A new pyridazine series of GABAA α5 ligands

Article abstract of DOI:10.1021/jm050249x

Screening of the Merck compound collection identified 6 as an unusually simple, low molecular weight hit with moderate affinity for GABA_A receptors. The structural novelty of 6, compared to our advanced series of GABA_A α5 inverse ago

Full text of DOI:10.1021/jm050249x

6004
J. Med. Chem. 2005, 48, 6004-6011
A New Pyridazine Series of GABAA r5 Ligands
Monique B. van Niel,* Kevin Wilson,* Charles H. Adkins, John R. Atack, Jose´ L. Castro, Dawn E. Clarke,
Stephen Fletcher, Ute Gerhard, Mark M. Mackey, Sallie Malpas, Karen Maubach, Robert Newman,
Desmond O’Connor, Gopalan V. Pillai, Peter B. Simpson, Steven R. Thomas, and Angus M. MacLeod
Departments of Medicinal Chemistry, Biochemistry and Pharmacology, Merck Sharp & Dohme Research Laboratories,
Terlings Park, Eastwick Road, Harlow, Essex, CM20 2QR, United Kingdom
Received March 18, 2005
Screening of the Merck compound collection identified 6 as an unusually simple, low molecular
weight hit with moderate affinity for GABAA receptors. The structural novelty of 6, compared
to our advanced series of GABAA R5 inverse agonists, made it an attractive molecule for further
exploration. This paper will describe the evolution of 6 into a new series of ligands with
nanomolar affinity and functional selectivity for GABAA R5 receptor subtypes.
Introduction
subunits (R1-R6, â1-â3, γ1-γ3, δ, ꢀ, θ, π).2,3 Evolution
has favored a small number of combinations and the
more abundant assemblies R1â2γ2, R2â3γ2, R3â3γ2,
and R5â3γ2 show distinct regional distribution in the
brain. Ion channels containing an R5 subunit account
for 25% of GABA receptors in the hippocampus, an area
of the brain implicated in learning and memory.4
Disease states such as dementias do not affect the
population density of R5-containing receptors in the
hippocampus, making them an attractive target for
therapeutic intervention. Additional evidence for the
role of this subtype in hippocampal function has come
from transgenic mouse models in which R5 subunits are
absent or decreased in the hippocampus. These animals
showed enhanced performance in learning and memory
tests5,6 giving support to the hypothesis that R5 inverse
agonists would be cognition enhancers.
After three decades of benzodiazepine use, it was
established through use of recombinant receptors that
binding of these drugs occurs exclusively at the interface
of the R and γ subunits.7-10 The commonality of the
binding contribution made by the γ2 subunit this makes
the discovery of binding selective compounds a signifi-
cant challenge. Unpleasant side effects, such as sedation
seen with benzodiazepine agonists or seizures observed
with inverse agonists such as DMCM, are a consequence
of the nonselective nature of these compounds and
targeting drugs with single receptor subtype selectivity
may result in drugs with improved profiles. The trans-
genic mice described above do not show an increase in
seizure activity, suggesting that selective R5 inverse
agonists would be cognition enhancers devoid of un-
wanted side effects. Following on from elegant biology
came the challenge to medicinal chemists to identify
subtype selective molecules11 which would enable clini-
cal evaluation of the hypothesis.
GABAA ligand-gated ion channels play a key role
during fast inhibitory neurotransmission in the mam-
malian central nervous system and are activated by
the endogenous neurotransmitter γ-aminobutyric acid
(GABA). Additional binding sites have been found on
the GABAA ion channel at which synthetic compounds,
including anaesthetics, avermectins, barbiturates, ben-
zodiazepines and loreclezole exert their predominant
pharmacological actions. Some of these binding sites
influence the binding of GABA, and this allosteric,
pharmacological modulation, results in changes in flux
of chloride through the ion channel, with concomitant
effects on neuronal activity and disease states.
Of particular interest is the site that binds benzo-
diazepines, for which no endogenous ligand has so far
been identified. At this site, exogenous ligands can be
classed as positive allosteric modulators (full agonists,
e.g. diazepam) which enhance the effects of GABA and
increase the chloride flux: negative allosteric modula-
tors (inverse agonists, e.g. 6,7-dimethoxy-4-ethyl-â-
carboline-3-methoxylate, DMCM) which decrease chlo-
ride flux: or antagonists (e.g. flumazenil) which cause
no overall change in chloride flux but can prevent
agonists or inverse agonists from binding.
Full agonists such as diazepam have been in clinical
use as anxiolytics, anticonvulsants and sedatives since
the 1960s. Memory impairment and dependence are
associated adverse side effects. Inverse agonists produce
converse behaviors including anxiety and convulsions,
and they can augment the activity of convulsant com-
pounds when they are coadministered (proconvulsant
profiles). For these reasons, inverse agonists have, to
date, only been used as research tools. Of particular
interest is the observation that inverse agonists enhance
learning and memory in animals and this indication has
been actively studied in several laboratories.1
With these aims in mind, we have recently reported
the identification of 1,12 2,13 314 and 415 as R5 GABAA
inverse agonists which are cognition enhancing in
rodent models (Chart 1). As part of a back-up strategy,
a search for truly structurally diverse leads to comple-
ment existing series was initiated by screening of
compounds in the Merck collection for affinity at R1, R3
and R5 GABAA receptor subtypes. Methyl 1,5-diphenyl-
Molecular biology studies in the late 1980s and early
1990s demonstrated that the GABAA channels are
formed from pentameric assemblies of 16 different
* To whom correspondence should be addressed. Tel +44 (0)1279
440000; Fax +44 (0) 1279 440390; e-mail: monique_vanniel@merck.com
and kevin_wilson2@merck.com.
10.1021/jm050249x CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/23/2005

Pyridazine Series of GABAA R5 Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6005
Chart 1
Scheme 1a
a Reagents: (i) Pd(PPh3)4, THF, ArBr, 150 °C, 600 s, microwave
reactor; (ii) I2, THF; (iii) Pd(PPh3)4, THF, 2N Na2CO3, ArB(OH)2,
150 °C, 600 s, microwave reactor.
above. Our in-house diversity program24 was used to
select a set of 50 aryl boronic acids and halides to fully
capture a diverse selection of reagents to explore chemi-
cal space at the 5-position of the pyridazine. A smaller
set of 10 proprietary and privileged fragments to add
further diversity coverage and novelty was also selected.
Following simple workup, products were purified by
HPLC with mass triggered fraction collection or crystal-
lization. The overall process was very efficient, with the
design and synthesis of several small targeted libraries,
through iterative loops, being achieved in a short time
frame.
2,3-diazabicyclo[3.1.0]hex-2-ene-6-carboxylate (5)16,17 was
identified as an interesting lead with good affinity for
GABAA R5 receptors (Chart 2). Subsequent assessment
Chart 2
(b) 3,5-Diarylpyridazines. Following identification
of a number of interesting leads from this initial work
(see below), our attention also focused on modifications
to the 3-aryl substituent. Accordingly, 5-chloropyridazin-
3(2H)-one 1025 was identified as a valuable intermediate
that enabled the incorporation of 5-aryl substituents
using a Suzuki coupling (Scheme 2). Only minor modi-
of the original stored solution revealed that the sample
had rearranged and oxidized to form the pyridazine 6.18
Rescreening of a sample prepared by independent
synthesis confirmed that 6 was responsible for the
observed binding affinity. The low molecular weight and
intriguingly simple structure of 6 made it an attractive
molecule for further exploration.19 This paper will
describe the evolution of 6 into a new series of ligands
with nanomolar affinity and functional selectivity for
GABAA R5 receptor subtypes.
Scheme 2a
Synthetic Chemistry
(a) 3-Phenyl-5-arylpyridazines. Chemistry was
initiated with a resynthesis of screening lead 6.18
Having confirmed that 6 was responsible for the ob-
served binding affinity at GABA receptors, further work
was instigated to explore preliminary structure-activity
relationships (SAR) of this lead. To determine the
importance of the ester for binding affinity, pyridazine
7 was prepared using stannane 820 (Scheme 1). The
finding that 7 retained good binding affinity and was
accessible via an advanced intermediate made this
molecule a more attractive lead for further evalua-
tion.21,22 The low molecular weight of pyridazine 7 would
also accommodate a number of options for optimizing
in vitro and in vivo profiles by increasing the complexity
of the molecular architecture. It was also anticipated
that the stannane 8 could be reacted in parallel using
solution phase synthesis in a microwave reactor.23
Multiple Stille couplings could be performed in a short
time frame using simple apparatus. The stannane 8 was
also converted to the iodide 920 (Scheme 1) which
allowed incorporation of a complementary set of aryl
boronic acids using a microwave reactor as described
a Reagents: (i) dichloro-(1,4-bis(diphenylphosphino)butane),
PhCH3, Na2CO3, ArB(OH)2, 150 °C, 600 s, microwave reactor; (ii)
POCl3; (iii) dichloro-(1,4-bis(diphenylphosphino)butane)palladium,
THF, Na2CO3, Ar1B(OH)2, 150 °C, 600 s, microwave reactor.
fications to the catalyst and solvent used in the original
protocol were required. Conversion of the 5-aryl-
pyridazinones 11 to the chloride 12 enabled further
functionalization to be carried out at the 3-position using
an additional Suzuki coupling. This sequence gave
efficient access to a range of 3,5-diaryl substituted
pyridazines for evaluation.
(c) 3-Amino- and 3-Alkoxy-5-arylpyridazines.
5-Aryl-3-chloropyridazines 12 were also ideal substrates

6006 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19
van Niel et al.
Scheme 3a
respectable binding affinity. However, whereas 15 and
16 had been equipotent in the biaryl pyridazine series,
33 was 7-fold less active at R5 than the corresponding
bromide 29. Introduction of some simple alkyl amino
substituents such as 36, 37 and 40 were also tolerated.
Interestingly, the affinity for 36 and 38 tracked that
observed with the corresponding alkoxide pairs 29 and
33 , but in contrast to this, 40 had higher affinity than
39. Thus, amine and alkoxide substitution could offer
useful alternative scaffolds with additional structural
variation, for further optimization.
A comparison of 41 with 15 illustrates that substit-
uents are not transferable from the 5-position to the
3-position of the pyridazine. This observation supported
our earlier conclusions regarding the subtle directional
influences of the nitrogen atoms.
a Reagents: (i) ROH, Na; (ii) R2NH, THF, 170 °C, microwave
reactor, 600 s.
to explore amine and alkoxide substitution at the
3-position. Displacement of the chloride occurred rapidly
with freshly generated alkoxides at room temperature
or with amines under microwave conditions (Scheme 3).
Results and Discussion
Functional Efficacy. The pyridazines, in common
with our previously reported structural classes,12-15
have modest binding selectivities for the receptor sub-
types. With binding selectivities of less than an order
of magnitude, discrimination between the receptor
subtypes was targeted through functional selec-
tivity. Patch-clamp electrophysiological techniques were
used to determine functional efficacy. Although this
generates high quality data, it has low capacity for
screening purposes, and the multiple compounds gener-
ated from parallel synthesis could not be evaluated
efficiently in this manner. Primary functional screening
data was thus generated using a voltage/ion probe
reader (VIPR)26,27 assay as an aid in the selection of
compounds which would be more fully evaluated.
We were encouraged to note that while 7 retained a
respectable in vitro binding profile it was also a weak
inverse agonist at R5 receptors in the VIPR functional
assay. It became apparent from the early analogues
which were prepared in this pyridazine series that R5
functional selectivity could be modulated through in-
corporation of simple substituents in the C5 phenyl ring.
In addition to improving binding affinity, the m-cyano
(15), methyl carboxylate (18) and triazole analogue (20)
were also functionally selective inverse agonists for R5
receptor subtypes. The nature of the substituent clearly
plays a crucial role in influencing efficacy profiles, with
analogues 16, 17, 19, 21 and 22 demonstrating a range
of efficacies at R5 receptors and functional selectivities
over other receptor subtypes.
Assessment of fluorinated analogues 23-27 in the
VIPR assay indicated that only 23 retained a balance
of binding affinity and R5 functional selectivity. There
were indications that heteroatom-linked pyridazines
such as 29, 36, 37 and 40 could attain good affinity at
R5 receptors although only 36 showed R5 functional
selectivity.
Clearly, the primary binding and functional assays
acted as efficient, but stringent, filters for the com-
pounds generated using parallel synthesis. We were
encouraged to identify 15, 23 and 36 since they had
desired in vitro profiles. Analogue 15 was selected for
full evaluation of efficacy profiles. Measurements in
mouse fibroblast L(tk-) cells expressing human GABAA
receptor subtypes demonstrated that 15 was an R5
inverse agonist (efficacy ) -37%) as predicted using
VIPR (Table 2). Reference compounds include non
selective DMCM which is regarded as a R5 full inverse
Binding Affinity. Although removal of the ester of
6 to form 7 did not give an enhanced in vitro binding
profile, it did clearly simplify the pyridazine core struc-
ture and improved synthetic access. Both pyridazine
nitrogen atoms are absolutely required to achieve good
binding affinity, and their role is subtle. Thus 2,4- and
3,5-diphenylpyridine and 4,6-diphenylpyrimidine have
no affinity for any receptor subtypes (Ki > 600 nM). 3,5-
Diphenyl-1,2,4-triazine and 4,6-diphenyl-1,2,3-triazine
were the only alternative heterocycles that retained
weak affinity (R5: Ki ) 160 nM and 300 nM, respec-
tively), and our SAR was thus confined to the original
3,5-diphenylpyridazine (7). Preliminary efforts were
focused on modifications to the 5-phenyl group. From
our first parallel synthesis run it became apparent that
meta-substitution of this phenyl ring was favored over
the ortho or para positions. For example, the m-cyano
substituent 15 had nanomolar binding affinity at R5
receptors whereas 13 and 14 were at least 10 fold
weaker (Table 1). Additional substituents that followed
this observation include bromo (16), isopropoxy (17) and
methyl carboxylate (18) groups. Thiophene 19 was also
identified in the first library and had a similar binding
profile when compared to the phenyl isostere 18. The
strategy to use a diverse set of aryl halides reagents
during the first targeted library was clearly valuable
in identifying further interesting lead compounds with
improved binding affinities at R5 receptors. Another
beneficial strategy was to include proprietary fragments.
Compounds 20 and 21, substituted with heterocycles,
also showed improved in vitro profiles, with 22 having
the highest observed affinity at all receptor subtypes.
On the basis of binding affinity and efficacy profiles
(see below), 15 was selected for further elaboration.
Fluorine plays a pivotal role in the life sciences industry
as a result of its ability to modify molecular properties
including lipophilicity, metabolic stability and confor-
mation in addition to binding affinities. This substituent
was therefore incorporated on the 3-phenyl ring. We
observed that only 2-fluoro (23) and 3-fluoro (24)
substitution retained low nanomolar binding affinity for
the R5 receptor subtype. Derivatives 25-27 had weaker
binding affinity, and this gave us preliminary indica-
tions about substitution patterns which may be toler-
ated.
Replacement of the 3-phenyl group with alkoxides and
amines was also pursued. Brief exploration with alkox-
ide substituents showed that ethoxide (29) did retain

Pyridazine Series of GABAA R5 Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6007
Table 1. Binding Affinity and Efficacy of Pyridazines at Human Recombinant GABAA Rxâ3γ2 Receptor Subtypes
binding affinity Ki, nMa
efficacyb
R1
R2
R3
R1
R3
R5
R3
n.d
-20 ( 11
n.d.
49 ( 11
-4 ( 7
-2 ( 9
22 ( 10
-7 ( 6
R5
6
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
CO2CH3
Ph
Ph
154 ( 48
36 ( 4
n.d
27 ( 3
7.0 ( 3.7
7.5 ( 1.9
12 ( 4
n.d
64 ( 25
45 ( 6
12 ( 3
29 ( 6
323 ( 33
56 ( 7
5.2 ( 1.3
7.9 ( 2.2
8.4 ( 2.5
11 ( 1
27 ( 6
-14 ( 10
n.d.
7
H
H
H
H
H
H
H
13
14
15
16
17
18
2-CN-Ph
440 ( 41
81 ( 9
4-CN-Ph
45 ( 9
3-CN-Ph
14 ( 4
-32 ( 6
-2 ( 9
18 ( 9
3-Br-Ph
15 ( 3
3-OiPr-Ph
3-CO2CH3-Ph
27 ( 8
29 ( 10
-48 ( 7
19
Ph
H
88 ( 17
45 ( 8
41 ( 3
12 ( 3
-44 ( 6
-39 ( 5
-31 ( 6
20
Ph
H
n.d
12 ( 2
13 ( 11
21
22
Ph
Ph
H
H
8.5 ( 1.1
10.4 ( 1.0
8.8 ( 1.4
1.6 ( 0.2
43 ( 14
-49 ( 4
-5 (16
1.3 ( 0.1
3.1 ( 0.1
-62 ( 14
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
2-F-Ph
3-F-Ph
4-F-Ph
2,4-diF-Ph
3,4-diF-Ph
OMe
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
3-CN-Ph
3-CN-Ph
3-CN-Ph
3-CN-Ph
3-CN-Ph
3-Br-Ph
3-Br-Ph
3-Br-Ph
3-Br-Ph
3-CN-Ph
3-CN-Ph
3-Br-Ph
3-Br-Ph
3-Br-Ph
3-Br-Ph
3-CN-Ph
3-Br-Ph
3-CN-Ph
Ph
n.d.
n.d.
7.9 ( 0.9
7.2 ( 1.6
26 ( 12
23 ( 1
3.4 ( 0.3
4.0 ( 0.4
14 ( 3
-17 ( 10
3.6 ( 9
n.d.
-31 ( 14
0 ( 10
n.d.
-57 ( 7
-15 ( 12
-18 ( 14
4 ( 7
12 ( 12
n.d.
n.d.
-9 ( 6
-26 ( 5
-62 ( 14
-12 ( 7
14 ( 9
-29 ( 10
14 ( 7
17 ( 16
n.d.
n.d.
n.d.
17 ( 7
n.d.
n.d.
41 ( 9
13 ( 4
32 ( 15
42 ( 14
39 ( 5
28 ( 9
32 ( 13
60 ( 10
19 ( 10
3 ( 6
-49 ( 4
2 ( 8
25 ( 10
8 ( 13
66 ( 10
59 ( 11
n.d.
n.d.
205 ( 81
19 ( 1
60 ( 9
302 ( 100
60 ( 8
72 ( 9
54 ( 18
5.7 ( 3.1
36 ( 4
OEt
8.6 ( 1.2
26 ( 1
n.d.
OiPr
OBn
63 ( 23
17 ( 2
OBn
n.d.
n.d.
OEt
36 ( 4
NHMe
NMe2
49 ( 19
55 ( 6
3.5 ( 1.2
7.3 ( 0.2
n.d.
113 ( 12
120 ( 13
7.3 ( 1.6
22 ( 9
27 ( 11
51 ( 1
NHEt
NHiPr
NHEt
NHBn
NHBn
3-CN-Ph
3.8 ( 1.1
7.2 ( 2.0
33 ( 5
79 ( 6
n.d.
51 ( 5
19 ( 7
6.1 ( 0.1
>666
n.d
15 ( 5
>666
>666
a Displacement of [3H] Ro 15-1788 binding from human recombinant GABAA Rxâ3γ2 receptor subtypes (x ) 1, 3 or 5) stably expressed
in mouse fibroblast L(tk-) cells. Ki values are the mean ( SEM of at least three independent determinations. b Primary functional screening
datawere generated using a Voltage/Ion Plate reader assay26,27 at compound concentrations of 300 nM. n.d., not determined.
Table 2. Maximum Inhibition and pEC50 Values of a GABA
EC20 Response Produced by 15 at Human GABAA Receptor
Subtypes Expressed in L(tk-) Cells
resulting from activation of these receptors. However,
there was a window of functional selectivity, and this
provided the opportunity to evaluate this compound
further in our screening cascade.
subunit
R1
R3
R5
maximum
-23 ( 2b
-16 ( 2c
-37 ( 5b
Pharmacokinetics. The pyridazine 15 was evalu-
ated for pharmacokinetic (PK) properties in rat. A low
volume of distribution together with clearance of ap-
proximately half-liver blood flow resulted in a half-life
of 0.6 h. Exposure of pyridazine 15 to isolated rat liver
microsomes showed that 46% of the material was
metabolized in 15 min (Table 3). These data was useful
to correlate with the observed pharmacokinetics. Ad-
ditionally, microsomal turnover in higher species dem-
onstrated that 15 was extensively metabolized in dog,
rhesus and human liver microsomes. If turnover data
is assumed to be a good predictor for pharmacokinetics
in rat, then it may be extrapolated that PK parameters
are likely to be poor in the higher species. Our attempts
to address potential metabolism issues, by incorporation
inhibition (%)a
d
pEC50
-8.53 ( 0.27 -8.02 ( 0.47 -8.13 ( 0.03
a Values are the mean maximum modulation ( SEM from bn
) 4; cn ) 3 concentration-response curves for each receptor
subtype. d Negative logarithm of the concentration that leads to
50% maximal response.
agonist (efficacy ) -52%),13 the full agonist chlor-
diazepoxide (efficacy ) +134%)13 and in house clinical
candidates 213 (efficacy ) -40%) and 415 (efficacy )
-55%). 15 was found to have a weak partial inverse
agonist profile at R1 and R3 subtypes. An ideal profile
at these receptor subtypes would be an antagonist
profile in order to prevent undesirable biological activity

6008 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19
van Niel et al.
of specific [3H]Ro15-1788 binding. From the IC50, Ki was
calculated assuming respective KD values of [3H]Ro15-1788
binding of 0.92, 0.58 and 0.45 nM at the R1â3γ2, R3â3γ2 or
R5â3γ2 receptor subtypes.10
Table 3. Pharmacokinetics and Microsomal Turnover for 15
NADPH dependent turnover
in liver microsomes (%)a
pharmacokineticsb
d
e
rat
dog
rhesus human
Clc
T1/2
Vdss
Functional Efficacy Assays. (a) Voltage Ion Plate
Reader (VIPR) Assay for GABA-A r-3 and r-5 Subtypes.
Primary functional screening data were generated using a
Voltage/Ion Probe Reader assay (VIPR; Aurora Biosciences,
CA) largely as previously described.26,27 Briefly, Ltk- cells
stably expressing GABAA receptor subunit combinations were
seeded into black-sided Porvair 96 well plates, and receptor
expression was induced 24 h prior to experiment with 1 µM
dexamethasone (Sigma, UK). Cells were washed in low-Cl-
buffer (160 mM sodium D-gluconate, 4.5 mM potassium
D-gluconate, 2 mM CaCl2, 1 mM MgCl2, 10 mM D-glucose, 10
mM HEPES pH 7.4) and dye-loaded for 30 min to give final
concentrations of 4 µM chlorocoumarin-2-dimyristoyl phos-
phatidylethanolamine (CC2-DMPE; FRET donor) (Aurora Bio-
sciences Corp, CA) and 1 µM bis(1,3-diethyl-2-thiobarbiturate)-
trimethineoxonol (DiSBAC2(3); FRET acceptor) (Molecular
Probes), with 0.5 mM tartrazine (Sigma, UK) present extra-
cellularly. Plate preparation was automated using a CCS-
Packard Platetrak. Plates were then placed in a VIPR which
performs automated liquid additions and records fluorescence
emission from eight wells. A 400DF15 filter was used in the
excitation pathway, and 460DF45 and 580DF60 filters were
used in the dual emission pathways. Rapid ratiometric FRET
measurements were made of GABA-evoked depolarizations as
previously described, and the ability of compounds to modulate
an EC50 response to GABA was examined. Basal fluorescence
was read for 8s before addition of test compounds, and a half-
maximal concentration of GABA was added 22s later, with
fluorescence emissions recorded at 1 Hz.
For each time-point, background fluorescence was sub-
tracted (recorded from wells without cells in the same plate)
and the ratio of fluorescence at 460 nm to 580 nm was
calculated. GABA-evoked depolarizations were then expressed
as a fractional change in this ratio. In house algorithms written
as Excel 97 (Microsoft Corp.) macros were used for calculation
of fluorescence ratio and GABA responses. All compound plates
contained an in-house inverse agonist at the R5 receptor, a
benzodiazepine site agonist (alprazolam) and a nonselective
benzodiazepine site inverse agonist (DMCM). Test compound
efficacy was normalized to that of DMCM on the same plate,
which was always taken as -100%. Compound efficacy was
measured at these receptors at 300 nM to minimize direct or
fluorescence effects of some compounds at higher concentra-
tions and was derived from at least three plates/day screened
over a minimum of 2 days.
(b) Whole Cell Patch-Clamp of L(tk-) Cells Stably
Transfected with Human GABA-A Receptors. Experi-
ments were performed on L(tk-) cells expressing human cDNA
combinations R1â3γ2s, R3â3γ2s and R5â3γ2s. Glass cover-slips
containing the cells in a monolayer culture were transferred
to a Perspex chamber on the stage of Nikon Diaphot inverted
microscope. Cells were continuously perfused with a solution
containing 124 mM NaCl, 2 mM KCl, 2 mM CaCl2, 1 mM
MgCl2, 1.25 mM KH2PO4, 25 mM NaHCO3, 11 mM D-glucose,
at pH 7.2, and observed using phase-contrast optics. Patch-
pipets were pulled with an approximate tip diameter of 2 µm
and a resistance of 4 MΩ with borosilicate glass and filled with
130 mM CsCl, 10 mM HEPES, 10 mM EGTA, 3 mM Mg+-
ATP, pH adjusted to 7.3 with CsOH. Cells were patch-clamped
in whole-cell mode using an Axopatch-200B patch-clamp
amplifier. Drug solutions were applied by a double-barreled
pipet assembly, controlled by a stepping motor attached to a
Prior manipulator, enabling rapid equilibration around the
cell. Increasing GABA concentrations were applied for 5 s
pulses with a 30 s interval between applications. Curves were
fitted using a nonlinear square-fitting program to the equation
f(x) ) Bmax/[1 + (EC50/x)n] where x is the drug concentration,
EC50 is the concentration of drug eliciting a half-maximal
response, and n is the Hill coefficient. Allosteric modulation
of GABA receptors was measured relative to a GABA EC20,
15 46 ( 8 86 ( 1 89 ( 1 85 ( 0
36
0.6
1.5
a n ) 3, mean ( SD. b 1 mg/kg in a single rat of a 3:1 PEG 300:
water solution given iv (1 mL/kg). c mL/min/kg from plasma.
d Hours, calculated from 1 to 3 h data. e L/kg.
of fluorine substituents, provided limited access to
analogues with both the desired in vitro binding and
functional efficacy profiles. Our conclusions from pro-
gressing 15 down the screening cascade were that there
are number of challenges which need to be addressed
in developing the SAR of this lead.
Conclusions
We have described how screening hit 6 was identified
as a pyridazine with moderate affinity for the R5
receptor subtype of the GABAA ligand-gated ion chan-
nel. Initial SAR resulted in the identification of the low
molecular weight, structurally simple 3,5-diphenyl-
pyridazine (7) which retained respectable binding af-
finity and was an inverse agonist at R5 receptors. This
lead was explored using parallel synthesis which re-
sulted in the identification of compounds such as 15 and
23 with improved binding affinity which were more
efficacious inverse agonists at R5 receptors. The iden-
tification of functionally selective leads such as 15 may
be useful in further delineating the complex biology of
the GABA-A receptor complex in vivo. Our observations
that simple, low molecular weight structures could
provide functionally selective ligands were unexpected
when we started to evaluate the original screening hit.
The results reported above encouraged us to carry out
additional extensive studies, the results of which will
be described in a following paper.
Experimental Section
Radioligand Binding Assay at the Benzodiazepine
Site. The binding affinities of synthesized molecules for
various subtypes of GABAA receptors were determined by
automated radioligand binding.10 Membrane pellets, prepared
from Ltk- cells expressing GABAA receptors of subunit com-
position R1â3γ2, R3â3γ2 or R5â3γ2, were resuspended in
different volumes of ice-cold 50 mM Tris assay buffer, depend-
ing on the total cpm they gave when batch tested, using a
Polytron for 10 s. Test compounds were prediluted from 2 mM
stocks to 100 µM in DMSO followed by a final dilution into
buffer, and the incubation was started with the addition of
membranes. Inhibition of specific [3H]Ro15-1788 binding
(specific activity 78.6Ci/mmol, NEN) was measured for 30 min
at 24 °C in 50 mM Tris buffer, pH 7.4. The final concentration
of radioligand was 1.8 nM. Nonspecific binding was defined
with 10 µM flunitrazepam (Sigma, UK), and total binding was
determined in the presence of 0.1% DMSO. The assays were
performed using a Sagian Robotic system. The assay was
terminated by filtration over prewetted Unifilter GFB plates
(Packard, UK) using an automated Brandel harvestor, wash-
ing with Tris buffer. Filter plates were dried for 4-6 h at room
temperature, and the robot then added 50 mL of Microscint
‘O’ to each well of the filter plate, incubating for 1 h. Following
this, plates were counted for 1 min/well in a TriLux to detect
filter-retained radioactivity. Results were generated using
ActivityBase and the XLfit curve fitting program for Microsoft
Excel (IDBS, Guildford, Surrey) and expressed as % inhibition

Pyridazine Series of GABAA R5 Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6009
individually determined for each cell to account for differences
in GABA affinity.
Purity for target compounds was g95% as judged by NMR and
HPLC. Representative procedures for the synthesis of com-
pounds 15, 23, 29, and 36 are given below. Alternative
procedures to 5-(3-bromophenyl)-(2H)-pyridazin-3-one (11, Ar
) 3-Br-Ph), 5-(3-bromophenyl)-3-chloropyridazine (12, Ar )
3-Br-Ph) and 3-(6-ethylaminopyridazin-4-yl)-benzonitrile (38)
is available in Supporting Information. 2,4-Diphenylpyridine,
3,5-diphenylpyridine, 4,6-diphenylpyrimidine and 3,5-diphen-
yl-1,2,4-triazine are commercially available. 4,6-Diphenyl-
1,2,3-triazine was prepared using literature procedures.28
Stability in Liver Microsomes. Test compound at a
concentration of 1 µM in pH 7.4 buffer containing 1% DMSO
was incubated at 37 °C for 15 min with liver microsomes (0.5
mg/mL protein concentration) in the presence or absence of
NADPH (1 mM). The reaction was quenched by the addition
of MeCN and after vortex mixing and centrifugation, samples
were analyzed by LC-MS/MS (HP1050/CTC autosampler;
KR100 5C18 column, 50 mm × 2.1 mm i.d.; with gradient
elution using MeCN in 25 mM ammonium formate pH 3; flow
0.4 mL/min; injection volume 20 µL). A Micromass Quattro
LC was used as the mass spectrometer. Turnover was calcu-
lated by comparing the peak areas for incubations performed
in the presence of NADPH with those performed in the absence
of NADPH.
Pharmacokinetic Determinations. The jugular vein of
a male Sprague Dawley rat (approximate weight 300 g) was
cannulated under isofluorane anaesthesia. The rat was treated
with a 100 unit dose of heparin (0.1 mL, 1000 units/mL) via
the cannula and then allowed to recover for 48 h. The animal
was deprived of food overnight and then given a 1 mg/kg dose
of the test compound iv, via bolus injection. The test compound
was formulated as a solution (1 mg/mL) in a PEG 300:H2O
(3:1). Serial blood samples (600 µL) were collected into hep-
arinized tubes at several time points up to 6 h after dosing.
Plasma was harvested from the blood samples by centrifuga-
tion, and the samples were stored at -80 °C until analysis.
To 10 µL aliquots of 1 ng/µL internal standard solution
(diazepam) on a conical bottomed 96-well plate were added
plasma (50 µL) and MeCN (150 µL). The plate was sealed,
mixed and centrifuged. A portion of the supernatant (100 µL)
was added to 25 mM ammonium formate pH3 (50 µL) on a
fresh 96-well plate. The resulting samples were analyzed using
the same LC-MS/MS method described above. Detection of test
compound was achieved by MRM ESP+ using the transition
258.3 to 77.0. Quantitation of test compound was achieved by
comparing peak areas of samples against a calibration curve
obtained by spiking control rat plasma with known amounts
of test compounds. All pharmacokinetic analysis was model
independent and used standard formulas in a Microsoft Excel
spreadsheet.
General Chemical Experimental Details. 1H NMR
spectra were recorded on Bruker AMX500, AV500, DPX400
or DPX360 spectrometers. Chemical shifts are reported in
parts per million (δ) downfield from tetramethylsilane as
internal standard and coupling constants in Hertz. Melting
points were determined on a Reichert Thermovar hot stage
apparatus and are uncorrected. Mass spectra were recorded
using a HP1100 flow inject system (50:50 1% formic acid-
H2O:MeCN; 30 s run; flow rate 0.3 mL/min) into an electro-
spray source of a Waters ZMD2000. Spectra were recorded
between 100 and 800 Da. Standard abbreviations are used for
common solvents. During workup procedures, organic solvents
were dried over MgSO4 prior to concentration. Anhydrous
THF, DMF, Et2O, MeOH, and toluene were purchased from
the Aldrich Chemical Co. EtOH refers to absolute EtOH. Flash
chromatography and dry flash chromatography were per-
formed on silica gel Merck Kieselgel (230-400) mesh and
Merck 15111, respectively. Thin-layer chromatography (TLC)
was carried out on Merck 5 × 10 cm plates with silica gel 60
F254 as sorbant. Preparative thin-layer chromatography was
carried out using 20 × 20 cm silica gel GF tapered plates
supplied by Analtech Inc, Newark. Phase separation cartridges
were supplied by Whatman (12 mL volume with 5 µM pore
size PTFE membrane). Mass triggered preparative LC-MS was
run on a Waters Fraction Lynx running 10 min gradients (20
mL/min, 0.1% TFA-H2O/MeCN) through a Supelco ABZ+
column. Microanalysis was determined by Butterworth Labo-
ratories, 54-56 Waldegrave Road, Teddington, U.K. Micro-
wave reactions were carried out, unless indicated, using a
Smith Synthesizer supplied by Personal Chemistry, Uppsala,
Sweden. Details of HPLC conditions used to analyze target
compounds is provided in the Supporting Information section.
Representative Stille and Suzuki Couplings for
Synthesis of 3-Phenyl-5-arylpyridazines (13-22). 3-(6-
Phenylpyridazin-4-yl)benzonitrile (15). (a) 5-(Tri-n-butyl-
stannyl)-3-phenyl pyridazine20 (8) (100 mg, 0.22 mmol), 3-bromo-
benzonitrile (48 mg, 0.27 mmol) and palladium(0) tetrakis-
(triphenylphosphine) (5 mg) in 4 mL of THF were combined
and heated to 150 °C for 600 s in microwave reactor. The
reaction was diluted with 6 mL of CH2Cl2 and 2 mL of H2O
then poured into a phase separation cartridge. The organic
phase was collected and concentrated to leave ∼100 mg of
crude product. Part of the sample was purified by HPLC with
mass triggered fraction collection to give (15) (2.2 mg).
Purification could also be carried out by preparative thin-layer
chromatography using 1:1 EtOAc-hexanes as eluent or re-
i
crystallization from MeOH or PrOH.
(b) 5-Iodo-3-phenylpyridazine20 (9) (480 mg, 1.17 mmol),
3-cyanophenylboronic acid (200 mg, 1.17 mmol), and pal-
ladium(0) tetrakis(triphenylphosphine) (40 mg) in 4 mL of THF
and 0.5 mL of 2 N Na2CO3 were reacted and purified as
described above to give (15) (84 mg, 28%). δH (400 MHz,
DMSO-d6) 7.57-7.64 (3H, m), 7.81 (1H, t, J ) 8 Hz), 8.03 (1H,
d, J ) 8 Hz), 8.30-8.35 (2H, m), 8.41 (1H, d, J ) 8 Hz), 8.59
(1H, d, J ) 4 Hz), 8.63 (1H, s), 9.69 (1H, d, J ) 4 Hz). m/z
(ES+) 258 (MH+). Mp. 170-171 °C. Anal. (C17H11N3) C, H, N.
Representative Synthesis of 3,5-Diarylpyridazines
(23-27, 41). 3-[6-(2-Fluorophenyl)-pyridazin-4-yl]benzo-
nitrile (23). (a) 5-Chloropyridazin-3(2H)-one (10) (2.0 g, 15.3
mmol), dichloro-(1,4-bis(diphenylphosphino)butane) (0.1 g, 0.16
mmol), 3-cyanophenylboronic acid (2.71 g, 18.4 mmol) and
saturated Na2CO3 solution (5 mL) in 20 mL of toluene were
combined and heated to 120 °C for 15 min in a Milestone
microwave reactor. The reaction was filtered and the solid
i
washed with PrOH (20 mL) then dried to give 5-(3-cyano-
phenyl)pyridazine-3-one (11, Ar ) 3-CN-Ph) as a yellow solid
(1.8 g, 50%). δH (400 MHz, DMSO-d6) 7.20 (1H, m), 7.79 (1H,
t, J ) 8 Hz), 8.05 (1H, m), 8.33 (1H, m), 8.54 (1H, m), 8.37
(1H, m), 13.1 (1H, br s). m/z (ES+)197 (MH+).
(b) 5-(3-Cyanophenyl)pyridazin-3-one (11, Ar ) 3-CN-Ph)
(1.8 g, 9.1 mmol) was suspended in phosphorus oxychloride
(15 mL, 12 mmol) and heated to 100 °C for 1 h. The reaction
was cooled to room temperature and concentrated under
reduced pressure before being poured onto ice. The pH was
adjusted to 8 by portionwise addition of saturated Na2CO3
solution, and the resultant solid was collected by filtration,
washed with H2O, and dried to give 3-chloro-5-(3-cyanophenyl)-
pyridazine (12, Ar ) 3-CN-Ph) (1.5 g, 76%). δH (400 MHz,
DMSO-d6) 7.79 (1H, t, J ) 8 Hz), 8.05 (1H, m), 8.33 (1H, m),
8.41 (1H, d, J ) 2 Hz), 8.54 (1H, m), 9.75 (1H, m). m/z (ES+)
216, 218 (MH+).
(c) 3-Chloro-5-(3-cyanophenyl)pyridazine (12, Ar ) 3-CN-
Ph) (0.1 g, 0.46 mmol), 2-fluorophenyl boronic acid (80 mg,
0.572 mmol), palladium(0) tetrakis(triphenylphosphine) (25
mg) and saturated Na2CO3 solution (0.5 mL) in 1.5 mL of THF
were combined and heated to 150 °C for 600 s in a microwave
reactor. The reaction was diluted with 6 mL of CH2Cl2 and 2
mL of saturated NH4Cl then poured into a phase separation
cartridge. The organic phase was concentrated and purified
by column chromatography using 30 to 40% EtOAc/hexanes
as eluent to give (23) as a colorless solid (65 mg, 51%). δH (400
MHz, DMSO-d6) 7.43-7.48 (2 H, m), 7.60-7.66 (1 H, m), 7.80
(1 H, t, J ) 7.8 Hz), 7.98-8.02 (1 H, m), 8.02-8.04 (1 H, m),
8.34-8.37 (1 H, m), 8.42 (1H, t, J ) 2.0 Hz), 8.56-8.57 (1 H,
m), 9.75 (1 H, d, J ) 2.0 Hz). m/z (ES+) 276 (MH+).

6010 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19
van Niel et al.
(6) Crestani, F.; Keist, R.; Fritschy, J.-M.; Benke, D.; Vogt, K.; Prut,
L.; Bluthmann, H.; Mohler, H.; Rudolph, U. Trace fear condi-
tioning involves hippocampal R5 GABAA receptors. Proc. Natl.
Acad. Sci. U.S.A. 2002, 99 (13), 8980-8985.
Representative Synthesis of 3-Alkoxy-5-arylpyridazines
(28-31). 5-(3-Bromophenyl)-3-ethoxy pyridazine (29). (a)
Sodium metal (0.13 g, 5.65 mmol) was added portionwise to
dry ethanol (10 mL) using external H2O bath cooling. On
addition of all the sodium, the reaction was allowed to attain
room temperature and the solution stirred for 1 h before
adding 5-(3-bromophenyl)-3-chloropyridazine (12, Ar ) 3-Br-
Ph) (0.15 g, 0.557 mmol) in EtOH (5 mL). The suspension was
stirred for 18 h before adding H2O and removing the EtOH in
vacuo. The products were extracted into CH2Cl2, concentrated,
and purified by preparative TLC using 25% EtOAc/hexanes
as eluent to give (29) as a colorless solid (30 mg, 19%). δH (400
MHz, DMSO-d6) 1.41 (3 H, t, J ) 7.0 Hz), 4.55 (2 H, q, J ) 7.0
Hz), 7.50 (1 H, t, J ) 8.0 Hz), 7.55 (1 H, d, J ) 2.0 Hz), 7.71-
7.74 (1 H, m), 7.91-7.94 (1 H, m), 8.14 (1 H, t, J ) 1.8 Hz),
9.30 (1 H, d, J ) 2.0 Hz). 1H. m/z (ES+) 279, 281 (MH+).
Representative Synthesis of 3-Alkylamino-5-aryl-
pyridazines (34-40). [5-(3-Bromophenyl)-N-pyridazin-3-
yl]ethylamine (36). To 5-(3-bromophenyl)-3-chloropyridazine
(12, Ar ) 3-Br-Ph) (1.2 g, 4.46 mmol) was added to a 2 M
solution of ethylamine in THF (7 mL) and heated to 170 °C
for 1 h in a microwave reactor. The organic phase was diluted
with EtOAc, washed with H2O, dried and concentrated while
loading onto silica. Column chromatography using a gradient
of 20% EtOAc/hexanes to EtOAc as eluent gave (36) as an
orange solid (0.45 g, 36%). δH (360 MHz, DMSO-d6) 1.20 (3 H,
t, J ) 7.0 Hz), 3.41 (2 H, qd, J ) 5.5, 7.0 Hz), 6.84 (1 H, t, J
) 5.5 Hz), 6.99 (1 H, d, J ) 2.0 Hz), 7.49 (1 H, t, 7.8 Hz), 7.68-
7.71 (1 H, m), 7.74-7.77 (1 H, m), 7.95 (1 H, t, J ) 1.8 Hz),
8.79 (1 H, d, J ) 2.0 Hz). m/z (ES+) 278, 280 (MH+).
(7) Whiting, P. J. GABA-A receptor subtypes in the brain:
a
paradigm for CNS drug discovery? Drug Discovery Today 2003,
8 (10), 445-450.
(8) Whiting, P. J.; Wafford, K. A.; McKernan, R. M. Pharmacologic
subtypes of GABAA receptors based on subunit composition.
GABA Nerv. Sys. 2000, 113 (126), 113-116.
(9) Quirk, K.; Gillard, N. P.; Ragan, C. I.; Whiting, P. J.; McKernan,
R. M. Model of subunit composition of gamma-aminobutyric acid
A receptor subtypes expressed in rat cerebellum with respect to
their alpha and gamma/delta subunits. J. Biol. Chem. 1994, 269,
16020-16028.
(10) Hadingham, K. L.; Wingrove, P.; Le Bourdelles, B.; Palmer, K.
J.; Ragan, C. I.; Whiting, P. J. Cloning of cDNA sequences
encoding human R-2 and R3 γ-aminobutyric acid A receptor
subtypes and characterisation of the benzodiazepine pharmacol-
ogy of recombinant of R1-, R2-, R3 and R5-containing human
γ-aminobutyric acid A receptors. Mol. Pharmacol. 1993, 43, 970-
2975.
(11) Cooke, A. J.; Hamilton, N. M. Expert Opin. Ther. Pat. 2002, 12
(10), 1491-1501.
(12) Chambers, M. S.; Atack, J. R.; Broughton, H. B.; Collinson, N.;
Cook, S.; Dawson, G. R.; Hobbs, S. C.; Marshall, G.; Maubach,
K. A.; Pillai, G. V.; Reeve, A. J.; MacLeod, A. M. Identification
of a novel, selective GABAA R5 receptor inverse agonist which
enhances cognition. J. Med. Chem. 2003, 46 (11), 2227-
2240.
(13) Sternfeld, F.; Carling, R. W.; Jelley, R. A.; Ladduwahetty, T.;
Merchant, K. J.; Moore, K. W.; Reeve, A. J.; Street, L. J.;
O’Connor, D.; Sohal, B.; Atack, J. R.; Cook, S.; Seabrook, G.;
Wafford, K.; Tattersall, F. D.; Collinson, N.; Dawson, G. R.;
Castro, J. L.; MacLeod, A. M. Selective, orally active γ-amino-
butyric acid-A R5 receptor inverse agonists as cognition enhanc-
ers. J. Med. Chem. 2004, 47 (9), 2176-2179.
(14) Street, L. J.; Sternfeld, F.; Jelley, R. A.; Reeve, A. J.; Carling,
R. W.; Moore, K. W.; McKernan, R. M.; Sohal, B.; Cook, S.; Pike,
A.; Dawson, G. R.; Bromidge, F. A.; Wafford, K. A.; Seabrook,
G. R.; Thompson, S. A.; Marshall, G.; Pillai, G. V.; Castro, J. L.;
Atack, J. R.; MacLeod, A. M. Synthesis and biological evaluation
of 3-heterocyclyl-7,8,9,10-tetrahydro-(7,10-ethano)-1,2,4-triazolo-
[3,4-a]phthalazines and analogues as subtype-selective inverse
agonists for the GABAAR5 benzodiazepine binding site. J. Med.
Chem. 2004, 47 (14), 3642-3657.
(15) Chambers, M. S.; Atack, J. R.; Carling, R. W.; Collinson, N.;
Cook, S. M.; Dawson, G. R.; Ferris, P.; Hobbs, S. C.; O’Connor,
D.; Marshall, G.; Rycroft, W.; MacLeod, A. M. An orally bio-
available, functionally selective inverse agonist at the benzodi-
azepine site of GABAA R5 receptors with cognition enhancing
properties. J. Med. Chem. 2004, 47 (24), 5829-5832.
(16) Komendantov, M. I.; Bekmukhametov, R. R. 1,3-Dipolar cyclo-
addition of diazomethane to the cyclopropene double bond, and
new method for synthesizing bicyclobutane derivatives. Zh. Org.
Khim. 1971, 7 (2), 423-424.
Acknowledgment. We thank Philip Jones, Mat-
thew Lindon, Elizabeth Norris, Jonathan Rose and Paul
Scott-Stevens for their contributions to this work.
Supporting Information Available: NMR and mass
spectral data for compounds 13, 14, 16-22, 24-28, 30-35, 37,
39-41, microanalysis data for 15 and HPLC data for 13, 14,
16-41. Alternative procedures to 5-(3-bromophenyl)-(2H)-
pyridazin-3-one (11, Ar ) 3-Br-Ph), 5-(3-bromophenyl)-3-
chloropyridazine (12, Ar ) 3-Br-Ph) and 3-(6-ethylamino-
pyridazin-4-yl)benzonitrile (38). This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(17) Huber, M. K.; Martin, R.; Rey, M.; Dreiding, A. S. The aminal-
enamine equilibrium in cyclopropane carbaminals. Helv. Chim.
Acta 1977, 60 (5), 1781-1800.
(18) Komendantov, M. I.; Bekmukhametov, R. R.; Novinskii, V. G.
1,3-Dipolar cycloaddition of diazomethane to the cyclopropene
double bond. New method for the synthesis of bicyclobutane
derivatives. Zh. Org. Khim. 1976, 12 (4), 801-805.
(19) Goodnow, R. Small molecule lead generation processes for drug
discovery. Drugs Future 2002, 27 (12), 1165-1180.
(20) Sauer, J.; Heldman, D. K. Synthesis of 3,5-disubstituted
pyridazines by regioselective [4+2] cycloadditions with ethynyl-
tributyltin and subsequent replacement of the organotin sub-
stituent. Tetrahedron 1998, 54 (17), 4297-4312.
(21) Oprea, T. I.; Davis, A. M.; Teague, S. J.; Leeson, P. D. Is There
a difference between leads and drugs? A historical perspective.
J. Chem. Inf. Comput. Sci. 2001, 41, 1308-1315.
(22) Hann, M. M.; Leach, A. R.; Harper, G. Molecular complexity and
its impact on the probability of finding leads for drug discovery.
J. Chem. Inf. Comput. Sci. 2001, 41, 856-864.
(23) The use of microwaves in synthesis is gaining momentum with
the introduction of sophisticated reactors that enable multiple
reactions to be run safely, with short reaction times: Lidstrom,
P.; Tierney, J.; Wathey, B.; Westman, J. Microwave assisted
organic synthesis- a review. Tetrahedron 2001, 57, 9225-
9283.
(1) McNamara, R. K.; Skelton, R. W. The neuropharmacological and
neurochemical basis of place learning in the Morris water maze.
Brain Res. Rev. 1993, 18, 33-49. DeLorey, T. M.; Lin, R. C.;
McBrady, B.; He, X.; Cook, J. M.; Lameh, J.; Loew, G. H.
Influence of benzodiazepine binding site ligands on fear-
conditioned contextual memory. Eur. J. Pharmacol, 2001, 426
(1, 2), 45-54. Bailey, D. J.; Tetzlaff, J. E.; Cook, J. M.; He, X.;
Helmstetter, F. J. Effects of hippocampal injections of a novel
ligand selective for the R5â2γ2 subunits of the GABA/benzo-
diazepine receptor on Pavlovian conditioning. Neurobiol. Learn.
Mem. 2002, 78 (1), 1-10. Liu, R.; Hu, R. J.; Zhang, P.; Skolnick,
P.; Cook, J. M. Synthesis and pharmacological properties of novel
8-substituted imidazobenzodiazepines: high-affinity, selective
probes for alpha 5-containing GABAA receptors. J. Med. Chem.
1996, 39 (9), 1928-34.
(2) Whiting, P. J. The GABA-A receptor gene family: new targets
for therapeutic intervention. Neurochem. Int. 1999, 34 (5), 387-
390.
(3) Pritchett, D. B.; Sontheimer, H.; Shivers, B. D.; Ymer, S.;
Kettenmann, H.; Schofield, P. R.; Seeburg, P. H. Importance of
a novel GABAA receptor subunit for benzodiazepine pharmacol-
ogy. Nature 1989, 338, 582-585.
(4) Wisden, W.; Laurie, D. J.; Monyer, H.; Seeburg, P. H. The
distribution of 13 GABAA receptor subunit mRNA’s in the rat
brain. J. Neurosci. 1992, 12, 1040-1062.
(24) The selection of the commercially available aryl halides was
made by searching the Available Chemicals Directory (ACD) for
all mono-bromo or chloro substituted phenyl derivatives that did
not contain additional reactive functional groups. The list was
further reduced by considering only preferred suppliers and
using a molecular weight <300. This set of structures was then
(5) Collinson, N.; Kuenzi, F. M.; Jarolimek, W.; Maubach, K. A.;
Cothliff, R.; Sur, C.; Smith, A.; Otu, F. M.; Howell, O.; Atack, J.
R.; McKernan, R. M.; Seabrook, G. R.; Dawson, G. R.; Whiting,
P. J.; Rosahl, T. W. Enhanced learning and memory and altered
GABAergic synaptic transmission in mice lacking the R5 subunit
of the GABAA receptor. J. Neurosci. 2002, 22 (13), 5572-5580.

Pyridazine Series of GABAA R5 Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6011
clustered into 10 clusters using k-means clustering on topological
similarities calculated from our in-house 2D descriptors: Kears-
ley, S. K.; Sallamack, S.; Fluder, E. M.; Andose, J. D.; Mosley,
R. T.; Sheridan, R. P. J. Chem. Inf. Comput. Sci. 1996, 36, 118.
Any cluster with more than five-members was itself clustered
and so forth in a recursive fashion. A set of 50 aryl bromides
was then selected from across the clusters and the similarity
matrix of the final selection was inspected to confirm that a
diverse set of reagents had been chosen. This set was used in
the first parallel run and subsequent iterative parallel runs used
other members of the reagent sets when the first reagent used
was demonstrated to lead to high affinity pyridazines.
(25) Francisco, F. P.; Celia, E. G. 3(2H)-Pyridazinones. ES 454136,
1977.
(26) Adkins, C. E.; Pillai, G. V.; Kerby, J.; Bonnert, T. P.; Haldon,
C.; McKernan, R. M.; Gonzalez, J.; Oades, K.; Whiting, P. J.;
Simpson, P. B. R4â3δ GABAA receptors characterized by fluo-
rescence resonance energy transfer-derived measurements of
membrane potential. J. Biol. Chem. 2001, 276, 38934-38939.
(27) Smith A. L.; Oxley, B.; Malpas, S.; Pillai, G. V.; Simpson, P. B.
Compounds exhibiting selective efficacy for different â subunits
of human recombinant GABAA receptors. J. Pharmacol. Exp.
Ther. 2004, 311, 601-609.
(28) Ohsawa, A.; Arai, H.; Ohnishi, H.; Itoh, T.; Kaihoh, T.; Okada,
M.; Igeta, H. Oxidation of 1-aminopyrazoles and synthesis of
1,2,3-triazines. J. Org. Chem. 1985, 50 (26), 5520-5523.
JM050249X