Discovery and clinical evaluation of 1-{N-[2-(amidinoaminooxy)ethyl]amino} carbonylmethyl-6-methyl-3-[2,2-difluoro-2-phenylethylamino]pyrazinone (RWJ-671818), a thrombin inhibitor with an oxyguanidine P1 motif

Article abstract of DOI:10.1021/jm901802n

We have identified RWJ -671818 (8) as a, novel, low molecular weight, orally active inhibitor of human a-thrombin (K_i = 1.3 nM) that is potentially useful for the acute and chronic treatment of venous and, arterial thrombosis. In a rat deep ven

Full text of DOI:10.1021/jm901802n

J. Med. Chem. 2010, 53, 1843–1856 1843
DOI: 10.1021/jm901802n
Discovery and Clinical Evaluation of 1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-
6-methyl-3-[2,2-difluoro-2-phenylethylamino]pyrazinone (RWJ-671818), a Thrombin Inhibitor
with an Oxyguanidine P1 Motif
Tianbao Lu, Thomas Markotan, Shelley K. Ballentine, Edward C. Giardino, John Spurlino, Kathryn Brown,
Bruce E. Maryanoff, Bruce E. Tomczuk, Bruce P. Damiano, Umesh Shukla, David End, Patricia Andrade-Gordon,
Roger F. Bone, and Mark R. Player*
Johnson & Johnson Pharmaceutical Research and Development, Welsh and McKean Roads, Spring House, Pennsylvania 19477-0776
Received December 4, 2009
We have identified RWJ-671818 (8) as a novel, low molecular weight, orally active inhibitor of human
R-thrombin (Ki = 1.3 nM) that is potentially useful for the acute and chronic treatment of venous and
arterial thrombosis. In a rat deep venous thrombosis model used to assess antithrombotic efficacy, oral
administration of 8 at 30 and 50 mg/kg reduced thrombus weight by 87 and 94%, respectively. In an
anesthetized rat antithrombotic model, where electrical stimulation of the carotid artery created a
thrombus, 8 prolonged occlusion time 2- and 3-fold at 0.1 and 1.0 mg/kg, iv, respectively, and more than
doubled activated clotting time and activated partial thromboplastin time at the higher dose. This
compound had excellent oral bioavailability of 100% in dogs with an estimated half-life of approxi-
mately 3 h. On the basis of its noteworthy preclinical data, 8 was advanced into human clinical trials and
successfully progressed through phase 1 studies.
Introduction
prevention of systemic thrombosis after heart valve replace-
ment or atrial fibrillation; and treatment and prevention of all
forms of VTE [deep venous thrombosis (DVT), pulmonary
embolism, and disseminated intravascular coagulation], sec-
ondary myocardial infarction, and restenosis.
Thromboembolic diseases remain a leading cause of
morbidity and mortality in the developed world. Venous
thromboembolism (VTE)a affects more than 2 million people
in the United States each year, progressing to pulmonary
embolism in approximately 600 000 of these patients and
becoming fatal in 200 000.1 VTE is the third leading cause
of cardiovascular-related death, after myocardial infarc-
tion and stroke.2 Nonvalvular atrial fibrillation is among
the most common dysrhythmia seen in clinical practice
and is responsible for up to 100 000 ischemic strokes
annually in the United States and more than a million
worldwide.3
The conditions amenable to treatment with anticoagulants
are broad, reflecting the importance of thrombosis in patho-
physiology. They includethe following: primary prevention of
acute myocardial infarction, stroke, and cerebral ischemia;
treatment of unstable angina; prevention of occlusion follow-
ing percutaneous transluminal coronary angioplasty and
coronary artery bypass grafting; prevention of reocclusion
after acute coronary intervention using intravascular stents;
Anticoagulants have great promise for the prophylaxis of
arterial and venous thrombotic disorders. Injectable agents
have been in use since unfractionated heparin became com-
mercially available in the 1940s. Low molecular weight hepar-
ins weredevelopedinthe early 1990s. The first directthrombin
inhibitor (DTI) was introduced in 1998, and the first synthetic
indirect factor Xa (fXa) inhibitor, fondaparinux, became
available in 2001. However, until very recently, oral agents
have been limited to vitamin K antagonists such as warfarin.
Vitamin K antagonists are highly efficacious but suffer from a
slow onset of action, genetic variations of metabolism, com-
plex food and drug interactions, and the costly requirement of
monitoring for dose adjustments. Thus, there is a significant
unmet need for new oral anticoagulants with rapid onset of
action and predictable pharmacokinetics and pharmacody-
namics.
The central role of the serine protease thrombin in throm-
bosis and hemostasis4 makes it an attractive target for antith-
rombotic therapy. As a key enzyme in the blood coagulation
pathway, thrombin acts on substrates such as fibrinogen,
factor V, factor VIII, factor XI, and factor XIII to induce or
accelerate the formation of thrombi.4 Thrombin very effi-
ciently initiates fibrin formation and activates factor XIII, a
transglutaminase that catalyzes the formation of cross-links
between fibrin monomers. In addition, thrombin effectively
promotes platelet activation via enzymatic cleavage of the
aminoterminusofprotease-activatedreceptors(PARs) on the
platelet surface (PAR-1, PAR-3, and PAR-4), unveiling a
*To whom correspondence should be addressed. Tel: 215-628-7860.
Fax: 215-540-4616. E-mail: mplayer@its.jnj.com.
aAbbreviations: ACT, activated clotting time; aPTT, activated partial
thromboplastin time; argatroban, (2R,4R)-1-[(2S)-5(diaminomethy-
lideneamino)-2-[[(3R)-3-methyl-1,2,3,4-tetrahydroquinolin-8-yl]sulfonyl-
amino]pentanoyl]-4-methyl-piperidine-2-carboxylic acid; BOP, benzo-
triazole-1-yl-oxy-tris-(dimethylamino)-phosphonium hexafluorophosphate;
CYP, cytochrome P450; DTI, direct thrombin inhibitor; DVT, deep venous
thrombosis; ECAT, electrical current-induced arterial thrombosis; fXa,
factor Xa; hERG, human ether-a-go-go related gene; HLM, human liver
microsome; PAR, protease-activated receptor; PT, prothrombin time;
RGD, arginine-glycine-aspartic acid; RLM, rat liver microsome; VTE,
venous thromboembolism.
r
2010 American Chemical Society
Published on Web 01/26/2010
pubs.acs.org/jmc

1844 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Lu et al.
tethered receptor ligand mechanism, which in turn activates
the receptor to elicit platelet activation. For example, mice
deficient in PAR-4 demonstrate reduced thrombus formation
at sites of arterial injury.5 In contrast, through its interaction
with thrombomodulin, thrombin activates protein C, initiat-
ing negative feedback of the coagulation cascade. Activated
protein C inactivates factors Va and VIIIa and regulates
fibrinolysis via activation of a fibrinolysis inhibitor.6 Anti-
thrombin is the physiological inhibitor of thrombin; however,
thrombin bound to a fibrin clot is inaccessible to the actions of
antithrombin. Thrombin, via activation of PARs, is also a
modulatorof endothelialand smoothmusclecell proliferation
and migration. It induces inflammatory responses, acting as a
chemotactic to peripheral blood monocytes and as a stimu-
lator of neutrophil adhesion.7 Thus, a potent inhibitor of
thrombin has potential as an effective antithrombotic drug in
both venous and arterial environments. Herein, we describe a
series of potent pyrazinone thrombin inhibitors, which culmi-
nated in the discovery and clinical evaluation of 8.
in refluxing toluene followed by saponification provided the
3-alkylamino-6-substituted pyrazinone acetic acids. The
coupling of the acid and oxyguanidine intermediate 4 was
accomplished with benzotriazole-1-yl-oxy-tris-(dimethyl-
amino)-phosphonium hexafluorophosphate (BOP). The
crude coupling product was purified by normal phase chro-
matography and then deprotected by treatment with tri-
fluoroacetic acid. Subsequent purification of the product by
normal phase chromatography followed by recrystallization
of the HCl salt from ethanol provided 7-18. To obtain the
3-alkylamino-6-chloropyrazinone acetic acids, coupling of
the amine took place with 3-bromopyrazinone acetic acid
ester 199 (Scheme 3) in refluxing toluene followed by chlori-
nation using N-chlorosuccinimide, and subsequent saponi-
fication afforded 20. Installation of the P1 oxyguanidine
then took place in an analogous manner as described
for 7-18. Methylation of the oxyguanidine on the proximal
nitrogen was accomplished using iodomethane to yield 22.
Ligand Design. Many early approaches to thrombin in-
hibitors relied upon the active site recognition motif D-Phe-
Pro-Arg, which mimics a portion of the natural substrate.10
Each inhibitor may be thought of as being composed of three
components: a basic P1 substituent, a central group that acts
as a scaffold and/or S2 binding fragment, and an aryl
substituent, P3. Modeling studies11 involving the docking
of a similar inhibitor Ac-Ala-Pro-Val-TFMK (trifluoro-
methylketone) into the active site of human neutrophil
elastase suggested that it might be possible to alter the
alanine R-carbon hybridization from sp3 to sp2 and construct
a hydrocarbon bridge from it to the adjacent proline. This
modification, combined with cleavage and deletion of two
carbons of the proline ring, results in a substituted pyridone
that retains many of the important hydrogen bond recogni-
tion elements of the tripeptides. However, many 3-amino-
2-pyridones show inherent chemical instability due to the
electron-rich nature of the pyridone ring.12 Stabilization of
the heterocycle has been addressed in three ways: insertion of
a strongly electron-withdrawing substituent such as CF3 at
the 4-position, derivatization of the 3-amino group as a
sulfonamide, or addition of a nitrogen to convert the pyr-
idone into a pyrazinone core.11 These strategies, particularly
the conversion to a pyrazinone core, are applicable to
thrombin inhibitors and have been used to design com-
pounds with a balance of thrombin affinity and physical
properties that permit desirable pharmacokinetics.
Results and Discussion
Synthetic Chemistry. Preparation of the P1 intermediate 4
(Scheme 1) started with the reaction of (2-hydroxy-ethyl)-
carbamic acid benzyl ester with N-hydroxyphthalimide to
afford 1, which was deprotected with methylamine to give
oxyamine 2. Guanylation of 2 provided the bis-BOC inter-
mediate 3, which was deprotected to afford key intermediate
oxyguanidine 4 in 49% overall yield beginning with the alcohol.
Reaction of 3-halo-6-substituted pyrazinone intermedi-
ates 516b and 68 (Scheme 2) with the appropriate alkylamine
Scheme 1a
a Reagents and conditions: (a) N-Hydroxyphthalimide, PPh3,
DEAD, room temperature, 10 h. (b) MeNH2, EtOH/THF, room
temperature, 1 h. (c) [N,N0-Di(tert-butoxycarbonyl)]amidinopyrazole,
room temperature, 10 h. (d) 10% Pd/C, H2(g), EtOH/THF, room
temperature, 30 min.
Scheme 2a
a Reagents and conditions: (a) PhMe, 60 °C, 3 h. (b) (i) Dioxane/1 N NaOH(aq), room temperature, 1.5 h; (ii) 10% HCl(aq). (c) Castro’s reagent,
TEA, DMF, room temperature, 18 h. (d) DCM, TFA, room temperature, 8 h.
Scheme 3a
a Reagents and conditions: (a) PhMe, reflux, 12 h and then room temperature, 5 days. (b) NCS, DCE, reflux, 2.5 h. (c) (i) dioxane/1 N NaOH(aq),
room temperature, 18 h; (ii) 10% HCl(aq). (d) Castro’s reagent, TEA, DMF, room temperature, 3 days. (e) DCM, TFA, room temperature, 24 h.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1845
Table 1. Effect of P1-P3 Substitution on Thrombin (thr) and Trypsin
(tryp) Inhibition (Ki) of 7-18, 21, and 22
Table 2. Rat and Human Microsomal Stability, Caco-2 Flux, and
Recombinant CYP3A4 Inhibition of 7-18, 21, and 22
t1/2 (min)
Caco-2 Papp
(AfB/BfA)
(ꢀ10-6 cm/s)
RLM
stability
HLM
stability
CYP3A4
inhibition (μM)
compd
7
>100
>100
16.7
13.8
14.4
44.7
32.9
>10
52.3
32.6
38.8
8.3
>100
>100
40.1
85.8
57.6
16.6
19.1
34.1
24.5
22.8
24.4
25.9
>100
59.2
0.5, 2.26
1.37, 6.45
0.55, 1.48
0.92, 1.84
0.5, 1.91
>10
>10
5.9
8
9
10
11
12
13
14
15
16
17
18
21
22
2.1
2.9
a
thr Ki tryp Ki
compd
R
Y
R1 (nM)
(nM)
0.74, 3.32
0.28, 3.18
0.87, 0.36
0.29, 10.3
0.67, 4.29
0.32, 11.81
0.14, 12.98
0.5, 3.04
3.0
3.5
7
8
2,2-difluoro-2-phenylethyl
2,2-difluoro-2-phenylethyl
2-(4-fluorophenyl)ethyl
CN
Me
Me
Me
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
2.3
1.3
105
128
6.7
3.4
9
13
1600
850
4.7
10
11
12
13
14
15
16
17
18
21
22
2-(3,4-difluorophenyl)ethyl
2-(2,4-difluorophenyl)ethyl
47
15
0.96
1.2
3400
2800
3000
3100
2200
2500
500
2-(4-trifluoromethylphenyl)ethyl Me
47
11
17.7
24.8
1.5
3.5
2-(4-methoxyphenyl)ethyl
2-(3,4-dimethoxyphenyl)ethyl
2-(4-ethylphenyl)ethyl
2-(5-indanyl)ethyl
Me
Me
Me
Me
Me
Me
Cl
0.27, 0.6
120
66
aromatic ring and thereby enhance the σ-π edge-to-face
interaction of the ring with the protein at electron-rich
Trp215,15 in this series, such a change was not beneficial,
as demonstrated by the 4-trifluoromethyl analogue 12 (Ki =
47 nM). Further comparing compounds with different induc-
tive effects at the 4-position, the trend in Ki values was OMe
(13) < F (9) < CF3 (12) < ethyl (15) (11, 13, 47, and 66 nM,
respectively). The potencies of these substitutions are likely due
to their presence only at the 4-position, allowing a slight
structural rearrangement of the binding pocket to accommo-
date the fit. Compound 14, the 3,4-dimethoxy analogue, is even
less potent (Ki = 120 nM) as it requires even more structural
adjustment at the P3 site.
Analogues with larger hydrophobic substituents, such
as 2,2-diphenylethyl (18, Ki = 6.8 nM), 2-(1-napthyl)ethyl
(17, Ki = 12 nM), and 4-methylphenyl (22, Ki = 37 nM)
maintained potency, as they may project out of the P3 pocket
into solvent. Compounds 18 and 17 concurrently displayed
62- and 4-fold selectivity over trypsin, respectively. 2-(5-
Indanyl)ethyl (16, Ki = 44 nM) was approximately as potent
as other 3,4-disubstituted phenyl analogues such as 10 and
14. Compound 22 also featured a methyl group at the
proximal nitrogen of the oxyguanidine (R to the oxygen),
which did not negatively impact potency or selectivity over
trypsin (68-fold) (vide infra).
44
12
2-(1-naphthyl)ethyl
2,2-diphenylethyl
6.8
1.7
37
420
370
2,2-difluoro-2-phenylethyl
2-(4-methylphenyl)ethyl
Me Me
2500
a Ki values are averaged from multiple determinations (n g 3), and the
standard deviations are <20% of the mean.
Arginine- or guanidine-based P1 groups provide mole-
cules with high affinity by virtue of a strong salt bridge
interaction with Asp189 in the S1 specificity pocket. The
strong basicity of such groups, however, often causes low
oral bioavailability due to poor intestinal absorption. In
addition, some inhibitors with aminocyclohexane P1s dis-
play in vivo toxicity, the severity of which is a direct function
of the pKa of the aminocyclohexane.13 To address these
problems, we have previously reported on a series of throm-
bin inhibitors exhibiting a novel oxyguanidine moiety, which
serves as an arginine surrogate of the arginine-glycine-aspar-
tic acid (RGD) type thrombin inhibitors.14 The oxyguani-
dine possesses a greatly reduced pKa of 7.0-7.5 versus 13-14
for guanidine, and this distinguishing characteristic is re-
flected in favorable in vitro permeability and pharmaco-
kinetics. The current series combines the features of the
pyrazinone core with the oxyguanidine moiety.
Structure-Activity Study. We conducted a survey of a
variety of lipophilic P3 substituents on the pyrazinone core.
The effects of methyl, chloro, and cyano substitution on the
6-position of the pyrazinone ring were also addressed. A
range of compounds (7-18, 21, and 22) were assessed for the
inhibition of human R-thrombin and the key enzyme trypsin
(Table 1). Insights emerged regarding the effect of P3 group
substitution on thrombin binding and on selectivity with
respect to trypsin. 2-(4-Fluorophenyl)ethyl compound 9 (Ki =
13 nM) was essentially equipotent to 11 (Ki = 15 nM), a 2,4-
difluoro-substituted analogue; however, 3,4-difluoro-substitu-
tion, as in 10, decreased the potency 3-fold (Ki = 47 nM). The
substitution pattern more significantly affected the selectivity
of this set of compounds against trypsin, however: 10, 9, and 11
were 18-, 120-, and 230-fold more potent for thrombin than for
trypsin. The same preference for monosubstitution as com-
pared to 3,4-substitution was seen in methoxy derivatives 13
and 14, which exhibited Ki values of 11 and 120 nM, res-
pectively. Although it has been hypothesized that electron-
withdrawing substituents on P3 decrease electron density in the
The addition of a gem-difluoromethyl group on the
benzylic position of the P3 moiety not only conferred greater
thrombin affinity but also higher metabolic stability, as that
position was shown to be one important site of oxidative
metabolism.16 This phenomenon was exemplified by 7, 8,
and 21. They were among the most potent compounds
studied, with Ki values of 2.3, 1.3, and 1.7 nM, respectively.
In human liver microsomal (HLM) preparations, the half-
lives of each of these three compounds exceeded 100 min
(Table 2). Without this modification to attenuate the meta-
bolic liability, half-lives shortened considerably, ranging
from 16.6 min for 12 to 85.8 min for 10. Within that range,
most HLM t1/2 values were between 20 and 30 min.
Substitution in the 6-position of the pyrazinone ring
improved potency via interaction with the Tyr-Pro-Pro-
Trp insertion loop near the S2 pocket of thrombin. In the
past, substituents such as methyl appeared to be metaboli-
cally labile,17 although given the high rat liver microsome
(RLM) and HLM half-lives, particularly of 8, that does not
appear to be the case throughout this series. Compound 21,

1846 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Lu et al.
which contains a 6-Cl substitution on the pyrazinone, was
1.7 nM against thrombin and 370 nM against trypsin. 6-Cyano
and 6-methyl groups (7 and 8), despite decreasing trypsin
selectivity (Ki = 105 and 128 nM, respectively), maintained
high affinity for thrombin (Ki = 2.3 and 1.3 nM, respectively).
The electron-withdrawing Cl and CN and the hydrophobic
methyl were all well-tolerated within the relatively hydropho-
bic section of protein in the S2 insertion loop.
The crystal structure of 8 reveals the archetypical binding
mode for the series of pyrazinone compounds. Initial differ-
ence electron density maps, calculated from atomic coordi-
nates after positional and temperature factor refinement
without the inhibitor, clearly revealed the positions of the
inhibitor as well as additional solvent molecules that were
not included in the difference Fourier calculation. The
central pyrazinone scaffold makes hydrogen bonds with
the backbone NH and CO of Gly216 as a peptide mimetic
(Figure 1a,b; PDB ID: 3LDX). The methyl substitution
projects into the pocket formed by His57, Tyr60A, and
Trp60D, mostly filling this pocket, and there is space for
the slightly larger substitutions of Cl or CN with minimal
rearrangement required. The amide NH also contributes
a hydrogen bond to the backbone of the protein (Ser214)
to complete the scaffold interactions.
The oxyguanidine binds in an asymmetric mode to
Asp189. Additional hydrogen bonds are made with the
backbone carbonyl of Gly219 and two water molecules to
fully coordinate the oxyguanidine moiety. Water 49 is fully
coordinated making hydrogen bonds with oxygen and one of
the terminal guanidine nitrogens as well as with the back-
bone carbonyl of Gly216. Water 3 is associated with the
proximal nitrogen and the other terminal nitrogen of the
guanidine. The use of water molecules to bridge hydrogen-
bonding interactions with the hydrogen atom of the prox-
imal nitrogen allows flexibility in this region of the molecule.
An alternate symmetrical binding mode of an orcinol-based
inhibitor with the hydrogen of the proximal nitrogen flipped
was seen in a previously reported oxyguanidine structure
(Figure 1c; PDB ID: 1T4U).18 This flexibility, along with the
use of water-bridged hydrogen bonding, explains the ability
to substitute a methyl group for the hydrogen atom on the
proximal nitrogen (22) without a significant loss of affinity.
Selectivity for thrombin versus trypsin is important be-
cause of the central location of trypsin in the gut and its likely
role in protease-mediated feedback control of endogenous
pancreatic enzyme secretion.19 Recognizing that the S1
pockets of thrombin and trypsin are identical with the
exception of residue 190, which is alanine and serine, respec-
tively, one possible approach would take advantage of the
slightly larger size of the pocket in thrombin by decreasing
the number of hydrophilic groups and increasing the hydro-
phobic surface area until the limit of the pocket is reached.20
Sanderson et al., utilizing this strategy, have disclosed affi-
nity and trypsin selectivity data for related series of pyrazi-
none thrombin inhibitors.21 A lead compound, featuring an
2-amino-6-methyl pyridine at P1, had a Ki of 0.8 nM for
thrombin and 1800 nM for trypsin. Other P1 groups have
also been employed as follows: imidazopyridazines, 5-linked
indoles, and azaindoles21 (the affinity of the three isomers
correlates with the acidity of the conjugate acid), azoles,22
imidazole acetamides,23 and aminopyridines.11
Figure 1. (a) Hydrogen-bonding interactions of 8 with thrombin
(PDB ID: 3LDX). The atoms are colored by atom type. Primary
interactions are shown as dotted lines, and secondary shell interac-
tions are shown as thin solid lines. (b) Accessible surface representa-
tion of thrombin is shown with the surface colored by atom type
with 8 shown as a stick representation. (c) Accessible surface
representation of thrombin with the inhibitor (PDB ID: 1T4U)
shown as a stick figure.
and hydroxylamines; (2) 2-aminopyrazines; and (3) 2-amino-
pyrimidines and 2-aminotetrahydropyrimidines.24 Those
containing benzylamine and thiophene were often single
digit nanomolar in potency for thrombin but triple digit
nanomolar for trypsin; however, those containing amino-
pyrimidines or aminopyrazines were considered inactive for
trypsin, thus more selective.
Advanced in Vitro Evaluation. Our exploration of the space
occupied by P3 led to 7, 8, and 21, which were 42-, 32-, and
217-fold selective versus trypsin, respectively (Table 1).
Compound 8 did have slight inhibitory potential for tissue
type plasminogen activator (tPA, Ki = 2.4 μM) and fXa
In addition, Reiner et al. have reported pyrazinone throm-
bin inhibitors embodying three classes of monocyclic P1
arginine surrogates: (1) (hetero)aromatic amidines, amines,

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1847
Table 3. Human Plasma Protein Binding (hPPB), in Vitro Anticoagulant Potency (2ꢀ aPTT), and Rat and Dog Pharmacokinetic Parameters of 7, 8,
and 21
rat
dog
t1/2 (h)b
compd
hPPB (%)
2ꢀ aPTT (nM)
Cmax (μM)a
t1/2 (h)b
F (%)
Cmax (μM)a
F (%)
7
8
49
67
88
342
510
565
0.2
3.3c
0.2
1.6
3.3
3.7
11
23
9
3.3
5.7
1.6
3.9
2.6
2.1
57
103
60
21
a Dose = 10 mg/kg unless otherwise noted in fasted animals. b po half-life. c Dose = 30 mg/kg.
(Ki = 3.3 μM); however, the compounds were inactive
(>26 μM) against many other important enzymes, that is,
factor VIIa, urokinase, chymotrypsin, cathepsin G, human
leukocyte elastase, kallikrein, chymase, and plasmin. Of
these, 7 had slight inhibitory potential for fXa (Ki = 6.5 μM).
Further in vitro evaluation of 7, 8, and 21 was undertaken
to assess the metabolic stability, permeability, and recombi-
nant cytochrome P450 (CYP) inhibition of these three lead
compounds as compared to others in the series (Table 2).
Test compounds were equilibrated with RLMs and HLMs at
37 °C, NADPH was added to initiate the reaction, and
aliquots were withdrawn and quenched at 0, 15, 30, and 60
min. The percent of parent compound remaining in the
incubation mixture was plotted as a function of time. The
elimination half-lives of 7 and 8 associated with the disap-
pearance of test article were determined to be >100 min in
both species, suggesting that 7 and 8 are not susceptible to
rapid metabolism by this mechanism. The half-life for 21
was >100 min in HLMs but dropped to 17.7 min in RLMs.
Permeability was measured by movement of compound
through Caco-2 cell monolayers grown on collagen-coated,
microporous, polycarbonate membranes. Compound 7 ex-
hibited low absorption potential, having movement from
apical to basolateral (A to B) of 0.5 ꢀ 10-6 cm/s and from B
to A of 2.26 ꢀ 10-6 cm/s, while the absorption potential for 8
was considered high (A to B = 1.37 ꢀ 10-6 cm/s and B to
A = 6.45 ꢀ 10-6 cm/s). The absorption potential for 21 was
similar to that of 7 (A to B = 0.5 ꢀ 10-6 cm/s and B to A =
3.04 ꢀ 10-6 cm/s).
intrinsic and common pathways, and is measured as follows:
Blood samples are buffered by sodium citrate at a final
concentration of 0.105-0.109 M and treated with phospho-
lipids, surface activators, and calcium to shorten clotting
time and initiate the coagulation reactions. A coagulometer
is then used to measure the optical density of the mixture and
record the time required to reach the threshold density at
which a clot is said to have formed. Therapies prolonging
aPTT from 1.5 to 2.5 (generally 2-fold) times the midpoint of
the control range are associated with a reduction in the risk
of recurrent thrombosis. Measurements of aPTT can be
sensitive to the exact nature of the reagents used as well as
sample storage/preparation protocols. Nevertheless, both
prolonged and shortened aPTT values can have significant
clinical implications with respect to bleeding and hyperco-
agulation, respectively. In the present study, fresh whole
blood from humans, dogs, and rats was centrifuged to obtain
plasma, and 8 was added to the plasma in concentrations
from 0.001 to 100 μM. After a short stabilization period, the
samples were placed in an ACL 100 microsampler coagula-
tion analyzer to determine clotting times in seconds. In
human-, rat-, and dog-spiked plasma, 8 doubled aPTT
coagulation time at concentrations of 0.51 ( 0.08, 0.85 (
0.12, and 2.03 ( 0.48 μM, respectively. Compound 7 doubled
aPTT coagulation time at a concentration of 0.34 ( 0.02 μM
in human plasma.
To further define the antithrombotic efficacy of 7 and 8, an
electrical current-induced arterial thrombosis (ECAT) mod-
el in rats was used to evaluate the ability of the compounds to
prolong time to occlusion.25 Because arterial thrombosis in
humans usually occurs in areas of medium to high blood flow
and high sheer stress with a triggering factor of vascular
injury such as atherosclerotic plaque rupture, the ECAT
model may have strong pathophysiological relevance to
human disease. Wong et al.25c utilized scanning electron
microscopy to confirm vessel wall injury at the site of
electrical stimulation in this model: The de-endothelialized
layer of the vessel showed both platelet and fibrin deposition,
likely due to activation of the coagulation cascade by ex-
posed tissue factor. Intravenous administration of 7 at doses
of 0.03, 0.1, 0.3, and 1.0 mg/kg resulted in a prolongation of
time to occlusion from a control of 11 ( 1 to 15.3 ( 3.2,
19.3 ( 3.8, 24.9 ( 2.6, and 30 ( 0 min, respectively, while 8
(iv) resulted in a prolongation of time to occlusion to 18.4 (
7.1, 20.7 ( 3.7, 14.7 ( 5.0, and 30 ( 0 min, respectively
(Figure 2). aPTT was also prolonged in a dose-related
manner (Figure 3). For 8, the increases in aPTT from a
control of 21 ( 2 s were 23 ( 1, 31 ( 4, 35 ( 8, and 66 ( 42 s,
respectively. A doubling of aPTT occurred at a dose of
between 0.3 to 1.0 mg/kg, which was associated with plasma
concentrations of 8 between 0.12 and 0.31 μM. For 7, the
increases in aPTT were 28 ( 2, 30 ( 2, 37 ( 9, and 47 ( 4 s,
respectively. A doubling of aPTT also occurred at a dose of
between 0.3 and 1.0 mg/kg, which was associated with
The HLM CYP inhibition assay was performed using
cDNA-expressed human enzymes of the P450 isoforms most
commonly responsible for the metabolism of drugs and
associated drug-drug interactions. Of particular interest
was the CYP3A4 isoform, as a few compounds in the series,
including 21, demonstrated 3A4 inhibition e3 μM. The
values for 7 and 8, however, were both >10 μM, and there
was no significant inhibition of any other P450 enzyme
isoform tested.
Pharmacokinetics. Table 3 summarizes the results of the
plasma protein binding (PPB) and rat and dog pharmaco-
kinetic studies for 7, 8, and 21. PPB was low (49, 67, and
88%, respectively). Compound 7, when administered to dogs
at 10 mg/kg po, displayed a Cmax of 3.3 μM, a half-life of
3.9 h, and an oral bioavailability of 57%. At the same dose
level and route of administration of 8, the Cmax was 5.7 μM,
the half-life was 2.6 h, and the oral bioavailability was 100%.
In contrast, the values for 21 were 1.6 μM, 2.1 h, and 60%,
respectively.
Anticoagulant and Antithrombotic Studies. On the basis of
the results of the aforementioned in vitro and in vivo studies,
7 and 8 were chosen for continued assessment. Activated
partial thromboplastin time (aPTT) is a global clotting test
sensitive to plasma concentration variations in several clot-
ting factors, particularly factors I, V, VII, IX, and XII of the

1848 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Lu et al.
Figure 2. Time to carotid artery occlusion as a function of dose for
7 and 8 in the rat ECAT model.
Figure 4. Antithrombotic effect of 7 and 8 after an oral dose of 3
mg/kg in the canine arteriovenous shunt model.
40 ( 13 to 73 ( 2% (Figure 4) during the 3 h testing period
starting at 2 h postoral dosing. There were no significant
changes in bleeding times as compared to vehicle experi-
ments. Compound 8 administered orally at 3 mg/kg had no
overt side effects in three out of three dogs during the 1 h
period before they were anesthetized. aPTT was increased
1.6-fold for 8 at 1 h postoral dose as compared to 1.4-fold for
7. Plasma concentrations at the 1 h time point were 0.87 (
0.3 μM for 8 as compared to 0.48 ( 0.16 μM for 7.
To further define the intravenous and oral antithrombotic
efficacy of 8, a rat deep vein thrombosis model was used to
evaluate the compound’s ability to inhibit venous thrombus
formation (Table 4).28 All experimental animal models of
venous thrombosis include a trigger of the thrombogenic
state, which is most often composed of two components of
Virchow’s triad: hypercoaguability, reduced blood flow, and
vessel wall injury. Hypercoaguability is induced by treatment
with an agent such as tissue factor or thromboplastin. Blood
flow reduction can be achieved by complete stasis, in which
case the concentration of blood components and antithrom-
botic agent is constant within the occluded segment. Re-
duced blood flow may also result from a partial stasis, which
allows flow of test compound and blood components around
the formed thrombus. Vessel wall injury plays a major role in
the partial stasis paradigm or can be generated separately by
means of ferric chloride or saline. Intravenous administra-
tion of 8 resulted in a 66 and 98% inhibition in thrombus
weight at the 0.3 and 1 mg/kg dose, respectively (Table 4) as
well as dose-related changes in coagulation times. Oral
administration of 30 and 50 mg/kg resulted in 87 and 94%
inhibition of thrombus weight as well as dose-related
changes in coagulation.
Figure 3. aPTT as a function of dose for 7and 8in the rat ECAT model.
plasma concentrations of 7 between 0.085 and 0.38 μM.
In this model, 7 and 8 were 10-fold more potent than
the reference DTI (2R,4R)-1-[(2S)-5(diaminomethylidene-
amino)-2-[[(3R)-3-methyl-1,2,3,4-tetrahydroquinolin-8-yl]-
sulfonylamino]pentanoyl]-4-methyl-piperidine-2-carboxylic
acid (argatroban) and equipotent to one another by dose.
An arteriovenous shunt procedure was employed as a
suitable model to test the antithrombotic efficacy of the
two lead thrombin inhibitors against a thrombus of mixed
platelet/fibrin/red blood cell composition. In this model, a
thrombus forms on a section of silk thread placed in an
extracorporeal shunt between the femoral artery and the
vein.26 Imaging studies27 using a 99Tc-labeled platelet glyco-
protein IIb/IIIa receptor antagonist have shown that the
deposition on the thrombogenic surface mimics a high-shear,
platelet-rich thrombus indicative of arterial conditions, and
the platelet-poor tail mimics that of a venous thrombus
stemming from a platelet-poor, low-shear environment.
Studies consisting of seven shunt periods in dogs starting
at 2 h postdrug and repeating every 30 min were performed.
The thrombus weight was measured after each shunt period,
and bleeding times and coagulation times were measured at
2, 3, 4, and 5 h and compared to control values.
Cardiovascular Safety. As part of an assessment of cardi-
ovascular safety, 8 was evaluated in vitro for its effects on the
human ether-a-go-go related gene (hERG) potassium chan-
nel. In the direct hERG channel binding assay, 8 demon-
3
strated 50% inhibition of H-astemizole binding at 10 μM.
However, in an assay measuring percent inhibition of the
Ikr-like current in hERG-transfected human embryonic
kidney cells, 8 had only a minimal effect, showing a mean
value of 9 ( 2%, which suggests that it has limited functional
activity and should not cause QT interval prolongation.
Table 5 summarizes the effects of 7 and 8 on hemo-
dynamics and electrocardiographic intervals in anesthetized
Thrombus formation (thrombus weight) by 7 was inhib-
ited 38 ( 9 to 45 ( 12%, while for 8 inhibition ranged from

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1849
Table 4. Effect of 8 on Thrombus Weight, aPTT, and Plasma Concentration in the Rat DVT Model
dose/route
(mg/kg)
mean thrombus weight
(mg/100 g body weight ( SEM)
decrease in thrombus
weight vs vehicle (%)
compd
vehicle
argatroban
aPTT
plasma concentration (μM)
21.8 ( 0.4 (4)a
1.7 ( 0.3 (5)
24.7 ( 3.0 (3)
7.5 ( 2.4 (5)
0.5 ( 0.5 (3)
23.0 ( 1.8 (5)
2.9 ( 0.6 (6)
1.3 ( 0.2 (3)
NDb
56 ( 7
35
1.0, iv
0.1, iv
0.3, iv
92
0
ND
0.31 ( 0.21
0.54 ( 0.07
0.80 ( 0.06
0.10 ( 0.05
0.76 ( 0.20
0.73 ( 0.10
8
66
98
0
46 ( 8
49 ( 8
33 ( 3
41 ( 6
39 ( 6
1.0, iv
8
10.0, po
30.0, po
50.0, po
87
94
a Number of animals in parentheses. b ND, not determined.
Table 5. Effects of 7 and 8 on Hemodynamics and ECG Intervals in Anesthetized Guinea Pigs
baseline
0.1 mg/kg
0.3 mg/kg
1 mg/kg
3 mg/kg
10 mg/kg
MAPa (mm Hg)
heart rate (beats/min)
QT (ms)
8
51 ( 4
4 ( 2
0
5 ( 2
-5
-2 ( 9
2 ( 2
-9
-4 ( 6
2 ( 2
9
5 ( 2
-10
-2 ( 6
1 ( 1
-12
-2 ( 7
2 ( 1
11
2 ( 2
-15
-3 ( 3
-2 ( 1
-17
-4 ( 5
5 ( 1
14
-12 ( 9
-21
-2 ( 4
-11 ( 2
-21
-6 ( 5
9 ( 1
21
7
vehicle
47 ( 12
219 ( 6
3 ( 5
0 ( 1
-5
-1 ( 2
1 ( 1
4
8
7
vehicle
242 ( 12
176 ( 3
8
7
vehicle
175 ( 18
336 ( 4
2 ( 1
1 ( 0
4
2 ( 2
2 ( 1
4
3 ( 2
3 ( 1
5
6 ( 3
3 ( 2
5
8 ( 4
2 ( 1
10
QTc (ms)
8
7
vehicle
353 ( 9
72 ( 3
-3 ( 5
2 ( 2
2
0 ( 9
7 ( 5
1
2 ( 9
7 ( 3
6
2 ( 8
6 ( 5
8
4 ( 7
4 ( 7
13
PR (ms)
8
7
vehicle
68 ( 2
27 ( 4
3 ( 3
0 ( 0
3
4 ( 2
0 ( 0
2
2 ( 2
-2 ( 1
2
1 ( 2
-1 ( 2
3
0 ( 2
0 ( 3
6
QRS (ms)
8
7
vehicle
36 ( 1
-2 ( 3
-3 ( 3
-4 ( 3
-5 ( 3
-4 ( 3
a Mean arterial pressure.
guinea pigs at doses of 0.1, 0.3, 1, 3, and 10 mg/kg. At 10 mg/
kg iv, 7 increased the QT interval 21% from baseline, and
QTc, PR, and QRS intervals were lengthened by 10, 13, and
6%, respectively. Compound 8 had no significant effects on
mean arterial pressure through 10 mg/kg (the apparent
decrease was due to nondose-dependent blood pressure
decreases in one animal) and no effect on heart rate except
for a moderate decrease at 10 mg/kg. QT interval, QTc, PR
interval, and QRS duration were not significantly affected by
8 at any dose evaluated. Furthermore, at concentrations up
to 10 μM, 8 had no chronotropic or inotropic effects in the
isolated right atrium of the guinea pig. Therefore, 8 was
deemed to have the more favorable profile and was chosen
for continued in vivo assessment.
Despite the slightly increased potency of 7 in vitro [2ꢀ
aPTT = 342 nM vs 510 nM for 8 (Table 3)], the more
favorable hERG channel binding value (25% inhibition at
10 μM vs 50% inhibition at 10 μM for 8), and the lower level
of human plasma protein binding (49 vs 67% for 8), 7
compared less favorably to 8 in other respects. As mentioned
above, the rat and dog PK parameters of 8 were generally
superior to those of 7 with the exception of the t1/2 in dogs.
Compound 8 was studied for hemodynamic and electro-
physiological effects after oral administration (20 mg/kg) in
conscious dogs. Relative to solvent, 8 had no statistically
significant effect on heart rate, systolic blood pressure,
pressure rate product, cardiac output, stroke volume, the
duration of the PQ, QRS, QT, QTcB, QTcF, and QTcVdW
intervals, QT dispersion, or on ECG morphology. It tended
to decrease diastolic blood pressure (peak effect -20% vs
baseline at 165 min) and systemic vascular resistance (peak
effect -26% vs baseline at 75 min) and increase cardiac
contractility (LV dp/dt max, LV dp/dt max/pd) and relaxa-
tion (LV dp/dt min).
Target Selectivity for 8. Compound 8 was tested against a
panel of 50 receptors and ion channels (Cerep International,
France) and did not inhibit paradigm compound binding to
any of the receptors by more than 26% at a concentration of
1 μM. Receptors that demonstrated inhibition >10% at 1
μM (AT1, H1central, MC4, NK2, and κ) and all ion channels
were reassessed at 10 μM. Even at the higher concentration, 8
inhibited paradigm compound binding to only one receptor
(L-type Ca2þ channel) more than 29%.
Human Clinical Studies. On the basis of these noteworthy
preclinical data, 8 was advanced into phase 1 safety and
biomarker assessment trials in humans. A double-blind,
randomized, placebo-controlled study was conducted to
evaluate the safety, tolerability, pharmacokinetics, and phar-
macodynamics of single oral doses of 8 (solution for-
mulation) in healthy male volunteers. Eight cohorts of
subjects were included. In each cohort, six subjects were
randomized to receive a single oral dose of 8 (4, 12, 36, 100,
200, 300, 400, or 500 mg), and two subjects were randomized
to receive placebo. Ascending doses of 8 were administered
to each cohort. Subjects received a single oral dose of 8 as a
solution or matching placebo on the morning of day 1 with
200 mL of water. All subjects had fasted for at least 10 h prior
to dose administration. Venous blood samples (5 mL) were
collected predose (within 60 min prior to dosing) and at 0.5,
1, 1.5, 2, 2.5, 3, 4, 6, 8, 12, 24, and 48 h postdosing for

1850 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Lu et al.
determination of plasma concentrations of 8. The pharma-
codynamic parameters assessed in this study focused on
evaluating the effect of 8 on coagulation and hemostasis.
Evaluated were the following parameters: aPTT, expressed
in seconds and as a ratio relative to individual subject base-
line (-1 h) values; prothrombin time (PT), expressed in
seconds and International Normalized Ratio; ecarin clotting
time (ECT), expressed in seconds; and activated clotting time
(ACT), expressed in seconds.
Compound 8 was rapidly absorbed into the plasma: the
median tmax ranged from 1.25 to 2.50 h (Figure 5; linear and
semilogarithmic scales shown). However, subjects generally
showed more than one peak in the pharmacokinetic profile.
Multiple peaks were also observed in rat studies after oral
dosing and are hypothesized to be due to enterohepatic
recirculation, supported by results of an additional rat
biliary excretion study (data not shown).
The mean terminal half-life was consistent across the
dose range and ranged from 3.85 to 5.57 h; however, due
to minimal sampling times in the 12-48 h period, these
data should be interpreted with caution. The majority of 8
was removed from the plasma within 12 h postdosing:
Concentrations at 12 h postdose were <17% of the Cmax
.
The mean apparent volume of distribution (V/F) ranged
from 530 to 806 L, and the mean apparent clearance (CL/F)
varied from 85.3 to 109 L/h across the 100-500 mg dose
range.
As compared with placebo, mean aPTT values increa-
sed in subjects receiving 8 at doses of 100 mg or higher
(Figure 6). Mean aPTT values peaked between 0.5 and 4 h
postdosing (around the same time maximum plasma con-
centrations were observed) and returned to baseline at
around 24 h postdose. The overall effect of 8 on aPTT
increased with higher doses, with the greatest effect seen
in the 500 mg dose group. A similar pattern of results
was observed for the other ex vivo clotting parameters
(PT, ACT, and ECT).
Figure 5. Mean plasma concentration-time profiles of 8 in healthy
human volunteers.
Figure 6. Mean plasma aPTT-time profiles as a function of 8 in healthy human volunteers.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1851
Conclusion
acetate (200 mL) was added, washed with saturated NaHCO3
(2 ꢀ 100 mL) and brine (100 mL), dried over Na2SO4, and
filtered. After the filtrate was evaporated, the residue was
purified by flash column chromatography (methylene chloride/
ethyl acetate, 1:0 to 96:4, v/v) to give compound 1 as a white solid
(9.3 g, 91%). 1H NMR (CDCl3): δ 7.84 (m, 2H), 7.78 (m, 2H),
7.37(m, 5H), 5.97 (br s, 1H), 5.14 (s, 2H), 4.27 (t, J = 4.9 Hz, 2H),
3.51 (q, J = 5.2 Hz, 2H).
2-(Benzyloxycarbonylamino)ethoxyamine (2). To a solution
of 1 (1.4 g, 4.0 mmol) in ethanol (20 mL) and tetrahydrofuran
(20 mL) was added 40% methylamine (2.0 mL, 25 mmol), and
the mixture stirred at room temperature for 1 h. After the solvent
was evaporated, the residue was purified by flash column
chromatography (hexane/ethyl acetate, 3:1 to 0:1, v/v) to give
2 as a white solid (800 mg, 95%). 1H NMR (CDCl3): δ 7.36 (m,
5H), 5.47 (br s, 2H), 5.21 (br s, 1H), 5.10 (s, 2H), 3.72 (t, J = 5.0
Hz, 2H), 3.44 (q, J = 5.0 Hz, 2H).
[N,N0-Di-(tert-butoxycarbonyl)]-2-(benzyloxycarbonylamino)-
ethoxyguanidine (3). To a solution of 2 (0.78 g, 3.7 mmol) in N,N-
dimethylformamide (20 mL) was added [N,N0-di-(tert-butoxy-
carbonyl)]amidinopyrazole (1.3 g, 4.0 mmol). The mixture was
stirred at room temperature overnight, and the solvent was
evaporated under high vacuum. The residue was purified by
flash column chromatography (methylene chloride/ethyl acet-
ate, 1:0 to 95:5, v/v) to give 3 as a colorless oil (1.55 g, 93%). 1H
NMR (CDCl3): δ 9.08 (s, 1H), 7.33 (m, 5H), 6.21 (br s, 1H), 5.21
(br s, 1H), 5.11 (s, 2H), 4.12 (t, J = 4.8 Hz, 2H), 3.54 (q, J = 4.9
Hz, 2H), 1.49 (s, 9H), 1.46 (s, 9H).
[N,N0-Di-(tert-butoxycarbonyl)]-2-aminoethoxyguanidine (4).
A mixture of 4 (730 mg, 1.5 mmol), as prepared in the preceding
step, 10% palladium on carbon (70 mg) in ethanol (20 mL), and
tetrahydrofuran (20 mL) was hydrogenated under hydrogen
(balloon) for 30 min. The catalysts were removed by filtration
through diatomaceous earth, and the filtrate was concentrated
in vacuo. The residue was purified on a 10 g Waters Sep-Pak
(methylene chloride/methanol saturated with ammonia, 95:5,
v/v) to give the title compound as a colorless oil (290 mg, 61%).
1H NMR (CDCl3): δ 9.08 (br s, 1H), 4.08 (t, J = 5.2 Hz, 2H),
2.99 (q, J = 5.1 Hz, 2H), 1.50 (s, 9H), 1.48 (s, 9H).
We have explored a series of potent nonpeptidic pyrazi-
none-based thrombin inhibitors featuring a P1 oxyguanidine
moiety, which led to 8 as a promising clinical candidate. This
compound exhibited good binding affinity, notable results in a
variety of in vitro screening assays, greatly improved PK as
compared to thrombin inhibitors with highly basic guanidine/
amidine P1 groups, and impressive efficacy in several in vivo
thrombosis models, such as ECAT, AV shunt, and DVT.
Given the favorable preclinical efficacy and safety profiles for
8, it was advanced into human clinical studies. The initial
results from the successful completion of phase I clinical
safety, tolerability, PK, and pharmacodynamic studies of 8
were promising.
Experimental Section
Chemistry. Reagents and solvents were obtained from com-
mercial suppliers and used without further purification. Flash
chromatography was performed using Fisher Chemical Silica
Gel Sorbent (230-400 mesh, grade 60). 1H NMR spectra were
recorded on a Bruker B-ACS-120 (400 MHz) spectrophot-
ometer at room temperature. Chemical shifts are given in ppm
(δ), coupling constants (J) are in hertz (Hz), and signals are
designated as follows: (s) singlet, (d) doublet, (t) triplet, (q)
quartet, (m) multiplet, (br s) broad singlet, (br t) broad triplet,
(br m) broad multiplet, (dd) doublet of doublet, (dt) doublet of
triplet, (td) triplet of doublet, (dq) doublet of quartet, (tm)
triplet of multiplet, (ddd) doublet of doublet of doublet, (ddt)
doublet of doublet of triplet, and (dddd) doublet of doublet of
doublet of doublet. LC-MS was performed on a system consist-
ing of an electrospray source on a Finnigan LCQ ion trap mass
spectrometer, a SEDEX 75C evaporative light scattering detec-
tor, a Shimadzu LC-10ADvp binary gradient pumping system, a
Gilson 215 configured as an autosampler, and a Princeton
Chromatography HTS HPLC column (5 μm, 50 mm ꢀ 3.0
mm). Accurate mass measurements were performed using either
of two methods: (1) A Micromass (Manchester, United King-
dom) Autospec E OA-TOF high-resolution magnetic sector
mass spectrometer was tuned to a resolution of >5000 using
the 5% valley definition. The ions were produced in a fast atom
bombardment (FAB) ion source at an accelerating voltage of 8
kV. Linear voltage scans at 33 Da/s were collected to include the
protonated sample ion and two polyethylene glycol (PEG) or
PEG monomethyl ether ions, which were used as internal
reference standards. The molecular mass was calculated using
a linear extrapolation method. Reported mass values are
averages over 100 s. (2) Positive ion electrospray (ESI) high-
resolution mass spectrometry was carried out with a Micromass
Q-Tof API US hybrid quadrupole/time-of-flight mass spectro-
meter. The mass spectrometer was externally calibrated using
cesium iodide. PEG 400 was added to the samples as an internal
calibrant. The purities of some key target compounds were
determined on a Varian HPLC system equipped with two
Varian PrepStar delivery pumps (model 218) and a Varian
UV-vis detector (model 345). The HPLC method used was as
follows: column, PrincetonSPHER-100 Phenyl, 4.6 mm ꢀ 150
mm, 5 μm; column temperature, ambient; flow rate, 1.0 mL/
min; gradient, 20% acetonitrile (0.01% TFA and 0.2% formic
acid) in water (0.01% TFA and 0.2% formic acid) to 80%
acetonitrile (0.01% TFA and 0.2% formic acid) in water (0.01%
TFA and 0.2% formic acid) in 30 min. All of the final com-
pounds had 95% or greater purity.
Representative Procedure for the Synthesis of 1-{N-[2-(Amidino-
aminooxy)ethyl]amino}carbonylmethyl-6-substituted-3-alkylpyra-
zinones (7-18). Synthesis of 7. Step A: [6-Cyano-3-(2,2-difluoro-
2-phenyl-ethylamino)-2-oxo-2H-pyrazin-1-yl]acetic Acid Ethyl
Ester. Compound 5 (0.20 g, 0.83 mmol) was dissolved in toluene
(5 mL), treated with 2,2-difluoro-2-phenylethylamine (0.80 g, 5.1
mmol), and heated to 60 °C for 3 h. The reaction was concen-
trated to dryness, and the residue was dissolved in dichloro-
methane, washed with 5% aqueous HCl, saturated NaHCO3,
and brine, dried over Na2SO4, and filtered. The filtrate was
concentrated to dryness, adsorbed onto silica gel, poured onto
a 7 cm ꢀ 3 cm pad of silica gel, and eluted with ethyl acetate/
dichloromethane, 0:1 to 1:9, v/v. The fractions were concentrated
in vacuo giving [6-cyano-3-(2,2-difluoro-2-phenyl-ethylamino)-
2-oxo-2H-pyrazin-1-yl]acetic acid ethyl ester as a light orange
1
solid (0.29 g, 95%). H NMR (CDCl3): δ 7.53 (m, 2H), 7.46
(m, 3H), 7.38 (s, 1H), 6.95 (br m, 1H), 4.79 (s, 2H), 4.29 (q, 2H,
J = 7.1 Hz), 4.16 (td, 2H, J = 6.5 Hz, 14.2 Hz), 1.32 (t, 3H,
J = 7.1 Hz).
Step B: [6-Cyano-3-(2,2-difluoro-2-phenyl-ethylamino)-2-oxo-
2H-pyrazin-1-yl]acetic Acid. The product of step A (0.28 g, 0.77
mmol) was dissolved in 1,4-dioxane (10 mL) and treated with 1
N NaOH (4 mL) at ambient temperature. After it was stirred for
1.5 h, the reaction was concentrated to dryness, dissolved in hot
water (10 mL), acidified with 10% aqueous HCl, and cooled to
ambient temperature, and the precipitate was filtered. The solids
were washed with water, dissolved in tetrahydrofuran, and
filtered, and the filtrate was concentrated in vacuo giving
[6-cyano-3-(2,2-difluoro-2-phenyl-ethylamino)-2-oxo-2H-pyr-
azin-1-yl]acetic acid as a gummy red-orange solid (0.27 g, 100%
for the hydrate). 1H NMR (DMSO-d6): δ 8.64 (t, 1H,
N-[2-(Benzyloxycarbonylamino)ethoxy]phthalimide (1). To a
solution of benzyl N-(2-hydroxyethyl)carbamate (5.9 g, 30
mmol), N-hydroxyphthalimide (4.9 g, 30 mmol), and triphenyl-
phosphine (7.9 g, 30 mmol) in tetrahydrofuran (100 mL) was
added diethyl azodicarboxylate (5.2 g, 30 mmol). The reaction
mixture was stirred at room temperature overnight. Ethyl

1852 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Lu et al.
J = 6.6 Hz), 7.62 (s, 1H), 7.53 (m, 5H), 4.72 (s, 2H), 4.14 (td,
2H, J = 6.6 Hz, 15 Hz). MS m/z [MHþ] 334.9.
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-(2-[4-trifluoromethylphenyl]ethylamino)pyrazinone Bis-
trifluoroacetic Acid Salt (12). 1H NMR (DMSO-d6): δ 10.98 (s,
1H), 8.45 (t, 1H, J = 5.5 Hz), 7.74 (br s, 4H), 7.66 (d, 2H, J = 8.1
Hz), 7.47 (d, 2H, J = 8.0 Hz), 6.68 (s, 1H), 4.62 (s, 2H), 3.81 (t,
2H, J = 5.4 Hz), 3.50 (m, 2H), 3.38 (q, 2H, J = 5.4 Hz), 2.97 (t,
Step C: 1-{N-[2-(N0,N00-Bis{tert-butoxycarbonyl}amidino-
aminooxy)ethyl]amino}carbonylmethyl-6-cyano-3-[2,2-difluoro-
2-phenylethylamino]pyrazinone. The product of step B (0.27 g,
0.77 mmol) was dissolved in anhydrous N,N-dimethylforma-
mide (10 mL) and treated with 4, Castro’s reagent (0.42 g, 0.95
mmol), and triethylamine (0.30 mL, 2.2 mmol). After it was
stirred at room temperature for 18 h, the reaction was concen-
trated to an oil, dissolved in dichloromethane, washed with 10%
aqueous citric acid, water, and brine, dried over Na2SO4, and
filtered. The filtrate was concentrated to dryness, the crude
product was purified by flash column chromatography on silica
gel with ethyl acetate/hexanes, 1:1 to 2:1, v/v, and the fractions
were concentrated in vacuo giving 1-{N-[2-(N0,N00-bis{tert-
butoxycarbonyl}amidinoaminooxy) ethyl]amino}carbonylmethyl-
6-cyano-3-[2,2-difluoro-2-phenylethylamino]pyrazinone as a pale
yellow solid (0.35 g, 71%). 1H NMR (CDCl3): δ 9.19 (br s, 1H),
8.78 (br s, 1H), 7.55 (m, 2H), 7.48 (m, 3H), 7.37 (s, 1H), 6.94 (t, 1H,
J = 6.3 Hz), 4.88 (s, 2H), 4.15 (m, 4H), 3.66 (m, 2H), 1.54 (s, 9H),
1.51 (s, 9H). MS m/z [MHþ] 634.8.
2H, J = 7 Hz), 2.08 (s, 3H). MS m/z [MHþ] 456.2. HPLC tR
=
6.32 min, >98% pure.
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-(4-methoxyphenethylamino)pyrazinone Bis-trifluoro-
1
acetic Acid Salt (13). H NMR (DMSO-d6): δ 10.88 (s, 1H),
8.42 (t, J = 5.5 Hz, 1H), 7.68 (br s, 4H), 7.15 (d, J = 8.6 Hz, 2H),
6.85 (d, J = 8.7 Hz, 2H), 6.67 (s, 1H), 4.61 (s, 2H), 3.81 (t, J =
5.3 Hz, 2H), 3.72 (s, 3H), 3.50 (m, 2H), 3.38 (m, 2H), 2.79 (t, J =
6.9 Hz, 2H), 2.07 (s, 3H). MS m/z [MHþ] 418.3. Anal.
(C19H27N7O4 2C2HF3O2 0.75H2O) C, H, N.
3
3
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-(3,4-dimethoxyphenylethylamino)pyrazinone Bis-tri-
1
fluoroacetic Acid Salt (14). H NMR (DMSO-d6): δ 10.95 (s,
1H), 8.44 (t, J = 5.6 Hz, 1H), 7.72 (br s, 4H), 6.85 (t, J = 8.1 Hz,
2H), 6.73 (d, J = 8.1 Hz, 1H), 6.67 (s, 1H), 4.61 (s, 2H), 3.81 (t,
J = 5.3 Hz, 2H), 3.73 (s, 3H), 3.71 (s, 3H), 3.39 (m, 4H), 2.79 (t,
J = 7.2 Hz, 2H), 2.07 (s, 3H). MS m/z [MHþ] 448.2. HPLC
Step D: 1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonyl-
methyl-6-cyano-3-[2,2-difluoro-2-phenylethylamino]pyrazinone
Dihydrochloride (7). The product of step C (0.34 g, 0.54 mmol)
was dissolved in dichloromethane (10 mL), treated with trifluor-
oacetic acid (5 mL), and stirred at ambient temperature for 8 h.
The reaction was concentrated to dryness, the crude product
was purified by flash column chromatography on silica gel
with methanol/dichloromethane saturated with ammonia gas,
85:15, v/v, and the fractions were concentrated in vacuo. The
column fractions were treated with 4 N HCl in ethanol, con-
centrated under high vacuum, and filtered, giving 7 as a white
solid (0.19 g, 82%). 1H NMR (DMSO-d6): δ 8.50 (br t, 1H, J =
6.1 Hz), 8.41 (t, 1H, J = 5.5 Hz), 7.57 (s, 1H), 7.52 (m, 5H), 5.13
(br s, 2H), 4.62 (s, 2H), 4.40 (br s, 2H), 4.13 (td, 2H, J = 5.5 Hz,
15 Hz), 3.63 (t, 2H, J = 5.6 Hz), 3.30 (q, 2H, J = 5.6 Hz).
MS m/z [MHþ] 435.1. Anal. (C18H20F2N8O3 2HCl 0.1H2O
t
R = 4.84 min, >98% pure.
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-(4-ethylphenethylamino)pyrazinone Bis-trifluoro-
acetic Acid Salt (15). 1H NMR (DMSO-d6): δ 10.97 (s, 1H),
8.44 (t, J = 5.6 Hz, 1H), 7.73 (br s, 4H), 7.14 (s, 4H), 6.67 (s,
1H), 4.61 (s, 2H), 3.81 (t, J = 5.2 Hz, 2H), 3.39 (m, 4H), 2.82
(t, J = 7.5 Hz, 2H), 2.56 (q, J = 7.6 Hz, 2H), 2.07 (s, 3H), 1.16
(t,
(C20H29N7O3 2C2HF3O2 0.5H2O) C, H, N.
J =
7.5 Hz, 3H). MS m/z [MHþ] 416.2. Anal.
3
3
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-(2-[5-indanyl]ethylamino)pyrazinone Dihydrochlo-
1
ride (16). H NMR (DMSO-d6): δ 11.13 (s, 1H), 8.71 (t, 1H,
J = 5.5 Hz), 7.78 (br s, 4H), 7.19 (s, 1H), 7.14 (d, 1H, J = 7.7
Hz), 7.07 (m, 1H), 6.70 (s, 1H), 4.67 (s, 2H), 3.82 (t, 2H, J =
5.3 Hz), 3.65 (m, 2H), 2.84 (m, 6H), 2.13 (s, 3H), 1.99 (pentet,
2H, J = 7.4 Hz). MS m/z [MHþ] 428.3. HPLC tR = 5.20 min,
>98% pure.
3
3
3
0.17CH3N) C, H, N.
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-[2,2-difluoro-2-phenylethylamino]pyrazinone Dihydro-
chloride (8). Yield, 37%; mp 211-214 °C. 1H NMR (400
MHz, DMSO-d6): δ 10.89 (s, 1H), 8.46 (t, J = 5.5 Hz, 1H),
7.69 (br s, 4H), 7.53 (m, 2H), 7.50 (m, 3H), 6.84 (t, J = 6.5 Hz,
1H), 6.64 (s, 1H), 4.62 (s, 2H), 4.06 (dt, J = 6.5, 14.4 Hz,
2H), 3.81 (t, J = 5.4 Hz, 2H), 3.38 (m, 2H), 2.06 (s, 3H).
MS m/z [MHþ] 424.4. Anal. (C18H23F2N7O3 2HCl 0.15H2O)
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-[2-(1-naphthalene)ethyl]aminopyrazinone Dihydro-
chloride (17). 1H NMR (400 MHz, DMSO-d6): δ 11.1 (s,
1H), 8.71 (t, 1H, J = 5.3 Hz), 8.26 (d, 1H, J = 8.2 Hz), 7.94
(d, 1H, J = 8.1 Hz), 7.83 (d, 1H, J = 8.2 Hz), 7.78 (br s, 4H),
7.56 (m, 3H), 7.46 (t, 1H, J = 7.5 Hz), 6.73 (s, 1H), 4.69 (s,
2H), 3.80 (m, 4H), 3.42 (m, 4H), 2.14 (s, 3H). MS m/z [MHþ]
438.2. Anal. (C22H27N7O3 2HCl 0.75H2O) C, H, N.
3
3
C, H, N.
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-(4-fluorophenethylamino)pyrazinone Bis-trifluoro-
acetic Acid Salt (9). 1H NMR (DMSO-d6): δ 10.97 (s, 1H),
8.45 (t, J = 5.6 Hz, 1H), 7.74 (br s, 4H), 7.26 (m, 2H), 7.15 (t, J =
8.8 Hz, 2H), 6.68 (s, 1H), 4.61 (s, 2H), 3.81 (t, J = 5.4 Hz, 2H),
3.39 (m, 4H), 2.85 (t, J = 7.3 Hz, 2H), 2.07 (s, 3H). MS m/z
[MHþ] 406.6. HPLC tR = 5.06 min, >98% pure.
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-(2-[3,4-difluorophenyl]ethylamino)pyrazinone Dihydro-
chloride (10). 1H NMR (DMSO-d6): δ 11.05 (s, 1H), 8.63 (t,
1H, J = 5.5 Hz), 7.74 (br s, 4H), 7.46 (t, 1H, J = 10 Hz), 7.36 (dt,
1H, J = 10.9 Hz, 8.5 Hz), 7.15 (m, 1H), 6.70 (s, 1H), 4.66 (s, 2H),
3.82 (t, 2H, J = 5.4 Hz), 3.66 (m, 2H), 3.38 (m, 2H), 2.91 (t, 2H,
J = 7.4 Hz), 2.13 (s, 3H). MS m/z [MHþ] 424.4. HPLC tR = 5.33
min, >98% pure.
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-
6-methyl-3-(2-[2,4-difluorophenyl]ethylamino)pyrazinone Bis-tri-
fluoroacetic Acid Salt (11). 1H NMR (DMSO-d6): δ 11.03 (s,
1H), 8.46 (t, J = 5.6 Hz, 1H), 7.77 (br s, 4H), 7.34 (q, J = 7.8 Hz,
1H), 7.18 (t, J = 9.5 Hz, 1H), 7.03 (t, J = 8.0 Hz, 1H), 6.67 (s, 1H),
4.62 (s, 2H), 3.81 (t, J = 5.4 Hz, 2H), 3.51 (m, 2H), 3.39 (t, J = 5.5
Hz, 2H), 2.89 (t, J = 7.1 Hz, 2H), 2.08 (s, 3H). MS m/z [MHþ]
424.4. HPLC tR = 5.81 min, >98% pure.
3
3
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
methyl-3-(2,2-diphenylethylamino)pyrazinone Bis-trifluoro-
acetic Acid Salt (18). 1H NMR (300 MHz, DMSO-d6): δ
10.92 (s, 1H), 8.44 (t, J = 5.6 Hz, 1H), 7.71 (br s, 4H), 7.11
(s, 4H), 6.67 (s, 1H), 4.61 (s, 2H), 3.81 (t, J = 5.4 Hz, 2H), 3.50
(m, 2H), 3.38 (m, 2H), 2.81 (t, J = 6.9 Hz, 2H), 2.27 (s, 3H),
2.07 (s, 3H). MS m/z [MHþ] 402.1. Anal. (C24H29N7O3
3
2C2HF3O2 1.0H2O) C, H, N.
3
[6-Chloro-3-(2,2-difluoro-2-phenyl-ethylamino)-2-oxo-2H-
pyrazin-1-yl]acetic Acid (20). Compound 19 (0.78 g, 3.0 mmol)
was dissolved in toluene (10 mL), treated with 2,2-difluoro-2-
phenylethylamine (1.0 g, 6.4 mmol), heated to reflux for 12 h,
and then stirred at ambient temperature for 3 days. The reaction
was concentrated to dryness in vacuo, the crude product was
purified by flash column chromatography on silica gel with
hexanes/ethyl acetate, 1:1, v/v as eluant, and the column frac-
tions were concentrated in vacuo giving [3-(2,2-difluoro-2-
phenyl-ethylamino)-2-oxo-2H-pyrazin-1-yl]acetic acid ethyl es-
1
ter as a pale orange solid (0.75 g, 75%). H NMR (400 MHz,
CDCl3): δ 7.56 (m, 2H), 7.45 (m, 3H), 6.83 (d, 1H, J = 4.7 Hz),
6.42 (d, 1H, J = 4.7 Hz), 6.36 (m, 1H), 4.57 (s, 2H), 4.26 (q, 2H,
J = 7 Hz), 4.13 (td, 2H, J = 6 Hz, 14 Hz), 1.31 (t, 3H, J = 7 Hz).

Article
Next, a solution of [3-(2,2-difluoro-2-phenyl-ethylamino)-2-
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1853
at ambient temperature, the solution was concentrated in vacuo,
and the residue was treated with pH 1 brine and made basic with
solid KOH. The resulting precipitate was filtered, washed with
water, and removed from the filter with methanol. The filtrate was
concentrated in vacuo, dissolved in methanol/dichloromethane,
1:1, v/v, and filtered. The filtrate was evaporated, and the residue
was treated with 2 mL of 4 N HCl in ethanol, dissolved in
methanol/dichloromethane, and refiltered, giving the title com-
oxo-2H-pyrazin-1-yl]acetic acid ethyl ester (0.75 g, 2.2 mmol)
and N-chlorosuccinimide (0.30 g, 2.3 mmol) in 1,2-dichlor-
oethane (40 mL) was heated to reflux for 2.5 h. After it was
cooled to room temperature, the reaction was washed with
saturated aqueous NaHCO3 and brine, dried over Na2SO4,
and filtered, and the filtrate was concentrated to dryness. The
crude product was then adsorbed onto silica gel, poured onto a 3
cm ꢀ 6 cm pad of silica gel, eluted with hexanes/ethyl acetate,
3:1 to 2:1, v/v, and the fractions were concentrated in vacuo
giving [6-chloro-3-(2,2-difluoro-2-phenyl-ethylamino)-2-oxo-
2H-pyrazin-1-yl]acetic acid ethyl ester as a pale orange solid
(0.67 g, 81%). 1H NMR (400 MHz, CDCl3): δ 7.53 (m, 2H), 7.43
(m, 3H), 6.90 (s, 1H), 6.25 (br t, 1H, J = 5.8 Hz), 4.89 (s, 2H),
4.26 (q, 2H, J = 7 Hz), 4.09 (td, 2H, J = 6 Hz, 14 Hz), 1.30 (t,
3H, J = 7.2 Hz). MS m/z [MHþ] 352.0.
1
pound as a yellow solid (0.08 g, 86%). H NMR (400 MHz,
DMSO-d6): δ 8.62 (t, 1H, J = 5.3 Hz), 7.90 (br s, 3H), 7.18 (d, 2H,
J=7.9Hz), 7.11(d, 2H, J=7.9Hz), 6.70(s, 1H), 4.64(s, 2H), 3.91
(t, 2H, J = 5.2 Hz), 3.39 (q, 2H, J = 5.2 Hz), 3.25 (s, 3H), 2.86 (t,
2H, J = 7.5 Hz), 2.27 (s, 3H), 2.12 (s, 3H). MS m/z [MHþ] 416.2.
HPLC tR = 6.21 min, >98% pure.
Enzymology. All compounds were assessed for inhibitory
activity toward thrombin, as well as the key enzymes fXa and
trypsin, by kinetic analysis using p-nitroaniline chromogenic
substrates. The assay buffer employed was 50 mM Hepes, pH
7.5, 200 mM NaCl, and fresh 0.05% n-octyl β-D-glucopyrano-
side. DMSO was present at a final concentration of 4%. In a 96-
well low-binding polystyrene plate, 280 μL of substrate in assay
buffer was preincubated at 37 °C for 15 min with 10 μL of test
compound in DMSO to obtain final test compound concentra-
tions that bracketed the Ki. Reactions were initiated by addition
of 10 μL of protease, and the increase in absorbance due to
proteolytic cleavage of substrate was kinetically monitored at
37 °C and 405 nm with a Molecular Devices Spectramax 340
platereader. Initial velocities were determined by analysis of the
initial linear portion of the reactions. Plots of vo/vi versus
inhibitor concentration, where vo = velocity without inhibitor
and vi = inhibited velocity, were fit to a linear regression line,
and the IC50 value was determined from the reciprocal of the
slope. Alternatively, IC50 values were determined using Graph-
Pad Prism software and a four-parameter logistics equation. Ki
was calculated from the IC50 using the Ki factor specific for the
assay as: Ki = IC50 ꢀ Ki factor or Ki = IC50 ꢀ {1/(1 þ [S]/Km)},
where S is the substrate concentration in the assay and Km is the
Michaelis constant for the substrate.30
[6-Chloro-3-(2,2-difluoro-2-phenyl-ethylamino)-2-oxo-2H-
pyrazin-1-yl]acetic acid ethyl ester (0.67 g, 1.8 mmol) was then
dissolved in 1,4-dioxane (24 mL) and treated with 1 N aqueous
NaOH (8.0 mL, 8.0 mmol). After it was stirred at ambient
temperature for 18 h, the reaction was concentrated to dryness,
dissolved in water, and treated with 10% aqueous HCl to pH 4,
and the white precipitate was filtered, washed thrice with water
and once with hexane, and dried in vacuo giving 20 as a white
solid (0.56 g, 90%). 1H NMR (400 MHz, DMSO-d6): δ 7.50 (m,
5H), 6.95 (s, 1H), 4.79 (s, 2H), 4.06 (td, 2H, J = 6.5 Hz, 15 Hz).
1-{N-[2-(Amidinoaminooxy)ethyl]amino}carbonylmethyl-6-
chloro-3-[2,2-difluoro-2-phenylethylamino]pyrazinone Dihydro-
chloride (21). Compound 20 (0.56 g, 1.6 mmol) was dissolved
in anhydrous N,N-dimethylformamide (25 mL) and treated with
[N,N0-di(tert-butoxycarbonyl)]-2-aminoethoxyguanidine (0.64
g, 2.0 mmol), Castro’s reagent (0.89 g, 2.0 mmol), and triethy-
lamine (1.0 mL, 7.2 mmol). After it was stirred at ambient
temperature for 3 days, the reaction was concentrated to dryness
in vacuo, the residue was dissolved in dichloromethane, washed
with 10% aqueous citric acid, water, saturated aqueous NaHCO3,
and brine, dried over Na2SO4, and filtered, and the filtrate
was concentrated in vacuo. The crude product was adsorbed
onto silica gel, poured onto a 7 cm ꢀ 3 cm pad of silica gel, and
eluted with hexanes/ethyl acetate, 2:1 to 0:1, v/v, and the
fractions were concentrated in vacuo giving 1-{N-[2-(N0,N00-
bis{tert-butoxycarbonyl}amidinoaminooxy)ethyl]amino}carbonyl-
methyl-6-chloro-3-[2,2-difluoro-2-phenylethylamino]pyrazinone as
The thrombin assay incorporated substrate SucAAPR
pNA (Bachem L-1720, [S] = 100 μM final, Km = 320 μM,
and Ki factor = 0.76). Human R-thrombin (Enzyme Research
Laboratories HT1002a) was diluted in assay buffer for a final
assay concentration of 1.1 nM. The fXa assay incorporated
substrate S-2765 (Diapharma/Chromogenix S-2765, Z-D-Arg-
1
a light yellow solid (0.91 g, 86%). H NMR (400 MHz, CDCl3):
Gly-Arg-pNA 2HCl, [S] = 100 μM final, Km = 260 μM, and Ki
δ 9.15 (s, 1H), 8.55 (br t, 1H, J = 5 Hz), 7.53 (m, 3H [Aryl þ NH]),
7.43 (m, 3H), 6.88 (s, 1H), 6.25 (t, 1H, J = 6.4 Hz), 4.96 (s, 2H), 4.11
(m, 4H), 3.63(dd, 2H, J = 4.9 Hz, 8.5 Hz), 1.51 (s, 9H), 1.49 (s, 9H).
1-{N-[2-(N0,N00-Bis{tert-butoxycarbonyl}amidinoaminooxy)-
ethyl]amino}carbonylmethyl-6-chloro-3-[2,2-difluoro-2-phenyl-
ethylamino]pyrazinone (0.91 g, 1.4 mmol) was then dissolved
in dichloromethane (10 mL), treated with trifluoroacetic acid
(5 mL), and stirred at ambient temperature for 24 h. The reaction
was concentrated in vacuo, the residue purified by flash column
chromatography on silica gel with methanol/dichloromethane
saturated with ammonia gas, 1:9 to 15:85, v/v, and the fractions
were concentrated in vacuo. The column fractions were treated
with 4 N HCl in ethanol, concentrated under high vacuum, and
filtered, giving 21 as a white solid (0.53 g, 85%). 1H NMR (400
MHz, DMSO-d6): δ 8.32 (m, 1H), 7.52 (m, 5H), 7.40 (t, 1H, J =
7 Hz), 6.92 (s, 1H), 5.16 (br m, 1H), 4.72 (s, 2H), 4.42 (br m, 1H),
4.06 (td, 2H, J = 6.6 Hz, 15.2 Hz), 3.62 (t, 2H, J = 5.5 Hz), 3.28
(q, 2H, J = 5.5 Hz). MS m/z [MHþ] 444.1. Anal. (C17H20-
ClF2N7O3 2HCl 0.33H2O) C, H, N.
3
factor = 0.72). Human fXa (Enzyme Research Laboratories
HFXa 1500) was diluted in assay buffer for a final assay
concentration of 0.53 nM.
The trypsin assay also incorporated substrate S-2765 ([S] =
60 μM final, Km = 61 μM, and Ki factor = 0.50). Human trypsin
(Calbiochem 650275) was diluted in assay buffer for 333 pM
final concentration. The within-run assay coefficient of varia-
tion (CV) was generally <10%; the between-run CV was
<20%. The inhibitory capacity of 7 and 8 for other proteases
was determined by CEREP protease assay service.
Crystallization, Data Collection, and Structure Determination
and Refinement. Human R-thrombin was purchased from En-
zyme Research Laboratories and crystallized at 4 °C in a complex
with sulfated hirudin 53-65 (Bachem Biosciences) as previously
described.31 Complexes of thrombin with the hirudin fragment
and active site-directed inhibitors were formed by adding inhi-
bitor solutions to standing vapor diffusion drops (2 μL) contain-
ing large single crystals of μ-thrombin. Crystals were undamaged
by this procedure and were mounted for data collection 4-5 h
after the addition of inhibitors. Compound 8 was dissolved in
DMSO at a concentration of 200 mM and then diluted to 4 mM
in 28% PEG 8000 containing 100 mM phosphate buffer (pH 7.3),
0.15 M NaCl, and 3 mM NaN3. An aliquot (1 μL) of this solution
was added to a drop (2 μL) containing single thrombin crystals to
produce a final inhibitor concentration of 1.3 mM.
3
3
1-{N-[2-(Amidino-N0-methylaminooxy)ethyl]amino}carbonyl-
methyl-6-methyl-3-(2-[4-methylphenyl]ethylamino)pyrazinone
Dihydrochloride (22). To a mixture of 1-{N-[2-(amidinoamino-
oxy)ethyl]amino}carbonylmethyl-6-methyl-3-(4-methylphenethyl-
amino)pyrazinone29 (0.076 g, 0.19 mmol) and sodium bicarbonate
(0.53 g, 6.3 mmol) in N,N-dimethylformamide (10 mL) was added
iodomethane (0.040 mL, 0.64 mmol). After it was stirred for 2 days

1854 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Lu et al.
The crystals were then transferred to a cryoprotectant solu-
tion containing 40% PEG 3350, 10 mM CaCl2, and 100 mM
Tris-maleate at a pH of 5.5 and flash frozen by immersion in
at 70 min postdosing. In both the intravenous and the oral
studies, a blood sample was taken at the termination of the
experiment to measure coagulation times and drug plasma
concentration.
˚
liquid nitrogen. X-ray diffraction data to a resolution of 2.06 A
Rat ECAT Model. Male Sprague-Dawley rats (350-450 g)
were anesthetized with pentobarbital 65 mg/mL (1 mL/kg, i.p.).
A midline incision was made in the neck to expose and isolate the
blood vessels. The right carotid artery was cannulated with a
saline-filled PE-50 catheter attached to a statham pressure
transducer to measure blood pressure and heart rate. The left
carotid artery was isolated, and a pulsed Doppler flow probe
(HVPD-20, Crystal Biotech, Hopkinton, MA) was placed dis-
tally on the artery to measure blood flow. The left jugular vein
was cannulated for administration of the compound or vehicle.
To prevent thrombosis, the arterial catheters were coated with
Sigmacote and air-dried. Proximal to the flow probe, the carotid
artery was laid across the exposed electrode wire in the hook end
of a bipolar Dastre electrode (Harvard Apparatus, Holliston,
MA). The electrode was fitted onto a micromanipulator and
positioned with slight upward tension but so that flow was
unaffected.
Once a steady baseline blood flow was achieved, solutol
vehicle (10%) or 8 (0.03, 0.1, 0.3, or 1.0 mg/kg) was administered
by slow infusion into the left jugular vein for 20 min at a rate of
0.05 mL/min. Electrical stimulation was begun after 5 min of
infusion and continued for 5 min. The flow crystal sensor was
turned off for the duration of electrical stimulation. Occlusion
times were recorded during the 30 min observation period
beginning at the conclusion of electrical stimulation. Differences
in time to occlusion between compound-treated and vehicle-
treated rats were compared using an unpaired t test. P < 0.05
was considered statistically different.
Dog Arteriovenous Shunt Model. Adult mongrel dogs of
either sex (13.2-14.5 kg) were dosed with vehicle or 8 or 7.
Following a 1 h observation period, the dogs were anesthetized
with pentobarbital sodium (35 mg/kg, iv) and ventilated with
room air via an endotracheal tube (15 strokes/min, 25 mL/kg).
The heart rate was monitored using a cardiotachometer trig-
gered from a lead II electrocardiogram generated by limb
leads. A Millar pressure-tipped catheter (Millar Instruments,
Houston, TX) was placed in the left femoral artery for mea-
surement of blood pressure. A carotid artery was cannulated
(PE-200) for blood withdrawal. The right femoral artery and
the right femoral vein were cannulated with silicon-treated
saline-filled polyethylene tubing (PE-200) and connected with
a 6 cm section of silicon-treated tubing (PE-240) to form an
extracorporeal arteriovenous shunt. The shunt patency was
monitored using a Doppler flow system and flowprobe
(2.4-2.8 mm) placed proximal to the locus of the shunt. All
parameters were monitored continuously using an LDS
PoNeMah Life Science Suite data acquisition system (LDS
Test and Measurement, Middleton, WI). On completion of a
15 min postsurgical stabilization period, an occlusive throm-
bus was formed by the introduction of a thrombogenic surface
(0 braided silk thread, 6 cm in length, Ethicon Inc., Somerville,
NJ) into the shunt. Seven 15 min shunt periods were employed
with the first starting at 2 h postoral dosing and repeated every
30 min out to 5 h postdrug. At the end of each 15 min shunt
period, the silk was carefully removed and weighed. The
thrombus weight was calculated by subtracting the weight of
the silk (9 mg) prior to placement from the total wet weight of
the silk on removal from the shunt. Arterial blood was with-
drawn prior to the oral dose and at 2-5 h postdrug for
determination of aPTT. The template bleeding time was also
performed at 2-5 h postdrug by making an incision into the
gum (Surgicutt, ITC Corp., Edison, NJ), and the time to
clot formation was monitored. Coagulation parameters were
determined using plasma in an ACL-100 microsampler coagu-
lation analyzer. Plasma samples were measured for drug
plasma concentration.
were collected on a Bruker AXS Proteum 6000 detector. Dif-
fraction data were indexed, integrated, and scaled using
the Proteum Processing Program suite from Bruker AXS. The
crystal belongs to the C2 space group, with unit cell para-
meters a = 71.1 A, b = 72.0 A, c = 73.41 A, and β = 101.06.
The structure was determined by molecular replacement with
CNX32 using the PDB33 coordinates 1QBV as the search model.
All model building was done using the computer program
CHAIN;34 refinement and map calculations were carried out
using CNX.34 The final structure was refined to an Rfactor of
17.8 and Rfree of 22.8.
Platelet Aggregation. PRP concentrate (Biological Special-
ties, Inc.) was diluted with platelet-poor plasma (PPP) to yield a
final plate count per well of 300 000 platelets/μL. PPP was
obtained by spinning PRP at 3000 rpms for 10 min at 25 °C.
H-Gly-Pro-Arg-Pro-NH2 (GPRP-NH2, final concentration of
4 mM) was added to the PRP to minimize the effects of
fibrinogen. Aggregation was run in a 96-well assay plate with
a total volume of 150 μL (120 μL of PRP, 15 μL of inhibitor, and
15 μL of agonist) in each well.
Animal Studies. All procedures involving the use of animals
were performed in accordance with the Guide for the Care and
Use of Laboratory Animals (1996) and the Animal Care and
Use Committee, Johnson and Johnson Pharmaceutical Re-
search and Development.
˚
˚
˚
In Vitro Coagulation Assays. The plasma concentrations of 7
and 8 needed to produce a 2-fold prolongation of aPTT were
determined in human plasma. Fresh whole blood from
humans, dogs, and rats was centrifuged to obtain plasma, and
the thrombin inhibitors were added to yield concentrations
from 0.001 to 100 μM. After a short stabilization period, the
samples were placed in an automated coagulation analyzer
(Instrumentation Laboratory ACL100, Beckman Coulter, Miami,
FL) to determine clotting times in seconds.
Rat DVT Model. Male Sprague-Dawley Rats (300-450 g)
were anesthetized with a mixture of ketamine and xylazine
(1 mL/kg, IM). A midline incision was made in the neck to expose
and isolate the left carotid and right jugular. The left carotid artery
was cannulated with a 24 ga/0.75 in angiocath catheter for the
withdrawal of blood samples. The right jugular vein was cannu-
lated with saline-filled PE50 tubing for administration of the
compound or vehicle. The left femoral vein was isolated and
cannulated with saline-filled PE50 tubing for administration
of thromboplastin. For thrombus formation, a ventral midline
incision was made, and the descending vena cava was carefully
exposed and separated from the aorta. Just distal to the renal
veins, a suture was looped under the vena cava, and a second loop
was placed around the vena cava just proximal to the iliac veins.
The main iliolumbar and internal spermatic branches of the
vena cava were tied off. At this time, a thrombus was formed
by injecting 0.2 mL/kg thromboplastin (ThromboMAX with
calcium, Sigma Diagnostics T-9902) into the femoral vein over
10 s followed by a 10 s waiting period. Then, the proximal loopjust
below the renal veins was tied, and after a 15 min waiting period,
the distal loop was tied. At this time, the section of vena cava was
removedandcarefully cutopenfollowedbyremovalandweighing
of the thrombus.
In the intravenous studies, vehicle (solutol 10% and D5W) or
8 was infused over a 45 min time period at 0.04 mL/min. The
total cumulative doses of 8 studied were 0.1, 0.3, and 1.0 mg/kg,
iv. The procedure for thrombus formation detailed above was
begun at the 30 min point of the infusion. In the oral dose
studies, 8 at 10, 30, and 50 mg/kg was solubilized in 0.5%
methocel in a total volume of 5 mL/kg and administered by an
oral dosing tube. The rats were anesthetized 45 min after the oral
dose, and the above thrombus formation procedure was begun

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1855
References
Lynch, J. J., Jr.; Yan, Y.; Chen, Z.; Kuo, L.; Gardell, S. J.; Shafer, J.
A.; Vacca, J. P. Pharmacokinetic optimization of 3-amino-6-
chloropyrazinone acetamide thrombin inhibitors. Implementation
of P3 pyridine N-oxides to deliver an orally bioavailable series
containing P1 N-benzylamines. Bioorg. Med. Chem. Lett. 2003, 13,
1353–1357.
(1) (a) Lilienfeld, D. E.; Chan, E.; Ehland, J.; Godbold, J. H.;
Landrigan, P. J.; Marsh, G. Mortality from pulmonary embolism
in the United States: 1962 to 1984. Chest 1990, 98, 1067–1072.
(b) Lilienfeld, D. E. Decreasing mortality from pulmonary embolism in
the United States, 1979-1996. Int. J. Epidemiol. 2000, 29, 465–469.
(2) Melnikova, I. The anticoagulants market. Nat. Rev. Drug Discov-
ery 2009, 8, 353–354.
(3) Halperin, J. L. Ximelagatran: Oral direct thrombin inhibition as
anticoagulant therapy in atrial fibrillation. J. Am. Coll. Cardiol.
2005, 45, 1–9.
(4) (a) Borrissoff, J. I.; Spronk, H. M. H.; Heeneman, S.; tenCate, H. Is
thrombin a key player in the ‘coagulation-antherogenesis’ maze?
Cardiovasc. Res. 2009, 82, 392–403. (b) Tanaka, K. A.; Key, N. S.;
Levy, J. H. Blood coagulation: Hemostasis and thrombin regulation.
Anesth. Analg. 2009, 108, 1433–1446. (c) Santilli, F.; Dav, G.
Thrombin as a common downstream target blocking both platelet and
monocyte activation. Thromb. Haemostasis 2009, 101, 220–221.
(5) Sambrano, G. R.; Weiss, E. J.; Zheng, Y. W.; Huang, W.; Soughlin,
S. R. Role of thrombin signaling in platelets in haemostasis and
thrombosis. Nature 2001, 413, 74–78.
(6) Crowther, M. A.; Weitz, J. I. Ximelagatran: The first oral direct
thrombin inhibitor. Expert Opin. Invest. Drugs 2004, 13, 403–413.
(7) Patterson, C.; Stouffer, G. A.; Madamanchi, N.; Runge, M. S. New
tricks for old dogs: Nonthrombotic effects of thrombin in vessel
wall biology. Circ. Res. 2001, 88, 987–997.
(8) Fleitz, F. J.; Lyle, T. A.; Zheng, N.; Armstrong, J. D.; Volante, R.
P. Kilogram scale synthesis of the pyrazinone acetic acid core of an
orally efficacious thrombin inhibitor. Synth. Commun. 2000, 30,
3171–3180.
(9) (a) Chung, J. Y. L.; Cvetovick, R. J.; Tsay, F.-R.; Dormer, P. G.;
DiMichele, L.; Mathre, D. J.; Chilenski, J. R.; Mao, B.; Wenslow,
R. Synthesis of 3-aminopyrazinone mediated by 2-pyridylthioimi-
date-ZnCl2 complexes. Development of an efficient route to a
thrombin inhibitor. J. Org. Chem. 2003, 68, 8838–8846. (b) Burgey,
C. S.; Robinson, K. A.; Williams, P. D.; Coburn, C. A.; Lyle, T. A.;
Sanderson, P. E. PCT Int. Appl. WO 0075134, 2000. (c) CAUTION!
Solid bromopyrazinone is known to be thermally unstable with an
exothermic transition at 50 °C, is potentially shock sensitive, and
gradually decomposes at room temperature.
(16) (a) Riffel, K. A.; Hengchang, S.; Gu, X.; Yan, K.; Lo, M.-W.
Simultaneous determination of a novel thrombin inhibitor and its
two metabolites in human plasma by liquid chromatography/
tandem mass spectrometry. J. Pharm. Biomed. Anal. 2000, 23,
607–616. (b) Burgey, C. S.; Robinson, K. A.; Lyle, T. A.; Sanderson,
P. E. J.; Lewis, S. D.; Lucas, B. J.; Krueger, J. A.; Singh, R.; Miller-
Stein, C.; White, R. B.; Wong, B.; Lyle, E. A.; Williams, P. D.; Coburn,
C. A.; Dorsey, B. D.; Barrow, J. C.; Stranieri, M. T.; Holahan, M. A.;
Sitko, G. R.; Cook, J. J.; McMasters, D. R.; McDonough, C. M.;
Sanders, W. M.; Wallace, A. A.; Clayton, F. C.; Bohn, D.; Leonard, Y.
M.; Detwiler, T. J.; Lynch, J. J.; Yan, Y.; Chen, Z.; Kuo, L.; Gardell, S. J.;
Shafer, J. A.; Vacca, J. P. Metabolism-directed optimization of 3-
aminopyrazinone acetamide thrombin inhibitors. Development of an
orally bioavailable series containing P1 and P3 pyridines. J. Med.
Chem. 2003, 46, 461–473.
(17) Young, M. B.; Barrow, J. C.; Glass, K. L.; Lundell, G. F.; Newton,
C. L.; Pellicore, J. M.; Rittle, K. E.; Selnick, H. G.; Stauffer, K. J.;
Vacca, J. P.; Williams, P. D.; Bohn, D.; Clayton, F. C.; Cook, J. J.;
Krueger, J. A.; Kuo, L. C.; Lewis, S. D.; Lucas, B. J.; McMasters,
D. R.; Miller-Stein, C.; Pietrak, B. L.; Wallace, A. A.; White, R. B.;
Wong, B.; Yan, Y.; Nantermet, P. G. Discovery and evaluation of
potent P1 aryl hetercycle-based thrombin inhibitors. J. Med. Chem.
2004, 47, 2995–3008.
(18) Lu, T.; Markotan, T.; Coppo, F.; Tomczuk, B.; Crysler, C.;
Eisennagel, S.; Spurlino, J.; Gremminger, L.; Soll, R. M.; Giardino,
E. C.; Bone, R. Oxyguanidines. Part 2: Discovery of a novel orally
active thrombin inhibitor through structure-based drug design and
parallel synthesis. Bioorg. Med. Chem. Lett. 2004, 14, 3727–3731.
(19) (a) Walkowiak, J.; Witmanowski, H.; Strzykala, K.; Bychowiec, B.;
Songin, T.; Borski, K.; Herzig, K.-H. Inhibition of endogenous
pancreatic enzyme secretion by oral pancreatic enzyme treatment.
Eur. J. Clin. Invest. 2003, 33, 65–69. (b) Obourn, J. D.; Frame, S. R.;
Shiu, T.; Solomon, T. E.; Cook, J. C. Evidence that A8947 enhances
pancreas growth via a trypsin inhibitor mechanism. Toxicol. Appl.
Pharmacol. 1997, 146, 116–126.
(20) Chen, Z.; Li, Y.; Mulichak, A. M.; Lewis, S. D.; Shafer, J. A.
(10) (a) Costanzo, M. J.; Maryanoff, B. E.; Hecker, L. R.; Schott, M. R.;
Yabut, S. C.; Zhang, H.-C.; Andrade-Gordon, P.; Kaufmann, J.
A.; Lewis, J. M.; Krishnan, R.; Tulinsky, A. Potent thrombin
inhibitors that probe the S1’ subsite: Tripeptide transition state
analogues based on a heterocycle-activated carbonyl group. J.
Med. Chem. 1996, 39, 3039–3043. (b) Costanzo, M. J.; Almond, H.
R., Jr.; Hecker, L. R.; Schott, M. R.; Yabut, S. C.; Zhang, H.-C.;
Andrade-Gordon, P.; Corcoran, T. W.; Giardino, E. C.; Kaufmann, J. A.;
Lewis, J. M.; de Garavilla, L.; Haertlein, B. J.; Maryanoff, B. E. In-
depth study of tripeptide-based R-ketoheterocycles as inhibitors of
thrombin. Effective utilization of the S10 subsite and its implications
to structure-based drug design. J. Med. Chem. 2005, 48, 1984–2008.
(11) Brown, F. J.; Andisik, D. W.; Bernstein, P. R.; Bryant, C. B.;
Ceccarelli, C.; Damewood, J. R., Jr.; Edwards, P. D.; Ealey, R. A.;
Feeney, S.; Green, R. C.; Gomes, B.; Kosmider, B. J.; Krell, R. D.;
Shaw, A.; Steelman, G. B.; Thomas, R. M.; Vacek, E. P.; Veale, C.
A.; Tuthill, P. A.; Warner, P.; Williams, J. C.; Wolanin, D. J.;
Woolson, S. A. Design of orally active, non-peptidic inhibitors of
human leukocyte elastase. J. Med. Chem. 1994, 37, 1259–1261.
(12) Sanderson, P. E. J.; Lyle, T. A.; Cutrona, K. J.; Dyer, D. L.;
Dorsey, B. D.; McDonough, C. M.; Naylor-Olsen, A. M.; Chen,
I.-W.; Chen, Z.; Cook, J. J.; Cooper, C. M.; Gardell, S. J.; Hare,
T. R.; Krueger, J. A.; Lewis, S. D.; Lin, J. H.; Lucas, B. J., Jr.; Lyle,
E. A.; Lynch, J. J., Jr.; Stranieri, M. T.; Vastag, K.; Yan, Y.; Shafer,
J. A.; Vacca, J. P. Efficacious, orally bioavailable thrombin inhibitors
based on 3-aminopyridone or 3-aminopyrazinone acetamide pep-
tidomimetic templates. J. Med. Chem. 1998, 41, 4466–4474.
(13) Isaacs, R. C. A.; Solinsky, M. G.; Cutrona, K. J.; Newton, C. L.;
Naylor-Olsen, A. M.; Krueger, J. A.; Lewis, S. D.; Lucas, B. J.
Structure-based design of novel groups for use in the P1 position of
thrombin inhibitor scaffolds. Part 1: Weakly basic azoles. Bioorg.
Med. Chem. Lett. 2000, 16, 338–342.
Crystal structure of human R-thrombin complexed with hirugen
and p-amidinophenylpyruvate at 1.6 A resolution. Arch. Biochem.
˚
Biophys. 1995, 322, 198–203.
(21) Sanderson, P. E.; Stanton, M. G.; Dorsey, B. D.; Lyle, T. A.;
McDonough, C.; Sanders, W. M.; Savage, K. L.; Naylor-Olsen, A.
M.; Krueger, J. A.; Lewis, S. D.; Lucas, B. J.; Lynch, J. J.; Yan, Y.
Azaindoles: Moderately basic P1 groups for enhancing the selec-
tivity of thrombin inhibitors. Bioorg. Med. Chem. Lett. 2003, 13,
795–798.
(22) Isaacs, R. C. A.; Solinsky, M. G.; Cutrona, K. J.; Newton, C. L.;
Naylor-Olsen, A. M.; Krueger, J. A.; Lewis, S. D.; Lucas, B. J.
Structure-based design of novel groups for use in the P1 position of
thrombin inhibitor scaffolds. Part 1: Weakly basic azoles. Bioorg.
Med. Chem. Lett. 2006, 16, 338–342.
(23) Isaacs, R. C. A.; Solinsky, M. G.; Cutrona, K. J.; Newton, C. L.;
Naylor-Olsen, A. M.; McMasters, D. R.; Krueger, J. A.; Lewis, S.
D.; Lucas, B. J.; Kuo, L. C.; Yan, Y.; Lynch, J. J.; Lyle, E. A.
Structure-based design of novel groups for use in the P1 position of
thrombin inhibitor scaffolds. Part 2: N-acetamidoimidazoles.
Bioorg. Med. Chem. Lett. 2008, 18, 2062–2066.
(24) Reiner, J. E.; Siev, D. V.; Araldi, G.-L.; Cui, J. J.; Ho, J. Z.; Reddy,
K. M.; Mamedova, L.; Vu, P. H.; Lee, K-S. S.; Minami, N. K.;
Gibson, T. S.; Anderson, S. M.; Bradbury, A. E.; Nolan, T. G.;
Semple, J. E. Non-covalent thrombin inhibitors featuring P3-
heterocycles with P1-monocyclic arginine surrogates. Bioorg.
Med. Chem. Lett. 2002, 12, 1203–1208.
(25) (a) Schumacher, W. A.; Heran, C. H.; Steinbacher, T. E.; Megill, J.
R.; Bird, J. E.; Giancarli, M. R.; Durham, S. K. Thrombin
inhibition compared with other antithrombotic drugs in rats.
Thromb. Res. 1992, 68, 157–66. (b) Berry, C. N.; Girard, D.; Lochot,
S.; Lecoffre, C. Antithrombotic actions of argatroban in rat models of
venous, `mixed' and arterial thrombosis, and its effects on the tail
transaction bleeding time. Br. J. Pharmacol. 1994, 113, 1209–1214.
(c) Wong, P. C.; Crain, E. J.; Knabb, R. M.; Meade, R. P.; Quan, M. L.;
Watson, C. A.; Wexler, R. R.; Wright, M. R.; Slee, A. M. Nonpeptide
factor Xa inhibitors II. Antithrombotic evaluation in a rabbit model of
electrically induced carotid artery thrombosis. J. Pharmacol. Exp.
Ther. 2000, 295, 212–218.
(14) Tomczuk, B.; Lu, T.; Soll, R. M.; Fedde, C.; Wang, A.; Murphy,
L.; Crysler, C.; Dasgupta, M.; Eisennagel, S.; Spurlino, J.; Bone, R.
Oxyguanidines: Application to non-peptide phenyl-based throm-
bin inhibitors. Bioorg. Med. Chem. Lett. 2003, 13, 1495–1498.
(15) Burgey, C. S.; Robinson, K. A.; Lyle, T. A.; Nantermet, P. G.;
Selnick, H. G.; Isaacs, R. C. A.; Lewis, S. D.; Lucas, B. J.; Kueger,
J. A.; Singh, R.; Miller-Stein, C.; White, R. B.; Wong, B.; Lyle, E.
A.; Stranieri, M. T.; Cook, J. J.; McMasters, D. R.; Pellicore, J. M.;
Pal, S.; Wallace, A. A.; Clayton, F. C.; Bohn, D.; Welsh, D. C.;
(26) Giardino, E. C.; Costanzo, M. J.; Kauffman, J. A.; Li, Q.-S.;
Maryanoff, B. E.; Andrade-Gordon, P. Antithrombotic properties

1856 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Lu et al.
of RWJ-50353, a potent and novel thrombin inhibitor. Thromb.
Res. 2000, 98, 83–93.
inhibition (IC50) of an enzymatic reaction. Biochem. Pharmacol.
1973, 22, 3099–3108.
(27) Barrett, J. A.; Crocker, A. C.; Damphousse, D. J.; Heminway, S. J.;
Liu, S.; Edwards, D. S.; Lazewatsky, J. L.; Kagan, M.; Mazaika, T.
J.; Carroll, T. R. Biological evaluation of thrombus imaging agents
utilizing water soluble phosphines and tricine as coligands when
used to label a hydrazinonicotinamide-modified cyclic glycopro-
tein IIb/IIIa receptor antagonist with 99mTc. Bioconjugate Chem.
1997, 8, 155–160.
(28) Lorrain, J.; Millet, L.; Lechaire, I.; Lochot, S.; Ferrari, P.;
Visconte, C.; Sainte-Marie, M.; Lunven, C.; Berry, C. N.; Schaef-
fer, P.; Herbert, J.-M.; O’Connor, S. E. Antithrombotic properties
of SSR182289A, a new, orally active thrombin inhibitor. J. Phar-
macol. Exp. Ther. 2003, 304, 567–574.
(31) (a) Webber, P. C.; Lee, S.-L.; Lewandowski, F. A.; Schadt, M. C.;
Chang, C.-H.; Kettner, C. A. Kinetic and crystallographic studies of
thrombin with Ac-(D)Phe-Pro-boroArg-OH and its lysine, amidine,
homolysine and ornithine analogs. Biochemistry 1995, 34, 3750–
3757. (b) Skrzypczak-Jankun, E.; Carperos, V.; Ravichandran, K. G.;
Tulinsky, A.; Westbrook, M.; Maraganore, J. M. Structure of the hirugen
and hirulog 1 complexes of R-thrombin. J. Mol. Biol. 1991, 221, 1379–
1393.
(32) Brunger, A. T.; Kuriyan, J.; Karplus, M. Crystallographic R-factor
refinement by molecular dynamics. Science 1987, 235, 458–460.
(33) Bernstein, F. C.; Koetzle, T. F.; Meyer, E. F., Jr.; Brice, M. D.;
Rodgers, J. R.; Kennard, O.; Shimanouchi, T.; Tasumi, M. The
protein databank: A computer based archival file for macro-
molecular structures. J. Mol. Biol. 1977, 112, 535–542.
(34) Sack, J. S. CHAIN;A crystallographic modeling program.
J. Mol. Graphics 1988, 249, 224–225.
(29) Lu, T.; Tomczuk, B. T.; Markotan, T. P. U.S. Patent 6,204,
263 B1, 2001.
(30) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes 50%