Discovery of potent, orally bioavailable, selective 5-HT1A/B/D receptor antagonists

Article abstract of DOI:10.1021/jm8001444

5-HT₁ receptor antagonists have been discovered with good selectivity over the 5-HT transporter. This is the first report of highly potent, selective ligands for the 5-HT_1A/B/D receptors with low intrinsic activity, which represent a

Full text of DOI:10.1021/jm8001444

J. Med. Chem. 2008, 51, 2887–2890
2887
Discovery of Potent, Orally Bioavailable,
Selective 5-HT1A/B/D Receptor Antagonists
Simon E. Ward,* Peter J. Eddershaw,# Claire M. Scott,
Laurie J. Gordon, Peter J. Lovell, Susan H. Moore,
Paul W. Smith, Kathryn R. Starr, Kevin M. Thewlis, and
Jeannette M. Watson
Figure 1. Structures of 1, a dual 5-HT1A receptor antagonist and 5-HT
reuptake inhibitor, and 2, a selective 5-HT1D receptor antagonist.
Psychiatry Centre of Excellence for Drug DiscoVery and Molecular
DiscoVery Research, GlaxoSmithKline, New Frontiers Science Park,
Third AVenue, Harlow, Essex, CM19 5AW, U.K.
Scheme 1a
ReceiVed February 11, 2008
Abstract: 5-HT1 receptor antagonists have been discovered with good
selectivity over the 5-HT transporter. This is the first report of highly
potent, selective ligands for the 5-HT1A/B/D receptors with low intrinsic
activity, which represent a useful set of molecules for further
understanding the roles of the 5-HT1 receptor subtypes and providing
new approaches for the treatment of depression.
Of the biological systems that have been investigated as
targets for the treatment of depression and related psychiatric
diseases, there has been, and continues to be, a considerable
focus in targeting pathways that are involved in the regulation
of synaptic 5-hydroxytryptamine (5-HT) levels. Indeed, there
is a large body of preclinical and clinical evidence that indicates
a link between reduced synaptic 5-HT concentrations and
depression, and consequently intervention to bring about an
elevation of 5-HT levels should alleviate those symptoms
associated with the disease.1 The most established approach to
this is demonstrated by block of 5-HT reuptake through
inhibition of the 5-HT transporter (SerT) giving rise to the
successful group of antidepressants, the SSRIs.a
Although SSRIs are an effective means of elevating synaptic
5-HT levels and thus of treating depression, there still exist a
number of challenges to be met. Specifically, improvements are
sought in the responder rate, the latency to onset of action, and
the side effect profile, including nausea and sexual dysfunction.
In particular, to address the latency to therapeutic onset that
typically requires several weeks of treatment, studies have been
conducted to investigate the role of 5-HT1 receptors in this
process. This work has led to the hypothesis that the latency to
therapeutic onset of SSRI action is attributable to time required
for the 5-HT1 receptors to desensitize,2 and consequently
antagonism of one or more of these 5-HT1 autoreceptors alone
or in combination with inhibition of SerT may offer alternative
antidepressant drugs.3 Recent publications from our group in
this field have described the identification of selective 5-HT1D
antagonists,4 mixed 5-HT1A antagonist/SSRIs,5,6 and 5-HT1A/
B/D ligands,7 and clearly there exists within this chemotype the
ability to identify a range of distinct pharmacological profiles.
In particular, to complement the existing profiles studied in vivo,
we were keen to prepare further analogues of 1 (SB-649915,6
Figure 1) to explore the substitution pattern around the phenyl
a Reagents and conditions: (i) BrCH2CH2Br, K2CO3, butan-2-one, 80
°C, 91%; (ii) 1-Boc-piperazine, K2CO3, DMF, 70 °C, 37%; (iii) 1 M HCl
in Et2O, EtOH, 91%; (iv) X-PhCHO, NaBH(OAc)3, dichloroethane,
45-75%.
ring present in the benzoxazinone group with the aim of
identifying potent, selective 5-HT1 receptor antagonists, i.e.,
molecules with pKi > 8 against the 5-HT1A, 5-HT1B, and 5-HT1D
receptors and importantly low intrinsic activity and good
selectivity over other receptors and transporters, including the
serotonin transporter (SerT). Compounds of this profile should
thus be expected to be free of the side effects associated with
the SSRIs (vide supra).
The work that had focused on identification of a selective
5-HT1D receptor antagonist and led to the discovery of 2 (SB-
714786,4 Figure 1) had clearly demonstrated that replacement
of the benzoxazinone group with a simply substituted phenyl
led to a loss in affinity for the 5-HT1A and 5-HT1B receptors
and afforded potent, selective 5-HT1D antagonists. This work
was continued by broadening the initial array run to explore
reductive alkylation products from starting piperazine B, which
could mimic the properties of the parent benzoxazinone to
recover affinities for the 5-HT1A and 5-HT1B receptors. For this
initial work outlined in Scheme 1, 1,2-dibromoethane was
reacted sequentially with 5-hydroxy-2-methylquinoline and then
1-(tert-butoxycarbonyl)piperazine to afford intermediate A.
Deprotection of this product afforded key intermediate B, which
was further elaborated by reductive alkylation to afford a set of
substituted benzylamines C.
Consistent with previous observations, the results of this array
demonstrated that high affinity for the 5-HT1D receptor can be
maintained even for the unsubstituted phenyl derivative 3 and
that this modification also resulted in a loss of affinity for the
5-HT1A and 5-HT1B receptors and the 5-HT transporter. Interest-
ingly, substitution of this phenyl ring with the constituent
functional groups present in the parent 3, namely, the m-
acetamide 4 or p-methoxy 5, resulted in some restoration of
affinities for the 5-HT1A receptor, particularly for the acetamide
(Table 1).
* To whom correspondence should be addressed. Phone: 44 (0)1279
622894. Fax: 44 (0)1279 622790. E-mail: Simon.E.Ward@gsk.com.
#
Current address: UCB, Granta Park, Great Abingdon, Cambridge, CB21
6GS, U.K.
a Abbreviations: SSRI, selective serotonin reuptake inhibitor; SerT, 5-HT
transporter; DMF, N,N-dimethylformamide; THF, tetrahydrofuran; PSA,
polar surface area.
10.1021/jm8001444 CCC: $40.75
2008 American Chemical Society
Published on Web 04/24/2008

2888 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 10
Letters
Table 1. Receptor Binding Affinity (pKia) for 5-HT1A, 5-HT1B, 5-HT1D
and Functional Inhibition of SerT for Novel Compounds
further by preparation of tertiary amide 9. This modification
had limited impact on the pharmacological profile, confirming
the key interaction is presumably as a hydrogen bond acceptor
rather than donor, which is in line with previous studies on an
alkylated series of benzoxazinones.7
During the course of this work, the binding assay for the
5-HT1A receptor was revised from displacement of [3H]8-OH-
DPAT (a 5-HT1A agonist) to displacement of [3H]WAY-100635
(a 5-HT1A antagonist). It has been well reported that h5-HT1A
receptors that have been expressed in cultured cell lines can
exist in low and high agonist affinity states, with the overall
receptor density being considerably higher in the recombinant
versus native tissue environment. Furthermore, the difference
between the measured pKi from displacement of a 5-HT1A
agonist such as [3H]8-OH-DPAT and a 5-HT1A antagonist such
as [3H]MPPF or [3H]spiperone can be used to predict the
functional efficacy at h5HT1A receptors expressed in HEK293
cells.8 As such and in order to ensure we were targeting
molecules with low intrinsic activity against the 5-HT1A receptor,
we generated data for the displacement of an antagonist
radioligand and compared the results to those previously
obtained for displacement of an agonist radioligand (Table 2).
5-HT1A 5-HT1A
compdb
o
m
p
[D]c
[W]d 5-HT1B 5-HT1D SerT
1
2
3
4
5
6
7
8
9
9.5
8.6
ND
6.7
6.7
7.3
ND
ND
5.9
6.6
8.0
6.7
6.6
7.7
6.3
6.8
5.9
5.8
7.3
8.8
9.1
9.0
9.3
7.9
8.6
6.7
8.3
9.5
8.0
6.5
6.1
5.9
6.4
6.2
5.8
5.0
6.0
6.5
7.6
8.7
H
H
H
H
H
H
H
H
NHAc
H
OMe
H
OMe 8.3
7.9
NHAc 6.8
H
NHAc H
H
ND
ND
H
NMeAc H
a Radioligand binding assay to determine affinity at human recombinant
5-HT receptors and functional [3H]5-HT uptake assays in rat cortical
synaptosomes to determine potency for SerT. Each determination lies within
0.3 log units of the mean with a minimum of three replicates. ND ) not
determined. b All compounds were characterized and purity was assessed
using 1H NMR and LCMS. c Radioligand used was [3H]8-OH-DPAT [D].
d Radioligand used was [3H]WAY100635 [W]. Radioligand for 5-HT1B/D
is [3H]5-HT.
Despite the high binding affinities for the acetamide 4, it was
clear that its level of intrinsic activity for the 5-HT1A receptor
was too high. As such, it was decided to continue exploration
in the piperidine core series, which had previously shown lower
levels of intrinsic activity.6 For this exploration (Scheme 2),
the key intermediate D was prepared by Horner-Wadsworth-
Emmons reaction of the phosphonate ester derived from
3-nitrobenzyl bromide. This key intermediate could then be
coupled with the quinolinyl fragment to afford intermediate E,
which required reduction of alkene and nitro functionalities to
generate the aniline which could be derivatized further as shown.
Gratifyingly, this modification had the desired effect, and the
direct piperidine analogue 10 showed not only lower intrinsic
activity but also higher measured affinity for displacement of
the antagonist radioligand in line with our earlier set of data.7
It is noteworthy, however, that although affinities for the 5-HT1A
and 5-HT1D receptors are generally improved in this investiga-
tion, the target 5-HT1B receptor affinity of pKi > 8 proved
routinely difficult to achieve (Table 3). Standard exploration of
the amide functionality encompassing the size of the alkyl group
(11 and 12) and the introduction of an additional substituent
on the nitrogen (13 and 14) showed limited differentiation from
the parent acetamide 10. Increasing the alkyl size to ethyl 11
Table 2. Receptor Radioligand Binding Affinity for Human
Recombinant 5-HT1A Receptor (pKia)
compd
5-HT1A [D]b
5-HT1A [W]c
∆pKi
5-HT1A IAd
WAY-100635
1
4
8-OH-DPAT
5-HT
9.8
9.5
8.7
9.1
8.9
9.1
8.6
6.7
5.4
5.0
0.7
0.9
2.0
3.7
3.9
0
0.1
0.5
0.9
1
a Each determination lies within 0.3 log units of the mean with a
minimum of three replicates. b Radioligand used was [3H]8-OH-DPAT [D].
c Radioligand used was [3H]WAY100635 [W]. d IA ) intrinsic activity.
Transposition of the m-acetamide group to the ortho 8 and
para 7 positions resulted in a reduction in affinity for all 5-HT
receptors, although interestingly the p-methoxy group could be
transposed to the meta position 6 with comparable binding
affinities to the para substituted analogue. Furthermore, it was
clear that selectivity over the 5-HT transporter (SerT) could be
more easily achieved, as all groups replacing the benzoxazinone
showed considerably reduced SerT affinity, including those such
as the m-acetamide or p-methoxy just described. Given the
superiority for the m-acetamide over the other substituents
investigated, the role of the secondary amide was investigated
Scheme 2a
a Reagents and conditions: (i) P(OEt)3, toluene, reflux 24 h, (68%); (ii) N-Boc-piperid-4-one, NaH, THF, room temp, 4 h (100%); (iii) 1 M HCl in Et2O,
72 h (85%); (iv) D, K2CO3, DMF, 100 °C, 16 h, 78%; (v) 10% Pd on C, H2, 0.5 M KOH in EtOH, 24 h, 54%; for amides, (vi) R1COCl or (R1CO)2O, NaH,
THF, 0 °C (60-100%); for sulfonamides, (vi) R1SO2Cl or (R1SO2)2O, NEt3, CH2Cl2 (65-90%); for ureas, (vi) RNCO, NEt3, toluene, 60 °C (48-59%);
(vii) NaH, THF, R2I, 0 °C (63-74%).

Letters
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 10 2889
Table 3. Receptor Binding Affinity (pKia) for 5-HT1A, 5-HT1B, 5-HT1D
and SerT for Novel Aniline Derivatives
A number of these molecules underwent pharmacokinetic
evaluation in rat to assess suitability for use as tool compounds
in vivo.10 From the data presented (Table 4) it was clear that
the original acetamide analogue 10 and the secondary amide
analogues 12-14 as well as tertiary amide 15 gave similar
pharmacokinetic profiles with modest blood clearance and
elimination half-lives and brain to blood ratios consistent with
free passage across the blood-brain barrier.
Although the sulfonamides had been advantageous for the
primary in Vitro potencies, we suspected that their increased
polarity would be detrimental with respect to their levels of CNS
penetration. Indeed, comparison of the polar surface areas (PSA)
of acetamide 10 and sulfonamide 18 showed an increase from
54 to 72 which translated, in ViVo, into a reduction in brain/
blood ratio from 1.1 to 0.2 (although brain Cmax levels
remained high due to improved blood concentrations). Accord-
ingly, the less polar tertiary sulfonamide 21 gave a reduction
in PSA to 63 and a corresponding increase in brain/blood ratio
to 0.9.
c
compdb
R1
COMe
R2 R3 5-HT1A 5-HT1B 5-HT1D SerT
10
11
12
13
14
15
16
17
18
19
20
21
22
H
H
H
Me
Et
H
H
H
H
H
H
H
H
H
H
H
H
7.6 (0.3) 7.7 (0.3) 9.3 (0.5) 6.3
7.6 (0.2) 7.8 (0.3) 9.2 (0.3) 6.1
COEt
COiPr
COMe
COMe
8.2 (0)
7.6 (0)
8.3 (0)
8.2 (0)
7.3 (0.3) 8.9 (0.4) 6.3
7.4 (0.3) 8.9 (0.6) 7.1
7.0 (0.3) 8.8 (0.6) 6.1
7.7 (0.4) 8.7 (0.5) 6.7
CH2CH2CH2CO
COCH2NMe2
CONHEt
SO2Me
H
H
H
Me
H
7.8 (inv) 8.2 (0.2) 9.0 (0.4) 6.4
7.5 (0)
8.6 (0)
7.9 (0)
8.6 (0)
8.1 (0)
7.6 (0.2) 9.6 (0.3) 6.6
8.7 (0.3) 9.3 (0) 6.0
7.9 (0) 8.8 (0.5) 6.0
8.9 (0.3) 9.4 (0.4) 6.4
7.9 (0.5) 8.8 (0.4) 6.1
SO2Me
i
SO2 Pr
CH2CH2CH2SO2
SO2Me
H
OMe 6.9
7.4
8.2
6.2
a Radioligand binding assay to determine affinity at human recombinant
5-HT receptors and functional [3H]5-HT uptake assays in rat cortical
synaptosomes to determine potency for SerT. Each determination lies within
0.3 log units of the mean with a minimum of three replicates. Values in
parentheses are intrinsic activity values. b All compounds were characterized
and purity was assessed using 1H NMR and LCMS c Radioligand used was
[3H]WAY100635.
Further affinity profiling of 18 and 21 across the various 5-HT
subunits (Table 5) indicated that they had highest affinity for
the 5-HT1A/B/D receptors, moderate affinity for 5-HT7 and D2,
and little or no affinity against the others, and as such, these
represented our first compounds that had potent affinity for the
5-HT1A/B/D receptors with approximately 1000-fold selectivity
against SerT.
and then isopropyl 12 led to a slight increase in affinity for the
5-HT1A receptor which was offset by a comparable decrease in
5-HT1B receptor binding. Similarly, although substitution on the
nitrogen of the acetamide was clearly tolerated, increasing the
size of the alkyl group from methyl to ethyl (from 13 to 14)
again led to a slight improvement in 5-HT1A affinity and a
comparable decrease in that for 5-HT1B. Interestingly, this
position of the molecule could also tolerate introduction of a
basic group, as demonstrated by dimethylamine 16. The
extension of these investigations into pyrrolidinone 15 gave a
good overall profile that we hoped would also lead to a good
pharmacokinetic profile. A wider investigation of the aniline
derivatives showed no clear advantage for urea 17, although
replacement of the amide with a sulfonamide 18 finally gave
us the improvement in 5-HT1B affinities which allowed us to
identify molecules that now met our primary in vitro criteria.
As for the amides, larger alkyl groups 20 and substitution on
the sulfonamide nitrogen 19 were tolerated and gave molecules
with pKi greater than our target of 8, although the formation of
the cyclic sultam 21 led to a drop in the 5-HT1B affinity.
Surprisingly, reintroduction of a p-methoxy substituent giving
22 to mimic the functionality present in the benzoxazinone led
to a marked decrease in binding affinities relative to 18.
With these discoveries, we proceeded to evaluate this
pharmacological and pharmacokinetic profile in vivo with the
aim of not only establishing an in vivo pharmacological effect
but also further exploring the 5-HT1B receptor, which, unlike
the 5-HT1A and 5-HT1D receptors, showed an intrinsic activity
greater than 0 in the recombinant cell line. In order to fulfill
both requirements, we investigated the activity of the first
analogue prepared, 18, in a 5-HT1B pharmacodynamic model
that involves administration of a 5-HT1B/D agonist, SKF99101
(15 mg/kg, ip) to induce an increase in seizure threshold in the
maximal electroshock threshold test in the rat.9 Gratifyingly,
oral administration of 18 demonstrated a clear, dose-dependent
reversal of the agonist induced increase in seizure threshold in
line with that achieved for the selective 5-HT1B receptor
antagonist SB-224289, with an approximate ED50 of 1 mg/kg
with exposures in line with data observed from the rat PK
screen. This result is particularly significant, as it confirms 18
was acting as a 5-HT1B antagonist in vivo.
This work has therefore allowed us to generate a new profile,
achieving potent affinities across the 5-HT1A/B/D receptors.
Furthermore, we have achieved an appropriate pharmacokinetic
profile in rat to allow for demonstration of functional antagonism
Table 4. Pharmacokinetic and in Silico Profiles of Aniline Derivatives in Rata
compd
R1
R2
CLb ((mL/min)/kg)
t1/2 (h)
CNS br/bl
Vss (L/kg)
brain Cmax (ng/g)
clogD pH 7.4
PSA
10
12
13
14
15
18
21
COMe
COiPr
COMe
COMe
CH2CH2CH2CO
SO2Me
H
H
Me
Et
47
45
37
34
35
38
39
0.6
0.9
1.2
1.1
1.3
0.6
0.9
1.1
2.3
ND
ND
1.2
0.2
0.9
1.2
1.8
2.2
1.9
1.5
1.6
0.9
221
175
190
234
410
346
287
3.6
4.5
3.6
4.1
3.9
3.5
3.4
54
54
46
46
46
72
63
H
CH2CH2CH2SO2
a In vivo data determined by 1 mg/kg iv study in rat. Brain Cmax and brain/blood ratio determined by additional 3 mg/kg oral rat PK study (br/bl from
whole brain AUC concentration measurements). ND ) not determined.
Table 5. In Vitro Cross-Screening Characterization (pKi) of 18 and 21 at h5-HT Receptors and Representative Other Monoaminergic Receptorsa
1A
1B
1D
1E
5.6
<5
1F
2A
5.6
<5.2
2B
5.7
<6
2C
<6
<5.4
4
5A
6
7
SerT
D2
R1B
ꢀ2
18
21
8.6
8.1
8.7
7.9
9.3
8.8
6.3
6.0
<5
5.1
5.9
5.8
<5
5.8
7.7
7.8
5.8
6.4
7.2
6.3
6.7
6.8
6.3
5.8
a Radioligand binding assays from cloned human 5-HT receptors subtypes.

2890 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 10
Letters
nists and serotonin reuptake inhibitors. Bioorg. Med. Chem. Lett. 2007,
17, 1033–1036.
in vivo. These molecules and this pharmacological profile are
now being further investigated for their potential to deliver a
differentiated antidepressant profile.
(6) Atkinson, P. J.; Bromidge, S. M.; Duxon, M. S.; Gaster, L. M.; Hadley,
M. S.; Hammond, B.; Johnson, C. N.; Middlemiss, D. N.; North, S. E.;
Price, G. W.; Rami, H. K.; Riley, G. J.; Scott, C. M.; Shaw, T. E.;
Starr, K. R.; Stemp, G.; Thewlis, K. M.; Thomas, D. R.; Thompson,
M.; Vong, A. K.; Watson, J. M. 3,4-Dihydro-2H-benzoxazinones are
5-HT(1A) receptor antagonists with potent 5-HT reuptake inhibitory
activity. Bioorg. Med. Chem. Lett. 2005, 15, 737–741.
Supporting Information Available: Experimental procedures
and data for final compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
(7) Ward, S. E.; Johnson, C. N.; Lovell, P. J.; Scott, C. M.; Smith, P. W.;
Stemp, G.; Thewlis, K. M.; Vong, A. K.; Watson, J. M. Studies on a
series of potent, orally bioavailable, 5-HT1 receptor ligands. Bioorg.
Med. Chem. Lett. 2007, 17, 5214–5217.
(8) Watson, J.; Collin, L.; Ho, M.; Riley, G.; Scott, C.; Selkirk, J. V.;
Price, G. W. 5-HT(1A) receptor agonist-antagonist binding affinity
difference as a measure of intrinsic activity in recombinant and native
tissue systems. Br. J. Pharmacol. 2000, 130, 1108–1114.
(9) Stean, T. O.; Atkins, A. R.; Heidbreder, C. A.; Quinn, L. P.; Trail,
B. K.; Upton, N. Postsynaptic 5-HT 1B receptors modulate elec-
troshock-induced generalised seizures in rats. Br. J. Pharmacol. 2005,
144, 628–635.
References
(1) Hirschfeld, R. M. History and evolution of the monoamine hypothesis
of depression. J. Clin. Psychiatry 2000, 61, 4–6.
(2) de Montigny, C.; Chaput, Y.; Blier, P. Modification of serotonergic
neuron properties by long-term treatment with serotonin reuptake
blockers. J. Clin. Psychiatry 1990, 51, 4–8.
(3) Roberts, C.; Price, G. W.; and Middlemiss, D. N. Ligands for the
investigation of 5-HT autoreceptor function. Brain Res. Bull. 2001,
56, 463–469.
(4) Ward, S. E.; Harrington, F. P.; Gordon, L. J.; Hopley, S. C.; Scott,
C. M.; Watson, J. M. Discovery of the first potent, selective
5-hydroxytryptamine1D receptor antagonist. J. Med. Chem. 2005, 48,
3478–3480.
(10) All animal work was conducted in compliance with the Home Office
guidance on the operation of the Animals (Scientific Procedures) Act
1986.
(5) Lovell, P. J.; Goodacre, C. J.; Smith, P. W.; Scott, C. M.; Starr, K. R.;
Thewlis, K. M.; Vong, A. K. K.; Ward, S. E.; Watson, J. M. 3,4-
Dihydro-2H-benzoxazinones as dual-acting 5-HT1A receptor antago-
JM8001444