Synthesis, stability, and pharmacological evaluation of nipecotic acid prodrugs

Article abstract of DOI:10.1021/js980302n

Nipecotic acid (1), one of the most potent in vitro inhibitors of neuronal and glial γ-amino butyric acid (GABA) uptake, is inactive as an anticonvulsant when administered systemically. To obtain in vivo active prodrugs of (1), we synthesized four new nipecotic acid esters (3-6), which were obtained by chemical conjugation with glucose, galactose, and tyrosine. These compounds were assayed to evaluate their in vitro chemical and enzymatic hydrolysis. In addition, their anticonvulsant activity was evaluated in vivo in Diluted Brown Agouti (DBA)/2 mice, an excellent animal model for the study of new anticonvulsant drugs. Esters (3-6) appeared stable, at various temperatures, in a pH 7.4 buffered solution and showed susceptibility to undergoing in vitro enzymatic hydrolysis. Intraperitoneally injected nipecotic acid (1)and esters (3-5)did not protect mice against audiogenic seizures; conversely, nipecotic tyrosine ester (6) showed a significant dose-dependent anticonvulsant activity. The in vivo protective activity of the ester (6) and the inefficiency of nipecotic acid (1) in the same experimental conditions suggest that this ester prodrug could be actively transported intact across the blood-brain barrier, beyond which it could be hydrolyzed.

Full text of DOI:10.1021/js980302n

Synthesis, Stability, and Pharmacological Evaluation of Nipecotic Acid
Prodrugs
,
†
‡
‡
§
|
FRANCESCO PAOLO BONINA,* LOREDANA ARENARE, FRANCESCO PALAGIANO, ANTONELLA SAIJA, FELICE NAVA,
§
‡
DOMENICO TROMBETTA, AND PAOLO DE CAPRARIIS
Contribution from Dipartimento di Scienze Farmaceutiche, Facolt a` di Farmacia, Universit a` di Catania, Catania, Italy,
Dipartimento di Chimica Farmaceutica e Tossicologica, Facolt a` di Farmacia, Universit a` di Napoli Federico II, Napoli, Italy, and
Dipartimento Farmaco-Biologico, Facolt a` di Farmacia, and Istituto di Farmacologia, Facolt a` di Medicina e Chirurgia, Universit a`
di Messina, Messina, Italy.
Received July 27, 1998.
Final revised manuscript received January 20, 1999.
Accepted for publication January 22, 1999.
actively transported across the blood-brain barrier (BBB)
represents a plausible means of improving their brain
delivery by providing suitable substrates for active mem-
brane transport. In particular, conjugation with tyrosine
or glucose has recently been shown to be a successful means
of selective drug delivery; for example, phosphonoformate-
Abstract 0 Nipecotic acid (1), one of the most potent in vitro inhibitors
of neuronal and glial γ-amino butyric acid (GABA) uptake, is inactive
as an anticonvulsant when administered systemically. To obtain in
vivo active prodrugs of (1), we synthesized four new nipecotic acid
esters (3−6), which were obtained by chemical conjugation with
glucose, galactose, and tyrosine. These compounds were assayed
to evaluate their in vitro chemical and enzymatic hydrolysis. In addition,
their anticonvulsant activity was evaluated in vivo in Diluted Brown
Agouti (DBA)/2 mice, an excellent animal model for the study of new
anticonvulsant drugs. Esters (3−6) appeared stable, at various
temperatures, in a pH 7.4 buffered solution and showed susceptibility
to undergoing in vitro enzymatic hydrolysis. Intraperitoneally injected
nipecotic acid (1) and esters (3−5) did not protect mice against
audiogenic seizures; conversely, nipecotic tyrosine ester (6) showed
a significant dose-dependent anticonvulsant activity. The in vivo
protective activity of the ester (6) and the inefficiency of nipecotic
acid (1) in the same experimental conditions suggest that this ester
prodrug could be actively transported intact across the blood−brain
barrier, beyond which it could be hydrolyzed.
4
L-tyrosine conjugate is actively transported, by means of
active amino acid carriers, through monolayers of porcine
brain microvessel endothelial cells, and a glycosyl phos-
photriester prodrug of 3′-azido-3′-deoxythymidine (AZT)
5
shows good delivery to the CNS. Similarly, conjugating
glucose or galactose to poorly absorbable drugs can improve
their intestinal absorption by means of glucose transport
carriers in the small intestine;6 for example, tocopherol
conjugation to a monocarboxylate or glycoside moiety has
appeared to provide suitable substrates for active eryth-
rocyte membrane transport.9
-8
Nipecotic acid (1) is one of the most potent in vitro
inhibitors of neuronal and glial γ-aminobutyric acid (GABA)
10
uptake. Disorders such as Parkinson’s disease, Hunting-
ton’s chorea, and epilepsy may result from abnormalities
in the GABA system and, thus, enhancement of the activity
of GABA may be a useful treatment for these disorders.
Unfortunately, nipecotic acid is a highly polar compound
that does not easily penetrate the BBB, and is therefore
inactive as an anticonvulsant when administered systemi-
cally.1
Introduction
1,12
Several potentially central nervous system (CNS)-active
drugs have relatively unfavorable physicochemical char-
acteristics that hinder their transport into the brain. In
fact, the brain microvessel endothelium provides a barrier
to the passive transport of hydrophilic drugs into the brain.
Several strategies have been developed to overcome this
problem.1 The prodrug approach represents a very prom-
ising method to enhance drug delivery to the brain. In fact,
prodrugs are normally inactive and must generate active
drug at their target by enzymatically or chemically medi-
ated cleavage of their promoiety. The development of CNS-
active prodrugs has been generally aimed at obtaining an
improvement in the lipophilic character of the drug by
transiently masking ionized group(s) of the parent drug.
However, the preferential delivery of drugs or their deriva-
tives to brain may be improved by using endogenous
facilitated transport systems present at the blood-brain
interface. Thus, chemical conjugation of potentially CNS-
active drugs with an amino acidic or glycoside moiety
Successful prodrugs of nipecotic acid should be readily
transported into the CNS and hydrolyzed there to the
parent drug. Several esters of nipecotic acid (e.g., alkyl-,
substituted phenyl-, and triaryl-nipecotic esters) have been
synthesized with the goal to increase drug lipophilicity and
have been reported to have varying degrees of anticonvul-
sant activity and to inhibit in vitro GABA uptake after
conversion to the parent drug.1
-3
3-15
The purpose of our research was to synthesize new
nipecotic acid esters (3-6) obtained by chemical conjuga-
tion with essential nutrients, such as glucose, galactose,
or tyrosine, that are actively transported across the BBB.
These compounds were assayed to evaluate their in vitro
chemical stability and enzymatic hydrolysis. In addition,
their anticonvulsant activity was evaluated in vivo in
Diluted Brown Agouti (DBA)/2 mice, an excellent animal
model for the study of certain kinds of epilepsy and for
testing new anticonvulsant drugs.1
6,17
*
Author to whom correspondence should be addressed at Diparti-
Methods and Materials
mento di Scienze Farmaceutiche, Viale A. Doria 6, 95125 Catania,
Italy. Phone/Fax: 0039-095-222258. E-mail: boninaf@mbox.unict.it.
†
Ma ter ia lssNipecotic acid was obtained from Sigma Chemical
Company, Boc-L-tyrosine was purchased from Fluka Chemical
Company. All other reagent chemicals were from Aldrich Chemical
Company.
Dipartimento di Scienze Farmaceutiche.
Dipartimento di Chimica Farmaceutica e Tossicologica.
ipartimento Farmaco-Biologico.
‡
§
|
Istituto di Farmacologia.
©
1999, American Chemical Society and
10.1021/js980302n CCC: $18.00
Published on Web 03/19/1999
Journal of Pharmaceutical Sciences / 561
American Pharmaceutical Association
Vol. 88, No. 5, May 1999

Scheme 1sSynthesis of nipecotic acid eters (3−6): synthetic routes and chemical structures of the obtained compounds (abbreviations: TMCG,
,2,3,4-tetramethylcarbonate-D-glucose; DAG, diacetone-D-glucose; DIPG, 1,2,3,4-diisopropylidene-D-galactose).
1
Liquid Chromatographic (LC) grade acetonitrile and water, used
in high-performance liquid chromatography (HPLC), were ob-
tained from Merck and Carlo Erba, respectively.
N-Boc-L-tyrosine, afforded the corresponding esters (7-10). These
esters, submitted to deprotection, allowed us to obtain compounds
3-6.
Ap p a r a tu ssMelting points were taken on Buchi 510 capillary
melting point apparatus and are uncorrected.
N-Boc-Nipecotic Acid (2)sNipecotic acid (1; 1 g, 7.75 mmol) was
dissolved in dioxane (10.7 mL) and NaOH solution (1 N, 9.6 mL,
9.6 mmol). The resulting solution was stirred for 5 h, with the
addition of the di-tert-butyl-dicarbonate (1.9 g, 10.0 mmol). The
solvent was evaporated, and the resulting aqueous mixture was
layered with ethyl acetate (EtOAc; 15 mL). Then, 1 N HCl solution
was added with vigorous stirring until the pH of the aqueous phase
was 2. The layers were separated, and the aqueous layer was
extracted with EtOAc (3 × 10 mL). The EtOAc phases were
combined, washed with brine (6 mL), dried over Na2SO4, filtered,
and concentrated to give 1.42 g of (2) as a white solid (80% yield).
1H NMR (CDCl3): δ 4.1 + 3.85 (m, 2H, 2-H), 3.05 + 2.85 (m,
2H, 6-H), 2.50 (m, 1H, 3-H), 2.05 + 1.72 (m, 2H, 4-H), 1.70 + 1.46
(m, 2H, 5-H), 1.45 (m, 9H, -Boc).
N-Boc-Nipecotic Acid (1′,2′,3′,4′-O-Tetramethyl carbonate)-D-
glucopyranos-6′-yl Ester (7)sTo a stirred solution of N-Boc nipe-
cotic acid (2; 1 g, 4.36 mmol) in CH2Cl2-dimethyl formamide (DMF)
(9:1, 30 mL) was added CDI (706 mg, 4.36 mmol), and the mixture
was stirred for 1 h at room temperature (RT). A solution of 1,2,3,4-
tetramethyl carbonate-D-glucose (TMCG; 2.17 g, 5.26 mmol) in
CH2Cl2/DMF (1:5, 30 mL) was added in a dropwise manner and
The H NMR and 13C spectra were recorded on a Bruker WM
1
2
50 and on a Bruker AMX 500, respectively, using CDCl3 as
solvent. Chemical shifts are reported in ppm (δ) relative to
1
tetramethylsilane as internal standard for H NMR and in ppm
1
3
relative to the solvent for C NMR. Elemental analyses were
performed on a Carlo Erba model 1108 elemental analyzer.
The flash chromatography was performed on Merck silica gel
(
0.040-0.063 mm) column. The HPLC apparatus consisted of a
Varian 5000 system (Varian, Walnut Creek, CA) equipped with a
0-µL loop and an HP 1046A fluorescence detector (Hewlett-
2
Packard, D-76337 Waldbrom, Germany). Integration of the chro-
matographic peaks was achieved with a 4290 integrator (Varian).
Gen er a l Syn th etic Meth od sAll esters (3-6) were prepared
as outlined in Scheme 1.
Initially we started from N-Boc nipecotic acid (2) obtained by
treatment of (1) with di-tert-butyl-dicarbonate. Esterification of
1
(
2), in the presence of 1,1 -carbonyldiimidazole (CDI), with 1,2,3,4-
O-tetramethyl carbonate-D-glucose (TMCG), diacetone-D-glucose
DAG), 1,2:3,4-di-O-isopropylidene-D-galactopyranose (DIPG), or
(
5
62 / Journal of Pharmaceutical Sciences
Vol. 88, No. 5, May 1999

1
1
1
stirred for 24 h at RT. Most of the DMF was removed in vacuo,
and the reaction mixture was taken up in EtOAc (30 mL), washed
with water (20 mL) and brine (20 mL), and dried. Solvent was
removed in vacuo and the crude was purified by flash chromatog-
raphy eluting with hexane/EtOAc (6:4) to provide 1.90 g of (7) as
a pale yellow oil (69.5% yield).
H NMR (CDCl3): δ 5.50 (m, 1H, 1 -H), 4.60 (m, 1H, 3 -H), 4.32
1 1 1 1
(m, 1H, 2 -H), 4.20 (m, 3H, 6 -H and 5 -H), 4.00 (m, 1H, 4 -H),
2.95 (m, 2H, 2-H), 2.75 (m, 1H, 3-H), 2.75 + 2.50 (m, 2H, 6-H),
2.05 + 1.70 (m, 2H, 4-H), 1.55 + 1.40 (m, 2H, 5-H), 1.40 (m, 21H,
Boc and ketals).
Nipecotic Acid D-Galactos-6′-yl Ester (5)sTo a solution of 9 (1.75
g, 3.71 mmol) in CH2Cl2 (5 mL), TFA (5 mL) was added, and the
mixture was stirred at RT for 2 h. Evaporation of the solvent gave
a residue that was taken up in water (20 mL), neutralized with
10% aqueous NH4OH, diluted with more water, and washed with
CHCl3 (3 × 15 mL). The aqueous layers were concentrated in
vacuo, and the crude was submitted to cation-exchange chroma-
tography, eluting with 10% pyridine in water to give 915 mg of 5
(84% yield) as pale yellow solid (mp 138-139 °C).
1
1
1
H NMR (CDCl3): δ 6.00 (m, 1H, 1 -H), 5.40 (m, 1H, 3 -H), 4.92
1
1
1
1
(
m, 1H, 2 -H), 4.60 (m, 2H, 6 -H), 4.15 (m, 2H, 4 -H and 5 -H),
4
.05 + 3.85 (m, 2H, 2-H); 3.65 + 2.75 (m, 2H, 6-H), 3.55 (m, 12H,
methyl carbonate groups), 2.55 (m, 1H, 3-H), 2.15 + 1.80 (m, 2H,
4
-H), 1.45 + 1.35 (m, 2H, 5-H), 1.20(m, 9H, Boc).
Nipecotic Acid D-glucos-6′-yl Ester (3)sTo a solution of 7 (1.90
g, 3.04 mmol) in CH2Cl2 (5 mL), 5 mL of trifluoroacetic acid (TFA)
was added, and the mixture was stirred at RT for 2 h. Evaporation
of the solvent gave a residue that was poured into a saturated
solution of NaHCO3 (20 mL) and extracted with CHCl3 (3 × 15
mL). The organic layers were dried over Na2SO4 and concentrated
in vacuo. This material (1.5 g) was taken directly into the next
reaction without further purification. To a solution of crude (1.5
g) in MeOH (10 mL) was added K2CO3 (100 mg), and the mixture
was stirred at RT for 12 h. Evaporation of the solvent gave a
residue that was taken up in water and neutralized with 6 N
aqueous HCl, diluted with more water, and washed with CHCl3.The
aqueous layers were concentrated in vacuo, and the residue was
submitted to cation-exchange chromatography, eluting with 10%
pyridine in water to give 610 mg of (3) (69% yield) as a pale yellow
solid (mp 145-146 °C).
1
1
1
H NMR (D2O): δ 4.95 (m, 1H, 1 -H), 4.28 (m, 1H, 3 -H), 3.60
1
1
1
(m, 1H, 2 -H), 3.42 (m, 2H, 6 -H), 3.35 (m, 1H, 5 -H), 3.25 (m, 1H,
1
4 -H), 3.10 + 2.95 (m, 2H, 2-H), 2.80 (m, 2H, 6-H), 2.33 (m, 1H,
3-H), 1.75 + 1.65 (m, 2H, 4-H), 1.45 (m, 2H, 5-H).
1
3
1
C NMR (D2O + MeOD): δ 179.41 (sCOOR), 96.72 (C-1 ),
1
1
1
1
1
75.15 (C-3 ), 72.04 (C-2 ), 70.40 (C-5 ), 68.82 (C-4 ), 60.97 (C-6 ),
46.01 (C-2), 43.99 (C-6), 40.76 (C-3), 25.73 (C-4), 21.01 (C-5).
N-Boc-Nipecotic Acid N-Boc-L-Tyrosyl Ester (10)sCDI (0.71 g,
4.37 mmol) was added to a solution of 2 (1 g, 4.37 mmol) in CH2-
Cl /DMF (3:1, 50 mL), and the resulting mixture was stirred at
2
RT. After disappearance of the starting materials (determined by
thin-layer chromatography), N-Boc-L-tyrosine (1.11 g, 4.15 mmol)
in CH Cl /DMF (3:1, 150 mL) was added in a dropwise manner to
2
2
1
1
1
H NMR (D2O): δ 4.65 (m, 1H, 1 -H), 4.60 (m, 1H, 3 -H), 3.70
the mixture over 120 min, and stirring was then continued for 12
h. After evaporation of the solvent, the residue was taken up in
EtOAc (60 mL) and washed with water (2 × 30 mL). The organic
layer was dried on Na2SO4 and evaporated to give a residue (3 g)
that was submitted to flash chromatography, eluting with CHCl3/
MeOH/AcOH (95:4.5:0.5) to afford the title compound (10) as a
pale yellow oil (1.5 g, 71%).
1
1
1
1
(
m, 1H, 2 -H), 3.50 (m, 3H, 6 -H and 5 -C), 3.15 (m, 3H, 3 -C and
2
2
-H), 2.95 + 2.75 (m, 2H, 6H), 2.45 (m, 1H, 3-H), 1.90 + 1.65 (m,
H, 4-H), + 1.55 (m, 2H, 5-H).
1
3
1
C NMR (D2O + MeOD): δ 172 (-COOR), 96.84 (C-1 ), 76.86-
1
1
1
1
7
6
5
6.66 (C-3 ), 75.71 (C-2 ), 75.31-75.08 (C-5 ), 70.56 (C-4 ), 63.57-
3.33 (C-6 ), 48.78 (C-2), 45.02-44.91 (C-6), 39.57 (C-3), 26.63 (C-
1
1
1
1
), 21.98 (C-4).
H NMR (CDCl3): δ 6.85 (d, 2H J ) 7 Hz, 2 -H and 6 -H), 6.64
1 1
N-Boc Nipecotic Acid Diacetone-R-D-glucofuranos-3′-yl Ester (8)s
(d, 2H J ) 7 Hz, 3 -H and 5 -H), 4.91 (br, 1H, -NH-Boc), 4.45
1 1
To a stirred solution of N-Boc-nipecotic acid (2; 1 g, 4.36 mmol) in
CH2Cl2/DMF (9:1, 30 mL) was added CDI (706.97 mg, 4.36 mmol),
and the mixture was stirred for 1 h at RT. A solution of DAG (1.36
g, 5.23 mmol) in CH2Cl2/DMF (9:1, 30 mL) was added in a dropwise
manner and stirred for 24 h at RT. Most of the DMF was removed
in vacuo, and the reaction mixture was taken up in EtOAc (30
mL), washed with water (20 mL) and brine (20 mL), and dried.
Solvent was removed in vacuo, and the crude was purified by flash
chromatography, eluting with hexane/Et2O (8:2) to provide 1.34 g
of 8 as a white oil (65% yield).
(m, 1H, 8 -H), 4.15 + 3.80 (m, 2H, 2-H), 2.85 (m, 2H, 7 -H), 2.84
+ 2.75 (m, 2H, 6-H), 2.40 (m, 1H, 3-H), 2.00 + 1.43 (m, 2H, 4-H),
1.40 (m, 20H, 5-H and -Boc).
Nipecotic Acid L-Tyrosyl Ester (6)sTo a solution of 10 (1.75 g,
3.71 mmol) in CH2Cl2 (5 mL), TFA (5 mL) was added, and the
mixture was stirred at RT for 2 h. Evaporation of the solvent gave
a residue that was taken up in water (20 mL), neutralized with
10% aqueous NH4OH, diluted with more water, and washed with
CHCl3 (3 × 15 mL). The aqueous layers were concentrated in
vacuo, and the crude was submitted to cation-exchange chroma-
tography, eluting with 10% pyridine in water to give 760 mg of 6
(73% yield) as a white solid (mp 134-135 °C).
1
1
1
H NMR (CDCl3): δ 5.85 (m, 1H, 1 -H), 5.25 (m, 1H, 3 -H), 4.41
1
1
1
1
(
m, 1H, 2 -H), 4.15 (m, 3H, 6 -H and 5 -H), 4.05 (m, 1H, 4 -H),
1
1
1
3
2
.95 (m, 2H, 2-H), 2.85 + 2.72 (m, 2H, 6-H), 2.48 (m, 1H, 3-H),
H NMR (DMSO): δ 7.30 (d, 2H J ) 7 Hz, 2 -H and 6 -H), 7.20
1 1 1
.00 + 1.70 (m, 2H, 4-H), 1.70 + 1.55 (m, 2H, 5-H), 1.40 (m, 21H,
(d, 2H J ) 7 Hz, 3 -H and 5 -H), 4.00 (m, 1H, 7 -H), 3.50 + 3.30
1
-Boc and ketals).
(m, 2H, 2-H), 3.19 (m, 2H, 6 -H), 3.05 (m, 2H, 6-H), 2.95 + 1.82
Nipecotic Acid D-Glucos-3′-yl Ester (4)sTo a solution of (8; 1.34
(m, 2H, 4-H), 1.82 + 1.70 (m, 2H, 5-H).
1
3
g, 2.84 mmol) in CH2Cl2 (5 mL) was added TFA (5 mL), and the
mixture was stirred at RT for 12 h. Evaporation of the solvent
gave a residue that was taken up in water (20 mL) and neutralized
with 10% aqueous NH4OH, diluted with more water, and washed
with CHCl3 (3 × 15 mL). The aqueous layers were concentrated
in vacuo, and the residue was submitted to cation-exchange
chromatography, eluting with 10% pyridine in water to give 690
mg of 4 (84% yield) as a pale yellow solid (mp 135-136 °C).
C (DMSO): δ 173.50 (-COOH); 172.84 (-COOs), 156.35 (C-
1
1 1 1 1
4 ), 134.64 (C-1 ), 131.75-131.63 (C-2 and C-6 ), 116.79 (C-3 and
1
1
C-5 ), 55.87 (C-7 ), 44.96 (C-2), 44.84 (C-6), 39.20 (C-3), 36.31-
1
36.06 (C-6 ), 25.52 (C-5), 21.73 (C-4).
Ch em ica l Sta bilitysNipecotic ester solutions were prepared
by dissolving an aliquot of 3-6 in pH 7.4 phosphate buffer to give
-
5
a final concentration of ≈10 M. The solution was maintained at
37 °C, and aliquots were withdrawn every 2 h for the initial 12 h
of incubation and successively every 12 h for 7 days. The disap-
pearance of the nipecotic esters was followed by HPLC analysis
using the method reported later. Pseudo-first-order rate constants
for chemical hydrolysis were determined from slopes of linear plots
obtained by reporting the logarithm of residual nipecotic ester
against time. All experiments were carried out in triplicate.
En zym a tic Sta bilitysEnzymatic hydrolysis of nipecotic esters
(3-6) was determined using the procedure described in the
literature.18 Porcine liver esterase (obtained from Sigma) was
diluted 10 times with phosphate buffer and used to hydrolyze
nipecotic esters. Nipecotic ester solutions were prepared by
dissolving an aliquot of 3-6 in phosphate buffer to give a final
concentration of ≈10 M. The solution was maintained at 37 °C,
and 325 µL of porcine esterase were added to achieve a concentra-
tion of 1.3 U/mL. Aliquots of 300 µL were withdrawn every hour
for 16 h and combined with 600 µL of 0.01 N HCl in methanol.
After centrifugation at 5000 × g for 5 min, an aliquot of
supernatant was derivatized by the method reported later and
1
1
1
H NMR (D2O): δ (m, 1H, 1 -H), 4.41 (m, 1H, 3 -H), 3.65 (m,
1
1
1
1
4
1
H, 2 -H), 3.50 (m, 2H, 6 -H), 3.21 (m, 1H, 5 -H), 3.15 (m, 2H,
1
-H), 3.15 + 3.03 (m, 2H, 2-H), 2.90 + 2.81 (m, 2H, 6-H), 2.42 (m,
H, 3-H); 1.81 + 1.67 (m, 2H, 4-H), 1.51 (m, 2H, 5-H).
1
3
C NMR (D2O + MeOD): δ 180.03 (sCOOR), 97.17-93.30 (C-
11
6
), 77.96 (C-3 ), 77.06 (C-2 ), 75.44 (C-5 ), 70.85 (C-4 ), 61.87 (C-
), 48.53 (C-2), 46.78 (C-6), 39.44 (C-3), 26.60 (C-5), 22.01 (C-4).
N-Boc-Nipecotic Acid (1′,2′:3′,4′-diisopropylidene)-R-D-galacto-
1
1
1
1
1
pyranos-6′-yl Ester (9)sTo a stirred solution of N-Boc-nipecotic acid
(2; 1 g, 4.36 mmol) in CH2Cl2/DMF (9:1, 30 mL) was added CDI
(706 mg, 4.36 mmol), and the mixture was stirred for 1 h at RT.
A solution of DIPG (1.36 g, 5.23 mmol) in CH2Cl2/DMF (1:5, 30
mL) was added in a dropwise manner and stirred for 24 h at RT.
Most of the DMF was removed in vacuo, and the reaction mixture
was taken up in EtOAc (30 mL), washed with water (20 mL) and
brine (20 mL), and dried. Solvent was removed in vacuo, and the
crude was purified by flash chromatography eluting with hexane/
Et2O (8:2) to provide 1.75 g of 9 (85% yield) as a white oil.
-
5
Journal of Pharmaceutical Sciences / 563
Vol. 88, No. 5, May 1999

then monitored by HPLC. Pseudo-first-order rate constants for
enzymatic hydrolysis were determined from the slopes of linear
plots obtained by reporting the logarithm of residual nipecotic ester
against time. All experiments were carried out in triplicate.
Table 1sChemical and Enzymatic Stability of Esters 3−6
t
1/2 (h)a
drug
pH 7.4 buffer
esterase (1.3 U/mL)
An a lytica l P r oced u r essFor dansyl derivatization of nipecotic
acid (1) and of 3-5, 100 µL of sample or of stock standard solutions
3
4
5
6
150.00
169.42
197.48
27.37
8.30
10.70
14.0
2.5
(prepared in pH 7.4 phosphate buffer) were combined with 9 µL
of dansyl chloride (1.5 mg/mL in acetonitrile) and 100 µL of 80
mM Li2CO3, pH 9.5. The mixture was kept at room temperature
for 2 h in darkness. The reaction was stopped by adding 10 µL of
a
1
1
t
1/2 was calculated from the equation: t1/2 ) (ln 0.5)/K , where K is the
2
% methylamine hydrochloride solution. For derivatization of
pseudo-first-order rate constant.
nipecotic tyrosine ester (6), 125 µL of sample or of stock standard
solutions (prepared in pH 7.4 phosphate buffer) were combined
with 55 µL of fluorescamine (2 mg/mL in acetone) and 125 µL of
Results and Discussion
1
20 mM phosphate buffer, pH 8.0; then, the mixture was mixed
by vortex for a few minutes. An aliquot of derivatized drug
solutions (20 µL) was analyzed by the chromatographic method
described next.
Data concerning chemical and enzymatic stability of
ester prodrugs 3-6 are reported in Table 1. Figure 1 and
Figure 2 show the times courses of disappearance of
nipecotic esters (3-6) in pH 7.4 buffer solution and in the
presence of porcine esterase. When we evaluated their
chemical stability, the compounds appeared stable in a pH
HPLC analysis was performed with an ODS Hypersil column
(
particle size, 5 µm; 125 × 4 mm i.d.; Hewlett-Packard, D-76337
Waldbrom, Germany). The mobile phase consisted of sodium
acetate buffer (100 mM)/acetonitrile (75:25). The flow rate was set
at 1.0 mL/min. Each sample was filtered prior to injection with a
Millex HV13 filter (Waters-Millipore Corporation, Milford, MA),
and an aliquot (20 µL) was injected into the HPLC apparatus.
Dansyl derivatives of nipecotic acid (1) and 3-5 were monitored
at 330 nm (excitation wavelength) and 510 nm (emission wave-
length), and the fluorescamine derivative of the nipecotic tyrosine
ester (6) was monitored at 260 nm (excitation wavelength) and
7
.4 buffered solution; however, prodrugs 3-5 disappeared
more slowly than the ester 6.
To confirm that the tested prodrugs can be enzymatically
hydrolyzed, we evaluated their stability in the presence of
porcine esterase. The findings indicate that each of the four
esters 3-6 is capable of being cleaved in vitro by esterase.
Half-life times concerning enzymatic stability were
notably lower than those obtained in buffer solution.
Moreover, we found a good correlation between chemical
and enzymatic decomposition rates of the prodrugs exam-
ined.
Table 2 shows the results obtained in in vivo experiments
with a genetically seizure-prone strain (DBA/2) of mice.
DBA/2 mice are a useful model to test compounds acting
on the GABAergic system;20 furthermore, this model avoids
the problem of possible effects of tested substances on the
4
70 nm (emission wavelength). The retention times of dansyl
derivatives of 1 and 3-5 were 6.29, 6.7, 6.9, and 7.37 min,
respectively. The retention time of the fluorescamine derivative
of ester 6 was 4.65 min. The limit of sensitivity was <2 µM for
detection of dansyl derivatives of 1 and 3-5 and <3 µM for
detection of the fluorescamine derivative of the ester 6.
An ticon vu lsiva n t ActivitysDBA/2 mice, purchased from
Charles River (Calco, Como, Italy), were used to evaluate the
anticonvulsant activity of 1 and 3-6. The DBA strain of the house
mouse, Mus musculus, inbred since 1909, has been known since
1
947 to be susceptible to sound-induced seizures. Nearly 100% of
absorption or metabolism of the convulsant compound that
arises with drug-induced seizures.2
1,22
According to data
the males and females of the DBA/2 strain undergo an age-
dependent, often fatal, sequence of convulsions (a wild running
phase, followed by clonic convulsions and tonic extensions, ending
in respiratory arrest or full recovery) when initially exposed to a
loud mixed-frequency sound (12-16 kHz; 90-120 dB) such as a
door bell. The age of maximum susceptibility of the DBA/2 strain
to the sound-induced seizures has been reported as 20-39 days,
reported by other authors, we confirmed that nipecotic acid
(1; 0.75 mmol/kg), given systemically 30 min before sound
11,12
exposure, possesses no anticonvulsant effect.
Also, 3-5,
intraperitoneally injected in the same experimental sched-
ule at dose levels up to 0.75 mmol/kg, did not protect mice
against all phases of audiogenic seizures. Conversely, when
the ester prodrug 6 was administered systemically, a
significant dose-dependent anticonvulsant activity was
observed. In fact, a marked reduction in the incidence of
each phase (wild running, clonus, tonus, respiratory arrest)
of audiogenic seizures was evident at the dose levels
studied (0.125, 0.17, 0.21, 0.25, 0.5, and 0.75 mmol/kg).
Furthermore, no behavior alteration or change in rectal
temperature in comparison with control animals were
recorded after administration of 3-6 (data not shown).
The in vivo protective activity of systemically adminis-
tered nipecotic tyrosine ester 6 against audiogenic seizures
and the inefficiency of nipecotic acid (1) under the same
experimental conditions suggest that this ester prodrug is
transported intact across the BBB, beyond which it could
be hydrolyzed. Tyrosine enters the brain through the large
neutral amino acid transporter system. This system,
present in the brain microvessel endothelial cells, has been
extensively characterized and is of particular interest in
2
1-28 days, or 16-26 days. Other strains of audiogenic mice often
require acoustic priming in the neonatal period to attain a similar
degree of seizure susceptibility.
1
9
In our study, male DBA/2 mice (10 for each experimental group),
1-28 days old, were exposed to auditory stimulation (109 dB for
0 s or until tonic extension occurred), 30 min following injection
2
6
of the drugs tested 1 and 3-6 or of a same volume of their vehicle.
Compounds 1 and 3-5 were dissolved in isotonic phosphate buffer,
pH 7.4; 6 was dissolved in 20% EtOH/H2O (v/v). The injection
volume (both for drug solutions and the vehicle alone) was 0.1 mL/
1
0 g body weight.
The intensity of seizure response (SR) was assessed on the
following scale: 0 ) no response, 1 ) wild running, 2 ) clonus, 3
tonus, and 4 ) respiratory arrest. The maximum response was
)
recorded for each animal. Rectal temperature was recorded
immediately prior to auditory testing using an Elektrolaboratoriet
thermometer type T.E.3. Behavioral changes (spontaneous activ-
ity, explorative behavior, tremor, spontaneous convulsions) were
observed during the period between drug administration and
auditory testing.
2
3,24
the transport of amino acid drugs into the brain.
Statistical comparisons between controls and drug-treated
groups were carried out with Fisher’s exact probability test
4
Moreover, Walker and co-workers have recently demon-
strated that a phosphonoformate-L-tyrosine conjugate
possesses a high affinity for large amino acid transporters
in monolayers of porcine brain microvessel endothelial cells
and so might be a substrate for facilitated transport at the
BBB. As to our findings, one tempting hypothesis is that
the tyrosine moiety acts as a vector, which should allow
(incidence of seizure phases) and the Mann-Whitney U test. The
percentage of incidence of each seizure phase was calculated for
the different doses of the drugs administered. Fifty percent
efficient doses (ED50s) with 95% confidence limits (95% CL) of the
nipecotic tyrosine ester (6) were estimated, concerning tonus and
clonus phases, by the method of Litchfield and Wilcoxon.
5
64 / Journal of Pharmaceutical Sciences
Vol. 88, No. 5, May 1999

Figure 1sTime courses of disappearance of nipecotic esters (3−6) in pH 7.4 buffer solution obtained by plotting time (h) versus drug concentration (µM). Each
point represents the mean ± SD of three experiments.
the transport of the nipecotic tyrosine ester (6) across the
BBB; as the drug appears to achieve pharmacologically
active concentrations in the brain, it could be supposed to
be endowed with a high affinity for the amino acid carriers.
Unfortunately, the almost absolute insolubility of nipecotic
tyrosine ester (6) in water made it impossible to calculate
its water/oil partition coefficient. Of course, further studies
are needed to clarify if the ability of 6 to cross the BBB is
dependent also on its lipophilicity. For example, Stark et
inhibition, induced by various esters of nipecotic acid, are
mostly due to free nipecotic acid generated after hydrolysis
in the brain.2
7,28
Similar results have also been reported for esters of
isoguvacine, a GABA uptake inhibitor, structurally related
to nipecotic acid.29 Finally we can exclude that the observed
anticonvulsant effect of 6 is due to the free promoiety
(tyrosine) generated in the brain; in fact, tyrosine, systemi-
cally injected under the same experimental conditions at
a dose level of 0.75 mmol/kg, elicited no protective effect
against audiogenic seizures (data not shown).
2
5
al. have demonstrated that des-tyrosine1-D-phenylala-
nine3-beta-casomorphin is able to cross the BBB by para-
cellular transport without using a carrier system. One
could to point out that the nipecotic tyrosine ester (6) seems
more potent than other lipophilic nipecotic acid esters, such
Concerning prodrugs 3-5, several possible explanations
can be offered for their lack of anticonvulsant activity.
First, â-D-glucose and galactose enter the brain through
1
2
11
30
as (+)ethyl nipecotate or pivaloyloxymethyl ester; how-
the transporter GLUT-1. A high transporter affinity is
ever, its ED50s calculated for tonus (ED50, 0.130 mmol/kg;
required, so that a drug may be efficiently transported by
the carrier across the BBB. Prodrugs 3-5 might have
insufficient affinity for this carrier to achieve significant
brain uptake and/or their transport through the BBB might
be limited by competition with the endogenous substrate.
In addition, the hydrophilic character of esters 3-5 would
preclude their passive transport through the BBB.
9
5% CL, 0.115-0.146 mmol/kg) and clonus (ED50, 0.173
mmol/kg, 95% CL, 0.162-0.186 mmol/kg) are very higher
than those of tiagabine (a GABA uptake inhibitor, struc-
turally related to nipecotic acid), as measured in the same
experimental models26 (tonus: ED50, 1 µmol/kg; clonus:
ED50, 5 µmol/kg).
Furthermore, the present findings do not allow us to
establish exactly whether the observed anticonvulsant
activity of 6 could be due to nipecotic acid derived from its
enzymatic hydrolysis or, partially at least, to the intact
compound. However, several papers have provided infor-
mation that the anticonvulsant effect and GABA uptake
Second, carbohydrate receptor-mediated targeting of
drugs has been proposed as a potential method for site-
specific delivery of drugs to cells possessing carbohydrate
receptors on their surface. Several types of galactosylated
drugs are reported to be interesting candidates for suc-
cessful delivery to liver cells.31 So, nipecotic derivatives
Journal of Pharmaceutical Sciences / 565
Vol. 88, No. 5, May 1999

Figure 2sTime courses of disappearance of nipecotic esters (3−6) in the presence of porcine esterase obtained by plotting time (h) versus drug concentration
µM). Each point represents the mean ± SD of three experiments.
(
Table 2sEffects of Intraperitoneal Injection of 1, 3, 4, 5, and 6 on
Audiogenic Seizures in DBA/2 Mice
after systemic injection. Therefore, these derivatives are
hindered in the achievement of pharmacologically active
concentrations in the brain.
a
%
response
Finally, a correlation between in vitro rates of enzymatic
hydrolysis and its pharmacologic activity is generally
expected for CNS-active prodrugs.29 In fact, the rate of
hydrolysis in plasma would have to be sufficiently slow to
allow the prodrug to gain access to the brain, but not so
slow as to remain inactive within the brain. Compounds
drug
dose, mmol/kg W.R.b clonus tonus R.A.b SRb nb
controls Ac
vehicle (0)
0.750
0.750
0.750
0.750
vehicle (0)
0.125
0.170
0.210
0.250
100
100
100
100
100
100
75
100
100
100
100
100
100
75
80
60
3.40 10
1
3
4
5
88.8 55.5 3.44
88.8 55.5 3.44
88.8 55.5 3.44
88.8 55.5 3.44
9
9
9
3
-5 appeared much more resistant to enzyme-catalyzed
9
8
8
10
10
d
hydrolysis (t1/2 ) 8.30, 10.70, and 14.0 h, respectively, for
3, 4, and 5) than the prodrug 6 (t1/2 ) 2.5 h), but unlike 6
do not give protection against audiogenic seizures 30 min
after administration. Moreover, no anticonvulsant effect
was observed when audiogenic seizures were induced 1, 2,
or 3 h after injection of 3-5 (data not shown). Further
investigations would clarify if a significant pharmacological
effect may be detectable at longer times after administra-
tion of the very slowly hydrolyzed esters 3-5.
controls B
87.5 50
3.37
2.25
6
6
6
6
6
6
50
40
30
25
10
0
e
60
40
50
1.6e
40
1.1
d
d
d
e
20
20
20
0
0.60 10
0.30e 10
0.40e 10
d
d
d
d
0.500
0.750
20
10
0
0
d
d
d
d
20
20
0
0
a
The incidence of each seizure phase is expressed as the percentage of
mice in each group displaying that phase. b W.R. ) wild running; R.A. )
respiratory arrest; SR ) seizure response expressed as arithmetic mean of
In conclusion, the present findings indicate that nipecotic
tyrosine ester 6 is an excellent prodrug for potential
application in the treatment of epilepsy and of other
disorders resulting from abnormalities in the GABA sys-
tem. Further studies are in progress to investigate the
bioavailability of this compound and to clarify its mecha-
nism of action.
c
the maximum individual response in each group; n ) number of mice. Mice
d
given buffer, pH 7.4, 10 mL/kg, ip. Mice given 20% EtOH/H
2
O solution, 10
e
mL/kg, ip. p < 0.05 from controls B, as assessed by Fisher’s exact probability
f
test. p < 0.05 from controls B, as assessed by Mann−Whitney U test.
3
-5, obtained by the conjugation with glucose and galac-
tose, could be rapidly taken up and degraded by the liver
5
66 / Journal of Pharmaceutical Sciences
Vol. 88, No. 5, May 1999

1
6. Chapman, A. G.; Croucher, M. J .; Meldrum, B. S. Evaluation
of anticonvulsant drugs in DBA/2 mice with sound-induced
seizures. Arzneim.-Forsch./ Drug Res. 1984, 34, 1261-1264.
References and Notes
1
2
3
4
5
6
. Greig, N. H. Drug entry into the brain and its pharmacologi-
cal manipulation. In: Physiology and Pharmacology of the
Blood-Brain Barrier. Bradbury, M. W. B., Ed.; Springer-
Verlag: New York, 1992; pp 487-524.
17. Seyfried, T. N.; Glaser, G. H. A review of mouse mutants as
genetic models of epilepsy. Epilepsia 1985, 26, 143-150.
1
8. Bonina, F. P.; Montenegro, L.; Guerrera, F. Naproxen 1-alkyl-
azacicloalkan-2-one esters as dermal prodrugs: in vitro
evaluation. Int. J . Pharm. 1993, 100, 99-105.
. Robinson, P. J .; Rapoport, S. I. Transport of drugs. In:
Physiology and Pharmacology of the Blood-Brain Barrier.
Bradbury, M. W. B., Ed.; Springer-Verlag: New York, 1992;
pp 279-300.
1
2
9. Fischer, R. S. Animal models of the epilepsies. Brain Res.
Rev. 1989, 14, 245-278.
. Greenwood, J . Experimental manipulation of the blood-brain
and blood-retinal barriers. In: Physiology and Pharmacology
of the Blood-Brain Barrier. Bradbury, M. W. B., Ed.; Springer-
Verlag: New York, 1992; pp 459-486.
0. Horton, R. W.; Prestwich, S. A.; Meldrum, B. S. Gamma-
aminobutyric acid and benzodiazepine binding sites in au-
diogenic seizure-susceptible mice. J . Neurochem. 1982, 39,
8
64-870.
. Walker, I.; Nicholls, D.; Irwin, W. J .; Freeman, S. Drug
delivery via active transport at the blood-brain barrier:
affinity of a prodrug of phosphonoformate for the large amino
acid transporter. Int. J . Pharm. 1994, 104, 157-167.
. Namane, A.; Gouyette, C.; Fillion, M.; Fillion, G.; Huynh-
Dinh, T. Improved brain delivery of AZT using a glycosyl
phosphotriester prodrug, J . Med. Chem. 1992, 35, 3039-
2
1. De Sarro, G. B.; Chimirri, A.; De Sarro, A.; Gitto, R.; Grasso,
S.; Giusti, P.; Chapman. A. G. GYKI 52466 and related 2,3-
benzodiazepines as anticonvulsant agents in DBA/2 mice.
Eur. J . Pharmacol. 1995, 294, 411-422.
22. Suzdak, P. D.; J ansen, J . A. A review of the preclinical
pharmacology of tiagabine: a potent and selective anticon-
vulsant GABA uptake inhibitor. Epilepsia 1995, 36, 612-
626.
23. Smith, Q. R.; Momma, S.; Aoyagi, M.; Rapoport, S. I. Kinetics
of neutral amino acid transport across the blood-brain
barrier. J . Neurochem. 1987, 49, 1651-1658.
3
044.
. Mizuma, T.; Ohta, K.; Hayashi, M.; Awazu, S. Intestinal
active absorption of sugar-conjugated compounds by glucose
transport system: implication of improvement of poorly
adsorbable drugs, Biochem. Pharmacol. 1992, 43, 2037-
2
039.
24. Reichel, A.; Begley, D. C.; Ermisch, A. Arginine vasopressin
reduces the blood-brain transfer of L-tyrosine and L-valine:
further evidence of the effect of the peptide on the L-system
transporter at the blood-brain barrier. Brain Res. 1996, 713,
232-239.
7
8
9
. Mizuma, T.; Ohta, K.; Hayashi, M.; Awazu, S. Comparative
study of active absorption by the intestine and disposition
of anomers of sugar-conjugated compounds. Biochem. Phar-
macol. 1993, 45, 1520-1523.
. Mizuma, T.; Ohta, K.; Awazu, S. The beta-anomeric and
glucose preferences of glucose transport carrier for intestinal
active absorption of monosaccharide conjugates. Biochim.
Biophys. Acta 1994, 1200, 117-122.
2
5. Stark, H.; Van Bree, J . B.; de Boer, A. G.; J aehde, U.;
Breimer, D. D. In vitro penetration of des-tyrosine 1-D-
phenylalanina-3-beta-casomorphin across the blood-brain
barrier. Peptides 1992, 13, P47-51.
6. Dalby, N. O.; Nielsen, E. B. Comparison of the preclinical
anticonvulsant profiles of tiagabine, lamotrigine, gabapentin
and vigabatrin. Epilepsy Res. 1997, 28, 63-72.
7. Crider, A. M.; Tita, T. T.; Wood, J . D.; Hinko, C. N. Esters of
nipecotic and isonipecotic acids as potential anticonvulsants.
J . Pharm. Sci. 1982, 71, 1214-1219.
8. Nassereddine-Sebaei, M.; Crider, A. M.; Carrol, R. T.; Hinko,
C. N. Determination of m-nitrophenol and nipecotic acid in
mouse tissues by high-performance liquid chromatography
after administration of the anticonvulsivant m-nitrophenyl-
. Bonina, F. P.; Lanza, M.; Montenegro, L.; Salerno, L.;
Smeriglio, P.; Trombetta, D.; Saija, A. Transport of alpha-
tocopherol and its derivatives through erytrocyte membranes.
Pharm. Res. 1996, 13, 1342-1346.
2
2
2
1
1
1
1
0. Krogsgaard-Larsen, P.; J ohnston, G. A. R. Inhibition of
GABA uptake in rat brain slices by nipecotic acid, various
isoxazoles and related compounds. J . Neurochem. 1975, 25,
7
97-802.
1. Meldrum, B. S.; Croucher, M. J .; Krogsgaard-Larsen, P.
GABA-uptake inhibitors as anticonvulsant agents. In: Prob-
lems in GABA Research; Okada, Y.; Roberts, E., Eds.;
Exerpta Medica: Amsterdam, 1982; pp 182-191.
3
8
-piperidinecarboxylate hydro-chloride. J . Pharm. Sci. 1993,
2, 39-43.
2. Horton, R. W.; Collins, J . F.; Anlezark, G. M.; Meldrum, B.
S. Convulsant and anticonvulsant actions in DBA/2 mice of
compounds blocking the reuptake of GABA. Eur. J . Phar-
macol. 1979, 59, 75-83.
29. Falch, E.; Krogsgaard-Larsen, P. Esters of isoguvacine as
potential prodrugs. J . Med. Chem. 1981, 24, 285-289.
30. Fuglsang, A.; Lomholt, M.; Gjedde, A. Blood-brain transfer
of glucose and glucose analogues in newborn rats. J . Neu-
rochem. 1986, 46, 1417-1428.
3. Hinko, C. N.; Seibert, K.; Crider, A. M. Anticonvulsant
activity of the nipecotic acid ester, (()-m-nitrophenyl-3-
piperidinecarboxylate, Neuropharmacology 1984, 23, 1009-
3
1. Takakura, Y.; Hashida, M. Macromolecular carrier systems
for targeted drug delivery: pharmacokinetics considerations
on biodistribution. Pharm. Res. 1996, 13, 820-831.
1
014.
1
1
4. Hinko, C. N.; Crider, A. M.; Wood, J . D. A comparison of
prodrug esters of nipecotic acid. Neuropharmacology 1988,
2
7(5), 475-483.
5. Dhar, T. G.; Borden, L. A.; Tyagarajan, S.; Smith, K. E.;
Branchek, T. A.; Weinshank, R. L.; Gluchowski, C. Design,
synthesis and evaluation of substituted triaryl nipecptic acid
derivatives as GABA uptake inhibitors: identification of a
ligand with moderate affinity and selectivity for the cloned
human Gaba transporter GAT.3. J . Med. Chem. 1994, 37-
Acknowledgments
This paper was partially supported by the Italian M.U.R.S.T.
(
40%).
(15), 2334-2342.
J S980302N
Journal of Pharmaceutical Sciences / 567
Vol. 88, No. 5, May 1999