Identification of a new series of non-peptidic NK3 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.12.135

The identification and structure-activity relationships of 2-aminomethyl-1-aryl cyclopropane carboxamides as novel NK₃ receptor antagonists are reported. The compound series was optimized to give analogues with low nanomolar binding to the NK₃ receptor and brain exposure, leading to activity in vivo in the senktide-induced hypoactivity model in gerbils.

Full text of DOI:10.1016/j.bmcl.2010.12.135

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1498–1501
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of a new series of non-peptidic NK3 receptor antagonists
Karsten Juhl a, , Tore Hansen a, , Jan Kehler a, Nikolay A. Khanzhin a, Morten B. Nørgaard a, Thomas Ruhland a,
⇑
Dorrit B. Larsen a, Klaus G. Jensen a, Björn Steiniger-Brach b, Søren M. Nielsen b, Klaus B. Simonsen a
a Discovery Chemistry and DMPK, H. Lundbeck A/S, Otilliavej 9, DK 2500 Valby, Denmark
b Discovery Pharmacology Research, H. Lundbeck, Otilliavej 9, DK 2500 Valby, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history:
The identification and structure–activity relationships of 2-aminomethyl-1-aryl cyclopropane carboxam-
ides as novel NK3 receptor antagonists are reported. The compound series was optimized to give
analogues with low nanomolar binding to the NK3 receptor and brain exposure, leading to activity
in vivo in the senktide-induced hypoactivity model in gerbils.
Received 2 November 2010
Revised 29 December 2010
Accepted 31 December 2010
Available online 13 January 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
NK3 antagonist
Schizophrenia
Cyclopropane
Senktide induced hypoactivity
Three non-peptide based NK3 receptor antagonists have entered
The NK3 receptor is one of three known neurokinin receptors. The
NK3 receptor is expressed both centrally and in some peripheral or-
gans.1–3 The expression in the central nervous system (CNS) and the
modulatory effect on various monoaminergic transmitter systems4
make NK3 receptor antagonists interesting as potential therapeutic
agents for the treatment of CNS diseases such as schizophrenia,5 Par-
kinsons disease,6 panic attacks and major depressive disorder.7
the clinical trials: Talnetant and AZD2624 from a common 2-aryl-
4-carboxamide-quinoline core structure and Osanetant from a
piperidine amide series. All three compounds have been with-
drawn from further clinical development.
As a result of an NK3-focused screening campaign including
compounds from a previous neurokinin project, we identified a
new structural class of non-peptidic NK3 receptor antagonists.
Specifically, we found lead compound 1a and structurally related
analogues that derived from
a tri-substituted cyclopropane
scaffold. This scaffold is capable of arranging three distinct func-
tionalities (a carboxamide, an aryl and an aminomethyl) in a
well-defined orientation. Compound 1a displays high affinity for
the NK3 receptor (Ki = 5.9 nM) and selectivity towards the NK1
and NK2 receptors (Ki = 130 nM and Ki = 88 nM, respectively).8
The synthesis of 1a (and derivatives) commenced from a diazo-
transfer from p-acetamidobenzenesulfonyl azide (p-ABSA) to the
phenyl acetic acid allyl ester 2 (see Fig. 1). The resulting diazo com-
pound 3 was converted through intramolecular cyclopropanation to
the racemic lactone 4. The racemate was resolved either through
chiral chromatography separation of the two enantiomers (1S,5R)-
4 and (1R,5S)-4 or through a series of chemical modifications involv-
O
NH
OH
O
NH
O
S
O
H
N
N
N
Talnetant
N
AZD2624
O
N
N
O
ing (a) lewis acid mediated lactone opening with (R)-a-methyl
benzyl amine, (b) separation of the resulting two diastereomers 5a
and 5b, (c) acid-mediated conversion of 5a and 5b to the enantio-
meric pure lactones. Alternatively, analogues of (1S,5R)-4 were
synthesized from (R)-(À)-epichlorohydrin and the appropriate
benzyl cyanide.9 The synthesis continued by opening the lactone
ring with various benzyl amines mediated by AlCl3. Conversion of
the alcohol 6 to the bromide and subsequent treatment with an
amine completed the synthesis of 1a.
Cl
Osanetant
Cl
⇑
Corresponding author. Tel.: +45 36434041; fax: +45 36438237.
E-mail address: kaju@lundbeck.com (K. Juhl).
Present address: University of Oslo, Department of Chemistry, Sem Sælands vei 26,
0315 Oslo, Norway.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.12.135

K. Juhl et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1498–1501
Cl
1499
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
NH2
N2
O
a
b
+
H
N
H
OH
N
OH
c
O
O
O
O
O
O
O
rac-4
2
3
5a
e
5b
d
e
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
g or h
NHMe
NR2
OH
N
N
f
O
O
O
O
O
O
1a
6
(1S,5R)-4
(1R,5S)-4
Figure 1. Reagents and conditions: (a) p-ABSA (1.2 equiv), DBU (1.3 equiv) MeCN, 0 °C to rt, 20 h; (b) Rh2(Oct)4 (0.005 equiv), DCM, rt, 4 h; (c) AlCl3 (2 equiv), DCE, column
chromatography; (d) chiral HPLC separation; (e) H2SO4; (f) benzyl amine, AlCl3 (2 equiv), DCE, 30 min; 4, 60 min; (g) (i) PBr3 (3 equiv), DCM, 3 h (ii) amine, Et(iPr)3N, MeCN,
85 °C, 3 h or (i) IBX, 1,2-dichloroethane, (ii) amine, Na(CN)BH3.
The variation of the carboxamide group was the most syntheti-
cally feasible approach to systematically analyze the structure–
activity relationships (SAR) of this part of the scaffold. In parallel
manner, we varied the N-methyl-N-benzyl amide moiety while
keeping the remainder of the molecule unchanged as shown in Table
1. The results indicate that the NK3 receptor affinities are not dra-
matically influenced by substituents on the N-benzyl moiety. A
screen of other amides with substituents different from N-methyl
and N-benzyl was also performed. However, this led to compounds
with lower affinity for the NK3 receptor (data not shown). On the
basis of these SAR, the N-benzyl-N-methyl and N-4-fluorobenzyl-
N-methyl amide moieties were selected to be the optimal substitu-
tion patterns. Furthermore, the data in Table 1 indicate a relatively
high degree of stereoselectivity in the cyclopropane scaffold with
the (1S,2R) configuration giving superior affinity for the NK3 recep-
tor. Therefore the (1S,2R) isomer was chosen for further SAR studies.
The aryl group directly bond to the cyclopropane ring is most ste-
rically restricted and we speculated that substituents in this part of
the scaffold would likely effect receptor affinity. Therefore our next
design iteration was devoted to vary the substitution pattern of this
aromatic ring. (Table 2).10 In this study, it was found that the original
3,4-dichloro substitution pattern was superior. While compounds
(e.g., 7, 8 and 9) with either neutral or electron-withdrawing substit-
uents maintain NK3 receptor affinity, more polar substituents such
Table 1
SAR of N-benzyl amide analogues
Cl
Cl
O
HN
N
N
Table 2
Ar
O
SAR of aromatic ring on cyclopropane core
O
Ar
HN
#
Stereo-chemistry
Ar
hNK3R
Ki (nM)a
N
N
F
O
1a
1b
7
8
9
(1S, 2R)
(1R, 2S)
rac
rac
rac
rac
rac
rac
rac
rac
(1S, 2R)
rac
rac
rac
rac
rac
rac
rac
C6H5
C6H5
2-Cl-C6H4
2-F-C6H4
2-Me-C6H4
2-OMe-C6H4
3-F-C6H4
5.9
83
20
18
5.3
14
12
24
33
50
5.8
11
7.5
15
4.6
28
24
16
13
23
5.9
67
32
70
14
#
Ar
hNK3R
Ki (nM)a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
4-F-C6H4
4-Cl-C6H4
67
36
37
350
1000
270
140
32
40
56
210
1900
47
3-Cl-C6H4
3-OMe-C6H4
3-CF3-C6H4
4-F-C6H4
4-Me-C6H4
4-OMe-C6H4
4-CF3-C6H4
4-CN-C6H4
3-F-C6H4
4-Cl-C6H4
4-Me-C6H4
4-OMe-C6H4
4-CF3-C6H4
4-SMe-C6H4
2,3-Cl2-C6H3
2,4-Cl2-C6H3
3,4-Cl2-C6H3
3,5-Cl2-C6H3
2,4-OMe2-C6H3
3,4-OMe2-C6H3
3,5-OMe2-C6H3
2-Thiophenyl
3-Thiophenyl
3-Cl-C6H4
3-Me-C6H4
3-OMe-C6H4
3-CF3-C6H4
3-CN-C6H4
3,4-F2-C6H3
3,4-Cl2-C6H3
3,5-Cl2-C6H3
3,4-Me-C6H3
3,4-OCH2O-C6H3
2-Thiphenyl
3-Thiophenyl
rac
rac
rac
rac
rac
rac
rac
5.8
90
7.3
140
93
110
a
Displacement of 3H-labelled Lu AE93103 from the cloned hNK3 receptor
a
Displacement of 3H-labelled Lu AE93103 from the cloned hNK3 receptor
expressed in BHK cells. Data are means (n = 2 or more).
expressed in BHK cells. Data are means (n = 2 or more).

1500
K. Juhl et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1498–1501
Table 3
Table 4
SAR of phenyl piperidine analogues
SAR of piperazine analogues
Cl
N
Cl
Cl
Cl
R
NR1R2
F
O
N
N
O
#
NR1R2
hNK3R Intrinsic
Caco-2
Caco-2
efflux
ratioc
Kia
(nM)
clearanceb
(L/min)
Pappc
#
R
hNK3R Intrinsic
Caco-2 Pappc
Caco-2 efflux
ratioc
(Â10À6 cm/
s)
Kia
clearanceb
(L/min)
(Â10À6 cm/s)
(nM)
O
O
H
HN
N
57
58
59
3.3
3.7
1.8
16
4.3
12
35
NA
27
1.5
NA
1a
49
50
5.9
4.2
6.1
3.5
19
5.6
0.8
0.3
2.6
12
N
N
N
N
O
O
O
HN
HN
N
N
N
O
1.3
>20
17
N
O
O
HN
N
N
H
N
60
61
17
8.5
10
28
1.1
4.2
51
52
1.5
2.8
6.3
0.4
NA
>100
NA
O
O
O
O
O
HN
HN
HN
N
N
N
H
7.3
2.9
N
N
>20
N
O
62
63
64
12
5.6
>20
5.7
25
3.5
1.5
N
H
N
O
S
53
54
5.6
12
6.8
2
0.6
0.6
42
85
N
H
O
N
7.6
5.8
O
O
S
H
N
HN
HN
N
N
N
>20
NA
NA
O
NH2
O
O
65
66
8.7
1.9
15
12
NA
NA
NA
NA
N
N
55
56
3.6
7.8
6.9
15
1.3
0.1
13
NH2
O
S
O
HN
N
N
N
O
O
S
H
N
H2N
H2N
67
68
150
140
0.9
0.9
5.4
11
2.7
1.5
120
O
a
Displacement of 3H-labelled Lu AE93103 from the cloned hNK3 receptor
a
Displacement of 3H-labelled Lu AE93103 from the cloned hNK3 receptor
expressed in BHK cells. Data are means (n = 2 or more).
expressed in BHK cells. Data are means (n = 2 or more).
b
b
See Ref. 9.
See Ref. 10.
See Ref. 9.
See Ref. 10.
c
c
as a cyano group, resulted in compounds (e.g., 35 and 41) with lower
affinity.
However, in contrast to binding, the ADME profile of compound
1a was found to be poor. The intrinsic clearance in human liver
microsomes11 was high relative to the human liver-blood flow
(1.4 L/min). Furthermore, a high efflux ratio and low permeability
were found in the Caco-2 assay12 (Table 3). These results are indica-
tions of poor systemic exposure and the results from the Caco-2 as-
say also imply poor blood–brain barrier penetration. The first
attempt to improve the ADME profile of the lead series, was to vary
the amide functionality located in the phenyl-piperidine-fragment.
As shown in Table 3, several non-basic nitrogen containing
functional groups were tolerated at the 4-position of the piperidine
ring. However, only in one case the clearance was improved; a urea
group rather than an acetamide group gave an intrinsic clearance of
2 L/min, but for this compound an even higher efflux ratio in the
Caco-2 assay was found (54, Table 3).
We speculated that the efflux issue of the compound series might
derive from the phenyl ring in the piperidine fragment. This
prompted us to explore the SAR and ADME properties of these
molecules with a selection of amines that did not carry aromatic
substituents. As shown in Table 4, high affinity for the NK3 receptor
was maintained with a series of N-substituted piperazines. More
Figure 2. Senktide induced hypoactivity (gerbils). Compound 63 was administered
sc, 60 min prior to senktide. Senktide was administered icv, directly before
behavioural analysis was started. Data shown are the mean activity counts over a
period of 30 min after senktide administration. ⁄⁄p <0.01 versus vehicle/senktide
treatment. n = 8–16 per group.

K. Juhl et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1498–1501
1501
importantly, the Caco-2 permeability and efflux were significantly
improved and the efflux ratio approached unity (Papp >10 and efflux
ratio <2 are indicative of good permeability, and that no efflux is in-
volved). However, the stability of the compounds in human liver
microsomes was not improved.
barrier penetration as verified by in vivo activity in the senktide-
induced hypoactivity model. However, to advance this compound
series further, an improvement in metabolic stability needs to be
obtained.
With the aim to improve the metabolic stability, a broader var-
iation of amine substituents NR2 on the structure shown in Table 4
was also performed. The best results from that variation came from
small aliphatic amines such as compounds 67 and 68 (Table 4).
These compounds have a low intrinsic clearance and a relatively
small Caco-2 efflux ratio. However, the affinity for the NK3 receptor
could not be maintained.
Acknowledgements
Maibritt Glarø, Heidi E. Daugaard, Jacob Bøgesvang, Dorthe
Belling, Peter Bregnedal, Bitten Hansen, Mette Larsen, Louise H.
Rosendal are acknowledged for their contributions to this work.
References and notes
We selected compound 63 for in vivo PK studies in rat. Intrinsic
clearance in rat microsomes was 210 mL/min and correlated well
with high in vivo clearance of 108 mL/min/kg. Oral bioavailability
was <5% which is most likely explained by first-pass metabolism,
since the Caco-2 data suggested high permeability and low efflux
ratio. Compound 63 was also selected for in vivo efficacy in the
senktide-induced hypoactivity assay in gerbils. The selection of
this particular compound was based on its low Caco-2 efflux ratio
and high permeability leading to expectation of good CNS penetra-
tion as well as on the high NK3 receptor affinity. Compound 63 was
also found to display selectivity towards the NK1 and NK2 receptors
(Ki = 340 nM and Ki = 75 nM, respectively).8 In this model, central
administration of the selective NK3 receptor agonist senktide in-
duces a reduction of explorative behavior (hypoactivity), which
lasts approximately 30 min.13 The reversal of this functional read
out of NK3 receptor agonism indicates central mediated in vivo
activity of the test compound. Compound 63 was administered sc
in order to avoid expected first-pass metabolism, 60 min prior to
senktide administration and fully reversed senktide-induced hypo-
activity with an ED50 = 14.2 mg/kg ( Fig. 2). Subsequent to the
behavioral test, that is, 90 min after drug administration, four ani-
mals of each dose group were sacrificed for plasma and brain sam-
pling. Compound 63 was found to enter well into the brain, with an
average brain/plasma ratio of 1.9.
1. Langlois, X.; Wintmolders, C.; te Riele, P.; Leysen, J. E.; Jurzak, M.
Neuropharmacology 2001, 40, 242.
2. Nagano, M.; Saitow, F.; Haneda, E.; Konishi, S.; Hayashi, M.; Suzuki, H. Br. J.
Pharmacol. 2006, 147, 316.
3. Rigby, M.; O’Donnel, R.; Rupniak, N. M. J. J. Comp. Neurol. 2005, 490, 335.
4. (a) Albert, J. S.; Potts, W. Expert Opin. Ther. Patents 2006, 16, 925; (b) Smith, P.
W.; Dawson, L. A. Recent Pat. CNS Drug Discov. 2008, 3, 1; (c) Dawson, L. A.;
Smith, P. W. Curr. Pharm. Des. 2010, 16, 344; (d) Simonsen, K. B.; Juhl, K.;
Steiniger-Brach, B.; Nielsen, S. M. Curr. Opin. Drug Discovery Dev. 2010, 13, 379.
5. Meltzer, H. Y.; Arvanitis, L.; Bauer, D.; Rein, W. Am. J. Psychiatry 2004, 161, 975.
6. Mesnage, V.; Houeto, J. L.; Bonnet, A. M.; Clavier, I.; Arnulf, I.; Cattelin, F.; Le, F.
G.; Damier, P.; Welter, M. L.; Agid, Y. Clin. Neuropharmacol. 2004, 27, 108.
7. Kronenberg, G.; Berger, P.; Tauber, R. F.; Bandelow, B.; Henkel, V.; Heuser, I.
Pharmacopsychiatry 2005, 38, 24.
8. NK1 binding: Displacement of 125I-labelled SP from the cloned hNK3 receptor.
NK2 binding: Displacement of 125I-labelled NKA from the cloned hNK3 receptor.
9. Shuto, S.; Ono, S.; Hase, Y.; Kamiyama, N.; Takada, H.; Yamashita, K.; Matsuda,
A. J. Org. Chem. 1996, 61, 915.
10. The lactone-precursors to 31–42, 44–48 (analogues of 4) were synthesized
from (R)-(À)-epichlorohydrin and the appropriate benzyl cyanide as described
in Ref. 7.
11. The stability of compounds in liver microsomes is determined by the t½
method, that is, the disappearance of 1 lM drug over time was measured by
LCMS. Using 0.5 mg/ml of microsomal protein (liver microsomes from several
donors pooled to obtain an average enzyme content) in a NADPH generating
system (1.3 mM NADP, 3.3 mM glucose 6-phosphate and 0.4 U/ml glucose 6-
phosphate dehydrogenase), 3.3 mM MgCl2, 0.1 M potassium phosphate buffer
(pH 7.4), in a total volume of 100 ll, and stopping the incubations at time
points 0, 5, 15, 30 and 60 min with 1:1 v/v acetonitrile. t½ is subsequently
scaled to the metabolic competence of a whole liver using 45 mg microsome/g
liver, 20 g liver/kg and standard body weight 70 kg.
In this Letter, we have described the discovery of a series of tri-
substituted cyclopropanes as novel non-peptidic NK3 receptor
antagonists. Extensive SAR analysis and targeted optimization of
the ADME profile resulted in compounds with good blood–brain
12. Larsen, M.; Holm, R.; Jensen, K. G.; Sveigaard, C.; Brodin, B.; Nielsen, C. U. Eur. J.
Pharm. Sci. 2010, 39, 68.
13. Jung, M.; Michaud, J. C.; Steinberg, R.; Barnouin, M. C.; Hayar, A.; Mons, G.;
Souilhac, J.; Emonds-Alt, X.; Soubrié, P.; Le Fur, G. Neuroscience 1996, 74, 403.