Novel imidazobenzazepine derivatives as dual H1/5-HT 2A antagonists for the treatment of sleep disorders

Article abstract of DOI:10.1016/j.bmcl.2010.07.029

A novel imidazobenzazepine template (5a) with potent dual H ₁/5-HT_2A antagonist activity was identified. Application of a zwitterionic approach to this poorly selective and poorly developable starting point successfully delivered a class of high quality leads, 3-[4-(3-R¹-2-R-5H-imidazo[1,2-b][2]benzazepin-11-yl)-1-piperazinyl]- 2,2-dimethylpropanoic acids (e.g., 9, 19, 20, and 21), characterized by potent and balanced H₁/5-HT_2A receptor antagonist activities and good developability profiles.

Full text of DOI:10.1016/j.bmcl.2010.07.029

Bioorganic & Medicinal Chemistry Letters 20 (2010) 5069–5073
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel imidazobenzazepine derivatives as dual H1/5-HT2A antagonists
for the treatment of sleep disorders
a
a
a
a
Massimo Gianotti a, , Corrado Corti , Sonia Delle Fratte , Romano Di Fabio , Colin P. Leslie ,
*
Francesca Pavone a, Laura Piccoli a, Luigi Stasi a, , Mark J. Wigglesworth b
a Neurosciences CEDD, GlaxoSmithKline, Medicines Research Centre, Via A. Fleming 4, 37135 Verona, Italy
b GlaxoSmithKline Medicines Research Centre, NFSP, Harlow, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel imidazobenzazepine template (5a) with potent dual H1/5-HT2A antagonist activity was identified.
Application of a zwitterionic approach to this poorly selective and poorly developable starting point suc-
cessfully delivered a class of high quality leads, 3-[4-(3-R1-2-R-5H-imidazo[1,2-b][2]benzazepin-11-yl)-
1-piperazinyl]-2,2-dimethylpropanoic acids (e.g., 9, 19, 20, and 21), characterized by potent and balanced
H1/5-HT2A receptor antagonist activities and good developability profiles.
Received 15 June 2010
Revised 7 July 2010
Accepted 8 July 2010
Available online 11 July 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Histamine (H1) antagonists
5-Hydroxytryptamine 2A (5-HT2A)
antagonists
Zwitterions
Hypnotics
Sleep disorders
Histamine H1 and serotonin 5-HT2A receptors mediate two dif-
ferent mechanisms involved in sleep regulation. Brain histamine is
involved in the regulation of the sleep–wake cycle, arousal, cogni-
tion, and memory mainly through H1 receptors1 and H1 antago-
nists act as sleep inducers. On the other hand selective blockade
of 5-HT2A receptor has been proven in both preclinical2 and clini-
cal3 studies to be efficacious in reducing wake after sleep onset
(WASO), increasing slow wave sleep (SWS), and total sleep time
(TST) therefore providing consolidation of sleep. It can be hypoth-
esized that the combination of these two mechanisms may offer an
improved hypnotic profile with respect to marketed gold standard
benzodiazepine-like drugs.4
As part of a discovery program to identify dual H1/5-HT2A antag-
onists the novel imidazobenzazepine compound 5a (Fig. 1) was de-
signed and prepared in house. Upon characterization it was found
to be endowed with good H1/5-HT2A potency but poor selectivity,
in particular towards adrenergic targets. In designing a specific
optimization strategy for this compound we reasoned that the
introduction of an aminoacid moiety in the top region of the mol-
ecule may modulate the developability profile of the series
appropriately.
It was well known that zwitterion moieties were tolerated by
H1 receptor antagonists,5 as they have been originally introduced
to reduce the sedating properties of centrally acting first genera-
tion anti-histaminic, by reducing their brain penetrating capacities.
Furthermore, in the second generation H1 antagonists (e.g., Cetiri-
zine and Fexofenadine), the introduction of the zwitterion group
was responsible for the reduction of the drug–drug interactions
and enhanced receptor selectivity profiles with respect of first gen-
eration drugs.6
Despite the fact that zwitterions were initially introduced to
avoid brain penetration of the first generation anti-histamines, sci-
entists at Hypnion, now Lilly, recently demonstrated that zwitter-
ionic structures can penetrate the brain and induce hypnotic
activity. In fact they shaped their sleep programme strategy by
H
N
In-vitro pharmacology
N
h-fpKi H1a / 5-HT2Ab = 9.2/8.1
h-fpKia 5-HT2B/C = 6.5/7.7
N
N
h-fpKib α1A/B = 8.4/8.5
5a
* Corresponding author.
Figure 1. In vitro data of compound 5a. af-pKi, functional pKi obtained from the
FLIPR (fluorescent imaging plate reader) assay.11 bf-pKi, functional pKi obtained
from the aequorin assay (intracellular Ca luminescence).12
E-mail addresses: massimo.gianotti@aptuit.com, masgian@libero.it (M. Gianotti).
Present address: Rottapharm Madaus, R&D Division, Via Valosa di Sopra 9, 20052
Monza, Italy.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.07.029

5070
M. Gianotti et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5069–5073
R3
introducing zwitterion moieties to known H1/5-HT2A antagonist
scaffolds.7 This approach led Hypnion to quickly progress several
compounds into the clinical phase, one of which, HY-10275 (LY-
2624803), successfully achieved its Proof of Concept as a sleep
drug.8
Herein we report our results applying the zwitterionic approach
that led to the identification of several advanced leads with excel-
lent developability profiles.
A novel synthetic pathway was put in place for the synthesis of
the compounds reported in this paper; in particular the first two
steps that allow access to the key intermediate 4 (Scheme 1)
proved critical.
The first step is the crucial one, in particular for substituted imi-
dazoles 2c–g, in order to obtain the target cycloketone 4 with the
desired regiochemistry of substituents R and R1. Imidazole 2a
and symmetric dimethyl imidazole 2b, were reacted with 1 in
DMF, without the use of any additional base, to give the correspon-
dent alkylated imidazoles 3a and 3b.
R3
A
R
H
N
N
R2
R2
[ ]n
N
N
[ ]n
N
i
ii
4
N
iii
N
N
R
R1
8 - 22
R1
5: R2 = R3 = H, n = 1
6: R2 = R3 = H, n = 2
7: R2 = R3 = Me (cis), n = 1
Scheme 2. Reagents and conditions: (i) appropriate piperazine derivative (see
Table 1), TMSOTf, 130°C, 2 h; (ii) appropriate ester (A, see Tables 1), DCE,
NaBH(OAc)3, rt, 12 h; (iii) KOH or LiOH, MeOH/H2O; reflux, 3 h.
on the tricyclic portion of the compound, in particular on the imid-
azole ring, with the aim to mask and reduce its basicity.
All the newly prepared chemical entities were assayed for their
agonistic and antagonistic properties using two kinds of functional
assays: FLIPR11 (fluorescent imaging plate reader) assays, using
either adherent transfected Chinese hamster ovary (CHO) cells
stably expressing the recombinant human H1 receptor or adherent
SHSY5Y cells stably expressing recombinant human 5-HT2B or
5-HT2C receptors; luminescence12 (aequorin) assays using either
frozen Chinese hamster ovary (CHO) cells stably expressing the
human H1 receptor and aequorin apo-protein or frozen human
embryonic kidney (HEK) cells stably expressing the human 5-
HT2A, 5-HT2B, or 5-HT2C receptors and aequorin apo-protein.
Before going into a detailed discussion of the results obtained, a
few general observations need to be made. No agonistic activity
was observed for any of the amino acids considered; high selectiv-
ity versus 5-HT2B and 5-HT2C receptors proved elusive, however
this was not considered a criterion to preclude compound progres-
sion. While there are few evidences of slight increases in wakeful-
ness and motor activity after blockade of the 5-HT2B receptor,13 no
significant alteration in the percentage distribution of any sleep
Subsequent cyclization with LDA gave 4a and 4b, respectively,
exploiting the acidity of the hydrogen in the 2-position of the imid-
azole. Regarding the mono-substituted imidazoles, two approaches
were developed for the selective syntheses of the imidazole deriv-
atives 3. To obtain the 5-substituted benzyl imidazoles 3c, 3d, and
3g, at least as the major regioisomers, we employed the same base
free conditions as reported above, while for the 4-substituted ben-
zyl imidazoles 3e and 3f, we were able to obtain high levels of
selectivity by introducing a base to the reaction mixture. An optimi-
zation process was required for this reaction as the bases examined
initially, such as K2CO3 and NaH, allowed formation of the desired
4-substituted product with satisfactory selectivity but in low yield
as these bases promoted oligomerisation of the bromoester 1. This
problem was overcome by the use of a Verkade proazaphosphatra-
ne9 base to effect the alkylation. This nonionic reagent is a strong
base readily able to deprotonate the imidazole and hence favors
the 4-substituted products but it is a very weak Lewis acid and
therefore avoids the oligomerisation side reaction. The ring closure
step of compounds 3c–g, to the corresponding tricyclic ketones
4c–g, was carried out using LDA as described for 4a,b. The ketones
4 obtained were then treated with trimethylsilyl triflate (TMSOTf)
in the presence of the appropriate piperazine derivative to give
the intermediates 5, 6 and 710 (Scheme 2). The amino acid side
chain was introduced either by means of a reductive amination
reaction or, in the case of compounds 8 and 11, an aza-Michael
reaction of the intermediates 5a and 7a, respectively, with an acry-
late. Compound 12 was synthesized using methyl 2-(bromo-
methyl)-2-propenoate. Finally, hydrolysis of the esters furnished
the desired final zwitterionic compounds 8–22 (see Table 1).
The SAR exploration was principally focused on three elements:
first, substitution/homologation of the piperazine in order to study
the influence of the most basic center of the compounds; second,
modulation of the acidic moiety; third, introduction of substituents
stage, arousal and light SWS (SWS1), was observed with the inhi-
14
bition of the 5-HT2C
.
Therefore the presence of 5-HT2B and 5-
HT2C activity in our compounds may not represent a potential issue
for their hypnotic profile. With regards to selectivity versus adren-
ergic a1A and a1B receptors, which was a concern for our initial hit
5a (Fig. 1), for the zwitterionic compounds synthesized and tested
in this assay, we never observed any significant activity on these
two receptors.15
From the outset of our exploration we noticed that in terms of
activity versus the primary targets the zwitterionic modification
always led to a reduction of both H1 and 5-HT2A activities, but
while a suitable level of potency is generally maintained at H1, po-
tency at 5-HT2A is much more sensitive, thus since the beginning it
was clear that the challenge for the exploration was to find com-
pounds with balanced H1/5-HT2A activity. The first two zwitter-
ionic compounds in the table are testimony to this behavior.
Compound 8 shows good H1 activity while its 5-HT2A activity is
unsatisfactory for progression. In contrast, compound 9, character-
O
O
N
O
O
ized by additional geminal dimethyl groups
a to the acidic moiety
O
2
N
i or ii
iii
that have the dual function of increasing lipophilicity and ‘mask-
ing’ the acid, demonstrates good H1/5-HT2A activity and selectivity
versus 5-HT2B but not versus 5-HT2C. Due to its good overall
in vitro profile, compound 9 was progressed into a rat pharmacoki-
netics (PK) study,16 see Table 2. It showed moderate blood clear-
ance, moderate volume of distribution, a half-life of 0.7 h and
good oral bioavailability. Compound 9 was also characterized by
a high brain fraction unbound (>50%) and low brain penetration
(0.02). As a result of this low brain penetration there is a substan-
tial imbalance between the free drug concentrations in the brain
and in the blood (free blood concentrations ꢀ50-fold higher than
N
4
HN
+
N
R
5
N
4a-g
R
R1
R
1
3a-g
R1
Br
2a-g
a: R=H, R1=H
R1
e: R=Cl, R1=H
b: R=Me, R1=Me f: R=F, R1=H
c: R=Me, R1=H g: R=H, R1=OMe
d: R=H, R1=Cl
Scheme 1. Reagents and conditions: (i) (used for compounds 3a, 3b, 3c, 3d, and 3g)
DMF, room temperature, 8 h; (ii) (used for compounds 3e and 3f) THF, Verkade’s
Superbase, room temperature, 2 h; (iii) THF, LDA, from À78 °C to room temperature,
12 h.

M. Gianotti et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5069–5073
5071
Table 1
c
b
b
Functional activity (f-pKia) at the human H1,b 5-HT2A
,
5-HT2B
,
and 5-HT2C receptors, for compounds 5a and 8–22
R3
A
N
R2
N
[ ]n
N
N
R
R1
Compound
R
R1
R2
R3
n
A
f-pKia
b
c
b
b
H1
5-HT2A
8.1
5-HT2B
6.5
5-HT2C
5a
8
H
H
H
H
H
H
H
H
1
1
H
9.2
7.7
O
O
O
O
O
7.3c
6.5
7.1
<6.1c
<6.1c
OH
OH
OH
OH
OH
9
10
11
12
H
H
H
H
H
H
H
H
H
H
H
1
2
1
1
7.6
7.3
7.7
7.5c
5.8
7.3
<5.7
6.8
H
<5.3
7.4
<5.7
6.2
Me
H
Me
H
6.7
<6.1
6.3
HO
O
13
14
H
H
H
H
H
H
H
H
1
1
7.9c
7.6c
<6.1
<6.1
<6.1
<6.1
<6.1
<6.1
isomer 1
HO
O
isomer 2
HO
O
15
H
H
H
H
1
6.9
6.2
7.4
<5.7
isomer 1
O
HO
16
17
H
H
H
H
H
H
1
1
7.2
7.4
6.3
6.4
7.6
7.8
<5.7
7.2
isomer 2
OH
O
O
O
Me
Me
OH
OH
OH
OH
OH
18
19
20
21
22
Me
Cl
H
H
H
H
H
H
H
H
H
H
H
H
1
1
1
1
1
7.3
6.9
7.3
7.2
7.0
6.6
7.2
7.5
H
7.1c
7.4c
7.4c
6.9c
7.0c
7.6c
6.5c
6.8c
6.3c
6.9c
6.5c
6.3c
O
O
Cl
F
H
O
H
MeO
a
b
c
f-pKi, functional pKi obtained from the FLIPR and aequorin assays. SEM for H1, 5-HT2A, 5-HT2B, and 5-HT2C data sets is <0.2 with a minimum of two replicates.
FLIPR, fluorescent imaging plate reader.11
Aequorin assay (intracellular Ca luminescence).12

5072
M. Gianotti et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5069–5073
Table 2
Rat pharmacokinetic profile for compounds 9, 19, 20, and 21
Compound
Br/Bl fua (%)
In vivo PKb
t1/2 (h)
CLb (ml/min/kg)
Vss (L/kg)
Fpo (%)
Br:Bl
9
19
20
21
>50/>50
30/34
18/22
45/40
38
11
3
1.9
1.9
1.1
1.1
0.7
2.4
5.0
1.8
61
58
37
45
0.02
0.07
0.07
0.04
10
a
fu, fraction unbound in brain/blood rat.
b
In vivo data determined following 0.5 mg/kg iv and 1 mg/kg po administration in rat, for compounds 19, 20, and 21 and by 1 mg/kg iv and 3 mg/kg po administration in
rat for compound 9. Br:Bl measured as brain to blood AUC0–8h ratio following po dosing.
free brain concentrations) that will greatly reduce the possibility of
achieving an acceptable therapeutic index. Therefore, an additional
goal for the exploration was to modify the physicochemical prop-
erties of the newly synthesized derivative 9 in order to obtain a
compound with a more balanced profile in terms of the free con-
centration in blood and brain. This topic will be discussed in detail
below.
With regard to modification of the piperazine, two compounds
were prepared: the homopiperazine derivative 10, which main-
tained comparable H1 activity but completely lost 5-HT2A potency,
and the dimethylpiperazine derivative 11, that was characterized
by balanced potency at the target receptors and reasonable selec-
Additionally, all of them were characterized by high fraction un-
bound in brain and blood and a brain penetration comprised be-
tween 0.04 and 0.07.
To understand if the low brain penetration observed is under
the control of an efflux system, the interaction with P-gp (P-glyco-
protein) was investigated in vivo in rat with a pre-treatment with a
known inhibitor of efflux transporters (GF120918).17 Compounds
19, 20, and 21 were tested in this paradigm and while the fold
change in the Br:Bl ratio in the presence and in the absence of
the inhibitor was significant for 20 and 21, suggesting a possible
role of efflux processes, for compound 19 it was not, suggesting
no role of efflux processes.
tivity versus 5-HT2B
.
Compounds 19, 20, and 21 were specifically prepared to test the
hypothesis that reducing the basicity of the imidazole ring and
increasing lipophilicity would have a positive impact both on brain
penetration and on the half-life, via reduction of the volume of
distribution.
Recalling that compounds 19, 20, and 21 were prepared to test a
specific hypothesis, an analysis of the rat PK data demonstrates
that these modifications have indeed led to improvements in brain
penetration (Br:Bl = 0.04–0.07 vs 0.02 for 9), attenuating some-
what the imbalance between free concentrations in brain and
blood.
Furthermore, the volumes of distribution for two of the three
derivatives were lower than for the parent (Vss = 1.1 L/kg vs 1.9
for 9) as speculated, however this did not translate into a shorter
half-life due to greater metabolic stability (half-life = 1.8–5.0 h vs
0.7 h for 9).
Some efforts to move away from the 2,2-dimethylpropanoic
acid moiety were also made. A propenoic acid derivative 12, the
two diastereoisomers of cyclobutanecarboxylic acid 13, 14 and
the two diastereoisomers of cyclohexanecarboxylic acid 15, 16
were synthesized. In general they presented good H1 potency but
had poor activity at 5-HT2A; the only compound from this series
that showed interesting target activities was compound 12.
The first attempts to introduce substituents onto the imidazole
ring gave compounds 17 and 18. These methylated derivatives
were prepared with the objective of increasing lipophilicity thus
potentially improving brain penetration with respect to 9. In both
cases the target receptor potencies were suboptimal and they were
not progressed into PK studies. Attention then turned to substitu-
ents that could modulate more effectively the electronic nature,
and therefore the basicity, of the imidazole ring.
Introduction of the electron donating methoxy substituent in
compound 22 was not well accepted, leading to a slight drop in
activity at both target receptors. In contrast, electron withdrawing
substituents appear to be better tolerated with halogenated deriv-
atives of 9 maintaining similar levels of activity to the parent. Com-
pounds 19 and 20, carrying a chlorine in the 5- and the 4-position
of the imidazole, respectively, and compound 21, carrying a fluo-
rine in the 4-position, all demonstrated good H1/5-HT2A activity
In conclusion, a novel imidazobenzazepine template (5a) with
potent dual H1/5-HT2A antagonist activity was identified. Applica-
tion of a zwitterionic approach to this poorly selective and poorly
developable starting point successfully delivered a class of high
quality leads, 3-[4-(3-R1-2-R-5H-imidazo[1,2-b][2]benzazepin-11-
yl)-1-piperazinyl]-2,2-dimethylpropanoic acids (e.g., 9, 19, 20,
and 21), characterized by potent and balanced H1/5-HT2A receptor
antagonist activities and good developability profiles. Compound 9
was also progressed into a pharmacodynamic model for sleep
disorders,16,18 where it exhibited a robust effect (statistical signif-
icant effect in the TST was reached at 1.0 mg/kg), thus confirming a
possible application as an alternative therapy to currently available
hypnotic drugs with the potential for a reduced side-effect burden.
but little or no selectivity versus 5-HT2B and 5-HT2C
.
As remarked earlier, the parent compound 9 is characterized by
low brain penetration and a goal was set to modify the physico-
chemical properties with the objective of improving brain penetra-
tion. In particular, compounds 19, 20, and 21 were the result of a
tailored medicinal chemistry strategy focused on reducing the ba-
sicity of the imidizole moiety and increasing the lipophilicity of
compound 9 through substitution with electron withdrawing
groups; it was hypothesized that these modifications should lead
to a reduction of the volume of distribution, and consequently a
reduction in half-life, as well as reducing the imbalance between
free concentrations in brain and blood.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.07.029.
References and notes
Results obtained in in vivo PK studies with these halogenated
derivatives of 9 are detailed in Table 2. All three compounds
showed low clearance, a moderate volume of distribution, a half-
life comprised between 1.8 and 5 h and good oral bioavailability.
1. (a) Monti, J. M.; Monti, D. CNS Drugs 2000, 13, 87; (b) Tashiro, M.; Mochizuki,
H.; Iwabuchi, K.; Sakurada, Y.; Itoh, M.; Watanabe, T.; Yanai, K. Life Sci. 2002, 72,
409.

M. Gianotti et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5069–5073
5073
2. Popa, D.; Léna, C.; Fabre, V.; Prenat, C.; Gingrich, J.; Escourrou, P.; Hamon, M.;
Adrien, J. J. Neurosci. 2005, 25, 11231.
3. Viola, A. U.; Brandenberger, G.; Toussaint, M.; Bouhours, P.; Macher, J. P.;
Luthringer, R. Clin. Neurophysiol. 2002, 113, 429.
4. Teegarden, B.; Al Shamma, H.; Xiong, Y. Curr. Top. Med. Chem. 2008, 8, 969.
5. Fonquerna, S.; Miralpeix, M. Expert Opin. Ther. Patents 2006, 16, 109.
6. Bousquet, J.Aria Workshop Group J. Allergy Clin. Immunol. 2001, 108, S147.
7. For example: (a) Edgar, D. M.; Hangauer, D. G.; Leighton, H. J.; Mignot, E. J. M.;
White, J. F.; Solomon, M.; Shiosaki, K. Patent WO2005/103041, 2005; (b) Edgar,
D. M.; Hangauer, D. G.; White, J. F.; Solomon, M.; Shiosaki, K. Patent WO2006/
034414, 2006.
. In Leifert, W., Ed., 5th ed.; Humana Press: New York, 2009; p 181.
13. Kantor, S.; Jakus, R.; Balogh, B.; Benko, A.; Bagdy, G. Br. J. Pharmacol. 2004, 142,
1332.
14. Smith, M. I.; Piper, D. C.; Duxon, M. S.; Upton, N. Pharmacol., Biochem. Behav.
2002, 71, 599.
15. For example, for compounds 9, 19, 20, and 21 h-fpKia1A/B 6 6 (aequorin assay).
16. All work involving animals was carried out in accordance with European
directive 86/609/EEC governing animal welfare and protection, which is
acknowledged by Italian Legislative Decree No. 116, 27 January 1992, and
according to internal review performed by the GlaxoSmithKline Committee on
Animal Research & Ethics (CARE) and to the company Policy on the Care and
Use of Laboratory Animals.
8. Sullivan, S.; Guilleminault, C. Expert Opin. Emerging Drugs 2009, 14, 411.
9. Verkade, J. G.; Kisanga, P. B. Aldrichim. Acta 2004, 37, 3.
17. Breedveld, P.; Beijnen, J. H.; Schellens, J. H. M. Trends Pharmacol. Sci. 2006, 27,
17.
10. Campiani, G.; Butini, S.; Gemma, S.; Nacci, V.; Fattorusso, C.; Catalanotti, B.;
Giorgi, G.; Cagnotto, A.; Goegan, M.; Mennini, T.; Minetti, P.; Di Cesare, M. A.;
Mastroianni, D.; Scafetta, N.; Galletti, B.; Stasi, M. A.; Castorina, M.; Pacifici, L.;
Ghirardi, O.; Tinti, O.; Carminati, P. J. Med. Chem. 2002, 45, 344.
11. Wigglesworth, M. J.; Lawless, K. J.; Standing, D. J.; Mackenzie, E. K.; Kitchen, V.
R.; Mckay, F.; Ward, E.; Brough, S. J.; Stylianou, M.; Jewitt, F. R.; Mclaren-
Douglas, A. M.; Jowet, M. I.; Tamayama, N.; Finnigan, D.; Ding, J.; Wise, A. J.
Biomol. Screen. 2008, 13, 354.
18. The effects of
9
were assessed using telemetric recording of
electroencephalogram (EEG) and electromyogram (EMG) in the rat.
Simultaneous recording of EEG and EMG allows the accurate derivation of
the sleep parameters. The test compound effects were assessed for 5 h, in the
active phase, starting at circadian time [CT] 18 (light off at circadian time [CT]
12) of the rat light–dark cycle. CT18 was chosen to allow a maximal window to
assess the hypnotic effects of the compound.
12. Brough, S. J.; Shah, P. Use of Aequorin for G-Protein Coupled Receptor Hit
Identification and Compound Profiling. In The Methods in Molecular Biology S52,