Discovery of Potent Protease-Activated Receptor 4 Antagonists with in Vivo Antithrombotic Efficacy

Article abstract of DOI:10.1021/acs.jmedchem.9b00186

In an effort to identify novel antithrombotics, we have investigated protease-activated receptor 4 (PAR4) antagonism by developing and evaluating a tool compound, UDM-001651, in a monkey thrombosis model. Beginning with a high-throughput screening hit, we identified an imidazothiadiazole-based PAR4 antagonist chemotype. Detailed structure-activity relationship studies enabled optimization to a potent, selective, and orally bioavailable PAR4 antagonist, UDM-001651. UDM-001651 was evaluated in a monkey thrombosis model and shown to have robust antithrombotic efficacy and no prolongation of kidney bleeding time. This combination of excellent efficacy and safety margin strongly validates PAR4 antagonism as a promising antithrombotic mechanism.

Full text of DOI:10.1021/acs.jmedchem.9b00186

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC
Article
Discovery of Potent Protease-Activated Receptor
Antagonists with in vivo Antithrombotic Efficacy
4
Michael M. Miller, Jacques Banville, Todd Friends, Mark Gagnon, Jon J. Hangeland, Jean-Francois
Lavalee, Alain Martel, Harold O'Grady, Roger Remillard, Edward H. Ruediger, Francois Tremblay, Shana
Posy, Nick Allegretto, Victor R. Guarino, David Harden, Timothy W. Harper, Karen Hartl, Jonathan
Josephs, Sarah Malmstrom, Carol A Watson, Yanou Yang, Ge Zhang, Pancras C Wong, Jing Yang, Michel
Bouvier, Dietmar A Seiffert, Ruth R. Wexler, R. Michael Lawrence, E. Scott Priestley, and Anne Marinier
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00186 • Publication Date (Web): 27 Jun 2019
Downloaded from http://pubs.acs.org on June 28, 2019
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 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 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.
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 86
Journal of Medicinal Chemistry
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 86
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
Discovery of Potent Protease-Activated Receptor 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
Antagonists with in vivo Antithrombotic Efficacy
Michael M. Miller,1‡† Jacques Banville, Todd J. Friends, Mark Gagnon, Jon J.
2‡
1ǂ
2
1ǂ
2
2
1
Hangeland, Jean-François Lavallée, Alain Martel, Harold O’Grady, Roger
2
2
2
3
1
Rémillard, Edward Ruediger, François Tremblay, Shana L. Posy, Nick J. Allegretto,
1
4#
1
1
Victor R. Guarino, David G. Harden, Timothy W. Harper, Karen Hartl, Jonathan
1
1
1
1
1
Josephs, Sarah Malmstrom, Carol Watson, Yanou Yang, Ge Zhang, Pancras
1
1
2,5
1†
1ǂ
Wong, Jing Yang, Michel Bouvier, Dietmar A. Seiffert, Ruth R. Wexler, R.
1ǂ
1ǂ
2
Michael Lawrence, E. Scott Priestley, * Anne Marinier *
1Bristol-Myers Squibb Research & Development, 311 Pennington-Rocky Hill Road,
2
Pennington, NJ 08534, USA. Institute for Research in Immunology and Cancer,
Université de Montréal, P.O. Box 6128, Downtown Station, Montréal, Québec H3C 3J7,
3
Canada. Bristol-Myers Squibb Research & Development, 3551 Lawrenceville Road,
4
Princeton, NJ 08540. Bristol-Myers Squibb Research & Development, 5 Research
ACS Paragon Plus Environment
1

Page 3 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
Parkway, Wallingford, CT 06492 USA. Department of Biochemistry and Molecular
Medicine, Université de Montréal, Montréal, Québec H3C 3J7, Canada.
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
ABSTRACT. In an effort to identify novel antithrombotics, we have investigated PAR4
antagonism by developing and evaluating a tool compound, UDM-001651, in a monkey
thrombosis model. Beginning with a high-throughput screening hit, we identified an
imidazothiadiazole-based PAR4 antagonist chemotype. Detailed structure-activity
relationship (SAR) studies enabled optimization to a potent, selective, and orally
bioavailable PAR4 antagonist, UDM-001651. UDM-001651 was evaluated in a monkey
thrombosis model and shown to have robust antithrombotic efficacy and no prolongation
of kidney bleeding time. This combination of excellent efficacy and safety margin strongly
validates PAR4 antagonism as a promising antithrombotic mechanism.
Introduction
ACS Paragon Plus Environment
2

Journal of Medicinal Chemistry
Page 4 of 86
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
Atherothrombotic cardiovascular and cerebrovascular diseases are the leading cause
of morbidity and mortality in developed countries and claim more lives each year than
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
cancer in the United States. Atherothrombotic diseases, such as acute coronary
syndrome and stroke, originate with a gradual build-up of plaque in the arterial wall.
The eventual rupture of an atherosclerotic plaque exposes a prothrombotic milieu to the
blood and activates platelets at the site of the injury, leading to the formation of a
potentially occlusive thrombus and subsequent ischemia. Antiplatelet agents, such as
aspirin and clopidogrel, inhibit thrombus formation and significantly reduce ischemic
events; thus, they are cornerstone therapies for the medical management of
2
atherothrombotic diseases. Recently, more potent and efficacious platelet P2Y12
receptor antagonists, such as prasugrel and ticagrelor, have been studied extensively
and found to be associated with increased bleeding risks, limiting their utility to improve
clinical outcomes.3-7 Significant unmet medical need remains and novel antiplatelet
agents with better efficacy and reduced bleeding risk are an important research goal.
ACS Paragon Plus Environment
3

Page 5 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Thrombin, the central enzyme in the coagulation cascade, activates platelets with
subnanomolar potency. It binds to two G-protein coupled receptors on human
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
platelets,8, 9 the protease-activated receptor 1 (PAR1) and protease-activated receptor 4
10
(
PAR4), and cleaves at a specific site within the N-terminal sequence of each
receptor. The newly exposed, shortened N-termini serve as tethered ligands that
activate their respective receptors through intramolecular interactions. Synthetic
peptide analogs based on the thrombin cleaved N-terminal sequence can also activate
the corresponding receptors in the absence of proteolytic activation.11, 12 Targeting
PAR1 or PAR4 with an antagonist is expected to reduce platelet activation without
inhibiting the other coagulation functions of thrombin.
PAR1 has generated widespread interest as an antithrombotic target, as evidenced
by pre-clinical research activity and the development of two PAR1 antagonist clinical
candidates, vorapaxar (1) and atopaxar (2), Figure 1. Phase III clinical studies of
vorapaxar in combination with aspirin and clopidogrel demonstrated incremental
reduction of ischemic events at the price of a significant increase in intracranial
ACS Paragon Plus Environment
4

Journal of Medicinal Chemistry
Page 6 of 86
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
bleeding.13, 14 In 2014, the U.S. Food and Drug Administration approved use of
vorapaxar to reduce thrombotic cardiovascular events in patients with prior myocardial
infarction or peripheral arterial disease. The drug is contraindicated for patients with a
history of stroke, transient ischemic attack, or intracranial hemorrhage because of
bleeding risk.15 Atopaxar has completed phase II clinical trials in acute coronary
syndromes and coronary artery disease, and demonstrated reductions in ischemia and
numerical increases in major bleeding. Liver enzyme elevations and QT interval
prolongation were also noted in atopaxar treated patients.16, 17 Further development of
atopaxar has not been reported and is presumed to have been discontinued.
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
O
H
H
H
H
H
O
N
OEt
OEt
O
But
O
F
NH O
N
Me
EtO
EtO
OMe
N
N
N
1
2
N
3
Vorapaxar
Atopaxar
O
YD-3
F
Figure 1. Chemical structures of PAR1 and PAR4 antagonists
ACS Paragon Plus Environment
5

Page 7 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
It is hypothesized that improved separation between effects on thrombosis and
hemostasis may be achieved by selectively targeting the late stage of thrombus
formation and vessel occlusion.18, 19 PAR1 is proteolytically activated at low
concentrations of thrombin and induces a rapid and transient calcium signal, whereas
PAR4 is cleaved at relatively high concentrations of thrombin and induces a sustained
calcium signal with a slower onset.20-22 Based on these signaling differences, it appears
that PAR1 is important for triggering platelet activation, while PAR4 is required for
sustaining platelet activation. Thus, inhibiting PAR4, while maintaining thrombin
signaling through PAR1, could potentially inhibit occlusive thrombosis while preserving
hemostasis. Selective PAR4 antagonism may comprise a novel antiplatelet mechanism
with a lower risk of bleeding than existing antiplatelet therapies. We have recently
disclosed preclinical studies23, 24 that demonstrate an outstanding efficacy and bleeding
profile for BMS-986120, a PAR4 antagonist clinical candidate. The early stages of the
medicinal chemistry effort that led to BMS-986120 are described in this manuscript.
Further optimization culminating in BMS-986120 will be reported subsequently.
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
ACS Paragon Plus Environment
6

Journal of Medicinal Chemistry
Page 8 of 86
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
In contrast to the vigorous investigation of PAR1 as an antithrombotic target, there
were very few reported PAR4 antagonists at the outset of this work. Only a single
chemical class of indazoles, typified by the compound YD-3 (3, Figure 1) had been
reported.25-27 YD-3 was shown to possess modest activity as judged by the inhibition
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
26, 28
observed in a GYPGKF-induced platelet aggregation assay (IC 130 nM).
Recent
50
work on the YD-3 lead led to a nitroindole, ML354, with 70 nM IC in a GYPGKF-
50
induced platelet aggregation assay, low in vitro metabolic stability, and 70-fold
selectivity vs. PAR1.29,30 Further optimization of this indole chemotype has not yet led
to compounds with activity against platelet aggregation in whole blood or in vivo
antithrombotic efficacy.31,32
In this report, we describe the identification and optimization of PAR4 antagonists with
potent activity in platelet rich plasma from an imidazothiadiazole high-throughput
screening lead. We also demonstrate in vivo antithrombotic efficacy and low bleeding
liability in a cynomolgous (cyno) monkey electrolytic carotid artery thrombosis (ECAT)
ACS Paragon Plus Environment
7

Page 9 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
model, confirming the therapeutic potential of PAR4 antagonism as a novel antiplatelet
approach.
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
Results and Discussion
Like the rest of the PAR family, PAR4 couples to the G protein Gq, which upon
receptor activation induces intracellular signaling events resulting in an increase in
cytosolic calcium.10 To screen for molecules possessing PAR4 antagonist activity, a
calcium mobilization assay (PAR4 fluorescent imaging plate reader (FLIPR)) was
33
developed using a HEK293-PAR4 cell line, an aequorin reporter, and a PAR4 agonist
peptide AYPGKF.11 YD-3 showed inhibitory activity in this assay (PAR4 FLIPR IC 410
50
nM). From a screening campaign of the BMS compound collection, 199 compounds
were identified which inhibited PAR4 agonist peptide-induced calcium signaling and did
not elicit agonist activity when tested in the absence of an added PAR4 agonist. The
PAR4-dependent antiplatelet effects of these compounds were then evaluated in a -
thrombin-induced washed platelet aggregation assay (-Thr WP). -Thrombin is a
proteolytic fragment of thrombin which specifically activates PAR4.34 The imidazo[2,1-
ACS Paragon Plus Environment
8

Journal of Medicinal Chemistry
Page 10 of 86
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][1,3,4]-thiadiazole(IDT)-based compound 4 (Figure 2) was one of the most potent hits
identified by the high throughput screen (HTS) (PAR4 FLIPR IC 17 nM, -Thr WP IC
50
50
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
500 nM).
S
N
N
MeS
N
O
4
Figure 2. Chemical structure of the PAR4 antagonist HTS hit
Lead profiling of 4 demonstrated poor in vitro liver microsomal stability across species,
and no significant activity in CYP, hERG, or PXR assays. Biotransformation studies
indicated that the methylthio group was oxidized to the sulfoxide and sulfone. Both
metabolites, but not the parent methyl thioether, react rapidly with glutathione and N-
acetylcysteine. In addition, oxidative metabolites of the benzofuran ring were detected.
Early SAR exploration was directed at improving potency in the PAR4 FLIPR and
platelet aggregation assays and improving metabolic stability.
Initial SAR optimization of the IDT scaffold was conducted using the synthetic
approach shown in Scheme 1. IDT 4 was obtained by condensation of 5-(methylthio)-
ACS Paragon Plus Environment
9

Page 11 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
,3,4-thiadiazol-2-amine 5 with bromoketone 6. Activation of the thiomethyl group with
mCPBA gave sulfone 7, which was converted to a series of methylthio replacements
upon treatment with various nucleophiles (thiols, amines, alcohols) in the presence of a
base. The regiochemistry of the cycloaddition was verified by an x-ray crystal structure
of compound 4 (supplementary material).
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
Scheme 1. General synthetic route for imidazothiadiazoles with 2-position modifications
(c)
XSH
S
O
S
MeS
NH2
Me
(d)
(e)
3
CH NH2
R
S
MeS
S
S
N
N N
N
O
(
a)
N
(b)
N
XOH
N
N
N
N
5
N
Br
40%
40-60%
7
6-80%
O
4 O
7
O
8-13
O
O
6
o
Reagents and conditions: (a) EtOH, wave: 150 C, 5 min; (b) m-CPBA, THF; (c) XSH,
o
o
MeOH, 70 C; (d) CH NH , DMF, wave: 70 C, 10 min (e) NaOMe, MeOH, rt or XOH,
3
2
K CO , DMF, rt.
2
3
Table 1 summarizes SAR for the 2-position of the IDT scaffold generated by this
synthetic approach. Incorporation of increased polarity as in sulfone 7 or steric bulk as
in the ethylthio and phenylthio analogs 8 and 9, respectively, decreased potency by 74-
to 260-fold, relative to the parent methylthio compound 4. Importantly, the methoxy
ACS Paragon Plus Environment
10

Journal of Medicinal Chemistry
Page 12 of 86
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
analog 10 retained the potent activity of 4 in both the PAR4 FLIPR and -Thr WP
assays, and possessed improved human liver microsome metabolic stability based on
percent of parent remaining at 10 min. Follow up in a five timepoint kinetic assay,
however, showed no improvement in human liver microsome t (compound 4, t = 5.4
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
1/2
1/2
min vs compound 10, t = 6.9 min). As observed for the thioethers, larger ether
1/2
substituents had successively lower potency (10 vs. 11 and 12). The methylamine
analog 13 also had modest activity, especially in the platelet aggregation assay,
suggesting that amines are not favored.
Table 1. Imidazothiadiazole 2-position structure-activity relationships
S
N
N
R
N
O
PAR4
FLIPR
IC50
Liver microsome
stability

-Thr WP
Cp
d
R
IC50
(% remaining, H,
R)
nM)a
(
nM)a
(
4
7
8
9
MeS
MeSO2
EtS
11
270
25, 4
ND
820
>30000
ND
2900
1900
45, 35
ND
PhS
ND
ACS Paragon Plus Environment
11

Page 13 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
0
1
MeO
EtO
4.0
67
240
80, 31
1
2100
57, 37
ND
12
i-PrO
MeNH
Me
760
110
37
>30000
>30000
>30000
2400
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
13
65, 2
20, 4
30, 3
62, 53
14
15
Et
38
16
c-Pr
530
>30000
a
Reported IC50 values are typically the mean of at least 2
experiments. ND = not determined
Results from the exploration of carbon-based replacements for sulfur are also shown
in Table 1.23 Methyl analog 14 shows at least 3-fold reduced potency relative to 4 and
1
0. Increasing the size of the alkyl group to ethyl (compound 15) maintained potency,
with continued poor metabolic stability, while the cyclopropyl analog 16 lost 14-fold
potency. A detailed biotransformation analysis on 15 identified the ethyl group as the
principal site of metabolism. While metabolic stability can be improved with substitution
on the ethyl moiety, both the putative hydroxy metabolites and bulkier alkyl alternatives
were found to possess weak PAR4 inhibitory activity.
ACS Paragon Plus Environment
12

Journal of Medicinal Chemistry
Page 14 of 86
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
In order to improve potency by gaining additional binding interactions and to modulate
the benzofuran metabolism, we next explored substitution on the benzofuran ring.
Analogs with benzofuran substitution were synthesized according to Scheme 2 and are
shown in Table 2. The substituted benzofurans were obtained by condensing
appropriately substituted 2-hydroxy-benzaldehydes 17 with chloroacetone, followed by
bromination of the resulting 2-acetylbenzofurans 18. Reaction of 2-amino-5-bromo-
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
1,3,4-thiadiazole with the resulting -bromomethylketones 19 provided the bromo-IDT
derivatives 20. Displacement of the bromide with sodium methoxide gave the methoxy-
IDT derivatives 21-29.
To prepare benzofuran analogs bearing alkoxy substituents at the 4- and 6-positions,
acetonide 30 was regioselectively derivatized at the p-phenol with the desired alcohol
using a Mitsunobu reaction as described by Kamisuki, et al.35 Subsequent alkylation at
the remaining phenol with an appropriate bromide provided the bis-alkoxy intermediate
32, which was reduced with DIBAL and hydrolyzed to afford the desired 2-hydroxy-4,6-
dialkoxy-benzaldehydes 33. These benzaldehydes were then reacted as shown for
ACS Paragon Plus Environment
13

Page 15 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
intermediate 17 to give analogs 34-38. Alternatively, analogs 39-58 were prepared from
analog 34 after benzyl deprotection and alkylation or Mitsunobu reaction of the resulting
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
phenol 35. In the benzyl deprotection step, the use of BCl with pentamethylbenzene
3
as cation scavenger was crucial for high yields.36
Scheme 2. General synthetic route for the preparation of benzofuranyl
imidazothiadiazoles with benzofuran modifications.
O
CHO
OH
Cl
(c, d)
O
O
O
R
R
R
(a, b)
O
Br
17
18
19
N N
S
NH2
S
N
N
(g)
S
N
Br
Br
R
MeO
R
N
N
(e, f)
O
N
O
20
21-29
OH
O
O
OH
31
O
O
BnO
O
O
O
BnO
33
(h)
(i)
(j, k)
CHO
OH
O
O
HO
MeO
MeO
MeO
30
32
O
OH
O
Ar
S
N
N
(l)
S
N
N
(m or n)
S
N
MeO
MeO
MeO
N
N
N
O
80%
O
N
O
36-58
OMe
OMe
OMe
34
35
o
Reagents and conditions: (a) Cs CO , DMF, 22 C, 18 h; (b) p-TsOH, 0.1 eq, THF, rt, 1
2
3
o
o
o
h; (c) LiHMDS, TMSCl, THF, -78 C; (d) NBS, -25 to 0 C, 1.5 h; (e) IPA, 88 C, 19 h; (f)
o
1
50 C, 30 min, wave; (g) NaOMe, MeOH, DCM, rt, 1 h; (h) DIAD, Ph P, MeOH, THF,
3
o
rt; (i) BnBr, K CO , DMF (j) DIBAL, -78 C, CH Cl (k) MeOH, HCl; (l) BCl ,
2
3
2
2
3
ACS Paragon Plus Environment
14

Journal of Medicinal Chemistry
Page 16 of 86
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
o
pentamethylbenzene, -78 C, DCM; (m) ArCH Br, K CO , DMF, rt; (n) ArCH OH, Bu P,
2
2
3
2
3
ADDP, THF, rt
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
As the potency of our PAR4 antagonists improved, we began to measure antiplatelet
activity with a -thrombin-induced platelet-rich plasma aggregation assay (-Thr PRP),
enabling evaluation of antiplatelet effects in the presence of plasma proteins. As shown
in Table 2, addition of methoxy substituents at the 4-, 6-, and 7-positions of the
benzofuran ring maintains single digit nanomolar potency in the PAR4 FLIPR assay,
and importantly, improves inhibitory potency 4- to 30-fold in the -Thr PRP assay
(
compounds 21, 24, and 26). However, a methoxy group is not tolerated at the 5-
position (compound 23). Increasing the size of the substitution, as with benzyloxy
groups, was detrimental to potency but beneficial for improving microsomal stability
(compounds 22 and 25).
Exploration of several di-substitution patterns (compounds 27-29 and 34) showed that
4
,6-disubstitution provided the best antagonist activity. The 4,6-dimethoxy and 4-
benzyloxy-6-methoxy analogs 29 and 34 were noteworthy in that they possess low
ACS Paragon Plus Environment
15

Page 17 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
micromolar -Thr PRP activity with significantly improved metabolic stability relative to
unsubstituted 10.
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
Table 2. Benzofuran substituent structure-activity relationships
4
S
N
N
MeO
R
N
O
7
PAR4
Liver microsome
stability

-Thr WP
-Thr
Cp
d
FLIPR
IC50
R
IC50
PRP IC50
(% remaining, H,
R)
(nM)a
(
nM)a
nM)a
(
1
2
2
0
1
2
H
4.0
240
350
15000
1300
80, 31
40, 42
85, 75
30, 38
57, 55
94, 68
25, 11
4-OMe
4-OBn
5-OMe
6-OMe
6-OBn
7-OMe
7.6
24
3100
4100
23
220
1.2
26
>30000
120
>30000
480
24
25
26
27
1800
240
>30000
3500
2.8
6-OMe,7-
OMe
1800
>30000
2100
>30000
10000
20, 3
28
4-OMe,7-
OMe
11
55, 28
ACS Paragon Plus Environment
16

Journal of Medicinal Chemistry
Page 18 of 86
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
9
4-OMe,6-
OMe
0
.99
74
530
86, 70
34
4-OBn, 6-
OMe
4.2
1000
1600
100, 77
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
Reported IC values are typically the mean of at least 2 experiments.
50
In light of their combination of anti-platelet potency and microsomal stability, the
pharmacokinetic profiles of compounds 29 and 34 were assessed in rat as summarized
in Table 3. The 4,6-dimethoxy analog 29 demonstrated high clearance and poor oral
bioavailability. Compound 34, bearing a bulky benzyloxy substituent at the benzofuran
4
-position, showed a promising pharmacokinetic profile with low clearance, long half-
life, and appreciable bioavailability.
a
Table 3. Metabolic stability and pharmacokinetic data for compounds 29 and 34
OMe
O
S
N
N
S
N
N
MeO
MeO
N
N
O
O
OMe
OMe
29
34
Compound
liver microsome t , min (H, R)
29
34
17, 22
33
>120, 40
3.7
1/2
Rat CL, mL/min/kg
t , h
3.2
12
1/2
ACS Paragon Plus Environment
17

Page 19 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
F, %
1
27
a
iv dose 0.5 mpk, iv vehicle 25/75 propylene glycol/PEG400 at 2 ml/kg, 10 min infusion; po dose 2 mpk, po
vehicle 5/5/90 EtOH/TPGS/PEG300 at 5 mL/kg
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
With a promising combination of -Thr PRP potency and favorable pharmacokinetic
properties established for compound 34, a series of extended aryl analogs were
prepared in an attempt to further drive potency (Table 4, compounds 36-38). Although
the -Thr PRP potency improved up to 3-fold, metabolic stability dropped significantly in
the extended aryl analogs, focusing further investigation on compounds that retained
the 4-benzyloxy group of compound 34.
Table 4. Structure activity relationships at the benzofuran 4-position
O
R
S
N
N
MeO
N
O
OMe
PAR4
FLIPR
IC50
-Thr
PRP
IC50
Liver microsome
stability
Cp
d
R
(% remaining, H,
R)
(
nM)a
(nM)a
3
4
6
-Ph
4.2
1600
100, 77
89, 43
3
-CH Ph
1.1
780
2
ACS Paragon Plus Environment
18

Journal of Medicinal Chemistry
Page 20 of 86
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
7
-CH OPh
0.81
.8
480
500
65, 36
75, 25
2
38
-
1
CH OCH Ph
2
2
a
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
Reported IC values are typically the mean of at least 2 experiments.
50
With the objective of improving potency and maintaining pharmacokinetic properties,
we investigated substitution on the benzyloxy moiety of compound 34. Table 5
summarizes the biological activity of analogs bearing small substituents, which were
synthesized according to the second synthetic sequence described in Scheme 2, from
key intermediate 35. Addition of substituents at the ortho position of the 4-benzyloxy
substituent in compound 34 did not improve -Thr PRP potency (compounds 39-42),
despite a slight gain in PAR4 FLIPR activity for the methyl and chloro analogs 39 and
4
0. Similar results were observed with substitution at the para position (compounds 49-
51). In contrast, introduction of substituents at the meta position generated more
promising results (compounds 43-48). Two analogs bearing lipophilic substituents
displayed submicromolar potency in the -Thr PRP assay along with good metabolic
stability (compounds 43 and 48), while more polar analogs bearing a hydroxyl group (45
ACS Paragon Plus Environment
19

Page 21 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and 46) were less active. The cyano analog 47 was slightly more potent than the
unsubstituted analog 34 in the same assay, while the methoxy analog 44 lost potency.
It is noteworthy that in most of these mono-substituted analogs, stability in human and
rat liver microsomes was similar to parent compound 34.
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
Table 5. Structure-activity relationships for 4-benzyloxy analogs
O
R
S
N
N
MeO
N
O
OMe
PAR4
Liver microsome
stability
Cp
d
-Thr PRP
R
H
FLIPR
a
IC (nM)
5
0
a
IC (nM)
(% remaining, H, R)
50
34
39
4.2
1.6
2.3
19
1600
3600
100, 77
90, 46
95, 78
82, 28
ND
2-Me
2-Cl
40
41
42
> 3000
>1000
>1000
680
2-OMe
2-CF3
3-Cl
3.4
3.3
9.4
3.1
27
43
82, 79
69, 34
97, 49
ND
44
45
46
3-OMe
3-OH
> 1000
2700
3-CH OH
> 1000
820
2
47
3-CN
0.98
3.1
100, ND
93, 79
48
3-CF3
440
ACS Paragon Plus Environment
20

Journal of Medicinal Chemistry
Page 22 of 86
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
4
9
4-CF3
4-Cl
4.3
11
1800
79, 90
ND
50
> 1000
> 3000
51
4-OMe
1.6
73, 64
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
Reported IC50 values are typically the mean of at least 2 experiments. ND = not
determined
In light of the enhanced -Thr PRP activity for certain analogs with meta substituents
on the 4-benzyloxy moiety, a more thorough investigation of structure-activity
relationships at this position was undertaken. The synthesis of these analogs followed
the second synthetic sequence described in Scheme 2. The results for key analogs are
summarized in Table 6. The addition of a phenyl ring attached with various linkers at the
meta position of the benzyloxy side chain was found to be beneficial. The PAR4 FLIPR
activity of these compounds remained in the low nanomolar range and the nature of the
linker was found to affect inhibition of platelet aggregation in plasma. Single atom
linkers, as in the aryl ether and thioether analogs 52 and 53, show 4-fold improved -Thr
PRP potency relative to 34. Most notably, however, extension to a benzyloxy group
(
compound 54, UDM-001651) enhanced the antiplatelet potency by 64-fold as
ACS Paragon Plus Environment
21

Page 23 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compared to the unsubstituted analog 34, with a -Thr PRP IC of 25 nM. Alternative
50
two atom linkers, such as the reverse ether 55 or ethylene 56 significantly impaired the
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
-Thr PRP activity relative to UDM-001651, indicating that both linker conformation and
polarity is important for antiplatelet activity. Lastly, the ortho and para benzyloxy
regioisomers 57 and 58 showed similar potency to parent compound 34.
Table 6. Structure-activity relationships for 4-benzyloxy analogs with aryl substituents
O
R
S
N
N
MeO
N
O
OMe
PAR4
FLIPR IC50
nM)a
Liver microsome
stability

-Thr PRP
Cpd
R
a
IC (nM)
50
(
(% remaining, H, R)
3
4
H
4.2
1.2
0.89
2.4
3.2
2.4
3.2
1.3
1600
390
410
25
100, 77
86, 42
80, 45
85, 76
86, 64
99, 68
94, 22
88, 84
52
3-OPh
3-SPh
53
54
55
3-OCH Ph
2
3-CH OPh
600
1200
2500
550
2
56
3-(CH ) Ph
2
2
5
7
8
2-OCH Ph
2
5
4-OCH Ph
2
a
Reported IC values are typically the mean of at least 2 experiments.
50
ACS Paragon Plus Environment
22

Journal of Medicinal Chemistry
Page 24 of 86
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
Given its excellent inhibitory potency in platelet-rich plasma, UDM-001651 was further
characterized both in vitro and in vivo. Direct binding to PAR4 expressed on HEK293
cell membranes was studied using a previously described binding assay (Table 7).24
The data demonstrate that UDM-001651 has single digit nanomolar affinity for PAR4
using both saturation binding and kinetic assays, with rapid and reversible binding
kinetics.
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
Table 7. Binding parameters for UDM-001651 to PAR4 a
Saturation binding K (nM): 1.4 ± 0.4
d
Association
kinetics
K (nM): 3.0 ± 0.3
d
-1
-1
K (nM min ): 0.004 ±
on
0.001
-1
K (min ): 0.013 ± 0.001
off
a
Reported values are means ± standard deviation, n = 3.
In vitro profiling data are summarized in Table 8. UDM-001651 was found to have
excellent in vitro stability in human, rat, mouse, dog, and monkey liver microsomes and
did not have any CYP, PXR, cardiac ion channel, or cytotoxicity issues. The compound
ACS Paragon Plus Environment
23

Page 25 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
showed very high protein binding across species. The selectivity of UDM-001651 was
tested using a variety of platelet agonists in a human platelet aggregation assay (Table
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
9). UDM-001651 showed potent inhibition of platelet aggregation triggered by a PAR4
agonist peptide, but no measurable inhibition of the effects of a PAR1 agonist peptide,
ADP, collagen, or the thromboxane A2 receptor agonist U46619 on platelets. Taken
together, these in vitro data indicate UDM-001651 has excellent properties as a tool
compound for mechanistic investigations of PAR4 inhibition. Pharmacokinetic data for
UDM-001651 are summarized in Table 10. The compound shows low clearance, 6-7 h
half-life, and modest bioavailability in rat, dog, and monkey. Importantly, the oral
exposure was sufficient to enable in vivo efficacy studies in our monkey ECAT model.37
Table 8. In vitro profiling assay data for UDM-001651
LM t , min (H, R, M, D, Mk)
86, 20, 120, 100, 48
1/2
HLM CYP IC , M
all > 40
5
0
PXR EC , M
> 50
5
0
cytotoxicity TC , M
hERG, Na patch clamp % inh. @ 10 M
> 50
≤ 10
50
Protein binding, % bound (H, R, M, D Mk)
99.6, 99.8, 99.5, ≥ 99.8, >
ACS Paragon Plus Environment
24

Journal of Medicinal Chemistry
Page 26 of 86
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
Table 9. Inhibition of human platelet aggregation by UDM-001651
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
platelet agonist
PAR4 agonist peptide
PAR1 agonist peptide
ADP
a
platelet aggregation IC (nM)
5
0
8
> 30,000
> 30,000
> 30,000
> 30,000
collagen
U46619
a
Reported IC values are typically the mean of at least 2 experiments.
50
Table 10. Pharmacokinetic data for UDM-001651
rat a
dog b
monkey b
species
CL, mL/min/kg
t , h
10
6.9
11
1.6
7.0
26
5.2
6.2
11
1
/2
F, %
a
iv dose 0.5 mpk, iv vehicle 25/75 propylene glycol/PEG400 at 2 ml/kg, 10 min infusion; po dose 2 mpk, po
vehicle 5/5/90 EtOH/TPGS/PEG300 at 5 mL/kg
b
iv dose 0.5 mpk, iv vehicle PEG400 at 0.8 ml/kg, 10 min infusion; po dose 2 mpk, po vehicle 40/60
TPGS/PEG400 at 0.5 mL/kg
In the monkey ECAT model, UDM-001651 was dosed orally at 0.4, 2, and 10 mg/kg (n
=
=
7 per dose) while the vehicle (40/60 TPGS/PEG400) was dosed orally at 2.5 ml/kg (n
8). As shown in Figure 3, UDM-001651 demonstrated dose-dependent antithrombotic
ACS Paragon Plus Environment
25

Page 27 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
efficacy, achieving maximal thrombus weight reduction of 38% at 2 and 10 mg/kg.
Furthermore, the compound did not increase monkey kidney bleeding time even at the
highest dose (Figure 4, P = 0.2, 10 mg/kg versus vehicle). Overall, UDM-001651
showed an effective antithrombotic profile with no prolongation of bleeding time,
demonstrating that PAR4 inhibition is an attractive approach to antithrombotic therapy.
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
10
*
P < 0.05 vs. vehicle
8
6
4
2
0


-9%
-38%
-38%
Vehicle
0.4
2
10
UDM-001651 (mg/kg PO)
Figure 3. Antithrombotic efficacy of UDM-001651 in the monkey ECAT model
ACS Paragon Plus Environment
26

Journal of Medicinal Chemistry
Page 28 of 86
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
000
800
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
600
400
200
0
Vehicle
0.4
2
10
UDM-001651 (mg/kg PO)
Figure 4. Profile of UDM-001651 in the monkey kidney bleeding time model
Conclusion
In an effort to advance antithrombotic therapy, we have explored PAR4 antagonism as
a novel antiplatelet mechanism, and developed a tool compound, UDM-001651,
suitable for in vivo target validation. We identified an IDT-based chemotype from an
HTS designed to detect PAR4 antagonism, with compound 4 as the initial hit. Detailed
SAR investigations around 4 led to the discovery of UDM-001651, a lead compound
with nanomolar antagonist potency against PAR4 in platelet rich plasma, excellent
ACS Paragon Plus Environment
27

Page 29 of 86
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
selectivity, and oral bioavailability. Evaluation of UDM-001651 in the monkey ECAT
model revealed robust antithrombotic efficacy and no prolongation of bleeding time,
demonstrating an excellent safety margin for PAR4 antagonism. These results validated
PAR4 as an antithrombotic target for our drug discovery effort, and strongly supported
further investigation of this mechanism towards the discovery of PAR4 antagonist
clinical candidates.
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
Experimental
HTS protocol. Screening of the BMS compound collection for PAR4 antagonists was
accomplished via aequorin-based detection of agonist induced calcium
mobilization.38 The synthetic peptide AYPGKF, corresponding to the activation
sequence derived from native N-terminus cleavage of the PAR4 receptor, served as the
agonist in the assay. Following confirmation of primary hits, potency of leads was
confirmed in a concentration response manner against the PAR4 receptor. Compounds
were further evaluated in calcium signaling assays to rule out PAR1 antagonist activity,
ACS Paragon Plus Environment
28