Design and synthesis of aminophenol-based factor Xa inhibitors

Article abstract of DOI:10.1016/S0960-894X(02)00142-7

A novel potent and selective aminophenol scaffold for fXa inhibitors was developed from a previously reported benzimidazole-based naphthylamidine template. The aminophenol template is more synthetically accessible than the benzimidazole template, which si

Full text of DOI:10.1016/S0960-894X(02)00142-7

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1307–1310
Design and Synthesis of Aminophenol-Based Factor Xa Inhibitors
Shung Wu, William J. Guilford,* Yuo-Ling Chou, Brian D. Griedel, Amy Liang,
Steve Sakata, Kenneth J. Shaw, Lan Trinh, Wei Xu, Zuchun Zhao
and Michael M. Morrissey
Discovery Research, Berlex Biosciences, 15049 San Pablo Avenue, PO Box 4099, Richmond, CA 94804-0099, USA
Received 20 December 2001; accepted 8 February 2002
Abstract—A novel potent and selective aminophenol scaffold for fXa inhibitors was developed from a previously reported benzi-
midazole-based naphthylamidine template. The aminophenol template is more synthetically accessible than the benzimidazole
template, which simplified the introduction of carboxylic acid groups. Substitution of a propenyl-para-hydroxy-benzamidine group
on the aminophenol template produced selective, sub-nanomolar fXa inhibitors. The potency of the inhibitors is partially explained
with the aid of a trypsin complex crystal structure. # 2002 Elsevier Science Ltd. All rights reserved.
Several excellent recent reviews1 have described the his-
tory and utility of factor Xa (fXa) inhibitors as safe and
effective anticoagulants. Previously, we described the
development of subnanomolar benzimidazole-based
factor Xa (fXa) inhibitors,2,3 which suffered from poor
selectivity versus trypsin and a poor pharmacokinetic
(PK) profile. Previous efforts to improve the selectivity
and PKprofile by introduction of a carboxylic acid
group, which had some success in our pyridine-based
fXa inhibitors,4 or by substitution of the naphthyl-
amidine group with either a biphenylamidine or a pro-
penylbenzamidine group had limited success.5 In this
communication, we present fXa inhibitors in which the
benzimidazole template is replaced with the more
synthetically accessible aminophenol template6 (Fig. 1).
naphthylamidine with 2-methylpropenylbenzamidine
afforded inhibitors with similar potency and selectivity
(2 vs 3c).
Optimization of the aminophenol ring substituents was
examined next (Table 2). When the 3-nitro group of the
phenyl ring was replaced with H, F, or CO2Et the inhi-
bitors lost potency and selectivity (3b vs 4a–c). How-
ever, replacement of the nitro group with
a
trifluoromethyl group resulted in only a 3-fold decrease
in potency and selectivity (3b vs 4d). Potency and selec-
tivity levels were regained by exploring substitution on
the aniline nitrogen. Mono- or di-acid moieties appen-
ded to the aniline afforded subnanomolar fXa inhibitors
with enhanced potency and selectivity (4d vs 4e–f).
By using the less rigid aminophenol template (Fig. 1, 2),
fXa inhibitors of similar potency and selectivity as the
benzimidazole analogues (1, fXa Ki=0.3 nM, fII
Ki>5000 nM, Trp Ki=3 nM) were realized. Previous
work towards a less rigid inhibitor molecule involved
replacement of the naphthylene ring with either a
biphenyl or propenylbenzene ring.3 Replacement of
naphthylamidine with either biphenylamidine or pro-
penylbenzamidine resulted in a minor loss of fXa
potency and fIIa selectivity with a slight increase in
trypsin selectivity (see Table 1). Replacement of
With our pyridine-based fXa inhibitors, introduction of
a hydroxy or an amino group para to the amidine
enhances both inhibitor potency and selectivity. Table 3
lists inhibitors with and without the para-hydroxyl
group. The hydroxy group improves fXa activity and
selectivity to afford sub-nanomolar inhibitors (4d vs 5a).
*Corresponding author. Tel.: +1-510-669-4065; fax: +1-510-669-
4310; e-mail: william_guilford@berlex.com
Figure 1. Replacement of benzimidazole with an aminophenol.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00142-7

1308
S. Wu et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1307–1310
Similar effects are also observed with acylated aniline
analogues such as the ester (5b vs 5c) or the corre-
sponding acid (4e vs 5d) which showed 51,000-fold
selectivity against thrombin and 2800-fold selectivity
against trypsin. The binding of the aminophenol tem-
plate is described by the X-ray crystal structure of 5a in
bovine trypsin (Fig. 2).7 The propenylbenzamidine
moiety binds to the aspartic acid residue (D189) in the
S1 pocket and the alkylamidine group binds to the S4
pocket; a binding mode analogous to DX-9065a on
both trypsin and fXa. The para-hydroxy group forms a
hydrogen bond to the Og of Ser195 of the catalytic triad
through a bridging water molecule.
Table 1. Replacement of naphthylamidine
In general, dibasic fXa inhibitors have demonstrated
poor oral availability. Methods to improve the oral
availability of aminophenol-based fXa inhibitors by
R
fXa Ki (nM)a fIIa Ki (nM) Trp Ki (nM)
2
0.08
0.17
0.27
0.07
1960
400
7
42
27
5
3a
3b
3c
2900
1550
aKi values for these competitive inhibitors are averaged from multiple
determinations (n52) and the standard deviations are <30% of the mean.
Figure 2. X-ray structure of 5a in trypsin.
Table 2. Bioassay results for substitution effect on the phenyl ring and the aniline nitrogen
No.
R
R1
fXa Ki (nM)a
fIIa Ki (nM)
Trp Ki (nM)
PT (mM)b
TT (mM)c
4a
4b
4c
4d
4e
4f
H
F
CO2Et
CF3
CF3
CF3
H
H
H
H
7.5
4.1
6.8
0.88
0.28
0.56
5000
4700
480
2140
5000
3260
80
70
40
50
110
10
NDd
0.73
2.6
0.80
0.75
0.52
ND
240
15
220
540
200
COCH2CH2CO2H
COPh-3,4-di-CO2H
aKi values for these competitive inhibitors are averaged from multiple determinations (nꢀ2) and the standard deviations are <30% of the mean.8
bConcentration of inhibitor required to double the prothrombin based clotting time in human plasma.
cTwo-fold increase in thrombin clotting time.
dNot determined.
Table 3. Bioassay results on hydroxy substitution effect on the aryl amidine ring
Entry
R1
R2
fXa Ki (nM)a
fIIa Ki (nM)a
Trp Ki (nM)a
Trp/fXa selectivity
PTb(mM)
TTc (mM)
5a
5b
5c
5d
H
OH
H
OH
OH
0.10
0.55
0.07
0.10
1100
1000
2280
5140
200
76
230
280
2000
140
3300
2800
1.05
0.87
0.8
210
32
265
174
COCH2CH2CO2Et
COCH2CH2CO2Et
COCH2CH2CO2H
0.54
aKi values for these competitive inhibitors are averaged from multiple determinations (n52) and the standard deviations are <30% of the mean.8
bConcentration of inhibitor required to double the prothrombin based clotting time in human plasma.
cTwo-fold increase in thrombin clotting time.

S. Wu et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1307–1310
Table 4. Bioassay results upon replacement of the alkylamidine group
1309
Entry
R1
R2
fXa Ki (nM)a
FIIa Ki (nM)
Trp Ki (nM)
PTb (mM)
TTc (mM)
6a
6b
6c
6d
6e
6f
MeCO
NH2CO
Ms
H
H
H
H
18.5
7.3
44
0.58
3.0
0.34
0.28
480
250
890
1820
>5500
>5000
>5000
830
520
1250
630
891
870
765
21
151
ND
ND
156
173
ND
ND
NDd
ND
0.98
3.0
Me
MeCO
CH2CO2H
Me
COCH2CH2CO2H
COCH2CH2CO2H
COCH2CH2CO2H
ND
ND
6g
aKi values for these competitive inhibitors are averaged from multiple determinations (n52) and the standard deviations are <30% of the mean.8
bConcentration of inhibitor required to double the prothrombin based clotting time in human plasma.
cTwo-fold increase in thrombin clotting time.
dNot determined.
Scheme 1. Reagents and conditions: (i) N-t-butoxycarbonyl-4-hydroxyl-piperidine, DEAD, PPh3; (ii) NaCNBH3; (iii) (a) HCl(g), EtOH; (b) NH3,
EtOH; (iv) MeC(NH)OEt, Et3N, MeOH; (v) EtOOCCH2CH2COCl, Et3N; (vi) (a) LiOH, MeOH, H2O; (vii) AcCl, Et3N, CH2Cl2 or BrCH2COOH,
CH2Cl2, Et3N or MeI, Et3N, CH2Cl2 or HCO2H, NaCNBH3.
reducing the overall basicity of the molecule include
addition of a carboxylic acid group, as done with our
pyridine-based fXa inhibitors, or elimination of a basic
group. The alkyl amidine non-basic and less-basic
replacements we explored are listed in Table 4. Acet-
amide, urea, or methane-sulfonamide groups afforded
inhibitors 10- to 400-fold less potent than the alkyl
amidine (5a vs 6a–c and 4e vs 6e). N-Methyl or N-car-
boxymethyl substituents had minimal effect on
potency and selectivity (5a vs 6d and 5d vs 6f, 6g).
Unfortunately, all of the analogues in the aminophe-
nol template exhibited poor oral availability when dosed
in dogs.
All compounds listed in Tables 2–4 were prepared by
the general routes shown in Scheme 1. 2-Substituted

1310
S. Wu et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1307–1310
aminophenol 7 was coupled with N-t-butoxycarbonyl-4-
hydroxypiperidine under Mitsunobu conditions to
afford intermediate 8. (route A). Reaction of aniline 8
with aldehyde 9, under reductive amination conditions,
affords 10, which could be converted to the alkyl ami-
dine using previously described conditions2 to obtain
4a–d and 5b. The synthesis of carboxylate-containing
inhibitors is shown in Scheme 1, route B. Acylation of
aniline 10 with either ethyl succinyl chloride or trimel-
litic anhydride chloride (intermediate to prepare 4f, not
shown) gave intermediate 11. Conversion of the nitrile
to the corresponding amidine with concomitant loss of
the para-methoxylbenzyl- and the Boc-group gave
intermediate 12. The piperidine nitrogen of intermediate
12 can be selectively alkylated with either ethyl acet-
imidate, iodomethane, or methyl bromoacetate to pre-
pare analogues shown in Tables 2–4. Alternatively
intermediate 12 can be acylated with acetyl chloride
or mesyl chloride to afford the acylated inhibitors in
Table 4.
References and Notes
1. Fevig, J. M.; Wexler, R. R. Annu. Rep. Med. Chem. 1999,
34, 81.
2. Zhao, Z.; Arnaiz, D.; Griedel, B.; Sakata, S.; Dallas, J.;
Whitlow, M.; Trinh, L.; Post, J.; Liang, A.; Morrissey, M.;
Shaw, K. Bioorg. Med. Chem. Lett. 2000, 10, 963.
3. Shaw, K.; Guilford, W. J.; Griedel, B.; Sakata, S.; Trinh,
L.; Xu, W.; Wu, S.; Zhao, Z; Morrissey, M. M. Bioorg. Med.
Chem. Lett. 2002, 12, 1311.
4. (a) Light, D. R.; Guilford, W. J. Curr. Top. Med. Chem.
2001, 1, 121. (b) Phillips, G. B.; Buckman, B. O.; Davey, D. D.;
Eagen, K. E.; Guilford, W. J.; Hinchman, J.; Ho, E.; Koo-
vakkat, S.; Liang, A.; Light, D. R.; Mohan, R.; Ng, H. P.;
Post, J. M.; Shaw, K. J.; Smith, D.; Subramanyam, B.; Sulli-
van, 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.
5. Arnaiz, D.; Chou, Y. L.; Griedel, B.; Sakata, S.; Shaw, K.;
Zhao, Z; Koovakkat, S.; Whitlow, M.; Liang, A.; Trinh, L.;
Hinchman, J.; Post, J.; Ho, E.; Subramanyam, B.; Vergona,
R.; Walters, J.; White, K.; Sullivan, M.; Morrissey, M.
Abstract of Papers, 215th National Meeting of the American
Chemical Society, Dallas, TX; American Chemical Society:
Washington, DC, 1998; Med Chem 129.
6. For structurally related fXa inhibitor, see: Taniuchi, Y.;
Sakai, Y.; Hisamichi, N.; Kayama, M.; Mano, Y.; Sato, K.;
Hirayama, F.; Koshio, H.; Matsumoto, Y.; Kawasaki, T.
Thromb. Haemost. 1998, 79, 543.
In summary, the aminophenol template is an easily
accessible scaffold for the development of potent and
selective fXa inhibitors. The aniline nitrogen can be
substituted with either a naphthylamidine, biphenyl-
amidine or propenyl-benzamidine group with minimal
change in vitro activity. A para-hydroxy substituent on
the propenylbenzamidine ring yields subnanomolar fXa
inhibitors with 50,000- and 3000-fold selectivity against
thrombin and trypsin, respectively. The potency and
selectivity of the aminophenol-based inhibitors can be
modulated with substituents on the piperidine nitrogen.
The potency of the inhibitors can be partially explained
with the aid of a trypsin complex crystal structure.
7. 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 Cryst. D
1999, 55, 1395.
8. (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.