Non-basic ligands for aminergic GPCRs: The discovery and development diaryl sulfones as selective, orally bioavailable 5-HT2A receptor antagonists for the treatment of sleep disorders

Article abstract of DOI:10.1016/j.bmcl.2010.04.090

Scaffold hopping from a non-basic series of 5-HT_2A receptor antagonists developed in-house that possessed reduced activity in vivo enabled the discovery of a novel series of diaryl sulfones that gave excellent occupancy on oral dosing. Not only does this work further demonstrate that oral bioavailability of a given series can be enhanced by improving physicochemical parameters such as log P, but it corroborates the growing evidence that a protonated amine is not essential for affinity at aminergic GPCRs.

Full text of DOI:10.1016/j.bmcl.2010.04.090

Bioorganic & Medicinal Chemistry Letters 20 (2010) 3708–3712
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Non-basic ligands for aminergic GPCRs: The discovery and development diaryl
sulfones as selective, orally bioavailable 5-HT2A receptor antagonists for the
treatment of sleep disorders
,
*
Tammy Ladduwahetty , Myra Gilligan, Alexander Humphries, Kevin J. Merchant, Rebecca Fish,
George McAlister, Magnus Ivarsson, Maria Dominguez, Desmond O’Connor, Angus M. MacLeod
Merck Sharp & Dohme, Eastwick Road, Harlow, Essex CM20 2QR, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Scaffold hopping from a non-basic series of 5-HT2A receptor antagonists developed in-house that pos-
sessed reduced activity in vivo enabled the discovery of a novel series of diaryl sulfones that gave excel-
lent occupancy on oral dosing. Not only does this work further demonstrate that oral bioavailability of a
given series can be enhanced by improving physicochemical parameters such as log P, but it corroborates
the growing evidence that a protonated amine is not essential for affinity at aminergic GPCRs.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 28 February 2010
Revised 19 April 2010
Accepted 19 April 2010
Available online 24 April 2010
Keywords:
5-HT2A
GPCRs
Non-basic
Diaryl sulfones
G-protein Coupled Receptors (GPCRs) are membrane spanning
proteins, consisting of seven transmembrane (7TM) domains, that
elicit an intracellular signal transduction cascade through the acti-
vation of a G-protein. More than 30% of marketed drugs act on
GPCRs, making them among the most important drug targets for
the pharmaceutical industry.1 Of these, the Rhodopsin (or Family
A) class of GPCRs contain the largest number of receptors targeted
by clinically used drugs2 and include the biogenic amine GPCRs
that are activated by the binding of small, basic amines such as
dopamine, histamine, noradrenaline and serotonin.
For many years, aminergic GPCRs have been postulated to bind
their endogenous ligands in a binding pocket within the 7TM re-
gion.3 Mutagenesis studies have shown that amino acids in the
TM5, TM6 and TM7 are involved in the binding of these ligands4
and, in particular, the interaction between a conserved acidic ami-
no acid (Asp 113 in the D3 receptor: D3.32) on TM3 with the basic
amine of the ligand has been shown crucial to the binding of both
agonists and antagonists. The recent publication of the crystal
structure of the inverse agonist carazolol bound to the b-2 adren-
ergic receptor has confirmed these early hypotheses.5
Among the aminergic GPCRs, the 5-HT2A receptor has enjoyed
particular prominence due to its role in the efficacy of atypical
antipsychotics such as risperidone and olanzapine (Fig. 1).6 The
number of structurally diverse compounds with affinity for the
5-HT2A receptor subtype has allowed the designation of a common
pharmacophore for binding to this receptor.7 Several groups have
recognized the requirement for two aromatic rings and, in com-
mon with other aminergic GPCRs, a basic nitrogen for the binding
of both 5-HT2A agonists and antagonists.8 Furthermore, the optimal
distance between the basic nitrogen and the two aromatic rings, as
well as that between the two aromatic rings, has been defined.9
Contrary to established theory, however, work in these laborato-
ries has shown that deletion of the basic nitrogen from at least one of
the establishedseries of 5-HT2A antagonists can result in compounds
that retain high affinity for the receptor (e.g., 2 and 3, Fig. 1).10 This
demonstration that a basic nitrogen is not crucial for binding al-
lowed us to design structure types that have been hitherto not been
considered in the development of SAR for 5-HT2A antagonists. Fur-
thermore, this finding gave greater scope to avoid the off-target
activities characteristic of many 5-HT2A receptor ligands. For in-
stance, the pharmacophores for binding to the HERG channel, and
dopamine and adrenergic receptors, have much similarity to the
classical binding mode for 5-HT2A receptor ligands.11
* Corresponding author. Tel.: +44 1799 533 552; fax: +44 1799 531 495.
E-mail address: Tammy.Ladduwahetty@glpg.com (T. Ladduwahetty).
Present address: BioFocus, Chesterford Research Park, Saffron Walden, CB10 1XL,
UK.
Although we were able to identify low nanomolar compounds
in the non-basic sulfonylpiperidine series, lead optimization based
on compounds such as 2 and 3 did not result in achieving the
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.04.090

T. Ladduwahetty et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3708–3712
3709
O
duces, to our knowledge for the first time, non-basic 5-HT2A recep-
tor antagonists with high oral bioavailability and in vivo occupancy
comparable to bench-mark compounds. In keeping with other 5-
HT2A antagonists such as 1b reported by these laboratories12, a
compound from this class (11) was also shown to enhance Slow
Wave Sleep (SWS) and reduce the number of awakenings in rats.
On the basis of its excellent in vitro and in vivo profile, 11 was se-
lected as a clinical candidate for the treatment of insomnia and
other sleep disorders (Vide Infra).
N
CH3
N
O
N
F
N
N
N
N
CH3
N
H
CH3
S
Risperidone
Olanzapine
HO
H3C
O
O
In the absence of the basic nitrogen, key features in compounds
such 2 and 3 that may be considered important for activity are the
H-bond acceptor (sulfone or amide) and the two aryl rings. Hence,
a number of ways of linking these key binding elements were ex-
plored. One of the first groups considered as a replacement for the
metabolically labile piperidine was a phenyl ring, giving a com-
pound such as structure 5 (Fig. 1). A phenyl ring has the advantage
that metabolic oxidation may be controllable by appropriate sub-
stitution. Furthermore, this modification removes the potential
for one of principal routes of metabolism of 3 which involves cleav-
age of the piperidine amide.
H3C
N
O
O
F
F
F
S
R
MDL100907
N
F
F
F
1a R=H
1b R=F
O
O
F
O
S
N
N
N
F
F
2
3
The synthesis of the diarylsulfones was initially accomplished
O
using the route shown in Schemes 1 and 2.13
O
S
F
SNAr displacement of the fluorine in 4-fluorobenzaldehyde was
followed by reduction of the aldehyde to prevent further oxidation
of this group to the acid during the oxidation of the sulfide to the
sulfone. Hence oxidation of the sulfide, followed by oxidation of
the primary alcohol to the aldehyde and Wittig reaction with the
requisite phosphonate gave the trans stilbene as the major isomer.
The reduced compound 5 was obtained by hydrogenation of 4 un-
der standard conditions. The substituted compounds 6b–f were
obtained by manipulation of nitrile functionality in 6a according
to the route shown in Schemes 1 and 2.
A convergent route to these molecules was also developed by
synthesizing the aryl sodium sulfinate incorporating the 4-fluoro-
phenyl stilbene (Scheme 3). The coupling of sodium 4-bromophenyl
sulfinate directly with 4-fluorophenylstyrene gave the coupled
product in only 44% yield. Purification of 7, either by recrystallisa-
tion or chromatography, was also difficult hence compromising
the viability of this approach for the large scale synthesis of this
intermediate. Utilising the b-keto protecting group for sulfinates
developed by Baskin and Wang14 gave 8a which was subjected to
the Heck conditions and once again gave the sulfinate 7 in moderate
yield due to cleavage of the protecting group. Reasoning that the
F
5
Figure 1. 5-HT2A receptor antagonists.
sub-nanomolar affinity that is characteristic of bench-mark com-
pounds such as MDL100907 (Table 1). Moreover, one of the com-
pounds from the non-basic series, 3, achieved only 40%
occupancy of 5-HT2A receptors in the CNS after oral dosing in an
in vivo occupancy assay at 10 mg/kg. The relatively low occupancy
of this compound compared to 1b could be a function both of its
lower affinity for 5-HT2A and its low oral bioavailability. In vitro
metabolism studies on 3 showed extensive oxidation of the piper-
idine ring, the phenethyl side chain and cleavage of the amide
bond.
The microsomal instability of the piperidine core of 3 prompted
us to examine potential replacements. This communication
describes this work which has led to the identification of a novel
series of 5-HT2A ligands with subnanomolar affinity and intro-
Table 1
In vitro and occupancy data
Compound
h5-HT2A
h5-HT2C
Selectivityb
hIKr
% Turnover (HLM)c
% Occupancy (dose mg/kg)d
Ki/nMa
Ki/nMa
Ki/nMa
M1009013
0.31
0.42
0.39
2.10
3.90
1.32
1.43
0.65
0.20
0.65
0.07
0.10
0.22
13
39
69
184
100
65
526
127
21
41
7
13
47
42
92
176
84
28
49
368
195
106
63
100
130
213
1100
710
ND
ND
ND
ND
99
33
84
15.8
3
ND
ND
98 (10)
ND
ND
1a
1b
2
3
4
5561
>9000
>9000
4342
>9000
>9000
>9000
7919
5953
4000
8252
40 (10)
ND
5
6b
6c
6d
6f
6f
11
58 (10)
50 (10)
>100 (10)
99 (5)
85 (10)
83 (1)
0
28
<10
45
a
b
c
h5-HT2A, h5-HT2C and hIKr were determined as described in Ref. [6] (n = 2).
h5-HT2C/h5-HT2A
Microsomal turnover data were obtained by incubation of the test compound (and diazepam as a control compound) at 1 M concentration with liver microsomes pooled
.
from a number of individuals, at a protein concentration of 0.5 mg/mL for 0.25 h. Concentration of test compound remaining after incubation was determined by LC/MS/MS
analysis. Turnover values are the mean of three determinations, with a standard deviation of </) 5% for the values quoted.
d
Determined after oral dosing at 0.5 h time point. See Ref. [12] for details of protocol

3710
T. Ladduwahetty et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3708–3712
O
H
O
O
O
F
S
OH
S
O
S
SH
H
a
c
b
R
R
O
R
R=H, CN
O
O
d
S
O
F
(EtO)2OP
S
4
F
F
F
e
R
F
5
4 R=H
F
6a
R=2-CN
f
6a
O
S
O
F
R
6b R=2-CONH2
6c R=3-CONH2
F
6d
R=4-CONH2
Scheme 1. Reagents and conditions: (a) DMSO, 120 °C; (b) NaBH4, EtOH; (c) (i) AcOH, H2O2; (ii) ClCOCOCl, DMSO, CH2Cl2, À78 °C; (d) NaH, 15-Crown-5, THF, 0–25 °C; (e) H2,
10% Pd/C, EtOH; (f) 4 N NaOH, EtOH, 80 °C; (g) (i) DIBAL-H, CH2Cl2, À78 °C; (ii) NaBH4, EtOH; (h) TPAP, CH2Cl2, (ii) MeMgBr, THF.
(EtO)2OP
F
O
S
O
S
O
O
S
O
O
O
HO
F
F
NC
a
b
CN
6e
6a
c
OH
O
O
S
H3C
F
6f
Scheme 2. Reagents and conditions: (a) NaH, 15-Crown-5, THF, 0–25 °C; (b) (i) DIBAL-H, CH2Cl2, À78 °C; (ii) NaBH4, EtOH; (c) TPAP, CH2Cl2, (ii) MeMgBr, THF.
O
O
O
O
F
F
8b
R1
S
S
S
Br
Br
S
+Na-O
-O+Na
O
O
c
b
74%
a
8a R1=COCH3
8b R1=CN
9
R1
7
NC
94%
b
44%
8a
OH
b, 45%
I
d
10
O
F
S
+Na-O
OH
O
F
S
O
7
11
Scheme 3. Reagents and conditions: (a) AcOH, H2O, 100 °C, 1.5 h; (b) Pd(OAc)2, NaOAc, NMP, 135 °C, 35 min; (c) NaOMe, THF, MeOH, 25 °C, 1 h; (d) CuI (1.3 equiv.), DMSO, 90
°C, 88 h.

T. Ladduwahetty et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3708–3712
3711
Table 2
may be become an issue again. Hence, other approaches to improv-
ing the absorption in this series were explored. The propertyl pro-
file of the lead 4 (MW= 356, H-bond donor = 0, H-bond acceptor = 2,
c log P = 5.1) prompted us to examine whether the high log P is a
contributory factor to its sub-optimal in vivo profile. Therefore
incorporation of a primary carboxamide, a polar group shown to
be tolerated in the piperidine series exemplified by 1a15, at the
2-, 3- and 4-positions of the aryl sulfone was undertaken with
the view to reduce log P.
Comparison of dosing vehicles for 4a
Dosing vehicle
Hpv AUC
M h)
Plasma AUC
M h)
Brain AUC
(lM h)
(l
(
l
Methocel suspension
Imwittor/Tween
0.46
6.39
0.09
0.79
0.38
1.96
a
Concentration of compound in rat plasma obtained from hepatic portal vein
(Hpv), cardiac (systemic) blood samples and brain homogenate following po
administration at 5 mg/kg (four animals per group) and measured by electrospray
mass spectrometry. See Ref. [6] for details.
All three carboxamide analogues gave subnanomolar affinity at
the 5-HT2A receptor, with excellent selectivity over both 5-HT2C
and HERG. Interestingly, there were significant differences in both
plasma and brain levels for the three regioisomers when they were
dosed orally in an in vivo occupancy assay (Table 3). Intriguingly,
the receptor occupancies for 6c and 6d did not reflect the differ-
ences in their brain concentrations. The 2-carboxamide analogue
6b gave significantly higher plasma and brain levels than both of
its regioisomers, the high brain levels being reflected in the signif-
icantly higher occupancy of 5-HT2A receptors in the CNS. There is
no obvious explanation for this difference in the in vivo profiles
of the carboxamide isomers. We speculate the proximity of the
amide to the sulfone in 6b may affect ease of solvation/desolvation
of both functional groups and influence the ability of the molecule
to cross membranes, including the blood–brain barrier.
Table 3
Comparison of occupancy, brain and plasma levels for 6b, 6c and 6d
6b
10
100
2727
5770
6c
6d
10
53
300
620
Dose (mg/kg)
10
50
40
40
Occupancy (% inhibition)a
Plasma concentration (nM)
Brain concentration (nM)
a
Determined after oral dosing a methocel suspension at the 0.5 time point. See
Refs. [6] and [12] for details of protocol.
Table 4
In vitro and in vivo pharmacokinetic properties for 6b and 11a.
Further development of 6b was disappointingly prevented by
its low microsomal turnover which was reflected in the exception-
ally long half-lives in all three preclinical species (Table 4). Com-
pound 6b was also shown to be an activator of the nuclear
pregnane X (PXR) receptor16 which is indicative of its potential
b
Compound
Species
% Turnover
T1/2 (h)
Clp
3
6b
Rat
Dog
Rhesus
0
28
134
85
<10
23
0.3
0.13
95c
8
32
11
Rat
Dog
Rhesus
Human
50
34
44
45
3.8
3.15
2.13
to induce CYP3A4 enzymes (PXR activation 33–158% @ 10 lM rel-
ative to the positive control Rifampicin). The effect of the 2-carbox-
amido group on improving the absorption in the diaryl sulfone
series prompted us to examine other polar groups at this position
that would also have the effect of lowering the log P of the original
lead 6b. We examined the primary alcohol 6e which, like 6b, gave
excellent in vitro and in vivo profiles (Table 1). The CYP induction
potential assay showed that unlike 6b, 6e had a negligible effect
a
Determined in 15 male Sprague–Dawley rats, 3 dosed i.v. (0.1 mg/kg) and 3
dosed p.o. (1 mg/kg). See Ref. [6] for details of protocol.
b
mL/min/kg.
Vdss 19L/kg.
c
at the same concentration (0.9% @ 10 lM). In contrast, its CYP inhi-
cleavage occurred most probably due to the basicity of the reaction
conditions, we decided to synthesise the nitrile analogue of 8a since
it may be more stable to the Heck conditions, the pKa of the protons
adjacent to a nitrile being higher compared to the ketone (pKa
CH2CN 25 vs. CH2COCH3 20). Nucleophilic addition of sodium-4-
bromophenylsulfinate to acrylonitrile gave the protected sulfinate
8b which was coupled with 4-fluorophenylstyrene to give the stil-
bene 9 in good yield, after a single recrystallisation. The pure sulfi-
nate 7 could be obtained by deprotection of the sulfone. The
substituted sulfinates were coupled under copper catalysis with aryl
iodides utilizing conditions developed by Suzuki and Hajime15 as
shown for the synthesis of 11, the (S)-enantiomer of 6f.
The first compound in this series, 5, gave nanomolar affinity at
the 5-HT2A receptor with moderate selectivity over 5-HT2C (Table
1). Metabolism in human liver microsomes (HLM), however, was
high. The stilbene analogue 4 gave significantly improved selectiv-
ity over 5-HT2C as well as reduced turnover on HLM. However, dos-
ing 4 orally as a methocel suspension at 10 mg/kg once again gave
only moderate occupancy (40%) of 5-HT2A receptors. An in vivo
absorption study with 4 dosed orally as a methocel suspension
and, separately, as a solution in Imwittor/Tween revealed that sig-
nificantly improved plasma levels were obtained when it was for-
mulated as a solution (Table 2).
bition profile indicated that 6e was a potent inhibitor of the
CYP2C9 P450 isozyme (IC50 CYP2C9 100 nM).
Fortunately it was not necessary to make significant changes to
the structure of 6e to balance the CYP induction and inhibition pro-
files whilst retaining the excellent affinity and selectivity of this
series towards the 5-HT2A receptor. The secondary alcohol 6f was
synthesised and resolved. The more active (S)-enantiomer 1117
was shown to have subnanomolar 5-HT2A affinity with selectivity
over 5-HT2C, HERG, dopamine and adrenergic receptors. In a MDS
Pharma screen 11 was also shown to have no significant off target
activities against any other enzymes, ion channels or GPCRs.
A rising dose study with 11 showed that plasma level of 60 nM
(at a dose of <1 mg/kg) in rat resulted in 80% occupancy of 5-HT2A
The removal of the basic nitrogen resulted in reduced solubility
of the non-basic compounds and so consideration was given to re-
incorporation of a basic group in another part of the molecule, this
time not to enhance 5-HT2A binding, but as a solubilising group.
One disadvantage in this approach is that HERG channel affinity
Figure 2. Oral dose–response for 11 2h post-dose.

3712
T. Ladduwahetty et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3708–3712
Figure 3. Telemetry studies in rat for compound 11 (3 mg/kg, p.o.). See Ref. [12] for details of protocol.
receptors in the CNS, with full occupancy being achieved at a dose
of 3 mg/kg18 (Fig. 2). Its CYP induction (8% of control @ 10
M) and
CYP inhibition (CYP2C9 6 M, CYP1A 4.8 M, other CYPs >30 M)
profiles were acceptable in light of the plasma levels needed for
complete occupancy of 5-HT2A receptors. The cardiovascular profile
of 11 in dog was benign with no effect seen on QTc or heart rate
after i.v. infusion of up to 10 mg/kg (cumulative mean plasma level
References and notes
l
1. Schlyer, S.; Horuk, R. Drug Discov. Today 2006, 5, 481.
2. Lagerstrom, M. C.; Schioth, H. B. Nat. Rev. Drug. Discov. 2008, 13, 339.
3. Costanzi, S. J. Med. Chem 2008, 51, 2907.
4. Ravina, E.; Negreira, J.; Cid, J.; Masaguer, C. F.; Rosa, E.; Riva, M. E.; Fontenla, J.
A.; Loza, M. I.; Tristan, H.; Cadavid, M. I.; Sanz, F.; Lozoya, E.; Carotti, A.; Carrieri,
A. J. Med. Chem. 1999, 42, 2774.
l
l
l
5. Cherezov, V. et al Science 2007, 318, 1258.
80.9
lM). Consistent with other 5-HT2A antagonists developed in
6. Fletcher, S. R.; Burkamp, F.; Blurton, P.; Cheng, S. K. F.; Clarkson, R.; O’Connor,
D.; Spinks, D.; Tudge, M.; van Niel, M. B.; Patel, S.; Chapman, K.; Marwood, R.;
Shepheard, S.; Bentley, G.; Cook, G. P.; Bristow, L. J.; Castro, J. L.; Hutson, P. H.;
MacLeod, A. M. J. Med. Chem. 2002, 45, 492.
7. Westkaemper, R. B.; Glennon, R. A. Curr. Top. Med. Chem. 2002, 2, 575.
8. (a) Fritz, J.; Bandelow, B. Int. J. Psychiatry Clin. Pract. 1998, 2, 2133–2155; (b)
Rampe, D.; Murawsky, M. K.; Grau, J.; Lewis, E. W. J. Pharmacol. Exp. Ther. 1998,
286, 788.
9. Kristiansen, K.; Kroeze, W. K.; Willins, D. L.; Gelber, E. I.; Savage, J. E.; Glennon,
R. A.; Roth, B. L. J. Pharmacol. Exp. Ther. 2000, 293, 735.
10. Ladduwahetty, T.; Boase, A. L.; Mitchinson, A.; Quin, C.; Patel, S.; Chapman, K.;
MacLeod, A. M. Biorg. Med. Chem. Lett. 2006, 16, 3201.
11. Du, L.-P.; Tsai, K.-C.; Li, M.-Y.; You, Q.-D.; Xia, L. Bioorg. Med. Chem. Lett. 2004,
14, 4771.
12. Fish, L. R.; Gilligan, M. T.; Humphries, A. C.; Ivarsson, M.; Ladduwahetty, T.;
Merchant, K. J.; O’Connor, D.; Patel, S.; Phillips, E.; Vargas, H. M.; Hutson, P. H.;
MacLeod, A. M. Bioorg. Med. Chem. Lett. 2005, 15, 3665.
13. Experimental details for the synthesis of the compounds referred to in this
work can be found in WO 2006021805.
14. Baskin, J. M.; Wang, Z. Tetrahedron Lett. 2002, 43, 8479.
15. Suzuki, H.; Hajime, A. Tetrahedron Lett. 1995, 36, 6239.
16. Goodwin, B.; Redinbo, M. R.; Kliewer, S. A. Ann. Rev. Pharmacol. Toxicol. 2002,
42, 1.
17. The in vitro profile for the less active enantiomer of 6f was 5-HT2A 1 nM, 5-HT2C
150 nM, selectivity for 5-HT2A 150-fold.
18. A PET study showed that an Occ80 was achieved at plasma levels of 170 nM in
rhesus monkey. The authors would like to thank the Imaging Group at Merck &
Co., West Point, PA for these results.
19. Good dose proportional increases in plasma levels after oral dosing at 1–100
mg/kg in rats was seen although some variability was observed at the top dose.
20. Compound 11 was shown to be stable both as a solid and in solution in
acetonitrile/water (pH 1–10, 1500 °C, 5 days). In a phototoxicity assay, after
oral dosing of 15 to mice at 10–2000 mg/kg for 13 days showed slight
erythema at the top dose on day 1 only.
these laboratories, 11 was shown to significantly increase Slow
Wave Sleep (SWS) and decrease the number of awakenings in rats
over a 15 h period, as measured by EEG and EMG recordings and
compared to vehicle, at a dose of 3 mg/kg (Fig. 3). No effects on Ra-
pid Eye Movement (REM) sleep or sleep onset were observed indi-
cating 11 to be devoid of sedative side effects. Compound 11 was
also shown to be bioavailable in three pre-clinical species (F% rat
44%, dog 45%, rhesus 10%) with acceptable half-lives that would
enable its development as a treatment for sleep disorders in man
(Table 4).19
This account described our work on utlising a series of piperi-
dine sulfonamide/amide non-basic 5-HT2A ligands, which did not
have optimal in vitro and in vivo profiles, in order to identify a ser-
ies of novel, non-basic bis-aryl sulfones with subnanomolar affinity
for the 5-HT2A receptor and selectivity over 5-HT2C, HERG and a
broad panel of other GPCRs and ion channels. Our work culminated
in the identification of 11, a compound that was demonstrated to
be bioavalable in three preclinical species to give 80% receptor
occupancy of central 5-HT2A receptors at doses <1 mg/kg whilst
giving a benign cardiovascular profile in dog up to plasma levels
well above those required for a therapeutic effect. On the basis of
its excellent in vitro and in vivo profile, 1120 was selected as a clin-
ical candidate for the treatment of sleep disorders.
Acknowledgements
The authors thank Kevin Wilson, Adam Beard and Peter H. Hut-
son for contributions towards the preparation of this manuscript.