Design, synthesis, and SAR of substituted acrylamides as factor Xa inhibitors.

Article abstract of DOI:10.1016/S0960-894X(02)00199-3

Substituted acrylamides were used as templates that bridge P1 and P4 binding elements, resulting in a series of potent (sub-nanomolar) and selective factor Xa inhibitors. In this template, cis-geometry of P1 and P4 ligands is highly preferred. SAR on the substituting groups, as well as on modification of P1 and P4 moieties is described. Compounds in this series show good in vivo efficacy in animal models.

Full text of DOI:10.1016/S0960-894X(02)00199-3

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1511–1515
Design, Synthesis, and SAR of Substituted Acrylamides
as Factor Xa Inhibitorsy
Yonghong Song,* Lane Clizbe, Chhaya Bhakta, Willy Teng, Wenhao Li, Yanhong Wu,
Zhaozhong Jon Jia, Penglie Zhang, Lingyan Wang, Brandon Doughan, Ting Su,
James Kanter, John Woolfrey, Paul Wong, Brian Huang, Katherine Tran,
Uma Sinha, Gary Park, Andrea Reed, John Malinowski, Stan Hollenbach,
Robert M. Scarborough and Bing-Yan Zhu
Millennium Pharmaceuticals, Inc., 256 East Grand Avenue, South San Francisco, CA 94080, USA
Received 13 December 2001; accepted 28 February 2002
Abstract—Substituted acrylamides were used as templates that bridge P1 and P4 binding elements, resulting in a series of potent
(sub-nanomolar) and selective factor Xa inhibitors. In this template, cis-geometry of P1 and P4 ligands is highly preferred. SAR on
the substituting groups, as well as on modification of P1 and P4 moieties is described. Compounds in this series show good in vivo
efficacy in animal models. # 2002 Elsevier Science Ltd. All rights reserved.
Thrombotic diseases are a leading cause of mortality and
morbidity.2 Improvement in the treatment and preven-
tion of thrombotic diseases remains an important medi-
cal need. Current therapeutic drugs in the area such as
warfarin and heparin suffer from major problems such as
patient variability and the need for periodic monitoring,
making these drugs undesirable. Factor Xa is a serine
protease responsible for conversion of prothrombin to
thrombin in the blood coagulation cascade. In recent
years factor Xa has emerged as an attractive target for
the development of new antithrombotic agents.3 Our
research objective is to develop factor Xa inhibitors for
treatment of thrombotic diseases.
we thought that the heterocycles could be replaced with
a substituted double bond. Therefore, we investigated
the use of substituted acrylamides as the central tem-
plate (Fig. 1). As a result, a series of potent and selective
factor Xa inhibitors were discovered. In this paper, we
will present the structure–activity relationship (SAR) of
R1 and R2 substitution on the double bond of acrylam-
ide, as well as the selected modifications of P1 and P4
moieties.
We initially synthesized the unsubstituted cinnamamide
analogues as pure Z and E isomers (Fig. 2). The Z-iso-
mer 1 showed good Xa activity (IC50=17 nM) in our
assays,9 while the E-isomer 2 is 180-fold less active,
indicating the importance of cis-geometry of P1 and P4
ligands.
The active site of factor Xa, as shown by X-ray struc-
ture,4 contains a deep S1 pocket and a box-like S4
pocket. Going from S1 to S4 pocket requires a 90-degree
turn. Most of factor Xa inhibitors in literature contain an
S1 binding element (P1) and an S4 ligand (P4), which are
connected by a central template (Fig. 1).5ꢀ8 Among these
templates, heterocycles such as pyrazole,5 have been
shown to be effective scaffolds that provide Xa inhibi-
tors of high potency as exemplified by DuPont’s SN 429
(Ki=0.013 nM). In our search for factor Xa inhibitors,
§For preliminary account of this work, see: ref. 1.
*Corresponding author. Fax: +1-650-244-9287; e-mail: yonghong.
song@mpi.com
Figure 1. Schematic representation of structure of factor Xa inhibitors.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00199-3

1512
Y. Song et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1511–1515
Table 2. SAR of R1 substitution
Figure 2. Activity of E- and Z-isomers.
R1
Compd
FXa (IC50, nM)
After identification of compound 1, we studied the
effects of substitution on the benzamidine ring. As
shown in Table 1, substitution by fluoro and methoxy
groups at 4-position (4 and 6) was more tolerant than
the same substitution at 2-position (3 and 5). Actually,
4-fluoro analogue 4 is equally potent as compound 1,
while 4-methoxy analogue 6 is 12-fold less active. Inter-
estingly, 4-methoxyethoxy analogue 7 is 2-fold more
potent than compound 1.
H
CH3
CH3CH2–
iPr
iBu
1
8
9
10
11
12
13
17
4
7
35
5
10
7
CF3
CH3OCH2–
14
48
We then set out to study the effects of substituents on
the acrylamide double bond. The SAR of R1 substitu-
tion, as shown in Table 2, indicated that alkyl groups at
this position generally increase activity (8, 9, and 11–13)
except for the isopropyl group (10). The methyl sub-
stituted analogue 8 gave 4-fold increase of activity over
the unsubstituted analogue 1. Appendages (pyrazolyl
and tetrazolyl, 14–16) on the methyl group tended to
decrease activity, and 2-furyl analogue 17 gave 11-fold
reduction in potency as compared with 1. The tri-
fluoromethyl compound 12 was 2-fold less active than
the methyl analogue 8.
15
16
17
53
245
184
Table 3. SAR of R2 substitution
The SAR of R2 substitution is shown in Table 3. Gen-
erally, substitution at this position with a variety of dif-
ferent groups (18–23) all resulted in decreased potency.
This seemed to suggest that groups larger than hydro-
gen would decrease activity. This observation led us to
try fluorine as R2, since fluorine is similar to hydrogen
in size. The fluoro compound 24 (IC50=1 nM) displays
increased potency by 4-fold over 8 (where R2=H).
R1
R2
Compd
FXa (IC50, nM)
H
H
CH3
–CO2CH3
18
19
362
759
H
H
20
21
628
664
Previous literature has shown that halogen substitution
at the biphenyl P4 ligand generally increases anti-Xa
activity.10 A similar trend was observed in our study (25–
27), as shown in Table 4, with fluorine substitution (25)
giving the best potency (IC50=0.2 nM). These com-
pounds (24–27) are also selective against other serine
proteases such as thrombin (>1000-fold) and trypsin
(>360-fold) (Table 4).
CH3
CH3
CH3
–CH2CONH2
–CH2CO2H
F
22
23
24
99
234
1
Table 1. Substitution on benzamidine ring
Table 4. Activity and selectivity
R3
Compd
FXa (IC50, nM)
H
2-F
4-F
2-CH3O–
4-CH3O–
4-CH3OCH2CH2O–
1
3
4
5
6
7
17
99
18
6470
215
9
X
Compd
FXa
Ila (IC50, nM)
Trypsin
H
F
Cl
Br
24
25
26
27
1
3190
2200
650
259
194
188
189
0.2
0.5
0.5
503

Y. Song et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1511–1515
1513
After optimization of R2 with fluorine, we set out to do
further R1 modification. As shown in Table 5, append-
ing other functional groups (28–31) to the methyl group
at R1 failed to improve potency. Some polar groups
(28–30) are moderately tolerated at this position, but
basic dimethylamino (31) is least favored. Again the
trifluoromethyl compound 32 is 3-fold less active than
the methyl analogue 24.
Molecular modeling of compound 24 into the active site
of Xa crystal structure, as shown in Figure 3, indicated
that the amidine group of 24 interacts with the carbox-
ylate of Asp189 in S1 cavity, while the biphenylsulfo-
namide moiety fits into the box-like S4 pocket defined
by Phe174, Tyr99 and Trp215. The vinyl methyl group
fits into a pocket defined by the Cys191-Cys220 disulfide
bond (yellow surface in Fig. 3). Branching at this carbon
results in a reduction of activity, as evidenced by com-
pounds 10 and 11. The fluorine atom is in close proximity
to the Gln192 side chain. It is conceivable that groups
larger than fluorine would cause unfavorable interaction
with the side chain, resulting in reduced potency, which
is consistent with the SAR seen in Table 3.
With the goal of reducing basicity of benzamidine P1
moiety, electron-withdrawing substituents (nitro, fluoro)
were placed on the benzene ring, as shown in Table 6. A
nitro group (33) at 4-position reduced potency by 75-fold
as compared with compound 25, but reduction of nitro
to amine (34) regained some activity. However, fluorine
at 2-position (35–38) was well tolerated. Again halogen
substitution on the biphenyl increases potency. Com-
pound 37 shows subnanomolar activity with Ki of
0.18 nM.
The acrylamides of this study were generally synthesized
according to Scheme 1. An aldehyde or ketone was
condensed with an acetate equivalent to give the acry-
lates, usually as a mixture of cis and trans isomers. In
most cases, the two isomers were separated at this stage.
Weinreb amidation of the acrylate with the biphenyla-
mine yielded the acrylamide, which was then converted
to the benzamidine product under Pinner conditions.
We examined the antithrombotic activity of the com-
pounds in our modified rabbit deep vein thrombosis
(DVT) model.11 For example, compound 37 was found
to inhibit thrombus formation by 40%, at an iv bolus
dose of 2 mg/kg followed by infusion to maintain an
average plasma concentration of 0.66 mM during the 2-h
treatment period (giving an average 1.5-fold increase of
PT). In comparison, the maximum inhibition caused by
low-molecular weight heparin (Enoxaparin) was usually
about 45–50% in the model.11 We also determined oral
bioavailability of the compounds in rats. However, they
were usually below 3%.
To synthesize the (Z) cis-isomers 1 and 3–7 in Table 1,
the cis-cinnamates were prepared, as shown in Scheme 2.
Horner–Emmons reaction of the cyanobenzaldehyde
with methyl bis(2,2,2-trifluoroethyl)phosphono acetate12
gave nearly exclusively the cis-cinnamate. The (E) trans-
cinnamate leading to compound 2, was synthesized by
Table 5. Further SAR of R1 substitution
R1
X
Compd
FXa (IC50, nM)
CH3
H
H
H
H
F
24
28
29
30
31
32
1
3.7
13
10
191
3
HOCH2–
EtSCH2–
CH3SO2CH2–
(CH3)2NCH2–
CF3
Figure 3. Proposed binding mode of compound 24 with factor Xa.
Connolly surface created using InsightII (Accelrys, Inc.). Red, blue,
and yellow represent the surface close to oxygen, nitrogen, and sulfur
atoms, respectively. The fluorine atom on 24 is green.
H
Table 6. Modification on benzamidine ring
R4
X
Compd
FXa (IC50, nM)
4-NO2
4-NH2
2-F
2-F
2-F
F
F
H
F
Cl
Br
33
34
35
36
37
38
15
2.6
3.6
1.5
Scheme 1. General synthesis. Reagents and conditions: (a) base (e.g.,
KN(Me3Si)2, piperidine, KOtBu); (b) AlMe3, CH2Cl2, X=H, F, Cl,
Br; (c) MeOH/HCl (g); (d) NH4OAc.
0.74 (Ki=0.18 nM)
0.75
2-F

1514
Y. Song et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1511–1515
Scheme 2. Synthesis of compounds 1–7. Reagents and conditions:
(a) KN(Me3Si)2, 18-Crown-6, THF, ꢀ78 ꢁC; (b) NaH.
Wittig reaction of 3-cyanobenzaldehyde with (carbo-
methoxymethyl)triphenyl phosphonium ylide.
Scheme 5. Synthesis of compounds in Tables 4–6. Reagents and con-
ditions: (a) triethyl 2-fluoro-2-phosphonoacetate, KN(Me3Si)2,
ꢀ78 ꢁC; (b) biphenylamine, AlMe3; (c) Nu (EtSH for 29, CH3SO2Na
for 30, (CH3)2NH for 31); (d) pyridinium tribromide, HOAc.
Compounds 8–14 and 17 in Table 2 were synthesized
according to Scheme 3. Ketoacetates were treated with
Tf2O in the presence of TEA to give the enol triflates,
which were then coupled with 3-cyanobenzeneboronic
acid under Pd(0) catalysis to yield the R1 substituted
cinnamates. The cis-geometry of R1 to the vinyl proton
was confirmed by NOE analysis.
products 28–31 by Pinner reaction. Compound 28 was
obtained as a side product in the preparation of 30.
In summary, substituted acrylamides were found to be
effective templates which gives rise to a series of potent
(sub-nanomolar anti-Xa activity) and selective mono-
benzamidino factor Xa inhibitors. Within this template,
cis-geometry of P1 and P4 ligands is highly preferred.
Compounds in this series show good in vivo efficacy in
animal models. However, these compounds suffer from
low bioavailability. Replacement of the benzamidine P1
moiety with neutral or less basic groups to achieve good
bioavailability will be the subject of additional
communications from our laboratories.
To synthesize compounds 19–23 in Table 3, the sub-
stituted acrylates were made, as shown in Scheme 4.
Knoevenagel reaction of the cyanobenzaldehyde with
dimethyl malonate (or pyridylacetate) gave the desired
acrylates as a mixture of separable E/Z isomers. Stubbe
condensation of the cyanoacetophenone with dimethyl
succinate gave the acrylate as the expected monoester,
which gave rise to final products 22 and 23.
Synthesis of compounds 24–27 in Table 4, 33–38 in
Table 6 began with Horner–Emmons reaction of the
cyanoacetophenone 39a with triethyl 2-fluoro-2-phos-
phonoacetate, giving the acrylates in favor of the desired
isomer (with the cyanophenyl cis to the carboxylate)
(Scheme 5, Y=H).
References and Notes
1. Song, Y.; Clizbe, L.; Bhakta, C.; Teng, W.; Li, W.; Wu, Y.;
Jia, Z. J.; Zhang, P.; Wang, L.; Doughan, B.; Su, T.; Kanter, J.;
Woolfrey, J.; Wong, P.; Huang, B.; Tran, K.; Sinha, U.; Park,
G.; Reed, A.; Malinowski, J.; Hollenbach, S.; Scarborough, R.
M.; Zhu, B-Y. Books of Abstracts, Part 2, 221th National Meet-
ing of the American Chemical Society, San Diego, CA, April 1–
5, 2001; American Chemical Society: Washington, DC, 2001;
MEDI 063.
Similarly, Horner–Emmons reaction of the bromoace-
tophenone 39b, which was derived from 39a by bromi-
nation with pyridinium tribromide, gave the acrylates as
allyl bromides (Scheme 5). Weinreb amidation of these
acrylates generated the acrylamides. Substitution of the
allyl bromides with nucleophiles (Nu) produced the
intermediates 40, which were then converted to the final
2. (a) Scarborough, R. M. Annu. Rep. Med. Chem. 1995, 30,
71. (b) Edmunds, J. J.; Rapundalo, S. T. Ann. Rep. Med.
Chem. 1996, 31, 51.
3. (a) Zhu, B.-Y.; Scarborough, R. M. Annu. Rep. Med.
Chem. 2000, 35, 83. (b) Zhu, B.-Y.; Scarborough, R. M. Curr.
Opin. Cardiovasc., Pulm. Renal Invest. Drugs 1999, 1, 63. (c)
Sanderson, P. E. J. Annu. Rep. Med. Chem. 2001, 36, 79.
4. (a) Brandstetter, H.; Kuhne, A.; Bode, W.; Huber, R.; von
der Saal, W.; Wirthensohn, K.; Engh, R. A. J. Biol. Chem.
1996, 271, 29988. (b) Kamata, K.; Kawamoto, H.; Honma, T.;
Iwama, T.; Kim, S. Proc. Natl. Acad. Sci. U.S.A. 1998, 95,
6630.
Scheme 3. Synthesis of compounds in Table 2. Reagents and con-
ditions: (a) Tf2O, Et3N, CH2Cl2, ꢀ78 ꢁC; (b) 3-cyanobenzeneboronic
acid, Pd(Ph3P)4, dioxane, K3PO4, reflux.
5. Pinto, D. J. P.; Orwat, M. J.; Wang, S.; Fevig, J. M.; Quan,
M. L.; Amparo, E.; Cacciola, J.; Rossi, K. A.; Alexander,
R. S.; Smallwood, A. M.; Luettgen, J. M.; Liang, L.; Aungst,
B. J.; Wright, M. R.; Knabb, R. M.; Wong, P. C.; Wexler,
R. R.; Lam, P. Y. S. J. Med. Chem. 2001, 44, 566.
6. Masters, J. J.; Franciskovich, J. B.; Tinsley, J. M.; Camp-
bell, C.; Campbell, J. B.; Craft, T. J.; Froelich, L. L.; Gifford-
Moore, D. S.; Hay, L. A.; Herron, D. K.; Klimkowski, V. J.;
Kurz, K. D.; Metz, J. T.; Ratz, A. M.; Shuman, R. T.; Smith,
Scheme 4. Synthesis of compounds in Table 3. Reagents and con-
ditions: (a) base (piperidine, or KOtBu).

Y. Song et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1511–1515
1515
G. F.; Smith, T.; Towner, R. D.; Wiley, M. R.; Wilson, A.;
Yee, Y. K. J. Med. Chem. 2000, 43, 2087.
10. Han, Q.; Dominguez, C.; Stouten, P. F. W.; Park, J. M.;
Duffy, D. E.; Galemmo, R. A. Jr.; Rossi, K. A.; Alexander,
R. S.; Smallwood, A. M.; Wong, P. C.; Wright, M. M.;
Luettgen, J. M.; Knabb, R. M.; Wexler, R. R. J. Med. Chem.
2000, 43, 4398.
7. Phillips, G. B.; Buckman, B. O.; Davey, D. D.; Eagen, K. A.;
Guilford, W. J.; Hinchman, J.; Ho, E.; Koovakkat, S.; Liang, A.;
Light, D. R.; Mohan, R.; Ng, H. P.; Post, J. M.; Shaw, K. J.;
Smith, D.; Subramanyam, B.; Sullivan, M. E.; Trinh, L.; Ver-
gona, R.; Walters, J.; White, K.; Whitlow, M.; Wu, S.; Xu, W.;
Morrissey, M. M. J. Med. Chem. 1998, 41, 3557.
8. Ewing, W. R.; Becker, M. R.; Manetta, V. E.; Davis, R. S.;
Pauls, H. W.; Mason, H.; Choi-Sledeski, Y. M.; Green, D.;
Cha, D.; Spada, A. P.; Cheney, D. L.; Mason, J. S.; Maignan,
S.; Guilloteau, J.-P.; Brown, K.; Colussi, D.; Bently, R.;
Bostwick, J.; Kasiewski, C. J.; Morgan, S. R.; Leadley, R. J.;
Dunwiddie, C. T.; Perrone, M. H.; Chu, V. J. Med. Chem.
1999, 42, 3557.
9. The bioassays were performed as described in reference:
Sinha, U.; Ku, P.; Malinowski, J.; Zhu, B. Y.; Scarborough,
R. M.; Marlowe, C. K.; Wong, P. W.; Lin, P. H.; Hollenbach,
S. J. Eur. J. Pharmacol. 2000, 395, 51 Measurements are aver-
age of triplicates with standard deviation of ꢂ20%.
11. In our modified model, a cotton thread attached to a
copper wire was introduced into the abdominal vena cava via
the femoral vein of an anesthetized rabbit. At the same time as
the introduction of the cotton thread, compounds were admi-
nistered intravenously by a bolus dose followed by infusion at
a constant rate to maintain the targeted plasma concentration
during the 2-h treatment period. Inhibition of thrombus for-
mation was determined by weight of harvested thrombi
formed on the cotton thread. In the positive control, Enox-
aparin was administered at a dose of 200 U/kg followed by an
infusion of 2 U/kg/min, attaining about 45–50% inhibition.
For our original model, see: Hollenbach, S.; Sinha, U.; Lin,
P. H.; Needham, K.; Frey, L.; Hancock, T. E.; Wong, A.;
Wolf, D. L. Thromb. Haemost. 1994, 71, 357.
12. Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.