Conformationally constrained ortho- Anilino diaryl ureas: Discovery of 1-(2-(1′-Neopentylspiro[indoline-3,4′-piperidine]-1-yl)phenyl) -3-(4-(trifluoromethoxy)phenyl)urea, a potent, selective, and bioavailable P2Y1 antagonist

Article abstract of DOI:10.1021/jm4013906

Preclinical antithrombotic efficacy and bleeding models have demonstrated that P2Y₁ antagonists are efficacious as antiplatelet agents and may offer a safety advantage over P2Y₁₂ antagonists in terms of reduced bleeding liabilities. In this article, we describe the structural modification of the tert-butyl phenoxy portion of lead compound 1 and the subsequent discovery of a novel series of conformationally constrained ortho-anilino diaryl ureas. In particular, spiropiperidine indoline-substituted diaryl ureas are described as potent, orally bioavailable small-molecule P2Y₁ antagonists with improved activity in functional assays and improved oral bioavailability in rats. Homology modeling and rat PK/PD studies on benchmark compound 3l will also be presented. Compound 3l was our first P2Y₁ antagonist to demonstrate a robust oral antithrombotic effect with mild bleeding liability in the rat thrombosis and hemostasis models.

Full text of DOI:10.1021/jm4013906

Subscriber access provided by LAURENTIAN UNIV
Article
Conformationally Constrained ortho-Anilino Diaryl
Ureas. Discovery of 1-(2-(1'-Neopentylspiro[indoline-3,4'-
piperidine]-1-yl)phenyl)-3-(4-(trifluoromethoxy)phenyl)urea;
1
a Potent, Selective and Bioavailable P2Y Antagonist
Jennifer X Qiao, Tammy C Wang, Réjean Ruel, Carl Thibeault, Alexandre L’Heureux, William
A Schumacher, Steven A. Spronk, Sheldon Hiebert, Gilles Bouthillier, John Lloyd, Zulan
Pi, Dora M Schnur, Lynn M. Abell, Ji Hua, Laura A Price, Eddie Liu, Qimin Wu, Thomas E
Steinbacher, Jeffrey S Bostwick, Ming Chang, Joanna Zheng, Qi Gao, Baoqing Ma, Patricia
A McDonnell, Christine S Huang, Robert Rehfuss, Ruth R. Wexler, and Patrick Y. S. Lam
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 28 Oct 2013
Downloaded from http://pubs.acs.org on November 3, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Conformationally Constrained ortho-Anilino Diaryl
Ureas. Discovery of 1-(2-(1'-Neopentylspiro[indoline-
3,4'-piperidine]-1-yl)phenyl)-3-(4-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(trifluoromethoxy)phenyl)urea; a Potent, Selective and
Bioavailable P2Y Antagonist
1
*
Jennifer X. Qiao, Tammy C. Wang, Réjean Ruel, Carl Thibeault, Alexandre L’Heureux, William A.
Schumacher, Steven A. Spronk, Sheldon Hiebert, Gilles Bouthillier, John Lloyd, Zulan Pi, Dora M.
Schnur, Lynn M. Abell, Ji Hua, Laura A. Price, Eddie Liu, Qimin Wu, Thomas E. Steinbacher, Jeffrey S.
Bostwick, Ming Chang, Joanna Zheng, Qi Gao, Baoqing Ma, Patricia A. McDonnell, Christine S.
Huang, Robert Rehfuss, Ruth R. Wexler, and Patrick Y. S. Lam
Research and Development, Bristol-Myers Squibb Company, 311 Pennington-Rocky Hill Road,
Pennington, NJ 08534
RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required
according to the journal that you are submitting your paper to)
ABSTRACT
1
Preclinical antithrombotic efficacy and bleeding models have demonstrated that P2Y antagonists are
efficacious as antiplatelet agents and may offer a safety advantage over P2Y12 antagonists in terms of
reduced bleeding liabilities. In this paper, we describe the structural modification of the tert-butyl
ACS Paragon Plus Environment

Page 3 of 52
Journal of Medicinal Chemistry
phenoxy portion of lead compound 1 and subsequent discovery of a novel series of conformationally
constrained ortho-anilino diaryl ureas. In particular, spiropiperidine indoline substituted diaryl ureas are
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
described as potent, orally bioavailable small molecule P2Y antagonists with improved activity in
1
functional assays and improved oral bioavailability in rats. Homology modeling and rat PK/PD studies
1
on the benchmark compound 3l will also be presented. Compound 3l was our first P2Y antagonist to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
demonstrate a robust oral antithrombotic effect with mild bleeding liability in the rat thrombosis and
haemostasis models.
KEYWORDS
Antiplatelet, P2Y antagonists, conformationally constrained ortho-anilino diaryl ureas, spiropiperidine
1
1
indolines, homology modeling of P2Y , pharmacokinetic, rat models of thrombosis and bleeding
INTRODUCTION
Adenosine 5′-diphosphate (ADP) acts as a key activator of platelets in the process of haemostasis and
thrombosis. Platelet activation is pivotal in thrombus formation under high shear stress such as in
arterial circulation. Recent data also suggest that platelet activation may be important in mediating
thrombus formation under lower shear stress, such as in venous circulation. ADP activates platelets by
simultaneously interacting with two platelet G protein-coupled receptors (GPCRs), P2Y
1
and P2Y12, to
and
produce two intracellular signals that synergistically induce complete platelet activation. Both P2Y
1
P2Y must be present and functional for the complete effect of ADP-induced platelet aggregation. The
1
2
2+
P2Y
1
q
receptor, coupled to G-protein q (G ), mobilizes intracellular Ca . This transitory increase of
2+
intracellular free Ca leads to a change of the platelet shape and initiates a weak, transient platelet
i
aggregation in response to ADP. The P2Y12 receptor, on the other hand, coupled to G-protein i (G ),
completes and amplifies platelet aggregation in response to ADP and generates a stable platelet
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 52
1
aggregate. The blockade of either receptor, therefore, significantly decreases both in vitro and ex vivo
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
ADP-induced platelet aggregation and thrombosis formation. As a result, an antagonist that specifically
binds to either of the two receptors could theoretically halt ADP-induced platelet aggregation.
The P2Y receptor is a well-established and clinically validated target for antithrombotic therapy
1
2
because of its central role in the formation and stabilization of a thrombus and its relatively limited
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
tissue distribution in platelets compared to P2Y
1
. The irreversible P2Y12 antagonist clopidogrel (Figure
4
1
A), a thienopyridine analog acting as a prodrug, is in world-wide clinical use for the treatment of
peripheral artery disease and acute coronary syndrome as well as for the secondary prevention of
vascular ischemic events, stroke, myocardial infarction, and vascular death. Combination therapy of
clopidogrel with aspirin is the current standard of care for most patients with non-ST-elevated acute
coronary syndromes (NSTE-ACS), percutaneous coronary intervention (PCI) and stent replacement.
Besides clopidogrel, additional P2Y antagonists were recently approved in the United States including
1
2
5
6
prasugrel and ticagrelor (Figure 1A), with others in development.
In 1999, preclinical studies using P2Y
1
-deficient (P2Y −/−) mice and selective adenine nucleotide-
1
derived small molecule P2Y
1
antagonists introduced P2Y
1
antagonism as a new antiplatelet
7
a−g
−/−
mechanism.
In that year, the Leon laboratory reported that P2Y
1
mice showed resistance to
thromboembolism induced by infusion of ADP or a mixture of collagen and adrenaline with no
7
a
spontaneous bleeding tendency; the lab of Fabre independently confirmed the phenotyping of the
−/−
7b
P2Y
1
mice in a separate model of thromboembolism induced by a mixture of collagen and ADP.
P2Y −
/−
mice were also shown to be resistant to thrombin-dependent tissue factor-induced
1
7
c
thromboembolism by injection of thromboplastin and displayed significantly less localized arterial
7
d
thrombosis generated by ferric chloride- or laser-induced injury of mouse mesenteric arteries. These
results were subsequently extended to include the inhibition of both venous and arterial thrombosis in
−/−
P2Y
1
rats. Similar results were observed in mice or rats treated with P2Y
1
antagonists, the
7
c−f
7g
bisphosphate nucleotide derivatives MRS2179
and MRS2500 (Figure 1B). The latter, the more
ACS Paragon Plus Environment

Page 5 of 52
Journal of Medicinal Chemistry
potent and selective P2Y
1
antagonist of the two, strongly inhibited both systematic thromboembolism
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
after intravenous injection of collagen and epinephrine and thrombus formation in laser-injured
mesenteric arteries in vivo. Experimental data also showed that additive effects of inhibition of
7d
thrombosis formation were observed when clopidogrel was combined with either P2Y -deficiency or a
1
7g
P2Y
1
antagonist.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
F
F
A
NH
O
N
H
COOCH3
N
N
N
N
N
H COCO
3
HO(H C) O
N
2 2
S(CH ) CH3
S
2
2
Cl
clopidogrel
S
F
ticagrelor
prasugrel
HO
OH
B
O
N
NHCH3
O
N
NHCH3
N
HO P O
N
HO P O
N
R
OH
O
HO P O
OH
O
N
N
OH
O
N
HO P O
OH
MRS2500 R = I
MRS2279 R = Cl
MRS2179
Figure 1. A. Three P2Y12 antagonists in clinical use: thienopyridines clopidogrel and prasugrel and reversible ticagrelor. B.
Three representative nucleotide-derived P2Y antagonists from the MRS series.
1
Our in-house animal models of antithrombotic efficacy and bleeding with selective P2Y
such as t-butylphenoxypyridine urea analog 1, as well as clopidogrel, demonstrated that P2Y
inhibition provided equivalent antithrombotic efficacy in terms of blocking aggregation and reducing
1
antagonists
1
and P2Y12
1
thrombus weight, while P2Y antagonists may offer safety advantages in terms of a reduced bleeding
8a-d
liability.
1
P2Y may therefore represent a promising new target for the development of new
antiplatelet therapies.
1
The aforementioned bisphosphate adenine nucleotide analogs, although potent P2Y antagonists
exhibit poor pharmacokinetic profiles and limited oral bioavailability. To date, the non-nucleotide P2Y1
9
antagonists reported in the literature have not been sufficiently optimized. Therefore, the discovery of
potent orally bioavailable small molecule P2Y antagonists remains an important goal.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 52
Researchers at Bristol-Myers Squibb recently disclosed a series of 2-(phenoxypyridine)-3-phenylureas
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with good in vitro potency, represented by compound 1, which was derived from a HTS hit series.
Compound 1 was the first in-house P2Y antagonist to demonstrate thrombus weight reduction and a
1
8d
lower bleeding liability than clopidogrel in rat models of thrombosis and bleeding. Compound 1 was
1
0
i 1
relatively potent (K of 6 nM) in the P2Y primary membrane binding assay, but had only moderate in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
vitro antiplatelet activity in the ADP-induced platelet aggregation assay in platelet-enriched human
plasma (PA IC50 = 2.1 μM against 2.5 μM of ADP). The platelet aggregation assay is thought to be
more predictive of in vivo activity than the membrane binding assay and is sensitive to plasma protein
binding. In addition, poor aqueous solubility and low bioavailability precluded compound 1 from being
further progressed. Subsequent SAR studies including substitutions on all parts of the molecule,
including changing the urea linker to other one- to four-atom linear linkers as well as urea mimetic
linkers such as aminoheterocycles, and incorporation of polar groups to reduce plasma protein binding,
failed to produce a compound with significantly improved antiplatelet activity, making it difficult to
1
2
advance this series.
Design Strategy
Concurrent to the lead optimization efforts on the t-butylphenoxypyridinyl phenylurea series, we also
1
3
searched for novel structurally diverse small molecule P2Y
molecular modeling studies indicated that compound 1 and the bisphosphate nucleotide derivatives
bound at least in part to different regions of P2Y , therefore the published SAR and homology models of
1
antagonist chemotypes. Preliminary
1
the MRS series was difficult to apply to our design of novel bioavailable P2Y antagonists.
1
Several structural modifications to constrain the t-butylphenoxy pyridine region of compound 1 were
explored. In the most successful approach, we rigidified the t-butylphenoxy pyridinyl region and
reduced the rotatable bonds by replacing the ether oxygen atom in the t-butylphenoxy group with a
nitrogen and conformationally constrained the resulting 2-anilino diaryl urea by tying the nitrogen atom
back to the quaternary carbon of the t-butyl group (Figure 2). This led to the formation of N-arylated
ACS Paragon Plus Environment

Page 7 of 52
Journal of Medicinal Chemistry
bicyclic [6,5] and [6,6] ring systems 2a−2j, such as indoline/tetrahydroquinoline analogs 2a−2d. Further
modification of the 3,3-dimethyl region in indolinyl analog 2a by the addition of bulkier spirocycloalkyl
rings (2k−2n), introduction of a protonatable nitrogen into potent spiro[cyclohexane-1,3'-indoline]
analog 2n, and subsequent substitution of that nitrogen led to the discovery of the spiro[indoline-3,4'-
piperidine] analogs 3a−3l. Nitrogen walk on the indoline ring of compound 3l led to azaindoline analogs
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
m−3p. Changing the spirocycle ring size and the position of the nitrogen atom resulted in compounds
q−3s.
Conformationally constrained ortho-anilino diaryl ureas
OCF
3
OCF3
O
O
X
O
X
N
N
OCF3
N
H
N
H
H
N
H
N
N
N
N
N
O
H
H
OCF3
R
N
O
N
X
H
2a, X = CH
R 2e, X = CH
2f, X = CH
2g, X = CHMe, R = H
2
CH
, R = H
2
2
, R = H
2
c, X = CH
H
2
b, X = N
2d, X = N
1
2
2
h, X = O, R = H
i X = O, R = Me
Early lead
2j, X = NMe, R = H
P2Y1 K 6 nM
i
P2Y1 Flipr IC50 29 nM
PA IC50 2.0 μM
O
N
@
2.5 μM ADP
privileged
substructure
OCF3
N
N
H
H
2k, n=1
n
2l, n=2
2
m, n=3
O
2
-Aryl indole as privileged substructure
2n, n=4
7
OCF
3
N
OCF3
6
N
H
N
O
increasing solubility by insertion of
a protonatable nitrogen
N
H
X
5
N
N
3m,N6
3n, N4
3
3
H
H
4
NH
azaindolines
o, N5
p, N7
N
O
4
5
1 i
, X = CH, P2Y K = 75 nM
, X = N, P2Y1 Ki = 227 nM
OCF3
N
N
N
H
OCF3
H
O
alternate ring size
N
N
H
H
N
N
N
R
3q
Spiropiperidine indoline
Spiro[indoline-3,4'-piperidine])
(
3
a, R = CBz
3b, R = H
3
3
3
c, R = CH COOH
2
alternate nitrogen
position
d, R = CH CH OH
2
2
O
e, R = CH CH OMe
2
2
OCF3
3f, R = COOMe
N
N
N
H
3
3
3
g, R = CONHCH(CH3)2
h, R = Me
i, R = CH(CH3)2
H
3r, m=2
3s, m=1
3j, R = CH(CH )CH CH
3k, R = CH Ph
m
3
2
3
N
2
3
l, R = CH C(CH )
2 3 3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 52
Figure 2. Design of conformationally constrained ortho-anilino diaryl ureas as t-butylphenoxy diaryl urea replacements
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
Compounds of tables 1−4).
Because the phenoxypyridinyl group can be viewed as a privileged substructure for GPCR targets, we
14
also investigated other privileged structures, such as 2-aryl indoles as phenoxypyridinyl replacements.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The initial 2-phenylindole analog 4 and 2-pyridinyl indole analog 5 gave 75 nM and 227 nM binding
affinity towards P2Y , but neither was active in the ADP-induced platelet aggregation assay. Moreover,
1
as observed in the t-butylphenoxy series, further modification of these two analogs did not yield
compounds with significantly improved binding affinity and these compounds did not show in vitro
antiplatelet activity.
Herein, we describe the structural modification of the tert-butylphenoxy portion of compound 1 and
the discovery of a novel series of conformationally constrained ortho-anilino diaryl ureas. In particular
this work resulted in a series of potent and orally bioavailable spiropiperidine indoline substituted diaryl
ureas. The SAR of the spiropiperidine indoline ring, including piperidinyl nitrogen substitution leading
to 3l (3a−3l), azaindolinyl analogs (3m−3p), and variations of ring size and the position of the nitrogen
atom (3q−3s), will also be discussed. Finally, homology modeling and PK/PD studies on the benchmark
compound 3l will be presented.
Results and Discussion
CHEMISTRY
Schemes 1−10 illustrate the synthesis of compounds described herein. Scheme 1 depicts the
synthesis of various 3,3-disubstituted indolinyl diaryl ureas as conformationally constrained ortho-
anilino systems (2a, 2b, and 2l−2o). Calcium hydride-induced Brunner reaction of
cyclobutanecarboxylic acid 2-phenyl-hydrazide led to the formation of spiro(cyclobutane-1,3'-oxindole)
6
, whereas 3,3-dimethyl (8) and spiro(cyclopropane-, cyclopentane-, and cyclohexane-1,3'-oxindoles)
(
9, 12, and 13) were prepared via double alkylation of 1-acetyloxindole or oxindole with the
ACS Paragon Plus Environment

Page 9 of 52
Journal of Medicinal Chemistry
corresponding alkyl iodides or alkyl dibromides using LiHMDS as the base. Reduction of oxindoles 6,
, 9, 12, and 13 with lithium aluminum hydride afforded indolines 7, 10, 11, 14 and 15, respectively.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
8
Heating the neat mixture of indolines and 2-fluoronitrobenzene or 2-fluoro-3-nitropyridine under
microwave irradiation gave the nitro analogs 16a, 16b, and 17−20. Reduction followed by urea
formation led to the indolinyl analogs 2a, 2b, and 2k−2n.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 1
H
H
N
O
N
O
a,b,c
d
+
Cl
NH
NH
2
6
7
H
H
N
Ac
N
N
e
d
O
R2
O
2
R
R1
R1
R1 = R = CH
2
10 R1 = R = CH
2
8
3
3
1
2
1
2
9 R ,R = -(CH
2
)
2
-
2 2
11 R ,R = -(CH ) -
H
N
H
N
H
N
e
d
O
O
X
X
1
2 X = CH
3 X = CH
2
2
,
14 X = CH
15 X = CH
2
2
,
1
CH
2
CH
2
H
X
X
O
N
NO
2
OCF
3
f
NO
g,h
N
2
+
N
N
H
N
R2
H
X
F
R1
R2
R2
2a, 2k−2n
R1
R1
7
, 10, 11, 14, 15
1
2
2a X =CH, R
2b. X =N, R
2k X =CH,R
1
= R
= R
,R
2
= CH3
1
1
6a X =CH, R = R = CH
3
1
2
1
2
= CH3
6b X =N, R = R = CH
3
1
2
1
2
= -(CH ) -
= -(CH ) -
17 X =CH,R ,R = -(CH
2
2
2
2
)
2
3
4
5
-
2 2
1
2
2l X =CH,R
2m X =CH,R
2n X =CH, R
1
,R
2
1
1
8 X =CH,R ,R = -(CH
)
)
)
-
-
-
2 3
1
2
1
1
,R
,R
2
2
2 4
= -(CH ) -
= -(CH ) -
2 5
9 X =CH,R ,R = -(CH
1
2
20 X =CH,R ,R = -(CH
a
Reagents and conditions: (a) Et
3
N, CH
−X−CH
Cl, MeOH, 67−95%; (h) OCN-Ph-p-OCF
2
Cl
CH
2 2
2
; (b) CaH
Br, LiHMDS, 87−90%; (f) neat, microwave, 175 °C, 1200s; 40−95%; (g) Zn,
, CH Cl , 78−93%.
2
, 240 °C; (c) HCl, MeOH, H
2
O; 78−86% for 3 steps; (d) LiAlH
4
,
THF, 77−92%; (e) MeI or BrCH
NH
2
4
3
2
2
Scheme 2 shows the synthesis of several bicyclic [6,6] ring systems such as compounds 2c and 2j.
Treatment of aniline and 3-methylbut-2-enoyl chloride followed by Friedel-Crafts alkylation on the
resulting amide with trimethylaluminum and LAH reduction afforded the 4,4-dimethyl
tetrahydroquinoline 21. Nucleophilic aromatic substitution of 21 with 2-fluoronitrobenzene under the
same conditions as employed to prepare the indolinyl analogs above (Scheme 1) did not yield any
desired product. However, we found that using 2,4,6-lutidine as the base at high temperature (>200 °C)
under microwave irradiation, the reaction proceeded to the desired nitro analog 23, which after
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 52
hydrogenation gave the aniline 22 in 35% yield. On the other hand, Buchwald-Hartwig cross-coupling
of 21 with 2-bromonitrobenzene in the presence of Pd (dba) and rac-BINAP followed by zinc
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
3
reduction afforded the aniline 22 in an improved 73% yield. Surprisingly, an attempt to further improve
the yield by replacing the 2-bromonitrobenzene with 2-iodonitrobenzene gave rise to a C−C coupled
product 23 in 18% yield and only a 2% yield of the desired nitro product was obtained in the presence of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pd(OAc) , rac-BINAP and KO-t-Bu in refluxing toluene. Addition of bromo ethylacetate to o-
2
phenylenediamine led to tetrahydroquinoxalinone intermediate 24. Reductive amination of 24 with
paraformaldehyde and sodium cyanoborohydride gave the N-methylated derivative 25, which was
reduced with LAH to afford 1-methyl-1,2,3,4-tetrahydroquinoxaline (26). Nucleophilic aromatic
substitution of 2-fluoronitrobenzene with 26 using KO-t-Bu as the base in DMSO afforded the nitro
intermediate, and 2j was subsequently obtained using similar transformations leading to 2c. Compounds
2
d−2i (Figure 2) were synthesized similarly following the procedures of 2c or 2j.
a
Scheme 2
OCF
3
O
H
2
O N
NH
2
H
N
NH
2
N
N
N
H
a,b
c or d
e
H
N
N
22
2c
23
21
OCF
3
O
H
H
H
N
N
N
NH
2
2
f
N
O
N
O
i,j,k
H
H
g
h
N
N
NH
N
H
N
N
26
2j
24
25
a
Reagents and conditions: (a) 3-methylbut-2-enoyl chloride, CHCl
, THF, 0 °C, 93%; (c) with 2-fluoronitrobenzene, neat, 2,4,6-lutidine, microwave, 220−250 °C, 3600s, then Pd/C
10%), MeOH, 35%; (d) with 2-bromonitrobenzene, Pd (dba) , rac-BINAP, Cs CO , toluene, 100 ºC, 16 h; then Zn, NH Cl,
3 3
, reflux, 2 h; then AlCl , toluene, 80 °C, 2.5 h, 80%; (b)
LiAlH
(
4
2
3
2
3
4
MeOH, 82%; (e) OCN-p-Ph-OCF
NaBH CN, (CHO) , MeOH, HOAc, 98%; (h) LiAlH
3 n 4
3
, CH
2
Cl
2
, 77%; (f) BrCH
,THF, rt, 95%; (i) 2-fluoronitrobenzene, KOt-Bu, DMSO, 80 °C, 16 h,
5%; (j) Pd-C (10%), H , rt, 2 h, 90%; (k) OCN-p-OCF -Ph, CH Cl , 76%.
2 3
COOEt, Et N, DMF, rt, 16 h then 80 °C, 3 h, 80%; (g)
4
2
3
2
2
The initial synthesis of the N-isobutyl and N-neopentyl spiropiperidine indoline diaryl ureas 3j and
l is shown in Scheme 3. Spiro[indoline-3,4'-piperidine] (27) was prepared according to modified
3
literature procedures by dialkylation of 2-fluorobenzeneacetonitrile with bis(2-chloroethyl)carbamic
acid 1,1-dimethylethyl ester followed by deprotection of the Boc group and then reductive
1
5
spirocyclization. The more basic nitrogen in 27 was selectively protected with benzyl 2,5-
ACS Paragon Plus Environment

Page 11 of 52
Journal of Medicinal Chemistry
dioxopyrrolidin-1-yl carbonate to give 28. Heating the neat mixture of 28 and 2-fluoronitrobenzene
under microwave irradiation gave the corresponding nitro intermediate in 70% yield, which underwent
reduction to give the aniline 29 followed by urea formation to afford the urea 3a. Removal of the Cbz
group in 3a and then reductive amination with either dimethyl- or trimethylacetaldehyde provided
compounds 3j and 3l, respectively. Single crystal X-ray analysis confirmed the structure of the HCl salt
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
6
of 3l. In the crystal structure, the piperidinyl nitrogen is protonated and the Cl anion forms three
hydrogen bonds with the protonated piperidinyl nitrogen and the two urea NH, and connects individual
molecules into a one-dimensional chain. The urea group adopts a cis orientation (i.e., “W” shaped) in
the solid state.
a
Scheme 3
H
H
N
Cl
Cl
N
HCl
HN
NH2
a,b,c,d
f,g
N
e
F
N
N
CN
H
NO
2
Cbz
28
N
2
9
2
7
F
Cbz
O
O
OCF
3
OCF
3
N
H
N
H
N
N
h
i
N
N
H
H
3
b R = H
3j R = isobutyl
l R = neopentyl
j
N
Cbz
k
N
R
3
3
a
a
2 3 2 2
Reagents and conditions: (a) (Boc) O, Et N, CH Cl , 90%; (b) 2-fluorophenyl-acetonitrile, NaH, THF, 65%; (c) 4 M
HCl/dioxane, reflux, 75%; (d) LiAlH(OEt) (freshly prepared by EtOH+LAH in ethylene glycol/dimethyl ether), reflux, 2
3
days, 77%; (e) benzyl 2,5-dioxopyrrolidin-1-yl carbonate, THF, rt, 62%; (f) 200 °C, microwave, 20 min, 70%; (g) Zn,
NH
4 3 2
Cl, MeOH; (h) OCN-Ph-p-OCF , THF, reflux, 67% for two steps; (i) H , 10% Pd/C, MeOH, HOAc, 96%; (j)
(CH ) CHO, NaB(OAc) H, ClCH CH Cl, HOAc, 72%; (k) t-BuCHO, CH(OMe) , HOAc, NaBH(OAc) , NMP, 58%.
3
2
3
2
2
3
3
A larger scale synthesis of compound 3l was carried out as depicted in Scheme 4. Fischer indole
synthesis followed by in situ reduction of the indolenine intermediate was applied to construct the
spiropiperidine indoline core. Treatment of 1-neopentylpiperidine-4-carbaldehyde (30) with phenyl
.
hydrazine while heating at 50 °C in HOAc followed by addition of BF
3
Et
2
O and heating at 80 °C
yielded the indolenine intermediate, which underwent in situ reduction with NaBH in MeOH to afford
4
the desired indoline 31. The spiroindoline-3,3'-piperidine analog 3r (Figure 2) was prepared via a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 52
similar sequence starting from the Fischer indole synthesis of benzyl 3-formylpiperidine-1-carboxylate
and phenylhydrazine followed by in situ reduction of the resulting imine.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 4
H
OH
N
CHO
N
CO Et
NH2
2
HN
c
d-f
a, b
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
N
N
N
H
31
30
O
OCF3
NO2
N
h,i
g
N
N
H
N
H
N
N
32
3l
a
Reagents and conditions: (a) (CH
3
)
3
CCOCl, CH
2 2 3 4 2 2 2
Cl , Et N, 94%; (b) LiAlH , THF, 91%; (c) (COCl) , DMSO, CH Cl ,
9
5%; (d) HOAc, 50 °C, 2 h; (e) BF −Et O (3 equiv), 80 °C, 1.5 h; (f) NaBH , MeOH, rt, 38% over 3 steps; (g) 2-
3
2
4
bromonitrobenzene, Pd (dba) , rac-BINAP, Cs CO , toluene, 100 °C, 80%; (h) H , Pd-C (10%), MeOH/EtOAc (2:3), 88%;
2
3
2
3
2
(
3
i) NCO-Ph-p-OCF , THF, 99%.
Schemes 5−8 illustrate the synthesis of azaindolinyl analogs 3m−3p. The synthetic route for the
spiro[6-azaindoline 3,4'-piperidine]diphenyl urea analog 3m is illustrated in Scheme 5. Nucleophilic
aromatic substitution of the anion derived from nitrile 33 and KHMDS as the base with 3-fluoro-4-
chloropyridine selectively displaced the chlorine atom at the 4-position of the pyridine ring to provide
the fluoride 34. Intramolecular cyclization of 34 to form the Boc-protected spiropiperidine 6-azaindoline
35 was achieved using lithium tri-tert-butoxy aluminum hydride in 1,4-dioxane heating at 130 °C under
microwave irradiation. Urea 36 was prepared using the same sequence as that employed in 3l.
Transforming the Boc group to the pivaloyl group followed by reduction afforded the desired 6-aza
indolinyl analog 3m.
a
Scheme 5
ACS Paragon Plus Environment

Page 13 of 52
Journal of Medicinal Chemistry
H
1
2
3
4
5
6
7
N
N
O
N
Cl
N
CN
OCF
3
N
F
c, d, e
N
H
N
a
b
N
H
F
N
N
NC
NBoc
Boc
Boc
36. R = Boc
N
34
3
5
f, g
3
3
R
3m. R = neopentyl
8
a
Reagents and conditions: (a) KHMDS, toluene, 0−23 °C, 64%; (b) lithium tri-tert-butoxy aluminum hydride, 1,4-dioxane,
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
microwave, 130 °C, 18%; (c) 2-bromonitrobenzene, Pd
2
(dba)3, rac-BINAP, Cs
2
CO
3
2
, toluene, 100 °C, 80%; (d) H , Pd-C
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
6
10%), MeOH/EtOAc (2:3), 96%; (e) OCN-Ph-p-OCF , THF, 74%; (f) TFA, CH Cl , 68%; (g) Piv-Cl, DIPEA, CH Cl ,
3%; (h) LiAlH , THF, 57%.
4
3
2
2
2
2
6
-Chloro-4-azaindolinyl analog 37 was prepared in an analogous fashion to 35 starting from 2,3-
difluoro-4-chloropyridine (Scheme 6). The chlorine in 37 was then reduced at the aniline stage (38 to
9) and the neopentyl group was introduced at the end of the synthesis via reductive amination with
3
polymer supported cyanoborohyride to afford spiro[4-azaindoline 3,4'-piperidine] diphenyl urea 3n.
a
Scheme 6
H
Cl
F
Cl
N
Cl
F
F
NC
a
b
c, d
+
N
N
N
N
Boc
NC
N
Boc
N
33
3
7
Boc
O
NH2
NH2
Cl
OCF3
N
N
N
f,g,h
e
N
H
N
H
N
N
N
N
N
N
3n
Boc
Boc
38
3
9
a
Reagents and conditions: (a) KHMDS, toluene, 0−23 °C, 96%; (b) lithium tri-tert-butoxy aluminum hydride, dioxane, 100
°
C, 21%; (c) 2-bromonitrobenzene, Pd (dba) rac-BINAP, Cs CO , toluene, 100 °C, 58%; (d) H , Pd-C (10%),
2 3, 2 3 2
2 3
MeOH/EtOAc (2:3), 64%; (e) H , Pd-C (10%), Degussa (wet), MeOH/EtOAc (2:3), 42 psi, 29%; (f) OCN-Ph-p-OCF , THF,
3
+
-
1%; (g) TFA, CH Cl ; (h) HCO-tBu, PSN(CH ) BH CN , CH COOH, CH Cl , 15% for 2 steps.
2 2 3 3 3 3 2 2
Scheme 7 illustrates the synthetic route to 5-azaindoline diphenyl urea 3o. The chloromethyl
17
dihydropyridine 40, prepared according to a literature method in three steps was used as an
electrophile to react with 3-bromo-2-chloropyridin-4-amine using LiHMDS as the base. The secondary
amine 41 was protected with a trifluoromethylacetyl group followed by an intramolecular Heck reaction
which afforded spiro[indoline-3,4'-tetrahydropyridine] intermediate 42 in 49% yield. Buchwald-Hartwig
cross coupling followed by reduction of the nitro group and urea formation provided 43. The double
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 52
bond in 43 was reduced with triethylsilane in the presence of trifluoroacetic acid, generating the
spiropiperidine 44. Reductive amination resulted in N-neopentyl analog 45. High pressure
hydrogenation of 45 afforded the spiro[5-azaindoline 3,4'-piperidine] diphenyl urea 3o.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 7
N
Cl
Br
NH2
Br
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H
Cl
O
N
N
g,h,i
a,b,c
e,f
HN
Cl
N
NO2
Br
N
d
Cl
N
N
Boc
Boc
40
N
41
42
Boc
Boc
O
O
O
OCF3
OCF3
j
OCF3
N
N
H
N
H
N
N
N
N
N
N
H
H
k
H
H
N
N
N
H
R
Cl
Cl
45 (R = Cl)
3o (R = H)
N
N
N
O
l
H
4
3
44
Boc
a
Reagents and conditions: (a) tetramethylpiperidine, n-BuLi, THF, -78 °C, 30 min, then methylphenyl sulfoxide, -78 to 23
C; (b) KO-tBu, tBuOH, 80 °C, 2 h, 31% for 2 steps; (c) SOCl , 2,6-lutidine, toluene, 40 °C, 30 min, 85%; (d) LiHMDS,
°
2
THF, 47%; (e) TFAA, Et N, CH Cl , 79%; (f) Pd(OAc) , Et N. tBu NBr, DMA, 90 °C, 49%; (g) Pd (dba) , rac-BINAP
3
2
2
2
3
4
2
3
Cs CO , toluene, 100 °C, 80%; (h) H , Pd-C (10%), EtOAc, 60 °C, 98%; (i) OCN-Ph-p-OCF , THF, 99%; (j) Et SiH, TFA,
2
3
2
3
3
+
+
CH
2
Cl
2
, 50%; (k) PSN (CH
3
)
3
B H
3
3 2 2 2
CN, CH COOH, CH Cl , 70%; (l) H , Pd-C (10%), Degussa (wet), EtOH, 60 psi, 80%.
The spiro[7-azaindoline 3,4′-piperidine] diphenyl urea 3p was prepared via dialkylation of 2-
(trimethylsilyl)ethoxymethyl (SEM)-protected 1,3-dihydro-2H-pyrrolo[2,3-b]pyridin-2-one (47)
Scheme 8). It is worth mentioning that the Fischer indole synthesis was an efficient method for the
(
construction of compound 3l and substituted indolinyl analogs; however, reaction of 2-
pyridinylhydrazine with N-Boc-4-piperidinylaldehyde failed to give any desired cyclization product
under several reaction conditions. The dialkylation of 47 was carried out using cesium carbonate in
DMF to afford the SEM-protected spiro[7-azaindoline 3,4′-piperidine] 48. Removal of the SEM group
was achieved in a two-step one-pot process (48 to 50). Acylation of 50 followed by reduction afforded
the N-neopentyl spiro[7-azaindoline 3,4′-piperidine] 51.
a
Scheme 8
ACS Paragon Plus Environment

Page 15 of 52
Journal of Medicinal Chemistry
SEM
R
N
SEM
N
1
2
3
4
5
6
7
8
9
SEM
N
H
N
O
N
N
N
d
e
N
a,b
c
N
O
N
O
O
Br
Br
46
N
4
8
N
47
Boc
H
4
5
9 R = CH
2
OH
0 R = H
O
H
OCF
3
N
N
H
N
N
g,h,i,j
N
N
f
H
N
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
1
3p
a
+
-
Reagents and conditions: (a) NaH, SEM-Cl, DMF, 100%; (b) Br
CH CH Cl) NBoc, Cs CO , DMF, 80 °C, 30%; (e) 4 N HCl in dioxane, rt, 1h, then ethylene glycol, 70 °C, 15 h; (f)
trimethylacetyl chloride, Hunig’s base, CH Cl , rt, 4 h, 75% for 2 steps; (g) LiAlH , THF, rt, 25 min; (h) 2-
Bromonitrobenzene, Pd dba , rac-BINAP, Cs CO , toluene, 100 °C, 16 h, 36% for 3 steps; (i) Zn, NH Cl, MeOH/EtOAc
2 4
, PyH Br , dioxane; (c) Zn, NH Cl; 77% for 3 steps; (d)
(
2
2
2
2
3
2
2
4
2
3
2
3
4
(
3
1:1), rt, 16 h; (j) OCN-Ph-p-OCF , THF, rt, 16 h, 9% for 2 steps.
2
',3'-Dihydro-spiro[piperidine-4,4'(1'H)-quinoline] diphenyl urea 3q was prepared as illustrated in
Scheme 9. The Boc protecting group in 1-oxospiro[indane-3,4'-piperidine] analog 52 was switched to
the trifluoroacetyl group. Ring expansion of the indene-3-one 53 via Schmidt rearrangement with
hydrogen azide, generated in situ from sodium azide and conc. sulfuric acid, gave the spiro[piperidine-
4
,4'(1'H)-quinolin]-2'(3'H)-one (54) in 54% yield after separation of the two regioisomers via flash
column chromatography. The trifluoromethylacyl group was then changed to the pivaloyl group and
treatment of 55 with LAH gave 56, which was converted to urea 3q as above.
The spiro[indoline-3,3'-pyrrolidine] diphenyl urea 3s was prepared as shown in Scheme 10.
Treatment of tetrahydro-β-carboline 57 with tert-butylhypochloritein the presence of triethylamine
afforded the chloroindolenine intermediate 58, which was poured directly into a refluxing mixture of
NaOH in MeOH to give the complete rearranged product, β-spiro[pyrrolidinoindolenine] 59. Reduction
of 59 with LAH afforded the spiropyrrolidine indoline 60.
a
Scheme 9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 52
O
N
H
N
O
H
N
O
O
O
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
NH
a-b
c
d-e
N
N
N
N
Boc
CF
3
O
O
O
CF
3
O
CF3
5
2
5
3
5
4
55
OCF3
O
H
N
N
N
H
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H
f
g-i
N
N
N
5
6
3q
a
Reagents and conditions: (a) TFA, CH
two steps; (c) NaN , conc. H SO , H O, CHCl
rt, 16 h; (e) t-BuCOCl, Et N, CH Cl , rt, 4 h, 79% for 2 steps; (f) LiAlH
Pd (dba) , rac-BINAP, Cs CO , toluene, 100 °C, 16 h, 90%; (h) Zn, NH Cl, EtOH, rt, 16 h, 63%; (i) OCN-Ph-p-OCF , THF,
2
Cl2, 0 °C to rt, 30 min; (b) (CF
3
CO)
2
O, Et
SO , rt to 50 °C, 54%; (d) KOH, H
4
, THF, rt, 16 h, 81%; (g) 2-bromonitrobenzene,
3
N, DMAP (cat.), CH
2
Cl
2
, rt, 2 h, 92% for
3
2
4
2
3
, 0 °C to rt, 3 h, then conc. H
2
4
2
O, MeOH,
3
2
2
2
3
2
3
4
3
rt, 3 days, 42%.
a
Scheme 10
N
H
N
N
OCH
3
a
b
c
N
N
N
Cl
5
7
58
59
O
H
N
N
OCF
3
d,e,f
N
H
N
H
N
N
60
3s
a
Reagents and conditions: (a) tBu-OCl, Et
3
N, CH
2
Cl
2
, rt; (b) NaOH, MeOH, reflux, 93% for 2 steps; (c) LiAlH
4 2
, Et O,
88%; (d) 2-bromonitrobenzene, Pd
66% for 2 steps.
2
(dba)
3
, rac-BINAP, Cs
2
CO , toluene, 75%; (e) H , Pd/C, (f) OCN-Ph-p-OCF , THF, rt,
3
2
3
In vitro biological results
The 3,3-dimethylindoline analog 2b was the first conformationally constrained ortho-anilino diaryl
urea made according to the aforementioned design strategy. It showed a three-fold drop of P2Y binding
1
affinity compared to 1 and had IC50 >10 μM in the ADP-induced platelet aggregation (PA) assay.
Surprisingly, changing the central 2-pyridyl ring to the phenyl ring (2a) improved both binding affinity
(K = 8.7 nM) and in vitro antiplatelet activity (PA IC50 = 8.5 μM at 2.5 μM of ADP). The same SAR
i
trend was observed with 4,4-dimethyl tetrahydroquinoline analogs 2c and 2d. Compound 2c had very
ACS Paragon Plus Environment

Page 17 of 52
Journal of Medicinal Chemistry
similar binding affinity and antiplatelet activity (K = 7.3 nM and PA IC = 2.2 μM, respectively) as the
i
50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
t-butylphenoxypyridine urea compound 1. Both methyl groups in 2c were required for binding, as
removing either one (2g) or both (2f) showed reduced potency. Ring expansion to the [6,7] bicyclic
compound 2e resulted in much lower binding affinity. Because of the promising activity shown with 2c,
we looked at several structurally related [6,6] bicyclic analogs of 2c hoping to reduce lipophilicity while
maintaining activity. While changing from the 4-carbon atom in 2f to an oxygen or N-methyl group (2h,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8
and 2j) did reduce HPLC logP (HPLC logP of 2h and 2j is 5.7 and 5.5 respectively, compared with >
6.2 for 2c), these compounds showed reduced binding affinity and significant loss of antiplatelet
activity. Adding a gem-dimethyl group at the C−3 position of the benzomorpholine 2h, which provided
2i, did not result in much improvement in potency.
Table 1. Human P2Y receptor binding (K ) and in vitro platelet aggregation (PA) activity of
1
i
a
selected N-aryl indolines, tetrahyhydroquinolines and related bicyclic analogs
OCF3
O
X
N
N
H
H
M
PA IC
a
50
Compd #
1
M
X
P2Y K (nM)
1 i
b
(
μM)
6.0±0.6
8.7±1.7
20.3±4.6
2.1±0.3 (n=7)
N
2a
CH
N
8.5
Me
Me
Me
N
2
2
b
c
>10
Me
N
CH
7.3±0.6
2.2
Me Me
N
2d
N
16.0±2.7
11.1±0.2
Me Me
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 52
N
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
e
f
CH
CH
830±5
NT
N
N
36.5±5.3
>10
2
g
CH
29.5±5.5
>10
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Me
N
2
2
h
i
CH
CH
65±15
39±11
>10
>10
O
N
O
N
2
j
CH
94±15
>10
N
Me
a
b
K
i
values are reported as the mean ±SD, n≥2. See reference 10 for details. PA activity was tested against 2.5 μM of ADP in
human platelet-rich plasma, n=1 unless stated. See reference 11 for details. NT: not tested.
The tetrahydroquinoline analog 2c had relatively low metabolic stability (percentage remaining after
1
0 min liver microsomal incubation: 51% (human), 6% (rat), and 63% (mouse)) presumably because of
the extensive oxidation on the saturated carbons of the 4,4-dimethyltetrahydroquinoline ring. Therefore,
subsequent SAR studies focused on analogs of the indolinyl 2a. Previous SAR suggested that the
hydrophobicity of the gem-dimethyl group in 2a was important for binding affinity and in vitro
functional activity. This was further explored with spirocycloalkyl analogs by varying the spirocycle
ring size. As shown in Table 2, increasing the carbocyclic ring size of the spirocycle from cyclopropyl
2k to cyclobutyl 2l to cyclopentyl 2m to cyclohexyl 2n led to improved binding potency and platelet
aggregation activity. Spirocyclopentyl analog 2m (K
analog 2n (K = 5.9 nM, PA IC50 = 2.0 μM) were equipotent to the t-butyl phenoxy lead 1 in both the
binding and the platelet aggregation assays. Insertion of a 4-nitrogen atom to introduce polarity into the
cyclohexyl ring in 2n, resulted in 3b with a three-order-of-magnitude reduction of P2Y binding affinity
but with improved aqueous solubility (2n: <1 μg/mL vs. 3b: 84 μg/mL at pH 6.5, amorphous).
i
= 6.6 nM, PA IC50 = 2.4 μM) and spirocyclohexyl
i
1
1
9
Interestingly, the synthetic precursor, the Cbz-protected analog 3a only showed a six-fold reduction in
ACS Paragon Plus Environment

Page 19 of 52
Journal of Medicinal Chemistry
binding affinity, indicating that the protonatable nitrogen atom may be tolerated if substituted
appropriately. Around this time, a new fluorescent imaging plate reader (FLIPR) P2Y functional
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
0
assay was developed. It measured the activation of the calcium signal transduction pathway by P2Y
1
in washed human platelets in a more physiologically relevant setting using the endogenously expressed
receptor and a reduced amount of protein (~0.1% BSA). In this assay, compound 1 had an IC50 of 29
nM, whereas the spirocyclohexyl analog 2n demonstrated a 10-fold increase in in vitro functional
activity (IC of 2.2 nM).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
1 i
Table 2. Human P2Y receptor binding (K ) and in vitro platelet aggregation (PA) activity
spirocycloalkyl analogs 2k−2n, and spiropiperidine indolinyl analogs 3a and 3b compared to
compound 1, and 2a
OCF
3
O
N
N
H
H
M
P2Y
(nM)
1
P2Y
(nM)
1
bFLIPR IC50
PA IC50 μM
Compd #
1
M
a
c
K
i
ADP at 2.5 μM
6.0±0.6
29±3 (n=4)
NT
2.1±0.3 (n=7)
N
2a
8.7±1.7
8.5
Me
Me
N
N
N
2
2
k
l
26.0±2.2
7.6±0.7
NT
>10
4.1
10.5±0.9
2
2
m
n
6.6±0.3
5.9±0.4
NT
2.4
2.0
N
2.2±0.2
N
3a
36.6±9.9
5.9±0.2
>10
N
Cbz
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 52
N
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
b
4500±450
NT
NT
NH
a
b
K
i
values are reported as the mean ±SD, n≥2. See reference 10 for details. FLIPR was tested with 0.5 nM of ADP, and IC50
values are reported as the mean±SD, n=2 unless stated, see reference 20 for details. PA activity was tested against 2.5 μM of
ADP in human platelet-rich plasma, n=1 unless stated. See reference 11 for details. NT: not tested.
c
We continued to investigate the N-1' substitution on the piperidinyl ring in 3b (see Table 3).
Polar substitutions such as acid (3c) and alcohol (3d) functional groups were not well tolerated. The
methylethyl ether analog 3e showed similar activity to the Cbz analog 3a but with improved aqueous
solubility. Neutralization of the basic NH in 3b with small functional groups such as carbamate (3f) and
urea (3g) groups improved binding affinity compared with the unsubstituted 3b, with methylcarbamate
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
f being the most potent (K = 13 nM). Larger liphophilic groups were more potent, a trend visible from
i
the N-methyl substituted 3h to N-isobutyl substituted 3j to the N-neopentyl substituted 3l; although as
expected with an increase in logP, reduced solubility was observed. A methylene linker between the
piperidinyl nitrogen atom and the aryl group (3k) or the bulky alkyl group (3l) was preferred, with the
neopentyl substitution in 3l being optimal.
A threshold concentration of 2.5 μM level of agonist ADP was initially used in the platelet
aggregation assay due to the relative low potency and high protein binding of the compounds tested.
1
Because the significant potency improvement seen in compounds such as 2n and 3l in the P2Y FLIPR
functional assay (IC50 = 2.2 and 2.5 nM for 2n and 3l, respectively), the agonist concentration for the PA
assay was increased to 10 μM, a concentration believed to be clinically relevant because it is the lowest
8
c
concentration that produces a maximum aggregation response in rat platelet rich plasma. To our
delight, compound 3l had a PA IC50 of 0.67 μM against 2.5 μM ADP and a PA IC50 of 4.9 μM against
1
0 μM ADP. This was the first compound in the program that demonstrated good antiplatelet activity
against 10 μM ADP.
Table 3. In vitro data of N-1' substitution of spiropiperidine indolinyl analogs 3b−3l
ACS Paragon Plus Environment

Page 21 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
O
O
OCF
3
insert a protonatable
nitrogen
OCF3
N
N
N
N
N
H
H
N
H
H
2
n
3
b−3l
N
R
b
b
Compd#
R
P2Y
1
PA IC50 μM
PA IC50 μM
Solubility
μg/ml
18
a
19
HPLC logP
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
K (nM)
at 2.5 μM ADP
at 10 μM ADP
i
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
b
H
4500±1900 NT
NT
84 (partial cryst.)
5 (amorphous)
123 (partial cryst.)
66 (amorphous)
1 (cryst.)
4.3
3c
CH
CH
CH
2
2
2
COOH
13500
NT
NT
4.4
4.6
5.1
6.0
5.7
4.6
4.5
5.5
>6.3
>6.2
3d
CH
CH
2
2
OH
4500±450
390±80
13.3±5.5
NT
NT
3
3
3
3
3
e
f
OMe
25.4
NT
COOMe
5.8
54.2
g
h
i
CONH-isopropyl 200±30
>10
NT
1 (partial cryst.)
100 (amorphous)
71 (partial cryst.)
7 (partial cryst.)
<1(partial cryst.)
Me
690±140
300±260
20.7±3.2
18.6±1.1
4.3±2.6
NT
NT
isopropyl
isobutyl
4.7±0.8 (n=2)
2.2±0,2 (n=3)
3.3±1.0 (n=3)
29.2±0.6 (n=2)
8.8±2.7 (n=2)
18.6
3j
3
k
CH
2
Ph
3
l
neopentyl
0.67±0.31(n=6) 4.9±1.7 (n=267) 4 (cryst.)
a
b
K
i
values are reported as the mean ±SD, n≥2. See reference 10 for details. PA activity was tested against either 2.5 μM of
ADP or 10 μM of ADP in human platelet-rich plasma, n=1 unless stated. See reference 11 for details. NT: not tested.
A nitrogen walk on the phenyl ring of the indoline ring provided four azaindoline analogs 3m−3p
(
Table 4). Among them, compounds 3m and 3n were much more potent in binding and more active in
the platelet activation assay compared with the other two azaindoline analogs 3o and 3p. In addition to
being most potent, the spiropiperine-6-azaindoline analog 3m also showed improved aqueous solubility
(
25 μg/mL at pH=6.5, albeit with amorphous compound). Modifying the spirocycle ring size of 3l led to
the [6,6]spirocycle analog 3q, which had activity comparable to 3l. Changing the position of the
piperidinyl nitrogen in 3l as shown in the spiro[indoline-3,3'-piperidine] 3r resulted in much reduced
activity compared with 3l, and the [5,5]spirocycle analog 3s, which, despite having similar binding
affinity as 3l, was 4-fold less active in PA assay.
Table 4. In vitro data of N-neopentyl substituted analogs 3m−3s
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 52
O
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
OCF3
N
N
M
H
H
N
N
N
N
N
N
N
N
N
N
N
N
N
N
M =
N
N
N
N
N
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3l
3m
3n
3o
3p
3q
3r
3s
19
P2Y
1
P2Y
1
bFLIPR IC50 PA IC50 (μM) c
Solubility
18
Compd #
a
HPLC logP
d
K (nM)
(nM)
at 10 μM ADP
μg/mL
i
3
3
3
3
3
3
3
l
4.3±2.6
2.5±0.5
10.7±3.8
NT
4.9±1.7 (n=267)
4
25
5
428
NT
16
<1
1
>6.2
>6.0
>6.2
5.1
m
n
o
p
q
r
11.5±1.0
28.7±7.8
617.2±99.8
253±35
4.0±2.7 (n=2)
15.1±2.7 (n=2)
346
>40
>40
6.3
>40
18.5
503±249
9.9±0.9
NT
NT
4.6±0.9
>6.2
>6.2
>6.2
12.8±4.0
5.8±0.9
3
s
NT
a
b
K
i
values are reported as the mean±SD, n≥2. See reference 10 for details. FLIPR was tested with 0.5 nM of ADP, and IC50
c
values are reported as the mean±SD, n=2, see reference 20 for details. platelet aggregation (PA) was tested with 10 μM of
d
ADP in human platelet-rich plasma, n=1 unless stated. See reference 11 for details. solubility data reported are from
amorphous material except for compound 3l which was crystalline. NT: note tested.
Pharmacokinetics
Based on potency in the in vitro antiplatelet assay and metabolic stability, compounds 3j, 3l, 3m and
3
q were selected for PK studies. Table 5 illustrates the in vitro profile and the rat PK data for these
compounds. In rats, compound 3l had the best profile with relatively low clearance (CL) and medium
volume of distribution (Vdss,3.1 L/kg), slightly high PO exposure, and 32% oral bioavailability. In dogs
using the same vehicle as the rat PK studies (0.1 mg/kg IV and 0.2 mg/kg PO), compound 3l
demonstrated low CL (5.3 mL/min/kg) and relatively small Vdss (2.8 L/kg) with an oral half-life of 9.7 h
and oral bioavailability of 55%. Compound 3l was highly bound to serum proteins with a free fraction
less than 0.2% in all five species tested (human: < 0.1%, rat: < 0.1%, mouse: 0.2%, rabbit: 0.1%, and
dog: 0.2%).
a
Table 5. In vitro potency and rat PK profile of representative compounds 3j, 3l, 3m, and 3q
d
d
P2Y
1
PA IC50
HLM
RLM
IV CL
IV Vdss
T
1/2
PO
F%
ACS Paragon Plus Environment

Page 23 of 52
Journal of Medicinal Chemistry
K
i
(nM)
(μM) at 10
μM ADP
(mL/min/kg) (L/kg)
(h)
AUC total
nM*h)
1
2
(
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
b
3
j
20.7
4.3
8.8
4.9
4.0
6.3
79
89
67
90
84
87
73
87
27.1
9.1
12.8
2.8
7.5
7.5
6.1
7
147
587
167
278
26
32
19
33
c
3l
b
3m
11.5
17.7
5.6
b
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
q
4.6
14.5
4.1
a. IV bolus and PO administration: dose: 0.5 mg/kg; vehicle: 10% Cremophor/10% N,N-dimethylacetamide
DMAC)/10%EtOH/70%H O. b. Cassette PK. c. Discrete PK. d. Percentage remaining after 10 min incubation. HLM:
human liver microsomes, RLM: rat liver microsomes.
(
2
Compound 3l was selective against P2Y12 and P2Y
2
with IC50 > 20 μM in the P2Y12 cell binding
2
1
22
assay and IC > 10 μM in the P2Y FLIPR functional assay and achieved > 600-fold selectivity
5
0
2
23
against P2X
1
1
in the FLIPR functional assay (IC50 = 1.6 μM for P2X ). Compound 3l was clean against
a broad kinase panel. In the MDS panel, 3l showed much cleaner profile compared to the previous
12a,d
disclosed lead molecules bearing amphiphilic amines,
but did show inhibition against opiate receptor
(non selective, IC50 = 2.7 μM), rat Ca channel L-type and Na channel site 2 (78 % and 88% at 10 μM
respectively. Despite these potential selectivity issues, compound 3l provided the best overall profile
and was selected as a benchmark compound and advanced into in vivo studies in rat models of
thrombosis and bleeding.
Binding model of Compound 3l
The single crystal X-ray structure of the HCl salt of the N-neopentyl spiropiperidine indoline 3l
showed that the urea group adopted a “cis” orientation. A cis urea was also preferred by 3l in solution,
as determined by 2D NMR experiments including ROESY and NOESY in either DMSO-d
O:DMSO-d (see supporting information for experimental details).
6
or 9:1
H
2
6
1
Efforts to develop a binding model of P2Y bound to compound 3l began with an externally
developed model of the receptor bound to antagonist MRS2279, a bisphosphate nucleotide derivative
24
(
Figure 1B), which was kindly provided by Kenneth Jacobson and Stefano Costanzi. The MRS model
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 52
was based on the structure of bovine rhodopsin, which was the only GPCR that had been crystallized at
the time. It was evident that refinement of the model was required, primarily because the binding site
was too small to fit compound 3l and its analogs, and the two antagonist series have little obvious
chemical similarity (for example, the MRS analogs have a high negative charge density, with two
phosphate groups, in contrast to the relatively hydrophobic and cationic compound 3l).
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Several mutants of residues in the binding site in the rhodopsin-based model —Tyr100Ala,
Phe131Ala, His132Ala, Asn197Ala, Thr201Leu, Thr221Ala, Phe226Ala, Phe276Ala, and Ser314Ala—
2
4a
were produced to provide experimental data for model refinement. The mutants were tested in the
33
aforementioned primary binding assay using [ P] 2-methylthioADP (2MeSADP) as a probe with a
wide range of P2Y antagonists from several chemotypes, including conformationally constrained
ortho-anilino diphenylurea compounds (e.g., 3l) and bisphosphate derivatives. Almost without
exception, the only ligand/mutant combinations for which the K values differed more than 10-fold from
1
i
the wild-type were diphenylurea compounds with the Phe131Ala mutant. It should be noted that five of
these mutants had been previously reported to be less responsive to 2MeSADP in an inositol phosphate
24a
accumulation functional assay. The difference in the mutant data may be due to the different assays:
these residues may be important for agonism but contribute little to overall binding. Nevertheless, the
fact that so many of these residues appear to be uninvolved in binding any of the chemotypes is strong
1
evidence that rhodopsin is not a suitable template for a homology model of P2Y .
25
A new homology model was constructed from the crystal structure of the CXCR4 receptor, which
has the greatest residue identity in the transmembrane domain (TM) regions with P2Y (24%) among all
1
receptors that had been crystallized to date. The CXCR4 was also recently found by another group to be
26
the best template for a reversible series of P2Y12 antagonists.
A binding model of compound 3l (Figure 3, cyan) was obtained by docking and refined by
molecular dynamics (MD). Consistent with the crystallography and NMR results, the ligand pose
contains a cis urea. Consistent with mutagenesis results, most of the mutated residues (Figure 3, green)
ACS Paragon Plus Environment

Page 25 of 52
Journal of Medicinal Chemistry
are relatively far from the ligand binding site and do not interact with the ligand in the model. However,
Phe131 (Figure 3, red) forms two crucial aromatic edge-to-face interactions via both the indoline phenyl
ring and the adjacent phenyl ring.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. A. Compound 3l (cyan carbons) bound in P2Y
1
. Key complex interactions are shown as orange dashes between 3l
and the labeled residues. Residues for which P2Y mutants were tested but had no effect on 3l binding are shown as green
1
spheres. B. The binding model with the protein represented by ribbons and surface. The view is rotated slightly from A, but
the colors of the residues are the same.
Key SAR data can be explained by the binding model. The fact that at least one hydrogen bond
donor is required in the region of the urea may be due to the hydrogen bonding with the side chain of
Asp204 on extracellular loop 2 (ECL2) (Figure 3, magenta). The original pose based on docking hinted
at these hydrogen bonds, and the dynamics indicated that they were reasonably stable. At least one
direct hydrogen bond between the urea and Asp204 was retained for more than 8 ns of simulation, and
even after both H-bonds were broken, the Asp204 side chain remained near, suggesting that the
interactions could break and reform easily. An alternative explanation for the requirement for the urea
“
left-hand” NH is that intramolecular interactions between this donor and the indoline N lone pair help
stabilize the bioactive conformation of the core. The phenyl ring of the OCF -phenyl may form a π-
3
stacking interaction with Tyr203 on ECL2 (Figure 3, purple), although this interaction was less
prevalent in the MD than the hydrogen bonds with Asp204. It should be noted that residues on ECL2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 52
are not always well-placed in GPCR homology models because of variability in this loop, but Tyr203
and Asp204 are directly adjacent to Cys202, which generally can be placed with greater certainty
because of the disulfide linkage formed with Cys124 on TM3. Mutagenesis of these two residues would
be required to confirm these interactions.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Another feature of the SAR explained by the model is the requirement for hydrophobic bulk in the
basic piperidinyl series. The only hydrogen bond acceptors in the region of the basic NH are backbone
carbonyls of TM7, so binding must result in either the uncompensated desolvation of the cationic amine
or the disruption of the α-helical hydrogen bonding. Either of these would probably be so unfavorable
that compensation by hydrophobic interactions alone would require a rather large hydrophobic contact
area. The model has a deformed peptide bond between Arg310 and Gly311 that allows the carbonyl of
Arg310 to accept the hydrogen bond from the ligand, although the formation of this interaction (as
opposed to maintenance of the usual peptide bond and the absence of any hydrogen bond with the
ligand) is not a crucial part of the model. Directly adjacent to the basic amine is a hydrophobic pocket
formed by the side chains of Leu54, Tyr58, and Tyr111 and the TM7 backbone, including Gly311
(
Figure 3, yellow). This pocket is filled up almost completely by the neopentyl group of compound 3l,
satisfying the requirement for a significant burial of hydrophobic surface. In addition, it is likely that the
alpha CH in the neopentyl group of 3l forms a cation−π interaction with the phenyl ring of Tyr111
2
2
7
based on the proximity of these two groups.
Scan of phenyl−urea dihedral angle
The urea group of compound 3l in the binding model is considerably out of plane with respect to the
core phenyl ring, with a dihedral angle of 123° between the urea’s proximal N−C bond and the phenyl
plane. To further ensure that this was a reasonable conformation, a torsional scan of a model of
compound 3l was performed, with energies determined as the sum of the quantum mechanical energy
(at the RIMP2/cc-pVTZ//B3LYP/6-31+G** level of theory) and the free energy of solvation (calculated
ACS Paragon Plus Environment

Page 27 of 52
Journal of Medicinal Chemistry
2
8
with Cosmo-RS methodology ) (Figure 4). The scan confirms that the conformer with this dihedral
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
angle has a low energy (0.25 kcal/mol) relative to the global minimum.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. A. Torsional scan relating the energy of the compound 3l model system (inset) to the dihedral angle τ. There is a
relatively broad well from -180° to -120°. The binding pose of 3l has τ = -123.5° (purple), which falls within this well. B.
The conformation of the model system after minimization with τ constrained at -123.5° (purple carbons) overlaid onto the
binding pose (cyan carbons). The urea, phenyl, and dihydropyrrole groups align well.
Pharmacology
The effect of the benchmark compound 3l on arterial and venous thrombosis and provoked bleeding
29
time was determined in pentobarbital-anesthetized rats using previously described procedures.
Occlusive thrombosis was induced by a topical application of FeCl
2
-saturated filter paper to either the
carotid artery (50% FeCl solution for 10 min) or vena cava (15% FeCl solution for 2 min), while
2
2
bleeding time was measured by needle puncture of mesenteric arteries. Given the good solubility and
bioavailablity of compound 3l observed in rat PK studies, the same vehicle was also used for both IV
and PO routes in the PD studies and compared to compound 1, which was dosed only IV.
Compound 3l inhibited arterial thrombus formation and improved blood flow in a dose-dependent
manner when administered by either IV (Figure 5) or PO (Figure 6) routes. Maximum thrombus weight
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 52
reductions of 66% and 64% were observed in comparison to vehicle at doses of 0.6 mg/kg + 2.25
mg/kg/hr, IV and 10 mg/kg, PO, respectively. At these dose levels, vascular occlusion was prevented in
all drug-treated rats and integrated blood flow was maintained at >95% of control levels.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
6
4
4
/6
/4
/6
Vehicle
6
5
4
3
2
1
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.02 + 0.075
0.06 + 0.225
0.2 + 0.75
0.6 + 2.25
-41%
proportion of the total
number of rats in which
the artery occluded
*
Compound3l
Thrombus
Weight
1/7*
(mg/kg + mg/kg/hr, i.v.)
-66%
0
/6
*
(mg)
*
*
p<0.05 compared to Vehicle
=SEM
6
5
*
*
1
00
4
Carotid
Blood
Flow
8
0
3
Integrated
Blood 60
Flow Over
(ml/min)
2
1
0
6
0 min
4
0
0
0
(% Control)
5
0% FeCl
2
2
0
10
20
30
40
50
60
Time (min )
Figure 5. Dose-dependent effect of compound 3l on arterial thrombosis in rats following parenteral dosing. Vehicle: 10%
Cremophor/10% DMAC/10% EtOH/70% H O. Efficacy was demonstrated by reduction in thrombus weight, prevention of
2
vascular occlusion and improved integrated blood flow; plasma exposure was determined at the end of the experiment.
7
7/8
5/8
Vehicle
0.3
6
5
4
3
2
1
0
-45%
proportion of the total
number of rats in which 7/8
the artery occluded 2/8*
1
Compound3l
*
Thrombus
Weight
(mg/kg, p.o.)
3
-64%
0/7
*
10
*
(mg)
*
p<0.05 compared to Vehicle
=SEM
6
5
4
3
2
1
0
*
*
1
00
Carotid
Blood
Flow
80
60
40
Integrated
Blood
Flow Over
60 min
(
ml/min)
(% Control)
5
0% FeCl
2
20
0
0
10
20
30
40
50
60
Time (min )
Figure 6. Dose-dependent effect of compound 3l on arterial thrombosis in rats following oral dosing. Vehicle: 10%
Cremophor/10% DMAC/10% EtOH/70% H O. Efficacy was demonstrated by reduction in thrombus weight, prevention of
vascular occlusion and improved integrated blood flow.
2
ACS Paragon Plus Environment

Page 29 of 52
Journal of Medicinal Chemistry
All doses of compound 3l which inhibited arterial thrombosis (Figure 5 and Figure 6) significantly
inhibited ex vivo platelet aggregation responses to ADP by either parenteral (Figure 7A) or oral routes
(Figure 7B). Thus, the 0.6 mg/kg + 2.25 mg/kg/hr dose of compound 3l was associated with strong 85%
inhibition of platelet aggregation response to 10 μM ADP and a modest increase in mesenteric bleeding
time (MBT) (2.3 times control). A similar profile had been observed with maximum efficacious doses
of compound 1 (10 mg/kg + 10 mg/kg/hr) for inhibition of arterial thrombosis (68%), platelet
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
d
aggregation (99%) and prolongation of MBT (3.1 times control). Following IV administration, the
dose needed to achieve maximum thrombosis reduction was four times less for compound 3l than for
the lead 1. Following PO administration, the 3 mg/kg dose of compound 3l was associated with 50%
inhibition of platelet aggregation response to 10 μM ADP (Figure 7B, bleeding time was not measured
in this experiment). It is noteworthy that compound 3l is the first P2Y antagonist reported which shows
1
potent oral antithrombotic effects in an animal model, providing preclinical proof-of-concept for an
1
orally bioavailable P2Y antagonist.
Compound 3l also inhibited venous thrombosis at the same IV doses that were efficacious in arterial
thrombosis (Figure 8A; blood flow was not measured in these experiments). This demonstrates a broad
1
antithrombotic impact for the P2Y mechanism. The IV doses which showed antithrombotic efficacy
also prolonged MBT, but these increases were limited to less than three-fold (Figure 8B).
In these same rat models, a 20 mg/kg oral dose of clopidogrel reduced carotid artery thrombus weight
2
9a
by 67% and prolonged mesenteric bleeding time to 8.2-fold. These results suggest that inhibition of
P2Y by 3l may compare favorably to the antithrombotic and haemostatic effects of a clinically
established P2Y12 receptor antagonist.
1
ACS Paragon Plus Environment