Synthesis and anticonvulsant activity of N-benzyloxycarbonyl-amino acid prodrugs of phenytoin

Article abstract of DOI:10.1211/0022357991772835

Glycine, which has weak anticonvulsant properties, has been shown to potentiate the activity of several antiepileptic drugs but not phenytoin. Recently, studies have shown that N-(benzyloxycarbonyl)glycine (Z-glycine) antagonized seizures more than glycine in addition to possessing activity in the maximal electroshock test, a convulsive model in which glycine is inactive. In the present study esters of 3-hydroxymethylphenytoin, a phenytoin prodrug, and Z-glycine as well as the homologous N-(benzyloxycarbonyl)-ω-amino acids, z-β-alanine and Z-γ-aminobutyric acid (Z-GABA), were prepared and tested for their anticonvulsant and acute neurotoxic activities. The phenytoin prodrugs were obtained by esterification of bis(2-oxo-3-oxazolidinyl)phosphinic acid chloride-mediated esterification of 3-hydroxymethylphenytoin with the respective N-benzyloxycarbonyl-protected amino acid. The Z-glycine-phenytoin ester was the most active anticonvulsant derivative. Compared with phenytoin the compound exhibited a decreased median effective dose (ED50) in the MES test and an increased median toxic dose (TD50), resulting in an significantly improved protective index expressed as the ratio between TD50 and ED50. The present data suggest that covalent binding of phenytoin to Z-glycine results in an improved pharmacological profile of the drug.

Full text of DOI:10.1211/0022357991772835

J. Pharm. Pharmacol. 1999, 51: 549±553
Received October 23, 1998
Accepted January 4, 1999
# 1999 J. Pharm. Pharmacol.
Synthesis and Anticonvulsant Activity of N-Benzyloxycarbonyl-
Amino Acid Prodrugs of Phenytoin
GERHARD K. E. SCRIBA AND DIDIER M. LAMBERT*
È
È
Department of Pharmaceutical Chemistry, University of Munster, Hittorfstrasse 58-62, 48149 Munster,
Germany and *Laboratory of Medicinal Chemistry and Radiopharmacy, Catholic University of Louvain,
Avenue E. Mounier 73, CMFA 73.40, 1200 Brussels, Belgium
Abstract
Glycine, which has weak anticonvulsant properties, has been shown to potentiate the
activity of several antiepileptic drugs but not phenytoin. Recently, studies have shown that
N-(benzyloxycarbonyl)glycine (Z-glycine) antagonized seizures more than glycine in
addition to possessing activity in the maximal electroshock test, a convulsive model in
which glycine is inactive. In the present study esters of 3-hydroxymethylphenytoin, a
phenytoin prodrug, and Z-glycine as well as the homologous N-(benzyloxycarbonyl)-o-
amino acids, Z-b-alanine and Z-g-aminobutyric acid (Z-GABA), were prepared and tested
for their anticonvulsant and acute neurotoxic activities.
The phenytoin prodrugs were obtained by esteri®cation of bis(2-oxo-3-oxazolidinyl)-
phosphinic acid chloride-mediated esteri®cation of 3-hydroxymethylphenytoin with the
respective N-benzyloxycarbonyl-protected amino acid. The Z-glycine-phenytoin ester was
the most active anticonvulsant derivative. Compared with phenytoin the compound
exhibited a decreased median effective dose (ED50) in the MES test and an increased
median toxic dose (TD50), resulting in an signi®cantly improved protective index
expressed as the ratio between TD50 and ED50.
The present data suggest that covalent binding of phenytoin to Z-glycine results in an
improved pharmacological pro®le of the drug.
Glycine is one of the major inhibitory amino acids
in the central nervous system (CNS) acting via the
strychnine-sensitive glycine receptor localized
mainly in the brain stem and the spinal cord (Betz
1992). Simultaneously, glycine acts as a co-agonist
of the excitatory neurotransmitter glutamate by
interaction with the strychnine-insensitive glycine
binding site of the N-methyl-D-aspartate (NMDA)
receptor complex (Leeson & Iversen 1994).
In-vivo, glycine displays weak anticonvulsant
activity in several seizure models including the 3-
mercaptopropionic acid test and the strychnine test,
while it is inactive in the maximal electroshock test
and the pentylenetetrazole test (Lapin 1981; Toth et
al 1983). Recently, it was shown that N-(benzy-
loxycarbonyl)glycine (Z-glycine) exhibits a higher
anticonvulsant activity than glycine after intraper-
itoneal administration, especially in the maximal
electroshock test (Lambert et al 1994). Amide and
ester derivatives of Z-glycine further increased the
seizure-antagonizing action (Lambert et al 1995;
Geurts et al 1998).
Co-administration of glycine potentiates the
activity of several anticonvulsant drugs such as
phenobarbital, carbamazepine, diazepam and
valproate (Seiler & Sarhan 1984a, b; Toth & Lajtha
1984; Seiler et al 1985; Liu et al 1990; Peterson et
al 1990; Peterson 1991). However, co-administra-
tion of glycine does not alter the anti-convulsant
activity of phenytoin (Peterson et al 1990), one of
the most frequently used antiepileptic drugs for the
treatment of generalized tonic-clonic seizures (Rall
& Schleifer 1990). Moreover, large doses of gly-
1
cine ranging between 3 and 11 g kg
administered
(40±
sub-
1
150 mmol kg )
either
cutaneously or orally were necessary to exert
potentiation. Structural homologues of o-amino
acids such as b-alanine and g-aminobutyric acid
(GABA) were less effective than glycine (Seiler
& Sarhan 1984a). GABA is an inhibitory
Correspondence: G. K. E. Scriba, Department of Pharma-
ceutical Chemistry, University of MuÈnster, Hittorfstrasse 58-
62, 48149 MuÈnster, Germany.
E-Mail: scriba@uni-muenster.de

550
GERHARD K. E. SCRIBA AND DIDIER M. LAMBERT
neurotransmitter in the CNS while b-alanine acts as
an agonist at the strychnine-sensitive glycine
receptor (Marvizon et al 1986).
The present study was conducted to evaluate the
potential of prodrugs combining phenytoin and Z-
amino acids with regard to a synergism of the
compounds. The derivatives were synthesized by
5Á6 Hz); 13C NMR (50 MHz) CDCl3 d: 172Á2,
169Á1, 156Á3, 154Á6, 138Á5, 136Á6, 128Á9, 128Á8,
128Á6, 128Á2, 128Á1, 126Á8, 70Á4, 67Á2, 62Á4, 42Á5;
‡
MS m=z: 473 (M ), 266, 237, 194, 180, 165, 107,
91.
3-(N-Benzyloxycarbonylamino)-[(2,4-dioxo-5,5-
diphenyl-3-imidazolidinyl)methoxy]propionic acid
ester (Z-b-alanine-phenytoin). Yield 2Á0 g, 58%;
esteri®cation of 3-hydroxymethylphenytoin,
a
phenytoin prodrug (Varia et al 1984), with Z-gly-
cine, the homologous Z-b-alanine and Z-GABA
and with Z-phenylalanine as an amino acid devoid
of CNS activity.
1
mp 49±50ꢀC, H NMR (200 MHz) CDCl3 d: 7Á65
(1H, s), 7Á38±7Á30 (15H, m), 5Á55 (2H, s), 5Á37 (1H,
t, J ˆ 6 Hz), 3Á34 (2H, dt, J ˆ 6, 5Á8 Hz), 2Á48 (2H,
t, J ˆ 5Á8 Hz); 13C NMR (50 MHz) CDCl3 d: 172Á3,
171Á1, 156Á3, 154Á6, 138Á6, 136Á5, 129Á3, 128Á9,
128Á8, 128Á5, 128Á1, 126Á8, 70Á4, 66Á8, 62Á1, 36Á4,
‡
Materials and Methods
34Á2; MS m=z: 487 (M ), 266, 237, 194, 180, 120,
91.
General methods
4-(N-Benzyloxycarbonylamino)-[(2,4-dioxo-5,5-
diphenyl-3-imidazolidinyl)methoxy]butanoic acid
ester (Z-GABA-phenytoin). Yield 2Á7 g, 76%; mp
54±55ꢀC, 1H NMR (200 MHz) CDCl3 d: 8Á04 (1H),
7Á40±7Á30 (15H, m), 5Á51 (2H, s), 5Á08 (1H, s), 5Á03
(2H, s), 3Á12 (2H, t, J ˆ 7Á2 Hz), 2Á28 (2H, t,
J ˆ 7Á2 Hz), 1Á73 (2H, m); 13C NMR (50 MHz)
CDCl3 d: 172Á4, 171Á9, 156Á6, 154Á8, 138Á7, 136Á6,
128Á8, 128Á7, 128Á5, 128Á2, 127Á5, 126Á6, 70Á3, 66Á7,
Melting points were determined on a Ko¯er melting
point apparatus and are uncorrected. NMR spectra
were recorded on a Varian Gemini 200 (Varian
AG, Bremen, Germany). The chemical shifts are
reported as d (ppm) values relative to tetra-
methylsilane. Mass spectra were obtained on a
Varian MAT 44S (Varian AG, Bremen, Germany)
at 70 eV. Z-glycine, Z-phenylalanine and bis(2-
oxo-3-oxazolidinyl)phosphinic acid chloride were
obtained from Fluka (Deisenhofen, Germany). 3-
Hydroxymethylphenytoin (Varia et al 1984), Z-b-
alanine (Fiedler et al 1993) and Z-GABA (Blan-
kespoor et al 1984) were synthesized as described.
Methylene chloride was distilled over P2O5, trie-
thylamine was distilled over KOH.
‡
62Á0, 40Á2, 31Á0, 24Á9; MS m=z: 501 (M ), 394,
266, 237, 194, 180, 91.
2-(N-Benzyloxycarbonylamino) -3-phenyl-[(2,4-
dioxo-5,5-diphenyl-3-imidazolidinyl)methoxy]pro-
pionic acid ester (Z-phenylalanine-phenytoin).
Yield 2Á0 g, 49%; mp 67±68ꢀC,1H NMR
(200 MHz) CDCl3 d: 8Á01 (1H, s), 7Á40±7Á02
(20H, m), 5Á63 (1H, d, J ˆ 10Á3 Hz), 5Á53 (1H, d,
J ˆ 10Á3 Hz), 5Á32 (1H, d, J ˆ 8Á3 Hz), 5Á06 (1H, d,
J ˆ 12 Hz), 4Á98 (1H, d, J ˆ 12 Hz), 4Á64 (1H, m),
3Á00 (2H, m); 13C NMR (50 MHz) CDCl3 d: 172Á1,
170Á4, 155Á6, 154Á6, 138Á6, 138Á5, 135Á3, 129Á3,
128Á9, 128Á7, 128Á6, 128Á5, 128Á1, 128Á0, 127Á9,
127Á1, 126Á8, 70Á4, 67Á0, 62Á6, 54Á7, 37Á9; MS
Prodrug synthesis
Two grams (7Á1 mmol) of 3-hydroxmethylpheny-
toin, 7Á1 mmol of the respective Z-amino acid, 1Á5 g
(15 mmol) of triethylamine and 1Á9 g (7Á5 mmol) of
bis(2-oxo-3-oxazolidinyl)phosphinic acid chloride
were stirred in 50 mL methylene chloride under
nitrogen at room temperature for 2 h. The mixture
was poured into ice-cold 0Á05 M HCl and extracted
twice with methylene chloride. The organic layer
was dried over Na2SO4 and evaporated under
reduced pressure. Column chromatography on
silica gel (Merck, Darmstadt, Germany) using
methylene chloride±methanol (100 : 1, v=v) as
eluent yielded amorphous solids.
‡
m=z: 563 (M ), 413, 266, 237, 208, 194, 180,
160, 131, 91.
Prodrug hydrolysis
Hydrolysis was performed at 20Æ 0Á5ꢀC in 5% rat
plasma diluted with phosphate-buffered saline and
50 mM phosphate buffer, pH 7Á4 containing 30%
acetonitrile. Stock solutions of the compounds in
acetonitrile were added to the prewarmed incuba-
tion mixtures to give a ®nal concentration of 1 mM.
The reaction mixtures were vigorously stirred
during the kinetic run. Samples (100 mL) were
N-Benzyloxycarbonyl-[(2,4-dioxo-5,5-diphenyl-3-
imidazolidinyl)methoxy]acetic acid ester (Z-gly-
cine-phenytoin). Yield 2Á2 g, 65%; mp 62±
63ꢀC,1H NMR (200 MHz) CDCl3 d: 7Á66 (1H, s),
7Á38±7Á30 (15H, m), 5Á58 (2H, s), 5Á38 (1H, t,
J ˆ 5Á6 Hz), 5Á07 (2H, s), 3Á93 (2H, d, J ˆ

N-BENZYLOXYCARBONYL-AMINO ACID PRODRUGS OF PHENYTOIN
551
quenched by addition to 50 mL ice-cold 0Á5 M per-
chloric acid followed by dilution with 850 mL
water±acetonitrile (70 : 30, v=v) and assayed by
HPLC using a LiChrospher RP Select B column
(125 6 4Á6 mm, 5 mm) (Merck, Darmstadt, FRG)
equipped with a guard column (25 6 4Á6 mm,
5 mm) ®lled with the same adsorbent. The mobile
phase consisted of 0Á05 M phosphate buffer, pH 5Á0,
containing 40% acetonitrile (v=v). The ¯ow rate
1
was 1Á0 mL min . Phenytoin and prodrug con-
centrations were calculated from calibration curves
obtained by analysis of the pure compounds under
identical chromatographic conditions.
Pharmacology
Anticonvulsant testing was provided by the Anti-
epileptic Drug Development Program, Epilepsy
Branch, Division of Convulsive, Developmental
and Neuromuscular Disorders, National Institutes
of Health, according to standard procedures (Porter
et al 1984) and included the maximal electroshock
test and the seizure threshold test with sub-
cutaneous pentylenetetrazole. The acute neurolo-
gical toxicity was determined in the rotorod test.
For all these evaluations the compounds were dis-
solved or suspended in 0Á5% aqueous methyl cel-
lulose.
Figure 1. Structures of the Z-amino acid phenytoin esters.
doses (TD50) in the rotorod test, upon intraper-
itoneal administration of the compounds to mice,
are summarized in Table 1. All esters, as well as the
1 : 1 mixture, exhibited anticonvulsant activity in
the maximal electroshock test. Z-Glycine-pheny-
toin was the most effective derivative. Compared
with the administration of phenytoin on a molar
basis, the ED50 was reduced by about 10% while
the TD50 increased signi®cantly (approx. 2Á6-fold).
As a net result the protective index (the ratio
between TD50 and ED50) increased by a factor of
1Á7. Based on these data, Z-glycine-phenytoin is a
safer drug than the parent compound pheny-
toin. Interestingly, co-administration of equi-
molar amounts of Z-glycine and 3-hydroxy-
methylphenytoin also increased the protective
index due to a slight increase of the TD50, while
the ED50 in the MES test remained unaltered.
However, neither increase seen with the mixture
was signi®cantly different from phenytoin itself.
The Z-b-alanine and Z-GABA derivatives also
had an increased protective index, although to a
lesser extent than Z-glycine-phenytoin. On a molar
basis, the compounds were about half as active as
phenytoin, but also less toxic than phenytoin itself.
Both b-alanine and GABA were found to be less
effective at potentiating the seizure-antagonizing
activity of the GABA uptake inhibitor g-vinyl-
GABA in the 3-mercaptopropionic acid test when
compared with glycine (Seiler & Sarhan 1984a).
Esteri®cation with Z-phenylalanine, an amino acid
Results and Discussion
The structures of the phenytoin Z-amino acid esters
are given in Figure 1. They were readily obtained
by
bis(2-oxo-3-oxazolidinyl)phosphinic
acid
chloride-mediated esteri®cation of 3-hydroxy-
methylphenytoin with the respective N-benzyl-
oxycarbonyl-protected amino acid. The compounds
degraded slowly in aqueous buffer, pH 7Á4, with
half-lives between 208 h (Z-glycine-phenytoin) and
313 h (Z-GABA-phenytoin). In contrast, phenytoin
was quickly liberated from all esters by rat plasma
esterases (Table 1). Phenytoin was the only
detectable degradation product in the incubations.
The primary product of the hydrolysis, 3-hydroxy-
methylphenytoin, has been shown to liberate phe-
nytoin at physiological pH with a half-life of 2±4 s
(Varia et al 1984).
Pharmacological testing included the maximal
electroshock test and the subcutaneous pentylene-
tetrazole test as well as the determination of the
acute neurotoxicity using the rotorod test. For
comparison, an equimolar mixture of 3-hydroxy-
methylphenytoin and Z-glycine was also investi-
gated. The median effective doses (ED50) in the
maximal electroshock test and the median toxic

552
GERHARD K. E. SCRIBA AND DIDIER M. LAMBERT
Table 1. Median effective dose (ED50), median toxic dose (TD50) and protective index of phenytoin and the Z-amino acid
phenytoin esters after intraperitoneal administration to mice and half-lives of the plasma-esterase-catalysed in-vitro hydrolysis of
the ester derivatives.
1 a
)
1 a
)
Compound
MW
ED50 (mg kg
TD50 (mg kg
Protective index
(TD50=ED50)
Plasma half-life
(min)b
Phenytoinc
252Á3
491Á5
473Á5
487Á5
501Á5
563Á6
6Á5 (5Á6±7Á2) [2 h]
12Á4 (10Á1±15Á1) [2 h]
10Á6 (9Á2±12Á1) [2 h]
29Á0 (22Á5±35Á0) [8 h]
26Á2 (18Á8±35Á0) [2 h]
32Á6 (26Á1±42Á0) [2 h]
42Á8 (36Á4±47Á5) [2 h]
105.(90Á3±167) [2 h]
118.(100±138) [4 h]
256.(213±306) [8 h]
245.(220±262) [2 h]
193.(142±271) [4 h]
6Á6
8Á5
±
±
Phenytoin ‡ Z-glycined
Z-Glycine-phenytoin
Z-b-Alanine-phenytoin
Z-GABA-phenytoin
Z-Phenylalanine-phenytoin
11Á2
8Á8
1Á0Æ 0Á1
2Á3Æ 0Á4
1Á1Æ 0Á1
1Á6Æ 0Á3
9Á3
5Á9
aThe data were calculated from 5±6 doses (n ˆ 8±16 animals per dose). 95% con®dence intervals are given in parentheses. The
b
time of testing is listed in square brackets. 5% heparinized rat plasma in phosphate-buffered saline, 20ꢀC, meanÆ s.d., n ˆ 3±4.
d
cData from Dimmock & Baker (1994). Equimolar mixture of 3-hydroxymethylphenytoin and Z-glycine.
devoid of CNS activity, did not increase the pro-
tective index. Thus, combination of phenytoin with
an inhibitory amino acid appears to be necessary
for an improvement of the drug. Z-b-alanine, Z-
GABA and Z-phenylalanine did not display any
effect in the maximal electroshock test nor any
while the pharmacological pro®le remained unal-
tered. Z-Glycine-phenytoin was the most effective
compound. Co-administration of equimolar
amounts of Z-glycine and 3-hydroxymethyl-
phenytoin also increased the protective index
although the synergism was not as effective as by
covalent binding.
1
toxicity up to a dose of 300 mg kg .
As with phenytoin itself (Porter et al 1984;
Dimmock & Baker 1994), none of the investigated
compounds or the equimolar mixture of Z-glycine
and 3-hydroxymethylphenytoin, exhibited any sig-
ni®cant activity in the subcutaneous pentylene-
tetrazole test up to a dose of 300 mg kg 1. Thus,
combinations of phenytoin and Z-amino acids did
not alter the pharmacological pro®le of the drug, as
has been reported for dihydropyridine esters of 3-
hydroxymethylphenytoin (Shek et al 1989). The
dihydropyridine derivatives only slightly increased
the protective index after intravenous administra-
tion to rats (Shek et al 1989).
It is interesting to note that the time of the peak
effect of the maximal electroshock activity of Z-b-
alanine-phenytoin is 8 h compared with 2 h for the
other esters and phenytoin itself. The peak of the
neurotoxicity of all amino acid esters, as well as the
equimolar mixture of 3-hydroxymethylphenytoin
and Z-glycine, is also retarded compared with
phenytoin, although plasma esterase-catalysed
hydrolysis proceeded extremely quickly for all
esters. The rapid hydrolysis suggests a prodrug
mechanism for the Z-amino acid phenytoin esters.
However, the retardation of the peak effects sug-
gests that factors other than just hydrolysis into the
individual components of the prodrugs might con-
tribute to the pharmacological activity.
Acknowledgements
The anticonvulsant and toxicity testing by Dr J. P.
Stables and his staff at the NINDS Epilepsy
Branch, National Institutes of Health, Bethesda,
MD, and the ®nancial support by the Fonds der
Chemischen Industrie are gratefully acknowledged.
References
Betz, H. (1992) Structure and function of inhibitory glycine
receptors. Q. Rev. Biophys. 25: 381±394
Blankespoor, R. L., Lau, A. N. K., Miller, L. L. (1984)
Electroreductive cleavage of 2-methylene-9,10-anthraqui-
none (Maq) esters of carboxylic acids and N-substituted
carbamic acids: protecting groups of carboxylic acids and
primary amines. J. Org. Chem. 49: 4441±4446
Dimmock, J. R., Baker, G. B. (1994) Anticonvulsant activities
of 4-bromobenzaldehyde semicarbazone. Epilepsia 35: 648±
655
Fiedler, W. J., Guggisberg, A., Hesse, M. (1993) Synthetische
Analoga von niedermolekularen Spinnentoxinen mit Acyl-
polyamin-Struktur. Helv. Chim. Acta 76: 1167±1181
Geurts, M. A., Poupaert, J. H., Scriba, G. K. E., Lambert, D.
M. (1998) N-(Benzyloxycarbonyl)glycine esters and amides
as new anticonvulsants. J. Med. Chem. 41: 24±30
Lambert, D. M., Poupaert, J. H., Maloteaux, J.-M., Dumont, P.
(1994) Anticonvulsant activities of N-benzyloxycarbonyl-
glycine after parenteral administration. Neuroreport 5: 777±
780
Lambert D. M., Geurts, M. A., Scriba, G. K. E., Poupaert, J.
Â
In conclusion, esteri®cation of phenytoin with N-
benzyloxycarbonyl derivatives of inhibitory amino
acids resulted in an increase of the protective index
compared with the parent compound, phenytoin,
Â
Â
Â
H., Dumont, P. (1995) Derives simples d'acides amines
Â
neurotransmetteurs. Evaluation anticonvulsivante de derives
amides, carbamates et esters de la glycine et de la b-alanine.
J. Pharm. Belg. 50: 194±203

N-BENZYLOXYCARBONYL-AMINO ACID PRODRUGS OF PHENYTOIN
553
Lapin, I. P. (1981) Antagonism of glycine to seizures induced
byL-kynurenine and strychnine in mice. Eur. J. Pharmacol.
71: 495±498
Leeson, P. D., Iversen, L. L. (1994) The glycine site on
the NMDA receptor: structure-activity relationships
and therapeutic potential. J. Med. Chem. 37: 4053±
4067
Rall, T. W., Schleifer, L. S. (1990) Drugs effective in the therapy
of the epilepsies. In: Goodman-Gilman, A., Rall, T. W., Nies,
A. S., Taylor, P. (eds) The Pharmacological Basis of Ther-
apeutics. 8th edn, Pergamon Press, New York, pp 439±443
Seiler, N., Sarhan, S. (1984a) Synergistic anticonvulsant
effects of a GABA-T inhibitor and glycine. Naunyn-
Schmiedberg's Arch. Pharmacol. 326: 49±57
Liu, Z., Seiler, N., Marescaux, C., Depaulis, A., Vergnes, M.
(1990) Potentiation of g-vinylGABA (vigabatrin) effects by
glycine. Eur. J. Pharmacol. 182: 110±114
Seiler, N., Sarhan, S. (1984b) Synergistic anticonvulsant
effects of a GABA agonist and glycine. Gen. Pharmacol.
15: 367±369
Marvizon, J. C. G., Vasquez, J., Calvo, M. G., Mayor, T.,
Rui-Gomez, A., Valdivieso, F., Benavides, J. (1986)
The glycine receptor: pharmacological studies and mathe-
matical modeling of the allosteric interaction between the
glycine and strychnine binding sites. Mol. Pharmacol. 30:
590±597
Peterson, S. L. (1991) Glycine potentiation of anticonvulsant
drugs in pentylenetetrazole seizures in rats. Brain Res. Bull.
26: 43±47
Peterson, S. L., Trzeciakowski, J. P., Frye, G. D., Adams, H. R.
(1990) Potentiation by glycine of anticonvulsant drugs in
maximal electroshock seizures in rats. Neuropharmacology
29: 399±409
Porter, R. J., Cereghino, J. J., Gladding, G. D., Hessie, B. J.,
Kupferberg, H. J., Scoville, B., White, B. G. (1984) Anti-
epileptic drug development program. Cleveland Clin. Q. 51:
293±305
Seiler, N., Sarhan, S., Krogsgaard-Larsen, P., Hjeds, H.,
Schousboe, A. (1985) Ampli®cation by glycine of the anti-
convulsant effect of THPO, a GABA uptake inhibitor. Gen.
Pharmacol. 16: 509±511
Shek, E., Murakami, T., Nath, C., Pop, E., Bodor, N. S. (1989)
Improved anticonvulsant activity of phenytoin by a redox
brain delivery system III: brain uptake and pharmacological
effects. J. Pharm. Sci. 78: 837±843
Toth, E., Lajtha, A., Sarhan, S., Seiler, N. (1983) Anticonvul-
sant effects of some inhibitory neurotransmitter amino acids.
Neurochem. Res. 8: 291±302
Toth, E., Lajtha, A. (1984) Glycine potentiates the action of
some anticonvulsant drugs in some seizure models. Neuro-
chem. Res. 9: 1711±1718
Varia, S. A., Schuller, S., Sloan, K. B., Stella, V. J. (1984)
Phenytoin prodrugs III: water soluble prodrugs for oral
and=or parenteral use. J. Pharm. Sci. 73: 1068±1073