Structure-activity relationship of salicylic acid derivatives on inhibition of TNF-α dependent NFκB activity: Implication on anti-inflammatory effect of N-(5-chlorosalicyloyl)phenethylamine against experimental colitis

Article abstract of DOI:10.1016/j.ejmech.2011.11.030

To develop a more potent NFκB inhibitor from salicylic acid which is known to inhibit activity of NFκB, a transcription factor regulating genes involved in immunity, inflammation and tumorigenesis, derivatives of salicylic acid (SA) where the 5 position, carboxyl or hydroxyl group was modified were treated in HCT116 cells transfected with an NFκB dependent luciferase gene and LPS-stimulated RAW264.7 cells. Amidation of the carboxylic group or substitution of chlorine at the 5 position increased the ability of SA to suppress the expression of NFκB dependent luciferase and inducible nitric oxide synthase, a product of an NFκB target gene. Moreover, simultaneous amidation and chlorination of SA (5-chlorosalicylamide; 5-CSAM) conferred an additive NFκB inhibitory activity on SA. To further enhance the inhibitory activity, N-modification was imposed on 5-CSAM. N-(5-chlorosalicyloyl) phenethylamine (5-CSPA), N-(5-chlorosalicyloyl)3-phenylpropylamine (5-CSPPA) and N-(5-chlorosalicyloyl)4-hydroxyphenylethylamine (5-CSHPA) showed greater potencies for inhibiting NFκB activity than other derivatives. Their IC₅₀s' in the luciferase assay measured 15 μM (5-CSPA), 17 μM (5-CSPPA) and 91 μM (5-CSHPA). Rectal administration of 5-CSPA ameliorated TNBS-induced rat colitis, which was more effective than a conventional drug, 5-aminosalicylic acid. These data may provide useful information for development of a therapeutic agent for treatment of diseases where NFκB plays a critical role in the pathogenic progresses.

Full text of DOI:10.1016/j.ejmech.2011.11.030

European Journal of Medicinal Chemistry 48 (2012) 36e44
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Structureeactivity relationship of salicylic acid derivatives on inhibition
of TNF- dependent NF B activity: Implication on anti-inflammatory effect
a
k
of N-(5-chlorosalicyloyl)phenethylamine against experimental colitis
*
Jihye Kim, Sookjin Kang, Sungchae Hong, Soowhan Yum, Young Mi Kim, Yunjin Jung
Laboratory of Biomedicinal Chemistry, College of Pharmacy, Pusan National University, Busan 609-735, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
To develop a more potent NFkB inhibitor from salicylic acid which is known to inhibit activity of NFkB,
Received 10 August 2011
Received in revised form
2 November 2011
Accepted 16 November 2011
Available online 25 November 2011
a transcription factor regulating genes involved in immunity, inflammation and tumorigenesis, deriva-
tives of salicylic acid (SA) where the 5 position, carboxyl or hydroxyl group was modified were treated in
HCT116 cells transfected with an NF
Amidation of the carboxylic group or substitution of chlorine at the 5 position increased the ability of SA
to suppress the expression of NF B dependent luciferase and inducible nitric oxide synthase, a product of
an NF B target gene. Moreover, simultaneous amidation and chlorination of SA (5-chlorosalicylamide; 5-
CSAM) conferred an additive NF B inhibitory activity on SA. To further enhance the inhibitory activity, N-
kB dependent luciferase gene and LPS-stimulated RAW264.7 cells.
k
k
Keywords:
k
Salicylic acid derivatives
Structureeactivity relationship
Inflammatory bowel disease
modification was imposed on 5-CSAM. N-(5-chlorosalicyloyl)phenethylamine (5-CSPA), N-(5-
chlorosalicyloyl)3-phenylpropylamine (5-CSPPA) and N-(5-chlorosalicyloyl)4-hydroxyphenylethylamine
(5-CSHPA) showed greater potencies for inhibiting NF
k
B activity than other derivatives. Their IC50s’ in
M (5-CSHPA). Rectal adminis-
NFkB
the luciferase assay measured 15 M (5-CSPA), 17 M (5-CSPPA) and 91 m
m
m
tration of 5-CSPA ameliorated TNBS-induced rat colitis, which was more effective than a conventional
drug, 5-aminosalicylic acid. These data may provide useful information for development of a therapeutic
agent for treatment of diseases where NF
k
B plays a critical role in the pathogenic progresses.
Ó 2011 Elsevier Masson SAS. All rights reserved.
1. Introduction
cytokines and chemokines that are pivotal mediators of the
immune and inflammatory responses.
NF
involved in immunity and inflammation [1]. Stimulation with pro-
inflammatory cytokines such as TNF- initiates an intracellular
signaling cascade, resulting in the phosphorylation and subsequent
degradation of I B by the 26S-proteasome [2]. The degradation of
releases NF B allowing it to translocate into the nucleus and
k
B is an important transcription factor that regulates genes
NF
kB activity and levels of the pro-inflammatory cytokine TNF-
a
have been shown to be increased in the colon epithelial cells and
mucosa of patients with inflammatory bowel disease (IBD) which
comprises ulcerative colitis and Crohn’s disease [3]. It is known that
expression of TNF-
regulated by NF B. This provides a positive autoregulatory loop
a
k
k
a, which strongly activates NFkB, is itself up-
Ik
Ba
k
transactivates target genes such as cyclooxygenase-2 (COX-2),
that amplifies the inflammatory response and perpetuates chronic
intestinal inflammation [3]. For this reason, therapeutic interven-
tion against TNF-
of inflammatory bowel disease (IBD) [4]. In fact, inhibition of NF
activity has been suggested to be a major component of the anti-
inflammatory activity of glucocorticoid and 5-aminosalicylic acid
(5-ASA) both of which are frequently used for treatment of chronic
intestinal inflammation [5,6]. Moreover, new strategies that
specifically regulate NF
oligonucleotides, proteasome inhibitors or adenoviral I
sion vector show beneficial therapeutic effects in experimental
colitis [7,8].
a or NFkB activation has been used for treatment
Abbreviation: IBD, inflammatory bowel disease; SA, salicylic acid; 5-CSAM, 5-
chlorosalicylamide; 5-CSPA, N-(5-chlorosalicyloyl)phenethylamine; 5-CSPPA, N-
kB
(5-chlorosalicyloyl)3-phenylpropylamine;
5-CSHPA,
N-(5-chlorosalicyloyl)4-
hydroxyphenylethylamine; 5-ASA, 5-aminosalicylic acid; TNF, tumor necrosis
factor; COX-2, cyclooxygenese-2; iNOS, inducible nitric oxide synthase; 5-DFPSA,
diflunisal; 5-NSA, 5-nitrosalicylic acid; 5-CSA, 5-chlorosalicylic acid; ASP, ace-
tylsalicylic acid; MSA, methylsalicylate; SAM, salicylamide; CINC, cytokine-induced
neutrophil chemoattractant; TNBS, 2,4,6-trinitrobenzene sulfonic acid; MPO,
myeloperoxidase.
k
B activity using p65 (RelA) antisense
B expres-
k
* Corresponding author. Tel.: þ82 51 510 2527; fax: þ82 51 513 6754.
E-mail address: jungy@pusan.ac.kr (Y. Jung).
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.11.030

J. Kim et al. / European Journal of Medicinal Chemistry 48 (2012) 36e44
37
Table 1
the anti-inflammatory activity of one of the derivatives, 5-
chlorosalicylic acid phenethylamide (5-CSPA), was evaluated in
a TNBS-induced rat colitis model, which was compared to 5-ASA,
the most widely used drug for the treatment of IBD.
Structures of salicylic acid derivatives.
2. Results
2.1. Structureeactivity relationship of salicylic acid derivatives on
inhibition of TNF-a dependent NFkB activity
Compd. name
R1
R2
R3
SA
H
H
H
H
H
OH
OH
OH
OH
5-NSA
5-ASA
5-CSA
NO2
NH2
Cl
To investigate structureeactivity relationship of salicylates on
kB inhibition, salicylic acid derivatives (structures of the deriv-
NF
atives used are shown in Table 1), diflunisal (5-DFPSA), 5-
nitrosalicylic acid (5-NSA), 5-aminosalicylic acid (5-ASA), 5-
chlorosalicylic acid (5-CSA), acetylsalicylic acid (ASP), methyl-
salicylate (MSA) and salicylamide (SAM), were subjected to a lucif-
erase assay using colon carcinoma HCT116 cells transfected with an
5-DFPSA
H
OH
MSA
SAM
ASP
H
H
H
CH3
H
COCH3
OH
NH2
OH
NFk
B dependent luciferase gene. As shown in Fig. 1A, stimulation
induced approximately 18-fold luciferase activity in
with TNF-
a
cells, which was attenuated to a varied degree by treatment with
salicylates. While the potencies of 5-DFPSA, 5-CSA, MSA and SAM
were greater than that of salicylic acid (SA), the other derivatives,
ASP, 5-NSA and 5-ASA, showed similar or less potencies than that of
SA. Because modifications of SA such as substitution of chlorine at 5
position or amidation of the carboxyl group elicited the positive
effect on the potency, 5-CSA was amidated to afford 5-
chlorosalicylamide (5-CSAM), expecting to further increase the
potency of the inhibitory activity. As shown in Fig. 1B, consistent
with the hypothesis, the ability of 5-CSAM to suppress the lucif-
erase expression was greater than that of either 5-CSA or SAM. To
confirm NFkB inhibition by the salicylates, SA, 5-CSA, 5-SAM and 5-
CSAM, in cells, RAW264.7 cells were treated with LPS in the pres-
ence of the salicylates and inducible nitric oxide synthase (iNOS)
protein expression, a target gene product of NFkB [12], was moni-
tored. As shown in Fig. 1C, all the salicylates attenuated iNOS
SA: salicylic acid; 5-NSA: 5-nitrosalicylic acid; 5-ASA: 5-aminosalicylic acid; 5-CSA:
5-chlorosalicylic acid; 5-DFPSA: 5-(2,4 difluorophenyl)salicylic acid; MSA: methyl-
salicylate; SAM: salicylamide; ASP: acetylsalicylic acid.
Recent papers reported that NFkB inhibition is one of the anti-
inflammatory mechanisms of salicylates such as sulfasalazine,
salicylate, aspirin and 5-ASA [9e11]. Although the salicylates whose
structures are different possess a common biological activity, there
has been no prior study on structureeactivity relationship on the
NF
acid were purchased or prepared and structureeactivity relation-
ship of salicylic acid derivatives on inhibition of TNF- dependent
NF B activity was investigated, which could provide valuable
information for development of a new therapeutic agent for
treatment of diseases where NF B plays a critical role as a patho-
logical mechanism. Based on the potential therapeutic application,
kB inhibition activity. In the present work, derivatives of salicylic
a
k
k
expression and their potencies were on the order of 5-CSAM > 5-
Fig. 1. Structureeactivity relationship of salicylic acid derivatives on TNF-
a
mediated expression of NF
k
B dependent luciferase. A. HCT116 cells were cotransfected with NF
kB
dependent luciferase plasmid (0.4 g) and CMV Renilla luciferase plasmid (4 ng) and subsequently treated with TNF-
m
a
(1 ng/ml) in the presence of salicylic acid derivatives (1 mM)
for 6 h. Reporter activities were measured and normalized to CMV Renilla luciferase activity. SA: salicylic acid, 5-NSA: 5-nitrosalicylic acid, 5-ASA, 5-CSA, 5-DFPSA, MSA, SAM, ASP. B.
The same experiment as in “A” was done in the presence of 0.5 mM of SA, 5-CSA, 5-CSAM or SAM. C. RAW264.7 cells were pretreated with various concentration of SA, 5-CSA, 5-
CSAM or SAM for 1 h followed by stimulation with LPS (1 ng/ml) for 6 h. Levels of iNOS protein in the whole cell lysates were analyzed by Western blot.

38
J. Kim et al. / European Journal of Medicinal Chemistry 48 (2012) 36e44
CSA/5-SAM > SA, which is in good agreement with the results of
the luciferase experiments. Although 5-CSAM showed an enhanced
dependent luciferase gene expression in a dose dependent
manner and IC50 values measured 15, 17 and 91 M for 5-CSPA, 5-
CSPPA and 5-CSHPA, respectively. To confirm the NF B inhibitory
activity of the derivatives, LPS-mediated iNOS protein expression
was monitored in the presence of 5-CSPA, 5-CSATY and 5-CSAPPA
in RAW264.7 cells. Consistent with the results of the luciferase
assay, the derivatives attenuated the expression of iNOS protein
(Fig. 2C).
m
inhibitory activity against NF
0.34 mM for 50% inhibition of NF
k
B, it still required approximately
B. For further improvement of
k
k
therapeutic activity of 5-CSAM, 5-CSAM was modified to afford N-
substituted derivatives. A variety of derivatives of 5-CSAM were
prepared (structures of the derivatives are shown in Table 2). They
were subjected to an NF
stimulated with TNF-
k
B dependent luciferase assay. Cells were
a
in the presence of the derivatives. As
shown in Fig 2A, of the derivatives, N-(5-chlorosalicyloyl)benzyl-
amine 9, N-(5-chlorosalicyloyl)phenethylamine (5-CSPA) 10, N-
(5-chlorosalicyloyl) 3-phenylpropylamine (5-CSPPA) 11, N-(5-
chlorotsalicyloyl) 4-hydroxyphenylethylamine (5-CSHPA) 12 and
2.2. N-(5-chlorosalicyloyl)phenethylamine administered rectally
ameliorates TNBS-induced colitis rats
To examine whether N-(5-chlorosalicyloyl)phenethylamine (5-
CSPA) that was assessed as one of the most potent derivatives for
N-(5-chlorosalicyloyl)hexylamine
3 elicited greater potencies
than 5-CSAM. To determine IC50 of 5-CSPA, 5-CSHPA and 5-CSPPA,
the luciferase assay was performed at various concentrations
NF
inflammatory bowel disease (IBD) whose pathological process is
associated profoundly with increased NF B activity [3], 5-CSPA was
kB inhibition is therapeutically active against diseases such as
of the derivatives. As shown in Fig. 2B, they inhibited NF
k
B
k
Table 2
Synthetic reactions and structures of N-substituted 5-chlorosalicylamide (CSAM) derivatives.
Compd. no.
R1
R2
IC50 (m
M)a
1
2
3
H
CH3
H
eCH3
eCH3
e(C6H13
>250
>250
95
)
n
9
H
H
H
105
15
10
11
17
12
H
91
Compd. no.
X
Y
IC50
4
5
6
O
N
N
e
>250
>250
>250
H
eCH3
7
8
N
C
>250
>250
H2
a
IC50 was measured in an NF
kB dependent luciferase assay.

J. Kim et al. / European Journal of Medicinal Chemistry 48 (2012) 36e44
39
Fig. 2. Structureeactivity relationship of N-substituted 5-CSAM derivatives on TNF-
a
mediated expression of NF
k
B dependent luciferase A. A. HCT116 cells were cotransfected with
(1 ng/ml) in the presence of N-substituted 5-
NF B dependent luciferase plasmid (0.4 g) and CMV Renilla luciferase plasmid (4 ng) and subsequently treated with TNF-a
k
m
chlorosalicylamide derivatives (0.25 mM) for 6 h. Reporter activities were measured and normalized to CMV Renilla luciferase activity. B. The same experiment as in “A” was
done in the presence of various concentrations of 5-CSPA, 5-CSPPA or 5-CSHPA. C. RAW264.7 cells were 1 h pretreated with various concentration of SA (5 mM), 5-CSPA (0.1 mM), 5-
CSPPA (0.1 mM) or 5-CSHPA (0.25 mM) followed by stimulation with LPS (1 ng/ml) for 6 h. Levels of iNOS protein in the whole cell lysates were analyzed by Western blot.
administered rectally to rats with colitis induced by TNBS once
a day for 6 days, which was started 3 days after the induction of
inflammation. The same experiment was done with 5-
aminosalicylic acid that is the active agent of anti-IBD drugs, sul-
fasalazine and mesalazine, to compare potency and effectiveness.
Rectal route instead of oral route was used to simplify analysis of
the therapeutic effect by minimizing pharmacokinetic influence.
showed no damage but the control colon, the inflamed colon with
no medication, was severely damaged showing scab by the
hemorrhagic necrosis of the mucosa, stricture and extensive serosal
adhesion to other organs. Rectal administration of 5-CSPA signifi-
cantly healed the damaged colon in a dose dependent manner, and
5-CSPA (300 mM) seemed more effective than 5-ASA (photographs
of the colonic tissues are shown in Supplementary data). Fig. 3B
showed that 5-CSPA lowered the level of MPO activity to about 60%
Dose of 5-ASA was 30 mM in a buffer solution (500
a previous report [13] and the doses of 5-CSPA were 100 and
300 M in the same volume of the buffer. The anti-inflammatory
mL) according to
(for 100
medication) and the effectiveness of 5-CSPA (300
than that of 5-ASA, which is in accordance with the recovery of the
colonic damage. Expression of NF B target gene products involved
mM) and 35% (for 300
mM) of the control (group with no
m
mM) was greater
activity of the drugs was assessed by colonic damage score (CDS)
and myeloperoxidase (MPO) activity. Moreover, protein levels of
k
pro-inflammatory NF
k
B
target genes such as cyclooxygenase
in inflammation, CINC-3 and COX-2 protein, was examined in the
inflamed tissue. As shown in Fig. 4A and B, levels of the inflam-
matory mediators were elevated in the inflamed colonic tissue
compared with the normal colonic tissue. Consistent with MPO and
CDS results, 5-CSPA decreased the levels of the inflammatory
mediators in the inflamed tissue in a dose dependent manner, and
(COX)-2 and, cytokine-induced neutrophil chemoattractant (CINC)-
3 were measured in the inflamed colonic tissues to examine
whether the anti-inflammatory effect was mediated by inhibition
of NF
kB. As shown in Fig. 3A, in which CDS represent the extent of
colonic injury by TNBS-induced inflammation, the normal colon

40
J. Kim et al. / European Journal of Medicinal Chemistry 48 (2012) 36e44
Fig. 3. Rectally administered 5-CSPA ameliorates TNBS-induced colitis of rats. A. Colonic damage score (CDS) was assigned for each rat by four independent observers who were
blind to the experimental condition according to the modified CDS scoring system. The data are means ꢀS.E. (n ¼ 5), *P < 0.05 vs. control, **P < 0.001 vs. control. B. MPO activities
were measured in the distal colon segment (4 cm) as described under “Experimental part”. The data are means ꢀS.E. (n ¼ 5), *P < 0.005 vs. control, **P < 0.01 vs. control, ***P < 0.05.
the suppressive effect of 5-CSPA (300
was greater than 5-ASA.
mM) on the molecular indices
salicylic acid (SA) against NF
showed an additive effect on NF
k
B, where simultaneous modification
kB inhibition. N-monosubstituted
derivatives of 5-chlorosalicylamide (5-CSAM), N-(5-chlorosalicyloyl)
benzylamine, N-(5-chlorosalicyloyl)phenethylamine (5-CSPA), N-(5-
chlorosalicyloyl) 3-phenylpropylamine, N-(5-chlorosalicyloyl) 4-
hydroxyphenylethylamine and N-(5-chlorosalicyloyl)hexylamine,
further increased the inhibitory activity. Furthermore, 5-CSPA, one
3. Discussion
In this study, it was demonstrated that amidation or/and 5-
chlorination of salicylic acid increased inhibitory activity of
Fig. 4. Rectally administered 5-CSPA suppressed the expression of pro-inflammatory NF
k
B target gene products A. Tissues were homogenized with a dounce homogenizer and
incubated on ice for 30 min. The homogenates were centrifuged at 13,000 rpm and 4 ꢁC for 10 min. The supernatant was subjected to Western blot analysis and COX-2 protein was
detected using a monoclonal anti-COX-2 antibody. B. The inflamed distal colon (1 g) was removed and mixed with RIPA buffer (3 mL) followed by homogenization. The homog-
enates were centrifuged at 13,000 rpm for 10 min and the supernatants (100
supernatants was subjected to CINC-3 ELISA. The data are means ꢀS.E. (n ¼ 5), *P < 0.001 vs. control, **P < 0.005.
m
L) were centrifuged again at 13,000 rpm and 4 ꢁC for 10 min. An appropriate volume of the

J. Kim et al. / European Journal of Medicinal Chemistry 48 (2012) 36e44
41
of the most potent derivatives, ameliorated TNBS-induced colitis
of rats, which was more effective than 5-ASA, a conventional anti-
IBD drug.
Inhibitory activity of SA against NFkB could be enhanced by
chemical modification such as 5-substitution or/and amidation.
needs to be confirmed by testing more derivatives in elaborated
conditions where the pharmacokinetic factors are well-controlled.
5-CSPA elicited a substantial anti-inflammatory activity against
TNBS-induced colitis of rats as shown in the data demonstrating
the improvement of colonic damage score and attenuation of MPO
activity. The therapeutic activity of 5-CSPA seems to be relevant to
This is shown by the data demonstrating that the potencies of 5-
CSA and salicylamide (SAM) for inhibiting NF
k
B was greater than
inhibition of NF
inflamed tissue is supported by showing that 5-CSPA substantially
suppressed the expression of pro-inflammatory NF B target gene
products such as COX-2 and CINC-3, which is in line with the
in vitro results of the NF B luciferase assay and iNOS expression.
This hypothesis on a molecular mechanism of therapeutic action is
consistent with the recent findings that NF B inhibition is a prom-
ising therapeutic target for treatment of IBD [4,15]. The data
showing that 5-CSPA (300 M) was more effective at ameliorating
kB in the inflamed tissue. NFkB inhibition in the
SA and simultaneous modification to produce 5-CSAM further
increased the inhibitory potency, which suggests that the structural
effect of each modification is additive. Although substitution of 2,4
difluorophenylgroup at 5 position of SA, namely diflunisal, afforded
an equivalent inhibitory effect to that of chlorine, the modification
was not adopted as a lead for further modification because its
solubility became too low on modification of the carboxylic group
of diflunisal. N-modification of 5-CSAM was found to further
k
k
k
m
increase the potency for NF
tution, was taken by previous papers demonstrating that the bio-
logical activities (including NF B inhibition) of caffeic acid (3,4
k
B inhibition. The direction, N-substi-
the rat colitis than 5-ASA (30 mM) imply that 5-CSPA is much more
potent than 5-ASA. Although molecular mechanisms of 5-ASA
underlying anti-IBD activity are still controversial [16], previous
k
dihydroxycinnamic acid) are enhanced by esterification with alkyl
or arylalkyl alcohols [12,14]. It seems that N-monosubstituted
derivatives elicit greater potency than N-disubstituted derivatives.
However, it is not certain whether the difference is derived from
pharmacokinetic or pharmacodynamic factors. This observation
papers provide a number of evidence that NFkB inhibition is crucial
for its therapeutic activity [6,9]. Therefore, their difference in
therapeutic effectiveness may come from that in the inhibitory
potency on NFkB. Recently, peroxisome proliferator-activated
receptor-r (PPAR-r), which is involved in the regulation of colon
Table 3
Physical data of N-substituted CSAM derivatives.
Compd. no.
1
Mp (ꢁC)
IR (Nujol,
1H NMR (DMSO-d6, ppm)
M.S. m/z
[M þ H]þ
eC]O, cmꢂ1
)
E.A (calculated/found)
183e185
1640
1583
1635
2.80 (s, 3H, CH3), 6.83 (d, J ¼ 8.8 Hz, 1H, H-3), 7.29 (dd, J ¼ 8.7, 2.7 Hz, 1H, H-4), 7.82
(d, J ¼ 3.0 Hz, 1H, H-6)
196
210
256
C, 51.77; H, 4.34; N, 7.55/C, 51.53; H, 4.31; N, 7.57
2
3
166e168
2.78e2.92 (br d, 6H, N(CH3)2), 6.86 (d, J ¼ 8.8 Hz, 1H, H-3), 7.10 (d, J ¼ 2.6, 1H, H-6), 7.23
(dd, J ¼ 8.8, 2.7 Hz, 1H, H-4)
C, 54.15; H, 5.05; N, 7.02/C, 54.23; H, 5.07; N, 7.05
65
0.84 (t, J ¼ 7.1 Hz, 3H, CH3), 1.22e1.31 (m, 6H, CH2CH2(CH2)3CH3, 1.51 (p, J ¼ 7 Hz, 2H,
CH2CH2(CH2)3CH3, 3.25 (q, J ¼ 5.9 Hz, 2H, NCH2CH2(CH2)3CH3), 6.91
(d, J ¼ 9.0 Hz, 1H, H-3), 7.41 (dd, J ¼ 8.8, 2.7 Hz, 1H, H-4), 7.93 (d, J ¼ 2.7, 1H, H-6)
C, 61.05; H, 7.09; N, 5.48/C, 61.16; H, 7.04; N, 5.44
4
5
6
7
218e221
187e188
164e165
200e202
1570
1574
1576
1572
3.16e3.58 (br m, 8H, morpholine), 6.86 (d, J ¼ 8.8 Hz, 1H, H-3), 7.15 (d, J ¼ 2.7 Hz, 1H, H-6),
253
241
255
317
7.24 (dd, J ¼ 8.7, 2.5 Hz, 1H, H-4)
C, 54.67; H, 5.00; N, 5.80/C, 54.74; H, 5.03; N, 5.76
3.05e3.98 (br m, 8H, piperazine), 6.84 (d, J ¼ 8.6 Hz, 1H, H-3), 7.09 (d, J ¼ 2.7 Hz, 1H, H-6),
7.18 (dd, J ¼ 8.8. 2.7 Hz, 1H, H-4)
C, 54.89; H, 5.44; N, 11.64/C, 54.81; H, 5.47; N, 11.59
2.16 (s, 3H, N(CH3)), 2.28 (br m, 4H, N(CH2)2), 3.33 (br t, 4H, OCN(CH2)2), 6.85
(d, J ¼ 8.8 Hz, 1H, H-3), 7.1 (d, J ¼ 2.7 Hz, 1H, H-6), 7.2 (dd, J ¼ 8.6, 2.9 Hz, 1H, H-4)
C, 56.58; H, 5.94; N, 11.00/C, 56.77; H, 5.98; N, 11.05
3.13 (br m, 4H, N(CH2)2), 3.32 (br m, 4H, OCN(CH2)2), 6.89 (d, J ¼ 8.8 Hz, 1H, H-3), 7.18
(d, J ¼ 2.7 Hz, 1H, H-6), 7.27 (dd, J ¼ 8.5, 2.7 Hz, 1H, H-4),6.79 (t, J ¼ 7.3 Hz,
1H, H-40), 6.93 (d, J ¼ 8.1 Hz, 2H, H-20,60), 7.21 (t, J ¼ 8.3 Hz, 2H, H-30,50)
C, 64.46; H, 5.41; N, 8.84/C, 64.71; H, 5.38; N, 8.89
8
192e194
146e147
120e121
1565
1638
1639
1.46e1.56 (m, 6H, (CH2)3, 3.13e3.53 (br d, 4H, N(CH2)2), 6.85 (d, J ¼ 8.7 Hz, 1H, H-3),
7.09 (d, J ¼ 2.7 Hz, 1H, H-6), 7.21 (dd, J ¼ 8.8, 2.7 Hz, 1H, H-4)
C, 60.13; H, 5.89; N, 5.84/C, 60.31; H, 5.85; N, 5.83
4.50 (d, J ¼ 5.9 Hz, 2H, NCH2), 6.94 (d, J ¼ 8.8 Hz, 1H, H-3), 7.4 (dd, J ¼ 8.8, 2.4 Hz, 1H,
H-4), 7.98 (d, J ¼ 2.7 Hz, 1H, H-6), 7.25 (m, 1H, H-40), 7.32 (m, 4H, H-20, 30, 50, 60)
C, 64.25; H, 4.62; N, 5.35/C, 64.41; H, 4.64; N, 5.32
2.84 (t, J ¼ 7.1 Hz, 2H, NCH2CH2), 3.51 (m, 2H, NCH2CH2), 6.92 (d, J ¼ 8.8 Hz, 1H, H-3),
7.4 (dd, J ¼ 8.8, 2.7 Hz, 1H, H-4), 7.9 (d, J ¼ 2.4 Hz, 3H, 6-H), 7.19 (t, J ¼ 8.5, 1H, H-40),
7.22e7.30 (m, 4H, H-20, 30, 50, 60)
251
272
286
9
10
C, 65.34; H, 5.12; N, 5.08/C, 65.42; H, 5.09; N, 5.05
11
12
99e100
1638
1638
1.83 (p, J ¼ 7.3 Hz, 2H, NCH2CH2CH2), 2.62 (t, J ¼ 7.9 Hz, 2H, NCH2CH2CH2), 3.30
(q, J ¼ 6.8 Hz, 2H, NCH2CH2CH2), 6.92 (d, 8.8 Hz, 1H, H-3), 7.41 (dd, J ¼ 8.8, 2.7 Hz, 1H, H-4),
7.9 (d, J ¼ 2.4 Hz, 1H, H-6), 7.15 (t, J ¼ 7.4 Hz, 1H, H-40), 7.20e7.28 (m, 4H, H-20, 30, 50, 60)
C, 66.32; H, 5.57; N, 4.83/C, 66.60; H, 5.55; N, 4.85
290
292
176e178
2.71 (t, J ¼ 7.3 Hz, 2H, NCH2CH2), 3.44 (q, J ¼ 6.4 Hz, 2H, NCH2CH2), 6.91 (d, 8.8 Hz, 1H, H-3),
7.41 (dd, J ¼ 8.8, 2.7 Hz, 1H, H-4), 7.89 (d, J ¼ 2.5 Hz, 1H, H-6), 6.67 (d, J ¼ 8.5 Hz, 2H,
H-20, 60), 7.01 (d, J ¼ 8.6 Hz, 2H, H-30, 50)
C, 61.76; H, 4.84; N, 4.80/C, 61.77; H, 4.81; N, 4.79
CSAM: 5-chlorosalicylamide.

42
J. Kim et al. / European Journal of Medicinal Chemistry 48 (2012) 36e44
inflammation [17], is identified as a therapeutic target of 5-ASA for
treatment of intestinal inflammation [18]. Thus, it would be inter-
esting to investigate that N-substituted derivatives including 5-
CSPA can activate the nuclear receptor.
(C]O) cmꢂ1
;
1H NMR (DMSO-d6):
d
¼ 3.86 (s, 3H, CH3), 7.01 (d,
J ¼ 8.8 Hz, 1H, H-3), 7.53 (dd, J ¼ 8.4, 2.7 Hz, 1H, H-4), 7.69 (d,
J ¼ 2.4 Hz, 1H, H-6); C8H7O3Cl1 (186.59) calcd: C, 51.49; H, 3.78;
found: C, 51.82; H, 3.72.
It is not sure that the therapeutic activity of 5-CSPA adminis-
tered rectally could be reproduced upon administration via oral
route. A drug administered orally must come through absorption,
distribution, metabolism and excretion to reach its drug target.
Considering pharmacokinetic factors and the potency in animal
study, 5-CSPA still needs to be polished to develop to a therapeutic
agent for treatment of IBD. However, 5-CSPA could be subjected to
a preclinical study by adoption of a colon-specific drug delivery
system, which might present a similar therapeutic situation to
rectal administration. Colon-specific delivery of 5-CSPA implies
that 5-CSPA administered orally is delivered to the large intestine
without significant loss in the upper intestine leading to greater
concentration at the large intestine and lower concentration in the
blood, generally resulting in increasing therapeutic effectiveness
and lowering systemic side effects [19]. Development of a colon-
specific drug delivery system of 5-CSPA using pharmaceutical
formulation and a prodrug approach is under study.
5.2.2. 5-Chlorosalicylamide (CSAM)
Methyl 5-chlorosalicylate (0.5 g) was dissolved in 25% ammonia
solution (10 mL) and stirred at RT for 2 days. The reaction mixture
was cooled down in an ice bath to afford CSAM as white precipitate.
Yield: 0.35 g (78%); mp: 200e202 ꢁC; IR (Nujol): 1655 (C]O) cmꢂ1
;
1H NMR (DMSO-d6):
J ¼ 8.4, 2.4 Hz, 1H, H-4), 7.94 (d, J ¼ 3.0 Hz, 1H, H-6); C7H6N1O2 Cl1
(171.58) calcd: C, 49.00; H, 3.52; N, 8.16; found: C, 49.12; H, 3.55; N,
8.15.
d
¼ 6.90 (d, J ¼ 8.9 Hz, 1H, H-3), 7.43 (dd,
5.2.3. N-substituted 5-chlorosalicylamide derivatives
Methyl 5-chlorosalicylate (0.5 g) was dissolved in methanol
(15 mL) followed by addition of 10 fold molar concentration of
methylamine, dimethylamine, hexylamine, piperidine, piperazine,
morpholine, 1-methylpiperazine, 1-phenylpiparzine, benzylamine,
phenethylamine, 3-phenylpropylamine or (4-hydroxyphenyl)ethyl-
amine, which was evaporated after stirring at RT for 36 h. The
residue was dissolved in ethylacetate, which was washed with 1 M
HCl and 5% NaHCO3 and subsequently dried over anhydrous Na2SO4.
Flash evaporation of ethylacetate afforded N-(5-chlorosalicyloyl)
methylamine 1 (yield: 0.46 g, 88%), N-(5-chlorosalicyloyl)dimethyl-
amine 2 (yield: 0.47 g, 85%), N-(5-chlorosalicyloyl)hexylamine 3
(yield: 0.58 g, 85%), N-(5-chlorosalicyloyl)morpholine 4 (yield:
4. Conclusion
Our data may provide useful information for development of
a therapeutic agent for treatment of diseases where NF
a critical role in the pathogenic progresses.
kB plays
5. Experimental part
0.51 g, 76%), 4-(5-chlorosalicyloyl)1-phenylpiperazine 7 (yield:
0.54 g, 64%), N-(5-chlorosalicyloyl)piperidine 8 (yield: 0.52 g, 78%),
N-(5-chlorosalicyloyl)benzylamine 9 (yield: 0.64 g, 89%), N-(5-
chlorosalicyloyl)phenethylamine 10 (yield: 0.62 g, 81%), N-(5-
chlorosalicyloyl) 3-phenylpropylamine 11 (yield: 0.64 g, 83%) or N-
(5-chlorosalicyloyl) 4-hydroxyphenylethylamine 12 (yield: 0.61 g,
79%) as white powders, respectively. To obtain 4-(5-chlorosalicyloyl)
piperazine 5 and 4-(5-chlorosalicyloyl)1-methylpiperazine 6, the
reaction mixtures were evaporated at 90 ꢁC followed by the washing
process (distilled water was used instead of 1 M HCl) and subse-
quent flash evaporation, which afforded the final products as white
powder (yield of 5: 0.32 g, 51%; yield of 6: 0.44 g, 65%). The physical
5.1. Reagents
Recombinant human TNF-a was obtained from R & D systems
(MN, USA). LPS and 2,4,6-trinitrobenzene sulfonic acid (TNBS)
were purchased from SigmaeAldrich (MO, USA). Salicylic acid
(SA), 5-nitrosalicylic acid (5-NSA) and 5-aminosalicylic acid (5-
ASA) were obtained from Tokyo Kasei Kogyo Co. (Tokyo, Japan).
Acetylsalicylic acid (ASP), 5-chlorosalicylic acid (5-CSA), diflu-
nisal (DFA), methylsalicylate (MSA), salicylamide (SAM),
methylamine, dimethylamine, benzylamine, phenethylamine, 3-
phenpropylamine, (4-hydroxyphenyl)ethylamine, hexylamine,
piperidine, morpholine, piperazine, 1-methylpiperazine and 1-
phenylpiperazine were purchased from SigmaeAldrich. Reac-
tion solvents were obtained from Junsei chemical Co. (Tokyo,
Japan). All other chemicals were reagent-grade, commercially
available products. IR spectra were recorded on a Varian FT-IR
spectrophotometer (Varian, CA, USA), respectively. 1H NMR
spectra were taken on a Varian AS 500 spectrometer. Elemental
analysis was carried out by an Elemental Analyzer System
(Profile HV-3, Manchester, UK). Mass spectra were obtained
using an Agilent QQQ 6460 mass spectrometer (Agilent Tech-
nologies, CA, USA). The animal protocol used in this study has
been reviewed and approved by the Pusan National Uni-
versityeInstitutional Animal Care and Use Committee (PNU-
IACUC) on their ethical procedures and scientific care.
data and IC50 values measured in an NFkB-dependent luciferase
assay and structures of N-substituted CSAM derivatives were shown
in Tables 2 and 3.
5.3. Cell lines
Human colon carcinoma HCT116 cells (ATCC, MD, USA) and
murine macrophage RAW264.7 cells (ATCC) were grown in DMEM
medium (Hyclone, Utah, USA) supplemented with 10% fetal bovine
serum and penicillin/streptomycin (Hyclone).
5.4. Luciferase assay
Cells were plated in 6-well plates to be 50e60% confluent on
the day of transfection with NF
kB dependent luciferase plasmid
5.2. Synthesis
(0.5 g, a gift from Dr. M. Birrer, NCI) and cytomegalovirus
m
(CMV) Renilla luciferase plasmid (4 ng, Promega, WI, USA).
Fugene (Roche, CA, USA) was used as a transfection reagent.
Twenty-four hours post-transfection, cells were treated with
5.2.1. Methyl 5-chlorosalicylate (MCSA)
5-Chlorosalicylic acid (1 g) was dissolved in methanol (15 mL)
and concentrated H2SO4 (approximately 20
m
L) was added, which
TNF-a in the presence of the reagent at the indicated concen-
was evaporated after 1 h reflux. The residue dissolved in
ethylacetate was washed with 5% NaHCO3 and subsequently dried
over anhydrous Na2SO4, which was subjected to flash evaporation
to obtain MCSA. Yield: 0.96 g (89%); mp: 46 ꢁC; IR (Nujol): 1682
trations in the figure legends. Cells were lysed 6 h later and
luciferase activities were measured and normalized to CMV
Renilla luciferase activities using a Dual Luciferase reporter assay
system (Promega).

J. Kim et al. / European Journal of Medicinal Chemistry 48 (2012) 36e44
43
5.5. Immunoblot analysis
myeloperoxidase activity is defined as that degrading 1
mmol of
peroxide per minute at 25 ꢁC.
Cells were lysed and cell lysates were prepared as described
previously [20]. To prepare tissue total lysates, the inflamed distal
colon was removed and mixed with 3 ml of ice cold RIPA buffer
(50 mM TriseHCl (pH 7.4), 1 mM EDTA, 0.7% Na deoxycholate, 1%
5.8. ELISA for cytokine-induced neutrophil chemoattractant
(CINC)-3
NP-40, 150 mM NaCl, 0.3 mM Aprotinin, 1 mM Pepstatin and 1 mM
To measure CINC-3 levels in the tissue, the inflamed distal colon
(1 g) was removed and mixed with RIPA buffer (3 mL) followed by
homogenization. The homogenates were centrifuged at 13,000 rpm
PMSF) per gram of tissue. Tissues were further disrupted and
homogenized with a homogenizer and incubated on ice for 30 min.
Tissues were centrifuged at 13,000 rpm and 4 ꢁC for 10 min.
Supernatants were removed and centrifuged again. Protein
concentration in the supernatants was determined by the BCA
method. The extracts were transferred to a fresh tube and stored
at ꢂ80 ꢁC until used. Cell or tissue extracts were electrophoretically
separated using 7.5 or 10%. Proteins were transferred to nitrocel-
lulose membranes (Protran, Schleicher & Schuell, NH, USA). COX-2
protein in tissues homogenates was detected using a monoclonal
anti-COX-2 antibody (Santa Cruz, CA, USA). iNOS protein was
and 4 ꢁC for 2 min. The supernatants (100
mL) transferred to a fresh
microtube and then were centrifuged at 13,000 rpm and 4 ꢁC for
10 min. An appropriate volume of the supernatants was subjected
to CINC-3 ELISA (R&D Systems, MN, USA).
5.9. Statistical analysis
The results are expressed as the mean ꢀ S.D. One-way ANOVA
followed by Tukey’s (HSD) test was used for testing the difference
between data. Differences with p < 0.05 were considered signifi-
cant. The XLSTATÒ Software (Addinsoft, Inc, NY, USA) was used for
the statistical analysis.
detected in whole cell lysates (30e40 mg) using a polyclonal anti-
iNOS antibody (Santa Cruz). Signals were visualized using the
Supersignal chemiluminescence substrate (Pierce, IL, USA). Exper-
iments were performed in duplicate and equivalent loading was
confirmed by probing the blots with anti-a-tubulin antibody (Santa
Cruz).
Acknowledgments
5.6. TNBS-induced inflammation
This work was supported by a grant of the Korea Healthcare
technology R & D Project, Ministry for Health, Welfare and Family
Affairs, Republic of Korea (A080640) awarded to Y.J.
Inflammation was induced by the method of Morris et al. [21]
and Yano et al. [22]. Briefly, before induction of colitis, rats were
starved for 24 h but had free access to water. The rats were lightly
anesthetized with ether. A rubber cannula (OD, 2 mm) was inserted
rectally into the colon such that the tip was 8 cm proximal to the
anus, approximately at the splenic flexture. TNBS dissolved in 50%
(v/v) aqueous ethanol was instilled into the colon via the rubber
cannula (15 mg/0.3 ml/rat).
Appendix. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ejmech.2011.11.030.
References
5.7. Evaluation of TNBS-induced colitis
[1] Q. Li, I.M. Verma, NF-kappaB regulation in the immune system, Nat. Rev.
Immunol. 2 (2002) 725e734.
[2] K. Tanaka, T. Kawakami, K. Tateishi, H. Yashiroda, T. Chiba, Control of Ikap-
paBalpha proteolysis by the ubiquitin-proteasome pathway, Biochimie 83
(2001) 351e356.
[3] M.F. Neurath, C. Becker, K. Barbulescu, Role of NF-kappaB in immune and
inflammatory responses in the gut, Gut 43 (1998) 856e860.
[4] Y. Yamamoto, R.B. Gaynor, Therapeutic potential of inhibition of the NF-
kappaB pathway in the treatment of inflammation and cancer, J. Clin.
Invest. 107 (2001) 135e142.
Three days after induction of inflammation, N-(5-
chlorosalicyloyl)phenethylamine (100, 300
mM, 500 mL) or sodium
5-aminosalicylate (30 mM, 500 L) dissolved in PBS was adminis-
m
tered rectally once a day and the rats were sacrificed after the
treatment for 6 days. A gross colonic damage score (CDS) was
calculated according to the criteria set forth previously [21,22]. The
modified scoring system is: 0, normal appearance; 1, localized
hyperemia but no ulcer; 2, linear ulcers without significant
inflammation; 3, 2e4 cm site of inflammation and ulceration
without scab; 4, serosal adhesion to other organs, 2e4 cm site of
inflammation and ulceration with scab; 5, stricture, serosal adhe-
sion involving several bowel loops, <4 cm site of inflammation and
ulceration with scab. Four independent observers blinded to the
treatment did the assessment of colonic damage score. Using the
distal colon (4 cm), MPO activity was measured as described
previously [23]. The distal colon segment, which was kept at
ꢂ80 ꢁC, was finely minced in a vial containing 1 ml of 0.5% (HTAB,
pH 6.0) and homogenized using a polytron homogenizer on ice.
After homogenization, the homogenizer was rinsed with HTAB
solution. HTAB was added to adjust the pooled homogenates to
100 mg tissue per ml which was sonicated for 10 s, frozen-thawed
three times, and centrifuged at 13,000 rpm at 4 ꢁC. The superna-
tants, 0.1 ml, were combined with 2.9 ml of 0.050 M phosphate
buffer solution (pH 6.0) containing 0.167 mg/ml o-dianisidine
hydrochroride and 0.0005% hydrogen peroxide. The change in
absorbance at 460 nm was measured by an UV-2101 PC spectro-
photometer (Shimadzu, Tokyo, Japan) for 5 min at 25 ꢁC. One unit of
[5] N. Auphan, J.A. DiDonato, C. Rosette, A. Helmberg, M. Karin, Immunosup-
pression by glucocorticoids: inhibition of NF-kappa
B activity through
induction of I kappa B synthesis, Science 270 (1995) 286e290.
[6] F. Yan, D.B. Polk, Aminosalicylic acid inhibits IkappaB kinase alpha phos-
phorylation of IkappaBalpha in mouse intestinal epithelial cells, J. Biol. Chem.
274 (1999) 36631e36636.
[7] M.F. Neurath, S. Pettersson, K.H. Meyer zum Buschenfelde, W. Strober, Local
administration of antisense phosphorothioate oligonucleotides to the p65
subunit of NF-kappa B abrogates established experimental colitis in mice, Nat.
Med. 2 (1996) 998e1004.
[8] E.M. Conner, S. Brand, J.M. Davis, F.S. Laroux, V.J. Palombella, J.W. Fuseler,
D.Y. Kang, R.E. Wolf, M.B. Grisham, Proteasome inhibition attenuates nitric
oxide synthase expression, VCAM-1 transcription and the development of
chronic colitis, J. Pharmacol. Exp. Ther. 282 (1997) 1615e1622.
[9] G.C. Kaiser, F. Yan, D.B. Polk, Mesalamine blocks tumor necrosis factor growth
inhibition and nuclear factor kappaB activation in mouse colonocytes,
Gastroenterology 116 (1999) 602e609.
[10] M.J. Yin, Y. Yamamoto, R.B. Gaynor, The anti-inflammatory agents aspirin and
salicylate inhibit the activity of I(kappa)B kinase-beta, Nature 396 (1998)
77e80.
[11] C.K. Weber, S. Liptay, T. Wirth, G. Adler, R.M. Schmid, Suppression of NF-
kappaB activity by sulfasalazine is mediated by direct inhibition of IkappaB
kinases alpha and beta, Gastroenterology 119 (2000) 1209e1218.
[12] Y. Lee, D.H. Shin, J.H. Kim, S. Hong, D. Choi, Y.J. Kim, M.K. Kwak, Y. Jung, Caffeic
acid phenethyl ester-mediated Nrf2 activation and IkappaB kinase inhibition
are involved in NFkappaB inhibitory effect: structural analysis for NFkappaB
inhibition, Eur. J. Pharmacol. 643 (2010) 21e28.

44
J. Kim et al. / European Journal of Medicinal Chemistry 48 (2012) 36e44
[13] Y. Jung, H.H. Kim, H. Kim, H. Kong, B. Choi, Y. Yang, Y. Kim, Evaluation of 5-
aminosalicyltaurine as a colon-specific prodrug of 5-aminosalicylic acid for
treatment of experimental colitis, Eur. J. Pharm. Sci. 28 (2006) 26e33.
[14] D. Choi, J. Han, Y. Lee, J. Choi, S. Han, S. Hong, H. Jeon, Y.M. Kim, Y. Jung, Caffeic
acid phenethyl ester is a potent inhibitor of HIF prolyl hydroxylase: structural
analysis and pharmacological implication, J. Nutr. Biochem. 21 (2010)
809e817.
[15] I. Atreya, R. Atreya, M.F. Neurath, NF-kappaB in inflammatory bowel disease,
J. Intern. Med. 263 (2008) 591e596.
[16] P. Del Soldato, M. Campieri, C. Brignola, G. Bazzocchi, P. Gionchetti,
G.A. Lanfranchi, M. Tamba, A possible mechanism of action of sulfasalazine
and 5-aminosalicylic acid in inflammatory bowel diseases: interaction with
oxygen free radicals, Gastroenterology 89 (1985) 1215e1216.
P. Bulois, A. Cortot, J.F. Colombel, P. Desreumaux, Intestinal antiinflammatory
effect of 5-aminosalicylic acid is dependent on peroxisome proliferator-
activated receptor-gamma, J. Exp. Med. 201 (2005) 1205e1215.
[19] Y. Jung, Y.M. Kim, What should be considered on design of a colon-specific
prodrug? Expert Opin. Drug Deliv. 7 (2010) 245e258.
[20] N.C. Andrews, D.V. Faller, A rapid micropreparation technique for extraction
of DNA-binding proteins from limiting numbers of mammalian cells, Nucleic
Acids Res. 19 (1991) 2499.
[21] G.P. Morris, P.L. Beck, M.S. Herridge, W.T. Depew, M.R. Szewczuk, J.L. Wallace,
Hapten-induced model of chronic inflammation and ulceration in the rat
colon, Gastroenterology 96 (1989) 795e803.
[22] H. Yano, F. Hirayama, M. Kamada, H. Arima, K. Uekama, Colon-specific
delivery of prednisolone-appended alpha-cyclodextrin conjugate: alleviation
of systemic side effect after oral administration, J. Control. Release. 79 (2002)
103e112.
[17] L. Dubuquoy, C. Rousseaux, X. Thuru, L. Peyrin-Biroulet, O. Romano,
P. Chavatte, M. Chamaillard, P. Desreumaux, PPARgamma as a new thera-
peutic target in inflammatory bowel diseases, Gut 55 (2006) 1341e1349.
[18] C. Rousseaux, B. Lefebvre, L. Dubuquoy, P. Lefebvre, O. Romano, J. Auwerx,
D. Metzger, W. Wahli, B. Desvergne, G.C. Naccari, P. Chavatte, A. Farce,
[23] M.W. Son, J.I. Ko, H.M. Doh, W.B. Kim, T.S. Park, M.J. Shim, B.K. Kim, Protective
effect of taurine on TNBS-induced inflammatory bowel disease in rats, Arch.
Pharm. Res. 21 (1998) 531e536.