Dual neurokinin NK1/NK2 antagonists: N-[(R,R)-(E)-1-arylmethyl-3-(2-oxo-azepan-3-yl)carbamoyl]allyl-N-methyl-3,5- bis(trifluoromethyl)benzamides and 3-[N′-3,5-bis(trifluoromethyl)benzoyl-N-arylmethyl-N′- methylhydrazino]-N-[(R)-2-oxo-azepan-3-yl]propionamides

Article abstract of DOI:10.1016/S0960-894X(01)00631-X

Based on the structure of N-[(R,R)-(E)-1-(4-chlorobenzyl)-3-(2-oxoazepan-3-yl)carbamoyl]allyl-N- methyl-3,5-bis(trifluoromethyl)benzamide (1), attempts to improve the NK₂ affinity have resulted in the discovery of N-[(R,R)-(E)-1-(3,4-dichlorobenzyl)-3-(2-oxoazepan-3-yl)carbamoyl]allyl-N- methyl-3,5-bis(trifluoromethyl)benzamide (9, DNK333) exhibiting a 5-fold improved affinity to the NK₂ receptor in comparison to 1. Simplification of the structure via elimination of a chiral centre led to 3-[N′-3,5-bis(trifluoromethyl)benzoyl-N-(3,4-dichlorobenzyl)-N′- methylhydrazino]-N-[(R)-2-oxo-azepan-3-yl]propionamide (22), a potent and fairly balanced NK₁/NK₂ antagonist.

Full text of DOI:10.1016/S0960-894X(01)00631-X

Bioorganic & Medicinal Chemistry Letters 11 (2001) 3081–3084
Dual Neurokinin NK1/NK2 Antagonists: N-[(R,R)-(E)-1-
arylmethyl-3-(2-oxo-azepan-3-yl)carbamoyl]allyl-N-methyl-3,5-
bis(trifluoromethyl)benzamides and 3-[N0-3,5-
bis(trifluoromethyl)benzoyl-N-arylmethyl-N0-methylhydrazino]-N-
[(R)-2-oxo-azepan-3-yl]propionamides
Marc Gerspacher,a,* Luigi La Vecchia,a Robert Mah,a Andreas von Sprecher,a
Gary P. Anderson,b Natarajan Subramanian,a Kathleen Hauser,a Heinrich Bammerlin,a
Sabine Kimmel,a Viviane Pawelzik,a Karin Ryffela and Howard A. Balla
aPharma Research, Novartis Pharma AG, CH-4002 Basel, Switzerland
bDepartment of Pharmacology, University of Melbourne, Parkville, 3052 VIC, Australia
Received 19 July 2001; accepted 19 September 2001
Abstract—Based on the structure of N-[(R,R)-(E)-1-(4-chlorobenzyl)-3-(2-oxoazepan-3-yl)carbamoyl]allyl-N-methyl-3,5-bis(tri-
fluoromethyl)benzamide (1), attempts to improve the NK2 affinity have resulted in the discovery of N-[(R,R)-(E)-1-(3,4-dichloro-
benzyl)-3-(2-oxoazepan-3-yl)carbamoyl]allyl-N-methyl-3,5-bis(trifluoromethyl)benzamide (9, DNK333) exhibiting a 5-fold improved
affinity to the NK2 receptor in comparison to 1. Simplification of the structure via elimination of a chiral centre led to 3-[N0-3,5-
bis(trifluoromethyl)benzoyl-N-(3,4-dichlorobenzyl)-N0-methylhydrazino]-N-[(R)-2-oxo-azepan-3-yl]propionamide (22),
and fairly balanced NK1/NK2 antagonist. # 2001 Elsevier Science Ltd. All rights reserved.
a potent
Neurokinins, also known as tachykinins, are a family of
small peptides and are widely distributed throughout
the central and peripheral nervous system where they
act as neurotransmitters and neuromodulators. Neuro-
kinins exhibit their effects via 7-transmembrane G-pro-
tein-coupled receptors (NK1, NK2 and NK3 receptors).
Neurokinins are proposed to be involved in a number of
pathological conditions (pain, arthritis, migraine,
emesis, cancer, anxiety, depression, schizophrenia,
asthma) and NKreceptor antagonists have been pro-
posed to have potential clinical benefits.1 There is evi-
dence that neurokinins play an important role in airway
disease induction and progression via the activation of
NK1 and NK2 receptors.2 A number of studies suggest
that neurokinin receptor antagonists, especially dual
NK1/NK2 antagonists, may represent a new treatment
option for asthma and other airway diseases, particu-
larly since lung tissue from asthma patients has been
shown to overexpress NK1 and NK2 receptors.3 As a
consequence, many pharmaceutical companies have
shifted their efforts aiming at selective NKantagonists
towards the discovery of dual NK1/NK2 antagonists.4
Recently we have reported the discovery of an orally
active NK1/NK2 antagonist. N-[(R,R)-(E)-1-(4-chloro-
benzyl)-3-(2-oxo-azepan-3-yl)carbamoyl]allyl-N-methyl-
3,5-bis(trifluoromethyl)benzamide (1)5 is derived from a
series of N-[1-(4-chlorobenzyl)-3-carbamoyl]allyl-N-methyl-
benzamides (Fig. 1).
*Corresponding author. Tel.: +41-61-696-7636; fax: +41-61-696-
5966; e-mail: marc.gerspacher@pharma.novartis.com
Figure 1. N-[1-(4-chlorobenzyl)-3-carbamoyl]allyl-N-methyl-benzamides.
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00631-X

3082
M. Gerspacher et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3081–3084
Compound 1 exhibits a highly potent affinity to the
NK1 receptor (inhibition of 3H-sar9-substance P binding
to bovine retinal membranes: IC50=0.5 nM) and a good
affinity to the NK2 receptor (inhibition of 125I-NKA
binding to human NK2–CHO cells: IC50=24 nM). In
addition to the potent binding affinities, 1 also shows
oral activity in guinea pig bronchoconstriction.5
Condensation with enantiomerically pure d-a-amino-e-
caprolactam5,8 in the presence of EDC, removal of the
BOC protecting group and subsequent N-acylation,
followed by simple chromatography on silica gel to
remove the undesired diastereoisomer (diastereoisomers
are easily separable in all cases) led to the final products
as enantiomerically pure R,R-compounds.9
Although 1 in terms of NK1/NK2 antagonistic activity
is very promising, we wanted to explore possible ways
to improve the NK2 affinity and to simplify the struc-
ture (elimination of chiral centres). In order to explore
the effect of various benzyl substituents on the binding
properties, compounds were prepared as outlined in
Scheme 1. Starting from the corresponding amino acid
methylesters 2a,b, followed by BOC-protection and N-
methylation (methylation of the indole-nitrogen also
occurs in this step) the N-BOC-N-methylaryl-alanine
derivatives (5) were prepared. For compounds where
the amino acid-methylesters are not commercially
available, alkylation of N-BOC-sarcosine methylester6
using commercially available aryl-methyl bromides and
15a,b7 were used to prepare 5.
Compounds containing 4-chlorophenyl (1), dichloro-
thienyl (7), 3,4-dichlorophenyl (9), 2-naphthyl (10), and
3-(1-methyl)indolyl (11) substituents exhibit binding
affinities to the NK1 receptor in the low nanomolar
region, whereas the dibromothiophene containing com-
pound 8 has a somewhat decreased binding affinity to
the NK1 receptor. NK2 binding affinities stay between
17–60 nM, except for compound 9 where the additional
chlorine leads to a 5 times more potent affinity to the
NK2 receptor in comparison with the monochloro-deri-
vative 1. Both the dichlorophenyl- and the dibromo-
thiophene derivatives 9 and 8 exhibit balanced binding
affinities to both receptors (Table 1).
We also aimed at the simplification of the structure.
First (in a limited effort), ways to eliminate the chiral cen-
tre in the caprolactam residue were explored. Compounds
12 and 13 were prepared by reacting 6 (aryl=4-Cl–Ph)
Reduction of the esters to the aldehydes followed by
chain elongation led to the a,b-unsaturated acids 6.
Table 1. In vitro binding affinities of compounds 1, 7–11 to NK1-
and NK2-receptors
Aryl
compound no.
NK1 binding10
IC50 (nM)a
NK2 binding11
IC50 (nM)a
0.5
3.6
24
17
1
7
20
26
8
4.8
1.1
5.5
60
9 (DNK333)12
10
Scheme 1. Preparation of compounds 1, 7–13. (a) (BOC)2O, Et3N; (b)
MeI, Ag2CO3, DMF; (c) LDA, THF, À78 ꢀC to rt; (d) DIBAH, tol,
À78 ꢀC; (e) n-BuLi, trimethylsilyl-P,P-diethylphos-phonoacetate, then
citric acid; (f) R-NH2, EDC, DMAP, CH2Cl2; (g) TFA, CH2Cl2; (h)
3,5-bis(trifluoromethyl)benzoyl chloride; Et3N, CH2Cl2, chromato-
graphy; (i) DIBAH, tol; (j) PBr3, CH2Cl2.
0.5
24
11
aValues are means of three experiments.

M. Gerspacher et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3081–3084
3083
with the corresponding amines instead of d-a-amino-e-
caprolactam as outlined in Scheme 1.
residue leads to a compound that exhibits only a minor
decrease of binding affinities.
As can be seen from the data shown in Table 2,
removal/replacement of the heteroatoms from the 7-mem-
bered ring leads to a significant decrease of binding affi-
nities to both receptors. Replacement of the chiral amino-
caprolactam with an amino-cyclohexylcarboxamide
Another way to simplify the structure is the elimination
of the chiral centre in the backbone of the compounds.
Replacing the chiral carbon atom with a nitrogen leads
to hydrazine derivatives that can be prepared as
outlined in Scheme 2.
Table 2. In vitro binding affinities of compounds 1, 12, 13 to NK1-
and NK2-receptors
Condensation of the commercially available 3,5-bis(tri-
fluoromethyl)benzoyl hydrazide with the corresponding
commercially available aldehydes followed by N-
methylation and reduction led to hydrazides 19a–c.
Reaction of the latter compounds with 3-oxopropionic
acid ethylester13 followed by another reduction led to
the esters 20a–c. Subsequent saponification and reaction
with d-a-amino-e-caprolactam in the presence of EDC
yielded enantiopure hydrazine-derivatives 21–23. The
final compounds can be prepared in overall yields of up
to 43%.14
R Compound no.
NK1 binding10
IC50 (nM)a
NK2 binding11
IC50 (nM)a
Replacement of the chiral carbon atom with a nitrogen
did not have a dramatic impact on the binding affinities
of the resulting hydrazine derivatives to the NK1 and
NK2 receptors. The 3,4-dichloro-benzyl-derivative 22
exhibiting an IC50 value of 1.4 nM (NK1) and an IC50
value of 5.5 nM (NK2) is as expected the most potent
and balanced dual NK1/NK2 antagonist from this small
series of hydrazine analogues (Table 3).
0.5
12
24
1
203
12b
1
66
Based on the structure of N-[(R,R)-(E)-1-(4-chloroben-
zyl)-3-(2-oxoazepan-3-yl)carbamoyl]allyl-N-methyl-3,5-
bis(trifluoromethyl)benzamide (1), attempts to improve
the NK2 affinity have resulted in the discovery of the
balanced NK1/ NK2 antagonist N-[(R,R)-(E)-1-(3,4-
dichlorobenzyl)-3-(2-oxoazepan-3-yl)carbamoyl]allyl-N-
methyl-3,5-bis(trifluoromethyl)benzamide (9) exhibiting
a 5 times improved affinity to the NK2 receptor in
13b
aValues are means of three experiments.
b12 and 13 are racemic compounds.
Table 3. In vitro binding affinities of compounds 21–23 to NK1- and
NK2-receptors
Aryl compound no.
NK1 binding10
IC50 (nM)a
NK2 binding11
IC50 (nM)a
18
19.5
21
1.4
6.4
5.5
55
22
23
Scheme 2. Preparation of compounds 21–23. (a) 3,4-dichloro-
benzaldehyde, EtOH, heat; (b) MeI, K2CO3, acetone, heat; (c)
NaCNBH3, H+; (d) 3-oxopropionic acid ethylester, EtOH, HOAc; (e)
NaCNBH3, H+; (f) NaOH, THF, MeOH, H2O; (g) d-a-amino-e-
caprolactam, EDC, DMAP, CH2Cl2.
aValues are means of three experiments.

3084
M. Gerspacher et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3081–3084
comparison to 1. In addition to this, removal of the
chiral centre in the backbone of 9 resulted in the dis-
covery of the potent dual NK1/NK2 antagonist 3-[N0-
3,5-bis(trifluoromethyl)benzoyl-N-(3,4-dichlorobenzyl)-
N0 -methylhydrazino]-N-[(R)-2-oxoazepan-3-yl]propion-
amide (22) exhibiting balanced affinity for both receptors.
well plates (Nunclon) containing 200 mL 20 mM HEPES buf-
fer, pH 7.4 containing 2 mM MnSO4 and 6 mM MgCl2, 3Â105
h NK2 CHO cells, 0.05 nM [125I]-NKA (2200 Ci mmolÀ1) and
various drug concentrations. Non-specific binding was esti-
mated in the presence of 50 nM NKA. The mixture was incu-
bated for 20 min at room temperature after which the
unbound ligand was removed by rapid filtration and washed
four times with ice cold Tricine buffer. Filter bound radio-
activity was counted in Microscint 20 in a scintillation coun-
ter. All samples were measured in triplicate. Culture
conditions and cell isolation for h NK2 CHO cells: Sub-
ramanian, N.; Ruesch, C.; Bertrand, C. Biochem. Biophys.
Res. Comm. 1994, 200, 1512.
References and Notes
1. (a) Longmore, J.; Swain, C. J.; Hill, R. G. Drug News Per-
spec. 1995, 8, 5. (b) Kucharczyk, N. Exp. Opin. Invest. Drugs
1995, 4, 299. (c) Elliott, J.; Seward, E. M. Exp. Opin. Ther. Pat.
1997, 7, 43. (d) Longmore, J.; Hill, R. G.; Hargreaves, R. J.
Can. J. Phys. Pharmacol. 1997, 75, 612. (e) von Sprecher, A.;
Gerspacher, M.; Anderson, G. P. Idrugs 1998, 1, 73.
2. (a) Ford-Hutchinson, A. W.; Rodger, I. W.; Jones, T. R.
Drug News Perspec. 1992, 5, 542. (b) Geppetti, P.; Bertrand,
C.; Ricciardolo, F. M. L.; Nadel, J. A. Can. J. Physiol. Phar-
macol. 1995, 7, 843. (c) Advenier, C.; Lagente, V.; Boichot, E.
Eur. Respir. J. 1997, 10, 1892. (d) Chapman, R. W.; Hey, J. A.;
McLeod, R.; Minnicozzi, M.; Rizzo, Ch. Drug News Perspect.
1998, 11, 480.
3. (a) Murai, M.; Morimoto, H.; Maeda, Y.; Kiyotoh, S.;
Nishikawa, M.; Fujii, T. J. Pharmacol. Exp. Ther. 1992, 262,
403. (b) Joos, G. F.; Van Schoor, J.; Kips, J. C.; Pauwels, R. A.
Am. J. Respir. Crit. Care Med. 1996, 153, 1781.
4. Gerspacher, M.; von Sprecher, A. Drugs Future 1999, 24,
883.
5. Gerspacher, M.; von Sprecher, A.; Mah, R.; Anderson,
G. P.; Bertrand, C.; Subramanian, N.; Hauser, K.; Ball, H.
Bioorg. Med. Chem. Lett. 2000, 10, 1467.
6. Bonora, G. A.; Toniolo, C. Gazz. Chim. Ital. 1977, 107, 381.
7. 15a,b were prepared from the corresponding methylesters
(Scheme 1). 2,3-dibromothiophene-5-carboxylic acid methy-
lester is commercially available, whereas 2,3-dichloro-
thiophene-5-carboxylic acid methylester can be prepared as
described by Stanetty, P.; Puschantz, E. Monatsh. Chem. 1987,
120, 65.
12. Detailed pharmacology data of DNK333 will be published
in due course.
Analytical data,
9
(DNK333): mp 127–129 ꢀC.
[a]2D0=+39.9ꢀ (c 0.94, EtOH). ee value=99.3 (Rt=21.64 min,
Chiralcel OJ, hexane/isopropanol=85:15+0.1% TFA, flow
rate 1 mL/min). 1H NMR (400 MHz, d, DMSO, +150 ꢀC)
8.02 (s, 1H); 7.65 (s, 2H); 7.55 (s,b, 1H); 7.44 (m, 2H); 7.28 (b,
1H); 7.23 (b, 1H); 6.71 (dd, 1H); 6.32 (dd, 1H); 5.03 (b, 1H);
4.51 (m, 1H); 3.18 (m, 2H); 3.05 (m, 2H); 2.82 (s, 3H); 1.96 (m,
2H); 1.75 (m, 2H); 1.45 (m, 1H); 1.3 (m, 1H). Anal. calcd for
C27H26ClF6N3O3: C: 51.94, H: 4.04, N: 6.78. Found: C: 51.66,
H: 4.05, N: 6.73. IR (CH2Cl2, cmÀ1): 1668 (–CH¼CH–C¼O),
1643 (Ph–C¼O), 1286 (Ph–CF3), 975 (C–H, trans disubst.
C¼C).
13. Sato, M.; Yoneda, N.; Katagiri, N.; Watanabe, H. Synth-
esis 1986, 672.
14. For the preparation of larger quantities of the hydrazine
derivative 22, it proved possible to directly use the Meldrum’s
acid derivative 24 in the enamine-formation step thus avoiding
the preparation of the unstable 3-oxopropionic acid ethyl-
ester.13 In addition, in the enamine reductions, the triethyl-
silane/TFA-system15 was used instead of sodium
cyanoborohydride:
8. Rezler, E. M.; Fenton, R. R.; Esdale, W. J.; McKeage,
M. J.; Russell, P. J.; Hambley, T. W. J. Med. Chem. 1997, 40,
3508.
9. The assignment of the stereochemistry of the backbone
chiral centre (R or S) was done on the basis of chromato-
graphical behaviour on silica gel and NKreceptor binding
profile of the two diastereoisomers in comparison with 1.5
Stereoselective synthesis of the R,R-isomers of compounds 1,5
10 and 11 starting from the corresponding d-amino acids
confirmed the assignment of the absolute stereochemistry.
10. For experimental details, see: Bittiger H., Heid J. In Sub-
stance P; Skrabanek, P., Powell, D., Eds.; Boole: Dublin,
1983; p 198.
(a) EtOH, HOAc, heat (b) Et3SiH, TFA; (c) NaOH, THF,
MeOH, H2O; (d) d-a-amino-e-caprolactam, EDC, DMAP,
CH2Cl2
11. Inhibition of [125I]-NKA binding to transfected chinese
hamster ovary cells (CHO cells) expressing recombinant
human neurokinin 2 receptors: The assay was performed in 96
15. Wu, P.-L.; Peng, S. Y.; Magrath, J. Synthesis 1995, 435.