The characterization of a novel V1b antagonist lead series

Article abstract of DOI:10.1016/j.bmcl.2010.11.061

The SAR around a V1b antagonist HTS hit 3 was explored to produce a series of thiazole sulfonamides as a lead series with selectivity over the related V1 and oxytocin receptors.

Full text of DOI:10.1016/j.bmcl.2010.11.061

Bioorganic & Medicinal Chemistry Letters 21 (2011) 92–96
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The characterization of a novel V1b antagonist lead series
Chris A. Smethurst a, Jennifer A. Borthwick a, , Simon Gaines a, Steve Watson a, Andrew Green a,
⇑
Mark J. Schulz b, George Burton b, Alberto A. Buson c, Roberto Arban c
a GlaxoSmithKline Research and Development, New Frontiers Science Park, Harlow, UK
b GlaxoSmithKline Research and Development, Upper Providence, USA
c GlaxoSmithKline Research and Development, Verona, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The SAR around a V1b antagonist HTS hit 3 was explored to produce a series of thiazole sulfonamides as a
lead series with selectivity over the related V1 and oxytocin receptors.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 29 October 2010
Revised 12 November 2010
Accepted 15 November 2010
Available online 19 November 2010
Keywords:
Vasopressin
Anxiety
Depression
Arginine vasopressin (AVP) is a neuropeptide produced in the
hypothalamus. It is involved in many biological processes of both
the circulatory system and the central nervous system (CNS),
including water homeostasis, renal function, mediation of cardio-
vascular function, non-opioid mediation of tolerance for pain,
regulation of temperature in mammals and modulation of cortico-
tropin release from the pituitary.1–3 AVP acts on three vasopressin
receptor subtypes, designated V1a, V2 and V1b (V3) and has also
been shown to agonize the closely related oxytocin (OT) receptor.4
The V1b receptor is expressed mainly in the anterior pituitary and
treatment with a Vlb receptor antagonist can inhibit stress-
induced corticotropin releasing hormone (CRH) stimulated adreno-
corticotropic hormone (ACTH) secretion.5 Behavioral studies of
V1b knockout mice, which are healthy and viable, have shown that
the V1b receptor is implicated in anxiety, depression and aggres-
sion.6,7 Therefore selective antagonism of the Vlb receptor may
have significant clinical potential as a treatment for certain types
of depression and stress-related affective disorders. It is believed
that Vlb antagonists that penetrate the CNS may have greater ther-
apeutic potential for stress-related affective disorders. Some V1b
receptor antagonists have been described in the literature, among
them those by Sanofi 18 and Organon/Schering-Plough 29,10 (Fig. 1).
The reported V1b antagonists tend to have high molecular weight
and high topological surface area (tPSA) which are molecular prop-
erties usually associated with poor CNS penetration.11 They also
generally show poor selectivity over the V1a and OT receptors.12
Herein we report the development of a novel, selective lead series
of V1b antagonists with improved physicochemical properties.
Compound 3 was identified as a V1b antagonist from a high
throughput screen performed on the GSK compound collection
(Fig. 1).13 The compound was found to be >10-fold selective over
the OT and V1a receptors and had lower molecular weight and
tPSA than previously identified V1b antagonists. Compound 3
was therefore an attractive starting point for hit to lead optimiza-
tion. The SAR was investigated with the aim of increasing the V1b
in vitro potency and improving physicochemical properties, espe-
cially the high calculated log D at pH 7.4 (c log D7.4)
14 whilst main-
taining the selectivity profile over the V1a and OT receptors.
Many of the compounds were synthesized from commercially
available ethyl 2-bromothiazole-4-carboxylate, 4, using the route
shown (Scheme 1). The key intermediate 5 could be subsequently
reacted in Suzuki cross-coupling reactions or in Buchwald
amidations.15 The sulfonylation, Suzuki and Buchwald couplings
and the two reductive alkylation steps were all amenable to paral-
lel synthesis, which allowed for rapid SAR exploration.
The SAR of the thiazole-2-position was explored. Substituents in
the ortho and meta positions were found to be generally less potent
than para substituents (data not shown) so the para position was
explored in more detail. A series of simple alkyl substitutions, butyl-
amine analogs and benzylic aromatics were prepared (Table 1,
compounds 6–12) as alternatives to the terminal butylamine. An in-
crease in target potency was observed with 6, V1b pKi = 7.6. It was
shown by preparation of 7, V1b pKi = 7.0 that the activity could also
be retained by formation of the tertiary amine. Compounds 10–12
demonstrated that the potency was not simply driven by
⇑
Corresponding author.
E-mail address: jennifer.a.borthwick@gsk.com (J.A. Borthwick).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.11.061

C. A. Smethurst et al. / Bioorg. Med. Chem. Lett. 21 (2011) 92–96
93
OH
S
OMe
O
S
H
N
N
N
N
O
H
N
N
O
Cl
N
O
O
O
N
N
S
O
SO2
OMe
SO2
MeO
OMe
MeO
3
1 SSR149415
2
V1b pKi=6.8, V1a pIC50=5.6, OT pIC50=5.3,
Molecular weight = 513, tPSA = 71.5 Å2,
cLogD7.4= 4.3
V1b pKi = 9.34, V1a pKi = 7.23, OT pKi 8.82
Molecular weight = 630 tPSA = 125.9 Å2
clogD7.4 = -0.2.
V1b pKi =7.36, V1a pKi = 5.25,OT pKi=6.50
Molecular weight = 619, tPSA = 105.3 Å2,
clogD7.4 = 3.0
Figure 1. Selected literature V1b antagonists and HTS hit 3.
S
H
S
S
O
O
a
S
b
c
O
N
N
N
Br
Br
N
NH
R1
Br
S
Ar
N
Br
H
O
R1
O
4
5
R2
O
S
S
N
O
S
d
O
S
e
R3
N
+
N
Ar
N
Br
N
Ar
R1
R1
O
B(OH)2
O
B(OH)2
5
N
R3
R2
S
S
O
O
O
O
S
f
+
N
N
N
S
Ar
Br
R4
R4
N
H
N
NH2
Ar
R1
R1
O
O
5
Scheme 1. Representative synthesis of compounds. Reagents and conditions: (a) DIBAL-H, THF, CH2Cl2, À78 °C; (b) NH2R1, NaBH(OAc)3, CH2Cl2; (c) Ar-SO2Cl, solid supported-
1,5,7-triazabicyclo[4.4.0]dec-5-ene, DIPEA, CH2Cl2, DMA, microwave 60 °C, 20 min; (d) NHR2R3, NaBH4, MeOH; (e) Pd(PPh3)2Cl2, LiCl, Na2CO3, 1:1 toluene, ethanol,
microwave, 130 °C, 30 min; (f) Pd(OAc)2, Xantphos, Cs2CO3, THF, microwave, 130 °C, 15 min.
Table 1
Variation of thiazole-2-substituent
S
O
N
*
N
O
S
O
R
V1b pKi
V1a pKi
OT pKi
Molecular weight
513
tPSA (Å2)
71.5
c log D7.4
N
H
3
6
7
6.8
7.6
7.0
5.6
5.3
4.3
*
N
H
6.0
5.7
5.6
5.4
513
513
71.5
62.7
4.2
3.9
*
N
*
N
H
8
9
7.0
7.0
5.1
5.7
4.6
5.2
527
499
71.5
71.5
4.7
3.9
*
*
N
H
(continued on next page)

94
C. A. Smethurst et al. / Bioorg. Med. Chem. Lett. 21 (2011) 92–96
Table 1 (continued)
R
V1b pKi
V1a pKi
OT pKi
Molecular weight
553
tPSA (Å2)
71.5
c log D7.4
N
H
10
6.5
5.5
5.2
5.2
*
N
H
11
6.4
5.6
5.2
4.4
4.7
537
577
84.7
80.8
5.0
5.0
O
*
N
H
12
<4.3
MeO
*
O
H2N
Me
13
14
15
6.3
6.2
6.1
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
471
499
472
102.6
79.8
96.8
4.1
4.3
1.9
*
O
N
Me
*
O
HO
*
O
*
N
16
17
7.0
6.8
n.d.
n.d.
n.d.
n.d.
473
473
101.5
101.5
3.4
2.9
H
N
O
*
N
H
N
O
*
N
18
19
6.4
6.5
n.d.
n.d.
n.d.
n.d.
502
438
97.8
88.6
5.5
4.4
H
MeO
O
*
N
H
Effect on V1b in vitro potency, n.d. = not determined.
Table 2
Variation of thiazole core
Table 2 (continued)
Core
V1b pKi
Molecular
weight
tPSA (Å2)
71.5
c log D7.4
O
R
CH3
21
5.9
521
5.2
het
Ar
Ar
N
X
N
S
N
O
O
22
23
24
6.1
5.5
6.0
476
524
524
41.5
58.6
58.6
3.1
5.1
5.2
H
N
S
X = C, N, S
R = H, Me
N
F
Core
V1b pKi
Molecular
weight
tPSA (Å2)
71.5
c log D7.4
Ar
N
S
N
Ar
F
3
6.8
513
507
4.3
4.7
N
N
Ar
N
20
6.4
71.5
Ar
Effect on V1b in vitro potency.
N

C. A. Smethurst et al. / Bioorg. Med. Chem. Lett. 21 (2011) 92–96
95
Table 3
Alteration of sulfonamide
4-substitution. It appears that there is a tight size constraint at
the para position as replacement of the methoxy group in 3 by tri-
fluoromethoxy 29 or isopropoxy 28 groups resulted in decreased
potency, whilst replacement by fluoro or H in compounds 30 and
31 resulted in little or no decrease. The drop-off in potency
observed with the disubstituted compound 27 also suggested that
there is a strong steric component to binding in this region. No
increase in potency from 3 was observed although it was encour-
aging that in vitro potency could be maintained in 32 with a
simultaneous reduction in c log D7.4 (3.7 compared to 4.3 for 3).
Compound 3 was studied in a rat DMPK experiment (Table 4)16
and was found to have good brain penetration which was consis-
tent with predictions based on its relatively low tPSA value. How-
ever 3 was also found to have low oral bioavailability, which was
attributed to its high clearance (>liver blood flow). A metabolite
ID study of 3 (data not shown) was therefore conducted and iden-
tified that the cyclopentyl moiety was a key site of metabolism for
this compound. We postulated that the introduction of heteroat-
oms into this ring would stabilise the ring to oxidative metabolism.
Several analogs were synthesized and a number of them retained
V1b potency. These analogs were progressed to DMPK experiments
where, gratifyingly, a reduction in in vitro and in vivo clearance
was observed. However, the compounds did not show an improve-
ment in measured oral bioavailability from 3. Although the high
clearance had been addressed the low oral bioavailability suggests
that the compounds also have poor absorption which would re-
quire further optimization. By introducing additional heteroatoms
to reduce clearance, the tPSA of these compounds was increased
and a trend for decreasing CNS penetration with increasing tPSA
was observed.
S
O
S
N
N
*
N
O
R
V1b pKi
Molecular weight
tPSA (Å2)
c log D7.4
3
25
26
27
28
29
30
31
32
4-MeOPh
3-MeOPh
2-MeOPh
3,4-MeOPh
4-iPrOph
4-OCF3Ph
4-FPh
6.8
6.3
5.9
5.9
5.9
5.7
6.4
6.2
6.6
513
513
513
543
541
567
501
483
540
71.5
71.5
71.5
80.7
71.5
71.5
62.3
62.3
91.4
4.3
4.8
4.2
4.2
5.2
5.0
4.0
4.4
3.7
Ph
3-NHAcPh
Effect on V1b in vitro potency.
lipophilicity as increasing c log D7.4 lowered the V1b potency. The
selectivity of all of these compounds for V1b over V1a and OT was
maintained at 10- to 100-fold.
These results encouraged us to make further changes to the left-
hand portion of the molecule, replacing the butylamine with non-
basic substituents (compounds 13–19). The compounds prepared
maintained V1b antagonist potency and demonstrated that the
c log D7.4 could be reduced substantially, as demonstrated in the
acidic compound 15, V1b pKi = 6.1, c log D7.4 = 1.9. The molecular
weight could also be reduced substantially whilst maintaining
V1b potency as demonstrated by compounds such as 19, V1b
pKi = 6.5, molecular weight = 438.
Replacement of the thiazole ring in 3 with alternative aryl rings
was investigated using analogous chemistry to that described in
Scheme 1. These replacements generally had similar or slightly
lower potency than 3 (Table 2) and were not pursued further at
this stage.
In summary, a novel series of selective V1b antagonists has
been identified with lower tPSA and molecular weight than previ-
ously identified V1b antagonists. The hit to lead effort demon-
strated that a number of properties of the HTS hit 3 could be
improved. An improvement to in vitro potency at the V1b receptor
was achieved with 6. The physicochemical properties of the hit
were improved, for example, c log D7.4 was substantially reduced
in 15 and molecular weight was reduced in 19. The DMPK profile
was partially addressed with whole blood clearance substantially
reduced for 33, 34 and 35. These molecules represent useful tools
for in vivo experimentation. However, further optimization is now
Substitution of the aryl sulfonamide ring was investigated
(Table 3). Comparison of compounds 25 and 26 with 3 demon-
strates that there is a slight preference in the binding site for
Table 4
Alteration of sulfonamide N-substituent
S
O
N
N
O
S
H
N
*
O
R
V1b pKi
Molecular weight
tPSA
(Å2)
c log D7.4
Rat Clint
(ml/min/g)
Clb
Vdss
(L/kg)
T1/2
(h)
F%
Ratio
[Brain]:[Blood]
(ml/min/kg)
*
3
6.8
6.2
513
515
71.5
80.7
4.3
2.7
17.3
6.8
129
4.5
4.0
0.8
1.1
<1%
<1%
6.3:1
0.9:1
*
33
62
48
O
O
O
*
34
35
6.9
5.9
529
563
80.7
3.1
1.9
3.5
1.4
1.5
2.0
0.6
1.4
<1%
<1%
1.1:1
0.4:1
*
105.6
33
S
O
Effect on V1b in vitro potency and in vivo DMPK. Rat Clint is intrinsic clearance measured in rat microsomes; Clb refers to whole blood clearance; Vdss is volume of distribution
at steady state; T1/2 is the half life of the test compound; F% is the oral bioavailability expressed as a percentage; ratio [Brain]:[Blood] is the ratio of test compound
concentration in the brain and blood.

96
C. A. Smethurst et al. / Bioorg. Med. Chem. Lett. 21 (2011) 92–96
11. Van de Waterbeemd, H.; Camenisch, G.; Folkers, G.; Chretien, J. R.; Raevsky, O.
A. J. Drug Target. 1998, 6, 151.
12. Griffante, C.; Green, A.; Curcuruto, O.; Haslam, C. P.; Dickinson, B. A.; Arban, R.
Br. J. Pharmacol. 2005, 146, 744.
required to identify molecules with lower clearance combined
with improved physicochemical properties which could result in
improved oral bioavailability.
13. All V1b compound screening and profiling activities were conducted using a
Chinese Hamster Ovary (CHO) host stably expressing a recombinant V1b
receptor cell. A 384 well plate based FLIPR Ca2+ assay (Fluorescence Imaging
References and notes
Plate Reader) was also configured to measure compound inhibition of
a
1. Ali, F.; Guglin, M.; Vaitkevicius, P.; Ghali, J. K. Drugs 2007, 67, 847.
2. Thibonnier, M. Expert Opin. Investig. Drugs 1998, 7, 729.
3. Steckler, T.; Holsboer, F.; Reul, J. M. Baillieres Best Pract. Res. Clin. Endocrinol.
Metab. 1999, 13, 597.
4. Thibonnier, M.; Berti-Mattera, L. N.; Dulin, N.; Conarty, D. M.; Mattera, R. Prog.
Brain Res. 1998, 119, 147.
5. 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.
6. Tanoue, A.; Honda, K.; Oshikawa, S.; Kitigawa, Y.; Koshimizu, T.; Mori, T. J. Clin.
Invest. 2004, 113, 302.
7. Wersinger, S. R.; Ginns, E. I.; O’Carroll, A.-M.; Lolait, S. J.; Young, W. S. Mol.
Psychiatry 2002, 7, 975.
vasopressin mediated response. 10 K cells per well were plated overnight in
culture media before being washed in assay buffer and then loaded with
calcium fluorophore, Fluo4-AM for 1 h. Test compounds were equilibrated with
cells at 1% DMSO for 10 min before addition of EC80 vasopressin. For compound
screening, compounds were tested at 10 lM and hits were identified as those
compounds capable of giving >50% inhibition of the vasopressin response. For
further characterization of hits, concentration response curves were
constructed versus EC80 of vasopressin, and the concentration at which 50%
of the maximal inhibition is achieved was calculated (IC50). All quoted data is
for a minimum of n = 3 determinations of IC50. pKi values are calculated from
the IC50 values by using the Cheng and Prusoff equation.
14. Calculated using log D suite version 11. Advanced Chemistry Development,
Inc., Toronto, Ontario, Canada (http://www.acdlabs.com/).
15. Yin, J.; Buchwald, S. L. Org. Lett. 2000, 2, 1101.
16. Pharmacokinetics measured in male Sprague–Dawley rats following
intravenous (iv) and oral (po) administration. Dose was 1 mg/kg iv and 5 mg/
kg po for compound 3. Dose was 1 mg/kg iv and 1 mg/kg po for all other
compounds. These studies used at least three animals for each (iv/po) leg.
8. Serradeil-Le Gal, C.; Wagnon, J.; Tonnerre, B.; Roux, R.; Garcia, G.; Griebel, G.;
Aulombard, A. CNS Drug Rev. 2005, 11, 53.
9. 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.
10. Scott, J. D.; Miller, M. W.; Li, S. W.; Lin, S.-I.; Vaccaro, H. A.; Hong, L.; Mullins, D.
E.; Guzzi, M.; Weinstein, J.; Hodgson, R. A.; Varty, G. B.; Stamford, A. W.; Chan,
T.-Y.; McKittrick, B. A.; Greenlee, W. J.; Priestley, T.; Parker, E. M. Bioorg. Med.
Chem. Lett. 2009, 19, 6018.