Discovery of novel tetrahydroisoquinoline derivatives as potent and selective factor Xa inhibitors

Article abstract of DOI:10.1016/j.bmcl.2004.10.033

A series of novel 2,7-disubstituted tetrahydroisoquinoline derivatives were designed and synthesized. Among these derivatives, compounds 1 and 2 (JTV-803) exhibited potent inhibitory activity against FXa and good selectivity with respect to other serine proteases (thrombin, plasmin, and trypsin). In addition, compound 2 exhibited potent anti-FXa activity after intravenous and oral administration to cynomolgus monkey, and showed a dose-dependent antithrombotic effect in a rat model of venous thrombosis.

Full text of DOI:10.1016/j.bmcl.2004.10.033

Bioorganic & Medicinal Chemistry Letters 15 (2005) 185–189
Discovery of novel tetrahydroisoquinoline derivatives as
potent and selective factor Xa inhibitors
Hiroshi Ueno, Katsuyuki Yokota, Jun-ichi Hoshi, Katsutaka Yasue, Mikio Hayashi,
Itsuo Uchida, Kazuo Aisaka, Yasunori Hase, Susumu Katoh and Hidetsura Cho*
Central Pharmaceutical Research Institute, Japan Tobacco Inc., 1-1, Murasaki-cho, Takatsuki, Osaka 569-1125, Japan
Received 26 August 2004; accepted 5 October 2004
Available online 5 November 2004
Abstract—A series of novel 2,7-disubstituted tetrahydroisoquinoline derivatives were designed and synthesized. Among these deriva-
tives, compounds 1 and 2 (JTV-803) exhibited potent inhibitory activity against FXa and good selectivity with respect to other
serine proteases (thrombin, plasmin, and trypsin). In addition, compound 2 exhibited potent anti-FXa activity after intravenous
and oral administration to cynomolgus monkey, and showed a dose-dependent antithrombotic effect in a rat model of venous
thrombosis.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Activation of the intravascular coagulation system is in-
volved in a number of cardiovascular diseases such as
deep vein thrombosis (DVT), disseminated intravascular
coagulation (DIC), pulmonary embolism, ischemic
stroke, and unstable angina. Interruption of the clotting
cascade has been investigated to inhibit clot formation
and for the prevention or treatment of these thrombotic
disorders. Heparin and warfarin are the most widely
used anticoagulants for the prophylaxis and treatment
of thrombus-based diseases. However, these anticoagu-
lants have clinical limitations due to dependence on anti-
thrombin III and antagonism of vitamin K, respectively.
For instance, heparin administration is sometimes asso-
ciated with the development of an antibody to platelet
factor (PF4) that causes thrombocytopenia, while war-
farin requires a longer time for the onset of its action,
needs continual monitoring, and sometimes interacts
with food or other drugs. Factor Xa (FXa) is a tryp-
sin-like serine protease that is an important enzyme in
the clotting cascade, since it represents the confluence
of the intrinsic and extrinsic pathways. Factor Xa forms
the prothrombinase complex, together with nonenzy-
matic cofactor Va and Ca2+ on the surface phospholi-
pids of platelets or endothelial cells, and this complex
is responsible for the conversion of prothrombin to
thrombin. Then thrombin catalyzes the cleavage of
fibrinogen to fibrin, initiating a process that ultimately
leads to clot formation. Therefore, we have been inter-
ested in achieving inhibition of FXa by the development
of a novel FXa inhibitor.
Since the discovery of compound 3 (DX-9065a),1 a vari-
ety of other compounds have been reported.2 The re-
lease of fondaparin sodium in 2002, which is a
synthetic pentasaccharide, proved the clinical effective-
ness of FXa inhibitors as anticoagulants.3 This situation
prompted us to disclose our experimental results on
novel FXa inhibitors (Fig. 1).
From analysis of the binding of FXa to compound 3,4 it
has been demonstrated that the basic part of both ends
of the compound is important for inhibition of FXa. We
focused on the two basic parts for S1 and S4 sites, and
designed a fundamental skeleton (compound 4) as fol-
lows. At first, we selected an N-amidinotetrahydroiso-
quinoline ring that was expected to completely fill the
S1 pocket of FXa and to interact with Asp 189.5 Then
a 4-piperidinylmethyloxy group was selected for posi-
tion-7 of the tetrahydroisoquinoline ring as a spacer
for the introduction of basic substituents because the
nitrogen atom of the piperidinyl group could be oriented
toward the S4 site.
Keywords: Potent and selective factor Xa (FXa) inhibitor; Tetrahydro-
isoquinoline derivatives; JTV-803; Antithrombotic effect on venous
thrombosis in rats; Inhibition of human FXa.
*
Corresponding author. Tel.: +81 72 681 9700; fax: +81 72 681
9725; e-mail: hidetsura.cho@ims.jti.co.jp
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.10.033

186
H. Ueno et al. / Bioorg. Med. Chem. Lett. 15 (2005) 185–189
N
O
Me
NH
NH
NH
N
N
O
H2N
H2N
N
COOH
COOH
HCl 5H2O
3 DX-9065a
1
2
2HCl
MsOH
H
NH
2
N
1
O
H2N
N
7
4
2HCl
Figure 1.
We initially prepared compound 4, which showed mod-
erate FXa inhibitory activity (IC50 = 3.0lM),1a demon-
strating that it had sufficient potential for further studies
on structure–activity relationships. Introduction of the
basic substituents on a nitrogen atom of the piperidine
ring was carried out to obtain compounds 6–11 (Scheme
1 and Table 1).6 Among them, compound 10 with a 4-
pyridinyl group showed potent FXa inhibitory activity
(IC50 = 0.06lM) without inhibition of factor IIa
(thrombin) (>10lM). The potent activity of compound
10 might be explained by the interaction of protonated
pyridine ring with S4 site which favors positive charge7
and by hydrogen bonding of protonated nitrogen atom
of pyridine with Glu 97, as expected from the docking
simulation of the compounds with FXa.
However, most of the compounds listed in Table 1
showed fatal acute toxicity. When these compounds
OHC
OMe
HN
MeO
OMe
OMe
N
N
b
a
OMe
OH
Boc
OH
e
c, d
N
HBr
R
N
f
k
HO
R
NBoc
N
OH
Boc
O
BocHN
N
N
R=4-pyridinyl
R=quinolin-4-yl
l, m
Cbz
N
g, h
Br
R
NBoc
N
NH
NH
N
i, j
O
O
BocHN
H2N
N
4-11
Scheme 1. Synthesis of compounds 4–11. Reagents and conditions:6 (a) aminoacetal, toluene, reflux; (b) 76% H2SO4 aq, 0ꢁC; (c) H2 (3atm), PtO2
(cat.), AcOH, rt; (d) HBr aq, reflux; (e) Boc2O, NaOH, dioxane-H2O, rt; (f) 1H-pyrazole-1-(N,N0-bis-tert-butoxycarbonyl) carboxamidine, Et3N,
THE, rt; (g) NaOH, DMSO, rt; (h) H2 (3atm), 7.5% Pd/C (cat.), THF–EtOH, rt; (i) ethyl acetoimidate hydrochloride, Et3N, THF–EtOH, rt for 5;
nitro-benzyl bromide, diisopropylethylamine (DIPEA),THF–DMF, rt, then H2 (3atm), 7.5% Pd/C for 6 and 7; picolyl chloride, DIPEA THF–DMF,
rt for 8 and 9; (j) TFA, CHCl3, rt, then HCl–MeOH; (k) diisopropyl azodicarboxylate (DIPAD), PPh3, THE, rt; (l) TFA, CHCl3, rt; (m) 1H-
pyrazole-1-carboxamidine hydrochloride, DIPEA, DMF, rt, then HCl–MeOH for 10 and 11.

H. Ueno et al. / Bioorg. Med. Chem. Lett. 15 (2005) 185–189
187
Table 1. Inhibitory activity for FXa and FIIa (Thrombin),1a and acute toxicity in mice (mortality)
R1
N
NH
1
O
7
H2N
N
2
R2
a
n HCl
Compounda
R1
R2
FXa IC50 (lM)b
FIIa IC50 (lM)b
Mortality (10mg/kg, iv)
4
H
H
H
H
H
H
H
H
H
H
H
H
H
3.0
0.5
3.0
4.0
1.0
0.8
0.06
0.4
0.2
2.0
0.2
0.2
0.08
0.03
0.2
0.1
1.0
>10b
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
NTc
3/3d
NT
NT
NT
3/3
3/3
3/3
3/3
0/3
3/3
0/3
3/3
0/3
0/3
0/3
NT
5
C(@NH)Me
2-Aminobenzyl
3-Aminobenzyl
6
7
8
Pyridin-2-ylmethyl
Pyridin-4-ylmethyl
Pyridin-4-yl
9
10
11
12
13
14
15
16
1
Quinolin-4-yl
(3–CO2Me)pyridin-4-yl
(3–CO2H)pyridin-4-yl
(3–(CH2)2CO2Et)pyridin-4-yl
(3–(CH2)2CO2H)pyridin-4-yl
Pyridin-4-yl
CO2Et
CO2H
CO2H
CO2H
CO2H
Pyridin-4-yl
17
18
19
2-Methylpyridin-4-yl
1,3-Pyrimidin-4-yl
4-(1-Methylpyridinium)chloride
a Compounds 1, 4, 5, 10–18: 2HCl salt, compounds 6–9: 3HCl salt, compound 19: HCl salt.
b n = 2 (number of animals).
c Not tested.
d Number of deaths/number of animals tested.
Cbz
N
b, c
a
Boc
O
Boc
Boc
OH
N
N
N
Cbz
N
N
R
Br
N
N
NH
N
i
g, h
O
O
R'
H2N
N
R= CHO
d
12; R'= CO2Me
13; R'= CO2H
e, f
R= CO2Me
R= (CH2)2CO2Et
14; R'= (CH2)2CO2Et
15; R'= (CH2)2CO2H
i
Cbz
OH
j, k
Cbz
O
Cl
l
N
N
N
N
EtO2C
N
N
N
NH
N
m, n
Cbz
O
O
N
H2N
N
CO2Et
CO2R''
16; R''= Et
o
1; R''= H
Scheme 2. Synthesis of compounds 1 and 12–16. Reagents and conditions:6 (a) NaH, DMF, rt; (b) H2 (3atm), 7.5% Pd/C (cat.), THF–EtOH, rt; (c)
4-chloro-3-formylpyridine, Et3N, EtOH, reflux; (d) MnO2, NaCN (cat.), CHCl3–MeOH, rt; (e) (EtO)2POCH2CO2Et, NaH, THF, rt; (f) H2 (1atm),
7.5% Pd/C (cat.), THF–EtOH, rt; (g) TFA, CHCl3, rt; (h) 1H-pyrazole-1-carboxamidine hydrochloride, DIPEA, DMF, rt; (i) NaOH, MeOH–H2O,
rt, then HCl aq; (j) ClCH2SMe, NaH, DMF, rt; (k) SO2Cl2, CH2Cl2, 0ꢁC; (1) LDA, THF, ꢀ70ꢁC rt; (m) 30% HCl–EtOH, 85ꢁC; (n) 1H-pyrazole-1-
carboxamidine, NaHCO3 aq, acetone, rt; (o) c-HCl, reflux.

188
H. Ueno et al. / Bioorg. Med. Chem. Lett. 15 (2005) 185–189
were administered intravenously at a bolus dose of
10mg/kg, severe convulsions occurred immediately,
and the mice died in a few minutes. Such fatal acute toxi-
city was also observed in our previous research on other
FXa inhibitors, and was avoided by introduction of an
acidic functional group.8 Hence, we tried to introduce
a carboxyl group to the most active compound 10, as
shown in Scheme 2 and Table 1 (compounds 13, 15,
and 1). As expected, acute toxicity was suppressed when
the compounds were given carboxyl groups, while the
corresponding esters still showed severe toxicity. Loss
of toxicity after introduction of a carboxyl group at dif-
ferent positions suggested that one reason for severe toxi-
city might be the strong basicity of the compound.
Similar to compound 10, compound 1 exhibited potent
inhibitory activity for FXa (IC50 = 0.03lM). This find-
ing was explained by conformational analysis of com-
pounds 10 and 1. That is, introduction of the carboxyl
group at the tertiary carbon of the piperidinyl group
did not influence the conformation of compound 10
and the introduced hydrophilic group was oriented to-
wards the aqueous area. (Fig. 2).9 On the other hand,
the introduction of substituents (R1) at the position-3
of pyridine ring causes the change of the torsional angle
between the pyridine ring and the piperidine ring. The
loss in activity of compounds 12–15 might be due to this
fact.
100
80
60
40
20
0
1 mg/kg i.v.
10 mg/kg p.o.
0
120
240
time(min)
360
480
Figure 3. Percent inhibition of human factor Xa after intravenous and
oral administration of compound 2 to cynomolgus monkeys.10 Data
represent the mean SEM (n = 6).
2, there was selectivity of 2 for FXa relative to other ser-
ine proteases (thrombin, plasmin, and trypsin). The anti-
FXa activity of compound 2 was evaluated after intrave-
nous and oral administration to cynomolgus monkey
based on inhibition of human FXa in plasma (see Fig.
3).10 After oral administration of compound 2 at a dose
of 10mg/kg, inhibition increased to a maximum of
41 4% at 120min and then gradually declined.
The plasma concentration (Cmax) was 0.39lg/mL at
120min.11 Thus, compound 2 showed anti-FXa activity
after intravenous and oral administration to cynomol-
gus monkey. Subsequently, the antithrombotic effect of
2 was examined in a rat venous thrombosis model.12
As shown in Figure 4, compound 2 had a dose-depend-
ent antithrombotic effect at doses of 0.1–1mg/kg/h, and
the effect was statistically significant at 0.3 or 1mg/kg/h.
Low molecular weight heparin (LMWH) also showed a
dose-dependent antithrombotic effect in this model at
doses of 30–300U/kg/h.
For further pharmacological evaluation, compound 2
(MsOH salt) was used instead of compound 1 because
stable crystals could be obtained. As shown in Table
In conclusion, a series of tetrahydroisoquinoline deriva-
tives were designed and synthesized. Among them, com-
pounds 1 and 2 exhibited the most potent inhibition of
FXa. The selectivity of compound 2 (JTV-803) for
FXa relative to other serine proteases (thrombin, plas-
min, and trypsin) was demonstrated. Moreover, com-
pound 2 showed good efficacy when pharmacological
evaluation was done in a rat venous thrombosis model,
and it displayed oral activity in cynomolgus monkeys.
After further evaluation of toxicology and physical
properties, compound 2 was finally selected as a candi-
date of FXa inhibitor for clinical studies.
Figure 2. Complex model of factor Xa (green) and compound 1
(orange).
Table 2. Selectivity of compound 2 for serine proteases; inhibitory activity (Ki value: lM) for Factor Xa, Thrombin, Plasmin, and Trypsin
Compound
Factor Xa
Thrombin
Plasmin
Trypsin
2
3 (ref.)
0.019 0.001lM
0.041 0.002lM
>100
>100
78.2 2.8
23.0 0.8
13.6 1.8
0.62 0.08
The Ki values for each enzyme were determined from DixonÕs plot constructed at two substrate concentrations. Data represent the mean SEM
(n = 3).

H. Ueno et al. / Bioorg. Med. Chem. Lett. 15 (2005) 185–189
189
5
4
3
2
1
0
5
4
*
**
3
2
1
0
*
Vehicle
0.1
0.3
2
1
30
100
300
Vehicle
Compound
(mg/kg/hr)
LMWH(U/kg/hr)
Figure 4. Effect of compound 2 and LMWH on venous thrombosis in rats. (Data represent the mean SEM, **P < 0.01, *P < 0.05, DunnettÕs test,
n = 5–7).12
10. Ex vivo assessment of inhibition of human factor Xa in
cynomolgus monkey plasma––Compound 2 was dissolved
in physiological saline and was administrated intrave-
nously to male cynomolgus monkeys at a dose of 1mg/kg.
Before administration and at 5, 10, 15, 30, 60, and 120min
after administration, 1500lL blood samples were collected
from the saphenous vein of each monkey into a syringe
containing 300lL of 3.8% citric acid. Plasma was
prepared from each sample by centrifugation at 2000 · g
for 10min at 4ꢁC. Compound 2 was dissolved in deionized
distilled water and administered orally to fasting male
cynomolgus monkeys at a dose of 10mg/kg. Before
administration and at 15, 30, 60, 120, 240, 360, and
480min after oral administration, 1500lL blood samples
were collected and plasma was obtained as described
above. Forty microliters of human factor Xa (0.5U/mL)
and 40lL of a 4-fold diluted plasma sample were
incubated in 40lL of 0.1M Tris–0.2M NaCl buffer (pH
8.4) at 37ꢁC for 10min. Then, 40lL of a synthetic
substrate (S-2222, adjusted to 0.8mM) was added and the
mixture was incubated at 37ꢁC for 3min. The reaction was
stopped by addition of 60% acetic acid and the absorbance
at 405nm was measured with a spectrometer (Model 3550,
BIO-RAD, Hercules, USA). As the control, plasma
obtained prior to administration of compound 2 was
measured. Human factor Xa inhibitory activity was
calculated as the percent inhibition relative to the control.
11. Pharmacokinetic parameters in cynomolgus monkeys:
Cmax 0.39lg/mL, T1/2 3.6h, Bioavailability (B.A.) 10.7%
(10mg/kg po), Clearance 0.25L/h/kg, Volume of Distri-
bution 0.32L/kg (1mg/kg iv).
12. Assessment of the effect on venous thrombosis in rats––
Male SD rats were anesthetized with urethane (1.3g/kg
ip). About 1cm of an abdominal vein was carefully
dissected at a site below the left renal vein, and Parafilm
(Parafilm M, ANC) was placed on the dorsal aspect of the
vein. A 2 · 3mm piece of filter paper (No. 1, Whattman)
containing 25% FeCl3 was applied to the detached vein
and removed after 20min. Immediately after removing the
filter paper, a 5mm length of the abdominal vein was
resected and weighed. The thrombus weight was calcu-
lated by subtracting the weight of the vessel walls from the
total measured weight. Compound 2 (0.1, 0.3, 1mg/kg/h)
was administered by continuous intravenous infusion
from 1h prior to placement of the filter paper.
References and notes
1. (a) Hara, T.; Yokoyama, A.; Ishihara, H.; Yokoyama, Y.;
Nagahara, T. Thromb. Haemost. 1994, 71, 314; (b)
Nagahara, T.; Yokoyama, Y.; Inamura, K.; Katakura,
S.; Komoriya, S.; Yamaguchi, H.; Hara, T.; Iwamoto, M.
J. Med. Chem. 1994, 37, 1200.
2. (a) Betz, A. Exp.Opin. Ther. Patents 2001, 11, 1007; (b)
Rai, R.; Sprengeler, P. A.; Elrod, K. C.; Young, W. B.
Curr. Med. Chem. 2001, 8, 101; (c) Ewing, W. R.; Pauls,
H. W.; Spada, A. P. Drugs Future 1999, 24, 771; (d) Zhu,
B. Z.; Scarborough, R. M. Annu. Rep. Med. Chem. 2000,
35, 83; (e) Hirayama, F.; Koshio, H.; Ishihara, T.;
Watanuki, S.; Hachiya, S.; Kaizawa, H.; Kuramochi, T.;
Katayama, N.; Kurihara, H.; Taniuchi, Y.; Sato, K.;
Sakai-Moritani, Y.; Kaku, S.; Kawasaki, T.; Matsumoto,
Y.; Sakamoto, S.; Tsukamoto, S. Bioorg. Med. Chem.
2002, 10, 2597; (f) Quan, M. L.; Ellis, C. D.; He, M. Y.;
Liauw, A. Y.; Woerner, F. J.; Alexander, 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.
3. Hobbelen, P. M.; van Dinther, T. G.; Vogel, G. M. T.; van
Boeckel, C. A. A.; Moelker, H. C. T.; Meuleman, D. G.
Thromb. Haemost. 1990, 63, 265.
4. (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.
5. Kucznierz, R.; Grams, F.; Leinert, H.; Marzenell, K.;
Engh, R. A.; von der Saal, W. J. Med. Chem. 1998, 41,
4983.
6. Katoh, S.; Yokota, K.; Hayashi, M., PCT Int. Appl.
WO9952895, 1999.
7. S4 site consists of three aromatic residues (Phe174, Trp215
,
Tyr99) favorably interacts with a positive charge. Dough-
erty, D. A. Science 1996, 271, 163.
8. Ueno, H.; Katoh, S.; Yokota, K.; Hoshi, J.; Hayashi, M.;
Uchida, I.; Aisaka, K.; Hase, Y.; Cho, H. Bioorg. Med.
Chem. Lett. 2004, 14, 4281.
9. FXa-compound 1 complex model was built by docking
compound 1 into the S1 and the aryl binding site on FXa
crystal structure (PDB code 1fax). Discover and Insight II
program from Accerlys were used for energy calculation
and graphical displays, respectively.