Efficacious and orally bioavailable thrombin inhibitors based on a 2,5-thienylamidine at the P1 position: Discovery of N-carboxymethyl-D-diphenylalanyl-L-prolyl[(5-amidino-2- thienyl)methyllamide

Article abstract of DOI:10.1021/jm030025j

Thrombin, a crucial enzyme in the blood coagulation, has been a target for antithrombotic therapy. Orally active thrombin inhibitors would provide effective and safe prophylaxis for venous and arterial thrombosis. We conducted optimization of a highly efficacious benzamidine-based thrombin inhibitor LB30812 (3, Ki = 3 pM) to improve oral bioavailability. Of a variety of arylamidines investigated at the P1 position, 2,5-thienylamidine effectively replaced the benzamidine without compromising the thrombin inhibitory potency and oral absorption. The sulfamide and sulfonamide derivatization at the N-terminal position in general afforded highly potent thrombin inhibitors but with moderate oral absorption, while the well-absorbable N-carbamate derivatives exhibited limited metabolic stability in S9 fractions. The present work culminated in the discovery of the N-carboxymethyl- and 2,5-thienylamidine-containing compound 22 that exhibits the most favorable profiles of anticoagulant and antithrombotic activities as well as oral bioavilability (K_i = 15 pM; F = 43%, 42%, and 15% in rats, dogs, and monkeys, respectively). This compound on a gravimetric basis was shown to be more effective than a low molecular weight heparin, enoxaparin, in the venous thrombosis models of rat and rabbit. Compound 22 (LB30870) was therefore selected for further preclinical and clinical development.

Full text of DOI:10.1021/jm030025j

3612
J . Med. Chem. 2003, 46, 3612-3622
Effica ciou s a n d Or a lly Bioa va ila ble Th r om bin In h ibitor s Ba sed on a
2,5-Th ien yla m id in e a t th e P 1 P osition : Discover y of
N-Ca r boxym eth yl-D-d ip h en yla la n yl-L-p r olyl[(5-a m id in o-2-th ien yl)m eth yl]a m id e
Koo Lee,* Cheol Won Park, Won-Hyuk J ung, Hee Dong Park, Sun Hwa Lee, Kyung Ha Chung, Su Kyung Park,
O Hwan Kwon, Myunggyun Kang, Doo-Hee Park, Sang Koo Lee, Eunice E. Kim, Suk Kyoon Yoon, and Aeri Kim
LG Life Sciences Ltd./ R&D Park, P.O. Box 61 Yu-Sung, Science Town, Taejon 305-380, Korea
Received J anuary 15, 2003
Thrombin, a crucial enzyme in the blood coagulation, has been a target for antithrombotic
therapy. Orally active thrombin inhibitors would provide effective and safe prophylaxis for
venous and arterial thrombosis. We conducted optimization of a highly efficacious benzamidine-
based thrombin inhibitor LB30812 (3, Ki ) 3 pM) to improve oral bioavailability. Of a variety
of arylamidines investigated at the P1 position, 2,5-thienylamidine effectively replaced the
benzamidine without compromising the thrombin inhibitory potency and oral absorption. The
sulfamide and sulfonamide derivatization at the N-terminal position in general afforded highly
potent thrombin inhibitors but with moderate oral absorption, while the well-absorbable
N-carbamate derivatives exhibited limited metabolic stability in S9 fractions. The present work
culminated in the discovery of the N-carboxymethyl- and 2,5-thienylamidine-containing
compound 22 that exhibits the most favorable profiles of anticoagulant and antithrombotic
activities as well as oral bioavilability (Ki ) 15 pM; F ) 43%, 42%, and 15% in rats, dogs, and
monkeys, respectively). This compound on a gravimetric basis was shown to be more effective
than a low molecular weight heparin, enoxaparin, in the venous thrombosis models of rat and
rabbit. Compound 22 (LB30870) was therefore selected for further preclinical and clinical
development.
In tr od u ction
shorter duration of action, and thus, its further develop-
ment was limited. These clinical results, after all, led
us to seek orally active thrombin inhibitors with suf-
ficient efficacy that would clearly satisfy the criteria for
clinically useful anticoagulants.
Thrombin is a trypsin-like serine protease that plays
a central role in thrombosis by cleaving fibrinogen to
insoluble fibrin in the blood coagulation cascade. This
enzyme also potently activates platelet aggregation by
proteolytic activation of the thrombin receptor. Current
anticoagulant therapy is limited to indirect thrombin
inhibitors including heparin, low molecular heparins
(LMWH), and a vitamin K antagonist, warfarin. How-
ever, neither heparin nor LMWH can be administered
orally, and both heparin and warfarin require careful
monitoring for dose adjustments to prevent bleeding
complications. Accordingly, direct-acting and orally ac-
tive thrombin inhibitors would be effective and safe in
the treatment as well as prophylaxis of venous and
arterial thrombosis.1
Small-molecule inhibitors of thrombin have been of
considerable interest to medicinal chemists over the past
decade.2 Some years ago, we undertook a research
program aimed at the discovery of orally active small-
molecule thrombin inhibitors. Our earlier efforts focus-
ing on arylsulfonylphenylalanine amides (so-called ar-
gatroban type) led us to a highly potent, selective, and
orally active compound, 1 (LB30057, Ki ) 0.4 nM), in
the benzamidrazone P1 class,3,4 along with other com-
pounds in the benzylamine series derived from the same
template (e.g., 2) that are also highly orally bioavailable
in rats.5 However, the phase I clinical study of 1
revealed less than desired anticoagulant activity and
Our continuing research efforts have been devoted to
a class of D-Phe-Pro-agmatine-type6 tripeptides, given
that noncovalent inhibitors lacking an electrophilic
serine trap are fast-binding and more efficacious and
thus have a better therapeutic index than covalent
inhibitors.7,8 One strategic approach in this work was
to judiciously combine the intrinsically potent D-diphen-
yl-Ala-Pro dipeptide9,10 with our previously defined P1
elements such as benzamidrazone, benzylamine, and
* To whom correspondence should be addressed. Phone: +82-42-
866-2258. Fax: +82-42-861-2566. E-mail: kleec@lgls.co.kr.
10.1021/jm030025j CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/12/2003

Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3613
Ta ble 1. Thrombin Inhibitory Activity and Pharmacokinetic Parameters for Compounds 3-25
plasma concnb (n ) 3)
aromatic
group
thrombina
Ki (nM)
compd
R
Cmax (µg/mL)
AUC (µg‚min/mL)
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
H2NS(O)2
H2NS(O)2
H2NS(O)2
H2NS(O)2
H2NS(O)2
H2NS(O)2
H2NS(O)2
H2NS(O)2
H2NS(O)2
H2NS(O)2
MeS(O)2
a
b
c
d
e
f
g
h
i
0.003
0.008
0.010
0.015
0.004
0.02
0.16
0.43
0.02
0.12
0.003
0.005
0.03
0.017
0.015
0.17
0.60
1.4
0.013
0.015
0.10
2.8
2.4
1.1
1.6
2.2
2.0
0.5
1.0
NTc
NTc
2.8
1.3
4.0
272
206
89
246
282
254
136
NTc
NTc
263
85
594
1564
426
850
1297
1113
120
1390
212
j
a
e
a
d
e
f
g
i
a
e
g
j
MeS(O)2
MeOC(O)
MeOC(O)
MeOC(O)
MeOC(O)
MeOC(O)
MeOC(O)
HO2CCH2
HO2CCH2
HO2CCH2
HO2CCH2
HO2CCH2CH2
14
4.1
4.6
8.6
7.1
1.3
9.5
4.6
11.1
4.9
0.64
0.010
4.0
1126
503
e
argatroban
a
b
Human thrombin. Concentration after oral dosing at 30 mg/kg in rats as a water solution or as a 20-30% cosolvent (PEG400/
EtOH/Tween20 ) 85:10:5) solution in water. c NT: not tested.
benzamidine as a means to improve oral bioavailability
(pKa of ∼8.9,5 ∼8.8,11 and ∼10.2,12 respectively). These
P1 functionalities are substantially less basic than the
agmatine but are still capable of strong binding in the
S1 specificity pocket of thrombin via ionic interaction
with Asp189. To this end, incorporation of the benza-
midine P1 led to the identification of a series of
extremely potent and orally absorbable thrombin inhibi-
tors with excellent antithrombotic activity as exempli-
fied by the sulfamide compound 3 (LB30812, Ki ) 3
pM).13 However, because of limited bioavailability (e.g.,
∼10% in rats for 3) and a short period of duration after
oral administration, these compounds were unsuitable
for further development. In this paper, we describe
potent thrombin inhibitors with improved oral phar-
macokinetic profile that resulted from further optimiza-
tion studies with the amidine P1-based noncovalent
tripeptides.
The study focused herein is the investigation of a
variety of arylamidines as replacements for the benza-
midine at P1, and the structure-activity and structure-
absorption relationships study began with the sulfamide
lead 3. On the basis of our previous experience that a
fluorine substituent on the P1 phenylene ring of the
amidrazone compound 1 has an effect on enhancing oral
absorption,14 fluorophenylene P1 linkers were examined
for the amidine class of compounds. While both com-
pounds 4 and 5 maintained thrombin inhibitory potency,
there was no improvement in absorption in either case.
The pyridine heterocyle as in compound 6 was beneficial
to the aqueous solubility but did not improve the oral
absorption. For the analogue 7 possessing 2,5-thiophene,
both potency and oral absorption were found to be
substantially comparable to those of 3, consistent with
the SAR for the sulfonate derivatives 13 and 14.15 On
the other hand, incorporation of its regioisomeric
thiophene nuclei as in the analogues 8 and 9 was
detrimental to the potency. A potency decrease also
resulted upon employing the furan and thiazole replace-
ments as in compounds 10-12. A relatively dramatic
potency drop for the 2,5-furan analogue 10 can be
rationalized by molecular modeling, which suggests that
the highly bent bond angle of side chains on the furan
ring (∼135°) causes unfavorable interaction of the
Resu lts a n d Discu ssion
As shown in Table 1, the in vitro activity of com-
pounds was expressed as the inhibition constant Ki. The
oral absorption behavior of compounds in this study was
determined primarily by the peak plasma concentration
(Cmax) and the area under the curve (AUC) measured
after oral administration in rats.

3614 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Lee et al.
Ta ble 2. Pharmacokinetic Parameters and Antithrombotic Effect of Selected Compounds in Rats (n ) 3-4)
a
a
a
a
b
c
Cl
Vdss
t1/2
AUC
F
DVT
ACT
compd
(mL/min/kg)
(mL/kg)
(min)
(µg‚min/mL)
(%)
(rat, %)
(rat, n-fold)
3
7
8.6
11.6
3.0
2.1
5.4
1.7
9.6
6.0
297
310
174
139
152
230
258
189
45
34
69
81
53
45
47
50
1181
864
8
11
6
11
8
7
43
21
100
100
100
100
93
11
5.0
6.8
6.0
5.2
3.7
12
15
16
17
19
22
24
3407
4823
1870
6139
1059
1746
98
100
NT
82
NT
1.9
argatroban
a
Parameter obtained after iv application at 10 mg/kg. b Inhibition of thrombus formation in thromboplastin-induced deep vein thrombosis
(DVT) rat model after iv bolus dosing at 1 mg/kg. c Ex vivo activated clotting time (ACT) over the control (ca. 70 s).
Ta ble 3. In Vitro Metabolic Stability of Selected Compounds
amidine group with Asp189 in the S1 specificity pocket
of thrombin. On the other hand, the 2,5-thienylamidine
P1 (bond angle of ∼158°) is well tolerated in the
specificity pocket with the amidine group still forming
a bidentate salt-bridge interaction with Asp189 (vide
supra).
recoverya (%)
monkey
compd
rat
dog
human
3
7
90
84
80
60
49
87
89
93
96
65
91
83
95
99
87
81
39
<10
<10
80
92
86
86
38
13
97
95
15
17
19
21
22
Our next P1 variation study was conducted with the
methyl carbamate and carboxymethyl derivatives, 15
and 21. Compound 15 in our previous paper was
superior to the sulfamide 3 in oral absorption,13 and the
carboxymethyl functionality,16,17 as exemplified by mel-
agatran,18 has been widely utilized in the thrombin
inhibitor area as a means of improving aqueous solubil-
ity as well as oral bioavailability, although this element
in the case of the benzamidine 21 was not advantageous
for oral absorption. As the P1 aryl was varied, a potency
trend in the carboxymethyl series of analogues 21-24
was very similar to that observed for the sulfamide
series while a slightly different trend was seen for the
carbamate series in which 15-17 are all similarly
potent. When evaluated for oral absorption in rats, the
methyl carbamate series displayed a better profile than
the sulfamide series. In particular, compounds 16 and
19 in this series displayed 3- to 5-fold higher Cmax and
AUC values compared to 3. In the series of carboxy-
methyl derivatives, good oral absorption behavior was
displayed by compounds 22 and 24; this is in contrast
to that shown by compounds 21 and 23 despite that they
are all very similar in structure.19 These results suggest
that these hydrophilic compounds are presumably ab-
sorbed by an active transport mechanism. Most notable
in this series is compound 22, which has an excellent
profile of thrombin inhibition and oral absorption as
well. Extension of the carboxymethyl chain as in 25
maintained potency compared to 22. These results are
consistent with those of X-ray crystallography of 22,
where the carboxylic acid group is directed toward the
solvent, making no interactions with the thrombin
active site (vide supra).
83
a
Recovery of compound after 30 min incubation in liver S9 at
the initial concentration of 20 µM.
thrombosis model at an intravenous dose of 1 mg/kg
(Table 2). Like compound 3, compounds 7, 15, 16, and
22 completely inhibited thrombus formation, reflecting
their strong in vitro potency, whereas argatroban used
as a reference was 82% effective and our earlier
compound 1 (LB30057) was only 50% effective in this
model. Activated clotting time (ACT) was also measured
ex vivo from the rat plasma of the same animal setting
to determine potential anticoagulant activity. Consistent
with the results of antithrombotic activity, the ACT for
each compound shown in Table 2 was highly increased
and at least a 5-fold ACT over the control (ca. 70 s) was
required for complete inhibition of thrombosis. For this
series of thrombin inhibitors, it appears that the in vitro
potency was well translated into ACT anticoagulant
activity. Inhibitors with relatively lower Ki potency such
as 10 and 20 exhibited a less than 2-fold ACT prolonga-
tion over the control (data not shown) despite them
being more potent than argatroban.
Compounds were evaluated for their metabolic stabil-
ity using S9 fractions from rat, dog, monkey, and human
livers (Table 3). In this assay, the sulfamoyl and
carboxymethyl derivatives were generally quite stable
in all species, but the methyl carbamate analogues were
relatively unstable, particularly in the monkey. Conse-
quently, such poor metabolic stability, together with low
oral bioavailability, precluded further characterization
of the carbamate analogues.
A set of key compounds was selected to determine
absolute oral bioavailability and antithrombotic activity
in rats (Table 2). Upon intravenous dosing, the carbam-
ate derivatives 15-19 revealed relatively low clearance
and moderate oral bioavailability despite their excellent
Cmax and AUC values observed after oral dosing. The
carboxymethyl derivatives 22 and 24 displayed a similar
level of clearance but superior oral bioavailability (43%
and 21%, respectively), while the sulfamide derivatives
3 and 7 exhibited modest bioavailability. The anti-
thrombotic efficacy for primary in vivo screening was
determined using the thromboplastin-induced rat venous
The selected compounds were evaluated in vitro for
their ability to prolong activated partial thromboplastin
time (APTT) in the rat and human plasma. Compared
to argatroban, compounds 3, 7, and 22 exhibited excel-
lent anticoagulant activity, doubling APTT at 0.14, 0.19,
and 0.23 µM, respectively, in human plasma (Table 4).
The superiority of these compounds was further dem-
onstrated by their strong ability for ex vivo ACT
prolongation in rats. Compound 22 doubled ACT at 0.85
µM. Hence, despite the 5-fold lower Ki potency in
inherent thrombin inhibition, the anticoagulant efficacy

Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3615
Ta ble 4. In Vitro and Ex Vivo Anticoagulant Activity of
pound 22 was well absorbed in dogs after oral dosing
at 5 mg/kg (at 5 mg/kg, Cmax ) 3.8 µg/mL, F ) 42%, t1/2
) 68 min). In cynomolgus monkeys, 22 was also orally
bioavailable (at 20 mg/kg, F ) 15%, t1/2 ) 58 min).
Finally, 22 was compared with the most commonly used
therapeutic anticoagulant LMWH in the rat and rabbit
venous thrombosis models. Upon intravenous infusion
at four different doses in the rats, compound 22 inhib-
ited thrombus formation in a dose-dependent manner,
showing complete inhibition at a bolus dose of 0.2 mg/
kg with an intravenous infusion dose of 8 µg/(kg‚min),
which compares favorably to a dose of 2.5 mg/kg + 100
µg/(kg‚min) for enoxaparin with an average MW of 5000
(Figure 1). In this study, the APTT of 22 was increased
by approximately 1.7-, 2.8-, 3.6-, and 5.6-fold over the
control. Consistent with these results, oral administra-
tion of this compound in the same rat model caused
dose-dependent inhibition of thrombosis with an ID50
of 10.3 mg/kg and prolonged APTT by approximately
1.2-, 1.2-, 1.8- and 2.4-fold over the control, respectively
(Figure 2). In a rabbit model, thrombosis was induced
by insertion of silk threads saturated with thrombo-
plastin into a vena cava. Compound 22 significantly
inhibited rabbit venous thrombosis at a dose of 12.5 µg/
kg + 0.5 µg/(kg‚min), which caused a 1.5-fold increase
in APTT (Figure 3). On the other hand, enoxaparin, a
positive reference used in comparison, required a much
higher dose on a gravimetric basis for a similar effect
of thrombosis inhibition in rabbits (300 µg/kg + 5 µg/
(kg‚min)).
Selected Compoundsa
anticoagulant activity (µM)
compd
2 × APTT (human) 2 × APTT (rat) 2 × ACT (rat)
3
7
17
19
22
0.14
0.19
0.34
0.37
0.23
0.57
1.24
1.50
2.85
5.20
0.97
2.50
1.23
2.09
3.75
6.86
0.85
1.97
argatroban
a
At concentration of inhibitor required to double APTT and
ACT.
Ta ble 5. Selectivity Profile for Compound 22
enzymea
Ki (nM)
selectivity (n-fold)
thrombin
factor Xa
trypsin
trypsinb
plasmin
t-PA
0.015
2250
0.30
3.0
7400
433
1
>10000
20
200
>10000
>10000
a
b
Human. Bovine.
of 22 compares favorably to that of 3, which is possibly
attributed to their difference in the plasma protein
binding. Compound 22 exhibited 82% and 49% protein
binding in the rat and human plasma, respectively,
without significant variation in a concentration range
of 1-30 µg/mL. On the other hand, compound 3 was
95% protein-bound in both species.
Because of its superior pharmacokinetic and phar-
macological properties, compound 22 was selected for
further in vitro and in vivo characterization. In addition
to its excellent thrombin inhibitory activity, this com-
pound displayed good selectivity toward other serine
proteases important to hemostasis and thrombolysis
(Table 5). Pharmacokinetic parameters for compound
22 in dogs and monkeys are shown in Table 6. Com-
The crystal structure of human thrombin complexed
with 22 has been obtained at 1.8 Å (Figure 4).20 The
inhibitor is bound in the active site in a fashion very
similar to that of the previously reported compounds.9,10a
As expected, the diphenyl and proline rings span the
distal and proximal hydrophobic pockets, respectively.
Ta ble 6. Pharmacokinetic Parameters for Compound 22
dose
Cl
Vdss
t1/2
Cmax
Tmax
AUC
F
species
dog
(mg/kg)
n
route
(mL/(min‚kg))
(mL/kg)
(min)
(µg/mL)
(min)
(µg‚min/mL)
(%)
5
2.5
20
2
3
3
2
2
po
iv
po
iv
68
54
58
23
3.8
30
90
363
457
309
205
42
15
5.9
9.8
223
230
monkey
2.1
F igu r e 1. Dose-dependent antithrombotic effect of 22 (LB30870) and LMWH after iv administration in the rat venous thrombosis
model: bolus (mg/kg)/infusion (µg/(kg‚min)); n ) 10 for the control group; n ) 6 for each inhibitor treatment group; (//) p < 0.01
vs control; ANOVA, followed by Dunnett’s multiple comparison test.

3616 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Lee et al.
preceded by catalytic hydrogenation of the amidoxime
intermediates, 28 and 29, under acetylation condi-
tions.21
The synthesis of the requisite P1 precursors, CN-
Ar-CH2NH2 30b-j, is also outlined in Scheme 1. As
previously described,13 compounds 30b-e were typically
prepared from the commercially available materials of
general structure CH3-Ar-Br in several steps involving
CuCN-mediated cyanation, radical bromination with
N-bromosuccinimide and a Gabriel-type amination22
with di-tert-butyl iminodicarboxylate. Synthesis of com-
pounds 30f-i began with aldehyde reduction of OHC-
Ar-Br (iodoaldehyde23 for 30g), and subsequent PPh3/
CBr4 bromination followed by cyanation afforded the
common intermediate BrCH2-Ar-CN. Unlike this path-
way, the thiazolecarbonitrile 30j was prepared from the
amide precursor. This thiazole precursor was readily
prepared by conversion of N-Boc-glycinamide into a
thioamide and subsequent cyclocondensation with 3-bro-
mo-2-oxopropionic acid, followed by amidation with
aqueous ammonia as previously described by J ung.24
For the practical preparation of the 2,5-thiophene 30e,
which is a precursor of 22, the bromide 31 was trans-
formed to an azide followed by iminophosphorylation
and subsequent hydrolysis.
F igu r e 2. Dose-dependent antithrombotic effect of 22
(LB30870) after oral administration in the rat venous throm-
bosis model: n ) 6 for the control and inhibitor treatment
groups; (//) p < 0.01 vs control; ANOVA, followed by Dunnett’s
multiple comparison test.
As discussed above, the thienylamidine P1 chain binds
in the S1 specificity pocket with the amidine making
the bidentate salt bridge interaction with Asp189. The
amidine is also hydrogen-bonded to Gly219 and a water
molecule. On the other hand, the carboxylic acid moiety
is solvent-exposed without interacting with the throm-
bin active site.
Con clu sion
We have conducted optimization of a highly efficacious
noncovalent tripeptidic inhibitor of thrombin,
3
Ch em istr y
(LB30812), focusing on improving oral bioavailability.
Of a variety of aryl replacements explored at the
amidine P1 position, the 2,5-thienylamidine could ef-
fectively replace the benzamidine without compromising
thrombin inhibitory potency and oral absorption behav-
ior. The sulfamide and sulfonamide derivatization at the
N-terminal position, in general, afforded highly potent
thrombin inhibitors with good metabolic stability while
displaying a moderate level of Cmax and AUC values
upon oral administration to rats. The N-carbamate
derivatives exhibited superior absorption profile but at
the expense of metabolic stability in S9 fractions. The
N-carboxymethyl derivative 22 demonstrated the best
The target compounds were prepared as exemplified
in Scheme 1. The N-protected dipeptide 26 was coupled
to the P1 precursors 30a -i, and the N-Boc group was
removed. The free amines 27 were reacted with R-Cl
(sulfamoyl chloride, sulfonyl chloride, and chlorofor-
mate) in the presence of base. The resulting intermedi-
ates were then converted to the amidines 3-20 by a
modified Pinner reaction that proceeds via a thioimidate
intermediate. Compounds 21-25 were prepared from
27 by N-alkylation with alkyl bromides, nitrile to
amidine conversion, and hydrolysis of the ester groups.
In this case, the amidine formation was more efficiently
F igu r e 3. Dose-dependent antithrombotic effect of 22 (LB30870) and LMWH after iv administration in the rabbit venous
thrombosis model: bolus (µg/kg)/infusion (µg/(kg‚min)); n ) 11 for the control group; n ) 4-8 for each inhibitor treatment group;
(//) p < 0.01 vs control; ANOVA, followed by Dunnett’s multiple comparison test.

Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3617
performed by LG Chem. Analysis Center, Korea, and were
within 0.4% of the calculated values.
5-(Am in om eth yl)th iop h en e-2-ca r bon itr ile Hyd r och lo-
r id e (30e). A mixture of 5-methylthiophene-2-carbonitrile (9.9
g, 80.5 mmol), benzoyl peroxide (0.23 g, 0.95 mmol), and
N-bromosuccinimide (15 g, 84.3 mmol) in carbon tetrachloride
(200 mL) was heated at reflux for 6 h. The resulting suspension
was filtered, and the filtrate was diluted with dichloromethane
(400 mL), washed with a saturated sodium bicarbonate solu-
tion, dried over magnesium sulfate, and concentrated in vacuo.
The residue was purified by column chromatography (EtOAc/
n-hexane, 1:4) to give 5-(bromomethyl)thiophene-2-carboni-
trile17 as a yellow oil (14.0 g, 86%).
To a cold solution of 5-(bromomethyl)thiophene-2-carboni-
trile (4.2 g, 20.8 mmol) in THF (500 mL) was added sodium
hydride (60% dispersion in oil, 1.0 g, 25 mmol) in portions. To
this suspension was added in portions di-tert-butyl iminodi-
carboxylate (4.9 g, 22.9 mmol, purchased from TCI, J apan).
After being stirred for 3 h, the resulting solution was diluted
with ethyl acetate (400 mL), washed with water, dried over
magnesium sulfate, and concentrated in vacuo. The residue
was purified by column chromatography (EtOAc/n-hexane, 1:4)
to give 5-(N,N-Boc2-aminomethyl)thiophene-2-carbonitrile as
a yellow foam. This solid was dissolved in ethyl acetate (150
mL) and cooled to 0 °C. HCl gas was bubbled through the
solution for 10 min, and the mixture was allowed to warm to
room temperature. The solvent was removed in vacuo to give
the title compound as a pale-yellow solid (2.4 g, 86%). 1H NMR
(CD3OD): δ 7.48 (d, 1H), 7.20 (d, 1H), 4.21 (s, 2H). FAB MS,
m/z: 175 [M + 1]+.
Alter n a tive P r ep a r a tion of 30e. To a solution of 5-(bro-
momethyl)thiophene-2-carbonitrile (14.4 g, 71.3 mmol) in
benzene (71 mL) was added tetrabutylammonium chloride
(0.46 g, 1.43 mmol) and sodium azide (5.57 g, 85.6 mmol), and
the mixture was stirred for 1 h at room temperature. The
resulting solution was washed with water (20 mL × 2) and
cooled to 15 °C. Triphenylphosphine was added portionwise
for 1 h, and the mixture was stirred for 1.5 h. After the reaction
was completed, 12.8 mL of water was added and the solution
was heated at 70 °C for 3 h and extracted with 2 N HCl (30
mL × 2). The aqueous layer was washed with dichloromethane
(30 mL), basified with NaOH (4.6 g), and extracted with
dichloromethane (30 mL × 2). The combined extracts were
concentrated, the residue was dissolved in diethyl ether (70
mL), and 3 N HCl in ethyl acetate (30 mL) was added dropwise
keeping the temperature below 25 °C. The white precipitates
were collected by filtration and the filter cake was washed with
diethyl ether and ethyl acetate to give 30e as a white solid
(10.9 g, 88%).
D-Dip h en yla la n yl-L-p r olyl-[(5-cya n o-2-th ien yl)m eth yl]-
a m id e Hyd r och lor id e (27e). To a solution of N-Boc-D-
diphenylalanyl-L-proline (2 g, 4.8 mmol) in DMF (20 mL) was
added 5-(aminomethyl)thiophene-2-carbonitrile hydrochloride
(30e) (1 g, 5.7 mmol), EDC (1.4 g, 7.2 mmol), HOBT (0.8 g, 6.2
mmol), and triethylamine (2 mL, 14.4 mmol), and the mixture
was stirred overnight at room temperature. The solvent was
removed in vacuo, and the residue was dissolved in EtOAc and
washed sequentially with a saturated sodium bicarbonate
solution, 1 N HCl, and brine. After the solvent was dried over
magnesium sulfate and concentrated in vacuo, the residue was
purified by column chromatography (EtOAc/n-hexane, 2:1) to
give N-Boc-D-diphenylalanyl-L-prolyl[(5-cyano-2-thienyl)methyl]-
amide (2.3 g, 86%). 1H NMR (CDCl3): δ 7.83 (s, 1H), 7.55-
7.10 (m, 11H), 6.92 (d, 1H), 4.98 (d, 1H), 4.80 (m, 1H), 4.53
(m, 2H), 4.37 (m, 2H), 3.68 (m, 1H), 2.54 (q, 1H), 2.09 (m, 1H),
1.66 (m, 1H), 1.45 (m, 1H), 1.37 (m, 1H), 1.31 (s, 9H). FAB
MS, m/z: 558 [M + 1]+.
F igu r e 4. Compound 22 is complexed with human thrombin
as determined by X-ray crystallography. (a) The final electron
density (2Fo - Fc) around the P1 pocket is contoured at the
1.2σ level. Hydrogen bonds are indicated as dashed lines.
There is a tightly bound water molecule that serves as a
bridge. (b) 22 is bound to thrombin in a surface representation.
The diphenyl moiety fits very nicely near the P4 site. Oxygen
atoms are shown in red, nitrogen atoms are shown in blue,
sulfur is shown in yellow and carbons are shown in either
green or dark-gray.
overall profile in this series, which is potent, selective,
metabolically stable, and orally bioavailable. This com-
pound also displayed excellent anticoagulant and anti-
thrombotic activity in venous thrombosis models of rat
and rabbit. Compound 22 (LB30870) was therefore
selected for further preclinical and clinical development.
Exp er im en ta l Section
Ch em istr y. The solvents employed in these experiments
were from commercial grade and used without further puri-
fication. Proton NMR spectra were recorded on a J EOL 500
spectrometer. Chemical shifts are reported in δ (ppm) relative
to tetramethylsilane. Column chromatography was carried out
with Merck grade 60 silica gel (230-400 mesh). Purification
of the final products was achieved by reverse-phase prepara-
tive HPLC on a Delta-Pak C18 column, 100 Å pore size, with
trifluoroacetic acid (0.1%)-H2O-MeOH solvent systems using
various linear gradients. The pure fractions were lyophilized
to give a white solid. FABMS and HR FABMS were obtained
on a J EOL DX300 mass spectrometer. Microanalyses were
To a cooled solution of this compound (2 g, 3.7 mmol) in
EtOAc (20 mL) was added 4 N HCl in EtOAc (5 mL). After
being stirred for 3 h at room temperature, the solution was
removed to dryness to give the title compound as a white solid
(2.71 g, 100%). 1H NMR (CDCl3): δ 8.52 (m, 1H), 7.55-7.12
(m, 11H), 6.84 (d, 1H), 4.77 (d, 1H), 4.57 (d, 1H), 4.50 (dd, 1H),
4.23 (d, 1H), 4.18 (dd, 1H), 3.77 (t, 1H), 2.43 (q, 1H), 1.94 (m,

3618 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Lee et al.
Sch em e 1a
a
(a) CN-Ar-CH2NH2 (30a -i), NMM, EDC, HOBT, DMF, quantitaive; (b) TFA, CH2Cl2 or 4 N HCl/EtOAc, EtOAc; (c) R-Cl
(methanesulfonyl chloride, sulfamoyl chloride, methyl chloroformate), Et3N, CH2Cl2; (d) (i) H2S, Py, Et3N, (ii) MeI, CH3CN, reflux, (iii)
NH4OAc, MeOH, reflux; (e) t-BuO2CCH2Br or EtO2CCH2CH2Br, DIPEA, NaI (cat.), CH3CN, room temp, 36 h; (f) NH2OH‚HCl, Na2CO3,
EtOH/H2O; (g) H2, 10% Pd-C, Ac2O, MeOH; (h) 6 N HCl, MeOH for 21-24; LiOH, H2O for 25; (i) CuCN, DMF, reflux; (j) NBS, benzoyl
peroxide, CCl4, reflux; (k) (Boc)2NH, NaH, THF; (l) n-BuLi, DMF, THF, -78 °C; (m) NaBH4, MeOH, room temp; (n) CBr4, PPh3, CH2Cl2;
(o) Lawesson’s Reagent, DME, room temp; (p) 3-bromo-2-oxopropionic acid, CaCO3, EtOH, room temp; (q) isobutyl chloroformate, NH4OH,
NMM, THF; (r) (CF3CO)2O, Et3N, CH2Cl2; (s) NaN3, Bu4NCl, benzene, room temp; (t) PPh3, H2O, benzene, reflux.
1H), 1.62 (m, 1H), 1.48 (m, 1H), 1.30 (m, 1H). FAB MS, m/z:
459 [M + 1]+. Anal. Calcd for C26H26N4O2S‚1.0C4H8O2‚
1.0HCl: C, 61.79; H, 6.05; N, 9.61. Found: C, 61.6; H, 5.9; N,
9.8.
hexane, 3:1) to give N-aminosulfonyl-D-diphenylalanyl-L-prolyl-
[(5-cyano-2-thienyl)methyl]amide (3.1 g, 88%). 1H NMR
(CDCl3): δ 7.77 (d, 1H), 7.51-7.17 (m, 10H), 7.15 (d, 1H), 4.98
(d, 1H), 4.62 (m, 1H), 4.59 (m, 1H), 4.31 (m, 1H), 4.05 (m, 1H),
3.76 (m, 1H), 2.90 (m, 1H), 1.85 (m, 1H), 1.73 (m, 1H), 1.62
(m, 1H), 1.46 (m, 1H). MS, m/z: 538 [M + 1]+.
A solution of the coupling compound (0.2 g, 0.37 mmol) in
pyridine (2 mL) and Et3N (0.2 mL) was saturated with gaseous
H2S. After the mixture stood for 1 day, the solvent was
removed in vacuo to obtain the thioamide as a yellow solid.
To this material was added acetonitrile (2 mL) and io-
domethane (0.07 mL, 1.12 mmol), and the mixture was heated
at reflux for 1 h. After the solvent was evaporated in vacuo,
the resulting methyl thioimidate was dissolved in acetonitrile
(2 mL). To this solution was added ammonium acetate (0.09
g, 1.12 mmol) over 10 min, and the mixture was heated at
reflux for 1 h. The solution was cooled and concentrated and
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl[(5-am idin o-
2-th ien yl)m eth yl]a m id e Tr iflu or oa cetic Acid (7). To a
stirring solution of chlorosulfonyl isocyanate (6.3 g, 45 mmol)
in dichloromethane (25 mL) was added dropwise formic acid
(2.13 g, 45 mmol). The mixture was heated at reflux for 5 h
and cooled to obtain a 1.8 N solution of sulfamoyl chloride in
dichloromethane. To a cooled (0 °C) solution of 27e (3.2 g, 6.47
mmol) in dichloromethane (100 mL) was added the 1.8 N
sulfamoyl chloride solution (6 mL) and triethylamine (2.7 mL).
After the reaction was completed, the resulting solution was
diluted with dichloromethane (40 mL), washed with brine,
dried over magnesium sulfate, and concentrated in vacuo. The
residue was purified by column chromatography (EtOAc/n-

Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3619
the residue was purified by column chromatography using 10%
methanol in chloroform to give the title compound, which was
further purified by preparative HPLC (TFA (0.1%)-H2O-
vacuo, and the residue was purified by column chromatography
(EtOAc/n-hexane, 7:3) to give N-[2-(methoxycarbonyl)ethyl]-
D-diphenylalanyl-L-prolyl[(5-cyano-2-thienyl)methyl]amide (180
mg, 41%). This compound (160 mg) was hydrogenated in the
presence of palladium-on-carbon and acetic anhydride using
the procedure as described in the preparation of 22 to give
N-[2-(methoxycarbonyl)ethyl]-D-diphenylalanyl-L-prolyl[(5-ami-
dino-2-thienyl)methyl]amide diacetic acid 29 (91 mg, 54%). 1H
NMR (CD3OD): δ 7.74 (d, 1H), 7.35-7.11 (m, 10H), 6.98 (d,
1H), 4.72 (dd, 1H), 4.30 (m, 1H), 4.17 (s, 2H), 4.16 (m, 1H),
3.53 (s, 3H), 3.52 (m, 1H), 2.89 (m, 2H), 2.75 (m, 1H), 2.38 (m,
2H), 1.89 (m, 2H), 1.62 (m, 1H), 1.40 (m, 1H). FAB MS, m/z:
562 [M + 1]+
A mixture of the above product (91 mg, 0.16 mmol), 0.5 N
LiOH (10 mL), and water (3 mL) was stirred for 3 h at room
temperature. The reaction mixture was neutralized with 1 N
HCl and concentrated in vacuo. The residue was purified by
preparative HPLC (TFA (0.1%)-H2O-MeOH gradient) to give
the title compound (55 mg, 44%). 1H NMR (CD3OD): δ 7.81
(d, 1H), 7.62 (m, 2H), 7.50 (m, 2H), 7.42-7.32 (m, 6H), 7.21
(d, 1H), 5.12 (d, 1H), 4.61 (dd, 2H), 4.50 (d, 1H), 4.11 (m, 1H),
3.60 (m, 1H), 3.31 (m, 2H), 2.78 (m, 1H), 2.68 (m, 2H) 1.83 (m,
3H), 1.30 (m, 1H). FAB MS, m/z: 548 [M + 1]+. Anal. Calcd
for C29H33N5O4S‚2.0C2HF3O2‚0.3H2O: C, 50.7; H, 4.6; N, 8.9.
Found: C, 50.5; H, 4.5; N, 8.9.
1
MeOH gradient). H NMR (CD3OD): δ 8.21 (m, 1H), 7.77 (d,
1H), 7.51-7.25 (m, 11H), 4.98 (d, 1H), 4.62 (dd, 1H), 4.59 (dd,
1H), 4.31 (d, 1H), 4.05 (dd, 1H), 3.76 (m, 1H), 2.90 (q, 1H),
1.85 (m, 1H), 1.73 (m, 1H), 1.62 (m, 1H), 1.46 (m, 1H). FAB
MS, m/z: 555 [M + 1]+. Anal. Calcd for C26H30N6O4S2‚1.3C2-
HF3O2‚0.3H2O: C, 48.3; H, 4.5; N, 11.8. Found: C, 48.3; H,
4.6; N, 11.6.
N-(ter t-Bu t oxyca r b on yl)m et h yl-D-d ip h en yla la n yl-L-
p r olyl[(5-h yd r oxya m id in o-2-th ien yl)m eth yl]a m id e (28e).
To a cooled (0 °C) solution of 27e (3.03 g, 6.06 mmol) in
acetonitrile (60 mL) was added diisopropylethylamine (4.22
mL, 24.24 mmol) and tert-butyl bromoacetate (4.22 mL, 9.09
mmol), and the mixture was stirred for 2 days at room
temperature. After the reaction mixture was concentrated in
vacuo, the residue was purified by column chromatography
(EtOAc/n-hexane, 7:3) to give N-(tert-butoxycarbonyl)methyl-
D-diphenylalanyl-L-prolyl[(5-cyano-2-thienyl)methyl]amide (2.51
1
g, 72%). H NMR (CDCl3): δ 8.12 (t, 1H), 7.37 (m, 5H), 7.19
(m, 6H), 6.93 (d, 1H), 4.61 (dd, 1H), 4.49 (dd, 1H), 4.25 (m,
2H), 4.12 (dd, 1H), 3.24 (m, 3H), 2.67 (m, 1H), 2.07 (m, 1H),
1.67 (m, 1H), 1.43 (m, 1H), 1.37 (s, 9H), 1.25 (m, 1H). FAB
MS, m/z: 573 [M + 1]+.
In
a procedure similar to the synthesis of 7 and 22,
To a solution of the above product (2.31 g, 4.03 mmol) in a
4:1 mixture of ethanol and water (60 mL) was added hydroxy-
lamine hydrochloride (1.04 g, 14.91 mmol) and sodium carbon-
ate (726 mg, 6.85 mmol), and the mixture was heated at reflux
for 1 h. After the reaction mixture was concentrated in vacuo,
the residue was diluted with EtOAc, washed with brine, dried
over magnesium sulfate, and concentrated in vacuo to give the
title compound (2.36 g, 96%). 1H NMR (CD3OD): δ 8.13 (t, 1H),
7.42-7.35 (m, 4H), 7.27-7.11 (m, 7H), 7.01 (d, 1H), 6.83 (d,
1H), 4.87 (s, 2H), 4.56 (m, 1H), 4.44 (m, 1H), 4.27 (m, 3H),
3.32 (m, 1H), 3.25 (dd, 2H), 2.73 (m, 1H), 1.80-1.42 (m, 4H),
1.41 (s, 9H). FAB MS, m/z: 606 [M + 1]+. Anal. Calcd for
compounds 3-6, 8-21, 23, and 24 were prepared from the
corresponding amines, CN-Ar-CH2NH2.
N-Am in osu lfon yl-D-d ip h en yla la n yl-L-p r olyl-[(4-a m id i-
n op h en yl)m et h yl]a m id e Tr iflu or oa cet ic Acid (3). 1H
NMR (CD3OD): δ 7.73 (d, 2H), 7.55-7.18 (m, 12H), 4.91 (d,
1H), 4.45 (d, 1H), 4.39-4.31 (t, 2H), 4.09 (m, 1H), 3.78 (m,
1H), 2.91 (q, 1H), 1.85 (m, 1H), 1.71 (m, 1H), 1.61 (m, 1H),
1.46 (m, 1H). FAB MS, m/z: 549 [M + 1]+. Anal. Calcd for
C
28H32N6O4S‚1.3C2HF3O2‚0.3H2O: C, 52.1; H, 4.9; N, 11.9.
Found: C, 52.0; H, 4.9; N, 12.1.
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl[(4-am idin o-
3-flu or op h en yl)m eth yl]a m id e Tr iflu or oa cetic Acid (4).
1H NMR (CD3OD): δ 7.60 (m, 1H), 7.45-7.21 (m, 12H), 5.05
(d, 1H), 4.50 (dd, 1H), 4.35 (m, 2H), 4.07 (m, 1H), 3.78 (m, 1H),
2.88 (m, 1H), 1.85 (m, 1H), 1.72 (m, 1H), 1.61 (m, 1H), 1.47
C
28H31N5O4S‚1.5H2O: C, 60.0; H, 6.1; N, 12.5. Found: C, 60.2;
H, 6.3; N, 12.6.
N-Ca r boxym eth yl-D-d ip h en yla la n yl-L-p r olyl[(5-a m id i-
n o-2-th ien yl)m eth yl]a m id e Ditr iflu or oa cetic Acid (22).
To a solution of compound 28e (2.36 g, 3.89 mmol) in methanol
(45 mL) was added 10% palladium-on-carbon (240 mg) and
acetic anhydride (0.74 mL, 7.78 mmol), and the mixture was
stirred for 8 h under H2 (1 atm). The reaction mixture was
filtered through Celite, and the filtrate was concentrated in
vacuo. The residue was purified by preparative HPLC to give
the N-(tert-butoxycarbonyl)methyl-D-diphenylalanyl-L-prolyl-
[(5-amidino-2-thienyl)methyl]amide, acetic acid salt (1.6 g,
70%). 1H NMR (CD3OD): δ 7.77 (d, 1H), 7.59 (m, 2H), 7.44
(m, 2H), 7.31-7.18 (m, 6H), 7.16 (d, 1H), 4.61 (dd, 2H), 4.49
(d, 1H), 4.20 (d, 1H), 4.03 (m, 1H), 3.53 (m, 1H), 2.99 (m, 1H),
1.89 (s, 3H), 1.85-1.68 (m, 3H), 1.42 (s, 9H), 1.35 (m, 1H). FAB
MS, m/z: 590 [M + 1]+.
To a cooled (0 °C) solution of the above compound (1.4 g,
2.37 mmol) in dichloromethane (15 mL) was added TFA (15
mL), and the mixture was stirred for 3.5 h. After the reaction
mixture was concentrated in vacuo, the residue was purified
by reverse-phase preparative HPLC to give the title compound
(1.2 g, 80%). The analytical sample was obtained as a crystal-
line solid by neutralizing the TFA salt with 2 N NaOH. 1H
NMR (CD3OD): δ 7.80 (d, 1H), 7.65 (m, 2H), 7.49 (m, 2H),
7.39-7.22 (m, 6H), 7.19 (d, 1H), 5.33 (d, 1H), 4.60 (dd, 2H),
4.58 (d, 1H), 4.06 (m, 1H), 3.84 (dd, 2H), 3.49 (m, 1H), 2.85
(m, 1H), 1.89-1.73 (m, 3H), 1.32 (m, 1H). FAB MS, m/z: 534
[M + 1]+. Anal. Calcd for C28H31N5O4S‚1.5H2O: C, 60.0; H,
6.1; N, 12.5. Found: C, 60.2; H, 6.3; N, 12.6.
(m, 1H). FAB MS, m/z: 567 [M + 1]+. Anal. Calcd for C28H31
-
FN6O4S‚1.2C2HF3O2‚0.5H2O: C, 51.0; H, 4.7; N, 11.7. Found:
C, 50.8; H, 4.8; N, 11.4.
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl[(4-am idin o-
2-flu or op h en yl)m eth yl]a m id e Tr iflu or oa cetic Acid (5).
1H NMR (CD3OD): δ 7.71-7.21 (m, 13H), 4.98 (d, 1H), 4.56
(m, 1H), 4.35 (m, 2H), 4.08 (m, 1H), 3.78 (m, 1H), 2.90 (m,
1H), 1.84 (m, 1H), 1.75 (m, 1H), 1.61 (m, 1H), 1.47 (m, 1H).
FAB MS, m/z: 567 [M + 1]+. Anal. Calcd for C28H31
-
FN6O4S‚1.2C2HF3O2‚0.7H2O: C, 50.7; H, 4.7; N, 11.6. Found:
C, 50.7; H, 4.8; N, 11.5.
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl(6-am idin o-
3-p icolyl)a m id e Tr iflu or oa cetic Acid (6). 1H NMR (CD3-
OD): δ 8.72 (s, 1H), 8.08 (m, 1H), 7.99 (m, 1H), 7.44-7.35 (m,
4H), 7.26-7.24 (m, 6H), 4.98 (d, 1H), 4.59 (d, 1H), 4.33-4.30
(m, 2H), 4.06 (m, 1H), 3.76 (m, 1H), 2.90 (m, 1H), 1.84-1.45
(m, 4H). FAB MS, m/z: 550 [M + 1]+. Anal. Calcd for
C
27H31N7O4S‚2.7C2HF3O2‚0.7H2O: C, 44.9; H, 4.1; N, 11.3.
Found: C, 44.7; H, 4.2; N, 11.5.
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl[(4-am idin o-
2-th ien yl)m eth yl]a m id e Tr iflu or oa cetic Acid (8). 1H NMR
(CD3OD): δ 8.10 (m, 1H), 7.51-7.25 (m, 11H), 4.99 (d, 1H),
4.45 (dd, 2H), 4.31 (d, 1H), 4.06 (m, 1H), 3.76 (m, 1H), 2.87
(m, 1H), 1.87 (m, 1H), 1.71 (m, 1H), 1.60 (m, 1H), 1.47 (m,
1H). FAB MS, m/z: 555 [M + 1]+. Anal. Calcd for C28H31N5O4S‚
1.3C2HF3O2‚0.5H2O: C, 48.1; H, 4.6; N, 11.7. Found: C, 48.3;
H, 4.7; N, 11.5.
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl[(5-am idin o-
3-th ien yl)m eth yl]a m id e Tr iflu or oa cetic Acid (9). 1H NMR
(CD3OD): δ 7.83 (m, 1H), 7.45-7.35 (m, 4H), 7.30-7.15 (m,
7H), 4.95 (d, 1H), 4.45 (m, 1H), 4.32 (d, 1H), 4.21 (d, 1H), 4.04
(m, 1H), 3.76 (m, 1H), 2.89 (m, 1H), 1.85 (m, 1H), 1.77 (m,
1H), 1.61 (m, 1H), 1.46 (m, 1H). FAB MS, m/z: 555 [M + 1]+.
N-(2-Ca r boxyeth yl)-D-d ip h en yla la n yl-L-p r olyl-[(5-a m i-
d in o-2-th ien yl)m eth yl]a m id e Ditr iflu or oa cetic Acid (25).
A mixture of 27e (400 mg, 0.81 mmol), sodium carbonate (690
mg, 6.5 mmol), sodium iodide (609 mg, 4.1 mmol), tetrabutyl-
ammonium bromide (79 mg, 0.244 mmol), and methyl 3-bromo-
propionate (0.2 mL, 1.626 mmol) in toluene (8 mL) was heated
at reflux for 5 h. The reaction mixture was concentrated in

3620 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Lee et al.
Anal. Calcd for C26H30N6O4S2‚1.4C2HF3O2‚0.4H2O: C, 47.9; H,
4.5; N, 11.6. Found: C, 48.1; H, 4.6; N, 11.6.
1.0C2HF3O2‚3.0H2O: C, 51.3; H, 5.5; N, 10.0. Found: C, 51.3;
H, 5.7; N, 10.2.
N-Meth oxyca r bon yl-D-d ip h en yla la n yl-L-p r olyl[(5-a m i-
d in o-3-th ien yl)m eth yl]a m id e Tr iflu or oa cetic Acid (19).
1H NMR (CD3OD): δ 7.49 (d, 1H), 7.39 (m, 2H), 7.34 (m, 2H),
7.25 (m, 6H), 6.57 (d, 1H), 5.15 (d, 1H), 4.59 (m, 1H), 4.37 (m,
2H), 4.05 (m, 1H), 3.85 (m, 1H), 3.39 (s, 3H), 2.89 (m, 1H),
1.82-1.46 (m, 4H). FAB MS, m/z: 516 [M + 1]+. Anal. Calcd
for C28H31N5O4S‚1.2C2HF3O2‚0.2H2O: C, 54.2; H, 4.9; N, 10.4.
Found: C, 54.5; H, 4.7; N, 10.4.
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl[(5-am idin o-
2-fu r a n yl)m eth yl]a m id e Tr iflu or oa cetic Acid (10). 1H
NMR (CD3OD): δ 7.78 (m, 3H), 7.52 (m, 2H), 7.34 (m, 6H),
6.60 (d, 1H), 4.99 (d, 1H), 4.49 (m, 2H), 4.24 (d, 2H), 4.11 (m,
1H), 3.76 (m, 1H), 2.95 (m, 1H), 1.81-1.37 (m, 4H). FAB MS,
m/z: 538 [M + 1]+. Anal. Calcd for C26H30N6O5S‚1.2C2HF3O2‚
0.2H2O: C, 49.9; H, 4.7; N, 12.3. Found: C, 49.6; H, 4.6; N,
12.5.
N-Meth oxyca bon yl-D-d ip h en yla la n yl-L-p r olyl[(5-a m i-
d in o-2-fu r a n yl)m eth yl]a m id e Tr iflu or oa cetic Acid (20).
1H NMR (CD3OD): δ 7.49 (d, 1H), 7.39 (m, 2H), 7.34 (m, 2H),
7.25 (m, 6H), 6.57 (d, 1H), 5.15 (d, 1H), 4.59 (m, 1H), 4.37 (m,
2H), 4.05 (m, 1H), 3.85 (m, 1H), 3.39 (s, 3H), 2.89 (m, 1H),
1.82-1.46 (m, 4H). FAB MS, m/z: 518 [M + 1]+. Anal. Calcd
for C28H31N5O5‚1.2C2HF3O2‚0.3H2O: C, 55.5; H, 5.0; N, 10.7.
Found: C, 55.4; H, 5.1; N, 10.9.
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl[(5-am idin o-
3-fu r a n yl)m eth yl]a m id e Tr iflu or oa cetic Acid (11). 1H
NMR (CD3OD): δ 7.83 (s, 1H), 7.47 (s, 1H), 7.43 (m, 2H), 7.35
(m, 2H), 7.25 (m, 6H), 4.93 (d, 1H), 4.36 (m, 1H), 4.30 (d, 1H),
4.11 (m, 1H), 4.02 (m, 1H), 3.75 (m, 1H), 2.88 (m, 1H), 1.84
(m, 1H), 1.75-1.55 (m, 2H), 1.43 (m, 1H). FAB MS, m/z: 539
[M + 1]+. Anal. Calcd for C26H30N6O5S‚1.3C2HF3O2‚0.3H2O:
C, 49.4; H, 4.6; N, 12.1. Found: C, 49.6; H, 4.5; N, 11.9.
N-Ca r boxym eth yl-D-d ip h en yla la n yl-L-p r olyl[(4-a m id i-
N-Am in osu lfon yl-D-diph en ylalan yl-L-pr olyl[(4-am idin o-
1
n op h en yl)m eth yl]a m id e Ditr iflu or oa cetic Acid (21). H
1
2-th ia zolyl)m eth yl]a m id e TF A (12). H NMR (CD3OD): δ
NMR (CD3OD): δ 7.76 (d, 2H), 7.67 (m, 2H), 7.57 (d, 2H), 7.51
(m, 2H), 7.41-7.26 (m, 6H), 5.33 (d, 1H), 4.59 (d, 1H), 4.49
(dd, 2H), 4.10 (m, 1H), 3.82 (dd, 2H), 3.49 (m, 1H), 2.87 (m,
1H), 1.91-1.72 (m, 3H), 1.30 (m, 1H). FAB MS, m/z: 528 [M
+ 1]+. Anal. Calcd for C30H33N5O4‚2.3C2HF3O2‚0.7H2O: C,
51.7; H, 4.6; N, 8.7. Found: C, 51.9; H, 4.3; N, 8.4.
8.56 (s, 1H), 7.45 (m, 2H), 7.35 (m, 2H), 7.25 (m, 6H), 4.94 (d,
1H), 4.67 (m, 2H), 4.33 (d, 1H), 4.11 (m, 1H), 3.77 (m, 1H),
2.89 (m, 1H), 1.88 (m, 1H), 1.80-1.61 (m, 2H), 1.47 (m, 1H).
FAB MS, m/z: 556 [M + 1]+. Anal. Calcd for C25H29N7O4S2‚
1.5C2HF3O2‚0.2H2O: C, 46.1; H, 4.3; N, 13.4. Found: C, 46.2;
H, 4.5; N, 13.2.
N-Ca r boxym eth yl-D-d ip h en yla la n yl-L-p r olyl[(5-a m id i-
n o-3-th ien yl)m eth yl]a m id e Ditr iflu or oa cetic Acid (23).
1H NMR (CD3OD): δ 7.91 (s, 1H), 7.87 (s, 1H), 7.67 (m, 2H),
7.50 (m, 2H), 7.38-7.25 (m, 6H), 5.34 (d, 1H), 4.58 (d, 1H),
4.49 (d, 1H), 4.36 (d, 1H), 4.07 (m, 1H), 3.78 (dd, 2H), 3.53 (m,
1H), 2.89 (m, 1H), 1.83-1.72 (m, 3H), 1.31 (m, 1H). FAB MS,
m/z: 534 [M + 1]+. Anal. Calcd for C28H31N5O4S‚2.2C2HF3O2‚
0.5H2O: C, 49.2; H, 4.4; N, 8.9. Found: C, 49.0; H, 4.4; N, 9.0.
N-Ca r boxym eth yl-D-d ip h en yla la n yl-L-p r olyl[(4-a m id i-
n o-2-th ia zolyl)m eth yl]a m id e Ditr iflu or oa cetic Acid (24).
1H NMR (CD3OD): δ 8.63 (s, 1H), 7.65 (m, 2H), 7.52 (m, 2H),
7.45-7.23 (m, 6H), 5.31 (d, 1H), 4.74 (s, 2H), 4.55 (d, 1H), 4.13
(m, 1H), 3.78 (dd, 2H), 3.52 (m, 1H), 2.86 (m, 1H), 1.81 (m,
3H), 1.37 (m, 1H). FAB MS, m/z: 535 [M + 1]+. Anal. Calcd
for C27H30N6O4S‚2.2C2HF3O2‚0.8H2O: C, 47.0; H, 4.2; N, 10.4.
Found: C, 46.8; H, 4.5; N, 10.5.
Deter m in ation of En zym e In h ibition Con stan ts. Throm-
bin assay was performed according to the same procedure as
described previously25 except that the concentration of Chro-
mozym TH and thrombin was set to 80 µM and 1.5 mU/mL,
respectively, and the hydrolysis reaction was monitored for
1.5 h. Assays for trypsin, factor Xa, plasmin and t-PA were
performed according to published procedures.
Det er m in a t ion of P h a r m a cok in et ics. Male Sprague-
Dawley rats (250-300 g) were restrained individually on a
surgical plate in a supine position. The femoral artery and the
femoral vein (iv only) of rats were cannulated with polyeth-
ylene tubing (PE-50, Clay Adams, Parsippany, NJ ) under light
ether anesthesia. After complete recovery from anesthesia, rats
were orally given 30 mg/kg of the test compound dissolved in
distilled water via gavage or given 10 mg/kg via the femoral
vein for intravenous (iv) study. Blood samples (0.25 mL) were
collected from the femoral artery at 0 (for control), 1 (iv only),
5, 15, 30, 60, 90 (iv only), 120, 180, and 240 min after dosing.
Male beagle dogs (7-10 kg, Hazleton Research Product Inc.,
Calamazoo, MI) were housed individually in a metabolic cage
for the plasma disposition study. Dogs were orally given 10
mg/kg of the test compound dissolved in distilled water via
gavage or injected with 2 mg/(0.2 mL‚kg) via the cephalic vein
using INTROCAN. Blood samples were withdrawn via the
cephalic vein at 0 (for control), 1, 5 (iv only), 15, 30, 60, 90,
120, 180, 240, 360 (po only), and 480 (po only) min after dosing.
Blood samples were placed into heparinized tubes (25
U/mL), deproteinized with 2 volumes of methanol, and cen-
trifuged. The resulting supernatant (60 µL) was analyzed by
HPLC, eluting with a mixture of 0.1% trifluoroacetic acid
aqueous solution and acetonitrile with a ratio of 81% to 19%.
Plasma concentration of the test compound was recorded and
N-Meth ylsu lfon yl-D-d ip h en yla la n yl-L-p r olyl[(4-a m id i-
n op h en yl)m eth yl]a m id e Tr iflu or oa cetic Acid (13). 1H
NMR (CD3OD): δ 7.76 (m, 2H), 7.52 (m, 4H), 7.48-7.18 (m,
8H), 5.02 (d, 1H), 4.44 (m, 2H), 4.34 (d, 1H), 4.09 (m, 1H), 3.72
(m, 1H), 2.98 (m, 1H), 2.84 (s, 3H), 1.80 (m, 2H), 1.67 (m, 1H),
1.40 (m, 1H). FAB MS, m/z: 548 [M + 1]+ Anal. Calcd for
.
C
29H33N5O4S‚1.5C2HF3O2‚1.2H2O: C, 51.9; H, 5.0; N, 9.5.
Found: C, 51.8; H, 5.1; N, 9.6.
N-Meth ylsu lfon yl-D-diph en ylalan yl-L-pr olyl[(5-am idin o-
2-th ien yl)m eth yl]a m id e Tr iflu or oa cetic Acid (14). 1H
NMR (CD3OD): δ 7.78 (d, 1H), 7.50 (m, 2H), 7.42-7.18 (m,
9H), 5.02 (d, 1H), 4.58 (m, 2H), 4.31 (d, 1H), 4.04 (m, 1H), 3.72
(m, 1H), 2.95 (m, 1H), 2.87 (s, 3H), 1.81 (m, 2H), 1.64 (m, 1H),
1.42 (m, 1H). FAB MS, m/z: 554 [M + 1]+. Anal. Calcd for
C
27H31N5O4S2‚0.5C2HF3O2‚2H2O: C, 52.0; H, 5.5; N, 10.8.
Found: C, 52.0; H, 5.7; N, 10.7.
N-Meth oxyca r bon yl-D-d ip h en yla la n yl-L-p r olyl[(4-a m i-
1
d in op h en yl)m eth yl]a m id e Tr iflu or oa cetic Acid (15). H
NMR (CD3OD): δ 7.82 (m, 2H), 7.18-7.52 (m, 12H), 5.10 (d,
1H), 4.53 (m, 1H), 4.36 (m, 2H), 4.15 (m, 1H), 3.83 (m, 1H),
3.23 (s, 3H), 2.86 (m, 1H), 1.85 (m, 2H), 1.60 (m, 1H), 1.47 (m,
1H). FAB MS, m/z: 527 [M + 1]+. Anal. Calcd for C30H33N5O4‚
1.2C2HF3O2‚0.2H2O: C, 58.5; H, 5.2; N, 10.6. Found: C, 58.3;
H, 5.3; N, 10.8.
N-Meth oxyca r bon yl-D-d ip h en yla la n yl-L-p r olyl(5-a m i-
1
d in o-2-p icolyl)a m id e Tr iflu or oa cetic Acid (16). H NMR
(CD3OD): δ 8.67 (s, 1H), 8.09 (d, 1H), 7.96 (d, 1H), 7.41 (m,
2H), 7.35 (m, 2H), 7.21 (m, 6H), 5.15 (d, 1H), 4.59 (m, 1H),
4.42 (m, 1H), 4.38 (d, 1H), 4.09 (m, 1H), 3.82 (m, 1H), 3.33 (s,
3H), 2.90 (m, 1H), 1.91-1.75 (m, 2H), 1.69 (m, 1H), 1.47 (m,
1H). FAB MS, m/z: 529 [M + 1]+. Anal. Calcd for C29H32N6O4‚
2.2C2HF3O2‚0.2H2O: C, 50.9; H, 4.4; N, 10.6. Found: C, 50.7;
H, 4.5; N, 10.8.
N-Meth oxyca r bon yl-D-d ip h en yla la n yl-L-p r olyl[(5-a m i-
d in o-2-th ien yl)m eth yl]a m id e Tr iflu or oa cetic Acid (17).
1H NMR (CD3OD): δ 7.81 (m, 1H), 7.52-7.18 (m, 11H), 5.14
(d, 1H), 4.66 (m, 1H), 4.52 (m, 1H), 4.36 (dd, 1H), 4.07 (m, 1H),
3.84 (m, 1H), 3.30 (s, 3H), 2.91 (m, 1H), 1.85 (m, 2H), 1.65 (m,
1H), 1.49 (m, 1H). FAB MS, m/z: 533 [M + 1]+. Anal. Calcd
for C28H31N5O4S‚1.0C2HF3O2‚2.2H2O: C, 52.4; H, 5.3; N, 10.2.
Found: C, 52.7; H, 5.7; N, 10.3.
N-Meth oxyca r bon yl-D-d ip h en yla la n yl-L-p r olyl[(4-a m i-
d in o-2-th ien yl)m eth yl]a m id e Tr iflu or oa cetic Acid (18).
1H NMR (CD3OD): δ 8.26 (m, 1H), 7.48-7.18 (m, 11H), 5.14
(d, 1H), 4.59 (q, 1H), 4.36 (d, 2H), 4.07 (m, 1H), 3.83 (m, 1H),
3.45 (s, 3H), 2.92 (m, 1H), 1.82 (m, 2H), 1.65 (m, 1H), 1.49 (m,
1H). FAB MS, m/z: 516 [M + 1]+. Anal. Calcd for C28H31N5O4S‚

Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3621
used to calculate the pharmacokinetic parameters: maximum
plasma concentration of test compound (Cmax), time of maxi-
mum plasma concentration (Tmax), area under the curve (AUC),
and fraction of test compound absorbed (F).
was excised by surgical dissection, and wet weights of throm-
bus were measured.
Deter m in a tion of An ticoa gu la n t Activity. The antico-
agulant activity for activated partial thromboplastin time
(aPTT) was determined using an automatic coagulometer
(Coag-A-Mate RA4, Organon Teknika). For the aPTT, 100 µL
of plasma and the inhibitor were mixed with the same volume
of the aPTT reagent (Platelin LS, Organon Teknika), and the
mixture was incubated at 37 °C for 5 min, followed by the
addition of 100 µL of 25 mM CaCl2. For determination of ex
vivo coagulation parameters of activated clotting time (ACT),
blood samples were immediately collected from the carotid
artery, and the clotting time was measured in a Hemotech
automate (ACT II, Medtronic. Inc., Minneapolis, MN).
Deter m in a tion of Meta bolic Sta bility. A portion of liver
of rats (n ) 2, pooled), dogs (n ) 2, pooled), monkeys (n ) 2,
pooled), and humans (n ) 2) was excised, chilled, and blotted
dry on the paper tissue. After homogenization by a Tempest
tissue homogenizer (The Virtis Company, NY) with 2 volumes
of 0.1 M Tris-acetate buffer (pH 7.4) containing 0.1 M KCl and
1 mM EDTA, the homogenate was centrifuged twice at 9000g
for 20 min using Beckman model J 2-M1 (CA). A portion of the
resulting supernatant was used as the S9 fraction. The
metabolic reaction was initiated by adding the test compound
(final concentration of 20 µM) to a mixture of the NADPH
generating system (1 mM NADP, 3 mM G6P, 0.2 units/mL
G6PD, and 5 mM MgCl2) and liver S9 (6.72 mg/mL) in 100
mM potassium phosphate buffer (pH 7.4) incubated at 37 °C
and 150 opm. The volume of the total reaction mixture was
0.5 mL. Aliquots of 100 µL were withdrawn at 0, 10, 30, and
60 min, and 2 volumes of methanol were added to deproteinize
the samples for HPLC analysis. The metabolic activity was
also measured in the absence of the NADPH generating
system to serve as a control.
Ack n ow led gm en t. We cordially thank Drs. Yong Zu
Kim and In Chul Kim for their continuous support of
this work. We also thank Tae Hun Kim for determining
some enzyme inhibition constants, LG Chemical Analy-
sis Center, Korea, for performing microanalyses, and
Dr. J ohn Kim for some helpful discussion in the
preparation of this manuscript.
Refer en ces
Deter m in ation of An tith r om botic Activity. Rat m odel.26
Male Sprague-Dawley rats (300-350 g, six per group) were
anesthetized by intraperitoneal injection with urethane solu-
tion (1.25 g/kg). The abdomen was surgically opened by a
midline incision, and the inferior venae cava was carefully
dissected free from surrounding connective tissue. The venae
iliolumbar and spermatica were ligated with a silk thread.
Thrombus formation was initiated by infusion of a thrombo-
plastin preparation (Simplastin) using an infusion pump
(model 100, IITC Life Science) via the left femoral vein at 0.5
mL/(kg‚min). Simplastin (Organon Teknika) was reconstituted
with 4 mL of distilled water and then given diluted at 1:2.5 in
distilled water. At the start of infusion for 30 s, the vena cava
was ligatured below the left renal vein. After the end of
infusion, the vena cava was also ligatured above the iliac veins
16 mm apart from upper ligature. After 15 min of stasis, the
thrombus formed inside the vessel was carefully removed and
weighed. Before the sample was weighed, the excess blood was
removed by blotting the wet clot on the wet Whitman filter
paper.
(1) (a) Kimball, S. D. Challenges in the Development of Orally
Bioavailable Thrombin Active Site Inhibitors. Blood Coagulation
Fibrinolysis 1995, 6, 511-519. (b) Shafer, J . A. Cardiovascular
Chemotherapy: Anticoagulants. Curr. Opin. Chem. Biol. 1998,
2, 458-465. (c) Adang, A. E. P.; Rewinkel, J . B. M. A New
Generation of Orally Active Antithrombotics: Comparing Strat-
egies in the GPIIb/IIIa, Thrombin and Factor Xa Areas. Drugs
Future 2000, 25, 369-383. (d) Weitz, J . I.; Crowther, M. Direct
Thrombin Inhibitors. Thromb. Res. 2002, 106, V275-V284.
(2) For recent reviews, see the following. (a) Hauptmann, J .;
Sturzebecher, J . Synthetic Inhibitors of Thrombin and Factor
Xa: From Bench to Bedside. Thromb. Res. 1999, 93, 203-241.
(b) Sanderson, P. E. J . Small, Noncovalent Serine Protease
Inhibitors. Med. Res. Rev. 1999, 19, 179-197. (c) Vacca, J . P.
New Advances in the Discovery of Thrombin and Factor Xa
Inhibitors. Curr. Opin. Chem. Biol. 2000, 25, 369-383.
(3) (a) Oh, Y. S.; Yun, M.; Hwang, S. Y.; Hong, S.; Shin, Y.; Lee, K.;
Yoon, K. H.; Yoo, Y. J .; Kim, D. S.; Lee, S. H.; Lee, Y. H.; Park,
H. D.; Lee, C. H.; Lee, S. K.; Kim, S. Discovery of LB30057, a
Benzamidrazone-Based Selective Oral Thrombin Inhibitor. Bioorg.
Med. Chem. Lett. 1998, 8, 631-634. (b) Kim, I.-C.; Oh, Y. S.;
Yun, M.; Hwang, S. Y.; Hong, S.; Lee, Y. H.; Lee, K.; Kim, S.;
Yoo, Y. J .; Yoon, K. H.; Kim, D. S.; Lee, C. H. Circulation 1997,
96, I-41.
(4) (a) Chi, L.; Mertz, T. E.; Rogers, K. L.; J aniczek, N.; Peng, Y.
W.; Saganek, L.; Bousley, R. F.; J uneau, P. L.; Uprichard, A.
C.; Gallogher, K. P. Antithrombotic Effect of LB30057 (CI-1028),
a New Synthetic Thrombin Inhibitor, in a Rabbit Model of
Thrombosis: Comparison with Inogatran. J . Thromb. Throm-
bolysis 2001, 11, 19-31. (b) McClanahan, T. B.; Hicks, G. W.;
Ignasiak, D. P.; Bousley, R. F.; Mertz, T. E.; J uneau, P. L.;
J aniczek-Dolpihin, N.; Kim, I. C.; Gallogher, K. P. The Anti-
thrombotic Effect of CI-1028, an Orally Bioavailable Direct
Thrombin Inhibitor, in a Canine Model of Venous and Arterial
Thrombosis. J . Thromb Thrombolysis 2000, 10, 277-284.
(5) Lee, K.; J ung, W.-H.; Park, C. W.; Hong, C. Y.; Kim, I. C.; Kim,
S.; Oh, Y. S.; Kwon, O. H.; Lee, S.-H.; Park, H. D.; Kim, S. W.;
Lee, Y. H.; Yoo, Y. J . Benzylamine-Based Selective and Orally
Bioavailable Inhibitors of Thrombin. Bioorg. Med. Chem. Lett.
1998, 8, 2563-2568.
(6) Wiley, M. R.; Chirgadze, N. Y.; Clawson, D. K.; Craft, T. J .;
Gifford-Moore, D. S.; J ones, N. D.; Olkowski, J . L.; Schacht, A.
L.; Weir, L. C.; Smith, G. F. D-Phe-Pro-p-Amidinobenzylamine:
A Potent and Highly Selective Thrombin Inhibitor. Bioorg. Med.
Chem. 1995, 5, 2835-2391.
(7) Stone, S. R.; Tapparelli, C. Thrombin Inhibitors as Antithrom-
botic Agents: The Importance of Rapid Inhibition. J . Enzyme
Inhib. 1995, 9, 3-15.
(8) Elg, M.; Gustafson, D.; Deinum, J . The Importance of Enzyme
Inhibition Kinetics for the Effect of Thrombin Inhibitors in a
Rat Model of Arterial Thrombosis. Thromb. Haemostasis 1997,
78, 1286-1292.
(9) Lee, K.; J ung, W.-H.; Kang, M.; Lee, S.-H. Noncovalent Thrombin
Inhibitors Incorporating an Imidazolylethynyl P1. Bioorg. Med.
Chem. Lett. 2000, 10, 2775-2278.
(10) (a) Tucker, T. J .; Lumma, W. C.; Mulichak, A. M.; Chen, Z.;
Naylor-Olsen, A. M.; Lewis, S. D.; Lucas, R.; Freidinger, R. M.;
Kuo, L. C. Design of Highly Potent Noncovalent Inhibitors That
Utilize a Novel Lipophilic Binding Pocket in the Thrombin Active
Saline (control) or the test compound (1 mg/kg) was given
by either bolus iv injection only or bolus injection plus infusion
via the femoral vein 5 min before the thromboplastin infusion.
The bolus injection volume was 0.5 mL/kg. For oral study, the
test compound was given 120 min prior to thrombus formation.
Antithrombotic activity (%) was expressed as 100 × (A - B)/
A, where A is the mean clot weight of control group and B is
the mean clot weight of test compound group.
Ra bbit Mod el. New Zealand White rabbits (2.5-3.5 kg)
were anesthetized by intravenous administration of a mixture
of ketamine and xylazine. Maintenance of anesthesia was
achieved by periodic intravenous administration of pentobar-
bital sodium. The animals were administered with either
saline or compound 22 via intravenous bolus injection, followed
by infusion for 60 min. During the infusion, the vena cava and
left femoral vein were isolated for insertion of silk threads.
Eight strands of silk thread (USP 4-0, Ethicon), which served
as a stimulus for clot formation, were tied at the end of a 2F
arterial embolectomy catheter (Baxter) and trimmed to 3 cm
in length. The threads were wetted for about 20 min in a
diluted thromboplastin solution (1:4 dilution of PT reagents).
Fifteen minutes after the onset of infusion, the threads were
inserted into the abdominal vena cava via the left femoral vein
and were positioned approximately 2 cm below the branch of
the left renal vein. Distal and proximal ligatures with PE-230
catheters occluded the vena cava segment (4 cm in length)
containing silk threads. A ligature stenosis was formed by
removal of the catheter. Infusion continued for a further 45
min to allow the growth of thrombus. The resulting thrombus

3622 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Lee et al.
Site. J . Med. Chem. 1997, 40, 830-832. (b) Feng, D.-M.; Gardell,
S. J .; Lewis, S. D.; Bock, M. G.; Chen, Z.; Freidinger, R. M.;
Naylor-Olsen, A. M.; Ramjit, H. G.; Woltmann, R.; Baskin, E.
P.; Lynch, J . J .; Lucas, R.; Shafer, J . A.; Chen, I.-W.; Dancheck,
K. B.; Mao, S.-S.; Krueger, J . A.; Hare, T. R.; Mulichak, A. M.;
Vacca, J . P. Discovery of a Novel Selective and Orally Bioavail-
able Class of Thrombin Inhibitors. J . Med. Chem. 1997, 40,
3726-3733. (c) Cheng, L.; Goodwin, C. A.; Schully, M. F.;
Kakkar, V. V.; Claeson, G. J . Med. Chem. 1992, 35, 3364.
(11) Pinto, D. J . P.; Orwat, M. J .; Wang, S.; Fevig, J . M.; Quan, M.
L.; Amparo, E.; Cacciola, J .; Rossi, K. A.; Alexander, R. S.;
Smallwood, A. M.; Luettgen, J . M.; Liang, L.; Aungst, B. J .;
Wright, M. R.; Knabb, R. M.; Wong, P. C.; Wexler, R. R.; Lam,
P. Y. S. Discovery of 1-[3-(Aminomethyl)phenyl]-N-[3-fluoro-2′-
(methylsulfonyl)-[1,1′-biphenyl]-4-yl]-3-(trifluoromethyl)-1H-pyra-
zole-5-carboxamide (DPC423), a Highly Potent, Selective, and
Orally Bioavailable Inhibitor of Blood Coagulation Factor Xa.
J . Med. Chem. 2001, 44, 566-578.
Brown, A. D.; Fish, P. V.; Hawkeswood, E.; Holland, S. J .; J ames,
K.; McElroy, A. B.; Overington, J .; Powling, M. J .; Rance, D. J .
Design of Selective Thrombin Inhibitors Based on the (R)-
PhePro-Arg Sequence. J . Med. Chem. 2002, 45, 2432-2453. (c)
Hauel, N. H.; Nar, H.; Priepke, H.; Ries, U.; Stassen, J .-M.;
Wienen, W. Structure-Based Design of Novel Thrombin Inhibi-
tors. J . Med. Chem. 2002, 45, 1757-1766. (d) Wiley, M. R.; Weir,
L. C.; Bliggs, S. L.; Chirgadze, N. Y.; Clawson, D.; Gifford-Moor,
D. S.; Smith, G. F.; Vasudevan, V.; Zornes, L. L.; Klimkowski,
V. J . The Design of Potent, Selective, Non-covalent, Peptide
Thrombin Inhibitors Utilizing Imidazole as a S1 Binding Ele-
ment. Bioorg. Med. Chem. Lett. 1999, 9, 2767-2772. (e) Adang,
A. E. P.; Lucas, H.; de Man, A. P. A.; Engh, R. A.; Grootenhuis,
P. D. J . Novel Acylguanidine Containing Thrombin Inhibitors
with Reduced Basicity at the P1 Moiety. Bioorg. Med. Chem.
Lett. 1998, 8, 3603-3608.
(18) (a) Gustafson, D.; Antonsson, T.; Bylund, R.; Eriksson, U.;
Gyzander, E.; Nilsson, I.; Elg, M.; Mattsson, C.; Deinum, J .;
Pehrsson, S.; Karlsson, O.; Nilsson, A.; Sorensen, H. Effects of
Melagatran, a New Low Molecular Weight Thrombin Inhibitor,
on Thrombin and Fibrinolytic Enzymes. Thromb. Haemostasis
1997, 78, 1286-1292. (b) Eriksson, B. I.; Carlsson, S.; Halva-
rsson, M.; Risberg, B.; Mattsson, C. Antithrombotic Effect of Two
Low Molecular Weight Thrombin Inhibitors and a Low Molec-
ular Heparin in a Caval Vein Thrombosis Model in the Rat.
Thromb. Haemostasis 1997, 78, 1404-1407. (c) Sorbera, L. A.;
Bayes, M.; Castaner, J .; Silvestre, J . Melagatran and Ximela-
gatran. Drugs Future 2001, 26, 1155-1170.
(19) Adang et al. also reported different oral bioavailability profiles
among structurally similar carboxymethyl derivatives of tri-
peptidic thrombin inhibitors (see ref 17a).
(20) Details of crystallogarphic data will be published elsewhere.
(21) J udkins, B. D.; Allen, D. G.; Cook, T. A.; Evans, B.; Sardharwala,
T. A. Synth. Commun. 1996, 26, 4351.
(12) Ho, J . Z.; Gibson, T. S.; Semple, J . E. Novel, Potent Non-Covalent
Thrombin Inhibitors Incorporating P3-Lactam Scaffolds. Bioorg.
Med. Chem. 2002, 12, 743-748.
(13) Lee, K.; J ung, W.-H.; Hwang, S. Y.; Lee, S.-H. Noncovalent
Tripeptidic Thrombin Inhibitors Incorporating Amidrazone,
Amine and Amidine Functions at P1. Bioorg. Med. Chem. Lett.
2002, 12, 1017-1022.
(14) Lee, K.; J ung, W.-H.; Park, C. W.; Park, H. D.; Lee, S. H.; Kwon,
O. W. Fluorobenzamidrazone Thrombin Inhibitors: Influence of
Fluorine on Enhancing Oral Absorption. Bioorg. Med. Chem.
Lett. 1999, 9, 2483-2486.
(15) The use of 2,5-thienylamidine P1 as a benzamidine surrogate
for thrombin inhibitors was described in our previous com-
munications; see ref 11 and the following. Lee, K.; Hwang, S.
Y.; Yun, M.; Kim, D. S. Potent and Efficacious Thienylamindine-
Incorporated Thrombin Inhibitors. Bioorg. Med. Chem. 1998, 6,
1683-1686. This P1 element was also utilized by other research
groups; see the following. Reiner, J . E.; Sieve, 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.;
Brabury, A. E.; Nolan, T. G.; Semple, J . E. Non-Covalent
Thrombin Inhibitors Featuring P3-Heterocycles with P1-Mono-
cyclic Arginine Surrogates. Bioorg. Med. Chem. 2002, 12, 1203-
1208. Lange, U. E. W.; Zechel, C. Solid-Phase Synthesis of
Thrombin Inhibitors. Bioorg. Med. Chem. 2002, 12, 1571-1573.
(16) (a) Teger-Nilsson, A. C.; Gyzander, E.; Anderson, S.; Englund,
M.; Mattson, C.; Ulvinger, J . C.; Gustafsson, D. In Vitro
Properties of Inogatran, a Selective Low Molecular Weight
Inhibitor of Thrombin. Thromb. Haemostasis 1997, 73, 1633. (b)
Preville, P.; He, J . X.; Tarazi, M.; Siddiqui, M. A.; Cody, W. L.;
Doherty, A. M. An Efficient Preparation of the Potent and
Selective Pseudopeptide Thrombin Inhibitor, Inogatran. Bioorg.
Med. Chem. 1997, 7, 1563-1566.
(22) Schacht, A. L.; Smith, G. F.; Wiley, M. F. Antithrombotic Agents.
U.S. Patent 5,710,130, 1998.
(23) Ewing, W. R.; Becker, M. R.; Manetta, V. A.; Davis, R. S.; Pauls,
H. W.; Mason, H.; Choi-Sledeski, Y. M.; Green, D.; Cha, D.;
Spada, A. P.; Cheney, D. L.; Mason, J . S.; Maignan, S.; Guillo-
teau, J .-P.; Brown, K.; Colussi, D.; Bentley, R.; Bostwick, J .;
Kasiewski, C. J .; Morgan, S. R.; Leadley, R. J .; Dunwiddie, C.
T.; Perrone, M. H.; Chu, V. Design and Structure-Activity
Relationships of Potent and Selective Inhibitors of Blood Co-
agulation Factor Xa. J . Med. Chem. 1999, 42, 3557-3571.
(24) Videnov, G.; Kaiser, D.; Kempter, C.; J ung, G. Synthesis of
Naturally Occurring, Conformationally Restricted Oxazole- and
Thiazole-Containing Di- and Tripeptide Mimics. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 1503-1506.
(25) Lee, K.; Hwang, S. Y.; Hong, S. W.; Hong, C. Y.; Lee, C.-S.; Shin,
Y.; Kim, S.; Yun, M.; Yoo, Y. J .; Kang, M.; Oh, Y. S. Structural
Modification of an Orally Active Thrombin Inhibitor, LB30057:
Replacement of the D-Pocket-Binding Naphthyl Moiety. Bioorg.
Med. Chem. 1998, 6, 869-876.
(26) Millet, J .; Theveniaux, J .; Brown, N. L. The Venous Antithrom-
botic Effect of LF 1351 in the Rat following Oral Administration.
Thromb. Haemostasis 1992, 67, 176-179.
(17) (a) Adang, A. E. P.; de Man, A. P. A.; Vogel, G. M. T.;
Grootenhuis, P. D. J .; Smit, M. J .; Peters, C. A. M.; Visser, A.;
Rewinkel, J . B. M.; van Dinther, T.; Lucas, H.; Kelder, J .; van
Aelst, S.; Meiuleman, D. G.; van Boeckel, C. A. A. Unique
Overlap in the Prerequisites for Thrombin Inhibition and Oral
Bioavailability Resulting in Potent Oral Antithrombotics. J . Med.
Chem. 2002, 45, 4419-4432. (b) Danilewicz, J . C.; Abel, S. M.;
J M030025J