1,4-Dihydro-2,3-quinoxalinediones as potential flavin metabolites and excitatory amino acid receptor ligands. Part 1: Synthesis and pharmacological evaluation of the benzylic oxidation series of 1,4-dihydro-6,7-dimethyl-2,3- quinoxalinedione

Article abstract of DOI:10.1016/S0968-0896(97)10032-3

A series of five 6,7-disubstituted 1,4-dihydro-2,3-quinoxalinediones was prepared, two of which are known microbial flavin metabolites and three of which are potential flavin metabolites. Four of the five compounds inhibited specific binding of [³H]-amino-3-hydroxy-5-methyl-4-isoxazolepropanoic acid ([³H]AMPA), [³H]kainic acid, and [³H]6-cyano-1,4-dihydro-7-nitro-2,3- quinoxalinedione ([³H]CNQX) in rat brain homogenate fractions, with IC₅₀ values in the low micromolar range (the fifth compound competed only with [³H]CNQX). Two of the compounds were moderately potent AMPA antagonists in an in vitro functional test.

Full text of DOI:10.1016/S0968-0896(97)10032-3

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 271±282
1,4-Dihydro-2,3-quinoxalinediones as Potential Flavin
Metabolites and Excitatory Amino Acid Receptor Ligands.
Part 1: Synthesis and Pharmacological Evaluation of the
Benzylic Oxidation Series of
1,4-Dihydro-6,7-dimethyl-2,3-quinoxalinedione
Ajita Bhat,ay Hui-Min Chang,a Lane J. Wallace,b David M. Weinstein,b
Gamal Shams,b Cynthia C. Garrisa and Ronald A. Hilla*
aDepartment of Molecular Pharmacology, School of Pharmacy, Northeast Louisiana University, Monroe, LA 71209-0470, U.S.A
bDivision of Pharmacology, College of Pharmacy, The Ohio State University, Columbus, OH 43210, U.S.A
Received 5 February 1997; accepted 14 October 1997
AbstractÐA series of ®ve 6,7-disubstituted 1,4-dihydro-2,3-quinoxalinediones was prepared, two of which are
known microbial ¯avin metabolites and three of which are potential ¯avin metabolites. Four of the ®ve compounds
inhibited speci®c binding of [3H]-amino-3-hydroxy-5-methyl-4-isoxazolepropanoic acid ([3H]AMPA), [3H]kainic acid,
and [3H]6-cyano-1,4-dihydro-7-nitro-2,3-quinoxalinedione ([3H]CNQX) in rat brain homogenate fractions, with IC50
values in the low micromolar range (the ®fth compound competed only with [3H]CNQX). Two of the compounds
were moderately potent AMPA antagonists in an in vitro functional test. # 1998 Elsevier Science Ltd. All rights
reserved.
Introduction
number of compounds with structural similarities to the
2,3-quinoxalinediones or 3-oxoquinoxaline-2-carboxylic
acids (3, Figure 1) have been synthesized as potential
therapeutic agents;3 such compounds appear especially
promising as early intervention agents in stroke.4
Certain synthetic 1,4-dihydro-2,3-quinoxalinediones,
such as 6-cyano-1,4-dihydro-7-nitro-2,3-quinoxaline-
dione (CNQX, 1; Figure 1) and the 6,7-dinitro analog
(DNQX, 2), antagonize excitatory responses elicited
from many neurons in the mammalian central nervous
system by glutamate and by the receptor-subtype-selec-
tive glutamate agonists AMPA and kainic acid.1 Exci-
tations elicited by N-methyl-d-aspartate (NMDA) are
also antagonized to a lesser extent, apparently due to
competition for the glycine coagonist binding site asso-
ciated with the NMDA receptor complex.2 A substantial
To date, no endogenous ligand for `quinoxalinedione
receptors' has been identi®ed. A few naturally occurring
2,3-quinoxalinediones have been reported in the
literature, either in plant sources or as microbial meta-
bolites, or precursors, of ribo¯avin (see below), and we
are therefore addressing the question of whether ¯avin
metabolism is somehow relevant to excitatory amino
acid neurotransmission; hypothetically, there might be
endogenous or naturally occurring xenobiotic, ¯avin-
related compounds that play neuroexcitatory, neuro-
modulatory, or neuroprotective roles, or act as
neurotoxins.
Key words: Quinoxalinedione; ribo¯avin; metabolite; gluta-
mate.
*Corresponding author. Tel: (318) 342-1706; fax: (318) 342-
1606; e-mail: pyhill@alpha.nlu.edu
yPresent address: University of Georgia at Athens, Department
of Medicinal Chemistry, School of Pharmacy, Athens, GA
30602, USA
In the late 1950s and early 1960s, Stadtman and
coworkers5±8 showed that a strain of Pseudomonas
0968-0896/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(97)10032-3

272
A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
ribo¯avinus (designated as Pseudomonas RF by these
workers), an aerobic soil bacterium that utilizes ribo-
¯avin as its sole source of energy, degrades ribo¯avin
via 2,3-quinoxalinedione derivatives 4 and 5 to 3,4-
dimethyl-6-carboxy-a-pyrone (6), ribose or ribitol, oxa-
mide, carbon dioxide, and urea (Figure 2). Kato and
Makida9 also reported 4 and 5 among the ribo¯avin
degradation products formed by a soil bacillus. Chemi-
cally, alkaline hydrolysis of ribo¯avin gives urea and the
1-d-ribityl-3-quinoxalinecarboxylic acid 7 (Figure 3).10
Alkaline hydrolysis with peroxide gives 4 or 5, depend-
ing on the conditions.11,12 It also was reported by
Miles et al.13 that the carboxylic acid 7 could be
oxidatively converted to 4 by hydrogen peroxide in
acetic acid (Figure 3); however, 7 may not be an inter-
mediate in the microbial conversion of ribo¯avin to 4.
In a model for such biological oxidations, alkaline
ferricyanide chemical oxidation of 414 gave 5 and the
quinoxaline-6-carboxylic acid 8, among other products
(Figure 4).
The 2,3-quinoxalinediones are not yellow, and have a
¯uorescence emission range of 200±350 nm. To date, it
appears that mammalian ¯avin metabolism studies have
concentrated on yellow-colored compounds with ¯uor-
escence emission at wavelengths above 500 nm. Fur-
thermore, radioisotopic studies of mammalian
ribo¯avin metabolism have only been carried out with
[2-14C]ribo¯avin; therefore, any 2,3-quinoxalinedione
metabolite would be lacking a radiolabel. ([14C]Urea has
been detected in various studiesÐsee Discussion sec-
tion.) 2,3-Quinoxalinediones as a class are polar com-
pounds, but are not very water-soluble. Their polarity
causes them not to partition eciently from aqueous
phases into water-immiscible organic solvents (such as
chloroform or phenol, two solvents that have commonly
been used in ¯avin metabolism studies); neither are they
readily soluble in such solvents.
A number of 2,3-quinoxalinediones and 3-oxoquin-
oxaline-2-carboxylic acids that are potential ¯avin
Figure 1. 2,3-Quinoxalinedione and 3-oxoquinoxaline-2-carboxylic acid excitatory amino acid receptor ligands.
Figure 2. Metabolism of ribo¯avin by Pseudomonas ribo¯avinus.

A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
273
metabolites have not been synthesized previously. The
availability of synthetic compounds as analytical stan-
dards facilitates the development of methods for their
detection and quantitation in biological samples.
Herein, we report the synthesis of the compounds
representing the benzylic oxidation series of 5, along
with the results of an initial pharmacological evaluation
that provides one index of the likelihood that these
compounds could play a neuromodulatory role in the
mammalian central nervous system.
Figure 3. Chemical degradation pathways of ribo¯avin leading to 2,3-quinoxalinediones and 3-oxoquinoxaline-2-carboxylic acids.
Figure 4. Alkaline ferricyanide oxidation as a model for oxidative metabolism of compound 4 by bacteria (from Ref. 17).

274
A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
Results
A number of approaches were pursued for synthesizing
the known (and deceptive by its apparent simplicity)
7-methylquinoxaline-6-carboxylic acid 8, and the result-
ing body of work will be the subject of another paper.
Herein we report the synthesis of this compound from
2-methyl-4,5-dinitrobenzoic acid (16), prepared essen-
tially by the method of Goldstein and Tardent,21 except
that the nitrile hydrolysis was carried out with basic
peroxide (Figure 7). Reduction to the diamine could be
accomplished either with hydrazine and Raney nickel or
tin metal in aqueous HCl. Careful optimization of the
reaction conditions was required with both methods
because further conversion of the diamine 17 to at least
one unidenti®ed material was problematic. The diamine
free base was unstable in neutral or basic solutions, and
such solutions quickly turned red and then brown.
Straightaway cyclization of the free base prepared by
the hydrazine/Raney nickel reduction to the 2,3-quin-
oxalinedione 8 was readily accomplished with diethyl
oxalate in re¯uxing THF, however.
Synthetic chemistry
The known15 compound 5 was synthesized by re¯uxing
3,4-dimethyl-1,2-phenylenediamine with diethyl oxalate
in tetrahydrofuran (THF) under a positive pressure of
argon gas (Figure 5).16 Compound 5 is poorly soluble in
solvents other than dimethylformamide or dimethyl
sulfoxide; to allow free-radical bromination, it was
therefore converted to the corresponding 2,3-dimethoxy-
quinoxaline: the 2,3-quinoxalinedione 5 was treated with
phosphorus oxychloride to provide the corresponding
2,3-dichloro derivative 9,17 which was further treated
with sodium methoxide in anhydrous methanol to yield
the dimethoxy compound 10.18 Free-radical bromina-
tion was carried out using N-bromosuccinimide in car-
bon tetrachloride, which gave compounds 11 and 12,
always together with unreacted starting material (10).
The bromomethyl and 6,7-bis(bromomethyl) com-
pounds 11 and 12 were separated from each other and
from 10 by column chromatography. Compound 11 was
hydrolyzed (Figure 6) in 1.2 N hydrochloric acid19 to
provide the 6-hydroxymethyl-7-methyl-2,3-quinox-
alinedione 13. Similarly, the bis(bromomethyl) deriva-
tive 12 was hydrolyzed to 14. The bromomethyl
compound 11 was also reacted with hexamethylenete-
tramine to prepare the aldehyde derivative 15 in a
Sommelet reaction (Figure 6).20
Pharmacological evaluations
The 6,7-dimethyl (5), 6-hydroxymethyl-7-methyl (13), 6-
formyl-7-methyl (15), and 6,7-bis(hydroxymethyl) (14)
2,3-quinoxalinediones were inhibitors of [3H]CNQX
and [3H]AMPA speci®c binding in rat brain homo-
genates, although not exceedingly potent; IC50 values
were in the low micromolar range (Table 1). The
Figure 5. Synthesis of intermediates 11 and 12. Reagents and conditions: (a) EtOCOCO2Et, THF, CH3CO2H (cat.), Ar(g), re¯ux; (b)
POCl3, N,N-dimethylaniline, re¯ux; (c) CH3ONa, CH3OH, re¯ux; (d) NBS, BzOOBz (cat.), CCl4, re¯ux; (e) silica chromatography.

A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
275
Figure 6. Synthesis of target compounds 13±15. Reagents and conditions: (a) 1.2 N HCl, Á; (b) C6H12N4; (c) CH3CO2H, re¯ux; (d)
HCl, re¯ux.
compounds were all about tenfold less potent as inhibi-
tors of speci®c [3H]kainate binding. Unlabelled CNQX
(1) and DNQX (2) were tested as reference compounds,
and were somewhat more potent in inhibiting the spe-
ci®c binding of all three radioligands. This is consistent
with the well-documented observation that electron-
withdrawing functional groups in the R6 or R7 positions
confer greater receptor anity, and in fact the most
potent compound in our series is the aldehyde 15. The
diol 14 was more active than we had expected, however.
Compounds 14 and 15 were modestly potent antago-
nists of AMPA-receptor-stimulated [3H]norepinephrine
release from rat hippocampal nerve endings in vitro,
though again somewhat less active than the standard
compound DNQX.
A nitro group is present at the R6 or R7 position in
many of the 2,3-quinoxalinediones known to exhibit the
most potent AMPA-receptor antagonism. Both DNQX
and CNQX are suciently acidic as to be pre-
dominantly ionized at physiological pH. The nitro
group would be a locus for a considerable portion of the
resulting negative charge (Figure 8). Even in the absence
of ionization, the electron-withdrawing nitro group is a
region of negative charge buildup. We therefore antici-
pated that the carboxylic acid 8 might be a potent
AMPA receptor ligand, because ionization of the carboxyl
group is essentially total under physiological conditions.
This compound proved to be inactive, however, as an
inhibitor of speci®c [3H]AMPA and [3H]kainate bind-
ing, and its modest inhibition of speci®c [3H]CNQX
binding might be due to activity at the NMDA-recep-
tor-associated glycine coagonist site.2 The isoalloxazines
ribo¯avin and lumi¯avin (18) were inactive, as was the
alloxazine lumichrome (19).
Figure 7. Synthesis of quinoxaline-6-carboxylic acid 8.
Reagents and conditions: (a) H2O2, 10% NaOH, re¯ux; (b)
HNO3, H2SO4 (1:1); (c) NH2NH2, Ra(Ni); (d) EtOCOCO2Et,
THF, CH3CO2H (cat.), Ar(g), re¯ux.
Discussion
Tillotson and Karcz22 con®rmed the presence in rat
urine of at least ten [14C]¯avin-related compounds

276
A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
Table 1. Radioligand binding and in vitro functional studies
Compound
R6
Ð
R7
Ð
IC50 [3H]CNQXa
IC50 [3H]AMPAb
IC50 [3H]KAINc
IC50 [3H]NE released
(ꢀM)
(ꢀM)
(ꢀM)
(ꢀM)
Ribo¯avin
18 (lumi¯avin)
19 (lumichrome)
Ie
I
I
I
I
NDf
ND
I
I
I
I
ND
1000
5
13
CH3
CH2OH
CHO
CH3
CH3
CH3
CH3
6.4‹1.3
33‹5.6
4‹1
16‹2.7
13‹2.9
11‹3.2
>100
>100
>100
54‹9
>100
>100
2.3‹0.7
2.2‹0.4
>1000
158‹43
ND
15
8
14
CO2H
20‹11
9.2‹3.4
0.2‹0.0
1.4‹0.1
CH2OH CH2OH
9.3‹6.4
0.56‹0.24
0.93‹0.42
450‹72
ND
41‹6
1 (CNQXa)
2 (DNQXg)
CN
NO2
NO2
NO2
Data are expressed as the mean‹sem of at least three independent determinations.
aCNQX: 6-cyano-1,4-dihydro-7-nitro-2,3-quinoxalinedione.
bAMPA: a-amino-3-hydroxy-5-methyl-4-isoxazolepropanoic acid.
cKAIN refers to kainic acid.
dAntagonism of [3H]norepinephrine release elicited by 100 mM AMPA from hippocampal slices; see text for details.
eNo detectable competition at 100 mM concentration.
fNot determined.
gDNQX: 6,7-dinitro-1,4-dihydro-2,3-quinoxalinedione.
Figure 8. Comparison of the conjugate base of CNQX (1) with the carboxylate anion of compound 8.

A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
277
resulting from metabolism and chemical degradation of
Experimental
orally administered [2 14C]ribo¯avin. These com-
pounds have not all been identi®ed, and only about 10±
20% of the total 14C was excreted in the urine; as much
as 9% of this urinary 14C appeared as [14C]urea.
[14C]Urea was also identi®ed in rat urine in
[2 14C]ribo¯avin metabolism studies by Christensen,23
though in smaller amounts. These results con®rm the
biological degradation of the pyrimidinoid portion of
the isoalloxazine ring of ¯avins, although it is still not
clear whether such degradation occurs somatically or
(more likely) through the action of microorganisms in
the gut in mammals. Various reports suggest (by physi-
cal descriptions of spots on paper and thin-layer chro-
matograms) the likely presence of quinoxalinediones in
human feces,24 rabbit urine,25 and the urines of goats,
calves, and sheep.26
Synthetic chemistry general procedures
Melting points were determined using a Thomas±
Hoover capillary melting point apparatus and are
reported without an emergent stem correction. Infrared
spectra of samples (as KBr pellets unless otherwise
noted) were obtained on a Nicolet Analytical 5MX FT
or Nicolet Impact 400D FT spectrometer. NMR spectra
were obtained with a JEOL FX-90 (90 MHz) spectro-
meter or a Bruker AM-400 (400 MHz) spectrometer.
Chemical shifts were reported in parts per million (d)
relative to tetramethylsilane (unless otherwise noted).
Electron-impact mass spectra were recorded on a
Finnigan MAT TQS 4510 or Finnigan MAT 900 double-
focusing magnetic sector spectrometer. Fast-atom bom-
bardment mass spectra (FABMS) were obtained on a
Finnigan MAT TSQ-70 equipped with an ANTEK
cesium ion source and an acceleration voltage of 5 kV;
or a Finnigan MAT 900 double-focusing instrument
with a Finnigan cesium ion source and an acceleration
voltage of 6 kV. HPLC analyses were performed on an
ISCO binary gradient chromatograph equipped with
two Model 2350 pumps, a Valco C6W injection valve
(VICI, Inc.), and a UA-5 ®xed-wavelength detector. All
analyses were carried out with wavelength cuto€ ®lters
to select the 254 nm or 280 nm mercury emission lines,
and using an Altex UltrasphereTM-Octyl 5 mm C8
(25 cmÂ4.6 mm) or ISCO 5 mm Spherisorb1 C18
(25 cmÂ4.6 mm) column. Mobile phases were isocratic
(compositions noted in each case below) or gradient
combinations of the following solvents: Solvent A, 5%
CH3CN/0.1% CF3COOH/H2O; Solvent B, 75%
CH3CN/0.1% CF3COOH/H2O. Elemental analyses were
performed by Galbraith Laboratories Inc., Knoxville,
TN, Oneida Research Services, Whitesboro, NY, or
Atlantic Microlabs, Inc., Norcross, GA. The abbrevia-
tions used are tetrahydofuran, THF; N,N-dimethyl
formamide, DMF; dimethylsulfoxide, DMSO. All
reagents and solvents were used as obtained without
further puri®cation, unless otherwise noted. All solvents
for recrystallization were analytical grade or better.
The 1-ribityl-2,3-quinoxalinedione 4 (see Figs 2±4) was
reported to be present in the pea plant and several other
plant species,27,28 and could apparently serve as a ribo-
¯avin precursor. Furthermore, juvenile rats fed com-
pound 4 exhibited considerably better growth than
those fed ¯avin-de®cient diets or diets containing 5.29
Although the latter could be due to a `¯avin-sparing
e€ect',30 these two observations suggest an alternate
biosynthetic pathway for ribo¯avin involving quinox-
alinone intermediates, essentially the reverse of the micro-
bial degradation pathway discussed above, which may
occur more widely in nature than heretofore recognized.
We have developed methods for extracting and analyz-
ing the 2,3-quinoxalinediones 5, 13, 15, and 8 in rat
brain, rat liver, rat urine, rat feces, rabbit urine, and
rabbit feces. The availability of synthetic analytical
standards was essential to this work. Preliminary results
indicate the presence of each of these compounds, in the
high parts-per-billion to low parts-per-million range, in
one or more of the matrices examined. To the best of
our knowledge, these compounds have never been
reported to be present in these matrices, or in any other
samples of mammalian or animal origin, and our methods
and results will be reported separately in the near
future.
Synthetic methods
1,4-Dihydro-6,7-dimethyl-2,3-quinoxalinedione (5). 4,5-
Dimethyl-1,2-phenylenediamine (3.54 g, 26.0 mmol) was
dissolved in 52 mL of THF; diethyl oxalate (24.2 g,
165 mmol) and six drops of glacial acetic acid were then
added. The ¯ask was ¯ushed with argon gas and main-
tained under a positive pressure of argon for three days
of re¯ux, during which time the product crystallized out
of the boiling solution. Upon cooling the reaction mix-
ture, diethyl ether (200 mL) was added, and the suspen-
sion was shaken vigorously and ®ltered to provide a
Conclusions
The four compounds representing the benzylic oxida-
tion series of 1,4-dihydro-6,7-dimethyl-2,3-quinoxaline-
dione have been synthesized. Three of the compounds
(but not the carboxylic acid 8) compete with radioligands
for binding to AMPA/kainate receptors, with IC50
values in the low micromolar range. Any possible
neurophysiological signi®cance remains to be demon-
strated.

278
A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
cream-colored solid. This solid was resuspended in di-
ethyl ether (200 mL), shaken vigorously, and ®ltered
again. This procedure was repeated twice more, and the
collected product was dried in air to a constant weight
to yield 4.76 g (96%) of 5 as a tan powder. Recrystalli-
zation from a large volume of methanol/ethanol gave
white plates, mp >340 ꢀC [lit.15a>250 ꢀC; lit.15b 340 ꢀC
(dec); lit.15c>325 ꢀC (acetic acid); lit.15d 340 ꢀC (dec)];
1H NMR (DMSO-d6) d 11.8 (s, 2H, NH), 6.89 (s, 2H,
under a dry argon atmosphere for 14 h. After cooling,
the resulting suspension was passed through glass wool
to remove the crystallized succinimide, and the ®ltrate
was concentrated in vacuo to a small volume. This
solution was chromatographed on a silica gel column
(45 cmÂ5.1 cm), eluting with benzene. The fractions
containing compound 11 were combined, as were those
for compound 12, and the benzene was distilled o€ on a
rotary evaporator to provide the compounds 11 (0.37 g,
54%) and 12 (0.14 g, 16%). Recrystallization of each
compound from ethyl acetate/hexane gave analytical
samples: 11, mp 148.5±149.5 ꢀC; 12, mp 154±155.5 ꢀC.
ArH), 2.18 (s, 6H, CH3); these shift values are consistent
1
with those reported by Cai et al.;15a IR (KBr, cm
3151 (NH), 1630 (C=O).
)
1
6,7-Dimethyl-2,3-dichloroquinoxaline (9). A solution of
compound 5 (2.38 g, 0.0125 mol, dried overnight in an
oven) in phosphorus oxychloride (10 mL, 0.11 mol) and
N,N-dimethylaniline (2.6 g, 0.021 mol) was re¯uxed for
8 h under a dry argon atmosphere. The reaction mixture
was then added very slowly with stirring into crushed ice
(300 g). The resulting precipitate was collected by suc-
tion ®ltration, washed with water, and dried to provide
an o€-white amorphous solid (2.6 g, 94%), mp 177±
179 ꢀC; mp 180±181 ꢀC after recrystallization from pet-
roleum ether [lit.31 mp 183 ꢀC; lit.32 mp 182 ꢀC; lit.15b
mp 270 ꢀC (petroleum ether; this value seems likely to be
Analytical data for compound 11: H NMR (CDCl3) d
7.73 (s, 1H, H5), 7.57 (s, 1H, H8), 4.64 (s, 2H, CH2),
1
4.13 (s, 6H, OCH3), 2.55 (s, 3H, CH3); IR (KBr, cm
)
2987 (CH), {1580, 1518, 1483, 1385} (ring stretches),
{1331, 1261} (C-O); MS m/e 298 (M+, 81Br), 296 (M+,
79Br), 217 (base, ± Br). Anal. calcd for C12H13BrN2O2:
C, 48.50; H, 4.41; N, 9.43; found: C, 48.52; H, 4.40; N,
9.35.
1
Analytical data for compound 12: H NMR (CDCl3) d
7.8 (s, 2H, H5), 4.83 (s, 4H, CH2), 4.15 (s, 6H, OCH3);
IR (KBr, cm 1) 2986 (CH), {1585, 1518, 1483, 1452,
1381} (ring stretches), {1336, 1260} (C-O); MS m/e 378
(M+, 81Br2), 376 (M+, 79Br81Br), 374 (M+, 79Br2), 297
1
erroneously reported)]; H NMR (CDCl3) d 7.8 (s, 2H,
ArH), 2.5 (s, 6H, CH3); IR (KBr, cm 1) {1618, 1540,
1478} (ring stretches), 1250s, 1195m, 1132s, 987s.
(base,
79Br), 295 (base,
81Br). Anal. calcd for
C12H12Br2N2O2: C, 38.33; H, 3.22; N, 7.45; found: C,
38.45; H, 3.23; N, 7.46.
2,3-Dimethoxy-6,7-dimethylquinoxaline (10). Compound
9 (1.63 g, 7.2 mmol) was added to a mixture of sodium
methoxide (0.90 g, 16.7 mmol) in dry methanol (10 mL),
and the mixture was re¯uxed under a positive pressure
of dry argon gas for 5 h. The reaction was cooled to
room temperature, ®ltered, and the ®lter cake was dried
overnight to obtain a yellow product. This product was
suspended in a minimal amount of ether, the suspension
was stirred for 30 min, and the product was collected by
suction ®ltration to yield a light-yellow amorphous solid
(1.48 g, 94%), mp 123±125 ꢀC. Further puri®cation
could be achieved either by (1) direct recrystallization
from ethanol or (2) partitioning between ethyl acetate
and water, evaporating the (dried) ethyl acetate in
vacuo, and recrystallizing the residue from ethanol; mp
139±140 ꢀC (ethanol), [lit18. 142±144 ꢀC (ethanol)]; 1H
NMR (CDCl3) d 7.67 (s, 2H, H5, H8; lit18. d 7.53 in
DMSO-d6), 4.15 (s, 6H, OCH3), 2.42 (s, 6H, CH3); IR
(KBr, cm 1) {1640, 1576, 1560, 1479, 1390} (ring stret-
ches), {1321, 1251} (C-O).
1,4-Dihydro-6-hydroxymethyl-7-methyl-2,3-quinoxaline-
dione (13). Compound 11 (0.50 g, 1.7 mmol) was
re¯uxed in 1.2 N HCl (30 mL) overnight, and then
allowed to cool. The precipitated solid was collected by
vacuum ®ltration. Chloroform (15 mL) was added to
this solid and the suspension was stirred at room tem-
perature for 10 min. The solid was again collected by
®ltration, dried in a stream of air, and recrystallized
from 95% ethanol to yield the dihydrate (0.20 g, 49%),
mp >250 ꢀC; 1H NMR (DMSO-d6) d 11.83 (bs, 2H,
NH), 7.18 (s, 1H, ArH), 6.87 (s, 1H, ArH), 5.14 (s, 1H,
OH), 4.45 (s, 2H, CH2), 2.18 (s, 3H, CH3); IR (KBr,
cm 1) 3440 (NH), 1685 (C=O), 1656 (ring stretch); MS
(FAB) m/e 206, 207 (M+1). A separate analytical sam-
ple of the monohydrate was prepared by recrystalliza-
tion from 95% ethanol. Anal. calcd for
.
C10H10N2O3 H2O: C, 53.57; H, 5.39; N, 12.49; found:
C, 53.61, 53.52; H, 5.42, 5.45; N, 12.55, 12.49.
6-Bromomethyl-7-methyl-2,3-dimethoxyquinoxaline (11)
and
(12). Compound 10 (0.50 g, 2.3 mmol) was dissolved in
carbon tetrachloride (10 mL). N-Bromosuccinimide
(0.41 g, 2.3 mmol) and benzoyl peroxide (0.079 g,
0.33 mmol) were added, and the solution was re¯uxed
1,4-Dihydro-6,7-bis(hydroxymethyl)-2,3-quinoxalinedione
(14). In a similar fashion as for 11, compound 12 (0.50 g,
1.3 mmol) was re¯uxed in 1.2 N HCl (30 mL) for 10 h.
On cooling, the solid precipitate was collected by ®ltra-
tion, resuspended in chloroform (15 mL), and stirred at
room temperature for 10 min. The product was again
6,7-bis(bromomethyl)-2,3-dimethoxyquinoxaline

A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
279
collected by ®ltration and dried in a stream of air to give
a
ArH), 8.23 (s, 1H, ArH), 2.68 (s, 3H, ArCH3); IR
(KBr, cm 1): 1712 (C=O), 1547 (NO2), 1368 (NO2),
1275 (C-O).
shiny, light-tan-colored solid (0.22 g, 69%), mp
>250 ꢀC; 1H NMR (DMSO-d6) d 11.89 (s, 2H, NH),
1
7.19 (s, 2H, Ar-H), 4.48 (s, 4H, CH2); IR (KBr, cm
)
1682 (C=O), 1390 (ring stretch), 1105 (C-O); MS
4,5-Diamino-2-methylbenzoic acid (17).
A
250 mL
.
(FAB) m/e 223 (M+1). Anal. calcd for C10H10N2O4 0.9
round-bottomed ¯ask was charged with 4,5-dinitro-2-
methylbenzoic acid (16; 1.25 g, 5.52 mmol), hydrazine
hydrate (100%, 3.2 mL), and ethanol (65 mL). A cata-
lytic amount of Raney nickel was added portionwise
with stirring while the reaction mixture was heated
under re¯ux.37 The reaction was monitored by HPLC
(30% CH3CN/0.1% CF3COOH/H2O, 1.2 mL/min;
tR ˆ 15:6 min and 2.2 min for the starting material and
product, respectively), and continued just to the point of
completion, about 2±3 h. The reaction mixture was ®l-
tered, and the ®ltrate concentrated under reduced pres-
sure to a semisolid residue, which was recrystallized
from ethanol to give 17 (688 mg, 74.8%). The pure
compound is presumably colorless, but the isolated
pinkish solid quickly turns red and eventually brown
upon standing, particularly when exposed to light; 1H
NMR (DMSO-d6) d 7.18 (s, 1H, ArH), 6.32 (s, 1H,
ArH), ꢁ4.25 (bs, NH), 2.33 (s, 3H, ArCH3). Elemental
analysis was obtained for the dihydrochloride salt (see
below).
H2O: C, 50.38; H, 4.99; N, 11.75; found: C, 50.57; H,
4.97; N, 11.75. (Water of hydration was con®rmed with
a micro loss-on-drying analysis).
6-Formyl-7-methyl-2,3-quinoxalinedione (15). Compound
11 (1.50 g, 5.05 mmol) was dissolved in chloroform
(8.4 mL), and a solution of hexamethylenetetramine
(2.40 g, 17.9 mmol) in chloroform (4.2 mL) was added
with vigorous stirring. A white solid separated almost
immediately. This suspension was re¯uxed for 30 min,
cooled, and the precipitate collected by ®ltration and
washed with light petroleum (5 mL) to give a white
solid. This solid was then combined with 50% aqueous
acetic acid (20 mL), and the mixture was re¯uxed for
2 h. Concd aq HCl (4.2 mL) was then added, and the
solution was re¯uxed for 5 min. The reaction mixture
was cooled and kept in a freezer overnight, and the
resulting solid was ®ltered o€ and recrystallized from
ethanol to provide light-tan crystals (0.80 g, 76%), mp
>250 ꢀC; 1H NMR (DMSO-d6) d 12.16 (s, 1H, N(1)H),
12.02 (s, 1H, N(3)H), 10.13 (s, 1H, CHO), 7.52 (s, 1H,
H5), 6.97 (s, 1H, H8), 2.57 (s, 3H, CH3); IR (KBr,
cm 1) 3590 (NH), 1716 (aldehyde C=O), 1685 (qui-
noxalinedione C=O); MS (FAB) m/e 204. Anal. calcd
4,5-Diamino-2-methylbenzoic acid dihydrochloride.
A
15 mL round-bottomed ¯ask was charged with 4,5-di-
nitro-2-methylbenzoic acid (16; 417 mg, 1.84 mmol) and
powdered tin (712 mg, 6.0 mmol), and equipped with a
re¯ux condenser. Concd HCl (4.5 mL) was added
slowly; the reaction is quite vigorous. The progress was
monitored by HPLC (¯ow rate 1.2 mL/min, eluting
3 min with Solvent A, then a linear gradient to Solvent B
over 9 min, then Solvent B for 4 min; starting material:
tR ˆ 14 min, product: tR ˆ 10 min), and the reaction
was continued just to the point of completion, about
25±40 min. The resulting mixture was diluted with water
(200 mL) to yield a clear solution, and hydrogen sul®de
gas was passed into the liquid until all the tin was pre-
cipated as sul®de (to monitor this process, a small sam-
ple was ®ltered from time to time and treated with
hydrogen sul®de to check for additional precipitate for-
mation). The precipitate was allowed to settle overnight,
the clear liquid was decanted, and the remaining residue
was gravity-®ltered through paper. The combined ®l-
trate and decantated liquid was evaporated in vacuo
until crystals appeared, and then cooled in ice. The
crystals were collected by suction ®ltration, washed with
a little concd HCl, and dried in air to yield 321 mg
(72.9%) of dihydrochloride, dp 218 ꢀC; 1H NMR
(DMSO-d6) d 7.78 (s, 1H, ArH), 6.64 (s, 1H, ArH), ꢁ6.5
.
for C10H8N2O3 0.25 H2O: C, 57.55; H, 4.11; N, 13.42;
found: C, 57.66; H, 4.26; N, 13.08.
2-Methyl-4-nitrobenzoic acid.
A
mixture of 6.7 g
(41 mmol) of 2-methyl-4-nitrobenzonitrile,21,33 10% aq
sodium hydroxide (81 mL), and 10% hydrogen peroxide
(42 mL) was brought to re¯ux. After 2.5 h, the water
circulation in the condenser was halted for 5±10 min to
remove dissolved ammonia. The water ¯ow was
restored, re¯ux was continued for 1 h, and the reaction
mixture was then allowed to cool. Concd HCl was
added until the product precipitated completely; it was
then collected by suction ®ltration and washed with a
little cold water. Recrystallization from hot water yiel-
ded 6.17 g (83%) of light-yellow needles, mp 149±152 ꢀC
(lit.33 151 ꢀC (water), lit.34 152 ꢀC (water(?)), lit.35 153±
154 ꢀC; lit.36 153±154 ꢀC); 1H NMR (DMSO-d6) d 8.21±
7.97 (m, 3H, ArH), 2.64 (s, 3H, ArCH3); IR (KBr,
cm 1): 1698 (C=O), 1526 (NO2), 1354 (NO2), 1275 (C-O).
4,5-Dinitro-2-methylbenzoic acid (16). The procedure
was that of Goldstein and Tardent.21 From 1.81 g
(1.00 mmol) of 2-methyl-4-nitrobenzoic acid, 2.02 g
(89%) of 16 was obtained (after recrystallization from
hot water) as pale-yellow needles, mp 184±189 ꢀC (lit.21
186 ꢀC (water)); 1H NMR (DMSO-d6) d 8.51 (s, 1H,
+
(v br s, NH3 and H2O), 2.41 (s, 3H, ArCH3). Anal.
.
.
calcd for C8H10N2O2 2HCl 0.75H2O: C, 38.34; H, 4.63;
N, 11.18; Cl, 28.29; found: C, 38.75, 38.69; H, 4.89, 4.92;
N, 11.26, 11.29; Cl, 28.25.

280
A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
1,4-Dihydro-7-methyl-2,3-dioxoquinoxaline-6-car-boxylic
acid (8). 4,5-Diamino-2-methylbenzoic acid (17, 436 mg,
2.63 mmol) was dissolved in THF (5.2 mL). Diethyl
oxalate (2.3 mL) was added, followed by two drops of
glacial acetic acid, and the mixture was re¯uxed under a
positive pressure of argon gas for three days. Upon cool-
ing the reaction mixture, diethyl ether (30 mL) was added
to make a suspension, which was shaken vigorously and
then ®ltered to provide a cream-colored solid. This solid
was resuspended in diethyl ether (30 mL) and ®ltered o€
again, and the collected product was dried in air over-
night to give 535 mg (91%) of 8 as an o€-white powder,
mp >250 ꢀC; 1H NMR (DMSO-d6) d 11.92 (bs, 2H,
NH), 7.68 (s, 1H, ArH), 6.95 (s, 1H, ArH), 2.48 (s, 3H,
CH3). Recrystallization from 95% ethanol gave small,
light-tan rhombic crystals of the monoethanolate; 1H
NMR (DMSO-d6) d 12.74 (bs, 1H, COOH), 12.04 (bs,
1H, NH), 11.92 (bs, 1H, NH), 7.68 (s, 1H, ArH), 6.95 (s,
1H, ArH), 4.36 (bs, 1H, CH3CH2OH), 3.37 (m, 2H,
CH3CH2OH), 2.48 (s, 3H, ArCH3), 1.05 (t, 3H, J ˆ 7:0
Hz, CH3CH2OH); 13C NMR (DMSO-d6) 168.2, 156.0,
155.4, 135.2, 129.0, 124.7, 124.0, 118.2, 117.8, 56.6
(CH3CH2OH), 21.7 (ArCH3), 19.7 (CH3CH2OH). Anal.
exactly the same as for the AMPA studies to the point
of the freeze/thaw cycle. After thawing, the membranes
were washed twice with 50 mM tris-HCl bu€er contain-
ing 2.5 mM CaCl2, pH 7.1. Binding was then performed
by incubating membranes with [3H]kainate (1 nM ®nal
concentration), and solutions of test compounds at
appropriate concentrations, in a 50 mM Tris-citrate
bu€er, pH 7.4. The mixture was incubated for 60 min on
ice before separating bound and free ligand by the
method described above.
CNQX binding assay. Inhibition of [3H]CNQX speci®c
binding was assessed according to the procedure of
Honore et al.39
Neurotransmitter release assay. This assay is based on
the fact that AMPA receptor activation induces nor-
epinephrine release from hippocampal nerve endings.40
The following procedure has been adapted from the
method of Desai et al.41 Brie¯y, mouse hippocampi were
dissected out and chopped into 0.3 mmÂ0.3 mm slices.
Slices were incubated with 0.2 mM [3H]norepinephrine in
Krebs bu€er for 30 min, and then transferred to super-
fusion chambers and washed for 60 min with warmed,
oxygenated Krebs bu€er at 0.3 mL/min. Following this,
ten 5 min fractions were collected. After the 3rd collec-
tion (i.e., fraction 3), bu€er containing 50 mM cyclo-
thiazide was introduced. The presence of cyclothiazide
diminishes AMPA receptor desensitization, thus allow-
ing for a more readily quanti®able AMPA response.
After the collection of fraction 4, bu€er containing
drug, cyclothiazide, and 100 mM AMPA was introduced
for 5 min, followed by normal bu€er until the end. For
initial screening purposes, tissue slices were exposed to
1 mM test compound alone (test for possible agonist
activity) run in parallel with tissue slices exposed to test
compound plus 100 mM AMPA (test for antagonist).
Each condition was run in triplicate. At least two
experiments were conducted to generate an n ˆ 6 for
each treatment. For dose-response data, a minimum of
four concentrations of each test compound were run
against 100 mM AMPA (for antagonists) or alone (for
agonists). Determination of IC50 values was accom-
plished utilizing median e€ect plot analysis (Chou and
Talalay, 1983).42
.
calcd for C10H8N2O4 CH3CH2OH: C, 54.13; H, 5.30; N,
10.52; found: C, 54.46; H, 5.50; N, 10.62.
Pharmacological testing
AMPA binding assay. The interaction of compounds
with AMPA receptors was assessed by inhibition of
speci®c [3H]AMPA binding in washed membrane pre-
parations of rat brain following the procedure originally
published by Honore et al.38 After the freeze/thaw cycle
and two additional washings, the preparation was incu-
bated with 0.04% Triton X-100 at 37 ꢀC for 30 min to
remove an inhibitor protein. After another wash, mem-
branes were incubated with [3H]AMPA (1.4 nM ®nal
concentration) and solutions of test compounds at
appropriate concentrations in 50 mM tris-HCl bu€er
containing 2.5 mM CaCl2 and 100 mM KSCN, pH 7.2.
The mixture was incubated on ice for 60 min with peri-
odic shaking, following which bound and free ligand
were separated by rapid ®ltration through Whatman
GF/C ®lter paper using a Brandel M-12R cell harvester
(Gaithersburg, MD). The resulting ®lter disks, contain-
ing membrane-bound radioligand, were dissolved in
Scintiverse-E1 cocktail for scintillation counting. Non-
speci®c binding was determined in the presence of 1 mM
glutamate. For determination of potency, a range of test
compound concentrations (0.01±100 mM) was tested.
IC50 values were calculated using nonlinear regression
analysis (InPlot1 software, GraphPad, Inc.).
Acknowledgements
This work was funded by NIH-NIGMS R15 Grant
#GM51040-01A1, and by Northeast Louisiana Uni-
versity. R.A.H. is heavily indebted to Professor Duane
D. Miller, University of Tennessee Health Sciences
Center, Memphis, as this project was conceived whilst
working in his group. The authors would like to grate-
fully acknowledge Professor Donald B. McCormick,
Emory University (consulting); Dr Jack Lay, National
Kainate binding assay. Interaction of compounds with
kainate receptors was assessed by inhibition of speci®c
[3H]kainate binding sites. Tissue preparation was

A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
281
Center for Toxicology Research, Je€erson, AR (mass
spectrometry); Drs Rolly Singh and Tracy McCarley,
Louisiana State UniversityÐBaton Rouge (mass spec-
trometry); Professor Upali Siriwardane, Louisiana Tech
UniversityÐRuston (NMR spectroscopy); Professor
Nikolaus Fischer and Dr Tiansheng Lu, Louisiana State
UniversityÐBaton Rouge (NMR spectroscopy); Dr
Satyavijayan N. Danthi, Ms Adrienne Butcher, Ms
Courtney Manasco, Ms Valerie Robinson, Ms Debbie
Hutto, M. Melinda Matthews, Professor A. Michael
Crider, Dr Leander J. Valdes, and Professor William J.
Keller.
7. Harkness, D. R.; Tsai, L.; Stadtman, E. R. Arch. Biochem.
Biophys. 1964, 108, 323.
8. Harkness, D. R.; Stadtman, E. R. J. Biol. Chem. 1965, 240,
4089.
9. Kato, Y.; Makida, H. Gifu Yakka Daigaku Kiyo 1964, 14,
38. (Also referenced in Chem. Abstr. 1966, 64, 16309g.).
10. Surrey, A. R.; Nachod, F. C. J. Am. Chem. Soc. 1951, 73,
2336.
11. Hotta, K.; Terao, M. Bitamin. 1959, 17, 539. (Also refer-
enced in Chem. Abstr. 1963, 59, 9018c).
12. (a) Ina, S. Tohoku Igaku Zassi 1959, 59, 13. (b) Idem. Ibid
1959, 59, 19.
13. Miles, H. T.; Smyrniotis, P. Z.; Stadtman, E. R. J. Am.
Chem. Soc. 1959, 81, 1946.
14. Saito, I.; Matsuura, T. Biochemistry 1967, 6, 3602.
References
15. (a) Cai, S. X.; Weber, E.; Keana, J. F. W.; Kher, S. PCT
Int Appl WO 95 12,417 (Cl. A61K51/00), 11 May 1995, US
Appl 148 259, 5 Nov 1993, 94,. (b) Oguchi, S.; Yamamoto, M.;
Aida, T.; Yoshino, R. Tokyo Gakugei Daigaku Kiyo, Dai-4-
bumon 1981, 33, 115. (c ) Ahmad, Y.; Habib, M. S.; Iqbal, M.;
Qureshi M. I. Ziauddin. Can. J. Chem. 1965, 43, 3424. (d)
Kazimierczuk, Z.; P¯eiderer, W. Liebigs. Ann. Chem. 1982, 754.
(e) Cheeseman, G. H. W. J. Chem. Soc. 1961, 1246.
1. (a) Blake, J. F.; Yates, R. G.; Brown, M. W.; Collingridge,
G. L. Br J Pharmacol. 1989, 97, 71. (b) Honore, T.; Davies, S.
N.; Drejer, J.; Fletcher, E.; Jacobsen, P.; Lodge, D.; Nielsen, F.
E. Science 1988, 241, 701. (c) Watkins, J. C.; Krogsgaard-Lar-
sen, P.; Honore, T. Trends Pharmacol. Sci. 1990, 11, 25.
2. (a) Verdoorn, T. A.; Kleckner, N. W.; Dingledine, R. Mol
Pharmacol. 1989, 35, 360. (b) Goldstein, J. M.; Litwin, L. C.
Mol. Chem. Neuropathol. 1993, 18, 145. (c) Randle, J. R. C.;
Guet, T.; Bobichon, C.; Moreau, C.; Curutchet, P.; Lambolezz,
B.; Carvalho, L. P. D.; Cordi, A.; Lepagnol, J. M. Molec
Pharmacol. 1992, 41, 337. (d) Lester, R. A.; Quarum, M. L.;
Parker, J. D.; Weber, E.; Jahr, C. E. Mol. Pharmacol. 1989, 35,
565.
16. Hill, Ronald A. Ph D Dissertation, The Ohio State Uni-
versity, 1991. Hill, R. A.; Wallace, L. J.; Miller, D. D.; Bhat, A.
R.; Supko, D.; Shams, G.; Tai, H.; Layer R. T.; Weinstein, D.;
Willins, D.; Uretsky, N. J. Document in preparation.
17. Cf.: Sastry, C. V. R.; Jogibhukta, M.; Krishnan, V. S. H.;
Rao, K. S.; Vemana, K.; Shridhar, D. R.; Tripathi, R. M.;
Verman, R. K.; Kaushal, R. Indian J. Chem. Sect. B 1988, 27B,
1110.
3. See, for example (and references therein): (a) Desos, P.;
Lepagnol, J. M.; Morain, P.; Lestage, P.; Cordi, A. A. J. Med.
Chem. 1996, 39, 197. (b) Guzikowski, A. P.; Cai, S. X.; Espita,
S. A.; Hawkinson, J. E.; Huettner, J. E.; Nogales, D. F.; Tran,
M.; Woodward, R. M.; Weber, E.; Keana, J. F. W. J. Med.
Chem. 1996, 39, 4643. (c) Ohmori, J.; Shimizu-Sasamata, M.;
Okada, M.; Sakamoto, S. J. Med. Chem. 1996, 39, 3971. (d)
Hays, S. J; Boxer, P. A.; Taylor, C. P.; Vartarian, M. G.;
Robichaud, L. J.; Nielsen, E. é. Bioorg. Med. Chem. Lett.
1993, 3, 77. (e) Watjen, F.; Nielsen, E, é.; Drejer, J.; Jensen, L.
H. Bioorg. Med. Chem. Lett. 1993, 3, 105. (f) Carling, R. W.;
Leeson, P. D.; Moore, K. W.; Smith, J. D.; Moyes, C. R.;
Mawer, I. M.; Thomas, S.; Chan, T.; Baker, R.; Foster, A. C.;
Grimwood, S.; Kemp, J. A.; Marshall, G. R.; Tricklebank, M.
D.; Saywell, K. L. J. Med. Chem. 1993, 36, 3397.
18. Leeson, P. D.; Baker, R.; Carling, R. W.; Curtis, N. R.;
Moore, K. W.; Williams, B. J.; Foster, A. C.; Donald, A. E.;
Kemp, J. A.; Marshall, G. R. J. Med. Chem. 1991, 34, 1243.
19. Cf.: Ohkawa, H.; Ohishi, N.; Yagi, K. J. Biol. Chem. 1983,
258, 5623.
20. Furniss, B. S.; Hanna€ord, A. J. editors. Vogel's Textbook
of Practical Organic Chemistry, 5th ed. Harlow:English Lan-
guage Book Society/Longman, UK, 1989, 1003±1004.
21. Goldstein, H.; Tardent, A. Helv. Chim. Acta. 1951, 34,
149.
22. Tillotson, J. A.; Karcz, M. S. J. Nutr. 1977, 107, 1269.
23. Christensen, S. Acta. Pharmacol. Toxicol. 1971, 29, 428.
See also Idem Ibid 1971, 30, 141; Idem Ibid 1971, 30, 177; Idem
Ibid 1973, 32, Suppl.), 3.
4. (a) Buchan, A. M.; Lesiuk, H.; Barnes, K.A.; Li, H.; Huang,
Z. G.; Smith, K. E.; Xue, D. Stroke 1993, 24 [suppl I] I148.
(b) Gill, R. Cerebrovasc. Brain. Metab. Rev. 1994, 6, 225. (c)
Shimizu-Sasamata, M.; Kawasaki-Yatsugi, S.; Okada, M.;
Sakamoto, S.; Yatsugi, S.; Togami, J.; Hatanaka, K.; Ohmori,
J.; Koshiya, K.; Usuda, S.; Murase, K. J. Pharmacol. Exp.
Ther. 1996, 276, 84. (d) Graham, S. H.; Chen, J.; Lan, J. Q.;
Simon, R. P. J. Pharmacol. Exp. Ther. 1996, 276, 1. (e) Gins-
berg, M. D. Neuroscientist 1995, 1, 95. (f) Sheardown, M. J.;
Suzdak, P. D.; Nordholm, L. Eur. J. Pharmacol. 1993, 236,
347.
24. Hotta, K.; Munetaka, M.; Hayashi, F.; Kato, S.; Sugito,
R. Bitamin 1958, 14, 57. (Also referenced in Chem. Abstr. 1960,
54, 22918a).
25. Owen, E. C.; West, D. W.; Coates, M. Br. J. Nutr. 1970,
24, 259.
26. Owen, E. C.; West, D. W. Br. J. Nutr. 1970, 24, 45.
27. Hotta, K.; Ando, O. J Vitaminol. (Kyoto) 1961, 7, 196.
28. Hotta, K.; Ando, O. Bitamin. (Kyoto) 1961, 23, 247. (Also
referenced in Chem. Abstr. 1964, 61, 16416a.).
5. Tsai, L.; Stadtman, E. R. Methods Enzymol. 1971, 18, 557.
29. Hotta, K.; Ishiguro, I.; Tanaka, K.; Ando, O.; Imaizumi, S.
Nagoya J. Med. Sci. 1958, 21, 46.
6. Tsai, L.; Smyrniotis P. Z.; Harkness, D.; Stadtman, E. R.
Biochem. Z 1963, 338, 561.

282
A. Bhat et al./Bioorg. Med. Chem. 6 (1998) 271±282
30. Ogunmodede, B. K.; McCormick, D. B. Proc. Soc. Exp.
Biol. Med. 1966, 122, 845.
37. Cf.: Sastry, C. V. R.; Rao, K. S.; Krishana, V. S. H.; Ras-
togi, K.; Jain, M. L. Indian J. Chem. Sect. B 1989, 28B, 888.
31. Papavassiliou, G. C.; Yiannopoulos, S. Y.; Zambounis, J.
S. Chem. Scr. 1987, 27, 265.
38. Honore, T.; Laridsen, J.; Krogsgaard-Larsen, P. J. Neu-
rochem. 1982, 38, 173.
32. Eicken, K.; Spiegler, W.; Wuerzer, B.; Meyer, N. Ger O€en
DE 3,533,791 (Cl. C07D241/44), 26 March 1987, 7.
39. Honore, T.; Drejer, J.; Nielsen, E. é.; Nielsen, M. Biochem.
Pharmacol. 1989, 38, 3207.
33. Mayer, M. J. Prakt. Chem. (2) 1915, 92, 137.
34. Jacobsen, O. Ber. Dtsch. Chem. Ges. 1884, 17, 162.
40. Pittaluga, A.; Raiteri, M. J. Pharmacol. Exp. Ther. 1992,
260, 232.
41. Desai, M. A.; Burnette, J. P.; Schoepp, D. D. J. Neu-
rochem. 1994, 63, 231.
35. Peltier, D. Bull. Soc. Sci. Bretagne. 1956, 31, 7. See also
Peltier, D. Compt. Rend. 1952, 235, 1660.
36. Hannig, E.; Wilhelm, H. Pharmazie 1969, 24, 29.
42. Chou, T. C.; Talalay, P. Trends Pharmacol. Sci. 1983, 4, 450.