3-Heteroaryl-2-pyridones: Benzodiazepine site ligands with functional selectivity for α2/α3-subtypes of human GABAa receptor-ion channels

Article abstract of DOI:10.1021/jm0110789

A novel series of 3-heteroaryl-5,6-bis(aryl)-1-methyl-2-pyridones were developed with high affinity for the benzodiazepine (BZ) binding site of human γ-aminobutyric acid (GABA_A) receptor ion channels, low binding selectivity for α2- and/or α3- over α1-containing GABA_A receptor subtypes and high binding selectivity over α5 subtypes. High affinity appeared to be associated with a coplanar conformation of the pyridone and sulfur-containing 3-heteroaryl rings resulting from an attractive S...O intramolecular interaction. Functional selectivity (i.e., selective efficacy) for α2 and/or α3 GABA_A receptor subtypes over α1 was observed in several of these compounds in electrophysiological assays. Furthermore, an α3 subtype selective inverse agonist was proconvulsant and anxiogenic in rodents while an α2/α3 subtype selective partial agonist was anticonvulsant and anxiolytic, supporting the hypothesis that subtype selective BZ site agonists may provide new anxiolytic therapies.

Full text of DOI:10.1021/jm0110789

J . Med. Chem. 2002, 45, 1887-1900
1887
3-Heter oa r yl-2-p yr id on es: Ben zod ia zep in e Site Liga n d s w ith F u n ction a l
Selectivity for r2/r3-Su btyp es of Hu m a n GABAA Recep tor -Ion Ch a n n els
Ian Collins,* Christopher Moyes, William B. Davey, Michael Rowley, Frances A. Bromidge, Kathleen Quirk,
J ohn R. Atack, Ruth M. McKernan, Sally-Ann Thompson, Keith Wafford, Gerard R. Dawson, Andrew Pike,
Bindi Sohal, Nancy N. Tsou,† Richard G. Ball,† and J ose´ L. Castro
Merck Sharp & Dohme Research Laboratories, The Neuroscience Research Centre, Terlings Park, Eastwick Road,
Harlow, CM20 2QR, United Kingdom
Received October 29, 2001
A novel series of 3-heteroaryl-5,6-bis(aryl)-1-methyl-2-pyridones were developed with high
affinity for the benzodiazepine (BZ) binding site of human γ-aminobutyric acid (GABAA) receptor
ion channels, low binding selectivity for R2- and/or R3- over R1-containing GABAA receptor
subtypes and high binding selectivity over R5 subtypes. High affinity appeared to be associated
with a coplanar conformation of the pyridone and sulfur-containing 3-heteroaryl rings resulting
from an attractive S...O intramolecular interaction. Functional selectivity (i.e., selective efficacy)
for R2 and/or R3 GABAA receptor subtypes over R1 was observed in several of these compounds
in electrophysiological assays. Furthermore, an R3 subtype selective inverse agonist was
proconvulsant and anxiogenic in rodents while an R2/R3 subtype selective partial agonist was
anticonvulsant and anxiolytic, supporting the hypothesis that subtype selective BZ site agonists
may provide new anxiolytic therapies.
In tr od u ction
continuum of ligand activities is possible, including
antagonists (neutral allosteric modulators), which have
no effect on the ion channel response to GABA but will
competitively displace other BZ site ligands from the
receptor. These different efficacies are reflected in the
behavioral effects of BZ site ligands. For example,
agonists show anxiolytic effects whereas inverse ago-
nists are anxiogenic, and antagonists have no effect on
anxiety behavior. Classical BZ drugs for the treatment
of anxiety, such as diazepam, are full agonists at the
BZ site of the GABAA receptor with no subtype selectiv-
ity. Although clinically useful, such compounds often
exhibit undesirable effects,5 particularly sedation, ataxia,
and potentiation with alcohol. There is also a risk of
tolerance and dependence developing with chronic use.
The need for improved anxiolytic therapies is therefore
clear, and potentially fruitful research has grown from
an increased understanding of the molecular biology of
GABAA receptors.6,7
Neurotransmission by γ-aminobutyric acid (GABA) is
a major inhibitory mechanism in the human central
nervous system (CNS). Given the ubiquity of GABA-
ergic pathways in the CNS, it is unsurprising that
pharmacological intervention in this system can elicit
diverse responses, ranging from changes in motor
activity, muscle tone, and seizure behavior, through
sedation and hypnotic effects, to modulation of cognition
and mood, particularly anxiety states.1 Of the three
2
pharmacological classes of GABA receptors, GABAA
3
and GABAC are ligand gated chloride ion channels,
whereas GABAB is a G-protein linked receptor. Postsyn-
aptic GABAA ion channels open in response to the
binding of GABA, and the ensuing chloride flux hyper-
polarizes the membrane of the postsynaptic neurone
inhibiting further neuronal activity. Recently, subtypes
of the GABAA receptors have been identified, as dis-
cussed in detail below. In addition to the GABA binding
site, allosteric modulatory sites for numerous ligands
have been identified on GABAA receptors (benzodiaze-
pines (BZs), barbiturates, ethanol, steroids, avermectins,
picrotoxin, zinc cations, and loreclezole), of which the
BZ site has historically been of the most pharmacologi-
cal interest.4
Ligands at the BZ binding site of GABAA receptors
fall into functional classes categorized by their effect on
the ion current elicited by the endogenous neurotrans-
mitter GABA: Agonists (positive allosteric modulators)
give an increased ion flux, resulting in a greater
inhibitory effect on the postsynaptic neurone, while
inverse agonists (negative allosteric modulators) cause
a decrease in the ion flux in response to GABA. A
GABAA receptors are oligomeric assemblies of a large
range of subunits (R1-6, â1-3, γ1-3, δ, ꢀ, π, and θ). In
vitro, the pharmacology of the natural receptors can be
replicated with pentameric combinations of cloned
receptor subtypes, provided at least one of each of the
R, â, and γ subunits is present, and it is currently
thought that the dominant stoichiometry in vivo is (R)2-
(â)2(γ).7,8 The subtypes R1ânγ2 correspond pharmaco-
logically to the GABAA receptors previously classified
as BZ-I (or ω1), while R2ânγ2, R3ânγ2, and R5ânγ2 have
been identified as comprising the BZ-II (or ω2) class.
Subtypes containing the R4 and R6 subunits are insen-
sitive to BZs. The γ2 subunit is the predominant γ
subunit in the brain. The nature of the â subunits in
the receptor complex does not affect the BZ pharmacol-
ogy. Although characterization of native GABAA recep-
tor subtypes is far from complete, through in situ mRNA
hybridization, subunit specific immunoprecipitation,
* To whom correspondence should be addressed. Tel.: (+1279)-
440000. Fax: (+1279)440187. E-mail: ian_collins@merck.com.
† Merck Research Laboratories, POB 2000, Rahway, NJ 07065.
10.1021/jm0110789 CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/29/2002

1888 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Collins et al.
Sch em e 1
and immunoaffinity chromatography, significant progress
has been made in determining the distribution of the
individual subunits in the brain, and from these data,
the distribution of the GABAA subtypes has been
inferred:7,8 While the dominant R1â2γ2 subtypes are
present in both cerebellum and cortex, the R2ânγ2 and
R3ânγ2/3 assemblies are found mainly in the cortex and
hippocampus, and the R5â3γ2 and R5â3γ3 are largely
confined to the hippocampus. The differential localiza-
tion of subtypes within the brain suggests that different
subtypes may be associated with different physiological
processes, and this has been observed in transgenic mice
where the R1 subunit was rendered BZ insensitive.9 In
these mice, the sedative and amnesic properties of BZ
full agonists were abolished; yet, the anxiolytic, myo-
relaxant, and partial anticonvulsant effects were re-
tained, clearly identifying selective mediation of the
sedative response through R1 subunits.10
In light of this, subtype selective ligands should
discriminate between the many behaviors mediated by
GABA-ergic transmission, and this is supported by data
obtained on the limited number of selective compounds
known. For example, while unselective full agonist BZs
such as diazepam are anxiolytics with sedative side
effects apparent at higher doses, the BZ site full agonist
zolpidem that has some binding selectivity for R1-
containing receptors over other GABAA subtypes is a
clinical hypnotic, again implicating R1-containing sub-
types in the sedative response. Other compounds with
moderate binding selectivity for R1 subtypes have been
described recently, including pyrrolo[2,3-c]pyridines,11
indoleglyoxamides,12 and â-carbolines,13 but there are
fewer reports of R2/R3 selective ligands.13,14 A class of
agonist imidazo[1,5-a]quinoxalines15 have been re-
ported, some of which showed functional selectivity for
R1 subtypes over R3. The discovery of functionally
selective R2/R3 partial agonist triazolo[4,3-b]pyridazines16
in our laboratory was recently communicated,17 along
with evidence that such compounds are anxiolytic but
not sedating or ataxic in animal models, supporting the
hypothesis that the sedative effects of BZ site ligands
are mediated through R1-containing receptor sub-
types.10,18 In this paper, we describe a new class of
2-pyridone GABAA BZ site ligands based on the screen-
ing lead 5a and their adaptation to functionally selective
compounds for R2/R3 subtypes.19
Sch em e 2
Ch em istr y
The well-precedented regioselective condensation of
2-cyanoacetamide with 3-(dimethylamino)-2-propenones
under basic conditions was used for the construction of
the trisubstituted 2-pyridones,20 requiring the initial
preparation of various 1,2-diarylethanones 2 (Scheme
1). The carbanions formed by deprotonation of aryl and
heteroaryl O-trimethylsilylcyanohydrins 1 were alkyl-
ated with benzylic halides to give the ketones 2a -e after
acidic hydrolysis (50-72%, route A). An alternative
procedure began with a Claisen condensation of 2-aryl
acetate esters and heteroaroyl imidazolides to give the
â-ketoesters 3. Decarboxylation under Krapcho’s condi-
tions21 afforded the ketones 2f-h (2-36%, route B). In
the case of 4-picoline, deprotonation of the 4-methyl
group20d and addition to 4-cyanopyridine produced the
ketone 2i directly, albeit in low yield (3%) (route C).
Treatment of the ketones 2 with dimethylformamide-
dimethylacetal (DMF-DMA) gave the corresponding
3-dimethylaminopropen-2-ones, and generally without
purification, these were condensed with 2-cyanoacet-
amide to give the 3-cyano-2-pyridones 4 in moderate to
good yields (47-80%) (Scheme 2). Alkylation of the
pyridone 4b with methyl iodide and cesium carbonate
gave a readily separable mixture of the N-methylated
lead pyridone 5a and the isomeric 2-methoxypyridine
6a .

3-Heteroaryl-2-pyridones
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1889
Sch em e 3
Sch em e 5
Sch em e 4
ketones derived from 2 to give the fully functionalized
pyridones 13 (Scheme 5). This direct approach to
3-heteroaryl-2-pyridones, which was particularly ef-
ficient with sulfur-containing heterocycles, does not
appear to have been exploited before.20a,b
The intermediate 3-cyano-2-pyridones 4 were hydro-
lyzed upon refluxing in strong aqueous acid to give the
carboxylic acids 7. These intermediates were not gener-
ally purified but were methylated to give the 1-methyl-
3-methoxylcarbonyl-2-pyridones 8, again accompanied
by the isomeric 2-methoxypyridines (Scheme 3).
To obviate the need for separation of these alkylation
products, the condensation of the eneaminoketone de-
rived from ketone 2f was carried out with N-methyl
2-cyanoacetamide to give the 1-methyl-3-cyano-2-pyri-
done 5b directly in good yield (50-60%, Scheme 4).
Hydrolysis to the acid 9 allowed formation of the
3-methyl-1,2,4-oxadiazole 10.45 The isomeric 5-methyl-
1,2,4-oxadiazole 11a and its cyclopropyl analogue 11b
were prepared from the nitrile 5b by standard chemis-
try.
The amides 12 were readily prepared from the corre-
sponding ethyl esters upon treatment with methyl-
amine, and several methods were used to reach these
precursors (Scheme 6). The 3-methyl-1,2,4-thiadiazol-
5-yl acetamide 12a was built up from acetamidine by
condensation with trichloromethanesulfenyl chloride to
give 5-chloro-3-methyl-1,2,4-thiadiazole, followed by SN-
Ar displacement with 2-lithio ethyl acetate.22 For the
unsubstituted thiazol-2-yl acetamide 12b and its 5-
methyl analogue 12c, the diesters 14 were synthesized
from the thiazoles according to Dondoni’s procedure.23
Krapcho decarboxylation then afforded the monoesters.
Selective lateral lithiation24 of 2-methyl-4-methylthia-
zole and quenching with ethyl chloroformate provided
a fast route to the desired ester precursor for 12d , but
monoacylation was difficult to control and was conse-
For detailed investigation of heterocyclic groups at the
3-position of the pyridone, a more convenient synthesis
was sought, and fortunately, a variety of N-methyl
heteroarylacetamides 12 were found to undergo the
condensation and cyclization reaction with the enamino-

1890 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Collins et al.
Sch em e 6
Sch em e 7
gave the 3-phenyl-2-pyridone 13l and the 3-(3-thienyl)-
2-pyridone 13m .
Biologica l Meth od s
The affinities of test compounds for cloned human
R1â3γ2, R2â3γ2, R3â3γ2, and R5â3γ2 GABAA receptors,
stably expressed in L(tk-) cells, were determined by
displacement of [3H]Ro15-1788.33 The efficacies of se-
lected compounds were determined at cloned human
R1â3γ2, R2â3γ2, and R3â3γ2 GABAA receptors tran-
siently expressed in Xenopus oocytes by measurement
of the modulatory effect on the GABA EC20 ion current
using two electrode voltage-clamp electrophysiology.34
In all cases, the γ2S splice-variant of the γ subunit was
used. The nonselective full agonist chlordiazepoxide and
the nonselective full inverse agonist methyl 6,7-dimeth-
oxy-4-ethyl-â-carboline-3-carboxylate (DMCM) were in-
cluded as standards in the determination of binding
affinities and efficacies. The pharmacokinetic param-
eters of selected compounds following intravenous and
oral dosing in rats were investigated, and occupancy at
the GABAA receptors in rodent brain was determined
by inhibition of in vivo [3H]Ro15-1788 binding.35 Anti-
convulsant and anxiolytic properties were investigated
in a mouse model of epilepsy36 and the rat elevated plus
maze,37 respectively. In brief, pro- or anticonvulsant
activity was measured in mice (n ) 8/group) dosed with
test compound or vehicle (i.p. in 0.5% methylcellulose)
followed immediately after by various doses of pentyl-
enetetrazole (PTZ; s.c. in 0.9% saline vehicle). Animals
were then observed for 30 min, and the number of mice
exhibiting tonic convulsions was scored; from these data,
the dose corresponding to that required for 50% of
animals to demonstrate tonic convulsions (ED50) was
calculated. In this assay, a drug with proconvulsant
activity will reduce the PTZ ED50 whereas an anticon-
vulsant drug will increase the ED50. In the elevated plus
maze, rats (n ) 18/group) received either vehicle (i.p.
0.5% aqueous methylcellulose; pretreatment time ) 5
quently low-yielding.25 Lithiation of 2-methyl-4-cyclo-
propylthiazole gave predominantly metalation at C-5 of
the thiazole ring, as expected from electronic consider-
ations,24 leading to ethyl 2-methyl-4-cyclopropylthiazole-
5-carboxylate (49%). However, some of the required
ethyl 2-(4-cyclopropylthiazol-2-yl)acetate (ca. 6%) was
isolated in a crude mixture with unreacted starting
material and was successfully purified after conversion
to the amide 12e. In contrast, lithiation of 3-methyl-
thiophene26 provided a 4:1 mixture of 4-methylthio-
phene-2-carboxaldehyde and 3-methylthiophene-2-car-
boxaldehyde in high yield (93%) on quenching with
DMF. The mixture was carried through the subsequent
one carbon chain extension using methyl (methylthio)-
methyl sulfoxide27 and separated after formation of the
amide 12f.
The direct condensation-cyclization route failed in
some cases, notably with substituted phenyl acetamides,
and an alternative synthetic route was developed to
allow introduction of the 3-substituent on the pyridone
in a rapid analogue fashion as the final step of the
synthesis (Scheme 7).28 Thus, the ketone 2f was con-
verted to the N-methyl imine using titanium tetra-
chloride as a dehydrating agent.29 Acylation of the crude
imine with 2-benzyloxyacetyl chloride, in the absence
of any additional base,30 gave the N-acyl enamine 15
(58% over two steps). Vilsmeier formylation and in-
tramolecular cyclization of 15 according to the procedure
developed by Meth-Cohn31 gave the 3-benzyloxy-2-
pyridone 16 in moderate yield (28-48%). Hydrogenolytic
deprotection was followed by formation of the triflate
17. Various palladium-catalyzed coupling reactions to
17 were successful;28 for example, Suzuki couplings32

3-Heteroaryl-2-pyridones
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1891
Ta ble 1. Binding Affinities and Efficacies of 3-Cyano- and 3-Carbomethoxy-2-pyridones at Cloned Human GABAA Receptor Subtypes
Ki (nM)a modulation of GABA EC20 current (%)b
no.c
R1â3 γ2
R2â3γ2
1800 ((320)
R3â3 γ2
R5â3γ2
>667d
>667
>667
>667g
>667
393g
R1â3γ2
R2â3γ2
R3â3γ2
5a
4b
6a
8a
8b
8c
8d
8e
7000 ((1200)
860 ((180)
+7 ((3) 30
f
-16 ((3) 30
+8 ((2) 30
>30 000e
>30 000
>30 000
>30 000
>30 000
630 ((120)
>3000h
625i (460, 850)
520 (510, 540)
>3000
>30 000
470g
270 ((70)
600 ((340)
340 ((31)
250 ((97)
+181 ((15) 30
+81 ((13) 30
+132 ((27) 30
210 ((27)
270 (210, 350)
>667
510 (470, 570) >667
8f
8g
680 (670, 700)
1020 ((97)
560 (410, 780)
5.7
76 (60, 100)
320 ((19)
460 (380, 560)
8.3
73 ((11)
79 ((11)
275 (250, 310)
4.0
>667g
390 (280, 540)
320 (290, 350) +173 ((10) 10
1.0 -34 ((4)
+3 ((7) 10
+8 ((5) 10
-25 ((4)
+165 ((6) 10
0.3 -35 ((4) 0.3
1
CDZj
DMCM
+101 ((8) 10
0.3 -36 ((4)
a
Displacement of [3H]Ro15-1788 from hGABAA receptor subtypes stably expressed in L(tk-) cells, mean ((SEM) for n ) 3-5, expressed
to two significant figures. Six to eight concentrations of each test compound incremented in half-log units were used in the determinations.
Modulatory effects of compounds on coapplication with GABA at the GABA EC20 (concentration eliciting 20% of maximum GABA current,
b
range 4-30 µM) determined in Xenopus oocytes transiently expressing hGABAA receptors; mean ((SEM) for n ) 3-7, concentration of
d
test compound (µM). c For structures, see Schemes 2 and 3. Denotes <50% inhibition at 2 µM, n ) 2. e Denotes <50% inhibition at 100
g
h
i
µM, n ) 2. f Value not determined. n ) 1. Denotes <50% inhibition at 10 µM, n ) 2. n ) 2, individual determinations in parentheses.
j
CDZ ) chlordiazepoxide.
min), 10, 30, or 60 mg/kg test compound (i.p. or p.o. as
indicated in text; pretreatment time ) 5 min). Animals
were then placed on the elevated plus maze, and the
time spent in the closed or open arms during a 5 min
trial period was recorded. Increased time spent in the
closed arms was used as an index of anxiogenesis, and
increased time spent on the open arms was used as an
index of anxiolysis.37 In the case of anxiogenic com-
pounds, the nonselective partial inverse agonist FG
7142 (30 mg/kg i.p.; pretreatment time ) 30 min) was
used as positive control, and for anxiolytic compounds,
the positive control was the nonselective full agonist
diazepam (5 mg/kg i.p.; pretreatment time ) 30 min).
Comparisons between groups were made using an
analysis of variance with a post hoc Dunnett’s t-test to
test for significant differences between individual groups
and control (vehicle-treated) animals.
verse agonist character of 5a to a strong agonist across
all three subtypes.
In an attempt to improve affinity and selectivity at
R2/R3-containing receptors, the 3-methyl ester was
retained, and variations of the pendant aryl substitu-
ents at C-5 and C-6 of the 2-pyridone were examined.
Because deletion of these units in 5a had earlier proved
fruitless, simple changes to the aromatic substitution
patterns were pursued. In the event, the presence or
absence of the 4-methoxy and 4-dimethylamino substit-
uents was irrelevant for affinity at the GABAA receptor
subtypes and compounds 8a , 8c, and 8d had affinities
similar to the prototype ester 8b.
To increase the basicity of the molecules and poten-
tially improve aqueous solubility, the introduction of
pyridine rings to the C-6 position of the 3-carbomethoxy-
2-pyridone was investigated. Although the 2-pyridyl
isomer 8e was of generally lower affinity at the GABAA
receptors, the 3- and 4-pyridyl compounds 8f and 8g
showed significant improvements in affinity at the R3
subtype over the lead ester 8b while retaining ca. 10-
fold selectivity over R1. Additionally, the 4-pyridyl
isomer 8g showed some weak discrimination (ca. 4-fold)
for R3 vs R2 and R5. The effect of the 4-pyridyl isomer
8g on the GABA EC20 ion current was measured,
revealing an intriguing functional selectivity for this
compound. Whereas 8g was an antagonist at R1 and
R2 receptors, inverse agonist efficacy was indicated at
the R3 subtype (-25% as compared to -35% for the
standard full inverse agonist DMCM in this assay).
Resu lts a n d Discu ssion
The lead 1-methyl-3-cyano-2-pyridone 5a (Table 1)
emerged from screening of the corporate sample collec-
tion as a low-affinity ligand with weak binding selectiv-
ity (3-8-fold) for R2 and R3 GABAA subtypes over R1.
Further profiling of 5a showed it to be an antagonist at
R1 and R3 subtypes, with some partial inverse agonist
activity at R2 subtypes. In view of the hypotheses
described in the Introduction, our initial aim was to
improve the affinity and binding selectivity of the lead
for R2 and/or R3 subtypes over R1 and to achieve agonist
efficacy in this series of compounds.
Simple changes to the lead structure rapidly estab-
lished that the N-1 methyl group was essential for
affinity at human GABAA receptors, shown by the lack
of affinity of the N-unsubstituted compound 4b, and
that the isomeric 2-methoxypyridine 6a would not serve
as a replacement for the 2-pyridone (Table 1). Removal
of either of the pendant aryl groups and replacement
with a methyl group also led to complete loss of affinity
(Ki > 30 µM at R1 and R3 subtypes; compounds not
shown). However, modification of the 3-cyano group of
5a to a methyl ester gave 8b, which had higher affinities
than the lead at R1, R2, and R3 subtypes. On measuring
the efficacy of 8b, the change from nitrile to ester was
observed to have converted the antagonist/partial in-
The functional selectivity of compound 8g as an
inverse agonist at R3 receptors and an antagonist at R1
and R2 receptors provided the opportunity to look for
the effects of selectively modulating GABA neurotrans-
mission through R3 receptors in vivo (Table 3). The
pharmacokinetic behavior of 8g in rats was poor, with
rapid metabolism observed, but CNS penetration of the
compound was nonetheless encouraging: On oral dosing
as a suspension in aqueous 0.5% methylcellulose at 10
mg/kg, the levels of 8g in the rat brain reached 170
((59) ng/g at 30 min postdose, sufficient to progress to
testing for pro- or anticonvulsant activity in a mouse
model of epilepsy.36 In this assay, 8g (30 mg/kg i.p.,
suspension in aqueous 0.5% methylcellulose) was found

1892 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Collins et al.
Ta ble 2. Binding Affinities and Efficacies of 3-Heteroaryl- and 3-Aryl-2-pyridones at Cloned Human GABAA Receptor Subtypes
Ki (nM)a modulation of GABA EC20 current (%)b
no.c
R1â3γ2
>1000d
R2â3γ2
R3â3γ2
>1000
R5â3γ2
R1â3γ2
R2â3γ2
R3â3γ2
10
>667e
>667
f
11a
11b
13a
13b
13c
13d
13e
13f
13g
13h
13i
13j
13k
13l
13m
260 ((110)
310 ((34)
90 ((7)
45 ((3)
330 (290, 360)
17 ((1)
31 ((23)
30 ((9)
41 ((1)
13 ((4)
24 (24, 25)
12.8 ((0.3)
17 ((2)
83 ((18)
39 (39, 40)
1100g (910, 1300)
240 ((50)
42 ((8)
320 ((50)
60 ((15)
25 ((4)
6.2 ((1.6)
250 ((57)
3.4 ((1.3)
12 ((3)
15 ((3)
23 ((2)
1300h
+111 ((28) 30
+81 ((10) 10
+120 ((17) 30 +142 ((33) 30 +150 ((31) 10
>667
>1000
>1000
320 ((38)
>1000
350 (310, 390)
+117 ((20) 10
+56 ((3)
+97 ((24) 10 +138 ((22)
3
1
37 ((4)
3
+67 ((7)
+80 ((13)
+15 ((6)
3
1
1
+32 ((3)
380 ((47)
8.6 ((1.8)
24 ((6)
+7 ((2)
+29 ((8)
+9 ((4)
1
1
1
+43 ((3)
+35 ((6)
+36 ((3)
1
1
1
23 ((1)
38 ((7)
340 (270, 410)
380 ((18)
270 (260, 280)
8.0 ((1.3)
20 (18, 23)
5.0 ((0.4)
6.6 ((2.2)
130 ((37)
51 (50, 52)
3.5 ((1.0)
6.7 (5.7, 7.8)
3.8 ((0.1)
4.1 (2.5, 6.6)
24 ((6)
+57 ((10)
-23 ((1)
+78 ((10)
+7 ((5)
3
3
1
3
3
+41 ((10)
-4 ((3)
+77 ((14)
+34 ((7)
+63 ((5)
+18 ((2)
1
1
3
1
3
1
>667
330 ((110)
>1000
450 (440, 450)
+33 ((6)
10
+45 ((8)
12 (12, 12)
a
Displacement of [3H]Ro15-1788 from hGABAA receptor subtypes stably expressed in L(tk-) cells, mean ((SEM) for n ) 3-5, expressed
to two significant figures. Six to eight concentrations of each test compound incremented in half-log units were used in the determinations.
Modulatory effects of compounds on coapplication with GABA at the GABA EC20 (concentration eliciting 20% of maximum GABA current,
b
range 4-30 µM) determined in Xenopus oocytes transiently expressing hGABAA receptors; mean ((SEM) for n ) 3-7, concentration of
d
test compound (µM). c For structures, see Schemes 4-7. Denotes <50% inhibition at 3 µM, n ) 2. e Denotes <50% inhibition at 2 µM,
g
h
n ) 2. f Value not determined. n ) 2, individual determinations in parentheses. n ) 1.
Ta ble 3. Qualitative Summary of Biological Data for Compounds 8g, 13d , and 13k
effects in vivo
rat elevated plus maze
compd
efficacy profile in vitro
mouse PTZ assay
8g
13d
13k
R3 selective inverse agonist
R2/R3 selective partial agonist
R2/R3 selective partial agonist
proconvulsant
anticonvulsant
anticonvulsant
anxiogenic
anxiolytic
to be proconvulsant, potentiating the action of PTZ in
inducing tonic seizures and reducing the PTZ ED50 from
102 (88-116 mg/kg; 95% confidence limits) to 63 mg/
kg (55-74 mg/kg; 95% confidence limits). The effects of
the compound in the rat elevated plus maze model37 for
anxiety were also examined, and 8g (30 mg/kg i.p.,
suspension in aqueous 0.5% methylcellulose) was found
to have a statistically significant anxiogenic effect
relative to vehicle-dosed control animals.
These experiments provided evidence for a functional
role for the R3 subtype in vivo, with the effects observed
in agreement with those expected from a BZ inverse
agonist and indicating that the R3 subtype is at least
partly involved in mediating anxiety behavior. A caveat
to this interpretation would be possible interspecies
differences in the GABAA subtype selectivity; however,
where we have been able to test other BZ ligands for
affinities or efficacies at mouse and rat GABAA sub-
types, we have not observed significant deviations from
the profiles observed in vitro with human GABAA
receptors. This is consistent with the high sequence
homology (>90%) of GABAA subunit subtypes between
human and rodent.2a
A few simple ester analogues of 8g were prepared (not
shown in tables) and demonstrated that groups larger
than methyl (e.g., allyl, tert-butyl) were tolerated in
terms of affinity and provided a general increase in the
level of efficacy at R1, R2, and R3 subtypes. Capitalizing
on this, we sought to replace the metabolically labile
methyl ester of 8g with aromatic heterocycles (Table 2).
1,2,4-Oxadiazoles have become generally established as
bioisosteres for esters, and this equivalence has been
investigated in other series of BZ site ligands.38 While
the 3-methyl-1,2,4-oxadiazol-5-yl isomer 10 showed
negligible affinity for GABAA subtypes, the 5-methyl-
1,2,4-oxadiazol-3-yl derivative 11a retained some affin-
ity, although the 10-fold binding selectivity for R3
against R1 of 8g was lost. By increasing the bulk of the
heterocyclic substituent to the cyclopropyl oxadiazole
11b, affinity was regained. Both 11a and 11b were full
agonists at all GABAA subtypes, a sweeping change
from the antagonist/inverse agonist profile of 8g and
paralleling earlier observations with simple homolo-
gated esters at C-3 of the pyridone.
Although the 3-methyl-1,2,4-oxadiazole 10 had shown
negligible BZ site affinity, the corresponding 1,2,4-
thiadiazole 13a showed excellent binding and was a full
agonist at all subtypes. Further improvements in af-
finity were seen on deletion of nitrogen in the thiadia-
zole to generate a group of 3-(2-thiazolyl)-2-pyridones
13b-e. Thus, the unsubstituted thiazole 13b proved a
very high affinity, partial agonist at R3 subtypes, with
ca. 6-7-fold binding selectivity over the R1 and R2
subtypes, and inactive at R5 subtypes. The thiazoles
were sensitive to substitution pattern: whereas intro-
duction of a 5-methyl group to give 13c reduced affinity
severely, the isomeric 4-methylthiazole 13d marginally
improved affinity at all subtypes, while maintaining a
low level of binding selectivity for R3 subtypes over R1
and R2. Most importantly, 13d had functional selectivity
against R1-containing receptors, being an antagonist at
this subtype while a partial agonist at R2 and R3
receptors. Increasing the size of the thiazole 4-substitu-
ent to cyclopropyl, as in 13e, somewhat reduced affinity
at all subtypes and brought back R1 efficacy.
The apparent preference for sulfur-containing hetero-
cycles was further demonstrated with the thiophen-2-
yl derivatives 13f and 13g, although with some loss of

3-Heteroaryl-2-pyridones
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1893
Ta ble 4. Pharmacokinetics in Rat of Pyridones 13d and 13k
dose
T1/2
Cmax
Tmax
Clp
Vdss
Foral
no. LogD (mg/kg) (h) (ng/mL) (h) (ng mL/h) (L/kg) (%)
13d
13k
3.6
2.7
2
1.5
1.7
1.1
83
170
0.3
0.5
44
35
3.3
2.5
25
65
ated with the C-3 substituent that drives BZ site
binding affinity.
F igu r e 1. X-ray crystal structure of 13d (hydrogens omitted
The models of ligand binding to the BZ site derived
so far do not take account of the subtype selective
efficacy that can be achieved with these 2-pyridones.
However, many BZ site agonist pharmacophore models,
although derived from data on mixed populations of
GABAA receptors, contain common elements including
interactions between the ligands and two hydrogen bond
donors in the receptor.4a,39 This general arrangement
has been maintained in models derived for single
subtypes of the GABAA receptor.14,43 Assuming that the
pyridone carbonyl participates in one such interaction,
compounds such as 13f, 13g, 13l, and 13m may lack a
second H bond acceptor site. In the thiazoles, 1,3,4-
thiadiazole, and 1,2,4-oxadiazoles, additional hetero-
atoms in the C-3 substituent may contribute to binding
by fulfilling this role. Binding to distinct lipophilic
pockets in the receptor appears to be a major determi-
nant of BZ site binding and efficacy.39 The C-3 substitu-
ent of the 2-pyridones may occupy one such region, the
gross electronic and steric properties of the heterocycle
determining the strength of the binding interaction and
the efficacy elicited.
The 2-pyridones described here showed limited bind-
ing selectivity between R1, R2, and R3 subtypes, typi-
cally 10-fold at greatest, although better discrimination
against R5 subtypes could be achieved, a pattern
encountered in other BZ site ligands such as zolpidem
and a series of indoleglyoxamides.12 Studies using site-
directed mutagenesis or photoaffinity labeling of amino
acid residues on both R and γ subunits have located the
BZ binding site at the interface of the R and γ subunits
in the GABAA receptor assembly.44 The lack of binding
selectivity might suggest that the 2-pyridones interact
with amino acid residues that are conserved or mini-
mally changed between the different R subunits. It is
also possible that binding to the γ2 subunit, which is
invariant in the constructs examined here, dominates
the affinity and could account for the limited R subunit
binding selectivity. However, it is clear that interactions
with the R subunits are important for determining the
efficacy of BZ site ligands and can lead to in vitro
functional selectivity that correlates with useful in vivo
pharmacology.
for clarity).
affinity. The unsubstituted thiophene 13f showed func-
tional selectivity for R3 over R1, albeit with reduced
efficacy for R2 as compared to 13d . The thiophene sulfur
atom could be translated to the 3-position in 13m
without further loss of affinity. Congruent with this,
replacement of the heterocycle with a phenyl ring in 13l
was tolerated, but this compound was an unselective
partial agonist at R1 and R3 subtypes.
Examination of the crystal structure of 13d threw
some light on the structure-activity of the C-3 hetero-
cycles (Figure 1). Adjacent pendant aromatic groups
might be generally expected to twist out-of-plane rela-
tive to the core pyridone to reduce steric repulsions, and
this was seen in the crystal structure of 13d for the C-5
and C-6 aromatic rings (interplanar torsions ca. 58 and
63°, respectively). This conformation in itself may be
important for effectively filling the out-of-plane lipo-
philic pockets postulated in many BZ agonist pharma-
cophore models.39 For the sterically less-demanding five-
membered C-3 heterocycles, flanked only by a proton
and a carbonyl, electronic considerations are important.
In 13d , the thiazole was clearly coplanar with the
pyridone ring (interplanar torsion ca. 3°), with a close
intramolecular contact of the sulfur and the carbonyl
oxygen (S...O ) 2.76 Å; van der Waals radii40 S ) 1.80
Å, O ) 1.52 Å). This type of attractive intramolecular
interaction between a thiazole or thiazolidine S and an
adjacent carbonyl O at four bonds distance has been
observed before41 and can be viewed simplistically as a
Coulombic attraction between the positively charged S
and the negatively charged O.41b Similar CdO...S
stabilizing interactions have been proposed from the
crystal structures of analogous 1,3,4-thiadiazoles41b and
2-thiophenes.42
As has been remarked elsewhere,4b these 2-pyridones
are unusual among the many known BZ site ligands in
having a monocyclic rather than a fused bi- or tricyclic
core, but the presence of a potential conformational lock
in 13d , and by extension the other 2-thiazoles, 2-thio-
phenes, and 1,3,4-thiadiazoles examined here, arguably
renders these compounds pseudotricyclic and may ac-
count in part for their higher affinity relative to other
five-membered heterocycles. The difference in affinities
between compounds 13c and 13d thus reflects the
positioning of the methyl substituent with respect to the
GABA receptor and implies a size-limited binding
pocket of defined position relative to the pyridone core.
Other five-membered C-3 heterocycles should be able
to reach the coplanar orientation but without the benefit
of a positive stabilizing interaction. The only slightly
reduced affinities of the 3-phenyl-2-pyridone 13l, which
cannot reach coplanarity, and the 3-thienyl analogue
13m , which does not have the conformational lock,
indicate that conformation is not the only factor associ-
The R2/3 selective partial agonist 13d was investi-
gated in vivo for the anticonvulsant activity expected
of a compound with this efficacy profile (Table 3).
Indeed, 13d gave 50% protection against PTZ-induced
tonic seizures in mice after i.p. administration at 30 mg/
kg as a suspension in aqueouse 0.5% methylcellulose,
a result complementing the R3-mediated proconvulsant
activity of 8g. However, attempts to look for anxiolytic
activity of 13d in the rat elevated plus maze model (60
mg/kg p.o. dosing) were unsuccessful. This failure was
attributed to the poor pharmacokinetic and pharmaco-
dynamic properties of 13d (Table 4). Determination of
the in vivo occupancy35 of rat GABAA receptors demon-
strated the low levels of compound reaching the phar-

1894 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Collins et al.
macological target: Administration of 13d in aqueous
70% PEG300 at a dose of 50 mg/kg p.o. gave only 16
((8)% occupancy.
subtypes. Moreover, the ultimate validation of the
hypothesis that a particular subtype mediates anxioly-
sis, or any other aspect of the clinical diversity of
nonselective BZ actions, will necessitate the evaluation
of such compounds in the clinic.
To improve the pharmacokinetic properties of 13d ,
conservative variations of the pendant aryl groups were
made, hoping to retain high affinity by keeping the
4-methylthiazol-2-yl group at C-3 of the pyridone for the
reasons discussed above. This proved fruitful, as shown
by the four examples 13h -k , where similar affinities
to 13d were observed at all GABAA subtypes. These
examples also served to show that even small changes
to the C-5 and C-6 substituents were found to signifi-
cantly alter the efficacy profile. Thus, modification of
the 4-methoxy group in the C-5 phenyl ring to fluorine
gave 13h , an unselective partial agonist at R1 and R3
subtypes, and replacement of this aryl group by 4-pyri-
dyl produced the R1 inverse agonist 13i. The substitu-
tion of 3-pyridyl for 4-pyridyl at the C-6 position of the
pyridone, a change that had been shown to maintain
affinity at R3 in the series of C-3 esters (8f vs 8g), also
modified the efficacy to give the unselective agonist 13j.
However, combining the 3- and 4-pyridyl motif to give
the 4-pyridazinyl-substituted compound 13k restored
the functional selectivity, providing an antagonist at the
R1 subtype with equal partial agonist activity at R2 and
R3 subtypes. Moreover, the oral bioavailability of 13k
was improved over the original R2/3 selective agonist
13d (Table 4).
Con clu sion s
A novel series of 3-heteroaryl-5,6-bis(aryl)-2-pyridones
were developed with high affinity for the BZ binding
site of human GABAA receptor ion channels and low
binding selectivity for R2- and/or R3- over R1-containing
GABAA receptor subtypes. Good binding selectivity was
observed over R5-containing subtypes. High affinity in
this series appeared to be associated with an attractive
intramolecular S...O interaction between the pyridone
carbonyl and adjacent thiacycles, evident in the crystal
structure of the thiazole 13d , which favored a coplanar
conformation of these two rings. Functional selectivity
at GABAA receptor subtypes was observed in several of
these compounds, giving in particular cases no efficacy
at the R1 subtype and varying degrees of efficacy at R2
and R3 subtypes. The selective R3 inverse agonist ester
8g was proconvulsant and anxiogenic in rodents, sug-
gesting that modulation of GABA-ergic transmission
through the R3 subtype alone is capable of mediating
anxiety behavior. The R2/R3 selective partial agonist
thiazoles 13d and 13k were anticonvulsant, and the
more bioavailable compound 13k was also anxiolytic on
oral dosing. These results are encouraging for the
hypothesis that subtype selective BZ site agonists will
provide new anxiolytic therapies.
The moderately improved bioavailability of 13k was
reflected in its in vivo profile (Table 3). Compound 13k
(i.p. dosing as a suspension in 0.5% aqueous methyl-
cellulose) was anticonvulsant in the mouse model of
epilepsy, with an ED50 of 3.3 mg/kg for protection
against PTZ-induced tonic seizures, significantly more
potent than 13d (50% protection after 30 mk/kg i.p.).
Furthermore, as anticipated from the selective partial
agonist profile and improved pharmacokinetic param-
eters relative to 13d , compound 13k (30 mg/kg p.o.,
suspension in 0.5% aqueous methylcellulose) showed a
statistically significant anxiolytic effect in the rat
elevated plus maze.
Exp er im en ta l Section
1H nuclear magnetic resonance (NMR) spectra were re-
corded on Brucker AM360 or AC250 spectrometers. Chemical
shifts (δ), from Me4Si as internal standard, are given in parts
per million and coupling constants (J ) in Hertz. Mass spectra
were obtained with a VG Quattro spectrometer operating in
electrospray positive ion mode. Anhydrous solvents were
purchased from the Aldrich Chemical Co. Organic solutions
were dried over Na2SO4 or MgSO4. Flash chromatography was
performed on silica gel Fluka Art. No. 60738. Thin-layer
chromatography (TLC) was carried out on Merck 5 cm × 10
cm plates with silica gel 60 F254 as sorbant. Preparative TLC
was carried out using 20 cm × 20 cm silica gel GF tapered
plates supplied by Analtech Inc., Newark. Microanalyses were
determined by Butterworth Laboratories, 54-56 Waldegrave
Road, Teddington, U.K. Commercially available starting ma-
terials were used as supplied.
2-(4-Meth oxyph en yl)-1-(4-dim eth ylam in oph en yl)eth an -
1-on e 2a . Solid ZnI2 (0.5 g, 1.6 mmol) was added to a stirred
solution of 4-dimethylaminobenzaldehyde (53.7 g, 360 mmol)
and TMSCN (45 mL, 360 mmol) in CH2Cl2 (500 mL) at 0 °C.
The mixture was warmed to room temperature and stirred for
6 h and then concentrated. The residue was redissolved in
Et2O and filtered through activated charcoal. The filtrate was
dried and concentrated to give the crude cyanohydrin 1a (86
g, quantitative). A portion of the cyanohydrin (65 g, 280 mmol)
was dissolved in dry tetrahydrofuran (THF; 150 mL) and
added via cannula to a stirred solution of LDA (280 mmol) in
dry THF (500 mL) at -78 °C under N2. The mixture was
stirred for 0.5 h followed by addition of 4-methoxybenzyl
chloride (38 mL, 280 mmol). The cooling bath was removed,
and the mixture was stirred at room temperature for 18 h,
then poured into H2O (1 L), and extracted with EtOAc (3 ×
350 mL). The extracts were washed with H2O and brine, dried,
and concentrated. The residue was dissolved in MeOH (300
mL), and 1 M H2SO4 (400 mL) was added. After it had stirred
at room temperature for 18 h, the mixture was basified to pH
The present study, together with the identification of
an R2/R3 subtype selective agonist triazolo[4,3-b]pyri-
dazine in our laboratories,16 demonstrates that func-
tional selectivity between GABAA receptor subtypes can
be achieved in distinct chemical series. While these data
suggest that R3 subunit-containing GABAA receptors
play a role in mediating the anxiolytic properties of
nonselective BZs, it is not possible to estimate the
relative contribution of this receptor subtype as com-
pared to other subtypes. Indeed, experiments with
transgenic mice have implicated receptors containing
an R2 rather than R3 subunit.46 The reason for the
discrepancy between the pharmacological approach in
which R3-containing receptors are involved in anxiolysis
and the molecular genetic approach in which anxiolysis
appears to be mediated via R2- rather than R3-contain-
ing receptors is unclear but may, in part, be method-
ological.47,48 Attributing the anxiolytic properties of BZs
to particular GABAA receptor subtypes would be further
clarified by additional subtype selective compounds.
Nevertheless, the present study clearly demonstrates
that it is possible to develop compounds that discrimi-
nate, on the basis of efficacy, between GABAA receptor

3-Heteroaryl-2-pyridones
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1895
10 with 4 M NaOH and extracted with EtOAc (3 × 300 mL).
The extracts were washed with H2O and brine, dried, and
concentrated. The residue was triturated with Et2O to give
2a (50 g, 186 mmol, 67%) as a yellow solid. δH (250 MHz,
CDCl3): 3.05 (6H, s), 3.77 (3H, s), 4.13 (2H, s), 6.63 (2H, d, J
9), 6.84 (2H, d, J 9), 7.19 (2H, d, J 9), 7.92 (2H, d, J 9).
2-(4-Meth oxyph en yl)-1-ph en yleth an -1-on e 2b. Yield 64%.
δH (360 MHz, CDCl3): 3.79 (3H, s), 4.23 (2H, s), 6.87 (2H, d,
J 9), 7.18 (2H, d, J 9), 7.45 (2H, dd, J 8, 8), 7.55 (1H, t, J 8),
8.00 (2H, d, J 8).
1-(4-Dim eth yla m in op h en yl)-2-p h en yleth a n -1-on e 2c.
Yield 72%. δH (250 MHz, dimethyl sulfoxide (DMSO)-d6): 3.00
(6H, s), 4.19 (2H, s), 6.71 (2H, d, J 9), 7.16-7.33 (5H, m), 7.88
(2H, d, J 9).
2-(4-Meth oxyp h en yl)-1-(2-p yr id yl)eth a n -1-on e 2d . Yield
64%. δH (360 MHz, CDCl3): 3.78 (3H, s), 4.49 (2H, s), 6.85 (2H,
d, J 9), 7.24 (2H, J 9), 7.44-7.48 (1H, m), 7.78-7.84 (1H, m),
8.01-8.05 (1H, m), 8.73 (1H, d, J 8); m/z (ES+) 228 (M + H+).
2-(4-Meth oxyp h en yl)-1-(3-p yr id yl)eth a n -1-on e 2e. Yield
50%. δH (250 MHz, CDCl3): 3.78 (3H, s), 4.23 (2H, s), 6.87 (2H,
d, J 9), 7.18 (2H, J 9), 7.40 (1H, dd, J 8, 5), 8.25 (1H, ddd, J 8,
2, 2), 8.76 (1H, dd, J 5, 2), 9.22 (1H, d, J 2); m/z (ES+) 455 (M2
+ H+).
2-(4-Meth oxyp h en yl)-1-(4-p yr id yl)eth a n -1-on e 2f. N,N-
Carbonyldiimidazole (20.3 g, 125 mmol) was added portionwise
to a stirred suspension of isonicotinic acid (15.4 g, 125 mmol)
in dry THF (200 mL). The gently effervescent mixture was
stirred at room temperature for 1.5 h forming an orange
solution. Meanwhile, a solution of methyl 2-(4-methoxyphenyl)-
acetate (45 g, 250 mmol) in dry THF (50 mL) was added
dropwise at -78 °C to a stirred solution of LDA (250 mmol) in
dry THF (250 mL). The yellow enolate solution was stirred at
-78 °C for 20 min, followed by addition of the imidazolide
solution dropwise via cannula. After it had stirred at -78 °C
for 45 min, the bright yellow suspension was warmed to room
temperature, poured into 1.3 M HCl (700 mL), and washed
with hexane (200 mL). The aqueous solution was brought to
pH 3 with 4 M NaOH and then to pH 7 with saturated
NaHCO3. The neutralized solution was extracted with CH2-
Cl2 (2 × 400 mL). The organic extracts were washed with brine
(300 mL), dried, and concentrated to give a brown oil that
crystallized on standing. The material was triturated with
EtOAc-Et2O (50 mL) and filtered to give 2-(4-methoxyphenyl)-
3-oxo-3-(pyridin-4-yl)propionic acid methyl ester 3 as an off-
white solid (14.7 g, 52 mmol, 42%). δH (250 MHz, CDCl3): 3.77
(3H, s), 3.79 (3H, s), 5.48 (1H, s), 6.90 (2H, d, J 9), 7.28 (2H,
d, J 9), 7.70 (2H, d, J 6), 8.77 (2H, d, J 6). A mixture of the
ester (14.7 g, 53 mmol), NaCl (6.0 g, 103 mmol), and H2O (1.9
mL, 105 mmol) in DMSO (200 mL) was stirred at 150 °C for
2 h. The solution was cooled to room temperature, poured into
H2O (600 mL), and extracted with EtOAc-Et2O (1:1; 2 × 400
mL). The extracts were dried and concentrated to give 2f as a
brown solid (10.2 g, 45 mmol, 87%). δH (250 MHz, CDCl3): 3.79
(3H, s), 4.44 (2H, s), 6.88 (2H, d, J 9), 7.16 (2H, d, J 9), 7.77
(2H, d, J 6), 8.80 (2H, d, J 6); m/z (ES+) 228 (M + H+).
2-(4-F lu or op h en yl)-1-(4-p yr id yl)eth a n -1-on e 2g. Yield
21%. δH (360 MHz, CDCl3): 4.26 (2H, s), 7.00-7.06 (2H, m),
7.18-7.23 (2H, m), 7.75 (2H, d, J 6), 8.81 (2H, d, J 6); m/z
(ES+) 216 (M + H+).
CH2Cl2-MeOH, gave a yellow oil (5.4 g). EtOAc (10 mL) was
added, and the solution stood at room temperature for 18 h.
The solid was collected and washed with EtOAc to give 2i (0.80
g, 4 mmol, 3%) as a light yellow powder. δH (250 MHz,
CDCl3): 4.30 (2H, s), 7.20 (2H, d, J 6), 7.77 (2H, d, J 6), 8.60
(2H, d, J 6), 8.85 (2H, d, J 6).
5,6-Dip h en yl-3-cya n o-1H-p yr id in -2-on e 4a . A solution of
deoxybenzoin (10 g, 51 mmol) and DMF-DMA (27 mL, 204
mmol) in dry DMF (200 mL) was stirred at 85 °C for 18 h.
Solvent and excess reagent were removed by evaporation to
give a yellow oil. A solution of the oil in dry DMF (100 mL)
with cyanoacetamide (4.75 g, 57 mmol) and MeOH (4.54 mL,
112 mmol) was added via cannula to a stirred suspension of
NaH (60%; 4.5 g, 112 mmol) in dry DMF (100 mL) at room
temperature under N2. After it had stirred at room tempera-
ture for 15 min, the mixture was stirred at 95 °C for 18 h.
The mixture was cooled, and solvents were removed by
evaporation. Saturated NH4Cl(aq) was added to the residues,
and the resulting solids were collected, washed with H2O and
Et2O, and dried under vacuum to give 4a (7.5 g, 28 mmol,
54%). δH (360 MHz, DMSO-d6): 7.03-7.06 (2H, m), 7.17-7.25
(5H, m), 7.29-7.33 (2H, m), 7.36-7.41 (1H, m), 8.23 (1H, s),
12.79 (1H, br s).
6-(4-Dim eth yla m in op h en yl)-5-(4-m eth oxyp h en yl)-3-cy-
a n o-1H-p yr id in -2-on e 4b. Yield 95%. δH (360 MHz, CDCl3):
3.00 (6H, s), 3.81 (3H, s), 6.57 (2H, d, J 9), 6.83 (2H, d, J 9),
7.03 (2H, d, J 9), 7.11 (2H, d, J 9), 7.85 (1H, s), 10.50 (1H, br
s).
5-(4-Met h oxyp h en yl)-6-p h en yl-3-cya n o-1H-p yr id in -2-
on e 4c. Yield 47%. δH (360 MHz, CDCl3): 3.80 (3H, s), 6.81
(2H, d, J 9), 6.95 (2H, d, J 9), 7.25-7.27 (3H, m), 7.35 (2H, d,
J 9), 7.96 (1H, s), 11.90 (1H, br s).
6-(4-Dim eth yla m in op h en yl)-5-p h en yl-3-cya n o-1H-p y-
r id in -2-on e 4d . Yield 80%. δH (250 MHz, DMSO-d6): 3.00 (6H,
s), 7.03-7.06 (2H, m), 7.17-7.25 (4H, m), 7.29-7.33 (2H, m),
7.36-7.41 (1H, m), 8.23 (1H, s), 12.79 (1H, br s).
6-(4-Dim eth yla m in op h en yl)-5-(4-m eth oxyp h en yl)-3-cy-
a n o-1-m eth yl-1H-p yr id in -2-on e 5a a n d 6-(4-Dim eth yl-
a m in op h en yl)-5-(4-m eth oxyp h en yl)-3-cya n o-2-m eth oxy-
p yr id in e 6a . MeI (0.203 mL, 3.2 mmol) was added to a stirred
suspension of 4b (1.0 g, 2.9 mmol) and Cs2CO3 (1.04 g, 3.2
mmol) in dry DMF (20 mL) at room temperature under N2.
After 1 h, the mixture was partitioned between H2O and
EtOAc (× 3). The organic solution was washed with H2O and
brine, dried, and concentrated. The residual solid was boiled
in Et2O (25 mL) and cooled, and the solid was collected. The
material was recrystallized from CHCl3-hexane to give 5a
(0.609 g, 1.7 mmol, 59%) as yellow cubes. The mother liquors
were concentrated and purified by flash column chromatog-
raphy, eluting with 99:1 and then 95:5 CH2Cl2-MeOH, to give
first 6a (0.304 g, 0.84 mmol, 29%) as a yellow foam that gave
light yellow needles on recrystallization from CHCl3-hexane.
Continued elution gave further 5a (0.10 g, 0.3 mmol, 10%).
5a : δH (360 MHz, CDCl3): 2.97 (6H, s), 3.42 (3H, s), 3.75 (3H,
s), 6.60 (2H, d, J 9), 6.70 (2H, d, J 9), 6.83 (2H, d, J 9), 6.89
(2H, d, J 9), 7.84 (1H, s); m/z (ES+) 360 (M + H+). Anal.
(C23H21N3O2) C, H, N. 6a : δH (360 MHz, CDCl3): 2.97 (6H, s),
3.83 (3H, s), 4.13 (3H, s), 6.57 (2H, d, J 9), 6.86 (2H, d, J 9),
7.11 (2H, d, J 9), 7.38 (2H, d, J 9), 7.75 (1H, s); m/z (ES+) 360
(M + H+). Anal. (C23H21N3O2) C, N; H: calcd, 5.89; found, 5.35.
6-(4-Dim eth ylam in oph en yl)-5-ph en yl-3-car bom eth oxy-
1-m eth yl-1H-p yr id in -2-on e 8a . A solution of 4d (1.5 g, 4.76
mmol) in 6 M HCl (50 mL) was refluxed under N2 for 72 h.
The mixture was cooled to yield a precipitate that was
collected, washed with Et2O, and dried under vacuum to give
the carboxylic acid 7a (1.36 g, 4.07 mmol, 86%). δH (250 MHz,
DMSO-d6): 2.93 (6H, s), 6.64 (2H, d, J 9), 7.09-7.16 (4H, m),
7.24-7.34 (3H, m), 8.20 (1H, s), 13.23 (1H, br s). A suspension
of the acid (1.0 g, 3.0 mmol), Cs2CO3 (2.14 g, 6.6 mmol), and
MeI (0.41 mL, 6.6 mmol) in dry DMF (20 mL) was stirred at
room temperature under N2 for 18 h. The mixture was
partitioned between H2O and EtOAc (× 3), and the organic
extracts were washed with H2O and brine, then dried, and
concentrated. Flash column chromatography, eluting with 98:2
2-(4-Meth oxyp h en yl)-1-(4-p yr id a zin yl)eth a n -1-on e 2h .
Yield 2%. δH (360 MHz, CDCl3): 3.79 (3H, s), 4.24 (2H, s), 6.88
(2H, d, J 9), 7.16 (2H, d, J 9), 7.84 (1H, dd, J 5, 2), 9.42 (1H,
d, J 5), 9.60 (1H, d, J 2).
1,2-Bis(4-p yr id yl)eth a n -1-on e 2i. n-BuLi (2.5 M, 52 mL)
was added at -78 °C to a stirred solution of picoline (12 g,
129 mmol) in dry THF (300 mL) under N2. After the mixture
had stirred for 1 h, a solution of 4-cyanopyridine (9 g, 86 mmol)
in dry THF (100 mL) was added over 10 min. The red
suspension was warmed to room temperature and stirred for
1 h, followed by addition of 1 M HCl (150 mL) and further
stirring for 1 h. The mixture was neutralized with saturated
NaHCO3(aq) and separated. The aqueous layer was extracted
with CH2Cl2, and the combined organic layers were dried and
concentrated. Flash column chromatography, eluting with 97:7

1896 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Collins et al.
and then 96:4 CH2Cl2-MeOH, followed by recrystallization
from EtOAc, gave 8a (0.34 g, 0.94 mmol, 31%). δH (360 MHz,
DMSO-d6): 2.89 (6H, s), 3.25 (3H, s), 3.77 (3H, s), 6.63 (2H, d,
J 9), 7.00-7.05 (4H, m), 7.10-7.20 (3H, m), 7.99 (1H, s); m/z
(ES+) 363 (M + H+). Anal. (C22H22N2O3) C, H, N.
6-(4-Dim et h yla m in op h en yl)-5-(4-m et h oxyp h en yl)-3-
ca r bom eth oxy-1-m eth yl-1H-p yr id in -2-on e 8b. Yield 44%.
δH (360 MHz, CDCl3): 2.97 (6H, s), 3.41 (3H, s), 3.75 (3H, s),
3.93 (3H, s), 6.61 (2H, d, J 9), 6.69 (2H, d, J 9), 6.88 (2H, d, J
9), 6.92 (2H, d, J 9), 8.22 (1H, s); m/z (ES+) 393 (M + H+).
Anal. (C23H24N2O4‚0.25(H2O)) C, H, N.
5-(4-Meth oxyph en yl)-6-ph en yl-3-car bom eth oxy-1-m eth -
yl-1H-p yr id in -2-on e 8c. Yield 41%. δH (360 MHz, CDCl3):
3.37 (3H, s), 3.72 (3H, s), 3.94 (3H, s), 6.66 (2H, d, J 9), 6.85
(2H, d, J 9), 7.10-7.12 (2H, m), 7.34-7.36 (3H, m), 8.23 (1H,
s); m/z (ES+) 350 (M + H+). Anal. (C21H19NO4) C, N; H: calcd,
5.48; found, 4.95.
5,6-Dip h en yl-3-ca r bom eth oxy-1-m eth yl-1H-p yr id in -2-
on e 8d . Yield 7%. δH (360 MHz, DMSO-d6): 3.21 (3H, s), 3.78
(3H, s), 7.00-7.02 (2H, m), 7.11-7.17 (3H, m), 7.29-7.33 (2H,
m), 7.35-7.39 (3H, m), 8.01 (1H, s); m/z (ES+) 320 (M + H+).
Anal. (C19H17NO3) C, H, N.
6-(2-P yr id yl)-5-(4-m eth oxyp h en yl)-3-ca r bom eth oxy-1-
m eth yl-1H-p yr id in -2-on e 8e. Yield 15%. δH (360 MHz,
CDCl3): 3.37 (3H, s), 3.73 (3H, s), 3.94 (3H, s), 6.67 (2H, d, J
9), 6.87 (2H, d, J 9), 6.96-7.00 (1H, m), 7.26-7.30 (1H, m),
7.56-7.62 (1H, m), 8.26 (1H, s), 8.74 (1H, d, J 8); m/z (ES+)
351 (M + H+). Anal. (C20H18N2O4) C, H, N.
6-(3-P yr id yl)-5-(4-m eth oxyp h en yl)-3-ca r bom eth oxy-1-
m et h yl-1H -p yr id in -2-on e 8f. Yield 3%. δH (360 MHz,
CDCl3): 3.38 (3H, s), 3.73 (3H, s), 3.94 (3H, s), 6.68 (2H, d, J
9), 6.84 (2H, d, J 9), 7.30 (1H, dd, J 8, 5), 7.46 (1H, ddd, J 8,
2, 2), 8.22 (1H, s), 8.42 (1H, d, J 2) 8.60 (1H, dd, J 5, 2); m/z
(ES+) 351 (M + H+). Anal.(C20H18N2O4) C, H, N.
6-(4-P yr id yl)-5-(4-m eth oxyp h en yl)-3-ca r bom eth oxy-1-
m et h yl-1H -p yr id in -2-on e 8g. Yield 4%. δH (360 MHz,
CDCl3): 3.36 (3H, s), 3.73 (3H, s), 3.94 (3H, s), 6.68 (2H, d, J
9), 6.85 (2H, d, J 9), 7.09 (2H, d, J 6), 8.20 (1H, s), 8.64 (2H,
d, J 6); m/z (ES+) 351 (M + H+). Anal. (C20H18N2O4) H, N. C:
calcd, 68.56; found, 68.99. High-performance liquid chroma-
tography (HPLC) > 99.5%.
6-(4-P yr id yl)-5-(4-m et h oxyp h en yl)-3-(3-m et h yl-1,2,4-
oxa d ia zol-5-yl)-1-m eth yl-1H-p yr id in -2-on e 10. A solution
of 2f (10.0 g, 44 mmol) and DMF-DMA (25 mL, 188 mmol) in
dry DMF (100 mL) was stirred for 16 h at 80 °C under N2.
Solvent was removed by evaporation, and the brown solid was
recrystallized from EtOAc to give 3-(dimethylamino)-2-(4-
methoxyphenyl)-1-(pyridin-4-yl)propen-1-one (10.3 g, 36.5 mmol,
83%) as a light orange solid. δH (250 MHz, CDCl3): 2.75 (6H,
br s), 3.77 (3H, s), 6.81 (2H, d, J 9), 7.04 (2H, d, J 9), 7.22 (2H,
d, J 6), 7.38 (1H, s), 8.54 (2H, d, J 6); m/z (ES+) 283 (M + H+).
A portion of this material (5.0 g, 17.7 mmol) and N-methyl-
cyanoacetamide (1.75 g, 17.8 mmol) in dry MeOH (40 mL) was
added via cannula at 0 °C to a stirred solution of freshly
prepared NaOMe (1.94 g, 36 mmol) in dry MeOH (25 mL)
under N2. The suspension was warmed to room temperature
and stirred for 18 h and then refluxed for 4.5 h. The mixture
was cooled, and solvent was removed by evaporation. The
residue was dissolved in CH2Cl2-MeOH-NH3 (90:9:1) and
filtered through a silica plug. The filtrate was concentrated
to give a brown tar, which was triturated and washed with
EtOAc to give 5b (2.65 g, 8.36 mmol, 47%) as a yellow solid.
δH (360 MHz, CDCl3): 3.37 (3H, s), 3.74 (3H, s), 6.70 (2H, d,
J 9), 6.82 (2H, d, J 9), 7.07 (2H, d, J 5), 7.88 (1H, s), 8.64 (2H,
d, J 5); m/z (ES+) 318 (M + H+). A solution of 5b (1.63 g, 5.14
mmol) in 8 M HCl (20 mL) was refluxed for 24 h. The cooled
solution was basified with concentrated aqueous NH3 (100
mL), and solvent and excess ammonia were removed by
evaporation to give a gray solid. The solid was extracted with
hot EtOAc, and the filtered extract was concentrated to give
the acid 9 (0.78 g, 2.33 mmol, 46%) as a pale yellow solid. δH
(250 MHz, CDCl3): 3.48 (3H, s), 3.74 (3H, s), 6.70 (2H, d, J 9),
6.86 (2H, d, J 9), 7.12 (2H, d, J 6), 8.60 (1H, s), 8.68 (2H, d, J
6), 14.36 (1H, br s); m/z (ES+) 336 (M + H+). A mixture of 9
(0.30 g, 0.89 mmol) and 1,1′-carbonyldiimidazole (0.29 g, 1.78
mmol) in dry THF (20 mL) was refluxed under N2 for 48 h.
The mixture was cooled and partitioned between H2O and
EtOAc. The organic layer was dried and concentrated to give
an orange foam that was redissolved in dry DMF (5 mL). NaH
(55%, 0.04 g, 0.92 mmol) was added to a stirred suspension of
acetamide oxime (0.073 g, 0.99 mmol) and freshly activated 4
Å molecular sieves in dry DMF (20 mL) at room temperature
under N2. After it had stirred for 30 min, the solution of the
imidazolide was added and the mixture was stirred at 80 °C
for 2 h. The mixture was cooled and partitioned between
EtOAc and H2O. The organic extracts were washed with brine,
dried, and concentrated to give a yellow solid. Recrystallization
from EtOAc gave 10 (0.10 g, 0.267 mmol, 30%) as a pale yellow
solid. δH (360 MHz, CDCl3): 2.50 (3H, s), 3.44 (3H, s), 3.74
(3H, s), 6.70 (2H, d, J 9), 6.88 (2H, d, J 9), 7.13 (2H, d, J 6),
8.39 (1H, s), 8.64 (2H, d, J 6); m/z (ES+) 375 (M + H+). HPLC
99.5%.
6-(4-P yr id yl)-5-(4-m et h oxyp h en yl)-3-(5-m et h yl-1,2,4-
oxa d ia zol-3-yl)-1-m eth yl-1H-p yr id in -2-on e 11a . NH2OH‚
HCl (0.24 g, 3.5 mmol) and 5b (1.0 g, 3.15 mmol) were added
to a stirred solution of freshly prepared NaOMe (0.40 g, 7.4
mmol) in dry MeOH (10 mL), and the yellow suspension was
refluxed under N2 for 2 h. Further NaOMe (0.16 g, 3.0 mmol)
was prepared and added to the cooled solution, followed by
further NH2OH‚HCl (0.10 g, 1.5 mmol). The mixture was
refluxed for a further 2 h, then cooled, and diluted with
saturated NH4Cl(aq) (100 mL). The mixture was extracted
with CH2Cl2 (2 × 100 mL), and the extracts were washed with
brine (50 mL), dried, and concentrated to give the amide oxime
(1.01 g, 2.87 mmol, 91%) as a yellow solid. δH (360 MHz,
CDCl3): 3.37 (3H, s), 3.73 (3H, s), 6.36 (2H, br s), 6.67 (2H, d,
J 9), 6.84 (2H, d, J 9), 7.08 (2H, d, J 6), 8.14 (1H, s), 8.62 (2H,
d, J 6). A solution of the amide oxime (0.36 g, 1.03 mmol) in
dry DMF (7 mL) was stirred at room temperature under N2
with freshly activated 4 Å molecular sieves. NaH (55%; 0.05
g, 1.2 mmol) was added, and the mixture was stirred for 30
min, followed by the addition of EtOAc (1 mL, 10.2 mmol).
The mixture was stirred at 80 °C for 2 h and then poured into
H2O (50 mL) and extracted with CH2Cl2-MeOH (90:10) (3 ×
30 mL). The extracts were dried and concentrated. Preparative
TLC, eluting with CH2Cl2-MeOH (90:10), followed by recrys-
tallization from EtOAC-EtOH (90:10), gave 11a (0.157 g, 0.42
mmol, 41%). δH (360 MHz, DMSO-d6): 2.65 (3H, s), 3.27 (3H,
s), 3.69 (3H, s), 6.76 (2H, d, J 9), 6.99 (2H, d, J 9), 7.43 (2H, d,
J 6), 8.13 (1H, s), 8.64 (2H, d, J 6); m/z (ES+) 375 (M + H+).
HPLC 99.1%.
6-(4-P yr id yl)-5-(4-m et h oxyp h en yl)-3-(5-cyclop r op yl-
1,2,4-oxa d ia zol-3-yl)-1-m eth yl-1H-p yr id in -2-on e 11b. Yield
63%. δH (360 MHz, CDCl3): 1.21-1.30 (4H, m), 2.20-2.30 (1H,
m), 3.41 (3H, s), 3.74 (3H, s), 6.69 (2H, d, J 9), 6.89 (2H, d, J
9), 7.11 (2H, d, J 6), 8.25 (1H, s), 8.64 (2H, d, J 6); m/z (ES+)
401 (M + H+). Anal. (C23H20N4O3) C, H, N.
N-Meth yl-2-(3-m eth yl-1,2,4-th ia d ia zol-5-yl)a ceta m id e
12a . NaOH (60 g, 1.5 mol) in H2O (100 mL) was added
dropwise at -5 °C over 2 h to a stirred mixture of acetamidine
hydrochloride (30 g, 320 mmol) and Cl3CSCl (59 g, 320 mmol)
in CH2Cl2 (300 mL). The mixture was stirred at 0 °C for 0.5 h,
then warmed to room temperature, and filtered. The biphasic
filtrate was separated, and the aqueous layer was extracted
with CH2Cl2. The combined organic layers were washed with
brine and dried, and solvent was removed by distillation at
atmospheric pressure. The residual oil was distilled in vacuo
(20 mm Hg, 53-55 °C) to give 5-chloro-3-methyl-1,2,4-thia-
diazole (14.5 g, 112 mmol, 35%) as a yellow liquid. δH (250
MHz, CDCl3): 2.65 (s). A portion of this material (5 g, 37 mmol)
in dry THF (15 mL) was added at -78 °C to a stirred solution
of lithio ethyl acetate (2.91 g, 31 mmol) (prepared from LDA
and EtOAc at -78 °C) in dry THF (70 mL) under N2. The
mixture was warmed to room temperature and stirred for 2.5
h and then partitioned between EtOAc and H2O. The organic
extract was washed with brine, dried, and concentrated. Flash
column chromatography, eluting with hexanes-EtOAc (80:20),
gave ethyl 2-(3-methyl-1,2,4-thiadiazol-5-yl)acetate (0.60 g,

3-Heteroaryl-2-pyridones
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1897
3.23 mmol, 9%). δH (360 MHz, CDCl3): 1.34 (3H, t, J 7), 2.66
(3H, s), 4.16 (2H, s), 4.29 (2H, q, J 7). The ester (0.60 g, 3.23
mmol) was dissolved in a solution of MeNH2 in EtOH (8 M,
40 mL, 320 mmol) and briefly warmed to reflux. After it stood
at room temperature for 2 h, solvent was removed by evapora-
tion. Flash column chromatography, eluting with CH2Cl2-
MeOH-NH3 (92:7:1), gave 12a (0.29 g, 1.71 mmol, 53%) as a
brown solid. δH (250 MHz, CDCl3): 2.67 (3H, s), 2.90 (3H, d, J
5), 4.04 (2H, s), 6.38-6.58 (1H, br m).
12d (1.47 g, 8.65 mmol, 18%) as an orange oil. δH (360 MHz,
CDCl3): 2.45 (3H, d, J 1), 2.84 (3H, d, J 5), 3.91 (2H, s), 6.82
(1H, q, J 1), 7.00-7.20 (1H, br m); m/z (ES+) 171 (M + H+).
N-Meth yl-2-(4-cyclop r op yl-2-th ia zolyl)a ceta m id e 12e.
Compound 12e was prepared as above from 4-cyclopropyl-
thiazole. Yield 4%. δH (250 MHz, CDCl3): 0.82-0.89 (2H, m),
0.91-1.01 (2H, m), 2.00-2.11 (1H, m), 2.83 (3H, d, J 5), 3.88
(2H, s), 6.77 (1H, s), 7.00-7.20 (1H, br m); m/z (ES+) 197 (M
+ H+).
N-Meth yl-2-(4-m eth yl-2-th ien yl)a ceta m id e 12f. Com-
pound 12f was prepared from 2-formyl-3-methylthiophene26
by homologation27 to the methyl ester and treatment with 8
M MeNH2 in EtOH to give 12f (26%) as a brown oil. Purity
90% by NMR, remainder of material consisted of regioisomeric
N-methyl-2-(3-methyl-2-thienyl)acetamide. Data for 12f: δH
(250 MHz, CDCl3): 2.24 (3H, s), 2.78 (3H, d, J 5), 3.72 (2H, s),
5.50-5.80 (1H, br m), 6.74 (1H, s), 6.81 (1H, s); m/z (ES+) 170
(M + H+).
6-(4-P yr id yl)-5-(4-m et h oxyp h en yl)-3-(3-m et h yl-1,2,4-
th iadiazol-5-yl)-1-m eth yl-1H-pyr idin -2-on e 13a. See prepa-
ration of 13b. Yield 25%. δH (360 MHz, CDCl3): 2.70 (3H, s),
3.53 (3H, s), 3.75 (3H, s), 6.71 (2H, d, J 9), 6.92 (2H, d, J 9),
7.13 (2H, d, J 6), 8.63-8.66 (3H, m); m/z (ES+) 391 (M + H+).
Anal. (C21H18N4O2S) C, H, N.
6-(4-P yr id yl)-5-(4-m et h oxyp h en yl)-3-(t h ia zol-2-yl)-1-
m eth yl-1H-p yr id in -2-on e 13b. A solution of 2f (10.2 g, 45
mmol) and DMF-DMA (20 mL, 150 mmol) in dry DMF (50
mL) was stirred at 80 °C for 4 h. The mixture was cooled,
poured into water (500 mL), and extracted with EtOAc (3 ×
200 mL). The combined extracts were washed with brine (100
mL), dried, and concentrated to give a yellow solid. The
material was washed well with Et2O-hexane (1:9; 2 × 100
mL) and dried in vacuo to give 3-(dimethylamino)-2-(4-meth-
oxyphenyl)-1-(pyridin-4-yl)propen-1-one as a beige solid (7.7
g, 27 mmol, 60%). δH (250 MHz, CDCl3): 2.78 (6H, br s), 3.79
(3H, s), 6.79 (2H, d, J 9), 7.04 (2H, d, J 9), 7.24 (2H, d, J 6)
7.40 (1H, s), 8.51 (2H, d, J 6). A portion of this material (2.53
g, 8.97 mmol), 12b (1.70 g, 10.9 mmol), and MeOH (0.73 mL,
18 mmol) was dissolved in dry DMF (20 mL) and added via
cannula at room temperature to a stirred suspension of NaH
(55%, 0.80 g, 18 mmol) in dry DMF (10 mL) under N2. The
mixture was stirred at 40 °C for 18 h and then poured into
water (200 mL). The yellow precipitate was collected and
recrystallized from EtOAc to give 13b (2.18 g, 5.81 mmol, 65%)
as a yellow powder. δH (360 MHz, CDCl3): 3.50 (3H, s), 3.74
(3H, s), 6.70 (2H, d, J 9), 6.93 (2H, d, J 9), 7.15 (2H, d, J 6),
7.50 (1H, d, J 3), 7.96 (1H, d, J 3), 8.64 (2H, d, J 6), 8.72 (1H,
s); m/z (ES+) 376 (M + H+). Anal. (C21H17N3O2S) C, H, N.
6-(4-P yr id yl)-5-(4-m eth oxyp h en yl)-3-(5-m eth ylth ia zol-
2-yl)-1-m eth yl-1H-p yr id in -2-on e 13c. Yield 16%. δH (360
MHz, CDCl3): 2.56 (3H, s), 3.48 (3H, s), 3.74 (3H, s), 6.69 (2H,
d, J 9), 6.93 (2H, d, J 9), 7.14 (2H, d, J 6), 7.62 (1H, s), 8.64
(2H, d, J 6), 8.72 (1H, br s); m/z (ES+) 390 (M + H+). Anal.
(C22H19N3O2S‚0.5(C4H8O2)‚0.5(H2O)) C, H, N.
6-(4-P yr id yl)-5-(4-m eth oxyp h en yl)-3-(4-m eth ylth ia zol-
2-yl)-1-m et h yl-1H-p yr id in -2-on e 13d . Yield 42%. δH (360
MHz, CDCl3): 2.58 (3H, s), 3.50 (3H, s), 3.74 (3H, s), 6.70 (2H,
d, J 9), 6.95 (2H, d, J 9), 7.08 (1H, s), 7.15 (2H, d, J 6), 8.65
(2H, d, J 6), 8.90 (1H, br s); m/z (ES+) 390 (M + H+). Anal.
(C22H19N3O2S) C, H, N.
6-(4-P yr id yl)-5-(4-m eth oxyp h en yl)-3-(4-cyclop r op ylth i-
a zol-2-yl)-1-m eth yl-1H-p yr id in -2-on e 13e. Yield 43%. δH
(360 MHz, CDCl3): 0.91-0.99 (4H, m), 2.18-2.26 (1H, m), 3.48
(3H, s), 3.75 (3H, s), 6.72 (2H, d, J 9), 6.94 (2H, d, J 9), 7.98
(1H, s), 7.16 (2H, d, J 6), 8.65 (2H, d, J 6), 8.76 (1H, br s); m/z
(ES+) 416 (M + H+). Anal. (C24H21N3O2S‚0.6(H2O)) C, H, N.
6-(4-P yr id yl)-5-(4-m eth oxyp h en yl)-3-(th iop h en -2-yl)-1-
m eth yl-1H-p yr id in -2-on e 13f. Yield 34%. δH (360 MHz,
CDCl3): 3.45 (3H, s), 3.75 (3H, s), 6.72 (2H, d, J 9), 6.91 (2H,
d, J 9), 7.12 (1H, dd, J 5, 4), 7.16-7.19 (2H, m), 7.43 (1H, dd,
J 5,1), 7.71 (1H, dd, J 4,1), 7.90 (1H, s), 8.58-8.66 (2H, m);
m/z (ES+) 375 (M + H+). Anal. (C22H18N2O2S) C, H, N.
6-(4-P yr idyl)-5-(4-m eth oxyph en yl)-3-(4-m eth ylth ioph en -
2-yl)-1-m eth yl-1H-p yr id in -2-on e 13g. Yield 34%. δH (360
N-Meth yl-2-(2-th ia zolyl)a ceta m id e 12b. EtOCOCl (11.5
mL, 120 mmol) was added dropwise to a stirred solution of
thiazole (7.4 mL, 104 mmol) in dry THF (200 mL) at 0 °C under
N2. After 1 h, a freshly prepared solution of lithio diethyl-
malonate (19.9 g, 120 mmol) in dry THF (150 mL) was added
via cannula and the mixture was stirred at room temperature
for 18 h. The mixture was diluted with Et2O (100 mL) and
washed with saturated NH4Cl(aq) (250 mL) and brine (200
mL), dried, and concentrated. Flash column chromatography,
eluting with hexanes-EtOAc (8:2), gave a mixture of the
thiazoline and diethylmalonate (28.9 g, 4:1 by NMR) as a
colorless oil. A portion of this material (22 g) was stirred at 0
°C in dry CH2Cl2 (200 mL), and o-chloranil (15.5 g, 63 mmol)
was titrated portionwise into the mixture. After complete
addition, the mixture was stirred for 1 h and then washed with
aqueous NaHCO3 (400 mL) and brine (200 mL), dried, and
concentrated. Flash column chromatography, eluting with 6:4
hexanes-EtOAc, gave 14a (14.7 g, 60 mmol, 58%) as a beige
solid. δH (360 MHz, DMSO-d6): 1.25 (6H, t, J 7), 4.14 (4H, q,
J 7), 7.07 (1H, d, J 4), 7.81 (1H, d, J 4), 12.92 (1H, br s). A
solution of 14a (14.7 g, 60 mmol), NaCl (7 g, 120 mmol), and
H2O (2.2 mL, 120 mmol) in DMSO (250 mL) was stirred at
180 °C for 30 min, then cooled, diluted with H2O (500 mL),
and extracted with Et2O-EtOAc (1:1, 400 mL). The extract
was dried and concentrated. Flash column chromatography,
eluting with 7:3 and then 6:4 hexanes-EtOAc, gave ethyl
2-(thiazol-2-yl)acetate (5.0 g, 27 mmol, 48%) as a brown oil.
δH (360 MHz, CDCl3): 1.29 (3H, t, J 7), 4.10 (2H, s), 4.24 (2H,
q, J 7), 7.33 (1H, d, J 3), 7.76 (1H, d, J 3). A solution of Me2-
NH in EtOH (8 M, 40 mL, 320 mmol) was added to a stirred
solution of ethyl 2-(thiazol-2-yl)acetate (5.0 g, 29 mmol) in
EtOH (20 mL) at room temperature After 5 h, solvents and
excess amine were removed by evaporation. Flash column
chromatography, eluting with EtOAc and then CH2Cl2-
MeOH-NH3 (90:9:1), gave partly purified product that was
redissolved in EtOAc (50 mL) and filtered, and the filtrate was
concentrated to yield 12b (2.95 g, 18.9 mmol, 65%) as a brown
oil. δH (250 MHz, CDCl3): 2.85 (3H, d, J 5), 3.98 (2H, s), 7.05-
7.25 (1H, br m), 7.31 (1H, d, J 3), 7.76 (1H, d, J 3).
N-Meth yl-2-(5-m eth yl-2-th ia zolyl)a ceta m id e 12c. Pre-
pared as above from 5-methylthiazole to give 14b. Yield 57%.
δH (250 MHz, DMSO-d6): 1.23 (6H, t, J 7), 2.25 (3H, d, J 1),
4.12 (4H, q, J 7), 7.15 (1H, q, J 1), 12.61 (1H, br s); m/z (ES+)
258 (M + H+). 12c: Yield 83%. δH (360 MHz, CDCl3): 2.45
(3H, s), 2.83 (3H, d, J 5), 3.87 (2H, s), 7.10-7.25 (1H, br m),
7.37 (1H, s); m/z (ES+) 171 (M + H+).
N-Meth yl-2-(4-m eth yl-2-th ia zolyl)a ceta m id e 12d . A so-
lution of n-BuLi (2.5M, 19 mL) in hexanes was added dropwise
to a stirred solution of 2,4-dimethylthiazole (5 mL, 47 mmol)
in dry THF (100 mL) at -78 °C under N2. The bright yellow
suspension was stirred for 45 min followed by addition of
EtOCOCl (4.8 mL, 50 mmol). After the mixture had stirred at
-78 °C for 1 h, the reaction was quenched with saturated NH4-
Cl(aq) (100 mL) and H2O (100 mL) and extracted with CH2-
Cl2 (100 mL). The extract was dried and concentrated. Flash
column chromatography, eluting with 97:3 and then 95:5 CH2-
Cl2-MeOH, gave a mixture of acylation products and starting
material (monoacylation:diacylation:starting material 4:2:1 by
NMR) as an orange oil (2.9 g). The oil was added to a solution
of MeNH2 (8M, 20 mL, 160 mmol) in EtOH and stood for 18 h
at room temperature. Solvent and excess amine were removed
by evaporation. Flash column chromatography, eluting suc-
cessively with CH2Cl2-MeOH (97:3, 96:4, 90:10), gave partly
purified product, which was triturated with Et2O and filtered
to remove the white solid. The filtrate was concentrated to give

1898 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Collins et al.
MHz, CDCl3): 2.30 (3H, s), 3.43 (3H, s), 3.75 (3H, s), 6.71 (2H,
d, J 9), 6.90 (2H, d, J 9), 7.00 (1H, s), 7.11 (2H, d, J 6), 7.56
(1H, s), 7.85 (1H, s), 8.62 (2H, d, J 6); m/z (ES+) 389 (M +
H+). Anal. (C23H20N2O2S) C, H, N.
6-(4-P yr id yl)-5-(4-flu or op h en yl)-3-(4-m eth ylth ia zol-2-
yl)-1-m eth yl-1H-p yr id in -2-on e 13h . Yield 21%. δH (360
MHz, CDCl3): 2.51 (3H, s), 3.49 (3H, s), 6.85-6.90 (2H, m),
6.97-7.01 (2H, m), 7.06 (1H, s), 7.11 (2H, d, J 6), 8.62-8.66
(3H, m); m/z (ES+) 378 (M + H+). Anal. (C21H16N3OSF) C, H,
N.
5,6-Bis(4-p yr id yl)-3-(4-m et h ylt h ia zol-2-yl)-1-m et h yl-
1H-p yr id in -2-on e 13i. Yield 14%. δH (360 MHz, CDCl3): 2.52
(3H, s), 3.50 (3H, s), 6.98 (2H, d, J 6), 7.08 (1H, s), 7.15 (2H,
d, J 6), 8.44 (2H, d. J 6), 8.64 (1H, s), 8.70 (2H, d, J 6); m/z
(ES+) 361 (M + H+). Anal. (C20H16N4OS‚0.4(H2O)) C, N; H:
calcd, 4.61; found, 4.19. HPLC 99.5%.
6-(3-P yr id yl)-5-(4-m eth oxyp h en yl)-3-(4-m eth ylth ia zol-
2-yl)-1-m eth yl-1H-p yr id in -2-on e 13j. Yield 32%. δH (250
MHz, DMSO-d6): 2.42 (3H, s), 3.37 (3H, s), 3.68 (3H, s), 6.77
(2H, d, J 9), 6.99 (2H, d, J 9), 7.34 (1H, s), 7.45 (1H, dd, J 8,
5), 7.88 (1H, ddd, J 8, 2, 2), 8.42 (1H, s), 8.49 (1H, d, J 2), 8.56
(1H, dd, J 5, 2); m/z (ES+) 390 (M + H+). HPLC 95.5%.
was filtered and the filter cake was washed with 1 M HCl (150
mL). The filtrate was neutralized with NaHCO3 and extracted
with CH2Cl2 (2 × 100 mL). The extracts were dried and
concentrated to give 3-(4-methoxyphenyl)-6-(4-pyridyl)-3-hy-
droxy-1-methyl-1H-pyridin-2-one (0.62 g, 2.01 mmol, 75%) as
a pale pink solid. A sample was recrystallized from EtOAC.
δH (360 MHz, CDCl3): 3.40 (3H, s), 3.73 (3H, s), 6.67 (2H, d,
J 9), 6.86 (2H, d, J 9), 6.90 (1H, br s), 6.93 (1H, s), 7.06 (2H,
d, J 6), 8.59 (2H, d, J 6); m/z (ES+) 309 (M + H+). (CF3SO2)2O
(0.60 mL, 0.60 mmol) was added dropwise at -78 °C to a
stirred suspension of the alcohol (0.60 g, 1.95 mmol) and
pyridine (0.40 mL, 3.57 mmol) in dry CH2Cl2 (30 mL). The
brown suspension was briefly warmed to dissolve all material
and then stirred at -78 °C for 1 h. The mixture was warmed
to room temperature, poured into saturated NaHCO3(aq) (50
mL), and extracted with CH2Cl2 (2 × 30 mL). The extracts
were washed with H2O (50 mL) and brine (30 mL), dried, and
evaporated to give a brown oil. Trituration with Et2O gave 17
(0.78 g, 1.77 mmol, 90%) as a brown solid. δH (360 MHz,
CDCl3): 3.39 (3H, s), 3.74 (3H, s), 6.70 (2H, d, J 9), 6.84 (2H,
d, J 9), 7.09 (2H, d, J 7), 7.46 (1H, s), 8.64 (2H, d, J 7); m/z
(ES+) 441 (M + H+). A mixture of 17 (0.055 g, 0.125 mmol),
PhB(OH)2 (0.25 g, 2.05 mmol), and Na2CO3(aq) (2 M, 2 mL, 4
mmol) in (MeOCH2)2 (5 mL) was degassed and purged with
N2 at room temperature. Pd(PPh3)4 (0.1 g, 0.087 mmol) was
added, and the mixture was refluxed under N2 for 2 h. The
mixture was diluted with H2O (10 mL) and extracted with
EtOAc (3 × 10 mL). The extracts were dried and concentrated.
Preparative TLC, eluting with CH2Cl2-MeOH (95:5), gave 13l
(0.028 g, 0.076 mmol, 61%) as a pale pink foam; this was
recrystallized from EtOAc-hexane to give a beige powder. δH
(360 MHz, CDCl3): 3.39 (3H, s), 3.73 (3H, s), 6.69 (2H, d, J 9),
6.90 (2H, d, J 9), 7.13 (2H, d, J 6), 7.35-7.44 (3H, m), 7.59
(1H, s), 7.77 (2H, d, J 8), 8.60 (2H, d, J 6); m/z (ES+) 369 (M
+ H+). Anal. (C24H20N2O2‚0.25(H2O)) C, H. N; calcd, 7.51;
found, 6.99. HPLC 98.0%.
6-(4-P yr id a zin yl)-5-(4-m eth oxyp h en yl)-3-(4-m eth ylth i-
a zol-2-yl)-1-m eth yl-1H-p yr id in -2-on e 13k . Yield 6%. δH
(360 MHz, CDCl3): 2.51 (3H, s), 3.50 (3H, s), 3.74 (3H, s), 6.71
(2H, d, J 9), 6.89 (2H, d, J 9), 7.08 (1H, s), 7.27 (1H, dd, J 5,
2), 8.66 (1H, s), 9.06 (1H, dd, J 2, 1), 9.20 (1H, dd, J 5, 1); m/z
(ES+) 391 (M + H+). Anal. (C21H18N4O2S) C, H, N.
6-(4-P yr id yl)-5-(4-m eth oxyp h en yl)-3-p h en yl-1-m eth yl-
1H-p yr id in -2-on e 13l. A stirred solution of 2f (17.5 g, 77
mmol) in CHCl3 (200 mL) was saturated with MeNH2 gas at
0 °C. A solution of TiCl4 in CH2Cl2 (1 M, 40 mL, 40 mmol) was
added via syringe at 0 °C under N2. After the solution had
stirred at room temperature for 18 h, anhydrous Na2SO4 (5 g)
was added and the yellow suspension was filtered through
Celite, washing with CH2Cl2. Evaporation of the filtrate gave
the crude imine (16.7 g, 70 mmol, 90%) as an orange oil. δH
(250 MHz, CDCl3): 3.48 (3H, br s), 3.77 (3H, s), 4.04 (2H, br
s), 6.76-6.85 (2H, m), 6.96-7.05 (2H, m), 7.26 and 7.58 (2H,
2 × d, J 6), 8.09 (2H, m). A solution of BnOCH2COCl (11.7 g,
68 mmol) in dry THF (50 mL) was added via cannula to a
stirred solution of the imine (16.7 g, 70 mmol) in dry THF (150
mL) at -78 °C under N2. After complete addition, the purple
solution was warmed to 0 °C and stirred for 1.5 h. The brown
suspension was poured into ice/NaHCO3(aq) (500 mL) and
extracted with EtOAc (2 × 250 mL). The extracts were washed
with brine (100 mL), dried, and concentrated. Flash column
chromatogaphy on silica, eluting with CH2Cl2-MeOH (95:5),
gave 15 (17.1 g, 44 mmol, 64%) as a brown gum. δH (250 MHz,
CDCl3): 3.16 and 3.24 (3H, 2 × s), 3.68 and 3.73 (3H, 2 × s),
3.85 (2H, s), 4.36-4.50 (2H, m), 6.88-7.02 (3H, m), 7.10-7.40
(9H, m), 8.44 and 8.62 (2H, 2 × d, J 6); m/z (ES+) 389 (M +
H+). A solution of 15 (7.37 g, 19 mmol) in the minimum volume
of CH2Cl2 (10 mL) was added slowly at room temperature to
stirred POCl3 (12 mL, 129 mmol). The brown solution was
stirred for 1 h and then cooled to 0 °C, and DMF (3 mL) was
added dropwise. The mixture was stirred at room temperature
for 1 h and then at 75 °C for 1.5 h. The mixture was cooled to
0 °C, and further DMF (1.5 mL) was added dropwise. The
mixture was heated at 75 °C for 3 h and then cooled and
poured onto ice (300 mL). NaOH (2M, 150 mL, 300 mmol) was
added with vigorous stirring, followed by additional base to
adjust the solution to pH 9-10. The mixture was extracted
with CH2Cl2 (3 × 100 mL). The extracts were washed with
H2O (100 mL) and brine (100 mL), dried, and concentrated.
Flash column chromatogaphy on silica, eluting with CH2Cl2-
MeOH (96:4), gave 16 (2.12 g, 5.33 mmol, 28%) as a brown
foam. A sample was recrystallized from EtOAc. δH (360 MHz,
CDCl3): 3.36 (3H, s), 3.72 (3H, s), 5.18 (2H, s), 6.65 (2H, d, J
9), 6.76 (1H, s), 6.78 (2H, d, J 9), 7.04 (2H, d, J 6), 7.30-7.47
(5H, m), 8.56 (2H, d, J 6); m/z (ES+) 399 (M + H+). 10% Pd-C
(0.5 g) was added at room temperature to a stirred solution of
16 (1.07 g, 2.69 mmol), HCO2NH4 (1.26 g, 20 mmol), and AcOH
(10 mL, 175 mmol) in MeOH (20 mL). After 3.5 h, the mixture
6-(4-P yr id yl)-5-(4-m eth oxyp h en yl)-3-(th iop h en e-3-yl)-
1-m eth yl-1H-p yr id in -2-on e 13m . Yield 67%. δH (360 MHz,
CDCl3): 3.42 (3H, s), 3.74 (3H, s), 6.70 (2H, d, J 9), 6.90 (2H,
d, J 9), 7.11 (2H, d, J 6), 7.35 (1H, dd, J 5, 3), 7.54 (1 H, d, J
5), 7.76 (1H, s), 8.36 (1H, d, J 3), 8.61 (2H, d, J 6); m/z (ES+)
375 (M + H+). HPLC 97.0%.
Su p p or tin g In for m a tion Ava ila ble: Data for the X-ray
crystal structure determination of 13d . This material is
available free of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) For recent reviews of therapeutic areas addressed through
modulation of GABA neurotransmission, see (a) Krogsgaard-
Larsen, P.; Frøland, B.; Kristiansen, U.; Frydenvang, K.; Ebert,
B. GABAA and GABAB receptor agonists, partial agonists,
antagonists and modulators: design and therapeutic prospects.
Eur. J . Pharm. Sci. 1997, 5, 355-384. (b) Chebib, M.; J ohnston,
G. A. R. GABA-Activated Ligand Gated Ion Channels: Medicinal
Chemistry and Molecular Biology. J . Med. Chem. 2000, 43,
1427-1447. (c) Martin, I. L.; Lattman, E. Benzodiazepine
recognition site ligands and GABAA receptors. Exp. Opin. Ther.
Pat. 1999, 9, 1347-1358.
(2) For recent reviews of the molecular biology and pharmacology
of GABAA receptors, see (a) Stephenson, F. A. The GABAA
receptors. Biochem. J . 1995, 310, 1-9. (b) Mehta, A. K.; Ticku,
M. K. An update on GABAA receptors. Brain Res. Rev. 1999,
29, 196-217.
(3) It has been proposed that GABAC receptors could be classified
as a subset of the GABAA receptors, primarily on the grounds
of similarities in molecular biology. GABAC receptors are
insensitive to benzodiazepines. For a discussion, see ref 1b.
(4) For recent surveys of ligands acting at the GABAA BZ binding
site, see (a) Wang, Q.; Han, Y.; Xue, H. Ligands of the GABAA
Receptor Benzodiazepine Binding Site. CNS Drug Rev. 1999, 5,
125-144. (b) Teuber, L.; Wa¨tjen, F.; J ensen, L. H. Ligands for
the Benzodiazepine Binding Sitesa Survey. Curr. Pharm. Des.
1999, 5, 317-343.
(5) Korpi, E. R.; Mattila, M. J .; Wisden, W.; Lu¨ddens, H. GABAA
Receptor Subtypes: Clinical Efficacy and Selectivity of Benzo-
diazepine Site Ligands. Ann. Med. 1997, 29, 275-282.
(6) Whiting, P. J . The GABAA receptor gene family: new targets
for therapeutic intervention. Neurochem. Int. 1999, 34, 387-
390.

3-Heteroaryl-2-pyridones
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1899
(7) Sieghart, W.; Fuchs, K.; Tretter, V.; Ebert, V.; J echlinger, M.;
Ho¨ger, H.; Adamiker, D. Structure and subunit composition of
GABAA receptors. Neurochem. Int. 1999, 34, 379-385.
(8) McKernan, R. M.; Whiting, P. J . Which GABAA-receptor sub-
types really occur in the brain? Trends Neurosci. 1996, 19, 139-
143.
(9) Rudolph, U.; Crestani, F.; Benke, D.; Bru¨nig, I.; Benson, J . A.;
Fritschy, J .-M.; Martin, J . R.; Bluethmann, H.; Mo¨hler, H.
Benzodiazepine actions mediated by specific γ-aminobutyric
acidA receptor subtypes. Nature 1999, 401, 796-800.
(10) Sieghart, W. Unraveling the function of GABAA receptor sub-
types. Trends Pharmacol. Sci. 2000, 21, 411-413.
(11) Doisy, X.; Dekhane, M.; Le Hyaric, M.; Rousseau, J .-F.; Singh,
S. K.; Tan, S.; Guilleminot, V.; Schoemaker, H.; Sevrin, M.;
George, P.; Potier, P.; Dodd, R. H. Synthesis and Benzodiazepine
Receptor (w Receptor) Affinities of 3-Substituted Derivatives of
Pyrrolo[2,3-c]pyridine-5-carboxylate, a Novel Class of ω1 Selec-
tive Ligands. Bioorg. Med. Chem. 1999, 7, 921-932.
(12) Collins, I.; Davey, W. B.; Rowley, M.; Quirk, K.; Bromidge, F.
A.; McKernan, R. M.; Thompson, S.-A.; Wafford, K. A. N-(Indol-
3-ylglyoxylyl)piperidines: High Affinity Agonists of Human
GABA-A Receptors Containing the R1 Subunit. Bioorg. Med.
Chem. Lett. 2000, 10, 1381-1384.
(13) Huang, Q.; Cox, E. D.; Gan, T.; Ma, C.; Bennett, D. W.;
McKernan, R. M.; Cook, J . M. Studies of Molecular Pharma-
cophore/Receptor Models for GABAA/Benzodiazepine Receptor
Subtypes: Binding Affinities of Substituted â-Carbolines at
Recombinant Rxâ3γ2 Subtypes and Quantitative Structure-
Activity Relationship Studies via a Comparative Molecular Field
Analysis. Drug Des. Discovery 1999, 16, 55-76.
(14) Yu, S.; He, X.; Ma, C.; McKernan, R.; Cook, J . M. Studies in
search of R2 selective ligands for GABAA/BzR receptor subtypes.
Part 1. Evidence for the conservation of pharmacophoric descrip-
tors for DS subtypes. Med. Chem. Res. 1999, 9, 186-202.
(15) J acobsen, E. J .; Stelzer, L. S.; TenBrink, R. E.; Belonga, K. L.;
Carter, D. B.; Im, H. K.; Im, W. B.; Sethy, V. H.; Tang, A. H.;
VonVoigtlander, P. F.; Petke, J . D.; Zhong, W.-Z.; Mickelson, J .
W. Piperazine Imidazo[1,5-a]quinoxaline Ureas as High-Affinity
GABAA Ligands of Dual Functionality. J . Med. Chem. 1999, 42,
1123-1144.
(24) Knaus, G.; Meyers, A. I. Chemistry of metalated heterocycles.
Site of metalation of 2-methyl-4-substituted 1,3-thiazoles. Elec-
tronic, steric and isotope effects. J . Org. Chem. 1974, 39, 1192-
1195.
(25) Quenching of 2-(lithiomethyl)-4-methylthiazole with ethyl chlo-
roformate provided a mixture of the desired ethyl 2-(4-meth-
ylthiazol-2-yl) acetate, diethyl 2-(4-methylthiazoly-2-yl)malonate
and starting material in a 4:2:1 mixture (27%). The monoacyl
derivative was isolated after conversion to the amide.
(26) Sice´, J . Substituted thenoic acids. J . Org. Chem. 1954, 19, 70-
73.
(27) Ogura, K.; Ito, Y.; Tsuchihashi, G. A new synthesis of arylacetic
esters starting from aromatic aldehyde by the use of methyl
(methylthio)methyl sulfoxide. Bull. Chem. Soc. J pn. 1979, 52,
2013-2022.
(28) Collins, I.; Castro, J . L. A Convenient Synthesis of Highly
Substituted 2-Pyridones. Tetrahedron Lett. 1999, 40, 4069-4072.
(29) Martin, G.; Guitian, E.; Castedo, L.; Saa, J . M. Intermolecular
benzyne cycloaddition (IBC), a versatile approach to benzo-
phenanthridine antitumour alkaloids. Formal synthesis of niti-
dine and chelerythrine. J . Org. Chem. 1992, 57, 5907-5911.
(30) Inclusion of an equivalent of base (Et3N) in the acylation reaction
gave significant â-lactam formation (40%, with 40% acyl enam-
ine also produced) as a result of competing elimination of the
acid chloride to a ketene and subsequent [2 + 2] cycloaddition
to the imine. In the absence of base, the proportion of this side
product was reduced to <10%. See ref 28.
(31) Meth-Cohn, O.; Westwood, K. T. A versatile new synthesis of
quinolines and related fused pyridines. Part 12. A general
synthesis of 2-chloropyridines and 2-pyridones. J . Chem. Soc.,
Perkin Trans. 1 1984, 1173-1182.
(32) Miyaura, N.; Yanagi, T.; Suzuki, A. The palladium-catalyzed
cross-coupling reaction of phenylboronic acid with haloarenes
in the presence of bases. Synth. Commun. 1981, 11, 513-519.
(33) Hadingham, K. L.; Wingrove, P.; Le Bourdelles, B.; Palmer, K.
J .; Ragan, C. I.; Whiting, P. J . Cloning of cDNA sequences
encoding human R2 and R3 γ-aminobutyric acidA receptor
subunits and characterization of the benzodiazepine pharmacol-
ogy of recombinant R1-, R2-, R3-, and R5-containing human
γ-aminobutyric acidA receptors. Mol. Pharmacol. 1993, 43, 970-
975.
(34) Wafford, K. A.; Whiting, J . J .; Kemp, J . A. Differences in affinity
and efficacy of benzodiazepine receptor ligands at recombinant
γ-aminobutyric acid A receptor subtypes. Mol. Pharmacol. 1993,
43, 240.
(35) Atack, J . R.; Smith, A. J .; Emms, F.; McKernan, R. M. Regional
Differences in the Inhibition of Mouse In Vivo [3H]Ro15-1788
Binding Reflect Selectivity for R1 vs R2 and R3 Subunit-
Containing GABAA Receptors. Neuropsychopharmacology 1999,
20, 255-262.
(36) Dawson, G. R.; Wafford, K. A.; Smith, A.; Marshall, G. R.; Bayley,
P. J .; Schaeffer, J . M.; Meinke, P. T.; McKernan, R. M. Anti-
convulsant and Adverse Effects of Avermectin Analogues in Mice
Are Mediated through the γ-Aminobutyric AcidA Receptor. J .
Pharmacol. Exp. Ther. 2000, 295, 1051-1060.
(37) Pellow, S.; File, S. E. Anxiolytic and Anxiogenic Drug Effects
on Exploratory Activity in an Elevated Plus-Maze: a Novel Test
of Anxiety in the Rat. Pharmacol., Biochem. Behav. 1986, 24,
525-529.
(38) (a) Watjen, F.; Baker, R.; Engelstoff, M.; Herbert, R.; MacLeod,
A.; Knight, A.; Merchant, K.; Moseley, J .; Saunders, J .; Swain,
C. J .; Wong, E.; Springer, J . P. Novel Benzodiazepine Receptor
Partial Agonists: Oxadiazolylimidazobenzodiazepines. J . Med.
Chem. 1989, 32, 2282-2291. (b) Tully, W. R.; Gardner, C. R.;
Gillespie, R. J .; Westwood, R. 2-(Oxadiazolyl)- and 2-(Thiazolyl)-
imidazo[1,2-a]pyrimidines as Agonists and Inverse Agonists at
Benzodiazepine Receptors. J . Med. Chem. 1991, 34, 2060-2067.
(c) TenBrink, R. E.; Im, W. B.; Sethy, V. H.; Tang, A. H.; Carter,
D. B. Antagonist, Partial Agonist, and Full Agonist Imidazo-
[1,5-a]quinoxaline Amides and Carbamates Acting through the
GABAA/Benzodiazepine Receptor. J . Med. Chem. 1994, 37, 758-
768.
(39) Zhang, W.; Koehler, K. F.; Zhang, P.; Cook, J . M. Development
of a Comprehensive Pharmacophore Model for the Benzodiaz-
epine Receptor. Drug Des. Discovery 1995, 12, 193-248.
(40) Bondi, A. van der Waals Volumes and Radii. J . Phys. Chem.
1964, 68, 441-451.
(41) (a) Maurin, J . K.; Czarnocki, Z.; Paluchowska, B. Structure
of N2-[4-(2-Chlorophenyl)-1,3-thiazol-2-yl]-(1R,1S)-6,7-dimeth-
oxy-1-(3-pyridylmethyl)-1,2,3,4-tetrahydro-2-isoquinolinecarboxo
Amide. Pol. J . Chem. 1999, 73, 377-383. (b) Hansell, D. P.;
Robinson, K. J .; Wallis, J . D.; Povey, D. C. N-[3-Methyl-2-(3H)-
thiazolylidene]benzamide Containing a Short Intramolecular
S...O Contact. Acta Crystallogr. 1996, C52, 136-139.
(16) Broughton, H. B.; Carling, W. R.; Castro Pineiro, J . L.; Guiblin,
A. R.; Madin, A.; Moore, K. W.; Russell, M. G.; Street. L. J .
Patent Application WO 98/04559.
(17) McKernan, R. M.; Rosahl, T. W.; Reynolds, D. S.; Sur, C.;
Wafford, K. A.; Atack, J . R.; Farrar, S.; Myers, J .; Cook, G.;
Ferris, P.; Garrett, L.; Bristow, L.; Marshall, G.; Macaulay, A.;
Brown, N.; Howell, O.; Moore, K. W.; Carling, R. W.; Street, L.
J .; Castro, J . L.; Ragan, C. I.; Dawson, G. R.; Whiting, P. J .
Sedative but not anxiolytic properties of benzodiazepines are
mediated by the GABAA receptor R1 subtype. Nat. Neurosci.
2000, 3, 587-592.
(18) Some unselective partial agonist BZ site ligands have been
shown to be anxiolytic but not sedating in animal models.
However, sedation was observed in human clinical trials of one
of these compounds, bretazenil; see ref 4.
(19) Collins, I. J .; Fletcher, S. R.; Harrison, T.; Leeson, P. D.; Moyes,
C. R.; Nadin, A. J .; Rowley, M.; Sparey, T. J .; Teall, M. R. U.S.
Patent 6,200,982.
(20) For surveys of this and related routes to 2-pyridones, see (a)
J ones, G. In Comprehensive Heterocyclic Chemistry II; Katritzky,
A. R., Rees, C. W., Scriven, E. F. V., Eds.; Pergamon: Oxford,
1996; Vol. 5, pp 167-243. (b) J ones, G. In Comprehensive
Heterocyclic Chemistry; Katritzky, A. R., Rees, C. W., Eds.;
Pergamon: Oxford, 1984; Vol. 2, pp 395-510. For specific
examples of this condensation-cyclisation, see (c) Sircar, I.;
Duell, B. L.; Bristol, J . A.; Weishaar, R. E.; Evans, D. B.
Cardiotonic agents. 5. 1,2-Dihydro-5-[4-(1H-imidazol-1-yl)phe-
nyl]-6-methyl-2-oxo-3-pyridinecarbonitriles and related com-
pounds. Synthesis and inotropic activity. J . Med. Chem. 1987,
30, 1023-1029 and references therein. (d) Robertson, D. W.;
Beedle, E. E.; Schwartzendruber, J . K.; J ones, N. D.; Elzey, T.
K.; Kauffman, R. F.; Wilson, H.; Hayes, J . S. Bipyridine
cardiotonics: the three-dimensional structures of amirinone and
milrinone. J . Med. Chem. 1986, 29, 635-640 and references
therein.
(21) Krapcho, A. P.; Lovey, A. J . Decarboxylations of geminal diesters,
â-keto esters, and R-cyano esters effected by sodium chloride in
dimethyl sulfoxide. Tetrahedron Lett. 1973, 957-960.
(22) MacLeod, A. M.; Baker, R.; Freedman, S. B.; Patel, S.; Merchant,
K. J .; Roe, M.; Saunders, J . Synthesis and muscarinic activities
of 1,2,4-thiadiazoles. J . Med. Chem. 1990, 33, 2052-2059.
(23) Dondoni, A.; Dall’Occo, T.; Fantin, G.; Fogagnolo, M.; Medici,
A. Regioselective carbon-carbon bond formation at C-2 of 1,3-
thiazole by reaction of N-ethoxycarbonylthiazolium chloride with
C-nucleophiles. Tetrahedron Lett. 1984, 25, 3633-3636.

1900 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Collins et al.
(42) (a) Wouters, J .; Creuven, I.; Norberg, B.; Evrard, G.; Durant,
F.; van Aerschot, A.; Herdewijn, P. trans- and cis-S-C-C-S
Conformations in 5-(2,2′-Dithien-5-yl)-2′-deoxyuridine. Acta Crys-
tallogr. 1997, C53, 892-895. (b) Creuven, I.; Norberg, B.; Olivier,
A.; Evrard, C.; Evrard, G.; Wigerinck, P.; Herdewijn, P, Durant,
F. 5-(thien-2-yl) uracil analogues: 5-(5-methylthien-2-yl)-2′-
deoxyuridine, 5-(5-thien-2-yl)-2′-deoxyuridine, and 5-(5-bro-
mothien-2-yl)-2′deoxyuridine. J . Chem. Crystallogr. 1996, 26,
777-789.
(43) Huang, Q.; He, X.; Ma, C.; Liu, R.; Yu, S.; Dayer, C. A.; Wenger,
G. R.; McKernan, R.; Cook, J . M. Pharmacophore/Receptor
Models for GABA/BzR Subtypes (R1â3γ2, R5â3γ2, and R6â3γ2)
via a Comprehensive Ligand-Mapping Approach. J . Med. Chem.
2000, 43, 71-95.
(44) Sigel, E.; Buhr, A. The benzodiazepine binding site of GABAA
receptors. Trends Pharmacol. Sci. 1997, 18, 425-429.
(45) Liang, G.-B.; Feng, D. D. An Improved Oxadiazole Synthesis
Using Peptide Copuling Reagents. Tetrahedron Lett. 1996, 37,
6627-6630.
(46) Lo¨w, K.; Crestani, F.; Keist, R.; Benke, D.; Bru¨nig, I.; Benson,
J . A.; Fritschy, J .-M.; Ru¨licke, T.; Bluethmann, H.; Mo¨hler, H.;
Rudolph, U. Molecular and Neuronal Substrate for the Selective
Attenuation of Anxiety. Science 2000, 290, 131-134.
(47) Reynolds, D. S.; McKernan, R. M.; Dawson, G. R. Anxiolytic-
like action of diazepam: which GABAA receptor subtype is
involved? Trends Pharmacol. Sci. 2001, 22, 402-403.
(48) Rudolph, U.; Crestani, F.; Mo¨hler, H. GABAA receptor sub-
types: dissecting their pharmacological functions. Trends Phar-
macol. Sci. 2001, 22, 188-194.
J M0110789