Optimization of a screening lead for factor VIIa/TF

Article abstract of DOI:10.1016/S0960-894X(01)00420-6

The structure-based design and progression of a screening lead to a 3 nM factor VIIa/TF inhibitor with improved selectivity versus related enzymes is described.

Full text of DOI:10.1016/S0960-894X(01)00420-6

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2253–2256
Optimization of a Screening Lead for Factor VIIa/TF
Wendy B. Young,* Aleksandr Kolesnikov, Roopa Rai, Paul A. Sprengeler,
Ellen M. Leahy, William D. Shrader, Joan Sangalang, Jana Burgess-Henry, Jeff Spencer,
Kyle Elrod and Lynne Cregar
Departments of Medicinal Chemistry, Structural Chemistry, and Enzymology, Axys Pharmaceuticals, Inc.,
385 Oyster Point Blvd., South San Francisco, CA 94080, USA
Received 2 April 2001; accepted 18 May 2001
Abstract—The structure-based design and progression of a screening lead to a 3 nM factor VIIa/TF inhibitor with improved
selectivity versus related enzymes is described. # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
a potent inhibitor of factor VIIa/TF and several serine
proteases.16 This compound showed good inhibition for
0
factor VIIa/TF (Ki =78 nM), although also demon-
Factor VIIa, when complexed with Tissue Factor (TF),
rapidly activates coagulation factors IX and X to fac-
tors IXa and Xa, respectively, which in turn are then
responsible for the activation of prothrombin to
thrombin.1 Active thrombin then cleaves fibrinogen to
fibrin which forms the structural basis of all blood clots.
Much research has focused on the generation of potent
and selective inhibitors of factor Xa or thrombin as
potentially valuable therapeutic agents for the treatment
of thromboembolic disorders.2ꢀ4 More recently, the
fVIIa/TF complex has been deemed a worthwhile point
of intervention.5 Studies using anti-factor VIIa anti-
bodies,6 active-site-inhibited factor VIIa,7 and recombi-
nant anticoagulant tissue factor pathway inhibitor
(TFPI)8 have shown promising results and prompted
the attention of many research groups. Accordingly,
there is increasing interest in the generation of potent
and selective small molecule factor VIIa inhibitors in the
search for novel antithrombotics.9ꢀ11 Herein, we
describe our efforts in the design and synthesis of a ser-
ies of novel, potent and selective factor VIIa inhibitors.
strated submicromolar inhibition for factor Xa and
urokinase-type plasminogen activator (uPa) and low
micromolar inhibition versus thrombin (fIIa), plasmin,
and trypsin.17 The structure-based design and progres-
sion of 1 to a 3 nM inhibitor of factor VIIa/TF with
improved selectivity versus relevant anti-targets is
described.
Utilizing the crystal structures of related compounds in
uPa16,18 as well as other serine proteases, a model of 1
bound to factor VIIa was constructed (Fig. 1a). The
binding mode, which is consistent between all these
structures, involves a network of short and normal
hydrogen bonds between the phenol group, the Ser195
and His57 residues, and a water molecule bound in the
oxyanion hole as depicted in Figure 1b.16 The 5-amidino-
benzimidazole binds as expected in the S1 pocket with a
salt-bridge between the amidine and Asp189. Occasion-
ally, a water molecule spans the two moieties instead of
through direct hydrogen bonds. Due to the smaller size
of the factor VIIa S1 pocket (as compared to uPa), we
built the model with direct hydrogen bonds between
Since factor VIIa is a trypsin-like serine protease we
began our search by screening our collection of pro-
prietary serine protease inhibitors.12ꢀ15 From this
library, we identified 1, 2-(30-amino-5-chloro-2-hydroxy-
biphenyl-3-yl)-1H-benzoimidazole-5-carboxamidine, as
*Corresponding author. Tel.: +1-650-829-1130; fax: +1-650-829-
1019; e-mail: wendy_young@axyspharm.com
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00420-6

2254
W. B. Young et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2253–2256
Asp189 and 1. The model indicates that the C30 phenyl
ring is bound in the S10 pocket of factor VIIa and, in
addition to hydrophobic interactions with the Cys40-
Cys58 disulfide, appears to be involved in edge to face
b-stacking with the catalytic imidazole ring of His57.
The model also indicates that the methylene chain of
Lys60A lies immediately above this ring and is directed
towards the central phenolic ring of the inhibitor. The
central phenolic ring is packed against residue 192,
which is a lysine in factor VIIa. Interestingly, most
serine proteases have a glutamine at this site with the
notable exception of thrombin which has a glutamic
acid. Interaction between this residue and an acidic
functionality might be expected to afford a degree of
selectivity versus thrombin.
potentially, Lys60A as well. Additionally, we antici-
pated that an anionic charge here would repel the
Glu192 of thrombin imparting target selectivity. These
changes are represented in compound 2 as depicted in
Table 1. Compound 2 did prove to have a stronger
0
binding affinity toward factor VIIa/TF (Ki =12 nM)
and, as expected, improved selectivity versus thrombin.
The selectivity versus the other anti-targets studied
remained fairly consistent.
In an attempt to further improve the overall binding
affinity of these factor VIIa inhibitors, we surveyed the
effect of meta substitution on either affecting the
strength of the b-stacking interaction of His57 or
increasing S10 pocket complementarity. Substitution at
the ortho position might be expected to adversely affect
the dihedral angle between the two phenyl rings while
the model indicated that there was very little room for
para substitution. Hence, a small set of meta substituted
analogues was generated. Of these compounds, the 3-
nitroaryl analogue, 6, showed the strongest binding
Based on this model and previous work with this class
of inhibitors,16,18 we believed that immediate improve-
ment of the inhibition for factor VIIa by 1 could be
realized in two ways. First, we transformed the benzi-
midazole ring of 1 into an indole. Earlier results from
our proprietary serine protease inhibitor library clearly
indicate that indole rings provide improved inhibition
for factor VIIa as compared to benzimidazole rings.
Secondly, we exchanged the chlorine at C50 of 1 with an
acetic acid moiety to interact with Lys192 and,
0
affinity to factor VIIa/TF (Ki =3 nM) and the best
selectivity versus thrombin, plasmin, trypsin, and uPa.
Acidic (7–9) and basic (10) analogues show decreased
factor VIIa/TF inhibition most likely due to the hydro-
phobic nature of the site created by the alkyl chain of
Lys60A and the Cys40-Cys58 disulfide.
Having identified 6, which has a C30 moiety that fits well
into the S10 pocket and affords increased potency for
factor VIIa/TF and enhanced selectivity for relevant
anti-targets, we further probed the interactions with the
Lys192 of factor VIIa. Anticipating that optimization of
an anionic charge near Lys192 would further enhance
the selectivity profile, a series of derivatives that incor-
porate anionic groups attached to the C50 position of 6
were generated (Table 2). Compounds 6, 11, and 13
maintain similar inhibition for factor VIIa/TF, while 12,
14–15 are less effective inhibitors. These results would
indicate that Lys192 can accommodate acid tethers
Table 1. C30 modifications
0
Ki (nM)
fVIIa/TF
Compd
R1
Selectivity ratio
fXa fIIa Plasmin Trypsin uPa
2
3
4
6
7
8
NH2
H
Cl
NO2
CO2H
OPO3H2
12
41
33
3
320
950
26 550
27
6
17
90
5
83
1
3
160
1
Na
6
1
4
64
6
4
4
5
49
112
20 960
0.5 375
1
87
4
9
480
430
0.5 100
2
3
3
2
2
10
4
140
0.6
Figure 1. (a) Model of 1 bound in factor VIIa; (b) H-bonding array
displayed by 1 with the catalytic residues of uPa in the crystal structure.

W. B. Young et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2253–2256
2255
lengths of both 6 and 11, while the length of 12 may not
be optimal. The anionic charge on compound 13 is
likely extended out far enough to interact with Lys60A
and this may explain its improved potency.
levels while 14 is slightly more effective. Compound 15
was the least effective. In-house studies have indicated
that by effectively decreasing the lipophilicity of such
compounds (i.e., increasing water solubility/decreasing
plasma protein binding) a marked improvement in the
anticoagulant effect has been shown. It is also realized
that the selectivity profile of the inhibitors plays a role
in the overall anticoagulant effect.
The in vitro anticoagulant effect (prothrombin times,
PTs) of several lead analogues is listed in Table 2.
Compounds 6, 11–12 double the PT at low micromolar
The synthesis of 6, starting from commercially available
methyl 4-hydroxyphenylacetate (16), is illustrated in
Scheme 1. Formylation of 16 with MgCl2 and paraform-
aldehyde, bromination with NBS, and subsequent
treatment with methoxyethoxymethyl chloride afforded
17. A Suzuki coupling of 17 with the commercially
available boronic acid 18 followed by treatment with
dimethyl-1-diazo-2-oxopropylphosphonate19 afforded
alkyne 19. A palladium(0) catalyzed coupling of 19 with
the aryl iodo moiety 20 then gave 21. Removal of the
protecting groups and transformation of the nitrile into
an amidine via standard Pinner/aminolysis conditions
afforded 6. All other analogues described were syn-
thesized in an analogous manner to 6, although the
requisite boronic acids or a modified starting phenol
moiety was utilized.12
Table 2. C50 modifications
0
Ki (nM)
FVIIa/TF
Compd
R
Selectivity ratio
2xPT
(mM)
fXa fIIa Plasmin Trypsin uPa
11
6
5
3
42 420
75
90
6
70
160
37
63
5
44 4.1
64 1.7
20 960
12
13
14
15
27
3
2
5
59
6
2.2
The optimization of a nonselective serine protease
screening lead to a 3 nM factor VIIa/TF inhibitor,
which possesses improved selectivity versus relevant
anti-targets, is described. Pharmacokinetics, in vivo
animal efficacy, and non-amidino P1 analogue develop-
ment will be the subject of a future publication.
870
33
3
21 na
50
50
0.5 40
0.5 140
1
0.73
20
40
15 16
Acknowledgements
The authors wish to thank Michael Venuti and Jochen
Knolle for helpful scientific discussions, Liling Fang for
analytical support and Barbara Nielsen for aid in
preparing this manuscript.
References and Notes
1. Davie, E. W.; Fujikawa, K.; Kisiel, W. Biochemistry 1991,
30, 10363.
2. Hauptmann, J.; Sturzebecher, J. Thromb. Res. 1999, 93,
203.
3. Zhu, B. Y.; Scarborough, R. M. Annu. Rep. Med. Chem.
2000, 35, 83.
4. Schmaier, A. H. Handb. Exp. Pharmacol. 2000, 140, 277.
5. Johnson, K.; Hung, D. Cor. Art. Dis. 1998, 9, 83.
6. Biemond, B. J.; Levi, M.; Ten Cate, H.; Soule, H. R.;
Morris, L. D.; Foster, D. L. Thromb. Haemost. 1995, 73, 223.
7. Arnljots, B.; Ezban, M.; Hedner, U. J. Vasc. Surg. 1997,
25, 341.
8. Bajaj, M. S.; Bajaj, S. P. Thromb. Haemost. 1997, 78, 471.
9. Jakobsen, P.; Pederson, B. R.; Persson, E. Bioorg. Med.
Chem. 2000, 8, 2095.
10. Kohrt, J. T.; Filipski, K. J.; Rapundalo, S. T.; Cody,
W. L.; Edmunds, J. J. Tetrahedron Lett. 2000, 41, 6041.
11. Gallagher, K. P.; Mertz, T. E.; Chi, L.; Rubin, J. R.;
Uprichard, A. C. G. Handb. Exp. Pharmacol. 1999, 132, 421.
Scheme 1. Palladium(0) catalyzed route to 6. Reagents: (a) MgCl2,
(CH2O)n, MeCN; (b) NBS, DMF; (c) MemCl, Hunig’s base; (d) 18,
Pd(PPh3)4 1 M Na2CO3, toluene, reflux; (e) dimethyl-1-diazo-2-oxo-
propylphosphonate, K2CO3, MeOH; (f) 20, Pd(PPh3)2Cl2, CuI, NEt3,
MeCN, 80 ꢁC; (g) NaOH, THF; (h) EtOH, HCl (anhyd); (i) NH3,
EtOH; (j) 1 N HCl, reflux.

2256
W. B. Young et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2253–2256
12. Allen, D.A.; Hataye, J. M.; Hruzewicz, W. N.; Kolesni-
kov, A.; Mackman, R. L.; Rai, R.; Shrader, W. D.; Spencer, J.
R.; Verner, E. J.; Young, W. B. International Patent 00/35886,
published June 22, 2000.
13. Day, R. F; Young, W. B. International Patent 99/26933,
published June 3, 1999.
14. Fatheree, P. R.; Jenkins, T. E.; Li, Y.; Linsell, M. S.; Rai,
R.; Shrader, W. D.; Trapp, S. G.; Young, W. B. International
Patent 99/26932, published June 3, 1999.
15. Fatheree, P. R.; Jenkins, T. E.; Linsell, M. S.; Trapp, S.
G.; Verner, E. J.; Young, W. B. US6150379, Nov 21, 2000.
16. Verner, E., Katz, B.; Spencer, J.; Allen, D.; Hataye, J.;
Hruzewicz, W.; Hui, H.; Kolesnikov, A.; Li, Y.; Luong, C.;
Martelli, A.; Radika, K.; She, M.; Shrader, W.; Sprengeler, P.
A.; Trapp, S.; Wang, J.; Young, W.; Mackman, R. L. J. Med.
Chem., in press.
17. Inhibition assays for factor Xa, thrombin, plasmin and
trypsin were performed as described (Cregar, L.; Elrod, K. C.;
Putnam, D.; Moore, W. R. Arch. Biochem. Biophys. 1999, 366,
125) with the pH adjusted to 7.4. Factor VIIa assay was
performed and analyzed as above using 7 nM VIIa and Pen-
tapharm VIIa as the respective substrates. The buffer for the
factor VIIa assay was supplemented with 25 nM relipidated
tissue factor and 5 mM CaCl2. Coagulation experiments were
carried out with human plasma and a coagulation meter (Elec-
tra 800) in accordance with the manufacturer’s instructions.
18. Katz, B. A.; Elrod, K.; Luong, C.; Rice, M.; Mackman,
R. L.; Sprengeler, P. A.; Spencer, J.; Hataye, J.; Janc, J.; Link,
J.; Litvak, J.; Rai, R.; Rice, K.; Sideris, S.; Verner, E.; Young,
W. J. Mol. Biol. 2001, 307, 1451.
19. Muller, S.; Liepold, B.; Roth, G. J.; Bestmann, H. J.
Synlett 1996, 521.