Discovery of nonbenzamidine factor VIIa inhibitors using a biaryl acid scaffold

Article abstract of DOI:10.1016/j.bmcl.2013.06.028

In this Letter, we describe the synthesis of several nonamidine analogs of biaryl acid factor VIIa inhibitor 1 containing weakly basic or nonbasic P1 groups. 2-Aminoisoquinoline was found to be an excellent surrogate for the benzamidine group (compound 2) wherein potent inhibition of factor VIIa is maintained relative to most other related serine proteases. In an unanticipated result, the m-benzamide P1 (compounds 21a and 21b) proved to be a viable benzamidine replacement, albeit with a 20-40 fold loss in potency against factor VIIa.

Full text of DOI:10.1016/j.bmcl.2013.06.028

Bioorganic & Medicinal Chemistry Letters 23 (2013) 5239–5243
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of nonbenzamidine factor VIIa inhibitors using a biaryl
acid scaffold
e,
c
f
d
Scott A. Bolton a, , James C. Sutton , Rushith Anumula , Gregory S. Bisacchi , Bruce Jacobson ,
William A. Slusarchyk b, Uwe D. Treuner a, Shung C. Wu a, Guohua Zhao a, Zulan Pi a, Steven Sheriff b,
Rebecca A. Smirk a, Sharon Bisaha a, Daniel L. Cheney a, Anzhi Wei b, William A. Schumacher a,
Karen S. Hartl a, Eddie Liu a, Robert Zahler g, Steven M. Seiler b
⇑
⇑
a Bristol-Myers Squibb Research & Development, P.O. Box 5400, Hopewell, NJ 08534-5400, USA
b Bristol-Myers Squibb Research & Development, P.O. Box 4000, Princeton, NJ 08543-4000, USA
c Bristol-Myers Squibb Biocon Research Center, Hosur Road, Electronics City, Bangalore 560 100, India
d St. Cloud State University, Department of Biology, 720 Fourth Avenue South, St. Cloud, MN 56301-4498, USA
e Novartis Institute of Biomedical Research, 4560 Horton St., Emeryville, CA 94608, USA
f AstraZeneca R&D Boston, 35 Gatehouse Drive, Waltham, MA 02451, USA
g PharmD Consulting, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
In this Letter, we describe the synthesis of several nonamidine analogs of biaryl acid factor VIIa inhibitor 1
containing weakly basic or nonbasic P1 groups. 2-Aminoisoquinoline was found to be an excellent sur-
rogate for the benzamidine group (compound 2) wherein potent inhibition of factor VIIa is maintained
relative to most other related serine proteases. In an unanticipated result, the m-benzamide P1 (com-
pounds 21a and 21b) proved to be a viable benzamidine replacement, albeit with a 20–40 fold loss in
potency against factor VIIa.
Available online 20 June 2013
Dedicated to the memory of Steven M.
Seiler, Ph.D. Steve’s untimely passing will
not diminish the impact of his drug
discovery research or the admiration of his
fellow scientists
Ó 2013 Published by Elsevier Ltd.
Keywords:
Factor VIIa
2-Aminoisoquinoline
Inhibition
Thromboembolic cardiovascular disease remains
a
leading
or in post-operative settings. Exposure of TF to blood leads
to the formation of the catalytically active TF/FVIIa complex.
TF/FVIIa then initiates coagulation by activation of factor
X–Xa, both directly by proteolytic action on factor X and indi-
rectly by converting factor IX–IXa which binds to surface
bound VIIIa to form the tenase complex, which in turn acti-
vates factor X.4
Factor VIIa alone is a weakly active trypsin-like serine protease,
and becomes fully competent in the TF/FVIIa complex. Like other
trypsin-like serine proteases, factor VIIa has a conserved aspartic
acid in the bottom of the S1 binding pocket.5 At the time of our en-
try into the field, the bulk of reported small-molecule factor VIIa
inhibitors were peptidic in origin and obtained potency through
the use of strongly basic guanidine or amidine P1 groups, features
that hamper oral bioavailability. A notable nonpeptidic exception
is compound 1, reported by Ono Pharmaceuticals.6 As part of our
interest in the development of selective factor VIIa inhibitors, we
have focused extensively on identifying weakly basic or nonbasic
P1 groups applicableacross multiple chemotypes. In this Letter,
we report the synthesis of nonamidine analogs of 1, leading to
cause of hospitalization and mortality in developed countries.1
Existing anticoagulant and antiplatelet agents are effective in
treating thromboembolic disorders; however, associated bleeding
side effects, narrow therapeutic indices, and limitations in route
of administration continue to highlight the need for agents with
an improved balance in efficacy, safety and convenience.2 A current
area of inquiry is the development of selective inhibitors of the
tissue factor/factor VIIa complex (TF/FVIIa), a key initiator of coag-
ulation, which may offer antithrombotic efficacy with preserved
hemostasis.3
Under normal physiological conditions, membrane associated
tissue factor (TF) is not exposed to blood flow, but becomes
exposed appropriately upon vascular injury. However, exposure
of TF can be significantly increased under pathological condi-
tions, such as through the rupture of atherosclerotic plaque
⇑
Corresponding authors. Tel.: +1 609 818 5493; fax: +1 609 818 3545.
E-mail addresses: scott.bolton@bms.com (S.A. Bolton), james.sutton@novartis.
com (J.C. Sutton).
0960-894X/$ - see front matter Ó 2013 Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.bmcl.2013.06.028

5240
S. A. Bolton et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5239–5243
H
N
H
N
provided intermediates 4a and 4b. EDC mediated coupling of the
carboxylic acids to neopentylamine provided triflates 5a and 5b.
DCC mediated coupling of the carboxylic acids to (S)-1-(tert-butyl-
dimethylsilyloxy)-3,3-dimethylbutan-2-amine provided triflates
5c and 5d. Tributyl stanannes 7a and 7b were prepared from picol-
inaldehydes 6a and 6b following methods described by Comins
and Kelly.8 Compound 7b was then transformed to compound 8
in three steps beginning with conversion to the carboxylic acid
by treatment with sodium chlorite. Esterification by treatment
with MeI and Cs2CO3 followed by mCPBA oxidation provided the
methyl N-oxo-3-stannyl picolinate 8.
O
N
O
NH
OH
OH
NH2
OH
OH
H2N
O
N
O
O
O
N
H
N
H
N
OMe
OMe
2
1
Figure 1. Structures of compounds 1 and 2.
Biaryl intermediates 9a and 9b and 10a and 10b were prepared
following general Stille coupling conditions where appropriately
substituted triflates 5a–d and stannane 7a were treated with
PdCl2(PPh3)2 in the presence of CuO in DMF. Oxidation with so-
dium chlorite then provided the carboxylic acids 9a and 9b and
10a and 10b. Coupling of 10a with the mono-Cbz protected
4-amidinoaniline 11a using DCC and catalytic DMAP in DCM
followed by removal of the TBS group with 3:1 HOAc/water and
catalytic hydrogenation afforded compound1. In similar fashion,
the identification of the potent factor VIIa inhibitor 2, utilizing a
novel 2-aminoisoquinoline P1 group (Fig. 1).
General methods for the preparation of 2-(pyridin-3-yl)benzoic
acid analogs of compound 1, are described in Scheme 1, following a
previously described route.7 Intermediates 4a and 4b were pre-
pared in three steps starting with esterification of 5-formyl sali-
cylic acid with benzylbromide or t-butanol. Treatment with Tf2O
followed by oxidation of the aldehyde with sodiumchlorite then
H
O
N
SnBu3
SnBu3
N+
CHO
CO2H
OHC
MeO2C
-O
OHC
R2
c,f,g
e
a,b,c
d
N
N
CO2R1
CO2R1
CO2H
R3
R3
OH
OTf
OTf
8
5a R1 = Bn, R2 = H
6a R3 = OMe
7a R3 = OMe
7b R3 = H
3
4a R1 = Bn
5b R1 = t-Bu, R2 = H
5c R1 = Bn, R2 = OTBS
5d R1 = t-Bu, R2 = OTBS
6b R3 = H
4b R1 = t-Bu
H
N
NH
H
H
O
O
O
N
O
N
HN
Cbz
NH2
OH
OH
i,j,k
HN
NH2
OTBS
CO2R1
h,c
h,c
5c + 7a
5d + 7a
5a + 7a
5b + 7a
11a
or
NH
O
CO2R1
O
O
N
HO
DMB
H
HO
N
1
N
N
NH2
OMe
N
OMe
i,l, m
OMe
N
9a R1 = Bn
NH2
10a R1 = Bn
2
11b
9b R1 = t-Bu
10b R1 = t-Bu
Scheme 1. Preparation of derivatives 1 and 2. Reagents and conditions: (a) BnBr, NaHCO3, DMF, 78% or t-BuOH, DCC, DMAP, THF, 93%; (b) Tf2O, pyridine, DCM, ꢀ10 °C, 83%;
(c) NaClO2, NaH2PO4ꢁH2O, 2-methyl-2-butene, tBuOH, CH3CN, H2O, 87%; (d) neopentylamine, EDAC, HOAt, NMM, DCM, 92% or ((S)-1-(tert-butyldimethylsilyloxy)-3,3-
dimethylbutan-2-amine), DCC, HOBt, DMAP, Et2O, 92%; (e) N,N,N0-trimethylethylenediamine, nBuLi (2.5 M in hexanes), tributyltin chloride, THF, 53%; (f) MeI, Cs2CO3, DMF,
72%; (g) mCPBA, DCM, 95%; (h) PdCl2(PPh3)2, CuO, DMF, 110 °C, 68–74%; (i) DCC, HOBt, DMAP, DMF, 53–85%; (j) AcOH–H2O (3:1), 70%; (k) H2, Pd/C, 1 N HCl, MeOH, 90%; (l)
TFA, anisole; (m) 1 N NaOH, 1,4-dioxane, 50% for two steps.
H
N
H
N
H
N
O
N
O
N
DMB
O
N
HN
H
N
N
DMB
O
HN
SnBu3
N+
NH2
MeO2C
-O
O
CO2tBu
CO2tBu
N
CO2tBu
a, b, c
O
e
11b
+
HO
d
N
H
CO2tBu
OTf
5b
8
OTf
OH
OH
13
14
12
g
,
f
j
,
i
r
o
g
,
h
k,l
O
H
N
H
N
O
N
O
N
H
N
NH2
NH2
NH2
N
O
CO2H
N
O
CO2H
N
O
CO2H
N
H
N
H
N
H
N
OR4
N
R6
R5
15
17
16
Scheme 2. Preparation of generic analogs 15 and 16 and compound 17. Reagents and conditions: (a) PdCl2(PPh3)2, CuO, DMF, 110 °C, 74–82%; (b) (CF3CO)2O, DMF, 95%; (c)
1 N NaOH, THF, H2O, 86%; (d) EDAC, HOAt, DMAP, DMF, 69%; (e) N-phenyltrifluoromethanesulfonimide, TEA, DCM, 81%; (f) R4-OH, DIAD, PPh3, THF; (g) TFA, DCM, 59–88% for
two steps; (h) R5(R6)NH, DMSO, 100 °C; (i) Bz(R6)NH, DMSO, 100 °C; (j) conc H2SO4, 28–34% for two steps; (k) tributyl(vinyl)tin, PdCl2(PPh3)2, CuO, DMF, 110 °C, 77%; (l) TFA,
anisole, 78%.

S. A. Bolton et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5239–5243
5241
Table 2
compound 2 was prepared from 10a by coupling with the 2,4-
Factor VIIa inhibition of compounds 17 and 22–3210
dimethoxybenzyl protected aminoisoquinoline 11b9 and subse-
quent deprotection with TFA/anisole followed by ester hydrolysis
using 1 N NaOH. Using similar chemistry as described above, inter-
mediates 9a and 9b were used to prepare compounds 18 and 19.
Compounds 20a and 20b and 21a and 21b were prepared by cou-
pling of Boc protected 2-aminobenzimidazole or 3-aminobenzam-
ide to the appropriate biaryl acid followed by deprotection.
Intermediate 12, Scheme 2, was preparedby coupling triflate 5b
and stannane 8 by treatment with PdCl2(PPh3)2 in the presence of
CuO in DMF. Treatment with trifluoroacetic anhydride to rearrange
the N-oxide to the pyridone, followed by hydrolysis with sodium
hydroxide provided the pyridone carboxylic acid 12. Coupling to
the protected aminoisoquinoline 11b provided compound 13.
Ether formation following standard Mitsunobu protocols and
deprotection provided compounds 22–29 (Table 2). Conversion of
13 to triflate 14 provided access to compounds 30–32 (Table 2)
through displacement with primary or secondaryamines followed
by global deprotection with TFA or H2SO4. Compound 17 was ob-
tained from triflate 14 by palladium coupling with tribu-
tyl(vinyl)stannane and subsequent deprotection.
H
O
N
NH2
OH
N
O
O
N
H
N
R'
Compound
R1
FVIIa IC50 (lM)
17
22
0.023
0.011
O
O
23
24
25
26
0.0125
0.111
0.034
0.113
O
O
O
O
O
27
28
29
0.437
0.357
0.499
Discussion: Following the general methods of Scheme 1, a num-
ber of compounds having various weakly basic and neutral P1
groups, such as benzamides, phenyl sulfones, sulfonamides, ani-
lines, phenols, and heterocycles were synthesized and screened
for activity against human factor VIIa and a panel of related serine
proteases: factor IXa, factor Xa, factor XIa, thrombin, trypsin, uro-
kinase (uPA) and tissue plasminogen activator (tPA).
O
N
30
31
32
0.101
0.051
0.009
N
H
Table 1 shows data for compounds containing P1 groups with
factor VIIaIC50’s below 0.5 lM. The results demonstrate that, in
N
H
the context of the 2-(pyridin-3-yl) benzoic acid scaffold, 2-amino-
isoquinoline, 2-aminobenzimidazole, and m-benzamide can re-
place benzamidine while retaining significant inhibition of factor
VIIa. The 2-aminoisoquinoline compounds 2 and 19 are equipotent
to 1 and 18, and likewise show good selectivity against related ser-
ine proteases, with the exceptions of factor IXa and factor Xa where
selectivity is only modest. By comparison, the 2-aminobenzimidaz-
ole compounds 20a and 20b are 12- and 15-fold less potent against
factor VIIa than 1 and 18, respectively. Selectivity is generally ob-
served with the notable exception of factor IXa, where compounds
20a and 20b were found to be moderately morepotent than factor
VIIa. Interestingly, the neutral m-benzamide P1 (compounds 21a
and 21b) proved to be a viable benzamidine replacement, albeit
Table 1
Factor VIIa inhibition for compounds with varied P1 groups10
H
O
N
N
R2
O
OH
O
R1
N
H
OMe
Compd
R1
R2
FVIIa
IC50 (lM)
FIXa
FXa
FXIa
Thrombin
Trypsin
lPA
tPA
NH
1
18
CH2OH
H
0.005
0.011
0.044
0.051
0.035
0.10
3.9
4.6
24.0
21.7
1.6
0.64
5.9
15.5
>33
>33
H2N
NH2
2
19
CH2OH
H
0.007
0.013
0.066
0.180
0.27
1.7
10.9
7.2
>33
>33
7.9
10.8
>33
31.5
>33
>33
N
N
20a
20b
CH2OH
H
0.060
0.164
0.013
0.067
>33
>33
>33
20.8
>33
>33
>33
>33
>33
>33
>33
>33
H2N
H2N
N
H
O
21a
21b
CH2OH
H
0.175
0.265
20
—
>33
—
>33
—
>33
—
>33
—
>33
—
>33
—

5242
S. A. Bolton et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5239–5243
Figure 4. (A) Overlay of compounds 1 and 2 depicting key interactions in the S1
pocket. (B) Key interactions of the benzamide of 21a in the S1 pocket of factor VIIa.
isoquinoline P1. A number of analogues were prepared as outlined
in Scheme 2 in which the methoxy group in 19 was replaced by
small ether groups, alkyl amine groups or vinyl (Table 2). The body
of data from Table 2 shows that small groups are well tolerated,
but as group size increases, factor VIIa inhibition decreases.
This suggests the region of the factor VIIa binding pocket occupied
by the methoxy group of 19 is sterically limited, which is consis-
tent with the crystallographic structures described below. Simi-
larly, it was also found that increasing the size of groups in the
S2 pocket leads to improved selectivity relative to factor IXa. For
example, compounds 19,22 and 24 containing methoxy, ethoxy
and isopropoxy R2 groups have factor IXa IC50 values of 0.18,
Figure 2. Overlay of 1 (white), 2 (green), and 21a (orange) extracted from their
crystallographic complexes with factor VIIa.11 The surface was generated using the
cognate protein structure bound to 1. Graphics in Figures 2–4 were generated using
the program PyMol.15
with a 20–40 fold loss in potency against factor VIIa, which is pre-
sumably due to loss of the salt bridge interaction of the basic P1
group with ASP189 (see the Modeling and structure below). Over-
all selectivity was improved in that all other serine proteases
1.80 and 6.70
lM, respectively (data not shown) compared to
0.051 M for compound 18. No significant change in inhibition
l
was seen versus other related serine proteases as group size
increased.
including factor IXa showed IC50 values greater than 33 lM. It
should be noted that as part of the SAR development for this series,
variation of the (S)-tert-leucinol amide group was evaluated,
and several examples showed comparable potency for both the
(S)-tert-leucinol and related neopentyl amides (data not shown).
We then decided to explore the S2 pocket of factor VIIausing the
potent and selective compound 19, which contains the 2-amino-
Modeling and structure: An overlay of crystallographic structures
of compounds 1, 2 and 21a (Fig. 2) indicate that all three analogues
are oriented similarly in the factor VIIa binding site.11,12 Further-
more, the binding mode of 1 in factor VIIa is indistinguishable from
that reported earlier fortrypsin.13 The methoxy pyridine moiety
occupies the S2 pocket defined by His57, Thr99 and Try215 side-
chains (Fig. 3). Hydrogen bonds are observed between the carbox-
ylic acid and His57 and Ser195 side-chains, and Gly193 oxygen.
The tert-butyl group is bound to a small hydrophobic recess de-
fined by amino acids Thr151, Gln40 and Asp193 and located on
the prime side of the catalytic triad. This pocket, rarely discussed
in the literature, is also involved in binding Leu17 sidechain of
BPTI.14
Although the phenyl rings of the benzamidine, 2-aminoisoquin-
oline and benzamide groups of 1, 2, and 21a, respectively, are ori-
ented similarly in the S1 pocket, the polarinteractions vary in
interesting ways (Fig. 4). The amidine moiety of 1 forms a biden-
tate salt bridge with Asp189 sidechain and also hydrogen bonds
with Gly219 oxygen and a water located above Tyr228. The
2-aminoisoquinoline analogue, deduced from crystal contacts to
be protonated, engages in similar interactions except that chemical
replacement of the polar NH with an aromatic CH results in a sig-
nificant rotation of the Gly219 carbonyl such that it engages in a
hydrogen bond with the side chain oxygen of Thr221 (Fig. 4A).
Binding of the benzamide moiety to S1 differs significantly, in that
a pair of hydrogen bonds are established with Gly219 (Fig. 4B). An
additional hydrogen bond is formed with a second water molecule
that is otherwise displaced by compounds1 and 2. Interestingly,
the amide carbonyl forms a close contact (3.2 Å) with Gly219 car-
bonyl carbon at an angle of 110°, which is in range for the geome-
try of the carbonyl–carbonyl polar interaction previously described
by Paulini et al.16 Taken together, these observations suggest that
efficient binding in the factor VIIa S1 pocket does not require a salt
Figure 3. Crystallographic structure of 1 bound to factor VIIa with key interactions
noted.

S. A. Bolton et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5239–5243
5243
Kühne, H.; Banner, D. W.; Böhm, H. J.; Stahl, M.; Ackermann, J.; Alig, L.; Weber,
L.; Wessel, H. P.; Riederer, M. A.; Tschopp, T. B.; Lavé, T. Bioorg.Med. Chem. Lett.
2005, 15, 5344; (e) Klingler, O.; Matter, H.; Schudok, M.; Donghi, M.; Czech, J.;
Lorenz, M.; Nestler, H. P.; Szillat, H.; Schreuder, H. Bioorg. Med. Chem.Lett. 2004,
14, 3715; (f) Parlow, J. J.; Kurumbail, R. G.; Stegeman, R. A.; Stevens, A. M.;
Stallings, W. C.; South, M. S. Bioorg. Med. Chem.Lett. 2003, 13, 3721.
4. (a) Frederick, R.; Pochet, L.; Charlier, C.; Masereel, B. Curr. Med. Chem. 2005, 12,
397; (b) Rao, L. V. M.; Mackman, N. Thromb. Res. 2010, 125(Suppl. 1), S1.
5. (a) Schechter, I.; Berger, A. Biochem. Biophys. Res. Commun. 1967, 27, 157; (b)
Schechter, I. Curr. Protein Pept.Sci. 2005, 6, 501.
6. (a) Senokuchi, K.; Ogawa, K. WO 9,941,231, 1999; Chem. Abstr. 1999, 131,
184864.; (b) WO 9,941,231 A1 19990819; PCT Int. Appl. 1999.
7. Kohrt, J. T.; Filipski, K. J.; Rapundalo, S. T.; Cody, W. L.; Edmunds, J. J.
Tetrahedron Lett. 2000, 41, 6041.
8. (a) Comins, D.; Killpack, M. J. Org. Chem. 1990, 55, 69; (b) Kelly, T. R.; Kim, M. J.
Org.Chem. 1992, 57, 1593.
9. Zhao, R.; Chen, B. US2004077865; U.S. Pat. Appl. 2004.
10. Factor VIIa determinations were made in 0.005 M calcium chloride, 0.15 M
sodium chloride, 0.001 M CHAPS, and 0.02 M HEPES buffer containing 0.1% PEG
6000 at a pH of 7.4. Determinations were made using purified human factor
VIIa, n = 2 separate experiments (Enzyme Research Labs) at a final assay
concentration of 1 nM, innovin human tissue factor (Dade Behring), and D-Ile-
Pro-Arg-AFC (Enzyme Systems Products) as substrate. Inhibitor concentrations
of required for half maximal effect (IC50) were determined by a four-parameter
logistic fit. Activity against other serine proteases was determined as
describedin: Schumacher, W.; Seiler, S.; Steinbacher, T.; Stewart, A.;
Bostwick, J.; Hartl, K.; Liu, E.; Oletree, M. Eur. J.Pharmacol. 2007, 570, 167–
174.
11. Structures were determined by soaking crystals of des-Gla factor VIIa in
complex with benzamidine with each of the inhibitors 1, 2, and 21a.
Resolutions of the complexes were: 2.2 Å (1), 1.7 Å (2) and 2.0 Å (21a).
12. Co-ordinates of 1, 2 and 21a have been deposited in the Protein Data Bank
under ID codes 4JZD, 4JZE, and 4JZF, respectively.
13. Sherawat, M.; Kaur, P.; Perbandt, M.; Betzel, C.; Slusarchyk, W. A.; Bisacchi, G.
S.; Chang, C.; Jacobson, B. L.; Einspahr, H. M.; Singh, T. P. Acta Crystallogr. 2007,
D63, 500.
14. (a) PDBID 1FAK.; (b) Zhang, E.; St. Charles, R.; Tulinsky, A. J. Mol. Biol. 1999, 285,
2089.
15. The PyMOL Molecular Graphics System, Version 1.3, Schrödinger, LLC.
16. Paulini, R.; Müller, K.; Diederich, F. Angew. Chem., Int. Ed. 2005, 44, 1788.
17. (a) Tucker, T. J.; Brady, S. F., et al J. Med. Chem. 1998, 41, 3210; (b) Pinto, D. J. P.;
Galemmo, R. A., et al Bioorg. Med. Chem. Lett. 2006, 16, 5584.
bridge with Asp189, and that the development of high affinity neu-
tral factor VIIa inhibitors are possible, such as have been discov-
ered for other trypsin-like serine proteases.17
In summary, we have identified several weakly basic or nonba-
sic P1 groups that can replace the highly basic benzamidine moiety
contained in 1, while maintaining factor VIIa potency. Of particular
interest was the discovery of the 2-aminoisoquinoline group as an
equipotent surrogate for benzamidine, but like compound 1, these
compounds displayed substantial factor IXa and Xa inhibition.
Substituting aminobenzimidazole for benzamidine led to com-
pounds with less factor Xa activity but even more potent factor
IXa activity. Use of a m-benzamide P1 group, although less potent,
indicates that a salt bridge interaction with Asp189 is not required
for activity and that together with further exploration of the S2
pocket of factor VIIa may enable the discovery of new analogs de-
void of factor IXa activity with improved permeability and oral
bioavailability.
Acknowledgment
The authors thank Dr. E. Scott Priestley for reviewing this Letter.
References and notes
1. (a) Liu, H.; Fox, C. J.; Zhang, S.; Kaye, A. D. Anesthesiol. Clin. 2010, 28, 723; (b)
McManus, D. D.; Chinali, M.; Saczynski, J. S.; Gore, J. M.; Yarzebski, J.; Spencer,
F. A.; Lessard, D.; Goldberg, R. J. Am. J. Cardiol. 2011, 107, 353.
2. Levi, M.; Eerenberg, E.; Kamphuisen, P. W. J. Thromb. Haemost. 2011, 9, 1705.
3. (a) Ott, I. Thromb. Haemost. 2010, 103, 7; (b) Hu, H.; Kolesnikov, A.; Riggs, J. R.;
Wesson, K. E.; Stephens, R.; Leahy, E. M.; Shrader, W. D.; Sprengeler, P. A.;
Green, M. J.; Sanford, E.; Nguyen, M.; Gjerstad, E.; Cabuslay, R.; Young, W. B.
Bioorg. Med. Chem. Lett. 2006, 16, 4567; (c) Kohrt, J. T.; Filipski, K. J.; Cody, W. L.;
Cai, C.; Dudley, D. A.; VanHuis, C. A.; Willardsen, J. A.; Narasimhan, L. S.; Zhang,
E.; Rapundalo, S. T.; Saiya-Cork, K.; Leadley, R. J.; Edmunds, J. J. Bioorg. Med.
Chem. Lett. 2006, 16, 1060; (d) Zbinden, K. G.; Obst-Sander, U.; Hilpert, K.;