Highly efficacious factor Xa inhibitors containing α-substituted phenylcycloalkyl P4 moieties

Article abstract of DOI:10.1016/j.bmcl.2008.11.049

We previously disclosed a series of highly potent FXa inhibitors bearing α-substituted (CH₂NR¹R²) phenylcyclopropyl P4 moieties in the pyrazolodihydropyridone core system. Herein, we describe our continuous SAR efforts in

Full text of DOI:10.1016/j.bmcl.2008.11.049

Bioorganic & Medicinal Chemistry Letters 19 (2009) 462–468
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Highly efficacious factor Xa inhibitors containing
phenylcycloalkyl P4 moieties
a-substituted
*
Jennifer X. Qiao , Sarah R. King, Kan He, Pancras C. Wong, Alan R. Rendina, Joseph M. Luettgen,
Baomin Xin, Robert M. Knabb, Ruth R. Wexler, Patrick Y. S. Lam
Research and Discovery, Bristol-Myers Squibb Company, P.O. Box 5400, Princeton, NJ 08543-5400, USA
a r t i c l e i n f o
a b s t r a c t
We previously disclosed a series of highly potent FXa inhibitors bearing a
-substituted (CH2NR1R2) phe-
nylcyclopropyl P4 moieties in the pyrazolodihydropyridone core system. Herein, we describe our contin-
uous SAR efforts in this series. Effects of the C-3 substitution of the pyrazolodihydropyridone core and the
Article history:
Received 23 October 2008
Revised 10 November 2008
Accepted 12 November 2008
Available online 18 November 2008
a
-substitution (R group) of the cyclopropyl ring on FXa binding affinity (FXa Ki), human plasma antico-
agulant activity (PT EC2Â) and permeability are discussed. A set of compounds obtained from optimiza-
tion of the R group and the C-3 substituent were orally bioavailable in dogs. Furthermore, representative
compounds were highly efficacious in the rabbit arterio-venous shunt thrombosis model (EC50s = 29–
81 nM).
Keywords:
Factor Xa inhibitors
a-Substituted phenylcyclopropyls
Ó 2008 Elsevier Ltd. All rights reserved.
SAR
Pharmacokinetic
Rabbit arterio-venous shunt
Anticoagulants are key agents for the prophylaxis and treat-
ment of thromboembolic disorders. Limitations of the existing
anticoagulants warfarin and heparins establish a need for newer
therapies. The past decade saw the development of a class of novel
anticoagulants designed to inhibit thrombin generation by target-
ing the direct inhibition of coagulation factor Xa (FXa), a serine
protease at the conjunction of the intrinsic and the extrinsic path-
ways in the coagulation cascade. Both preclinical and clinical data
have shown that inhibition of FXa is an effective approach in the
treatment of venous and arterial thrombosis.1,2 Several selective
and orally active small-molecule FXa inhibitors, such as razax-
aban,3 apixaban,4 LY517717,2g YM150,2g DU-176b,2g and rivarox-
aban,2g have entered clinical development, some of which are in
the late-stage clinical trials for the prevention and treatment of
thromboembolic diseases.
to improve anticoagulant activity, and to modulate physicochemi-
cal properties such as permeability. A set of compounds in this ser-
ies underwent PK studies and four compounds were tested in the
rabbit arterio-venous shunt model for antithrombotic effects.
We began to investigate the effects of polar C-3 substituents of
the pyrazolodihydropyridone core on FXa potency and anticoagu-
lant activity in the early analogs5 (R = CH2NR1R2). Table 1 shows
a comparison of
a
-CH2NR1R2-substituted phenylcyclopropyl P4
analogs 1–6 bearing various C-3 groups. Replacing CF3 in 1a–5a
with a SO2Me, CONH2, or CN group resulted in compounds with in-
creased polarity, indicated by the increased polar surface area
(PSA) and lowered overall lipophilicity (clogP). Compounds 1b–
5b, 1c–4c, and 1d–4d bearing a SO2Me, CONH2, or CN group,
respectively, had similarly high FXa affinity as the corresponding
CF3 analogs 1a–5a (FXa Ki in the pM range). Furthermore, com-
pounds 1b–5b, 1c–4c, and 1d–4d generally had higher anticoagu-
lant activity (PT EC2Â) compared with the corresponding CF3
analogs 1a–5a. This is particularly true for the less basic com-
pounds, such as the morpholinyl analogs 4a–4d (PT EC2Â of CF3
We previously discussed5 the discovery of a series of highly po-
tent and selective pyrazole bicyclic-based FXa inhibitors bearing a-
substituted (CH2NR1R2) phenylcyclopropyl P4 moieties. These have
the general structure A3 shown in Figure 1, where the perpendicu-
lar conformation of the phenylcyclopropyl groups in A was used to
mimic the bioactive conformation of the ortho-substituted biphe-
nyl moieties in B (Fig. 1).5 We herein describe our continuous
SAR efforts on this series with a general structure C (Fig. 1) by vary-
ing both the C-3 substituent of the pyrazolodihydropyridone core
analog 4a was 14
lM, while PT EC2Â values of 4b (SO2Me), 4c
(CONH2) and 4d (CN) were 1.7, 1.5, and 4.1
lM, respectively).
The improved in vitro clotting activity can be attributable to lower
protein binding, that is, the increased free fraction in human plas-
ma caused by the more polar C-3 groups. On the other hand, both
the basic pyrrolidines 3b (SO2Me C-3) and 3c (CONH2 C-3) and the
neutral pyrrolidones 6b and 6c showed good FXa binding activity
and good PT anticoagulant activity.
and the
a-substituent (R group) on the phenylcyclopropyl group
Table 1 indicates that the Caco-2 apparent permeability values
(Papp, apical to basolateral) followed a general trend: compounds
* Corresponding author. Tel.: +1 609 818 5298; fax: +1 609 818 6810.
E-mail address: jennifer.qiao@bms.com (J.X. Qiao).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.11.049

J. X. Qiao et al. / Bioorg. Med. Chem. Lett. 19 (2009) 462–468
463
R1
R2
N
F3C
F3C
N
R1
N
R2
N
N
N
N
N
O
O
P4
P4
A
B
P1
P1
OMe
OMe
C3
N
R
N
N
O
P4
P1
C
OMe
Figure 1. General structures.
Table 1
Effect of C-3 substitution in phenylcyclopropylmethylamine derivatives (R = CH2NR1R2)a,b
C3
N
R
N
N
O
OMe
Compound
C3
R
FXa Ki (nM)
PT EC2Â
(
lM)
Caco-2 Papp (nm/s)c
HPLC LogP
PSA (Sq. Ang.) (PreCalc)e
clogP (PreCalc)e
1a
1b
1c
1d
CF3
CH2NHMe
CH2NHMe
CH2NHMe
CH2NHMe
0.18
0.24
0.055
0.15
2.9
1.2
0.86
1.5
26
1
1
3.2
—
—
58
95
99
75
3.54
1.31
1.24
2.14
SO2Me
CONH2
CN
6
2.1
2a
2b
2c
2d
CF3
CH2NMe2
CH2NMe2
CH2NMe2
CH2NMe2
0.035
0.051
0.041
0.048
1.3
0.83
1.2
85
8
5
3.4
—
1.6
2.5
47
80
89
67
4.08
1.85
1.78
2.68
SO2Me
CONH2
CN
0.79
46
3a
3b
3c
3d
CF3
CH2-N-pyrrolidinyl
CH2-N-pyrrolidinyl
CH2-N-pyrrolidinyl
CH2-N-pyrrolidinyl
0.021
0.054
0.074
0.027
1.4
70
10
10
72
3.4
—
—
46
82
89
65
4.88
2.65
2.58
3.48
SO2Me
CONH2
CN
0.73
0.70
1.1
—
4a
4b
4c
4d
CF3
CH2-N-morpholinyl
CH2-N-morpholinyl
CH2-N-morpholinyl
CH2-N-morpholinyl
0.064
0.079
0.042
0.093
14
ndsdd
143
90
—
—
3.3
4.0
56
94
99
76
4.23
2.00
1.93
2.83
SO2Me
CONH2
CN
1.7
1.5
4.1
137
5a
5b
CF3
SO2Me
CH2-2-Me-imidazol-1-yl
CH2-2-Me-imidazol-1-yl
0.025
0.073
2.7
0.79
70
11
—
—
56
91
4.32
2.09
6b
6c
SO2Me
CONH2
CH2-2-pyrrolidone-1-yl
CH2-2-pyrrolidone-1-yl
0.24
0.11
1.6
2.8
20
8
—
—
96
103
1.43
1.37
a
All compounds were purified by either reverse-phase HPLC or preparative LC/MS (water/acetonitrile or water/methanol gradient + 0.5% TFA). The basic compounds were
isolated as TFA salts following lyophilization. All compounds gave satisfactory spectral and analytical data.
b
Human purified enzymes were used. Values are averages from multiple determinations (n P 2). Ki values and PT EC2Â (the concentration of the inhibitor that doubles the
clotting time compared to the control in the prothrombin time assay) values were measured as described in Refs. 3 and 6. Same for all the tables in this publication.
c
Caco-2 Papp values were measured according to Ref. 7.
Not detected in sample or donor.
Polar surface area (PSA) and clogP (pH 7.4) values were calculated using the ADME profiler in SMART prototype.
d
e
bearing a CONH2 or SO2Me group had lower Caco-2 permeability
than those bearing a CN group, which in turn had lower Caco-2
permeability than those bearing a CF3 group.
Continuing the search for diverse P4 groups,8 we screened a
variety of -substituents (R groups) beyond the initial leads
(R = CH2NR1R2)5 using CF3 as the C-3 substituent (see Tables 2–4).
a

464
J. X. Qiao et al. / Bioorg. Med. Chem. Lett. 19 (2009) 462–468
Table 2
Phenylcyclopropylamines: The less basic phenylcyclopropyl
amine analogs 8a and 9a (Table 2) maintained potency (FXa Ki)
and slightly increased Caco-2 values, but showed decreased antico-
agulant activity (PT EC2Â) compared with the phenylcyclopropyl
methylamine analogs 1a and 2a. The dimethylamine 9a was
slightly more potent than the methylamine 8a, which was slightly
more potent than the primary amine 7a. Compared with the parent
methylamine 8a, the less basic compounds 10a and 11a showed
similar binding affinity, slightly decreased anticoagulant activity,
yet improved cell permeability. On the other hand, the neutral
amide 15a and sulfonamide 16a as well as the lactams 17a and
18a, were significantly less active than 8a, presumably due to the
steric conflict between the CO or SO2 group and the backbone of
E97 in the S4 pocket of FXa.
Homologous R groups: Table 3 shows that the methylene spacer
has little effect on the binding activity (FXa Ki) within a series of
homologues, for example, 9a versus 2a versus 20a. In general, com-
pounds with a trisubstituted amine (R = (CH2)1–2NR1R1) had better
anticoagulant activity (PT EC2Â) compared with the corresponding
compounds with a di-substituted amine (R = (CH2)1–2NHR1) with
the exception of the phenylcyclopropyl amine analogs 8a and 9a.
Heterocyclic R groups: Table 4 shows FXa Ki, PT EC2Â, and Caco-2
values for several compounds bearing five-membered heterocyclic
R groups as potential COOMe or COOH isosteres. Compared with
the methyl ester 23a, the basic cyclic amidine derivatives 24a
and 25a were highly potent (FXa Ki = 21 and 155 pM, and PT
SAR of phenylcyclopropylamines with CF3 as the C-3 substituent
F3C
N
R
N
N
O
OMe
Compound
R
FXa Ki
(nM)
PT EC2x
M)
Caco-2 Papp
(nm/s)
(
l
7a
8a
9a
NH2
NHMe
NMe2
0.78
0.16
0.085
0.12
0.14
5.1
3.0
11
1.5
2.9
6.4
4.4
9.2
15
7.1
—
—
—
11
—
—
—
nd
49
92
113
180
—
10a
11a
12a
13a
14a
15a
16a
17a
18a
N(Me)CH2CH2OH
N(Me)CH2CONH2
NHCOMe
NHSO2Me
NHCOOMe
—
—
N(Me)COMe
N(Me)SO2Me
N-Pyrrolidine-2-one
N-Piperidine-2-one
—
—
—
7.6
20
Table 3
SAR of homologous R groups with CF3 as the C-3 substituent
F3C
EC2Â = 0.69 and 1.76 lM, respectively), but neither had good per-
N
R
N
meability. The cyclic amidinyl sulfonamide 26a showed a 200-fold
drop in FXa potency compared with the unsubstituted 24a. The
neutral oxazoline derivative 27a was much less potent than 24a,
but had greater Caco-2 permeability. Dimethyl substitution on
the oxazoline ring as shown in 28a did not significantly improve
FXa potency. Aromatization of cyclic amidines 24a and 25a re-
sulted in much decreased FXa potency as shown in the imidazole
derivatives 29a and 30a.
Using a strategy similar to that demonstrated in Table 1, we re-
placed the CF3 group in analogs having FXa Ki less than 1 nM with a
more polar C-3 substitution, such as CONH2 or CN. As exemplified
in Table 5, the resulting compounds showed similarly high FXa
inhibitory activity and improved anticoagulant activity in human
plasma for CONH2 and CN analogs compared with the correspond-
ing CF3 analogs (e.g., 8b and 8c vs 8a, 9b and 9c vs 9a, 22b and 22c
vs 22a), suggesting lower protein binding for CONH2 and CN ana-
logs as previously observed in other series.4 In particular, replacing
the CF3 group in 31a and 33a with a CONH2 group in 31b and 33b
led to >fivefold improvement of anticoagulant activity.
N
O
OMe
Compound
R
FXa Ki (nM)
PT EC2Â (lM)
8a
1a
19a
NHMe
0.16
0.18
0.55
4.4
2.9
12
CH2NHMe
CH2CH2NHMe
9a
2a
20a
NMe2
0.085
0.035
0.020
9.2
1.3
2.7
CH2NMe2
CH2CH2NMe2
3a
21a
CH2N-pyrrolidinyl
CH2CH2N-pyrrolidinyl
0.024
0.020
1.4
2.3
4a
22a
CH2N-morpholinyl
CH2CH2N-morpholinyl
0.064
0.21
14
15
Table 4
By choosing the proper combination of R groups and C-3 groups,
we identified multiple compounds which maintained high FXa
binding affinity (FXa Ki < 1 nM) and improved anticoagulant activ-
SAR of heterocyclic R groups in the phenylcyclopropyl analogs with CF3 as the C-3
substituent
F3C
ity (PT EC2Â < 5 lM), while keeping Caco-2 permeability in a rea-
N
R
N
sonable range. Selection of compounds for dog PK studies was
based on the aforementioned activity and liability profile. Table 6
illustrates the pharmacokinetic profile of some representative
compounds.
N
O
The basic analogs with R = CH2NR1R2 such as CH2NMe2 and
CH2-N-pyrrolidine had large Vdss and moderate systematic CL in
dogs as shown in compounds 2d, 3b, 3c, and 3d. Compounds 2d,
3b, and 3d showed good bioavailability (F = 40–100%). Compound
3d with a CN C-3 group and a CH2-N-pyrrolidinyl phenylcyclopro-
pyl P4 group showed a PK profile similar to that of raxazaban.
Reducing the basicity of 3b by changing the pyrrolidine into a mor-
pholine led to reduced Vdss (7.7 in 3b vs 1.9 L/kg in 4b) and de-
creased CL (3.3 in 3b vs 0.8 L/h/kg in 4b). On the other hand,
compound 5b containing a 2-methylimidazole R group, had negli-
gible oral bioavailability and high clearance in both dogs and short
half-lives in liver microsomal incubation presumably caused by
OMe
Compound
R
FXa Ki PT EC2x Caco-2 Papp
(nM)
(
lM)
(nm/s)
23a
24a
25a
26a
COOMe
1.5
>20
ndsd
4,5-Dihydro-1H-imidazol-2-yl
4,5-Dihydro-1H-imidazol-1-methyl-2-yl 0.16
0.021 0.69
1
1.8
—
7
—
4,5-Dihydro-1H-imidazol-1-
methanesulfonyl-2-yl
4,5-Dihydro-oxazol-2-yl
4,5-Dihydro-4,4-diMe-oxazol-2-yl
1H-Imidazol-2-yl
4.9
27a
28a
29a
30a
2.5
1.1
1.1
6.4
>20
38
25
—
145
44
57
—
1H-Imidazol-1-methyl-2-yl

J. X. Qiao et al. / Bioorg. Med. Chem. Lett. 19 (2009) 462–468
465
Table 5
SAR of example compounds with different C-3 substitution
C3
N
R
N
N
O
OMe
Compound
R
C3
FXa Ki (nM)
PT EC2Â
(l
M)
Caco-2 Papp (nm/s)
8a
8b
8c
NHMe
NHMe
NHMe
CF3
CONH2
CN
0.16
0.065
0.31
4.4
0.60
2.0
49
39
110
9a
9b
9c
NMe2
NMe2
NMe2
CF3
CONH2
CN
0.085
0.043
0.084
9.2
1.1
1.9
92
125
170
22a
22b
22c
CH2CH2N-morpholinyl
CH2CH2N-morpholinyl
CH2CH2N-morpholinyl
CF3
CONH2
CN
0.21
0.14
0.74
15
2.8
4.0
—
45
—
27a
27b
4.5-Dihydro-oxazol-2-yl
4.5-Dihydro-oxazol-2-yl
CF3
CONH2
2.5
0.56
>20
13
145
62
31a
31b
COMe
COMe
CF3
CONH2
1.0
0.18
36
3.2
ndsd
227
32a
32b
32c
CH2SO2Me
CH2SO2Me
CH2SO2Me
CF3
CONH2
CN
0.32
0.071
0.29
11
2.5
5.5
—
145
152
33a
33b
CH2CONH2
CH2CONH2
CF3
CONH2
0.59
0.36
15
1.9
—
28
Table 6
Dog pharmacokinetic profiles of example
a
-substituted phenylcycloalkyl analogsa
C3
N
R
N
N
O
OMe
Compound
R
C3
FXa Ki
PT
HLM/DLM
Caco-2 Papp
CL
Vdss
t
F
½
(nM)
EC2Â lM)
t
(min)
(nm/s)
(L/h/kg)
(L/kg)
(po) (h)
(%)
½
2d
3b
3c
3d
4b
5b
6c
8a
CH2NMe2
CN
0.048
0.054
0.074
0.027
0.079
0.073
0.11
0.79
0.73
0.70
1.1
1.7
0.79
2.8
4.4
0.60
1.4
1.9
1.9
178/73
37/—
>200/60
>200/—
25/—
46
10
6
72
143
11
8
49
39
24
28
56
9
1.37
3.3
1.7
1.1
0.8
11.7
7.7
7.1
6.6
1.9
6.4
2.3
4
5.0
4.0
na
41
100
6
CH2-N-pyrrolidinyl
CH2-N-pyrrolidinyl
CH2-N-pyrrolidinyl
CH2-N-morpholinyl
CH2-2-Me-imidazol-1-yl
CH2-2-pyrrolidone-1-yl
NHMe
NHMe
N(Me)CH2CH2OH
CH2CONH2
SO2Me
CONH2
CN
SO2CH3
SO2CH3
CONH2
CF3
CONH2
CONH2
CONH2
—
48
61
0.2
3.1
100
15
67
27
84
58
5/—
4.2
4.0
10/11
123/83
146/36
47/180
97/99
—
0.25
0.46
1.8
1.2
0.49
1.1
0.20
13.9
4.4
4.1
0.65
5.3
0.99
24
0.16
8b
10b
33b
0.065
0.068
0.36
0.19
0.08
1.6
4.3
1.9
3.4
5.8
Raxazaban3
Apixaban4
—
—
—
3.8
0.02
0.2
a
Compounds were dosed as the TFA salts in an N-in-1 format at 0.5 mg/kg iv and 0.2 mg/kg po (n = 2).
extensive first pass metabolism. The metabolism for the phenylcy-
clopropyl methylamine derivatives (R = CH2NR1R2) mainly consists
of a-carbon oxidation of the cyclopropyl methylamines to form the
corresponding aldehydes/carboxylic acids. In addition, demethyla-
tion products were also formed when either R1 or R2 was a methyl
group.
The neutral pyrrolidinone 6c had very low Vdss (0.20 L/kg and
low CL (0.25 L/h/kg) compared with the corresponding basic pyrro-
lidinyl analog 3c. However, 6c was unstable in liver microsomal
incubations (HLM t1/2 = 10 min). The difference between the
in vitro intrinsic clearance and the whole body systemic clearance
may be due to protein binding or unknown in vitro clearance
mechanisms. The low oral bioavailability of 6c is presumably be-
cause of its relatively poor absorption and low Caco-2 value.
The cyclopropylamine analog 8a showed moderate CL (0.46 L/
kg/h) and high Vdss (13.9 L/kg) in dogs. The compound was 100%
bioavailable and showed a longer t1/2 (24 h) than the rest of the
compounds because of the combined effects of slightly reduced

466
J. X. Qiao et al. / Bioorg. Med. Chem. Lett. 19 (2009) 462–468
Table 7
(Table 6). Overall, compounds in this series generally had higher
Vdss and CL than apixaban, other than the neutral compounds 6c
and 33b, which were poorly bioavailable in dogs.
Representative compounds 2d, 3b, 3d, and 5b, when dosed
intravenously in the rabbit arterio-venous (A-V) shunt thrombosis
model,9 demonstrated concentration-dependent antithrombotic
effects. The ID50 values (doses that produce 50% inhibition of
thrombus formation) for 2d, 3b, 3d, and 5b were 0.27, 0.08, 0.19,
Anticoagulant activity in rabbits
Compound
PT EC2Â
(rabbit) (
Rabbit A-V shunt
EC50 (nM)
Rabbit A-V shunt
ID50 (lmol/kg/h)
lM)
2d
3b
3d
5b
<2.5
0.32
<2.5
0.75
1.9
47
29
68
81
340
325
0.27
0.08
0.19
0.16
1.6
Raxazaban3
Apixaban2e,4
and 0.16 lmol/kg/h, respectively, and the IC50 values were 47,
2.3
0.57
29, 68, 81 nM, respectively. Consistent with the high in vitro po-
tency (FXa Ki and PT EC2x),10 the four compounds provided strong
antithrombotic effects, and in fact, they were more efficacious than
razaxaban and apixaban in the A-V shunt model. The plasma pro-
tein binding for 3d was 68% in human serum and 72% in rabbit ser-
um, lower than that of raxazaban (93% in rabbit) and apixaban
(87% in rabbit) (Table 7).
CL and increased Vdss. Changing the CF3 in 8a to a CONH2 led to 8b
with reduced Vdss but shorter t1/2 and lower oral bioavailability.
The major metabolites for 8a and 8b were N-demethylated prod-
ucts. Being less basic, compound 10b with a N(Me)CH2CH2OH
group showed increased oral bioavailability, slightly decreased Vdss
and CL, and dramatically increased t1/2 in DLM incubation com-
pared with 8b. The neutral analog 33b with a primary acetamide
had relatively low CL (0.49 L/h/kg) and low Vdss (0.65 L/kg) com-
pared with other compounds bearing the same CONH2 C-3 group
in this series.
Scheme 1 illustrates the synthesis of compound 2b bearing an
a-CH2NMe2-substituted phenylcyclopropyl P4 group and a SO2Me
C-3 group by using procedures similar to those for the CF3 analogs
previously disclosed.5
Scheme 2 depicts the preparation of phenylcyclopropyl amine
derivatives using compounds 8b, 9b, 9c, and 10b as examples. Cur-
tius rearrangement of the carboxylic acid 34 with DPPA followed
by heating in t-BuOH and then methylation gave the Boc-protected
Some compounds were stable in human liver microsomal incu-
bation (e.g., 2d, 3c, and 8b), but had a short t1/2 in dog liver micro-
somal incubation, suggesting species differences in metabolism
MeO2S
MeO2S
Cl
SO2Me
NH
N
COOH
N
N
N
HN
b
a
N
N
NH
+
O
O
Cl
34
O
OMe
OMe
OMe
MeO2S
MeO2S
Me
N
Me
N
OH
N
N
N
d,e
c
N
N
O
O
2b
OMe
OMe
Scheme 1. Reagents and conditions: (a) Et3N, toluene, 85 °C, 15 h, 31%; (b) 1-(4-Iodophenyl)cyclopropanecarboxylic acid (34), K2CO3, CuI, 1,10-phenanthroline, 120 °C, 1day,
90%; (c) ClCOOEt, Et3N, THF, 0 °C, 20 min; then NaBH4, THF/MeOH (5:1), 0 °C, 20 min, 85%; (d) PCC, NaOAc, 4 Å MS, CH2Cl2, rt, 2 h, 90%; (e) NHMe2, NaBH(OAc)3, cat. HOAc,
ClCH2CH2Cl, rt, 2 h, 28% for two steps.
EtOOC
Boc
H2NOC
EtOOC
NH
N
NH
N
N
Boc
COOH
N
d,e
a,b
N
N
c
N
N
N
N
I
I
O
+
O
O
34
35
8b
36
37
OMe
OMe
OMe
f
h
NC
H2NOC
H2NOC
N
N
OH
N
N
N
N
N
g
N
N
N
N
O
N
O
O
9c
9b
10b
OMe
OMe
OMe
Scheme 2. Reagents and conditions: (a) DPPA, Et3N, rt, overnight, then t-BuOH, reflux, 3 h, 65%; (b) MeI, NaH, THF, rt, overnight, 95%; (c) K2CO3, CuI, 1,10-phenanthroline,
120 °C, 2 h, 82%; (d) NH3 in ethylene glycol, 80 °C, 3 h,%; (e) TFA, CH2Cl2, rt, 2 h, 98%; (f) Aqueous formaldehyde, NaBH3CN, HOAc, CH3CN, rt, 2 h, 92%; (g) SOCl2, DMF, CH2Cl2,
90%; (h) BrCH2CH2OH, K2CO3, DMF, 70 °C, 2 h, 65%.

J. X. Qiao et al. / Bioorg. Med. Chem. Lett. 19 (2009) 462–468
467
F3C
NH
N
COOH
CH2COO-t-Bu
a, b
N
c
O
I
I
+
39
34
38
OMe
O-t-Bu
F3C
F3C
O
N
N
N
O
N
d, e, f, g
N
N
N
O
21a
OMe
OMe
Scheme 3. Reagents and conditions: (a) (COCl)2, CH2Cl2, DMF, rt, 2 h; (b) TMSCHN2, CH3CN/THF (1:1), 0 °C to rt, 4 h; then t-BuOH, AgOBz, Et3N, reflux, 1 h, 35% for two steps;
(c) K2CO3, CuI, 1,10-phenanthroline, 120 °C, 3 h, 67%; (d) TFA, CH2Cl2, rt, 2 h, 87%; (e) ClCOOEt, Et3N, THF, 0 °C, 20 min; then NaBH4, THF/MeOH (5:1), 0 °C, 20 min; (f) PCC,
NaOAc, 4 Å MS, CH2Cl2, rt, 2 h; (g) Pyrrolidine, NaBH(OAc)3, cat. HOAc, ClCH2CH2Cl, rt, 3 h, 35% for three steps.
O
S
N
N
b
COOH
a
c
I
I
I
41
40
34
F3C
F3C
N
N
N
N
N
N
N
N
N
N
e
N
N
d
O
O
I
39
30a
42
25a
OMe
OMe
F3C
O
O
N
N
N
N
COOH
f,g
h
N
I
I
O
39
27a
43
OMe
Scheme 4. Reagents and conditions: (a) pentafluorophenol, DCC, CH2Cl2, rt, 2 h; then piperidine, overnight, 88%; (b) Lawesson’s reagent, toluene, 110 °C, 1.5 h, 81%; (c) MeI,
then NHMeCH2CH2NH2, MeOH, 2 days, 65%; (d) 39, K2CO3, CuI, 1,10-phenanthroline, 120 °C, overnight, 38%.; (e) KMnO4, 1,4-dioxane, 80 °C, 12 h, 72%; (f) (COCl)2, CH2Cl2, 0 °C,
1 h; then ethanolamine, rt, 1.5 h, 73%; (g) Burgess reagent, THF, 70 °C, 2 h, 74%; (h) 39, K2CO3, CuI, 1,10-phenanthroline, 120 °C, overnight, 50%.
cyclopropylamine 35, which was coupled with the lactam 364 un-
der Buchwald–Ullmann condition to afford intermediate 37. Ami-
nation of the ethylester in 37 followed by removing Boc group
afforded the methylamine 8b. Starting from 8b, compounds 9b,
9c, and 10b were synthesized as illustrated in Scheme 2.
Scheme 3 illustrates the synthesis of cyclopropyl-ethylamine
analogs using 21a as an example. Homologation of acid 34 with
TMSCHN2 followed by heating in t-BuOH in the presence of silver
benzoate and triethylamine afforded the Boc-protected homolo-
gous acid 38. Ullmann coupling of 38 with lactam 39 followed by
a similar sequence of transformations as described in Scheme 1
generated compound 21a.
obtained from the acid via pentafluorophenol/DCC, with piperidine
afforded the amide 40. Formation of the thioamide 41 using Lawes-
son’s reagent, then thioimidate via MeI, followed by cyclization of
the thioimidate intermediate with ethylamine diamine in refluxing
MeOH gave the methyl imidazoline intermediate 42. Buchwald
Ullmann coupling of 42 and lactam 39 yielded the imidazoline
25a. Aromatization of 25a with potassium permanganate in diox-
ane afforded the imidazole analog 30a. On the other hand, cycliza-
tion of the b-hydroxyamide with Burgess reagent in THF at reflux
afforded the oxazoline intermediate 43 in 74% yield. Utilizing
Buchwald–Ullmann chemistry, 43 was coupled with the lactam
39 affording the desired product 27a.
Scheme 4 illustrates the synthesis of phenylcyclopropyl analogs
bearing heterocyclic R groups. N-acylation of the pentafluoroester,
SAR studies of the C-3 substitution on the pyrazolodihydropyri-
done core and the R substitution on the phenylcyclopropyl group

468
J. X. Qiao et al. / Bioorg. Med. Chem. Lett. 19 (2009) 462–468
Schlemmer, K-H.; Reinemer, P.; Perzborn, E.; Roehrig, S. J. Med. Chem. 2005, 48,
5900; (c) Eriksson, B. L.; Borris, L.; Dahl, O. E.; Haas, S.; Huisman, M. V.; Kakkar,
A. K. J. Thromb. Haemost. 2006, 4, 121; (d) Hampton, T. JAMA 2006, 295, 743; (e)
Lassen, M. R.; Davidson, B. L.; Gallus, A.; Pineo, G.; Ansell, J.; Deitchman, D.
Thromb. Haemost. 2007, 5, 2368; (f) Wong, P. C.; Crain, E. J.; Watson, C. A.; Xin,
B.; Wexler, R. R.; Lam, P. S. Y.; Pinto, D. J.; Luettgen, J. M.; Knabb, R. M. J. Thromb.
Haemost. 2008, 6, 820–829; (g) Turpie, A. G. G. Arterioscler. Thromb. Vasc. Biol.
2007, 27, 1238. and references therein.
of a series of pyrazole bicyclics bearing
ylcycloalkyl groups resulted in a series of FXa inhibitors with high
FXa binding affinity (FXa Ki of some in the picomolar range) and
high anticoagulant activity (PT EC2Â < 3 lM) in vitro. C-3 substitu-
tion of the pyrazole core with more polar groups such as SO2Me,
CONH2, and CN, maintained FXa Ki potency and increased anticoag-
ulant PT potency compared with the CF3 C-3 analogs, suggestive of
improved free fraction in protein binding. A set of compounds, ob-
tained from optimization of the R group and the C-3 substituent,
were orally bioavailable in dog PK studies. Representative com-
a-substituted phen-
3. (a) Quan, M. L.; Lam, P. Y. S.; Han, Q.; Pinto, D. J. P.; He, M. Y.; Li, R.; Ellis, C. D.;
Clark, C. G.; Teleha, C. A.; Sun, J-H.; Alexander, R. S.; Bai, S.; Luettgen, J. M.;
Knabb, R. M.; Wong, P. C.; Wexler, R. R. J. Med. Chem. 2005, 48, 1729; (b) Wong,
P. C.; Crain, E. J.; Watson, C. A.; Wexler, R. R.; Lam, P. Y. S.; Quan, M.; Knabb, R.
M. J. Thromb. Thrombol. 2007, 24, 43.
4. Pinto, D. J. P.; Orwat, M. J.; Koch, S.; Rossi, K. A.; Alexander, R. S.; Smallwood, A.;
Wong, P. C.; Rendina, R. A.; Luettgen, J. M.; Knabb, R. M.; He, K.; Xin, B.; Wexler,
R. R.; Lam, P. Y. S. J. Med. Chem. 2007, 50, 5339.
5. Qiao, J. X.; Cheney, D. L.; Alexander, R. S.; King, S. R.; Rendina, A. R.; Luettgen, J.
M.; Knabb, R. M.; Wexler, R. R.; Lam, P. Y. S. Bioorg. Med. Chem. Lett. 2008, 18,
4118.
6. (a) Knabb, R. M.; Kettner, C. A.; Timmermans, P. B. M. W. M.; Reilly, T. M.
Thromb. Haemost. 1992, 67, 56; (b) Kettner, C. A.; Mersinger, L. J.; Knabb, R. M. J.
Biol. Chem. 1990, 265, 18289.
7. Hilgers, A. R.; Conradi, R. A.; Burton, S. Pharm. Res. 1990, 7, 902–910.
8. Qiao, J. X.; Cheng, X.; Smallheer, J. M.; Galemmo, R. A.; Spencer, D.; Pinto, D. J.
P.; Cheney, D. L.; He, K.; Wong, P. C.; Luettgen, J. M.; Knabb, R. M.; Wexler, R. R.;
Lam, P. Y. S. Bioorg. Med. Chem. Lett. 2007, 17, 1432.
9. (a) Wong, P. C.; Quan, M. L.; Crain, E. J.; Watson, C. A.; Wexler, R. R.; Knabb, R.
M. J. Pharmacol. Exp. Ther. 2000, 292, 351; (b) Wong, P. C.; Crain, E. J.; Knabb, R.
M.; Meade, R. P.; Quan, M. L.; Watson, C. A.; Wexler, R. R.; Wright, M. R.; Slee, A.
M. J. Pharmacol. Exp. Ther. 2000, 295, 212; (c) Wong, P. C.; Watson, C. A.; Crain,
E. J.; Zaspel, A. M.; Luettgen, J. M.; Bai, S. M.; Lam, P. Y.; Quan, M. L.; Wexler, R.
R.; Knabb, R. M. Blood 2003, 102, 813a.
pounds such as 3d and 8a showed moderate clearance, long t1/2
,
and good bioavailability, but high Vdss in dogs. Compounds 2d,
3b, 3d, and 5b in this series were also highly efficacious in the rab-
bit A-V shunt thrombosis model (EC50 < 85 nM). Though highly
efficacious, the pharmacokinetic properties of this series were less
optimal compared with that of apixaban, which has a superior
combination of exceptionally low CL and Vdss with a long half life
and good oral bioavailability.
Acknowledgments
The authors thank Dr. Kiho Lee for Caco-2 permeability and Jef-
frey M. Bozarth, Frank A. Barbera, Tracy A. Bozarth, and Karen S.
Hartl for in vitro assays.
10. (a) Compounds 2d, 3b, 3d and others in this series also had favorable, rapid
rates of association (on-rates) comparable to razaxaban, apixaban, and other
FXa inhibitors, near the diffusion controlled limit. Rapid onset of inhibition is
preferred and observed for coagulation.; (b) Elg, M.; Gustafsson, D.; Denium, J.
J. Thromb. Haemost. 1997, 78, 1286; (c) Rezaie, A. R. Thromb. Haemost. 2003, 89,
112; (d) Gould, W. R.; Cladera, E.; Harris, M. S.; Zhang, E.; Narasimhan, L.;
Thorn, J. M.; Leadley, R. L. Biochemistry 2005, 44, 9280; (e) Pinto, D. J. P.; Orwat,
M. J.; Quan, M. L.; Han, Q.; Galemmo, R. A.; Amparo, E.; Wells, B.; Smallwood,
A.; Wong, P. C.; Luettgen, J. M.; Rendina, A. R.; Knabb, R. M.; Mersinger, L.;
Kettner, C.; Bai, S.; He, K.; Wexler, R. R.; Lam, P. Y. S. Bioorg. Med. Chem. Lett.
2006, 16, 4141.
References and notes
1. For recent review papers of FXa inhibitors, see: (a) Quan, M.; Smallheer, J. Curr.
Opin. Drug Discov. Dev. 2004, 7, 460; (b) Walenga, J. M.; Jeske, W. P.;
Hoppensteadt, D.; Fareed, J. Curr. Opin. Investig. Drugs 2003, 4, 272; (c) Gould,
W. R.; Leadley, R. J. Curr. Pharm. Des. 2003, 9, 2337; (d) Quan, M. L.; Wexler, R. R.
Curr. Topics Med. Chem. 2001, 1, 137; (e) Zhu, B.; Scarborough, R. M. Ann. Rep.
Med. Chem. 2000, 35, 83.
2. (a) Lassen, M. R.; Davidson, B. L.; Gallus, A.; Pineo, G.; Ansell, J.; Deitchman, D.
Blood 2003, 102, 11; (b) Straub, A.; Pohlmann, J.; Lampe, T.; Pernerstorfer, J.;