Synthesis and SAR of sulfoxide substituted carboxyquinolines as NK3 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.11.003

The neurokinin-3 (NK3) receptor is regarded as a potential novel target for treating patients with schizophrenia. Herein we report the synthesis and SAR of a series of C3-alkylsulfoxide substituted quinolines as potent NK3 receptor antagonists. These compounds have excellent NK3 functional activity, good selectivity and drug-like properties. Several key compounds have good in vitro/in vivo DMPK characteristics, and are active in a gerbil locomotor activity model.

Full text of DOI:10.1016/j.bmcl.2010.11.003

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1896–1899
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and SAR of sulfoxide substituted carboxyquinolines as NK3
receptor antagonists
Hui Xiong, James Kang, James M. Woods, John P. McCauley Jr., Gerard M. Koether, Jeffrey S. Albert,
⇑
Lindsay Hinkley, Yan Li, Reto A. Gadient, Thomas R. Simpson
CNS Discovery, AstraZeneca Pharmaceuticals, 1800 Concord Pike, Wilmington, DE, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The neurokinin-3 (NK3) receptor is regarded as a potential novel target for treating patients with
schizophrenia. Herein we report the synthesis and SAR of a series of C3-alkylsulfoxide substituted quin-
olines as potent NK3 receptor antagonists. These compounds have excellent NK3 functional activity, good
selectivity and drug-like properties. Several key compounds have good in vitro/in vivo DMPK character-
istics, and are active in a gerbil locomotor activity model.
Received 24 August 2010
Accepted 1 November 2010
Available online 21 December 2010
Keywords:
Neurokinins
Ó 2010 Elsevier Ltd. All rights reserved.
NK3 receptor antagonists
Carboxyquinolines
Sulfoxides
Neurokinin B (NKB), a tachykinin peptide, mainly elicits its effect
through the G-protein coupled neurokinin-3 (NK3) receptor. The
NK3 receptor is almost exclusively expressed in the central nervous
system, and has been demonstrated to affect monoaminergic and
other neurotransmission. Therefore, it is considered as a potential
new and druggable target for treatment of patients with schizophre-
nia.1 In addition to treating schizophrenia, NK3 receptor (NK3r)
antagonists mayalso provide benefits in the treatmentof pain, mood
disorder, Parkinson’s disease and other CNS indications.2
side-effects sometimes observed with currently marketed
antipsychotics such as extra pyramidal symptoms, weight gain
and cognitive deficits.
The carboxyquinoline core of Talentant has been demonstrated
as a viable scaffold for selective NK3r antagonists,6 and can tolerate
a wide variety of substituents at the C3 position of the quinoline
ring. This position has been extensively explored to optimize
selectivity as well as drug-like properties.7 Due to its large dipolar
character and chirality, the sulfoxide is a unique functional group
in medicinal chemistry and can be found in several drugs, most
notably, esomeprazole (Nexium). This paper describes our entry
into this area through the discovery of novel, potent and drug-like
compounds based on C3-alkylsulfoxide substituted quinolines.
Chemistry: The synthesis generally started from the Pfitzinger
condensation8 of suitable isatins with aryl keto substituted alkyl
sulfides to give the corresponding C3-alkylthio substituted quino-
line carboxylic acids in good yields (Scheme 1). Amide coupling
and subsequent sulfide oxidation led to the desired sulfoxides as
a mixture of isomers. Various Pd-catalyzed coupling reactions
involving aryl halide groups enabled access to substitution at var-
ious positions on the quinoline ring. As shown in Scheme 2, bromo-
methyl substituted carboxyl quinoline, prepared according to
literature procedure,7a was converted into C3-methylsulfinylmeth-
yl substituted quinoline amides in several steps.
Cl
Cl
O
NH
OH
N
3
N
N
N
O
O
Osanetant
Talentant
Two structurally distinct and selective NK3r antagonists, Osane-
tant3 (Sanofi-Aventis) and Talentant4 (GlaxoSmithKline) have been
reported to show efficacy in improving the positive symptoms of
schizophrenia in phase II clinical studies. However, development
of both drug candidates have been terminated5 without clearly
demonstrating whether selective NK3r antagonists can provide
sufficient antipsychotic efficacy while avoiding the undesired
As shown in Scheme 3, para-methoxybenzyl substituted inter-
mediate, which was made in the fashion depicted in Scheme 1,
was converted to free thio when heated with TFA and anisol.
Alkylation and subsequent oxidation led to a variety of alkyl
substituted sulfoxides.
⇑
Corresponding author.
E-mail address: thomas.simpson1@verizon.net (T.R. Simpson).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.11.003

H. Xiong et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1896–1899
1897
O
OH
Table 2
O
R1
NK3 functional activity assay results for compounds 8–20
R1
S
a
S
n
+
n
O
X
X
Ph
N
Ar
O
N
Ar
H
NH
O
O
n = 0 or 2
S
R1
b, c
N
R2
R3
R4
R3
R4
NH
Compda
R1
R2
EC50 (nM)
Sol.b
(l
M)
a
O
NH
O
d
2
8
9
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Ethyl
i-Propyl
CH2CONMe2
Ph
4.9 2.6
81 14
18
48 11
1.7 0.6
1.8 1.3
1120 40
690 520
110 12
220 90
2500 740
6.9 1.8
230
53
100
88
47
28
360
470
41
390
420
170
83
R1
S
O
R1
S
O
2-F phenyl
3-F phenyl
4-F phenyl
2-Thiophenyl
3-Thiophenyl
2-Pyridyl
3-Pyridyl
2-Thiazol-4-yl
2-Thiazol-5-yl
3-Pyrazolyl
n
n
R2
X
3
N
Ar
N
Ar
10
11
12
13
14
15
16
17
18
19
20
Scheme 1. Reagents and conditions: (a) NaOH, EtOH/THF, reflux; (b) amine, EDCI,
HOBt, DIPEA, DCM; (c) NaIO4, MeOH/H2O,
D
; (d) Pd-catalyzed coupling reactions.
Ph
O
OMe
O
OH
O
NH
Ph
Ph
Ph
c, d
a, b
16
12 0.2
4
Br
Ph
S
S
52
N
N
Ph
O
a,b
N
Ph
See footnote a in Table 1.
Scheme 2. Reagents and conditions: (a) MeSNa, THF; (b) LiOH, THF/H2O; (c) amine,
EDCI, HOBt, DIPEA, DCM; (d) NaIO4, MeOH/H2O,
D
.
Table 3
NK3 functional activity assay results for compounds 21–29
Ph
Ph
Ph
Ph
O
NH
O
NH
SH
O
NH
SPMB
O
a
b, c
O
NH
O
S
S
R
7
N
Ph
R
N
Ph
N
Ph
N
Ph
8
Compda
R
EC50 (nM)a
Sol. (
l
M)b
Scheme 3. Reagents and conditions: (a) TFA, anisol,
D; (b) RI, K2CO3, DMF; (c)
NaIO4, MeOH/H2O,
D.
2
H
8-F
8-Me
7-F
7-Cl
7-EtO
7-Me2N
7-Br
7-Me
4.9 2.6
8.2 2.3
1900 800
13 10
230
160
16
58
34
69
44
15
160
460
21
22
23
24
25
26
27
28
29
Table 1
7.7 2.4
NK3 functional activity, solubility and permeability for 1–7
38
1
Ph
212 13
5.1 1.2
3.6 1.0
88 58
O
NH
R
7-Me2NCH2
a,b
See footnote a in Table 1.
N
Ph
Sol.b
(lM)
Pappc/efflux ratio
a
Compd
R
EC50 (nM)
Table 4
NK3 functional activity assay results for compounds 30–35
1
2
SMe
SOMe
SOMe (S)
SOMe (R)
SO2Me
36 10
4.9 2.6
3.7 3.4
240 120
<1
230
270
44
36
2
8
53
4
nde
9.4/2.7
28/1.3
17/2
11/2.6
5.9/1.1
10/2.6
9.9/3.7
11/3.1
1/21
2ad
2bd
3
R1 R2
O
NH
14
35
4
8
O
S
4
CH2SMe
5ad
5bd
6
CH2SOMe (R)
CH2SOMe (S)
CH2SO2Me
CH2CH2SOMe
1.8 0.8
5.3 2.7
2.5 0.7
3.3 0.5
N
Ph
7
360
a
Compd
R1
R2
Ethyl
EC50 (nM)
Sol.b
(lM)
a
Unless noted, compounds were tested as 1:1 mixtures of diasteromers; EC50’s
are reported as the mean SEM (n P 2) and without SEM where n = 1.
2
Ph
4.9 2.6
230
180
1.3
4.5
b
30
31
Ph
Ph
Cyclopropyl
(CH2)2
15
16
7
9
Equilibrium solubility.
Measured in MDCK cells expressing human MDR1; unit as 10À6 cm/s.
c
d
32
Ph
(CH2)3
38 15
>10,000
>9300
Single isomer.
nd = not determined.
33ac
33bc
34
4-Pyridyl
4-Pyridyl
3-Pyridyl
2-Pyridyl
2-Pyridyl
Cyclopropyl
Cyclopropyl
Ethyl
Ethyl
Ethyl
99
e
370
350
420
460
74
9
35ac
35bc
77 15
>5200
Results: The primary assay used to assess project SAR was a
functional assay employing FLIPR technology in CHO cells stably
expressing human recombinant NK3 receptor. Antagonism was
a,b,c
See footnote a in Table 1.

1898
H. Xiong et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1896–1899
Table 5
Further profiling of compounds 2a, 5a and 5b
Compd EC50 (nM) Selectivity
Selectivity
Sol. (lM) CLint
(
l
L/min/mg) Papp (10À6 cm/s) Efflux ratio Log D hERG (
lM) LMA IP ED50 (lmol/kg)
versus NK1 versus NK2
2a
5a
5b
3.7 3.4
1.8 0.8
5.3 2.7
>2700
nd
>660
1860
nd
170
280
8
53
23
5.2
8.4
28
10
9.9
1.3
2.6
3.7
3
3.5
3.5
>33
>33
>33
0.75
22
2
measured as inhibition of a transient Ca2+ flux induced by senktide,
an NK3r peptide agonist, which has >3000-fold selectivity for NK3r
over NK1r & NK2r. Key physical property (solubility) and DMPK
characteristics (permeability, Pgp efflux ratio, metabolic clearance)
were also measured. Most compounds were screened as a roughly
1:1 mixture of diastereomers. Key compounds were resolved into
single diastereomers and studied further.
As shown in Table 1, the sulfoxide substitution at the C3 posi-
tion of the quinoline ring generally gave very active and soluble
compounds. By changing the linker length and the stereochemistry
of sulfoxide, we were able to obtain superior compounds. The per-
meability and efflux ratios of these sulfoxide compounds was
greatly improved when the sulfoxide moiety was moved closer
to the quinoline ring. Therefore, shortening the methylene chain
length greatly reduced the efflux ratio. The sulfoxide stereochem-
istry had great influence not only on the NK3 functional activity,
but also on other physical properties. These compounds were quite
stable under thermal conditions, but had a short half-life under
light. Other oxidation states of sulfur atom were also explored. Sul-
fides gave very low solubility and very high clearance, while sulf-
ones produced very good functional activity, but were less
soluble than the sulfoxides.
Several different amides at the 4-carboxy position were
examined as well (Table 4). Cyclopropyl group replacement of
the ethyl group led to a slightly less active compound. Surprisingly,
introduction of either achiral 1-phenylcyclopropanamine (entry
31) or 1-phenylcyclobutanamine (entry 32) still gave fairly active
compounds, but this caused a precipitous drop in solubility.
Replacement of the phenyl ring with a pyridyl ring was intended
to introduce an ionizable basic nitrogen atom. However, 2- or
3-pyridyl groups in this position resulted in a loss of activity, while
switching to a 4-pyridyl group led to even more dramatic loss of
activity.
Several potent compounds were selected based on their in vitro
profiles and evaluated in a gerbil locomotor activity assay (Table
5). Injection of senktide, a selective NK3 peptide agonist, into the
lateral ventricle of male Mongolian gerbils acutely suppresses loco-
motion. The magnitude of suppression is dependent on the dose of
senktide injected. This effect can be prevented by systemic pre-
administration of NK3r antagonists. Alkylsulfoxide 2a, with its
excellent selectivity against other neurokinin receptors, good
in vitro DMPK profile and physical properties, was found to dose
dependently reverse the response to senktide by ip dosing (ED50
of 0.75 lmol/kg). Compound 2a exhibited acceptable pharmacoki-
After determining the optimal sulfur oxidation state and linker
length, we explored various aryl groups at the 2 position of quino-
line ring, as well as alkyl groups on the sulfoxide (Table 2). Fluorine
substitution on the C2 phenyl ring led to less active compounds.
Among them, meta F had the best activity (entry 9). Either 2-, or
3-thiophenyl groups gave highly active compounds, which were
metabolically unstable in human liver microsomes (hCLint ꢀ220
netics in vivo, in rat (F% ꢀ60%) with a high clearance in plasma
(ꢀ60 mL/min/kg), and moderate t1/2 (3 h). The total brain/plasma
ratio in rat was 0.4 @ 1 h post dose (3 lmol/kg iv dosing).
Interestingly, potent compound 5a was found to be very weak
(ED50 22 mol/kg) in the gerbil locomotor assay, presumably due
to its limited solubility and absorption. In contrast, its diastereo-
mol/kg)
l
mer 5b displayed excellent in vivo activity (ED50 ꢀ2
l
and 190
ll/min/mg, respectively). We attempted to further im-
and DMPK profile. The apparent oral bioavailability of this com-
pound was very high in dog (F% ꢀ89%); the plasma clearance at
2.3 mL/min/kg, and suitable half-life at 6.7 h.
prove aqueous solubility via introduction of nitrogen containing
heterocycles (pyridine, thiazole, and pyrrazole) at the C2 position.
This led to greater solubility, but it also resulted in greatly reduced
functional activity.Introduction of various simple alkyl groups on
the sulfoxide sulfur atom was tolerated for activity, but was shown
to decrease metabolic stability of the compounds in human liver
microsomes. Addition of an alkyl amide group to the sulfoxide
was tolerated (entry 20). However, attaching such a polar groups
to the sulfoxide led to low permeability (1.5 Â 10À6 cm/s) and high
efflux ratio (ꢀ20).
The chirality of the sulfur atom in compound 5b was deter-
mined by X-ray analysis to be (S)-configuration. In the crystal
structure, intra-molecular hydrogen bonds form between the
amide NH group and the sulfoxide O atom, resulting in an eight-
membered ring. The absolute stereochemistry of the sulfoxide chi-
ral center in compound 2a was determined to be (S)-configuration
as well by Vibrational Circular Dichroism (VCD) analysis.9
In conclusion, C3-alkylsulfoxide substituted carboxyquinolines
were readily prepared, starting from the Pfitzinger condensation
between arylketo sulfides and isatins. Several of these compounds
were shown to have excellent NK3 functional activity, and good
selectivity against other neurokinin receptors. One key SAR obser-
vation was the influence of linker length on permeability and efflux
ratios. Two key compounds, 2a and 5b, had excellent in vitro/
in vivo DMPK characteristics, and are active in the gerbil locomotor
activity model. The absolute stereochemistry of the sulfoxide chiral
center in compounds 2a and 5b was determined to be (S)-configu-
ration, by VCD and X-ray analysis, respectively.
Next we explored SAR of the C7 and C8 positions of the quino-
line ring (Table 3). Fluorine substitution at the C8 position of the
quinoline gave similar activity to simple phenyl analog 2. Surpris-
ingly, introduction of a methyl group in this position greatly re-
duced NK3 functional activity. This led us to hypothesize that the
receptor might be very sensitive to steric hindrance around the
C8 position.
Compared with C8 position, C7 position was much more toler-
ant of substitution. This position had a preference for small, lipo-
philic groups like halogen atoms or a methyl group. However,
introduction of these groups led to less soluble compounds. Instal-
lation of electron donating groups, such as ethoxy- or dimethyl-
amino- groups, afforded less active compounds. A basic amine
group was introduced to improve solubility and allow for possible
salt formation. However, this change also reduced the activity by
ꢀ20-fold.
Acknowledgements
The authors are grateful to Jennifer Van Anda and the purifica-
tion group for SFC separation of chiral sulfoxides, James Hall for
NMR support, Donald Pivonka and Steven Wesolowski for VCD

H. Xiong et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1896–1899
1899
4. (a) Evangelista, S. Curr. Opin. Investig. Drugs 2005, 6(7), 717; (b) Dawson, L. A.
et al Neuropsychopharmacology 2008, 33, 1642; (c) Flor, R.; Dawson, L. A.
Neuropharmacology 2009, 56, 342.
5. Quartara, L.; Altamura, M.; Evangelista, S.; Maggi, C. A. Expert Opin. Investig.
Drugs 2009, 1812, 1843.
6. (a) Giardina, G. A. M. et al J. Med. Chem. 1996, 39, 2281; (b) Giardina, G. A. M. et al
J. Med. Chem. 1997, 40, 1794; (c) Giardina, G. A. M. et al J. Med. Chem. 1999, 42,
1053.
support, Dedong Wu for X-ray analysis, My Linh Do for gerbil
locomotor assay, Mark Eisman for in vivo DMPK support. Peter
Bernstein, Marc Chapdelaine and Chad Elmore for helpful
discussion.
References and notes
7. (a) Blaney, F. E. et al J. Med. Chem. 2001, 44, 1675; (b) Smith, P. W. et al Bioorg.
Med. Chem. Lett. 2009, 19, 837; (c) Elliott, J. M. Bioorg. Med. Chem. Lett. 2006, 16,
5748. and 5752.
8. (a) Pfitzinger, W. J. Prakt. Chem. 1886, 33, 100; (b) Shvekhgeimer, M. G.-A. Chem.
Heterocycl. Compd. 2004, 40, 257.
1. (a) Spooren, W.; Riemer, C.; Meltzer, H. Nat. Rev. Drug Disc. 2005, 4, 967; (b)
Albert, J. S.; Potts, W. Expert Opin. Ther. Pat. 2006, 16, 925; (c) Quartara, L.;
Altamura, M. Curr. Drug Targets 2006, 7, 975; (d) Dawson, L. A.; Smith, P. W. Curr.
Pharm. Des. 2010, 16, 344.
2. Simonsen, K. B.; Juhl, K.; Steiniger-Brach, B.; Nielsen, S. M. Curr. Opin. Drug
Discov. Devel. 2010, 13, 379. and references therein.
3. Kamali, F. Curr. Opin. Investig. Drugs 2001, 2(7), 950.
9. Nafie, L. A.; Dukor, R. K. In Applications of Vibrational Spectroscopy in
Pharmaceutical Research and Development; Pivonka, D. E., Chalmers, J. M.,
Griffiths, P. R., Eds.; John Wiley & Sons, 2007; pp 129–154.