Chemical stability, enzymatic hydrolysis, and nasal uptake of amino acid ester prodrugs of acyclovir

Article abstract of DOI:10.1002/1520-6017(200105)90:5<617::AID-JPS1018>3.0.CO;2-5

The objective of this work was to improve nasal absorption of relatively impermeable small drug molecules via an amino acid prodrug approach. Acyclovir was selected as a model drug. L-Aspartate β-ester, L-lysyl, and L-phenylalanyl esters of acyclovir were synthesized to investigate their effectiveness in enhancing nasal absorption of acyclovir. A stability study was conducted in phosphate buffer under various pH conditions at 25 and 37°C. Enzymatic hydrolysis in rat nasal washings and plasma was conducted at 37°C. A rat in situ nasal perfusion technique was utilized in this investigation to examine the rate and extent of nasal absorption of amino acid prodrugs. The remaining analyte concentrations in the nasal perfusate were quantitated by reversed-phase high-performance liquid chromatography. The results revealed that the L-lysyl and L-phenylalanyl esters were less stable than L-aspartate β-ester. The stability of all three esters decreased with increasing pH and temperature. L-phenylalanyl ester is highly susceptible to plasma esterases, with an in vitro half-life 1.33 min. The rat in situ nasal perfusion study revealed that the extent of nasal absorption of acyclovir, L-lysyl and L-phenylalanyl esters was not significant (p < 1%). L-Aspartate β-ester was absorbed to the extent of ～8% over 90 min of perfusion at an initial drug concentration of 100 μM. Nasal absorption of L-aspartate β-ester of acyclovir was inhibited by L-asparagine but not by a dipeptide glycylsarcosine (Gly-Sar). The enhancement of acyclovir nasal absorption from the L-aspartate β-ester prodrug suggests that nasal uptake of this prodrug probably involves an active transport system.

Full text of DOI:10.1002/1520-6017(200105)90:5<617::AID-JPS1018>3.0.CO;2-5

Chemical Stability, Enzymatic Hydrolysis, and Nasal Uptake
of Amino Acid Ester Prodrugs of Acyclovir
CHUN YANG, HONGWU GAO, ASHIM K. MITRA
Division of Pharmaceutical Sciences, School of Pharmacy, University of Missouri, 5005 Rockhill Road, Kansas City,
Missouri 64110
Received 22 June 2000; revised 17 October 2000; accepted 25 October 2000
ABSTRACT: The objective of this work was to improve nasal absorption of relatively
impermeable small drug molecules via an amino acid prodrug approach. Acyclovir was
selected as a model drug. L-Aspartate b-ester, L-lysyl, and L-phenylalanyl esters of
acyclovir were synthesized to investigate their effectiveness in enhancing nasal absorp-
tion of acyclovir. A stability study was conducted in phosphate buffer under various pH
conditions at 25 and 37ꢀC. Enzymatic hydrolysis in rat nasal washings and plasma was
conducted at 37ꢀC. A rat in situ nasal perfusion technique was utilized in this
investigation to examine the rate and extent of nasal absorption of amino acid prodrugs.
The remaining analyte concentrations in the nasal perfusate were quantitated by
reversed-phase high-performance liquid chromatography. The results revealed that the
L-lysyl and L-phenylalanyl esters were less stable than L-aspartate b-ester. The stability
of all three esters decreased with increasing pH and temperature. L-phenylalanyl ester
is highly susceptible to plasma esterases, with an in vitro half-life 1.33 min. The rat in
situ nasal perfusion study revealed that the extent of nasal absorption of acyclovir, L-
lysyl and L-phenylalanyl esters was not signi®cant ( p<1%). L-Aspartate b-ester was
absorbed to the extent of ꢁ8% over 90 min of perfusion at an initial drug concentration of
100 mM. Nasal absorption of L-aspartate b-ester of acyclovir was inhibited by L-
asparagine but not by a dipeptide glycylsarcosine (Gly-Sar). The enhancement of
acyclovir nasal absorption from the L-aspartate b-ester prodrug suggests that nasal
uptake of this prodrug probably involves an active transport system. ß 2001 Wiley-Liss,
Inc. and the American Pharmaceutical Association J Pharm Sci 90:617±624, 2001
Keywords: nasal drug delivery; acyclovir; prodrug; active transport; amino acid
transporter; peptide transporter
INTRODUCTION
for examples, low permeability across the mucosa,
degradation of drug by enzymes in the nasal
cavity, and drug loss by rapid mucociliary clear-
ance. To improve systemic bioavailability through
nasal administration, two strategies are commonly
employed; these are, structural modi®cation
and formulation manipulation. Prodrug strategy
usually involves transient modi®cation by con-
necting particular functional group(s) of a drug
with a promoiety to obtain favorable physioche-
mical and biological properties for absorption (i.e.,
attaching biocleavable lipophilic moieties, thereby
thermodynamically favoring solute partition-
ing into the nasal membrane2). Previous studies
reported from our laboratory demonstrated that
Systemic drug delivery by the nasal route is
currently receiving considerable attention be-
cause this route offers many advantages, such as
(a) rapid absorption and onset of pharmacological
effect; (b) avoidance of liver ®rst- pass metabolism
and high systemic availability; and (c) an easy
administration route particularly suitable for
self-medication.1 There are, however, limitations;
Correspondence to: A.K. Mitra (Telephone: 816-235-1615;
Fax: 816-235-5190; E-mail: mitraa@umkc.edu)
Journal of Pharmaceutical Sciences, Vol. 90, 617±624 (2001)
ß 2001 Wiley-Liss, Inc. and the American Pharmaceutical Association
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 5, MAY 2001
617

618
YANG, GAO, AND MITRA
aliphatic ester prodrugs of acyclovir resulted in
improved nasal drug absorption.3 In this study,
an attempt has been made to demonstrate the
feasibility of utilizing amino acids as a promoiety
such that the conjugate may translocate a model
drug across a mucosal membrane by a carrier-
mediated transport process. The hypothesis
behind this approach is that the amino acid or
oligopeptide transport system in the nasal mucosa
will recognize the amino acid-conjugated drug.
Once the conjugate is absorbed into the systemic
circulation, it will revert enzymatically and/or
chemically to the parent drug.
biochem (San Diego, CA). Perchloric acid (70%)
was purchased from Allied Chemical (Morris-
town, NJ). Acyclovir was a gift from Burroughs
Wellcome, Inc. (Research Triangle Park, NC). L-
Asparagine and glycylsarcosine (Gly-Sar) were
obtained from Sigma Chemical Company (St.
Louis, MO). All other chemicals were of special
reagent grade and were received from Aldrich
Chemical Company, Inc. (Milwaukee, WI). Rever-
sed-phase thin layer chromatographic (TLC)
plates (MKC18F) were obtained from Whatman
International Ltd. (Maidstone, U.K.).
Acyclovir, an acyclic analogue of the natural
nucleoside 2'-deoxyguanosine, is clinically used in
the treatment of herpes simplex, varicella zoster,
cytomegalovirus, and Epstein±Barr virus infec-
tions.4 Absorption of orally administered acyclovir
is slow, variable, and incomplete, with a bioavail-
ability of ꢀ15±30%. It is impermeable across the
nasal mucosa.5 An in vitro study using porcine
buccal tissue indicated that buccal transport of
acyclovir occurs predominantly by a passive
diffusion mechanism, probably through the para-
cellular route.6 Therefore, this compound may
serve as a good model drug to study nasal absorp-
tion enhancement via transporters. Previous re-
ports from our laboratory have demonstrated that
L-phenylalanine is absorbed primarily through an
amino acid transporter expressed on the rat nasal
mucosa.7, 8 An electrogenic amino acid transpor-
ter is expressed on the apical membrane of
cultured human bronchial epithelial cells and
nasal polyp epithelial cells. Apical L-lysine was
effective in increasing the short-circuit current.9
In this study, L-aspartate b-ester, L-lysyl, and
L-phenylalanyl esters of acyclovir were synthe-
sized as potential prodrugs. A rat in situ nasal
perfusion technique was utilized to examine the
nasal uptake of these esters, and their chemical
stability and enzymatic hydrolysis were evalu-
ated. Possible mechanisms involved in the nasal
absorption of these prodrugs are discussed.
Synthesis of Amino Acid Ester Prodrugs
Amino acid ester prodrugs were synthesized by
following the procedure described by Beauchamp
et al.10 (Scheme I). Column chromatography was
carried out using Grade 62 (60±200 mesh) silica
gel, and elution was performed by methanol±
methylene chloride (1:9) solvent system. Melting
points were determined on a Thomas Hoover
capillary melting point apparatus and were
uncorrected. Proton nuclear magnetic resonance
(1H NMR) and 13C NMR spectra were measured
at 250 or 62 MHz (Bruker AC250). Electrospray
ionization (ESI) mass spectra were obtained on a
Finnigan TSQ-700 triple-quadrupole mass spec-
trometer equipped with an electrospray source.
Acyclovir (2.5 mmol) was dissolved in anhy-
drous dimethylformamide (DMF; 75 mL) by
warming on a steam bath. Successively, 4-
(dimethylamino) pyridine (DMAP; 0.3 mmol),
protected amino acid (4 mmol), and dicyclohex-
ylcarbodiimide (DCC; 4 mmol) were added to the
MATERIALS AND METHODS
Materials
High-performance liquid chromatography (HPLC)
grade acetonitrile (ACN) and NaH2PO4 were
obtained from Fisher Scienti®c Company (St.
Louis, MO). N-Carbobenzoxy (Cbz)-L-aspartic
acid a-benzyl ester, N-Cbz-L-phenylalanine, and
N-a-N-e-di-Cbz-L-lysine were obtained from Nova-
Scheme 1. Synthesis of amino acid ester prodrugs of
acyclovir.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 5, MAY 2001

ENHANCED NASAL ABSORPTION OF ACYCLOVIR FROM AMINO ACID ESTER PRODRUG
619
cooled solution. The solution was stirred under a
nitrogen atmosphere at ambient temperature for
24 h. The mixture was recharged with additional
DMAP (0.3 mmol), amino acid (4 mmol), and DCC
(4 mmol), and stirring was continued at ambient
temperature for 2 days. The mixture was ®ltered,
DMF was removed from the ®ltrate in vacuo, and
the residue was chromatographed on silica gel
using 1:9 MeOH±CH2Cl2 as the eluent, to
generate intermediate products (IIIa, IIIb, IIIc).
The1H NMR and mass spectrometry (MS) were
satisfactory for the desired structure. For depro-
tection, a solution of III (2 mmol) in MeOH (30
mL), tetrahydrofuran (THF; 15 mL), and water (5
mL) was added to 0.5 N aqueous HCl (4.5 mL) and
60 mg of 10% palladium on charcoal. The mixture
was shaken in a Parr apparatus under an initial
pressure of 50 psi of hydrogen at ambient
temperature for 4 h or until no reactant spot
was visible on TLC. The mixture was ®ltered, the
catalyst was washed with MeOH, and the com-
bined washings and ®ltrate were lyophilized. The
residue of IVb and IVc was recrystallized follow-
ing the method reported by Beauchamp et al.10
The residue of IVa was dissolved in a small
volume of water, and the pH of the solution was
adjusted to ꢀ5.0 with NaOH (1 N). Then the pre-
cipitate was recollected and washed with a small
volume of water. After repeating the same proce-
dure once more, the precipitate was dried in vacuo
to yield the desired amino acid ester product.
Purity of ®nal products was monitored by
reversed-phase TLC and HPLC. All of prodrugs
were at least 99% pure. Starting material (acy-
clovir) content was<0.5%.
(aC), 64.45 (CH2OCO), 66.22 (CH2O), 71.96
(NCH2O), 115.99 (C-5), 127.30 (aromatic), 128.57
(aromatic), 129.50 (aromatic), 134.60 (aromatic),
137.64 (C-8), 151.33 (C-4), 154.25 (C-2); 156.61 (C-
‡
6), 169.00 (CˆO); MS (ESI): 373 [M ‡ 1] (100).
IVc: foamy solid (lit. 142±154ꢁC); yield 72%; 1H
‡
NMR (DMSO-d6): d1.42 (m, 2H, COCH(NH3 )
‡
CH2 CH2CH2CH2NH3 ), 1.60 (m, 2H, OCH
‡
‡
(NH3 )CH2CH2CH2CH2 NH3 ), 1.79 (dm, 2H,
‡
‡
COCH(NH3 )CH2CH2CH2CH2NH3 ), 2.77 (bs, 2H,
COCH(NH3 )CH2CH2CH2 CH2NH3 ), 3.84 (bs,
2H, CH2O), 3.95 (bs, 1H, COCH(NH3 )CH2CH2
CH2CH2NH3 ), 4.30 (s, 2H, OCH2), 5.55 (s, 2H,
‡
‡
‡
‡
‡
NCH2O), 7.39 (bs, 2H, 2-NH2), 8.23 (s, 3H, NH3 ),
‡
8.76 (s, 3H, NH3 ), 9.03 (s, 1H, H-8), 11.74 (bs, 1H,
NH); 13C NMR (DMSO-d6): d 21.08, 26.05, 29.15
‡
‡
(COCH(NH3 ) CH2CH2CH2CH2NH3 ), 38.08 (CO-
‡
‡
CH(NH3 ) CH2CH2CH2CH2NH3 ), 51.47 (COCH-
‡
‡
(NH3 ) CH2CH2CH2CH2NH3 ), 64.24 (CH2OCO),
66.76 (CH2O), 72.98 (NCH2O), 110.74 (C-5),
137.80 (C-8), 150.22 (C-4), 154.41 (C-2), 155.19
‡
(C-6), 169.28 (CˆO); MS(ESI): 354 [M ‡ 1] (14),
270 (22), 186 (45), 169 (100).
Stability Study
Aqueous stability of prodrugs was examined at
pH 7.4 in isotonic phosphate buffer saline solution
at 25 and 37ꢁC, and at pH 5.5 and 3.0 in 50 mM
sodium phosphate buffer at room temperature.
Hydrolysis of the amino acid esters to acyclovir
was detected by HPLC. The reaction was followed
for 1 week. Apparent ®rst-order rate constants for
chemical hydrolysis were determined from the
slopes of semilogarithmic plots of percent prodrug
remaining versus time.
IVa: decomposition 195ꢁC; yield (two steps)
1
51.6%; H NMR (DMSO-d6 ‡ CF3CO2D, 10:1): d
2.89 (d, 2H, CH2CH(NH2)CO), 3.76 (m, 2H,
CH2O), 4.17 (m, 1H, aCH), 4.23 (m, 3H, COCH2),
5.53 (s, 2H, NCH2O), 7.23 (bs, 2-NH2), 8.41 (bs,
Analytical Procedure
The concentrations of acyclovir and its prodrugs
were quantitated by reversed-phase HPLC. Prior
to analysis, all of the biological samples were
treated with 10% volume of perchloric acid (70%
solution) to precipitate the proteins. Then, inter-
nal standard solution was added to a ®nal
concentration of 50 mM. After centrifugation at
100,000 Â g for 20 min, the supernatant was
injected onto the HPLC column (Rainin Micro-
sorb-MV C18 column, 5 mm, 250 mm).
The HPLC system was comprised of Varian
9012 Solvent Delivery System, 9100 AutoSam-
pler, and 9050 Variable Wavelength UV-VIS
Detector. The signal was monitored at 254 nm.
The ratio of integrated peak areas of drug to
‡
NH3 ), 9.27 (s, 1H, H-8), 11.67 (bs, HN), 13.73
(COOH ); 13C NMR (DMSO-d6 ‡ CF3COOD, 10:1):
d 34.23 (bC), 48.58 (aC), 63.95 (CH2OCO), 67.61
(CH2O), 74.20 (NCH2O), 108.26 (C-5), 138.66 (C-
8), 150.24 (C-4), 153.88 (C-2), 155.99 (C-6), 169.69
(bCˆO), 169.93 (terminal CˆO), MS (ESI): 341
‡
‡
[M ‡ 1] (100), 363 [M ‡ Na] (25).
IVb: mp, 210±215; yield, 52.2% 1HNMR
(DMSO-d6): d 3.09 (m, 2H, CH2CH(NH2)C), 3.61
(m, 2H, CH2O), 4.17(s, 1H, aCH), 4.24 (m, 2H,
COOCH2), 5.34 (s, 2H, NCH2O), 6.77(s, 2H, 2-
NH2), 7.20, 7.22, 7.26, and 7.29 (m, 5H, C6H5),
‡
7.90 (s, 1H, H-8), 8.67 (brs, 2H, NH3 ); 10.93 (s,
1H, NH); 13C NMR (DMSO-d6): d 35.78 (bC), 53.11
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 5, MAY 2001

620
YANG, GAO, AND MITRA
Table 1. HPLC Conditions for Analyses of Prodrugs
Retention Time (min)
Mobile Phase
Internal
Compound
and Flow Rate
Standard;
Prodrug
I.S.
ACV
ACV
3% ACN ‡ 97% Buffer 1.0 mL/min
1% ACN ‡ 99% Buffer 1.2 mL/min
3% ACN ‡ 97% Buffer 0.8 mL/min
12% ACN ‡ 88% Buffer 1.0 mL/min
Thymine
Thymine
Thymine
Caffeine
Ð
6.3
7.9
8.3
9.6
7.5
12.1
10.4
3.0
Asp±ACV
Lys±ACV
Phe±ACV
9.5
6.7
7.1
internal standard was used for quantitative pur-
poses. Mobile phase was prepared with ACN and
NaH2PO4 (20 mM, pH 3.0). Table 1 provides the
mobile phase composition, ¯ow rate, internal
standard used, and retention times.
4ꢁC) was mixed with nine volumes of prewarmed
nasal washing solution (37ꢁC) and vortexed. A
zero-time sample (100 mL) was taken immediately
and mixed with 10 mL of an HClO4 solution (70%)
and vortexed for 30 s to precipitate the proteins.
The mixture was incubated at 37ꢁC, and 100-mL
samples were withdrawn at predetermined time
intervals and were subjected to the same treat-
ment. Blood was collected from rat jugular vein.
After centrifugation at 10,000 rpm for 10 min,
plasma was collected and then stored at 80ꢁC
until further use. The incubation procedure with
rat plasma was similar to that with rat nasal
washings.
Method for Nasal Perfusion
The rat in situ nasal perfusion technique devel-
oped by Hirai et al.11 and Huang et al.12 was used.
Male Sprague-Dawley rats, weighing 250±350 g,
were anesthetized with an intraperitoneal injec-
tion (0.1 mL/100 g body weight) of a ketamine (90
mg/mL) and xylazine (10 mg/mL) mixture. After
an incision was made in the neck, the trachea was
cannulated with a polyethylene (PE-200) tube to
maintain respiration. Another PE-200 tube was
inserted through the esophagus toward the
posterior part of the nasal cavity and ligated.
The passage of the nasopalatine tract was sealed
with an adhesive agent to prevent drainage of the
solution from the nasal cavity to the month. The
cannula served to deliver the solution to the nasal
cavity. The perfusion medium (initial concentra-
tion 100 mM), which was prepared with isotonic
phosphate buffer saline solution, pH 7.4, was
circulated by means of a peristaltic pump at a ¯ow
rate of 2 mL/min. The perfusate was recollected
into a reservoir, which was maintained at a
temperature of 37 Æ 0.5ꢁC throughout the course
of an experiment. A constant perfusate volume of
5 mL was maintained throughout with constant
stirring and an aliquot (100 mL) was sampled at
predetermined time intervals.
RESULTS
Apparent ®rst-order rate constants for hydrolysis
and nasal absorption were determined from the
slopes of various semilogarithmic plots of percent
remaining of prodrugs in buffer solutions, nasal
washings, plasma and nasal perfusate versus
time curves. Table 2 summarizes the half-lives
of chemical hydrolyses of various prodrugs in
different buffer solutions. A longer half-life (t1/2) of
L-aspartate b-ester than that of the other two
esters indicated L-aspartate b-ester possessed a
better stability. The stability of three esters
decreased as the temperature and pH of aqueous
solution were raised.
Two processes contribute to the prodrug loss
from perfusate; they are, nasal uptake and
hydrolysis, including chemical and enzymatic
hydrolysis. The rate of prodrug loss from perfu-
sate is expressed by eq.1:
In Vitro Enzymatic Degradation Study with
Rat Nasal Washings and Plasma
dC
ˆ kappC ˆ khyd ‡ kabs
C
ꢂ1†
dt
Isotonic phosphate buffer saline solution, pH 7.4,
was perfused through the rat nasal cavity for 90
min. The perfusate at the end of the experiment
was collected and stored at 80ꢁC until further
use. One volume of drug stock solution (1.0 mM,
where C denotes drug concentration in perfusate,
kapp represents apparent ®rst-order rate constant
of drug loss from perfusate, khyd denotes apparent
®rst-order hydrolysis rate constant, and kabs
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 5, MAY 2001

ENHANCED NASAL ABSORPTION OF ACYCLOVIR FROM AMINO ACID ESTER PRODRUG
621
Table 2. Half-Lives (h) of Chemical Hydrolyses of Prodrugsa
pH 7.4
pH 3.0
25ꢁC
pH 5.0
25ꢁC
Compound
Asp±ACV
37ꢁC
25ꢁC
16.6 (0.3)
60.3 (0.7)
No measurable
degradation
during 1 week
Less than 10%
degraded during
1 week
Less than 15%
degraded during
1 week
Less than 10%
degraded during
1 week
Less than 10%
degraded during
1 week
Lys±ACV
Phe±ACV
3.9 (0.09)
4.2 (0.2)
11.4 (0.2)
Ð
136.6 (3.7)
aData are mean (SD) (n ˆ 3).
signi®es apparent ®rst-order absorption rate
constant. The apparent ®rst-order rate constants
of prodrug hydrolysis and drug loss from rat nasal
perfusate are given in Table 3. The t1/2 values of
prodrug hydrolysis in nasal washings were,
respectively, 16.96, 2.93, and 0.80 h for the Asp,
Lys, and Phe conjugates of acyclovir (Asp±ACV,
Lys±ACV, and Phe±ACV, respectively). A com-
parison of hydrolysis t1/2 in nasal washings with
hydrolysis t1/2 in isotonic phosphate buffer saline
solution indicated that enzymatic hydrolysis of L-
lysyl ester and L- phenylalanyl ester in rat nasal
washings solution was signi®cantly more rapid
than in isotonic phosphate buffer saline solution,
whereas that in L-aspartate b-ester exhibited no
signi®cant difference. Therefore, L-aspartate b-
ester appears to be less labile to the esterases in
rat nasal cavity (Table 3), an encouraging indica-
tion that may allow for enhanced nasal mucosal
transport of the intact prodrug because of its
longer possible residence time in contact with the
mucosal epithelium.
The rate constants for hydrolysis of L-lysyl and
L-phenylalanyl esters in rat nasal washing solu-
3
tion were 3.94 and 14.5 Â 10
min 1, res-
pectively (Figure 1 and Table 3). These values
were not signi®cantly different from the rate
constants of prodrug loss from nasal perfusate
3
that were 3.99 and 14.5 Â 10
min 1, respec-
tively (Figure 1 and Table 3). The absorption rate
constants of L-lysyl and L-phenylalanyl esters
were calculated according to eq. 1 and were found
5
1
to be 5.00 Â 10
min
and not measurable,
respectively. These results indicate that nasal
absorption of intact L-lysyl and L-phenylalanyl
esters was negligible. However, L-aspartate b-
ester prodrug was absorbed signi®cantly to the
extent of ꢀ8% over a 90-min perfusion at an
initial concentration of 100 mM. The absorption
rate constant of L-aspartate b-ester prodrug was
calculated to be 7.69 Â 10 4 min 1 (Figure 2 and
Table 3). Figure 3 depicts the loss of L-aspartate b-
ester from nasal perfusate and simultaneous
hydrolysis to acyclovir.
1
Table 3. Apparent First-Order Rate Constants (Â103 min
) of Enzymatic
Hydrolysis and Drug Loss from Perfusate and Percent Remaining after 90-min
Perfusiona
% Remaining
after 90-min
Perfusion
Drug
Disappearance
from Perfusate
Hydrolysis in
Nasal Washings
Plasma
Metabolism
Compound
ACV
Ð
Ð
99.06 (5.65)
86.72 (1.32)
69.34 (1.23)
29.06 (10.87)
0.142 (0.075)
1.45 (0.26)
3.99 (0.25)
14.5 (5.0)
Asp±ACV
Lys±ACV
Phe±ACV
0.681 (0.072)
3.94 (0.86)
14.5 (1.8)
8.94 (0.88)
12.2 (0.25)
519 (37)
Data are mean (SD) (n ˆ 3±5).
aExperiments were run at 37ꢁC in IPBS pH7.4.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 5, MAY 2001

622
YANG, GAO, AND MITRA
Figure 3. Loss of L-aspartate b-ester prodrug (Asp-
ACV) in nasal perfusate increased as a function of time
and simultaneous hydrolysis to acyclovir. Data are
mean Æ SD (n ˆ 3). Acyclovir generated from hydrolysis
was signi®cantly lower than Asp-ACV lost (p<0.005,
two-tailed student t test).
Figure 1. Time courses of disappearance of L-lysyl
and L-phenylalanyl esters of acyclovir from incubation
with nasal washing solution and perfusion through
nasal cavity. Data are mean Æ standard deviation
(SD; n ˆ 3±5).
Plasma hydrolysis study suggests that L-phe-
nylalanyl ester of acyclovir is highly susceptible to
degradation with a reversion t1/2 of 1.33 min,
whereas plasma bioconversion t1/2 for L-lysyl ester
and L-aspartate b-ester are 56.8 and 77.5 min,
respectively (Table 3).
To delineate the possible mechanism for nasal
uptake of L-aspartate b-ester of acyclovir, a
competitive inhibition study was conducted in
the presence of L-asparagine (2 mM) and the
dipeptide Gly-Sar (2 mM). The inhibitors were
dissolved in isotonic phosphate buffer saline
solution, pH 7.4, and the concentration of
L-aspartate b-ester of acyclovir was kept at 100
mM. Figure 4 illustrates the inhibitory effect of
L-asparagine and Gly-Sar on nasal absorption of
L-aspartate b-ester. After 90 min of perfusion, the
extent of L-aspartate b-ester loss from nasal
perfusate in the presence of L-asparagine showed
signi®cant difference from that in the absence of
L-asparagine (one-tail student t test, p<0.05).
However, there is no signi®cant difference in
prodrug loss from nasal perfusate in the pre-
sence and absence of Gly-Sar. These preliminary
Figure 4. Inhibitory effects of L-asparagine (Asn)
and Gly-Sar on nasal absorption of L-aspartate b-ester
of acyclovir. Percent remaining of prodrug in nasal
perfusate after a 90-min perfusion. Data are mean Æ SD
(n ˆ 3). L-Asparagine inhibited prodrug absorption
signi®cantly (p < 0.05, one-tailed student t test).
Figure 2. Time courses of disappearance of L-aspar-
tate b-ester of acyclovir from incubation with nasal
washing solution and perfusion through nasal cavity.
Comparison with the time course of nasal perfusion of
acyclovir. Data are mean Æ SD (n ˆ 3±5).
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 5, MAY 2001

ENHANCED NASAL ABSORPTION OF ACYCLOVIR FROM AMINO ACID ESTER PRODRUG
623
results indicate that L-asparagine inhibits nasal
uptake of the prodrug, whereas Gly-Sar exerts
no inhibitory effect on the nasal uptake of
Asp±ACV.
airway epithelial cell line, we have demonstrated
that L-aspartate b-ester of acyclovir inhibited L-
tyrosine uptake in a competitive manner.19 The
amino acid transport system expressed on nasal
epithelial membrane may require that substrate
possesses certain structural features (i.e., free
a-amino and terminal carboxyl groups). The
b-esters were not absorbable probably because of
their lack of free terminal carboxyl group. Our
previous study on nasal absorption of an L-tyro-
sine-like model compound also showed that the
L-tyrosine conjugate, which was formed through
ether linkage between hydroxyl group of parent
compound and aromatic hydroxyl group of tyro-
sine, was absorbed, whereas its parent compound
was not absorbed.20 Amino acids are absorbed in
the intestine by distinctly different mechanism
from the peptides.21 The L-valine ester prodrug of
acyclovir was taken up in the intestine by the
intestinal PepT1 peptide transporter. L-Lysyl
ester and L-phenylalanyl ester might be sub-
strates for peptide transport system. However,
the peptide transport systems may not be highly
expressed on the nasal epithelial cells. Previous
studies from our laboratory showed that nasal
absorption of the dipeptide Tyr-Gly and [-Arg2]-
kyotorphin was negligible.22
DISCUSSION
In recent years, different nutrient transporters
(e.g., peptide, amino acid, and glucose transpor-
ters) have been identi®ed and cloned. The active
nutrient transport systems have become a key
target for prodrug design. The preferential deliv-
ery of drugs or their derivatives to a target site
may be improved by using the active transport
system present at the target site. For example,
prodrug designed in the form of di/tripeptide
analogues can be absorbed across the intestinal
brush border membrane via the oligopeptide
transport system.13. Valacyclovir, the L-valyl
ester prodrug of acyclovir, has been demonstrated
to be absorbed in the intestinal tract by the PepT1
peptide transporters.14, 15 Amino acid transport
systems have also been utilized to improve the
absorption of amino acid-conjugated drugs or
17
amino acid derivatives.16,
Walker et al.17
demonstrated that the phosphonoformate±L-tyr-
osine conjugate is actively transported by means
of active amino acid transporters through mono-
layers of porcine brain microvessel endothelial
cells. Surendran et al.18 reported that PD 158473,
an NMDA antagonist and a phenylalanine deri-
vative with a phosphoric acid and naphthyl
substitutent on positions 5 and 3, respectively, is
a substrate for large neutral amino acid (LNAA)
transporter and dipeptide (hPEPT1) transporter.
In this study, two different classes of amino
acid esters, a-amino acid ester and amino acid
b-ester, were synthesized to investigate their
ability to improve nasal absorption of acyclovir.
The results suggest that L-lysyl ester and L-phe-
nylalanyl ester are not suf®ciently absorbed
through nasal pathway, although there is evi-
dence to show that L-lysine and L-phenylalanine
can be taken up by an amino acid transport
system in the nasal cavity.8,9 However, L-aspar-
tate b-ester can be absorbed signi®cantly by the
nasal route. Preliminary results from the inhibi-
tion of nasal absorption of L-aspartate b-ester by
L-asparagine but not by Gly-Sar suggest that
nasal uptake of this amino acid ester prodrug is
probably mediated by an amino acid transport
system in the rat nasal mucosa. In another report,
using an in vitro cell culture model of human
Rational design of prodrugs that can be taken
up by a speci®c transporter requires more infor-
mation on the substrate structural speci®city of
the transport system. It is currently unknown as
to which speci®c amino acid transport system is
involved in the nasal absorption of the L-aspartate
b-ester of acyclovir.
In conclusion, the L-aspartate b-ester prodrug
can be absorbed nasally and is least labile to
enzymatic hydrolysis in the nasal mucosa. This
report demonstrates that improvement in nasal
absorption of nonabsorbable acyclovir could be
achieved by attaching the drug to an amino acid
such that the amino acid transporter can recog-
nize the conjugated drug. This strategy appears to
have wide applicability in the design of potential
prodrugs of poorly permeable drugs for ef®cient
transmucosal delivery.
ACKNOWLEDGMENTS
The authors are grateful to Dr. Paul Brown,
Quintiles, Kansas City, Missouri, for his help in
obtaining mass spectra. This work was supported
by NIH Grants 2R01 EY 09171-06 and 2R01
EY10659-05.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 5, MAY 2001

624
YANG, GAO, AND MITRA
REFERENCES
14. Han HK, Devrueh RLA, Rhie JK, Covitz KMY,
Smith PL, Lee CP, Oh DM, Sadee W, Amidon GL.
1998. 5'-Amino acid ester of antiviral nucleosides,
acyclovir, and AZT are absorbed by the intestinal
PEPT1 peptide transporter. Pharm Res 15:1154±
1159.
15. Guo AL, Hu PD, Balimane PV, Leibach FH,
Sinko PJ. 1999. Interactions of a nonpeptidic
drug, valacyclovir, with the human intestinal
peptide transporter (hPEPT1) expressed in a mam-
malian cell line. J Pharmacol Exp Ther 289:448±
454.
1. Hussain AA. 1998. Intranasal drug delivery. Adv
Drug Delivery Rev 29:39±49.
2. Krishamoorthy R, Mitra AK. 1998. Prodrugs for
nasal drug delivery. Adv Drug Delivery Rev 29:
135±146.
3. Shao Z, Park GB, Krishnamoorthy R, Mitra AK.
1994. The physicochemical properties, plasma
enzymatic hydrolysis, and nasal absorption of
acyclovir and its 2'-ester prodrugs. Pharm Res
11:237±242.
16. 16. *.Bonina FP, Arenare L, Palagiano F, Saija
A, Nava F, Trometta D, de Caprariis P. 1999.
Synthesis, stability, and pharmacological evalua-
tion of nipecotic acid prodrugs. J Pharm Sci 88:
561±567.
17. Walker I, Nicholls D, Irwin WJ, Freeman S. 1994.
Drug delivery via active transport at the blood-
brain barrier: Af®nity of a prodrug of phosphono-
formate for the large amino acid transporter. Int J
Pharm 104:157±167.
18. Surendram N, Covitz, KY, Han H, Sadee W, Oh D,
Amidon GL, Williamson RM, Bigge CF, Stewart
BH. 1999. Evidence for overlapping substrate
speci®city between large neutral amino acid
(LNAA) and dipeptide (hPEPT1) transporters for
PD 158473, an NMDA antagonist. Pharm Res
16:391±395.
19. Yang C, Mitra AK. 2000. Mechanistic study on
nasal absorption of amino acid ester prodrug of
acyclovir: competitive inhibition of L-tyrosine
uptake by L-aspartic acid b-ester of acyclovir,
manuscript in preparation.
4. O'Brien JJ, Campoli-Richards DM. 1989. Acyclovir.
An updated review of its antiviral activity, phar-
macokinetic properties and therapeutic ef®cacy.
Drug 37:233±309.
5. Shao Z, Mitra, AK. 1994. Bile salt-fatty acid mixed
micelles as nasal absorption promoters: III. Effects
on nasal transport and enzymatic degradation of
acyclovir prodrugs. Int J Pharm 11:243±250.
6. Shojaei AH, Berner B, Li X. 1998. Transbuccal
Delivery of acyclovir: In vitro determination of
routesofbuccaltransport. PharmRes15:1182±1188.
7. Tengamnuay P, Mitra AK. 1988. Transport of
tyrosine and phenyalanine across the rat nasal
mucosa, Life Sci 43:585±593.
8. Tengamnuay P. 1989. Ph.D. Thesis, Purdue Uni-
versity, West Lafayette, IN.
9. Galietta LJV, Musante L, Romio L, Caruso U,
Fantasia A, Gazzolo A, Romano L, Sacco O, Rossi
GA, Varesio L, Zegarra-Moran O. 1998. An electro-
genic amino acid transporter in the apical mem-
brane of cultured human bronchial epithelial cells.
Am J Physiol 275: L917±L923.
20. Yang C, Mitra AK. 2000. Nasal absorption of
tyrosine linked model compounds. J Pharm Sci,
(in press)
21. Ganapathy V, Brandsch M, Leibach FH. 1994.
Intestinal transport of amino acids and peptides.
In: Johnson, LR, Editor. Physiology of the gastro-
intestinal tract, 3rd ed. New York: Raven Press, pp
1773±1794.
22. Tengamnuay P, Mitra AK. 1990. Bile salt-fatty
acid-mixed micelles as nasal absorption promoters
of peptides IÐEffects of counterions, adjuvant
composition of lipid structure on the nasal ab-
sorption of [-Arg2]-Kyotorphin. Pharm Res 7:129±
135.
10. Beauchamp LM, Orr GF, de Miranda P, Burnette
T, Krenitsky TA. 1992. Amino acid ester prodrugs
of acyclovir. Antiviral Chem Chemother 3:157±164.
11. Hirai S, Yashiki T, Matsuzawa T, Mima H. 1981.
Absorption of drugs from the nasal mucosa. Int J
Pharm 7:317±325.
12. Huang CH, Kimura R, Bawarshi-Nassar R, Hus-
sain A.1985. Mechanism of nasal absorption of
drugs I: Physicochemical parameters in¯uencing
the rate of in situ nasal absorption of drugs in rats.
J Pharm Sci 74:608±611.
13. Yang CY, Dantzig AH, Pidgeon C. 1999. Intestinal
peptide transport systems and oral drug availabil-
ity. Pharm Res 16:1331±1343.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 5, MAY 2001