New hypotheses for the binding mode of 4- and 7-substituted indazoles in the active site of neuronal nitric oxide synthase

Article abstract of DOI:10.1016/j.bmc.2012.06.025

Taking into account the potency of 4- and 7-nitro and haloindazoles as nNOS inhibitors previously reported in the literature by our team, a multidisciplinary study, described in this article, has recently been carried out to elucidate their binding mode in the enzyme active site. Firstly, nitrogenous fastening points on the indazole building block have been investigated referring to molecular modeling hypotheses and thanks to the in vitro biological evaluation of N₁- and N₂-methyl and ethyl-4-substituted indazoles on nNOS. Secondly, we attempted to confirm the importance of the substitution in position 4 or 7 by a hydrogen bond acceptor group thanks to the synthesis and the in vitro biological evaluation of a new analogous 4-substituted derivative, the 4-cyanoindazole. Finally, by opposition to previous hypotheses describing NH function in position 1 of the indazole as a key fastening point, the present work speaks in favour of a crucial role of nitrogen in position 2.

Full text of DOI:10.1016/j.bmc.2012.06.025

Bioorganic & Medicinal Chemistry 20 (2012) 5296–5304
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
New hypotheses for the binding mode of 4- and 7-substituted indazoles
in the active site of neuronal nitric oxide synthase
Elodie Lohou a, Jana Sopkova-de Oliveira Santos a, Pascale Schumann-Bard b, Michel Boulouard b,
Silvia Stiebing a, Sylvain Rault a, Valérie Collot a,
⇑
a Université de Caen Basse-Normandie, EA-4258 CERMN, FR CNRS 3038 INC3M, SF 4206 ICORE, UFR des Sciences Pharmaceutiques, Boulevard Becquerel, F-14032 Caen, France
b Université de Caen Basse-Normandie, EA-4259 GMPC, UFR des Sciences Pharmaceutiques, Campus 5 ‘Santé’, Jules Horowitz, Boulevard Becquerel, F-14032 Caen, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Taking into account the potency of 4- and 7-nitro and haloindazoles as nNOS inhibitors previously
reported in the literature by our team, a multidisciplinary study, described in this article, has recently
been carried out to elucidate their binding mode in the enzyme active site. Firstly, nitrogenous fastening
points on the indazole building block have been investigated referring to molecular modeling hypotheses
and thanks to the in vitro biological evaluation of N1- and N2-methyl and ethyl-4-substituted indazoles
on nNOS. Secondly, we attempted to confirm the importance of the substitution in position 4 or 7 by a
hydrogen bond acceptor group thanks to the synthesis and the in vitro biological evaluation of a new
analogous 4-substituted derivative, the 4-cyanoindazole. Finally, by opposition to previous hypotheses
describing NH function in position 1 of the indazole as a key fastening point, the present work speaks
in favour of a crucial role of nitrogen in position 2.
Received 29 March 2012
Revised 11 June 2012
Accepted 13 June 2012
Available online 3 July 2012
Keywords:
Neuronal nitric oxide synthase inhibitors
7-Nitroindazole
4-Substituted indazoles
N-Alkylation
Ó 2012 Elsevier Ltd. All rights reserved.
Microwave-promoted palladium-catalysed
cyanation
Binding mode in the nNOS active site
1. Introduction
NO is implicated in a wide range of physiological processes,
including regulation of vascular tone and cerebral blood flow8 as
Over the last decades, nitric oxide (NO) has grown on as an ori-
ginal biological messenger because of its free radical and gaseous
nature1 as well as its specific mode of action. This molecule is pro-
duced essentially in neurons, endothelial cells and macrophages
well as control of platelet aggregation.5 Indeed, it activates from
endothelial cells the soluble guanylate cyclase in smooth muscle
cells and platelets which induces calcium concentration
decrease.3,4 Moreover, NO is involved in glutamatergic neurotrans-
mission since NMDA receptor stimulation provokes nNOS activa-
tion, but it can also act as a direct messenger in the nervous
nitridergic system.3,9 As such, it intervenes in pain perception10–12
and memorization.12,13 Finally, NO takes part in fight against invad-
ing pathogens after a massive release by iNOS and thanks to its
cytotoxic free radical properties.3
However, overproduction of NO, especially by nNOS and iNOS,
plays a role in several disorders such as post-ischemic stroke dam-
age14–16 and neurodegenerative diseases.17,18 Indeed, it exacerbates
oxidative stress and glutamatergic excitotoxicity. Moreover, it is
also involved in migraine headaches2,19 and sustained hypotension
during septic shock20 since it provokes a massive vasodilatation.
Thus, it is crucial to develop new selective nNOS inhibitors with
potential antinociceptive properties and neuroprotective functions
during ischemia or in case of Alzheimer’s disease, but which would
not perturb eNOS and its beneficial vascular effects. Despite a quite
low homology between the three human NOS primary sequences
(approximately 50%), the overall folds and the active site are very
closed in the three NOS oxygenase domains (NOSox). With this in
mind, specific inhibition of one isoenzyme constitutes a significant
during the conversion of
three major isoforms of NO Synthases (NOS). The neuronal and
endothelial enzymes (nNOS and eNOS) are constitutive and Ca2+
L-arginine to L-citrulline catalyzed by
-
dependent, whereas the inducible one (iNOS) is Ca2+-independent
and mostly expressed in activated macrophages.2,3 Nevertheless,
both enzymes appear as heme homodimers through which an elec-
tron flux can circulate between a reductase domain and an oxygen-
ase one. The redox reaction series needs the presence of six
cofactors: NADPH, FAD, FMN, tetrahydrobiopterin, calmodulin
and calcium.3–5 As a gas NO easily diffuses until neighbouring tar-
get cells and can interact with various proteins thanks to a direct or
an indirect mechanism.3,6 In the first case, it bonds to the heme en-
zyme iron atom to modulate the biological activity (activation of
soluble guanylate cyclase or retroinhibition of NOS. . .). In the sec-
ond case, it can also react with other free radicals to lead to specific
molecular modifications like reversible S-nitrosylations7 (activa-
tion of tissue plasminogen activator. . .).
⇑
Corresponding author. Tel.: +33 2 31566815; fax: +33 2 31566803.
E-mail address: valerie.collot@unicaen.fr (V. Collot).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.06.025

E. Lohou et al. / Bioorg. Med. Chem. 20 (2012) 5296–5304
5297
challenge for structure-based drug design. Nevertheless, several
families of nNOS inhibitors have already been discovered. First,
indazole derivatives.37,38 All 3D structures show that these inhibi-
tors don’t directly bond as the substrate neither to the glutamic
acid, nor to the heme iron of the active site. In the published
7-NI/iNOSox crystal structure as well as in the 7-NI/eNOSox one,
two possible orientations of 7-NI were observed (Fig. 1).38 In both
cases, 7-NI stands parallel to the heme porphyrin ring with a
substrate analogues like
L
-arginine derivatives bearing a guani-
dine2,4,21–23 or an amidine group,2,4,24,25 sulphur analogues of
L
-cit-
rulline4,24 and isothioureas4 have been synthesized, but despite the
fact that some of these compounds appear as potent NOS inhibitors,
none of them has showed a real selectivity towards nNOS. Then,
heterocyclic inhibitors including imidazole24 and indazole26,27
derivatives have been developed revealing two corresponding lead
compounds, the 1-((2-trifluoromethyl)phenyl)imidazole (TRIM)4
and the 7-nitroindazole (7-NI).4,10 Finally, Jaroch et al., Silverman
et al. and Annedi et al. have recently respectively described in the
literature some other promising new dihydroquinoleine28, 2-ami-
nopyridine29 and indole30,31 derivatives which appear as potent
and above all for the last ones selective nNOS inhibitors.
p-stacking formation between them and the establishment of
two hydrogen bonds. In the first majority orientation (named
orientation A—Fig. 1), the nitro group of 7-NI interacts through a
hydrogen bond with NH of Met 589 of the enzyme and the NH
function in position 1 of the compound with carbonyl group of
Trp 587. In the second orientation (named orientation B—Fig. 1),
the nitro group interacts with NH of Gly 586 and the NH function
in position 1 with carbonyl group of Trp 587.40 However, a previ-
ously reported X-ray structure of 7-NI/nNOSox and 3-bromo-7-
NI/NOSox complexes presented only the orientation A.39 We have
therefore based our modeling studies only on the orientation A.
The 7-NI/nNOS crystal model (orientation A) was energy mini-
mized and this analysis showed that the binding energy of 7-NI
with protein was about ꢀ28.8 kcal/mol. The energy minimization
moved little the 7-NI towards Gly 586 (RMSDheavy atoms = 0.55 Å)
and the enzyme backbone part interacting with 7-NI (from Phe
584 to Glu 590) changed only little (RMSDheavy atoms = 0.50 Å), prin-
cipally a rearrangement in the hydrogen orientations was observed
(Fig. 2). Moreover, the calculation of interaction energy with nNOS
per 7-NI atom proved that crucial fastening points for recognition
in the protein binding site were the oxygens of the nitro group (O9
and O10) and surprisingly the N2 atom (Table 1). This is a contradic-
tory result with respect to generally proposed 7-NI binding mode
in NOS active site, in which the 7-NI is supposed to interact with
the enzyme backbone through the oxygens of the nitro group, as
we observed, but also through the NH function in position 1. Our
study showed that the interaction energy of 7-NI H1 atom with en-
zyme is near zero, contrary to the interaction energy of 7-NI N2
atom which is one of the highest. Furthermore, the energy analysis
on the 7-NI itself (from energy minimized complex) showed that
the interaction exists between O10 and H1 atoms. The calculated
interaction energy between O10 and H1 is about ꢀ21.12 kcal/mol
and between O10 and N1 about ꢀ17.06 kcal/mol. Then, a previous
work on the X-ray structure of 7-NI itself published by our team re-
vealed that the NH function of indazole was involved in an intra-
molecular hydrogen bond with one oxygen of the nitro group.39
Thus, all these data showed that the NH function in position 1 of
the indazole is involved in an intramolecular hydrogen bond with
its nitro group also in the complex and it is therefore unlikely that
this function be able to make at the same time another hydrogen
bond with the protein backbone. Moreover, the visualization of
minimized complex showed that N2 of 7-NI was situated in the
proximity of backbone Gly 586 NH group. The existence of a hydro-
In this context, considering the laboratory’s know-how in inda-
zole chemistry32,33 and the interesting activity of 7-NI
(IC50 = 0.71 l
M)26 which furthermore reveals a good in vivo selec-
tivity towards nNOS, design, synthesis and pharmacological evalu-
ation of new indazole libraries (Scheme 1) have been developed for
a few years in our team in order to elucidate the structure–activity
relationships of these compounds as nNOS inhibitors. Some results
describing extensive pharmacomodulations realized on all
positions of the indazole six-membered ring have already been
published.34–36 This large study has shown that many of 4- and
7-substituted derivatives appear as potent inhibitors in the oppo-
site of 5- and 6-substituted ones. In this work, we have not only
confirmed the interest of substitution in position
IC50 = 0.6 M; 7-bromoindazole, IC50 = 14.3 M), but also discov-
ered a new family of potent nNOS inhibitors, the 4-substituted
indazoles (4-nitroindazole (4-NI), IC50 = 3.1 M; 4-bromoindazole,
IC50 = 2.4 M). In fact, these compounds had been until this time
7 (7-NI,
l
l
l
l
poorly investigated because of difficulties in chemical access.36
It is interesting to notice that all active 4- and 7-substituted
similar compounds bear electron withdrawing groups and, above
all, that these substituents can be considered as hydrogen bond
acceptors. Thus, taking into account the analogous properties of
these 4- and 7-nitro and haloindazoles, we report here a multidis-
ciplinary study which aims to elucidate their binding mode in the
nNOS active site. Firstly, nitrogenous fastening points on the inda-
zole building block have been investigated referring to molecular
modeling hypotheses and thanks to the in vitro biological evalua-
tion of N1- and N2-methyl and ethyl-4-substituted indazoles on
nNOS. Secondly, we attempted to confirm the importance of the
substitution in position 4 or 7 by a hydrogen bond acceptor group
thanks to the synthesis and the in vitro biological evaluation of a
new analogous 4-substituted derivative, the 4-cyanoindazole.
2. Study of nitrogenous fastening points
2.1. Molecular modeling of 7-NI and 4-NI in the nNOS active site
Met 589
Currently, several crystallographic data are available in the Pro-
tein Data Bank (PDB) presenting the NOSox co-crystallised with
Trp 587
Trp 587
Gly 586
B
A
4
3
3a
5
2
N
R
6
N
1
7a
7
H
Figure 1. Structure of iNOSox co-crystallised with 7-NI with two different inhibitor
orientations: orientation A in which the nitro group of 7-NI interacts with NH of
Met 589 and orientation B in which it interacts with NH of Gly 586. Dashed lines
indicate electrostatic interactions. This figure was made with PYMOL (DeLano
Scientific, 2002, San Carlo, USA).
R = OMe, NO2, F, Cl, Br, I...
Scheme 1. Pharmacomodulations recently described on the indazole six-mem-
bered ring.

5298
E. Lohou et al. / Bioorg. Med. Chem. 20 (2012) 5296–5304
Met 589
Met 589
Met 589
A
C
A
Gly586
Gly 586
Figure 3. Two proposed orientations of 4-NI in the nNOS binding cavity from
docking studies. Dashed lines indicate electrostatic interactions. This figure was
made with PYMOL (DeLano Scientific, 2002, San Carlo, USA).
Figure 2. The superposition of 7-NI orientation A within the nNOS binding cavity
before (7-NI in green and enzyme part in yellow) and after energy minimization (7-
NI and enzyme part in cyan). Dashed lines indicate electrostatic interactions. This
figure was made with PYMOL (DeLano Scientific, 2002, San Carlo, USA).
Met 589
Met 589
Table 1
Interaction energy between 7-NI, but also 4-NI and nNOS calculated per ligand atom.
The strongest interactions (more negative values) are in bold
A
C
7-NI
4-NI
Atom Interaction energy (kcal/mol) Atom Interaction energy (kcal/mol)
Gly 586
C7a
N1
H1
N2
C3
H3
C3a
C4
H4
C5
H5
C6
0.00
2.32
0.19
C3a
C3
0.00
1.16
1.64
H3
N2
N1
H1
C7a
C7
H7
C6
H6
C5
ꢀ17.74
ꢀ19.18
0.91
3.80
1.95
5.04
Figure 4. The superposition of 4-NI orientations A and C within the nNOS binding
cavity before (4-NI in blue and enzyme part in yellow) and after energy
minimization (4-NI in pink and enzyme part in cyan). This figure was made with
PYMOL (DeLano Scientific, 2002, San Carlo, USA).
ꢀ1.05
ꢀ3.42
2.84
ꢀ3.51
ꢀ3.27
3.14
ꢀ2.63
ꢀ2.40
1.91
1.56
ꢀ1.74
ꢀ0.50
0.44
ꢀ1.59
ꢀ0.84
0.50
minimization (RMSDheavy atoms = 2.47 Å). After the energy minimi-
zation the 4-NI in solution C was deflected from coplanar orienta-
tion with respect to the heme and therefore the solution A was
favoured ( Fig. 4). Interestingly, the interaction energy analysis
per 4-NI atom certified that 4-NI at orientation A reproduces the
binding mode of 7-NI with three key interactions to the enzyme
through the oxygen atoms of the nitro group as well as through
the N2 atom (Table 1). Then, looking at our modeling hypotheses,
we observed that for orientation A there is a possibility to put a lar-
ger substituent, like a methyl group, on the N1 atom, but in the ori-
entation C this substitution is not conceivable. Furthermore, for
orientation A N2-substituted derivatives should appear as less
potent nNOS inhibitors losing a possibility of hydrogen bond. Thus,
for a better understanding of 4-substituted indazole binding in the
nNOS active site and to confirm the new modeling hypotheses,
synthesis and in vitro pharmacological evaluation of 1- and 2-alkyl
4-substituted indazole derivatives have been developed.
H6
C7
H5
C4
N8
O9
O10
12.93
ꢀ9.04
ꢀ13.13
N8
O9
O10
13.91
ꢀ9.78
ꢀ14.40
gen bond of the Nꢁ ꢁ ꢁNH type is possible and justifies the observed
importance of the N2 atom in the interaction with the protein
(Fig. 2).
On the basis of these results, we propose that the 7-NI binding
mode in nNOS active site is as follows: 7-NI interacts with the en-
zyme backbone through the oxygens of the nitro group and the N2
atom.
Furthermore, the discovery of 4-NI efficiency as nNOS inhibitor
strengthens this new suggested binding mode. Indeed, if the nitro
group is conserved as a fixed fastening point, NH function in posi-
tion 1 of 4-NI moves at the opposite side of the binding cavity
without any contact with the protein backbone. Molecular docking
studies were therefore carried out to position 4-NI in the heme
binding cavity using Gold program.40 The selected docking poses
could be divided into two groups: one close to the orientation A
of 7-NI and the second different to those observed for 7-NI (named
orientation C—Figure 3).
Both complexes were submitted to energy minimization (Fig. 4).
In the case of the 4-NI/nNOSox complex, the energy minimization
turned slightly the 4-NI (RMSDheavy atoms = 0.55 Å) in the binding
cavity for the solution A but changed globally the 4-NI orientation
(RMSDheavy atoms = 2.47 Å) in the solution C. The changes in the
interacting protein backbone concerned principally the hydrogen
atoms for both solutions as in the case of the 7-NI/nNOSox com-
2.2. Synthesis and in vitro biological evaluation of N1- and N2-
methyl and ethyl-4-substituted indazoles 3–6
In this context, we directly managed to prepare, thanks to the
indazole-ring prototropic tautomery, 1- and 2-methyl-4-nitroin-
dazoles 3a–b, but also 1- and 2-methyl-4-bromoindazoles 5a–b
nearly in a ratio of two thirds to one third starting respectively
from 4-nitro and 4-bromoindazoles 1 and 2a in the presence of
iodomethane and potassium carbonate in acetone reflux41
(Scheme 2, Table 2). Similarly, 1- and 2-ethyl-4-nitroindazoles
4a–b and 1- and 2-ethyl-4-bromoindazoles 6a–b have been syn-
thesized in good yields replacing iodomethane by iodoethane
(Scheme 2, Table 2). We can notice that Cheung et al. reported in
2003 a regioselective methylation of indazole derivatives in posi-
plex (RMSDheavy atoms = 0.51 Å for solution
A and RMSDheavy
atoms = 0.41 Å for solution C). The comparison of the 4-NI/protein
binding energy showed that it was the same level for orientations
A (ꢀ25.2 kcal/mol) and C (ꢀ24.6 kcal/mol). However, the orienta-
tion C seems to be visually less likely since the nitro group doesn’t
interact with the protein and it was instable during the energy
tion
2 using trimethyloxonium tetrafluoroborate as alkyling
agent.42 However, we chose to use a more classic method to obtain,
after a simple separation by chromatography, the both expected 1-
and 2-substituted compounds. Moreover, each of these products

E. Lohou et al. / Bioorg. Med. Chem. 20 (2012) 5296–5304
5299
R
R
R
i
N
N
R'
N
+
N
N
N
H
R'
R = NO2 (1) or Br (2a)
R' = CH3 or C2H5
3-6
Scheme 2. Reagents and conditions: (i) CH3I or C2H5I (1.2–1.8 equiv), K2CO3 (3 equiv), acetone, reflux, 3–8 h, 35–69%.
Table 2
Table 3
N1 and N2-alkylation of 4-nitro and 4-bromoindazoles
IC50 values of N1- and N2-alkyl-4-substituted indazoles and 4-cyanoindazole against
rat cerebellar nNOS
R
R0
Total yield (%)
60
Ratio (%)
Indazole
IC50
(l
M)
Inhibition at
Vehicle Number of
experiments
4-NO2
N1-CH3
N2-CH3
N1-C2H5
N2-C2H5
N1-CH3
N2-CH3
N1-C2H5
N2-C2H5
3a
3b
4a
4b
5a
5b
6a
6b
59
41
65
35
68
32
57
43
derivative
mean sem
100
lM (%)
57
69
35
3a
3b
4a
4b
5a
5b
6a
6b
9a
10.6 4.2
86
DMSO
0.4%
7
6
4
3
3
3
3
3
3
4-Br
>100
34
67
34
75
19
61
0
DMSO
0.4%
51.7 22.3
>100
DMSO
0.4%
DMSO
0.4%
28.2 7.1
>100
DMSO
0.4%
has been identified thanks to X-ray crystallography studies real-
ized in our laboratory (Fig. 5).
DMSO
0.4%
Effects of these different indazole derivatives on nNOS activity
were evaluated in vitro according to a procedure described in the
experimental section. N1-alkyl-4-substituted indazoles 3a, 4a, 5a
and 6a appeared as less potent nNOS inhibitors than 4-nitro and
4-bromoindazoles, but remained active in contrast to N2-alkyl-4-
substituted indazoles 3b, 4b, 5b and 6b (Table 3). Thus, the NH
function at position 1 doesn’t seem to be essential for 4-substituted
indazole binding with the enzyme, but the substitution at this
point isn’t always very well-tolerated considering the nNOS activ-
ity especially with bulky groups. Indeed, N1-ethyl-4-substituted
indazoles 4a and 6a are less efficient nNOS inhibitors than N1-
methyl-4-substituted indazoles 3a and 5a (Table 3). Therefore,
these results allow us to reject the orientation C proposed by dock-
ing ( Fig. 3). Moreover, N2 indisponibility for binding due to its
obstruction by alkyl group leads on the contrary to the complete
loss of inhibitory activity which makes of this nitrogen a crucial
fastening point. In fact, the similar potency of 7-NI and 4-NI as
nNOS inhibitors only seems to be conceivable thanks to this com-
mon fastening point which allows the rotation of the nitro group
for both compounds to the correct position. We have also discov-
ered in the litterature26,27 that N1- and N2-methyl-7-nitroindazoles
are poor nNOS inhibitors what proves another time the importance
of the nitrogen in position 2 in the binding. The N1-methyl deriva-
tive would just be here too much clustered to have a good affinity
72.5 27.5
>100
DMSO
0.4%
DMSO
0.4%
42.5 2.5
63
DMSO
0.4%
(Fig. 2). On the contrary, the methyl group in position 1 of the 4-NI
corresponding compound seems to have enough place to be ac-
cepted (Fig. 3). Finally, all these biological data confirm the new
modeling hypotheses and demonstrate that 4-substituted inda-
zoles as well as 7-NI bind to the nNOS active site through the ori-
entation A (Figs. 2 and 3): crucial role in their binding is played by
the N2 atom and not by the NH function at position 1 as it was pre-
viously proposed.26,27,35
3. Introduction of a new hydrogen bond acceptor group in
position 4
3.1. Synthesis and in vitro biological evaluation of 4-
cyanoindazole 9a
All the already described potent nNOS inhibitor indazole deriv-
atives bear a hydrogen bond acceptor substituent in position 4 or 7.
As we have previously demonstrated thanks to the molecular mod-
eling studies, this substituent appears essential for their binding.
For the halogen derivatives, it was recently observed that halogens
are able to establish two kinds of contacts in biomolecular sys-
tems: C-XꢁꢁꢁO and C–Xꢁ ꢁ ꢁH.43,44 Therefore, we can assume that hal-
ogens play a role of hydrogen bond acceptors, like the nitro group,
in our system.
To develop the 4-substituted indazole new nNOS inhibitor fam-
ily and to replace the methemoglobinisant and carcinogenic nitro
group of 4-NI, we aimed to synthesize the 4-cyanoindazole also
bearing a hydrogen bond acceptor substituent. Indeed, Cottyn
et al. have already proved the interest of 7-cyanoindazole as nNOS
Figure 5. ORTEP diagram of the X-ray structure of 1- and 2-ethyl-4-nitroindazoles
(4a–b) with the thermal ellipsoid at 50% probability.

5300
E. Lohou et al. / Bioorg. Med. Chem. 20 (2012) 5296–5304
CN
CN
Br
Br
iii
i
ii
N
N
N
N
N
N
N
N
THP
H
H
THP
9a
8a
2a
7a
Scheme 3. Reagents and conditions: (i) DHP (2.5 equiv), TFA (cat.), EtOAc, reflux, 6 h, 82%; (ii) Zn(CN)2 (1.13 equiv), Pd2(dba)3 (0.04 equiv), S-Phos (0.1 equiv), DMF/water
99:1, microwave heating at 150 °C, 40 min, 79%; (iii) EtOH–HCl, rt, 3 h, 76%.
inhibitor (IC50 = 25 l
M).27 However, this compound didn’t appear
Met 589
Met 589
as potent as 7-NI that’s why we tried to use the analogy between
4- and 7-substituted indazoles to design a more efficient deriva-
tive. Taking into account the laboratory’s know-how in metal-cat-
alysed cross coupling reactions,32,33 a new palladium-catalysed
cyanation procedure has been carried out from haloindazoles and
using microwave heating to reduce reaction time comparing to
the described synthesis of 7-cyanoindazole. In fact, we first
adapted the isoquinoleine cyanation method reported by Choba-
nian et al. in 200645 to 5-bromoindazole 2b because of the usually
better reactivity of position 5. Unfortunately, by reacting with zinc
cyanide, Pd2(dba)3 as the catalyst and S-Phos as the ligand in DMF-
water 99:1 at 150 °C in a microwave reactor for 40 min, we only
obtained 5-cyanoindazole with a 22% yield. However, we sug-
gested that the unprotected NH group could poison the catalyst
by complexation with the palladium. Thus, thanks to a previously
described work about indazole protection by a tetrahydropyran-2-
yl (THP) group32,33 a new assay starting from 1-THP-5-bromoin-
dazole 7b could be realized and offered the corresponding cyano
derivative 8b with an excellent yield of 79%. Then, the same reac-
tion conditions were applied to 1-THP-4-bromoindazole 7a also
with a very good yield of 79%. Finally, we obtained the expected
unprotected 4- and 5-cyanoindazoles 9a–b with a respective yield
of 76% and 79% using a treatment with ethanolic HCl at room tem-
perature already reported by our team for similar deprotections of
other indazole derivatives (Scheme 3).33
C
A
Gly 586
Figure 6. Two proposed orientations of 4-cyanoindazole in the nNOS binding cavity
from docking studies. Dashed lines indicate electrostatic interactions. This figure
was made with PYMOL (DeLano Scientific, 2002, San Carlo, USA).
3.2. Molecular modeling of 4-cyanoindazole in the nNOS active
site
As in the case of 4-NI, docking studies of 4-cyanoindazole in the
nNOS active site proposed two orientations: one analogous to the
orientation A of 7-NI and the second close to the orientation C al-
ready observed with 4-NI (Fig. 6). The 4-cyanoindazole/nNOSox
interaction energy is approximately the same level for both orien-
tations with about a difference of 5 kcal/mol in benefit of the orien-
tation C. However, in the orientation C the cyano group doesn’t
participate in the interaction with the protein. Therefore, by anal-
ogy with 4-NI binding mode we can assume that the orientation C
be again excluded. The energy minimization turned slightly the
4-cyanoindazole in a similar way to orientation A of 4-NI/nNOSox
complex. The interaction energy with protein calculated per
4-cyanoindazole atom for the orientation A showed that the N2
atom and the nitrogen of cyano group (N9) participated strongly
in the fixation (Table 4). Indeed, hydrogen bond type interactions
were observed between N2 and NH of Met 589 (dN–N ꢂ2.99 Å)
and the nitrogen of cyano group and NH of Gly 586 (dN–N
ꢂ3.21 Å). Finally, in the orientation A, the 4-cyanoindazole is able
to reproduce the two attachment points observed with 7-NI and
4-NI. These last results confirm the new molecular modeling
hypotheses for the binding of 4 and 7-substituted indazoles: the
N2 atom and the substitution in position 4 or 7 by a hydrogen bond
acceptor group play a crucial role for their interaction with nNOS.
As shown in Table 3, 4-cyanoindazoles (9a) displayed a modest
inhibitory effect against NOS activity. However, this data
confirmed at least the already suggested importance of indazole
nucleus substitution by hydrogen bond acceptor groups at position
4. With this in mind, molecular modeling studies were finally car-
ried out considering this new indazole derivative to analyze the
potential similarities with other 4-substituted indazole binding
mode.
Table 4
Interaction energy between 4-cyanoindazole 9a and nNOS calculated per ligand atom.
The strongest interactions (more negative values) are in bold.
4-Cyanoindazole
Atom
Interaction energy (kcal/mol)
C7a
N1
H1
N2
C3
0.00
1.10
1.50
4. Conclusion
ꢀ18.24
3.48
To conclude, several advances have been made in the study of
4-substituted indazoles as potent neuronal nitric oxide synthase
inhibitors. Firstly, this multidisciplinary work allowed us to eluci-
date the binding mode of these indazole derivatives as well as of
7-NI in the nNOS active site. Indeed, by opposition to previous
hypotheses26,27,35,37,38 describing the NH function in position 1 of
the indazole as a key fastening point, the present study speaks in
favour of a crucial role of the nitrogen in position 2. Secondly,
the interest of substitution at position 4 by a hydrogen bond accep-
tor has been confirmed thanks not only to the synthesis and the
H3
C3a
C4
H4
C5
H5
C6
H6
C7
4.66
ꢀ3.84
ꢀ3.65
3.04
ꢀ3.12
2.53
ꢀ2.19
0.11
0.05
6.20
ꢀ17.50
C8
N9

E. Lohou et al. / Bioorg. Med. Chem. 20 (2012) 5296–5304
5301
in vitro biological evaluation of a new compound in this series, the
4-cyanoindazole, but also by molecular modeling studies. Taking
into account these new data, advanced structure-activity relation-
ship studies could be performed in the future to explore notably
the substitution in position 1 of 4-substituted indazoles since 3-
bromo-7-NI appears in the litterature, by analogy, as a potent
J = 7.9 Hz, 1H), 8.10 (d, J = 8.3 Hz, 1H), 8.17 (d, J = 7.6 Hz, 1H),
8.54 (s, 1H), 13.92 (s, 1H). 13C NMR (DMSO-d6) d: 115.3, 118.3,
118.5, 125.7, 132.5, 139.7, 141.6. HRMS/EI m/z calcd for
C7H5N3O2 [M]+: 163.0382; found: 163.0386.
5.1.1.2. 4-Bromo-1H-indazole (2a).
Starting from the 3-bro-
nNOS inhibitor (IC50 = 0.17
l
M).26
mo-2-methylaniline (5.9 g, 31.7 mmol) and respecting the general
procedure, the 3-bromo-2-methylphenyldiazonium tetrafluorobo-
rate salt (9.1 g, 31.4 mmol) was first obtained, followed by the ex-
pected product 2a as a brown solid (3.6 g, 57% for two steps). Mp
170 °C. TLC Rf = 0.24 (Silica gel; EtOAc/cyclohexane 1:3). IR (KBr):
5. Experimental section
5.1. Chemistry
3438, 3142, 3048, 2857, 1621, 1354, 1183, 965, 916, 769 cmꢀ1
.
1H NMR (CDCl3) d: 7.23 (t, J = 10.7 Hz, 1H), 7.33 (d, J = 7.8 Hz,
1H), 7.46 (d, J = 7.8 Hz, 1H), 8.12 (s, 1H), 10.75 (s, 1H). 13C NMR
(CDCl3) d: 108.8, 114.6, 124.0, 124.5, 127.6, 135.1, 140.6. LRMS/EI
m/z (%): 199 (2.8), 198 (M+, 1.97), 197 (9), 196 (M+, 100), 117
(23). Anal. calcd for C7H5BrN2 (%): C, 42.67; H, 2.56; N, 14.22;
found: C, 42.76; H, 2.34; N, 13.82.
All commercial reagents were used as received without further
purification. Reaction mixtures were stirred magnetically and
monitored by TLC using 0.2 mm Macherey-Nagel Polygram SIL G/
UV254 precoated plates. Column chromatography was performed
using CarloErba-SDS 60A 70–200 lm silica gel. Melting points were
determinated on a Köfler melting point apparatus. IR spectra were
taken with a Perkin-Elmer spectrum X FT-IR spectrometer. 1H NMR
(400 MHz) and 13C NMR (100 MHz) spectra were recorded on a
JEOL Lambda 400 spectrometer. Chemical shifts (d) are expressed
in parts per million downfield from tetramethylsilane as an inter-
nal standard and the coupling constants are in Hertz. High and low
resolution electron impact (EI) mass spectra were recorded on a
JEOL JMS GCMate spectrometer with ionising potential of 70 eV
and with pfk as internal standard. Low resolution electrospray
(ESI) mass spectra were obtained on a LC/MS Waters alliance
2695 with a Xterra MSC 118/2,1 ꢃ 50 mm column coupled with a
Micromass ZMD 2000 spectrometer. Elemental analyses were per-
formed at the ‘Institut de Recherche en Chimie Organique Fine’
(Rouen, France).
5.1.2. General procedure for the synthesis of 1- and 2-methyl-4-
nitro and 4-bromo-1H-indazoles (3a–b and 5a–b)
To a solution of potassium carbonate (3 equiv) in acetone
(15-30 mL) was added the chosen 4-substituted indazole 1 or 2a
and the mixture was stirred for 30 min at room temperature. Then,
methyl iodide (1.2–1.8 equiv) was finally introduced in the reac-
tion mixture which was heated under reflux conditions for 3–8 h.
After evaporating acetone, the residue was dissolved in EtOAc
(30 mL) and the organic layer was washed with brine
(3 ꢃ 15 mL), dried over MgSO4, filtered and evaporated in vacuo.
The crude material was purified by flash column chromatography
on silica gel (EtOAc/cyclohexane, 1:2 or CH2Cl2) to give the ex-
pected 1- and 2-methyl-4-substituted-indazoles 3a–b and 5a–b.
Single crystals of compounds 4a–b suitable for X-ray crystallo-
graphic analysis were obtained by slow evaporation from cyclo-
hexane. Data for crystal structure analysis were collected at
296 K with a Bruker-Nonius Kappa CCD area detector diffractome-
5.1.2.1. 1-Methyl-4-nitro-1H-indazole (3a).
Starting from
the 4-nitro-1H-indazole 1 (0.5 g, 3.1 mmol) and following the gen-
eral procedure with a purification by flash column chromatogra-
phy on silica gel thanks to CH2Cl2, the product 3a was obtained
as a yellow solid (0.2 g, 35%). Mp 145 °C. TLC Rf = 0.28 (Silica gel;
CH2Cl2). IR (KBr): 1513, 1341, 1324, 1267, 999, 963, 808, 738,
ter
with
graphite-monochromatized
Mo
Kk
radiation
(k = 0.71073 Å). The structure was solved using direct methods
and refined by full-matrix least-squares analysis on F2. Program
used to solve and refine structure: SHELXS-97. Software used to pre-
pare material for publication: SHELXL-97.46
729 cmꢀ1 1H NMR (CDCl3) d: 4.18 (s, 3H), 7.52 (t, J = 7.8 Hz, 1H),
.
7.77 (d, J = 7.8 Hz, 1H), 8.16 (d, J = 7.8 Hz, 1H), 8.61 (s, 1H). 13C
NMR (CDCl3) d: 35.0, 114.9, 115.9, 117.0, 124.2, 131.5, 139.6,
140.4. LRMS/ESI: [M+H]+ 178. HRMS/EI m/z calcd for C8H7N3O2
[M]+: 177.0538; found: 177.0537.
5.1.1. General procedure for the synthesis of 4-nitro and 4-
bromo-1H-indazoles (1 and 2a)36
To a cooled solution of the chosen 2-methylaniline dissolved in
HBF4 (50% solution in water; 15–30 mL) was added at 0 °C drop-
wise a cooled aqueous solution of NaNO2 (1 equiv in the minimum
of water). After the end of the addition, the mixture was stirred 1 h
at 0 °C and 2 h at room temperature. Then, the resulting precipitate
was filtered, washed with Et2O (3 ꢃ 100 mL) and dried to obtain
the corresponding 2-methylphenyldiazonium tetrafluoroborate
salts which were directly added in one portion under nitrogen to
a stirred mixture of KOAc (2 equiv) and 18-crown-6 (0.05 equiv)
in dry CHCl3 (350–700 mL). After 2 h at room temperature, the
mixture was filtered, washed with CHCl3 (3 ꢃ 100 mL) and the or-
ganic filtrate was finally concentrated in vacuo. The residual gum
was purified by column chromatography on silica gel (EtOAc/
cyclohexane 1:3) to give the desired indazoles 1 and 2a.
5.1.2.2. 2-Methyl-4-nitro-1H-indazole (3b).
Starting from
the 4-nitro-1H-indazole 1 (0.5 g, 3.1 mmol) and following the gen-
eral procedure with a purification by flash column chromatogra-
phy on silica gel thanks to CH2Cl2, the product 3b was obtained
as a yellow solid (0.1 g, 25%). Mp 110 °C. TLC Rf = 0.31 (Silica gel;
CH2Cl2). IR (KBr): 2927, 1522, 1338, 1280, 1153, 783 cmꢀ1 1H
.
NMR (CDCl3) d: 4.32 (s, 3H), 7.41 (t, J = 7.8 Hz, 1H), 8.08 (d,
J = 8.8 Hz, 1H), 8.19 (d, J = 7.8 Hz, 1H), 8.56 (s, 1H). 13C NMR (CDCl3)
d: 40.9, 115.1, 120.6, 124.3, 125.2, 125.8, 140.6, 150.0. LRMS/ESI:
[M+H]+ 178. LRMS/EI m/z (%): 178 (13), 177 (M+, 100), 131 (40),
116 (5). HRMS/EI m/z calcd for C8H7N3O2 [M]+: 177.0538; found:
177.0530.
5.1.2.3. 1-Methyl-4-bromo-1H-indazole (5a).
Starting from
5.1.1.1. 4-Nitro-1H-indazole (1).
Starting from the 3-nitro-2-
the 4-bromo-1H-indazole 2a (0.5 g, 2.5 mmol) and following the
general procedure with a purification by flash column chromatog-
raphy on silica gel thanks to EtOAc/cyclohexane, 1:2, the product
5a was obtained as an orange oil (0.2 g, 47%). TLC Rf = 0.54 (Silica
gel; EtOAc/cyclohexane, 1:2). IR (KBr): 1615, 1561, 1494, 1432,
methylaniline (10 g, 65.7 mmol) and respecting the general proce-
dure the 3-Nitro-2-methylphenyldiazonium tetrafluoroborate salt
(11.7 g, 46.6 mmol) was first obtained, followed by the expected
product 1 as a brown solid (6.8 g, 63% for two steps). Mp 207 °C.
TLC Rf = 0.09 (Silica gel; EtOAc/cyclohexane 2:3). IR (KBr): 3311,
1262, 1187, 1114, 989, 915; 830, 770, 730, 627 cmꢀ1 1H NMR
.
1518, 1337, 1239, 942, 738 cmꢀ1 1H NMR (DMSO-d6) d: 7.62 (t,
.
(CDCl3) d: 4.08 (s, 3H), 7.24 (t, J = 6.8 Hz, 1H), 7.30 (d, J = 6.8 Hz,

5302
E. Lohou et al. / Bioorg. Med. Chem. 20 (2012) 5296–5304
1H), 7.35 (d, J = 7.8 Hz, 1H), 8.00 (s, 1H).13C NMR (CDCl3) d: 34.9,
107.0, 113.6, 122.3, 124.0, 126.0, 131.8, 139.4. LRMS/ESI: [M+H]+
211 and 213. HRMS/EI m/z calcd for C8H7BrN2 [M]+: 209.9793;
found: 209.9790.
6a was obtained as an orange oil (0.2 g, 20%). TLC Rf = 0.42 (Silica
gel; EtOAc/cyclohexane, 1:3). IR (KBr): 2980, 1612, 1560, 1492,
1447, 1363, 1292, 1177, 1122, 915, 828, 770, 731 cmꢀ1 1H NMR
.
(CDCl3) d: 1.50 (t, J = 6.8 Hz, 3H), 4.41 (q, J = 6.8 Hz, 2H), 7.20 (t,
J = 7.8 Hz, 1H), 7.27 (d, J = 6.8 Hz, 1H), 7.33 (d, J = 7.8 Hz, 1H),
8.00 (s, 1H).13C NMR (CDCl3) d: 14.8, 44.1, 108.0, 114.7, 123.2,
125.1, 126.8, 132.8, 139.5. LRMS/ESI: [M+H]+ 225 and 227. LRMS/
EI m/z (%): 226 (M+, 73), 224 (M+, 75), 213 (100), 212 (94), 205
(42), 161 (73), 146 (29). HRMS/EI m/z calcd for C9H9BrN2 [M]+:
225,0027; found: 225,0023.
5.1.2.4. 2-Methyl-4-bromo-1H-indazole (5b).
Starting from
the 4-bromo-1H-indazole 2a (0.5 g, 2.5 mmol) and following the
general procedure with a purification by flash column chromatog-
raphy on silica gel thanks to EtOAc/cyclohexane, 1:2, the product
5b was obtained as a brown oil (0.1 g, 22%). TLC Rf = 0.25 (Silica
gel; EtOAc/cyclohexane, 1:2). IR (KBr): 2944, 1625, 1534, 1400,
1369, 1283, 1190, 1157, 936, 770 cmꢀ1
.
1H NMR (CDCl3) d: 4.24
5.1.3.4. 2-Ethyl-4-bromo-1H-indazole (6b).
Starting from
(s, 3H), 7.14 (t, J = 6.8 Hz, 1H), 7.24 (d, J = 6.8 Hz, 1H), 7.63 (d,
J = 8.8 Hz, 1H), 7.93 (s, 1H). 13C NMR (CDCl3) d: 40.5, 112.7,
116.5, 124.1, 124.2, 124.6, 126.6, 148.9. LRMS/ESI: [M+H]+ 211
and 213. HRMS/EI m/z calcd for C8H7BrN2 [M]+: 209.9793; found:
209.9783.
the 4-bromo-1H-indazole 2a (1 g, 5.1 mmol) and following the
general procedure with a purification by flash column chromatog-
raphy on silica gel thanks to EtOAc/cyclohexane, 1:2, the product
6b was obtained as an orange oil (0.2 g, 15%). TLC Rf = 0.35 (Silica
gel; EtOAc/cyclohexane, 1:3). IR (KBr): 2982, 1625, 1533, 1444,
1413, 1371, 1294, 1188, 1153, 932, 769 cmꢀ1 1H NMR (CDCl3) d:
.
5.1.3. General procedure for the synthesis of 1- and 2-ethyl-4-
bromo and 4-nitro-1H-indazoles (4a–b and 6a–b)
1.62 (t, J = 6.8 Hz, 3H), 4.44 (q, J = 6.8 Hz, 2H), 7.12 (t, J = 6.8 Hz,
1H), 7.22 (d, J = 6.8 Hz, 1H), 7.64 (d, J = 8.8 Hz, 1H), 7.92 (s, 1H).
13C NMR (CDCl3) d: 15.6, 48.6, 112.7, 116.5, 122.8, 123.7, 124.0,
126.4, 148.6. LRMS/ESI: [M+H]+ 225 and 227. LRMS/EI m/z (%):
226 (M+, 93), 224 (M+, 100), 198 (52), 196 (55). Anal. calcd for
C9H9BrN2 (%): C, 48.03; H, 4.03; Br, 35.50; N, 12.45; found: C,
48.26; H, 4.15; Br, 35.83; N, 12.40.
To a solution of potassium carbonate (3 equiv) in acetone (15–
30 mL) was added the chosen 4-substituted indazole 1 or 2a and
the mixture was stirred for 30 min at room temperature. Then,
ethyl iodide (1.2–1.8 equiv) was finally introduced in the reaction
mixture which was heated under reflux conditions for 3–8 h. After
evaporating acetone, the residue was dissolved in EtOAc (30 mL)
and the organic layer was washed with brine (3 ꢃ 15 mL), dried
over MgSO4, filtered and evaporated in vacuo. The crude material
was purified by flash column chromatography on silica gel
(EtOAc/cyclohexane, 1:2 or CH2Cl2) to give the expected 1- and
2-ethyl-4-substituted-indazoles 4a–b and 6a–b.
5.1.4. General procedure for the synthesis of 4 and 5-bromo-1-
32
(tetrahydro-2H-pyran-2-yl)-1H-indazoles (7a–b)
The chosen bromoindazole 2a–b dissolved in ethyl acetate or
chloroform (20 mL), a catalytic amount of TFA and 3,4-dihydro-
2H-pyrane (2.5 equiv) were heated under reflux conditions for
6 h. Then, the solution was concentrated in vacuo and EtOAc
(40 mL) was added to the residue. The organic layer was succes-
5.1.3.1. 1-Ethyl-4-nitro-1H-indazole (4a).
Starting from the
4-nitro-1H-indazole 1 (0.5 g, 3.1 mmol) and following the general
procedure with a purification by flash column chromatography
on silica gel thanks to CH2Cl2, the product 4a was obtained as a yel-
low solid (0.2 g, 37%). Mp 119 °C. TLC Rf = 0.35 (Silica gel; CH2Cl2).
IR (KBr): 2978, 1522, 1336, 1289, 1197, 1027, 951, 804, 788,
sively washed with a satured potassium carbonate solution
(3 ꢃ 20 mL) and brine (3 ꢃ 20 mL), dried over MgSO4, filtered and
evaporated in vacuo. The crude material was purified by flash col-
umn chromatography on silica gel (EtOAc/cyclohexane, 1:4) to give
the expected 1-THP-bromoindazoles 7a–b.
738 cmꢀ1 1H NMR (CDCl3) d: 1.57 (t, J = 7.0 Hz, 3H, CH3), 4.48 (q,
.
J = 7.8 Hz, 2H), 7.50 (t, J = 8.8 Hz, 1H), 7.79 (d, J = 7.8 Hz, 1H), 8.14
(d, J = 6.8 Hz, 1H), 8.61 (s, 1H). 13C NMR (CDCl3) d: 15.0, 44.4,
116.0, 117.1, 118.1, 125.2, 132.6, 140.6, 140.8. LRMS/ESI: [M+H]+
192. LRMS/EI m/z (%): 192 (23), 191 (M+, 100), 190 (4), 177 (17),
176 (100), 163 (24), 146 (5), 145 (36), 144 (6), 133 (15), 131 (6),
130 (53), 129 (8), 119 (25), 118 (27), 117 (20), 116 (9), 105 (15),
103 (16), 102 (11). HRMS/EI m/z calcd for C9H9N3O2 [M+H]+:
192,0773; found: 192,0776.
5.1.4.1.
(7a).
4-Bromo-1-(tetrahydro-2H-pyran-2-yl)-1H-indazole
Starting from the 4-bromo-1H-indazole 2a (2 g,
10.1 mmol) and following the general procedure in ethyl acetate,
the product 7a was obtained as a brown solid (2.3 g, 82%). Mp
72 °C. TLC Rf = 0.50 (Silica gel; EtOAc/cyclohexane 1:3). IR (KBr):
3435, 2963, 2944, 2850, 2866, 1611, 1567, 1493, 1446, 1418,
1249, 1170, 1079, 1059, 1040, 914, 772 cmꢀ1 1H NMR (CDCl3) d:
.
2.18-2.22 (m, 5H), 2.50-2.56 (m, 1H), 3.71-3.77 (m, 1H), 4.00-
4.07 (m, 1H), 5.72 (dd, J = 6.6 Hz and J’ = 2.7 Hz, 1H), 7.24 (t,
J = 8.1 Hz, 1H), 7.32 (d, J = 7.3 Hz, 1H), 7.55 (d, J = 8.3 Hz, 1H),
8.04 (s, 1H). 13C NMR (CDCl3) d: 22.4, 25.1, 29.4, 67.4, 85.7,
109.4, 114.5, 124.1, 125.5, 127.4, 133.9, 141.1. LRMS/EI m/z (%):
283 (38), 281 (M+, 38), 240 (98), 238 (100), 196 (21), 117 (21).
5.1.3.2. 2-Ethyl-4-nitro-1H-indazole (4b).
Starting from the
4-nitro-1H-indazole 1 (0.5 g, 3.1 mmol) and following the general
procedure with a purification by flash column chromatography
on silica gel thanks to CH2Cl2, the product 4b was obtained as a yel-
low solid (0.1 g, 20%). Mp 84 °C. TLC Rf = 0.28 (Silica gel; CH2Cl2). IR
(KBr): 1521, 1499, 1429, 1340, 1295, 1235, 1149, 797, 737 cmꢀ1. 1H
NMR (CDCl3) d: 1.70 (t, J = 6.8 Hz, 3H), 4.57 (q, J = 6.8 Hz, 2H), 7.40
(t, J = 7.8 Hz, 1H), 8.09 (d, J = 8.8 Hz, 1H), 8.18 (d, J = 7.8 Hz, 1H),
8.58 (s, 1H). 13C NMR (CDCl3) d: 15.7, 49.2, 114.9, 120.5, 123.6,
124.2, 125.9, 140.6, 149.8. LRMS/ESI: [M+H]+ 192. LRMS/EI m/z
(%): 192 (11), 191 (M+, 100), 176 (18), 163 (39), 145 (12), 133
(14), 105 (10). HRMS/EI m/z calcd for C9H9N3O2 [M]+: 191.0695;
found: 191.0693.
HRMS/EI m/z calcd for
280.022.
C
12H13BrN2O [M]+: 280.0211; found:
5.1.4.2.
(7b).
5-Bromo-1-(tetrahydro-2H-pyran-2-yl)-1H-indazole
Starting from the 5-bromo-1H-indazole 2b (0.7 g,
3.7 mmol) and following the general procedure in chloroform,
the product 7b was obtained as an orange oil (0.7 g, 70%). TLC
Rf = 0.35 (Silica gel; EtOAc/cyclohexane 1:8). IR (KBr): 3429, 3100,
2942, 2859, 1733, 1611, 1486, 1441, 1418, 1275, 1210, 1179,
1116, 1082, 1043, 994, 910, 801, 783 cmꢀ1 1H NMR (CDCl3) d:
.
5.1.3.3. 1-Ethyl-4-bromo-1H-indazole (6a).
Starting from
1.75-1.77 (m, 3H), 2.07-2.09 (m, 2H), 2.50-2.52 (m, 1H), 3.72-
3.77 (m, 1H), 4.00-4.03 (m, 1H), 5.69 (dd, J = 6.6 Hz and
J’ = 2.7 Hz, 1H), 7.46 (dd, J = 8.1 Hz and J’ = 1.7 Hz, 1H), 7.50 (d,
J = 8.8 Hz, 1H), 7.87 (d, J = 1.7 Hz, 1H), 7.96 (s, 1H). 13C NMR (CDCl3)
the 4-bromo-1H-indazole 2a (1 g, 5.1 mmol) and following the
general procedure with a purification by flash column chromatog-
raphy on silica gel thanks to EtOAc/cyclohexane, 1:2, the product

E. Lohou et al. / Bioorg. Med. Chem. 20 (2012) 5296–5304
5303
d: 22.4, 25.1, 29.3, 67.4, 85.5, 111.7, 114.3, 123.4, 126.2, 129.6,
133.0, 138.2. LRMS/EI m/z (%): 282 (M+, 9), 280 (M+, 9), 198 (73),
196 (77), 117 (24), 90 (26), 85 (100), 84 (85), 76 (13). HRMS/EI
m/z calcd for C12H13BrN2O [M]+: 280.0211; found: 280.0198.
was obtained as a pinkish solid (0.06 g, 76%). Mp 132 °C. TLC
Rf = 0.26 (Silica gel; EtOAc/cyclohexane 1:2). IR (KBr): 3434, 3165,
2925, 2854, 2228, 1615, 1370, 1266, 1089, 1031, 949, 784, 756 cm
ꢀ1
.
1H NMR (DMSO-d6) d: 7.51 (t, J = 6.8 Hz, 1H), 7.83 (d, J = 6.8 Hz,
1H), 7.94 (d, J = 8.8 Hz, 1H), 8.39 (s, 1H), 13.72 (s, 1H). 13C NMR
(DMSO-d6) d: 102.1, 116.1, 117.5, 122.2, 126.0, 127.0, 131.7,
139.6. LRMS/ESI: [M+H]+ 144. Anal. calcd for C8H5N3 (%): C,
67.13; H, 3.52; N, 29.35; found: C, 67.51; H, 3.50; N, 29.24.
5.1.6.2. 1H-Indazole-5-carbonitrile (9b). Starting from the 1-
(tetrahydro-2H-pyran-2-yl)-1H-indazole-5-carbonitrile 8b (0.1 g,
0.4 mmol) and following the general procedure, the product 9b
was obtained as a beige solid (0.05 g, 79%). Mp 177 °C. TLC
Rf = 0.18 (Silica gel; EtOAc/cyclohexane 1:2). IR (KBr): 3435, 3132,
5.1.5. General procedure for the synthesis of 1-(tetrahydro-2H-
pyran-2-yl)-1H-indazole-4 and 5-carbonitriles (8a–b)
The chosen 1-THP-bromoindazole 7a–b dissolved in DMF/water
99:1 (5 mL), SPhos (0.1 equiv), Pd2(dba)3 (0.04 equiv) and zinc cya-
nide (1.13 equiv) were combined in a microwave tube fitted with a
crimped septum cap. The tube was evacuated and then filled with
argon and stirred at room temperature for 15 min. The reaction
mixture was heated in a microwave at 150 °C for 40 min. After
evaporating DMF in vacuo, EtOAc (30 mL) was added to the residue
and the organic layer was successively washed with a 1 N sodium
hydroxide solution (3 ꢃ 15 mL) and brine (3 ꢃ 15 mL), dried over
MgSO4, filtered and concentrated in vacuo. The crude material
was purified by flash column chromatography on silica gel
(EtOAc/cyclohexane, 1:4 to 1:3) to give the attempted 1-THP-inda-
zole-carbonitriles 8a-b.
2954, 2873, 2221, 1625, 1292, 1076, 951, 911, 801 cmꢀ1 1H NMR
.
(DMSO-d6) d: 7.70 (d, J = 8.8 Hz, 1H), 7.76 (d, J = 8.8 Hz, 1H), 8.31
(s, 1H), 8.44 (s, 1H), 13.62 (s, 1H). 13C NMR (DMSO-d6) d: 102.7,
111.8, 119.9, 122.4, 127.7, 128.0, 134.8, 140.8. LRMS/ESI: [M+H]+
144. HRMS/EI m/z calcd for C8H5N3 [M]+: 143.0483; found:
143.0484.
5.2. In vitro biological evaluation
5.1.5.1. 1-(Tetrahydro-2H-pyran-2-yl)-1H-indazole-4-carboni-
trile (8a).
Starting from the 4-bromo-1-(tetrahydro-2H-pyr-
Effects of the different previously synthetized indazole deriva-
tives on nNOS activity were evaluated in vitro on rat cerebellum
homogenates. IC50 were determined from the NOS inhibition
curves constructed with five concentrations (0.01, 0.1, 1, 10 and
an-2-yl)-1H-indazole 7a (0.3 g, 1 mmol) and following the
general procedure, the product 8a was obtained as a yellow solid
(0.2 g, 79%). Mp 88 °C. TLC Rf = 0.28 (Silica gel; EtOAc/cyclohexane
1:4). IR (KBr): 2913, 2855, 2226, 1604, 1449, 1374, 1225, 1080,
100
conversion of
viously described method (1 mM CaCl2, 200
-arginine, 0.12
-[3H]arginine, 15 min at 37 °C).34,36,47 nNOS
l
M). Enzymatic activities were assayed by monitoring the
-[3H]arginine to -[3H]citrulline according to a pre-
M b-NADPH, 0.88
1037, 967, 916, 787 cmꢀ1 1H NMR (CDCl3) d: 1.69–1.81 (m, 3H),
.
L
L
2.04–2.14 (m, 2H), 2.49–2.55 (m, 1H), 3.76 (t, J = 7.8 Hz, 1H), 3.99
(d, J = 11.7 Hz, 1H), 5.78 (dd, J = 9.3 Hz and J’ = 2.9 Hz, 1H), 7.45
(t, J = 6.8 Hz, 1H), 7.56 (d, J = 6.8 Hz, 1H), 7.89 (d, J = 7.8 Hz, 1H),
8.21 (s, 1H). 13C NMR (CDCl3) d: 22.1, 25.0, 29.3, 67.3, 85.9,
104.1, 115.5, 117.2, 124.7, 126.0, 127.1, 132.1, 139.2. LRMS/ESI:
[M+H]+ 228. LRMS/EI m/z (%): 228 (16), 227 (M+, 100), 199 (10),
168 (8), 156 (9), 144 (45), 143 (38). HRMS/EI m/z calcd for
l
lM
L
lM L
basal activity represented 120 15 pmol citrulline formed/mg pro-
tein/h and the already described lead compound, 7-NI (dissolved in
DMSO 0.4%), exhibited an IC50 value of 0.6 0.2 lM (mean SEM;
n = 5 experiments).
C
13H13N3O [M]+: 227.1059; found: 227.1053.
5.3. Molecular modeling
5.1.5.2. 1-(Tetrahydro-2H-pyran-2-yl)-1H-indazole-5-carboni-
trile (8b). Starting from the 5-bromo-1-(tetrahydro-2H-pyr-
For each docked compound a preliminary calculation on its pro-
tonation state at pH 7.4 was carried out using standard tools of the
ChemAxon Package (http://www.chemaxon.com/) and the major-
ity microspecie (corresponding to among 94–100% for different
compounds) at this pH was used for docking studies.
The crystallographic coordinates of rat nNOSox used for docking
studies were obtained from X-ray structure of the 3-bromo-7-NI/
nNOSox complex (PDB ID 1OM5, a structure refined to 2.3 Å with
an R factor of 23.1%).
Docking of compounds into nNOSox was carried out with the
GOLD program (v 5.0) using the default parameters.40 This pro-
gram applies a genetic algorithm to explore conformational spaces
and ligand binding modes. To evaluate the proposed ligand poses
the ChemScore fitness function was applied in the docking studies.
The binding site in the nNOSox was defined as a 10 Å sphere cen-
tered on the heme iron atom.
an-2-yl)-1H-indazole 7b (0.3 g, 1 mmol) and following the
general procedure, the product 8b was obtained as an orange solid
(0.2 g, 79%). Mp 85 °C. TLC Rf = 0.31 (Silica gel; EtOAc/cyclohexane
1:3). IR (KBr): 2926, 2222, 1616, 1502, 1428, 1209, 1081, 1042,
994, 911, 808 cmꢀ1 1H NMR (CDCl3) d: 1.62-1.77 (m, 3H), 2.10-
.
2.15 (m, 2H), 2.50-2.55 (m, 1H), 3.77 (t, J = 9.8 Hz, 1H), 4.02 (d,
J = 11.7 Hz, 1H), 5.76 (dd, J = 9.3 Hz and J’ = 2.9 Hz, 1H), 7.59 (d,
J = 8.8 Hz, 1H), 7.71 (d, J = 8.8 Hz, 1H), 8.12 (s, 1H), 8.13 (s, 1H).
13C NMR (CDCl3) d: 22.2, 25.0, 29.4, 67.4, 85.8, 104.8, 111.6,
119.5, 124.3, 127.4, 128.6, 134.5, 140.3. LRMS/ESI: [M+H]+ 228.
HRMS/EI m/z calcd for C13H13N3O [M]+: 227.1059; found:
227.1057.
5.1.6. General procedure for the synthesis of 1H-indazole-4 and
5-carbonitriles (9a–b)
Molecular mechanics calculations on the nNOSox/indazole
derivative complexes were performed using CHARMM program
with potential function parameter set 22.48 The Discovery Studio
program was used to derive CHARMM force field parameters for
the indazole derivatives applying MMFF partial charges.49 During
all calculations all atoms that lay 8 Å or further from any indazole
derivative atoms were fixed.
For all energy minimizations, 500 steps of steepest descent, fol-
lowed by 1000 steps of conjugate gradient and by n steps of
Adopted Basis Newton-Raphson minimization were carried out un-
til an average energy gradient, during a cycle of minimization, less
than 10–3 kcal/(mol Å).
Indazole 8a or 8b was introduced in ethanolic HCl (5–15 mL)
and the preparation was stirred for 3–6 h at room temperature.
Then, the acidic mixture was neutralized thanks to a 2 N sodium
hydroxide solution and ethanol was evaporated in vacuo. After
addition of EtOAc (20 mL) to the residue, the organic layer was
washed with brine (3 ꢃ 10 mL), dried over MgSO4, filtered and
evaporated in vacuo. Finally, the expected unprotected indazole-
carbonitriles 9a–b were in this case directly isolated.
5.1.6.1. 1H-Indazole-4-carbonitrile (9a).
Starting from the 1-
(tetrahydro-2H-pyran-2-yl)-1H-indazole-4-carbonitrile 8a (0.1 g,
0.5 mmol) and following the general procedure, the product 9a

5304
E. Lohou et al. / Bioorg. Med. Chem. 20 (2012) 5296–5304
25. Maccallini, C.; Patruno, A.; Lannutti, F.; Ammazzalorso, A.; De Filippis, B.;
Acknowledgment
Fantacuzzi, M.; Franceschelli, S.; Giampietro, L.; Masella, S.; Felaco, M.; Re, N.;
Amoroso, R. Bioorg. Med. Chem. Lett. 2010, 20, 6495.
We thank Dr. Rémi Legay for the mass spectrometry measure-
ments. We thank the CRIHAN (Centre de Ressources Informatiques
de Haute-Normandie) and the European Community (FEDER) for
the molecular modelling softwares.
26. Claramunt, M. R.; Lopez, C.; Perez-Medina, C.; Perez-Torralba, M.; Elguero, J.;
Escames, G.; Acuna-Castroviejo, D. Bioorg. Med. Chem. 2009, 17, 6180.
27. Cottyn, B.; Acher, F.; Ramassamy, B.; Alvey, L.; Lepoivre, M.; Frapart, Y.; Stuehr,
D.; Mansuy, D.; Boucher, J.-L.; Vichard, D. Bioorg. Med. Chem. 2008, 16, 5962.
28. Jaroch, S.; Hölscher, P.; Rehwinkel, H.; Sülzle, D.; Burton, G.; Hillmann, M.;
McDonald, F. M. Bioorg. Med. Chem. Lett. 2003, 13, 1981.
29. Xue, F.; Fang, J.; Lewis, W. W.; Martasek, P.; Roman, L. J.; Silverman, R. B. Bioorg.
Med. Chem. Lett. 2010, 20, 554.
30. Maddaford, S.; Renton, P.; Speed, J.; Annedi, S. C.; Ramnauth, J.; Rakhit, S.;
Andrews, J.; Mladenova, G.; Majuta, L.; Porreca, F. Bioorg. Med. Chem. Lett. 2011,
21, 5234.
References and notes
1. Bryan, N. S.; Bian, K.; Murad, F. Front Biosci. 2009, 14, 1.
2. Salerno, L.; Sorrenti, V.; Di Giacomo, C.; Romeo, G.; Siracusa, M. A. Curr. Pharm.
Des. 2002, 8, 177.
3. Arzumanian, V.; Stankevicius, E.; Laukeviciene, A.; Kevelaitis, E. Medicina 2003,
39, 535.
4. Alderton, W. K.; Cooper, C. E.; Knowles, R. G. Biochem. J. 2001, 357, 593.
5. Förstermann, U.; Münzel, T. Circulation 2006, 113, 1708.
6. Villanueva, C.; Giulivi, C. Free Radical Biol. Med. 2010, 49, 307.
7. Brüne, B. Pharmacol. Res. 2010, 61, 265.
8. Toda, N.; Ayajiki, K.; Okamura, T. Pharmacol. Rev. 2009, 61, 62.
9. Vincent, S. R. Prog. Neurobiol. 2010, 90, 246.
10. Schmidtko, A.; Tegeder, I.; Geisslinger, G. Trends Neurosci. 2009, 32, 339.
11. Xu, L.; Mabuchi, T.; Katano, T.; Matsumura, S.; Okuda-Ashitaka, E.; Sakimura,
K.; Mishina, M.; Ito, S. Nitric Oxide 2007, 17, 18.
12. Freire, M. A. M.; Guimaraes, J. S.; Gomes-Leal, W.; Pereira, A. Front. Neurosci.
2009, 3, 175.
13. Overeem, K. A.; Ota, K. T.; Monsey, M. S.; Ploski, J. E.; Schafe, G. E. Front. Behav.
Neurosci. 2010, 4, 1.
14. Cheng, B.-C.; Chang, C.-P.; Lin, M.-T.; Lee, C.-C. Neuropharmacology 2007, 52,
297.
31. Annedi, S. C.; Maddaford, S. P.; Ramnauth, J.; Renton, P.; Speed, J.; Rakhit, S.;
Andrews, J.; Porreca, F. Bioorg. Med. Chem. Lett. 2012, 22, 1980.
32. Crestey, F.; Lohou, E.; Stiebing, S.; Collot, V.; Rault, S. Synlett 2009, 4, 615.
33. Lohou, E.; Collot, V.; Stiebing, S.; Rault, S. Synthesis 2011, 16, 2651.
34. Schumann, P.; Collot, V.; Hommet, Y.; Gsell, W.; Dauphin, F.; Sopkova, J.;
MacKenzie, E.; Duval, D.; Boulouard, M.; Rault, S. Bioorg. Med. Chem. Lett. 2001,
11, 1153.
35. Collot, V.; Sopkova-de Oliveira Santos, J.; Schumann, P.; Colloc’h, N.;
MacKenzie, E.; Rault, S. J. Enz. Inh. Med. Chem. 2003, 18, 195.
36. Boulouard, M.; Schumann-Bard, P.; Butt-Gueulle, S.; Lohou, E.; Stiebing, S.;
Collot, V.; Rault, S. Bioorg. Med. Chem. Lett. 2007, 17, 3177.
37. Li, H.; Martasek, P.; Masters, B. S.; Poulos, T. L.; Raman C. S. To be published
(PDB ID: 1OM5).
38. 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.
39. Sopkova-de Oliveira Santos, J.; Collot, V.; Rault, S. Acta Cryst. C 2000, 56, 1503.
40. Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. J. Mol. Biol. 1997, 267,
727.
41. Bouissane, L.; El Kazzouli, S.; Leonce, S.; Pfeiffer, B.; Rakib, E. M.; Khouili, M.;
Guillaumet, G. Bioorg. Med. Chem. Lett. 2006, 14, 1078.
42. Cheung, M.; Boloor, A.; Stafford, J. A. J. Org. Chem. 2003, 68, 4093.
43. Lu, Y.; Wang, Y.; Xu, Z.; Yan, X.; Luo, X.; Jiang, H.; Zhu, W. J. Phys. Chem. B 2009,
113, 12615.
44. Baumli, S.; Endicott, J. A.; Johnson, L. N. Chem. Biol. 2010, 17, 931.
45. Chobanian, H. R.; Fors, B. P.; Lin, L. S. Tetrahedron Lett. 2006, 47, 3303.
46. Sheldrick, G. M. Acta Cryst. 2008, A64, 112.
47. Dwyer, M. A.; Bredt, D. S.; Snyder, S. H. Biochem. Biophys. Res. Commun. 1991,
176, 1136.
15. Moro, M. A.; Cardenas, A.; Hurtado, O.; Leza, J. C.; Lizasoain, I. Cell Calcium 2004,
36, 265.
16. Martinez-Murillo, R.; Fernandez, A. P.; Serrano, J.; Rodrigo, J.; Salas, E.;
Mourelle, M.; Martinez, A. Neurosci. Lett. 2007, 415, 149.
17. Singh, S.; Dikshit, M. Brain Res. Rev. 2007, 54, 233.
18. Chung, K. K. K.; David, K. K. Nitric Oxide 2010, 22, 290.
19. Fidana, I.; Yuksel, S.; Ymir, T.; Irkeç, C. Nur Aksakal F. J. Neuroimmunol. 2006,
171, 184.
20. Chandra, A.; Enkhbaatar, P.; Nakano, Y.; Traber, L. D.; Traber, D. L. Clinics 2006,
61, 71.
21. Seo, J.; Martasek, P.; Roman, L. J.; Silverman, R. B. Bioorg. Med. Chem. 2007, 15,
1928.
22. Erdal, E. P.; Martasek, P.; Roman, L. J.; Silverman, R. B. Bioorg. Med. Chem. 2007,
15, 6096.
23. Lee, K.-S.; Lee, D.-K.; Jeoung, D.; Lee, H.; Choe, J.; Ha, K.-S.; Wond, M.-H.; Kwon,
Y.-G.; Kim, Y.-M. Biochem. Biophys. Res. Commun. 2012, 418, 49.
24. Litzinger, E. A.; Martasek, P.; Roman, L. J.; Silverman, R. B. Bioorg. Med. Chem.
2006, 14, 3185.
48. Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; Sates, D. J.; Swaninathan, S.;
Karplus, M. J. Comput. Chem. 1983, 4, 187.
49. MacKerell, J. A. D.; Bashford, D.; Bellott, M.; Dunbrack, J. R. L.; Evanseck, J. D.;
Field, M. J.; Fischer, S.; Gao, J.; Guo, H.; Ha, S.; Joseph-McCarthy, D.; Kuchnir, L.;
Kuczera, K.; Lau, F. T. K.; Mattos, C.; Michnick, S.; Ngo, T.; Nguyen, D. T.;
Prodhom, B.; Reiher, I. W. E.; Roux, B.; Schlenkrich, M.; Smith, J. C.; Stote, R.;
Straub, J.; Watanabe, M.; Wiorkiewicz-Kuczera, J.; Yin, D.; Karplus, M. J. Phys.
Chem. B 1988, 102, 3586.