Rational design, synthesis, and structure-activity relationships of novel factor Xa inhibitors: (2-substituted-4-amidinophenyl)pyruvic and -propionic acids

Article abstract of DOI:10.1021/jm020485x

An inhibitor of factor Xa (fXa), the m-substituted benzamidine AXC1578 (1a), was structurally modified with the aim of increasing its potency. In particular, pyruvic acid and propionic acid substituents were incorporated into the P1 benzamidine moiety to

Full text of DOI:10.1021/jm020485x

J . Med. Chem. 2003, 46, 1845-1857
1845
Ra tion a l Design , Syn th esis, a n d Str u ctu r e-Activity Rela tion sh ip s of Novel
F a ctor Xa In h ibitor s: (2-Su bstitu ted -4-a m id in op h en yl)p yr u vic a n d
-p r op ion ic Acid s1
Kazuyuki Sagi, Tadakiyo Nakagawa, Masahiro Yamanashi, Shingo Makino, Mitsuo Takahashi,
Masaru Takayanagi, Kaoru Takenaka, Nobuyasu Suzuki, Seiji Oono, Noriyasu Kataoka, Kohki Ishikawa,
Sayaka Shima, Yumiko Fukuda, Takashi Kayahara, Shunji Takehana, Yoichiro Shima, Kazumi Tashiro,
Hiroshi Yamamoto, Ryota Yoshimoto, Seinosuke Iwata, Takashi Tsuji, Kuniya Sakurai, and Masataka Shoji*
Pharmaceutical Research Laboratories, Ajinomoto Company Inc., 1-1 Suzuki-cho, Kawasaki-ku, Kawasaki-shi 210-8681, J apan
Received October 29, 2002
An inhibitor of factor Xa (fXa), the m-substituted benzamidine AXC1578 (1a ), was structurally
modified with the aim of increasing its potency. In particular, pyruvic acid and propionic acid
substituents were incorporated into the P1 benzamidine moiety to introduce a favorable
interaction with the oxy-anion hole in the catalytic triad region of fXa. This strategy was based
on computational docking studies using the extracted active site of fXa. The validity of the
computational model was supported by the acquisition of X-ray crystal structures of the 1a -
trypsin and 3b-trypsin complexes (the homology around the active sites of fXa and trypsin is
high). The above modifications significantly increased the inhibitory activity toward fXa,
whereas the high selectivity for fXa versus thrombin was maintained or enhanced. Compounds
3b, 3c, 3e, and 4b are considered to be potential lead compounds for the development of orally
active anticoagulant drugs because they demonstrated potent activity when administered orally
to cynomolgus monkeys.
In tr od u ction
Figure 1a. To support the proposed model, the 1a -
trypsin complex structure was elucidated by X-ray
crystallography. Trypsin was considered to be a reason-
able surrogate protein, because the homology around
the active sites of fXa and trypsin is high.8 The similari-
ties in the two structures support the validity of the
computational model (Figure 1b).
Factor Xa (fXa) is an attractive target for the develop-
ment of orally active anticoagulants.2 Although vitamin
K antagonists, for example, warfarin, have been used
in chronic oral anticoagulant therapy for thromboem-
bolic diseases, there have been numerous reports of
serious risks, especially bleeding, associated with the
use of these antagonists.3 Because of the central role of
thrombin in thrombosis, thrombin inhibitors have been
investigated as alternatives to warfarin. However, direct
thrombin inhibitors increase the likelihood of bleeding
complications.4 Factor Xa is a trypsin-like serine pro-
tease that converts the prothrombin zymogen to its
active form, thrombin. On the basis of studies in
preclinical efficacy models, we considered that selective
fXa inhibitors might have lower bleeding risks and a
more favorable safety/efficacy ratio than thrombin
inhibitors.5
Initially, we focused on preparing and evaluating new
bisamidine compounds because some bisamidine com-
pounds are already known to be selective inhibitors of
fXa.6 This program resulted in the discovery of the
m-substituted benzamidine AXC1578 (1a ) as a selective
fXa inhibitor (Ki for fXa ) 0.22 µM, Ki for thrombin )
710 µM). The simple structure of 1a encouraged further
study aimed at increasing the potency and selectivity.
In h ibitor Design Ba sed on Com p u ta tion a l Mod -
els. To facilitate inhibitor design, the structure of the
1a -fXa complex was modeled computationally by using
the X-ray structure of fXa (1HCG),7 as described in the
Experimental Section. The results are illustrated in
According to this model, the S4 pocket of fXa is occu-
pied by the 4-amidinophenyl moiety of 1a , and the S1
pocket is occupied by the other amidinophenyl moiety
(see Chart 1). Therefore, we examined the effect of
variation of the P4 moiety on the affinity of 1a for fXa
(1b,9 1c,10 1d , and 1e in Table 1). Next, we attempted
carboxymethyl substitution on the linker part (1f, 1g
in Table 1), on the basis of a visual inspection of the
1a -fXa complex structure. According to the model of
the R-isomer 1f-fXa complex (Figure 2), the carboxylic
acid group of 1f might interact favorably with Gln19211
and Arg143 of fXa through electrostatic attraction, in
contrast to the electrostatic repulsion with Glu192 of
thrombin (1DWC).12 The S-isomer 1g would not interact
with Gln192 and Arg143 of fXa, according to the
computer modeling (data not shown).
Further study of the 1a -fXa complex structure
indicated the presence of an empty region around the
catalytic triad (CT) of fXa (Figure 1a). Appropriate
substitutions at the para position of the P1 benzamidine
might introduce favorable interactions with the CT
region, especially the oxy-anion hole. In connection with
this, 4-amidinophenylpyruvic acid (APPA) (2) has been
reported as a trypsin inhibitor with oral bioavailabil-
ity.13 The X-ray crystal structure of the APPA-trypsin
complex indicates that the pyruvic acid moiety of APPA
interacts with the oxy-anion hole.14 The benzamidine
* To whom correspondence should be addressed. Tel: +81-44-210-
5826. Fax: +81-44-210-5876. E-mail: masataka_shouji@ajinomoto.com
10.1021/jm020485x CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/11/2003

1846 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Sagi et al.
F igu r e 1. (a) Calculated docking mode of AXC1578 (1a ) in factor Xa (1HCG). Both the hydrogen and carbon atoms of the inhibitor
are colored green. Specific residues relevant to the binding of 1a are colored. Asp189 at the bottom of the S1 pocket is shown in
magenta. Gln192 and Arg143 at the top of the S1 pocket are in yellow. Ser195, His57, and Asp102 forming the catalytic triad and
Gly 193 forming the oxy-anion hole are in orange. Residues forming the S4 pocket, Tyr99, Phe174, and Trp215 are in cyan. In
this docking mode, the amidine of 1a hydrogen bonds with both the carboxyl group of Asp189 and the carbonyl group of Gly218,
and the carbonyl group of 1a hydrogen bonds with the amide NH of Gly218. (b) The crystal structure of the complex of 1a with
bovine trypsin is superposed onto Figure 1a. Compound 1a and the residues of the crystal structure are colored magenta and
gray, respectively. The residues that differ between factor Xa and trypsin are depicted as factor Xa/trypsin. The hydrogen bonds
of Figure 1a are apparent in this crystal structure.
Ta ble 1. Enzyme-Inhibitory Activity of AXC1578 Analogues without Substitution at the Para-Position of Benzamidine
a
See reference 29.
moieties of 1a and APPA appear to occupy similar sites
of fXa and trypsin, respectively. Therefore, the introduc-
tion of a pyruvic acid moiety into 1a (3a in Table 2)
should increase the affinity of the compound for the oxy-
anion hole of fXa. The computational model of 3a -fXa
indicated that the pyruvic acid moiety of 3a can interact

Relationships of Novel Factor Xa Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1847
that the propionic acid moiety might also interact
favorably with Gly193 of the oxy-anion hole and Gln192
of fXa (Figure 4). Because 3b showed good anticoagulant
activity after oral administration, a compound having
the same substituent R1 as 3b (4a in Table 2) was
utilized in this modeling.
Ch em istr y. AXC1578 analogues without substitution
at the para position of benzamidines were prepared as
follows (Scheme 1). The hydroxyl group of 3-hydroxy-
benzonitrile (5) was alkylated with either alkyl halides
or alcohols. Next, the Boc group was removed under
acidic conditions, followed by acylation with various
carboxylic acids to give 7. Finally, the nitriles 7 were
converted into the target amidines 1 by means of the
Pinner reaction.15
The benzamidine compounds with pyruvic acid sub-
stitution were synthesized as shown in Scheme 2.
3-Hydroxybenzoic acid (8) was iodinated to give the
4-iodo compound (9).16 Then, the carboxyl group of 9 was
converted into a nitrile group (10). Direct iodination of
5 to afford 10 did not proceed well. Compound 10 was
subsequently alkylated to afford 11 by using a procedure
similar to that in Scheme 1. Compound 11 was coupled
with methyl 2-acetamidoacrylate by means of the Heck
reaction17 to give 3-phenyl-2-acetamidoacrylate 12. Com-
pound 12 was converted into 13 by using a procedure
similar to that shown in Scheme 1. The nitrile group of
13 was converted into an amidino group, and the target
molecules 3 were obtained by simultaneous hydrolysis18
of both the enamido and the ester moieties of the
2-acetamido acrylate moiety.
The benzamidine compounds with propionic acid
substitution were synthesized according to Scheme 3.
The key intermediate 11a was coupled with ethyl
acrylate to afford 14. Subsequently, amidines 16 were
prepared by using a procedure similar to that shown in
Scheme 1. Finally, the target compounds 4 were ob-
tained by hydrogenation of the double bond, followed
by hydrolysis of the ester.
F igu r e 2. Calculated docking mode of 1f with factor Xa. The
same color scheme as in Figure 1a is adopted. For comparison,
the residues of thrombin (1DWC) are superposed and colored
gray. The residues that differ between factor Xa and thrombin
are depicted as factor Xa or thrombin. The carboxyl group of
1f is placed in the vicinity of Gln 192 and Arg143 of fXa, which
is consistent with hydrogen-bond formation, taking into ac-
count the flexibility of the side chains of these amino acids.
with the oxy-anion hole of fXa without interfering with
the linker moiety (Figure 3a). According to the X-ray
structure of the APPA-trypsin complex, the effect may
even be enhanced by nucleophilic interaction of the
oxygen atom of the Ser195 side chain with the pyruvic
acid moiety. Compound 3a was therefore synthesized,
together with the R1-modified analogue 3b (Table 2).
The X-ray crystal structure determination of the 3b-
trypsin complex provided support for the proposed
binding mode (Figure 3b) based on the high homology
between the binding sites of fXa and trypsin.
Because compounds into which a pyruvic acid moiety
had been introduced showed good results in the biologi-
cal activity evaluation (see below), we extended the
modification to propionic acid substitution in place of
pyruvic acid substitution. Computer modeling indicated
Biologica l Resu lts a n d Discu ssion
Table 1 summarizes the results of replacement of the
4-amidinobenzoyl group in 1a (1b-e). Compounds 1b
F igu r e 3. (a) Calculated docking mode of 3a in factor Xa. The same color scheme as in Figure 1a is adopted. The docking calculation
was done assuming formation of a noncovalent Michaelis-Menten complex before the formation of a plausible covalent complex,
as observed in the crystal structure of APPA-trypsin complex. (b) Crystal structure of the complex of 3b with bovine trypsin. The
same color scheme as above is adopted. A covalent bond is formed between the carbonyl carbon of 3b and Oγ of Ser195 of trypsin.

1848 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Sagi et al.
Ch a r t 1
(fXa Ki ) 0.17 µM) and 1c (0.29 µM) are equipotent to
the parent compound 1a (0.22 µM). When R1 is basic
(1a -1c), the compounds showed high affinity, probably
owing to cation-π stabilization between the cationic
portions of the protonated R1 and the aromatic side
chains of Tyr99, Phe174, and Trp215 in the S4 pocket
(Figure 1a).19 Especially in the case of 1b, the affinity
may also be owed to hydrogen bonding of the long R1
moiety with the flexible acidic side chain of Glu97
hanging over the end of the S4 pocket, by induced-fitting
as suggested in the DX-9065a-fXa complex.20 The
compound having a neutral R1, 1d , is an order of
magnitude more potent than 1a (fXa Ki ) 20 nM
compared to 0.22 µM). The high potency of 1d may be
attributed to favorable hydrophobic and van der Waals
interactions of the pyrrolidine-carbonyl group with the
three aromatic side chains mentioned above. Taking into
account the low affinity of 1e (R1 ) Ph, 2.8 µM), it is
clear that the R1 group in 1a -1d greatly enhances the
affinity of these compounds for fXa. Table 1 also shows
the result of incorporation of a carboxymethyl group in
the linker part of 1a (1f, 1g). As we expected, the affinity
for fXa of the R-isomer 1f was 10-fold greater than
thatof 1a , and the selectivity over thrombin was con-
siderably increased (more than 50000-fold selectivity).
On the other hand, the affinity of 1a was not improved
by S-carboxymethyl substitution (1g).
1a -d and 3a -d ). In all cases, the introduction of the
pyruvic acid moiety resulted in an approximately 10-
fold enhancement of the affinity, indicating that the P4
and pyruvic acid moieties had predominantly indepen-
dent effects in terms of structure-activity relationships.
However, the affinity for thrombin, as well as fXa, was
increased by the introduction of the pyruvic acid sub-
stituent, presumably because the CT region is highly
conserved among these serine proteases. Thus, the
selectivity of 3a -d for fXa over thrombin was little
changed from that of the corresponding compounds 1a -
d . To increase the selectivity, the R-carboxymethyl
group was further attached to the linker of 3b and 3c
(3e and 3f in Table 2). To our surprise, this modification
did not significantly enhance the affinity for fXa in
contrast to the conversion from 1a to 1f, but nonethe-
less, the selectivity for fXa was markedly improved.
Thus, the R-carboxymethyl and pyruvic acid moieties
did not have independent effects in this case, possibly
because the binding mode of the compound could not
be optimum for both moieties.
Although the pyruvic acid moiety is an excellent
substituent in terms of affinity, its electrophilic site may
not be favorable from the viewpoint of pharmacology.21
Therefore, we also designed and synthesized corre-
sponding compounds with propionic acid substitution
(4a , 4b in Table 2). Their affinity was slightly decreased
compared with that of the compounds with pyruvic acid
substitution (3b, 3c); however, the affinities of these
compounds (4a : 33 nM, 4b: 43 nM) were much higher
than those of the compounds without substitution (1b:
0.17 µM, 1c: 0.29 µM), presumably because of the
interactions discussed in the design section. On the
The data in Table 2 demonstrate that the affinity for
fXa of 1a (0.22 µM) was greatly enhanced by the
incorporation of the pyruvic acid moiety on P1 benza-
midine (3a , 26 nM). Next, the R1 group of 3a was
replaced with the above P4 moieties to estimate the
effect of pyruvic acid introduction (comparison between

Relationships of Novel Factor Xa Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1849
Ta ble 2. Enzyme-Inhibitory Activity of Benzamidine Compounds with Pyruvic Acid or Propionic Acid Substitution
other hand, the compounds with propionic acid substi-
tution did not show enhanced potency against thrombin,
and therefore, the selectivity for fXa over thrombin was
significantly improved to more than 35000-fold. This
may be due to the repulsive interaction with the
corresponding Glu192 of thrombin (Figure 4).
Table 3 shows the concentrations of inhibitors re-
quired to double the prothrombin time (PT) in plasma
clotting. The selected compounds (3b, 3c, 3e, and 4b)
showed potent anticoagulant activities. Therefore, the
ex vivo anticoagulant activities were further examined
in cynomolgus monkeys. In terms of PT prolongation,
these compounds exhibited potent anticoagulant activity
at the oral dose of 10 mg/kg (Figure 5). The anticoagu-
lant effects of the compounds with pyruvic acid substi-
tution (3b, 3c, 3e) peaked at 1-2 h and disappeared
within 8 h, whereas the compounds with propionic acid
substitution (4b) had a longer duration of action. The
oral bioavailability of 4b, evaluated by measuring the
drug concentration in plasma, was 6%. Compounds 3b,
3c, and 3e had similar bioavailabilities of about 12%
and might therefore be potential lead compounds for the
F igu r e 4. Calculated docking mode of 4a in fXa. The residues
of thrombin are superposed, and the color scheme is the same
as in Figure 2. The propionic acid moiety of 4a is placed in
the vicinity of Gly193 and Gln192 of fXa and appears to
interact with them.

1850 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Sagi et al.
Sch em e 1a
a
Reagents and conditions: (A) X ) Cl, K2CO3, DMF 50 °C; (B) X ) OH, DEAD, PPh3, THF; (C) HCl, dioxane; (D) condensation
reaction; (E) i. HCl, EtOH, dioxane; ii. (NH4)2CO3, EtOH; iii. ethyl acetimidate hydrochloride, NEt3 for 1b; iv. cHCl, 40 °C for 1f and 1g.
Sch em e 2a
a
Reagents and conditions: (A) ICl, AcOH; (B) i. ClCOOEt, TEA, THF; ii. NH3, THF; iii. (CF3CO)2O, pyridine, dioxane; (C) X ) Cl,
K2CO3, DMF, 50 °C; (D) X ) OH, N,N,N′,N′-tetramethylazodicarboxamide, PPh3, THF; (E) Pd(OAc)2, PPh3, DMF, 100 °C; (F) HCl, dioxane;
(G) EDC, TEA, HOBt, DMF; (H) i. HCl, EtOH, dioxane; ii. (NH4)2CO3, EtOH; iii. ethyl acetimidate hydrochloride, TEA for 13b and 13e;
(I) HCl aq, 80 °C.
development of anticoagulant drugs for oral administra-
tion.
selectivity; and (3) pyruvic and propionic acid moieties
were incorporated onto the P1 benzamidine to interact
with the CT region, with the aim of enhancing the
affinity for fXa.
Con clu sion
We have computationally designed selective fXa
inhibitors based on the three-dimensional structures of
fXa and thrombin. This structure-based drug design
strategy to improve the affinity and selectivity of 1a may
be summarized as follows: (1) The 4-amidinophenyl
group was successfully replaced with other groups
designed to interact better with the S4 pocket; (2)
R-carboxymethyl substitution was introduced into the
linker moiety because the computational models pre-
dicted electrostatic attraction with fXa and repulsion
with thrombin, which was expected to improve the
As a result of this approach, selective fXa inhibitors
with high affinity were successfully developed. Some of
the compounds (3b, 3c, 3e, and 4b) were further
examined for anticoagulant activity after oral admin-
istration to cynomolgus monkeys, and the results indi-
cated that several of the inhibitors could be potential
lead compounds for the development of orally active
anticoagulant drugs. Further investigations to improve
the pharmacokinetic profile of these compounds are in
progress.

Relationships of Novel Factor Xa Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1851
Sch em e 3a
a
Reagents and conditions: (A) Pd(OAc)2, PPh3, DMF, 100 °C; (B) HCl, dioxane; (C) EDC, TEA, HOBt, DMF; (D) i. HCl, EtOH, dioxane;
ii. (NH4)2CO3, EtOH; iii. ethyl acetimidate hydrochloride, TEA for 15a ; (E) i. Pd/C, H2; ii. cHCl, 50 °C.
Ta ble 3. Anticoagulant Activities of fXa Inhibitors
2× PT (µM)a
Ch em istr y. Workup included drying over magnesium
sulfate, filtering, and concentrating in vacuo. Column chro-
matography of intermediates was performed using silica gel
(Merck, particle size 0.063-0.200 mm). Final products were
purified by preparative reversed-phase HPLC, which was
performed on a Waters 600 system with a C-18 reversed-phase
column (Inertsil ODS-3, GL Sciences Inc.), using a mixture of
acetonitrile and water, both of which contained 0.1% TFA.
Fractions containing the desired material were concentrated
and lyophilized to obtain the final products as a white solid.
NMR spectra were recorded on a Varian EM-390 at 300 MHz.
Keto-enol tautomerism was observed in the NMR spectra of
all the benzamidine compounds with pyruvic acid substitution
(3). Mass spectra (ESI and FAB) were measured on J EOL
J MS-DX300 instruments. Where analyses are indicated only
by the symbols of the elements, the results obtained were
within 0.4% of the theoretical values. Reagent abbreviations:
HOBT, 1-hydroxybenzotriazole; EDC, 1-ethyl-3-(3-dimethyl-
aminopropyl)carbodiimide HCl; TEA, triethylamine; TFA,
trifluoroacetic acid.
compound
human
monkey
3b
3c
3e
4b
0.20
0.22
0.38
0.42
0.12
0.22
0.38
0.70
a
Concentration required to double the prothrombin time (PT).
Exp er im en ta l Section
Molecu la r Mod elin g. The coordinate sets for the protein
structures used in this investigation were 1HCG7 for fXa and
1DWC12 for thrombin. Compounds 1a , 1f, 3a , and 4a were
docked into the fXa active site using an in-house docking
program. This method is composed of three steps. Initially,
alignment of an anchor fragment in a target protein is
searched, namely P1 benzamidine in the case of fXa. Second,
the remaining part (excluding the anchor) is put into incre-
mental conformational searches in the protein environment.
Finally, the alignment of the resulting whole ligand is opti-
mized. The docking program was created using the Sybyl
Programming Language and utilizes the Sybyl commands,
systematic search, and Leapfrog search.24 Full details of these
docking studies have been reported.1c,25
Deter m in a tion of Tr yp sin -In h ibitor Cr ysta l Str u c-
tu r e. Cr ysta lliza tion a n d In h ibitor Soa k in g. Bovine pan-
creatic trypsin (Sigma T8003) was solubilized to final concen-
tration of 30 mg/mL in a buffer containing 0.5 mM CaCl2, 20
mM benzamidine, and 50 mM cacodylate at pH 6.0. The
sitting-drop crystallization method was used to form crystals.
Single crystals were obtained with ammonium sulfate (2.0-
2.2 M) at pH 6.0 in the open form (P212121, a ) 63.7 Å, b )
63.5 Å, c ) 68.9 Å), wherein the binding site is free of any
contact with crystallographic symmetry-related molecules.26
The crystals were soaked in 500 µL of a buffer containing 3.0
M ammonium sulfate, 1 mM CaCl2, and 50 mM cacodylate at
pH 6.0 for 1 h five times to exclude benzamidine, and then
they were soaked in 500 µL of the same solution with 2 mM
1a or 3b for 2 d.
ter t-Bu tyl (2-Ch lor oeth yl)ca r ba m a te (17a ). A mixture
of TEA (30 mL, 0.2 mol), di-tert-butyl carbonate (40 g, 0.18
mol), and 2-chloroethylamine HCl (25 g, 0.2 mol) in dichlo-
romethane (300 mL) was stirred for 3 d. Then 0.5 M HCl was
added to the mixture. The organic layer was separated, washed
with saturated NaCl, and worked up to yield 17a (35.6 g,
99%): 1H NMR (CDCl3) δ 1.45 (9H, s), 3.45 (2H, dt), 3.60 (2H,
t), 5.00 (1H, br).
Ben zyl (3R)-3-ter t-Bu toxyca r bon yla m in o-4-h yd r oxy-
bu tyr a te (17b). Ethyl chloroformate (3.08 mL, 32.2 mmol)
was added to a mixture of Boc-D-aspartic acid â-benzyl ester
(10.4 g, 32.3 mmol) and TEA (4.49 mL, 32.2 mmol) in
tetrahydrofuran (100 mL) at 0 °C. The reaction mixture was
stirred for 10 min and then filtered to remove precipitates. A
small amount of crushed ice and NaBH4 (1.23 g, 32.5 mmol)
in water was added to the filtrate at 0 °C. The mixture was
stirred for 3 min and then diluted with 1 N HCl. EtOAc was
added to the solution, and then the organic layer was washed
with saturated NaCl and worked up. The crude product was
purified by chromatography (EtOAc/hexane) to give 17b (5.28
g, 51%): 1H NMR (CDCl3) δ 1.42 (9H, s), 2.66 (2H, d), 3.65
(2H, dd), 4.00 (1H, ddt), 5.14 (2H, s), 7.35-7.40 (5H, m).
Da t a Collect ion a n d P r ocessin g. X-ray intensity data
were collected using an R-AXIS II at room temperature
(Rigaku, J apan). Data processing was done with PROCESS
(Rigaku), and structure refinement was performed using
X-PLOR.3.851.27 The starting point was the uncomplexed
trypsin structure at 1.50 Å resolution (1TLD). The final
R-factors of 1a and 3b were 0.188 and 0.193, respectively, in
the resolution range of 10-1.88 Å.
Ben zyl (3S)-3-ter t-Bu toxyca r bon yla m in o-4-h yd r oxy-
bu tyr a te (17c). The title compound was prepared as described
above, with Boc-L-aspartic acid â-benzyl ester as starting
material.
3-(2-Am in oeth oxy)ben zon itr ile HCl (6a ). P a r t A. A
mixture of 3-cyanophenol (6.64 g, 55.7 mmol), K2CO3 (7.70 g,

1852 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Sagi et al.
Ben zyl (3R)-3-Am in o-4-(3-cya n op h en oxy)bu tyr a te HCl
(6b). P a r t A. Diethyl azodicarboxylate 40% solution in toluene
(7.84 g, 18 mmol) was added to a mixture of 17b (5.28 g, 18
mmol), 3-cyanophenol (2.38 g, 20 mmol), and triphenylphos-
phine (4.72 g, 18 mmol) in tetrahydrofuran (60 mL) at 0 °C.
The mixture was stirred for 2 d and concentrated in vacuo.
The residue was diluted with EtOAc, washed with 1 N NaOH
and saturated NaCl, and then worked up. The crude product
was purified by chromatography (EtOAc/hexane) to give ben-
zyl (3R)-3-tert-butoxycarbonylamino-4-(3-cyanophenoxy)bu-
tyrate (4.32 g, 58%): 1H NMR (CDCl3) δ 1.46 (9H, s), 2.79 (2H,
d), 4.00 (1H, dd), 4.06 (1H, dd), 4.41 (1H, br), 5.13 (2H, s), 5.56
(1H, br), 7.05-7.18 (4H, m), 7.21-7.38 (5H, m).
P a r t B. The product of Part A was treated as described for
6a , Part B, to give 6b: 1H NMR (DMSO-d6) δ 2.90 (2H, d),
3.90 (1H, m), 4.15-4.30 (2H, m), 5.20 (2H, s), 7.30-7.40 (7H,
m), 7.40-7.60 (3H, m), 8.35 (3H, br).
Ben zyl (3S)-3-Am in o-4-(3-cya n op h en oxy)bu tyr a te HCl
(6c). The title compound was prepared as described above,
with benzyl (3S)-3-tert-butoxycarbonylamino-4-hydroxybu-
tyrate (17c) as the starting material.
4-(1-ter t-Bu toxyca r bon yl-4-p ip er id yloxy)ben zoic Acid
(18a ). P a r t A. Diethyl azodicarboxylate (1.62 g, 9.3 mmol),
ethyl 4-hydroxybenzoate (1.70 g, 10.2 mmol), 1-tert-butoxy-
carbonyl-4-hydroxypiperidine (1.76 g, 9.3 mmol), and triphen-
ylphosphine (2.44 g, 9.3 mmol) were treated as described for
6b, Part A, to give ethyl 4-(1-tert-butoxycarbonyl-4-piperidyl-
oxy)benzoate (1.57 g, 44%): 1H NMR (CDCl3) δ 1.38 (3H, t),
1.50 (9H, s)1.70-1.80 (2H, m), 1.90-2.00 (2H, m), 3.30-3.41
(2H, m), 3.63-3.75 (2H, m), 4.35 (2H, q), 4.55 (1H, m), 6.90
(2H, d), 8.00 (2H, d).
P a r t B. A 1.0 N solution of NaOH (5 mL) was added to a
solution of the product of Part A (847 mg, 2.43 mmol) in
ethanol (50 mL). The reaction mixture was stirred for 3 d and
concentrated in vacuo. The residue was diluted with EtOAc,
then washed with 1 N HCl and saturated NaCl, and worked
up to yield 18a (697 mg, 92%): 1H NMR (CDCl3) δ 1.50 (9H,
s), 1.70-2.00 (4H, m), 3.30-3.40 (2H, m), 3.65-3.75 (2H, m),
4.60 (1H, s), 6.95 (2H, d), 8.05 (2H, d).
4-(1-P yr r olid in eca r bon yl)ben zoic a cid (18b). A mixture
of methyl 4-chlorocarbonylbenzoate (29.0 g, 0.146 mmol),
pyrrolidine (14.2 g, 200 mmol), and TEA (21.0 g, 208 mmol)
in dichloromethane (300 mL) was stirred. The solution was
washed with 1 N HCl, saturated NaHCO3, and saturated
NaCl, and then it was worked up to give crude methyl 4-(1-
pyrrolidinecarbonyl)benzoate. NaOH (12.0 g, 0.3 mmol) was
added to a solution of the crude ester in a mixture of water,
methanol, and tetrahydrofuran (70 mL:70 mL:70 mL). The
reaction mixture was stirred and concentrated in vacuo. The
residue was diluted with CH2Cl2, washed with saturated NaCl,
and then worked up to yield 18b (23.7 g, 74%): 1H NMR
(DMSO-d6) δ 1.75-1.90 (4H, m), 3.30-3.50 (4H, m), 7.62 (2H,
d), 7.99 (2H, d), 13.14 (1H, br).
F igu r e 5. Anticoagulant activities of fXa inhibitors after oral
administration at 10 mg/kg in cynomolgus monkeys. The
results are expressed as mean ( SEM of three determinations
for (A) 3b, (B) 3c, (C) 3e, and (D) 4b.
N-[2-(3-Cya n op h en oxy)eth yl]-4-cya n oben za m id e (7a ).
Ethyl chloroformate (0.67 mL, 7.1 mmol) was added to a
solution of 4-cyanobenzoic acid (18c) (1.13 g, 7.68 mmol) and
N-methylmorpholine (1.6 mL, 14.1 mmol) in DMF at 0 °C.
After 10 min, 6a (1.27 g, 6.41 mmol) was added, and the
mixture was stirred for 5 h. The reaction mixture was diluted
with EtOAc, washed with 1 N HCl, 1 N NaOH, and saturated
NaCl, and then worked up. The residue was washed with Et2O
and EtOAc to give 7a (1.29 g, 69%): 1H NMR (CDCl3) δ 3.91
(2H, dt), 4.19 (2H, t), 6.78 (1H, br), 7.14 (1H, d), 7.17 (1H, s),
7.28 (1H, d), 7.39 (1H, t), 7.75 (2H, d), 7.90 (2H, d).
55.7 mmol), and 17a (10 g, 55.7 mmol) in DMF (100 mL) was
heated to 60 °C for 4 d. The solution was concentrated in vacuo
and diluted with EtOAc. The solution was washed with 1 N
HCl, 1 N NaOH, and saturated NaCl and then worked up.
The crude product was purified by chromatography (EtOAc/
hexane) to give 3-[2-(tert-butoxycarbonylamino)ethoxy]ben-
zonitrile (10.9 g, 75%): 1H NMR (CDCl3) δ 1.44 (1H, s), 3.55
(2H, dt), 4.05 (2H, t), 4.95 (1H, br), 7.12 (1H, d), 7.14 (1H, s),
7.26 (1H, d), 7.38 (1H, t).
N-[2-(3-Cyan oph en oxy)eth yl]-4-(1-ter t-bu toxycar bon yl-
4-p ip er id yloxy)b en za m id e (7b ). The title compound was
prepared from 18a (211 mg, 0.65 mmol) and 6a (129 mg, 0.65
mmol) as described for 7a . The crude product was purified by
chromatography (EtOAc/hexane) to give 7b (167 mg, 55%): 1H
NMR (CDCl3) δ 1.50 (9H, s), 1.65-1.80 (2H, m), 1.85-2.00 (2H,
m), 3.30-3.40 (2H, m), 3.60-3.75 (2H, m), 3.90 (2H, dt), 4.20
(2H, t), 4.55 (1H, m), 6.45 (1H, t), 6.94 (2H, d), 7.15 (1H, d),
7.17 (1H, s), 7.26 (1H, d), 7.38 (1H, t), 7.74 (2H, d).
P a r t B. A solution of 4.0 M HCl in 1,4-dioxane (100 mL)
was added to the product of Part A (23.9 g, 91.2 mmol). The
reaction mixture was stirred for 2 h and concentrated in vacuo.
The residue was washed with EtOAc to give 6a (15.4 g, 85%):
1H NMR (DMSO-d6) δ 3.20 (2H, t), 4.30 (2H, t), 7.35 (1H, d),
7.45 (1H, d), 7.50 (1H, s), 7.55 (1H, t), 8.35 (3H, br).

Relationships of Novel Factor Xa Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1853
N-[2-(3-Cya n op h en oxy)eth yl]-1-(4-p yr id yl)p ip er id in e-
4-ca r boxa m id e (7c). A mixture of 1-(4-pyridyl)piperidine-4-
carboxylic acid HCl (18d )10 (412 mg, 1.48 mmol), 6a (350 mg,
1.77 mmol), TEA (0.25 mL, 1.77 mmol), HOBT (240 mg, 1.77
mmol), and EDC (340 mg, 1.77 mmol) in DMF (3 mL) was
stirred overnight. The mixture was diluted with EtOAc,
washed with 1 N NaOH and saturated NaCl, and then worked
up to give 7c (470 mg, 91%): 1H NMR (DMSO-d6) δ 1.52 (2H,
m), 1.75 (2H, d), 2.40 (1H, m), 2.80 (2H, m), 3.40 (2H, dt), 3.90
(2H, m), 4.08 (2H, t), 6.80 (2H, d), 7.31 (1H, d), 7.40 (1H, d),
7.42 (1H, s), 7.51 (1H, t), 8.09 (1H, t), 8.13 (2H, d).
N-[2-(3-Cya n op h en oxy)et h yl]-4-(p yr r olid in e-1-ca r b o-
n yl)ben za m id e (7d ). The title compound was prepared from
6a and 18b as described for 7c: 1H NMR (CDCl3) δ 1.80-
2.00 (4H, m), 3.30-3.70 (4H, m), 3.85 (2H, dt), 4.20 (2H, t),
7.14-7.28 (4H, m), 7.38 (1H, t), 7.48 (2H, d), 7.79 (2H, d).
N-[2-(3-Cya n op h en oxy)eth yl]ben za m id e (7e). Benzoyl
chloride (421 mg, 3.00 mmol) was added to a solution of 6a
and TEA (2.09 mL, 15.0 mmol) in DMF (15 mL) at 0 °C. The
reaction mixture was stirred overnight and diluted with
EtOAc. The solution was washed with 1 N HCl, saturated
NaHCO3, and saturated NaCl, and then worked up. The
residue was purified by chromatography to give 7e (710 mg,
89%): 1H NMR (CDCl3) δ 3.91 (2H, q), 4.19 (2H, t), 6.54 (1H,
m), 7.18 (2H, d), 7.28 (1H, d), 7.38 (1H, d), 7.46 (2H, d), 7.52
(1H, d), 7.78 (2H, d).
(460 mg, 1.31 mmol) was treated as described for 1a to yield
1c (402 mg, 52%): 1H NMR (DMSO-d6) δ 1.57 (2H, m), 1.82
(2H, m), 2.51-2.60 (1H, m), 3.20 (2H, m), 3.40 (2H, dt), 4.09
(2H, t), 4.23 (2H, m), 7.18 (2H, d), 7.30 (1H, d), 7.40 (1H, d),
7.40 (1H, s), 7.55 (1H, t), 8.20 (3H, m), 9.25 (2H, br), 9.29 (2H,
br). MS (ESI): 368 (MH+). Anal. (C20H25N5O2‚2.0TFA‚1.5H2O)
C, H, N.
N-[2-(3-Am id in op h en oxy)eth yl]-4-(1-p yr r olid in eca r bo-
n yl)ben za m id e Tr iflu or oa ceta te (1d ). To 4.0 M HCl solu-
tion in 1,4-dioxane (5 mL) and ethanol (0.5 mL) was added 7d
(270 mg, 0.74 mmol). The reaction mixture was stirred for 5 d
and concentrated in vacuo. The crude imidate was diluted with
ethanol (10 mL), and ammonium carbonate (300 mg) was
added to the solution. The reaction mixture was stirred for 2
d and concentrated in vacuo. Purification by reversed-phase
HPLC yielded 1d (238 mg, 65%): 1H NMR (DMSO-d6) δ 1.75-
1.90 (4H, m), 3.30-3.50 (4H, m), 3.70 (2H, dt), 4.20 (2H, t),
7.34 (1H, d), 7.39 (1H, d), 7.40 (1H, s), 7.54 (1H, t), 7.59 (2H,
d), 7.91 (2H, d), 8.80 (1H, t), 9.10 (2H, br), 9.30 (2H, br). MS
(ESI): 381 (MH+). Anal. (C21H24N4O3‚TFA‚0.35H2O) C, H, N.
N-[2-(3-Cya n op h en oxy)eth yl]ben za m id e Tr iflu or oa ce-
ta te (1e). Compound 7e (460 mg, 1.73 mmol) was treated as
described for 1a to yield 1e (117 mg, 14%): 1H NMR (DMSO-
d6) δ 3.69 (2H, dt), 4.23 (2H, t), 7.35-7.53 (7H, m), 7.89 (2H,
d), 8.78 (1H, br), 9.21 (2H, br), 9.33 (2H, br). MS (ESI): 284
(MH+). Anal. (C16H17N3O2‚TFA‚0.5H2O) C, H, N.
(3R)-3-(4-Am idin oben zoylam in o)-4-(3-am idin oph en oxy)-
bu tyr ic Acid Bistr iflu or oa ceta te (1f). P a r t A. Compound
7f (1.21 g, 2.75 mmol) was treated as described for 1a to yield
ethyl (3R)-3-(4-amidinobenzoylamino)-4-(3-amidinophenoxy)-
butyrate bistrifluoroacetate (456 mg, 26%): 1H NMR (DMSO-
d6) δ 1.15 (3H, t), 2.82 (2H, d), 4.07 (2H, q), 4.12 (1H, dd), 4.24
(1H, dd), 4.72 (1H, br), 7.33 (1H, d), 7.39 (1H, s), 7.40 (1H, d),
7.54 (1H, dd), 7.91 (2H, d), 8.02 (2H, d), 8.84 (1H, d), 9.16 (2H,
s), 9.28 (4H, s), 9.42 (2H, s). MS (ESI): 412 (MH+).
P a r t B. A mixture of the Part A product (466 mg, 0.73
mmol) and hydrochloric acid (10 mL) was stirred at 40 °C for
6 h and concentrated in vacuo. The residue was purified by
reversed-phase HPLC to give 1f (151 mg, 46%): 1H NMR
(DMSO-d6) δ 2.74 (2H, d), 4.13 (1H, dd), 4.24 (1H, dd), 4.69
(1H, ddt), 7.35 (1H, d), 7.40 (1H, d), 7.41 (1H, s), 7.55 (1H,
dd), 7.91 (2H, d), 8.03 (2H, d), 8.81 (1H, d), 9.20 (2H, s), 9.28
(2H, s), 9.33 (2H, s), 9.43 (2H, s). MS (ESI): 384 (MH+). Anal.
(C19H21N5O4‚2.0TFA‚1.0H2O) C, H, N.
(3S)-3-(4-Am idin oben zoylam in o)-4-(3-am idin oph en oxy)-
bu tyr ic Acid Bistr iflu or oa ceta te (1 g). The title compound
was prepared as described above, with benzyl (3S)-3-amino-
4-(3-cyanophenoxy)butyrate HCl (6c) as the starting material.
Anal. (C19H21N5O4‚2.0TFA‚1.0H2O) C, H, N.
3-Hyd r oxy-4-iod oben zoic Acid (9). Iodine monochloride
(53.0 g, 326 mmol) was added to a solution of 3-hydroxybenzoic
acid (8) (30.0 g, 217 mmol) in acetic acid (200 mL). The reaction
mixture was stirred at 45 °C for 15 h, and concentrated in
vacuo. The residue was washed with 1% sodium thiosulfate
and water to give 9 (17.2 g 30%):1H NMR (DMSO-d6) δ 7.13
(1H, dd), 7.43 (1H, d), 7.80 (1H, d); the spectrum was identical
to the reported spectrum.16b
3-Hyd r oxy-4-iod oben zon itr ile (10). P a r t A. Ethyl chlo-
roformate (19.7 mL, 206 mmol) was added to a solution of 9
(22.3 g, 89.7 mmol) and TEA (28.7 mL, 206 mmol) in tetrahy-
drofuran (300 mL) at 0 °C. The reaction mixture was stirred
for 15 min and filtered to remove precipitates. The filtrate was
added to a NH3 solution in tetrahydrofuran at 0 °C. The
mixture was stirred for 10 h and concentrated in vacuo to give
the crude benzamide.
P a r t B. The product of Part A was diluted with dioxane
(450 mL), and then trifluoroacetic anhydride (17.4 mL, 117
mmol) and pyridine (21.8 mL 269 mmol) were added to the
solution at 0 °C. The reaction mixture was stirred for 18 h at
room temperature and concentrated in vacuo. The residue was
diluted with CHCl3, washed with saturated NaCl, and worked
up to give the crude benzonitrile.
Ben zyl (3R)-3-(4-Cya n oben zoyla m in o)-4-(3-cya n op h en -
oxy)bu tyr ic Acid (7f). Compound 6b (1.52 g, 4.38 mmol),
4-cyanobenzoyl chloride (1.09 g, 6.58 mmol), TEA (1.22 mL,
8.76 mmol), and CH2Cl2 (5 mL) were treated as described for
7e to give 7f (1.21 g, 63%): 1H NMR (CDCl3) δ 2.86 (1H, dd),
2.95 (1H, dd), 4.12 (1H, dd), 4.20 (1H, dd), 4.85 (1H, br), 5.16
(2H, s), 7.09 (1H, d), 7.11 (1H, dd), 7.24-7.40 (7H, m), 7.72
(2H, d), 7.83 (2H, d).
Ben zyl (3S)-3-(4-Cya n oben zoyla m in o)-4-(3-cya n op h en -
oxy)bu tyr ic Acid (7g). The title compound was prepared as
described above, with benzyl (3S)-3-amino-4-(3-cyanophenoxy)-
butyrate HCl (6c) as the starting material.
N -[2-(3-Am id in o p h e n o x y )e t h y l]-4-a m id in o b e n za -
m id e Bistr iflu or oa ceta te (1a ). A mixture of 30% HCl sol-
ution in ethanol (24 mL), 7a (2.43 g, 8.35 mmol), and a 4.0 M
HCl solution in 1,4-dioxane (56 mL) was stirred for 4 d and
concentrated in vacuo. The crude imidate was diluted with 10%
ammonia solution in ethanol. The reaction mixture was stirred
for 24 h and concentrated in vacuo to yield the crude product.
Purification by reversed-phase HPLC yielded 1a (1.19 g,
26%): 1H NMR (DMSO-d6) δ 3.69 (2H, dt), 4.24 (2H, t), 7.32
(1H, d), 7.39 (1H, d), 7.40 (1H, s), 7.53 (1H, t), 7.90 (2H, d),
8.05 (2H, d), 9.02 (1H, t), 9.18 (2H, br), 9.30 (4H, br), 9.43 (2H,
br). MS (FAB): 326 (MH+). Anal. (C17H19N5O2‚2.0TFA‚2.0H2O)
C, H, N.
N-[2-(3-Am id in op h en oxy)eth yl]-4-(1-a cetim id oyl-4-p i-
p er id yloxy)ben za m id e Bistr iflu or oa ceta te (1b). P a r t A.
Compound 7b (165 mg, 0.35 mmol) was treated as described
for 1a to yield N-[2-(3-amidinophenoxy)ethyl]-4-(4-piperidy-
loxy)benzamide bistrifluoroacetate (124 mg, 0.20 mmol): 1H
NMR (DMSO-d6) δ 1.80-1.90 (2H, m), 2.08-2.18 (2H, m),
3.02-3.30 (4H, m), 3.62 (2H, q), 4.21 (2H, t), 4.75 (1H, m),
7.06 (2H, d), 7.30-7.42 (3H, m), 7.53 (1H, t), 7.85 (2H, d), 8.58
(2H, br), 8.61 (1H, br), 9.12 (2H, br), 9.28 (2H, br). MS (ESI):
383 (MH+).
P a r t B. Ethyl acetimidate HCl (147 mg, 1.2 mmol) was
added to a solution of the Part A product (124 mg, 0.20 mmol)
and TEA (183 mg, 1.8 mmol) in ethanol (5 mL). The reaction
mixture was stirred for 6 d and concentrated in vacuo. The
residue was purified by reversed-phase HPLC to give 1b (120
mg, 92%): 1H NMR (DMSO-d6) δ 1.70-1.82 (2H, m), 2.02-
2.14 (2H, m), 2.30 (3H, s), 3.50-3.60 (2H, m), 3.65 (2H, q),
3.70-3.80 (2H, m), 4.20 (2H, t), 4.80 (1H, m), 7.07 (2H, d),
7.30-7.40 (3H, m), 7.53 (1H, t), 7.85 (2H, d), 8.57-8.63 (2H,
m), 9.11-9.18 (3H, m), 9.28 (2H, br). MS (ESI): 424 (MH+).
Anal. (C23H29N5O3‚2.0TFA‚1.2H2O) C, H, N.
N-[2-(3-Am idin oph en oxy)eth yl]-1-(4-pyr idyl)piper idin e-
4-ca r b oxa m id e Bist r iflu or oa cet a t e (1c). Compound 7c
P a r t C. The residue of Part B was diluted with a mixture
(180 mL) of tetrahydrofuran and methanol (1:1). A 1 N NaOH

1854 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Sagi et al.
solution (90 mL) was added, and the reaction mixture was
stirred for 4 h to hydrolyze the remaining ethyl phenyl
carbonate. The solution was concentrated, and the residue was
diluted with EtOAc. This solution was washed with 1 N HCl
and saturated NaCl and worked up. The crude product was
purified by chromatography (EtOAc/hexane) to give 10 (9.29
g, 42%): 1H NMR (CDCl3) δ 5.63 (1H, br), 6.96 (1H, dd), 7.23
(1H, d), 7.79 (1H, d).
1.80 (2H, m), 1.85-2.00 (2H, m), 2.02 (3H, s), 3.30-3.40 (2H,
m), 3.60-3.75 (2H, m), 3.80 (3H, s), 4.35 (2H, t), 4.55 (1H, m),
6.82 (2H, d), 6.99 (1H, t), 7.18-7.22 (2H, m), 7.33 (1H, s), 7.44
(1H, s), 7.69 (2H, d), 7.87 (1H, d).
Meth yl 2-Acetam ido-3-[2-[2-[1-(pyr idin -4-yl)piper idin e-
4-ca r bon yla m in o]eth oxy]-4-cya n op h en yl]a cr yla te Tr i-
flu or oacetate (13c). Bromotripyrrolidinophosphonium hexaflu-
orophosphate (1.51 g, 3.24 mmol) was added to a mixture of
1-(pyridin-4-yl)piperidine-4-carboxylic acid HCl (18d ) (0.79 g,
3.24 mmol), 19a (1.0 g, 2.94 mmol), and TEA (1.64 mL) in DMF
at 0 °C. The reaction mixture was stirred overnight at ambient
temperature and concentrated in vacuo. The residue was
purified by HPLC to give 13c (360 mg, 25%): 1H NMR (DMSO-
d6) δ 1.50-1.70 (2H, m), 1.75-1.90 (2H, m), 1.95 (3H, s), 2.50
(1H, m), 3.15-3.30 (2H, m), 3.42 (2H, dt), 3.65 (3H, s), 4.10-
4.22 (4H, m), 7.15-7.21 (3H, m), 7.44 (1H, d), 7.58 (1H, s),
7.68 (1H, d), 8.09 (1H, t), 8.21 (2H, d), 9.65 (1H, s).
Meth yl 2-Aceta m id o-3-[2-[2-[4-(p yr r olid in e-1-ca r bon -
yl)ben zoyla m in o]eth oxy]-4-cya n op h en yl]a cr yla te (13d ).
Compound 18b (1.56 g, 8.1 mmol), 19a (2.5 g, 7.4 mmol),
HOBT (1.1 g, 8.1 mmol), TEA (1.53 mL), EDC (1.55 g, 8.1
mmol), and DMF were treated as described for 7c to give 13d
(1.80 g, 50%): 1H NMR (DMSO-d6) δ 1.78-1.90 (4H, m), 1.95
(3H, s), 3.30-3.50 (4H, m), 3.60-3.65 (3H, s), 3.70 (2H, dt),
4.30 (2H, t), 7.20 (1H, s), 7.41 (1H, d), 7.55-7.70 (4H, m), 7.85
(2H, d), 8.75 (1H, t), 9.65 (1H, s).
ter t-Bu tyl [2-(5-Cya n o-2-iod op h en oxy)eth yl]ca r ba m ic
Acid (11a ). Compound 10 (28.7 g, 117 mmol), 17a (25 g, 140
mmol), K2CO3 (48.4 g, 350 mmol), and KI (39 g, 235 mmol)
were treated as described for 6a , Part A, to give 11a (37 g,
1
81%). H NMR (CDCl3) δ 1.45 (9H, s), 3.62 (2H, dt), 4.10 (2H,
t), 5.05 (1H, br), 6.96-7.06 (2H, m), 7.90 (1H, d).
Ben zyl (3R)-3-ter t-Bu t oxyca r b on yla m in o-4-(5-cya n o-
2-iod op h en oxy)bu tyr ic Acid (11b). N,N,N′,N′-Tetramethyl-
azodicarboxamide28 (7.4 g, 42.7 mmol) was added to a mixture
of 17b (10.16 g, 32.8 mmol), 10 (10.5 g, 42.7 mmol), and
triphenylphosphine (11.2 g, 42.7 mmol) in toluene (100 mL)
at 0 °C. The reaction mixture was stirred overnight and
concentrated in vacuo. The residue was purified by chroma-
tography (EtOAc/hexane) to give 11b (11.9 g, 67%): 1H NMR
(CDCl3) δ 2.93 (1H, dd), 3.08 (1H, dd), 4.13 (1H, dd), 4.30 (1H,
dd), 4.94 (1H, m), 5.14 (1H, d), 5.19 (1H, d), 6.96 (1H, d), 7.03
(1H, dd), 7.28-7.33 (5H, m), 7.72 (2H, d), 7.86 (2H, d), 7.88
(1H, d).
Meth yl 2-Aceta m id o-3-[2-[(2R)-2-[4-(1-ter t-bu toxyca r -
bon ylp ip er id yl-4-oxy)ben zoyl]a m in o-3-ben zyloxyca r bo-
n ylpr opoxy]-4-cyan oph en yl]acr ylate (13e). Compound 19b
was treated as described for 13b to give 13e: 1H NMR (CDCl3)
δ 1.50 (9H, s), 1.65-2.05 (4H, m), 2.00 (3H, s), 2.80-3.10 (2H,
m), 3.30-3.40 (2H,m), 3.60-3.80 (2H, m), 3.80 (3H, s), 4.25
(2H, d), 4.55 (1H, m), 4.85 (1H, m), 5.15 (2H, s), 6.80-7.90
(15H, m).
Meth yl 2-Aceta m id o-3-[2-[(2R)-2-[1-(4-p yr id yl)p ip er i-
d in e-4-ca r bon yl]a m in o-3-ben zyloxyca r bon ylp r op oxy]-4-
cya n op h en yl]a cr yla te Tr iflu or oa ceta te (13f). Compound
19b (2.84 g, 5.9 mmol), 1-(4-pyridinyl)piperidine-4-carboxylic
acid (18d ) (1.4 g, 5.9 mmol), bromotripyrrolidinophosphonium
hexafluorophosphate (2.80 g, 6.0 mmol), TEA (3.3 mL, 23.7
mmol), and DMF (50 mL) were treated as described for 13c
to give 13f (1.78 g, 40%): 1H NMR (DMSO-d6) δ 1.45-1.65
(2H, m), 1.70-1.85 (2H, m), 1.95 (3H, s), 2.55-2.80 (2H, m),
3.16 (1H, m), 3.24 (1H, br d), 3.60 (2H, m), 3.67 (3H, s), 4.05
(1H, m), 4.17 (2H, m), 4.50 (1H, m), 5.09 (2H, d), 7.18 (3H, m),
7.32-7.40 (5H, m), 7.45 (1H, m), 7.58 (1H,m), 7.70 (1H, m),
8.10 (1H, d), 8.22 (2H, d), 9.69 (1H, br).
3-[4-Am id in o-2-[2-(4-a m id in ob en zoyla m in o)et h oxy]-
p h en yl]-2-oxop r op ion ic Acid Bist r iflu or oa cet a t e (3a ).
P a r t A. Compound 13a (120 mg, 0.3 mmol) was treated as
described for 1d to give the crude amidine.
P a r t B. The crude amidine was diluted with 3 N HCl. The
mixture was stirred at 80°C and concentrated in vacuo.
Purification by reversed-phase HPLC gave 3a (65 mg, 34%):
1H NMR (DMSO-d6) δ 3.72 (2H, dt), 4.15 (2H, s, keto form),
4.31 (2H, t), 6.81 (1H, s, enol form), 7.36-7.48 (2H, m), 7.88
(2H, d), 8.04 (2H, d), 8.33 (1H, d), 9.02 (1H, t), 9.07 (2H, m),
9.25 (4H, br), 9.40 (2H, br). MS (FAB): 412 (MH+). Anal.
(C20H21N5O5‚2.0TFA‚1.0H2O) C, H, N.
Meth yl 2-Aceta m id o-3-[2-(2-ter t-bu toxyca r bon yla m i-
n oeth oxy)-4-cya n op h en yl]a cr yla te (12a ). A mixture of 11a
(15.33 g, 39.5 mmol), TEA (7.99 g, 79.0 mmol), methyl
2-acetamidoacrylate (9.05 g, 63.2 mmol), tri-o-tolylphosphine
(7.30 g, 24 mmol), and Pd(OAc)2 in CH3CN (150 mL) was
heated to reflux overnight. The reaction mixture was filtered
through Celite, and the filtrate was concentrated in vacuo. The
residue was diluted with EtOAc, and the organic layer was
washed with 1 N HCl and saturated NaCl and worked up. The
crude product was washed with EtOAc/hexane to give 12a
(9.63 g, 60%): 1H NMR (DMSO-d6) δ 1.38 (9H, s), 1.95 (3H,
s), 3,35 (2H, dt), 3.70 (3H, s), 4.10 (2H, t), 7.03 (1H, t), 7.20
(1H, s), 7.43 (1H, d), 7.55 (1H, s), 7.68 (1H, d), 9.65 (1H, s).
Meth yl 2-Acetam ido-3-[2-(2-am in oeth oxy)-4-cyan oph en -
yl]a cr yla te HCl (19a ). Compound 12a (9.63 g, 23.9 mmol)
was treated as described for 6a , Part B, to give 19a (6.31 g,
78%): 1H NMR (DMSO-d6) δ 1.95 (3H, s), 3.25 (2H, dt), 3.70
(3H, s), 4.30 (2H, t), 7.28 (1H, s), 7.48 (1H, d), 7.62 (1H, s),
7.70 (1H, d), 8.20 (3H, br), 9.75 (1H, s).
Meth yl 2-Acetam ido-3-[(2-bu toxycar bon ylam in o-3-ben -
zyloxyca r bon ylp r op oxy)-4-cya n op h en yl]a cr yla te (12b).
Compound 11b (12.0 g, 22.4 mmol), TEA (6.24 mL), methyl
2-acetamidoacrylate (6.40 g, 44.7 mmol), tri-o-tolylphosphine
(1.47 g, 4.83 mmol), Pd(OAc)2 (0.49 g, 10 mol %), and DMF
(100 mL) were treated as described in 12a to give 12b (9.8 g,
79%): 1H NMR (DMSO-d6) δ 1.35 (9H, s), 1.95 (3H, s), 2.61-
2.73 (2H, m), 3.68 (3H, s), 4.05 (2H, m), 4.24 (1H, m), 5.10
(2H, s), 7.11 (1H, d), 7.19 (1H, s), 7.35 (5H, m), 7.44 (1H, d),
7.56 (1H, s), 7.67 (1H, d), 9.69 (br, 1H).
Meth yl 2-Aceta m id o-3-[(2-a m in o-3-ben zyloxyca r bon -
ylp r op oxy)-4-cya n op h en yl]a cr yla te HCl (19b). 12b (530
mg, 0.96 mmol) was treated as described for 6a , Part B, to
give 19b (470 mg, quant.): MS (ESI): 452 (MH+).
3-[4-Am id in o-2-[2-[4-(1-a cet im id oyl-4-p ip er id yloxy)-
ben zoyla m in o]eth oxy]p h en yl]-2-oxop r op ion ic Acid Bis-
tr iflu or oa ceta te (3b). P a r t A. Compound 13b (10.3 g, 17.0
mmol) was treated as described in 1d to give crude ethyl
2-acetamido-3-[2-[2-[4-(4-piperidyloxy)benzoylamino]ethoxy]-
4-cyanophenyl]acrylate.
P a r t B. The crude product of Part A was treated as
described for 1b, Part B to give crude methyl 2-acetamido-3-
[2-[2-[4-(1-acetimidoyl-4-piperidyloxy)benzoylamino]ethoxy]-4-
cyanophenyl]acrylate.
P a r t C. The crude product in Part B was treated as
described for 3a , Part B, to give 3b (4.09 g, 33%): 1H NMR
(DMSO-d6) δ 1.65-1.85 (2H, m), 2.02-2.19 (2H, m), 2.25 (3H,
s), 3.58-3.82 (6H, m), 4.23 (2H, s, keto form), 4.30 (2H, t), 4.79
(1H, m), 6.80 (1H, s, enol form), 7.07 (2H, d), 7.38-7.47 (2H,
Meth yl 2-Acetam ido-3-[2-[2-(4-cyan oben zoylam in o)eth -
oxy]-4-cya n op h en yl]a cr yla te (13a ). 4-Cyanobenzoic acid
(18c) (44 mg, 0.3 mmol), 19a (100 mg, 0.3 mmol), TEA (61
mg, 0.6 mmol), HOBT (46 mg, 0.3 mmol), EDC (58 mg, 0.3
mmol), and DMF (5 mL) were treated as described for 7c to
give 13a (120 mg, 92%): 1H NMR (DMSO-d6) δ 1.95 (3H, s),
3.65 (3H, s), 3.70 (2H, dt), 4.25 (2H, t), 7.20 (1H, s), 7.45 (1H,
d), 7.65 (1H, s), 7.70 (1H, d), 8.00 (4H, s), 8.90 (1H, t), 9.65
(1H, s).
Meth yl 2-Aceta m id o-3-[2-[2-[4-(1-ter t-bu toxyca r bon yl-
4-p ip er id yloxy)b en zoyla m in o]et h oxy]-4-cya n op h en yl]-
a cr yla te (13b). Compound 18a (6.8 g, 21 mmol), 19a (6.75 g,
20 mmol), HOBT (3.42 g, 21 mmol), TEA (9 mL), EDC (58 mg,
0.3 mmol), and DMF (80 mL) were treated as described for 7c
to give 13b (10.3 g, 85%): 1H NMR (CDCl3) δ 1.45 (9H,s), 1.65-

Relationships of Novel Factor Xa Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1855
m), 7.83 (2H, d), 8.33 (1H, d), 8.55-8.67 (2H, m), 9.05-9.34
(5H, br), 9.75 (1H, br, enol form). MS (ESI): 510 (MH+). Anal.
(C26H31N5O6‚2.0TFA‚1.0H2O) C, H, N.
Eth yl 3-[2-[2-[4-(1-ter t-Bu toxycar bon yl-4-piper idyloxy)-
ben zoylam in o]eth oxy]-4-cyan oph en yl]acr ylate (15a). Com-
pound 20 (2.24 g, 7.56 mmol), 18a (2.67 g, 8.32 mmol), EDC
(1.59 g, 8.32 mmol), HOBT (1.12 g, 8.32 mmol), TEA (3.16 mL,
22.7 mmol), and CH2Cl2 (50 mL) were treated as described
for 7c to give 15a (3.0 g, 71%): 1H NMR (CDCl3) δ 1.33 (3H,
t), 1.47 (9H, s), 1.64-1.79 (2H, m), 1.86-1.98 (2H, m), 3.24-
3.42 (2H, m), 3.60-3.73 (2H, m), 3.92 (2H, dt), 4.24 (2H, q),
4.28 (2H, t), 4.45-4.53 (1H, m), 6.57 (1H, d), 6.77 (1H, t), 6.88
(2H, d), 7.18 (1H, d), 7.23 (1H, d), 7.58 (1H, d), 7.77 (2H, d),
7.97 (1H, d).
Eth yl 3-[2-[2-[1-(4-P yr id yl)p ip er id in e-4-ca r bon yla m i-
n o]eth oxy]-4-cya n op h en yl]a cr yla te (15b). Compound 20
(2.47 g, 8.33 mmol), 1-(4-pyridyl)piperidine-4-carboxylic acid
HCl (18d ) (2.22 g, 9.17 mmol), bromotripyrrolidinophospho-
nium hexafluorophosphate (4.27 g, 9.17 mmol), TEA (3.48 mL,
25.0 mmol), and DMF (50 mL) were treated as described for
13c. The crude product was purified by chromatography to give
15b (2.3 g, 62%): 1H NMR (DMSO-d6) δ 1.26 (3H, t), 1.50-
1.68 (2H, m), 1.68-1.73 (2H, m), 2.62-2.68 (1H, m), 2.94-
3.06 (2H, m), 3.40-3.53 (2H, m), 3.95-4.25 (6H, m), 6.76 (1H,
dd), 6.94 (2H, d), 7.44 (1H, dd), 7.62 (1H, br), 7.83 (1H, dd),
7.90 (1H, d), 9.01 (1H, t), 8.15 (2H, d).
3-[4-Am id in o-2-[2-[1-(p yr id in -4-yl)p ip er id in e-4-ca r bo-
n yla m in o]eth oxy]p h en yl]-2-oxop r op ion ic Acid Bistr i-
flu or oa ceta te (3c). Compound 13c (360 mg, 0.73 mmol) was
treated as described for 3a to give 3c (205 mg, 41%): 1H NMR
(DMSO-d6) δ 1.50-1.70 (2H, m), 1.80-1.95 (2H, m), 2.60 (1H,
m), 3.18-3.30 (2H, m), 3.43-3.60 (2H, m), 4.10-4.30 (4H, m),
6.80 (1H, s, enol form), 7.18 (2H, d), 7.35-7.48 (2H, m), 8.22
(2H, d), 8.34 (1H, d), 9.15 (2H, br), 9.30 (2H, br), 9.80 (1H, br,
enol form). MS (ESI): 454 (MH+). Anal. (C23H27N5O5‚2.0TFA‚
2.0H2O) C, H, N.
3-[4-Am id in o-2-[2-[4-(p yr r olid in e-1-ca r bon yl)ben zoyl-
a m in o]eth oxy]p h en yl]-2-oxop r op ion ic Acid Tr iflu or o-
a ceta te (3d ). Compound 13d (1.80 g, 3.68 mmol) was treated
as described for 3a to give 3d (345 mg, 16%): 1H NMR (DMSO-
d6) δ 1.75-1.95 (4H, m), 3.30-3.50 (4H, m), 3.72 (2H, dt), 4.25
(2H, s, keto form), 4.30 (2H, t), 6.81 (1H, s, enol form), 7.35-
7.50 (2H, m), 7.60 (2H, d), 7.89 (2H, d), 8.33 (1H, d), 8.85 (1H,
br), 9.01 (2H, br), 9.27 (2H, br), 9.80 (1H, br, enol form). MS
(ESI): 467 (MH+). Anal. (C24H26N4O6‚TFA‚2.0H2O) C, H, N.
3-[2-[2-[(1-Acetim id oyl-4-p ip er id yloxy)ben zoyla m in o]-
eth oxy]-4-a m id in op h en yl]p r op ion ic Acid Bistr iflu or o-
a ceta te (4a ). P a r t A. Compound 15a (3.0 g, 5.33 mmol) was
treated as described for 1d to give crude ethyl 3-[2-[2-[4-(4-
piperidyloxy)benzoylamino]ethoxy]-4-amidinophenyl]acryl-
ate. The crude product was treated as described for 1b, Part
B, to give ethyl 3-[2-[2-[(1-acetimidoyl-4-piperidyloxy)benzoyl-
amino]ethoxy]-4-amidinophenyl]acrylate bistrifluoroacetate
(16a ) (2.3 g, 37%): 1H NMR (DMSO-d6) δ 1.23 (3H, t), 1.67-
1.87 (2H, m), 2.00-2.25 (2H, m), 2.29 (3H, s), 3.45-3.60 (2H,
m), 3.66-3.77 (4H, m), 4.17 (2H, q), 4.34 (2H,t), 4.73-4.76 (1H,
m), 6.79 (1H, d), 7.05 (2H, t), 7.43 (1H, d), 7.56 (1H, d), 7.84
(2H, d), 7.92 (1H, br), 7.97 (1H, d), 8.64 (2H, br), 9.19 (1H, br),
9.37 (4H, br). MS (ESI, m/z): 522 (MH+).
P a r t B. A mixture of 10% Pd-C (100 mg) and 16a (1.2 g,
1.60 mmol) in methanol (50 mL) was stirred under a hydrogen
atmosphere overnight. The reaction mixture was filtered
through Celite, and the filtrate was concentrated in vacuo. The
residue was purified by reversed-phase HPLC to give ethyl
3-[2-[2-[(1-acetimidoyl-4-piperidyloxy)benzoylamino]ethoxy]-4-
amidinophenyl]propionate bistrifluoroacetate (1.1 g, 91%): 1H
NMR (DMSO-d6) δ 1.16 (3H, t), 1.67-1.76 (2H, m), 2.00-2.25
(2H, m), 2.29 (3H, s), 2.58 (2H, t), 2.90 (2H, t), 3.46-3.58 (2H,
m), 3.63-3.84 (4H, m), 3.98 (2H, q), 4.23 (2H, t), 4.74-4.87
(1H, m), 7.08 (1H, d), 7.36 (1H, br), 7.38 (2H, d), 7.84 (2H, d),
8.61 (2H, br), 9.11 (2H, br), 9.16 (1H, br), 9.24 (2H, br). MS
(ESI, m/z): 524 (MH+).
P a r t C. A mixture of the product from Part B (600 mg, 0.8
mmol) and hydrochloric acid (10 mL) was stirred at 50 °C for
4 h and concentrated in vacuo. The residue was purified by
reversed-phase HPLC to give 4a (480 mg, 77%): 1H NMR
(DMSO-d6) δ 1.66-1.87 (2H, m), 2.02-2.17 (2H, m), 2.29 (3H,
s), 2.52 (2H, t), 2.78 (2H, t), 3.44-3.62 (2H, m), 3.64-3.75 (4H,
m), 4.42 (2H, t), 4.74-4.86 (1H, m), 7.06 (2H, d), 7.37 (1H,
br), 7.39 (2H, d), 7.85 (2H, d), 8.64 (2H, br), 9.19 (1H, d), 9.23
(2H, br), 9.25 (2H, br). MS (ESI, m/z): 496 (MH+). Anal.
(C26H33N5O5‚2.0TFA‚2.0H2O) C, H, N.
3-[2-[2-[1-(4-P yr id yl)p ip e r id in e -4-ca r b on yla m in o]-
eth oxy]-4-a m id in op h en yl]p r op ion ic Acid Bistr iflu or o-
a ceta te (4b). P a r t A. Compound 15b (2.3 g, 5.13 mmol) was
treated as described for 1d to give the crude amidine (16b).
The amidine was treated as described for 4a , Part B, to give
ethyl 3-[2-[2-[1-(4-pyridyl)piperidine-4-carbonylamino]ethoxy]-
4-amidinophenyl]propionate bistrifluoroacetate (1.5 g, 42%):
1H NMR (DMSO-d6) δ 1.15 (3H, t), 1.50-1.67 (2H, m), 1.76-
1.81 (2H, m), 2.52-2.60 (1H, m), 2.62 (2H, dd), 2.89 (2H, dd),
3.15-3.28 (2H, m), 3.49 (2H, dt), 4.03 (2H, q), 4.12 (2H, t),
4.20 (2H, d), 7.19 (2H, d), 7.37 (3H, br), 8.18 (1H, d), 8.21 (2H,
d), 9.23 (2H, br), 9.25 (2H, br). MS (ESI, m/z): 468 (MH+).
(3R)-4-[5-Am id in o-2-(2-ca r boxy-2-oxoeth yl)p h en oxy]-
3-[4-(1-a cetim id oylp ip er id yl-4-oxy)ben zoyla m in o]bu tyr -
ic Acid Bistr iflu or oa ceta te (3e). P a r t A. Compound 13e
(12.0 g, 16.0 mmol) was treated as described for 1d to give
methyl 2-acetamido-3-[4-amidino-2-[(2R)-3-ethoxycarbonyl-2-
[4-(piperidyl-4-oxy)benzoylamino]propoxy]phenyl]acrylate bis-
trifluoroacetate (3.35 g, 25%): 1H NMR (DMSO-d6) δ 1.16 (3H,
t), 1.72-1.89 (2H, m), 1.95 (3H, s), 2.04-2.17 (2H, m), 2.79
(2H, d), 3.03-3.18 (2H, m), 3.20-3.32 (2H, m), 3.66 (3H, s),
4.04 (2H, q), 4.23 (2H, d), 4.66-4.79 (1H, m), 7.06 (2H, d), 7.28
(1H, s), 7.44 (1H, dd), 7.53 (1H, d), 7.73 (1H, d), 7.81 (2H, d),
8.46 (1H, d), 9.18 (2H, s), 9.34 (2H, s), 9.67 (1H, br s).
P a r t B. The product of Part A (3.35 g, 5.5 mmol) was treated
as described for 1b, Part B, and subsequently the crude
product was treated as described for 3a , Part B, to give 3e
(1.73 g, 39%): 1H NMR (DMSO-d6) δ 1.70-1.87 (2H, m), 2.02-
2.15 (2H, m), 2.29 (3H, s), 2.68 (1H, d), 2.74 (1H, d), 3.48-
3.58 (2H, m), 4.11-4.28 (2H, m), 4.60-4.72 (1H, m), 4.81 (1H,
br), 6.80 (1H, s, enol form), 7.07 (2H, d), 7.45 (1H, d), 7.48 (1H,
s), 7.82 (2H, d), 8.33 (1H, d), 8.62 (1H, s), 9.14 (2H, s), 9.17
(1H, s), 9.26 (2H, s). MS (ESI): 568 (MH+). Anal. (C28H33N5O8‚
2.0TFA‚1.5H2O) C, H, N.
(3R)-4-[5-Am id in o-2-(2-ca r boxy-2-oxoeth yl)p h en oxy]-
3-[1-(p yr id in -4-yl)p ip er id in e-4-ca r b on yla m in o]b u t yr ic
Acid Bistr iflu or oa ceta te (3f). Compound 13f (1.78 g, 2.8
mmol) was treated as described for 3a to give 3f (745 mg,
43%): 1H NMR (DMSO-d6) δ 1.79 (2H, m), 1.82 (2H, m), 2.55
(2H, m), 3.25-3.70 (5H, m), 4.00-4.25 (4H, m), 4.45 (1H, m),
6.76 (1H, s, enol), 7.20 (2H, m), 7.42 (3H, m), 8.23 (2H, d),
8.35 (1H, d), 9.15 (2H, br), 9.28 (2H, br), 9.80 (1H, br). MS
(ESI): 512 (MH+). Anal. (C25H29N5O7‚2.0TFA‚1.5H2O) C, H,
N.
Eth yl 3-[2-[2-(1-ter t-Bu toxyca r bon yla m in o)eth oxy]-4-
cya n op h en yl]a cr yla te (14). A mixture of 11a (11.0 g, 28.4
mmol), ethyl acrylate (15.4 mL, 142 mmol), TEA (20 mL, 142
mmol), and Pd(OAc)2 (127 mg, 0.57 mmol) in DMF (200 mL)
was stirred overnight at 100 °C. The reaction mixture was
conentrated in vacuo and diluted with EtOAc. The solution
was washed with 1 N HCl, saturated NaHCO3 and saturated
NaCl, and then worked up. The residue was purified by
chromatography to give 14 (9.6 g, 94%): 1H NMR (CDCl3) δ
1.38 (3H, t), 1.46 (9H, s), 3.62 (2H, dt), 4.16 (2H, t),4.28 (2H,
q), 6.56 (1H, d), 7.16 (1H, d), 7.27 (1H, d), 7.60 (1H, d), 7.96
(1H, d).
Eth yl 3-[2-(2-am in oeth oxy)-4-cyan oph en yl]acr ylate HCl
(20). Compound 14 was treated as described for 6a , Part B,
to give 20: 1H NMR (DMSO-d6) δ 1.27 (3H, t), 3.25 (2H, br),
4.20 (2H, q), 4.40 (2H, t), 6.75 (1H, d), 7.50 (1H, d), 7.65 (1H,
d), 7.95 (1H, d), 8.00 (1H, d), 8.40 (3H, br).
P a r t B. The product of Part A (250 mg, 0.36 mmol) was
treated as described for 4a , Part C, to give 4b (480 mg, 92%):

1856 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10
Sagi et al.
1H NMR (DMSO-d6) δ 1.50-1.67 (2H, m), 1.76-1.92 (2H, m),
2.54 (2H, dd), 2.55-2.67 (1H, m), 2.88 (2H, dd), 3.12-3.29 (2H,
m), 3.49 (2H, dt), 4.12 (2H, t), 4.20 (2H, d), 7.18 (2H, d), 7.36
(2H, br), 7.37 (1H, d), 8.18 (1H, d), 8.20 (2H, d), 9.14 (2H, br),
9.24 (2H, br). MS (ESI, m/z): 440 (MH+). Anal. (C23H29N5O4‚
2.0TFA) C, H, N.
En zym e Assa ys. Human fXa and thrombin were obtained
from Enzyme Research Laboratories, Inc. and Sigma Chemical
Co., respectively. The chromogenic substrates used were
S-2222 (Daiichi Pure Chemical, J apan) for fXa and S-2238
(Daiichi Pure Chemical) for thrombin. FXa and thrombin were
assayed in a buffer containing 0.1 M Tris, 0.2 M NaCl, 0.1%
PEG-6000, and 0.02% Tween20 at pH 8.4.
(2) Much effort has been focused on orally available fXa inhibitors:
(a) Al-Obeidi, F.; Ostrem, J A. Factor Xa inhibitors. Exp. Opin.
Ther. Patents 1999, 9, 931. (b) Betz, A. Recent advances in factor
Xa inhibitors. Exp. Opin. Ther. Patents 2001, 11, 1007.
(3) (a) Hirsh, J .; Fuster, V. Guide to anticoagulant therapy. Part 2:
Oral anticoagulants. Circulation 1994, 89, 1469-1480. (b)
Amerena, J .; Mashford, M. L.; Wallace, S. Adverse effects of
anticoagulants. Adverse Drug React. Acute Poisoning Rev. 1990,
9, 1-36.
(4) (a) Neuhaus, K.; Von Essen, R.; Tebbe, U.; J essel, A.; Heinrichs,
H.; Ma¨urer, W.; Do¨ring, W.; Harmjanz, D.; Ko¨tter, V.; Kalham-
mer, E.; Simon, H.; Horacek, T. Safety observations from the
pilot phase of the randomized r-hirudin for improvement of the
thrombolysis (HIT-III) study. Circulation 1994, 90, 1638-1642.
(b) GUST IIA investigators. Randomized trial of intravenous
heparin versus recombinant hirudin for acute coronary syn-
dromes. The global use of strategies to open occluded coronary
arteries (GUSTO) IIa investigators. Circulation 1994, 90, 1631-
1637. (c) Antman, E. Hirudin in acute myocardial infarction.
Safety report from the thrombolysis and thrombin inhibition in
myocardial infarction (TIMI) 9A trial. Circulation 1994, 90,
1624-1630.
(5) (a) Sitko, G. R.; Ramjit, D. R.; Stabilito, I. I.; Lehman, D.; Lynch,
J . J .; Vlasuk, G. P. Conjunctive enhancement of enzymatic
thrombolysis and prevention of thrombotic reocclusion with the
selective factor Xa inhibitor, tick anticoagulant peptide. Com-
parison to hirudin and heparin in a canine model of acute
coronary artery thrombosis. Circulation 1992, 85, 805-815. (b)
Herbert, J . M.; Bernat, A.; Dol, F.; Herault, J . P.; Crepon, B.;
Lormeau, J . C. DX-9065a, a novel, synthetic selective and orally
active inhibitor of factor Xa: In vitro and in vivo studies. J .
Pharmacol. Exp. Ther. 1996, 276, 1030-1038. (c) Harker, L. A.;
Hanson, S. R.; Kelly, A. B. Antithrombotic benefits and hemor-
rhagic risks of direct thrombin antagonists. Thrombosis Hae-
mostasis 1995, 74, 464-472. (d) Kaiser, B. Factor Xa versus
factor IIa inhibitors. Clin. Appl. Thromb./Hemostasis 1997, 3,
16-24.
The final substrate concentrations in the reaction were 200
and 100 µM for S-2222 and S-2238, respectively. All enzyme
assays were carried out at room temperature in 96-well
microtiter plates with final concentrations of 4.5 nM for fXa
and 1.1 nM for thrombin. Compounds at appropriate dilutions
were added to wells containing buffer and enzyme and
preincubated for 10 min. The enzyme reactions were initiated
by the addition of substrate, and color development, due to
the release of p-nitroaniline from each chromogenic substrate,
was monitored continuously for 16 min at 405 nm on a
microplate reader (model 3550-UV, Bio-Rad). The measured
initial velocities were used to determine the amount of
inhibitor required to reduce the control velocity by 50%; this
concentration was defined as the IC50 of the inhibitor. The
apparent Ki values of compounds were then calculated accord-
ing to the relationship: (IC50)1/(IC50
)
) (Ki)1/(Ki)2.22 In this
2
equation, the Ki value of 3e, which was determined from a
double reciprocal plot,23 was used as the standard.
Coa gu la tion Assa y. PT was measured with coagulometer
CA3000 (Sysmex, J apan). Citrated human plasma, which was
collected from healthy volunteers, and cynomolgus monkey
plasma (A.T. VIRI, Inc., the Philippines) were used in the
assays. Plasma (45 µL) was mixed with 5 µL of compound
dilutions followed by the automatic addition of 100 µL of
prothrombin reagent (Thromboplastin C plus, Dade Behring).
The concentrations required to double the clotting time (2 ×
PT) were estimated from each dose-response curve.
An ticoa gu la n t a n d P h a r m a cok in etic Stu d ies in Cy-
n om olgu s Mon k eys. Male cynomolgus monkeys were fasted
overnight. Compounds were administered orally at 10 mg/kg
using a gastric tube and intravenously at 0.3 mg (3b, 3c, and
4b) or 1.0 mg/kg (3e). Blood (1.5 mL) was collected from the
saphenous vein in the leg prior to treatment and at multiple
time points for up to 24 h after treatment. Collected blood
samples were immediately transferred into sample tubes
containing 167 µL of 3.8% trisodium citrate solution, and
platelet-poor plasma was prepared by centrifugation to mea-
sure PT. Plasma drug concentrations were also measured by
HPLC.
(6) (a) Tidwell, R. R.; Webster, W. P.; Shaver, S. R.; Geratz, J . D.
Strategies for anticoagulation with synthetic protease inhibi-
tors: Xa inhibitors versus thrombin inhibitors. Thromb. Res.
1980, 19, 339-349. (b) Sturzebecher, J .; Sturzebecher, U.;
Vieweg, H.; Wagner, G.; Hauptmann, J .; Markwardt, F. Syn-
thetic inhibitors of bovine factor Xa and thrombin: Comparison
of their anticoagulant efficiency. Thromb. Res. 1989, 54, 245-
252.
(7) Padmanabhan, K.; Padmanabhan, K. P.; Tulinsky, A.; Park, C.
H.; Bode, W.; Huber, R.; Blankenship, D. T.; Cardin, A. D.; Kisiel,
W. Structure of human des(1-45) factor Xa at 2.2 A resolution.
J . Mol. Biol. 1993, 232, 947-966.
(8) The use of trypsin as a surrogate enzyme for fXa is a well-
established practice: (a) Renatus, M.; Bode, W.; Huber, R.;
Sturzebecher, J .; Stubbs, M. T. Structural and functional
analyses of benzamidine-based inhibitors in complex with
trypsin: Implications for the inhibition of factor Xa, tPA, and
urokinase. J . Med. Chem. 1998, 41, 5445-5456. (b) Choi-
Sledeski, Y. M.; McGarry, D. G.; Green, D. M.; Mason, H. J .;
Becker, M. R.; Davis, R. S.; Ewing, W. R.; Dankulich, W. P.;
Manetta, V. E.; Morris, R. L.; Spada, A. P.; Cheney, D. L.; Brown,
K. D.; Colussi, D. J .; Chu, V.; Heran, C. L.; Morgan, S. R.;
Bentley, R. G.; Leadley, R. J .; Maignan, S.; Guilloteau, J .-P.;
Dunwiddie, C. T.; Pauls, H. W. Sulfonamidopyrrolidinone factor
Xa inhibitors: Potency and selectivity enhancements via P-1 and
P-4 optimization. J . Med. Chem. 1999, 42, 3572-3587. (c)
Phillips, G. B.; Buckman, B. O.; Davey, D. D.; Eagen, K. A.;
Guilford, W. J .; Hinchman, J .; Ho, E.; Koovakkat, S.; Liang, A.;
Light, D. R.; Mohan, R.; Ng, H. P.; Post, J . M.; Shaw, K. J .;
Smith, D.; Subramanyan, B.; Sullivan, M. E.; Trinh, L.; Vergona,
R.; Walters, J .; White, K.; Whitlow, M.; Wu, S.; Xu, W.;
Morrissery, M. M. Discovery of N-[2-[5-[amino(imino)methyl]-
2-hydroxyphenoxy]-3,5-difluoro-6-[3-(4,5-dihydro-1-methyl-1H-
imidazol-2-yl)phenoxy]pyridine-4-yl]-N-methylglycine (ZK-807834):
A potent, selective, and orally active inhibitor of the blood
coagulation enzyme factor Xa. J . Med. Chem. 1998, 41, 3557-
3562.
Ack n ow led gm en t. We thank members of the Ana-
lytical Chemistry Department for mass spectral mea-
surements.
Refer en ces
(1) Parts of this work have already been presented in the following
publications: (a) Takayanagi, M.; Sagi, K.; Nakagawa, T.;
Yamanashi, M.; Kayahara, T.; Takehana, S.; Fukuda, Y.; Ta-
kahashi, M.; Shoji, M. WO 9831661, 1998. (b) Nakagawa, T.;
Sagi, K.; Yoshida, K.; Fukuda, Y.; Shoji, M.; Takehana, S.;
Kayahara, T.; Takahara, A. WO 9964392, 1999. (c) Takahashi,
M.; Sagi, K.; Yamanashi, M.; Makino, S.; Suzuki, N.; Takay-
anagi, M.; Nakagawa, T.; Ishikawa, K.; Shoji, M. Docking study
and structure-based design of novel factor Xa inhibitor having
augumented interaction with oxy-anion hole. Presented at the
13th European Symposium on Quantitative Structure-Activity
Relationships, Du¨sseldorf, August 2000. (d) Kayahara, T.;
Fukuda, Y.; Takehana, S.; Takahara, A.; Yamanashi, M.; Sagi,
K.; Takayanagi, M.; Nakagawa, T.; Shoji, M.; Iwata, S. The in
vitro anticoagulant activities of a novel selective factor Xa
inhibitor AX1826; and Kayahara, T.; Shima, Y.; Takehana, S.;
Takahara, A.; Gonsho, A.; Shoji, M.; Iwata, S. Oral anticoagulant
effects of a selective FXa inhibitor AX1826 in cynomolgus
monkeys. 73rd Annual Meeting of the J apanese Pharmacological
Society, Yokohama, March 2000; P-375 and P-376, respectively.
(9) FXa inhibitors containing the acetimidoylpiperidyloxy group
have been reported: (a) Nagahara, T.; Kanaya, N.; Inamura,
K.; Yokoyama, Y. EP 540051, 1993. (b) Hirayama, F.; Koshio,
H.; Matsumoto, Y.; Kawasaki, T.; Kaku, S.; Yanagisawa, I. WO
9616940, 1996.
(10) FXa inhibitors containing the pyridinylpiperidinyl group have
been reported: Faull, A. W.; Mayo, C. M.; Preston, J .; Stocker,
A. WO 9610022, 1996.
(11) Residues are numbered with chymotrypsin-based numbering:
Bode, W.; Mayr, I.; Baumann, U.; Huber, R.; Stone, S. R.;
Hofsteenge, J . The refined 1.9 Å crystal structure of human
R-thrombin: Interaction with D-Phe-Pro-Arg chloromethyl ke-
tone and significance of the Tyr-Pro-Pro-Trp insertion segment.
EMBO J . 1989, 8, 3467-3475.

Relationships of Novel Factor Xa Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 10 1857
(12) Banner, D. W.; Hadvary, P. Crystallographic analysis at 3.0-A
resolution of the binding to human thrombin of four active site-
directed inhibitors. J . Biol. Chem. 1992, 266, 20085-20093.
(13) (a) Markwardt, F.; Hauptmann, J .; Richter, M.; Richer, P.
Tierexperimentelle untersuchungen zur phamakokinetik von
4-amidinophenylbrenztraubensa¨ure. (Animal experiments on the
pharmacokinetics of 4-amidinophenylpyruvic acid (APPA)). Phar-
mazie 1979, 34, 178-181. (b) Nagahara, T.; Yokoyama, Y.;
Inamura, K.; Katakura, S.; Komoriya, S.; Yamaguchi, H.; Hara,
T.; Iwamoto, M. Dibasic (amidinoaryl)propanoic acid derivatives
as novel blood coagulation factor Xa inhibitors. J . Med. Chem.
1994, 37, 1200-1207.
(21) The presence of a reactive electrophilic site in inhibitors might
increase the likelihood of pharmacological problems. (a) Sand-
erson, P. E. Small, noncovalent serine protease inhibitors. Med.
Res. Rev. 1999, 19, 179-197. (b) Leung, D.; Abbenante, G.;
Fairlie, D. P. Protease inhibitors: Current status and future
prospects. J . Med. Chem. 2000, 43, 305-341.
(22) Cheng, Y. C.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes
50% inhibition (IC50) of an enzymatic reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(23) Lineweaver, H.; Burk D. The determination of enzyme dissocia-
tion constants. J . Am. Chem. Soc. 1934, 56, 658-666.
(24) SYBYL User’s Manual, ver. 6.2; Tripos, Inc.: St. Louis, MO,
1995.
(25) Takahashi, M.; Shoji, M.; Makino, S. Docking mode search of
flexible ligand to binding site of protein. 24th Symposium on
Structure-Activity Relationships, Osaka, November 1996;
13SP05.
(26) Bartunik, H. D.; Summers, L. J .; Bartsch, H. H. Crystal structure
of bovine beta-trypsin at 1.5 Å resolution in a crystal form with
low molecular packing density. Active site geometry, ion pairs
and solvent structure. J . Mol. Biol. 1989, 210, 813-828.
(27) Bru¨nger, A. T.; Kuriyan, J .; Karplus, M. Crystallographic
R-factor refinement by molecular dynamics. Science 1987, 235,
458-460.
(28) Tsunoda, T.; Yamamiya, Y.; Kawamura, Y.; Ito, S. Mitsunobu
acylation of sterically congested secondary alcohols by N,N,N′,N′-
tetramethylazodicarboxamide-tributylphosphine reagents. Tet-
rahedron Lett. 1995, 36 (14), 2529-2530.
(29) 3-(3-Benzoylaminopropoxy)benzamidine is reported to be a weak
inhibitor of bovine fXa, and the Ki against human fXa is 6.6 µM
under our assay conditions: Robison, D. J .; Furie, B.; Furie, B.
C. Active site of bovine factor Xa. J . Biol. Chem. 1980, 2555,
2014-2021.
(14) Marquart, M.; Walter, J .; Deisenhofer, J .; Bode, W.; Huber, R.
The geometry of the reactive site and of the peptide groups in
trypsin, trypsinogen and its complexes with inhibitors. Acta
Crystallogr. 1983, B39 , 480-490.
(15) Pinner, A.; Klein, F. Ber. 1877, 10, 1889.
(16) (a) Brenans, N.; Prost, S. C. R. Hebd. Seances Acad. Sci. 1924,
178, 1285. (b) Pilling, R. J .; Whiting, D. A. FIQ and FIQ2, New
Q-site inhibitors for photosynthetic electron transport: Synthesis
and the relationship between stereochemistry and biological
activity. J . Chem. Soc., Perkin Trans. 1, 1999, 2077-2086.
(17) (a) Cutoro, M.; Fiandanese, V.; Naso, F.; Sciacovelli, O. Synthesis
of dehydrophenylalanine derivatives by palladium catalyzed
arylation of 2-acetamidoacrylic acid. Tetrahedron Lett. 1983, 24,
4603-4606. (b) Carlsto¨m, A.-S.; Frejd, T. Palladium-catalyzed
synthesis of didehydroamino acid derivatives. Synthesis 1989,
414-418. (c) Harrington, P. J .; Hegedus, L. S. J . Palladium-
catalyzed reactions in the synthesis of 3- and 4- substituted
indoles. J . Org. Chem. 1984, 49, 2657.
(18) Richter, P.; Wagner, G. Synthesen von 4-amidinophenylbrenz-
traubenztraubensa¨ure (APPA). Pharmazie 1973, 28, 514-519.
(19) Lin, Z.; J ohnson, M. E. Proposed cation-π mediated binding by
factor Xa: A novel enzymatic mechanism for molecular recogni-
tion. FEBS Lett. 1995, 370, 1-5.
(20) Brandstetter, H.; Ku¨hne, A.; Bode, W.; Huber, R.; Saal, W.;
Wirthensohn, K.; Engh, R. A. X-ray structure of active site-
inhibited clotting factor Xa. J . Biol. Chem. 1996, 271, 29988-
29992.
J M020485X