1-(2-Naphthyl)-1H-pyrazole-5-carboxylamides as potent factor Xa inhibitors. Part 3: Design, synthesis and SAR of orally bioavailable benzamidine-P4 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2003.12.054

Using N,N-dialkylated benzamidines as the novel P4 motifs, we have designed and synthesized a class of 1-(2-naphthyl)-1H-pyrazole-5-carboxylamides as highly potent and selective fXa inhibitors with significantly improved hydrophilicity and in vitro anticoagulant activity. These benzamidine-P4 fXa inhibitors have displayed excellent oral bioavailability and long half-life.

Full text of DOI:10.1016/j.bmcl.2003.12.054

Bioorganic & Medicinal Chemistry Letters 14 (2004) 1229–1234
1-(2-Naphthyl)-1H-pyrazole-5-carboxylamides as
potent factor Xa inhibitors. Part 3: Design, synthesis and
SAR of orally bioavailable benzamidine-P4 inhibitors
Zhaozhong J. Jia,*,y Yanhong Wu, Wenrong Huang, Penglie Zhang, Yonghong Song,
John Woolfrey, Uma Sinha, Ann E. Arfsten, Susan T. Edwards, Athiwat Hutchaleelaha,
Stanley J. Hollennbach, Joseph L. Lambing, Robert M. Scarborough and
Bing-Yan Zhu*,{
Millennium Pharmaceuticals, Inc., 256 East Grand Ave., South San Francisco, CA 94080, USA
Received 19 June 2003; revised 5 December 2003; accepted 12 December 2003
Abstract—Using N,N-dialkylated benzamidines as the novel P4 motifs, we have designed and synthesized a class of 1-(2-naphthyl)-
1H-pyrazole-5-carboxylamides as highly potent and selective fXa inhibitors with significantly improved hydrophilicity and in vitro
anticoagulant activity. These benzamidine-P4 fXa inhibitors have displayed excellent oral bioavailability and long half-life.
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
inhibitors with N,N-dialkylated benzamidines as the
P4 motifs.
The design and discovery of orally active small molecule
competitive human factor Xa (fXa) inhibitors as novel
therapies for thromboembolic disorders have been a
major focus in the pharmaceutical industry.1 We have
recently reported a series of 1-(2-naphthyl)-1H-pyrazole-
5-carboxylamides as highly potent and selective fXa
inhibitors with respectable oral bioavailability and half-
life.2,3 However, compared to our benzamidine-P1 fXa
inhibitors with similar binding affinity, many of the 2-
naphthyl-P1 fXa inhibitors are much weaker in vitro
anticoagulants judged by the thrombin generation (TG)
assay in human plasma.2b,4,5 The poor in vitro anti-
coagulant activity has been believed due to their low
hydrophilicity which could lead to extensive protein
binding in plasma. We have explored a variety of
P4 strategies to increase hydrophilicity and in vitro
anticoagulant activity.2b In this communication, we
describe our discovery and SAR study of a class of highly
potent 1-(2-naphthyl)-1H-pyrazole-5-carboxylamide fXa
2. Results and discussion
Unsubstituted benzamidines have long been employed
in fXa inhibitors as binding ligands in both the S1 and
S4 pockets.1 However, most of these benzamidine com-
pounds suffer poor oral bioavailability and short half-
life, and are not good development candidates for oral
fXa inhibitors. The N,N-dialkylated benzamidine moi-
eties have been known to improve oral absorptions
from the unsubstituted benzamidine in the platelet gly-
coprotein IIb-IIIa antagonist literature.6,7 Moreover,
Berlex compound ZK-8078348 has demonstrated N-
methylimidazoline, a special form of N,N-dialkylami-
dine, to be a highly potent fXa P4 ligand. Thus, we
hypothesized that the N,N-dialkylated benzamidine-P4
1-(2-naphthyl)-1H-pyrazole-5-carboxylamides (1, 2 and
3) could be highly potent fXa inhibitors with desired
hydrophilicity and good oral PK properties. Our com-
Keywords: Factor Xa inhibitors.
* Corresponding authors. Tel.: +1-650-246-7073; fax: +1-650-246-
7776 (Z.J.J.); tel.: +1-510-402-4032 (B.-Y.Z.); e-mail: zjia@portola.
com; bzhu@qbius.com
y
Present address: Portola Pharmaceuticals, Inc., 270 East Grand
Ave., Suite 22, South San Fransisco, CA 94080, USA.
Present address: Quark Biotech, Inc., 6536 Kaiser Dr., Fremont, CA
94555, USA.
{
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.12.054

1230
Z. J. Jia et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1229–1234
Table 2. Effects of amidine substitution on fXa potency
Table 1. Effects of the aromatic moiety on fXa potency
Compd
fXa IC50 (nM)
Compd
Ar
fXa IC50 (nM)
511
4
8
22
5
6
7
377
479
18
9
777
114
22
10
11
12
13
14
8
22
puter-based modeling suggested that the fXa S4 pocket
could accommodate a variety of substituted amidines.
At physiological pH, the N,N-dialkylated amidine
would be positively charged. This charged P4 ligand not
only would increase the inhibitors hydrophilicity, but
also would increase its binding affinity through p-cation
interactions in fXa S4 pocket.
34
110
48
15
16
80
50
To examine our hypothesis, we prepared the 3-fluoro-2-
naphthyl analogues 4–8 using the simplest N,N-dimethyl-
amidine as the P4 element. As shown in Table 1, the low
nanomolar fXa inhibitory activity of compounds 7 (18
nM) and 8 (22 nM) has proved our hypothesis. It has
also been clear that the 1,4-phenyl linkage is optimal for
fXa binding affinity.
17
18
6
8
To explore the effects of P4 amidine substitution on fXa
activity, we then synthesized compounds 9–21. As listed
in Table 2, unsubstituted amidine 9 (777 nM) and N-
monosubstituted amidine 10 (114 nM) are not very
active. The N-ethyl-N-methyl substituted analogue 11
(22 nM), pyrrolidine containing analogue 17 (6 nM),
piperidine containing analogue 18 (8 nM) and N-
methylimidazoline analogue 21 (18 nM) display equal
or slightly better fXa activity than lead 8 (22 nM). All
other amidine substitutions have resulted in a loss in
fXa binding potency.
19
20
134
50
21
18
To improve fXa activity, we prepared compounds 22–25
to incorporate a trifluoromethyl group into the pyrazole.9
We also synthesized the corresponding tetrazole analo-
gues 26–29 to probe the modification of the central
nucleus.10 As summarized in Table 3, the trifluoromethyl
group improves fXa potency only for N,N-dimethyl-
amidine analogue 22 (4 nM). For the other analogues
(23–25), there is no gain by the trifluoromethyl sub-
stituent. The tetrazole analogues (26–29) are not as potent
as the corresponding 3-methylpyrazole fXa inhibitors
(8, 17, 18 and 21).
analogues (30–41) are more potent than the 3-fluoro-2-
naphthyl (8–29) and the 3-methylsulfonyl-2-naphthyl
(42–45) analogues. This trend matches what we
observed before in the 1-(2-naphthyl)-1H-pyrazole-5-
carboxylamide-based fXa inhibitors with a variety of
other P4 motifs.2b For the 6-chloro-2-naphthyl analogues
with N,N-dialkylated amidines in P4 (34–37), the tri-
fluoromethyl substituent in the pyrazole scaffold has
enhanced the fXa inhibitory activity.9 Meanwhile, the
displacement of the 3-methylpyrazole nucleus by tetra-
zole has little effect on fXa activity for the 6-chloro-2-
naphthyl analogues (38–41).
To survey the effects of the 2-naphthyl substitution
along with nucleus modification on fXa potency, we
prepared compounds 30–45. As shown in Table 4, in
terms of fXa inhibitory activity, the 6-chloro-2-naphthyl
The Ki value and in vitro anticoagulant activity of the
leading fXa inhibitors disclosed above are summarized
in Table 5. Interestingly, by fXa inhibitory activity, tri-

Z. J. Jia et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1229–1234
1231
Table 3. Effects of central nucleus modification on fXa potency
Table 4. Effects of 2-naphthyl substitution and nucleus modification
on fXa potency
Compd
fXa IC50 (nM)
Compd
Z3
H
Z6
Cl
fXa IC50 (nM)
22
3
30
5
23
24
10
11
31
32
33
34
35
36
37
38
39
40
H
H
H
H
H
H
H
H
H
H
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
7
7
8
1
2
4
2
2
7
8
25
26
31
16
27
28
24
76
29
110
fluoromethyl substituted compounds 34 (IC50 1 nM; Ki
0.9 nM) and 37 (IC50 2 nM; Ki 1.2 nM) are more potent
than analogues 30 (IC50 5 nM; Ki 0.7 nM) and 33 (IC50
8 nM; Ki 1.0 nM). However, in terms of fXa Ki, there is
no difference at all in regard to their potency. Overall,
this class of fXa inhibitors has displayed good to excellent
in vitro anticoagulant activity in our human plasma TG
assays. As expected, the 6-chloro-2-naphthyl analogues
are the most active, judged by both fXa IC50 and Ki
values. But, due to relatively lower hydrophilicity, their
strong fXa potency has not been translated into the
desired in vitro anticoagulant activity. Despite of rela-
tively weaker activity (fXa IC50 and Ki), the more
hydrophilic 3-methylsulfonyl-2-naphthyl analogues 42
(2ꢀTG 0.53 mM) and 45 (2ꢀTG 0.78 mM) are however
the most potent in vitro anticoagulants. Moreover, it
seems that the trifluoromethyl substituent on the pyra-
zole plays a detrimental effect on the in vitro anti-
coagulant activity. For example, judged by the TG
assay outcome, inhibitors 34 (2ꢀTG 1.8 mM) and 37
(2ꢀTG 4.5 mM) are less active in vitro anticoagulants
than 30 (2ꢀTG 0.83 mM) and 33 (2ꢀTG 0.91 mM), even
though their Ki values are about equal. Furthermore,
the tetrazole nucleus is also disadvantageous to in vitro
anticoagulant activity, as demonstrated by inhibitor 38
(Ki 0.7 nM; 2ꢀTG 2.5 mM).
41
42
H
Cl
H
15
12
SO2Me
43
44
45
SO2Me
SO2Me
SO2Me
H
H
H
15
28
14
The enzyme selectivity and rat PK profiles11 of four
representative fXa inhibitors (8, 21, 22 and 30) are

1232
Z. J. Jia et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1229–1234
Table 5. The in vitro anticoagulant activity results of the leading
N,N-dialkylated benzamidine-P4 fXa inhibitors
Table 6. The selectivity and PK profiles of four fXa inhibitors
Compd
8
21
22
30
Compd
Z3
F
Z6
H
fXa IC50 fXa Ki 2ꢀTG
Z3
Z6
Y
F
H
CH3
F
H
CH3
F
H
CF3
H
Cl
CH3
(nM)
(nM)
(mM)
8
22
3.9
1.0
fXa IC50 (nM)
22
>10
>10
>10
>10
>10
>10
18
>10
>10
>10
>10
>10
>10
3
>10
>10
>10
>10
>10
>10
5
>10
>10
>10
>10
>10
>10
Thrombin IC50 (mM)
Trypsin IC50 (mM)
t-PA IC50 (mM)
aPC IC50 (mM)
Plasmin IC50 (mM)
Kallikrein IC50 (mM)
17
18
F
F
H
H
6
8
4.1
5.2
1.4
1.5
F(%)
24.3
3.1
49.4
54.3
43.9
1.8
7.6
49.3
46.9
13.8
18.0
82.5
55.9
21
22
30
33
34
37
38
42
45
F
H
H
18
3
8.0
0.7
0.7
1.0
0.9
1.2
0.7
2.8
3.9
1.7
T1/2 (IV) (h)
Vd (L/kg)
CL (mL/min/kg)
8.8
11.0
14.7
F
H
1.4
in the rabbit deep vein thrombosis (DVT) model.12
Infusion of fXa inhibitor 22 produced a 51% inhibition
of thrombus accretion at a plasma level of 4.6 mM and
an ex vivo prothrombin time (PT) change of 1.53-fold.
At a plasma concentration of 3.9 mM, 20% inhibition
and 1.43-fold ex vivo PT change were found. Infusion of
fXa inhibitor 30 generated a 56% inhibition of thrombus
accretion at a plasma level of 4.5 mM and an ex vivo PT
change of 1.66-fold. At a plasma concentration of 3.7
mM, 20% inhibition and 1.56-fold ex vivo PT change
were produced.
Cl
Cl
Cl
Cl
Cl
H
5
0.83
0.91
1.8
H
8
H
1
H
2
4.5
H
2
2.5
Computer-based docking studies using GOLD13 as well
as both published and in-house fXa crystal data sug-
gested a reasonable binding mode hypothesis for this
class of N,N-dialkylated benzamidine-P4 fXa inhibitors.
As shown in Figure 1 for compound 30 (fXa Ki 0.7 nM),
the 6-chloro-2-naphthyl P1 moiety fits in the fXa S1
pocket and points into the aromatic ring of Tyr228 at the
bottom of the pocket. The N,N-dimethylamidine is posi-
tively charged at physiological pH. This positive charge
facilitates the p-cation stacking interaction of the benzami-
dine with the aromatic functionalities in the S4 pocket.
SO2Me
SO2Me
12
14
0.53
0.78
H
displayed in Table 6. Like our other 2-naphthyl-P1 fXa
inhibitors,2 the above N,N-dialkylated benzamidine 1-
(2-naphthyl)-1H-pyrazole-5-carboxylamides have excellent
fXa selectivity (>1000-fold by IC50) against thrombin,
trypsin, t-PA, aPC, plasmin and kallikrein. All of these
four compounds have excellent oral bioavailability and
long half-life. In regard to the P4 modification, the N-
methylimidazoline group (8) gives higher bioavailability
and lower volume distribution than the N,N-dimethyl-
amidine group (21). In regard to the pyrazole nucleus
modification, the trifluoromethyl substituent (22) provides
considerable improvement for all PK parameters as we
expected. As for the 6-chloro-2-naphthyl-P1 analogue
(30), it displays a very long half-life, a relatively large
volume distribution and fairly good bioavailability.
In conclusion, by using N,N-dialkylated benzamidines as
the P4 motifs, we have discovered a class of 1-(2-naphthyl)-
1H-pyrazole-5-carboxylamides as highly potent and
selective fXa inhibitors with excellent oral bioavailablity,
long half-life, and excellent in vitro anticoagulant potency.
Our further optimization on the 2-naphthyl P1 moiety using
two substitution groups will be reported in due course.
3. Chemistry
To evaluate our fXa inhibitors in vivo, we examined
compounds 22 (2ꢀTG 1.4 mM) and 30 (2ꢀTG 0.83 mM)
The preparation of the N,N-dialkylated benzamidine-P4
fXa inhibitors is illustrated by the synthesis of

Z. J. Jia et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1229–1234
1233
Scheme 2. (a) (1) NaNO2 (1.1 equiv), concd HCl, 0 ꢁC, 30 m; (2)
Figure 1. Computer modeling of compound 30 docked in fXa active
enzyme pocket.
ꢁ
.
SnCl2 2H2O (3 equiv), concd HCl, 0 C, 1 h; filtration; (b) THF/HOAc
(1:2), reflux, 2 h; (c) KMnO4 (5 equiv), acetone/water (2:1), 60 ꢁC, 3 h;
(d) EtO2CCOCl (1.2 equiv), Et3N (5 equiv), DCM, rt; (e) Ph3P (2
equiv), CCl4, reflux, overnight; (f) NaN3 (1.2 equiv), MeCN, rt, over-
night; (g) CuCl (10 equiv), CuI (1 equiv), DMF, reflux; (h) (1) DPPA
(1.5 equiv), Et3N (1.5 equiv), DMF, rt, 3 h; (2) water, rflx, 1 h.
compounds 8 and 21. Ethyl 1-(3-fluoro-2-naphthyl)-3-
methyl-1H-pyrazole-5-carboxylate (47) was prepared
from 3-fluoro-2-naphthylamine (46) as described in our
previous communication.2a Compound 48, assembled
via a Weinreb reaction, was employed to produce 8 and
21 in two routes through methyl imidate 49 and methyl
thioimidate 50.
oxalate to give 53, which was converted to 54 by reacting
with triphenylphosphine in carbon tetrachloride. Com-
pound 54 then reacted with sodium azide to smoothly
produce building block 55 in high yield. Ethyl 1-(6-
chloro-2-naphthyl)-1H-tetrazole-5-carboxylate (59) w as
synthesized similarly from 6-chloro-2-naphthylamine
(58), which had been prepared from commercial 6-
bromo-2-naphthoic acid (56) in two steps. The synthesis
of ethyl 1-(6-chloro-2-naphthyl)-3-methyl-1H-pyrazole-
5-carboxylate and ethyl 1-(3-methylsulfonyl-2-naph-
thyl)-3-methyl-1H-pyrazole-5-carboxylate have been
reported.2 All the corresponding N,N-dialkylated benz-
amidine-P4 fXa inhibitors were made using the chem-
istry exemplified in Scheme 1.
As shown in Scheme 2, the synthesis of 1-(3-fluoro-2-
naphthyl)-3-trifluoromethyl-1H-pyrazole-5-carboxylic
acid (52) was accomplished through oxidative cleavage
of the furyl group in compound 51. To prepare ethyl
1-(3-fluoro-2-naphthyl)-1H-tetrazole-5-carboxylate (55),
3-fluoro-2-naphthylamine (46)2a was treated with ethyl
Acknowledgements
We would like to thank Jaya Kothule, Paul Wong,
Brian Huang, Bridget May and Shannon Poole for
performing the biological assays. We would like to
thank Keith Abe and Gail Siu for performing the rabbit
DVT experiment.
References and notes
1. (a) Zhu, B.-Y.; Scarborough, R. M. Annu. Rep. Med.
Chem. 2000, 35, 83. (b) Zhu, B.-Y.; Scarborough, R. M.
Curr. Opin. Car. Pulmon. Renal Invest. Drugs 1999, 1, 63.
(c) Rai, R.; Spengeler, P. A.; Elrod, K. C.; Young, W. B.
Curr. Med. Chem. 2001, 8, 101. (d) Sanderson, P. E. J.
Annu. Rep. Med. Chem. 2001, 36, 79. (e) Walenga, J. M.;
Jeske, W. P.; Hoppensteadt, D.; Fareed, J. Curr. Opin.
Invest. Drugs 2003, 4, 272.
Scheme 1. (a) (1) NaNO2 (1.1 equiv), concd HCl, 0 ꢁC, 30 m; (2)
ꢁ
.
SnCl2 2H2O (3 equiv), concd HCl, 0 C, 1 h; filtration; (b) THF/HOAc
(1:2), reflux, 2 h; (c) Me3Al (2M in hexane, 5 equiv), DCM, overnight;
(d) dry MeOH, saturated with HCl (g), 0 ꢁC to rt, overnight; (e) Et3N/
dioxane (1:10), saturated with H2S (g), rt, overnight; (f) MeI (10
equiv), acetone, reflux, 1–2 h; (g) Me2NH (2M in MeOH, 5 equiv), dry
MeOH, reflux, 1 h; (h) MeNHCH2CH2NH2 (3 equiv), dry MeOH,
reflux, 1 h; (i) Me2NH (2M in MeOH, 3 equiv), HOAc (5 equiv), dry
MeOH, reflux, 1 h; (j) MeNHCH2CH2NH2 (3 equiv), HOAc (8 equiv),
dry MeOH, reflux, 1 h.
2. (a) For precedent work, see: Jia, Z. J.; Wu, Y.; Huang,
W.; Goldman, E.; Zhang, P.; Woolfrey, J.; Wong, P.;

1234
Z. J. Jia et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1229–1234
Huang, B.; Sinha, U.; Park, G.; Reed, A.; Scarborough,
8. Phillips, G. B.; Buckman, B. O.; Davey, D. D.; Eagen,
K. A.; Guilford, W. J.; Hinchman, J.; Ho, E.; Koovakkat,
S.; Liang, A.; Light, D. R.; Kohan, 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.
9. The fXa potency enhancing effect of the trifluoromethyl
group was first demonstrated by the DuPont team. For
reference, see: Pinto, D. J.; Orwat, M. J.; Wang, S.; Fevig,
J. M.; Quan, M. L.; Amparo, E. C.; Cacciola, J.; Rossi,
K. A.; Alexander, R. S.; Smallwood, A. S.; Luettgen,
J. M.; Liang, L.; Aungst, B. J.; Wright, M. R.; Knabb,
R. M.; Wong, P. C.; Wexler, R. R.; Lam, P. Y. S. J. Med.
Chem. 2001, 44, 566.
10. A series of tetrazole nucleus-based fXa inhibitors was
recently published by the BMS team: Quan, M. L.; Ellis,
C. D.; He, M. Y.; Liauw, A. Y.; Woerner, F. J.; Alex-
ander, R. S.; Knabb, R. M.; Lam, P. Y. S.; Luettgen,
J. M.; Wong, P. C.; Wright, M. R.; Wexler, R. R. Bioorg.
Med. Chem. Lett. 2003, 13, 369.
11. These compounds were administered by IV bolus injec-
tion to conscious Sprague–Dawley rats at a dose of 0.2 or
0.4 mg/kg in a solution of 50% PEG-300. They were
administered orally at a dose of 4 or 6 mg/kg in a sus-
pension of 25% PEG-300. The plasma samples were ana-
lyzed using LC/MS/MS.
R. M.; Zhu, B.-Y. Bioorg. Med. Chem. Lett. 2002, 12,
1651. (b) Jia, Z. J.; Wu, Y.; Huang, W.; Zhang, P.; Clizbe,
L. A.; Goldman, E. A.; Sinha, U.; Arfsten, A. E.;
Edwards, S. T.; Alphonso, M.; Hutchaleelaha, A.; Scar-
borough, R. M.; Zhu, B.-Y. Bioog. Med. Chem. Lett.
2004, 14, the preceding communication. doi:10.1016/
j.bmcl.2003.12.053.
3. Part of the work described in this communication has been
previously presented at the 226th ACS National Meeting
in September 2003 in NewYork, NewYork (MEDI-32).
4. Thrombin generation in human plasma (TG) assay:
Potential fXa inhibitors are dissolved in DMSO, and
serial dilutions are made in wells of 96 well plates. Reptilase-
treated human plasma is added to the inhibitors, followed
by substrate (H-b-Ala-Gly-Arg-p-nitroanilide) and CaCl2.
The thrombin generation reaction is started by addition of
640 pmol/L of recombinant tissue factor. The absorbance is
monitored at 405 nm at 37 ꢁC for 22 min. The results are
expressed as 2ꢀTG, the concentration of an inhibitor
required to double the time of maximum thrombin genera-
tion. Aꢂ1 mM 2ꢀTG value is usually required for positively
efficacious response in our rabbit DVT model.
5. Sinha, U.; Lin, P. H.; Edwards, S. T.; Wong, P. W.; Zhu,
B.-Y.; Scarborough, R. M.; Su, T.; Jia, Z. J.; Song, Y.;
Zhang, P.; Clizbe, L. A.; Park, G.; Reed, A.; Hollenbach,
S. J.; Malinowski, J.; Arfsten, A. E. Arterios. Thromb.
Vasc. Biol. 2003, 23, 1098.
12. (a) Sinha, U.; Ku, P.; Malinowski, J.; Zhu, B.-Y.; Scar-
borough, R. M.; Marlowe, C. K.; Lin, P. H.; Hollen-
bach, S. J. Eur. J. Pharmacol. 2000, 395, 51. (b) Wong,
A. G.; Gunn, A. C.; Ku, P.; Hollenbach, S.; Sinha, U.
Thromb. Haemost. 1997, 77, 1143. (c) Hollenbach, S.;
Sinha, U.; Lin, P. H.; Needham, K.; Frey, L.; Hancock,
T. E.; Wong, A.; Wolf, D. L. Thromb. Haemost 1994, 71,
357.
6. Scarborough, R. M.; Gretler, D. D. J. Med. Chem. 2000,
43, 3453.
7. (a) Okumura, K.; Shimazaki, T.; Aoki, Y.; Yamashita,
H.; Tanaka, E.; Banba, S.; Yazawa, K.; Kibayashi, K.;
Banno, H. J. Med. Chem. 1998, 41, 4036. (b) Hayashi, Y.;
Katada, J.; Harada, T.; Tachiki, A.; Iijima, K.; Takiguchi,
Y.; Muramatsu, M.; Miyazaki, H.; Asari, T.; Okazaki, T.;
Sato, Y.; Yasuda, E.; Yano, M.; Uno, I.; Ojima, I. J.
Med. Chem. 1998, 41, 2345.
13. Jones, G.; Willet, P.; Glen, R. C.; Leach, A. R.; Taylor,
R. J. Mol. Biol. 1997, 267, 727.