Design, synthesis, and biological evaluation of potent and selective amidino bicyclic factor Xa inhibitors

Article abstract of DOI:10.1021/jm000113t

Thrombotic diseases are a major cause of death and morbidity. Factor Xa (fXa) plays a vital role in the regulation of normal homeostasis and abnormal intravascular thrombus development in the blood coagulation cascade. A novel series of fXa inhibitors inc

Full text of DOI:10.1021/jm000113t

4398
J . Med. Chem. 2000, 43, 4398-4415
Design , Syn th esis, a n d Biologica l Eva lu a tion of P oten t a n d Selective Am id in o
Bicyclic F a ctor Xa In h ibitor s
Qi Han,* Celia Dominguez,† Pieter F. W. Stouten,# J eongsook M. Park, Daniel E. Duffy, Robert A. Galemmo, J r.,
Karen A. Rossi, Richard S. Alexander, Angela M. Smallwood, Pancras C. Wong, Matthew M. Wright,
J oseph M. Luettgen, Robert M. Knabb, and Ruth R. Wexler
DuPont Pharmaceuticals Company, Experimental Station, P.O. Box 80500, Wilmington, Delaware 19880-0500
Received March 8, 2000
Thrombotic diseases are a major cause of death and morbidity. Factor Xa (fXa) plays a vital
role in the regulation of normal homeostasis and abnormal intravascular thrombus development
in the blood coagulation cascade. A novel series of fXa inhibitors incorporating an amidino
6,5-fused bicyclic moiety at the P1 position has been designed and synthesized based on
molecular modeling studies. Structure-activity relationship (SAR) studies have led to selective
subnanomolar fXa inhibitors. The most potent fXa inhibitor in this series (72, SE170) has a
potent inhibition constant (Ki ) 0.3 nM), is 350-fold selective for fXa over trypsin, and also
shows good in vivo efficacy in a rabbit arterio-venous thrombosis model (ID50 ) 0.14 µmol/kg/
h). An X-ray crystal structure of 72 complexed to bovine trypsin was completed, and a binding
mode of 72 with fXa has been proposed based on modeling with human des-Gla-fXa.
In tr od u ction
Thrombotic diseases, including stroke, deep vein
thrombosis, and pulmonary embolism, are a major cause
of death and morbidity. Recent understanding of the
causes and mechanisms of thromboembolic disorders
has grown rapidly. Current antithrombotic drugs have
their merits and limitations. Heparin, low-molecular-
F igu r e 1. Strucutre of DX-9065a (1).
weight heparins, and recombinant hirudin all lack oral
bioavailability.2 Coumadin (warfarin), the only mar-
keted oral anticoagulant agent, is being used in a
growing number of antithrombotic indications; however,
it requires individual dose titration and periodic moni-
toring.3 Therefore, new orally active anticoagulants,
which offer significant advantages over the agents
presently in use, are needed for prevention and treat-
ment of thrombotic diseases.
Factor Xa (fXa) plays a vital role in the regulation of
normal homeostasis and abnormal intravascular throm-
bus development in the blood coagulation cascade.4 fXa,
a trypsin-like serine protease, is located at the conver-
gent point of the intrinsic (activated by surface contact)
and extrinsic (activated by vessel injury) pathways of
coagulation.5 The major role of fXa is the generation of
thrombin from prothrombin.6 Thrombin, as the last
serine protease in the cascade, has several important
procoagulant functions including converting fibrinogen
to fibrin for clot formation and stimulating platelet
aggregation.7 Most of the research in the past decade
has been focused on thrombin inhibition. Recent re-
search indicates that the prothrombinase complex is
unaffected by direct thrombin inhibitors;8 by contrast,
inhibition of fXa should prevent continuing production
of thrombin while maintaining a basal level of thrombin
activity necessary for primary homeostasis. In addition,
one molecule of fXa generates many molecules of
thrombin,9 suggesting that inhibition of fXa may be
more efficacious than inhibition of thrombin. Moreover,
since fXa inhibitors affect coagulation specifically but
not platelet function, this mechanism should have less
potential to increase the risk of abnormal bleeding.
Recent in vivo studies with small-molecule fXa inhibi-
tors have demonstrated that fXa inhibitors are indeed
efficacious anticoagulants with a minimal effect upon
bleeding.10 As a result of these investigations, the
inhibition of fXa has begun to emerge as an attractive
strategy for the discovery of novel antithrombotic
agents.11
At the initiation of our program, Daiichi’s DX-9065a
(1; Figure 1) was reported as a potent nonpeptide fXa
inhibitor.12 Since that time, more amidines have been
reported as fXa inhibitors.13 An important feature
common to amidine-based, trypsin-like serine protease
inhibitors is that the highly basic amidino group re-
mains completely protonated under most physiological
conditions. Therefore, amidines are capable of forming
a symmetrical bidentate charge interaction with the
carboxylate side chain of Asp189 in the primary specific-
ity pocket (S1) of the serine proteases.14 In this paper,
we describe our progress toward the rational design and
synthesis of novel series of amidinobenzimidazole, ami-
dinoindole, and amidinoindazole derivatives as potent
and specific fXa inhibitors.
* Corresponding author. Tel: 302-695-8676. Fax: 302-695-8175.
E-mail: qi.han@dupontpharma.com.
Ch em istr y
† Amgen Inc., 1840 DeHavilland Dr., Thousand Oaks, CA 91320-
1789.
The preparative route to amidinobenzimidazole esters
12-15 is shown in Scheme 1. Ethyl 2-[2-(3-cyanophe-
# Pharmacia & Upjohn, Viale Pasteur 10, 20014 Nerviano (Mi), Italy.
10.1021/jm000113t CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/01/2000

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4399
Sch em e 1. Synthesis of Amidinobenzimidazole Ester Derivativesa
a
Reagents: (a) Zn, CuCN, 2 equiv LiCl; (b) K2CO3, DMF, 110 °C; (c) HCl(g), EtOH; (d) (NH4)2CO3, EtOH; (e) HPLC separation.
Sch em e 2. Synthesis of Amidinobenzimidazole Ketone Derivativesa
a
Reagents: (a) NaH, THF; (b) HPLC separation; (c) phenylboronic acid, Pd(PPh3)4, 2 M Na2CO3, THF-H2O; (d) 19, Pd(PPh3)4, 2 M
Na2CO3, THF-H2O; (e) HCl(g), EtOH; (f) (NH4)2CO3, EtOH.
nyl)ethyl]acrylate (5) and ethyl 2-[2-(4-cyanophenyl)-
ethyl]acrylate (6) were prepared, respectively, from a
carbon-carbon coupling reaction of ethyl (bromomethy-
l)acrylate (2) with either a m- or p-cyanotolyl copper
reagent, which in turn was prepared from the reaction
of the corresponding bromide (3 or 4) with zinc, followed
by treatment with copper(I) cyanide and lithium chlo-
ride. Michael addition of 5(6)-cyanobenzimidazole (7)15
to 5 or 6 gave the regioisomeric mixture 8 or 9,
respectively. Subsequent conversion of cyano precursors
8 and 9 to the corresponding amidino compounds 10 and
11 was carried out under standard Pinner reaction
conditions by bubbling hydrogen chloride gas into a
solution of the cyano compound in ethanol and subse-
quent amination of the formed imidate by treatment
with ammonium carbonate in ethanol. The regioisomeric
mixtures 8-11 could be separated on HPLC either at
the cyano precursor stage or at the amidino stage;
however, separation of the cyano precusors 8 and 9
proved to be more facile than that of amidino derivatives
10 and 11. The structures of the four compounds 12-
15 were confirmed by long-range proton-carbon cor-
relation NMR spectroscopy.
The synthesis of the substituted amidinobenzimida-
zole ketones 20-22 as shown in Scheme 2 began with
the alkylation of 7 with 2′,4-dibromoacetophenone (16)
in the presence of NaH. HPLC separation of the two
regioisomers provided 6-cyano-substituted isomer 17
and 5-cyano-substituted isomer 18. Suzuki coupling
reactions of 17 with phenylboronic acid or 2-(tert-

4400 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Han et al.
Sch em e 3. Synthesis of Amidinobenzimidazole Amide Derivativesa
a
Reagents: (a) K2CO3, 18-crown-6, THF; (b) HPLC separation; (c) 4-aminobiphenyl or 28, AlMe3, CH2Cl2; (d) HCl(g), EtOH; (e) (NH4)2CO3,
EtOH; (f) R-chloroacetyl chloride, K2CO3, CH3CN; (g) K2CO3, DMF, 7.
butylaminosulfonyl)phenylboronic acid (19),16 and 18
with phenylboronic acid, followed by the standard
conversion from a cyano to an amidine group as de-
scribed above, gave the desired amidino ketones 20-
22, respectively.
presence of the BOP coupling reagent and triethylamine
provided the cyano intermediates 40 and 42, respec-
tively. These intermediates were subjected to the stan-
dard Pinner reactions and aminations, followed by
HPLC purification, to give amidinoindole 41 and ami-
dinoindolines 43 and 44, respectively.
Two routes were employed to make the substituted
amidinobenzimidazole amides 30-33 as shown in
Scheme 3. Alkylation of 7 with methyl 2-bromoacetate
(23), followed by separation on HPLC, afforded the
6-cyano-substituted isomer 24 and 5-cyano-substituted
isomer 25. Weinreb reaction of ester 24 or 25 with the
aluminum salt of commercial 4-aminobiphenyl or N-(tert-
butyl)-4′-amino[1,1′-biphenyl]-2-sulfonamide (28)16 pro-
vided amides 26 and 27, respectively. The amides 26
and 27 were also prepared by alkylation of 7 with
2-chloroamide 29, which was obtained from acylation
of 28 with 2-chloroacetyl chloride, followed by HPLC
isolation. The latter route is preferred. Conversion of
the cyano intermediates 26 and 27 to the amidines 30-
33 was carried out under reaction conditions described
earlier.
The synthesis of the amidinoindole derivatives is
exemplified by the preparation of 41, while the prepara-
tion of the amidinoindoline derivatives is exemplified
by the preparation of 43 and 44 (Scheme 4). 5-Cyanoin-
dole (34) was converted to methyl (5-cyano-1H-indol-3-
yl)(oxo)acetate (35) by treatment with oxalyl chloride
and then methanol. N-Methylation of indole 35 with
methyl iodide in the presence of NaH provided methyl
(5-cyano-1-methyl-1H-indol-3-yl)(oxo)acetate (36). Re-
duction of 35 with triethylsilane in trifluoroacetic acid,
followed by saponification with KOH in MeOH, provided
(5-cyano-1H-indol-3-yl)acetic acid (37) and (5-cyano-2,3-
dihydroindol-3-yl)acetic acid (39). Similarly, reduction
of 36 followed by saponification provided 38. The reac-
tion of acid 37 or 39 with aniline 28 in DMF in the
The synthesis of the amidinoindazole derivative 50,
shown in Scheme 5, was accomplished by alkylation of
6-nitroindazole (45) with methyl 2-bromoacetate, fol-
lowed by hydrogenation of the nitro group to form
methyl (6-amino-1H-indazol-1-yl)acetate (46). Diazoti-
zation of 46 followed by quenching with CuCN/NaCN
and then saponification provided (6-cyano-1H-indozol-
1-yl)acetic acid (47). Coupling of acid 47 with 2-(6-
amino-3-pyridinyl)-N-(tert-butyl)benzenesulfonamide
(48),16 followed by the standard Pinner reaction and
amination, produced 50.
Generally, the 4-aminobiphenyl intermediates were
prepared via a palladium-catalyzed coupling reaction
(Scheme 6).17 A Suzuki coupling reaction of 4-bromoni-
trobenzene (51) with 19 provided nitrobiphenyl 52.
Alkylation of the sulfonamide NH of 52 with methyl
iodide, ethyl bromide, or propyl bromide in the presence
of NaH or K2CO3, followed by hydrogenation in the
presence of 10% palladium on carbon, afforded 53-55.
Removal of the tert-butyl group of 52 with TFA, followed
by dialkylation of the sulfonamide with methyl iodide
in the presence of NaH and then hydrogenation, gave
56. The 4-aminobiphenyl intermediates 58-62 were
synthesized via Suzuki reactions of appropriately sub-
stituted 4-bromoaniline 57 with 19. N-(tert-Butyl)-4′-
(methylamino)[1,1′-biphenyl]-2-sulfonamide (63) was
synthesized by acylation of 28 with ethyl formate,
followed by reduction with LAH.

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4401
Sch em e 4. Synthesis of Amidinoindole Amide Derivativesa
a
Reagents: (a) oxalyl chloride, Et2O; (b) MeOH; (c) NaH, MeI; (d) TFA, Et3SiH; (e) KOH, MeOH; (f) 28, BOP, Et3N, DMF; (g) HCl(g),
MeOH; (h) (NH4)2CO3, MeOH.
Sch em e 5. Synthesis of Amidinoindazole Amide Derivativesa
a
Reagents: (a) BrCH2CO2Me, K2CO3, DMF; (b) H2, 10% Pd-C; (c) 0.5 N HCl, NaNO2, Na2CO3, CuCN, NaCN, 0 °C; (d) KOH, MeOH-
H2O; (e) BOP, Et3N; (f) HCl(g), EtOH; (g) (NH4)2CO3, EtOH.
Resu lts a n d Discu ssion s
confirmed by the crystal structure of the fXa/DX-9065a
complex,19 include four key interactions (Figure 2). First,
the naphthyl amidine binds via a symmetrical bidentate
charge interaction with a single Asp189 at the bottom of
the S1 pocket; however, only Asp189 oxygen twin (naph-
thamidine) arrangement was observed in the crystal
structure of the fXa/DX-9065a complex. Second, the
Mod elin g Stu d y. At the initiation of our program
the crystal structure of human des-Gla-fXa dimer18 was
used in our molecular modeling studies and Daiichi’s
DX-9065a (1) was chosen as a model for our inhibitor
design. The key features of a proposed binding mode of
DX-9065a docked in fXa, which were subsequently

4402 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Sch em e 6. Synthesis of Aminobiphenylsa
Han et al.
a
Reagents: (a) Pd(PPh3)4, 2 M Na2CO3, THF-H2O, reflux, 4-18 h; (b) MeI, EtBr or PrBr; K2CO3 or NaH; (c) H2, 10% Pd-C; (d) TFA;
(e) HCO2Et, reflux, 2 h; (f) LAH.
F igu r e 2. Modeling studies of DX-9065a in fXa and design of our initial fXa inhibitors.
pyrrolidine moiety sits in the aryl-binding pocket, which
is formed by residues Phe174, Trp215, and Tyr99. Third,
the acetimidoyl functionality binds in the cation hole
at the back of the S4 pocket composed of the carbonyl
oxygens of Lys96, Thr98, and Glu97 and the side chain
carboxylate of Glu97, which binds the basic charged
group. Fourth, the carboxylic acid, which was believed
to be responsible for the specificity for fXa over throm-
bin, forms a hydrogen bond with the side chain of Arg143
and Gln192. This binding model was used to design our
more rigid and tighter binding inhibitors.
assesses the active site for receptor “target” points. The
selected interest points are required to interact with the
proposed inhibitors. The Sprout program does an ex-
haustive search based on electrostatic interactions by
selecting templates to connect these points and then
generates hundreds of skeletons. The S1 pocket of fXa
was chosen for template design since it is a well-defined
region with a short distance to span in order to join the
target points. The relevant electrostatic target points
for the S1 pocket of fXa, based on the modeling results
of the des-Gla-fXa/DX-9065a complex, include Asp189
(deep in the S1 pocket), Ser195 (at the active site), and
Arg143/Gln192 (hovering over the S1 pocket) (Figure 2).
The Sprout-generated skeleton 64 was chosen by an
The Sprout program20 was used as a de novo design
technique to produce frameworks as starting points in
the design of our initial fXa inhibitors. This program

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4403
Ta ble 1. SAR of Amidinobenzimidazole Ester Derivatives with Basic P4 Fragments
a
Human purified enzymes were used. Values are averages from multiple determinations (n g 2) and the standard deviations were
<20% of the mean.
evaluation of its tight fit in the active site and for its
rigidity and size. Skeleton 64 was transformed to
compound 65, which was reversed to ethyl 3-(6-amidino-
1H-benzimidazol-1-yl)propionate (66) because the struc-
ture offered more synthetic flexibility. Subsequently, 66
was optimized in the S1 pocket prior to expansion into
the S4 pocket, through energy minimization using
Sprout and 3D database searching. When 66 was
modeled in the S1 pocket of fXa, 66 formed favorable
basicity of the compounds (Figure 3). Our modeling
suggested that the benzamidine P4 fragment could be
replaced by a variety of aryl moieties, such as a
functionalized p-biphenyl. In addition, introduction of
a hydrogen-bond donor/acceptor on the ortho-position
of the distal phenyl ring ought to be capable of ad-
ditional hydrogen-bonding interactions with Tyr99 of
fXa. This strategy was found to be successful in our
earlier series of fXa inhibitors when we replaced the P4
benzamidine moiety in 67 with a biphenylsulfonamide
in 68.16 The modeling also suggested that the carboxy-
late in 12 could be replaced with a carbonyl as the
linking functionality in 69 to eliminate the chiral center.
In so doing, these inhibitors ought to also have an
additional hydrogen-bond interaction between the car-
bonyl oxygen and the Gly219 NH of fXa.
SAR Eva lu a tion . Initially, we were gratified to find
that the 6-amidinobenzimidazole derivative with a
biphenyl P4 fragment (20) shown in Table 2 had similar
potency and selectivity to benzamidine 12. As antici-
pated based on our previous results, these data sug-
gested that indeed a neutral biphenyl without cationic
character could replace the pharmacokinetically unde-
sirable amidine moiety. For comparison 5-amidinoben-
zimidazole derivative 21 was also prepared. The 6-iso-
mer 20 (fXa Ki ) 120 nM) was significantly more potent
than the 5-isomer 21 (fXa Ki ) 2,700 nM). We therefore
focused on the preparation of 6-amidinobenzimidazole
derivatives, employing 20 as the starting point to
systematically introduce changes into the P4 moiety. We
had previously shown that the o-biphenylsulfonamide
is an optimal P4 substituent.16 Unfortunately, as shown
by 22 (fXa Ki ) 240 nM), an o-sulfonamide on the distal
phenyl ring of this series showed 2-fold weaker affinity
than the unsubstituted biphenyl 20. To improve the
potency, we thought to extend the tether from a ketone
group to an amide group to allow the biphenyl ring to
sit deeper into the S4 pocket. To initially test this
interactions with all four electrostatic targets (Asp189
,
Arg143, Gln192, and Ser195) in the S1 pocket. To maximize
the interaction with the aryl pocket and the “cation hole”
in the S4 pocket of fXa, a benzamidine group as the P4
fragment was placed in the center of this area. We
modeled different length carbon chains as a tether for
connecting the P1 and P4 fragments without disturbing
the orientation of the fragments in their respective
pockets. A two-carbon chain connecting the R-position
of the ester group of 66 to either the meta- or para-
position of the benzamidine was found to be an optimal
tether.
The modeling studies led to the design and synthesis
of four initial compounds, 12-15 (Table 1). Compound
12 was found to be the most potent fXa inhibitor (Ki )
140 nM) among the four, with 15 being approximately
85-fold less potent (Ki ) 11 900 nM) than 12. Com-
pounds 12 and 13 containing a 3-benzamidine moiety
as the P4 fragment are more potent than 14 and 15
containing a 4-benzamidine moiety. This observation is
similar to our earlier findings and those reported in the
literature, in which 3-substituted benzamidines are
preferred over 4-substituted benzamidines.18,21
Since compounds 12-15 have two basic moieties, we
were concerned that these bisamidines would not have
adequate oral absorption or appropriate pharmacoki-
netics to be useful as oral antithrombotic agents.22
Therefore, our strategy was to replace the P4 benzami-
dine group with a neutral fragment to attenuate the

4404 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Han et al.
F igu r e 3. Introduction of neutral P4 fragments into amidinobenzimidazoles.
Ta ble 2. SAR of Amidinobenzimidazole Derivatives with Neutral P4 Fragments
a
Human purified enzymes were used. Values are averages from multiple determinations (n g 2) and the standard deviations were
<20% of the mean.
hypothesis, we prepared a pair of amides, the 6-amidi-
nobenzimidazole 30 (fXa Ki ) 220 nM) and 5-amidi-
nobenzimidazole 31 (fXa Ki ) 3900 nM), which did not
show improved potency. Fortunately, when the o-
sulfonamide group was introduced on the distal phenyl
ring in this series, the 6-substituted isomer 32 (fXa Ki
) 2.4 nM) and also the corresponding 5-isomer 33 (fXa
Ki ) 400 nM) showed significant improvement in
potency for fXa relative to unsubstituted biphenyl
derivatives 30 and 31. We believe that the o-sulfona-
mide on the distal phenyl ring in the P4 fragment causes
the two rings of the biphenyl to be more orthogonal due
to steric bulk. The orthogonal arrangement allows the
distal phenyl ring to form a stronger “two-sandwich”
π-interaction with residues Phe174 and Tyr99 in the S4
pocket of fXa. In addition, our model suggests that the
sulfonamide may be interacting with Tyr99 to further
enhance fXa potency.
As the modeling suggested, the benzimidazole P1
moiety could be replaced with other 6,5-fused bicyclic
templates, such as indole, with good retention of fXa
potency. The use of 6-amidinoindole as the P1 moiety

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4405
Ta ble 3. Comparison of Selectivity and Pharmacokineticsa of Amidinobenzimidazole 32 and Amidinoindole 41
a
Dosed intravenously in rabbit for 60 min at a dose of 10 mg/kg/h. bHuman purified enzymes were used. Values are averages from
multiple determinations (n g 2) and the standard deviations were <20% of the mean.
Ta ble 4. SAR of P4 Fragments in Amidinoindole Series
a
Human purified enzymes were used. Values are averages from multiple determinations (n g 2) and the standard deviations were
<20% of the mean.
has been shown to successfully result in potent throm-
bin inhibitors.24 As shown in Table 3, both indole 41
(fXa Ki ) 2.1 nM) and benzimidazole 32 have similar
affinity for fXa, but indole 41 has the advantage of being
more selective for fXa over trypsin. Indole 41 and
benzimidazole 32 were evaluated in pharmacokinetic
studies in rabbits via an intravenous infusion at a dose
of 10 mg/kg/h over 60 min. Indole 41 exhibited the better
pharmacokinetic profile, having a lower clearance and
a slightly longer half-life than benzimidazole 32. As a
result of these studies, further optimization was pur-
sued in the indole series.
Further efforts to define the SAR included manipula-
tion of the P4 side chain by alkylation of the amide NH
and exploration of biphenyl replacements (see Table 4).
The N-methylamide 70 was 20-fold less active than 41.
As anticipated, based on earlier results with the isox-
azoline series16, replacement of the proximal phenyl ring
with a pyridyl ring (71) resulted in a 2-fold enhancement
in potency. To further improve the potency and block
possible amide bond cleavage in the indole series,
halogen and alkyl substituents were introduced into the
ortho-position of the proximal phenyl ring (Table 4).
Introduction of a bromo (72), iodo (73), or chloro (74)
substituent in the ortho-position of the proximal ring
also afforded compounds with subnanomolar fXa inhibi-
tion and good selectivity (>200-fold) against related
serine proteases (thrombin and trypsin). Significantly,
bromide 72 showed a 6-fold improvement in fXa affinity
(fXa Ki ) 0.32 nM) and had improved selectivity (375-
fold selectivity over trypsin). Introduction of a methyl
or fluoro substituent (75 and 76) in the ortho-position
resulted in little change in fXa affinity and selectivity
of fXa over trypsin, but the fluoro analog (76) showed
better selectivity against thrombin compared to the
unsubstituted biphenyl derivative of this series (41).

4406 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Ta ble 5. Substitution in Distal Phenyl Ring
Han et al.
a
Human purified enzymes were used. Values are averages from multiple determinations (n g 2) and the standard deviations were
<20% of the mean.
Ta ble 6. SAR of Amidino 6,5-Fused Bicyclic fXa Inhibitors
a
Human purified enzymes were used. Values are averages from multiple determinations (n g 2) and the standard deviations were
<20% of the mean.
N-Alkylation of the sulfonamide was pursued in an
attempt to improve the pharmacokinetic profile. More
specifically the rationale for alkylation of the sulfona-
mide NH2 with methyl (77), ethyl (78), dimethyl (79),
and n-propyl (80) as illustrated in Table 5 was to
increase the lipophilicity of the compounds in an at-
tempt to increase the volume of distribution (Vdss), which
would result in an increased duration of action. Unfor-
tunately, as the size of the alkyl group on the sulfona-
mide was increased, the fXa inhibitory activity de-
creased. Additionally, the selectivity of fXa over thrombin
and trypsin of these compounds was also lower than
that of the parent 41. Therefore, the N-alkylated
derivatives were not evaluated further.
Efforts to further explore the P1 group are shown in
Table 6. The N-methylated indole 81 showed a 200-fold
decrease in affinity for fXa compared with 71. In an
effort to optimize the amidine interaction in the S1
pocket, additional heterobicyclic amidines were evalu-
ated. Amidinoindazole 50 was found to be 2-fold less

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4407
Ta ble 7. Dog Pharmacokinetics of Selective Amidinoindole fXa Inhibitors
a
From N-in-1 results.
Ta ble 8. A-V Shunt Results of Select Amidinoindole fXa Inhibitors
potent than the corresponding indole 71. Amidinoindo-
line (+)-43 showed a 3-fold enhancement in potency
relative to the corresponding indole 41, while the
negative isomer (-)-44 is approximately 25-fold less
potent than 41.
P h a r m a cok in etic An a lysis. Representative indoles
including 71, 72, 74, and 76 were evaluated in phar-
macokinetic studies following intravenous administra-
tion in the dog and compared to parent indole 41 (Table
7). The halogenated compounds 72, 74, and 76 all were
shown to have longer duration of action than 41.
However, only fluoro analog 76 had a t1/2 > 2 h. Overall,
76 demonstrated the best pharmacokinetic profile in the
series with the lowest clearance (Cl ) 1.1 L/kg/h), the
longest t1/2 (2.5 h), and an increase in Vdss (0.87 L/kg)
compared to the lead 41. However, 76 is 10-fold less
potent than bromide 72 (Ki values 3.1 nM vs 0.32 nM).
Bromide 72 and chloride 74 both showed increased t1/2
(1.8 and 1.75 h) compared to 41 (t1/2 ) 0.79 h) due to
the larger Vdss (0.67 and 2.01 L/kg), despite having
similar clearance (1.34 L/h/kg vs 1.22 L/h/kg). Even
though pyridine 71 showed an increase in Vdss (1.9 L/kg),
it had poorer exposure than the other compounds
evaluated because of increased clearance (3.7 L/h/kg).
An tith r om botic Affica cy Stu d y. The antithrom-
botic efficacy of indoles 41, 71, 72, and 74 was evaluated
in a rabbit arterio-venous thrombosis model25 (Table 8).
The compounds were administered by intravenous infu-
sion, and the antithrombotic effect was expressed as the
ID50 (dose which reduced the clot weight by 50%). As
anticipated based on the in vitro potency, bromide 72
was the most potent compound with an ID50 of 0.14
µmol/kg/h. Pyridine 71 (ID50 ) 0.3 µmol/kg/h) and

4408 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Han et al.
F igu r e 4. X-ray structure of 72 (SE170)/trypsin complex. Structure of 72 (SE170; thick bonds) complexed to bovine trypsin (thin
bonds). Selected active site residues are labeled. The inhibitor binds in both the P1 pocket and the P4 binding site.
indole ring and the side chain hydroxyl of Ser195. This
distance is too long to be considered a hydrogen bond
but may be indicative of a weakly polar interaction. A
hydrogen bond is observed between the carbonyl oxygen
of 72 and the backbone amide oxygen of Gly219. Only
one orientation is observed for the bromine on the
proximal phenyl ring and for the sulfonamide on the
distal phenyl ring. The biphenyl moiety is located in the
aryl-binding pocket (S4). This region includes the side
chains of Trp215 and Leu99
.
Superimposing the inhibitor coordinates on the struc-
ture of fXa, an overlap between the van der Waals
surfaces of the bromine and Tyr99 is observed. This
interaction may lead to the greater selectivity and
affinity of this molecule for fXa, as thrombin and trypsin
both have a leucine in position 99.
F igu r e 5. 72 (SE170) in rabbit A-V shunt model. Antithrom-
botic effect (expressed as % inhibition of thrombus formation)
of SE170 at 0.008-1 mg/kg/h in rabbit arterio-venous shunt
model; n ) 3-4/group and means ( SE.
Con clu sion
On the basis of molecular modeling studies employing
the active site coordinates of the fXa dimer, we have
designed and synthesized a novel class of potent and
selective fXa inhibitors. These inhibitors employ ami-
dine-substituted 6,5-fused heterocycles, such as amidi-
nobenzimidazole and amidinoindole, to occupy the S1
site of fXa. The X-ray structure of the 72/trypsin
complex supports the proposed binding model. These
compounds demonstrated potent inhibitory fXa activity
with the most potent compounds showing subnanomolar
affinity. These inhibitors also maintained high selectiv-
ity for fXa versus thrombin and trypsin.
chloride 75 (ID50 ) 0.35 µmol/kg/h) also showed better
in vivo efficacy than the lead 41 (ID50 ) 1.7 µmol/kg/h).
X-r a y An a lysis of 72/Tr yp sin Com p lex. Difficulties
were encountered in obtaining an X-ray crystal struc-
ture of fXa bound with an inhibitor. Instead, because
of the large degree of homology between fXa and
trypsin,26 the structure of 72 (SE170) complexed to
bovine trypsin was determined (Figure 4). The structure
has been refined to a crystallographic R-factor of 0.184
at 2.0 Å resolution. The binding of the inhibitor includes
interactions with both the S1 specificity pocket and the
S4 aryl-binding pocket. No major reorientations of the
active site residues were observed upon inhibitor bind-
ing. As expected, the nitrogens of the amidine group are
in close contact (2.8 and 2.9 Å) with the side chain of
the carboxylic acid oxygens of Asp189. Interestingly,
there is a 3.8 Å distance between the nitrogen of the
Exp er im en ta l Section
En zym e Affin ity Assa ys. fXa, thrombin, and trypsin Ki's
were obtained from human purified enzymes. All assays were
run in microtiter plates using a total volume of 250 µL in 0.1
M sodium phosphate buffer containing 0.2 M NaCl and 0.5%
poly(ethylene glycol) 6000 at pH 7.4. The compounds were run
at 10, 3.16, 1.0, 0.316, 0.1, 0.0316, 0.01, and 0.00316 µM. The

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4409
substrate for fXa and trypsin was S2222 and the substrate
for thrombin was S2366. Plates were read for 30 min at 405
nm. Rate of substrate hydrolysis was determined for the
controls (no inhibitor) and for the inhibitors. Percent (%)
enzyme activity was determined from these rates and used in
Additionally, the longest carbon chain was set to a length of
three atoms. The run resulted in thousands of skeletons. These
were manually reduced to approximately 200 that were
“reasonable”. Manual assignment of atom types in a few of
these skeletons was performed to allow for interactions with
specific residues. Once skeletons were atom-typed, these
structures were placed into Insight II for further evaluations.
The protein was held rigid while the ligands were minimized
using the cvff force field. The starting structures of the
potentially inhibitory molecules for fXa were directly chosen
from the Sprout results.
Ch em ica l a n d P h ysica l Meth od s. All reactions detailed
below were carried out under inert gas in oven-dried glassware
unless otherwise indicated. Solvents and reagents were ob-
tained from commercial vendors in the appropriate grade and
used without further purification unless otherwise indicated.
Proton and carbon NMR spectra were obtained on VXR or
Unity 300 or 400 MHz instruments (Varian Instruments, Palo
Alto) with TMS as an internal reference standard. Mass
spectra were measured with a Hewlett-Packard 5988A mass
spectrometer with particle beam interface using NH3 for
chemical ionization, a Finnigan “Navigator-aQa” system under
atmospheric pressure chemical ionization (APcI) conditions,
or a Micromass Platform II system under electrospray ioniza-
tion (ESI) conditions. High-resolution mass spectra were
obtained either on a VG70-VSE mass spectrometer under NH3-
CI conditions or on a Finnigan MAT95S mass spectrometer
under ESI conditions. Melting points were determined on a
Mettler SP61 apparatus and are uncorrected. All compounds
were determined to be homogeneous by TLC, elemental
analysis, and/or HPLC. Analytical HPLC was performed on a
C18 column using linear gradients of CH3CN/H2O/0.05% TFA
from 90% water and 0.05% TFA in CH3CN to 0.05% TFA in
CH3CN over 30 min. Preparative reverse-phase HPLC was
performed on a C18 column using gradients with compositions
of CH3CN/H2O/0.05% TFA over 45 min. Elemental analyses
were performed by Quantitative Technologies, Inc., Bound
Brook, NJ .
Eth yl 2-[2-(3-Cya n op h en yl)eth yl]a cr yla te (5) a n d Eth -
yl 2-[2-(4-Cya n op h en yl)eth yl]a cr yla te (6). To a stirred
suspension of zinc powder (1.43 g, 22 mmol) in THF (10 mL)
was added a catalytic amount of 1,2-dibromoethane (0.2 g) at
room temperature, and the mixture was stirred for 30 min. A
solution of 3-cyanobenzyl bromide (3) or 4-cyanobenzyl bromide
(4) (3.92 g, 20 mmol) in THF (25 mL) was slowly added at a
rate of 1 drop every 5 s at 5-10 °C. The resulting mixture
was stirred for 3 h at that temperature and then transferred
into a solution of CuCN (1.82 g, 20 mmol) and LiCl (1.68 g, 40
mmol) in THF (20 mL) at -78 °C. The resulting mixture was
warmed to -20 °C, stirred for 20 min, and then re-cooled to
-78 °C. After ethyl 2-(bromomethyl)acrylate (2) (3.86 g, 20
mmol) was slowly added, the mixture was stirred at -78 °C
for 2 h and then warmed to room temperature overnight. Ether
(100 mL) and aqueous saturated NH4Cl (50 mL) were added,
and the mixture was filtered. The filtrate was washed with
water and brine and dried over MgSO4. Concentration gave a
residue, which was purified by column chromatography with
a gradient of CH2Cl2-EtOAc to give pure 5 (3.6 g, 78.6%) and
6 (1.2 g, 26.2%), respectively.
the following formula to determine Ki: Ki ) 1000 × I(((Km
+
S) - S × ACT)/(ACT × Km) - 1); where I ) inhibitor
concentration, Km ) substrate concentration that yields half-
maximal velocity, S ) substrate concentration, and ACT ) %
enzyme activity in the presence of inhibitor. All compounds
were tested in duplicate studies and were compared with the
same internal standards. Standard deviation of all assays was
less than 20%. The detailed methodology is described in ref
23.
P h a r m a cok in etic Stu d ies. Plasma samples (100 µL) were
prepared by protein precipitation with acetonitrile (750 µL).
The supernatant was dried under nitrogen and reconstituted
in 100 µL of acetonitrile:water (5:95 v/v). Plasma levels were
determined by liquid chromatography-tandem mass spec-
trometry. The instrument used was a Sciex (Thornhill, On-
tario) model API III+ tandem mass spectrometer interfaced
with a turbo ion spray ionization source. The limit of quanti-
tation was 5 nM. Plasma concentration-time data were fitted
to a noncompartment kinetic model using Winnonlin phar-
macokinetic program (Scientific Consulting, Inc., Cary, NC).
Plasma half-lives were calculated using nonlinear regression.
The areas under the concentration curve (AUC) were calcu-
lated by trapezoidal approximation. The systemic plasma
clearance (Cl) was calculated from the concentration-time
data using the standard formula: Cl ) dose/AUC0finfinity. The
apparent volume of distribution at steady state (Vdss) was
calculated using statistical moment analysis by the equation:
Vdss ) Cl × MRT, where MRT is the mean residence time,
calculated as: MRT ) AUMC0finfinity/AUC0finfinity, where AUMC
is the area under the first moment curve.
Ar ter io-Ven ou s Sh u n t Th r om bosis Mod el. New Zealand
rabbits (2-4 kg) were anesthetized with ketamine (50 mg/kg
im) and xylazine (10 mg/kg im). These anesthetics were
supplemented as needed. The femoral artery, jugular vein, and
femoral vein were isolated and catheterized. A saline-filled
arterio-venous (A-V) shunt device was connected between the
femoral arterial and the femoral venous cannulae. The A-V
shunt device consisted of an outer piece of 8-cm Tygon tubing
(internal diamter ) 7.9 mm) and an inner piece of 2.5-cm
tubing (internal diamter ) 4.8 mm). The A-V shunt also
contained an 8-cm-long 2-0 silk thread (Ethicon, Somerville,
NJ ). Blood flowed from the femoral artery via the A-V shunt
into the femoral vein. The exposure of flowing blood to a silk
thread induced the formation of a significant thrombus. After
40 min, the shunt was disconnected and the silk thread
covered with thrombus was weighed. The compounds or saline
vehicle was given as continuous iv infusions via the jugular
vein starting 1 h before blood was circulated in the shunt and
continuing throughout the experiment (i.e. 100 min). The
percentage inhibition of thrombus formation was determined
for each treatment group. The ID50 values (dose which
produced 50% inhibition of thrombus formation) were esti-
mated by linear regression. Variation of antithrombotic effect
was less than 10%. See detailed description in ref 25.
Ester 5. 1H NMR (CDCl3): δ 7.51-7.36 (m, 4H), 6.17 (s,
1H), 5.48 (d, J ) 1.1 Hz, 1H), 4.22 (q, J ) 7.3 Hz, 2H), 2.84
(dd, J ) 8.4 Hz, J ) 7.0 Hz, 2H), 2.61 (dd, J ) 8.4 Hz, J ) 7.0
Hz, 2H),1.32 (t, J ) 7.0 Hz, 3H). 13C NMR (CDCl3): δ 166.80,
142.85, 139.41, 133.14, 132.05, 129.85, 129.17, 118.94, 112.43,
60.79, 34.50, 33.57, 14.22. LRMS: 247 (M + NH4)+.
Compound 72 (SE170) given at 0.008-1 mg/kg/h produced
a dose-dependent antithrombotic effect with an ID50 of 0.14
µmol/kg/h (Figure 5).
Com p u ta tion a l Molecu la r Mod elin g. DX-9065a was
manually placed into the fXa active site using Insight II
(Molecular Simulations, Inc., San Deigo, CA). The protein was
held rigid while the inhibitor was minimized using the cvff
1
Ester 6. H NMR (CDCl3): δ 7.58 (dd, J ) 8.4 Hz, J ) 1.8
Hz, 2H), 7.28 (d, J ) 8.4 Hz, 2H), 6.17 (d, J ) 1.1 Hz, 1H),
5.48 (dd, J ) 2.6 Hz, J ) 1.1 Hz, 1H), 4.22 (q, J ) 7.3 Hz, 2H),
2.86 (dd, J ) 8.6 Hz, J ) 7.1 Hz, 2H), 2.61 (dd, J ) 8.6 Hz, J
) 7.0 Hz, 2H),1.32 (t, J ) 7.0 Hz, 3H). MS: 247 (M + NH4)+.
Eth yl 2-(3-Am id in op h en yl)eth yl-3-(5-a m id in oben zim i-
d a zol-1-yl)p r op ion a te (12) a n d Eth yl 2-(3-Am id in op h e-
n yl)eth yl-3-(6-am idin oben zim idazol-1-yl)pr opion ate (13).
A mixture of 715 (288 mg, 2 mmol), 5 (438 mg, 2 mmol), and
force field. After the four target points (Asp189, Ser195, Arg143
/
Gln192, and Ser195) were selected in Sprout, templates to join
these target points were selected. A basic set using a single
point, double bond, single bond, phenyl ring, and unsaturated
five-membered ring were selected. No atom types are assigned
during this phase of the Sprout calculation. A maximum of
three fused rings and a total of two six-membered rings and
two five-membered rings were allowed in the templates.

4410 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Han et al.
1
K2CO3 (276 mg, 2 mmol) in DMF (10 mL) was heated at 90 °C
for 16 h. The mixture was diluted with EtOAc (150 mL),
washed with 1 N HCl and brine, and dried over MgSO4. After
filtration and concentration, the residue was purified by
column chromatography with a gradient of CH2Cl2-EtOAc to
give a mixture of two regioisomers 8 (570 mg, 76.4%) as a
colorless oil. 1H NMR (CDCl3): δ 8.13-7.36 (m, 8H), 4.55 (dd,
J ) 14.3 Hz, J ) 9.2 Hz, 1H), 4.28 (dd, J ) 14.3 Hz, J ) 5.5
Hz, 1H), 4.07 (q, J ) 7.0 Hz, 2H), 3.00-2.91 (m, 1H), 2.80-
2.64 (m, 2H), 2.18-2.07 (m, 1H), 1.92-1.82 (m, 1H), 1.12 (t, J
) 7.0, 3H).
6-Ca r bon itr ile 17. H NMR (CDCl3): δ 8.35 (s, 1H), 8.11
(dd, J ) 1.1 Hz, J ) 0.7 Hz, 1H), 8.08 (d, J ) 8.8 Hz, 2H), 7.81
(d, J ) 8.4 Hz, 2H), 7.75 (dd, J ) 8.4 Hz, J ) 0.7 Hz, 1H), 7.56
(dd, J ) 8.4 Hz, J ) 1.8 Hz, 1H), 6.16 (s, 2H). ESIMS: 340/
342 (M + H)+.
1
5-Ca r bon itr ile 18. H NMR (CDCl3): δ 8.31 (s, 1H), 8.13
(t, J ) 0.7 Hz, 1H), 8.07 (d, J ) 8.8 Hz, 2H), 7.81 (d, J ) 8.8
Hz, 2H), 7.75 (dd, J ) 8.4 Hz, J ) 0.7 Hz, 1H), 7.57 (dd, J )
8.4 Hz, J ) 1.1 Hz, 1H), 6.15 (s, 2H). ESIMS: 340/342 (M +
H)+.
1-[2-[2′-(Am in osu lfon yl)[1,1′-biph en yl]-4-yl]-2-oxoeth yl]-
1H -b en zim id a zole-6-ca r b oxim id a m id e (22). Compounds
20-22 were prepared as described in the following procedures.
To a solution of 17 (200 mg, 0.59 mmol) and 2-(tert-butylami-
nosulfonyl)phenylboronic acid (19) (1.52 g, 0.59 mmol) in THF
(9 mL) was added Na2CO3 (2 M in water, 9 mL), and the
mixture was purged with nitrogen. To the mixture was added
Pd(PPh3)4 (60 mg, 0.05 mmol), and the resulting mixture was
refluxed for 16 h. The mixture was diluted with EtOAc (200
mL), washed with water and brine, dried over MgSO4, and
concentrated. The resulting residue was crystallized from
EtOAc two times and further purified on preparative TLC
plates with 50% EtOAc in CH2Cl2 to provide 1N-(4′-(o-tert-
butylaminosulfonylphenyl)phenylcarbonyl)methyl-6-cyanoben-
zimidazole (270 mg, 97%) as a white solid. This solid was
further modified to the amidine as described above, followed
by purification on preparative HPLC to provide 22 (14 mg,
5.6%) and 1-[2-[2′-(tert-butylaminosulfonyl)[1,1′-biphenyl]-4-
yl]-2-oxoethyl]-1H-benzimidazole-6-carboximidamide (tert-bu-
tyl analogue of 22) (100 mg, 31%). For 22. Mp: 126-128 °C.
1H NMR (CD3OD): δ 8.55 (bs, 1H), 8.18 (d, J ) 8.4 Hz, 2H),
8.13 (dd, J ) 7.7 Hz, J ) 1.4 Hz, 1H), 8.09 (s, 1H), 7.94 (d, J
) 8.8 Hz, 1H), 7.71 (dd, J ) 8.4 Hz, J ) 1.4 Hz, 1H), 7.67 (d,
J ) 7.8 Hz, 1H), 7.65 (d, J ) 8.1 Hz, 2H), 7.60 (dd, J ) 7.8 Hz,
J ) 1.4 Hz, 1H), 7.36 (dd, J ) 7.3 Hz, J ) 1.4 Hz, 1H), 6.13 (s,
2H). MS: 217.7 (M + 2H)+2. HRMS: 434.1303 (obsd), 434.1287
(calcd) for C22H20N5O3S (M + H)+. For the tert-butyl analogue
of 22. Mp: 118-120 °C. 1H NMR (CD3OD): δ 8.60 (bs, 1H),
8.19 (d, J ) 8.4 Hz, 2H), 8.13 (dd, J ) 7.7 Hz, J ) 1.4 Hz, 1H),
8.09 (s, 1H), 7.95 (d, J ) 8.4 Hz, 1H), 7.76 (dd, J ) 8.4 Hz, J
) 1.4 Hz, 1H), 7.67 (d, J ) 8.4 Hz, 1H), 7.63 (dd, J ) 7.7 Hz,
J ) 1.5 Hz, 1H), 7.60 (dd, J ) 7.7 Hz, J ) 1.4 Hz, 1H), 7.34
(dd, J ) 7.7 Hz, J ) 1.4 Hz, 1H), 6.14 (s, 2H), 1.09 (s, 9H);
MS: 282.8 (M + 2H)+2. Anal. Calcd for C26H27N5O3S‚1.89TFA‚
1H2O‚0.23HCl: C, 49.19; H, 4.31; N, 9.62; F, 14.10; S, 4.41;
Cl, 1.22. Found: C, 49.14; H, 4.24; N, 9.60; F, 14.44; S, 4.17;
Cl, 1.12.
Compound 8 (570 mg, 1.53 mmol) in anhydrous EtOH (10
mL) was treated with HCl (gas) for 15 min at 0 °C and then
stirred at room temperature for 16 h. After evaporation of
excess HCl (gas) and EtOH, the residue was dissolved in
anhydrous EtOH (10 mL), and treated with (NH4)2CO3 (5
equiv) at room temperature for 24 h. Concentration gave a
residue, which was purified on preparative TLC plates with
10% MeOH in CH2Cl2 to give 10 (400 mg, 65.4%). Mp: 160-
165 °C. ESIMS: 204.2 (M + 2H)+2. HRMS: 407.2199 (obsd),
407.2195 (calcd) for C22H26N6O2.
Compound 10 containing two regioisomers was further
separated by HPLC on a Zorbax protein 10 silica column (2"
× 25 cm) with 90% CH3CN/10% H2O/0.2% TFA at flow rate of
60 mL/min to give 12 and 13. The two regioisomers, as well
as the following pair of 14 and 15, were identified by long-
range proton-carbon correlation spectroscopy. These four
compounds were shown 100% pure on reversed analytical
column and also on Zorbax silica analytical column with 90%
CH3CN/10% H2O/0.2% TFA at flow rate of 1 mL/min.
1
P r op ion a te 12. H NMR (CD3OD): δ 8.36 (s, 1H), 8.17 (s,
1H), 7.75-7.72 (m, 2H), 7.63 (bs, 2H), 7.50-7.48 (m, 2H),
0.4.66 (dd, J ) 9.5 Hz, J ) 14.3 Hz, 1H), 4.55 (dd, J ) 5.5 Hz,
J ) 14.2 Hz, 1H), 4.02-3.92 (m, 2H), 3.14-3.08 (m, 1H), 2.81
(t, J ) 7.0 Hz, 2H), 2.19-1.93 (m, 2H), 1.04 (t, J ) 7.0 Hz,
3H). ESIMS: 407 (M + H)+.
1
P r op ion a te 13. H NMR (CD3OD): δ 8.37 (s, 1H), 8.10 (s,
1H), 7.84 (d, J ) 8.4 Hz, 1H), 7.72 (d, J ) 8.4 Hz, 1H), 7.65-
7.62 (m, 2H), 7.55-7.46 (m, 2H), 4.68 (dd, J ) 9.5 Hz, J )
14.3 Hz, 1H), 4.56 (dd, J ) 5.5 Hz, J ) 14.2 Hz, 1H), 4.04-
3.94 (m, 2H), 3.24-3.18 (m, 1H), 2.83 (t, J ) 7.0 Hz, 2H), 2.19-
1.95 (m, 2H), 1.05 (t, J ) 7.0 Hz, 3H). ESIMS: 407 (M + H)+.
Eth yl 2-(4-Am id in op h en yl)eth yl-3-(5-a m id in oben zim i-
d a zol-1-yl)p r op ion a te (14) a n d Eth yl 2-(4-Am id in op h e-
n yl)eth yl-3-(6-am idin oben zim idazol-1-yl)pr opion ate (15).
By using the same method as described for compounds 12 and
13, Michael addition of 6 (492 mg, 2 mmol) to 7 (286 mg, 2
mmol), followed by Pinner reaction and amination gave
compound 11 (100 mg, 13% for two steps). Mp: 230 °C dec.
HRMS: 407.2200 (obsd), 407.2195 (calcd) for C22H26N6O2.
Compound 11 was further separated by HPLC on a chiral OJ
column with CO2/MeOH/TEA (80/20/0.1) to give 14 and 15.
P r op ion a te 14. 1H NMR (DMSO-d6): δ 9.43-9.08 (m, 6H),
7.74-7.65 (m, 2H), 7.40-7.38 (m, 2H), 7.35-7.00 (m, 4H),
4.67-4.55 (m, 2H), 4.06 (bs, 2H), 3.48 (bs, 2H), 3.20 (bs, 1H),
2.70 (bs, 2H), 1.00 (bs, 3H). ESIMS: 407.2 (M + H)+.
1-[2-[(1,1′-Biph en yl)-4-yl]-2-oxoeth yl]-1H-ben zim idazole-
1
6-ca r boxim id a m id e (20). Mp: 155-157 °C. H NMR (CD3-
OD): δ 8.44 (s, 1H), 8.23 (d, J ) 8.4 Hz, 1H), 8.07 (d, J ) 1.1
Hz, 1H), 7.91 (d, J ) 8.8 Hz, 1H), 7.88 (dd, J ) 8.4, 2H), 7.72
(dd, J ) 8.4 Hz, J ) 1.1 Hz, 4H), 7.52-7.41 (m, 3H), 6.10 (s,
2H). LRMS: 355 (M + H)+. HRMS: 355.1559 (calcd), 355.1554
(obsd). Anal. Calcd for C22H18N4O‚0.7TFA‚1.2HCl‚1.2H2O: C,
56.26; H, 4.50; N, 11.21; Cl, 8.52. Found: C, 56.37; H, 4.86;
N, 11.09; Cl, 8.54.
P r op ion a te 15. 1H NMR (DMSO-d6): δ 9.23-9.12 (m, 6H),
8.41 (s, 1H), 8.21 (s, 1H), 7.84-7.82 (m, 1H), 7.74 (d, J ) 8.1
Hz, 2H), 7.41 (d, J ) 7.8 Hz, 2H), 7.24 (bs, 1H), 4.58-4.56 (m,
2H), 3.95-3.89 (m, 2H), 3.10-3.00 (m, 1H), 2.73-2.71 (m, 2H),
1.90-1.88 (m, 2H), 0.98-0.96 (m, 3H). ESIMS: 407 (M + H)+.
1-[2-(4-Br om op h en yl)-2-oxoet h yl]-1H-b en zim id a zole-
6-ca r bon itr ile (17) a n d 1-[2-(4-Br om op h en yl)-2-oxoet-
h yl]-1H-ben zim id a zole-5-ca r bon itr ile (18). To a suspen-
sion of NaH (60% in mineral oil, 1.92 g, 48 mmol) in THF (80
mL) was added 7 (5.14 g, 36 mmol) and the mixture was stirred
at room temperature for 1 h. 2,4′-Dibromoacetophenone (16)
(10 g, 36 mmol) was added to the reaction mixture, and the
resulting mixture was stirred under reflux for 2 h. The mixture
was diluted with EtOAc (200 mL), washed with water, dried
over MgSO4, and concentrated to give a mixture of two
regioisomers (11 g, 90%). The mixture was isolated by HPLC
on a chiralcel OJ column with MeOH/CO2 (20%/80%) to give
17 and 18.
1-[2-[(1,1′-Biph en yl)-4-yl]-2-oxoeth yl]-1H-ben zim idazole-
1
5-ca r boxim id a m id e (21). Mp: 260-261 °C. H NMR (CD3-
OD): δ 8.41 (s, 1H), 8.22 (s, 1H), 8.20 (d, J ) 8.8 Hz, 2H), 7.87
(d, J ) 8.4 Hz, 2H), 7.73-7.70 (m, 4H), 7.51-7.41 (m, 3H),
6.10 (s, 2H). MS: 355.2 (M + H)+. HRMS: 355.1559 (calcd),
355.1538 (obsd). Anal. Calcd for C22H18N4O‚1.5TFA‚0.08HCl‚
1H2O: C, 54.96; H, 3.98; N, 10.22; Cl, 0.52. Found: C, 54.65;
H, 3.87; N, 10.56; Cl, 0.45.
2-(6-Am id in oben zim id a zol-1-yl)-N-[2′-(a m in osu lfon yl)-
(1,1′-bip h en yl)-4-yl]a ceta m id e (32) a n d 2-(5-Am id in oben -
zim id a zol-1-yl)-N-[2′-(a m in osu lfon yl)(1,1′-bip h en yl)-4-yl]-
acetam ide (33). Compounds 30-33 were prepared as described
in the following procedures. To a suspension of K2CO3 (331
mg, 2.4 mmol) in CH2Cl2 (10 mL) were added 4′-amino(1,1′-
biphenyl)-2-N-(tert-butylamino)sulfonamide (28) (608 mg, 2
mmol) and R-chloroacetyl chloride (271 mg, 2.4 mmol), and
the resulting mixture was stirred at room temperature for 16
h. The mixture was quenched with water (20 mL) and diluted

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4411
with EtOAc (100 mL). The organic layer was dried over MgSO4
and concentrated to give N-[2-(tert-butylaminosulfonyl)(1,1′-
biphenyl)-4′-yl]-2-chloroacetamide (29) (700 mg, 92%) as a
white solid. To a suspension of K2CO3 (552 mg, 4 mmol) in
DMF (10 mL) were added 7 (286 mg, 2 mmol) and 29 (760
mg, 2 mmol), and the mixture was stirred at room temperature
for 16 h. As described above, workup and purification on
preparative TLC plates, followed by preparative HPLC, pro-
vided 26 (240 mg, 56%) and 27 (160 mg, 37%). The two
regioisomers were further modified to the amidines followed
by purification on preparative HPLC to give 32 and 33 in
varied yields (15-50%), respectively.
mL) was added another solution of NaOH (1.56 g, 27.7 mmol)
in water (15 mL), and the resulting mixture was stirred at
room temperature for 4 h and then concentrated. The residue
was dissolved in water (100 mL), and the resulting mixture
was extracted with Et2O (2 × 50 mL). The aqueous layer was
acidified to pH 1 with 6 N HCl and extracted with Et2O (4 ×
50 mL). The organic layers were combined, dried over MgSO4,
and concentrated in vacuo to afford 39 (2.53 g, 90%). Mp: 120-
1
121 °C. H NMR (CD3OD): δ 7.30-7.26 (m, 2H), 6.53 (d, J )
8.1 Hz, 1H), 3.80 (t, J ) 9.2 Hz, 1H), 3.69-3.59 (m, 1H), 3.33-
3.28 (m, 1H), 2.71 (dd, J ) 16.3 Hz, J ) 6.0 Hz, 1H), 2.53 (dd,
J ) 16.1 Hz, J ) 8.4 Hz, 1H). 13C NMR (CD3OD): δ 174.08,
156.03, 133.24, 131.94, 127.18, 120.38, 107.68, 97.44, 52.52,
Aceta m id e 32. Mp: 134-136 °C. 1H NMR (CD3OD): δ 8.73
(bs, 1H), 8.15 (s, 1H), 8.10 (dd, J ) 7.6 Hz, J ) 1.2 Hz, 1H),
7.93 (d, J ) 8.3 Hz, 1H), 7.75 (d, J ) 7.4 Hz, 1H), 7.64 (d, J )
8.4 Hz, 2H), 7.60 (dd, J ) 7.6 Hz, J ) 1.2 Hz, 1H), 7.52 (td, J
) 7.6, J ) 1.4 Hz, 1H), 7.40 (d, J ) 8.8 Hz, 2H), 7.32 (dd, J )
7.6 Hz, J ) 1.2 Hz, 1H), 5.36 (s, 2H). 13C NMR (CD3OD): δ
168.79, 166.75, 143.05, 141.48, 138.93, 137.50, 133.63, 132.92,
131.28, 128.75, 128.59, 124.63, 123.35, 120.96, 120.53, 112.80,
47.51. HRMS: 449.1401 (obsd), 449.1396 (calcd) for C22H21N6O3S
(M + H)+. Anal. Calcd for C22H20N6O3S‚2.05TFA‚0.15HCl‚
1.5H2O: C, 43.86; H, 3.55; N, 11.76; F, 16.35; S, 4.49; Cl, 0.74.
Found: C, 43.92; H, 3.35; N, 12.00; F, 16.55; S, 4.15; Cl, 0.71.
38.40, 37.29. LRMS: 199 (M
C11H8N2O2: C, 65.34; H, 4.98; N, 13.85. Found: C, 65.41; H,
4.94; N, 13.65.
-
H)+. Anal. Calcd for
By using the same procedure, methyl (5-cyano-1H-indol-3-
yl)acetate was saponified in KOH/MeOH at room temperature
1
over 18 h to provide 37. Mp: 196.5-198.5 °C. H NMR (CD3-
OD): δ 7.89 (s, 1H), 7.45 (d, 1H, J ) 8.6 Hz), 7.38 (d, 1H, J )
8.6 Hz), 7.28 (s, 1H), 3.78 (s, 2H). Anal. Calcd for C11H10N2O2:
C, 66.00; H, 4.04; N, 13.99. Found: C, 65.71; H, 4.24; N, 13.94.
(5-Cya n o-1-m eth yl-1H-in d ol-3-yl)a cetic Acid (38). To a
solution of 35 (4 g, 17.5 mmol) in DMF (50 mL) was added
NaH (95%, 0.5 g, 21 mmol), and the mixture was stirred at
room temperature for 20 min. Methyl iodide (2.5 g, 17.5 mmol)
was added, and the resulting mixture was stirred at room
temperature for 2 h. The mixture was diluted with EtOAc,
washed with water (4×), dried over MgSO4, and concentrated
to give a crude of methyl (5-cyano-1-methyl-1H-indol-3-yl)(oxo)-
acetate (36; 6.98 g). The crude was reduced with triethylsilane
and by saponified in NaOH and aqueous methanol as described
1
Aceta m id e 33. Mp: 254 °C dec. H NMR (CD3OD): δ 8.55
(s, 1H), 8.22 (s, 1H), 8.08 (d, J ) 6.6 Hz, 1H), 7.83-7.75 (m,
2H), 7.62 (d, J ) 8.8, 2H), 7.59-7.52 (m, 2H), 7.39 (d, J ) 8.4
Hz, 2H), 7.31 (d, J ) 7.4 Hz, 1H), 5.33 (s, 2H). 13C NMR
(DMSO-D6): δ 165.73, 164.97, 147.72, 142.58, 142.19, 139.43,
138.17, 137.63, 135.23, 132.31, 131.35, 129.69, 127.39, 127.23,
122.20, 121.03, 120.96, 120.17, 118.21, 118.12, 111.24, 47.51.
HRMS: 449.1414 (obsd), 449.1396 (calcd) for C22H21N6O3S (M
+ H)+. Anal. Calcd for C22H20N6O3S‚1.8TFA‚0.25HCl‚1H2O: C,
45.16; H, 3.56; N, 12.34; F, 15.07; S, 4.71; Cl, 1.30. Found: C,
45.38; H, 3.57; N, 12.49; F, 15.19; S, 4.67; Cl, 1.20.
1
for example 37 to provide 38. H NMR (CD3OD): δ 7.97 (t, J
) 0.7 Hz, 1H), 7.43 (d, 1H, J ) 8.6 Hz), 7.37 (d, 1H, J ) 8.6
Hz), 7.29 (s, 1H), 3.81 (s, 3H), 3.74 (d, J ) 0.7 Hz, 2H). LRMS:
215.1 (M + H)+.
2-(6-Am id in ob en zim id a zol-1-yl)-N-[(1,1′′-b ip h en yl)-4-
1
yl]a ceta m id e (30). H NMR (CD3OD): δ 8.47 (bs, 1H), 8.08
2-(5-Am id in o-1H-in d ol-3-yl)-N-[2′-(a m in osu lfon yl)(1,1′-
bip h en yl)-4-yl]a ceta m id e (41). To a stirred solution of 37
(281 mg, 1.4 mmol) and BOP (928 mg, 2.1 mmol) in DMF (10
mL) were added 28 (640 mg, 2.1 mmol) and Et3N (212 mg, 2.1
mmol). After being heated at 50 °C for 18 h, the reaction was
diluted with water, extracted with EtOAc, and washed with
10% HCl (3×), 5% NaOH (2×), and brine (2×). The organic
layer was dried with MgSO4, filtered, and concentrated in
vacuo to afford 40 (370 mg, 54%). A solution of 40 (370 mg,
0.76 mmol) in TFA (10 mL) was refluxed for 2 h. Concentration
and purification via silica gel chromatography using 100%
EtOAc as the eluant, afforded the des-tert-butyl cyano inter-
mediate (280 mg, 86%). This cyano intermediate (880 mg, 2
mmol) was dissolved in dry MeOH (5 mL) and MeOAc (10 mL)
at 0 °C, and the solution was saturated with HCl (gas). The
resulting solution was stirred at room temperature for 18 h.
The mixture was concentrated in vacuo, redissolved in dry
MeOH, and solid (NH4)2CO3 (672.0 mg) was added. The
reaction flask was sealed, and the mixture was stirred at room
temperature for 18 h. The resulting suspension was filtered
through Celite, and the solids were rinsed with dry MeOH.
The filtrate was concentrated in vacuo, followed by purification
on preparative HPLC to afford 41 (450 mg, 50%). 1H NMR
(CD3OD): δ 8.22 (s, 1H), 8.06 (dd, J ) 7.7 Hz, J ) 1.5 Hz,
1H), 7.60-7.54 (m, 6H), 7.43 (s, 1H), 7.36 (d, J ) 8.8 Hz, 2H),
7.30 (d, J ) 7.7 Hz, 1H), 3.91 (s, 2H). HRMS: 448.1437 (obsd),
448.1443 (calcd) for C23H22N5O3S (M + H)+. Anal. Calcd for
(s, 1H), 7.85 (d, J ) 8.5 Hz, 1H), 7.67 (d, J ) 8.8 Hz, 1H), 7.58
(d, J ) 9.1 Hz, 2H), 7.52-7.48 (m, 4H), 7.32 (t, J ) 7.3 Hz,
2H), 7.21 (t, J ) 7.3 Hz, 1H), 5.28 (s, 2H). LRMS: 370.2 (M +
H)+.
2-(5-Am id in oben zim id a zol-1′-yl)-N-[(1,1′′-bip h en yl)-4-
1
yl]a ceta m id e (31). H NMR (CD3OD): δ 8.47 (bs, 1H), 8.21
(s, 1H), 7.77 (td, J ) 8.1 Hz, J ) 4.0 Hz, 2H), 7.65 (d, J ) 8.5
Hz, 2H), 7.57 (dd, J ) 8.1 Hz, J ) 1.1 Hz, 4H), 7.39 (t, J ) 7.3
Hz, 2H), 7.29 (t, J ) 7.3 Hz, 1H), 5.31 (s, 2H). LRMS: 370.2
(M + H)+.
Meth yl (5-Cya n o-1H-in d ol-3-yl)(oxo)a ceta te (35). To a
stirred solution of oxalyl chloride (103.8 g, 818 mmol) in dry
Et2O (500 mL) was added 1H-indole-5-carbonitrile (34) (49.8
g, 350 mmol) in small portions over 15 min at -10 °C, and
the resulting mixture was stirred at room temperature for 3.5
h. The mixture was filtered, and the residue was collected and
rinsed with Et2O. After dry MeOH (400 mL) was added in,
the resulting mixture was stirred at room temperature for 18
h and filtered. The solids were collected and rinsed with MeOH
and Et2O to afford 35 (74.9 g, 94%). Mp: 291-292 °C. 1H NMR
(DMSO-d6): δ 12.85 (bs, 1H), 8.69 (d, J ) 3.3 Hz, 1H), 8.65 (d,
J ) 3.3 Hz, 1H), 7.75 (d, J ) 8.1 Hz, 1H), 7.70 (dd, J ) 8.4 Hz,
J ) 1.4 Hz, 1H), 3.91 (s, 2H). LRMS: 229 (M + H)+.
(5-Cya n o-1H-in d ol-3-yl)a cetic Acid (37) a n d (5-Cya n o-
2,3-d ih yd r o-1H-in d ol-3-yl)a cetic Acid (39). To a solution
of triethylsilane (186.7 mL, 1167 mmol) in trifluoroacetic acid
(250 mL) at -10 °C was added 35 (76.16 g, 334 mmol) in small
potions over 25 min, and the resulting mixture was stirred at
room temperature for 20 h. The mixture was concentrated to
remove the excess solvent and reagent, and the residue was
neutralized with saturated NaHCO3 and extracted with CH2-
Cl2 (3 × 300 mL). The organic layer was concentrated and
purified by column chromatography eluted with MeOH in CH2-
Cl2 (0-5%) to provide methyl (5-cyano-1H-indol-3-yl)acetate
(16.18 g, 22.6%) and methyl (5-cyano-2,3-dihydro-1H-indol-3-
yl)acetate (30 g, 41.6%). To a solution of methyl (5-cyano-2,3-
dihydro-1H-indol-3-yl)acetate (3 g, 13.9 mmol) in MeOH (60
C
23H21N5O3S‚1.15TFA‚1.4H2O: C, 50.32; H, 4.16; N, 11.60; F,
10.85; S, 5.31. Found: C, 50.59; H, 3.91; N, 11.57; F, 10.57; S,
4.92.
The following compounds (70, 71, 77-81) were prepared by
the procedure described for the preparation of 41, and the
yields of the final step were in the range of 5-60%.
2-(5-Am id in o-1H-in d ol-3-yl)-N-[2′-(a m in osu lfon yl)(1,1′-
bip h en yl)-4-yl]-N-m eth yla ceta m id e (70). LRMS: 462 (M
1
+ H)+. H NMR (CD3OD): δ 8.14 (dd, J ) 7.9 Hz, J ) 1.1 Hz,
1H), 8.09 (s, 1H), 7.66 (t, J ) 7.7 Hz, 1H), 7.58 (t, J ) 7.5 Hz,
1H), 7.54-7.49 (m, 4H), 7.37 (d, J ) 7.7 Hz, 1H), 7.27 (d, J )

4412 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Han et al.
8.3 Hz, 2H), 7.00 (s, 1H), 3.79 (s, 2H), 3.31 (s, 3H). 13C NMR
(CD3OD): δ 173.62, 169.34, 144.46, 143.26, 141.69, 141.11,
133.44, 132.92, 132.10, 129.15, 128.67, 128.23, 127.84, 121.36,
allowed to stir at room temperature for 24 h. This final reaction
mixture was evaporated and purified on preparative HPLC.
After lyophilization, pure amidine products were obtained in
varied yields (15-50%). For 72: 1H NMR (CD3OD): δ 8.22 (s,
1H), 8.07 (d, J ) 7.7 Hz, 1H), 7.81 (d, J ) 8.4 Hz, 1H), 7.52-
7.61 (m, 6H), 7.36 (d, J ) 8.2 Hz, 1H), 7.27 (d, J ) 7.3 Hz,
1H), 4.01 (s, 2H). HRMS: 526.0548 (calcd), 526.0538 (obsd)
for C23H21N5SO3Br (M + H)+. Anal. Calcd for C23H20N5SO3Br‚
1.3TFA‚2H2O: C, 43.27; H, 3.59; N, 9.98; S, 4.51; F, 10.43.
Found: C, 43.20; H, 3.37; N, 9.85; S, 4.27; F, 10.29.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[2′-(a m in osu lfon yl)-3-
iod o(1,1′-bip h en yl)-4-yl]a ceta m id e (73). 1H NMR (CD3-
OD): δ 8.15 (s, 1H), 7.98 (d, J ) 7.7 Hz, 1H), 7.76 (s, 1H), 7.60
(d, J ) 8.4 Hz, 1H), 7.55-7.40 (m, 3H), 7.32 (m, 2H), 7.20 (t,
J ) 7.33 Hz, 2H), 3.90 (s, 2H). HRMS: 574.0410 (calcd),
574.0428 (obsd) for C23H21N5SO3I (M + H)+.
119.25, 113.20, 110.99, 38.02, 32.41. Anal. Calcd for C24H23
-
N5O3S‚1.3TFA‚1.4H2O‚0.4CH3CN: C, 50.01; H, 4.32; N, 11.41;
F, 11.04; S, 4.84. Found: C, 50.20; H, 3.97; N, 11.15; F, 10.98;
S, 4.45.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[5-[2-(a m in osu lfon yl)-
1
p h en yl]-2-p yr id in yl]a ceta m id e (71). H NMR (CD3OD): δ
8.30 (s, 1H), 8.23 (s, 1H), 8.11 (dd, J ) 7.7 Hz, J ) 1.5 Hz,
1H), 8.03 (d, J ) 8.8 Hz, 1H), 7.92 (d, J ) 9.1 Hz, 1H), 7.68-
7.53 (m, 4H), 7.49 (s, 1H), 7.37 (d, J ) 7.7 Hz, 1H), 4.04 (s,
2H). HRMS: 449.1396 (calcd), 449.1391 (obsd) for C22H21N6O3S
(M + H)+. Anal. Calcd for C22H20N6O3S‚2TFA‚1.5H2O: C,
44.39; H, 3.58; N, 11.94; F, 16.20; S, 4.56. Found: C, 44.61;
H, 3.39; N, 11.93; F, 16.25; S, 4.23.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[2′-(a m in osu lfon yl)-3-
ch lor o(1,1′-bip h en yl)-4-yl]a ceta m id e (74). H NMR (CD3-
2-(5-Am id in o-1H-in d ol-3-yl)-N-[2′-[(m eth yla m in o)su lfo-
n yl](1,1′-bip h en yl)-4-yl]a ceta m id e (77). 1H NMR (CD3-
OD): δ 8.22 (s, 1H), 7.99 (d, J ) 8.01 Hz, 1H), 7.64-7.49 (m,
6H), 7.44 (s, 1H), 7.36-7.30 (m, 3H), 3.91 (s, 2H), 2.34 (s, 3H).
HRMS: 462.1600 (calcd), 462.1591 (obsd) for C24H24N5O3S (M
+ H)+. Anal. Calcd for C24H23N5O3S‚1.2TFA‚2H2O: C, 49.98;
H, 4.48; N, 11.04; S, 5.05; F, 10.78. Found: C, 49.80; H, 4.03;
N, 11.00; S, 5.15; F, 10.64.
1
OD): δ 8.22 (s, 1H), 8.08 (d, J ) 7.69 Hz, 1H), 7.85 (d, J )
8.42 Hz, 1H), 7.5-7.6 (m, 5H), 7.45 (s, 1H), 7.31 (t, J ) 8.42
Hz, 2H), 4.01 (s, 2H). HRMS: 482.1054 (calcd), 482.1039 (obsd)
for C23H21N5SO3Cl (M + H)+.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[2′-(a m in osu lfon yl)-3-
m eth yl(1,1′-bip h en yl)-4-yl]a ceta m id e (75). 1H NMR (CD3-
OD): δ 8.25 (s, 1H), 8.09 (d, J ) 8.05 Hz, 1H), 7.56 (m, 3H),
7.49 (t, J ) 8.79 Hz, 2H), 7.32 (d, J ) 8.05 Hz, 1H), 7.25 (m,
3H), 3.96 (s, 2H), 2.12 (s, 3H). HRMS: 462.1600 (calcd),
462.1586 (obsd) for C24H24N5O3S (M + H)+.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[2′-[(et h yla m in o)su lfo-
n yl](1,1′-bip h en yl)-4-yl]a ceta m id e (78). 1H NMR (CD3-
OD): δ 8.23 (s, 1H), 8.01 (d, J ) 7.7 Hz, 1H), 7.63-7.48 (m,
6H), 7.44 (s, 1H), 7.35 (d, J ) 8.4 Hz, 2H), 7.31 (d, J ) 7.3 Hz,
1H), 3.92 (s, 2H), 2.70 (q, J ) 7.3 Hz, 2H), 0.92 (t, J ) 7.3 Hz,
3H). HRMS: 476.1759 (calcd), 476.1759 (obsd) for C25H26N5O3S
(M + H)+. Anal. Calcd for C25H25N5O3S‚1.5TFA‚0.8H2O: C,
51.36; H, 4.36; N, 10.69. Found: C, 51.20; H, 4.19; N, 10.88.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[2′-[(d im et h yla m in o)-
su lfon yl](1,1′-biph en yl)-4-yl]acetam ide (79). 1H NMR (CD3-
OD): δ 8.22 (s, 1H), 8.00 (dd, J ) 8.0 Hz, J ) 1.4 Hz, 1H),
7.64-7.28 (m, 10H), 3.91 (s, 2H), 2.37 (s, 6H). HRMS: 476.1556
(calcd), 476.1752 (obsd) for C25H26N5O3S (M + H)+. Anal. Calcd
for C25H25N5O3S‚1.5TFA‚2H2O‚2MeOH: C, 48.19; H, 5.32; N,
9.37; F, 11.43. Found: C, 48.25; H, 5.42; N, 9.28; F, 11.51.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[2′-[(p r op a n yla m in o)-
su lfon yl](1,1′-biph en yl)-4-yl]acetam ide (80). 1H NMR (CD3-
OD): δ 8.22 (s, 1H), 8.00 (d, J ) 7.7 Hz, 1H), 7.62-7.48 (m,
6H), 7.44 (s, 1H), 7.36 (d, J ) 8.4 Hz, 2H), 7.31 (d, J ) 7.43
Hz, 1H), 3.92 (s, 2H), 2.60 (t, J ) 7.3 Hz, 2H), 1.31 (q, J ) 7.3
Hz, 2H), 0.75 (t, 3H). HRMS: 490.1913 (calcd), 490.1910 (obsd)
for C26H28N5O3S (M + H)+.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[2′-(a m in osu lfon yl)-3-
1
flu or o(1,1′-bip h en yl)-4-yl]a ceta m id e (76). H NMR (CD3-
OD): δ 8.21 (s, 1H). 8.15 (d, J ) 7.69 Hz, 1H), 7.85 (t, J )
8.06 Hz, 1H), 7.5-7.6 (m, 4H), 7.45 (s, 1H), 7.3 (d, J ) 7.69
Hz, 1H), 7.25 (d, J ) 9.52 Hz, 1H), 7.18 (d, J ) 9.52 Hz, 1H),
3.98 (s, 2H). HRMS: 466.1349 (calcd), 466.1338 (obsd) for
C
23H21N5SO3F (M + H)+. Anal. Calcd for C23H20N5SO3F‚2TFA‚
0.15H2O: C, 46.20; H, 3.63; N, 10.21; F, 16.88. Found: C,
46.14; H, 3.46; N, 10.51; F, 16.98.
2-(5-Am id in o-2,3-d ih yd r o-1H-in d ol-3-yl)-N-[2′-(a m in o-
su lfon yl)(1,1′-bip h en yl)-4-yl]a ceta m id e (43, 44). To a solu-
tion of 39 (5.0 g, 25 mmol) in DMF (25 mL) was added BOP
(16.41 g, 37 mmol), and the resulting solution was stirred at
room temperature for 30 min. To this solution was added 28
(11.29 g, 37 mmol) followed by Et3N (5.2 mL, 37 mmol), and
the reaction mixture was heated at 60 °C for 18 h. The mixture
was poured into water and extracted with EtOAc (4×). The
organic extracts were washed with water and brine, dried over
MgSO4, and concentrated. The residue was purified by flash
chromatography on a silica gel column (300 g) eluted with
hexane:EtOAc (2:1) to give the racemic coupled product (3 g,
25%). The isomers were then separated by HPLC on a chiral
column eluted with 95% MeOH, 5% H2O, and 0.1% TEA at 7
mL/min and monitored at 230 nm to afford the (-)-enantiomer
((-)-42, 1.49 g) and the (+)-enantiomer ((+)-42, 1.49 g). For
(-)-42. LRMS: 489 (M + H)+ and 511 (M + Na)+. [R]: -52.00°
(c ) 0.400 g/dL in methanol at 25 °C). For (+)-42. LRMS: 489
(M + H)+ and 511 (M + Na)+. [R]: +49.75° (c ) 0.400 g/dL in
2-(5-Am id in o-1-m eth ylin d ol-3-yl)-N-[5-[2-(a m in osu lfo-
n yl)p h en yl]-2-p yr id in yl]a ceta m id e (81). 1H NMR (CD3-
OD): δ 8.29 (s, 1H), 8.22 (s, 1H), 8.10 (d, J ) 6.6 Hz, 1H), 8.00
(d, J ) 6.6 Hz, 1H), 7.91 (dd, J ) 7.7 Hz, J ) 2.6 Hz, 1H),
7.64-7.57 (m, 4H), 7.43 (s, 1H), 7.35 (d, J ) 6.2 Hz, 1H), 4.00
(s, 3H), 3.87 (s, 2H). HRMS: 463.1552 (calcd); 463.1570 (obsd)
for C23H23N6O3S (M + H)+.
2-(5-Am id in o-1H -in d ol-3-yl)-N-[2′-(a m in osu lfon yl)-3-
br om o(1,1′-bip h en yl)-4-yl]a ceta m id e (72). Compounds 72-
76 were made as described in the following procedures. To a
solution of 37 (1 mmol, 0.2 g) in anhydrous CH3CN (10 mL)
was added thionyl chloride (0.3 mL, 4 mmol). The reaction
mixture was warmed at 50 °C for 10 min and then allowed to
cool to room temperature and stirred for 2 h. The solvent and
excess thionyl chloride were removed in vacuo, and the residue
was dried in vacuo. To this residue was added a mixture of an
aniline (58-62) (1.0 equiv) and Et3N (0.14 mL, 1.0 equiv; 2.0
equiv for the HCl salt) in anhydrous CH2Cl2 (10 mL). This
reaction mixture was stirred at room temperature for 2 h, then
concentrated and purified by flash chromatography on a silica
gel column eluted with hexane:EtOAc (3:1) to give pure amides
in about 50% yields. A solution of the amide intermediate (1.24
mmol, 0.7 g) in anhydrous MeOAc (15 mL) and anhydrous
MeOH (0.5 mL) was saturated with dry HCl (gas) at -20 °C
for 20 min. The reaction mixture was left at room temperature
for 18 h, and was evaporated to remove any residual HCl and
the solvents. To the imidate in anhydrous MeOH (15 mL) was
added (NH4)2CO3 (10 equiv), and the reaction mixture was
1
methanol at 25 °C). H NMR (CDCl3): δ 8.16 (d, J ) 7.9 Hz,
1H), 7.79 (s, 1H), 7.62 (m, 2H), 7.56 (m, 2H), 7.46 (m, 2H),
7.34 (m, 2H), 6.55 (d, J ) 8.8 Hz, 1H), 3.93 (m, 1H), 3.68 (s,
1H), 3.49 (m, 1H), 2.83-2.62 (m, 2H), 1.02 (s, 9H).
The same procedure as described in 41 was used to make
(+)-43 from (+)-42 and (-)-44 from (-)-42.
1
Aceta m id e (+)-43. H NMR (CD3OD): δ 8.08 (d, J ) 7.3
Hz, 1H), 7.65-7.30 (m, 9H), 6.60 (d, J ) 8.1 Hz, 1H), 3.90-
3.72 (m, 2H), 3.45 (bs, 1H), 2.83-2.68 (m, 2H); [R]: +38.50°
(c ) 0.200 g/dL in methanol). HRMS: 450.1600 (calcd),
450.1596 (obsd) for C23H24N5O3S1 (M + H)+. Anal. Calcd for
C
23H23N5O3S1‚1.2TFA‚0.5 H2O: C, 51.24; H, 4.27; N, 11.76.
Found: C, 51.38; H, 4.07; N, 11.58.
1
Aceta m id e (-)-44. H NMR (CD3OD): δ 8.08 (d, J ) 7.3
Hz, 1H), 7.65-7.47 (m, 6H), 7.37 (d, J ) 8.4 Hz, 2H), 7.30 (d,
J ) 7.0 Hz, 1H), 6.62 (d, J ) 8.8 Hz, 1H), 3.89-3.77 (m, 2H),
3.44 (bs, 1H), 2.85-2.62 (m, 2H). [R]: +45.05 (c ) 0.202 g/dL
in methanol). HRMS: 450.1600 (calcd), 450.1596 (obsd) for

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4413
C23H24N5O3S1 (M + H)+. Anal. Calcd for C23H23N5O3S1‚
1.3TFA: C, 51.44; H, 4.10; N, 11.72; F, 12.40. Found: C, 51.12;
H, 4.10; N, 11.49; F, 12.40.
1.1 Hz, 1H), 7.69-7.60 (m, 3H), 7.58 (td, J ) 7.7 Hz, J ) 1.4
Hz, 1H), 7.33 (dd, J ) 7.3 Hz, J ) 1.4 Hz, 1H), 4.42 (s, 2H).
To a solution of the crude product (1.0 g, 3.59 mmol) in DMF
(20 mL) at 0 °C was added NaH (95%, 0.181 g, 7.2 mmol). After
the mixture was stirred for 15 min, methyl iodide (0.47 mL,
7.53 mmol) was added, and the resulting mixture was stirred
at room temperature for 18 h. The mixture was extracted with
EtOAc, and the combined extracts were washed with water
(5×), dried over MgSO4, and concentrated. The crude product
was purified by column chromatography eluted with hexane
and EtOAc (1:1) to give a N,N-dimethylsulfonamide (725 mg,
(6-Cya n o-1H-in d a zol-1-yl)a cetic Acid (47). To a solution
of 6-nitro-1H-indazole (45) (10 g, 88.5 mmol) in DMF (50 mL)
were added K2CO3 (30.55 g, 221 mmol) and methyl 2-bromoac-
etate (8.37 mL, 88 mmol), and the resulting mixture was
stirred at room temperatureerature for 18 h. The reaction was
quenched with water, and the precipitate was collected,
washed with water, and dried by air. The solid was dissolved
in hot EtOAc, decolorized with charcoal, filtered through
Celite, and concentrated to give methyl 2-(6′-nitroindazol-1′-
yl)acetate (7.01 g, 34%). The nitro compound (3 g, 12.8 mmol)
was dissolved in MeOH (70 mL) and CHCl3 (20 mL) and
treated with hydrogen at 1 atm in the presence of 10% Pd/C
(0.7 g) at room temperature for 4 h. After additional 20% Pd-
(OH)2/C (0.5 g) was added, the reaction was further treated
with hydrogen at room temperature for 3 h. The mixture was
filtered through Celite, and the filtrate was concentrated. The
residue was purified by flash chromatography with a gradient
mixture of 0% to 100% of EtOAc in CHCl3 to provide methyl
(6-amino-1H-indazol-1-yl)acetate (46) (1.33 g, 51%). To a
solution of 46 (1.33 g, 6.4 mmol) in 2.5 N HCl (20 mL) at 0 °C
was added NaNO2 (0.44 g, 6.4 mmol), and the mixture was
stirred for 1 h. The reaction was neutralized with saturated
Na2CO3 and then was added to a solution of CuCN (0.75 g)
and NaCN (0.632 g) in water (12 mL) and EtOAc (24 mL) at
0 °C. The mixture was stirred at 0 °C for 2 h and then warmed
to room temperature for 1 h. The resulting mixture was
extracted with EtOAc, and the organic extracts were concen-
trated. The residue was purified by flash chromatography on
silica gel with EtOAc:hexane (4:1) to provide methyl 2-(6′-
cyanoindazol-1′-yl)eacetate (0.64 g, 46%). The ester (0.63 g,
2.95 mmol) was saponified in MeOH (10 mL) and H2O (10 mL)
with KOH (215 mg, 3.84 mmol) at room temperature for 1.5
h. The solution then was concentrated in vacuo, diluted with
water, extracted with EtOAc, and then acidified to pH 3 with
1 N HCl. The resulting orange participate was filtered and
1
66%). H NMR (CDCl3): δ 8.28 (d, J ) 8.4 Hz, 2H), 8.08 (d, J
) 7.7 Hz, 1H), 7.64 (t, J ) 7.3 Hz, 1H), 7.61-7.56 (m, 3H),
7.29 (t, J ) 7.3 Hz, 1H), 2.47 (s, 6H). LRMS: 324 (M + H)+.
A solution of the intermediate (725 mg, 2.37 mmol) in MeOH
(20 mL) and CH2Cl2 (15 mL) was treated with hydrogen at
one atmosphere in the presence of 10% Pd/C (250 mg) at room
temperature for 2 h. The resulting mixture was then filtered
through Celite and concentrated to provide 56 (0.56 g, 85.6%).
1H NMR (CDCl3) δ 8.09 (dd, J ) 8.1 Hz, J ) 1.5 Hz, 1H), 7.53
(td, J ) 7.3 Hz, J ) 1.4 Hz, 1H), 7.44 (td, J ) 7.3 Hz, J ) 1.4
Hz, 1H), 7.28 (dd, J ) 7.7 Hz, J ) 1.4 Hz, 1H), 7.26 (s, 1H),
7.22 (dd, J ) 8.4 Hz, 2H), 6.75 (d, J ) 8.4 Hz, 2H), 2.38 (s,
6H). LRMS: 277.4 (M + H)+.
Am in obip h en yls 53-55. Compounds 53-55 were pre-
pared as described for the synthesis of 56. To a solution of 52
(828 mg, 2.48 mmol) in DMF (8 mL) at 0 (C was added NaH
(95%, 71 mg, 2.98 mmol). After the mixture was stirred for 15
min, bromopropane (0.27 mL, 2.97 mmol) was added, and the
resulting mixture was stirred at room temperature for 18 h.
The mixture was worked up, and the nitro intermediate was
hydrogenated as described for preparation of 56 to give 55 (651
mg, 75.9% for two steps).
4′-Am in o-N-p r op a n yl-N-(ter t-b u t yl)[1,1′-b ip h en yl]-2-
1
su lfon a m id e (55). H NMR (CDCl3): δ 8.14 (dd, J ) 8.1 Hz,
J ) 1.5 Hz, 1H), 7.55-7.21 (m, 5H), 6.76 (d, J ) 8.8 Hz, 2H),
2.90 (q, J ) 7.0 Hz, 2H), 1.13 (t, J ) 7.0 Hz, 3H), 1.22 (s, 9H).
LRMS: 347.2 (M + H)+.
1
dried under high vacuum to afford 47. Mp: 189-191 °C. H
4′-Am in o-N-m eth yl-N-(ter t-bu tyl)[1,1′-bip h en yl]-2-su l-
fon a m id e (53). LRMS: 319.2 (M + H)+.
NMR (CD3OD): δ 8.17 (s, 1H), 8.08 (s, 1H), 7.95 (dd, J ) 8.4
Hz, J ) 1.1 Hz, 1H), 7.40 (dd, J ) 8.4 Hz, J ) 1.1 Hz, 1H),
5.30 (s, 1H). LRMS: 202 (M + H)+.
4′-Am in o-N-et h yl-N-(ter t-b u t yl)[1,1′-b ip h en yl]-2-su l-
fon a m id e (54). 1H NMR (CDCl3): δ 8.00 (dd, J ) 8.1 Hz, J )
1.4 Hz, 1H), 7.47-7.21 (m. 5H), 6.73 (d, J ) 8.4 Hz, 2H), 2.71
(t, J ) 7.7 Hz, 2H), 1.57-1.51 (m, 2H), 1.22 (s, 9H), 0.73 (t, J
) 7.7 Hz, 3H). LRMS: 333.2 (M + H)+.
2-(6-Am id in o-1H-in d a zol-3-yl)-N-[5-[2-(a m in osu lfon yl)-
p h en yl]-2-p yr id in yl]a ceta m id e (50). The same procedure
as described in 41 was used to make 50 from the coupling of
47 with 48, followed by Pinner reaction and amination, to form
50. 1H NMR (CD3OD): δ 8.34 (d, J ) 2.6 Hz, 1H), 8.26 (s, 1H),
8.15 (s, 1H), 8.10 (dd, J ) 7.7 Hz, J ) 1.5 Hz, 1H), 8.03 (d, J
) 8.4 Hz, 2H), 7.81 (dd, J ) 8.8 Hz, J ) 2.2 Hz, 1H), 7.65 (dd,
J ) 7.3 Hz, J ) 1.1 Hz, 1H), 7.60 (dd, J ) 6.3 Hz, J ) 1.5 Hz,
1H), 7.55 (dd, J ) 7.3 Hz, J ) 1.5 Hz, 1H), 7.34 (dd, J ) 7.7
Hz, J ) 1.5 Hz, 1H), 5.52 (s, 2H). HRMS: 450.1348 (calcd),
450.1347 (obsd) for C21H20N7O3S1 (M + H)+. Anal. Calcd for
C21H19N7S1O3‚1.8TFA‚1.5H2O: C, 43.34; H, 3.52; N, 14.38, F,
4.70. Found: C, 43.28; H, 3.46; N, 14.18, F, 4.78.
N-(ter t-Bu tyl)-4′-n itr o[1,1′-biph en yl]-2-su lfon am ide (52).
To a solution of 4-bromonitrobenzene (51) (8.16 g, 40.6 mmol)
and 19 (10.43 g, 40.6 mmol) in THF (400 mL) was added
aqueous sodium carbonate (8.6 g in 200 mL of water). The
mixture was stirred at room temperature for 30 min while
being degassed with nitrogen, and then Pd(PPh3)4 (1.0 g, 0.87
mmol) was added. The reaction mixture was refluxed for 18 h
and filtered through Celite. The filtrate was concentrated. The
residue was diluted with water, and the resulting mixture was
extracted with EtOAc. The organic extracts were washed with
brine, dried over MgSO4, and concentrated. The residue was
recrystallized from EtOAc to give 52 (6.83 g, 50.4%). 1H NMR
(CDCl3) δ 8.3-7.2 (m, 8H), 3.6 (s, 1H), 1.2 (s, 9H). LRMS: 335
(M + H)+.
Am in obip h en yls 58-62. In a general procedure, a mixture
of 4-bromo-2-X-aniline (X ) halide or methyl) (57) (10 mmol),
2-(tert-butylaminosulfonyl)phenylboronic acid (19; 2.57 g, 10
mmol), Pd(PPh3)4 (0.23 g, 0.2 mmol), and Na2CO3 (3.18 g, 30
mmol) in THF (100 mL) and water (50 mL) was stirred at room
temperature for 30 min while degassed with nitrogen. The
reaction mixture was then heated at reflux for 18 h and filtered
through Celite. The filtrate was evaporated in vacuo and the
residue was diluted with water. The resulting mixture was
extracted with EtOAc. The organic extracts were washed with
brine, dried (MgSO4), and concentrated. The residue was
purified by flash chromatography on a silica gel column eluted
with hexane:EtOAc (2.5:1) (or hexane for the o-F compound)
to give pure products in 50-60% yields.
4′-Am in o-3′-m eth yl-N-(ter t-bu tyl)[1,1′-bip h en yl]-2-su l-
1
fon a m id e (58). H NMR (acetone-d6): δ 8.07 (d, J ) 7.9 Hz,
1H), 7.60 (t, J ) 7.3 Hz, 1H), 7.5 (t, J ) 7.7 Hz, 1H), 7.34 (d,
J ) 7.5 Hz, 1H), 7.15 (m, 2H), 6.78 (d, J ) 8.1 Hz, 1H), 4.72
(bs, 2H), 4.24 (bs, 1H), 2.20 (s, 3H), 0.98 (s, 9H). LRMS: 319
(M + H)+ and 336 (M + NH4)+.
4′-Am in o-3′-flu or o-N-(ter t-b u t yl)[1,1′-b ip h en yl]-2-su l-
1
fon a m id e (59). H NMR (DMSO-d6): δ 7.96 (d, J ) 7.8 Hz,
1H), 7.54 (t, J ) 7.5 Hz, 1H), 7.46 (t, J ) 7.7 Hz, 1H), 7.24 (d,
J ) 7.5 Hz, 1H), 7.03 (d, J ) 12.9 Hz, 1H), 6.88 (d, J ) 8.4 Hz,
1H), 6.74(t, J ) 9.2 Hz, 1H), 6.45 (s, 1H), 5.24 (s, 2H), 0.95 (s,
9H). HRMS: 323.1230 (calcd), 323.1215 (obsd) for C16H20N2-
SO2F (M + H)+.
4′-Am in o-N-dim eth yl[1,1′-biph en yl]-2-su lfon am ide (56).
A solution of 52 (6.83 g, 20.4 mmol) in CH2Cl2 (15 mL) was
treated with TFA (20 mL) under reflux for 3 h and then
concentrated to give a crude sulfonamide (5.87 g). 1H NMR
(CDCl3): δ 8.31 (d, J ) 8.8 Hz, 2H), 8.20 (dd, J ) 7.7 Hz, J )
4′-Am in o-3′-ch lor o-N-(ter t-bu tyl)[1,1′-bip h en yl]-2-su l-
fon a m id e (60). 1H NMR (CDCl3): δ 8.14 (dd, J ) 7.9 Hz, J )

4414 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Han et al.
1.3 Hz, 1H), 7.53 (t, J ) 7.7 Hz, 1H), 7.45 (t, J ) 7.9 Hz, 1H),
7.36 (d, J ) 2.2 Hz, 1H), 7.29-7.25 (m, 2H), 6.83 (d, J ) 8.0
Hz, 1H), 1.03 (s, 9H). LRMS: 339 (M + H)+ and 356 (M +
NH4)+.
Ack n ow led gm en t. We thank Dr. J ung-Hui Sun for
scaling up compounds 37 and 39.
Refer en ces
4′-Am in o-3′-br om o-N-(ter t-bu tyl)[1,1′-bip h en yl]-2-su l-
1
(1) (a) Bazan, J . F. Protein-protein interactions: big rigs in blood
coagulation. Nature 1996, 380, 21, 23. (b) Tapparelli, C.; Met-
ternich, R.; Ehrhardt, C.; Cook, N. S. Synthetic low-molecular
weight thrombin inhibitors: molecular design and pharmaco-
logical profile. Trends Pharmacol. Sci. 1993, 14, 366-76. (c)
Maraganore, J . M. Thrombin, thrombin inhibitors, and the
arterial thrombotic process. Thromb. Haemostasis 1993, 70,
208-11. (d) Tans, G.; Rosing, J .; Thomassen, M. C.; Heeb, M.
J .; Zwaal, R. F. Comparison of anticoagulant and procoagulant
activities of stimulated platelets and platelet-derived micropar-
ticles. Blood 1991, 77, 2641.
(2) Witting, J . I.; Bourdon, P.; Brezniak, D. V.; Maraganore, J . M.;
Fenton, J . W., II. Thrombin-specific inhibition by and slow
cleavage of hirulog-1. Biochem. J . 1992, 283, 737-43.
(3) Bona, R. D.; Hickey, A. D.; Wallace, D. M. Efficacy and safety of
oral anticoagulation in patients with cancer. Thromb. Haemo-
stasis 1997, 78, 137 and references cited therein.
(4) Coleman, R. W.; Marder, V. J .; Salzman, E. W. In Hemostasis
and Thrombosis: Basic Principles and Clinical Practice, 3rd ed.;
Coleman, R. W., Marder, V. J ., Salzman, E. W., Eds.; J . B.
Lippincott: Philadelphia, 1994; Chapters 1, 9, 57, and 80-86.
(5) Davie, E. W.; Fujikawa, K.; Kisiel, W. The coagulation cascade:
initiation, maintenance, and regulation. Biochemistry 1991, 30,
10363-70.
(6) Mann, K. G.; Nesheim, M. E.; Church, W. R.; Haley, P.;
Krishnaswamy, S. Surface-dependent reactions of the vitamin
K-dependent enzyme complexes. Blood 1990, 76, 1-16.
(7) Ahmad, S. S.; Rawala-Sheikh, R.; Walsh, P. N. Compounds and
assembly of the factor X activating complex. Semin. Thromb.
Hemostasis 1992, 18, 311-23.
(8) (a) Suttie, J . W.; J ackson, C. M. Prothrombin structure, activa-
tion and biosynthesis. Physiol. Rev. 1997, 57, 1-70. (b) Eisen-
berg, P. R.; Siegel, J . E.; Abendschein, D. R.; Miletich, J . P.
Importance of factor Xa in determining the procoagulant activity
of whole-blood clots. J . Clin. Invest. 1993, 91, 1877-83.
(9) Elodi, S.; Varadi, K. Optimization of conditions for the catalytic
effect of the factor IXa-factor VIII complex: probable role of the
complex in the amplification of blood coagulation. Thromb. Res.
1979, 15, 617-29.
(10) Harker, L. A.; Hanson, S. R.; Kelly, A. B. Antithrombotic
strategies targeting thrombin activities, thrombin receptors and
thrombin generation. Thromb. Haemostasis 1997, 78, 736-41.
(11) Al-Obeidi, F.; Ostrem, J . A. Factor Xa inhibitors by classical and
combinatorial chemistry. Drug Discovery Today 1998, 3, 223-
31.
(12) (a) Nagahara, T.; Yukoyama, Y.; Inamura, K.; Kayakura, S.;
Komoriya, S.; Yamaguchi, H.; Hara, T.; Iwamoto, M. Dibasic-
(amidinoaryl)propanoic acid derivatives as novel blood coagula-
tion factor Xa inhibitors. J . Med. Chem. 1994, 37, 1200-7. (b)
Nagahara, T.; Katakura, S.; Yokoyama, Y.; Kanaya, N.; Ina-
mura, K.; Komoriya, S.; Yamada, K.; Yamaguchi, H.; Tanabe,
K. Design, synthesis and biological activities of orally active
coagulation factor Xa inhibitors. Eur. J . Med. Chem. 1995, 30,
139-43.
fon a m id e (61). H NMR (CD3OD): δ 8.07 (dd, J ) 7.8 Hz, J
) 1.2 Hz, 1H), 7.57 (t, J ) 7.6 Hz, 1H), 7.48 (dd, J ) 7.9 Hz,
J ) 1.3 Hz, 1H), 7.45 (d, J ) 2.0 Hz, 1H), 7.28 (dd, J ) 7.6 Hz,
J ) 1.2 Hz, 1H), 7.20 (dd, J ) 8.2 Hz, J ) 2.1 Hz, 1H), 6.88 (d,
J ) 8.3 Hz, 1H), 1.02 (s, 9H). LRMS: 383/385 (M + H)+.
4′-Am in o-3′-iodo-N-(ter t-bu tyl)[1,1′-biph en yl]-2-su lfon a-
m id e (62). 1H NMR (CD3OD): δ 8.07 (dd, J ) 8.1 Hz, J ) 1.3
Hz, 1H), 7.66 (d, J ) 2.0 Hz, 1H), 7.59 (td, J ) 7.6 Hz, J ) 1.5
Hz, 1H), 7.48 (td, J ) 7.7 Hz, J ) 1.5 Hz, 1H), 7.30 (dd, J )
7.6 Hz, J ) 1.5 Hz, 1H), 7.23 (dd, J ) 8.3 Hz, J ) 2.0 Hz, 1H),
6.84 (d, J ) 8.3 Hz, 1H), 1.03 (s, 9H). LRMS: 431 (M + H)+
and 448 (M + NH4)+.
4′-(Met h yla m in o)-N-(ter t-b u t yl)[1,1′-b ip h en yl]-2-su l-
fon a m id e (63). A solution of 28 (2 g, 6.58 mmol) in ethyl
formate (30 mL) was refluxed for 2 h. The mixture was
concentrated in vacuum to afford a formamide intermediate
(2.24 g) as a white solid. To a stirred solution of the above
formamide in THF (70 mL) at 0 °C was added LAH (1 M in
THF, 20.25 mL, 20.3 mmol), and the resulting mixture was
stirred at room temperature for 3 h. The reaction was
quenched with Et2O and water (1:1), filtered through Celite,
and purified by silica gel chromatography with CHCl3 to 63
(1.76 g, 48% for two steps). 1H NMR (CD3OD): δ 8.04 (dd, J )
8.0, J ) 1.1 Hz, 1H), 7.56 (td, J ) 7.7 Hz, J ) 1.1 Hz, 1H),
7.43 (td, J ) 7.7 Hz, J ) 1.1 Hz, 1H), 7.30 (dd, J ) 7.7 Hz, J
) 1.1 Hz, 1H), 7.26 (d, J ) 8.4 Hz, 2H), 6.68 (d, J ) 8.4 Hz,
2H), 2.79 (s, 3H), 0.95 (s, 9H). LRMS 319 (M + H)+.
X-r a y Ma ter ia ls a n d Meth od s. Bovine trypsin was pur-
chased from Worthington (cat. #3707) and used without
further purification. Crystals were grown by vapor diffusion
using 20-µL hanging drops containing 10-30 mg/mL â-trypsin,
35 mM Tris pH 7.5, 2.5 mM benzamidine, 100 mM ammonium
sulfate, and 6-12% W/V PEG 8,000. The drops were equili-
brated at 5 °C over 50 mM Tris pH 7.5, 200 mM ammonium
sulfate, and 12-24% W/V PEG 8,000. Crystals appeared after
1 week. Benzamidine was removed by letting the crystals soak
overnight in a stabilizing solution containing 50 mM Tris pH
7.5, 200 mM ammonium sulfate, and 20% W/V PEG 8,000. The
crystals were transferred to a solution containing 20 mM
sodium phosphate pH 7.5, 20% PEG 8,000 and 0.1% glutaral-
dehyde for 30 min to cross-link. These crystals were then
transferred to a solution containing the inhibitor. The inhibitor
solution was prepared by first dissolving 1 mg of inhibitor in
5 µL of DMSO. This was followed by a 40-fold dilution of the
inhibitor/DMSO solution into the first stabilizing solution.
Data were collected 1 week after inhibitor addition.
A crystal of the trypsin-inhibitor complex was mounted and
sealed in a glass capillary. An R-AXIS image plate detector
was used for X-ray data acquisition. A Rigaku RU-200 rotating
anode X-ray generator operating at 50 kV/100 mA equipped
with a graphite monochromator was used for data collection.
The trypsin data were collected at 4 °C using an Enraf Nonius
cooling device. Data frames of 2° rotation about the spindle
axis, φ, were collected, with exposure times of 30 min/frame,
for total angular rotation ranges about φ of 90°. Data were
processed using the Raxis data processing software (Molecular
Structure Corp.). Crystals grew in space group P212121 with
unit cell parameters: a ) 54.8 Å, b ) 58.9 Å, c ) 67.3 Å. Data
greater than 1σ were used in refinement and were 89.7%
complete. The XPLOR27 program was used for crystallographic
refinement. Simulated annealing (at a maximum temperature
of 3000 °C) was followed by B-factor refinement. The refined
coordinates of trypsin28 were used to calculate the initial
phases for the enzyme-inhibitor structure. The inhibitor was
built with the program QUANTA (Molecular Simulations Inc.).
No major adjustments to the protein model were needed during
the course of the refinements. The final crystallographic
R-factor was 18.4%.
(13) At the time this work was carried out, there were relatively few
reports of small-molecule inhibitors of fXa. Subsequently, more
reports on small fXa inhibitors have published. See review and
references therein: Zhu, B.-Y.; Scarborough, S. M. Recent
advanced in inhibitors of factor Xa in the prothrombinase
complex. Curr. Opin. Cardiovasc. Pulmon. Renal Invest. Drugs
1999, 1, 63-87.
(14) (a) Bode, W.; Schwager, P. The refined crystal structure of bovine
beta-trypsin at 1.8 Å resolution. II. Crystallographic refinement,
calcium binding site, benzamidine binding site and active site
at pH 7.0. J . Mol. Biol. 1975, 98, 693-717. (b) Marquart, M.;
Walter, J .; Deisenhofer, J .; Bode, W.; Huber, R. The geometry
of the active site and of the peptide groups in trypsin, trypsi-
nogen and its complexes with inhibitors. Acta Crystallogr. Sect.
B 1983, 39, 480-90. (c) Banner, D. W.; Hadvary, P. Crystal-
lographic analysis at 3.0 Å - resolution of the binding to human
thrombin of four active site-directed inhibitors. J . Biol. Chem.
1991, 266, 20085-91.
(15) Hinde, A. L.; Radom, L.; Rasmussen, M. A model theoretical
study for alkylation in benzimidazoles: ab initio electrostatic
potentials. Aust. J . Chem. 1979, 32, 11-20.
(16) (a) Quan, M. L.; Liauw, A. Y.; Ellis, C. D.; Pruitt, J . R.; Carini,
D. J .; Bostrom, L. L.; Huang, P. P.; Harrison, K.; Knabb, R. M.;
Thoolen, M. J .; Wong, P. C.; Wexler, R. R. Design and synthesis
of isoxazoline derivatives as factor Xa inhibitors. 1. J . Med.

Amidino Bicyclic fXa Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4415
Chem. 1999, 42, 2752-9. (b) Quan, M. L.; Ellis, C. D.; Liauw,
A. Y.; Alexander, R. S.; Knabb, R. M.; Lam, G.; Wright, M. R.;
Wong, P. C.; Wexler, R. R. Design and synthesis of isoxazoline
derivatives as factor Xa inhibitors. 2. J . Med. Chem. 1999, 42,
2760-73.
C. A.; Mersinger, L. J .; Knabb, R. M. The selective inhibition of
thrombin by peptides of boroarginine. J . Biol. Chem. 1990, 265,
18289-97.
(24) (a) The DuPont group has previously employed the 5-amidinoin-
dole in the design of potent thrombin inhibitors; see: Dominguez,
C.; Duffy, D. E.; Han, Q.; Alexander, R. S.; Galemmo, R. A., J r.;
Park, J . M.; Wong, P. C.; Amparo, E. C.; Knabb, R. M.; Luettgen,
J .; Wexler, R. R. Design and synthesis of potent and selective
5,6-fused heterocyclic thrombin inhibitors. Bioorg. Med. Chem.
Lett. 1999, 9, 925-30. (b) After this work had been completed a
recent publication demonstrated successful use of 5-amidinoin-
dole in the design of both fXa and thrombin inhibitors; see: Zhou,
Y.; J ohnson, M. E. Comparative molecular modeling analysis of
5-amidinoindole and benzamidine binding to thrombin and
trypsin: specific H-bond formation contributes to high 5-ami-
dinoindole potency and selectivity for thrombin and factor Xa.
J . Mol. Recognit. 1999, 12, 235-41.
(25) (a) Wong, P. C.; Crain, E. J ., J r.; Nguan, O.; Watson, C. A.;
Racanelli, A. Antithrombotic actions of selective inhibitors of
blood coagulation factory Xa in rat models of thrombosis.
Thromb. Res. 1996, 83, 117-26. (b) Wong P. C.; Quan, M. L.;
Crain, E. J .; Watson, C. A.; Wexler, R. R.; Knabb, R. M.
Nonpeptide factor Xa inhibitors: I. Studies with SF303 and
SK549, a new class of potent antithrombotics. J . Pharmacol. Exp.
Ther. 2000, 292, 351-7.
(26) Bode, W.; Brandstetter, H.; Mather, T.; Stubbs, M. T., Compara-
tive analysis of haemostatic proteinases: structural aspects of
thrombin, factor Xa, factor IXa and protein C. Thromb. Haemost
1997, 78, 501-11.
(27) Bru¨nger, A. T.; Kuriyan, J .; Karplus, M. Crystallographic R
factor refinement by molecular dynamics. Science 1987, 235,
458-60.
(28) Krieger, M.; Kay, L. M.; Stroud, R. M. Structure and specific
binding of trypsin. Comparison of inhibited derivatives and a
model for substrate binding. J . Mol. Biol. 1974, 83, 209-30.
(17) Miyaura, N.; Yanagi, T.; Suzuki, A. The palladium-catalyzed
cross-coupling reaction of phenylboronic acid with haloarenes
in the presence of bases. Synth. Commun. 1981, 11, 513-9.
(18) 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.ANG.
resolution. J . Mol. Biol. 1993, 232, 947-66.
(19) (a) Brandstetter, H.; Kuehne, A.; Bode, W.; Huber, R.; von der
Saal, W.; Wirthensohn, K.; Engh, R. A. X-ray structure of active
site-inhibited clotting factor Xa. Implications for drug design and
substrate recognition. J . Biol. Chem. 1996, 271, 29988-92 and
references cited therein. (b) Stubbs, M. T., II. Structural aspects
of factor Xa inhibition. Curr. Pharm. Des. 1996, 2, 543-52.
(20) (a) Gillet, V.; J ohnson, A. P.; Mata, P.; Sike, S.; Williams, P.
SPROUT: A program for structure generation. J . Comput.-Aided
Mol. Des. 1993, 7, 127-53. (b) Gillet, V. J .; Newell, W.; Mata,
P.; Myatt, G.; Sike, S.; Zsoldos, Z.; J ohnson, A. P. SPROUT:
Recent developments in the de novo design of molecules. J .
Chem. Inf. Comput. Sci. 1994, 34, 207-17. (c) Gillet, V. J .; Myatt,
G.; Zsoldos, Z.; J ohnson, A. P. SPROUT, HIPPO and CAESA:
tools for de novo structure generation and estimation of synthetic
accessibility. Perspect. Drug Discovery Des. 1995, 3, 34-50.
(21) Maduskuie, T. P., J r.; McNamara, K. J .; Ru, Y.; Knabb, R. M.;
Stouten, P. F. W. Rational Design and synthesis of novel, potent
bis-phenylamidine carboxylate factor Xa inhibitors. J . Med.
Chem. 1998, 41, 53-62.
(22) Klein, S. I.; Czekaj, M.; Gardner, C. J .; Guertin, K. R.; Cheney,
D. L.; Spada, A. P.; Bolton, S. A.; Brown, K.; Colussi, D.; Heran,
C. L.; Morgan, S. R.; Leadley, R. J .; Dunwiddie, C. T.; Perrone,
M. H.; Chu, V. Identification and initial structure-activity
relationships of a novel class of nonpeptide inhibitors of blood
coagulation factor Xa. J . Med. Chem. 1998, 41, 437-50.
(23) (a) Knabb, R. M.; Kettner, C. A.; Timmermans, P. B. M. W. M.;
Reilly, T. M. In vivo characterization of a new synthetic thrombin
inhibitor. Thromb. Haemostasis 1992, 67, 56-9. (b) Kettner,
J M000113T