Identification of a potent and selective 5-HT1B receptor antagonist

Article abstract of DOI:10.1016/j.bmcl.2005.07.085

An SAR study around the mixed 5-HT_1ABD receptor antagonist SB-272183 found that introduction of cis-2,6-dimethyl substitution onto the piperazine ring was a key structural change, which imparted a combination of both excellent selectivity over

Full text of DOI:10.1016/j.bmcl.2005.07.085

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4708–4712
Identification of a potent and selective 5-HT1B receptor antagonist
Paul A. Wyman,* Howard R. Marshall, Sean T. Flynn, Ron J. King, Mervyn Thompson,
Paul W. Smith, Michael S. Hadley, Gary W. Price, Claire M. Scott and Lee A. Dawson
Psychiatry CEDD, New Frontiers Science Park, GlaxoSmithKline, Third Avenue, Harlow, Essex CM19 5AW, UK
Received 26 May 2005; revised 19 July 2005; accepted 27 July 2005
Available online 8 September 2005
Abstract—An SAR study around the mixed 5-HT1ABD receptor antagonist SB-272183 found that introduction of cis-2,6-dimethyl
substitution onto the piperazine ring was a key structural change, which imparted a combination of both excellent selectivity over
the 5-HT1A and 5-HT1D receptors and low intrinsic activity. This led to the identification of the selective 5-HT1B receptor antagonist
SB-616234.
Ó 2005 Elsevier Ltd. All rights reserved.
The 5-hydroxytryptamine (5-HT) family of receptors
comprises 14 distinct sub-types, which have been exten-
sively studied and categorized into seven main families
5-HT1–5-HT7 on the basis of their operational, structur-
al, signal transduction pathways and pharmacological
attributes.1,2 The largest family comprises the 5-HT1
receptors, which have been subdivided further into
5-HT1A, 5-HT1B, 5-HT1D, 5-HT1E and 5-HT1F receptors
showing approximately 40–60% homology between
members. Indeed, much interest within central nervous
system (CNS) research has been directed towards this
family since 5-HT1A, 5-HT1B and 5-HT1D receptors
have been identified as autoreceptors and are thus
considered potential targets for the modulation of
central serotonergic function.3,4
the terminal 5-HT autoreceptor and hence result in
an increase in extracellular levels of 5-HT.7–9 Chronic
blockade of the 5-HT uptake carrier by selective
serotonin re-uptake inhibitors (SSRIs) is also known
to increase extracellular 5-HT levels, and this is hypoth-
esized to result in the antidepressant and anxiolytic
activity of this class of drugs in the clinic. It is therefore
postulated that 5-HT1B receptor antagonists, by virtue
of increasing 5-HT levels, will have the same net
effect as SSRIs and hence may offer clinical utility
in treatment of mood disorders.5,9 Consequently, selec-
tive antagonists of the 5-HT1B receptor or mixed
5-HT1B/1D ligands have been studied alone or in combi-
nation with other serotonergic agents for their potential
utility in depression and anxiety. A number of reviews
have covered the potential therapeutic applications of
agonists and antagonists of these receptors.10,11
Within the CNS, 5-HT1B receptors are located both pre-
and postsynaptically. The presynaptic 5-HT1B receptors
are located on nerve terminals and in somatodendritic
regions (cell body or raphe) of the 5-HT neurone.5 As
the 5-HT1B receptor is negatively coupled to adenylate
cyclase, in both cases activation of these presynaptic
5-HT1B receptors serves to inhibit 5-HT release.5,6 The
location of 5-HT1B receptors on 5-HT nerve terminals
has resulted in a number of studies (including those car-
ried out on human brain tissue6), which has led to the
suggestion that blockade of this receptor will have the
effect of relieving the autoinhibitory action of 5-HT at
Here, we describe part of our research in the 5-HT area in
which we were interested in identifying a selective antag-
onist for the 5-HT1B receptor. Past literature concerning
5-HT1B and 5-HT1D receptors is complicated by changes
in nomenclature; 5-HT1D receptors were formerly classi-
fied as 5-HT1Da and 5-HT1B receptors as 5-HT1Db.12 Most
of the initial work in the 5-HT1B area was carried out
using mixed 5-HT1B/1D ligands, such as GR127935,13
which demonstrated silent antagonism in a rat glial
cell line, as determined by the inhibition of forskolin-
stimulated cAMP formation by 5-HT.14 However,
such compounds showed partial agonism in a human re-
combinant functional assay ([35S]GTPcS binding).15
Our earlier studies provided SB-224289 as the first truly
selective 5-HT1B antagonist16 and although this com-
Keywords: 5-HT1B receptor antagonist; SB-616234.
*
Corresponding author. Fax: +44 01279 622790; e-mail: paul.
wyman@gsk.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.07.085

P. A. Wyman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4708–4712
4709
Me
N
Me
N
Me
Me
Me
O
N
N
N
Me
N
O
N
N
O
O
OMe
N
H
O
SB-224289
GR127935
O
Me
N
O
H
Me
N
N
N
N
N
H
N
Cl
O
N
N
Me
AR-A000002
SB-272183
pound proved unsuitable for development it did serve
as a very useful tool for investigating the pharmacology
of this receptor. Recently, AR-A00002 has been reported
as a selective 5-HT1B receptor antagonist, although
this also shows partial agonism in a [35S]GTPcS assay.17
corresponding piperazine analogues (1, 6, and 8),
whereas they showed a consistent reduction in 5-HT1D
affinity together with a marked reduction in affinity at
the 5-HT1A receptor. A highly significant third finding
was the effect on intrinsic agonist activity of introducing
the cis-dimethyl substitution, as evaluated in the
[35S]GTPcS functional assay,19 which was shown to be
significantly reduced at both the 5-HT1B and 5-HT1D
receptors. Thus, compound 2 was a very encouraging
early lead with a good 5-HT1B/5-HT1D selectivity, low
5-HT1B intrinsic activity, and it also showed >100-fold
selectivity over a range of other 5-HT and dopamine
receptors. Although this compound subsequently
proved to have in vitro metabolic liability, it prompted
further SAR investigations around these molecules.
The starting point for our investigation came from SAR
studies and data mining around the previously reported
potent 5-HT1ABD receptor antagonist SB-272183.18
From these investigations, we identified similar struc-
tures in which cis-2,6-dimethyl substitution of the piper-
azine ring afforded improved selectivity for the 5-HT1B
receptor, as shown in Table 1.
Three significant findings emerged from this initial
investigation. First, the biaryl amide LHS could be re-
placed by a substituted phenyl urea or phenylacetamide
and maintain good 5-HT1B affinity. Second, compounds
2, 7 and 9 all showed comparable 5-HT1B affinity to the
As N-demethylation of 2 was indicated as one possible
source of instability a small investigation was conducted
on this compound. Increasing the size of the N-alkyl
Table 1. Receptor binding affinity (pKi)a, intrinsic activity19 and intrinsic clearance for 5-HT1A, 5-HT1B and 5-HT1D
R
O
Z
N
X
N
N
R
OMe
Compound
X
R
H
Z
Binding affinity pKi
(intrinsic activity)
Selectivity
Intrinsic clearance
(liver microsomes)
ml/min/g
5-HT1A 5-HT1B
5-HT1D
9.0
8.7 (0.9) 7.6 (0.9)
5-HT1B:5-HT1A 5-HT1B:5-HT1D Rat
Human
SB-272183
8.4
7.9
6.3
6.0
6.3
8.4
1
6
0.25
12
30
2.8
28
4.7
1
2
3
4
5
Me
—
46
12
45
—
—
34
14
47
—
Me Me
Me
Me Et
8.5 (0.3) 7.0 (inv)b 160
N
H
7.3
8.4
7.5
6.9
6.9
6.8
21
110
2
Me
Me n-Pr 7.1
H
6
7
H Me
Me Me
7.3
5.8
8.4 (0.6) 7.9 (0.7)
8.1 (0.1) 7.1 (0.1)
13
200
3
10
—
—
N
3.0
3.0
Cl
Cl
8
9
H Me
Me Me
6.8
5.7
8.1 7.5
7.8 (0.1) 6.6 (0.2)
20
125
4
15
—
—
—
—
O2N
a Radioligand binding assay from cloned human 5-HT receptors.
b Inverse agonism.

4710
P. A. Wyman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4708–4712
Table 2. Receptor binding affinity (pKi)a and intrinsic activity19 of
substituted benzamides, phenylacetamides, ureas and carbamates for
5-HT1B and 5-HT1D
N-methyl group did reduce the intrinsic clearance, but the
NH compound 3 also showed a reduced 5-HT1B affinity.
Me
Therefore, investigation was focused on the LHS. A ser-
ies of substituted benzamides, phenylacetamides, ureas
and carbamates was prepared and the effect on 5-HT1B
affinity, selectivity and intrinsic activity examined. It
was rapidly found that optimal affinity was obtained
with substitution by a 2- and/or 3-position electron-
withdrawing group. Table 2 shows a representative set
of compounds, which also indicates that very low intrin-
sic activity at the 5-HT1B receptor was maintained.
Me
Me
O
N
X
N
N
R
OMe
Compound
X
R
5-HT1B pKi
(intrinsic activity) pKi
5-HT1D
10
11
12
13
14
15
16
17
18
19
Bond 2-Cl, 3-Cl
Bond 3-Cl
7.9
7.7
6.6
6.6
7.0
6.7
6.9
6.7
6.8
7.3
7.5
7.6
CH2
CH2
CH2
NH
NH
NH
O
2-F, 3-Cl
8.7 (0.1)
This study revealed that the benzamides 10 and 11 were
less active at the 5-HT1B receptor. Phenylacetamides 12–
14 generally had a slightly greater affinity than the corre-
sponding ureas 15–17, whereas the carbamates 18 and 19
had good 5-HT1B affinity but generally these compounds
showed lower selectivity over the 5-HT1D receptor. Where
the corresponding NH piperazine analogues were pre-
pared, these showed reduced 5-HT1B affinity (data not
shown).
2-CF3, 3-F 8.5 (0.1)
2-F, 3-CF3 8.8 (0.1)
2-F, 3-Cl
8.1 (0.1)
2-F, 3-CF3 8.4 (0)
2-Cl, 3-CF3 8.6 (0)
2-F, 3-CF3 8.6 (0.1)
2-Cl, 3-CF3 8.6 (0.15)
O
a Radioligand binding assay from cloned human 5-HT receptors.
Compounds 14 and 16 were profiled further in vivo, but
both of them failed to meet all the requirements for
progression. Although the simple benzamides (e.g., 10)
group to ethyl (4) retained 5-HT1B affinity, but had no
effect on intrinsic clearance. The n-propyl analogue 5
was less potent at the 5-HT1B receptor. Removing the
Table 3. Receptor binding affinity (pKi)a, intrinsic activity19 and human intrinsic clearance (CLi) of biaryl LHSs for 5-HT1B and 5-HT1D
Me
X
O
R
N
N
N
Me
OMe
Compound
X
R
5-HT1B pKi (intrinsic activity)
5-HT1D pKi
CLi human
—
Me
20
Me
H
9.2 (0)
7.8
Me
O
21
8.5 (inv)b
6.4
1.9
Me
Me
N
22
23
24
25
26
H
8.1 (0)
8.3
6.3
7.8
6.6
6.4
6.6
6.6
—
Me
Me
Me
Me
Me
N
N
Me
H
O
O
N
N
8.5 (0)
8.3 (0)
8.3 (inv)b
1.0
0.9
0.6
O
Me
N
H
N
Me
N
N
H
a Radioligand binding assay from cloned human 5-HT receptors.
b Inverse agonism.

P. A. Wyman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4708–4712
4711
that seen originally in the corresponding aniline series.9
Significantly for compounds in this class, the N–H
piperazine analogues had similar 5HT1B affinity to the
N–Me piperazines, yet retained their superior 5-HT1B/
Ac
N
Ac
N
a
Ac
(2)
(1)
receptor selectivity and, therefore, offered examples
b, c
1D
avoiding potential metabolic instability due to N-de-
methylation. Consequently, a number of compounds
were taken up for in vivo evaluation from this series
and compound 24 (SB-616234) demonstrated to be the
best overall profile for further studies.
Ac
N
Ac
N
Br
d, e
OH
OMe
(4)
(3)
The synthesis of this series of indoline piperazines uti-
lised (6) (R = H or Me) (Scheme 1) as a key intermedi-
ate. This was prepared from 1-acetylindoline (1),
which was acetylated under Friedel–Crafts conditions
to give the 5-acetyl analogue (2). Baeyer–Villiger oxida-
tion gave the acetoxy intermediate, which readily hydro-
lysed to give the 5-hydroxy compound (3). Bromination
of this material with N-bromosuccinimide followed by
O-methylation, afforded (4) and the 1,2,6-trimethyl or
2,6-dimethyl piperazine could then be introduced to give
(5). In the case of dimethylpiperazine, only one isomer
was obtained.
f
Me
Me
R
R
N
N
Ac
N
H
N
g
N
N
Me
Me
OMe
OMe
(5)
(6)
Scheme 1. Synthesis of indoline (6). Reagents: (a) AcCl, AlCl3, DCM,
81%; (b) AcOOH, 90%; (c) NaOH, 100%; (d) NBS, AcOH, 64%; (e)
MeI, K2CO3, DMF, 92%; (f) cis-2,6-dimethylpiperazine (R = H, Me),
BINAP, Pd(OAc)2, Cs2CO3, dioxane, 63%; (g) 2 M HCl, 90%.
Acidic hydrolysis then gave the NH indoline (6) and the
final amides and phenylacetamides were prepared by
either acylation with the appropriate acid chloride or
via EDC/HOBt coupling with the acid. Carbamates
were prepared from the chloroformates and the ureas
were prepared by reaction with the appropriate isocya-
nate; those not commercially available were prepared
in situ from the appropriate amine and triphosgene
(Scheme 2).
had shown reduced potency at 5-HT1B, when a second
aryl ring was introduced, this gave a significant increase
in 5-HT1B receptor affinity and afforded compounds,
such as 20 (Table 3), with nanomolar potency and full
antagonism in the [35S]GTPcS functional assay.
However, these compounds showed either insufficient
5-HT1B/1D selectivity, inhibited P450 isoforms (data
not shown) or had high CLi in microsomes precluding
further studies. Introduction of a further ring onto the
biaryl side chain produced many analogues with excel-
lent 5HT1B affinity and selectivity. This SAR mimics
The oxadiazole biphenyl acid (9) required for SB-
616234 was prepared from the bromobenzonitrile (7)
via reaction with hydroxylamine, followed by acetic
Me
R
N
H
N
N
Me
OMe
(6)
a or b
c
Me
Me
N
H
O
R
O
R
N
N
X
Ar
N
Ar
N
N
N
Me
Me
OMe
OMe
Scheme 2. Synthesis of amides, ureas and carbamates. Reagents: (a) ArXCOCl, Et3N, DCM; (b) ArXCOOH, EDC, HOBt, DCM; (c) ArNCO,
DCM.
Me
Me
Me
a, b
Me
Me
c
N
N
N
N
NC
Br
COOH
Br
O
O
(9)
(7)
(8)
Scheme 3. Synthesis of oxadiazole biphenyl acid. Reagents: (a) H2NOH, MeOH, 90%; (b) Ac2O, 100%; (c) 4-carboxybenzeneboronic acid,
Pd(PPh3)4, Na2CO3, DME, H2O, 74%.

4712
P. A. Wyman et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4708–4712
Beattie, D.; Kelly, H. A.; Mitchell, W. L.; Oxford, A.
W.; Wadsworth, A. H.; Tyers, M. B. J. Med. Chem.
1994, 37, 2253.
anhydride to give (8). This was coupled with 4-carb-
oxybenzeneboronic acid under Suzuki reaction condi-
tions to afford (9). Other biaryl acids were prepared
by similar reported procedures20 (Scheme 3).
10. Clitherow, J. W.; King, F. D.; Middlemiss, D. N.;
Wyman, P. A. In Progress in Medicinal Chemistry; King,
F. D.; Oxford, A. W., Eds.; Elsevier Amsterdam. 2003,
Vol. 41, pp 129–165.
11. Slassi, A. Curr. Top. Med. Chem. 2002, 2, 559.
12. Barnes, N. M.; Sharp, T. Neuropharmacology 1999, 38,
1083.
13. Pauwels, P. J. CNS Drugs 1996, 2, 415.
14. Pauwels, P. J.; Colpaert, F. C. Neuropharmacology 1995,
34, 235.
In summary, we have been able to identify 5-HT1B
receptor antagonists with increased selectivity and re-
duced intrinsic agonist activity through the introduction
of cis-2,6-dimethyl substitution onto the piperazine ring
of a mixed 5-HT1A/1B/1D ligand. Details of the in vitro
and in vivo activity of SB-616234 (24) will be described
in forthcoming papers.21,22
15. Pauwels, P. J.; Palmier, C.; Wurch, T.; Colpaert, F. C.
Naunyn-Schmiedebergs Arch. Pharmacol. 1996, 353, 144.
16. Gaster, L. M.; Blaney, F. E.; Davies, S.; Duckworth, D.
M.; Ham, P.; Jenkins, S.; Jennings, A. J.; Joiner, G. F.;
King, F. D.; Mulholland, K. R.; Wyman, P. A.; Hagan, J.
J.; Hatcher, J.; Jones, B. J.; Middlemiss, D. N.; Price, G.
W.; Riley, G.; Roberts, C.; Routledge, C.; Selkirk, J.;
Slade, P. D. J. Med. Chem. 1998, 41, 1218.
17. Ahlgren, C.; Eriksson, A.; Tellefors, P.; Ross, S. B.;
Stenfors, C.; Malmberg, A. Eur. J. Pharmacol. 2004, 499,
67.
18. Watson, J.; Roberts, C.; Scott, C.; Kendall, I.; Collin, L.;
Day, N. C.; Harries, M. H.; Soffin, E.; Davies, C. H.;
Randall, A. D.; Heightman, T.; Gaster, L.; Wyman, P.;
Parker, C.; Price, G. W.; Middlemiss, D. N. Br.
J. Pharmacol. 2001, 133, 797.
References and notes
1. Hartig, P. R.; Hoyer, D.; Humphrey, P. P.; Martin, G. R.
Trends Pharmacol. Sci. 1996, 17, 103.
2. Hoyer, D.; Hannon, J. P.; Martin, G. R. Pharmacol.
Biochem. Behav. 2002, 71, 533.
3. Roberts, C.; Price, G. W. Neurosci. Lett. 2001, 300, 45.
4. Roberts, C.; Price, G. W.; Middlemiss, D. N. Brain Res.
Bull. 2001, 56, 463.
5. Stamford, J. A.; Davidson, D.; McLaughlin, D. P.;
Hopwood, S. Trends Neurosci. 2000, 23, 459.
6. Middlemiss, D. N.; Gothert, M.; Schlicker, E.; Scott, C.
M.; Selkirk, J. V.; Watson, J.; Gaster, L. M.; Wyman,
P.; Riley, G.; Price, G. Eur. J. Pharmacol. 1999, 375,
359.
7. Roberts, C.; Boyd, D. F.; Middlemiss, D. N.; Routledge,
C. Neuropharmacology 1999, 38, 1409.
19. Watson, J.; Burton, M. J.; Price, G. W.; Jones, B. J.;
Middlemiss, D. N. Eur. J. Pharmacol. 1996, 314, 365.
20. Marshall, H.; Thompson, M.; Wyman, P. A.
WO2001023374.
21. Scott, C.; Langmead, C. J.; Wyman, P.; Smith, P. W.;
Starr, K. R.; Dawson, L. A.; Price, G. W.; Watson, J.
Neuropharmacology, submitted.
8. Roberts, C.; Hatcher, P.; Hagan, J. J.; Jeffrey, P.; Wyman,
P.; Gaster, L. M.; Middlemiss, D. N. Naunyn-Schmiede-
bergs Arch. Pharmacol. 2000, 362, 177.
22. Dawson, L. A.; Hughes, Z. A.; Starr, K. R.; Storey, J. D.;
Bettelini, L.; Arban, R.; Hagan, J. J.; Middlemiss, D. N.;
Price, G. W. Neuropharmacology, submitted.
9. Clitherow, J. W.; Scopes, D. I. C.; Skingle, M.; Jordan,
C. C.; Feniuk, W.; Campbell, I. B.; Carter, M. C.;
Collington, E. W.; Connor, H. E.; Higgins, G. A.;