Enantioselective synthesis of an axially chiral 1,7-naphthyridine-6-carboxamide derivative having potent antagonist activity at the NK1 receptor

Article abstract of DOI:10.1039/a805333b

A new and highly potent NK₁ antagonist, (aR,9R)-3 [(aR,9R)-7-[3,5-bis(trifluoromethyl)benzyl]-8,9,10,11-tetrahydro-9-methyl-5-(4- methylphenyl)-7H-[1,4]diazocino[2,1-g][1,7]naphthyridine-6,13-dione], was atropdiastereoselectively synthesized in

Full text of DOI:10.1039/a805333b

View Article Online / Journal Homepage / Table of Contents for this issue
Enantioselective synthesis of an axially chiral 1,7-naphthyridine-6-carboxamide
derivative having potent antagonist activity at the NK1 receptor
Yoshinori Ikeura,a Takenori Ishimaru,b Takayuki Doi,a Mitsuru Kawada,c Akira Fujishimaa and
Hideaki Natsugari*a†
a Pharmaceutical Research Division, b Discovery Research Division and c Technology Development Department, Takeda
Chemical Industries, Ltd., 2-17-85, Jusohonmachi, Yodogawa-ku, Osaka 532-8686, Japan
A new and highly potent NK1 antagonist, (aR,9R)-3
[(aR,9R)-7-[3,5-bis(trifluoromethyl)benzyl]-8,9,10,11-tetra-
hydro-9-methyl-5-(4-methylphenyl)-7H-[1,4]diazocino[2,1-
g][1,7]naphthyridine-6,13-dione], was atropdiastereoselec-
tively synthesized in good yield by cyclization of the chiral
intermediate 6b.
racemization only after storage at 50 °C for ca. 70 h. Among
these four isomers of 1, the active isomer was shown to be
(trans,aR)-1. From a practical perspective, however, separation
of the active isomer (trans,aR)-1 is difficult, and further studies
using trans-1 as a racemate would encounter difficulties,
especially at the stage of pharmaceutical development.
Thus, in the search for new compounds with an improved
stereochemical profile, we designed cyclic analogues of 1, and
analogues with an eight-membered ring (e.g. 2 and 3) (Scheme
2) became target molecules based on the results of conforma-
tional studies on trans-1. Here we describe the atropdiaster-
eoselective synthesis of the potent NK1 receptor antagonist
In our preceding papers,1 we described the discovery1a,b and
stereochemical characterization1c of the potent NK1 antagonist2
1
{N-[3,5-bis(trifluoromethyl)benzyl]-7,8-dihydro-N,7-di-
methyl-5-(4-methylphenyl)-8-oxo-1,7-naphthyridine-6-carbox-
amide}. Since 1 has a tertiary carboxamide group at the
sterically hindered C6-position, it exhibits two notable ster-
eochemical properties (Fig. 1). First, the trans and cis amide
conformational isomers (rotamers) of 1 are separable at room
temperature;1a the compound isolated by conventional work-up
is the trans-isomer (trans-1), while the thermodynamically
unstable cis-isomer (cis-1) can also be isolated as a minor
product by careful separation procedures. Both isomers inter-
convert and reach an equilibrium state of a ca. 7:1 ratio in
solution. Second, trans-1 and cis-1 exist as a mixture of two
separable and stable enantiomers [(trans,aR)-1, (trans,aS)-1
and (cis,aR)-1, (cis,aS)-1, respectively]3 arising from restricted
rotation around the C6–C(O) bond (Fig. 1).1c Such ster-
eoisomerism due to restricted rotation is known as atropisomer-
ism among biaryl compounds and some sterically hindered
aromatic carboxamides.4 The atropisomers, (trans,aR)-1 and
(trans,aS)-1, which were separated by preparative high per-
formance liquid chromatography (HPLC) using a chiral col-
umn,‡ have significant stability in solution; e.g. they were not
interconverted in DMSO at 37 °C for 16 h and underwent
(aR,9R)-3
{(aR,9R)-7-[3,5-bis(trifluoromethyl)benzyl]-
8,9,10,11-tetrahydro-9-methyl-5-(4-methylphenyl)-7H-[1,4]-
diazocino[2,1-g][1,7]naphthyridine-6,13-dione} by cyclization
of the chiral intermediate 6b.
General synthesis of tricyclic analogues of 1 (2 and 3) is
outlined in Schemes 1 and 2. The key intermediates 6a–c
[7-(hydroxyalkyl)-1,7-naphthyridine-6-carboxamide deriva-
tives] were prepared from the pyrano[3,4-b]pyridine-6-car-
boxylic acid 4.5 First, the acid 4 was amidated with 3,5-bis(tri-
fluoromethyl)benzylamine via the acid chloride to provide the
amide. Treatment of the amide with the appropriate amines
5a–c,6 followed by deprotection with TsOH and dehydration
with DBU gave 6a–c (Scheme 1). Cyclization of 6a–c was
accomplished by mesylation followed by treatment with NaH in
THF to give 2 and 3 in good yields (Scheme 2).
For the atropisomers in the cyclic analogues of 1 (2 and 3), we
initially supposed that the flipping of this new ring would be too
rapid to enable the separation of stable isomers at room
temperature.7 However, chiral HPLC analysis of 2 showed two
peaks at room temperature, and 2 was separated by preparative
HPLC using a chiral column to give the atropisomers (aR)-2 and
(aS)-2, which have opposite [a]D values (+45.6 and 241.3,
respectively) and show considerable stability in solution; e.g.
they are gradually interconverted in DMSO to ca. 70% ee at
37 °C over 40 h and undergo racemization after storage at 50 °C
for ca. 60 h. These results indicate that 2 exists as a racemate,
making development as a clinical candidate difficult. Thus, we
O
O
Ar
Me
O
Me
O
N
N
Me
N
N
6
N
N
5
Me
Ar
trans-1
cis-1
Me
Me
O
R
O
O
O
OH
N
N
*
Me
R
Me
N
Me
N
Me
O
N
N
N
S
i–v
R
6
N
N
CO2H
H2N
CONH
5
O-THP
*
O
Ar
Ar
O
CF3
5
6
F3C
4
Me
Me
(trans, aS )-1
(trans, aR)-1
Me
Me
* = —
* = R
* = S
5,6a R = H,
b R = Me,
c R = Me,
(cis, aS)-1
(cis, aR)-1
Scheme 1 Reagents and conditions: i, SOCl2, THF, reflux, 1.5 h; ii,
3,5-bis(trifluoromethyl)benzylamine, Et3N, THF, room temp., 0.5 h (71%
from 4); iii, 5, THF–MeOH, room temp., 16 h ; iv, DBU, toluene–MeCN,
reflux, 1 h; v, TsOH, MeOH, room temp., 0.5 h
Ar = 3,5-bis(trifluoromethyl)phenyl
Fig. 1 The stereoisomers of 1
Chem. Commun., 1998
2141

View Article Online
O
O
around C5a–C6(O)–N7–CH2Ar is the important factor for
receptor recognition.
In summary, the axially chiral compound (aR,9R)-3 and its
enantiomer (aS,9S)-3 were atropdiastereoselectively synthe-
sized by cyclization of the chiral intermediates 6b and 6c,
respectively. Compound (aR,9R)-3 exhibited excellent NK1
antagonistic activities both in vitro and in vivo.
N
N
i, ii
N
N
6a
S
R
N
N
O
O
Ar
Ar
(aR)-2
(aS)-2
Me
Me
The authors thank Mr T. Tanaka for conformational analysis,
Ms F. Kasahara for NMR analysis, Mr I. Kamo for in vivo
screening and Mr Y. Tajima for in vitro screening.
O
O
Me
Me
N
N
N
i, ii
N
R
9
R
6b
S
R
N
N
7
ca. 98 : 2
Notes and References
O
O
Ar
Ar
† E-mail: natsugari_hideaki@takeda.co.jp
(aR, 9R)-3
(aS, 9R)-3
‡ Chiralpack AD, DAICEL Chemical Industries, Ltd., Japan.
§ The NK1 antagonistic activities were measured in vitro for inhibition of
Me
Me
(aS, 9S)-3
i, ii
[
125I] Bolton-Hunter-SP binding in human IM-9 cells (ref. 9) and in vivo for
(aR, 9S)-3
6c
inhibition of capsaicin-induced plasma extravasation in the trachea of
guinea pigs (ref. 10).
ca. 98 : 2
Ar = 3,5-bis(trifluoromethyl)phenyl
¶ Since the purity of (aS,9R)-3 is 98.6% de [i.e. it contains ca.1% of the
active isomer (aR,9R)-3], its intrinsic antagonistic activities may be lower
than those observed.
Scheme 2 Reagents and conditions: i, MsCl, Et3N, THF, 0 °C, 0.5 h; ii,
NaH, THF, reflux, 1 h [isolated yields from 6: 79% for 2, 69% for (aR,9R)-3
and 66% for (aS,9S)-3]
1 (a) H. Natsugari, Y. Ikeura, Y. Kiyota, Y. Ishichi, T. Ishimaru, O. Saga,
H. Shirafuji, T. Tanaka, I. Kamo, T. Doi and M. Otsuka, J. Med. Chem.,
1995, 38, 3106; (b) Y. Ikeura, T. Tanaka, Y. Kiyota, S. Morimoto, M.
Ogino, T. Ishimaru, I. Kamo, T. Doi and H. Natsugari, Chem. Pharm.
Bull., 1997, 45, 1642; (c) Y. Ikeura, Y. Ishichi, T. Tanaka, A. Fujishima,
M. Murabayashi, M. Kawada, T. Ishimaru, I. Kamo, T. Doi and H.
Natsugari, J. Med. Chem., in the press.
2 Recent review articles for tachykinin antagonists: S. MacLean, Med.
Res. Rev., 1996, 16, 297; J. A. Lowe III, Med. Res. Rev., 1996, 16,
527.
next designed the C9 methyl analogues of 2 (i.e. 3) as target
compounds, expecting asymmetric induction from the C9 chiral
center to obtain the desirable chirality arising from atropisomer-
ism, and achieved the stereoselective synthesis of the atropi-
somer by cyclization of an intermediate with a chiral methyl
group, 6b (Scheme 2). The product ratio of the atropisomer
(aR,9R)-3 to its isomer (aS,9R)-3 was ca. 98:2, and a single
recrystallization step gave (aR,9R)-3 with > 99% de. The minor
isomer (aS,9R)-3, with 98.6% de, was isolated as a powdery
substance by repeated preparative HPLC at 0 °C using the
mother liquor. Both atropisomers, (aR,9R)-3 and (aS,9R)-3,
were found to be gradually interconverted in solution to reach
the same equilibrium state [(aR,9R)-3:(aS,9R)-3 = ca. 98:2]
(e.g. in EtOH at 37 °C in ca. 60 h).
3 The italic letter a before the corresponding R and S denotes an axial
chirality as suggested by Cahn et.al.; R. S. Cahn, C. Ingold and V.
Prelog, Angew. Chem., Int. Ed. Engl., 1966, 5, 385.
4 As for synthesis and separation of atropisomers of some aromatic
carboxamides, see S. Thayumanavan, P. Beak and D. P. Curran,
Tetrahedron Lett., 1996, 37, 2899; J. Clayden, N. Westlund and F. X.
Wilson, Tetrahedron Lett., 1996, 37, 577; J. Clayden, J. H. Pink and
S. A. Yasin, Tetrahedron Lett., 1998, 39, 105; A. Ohno, M. Kashiwagi,
Y. Ishihara, S. Ushida and S. Oka, Tetrahedron, 1986, 42, 961.
5 H. Natsugari, T. Ishimaru, T. Doi, Y. Ikeura and C. Kimura, Eur. Pat.
Appl., EP733632, 1996 (Chem. Abst., 1997, 126, 8145).
6 K. Mori, Tetrahedron, 1983, 39, 3107; H. Mattes, K. Hamada and C.
Benezra, J. Med. Chem., 1987, 30, 1948.
7 As for atropisomerism in seven-membered heterocycles, see N. W.
Gilman, P. Rosen, J. V. Earley, C. Cook and L. J. Todaro, J. Am. Chem.
Soc., 1990, 112, 3969.
Single crystal X-ray analysis of (aR,9R)-38 revealed that the
N7–C8–C9–C10 moiety in the eight-membered ring is disposed
above the plane of the adjacent 1,7-naphthyridine ring, while
the amide oxygen (C6§O) is below the ring (i.e. aR ster-
eochemistry). The relative spatial orientation of the C9 methyl
group and the N-[3,5-bis(trifluoromethyl)benzyl] group in
(aR,9R)-3 was shown to be such that the two groups are
disposed in opposite directions. This is presumed to be a
thermodynamically stable form which is important for the high
atropdiastereoselectivity in the cyclization of 6b.
8 Crystal data for (aR, 9R)-3: C30H25F6N3O2, M
= 573.54, or-
thorhombic, 0.80 3 0.50 3 0.20 mm, a = 15.727(3), b = 22.972, c =
7.672(3) Å, V = 2771.6(10) Å3, T = 298 K, P212121 (#19), Z = 4,
m(Cu-Ka) = 9.59 cm21, reflections measured = 2386, R = 0.068.
CCDC 182/988.
9 M. A. Cascieri, E. Ber, T. N. Fong, S. Sadowski, A. Basal, C. Swain, E.
Seward, B. Frances, D. Burns and C. D. Strader, Mol. Pharmacol., 1992,
42, 458.
The enantiomer of (aR,9R)-3 [i.e. (aS,9S)-3], with > 99% de,
was similarly obtained by the cyclization of the corresponding
enantiomeric intermediate 6c followed by a single recrystalliza-
tion step (Scheme 2).
Compound (aR,9R)-3 exhibited excellent NK1 antagonistic
activities§ both in vitro (IC50 = 0.45 n ) and in vivo (ED50 =
M
4.3 mg kg21). The structure–activity relationships in the isomers
10 A. Eglezos, S. Giuliani, G. Viti and C. A. Maggi, Eur. J. Pharmacol.,
1991, 209, 277.
of 3 [for the atropisomer (aS,9R)-3: IC50 = 20 n
M
and ED50
=
26 mg kg21¶ and for the enantiomer (aS,9S)-3: IC50 = 340 n
M
and ED50 = > 300 mg kg21] indicate that the stereochemistry
Received in Cambridge, UK, 9th July 1998; 8/05333B
2142
Chem. Commun., 1998