Design, synthesis, and in vitro biological activity of benzimidazole based Factor Xa inhibitors

Article abstract of DOI:10.1016/S0960-894X(00)00139-6

Inhibitors based on the benzimidazole scaffold showed subnanomolar potency against Factor Xa with 500-1000-fold selectivity against thrombin and 50-100-fold selectivity against trypsin. The 2-substituent on the benzimidazole ring had a strong impact on th

Full text of DOI:10.1016/S0960-894X(00)00139-6

Bioorganic & Medicinal Chemistry Letters 10 (2000) 963±966
Design, Synthesis, and In Vitro Biological Activity of
Benzimidazole Based Factor Xa Inhibitors
Zuchun (Spring) Zhao,* Damian O. Arnaiz, Brian Griedel, Steven Sakata,
Jerry L. Dallas, Marc Whitlow, Lan Trinh, Joseph Post, Amy Liang,
Michael M. Morrissey and Kenneth J. Shaw
Discovery Research, Berlex Biosciences, 15049 San Pablo Avenue, PO Box 4099, Richmond, CA 94804-0099, USA
Received 13 December 1999; accepted 22 February 2000
AbstractÐInhibitors based on the benzimidazole sca€old showed subnanomolar potency against Factor Xa with 500±1000-fold
selectivity against thrombin and 50±100-fold selectivity against trypsin. The 2-substituent on the benzimidazole ring had a strong
impact on the FXa inhibitory activity. Crystallography studies suggest that the 2-substituent may have a conformational e€ect
favoring the extended binding conformation. # 2000 Elsevier Science Ltd. All rights reserved.
Factor Xa (FXa) is a trypsin-like serine protease that
plays a central role in the blood coagulation cascade
linking the intrinsic and extrinsic pathways to the ®nal
common pathway of coagulation. The primary role of
FXa is the proteolytic activation of thrombin (FIIa),
after combining with factor Va and calcium on a phos-
pholipid membrane to form the prothrombinase complex.
Prothrombinase catalyzes the formation of thrombin
from prothrombin. Thrombin, in turn, promotes blood
clot formation by catalyzing the formation of poly-
merizable ®brin from ®brinogen, as well as strongly
inducing platelet aggregation.1 Since direct inhibition of
thrombin has demonstrated a tendency to undesirably
prolong bleeding, the development of FXa inhibitors has
emerged as a alternative focus for the treatment and
prevention of thrombotic disorders.2 This is supported
by the recent reports of the low molecular weight FXa
inhibitors that have been found to be ecacious in animal
models of thrombosis, and o€er the potential for improved
treatment of a variety of thrombotic disorders.3
anticoagulants as measured by their prothrombin based
clotting time (PT) in plasma. The poor anticoagulant
activity was thought to be due to the low aqueous solu-
bility and high lipophilicity of these inhibitors that
could lead to extensive protein binding in plasma,
therefore the central sca€old was modi®ed in an attempt
to increase the hydrophilicity. In this communication,
we describe our initial series of inhibitors based on the
benzimidazole sca€old with improved potency, selectiv-
ity and anticoagulant activity.
A general synthesis of the benzimidazole based FXa
inhibitors described in Table 1 is shown in Scheme 1.5
The commercially available 4-amino-3-nitrophenol 1
was reduced to the bis-aniline, acylated with the appro-
priate acid chloride and then cyclized by re¯uxing in
concentrated hydrochloric acid to a€ord the various 2-
substituted benzimidazole derivatives 2. BOC protection
followed by Mitsunobu6 reaction with N-t-butoxy-
carbonyl-4-hydroxyl-piperidine and then selective removal
of the benzimidazole BOC group with methanolic
ammonia a€orded intermediate 4. Alkylation of 4 with
sodium hydride and 7-cyano-2-bromomethylnapthalene
5 a€orded a 1:1 mixture of 5 and 6 regioisomers (6).
This regioisomeric mixture was transformed to the cor-
responding amidines by sequential treatment with HCl
and ammonia gas in ethanol with concomitant depro-
tection of the Boc group.7 The regioisomeric mixtures
were separable by reverse-phase HPLC, but were gen-
erally carried on as a mixture and treated with ethyl
acetimidoimidate and triethylamine to a€ord the dibasic
benzimidazole analogues 7 and 8. The regioisomers
In a previous communication we identi®ed a series of
inhibitors based on indole and carbazole sca€olds.4
These studies identi®ed the 2-amidinonapthyl and 4-
iminoethylpiperidonyl groups as optimal substituents to
bind in the S1 and S4 pockets of Factor Xa, similar to
the reported studies with DX9065.3a Compounds from
these series were potent inhibitors of FXa when measured
against free FXa in vitro, however were weak functional
*Corresponding author. Tel.: +1-510-669-4457; fax: +1-510-669-4310;
e-mail: Spring_Zhao@berlex.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00139-6

964
Z. Zhao et al. / Bioorg. Med. Chem. Lett. 10 (2000) 963±966
Table 1. In vitro activities of 2-substituted benzimidazole regioisomers
6-Regioisomer, Ki (nM)a
2xPT(nM)b
5-Regioisomer, Ki (nM)
R
No.
hFXa
hFIIa
bTrp
No.
hFXa
hFIIa
bTrp
H
Me
Et
i-Pr
t-Bu
s-Bu
n-Pr
n-Bu
7a
7b
7c
7d
7e
7f
7g
7h
7i
2.9
810
230
60
32
48
160
70
230
2670
590
50
18
7.7
5.7
7.0
10
6.8
5.7
18
450
350
130
68
260
280
290
625
NDc
NDc
8a
8b
8c
8d
8e
8f
8g
8h
8i
6.3
6.2
4.0
2.0
2.0
3.3
4.5
2.8
5.9
3.6
>5000
3900
2100
870
8.2
16
17
16
18
13
16
14
14
10
0.27
0.087
0.026
0.046
0.12
0.23
0.23
1.3
950
2300
2400
2000
>5000
>5000
(CH2)2COOH
(CH2)2CONH2
7j
1.0
4.4
8j
aKi values for these competitive inhibitors are averaged from multiple determinations (n ꢀ 2) and the standard deviations are <30% of the mean.
bConcentration of inhibitor required to double the prothrombin based clotting time in human plasma.
cNot determined.
Scheme 1.
were separated and puri®ed by preparative reverse-phase
HPLC, and the regiochemistry of the isomers were
determined by NOE studies. In the 1,5-isomers cross
NOE e€ect was observed between the naphthyl-CH2
and the benzimidazole H-4, whereas, in the 1,6-isomers
NOE e€ect was observed between the CH2 and the H-7.
protected by reaction with one equivalent of sodium
hydride and p-toluenesulphonyl chloride, which allowed
for selective alkylation or acylation of the aniline nitrogen.
Selective deprotection of the benzimidazole ring nitro-
gen followed by alkylation with 5 a€orded a 1:1 mixture
of the 5 and 6 regioisomers (12) that were converted to
13 and 14 in a similar manner as described above.
A general synthesis of the 6-aminosubstituted inhibitors
described in Table 2 is illustrated in Scheme 2. Catalytic
hydrogenation of commercially available 5-nitrobenzi-
midazole 9 followed by reductive amination with N-BOC-
4-piperidone and sodium cyanoborohydride a€orded
10. The benzimidazole ring nitrogen was selectively
Examination of the results in Table 1 demonstrates that
the substituent at the 2-position of the benzimidazole
has a large in¯uence on the FXa inhibitory activity of
the 6-substituted regioisomers (7a±j), which is consistent
with our earlier studies with the indole and carbazole
Table 2. In vitro inhibitory activities of N-substituted benzimidazole regioisomers
6-Regioisomer,
Ki (nM)a
5-Regioisomer,
Ki (nM)
R
No.
hFXa
hFIIa
bTrp
No.
hFXa
hFIIa
bTrp
CH2COOMe
CH2CONH2
13a
13b
13c
13d
13e
13f
13g
13h
0.3
0.46
300
420
1.1
3.0
300
417
98
490
4610
5000
300
650
4600
5000
11
28
260
670
41
47
256
670
13a
13b
13c
13d
13e
13f
13g
13h
22
NDb
30
370
80
80
370
ND
2700
ND
7.6
ND
60
800
160
120
800
ND
CH2CH2CH2COOMe
CH2CH2CH2COOH
CH2CH(Me)COOMe
CH2CH(Me)COOH
COCH2CH2COOEt
COCH2CH2COOH
6000
5000
5000
5000
5000
ND
aKi values for these competitive inhibitors are averaged from multiple determinations (n ꢀ 2) and the standard deviations are <30% of the mean.
bNot determined.

Z. Zhao et al. / Bioorg. Med. Chem. Lett. 10 (2000) 963±966
965
Scheme 2.
inhibitors.4 Substitution of a methyl group at the
2-position (7b, FXa Ki=0.27 nM) a€ords a 10-fold
increase in the FXa inhibitory activity compared to the
unsubstituted analogue (7a FXa Ki=2.9 nM). An
additional 3-fold increase in activity is seen with the
ethyl analogue (7c, FXa Ki=0.087 nM) and another
3-fold enhancement is seen with the isopropyl substituent
(7d, FXa Ki=0.026 nM). The isopropyl substituent
appears to be the optimal alkyl substituent as the
n-propyl substituent (7g) or the butyl substituents (7e,f,h)
are less active. Introduction of a polar substituent had a
negative in¯uence on the FXa inhibitory activities as
demonstrated by the propionic acid and amide analo-
gues (7i and 7j). Overall, comparison of 7a with 7d
shows a 1000-fold increase of the inhibitory activity by
substitution of an isopropyl substituent for a hydrogen
atom at the 2-position of the benzimidazole. A similar
but less pronounced increase in potency is also seen
towards FIIa (25-fold) and trypsin (Trp) (approximately
10-fold), which improves the overall selectivity pro®le
for this series of inhibitors. In contrast to the results
with the 6-regioisomers, there is a maximum 3-fold
e€ect of the 2-substituent on the FXa inhibitory activ-
ities for the 5-regioisomers (8a±8j).
no contact with the protein but may exert a steric in¯uence
on the conformation of the naphthylamidine substituent
that would favor the extended L-shape conformation of
the molecule explaining the e€ect of this substituent.
The nitrogen substituted analogues in Table 2 were
prepared in an attempt to append substituents that
would either enhance the inhibitory activity, or provide
a site for substitution of a carboxylate substituent that
could enhance the oral activity for this series of com-
pounds. The results in Table 2 demonstrate that while
alkylation with acetate or acetamide (13a and 13b)
a€orded no signi®cant loss of the FXa inhibitor potency in
comparison to 7b, the larger substituents led to inhibi-
tors with signi®cantly lowered inhibitory activity. The
corresponding acetic acid analogue was prepared in an
attempt to evaluate the e€ect of this substituent on the
oral absorption, however the compound was not stable.
In summary, we have developed a series of FXa inhibitors
based on the benzimidazole sca€old. These inhibitors
The 1,6-benzimidazole based FXa inhibitors are potent
functional anticoagulants as measured by the concentra-
tion needed to double the clotting fold in plasma
(2ÂPT). All inhibitors extended the clotting time at
submicromolar concentrations (2ÂPT=68±450 nM) in
contrast to the weak anticoagulant activity of the indole
and carbazole based FXa inhibitors. The improved antic-
oagulant activity may be attributed to the increased FXa
inhibitory activity as well as improved aqueous solubility.
X-ray crystallography of 7b in bovine trypsin is shown
in Figure 1.8 This inhibitor is bound in the active site of
trypsin in an extended L-shaped conformation, similar
to the reported binding mode of DX-9065a in bovine
trypsin and human FXa.9 The naphthylamidine ring is
bound in the S1 pocket, making four hydrogen bonds to
Asp189, Ser 190, and Gly219. The central benzimidazole
ring sits above the catalytic His57 and makes no apparent
contact with the protein, and the iminoethylpiperidine
ring binds in the S4 pocket, de®ned by Thr98, Leu99,
Gln175, and Trp215. The 2-methyl group appears to make
Figure 1. X-ray crystal structure of 7b bound to the active site of
bovine trypsin.

966
Z. Zhao et al. / Bioorg. Med. Chem. Lett. 10 (2000) 963±966
have subnanomolar inhibitory activity against FXa with
500±1000-fold selectivity for thrombin and 50±100-fold
selectivity for trypsin. It was also demonstrated that the
2-substituent on the benzimidazole ring played a critical
role in enhancing inhibitor potency. X-ray crystal-
lography indicated that the inhibitor binds to trypsin in
an L-shaped con®rmation, and the 2-substituent on the
benzimidazole may have a conformational e€ect favor-
ing the extended L-shaped conformation. Substitution
of an exocyclic nitrogen with carboxyl derived sub-
stituents decreased the FXa inhibitory activity. This led
to a series of carboxylate analogueues derived from
substitution at the 4, 5, and 7 positions of the benzimi-
dazole ring with enhanced oral absorption. This work
will be the focus of a future communication.
3. (a). Nagahara, T.; Yokoyama, Y.; Inamura, K.; Katakura,
S.-I.; Komoriya, S.; Yamaguchi, H.; Hara, T.; Iwamoto, J.
Med. Chem. 1994, 31, 1200. (b). Phillips, G. B.; Buckman, B.
O.; Davey, D. D.; Eagen, K. E.; 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.; Vergona, R.; Walters, J.; White,
K.; Whitlow, M.; Wu, S.; Xu, W.; Morrissey, M. M. J. Med.
Chem. 1998. 41, 3557.
4. Arnaiz, D. O.; Shaw, K. J.; Zhao, Z. Bioorg. Med. Chem.
Lett., submitted.
5. Phillips, M. A. J. Chem. Soc. 1928, 2393.
6. Mitsunobu; Yamada. Bull. Chem. Soc. Jpn. 1937, 40, 2380.
7. Neilson In The Chemistry of Amidines and Imidates; Patai,
S. Ed.; Wiley: New York, 1975; pp 385±489.
8. Whitlow, M.; Arnaiz, D. O.; Buckman, B. O.; Davey, D.
D.; Griedel, B.; Guilford, W. J.; Koovakkat, S. K.; Liang, A.;
Mohan, R.; Phillips, G. B.; Seto, M.; Shaw, K. J.; Xu, W.;
Zhao, Z. Z.; Light, D. R.; Morrissey, M. M. Acta Crystallogr.
D. 1999, 55, 1395.
References and Notes
1. Davie, E. W.; Fujikawa, K.; Kisiel, W. Biochem. 1991, 30,
10363.
2. Zhu, B.; Scarborough, R. M. Cardiovascular, Pulmonary &
Renal Investigational Drugs 1999, 1, 63.
9. (a) Stubbs, M. T.; Huber, R.; Bode, W. FEBS Lett. 1995,
375, 103. (b) Brandstetter, H.; Kuhne, A.; Bode, W.; Huber,
R.; Von der Saal, W.; Wirthensohn, K.; Engh, R. A. J. Biol.
Chem. 1996, 271, 29988.