Design and synthesis of novel 9-substituted-7-aryl-3,4,5,6-tetrahydro-2H- pyrido[4,3-b]- and [2,3-b]-1,5-oxazocin-6-ones as NK1 antagonists

Article abstract of DOI:10.1016/j.bmcl.2004.12.091

Novel 9-substituted-7-aryl-3,4,5,6-tetrahydro-2H-pyrido[4,3-b]- and [2,3-b]-1,5-oxazocin-6-ones were designed and prepared as part of a search for NK₁ antagonists. Structure-activity relationship studies indicated that the conformational restri

Full text of DOI:10.1016/j.bmcl.2004.12.091

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1479–1484
Design and synthesis of novel
9-substituted-7-aryl-3,4,5,6-tetrahydro-2H-pyrido[4,3-b]- and
[2,3-b]-1,5-oxazocin-6-ones as NK1 antagonists
Shigeki Seto,* Asao Tanioka, Makoto Ikeda and Shigeru Izawa
Discovery Research Laboratories, Kyorin Pharmaceutical Co., Ltd, 2399-1, Nogi, Nogi-machi, Shimotsuga-gun,
Tochigi 329-0114, Japan
Received 6 December 2004; accepted 28 December 2004
Available online 29 January 2005
Abstract—Novel 9-substituted-7-aryl-3,4,5,6-tetrahydro-2H-pyrido[4,3-b]- and [2,3-b]-1,5-oxazocin-6-ones were designed and pre-
pared as part of a search for NK1 antagonists. Structure–activity relationship studies indicated that the conformational restriction
resulting from the incorporation of an oxazocine ring and the presence of a terminal heteroatom on the cyclic amino group at the C-
9 position play important roles in NK1, receptor recognition.
Ó 2005 Elsevier Ltd. All rights reserved.
6
1. Introduction
nists have been reported, such as L-733060, MK-869,7
and TAK-637 (Fig. 1).8
Tachykinin, a peptide neurotransmitter that was eventu-
ally characterized as the undecapeptide Arg-Pro-Lys-
Pro-Gln-Gln-Phe-Phe-Gly-Leu-Met-NH2,1 binds to a
series of three neurokinin receptors, NK1, NK2, and
NK3, which have selective affinity for substance P
(SP), neurokinin A, and neurokinin B, respectively.2
Among them, SP3 is known to exhibit a wide variety
of biological responses both centrally and peripherally.
SP binds to NK1 receptors and has been implicated in
the transmission of pain and stress signals, inflamma-
tion, and the contraction of smooth muscle. Therefore,
NK1 antagonists may be useful in the clinic for treat-
ment of a wide range of diseases. Among them, we were
interested in the relationship between tachykinin and the
activation of micturition-related reflexes,4 with a view to
possible application in the treatment of pollakiuria and
urinary incontinence.
A common feature of high-affinity NK1 antagonists is
the presence of an intramolecular face-to-face or edge-
to-face p–p interaction between two aromatic rings,
which may be important in stabilizing the bioactive con-
formation.9 A conformationally restricted system could
increase this interaction, and in TAK-637 this was
achieved by introducing an eight-membered ring into
the naphthyridine ring. Another important pharmaco-
phore is the bridgehead basic nitrogen, which mediates
NK1 receptor recognition through ion-pair site interac-
tion with the receptor.10
In our research directed toward novel NK1 antagonists,
we were interested in the 2-[3,5-bis(trifluoromethyl)benz-
yloxy]-1-phenylethylamine fragment 5 present in the
early Merck lead L733060 (2), which is the minimum ele-
ment required for binding to the NK1 receptor.11 By
incorporating this fragment into a pyridine ring, com-
pound 6 (KB = 25.7 nM) was designed as our starting
point. However, in vitro studies in rat microsomes
showed that compound 6 was metabolically labile
(remaining ratio = 9%). Conversion of the linking group
from an oxygen atom to an N-methyl carbamoyl group
led to compound 7, which showed a little less NK1
antagonist activity (KB = 151 nM) but much better met-
abolic stability (remaining ratio = 70%) than compound
The disclosure by Pfizer of the first two nonpeptide NK1
antagonists, CP-96345 and CP-99994,5 has spurred
intensive research in this field. During the last few years,
several other structural classes of NK1 receptor antago-
Keywords: Meisenheimer reaction; Pyrido[4,3-b]-1,5-oxazocin-6-one;
Pyrido[2,3-b]-1,5-oxazocin-6-one.
*
Corresponding author. Fax: +81 280 57 1293; e-mail: shigeki.seto@
mb.kyorin-pharm.co.jp
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.12.091

1480
S. Seto et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1479–1484
CF3
O
CF3
N
Me
MeO
Me
N
CF3
H
N
O
O
N
O
CF3
CF3
O
N
N
H
N
H
N
H
F
O
Me
CF3
N
N
H
CP-99994 (1)
L-733060 (2)
MK-869 (3)
TAK-637 (4)
Figure 1.
Ra
O
N
Rb
CF3
CF3
N
N
O
R1
H2N
CF3
O
N
CF3
A
8
CF3
Ra
N
CF3
O
N
Rb
CF3
CF3
6: A=O
7: A=CONMe
5
N
O
R1
9
Scheme 1.
6. Compound 7 is structurally new and simple; therefore,
we thought that it would be amenable to the introduc-
tion of multiple chemical groupings with the aim of pro-
ducing novel potent NK1 antagonists. By incorporating
the above mentioned common pharmacophore, an
eight-membered ring and a basic amine moiety, into pyr-
idine ring, the general targets 8 and 9 were designed
(Scheme 1).
pling of intermediates 15a and 15b with morpholine
under heat afforded the desired compounds 7b and 7c,
respectively.
Scheme 3 illustrates the synthesis of compounds 8a,b,
and 9a–i. Bicyclic intermediates 16a,b, and 17a–d were
prepared according to the method described previ-
ously.14 Compounds 8a,b, and 9a–i were synthesized
by treatment of 16a,b, and 17a–d via the Meisenheimer
reaction, followed by animation with various cyclic ani-
mes according to the synthesis of 7b and 7c.
In this paper, we report the molecular design and syn-
thesis of 8 and 9 as new lead compounds for novel
NK1 antagonists.
3. Biology
2. Chemistry
3.1. In vitro studies
The syntheses of a variety of arylpyridinecarboxamide
derivatives are summarized in Schemes 2 and 3.
NK1 receptor antagonist activity toward guinea pig
ileum was evaluated by the method previously de-
scribed15 with slight modification. The activity was ex-
pressed as KB values as determined by the Schild
method.16
Scheme 2 illustrates the synthesis of compounds 7a–c.
Treatment of 10 with 3,5-bis(trifluoromethyl)-N-meth-
ylbenzylamine afforded 11a and 11b. Alkoxylation of
11b with potassium methoxide afforded 12.12 The Suzuki
coupling reaction of 11a and 12 with aryl boronic acid
took place smoothly to give 7a and 13. Formation of
chlorides 15a and 15b via the Meisenheimer reaction13
was accomplished by treatment of 14a and 14b with
mCPBA followed by rearrangement with POCl3. Cou-
3.2. Metabolic stability in rat microsomes
Metabolic stability was expressed as the remaining ratio
determined after incubation for 60 min in a shaking
water bath at 37 °C.

S. Seto et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1479–1484
1481
CF3
R2
O
CF3
R2
R2
O
Me
a
c
N
N
Me
N
N
CF3
N
CO2H
CF3
Cl
Cl
10a: R2 = H
10b: R2 = I
11a: R2 = H, 26%
11b: R2 = I, 99%
12: R2 = OMe, 47%
7a: R2 = H, 90%
13: R2 = OMe, 96%
b
CF3
CF3
R2
O
R4
R2
Me
Me
N
d
e
N
N
N
CF3
CF3
O
O
14a: R2 = H, 89%
15a: R2 = H
f
14b: R2 = OMe, 99%
15b: R2 = OMe
7b: R2 = H 66% from 14a
f
7c: R2 = OMe, 27% from 14b
Scheme 2. Reagents and conditions: (a) (1) SOCl2, DMF, reflux, 2 h; (2) 3,5-bis(trifluoromethyl)-N-methylbenzylamine, Et3N, 0 °C, 1 h then rt, 3 h;
(b) MeOK, MeOH, rt, 24 h; (c) phenylboronic acid, 10 mol % Pd(PPh3)4, 2 M Na2CO3, toluene, dioxane, reflux, 1–5 h; (d) mCPBA, CH2Cl2, rt, 24 h;
(e) POCl3, reflux, 1–2 h; (f) morpholine, 140–150 °C, 5 h.
O
O
O
Y
Y
Cl
Y
CF3
CF3
CF3
CF3
CF3
CF3
X
X
X
a
b
N
N
N
R1
R1
R1
O
O
O
16a: X = N, Y = CH, R1 = H
16b: X = N, Y = CH, R1 =Me
17a: X = CH, Y = N, R1 = H
17b: X = CH, Y = N, R1 = Cl
17c: X = CH, Y = N, R1 = F
17d: X = CH, Y = N, R1 = Me
18a: X = N-O, Y = CH, R1 = H, 44%
18b: X = N-O, Y = CH, R1 =Me, 70%
19a: X = CH, Y = N-O, R1 = H, 54%
19b: X = CH, Y = N-O, R1 = Cl, 80%
19c: X = CH, Y = N-O, R1 = F, 72%
20a: X = N, Y = CH, R1 = H, 99%
20b: X = N, Y = CH, R1 = Me, 99%
21a: X = CH, Y = N, R1 = H, 99%
21b: X = CH, Y = N, R1 = Cl, 99%
21c: X = CH, Y = N, R1 = F, 99%
21d: X = CH, Y = N, R1 = Me, 99%
19d: X = CH, Y = N-O, R1 = Me, 83%
R4
R1
Y
O
Z1 =
N
O
CF3
c
X
Z2 =
Z3 =
Z4 =
Z5 =
Z6 =
N
N
N
N
N
N
N
O
CF3
8a: X = N, Y = CH, R1 = H, R4 = Z1, 57%
8b: X = N, Y = CH, R1 =Me, R4 = Z2, 39%
9a: X = CH, Y = N, R1 = H, R4 = Z1, 27%
9b: X = CH, Y = N, R1 = H, R4 = Z3, 47%
9c: X = CH, Y = N, R1 = H, R4 = Z4,11%
9d: X = CH, Y = N, R1 = H, R4 = Z5, 48%
9e: X = CH, Y = N, R1 = H, R4 = Z2, 48%
9f: X = CH, Y = N, R1 = H, R4 = Z6, 48%
9g: X = CH, Y = N, R1 = Cl, R4 = Z2, 48%
9h: X = CH, Y = N, R1 = F, R4 = Z2, 52%
9i: X = CH, Y = N, R1 = Me, R4 = Z2, 48%
NMe
N
Scheme 3. Reagents and conditions: (a) mCPBA, CH2Cl2, rt, 24 h; (b) POCl3, reflux, 1 h; (c) R2H, 150 °C, 5 h.

1482
S. Seto et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1479–1484
3.3. In vivo studies
The effect of the substituents at the ortho and para posi-
tions on the pyridine ring (R2 and R4) was examined
(7a–c). The NK1 antagonist activity of 7b, in which R4
is a morpholino group, was similar to that of 7a. Com-
pound 7c, carrying an additional methoxy group at R2,
showed 5 times more potent NK1 antagonist activity
than 7b.
Effective bladder capacity was measured by injection of
saline into spinalized guinea pigs according to the meth-
od previously described.17
4. Results and discussion
Next, the effect of cyclization between R2 and R3 and
that of the substituent at the R4 position was examined.
Compound 16a, which is cyclized between R2 and R3,
showed NK1 antagonist activity that was twice as potent
as that of 7a. Interestingly, compound 8a, bearing a
The NK1 antagonist activity and metabolic stability of
the present series of compounds are summarized in
Table 1, together with the results for a representative
NK1 antagonist, TAK-637.
Table 1. NK1 antagonist activity of Arylpyridinecarboxamide derivatives
CF3
R4
R1
Y
R2
O
R3
N
X
CF3
Compd
X
Y
R1
R2
R3
R4
NK1 antagonist
activitya KB (nM)
Metabolic stabilityb
remaining ratio (%)
7a
7b
N
N
CH
CH
H
H
H
H
Me
Me
H
151
129
70
100
O
N
N
7c
16a
8a
9a
9b
9c
9d
9e
9f
N
CH
CH
CH
N
H
H
H
H
H
H
H
H
H
F
OMe
Me
26.3
69.2
2.63
2.57
13.8
13.2
1.16
0.224
2.34
0.207
0.382
0.210
100
100
100
100
71
O
N
–O(CH2)3–
H
N
–O(CH2)3–
–O(CH2)3–
–O(CH2)3–
–O(CH2)3–
–O(CH2)3–
–O(CH2)3–
–O(CH2)3–
–O(CH2)3–
–O(CH2)3–
–O(CH2)3–
O
O
N
N
CH
CH
CH
CH
CH
CH
CH
CH
CH
N
N
N
70
N
N
100
86
MeN
N
N
N
N
N
100
89
N
N
9g
9h
9i
N
N
N
N
N
N
Cl
Me
100
100
N
N
N
8b
N
CH
Me
–O(CH2)3–
0.339
0.270
100
N
N
TAK-637
a Compounds were screened for antagonist activity on guinea pig ileum as described in the text.
b Compounds were screened for metabolic stability on rat microsomes as described in the text.

S. Seto et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1479–1484
1483
morpholino group at R4, showed NK1 antagonist activ-
ity that was 25 times more potent than that of 16a.
Moreover, the activity of 8a was 10 times more potent
than that of 7c, which is noncyclized compound. These
results indicate that the conformational restriction by
the oxazocine ring and introduction of a nitrogen atom
is important for NK1 receptor recognition, as shown in
the previous studies of NK1 antagonists. The NK1
antagonist activity of 9a, the regioisomer of the nitrogen
atom on the pyridine ring, was similar to that of 8a.
Table 2. Augmentative effects of 8b and 9i on effective bladder
capacity in guinea pigs
N
O
N
Y
CF3
X
N
Me
O
CF3
Compd
X
Y
NK1 antagonist
Effective bladder
activitya KB (nM) capacity increasing
To explore the effect of the terminal oxygen atom on
morpholino moiety, replacement of the cyclic amino
group (R4) was examined (9a–f). A terminal nitrogen
atom was favorable (9d,e,f). In particular, 9e, into which
a 4-(pyrrolidiny)piperidino group had been introduced,
showed excellent NK1 antagonist activity. These results
suggest that the terminal heteroatom on the cyclic
amino group at the C-9 position are very important
for NK1 receptor recognition.
ratio (0.3 mg/kg iv)b
8b
9i
N
CH
CH 0.339
N
3.97%
24.2%
0.210
TAK-637
0.270
12.0%
a Compounds were screened for antagonist activity on guinea pig ileum
as described in the text.
b Effective bladder capacity was measured as the volume of saline
injected into spinalized guinea pigs. The capacity-increasing effects of
the test compounds were expressed as the ratio of the increase in
effective bladder capacity compared with the predrug values.
The observed effect of introducing substituents at the
ortho position on the phenyl ring (R1) was not signifi-
cant (9e,g–i). On the other hand, the NK1, antagonist
activity of 8b,18 the regioisomer of 9i,19,20 was similar
to that of 9e.
pigs. Pharmacological studies of 9i will be reported in
due course.
As for the metabolic stability, the newly synthesized
compounds 7a–c, 8a,b, and 9a–i were considerably im-
proved in comparison with 6. Above all, 7b,c, 16a,
8a,b, 9a,d,f,h,i showed good stability. Taking NK1
antagonist activity and metabolic stability into consider-
ation, 9i was shown to be the best of all the compounds
we synthesized.
Acknowledgements
The authors wish to thank Dr. H. Miyachi, presently at
the University of Tokyo, for many valuable suggestions
and for encouragement, and also Dr. T. Ishizaki, Dr. Y.
Fukuda, Dr. Y. Takano, and Dr. M. Segawa of Kyorin
Pharmaceutical Co., Ltd, for helpful discussion.
We also evaluated the augmentative effect of 9i, the most
effective compound among those tested in the above in
vitro assay, and 8b, the regioisomer of 9i, on effective
bladder capacity of guinea pigs following iv administra-
tion (Table 2). Compound 8b was less potent than TAK-
637. In contrast, 9i showed the greatest in vivo activity
(24.2% at a dose of 0.3 mg/kg) of the present series of
compounds, although 8b, 9i, and TAK-637 showed al-
most the same in vitro activity. These results suggest
that a pyrido[2,3-b]-l,5-oxazocine core bearing a 4-(pyr-
rolidiny)piperidino moiety is favorable for enhanced
activity in vivo.
References and notes
1. Chang, M. M.; Leeman, S. E.; Niall, H. D. Nat. New Biol.
1971, 232, 86–87.
2. Guard, S.; Watson, S. P. Neurochem. Int. 1991, 18, 149.
3. (a) VonEuler, U. S.; Gaddum, J. H. J. Physiology 1931, 72,
74; (b) Chan, M. M.; Leeman, S. E.; Niall, H. D. Nat. New
Biol. 1971, 232, 86.
4. (a) Maggi, C. A. Gen. Pharmacol. 1991, 22, 1; (b) Lecci,
A.; Giuliani, S.; Garret, C.; Maggi, C. A. Neuroscience
1993, 54, 827.
In conclusion, we have succeeded in design and synthesis
of novel pyrido-oxazocine derivatives 8 and 9. The
structure–activity relationship study indicated that
NK1 receptor recognition was improved by conforma-
tional restriction arising from the incorporation of an
oxazocine ring and by the presence of a nitrogen atom
at C-9 position. Moreover, it was clarified that a termi-
nal heteroatom on the cyclic amino group at the C-9 po-
sition is important for NK1 receptor recognition.
Through these studies, we identified 9-[4-(pyrrolid-
iny)piperidino]-7-(2-methylphenyl)-3,4,5,6-tetrahydro-2H-
pyrido[2,3-b]-1,5-oxazocin-6-one (9i) as exhibiting
highly potent NK1 antagonist activity in the guinea
pig ileum contraction assay and good in vivo activity
in increasing the effective bladder capacity of guinea
5. Desai, M. C.; Lefkowitz, S. L.; Thadeio, P. R.; Longo, K.
P.; Snider, R. M. J. Med. Chem. 1992, 35, 3911.
6. Harrison, T.; Williams, B. J.; Swain, C. J.; Ball, R. G.
Bioorg. Med. Chem. Lett. 1994, 4, 2545.
7. Hale, J. J.; Mills, S. G.; MacCoss, M.; Finke, P. E.;
Cascieri, M. A.; Sadowski, S.; Ber, E.; Chicchi, G. G.;
Kurts, M.; Metzger, J.; Eiermann, G.; Tsou, N. N.;
Tattersall, F. D.; Rupniak, N. M. J.; Williams, A. R.;
Rycroft, W.; Hargreaves, R.; Maclntyre, D. E. J. Med.
Chem. 1998, 41, 4607.
8. Natsugari, H.; Ikeura, Y.; Kamo, I.; Ishimaru, T.; Ishichi,
Y.; Fujishima, A.; Tanaka, T.; Kasahara, F.; Kawada, M.;
Doi, T. J. Med. Chem. 1999, 42, 3982.
9. Swain, C. J.; Seward, E. M.; Cascieri, M. A.; Fong, T. M.;
Herbert, R.; MacIntyre, D. E.; Merchant, K. J.; Owen, S.
N.; Owens, A. P.; Sabin, V.; Teall, M.; VanNiel, M. B.;

1484
S. Seto et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1479–1484
s), 7.78 (1H, s). HRMS (El): calcd for C34H36F6N4O2 (M+)
646.2742, found 646.2723.
Williams, B. J.; Sadowski, S.; Strader, C.; Ball, R. G.;
Baker, R. J. Med. Chem. 1995, 38, 4793.
10. Lowe, J. A., III; Drozda, S. E.; Snider, R. M.; Longo, K.
P.; Zorn, S. H.; Morrone, J.; Jackson, E. R.; McLean, S.;
Bryce, D. K.; Borner, J.; Nagahisa, A.; Kanai, Y.; Suga,
O.; Tsuchiya, M. J. Med. Chem. 1992, 35, 2591.
11. Swain, C. J.; Cascieri, M. A.; Owens, A. P.; Saari, W.;
Sadowski, S.; Strader, C. D.; Teall, M.; VanNiel, M. B.;
Williams, B. J. Bioorg. Med. Chem. Lett. 1994, 4, 2161.
12. A regioisomer of 12, N-[3,5-bis(trifluoromethyl)benzyl]-4-
19. Compound 9i: This compound exists as a mixture of
slowly interconverting conformational isomers (atro-
pisomers). 1H NMR (400 MHz, CDCl3): d 1.50–2.01
(10H, m), 2.04–2.20 (1H, m), 2.20–2.36 (3H, m), 2.54–
2.67 (4H, m), 2.88–3.02 (2H, m), 3.11–3.23 (1H, m), 3.63–
3.83 (1H, m), 3.94 (0.5H, d, J = 15.3 Hz), 3.95 (0.5H, d,
J = 15.3 Hz), 4.20–4.51 (4H, m), 5.36 (0.5H, d,
J = 15.3 Hz), 5.39 (0.5H, d, J = 15.3 Hz), 6.26 (1H, s),
6.79 (0.5H, d, J = 7.3 Hz), 7.02–7.09 (0.5H, br), 7.10–7.16
(0.5H, br), 7.18–7.28 (2H, m), 7.28–7.33 (0.5H, br), 7.53
(2H, s), 7.76 (1H, s). HRMS (EI): calcd for C34H36F6N4O2
(M+) 646.2742, found 646.2704.
chloro-2-methoxy-N-methylpyridinecarboxamide,
obtained in 16% yield.
was
13. Meisenheimer, J. Ber 1926, 59, 1848.
14. Seto, S. Tetrahedron Lett. 2004, 45, 8475.
15. Dion, S.; DÕOrleans-Juste, P.; Drapeau, G.; Rhaleb, N.-E.;
Rouissi, N.; Tousignant, C.; Regoli, D. Life Sci. 1987, 41,
2269.
16. Arunlakshana, O.; Schild, H. O. Br. J. Pharmacol. 1959,
14, 48.
20. There is a possibility that 9i is an equilibrating mixture of
conformational isomers. In NMR spectroscopy, the signal
of the benzylic methylene protons of 9i19 was separated as
two pairs of doublets (d 3.94, 3.95, 5.36, 5.39; each 0.5H,
J = 15.3 Hz) and that of the aryl protons was broadened
and separated. These results might indicate that 9i suffers
from restricted rotation about the biaryl bond, which
might reflect the stacking conformation desirable for NK1
receptor recognition.8,21 This feature of the NMR spec-
trum of 9i, however, was not observed with compounds
8b.17.
17. Peterson, J. S.; Hanson, R. C.; Noronha-Blob, L. J.
Pharmacol. Methods 1989, 21, 231.
1
18. Compound 8b: H NMR (400 MHz, CDCl3): d 1.45–1.85
(6H, m), 1.85–2.08 (5H, m), 2.24 (3H, s), 2.52–2.65 (3H,
m), 2.82–2.97 (2H, m), 3.18–3.28 (1H, m), 3.78–3.88 (1H,
m), 3.93 (1H, d, J = 15.3 Hz), 4.06–4.16 (1H, m), 4.16–4.45
(2H, m), 5.34 (1H, d, J = 15.3 Hz), 6.13 (1H, s), 6.92–7.01
(1H, m), 7.01–7.09 (1H, m), 7.16–7.24 (2H, m), 7.58 (2H,
21. Ishichi, Y.; Ikeura, Y.; Natsugari, H. Tetrahedron 2004,
60, 4481.