N -[6-(4-Butanoyl-5-methyl-1 H -pyrazol-1-yl)pyridazin-3-yl]-5-chloro-1-[2-(4-methylpiperazin-1-yl)-2-oxoethyl]-1 H -indole-3-carboxamide (SAR216471), a Novel Intravenous and Oral, Reversible, and Directly Acting P2Y12 Antagonist

Article abstract of DOI:10.1021/jm500588w

In the search of a potential backup for clopidogrel, we have initiated a HTS campaign designed to identify novel reversible P2Y12 antagonists. Starting from a hit with low micromolar binding activity, we report here the main steps of the optimization process leading to the identification of the preclinical candidate SAR216471. It is a potent, highly selective, and reversible P2Y12 receptor antagonist and by far the most potent inhibitor of ADP-induced platelet aggregation among the P2Y12 antagonists described in the literature. SAR216471 displays potent in vivo antiplatelet and antithrombotic activities and has the potential to differentiate from other antiplatelet agents.

Full text of DOI:10.1021/jm500588w

Article
pubs.acs.org/jmc
N‑[6-(4-Butanoyl-5-methyl‑1H‑pyrazol-1-yl)pyridazin-3-yl]-5-chloro-
1-[2-(4-methylpiperazin-1-yl)-2-oxoethyl]‑1H‑indole-3-carboxamide
(SAR216471), a Novel Intravenous and Oral, Reversible, and Directly
Acting P2Y12 Antagonist
Christophe Boldron,*,† Angelina Besse,† Marie-Francoise Bordes,† Stephanie Tissandie,† Xavier Yvon,†
̧
́ ́
́
Benjamin Gau,† Alain Badorc,† Tristan Rousseaux,† Guillaume Barre,† Jerome Meneyrol,† Gernot Zech,∥
́
́
̂
Marc Nazare,⊥ Valerie Fossey,‡ Anne-Marie Pflieger,† Sandrine Bonnet-Lignon,† Laurence Millet,†
́
Christophe Briot,§ Frederique Dol,† Jean-Pascal Herault,† Pierre Savi,† Gilbert Lassalle,†
́
́
́
Nathalie Delesque,† Jean-Marc Herbert,† and Francoise Bono†
̧
†Sanofi R&D, 195 Route d’Espagne, 31036 Toulouse Cedex, France
‡Sanofi R&D, 1, Avenue Pierre Brossolette, 91385 Chilly-Mazarin Cedex, France
§Sanofi R&D, 371 Rue du Professeur Joseph Blayac, 34184 Montpellier, France
∥Sanofi R&D, Industriepark Hoechst, 65926 Frankfurt am Main, Germany
⊥AG Medizinische Chemie, Leibniz-Institut fur Molekulare Pharmakologie, 13125 Berlin, Germany
̈
S
* Supporting Information
ABSTRACT: In the search of a potential backup for clopidogrel, we
have initiated a HTS campaign designed to identify novel reversible
P2Y12 antagonists. Starting from a hit with low micromolar binding
activity, we report here the main steps of the optimization process
leading to the identification of the preclinical candidate SAR216471.
It is a potent, highly selective, and reversible P2Y12 receptor
antagonist and by far the most potent inhibitor of ADP-induced
platelet aggregation among the P2Y12 antagonists described in the
literature. SAR216471 displays potent in vivo antiplatelet and
antithrombotic activities and has the potential to differentiate from
other antiplatelet agents.
ADP-stimulated platelet activation of the glycoprotein (GP)
IIb/IIIa.2
Historically, thienopyridines (ticlopidine and clopidogrel)
INTRODUCTION
■
Platelets play a major role in the formation of arterial
thrombosis originating from a lesion of the vascular wall or
the rupture of an atherosclerotic plaque. Indeed they are
activated by multiple factors such as the release of endogenous
mediators from damaged endothelial cells, the exposure of
thrombogenic molecules of the extracellular matrix of the
damaged vessel, or high shear stress from stenosed vessels.
After activation, platelets adhere and aggregate at the site of the
vascular lesion, causing thrombi formation. Adenosine 5′-
diphosphate (ADP) liberated from damaged red blood cells or
endothelial cells is a key mediator of activation and aggregation
of platelets. ADP could bind two purinergic receptors P2Y1 and
P2Y12. P2Y1 is implicated in platelet shape changes, whereas
P2Y12 is involved in platelet aggregation, thrombin generation,
microparticles release, and thrombus stabilization.1 Although
binding of ADP to both receptors is required for complete
platelet aggregation, P2Y12 is the major receptor involved in
were reported as the first orally active ADP antagonists (Figure
1). They were shown to display high levels of antithrombotic
activity.3−5 Following the cloning of the P2Y12 receptor,6,7
clopidogrel (FDA-approved in 1997) has been further
demonstrated to bind irreversibly to the receptors through a
thiol active metabolite following two oxidative biotransforma-
tions. The therapeutic success of these P2Y12 receptor
inhibitors in preventing atherothrombotic events in patients
after an experience of acute coronary syndrome (ACS),
especially those undergoing percutaneous coronary interven-
tion (PCI), has definitely provided evidence that P2Y12
antagonism is a key therapeutic target in the management
and prevention of arterial thrombosis.8 Prasugrel (Figure 1),9
Received: April 15, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Reference P2Y12 antagonists.
approved in 2009, is a prodrug analogue of the first
biotransformed intermediate of clopidogrel, therefore requiring
only one oxidative step to generate a thiol active metabo-
lite.10,11 Similar to clopidogrel, prasugrel activity is exerted
through an irreversible binding to the P2Y12 receptor.
Ticagrelor, approved in 2011 for coronary artery disease
(CAD) and ACS, is the result of an impressive program of
medicinal chemistry aiming at transforming ATP, the natural
antagonist of ADP receptors, into an orally active, reversible
antagonist of P2Y12 (Figure 1).12 More recently, elinogrel
(Figure 1), originating from a HTS hit and further
optimization, was developed as both iv and po antithrombotic
agent until phase II studies, but its clinical development was
unfortunately stopped in 2012.13 Although clopidogrel is still a
treatment of choice for most of arterial diseases, we were
wondering whether an iv/po reversible antagonist displaying a
fast onset of action with less susceptibility for interindividual
variability and less bleeding risk would complement clopidogrel
treatments.
(compound A2) showed a slight improvement in the ADP-
induced platelet aggregation assay on washed platelets (IC50
=
4.5 μM). An additional optimization run was then initiated
(about 350 new molecules synthesized). Compound A3,
bearing a butanoic side chain connected through an alkoxy
group at the ortho position of the urea linker, revealed some
key structural features that allowed marked activity improve-
ments in the Bnd assay (Figure 2, IC50 = 80 nM). Then further
improvements were made possible by inserting a methyl group
in the para position of the anisole group. In the last part of the
optimization run, the metabolically labile pyrazole-4-carboxylic
ethyl ester was replaced by a pyrazole-4-butan-1-one motif, thus
leading to the identification of compound A4, a potent and
metabolically stable P2Y12 antagonist (Bnd, IC50 = 17 nM),
which was nominated as “intermediate lead”. However,
significant potency loss was observed with intermediate lead
A4 in the ADP-induced platelet aggregation assays (will be
referred as to AA in the text) in rat and human platelet rich
plasma (will be referred as to rPRP and hPRP, respectively).
Always bearing in mind our goal to identify orally
bioavailable P2Y12 antagonists, we have attempted simple
modifications such as monomethylation of both urea nitrogen
atoms, replacement of the urea by up and down amide,
acetamide, 2-oxo-acetamide, or 2-aminoimidazole linkers with-
out success (Figure 2). An important SAR element was the loss
of activity while replacing the oxygen atom from anisole by a
methylene group, thus making this o-aminoanisole group
mandatory for reaching a high level of activity (compounds
A3 and A4). We made the hypothesis that the molecule could
be locked in an active conformation because of a H-bonded
five-membered ring between the urea NH and the oxygen atom
in the ortho position (Figure 3). Such a H-bond stabilization
was already demonstrated by Etter et al. after X-ray diffraction
of monocrystals of o-alkoxybiphenylurea derivatives.14 To
experiment with this hypothesis, we simply replaced the o-
aminoanisole by an indole connected to the carbonyl group via
the 1-position (48) or via the 3-position (47). We were
delighted to see that these modifications did not affect the
biological activities (Figure 3). As 48 was shown to be sensitive
to degradation in mild basic and acidic conditions, we decided
to focus our next efforts around compound 47. Importantly,
this successful rescaffolding reduced the number of H-bond
donors and acceptors, making this new platform a good starting
Starting from a HTS hit with micromolar binding activity, we
report here how we improved in vitro and in vivo antiplatelet
activities together with several key ADME parameters, thus
leading to the identification of a preclinical candidate.
RESULTS AND DISCUSSION
■
Medicinal Chemistry Optimization. (Compound code
numbers are depicted in Table 1.) A high throughput screening
(HTS) campaign of the Sanofi compound collection was set
using as primary assay the measurement of the down regulation
of cAMP in the presence of Forskoline on CHO-CRE-Luc
stimulated by 2-methylthio-ADP. Hits were confirmed using a
binding assay involving two partners: radiolabeled thiomethyl-
ADP and P2Y12-transfected CHO cells (will be referred as to
Bnd assay in the text). Among the most potent hits, a 4-
pyrazol-1-ylbenzamide chemical series was identified (Figure
2). The best representative member of the family was
compound A1 (HTS hit) that displayed micromolar IC50 in
the binding assay. By use of this scaffold, novel substructure and
similarity searches were undertaken within the corporate
database as well as commercially available collections in order
to perform a backscreening aiming to test close analogues of
HTS hits that were not included in the HTS core list. A
backscreening hit bearing a urea in place of the amide linker
B
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
appeared to be deleterious for biological activities. Interestingly,
methyl insertion at C-6 positions (compound 50c), while
decreasing binding activity, was very much favorable regarding
the functional activity in both species. Surprisingly, the quite
acceptable and nonspecies dependent functional activity of 46a,
ester analogue of 50a, was unexpected given its modest IC50 of
binding. This result confirmed the role played by the carboxylic
acid group in term of both functional potency and species
discrepancy. When we were working on COOH-containing
urea derivatives, we had observed a significant decrease of the
binding activities by a factor of 5−50 when performing binding
experiments in the presence of 80 g/L human serum albumin
(HSA, mean normal blood concentration of ∼50 g/L),15 thus
revealing a strong capacity to bind plasma proteins. In this
regard and when compounds 50a and 46a are compared,
methyl ester removes and inhibits the interaction of free
carboxylic acid to serum albumin. In order to avoid bias related
to binding experiments conditions, we decided to make the lead
selection based on the in vitro functional activity (AA).
Table 1. Compound Code Numbers
Thus, 50c was considered as the new lead compound, as it
allowed reaching for the first time submicromolar activities in
the in vitro AA assay on hPRP. Because of the carboxylic acid
group, compounds 50a and 50c were found to be metabolically
stable in rat and human microsomal fractions but, by contrast,
not permeable on Caco2 cells (Table 3). As shown with
compound 46a, the carboxy group seemed not to be absolutely
mandatory for the functional activity. As we did not want to
move further through the optimization of an ester prodrug, we
exploited this finding by undertaking the preparation of amide
derivatives. Behaving in the same manner as 46a, compound 51
(dimethylcarboxamide analogue of 50c) exhibited submicro-
molar functional activity in hPRP despite a significant loss of
activity in the binding assay when compared to 50c (Table 3).
Interestingly, primary carboxamide and monomethylcarbox-
amide analogues displayed similar activities as 51 (Bnd, 31 and
47 nM respectively; AA on hPRP, 2.0 and 1.0 μM,
respectively). These results clearly confirmed carboxamide
derivatives as good candidates for carboxylic acid replacement.
In addition to retaining the functional activity, 51 also showed a
high level of permeation on Caco2 cells which was promising as
part of our aim to identify orally bioavailable compounds.
Unfortunately, 51 was found to be metabolically unstable on
microsomal fractions. In order to address the low solubility of
51, we then prepared the methylpiperazine analogue 52c. Very
interestingly, 52c appeared to be biologically more active than
51, quite permeable, but still sensitive to oxidative metabolism.
Suspecting benzylic positions to represent metabolic hot spots,
we removed one methyl group to generate a potent P2Y12
antagonist (compound 52a) displaying more acceptable levels
of oxidative metabolism in rat and human species and, because
of the piperazinyl motif, a large increase of aqueous solubility
up to 0.76 mg/mL (Table 3). 52a displayed a medium in vitro
permeability on Caco2 cells and was appointed as pharmaco-
logical tool for a preliminary in vivo evaluation. In the
meantime and building on these promising results achieved
with the methylpiperazine motif, we have prepared a series of
carboxamide derivatives also bearing a basic nitrogen atom
whose role was to ensure an acceptable solubility. The
representative and nonexhaustive list of compounds synthe-
sized is given at Table 4 and could be divided in three
subgroups: subgroup 1, compounds 53−63 ⇒ carboxamides
connected through a noncyclic amino group; subgroup 2,
compounds 64−77 ⇒ nonpiperazinyl-containing carboxamides
point on the path to orally bioavailable compounds. The next
stage of the optimization process was to improve the functional
activity. Interestingly, decreasing the length of the COOH-
bearing side chain of 47 by one methylene (49) or two
methylene groups (50a) dramatically improved biological
activities (Figure 4). However, a significant discrepancy should
be noted when comparing functional activities in humans and
rats. A “methyl scan” of 50a (IC50 values: Bnd 0.9 nM; AA
(hPRP) 1.8 μM; AA (rPRP) 9.7 μM) was performed in order
to better apprehend the active conformation and/or the key
points of interaction with the P2Y12 receptor (Table 2).
Methylation at the amide linker or at C-2 and C-4 positions of
the indole motif (compounds 50b, 50e, and 50d, respectively)
C
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
b
Figure 2. HTS phenylpyrazole hit series. Footnotes for the table are the following: P2Y12 binding (n = 2); ADP-induced platelet aggregation
([ADP] = 2.5 μM, n = 2), chuman platelet rich plasma (hPRP).
connected through a cyclic amino group; subgroup 3,
compounds 52a and 78−97 ⇒ piperazinylcarboxamide-
containing derivatives. All these compounds have been
biologically evaluated using “Bnd” and “AA in rPRP” assays.
Indeed, as mentioned before with compound 46a, carboxamide
derivatives did not display species discrepancies anymore (see
Table 3). Therefore, evaluation on rat samples (instead of
humans) was preferred for obvious reasons of plasma
accessibility. The selection of potential in vivo pharmacological
tools was done using an empirical value that we called the
eligibility factor (EF). EF was the result of the multiplication of
“Bnd” in nM by “AA in rPRP” in μM and reflected the balance
between functional and binding activities. We posed the rule
that compounds to be selected for in vivo evaluation should
display EF values below 20. This simple equation was born
from the observation that displaying submicromolar functional
activities was not a sufficient condition for a compound to be
active in vivo. Indeed, it had to be associated with a good
enough activity of binding. This rule has been tested and
challenged with success right from the start of the advanced
program phase. One typical example for illustration is
compound 85 which was found to be inactive ex vivo (15%
2) after a 10 mg/kg oral administration. This compound
displayed a high level of functional activity (IC50 = 0.4 μM), an
ADME profile very similar to those of ex vivo active
compounds displayed in Figure 5 but a relatively low activity
of binding (IC50 = 82 nM).
Among the 46 compounds displayed in Table 4, it appeared
very clearly that piperazinyl-containing derivatives were very
much preferred, as no member from subgroups 1 and 2 were
elected. Ethylenediamine-, aminopyrrolidine-, and aminopiper-
idine-containing derivatives all displayed high IC50 values of
binding that were not compensated by good enough AA in
rPRP, thus generating high EF. In addition, there seem not to
be any hydrogen bonding to the receptor in this region, as
shown by the similar levels of activity between non-, mono- and
dimethylated amines. Only compound 77 was found to be
borderline with an EF of 23. It was very interesting to note that
within subgroups 1 and 2, the 2-methyloctahydropyrrolo[3,4-
c]pyrrole motif from 77 was by far the most analogous to the
methylpiperazine motif from 52a in terms of “3D shape”. This
reinforced our desire to go further with piperazinyl-containing
derivatives. Removing the methyl from the methylpiperazine
motif (90), increasing the length from methyl to ethyl and n-
butyl (52a, 78, 79), or inserting ramified or cyclic alkyl groups
(82, 83, 84, and 85) was not beneficial. The insertion of an
D
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. o-Aminoanisole rescaffolding. Footnotes for the table are the following: achemical structure as shown below table; bP2Y12 binding (n = 2);
cADP-induced platelet aggregation ([ADP] = 2.5 μM, n = 2) in rat and human platelet rich plasma (rPRP and hPRP, respectively).
Table 2. “Methyl Scan” around Compound 50a
b
b
AA
AA
a
Bnd
(μM) in (μM)
hPRP in rPRP
compd
R1
R2
R4
R6
Z
(nM)
50a
50b
50e
50d
50c
46a
H
H
H
H
OH
OH
OH
OH
OH
OMe
0.9
304
84
1.8
>30
15
9.7
>30
16
Me
H
H
H
H
Figure 4. Modification of the COOH-bearing side chain. Footnotes
for the table are the following: aP2Y12 binding (n = 2); bADP-induced
platelet aggregation ([ADP] = 2.5 μM, n = 2) in rat and human
platelet rich plasma (rPRP and hPRP, respectively).
Me
H
H
H
H
Me
H
H
272
5.1
93
>30
0.4
4.1
>30
4.6
3.5
H
H
Me
H
H
H
H
a
b
P2Y12 binding (n = 2). ADP-induced platelet aggregation (n = 2,
[ADP] = 2.5 μM) in rat and human platelet rich plasma (rPRP and
hPRP, respectively).
ethyl side chain bearing one or two heteroatoms (80, 81, 91,
and 92) led to very potent compounds displaying EFs down to
6. The trifluoroethyl analogue 93 was found to be much less
active. From the set of compounds bearing one or two methyl
groups at the α position of the basic nitrogen (86, 87, 88, 89)
or a methylene group bridging the 2- and 5-piperazinyl
positions (95 and 96), only compound 87 was elected. Very
potent compounds were also identified bearing either a 3-
methyloctahydropyrrolo[1,2-a]pyrazin-7-ol motif (94) or a p-
fluorobenzyl group (97).
At this point in the optimization and regarding the high level
of in vitro functional activity reached, we moved to an in vivo
proof of concept (POC). Seven new compounds, in addition to
52a, met the EF rule (shaded lines in Table 4). We first
checked their liability on rat microsomal fractions, and as they
all displayed low to medium level of oxidative metabolism, we
decided to evaluate them in our standard ex vivo inhibition of
platelet aggregation model after an oral dosing of 10 mg/kg in
rats (Figure 5). Three out of 8 tested compounds (81, 91, and
97) exhibited high levels of ex vivo inhibition in good
correlation with their low EF values. 52a and 87 were found
to be borderline in term of oxidative metabolic stability on rat
microsomal fractions which may justify their lower ex vivo
activity in rats. Surprisingly, 92 was found to be much less
active in vivo than 91 despite that it was as active in vitro and it
displayed similar metabolic and permeability parameters. This
E
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 3. SAR Optimization: From Carboxylic Acid to Carboxamide Derivatives
a
b
P2Y12 binding (n = 2). ADP-induced platelet aggregation (n = 2, [ADP] = 2.5 μM) in rat and human platelet rich plasma (rPRP and hPRP,
c
d
respectively). Permeability measured on a Caco2 cells monolayer (expressed in 10−7 cm·s−1). iIncubation with hepatic microsomal fractions during
20 min in the presence of 1 mM NADPH (e.g., 5% metabolism means for 95% of unchanged compound).
Table 4. First Part of the Carboxamide Optimization (5-Methylindole Platform)
a
b
c
P2Y12 binding (n = 2). ADP-induced platelet aggregation (n = 2, [ADP] = 2.5 μM) in rat platelet rich plasma (rPRP). Eligibility factor = Bnd
(nM) × AA in rPRP (μM). Shaded cells correspond to compounds selected for in vivo evaluation, as they display EF values of <20.
might be the result of a decrease in solubility consequent to the
presence of the second fluorine atom that decreased the
piperidine pKa. However, we could not exclude the possibility
that 91 ex vivo activity was further enhanced because of an
irreversible binding to the receptor either through a direct
substitution of the aliphatic C−F bond or via the intermediary
formation of an electrophilic aziridinium (short duration of in
vitro assays would limit the formation of such a covalent
adduct). Considering this hypothesis, 92 was much less
amenable to such an irreversible binding. In doubt, we excluded
this fluoroethyl motif from any further investigations following
the initial specifications to identify a reversible P2Y12
antagonist. As suggested by their low Caco2 permeation values,
the capacity of compounds 80 and 94 to cross the
gastrointestinal barrier was probably hampered by the presence
of the hydroxyl group (Figure 5).
Unfortunately, 81 and 97 displayed a high level of oxidative
metabolism in human microsomal fractions, not making them
suitable for further investigations. However, this encouraging ex
vivo POC with 81, 91, and 97 comforted us in regard to the
translatability of in vitro results into ex vivo activities after oral
administration. We then focused our efforts through the
identification of oxidatively more stable compounds. Since the
most obvious metabolic hot spot shared by these last
compounds was the benzylic position, we undertook the
optimization of the indole substitution pattern. We decided to
proceed using a 4-methylpiperazinylcarboxamide platform,
since this motif appeared to provide the best balance between
F
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 5. Ex vivo inhibition of platelet agregation in rats 2 h after a 10 mg/kg oral dosing. Footnotes for the table are the following: aadministration
b
of a 10mg/kg oral dose; incubation with hepatic microsomal fractions during 20 min in the presence of 1 mM NADPH (e.g., 41% metabolism
c
means for 59% of unchanged compound); permeability measured on Caco2 cells monolayer (expressed in 10−7 cm·s−1).
Table 5. Indole Optimization
% metabo-
lism on
microsomal
e
fractions
a
Bnd
Caco2
permeation
ex vivo
f
b
b
c
d
compd
R5
R6
(nM)
AA (μM) in human PRP AA (μM) in rat PRP EF
human rat
inhibition
52c
52a
52g
52h
52l
Me
Me
Me
H
39
0.5
0.5
0.4
0.5
0.5
1.1
0.7
0.5
0.3
0.5
0.5
0.4
0.3
0.4
0.4
0.6
2.3
1.7
2.0
3.7
2.7
5.4
5.9
12
16
11
11
14
19
20
20
50
39
68
36
31
34
15
36
52
41
63
14
21
20
14
23
59
40
0
37
5
3
31
33
OMe
Cl
H
22
ND
50
H
27
16
11
28
9
5
Cl
F
48
inactive
52m
52n
52o
52p
52q
52r
52s
52t
Br
Me
H
48
21
26
31
4
8
2
CF3
Cl
50
Me
H
34
36
CN
Me
Me
Cl
76
175
200
276
366
383
Br
Cl
Cl
H
118
138
99
F
142
52u
52p′
H
Me
(CO)NH2
H
a
b
P2Y12 binding (n = 2). ADP-induced platelet aggregation (n = 2, [ADP] = 2.5 μM) in human or rat platelet rich plasma (hPRP or rPRP,
c
d
respectively). Eligibility factor = Bnd (nM) × AA in rPRP (μM). Permeability measured on a Caco2 cells monolayer (expressed in 10−7 cm·s−1).
e
Incubation with hepatic microsomal fractions during 20 min in the presence of 1 mM NADPH (e.g., 52% metabolism means for 48% of unchanged
compound). Ex vivo inhibition of platelet aggregation in rats (%), 2 h after a 10 mg/kg oral dosing.
f
oxidative metabolism and Caco2 permeation. The nonexhaus-
tive list of modified indole-containing derivatives is given in
Table 4. We followed the same decision tree as before (AA in
rPRP, yes/no → Bnd, yes/no → AA in hPRP, yes/no → ex
vivo when EF ≤ 20). Substituents such as a fluorine atom
(52t), a nitrile (52p), or a primary carboxamide (52p′) at
position 5 were not tolerated. Moving the methyl group from
position 5 in 52a to position 6 in 52u was detrimental, which
was somehow unexpected because the 5,6-dimethylindole
derivative 52c was found to be active. The combination of a
G
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 6. Second Part of the Carboxamide Optimization (5-Chloroindole Platform)
a
b
P2Y12 binding (n = 2). ADP-induced platelet aggregation (n = 2, [ADP] = 2.5 μM) in human or rat platelet rich plasma (hPRP or rPRP,
c
d
respectively). Eligibility factor = Bnd (nM) × AA in rPRP (μM). Permeability measured on a Caco2 cells monolayer (expressed in 10−7 cm·s−1).
e
Incubation with hepatic microsomal fractions during 20 min in the presence of 1 mM NADPH (e.g., 36% metabolism means for 64% of unchanged
compound).
methyl group at position 5 with a halogen atom at position 6
(52q and 52r) or the combination of two chlorine atoms (52s)
produced less active compounds. Interestingly, the methyl/
halogen combination the other way around (52m and 52o) was
much tolerated. Although it was found to be potent in vitro, the
5-OMe derivative 52g was not evaluated in vivo, as it displayed
both a high level of metabolic transformation on rat
microsomal fractions and a low Caco2 permeation. Five out
of six new in vitro active compounds displaying EF below 20
(52h, 5-Cl; 52l, 5-Cl, 6-F; 52m, 5-Br, 6-Me; 52n, 5-CF3; 52o,
5-Cl,6-Me) were found to be metabolically more stable than
52a, especially on rat microsomal fractions. We also observed a
slight loss of permeation capability on Caco2 cells. Although we
met our goal to identify in vitro active compounds less
susceptible to oxidative metabolism, stability on human
microsomal fractions was acceptable but not markedly reduced.
We then evaluated them in the ex vivo inhibition of platelet
aggregation model (2 h, 10 mg/kg oral, Table 5). Very
interestingly, the 5-chloroindole derivative 52h was found to be
significantly more active than compound 52a in the same
conditions and was shown to respond in a dose dependent
manner with an ED50 of 11.1 mg/kg (data not shown). The
other five compounds did not show any improvement and
remained stuck below 30% inhibition of ex vivo platelet
aggregation. Very interestingly, 52h was also found to be active
in this ex vivo model after iv administration (30 min, I% (1 mg/
kg) = 53 18, I% (3 mg/kg) = 70 8, and I% (10 mg/kg) =
86 5). Before moving forward, we were willing to know more
about the metabolic behavior of 52h in the human context and
we were very pleased to see a major improvement from a high
in vitro hepatic clearance on human hepatocytes with 52a
(0.150 mL/h) to a moderate one with 52h (0.100 mL/h). The
metabolic hot spots were revealed after incubation on human
microsomal fractions and human hepatocytes. Three main
metabolites were produced in equal proportions (UV−LCMS
detection) and were the result of (i) indole monohydroxylation,
(ii) piperazine demethylation, and (iii) n-propyl side chain
oxidation. Despite 52h displaying a potent and long-lasting ex
vivo activity together with an acceptable projection in term of
liver metabolism in the human context, we decided to operate
some chemical modifications aiming to reduce further the
metabolic liability. The optimization of the indole part of the
molecule was discussed above (Table 5), and we already
concluded that 5-chloroindole was the best motif. We then
focused on the methylpiperazinyl motif, and leveraging our
previous experience around compound 52a, we prepared more
than 100 piperazine-containing carboxamide derivatives.
Compounds 98−104 were found to be very much active in
vitro (EF values down to 2) and were selected as representative
examples, as they illustrate the difficulty we faced to adjust the
permeation to metabolism balance within this chemical series
(Table 6). Compound 98, bearing a methoxyethyl side chain,
displayed a high level of Caco2 permeation together with a
good metabolic stability on rat microsomal fractions leading to
a strong (60−70% platelet aggregation inhibition) and long-
lasting ex vivo activity (plateau up to 6 h after administration).
Unfortunately, this compound appeared to be much too
metabolized on human microsomal fractions (47% of liability)
and human hepatocytes (hepatic clearance of 0.124 mL/h) for
being further investigated. 101 and 102 were found to be
particularly potent after oral administration (72% and 80%
inhibition, respectively, 2 h after administration) but were not
H
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
selected for the same reasons as 98. As expected, the oxidative
metabolism was largely reduced while moving from methoxy to
hydroxy group (compound 99 = main metabolite formed after
incubation of 98 on microsomal fractions). However, as a
consequence of the polarity increase, we observed the
concomitant loss of Caco2 permeation and oral ex vivo activity.
Similar to 99, compound 100 bearing an acetamide fragment
was found to be inactive ex vivo. The two isolated enantiomers
from 102 displayed very similar in vitro biological activities,
Caco2 permeation, and metabolic stability. Indeed, both were
found to be very potent in vitro, confirming that the
stereochemistry in this region of the molecule did not affect
the interaction with the receptor. Very interestingly, the D-
enantiomer 103 was found to be much less potent in vivo than
its L-counterpart 104, revealing an impact of the stereo-
chemistry on the compound bioavailability that was apparently
not related to permeation and oxidative metabolism. As a
conclusion and similar to the work done around 52a, we had to
admit that the efforts to replace the methylpiperazine motif had
been vain.
At this stage of the optimization, it appeared clearly that 52h
was the best compromise between in vivo potency/duration of
action in rats and predicted metabolism in humans. Although
several compounds have shown higher levels of ex vivo
antiaggregant activity in rats (compounds 81, 91, 97, 98, 101,
and 104), we systematically had to face an extensive metabolic
liability on human liver components.
Projecting farther to the required conditions necessary before
considering a development candidate, we came to debate about
the presence of an aniline motif that could be released upon
cleavage of the central carboxamide linker of 52h in biological
media. This aniline fragment was found to be positive in the
Ames assay after metabolic activation.16 It is very important to
note that 52h was found to be negative in the same assay with
or without metabolic activation. Many ethical, biological, and
technical challenges did not encourage us to try to demonstrate
whether this aniline fragment and its mutagenic metabolite(s)
were generated or not in vivo. Indeed, there is no threshold
value for mutagenicity, which means that a single molecule
could be responsible for it. The diversity and the chemical
reactivity of the expected metabolites make it even more
complex. In view of that, it appeared inconceivable to reach
conclusions that would be strong enough to convince ourselves
and competent authorities about the long-term safety of
compound 52h. Therefore, we decided to initiate a second
MedChem optimization phase aiming to get rid of this aniline
fragment.
We already knew from previous studies in the original
biphenylurea series but also during some works around
compound 50a that the biological activity was very sensitive
to the replacement of the 1-(5-methyl-1H-pyrazol-4-yl)butan-1-
one motif. As a result, we were not prone to make dramatic
modifications and we finally focused on simple ones among
which the most representative are depicted at Table 7.
As generally observed, the aniline fragment was not found to
be mutagenic in the absence of metabolic activation. Because
bacteria are unable to metabolize chemicals like CYP does, a
key component for making the Ames test useful in that
particular case is the inclusion of S9 microsomal fractions as an
exogenous mammalian metabolic activation system.17 Indeed,
to exert their mutagenic potential, aromatic amines or their
acetylated analogues (from aniline acetylation by acetyl CoA)
have to be transformed into reactive electrophiles (Scheme 1).
The first metabolic step leading to these reactive species is a
CYP-mediated N-oxidation that produces hydroxylamine
intermediates that can undergo N−O bond cleavage to form
arylnitrenium ions. Alternatively, further bioactivation of the N-
hydroxylamine derivatives to N-acetoxy or N-sulfate esters
makes the N−O bond heterolysis easier and/or the DNA-
adducts formation through a SN2 reaction with DNA
possible.18 Then the nitrenium ion is trapped by DNA, where
it binds covalently through nitrenium N atom directly or
through the C atom at the ortho position, essentially at guanine
nucleic bases (Scheme 1).19 These genetic alterations might
then result in mutagenesis and carcinogenesis.
Several QSAR equations displaying accuracies of about 80−
90% have been drawn from the large aniline mutagenicity and
carcinogenicity databank. The identification of some key
molecular determinants came from these intensive investiga-
tions.20 The probability of inducing mutagenicity was found to
be closely related to physicochemical factors that modulate the
oxidation capacity of the amino group which is the first step on
the course to the generation of the arylnitrenium (Scheme 1).
The most relevant factors appeared to be the hydrophobicity,
the energy of the highest occupied orbital (HOMO) and of the
lowest unoccupied orbital (LUMO), and the steric hindrance
ortho to the amino group. Hydrophobicity is involved in the
absorption and the transport of the drugs in cells (hepatocytes
for instance) and organisms, as well as in the interaction
between drugs and the metabolizing enzymes. The electronic
Table 7. Optimization of Pyrazole Substitution
%
metabolism
on
microsomal
d
fractions
a
b
Bnd
AA (μM) in rat
c
compd
R1
(nM)
PRP
EF
11
human rat
52h
52i
52j
52k
CH2CH3
CH3
27
0.4
0.6
0.7
8
36
32
40
6
14
25
47
3
336
143
201
H
100
CH2CF3
574
4592
a
b
P2Y12 binding (n = 2). ADP-induced platelet aggregation (n = 2,
c
[ADP] = 2.5 μM) in rat platelet rich plasma (rPRP). Eligibility factor
= Bnd (nM) × AA in rPRP (μM). Incubation with hepatic
d
microsomal fractions during 20 min in the presence of 1 mM NADPH
(e.g., 36% metabolism means for 64% of unchanged compound).
Shortening of the alkyl side chain (52i and 52j) appeared
detrimental for the biological activity and did not protect from
oxidative metabolism. Very interestingly, the insertion of three
fluorine atoms at the extremity of the n-propyl side chain (52k)
produced a compound much less sensitive to oxidative
metabolism. Unfortunately, this modification was not appli-
cable, as 52k appeared to be inactive.
I
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Scheme 1. General Mechanisms Leading to Aniline Mutagenesis/Carcinogenesis
Table 8. Replacement of the Aniline Fragment: Mutagenicity Data
a
Measured using the Ames II Salmonella assay in the presence of S9 fractions (all compounds were found “not mutagenic” without S9 fractions).
parameters are measures of the chemical reactivity, hence of the
ability of undergoing oxidative metabolic transformations.
Electron releasing substituents on the aromatic amine increase
the electronic density on the aniline nitrogen, making it more
susceptible to oxidation, polarize the hydroxylamine bond,
favoring its heterolytic cleavage, and stabilize the nitrenium ion
formed thus increasing the risk of mutagenicity. Having this in
mind, we built a MedChem plan aiming to identify new and
safe chemical fragments in replacement of the mutagenic
aniline.
The insertion of nitrogen atoms within the aryl group was
performed with the objective to decrease the lipophilicity of the
fragment (Table 8, compounds 20w, 20x, 20y, and 20ag). 3-
Aminopyridine 20ag was found to be mutagenic, while
J
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 9. Replacement of the Aniline Fragment: (a) In vitro Biological Activities of Compounds Not Bearing a Mutagenic
Fragment and (b) Human Antiaggregant Activity and eADME Parameters of Compounds 52w, 52z, 52ad, and 52ae
a
b
c
P2Y12 binding (n = 2). ADP-induced platelet aggregation (n = 2, [ADP] = 2.5 μM) in rat platelet rich plasma. Eligibility factor = Bnd (nM) ×
d
e
AA in rPRP (μM). ADP-induced platelet aggregation (n = 2, [ADP] = 2.5 μM) in human platelet rich plasma. Incubation with hepatic microsomal
f
fractions during 20 min in the presence of 1 mM NADPH (e.g., 29% metabolism means for 71% of unchanged compound). Permeability measured
g
on a Caco2 cell monolayer, expressed in 10−7 cm·s−1). In vitro metabolism using human hepatocytes (incubation of 24 h), expressed in mL·h‑1·106
hep)−1.
aminopyridazine, 2-aminopyridine, and aminopyrazine deriva-
tives were not. Hence, the proximity of the N atom to the
amino group seemed mandatory. The aminothiazole 20ah
which might be considered as an 2-aminopyridine bioisoster
was found to be mutagenic. We did not find any correlation
between mutagenicity and calculated physicochemical param-
eter for these five fragments. The replacement of the phenyl
group of 20h by a trans-cyclohexyl (20ad) or a piperidine motif
(20ae) obviously prevented mutagenicity from the mechanisms
described at Scheme 1. 20ad is an aliphatic amine and was not
tested in the Ames assay. As some hydrazine derivatives were
described in the literature to be mutagenic,21 20ae was
challenged in the Ames II assay and was found to be negative
both with and without metabolic activation. Aniline replace-
ment by a phenol group produced the nonmutagenic 20af
fragment. 20ac was prepared as a bulky secondary amine that
was indeed not found to be mutagenic. The insertion of ortho-
substituents aiming to modulate both the polarity and the
oxidability of the aniline fragment was done. Polar and/or
electron withdrawing substituents such as nitrile (20z),
carboxylic acid (20aa), or primary carboxamide (20ab) allowed
prevention of mutagenicity, while the somehow lipophilic
chlorine atom insertion (20aj) produced a mutagenic aniline.
20ai was found to be mutagenic, while 20ab was not. This may
be due to the role played by the two methyl groups which
increase the electronic density on the aromatic system together
with the lipophilicity. At this stage, we were pleased to have in
hand 10 new nonmutagenic fragments that were all engaged in
the total synthesis of P2Y12 antagonists 52w to 52af (Table
9a). The synthesis of the 2-aminopyridine-containing final
derivative was not completed because in that specific case, the
central amide linker appeared to be sensitive in slightly acidic
and basic conditions. Five compounds out of nine (Table 9a,
52y, 52aa, 52ab, 52ac, and 52af) were found to be inactive in
both Bnd and rat AA assays. The four remaining compounds
containing pyridazine (52w), o-benzonitrile (52z), trans-
cyclohexyl (52ad), or piperidine (52ae) motifs, were elected
for being evaluated further, as they displayed EF values below
20. It must be noted that this optimization aiming to replace
the mutagenic aniline fragment was also very successful in
terms of in vitro biological activity. Indeed, 52w and 52ae were
the best methylpiperazine-containing compounds tested
throughout the project, with respective EF values of 2.6 and
0.6 (Table 9a). As already observed with piperazine-containing
actives, there was no discrepancy between human and rat
functional activities (Table 9b). With IC50 values below 100
nM, 52w, 52ad, and 52ae are by far the most potent in vitro
P2Y12 antagonist ever described in the literature. The four
selected compounds were shown to be stable on rat and mouse
microsomes. However, 52z, 52ad, and 52ae displayed high
levels of oxidative metabolism on human microsomes. These
results were well correlated with the high clearance on human
hepatocytes measured for 52z and 52ad (0.143 and 0.171 mL/
h, respectively). Interestingly, 52ae was much less cleared than
expected (0.050 mL/h) and was finally selected on these
criteria as well as 52w that was found fine on human
microsomes and hepatocytes. Additionally, the contribution
of CYP3A4 to the hepatic clearance was found to be low for
both compounds. Despite low permeations measured on Caco2
cells, we decided to perform a dose−response study with 52w
K
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 6. Dose−response study: (a) ex vivo inhibition of platelet agregation in rats 2 h after oral dosing with 1, 3, and 10 mg/kg 52w, 52ae, and
ticagrelor; (b) plasma concentrations measured for 52w and 52ae in the corresponding ex vivo experiment.
Figure 7. Kinetic study: (a) ex vivo inhibition of platelet agregation in rats over time after oral dosing of 10 mg/kg 52w and 3 mg/kg ticagrelor; (b)
plasma concentrations measured for 52w in the same ex vivo experiment.
and 52ae in our ex vivo platelet aggregation model in rats
(Figure 6a). This was done in comparison with ticagrelor, 2 h
after dosing. 52w and 52ae inhibited ADP-induced platelet
aggregation in a dose-dependent manner with ED50 values of
2.75 and 0.95 mg/kg, respectively. Nearly complete inhibition
of platelet aggregation (85%) was observed with 52w at 30 mg/
kg (data not shown). 52ae retained a significant level of activity
at very low dose (∼50% at 1 mg/kg), thus making it a very
potent antiplatelet agent. Ticagrelor exhibited an ED50 of 1.59
mg/kg, while clopidogrel and prasugrel displayed ED50 values
of 2.60 and 0.46 mg/kg in the same conditions, respectively.22
Plasma concentrations were determined from samples collected
during this dose−response study. In the case of 52w, the
circulating levels were very well correlated with both the level
of antiaggregating activity and the dose (Figure 6b). On the
other hand, the ex vivo activity of 52ae appeared to be high
regarding the very low drug plasma concentrations. This kind of
PK/PD profile may suggest a covalent mode of binding to the
L
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of Key Substituted 3-Carboxyindoles 3 and 4
a
(a) (CF3CO)2O, diethyl ether, −5 °C, 83−99%; (b) Br2, AcOH, rt, 81%; (c) KOH in H2O, reflux, 62−99%; (d) Et3N, DMAP, acetyl chloride,
DCM, rt, 56−98%; (e) NaH (2 equiv), DMF, −10 °C to rt, then methyl bromoacetate, −20 °C to rt, 61−92%; (f) dimethylformamide di-tert-
butylacetal, benzene, reflux, then benzyl bromoacetate, cesium carbonate, DMF, rt, 66% for two steps; (g) LiOH, THF/H2O, 0 °C, quantitative; (h)
EDC·HCl, pentafluorophenol, DMF, rt, then N-ethylmorpholine, 1-methylpiperazine, rt, 58%; (i) TFA, DCM, rt, quantitative. (j) For R2 = H: LiCl,
[1,4]benzoquinone, Pd(OAc)2 cat., benzyl acrylate, THF, rt, 51−98%. For R2 = Me: APTS, 3-oxobutyric acid benzyl ester, toluene, reflux in a
Dean−Stark apparatus, 88%. (k) For R2 = H: 1,4-diazabicyclo[2,2,2]octane, Pd(OAc)2 cat., DMF, 120 °C, 22−70%. For R2 = Me:
tri(otolyl)phosphine, Et3N, Pd(OAc)2 cat., acetonitrile, 140 °C under microwave, 97%. (l) K2CO3, methyl- or tert-butyl bromoacetate, DMF, rt, 96−
98%; (m) Pd/C cat., ammonium formate, MeOH, reflux, 58−88%; (n) Pd/C, NH2NH2·H2O, cyclohexane, 0 °C, then Et3N, acetyl chloride, 10 °C,
t
37%; (o) K2CO3, MeOH, rt, 80%; (p) DMAP, propynoic acid benzyl ester, THF, 0 °C, then aq NaOH, rt, 40% (6c, 29%); (q) BuOK,
Br(CH2)3COOEt, DMF, rt, 90%; (r) 2-methyl-2-butene, NaClO2, NaH2PO4, H2O/THF/tBuOH, 0 °C to rt, 88%.
receptor or the formation of hyperactive metabolites, the latter
option seeming unlikely as 52ae was shown to be stable on rat
microsomes (Table 9). These uncertainties around 52ae,
especially regarding the initial specifications we had imple-
mented (reversible and directly acting P2Y12 antagonist), did not
encourage us to move further in its in vivo characterization, and
we decided to focus on compound 52w. 52w was also found to
be active in this ex vivo model after iv administration (5 min
after injection, I% (0.1 mg/kg) = 37 22, I% (0.3 mg/kg) = 63
10, I% (1 mg/kg) = 75 7, I% (3 mg/kg) = 86 2, and I%
(10 mg/kg) = 92 6).
As a next step, we performed a kinetic study after a 10 mg/kg
oral administration of 52w that resulted in a rapid maximal
inhibition of platelet aggregation (less than 1 h, Figure 7a). The
extent of inhibition was maintained at relatively constant levels
during several hours with an average percent of platelet
inhibition of 74% during the first 6 h. Again, a high correlation
was found between the extent of ex vivo inhibition of ADP-
induced platelet aggregation and the circulating levels of 52w
over time (r = 0.861, data not shown).22 Very interestingly, the
ex vivo potency of 52w to inhibit ADP-induced platelet
aggregation was also in excellent correlation with its in vitro
M
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 3. Synthesis of Synthons 20h−k, 20w, 20x, and 20ad
a
(a) DMF−DMA, 2 **h, rt, 99%; (b) EtOH, reflux, 2 h, 72−76%; (c) LiOH or NaOH or KOH, THF/EtOH/H2O, reflux, 74−99%; (d) CDI, THF,
rt, 20 h, then magnesium bis(3-ethoxy-3- oxopropanoate), 55 °C, 59−92%; (e) NaH 60%, THF, 0 °C then R1X, rt, 24h, 82−96%, or K2CO3, TBAB,
R1X, THF, 55 °C, 82−97%, or Cs2CO3, EtI, DMF, rt, 99%; (f) Aqueous HCl 37%, reflux, 2 h, 78−100%, or Al2O3, 1,4-dioxane/H2O, 100 °C, 82%;
(g) Pd/C, cyclohexene, EtOH, reflux, 57−95%, or H-Cube, EtOAc/MeOH, 57%; (h) 4- methoxybenzylamine, 1,4-dioxane, reflux, 2 h, 72%; (i)
KOH, EtOH, 80 °C, 2 h, 98%; (j) DMAP, Et3N, di-tert-butyl dicarbonate, DMF, rt, then aqueous KOH, rt, 56%; (k) Pd(OAc)2, Cs2CO3, MePhos,
benzhydrylidenehydrazine, tBuOH, 80 °C, 74%; (l) aqueous HCl 37%, rt, 72%; (m) tert-butyl carbazate, MeOH, rt, quant; (n) NaBH3CN, AcOH/
H2O, rt, then HCl in EtOAc, rt, 63%; (o) HCl 4 M in 1,4-dioxane, rt, 90%.
potency in rats (IC50 = 62 nM, Figure 7b). A nearly identical
profile of ex vivo inhibition was observed with ticagrelor in the
same experimental conditions at a 3 mg/kg oral dose (Figure
7a) which was in good correlation with its plasma circulating
levels (r = 0,836).22 By contrast, its ex vivo activity was not
consistent with its in vitro potency (AA in rPRP, IC50 = 4160
nM), thus suggesting that ticagrelor may exert its activity
through complementary mechanisms of action independent of
its ability to antagonize P2Y12 receptors. A preclinical study in
a rat shunt thrombosis model demonstrated a dose-dependent
antithrombotic activity after oral administration of 52w
together with a favorable safety profile as demonstrated in a
rat tail bleeding model.22
52w has demonstrated an excellent selectivity (IC50 > 10
μM) against a large panel of receptors, enzymes, and ion
channels (see Supporting Information).22 No toxicity was
achieved and no damaged target organ identified in a 7-day oral
(30 and 100 mg/kg) and intravenous (3 and 10 mg/kg)
toxicity study in rats. In this study the plasma exposure was
found to be in the same range after a single as well as after a 7-
day repeated administration (AUC(0−24)(day 7)/AUC(0−24)(day
1) ratio of 1.3), suggesting the absence of autoinduction or
accumulation process. In addition, 52w, orally administered at
the 100 mg/kg dose to conscious telemetered mongrel dogs,
did not induce any major hemodynamic or electrocardiographic
effects and did not affect body temperature within 24 h after
treatment. Mean maximal plasma concentration measured 4 h
after dosing reached 2340 388 ng/mL (4.2 μM).
Preparation of Compounds. (Code numbers are given at
Table 1.) Depending on the commercial availability of starting
materials, synthons 3 and 4 were prepared following four
different methods described in Scheme 2 (methods A−D).
Method A was the most convenient, as it allowed getting most
of synthon 3 or 4, starting from commercially available 5,6-
substituted indoles. The first two steps were adapted from a
previously published work and consisted of the selective
trifluoroacetylation at position 3 using trifluoroacetic anhydride
in diethyl ether, thus producing intermediates 1 followed by the
forced hydrolysis of the trifluoroacetyl group to produce
compounds 2.23 1q was obtained after bromination of 1a.
Acetylation of intermediates 2 in mild conditions led to
compounds 3a, 3l, 3q, 3t, and 3u (method A-1), while
activation with 2 equiv of NaH followed by the thermodynami-
cally controlled N-indole substitution with 1 equiv of methyl
N
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 4. Synthesis of Compounds 20v, 20z, 20aa, 20ab, 20ac, 20af, 20ai, 20aj, and 20ak
a
(a) NCS, acetonitrile, reflux, 88%; (b) H2SO4, NaNO2, H2O, 0 °C to reflux, 80%; (c) NBS, acetonitrile, reflux, 76%; (d) ZnCN2, Pd(PPh3)4, DMF,
100 °C, 85%; (e) NaOH, H2O, reflux, 87%; (f) KHCO3, CH3I, DMF, rt, 87%; (g) NaOH, 1,4-dioxane/H2O, 100 °C, 66%; (h) N-
methylmorpholine, BOP, DCM, rt, then Me2NH·HCl, rt, 58%; (i) (CF3CO)2O, DCM, rt, 97%; (j) NaH 60%, THF, 0 °C, then methyl
bromoacetate, rt, 99%; (k) K2CO3, MeOH, then aq 2 N NaOH, then CDI, Et3N, THF, then NaBH4, rt, 37%; (l) Cs2CO3, 2-(di-tert-
butylphosphino)-1,1′-binaphthyl, Pd(OAc)2, toluene, 70 °C, 68%.
bromoacetate produced compounds 4f, 4g, 4h, and 4s (method
A-2). It was very interesting to note here that this original
regiocontrolled addition prevented a sequence of cumbersome
protection/deprotection steps. Alternatively, compound 4h′
was prepared starting from compound 2h, through a sequence
of protection/alkylation with benzyl bromoacetate/deprotec-
tion steps (compounds 11 and 12) followed by the
methylpiperazine insertion (13) and tert-butyl ester depro-
tection (method A-3). This 4h′ synthon was very promising, as
it gave the potential to go directly to final methylpiperazine-
containing compounds 52. This method was used to produce
compound 52ae (Scheme 6). Unfortunately, 4h′ was not stable
enough in the stronger reaction conditions required to couple
aromatic amines, and this method could not be generalized.
Method B was already described in the literature and provided
compounds 4 from correctly substituted anilines.24 Commer-
cially available o-halogenoanilines were engaged in a Pd-
catalyzed reaction with benzyl acrylate to form compounds 5c
and 5n. 5e was prepared from 2-bromo-4-methylphenylamine
and 3-oxobutyric acid benzyl ester in the presence of p-
toluenesulfonic acid. Compounds 5 were then submitted to a
Pd-catalyzed cyclization, thus producing compounds 6c, 6e,
and 6n. Method C is another option, allowing the generation of
the fully acetylated N-hydroxylamine 8 from the corresponding
nitro derivative. It was followed by a selective O-deacetylation
to afford 9, which was then submitted to a one step Pd-
catalyzed addition/cyclization with benzyl propionate to form
6d.25 In the absence of ortho substituents, the cyclization was
allowed at both sides, thus additionally producing the 6c
isomer. Compounds 7 were the point of convergence of
methods B and C and were obtained by alkylation with alkyl
bromoacetates. Finally, a catalytic hydrogenation over Pd/C
produced the key synthons 4c, 4d, 4e, and 4n. The very first
method we used to produce 5-methylindole derivatives was
from indole-3-carbaldehyde, a nonreactive precursor of the
indole-3-carboxylic acid motif. Thus, this allowed the
straightforward N-alkylation with ethyl 4-bromobutyrate
producing 10, followed by aldehyde oxidation using the sodium
chlorite and 2-methylbut-2-ene couple to form 4a′ (method
D).26
Compounds 20h−k, 20w, 20x, and 20ad were prepared
following a chemical strategy that was already described in the
literature to produce a large variety of alkyl ketones from
carboxylic acid derivatives (Scheme 3).27 The common 4-
carboxypyrazole scaffold of 16h, 16w, 16x, and 16ad is key in
this synthetic method. 16h was prepared in three steps from 3-
O
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 5. Synthesis of Synthons 20y, 20ae, 20ag, and 20ah
a
(a) K2CO3, TrCl, DMF, rt, then KOH, EtOH/H2O, 56%; (b) DMAP, BOP-Cl, DCM, then N,O-dimethylhydroxylamine hydrochloride, rt, 87%;
(c) n-PrMgCl, THF, −30 °C to rt, 85%; (d) HCl 4 M in 1,4-dioxane, rt, 62%; (e) TBDMSCl, imidazole, DCM, rt, 82%; (f) LiBH4, diethyl ether, rt,
quantitative; (g) oxalyl chloride, DMSO, DCM, −78 °C, then Et3N, 0 °C, then tert-butyl carbazate, NaBH(OAc)3, DCM, 0 °C, 51%; (h) TBAF,
THF, 0 °C to rt, 78%; (i) DMAP, Et3N, MeSO2Cl, DCM, rt, quantitative; (j) 36, tBuOK, DMF, 50 °C, 10%; (k) TFA, DCM, rt, 80%; (l) 36, NaH
60%, DMF, rt, then Pd/C, H2 (2 bar), THF, rt, 26% for two steps; (m) pyridine, benzyl chloroformate, acetonitrile, rt, 93%; (n) 36, K2CO3, CuI,
trans-N,N′-bismethyl-1,2-cyclohexane diamine, 85 °C, then TFA, reflux, 11% for two steps; (o) 36, K2CO3, proline, CuI, DMSO, 130 °C, 18%.
oxobutyric acid ethyl ester. A first condensation between N,N-
dimethylformamide dimethyl acetal and 3-oxobutyric acid ethyl
ester produced 14. A second condensation with (4-
nitrophenyl)hydrazine led to 15 that was further hydrolyzed
to afford 16h. 16w was obtained in three steps from its
chloropyridazine analogue.28 Reaction with p-methoxybenzyl-
amine as the solvent at high temperature produced 21 which
was then hydrolyzed to form the carboxypyrazole 22. Finally, as
we did not succeed in elongating the carboxylic acid of 22, the
secondary aromatic amine was Boc-protected to afford 16w.
16x was obtained in four steps from commercially available 5-
bromo-2-nitropyridine that was first reacted with benzhydryli-
denehydrazine under Pd(II) catalysis (23) followed by HCl
deprotection to produce 24 in its hydrochloride form.
Condensation with 14 afforded pyrazole 25 that was then
hydrolyzed to give 16x. 16ad was obtained in five steps from
(4-oxocyclohexyl)carbamic acid benzyl ester that was first
condensed with tert-butyl carbazate to produce 26. The
reduction of 26 using sodium cyanoborohydride in acidic
conditions produced the racemic cis/trans hydrazine mixture.
Very interestingly, the pure trans isomer 27 was isolated in its
hydrochloride form by precipitaton upon addition of an organic
HCl solution to the racemic mixture dissolved in EtOAc. Then
Boc removal in HCl/1,4-dioxane (28) was followed by
condensation with 14, leading to 29 which was hydrolyzed to
afford 16ad. The elongation of the 4-carboxypyrazole
derivatives 16h, 16w, 16x, and 16ad was performed using a
magnesium salt of 3-ethoxy-3-oxopropanoate, leading to
compounds 17h, 17w, 17x, and 17ad, respectively. The
activation of the malonate proton with sodium hydride allowed
the insertion of various alkyl groups through reaction with
halogenoalkanes (compounds 18h, 18i, 18w, 18x, and 18ad),
followed by a decarboxylation/deprotection step in strong
acidic conditions producing compounds 19h, 19i, 19x, 20w,
and 20ad. The reaction of 17h with 1,1,1-trifluoro-2-iodo-
ethane was not complete. The purification appeared to be
difficult because of the coelution of the expected product and
the starting material on silica gel. Therefore, the decarbox-
ylation step was performed on the crude, leading to a mixture
of 19j and 19k which were easily separated on silica gel. Key
intermediates 20h, 20i, 20j, 20k, and 20x were obtained by
reduction of the nitro group using Pd/C either in batch with
cyclohexene as the source of hydrogen or with the H-cube
apparatus that generates molecular hydrogen in situ from the
electrolysis of water under flow conditions. 20h was a key
intermediate, as it allowed the preparation of a variety of aniline
derivatives (Scheme 4). Ortho-halogenation using N-chloro- or
N-bromosuccinimide led to 20aj and 20v, respectively.
Preparation of the diazonium ion from 20h followed by
heating in water produced phenol 20af. Trifluoroacetylation of
20v (30) followed by insertion of a methyl acetate chain led to
31. Trifluoroacetyl removal, methyl ester hydrolysis, and
reduction of the activated carboxylic acid using NaBH4 afforded
the alcohol 32, which was then engaged in a Pd(II)-catalyzed
P
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 6. Synthesis of Compounds 44−47 and 49−104
a
(a) DMAP, 0 °C, then EDC·HCl, DCM, rt, 28−98%; (b) K2CO3, MeOH, rt, 30−89%; (c) K2CO3, methyl bromoacetate, DMF, rt, 42−95%; (d)
EDC·HCl, DMAP, 1,2- dichloroethane, rt, 11−96% or SOCl2, DMF cat., DCM, reflux then 4 Å molecular sieves, DMAP, 1,2-dichloroethane, 80 °C,
24−72%; (e) aq NaOH, MeOH or 1,4-dioxane or MeOH/1,4-dioxane or THF, rt, 41−98%, or (46ak → 50ak) aq LiOH, THF, rt, 68%, or (46e →
50e) TFA, DCM, rt, 25%; (f) HNR8R9, Et3 N, BOP-Cl, DCM, rt, 15−97%, or Excess HNR8R9, BOP-Cl, DCM, rt, 17−95%, or Excess HNR8R9,
pyridine, BOP-Cl, DMF, rt, 69%; (g) Et3N, BOP-Cl, DMF, 50 °C, then aq NaOH in MeOH, rt, 29%; (h) ethyl acrylate, benzyltrimethylammonium
hydroxide, 1,4-dioxane, then aq NaOH, rt, 25%; (i) Br2, AcOH, rt, 41%; (j) NiCl2, DMF, 200 °C under microwave, 73−80%; (k) NaH, DMF, 4 °C,
then CH3I, rt, 27%; (l) ZnCN2, Pd(PPh3)4, DMF, 100 °C in a sealed tube, 82−84%; (m) AcOH/aq 35% HCl, 90 °C, 44%; (n) Boc-protected
HNR8R9, Et3N, BOP-Cl, DCM, rt, then HCl in 1,4-dioxane/EtOAc/MeOH, rt, 15−89%; (o) piperazine-2-carboxylic acid methyl ester, Et3N, BOP-
Cl, DCM, rt, then tert-butyl methyl(2-oxoethyl)carbamate, sodium cyanoborohydride, AcOH, MeOH, rt, then TFA, rt, then aq NaOH, 57%; (p)
separation on a chiral column using supercritical conditions; (q) K2CO3, CH3I, DMF, rt, 13−78%; (r) aq LiOH, THF, rt, 60%; (s)
diisopropylethylamine, HOBt, EDC·HCl, DMF, rt, 33%.
cyclization step to form the morpholine motif of 20ac. 20v was
also the starting point leading to 20z through a Pd(II)-catalyzed
cyanation. Mild hydrolysis of 20z produced the primary
carboxamide 20ab, while stronger hydrolysis conditions led to
the carboxylic acid 20aa, which was either coupled with
dimethylamine to form 20ai or methylated to afford ester 20ak.
Another key synthon is pyrazole 36 which was prepared in
four steps from 5-methyl-1H-pyrazole-4-carboxylic acid methyl
ester (Scheme 5). Trityl insertion followed by methyl ester
hydrolysis led to 33 which was then activated through the
formation of the corresponding Weinreb amide 34. Reaction
with n-propylmagnesium chloride (35) followed by trityl
removal produced compound 36. Intermediate 20ae was
prepared in seven steps from 3-hydroxypentanedioic acid
dimethyl ester. Silylation (37) followed by ester reduction with
lithium borohydride produced 38. Alcohol oxidation under
Swern conditions followed by a double reductive amination
with tert-butyl carbazate in the presence of sodium triacetox-
yborohydride afforded 39. TBDMS removal (40) followed by
alcohol activation led to mesylate 41 which was reacted with 36
to produce 42. Reaction yield was low, as the formation of the
position isomer was sterically favored (alkylation at the second
pyrazole nitrogen). TFA-mediated BOC-deprotection finally
afforded compound 20ae. Pyrazole 36 was also reacted with 5-
Q
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
bromopyrazin-2-ylamine to afford 20y or with (5-bromothiazol-
2-yl)carbamic acid benzyl ester (43) to produce 20ah or with
2-fluoro-5-nitropyridine to produce a nitropyridine intermedi-
ate that was then reduced to 20ag.
surrogate in A4 (Figure 2). A clear breakthrough was the
successful rescaffolding of an o-aminoanisole into an indole
motif and the ensuing identification of compound 50a.
Carboxylic acid to carboxamide shift was also very decisive, as
it allowed reaching submicromolar functional activities together
with the limitation of species discrepancies. Carboxamide
derivatives showed a much higher level of permeation on Caco2
cells which was promising as part of our aim to identify orally
bioavailable compounds. After optimization, 52h was identified
that displayed a high level of in vivo antiplatelet activity. The
last sequence of the optimization process, aiming to get rid of a
mutagenic aniline fragment that we decided to avoid, led to
52w. 52w is a potent, highly selective, and reversible P2Y12
receptor antagonist (binding, 17 nM; in vitro antiaggregant
activity, 60 nM in rat and 100 nM in human). 52w is directly
active and does not require any metabolic activation which was
a prerequisite set in stone from the beginning of the project.
52w displays an ADME-T profile consistent with its MPK
development. Ex vivo studies showed a rapid onset of action
(within the first hour after oral administration) and a long
duration of antiplatelet activity after either intravenous or oral
administration. This duality in the mode of administration
offered by 52w is unique and gives an excellent opportunity for
the intravenous treatment of acute coronary syndrome and the
oral prevention of secondary thrombotic events which is a clear
advantage compared to current standards of care. 52w displays
a potent antithrombotic activity in an arterial thrombosis
model, with the best bleeding/antithrombotic balance when
compared to clopidogrel, prasugrel, and ticagrelor.22
Intermediates 3, 4, and 20 are the three key synthons for the
preparation of compounds 44 and 45 (Scheme 6). Preparation
of compounds 44 was done from synthons 20 and 3 using
EDC·HCl as the coupling agent. Then acetyl removal with
potassium carbonate in MeOH produced compounds 45a, 45l,
45q, 45t, and 45u. Bromination of 45u in AcOH afforded 45m.
Bromoaryl substitution by a chlorine atom was possible using
NiCl2 in DMF (45o from 45m and 45r from 45q).29 Then
reaction with methyl bromoacetate produced compounds 46a,
46l, 46m, 46o, 46q, 46r, 46t, and 46u. Alternatively, other
compounds 46 were prepared from corresponding synthons 20
and 4 (Scheme 6). Some intermediates 20 were not engaged in
coupling for various reasons. 20ag, 20ah, 20ai, 20aj were not
transformed in final compounds, as they were found to be
mutagenic. 20x was reacted. However, 46x, the expected
product of the reaction, was found to be unstable, and the
synthesis was discontinued. Probably because of the electron
withdrawing effect of the nitrile group, the amino group of 20z
was found to be totally nonreactive. Thus, 46z was obtained
from 46v. The same palladium-catalyzed cyanation procedure
was used to generate 46p from 46f. 46b was produced from
46a using the NaH/CH3I couple in DMF. Compounds 50
were the result of the mild hydrolysis of carboxylic esters 46
using aqueous NaOH in organic solvents (Scheme 6). In the
specific case of 46ak, bearing also an aromatic methyl ester,
aqueous LiOH was used to produce 50ak. The tert-butyl ester
from 46e was reacted in TFA/DCM to produce 50e. Mild
nitrile hydrolysis of 50p produced the corresponding
carboxamide 50p′. Alternatively, 47, bearing a butanoic side
chain, was prepared through a peptidic coupling between 20h
and 4a′ followed by ester hydrolysis, and 49, bearing a
propanoic side chain, was obtained by a Michael addition of
45a with ethyl acrylate followed by ester hydrolysis (Scheme
6).
Usually, final compounds were prepared directly from
compounds 50 through peptidic couplings (Scheme 6).
Sometimes more than one step was necessary. Hence, the
preparation of 53, 65−67, 69, 70, 72, 74, 76, 86, 88, 90, and 95
required a BOC-deprotection step after peptidic coupling,
sometimes followed by a methylation step to produce 62, 75,
77, 87, 89, and 96. Compound 102 was the result of a sequence
of four chemical steps from 50h (Scheme 6). Peptidic coupling
with piperazine-2-carboxylic acid methyl ester was followed by
a reductive amination with tert-butyl methyl(2-oxoethyl)-
carbamate. BOC deprotection in TFA followed by intra-
molecular cyclization upon basification produced 102 as a
racemic mixture. Separation on a chiral column led to the two
enantiomers 103 and 104. Finally, mild hydrolysis of 52ak
using aqueous LiOH in THF afforded 52aa.
Accordingly, 52w has the potential to differentiate from other
antiplatelet agents and was selected for development as the
preclinical candidate SAR216471.
EXPERIMENTAL METHODS
■
Reactions were run using commercially available starting materials
without further purification. All solvents were analytical grade, and
anhydrous reactions were performed in oven-dried glassware under an
atmosphere of argon or nitrogen. The microwave procedures were
carried out with a Biotage microwave. The proton nuclear magnetic
resonance spectra (1H NMR) were recorded on Bruker spectrometers
(250, 400, and 500 MHz) in DMSO-d6 or CDCl3. The chemical shifts
δ were expressed in parts per million (ppm). The following
abbreviations were used for interpreting the spectra: s, singlet; d,
doublet; t, triplet; q, quadruplet; quint, quintuplet; sext, sextuplet; m,
multiplet; dd, doublet of doublets; br, broad singlet. HPLC−UV−MS
(liquid chromatography/UV detection/mass spectrometry) and
melting points (mp) analyses were given hereafter only for compounds
that were involved in biological assays. The UV purity of the
compounds was assessed to be above 95%. Results of elemental
analysis were obtained within 0.3% of the theoretical values. Melting
points were determined using open capillary tubes on a Buchi 530
apparatus and are uncorrected. The HPLC−UV−MS equipment used
was composed of a chromatographic chain equipped with a diode array
detector and a quadripole mass spectrometer: Agilent 1100 series;
Symmetry C18 3.5 μm (2.1 mm × 50 mm, Waters); solvent A, water +
0.005% TFA; solvent B, MeCN + 0.005% TFA; 0.4 mL/min, 25 °C;
MSD SL (Agilent) ESI+.
CONCLUSION
■
Parameters for methods A and B are shown in Table 10.
In the search for a backup to clopidogrel, we initiated a HTS
campaign designed to identify novel reversible P2Y12
antagonists. Starting from a hit with low micromolar binding
activity, we had set up a rational and sequential SAR
optimization process leading to the identification of 52w. The
first sequence was focused on the optimization of the in vitro
binding activity leading to A3 followed by the key finding of the
pyrazole ketone motif as a metabolically stable pyrazole ester
Method C involved the following: Phenomenex Luna C18(2)
column, 10 mm × 2 mm, 3 μm; gradient, water + 0.05% TFA/
acetonitrile st 93:7 (0 min) to 5:95 (1.20 min) to 5:95 (1.40 min); 1.1
mL/min, 30 °C; Agilent series 1100 MSD.
Chemical names of the molecules have been generated using
ACDName, version 12, from Advanced Chemistry Development, Inc.
1-(5-Bromo-1H-indol-3-yl)-2,2,2-trifluoroethanone (1f). Tri-
fluoroacetic anhydride (21.3 mL, 151 mmol) in diethyl ether (70 mL)
R
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
1-Carboxymethyl-5-chloro-1H-indole-3-carboxylic Acid tert-
Butyl Ester (12). To a solution of 11 (9.80 g, 24.4 mmol) in a
mixture of THF/H2O (140 mL, 4/1; v/v), was added aqueous 2 N
LiOH (12.2 mL, 24.4 mmol) at 0 °C. After the mixture was stirred for
3 h at this temperature, conversion was complete and the reaction
mixture was adjusted to pH 6 by the addition of aqueous 1 N HCl.
Upon evaporation of the solvent a solid was formed which was
collected on a glass filter, yielding 4.3 g of a white solid. Back-
extraction of the wash solution yielded another batch of 4.0 g of
equally pure title compound (total amount of 12 obtained = 8.3 g,
quantitative). 1H NMR, DMSO-d6 (500 MHz), δ (ppm): 8.09 (s, 1H),
7.94 (d, J = 2.1 Hz, 1H), 7.55 (d, J = 8.0 Hz, 1H), 7.25 (dd, J = 8.0 Hz,
J = 2.1 Hz, 1H), 5.14 (s, 2H), 1.57 (s, 9H). LCMS (method C): M −
tBu + H+ (tR) = 254.2 (0.96 min).
Table 10
gradient (min)
solvent A
solvent B
Method A
0
10
15
100
0
100
100
0
0
Method B
0
30
35
100
0
0
100
100
0
was added dropwise to a cold solution (−5 °C) of 5-bromoindole (20
g, 101 mmol) in diethyl ether (250 mL). It was stirred for 2 h at −5
°C. The precipitate formed was drained and washed with diethyl ether.
1f was obtained as a white powder (24.5 g, 83%). 1H NMR, DMSO-d6
(250 MHz), δ (ppm): 12.85 (br, 1H), 8.55 (s, 1H), 8.31 (s, 1H), 7.58
(d, J = Hz, 1H), 7.50 (d, J = Hz, 1H).
1-(5-Chloro-1H-indol-3-yl)-2,2,2-trifluoroethanone (1h).
Same procedure was used as for 1f (37.9 g, 93%). 1H NMR,
DMSO-d6 (250 MHz), δ (ppm): 12.89 (br, 1H), 8.58 (q, J = 1.9 Hz,
1H), 8.16 (d, J = 2.1 Hz, 1H), 7.63 (d, J = 8.7 Hz, 1H), 7.39 (dd, J =
8.6, 2.1 Hz, 1H).
5-Bromo-1H-indole-3-carboxylic Acid (2f). 1f (24.0 g, 82.2
mmol) was added to a solution of potassium hydroxide (46.1 g, 822
mmol) in water (25 mL) and then heated for 4 h under reflux. The
reaction mixture was cooled and washed with diethyl ether. The
aqueous phase was cooled to 5 °C and then neutralized with a solution
of phosphate buffer preloaded with 35% hydrochloric acid. The
precipitate formed was drained, washed with water, and dried under
5-Chloro-1-[2-(4-methylpiperazin-1-yl)-2-oxoethyl]-1H-in-
dole-3-carboxylic Acid tert-Butyl Ester (13). To a solution of 12
(3.15 g, 10.0 mmol) in DMF (70 mL) were added EDC·HCl (2.93 g,
20.0 mmol) and pentafluorophenol (2.81 g, 15.0 mmol). After 20 min,
N-ethylmorpholine (3.80 mL, 30.0 mmol) and 1-methylpiperazine
(1.1 mL, 10.0 mmol) were added and the reaction mixture was stirred
for 12 h at room temperature. After this time the reaction mixture was
concentrated and taken up with DCM and washed twice with an
aqueous LiCl solution (4% w/w). The organic layer was dried over
MgSO4 and the crude product obtained after evaporation was purified
by chromatography on silica gel, eluting with a gradient of EtOAc/
MeOH to afford 13 (2.26 g, 58%). 1H NMR, DMSO-d6 (500 MHz), δ
(ppm): 8.05 (s, 1H), 7.99 (d, J = 2.1 Hz, 1H), 7.48 (d, J = 8.8 Hz,
1H), 7.23 (dd, J = 8.8 Hz, J = 2.1 Hz, 1H), 5.34 (s, 2H), 3.58 (m, 2H),
3.47 (m, 2H), 2.44 (m, 2H), 2.31 (m, 2H), 2.24 (s, 3H), 1.58 (s, 9H).
t
LCMS (method C): M − Bu + H+ (tR) = 336.2 (0.73 min).
5-Chloro-1-[2-(4-methylpiperazin-1-yl)-2-oxoethyl]-1H-in-
dole-3-carboxylic Acid (4h′). To a solution of 13 (2.26 g, 5.80
mmol) in DCM (30 mL) was added TFA (4.3 mL, 56.1 mmol). After
being stirred for 3 h, the reaction mixture was concentrated and co-
distilled twice with toluene. The residue thus obtained was dissolved in
MeCN/H2O and freeze-dried after addition of 7.3 mL of aqueous 2 N
HCl. This procedure was repeated once to give 4h′ in its
1
vacuum. 2f was obtained as a white powder (15.6 g, 79%). H NMR,
DMSO-d6 (250 MHz), δ (ppm): 12.10 (br, 1H), 12.02 (br, 1H), 8.14
(d, J = 1.9 Hz, 1H), 8.06 (s, 1H), 7.46 (d, J = 8.6 Hz, 1H), 7.32 (dd, J
= 8.6, 1.9 Hz, 1H).
5-Chloro-1H-indole-3-carboxylic Acid (2h). Same procedure
was used as for 2f (27.6 g, 92%). H NMR, DMSO-d6 (250 MHz), δ
1
1
(ppm): 12.10 (s, 1H), 11.95 (s, 1H), 8.07 (s, 1H), 7.97 (s, 1H), 7.49
(d, J = 8.5 Hz, 1H), 7.20 (d, J = 8.5 Hz, 1H).
hydrochloride form (2.33 g, quantitative). H NMR, DMSO-d6 (500
MHz), δ (ppm): 11.25 (br, 1H), 7.99 (s, 1H), 7.97 (d, J = 2.1 Hz,
1H), 7.54 (d, J = 8.3 Hz, 1H), 7.25 (dd, J = 8.3 Hz, J = 2.1 Hz, 1H),
5.45 (d, J = 17.3 Hz, 1H), 5.30 (d, J = 17.3 Hz, 1H), 4.37 (d, J = 13.1
Hz, 1H), 4.15 (d, J = 13.1 Hz, 1H), 3.62 (t, J = 12.5 Hz, 1H), 3.52 (d, J
= 12.1 Hz, 1H), 3.43 (d, J = 12.1 Hz, 1H), 3.13 (m, 2H), 2.96 (m,
1H), 2.79 (s, 3H). LCMS (method C): MH+ (tR) = 336.2 (0.51 min).
1-[6-(4-Methoxybenzylamino)pyridazin-3-yl]-5-methyl-1H-
pyrazole-4-carboxylic Acid Ethyl Ester (21). 4-Methoxybenzyl-
amine (0.31 mL, 2.37 mmol) was added to a solution of ethyl 1-(6-
chloropyridazin-3-yl)-5-methyl-1H-pyrazole-4-carboxylate (0.30 g,
1.13 mmol) in 1,4-dioxane (2 mL). Then it was heated at reflux for
2 h and evaporated to dryness. The residue was triturated with water,
leading to the formation of a precipitate which was drained, dried
under vacuum, taken up in isopropyl ether, drained again, and dried
5-Chloro-1-(2-methoxy-2-oxoethyl)-1H-indole-3-carboxylic
Acid (4h). A solution of 2h (10 g, 51.1 mmol) in DMF (110 mL) was
added dropwise to a suspension of NaH (60% in oil, 4.50 g, 112.5
mmol) in DMF (400 mL) at −10 °C. After it returned to room
temperature, it was stirred for 1 h. The reaction mixture was cooled to
−20 °C. Methyl bromoacetate (4.86 mL, 51.1 mmol) was added
dropwise, and then it was returned to room temperature for a period
of 5 h and stirred for 15 h. The reaction mixture was added to 1 L of
EtOAc/aqueous 1 N HCl mixture. The organic phase was collected,
and the aqueous phase was extracted with EtOAc. The combined
organic phases were washed with water and with brine, then dried over
Na2SO4 and evaporated to dryness to afford 4h as a white powder (8.9
1
g, 65%). H NMR, DMSO-d6 (250 MHz), δ (ppm): 12.3 (br, 1H),
8.12 (s, 1H), 7.98 (s, 1H), 7.57 (d, 1H), 7.26 (d, 1H), 5.22 (s, 2H),
3.70 (s, 3H).
1
under vacuum to yield 21 (0.30 g, 72%). H NMR, DMSO-d6 (250
MHz), δ (ppm): 8.03 (s, 1H), 7.67 (t, J = 5.5 Hz, 1H), 7.62 (d, J = 9.2
Hz, 1H), 7.32 (d, J = 6.7 Hz, 2H), 7.09 (d, J = 9.2 Hz, 1H), 6.91 (d, J
= 6.7 Hz, 2H), 4.55 (d, J = 5.5 Hz, 2H), 4.26 (q, J = 7.0 Hz, 2H), 3.74
(s, 3H), 2.69 (s, 3H), 1.30 (t, J = 7.0 Hz, 3H).
1-Benzyloxycarbonylmethyl-5-chloro-1H-indole-3-carbox-
ylic Acid tert-Butyl Ester (11). To a solution of 2h (5.00 g, 26.0
mmol) in benzene (200 mL) was added dimethylformamide di-tert-
butyl acetal (25 mL, 104 mmol). The reaction mixture was refluxed for
12 h. After this time another portion of the acetal (25 mL) was added
and the mixture stirred again under reflux for 12 h. It was
concentrated, diluted with DCM, and washed three times with a
saturated aqueous NaHCO3 solution. To the crude product (5.7 g, 22
mmol) in solution in DMF (150 mL) were added benzyl
bromoacetate (3.50 mL, 22.0 mmol) and cesium carbonate (11.1 g,
34.0 mmol). After the mixture was stirred for 12 h it was diluted with
DCM and washed three times with an aqueous LiCl solution (4% w/
w). The organic layer was dried over MgSO4 and the crude product
obtained after evaporation was purified by chromatography on silica
gel, eluting with a gradient of EtOAc/MeOH to afford 11 (5.80 g,
1-[6-(4-Methoxybenzylamino)pyridazin-3-yl]-5-methyl-1H-
pyrazole-4-carboxylic Acid (22). A solution of 21 (5 g, 13.6 mmol)
in EtOH (50 mL) was added to a solution of KOH (3.82 g, 68.2
mmol) in water (50 mL). It was heated at 80 °C for 2 h. It was then
evaporated to dryness, and the residue was taken up in water (100
mL). An aqueous 1 N HCl of solution (68.0 mL, 68.0 mmol) was
added dropwise under stirring and the precipitate that formed was
drained, washed with water, and then dried in a vacuum stove to afford
22 as a white powder (4.5 g, 98%). 1H NMR, DMSO-d6 (250 MHz), δ
(ppm): 8.42 (br, 1H), 8.04 (s, 1H), 7.81 (d, J = 9.5 Hz, 1H), 7.37−
7.30 (m, 3H), 6.93 (d, J = 6.9 Hz, 2H), 4.57 (s, 2H), 3.75 (s, 3H), 2.71
(s, 3H).
66%). LCMS (method C): M − Bu + H+ (tR) = 344.2 (1.22 min).
t
S
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
1-{6-[tert-Butoxycarbonyl-(4-methoxybenzyl)amino]-
pyridazin-3-yl}-5-methyl-1H-pyrazole-4-carboxylic Acid (16w).
DMAP (0.17 g, 1.38 mmol), Et3N (2.32 mL, 16.5 mmol), and di-tert-
butyl dicarbonate (3.01 g, 13.8 mmol) were successively added to a
solution of 22 (1.87 g, 5.51 mmol) in DMF (22 mL). Then it was
stirred for 20 h at room temperature. A solution of potassium
hydroxide (0.46 g, 8.27 mmol) in water (10 mL) was added, and it was
stirred for 20 h. Water (500 mL) was added, and it was washed with
ether. The aqueous phase was buffered with phosphate buffer
preloaded with aqueous 1 N HCl. The precipitate that formed was
drained, washed with water, and dried in a vacuum stove at 50 °C to
afford 16w (1.35 g, 56%). 1H NMR, DMSO-d6 (250 MHz), δ (ppm):
12.7 (br, 1H), 8.21−8.05 (m, 3H), 7.25 (d, J = 8.7 Hz, 1H), 6.88 (d, J
= 8.7 Hz, 1H), 5.19 (br, 2H), 3.71 (s, 3H), 2.83 (s, 3H), 1.43 (s, 9H).
3-(1-{6-[tert-Butoxycarbonyl-(4-methoxybenzyl)amino]-
pyridazin-3-yl}-5-methyl-1H-pyrazol-4-yl)-3-oxopropionic Acid
Ethyl Ester (17w). 1,1′-Carbonyldiimidazole (0.53 g, 3.25 mmol) was
added to a solution of 16w (1.1 g, 2.50 mmol) in THF (17 mL), and it
was stirred for 20 h at room temperature. Magnesium bis(3-ethoxy-3-
oxopropanoate) (0.82 g, 2.88 mmol) was added, and it was stirred at
55 °C for 20 h. EtOAc (50 mL) was added. It was washed with
aqueous 0.1 N NaOH, with brine, dried over Na2SO4, and evaporated
to dryness. It was purified by silica gel chromatography, eluting with
DCM and then eluting with DCM/MeOH mixture (95/5; v/v) to
It should be noted that some more expected compound was present in
the filtrate. H NMR, DMSO-d6 (250 MHz), δ (ppm): 7.61 (s, 1H),
7.43−7.34 (m, 9H), 7.09−7.03 (m, 6H), 2.33 (s, 3H).
1
5-Methyl-1-trityl-1H-pyrazole-4-carboxylic Acid Methoxy-
methylamide (34). To a solution of 33 (9.70 g, 26.3 mmol),
DMAP (10.3 g, 84.3 mmol), and BOP-Cl (10.3 g, 39.5 mmol) in
DCM (100 mL) was added N,O-dimethylhydroxylamine hydro-
chloride (3.856 g, 39.5 mmol). The reaction mixture was stirred at
room temperature for 1 h and evaporated to dryness. The residue was
dissolved in EtOAc, and the organic phase was washed with water,
with brine, dried over Na2SO4, and evaporated to dryness. The solid
residue was triturated with isopropyl ether, filtered, and dried under
1
vacuum at 50 °C to afford 34 as a white powder (10.4 g, 87%). H
NMR, DMSO-d6 (250 MHz), δ (ppm): 7.71 (s, 1H), 7.44−7.33 (m,
9H), 7.15−7.05 (m, 6H), 3.39 (s, 3H), 3.13 (s, 3H), 2.34 (s, 3H).
1-(5-Methyl-1-trityl-1H-pyrazol-4-yl)butan-1-one (35). To a
solution of 34 (10.4 g, 25.2 mmol) in anhydrous THF (130 mL) at
−30 °C was added drop by drop a 2 M n-propylmagnesium chloride
solution in ether (32,7 mL, 65.4 mmol). The mixture was allowed to
return to room temperature, and it was then stirred for 4 h. The
reaction mixture was cooled to −30 °C, and water (50 mL) was added
drop by drop. After the mixture returned to room temperature, an
aqueous 1 N HCl solution (250 mL) was added and the aqueous
phase was extracted with EtOAc (3×). The combined organic phases
were washed with water, with brine, dried over Na2SO4, and
evaporated to dryness. The solid residue was triturated with isopropyl
ether, filtered, and dried under vacuum at 50 °C to afford 35 as a white
powder (8.4 g, 85%). 1H NMR, DMSO-d6 (400 MHz), δ (ppm): 7.93
(s, 1H), 7.45−7.33 (m, 9H), 7.12−7.04 (m, 6H), 2.62 (t, J = 7.2 Hz,
2H), 2.34 (s, 3H), 1.52 (sext, J = 7.2 Hz, 2H), 0.84 (t, J = 7.2 Hz, 3H).
1-(5-Methyl-1H-pyrazol-4-yl)butan-1-one (36). 35 (8.4 g, 21.3
mmol) was stirred in a 4 N solution of HCl in 1,4-dioxane (50 mL)
during 6 h. The reaction mixture was evaporated to dryness, and the
solid residue was triturated with isopropyl ether, filtered, and dried
under vacuum at 50 °C to afford 36 as a white powder (2.00 g, 62%).
It should be noted that evaporation of the filtrate allowed isolation of a
new batch with slightly less purity (1.10 g, 34%). 1H NMR, DMSO-d6
(250 MHz), δ (ppm): 8.15 (s, 1H), 2.71 (t, J = 7.0 Hz, 2H), 2.39 (s,
3H), 1.58 (sext, J = 7.0 Hz, 2H), 0.90 (t, J = 7.0 Hz, 3H).
1
afford 23w in the form of a white powder (1.18 g, 92%). H NMR,
DMSO-d6 (250 MHz), δ (ppm): 8.41 (s, 1H), 8.22 (d, J = 9.5 Hz,
1H), 8.09 (d, J = 9.5 Hz, 1H), 7.25 (d, J = 6.5 Hz, 2H), 6.88 (d, J = 6.5
Hz, 2H), 5.19 (s, 2H), 4.13 (q, J = 7.2 Hz, 2H), 4.04 (s, 2H), 3.72 (s,
3H), 2.82 (s, 3H), 1.43 (s, 9H), 1.21 (t, J = 7.2 Hz, 3H).
2-(1-{6-[tert-Butoxycarbonyl-(4-methoxybenzyl)amino]-
pyridazin-3-yl}-5-methyl-1H-pyrazole-4-carbonyl)butyric Acid
Ethyl Ester (18w). Potassium carbonate (1.25 g, 9.03 mmol),
tetrabutylammonium bromide (0.98 g, 3.03 mmol), and iodoethane
(0.44 mL, 4.51 mmol) were added to a solution of 17w (1.15 g, 2.26
mmol) in THF (23 mL). Then it was stirred at 55 °C for 20 h. After it
returned to room temperature, EtOAc (150 mL) was added and it was
washed with water, with brine, dried over Na2SO4, and evaporated to
dryness. It was purified by silica gel chromatography, eluting with
DCM/MeOH mixture (97/3; v/v) to afford 18w in the form of a
white powder (1.18 g, 97%). 1H NMR, DMSO-d6 (250 MHz), δ
(ppm): 8.49 (s, 1H), 8.22 (d, J = 9.5 Hz, 1H), 8.09 (d, J = 9.5 Hz,
1H), 7.25 (d, J = 8.7 Hz, 2H), 6.88 (d, J = 8.7 Hz, 2H), 5.19 (s, 2H),
4.28 (t, J = 7.2 Hz, 1H), 4.10 (q, J = 7.2 Hz, 2H), 3.72 (s, 3H), 2.82 (s,
3H), 1.88 (quint, J = 7.2 Hz, 2H), 1.43 (s, 9H), 1.15 (t, J = 7.2 Hz,
3H), 0.92 (t, J = 7.2 Hz, 3H).
1-[1-(6-Aminopyridazin-3-yl)-5-methyl-1H-pyrazol-4-yl]-
butan-1-one (20w). A solution of 18w (1.18 g, 2.19 mmol) in
aqueous 37% HCl (4.4 mL) was heated at 105 °C for 6 h. After it
returned to room temperature, water was added (50 mL), and it was
washed twice with EtOAc. Then the aqueous phase was evaporated to
dryness. A 0.2 N solution (50 mL) was added to the solid residue, and
it was extracted with EtOAc. The organic phase was washed with
brine, dried over Na2SO4, and evaporated to dryness to afford 20w as a
white powder (0.42 g, 78%). 1H NMR, DMSO-d6 (250 MHz), δ
(ppm): 8.27 (s, 1H), 7.59 (d, J = 9.4 Hz, 1H), 7.00 (d, J = 9.4 Hz,
1H), 6.76 (br, 2H), 2.82 (t, J = 7.2 Hz, 2H), 2.65 (s, 3H), 1.62 (sext, J
= 7.2 Hz, 2H), 0.93 (t, J = 7.2 Hz, 3H).
5-Methyl-1-trityl-1H-pyrazole-4-carboxylic Acid (33). A sol-
ution of 5-methyl-1H-pyrazole-4-carboxylic acid methyl ester (6.60 g,
47.1 mmol), potassium carbonate (8.46 g, 61.22 mmol), and trityl
chloride (15.8 g, 56.5 mmol) in DMF (50 mL) was stirred at room
temperature during 5 days. EtOAc was added and the organic phase
was washed with water, with brine, dried over Na2SO4, and evaporated
to dryness. The oily residue was suspended in EtOH/H2O (100 mL,
50/50 v/v). Then potassium hydroxide (9.95 g, 177 mmol) was added,
and the reaction mixture was refluxed during 6 h. The hot mixture was
drained, and the filtrate was reduced to remove EtOH. An aqueous 1
N HCl solution was added and the precipitate that formed was filtered,
washed with a isopropyl ether/EtOAc mixture (50/50 v/v), and dried
at 50 °C under vacuum to afford 33 as a white powder (9.70 g, 56%).
3-(tert-Butyldimethylsilanyloxy)pentanedioic Acid Dimethyl
Ester (37). To a solution of TBDMSCl (28.2 g, 187 mmol) and
imidazole (30.2 g, 443 mmol) in DCM (225 mL) was slowly added a
solution of dimethyl 3-hydroxyglutarate (30 g, 170 mmol) in DCM
(225 mL). After the mixture was stirred overnight water was added,
the layers were separated, and the aqueous layer was extracted twice
with DCM. The combined organic layers were dried over MgSO4, and
the crude product obtained after evaporation was purified by
chromatography on silica gel, eluting with a gradient of EtOAc/n-
heptane. The fractions containing the product were combined and the
solvent was evaporated under reduced pressure to yield 37 (45.0 g,
1
91%). H NMR, CDCl3 (400 MHz), δ (ppm): 4.50 (quint, 1H), 3.62
(s, 6H), 2.50 (d, 6H), 0.80 (s, 9H), 0.00 (s, 6H).
3-(tert-Butyldimethylsilanyloxy)pentane-1,5-diol (38). To a
solution of 37 (13.5 g, 46.5 mmol) in ether (750 mL) was added
successively lithium borohydride (6.00 g, 279 mmol) at 0 °C. After the
mixture was stirred overnight saturated aqueous ammonium chloride
was added and the mixture extracted twice with DCM. After
evaporation of the solvent the crude product 38 was obtained (11.9
1
g, quantitative). H NMR, DMSO-d6 (400 MHz), δ (ppm): 4.28 (t, J
= 5.4 Hz, 2H), 3.86 (quint, J = 6.0 Hz, 1H), 3.41 (m, 4H), 1.52 (m,
4H), 0.82 (s, 9H), 0.00 (s, 6H).
[4-(tert-Butyldimethylsilanyloxy)piperidin-1-yl]carbamic
Acid tert-Butyl Ester (39). To a solution of oxalyl chloride (13.3 mL,
152 mmol) in DCM (700 mL) was added dropwise a solution of
DMSO (14.4 mL, 203 mmol) in 160 mL of DCM at −78 °C. After 30
min, a solution of 38 (11.9 g, 50.8 mmol) in DCM (160 mL) was
added at this temperature, followed by the addition of Et3N (70.7 mL,
508 mmol) after 30 min. After stirring for 1 h, it was warmed to 0 °C
and stirred an additional 30 min at this temperature. It was diluted
with toluene (92 mL), filtered via a glass frit, and the filtrate was
T
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
concentrated. After resuspension with pentane it was filtered via a plug
of Celite to give the crude dialdehyde (13.5 g, quantitative) after
evaporation of the solvent. To a solution of crude dialdehyde (12.0 g,
∼52.1 mmol) and tert-butyl carbazate (7.60 g, 57.3 mmol) in DCM
(350 mL) was added sodium triacetoxyborohydride (26.5 g, 125
mmol) at 0 °C. After the mixture was stirred overnight, saturated
aqueous NaHCO3 was added and the layers were separated. The
aqueous layer was extracted twice with DCM to give the crude product
after evaporation of the solvent. Purification was performed by
chromatography on silica gel, eluting with a gradient of EtOAc/n-
heptane. The fractions containing the product were combined and the
solvent was evaporated under reduced pressure to afford 39 (8.6 g,
50%). 1H NMR, DMSO-d6 (400 MHz), δ (ppm): 7.86 (br, 1H), 3.64
(m, 1H), 2.77 (m, 2H), 2.52 (m, 2H), 1.65 (m, 2H), 1.43 (m, 2H),
1.33 (s, 9H), 0.83 (s, 9H), 0.00 (s, 6H).
(4-Hydroxypiperidin-1-yl)carbamic Acid tert-Butyl Ester
(40). To a solution of 39 (8.60 g, 26.0 mmol) in THF (400 mL)
was added 1 M TBAF in THF (28.6 mL, 28.6 mmol) at 0 °C. After the
mixture was stirred for 12 h, additional 28.6 mL of TBAF solution was
added and the reaction mixture stirred for 12 h. It was then
concentrated, redissolved in DCM, and extracted three times with
water. The combined aqueous layers were reextracted with DCM/i-
PrOH 3:1, and the combined organic fractions were evaporated.
Purification was performed by chromatography on silica gel, eluting
with a gradient of EtOAc/n-heptane. The fractions containing the
product were combined and the solvent was evaporated under reduced
pressure, affording 40 (4.35 g, 78%). 1H NMR, DMSO-d6 (400 MHz),
δ (ppm): 7.89 (br, 1H), 4.55 (d, J = 3.8 Hz, 1H), 3.43 (m, 1H), 2.78
(m, 2H), 2.52 (m, 2H), 1.67 (m, 2H), 1.48−1.36 (m, 11H).
2H), 1.96 (d, J = 12.2 Hz, 2H), 1.57 (sext, J = 7.2 Hz, 2H), 0.89 (t, J =
7.2 Hz, 3H).
{3-[6-(4-Butyryl-5-methylpyrazol-1-yl)pyridazin-3-ylcarba-
moyl]-5-chloroindol-1-yl}acetic Acid Methyl Ester (46w). To a
solution of 4h (6.00 g, 22.4 mmol) in DCM (200 mL) were added
DMF (few drops) and thionyl chloride (6.60 mL, 89.7 mmol). After 3
h under reflux, the reaction mixture was evaporated to dryness and the
solid residue was triturated with DCM (80 mL), filtered, and washed
with DCM, thus producing (5-chloro-3-chlorocarbonylindol-1-yl)-
acetic acid methyl ester as a white powder (4.5 g).
A solution of 20w (0.35 g, 1.43 mmol) and DMAP (0.59 g, 4.84
mmol) was stirred for 30 min in the presence of 4 Å molecular sieves
(1.00 g) in 1,2-dichloroethane (20 mL). (5-Chloro-3-chlorocarbony-
lindol-1-yl)acetic acid methyl ester (0.92 g, 3.21 mmol) was added,
and the reaction mixture was stirred at 80 °C for 6 h. After the mixture
returned to room temperature, the molecular sieves were removed by
filtration. The filtrate was washed with water, with brine, dried over
Na2SO4, and evaporated under vacuum. The solid residue was purified
by silica gel chromatography, eluting with DCM/MeOH mixture
(gradient from 0% to 5% of MeOH), obtaining 46w as a white powder
1
(0.51 g, 72% from 20w). H NMR, DMSO-d6 (250 MHz), δ (ppm):
11.42 (br, 1H), 8.71 (d, J = 9.5 Hz, 1H), 8.65 (s, 1H), 8.41 (s, 1H),
8.23 (s, 1H), 8.13 (d, J = 9.5 Hz, 1H), 7.62 (d, J = 8.7 Hz, 1H), 7.32
(d, J = 8.7 Hz, 1H), 5.33 (s, 2H), 3.73 (s, 3H), 2.87 (t, J = 7.2 Hz,
2H), 2.82 (s, 3H), 1.64 (sext, J = 7.2 Hz, 2H), 0.94 (t, J = 7.2 Hz, 3H).
{3-[4-(4-Butyryl-5-methylpyrazol-1-yl)phenylcarbamoyl]-5-
methylindol-1-yl}acetic Acid (50a). To 46a (10.0 g, 21.1 mmol) in
MeOH (100 mL) was added aqueous 2 N NaOH (15.9 mL, 31.7
mmol). It was stirred at room temperature overnight and then
concentrated under vacuum. The residue was suspended in aqueous 1
N HCl, then filtered, washed with water, and dried under vacuum to
afford 50a as a white powder (9.47 g, 98%). 1H NMR, DMSO-d6 (250
MHz), δ (ppm): 13.19 (br, 1H), 10.02 (s, 1H), 8.24 (s, 1H), 8.23 (s,
1H), 8.02 (s, 1H), 7.95 (d, J = 9.0 Hz, 2H), 7.49 (d, J = 9.0 Hz, 2H),
7.39 (d, J = 8.5 Hz, 1H), 7.08 (d, J = 8.5 Hz, 1H), 5.13 (s, 2H), 2.82 (t,
J = 7.2 Hz, 2H), 2.50 (s, 3H), 2.43 (s, 3H), 1.64 (sext, J = 7.2 Hz, 2H),
0.94 (t, J = 7.2 Hz, 3H). LCMS (method A): MH+ (tR) = 459 (8.31
min). Mp = 201 °C.
Methanesulfonic Acid 1-tert-Butoxycarbonylaminopiperi-
din-4-yl Ester (41). To a solution of 40 (4.35 g, 19.7 mmol) in
DCM (100 mL) were successively added DMAP (0.24 g, 2.00 mmol),
Et3N (2.80 mL, 20.6 mmol), and methanesulfonyl chloride (1.50 mL,
19.7 mmol). After the mixture was stirred overnight, it was diluted
with DCM and washed twice with aqueous 0.1 M HCl. The aqueous
layer was reextracted with DCM, and the combined organic fractions
1
yielded the crude product 41 as colorless oil (6.0 g, quantitative). H
NMR, DMSO-d6 (500 MHz), δ (ppm): 8.39 (br, 1H), 4.76 (m, 1H),
3.24 (s, 3H), 2.92 (m, 2H), 2.78 (m, 2H), 1.95 (m, 2H), 1.83 (m, 2H),
1.40 (s, 9H).
{3-[6-(4-Butyryl-5-methylpyrazol-1-yl)pyridazin-3-ylcarba-
moyl]-5-chloroindol-1-yl}acetic Acid (50w). Same procedure was
1
used as for 50a in dioxane (2.07 g, 69%). H NMR, DMSO-d6 (400
[4-(4-Butyryl-5-methylpyrazol-1-yl)piperidin-1-yl]carbamic
Acid tert-Butyl Ester (42). To a solution of 36 (1.55 g, 10.2 mmol)
in DMF (15 mL) was added potassium tert-butoxide (1.20 g, 10.7
mmol), and the solution was stirred at 50 °C for 30 min. Then a
solution of 41 (3.00 g, 10.2 mmol) in DMF (5 mL) was added and
further stirred at this temperature. Additional potassium tert-butoxide
(3 × 300 mg after 12 h each) was added in order to increase the
conversion. It was concentrated, diluted with DCM, and washed with
water three times. The crude product thus obtained was purified by
preparative HPLC (C18 reverse phase column, elution with a H2O/
MeCN gradient with 0.1% TFA). The fractions containing the product
were evaporated and lyophilized to yield 0.96 g of a mixture of the title
compound and its position isomer. Further HPLC purification on
chiral stationary phase yielded the desired N-alkylated pyrazole 42
MHz), δ (ppm): 13.25 (br, 1H), 11.35 (s, 1H), 8.68 (d, J = 8.0 Hz,
1H), 8.66 (s, 1H), 8.40 (s, 1H), 8.23 (s, 1H), 8.13 (d, J = 8.0 Hz, 1H),
7.61 (d, J = 8.8 Hz, 1H), 7.30 (d, J = 8.8 Hz, 1H), 5.17 (s, 2H), 2.86 (t,
J = 7.2 Hz, 2H), 2.82 (s, 3H), 1.65 (sext, J = 7.2 Hz, 2H), 0.93 (t, J =
7.2 Hz, 3H).
Hydrochloride Salt of N-[6-(4-Butanoyl-5-methyl-1H-pyra-
zol-1-yl)pyridazin-3-yl]-5-chloro-1-[2-(4-methylpiperazin-1-yl)-
2-oxoethyl]-1H-indole-3-carboxamide (52w). 1-Methylpiperazine
(1.66 g, 16.5 mmol), pyridine (2.23 mL, 27.6 mmol), and BOP-Cl
(4.30 g, 16.5 mmol) were added successively to 50w (2.65 g, 5.51
mmol) in solution in DMF (80 mL). The reaction mixture was stirred
for 48 h. Then it was poured into a biphasic mixture of EtOAc and
saturated aqueous NaHCO3. The precipitate formed was filtered and
washed with isopropyl ether. The filtrate was transferred to a
separating funnel, and the organic phase was washed with water,
with brine and then dried over Na2SO4. The organic phase was
concentrated partially, and the precipitate formed was filtered. The
two precipitates were combined and dried in a vacuum stove. It was
suspended in MeOH (200 mL). Then a 1 N HCl solution in ether
(6.60 mL, 6.60 mmol) was added, and it was stirred for 1 h. The
precipitate that formed was drained. It was washed with isopropyl
ether and then dried in a vacuum stove at 40 °C, obtaining 52w (2.29
1
(0.36 g, 10%). H NMR, DMSO-d6 (500 MHz), δ (ppm): 8.49 (br,
1H), 8.03 (s, 1H), 4.22 (m, 1H), 3.05 (d, J = 10.4 Hz, 2H), 2.81 (t, J =
10.4 Hz, 2H), 2.73 (t, J = 7.2 Hz, 2H), 2.54 (s, 3H), 2.13 (m, 2H),
1.82 (m, 2H), 1.57 (sext, J = 7.2 Hz, 2H), 1.39 (s, 9H), 0.90 (t, J = 7.2
Hz, 3H).
Hydrochloride Salt of 1-[1-(1-Aminopiperidin-4-yl)-5-meth-
yl-1H-pyrazol-4-yl]butan-1-one (20ae). To a solution of 42 (0.54
g, 1.53 mmol) in DCM (12 mL) was added TFA (3.30 mL, 45.9
mmol). After stirring for 3 h at room temperature, the reaction mixture
was concentrated and co-distilled twice with toluene. The residue thus
obtained was dissolved in MeCN/H2O and freeze-dried after addition
of aqueous 2 N HCl (2.0 mL). This procedure was repeated once to
1
g, 69%) as a light yellow powder. H NMR, DMSO-d6 (400 MHz), δ
(ppm): 11.38 (s, 1H), 10.97 (br, 1H), 8.71 (d, J = 9.6 Hz, 1H), 8.58
(s, 1H), 8.41 (s, 1H), 8.24 (s, 1H), 8.14 (d, J = 9.6 Hz, 1H), 7.59 (d, J
= 8.8 Hz, 1H), 7.30 (d, J = 8.8 Hz, 1H), 5.50 (d, J = 16 Hz, 1H), 5.35
(d, J = 16 Hz, 1H), 4.39−3.41 (m, 5H), 3.25−2.93 (m, 3H), 2.89−
2.80 (m, 8H), 1.65 (sext, J = 7.2 Hz, 2H), 0.95 (t, J = 7.2 Hz, 3H).
LCMS (method A): MH+ (tR) = 563 (6.59 min). C 55.59%, H 5.35%,
1
give the crude hydrochloride salt of 20ae (0.35 g, 74%). H NMR,
DMSO-d6 (500 MHz), δ (ppm): 8.10 (s, 1H), 4.43 (m, 1H), 3.36 (m,
2H), 2.90 (br, 2H), 2.71 (t, J = 7.2 Hz, 2H), 2.55 (s, 3H), 2.15 (m,
U
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
N 18.46%, C28H31ClN8O3·HCl·0.1H2O. Mp = 287 °C (crystal
powder). Experimental log D = 2.71. Experimental pKa = 6.3.
Hydrochloride Salt of N-[4-(4-Butanoyl-5-methyl-1H-pyra-
zol-1-yl)piperidin-1-yl]-5-chloro-1-[2-(4-methylpiperazin-1-yl)-
2-oxoethyl]-1H-indole-3-carboxamide (52ae). To a solution of
4h′ (0.337 g, 1.00 mmol) in DMF (45 mL) were added HOBt (0.152
g, 1.10 mmol), diisopropylethylamine (0.60 mL, 3.50 mmol), and
EDC·HCl (0.191 g, 1.00 mmol). After 10 min, 20ae (0.330 g, 1.10
mmol) was added and the reaction mixture was stirred at room
temperature during 24 h. EDC·HCl (0.095 g), HOBt (0.075 g), and
diisopropylethylamine (0.30 mL) were added. After 24 h, the reaction
was complete. The reaction mixture was evaporated under vacuum,
and the residue was purified by preparative HPLC (reverse phase C18
column using a H2O/acetonitrile gradient in the presence of TFA
(0.1%). The solid residue was dissolved in a H2O/acetonitrile mixture
(1/1; v/v). Aqueous 1 N HCl (0.43 mL, 0.43 mmol) was added, and it
was lyophilized. This procedure was repeated once to afford 52ae
(0.186 g, 33%). 1H NMR, DMSO-d6 (500 MHz), δ (ppm): 11.30 (br,
1H), 8.22 (s, 1H), 8.11 (s, 1H), 8.09 (s, 1H), 7.57 (d, J = 9.0 Hz, 1H),
7.27 (d, J = 9.0 Hz, 1H), 5.48 (AB system, J = 17.0 Hz, 1H), 5.34 (AB
system, J = 17.0 Hz, 1H), 4.57 (br, 1H), 4.35 (d, J = 13.6 Hz, 2H),
4.17 (d, J = 13.6 Hz, 2H), 3.70−3.39 (m, 6H), 3.16 (m, 2H), 2.98 (m,
1H), 2.81 (d, J = 4.6 Hz, 3H), 2.75 (t, J = 7.2 Hz, 2H), 2.59 (s, 3H),
2.42 (m, 2H), 2.05 (m, 2H), 1.57 (sext, J = 7.2 Hz, 2H), 0.90 (t, J =
7.2 Hz, 3H).
aggregation amplitude expressed as height and are expressed as
percentage inhibition. The compounds have CI50 (of inhibition of
platelet aggregation) between 0.1 and 2 μM.
Inhibition of Platelet Aggregation in Vitro (Rat Blood). The
blood is taken from male rats of the Sprague−Dawley strain, weighing
250−300 g. The sample is taken on sodium citrate at 3.8% (1 volume
to 9 volumes of blood) by puncture of the abdominal aorta after
anesthetizing the animal with pentobarbital sodium. The platelet-rich
plasma (PRP) is obtained by centrifugation of the blood at 300g for 5
min, and the measurements of platelet aggregation are performed as
described above. The results are calculated using the area under the
curve of absorbance measured for 6 min and expressed as percentage
inhibition. The compounds have CI50 values (of inhibition of platelet
aggregation) between 0.02 and 1.5 μM.
Inhibition of Platelet Aggregation ex Vivo (Rat Blood). Male
rats of the Sprague−Dawley strain, weighing 250−300 g, are used at a
rate of 6 animals per batch. Each test compound is diluted in a solution
of glucose-containing water (glucose 5%) containing 5% of
Cremophor and 3% of glycofurol and administered by stomach tube
(10 mL/kg at 1 mg/mL) 2 h before taking the sample. The sample is
taken on sodium citrate at 3.8% (1 volume to 9 volumes of blood) by
puncture of the abdominal aorta after anesthetizing the animal with
pentobarbital sodium.
The platelet-rich plasma (PRP) is obtained by centrifugation of the
blood at 300g for 5 min, and the measurements of platelet aggregation
are performed as described above. The results are calculated using the
area under the curve of absorbance measured for 6 min and expressed
as percentage (%) of inhibition.
Binding Experiments. [33P]2-MeS-ADP binding was measured
on P2Y12-expressing CHO cells using a previously reported filtration
technique to separate the free from bound [33P]2-MeSADP.30
Incubations were performed in a total of 200 μL of binding buffer
(phosphate buffered saline [PBS], 1 mM EDTA 0.2%, w/v bovine
serum albumin) containing CHO cells (35 × 103/sample) and [33P]2-
MeS-ADP (0.16 nM, 0.15 × 106 dpm/sample). The tested compounds
were incubated for 30 min with cells. Triplicate incubations with
radioactive ligand were carried out at 20 °C for 15 min and were
terminated by the addition of 3 mL of ice-cold assay buffer followed by
rapid vacuum filtration over glass-fiber filters (Filtermats 11734,
Skatron Instruments Inc., Sterling, VA, USA). Filters were then
washed twice with 5 mL of ice-cold PBS, dried, and the radioactivity
was measured by scintillation counting. Nonspecific binding was
defined as the total binding measured in the presence of excess
unlabeled ADP (1 mM), and specific binding was defined as the
difference between total binding and nonspecific binding. The percent
inhibition was expressed as %I = [(total binding − total binding with
antagonist)/specific binding] × 100. The IC50 value was defined as the
concentration of inhibitor required to obtain 50% inhibition of the
specific binding. Concentration−response curves were analyzed with
internal software using the four-parameter logistic modeling according
to the method of Ratkowsky and Reedy.31 The adjustment was
obtained by nonlinear regression using the Marquardt algorithm in
SASR software (version 9.1 UNIX, SAS Institute, Cary, NC, USA).
Parameter estimations were obtained using % inhibition with lower
asymptotes and upper asymptotes constrained at 0 and 100,
respectively.
Inhibition of Platelet Aggregation in Vitro (Human Blood).
The blood is taken from healthy volunteers, using 20 mL syringes
containing 2 mL of buffered sodium citrate. The blood is transferred to
polypropylene tubes and centrifuged for 5 min (100g) at room
temperature (without using the brake of the centrifuge). The
supernatant platelet-rich plasma (PRP) is then removed, diluted, and
the platelets are counted before they are used in aggregation
measurements.
The measurements of platelet aggregation are performed at 37 °C
in glass tubes (Chrono-Log Kordia aggregometer). An amount of 4 μL
of the test compound (solution 100 times more concentrated than the
required final concentration, in DMSO) is mixed with 392 μL of fresh
PRP, and the mixture was incubated for 1 min with stirring. Then 4 μL
of a solution of ADP at 250 μM is added to the mixture. The
measurements of aggregation are monitored for 6−8 min, stirring
continuously, by recording the variations of optical density according
to the method of G. V. Born.32 The results are calculated using the
ASSOCIATED CONTENT
■
S
* Supporting Information
Selectivity data for SAR216471, analytical data (1H NMR) and
synthetic methods for reaction intermediates, analytical data
(1H NMR, LCMS, elemental analysis, melting points) and
synthetic methods for final compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: christophe.boldron@sanofi.com. Phone:
+33534633089.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Dr. J.-P. Maffrand for constant support and
strong scientific challenge, Dr. A. Lindenschmidt, Dr. V.
Wehner, and Dr. H. Heitsch (Sanofi R&D, Frankfurt,
Germany) for scientific challenge, Dr. P. Hamley and colleagues
(Sanofi R&D, Frankfurt, Germany) for library synthesis, Dr. C.
Ponthus, Dr. E. Cogo, Dr. J. Menegotto, and colleagues (Sanofi
R&D, Toulouse, France) for their essential support in structural
determinations, early formulation, and solid state studies, Dr. S.
Maignan and Dr. J. P. Rameau (Sanofi R&D, Toulouse, France)
for in silico chemistry support, J. C. Bourbier, G. Basquin, and
G. Manette (Sanofi R&D, Chilly-Mazarin, France) for synthetic
chemistry support. Also, special thanks are given to L.
Laplanche, F. Sadoun, E. Erdociain, P. Cardon, P. Dupont, G.
Fabre, and colleagues from the DSAR Department (Sanofi
R&D, Montpellier, France) for their support in ADME studies
and to A. R. Jost and K. Braun (Sanofi R&D, Frankfurt,
Germany) for the Ames II Salmonella evaluations.
V
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
2007, 17, 6013−6018. (c) Huber, K.; Hamad, B.; Kirkpatrick, P. Fresh
from the pipeline. Ticagrelor. Nat. Rev. Drug Discovery 2011, 10, 255−
256. (d) Gurbel, P. A.; Kereiakes, D. J.; Tantry, U. S. Ticagrelor for the
treatment of arterial thrombosis. Expert Opin. Pharmacother. 2010, 11,
2251−2259.
ABBREVIATIONS USED
■
AA, antiaggregant activity; aq, aqueous; Bnd, binding; BOP-Cl,
bis(2-oxo-3-oxazolidinyl)phosphinic chloride; Caco-2, colon
adenocarcinoma; CHO, Chinese hamster ovary; CRE-Luc,
luciferase-directed cAMP response elements; EDC·HCl, N-(3-
dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride;
EF, eligibility factor; HOBt, 1-hydroxybenzotriazole; MPK,
metabolism and pharmacokinetics; POC, proof of concept;
PRP, platelet rich plasma; TBDMSCl, tert-butyldimethylsilyl
chloride
(13) (a) Gurbel, P. A.; Kereiakes, D.; Tantry, U. S. Elinogrel
Potassium. Drugs Future 2010, 35, 885−892. (b) Ueno, M.; Rao, S. V.;
Angiolillo, D. J. Elinogrel: pharmacological principles, preclinical and
early phase clinical testing. Future Cardiol. 2010, 6, 445−453.
(c) Oestreich, J. H. Elinogrel, a reversible P2Y12 receptor antagonist
for the treatment of acute coronary syndrome and prevention of
secondary thrombotic events. Curr. Opin. Invest. Drugs 2010, 11, 340−
348. (d) Berger, J. S.; Roe, M. T.; Gibson, C. M.; Kilaru, R.; Green, C.
L.; Melton, L.; Blankenship, J. D.; Metzger, D. C.; Granger, C. B.;
Gretler, D. D.; Grines, C. L.; Huber, K.; Zeymer, U.; Buszman, P.;
Harrington, R. A.; Armstrong, P. W. Safety and feasibility of adjunctive
antiplatelet therapy with intravenous elinogrel, a direct-acting and
reversible P2Y12 ADP-receptor antagonist, before primary percuta-
neous intervention in patients with ST-elevation myocardial infarction:
the early rapid reversal of platelet thrombosis with intravenous
elinogrel before PCI to optimize reperfusion in acute myocardial
infarction (ERASE MI). Am. Heart J. 2009, 158, 998−1004.
(14) Etter, M. C.; Urbakzyk-Lipkowska, Z.; Zia-Ebrahimi, S. M.;
Panunto, T. W. Hydrogen bond directed cocrystallization and
molecular recognition properties of diarylureas. J. Am. Chem. Soc.
1990, 112, 8415−8426.
(15) Dennis, M. S.; Zhang, M.; Meng, Y. G.; Kadkhodayan, M.;
Kirchhofer, D.; Combs, D.; Damico, L. A. Albumin binding as a
general strategy for improving the pharmacokinetics of proteins. J. Biol.
Chem. 2002, 38, 35035−35043.
(16) Ames, B. N.; Mccann, J.; Yamasaki, E. Methods for detecting
carcinogens and mutagens with the Salmonella/mammalian-micro-
some mutagenecity test. Mutat. Res. 1975, 31, 347−364.
(17) Mortelmans, K.; Zeiger, E. The Ames Salmonella/microsome
mutagenicity assay. Mutat. Res. 2000, 455, 29−60.
REFERENCES
■
(1) Kunapuli, S. P. P2 receptors and platelet activation.
TheScientificWorldJournal 2002, 2, 424−433.
(2) Kauffenstein, G.; Bergmeier, W.; Eckly, A.; Ohlmann, P.; Leon,
C.; Cazenave, J. P.; Nieswandt, B.; Gachet, C. The P2Y12 receptor
induces platelet aggregation through weak activation of the αIIbβ3
integrina phosphoinositide 3-kinase-dependent mechanism. FEBS
Lett. 2001, 505 (2), 281−290.
(3) Maffrand, J. P. The story of clopidogrel and its predecessor,
ticlopidine: Could these major antiplatelet and antithrombotic drugs
be discovered and developed today? C. R. Chim. 2012, 15, 737−743.
(4) Angiolillo, D. J. The evolution of antiplatelet therapy in the
treatment of acute coronary syndromes: from aspirin to the present
day. Drugs 2012, 72 (16), 2087−2116.
(5) Feliste, R.; Delebassee, D.; Simon, M. F.; Chap, H.; Defreyn, G.;
Vallee, E.; Douste-Blazy, L.; Maffrand, J. P. Effect of PCR 4099 on
ADP-induced calcium movements and phosphatidic acid production in
rat platelets. Thromb. Res. 1987, 48 (4), 403−415.
(6) Hollopeter, G.; Jantzen, H. M.; Vincent, D.; Li, G.; England, L.;
Ramakrishnan, V.; Yang, R. B.; Nurden, P.; Nurden, A.; Julius, D.;
Conley, P. B. Identification of the platelet ADP receptor targeted by
antithrombotic drugs. Nature 2001, 409, 202−207.
(18) (a) Helmick, J. S.; Martin, K. A.; Heinrich, J. L.; Novak, M.
Mechanism of the reaction at carbon and nitrogen nucleophiles with
the model carcinogens O-pivaloyl-N-arylhydroxylamines: competing
SN2 substitution and SN1 solvolysis. J. Am. Chem. Soc. 1991, 113,
3459−3466. (b) Borosky, G. L. Ultimate carcinogenic metabolites
from aromatic and heterocyclic aromatic amines: a computational
study in relation to their mutagenic potency. Chem. Res. Toxicol. 2007,
20, 171−180.
(7) Robert, A. N. Identification of the P2Y12 receptor: a novel
member of the P2Y family of receptors activated by extracellular
nucleotides. Mol. Pharmacol. 2001, 60 (3), 416−420.
(8) (a) Savi, P.; Herbert, J. M. clopidogrel and ticlopidine: P2Y12
adenosine diphosphate-receptor antagonists for the prevention of
atherothrombosis. Semin. Thromb. Hemostasis 2005, 31, 174−83.
(b) CAPRIE Steering Committee.. A randomised, blinded, trial of
clopidogrel versus aspirin in patients at risk of ischaemic events
(CAPRIE). Lancet 1996, 348, 1329−1339. (c) The Clopidogrel in
Unstable Angina To Prevent Recurrent Events Trial Investigators..
Effects of clopidogrel in addition to aspirin in patients with acute
coronary syndromes without ST-segment elevation. N. Engl. J. Med.
2001, 345, 494−502.
(19) Dipple, A. DNA adducts of chemical carcinogens. Carcinogenesis
1995, 16, 437−441.
(20) (a) Benigni, R. Structure−activity relationship studies of
chemical mutagens and carcinogens: mechanistic investigations and
prediction approaches. Chem. Rev. 2005, 105, 1767−1800. (b) Benigni,
R.; Giuliani, A.; Franke, R.; Gruska, A. Quantitative structure−activity
relationship of mutagenic and carcinogenic aromatic amines. Chem.
Rev. 2000, 100, 3697−3714. (c) Kadlubar, F. F.; Beland, F. A.
Chemical Properties of Ultimate Carcinogenic Metabolites of
Arylamines and Arylamides. Polycyclic Hydrocarbons and Carcinogenesis;
American Chemical Society: Washington, DC, 1985; pp 341−370.
(d) Chung, K. T.; Kirkovsky, L.; Kirkovsky, A.; Purcell, W. P. Review
of Mutagenicity of Monocyclic Aromatic amines: quantitative
structure−activity relationships. Mutat. Res. 1997, 387, 1−16.
(e) Colvin, M. E.; Hatch, F. T.; Felton, J. S. Chemical and biological
factors affecting mutagen potency. Mutat. Res. 1998, 400, 479−492.
(f) Benigni, R.; Bossa, C.; Netzeva, T.; Rodomonte, A.; Tsakovska, I.
Mechanistic QSAR of aromatic amines: new models for discriminating
between homocyclic mutagens and nonmutagens, and validation of
models for carcinogens. Environ. Mol. Mutagen. 2007, 48, 754−771.
(g) Benigni, R.; Passerini, L. Mutat. Res. 2002, 511, 191−206.
(21) Kimball, R. F. The mutagenicity of hydrazine and some of its
derivatives. Mutat. Res. 1977, 39 (2), 111−126.
(9) Niitsu, Y.; Jakubowski, J. A.; Sugidachi, A.; Asai, F. Pharmacology
of CS-747 (prasugrel, LY640315), a novel, potent antiplatelet agent
with in vivo P2Y12 receptor antagonist activity. Semin. Thromb.
Hemostasis 2005, 31, 184−194.
(10) Savi, P.; Herbert, J. M.; Pflieger, A. M.; Dol, F.; Delebassee, D.;
Combalbert, J.; Defreyn, G.; Maffrand, J. P. Importance of hepatic
metabolism in the antiaggregating activity of the thienopyridine
clopidogrel. Biochem. Pharmacol. 1992, 44 (3), 527−532.
(11) Schroer, K. The basic pharmacology of ticlopidine and
clopidogrel. Platelets 1993, 4 (5), 252−261.
(12) (a) van Giezen, J. J. J.; Humphries, R. G. Preclinical and clinical
studies with selective reversible direct P2Y12 antagonists. Semin.
Thromb. Hemostasis 2005, 31, 195−204. (b) Springthorpe, B.; Bailey,
A.; Barton, P.; Birkinshaw, T. N.; Bonnert, R. V.; Brown, R. C.;
Chapman, D.; Dixon, J.; Guile, S. D.; Humphries, R. G.; Hunt, S. F.;
Ince, F.; Ingall, A. H.; Kirk, I. P.; Leeson, P. D.; Leff, P.; Lewis, R. J.;
Martin, B. P.; McGinnity, D. F.; Mortimore, M. P.; Paine, S. W.;
Pairaudeau, G.; Patel, A.; Rigby, A. J.; Riley, R. J.; Teobald, B. J.;
Tomlinson, W.; Webborn, P. J. H.; Willis, P. A. From ATP to
AZD6140: the discovery of an orally active reversible P2Y12 receptor
antagonist for the prevention of thrombosis. Bioorg. Med. Chem. Lett.
(22) Delesque-Touchard, N.; Pflieger, A. M.; Bonnet-Lignon, S.;
Millet, L.; Salel, V.; Boldron, C.; Lassalle, G.; Herbert, J. M.; Savi, P.;
Bono, F. SAR216471, an alternative to the use of currently available
W
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
P2Y12 receptor inhibitors? Thromb. Res. [Online early access]. DOI:
10.1016/j.thromres.2014.06.034. Published Online: July 11, 2014.
(23) Mackie, R. K.; Mhatre, S.; Tedder, J. M. Carbon-
trifluoroacylationa convenient route to carboxylic acids. J. Fluorine
Chem. 1977, 10 (6), 437−445.
(24) Sakamoto, T.; Nagano, T.; Kondo, Y.; Yamanaka, H. Condensed
heteroaromatic ring systems. XVII. Palladium-catalyzed cyclization of
β-(2-halophenyl)amino substituted α,β-unsaturated ketones and esters
to 2,3-disubstituted indoles. Synthesis 1990, 3, 215−218.
(25) Hwu, J. R.; Patel, H. V.; Lin, R. J.; Gray, M. O. Novel methods
for the synthesis of functionalized indoles from arylhydroxylamines
and activated acetylenes. J. Org. Chem. 1994, 59, 1577−1582.
(26) Duflosa, M.; Nourrissona, M. R.; Breleta, J.; Couranta, J.;
LeBauta, G.; Grimaudb, N.; PetitbDuflos, J. Y. N-Pyridinylindole-3-
(alkyl)carboxamides and derivatives as potential systemic and topical
inflammation inhibitors. Eur. J. Med. Chem. 2001, 36 (6), 545−553.
(27) Brooks, D. W.; Lu, L. D. L.; Masuniune, S. C-Acylation under
virtually neutral conditions. Angew. Chem., Int. Ed. Engl. 1979, 18, 72−
74.
(28) Hu, F. Z.; Zhang, G. F.; Liu, B.; Zou, X. M.; Zhu, Y. Q.; Yang,
H. Z. A new method for the synthesis and herbicidal activity of 3-
phenoxy-6-(1H-(substituted)pyrazol-1-yl) pyridazines. J. Heterocycl.
Chem. 2009, 46, 584−590.
(29) Arvela, K.; Leadbeater, N. E. Fast and easy halide exchange in
aryl halides. Synlett 2003, 8, 1145−1148.
(30) Savi, P.; Herbert, J. M. Use of radiolabeled 2-methylthio-ADP to
study P2Y receptors on platelets and cell lines. Methods Mol. Biol.
2004, 273, 115−124.
(31) Ratkowsky, D. A.; Reedy, T. J. Choosing near-linear parameters
in the four-parameter log model for radioligand and related assays.
Biometrics 1986, 42, 575−582.
(32) Born, G. V. Aggregation of blood platelets by adenosine
diphosphate and its reversal. Nature 1962, 194, 927−929.
X
dx.doi.org/10.1021/jm500588w | J. Med. Chem. XXXX, XXX, XXX−XXX