MRK-409 (MK-0343), a GABAA receptor subtype-selective partial agonist, is a non-sedating anxiolytic in preclinical species but causes sedation in humans

Article abstract of DOI:10.1177/0269881109354927

MRK-409 binds to α1-, α2-, α3- and α5-containing human recombinant GABA_A receptors with comparable high affinity (0.21-0.40 nM). However, MRK-409 has greater agonist efficacy at the α3 compared with α1 subtypes (respective efficacies relative to the full agonist chlordiazepoxide of 0.45 and 0.18). This compound readily penetrates the brain in rats and occupies the benzodiazepine site of GABA_A receptors, measured using an in vivo [³H]flumazenil binding assay, with an Occ₅₀ of 2.2 mg/kg p.o. and a corresponding plasma EC ₅₀ of 115 ng/mL. Behaviourally, the α3-preferring agonist efficacy profile of MRK-409 produced anxiolytic-like activity in rodent and primate unconditioned and conditioned models of anxiety with minimum effective doses corresponding to occupancies, depending on the particular model, ranging from ～35% to 65% yet there were minimal overt signs of sedation at occupancies greater than 90%. In humans, however, safety and tolerability studies showed that there was pronounced sedation at a dose of 2 mg, resulting in a maximal tolerated dose of 1 mg. This 2 mg dose corresponded to a C _max plasma concentration of 28 ng/mL, which, based on the rodent plasma EC₅₀ for occupancy of 115 ng/mL, suggested that sedation in humans occurs at low levels of occupancy. This was confirmed in human positron emission tomography studies, in which [¹¹C]flumazenil uptake following a single dose of 1 mg MRK-409 was comparable to that of placebo, indicating that occupancy of GABA_A receptor benzodiazepine binding sites by MRK-409 was below the limits of detection (i.e. <10%). Taken together, these data show that MRK-409 causes sedation in humans at a dose (2 mg) corresponding to levels of occupancy considerably less than those predicted from rodent models to be required for anxiolytic efficacy (～35-65%). Thus, the preclinical non-sedating anxiolytic profile of MRK-409 did not translate into humans and further development of this compound was halted. The Author(s) 2009.

Full text of DOI:10.1177/0269881109354927

Original Paper
MRK-409 (MK-0343), a GABA receptor
A
subtype-selective partial agonist, is a
non-sedating anxiolytic in preclinical
species but causes sedation in humans
Journal of Psychopharmacology
5(3) 314–328
The Author(s) 2009
2
!
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0269881109354927
jop.sagepub.com
1
1
1
1
1
JR Atack , KA Wafford , LJ Street , GR Dawson , S Tye ,
2
2
3
4
K Van Laere , G Bormans , SM Sanabria-Boh o´ rquez , I De Lepeleire ,
JN de Hoon , A Van Hecken , HD Burns , RM McKernan , MG Murphy
and RJ Hargreaves
5
5
3
1
6
3
Abstract
MRK-409 binds to a1-, a2-, a3- and a5-containing human recombinant GABA
MRK-409 has greater agonist efficacy at the a3 compared with a1 subtypes (respective efficacies relative to the full agonist chlordiazepoxide of 0.45
and 0.18). This compound readily penetrates the brain in rats and occupies the benzodiazepine site of GABA receptors, measured using an in vivo
H]flumazenil binding assay, with an Occ50 of 2.2 mg/kg p.o. and a corresponding plasma EC50 of 115 ng/mL. Behaviourally, the a3-preferring agonist
A
receptors with comparable high affinity (0.21–0.40 nM). However,
A
3
[
efficacy profile of MRK-409 produced anxiolytic-like activity in rodent and primate unconditioned and conditioned models of anxiety with minimum
effective doses corresponding to occupancies, depending on the particular model, ranging from ꢀ35% to 65% yet there were minimal overt signs of
sedation at occupancies greater than 90%. In humans, however, safety and tolerability studies showed that there was pronounced sedation at a dose of
2
mg, resulting in a maximal tolerated dose of 1 mg. This 2 mg dose corresponded to a Cmax plasma concentration of 28 ng/mL, which, based on the
rodent plasma EC50 for occupancy of 115 ng/mL, suggested that sedation in humans occurs at low levels of occupancy. This was confirmed in human
11
positron emission tomography studies, in which [ C]flumazenil uptake following a single dose of 1 mg MRK-409 was comparable to that of placebo,
indicating that occupancy of GABA receptor benzodiazepine binding sites by MRK-409 was below the limits of detection (i.e. <10%). Taken together,
A
these data show that MRK-409 causes sedation in humans at a dose (2 mg) corresponding to levels of occupancy considerably less than those predicted
from rodent models to be required for anxiolytic efficacy (ꢀ35–65%). Thus, the preclinical non-sedating anxiolytic profile of MRK-409 did not translate
into humans and further development of this compound was halted.
Keywords
A
benzodiazepine, GABA receptor, non-sedating anxiolytic, sedation, subtype-selective
The fact that the pharmacological effects of benzodiaze-
pines are mediated by four different GABA subtypes,
Introduction
A
The many pharmacological actions of benzodiazepines, namely those receptors containing either an a1, a2, a3 or
including sedation, anxiolysis, myorelaxation and the anticon- a5 subunit, has led to the search for compounds which
vulsant and cognition impairing activities, are mediated via
interactions with a specific recognition site on GABA
tors. These receptors are composed of various combinations of
the GABA gene family (a1-6, b1-3, g1-3, d, e, y, p; Simon
et al., 2004). More specifically, those that contain a benzodia-
zepine binding site comprise a b, g2 and either a1, a2, a3 and
a5 subunits (Sieghart, 2006) arranged in a pentamer with a
a:b:g subunit stoichiometry of 2:2:1 (Sieghart and Sperk,
A
recep-
1
Neuroscience Research Centre, Merck Sharp
Laboratories, Harlow, UK.
& Dohme Research
A
2
Department of Nuclear Medicine and Radiopharmacy, Katholieke
Universiteit Leuven, Leuven, Belgium.
3
Imaging Research and Pharmacology, Merck Research Laboratories, West
Point, PA, USA.
4
Clinical Pharmacology, Merck Sharp & Dohme Research Laboratories,
Brussels, Belgium.
2
GABAA receptors contain benzodiazepine binding
002). Around three-quarters of the total number of brain
a
5
Department of Clinical Pharmacology, Katholieke Universiteit Leuven,
Leuven, Belgium.
site (McKernan and Whiting, 1996), the location of which is
at the interface of the g2 and a subunits (Sieghart, 2006). The
pharmacology of this site is dictated to a large extent by the a
subunit (Sieghart, 2006); an observation best illustrated by the
fact that receptors containing either an a4 or a6 subunit do not
bind diazepam, a property that can be solely attributed to a
single amino acid (Wieland et al., 1992).
6
Clinical Pharmacology, Merck Research Laboratories, Bluebell, PA, USA.
Corresponding author:
Dr John R Atack, Neuroscience, Johnson & Johnson Pharmaceutical
Research and Development, Turnhoutseweg 30, B-2340 Beerse, Belgium
Email: JAtack1@its.jnj.com
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

Atack et al.
315
F
F
F
N N
N N
N N
O
NH
2
N
N
N
N
N
N
OH
Cl
Cl
HN
H N–NH
2
F
2
F
N
N
N
N
N
N
O
N
O
O
N
N
N
N
N
N
Cl
1
Cl
2
Cl
3
N
N
MRK-409
TPA-023
L-838417
F
Cl
Figure 1. Comparison of the structures of 7-cyclobutyl-6-(2-methyl-2H-
,2,4-triazol-3-ylmethoxy)-3-(2,6-difluorophenyl)-1,2,4-triazolo[4,3-b]
O
1
F
pyridazine (MRK-409, also known as MK-0343 (de Haas et al., 2008)),
TPA023 (Atack et al., 2006) and L-838417 (McKernan et al., 2000).
F
N
N
selectively interact with only certain of these subtypes and
might therefore possess novel pharmacological profiles
N
N
F
(Atack, 2005; Maubach, 2003). These studies have been facili-
Cl
4
tated by the use of transgenic mouse models (Rudolph and
Mohler, 2004) that have helped ascribe particular pharmaco-
HO
N
A
logical properties of benzodiazepines to certain GABA sub-
types. As a result, compounds which selectively interact with
the a2/a3 subtypes have been reported to be non-sedating
anxiolytics in preclinical species (Atack et al., 2006;
McKernan et al., 2000; Mirza et al., 2008) whilst a5 selective
compounds have been shown to enhance cognition in rodent
models (Ballard et al., 2009; Chambers et al., 2004; Dawson
et al., 2006). It should be emphasized that to date there is no
data to suggest that these novel preclinical profiles translate
into clinical efficacy and that the predictive value of the
preclinical assays of such novel compounds remains to be
established.
N
NH
N
N
N
N
N
N
5
6
7
F
N
N
N
N
F
O
N
N
N
In the present study, we describe the properties of a com-
pound, 7-cyclobutyl-6-(2-methyl-2H-1,2,4-triazol-3-ylmethoxy)-
MRK-409
3-(2,6-difluorophenyl)-1,2,4-triazolo[4,3-b]pyridazine(MRK-409,
also known as MK-0343; de Haas et al., 2008), which is struc-
turally related to the previously-described compounds TPA023
and L-838417 (Figure 1 and Atack et al., 2006; McKernan et al.,
Figure 2. Scheme for the synthesis of gram quantities of MRK-409.
Initial, smaller scale synthesis was carried out using methods described
previously for other triazolopyridazines in this series (Carling et al.,
2
000). More specifically, the anxiolytic-like and sedative proper-
ties of this compound were evaluated in preclinical species
rodents and non-human primates) following which the safety,
2
3
005). Briefly, the reaction of cyclobutane carboxylic acid with
,6-dichloropyridazine (1) produced 2 which was then reacted with
(
hydrazine to produce the hydrazinopyridazine 3, the latter of which was
reacted with 2,6-difluorobenzoyl chloride to produce the triazolopyrida-
zine 4. The 1,2,4-triazole 5 was first N-methylated (6) and then hydro-
xymethylated (7) before being coupled to the triazolopyridazine 4 to
produce MRK-409.
tolerability, pharmacokinetics and receptor occupancy of
MRK-409 were assessed in healthy young male volunteers.
Materials and methods
All animal procedures were performed in accordance with the (7-cyclobutyl-6-(2-methyl-2H-1,2,4-triazol-3-ylmethoxy)-3-
UK Animals (Scientific Procedures) Act, 1986 and associated (2,6-difluorophenyl)-1,2,4-triazolo[4,3-b]pyridazine; MRK-
guidelines.
409). Diazepam and chlordiazepoxide were purchased from
Sigma-Aldrich (Gillingham, UK) and bretazenil was synthe-
sized at Merck Research Laboratories (Harlow, UK).
Drugs
3
3
3
[
H]Flumazenil ([ H]Ro 15-1788) and [ H]Ro 15-4513 were
MRK-409 was synthesized on the milligram scale using meth- purchased from PerkinElmer Life and Analytical Sciences
ods analogous to those described elsewhere (Carling et al., (Boston, MA, USA).
2
005). The gram-scale synthesis was performed as summarized
in Figure 2. Briefly, 2,6-difluorobenzoyl chloride was reacted
with 3-chloro-4-cyclobutyl-6-hydrazinopyridazine (3) to pro-
duce 6-chloro-7-cyclobutyl-3-(2,6-difluorophenyl)-pyridazine
In vitro affinity and efficacy
(
(
4). This latter product was coupled with (2-methyl- The affinity of MRK-409 for the benzodiazepine binding site
1,2,4)triazol-3-yl)methanol to produce the final compound of GABA receptors was measured in mouse fibroblast
A
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

3
16
Journal of Psychopharmacology 25(3)
A
recep- vivo binding of [ H]flumazenil, separate groups of rats or
ꢁ
3
L(tk ) cells expressing human recombinant GABA
tors containing b3, g2 plus either a1, a2, a3, a4, a5 or a6 mice were given a dose of bretazenil (5 mg/kg i.p. in 100%
3
subunits using [ H]flumazenil for a1-, a2-, a3- and PEG 300) that occupies essentially all rat or mouse brain
3
a5-containing receptors or [ H]Ro 15-4513 for the a4- and benzodiazepine binding sites. In rat studies, trunk blood
a6-containing subtypes: (Hadingham et al., 1993, 1996; was collected into heparinized tubes immediately after decap-
Wafford et al., 1996). The affinity of MRK-409 for native itation. The plasma was separated by centrifugation, removed
ꢂ
rat cerebellum and spinal cord P2 membranes was measured and stored at ꢁ20 C for subsequent measurement of drug
3
using [ H]flumazenil as described elsewhere (Atack et al., concentrations using high performance liquid chromatogra-
2
006).
phy (HPLC) (Kromasil KR100 C18 150 mm ꢃ 3.2 mm
The same cell lines were used to measure the intrinsic column with a 2:3 ratio 25 mM ammonium formate pH
efficacy of MRK-409 using whole cell patch clamp electro- 3.0:acetonitrile mobile phase) and tandem mass spectrometry
physiology (Atack et al., 2006; Brown et al., 2002). Having detection (mass transition 398.2 to 370.1 a.m.u.).
separately established the concentration of GABA that pro-
The occupancy was defined as the extent to which the
3
duced a maximal response, the intrinsic efficacy of MRK-409 in vivo binding of [ H]flumazenil was reduced by prior treat-
was defined as the ability of the compound (applied for 30 sec ment with drug. Occupancy data from the different experi-
prior to a 5 sec application of a GABA) to modulate the ments was plotted as a function of either dose or plasma drug
currents produced by a GABA concentration equivalent to concentration and from these graphs the dose and plasma
an EC20. A concentration of the non-selective benzodiazepine drug concentrations corresponding to 50% occupancy
agonist chlordiazepoxide which produced a maximal modu- (Occ50 and EC50
, respectively) were calculated using
lation (3 mM) was used as an internal standard and gave an curve-fitting (GraphPad Prism).
increase in the GABA EC20-equivalent current amplitude in
the region of 75–150%. Concentration-effect curves were con-
structed for each cell with curve-fitting being performed using
a
Rat anxiolysis assays
non-linear least squares method (GraphPad Prism,
GraphPad Software Inc., San Diego, CA, USA). From In order to establish the anxiolytic properties of MRK-409,
these analyses, the maximum efficacy and an EC50 were deter- three different assays were employed, one using an ethological
mined for each cell (Atack et al., 2006; Brown et al., 2002). In measure of anxiety (the elevated plus maze) and two using
addition, the maximum efficacy was also expressed relative to conditioned models of anxiety (fear-potentiated startle and
that observed with chlordiazepoxide (3 mM).
conditioned suppression of drinking). An abbreviated
description of these methods is outlined below, although
more detailed methodology is presented elsewhere (Atack
et al., 2006).
MRK-409 GABA receptor occupancy
A
3
The inhibition of the in vivo binding of [ H]flumazenil by
MRK-409 was used to measure the occupancy of rat Elevated plus maze: Male Sprague-Dawley rats (approxi-
male 260–300g Sprague-Dawley; B&K Universal, Hull, mately 250–300g, n ¼ 17–18/group; B&K Universal, Hull,
UK) or mouse (male 23–31g Swiss-Webster; B&K UK) were given p.o. doses of either vehicle (0.5% methyl
Universal) brain GABA
(
A
receptor benzodiazepine binding cellulose; dose volume ¼ 1 mL/kg), and either in an initial
sites (Atack et al., 2006). Rat occupancy studies were per- experiment, 0.3, 1 or 3 mg/kg, or a subsequent experiment,
formed following p.o. dosing of MRK-409 (1 mL/kg in 1, 2 or 3 mg/kg MRK-409. Additional animals were given the
0
.5% methyl cellulose vehicle) in either separate in vivo bind- benzodiazepine full agonist chlordiazepoxide as a positive
ing studies (3, 10 and 30 mg/kg dose–response curve or control (5 mg/kg i.p. in isotonic saline). Half an hour later
mg/kg 1 h vs. 6 h comparison) or in conjunction with ele- rats were given a 5 min trial on the elevated plus maze
vated plus maze experiments (0.3, 1 and 3 mg/kg and 1, 2 and (Dawson and Tricklebank, 1995) with their movements
mg/kg studies). Mouse brain occupancy was determined being monitored by a video tracking system. The primary
immediately after rotarod experiments using oral doses of 1, measure of anxiolytic-like activity was the percent time
or 10 mg/kg MRK-409 (10 mL/kg in 0.5% methyl cellulose). spent on the open arms of the maze, with increases in this
3
3
3
Occupancy was also measured in animals used as posi- parameter being used as an index of anxiolytic activity. At
tive controls in the rat elevated plus maze (chlordiazepoxide, completion of the trial, alternate animals were taken and
5
1
mg/kg i.p.) and mouse rotarod (diazepam,
0 mg/kg p.o.).
Animals pretreated with either vehicle or drug were
3
or occupancy measured as described above.
3
given an intravenous (i.v.) injection of [ H]flumazenil Fear-potentiated startle: Male Sprague–Dawley rats
70–88 Ci/mmol diluted 1:150 with saline; 5 mL/g for mice, (260–310 g, n ¼ 15/group; B&K Universal) were trained to
mL/g for rats) via a tail vein and 3 min later were killed by associate the presentation of a light with a mild foot-shock.
(
1
stunning and decapitation. Brains were rapidly removed, Following this training period, rats were dosed (p.o.) with
homogenized and 300 mL aliquots filtered and washed over either vehicle (0.5% methyl cellulose, 1 mL/kg) or
Whatman GF/B filters. Washed filters were placed in scintil- MRK-409 (0.3, 1 or 3 mg/kg) and 0.5 h later placed in the
lation vials, scintillation fluid added and radioactivity startle apparatus (San Diego Instruments, San Diego, CA,
counted. In order to establish the extent of non-specific in USA). Rats were allowed to acclimatize for 5 min following
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

Atack et al.
317
which they received 10 ꢃ 100 dB tones (50 msec duration,
Primate anxiolysis and sedation assays
3
0 sec inter-trial interval) to partially habituate them to the
startle stimulus. During the test session itself, rats were pre- Conditioned emotional response: Twelve adult male
sented with random noise stimuli (10 ꢃ 100 dB, half given squirrel monkeys (Saimiri sciureus, 0.7–1.2 kg) were trained
3200 msec after the presentation of the conditioning, light to press a lever to obtain a fruit juice reward. In subsequent
stimulus and half given in darkness). All responses were mea- sessions, a red cue light was illuminated for 60 sec and during
sured over a 100 msec window beginning immediately upon the last 0.5 sec there was a 10% chance of a mild electric
presentation of the tone and the mean startle amplitude for tail-shock (1–7 mA) being delivered. The association between
each stimulus type was calculated. For each animal, the the red light and a possible tail-shock produced a marked
startle amplitude in the light minus the startle amplitude in reduction in lever pressing (Atack et al., 2006).
the dark (the ‘difference score’) was also calculated.
On drug-testing days, animals were given either vehicle
0.5% methyl cellulose, 2 mL/kg p.o.) or MRK-409 (0.03
and 0.1 mg/kg p.o. in Experiment 1 or 0.3, 1 or 3 mg/kg
(
Conditioned suppression of drinking: Water-deprived p.o. in Experiment 2) 0.5 h prior to testing in a pseudo-
male hooded-Lister rats (200–250 g; B&K Universal) were Latin square design in which every animal received vehicle
trained to lick a metal spout for a reward of 0.1 mL water and each dose of MRK-409. On test days, the schedule was
(
Dawson et al., 1994). Following this training period, a as above except that tail-shocks were not delivered when the
house light was illuminated for 60 sec and in the last second cue light was illuminated. The suppression of lever pressing
0.4-mA foot-shock was delivered; after only three produced by the cue light (ratio of responding during and
a
light-shock pairings, the licking rates (as measured by the prior to illumination of the red light) was measured with a
ratio of the lick rates during compared with prior to presen- anxiolytic-like activity being defined as a significant increase
tation of the light) were markedly suppressed. Rats that had in the suppression ratio relative to vehicle-treated animals.
learned the light–shock association (as judged by a suppres-
sion ratio of ꢄ0.25) were dosed with either vehicle
(
0.5% methyl cellulose, 1 mL/kg p.o.) or one of three Squirrel monkey lever pressing: This assay is analogous to
doses of MRK-409 (1, 3, 10 mg/kg p.o.) 0.5 h prior to the the rat chain pulling assay in so far as behavioural disruption,
test session (n ¼ 11–12/group). A significant increase in for example sedation or myorelaxation, is assessed by the
the suppression ratio was used as an indicator of ability of a compound to reduce the rate of responding to
anxiolytic-like activity.
obtain a fruit juice reward. Nine adult male squirrel monkeys
S. sciureus, 0.7–1.2 kg) were trained to press a lever to obtain
(
a fruit juice reward. On the test day, animals were dosed with
vehicle (0.5% methyl cellulose, 2 mL/kg p.o.) or MRK-409
Rodent sedation assays
Rat chain-pulling assay: Male PVG rats (270–350g; B&K (1, 3 or 10 mg/kg p.o.) in a pseudo-Latin square design with
Universal) were trained to pull a chain in order to receive a every animal receiving, on separate days, vehicle and each
food pellet reward. On the testing day, rats (n ¼ 12/group) dose of MRK-409. The rate of lever pressing measured on
were dosed p.o. with either vehicle (0.5% of methyl cellulose, the test day was expressed as a percentage of that observed on
1
mL/kg), MRK-409 (10 or 30 mg/kg) or diazepam (10 mg/kg) the preceding day.
and then immediately placed in the testing chamber with
chain-pulling activity being measured over the subsequent
Human studies
1-h period.
The imaging and safety and tolerability studies were per-
formed according to protocols approved by the relevant
Mouse rotarod assay: Male BKTO mice (26–30 g, n ¼ 7–8/ Institutional Review Board and with appropriate signed
group; B&K Universal) were trained to successfully complete informed consent forms.
a 2 min trial on a rotarod (Ugo basile, Comeno, Italy) revolv-
ing at a speed of 15 rpm. Mice were given either vehicle (0.5%
methyl cellulose, 10 mL/kg p.o.), MRK-409 (1, 3 or 10 mg/kg Positron emission tomography: PET studies were per-
p.o.) or diazepam (10 mg/kg p.o.) 0.5 h before being placed on formed at the Katholieke Universiteit Leuven, Belgium
the rotarod. The latency to fall from the rotarod was recorded using an HR+ PET camera (Siemens, Germany) as described
or if the mouse successfully completed the 2-min trial, the elsewhere (Van Laere et al., 2008). In brief, prior to the first
latency was recorded as 120 sec. Immediately following scan, a radial artery cannula was inserted under local anaes-
the rotarod trial, mice were given an i.v. injection of thesia for blood sampling and a venous cannula was inserted
3
[
measured as described above.
H]flumazenil via a tail vein and receptor occupancy was for tracer injection. For each scan a bolus of approximately
1
7.5 mCi of [ C]flumazenil was injected and scans consisted of
1
In order to assess the possible interaction of MRK-409 20 frames with a progressive increase in frame duration
with ethanol, the experiment described above was repeated (4 ꢃ 0.25, 4 ꢃ 1, 2 ꢃ 2.5, 10 ꢃ 5 min) to give a total duration
in mice given a subthreshold dose of ethanol (1.5 g/kg i.p.) of 1 h. A metabolite-corrected arterial input function for the
0
.5 h prior to initiation of the rotarod trial. In other words, tracer was generated from blood sampling during the scan by
1
1
ethanol was given immediately prior to p.o. doses of either measuring the plasma concentration of [ C]flumazenil and its
vehicle, MRK-409 or diazepam. labelled metabolites. Two 1-h scans were performed, one
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

3
18
Journal of Psychopharmacology 25(3)
ꢁ
commencing 1.5 to 0.5 h before and the second 1.5 to 2.5 h receptors expressed in mouse fibroblast L(tk ) cells by 20, 36,
after dosing with either placebo or 1 mg MRK-409.
74 and 26%, respectively (Figure 3, Table 2). When expressed
Separately, standard T1 weighted volumetric MRI scans in comparison with the non-selective agonist chlordiazepox-
were obtained for each subject and these were used for the ide, the respective relative efficacy values of MRK-409 were
manual anatomical delineation of the occipital cortex, frontal 0.18, 0.23, 0.45 and 0.18 at the a1, a2, a3 and a5 subtypes
cortex, cerebellum and pons as regions of interest (ROIs). The (inset, Figure 3). Hence, the maximal efficacy of MRK-409
1
1
[
C]flumazenil PET images obtained in the second scans were was higher at the a3-subtype than at other GABAA subtypes.
aligned with the first study in order to define the same ROIs A similar profile (i.e. a1 efficacy < a2 and/or a3) was also
1
1
and [ C]flumazenil time–activity curves (TACs) were gener- observed in human GABA
A
receptors transiently-expressed
1
1
ated for each ROI. These [ C]flumazenil TACs were used in Xenopus laevis oocytes, rat a1 and a3 subtypes transi-
together with the metabolite-corrected plasma curve and a ently-expressed in human embryonic kidney cells as well as
single-tissue compartment model to estimate the rate constant in L(tk ) stably-expressing the human subtypes but using a
ꢁ
3
36
ꢁ
for forward capillary exchange (K , expressed in mL/min/cm
[ Cl ] flux rather than electrophysiological assay (data not
) and shown).
is proportional to the ratio of In comparison, the prototypic non-selective benzodiaze-
benzodiazepine binding sites (Salmi et al., 2008) and the equi- pine diazepam has an affinity at the different GABAA recep-
librium dissociation constant, Bmax/K and can be estimated tors ranging from 11 to 20 nM (Atack et al., 1999). Moreover,
independent of any flow changes, which will be reflected in at each subtype diazepam has an efficacy comparable to
changes of K . Comparison of the V values obtained from chlordiazepoxide (a potentiation of a GABA EC20 in the
1
tissue), the total volume of distribution of flumazenil (V
the cerebral blood volume. V
T
T
d
1
T
1
1
the [ C]flumazenil scans provides information on receptor region of 100–150%; Dawson et al., 2006) such that at each
occupancy at the studied times.
Safety, tolerability and pharmacokinetics: An escalating
single oral dose study was performed in healthy young (18–45
1
00
0.45
0
0
0
0
.5
.4
.3
.2
years old) male volunteers using doses of 0.05, 0.1, 0.25, 0.5,
1
0.23
8
0
0.18
0
.18
.0, 1.5 and 2.0 mg MRK-409. Tolerability was assessed by
α3
careful questioning for adverse events, ECGs, monitoring of
vital signs and laboratory safety. Blood samples were
removed 0.5, 1, 2, 4, 6, 8, 12, 16 and 24 h after dosing and
plasma drug concentrations were measured.
0.1
0.0
60
α1
α2
α3
α5
4
0
0
0
α2
α5
α1
2
Results
In vitro affinity and efficacy of MRK-409
–10
–9
–8
–7
–6
MRK-409 has equivalent high affinity for the benzodiazepine
site of human recombinant GABA
A
receptors containing either
(MRK–409) (M)
an a1, a2, a3 or a5 subunit (Ki values ranging from 0.21 to
0
9
.40 nM) but its affinity for the a4 and a6 subtypes (78 and
80 nM, respectively) was much lower (Table 1). The affinities
Figure 3. Efficacy of MRK-409 at human recombinant GABA
A
receptors.
Potentiation of GABA EC20-equivalent currents in human recombinant
measured in native rat cerebellum and spinal cord receptors
0.28 and 0.27 nM, respectively; Table 1) were similar to those
GABA
A
receptors containing b3, g2 plus either a1, a2, a3 or a5 subunits
(
ꢁ
stably expressed in mouse L(tk ) cells was measured using whole-cell
patch clamp electrophysiology. Inset shows maximal potentiation at each
cell type expressed relative to the full agonist chlordiazepoxide. Data
shown is the mean ꢅ SEM (n ¼ 5–6).
seen in human recombinant receptors (0.21–0.40 nM).
Whole cell patch clamp electrophysiology showed that
MRK-409 potentiated the GABA EC -induced currents in
2
0
human recombinant a1-, a2-, a3- and a5-containing GABA
A
Table 1. Affinity of MRK-409 for the benzodiazepine site of recombinant human and native rat brain
GABA receptors
A
Ki, nM
Human recombinant GABA
A
receptors containing b3, g2 plus
Native rat brain receptors
a1 a2
a3
a4
a5
a6
Cerebellum
Spinal cord
0.22 ꢅ 0.02 0.40 ꢅ 0.07 0.21 ꢅ 0.03 78 ꢅ 12 0.23 ꢅ 0.04 980 ꢅ 160 0.28 ꢅ 0.06
0.27 ꢅ 0.04
Data shown is the mean ꢅ SEM (n ¼ 6–8 separate determinations).
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

Atack et al.
319
of the a1, a2, a3 and a5 subtypes its efficacy relative to chlor- at others. For example, a dose of 1 mg/kg gave occupancy
diazepoxide is ꢀ1.0.
of 11% and 31% in separate studies whereas at a dose of
3
7
mg/kg, data was much less variable (occupancy of 52, 62,
4 and 77%). Nevertheless, these data show that between
In vivo binding
doses of 0.3 and 30 mg/kg, and at 0.75–1 h post-dose, occu-
Following oral dosing, MRK-409 was rapidly absorbed, pancy was dose-dependent with an Occ50 of 2.2 mg/kg and a
with occupancy 1 h after a 3 mg/kg dose being 77 ꢅ 6% Hill slope of 1.65 (Figure 4B). Based upon these values, the
(
Figure 4A). However, by 6 h post-dose, occupancy fell to percent occupancy at doses ranging between 0.3 and 30 mg/kg
3
6 ꢅ 5%. Plasma drug concentrations were measured in ter- was calculated (Table 3).
minal samples collected from these same animals and showed
Plasma samples were collected from the various rat occu-
that the drop in occupancy from 77% at 1 h post-dose to 36% pancy studies and plasma drug concentrations measured. This
at 6 h was associated with a decrease in plasma drug concen- allowed occupancy to be plotted as a function of plasma
trations from 241 ꢅ 35 ng/mL to 47 ꢅ 9 ng/mL. This data is MRK-409 drug concentrations (Figure 5). This data was
consistent with more detailed pharmacokinetic studies in fitted to a single-site model with a Hill slope of 1.11 and the
which the time at which maximum plasma drug concentra- mean plasma level required to occupy 50% of brain receptors
tions were achieved following oral dosing was, depending was 115 ng/mL (equivalent to a plasma concentration of
upon dose, between 0.5 and 1 h (data not shown).
In order to calculate the Occ50 for MRK-409 at 0.75–1 h
after dosing, data from a number of separate studies
290 nM).
MRK-409 has anxiolytic-like activity in rats
(
the 1 and 6 h time course, the two elevated plus maze experi-
ments and a 3, 10, 30 mg/kg dose–response curve) were
combined (Figure 4B). There was degree of
experiment-to-experiment variability at some doses but not
Elevated plus maze: Two separate elevated plus maze
experiments were performed, the first using doses of 0.3, 1
and 3 mg/kg p.o. (Experiment 1) and the second 1, 2 and
a
3
mg/kg (Experiment 2), with Figure 6 showing the time
spent on the open arms of the maze expressed as a percent
of the total time (5 min). In both Experiment 1 and
A
Table 2. Efficacy of MRK-409 at human recombinant GABA receptors
containing different a subunits stably expressed in mouse fibroblast
ꢁ
L(tk ) cells
Human recombinant GABA
A
receptors containing b3, g2 plus
a2 a3 a5
20 ꢅ 2% 36 ꢅ 5% 74 ꢅ 5% 26 ꢅ 2%
Table 3. Occupancy of rat brain benzodiazepine
binding sites estimated for various p.o. doses of
MRK-409
a1
Maximum
modulation
Dose (mg/kg, p.o.)
Estimated % occupancy
Efficacy relative 0.18 ꢅ 0.02 0.23 ꢅ 0.03 0.45 ꢅ 0.04 0.18 ꢅ 0.02
0.3
1
4
to CDP
22
65
93
99
EC50 (nM)
3.2 ꢅ 0.6
3.1 ꢅ 0.3
3.4 ꢅ 0.5
3.0 ꢅ 0.6
3
1
3
0
0
CDP ¼chlordiazepoxide.
Values shown are mean ꢅ SEM (n ¼ 5–6).
(
A)
(B)
1
00
100
80
60
40
20
0
8
6
4
2
0
0
0
0
0
7
7 ± 6%
36 ± 5%
2
41 ± 35 ng/mL
47 ± 9 ng/mL
1
hr
6 hr
0
.3
1
3
10
30
Time after 3 mg/kg p.o. MRK-409
MRK-409 (mg/kg p.o.)
Figure 4. Occupancy of rat brain benzodiazepine binding sites by MRK-409. (A) Following a dose of 3 mg/kg (p.o.) occupancy was greater at 1 h
(
77 ꢅ 6%) compared with 6 h (36 ꢅ 5%) after dosing. Values shown are mean ꢅ SEM (n ¼ 6/group). (B) Occupancy 0.75–1 h after dosing was
dose-dependent with an Occ50 of 2.2 mg/kg. Values shown are mean ꢅ SEM (n ¼ 5–18 animals measured in 1–4 separate experiments).
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

3
20
Journal of Psychopharmacology 25(3)
Experiment 2, vehicle-treated animals spent a similar propor- responses recorded in the dark were not significantly different
tion of time on the open arms (16 ꢅ 2% and 17 ꢅ 1%, respec- in rats treated with any of the doses of MRK-409
tively). Similarly, chlordiazepoxide produced a significant (0.3–3 mg/kg p.o.), consistent with MRK-409 not having
increase in time spent on the open arms in both Experiment overt sedative properties (Figure 7A). However, there was a
1
(time on open arms, 35 ꢅ 4%) and Experiment 2 (31 ꢅ 3%) dose-dependent decrease in the startle responses recorded
at comparable levels of receptor occupancy (respective occu- during presentation of the conditioning stimulus (light) but
pancy values ¼ 30 ꢅ 2% and 24 ꢅ 2%). MRK-409 induced a this was only significantly from vehicle-treated rats at a dose
dose-proportional increase in the time spent on the open arms of 3 mg/kg (Figure 7A). With regard to the differences scores
which achieved significance at doses of 3 mg/kg in Experiment (i.e. the fear-potentiated startle) MRK-409 again produced a
1
and 2 mg/kg in Experiment 2, corresponding to receptor dose-dependent decrease which was significantly different
occupancies of 74 ꢅ 3% and 43 ꢅ 6%, respectively.
from vehicle at a dose of 3 mg/kg. Unlike the elevated plus
maze experiments, occupancy was not measured directly after
testing (Figure 7B). Nevertheless, in comparison with data
Fear-potentiated startle: Figure 7 shows that MRK-409 from other experiments, the minimal effective dose of 3 mg/
attenuated fear-potentiated startle in the rat. Baseline startle kg p.o. corresponds to 65% occupancy (Table 3).
Conditioned suppression of drinking: The rate at which
rats trained to associate the presentation of a light with an
1
00
electric shock licked a waterspout prior to and during presen-
tation of the conditioning stimulus (i.e. light) is shown in
Figure 8A. This data demonstrates that prior to the presen-
tation of the light stimulus the lick rate was not significantly
altered, even at a dose of 10 mg/kg MRK-409. In vehicle-
treated rats, presentation of the light produced a marked
8
6
4
2
0
0
0
0
0
(
77%) reduction in the lick rate. However, this suppression
of responding was alleviated in a dose-dependent manner by
MRK-409 such that the rate of responding at a dose of
10 mg/kg was significantly greater than in vehicle-treated
rats (Figure 8A).
From this primary data, the suppression ratio, which is an
index of the light (i.e. fear)-induced reduction in responding,
was calculated (Figure 8B). Analysis of this parameter
showed a dose-dependent reversal of the response suppression
such that responding at doses of 3 and 10 mg/kg p.o.
was significantly greater than in vehicle-treated animals.
The minimal effective dose (3 mg/kg) corresponds to
3
10
30
100 300 1000 3000 10000
Plasma MRK-409 (ng/mL)
Figure 5. Relationship between plasma drug concentrations and occu-
pancy in rat brain. Each data point represents an individual animal
(
n ¼ 90). The EC50 was 115 ng/mL with the Hill slope being 1.11.
(
A)
(B)
40
30
20
10
0
*
*
40
*
*
*
30
20
10
0
1
1%
31%
1
74%
3
30%
11%
1
43%
2
62%
3
24%
Veh.
0.3
CDP
Veh.
CDP
MRK-409 (mg/kg p.o.)
MRK-409 (mg/kg p.o.)
Figure 6. Effects of MRK-409 on the rat elevated plus maze. Date shown is the mean time ( ꢅ SEM, n ¼ 17–18/group) spent on the open arms
expressed as a percentage of the total 5 min trial time) after 0.5 h pre-treatment with: (A) 0.3, 1 or 3 mg/kg p.o. (Experiment 1) or 1, 2 or; (B) 3 mg/kg
(
p.o. (Experiment 2) MRK-409. CDP ¼chlordiazepoxide. *p < 0.05 versus vehicle-treated animals using ANOVA followed by Dunnett’s post-hoc t-test.
Figures within bars are the mean occupancy values obtained from a subset of animals tested (n ¼ 9).
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

Atack et al.
321
(
A)
Startle amplitude
(B)
Dark:light difference scores
†
1
1
1
50
25
00
*
70
60
†
†
5
4
3
2
1
0
0
0
0
0
0
7
5
2
5
0
5
0
†
*
(4%)
(22%)
(65%)
Veh.
0.3
1
3
Veh.
0.3
1
3
MRK-409 (mg/kg p.o.)
MRK-409 (mg/kg p.o.)
Figure 7. Effects of MRK-409 in rat fear-potentiated startle. Rats were conditioned to associate the presentation of light with a mild electric
foot-shock and then the extent to which they startled in response to a 100 dB tone was measured in the dark or the presence of light. Results are
expressed as the mean ꢅ SEM (n ¼ 15 group) of (A) the startle amplitude under non-threat (dark) or threat (light) conditions or (B) the mean of the
difference of each animal’s startle amplitude in the light compared with dark (safe) conditions (i.e. the extent of the fear-potentiated startle).
Figures within bars represent estimated occupancy values (Table 3). Within-group comparisons were made between the startle amplitude in the light
y
and dark ( p < 0.05 compared with startle responses measured in the dark) and the difference scores were analyses across groups using an analysis of
variance followed by Dunnett’s post hoc t-test). *p < 0.05 compared with startle responses in vehicle-treated control animals).
Lick rate
Lick rate supression ratio
*
*
0
.5
100
Pre CS
CS
*
0.4
80
60
40
20
0
0
0
.3
.2
0.1
0.0
(22%)
(65%)
(93%)
Veh.
1
3
10
Veh.
1
3
10
MRK-409 (mg/kg p.o.)
MRK-409 (mg/kg p.o.)
Figure 8. MRK-409 reverses fear-induced suppression of drinking in rats. Rats were dosed with either vehicle (0.5% methyl cellulose) or MRK-409 (1, 3
or 10 mg/kg p.o.) and 0.5 h later their rates of responding (water spout licking) were measured prior to and during a 60-sec presentation of a
light-stimulus that had previously been associated with an electric shock. Figures within bars represent estimated occupancy values (Table 3). Values
are mean ꢅ SEM (n ¼ 11–12/group). *p < 0.05 versus corresponding vehicle-treated animals (ANOVA followed by Dunnett’s post-hoc t-test).
occupancy of 65% whereas the maximum dose tested that at each 10-min interval during the 1-h testing ses-
(10 mg/kg p.o.), and at which no overt sedation was observed, sion, diazepam (10 mg/kg p.o.) significantly reduced
is equivalent to occupancy of 93% (Table 3).
chain-pulling rates. At a dose of 30 mg/kg p.o. MRK-409
the response rate at each 10-min time point was consistently
lower than in vehicle-treated rats. However, when the mean
chain pulls/min were averaged over the whole 1-h
MRK-409 is non-sedating in rodents
Rat chain-pulling: Figure 9A shows the effects of vehicle session (Figure 9B), there was not only a profound effect of
0.5% methyl cellulose), MRK-409 (10 and 30 mg/kg p.o.) diazepam but also a statistically significant, albeit modest
(
and diazepam (10 mg/kg p.o.) on the number of chain (25%), decrement in responding at the 30 mg/kg
pulls/minutes at successive 10 min intervals during the (but not 10 mg/kg) dose. Nevertheless, it should be noted
1-h session (expressed as a percentage of the baseline response that this modest effect on response rates was achieved
rate obtained from the previous drug-free session; left-hand only at a dose (30 mg/kg) that corresponds to 99% occupancy
panels). Statistical analysis of the chain-pulling rates showed (Table 3).
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

3
22
Journal of Psychopharmacology 25(3)
(
A)
(B)
80
100
*
Veh.
10
80
60
40
20
0
60
3
0
4
2
0
0
0
*
Diaz.
**
*
*
*
*
(93%)
(99%)
*
Veh.
10
30
Diaz.
0 mg/kg p.o.
0
10
20
30
40
50
60
1
Time (min)
MRK-409 (mg/kg p.o.)
Figure 9. Effects of MRK-409 in the rat chain-pulling assay of sedation. Rats were dosed p.o. with either vehicle (0.5% methyl cellulose),
MRK-409 (10 or 30 mg/kg) or diazepam (10 mg/kg) and the rate of chain pulling was recorded over the subsequent 1 h period with data being expressed
as a percentage of baseline responding. (A) The number of chain pulls/min at 10 min intervals during the 1 h session. (B) The mean chain
pulls/min averaged across the session. Figures within bars represent estimated occupancy values (Table 3). Values shown are mean ꢅ SEM
(
n ¼ 12/group). *,**p < 0.05 and p < 0.01 versus vehicle-treated animals (ANOVA followed by Student Newman-Keuls post-hoc t-test).
(
A)
Minus ethanol
(B)
Plus ethanol
*
1
20
120
90
60
30
0
9
6
3
0
0
0
0
*
*
*
2
%
28%
79%
66%
9%
32%
72%
47%
Veh.
1
3
10
Diaz.
Veh.
1
3
10
Diaz.
10 mg/kg p.o.
3 mg/kg p.o.
MRK-409 (mg/kg p.o.)
MRK-409 (mg/kg p.o.)
EtOH (1.5g/kg i.p.)
Figure 10. Effects of MRK-409 on rotarod performance in the absence and presence of ethanol. The latency to fall-off the rotarod during a 2-min trial
was measured in mice 0.5 h after p.o. dosing with either vehicle (0.5% methyl cellulose), MRK-409 (1, 3 or 10 mg/kg) or diazepam (3 or 10 mg/kg) in
(
A) the absence or (B) the presence of ethanol (1.5 g/kg i.p., 0.5 h pretreatment). Figures within bars are the mean occupancy values measured
immediately after completion of the trial. In these studies, the occupancy ED50 values were 5.5 and 5.4 mg/kg in the absence and presence of ethanol,
respectively. Values shown are mean ꢅ SEM (n ¼ 7–8/group). *p < 0.05 versus vehicle group.
Mouse rotarod: Mice were dosed p.o. with either vehicle rotarod performance (Figure 10B). However, ethanol
(
0.5% methyl cellulose), MRK-409 (1, 3 or 10 mg/kg p.o.) pre-treatment did potentiate the effects of not only diazepam
and diazepam (10 mg/kg p.o.) and then subjected to a trial but also MRK-409. More specifically, doses of MRK-409
on the rotarod lasting up to 2 min. Immediately following which did not impair performance in the absence of ethanol
completion of the trial, mice were taken and occupancy mea- (i.e. trial latency ¼ 120 sec) did so when mice had been
sured. Figure 10A shows that diazepam had a pronounced pre-treated with ethanol such that at 3 and 10 mg/kg average
effect on rotarod performance, with mice lasting on average trial durations were 78 ꢅ 12 and 42 ꢅ 15 sec, respectively.
2
6
1 ꢅ 7 sec before falling off. Occupancy in these animals was Furthermore, these effects were not a consequence of ethanol
6 ꢅ 2%. In contrast, MRK-409 had no effect on perfor- affecting MRK-409 occupancy since in the absence and
mance with all animals completing the 2-min trial, even at a presence of ethanol occupancy values at 3 mg/kg were
dose of 10 mg/kg that corresponds to occupancy of 79 ꢅ 4%. 28 ꢅ 3% and 32 ꢅ 5% and at 10 mg/kg were 79 ꢅ 4%
The same experiment described above was repeated but and 72 ꢅ 3%. Hence, MRK-409 demonstrates a significant
with pre-treatment using a dose of ethanol (1.5 g/kg i.p.) ethanol interaction at a dose (3 mg/kg) corresponding to
which did not in its own right produce any impairment of ꢀ30% occupancy.
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

Atack et al.
323
of 2000 ng/mL. Assuming further that the plasma occupancy
relationship in squirrel monkey is similar to that observed in
MRK-409 is a non-sedating anxiolytic in primates
MRK-409 is anxiolytic-like in the conditioned rat, the respective occupancies at the minimal effective dose
emotional response assay: Two separate conditioned for anxiolytic-like activity (0.1 mg/kg p.o., 69 ng/mL) and
emotional
(
response
Figure 11A) and in both, the presentation of the condition- 2000 ng/mL) are 35% and 95%.
experiments
were
performed at the highest non-sedating dose tested (10 mg/kg p.o.,
ing stimulus produced a marked reduction in the response
rate (response rates ¼ 6 ꢅ 2% and 4 ꢅ 1% of baseline
responding in Experiments 1 and 2, respectively). In the initial
experiment, 0.1 mg/kg MRK-409 p.o. gave a significant
increase in the rate of responding, to 35 ꢅ 10% of the baseline
Single-dose pharmacokinetics and tolerability of
MRK-409 in humans
responding rate, whereas in the second experiment, all of the Healthy male volunteers were given doses of MRK-409 ran-
doses of MRK-409 induced a statistically significant release in ging from 0.05 to 2.0 mg and plasma drug concentrations
lever pressing with respective response rates being 55 ꢅ 9%, were measured. As can be seen from Figure 12, the drug
60 ꢅ 9% and 60 ꢅ 10% at 0.3, 1 and 3 mg/kg p.o.
was rapidly absorbed with the maximum plasma drug con-
centrations being achieved within 0.8–1.3 h after dosing
(
Table 4). Thereafter, drug was rapidly cleared with a half-life
MRK-409 is non-sedating in the lever-pressing that ranged from 2.3 to 2.8 h (Table 4). There was a reason-
assay: Lever pressing rates on drug days, averaged through- ably linear relationship between dose and total drug exposure
out the 30 min test session, were expressed as a percentage of (measured as area under the curve (AUC); Figure 12, inset) as
the previous day’s performance (baseline). As illustrated in well as Cmax (Table 4).
Figure 11B, lever-pressing rates following vehicle treatment
The effect of food as measured by comparing the pharma-
were approximately 96 ꢅ 9% of those recorded on drug-free cokinetic profiles of a 0.5 mg dose of MRK-409 in fasted and
baseline days. MRK-409 did not have any significant effect on fed men showed that feeding reduced the Cmax and delayed
lever-pressing rates with doses of 1, 3 and 10 mg/kg p.o., the Tmax (3.9 ng/mL and 3.3 h) relative to fasted men
producing response rates of 88 ꢅ 9%, 90 ꢅ 11% and (7.6 ng/mL and 0.9 h). However, the total exposures were sim-
8
0 ꢅ 9%, respectively.
ilar in the fasted and fed subjects (respective AUCs of 29.6
In separate pharmacokinetic studies in squirrel monkeys and 27.7 ng.h/mL).
data not shown) the minimum effective dose of 0.1 mg/kg The most frequent adverse events associated with
gave plasma drug concentrations 1 h after dosing of MRK-409 are summarized in Table 4. Drowsiness/sleepiness
9 ng/mL whereas 3 mg/kg gave a plasma drug concentration and tiredness were dose-limiting at doses of 1.5 and 2 mg. At
(
6
of 699 ng/mL, which (and assuming linearity of exposure the highest single oral dose evaluated (2 mg) there was drows-
between 3 and 10 mg/kg) would correspond to a plasma iness in five out of six subjects and dose-limiting moderate and
drug concentration at a dose of 10 mg/kg in the region severe drowsiness in two subjects, one of whom had difficulty
(
A)
Anxiolysis
Expt. 1
(B)
Sedation
1
00
Expt. 2
*
100
80
8
6
4
2
0
0
0
0
0
*
*
6
4
0
0
*
20
0
Veh.
1
3
10
Veh. 0.03 0.1 Veh. 0.3
MRK-409
1
3
MRK-409 (mg/kg p.o.)
MRK-409
(mg/kg p.o.)
(
mg/kg p.o.)
Figure 11. MRK-409 had anxiolytic-like activity in a squirrel monkey conditioned emotional response assay but was devoid of overt sedation in a
lever-pressing response assay. (A) Two separate conditioned emotional response experiments were performed, using MRK-409 doses of either 0.03 and
0
.1 (Experiment 1) or 0.3, 1 and 3 mg/kg. MRK-409 or vehicle (0.5% methyl cellulose) was administered p.o. 0.5 h prior to testing during which the
suppression of lever pressing rates in response to a red cue light that had previously been associated with mild electric tail-shock was measured. Values
shown are mean ꢅ SEM (n ¼ 12/group). *p < 0.05 vs. vehicle control (paired t-test). (B) The number of lever presses per minute during the 1 h session
(expressed as a percentage of baseline responding) are shown following administration of MRK-409 (1–10 mg/kg, p.o.) given 0.5 h before the test
session. Values shown are mean ꢅ SEM (n ¼ 9/group). There were no statistically significant differences between groups as assessed using paired t-tests
p > 0.05 vs. vehicle treatment).
(
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

3
24
Journal of Psychopharmacology 25(3)
staying awake. The severe drowsiness commenced approxi- in preclinical species even at levels of receptor occupancy
mately one hour after dosing and lasted for three to six greater than 90%. However, it is possible that the blood–
hours. Other less frequent adverse experiences included brain barrier permeability and/or central nervous system
headache, tiredness and dizziness. There were no laboratory (CNS) distribution of MRK-409 is markedly different in
adverse events and none of the clinical adverse events was humans compared with rat and that as a consequence the
considered serious. Based upon these findings, the maxi- human plasma-occupancy relationship is different from that
mum tolerated single oral dose was determined to be 1 mg. in rats. As a consequence, the plasma-occupancy relationship
1
1
The 2 mg dose produced marked somnolence in five out of was evaluated in humans using [ C]flumazenil PET.
six subjects and this dose corresponds to a plasma Cmax of
28 ng/mL (Table 4). Using the rat plasma-occupancy relation-
ship (Figure 5), a plasma MRK-409 concentration of 28 ng/
11
[
C]Flumazenil PET studies
mL corresponds to an occupancy of around 20%, which is Figure 13 shows the distribution of brain radioactivity in
surprising given that no overt signs of sedation were observed three healthy volunteers given either placebo (n ¼ 1)
Scan time:
Placebo
–1 h
2 h
30
25
20
15
10
5
0
150
100
50
0
0
.0
0.5
1.0
1.5
2.0
Dose (mg)
MRK-409
MRK-409
2
1
0.25
0.1
1.5
0.5
0.05
0
4
8
12
Time (h)
Figure 12. Plasma drug concentrations following different single oral
doses of MRK-409 in healthy, fasted male volunteers. Values shown
are mean plasma concentrations at doses ranging from 0.05 to 2 mg
11
Figure 13. A pseudocolour representation of the [ C]flumazenil total
T
distribution volume V in the brains of three healthy human volunteers,
(
n ¼ 5–6/dose group) with error bars being omitted for clarity. However,
one of whom received placebo, the other two being given MRK-409
within each dose group, the total exposure (area under the curve, AUC)
and the maximum plasma concentration (Cmax) varied by less than 40%.
The inset shows that the AUC was linearly related to dose within the range
of 0.05–2.0 mg.
(
[
1 mg). In subjects receiving drug, there was no marked decrease in
11
C]flumazenil V demonstrating that there was limited occupancy of the
T
benzodiazepine binding site of human brain GABA
MRK-409.
A
receptors by
Table 4. Pharmacokinetic parameters and adverse events noted after administration of single oral doses of MRK-409 to young healthy male volunteers
Pharmacokinetic parameters Most frequent adverse events
a
AUC(0-1)
(ng.h/mL)
Dose (mg)
Placebo
N
T
max (h)
C
max (ng/mL)
T
½
(h)
Headache
3
Nausea
Tiredness
2
Drowsiness/ sleepiness
1
Dizziness
16
6
N/A
N/A
N/A
N/A
2.4
2.5
2.6
2.3
2.5
2.7
2.8
2.5
0
0
0
0
.05
1.2 ꢅ 0.4
0.8 ꢅ 0.3
1.3 ꢅ 0.5
0.9 ꢅ 0.2
3.3 ꢅ 1.0
1.2 ꢅ 0.7
0.8 ꢅ 0.3
0.9 ꢅ 0.2
0.6 ꢅ 0.1
1.7 ꢅ 0.3
3.4 ꢅ 0.6
7.6 ꢅ 1.8
3.9 ꢅ 0.8
2.5 ꢅ 0.6
.1
5
7.1 ꢅ 2.7
1
2
3
2
.25
.5
6
15.3 ꢅ 4.0
29.6 ꢅ 12.6
27.7 ꢅ 10.4
60.4 ꢅ 18.0
99.4 ꢅ 28.7
112.2 ꢅ 29.0
1
1
6
1
4
2
2
2
1
1
1
0
.5 (fed)
.0
6
2
1
2
5
1
6
13.5 ꢅ 4.6
23.2 ꢅ 5.4
27.8 ꢅ 4.1
b
1
.5
.0
6
b
2
6
NA ¼ not applicable.
a
AUC ¼ area under the (plasma concentration versus time) curve.
b
Doses highlighted in bold were poorly tolerated and defined the maximum tolerated dose as 1.0 mg.
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

Atack et al.
325
or 1.0 mg MRK-409 (n ¼ 2). As can be clearly seen, a dose of highlighting their poor CNS penetration. Accordingly, it is
.0 mg MRK-409 produced very little inhibition of the bind- assumed that the GABA -mediated pharmacological effects
1
A
1
1
ing of [ C]flumazenil to benzodiazepine binding sites in of MRK-409 are indeed associated with this compound rather
human brain and regional brain radioactivity in than a metabolite.
MRK-409-treated subjects was not distinguishable from
that observed for placebo.
In vivo properties of MRK-409 in preclinical species
Discussion
In rats, MRK-409 gave good occupancy after oral dosing
with an Occ50 of 2.2 mg/kg. However, this is around five-fold
less potent than TPA023, which had an Occ50 of 0.42 mg/kg
In vitro properties in MRK-409
MRK-409 is a triazolopyridazine that is structurally related (Atack et al., 2006). This difference in potency is not related
to TPA023 and the prototypic efficacy-selective compound to affinity since the Ki values of both compounds were com-
L-838417 (Figure 1 and Atack et al., 2006; McKernan parable against not only recombinant human but also native
et al., 2000). These compounds are all a2/a3 efficacy selective rat receptors (Ki values ranging from 0.21 to 0.40 nM for
compounds. Thus, although they all bind with equivalent MRK-409 and 0.19 to 0.41 nM for TPA023). Rather, the dif-
A
affinity to a1-, a2-, a3- and a5-containing GABA receptors, ference in potency seems to be related to blood–brain barrier
they have higher partial agonist efficacy at the a2 and/or a3 penetration and/or distribution within the CNS since the
compared with a1 subtypes. The a2 and/or a3 subtypes are plasma concentrations of MRK-409 required to give 50%
thought to be associated with the anxiolytic properties of occupancy (in rat 115 ng/mL) were around 5 times more
benzodiazepines whereas agonism at the a1 subtype is asso- than the corresponding value for TPA023 (25 ng/mL; Atack
ciated with sedation (Atack et al., 2005; Lo
McKernan et al., 2000; Morris et al., 2006; Rudolph et al.,
999; Savic et al., 2008). It is noteworthy, however, that effects not only in an unconditioned model (rat elevated plus
¨
w et al., 2000; et al., 2006).
In preclinical species, MRK-409 produced anxiolytic-like
1
whereas L-838417 and TPA023 have essentially antagonist maze) but also in conditioned anxiety models (rat
efficacy at the a1 subtype, MRK-409 has a degree of weak fear-potentiated startle and conditioned suppression of drink-
partial a1 agonist efficacy, albeit much lower than at the a3 ing and squirrel monkey conditioned emotional response; see
subtype (relative efficacies at the a1 and a3 subtypes of 0.18 Table 5). The minimal effective doses for these anxiolytic
and 0.45, respectively; Table 2).
effects corresponded to occupancies in the region of
In in vitro rat, dog, rhesus monkey and human liver micro- 35–65%, which is consistent with the fact that the lower effi-
some turnover assays, the major metabolite of MRK-409 was cacy compound TPA023 required higher levels of occupancy
the 3-hydroxycyclobutyl analogue and this compound had for anxiolytic-like efficacy (65–88%; Atack et al., 2006).
around 10-fold lower GABAA receptor affinity than the MRK-409 had no overt sedative properties at receptor occu-
parent (unpublished data). The O-dealkylated metabolite of pancies of up to 95% and it was only at a dose (30 mg/kg)
MRK-409 was detected in vivo in rat and dog but the corresponding to receptor occupancy of 99% that a mod-
GABAA receptor affinity of this compound was more than est sedative effect was observed in the rat chain-pulling
500-fold lower than the parent. In addition, neither metabo- assay (Table 5). Similarly, TPA023 had no sedative effect at
lite was detected in rat brain after a dose of 3 mg/kg, occupancies up to 99% (Atack et al., 2006).
Table 5. Summary of the preclinical non-sedating anxiolytic properties of MRK-409
a
Behaviour
Assay
Observation
Dose (mg/kg, p.o.)
Occupancy
Anxiety assays
b
Rat anxiety
Elevated plus maze
Anxiolysis
Anxiolysis
Anxiolysis
Anxiolysis
2
43%
65%
65%
35%
Conditioned suppression of drinking
Fear-potentiated startle
3
3
Primate anxiety
Sedation assays
Mouse sedation
Rat sedation
Conditioned emotional response
0.1
Rotarod
No effect
No effect
Slight effect
No effect
10
10
30
10
79%
93%
99%
95%
Chain-pulling
Chain-pulling
Response sensitivity
Rat sedation
Primate sedation
Other assay
Mouse EtOH interaction
Rotarod
Potentiates EtOH effects
3
32%
a
Occupancy for elevated plus maze and rotarod experiments are measured values, all others are predicted, either from the rat dose–response curve (Table 3) or, in the case of
the squirrel monkey, from the squirrel monkey plasma drug concentrations and the rat plasma-occupancy relationship (Figure 5).
b
Anxiolytic-like doses are minimal effective dose.
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

3
26
Journal of Psychopharmacology 25(3)
Consequently, MRK-409 behaved as a non-sedating anxioly- plasma Cmax of 5 ng/mL) showed little evidence of dissocia-
tic in not only rodents but also primates. tion between sedative effects measured using the VAS alert-
The non-sedating anxiolytic properties of MRK-409 in ness scale and saccadic eye movements with subjects tending
preclinical species can be attributed to its partial agonism at to fall asleep during testing (van Steveninck et al., 1996).
the a2 and a3 subtypes with lower agonism being measured Moreover, and again like MRK-409, bretazenil had pre-
at the a1 (‘sedative’) subtype. In this regard, MRK-409 is viously been shown to behave as a non-sedating anxiolytic
consistent with previous published data describing the in preclinical rodent and primate models (Atack, 2003).
non-sedating anxiolytic effects of a number of compounds
Although the sedation observed with MRK-409 was unex-
with higher agonist efficacy at a2- and/or a3- compared pected it nevertheless remained possible that these effects
with a1-containing receptors, such as L-838417, TPA023, might occur at high levels of occupancy and that anxiolysis
TPA003 and NS11394 (Atack et al., 2006; Dias et al., 2005; could occur at lower doses, thereby maintaining the separa-
McKernan et al., 2000; Mirza et al., 2008; Munro et al., 2008; tion between anxiolytic and sedative doses observed in pre-
Rowlett et al., 2005).
clinical species, albeit to a reduced extent. Consequently, it
was important to establish the level of receptor occupancy
produced by the maximum tolerated dose of MRK-409
Properties of MRK-409 in humans
(
1.0 mg) in healthy normal volunteers. This data (Figure 13)
In young healthy male volunteers, MRK-409 was rapidly shows that MRK-409 gave occupancy below the regional
1
1
absorbed having a Tmax in the region of 1 h, with total expo- test–retest values for [ C]flumazemil PET (i.e. <10%; Salmi
sure (Figure 12, inset) and Cmax (Table 4) both being dose- et al., 2008). This is much less than the occupancy produced
dependent (Cmax values at doses of 0.05 and 2.0 mg ¼ 0.6 and by a 2 mg dose of TPA023 (ꢀ50–60%). As a consequence of
2
7.8 ng/mL, respectively). This data agrees with that from a these results, the development of MRK-409 as a non-sedating
separate study performed to assess the effects of MRK-409 anxiolytic was halted.
also known as MK-0343) on saccadic eye movements using
single doses of 0.25 and 0.75 mg in which the Cmax occurred
(
Comparison of MRK-409 with TPA023
within the first hour, with values in the region of 3.5 and
9
ng/mL, respectively (de Haas et al., 2008). There was a clear difference between the properties of
Most notable, however, was the observation that MRK-409, which caused sedation at relatively low levels of
MRK-409 produced marked sedation in five out of six and occupancy (ꢀ10%), and TPA023, which was non-sedating at
two out of six subjects receiving the 2.0 and 1.5 mg dose, a dose (2 mg) that gave 50–60% occupancy (de Haas et al.,
respectively. A dose of 0.75 mg has been reported to impair 2007, 2008), raising the question of which aspect of the effi-
saccadic eye movement peak velocity, visual analogue scale cacy profiles of MRK-409 and TPA023 is responsible for the
(
VAS) alertness scores and postural stability (increased body sedation seen with MRK-409 but not with TPA023. The three
sway) and to increase saccadic latency in a manner compara- obvious differences in the intrinsic efficacy between these com-
ble to 2 mg lorazepam, although MRK-409 did not affect pounds are: (1) MRK-409 has partial agonist efficacy at the
memory performance whereas lorazepam did (de Haas et a1 subtype whereas TPA023 does not; (2) MRK-409 has rel-
al., 2008). Non-selective benzodiazepines usually show com- ative efficacy at the a2 and a3 subtypes (0.23 and 0.45) greater
parable effects on memory, alertness and postural stability than the corresponding values for TPA023 (0.11 and 0.21;
(
see discussion in de Haas et al., 2007) and therefore Atack et al., 2006); (3) MRK-409 has partial agonist efficacy
MRK-409 differentiates itself from non-selective benzodiaze- at the a5 subtype (relative efficacy ¼ 0.18) but TPA023 is
pines in that it showed no memory deficits. However, the essentially an antagonist at this subtype. In order to try to
significant effect of 0.75 mg MRK-409 on the VAS alertness resolve this issue, we characterized a compound which had an
score suggests a sedative effect of this compound at this dose. efficacy profile at the a2, a3 and a5 subtypes very similar to
Although MRK-409 has an effect on saccadic peak veloc- that of MRK-409 but was devoid of efficacy at the a1
ity that is comparable to lorazepam, it is uncertain whether subtype, and in this regard was more like TPA023. The
this is a biomarker for sedative or anxiolytic efficacy. Thus, properties of this compound, TPA023B are presented in the
although this effect might be related to sedation (de Visser accompanying manuscript and the data show that the sedat-
et al., 2003), an effect on saccadic peak velocity does not ing properties of MRK-409 are probably related to the weak
necessarily track the effects on VAS alertness scores. For partial agonism at the a1 subtype (Atack et al., 2010).
example, 0.5 and 1.5 mg of TPA023 (doses chosen based
upon the tolerability of TPA023 in healthy volunteers)
affected saccadic peak velocity to a similar extent as loraze-
pam (2 mg) yet both doses were without effect not only on the
VAS alertness score but also on memory performance and
postural stability (de Haas et al., 2007).
GABA subtype-selective modulators – does the
A
hypothesis need refining?
It is also important to note that not all compounds conform
It could be argued that aside from having efficacy at the a3 to the hypothesis that the a1 subtype mediates sedation and
subtype higher than the other subtypes, the efficacy profile of that high efficacy at this subtype is accompanied by sedation.
A
MRK-409 at a1-, a2-, a3- and a5-containing GABA recep- For example, the structurally related compounds Ocinaplon
tors (0.18, 0.23, 0.45 and 0.18) is similar to the non-selective and DOV 51892 have appreciable efficacy at the a1 subtype
partial agonist bretazenil (Atack, 2003). In this regard, and (which, in the case of DOV 51892 might be classified as
like MRK-409, a dose of 0.5 mg bretazenil (which gave a ‘super-agonism’) yet both are non-sedating anxiolytics in
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

Atack et al.
327
preclinical models (Lippa et al., 2005; Popik et al., 2006).
A better understanding of the clinical relevance of
A
Moreover, Ocinaplon showed no signs of sedation at anxio- GABA subtype selective modulators will emerge from the
lytic doses in humans (Lippa et al., 2005) and this cannot be clinical evaluation of additional compounds. In this regard,
attributed to an active metabolite since both Ocinaplon and the properties TPA023B are presented in the accompanying
the active metabolite DOV 315,090 possess similar in vitro manuscript. These data indicate that the sedating properties
efficacy profiles (Berezhnoy et al., 2008). Whilst the reason of MRK-409 are probably related to its weak partial agonism
for the lack of sedation in these apparently anomalous com- at the a1 subtype (Atack et al., 2010).
pounds is unclear, at the very least this data suggests that
efficacy as measured in recombinant a1-containing GABA
receptors may not necessarily be predictive of sedative liabil-
ity (Popik et al., 2006).
A
Abbreviations
MRK-409, 7-cyclobutyl-6-(2-methyl-2H-1,2,4-triazol-
3-ylmethoxy)-3-(2,6-difluorophenyl)-1,2,4-triazolo[4,3-b]-
pyridazine; PET, positron emission tomography; TPA023,
The pyridoindole SL65.1498 possesses appreciable efficacy
at the a1 subtype (relative efficacy of 0.45 versus zolpidem;
Griebel et al., 2001) yet even at a dose of 25 mg there were
no overt signs of sedation. However, although the 25 (but
not 2.5 or 7.5)-mg dose of SL65.1498 did decrease saccadic
peak velocity, these effects were much less than those produced
by not only lorazepam (de Haas et al., 2009) but also TPA023
and MRK-409 (de Haas et al., 2007, 2008). The maximal
plasma concentration achieved with 25 mg SL65.1498 was
7-(1,1-Dimethylethyl)-6-(2-ethyl-2H-1,2,4-triazol-3-ylmethoxy)-
3-(2-fluorophenyl)-1,2,4-triazolo[4,3-b]pyridazine; TPA023B,
6,2’-difluoro-5’-[3-(1-hydroxy-1-methylethyl)imidazo[1,2-b]
[1,2,4]triazin-7-yl]biphenyl-2-carbonitrile.
Acknowledgment
3
75 ng/mL, which was much higher than achieved for the high-
The authors would like to acknowledge Lingling Xue for excellent
support in the bioanalysis of clinical samples.
est tested doses of TPA023 (1.5 mg, Cmax ¼ 13 ng/mL) and
MRK-409 (0.75 mg, Cmax ¼ 9 ng/mL; de Haas et al., 2007,
2
008, 2009). Nevertheless, the affinity of SL65.1498 at the a1,
References
a2 and a3 subtypes (17–80 nM; Griebel et al., 2001) is ꢆ50-fold
less than that of either TPA023 and MRK-409 and therefore in
the absence of receptor occupancy data, the pharmacodynamic
effects of SL65.1498 are difficult to compare with those of
TPA023 or MRK-409.
In addition, the triazolopyrimidine Adipiplon (NG2-73) is
claimed to be an a3-preferring partial agonist that was being
evaluated in Phase II/III clinical studies for the treatment of
insomnia until development for this indication was halted in
July 2008 due to a higher than anticipated rate of unwanted
next-day effects (www.neurogen.com). The fact that a drug
that is claimed to be selective for the ‘anxiolytic’ a3 subtype of
Atack JR (2003) Anxioselective compounds acting at the GABA
receptor benzodiazepine binding site. Curr Drug Targets CNS
Neurol Disord 2: 213–232.
A
Atack JR (2005) The benzodiazepine binding site of GABAA recep-
tors as a target for the development of novel anxiolytics. Expert
Opin Investig Drugs 14: 601–618.
Atack JR, Hallett D, Tye S, et al. (2010) Preclinical and clinical phar-
A
macology of TPA023B, a GABA receptor a2/a3 subtype-
selective agonist. Accompanying article.
Atack JR, Hutson PH, Collinson N, et al. (2005) Anxiogenic proper-
ties of an inverse agonist selective for a3 subunit-containing
A
GABA receptors. Br J Pharmacol 144: 357–366.
Atack JR, Smith AJ, Emms F, McKernan RM (1999) Regional dif-
3
ferences in the inhibition of mouse in vivo [ H]Ro 15-1788 binding
GABA
trary to the prevailing hypothesis that the a1 GABA
A
receptors has hypnotic properties would appear con-
subtype
A
reflect selectivity for a1 versus a2 and a3 subunit-containing
GABAA receptors. Neuropsychopharmacology 20: 255–262.
Atack JR, Wafford KA, Tye SJ, et al. (2006) TPA023
mediates sedation whereas the a2 and/or a3 subtypes are
anxiolytic and a5-containing receptors mediate aspects of
cognition. However, when efficacy is expressed relative to
standard non-selective benzodiazepines (such as lorazepam,
triazolam, diazepam and alprazolam) rather than as a maxi-
mal potentiation, Adipiplon did not appear to have a marked
a3-selectivity (Krause et al., 2007). Indeed, this compound
had partial agonist efficacy at the a1 subtype that might
account for its clinical properties (Krause et al., 2007).
The apparently paradoxical observations with Ocinaplon
and, possibly SL65.1498 and Adipiplon, raises the issue of to
what extent the in vitro measurement of efficacy reflects the
in vivo situation. More specifically, the intrinsic efficacy of a
compound is measured as the extent to which the peak cur-
rent produced by a submaximal concentration of GABA
[
7-(1,1-dimethylethyl)-6-(2-ethyl-2H-1,2,4-triazol-3-ylmethoxy)-3-
2-fluorophenyl)-1,2,4-triazolo[4,3-b]pyridazine], an agonist selec-
tive for a2- and a3-containing GABA receptors, is a nonsedating
anxiolytic in rodents and primates. J Pharmacol Exp Ther 316:
10–422.
Ballard TM, Knoflach F, Prinssen E, et al. (2009) RO4938581, a
novel cognitive enhancer acting at GABA a5 subunit-containing
(
A
4
A
receptors. Psychopharmacology 202: 207–223.
Berezhnoy D, Gravielle MC, Downing S, et al. (2008)
Pharmacological properties of DOV 315,090, an ocinaplon meta-
bolite. BMC Pharmacol 8: 11.
Brown N, Kerby J, Bonnert TP, Whiting PJ, Wafford KA (2002)
Pharmacological characterization of a novel cell line expressing
human a
Carling RW, Madin A, Guiblin A, et al. (2005)
-(1,1-Dimethylethyl)-6-(2-ethyl-2H-1,2,4-triazol-3-ylmethoxy)-3-
2-fluorophenyl)-1,2,4-triazolo[4,3-b]pyridazine: functionally
) a2/a3-subtype selective
4 3 A
b d GABA receptors. Br J Pharmacol 136: 965–974.
(
(
normally an EC20) can be either potentiated or attenuated
benzodiazepine site agonism or inverse agonism, or, alterna-
7
(
A
tively, positive or negative allosteric modulation, respec-
tively). Such a measure does not take into account other
components of the electrophysiological response, such as
the rates of activation, desensitization and deactivation, that
might be more relevant to synaptic transmission.
A A
selective g-aminobutyric acid (GABA
agonist that exhibits potent anxiolytic activity but is not sedating
in animal models. J Med Chem 48: 7089–7092.
Chambers MS, Atack JR, Carling RW, et al. (2004) An orally bioa-
vailable, functionally selective inverse agonist at the
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015

3
28
Journal of Psychopharmacology 25(3)
benzodiazepine site of GABA
enhancing properties. J Med Chem 47: 5829–5832.
A
a5 receptors with cognition
A
Maubach KA (2003) GABA receptor subtype selective cognition
enhancers. Curr Drug Targets CNS Neurol Disord 2: 233–239.
Mirza NR, Larsen JS, Mathiasen C, et al. (2008) NS11394 [3’-[5-
(1-hydroxy-1-methyl-ethyl)-benzoimidazol-1-yl]-biphenyl-2-car-
Dawson GR, Tricklebank MD (1995) Use of the elevated plus maze
in the search for novel anxiolytic agents. Trends Pharmacol Sci 16:
3
3–36.
Dawson GR, Curnow R, Bayley P, Rambridge A, Tricklebank MD
1994) Lack of effect of flumazenil and CGS 8216 on the
anxiolytic-like properties of loreclezole. Eur J Pharmacol 252:
25–328.
Dawson GR, Maubach KA, Collinson N, et al. (2006) An inverse
agonist selective for a5 subunit-containing GABA receptors
A
bonitrile], a unique subtype-selective GABA receptor positive
allosteric modulator: in vitro actions, pharmacokinetic properties
and in vivo anxiolytic efficacy. J Pharmacol Exp Ther 327:
954–968.
(
3
Morris HV, Dawson GR, Reynolds DS, Atack JR, Stephens DN
(2006) Both a2 and a3 GABA
anxiolytic properties of benzodiazepine site ligands in the condi-
tioned emotional response paradigm. Eur Neurosci 23:
495–2504.
Munro G, Lopez-Garcia JA, Rivera-Arconada I, et al. (2008)
Comparison of the novel subtype-selective GABA
A
A
receptor subtypes mediate the
A
J
enhances cognition. J Pharmacol Exp Ther 316: 1335–1345.
De Haas SL, de Visser SJ, van der Post JP, et al. (2007)
Pharmacodynamic and pharmacokinetic effects of TPA023, a
2
GABA
and placebo in healthy volunteers.
74–383.
A
a2,3 subtype-selective agonist, compared to lorazepam
receptor-positive allosteric modulator NS11394 [3’-[5-(1-
hydroxy-1-methyl-ethyl)-benzoimidazol-1-yl]-
J
Psychopharmacol 21:
3
biphenyl-2-carbonitrile] with diazepam, zolpidem, bretazenil, and
gaboxadol in rat models of inflammatory and neuropathic pain.
J Pharmacol Exp Ther 327: 969–981.
De Haas SL, de Visser SJ, van der Post JP, et al. (2008)
Pharmacodynamic and pharmacokinetic effects of MK-0343, a
GABA
and placebo in healthy male volunteers. J Psychopharmacol 22:
4–32.
De Haas SL, Franson KL, Schmitt JAJ, et al. (2009) The pharma-
cokinetic and pharmacodynamic effects of SL65.1498, a GABA
2,3 selective agonist, in comparison with lorazepam in healthy
volunteers. J Psychopharmacol 23: 625–632.
A
a2,3 subtype selective agonist, compared to lorazepam
Popik P, Kostakis E, Krawczyk M, et al. (2006) The
anxioselective agent 7-(2-chloropyridin-4-yl)pyrazolo-[1,5-a]-pyri-
midin-3-yl](pyridin-2-yl)methanone (DOV 51892) is more effica-
2
cious than diazepam at enhancing GABA-gated currents at a
subunit-containing GABA receptors. J Pharmacol Exp Ther
19: 1244–1252.
Rowlett JK, Platt DM, Lelas S, Atack JR, Dawson GR (2005)
Different GABA receptor subtypes mediate the anxiolytic,
1
A
A
a
3
De Visser SJ, van der Post JP, de Waal PP, Cornet F, Cohen AF, van
Gerven JMA (2003) Biomarkers for the effects of benzodiazepines
in healthy volunteers. Br J Clin Pharmacol 55: 39–50.
A
abuse-related, and motor effects of benzodiazepine-like drugs in
primates. Proc Natl Acad Sci USA 102: 915–920.
Dias R, Sheppard WFA, Fradley RL, et al. (2005) Evidence for a
¨
A
Rudolph U, Mohler H (2004) Analysis of GABA receptor function
significant role of a3-containing GABA
the anxiolytic effects of benzodiazepines.
0682–10688.
A
receptors in mediating
and dissection of the pharmacology of benzodiazepines and gen-
eral anesthetics through mouse genetics. Annu Rev Pharmacol
Toxicol 44: 475–498.
J
Neurosci 25:
1
Griebel G, Perrault G, Simiand J, et al. (2001) SL651498: An anxio-
selective compound with functional selectivity for a2- and
Rudolph U, Crestani F, Benke D, et al. (1999) Benzodiazepine
actions mediated by specific g-aminobutyric acid
types. Nature 401: 796–800.
A
receptor sub-
a3-containing g-aminobutyric acid
J Pharmacol Exp Ther 298: 753–768.
A
(GABA
A
)
receptors.
Salmi E, Aalto S, Hirvonen J, et al. (2008) Measurement of GABA
A
Hadingham KL, Garrett EM, Wafford, KA, et al. (1996) Cloning of
cDNAs encoding the human g-aminobutyric acid type A receptor
a6 subunit and characterization of the pharmacology of
a6-containing receptors. Mol Pharmacol 49: 253–259.
Hadingham KL, Wingrove P, Le Bourdelles B, Palmer KJ, Ragan
CI, Whiting PJ (1993) Cloning of cDNA sequences encoding
11
receptor binding in vivo with [ C]Flumazenil: a test-retest study
in healthy subjects. Neuroimage 41: 260–269.
¨
A
Savic MM, Huang S, Furtmuller R, et al. (2008) Are GABA recep-
´
tors containing a5 subunits contributing to the sedative properties
of benzodiazepine site agonists? Neuropsychopharmacology 33:
332–339.
Sieghart W (2006) Structure, pharmacology, and function of GABA
receptor subtypes. Adv Pharmacol 54: 231–263.
A
human a2 and a3 g-aminobutyric acid receptor subunits and
A
characterization of the benzodiazepine pharmacology of recombi-
nant a1-, a2-, a3-, and a5-containing human g-aminobutyric
acid receptors. Mol Pharmacol 43: 970–975.
A
Sieghart W, Sperk G (2002) Subunit composition, distribution and
function of GABA receptor subtypes. Curr Top Med Chem 2:
95–816.
Simon J, Wakimoto H, Fujita N, Lalande M, Barnard EA (2004)
Analysis of the set of GABA receptor genes in the human
genome. J Biol Chem 279: 41422–41435.
Van Laere K, Bormans G, Sanabria-Bohorquez SM, et al. (2008) In
A
Krause JE, Matchett M, Crandall M, Yu J, Baltazar C, Brodbeck
RM, Kehne JH, Maynard GD, Shaw KR, Chenard BL,
Hutchison AJ, Xie L and Yu W (2007) I. Preclinical characteriza-
tion in vitro of NG2-73 as a potent and selective partial allosteric
7
A
A
activator of GABA receptors with predominant efficacy at the
´
a3 subunit. Program No. AAA26. 2007 Neuroscience Meeting
Planner. San Diego, CA: Society for Neuroscience, 2007. Online.
Lippa A, Czobor P, Stark J, et al. (2005) Selective anxiolysis pro-
vivo characterization and dynamic receptor occupancy imaging of
TPA023B, an a2/a3/a5 subtype selective g-aminobutyric acid-A
partial agonist. Biol Psychiatry 64: 153–161.
duced by ocinaplon, a GABA
Acad Sci USA 102: 7380–7385.
A
receptor modulator. Proc Natl
Van Steveninck AL, Gieschke R, Schoemaker RC, et al. (1996)
Pharmacokinetic and pharmacodynamic interactions of bretazenil
and diazepam with alcohol. Br J Clin Pharmacol 41: 565–573.
Wafford KA, Thompson SA, Thomas D, Sikela J, Wilcox AS,
Whiting PJ (1996) Functional characterization of human
g-aminobutyric acidA receptors containing the a4 subunit.
Mol Pharmacol 50: 670–678.
Low K, Crestani F, Keist R, et al. (2000) Molecular and neuronal
substrate for the selective attenuation of anxiety. Science 290:
¨
131–134.
McKernan RM, Whiting PJ (1996) Which GABA
A
-receptor subtypes
really occur in the brain? Trends Neurosci 19: 139–143.
McKernan RM, Rosahl TW, Reynolds DS, et al. (2000) Sedative but
not anxiolytic properties of benzodiazepines are mediated by the
¨
Wieland HA, Luddens H, Seeburg PH (1992) A single histidine in
GABAA receptors is essential for benzodiazepine agonist binding.
J Biol Chem 267: 1426–1429.
GABA
A
receptor a1 subtype. Nature Neurosci 3: 587–592.
Downloaded from jop.sagepub.com at FLORIDA ATLANTIC UNIV on May 25, 2015