Structural effects of N-aromatic acyl-amino acid conjugates on their deconjugation in the cecal contents of rats: Implication in design of a colon-specific prodrug with controlled conversion rate at the target site

Article abstract of DOI:10.1002/bdd.763

N-aromatic acyl-amino acid conjugates possess a colon-targeted property, implying that such conjugates are stable and are not absorbable until reaching the large intestine in which they are microbially converted (hydrolysed) to the parent drugs that are therapeutically active. To investigate the structural effect of N-aromatic acyl-amino acid conjugates on the large intestinal deconjugation, the hydrolysis of various N-aromatic acyl-amino acid conjugates was examined in the cecal contents. On incubation of conjugates with glycine,d or/andl forms of alanine or phenylalanine in the cecal contents, the conjugates withd amino acids were not hydrolysed. The other conjugates are susceptible to the hydrolysis, the rates of which decreased as the size of the substituent on the 2-position of the amino acids increased. The conjugates with alkyl analogs (2-4 carbons) of glycine and taurine were resistant to the hydrolysis, while taurine- and glycine-conjugates were hydrolysed effectively. The hydrolysis of N-aromatic acyl-glycine conjugates was enhanced by para-substitution of electron withdrawing groups on the aromatic acyl moiety and vice versa for electron-donating groups. While a methyl, methoxy or chloro group on the ortho-position retarded the hydrolysis, a hydroxyl group on the position accelerated it. Our data may provide useful information for the design of a colon-specific prodrug with controlled conversion rate in the large intestine.

Full text of DOI:10.1002/bdd.763

BIOPHARMACEUTICS & DRUG DISPOSITION
Biopharm. Drug Dispos. 32: 343–354 (2011)
Published online 28 July 2011 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/bdd.763
Structural effects of N-aromatic acyl-amino acid conjugates on
their deconjugation in the cecal contents of rats: implication in
design of a colon-specific prodrug with controlled conversion
rate at the target site
*
Hyesik Kong, Hyunjeong Kim, Heejeong Do, Yonghyun Lee, Sungchae Hong, Jeong-Hyun Yoon, Yunjin Jung ,
*
and Young Mi Kim
Laboratory of Biomedicinal Chemistry, College of Pharmacy, Pusan National University, Busan, Republic of Korea
ABSTRACT: N-aromatic acyl-amino acid conjugates possess a colon-targeted property, implying
that such conjugates are stable and are not absorbable until reaching the large intestine in which they
are microbially converted (hydrolysed) to the parent drugs that are therapeutically active. To
investigate the structural effect of N-aromatic acyl-amino acid conjugates on the large intestinal
deconjugation, the hydrolysis of various N-aromatic acyl-amino acid conjugates was examined in the
cecal contents. On incubation of conjugates with glycine, D or/and L forms of alanine or
phenylalanine in the cecal contents, the conjugates with D amino acids were not hydrolysed. The
other conjugates are susceptible to the hydrolysis, the rates of which decreased as the size of the
substituent on the 2-position of the amino acids increased. The conjugates with alkyl analogs (2–4
carbons) of glycine and taurine were resistant to the hydrolysis, while taurine- and glycine-
conjugates were hydrolysed effectively. The hydrolysis of N-aromatic acyl-glycine conjugates was
enhanced by para-substitution of electron withdrawing groups on the aromatic acyl moiety and vice
versa for electron-donating groups. While a methyl, methoxy or chloro group on the ortho-position
retarded the hydrolysis, a hydroxyl group on the position accelerated it. Our data may provide
useful information for the design of a colon-specific prodrug with controlled conversion rate in the
large intestine. Copyright © 2011 John Wiley & Sons, Ltd.
Key words: N-aromatic acyl-amino acid conjugates; colonic delivery; structural effect; colon
specific prodrug; drug design
Introduction
the site of action with limited systemic absorp-
tion, only a small dose is needed, which
subsequently results in reduced side effects.
Secondly, the colon could be the preferential site
of absorption for orally administered therapeutic
peptides and proteins. Thirdly, delayed absorp-
tion by the colonic targeting may also be useful
for the treatment of diseases susceptible to
diurnal rhythm such as asthma, gastric ulcer or
arthritis, which may have peak symptoms at bed-
time, by adjusting the plasma concentration–time
profile to the symptom [1–4].
Delivery of orally administered drugs specifically
to the colon is highly desirable for several
reasons. Firstly, for the efficient treatment of
diseases developed locally at the colon such as
ulcerative colitis, Crohn’s disease, constipation or
colorectal cancer. Being delivered specifically to
*Correspondence to: College of Pharmacy, Pusan National
University, Busan, Republic of Korea, 609–735.
E-mail: jungy@pusan.ac.kr; ymikim@pusan.ac.kr
Received 2 May 2011
Revised 31 May 2011
Accepted 16 June 2011
Copyright © 2011 John Wiley & Sons, Ltd.

344
The development of a prodrug is a way to
H. KONG ET AL.
synthesized N-nicotinoyl-2-(5-fluorouracil-1-yl)-
glycine (NFG) by substitution of 5-fluorouracil
(with a nucleophile). The NFG was designed as a
colon-specific prodrug of 5-fluorouracil because
2-(5-fluorouracil-1-yl)-glycine, which is formed
after the cleavage of the amide bond in NFG in
the cecal contents, decomposes spontaneously to
release 5-fluorouracil [17]. Substitution of other
drug with a nucleophile should produce a
compound with a general structure of N-
aromatic acyl-2-(drug)-glycine, which can be a
colon-specific prodrug [18]. In this case, N-
aromatic acyl-glycine acts as a colon-specific
promoiety and bioactivation of the prodrug
depends on the hydrolysis of the amide bond
in the colon. The bioactivation processes and
structures of the above prodrugs are shown in
Figure 1A and B.
deliver drugs specifically to the colon. A colon-
specific prodrug should be stable and non-
absorbable in the upper intestine so that it could
reach the colon in an intact form, and the linkage
between drug and promoiety should dissociate to
release the active drug in the colon for action [4].
In the prodrug approach, polymeric carriers have
been used to prevent the absorption of prodrugs
in the upper intestine [5–8]. There are cases in
which water-soluble small molecules have been
used as a colon-specific carrier [9–12]. After
being delivered to the colon, the prodrug is
presumed to be activated by the enzymes
originating from the microbes which are espe-
cially abundant in that portion of the alimentary
canal [13].
N-Aromatic acyl-amino acid conjugates are
known to be stable in the upper intestine and to
be dissociated by the microbial enzymes in the
colon [14]. For this reason, amino acid can be used
as a colon-specific promoiety for a drug with an
aromatic carboxylic acid such as 5-aminosalicylic
acid as reported by Jung et al. [15,16]. Lee et al.
In the present study, the structural effects of the
amino acids (or amino acid analogs) on the cecal
deconjugation (amide hydrolysis) of N-benzoyl-
amino acid conjugates were investigated. In
addition, the effects of various substituents on
the aromatic acyl moiety on the deconjugation
Figure 1. Bioactivation of N-aromatic acyl-amino acid conjugates in the large intestine. (A) Bioactivation of 5-aminosalicyloylamino
acid conjugates in the large intestine. (B) Bioactivation of N-nicotinoyl-2-(5-fluorouracil-1-yl)-D,L-glycine (NFG) in the large
intestine
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

CECAL METABOLISM OF N-AROMATIC ACYL-AMINO ACID CONJUGATES
345
of N-aromatic acyl-glycine were scrutinized
as well.
used most frequently for in vitro experiments to
examine activation of a colon specific prodrug
because the animal has similar intestinal microflora
composition to that of humans [19].
Materials and Methods
Materials and instruments
Preparation of N-aromatic acyl-amino acid
conjugate
Benzoic acid, salicylic acid, benzoyl chloride, 4-
methoxybenzoyl chloride, 2-methoxybenzoyl chlo-
ride, 2-chlorobenzoyl chloride, 4-chlorobenzoyl
chloride, D-alanine, glycine, taurine (2-aminoethan-
sulfonic acid), N-benzoyl-glycine, N-benzoyl
-L-alanine, N-benzoyl-D,L-alanine, N-benzoyl-L-phe-
nylalanine, N-benzoyl-D-phenylalanine, N-benzo-
yl-glycyl-D,L-phenylalanine, N-(4-nitrobenzoyl)
-glycine, N-(2-hydroxybenzoyl)-glycine, amino-
methansulfonic acid, 3-amino-1-propanesulfonic
acid, N-(4-methylbenzoyl)-glycine, N-(2-methyl-
benzoyl)-glycine and N,N′-carbonyldiimidazole
were purchased from Sigma-Aldrich Chemical
Co (St Louis, MO, USA) and used as received.
N-(4-Aminobenzoyl)-glycine, 4-amino-n-butyric
acid, N-benzoyl-D,L-phenylalanine and N-benzoyl-
b-alanine were purchased from Tokyo Kasei
(Tokyo, Japan) and used as received. Solvents
for NMR and HPLC were obtained from Merck
Inc. (Darmstadt, Germany). All other chemicals
were reagent grade, commercially available
products. The IR spectra were recorded on a
Bomem MB 100 FT-IR spectrophotometer (Bomem,
Ontario, Canada). The 1H-NMR spectra were taken
on a Varian AS 500 spectrometer (Santa Clara, CA)
and the chemical shifts are in ppm downfield from
tetramethylsilane. Elemental analysis was carried
out by an Elemental Analyzer System (Profile HV-3,
Manchester, UK). The HPLC system consisted of
Model 305, 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). Buffer solutions, hydro-
chloric acid buffer solution (pH 1.2), isotonic
phosphate buffer (pH 6.8), 0.067M phosphate buffer
(pH 4.5, pH 5.5) and 0.050M phosphate buffer (pH
6.0), were prepared as described in USP XXIII. The
animal protocol used in this study has been
reviewed and approved by the Pusan National
University–Institutional Animal Care and Use
Committee (PNU-IACUC) on their ethical proce-
dures and scientific care. The cecal content of rats is
Compounds for structure activity relationship
(SAR) study that are not commercially available
were prepared in our laboratory.
1. Preparation of N-benzoyl-2-aminoethansulfonic
acid. To the solution of 2-aminoethansulfonic acid
(1.3g, 10.5mmol) in 10ml of 1 N-sodium hydroxide,
benzoylchloride (1.4g, 10.0mmol) and 3 N-sodium
hydroxide were added separately from a separatory
funnel at such a rate that the solution was always
slightly alkaline while stirring and cooling below
30ꢀC. After the addition was completed, the
mixture was stirred for 2h longer. It was then
poured into 3ml of concentrated hydrochloric acid
in a 100ml beaker. The resulting precipitates were
collected by suction filtration, washed with cold
water, dried and recrystallized from ethanol–water.
mp: 251.3–253.8ꢀC, IR (nujol)nmax: 1193, 1051
(S=O, symmetric, asymmetric), 1640 (amide I:
C=O), 1537 (amide II: N-H), 1307 (amide III: C-
1
N), H NMR (D2O): d 3.08 (t, 2H), 3.64 (t, 2H),
7.39 (t, 2H), 7.48 (t, 1H), 7.61 (d, 2H). Elemental
analysis for C9H11SO4N: Calculated: C; 47.12, H;
4.80, S; 13.99, N; 6.11. Found: C; 47.53, H; 4.62, S;
13.64, N; 5.81.
2. Preparation of N-benzoyl-aminomethanesulfonic
acid. Aminomethanesulfonic acid (1.2g, 10.8mmol)
was reacted with benzoyl chloride (1.4g, 10.0mmol)
according to the procedure described in ‘1’). The
resulting precipitates were collected by suction
filtration, washed with cold water, dried and
recrystallized from ethanol–water. mp: 260–305ꢀC
(dec.), IR (nujol)nmax: 1187, 1047 (S=O, symmetric,
asymmetric), 1649 (amide I: C=O), 1547 (amide II:
N-H), 1297 (amide III: C-N), 1H NMR (D2O): d 4.49
(s, 2H), 7.42 (t, 2H), 7.53 (t, 1H). 7.70 (d, 2H).
Elemental analysis for C8H9SO4N: Calculated: C;
44.60, H; 4.18, S; 14.90, N; 6.50. Found: C; 44.78, H;
4.53, S; 14.34, N; 6.87.
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

346
H. KONG ET AL.
3. Preparation of N-benzoyl-3-amino-1-propanesulfonic
acid. 3-Amino-1-propanesulfonic acid (1.5g, 10.7
mmol) was reacted with benzoyl chloride (1.4g,
10.0mmol) according to the procedure described
in ‘1’). The resulting precipitates were collected
by suction filtration, washed with cold water,
dried and recrystallized from ethanol–water. mp:
254.7–255.7ꢀC, IR (nujol)nmax: 1197, 1050 (S=O,
symmetric, asymmetric), 1626 (amide I: C=O),
filtration, washed with cold water, dried and
recrystallized from ethanol–water. mp: 166.4–170ꢀ
C, IR (nujol)nmax 1741 (carboxylic acid: C=O), 1609
(amide I: C=O), 1564 (amide II: N-H), 1263 (amide
III: C-N), 1H NMR (D2O): d 3.85 (s, 3H), 4.0
(d, 2H), 7.0 (d, 2H), 7.90 (d, 2H). Elemental
analysis for C10H11O4N: Calculated: C; 57.37, H;
5.26, N; 6.69. Found: C; 57.57, H; 5.23, N; 6.94.
1
1543 (amide II: N-H), 1290 (amide III: C-N), H
7. Preparation of N-(2-methoxybenzoyl)-glycine. Gly-
cine (0.8g, 10.5mmol) was reacted with 2-meth-
oxybenzoyl chloride (1.7g, 10mmol) according to
the procedure described in ‘1’). The resulting
precipitates were collected by suction filtration,
washed with cold water, dried and recrystallized
from ethanol–water. mp: 102.2–104.5ꢀC, IR
(nujol)nmax 1693 (carboxylic acid: C=O), 1668
(amide I: C=O), 1599 (amide II: N-H), 1259 (amide
III: C-N), 1H NMR (DMSO-d6): d 3.79 (5H), 4.1 (d,
2H), 6.98 (t, 1H), 7.10 (d, 1H), 7.49 (m, 1H), 7.61 (m,
1H). Elemental analysis for C10H11O4N: Calculat-
ed: C; 57.41, H; 5.30, N; 6.70. Found: C; 56.98, H;
5.31, N; 6.88.
NMR (D2O): d 1.95 (q, 2H), 2.88 (t, 2H), 3.37
(t, 2H), 7.39 (t, 2H), 67.48 (t, 1H), 7.60 (d, 2H).
Elemental analysis for C10H13SO4N: Calculated:
C; 49.32, H; 5.34, S; 13.18, N; 5.75. Found: C; 49.57,
H; 5.63, S; 13.64, N; 5.34.
4. Preparation of N-benzoyl-D-alanine. D-Alanine
(0.95g, 10.5mmol) was reacted with benzoyl
chloride (1.4g, 10.0mmol) according to the proce-
dure described in ‘1’). The resulting precipitates
were collected by suction filtration, washed with
cold water, dried and recrystallized from ethanol–
water. mp: >350ꢀC (dec.); IR (nujol)nmax: 1732
(carboxylic acid: C=O), 1604 (amide I: C=O), 1557
1
(amide II: N-H), 1288 (amide III: C-N), H NMR
8. Preparation of N-(2-chlorobenzoyl)-glycine. Glycine
(0.8g, 10.5mmol)wasreactedwith2-chlorobenzoyl
chloride (1.75g, 10mmol) according to the proce-
dure described in ‘1’). The resulting precipi-
tates were collected by suction filtration, washed
with cold water, dried and recrystallized from
ethanol–water. mp: 186.9–188.2ꢀC, IR (nujol)nmax
1724 (carboxylic acid: C=O), 1626 (amide I: C=O),
(D2O): d 1.70 (d, 3H), 4.6 (m, 1H), 7.68 (t, 2H), 7.80
(t, 1H), 7.95 (d, 2H). Elemental analysis for
C10H11O3N: Calculated: C; 62.14, H; 5.70, N;
7.25. Found: C; 61.77, H; 5.98, N; 7.64.
5. Preparation of N-benzoyl-4-aminobutyric acid. 4-
Aminobutyric acid (1.1g, 10.9mmol) was reacted
with benzoyl chloride (1.4g, 10.0mmol) according
to the procedure described in ‘1’). The resulting
precipitates were collected by suction filtration,
washed with cold water, dried and recrystallized
from ethanol–water. mp: 65.3–66.3ꢀC, IR (nujol)
nmax 1691 (carboxylic acid: C=O), 1625 (amide I:
C=O), 1529 (amide II: N-H), 1300 (amide III: C-N),
1H NMR (CDCl3): d 1.92 (q, 2H, C-3), 2.43
(t, 2H, C-2), 3.47 (q, 2H, C-4), 7.36 (t/d, 2H), 7.45
(t/t, 2H), 7.76 (d/d, 2H). Elemental analysis for
C11H13O3N: Calculated: C; 63.69, H; 6.27, N; 6.76.
Found: C; 63.57, H; 6.63, N; 6.34.
1
1552 (amide II: N-H), 1228 (amide III: C-N), H
NMR (DMSO-d6): d 3.88 (d, 2H), 7.38-7.49 (4H),
7.85 Elemental analysis for C9H8O3N: Calculated:
C; 50.60, H; 3.77, N; 6.56. Found: C; 51.20, H; 3.79,
N; 6.61.
9. Preparation of N-(4-chlorobenzoyl)-glycine. Gly-
cine (0.8g, 10.5mmol) was reacted with 4-
chlorobenzoyl chloride (1.75g, 10mmol) according
to the procedure described in ‘1’). The resulting
precipitates were collected by suction filtration,
washed with cold water, dried and recrystal-
lized from ethanol–water. mp: 147.9–150.2ꢀC,
IR (nujol)nmax 1745 (carboxylic acid: C=O),
1684 (amide I: C=O), 1560 (amide II: N-H),
1219 (amide III: C-N), 1H NMR (DMSO-d6):
d 3.92 (d, 2H), 7.55 (m, 2H), 7.87 (m, 2H). Elemental
6. Preparation of N-(4-methoxybenzoyl)-glycine.
Glycine (0.8g, 10.5mmol) was reacted with 4-
methoxybenzoyl chloride (1.7g, 10mmol) accord-
ing to the procedure described in ‘1’). The
resulting precipitates were collected by suction
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

CECAL METABOLISM OF N-AROMATIC ACYL-AMINO ACID CONJUGATES
347
Table 1. HPLC analytical data
analysis for C9H8O3N: Calculated: C; 50.60, H;
3.77, N; 6.56. Found: C; 52.58, H; 3.57, N; 6.14.
No. of chemical
(Retention time in min)
% Methanol in mobile phase
(Methanol/5mM, pH 6.0 PBS)
TLC analysis
* 20 (22.2)
2.5
5
* 1 (14.6)
The TLCs were performed on Merck Kieselgel 60
F254 and spots were visualized by an UV monitor
at 254nm. The mobile phases consisted of
chloroform/methanol in a ratio of either 70/30
or 50/50.
* 15 (12.3)a
* 17 (18.3)
10
12.5
15
20
* 14 (12.1), 22 (11.1), 3 (10.7)
* 2 (8.7), 4 (8.7), 5 (7.6), 11 (7.7)
16 (8.2), 21 (7.0), 19 (9.2)
* 10 (6.7)
* 8 (14.9), 9 (14.7)
* 12 (5.76), 13 (12.05), 18 (11.09)
* Salicylic acid (12.5),
Benzoic acid (15.1)
30
gradientb
c
d
HPLC analysis
The concentration of N-aromatic acyl-amino acid
conjugates, benzoic acid and salicylic acid was
measured by a reverse-phase HPLC. Standard or
blank solution (1ml) was mixed on a vortex mixer
for 2min, centrifuged at 10000rpm for 5min and
filtered through a membrane filter (0.45mm). The
filtrate (20ml) was injected on a Lichrospher 100
RP-18 column (250ꢁ4.6mm, 5mm) from Merck
and eluted with a mobile phase at a flow rate of
0.7ml/min. The mobile phases consisted of
various ratios of methanol/5mM, pH 6.0 phos-
phate buffer solution, which were filtered through
a 0.45mm membrane filter before use. The eluate
was monitored by measuring the absorption at
224 or 254nm at a sensitivity of AUFS 0.01. The
Gilson 712 software was employed for the data
analysis. HPLC analytical data of the analytes are
listed in Table 1.
Flow rate: 0.7ml/min, Column: Lichrospher 100 RP-18 column (250ꢁ
4.6mm, 5mm).
aMobile phase contains 5mM tetrabutyl-ammonium chloride.
b0–4min, 20%; 4–5min, 20!50%; 5–12min, 50%; 12–13min, 50!20%;
13–25min, 20%.
c, Acetonitrile/pH 4.5 PBS=15/85.
d, Acetic acid/methanol/pH 4.5 PBS=4/35/65.
from the concentrations of the standard solutions
versus the peak areas on HPLC.
pH stability
The pH stability was determined by incubating
test compounds (100mg/ml) in either pH 1.2
hydrochloric acid buffer or 6.8 phosphate buffer
for 24h at 37ꢀC. At a predetermined time interval,
a 20ml portion of the solution was removed to
analyse the concentration of N-aromatic acyl-
amino acid conjugates by HPLC as described
previously.
Preparation of standard solutions in the rat cecal
contents
A male Sprague-Dawley rat was anesthetized by
diethyl ether and a midline incision was made.
The cecal content obtained was suspended in the
isotonic phosphate buffer (pH 6.8) solution to a
concentration of 10w/v% and filtered through
medical gauze to remove the fibrous mass. To
each 100ml portion of the suspension, 20, 40, 60,
100 or 200ml portion of each solution of either N-
aromatic acyl-amino acid conjugates (100mg/ml)
or aromatic acids (benzoic acid and salicylic acid,
100mg/ml) in isotonic phosphate buffer (pH 6.8)
and an appropriate amount of methanol were
added to make the final volume of 1ml. The
standard solutions in a concentration of 2.0, 4.0,
8.0, or 20.0mg/ml were obtained by the above
processes. A calibration curve was constructed
Incubation of an N-aromatic acyl-amino acid
conjugate with the rat cecal contents
A male Sprague-Dawley rat was anesthetized by
diethyl ether and a midline incision was made,
and the cecal content was collected in a glove box,
which was previously displaced by nitrogen. To a
0.1g portion of the cecal contents, each solution of
0.9ml of N-aromatic acyl-amino acid conjugates in
pH 6.8 isotonic phosphate buffer (100mg /0.9ml)
was added and the mixture was incubated at
37ꢀC. At appropriate time intervals, the samples
were centrifuged at 5000rpm for 3min. To a 0.1ml
portion of the supernatants, 0.9ml of methanol
was added, vortexed for 2min and centrifuged
for 5min at 10000ꢁg. The concentration of
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

348
H. KONG ET AL.
N-aromatic acyl-amino acid conjugates in a 20m
l portion of the supernatants was determined by
HPLC as described previously. In some experi-
ments, benzoic acid and salicylic acid as the cecal
metabolites of N-aromatic acyl-glycine conju-
gates were analysed by HPLC.
N-(4-nitrobenzoyl)-glycine or N-(4-aminobenzoyl)-
glycine was incubated for 24h at 37ꢀC in pH 1.2
hydrochloric acid buffer and pH 6.8 phosphate
buffer solution. None of them was hydrolysed
at these conditions, implying that the amount of
hydrolysis product produced during the incu-
bation with the rat cecal contents represents
solely that of microbial origin, not from the
chemical hydrolysis.
Results
Preparation of N-aromatic acyl-amino acid conjugates
Incubation of N-aromatic acyl-amino acid conjugates
with the rat cecal contents
Preparation of N-aromatic acyl-amino acid con-
jugates and N-aromatic acyl-aminoalkylsulfonic
acid conjugates were prepared by the addition of
acyl chloride to the solution of aqueous amino
acid or aminoalkylsulfonic acid keeping the
reaction mixture slightly alkaline throughout the
reaction period. After acidification and collection
of the resulting precipitates, the product was
obtained in good yield. The structure of the
product was confirmed from the data of IR and
1H-NMR spectra and the results from the elemen-
tal analysis. Compounds used for the present
study include N-benzoyl-glycine (1), N-benzoyl-L-
alanine (2), N-benzoyl-D-alanine (3), N-benzoyl-D,
L-alanine (4), N-benzoyl-b-alanine (5), N-benzoyl-
L-phenylalanine (6), N-benzoyl-D-phenylalanine
(7) N-benzoyl-D,L-phenylalanine (8), N-benzoyl-
glycyl-D,L-phenylalanine (9), N-(4-methylben-
zoyl)-glycine (10), N-(2-methylbenzoyl)-glycine
(11), N-(2-chlorobenzoyl)-glycine (12), N-(4-chlor-
obenzoyl)-glycine (13), N-(2-hydroxybenzoyl)
-glycine (14), N-(4-aminobenzoyl)-glycine (15),
N-(4-nitrobenzoyl)-glycine (16), N-(4-methoxyben-
zoyl)-glycine (17), N-(2-methoxybenzoyl)-glycine
(18), N-benzoyl-4-aminobutyric acid (19), N-ben-
zoyl-aminomethanesulfonic acid (20), N-benzoyl-
2-aminoethanesulfonic acid (21) and N-benzoyl-3
-aminopropanesulfonic acid (22). The structures of
the above compounds are shown in Figure 2.
N-Aromatic acyl-amino acid conjugates are
known to be stable in the upper intestine and to
be dissociated by the microbial enzymes in the
colon [14]. Bioactivation (deconjugation) of colon-
specific prodrugs that have the general structure
occurs by the enzymatic hydrolysis of the amide
bond in the large intestine [4].
To ensure whether the amide bonds in an N-
benzoyl-amino acid conjugate and an N-aromatic
acyl-glycine with a substituent on the aromatic
ring could be hydrolysed in the cecal contents,
either N-benzoylglycine or N-(2-hydroxybenzoyl)-
glycine was incubated in the cecal contents and
the hydrolysis of the conjugates and production
of the aromatic acids were monitored. As shown
in Figure 3A and B, the aromatic acids were
produced as much as the conjugates disap-
peared. This result indicates that the amide
bonds in N-aromatic acyl-amino acid conjugates
should be hydrolysed in the cecal contents.
Effect of the amino acid moiety on the cecal
hydrolysis of N-benzoyl-amino acid conjugates
Results from the incubation of N-benzoyl-amino
acid conjugates with the cecal contents of rats are
shown in Figure 4. It was observed that hydro-
lysis took place mostly with the L-amino acid
conjugate. Hydrolysis hardly occurred when N-
benzoyl-D-alanine or N-benzoyl-D-phenylalanine
was incubated with the cecal contents of rats. It is
seen from Figure 3 that the degree of hydrolysis
reached approximately 90% for glycine, 52% for L-
alanine, 37% for L-phenylalanine, 30% for D,L-
alanine and 19% for D,L-phenylalanine at 8h after
incubation. The degree of hydrolysis reached 94%,
88%, 85%, 45% and 47% for glycine, L-alanine,
Chemical stability of N-aromatic acyl-amino acid
conjugate
To investigate whether chemical hydrolysis of
the N-aromatic acyl-amino acid conjugate takes
place at the pHs of the various sections of
the gastrointestinal tract, N-benzoyl-glycine, N-
benzoyl-taurine, N-benzoyl-D,L-alanine, N-benzoyl-
D,L-phenylalanine, N-(2-hydroxybenzoyl)-glycine,
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

CECAL METABOLISM OF N-AROMATIC ACYL-AMINO ACID CONJUGATES
349
Figure 2. Structures of N-aromatic acyl-amino acid conjugates. (A) Structures of N-benzoyl-amino acid conjugates. (B) Structures of
N-aromatic acyl glycine conjugates. (C) Structures of N-benzoyl-aminoalkylsulfonic acids
L-phenylalanine, D,L-alanine and D,L-phenylala-
nine, respectively at 24h. As the size of the
substituent on the 2-position of amino acid
increased from H to CH3, CH2C6H5, the rate of
hydrolysis decreased, which appeared most
prominently at 8h. To confirm this, the degree
of hydrolysis of N-benzoyl-L-phenylalanine was
compared with that of N-benzoyl-glycyl-D,L-
phenylalanine with an insertion of a glycyl
moiety between the benzoyl group and glycine
as a spacer group. The degree of hydrolysis
increased from 37% to 49% at 8h. Glycine and
its isotere, taurine, are known to be good colon
specific carriers of 5-aminosalicylic acid [16,20].
To examine an effect of alkyl group length of the
amino acids on the hydrolysis, a series of alkyl
homologs of glycine was incubated with the cecal
contents of rats and the effect of chain length on
hydrolysis was observed. The hydrolysis profiles
are shown in Figure 5. Hydrolysis did not take
place with N-benzoyl-b-alanine and N-benzoyl-4-
aminobutyric acid. To compare the structural
effect of the substitution of amino acid with
aminoalkylsulfonic acid, a homologous series of
N-benzoyl-aminoalkylsulfonic acids were pre-
pared and incubated with the cecal contents of
rats. The hydrolysis profiles of the conjugates are
shown in Figure 6. While N-benzoyl-2-aminoetha-
nesulfonic acid (N-benzoyl-taurine) was hydro-
lysed to the extent of 86% in 24h, the others
were completely resistant to deconjugation in
the cecal contents. These results suggest that the
hydrolysis in the cecal contents of rats is largely
confined to the amide bond originating from the
natural amino acid or natural metabolite such
as taurine.
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

350
H. KONG ET AL.
Figure 3. Hydrolysis of N-benzoyl glycine and N-(2-hydroxybenzoyl) glycine and production of the aromatic acids during
incubation with the cecal contents of rats. Either N-benzoyl glycine or N-(2-hydroxybenzoyl) glycine was incubated with 10% cecal
contents in isotonic phosphate buffer (pH 6.8) at 37ꢀC (100mg of conjugates/0.1g cecal contents). The change in concentration of
the conjugates and their metabolites was measured. Data are meanꢂSE (n=3–5)
Figure 4. Hydrolysis (%) profiles of N-benzoyl-amino acid
conjugates during incubation with the cecal contents of rats.
N-benzoyl-amino acid conjugates were incubated with 10%
cecal contents in isotonic phosphate buffer (pH 6.8) at 37ꢀC
(100mg of conjugates/0.1g cecal contents). The change in
concentration of the conjugates was measured. Data are
meanꢂSE (n=3–5)
Figure 5. Hydrolysis (%) profiles of alkyl homologs of N-
benzoyl-glycine conjugates during incubation with the cecal
contents of rats. Alkyl homologs of N-benzoyl-glycine con-
jugates were incubated with 10% cecal contents in isotonic
phosphate buffer (pH 6.8) at 37ꢀC (100mg of conjugates/0.1g
cecal contents). The change in concentration of the conjugates
was measured. Data are meanꢂSE (n=3–5)
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

CECAL METABOLISM OF N-AROMATIC ACYL-AMINO ACID CONJUGATES
351
Figure 6. Hydrolysis (%) profiles of N-benzoyl-aminoalkylsul-
fonic acid conjugates during incubation with the cecal contents
of rats. N-benzoyl-aminoalkylsulfonic acid conjugates were
incubated with 10% cecal contents in isotonic phosphate buffer
(pH 6.8) at 37ꢀC (100mg of conjugates/0.1g cecal contents).
The change in concentration of the conjugates was measured.
Data are meanꢂSE (n=3–5)
Figure 7. Hydrolysis (%) profiles of N-benzoyl-glycine con-
jugates with substituent on 4-position during incubation with
the cecal contents of rats. N-aromatic acyl-glycine conjugates
were incubated with 10% cecal contents in isotonic phosphate
buffer (pH 6.8) at 37ꢀC (100mg of conjugates/0.1g cecal
contents). The change in concentration of the conjugates was
measured. Data are meanꢂSE (n=3–5)
at an early stage of incubation. The strong
electron-donating amino group elicited a greater
retarding effect on the hydrolysis than the weak
electron-donating groups methyl and methoxy
group. However, the degree of hydrolysis be-
came about the same level after 24h. On the other
hand, when a methyl, chloro or methoxy group
was introduced on the ortho-position, the hydro-
lysis was severely impaired. Ortho-substitution
with a hydroxyl group on the ortho-position,
namely N-(2-hydroxybenzoyl)-glycine, acceler-
ated the hydrolysis. The results are shown in
Figure 8.
Effect of substituents on the aromatic acyl moiety
on the cecal hydrolysis of N-aromatic acyl-glycine
conjugates
N-(Aromatic acyl)-glycine conjugates with a
substituent on the para position of the aromatic
ring were incubated with the cecal contents of
rats. Data from hydrolysis profiles are shown in
Figure 7. It is seen from Figure 7 that the degree of
hydrolysis reached 90% for N-(benzoyl)-glycine,
95% for N-(4-chlorobenzoyl)-glycine, 96% for N-(4-
nitrobenzoyl)-glycine, 57% for N-(4-methylbenzoyl)-
glycine, 56% for N-(4-methoxybenzoyl)-glycine and
41% for N-(4-aminobenzoyl)-glycine at 8h after
incubation. The degree of hydrolysis reached 94%
for N-(benzoyl)-glycine, 87% for N-(4-methylben-
zoyl)-glycine, 88% for N-(4-methoxybenzoyl)-
glycine and 90% for N-(4-aminobenzoyl)-glycine
at 24h. N-(4-Nitrobenzoyl)-glycine and N-(4-
chlorobenzoyl)-glycine were completely hydro-
lysed at 10h after incubation. Compared with the
hydrolysis rate of N-(benzoyl)-glycine, the hydro-
lysis was accelerated when electron-withdrawing
groups were substituted on the para-position and
vice versa for electron-donating groups especially
Discussion
In this study, it was demonstrated that N-aromatic
acyl-amino acid conjugates that possess various
amino acids and substituents on the aromatic ring
exhibit differences in the rates of hydrolysis in the
cecal contents. The stereochemistry, size of a
substituent at the 2 position and the alkyl chain
length of amino acids affect the cecal conversion
rates of the conjugates. Moreover, the rates are
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

352
H. KONG ET AL.
responsible for the hydrolysis of the conjugates
recognize natural amino acids in the conjugates.
This speculation is based on the findings that
glycine and taurine, but not alkyl homologs of
them, are susceptible to the hydrolysis.
In addition to amino acids in the conjugates, a
substituent on the aromatic ring influenced the
hydrolysis rate. Our results suggest that a
substituent on the para-position of the aromatic
ring elicited an electronic effect on the hydrolysis
of the conjugates. The hydrolysis occurred faster
when an electron-withdrawing group was substi-
tuted on the para-position and vice versa for
electron-donating groups, especially at an early
stage of incubation. The observation that sub-
stitutions on the ortho-position severely impaired
the hydrolysis implies that substituents at the
positions imposed steric hindrance. On the other
hand, although the steric effect of the ortho
substitution seemed to be predominant, an
electronic effect still confers an influence on the
hydrolysis rate. This is based on the observation
that substitution with a chloro group, which is
similar in size to a methyl group, exhibited faster
hydrolysis than that with a methyl group.
However, substitution on the ortho position does
not always impose such a steric barrier as the
conjugate with a hydroxyl group on the ortho
position is susceptible to the hydrolysis, which
occurred a little faster than the unsubstituted
conjugate. This observation is in parallel with a
previous finding that 5-aminosalicylglycine with
an ortho-hydroxyl group is deconjugated effec-
tively in the cecal contents [14]. The ortho-
hydroxyl group effect might be caused by some
special kind of interaction, such as intramolecular
hydrogen bonding.
As mentioned in the introduction, colon-specific
drug delivery technology is directed mainly for the
efficient treatment of colonic diseases such as
colorectal cancer, constipation, irritable and inflam-
matory bowel disease and intestinal infection,
chronotherapeutic treatment of asthma, hyperten-
sion, arthritis and gastric ulcer and orally active
peptide therapy. Therefore, all therapeutic agents
available for the diseases are candidate drugs for
modification to colon specific prodrugs [4]. A group
of drugs with aromatic carboxylic acid or sulfon-
amide could be designed as a colon specific
prodrug using various amino acids as a colon
Figure 8. Hydrolysis (%) profiles of N-benzoyl-glycine with
substituent on 2-position during incubation with the cecal
contents of rats. N-aromatic acyl-glycine conjugates were
incubated with 10% cecal contents in isotonic phosphate buffer
(pH 6.8) at 37ꢀC (100mg of conjugates/0.1g cecal contents).
The change in concentration of the conjugates was measured.
Data are meanꢂSE (n=3–5)
also modulated by the electronic and steric
properties of a substituent on the aromatic ring.
Our data showing stereo-specificity of the
hydrolysis for conjugates with L-amino acids are
in line with a previous observation that the
deconjugation of aromatic acid-amino acid con-
jugates occurs by microbial enzymes in the cecal
contents [4]. As the degree of hydrolysis of the
racemic conjugates was approximately half of the
corresponding L-conjugates, the D-conjugates do
not seem to affect the deconjugation of the
respective L-conjugates. Considering that an
enzymatic reaction is generally sensitive to steric
hindrance, this microbial enzyme-dependent
hydrolysis may rationalize our observation that as
the size of substituents at the 2-position of amino
acids increased, the rates of hydrolysis decreased.
The steric effect at the 2-position of amino acids
was confirmed by comparing the hydrolysis rate
of N-benzoyl-L-phenylalanine with that of N-
benzoyl-glycyl-D,L-phenylalanine with reduced
steric hindrance at the 2-position. The hydrolysis
rate of N-benzoyl-glycyl-D,L-phenylalanine was
greater. It seems that the enzymes that are
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

CECAL METABOLISM OF N-AROMATIC ACYL-AMINO ACID CONJUGATES
353
specific carrier (please see Figure 1A). On the other
hand, N-aromatic acyl-2-(acetoxy)-glycine could be
used as a platform for the design of a colon specific
prodrug (N-aromatic acyl-2-(drug)-glycine) of drugs
with a nucleophile (please see Figure 1B).
the aromatic ring, it is likely to design a colon
specific prodrug with controlled conversion rates
in the large intestine, which influences the
regional therapeutic availability of the colon
specific prodrug in the intestinal tract.
In a previous review [4], the rate of conversion
of a colon specific prodrug to its parent drug is
important in that it could affect the regional
availability of the prodrug in the large intestine.
The regional availability should influence the
therapeutic effect of prodrugs for the treatment
of colonic diseases whose lesion is located in
different regions in the large intestine. If the
conversion rate is rapid enough for a prodrug to
liberate most of the parent drug in the proximal
part of the large intestine, such a prodrug may not
provide a sufficient amount of the drug for the
distal part of the large intestine. This limitation in
distribution of the drug would be aggravated if
the drug is hydrophobic and/or is susceptible to
colonic metabolism. While a hydrophobic drug
would have trouble with diffusion in the contents
in the distal part of the large intestine where the
water content is very low, a drug susceptible to
colonic metabolism may be metabolically inacti-
vated before reaching the distal part. On the
contrary, if the conversion rate is too slow, a
sufficient amount of a drug would not be
supplied at any target site in the large intestine.
Thus, our data demonstrating structural effects
of the conjugates on the rate of the cecal conversion
might provide valuable information for the design
of a colon specific prodrug with a controlled-
release property at the target site. However, for the
development of a clinically useful colon specific
prodrug, other factors, species and inter-individual
difference of microflora and the possibility of
premature activation in small intestine, that influ-
ence the therapeutic activity of a colon specific
prodrug should be considered as well. A review
paper deals with these issues extensively [4].
Acknowledgements
This work was supported by the Korea Research
Foundation (KRF) grants funded by the Korea
government (MEST) (KRF-2008-314-E00315) and
(KRF-2007-314-E00245).
References
1. Kinget R, Kalala W, Vervoort L, van den Mooter G.
Colonic drug targeting. J Drug Target 1998; 6:
129–149.
2. Friend DR. Review article: issues in oral adminis-
tration of locally acting glucocorticosteroids for
treatment of inflammatory bowel disease. Aliment
Pharmacol Ther 1998; 12: 591–603.
3. Friend DR. New oral delivery systems for treat-
ment of inflammatory bowel disease. Adv Drug
Deliv Rev 2005; 57: 247–265.
4. Jung Y, Kim YM. What should be considered on
design of a colon-specific prodrug? Expert Opin
Drug Deliv 2010; 7: 245–258.
5. Brown JP, McGarraugh GV, Parkinson TM, et. al. A
polymeric drug for treatment of inflammatory
bowel disease. J Med Chem 1983; 26: 1300–1307.
6. Canevari M, Castagliuolo I, Brun P, et al. Poly
(ethylene glycol)-mesalazine conjugate for colon
specific delivery. Int J Pharm 2009; 368: 171–177.
7. Jung YJ, Lee JS, Kim HH, et al. Synthesis and
properties of dextran-5-aminosalicylic acid ester
as a potential colon-specific prodrug of 5-amino-
salicylic acid. Arch Pharm Res 1998; 21: 179–186.
8. McLeod AD, Friend DR, Tozer TN. Glucocorticoid-
dextran conjugates as potential prodrugs for colon-
specific delivery: hydrolysis in rat gastrointestinal
tract contents. J Pharm Sci 1994; 83: 1284–1288.
9. Jung Y, Kim HH, Kim H, et al. Evaluation of 5-
aminosalicyltaurine as a colon-specific prodrug of
5-aminosalicylic acid for treatment of experimental
colitis. Eur J Pharm Sci 2006; 28: 26–33.
Conclusion
10. Friend DR, Chang GW. Drug glycosides: potential
prodrugs for colon-specific drug delivery. J Med
Chem 1985; 28: 51–57.
11. Dhaneshwar SS, Gairola N, Kandpal M, et al.
Colon-specific, mutual azo prodrug of 5-amino-
salicylic acid with L-tryptophan: synthesis, kinetic
studies and evaluation of its mitigating effect in
The hydrolysis rates of N-aromatic acyl-amino
acid conjugates in the cecal contents depend on
the structure of the amino acid and a substituent
in the aromatic ring. By introduction of an
appropriate amino acid and/or a substituent in
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd

354
H. KONG ET AL.
trinitrobenzenesulfonic acid-induced colitis in
rats. Bioorg Med Chem 2007; 15: 4903–4909.
colon-specific prodrug of 5-aminosalicylic acid. J
Pharm Sci 2000; 89: 594–602.
12. Kim I, Kong H, Lee Y, et al. Dexamethasone 21-
sulfate improves the therapeutic properties of
dexamethasone against experimental rat colitis
by specifically delivering the steroid to the large
intestine. Pharm Res 2009; 26: 415–421.
13. Sinha VR, Kumria R. Microbially triggered drug
delivery to the colon. Eur J Pharm Sci 2003; 18: 3–18.
14. Jung YJ, Lee JS, Kim HH, et al. Synthesis and
evaluation of 5-aminosalicyl-glycine as a potential
colon-specific prodrug of 5-aminosalicylic acid.
Arch Pharm Res 1998; 21: 174–178.
15. Jung YJ, Lee JS, Kim YM. Colon-specific prodrugs of
5-aminosalicylic acid: synthesis and in vitro/in vivo
properties of acidic amino acid derivatives of 5-
aminosalicylic acid. J Pharm Sci 2001; 90: 1767–1775.
16. Jung YJ, Lee JS, Kim YM. Synthesis and in vitro/
in vivo evaluation of 5-aminosalicyl-glycine as a
17. Lee JS, Jung YJ, Kim YM. Synthesis and evaluation
of N-acyl-2-(5-fluorouracil-1-yl)-D,L-glycine as a
colon-specific prodrug of 5-fluorouracil. J Pharm
Sci 2001; 90: 1787–1794.
18. Kim D, Hong S, Jung S, et al. Synthesis and
evaluation of N-nicotinoyl-2-{2-(2-methyl-5-
nitroimidazol-1-yl)ethyloxy}-D,L-glycine as a
colon-specific prodrug of metronidazole. J
Pharm Sci 2009; 98: 4161– 4169.
19. Haeberlin B, Friend DR. Anatomy and physiology
of the gastrointestinal tract: implication for colonic
drug delivery. In Oral Colon-Specific Drug Delivery,
Friend DR (ed.). CRC Press: Florida, 1992; 1–43.
20. Jung YJ, Kim HH, Kong HS, Kim YM. Synthesis
and properties of 5-aminosalicyl-taurine as a
colon-specific prodrug of 5-aminosalicylic acid.
Arch Pharm Res 2003; 26: 264–269.
Copyright © 2011 John Wiley & Sons, Ltd.
Biopharm. Drug Dispos. 32: 343–354 (2011)
DOI: 10.1002/bdd