeOH (0.5 mL),
and DMF (4.5 mL). Trimethylsilylazide (0.140 mL,
1.0 mmol) was added and the mixture was heated at
120 °C for 48 h. The solvents were evaporated in vacuo
to afford a brown oil. Purification by column chroma-
tography on silica gel, eluting with ethyl acetate/petro-
leum ether (1:1) afforded 11 (125.0 mg, 96 %) as a
5.1.7. 7-(1H-Tetrazol-5-yl)-1H-indazole (9). A solution
of 6 (100 mg, 0.69 mmol) in 10 mL of toluene was added
dropwise to a solution of trimethylsilylazide (0.110 mL,
0.83 mmol) and trimethylaluminum (2.0 M in toluene,
0.420 mL, 0.84 mmol) in 2 mL of toluene at 0 °C. The
yellow mixture was stirred for 30 min at room tempera-
ture and then for 48 h at 80 °C. After cooling to 0 °C,
the mixture was poured into a biphasic mixture of
6 mL of ethyl acetate and 6 mL of 6 M HCl. The prod-
uct was extracted with ethyl acetate and the combined
organic layers were washed with water and brine, dried
over MgSO4 and evaporated under vacuum. Purification
by column chromatography on silica gel, eluting with
1
yellow powder. Mp: 211 °C. H NMR (200 MHz, ace-
tone-d6) d 2.96 (s, 1H), 4.04 (s, 1H), 7.25 (t, 1H,
J = 7.5), 7.82 (d, 1H, J = 7.5), 7.90 (d, 1H, J = 7.5),
8.15 (s, 1H), 12.40 (br s, 1H). 13C NMR (75 MHz, ace-
tone-d6) d 121.5, 121.6, 121.7, 123.9, 125.3, 129.6, 132.0,
133.5, 137.9. MS (CI-NH3) m/z 186 (M+H+). HRMS
Calcd for C9H8N5 (M+H+): 186.0780. Found: 186.0801.
5.1.10. 3-Bromo-7-iodo-1H-indazole (13). N-Bromosuc-
cinimide (730 mg, 4.10 mmol) and NaOH (170 mg,
4.25 mmol) were added to a solution of 5 (1.0 g,
4.10 mmol) in DMF (50 mL). The yellow mixture was
stirred for 48 hours at room temperature. After addition
of water, the product was extracted with ethyl acetate.
The combined organic layers were washed with water
and brine, dried over MgSO4 and concentrated in vacuo.
The crude mixture was purified by column chromatog-
raphy on silica gel eluting with diethyl ether/petroleum
ether (1:1) to afford 13 (1.04 g, 79%) as a white powder.
Mp: 161 °C. 1H NMR (200 MHz, CDCl3) d 7.03 (t, 1H,
J = 8.0), 7.63 (d, 1H, J = 8.0), 7.81 (d, 1H, J = 8.0),
10.10 (br s, 1H). 13C NMR (75 MHz, CDCl3) d 112.2,
120.4, 123.5, 124.3, 131.1, 136.7, 143.1. MS (EI) m/z
323 (M+), 243 (M-Br)+, 116 (M–Br–I)+. Anal. Calcd
for C7H4N2BrI: C 26.01, H 1.24, N 8.67, Found: C
26.23, H 1.28, N 8.61.
ethyl acetate/petroleum ether (1:1), afforded
9
(30.0 mg, 23%) as a brown powder. Mp: 320 °C. 1H
NMR (200 MHz, acetone-d6) d 3.72 (br s, 1H) 7.17 (t,
1H, J = 7.9), 7.72 (d, 1H, J = 7.9), 8.12 (s, 1H), 8.21
(d, 1H, J = 7.9), 12.52 (br s, 1H). 13C NMR (75 MHz,
acetone-d6) d 119.0, 120.0, 121.2, 123.0, 124.3, 124.5,
134.0, 163.5. MS (CI-NH3) m/z 187 (M+H+). HRMS
Calcd for C8H7N6 (M+H+): 187.0732. Found: 187.0757.
5.1.8. 7-Ethynyl-1H-indazole (10). A Schlenk flask was
charged with 5 (2.0 g, 8.2 mmol), tetrakis(triphenylphos-
phine)palladium (Pd(PPh3)4, 946 mg, 0.819 mmol,
10 mol %), and CuI (458.0 mg, 2.405 mmol). The flask
was flushed with argon, and degassed CH3CN
(30 mL), trimethylsilylacetylene (1.4 mL, 10 mmol),
and NEt3 (5 mL) were added. The orange mixture was
stirred for 6 hours at room temperature. The solution
was evaporated, and the product was extracted with
ethyl acetate. The organic layer was washed with water,
brine, dried over MgSO4 and evaporated in vacuo to af-
ford crude 7-((trimethylsilyl)ethynyl)-1H-indazole 19
(1.23 g, 70 %) as a yellow oil that was used directly in
5.1.11. 3-Bromo-1H-indazole-7-carbonitrile (14). Treat-
ment of a solution of 6 (20.0 mg, 0.140 mmol) in DMF
(5 mL) with N-bromosuccinimide (24.8 mg, 0.139 mmol)
and NaOH (5.6 mg, 0.14 mmol) for 5 hours at room
temperature was performed as described above for com-
pound 13. Purification of the crude product by column
chromatography on silica gel eluting with diethyl
ether/petroleum ether (1:1), afforded 14 (23.0 mg, 74%)
1
the next step. H NMR (200 MHz, CDCl3) d 0.32 (s,
9H), 7.13 (t, 1H, J = 8.1), 7.51 and 7.74 (2d, 2H,
J = 8.1), 8.10 (s, 1H). Potassium carbonate (570 mg,
4.12 mmol) was added to a solution of crude 19
(880 mg, 4.10 mmol) in 250 mL MeOH, and the mixture
was stirred for 16 hours at room temperature. The sol-
vent was evaporated, and the product was extracted
with ethyl acetate. The organic layer was washed with
water and brine, dried over MgSO4 and evaporated to
dryness in vacuo. Purification by column chromatogra-
phy on silica gel, eluting with ethyl acetate/pentane
(1:4), afforded 10 (337 mg, 60%) as brown needles.
as a white powder. Mp: 238 °C. IR (KBr) 2230 cmꢀ1
.
1H NMR (200 MHz, acetone-d6) d 7.45 (t, 1H,
J = 6.9), 8.00 (d, 2H, J = 6.9), 13.30 (br s, 1H). 13C
NMR (75 MHz, acetone-d6) d 95.5, 116.1, 122.6,
123.7, 124.7, 126.3, 134.3, 141.3. MS (EI) m/z 221
(M+), 142 (M–Br)+. Anal. Calcd for C8H4N3Br: C
43.24, H 1.80, N 18.92. Found: C 42.81, H 1.67, N
18.25.
5.1.12. 3-Bromo-1H-indazole-7-carboxylic acid (15). A
mixture of 14 (29.0 mg, 0.131 mmol) in 10 N HCl
(7 mL) was refluxed for 48 hours. After cooling to room
temperature, the white precipitate was filtered, washed
Mp: 77 °C. IR (KBr) 3280, 2190 cmꢀ1
.
1H NMR
(200 MHz, CDCl3) d 3.46 (s, 1H), 7.15 (t, 1H, J = 7.7),

B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
5971
with cold water and dried in vacuo to afford 15 (27.0 mg,
86 %) as a white powder. Mp: 280 °C. 1H NMR
(200 MHz, DMSO-d6) d 3.91 (br s, 1H), 7.35 (t, 1H,
J = 7.3), 7.87 (d, 1H, J = 7.3), 8.06 (d, 1H, J = 7.3),
13.47 (s, 1H). 13C NMR (75 MHz, DMSO-d6) d 114.4,
121.1, 121.4, 123.7, 124.8, 130.2, 139.0, 166.2. MS (ESI
negative mode) m/z 241 (M–H)ꢀ, 197 (M–CO2–H)ꢀ,
115 (M–CO2–Br–H)ꢀ. Anal. Calc. for C8H5N2O2Br: C
39.83, H 2.07, N,11.62. Found: C 39.87, H 1.97, N
12.11.
NADPH from Boehringer. Recombinant nNOS was
isolated and purified from the yeast Saccharomyces cere-
visiae transformed with a plasmid containing rat brain
nNOS as previously described.51 Full length recombi-
nant murine iNOS,52 bovine eNOS,53 and the heme do-
42
main of rat brain nNOS (nNOSoxy
)
were expressed in
Escherichia coli and purified in the absence of H4B and
L-arg as described previously. Protein concentrations
were determined by the method of Bradford using bo-
vine serum albumin as a standard and the Bradford re-
agent from Biorad.54 The heme concentrations of NOS
were determined optically from the [FeII-CO]– [FeII] dif-
5.1.13. 3-Bromo-1H-indazole-7-carboxamide (16). To a
cooled solution (ꢀ20 °C) of 15 (440 mg, 1.83 mmol)
and NEt3 (0.540 mL, 3.87 mmol) in CH2Cl2 (13 mL)
was added ethyl chloroformate (0.200 mL, 2.10 mmol).
The mixture was stirred for 40 min. Then, ammonia
was slowly bubbled through the mixture for 10 min at
ꢀ20 °C. The reaction mixture was poured into water
and the product was extracted with CH2Cl2. The organ-
ic layer was washed with water, 1 N HCl, brine, dried
over MgSO4 and evaporated to dryness in vacuo. The
product was purified by column chromatography on sil-
ica gel, eluting with CH2Cl2/ethanol (98:2), to afford 16
(414 mg, 95%) as a white powder. Mp: 287 °C. 1H NMR
(200 MHz, acetone-d6) d 7.18 (t, 1H, J = 6.3), 7.50 (br s,
2H), 7.53 (d, 1H, J = 6.3), 7.81 (d, 1H, J = 6.3), 12.25 (br
s, 1H). 13C NMR (75 MHz, acetone-d6) d 120.3, 120.5,
120.6, 120.9, 122.7, 128.4, 140.3, 168.5. MS (EI) m/z
239 (M+). HRMS Calc for C8H7N3OBr (M+H+):
239.9772. Found: 239.9773.
ference
spectrum
.
using
a
De445ꢀ480 nm
They were estimated to be more than
of
74 mMꢀ1 cmꢀ1
95% pure by SDS–PAGE electrophoresis.
55
5.2.2. Oxyhemoglobin assay. The rates of NO synthesis
were determined at 37 °C on an Uvikon 941 spectropho-
tometer using the oxyhemoglobin assay for NO.39 Usual
incubation mixtures were performed in 1-cm path length
cuvettes (total volume 150 lL) containing 50 mM Hepes
buffer (pH 7.4) with 0.1 M KCl supplemented with DTT
(5 mM), oxyhemoglobin (10-20 lM), 1000 U/mL each
of SOD and catalase, 10 lM each of H4B and L-arg,
CaCl2 (1 mM), CaM (10 lg/mL), NADPH (1 mM),
and inhibitors. Compounds 1–18 were added to the
incubation mixtures as 1.5 lL solutions in DMSO. The
mixtures were preincubated for 2 min at 37 °C prior to
initiation of the reaction by the addition of 2–5 lL ali-
quots of NOS to the sample cuvettes. The NO-mediated
conversion of oxyhemoglobin to methemoglobin was
monitored by repetitive scanning between 380 and
480 nm every 0.2 min, and quantitated using an extinc-
tion coefficient of 77 mMꢀ1.cmꢀ1 between peak at
401 nm and valley at 420 nm.39 The production of NO
was linear over all the entire reaction time for each iso-
form. Control incubations contained similar amounts
(1%) of DMSO without inhibitor. All values are ex-
pressed relative to the DMSO control and as
means SD from 3–4 experiments.
5.1.14. 1-Methyl-7-nitro-1H-indazole (17). KOH (56 mg,
1.0 mmol) and CH3I (0.150 mL, 2.4 mmol) were added to
a solution of 1 (163 mg, 1.0 mmol) in acetone (5 mL) at
0 °C. After stirring for 3 hours at room temperature, the
solution was evaporated. The crude product was dis-
solved in CH2Cl2, washed with water, brine, dried over
MgSO4, and the solvent wasevaporated in vacuo. The res-
idue was purified by column chromatography on silica
gel, eluting with diethyl ether/cyclohexane (1:1), to afford
17 (156 mg, 88%) as a yellow powder. Mp: 100 °C (99–
100 °C).36 1H NMR (200 MHz, acetone-d6) d 4.24 (s,
3H), 7.35 (t, 1H, J = 7.9), 8.16 (d, 1H, J = 7.9), 8.21 (d,
1H, J = 7.9), 8.30 (s, 1H).
5.2.3. NADPH oxidation by nNOS. The effects of com-
pounds 1–18 on the initial rates of NADPH oxidation
by nNOS were quantitated spectrophotometrically using
an extinction coefficient of 6.2 mMꢀ1 cmꢀ1 for the de-
crease in absorbance at 340 nm, as previously described.24
5.1.15. 3-Bromo-1-methyl-7-nitro-1H-indazole (18). A
solution of 12 (182 mg, 0.752 mmol) in acetone (3 mL)
was treated by CH3I (0.120 mL, 1.9 mmol) in the pres-
ence of the KOH (45 mg, 0.80 mmol) following the pro-
tocol described above for 17. Compound 18 (154 mg,
80%) was obtained as a yellow powder. Mp: 158 °C
(160–162 °C).36 1H NMR (200 MHz, CDCl3) d 4.26 (s,
3H), 7.32 (t, 1H, J = 7.6), 7.95 (d, 1H, J = 7.6), 8.22
(d, 1H, J = 7.6).
5.3. Spectroscopic studies
5.3.1. UV–visible spectroscopy. Optical spectra were re-
corded at room temperature in 1-cm path length cuv-
ettes containing 0.8–1.0 lM nNOSoxy in 150 lL of
50 mM Hepes pH 7.4. In some assays, ImH (1 mM)
was added to nNOSoxy to form the LS-heme FeIII
-
ImH complex. The baseline between 380 and 500 nm
was recorded and stepwise additions of the assayed com-
pounds dissolved in DMSO were performed. Equivalent
amounts of DMSO (maximal 2%) were added to the ref-
erence cuvette.
5.2. Biochemistry
5.2.1. General procedures. L-Arginine, L-citrulline, Nx-
nitro-L-arginine, dithiothreitol (DTT), hemoglobin,
superoxide dismutase (SOD), catalase, bovine serum
albumin, and porcine brain calmodulin were purchased
from Sigma. (6R)-5,6,7,8-Tetrahydrobiopterin and 7-NI
were purchased from Alexis (Coger, France) and
5.3.2. EPR spectroscopy. EPR spectra were recorded on
a Bruker Elexsys E500 EPR spectrometer operating at
X-band frequency (9.45 GHz) equipped with an shq

5972
B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
11. Flinspach, M. L.; Li, H.; Jamal, J.; Yang, W.; Huang, H.;
Hah, J. M.; Gomez-Vidal, J. A.; Litzinger, E. A.;
Silverman, R. B.; Poulos, T. L. Nat. Struct. Mol. Biol.
2004, 11, 54–59.
12. Wolff, D. J.; Gribin, B. J. Arch. Biochem. Biophys. 1994,
311, 300–306.
13. Wolff, D. J.; Lubeskie, A.; Umansky, S. Arch. Biochem.
Biophys. 1994, 314, 360–366.
14. Blasko, E.; Glaser, C. B.; Devlin, J. J.; Xia, W.; Feldman,
R. I.; Polokoff, M. A.; Phillips, G. B.; Whitlow, M.; Auld,
D. S.; McMillan, K.; Ghosh, S.; Stuehr, D. J.; Parkinson,
J. F. J. Biol. Chem. 2002, 277, 295–302.
0011 cavity and an Oxford Instrument liquid helium
probe. The following instrument settings were used:
modulation frequency, 100 kHz; modulation amplitude,
0.5 mT; time constant, 0.02 s; field sweep, 100 mT/min;
center field, 255 mT. Quartz tubes containing native
nNOSoxy (60 lL, 20–25 lM) or nNOSoxy containing
the studied compounds were frozen in cold ethanol
and then in liquid nitrogen. All spectra were recorded
at 12 2 K and 35 2 K, with four accumulations for
each experiment.
15. Paige, J. S.; Jaffrey, S. R. Curr. Top. Med. Chem. 2007, 7,
97–114.
5.4. Effects of compounds 1–18 on NO synthase activity in
RAW 264.7 cells
16. Moore, P. K.; Wallace, P.; Gaffen, Z.; Hart, S. L.;
Babbedge, R. C. Br. J. Pharmacol. 1993, 110, 219–224.
17. Babbedge, R. C.; Bland-Ward, P. A.; Hart, S. L.; Moore,
P. K. Br. J. Pharmacol. 1993, 110, 225–228.
18. Bland-Ward, P. A.; Moore, P. K. Life Sci. 1995, 57,
PL131–PL135.
19. Boelsterli, U. A.; Ho, H. K.; Zhou, S.; Leow, K. Y. Curr.
Drug Metab. 2006, 7, 715–727.
20. Purohit, V.; Basu, A. K. Chem. Res. Toxicol. 2000, 13,
673–692.
21. Boulouard, M.; Schumann-Bard, P.; Butt-Gueulle, S.;
Lohou, E.; Stiebing, S.; Collot, V.; Rault, S. Bioorg. Med.
Chem. Lett. 2007, 17, 3177–3180.
22. Schumann, P.; Collot, V.; Hommet, Y.; Gsell, W.;
Dauphin, F.; Sopkova, J.; MacKenzie, E. T.; Duval, D.;
Boulouard, M.; Rault, S. Bioorg. Med. Chem. Lett. 2001,
11, 1153–1156.
23. Collot, V.; Sopkova, J.; Schumann-Bard, P.; Colloc’h, N.;
MacKenzie, E. T.; Rault, S. J. Enzyme Inhib. Med. Chem.
2003, 18, 195–199.
The murine macrophage RAW 264.7 cells were plated at
100,000 cells/well in 96-well plates in 200 lL of RPMI
1640 culture medium supplemented with 5% FCS and
antibiotics, and incubated for 24 h at 37 °C under a
5% CO2 atmosphere.40 Fresh medium and various con-
centrations of compounds 1–18 were added 1 h before
the addition of 40 U/mL c-IFN and 50 ng/mL LPS.
After incubation for 20 h at 37 °C, 100 lL of the super-
natants were analyzed for nitrite formation using Griess’
reagent (1% sulfanilamide and 0.1% 1-naphthyl-ethy-
lenediamine in 0.5 M HCl) and absorbance at 550 nm
was measured on a Victor plate reader.41 Quantitation
of nitrite formation was performed with calibration
curves using known amounts of NaNO2 treated under
the same conditions.
24. Tuynman, A.; Perollier, C.; Frapart, Y.; Schumann-Bard,
P.; Collot, V.; Rault, S.; Boucher, J. L. Nitric Oxide 2003,
9, 86–94.
Acknowledgments
`
B.C. acknowledges the Ministere de l’Education Natio-
25. Sopkova, J.; Collot, V.; Rault, S. Acta Crystallogr. 2000,
C56, 1503–1504.
26. Sopkova-De Oliveira Santos, J.; Collot, V.; Rault, S. Acta
Crystallogr. 2002, C58, 688–690.
27. Raman, C. S.; Li, H.; Martasek, P.; Southan, G.; Masters,
B. S. S.; Poulos, T. L. Biochemistry 2001, 40, 13448–13455.
28. Rosenfeld, R. J.; Garcin, E. D.; Panda, K.; Andersson, G.;
Aberg, A.; Wallace, A. V.; Morris, G. M.; Olson, A. J.;
Stuehr, D. J.; Tainer, J. A.; Getzoff, E. D. Biochemistry
2002, 41, 13915–13925.
nale et de la Recherche for a fellowship. The authors
thank M.-A. Sari, and M. Jaouen (UMR 8601 CNRS,
Paris, France), and D. Durrad (Cleveland Clinic, Cleve-
land, USA) for their help in preparing proteins and per-
forming electrophoresis.
References and notes
29. Comins, D. L.; Dehghani, A. Tetrahedron Lett. 1992, 33,
6299–6302.
30. Ram, S.; Ehrenkaufer, R. E. Tetrahedron Lett. 1984, 25,
3415–3418.
31. Cottyn, B.; Vichard, D.; Terrier, F.; Nioche, P.; Raman,
C. S. Synlett 2007, 1203–1207.
32. Xie, W.; Herbert, B.; Schumacher, R.; Nguyen, T. M.;
Ma, J.; Gauss, C. M; Tehim, A. PCT Int. Appl. WO
2005092890, 2005.
33. Barber, C. G.; Bunnage, M. E.; Harvey, J. W.; Mathias, J.
P. A U.S. Patent Appl. Publ. US 2005020611, 2005.
34. Sokolov, Yu. A. Vest. Natsyyanal. Akad. Belar. 2000, 2,
76–80.
35. Huff, B. E.; Staszack, M. A. Tetrahedron Lett. 1993, 34,
8011–8014.
1. Kerwin, J. F.; Lancaster, J. R.; Feldman, P. F. J. Med.
Chem. 1995, 38, 4343–4362.
2. Pfeiffer, S.; Mayer, B.; Hemmens, B. Angew. Chem., Int.
Ed. 1999, 38, 1714–1731.
3. Davis, K. L.; Martin, E.; Turko, I. V.; Murad, F. Annu.
Rev. Pharmacol. Toxicol. 2001, 41, 203–236.
4. Masters, B. S. S.; McMillan, K.; Sheta, E. A.; Nishimura,
J. S.; Roman, L. J.; Martasek, P. FASEB J. 1996, 10, 552–
558.
5. Alderton, W. K.; Cooper, C. E.; Knowles, R. G. Biochem.
J. 2001, 357, 593–615.
6. Li, H.; Poulos, T. L. J. Inorg. Biochem. 2005, 99, 293–305.
7. Southan, G. J.; Szabo, C. Biochem. Pharmacol. 1996, 51,
383–394.
´
36. Bouissane, L.; El Kazzouli, S.; Leger, J. M.; Jarry, C.;
8. Babu, B. R.; Griffith, O. W. Curr. Opin. Chem. Biol. 1998,
2, 491–500.
9. Tinker, A. C.; Wallace, A. V. Curr. Top. Med. Chem. 2006,
6, 77–92.
10. Frey, C.; Narayanan, K.; McMillan, K.; Spack, L.; Gross,
S. S.; Masters, B. S.; Griffith, O. W. J. Biol. Chem. 1994,
269, 26083–26091.
Rakib, E. M.; Khouili, M.; Guillaumet, G. Tetrahedron
2005, 61, 8218–8225.
37. Dai, W. M.; Guo, D. S.; Sun, L. P. Tetrahedron Lett. 2001,
42, 5275–5278.
38. Penning, T. D.; Djuric, S. W.; Miyashiro, J. M.; Yu, S.;
Snyder, J. P.; Spangler, D.; Anglin, C. P.; Fretland, D. J.;

B. Cottyn et al. / Bioorg. Med. Chem. 16 (2008) 5962–5973
5973
Kachur, J. F.; Keith, R. H.; Tsai, B. S.; Villani-Price, D.;
Walsh, R. E.; Widomski, D. L. J. Med. Chem. 1995, 38,
858–868.
46. Mayer, B.; Klatt, P.; Werner, E. R.; Schmidt, K. Neuro-
pharmacology 1994, 33, 1253–1259.
47. Matsuoka, A.; Stuehr, D. J.; Olson, J. S.; Clark, P.; Ikeda-
Saito, M. J. Biol. Chem. 1994, 269, 20335–20339.
48. Abu-Soud, H. M.; Feldman, P. L.; Clark, P.; Stuehr, D. J.
J. Biol. Chem. 1994, 269, 32318–32326.
49. Klatt, P.; Heinzel, B.; John, M.; Kastner, M.; Bohme, E.;
Mayer, B. J. Biol. Chem. 1992, 267, 11374–11378.
50. Davies, R. R. J. Chem. Soc. 1955, 2412–2423.
51. Moali, C.; Boucher, J. L.; Sari, M. A.; Stuehr, D. J.;
Mansuy, D. Biochemistry 1998, 37, 10453–10460.
52. Wu, C.; Zhang, J.; Abu-Soud, H.; Ghosh, D. K.; Stuehr,
D. J. Biochem. Biophys. Res. Commun. 1996, 222, 439–444.
53. Ghosh, S.; Gachhui, R.; Crooks, C.; Wu, C.; Lisanti, M.
P.; Stuehr, D. J. J. Biol. Chem. 1998, 273, 22267–22271.
54. Bradford, M. M. Anal. Biochem. 1976, 72, 248–252.
55. Stuehr, D. J.; Ikeda-Saito, M. J. Biol. Chem. 1992, 267,
20547–20550.
39. Murphy, M. E.; Noack, E. Methods Enzymol. 1994, 233,
240–250.
40. Sekkai, D.; Guittet, O.; Lemaire, G.; Tenu, J. P.; Lepoivre,
M. Arch. Biochem. Biophys. 1997, 340, 117–123.
41. Lepoivre, M.; Flaman, J. M.; Henry, Y. J. Biol. Chem.
1992, 267, 22994–23000.
42. Abu-Soud, H. M.; Gachhui, R.; Raushel, F. M.; Stuehr,
D. J. J. Biol. Chem. 1997, 272, 17349–17353.
43. McMillan, K.; Masters, B. S. S. Biochemistry 1993, 32,
9875–9880.
44. Salerno, J. C.; Frey, C.; McMillan, K.; Williams, R. F.;
Masters, B. S.; Griffith, O. W. J. Biol. Chem. 1995, 270,
27423–27428.
45. Klatt, P.; Schmid, M.; Leopold, E.; Schmidt, K.; Werner,
E. R.; Mayer, B. J. Biol. Chem. 1994, 269, 13861–13866.