Synthesis and anticonvulsant evaluation of some new 2-substituted-3- arylpyrido[2,3-d]pyrimidinones

Article abstract of DOI:10.1016/j.bmc.2004.07.068

A series of 2-substituted-3-arylpyrido[2,3-d]pyrimidinones was prepared for evaluation as potential anticonvulsants. In murine screening, compounds 4a-c having a 2-oxo-2-(4-pyridyl)ethyl group in the 2-position and a 2-substituted phenyl moiety at the 3-position of the pyridopyrimidinone system displayed the most potent anti-seizure activity in both the maximal electroshock (MES) and pentylenetetrazol (scPTZ) tests at doses in the 3-10 mg/kg range. Compound 4c showed no agonist activity at the GABA_A receptor and was unable to block presynaptic sodium and calcium channels in vitro.

Full text of DOI:10.1016/j.bmc.2004.07.068

Bioorganic & Medicinal Chemistry 12 (2004) 5711–5717
Synthesis and anticonvulsant evaluation of some new
2-substituted-3-arylpyrido[2,3-d]pyrimidinones
David C. White,a Thomas D. Greenwood,a Aaron L. Downey,a
Jeffrey R. Bloomquistc and James F. Wolfea,b,
*
aDepartment of Chemistry, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0212, USA
bDivision of Biomedical Sciences, Edward Via Virginia College of Osteopathic Medicine, 2265 Kraft Drive,
Blacksburg, VA 24060, USA
cDepartments of Entomology and Chemistry, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0319, USA
Received 23 October 2003; accepted 15 July 2004
Available online 21 September 2004
Abstract—A series of 2-substituted-3-arylpyrido[2,3-d]pyrimidinones was prepared for evaluation as potential anticonvulsants. In
murine screening, compounds 4a–c having a 2-oxo-2-(4-pyridyl)ethyl group in the 2-position and a 2-substituted phenyl moiety
at the 3-position of the pyridopyrimidinone system displayed the most potent anti-seizure activity in both the maximal electroshock
(MES) and pentylenetetrazol (scPTZ) tests at doses in the 3–10mg/kg range. Compound 4c showed no agonist activity at the
GABAA receptor and was unable to block presynaptic sodium and calcium channels in vitro.
Ó 2004 Elsevier Ltd. All rights reserved.
1.Introduction
low neurotoxicity. The 2-oxo-2-(4-pyridyl) ethyl group
at the 2-position of the quinazolin-4-one ring system in
conjunction with a 2-substituted phenyl moiety at the
3-position were identified as key functional groups
(Fig. 1 and 2a–c). Quinazolinone 2d, which lacks a 2-
substituent in the 3-aryl ring was active only at doses
of 300mg/kg, which was 10 times less effective at con-
trolling scPTZ-induced seizures than 2a–c; 2d was com-
pletely inactive against seizures induced by maximal
electroshock.4 Unfortunately, this promising separation
of anticonvulsant and neurotoxic properties observed in
the ip scPTZ mice screens of compounds 2a–b with pro-
tective index values (PIs) of 5.6 and 18, respectively, was
negated in oral scPTZ assessment in rats (PIs of 1.1 and
0.94, respectively).4 An earlier study by Blanton and
co-workers,6 describing that 2-methyl-3-(2-methyl-
phenyl)pyrido[2,3-d]-4(3H)-pyrimidinone (3a) possessed
anti-scPTZ activity comparable to methaqualone (1),
but with lower neurotoxicity, prompted us to undertake
the preparation of a series of new pyrido[2,3-d]pyrimid-
inones 4 in the hope of achieving a more favorable sep-
aration of anticonvulsant and sedative characteristics
than displayed by our former quinazolinone counter-
parts 2. In addition, we were also interested in prepar-
ing selected 2-(2-arylethenyl)pyrido[2,3-d]pyrimidinones
5 in light of the reported antiseizure activity of simi-
larly substituted 3H-quinazolin-4-ones.1a A number of
For more than three decades, interest in 3H-quinazolin-
4-ones as potential sedative-hypnotic and/or anticonvul-
sant drugs has led to the preparation and pharmacolog-
ical evaluation of literally hundreds of such molecules.1
In spite of this effort, only one compound, 2-methyl-3-
(2-methylphenyl)-3H-quinazolin-4-one (methaqualone)
1,2 emerged as a clinically adopted, sedative-hypnotic
drug. However, marketing and clinical use of 1 was ter-
minated in 1984 when it was transferred to Schedule I of
the Controlled Substances Act because of its abuse lia-
bility.3 This, coupled with the tendency of 1 and related
3H-quinazolin-4-ones to present sedative–hypnotic
(neurotoxic) TD50s similar to their ED50s for anticon-
vulsant activity in traditional murine models has been
a major deterrent to the development of these com-
pounds as possible antiepileptic drugs (AEDs).
In a previous study,4 we discovered that several 2-substi-
tuted-3-aryl-3H-quinazolin-4-ones of structural type 25
possessed good activity against pentylenetetrazol
(scPTZ)-induced seizures in mice (ip) with relatively
*
Corresponding author. Tel.: +1 540 231 8200; fax: +1 540 231
8890; e-mail: wolfe@vt.edu
0968-0896/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.07.068

5712
D. C. White et al. / Bioorg. Med. Chem. 12(2004) 5711–5717
Scheme 1.
Scheme 2.
addition of sodium hydride in the presence of the re-
quired aromatic ester4 to afford 2-(2-oxoalkyl) deriva-
tives 4a–e (Scheme 1). Acid-catalyzed condensation of
3a with benzaldehyde and 4-pyridinecarboxaldehyde
gave excellent yields of compounds 5a and 5b, respec-
tively (Scheme 2).
2.2. Pharmacological evaluation
Pharmacological testing of pyridopyrimidinones 3–5
was performed by the Epilepsy Section of the National
Institute of Neurological Disorders and Stroke
(NINDS) using standard protocol adopted by the Anti-
epileptic Drug Development (ADD) program.10 Anti-
convulsant activity was evaluated in the maximal
electroshock seizure (MES) and subcutaneous pentylene-
tetrazol (scPTZ) tests; neurological impairment (TD50)
was assessed by the rotorod test (Table 1).
Figure 1.
Preliminary screening data for 2-methyl-3-arylpyridopyr-
imidinones 3a–d (Table 1) indicated that compounds
3a–c possessing a 2-substituted phenyl moiety at the
3-position of the pyridopyrimidinone ring system exhib-
ited modest activity at 100mg/kg against both MES-
and scPTZ-induced seizures, while 3-phenyl-substituted
compound 3d was active only in the scPTZ test. All four
compounds caused rotorod toxicity at the 100mg/kg
dosage level and therefore showed no separation be-
tween anticonvulsant activity and neurotoxicity.
2-(2-arylethenyl)quinazolinones structurally related to
5 have been prepared recently and were shown to be
effective noncompetitive antagonists of the excitatory
a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
(AMPA) receptor.7 In this same study, racemate 6 was
found to be a potent blocker of scPTZ-induced seizures
in mice (ED50 = 3.8mg/kg).8
Against this background, we now wish to describe the
synthesis of compounds 3–5 and the results of their
pharmacological evaluation as possible anticonvulsant
agents.
Introduction of the 2-oxo-2-(4-pyridyl)ethyl group at
the 2-methyl position of 2-substituted phenyl com-
pounds 3a–c to afford 4a–c resulted in as much as a
30-fold increase in anticonvulsant potency against sei-
zures induced by both MES and scPTZ when compared
to their precursors 3a–c (Table 1). Similar incorporation
of this group into 3-phenyl-substituted compound 3d
caused an increase in the MES activity for the resulting
4d. On the other hand, no enhancement in anticonvul-
sant activity was realized when the 2-methyl group of
3a was replaced by the analogous 2-oxo-2-phenethyl
moiety, cf. compound 4e. This activity trend, which we
2.Results and discussion
2.1. Synthesis
2-Methyl-3-arylpyridopyrimidinones 3a–d were pre-
pared from 2-methyl-pyrido[2,3-d][1,3]oxazin-4-one (7)9
and the appropriate aromatic amine in refluxing toluene
solution. Acylation of 3a–d was accomplished by the

D. C. White et al. / Bioorg. Med. Chem. 12(2004) 5711–5717
Table 1. Pharmacological activity of pyrido[2,3-d]pyrimidinones 3–5a
5713
Compd
MESb,e
sc PTZc,e
Toxicityd,e
0.5 h
4.0 h
0.5 h
4.0 h
0.5 h
4.0 h
3a
3b
3c
3d
4a
4b
4c
4d
4e
5a
5b
++(2/3)
++(3/3)
++(1/1)f
—
+(1/1)
++(2/3)
—
++(4/5)
++(1/1)
++++(1/5)
++(3/5)
+++++(1/4)
+++++(2/4)
+(1/1)
+(1/1)
++(5/5)
++(3/5)
—
+++(1/4)
++(8/8)
++(4/4)
+(2/2)
++(2/4)
—
—
—
++(3/8)
+++++(2/4)
+++++(2/4)
++++(3/4)
++(1/3)
—
+++(1/1)
+++(1/1)
+++(1/1)
++(3/3)
+(1/1)
++(1/3)
—
+++(1/1)
+++(1/1)
++++(1/4)
++(5/5)
+(2/5)
—
++++(6/8)
++++(5/8)
++++(2/4)
++(1/8)
+++(2/2)
+++(2/2)
+++(2/2)
++(4/4)
++(1/4)
++(2/4)
+(1/2)
++(4/5)
—
+(3/5)
++(1/8)
++(2/8)
+(1/4)
+(1/1)
+(1/1)
—
—
a The test compounds were administered ip to mice.
b Maximal electroshock seizure test (number of animals protected/number of animals tested).
c Subcutaneous pentylenetetrazol test.
d Neurologic toxicity (number of animals exhibiting toxicity/number of animals tested).
e Activity and toxicity at 3, 10, 30, 100, and 300mg/kg are indicated by +++++, ++++, +++, ++, and +, respectively; — denotes no antiseizure or
toxicity observed up to 300mg/kg.
f 2/3 animals were protected at 30mg/kg after 0.25h.
initially observed in the analogous quinazolinone series,4
again confirms the pharmacophoric nature of this pyr-
idyl containing moiety when combined with a 2-substi-
tuted phenyl substituent at position 3. The most active
compounds, 4a (R = CH3) and 4b (R = Cl), showed
anti-MES and anti-scPTZ activity at doses of 3mg/kg
in preliminary screening, while 4c (R = Br) was active
in both tests at 10mg/kg. However, all three compounds
also displayed neurological toxicity at dosage levels of
10mg/kg.
As illustrated in Table 2, compound 4c and its 2-chloro-
phenyl analog, 4b, exhibited similar pharmacological
profiles following po administration in rats, that is, each
displayed excellent antiseizure activity in the MES test
with nearly identical ED50 values of 2.56 and 2.49mg/
kg, and similar TD50 values of 8.61 and 8.86mg/kg,
respectively. A comparison of the ED50 values of 4c
and 2a–c from rat po testing reveals that these com-
pounds all have similar potencies against MES-induced
seizures while 4c is ca. 3 times less effective in controlling
seizures caused by scPTZ. These results contrast with
those obtained in mice ip testing as noted previously,
which indicated that 4c was much more potent than
2a–c in both tests.
In light of the similar pharmacological profiles of 4a–b,
it may be concluded that the electron releasing (Àr) ver-
sus electron withdrawing (+r) nature of the 3-phenyl
substituent (R = CH3 or Cl) has little effect on anticon-
vulsant potency. Rather, as observed previously in the
analogous quinazolinone series,4 the lipophilicity (+p)
and position of substitution appear to be the important
factors for determining antiseizure efficacy.
Replacement of the methyl group at the 2-position of 3b
with an arylethenyl group to afford pyridopyrimidinones
5a–b led to a general decrease in anticonvulsant potency
relative to 3b, with styryl-substituted compound 5a exhi-
biting the greater activity and neurotoxicity (Table 1).
These results stand in contrast to the reported enhance-
ment of MES activity observed by Boltze1 following
similar alterations of methaqualone (1).
On the basis of their preliminary anticonvulsant poten-
tial, compounds 4b and 4c were selected for quantitative
evaluation of their pharmacological parameters as
shown in Table 2, where data for compounds 1, 2a–c,
3c and phenobarbital are also included for comparison.
Median effective doses (ED50s) and median neurotoxic
doses (TD50s) were determined at the time of peak anti-
convulsant and neurotoxic effect. The mouse ip data for
4c corroborate the preliminary findings that this com-
pound is a powerful antiseizure agent in both the MES
(ED50 = 6.82mg/kg) and scPTZ (ED50 = 6.98mg/kg) as-
says and that it is much more potent in both screens
than its corresponding unelaborated 3-methyl analog,
3c, which exhibited ED50s of 60.3mg/kg and 30.9mg/
kg, respectively. However, as a consequence of its neuro-
toxicity potency (TD50 = 17.37mg/kg), 4c has marginal
protective index (PI) values of ca. 2.5 in both tests. Com-
parison of the ED50 values of ip administered 4c
with quinazolinones 2a–c in the same murine model
reveals that 4c is more potent in both MES and scPTZ
tests.
Finally, although 4a–e and 5a–b like compound 6 prob-
ably exist as racemates of potentially resolvable atrop-
isomers,8 their resolution was not attempted because
of the modest antiseizure activity exhibited by 4d–e
and 5a–b (100–300mg/kg) and marginal separation of
anticonvulsant activity and neurotoxicity in 4a–c.
As part of the present study we explored possible mech-
anisms by which the newly synthesized pyridopyrimid-
inones 4 as well as their quinazolinone analogs 2 might
be exerting their anticonvulsant action. Our earlier
observation4 that quinazolinone 2b provided protection
against picrotoxin-induced seizures in mice suggested a
possible mechanism of anticonvulsant action for this
class of compounds, and perhaps for the structurally
similar pyridopyrimidinones 4, involving agonist activ-
ity at GABAA receptors.

5714
D. C. White et al. / Bioorg. Med. Chem. 12(2004) 5711–5717
Table 2. Quantification of anticonvulsant activity and neurotoxicity
PId
a,b
ED50 mg/kg
b,c
TD50 mg/kg
Compd
Test type
MES
scPTZ
MES
1.06
scPTZ
1.64
1e
ip (mice)
ip (mice)
po (rat)
ip (mice)
po (rat)
ip (mice)
po (rat)
ip (mice)
ip (mice)
po (rat)
po (rat)
ip (mice)
po (rat)
52
(48–56)
148g
33.5
(28–40)
100g
55
(47–65)
561g
2af
2af
2bf
2bf
2c
3.8
5.6
1.1
(112–196)
1.73i
(56–197)
2.57i
(425–733)
2.8i
1.7
(1.27–2.21)
196g
(2.00–3.10)
14.5g
(2.36–3.54)
262g
1.3
18
(155–240)
1.44i
(5.6–24.3)
3.09i
(125–400)
2.91i
2.0
0.94
1.97
(1.17–1.73)
21.8j
(2.41–3.81)
16.0j
(2.48–3.40)
31.5j
1.44
0.94
0.82
2.54
3.36
3.55
4.82
6.68
(17.8–26.9)
4.13g
(9.6–30.0)
2.66g
(26.0–43.1)
3.87i
2c
1.45
(2.83–5.01)
60.3h
(1.48–4.27)
30.9j
(2.59–5.01)
49.4i
3c
1.60
(44.6–77.2)
6.82j
(22.0–42.8)
6.98h
(46.6–52.8)
17.4j
4c
2.49
(5.85–8.27)
2.56g
(5.39–8.57)
<8.60i
—
>31i
(14.3–21.0)
8.61k
4c
<1.00
<0.29
7.69
(1.45–4.09)
2.49k
(5.70–12.9)
8.86h
4b
(1.78–3.22)
20.1
—
12.6
(6.90–12.1)
96.8
PBl
PB
(14.8–31.6)
9.14
(7.58–11.9)
(7.99–19.1)
11.6
(7.74–15.0)
(79.9–115)
61.1
(43.7–95.8)
5.29
a Determined at the time of peak effect (TPE).
b Values in parentheses are 95% confidence intervals determined by probit analysis.
c Determined at the time of peak neurologic deficit (rotorod test).
d Protective index (TD50/ED50).
e See Ref. 6.
f See Ref. 4.
g TPE = 2h.
h TPE = 0.50h.
i TPE = 4h.
j TPE = 1 h.
k TPE = 0.25h.
l PB = phenobarbital.
Using a modification of an established assay11 for meas-
uring radioactive chloride flux at the GABAA receptor,
quinazolinones 2a–b were found to activate 36Cl uptake,
with compound 2b (EC50 = 48 2.1lM) being more
than twice as potent as 2a (EC50 > 100lM).12 Confirma-
tion that this activation resulted from actions on the
GABAA receptor was obtained from the observation
that chloride uptake stimulated by 2b was completely
blocked by the GABAA receptor antagonists picrotoxi-
nin and bicuculline at concentrations of 100 and
30lM, respectively.
Additional experiments were performed in order to
identify other possible target sites of action for 4c. We
tested the ability of 4c to block 30lM veratridine stim-
ulated release of preloaded [3H]serotonin from mouse
cortical synaptosomes,13 which would indicate possible
involvement of voltage-sensitive sodium and calcium
channels in the action of this compound. However,
when applied at a concentration of 100lM, 4c had no
effect on veratridine-stimulated release of labeled
serotonin.
Similar screening of pyridopyrimidinone 4c for GABAA
receptor activation revealed that this compound was not
effective in increasing the level of chloride uptake. On
the contrary, preincubation of 4c showed it to be an
antagonist of GABA-stimulated chloride transport hav-
ing an IC50 value of 28 1.3lM. In addition, dose–
response studies with GABA demonstrated that the
primary effect of 4c (10lM) was to suppress the maxi-
mal extent of GABA-dependent chloride uptake, which
was decreased 36% compared to GABA alone.
3.Conclusions
Of the pyrido[2,3-d]pyrimidinones prepared and
screened in this study, compounds 4a–c containing
the
pharmacophoric
2-[-2-oxo-2-(4-pyridyl)ethyl]
group and a 2-substituted 3-aryl moiety exhibited
potent anti-MES and anti-scPTZ activity in murine
models. Following ip administration in mice, these
compounds were significantly more potent than
their 4(3H)-quinazolinone counterparts 2a–c for sei-

D. C. White et al. / Bioorg. Med. Chem. 12(2004) 5711–5717
5715
zure control, but were also considerably more
neurotoxic.
4.2. General procedure for the preparation of 2-methyl-3-
arylpyrido[2,3-d]-4(3H)-pyrimidinones
In spite of their enhanced anticonvulsant potency rela-
tive to 2a–c, compounds of structural type 4 did not
show the separation of anticonvulsant and neurotoxic
activity we hoped for when this study was initiated.
Although the MES and scPTZ ED50 values of 4c com-
pare very favorably with those of the prototype
drug phenobarbital (Table 2), the greater neurotoxicity
of 4c results in a much less attractive margin of
protection.
To a 250mL three-neck round bottom flask equipped
with a Dean–Stark trap was added 2-methyl-pyr-
ido[2,3-d][1,3]oxazin-4-one (7) (4.86g, 30.0mmol), 2-
bromoaniline (5.16g, 30mmol) and toluene (150mL).
As the mixture was heated under nitrogen, solution
was achieved. Water was azeotroped off as the solution
was heated under reflux for 8h. Upon cooling, 2-ace-
tamidonicotinic acid (1.30g) precipitated, which was fil-
tered and washed with toluene, mp 189–190°C (lit.1b mp
184–188°C). The orange mother liquor was concen-
trated to ca. 50mL and 6.48g (68%) of 2-methyl-3-
(2-bromophenyl)pyrido[2,3-d]-4(3H)-pyrimidinone (3c)
crystallized, mp 200–202°C. Recrystallization from
AcOEt gave 3c as pale-yellow crystals, mp 204–205°C.
1H NMR (270MHz, CDCl3) d 2.31 (s, 3H), 7.45 (m,
3H), 7.55 (m, 1H), 7.82 (dd, 1H, J = 1.4, 6.0Hz) 8.61
(dd, 1H, J = 2.0, 7.9Hz), 9.01 (dd, 1H, J = 2.0,
4.6Hz). 13C NMR (360MHz, CDCl3) d 24.0, 115.9,
122.3, 122.6, 129.3, 129.7, 131.3, 134.1, 136.5, 136.8,
156.3, 157.6, 157.7, 161.7. Anal. Calcd for
C14H10N3OBr: C, 53.15; H, 3.19; N, 13.29. Found: C,
53.01; H, 3.38; N, 13.26.
In in vitro testing, quinazolinones 2a and 2b were shown
to be moderate agonists at the GABAA receptor, with 2b
being more than twice as effective as an agonist com-
pared to 2a. This difference is reflected in their relative
effectiveness for suppressing picrotoxin-induced seizures
in vivo; ED50 = 153mg/kg for 2b, while 2a was ineffec-
tive in doses up to 600mg/kg.4 This correlation supports
the premise that effects on the GABAA receptor may
underlie at least some of the anticonvulsant effects of
these compounds.
In contrast to 2a and 2b, pyridopyrimidinone 4c showed
no GABAA agonist activity. Compound 4c also failed to
block presynaptic sodium and calcium channels. This
was surprising in light of its structural similarity to
2a–b and its potency against scPTZ-induced seizures.
Perhaps, like 2b and 6, the anticonvulsant action of 4c
may be directed at a different target, such as the AMPA
receptor.7,14 However, in light of its considerable neuro-
toxicity, which severely limits the potential of 4c as a
viable anticonvulsant drug candidate, further in vitro
studies were not performed. The antagonism of GABA
receptor function by 4c was similar in potency to that
of picrotoxinin (IC50 = 11lM) in chloride flux assays.11
This effect may play a role in the neurotoxicity observed
with compound 4c in whole animal screens.
4.3. 2-Methyl-3-phenylpyrido[2,3-d]-4(3H)-pyrimidinone
(3d)
From (7) (14.23g, 87.8 mmol) and aniline (8.15g,
88mmol) was obtained 10.9g (52%) of 3d16 mp 203–
204°C. 1H NMR (270MHz, CDCl3) d 2.31 (s, 3H),
7.39 (m, 5H), 7.56 (dd, 1H, J = 1.6, 6.2Hz), 8.55 (dd,
1H, J = 2.0, 7.8Hz), 8.97 (dd, 1H, J = 2.0, 4.6Hz). Anal.
Calcd for C15H13N3O: C, 71.70; H, 5.21; N, 16.72.
Found: C, 71.47; H, 5.29; N, 16.58.
4.4. 2-Methyl-3-(2-methylphenyl)pyrido[2,3-d]-4(3H)-
pyrimidinone (3a)
From (7) (9.42g, 58.1mmol) and o-toluidine (7.10g,
66mmol) was obtained 7.60g (52%) of 3a, mp 178–
179°C (lit.15 mp 178–179°C). 1H NMR (270MHz,
CDCl3) d 2.12 (s, 3H), 2.25 (s, 3H), 7.10 (d, 1H,
J = 7.0Hz), 7.37 (m, 4H), 8.59 (dd, 1H, J = 1.9,
7.9Hz), 8.98 (dd, 1H, J = 1.9, 4.6Hz).
4.Experimental
4.1. General
Melting points were determined on a Thomas–Hoover
melting point apparatus and are uncorrected. 1H
NMR spectra were recorded using either a Bruker WP
270 (270MHz) or an AM 360 (360MHz) instrument.
Combustion analyses were performed by Atlantic
Microlab, Inc., Norcross, GA. Flash chromatography
was carried out using 230–400 mesh Merck Kieselgel
60 silica gel. Tetrahydrofuran (THF) was dried by
distillation from a solution of potassium/benzophe-
none ketyl. 2-Methyl-pyrido[2,3-d][1,3]oxazin-4-one (7)
was obtained in 54% overall yield in four steps from
quinolinic acid by a modification of the procedure of
Preuss and Hoffman.9 2-Styryl-3-(2-methylphenyl)pyr-
ido[2,3-d]-4(3H)-pyrimidinone (5a) was prepared by
acid-catalyzed condensation of 3a with benzaldehyde
in 94% yield using the procedure of Kassem and
Soliman.15
4.5. 2-Methyl-3-(2-chlorophenyl)pyrido[2,3-d]-4(3H)-
pyrimidinone (3b)
From (7) (8.79g, 54.2mmol) and 2-chloroaniline (7.02g,
55mmol) was obtained 10.8g (73%) of (3b), mp 192–
193°C (lit.15 mp 194–195°C). 1H NMR (270MHz,
CDCl3) d 2.31 (s, 3H), 7.42 (m, 3H), 7.52 (m, 1H),
7.79 (dd, 1H, J = 1.4, 6.3Hz), 8.60 (dd, 1H, J = 2.0,
7.6Hz), 9.00 (dd, 1H, J = 2.0, 4.6Hz).
4.6. General procedure for the acylation of 2-methyl-3-
arylpyrido[2,3-d]-4(3H)-pyrimidinones
A 250mL three-neck round-bottom flask was equipped
with a stopper, a pressure-equalizing funnel and a con-
denser connected to a water-filled gas buret through a

5716
D. C. White et al. / Bioorg. Med. Chem. 12(2004) 5711–5717
drying tube filled with Drierite. The flask was charged
with NaH (1.25g of a 60% mineral oil dispersion,
31.3mmol). The mineral oil was removed by washing
several times with hexane followed by decantation and
the NaH was immediately covered with dry THF
(75mL). Ethyl isonicotinate (1.00g, 6.60mmol) was then
added, the stirred mixture was heated to reflux, the gas
buret was adjusted to zero volume, and a solution of
2-methyl-3-(2-methylphenyl)pyrido[2,3-d]-4(3H)-pyrimi-
dinone (3a) (1.51g, 6.00mmol) in THF(75mL) was
added dropwise over a period of 20min. Hydrogen evo-
lution commenced immediately and continued through-
out the addition period. The resulting red-brown
reaction mixture was heated at reflux until the theoreti-
cal amount of hydrogen had evolved, ca. 4h, and then
cooled to 0°C in an ice bath and acetic acid (2.0mL)
was then cautiously added dropwise. The solvent was re-
moved under reduced pressure and the solid residue was
partitioned between CH2Cl2 and water. The layers were
separated, the organic layer was washed with saturated
NaHCO3 solution, dried (MgSO4) and concentrated to
afford a brown semisolid, which was chromatographed
on silica gel. Elution with CH2Cl2 gave a small amount
of the starting ester. The product was then eluted with
CH2Cl2:AcOEt 1:1 as a yellow solid, mp 222–224°C,
1.61g (75%). Recrystallization from AcOEt afforded
an analytical sample of 2-[2-oxo-2-(4-pyridyl)ethyl]-
4.9. 2-[2-Oxo-2-(4-pyridyl)ethyl]-3-phenylpyrido[2,3-d]-
4(3H)-pyrimidinone (4d)
From 3d (2.30g, 9.69mmol), NaH (1.25g, 52.0mmol),
and ethyl isonicotinate (1.66g, 11.0mmol) was obtained
1.83g (55%) of 4d, mp 204–205°C. 1H NMR (360MHz,
CDCl3) d 5.17 (s, 1H), 7.45 (m, 8H), 8.43 (dd, 1H,
J = 2.0, 4.6Hz), 8.62 (dd, 2H, J = 1.5, 4.3Hz), 8.73
(dd, 1H, J = 2.0, 7.6Hz), 15.00 (s, 1H). Anal. Calcd
for C20H14N4O2: C, 70.17; H, 4.12; N, 16.37. Found:
C, 70.09; H, 4.10; N, 16.24.
4.10. 2-(2-Oxo-2-phenylethyl)-3-(2-methylphenyl)pyr-
ido[2,3-d]-4(3H)-pyrimidinone (4e)
From (3a) (3.20g, 12.7mmol), NaH (2.30g, 95.8mmol),
and methyl benzoate (2.83g, 20.8mmol) was obtained
2.72g (60%) of 4e, mp 216.5–217°C. 1H NMR
(360MHz, CDCl3) d 2.18 (s, 3H), 5.09 (s, 1H), 7.17
(m, 2H), 7.29 (m, 2H), 7.39 (m, 4H), 7.55 (m, 2H),
8.41 (dd, 1H, 1.6, 7.6Hz), 8.69 (dd, 1H, 1.6, 5.0Hz),
14.92 (s, 1H). Anal. Calcd for C22H17N3O2: C,
74.35; H, 4.82; N, 11.82. Found: C, 74.43; H, 4.89; N,
11.87.
4.11. 2-[2-(4-Pyridyl)ethenyl]-3-(2-methylphenyl)pyr-
ido[2,3-d]-4(3H)-pyrimidinone (5b)
3-(2-methylphenyl)pyrido
[2,3-d]-4(3H)-pyrimidinone
(4a), mp 225–226°C. 1H NMR (360MHz, CDCl3) d
2.19 (s, 3H), 5.10 (s, 1H), 7.26 (m, 2H), 7.45 (m, 5H),
8.42 (dd, 1H, J = 2.0, 7.8Hz), 8.61 (dd, 2H, J = 1.6,
4.5Hz), 8.71 (dd, 1H, J = 2.0, 4.8Hz), 14.97 (s, 1H).
13C NMR (360MHz, CDCl3) d 17.3, 81.3, 112.0,
120.3, 120.6, 128.1, 128.3, 130.4, 131.9, 133.9, 136.0,
137.2, 146.0, 150.2, 150.6, 154.8, 155.9, 159.6, 185.1.
Anal. Calcd for C21H16N4O2: C, 70.78; H, 4.53; N,
15.72. Found: C, 70.64; H, 4.61; N, 15.74.
This compound was prepared in 97% yield according
1
to a literature procedure,15 mp 205–206°C. H NMR
(360MHz, CDCl3) d 2.14 (s, 3H), 6.32 (s, 1H), 6.38
(s, 1H), 7.36 (m, 7H), 8.33 (dd, 1H, J = 1.8,
7.4Hz), 8.61 (dd, 2H, J = 1.6, 4.8Hz), 9.02 (dd, 1H,
J = 1.8, 4.6Hz). 13C NMR (360MHz, CDCl3)
d
17.6, 116.4, 122.4, 122.5, 127.9, 128.4, 130.3, 131.8,
135.1, 135.9, 136.8, 139.9, 142.1, 150.3, 154.0, 156.7,
157.7, 161.9. Anal. Calcd for C21H16N4O: C, 74.10;
H, 4.74; N, 16.46. Found: C, 74.40; H, 4.79; N,
16.29.
4.7. 2-[2-Oxo-2-(4-pyridyl)ethyl]-3-(2-chlorophenyl)pyr-
ido[2,3-d]-4(3H)-pyrimidinone (4b)
4.12. Pharmacology
From (3b) (1.83g, 6.74mmol), NaH (1.22g, 51mmol),
and ethyl isonicotinate (1.24g, 8.00mmol) was obtained
1.17g (46%) of 4b, mp 216–217°C. 1H NMR (360MHz,
CDCl3) d 5.06 (s, 1H), 7.24 (m, 2H), 7.40 (m, 2H), 7.55
(m, 1H), 7.63 (m, 1H), 7.84 (dd, 1H, J = 1.2, 5.4Hz),
8.42 (dd, 1H, J = 1.8, 4.6Hz), 8.62 (dd, 2H, J = 1.8,
4.4Hz), 8.72 (dd, 1H, J = 1.9, 7.6Hz), 14.87 (s, 1H).
Anal. Calcd for C20H13N4O2Cl: C, 63.75; H, 3.48; N,
14.87. Found: C, 63.90; H, 3.53; N, 14.66.
Anticonvulsant evaluation of candidate compounds 3–5
was conducted by the Epilepsy Branch of NINDS using
standard protocol (see Ref. 4).
4.13. In vitro biological assays
Methods for chloride flux assay of GABAA receptor
function were similar to those of Bloomquist and
Soderlund,11 except that 15s incubations with agonist
were used. For presynaptic [3H]serotonin release stud-
ies, we essentially followed the methods of Kirby
et al.,13 except that the mice used in the present
study were albino male ICR strain. Concentration–
response curves were replicated at least 3 times and
were analyzed by nonlinear regression to a four
parameter logistic equation to determine EC50 and
maximal uptake parameters using Prism (GraphPad
Software, San Diego, CA). Single concentration
experiments were also replicated 3 times and analyzed
by T-test or ANOVA using InStat (GraphPad Soft-
ware, San Diego, CA).
4.8. 2-[2-Oxo-2-(4-pyridyl)ethyl]-3-(2-bromophenyl)pyr-
ido[2,3-d]-4(3H)-pyrimidinone (4c)
From (3c) (2.43g, 7.68mmol), NaH (1.51g, 62.9mmol),
and ethyl isonicotinate (1.40g, 9.26mmol) was obtained
1.02g (32%) of 4c, mp 218–219°C. 1H NMR (360MHz,
CDCl3) d 5.05 (s, 1H), 7.29 (m, 2H), 7.43 (m, 2H), 7.52
(m, 1H), 7.60 (m, 1H), 7.87 (dd, 1H, J = 1.1, 5.7Hz),
8.43 (dd, 1H, J = 1.8, 4.5Hz), 8.63 (dd, 2H, J = 1.6,
4.4Hz), 8.73 (dd, 1H, J = 1.8, 7.6Hz), 14.19 (s, 1H).
Anal. Calcd for C20H13N4O2Br: C, 57.03; H, 3.11;
N,13.30. Found: C, 57.23; H, 3.17; N, 13.35.

D. C. White et al. / Bioorg. Med. Chem. 12(2004) 5711–5717
5717
7. Welch, W. M.; Ewing, F. E.; Huang, J.; Menniti, F. S.;
Pagnozzi, M. J.; Kelly, K.; Seymour, P. A.; Guanowsky,
V.; Guhan, S.; Guinn, M. R.; Critchett, D.; Lazzaro, J.;
Ganong, A. H.; DeVries, K. M.; Staigers, T. L.; Chenard,
B. L. Bioorg. Med. Chem. Lett. 2001, 11, 177.
8. Compound 6 exists as a pair of ( ) atropisomers. The
ED50 value given is that of the racemate. The more potent
(+) atropisomer has an ED50 of 1.5mg/kg.
Acknowledgements
We are pleased to acknowledge the generous financial
support of this work by the Harvey W. Peters Research
Center for ParkinsonÕs Disease and Disorders of the
Central Nervous System Foundation, from the National
Institutes of Health through STTR Grant 442730 and
from VirginiaÕs Center for Innovative Technology,
Grant B10-99-006. We also thank James P. Stables for
providing pharmacological data through the Antiepilep-
tic Drug Development Program, National Institutes of
Health. We are grateful to Rebecca L. Barlow for tech-
nical assistance in carrying out the chloride flux
experiments.
9. Preuss, F. R.; Hoffmann, E. Dtsch. Apoth. Ztg. 1976, 116,
893.
10. (a) Anticonvulsant Screening Project, Antiepileptic Drug
Development Program; National Institutes of Health,
DHEW Publ (NIH) (US), 1978; pp 78–1093; (b) Krall,
R. L.; Penry, J. K.; White, B. G.; Kupferberg, H. J.;
Swinyard, E. A. Epilepsia 1978, 19, 409; (c) Porter, R. J.;
Cereghino, J. J.; Gladding, G. D.; Hessie, B. J.; Kupfer-
berg, H. J.; Scotville, B.; White, B. G. Cleveland Clin. Q.
1984, 51, 293; (d) Swinyard, E. A.; Wolf, H. H.; Franklin,
M. R.; Chweh, A. Y.; Woodhead, J. H.; Kupferberg, H. J.;
Gladding, G. D. Technical Report for 79146 on Contract
No. N01-N5-4-2361; Epilepsy Branch, National Institute
of Neurological and Communicative Disorders and
Stroke: Bethesda, MD, 1984.
11. Bloomquist, J. R.; Soderlund, D. M. Biochem. Biophys.
Res. Commun. 1985, 133, 37.
12. A precise EC50 for 2a could not be determined due to
limited solubility, which precluded testing at concentra-
tions >200lM.
13. Kirby, M. L.; Barlow, R. L.; Bloomquist, J. R. Toxicol.
Sci. 2001, 61, 100.
14. Chenard, B. L.; Menniti, F. S.; Pagnozzi, M. J.; Shenk, K.
D.; Ewing, F. E.; Welch, W. M. Bioorg. Med. Chem. Lett.
2000, 10, 1203.
References and notes
1. (a) Boltze, Von K.-H.; Dell, H.-D.; Lehwald, H.; Lorenz,
D.; Ruberg-Schweer, M. Arzneimittel-Forsch 1963, 13,
688; (b) Kretzschmar, E. Pharmazie 1980, 35, 253.
2. Ionescu-Pioggia, M.; Bird, M.; Orzack, M. H.; Benes, F.;
Beake, B.; Cole, J. O. Int. Clin. Psychopharmacol. 1988, 3,
97.
3. See Witkin, J. M.; Katz, J. L.; Barrett, J. E. J.
Pharmacol. Exp. Ther. 1981, 218, 1, and references cited
therein.
4. Wolfe, J. F.; Rathman, T. L.; Sleevi, M. C.; Campbell, J.
A.; Greenwood, T. D. J. Med. Chem. 1990, 33, 161.
5. Preference for the enamine tautomeric structure for 2 is
¨
1
based on H NMR and X-ray crystallographic data, see:
Duke, N. E. C.; Codding, P. W. Acta. Cryst. 1993, B49,
719.
15. Kassem, M. G.; Soliman, F. S. G. Monatsh. Chem. 1983,
114, 1197.
16. Reddy, M. S.; Ratnam, C. V. Indian J. Chem. 1984, 23B,
977.
6. Vaidya, N. A.; Panos, C. H.; Kite, A.; Iturrian, W. B.;
Blanton, C. D. J. Med. Chem. 1983, 26, 1422.