Factor VIIa inhibitors: A prodrug strategy to improve oral bioavailability

Article abstract of DOI:10.1016/j.bmcl.2006.01.039

We have developed a series of potent and selective factor VIIa inhibitors based on the 2-[5-(5-carbamimidoyl-1H-benzoimidazol-2-yl)-6-hydroxy-biphenyl-3- yl]-succinic acid scaffold. These amidine-containing compounds have low oral bioavailability. Herein, we describe our efforts to improve the oral bioavailability of the parent amidine via a prodrug strategy where the amidine basicity and polarity were reduced with either an alkoxy-amidine or a carbamate prodrug.

Full text of DOI:10.1016/j.bmcl.2006.01.039

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2224–2228
Factor VIIa inhibitors: A prodrug strategy to improve
oral bioavailability
Jennifer R. Riggs,* Aleksandr Kolesnikov, John Hendrix, Wendy B. Young,
William D. Shrader, Dange Vijaykumar, Robin Stephens, Liang Liu, Lin Pan,
Joyce Mordenti, Michael J. Green and Juthamas Sukbuntherng
Celera Genomics, 180 Kimball Way, South San Francisco, CA 94080, USA
Received 7 December 2005; revised 7 January 2006; accepted 9 January 2006
Available online 3 February 2006
Abstract—We have developed a series of potent and selective factor VIIa inhibitors based on the 2-[5-(5-carbamimidoyl-1H-ben-
zoimidazol-2-yl)-6-hydroxy-biphenyl-3-yl]-succinic acid scaffold. These amidine-containing compounds have low oral bioavailabil-
ity. Herein, we describe our efforts to improve the oral bioavailability of the parent amidine via a prodrug strategy where the
amidine basicity and polarity were reduced with either an alkoxy-amidine or a carbamate prodrug.
Ó 2006 Elsevier Ltd. All rights reserved.
The development of novel therapeutic agents for the
treatment of coagulation disorders, such as deep vein
thrombosis (DVT) and pulmonary embolism (PE), is a
major focus of research in the pharmaceutical indus-
try.1–3 A number of coagulation proteases have been tar-
geted for the development of an anticoagulant therapy,
most commonly thrombin and factor Xa. While inhibi-
tion of these enzymes has proven beneficial in the clinic,
safety concerns, such as bleeding risk, are problematic.
Studies evaluating bleeding tendency4 and surgical
blood loss5 have shown that inhibition earlier in the
coagulation cascade, e.g., factor VIIa, may provide an
increased window of safety over inhibition downstream
in the cascade with factor Xa and thrombin.5–7 These
possible safety advantages lead us to focus on the devel-
opment of an oral and selective factor VIIa inhibitor.
alytic Ser195, common to all proteases in this family.10–14
Second, this series utilizes a basic amidine to form a salt
bridgewith acidic Asp189 in the P1-pocket of the protease.
Unfortunately, the amidine is associated with sub-optimal
oral bioavailability due to its charged nature (pKa ꢀ 12).
One strategy to increase the oral bioavailability of an ami-
dine is tomake a prodrug. Ideally, a prodrug of an amidine
would mask its polar basicity, and upon absorption of the
prodrug from the gastrointestinal tract into circulation,
the prodrug would be cleaved to the active parent amidine.
This approach has been successfully demonstrated in the
clinic with a number of amidine-containing compounds,
most notably ximelagatran, the hydroxy amidine prodrug
of melagatran, which is being developed as a direct throm-
bin inhibitor.15
The parent amidine 1 and its associated des-amidine
counterpart 2 (Table 1) were administered intravenously
(IV) and orally (PO) to male Sprague–Dawley rats
(n = 3, rats/compound per route) to evaluate the oral
absorption and bioavailability of the base scaffold with
and without the amidine functionality.16 Both com-
pounds pass Lipinski’s Rule of 5 and demonstrate
acceptable values for molecular weight (MW < 350)
and for calculated polar surface area (PSA < 125).17,18
The amidine-containing analog demonstrated low oral
absorption and bioavailability (<1%), while the corre-
sponding non-amidino counterpart had improved oral
absorption (ꢀ100%) and bioavailability (16%). This ini-
tial study in rats confirmed that the amidine hindered
We have previously reported on the development of the 2-
[5-(5-carbamimidoyl-1H-benzoimidazol-2-yl)-6-hydroxy-
biphenyl-3-yl]-succinic acid scaffold as an effective factor
VIIa inhibitor.8,9 The potency of this series is mediated
by two principle binding interactions. First, a phenol moi-
ety forms a unique network of hydrogen bonds to the cat-
Keywords: Factor VIIa; Anticoagulants; Thrombosis; Serine protease;
Prodrug; Amidine; Amidoxime; Hydroxy amidine; Alkoxy amidine;
Carbamate prodrugs; Pharmacokinetics; Indole synthesis.
*
Corresponding author. Tel.: +1 415 846 3717; e-mail: jennifer.riggs@
yahoo.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.01.039

J. R. Riggs et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2224–2228
2225
Table 1. Comparison of oral bioavailability for the amidino (1) and
non-amidino (2) analogs in male rats16
oxidation. To this end, we chose to evaluate amidino
carbamates as prodrugs. The masking of amidines as
carbamates has been used successfully in various aryl
amidine cases to improve oral bioavailability.21 To as-
sess this prodrug strategy, a range of carbamate pro-
drugs (9–12) of base amidine 3 were evaluated both
in vivo and in vitro (Table 2). The carbamates (9–12)
exhibited acceptable PSA and MW values; however,
the aqueous solubilities of these prodrugs were lower
than those of the parent amidine or the corresponding
hydroxy- and alkoxy-amidine prodrugs. The carbamates
were stable in plasma (with the exception of 12 in rat)
and in simulated gastric fluid,22 minimizing the possibil-
ity of cleavage prior to absorption in vivo. Although the
conversion of the carbamate prodrugs to parent amidine
(i.e., 3–20%) was superior to the conversion of the hy-
droxy- and alkoxy-amidine prodrugs to parent (i.e.,
<1%) following IV administration in rats,16 the oral bio-
availability of the parent amidine after oral administra-
tion of these prodrugs was between 1% and 2% in all
cases. As before, the low oral bioavailability of the par-
ent amidine was due to low cleavage of the carbamate
prodrug. With the outcome of our prodrug evaluation
on our amidine fVIIa scaffold 3, we changed our strate-
gy and replaced the aryl-amidine P1 of 3 with a non-am-
idino heterocycle. The result of these efforts will be
disclosed in a future publication.
N
NH OH
R
HO
Analog
R
VIIa Ki PSA MW Abs (%)a F (%)b
(lM)
NH
1c
2
0.41
37.0
119
69
344
302
<1
<1
16
H2N
H
ꢀ100
a Abs = oral absorption, based on portal-vein drug concentrations.
b F = oral bioavailability, based on jugular-vein drug concentrations.
c 200-hydroxy-500-fluoro analog, exact analog represented also had low
% F.
oral absorption in this series. Following previous litera-
ture examples, we chose to utilize a prodrug strategy
wherein the amidine was converted to hydroxy and alk-
oxy amidines. Typically, these prodrugs are converted
back to parent upon absorption via a reductive cleavage
of the amidine N–O bond within the liver.19 The corre-
sponding prodrugs of our series are less basic (pKa ꢀ 5–
7) and more lipophilic than the parent amidine and were
envisioned to provide an increase in absorption and ulti-
mately oral bioavailability of the corresponding parent
amidine.
The indole scaffold (3) was completed according to
Scheme 1. The 2,20-bisphenol 13 was mono-protected
with MemCl and selectively ortho-formylated with para-
formaldehyde and MgCl2 using the conditions of Hof-
slokken et al. to yield aldehyde 14.23 Acid hydrolysis
of the Mem group followed by protection of the bis-phe-
nol as its bis-methyl ether with MeI yielded bis-anisole
15. Conversion of the aryl aldehyde to the correspond-
ing alkyne was accomplished using the diazophospho-
nate Ohira reagent 16. The indole ring was constructed
via a Sonagashira coupling between iodobenzonitrile
18 and alkyne 17 to produce a transient biaryl alkyne,
which spontaneously cyclized upon the acidic mesylate
nitrogen to produce the corresponding N-mesyl indole
19. The indole nitrogen sulfonamide was removed via
treatment with NaOH/MeOH. Pinner conversion of
the nitrile intermediate to the ethyl imidate followed
by treatment with ammonium carbonate provided the
desired amidine. The compound was globally deprotec-
ted with aq HBr to cleave the methyl ethers to provide
amidine 3. Compound 3 was purified by preparative re-
verse-phase HPLC, and isolated as its corresponding
HCl salt after lyophilization.
Hydroxy amidine 4 and a series of alkoxy amidines,
including ethoxy, allyloxy, isopropoxy, and tert-butoxy
(5–8), were synthesized and evaluated in vitro and in vivo
(Table 2). The hydroxy amidine 4 showed good physical
properties (PSA < 120, MW < 450), solubility, plasma
stability, and rapid cleavage to parent in liver micro-
somes. The related alkoxy-amidine prodrugs (5–8) main-
tained good physical properties; however, solubility and
cleavage in liver microsomes were reduced. These pro-
drugs were administered IV and PO to male Sprague–
Dawley rats (n = 3 rats/compound per route) and the
amount of parent and prodrug in plasma was quantified
as a function of time.16 As anticipated, all prodrugs
showed moderate to high oral absorption and bioavail-
ability; however, the percent of prodrug that was con-
verted to parent following absorption (i.e.,
bioavailability of the parent) was negligible for all alk-
oxy prodrugs.
It has been suggested that pig enzymes are better models
for the human enzymatic reductive processes which are re-
quired for prodrug conversion.19 Hydroxy amidine 4 was
evaluated in vitro in pig microsomes and in vivo in pigs,
and the results paralleled those observed in rats.20 In con-
clusion, both pigs and rats demonstrated low oral bio-
availability due to negligible conversion of the hydroxy
amidine prodrug to parent amidine after absorption.
The synthesis of the hydroxy and alkoxy amidine pro-
drugs began with deprotection of bis methyl ether 20
with BBr3 followed by Pinner conversion of the nitrile
to the corresponding ethyl imidate (Scheme 2). Displace-
ment of the imidate with either hydroxylamine to pro-
vide the hydroxyamidine or alkylhydroxyamines to
produce alkoxyamidines completed the first series of
prodrugs 21. Conversely, the simple hydroxyamidine
can be formed directly from the nitrile upon treatment
with hydroxylamine. The carbamate prodrugs 23 were
completed by reaction of p-nitrophenyl carbonate 22
The low reductive cleavage that was obtained with our
hydroxy- and alkoxy-amidine prodrugs led us to focus
on prodrugs that could be converted to parent by an
alternate mechanism, such as enzymatic hydrolysis or

Table 2. Physical properties, stability, and pharmacokinetic parameters of parent amidine and hydroxy-, alkoxy-, and carbamate-amidine prodrugs
R
N
NH
OH
HO
H2N
Analog
R
Calculated physical
properties
Solubility24 (lM)
Plasma stability25 (%
remaining at 2 h)
Liver microsome
stability26 (%
remaining at 1 h)
PK parameters in rats
Prodrug
Parent
PSA
MW
pH 7.4
83
Rat
100
Human
100
Rat
Human
61
Absa (%)
Fb (%)
Fb (%)
3c
4
H
106
115
343
359
100
9
—
—
<1
OH
380
43
94
97
41
50
39–87
15–39
67–81
1
O
O
5
104
104
104
104
118
118
118
145
387
399
401
416
415
429
519
473
100
100
70
42
69
78
53
49
9
53–100
<1
O
6
24
25
94
100
99
100
100
100
100
100
100
96
55
100
100
2
<1
<1
2
7
63
54
10–36
38–62
NDd
23
O
O
8
<10
<10
<10
<10
<10
100
37
60–100
NDd
53–81
21–31
NDd
9
95
O
O
O
10
11
12
100
89
NDd
87
<1
2
Cl
Cl
O
14
Cl
O
O
O
37
12
9
NDd
2
O
O
a Abs = oral absorption, based on portal-vein drug concentrations.
b F = oral bioavailability, based on jugular-vein drug concentrations.
c 200-hydroxy-500-fluoro analog, exact analog represented also had low % F.
d ND, not determined.

J. R. Riggs et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2224–2228
OH Omem
2227
a,b
c,d
H
O
OH
OH
13
14
O
O
P
O
NC
I
O
e
f
H
+
+
OMe
NHMs
OMe
N2
O
O
O
18
17
15
16
NH
N
NC
g,h,i,j
H2N
N
Ms
N
H
O
O
HO
HO
NH
O
O
Ms
19
3
Scheme 1. Synthesis of indole base scaffold 3. Reagents and conditions: (a) Mem-Cl, K2CO3; (b) MgCl2, TEA, Paraformaldehyde; (c) HCl/dioxane
10–30% over 3 steps; (d) MeI, NaH, 80%; (e) K2CO3, MeOH, 75%; (f) PdCl2(PPh3)2, TEA, CuI; (g) NaOH, MeOH, reflux, 55% over 2 steps; (h)
HCl(g), EtOH; (i) (NH4)2CO3; (j) HBr, 40% over 3 steps.
Alkoxyamidines
RO
N
N
a,b,c,
H2N
N
H
N
H
O
O
HO
HO
20
21
Carbamates
R
O
N
O
O
O
H2N
R
d
+
3
O
N
H
O2N
HO
HO
22
23
Scheme 2. Synthesis of hydroxy, alkoxy amidines and carbamates. Reagents: (a) BBr3; (b) HCl(g), EtOH; (c) R-ONH2, 30–50% over 3 steps; (d) TEA,
DMF, 20–40%.
with amidine 3.27 p-Nitrophenyl carbonates were made
from the corresponding commercially available
chloroformates.
exception of the N-acyloxyalkoxy amidine 12 in rat plas-
ma. The in vitro instability of analog 12 did not corre-
late with increased oral bioavailability of the parent
amidine in vivo.
The practice of employing amidine prodrugs as a means to
improve oral bioavailability of amidine compounds has
been a successful strategy in a number of relevant cases.
Hydroxy and alkoxy amidines were made in our factor
VIIa inhibitor scaffold to test our hypothesis for increased
oral bioavailability. The hydroxy amidine showed low
microsomal stability in vitro, suggesting good prodrug
conversion was possible; however, negligible cleavage to
the parent amidine was observed in vivo. Alternatively,
the alkoxy amidines remained relatively stable both in vi-
tro andin vivo, leadingto moderateabsorption ofthe pro-
drug, but very low conversion and thus very low oral
bioavailability of the parent amidine.
Many factors can affect the lack of conversion of our
amidine prodrugs. Successful applications of amidine
prodrugs have typically involved alkyl- and mono aryl-
amidine structures,28 unrelated to our biaryl scaffold
system. We hypothesize that our lack of prodrug con-
version to the parent amidine upon absorption is the re-
sult of our biaryl scaffold not being recognized as a
substrate for the enzymes necessary to effect the amidine
prodrug conversion.20
Acknowledgment
Additionally, the carbamate prodrugs also remained
very stable in both plasma and gastric fluid with the
The authors thank Colin O’Bryan for the synthesis of
key intermediates.

2228
J. R. Riggs et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2224–2228
bioavailability (F) in rats were evaluated in portal vein
References and notes
(PV) and jugular vein (JV) cannulated animals. The
plasma concentrations of the compounds in the portal
and jugular vein were quantified and used to calculate the
area-under-the-curve (AUC). Oral absorption and bio-
availability were calculated from dose-normalized AUC
values as follows: Abs = AUCPO,PV/AUCIV,JV and
F = AUCPO,JV/AUCIV,JV. Prodrug conversion was calcu-
lated by dividing the dose-normalized AUC of the parent
after IV prodrug administration by the AUC of the parent
after IV parent administration.
1. Ruef, J.; Katus, H. A. Expert Opin. Invest. Drugs 2003, 12,
78.
2. Walenga, J. M.; Jeske, W. P.; Hoppensteadt, D.; Fareed,
J. Methods Mol. Med. 2004, 93, 9.
3. Wieland, H. A.; Laux, V.; Kozian, D.; Lorenz, M. Curr.
Opin. Invest. Drugs 2003, 4, 26.
4. Harker, L. A.; Hanson, S. R.; Wilcox, J. N.; Kelly, A. B.
Haemostasis 1996, 26, 7.
5. Himber, J.; Kirchhofer, D.; Riederer, M.; Tschopp, T. B.;
Steiner, B.; Roux, S. P. Thromb. Haemostasis 1997, 78,
1142.
17. Ertl, P.; Rohde, B.; Selzer, P. J. Med. Chem. 2000, 43,
3714.
18. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P.
J. Adv. Drug Delivery Rev. (Netherlands), Mar 1 2001, 46, 3.
19. Clement, B. Drug Metab. Rev. 2002, 34, 565.
20. Liu, L.; Ling, Y.; Bashnick, L.; Young, W.; Rai, R.;
Kumar, D. V.; Riggs, J. R.; Ton, T.; Shaghafi, M.;
Graupe, D.; Mordenti, J.; Sukbuntherng, J. Species
comparison of in vitro and in vivo conversion of five N-
hydroxyamidine prodrugs of fVIIA inhibitors to their
corresponding active amidines. Presented at the 13th
North America ISSX Meeting, Maui, HI, Oct. 2005.
21. Blackburn, Brent; Olivero, Alan G.; Robarge, Kirk.
WO98/46576, October 22, 1998.
22. Simulated gastric fluid (SGF) was prepared by dissolving
1.0 g NaCl in ꢀ500 mL of water and 3.5 mL HCl followed
by sprinkling 1.6 g purified pepsin on the surface of the
solution. After thorough mixing, SGF has a pH about 1.2.
Stability in SGF was determined by incubating each
prodrug (maximum solubility) at 37 °C for 2 h. The
disappearance of the prodrug was monitored by HPLC/
UV. The percentage remaining at 2 h was calculated by
dividing the peak area of the prodrug at the end of the
incubation by the peak area of the prodrug at the start
(time 0).
6. Suleymanov, O. D.; Szalony, J. A.; Salyers, A. K.;
Lachance, R. M.; Parlow, J. J.; South, M. S.; Wood, R.
S.; Nicholson, N. S. J. Pharmcol. Exp. Ther. 2003, 306,
1115.
7. Olivero, A. G.; Eigenbrot, C.; Goldsmith, R.; Robarge,
K.; Artis, D. R.; Flygare, J.; Rawson, T.; Sutherlin, D. P.;
Kadkhodayan, S.; Beresini, M.; Elliott, L. O.; DeGuzman,
G. G.; Banner, D. W.; Ultsch, M.; Marzec, U.; Hanson, S.
R.; Refino, C.; Bunting, S.; Kirchhofer, D. J. Biol. Chem.
2005, 280, 9160.
8. Young, W. B.; Kolesnikov, A.; Rai, R.; Sprengeler, P. A.;
Leahy, E. M.; Shrader, W. D.; Sangalang, J.; Burgess-
Henry, J.; Spencer, J.; Elrod, K.; Cregar, L. Bioorg. Med.
Chem. Lett. 2001, 11, 2253.
9. Young, W. B.; Mordenti, J; Torkelson, S.; Shrader, W. D.;
Kolesnikov, A.; Rai, R.; Liu, L.; Hu, H.; Leahy, E. M.;
Green, M. J.; Sprengeler, P. A.; Katz, B.; Yu, C.; Janc, J.
W.; Elrod, K.; Marzec, U. M.; Hanson, S. R. Bioorg. Med.
Chem. Lett. (in press).
10. Shrader, W. D.; Costerison, J.; Hendrix, J.; Hu, H.;
Kolesnikov, A.; Kumar, V.; Leahy, E.; Rai, R.; Shaghafi,
M.; Ton, T.; Torkelson, S.; Wesson, K.; Young, W. B.;
Katz, B. A.; Sprengeler, P. A.; Yu, C.; Cabuslay, R.;
Gjerstad, E.; Janc, J.; Sanford, E. Abstracts of Papers,
229th ACS National Meeting, San Diego, CA, United
States, March 13–17, 2005, MEDI-251.
11. Katz, B. A.; Elrod, K.; Luong, C.; Rice, M. J.; 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.
12. Verner, E.; Katz, B. A.; Spencer, J. R.; Allen, D.; Hataye,
J.; Hruzewicz, W.; Hui, H. C.; Kolesnikov, A.; Li, Y.;
Luong, C.; Martelli, A.; Radika, K.; Rai, R.; She, M.;
Shrader, W.; Sprengeler, P. A.; Trapp, S.; Wang, J.;
Young, W. B.; Mackman, R. L. J. Med. Chem. 2001, 44,
2753.
13. Katz, B. A.; Spencer, J. R.; Elrod, K.; Luong, C.;
Mackman, R. L.; Rice, M.; Sprengeler, P. A.; Allen, D.;
Janc, J. J. Am. Chem. Soc. 2002, 124, 11657.
14. Katz, B. A.; Elrod, K.; Verner, E.; Mackman, R. L.;
Luong, C.; Shrader, W. D.; Sendzik, M.; Spencer, J. R.;
Sprengeler, P. A.; Kolesnikov, A.; Tai, V. W.-F.; Hui, H.
C.; Breitenbucher, J. G.; Allen, D.; Janc, J. W. J. Mol.
Biol. 2003, 329, 93.
15. Eriksson, U.; Bredberg, U.; Hoffmann, K.; Thuresson, A.;
Gabrielsson, M.; Ericsson, H.; Ahnoff, M.; Gislen, K.;
Fager, G.; Gustafsson, D. Drug Metab. Dispos. 2003, 31,
29.
16. Plasma concentrations of prodrugs and parent compounds
were determined by LC/MS/MS. The plasma sample was
processed using acetonitrile precipitation, then the super-
natant was injected onto the LC column. The limit of
quantitation of the assay was 0.1–6 nM. Pharmacokinetic
data were analyzed by WinNonlin-Pro (Pharsight Corp.),
using compartmental and non-compartmental analysis for
IV and PO data, respectively. Oral absorption (Abs) and
23. Hofslokken, N. U.; Skattebol, L. Acta Chem. Scand. 1999,
53, 258.
24. Kinetic solubility was determined in phosphate buffer
(PB), pH 7.4, containing 5% DMSO. Samples were
prepared by adding 10 lL of 10 mM DMSO stock to
190 lL PB. After votexing, samples were centrifuged at
4300 RPM for 10 min to precipitate any insoluble mate-
rials. The supernatant was then analyzed by HPLC/UV
and quantified against a calibration standard.
25. Frozen, pooled human and rat plasma in sodium heparin
were purchased from Bioreclamation (Hicksville, NY).
Plasma stability was determined by incubating compound
(10 lM) at 37 °C for 2 h. The disappearance of the
prodrug was monitored by LC/MS/MS. The percentage
remaining at 2 h was calculated by dividing the peak area
of the prodrug at the end of the incubation by the peak
area of the prodrug at the start (time 0).
26. Cryopreserved, pooled, male human and rat liver micro-
somes were purchased from Xenotech, LLC (Lenexa,
Kansas). Liver microsome stability was determined by
incubating compound (2 lM) at 37 °C for 1 h. The
disappearance of the prodrug was monitored by LC/MS/
MS. The percentage remaining at 1 h was calculated by
dividing the peak area of the prodrug at the end of the
incubation by the peak area of the prodrug at the start
(time 0).
27. Rahmathullah, S. M.; Hall, J. E.; Bender, B. C.;
McCurdy, D. R.; Tidwell, R. R.; Boykin, D. W. J. Med.
Chem 1999, 42, 3994.
28. Investigational Drugs Database (IDDB) Copyright Ó
2005 Thomson Scientific Limited and/or its suppliers, 34-
42 Cleveland Street, London, W1T 4JE, United Kingdom.
All rights reserved.