Orally efficacious thrombin inhibitors with cyanofluorophenylacetamide as the P2 motif

Article abstract of DOI:10.1016/j.bmcl.2008.03.087

2-Cyano-6-fluorophenylacetamide was explored as a novel P2 scaffold in the design of thrombin inhibitors. Optimization around this structural motif culminated in 14, which is a potent thrombin inhibitor (K_i = 1.2 nM) that exhibits robust effica

Full text of DOI:10.1016/j.bmcl.2008.03.087

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2865–2870
Orally efficacious thrombin inhibitors
with cyanofluorophenylacetamide as the P2 motif
Kevin D. Kreutter, Tianbao Lu, Lily Lee, Edward C. Giardino, Sharmila Patel,
Hui Huang, Guozhang Xu, Mark Fitzgerald, Barbara J. Haertlein, Venkatraman Mohan,
Carl Crysler, Stephen Eisennagel, Malini Dasgupta, Martin McMillan, John C. Spurlino,
Norman D. Huebert, Bruce E. Maryanoff, Bruce E. Tomczuk,
Bruce P. Damiano and Mark R. Player*
Johnson & Johnson Pharmaceutical Research & Development, Welsh & McKean Roads, Spring House, PA 19477, USA
Received 11 February 2008; accepted 31 March 2008
Available online 8 April 2008
Abstract—2-Cyano-6-fluorophenylacetamide was explored as a novel P2 scaffold in the design of thrombin inhibitors. Optimization
around this structural motif culminated in 14, which is a potent thrombin inhibitor (Ki = 1.2 nM) that exhibits robust efficacy in
canine anticoagulation and thrombosis models upon oral administration.
ꢀ 2008 Elsevier Ltd. All rights reserved.
Venous thromboembolism (VTE) and atrial fibrillation
(AF) account for approximately 100,000 deaths per year
in the U.S. VTE involves the formation of a deep venous
thrombus which may dislodge and move to the pulmon-
ary artery as a pulmonary embolus, and AF can induce
thrombus formation in the heart leading to strokes.1
VTE and AF-induced stroke are currently treated with
heparin derivatives, which require subcutaneous deliv-
ery, and warfarin, which requires dose titration to min-
imize bleeding complications.2
The weakly basic P1 oxyguanidine (pKa ꢀ 7), when com-
bined with a P2 phenyl scaffold (e.g., 1; Fig. 1), gave ro-
bust PK in dogs (F = 73%; iv t1/2 = 4.4 h) along with
good in vitro potency (Ki = 4.0 nM).7 However, the con-
centration of 1 required to double the activated partial
thromboplastin time (aPTT)8 in human plasma, an in vi-
tro measure of an anticoagulant’s potency, was rela-
tively high (2 · aPTT = 7.9 lM), presumably in part
due to >98% plasma protein binding.7 By contrast,
O
Since the serine protease thrombin catalyzes the conver-
sion of soluble fibrinogen to insoluble fibrin in the clot-
ting cascade, it is widely believed that an oral thrombin
inhibitor could provide a new standard of care in antico-
agulation therapy. Early efforts at thrombin inhibitor
design frequently employed a highly basic P1 moiety
such as guanidine (pKa ꢀ 13), but these usually suffered
from poor pharmacokinetics (PK).3 Strategies to im-
prove PK include the use of a prodrug-masked P1
(e.g., ximelagatran4 and dabigatran etexilate5) or a
weakly basic or nonbasic P1 isostere.6
H
N
N
NH2
O
O
1
O
NH
NH
O
O
O
S
N
O
2
N
N
H
N
H
NH2
H
O
CN
O
NH
O
14
N
N
N
N
NH2
H
H
H
F
F
F
Keywords: Thrombin inhibitor; Trypsin; Serine protease; Anticoagu-
lant.
P3
P2
P1
*
Corresponding author. Tel.: +1 610 458 6980; fax: +1 610 458
8258; e-mail: mplayer@prdus.jnj.com
Figure 1. Oxyguanidine-containing thrombin inhibitors.
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.03.087

2866
K. D. Kreutter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2865–2870
N
O
and slightly less plasma protein-bound than 4, it was
also significantly more potent in vitro (Table 1).
N
N
H
N
H
N
F
F
O
The synthesis of 5 is described in Scheme 1. Reaction of
sodium diethyl malonate with trifluorobenzonitrile 6,
followed by Krapcho decarboxylation, afforded a
1.35:1 mixture of regioisomers that were separated by
flash chromatography to afford key P2 intermediate 7
as the minor regioisomer in 34% overall yield. Fluori-
nated6 P3 amines 10 were accessed via reaction of a (het-
ero)aryl halide 8 with ethyl bromodifluoroacetate and
copper bronze to afford activated esters 9.14 The phe-
nyl-derived amines 10 were then efficiently obtained
via treatment of 9 with ammonia followed by borane
reduction of the amide. Since borane overreacted with
the analogous pyridyl-derived amides, pyridyl-derived
esters 9 (A = N) were reduced to the alcohol, converted
to the azide in two steps, and cleanly hydrogenated to
the amines 10 with palladium on carbon. The P3–P2
linkage was established by a straightforward SNAr reac-
tion between P3 amines 10 and P2 electrophile 7. Subse-
quent saponification, amide bond formation with a P1
amine (e.g., oxyguanidine-containing 13), and, if neces-
sary, Boc deprotection, afforded final targets such as 5.
The P1 amine 13 was obtained in a four-step sequence
NH2
NH
O
O
H
N
3
thrombin
R
O
O
N
H
N
H
N
H
NH2
F
F
F
O
O
4: R = Cl
5: R = CN
H
N
thrombin
Figure 2. Fluorophenyl compounds 4–5 as H-bond acceptors isosteric
with pyrazinone 3.
while pyridinone-based 29 displayed an improved
2 · aPTT value of 0.23 lM and improved efficacy over
1 (data not shown), it had only modest oral bioavailabil-
ity (dog F = 24%; rat F = 6%).9
Here we report on a novel 2-cyano-6-fluorophenylaceta-
mide series that led to 14, a molecule with comparable
preclinical oral efficacy to 2 but with improved canine
oral bioavailability (Fig. 1).
CN
CN
F
a, b
CO2Et
F
F
7
In previously published work on the chlorofluoropheny-
lacetamide P2 scaffold10 typified by 4 (Fig. 2), the fluo-
rine atom proved to be a competent hydrogen-bond
acceptor isosteric with the carbonyl oxygen of the pyrid-
inone11 and pyrazinone12 (e.g., 3) P2 scaffolds, as evi-
F
F
6
c
d
or e
CN
R
CO2Et
R
A
F
A
X
F
F
10
9
˚
8
denced by the F–H–N heteroatom distance of 3.17 A.
f, g, h
O
However, numerous examples of the chlorofluoropheny-
lacetamide series suffered from poor dog pharmacody-
namic parameters13 perhaps because of relatively high
plasma protein binding, and required multistep func-
tional group manipulation for P3 installation.10 It was
hypothesized that replacement of the chloro moiety with
a cyano group6 would increase the polarity for the ser-
ies, thereby potentially reducing plasma protein binding,
while allowing for a shorter, higher-yielding synthetic
route to targets. As expected, 5 was not only more polar
R
R
A
N
H
N
H
P1
NH
A
NH2
F
F
F
F
10
O
5: R = H, A = CH, P1 =
N
H
NH2
NBoc
NHBoc
k, l
Z
ONH2
i, j
O
Z
OH
N
H
H2N
N
H
N
H
12
11
13
Scheme 1. Reagents and conditions: (a) NaH (2.2 mol equiv), diethyl
malonate (2.2 mol equiv), THF, rt, 4 d, 98% (1.35:1 regioisomers); (b)
LiCl (1 mol equiv), water (1.3 mol equiv), DMSO, 120 ꢁC, 40 min, 35%
(desired regioisomer); (c) for A = CH, X = I or A = N, X = Br: Cu
bronze (2.2 mol equiv), ethyl bromodifluoroacetate (1.2 mol equiv),
DMSO, rt, 3 d, 56–97%; (d) for A = CH; i—NH3, MeOH, rt, 13 h,
95%; ii—borane, THF, reflux, 9 h, 67% after acid workup; (e) for
A = N; i—NaBH4, EtOH, 0 ꢁC, 3 h, 100%; ii—I2, imidazole, PPh3,
toluene/MeCN (2:1), 0 ꢁC ! 90 ꢁC, 33%; iii—NaN3, DMSO, 90 ꢁC,
2 d, 93%; iv—H2 (balloon), EtOAc, 10% Pd/C, 24 h, 86%; (f) 7
(1.25 mol equiv), (i-Pr)2NEt (1.3 mol equiv), DMSO, 110 ꢁC, 3 d, 64%;
(g) 1 M LiOH (aq)/MeOH/THF (1:1:2), 50 ꢁC, 1 h, 100%; (h) i—13
(1.2 mol equiv), BOP, (i-Pr)2NEt, CH2Cl2/MeCN (1:1), rt, 8 h, 83%;
ii—CH2Cl2/CF3CO2H/anisole (3:1:0.4, v/v), rt, 10 h, 84%; (i) N-
hydroxyphthalimide, PPh3, (CO2Et)N@N(CO2Et), THF, rt, o/n,
91%; (j) 40% MeNH2, THF/EtOH (1:1), rt, 1 h, 95%; (k) N,N0-bis-Boc-
1-guanylpyrazole, DMF, rt, o/n, 93%; (l) H2 (balloon), 10% Pd/C,
THF/EtOH (1:1), rt, 30 min, 61%.
Table 1. In vitro potency, plasma protein binding, and 2 · aPTT
values for compounds 1–5 using human thrombin and plasma
Compound
Ki thrombin
(nM)
Plasma
protein
binding (%)
clogP
2 · aPTT
(lM)
1a
2b
3c
4
4
4
>98
45.7
—
2.98
ꢁ1.35
1.52
7.9
0.23
0.29
—
0.1
47
2.3
95d
3.04
2.52
5
92.6d
0.64
a Ref. 7.
b Ref. 9.
c Ref. 12.
d Dog plasma protein binding.

K. D. Kreutter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2865–2870
Table 2. P3 structure–activity relationships
2867
5
CN
6
4
3
O
NH
R
O
A
2
N
H
N
H
N
H
NH2
F
F
F
b
Compound
R
A
Ki (nM) Trypsin 2 · aPTTb PPB
Cacoc
Dog PDd
t (h) P2 · aPTTe Fold aPTT maxf
ratioa
(lM)
(%)
5
H
CH
N
2.3
1.2
478
1170
770
0.64
0.36
—
93.7
56
1.1, 4.6
0.9, 8.5
—
—
5
—
14
15
16
17
18
19
20
21
22
23
24
25
26
H
2.83
—
6-Me
5-Me
4-Me
3-Me
4-Cl
N
10
—
—
—
—
N
2.9
1.2
148
0.43
0.41
0.22
—
—
—
N
4080
1110
6840
1580
760
97.6
—
0.2, 2.1 3.5
2.88
2.38
2.51
1.23
1.87
—
N
0.79
0.57
0.38
1.7
—
—
4.5
5.5
0
N
—
P3 = 8-quinolyl
3-Cl
3-F
0.44
0.76
0.59
0.68
—
99.6
95.2
85
1.7, —
0.8, 9.9
0.7, 4.7
0.5, 2.6
—
CH
CH
0
3.2
1.8
750
—
0
3,4-diF
6-SO2Me
H
CH
CH
2000
1850
770
99.0
—
1.67
1.04
1.89
1.20
0.65
1.3
0
N-Oxide
N-Oxide
—
—
—
—
—
—
0
4-Cl
0.66
5300
0
a Ki trypsin/Ki thrombin.
b In vitro assay using human plasma.
c A ! B, B ! A; Papp in 10ꢁ6 cm/s.
d Single 10 mg/kg oral dose in 2–5 dogs.
e Number of hours that the aPTT is P2 · the baseline value over the course of 8 h post-dose (7 time points).
f Peak aPTT multiple of baseline seen over 8 h.
from alcohol 11 in 49% overall yield, with the final step
requiring careful hydrogenolysis of the Cbz group to
avoid N–O bond cleavage.
halogenated derivatives 21–23. However, while selectiv-
ity increased for these analogues, Caco permeability be-
came somewhat worse. Finally, polar electron-
withdrawing substituents at the ortho position of the
P3 (hetero)aromatic ring were found to confer potency
(24–26) as expected.15 By contrast, modestly potent 15
may not be able to satisfy a binding mode seen in previ-
ous thrombin co-crystal structures whereby the ortho
polar moiety (e.g., N,N-oxide) faces solvent,15 with
Building upon the P3 SAR generated from the chloro-
fluorophenylacetamide series,10 it was discovered that
the 2-pyridyl analogue 14 provided a roughly 2-fold
improvement in human thrombin potency, 2 · aPTT va-
lue, and trypsin selectivity compared to phenyl analogue
5 (Table 2). In addition, 14 had significantly decreased
plasma protein binding compared to 5, in line with its
reduced clogP value of 1.61. However, neither 5 nor
14 had robust permeability as measured by apical to
basolateral transfer (A ! B) across a Caco cell mono-
layer (Papp = 1.1 and 0.9 · 10ꢁ6 cm/s, respectively), and
indeed, 14 demonstrated increased efflux as indicated
by the Caco B ! A/A ! B ratio of 9.7 (compared to
4.7 for 5). This was of potential concern, since Caco per-
meability had correlated with bioavailability for several
other thrombin inhibitor series, including those repre-
sented by 1 and 2 (1 A ! B Papp = 4.1 · 10ꢁ6 cm/s, with
rat F = 24% and 2 A ! B Papp = 0.7 · 10ꢁ6 cm/s, with
rat F = 6%).
In an attempt to improve the in vitro parameters of 14,
particularly Caco permeability, methyl substitution on
the pyridyl P3 moiety was explored. Analogues 17 and
18 proved to be potent and selective with 2 · aPTT val-
ues comparable to that seen for 14. However, methyl
substitution significantly increased plasma protein bind-
ing while showing no improvement in Caco permeability
(17, Table 2). Attempts were made to enhance the selec-
tivity and permeability of the phenyl analogue 5 with
Figure 3. Crystal structure of 14 bound to thrombin (PDB ID:2C27).

2868
K. D. Kreutter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2865–2870
3.0
2.5
2.0
1.5
1.0
90
80
1
2
14
19
27
70
60
50
40
30
20
10
0
vehicle, N=6
1, 10 mg/kg, N=3
2, 3 mg/kg, N=3
14, 3 mg/kg, N=3
N = 3-5/compound
0
1
2
3
4
5
6
7
8
2
3
4
5
Time Post Oral Administration (hr)
Time Post Oral Administration (hr)
Figure 4. aPTT following a single 10 mg/kg oral dose to conscious
dogs.
Figure 5. Oral antithrombotic efficacy of 1, 2, and 14 in the A-V shunt
model in anesthetized dogs.
a hERG ion channel patch–clamp study compared to
1 (0% inh. vs 85% inh. at 10 lM, respectively). In addi-
tion, 14 caused no significant heart rate-corrected QT
interval prolongation in guinea pigs up to 10 mg/kg iv
(data not shown).
the ortho hydrogen found in 24–26 (but not 15) access-
ing a favorable r–p interaction with Trp215.10,15
The X-ray crystal structure of 14 bound to human
thrombin is shown in Figure 3. As in previous struc-
tures, the oxyguanidine P1–S1 hydrogen-bonding net-
work is intact,10 and the hydrogen bonds with Gly216
and Ser214 are also maintained.6,10,15 However, the ni-
Therefore, the efficacy of 14 was explored in a dog arte-
riovenous (A-V) shunt model16 (Fig. 5) and a rat electri-
cally stimulated carotid artery (ESCA) model17 (Fig. 6).
Compound 14 inhibited thrombus formation by 67–51%
for at least 5 h following a 3 mg/kg oral dose in dogs
(Fig. 5). This is comparable to the antithrombotic effect
of 2, and is superior to that for 1. The rat intravenous
data mirrored this trend: 14 (iv ED50 ꢀ 40 lg/kg) and
2 have comparable efficacy, and both are substantially
superior to 1 as well as to the marketed iv antithrombot-
ic argatroban (Fig. 6).
˚
trile nitrogen of 14 additionally forms a 2.8 A hydrogen
bond with a water molecule that in turn forms H-bonds
˚
˚
with Lys60F (2.8 A) and His57 (2.9 A). This nitrile-
based H-bonding network may account in part for the
increased potency of the cyanofluorophenylacetamides
versus chlorofluorophenylacetamides.
Since a sampling of the compounds in Table 2 showed
good stability (69–100% remaining after a 10 min incu-
bation with human and dog microsomes at 37 ꢁC; four
examples), 10 of the analogues were profiled in an oral
dog pharmacodynamic (PD) assay (Fig. 4). Notably,
14, 17, and 19 demonstrated a P2.5-fold increase in
aPTT for at least 3 h following a single 10 mg/kg oral
dose (Table 2). This degree of anticoagulation is supe-
rior to that shown by 1, and indeed to that seen for 2
(Fig. 4). Interestingly, chlorofluorophenylacetamide 27
(Ki = 1.8 nM; clogP = 2.69), the chloro congener of 19
(Ki = 0.6 nM; clogP = 2.16), was significantly less active
than 19 (Fig. 4).
The PK parameters of several analogues are listed in Ta-
ble 3. The dog PK of 14, 17, and 18 support the dog PD
described in Table 2. All three have oral half-lives great-
er than 2 h, with Cmax values at least 7-fold higher than
the concentrations needed to double the activated par-
tial thromboplastin times of (human) plasma in vitro.
Maximum Occlusion Time = 30 min
30
25
In general, improvements in canine PD in this series cor-
relate with lower in vitro 2 · aPTT values and lower
plasma protein binding (or lower clogP). However,
the presence of additional polar heteroatoms (e.g.,
24–26) had a detrimental effect on PD, perhaps due to
an unacceptable loss of permeability.
20
Argatroban
1
15
2
14
Since 14 had robust PD as well as an attractive in vitro
profile across species (e.g., P86% remaining after
10 min incubation with human, dog, or rat microsomes;
669% PPB in human, dog, or rat plasma), it was
screened for off-target binding and cardiovascular
safety. Compound 14 showed little effect against a Cerep
panel of 50 receptors, ion channels, and transporters
(<40% inhibition at 10 lM), and was notably clean in
10
5
Vehicle Occlusion Time = 11 min
N = 5/dose
0.01
0.1
1
10
Dose (mg/kg, i.v.)
Figure 6. Intravenous antithrombotic efficacy of 1, 2, and 14 relative to
argatroban in the ESCA rat model.

K. D. Kreutter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2865–2870
Table 3. Dog, rat, and monkey PK of select analogues
2869
Speciesa
Cmax po (lM)
t1/2 po (h)
t1/2 iv (h)
Clb (mL/min/kg)
F (%)
c
Vdss (L/kg)
Compound
14
17
18
14
17
Dog
Dog
Dog
Ratd
Rat
3.5
2.8
2.5
2.5
3.2
—
2.9
3.6
2.8
1.0
1.3
13 (42%)
18 (58%)
15 (49%)
67 (122%)
30 (55%)
2.9
3.6
2.9
ꢀ6
2.3
49
45
65
2
3.3
0.03
0.25
2.5
2
a Unless otherwise noted: dog 10 mg/kg, po, 1 mg/kg, iv; rat 30 mg/kg, po, 3 mg/kg, iv.
b Clearance (as a percentage of liver blood flow in parentheses).
c Volume of distribution at steady state. Both clearance and Vd determined from iv dosing.
d 10 mg/kg, po, 2 mg/kg, iv.
Table 4. P1 structure–activity relationships
CN
O
P1
N
N
H
N
H
F
F
F
Compound
P1
Ki (nM)
Trypsin ratioa
640
PPBb (%)
Cacoc
NH
O
34
9
—
85
89
—
0.6, 0.3
N
NH2
Me
N
35
36
37
11
400
—
NH2
NH2
N
0.77
33
2000
>300
3.9, 13
—
F
N
O
N
38
19
>500
98.7
—
NH2
a Ki trypsin/Ki thrombin.
b In vitro assay using human plasma.
c A ! B, B ! A; Papp in 10ꢁ6 cm/s.
In addition, the dog bioavailabilities ranged from 45%
to 65%, an improvement over that observed for 2
(24%).9 However, 14 and 17 had unacceptably low bio-
availability in rats (2%) despite excellent rat microsomal
stability (>95% remaining after a 10 min incubation),
and the less polar analogues 21 and 23 had only margin-
ally better rat bioavailability. In addition, clearance
sometimes exceeded hepatic blood flow, indicating the
possibility of extrahepatic metabolism in the rat.18
into the ascending colon also yielded ꢀ2% bioavailabil-
ity as with oral dosing, indicating that the molecule is
not colonically absorbed (data not shown).
Since increased hydrogen-bond donor count was found to
correlate with decreased Caco permeability across throm-
bin chemical series within this discovery program, oxy-
guanidine was replaced with P1s containing fewer H-
bond donors (Table 4). The synthesis of N-methyl-oxy-
guanidine P1 31 found in 34 is described in Scheme 2,
while the P1 moieties in 35–38 were previously
reported.6,12,19
In order to ascertain whether the poor rat bioavailability
of 14 was caused by poor permeability across the intes-
tinal mucosa, or by metabolism upon first pass through
the liver, rats were dosed orally at 10 mg/kg and blood
samples from the portal vein (pre-liver) and jugular vein
(post-liver) were collected over 6 h. Neither sample site
contained significant amounts of 14, indicating that 14
had little propensity for diffusion across the small intes-
tine into the portal vein. Dosing 14 by cannula directly
Unfortunately, 34 provided no improvement in Caco
A ! B permeability, and potency dropped 7-fold rela-
tive to 14. Methylation at the distal oxyguanidine nitro-
gen had dramatically reduced thrombin potency in
another thrombin series (data not shown) and so was
not explored here. Compounds 35, 37, and 38 also

2870
K. D. Kreutter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2865–2870
4. Gustafsson, D.; Nystrom, J.; Carlsson, S.; Bredberg, U.;
OMs
BocHN
Eriksson, U.; Gyzander, E.; Elg, M.; Antonsson, T.;
Hoffmann, K.; Ungell, A.; Sorensen, H.; Nagard, S.;
Abrahamsson, A.; Bylund, R. Thromb. Res. 2001, 101,
171.
O
O
29
b,c,d
a
HO
O
N
OMe
BocHN
N
OMe
Me
Me
30
28
NH
CN
O
5. Hauel, N. H.; Nar, H.; Priepke, H.; Ries, U.; Stassen, J.-M.;
Wienen, W. J. Med. Chem. 2002, 45, 1757.
O
H2N
N
NH2
6. Burgey, C. S.; Robinson, K. A.; Lyle, T. A.; Sanderson, P.
E. J.; Lewis, S. D.; Lucas, B. J.; Krueger, J. A.; Singh, R.;
Miller-Stein, C.; White, R. B.; Wong, B.; Lyle, E. A.;
Williams, P. D.; Coburn, C. A.; Dorsey, B. D.; Barrow, J.
C.; Stranieri, M. T.; Holahan, M. A.; Sitko, G. R.; Cook,
J. J.; McMasters, D. R.; McDonough, C. M.; Sanders, W.
M.; Wallace, A. A.; Clayton, F. C.; Bohn, D.; Leonard, Y.
M., ; Detwiler, T. J., Jr.; Lynch, J. J., Jr.; Yan, Y.; Chen,
Z.; Kuo, L.; Gardell, S. J.; Shafer, J. A.; Vacca, J. P.
J. Med. Chem. 2003, 46, 461.
N
N
H
OH
Me
31
F
F
F
32
F
CN
F
F
f
e
O
34
N
N
H
O
F
F
F
F
F
33
Scheme 2. Reagents and conditions: (a) NaH (0.94 mol equiv), 29
(0.77 mol equiv), DMF, 0 ꢁC ! rt, 6%; (b) KOH (2.6 mol equiv), water,
90 ꢁC, 18 h, 85%; (c) N,N0-bis-Boc-1-guanylpyrazole, DMF, 50 ꢁC, o/n,
63%; (d) 12 N HCl (aq), rt, 30 min, 100%; (e) pentafluorophenol (3 mol
equiv), (i-Pr)N@N( i-Pr) (2 mol equiv), DMF, CH2Cl2, rt, 24 h, 50%; (f)
31, DMF, (i-Pr) 2NEt (4 mol equiv), rt, 12 h, 75%.
7. Lu, T.; Markotan, T.; Coppo, F.; Tomczuk, B.; Crysler,
C.; Eisennagel, S.; Spurlino, J.; Gremminger, L.; Soll, R.
M.; Giardino, E. C.; Bone, R. Bioorg. Med. Chem. Lett.
2004, 14, 3727.
8. Langdell, R. D.; Wagner, R. H.; Brinkhous, K. M. J. Lab.
Clin. Med. 1953, 41, 637.
9. Maryanoff, B. E.; McComsey, D. F.; Costanzo, M. J.;
Yabut, S. C.; Lu, T.; Player, M. R.; Giardino, E. C.;
Damiano, B. P. Chem. Biol. Drug Des. 2006, 68, 29.
10. Lee, L.; Kreutter, K. D.; Pan, W.; Crysler, C.; Spurlino, J.;
Player, M. R.; Tomczuk, B.; Lu, T. Bioorg. Med. Chem.
Lett. 2007, 17, 6266.
lacked sufficient potency for further profiling. Com-
pound 36 was potent and selective, with reasonable
PPB and improved Caco permeability (Table 4). Unfor-
tunately, its dog PK profile (oral Cmax = 0.8 lM; iv
t1/2 = 1.1 h; Cl = 34 mL/min/kg; 26% F at 10 mg/kg,
po, 1 mg/kg, iv) was inferior in all respects to that for
14, and it was inactive in the oral dog PD assay (aPTT
maximum multiple of baseline = 1.3, compared to 2.8
for 14). Perhaps phase II conjugation of the amino
group6 was responsible for the poor PK of 36, since it
had excellent dog microsomal stability (100% remaining
after 10 min at 37 ꢁC).
11. Tamura, S. Y.; Semple, J. E.; Reiner, J. E.; Goldman, E.
A.; Brunck, T. K.; Lim-Wilby, M. S.; Carpenter, S. H.;
Rote, W. E.; Oldeshulte, G. L.; Richard, B. M.; Nutt, R.
F.; Ripka, W. C. Bioorg. Med. Chem. Lett. 1997, 7, 1543.
12. Sanderson, P. E. J.; Lyle, T. A.; Cutrona, K. J.; Dyer, D.
L.; Dorsey, B. D.; McDonough, C. M.; Naylor-Olsen, A.
M.; Chen, I.-W.; Chen, Z.; Cook, J. J.; Cooper, C. M.;
Gardell, S. J.; Hare, T. R.; Krueger, J. A.; Lewis, S. D.;
Lin, J. H.; Lucas, B. J., Jr.; Lyle, E. A.; Lynch, J. J.;
Stranieri, M. T.; Vastag, K.; Yan, Y.; Shafer, J. A.; Vacca,
J. P. J. Med. Chem. 1998, 41, 4466.
In conclusion, the novel 2-cyano-6-fluorophenylaceta-
mide P2 scaffold, after suitable optimization of the P1
and P3 moieties, can provide potent thrombin inhibi-
tors. Thus, we obtained 14, as a potent thrombin inhib-
itor with robust dog PK and efficacy. However, within
this series of thrombin inhibitors, acceptable rat bio-
availability proved elusive.
13. Table 2e,f
.
14. Eto, H.; Kaneko, Y.; Sakamoto, T. Chem. Pharm. Bull.
2000, 48, 982.
15. Burgey, C. S.; Robinson, K. A.; Lyle, T. A.; Nantermet, P.
G.; Selnick, H. G.; Isaacs, R. C. A.; Lewis, S. D.; Lucas,
B. J.; Krueger, J. A.; Singh, R.; Miller-Stein, C.; White, R.
B.; Wong, B.; Lyle, E. A.; Stranieri, M. T.; Cook, J. J.;
McMasters, D. R.; Pellicore, J. M.; Pal, S.; Wallace, A. A.;
Clayton, F. C.; Bohn, D.; Welsh, D. C.; Lynch, J. J.; Yan,
Y.; Chen, Z.; Kuo, L.; Gardell, S. J.; Shafer, J. A.; Vacca,
J. P. Bioorg. Med. Chem. Lett. 2003, 13, 1353.
16. Sklar, A. H.; Bond, G. C.; Lightfoot, B.; Fowler, B.;
Ginsburg, J. J. Lab. Clin. Med. 1988, 111, 537.
17. Charlton, P. A.; Faint, R. W.; Bent, F.; Bryans, J.;
Chicarelli-Robinson, I.; Mackie, I.; Machin, S.; Bevan, P.
Thromb. Haemost. 1996, 75, 808.
References and notes
1. Rosamond, W.; Flegal, J.; Friday, G.; Furie, K.; Go, A.;
Greenlund, K.; Haase, N.; Ho, M.; Howard, V.; Kissela,
B.; Kittner, S.; Lloyd-Jones, D.; McDermott, M.; Meigs,
J.; Moy, C.; Nichol, G.; O’Donnell, C. J.; Roger, V.;
Rumsfeld, J.; Sorlie, P.; Steinberger, T. T.; Wasserthiel-
Smoller, S.; Hong, Y. Circulation 2007, 115, 69.
2. Ali, S.; Hong, M.; Antezano, E. S.; Mangat, I. Cardiovasc.
Hematol. Disord. Drug Targets 2006, 6, 233.
3. Masic, L. P. Curr. Med. Chem. 2006, 13, 3627.
18. Krishna, D. R.; Klotz, U. Clin. Pharmacokinet. 1994, 26,
144.
19. Lepore, S. D.; Wiley, M. R. J. Org. Chem. 2000, 65, 2924.