Diamino benzo[b]thiophene derivatives as a novel class of active site directed thrombin inhibitors. 5. Potency, efficacy, and pharmacokinetic properties of modified C-3 side chain derivatives

Article abstract of DOI:10.1021/jm9903388

A systematic investigation of the structure-activity relationships of the C-3 side chain of the screening hit la led to the identification of the potent thrombin inhibitors 23c, 28c, and 31c. Their activities (1240, 903, and 1271 x 10⁶ L/mol, respectively) represent 2200- and 2900-fold increases in potency over the starting lead la. This activity enhancement was accomplished with an increase of thrombin selectivity. The in vitro anticoagulant profiles of derivatives 28c and 31c were determined, and they compare favorably with the clinical agent H-R-1-[4aS,- 8aS]perhydroisoquinolyl-prolyl-arginyl aldehyde (D-Piq-Pro-Arg-H; 32). The more potent members of this series have been studied in an arterial/venous shunt (AV shunt) model of thrombosis and were found to be efficacious in reducing clot formation. However, their efficacy is currently limited by their rapid and extensive distribution following administration.

Full text of DOI:10.1021/jm9903388

J . Med. Chem. 2000, 43, 649-663
649
Dia m in o Ben zo[b]th iop h en e Der iva tives a s a Novel Cla ss of Active Site Dir ected
Th r om bin In h ibitor s. 5. P oten cy, Effica cy, a n d P h a r m a cok in etic P r op er ties of
Mod ified C-3 Sid e Ch a in Der iva tives
Daniel J . Sall,*,† Dianna L. Bailey,† J olie A. Bastian,† J ohn A. Buben,† Nickolay Y. Chirgadze,†
Amy C. Clemens-Smith,† Michael L. Denney,† Matthew J . Fisher,† Deborah D. Giera,† Donetta S. Gifford-Moore,†
Richard W. Harper,† Lea M. J ohnson,† Valentine J . Klimkowski,† Todd J . Kohn,† Ho-Shen Lin,†
J efferson R. McCowan,† Alan D. Palkowitz,† Michael E. Richett,† Gerald F. Smith,† David W. Snyder,†
Kumiko Takeuchi,† J ohn E. Toth,† and Minsheng Zhang*,‡
Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana 46285
Received J uly 1, 1999
A systematic investigation of the structure-activity relationships of the C-3 side chain of the
screening hit 1a led to the identification of the potent thrombin inhibitors 23c, 28c, and 31c.
Their activities (1240, 903, and 1271 × 106 L/mol, respectively) represent 2200- and 2900-fold
increases in potency over the starting lead 1a . This activity enhancement was accomplished
with an increase of thrombin selectivity. The in vitro anticoagulant profiles of derivatives 28c
and 31c were determined, and they compare favorably with the clinical agent H-R-1-[4aS,-
8aS]perhydroisoquinolyl-prolyl-arginyl aldehyde (D-Piq-Pro-Arg-H; 32). The more potent
members of this series have been studied in an arterial/venous shunt (AV shunt) model of
thrombosis and were found to be efficacious in reducing clot formation. However, their efficacy
is currently limited by their rapid and extensive distribution following administration.
In tr od u ction
provided a better understanding of the structural
requirements necessary for thrombin inhibition, leading
to the preparation of the more potent and selective
inhibitor 1b.7 The compounds from this series are
Thrombosis is one of the leading single causes of
morbidity and mortality in developed societies. The most
widely prescribed anticoagulants to treat thrombotic
disorders, parenteral heparins and oral vitamin K
antagonists (coumadin), are unfortunately less than
ideal due to well-documented liabilities.1,2 Accordingly,
over the last 10 years there has been an intense search
for novel anticoagulant agents with the most recent
efforts being aimed at discovering an oral anticoagulant
alternative to coumadin.
The most widely studied target for antithrombotic
intervention has been the serine protease thrombin
which performs a dual role in thrombogenesis, including
fibrin formation and platelet activation.3,4 Inhibitors of
this enzyme would have the potential to produce an
antithrombotic effect in both arterial and venous throm-
bosis.5 Not surprisingly, the development of direct acting
thrombin inhibitors has been the subject of extensive
research. Recent efforts have been focused on small
molecules possessing an arginine, benzamidine, or a
surrogate of the same, leading to the identification of a
number of potent and selective thrombin inhibitors.6 To
date however, none of these have successfully pro-
gressed through clinical trials as an oral anticoagulant.
As part of our effort to identify structurally novel,
orally active thrombin inhibitors, we undertook a broad
screening approach leading to the identification of the
2,3-disubstituted benzo[b]thiophene derivative 1a .7 Pre-
liminary structure-activity relationship (SAR) studies
distinguished from previously reported thrombin inhibi-
tors by their structural novelty (i.e. they are not derived
from arginine or benzamidine). Despite their unique
structure, these derivatives still displayed competitive
inhibition kinetics, indicating their interaction at the
active site of the enzyme, a result later confirmed by
X-ray crystallography. The promising bioavailability (up
to 50%) of these early derivatives warranted the further
development of this class of agents.
* To whom correspondence should be addressed. Tel: 317-276-6770.
Fax: 317-277-0892.
† Eli Lilly and Company.
To expand our understanding of the SAR around this
novel series of thrombin inhibitors, with an emphasis
‡ Present address: Bristol-Meyers Squibb, Pharmaceutical Research
Institute, Princeton, New J ersey 08543-4000.
10.1021/jm9903388 CCC: $19.00 © 2000 American Chemical Society
Published on Web 02/04/2000

650 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Sall et al.
F igu r e 1. Stereoview of the X-ray crystal structure of inhibitor 1b bound in the active site of human R-thrombin. The principal
interactions between the inhibitor and the active site residues are shown. The proximal binding site (S2) is formed by residues
Trp215, Leu99, His57, Tyr60A, and Trp60D while the distal pocket (S3) is composed of residues Trp215, Ile174, and Glu97A-
Leu99.
Sch em e 1a
a
Reagents: (a) Ph3P4Pd, 2 N aqueous Na2CO3, 1-[2-(4-bromophenoxy)ethyl]pyrrolidine; (b) aminoacyl chloride, TiCl4 or AlCl3; (c)
DIBAL-H or LiAlH4; (d) Et3SiH, TFA.
on identifying more potent derivatives, we conducted a
more systematic SAR study than previously described.
X-ray crystallographic studies provided a clearer un-
derstanding of the interaction of compound 1b in the
active site of thrombin. While the hydrophobic benzo-
[b]thiophene nucleus binds in the specificity pocket (S1)
of the enzyme, the C-3 side chain spans the proximal
and distal pockets of the active site (Figure 1).7 In
contrast, the interaction of the C-2 side chain with
thrombin is not as well defined. However, X-ray analysis
does indicate that the pyrrolidine ring is predominantly
exposed to solvent. Since the interaction of the C-3 side
chain within the active site of thrombin is better
defined, we chose to study the SAR in this region of the
molecule. This report represents the first detailed,
comprehensive account of both the synthesis and sys-
tematic SAR studies surrounding the C-3 side chain,
resulting in the discovery of inhibitors that are 2200-
and 2800-fold more potent than the original screening
hit 1a and are efficacious in a rat model of thrombosis.8
Also reported for the first time are the pharmacokinetic
properties of this novel series of agents.
Ch em istr y. Two general synthetic routes were em-
ployed for the preparation of a majority of the com-
pounds used in this study (Schemes 1 and 2). Deriva-
tives 1a , 4b-g, and 5a -d were prepared according to
Scheme 1, beginning with the commercially available
benzo[b]thiophene-2-boronic acid (2). Installation of the
entire basic C-2 side chain was accomplished through
a Suzuki coupling with 1-[2-(4-bromophenoxy)ethyl]-
pyrrolidine to afford the 2-substituted benzo[b]thiophene
3.9 In the second step, the desired C-3 side chains were
incorporated through Friedel-Crafts acylation10 with
the acid chlorides derived from a variety of amino
benzoic acids to provide the 2,3-disubstituted benzo[b]-
thiophene derivatives 1a , 4b-g. While a variety of
Lewis acids could be employed in the acylation step,
including AlCl3 and SnCl4, TiCl4 was the reagent of
choice since it gave cleaner reactions and higher product
yields. In those cases where the C-3 methylene deriva-

Active Site Directed Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 651
Sch em e 2a
a
Reagents: (a) p-fluorobenzoyl chloride, TiCl4; (b) RH, NaH; (c) DIBAL-H or LiAlH4; (d) Et3SiH, TFA.
Sch em e 3a
Sch em e 4a
a
Reagents: (a) p-anisoyl chloride, AlCl3; (b) methyltriphenyl-
phosphonium bromide, KOt-Bu; (c) pyridine hydrochloride; (d) 1-(2-
chloroethyl)pyrrolidine, Cs2CO3.
tives were required (5a -d ), reductive deoxygenation
was accomplished by a two-step procedure involving
initial hydride reduction to the secondary benzylic
alcohols (DIBAL-H or LiAlH4), followed by a second
reductive deoxygenation (TFA/Et3SiH) to give the de-
sired methylene products 5a -d . Attempts to affect both
the reduction and deoxygenation steps in a single
transformation using NaBH4/TFA were unsuccessful.
The second general synthetic approach that was
employed also utilized benzo[b]thiophene 3 (Scheme 2).
Acylation at the C-3 position with p-fluorobenzoyl
chloride afforded the common intermediate 6. Employ-
ing the methods of Schmid et al.,11 a diverse set of basic
C-4′′ side chains was installed through displacement of
the fluoride with a variety of oxygen-, sulfur-, and
nitrogen nucleophiles to yield derivatives 7a -l. Reduc-
tive deoxygenation to the C-3 methylene derivatives
(8a -i) was accomplished in a manner similar to that
described for the preparation of analogues 5a -d in
Scheme 1.
a
Reagents: (a) Br2; (b) n-BuLi; (c) bis(4-methoxyphenyl)disul-
fide; (d) BBr3; (e) 1-(2-hydroxyethyl)pyrrolidine, Ph3P, DEAD.
9 was conveniently prepared by a Suzuki coupling9 of
benzo[b]thiophene-2-boronic acid with p-bromoanisole.
Acylation with p-anisoyl chloride afforded the 2,3-
disubstituted derivative 10a which was subjected to the
Wittig reaction using a methyl phosphorane to give the
desired olefin 10b.12 Unfortunately, attempts to add
other Wittig or Horner-Wadsworth-Emmons reagents
to the C-3 ketone were unsuccessful, presumably due
to the steric congestion around this site. Deprotection
of the two methyl ethers using pyridine hydrochloride
afforded the corresponding diphenol (10c) which was
alkylated with 1-(2-chloroethyl)pyrrolidine in the pres-
ence of cesium carbonate to give the desired diamine
10d .
Thioether 12c was prepared according to Scheme 4.
Treatment of the 2-aryl benzo[b]thiophene derivative 9
with bromine afforded the 3-bromo intermediate 11.
Metal-halogen exchange gave the 3-lithio species which
Scheme 3 depicts the synthesis of the C-3 olefin
derivative 10d . The starting 2-aryl benzo[b]thiophene

652 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Sall et al.
Sch em e 5a
Sch em e 7a
a
Reagents: (a) 4-methoxyphenol, CuI, K2CO3; (b) n-BuLi; (c)
I2; (d) 4-methoxyphenylboronic acid, (Ph3P)4Pd, 2 N aqueous
Na2CO3; (e) pyridine hydrochloride; (f) 1-(2-chloroethyl)pyrrolidine,
Cs2CO3.
a
Reagents: (a) 6-chloronicotinyl chloride, AlCl3; (b) 1-(2-hy-
droxyethyl)pyrrolidine, Na; (c) AlCl3, EtSH; (d) 1-(2-chloroeth-
yl)pyrrolidine, Cs2CO3.
Sch em e 6a
Sch em e 8a
a
Reagents: (a) o-anisoyl chloride; AlCl3; (b) AlCl3, EtSH; (c)
1-(2-chloroethyl)pyrrolidine, Cs2CO3.
was quenched with commercially available bis(4-meth-
oxyphenyl)disulfide to afford the thioether linked prod-
uct 12a . Deprotection of the methyl ethers using BBr3
at room temperature followed by alkylation of the
resulting diphenol (12b) gave the desired diamine 12c.
The synthesis of the diaryl ether product 15c began
with 3-bromobenzo[b]thiophene (13; Scheme 5), which
was readily prepared by the methods of Cherry et al.13
Subjecting bromide 13 to Ullmann-type conditions
employing 4-methoxyphenol/CuI with sonication af-
forded the C-3 ether derivative 14a .14 Lithiation at the
2-position and subsequent quench with iodine gave the
2-iodo-3-phenoxy product 14b. Suzuki coupling9 with
4-methoxyphenylboronic acid gave the 2,3-disubstituted
derivative 15a . Conversion to the desired diamine
product 15c was accomplished by employing conditions
similar to those used in the preparation of analogue 10d
in Scheme 3.
Scheme 6 depicts the synthesis of isomer 16c which
possesses an ortho-substituted C-3 side chain. Initial
attempts to install the entire C-3 side chain via acyla-
tion of compound 9 with 2-[2-(1-pyrrolidinyl)ethoxy]-
benzoyl chloride failed, possibly due to steric congestion
caused by the ortho substituent. Accordingly, the 2-aryl
derivative 9 was acylated with o-anisoyl chloride/AlCl3
to give the 2,3-disubstituted product 16a . Bis deprotec-
tion of the methyl ethers using EtSH and AlCl3 afforded
the diphenol 16b which could be alkylated to give the
desired diamine 16c.
a
Reagents: (a) TFA.
6-chloronicotinoyl chloride afforded the 3-pyridyl-sub-
stituted compound 17. Installation of the remainder of
the C-3 side chain was accomplished by displacement
of the chloride with the sodium salt of 1-(2-hydroxyeth-
yl)pyrrolidine to give compound 18a . Deprotection of the
C-4′ methyl ether and subsequent alkylation afforded
the desired pyridyl product 18c.
The preparation of the C-4′′ ketone analogue 20 is
shown in Scheme 8. Treatment of the previously re-
ported acetylene 1915 with TFA afforded the desired
ketone product 20 in a single step.
Hybrid analogues which possess the optimal struc-
tural elements identified in this and previous studies
(23c, 28c, and 31c) were synthesized according to
Schemes 9-11. Preparation of the C-3′′ bromo-substi-
tuted derivative 23c began with 6-methoxybenzo[b]-
thiophene (21a ; Scheme 9), which was prepared accord-
ing to the methods of Graham et al.16 Treatment of 21a
with n-BuLi followed by quench with triisopropylborate
afforded the 2-boronic ester which was hydrolyzed under
acidic conditions to give the 6-methoxybenzo[b]thiophene-
2-boronic acid (21b). Coupling of 21b to 4-[2-(1-pyrro-
lidinyl)ethoxy]-1-bromobenzene under Suzuki condi-
The C-3 pyridyl derivative 18c was prepared from
intermediate 9 according to Scheme 7. Acylation with

Active Site Directed Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 653
Sch em e 9a
Sch em e 11a
a
Reagents (a) NaH, 1-(2-hydroxyethyl)pyrrolidine; (b) n-BuLi;
+
(c) MgBr2; (d) compound 27; (e) 10% Pd-C, 25% aqueous NH4
a
Reagents: (a) n-BuLi; (b) i-PrO)3B; (c) HCl; (d) [(Ph)3P]4Pd, 2
HCO2-; (F) DiBAL-H; (g) Et3SiH, TFA.
N aqueous Na2CO3, 4-[2-(1-pyrrolidinyl)ethoxy]-1-bromobenzene;
(e) 3-bromo-4-[(1-pyrrolidinyl)methyl]benzoyl chloride, AlCl3; (f)
LiAlH4; (g) Et3SiH, TFA; (h) EtSH, AlCl3.
derivative 26. Acylation with 3-methoxy-4-[(1-pyrrolidi-
nyl)methyl]benzoyl chloride in the absence of a Lewis
acid gave product 27 which possesses the entire C-3 side
chain. Treatment of 27 with the Grignard reagent
derived from 4-[2-(1-pyrrolidinyl)ethoxy]bromobenzene
installed the requisite C-2 side chain of the 2,3-disub-
stituted benzo[b]thiophene 28a . Reductive deoxygen-
ation of the C-3 ketone and deprotection of the C-6
methoxy group gave the desired product 28c.
Sch em e 10a
The final target of the study, the C-2 pyridyl deriva-
tive 31c, was prepared from 2,5-dibromopyridine (29)
according to Scheme 11. Treatment of pyridine 29 with
the sodium alkoxide of 1-(2-hydroxyethyl)pyrrolidine
afforded the incipient C-2 side chain 30. Conversion of
bromide 30 to the corresponding Grignard reagent and
its addition to the 2-dimethylbenzo[b]thiophene inter-
mediate 27 (Scheme 10) afforded the 2,3-disubstituted
benzo[b]thiophene analogue 31a . Conversion to the
desired target 31c was accomplished as discussed above
in Scheme 10.
Resu lts a n d Discu ssion
The thrombin inhibitory activities of the compounds
in this study were measured as apparent association
constants (Kass).7,18 The results of structure-activity
relationship (SAR) studies in the C-3 side chain are
shown in Tables 1-6, beginning at the C-3 linker and
extending to the basic pyrrolidine terminus of the side
chain. Conversion of the C-3 carbonyl to the correspond-
ing olefin (10d ) enhanced activity 2-fold while reductive
deoxygenation down to the methylene analogue 5a
increased thrombin inhibition 8-fold. Replacement of the
C-3 ketone with a sulfur or oxygen atom also increased
thrombin inhibitory activity relative to ketone 1a (12c
and 15c, respectively).
The data in Table 2 summarize a few of the structural
requirements surrounding the C-3 phenyl ring. While
the more conformationally flexible, acyclic analogue 4e
was significantly less potent than the parent 1a , the
pyridyl derivative 18c retained comparable activity.
Altering the substitution pattern of the C-3 aryl ring
in the form of the 1,3- and 1,2-disubstituted derivatives
4f and 16c, respectively, led to diminished potency.
a
Reagents: (a) LDA; (b) 4-benzyloxybenzaldehyde; (c) MeSO3H;
(d) 3-methoxy-4-[(1-pyrrolidinyl)methyl]benzoyl chloride, 135 °C;
(e) 4-[2-(1-pyrrolidinyl)ethoxy]phenyl magnesium bromide; (f) 10%
Pd-C, NH4+HCO2; (g) LiAlH4; (h) Et3SiH, TFA.
tions9 allowed installation of the entire C-2 side chain
in a single step, affording product 22. Acylation at the
3-position with 3-bromo-4-[(1-pyrrolidinyl)methyl]ben-
zoyl chloride/AlCl3 gave the 2,3-disubstituted derivative
23b. Reductive deoxygenation of the C-3 ketone (1:
LiAlH4; 2: Et3SiH, TFA) followed by cleavage of the
methyl ether at C-6 (EtSH, AlCl3) afforded target 23c.
The C-3′′ methoxy analogue 28c was prepared using
the methods of Godfrey et al.17 (Scheme 10). Treatment
of N,N-dimethylthioformamide with LDA and quench
of the resulting anion with 4-benzyloxybenzaldehyde
afforded condensation product 25. Acid-catalyzed cy-
clization gave the 2-dimethylamino benzo[b]thiophene

654 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Sall et al.
Ta ble 4. Effects of Modifying the Length of the C-4′′ Tether:
Ta ble 1. Effects of the C-3 Linker on Thrombin Inhibition:
Analogues 1a , 5a , 10d , 12c, and 15c
Analogues 5a -d and 8c
a
cmpd
X
Kass (L/mol × 106)
a
cmpd
X
Kass (L/mol × 106)
5a
5d
8c
5b
5c
O(CH2)2
O(CH2)3
O(CH2)4
(CH2)2
CH2
3.43 ( 0.55
0.61 ( 0.13
0.28 ( 0.02
6.19 ( 0.09
16.0 ( 2.30
1a
10d
5a
12c
15c
CdO
CdCH2
CH2
S
0.43 ( 0.07
0.77 ( 0.07
3.43 ( 0.55
3.08 ( 0.36
2.07 ( 0.21
O
a
Represents the apparent association constant as measured by
a
Represents the apparent association constant as measured by
the methods of Smith et al., ref 17.
the methods of Smith et al., ref 17.
Ta ble 5. Modification of the C-3 Terminal Amine: Derivatives
1a , 5a , 7j-l, and 8d -i
Ta ble 2. Structure-Activity Relationships of the C-3 Aryl
Ring: Derivatives 4e, 4f, 16c and 18c
Kass (L/mol × 106)
a
cmpd
X
Y
a
cmpd
R
Kass (L/mol × 106)
1a
7j
7k
7l
5a
8d
8e
8f
8g
8h
8i
CdO
CdO
CdO
CdO
H, H
H, H
H, H
H, H
H, H
H, H
H, H
1-pyrrolidinyl
1-piperidinyl
1-hexamethyleneimino
1-morpholinyl
1-pyrrolidinyl
NH2
NHEt
NEt2
NMe2
N(i-Pr)2
0.43 ( 0.07
0.16 ( 0.01
0.20 ( 0.02
0.30 ( 0.10
3.43 ( 0.55
0.24 ( 0.02
2.24 ( 0.13
12.80 ( 1.66
1.91 ( 0.12
6.75 ( 0.95
2.11 ( 0.46
1a
0.43 ( 0.07
4e
<0.005
18c
0.63 ( 0.19
N(n-Bu)2
4f
0.04 ( 0.002
<0.005
a
Represents the apparent association constant as measured by
the methods of Smith et al., ref 17.
16c
oxygen of lead 1a by carbon (3d ) led to a 2-fold increase
in activity, while incorporation of other heteroatoms
such as sulfur and nitrogen (8a and 8b, respectively)
had only a modest impact on activity relative to com-
pound 5a . Ketone derivative 20 was 3-fold less active.
Although the carbon analogue 3d was more potent than
its ether counterpart 1a , subsequent SAR studies were
conducted using ether linkages at C-4′′ due to the ease
of constructing aryl ether bonds relative to aryl alkyl
bonds.
a
Represents the apparent association constant as meas-ured
by the methods of Smith et al., ref 17.
Ta ble 3. Effects of the C-4′ Linker on Thrombin Inhibition:
Analogues 1a , 5a , 5d , 8a , 8b, 20
X-ray crystallography studies on benzo[b]thiophene
1b show that the C-3 side chains of inhibitors 1a and
5a span the proximal (S2) and distal sites of the
thrombin active site (Figure 1).7,19 The pyrrolidine ring
appears to interact at the lipophilic, distal site which is
composed of residues Trp215, Ile174, and Glu97A-
Leu99. In an effort to enhance this interaction, the
length of the tether between the pyrrolidine ring and
the C-4′′ site was varied (Table 4). While lengthening
the tether decreased potency (5d and 8c), shortening
the distance between the pyrrolidine and the C-3 phenyl
ring increased thrombin inhibitory activity (5b and 5c).
The greatest increase was observed for the pyrrolidi-
nylmethyl derivative 5c which was 5-fold more potent
than parent 5a .
Kass (L/mol × 106)
a
cmpd
X
Y
1a
3d
5a
8a
8b
20
CdO
CdO
CH2
CH2
CH2
CH2
O
0.43 ( 0.07
0.84 ( 0.07
3.43 ( 0.55
2.27 ( 0.16
3.04 ( 0.45
0.97 ( 0.01
CH2
O
S
NH
CdO
a
Represents the apparent association constant as measured by
the methods of Smith et al., ref 17.
The SAR studies progressed next to the linking group
between the 2-(1-(pyrrolidinyl)ethyl side chain and the
C-4′′ position (Y; Table 3). Replacement of the ether
The key structural elements of the pyrrolidine ring
were also modified to investigate interactions at the

Active Site Directed Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 655
Ta ble 6. Thrombin Inhibitory Activity and in Vitro Anticoagulant Profile of Hybrid Structures 23c, 28c, and 31c as Well as 32
a
cmpd
X
Y
Z
Kass (L/mol × 106)
2 × TTb (µg/mL)
2 × aPTTb (µg/mL)
2 × PTb (µg/mL)
23c
28c
31c
1a
OH
OH
OH
Br
OMe
OMe
CH
CH
N
1240 ( 249
903 ( 99
1271 ( 145
0.43 ( 0.07
531 ( 198
0.48
0.02
0.02
1.45
0.42
0.23
1.26
0.40
0.34
>44.5
>44.5
>44.5
D-Piq-Pro-Arg-Hc (32)
0.02
0.60
1.42
a
b
Represents the apparent association constant as measured by the methods of Smith et al. ref 17. Represents the concentration of
inhibitor necessary to double the thrombin time (TT), activated partial thromboplastin time (aPTT), and prothrombin time (PT) as measured
by the methods of Smith et al. refs 17 and 20. c The full chemical name is(H-R-1-[4aS,8aS]-perhydroisoquinolyl-prolyl-arginyl aldehyde.
distal pocket. Expanding the size of the pyrrolidine ring
to the six- and seven-membered analogues, as well as
incorporation of an oxygen atom (7j-l, respectively), led
to decreased activity relative to parent 1a . Decreasing
the lipophilic environment surrounding the amine, as
in the primary or secondary amines 8d and 8e, respec-
tively, also diminished potency relative to the pyrroli-
dine derivative (5a ). The importance of lipophilicity is
consistent with the interaction of the pyrrolidine with
the lipophilic, distal pocket. Opening the pyrrolidine
ring to the acyclic, tertiary diethylamine (8f) affected a
4-fold increase in potency. However, increasing the size
of the nitrogen substituents to di-i-propyl (8h ) or di-n-
butyl (8i) reduced potency relative to the diethyl ana-
logue, possibly due to unfavorable steric interactions at
the distal site.
In past and present structure-activity relationships
studies on the C-3 side chain,7 a number of structural
modifications which increase thrombin inhibitory activ-
ity have been made. Many of the optimal structural
elements identified have been incorporated together into
a single molecule in order to gauge the levels of
thrombin inhibition that might be achieved within this
series (23c, 28c, and 31c; Table 6). Each derivative
contains the reduced C-3 methylene linker (from Table
1) as well as the contracted C-4′′ side chain (from Table
4), both of which were discovered during the current
study. In addition, they possess a C-6 hydroxyl which
has been shown to increase thrombin inhibitory activity
by 6-fold, presumably through the formation of a
hydrogen bond with the carboxyl group of Asp189 in the
S1 specificity pocket of thrombin.7 Finally, each is
substituted at C-3′′ with either a bromine or methoxy
which have been reported to enhance potency by 8-fold
and 3.5-fold, respectively, presumably through interac-
tion at the lipophilic, proximal S-2 binding site of
thrombin.20 Analogue 31c is distinguished from deriva-
tive 28c by the fact that the C-2 aryl ring has been
replaced by a pyridine. As the data in Table 6 indicate,
combining the optimal structural elements into a single
molecule, as in derivatives 23c, 28c, and 31c, resulted
in very potent thrombin inhibition. Relative to the
screening hit 1a , thrombin inhibition has been enhanced
by 2800-, 2200-, and 2900-fold for derivatives 23c, 28c,
and 31c, respectively. On the basis of individual effects
of the C-6 hydroxyl (6-fold),7 the C-3′′ substituents (Br,
8-fold; OMe, 3.5-fold),20 the C-3 methylene (8-fold), and
the contracted C-4′′ side chain (4.5-fold), the activity of
hit 1a would be expected to be increased by 1700-fold
in the case of 23c and 750-fold in the case of 28c.
The determination of the in vitro anticoagulant
profiles for compounds 23c, 28c, and 31c was ac-
complished using the methods of Smith et al.17,21 (Table
6). Briefly, compounds 23c, 28c, and 31c were added
to human plasma over a range of concentrations to
determine prolongation effects on the thrombin time
(TT), activated partial thromboplastin time (aPTT), and
prothrombin time (PT) assays. The concentrations of
inhibitor that are required to double the clotting time
for each assay are reported as 2 × TT, 2 × aPTT, and 2
× PT, respectively. Whereas derivative 23c doubled the
TT at a concentration of 0.48 µg/mL, only 0.02 µg/mL
of both 28c and 31c were required. The activities of
thrombin inhibitors 23c, 28c, and 31c in the human
plasma aPTT and PT assays are also reported in Table
6. Consistent with the TT assay, lower concentrations
of 28c (0.42 and 0.40 µg/mL, respectively) and 31c (0.23
and 0.34 µg/mL, respectively) were required to double
the aPTT and PT, relative to 23c (1.45 and 1.26 µg/mL,
respectively). However, in each case the concentrations
of inhibitor necessary to double the aPTT and PT are
significantly higher than those in the TT assay. This is
consistent with the larger amounts of thrombin that are
generated in both the aPTT and PT assays. The lower
activity of the more lipophilic bromide 23c in the
anticoagulant assays, relative to 28c and 31c, may be
due in part to the increased interaction with plasma
lipids that are present in the assay.22
The thrombin inhibitory activity and in vitro antico-
agulant profile of 23c, 28c, and 31c were compared to
the tripeptide arginal, R-1-Piq-Pro-Arg-H (H-R-1-[4aS,-
8aS]-perhydroisoquinolyl-prolyl-arginyl aldehyde; 32),23
to assess the significance of the novel compounds from
this study. This potent tripeptide thrombin inhibitor
entered phase I clinical trials as an oral anticoagulant
candidate, and therefore, its in vitro activity could serve
as a benchmark for the current work. As reported in
Table 6, the thrombin times (TT) of benzo[b]thiophenes
28c and 31c were identical to that of the arginal 32,
doubling the TT at concentrations of 0.02 µg/mL. In the
aPTT and PT assays, however, thrombin inhibitors 28c
(0.42 and 0.40 µg/mL, respectively) and 31c (0.23 and
0.34 µg/mL, respectively) were more potent than com-
pound 32 (0.60 and 1.42 µg/mL, respectively), with the
pyridyl analogue being approximately 3-fold more ac-
tive. A number of factors could be responsible for the

656 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Sall et al.
F igu r e 2. Effects of thrombin inhibitors 28c, 31c, and 32 on
net clot weight in the rat arterial-venous (AV) shunt model.
F igu r e 3. Plasma drug concentrations following intravenous
administration of thrombin inhibitors 28c, 31c (5 mg/kg), and
32 (2.3 mg/kg) to Rats.
greater potency of derivatives 28c and 31c in the aPTT
and PT assays, including a faster on-rate relative to the
arginine aldehyde 32, inhibition of alternate elements
of the aPTT and PT pathways or, by affecting in a
different fashion, a thrombin-mediated function in the
aPTT and PT pathways.
relative to arginal 32 were due to the high volumes of
distribution to which they undergo (volumes of 37.4 and
27.5 L/kg compared to 0.4 L/kg; Table 8). Since the
majority of thrombin resides in the plasma component,
the relatively high volumes of distribution and resulting
low plasma levels of inhibitor may account for the
efficacy differences of inhibitors 28c and 31c relative
to arginal 32, despite the comparable in vitro antico-
agualnt profiles.
Compounds 28c and 31c were next evaluated for their
ability to inhibit in vivo blood clot formation in the rat
arterial/venous (AV) shunt model of thrombosis (Figure
2).24 The clinical agent 32 was also used as a benchmark
in this model. The average clot weight of 20 vehicle-
treated rats (control) was calculated to be 41.8 ( 1.6
mg (mean ( sem). Infusion of compound 32 (0.3-1.1
mg/kg/h for 30 min), beginning 15 min prior to opening
the shunt, reduced the average clot size in a dose-related
manner. The dose-response curve was steep with a
calculated ED50 of 0.7 mg/kg/h. Infusion of compound
28c (1-20 mg/kg/h) and compound 31c (2.25-18 mg/
kg/h) also reduced the average clot size in a dose-related
manner. The dose-response curves for 28c and 31c
were shallow, and the calculated ED50 values were 8.8
and 6.6 mg/kg/h, respectively. Despite the comparable
in vitro anticoagulant profiles, derivatives 28c and 31c
were approximately 10-fold less efficacious in this model
than the tripeptide arginal 32.
To better understand the discrepancy between the in
vitro anticoagulant profile and the observed in vivo
efficacy, rats were administered a single IV bolus dose
of each thrombin inhibitor (28c, 31c, and 32) and the
plasma levels of parent drug monitored over time
(Figure 3). The pharmacokinetic properties that were
calculated from these studies are reported in Table 8.25
Following a 2.3 mg/kg dose of arginal 32, maximal
plasma concentrations (Cmax) reached 11.57 µg/mL and
decreased monoexponentially with a half-life of 0.8 h.26
Area under the plasma concentration versus time curve
(AUC) and volume of distribution were 7.18 µg‚h/mL
and 0.4 L/kg, respectively.27 In contrast, a 2-fold higher
dose (5 mg/kg) of benzo[b]thiophenes 28c and 31c
resulted in significantly lower Cmax and AUC values
(Cmax’s of 0.32 and 0.64 µg/mL and AUC’s of 0.35 and
0.40 µg‚h/mL for derivatives 28c and 31c, respectively)
and moderate half-lives of 2-3 h. Perhaps the low
observed plasma levels of analogues 28c and 31c
To better understand the nature of the high volumes
of distribution observed for the benzo[b]thiophenes,
tissue distribution studies were undertaken with benzo-
[b]thiophene 31c. Samples of plasma, liver, kidney,
lung, and heart were obtained from male rats at various
times after a 5 mg/kg intravenous dose of 31c and
assayed for parent drug concentrations.28 The results
are shown in Figure 4. Only low levels of benzo[b]-
thiophene 31c were observed in the plasma from 15 min
to 5 h following drug administration (-9-; plasma 31c
concentrations were multiplied by 10 in order to show
them on the graph). In contrast, very high concentra-
tions were detected in each of the tissues. Maximal
tissue concentrations occurred 30 min posttreatment
and declined slowly. The highest concentrations were
found in the lung tissue (-b-; 77.7 µg/g), followed by
the kidney (-O-; 36.4 µg/g), heart (-1-; 21.3 µg/g), and
liver (-3-; 3.6 µg/g). Maximal plasma concentrations
(at 10 min) were 0.73 µg/mL. Tissue-to-plasma ratios
at 2 h posttreatment ranged from 30 for the liver to over
800 for the lung. The relatively low concentrations in
liver were likely the result of its capacity to metabolize
parent 31c. Appreciable amounts of a component whose
molecular weight corresponded to the glucuronide of 31c
were found in the liver extracts, but not in the kidney.29
The data in Figure 4 suggest that following intravenous
administration the benzo[b]thiophenes undergo both
rapid and extensive distribution to various tissues,
making them less available to inhibit thrombin in the
plasma.
A number of structural modifications were necessary
to achieve the high levels of thrombin inhibition dis-
played by derivatives 23c, 28c, and 31c. To determine

Active Site Directed Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 657
Ta ble 7. Serine Protease Selectivity of Derivatives 28c, 1a , 1b, and 32
apparent Kass values (L/mol × 106)a
cmpd
human thrombin
bovine trypsin
human fXa
human plasmin
human ntPA
human urokinase
28c
1a
1b
903 ( 99
0.43 ( 0.07
24.3 ( 2.0
531 ( 198
<0.009
0.003
<0.009
27.0
<0.002
0.01
0.015
2.70
0.007
0.003
0.009
1.51
0.008
0.03
0.009
0.03
<0.009
<0.009
<0.009
0.018
D-Piq-Pro-Arg-Hb (32)
a
Represents the apparent association constant as measured by the methods of Smith et al., ref 17. Kass values for each serine protease,
except thrombin, are the average of two determinations where the variation in the assay is (10%. Kass values for thrombin are the mean
b
of n ) 3, showing the standard deviation. The full chemical name is (H-R-1-[4aS,8aS]-perhydroisoquinolyl-prolyl-arginyl aldehyde.
activities of Kass ) 900-1200 × 106 L/mol. In addition,
Ta ble 8. Summary of the Pharmacokinetic Parameters for
Thrombin Inhibitors 28c, 31c, and 32 Following iv
Administration to Rats
the levels of thrombin selectivity for this series are
among the highest yet reported, being similar to that
found with hirudin.18 The observed levels of thrombin
compound
inhibition translated into potent in vitro anticoagulant
effects in the thrombin time (TT), activated partial
thromboplastin time (aPTT), and prothrombin time (PT)
assays. The in vitro thrombin inhibition and anticoagu-
lant activity achieved in derivatives 28c and 31c equals
or surpasses the potency of the clinical thrombin inhibi-
tor candidates, R-1-Piq-Pro-Arg-H and Efegatran.18 In
a rat model of arterial/venous thrombosis, compounds
28c and 31c were found to be efficacious in reducing
blood clot formation. However, the efficacy of the current
analogues is limited by their rapid and extensive
distribution following administration.
parameter
32
28c
31c
dose (mg/kg)
2.3
11.57
7.18
0.8
5.0
0.32
0.35
2.7
5.0
0.64
0.40
2.0
Cmax (µg/mL)
AUC0-4h (µg‚h/mL)
half-life (h)
volume (L/kg)
0.4
37.4
27.5
Exp er im en ta l Section
The experimental details for the synthesis and the physical
chemical data for the amino acids used in the preparation of
compounds 4-d , 4f, 4g, 23a , and 27 are included in the
Supporting Information. The physical chemical properties of
compounds 1a , 4b-g, 5a -d , 7a -l, 8a -i, 10a , 16a , and 17
are reported in the Supporting Information.
Melting points were determined on a Thomas-Hoover capil-
lary melting point apparatus calibrated with known com-
pounds and are uncorrected. Proton nuclear magnetic reso-
nance spectra (1H NMR) were obtained on either a Varion
Mercury-400 or General Electric QE-300 spectrometer using
either deuterated dimethyl sulfoxide (DMSO-d6), deuterated
chloroform (CDCl3), or deuterated methanol (CD3OD) as the
solvent, and chemical shifts are reported in parts per million
(ppm) relative to dimethyl sulfoxide (DMSO; 2.49 ppm),
chloroform (CHCl3; 7.25 ppm), or methanol (CH3OH; 3.30
ppm). Infrared spectra (IR) were run either as solutions in
CHCl3 or as pellets in KBr and were recorded on a Nicolet
510P spectrometer. Field desorption mass spectra (FDMS)
were obtained on a VG Analytical VG70SE mass spectrometer.
Fast atom bombardment high resolution mass spectra (FAB-
HRMS) were obtained on a VG Analytical VGZAB-2SE mass
spectrometer. Combustion analyses were performed on a
Control Equipment Corporation 440 elemental analyzer and
were within 0.4% of the calculated values. Medium pressure
liquid chromatography (MPLC) was performed using PrepPak
silica gel cartridges on a Waters Prep LC/System 500A. Flash
chromatography was performed over silica gel 60 (230-400
mesh ASTM). Preparative centrifugal thin-layer chromatog-
raphy (PCTLC) was performed on a Harrison model 7924
chromatotron with Merck silica gel 60 PF254 containing CaSO4‚
0.5H2O binder. All reactions, regardless of the solvent used,
were performed under a positive N2 flow with anhydrous
solvents.
F igu r e 4. Concentrations of analogue 31c in plasma and
selected tissues following a 5 mg/kg dose to rats.
whether these analogues retain the same levels of
thrombin specificity observed earlier for compounds 1a
and 1b, derivative 28c was tested for its ability to
inhibit other serine proteases, including the fibrinolytic
proteases as well as trypsin. As reported in Table 7,
analogue 28c is exceptionally selective for thrombin
relative to the other serine proteases tested, even
surpassing the high levels of specificity reported for
compound 1b.7 This series of agents represents one of
the most selective classes of thrombin inhibitors re-
ported to date.
Con clu sion s
Inhibitor activity against human R-thrombin and other
serine proteases were measured as apparent association
constants (Kass) which were derived from inhibition kinetics,
according to the methods of Smith et al.18 Briefly, enzyme
inhibition kinetics were performed in 96-well polystyrene
plates, and reaction rates were determined from the rate of
hydrolysis of appropriate p-nitroanilide substrates at 405 nm
This report describes a systematic SAR study sur-
rounding the C-3 side chain of a structurally novel class
of diamino benzo[b]thiophene thrombin inhibitors, de-
rived from the screening hit 1a . Through a series of
structural modifications, the inhibitory activity of the
hit 1a was enhanced by up to 2800-fold, achieving

658 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Sall et al.
using a Thermomax plate reader from Molecular Devices (San
Francisco, CA). The same protocol was followed for all enzymes
studied: 50 µL of 0.6 M Tris/0.15 M NaCl, pH 7.4, buffer was
added in each well, followed by 25 µL of inhibitor solution and
25 µL enzyme solution; within two minutes, 150 µL chromoge-
nic substrate was added to start the enzymatic reactions. The
rates of benzoyl-Phe-Val-Arg-p-NA hydrolysis reactions pro-
vide a linear relationship with human R-thrombin such that
free thrombin can be quantitated in reaction mixtures. Data
were analyzed directly as rates by the Softmax program to
produce [free enzyme] calculations for tight-binding Kass
determinations. For apparent Kass determinations, 5.9 nM
human thrombin or 1.4 nM bovine trypsin was used to
hydrolyze 0.2 mM BzPheValArg-pNA, 3.4 nM human plasmin
with 0.5 mM HD-Val-Leu-Lys-pNA, 1.2 nM human nt-PA with
0.81 mM HD-Ile-Pro-Arg-pNA, 0.37 nM urokinase with 0.30
mM pyro-gfsGlu-Gly-Arg-pNA, and 1.34 nM human factor Xa
with 0.18 mM BzIle-Glu-Gly-Arg-pNA.
addition of H2O (3 mL/mmol), 2 N aqueous NaOH (3 mL/
mmol), and H2O (3 mL/mmol). The two layers were separated,
and the aqueous layer was extracted with EtOAc (3 × 10 mL).
The combined organic layers were washed with brine (25 mL),
dried, and concentrated in vacuo.
The residue was taken up in CH2Cl2 (30 mL/mmol). The
mixture was cooled to 0 °C and was treated with triethylsilane
(10 equiv). After being stirred at 0 °C for 0.5 h, the reaction
mixture was treated with TFA (7 equiv) and stirred at 0 °C
until no further reaction was occurring as monitored by TLC.
The reaction mixture was poured into saturated aqueous
NaHCO3 (30 mL/mmol), and the two layers were separated.
The aqueous layer was extracted with CH2Cl2, and the
combined organic layers were dried, filtered, and concentrated
in vacuo. The products were purified by chromatography. The
physical chemical properties of compounds 5a -d and 8a -i are
reported in the Supporting Information.
4-Flu or op h en yl 2-[4-[2-(1-P yr r olid in yl)eth oxy]p h en yl]-
ben zo[b]th iop h en -3-yl Keton e (6). A 0 °C solution of benzo-
[b]thiophene 3 (3.54 g; 11.0 mmol) and 4-fluorobenzoyl chloride
(2.60 g; 16.4 mmol) in 130 mL of CH2Cl2 was protected from
light and treated with TiCl4 (8.30 g; 43.80 mmol) in a dropwise
manner. The mixture was stirred overnight while allowed to
warm to room temperature. The mixture was poured rapidly
into saturated aqueous NaHCO3 (300 mL). After being stirred
for 1 h, the mixture was diluted with H2O (200 mL) and the
mixture extracted with CH2Cl2 (2 × 150 mL). The combined
organic layers were dried (Na2SO4), filtered, and concentrated
in vacuo. Purification of the residue by MPLC (75% hexanes/
20% THF/5% Et3N) afforded 4.40 g (90%) of product 6: 1H
NMR (CDCl3) δ 7.95-7.89 (m, 1H), 7.87-7.77 (m, 3H), 7.46-
7.32 (m, 4H), 7.01-6.93 (m, 2H), 6.81 (d, J ) 8.6 Hz, 2H), 4.09
(t, J ) 5.9 Hz, 2H), 2.91 (t, J ) 5.9 Hz, 2H), 2.72-2.61 (m,
4H), 1.91-1.82 (m, 4H); IR (CHCl3) 1649, 1598, 1505, 1499,
2-[4-[2-(1-P yr r olid in yl)et h oxy]p h en yl]b en zo[b]t h io-
p h en e (3). To a solution of 2.53 g (9.4 mmol) of 1-(2-(4-
bromophenoxy)ethyl)pyrrolidine in 65 mL of benzene was
added 0.90 g (0.78 mmol) of tetrakis(triphenylphosphine)-
palladium(0). After the mixture was stirred for 5 min, a
solution of 1.40 g (7.90 mmol) of benzo[b]thiophene-2-boronic
acid (2) in 26 mL of EtOH was added followed by 8.80 mL of
2 N aqueous Na2CO3. The reaction was heated to 50 °C for 3
h, allowed to cool, and diluted with H2O and EtOAc (250 mL
each). The two layers were separated, and the aqueous layer
was extracted with EtOAc (3 × 100 mL). The combined organic
layers were dried (K2CO3), filtered, and concentrated in vacuo
to give 3.26 g of an oil. Purification by flash chromatography
(36:60:4 THF-hexanes-TEA) afforded 1.94 g (6.00 mmol;
76%) of product 3 as a white solid: 1H NMR (CDCl3) δ 7.80
(d, J ) 7.2 Hz, 1H), 7.74 (d, J ) 7.6 Hz, 1H), 7.64 (d, J ) 8.8
Hz, 2H), 7.38-7.26 (m, 3H), 6.97 (d, J ) 8.8 Hz, 2H), 4.17 (t,
J ) 5.9 Hz, 2H), 2.96 (t, J ) 5.9 Hz, 2H), 2.73-2.64 (m, 4H),
1.89-1.80 (m, 4H); IR (KBr) 2922, 1607, 1500, 1456, 1433,
1435, 1237, 1153 cm-1. FAB-HRMS: m/e, calcd for C27H25
FNO2S: 446.1590; Found: 446.1597 (M + 1). Anal. (C27H24
FNO2S) C, H, N.
-
-
1261, 1047, 822, 725 cm-1; FDMS 324 (M + 1). Anal. (C20H21
NOS‚0.1H2O) C, H, N.
-
Gen er a l P r oced u r e for th e Disp la cem en t of F lu or in e
fr om 4-F lu or op h en yl 2-[4-[2-(1-P yr r olid in yl)eth oxy]p h e-
n yl]ben zo[b]th iop h en -3-yl Keton e (6). P r ep a r a tion of
7a -l. A slurry of NaH (2-5 equiv of 60% NaH in mineral oil)
in DMF (3 mL/mmol NaH) was treated with a solution of the
appropriate alcohol, amine, or sulfide (2-5 equiv) in DMF in
a dropwise manner. After the mixture was stirred for 15 min
and gas evolution had ceased, a solution of fluoride 6 (1 equiv)
in DMF (2 mL/mmol) was added. The mixture was stirred at
room temperature for 5 h, then poured into water. The mixture
was extracted with EtOAc. The combined organic layers were
washed with brine, dried, filtered, and concentrated in vacuo.
The crude products were purified by chromatography. The
physical chemical properties of compounds 7a -l are reported
in the Supporting Information.
Gen er a l P r oced u r e for th e Acyla tion of 2-[4-[2-(1-
P yr r olid in yl)eth oxy]p h en yl]ben zo[b]th iop h en e (3) a n d
2-(4-Meth oxyp h en yl)ben zo[b]th iop h en e (9). P r ep a r a tion
of Der iva tives 1a , 4b-g, 10a , 16a , a n d 17. A mixture of the
aminocarboxylic acid (1.1 equiv) in 1,2-dichloroethane (10 mL/
mmol) was treated with a drop of DMF followed by SOCl2 (5.5
equiv). The reaction was heated to mild reflux for 2 h and
concentrated in vacuo. The residue was resuspended in 1,2-
dichloroethane and reconcentrated to afford the corresponding
acid chlorides. For the preparation of compounds 10a , 16a ,
and 17, the required acid chlorides were commercially avail-
able.
The acid chlorides were dissolved or suspended in 1,2-
dichloroethane (10 mL/mmol), and the mixture was cooled to
0 °C. Benzo[b]thiophene 3 (1 equiv) was added followed by the
portionwise addition of AlCl3 (8 equiv). The mixture was
stirred at 0 °C until TLC analysis showed no further reaction
(∼5 h) at which time it was carefully poured into a 0 °C
solution of saturated aqueous NaHCO3. The mixture was
extracted with EtOAc, and the combined organic extracts were
dried (K2CO3), filtered, and concentrated in vacuo. The crude
material was purified by chromatography to yield the desired
products. The free bases of the products were taken up in a
minimal amount of EtOAc and were treated with a solution
of 2.1 equiv of oxalic acid in EtOAc. The resulting solids were
filtered and dried. The physical chemical properties of com-
pounds 1a , 4b-g, 10a , 16a , 17 are reported in the Supporting
Information.
Meth yl 4-[1-[2-(4-Meth oxyp h en yl)ben zo[b]th iop h en -3-
yl]eth en yl]p h en yl Eth er (10b). A slurry of methyltriph-
enylphosphonium bromide (1.20 g; 3.36 mmol) in THF (50 mL)
was treated with potassium tert-butoxide (0.45 g; 4.01 mmol)
and the mixture stirred at room temperature for 0.5 h. To this
was added a solution of ketone 10a (0.80 g; 2.14 mmol) in THF
(10 mL) in a dropwise manner. The reaction was stirred at
room temperature for 18 h and then heated at gentle reflux
for 48 h. The reaction was quenched by the addition of brine
(100 mL). The two layers were separated and the organic layer
was dried (Na2SO4), filtered, and concentration in vacuo to give
1.36 g of an oil. Purification by PCTLC (10% EtOAc in hexanes)
afforded product 10b as an oil (0.61 g; 1.64 mmol; 77%): 1H
NMR (DMSO-d6) δ 7.99 (d, J ) 7.6 Hz, 1H), 7.53 (d, J ) 8.2
Hz, 2H), 7.40-7.27 (m, 5H), 6.95 (d, J ) 8.2 Hz, 2H), 6.88 (d,
J ) 8.2 Hz, 2H), 6.02 (s, 1H), 5.17 (s, 1H), 3.76 (s, 3H), 3.73 (s,
Gen er a l P r oced u r e for th e Red u ctive Deoxygen a tion
of Keton es 1a , 4b-d a n d 7a -l. P r ep a r a tion of P r od u cts
5a -d a n d 8a -i. A 0 °C slurry of LiAlH4 (3 equiv) in 10 mL of
THF was treated with a solution of the appropriate ketone (1
equiv) in THF (15 mL/mmol). The reaction was stirred at 0
°C until TLC showed complete consumption of starting mate-
rial (∼2 h). The reaction was quenched by the sequential
3H); IR (CHCl3) 1607, 1511, 1435, 1251, 1179, 1036 cm-1
FDMS 372 (M+; 100). Anal. (C24H20O2S) C, H.
;
4-[1-[2-(4-Hyd r oxyp h en yl)ben zo[b]th iop h en -3-yl]eth e-
n yl]p h en ol (10c). A mixture of dimethyl ether 10b (0.50 g;
1.34 mmol) and pyridine hydrochloride (10.0 g) was heated to
160 °C for 5 h. The reaction was allowed to cool and at 145 °C

Active Site Directed Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 659
was treated with ice (20 g). The mixture was extracted with
EtOAc (3 × 25 mL). The combined organic layers were washed
sequentially with H2O (25 mL), 1 N aqueous HCl (25 mL), and
H2O (25 mL) and were dried (Na2SO4), filtered, and concen-
trated in vacuo to a yellow solid. Purification by PCTLC
(gradient of 20-40% EtOAc in hexanes) afforded 310 mg (0.90
mmol; 67%) of diphenol 10c: 1H NMR (DMSO-d6) δ 9.70 (s,
1H), 9.53 (s, 1H), 7.92 (d, J ) 7.6 Hz, 1H), 7.38 (d, J ) 6.9 Hz,
2H), 7.32-7.23 (m, 3H), 7.14 (d, J ) 6.9 Hz, 2H), 6.71 (d, J )
6.9 Hz, 2H), 6.65 (d, J ) 6.9 Hz, 2H), 5.89 (s, 1H), 5.06 (s,
1H); IR (CHCl3) 3593, 3320, 1610, 1590, 1504, 1456, 1436,
1265, 1173, 838 cm-1; FDMS 344 (M+; 100). Anal. (C22H16O2S‚
0.5H2O) C, H.
1-[2-[4-[1-[2-[4-[2-(1-P yr r olid in yl)eth oxy]p h en yl]ben zo-
[b]th ien -3-yl]eth en yl]p h en oxy]eth yl]p yr r olid in e (10d ). A
mixture of phenol 10c (290 mg; 0.84 mmol) and Cs2CO3 (2.20
g; 6.75 mmol) in 10 mL of DMF was treated with 1-(2-
chloroethyl)pyrrolidine hydrochloride (430 mg; 2.53 mmol). The
mixture was heated to 80 °C for 16 h, cooled, filtered, and
concentrated in vacuo. The residue was partitioned between
H2O (10 mL) and EtOAc (10 mL). The two layers were
separated, and the aqueous layer was extracted with EtOAc
(2 × 10 mL). The combined organic layers were dried (K2CO3),
filtered, and concentrated in vacuo to give an oil. Purification
by PCTLC (gradient of 2 to 5 to 10% MeOH in CH2Cl2) afforded
301 mg (67%) of diamine 10d as an oil. The free base of product
10d was converted to the dioxalate salt: 1H NMR (DMSO-d6)
δ 7.95 (d, J ) 7.5 Hz, 1H), 7.49 (d, J ) 8.0 Hz, 2H), 7.38-7.16
(m, 5H), 6.95 (d, J ) 8.1 Hz, 2H), 6.87 (d, J ) 8.2 Hz, 2H),
5.99 (s, 1H), 5.14 (s, 1H), 4.32-4.10 (m, 4H), 3.54-3.36 (m,
4H), 3.29-3.12 (m, 8H), 1.98-1.72 (m, 8H); IR (KBr) 3425,
1719, 1606, 1510, 1404, 1245, 1181, 721 cm-1; FDMS 539 (M
+ 1; 100). Anal. (C34H38N2O2S‚2C2H2O4) C, H, N.
g (1.34 mmol; 61%) of the diphenol 12b as a solid: 1H NMR
(CDCl3) δ 9.92 (s, 1H), 9.47 (s, 1H), 7.86-7.78 (m, 2H), 7.61
(d, J ) 8.9 Hz, 2H), 7.34 (d, J ) 8.6 Hz, 2H), 6.96 (d, J ) 8.6
Hz, 2H), 6.88 (d, J ) 8.9 Hz, 2H), 6.64 (d, J ) 8.6 Hz, 2H); IR
(CHCl3) 3595, 3309, 1610, 1588, 1493, 1434, 1261, 1171, 827
cm-1; FDMS 350 (M+; 100). Anal. (C20H14O2S2‚0.5MeOH) C,
H.
1-[2-[4-[[2-[4-[2-(1-P yr r olid in yl)eth oxy]p h en yl]ben zo-
[b]th iop h en -3-yl]th io]p h en oxy]eth yl]p yr r olid in e (12c). A
solution of diphenol 12b (1.00 g; 2.90 mmol), tripenylphosphine
(1.90 g; 7.10 mmol), and 1-(2-hydroxyethyl)pyrrolidine (0.82
g; 7.1 mmol) in THF (30 mL) was treated with DEAD (1.50 g;
8.60 mmol) in a dropwise manner. After complete addition,
the reaction was stirred at ambient temperature for 16 h.
Additional tripenylphosphine and 1-(2-hydroxyethyl)pyrroli-
dine (2.50 equiv each) as well as DEAD (3 equiv) were added,
and stirring continued for 4 h. The mixture was concentrated
in vacuo and the residue purified by PCTLC (60:37:3 hexanes:
THF:TEA) to afford 0.72 g (1.33 mmol; 43%) of diamine 12c
as an oil. The free base of product 12c was converted to the
dioxalate salt: 1H NMR (DMSO-d6) δ 8.12-8.00 (m, 1H), 7.76-
7.65 (m, 3H), 7.47-7.38 (m, 2H), 7.13 (d, J ) 8.8 Hz, 2H), 7.00
(d, J ) 8.8 Hz, 2H), 6.87 (d, J ) 8.9 Hz, 2H), 4.36 (t, J ) 5.0
Hz, 2H), 4.19 (d, J ) 5.1 Hz, 2H), 3.56 (t, J ) 4.8 Hz, 2H),
3.47 (t, J ) 4.9 Hz, 2H), 3.42-3.18 (m, 8H), 2.03-1.82 (m, 8H);
IR (KBr) 3467, 1722, 1607, 1493, 1238, 1180, 706 cm-1; FDMS
545 (M + 1), 636 (M + 91, 100). Anal. (C32H36N2O2S2‚2C2H2O4‚
0.4H2O) C, H, N.
Ben zo[b]th iop h en -3-yl 4-Meth oxyp h en yl Eth er (14a ).
A mixture of 3-bromobenzo[b]thiophene (4.00 g; 19.7 mmol),12
4-methoxyphenol (4.96 g; 40 mmol), K2CO3 (5.52 g; 40 mmol),
and CuI (0.20 g; 1.0 mmol) was heated to 140 °C and sonicated
at this temperature for 2 h. The reaction mixture was allowed
to cool, taken up in CH2Cl2, and then washed several times
with 0.5 N NaOH. The organic layer was dried (Na2SO4),
filtered, and concentrated in vacuo to an oil which was
subjected to flash chromatography (gradient of 0-5% EtOAc
in hexanes). The fractions containing the desired product were
combined and evaporated in vacuo, and the residue was
recrystallized from hexanes to afford 500 mg (1.95 mmol; 10%)
of product 14a as a white solid: 1H NMR (CDCl3) δ 7.84-7.78
(m, 2H), 7.42-7.37 (m, 2H), 7.11 (d, J ) 8.9 Hz, 2H), 6.89 (d,
J ) 8.9 Hz, 2H), 6.45 (s, 1H), 3.82 (s, 3H); FDMS 256 (M+).
Anal. (C15H12O2S) C, H.
3-Br om o-2-(4-m eth oxyp h en yl)ben zo[b]th iop h en e (11).
A 0 °C slurry of 2-(4-methoxyphenyl)benzo[b]thiophene (9; 5.00
g; 20.8 mmol)30 in CHCl3 (400 mL) was treated slowly with
Br2 (1.60 mL; 31.0 mmol), resulting in a yellow solution. The
reaction was stirred at 0 °C for 1 h and then washed
sequentially with 1 N aqueous Na2S2O3 (200 mL), 1 N aqueous
NaHCO3 (200 mL), and H2O (200 mL). The solution was dried
(Na2SO4), filtered, and concentrated in vacuo to give analyti-
cally pure product 11 as an off-white solid (6.24 g; 19.6 mmol;
1
94%): mp 84.5-86.5 °C; H NMR (CDCl3) δ 7.86 (d, J ) 7.9
Hz, 1H), 7.80 (d, J ) 7.9 Hz, 1H), 7.72 (d, J ) 8.7 Hz, 2H),
7.50-7.37 (m, 2H), 7.02 (d, J ) 8.7 Hz, 2H), 3.88 (s, 3H); IR
(CHCl3) 1610, 1540, 1498, 1435, 1255, 1179, 1035, 889, 836
cm-1; FDMS 318 (100), 320 (M + 1). Anal. (C15H11BrOS) C, H.
2-Iod oben zo[b]th iop h en -3-yl 4-Meth oxyp h en yl Eth er
(14b). A -78 °C solution of aryl ether 14a (133 mg; 0.52 mmol)
in THF (3 mL) was treated with n-BuLi (0.33 mL; 1.6 M in
hexanes; 0.54 mmol). The mixture was stirred at -78 °C for
15 min and treated with a solution of I2 (138 mg; 0.54 mmol)
in THF (3 mL). The reaction was allowed to gradually warm
to room temperature and then partitioned between brine and
EtOAc/hexanes. The two phases were separated, and the
organic phase was washed with H2O, dried (Na2SO4), and
concentrated in vacuo. The residue was recrystallized from
hexanes to afford 143 mg (0.37 mmol; 72%) of iodide 14b as a
solid: 1H NMR (CDCl3) δ 7.73 (d, J ) 7.8 Hz, 1H), 7.46 (d, J
) 7.8 Hz, 1H), 7.34-7.19 (m, 2H), 6.91-6.78 (m, 4H), 3.78 (s,
3H); FDMS 382 (M+). Anal. (C15H11IO2S) C, H.
Meth yl4-[[2-(4-Meth oxyp h en yl)ben zo[b]th iop h en -3-yl]-
th io]p h en yl Eth er (12a ). To a -78 °C solution of bromide
11 (1.0 g; 3.1 mmol) in 20 mL of THF was added n-BuLi (2.90
mL; 1.6 M in hexanes; 4.7 mmol) in a dropwise manner. The
mixture was stirred at -78 °C for 10 min and treated with
solid bis(4-methoxyphenyl)disulfide (0.87 g; 3.13 mmol). The
reaction was stirred at -78 °C for 0.5 h and was allowed to
warm slowly to room temperature. The reaction was quenched
by the addition of saturated aqueous NH4Cl (1 mL) and MeOH
(1 mL) and was concentrated in vacuo. The residue was
partitioned between EtOAc (100 mL) and H2O (100 mL). The
organic layer was separated, dried (MgSO4), filtered, and
concentrated in vacuo to afford an oily solid. Partial purifica-
tion by flash chromatography (gradient of 1-5% EtOAc in
hexanes) afforded 0.82 g of product 12a as an oil: 1H NMR
(CDCl3) δ 7.94-7.85 (m, 2H), 7.72 (d, J ) 8.7 Hz, 2H), 7.43-
7.38 (m, 2H), 7.09-6.99 (m, 4H), 6.78 (d, J ) 8.7 Hz, 2H), 3.90
(s, 3H), 3.77 (s, 3H); IR (CHCl3) 1609, 1528, 1494, 1288, 1248,
1177, 1034 cm-1; FDMS 378 (M + 1; 100). Product 12a was
taken on without further purification.
2-(4-Meth oxyp h en yl)ben zo[b]th iop h en -3-yl 4-Meth ox-
yp h en yl Eth er (15a ). Employing the methods used in the
preparation of 3, product 15a was prepared from iodide 14b
and (4-methoxyphenyl)boronic acid in 70% yield following flash
chromatography (5% EtOAc in hexanes): 1H NMR (CDCl3) δ
7.80-7.72 (m, 3H), 7.42-7.36 (m, 1H), 7.32-7.21 (m, 2H),
6.95-6.86 (m, 4H), 6.81-6.75 (m, 2H), 3.82 (s, 3H), 3.74 (s,
3H); FDMS 363 (M + 1). Anal. (C22H18O3S) C, H.
4-[[2-(4-Hydr oxyph en yl)ben zo[b]th ioph en -3-yl]oxy]ph e-
n ol (15b). Employing the conditions described in the prepara-
tion of 10c, diphenol 15b was prepared from dimethyl ether
15a in 87% yield following PCTLC (25% EtOAc in hexanes):
1H NMR (CDCl3) δ 9.77 (s, 1H), 9.10 (s, 1H), 7.90 (d, J ) 8.0
Hz, 1H), 7.54 (d, J ) 8.6 Hz, 2H), 7.37-7.25 (m, 3H), 6.82-
6.73 (m, 4H), 6.63 (d, J ) 8.6 Hz, 2H); IR (KBr) 3394, 1611,
4-[[2-(4-H yd r oxyp h en yl)b en zo[b]t h iop h en -3-yl]t h io]-
p h en ol (12b). A 0 °C solution of dimethyl ether 12a (0.82 g;
2.2 mmol) in dichloroethane (50 mL) was treated with BBr3
(1.20 mL; 13 mmol). The reaction was stirred at 0 °C for 5 h
and quenched by the careful addition of MeOH (15 mL).
Concentration in vacuo gave a residue which was subjected
to flash chromatography (1% MeOH in CHCl3) to afford 0.47

660 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Sall et al.
1504, 1438, 1356, 1202, 1183, 1054, 825, 752 cm-1; FDMS 334
(M+, 100). Anal. (C20H14O3S) C, H.
2H), 7.18 (d, J ) 8.6 Hz, 2H), 6.74 (d, J ) 8.7 Hz, 1H), 6.66 (d,
J ) 8.6 Hz, 2H), 4.31 (t, J ) 5.7 Hz, 2H), 2.78-2.67 (m, 2H),
2.48-2.41 (m, 4H), 1.67-1.58 (m, 4H); IR (CHCl3) 2975, 1642,
1599, 1489, 1458, 1358, 1282, 1237 cm-1. HRMS calcd for
C26H24N2O3S: 445.1586. Found: 445.1569. Anal. (C26H24N2O3S‚
0.2H2O) C, H, N.
1-[2-[4-[[2-[4-[2-(1-P yr r olid in yl)eth oxy]p h en yl]ben zo-
[b]th ien -3-yl]m eth yl]ph en oxy]eth yl]pyr r olidin e (18c). Em-
ploying the conditions described in the preparation of 10d ,
phenol 18b was alkylated to provide diamine 18c in 84% yield
as an oil following PCTLC (gradient of 5-20% MeOH in THF).
The free base of product 18c was converted to the dioxalate
salt. 1H NMR (DMSO-d6) δ 8.33 (d, J ) 2.4 Hz, 1H), 8.09-
8.04 (m, 1H), 7.94 (dd, J ) 8.7 and 2.2 Hz, 1H), 7.66-7.62 (m,
1H), 7.47-7.38 (m, 2H), 7.29 (d, J ) 8.5 Hz, 2H), 6.84 (d, J )
6.1 Hz, 2H), 6.74 (d, J ) 8.7 Hz, 1H), 4.29 (t, J ) 4.8 Hz, 2),
4.02-3.95 (m, 2H), 2.74-2.66 (m, 4H), 2.53-2.38 (m, 8H),
1.69-1.58 (m, 8H); IR (KBr) 3434, 2966, 1729, 1644, 1599,
1458, 1239, 1051, 837, 698 cm-1; FDMS 542 (M + 1). Anal.
(C32H35N3O3S‚2C2H2O4‚1.5H2O) C, H, N.
4-[2-[4-[2-(1-P yr r olid in yl)e t h oxy]p h e n yl]b e n zo[b]-
th ioph en yl-3-yl]m eth ylph en yl 2-(1-P yr r olidin yl)eth yl Ke-
ton e (20). Acetylene 1914 (185 mg, 0.355 mmol) was dissolved
in TFA (1 mL) and the solution heated to reflux in a sealed
tube. After 2.5 h, the reaction mixture was cooled to 0 °C and
poured into saturated aqueous NaHCO3 (20 mL). The two
layers were separated, and the aqueous layer was extracted
with EtOAc (3 × 20 mL). The combined organic layers were
washed with brine, dried (Na2SO4), filtered, and concentrated
in vacuo. Purification by PCTLC (gradient of 85:10:5 to 65:
30:5 hexanes:THF:TEA) afforded ketone 20 as a yellow oil (30
mg; 0.056 mmol, 16%): 1H NMR (CDCl3) δ 7.85-7.88 (m, 2H),
7.45 (d, J ) 8.0 Hz, 1H), 7.38 (, J ) 8.6 Hz, 2H), 7.29-7.22
(m, 5H), 6.95 (d, J ) 8.6 Hz, 2H), 4.31 (s, 2H), 4.13 (t, J ) 5.9
Hz, 2H), 3.19 (t, J ) 6 Hz, 2H), 2.95-2.90 (m, 4H), 2.71-2.58
(m, 8H), 1.88-1.76 (m, 8H). The free base of product 20 was
converted to the dioxalate salt. IR (KBr) 1721, 1682, 1606,
1503, 1413, 1243, 1180, 1083, 720 cm-1; FDMS 539 (M + 1).
Anal. (C34H38N2O2S2‚C2H2O4‚0.5H2O) C, H, N.
6-Meth oxyben zo[b]th iop h en e-2-bor on ic Acid (21b). To
a -60 °C solution of 6-methoxybenzo[b]thiophene (21a ; 18.13
g, 0.111 mol)15 in THF (150 mL) was added n-BuLi (76.2 mL,
0.122 mol, 1.6 M solution in hexanes) in a dropwise manner.
After the mixture was stirred for 30 min, triisopropyl borate
(28.2 mL, 0.122 mol) was added dropwise. The resulting
mixture was allowed to gradually warm to 0 °C and then
partitioned between 1.0 N HCl and EtOAc (300 mL each). The
layers were separated, and the organic phase was dried (Na2-
SO4), filtered, and concentrated in vacuo. The white solid was
triturated from Et2O/hexanes and filtered to provide 16.40 g
(71%) of boronic acid 21b as a white solid which was used
without further purification: mp 200 °C (dec); 1H NMR
(DMSO-d6) δ 8.36 (br s, 2H), 7.86-7.75 (m, 2H), 7.53 (dd, J )
8.1 and 2.0 Hz, 1H), 6.98 (m, 1H), 3.82 (s, 3H); FDMS 208
(M+; 100).
1-[2-[4-[3-[4-[2-(1-P yr r olid in yl)eth oxy]p h en oxy]ben zo-
[b]th iop h en -2-yl]p h en oxy]eth yl]p yr r olid in e (15c). Em-
ploying the conditions described in the preparation of 10d ,
diphenol 15b was converted to diamine 15c in 52% yield
following PCTLC (gradient of 2-10% MeOH in CH2Cl2). The
free base of product 15c was converted to the dioxalate salt:
1H NMR (DMSO-d6) δ 7.96 (d, J ) 8.0 Hz, 1H), 7.68 (d, J )
8.5 Hz, 2H), 7.40-7.23 (m, 3H), 7.04 (d, J ) 8.5 Hz, 2H), 6.91-
6.80 (m, 4H), 4.38-4.13 (m, 4H), 3.55-3.41 (m, 4H), 3.36-
3.14 (m, 8H), 1.97-1.78 (m, 8H); IR (KBr) 1730, 1502, 1247,
1199, 833 cm-1; FDMS 529 (M + 1; 100). Anal. (C32H36N2O3S‚
2C2H2O4) C, H, N.
2-(4-Hyd r oxyp h en yl)ben zo[b]th iop h en -3-yl 2-Hyd r ox-
yp h en yl Keton e (16b). A 0 °C solution of dimethyl ether 16a
(1.11 g; 2.97 mmol) in CH2Cl2 (100 mL) was treated with AlCl3
(3.16 g; 23.7 mmol) followed by EtSH (2.63 mL; 35.6 mmol).
The cold bath was removed and the reaction allowed to stir at
ambient temperature for 2 h. The reaction was poured into a
separatory funnel containing saturated aqueous NaHCO3 (350
mL) and EtOAc (350 mL). The two layers were separated, and
the aqueous layer was extracted with EtOAc (250 mL). The
combined organic layers were dried (Na2SO4), filtered, and
concentrated in vacuo to give 1.00 g of diphenol 16b as a yellow
1
foam. H NMR (DMSO-d6) δ 9.81 (s, 1H), 8.03 (d, J ) 6.7 Hz,
1H), 7.53 (d, J ) 8.4 Hz, 1H), 7.44-7.32 (m, 4H), 7.26-7.18
(m, 3H), 6.89 (d, J ) 8.4 Hz, 1H), 6.72-6.63 (m, 3H); IR (KBr)
3594, 2977, 1625, 1611, 1501, 1484, 1437, 1267, 1237, 1175,
1155, 835, 808 cm-1; FDMS 346 (M+). Anal. (C21H14O3S) C, H.
2-[4-[2-(1-P yr r olid in yl)et h oxy]p h en yl]b en zo[b]t h io-
p h en -3-yl 2-[2-(1-P yr r olid in yl)-et h oxy]p h en yl Ket on e
(16c). Employing the conditions described in the preparation
of 10d , diphenol 16b was alkylated to afford diamine 16c in
97% yield following PCTLC (5% MeOH in CH2Cl2). The free
base of product 16c was converted to the dioxalate salt: 1H
NMR (DMSO-d6) δ 8.08-8.03 (m, 1H), 7.95-7.90 (m, 1H),
7.49-7.42 (m, 2H), 7.36-7.24 (m, 4H), 6.90-6.78 (m, 4H), 4.21
(t, J ) 4.9 Hz, 2H), 4.07 (t, J ) 4.9 Hz, 2H), 3.47 (t, J ) 4.9
Hz, 2H), 3.15 (t, J ) 4.9 Hz, 2H), 3.08 (t, J ) 6.3 Hz, 2H),
2.94-2.82 (m, 4H), 1.97-1.82 (m, 6H), 1.67-1.58 (m, 4H); IR
(KBr) 3054, 1727, 1660, 1609, 1457, 1349, 1289, 1249, 756,
701 cm-1; FDMS 541 (M + 1). Anal. (C33H36N2O3S‚1.4C2H2O4‚
2H2O) C, H, N.
2-(4-Meth oxyph en yl)ben zo[b]th iop h en -3-yl 6-[2-(1-P yr -
r olid in yl)eth oxy]p yr id -3-yl Keton e (18a ). A solution of
1-(2-hydroxyethyl)pyrrolidine (0.50 mL; 4.30 mmol) in xylenes
(10 mL) was treated with Na (50 mg; 2.20 mmol). The mixture
was heated to 50 °C until all the Na had disappeared. The
mixture was cooled to room temperature and treated with a
solution of chloride 17 (420 mg; 1.10 mmol) in xylenes (5 mL).
The reaction was heated to 50 °C for 2 h and was concentrated
in vacuo. The residue was partitioned between H2O (50 mL)
and EtOAc (50 mL). The organic layer was separated and the
aqueous layer extracted with EtOAc (2 × 50 mL). The
combined organic layers were dried (K2CO3), filtered, and
concentrated in vacuo to give 810 mg of a yellow solid.
Purification by PCTLC (gradient of 1-5% MeOH in CH2Cl2)
gave 525 mg (1.09 mmol; 99%) of product 18a as an amber
oil: 1H NMR (DMSO-d6) δ 8.33 (s, 1H), 8.06 (d, J ) 7.5 Hz,
1H), 7.97-7.91 (m, 1H), 7.67-7.62 (m, 1H), 7.47-7.39 (m, 2H),
7.32 (d, J ) 8.6 Hz, 2H), 6.86 (d, J ) 7.8 Hz, 2H), 6.75 (d, J )
8.8 Hz, 1H), 4.31 (t, J ) 5.8 Hz, 2H), 3.68 (s, 3H), 2.67 (t, J )
5.8 Hz, 2H), 2.49-2.38 (m, 4H), 1.68-1.57 (m, 4H); IR (CHCl3)
2970, 1647, 1599, 1485, 1458, 1435, 1281, 1254, 1238, 1178,
1034 cm-1; FDMS 459 (M+; 100). Anal. (C27H26N2O3S) C, H,
N.
6-Meth oxy-2-[4-[2-(1-p yr r olid in yl)eth oxy]p h en yl]ben -
zo[b]th iop h en e (22). Employing the conditions described in
the preparation of benzo[b]thiophene 3, the coupled product
22 was prepared from boronic acid 21b and 1-(2-(4-bromophe-
noxy)ethyl)pyrrolidine in 55% yield as an off-white solid
following MPLC (53:35:2 THF:hexanes:TEA): mp 151-154 °C;
1H NMR (CDCl3) δ 7.61 (d, J ) 8.8 Hz, 1H), 7.58 (d, J ) 8.8
Hz, 2H), 7.33 (s, 1H), 7.29 (d, J ) 2.3 Hz, 1H), 6.95 (d, J ) 8.7
Hz, 3H), 4.18 (t, J ) 5.9 Hz, 2H), 3.88 (s, 3H), 2.97 (t, J ) 5.9
Hz, 2H), 2.71 (br t, 4H), 1.85 (m, 4H); IR (CHCl3) 2941, 1610,
1501, 1479, 1250, 1176, 1059, 830 cm-1; FDMS 353 (M+). Anal.
(C21H23NO2S) C, H, N.
2-(4-Hyd r oxyph en yl)ben zo[b]th iop h en -3-yl 6-[2-(1-P yr -
r olid in yl)eth oxy)p yr id -3-yl Keton e (18b). Employing the
conditions described in the preparation of 16b, phenol 18b was
prepared from methyl ether 18a in 89% yield as a yellow solid
following PCTLC (5% MeOH in CH2Cl2): 1H NMR (DMSO-
d6) δ 9.81 (br s, 1H), 8.32 (s, 1H), 8.08-8.02 (m, 1H), 7.93 (dd,
J ) 2.2 and 8.8 Hz, 1H), 7.66-7.61 (m, 1H), 7.43-7.38 (m,
3-Br om o-4-[(1-p yr r olid in yl)m eth yl]p h en yl 6-Meth oxy-
2-[4-[2-(1-p yr r olid in yl)-eth oxy]ph en yl]ben zo[b]th iop h en -
3-yl Keton e (23a ). Employing the conditions described in the
preparation of 1a and 4b-g, product 23a was prepared from
3-bromo-4-[(1-pyrrolidinyl)methyl]benzoate31 and benzo[b]-
thiophene 22 in 55% yield following flash chromatography
(95:5 EtOAc:Et3N): 1H NMR (CDCl3) δ 7.90 (s, 1H), 7.70 (d, J

Active Site Directed Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 661
) 9.0 Hz, 1H), 7.61 (d, J ) 8.0 Hz, 1H), 7.35 (d, J ) 8.0 Hz,
1H), 7.33 (s, 1H), 7.26 (d, J ) 8.3, 2H), 7.02 (d, J ) 9.0 Hz,
1H), 6.75 (d, J ) 8.7 Hz, 2H), 4.03 (t, J ) 6.0 Hz, 2H), 3.90 (s,
3H), 3.70 (s, 2H), 2.85 (t, J ) 6.0 Hz, 2H), 2.60 (t, J ) 6.2 Hz,
4H), 2.51 (t, J ) 6.2 Hz, 4H), 1.80 (m, 8H); FDMS 620 (M +
1). Anal. (C33H35BrN2O3S) C, H, N.
NaOH (5.0 M), and extracted with CH2Cl2 (3 × 100 mL). The
combined organic layers were dried (Na2SO4), filtered, and
concentrated in vacuo. Purification by flash chromatography
(5% Et3N in EtOAc) afforded product 27 as a brown oil (3.70
g, 7.39 mmol; 84%): 1H NMR (CDCl3) δ 7.5-6.9 (m, 11H), 5.10
(s, 2H), 3.90 (s, 3H), 3.71 (s, 2H), 2.91 (s, 6H), 2.60 (m, 4H),
1.83 (m, 4H); IR (CHCl3) 1600, 1542, 1477, 1465, 1409, 1264
cm-1; FDMS 500 (M+). Product 27 was converted to the
monooxalate salt. Anal. (C30H32N2O3S‚1C2H2O4‚0.1H2O) C, H,
N.
3-[3-Br om o-4-[(1-p yr r olid in yl)m eth yl]ben zyl]-6-m eth -
oxy-2-[4-[2-(1-p yr r olid in yl)e t h oxy]p h e n yl]b e n zo[b]-
th iop h en e (23b). Employing the conditions described in the
preparation of 5a -d , ketone 23a was converted to product 23b
in 72% yield flash chromatography (gradient of 0-5% Et3N
6-Ben zyloxy-2-[4-[2-(1-pyr r olidin yl)eth oxy]ph en yl]ben -
zo[b]th iop h en -3-yl 3-Meth oxy-4-[(1-p yr r olid in yl)m eth yl]-
p h en yl Keton e (28a ). Magnesium turnings (0.30 g; 12.3
mmol) were placed in a two-neck 100 mL round-bottom flask
fitted with a reflux condenser and a magnetic stir bar. The
whole apparatus was flame-dried and allowed to cool to
ambient temperature. Dry THF (22 mL) and a small crystal
of iodine were then introduced, followed by slow addition of
1-[2-(4-bromophenoxy)ethyl]pyrrolidine (2.75 mL; 3.59 g; 13.3
mmol) while stirring at ambient temperature. The reaction
mixture was allowed to warm to a gentle reflux for 2 h or until
the magnesium turnings were completely consumed to give a
0.5 M solution of the Grignard reagent. This freshly prepared
Grignard solution (7 mL) was added slowly to a stirring
solution of intermediate 27 (1.10 g, 2.20 mmol) in THF (5.0
mL) at 0 °C under argon. The resulting mixture was stirred
at 0 °C for 2 h before quenching with saturated aqueous NH4-
Cl (50 mL) and extraction with CH2Cl2 (3 × 50 mL). The
combined organic layers were dried (Na2SO4), filtered, and
concentrated in vacuo. Purification by flash chromatography
(5:5:90 Et3N:MeOH:EtOAc) afforded product 28a as a colorless
oil (1.33 g, 2.06 mmol): 1H NMR (CDCl3) δ 7.60 (d, 1H), 7.47-
7.00 (m, 12H), 6.75 (d, 2H), 5.16 (s, 2H), 4.03 (t, 2H), 3.79 (s,
3H), 3.62 (s, 2H), 2.86 (t, 2H), 2.59 (m, 4H), 2.49 (m, 4H), 1.80
(m, 8H). Product 28a was taken on without further purifica-
tion.
6-Hyd r oxy-2-[4-[2-(1-p yr r olid in yl)eth oxy]p h en yl]ben -
zo[b]th iop h en -3-yl 3-Meth oxy-4-[(1-p yr r olid in yl)m eth yl]-
p h en yl Keton e (28b). Intermediate 28a (105 mg, 0.16 mmol)
in THF (5.0 mL) was treated sequentially with a solution of
ammonium formate (25% in H2O, 3 mL) and 10% Pd-C (50
mg). The resulting mixture was stirred at ambient tempera-
ture for 9 h and filtered, and the mixture was extracted with
CH2Cl2 (3 × 30 mL). The combined organic layers were washed
with water (50 mL), dried (Na2SO4), filtered, and concentrated
in vacuo. Purification by flash chromatography (5:10:85 Et3N:
MeOH:EtOAc) afforded product 28b as a yellow solid (80 mg,
0.15 mmol; 88%): 1H NMR (CDCl3) δ 7.56 (d, 2H), 7.32 (s, 1H),
7.24 (d, 1H), 7.20 (d, 1H), 7.15 (d, 2H), 6.98 (s, 1H), 6.79 (d,-
1H), 6.60 (d, 2H), 4.08 (t, 2H), 3.80 (s, 2H), 3.75 (s, 3H), 3.01
(t, 2H), 2.99 (m, 4H), 2.82 (m, 4H), 1.87 (m, 8H); FDMS 557
(M + 1). Anal. (C33H36N2O4S‚1.7C2H2O4‚3H2O) C, H, N.
6-Hydr oxy-3-[3-m eth oxy-4-[(1-pyr r olidin yl)m eth yl]ben -
zyl]-2-[4-[2-(1-p yr r olid in yl)e t h oxy]p h e n yl]b e n zo[b]-
th iop h en e (28c). Employing the conditions described in the
preparation of 5a -d , ketone 28b was converted to product 28c
in 76% yield following flash chromatography (5:10:85 Et3N:
MeOH:EtOAc): 1H NMR (CDCl3) δ 7.32 (d, 2H), 7.24 (d, 1H),
7.23 (d, 1H), 7.21 (s, 1H), 6.84 (d, 2H), 6.68 (d, 1H), 6.62 (s,
1H), 6.59 (d, 1H), 4.16 (s, 2H), 4.14 (t, 2H), 3.80 (s, 2H), 3.58
(s, 3H), 2.98 (t, 2H), 2.78 (m, 4H), 2.74 (m, 4H), 1.86 (m, 8H).
The free base of product 28c was converted to the dioxalate
salt: IR (KBr) 1719, 1612, 1548, 1504, 1462, 1420, 1241, 1215,
1179, 1032, 721 cm-1; FDMS 543 (M + 1). Anal. (C33H38N2O3S‚
2C2H2O4‚0.5H2O) C, H, N.
5-Br om op yr id -2-yl 2-(1-p yr r olid in yl)eth yl Eth er (30).
A solution of 2,5-dibromopyridine (29; 20.0 g; 84.4 mmol) in
DMF (400 mL) was treated with NaH (5.00 g; 60% in mineral
oil; 126.6 mmol). The mixture was cooled to 0 °C and treated
slowly with 1-(2-hydroxyethyl)pyrrolidine (10.0 mL; 85.5 mmol).
The reaction was stirred for 16 h while warming to room
temperature and was poured into brine (300 mL). The mixture
was diluted with H2O (150 mL) and EtOAc (200 mL). The two
layers were separated, and the aqueous phase was extracted
1
in EtOAc). H NMR (CDCl3) δ 7.34 (m, 6H), 7.03 (d, J ) 9.0
Hz, 1H), 6.95 (m, 3H), 4.16 (t, J ) 6.0 Hz, 2H), 3.90 (s, 3H),
3.70 (s, 2H), 2.95 (t, J ) 6.0 Hz, 2H), 2.65 (br s, 4H), 2.60 (br
s, 4H), 1.82 (m, 10H); IR (KBr) 3423, 1718, 1607, 1503, 1477,
1403, 1243, 1223, 1179, 1044, 1019, 834, 721 cm-1; FDMS 605
(M+). Product 23b was taken on directly to the next step.
3-[3-Br om o-4-[(1-p yr r olid in yl)m e t h yl]b e n zyl]-6-h y-
d r oxy-2-[4-[2-(1-p yr r olid in yl)et h oxy]p h en yl]b en zo[b]-
th iop h en e (23c). Employing the conditions described in the
synthesis of compound 16b, methyl ether 23b was deprotected
to afford phenol 23c as a white crystalline solid in 74%
following flash chromatography (gradient of 100% EtOAc to
85% EtOAc:10% MeOH:5% Et3N): 1H NMR (CDCl3) δ 7.34 (m,
5H), 7.14 (s, 1H), 7.06 (d, J ) 8.0 Hz, 1H), 7.00 (d, J ) 9.0 Hz,
2H), 6.80 (d, J ) 9.0 Hz, 1H), 4.12 (br t, J ) 6.0 Hz, 2H), 3.78
(s, 2H), 3.02 (s, 2H), 2.95 (t, J ) 6.0 Hz, 2H), 2.79 (br s, 4H),
2.63 (br s, 4H), 1.82 (m, 4H), 1.79 (m, 4H), no OH signal
observed; FDMS 591 (M+). The free base of product 23c was
converted to the dioxalate salt: mp 190-192 °C; IR (KBr)
1772, 1724, 1607, 1503, 1471, 1403, 1241, 1179 cm-1. Anal.
(C32H35BrN2O2S‚C2H2O4‚H2O) C, H, N.
N,N-Dim et h yl-2-h yd r oxy-2-(4-b en zyloxy)p h en ylt h io-
a ceta m id e (25). A -78 °C solution of diisopropylamine (22.9
mL; 174 mmol) in THF (400 mL) was treated with n-BuLi (100
mL; 1.6 M in hexanes; 160 mmol). After complete addition,
the mixture was stirred at -78 °C for 30 min then at 0 °C for
30 min. The mixture was cooled to -78 °C and treated with a
solution of 4-benzyloxybenzaldehyde (30.9 g; 146 mmol) and
N,N-dimethylthioformamide (24; 13.7 mL; 160.0 mmol) in THF
(50 mL) over 1 h. The reaction mixture was allowed to stir at
-78 °C for 16 h, poured into saturated aqueous NH4Cl (500
mL), and extracted with EtOAc (4 × 400 mL). The combined
organic layers were dried (MgSO4), filtered, and concentrated
in vacuo to give 23.0 g of a black oil which solidified on
standing. Purification by MPLC (25% EtOAc in hexanes)
afforded a solid which was recrystallized from hexanes/EtOAc
to afford the product 25 as a tan solid (14.90 g; 49.4 mmol;
34%): 1H NMR (CDCl3) δ 7.57-7.41 (m, 7H), 6.96 (d, J ) 8.6
Hz, 2H), 5.27 (s, 1H), 5.06 (s, 2H), 3.53 (s, 3H), 3.13 (s, 3H),
no OH signal observed; IR (CHCl3) 1609, 1508, 1386, 1245,
1230, 1176 cm-1; FDMS 301 (M+). Anal. (C17H19NO2S) C, H,
N.
2-Dim eth yla m in o-6-ben zyloxyben zo[b]th iop h en e (26).
A solution of alcohol 25 (7.00 g; 23.2 mmol) in CH2Cl2 (925
mL) was treated with MeSO3H (7.50 mL; 116.1 mmol) over 5
min. The reaction was stirred at ambient temperature for 2 h
and poured into saturated aqueous NaHCO3 (200 mL). The
two layers were separated, and the organic layer was dried
(MgSO4), filtered, and concentrated in vacuo to give 7.25 g of
a brown solid. Purification by MPLC (5% EtOAc in hexanes)
afforded product 26 as a solid (4.15 g; 14.7 mmol; 63%): 1H
NMR (CDCl3) δ 7.60-7.44 (m, 6H), 7.36 (d, J ) 4.5 Hz, 1H),
6.95 (dd, J ) 8.6 and 2.2 Hz, 1H), 5.96 (br s, 1H), 5.09 (s, 2H),
2.98 (s, 6H); IR (CHCl3) 3008, 2877, 1600, 1566, 1550, 1476,
1436, 1262, 1052 cm-1; FDMS 283 (M+). Anal. (C17H17NOS)
C, H, N.
6-Ben zyloxy-2-(d im et h yla m in o)b en zo[b]t h iop h en -3-
yl 3-Met h oxy-4-[(1-p yr r olid in yl)m et h yl]p h en yl Ket on e
(27). A solution of dimethylamine derivative 26 (2.50 g, 8.80
mmol) and 3-methoxy-4-[(1-pyrrolidinyl)methyl]benzoyl chlo-
ride31 (3.0 g, 1.3 equiv) in chlorobenzene (30 mL) was heated
at 135 °C under nitrogen for 2 h. The cooled reaction mixture
was diluted with brine (100 mL), neutralized with aqueous

662 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Sall et al.
with EtOAc (2 × 200 mL). The combined organic layers were
dried (Na2SO4), filtered, and concentrated in vacuo to give
25.80 g of an oil. Purification by MPLC (5:15:80 TEA:THF:
hexanes) afforded product 30 as a clear oil which solidified on
standing (21.50 g; 79.30 mmol; 94%): 1H NMR (CDCl3) δ 8.17
(d, J ) 2.4 Hz, 1H), 7.62 (dd, J ) 8.8 and 2.6 Hz, 1H), 6.70 (d,
J ) 8.8 Hz, 1H), 4.40 (t, J ) 5.8 Hz, 2H), 2.86 (t, J ) 5.9 Hz,
2H), 2.75-2.66 (m, 4H), 1.97-1.85 (m, 4H); IR (CHCl3) 2967,
1586, 1472, 1460, 1371, 1351, 1302, 1282, 828 cm-1; FDMS
m/e: 270, 272 (M+). Anal. (C11H15BrN2O) C, H, N.
Refer en ces
(1) (a) Rosenberg, R. D. Chemistry of the Hemostatic Mechanism
and Its Relation to the Action of Heparin. Fed. Proc., Fed. Am.
Soc. Exp. Biol. 1977, 36, 10-18. (b) Smith G. F.; Sundboom J .
L. Heparin and Protease Inhibition. II. The Role of Heparin in
the ATIII Inactivation of Thrombin, Plasmin, and Trypsin.
Thromb. Res. 1981, 22, 115-133.
(2) (a) O’Reilly, R. A. Vitamin K Antagonists. In Hemostasis and
Thrombosis: Basic Principles and Clinical Practice, 2nd ed.;
Colman, R. W., Hirsh, J ., Marder, V. J ., Salzman, E. W., Eds.;
J . B. Lippincott: Philadelphia, 1987; pp 1367-1372. (b) Hirsch
J . Oral Anticoagulants. N. Engl. J . Med. 1991, 324, 1865-1875.
(3) For a general discussion of the role of thrombin in thrombogen-
esis, see: (a) Machovich, R. Thrombin and Hemostasis. In The
Thrombin, Vol. 1; Machovich, R., Ed.; CRC Press: Boca Raton,
FL, 1984; pp 1-22. (b) Hirsh, J .; Salzman, E. W.; Marder, V. J .;
Colman, R. W. Overview of the Thrombotic Process and Its
Therapy. In Hemostasis and Thrombosis: Basic Principles and
Clinical Practice, 2nd ed.; Colman, R. W., Hirsh, J ., Marder, V.
J ., Salzman, E. W., Eds.; J . B. Lippincott: Philadelphia, 1987;
pp 1063-1072.
(4) (a) Hirsh, J .; Salzman, E. W. Pathogenesis of Venous Throm-
boembolism. In Hemostasis and Thrombosis: Basic Principles
and Clinical Practice, 2nd ed.; Colman, R. W., Hirsh, J ., Marder,
V. J ., Salzman, E. W., Eds.; J . B. Lippincott: Philadelphia, 1987;
pp 1199-1207. (b) Barnett, H. J . M. Thrombotic Processes in
Cerebrovascular Disease. In Hemostasis and Thrombosis: Basic
Principles and Clinical Practice, 2nd ed.; Colman, R. W., Hirsh,
J ., Marder, V. J ., Salzman, E. W., Eds.; J . B. Lippincott:
Philadelphia, 1987; pp 1301-1315.
(5) Stein, B.; Fuster, V.; Halperin, J . L.; Chesebro, J . H. Antithrom-
botic Therapy in Cardiac Disease. An Emerging Approach Based
on Pathogenesis and Risk. Circulation 1989, 80, 1501-1513.
(6) For recent reviews on approaches to the design of active site
directed thrombin inhibitors, see: (a) Kimball, S. D. Oral
Thrombin Inhibitors: Challenges and Progress.Handb. Exp.
Pharmacol. 1999, 132 (Antithrombotics), 367-396. (b) Menear,
K. Progress Towards the Discovery of Orally Active Thrombin
Inhibitors. Curr. Med. Chem. 1998, 5, 457-468. (c) Want, R.;
Liu, L.; Lai, L.; Tang, Y. Structure-affinity Relationship of
Thrombin Inhibitors. Wuli Huaxue Xuebao 1998, 14, 887-892.
(d) Sanderson, P. E. J ., Naylor-Olsen, A. M. Thrombin Inhibitor
Design. Curr. Med. Chem. 1998, 5, 289-304. (e) Kaiser, B.
Thrombin and Factor Xa Inhibitors. Drugs Future 1998, 23,
423-436. (f) Uzan, A. Antithrombotic Agents. Emerging Drugs
1998, 3, 189-208. (g) Wiley, M. R.; Fisher, M. J . Small Molecule
Thrombin Inhibitors. Exp. Opin. Ther. Patents 1997, 7, 1265-
1282. (h) Edmunds, J . J .; Rapundalo, S. T. Thrombin and Factor
Xa Inhibition. Ann. Rep. Med. Chem. 1996, 31, 51-60.
(7) (a) Sall, D. J .; Bastian, J . A.; Briggs, S. L.; Buben, J . A.;
Chirgadze, N. Y.; Clawson, D. K.; Denney, M. L.; Giera, D. D.;
Gifford-Moore, D. S.; Harper, R. W.; Hauser, K. L.; Klimkowski,
V. J .; Kohn, T. J .; Lin, H.-S.; McCowan, J . R.; Palkowitz, A. D.;
Smith, G. F.; Takeuchi, K.; Thrasher, K. J .; Tinsley, J . M.;
Utterback, B. G.; Yan, S.-CB.; Zhang, M. Dibasic Benzo[b]-
thiophene Derivatives as a Novel Class of Active Site Directed
Thrombin Inhibitors. 1. Determination of the Serine Protease
Selectivity, Structure-Activity Relationships and Binding Ori-
entation. J . Med. Chem. 1997, 40, 3489-3493. (b) Takeuchi K.;
Kohn T. J .; Sall D. J .; Denney M. L.; McCowan J . R.; Smith G.
F.; Gifford-Moore D. S. Dibasic Benzo[b]thiophene Derivatives
as a Novel Class of Active Site Directed Thrombin Inhibitors. 4.
SAR Studies on the Conformationally Restricted C-3-Side Chain
of Hydroxybenzo[b]thiophenes. Bioorg Med. Chem. Lett. 1999,
9, 759-764.
6-H yd r oxy-2-[6-[2-(1-p yr r olid in yl)et h oxy]p yr id -3-yl]-
b en zo[b]t h iop h en -3-yl 3-Met h oxy-4-[(1-p yr r olid in yl)-
m eth yl]p h en yl Keton e Dioxa la te (31b). A -78 °C solution
of pyridyl bromide 30 (3.25 g, 12.0 mmol) in THF (40 mL) was
treated with n-BuLi (1.6 M in hexanes; 8.10 mL, 13.00 mmol).
After 1 h, a slurry of MgBr2 [freshly prepared from Mg (365
mg, 15.0 mmol) and 1.3 mL of 1,2-dibromoethane] in THF (20
mL) was added. The reaction mixture was stirred at -78 °C
for an additional 10 min, and then the cold bath was removed.
After 45 min, the Grignard reagent was added dropwise via a
cannula to a 0 °C solution of benzo[b]thiophene 27 (5.00 g,
10.00 mmol) in THF (50 mL). The resulting mixture was
stirred for 2 h at 0 °C and then was allowed to warm to
ambient temperature. After 5 h, the reaction mixture was
poured into saturated aqueous NH4Cl (200 mL). The layers
were separated, and the aqueous phase was extracted with
CHCl3 (3 × 50 mL). The combined organic phases were dried
(Na2SO4), filtered, and concentrated in vacuo. Partial purifica-
tion was achieved by MPLC (gradient of 85:10:5 hexanes:THF:
TEA to 75:20:5 hexanes:THF:TEA) to afford 2.90 g (4.48 mmol;
45%) of semi-pure product 31a as a viscous orange oil: IR
(CHCl3) 2966, 1603, 1463, 1285, 1250 cm-1. FAB-HRMS: m/e,
calcd for C39H42N3O4S: 648.2896. Found: 648.2889 (M + 1).
Product 31a was taken on without further purification.
Intermediate ketone 31a (2.90 g; 4.48 mmol) was depro-
tected using the conditions described in the preparation of
compound 28b to afford phenol 31b as a yellow foam in 48%
yield following PCTLC (gradient of 75:20:5 hexanes:THF:TEA
to 60:35:5 hexanes:THF:TEA): 1H NMR (CDCl3) δ 8.12 (d, J
) 2.3 Hz, 1H), 7.47 (d, J ) 8.8 Hz, 1H), 7.47-7.39 (m, 2H),
7.20 (s, 2H), 6.77 (d, J ) 2.0 Hz, 1H), 6.64 (dd, J ) 8.9 and 2.1
Hz, 1H), 6.35 (d, J ) 8.7 Hz, 1H), 4.44 (t, J ) 5.3 Hz, 2H),
3.74 (s, 3H), 3.63 (s, 2H), 2.95 (t, J ) 5.3 Hz, 2H), 2.87-2.79
(m, 4H), 2.69-2.58 (m, 4H), 2.00-1.84 (m, 8H), no phenolic
OH observed. A sample of product 31b was converted to the
dioxalate salt: IR (KBr) 2967, 1718, 1602, 1525, 1466, 1413,
1285, 1271, 1207, 1029, 720 cm-1; FDMS 558 (M + 1). Anal.
(C32H35N3O4S‚2C2H2O4) C, H, N.
1-[2-[[5-[6-Hyd r oxy-3-[[3-m eth oxy-4-[(1-p yr r olid in yl)-
m et h yl]p h en yl]m et h yl]b en zo[b]t h iop h en -2-yl]p yr id -2-
yl]oxy]eth yl]p yr r olid in e (31c). Employing the conditions
described in the preparation of compounds 5a -d , only using
DIBAL-H as the reducing agent, product 31c was prepared
from ketone 31b in 68% yield following PCTLC (gradient of
75:20:5 hexanes:THF:TEA to 60:35:5 hexanes:THF:TEA): 1H
NMR (CDCl3) δ 8.22 (d, J ) 2.4 Hz, 1H), 7.53 (dd, J ) 8.4 and
2.3 Hz, 1H), 7.20 (d, J ) 8.7 Hz, 1H), 7.13 (d, J ) 7.5 Hz, 1H),
6.99 (d, J ) 2.1 Hz, 1H), 6.63 (d, J ) 11.2 Hz, 1H), 6.62-6.58
(m, 2H), 6.45 (dd, J ) 8.7 and 2.1 Hz, 1H), 4.49 (t, J ) 5.7 Hz,
2H), 4.13 (s, 2H), 3.65 (s, 2H), 3.51 (s, 3H), 2.95 (t, J ) 5.7 Hz,
2H), 2.69-2.61 (m, 8H), 1.86-1.78 (m, 8H), no phenolic OH
observed. The free base of the product was converted to the
dioxalate: IR (KBr) 1719, 1611, 1484, 1468, 1421, 1283, 1243,
1218 cm-1; FDMS 544 (M + 1). Anal. (C32H37N3O3S‚2C2H2O4)
C, H, N.
(8) A preliminary account of only limited aspects of this work has
appeared as a communication in Zhang, M.; Bailey, D. L.;
Bastian, J . A.; Briggs, S. L.; Chirgadze, N. Y.; Clawson, D. K.;
Denney, M. L.; Gifford-Moore, D. S.; Harper, R. W.; J ohnson, L.
M.; Klimkowski, V. J .; Kohn, T. J .; Lin, H. S.; McCowan, J . R.;
Richett, M. E.; Sall, D. J .; Smith, A. J .; Smith, G. F.; Snyder, D.
W.; Takeuchi, K.; Utterback, B. G.; Yan, S. C. B. Dibasic Benzo-
[b]thiophene Derivatives as a Novel Class of Active Site Directed
Thrombin Inhibitors: 2. Side chain Optimization and Demon-
stration of In Vivo Efficacy. Bioorg Med. Chem. Lett. 1999, 9,
775-780.
(9) For
a review of Suzuki coupling reactions see Suzuki, A.
Organoboron compounds in New Synthetic Reactions. Pure Appl.
Chem. 1985, 57, 1749-1758.
(10) For a review of Friedel-Crafts acylation, see: Olah, G. H.
Friedel-Crafts and Related Reactions; Wiley: New York, 1963;
Vol. 1, pp 91-115.
(11) Schmid, C. R.; Sluka, J . P.; Duke; K. M. Nucleophilic Aromatic
Substitution on 3-Aroyl-2-Arylbenzothiophenes. Rapid Access to
Raloxifene and Other Selective Estrogen Receptor Modulators.
Tetrahedron Lett. 1999, 40, 675-678.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails for the synthesis and the physical chemical data for the
amino acids used in the preparation of compounds 4b-d , 4f,
4g, 23a , and 27; physical chemical properties of compounds
1a , 4b-g, 5a -d , 7a -l, 8a -i, 10a , 16a , and 17. This material
is available free of charge via the Internet at http://pubs.
acs.org.

Active Site Directed Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 663
(12) For a general review of the Wittig reaction, see: Maryanoff, B.
E.; Reitz, A. B. The Wittig Olefination Reaction and Modifica-
tions Involving Phosphoryl-stabilized Carbanions. Stereochem-
istry, Mechanism, and Selected Synthetic Aspects. Chem Rev.
1989, 89, 863-927.
(23) Shuman, R. T.; Gesellchen, P. D. Development of an Orally
Active Tripeptide Arginal Thrombin Inhibitor. In Integration of
Pharmaceutical Discovery and Development: Case Studies; Bor-
chardt, R. T., Ed.; Plenum Press: New York, 1998; pp 57-80.
(24) Experiments with the rat AV shunt model were performed with
slight modifications of Smith, J . R.; White A. M. Br. J . Phar-
macol. 1982, 77, 29. Drug or vehicle was infused for 15 min prior
to opening the shunt and was continued for an additional 15
min after the shunt was opened. Net clot weights from drug-
treated animals were expressed as a percent of the vehicle-
treated animals run in the same experiments.
(13) Cherry, W. H.; Davies, W.; Ennis, B. C.; Porter, Q. N. The Base-
catalyzed Skeletal Rearrangement of a Thianaphthofluorene
Derivative. Aust. J . Chem. 1967, 20, 313-320.
(14) For a review of the Ullmann ether synthesis, see: Moroz, A. A.;
Shvartsberg, M. S. Ullmann’s Ester Condensation. Usp. Khim.
1974, 43, 1443-61.
(15) Bastian, J . A.; Chirgadze, N. Y.; Denney, M. L.; Gifford-Moore,
D. S.; Sall, D. J .; Smith, G. F.; Wikel, J . H.Diamino Benzo[b]-
thiophene Derivatives as a Novel Class of Active Site Directed
Thrombin Inhibitors. 3. Enhancing Activity by Imposing Con-
formational in the C-4′′ Side Chain. Bioorg Med. Chem. Lett.
1999, 9, 363-368.
(16) Graham, S. L.; Shepard, K. L.; Anderson, P. S.; Baldwin, J . J .;
Best, D. B.; Christy, M. E.; Freedman, M. B.; Gautheron, P.;
Habecker, C. N.; Hoffman, J . M.; Lyle, P. A.; Michelson, S. R.;
Ponticello, G. S.; Robb, C. M.; Schwam, H.; Smith, A. M.; Smith,
R. L.; Sondey, J . M.; Strohmaier, K. M.; Sugrue, M. F.; Varga,
S. L. Topically Active Carbonic Anhydrase Inhibitors. 2. Benzo-
[b]thiophenesulfon-amide Derivatives with Ocular Hypotensive
Activity. J . Med. Chem. 1989, 32, 2548-2554.
(17) Godfrey, A. G.; Bradley, D. A.; Schmid, C. R. Synergistic
Methodologies for the Synthesis of Selective Estrogen Receptor
Modulators. Rapid Access to 2-Aryl-3-Aroylbenzothiophenes,
2-Aryl-3-Aroylbenzofurans, and a Concise Synthesis of Ralox-
ifene. Submitted to J . Org. Chem.
(18) Apparent association constants (Kass) were measured according
to the methods of (a) Smith, G. F.; Gifford-Moore, D.; Craft, T.
J .; Chirgadze, N.; Ruterbories, K. J .; Lindstrom, T. D.; Satter-
white, J . H. Efegatran: A New Cardiovascular Anticoagulant.
In New Anticoagulants for the Cardiovascular Patient; Pifarre,
R., Ed.; Hanley & Belfus, Inc.: Philadelphia, 1997; pp 265-300
and (b) Smith, G. F.; Shuman, R. T.; Craft, T. J .; Gifford-Moore,
D.; Kurz, K. D.; J ones, N. D.; Chirgadze, N.; Hermann, R. B.;
Coffman, W. J .; Sandusky, G. E.; Roberts E.; Van J ackson, C. A
Family of Arginal Thrombin Inhibitors Related to Efegatran.
Sem. Thromb. Hemost. 1996, 22, 173-183.
(25) Plasma pharmacokinetic studies were conducted in male Spra-
gue-Dawley rats in which saline solutions of inhibitor 28c and
31c were administered intravenously at 5 mg/kg via tail vein
injection. Serial blood samples were obtained from the orbital
sinus at 5, 15, 30, 60, 120, 180, and 240 min post dose from 2 or
3 rats per time point. Plasma samples were extracted using Bond
Elute C8 (Varian) cartridges and assayed for drug concentrations
by an HPLC method using a methanol/30 mM ammonium
acetate buffer (pH 4) gradient.
(26) Ruterbories, K. J .; Hanssen, B. R.; Lindstrom, T. D. Unpublished
results.
(27) Maximal plasma concentrations (Cmax), area under the plasma
concentration versus time curve (AUC), half-life, and volume of
distribution were determined from the plasma concentration-
time profiles using standard analysis.
(28) Four male Fischer 344 rats were administered a 5 mg/kg dose
of compound 31c via the tail vein. At 10, 30, 120, and 300 min
after dosing, the rats were sacrificed and samples of blood, liver,
kidney, lung, and heart were obtained. Plasma was obtained by
centrifugation of whole blood. Tissue samples were rinsed in cold
water, blotted dry, weighed, and homogenized in one volume of
purified water. Protein was precipitated using methanol. The
methanolic supernates were made acidic and partitioned against
hexane, and then they were made basic and extracted using
methylene chloride. Sample extracts were concentrated, recon-
stituted, and analyzed by HPLC. Concentrations of 31c in each
sample were determined in relationship to standards prepared
in each tissue matrix. To determine whether glucuronide
metabolites were also present, the methanolic supernates from
liver and kidney samples were diluted with water and extracted
using C8 Bond Elute cartridges, prior to analysis by HPLC.
(29) Unpublished observations.
(30) J ones, C. D.; J evnikar, M. G.; Pike, A. J .; Peters, M. K.; Black,
L. J .; Thompson, A. R.; Falcone, J . F.; Clemens, J . A. Anties-
trogens. 2. Structure-activity studies in a Series of 3-Aroyl-2-
arylbenzo[b]thiophene Derivatives Leading to [6-Hydroxy-
2-(4-hydroxyphenyl)benzo[b]thien-3-yl]-[4-[2-(1-piperidinyl)-
ethoxy]phenyl]methanone Hydrochloride (LY 156758), a Re-
markably Effective Estrogen Antagonist with Only Minimal
Intrinsic Estrogenicity. J . Med. Chem. 1984, 27, 1057-1066.
(31) The preparation of the amino acids used in the synthesis of
compounds 4b, 4c, 4d , 4f, 4g, 23a , and 27 is included in the
Supporting Information.
(19) A detailed account of the X-ray crystallography studies per-
formed on this series of thrombin inhibitors will be provided in
a future publication.
(20) Sall, D. J .; Briggs, S.; Chirgadze, N. Y.; Clawson, D. K.; Gifford-
Moore, D. S.; Klimkowski, V. J .; McCowan, J . R.; Smith, G. F.;
Wikel, J . Dibasic Benzo[b]thiophene Derivatives as a Novel Class
of Active-Site Directed Thrombin Inhibitor. 2. Exploring Interac-
tions at the Proximal (S2) Binding Site. Bioorg Med. Chem. Lett.
1998, 8, 2527-2532.
(21) Smith, G. F.; Neubauer, B. L.; Sundboom, J . L.; Best, K. L.;
Goode, R. L.; Tanzer, L. R.; Merriman, R. L.; Frank, J . D.;
Herrmann, R. G. Correlation of the In Vivo Anticoagulant,
Antithrombotic, and Antimetastatic Efficacy of Warfarin in the
Rat. Thromb. Res. 1988, 50, 163-174.
(22) Unpublished observations.
J M9903388