Colon-specific prodrugs of 5-aminosalicylic acid: Synthesis and in vitro/in vivo properties of acidic amino acid derivatives of 5-aminosalicylic acid

Article abstract of DOI:10.1002/jps.1126

5-Aminosalicyl-L-aspartic acid (5-ASA-Asp) and 5-aminosalicyl-L-glutamic acid (5-ASA-Glu) were synthesized and their properties as colon-specific prodrugs of 5-aminosalicylic acid (5-ASA) were investigated employing rats as test animals. Incubation of 5-ASA-Asp and 5-ASA-Glu with the homogenates of tissue and contents of stomach or small intestine released no 5-ASA, indicating that they were stable in this condition. Incubation of 5-ASA-Asp with the cecal contents released 5-ASA 37%, whereas 5-ASA-Glu released only 8% of the dose in 16 h. Plasma concentration of 5-ASA-Asp after intravenous administration decreased rapidly and became undetectable in 60 min. No 5-ASA was detected in the blood, which indicated 5-ASA-Asp was stable in the plasma. After oral administration of 5-ASA-Asp, concentration of 5-ASA, its metabolite N-acetyl-5-ASA, and 5-ASA-Asp in the plasma, feces, and urine was determined. In the plasma, 5-ASA-Asp was not detected and the concentration of 5-ASA or N-acetyl-5-ASA was very low. About 33% of the administered dose was recovered as 5-ASA and N-acetyl-5-ASA and 43% as 5-ASA-Asp from feces, and 20% as 5-ASA and N-acetyl-5-ASA and 1% as 5-ASA-Asp from urine in 24 h. These results suggested that most of 5-ASA-Asp was delivered to the large intestine and about half of the administered dose was activated to liberate 5-ASA. After oral administration of free 5-ASA, fecal recovery was only 7% of the dose in 24 h and more than 80% was recovered from urine. Comparing 5-ASA-Asp and free 5-ASA, the amount of 5-ASA available in the large intestine was much larger, while the amount of 5-ASA in urine, which might be related to the systemic toxicity of 5-ASA, was much lower by the administration of 5-ASA-Asp than free 5-ASA.

Full text of DOI:10.1002/jps.1126

Colon-Speci®c Prodrugs of 5-Aminosalicylic Acid: Synthesis
and In Vitro/In Vivo Properties of Acidic Amino Acid
Derivatives of 5-Aminosalicylic Acid
YUN JIN JUNG, JEOUNG SOO LEE, YOUNG MI KIM
College of Pharmacy, Pusan National University, Pusan 609-735, Korea
Received 16 October 2000; revised 6 April 2001; accepted 15 June 2001
ABSTRACT: 5-Aminosalicyl-L-aspartic acid (5-ASA-Asp) and 5-aminosalicyl-L-glutamic
acid (5-ASA-Glu) were synthesized and their properties as colon-speci®c prodrugs of 5-
aminosalicylic acid (5-ASA) were investigated employing rats as test animals.
Incubation of 5-ASA-Asp and 5-ASA-Glu with the homogenates of tissue and contents
of stomach or small intestine released no 5-ASA, indicating that they were stable in this
condition. Incubation of 5-ASA-Asp with the cecal contents released 5-ASA 37%,
whereas 5-ASA-Glu released only 8% of the dose in 16 h. Plasma concentration of 5-ASA-
Asp after intravenous administration decreased rapidly and became undetectable in
60 min. No 5-ASA was detected in the blood, which indicated 5-ASA-Asp was stable in
the plasma. After oral administration of 5-ASA-Asp, concentration of 5-ASA, its meta-
bolite N-acetyl-5-ASA, and 5-ASA-Asp in the plasma, feces, and urine was determined.
In the plasma, 5-ASA-Asp was not detected and the concentration of 5-ASA or N-acetyl-
5-ASA was very low. About 33% of the administered dose was recovered as 5-ASA and N-
acetyl-5-ASA and 43% as 5-ASA-Asp from feces, and 20% as 5-ASA and N-acetyl-5-ASA
and 1% as 5-ASA-Asp from urine in 24 h. These results suggested that most of 5-ASA-
Asp was delivered to the large intestine and about half of the administered dose was
activated to liberate 5-ASA. After oral administration of free 5-ASA, fecal recovery was
only 7% of the dose in 24 h and more than 80% was recovered from urine. Comparing
5-ASA-Asp and free 5-ASA, the amount of 5-ASA available in the large intestine
was much larger, while the amount of 5-ASA in urine, which might be related to the
systemic toxicity of 5-ASA, was much lower by the administration of 5-ASA-Asp than
free 5-ASA. ß 2001 Wiley-Liss, Inc. and the American Pharmaceutical Association J Pharm Sci
90:1767±1775, 2001
Keywords: 5-aminosalicyl-L-aspartic acid; 5-aminosalicyl-L-glutamic acid; colon-
speci®c prodrug of 5-aminosalicylic acid; in¯ammatory bowel disease
INTRODUCTION
to avoid systemic absorption and reduce side
effects. Secondly, for those drugs for which ab-
sorption through the large intestine is bene®cial.
For orally administered therapeutic peptides and
proteins, the colon could be the preferential site of
absorption considering the low level of endogen-
ous enzymes and the long transit time. Thirdly,
delayed absorption by colonic targeting may also
be useful for the chronotherapy of diseases such
as asthma, gastric ulcer, or arthritis, which may
have peak symptoms at bedtime. In these cases,
the time-plasma level of the active drugs might
Colon-speci®c delivery of orally administered
drugs can be of advantage for the following
cases.1±6 Firstly, for the ef®cient treatment of
diseases that develop at the colonic site, in order
Correspondence to: Y.M. Kim (Telephone: 051-510-2807;
Fax: 051-513-6754; E-mail: ymikim@hyowon.pusan.ac.kr)
Journal of Pharmaceutical Sciences, Vol. 90, 1767±1775 (2001)
ß 2001 Wiley-Liss, Inc. and the American Pharmaceutical Association
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001
1767

1768
JUNG, LEE, AND KIM
coincide better with the symptoms considering
the transit time through the colon.
carbodiimide (DCC), and 10% Pd/C were pur-
chased from Sigma Chemical Co. (St. Louis, MO)
and used as received. Solvents for NMR and
HPLC were obtained from Merck, Inc. (Damstadt,
Germany). All other chemicals were reagent
grade, commercially available products. IR spec-
tra were recorded on a Bomem MB 100 FT-IR
spectrophotometer (Bomem). 1H-NMR spectra
were taken on a Brucker AC-200 spectrometer
(Brucker), and the chemical shifts were in ppm
down®eld from tetramethylsilane. Melting points
were taken on a Mel Tem II (Laboratory Devices,
Holliston, MA) and were uncorrected. A Parr 4562
pressure reactor (Parr Instrument Company,
Moline, IL) was used for catalytic hydrogenation.
An Eyela Mazela-Z tissue homogenizer (Tokyo
Rikakikai Co., LTD., Tokyo, Japan) was used for
homogenization of the gastrointestinal tracts of
rats. An Eppendorf Centrifuge 5415C (Eppendorf,
Hamburg, Germany) was used for centrifugation.
TLCs were performed on Merck Kieselgel 60 F254
and RP-8 F254s. Open column chromatography
and low-pressure chromatography were perform-
ed on Merck silica gel (70&230 and 230&400
5-aminosalicylic acid (5-ASA) is an active in-
gredient of agents used for the long-term main-
tenance therapy to prevent relapses of Crohn's
disease and ulcerative colitis. 5-ASA is absorbed
rapidly and extensively in the upper intestine and
hardly reaches to the colon.7 Systemically absorb-
ed 5-ASA is reported to induce nephrotic syn-
drome.8 Several prodrugs have been developed
aiming at the delivery of 5-ASA to the colon.9±14
A colon-speci®c prodrug should be chemically
and biochemically stable and nonabsorbable in
the upper intestine so that it can be delivered to
the colon in intact form, and the linkage between
drug and promoeity should be dissociated to
liberate the active drug in the colon.
Amide bonds of various N-aromatic acyl-amino
acid conjugates are stable in the upper intestine
and hydrolyzed when they are incubated with the
cecal contents of mammals.15±17 If a hydrophilic
amino acid conjugate of 5-ASA is administered
orally, transcellular absorption by way of lipid
membrane permeation might be limited in the
upper intestine, and a large fraction of the dose
might be delivered to the colon in intact form.
Once delivered to the colon, 5-ASA might be
released from N-5-aminosalicyl-amino acid by the
microbial degradation in the large intestine,
which should be available for the local action.
Our previous study showed that about 50% of
the orally administered dose of N-5-aminosalicyl-
glycine (5-ASA-Gly) was recovered as 5-ASA (N-
acetyl-ASA) from feces and about 28% as 5-ASA
(N-acetyl-ASA) and 14% as 5-ASA-Gly from
urine. We expect that an acidic amino acid con-
jugate of 5-ASA, such as 5-aminosalicyl-L-aspar-
tic acid (5-ASA-Asp) or 5-aminosalicyl-L-glutamic
acid (5-ASA-Glu), might be more hydrophilic than
5-ASA-Gly, and the transcellular absorption by
way of lipid membrane permeation in the upper
intestine might be decreased, accordingly. In this
report, synthetic methods and in vitro/in vivo
properties of 5-ASA-Asp and 5-ASA-Glu as colon-
speci®c prodrugs of 5-ASA are presented.
mesh) and Merck Lichroprep RP-8 size
B
(230&400 mesh) columns, respectively. The
HPLC system consisted of Model 305 and 306
pumps, a 117 variable UV detector, a Model 234
autoinjector, a Model 805 manometric module,
and a Model 811C dynamic mixer from Gilson
(Middleton, WI).
Calibration of the Compound in
Various Biological Specimens
The tissue and contents of the stomach and small
intestine and the contents of cecum or colon of a
male Sprague-Dawley rat were obtained and
homogenized separately. Each sample was dilut-
ed with pH 6.8 isotonic phosphate buffer to the
concentration of 10 w/v%. To each 100 mL portion
of the homogenate was added 10, 50, 100, or
200 mL of the solution of 5-ASA, N-acetyl-5-ASA,
5-ASA-Asp, or 5-ASA-Glu (50 mg equivalent of
5-ASA/mL) in pH 6.8 isotonic phosphate buffer,
and an appropriate amount of methanol was
added to make the ®nal volume of 1 mL. The
standard solutions of 5-ASA, N-acetyl-5-ASA,
5-ASA-Asp, or 5-ASA-Glu in concentrations of
0.5, 2.5, 5.0, or 10.0 ppm in various biological
specimens were obtained by the preceding pro-
cesses. A calibration curve for 5-ASA, N-acetyl-5-
ASA, 5-ASA-Asp, or 5-ASA-Glu was constructed
from the concentration of the standard solution
MATERIALS AND METHODS
Materials
5-Nitrosalicylic acid (5-NSA), L-aspartic acid
dimethyl ester hydrochloride, L-glutamic acid
dimethyl ester hydrochloride, N,N0-dicyclohexyl-
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001

COLON-SPECIFIC PRODRUGS OF 5-ASA
1769
versus the peak area on HPLC, which was
determined by the following procedure. Standard
or blank solution (1 mL) was mixed in a vortex
mixer for 2 min, centrifuged at 10,000 Â g for
5 min, and ®ltered through a membrane ®lter
(0.45 mm). The ®ltrate (20 mL) was injected on a
Synchropac ODS column (250 Â 4.6 mm, 5 mm)
and eluted with the mobile phase at a ¯ow rate of
1.5 mL/min and at a pressure of about 2000 psi.
The mobile phase consisted of 10% methanol in
phosphate buffer (pH 6.0) containing 0.5 mM
tetrabutylammonium chloride, which was ®ltered
through a 0.45 mm membrane ®lter before use.
The eluate was monitored at 254 nm by a UV
detector measuring the absorption with a sensi-
tivity of AUFS 0.01. The Gilson 712 software
was used for the data analysis. The retention
time of 5-ASA, N-acetyl-5-ASA, 5-ASA-Asp, and
5-ASA-Glu was 280 s, 820 s, 360 s, and 470 s,
respectively. Concentration of 5-ASA, N-acetyl-5-
ASA, 5-ASA-Asp, or 5-ASA-Glu in the sample was
calculated from the calibration curve.
pressure reactor at 50 psi for 1 h. After ®ltering
and removal of methanol, 1 N NaOH (30 mL) was
added and reacted for 5 h under nitrogen. On
adjustment of pH to 3&4, 5-ASA-Asp (0.58 g, 70%
yield) was obtained as white precipitates [mp:
&2608C (decomp); IR (nujol) nmax(C=O): 1640,
1
1690 cm 1; H NMR (DMSO-d6): 2.8 (d, 2H), 4.7
(q, 1H), 6.0&7.2 (m, 3H)]. 5-ASA-Glu (0.35 g, 38%
yield) was obtained by following the same proce-
dure [mp: 230&2328C (decomp); IR (nujol)
1
n
max(C=O): 1640, 1731 cm
;
1H NMR (DMSO-
d6): 2.1 (m, 2H), 2.3 (t, 2H), 4.4 (m, 1H), 6.6&7.5
(m, 3H)].
Apparent Partition Coef®cient
To a solution (10 mL) of 5-ASA-Asp or 5-ASA-Glu
(100 mg equivalent of 5-ASA/mL) in pH 6.8
isotonic phosphate buffer, 10 mL of chloroform
presaturated with pH 6.8 isotonic phosphate
buffer was added. The mixture was shaken for
10 h and left for 4 h at 378C. Concentration of 5-
ASA-Asp or 5-ASA-Glu in the aqueous phase was
analyzed by HPLC. The apparent partition coef®-
cients were calculated by employing the equation
Synthesis
5-NSA (5 g, 27.3 mmol) was dissolved in 170 mL of
anhydrous ethyl acetate and DCC (6.2 g,
30.0 mmol) was added in portions, with stirring,
at 08C for 1 h. To the reaction mixture, L-aspartic
acid dimethyl ester (4.45 g, 27.3 mmol) was added,
and the mixture was stirred mechanically for 3 h
at 08C and for 72 h at room temperature. After
®ltration, the ®ltrate was evaporated under re-
duced pressure to remove the solvent. The oily
residue was extracted with a saturated solution of
NaHCO3. The combined extract was acidi®ed
with 3N HCl, extracted with ethyl acetate, dried
with anhydrous Na2SO4, and the solvent was
removed under reduced pressure. The residue
was loaded on a silica gel open column and eluted
with CHCl3/MeOH (100/1.5), from which 5-nitro-
salicyl-L-aspartic acid dimethyl ester was obtain-
ꢀCo Cw†=Cw
where Co and Cw represent the initial and
equilibrium concentration of the drug in aqueous
phase, respectively.
pH Stability
A solution of 5-ASA-Asp or 5-ASA-Glu (100 mg
equivalent of 5-ASA/mL) in pH 1.2 hydrochloric
acid buffer or in pH 6.8 phosphate buffer was
incubated at 378C. At a predetermined time
interval, a 20 mL portion of the solution was
removed, and the concentration of 5-ASA was
analyzed by HPLC.
Incubation of 5-ASA-Asp or 5-ASA-Glu With the
Homogenate of Tissue and Contents of
Stomach or Small Intestine of Rats
ed [5.1 g, 58% yield; mp: 149&1508C; IR (nujol)
1
n
max(C=O): 1640, 1725, 1749 cm
;
1H NMR
(CDCl3): 2.9&3.2 (2H), 3.7 (s, 3H), 3.8 (s, 3H),
5.0 (m, 1H), 7.0&8.8 (m, 3H)]. 5-Nitrosalicyl-L-
glutamic acid dimethyl ester was obtained by
A male Sprague-Dawley rat was anesthetized by
ether and a midline incision was made. Sections of
stomach and small intestine were collected sepa-
rately, homogenized, and diluted to half concen-
tration with isotonic acetate buffer (pH 4.5)
for stomach and with isotonic phosphate buffer
(pH 6.8) for small intestine. To each microtube,
0.8 mL of 5-ASA-Asp or 5-ASA-Glu solution in pH
6.8 isotonic phosphate buffer (140 mg equivalent of
following the same procedure [mp: 75±778C; IR
1
(nujol) nmax(C=O): 1650, 1717, 1745 cm
;
1H
NMR (CDCl3): 2.2 (m, 2H), 2.5 (t, 2H), 3.7 (s, 3H),
3.8 (s, 3H), 4.7 (q, 1H), 7.0&8.5 (m, 3H)].
5-Nitrosalicyl-L-aspartic acid dimethyl ester
(1 g, 3.95 mmol) and 200 mg of 10% Pd/C in
methanol (20 mL) were hydrogenated in a Parr
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001

1770
JUNG, LEE, AND KIM
5-ASA/0.8 mL) and 0.2 g of the above homogenate
were added, mixed, and incubated under nitrogen
at 378C. At appropriate time intervals, a sample
was taken out and centrifuged at 5,000 rpm for
3 min. Methanol (0.9 mL) was added to the
supernatant (0.1 mL). It was vortexed for 2 min,
centrifuged for 5 min at 10,000 Â g, and the
amount of 5-ASA in a 20 mL portion of the
supernatant was analyzed by HPLC as described
previously.
anesthetized with ether and cannulated with
polyethylene tubing SP 45 (0.96 mm o.d.; Natsum,
Japan) through the femoral artery. 5-ASA-Asp
(50 mg equivalent of 5-ASA/kg), which was
dissolved in 1.0 mL of physiological saline solution
by adding a minimum amount of 1N-NaOH, was
administered by oral zonde. Blood collection and
analysis of 5-ASA-Asp, 5-ASA, and N-acetyl-5-
ASA in the blood were carried out by the
procedure described in the previous section.
Incubation of 5-ASA-Asp or 5-ASA-Glu With the
Cecal and Colonic Contents of Rats
Concentration of 5-ASA-Asp, 5-ASA, and
N-acetyl-5-ASA in Feces and Urine After Oral
Administration of 5-ASA-Asp
The cecal and colonic segments of the intestine
were cut open and their contents were collected
separately in a glove box, which was previously
displaced by nitrogen. A solution of 5-ASA-Asp or
5-ASA-Glu (140 mg equivalent of 5-ASA/0.9 mL) in
pH 6.8 isotonic phosphate buffer (0.9 mL) and the
gut contents (0.1 g) were placed in a microtube
and incubated at 378C. At appropriate time
intervals, the amount of 5-ASA and N-acetyl-5-
ASA released in the medium was analyzed by
HPLC as described previously.
A male Sprague-Dawley rat was placed in a
metabolic cage and starved for 24 h prior to use
in the experiments, but had free access to water.
5-ASA-Asp (50 mg equivalent of 5-ASA/kg) was
orally administered as described previously. The
fecal and urinary samples were collected sepa-
rately at 2 h intervals and stored immediately in a
freezer before being analyzed by the following
procedure. The fecal or urinary sample was
diluted with isotonic phosphate buffer solution
(pH 6.8) in 100- or 10-fold, respectively, and
centrifuged at 5,000 rpm for 3 min. Methanol
(0.9 mL) was added to the supernatant (0.1 mL).
The mixture was vortexed for 2 min, centrifuged
for 5 min at 10,000 Â g, and the amount of 5-ASA-
Asp, 5-ASA, and N-acetyl-5-ASA in 20 mL of the
supernatant was analyzed by HPLC as described
previously.
Plasma Concentration of 5-ASA-Asp, 5-ASA, and
N-acetyl-5-ASA After Intravenous
Administration of 5-ASA-Asp
Male Sprague-Dawley rats weighting 200&250 g
were anesthetized with ether and cannulated
with polyethylene tubing SP 45 (0.96 mm o.d.;
Natsum, Japan) through the femoral artery, and
were then administered 5-ASA-Asp dissolved in
4.0 mL of physiological saline solution via the
femoral vein (10 mg equivalent of 5-ASA/kg).
Blood samples were collected from the femoral
artery with a heparinized syringe at appropriate
time intervals and centrifuged at 5,000 rpm
for 3 min. To the 0.1 mL portion of separated
plasma, 0.9 mL of methanol was added. The
mixture was vortexed for 2 min and centrifuged
at 10,000 Â g for 5 min. The amount of 5-ASA
and N-acetyl-5-ASA in 20 mL of the super-
natant was analyzed by HPLC as described
previously.
RESULTS AND DISCUSSION
Preparation of 5-ASA-Asp and 5-ASA-Glu
5-ASA-Asp and 5-ASA-Glu were prepared from
5-NSA as shown in Scheme 1, where 5-NSA (1)
was directly reacted with dimethyl ester of amino
acid in the presence of DCC to produce dimethyl
ester of 5-NSA-Asp or 5-NSA-Glu (2). Reduction
with 10% Pd/C and subsequent hydrolysis of the
ester (3) afforded 5-ASA-Asp or 5-ASA-Glu (4).
Chemical Stability and Apparent
Partition Coef®cient
Plasma Concentration of 5-ASA-Asp, 5-ASA, and
N-acetyl-5-ASA After Oral Administration of
5-ASA-Asp
Apparent partition coef®cient of 5-ASA-Asp and
5-ASA-Glu in chloroform/phosphate buffer (pH
6.8) was 0.03 and 0.04, respectively, at 378C.
Absorption of these compounds by way of trans-
Male Sprague-Dawley rats had not been given
any food for 24 h, but were given water. They were
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001

COLON-SPECIFIC PRODRUGS OF 5-ASA
1771
Figure 1. Release pro®les of 5-ASA from prodrugs
(equivalent to 140 mg of 5-ASA) during incubation with
the cecal contents (1.0 mL of 10-fold dilution in pH 6.8
isotonic phosphate buffer) of rats at 378C. Data are
mean Æ S.E. (n ˆ 5).
Scheme 1. Synthesis of 5-aminosalicyl-L-aspartic
acid and 5-aminosalicyl-L-glutamic acid.
cellular passive diffusion might be limited due to
the low partition coef®cients, unless they are
absorbed by paracellular pathways or carrier-
mediated transport as in 5-ASA.18
5-ASA-Asp and 5-ASA-Glu were incubated at
378C in pH 1.2 hydrochloric acid buffer or pH 6.8
phosphate buffer solution representing pH of
the stomach or small intestine, respectively. No
5-ASA was detected during the 6 h incubation
period, which suggested that they might be
chemically stable during the transit through the
stomach and upper intestine.
incubated with the cecal contents, the degree of
5-ASA released was about 8% in 16 h, which was
much lower than with 5-ASA-Asp. As shown
in Figure 2, the degree of 5-ASA released from
5-ASA-Gly, 5-ASA-Asp, and 5-ASA-Glu upon
incubation with the cecal contents was in the
order of 5-ASA-Gly > 5-ASA-Asp > 5-ASA-Glu.
We believe that steric hindrance imposed by the
substituent at a-position inhibited the hydrolysis
of the amide bond by the microbes. N-acetyl-5-
ASA was not detected, which con®rms earlier
®ndings that acetylation of the aromatic amino
group does not take place by the bacterial systems
in the intestine.7,14,17
Incubation With Homogenates of Various
Segments of Gastrointestinal Tract of Rats
5-ASA-Asp and 5-ASA-Glu were incubated with
various segments of GI tracts of rats at 378C, and
the release pro®les of 5-ASA were determined.
The results are shown in Figures 1 and 2. No
5-ASA was detected when 5-ASA-Asp or 5-ASA-
Glu was incubated with the homogenate of tissue
and contents of stomach or small intestine,
suggesting that the prodrug was stable against
hydrolysis by peptidases in the upper intestine.
When 5-ASA-Asp was incubated with the cecal
and colonic contents, the degree of 5-ASA released
was about 37 and 9%, respectively in 16 h.
Prodrug activation took place most readily in the
rat cecum, where the bacterial counts are as high
as in the human colon. When 5-ASA-Glu was
Because the degree of prodrug conversion from
incubation of the cecal contents and 5-ASA-Glu
was only 8% in 16 h, we did not perform any
further studies with 5-ASA-Glu.
Plasma Concentration Pro®les After Intravenous
Administration of 5-ASA-Asp
5-ASA-Asp was administered intravenously and
plasma concentrations of 5-ASA-Asp, 5-ASA, and
N-acetyl-5-ASA were investigated. The results
are shown in Figure 3. Plasma concentration of
5-ASA-Asp decreased rapidly and became unde-
tectable in 60 min. 5-ASA or N-acety-5-ASA
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001

1772
JUNG, LEE, AND KIM
was not detected in the blood, implying that
conversion of 5-ASA-Asp to 5-ASA did not take
place in the blood.
Plasma Concentration Pro®les After Oral
Administration of 5-ASA-Asp
5-ASA-Asp was administered orally and plasma
concentrations of 5-ASA-Asp, 5-ASA, and N-
acetyl-5-ASA were determined. 5-ASA-Asp was
not detected in the plasma, which suggested that
absorption of 5-ASA-Asp by way of transcellular
passive diffusion is limited due to the low
partition coef®cient, and paracellular pathways
or carrier-mediated transport may not take place.
5-ASA and N-acetyl-5-ASA appeared at very
low levels during the 24 h experimental period
(Figure 4). Considering that 5-ASA-Asp was not
converted to 5-ASA in the plasma after intrave-
nous administration (Figure 3), 5-ASA and
N-acetyl-5-ASA determined in the plasma might
be originated from microbially derived 5-ASA
from 5-ASA-Asp in the cecum or colon. Thus,
5-ASA-Asp, which is delivered to the large intes-
tine and activated by microbial action, releases
5-ASA. A portion of the released 5-ASA was
absorbed in the large intestine and appeared
in the plasma. Elimination of 5-ASA and its
Figure 2. Release pro®les of 5-ASA during incuba-
tion of 5-ASA-Asp with homogenates of various seg-
ments of GI tracts of rats at 378C. 5-ASA-Asp
(equivalent to 140 mg of 5-ASA) in 1.0 mL of 10-fold
dilution of : *, cecal contents; &, colonic contents; ~,
stomach or small intestinal tissue contents in isotonic
phosphate buffer (pH 6.8). Data are mean Æ S.E. (n ˆ 5).
Figure 3. Plasma concentration pro®les of 5-ASA-
Asp after intravenous administration of 5-ASA-Asp (10
mg equivalent of 5-ASA/kg) via femoral vein of rats.
Blood was collected at predetermined intervals from a
cannulated femoral artery. 5-ASA and N-acetyl-5-ASA
were not detected in the plasma. Data are mean Æ S.E.
(n ˆ 5).
Figure 4. Plasma concentration pro®les of 5-ASA
(&) and N-acetyl-5-ASA (^) after oral administration
of 5-ASA-Asp (50 mg equivalent of 5-ASA/kg) in rats.
Blood was collected at predetermined intervals from a
cannulated femoral artery. 5-ASA-Asp was not detected
in the plasma. Data are mean Æ S.E. (n ˆ 5).
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001

COLON-SPECIFIC PRODRUGS OF 5-ASA
1773
and N-acetyl-5-ASA should appear in the plasma
at low levels for an extended period of time, as
shown in Figure 4.
Recovery of 5-ASA-Asp, 5-ASA, and
N-acetyl-5-ASA From Feces and Urine
After Oral Administration
After oral administration of 5-ASA-Asp, concen-
tration of 5-ASA-Asp, 5-ASA, and N-acetyl-5-ASA
in feces and urine was determined. The fraction of
the dose (%) recovered as 5-ASA, N-acetyl-5-ASA,
and 5-ASA-Asp from feces was 20.0, 12.5, and
43.0, respectively, and that from urine was 6.5,
13.0, and 1, respectively, in 24 h. These results
suggested that most of 5-ASA-Asp was delivered
to the large intestine and about one-half of them
were activated to liberate 5-ASA. A portion of the
liberated 5-ASA was absorbed in the large intes-
tine and eliminated through urine along with its
metabolite, N-acetyl-5-ASA. Only a small fraction
of 5-ASA-Asp was absorbed, which was elimi-
nated rapidly through urine (&1 %), rendering
the plasma concentration level of 5-ASA-Asp
undetectable (Figure 5). Figure 6 compares the
results from 5-ASA-Asp, 5-ASA, and sulfasala-
zine, one of the most commonly prescribed
prodrugs of 5-ASA. The fraction of the dose (%)
recovered as N-acetyl-5-ASA and 5-ASA in feces
Figure 5. Plasma concentration pro®le of 5-ASA and
N-acetyl-5-ASA after oral administration of 5-ASA (50
mg/kg) in rats. Blood was collected at predetermined
intervals from a cannulated femoral artery. Data are
mean Æ S.E. (n ˆ 5).
metabolite N-acetyl-5-ASA from the plasma was
very rapid after oral administration of free 5-ASA
(Figure 5). If 5-ASA is gradually released from
5-ASA-Asp and absorbed from the colon, 5-ASA
Figure 6. Percent recovery of 5-ASA and N-acetyl-5-ASA in feces and urine after oral
administration of (A) 5-ASA, (B) 5-ASA-Asp, or (C) sulfasalazine (50 mg equivalent of 5-
ASA/kg) in rats. Data are mean Æ S.E. (n ˆ 5).
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001

1774
JUNG, LEE, AND KIM
was only 7.0 and 0.3, respectively, for 5-ASA14
(A in Figure 6), 12.5 and 20.0, respectively, for
5-ASA-Asp (B in Figure 6), and 13 and 24,
respectively, for sulfasalazine14 (C in Figure 6).
Recovery of 5-ASA-Asp from feces was 43% and
that from urine was 1% (not shown in Figure 6).
According to Zhou et al.,18 part of the absorbed
5-ASA is metabolized to N-acetyl-5-ASA, which is
secreted into the intestinal lumen predominantly.
The fraction of 5-ASA absorbed transcellularly
and not metabolized or transported paracellularly
enters the general circulation, where it is meta-
bolized in the liver or kidney and ®nally excreted
through urine as an intact drug and metabolite.
They reasoned that metabolite levels in the urine
represent systemic or hepatic metabolism, and
lumenal and fecal metabolite levels represent a
measure of intestinal metabolism. Therefore, the
fraction of N-acetyl-5-ASA in the lumen or feces
might be related to therapeutic effect, while that
in the urine might be related to side effects.
prodrugs of 5-ASA. Inertness of a carrier molecule
(aspartic acid) might be an advantage of 5-ASA-
Asp compared with sulfasalazine, where undesir-
able side effects originated from the carrier
molecule (sulfapyridine) are reported.
ACKNOWLEDGMENTS
This work was partially supported by a grant from
Pusan National University.
REFERENCES
1. Mrsny RJ. 1992. Drug absorption in the colon: A
critical review. In: Friend DR, editor. Oral Colon
Speci®c Drug Delivery. Boca Raton, FL: CRC Press.
pp. 45±84.
2. Mcleod AD, Tozer NT. 1992. Kinetic perspectives in
colonic drug delivery. In: Friend DR, editor. Oral
Colon Speci®c Drug Delivery. Boca Raton, FL: CRC
Press. pp. 85±114.
The fraction of the dose recovered as N-acetyl-
5-ASA and 5-ASA in feces for 5-ASA-Asp was
32.5%, which is comparable to sulfasalazine
(37.3%). The fraction recovered as prodrugs from
feces and urine was 43.0 and 1%, respectively. If
we compare these results with those from 5-ASA-
Gly from our previous study,14 the degree of
prodrug conversion and the fraction of prodrug
absorbed (fraction recovered in the urine)
decreased by substituting the amino acid from
glycine (5-ASA-Gly) with aspartic acid (5-ASA-
Asp). Substitution of glycine with aspartic or
glutamic acid increased hydrophilicity as well as
the steric hindrance at the a-position of the amide
bond of the prodrug, which decreased intestinal
absorption of the drug by passive diffusion in the
upper intestine and the conversion of prodrug by
the microbes in the large intestine.
3. Wilding IR, Davis SS, O'Hagan DT. 1994. Target-
ing of drugs and vaccines to the gut. Pharmacol
Ther 62:97±124.
4. Davis SS. 1990. Overcoming barriers to the oral
administration of peptide drugs. Trends Pharmacol
Sci 11:353±355.
5. Saffran M. 1992. Oral colon-speci®c drug-delivery
with emphasis on insulin. In: Friend DR, editor.
Oral Colon Speci®c Drug Delivery. Boca Raton, FL:
CRC Press. pp. 115±142.
6. Kimura T, Sudo K, Kanzaki Y, Miki K, Takeichi Y,
Kurosaki Y, Nakayama T. 1994. Drug absorption
from largeintestine: Physicochemical factors govern-
ing drug absorption. Biol Pharm Bull 17: 327±333.
7. Crotty B, Jewel JP. 1992. Drug therapy of ulcera-
tive colitis. Br J Clin Pharmacol 34:189±198.
8. Novis BH, Korzets Z, Chen P, Bernheim J. 1988.
Nephrotic syndrome after treatment with 5-amino-
salicylic acid. Br Med J 296:1442±1442.
9. Brown JP, McGarraugh GV, Parkinson TM, Win-
gard RE, Onderdonk AB. 1983. A Polymeric Drug
for Treatment of In¯ammatory Bowel Disease. J
Med Chem 26:1300±1307.
CONCLUSION
10. Kopeckova P, Kopecek J. 1990. Release of
5-aminosalicylic acid from bioadhesive N-(2-hydro-
xypropyl)methacrylamide copolymers by azoreduc-
tase in vitro. Makromol Chem 191:2037±2045.
11. Jung YJ, Lee JS, Kim HH, Kim YT, Kim YM. 1998.
Synthesis and properties of dextran-5-aminosa-
licylic acid as a potential colon-speci®c prodrug of 5-
aminosalicylic acid. Arch Pharm Res 21:179±186.
12. Yamaguchi T, Sasaki K, Nakayama T, Kimura T.
1994. Biopharmaceutical evaluation of salicylazo-
sulfanilic acid as a novel colon-targeted prodrug of
5-aminosalicylic acid. J Drug Target 2:123±131.
Synthesis of 5-ASA-Asp and 5-ASA-Glu was
achieved in good yield by a simple synthetic
route. Incubation of 5-ASA-Asp and 5-ASA-Glu
with the cecal contents of rats released 5-ASA 37
and 8%, respectively, in 16 h, which was much
lower than 80% of 5-ASA-Gly in the same
conditions. The fraction of the dose recovered as
N-acetyl-5-ASA and 5-ASA in feces, which might
be related to therapeutic effect for 5-ASA-Asp,
was 32.5%, which is comparable to sulfasalazine
(37.3%), one of the most frequently prescribed
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001

COLON-SPECIFIC PRODRUGS OF 5-ASA
1775
13. Istran C, Gabor S, Ferene S. 1991. Glycosides
of 5-aminosalicylic acid. Magy Kem FOLY 97:
143±147.
of bacterial gut hydrolysis on the fate of orally
administered isonicotinuric acid in man. J Phar-
macokin Biopharm 2:211±237.
14. Jung YJ, Lee JS, Kim YM. 2000. Synthesis and in
vitro/in vivo evaluation of 5-aminosalicyl-glycine
as a colon-speci®c prodrug of 5-aminosalicylic acid.
J Pharm Sci 89:594±602.
15. Scheline RR. 1973. Metabolism of foreign com-
pounds by gastrointestinal microorganisms. Phar-
macol Rev 25:451±523.
17. Nakamura J, Asai K, Nishida K, Sasaki H. 1992. A
novel prodrug of salicylic acid, salicylic acid-
glutamic acid conjugate utilizing hydrolysis in
rabbit intestinal microorganisms. Chem Pharm
Bull 40:2164±2168.
18. Zhou SY, Fleisher D, Pao LH, Winward UB,
Zimmerman EM. 1999. Intestinal metabolism and
transport of 5-aminosalicylate. Drug Metab Dispos
27:479±485.
16. Boxenbaum HG, Jodhka GS, Ferguson AC,
Riegelman S, MacGregor TR. 1974. The in¯uence
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 11, NOVEMBER 2001