Non-peptide oxytocin agonists

Article abstract of DOI:10.1016/j.bmcl.2004.04.107

The first non-peptide, low molecular weight agonists of the hormone oxytocin (OT) are reported. The most potent compound, 39, showed an EC ₅₀ = 33 nM and was selective for the OT receptor. A library of compounds targeted to the vasopressin/oxytocin family of receptors was screened for activity at a cloned human oxytocin receptor using a reporter gene assay. Potency and selectivity were optimised to afford compound 39, EC₅₀ = 33 nM. This series of compounds represents the first disclosed, non-peptide, low molecular weight agonists of the hormone oxytocin (OT).

Full text of DOI:10.1016/j.bmcl.2004.04.107

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4585–4589
Non-peptide oxytocin agonists
Gary R. W. Pitt,a Andrzej R. Batt,a Robert M. Haigh,a Andrew M. Penson,a
a
Peter A. Robson, David P. Rooker, Andre L. Tartar, Julie E. Trim,
a
b
a
´
Christopher M. Yeaa and Michael B. Roea,
*
aFerring Research Ltd, Chilworth Science Park, Southampton SO16 7NP, UK
Faculte des sciences pharmaceutiques et biologiques, 3 rue du Pr. Laguesse, F-59006 Lille Cedex, France
b
´
Received 12 November 2003; revised 5 April 2004; accepted 8 April 2004
Abstract—A library of compounds targeted to the vasopressin/oxytocin family of receptors was screened for activity at a cloned
human oxytocin receptor using a reporter gene assay. Potency and selectivity were optimised to afford compound 39, EC50 =33nM.
This series of compounds represents the first disclosed, non-peptide, low molecular weight agonists of the hormone oxytocin (OT).
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
ical tools and as potential drugs. Furthermore, the pep-
tide vasopressin and oxytocin analogues such as
carbetocin (Duratocin) are poorly absorbed due to their
high hydrophilicities.6 Thus there is a need to develop
non-peptide compounds that may have potential to be
orally bioavailable. These may have utility as drugs
where OTfunction is compromised, in particular in
the treatment of various sexual disorders including
male erectile dysfunction. Also in promoting labour,
controlling post-partum bleeding and increasing milk
let-down.
Oxytocin (OT) is a cyclic nonapeptide that acts at the
OTreceptor, a member of the super-family of G-protein
coupled receptors (GPCRs). OTand the closely related
hormone arginine vasopressin (AVP) are neurohypo-
physeal hormones secreted by the posterior pituitary
gland.1 AVP acts at the vasopressin receptor of which
three subtypes, V1a, V1b and V2 have so far been identi-
fied. The OT receptor is approximately 10-fold selective
for OTover AVP whereas the vasopressin receptor sub-
types show stronger selectivity, up to 400-fold, in favour
of AVP.2,3 Within the periphery the OTreceptor is
localised in a number of different organs including the
uterus and mammary glands. In the uterus it is involved
in the onset and progress of labour. As such it has been
shown that OTantagonists have utility in the treatment
of pre-term labour and the peptide OTantagonist atosi-
ban (Tractocile) is marketed as a treatment for this indi-
cation.4 The OT receptor has also been localised in
central tissues such as the paraventricular nucleus where
it is involved in regulation of both male and female sex-
ual response.1,5 It is clear that OThas a range of phys-
iological roles that have not been fully elaborated. Thus
there is a need for new, potent, selective and efficacious
OTreceptor agonists that may be used as pharmacolog-
A number of non-peptide antagonists of OThave been
discovered.7–9 However, it is generally perceived to be
more difficult to identify non-peptide agonists, princi-
pally because of the much reduced molecular size com-
pared to the parent peptide hormone.10,11 However,
O
N
H
N
N
N
O
O
Cl
N
Cl
O
H
N
H
N
N
O
N
N
O
NMe2
Keywords: Oxytocin agonist; Non-peptide; Male erectile dysfunction.
*
1
2, WAY-VNA-932
3, OPC-51803
Corresponding author. Tel.: +44-2380-763400; fax: +44-2380-
766253; e-mail: michael.roe@ferring-research.co.uk
Figure 1. Known non-peptide V2 agonists.12–14
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.04.107

4586
G. R. W. Pitt et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4585–4589
S
Top group
CO2Me
CO2H
OH
CO2Me
O
N
N
a
c
b
O
Linker
O
R
+
+
HN
Cl
Cl
R
Cl
N
H
H
Cl
N
N
Br
BocNH
NH2
12
13
14
O
Bottom group
H2N
15
S
OMe
Scheme 1.
S
e
d
HN
N
N
16
OMe
OMe
O
non-peptide agonists of the V2 receptor have already
been reported by us (compound 1), as well as by
Wyeth-Ayerst (WAY-VNA-932, 2) and Otsuka
(OPC-51803, 3), the structures of which are shown in
Figure 1.12–14 Given the close relationship between OT
and AVP, we hypothesised that a library designed
around V2 agonists would yield OTagonists. To test this
hypothesis these V2 agonists were assayed for their OT
activity in a functional reporter gene assay, vide infra.
Compounds 2 and 3 showed <10% response at 3lM.
However, 1 gave a response relative to OTof 33% at
3lM. Therefore, a library of 50,000 structures related
to 1 was screened and the resulting OTagonists are de-
scribed in this paper.
Cl
O
O
N
OMe
Cl
H
N
H
N
N
O
O
O
N
O
OMe
N
31
17
Scheme 4. Reagents and conditions: (a) NBS, AIBN, CCl4, reflux,
84%; (b) (i) NH3, EtOH, (ii) (Boc)2O, NaOH, dioxan, H2O, 67%; (c) (i)
6, WSCÆHCl, Et2NiPr, DMAP, CH2Cl2, (ii) 4N HCl/dioxan, 73%; (d)
CDI, Et2NiPr, DMF, 98%; (e) (i) LiOH, H2O, THF, (ii) 1-methyl-
homopiperazine, WSCÆHCl, HOBt, Et3N, CH2Cl2, 76%.
prepared via a DIBAL mediated Beckmann rearrange-
ment from commercially available 4-keto-4,5,6,7-tetra-
hydrothianaphthene 4.15
2. Chemistry
The compounds related to 1 were prepared through the
synthesis of ꢀbuilding blocksꢁ, which were coupled to-
gether at the benzoyl-amide and at the urea as shown
schematically in Scheme 1.
The cyclisation of 9 that led to the synthesis of the fused
azepine tricycle 11 could be carried out in refluxing
2-propanol in the presence of acetic acid over 14 days
in a yield of 80%. More conveniently, the cyclisation
was carried out in basic conditions to achieve a yield
of 40% in just a few hours.16
The synthesis of the top groups is shown in Schemes 2
and 3. The building block 5,6,7,8-tetrahydro-4H-thie-
no[3,2-b] azepine, 6, present in compounds 28–38 was
The synthesis of the chloro substituted linker 14 is
shown in Scheme 4 and utilised a Boc protecting group
for the benzylic amine. It was coupled to the top group
by activation of the carboxylic acid using water soluble
carbodiimide hydrochloride (WSCÆHCl) and deprotect-
ed to give 15. This was elaborated by coupling to meth-
oxy-substituted proline 16 (prepared using literature
procedures) with carbonyl diimidazole (CDI) to give
the urea 17.17 The ester was hydrolysed and amide cou-
plings were carried out using WSCÆHCl to provide the
amides in Table 1, as exemplified with 1-methylhomo-
piperazine, which provided 31.
S
S
S
a
b
N
H
6
N
O
HO
4
5
Scheme 2. Reagents and conditions: (a) HONH2, NaOH, EtOH, H2O,
80%; (b) DIBAL, CH2Cl2, 78%.
H
N
N
N
N
NC
a
b, c
H2N
N
H
N
N
EtO2C
OEt
N
N
N
NH2
OEt
H
N
O
OH
O
OEt
O
Br
7
8
9
d
a
b
c
N
H
N
H
N
N
N
O
N
e
N
N
O
CN
CN
19
CN
N
H
10
N
H
11
O
H2N
18
20
21
Scheme 3. Reagents and conditions: (a) MeNHNH2, EtOH, 85%; (b)
2-fluoronitrobenzene, NaH, THF, 62%; (c) H2, Pd/C, MeOH, 92%; (d)
NaCH2SOMe, DMSO, 75°C, 2.5h, 40%; (e) LiAlH4, THF, 98%.
Scheme 5. Reagents and conditions: (a) (i) n-BuLi, THF, À78°C, (ii)
solid CO2, À78°C, 73%; (b) (i) SOCl2, toluene, (ii) 11, Et3N, CH2Cl2,
87%; (c) NaBH4, CoCl2Æ6H2O, MeOH, 64%.

G. R. W. Pitt et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4585–4589
4587
Table 1. Activities at the OTand V receptors of compounds containing a basic amine
2
S
N
O
Cl
O
H
N
N
O
R
X
Compounds
R
X
OTEC (nM)
50
V2 EC50 (nM)
N
29
34
O
S
1500 1300
470 330
570 0.11
200 79
N
N
30
35
O
S
2300 1400
480 150
1100 270
100 35a
N
31
36
O
S
2800 2100
430 220
1300 210
230 5a
N
N
N
32
O
O
2200 700
>10,000
1200 330a
1800 450a
N
N
33
N
a Reported as mean range.
The synthesis of the methyl substituted linker 19 is
shown in Scheme 5 and in this case the amine was
masked as a nitrile group. Carboxylation was carried
out at À78°C, which was followed by an amide coupling
via an acid chloride to the top group 11. Unmasking of
the nitrile group was carried out using sodium borohy-
dride activated by cobalt(II) chloride to give 21.18
perizine 23a and the resulting amide 24 was treated with
Lawessonꢁs reagent. Removal of the Boc-group, N-
methylation and removal of the Cbz-group afforded
the thioamide 26. Coupling to the amine 21 with CDI
gave the urea 39.19
3. Biology
The thioamides shown in Tables 1 and 2 were prepared
as exemplified for compound 39, Scheme 6. Cbz pro-
tected proline 22 was coupled to Boc-protected homopi-
The human OT and V2 receptors were cloned from
ovary total RNA and kidney total RNA (BD Clontech)
using the reverse transcriptase polymerase chain reac-
tion (RT-PCR). The receptors were subcloned into the
KpnI and EcoRI sites of pcDNA3. The cAMP response
element (CRE) and the nuclear factor of activated
T-cells (NFAT) element was subcloned into the BamHI
and BglII sites of pGL3 promoter (Promega), which
contains the Firefly Luciferase gene. Plasmids were
amplified in Novablue E. coli (Merck Biosciences Ltd)
and plasmid DNA was purified using UltraMobiuse
Endotoxin free plasmid purification kits (Merck Bio-
sciences Ltd).
a
b
CbzN
O
CbzN
S
CbzN
O
NBOC
NBOC
N
N
HN
NBOC
OH
23a
22
24
25
N
N
H
N
c
N
HN
S
d
O
The cell lines CHO (Chinese hamster ovary) and 293
(human embryo kidney) were maintained in Dulbeccoꢁs
Modified Eagles Medium (DMEM) without phenol red,
supplemented with penicillin and 10% foetal calf serum.
H
N
N
N
N
O
S
N
N
39
26
Cells were dual transfected with the receptor and repor-
ter constructs using Genejuice (Merck Biosciences Ltd)
and incubated at 37°C with 5% CO2 for 24h before plat-
ing into 96 or 384 well white plates. CHO cells were
Scheme 6. (a) HOBt, WSCÆHCl, Et3N, CH2Cl2, 100%; (b) Lawessonꢁs
reagent, toluene, 85°C, 67%; (c) (i) 4N HCl/dioxan, (ii) HCHO,
NaCNBH3, MeOH, AcOH, (iii) HBr, AcOH, 88%; (d) 21, CDI,
Et2NiPr, DMF, 63%.

4588
G. R. W. Pitt et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4585–4589
Table 2. Activities at the OTand V receptors of analogues of the thioamide 36
2
R
OTEC (nM)
50
V2 EC50 (nM)
S
N
O
Cl
O
230 5a
H
36
430 220
N
N
O
N
N
S
S
S
N
N
Cl
O
O
H
N
37
38
810 270
1800 330a
N
O
O
N
N
N
N
S
640 340
890 280a
H
N
N
S
N
H
N
N
N
39
33 15
850 280
O
H
N
N
O
N
N
S
a Reported as mean range.
H
N
transfected with OTreceptor and the NFAT-luciferase
constructs, 293 with the V2 receptor and CRE-luciferase
constructs. The plates were incubated for a further 24h
at 37°C with 5% CO2 before testing for functional agon-
ism.
N
S
N
N
O
N
Cl
O
H
N
O
F
H
N
H
N
N
O
N
Compounds were prepared in DMSO to a concentration
of 10mM and dilutions thereof to give the appropriate
concentrations when added to the cells. Final DMSO
concentrations were 0.1%. Levels of luciferase synthesis
were measured following 5 hours incubation at 37°C
using luclite kits (Perkin Elmer). The data were fitted
to a 4-parameter logistic equation and EC50 values
determined. In most cases data are reported as
mean standard deviation (n P 3). Where n=2 the data
are reported as mean range of data. OTreturned a
value of EC50 =0.18 0.069nM (n=20) and AVP of
EC50 =0.14 0.062nM (n=56).
O
O
F
27
28
Figure 2. Structures of the initial screening hits.
relative responses compared to OTof these hit
compounds, determined at a concentration of 3lM,
were 41% for 27 and 80% for 28. Thus 28 was selected
for hit-to-lead investigations. The bottom group was
selected for analoging because the amide could be pre-
pared rapidly. More than one hundred diverse amides
were prepared and screened (not shown) to provide ini-
tial SAR. The majority of these compounds were active
at the V2 receptor but gave only a poor response at the
OTreceptor. Some compounds were approximately
equipotent to both receptors and these appeared to con-
tain a basic amine in the bottom group (Table 1, com-
pounds 29–33). It was found that potency at the OT
receptor could be improved by preparing the thioamide
analogues, compounds 34–36. However, potency at the
V2 receptor was also improved. Selectivity for the V2 re-
ceptor ranged from 5-fold for compound 35 to 2-fold for
4. Results and discussion
Screening of our own in-house assembled vasopressin
targeted library for OTactivity identified two hits, 27
and 28 (Fig. 2), which were related to 1 (Fig. 1). Com-
mon to all three compounds was the substituted benzoyl
linker and a bicyclic or tricyclic fused azepine as a top
group. A urea functionality attached to an aryl ring or
an amino acid amide formed the bottom group. The

G. R. W. Pitt et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4585–4589
4589
against the V2 receptor. It is currently under further
investigation as a pharmacological tool in animal mod-
els of OTactivity, to be reported in subsequent papers,
and as a potential drug for diseases where oxytocin
activity is compromised.
References and notes
1. Gimpl, G.; Fahrenholz, F. Physiol. Rev. 2001, 81, 629.
2. Postina, R.; Kojro, E.; Fahrenholz, F. J. Biol. Chem. 1996,
217, 31593.
Figure 3. Dose response curves in the human OTreporter gene assay
for OT( n) and compounds 36 (h), 37 (j), 38 (s) and 39 (d). Percent
response is quoted relative to the OTresponse. Each data point is
mean SD (n=3).
3. Serradeil-LeGal, C.; Wagnon, J.; Valette, G.; Garcia, G.;
Pascal, M.; Maffrand, J. P.; Le Fyr, G. In Progress in
Brain Research; Poulain, D., Oliet, S., Theodosis, D., Eds.;
Elsevier: Amsterdam, 2002; Vol. 139, pp 197–210.
4. Coomarasamy, A.; Knox, E. M.; Gee, H.; Khan, K. S.
Med. Sci. Monit. 2002, 8, RA268.
5. Andersson, K.-E. Phamacol. Rev. 2001, 53, 417.
6. Brondsted, H.; Nielsen, H. M.; Hovgaard, L. Int. J.
Pharm. 1995, 114, 151.
compound 36 and 36 was selected for further investiga-
tion as shown in Table 2.
Compound 37 (Table 2), where the stereogenic methoxy
group on proline had been removed, showed a reduction
in potency compared to 36. However it was 2-fold selec-
tive for the OTreceptor and this was the first compound
to show OTselectivity. Replacement of the chloro sub-
stituent by methyl was investigated to give 38,
EC50 =640nM but this showed no improvement. Final-
ly, attention turned towards the top end group. Re-
examination of the screening data led to the hypothesis
that the fused azepine tricycle present in 27 may be
responsible for some of the efficacy of this compound
in the OTassay. Therefore it was incorporated into
the new thioamide series to give compound 39. This
compound showed a significant increase in potency at
the human OTreceptor to give an EC 50 =33nM. Fur-
thermore, this compound maximally stimulated the
OTreceptor to the same degree as OT( Fig. 3). There
was no such increase in potency at the V2 receptor thus
selectivity improved significantly to >25-fold. No agon-
ist activity was detected in reporter gene assays utilising
human V1a or V1b receptors.
7. Williams, P. D.; Bock, M. G.; Evans, B. E.; Freidinger, R.
M.; Gallichio, S. N.; Guidotti, M. T.; Jacobson, M. A.;
Kuo, M. S.; Levy, M. R.; Lis, E. V.; Michelson, S. R.;
Pawluczyk, J. M.; Perlow, D. S.; Pettibone, D. J.; Quigley,
A. G.; Reiss, D. R.; Salvatore, C.; Stauffer, K. J.; Woyden,
C. J. Bioorg. Med. Chem. Lett. 1999, 9, 1311.
8. Wyatt, P. G.; Allen, M. J.; Chilcott, J.; Foster, A.;
Livermore, D. G.; Mordaunt, J. E.; Scicinski, J.; Wool-
lard, P. M. Bioorg. Med. Chem. Lett. 2002, 12, 1399.
9. Wyatt, P. G.; Allen, M. J.; Chilcott, J.; Hicken, G.; Miller,
N. D.; Woollard, P. M. Bioorg. Med. Chem. Lett. 2001, 11,
1301.
10. Patchett, A. A.; Nargund, R. P. In Annual Reports in
Medicinal Chemistry; Doherty, A. M., Ed.; Academic: San
Diego, 2000; Vol. 35, pp 289–298.
11. Beeley, N. R. A. Drug Discovery Today 2000, 5, 354.
12. Pitt, G. R.; Abstracts of Papers, RSC-SCI Symposium on
Medicinal Chemistry, September 9–12, 2001.
13. Caggiano, T. J. Drugs Fut. 2002, 27, 248.
14. Kondo, K.; Kan, K.; Tanada, Y.; Bando, M.; Shinohara,
T.; Kurimura, M.; Ogawa, H.; Nakamura, S.; Hirano, T.;
Yamamura, Y.; Kido, M.; Mori, T.; Tominaga, M.
J. Med. Chem. 2002, 45, 3805.
15. Cho, H.; Murakami, K.; Nakanishi, H.; Isoshima, H.;
Hayakawa, K.; Uchida, I. Heterocycles 1998, 5, 919.
16. Chakrabarti, J. K.; Hotten, T. M.; Puller, I. A.; Tye, N. C.
J. Med. Chem. 1989, 32, 2573.
17. Smith, E. M.; Swiss, G. F.; Neustadt, B. R.; Gold, E. H.;
Sommer, J. A.; Brown, A. D.; Chiu, P. J. S.; Moran, R.;
Sybertz, E. J.; Baum, T. J. Med. Chem. 1988, 31, 875.
18. Satoh, T.; Suzuki, S.; Suzuki, Y.; Miyaji, Y.; Imai, Z.
Tetrahedron Lett. 1969, 10, 4555.
19. Analytical data for 39: NMR (270MHz, CDCl3) d 1.78–
2.84 (13H, m), 3.20–4.46 (14H, m), 4.50–4.68 (1H, m),
4.98–5.16 (1H, m), 5.88 (1H, d, J=14.6Hz), 6.58–6.70
(2H, m), 6.72–7.22 (5H, m), 7.22 (1H, s) ppm. MS m/z
601.3 [M+H]+.
The compounds herein are the first reported non-pepti-
dic, small molecule agonists of the hormone oxytocin.
The successful screening strategy demonstrates the effec-
tiveness of a functional assay for screening for agonists.
The [6,7]-fused ring template, exemplified by 2,3,4,5-
tetrahydro-1H-benzo[b]azepine in compound 1 and by
5,6,7,8-tetrahydro-4H-thieno[3,2-b] azepine 6 and the
fused tricycle 11, appears to be important for activity
at the AVP and OTreceptors. Thus it may be considered
a ꢀprivileged structureꢁ.10 It remains to be seen whether
this structure is privileged for other GPCRs. Compound
39 is potent against the OTreceptor and selective