Tetrahydroquinoline sulfonamides as vasopressin 1b receptor anatgonists

Article abstract of DOI:10.1016/j.bmcl.2009.09.050

Vasopressin 1b (V1b) antagonists have been postulated as possible treatments for depression and anxiety. A novel series of potent and selective V1b antagonists has been identified starting from an in-house screen hit. The incorporation of a sulfonamide li

Full text of DOI:10.1016/j.bmcl.2009.09.050

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6018–6022
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Tetrahydroquinoline sulfonamides as vasopressin 1b receptor anatgonists
a
a
a
a
a
Jack D. Scott a, , Michael W. Miller , Sarah W. Li , Sue-Ing Lin , Henry A. Vaccaro , Liwu Hong ,
Deborra E. Mullins b, Mario Guzzi b, Jay Weinstein a, Robert A. Hodgson c, Geoffrey B. Varty c,
Andrew W. Stamford a, Tin-Yau Chan a, Brian A. McKittrick a, William J. Greenlee a,
Tony Priestley b, Eric M. Parker b
*
a Department of Chemical Research, Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
b Department of Central Nervous System, Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
c Department of Neurobiology, Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Vasopressin 1b (V1b) antagonists have been postulated as possible treatments for depression and anxi-
ety. A novel series of potent and selective V1b antagonists has been identified starting from an in-house
screen hit. The incorporation of a sulfonamide linker between a tetrahydroisoquinoline core and amino
piperidine lead to the identification of a V1b antagonist with similar affinity for human and rat receptors.
Further optimization of the right hand portion afforded potent V1b antagonists that possessed moderate
to high selectivity over other receptors.
Received 26 August 2009
Revised 11 September 2009
Accepted 14 September 2009
Available online 17 September 2009
Keywords:
Ó 2009 Elsevier Ltd. All rights reserved.
Vasopressin
V1b antagonist
V1b receptor
Anxiety
Depression
Vasopressin (AVP) is a nonapeptidic hormone that plays a role in
several physiological processes. Three vasopressin receptors, V1a,
V1b and V2, have been identified and cloned to date. Vasopressin
V1a receptors are primarily responsible for vasoconstriction while
the V2 receptor regulates water homeostasis. Vasopressin V1b
receptors (also known as V3) are expressed at high levels in the
anterior pituitary and at lesser levels elsewhere in the brain. Vaso-
pressin has been shown to regulate the hypothalamic–pituitary–
adrenal (HPA) axis by stimulating the release of ACTH1 and acts
synergistically with corticotropin releasing factor (CRF).2 Over-
activity of the HPA axis has been linked to stress-related disorders,
including depression and anxiety. A role for the V1b receptor in
depression was identified through the use of V1b receptor knock-
out mice.3 ACTH levels were significantly lower in the knock-out
mice than in the control wild type mice following a forced swim
stress. It is believed that V1b receptor antagonists would be effec-
tive therapeutic agents for the treatment of anxiety and/or depres-
sion. Furthermore, structurally unique V1b antagonists are of great
interest as tools for pharmacological characterization.
ulate the release of ACTH. The selectivity of SSR149415 over other
receptor subtypes that bind to vasopressin (V1a, V2 and oxytocin5)
is high in the rat; however, it has been reported that this com-
pound is also a potent antagonist of the human oxytocin receptor.6
Recently, a selective and potent V1b antagonist, Org, has been de-
scribed.7 This compound has also been shown to regulate the re-
lease of ACTH in rats in response to various acute stressors.
In-house screening identified the tetrahydroquinoline sulfon-
amide (2)8 as a moderately active V1b receptor antagonist with a
human Ki value of 114 nM. However, there was little activity
against the rat receptor (Ki = 4456 nM). A goal of this project was
to identify selective V1b antagonists that possessed similar affini-
ties for both the human and rat receptors.9 This would allow for
the use of in vivo rodent models to evaluate compounds of interest.
Initially, an effort was made to prepare analogs of 2 by replacing
the carbamate moiety and varying the linker length between the
tetrahydroquinoline and the aminopiperidine (see Fig. 1).
Amide linked analogs (Scheme 1) were prepared from a com-
mon tetrahydroquinoline methyl ester 9 which was readily pre-
pared from quinolinone 6. Quinolinone 6 was protected as the
carbamate followed by reduction of the lactam to afford the ami-
no alcohol 7. Oxidation of the alcohol to form aminol 8 was
accomplished using sulfur trioxide pyridine complex. Ester 9
was prepared via treatment of 8 under Wittig conditions. Amide
11 was prepared directly from 9 first by hydrolysis of the ester
The potent V1b antagonist SSR149415 (1) has been shown to
have both anxiolytic and anti-depressive properties in several ro-
dent models.4 This antagonist has also been shown to directly reg-
* Corresponding author. Tel.: +1 908 740 3462; fax: +1 908 740 7441.
E-mail address: jack.scott@spcorp.com (J.D. Scott).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.09.050

J. D. Scott et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6018–6022
6019
yields. The sulfonamides 23a–23c were completed via coupling
with the commercially available piperidine ketal followed by ke-
tone formation and finally reductive amination. The urea analogs
were prepared starting with alcohol 218 (Scheme 3). Formation
of the amines 24 and 25 was followed by condensation with the
activated urea of piperidine 5 to provide the ureas 26a and 26b.
The binding data for these analogs are summarized in Table 1.10
The optimum linker length of these compounds was found to be 3
atoms between the piperidine nitrogen and the tetrahydroquino-
line. This trend was observed in both the amide and sulfonamide
linker series, compounds 14 and 23b respectively. Overall, com-
pared to the carbamate 2, the human V1b affinities decreased for
all of the analogs that were prepared. The urea 26b did have a com-
parable affinity for the human receptor, but the affinity for the rat
receptor was not improved. The only compound that achieved sim-
ilar affinities between the human and rat receptors, albeit with five
times lower affinity compared to carbamate 2, was the sulfon-
amide linked 23b. All further structure–activity investigations
were based on this general structure.
Conversion to the 2,4-dimethoxybenzene sulfonamide, as is
present on SSR149415 (1), provided analogs with a substantial
improvement in both human and rat V1b binding affinities. As
shown in Table 2, a 10-fold improvement in binding was observed
between 30a (hV1b Ki = 66 nM) compared to the 4-chlorobenzene
sulfonamide 23b (hV1b Ki = 600 nM). With the 2,4-dimethoxyben-
zene sulfonamide moiety fixed, substitution of the basic amine was
explored (Scheme 4). To facilitate this, the chemistry route was
modified to include a more robust and reproducible formation of
the sulfonyl chloride. Ester 9 was reduced and the t-butyl carba-
mate was exchanged for a methyl carbamate that was compatible
OMe
OH
NMe2
S
H
N
Cl
N
O
O
N
SO2
N
O
O
SO2
Cl
MeO
OMe
1 SSR149415
2
hV1b Ki = 0.6 nM
rV1b Ki = 1 nM
hV1a Ki = 22 nM
hV2 Ki = 325 nM
hoxt Ki = 19 nM
hV1b Ki = 114 nM
rV1b Ki = 4456 nM
Figure 1.
followed by amide coupling with piperidine 5 to afford 10. Cleav-
age of the t-butyl carbamate of 10 followed by sulfonamide for-
mation and allyl carbamate deprotection afforded amide 11.
Amides 14 and 16 were prepared from homologated analogs of
9 using similar chemistry.
Amine, sulfonamide and urea linked analogs were also prepared
(Schemes 2 and 3). Amine 18 was prepared through a reductive
amination of aldehyde 17 with the amino piperidine 5, while the
homologated amine 20 was prepared through a mesylate displace-
ment. The three sulfonamide analogs were prepared by the activa-
tion of the hydroxyl moiety of alcohols 12, 19 and 218 as a
mesylate or iodide. Displacement of the leaving group by either a
sulfate or a thiolate was followed by oxidation and/or chlorination
to afford the sulfonyl chlorides in moderate, but not reproducible
S
O
AllocN
S
a-c
+
NH2
N
N
H
Boc
3
4
5
OH
g
d-e
f
CO2Me
N
O
NH
Boc
N
OH
N
H
Boc
Boc
6
7
8
9
O
O
j-l
h, i
N
N
SO2
N
S
N
N
S
Boc
N
H
Alloc
Cl
10
11
S
H
N
N
o,p
m,k,n
i,l
N
SO2
OH
N
SO2
CO2H
N
SO2
9
O
Cl
Cl
Cl
12
13
14
O
q,g
r,s,i,l
N
SO2
N
CO2Me
9
S
N
N
H
Boc
15
Cl
16
Scheme 1. Reagents and conditions: (a) NaBH(OAc)3, AcOH, DCE, rt, 16 h (48%); (b) AllocCl, K2CO3, EtOAc/H2O, rt, 16 h (99%); (c) TFA, CH2Cl2, 1,3-dimethoxybenzene, rt, 3 h
(69%); (d) Boc2O, DMAP, Et3N, CH2Cl2, 0 °C–rt, 16 h; (e) NaBH4, EtOH, 0 °C–rt, 3 h (57%, two steps); (f) SO3–pyridine, DMSO, rt, 16 h (99%); (g) sodium
trimethylphosphonoacetate, THF, reflux 2.5 h (56–77%); (h) 2 M LiOH (aq), EtOH, rt, 3 h; (i) 5, EDCI, HOBt, iPr2NEt, MeCN, rt, 16 h (53–83%); (j) TFA, CH2Cl2, 3 h, rt; (k)
4-Cl–PhSO2Cl, DMAP, pyr (39–46%); (l) Pd(OAc)2; Et2NH, P(m-SO3Na–Ph)3, H2O, MeCN (69%); (m) 4 N HCl (dioxane), MeOH; (n); DiBAL, toluene, 0 °C (o) acetonecyanohydrin;
DIAD, PPh3, THF À20 °C to rt (93%); (p) 20% KOH (aq); EtOH, reflux (85%); (q) DiBAL, tol, À78 °C; (r) H2, Pd/C, MeOH; (s) 1 M NaOH (aq), MeOH, 65 °C (96% two steps).

6020
J. D. Scott et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6018–6022
a
N
N
SO2
S
CO2Me
CHO
N
N
N
H
Boc
Boc
Cl
9
17
18
S
H
N
f,g
OH
N
h-l
N
SO2
CO2H
N
SO2
N
SO2
Cl
Cl
Cl
13
19
20
O2
S
h,m,n
or
h, p-r
s,k,l
OH
n
SO2Cl
n
N
N
SO2
N
N
SO2
S
n
SO2
N
H
Cl
Cl
Cl
21 n = 1
12 n = 2
19 n = 3
22a n = 1
22b n = 2
22c n = 3
23a n = 1
23b n = 2
23c n = 3
Scheme 2. Reagents and conditions: (a) DiBAL, tol, À78 °C 3 h; (b) 5, NaBH(OAc)3, CH2Cl2, 16 h (56%); (c) TFA, CH2Cl2, 3 h (94%); (d) 4-Cl–PhSO2Cl, DMAP, pyr; (e) Pd(OAc)2;
Et2NH, P(m-SO3Na–Ph)3, H2O, MeCN, 16 h; (f) MeOH, H2SO4, reflux, 16 h (63%); (g) DiBAL, toluene 0 °C–rt (99%); (h) MsCl, Et3N, CH2Cl2 16 h; (i) NaI, acetone, reflux, 24 h; (j)
1,4-dioxa-8-azaspiro[4.5]decane, Cs2CO3, MeCN, reflux 5 h (74%); (k) 4 N HCl (aq), dioxane, reflux, 16 h (77%); (l) 2-thiophenemethylamine, NaBH(OAc)3, CH2Cl2, 16 h (63%);
(m) Na2SO3, EtOH, H2O, reflux 16 h, (n) oxalyl chloride, DMF, PhH, reflux; (o) MsCl, Et3N, CH2Cl2; (p) KSAc, DMF 55 °C; (q) NaOMe, MeOH 55 °C, 2 h; (r) KNO3, SO2Cl2, MeCN
0 °C (74%, three steps); (s) 1,4-dioxa-8-azaspiro[4.5]decane, iPr2NEt, CH2Cl2, rt, 16 h (28–71%).
OH
NH2
NHMe
a-c
d-f
N
SO2
N
SO2
N
SO2
Cl
Cl
Cl
25
21
24
S
H
S
N
AllocN
R
g-j
N
N
N
SO2
O
N
H
Cl
5
26a R = H
26b R = Me
Scheme 3. Reagents and conditions: (a) MsCl, Et3N; (b) NaN3, 18-C-6, DMF; (c) PS–PPh3, THF, H2O; (90% two steps); (d) Boc2O, CH2Cl2 (89%); (e) NaH, MeI, DMF (92%); (f) HCl,
dioxane (100%); (g) CDI, THF; (h) MeI, MeCN; (i) 24 or 25; (j) Pd(OAc)2; Et2NH, P(m-SO3Na–Ph)3, H2O, MeCN, 16 h (47–70% three steps).
with the acidic conditions used to prepare the sulfonyl chloride.
Conversion of the alcohol to the thioacetate followed conditions
described previously. A one-step cleavage of the thioacetate and
oxidation to the sulfonic acid was accomplished in high yield using
H2O2 in acetic acid.11 Treatment with Pd/C eliminated the excess
hydrogen peroxide. Chlorination was accomplished using phos-
gene to afford the sulfonyl chloride 28 in excellent yields. Coupling
with the previously mentioned piperidine ketal was followed by
methyl carbamate cleavage and sulfonylation. Subsequent ketal
cleavage afforded the ketone 29 which was treated with primary
amines under reductive amination conditions to afford the final
targets. These reductive aminations were either accomplished in
solution (Method A) or via high throughput library synthesis
(Method B).
Table 1
Structure–activity relationship of various linkers
X
n
N
N
SO2
S
N
H
Cl
Compound
n
X
hV1ba Ki (nM) rV1ba Ki (nM) hV1aa Ki (nM)
2
11
14
16
18
20
23a
23b
23c
26a
26b
1
1
2
3
2
3
1
2
3
1
1
–OC(O)–
–C(O)–
–C(O)–
–C(O)–
—
114 40
1523 77
715 37
1137 28
7196 376
1184 25
748 58
600 116
1036 21
416 103
4456 356
>10,000
>10,000
3244 408
9587 998
2776 500
2690 90
813 76
6164 1234
3654 85
6888 472
>8000
>10,000
—
4617 324
8734 703
4963 882
5564 657
>10,000
–S(O)2–
–S(O)2–
–S(O)2–
–N(H)C(O)–
As shown in Table 2, the two methylenethiophene analogs, 30a
and 30b, and the unsubstituted benzyl 30c analogs exhibited sim-
ilar affinities for both species. Substitution of the benzyl group de-
creased the affinity especially in the 4-methoxy example 30e
3941 74
>10,000
>10,000
–N(Me)C(O)– 295 43
8866 2309
(hV1b Ki = 939 nM). Furthermore, substitution
group also decreased the affinity significantly (examples 30g and
30l). Aliphatic side chains were equipotent to the benzyl analogs
a to the amine
a
Inhibition of 2 nM [3H]Arg8-vasopressin (AVP) binding to human or rat
vasopressin receptors. All values reported are an average of three independent
results.

J. D. Scott et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6018–6022
6021
Table 2
Structure–activity relationship of right-hand amines
NHR
N
N
SO2
S
O2
MeO
OMe
Compound
R
Methoda
hV1bb Ki (nM)
rV1bb Ki (nM)
hV1ab Ki (nM)
S
30a
A
B
B
B
B
66
96
4
7
7
23
3
720 60
33 97
30b
30c
30d
30e
145 29
S
104
83
8
8
192 26
395 23
4926 1092
218 12
939 58
158
F
637 29
OMe
30f
30g
30h
B
B
B
279 24
717 52
270
9
2600 477
4803 920
486 46
OMe
999 108
122
7
91 11
30i
B
B
B
B
A
A
152
35
3
2
2
227
27
7
3
4
1268 112
947 85
1257 125
709 86
553 49
188 16
30j
30k
30l
78
62
413 39
138 13
557 27
30m
30n
56
6
O
70
4
16
5
a
Method of preparation as listed in Scheme 4 step k.
Inhibition of 2 nM [3H]Arg8-vasopressin (AVP) binding to human or rat vasopressin receptors. All values reported are an average of three independent results.
b

6022
J. D. Scott et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6018–6022
O
NHR
g-j
a-c
d-f
k
N
N
9
N
SO2
S
N
SO2
S
OH
SO2Cl
N
N
O2
O2
CO2Me
CO2Me
MeO
OMe
MeO
OMe
27
28
29
30a-n
Scheme 4. Reagents and conditions: (a) LiBH4, EtOH, THF; (b) 4 N HCl (dioxane), MeOH; (c) ClCO2Me, K2CO3, EtOAc, H2O (80% three steps); (d) MsCl, Et3N, CH2Cl2; (e) KSAc,
DMF (85% two steps); (f) H2O2, AcOH; Pd/C, then phosgene, DMF, CH2Cl2 (100%); (g) 1,4-dioxa-8-azaspiro[4.5]decane, iPr2NEt, CH2Cl2 (79%); (h) LiSBu, HMPA (63%); (i) 2,4-
dimethoxybenzenesulfonyl chloride, DMAP, pyr (79%); (j) 4 N HCl (aq), dioxane (81%); (k) Method A: RNH2, NaBH(OAc)3, AcOH, CH2Cl2; Method B: RNH2, PS-BH(OAc)3, DCE,
MeCN, AcOH; PS-b-keto ester; MP-TsOH; NH3 (MeOH).
Table 3
Structure–activity relationship of separated enantiomers of 30n
H
N
30o
(Enantiomer A)
30p
(Enantiomer B)
+
Chiralpak AS
N
N
SO2
S
O2
MeO
OMe
30n
Compound
hV1ba Ki (nM)
rV1ba Ki (nM)
hV1aa Ki (nM)
rV1aa Ki (nM)
hV2a Ki (nM)
hOxtb Ki (nM)
30o
30p
208 16
874 62
1129 123
314 25
3718 169
1912 502
>10,000
>10,000
>10,000
5525 372
44
1
21
1
a
Inhibition of 2 nM [3H]Arg8-vasopressin (AVP) binding to human or rat vasopressin receptors. All values reported are an average of three independent results.
Inhibition of 2 nM [3H]Oxytocin binding to human oxytocin receptor. All values reported are an average of three independent results.
b
4. For a review of SSR149415 see: Serradeil-Le Gal, C.; Wagnon, J.; Tonnerre, B.;
when there was branching in the b- or
amine group (30k and 30j, respectively).
c- position relative to the
Roux, R.; Garcia, G.; Griebel, G.; Aulombard, A. CNS Drug Rev. 2005, 11, 53.
5. Oxytocin is a nonapeptidic hormone with similar structure to vasopressin.
6. (a) Hodgson, R. A.; Higgins, G. A.; Guthrie, D. H.; Lu, S. X.; Pond, A. J.; Mullins, D.
E.; Guzzi, M. F.; Parker, E. M.; Varty, G. B. Pharmacol. Biochem. Behav. 2007, 86,
431; (b) Griffante, C.; Green, A.; Curcuruto, O.; Haslam, C. P.; Dickinson, B. A.;
Arban, R. Br. J. Pharmacol. 2005, 146, 744.
7. (a) Craighead, M.; Milne, R.; Campbell-Wan, L.; Watson, L.; Presland, J.;
Thomson, F. J.; Marston, H. M.; MacSweeney, C. P. Prog. Brain Res. 2008, 170,
527; (b) Spiga, F.; Harrison, L. R.; Wood, S.; Knight, D. M.; MacSweeney, C. P.;
Thomson, F.; Craighead, M.; Lightman, S. L. J. Endocrinol. 2009, 200, 273; (c)
Spiga, F.; Harrison, L. R.; MacSweeney, C. P.; Thomson, F. J.; Craighead, M.;
Lightman, S. L. J. Endocrinol. 2009, 200, 285.
8. Asberom, T.; Bara, T. A.; Clader, J. W.; Greenlee, W. J.; Guzik, H. S.; Josien, H. B.;
Li, W.; Parker, E. M.; Pissarnitski, D. A.; Song, L.; Zhang, L.; Zhao, Z. Bioorg. Med.
Chem. Lett. 2007, 17, 205.
9. The human and rat V1b receptors have 83% homology: Lolait, S. J.; O’Carroll, A.-
M.; Mahan, L. C.; Felder, C. C.; Button, D. C.; Young, W. S., III; Mezey, E.;
Brownstein, M. J. Proc. Natl. Acad. Sci. 1995, 92, 6783.
To determine the importance of the stereogenic chiral center on
affinity, the enantiomers of 30n were separated via preparative
chiral HPLC.12 The slower eluting enantiomer, 30p, was found to
have greater affinity at both the human and rat V1b receptors com-
pared to the faster eluting enantiomer, 30o (Table 3). Furthermore,
for use as a possible tool in a rat in vivo model to evaluate V1b po-
tency, 30p was found to be 90-fold more selective for the rat V1b
receptor over the rat V1a receptor. The human V2 and oxytocin
affinities10 were also determined for these enantiomers to deter-
mine the selectivities compared to the human V1b receptor. Com-
pound 30p possessed greater than 100-fold V1b selectivity
compared to those two receptors.
In conclusion, a tetrahydroquinolinecore thatcontained a sulfon-
amide linker was identified that possessed equipotent activities in
both human and rat V1b receptor binding assays. From this core po-
tent analogs were prepared, several with double digit nanomolar
Ki’s. Compound 30p was also identified as a possible tool for use in
an in vivo rat model. Subsequent reports will describe the evolution
of this series to provide V1b receptor antagonists with in vivo activ-
ity along with further improvements in affinity and selectivity.
10. Competition binding assays were performed using membranes from CHO-K1
cells expressing the various vasopressin or oxytocin receptors, 2 nM [3H]Arg8-
vasopressin (AVP) or 2 nM [3H]Oxytocin (for the human oxytocin receptor) and
various concentrations of antagonist in 50 mM BisTrisPropane, pH 7.4, 5 mM
MgCl2 and 0.1% BSA (200
l
l
total volume). Non-specific binding was
oxytocin (for the human oxytocin
determined using AVP or 1 lM
1
l
M
receptor). The reaction mixtures were incubated for 60 min at room
temperature, and then filtered through Millipore MAFC glass fiber filter
plates presoaked in 0.5% (v/v) polyethyleneimine. The filters were washed
three times with 150
ll of ice cold 50 mM BisTrisPropane, pH 7.4, and filter-
associated radioactivity was measured in a Packard TopCount scintillation
counter. The data were analyzed using GraphPad PRISM
determined using the Cheng–Prusoff equation.
, and the Ki was
References and notes
11. Brouwer, A. J.; Monnee, M. C. F.; Liskamp, R. M. J. Synthesis 2000, 11, 1579.
12. Analytical chiral HPLC conditions: Chiralpak AS column, 70:30 hexanes/EtOH,
flow 1 mL/min, UV detector: 254 nm. Retention time enantiomer (A) 10 min,
enantiomer (B) 13.6 min. Each enantiomer from preparative scale separation
was determined to be P98% ee via analytical HPLC.
1. Aguilera, G.; Rabadan-Diehl, C. Regul. Pept. 2000, 96, 23.
2. Gillies, G. E.; Linton, E. A.; Lowry, P. J. Nature 1982, 299, 355.
3. Tanoue, A.; Ito, S.; Honda, K.; Oshikawa, S.; Kitagawa, Y.; Koshimizu, T.; Mori,
T.; Tsujimoto, G. J. Clin. Invest. 2004, 113, 302.