Inhibition of colitis by ring-modified analogues of 6-acetamido-2,4,5-trimethylpyridin-3-ol

Article abstract of DOI:10.1016/j.bioorg.2020.104130

6-Aminopyridin-3-ol scaffold has shown an excellent anti-inflammatory bowel disease activity. Various analogues with the scaffold were synthesized in pursuit of the diversity of side chains tethering on the C(6)-position. Structure-activity relationship among the analogues was investigated to understand the effects of the side chains and their linkers on their anti-inflammatory activities. In this study, structural modification moved beyond side chains on the C(6)-position and reached to pyridine ring itself. It expedited us to synthesize diverse ring-modified analogues of a representative pyridine-3-ol, 6-acetamido-2,4,5-trimethylpyridin-3-ol (9). In the evaluation of compounds on their inhibitory actions against TNF-α-induced adhesion of monocytic cells to colonic epithelial cells, an in vitro model mimicking colon inflammation, the effects of compounds 9, 17, and 19 were greater than tofacitinib, an orally available anti-colitis drug, and compound 17 showed the greatest activity. In addition, TNF-α-induced angiogenesis, which permits more inflammatory cell migration into inflamed tissues, was significantly blocked by compounds 17 and 19 in a concentration-dependent manner. In the comparison of in vivo therapeutic effects of compounds 9, 17, and 19 on dextran sulfate sodium (DSS)-induced colitis in mice, compound 17 was the most potent and efficacious, and compound 19 was better than compound 9 which showed a similar degree of inhibitory effect to tofacitinib. Taken together, it seems that either the trimethyl system or the hydroxyl group on the pyridinol ring is essential to the activity. This finding might become a new milestone in the development of pyridinol-based anti-inflammatory bowel disease agents.

Full text of DOI:10.1016/j.bioorg.2020.104130

Journal Pre-proofs
Inhibition of colitis by ring-modified analogues of 6-acetamido-2,4,5-trime-
thylpyridin-3-ol
Chhabi Lal Chaudhary, Prakash Chaudhary, Sadan Dahal, Dawon Bae, Tae-
gyu Nam, Jung-Ae Kim, Byeong-Seon Jeong
PII:
S0045-2068(20)31427-9
DOI:
https://doi.org/10.1016/j.bioorg.2020.104130
YBIOO 104130
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
10 April 2020
2 July 2020
20 July 2020
Please cite this article as: C. Lal Chaudhary, P. Chaudhary, S. Dahal, D. Bae, T-g. Nam, J-A. Kim, B-S. Jeong,
Inhibition of colitis by ring-modified analogues of 6-acetamido-2,4,5-trimethylpyridin-3-ol, Bioorganic
Chemistry (2020), doi: https://doi.org/10.1016/j.bioorg.2020.104130
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
©
2020 Published by Elsevier Inc.

Title
Inhibition of colitis by ring-modified analogues of 6-acetamido-2,4,5-trimethylpyridin-3-ol
Author names and affiliations
a
a
a
a
b,
Chhabi Lal Chaudhary , Prakash Chaudhary , Sadan Dahal , Dawon Bae , Tae-gyu Nam *,
a,
a,
Jung-Ae Kim *, Byeong-Seon Jeong *
a
College of Pharmacy and Institute for Drug Research, Yeungnam University, Gyeongsan
3
8541, Republic of Korea
b
Department of Pharmacy and Institute of Pharmaceutical Science and Technology, Hanyang
University, Ansan, Gyeonggi-do 15588, Republic of Korea
Corresponding authors
Byeong-Seon Jeong: Tel.: +82 53 810 2814; Fax: +82 53 810 4654; E-mail: jeongb@ynu.ac.kr
Jung-Ae Kim: Tel.: +82 53 810 2816; Fax: +82 53 810 4654; E-mail: jakim@yu.ac.kr
Tae-gyu Nam: Tel.: +82 31 400 5807; Fax: +82 31 400 5958; E-mail: tnam@hanyang.ac.kr
Highlights

Ring-modified analogues of 6-acetamido-2,4,5-trimethylpyridin-3-ol were
synthesized


Structure-activity relationship of the ring-modified analogues were established
Compound 17 inhibited TNF-α-induced responses in colonic epithelial cells better
than tofacitinib

Anti-colitis efficacy of compound 17 was much greater than tofacitinib in mice

Abstract
6
-Aminopyridin-3-ol scaffold has shown an excellent anti-inflammatory bowel disease activity.
Various analogues with the scaffold were synthesized in pursuit of the diversity of side chains
tethering on the C(6)-position. Structure-activity relationship among the analogues was
investigated to understand the effects of the side chains and their linkers on their anti-
inflammatory activities. In this study, structural modification moved beyond side chains on the
C(6)-position and reached to pyridine ring itself. It expedited us to synthesize diverse ring-
modified analogues of a representative pyridine-3-ol, 6-acetamido-2,4,5-trimethylpyridin-3-ol
(9). In the evaluation of compounds on their inhibitory actions against TNF-α-induced adhesion
of monocytic cells to colonic epithelial cells, an in vitro model mimicking colon inflammation,
the effects of compounds 9, 17, and 19 were greater than tofacitinib, an orally available anti-
colitis drug, and compound 17 showed the greatest activity. In addition, TNF-α-induced
angiogenesis, which permits more inflammatory cell migration into inflamed tissues, was
significantly blocked by compounds 17 and 19 in a concentration-dependent manner. In the
comparison of in vivo therapeutic effects of compounds 9, 17, and 19 on dextran sulfate sodium
(DSS)-induced colitis in mice, compound 17 was the most potent and efficacious, and
compound 19 was better than compound 9 which showed a similar degree of inhibitory effect
to tofacitinib. Taken together, it seems that either the trimethyl system or the hydroxyl group
on the pyridinol ring is essential to the activity. This finding might become a new milestone in
the development of pyridinol-based anti-inflammatory bowel disease agents.

Graphical abstract
Keywords
Ring modification; 6-Aminopyridin-3-ol; Structure-activity relationship; TNF-α; Adhesion;
Angiogenesis; Inflammatory bowel disease

1
. Introduction
Inflammatory bowel disease (IBD) is a group of intestinal disorders that cause chronic
relapsing inflammation of the gastrointestinal tract, with major subtypes of Crohn's disease
(CD) and ulcerative colitis (UC) [1,2]. CD can occur throughout the gastrointestinal tract, from
the mouth to the anus, and affects all layers of the intestinal wall [3]. UC typically occurs in
the lower tract as the colon and rectum, and the area of injury usually begins in the rectum and
can spread further through the colon [4]. Common symptoms of IBD are relapsing cycles of
diarrhea, abdominal pain, rectal bleeding, bloody stools, weight loss, and fatigue. IBD,
although known as a low mortality disease, severely impairs patients' quality of life. IBD is
currently estimated to affect about 5 million people worldwide, including approximately 2.5
million people in Europe and 1.5 million in the United States, and is now rapidly spreading
around the world [5].
The exact etiology of IBD is still incompletely understood, but it is commonly known to be
caused by inappropriate immune responses to foreign organisms such as bacteria, viruses, or
antigens in the intestinal tract in people with a genetic predisposition [6]. Defects of the
intestinal mucosa and abnormalities of the innate immune system are considered to be the main
causes of IBD, which interact to activate acquired immune T cells [7]. One that exists at the
heart of this situation is the powerful inflammatory cytokine TNF-α (tumor necrosis factor-α).
In relation to inflammation, TNF-α acts on vascular endothelial cells to induce angiogenesis
[8] and promote more inflammatory cytokine secretion from macrophages [9]. In addition, it
induces the death of intestinal epithelial cells [10], and activates MMPs of myofibroblasts to
degrade the matrix [11] and promote the survival of Th17 cells [12]. Since the central role of
TNF-α has been identified, anti-TNF-α monoclonal antibodies such as infliximab, adalimumab,
and golimumab have been developed [13]. Due to the greater therapeutic efficacy of these
antibody drugs than oral drugs such as salicylates, immunosuppressants, and corticosteroids,

these biologics play a major role in the treatment of IBD. However, such therapies using anti-
TNF-α monoclonal antibodies are generally quite expensive and have the inconvenience of
administration which must be given intravenously or subcutaneously. In addition, treatment
failure, resistance, and side effects such as fluid reactions or infectious complications have been
reported in many patients [14]. Regrettably, a considerable number of patients may fail to
overcome IBD with anti-TNF-α biologics in the long run.
Consequently, there is an urgent need to develop new drugs that are more effective and safer
than conventional ones. In particular, the development of small-molecule drugs of easier
administration and lower cost compared to antibody biologics is also an important task to be
achieved at the moment. Recently, several new small-molecule drugs have been developed and
are currently being tested in phase II or III clinical trials [15–18]. The representative one is
tofacitinib, a JAK (Janus kinase) inhibitor [19]. Tofacitinib has been used to treat rheumatoid
arthritis since 2012. And it was approved for the treatment of adult patients in the United States
with moderately to severely active UC in 2018 [20]. Although it is currently in phase II clinical
trials for CD, it seems not effective for the treatment of CD [21,22].
CO H
2
HO
CO H
2
N
N
N
CN
HO
N
O
N
H
N
N
N
NH2
S
N
H
O O
Tofacitinib
Mesalazine
Sulfasalazine
A variety of molecular targets have been discussed due to the complex etiology and interactions
between various cells in the study of IBD drug discovery [23], hence a strategy for new small-
molecule drug discovery aiming only one target molecule may occasionally carry a risk of
failure. In view of this point, we have established a cell-based in vitro screening system as a
phenotype-based drug discovery strategy [24,25]. We set up an assay model that mimics tissue

damage caused by persistent influx of inflammatory cells such as monocytes, macrophages,
and lymphocytes which are induced by TNF-α, a key factor in IBD pathogenesis. One
characteristic occurrence observed in IBD is that inflammatory cells infiltrate and adhere to
colon epithelium, which is induced by chiefly TNF-α. Treatment of colon epithelial cells with
TNF-α leads to activation of NF-κB (nuclear factor κB), the most well-known transcription
factor for up-regulating chemokines that recruit circulating leukocytes to the site of
inflammation and adhesion molecules [26,27]. In the TNF-α-induced monocyte-epithelial cell
adhesion assay system for quantification of inhibitory activity of test compounds against the
adhesion process, colon epithelial cells (HT-29 cells) and monocytic cells (U937 cells) were
co-cultured in the presence of TNF-α, and cell adhesion level was determined by measuring
fluorescence signal subsequent to washing off unattached cells.
G =
HO
3
O
O
R2
N
R
O
R
N 6
G
H
O
N
H
4
O
R
R1
6
-G-2,4,5-trimethylpyridin-3-ol
1
2
3
HO
S
O O
O
S
R
R
N
H
R
N
H
N
N
H
N
N
H
N
N
H
H
H
5
6
7
8
9
With this phenotype-based strategy using the in vitro assay as the primary screening system,
we have been studying on the discovery of novel promising small-molecule compounds for the
treatment of IBD for over a decade [28–35]. 2,4,5-Trimethylpyridin-3-ol (1) is the common
skeleton of the compounds which we have been interested in and dealt with for this study so
far [30,33,35–46]. We have prepared hundreds of its derivatives mostly by introducing several
functional groups with various substituents at its C(6)-position, and are still designing and
synthesizing new derivatives. From this 2,4,5-trimethylpyridin-3-ol library, we could pick out

some promising compounds that showed excellent efficacy against angiogenesis, cancers, or
colitis, and so forth. Regarding IBD, we reported our findings in a series of papers to date on
the 2,4,5-trimethylpyridin-3-ol derivatives with amino- (2), alkoxy- (3), carbamato- (4),
sulfonamido- (5), thioureido- (6), or ureido- (7) functional groups [30,33,35]. We observed that
these groups of compounds showed very potent inhibitory activity from the in vitro assays.
Compared to mesalazine which is the active metabolite of the sulfasalazine (SSZ), a clinical
drug for the treatment of IBD, they showed up to nearly 100,000 times better inhibition based
on the IC values. Some compounds selected from the in vitro assay were subjected to in vivo
5
0
experiments using acute colitis rat models, and all of the tested compounds administered were
far superior to the positive control, SSZ.
Inflammatory cells including macrophages and mast cells can activate endothelial cells and
proinflammatory cytokines produced from the cells such as TNF-α can induce angiogenesis,
which recruits more inflammatory cells and worsens tissue damage [47]. Thalidomide, one of
the old anti-TNF drugs, demonstrated to be effective in refractory pediatric IBD patients, and
its effectiveness is known to be mediated through antiinflammatory and antiangiogenic effects
[48]. Based on these reports, we have hypothesized that 6-amido-2,4,5-trimethylpyridin-3-ols
(8) which were reported to show good antiangiogenic effects should have anti-IBD activity.
Indeed, various amido compounds 8 clearly showed the activity (unpublished results). While
those analogues have mainly featured structural variation on the C(6)-position, in this study,
we aimed to investigate the effect of structural changes in the 2,4,5-trimethylpyridin-3-ol ring
itself on the activity. For this purpose, we needed to fix C(6)-tethering group as acetamido
group in order to reduce variables and grab clear idea on the structure-activity relationship
(SAR). Compound 9 (6-acetamido-2,4,5-trimethylpyridin-3-ol) was considered a suitable
reference for this study. Not only the acetamido group is the simplest amido-containing
structure to provide us with convenient synthetic approaches but also acetamido-containing

compound 9 has an intermediate level of activity to afford us enough activity margin to explore
a wider range of variation in activities that can be expected from new modifications.
2
2
. Results and Discussion
.1. Chemistry
To investigate the effect of pyridine ring on the activity, we have conducted ring modification
in several different directions (Figure 1). The C(3)-OH group is either removed or blocked or
replaced with other functional groups (compounds 10, 12, 17, and 18). Electron density of the
pyridine ring was controlled as either N-oxide or pyrimidine/phenyl replacement (compounds
1
1, 15, and 16). Trimethyl group on pyridine ring were either modified to hydroxymethyl group
or removed (compounds 13, 14, and 19). Detrimethyl analogue (compound 19) was further
modified to C(3)-F analogue or phenyl analogue (compounds 20 and 21).

HO
Ring modification
O
O
G
Ar
N
N
N
H
H
9
10~21
O
HO
MeO
O
O
O
O
N
N
N
N
N
N
H
H
H
O
10
11
12
HO
HO
OH
OH
HO
HO
O
O
N
O
N
N
H
N
N
H
N
N
H
13
14
15
HO
HO
Br
O
O
O
O
O
N
H
N
N
H
N
N
H
1
1
6
17
18
F
HO
O
N
N
H
N
N
H
N
H
9
20
21
Figure 1.
As shown in Scheme 1, we synthesized the compounds 9 and 10 using the key intermediate 29,
which was obtained by the well-known route developed by us [38]. The synthesis started from
inexpensive starting materials, pyridoxine hydrochloride (vitamin B , 22). Firstly, the primary
6
alcohols on 22 were converted to chlorines using a catalytic amount of DMF in SOCl2.
Secondly, the chlorines were reductively cleaved using Zn dust in acetic acid to obtain

trimethyl compound 24. Then, the bromine at C(6)-position on 25 was introduced by
electrophilic aromatic bromination reaction of 24 using 1,3-dibromo-5,5-dimethylhydantoin
(DBDMH) and then the phenolic alcohol at C(3)-position was protected by benzyl bromide to
give 26. Next, the bromide of 26 was substituted by Buchwald-Hartwig amination with
benzophenone imine to give imino-compound 27 followed by cleavage of the imine with
methanolic HCl to give amino-compound 28. Finally, the amino group in 28 was acetylated
with acetyl chloride to afford 29. The benzyl group of the intermediate 29 was deprotected to
afford the compound 9 which was then acetylated again to afford methyl ester 10.
HO
HO
Cl
OH
Cl
HO
HO
c
a
b
N
HCl
N
HCl
N
22
23
24
HO
BnO
BnO
d
e
Ph
Ph
N
Br
N
Br
N
N
25
26
27
BnO
BnO
f
g
h
O
N
NH2
N
N
H
28
29
HO
O
i
O
O
O
N
N
N
N
H
H
9
10
Scheme 1. Reagents and Conditions: (a) DMF, SOCl , 80 ℃, 4 h, 93%; (b) Zn, AcOH, 120 ℃,
2
4
h, 92%; (c) DBDMH, THF, r.t., 3 h, 70%; (d) PhCH Br, K CO , DMF, r.t., 12 h, 97%; (e)
2 2 3
t
Ph C=NH, Pd (dba) , BINAP, NaO Bu, PhMe, reflux, 6 h, 98%; (f) HCl/MeOH, THF, r.t., 1
2
2
3

h, 90%; (g) AcCl, NaOAc, Me CO, r.t., 4 h, 94%; (h) BCl , DCM, 0 ℃ to r.t., 2 h, 97%; (i)
2
3
AcCl, NaOAc, Me CO, r.t., 3 h, 79%.
2
The direct synthesis of N-oxide analogue 11 from the compound 9 using m-CPBA or hydrogen
peroxide was unsuccessful. Thus, the intermediate 29 was used as a starting substrate, as shown
in Scheme 2. The intermediate 29 was treated with m-CPBA to give N-oxide compound 30 and
was debenzylated by hydrogenolysis to afford compound 11. Methoxy analogue 12 was
synthesized via compound 33. Previously, 3-methoxy-6-bromo-compound 31 was obtained by
methylation of aromatic ‒OH of the 6-bromo compound 25 using iodomethane. Then, the
amino compound 33 was obtained by Buchwald-Hartwig amination, followed by acidic
methanolysis of the imine intermediate 32, in two step sequences. The compound 33 was then
acetylated to afford the acetamido compound 12 (Scheme 2).
BnO
BnO
HO
a
b
O
O
O
N
N
N
N
N
N
H
H
H
O
O
29
30
11
HO
MeO
MeO
c
d
Ph
Ph
N
Br
N
Br
N
N
25
31
32
MeO
MeO
e
f
O
N
NH2
N
N
H
33
12
Scheme 2. Reagents and Conditions: (a) m-CPBA, CHCl , r.t., 30 min, 97%; (b) H , Pd/C,
3
2

MeOH, r.t., 1 h, 80%; (c) MeI, K CO , THF-H O (1:1), r.t., 21 h, 83%; (d) Ph C=NH, Pd (dba) ,
2
3
2
2
2
3
t
BINAP, NaO Bu, PhMe, reflux, 6 h, 96%; (e) HCl/MeOH, THF-MeOH (1:1) , r.t., 2 h, 78%;
(
f) AcCl, Et N, DCM, r.t., 5 h, 73%.
3
The synthesis of hydroxylated analogues, 13 and 14, are given in Scheme 3. 6-
Aminopyridoxine hydrochloride (35) was synthesized by the reported procedure [49].
Pyridoxine hydrochloride (22) was reacted with diazotized aniline to afford 6-
phenylazopyridoxine hydrochloride (34), which was then hydrogenated to yield amino
compound 35. Treatment of 35 with excess acetic anhydride in the presence of a catalytic
amount of sulfuric acid produced the acetamide 13. 6-Amino-4-dehydroxypyridoxine
hydrochloride (38) was synthesized by the method reported by us [36,50]. Pyridoxine
hydrochloride (22) was refluxed with Zn dust in acetic acid, followed by HCl in refluxing
methanol gave 4'-dehydroxypyridoxine hydrochloride (36). The introduction of nitrogen
moiety at C(6)-position of 36 was done by azo-coupling reaction with diazotized aniline to give
azo-compound 37. Azo bond was then cleaved by catalytic hydrogenolysis to afford amino-
compound 38. Finally, N-acetylation was carried out under the same reaction conditions as in
the synthesis of 13 to obtain 14.

HO
HO
HO
HO
HO
HO
OH
OH
OH
a
b
N
HCl
N
HCl
N N Ph
N
HCl
NH2
22
34
35
HO
HO
OH
c
O
N
N
H
13
HO
HO
OH
OH
OH
HO
HO
d
e
N
HCl
N
HCl
N
HCl
N N Ph
22
36
37
OH
OH
HO
HO
f
g
O
N
HCl
NH2
N
N
H
38
14
Scheme 3. Reagents and Conditions: (a) PhNH , NaNO , HCl, NaOH, H O, 0 ℃ to r.t., 1 h,
2
2
2
8
2%; (b) H , Pd/C, MeOH, r.t., 6 h, 45%; (c) Ac O, H SO , AcOH, 100 ℃, 30 min, 88%; (d)
2
2
2
4
i) Zn, AcOH, 120 ℃, 3 h, ii) HCl/MeOH, MeOH, 60 ℃, 1 h, 93%; (e) PhNH , NaNO , HCl,
2
2
NaOH, H O, 0 ℃ to r.t., 1 h, 95%; (f) H , Pd/C, MeOH, r.t., 6 h, 90%; (g) Ac O, H SO , AcOH,
2
2
2
2
4
1
00 ℃, 30 min, 48%.
Scheme 4 shows the synthesis of pyrimidine analogue 15 of the compound 9. 5-Benzyloxy-
,6-dimethylpyrimidin-2-amine (43) was obtained by slight modification of the known
procedures [51,52]. 2,4-Dioxopentan-3-yl acetate (40) was synthesized by replacement of
4

chloride in the compound 39 by acetate using a microwave reactor. 2-Amino-5-
hydroxypyrimidine ring was formed by condensation reaction of 40 with guanidine (41) to give
4
2 which was further treated with benzyl chloride to get benzyloxy-compound 43. Next, the 2-
amino group was reacted with acetyl chloride to afford acetamido-compound 44. Final
debenzylation using BCl provided the pyrimidine analogue 15.
3
NH
Cl
a
AcO
b
O
O
+
H N
2
NH2
O
O
39
40
41
BnO
HO
c
BnO
d
N
O
N
N
N
N
N
NH2
HO
N
NH2
H
42
43
44
e
N
O
N
N
H
15
Scheme 4. Reagents and Conditions: (a) NaOAc, DMSO, MW, 50 ℃, 20 min; (b) NaOAc,
DMSO, MW, 80 ℃, 30 min, 40% (2 steps); (c) PhCH Cl, NaOH, EtOH-DMF, 80 ℃, 12 h,
2
7
6%; (d) AcCl, Et N, DCM, r.t., 3 h, 39%, (e) BCl , C HMe , DCM, r.t., 30 min, 49%.
3 3 6 5
The phenyl analogue 16 of the compound 9 was prepared starting from a hydroquinone 45 as
shown in Scheme 5. 2,3,5-Trimethylbenzene-1,4-diol (45) was oxidized to 1,4-benzoquinone
4
6 and then converted to mono-oxime compound 47 using hydroxylamine·HCl. Mono-oxime
formation has been known to be selective to the steric environment of carbonyl moiety in
quinone system [53]. Of course, in the compound 46, the less hindered carbonyl was the

primary site for the reaction. Reduction of the compound 47 with SnCl gave p-aminophenol
2
compound 48 which was finally treated with acetyl chloride to afford 16.
HO
O
O
a
b
OH
O
N OH
45
46
47
HO
HO
c
d
O
NH HCl
N
2
H
48
16
Scheme 5. Reagents and Conditions: (a) PhI(OAc) , MeOH, r.t., 1 h; (b) NH OH•HCl, THF,
2
2
reflux, 2 d, 72% (2 steps); (c) SnCl , HCl, DCM, r.t., 2 h, 32%; (d) AcCl, NaOAc, Me CO, r.t.,
2
2
3
h, 41%.
The synthesis of dehydroxylated analogue 17 of the compound 9 was shown in Scheme 6. This
synthetic strategy was designed based on the known procedure on phenolic C‒O bond cleavage
[54]. The 6-bromopyridin-3-ol 25 was taken as the starting material and reacted with 5-chloro-
1
-phenyltetrazole (49), the key reagent for this synthesis, to give 50. Amino group was
introduced by the two sequential reactions; imination under the Buchwald-Hartwig conditions
to give 51, and cleavage of the imine under acidic methanolysis conditions to give 52. Then,
hydrogenolysis of the phenolic C‒O bond in 52 gave the deoxygenated product 53 which was
then acetylated to afford the final compound 17, the deoxy-analogue of 9. For the 3-bromo
analogue 18, two-step procedure was needed from the intermediate 53; bromination and
acetylation.

Ph
N
N
Ph
HO
O
N
a
b
N
N
+
Cl
N
O
N
N
Br
N
N
Br
25
49
50
Ph
Ph
N
O
N
c
d
N
N
N
Ph
Ph
N
N
N
N
N
N
NH2
5
1
52
e
O
N
NH2
N
N
H
53
17
f
Br
g
Br
O
N
NH2
N
NH2
N
N
H
53
54
18
Scheme 6. Reagents and Conditions: (a) NaOH, MeCN, MW, 200W, 100 ℃, 1 h, 92%; (b)
t
Ph C=NH, Pd (dba) , BINAP, NaO Bu, PhMe, reflux, 3 h; (c) HCl/MeOH, MeOH-THF (1:1),
2
2
3
r.t., 3 h, 70% (2 steps); (d) H , Pd/C, MeOH, r.t., 3 d, 51%; (e) AcCl, NaOAc, Me CO, r.t., 4
2
2
h, 25%; (f) NBS, NH OAc, MeCN, r.t., 10 min, 41%; (g) AcCl, Et N, Me CO, r.t., 24 h, 47%.
4
3
2
For the demethylated analogue 19 of the compound 9, synthesis started from 3-
hydroxypyridine (25) which was treated with diazotized p-nitroaniline (55) to give azo-
compound 56. After cleavage of azo bond affording 57, acetylation finally gave 19 [55].
Similarly, commercially available 5-fluoropyridin-2-amine (58) was simply acetylated to give
3
-fluoro-demethylated analogue 20. The demethyl-phenyl analogue 21 (paracetamol) was
purchased. (Scheme 7)

HO
H N
2
HO
a
+
N
NO2
N
N N
NO2
25
55
56
HO
HO
b
c
O
N
HCl
NH2
N
N
H
57
19
F
F
d
O
N
NH2
N
N
H
58
20
HO
O
N
H
21 (purchased)
Scheme 7. Reagents and Conditions: (a) NaNO , HCl, NaOH, H O, r.t., 1.5 h, 96%; (b) H ,
2
2
2
Pd/C, MeOH, r.t., 12 h, 64%; (c) i) AcCl, Et N, DCM, r.t., 3 h, ii) NaOH, H O, r.t., 30 min,
3
2
3
0%; (d) AcCl, NaOAc, Me CO, r.t., 30 min, 38%.
2
2
.2. Biology
Inhibitory activity of the compounds against TNF-α-induced cell adhesion is summarized in
Table 1. Compound 9 is the parent compound on which ring modification was conducted to
prepare the analogues 10–21. First, we investigated the effect of modification in electron
density of the pyridine ring on the inhibitory activity. A parent compound 9 showed 56%
inhibition which is almost similar to tofacitinib, a positive control and JAK/STAT inhibitor,

and tremendously higher than mesalazine, an approved drug for IBD. Its direct phenyl analogue
6 significantly lost the activity indicating that the absence of nitrogen atom in pyridine ring
1
leading to increase some degree of electron density, decreased adhesion-inhibitory activity.
However, decrease in electron density of the pyridine ring of 9 in a form of N-oxide 11 and
pyrimidine 15 also resulted in decrease of the biological activity. Such discrepancy between
electron density and the activity imply that other factors, e.g., steric factor also contribute the
activity. Removal of the trimethyl groups on pyridine ring of 9 resulted in astonishing increase
in the adhesion-inhibitory activity (69.5%, 19). Further removal of nitrogen from pyridine ring
decreased the activity (21.4%, 21, paracetamol, an antipyretic analgesic drug). Such decrease
by 48.1% (19 vs. 21) was comparable to 44.6% decrease (9 vs. 16) which are commonly
reflected by the absence of pyridine ring nitrogen. As for steric factors, shrunk steric bulkiness
in 19, compared to 9, significantly increased the activity while removal of ring nitrogen as in
1
6 resulted in dramatic reduction in the activity, compared to 9. Modification of trimethyl
groups into hydroxymethyl groups afforded two analogues, 13 and 14. The former has
hydroxymethyl groups both on C(3)- and C(4)-positions to cause quite a decrease in activity,
compared to 9, while installation of hydroxymethyl group only in C(3)-position (14) restored
the activity to almost comparable level to 9. Next, role of C(3)-OH group on the activity was
investigated. Acetyl capping on C(3)-OH group of 9 afforded the slight loss of the activity (10).
Both methyl capping and the replacement of C(3)-OH group with bromide resulted in almost
the same degree of decrease (12 and 18, respectively). Surprisingly, removal of C(3)-OH group
afforded increase in the activity to considerable extent (77.6%, 17). It is indicated that either
an intact hydroxy group or a hydrogen should be attached on C(3)-position for the maximum
activity. Besides, replacement of C(3)-OH group of 19 which showed a high adhesion-
inhibitory activity (69.5%) with fluoride (20) again resulted in decrease in the activity. It is
noteworthy that 17 (dehydroxylated analogue of 9) and 19 (detrimethylated version of 9)

showed comparable activities (77.6% and 69.5%, respectively) which are higher than 9. Taken
together, it is suggested that either C(3)-OH group or trimethyl group is essential for the activity
individually and both functional groups decreased the activity a bit when installed together.
Mechanistically, cell adhesion assay in this report is not based on a specific target but only
observed TNF-α-induced adhesion as a phenotypic experiment. It is unclear at this point if our
compounds inhibited the interaction between TNF-α and its receptor on the cell surface or
penetrated cell membrane to inhibit cell adhesion via intracellular mechanism. One of the
reliable predictions for cell permeability is the Lipinski’s rule of five for drug-likeness. It
indicates that logP should not exceed 5 in order not to be too lipophilic. Calculated logP (cLogP)
values of test compounds range from –1.098 to 1.728 except for 2.267 for compound 18.
Therefore they conform to the Lipinski’s rule and can be expected to possibly cross cell
membrane. On the other hand, their low cLogP values also indicate that they are quite polar
and their permeability will be limited to some extent.
Table 1. Inhibitory activity against TNF-α-induced adhesion of human monocytic cells (U937)
to human colonic epithelial cells (HT-29).
Compound
Concentration
1 μM
% Inhibitiona
56.0 ± 6.3*
39.5 ± 2.8*
21.9 ± 4.9*
34.7 ± 6.5*
16.0 ± 2.3*
51.8 ± 4.3*
28.8 ± 4.2*
11.4 ± 5.8
IC50
9
0.70 ± 0.10 μM
1
1
1
1
1
1
1
0
1
2
3
4
5
6
1 μM
n/ab
n/ab
n/ab
n/ab
n/ab
n/ab
n/ab
1 μM
1 μM
1 μM
1 μM
1 μM
1 μM

1
1
1
2
2
7
1 μM
77.6 ± 4.7*
37.2 ± 5.5*
69.5 ± 4.3*
20.3 ± 7.0
21.4 ± 5.3
50.8 ± 3.9*
37.3 ± 3.3*
0.32 ± 0.10 μM
n/ab
8
1 μM
9
1 μM
0.48 ± 0.14 μM
n/ab
0
1 μM
1 (Paracetamol)
1 μM
n/ab
Tofacitinib
Mesalazine
1 μM
0.71 ± 0.10 μM
21.34 ± 2.28 mM
10,000 μM
a
Data are shown as 'mean ± SEM' of at least three independent experiments performed in
triplicate.
*
P <0.05 versus vehicle-treated group.
b
n/a: not applicable
To determine cytotoxicity of compounds 9, 17, and 19, we compared the effects of the
compounds, tofacitinib and mesalazine on the viability of CCD-841 normal human colon
epithelial cells. Tofacitinib significantly decreased the viability of CCD-841 in a concentration-
dependent manner (Figure 2). However, compounds 17 and 19 were much less cytotoxic than
tofacitinib, while mesalazine was much safer than 17 and 19 in CCD-841 cells (Figure 2).

Figure 2. Cytotoxic effects of the selected compounds (9, 17 and 19), tofacitinib, and
mesalazine on the viability of CCD-841 normal human colon epithelial cells. Results are
presented as the mean ± SEM of three independent experiments. *P < 0.05 versus vehicle-
#
&
treated controls. P < 0.05 versus tofacitinib-treated group. P < 0.05 versus mesalazine-treated
group.
Although the effect of TNF-α on angiogenesis is controversial, temporal expression of TNF-α
is important for induction of angiogenesis. Continuously exposed to TNF-α, endothelial cells
are sprouting and undergo angiogenesis [56], which resembles the relapsing event of
inflammation in IBD condition. Based on these notion, we examined inhibitory actions of the
effective compounds on TNF-α-induced angiogenesis. In an in vivo angiogenesis assay using
chick chorioallantoic membrane (CAM) in which newly formed vessel branch points were
counted, an angiogenic ability of TNF-α was comparable to that of vascular endothelial growth
factor (VEGF) (Figure 3). Compounds 17 and 19 significantly inhibited TNF-α-induced
angiogenesis in a concentration-dependent manner up to the vehicle-treated controls (Figure
3
).

Figure 3. Inhibitory effects of the selected compounds (17 and 19) on TNF-α-induced
angiogenesis of CAM as an in vivo model (n = 8).

Figure 4. Compounds 9, 17 and 19 ameliorates the clinical features of DSS-induced colitis in
mice (n = 5). Colitis was induced by oral administration of DSS dissolved in drinking water.
Data represent the mean ± SEM for five mice per group. *P < 0.05 versus vehicle-treated
#
control group. P < 0.05 versus vehicle-treated DSS group.
Next, we examined ameliorating effects of compounds 17 and 19 together with 9 on dextran
sulfate sodium (DSS)-induced colitis in mice. To compare therapeutic effects, compounds were
administered after mice were given DSS treatment over five consecutive days. DSS treatment
resulted in the development of colitis assessed by decrease in body weight (Figure 4a) and
increases in colon-weight/unit-length (Figure 4b) and levels of myeloperoxidase (MPO) levels,
a biochemical marker of colon tissue inflammation (Figure 4c). Because the purpose of this
study is to confirm the efficacy in vivo, the test compounds were administered intraperitoneally
(IP) to bypass the first-pass effect accompanied by oral administration (PO). Although
tofacitinib is an orally available drug, it was administered both IP and PO as a positive control.
Compound 17 with the best in vitro effect which was administered intraperitoneally at two
doses, 3 and 10 mg/kg significantly blocked colitis and showed a recovery rate of 93% at 10
mg/kg dose. At 10 mg/kg dose, 19 was less effective than 17, but much greater than tofacitinib
and 9. Anti-colitis effects of all compounds were greater than sulfasalazine (300 mg/kg) which
was orally administered to mice. When DSS (2w/v%) treatment was extended one more day
(total 7 days), body weight reduction was much greater than 6-day treatment, indicating 7-day
DSS treatment induced much more severe colitis in mice (Figure 4a vs. Figure 5a). The mice
treated with DSS alone became very weak resulting in more than half (67%) of mice not
surviving by the end point. Administration of SSZ (300 mg/kg, PO), tofacitinib (10 mg/kg, IP
and 30 mg/kg, PO), and compound 9 (10 mg/kg, IP) did not restore body weight significantly
(Figure 5a). Particularly, treatment with tofacitinib (IP, 10 mg/kg) caused DSS-induced death

at earlier time point than the vehicle-treated group (Figure 5b), and tofacitinib PO (30 mg/kg)
administration group showed a higher DSS-induced mortality rate than the IP (10 mg/kg)
administration group. However, compound 17 significantly recovered body weight at 4 days
after administration, and the recovery effect was dose-dependent (Figure 5a).
Figure 5. Compound 17 ameliorates severe murine colitis induced by DSS (n = 6). Severe
colitis was induced by oral administration of DSS (2 w/v%) dissolved in drinking water for 7
days. Data represent the mean ± SEM for six mice per group. *P < 0.05 versus vehicle-treated
#
normal control group. P < 0.05 compared to vehicle-treated DSS group.
3
. Conclusion

Various ring-modified analogues of N-(5-hydroxy-3,4,6-trimethylpyridin-2-yl)acetamide (9)
were synthesized in order to investigate the effects of ring structure on the anti-IBD activity.
The results showed that either the trimethyl group or the hydroxyl group on the pyridine ring
seem to play a critical role in the biological activities based on in vitro and in vivo models. The
results that the dehydroxy analogue (17) and the detrimethyl analogue (19) are stronger than
the parent compound (9) might provide an invaluable piece of information for the next
generation anti-IBD agents. It also would be of great interest to investigate if the trend is valid
in a side chain independent manner.
4
4
. Experimental
.1. Chemistry
General
Unless noted otherwise, materials were purchased from commercial suppliers and used without
further purification. Air or moisture-sensitive reactions were carried out under an inert gas
atmosphere. The reaction progress was monitored by thin layer-chromatography (TLC) using
silica gel F254 plates. The products were purified by flash column chromatography using silica
gel 60 (70–230 mesh) or by using the Biotage ‘Isolera One’ system with indicated solvents.
Melting points were determined using a Fisher–Johns melting point apparatus and were not
1
corrected. NMR spectra were obtained using a Bruker-250 spectrometer 250 MHz for H-NMR,
1
3
and 62.5 MHz for C-NMR. Chemical shifts (δ) were expressed in ppm using a solvent as an
internal standard and the coupling constant (J) in hertz. Low-resolution mass spectra (LRMS)
were obtained using an Advion Expression CMS and recorded in a positive ion mode with an
electrospray (ESI) source. High-resolution mass spectra (HRMS) were obtained using a

Finnigan LTQ Orbitrap mass spectrometer (Thermo Fisher Scientific Inc, MA, USA) operated
in positive-ion electrospray mode.
6
-Acetamido-2,4,5-trimethylpyridin-3-yl acetate (10)
To a suspension of N-(5-hydroxy-3,4,6-trimethylpyridin-2-yl)acetamide hydrochloride (9, 88
mg, 0.38 mmol) in acetone (2 mL) was added sodium acetate (110 mg, 1.15 mmol) and acetyl
chloride (41 µL, 0.57 mmol) dropwise then stirred at room temperature for 3 h. The reaction
mixture was diluted with EtOAc and H O, the aqueous layer was extracted with EtOAc. The
2
combined EtOAc layer was dried over MgSO , filtered and, concentrated. The concentrated
4
residue then purified by silica gel column chromatography (DCM/MeOH = 20/1) to afford 10
(
71 mg, 79%). Pale yellow solid; TLC R 0.43 (DCM/MeOH = 9/1); m.p. 165 ℃; LRMS (ESI)
f
+
1
m/z 237 [M+H] ; H NMR (CDCl ) δ 8.62 (s, 1H), 2.35 (s, 3H), 2.28 (s, 3H), 2.16 (s, 3H), 2.14
3
1
3
(
s, 3H), 2.10 (s, 3H); C NMR (CDCl ) δ 168.65 (2C), 146.71 (2C), 145.93, 143.34, 141.09,
3
+
2
2
3.48, 20.60, 18.94, 15.11, 13.47; HRMS (ESI) m/z calculated for C H N O [M+H]+
1
2
17
2
3
37.1234, found 237.1236.
2
-Acetamido-5-(benzyloxy)-3,4,6-trimethylpyridine 1-oxide (30)
To a solution of N-(5-(benzyloxy)-3,4,6-trimethylpyridin-2-yl)acetamide (29, 145 mg, 0.51
mmol) in CHCl (5 mL) was added m-CPBA (106 mg, 0.61 mmol) and stirred at room
3
temperature for 30 minutes. The reaction mixture was quenched with sodium sulfite solution,
stirred at room temperature for additional 30 minutes and extracted with CHCl . The combined
3
CHCl layer was washed with sodium carbonate solution then dried over MgSO , filtered and,
3
4
concentrated. The concentrated crude was purified by silica gel column chromatography
(hexanes/EtOAc = 1/1 to EtOAc only) to afford 30 (179 mg, 97%). White solid; TLC R 0.48
f
+
1
(
DCM/MeOH = 9/1); m.p. 162 ℃; LRMS (ESI) m/z 301 [M+H] ; H NMR (CDCl ) δ 9.01 (s,
3

1
3
1
H), 7.56–7.34 (m, 5H), 4.79 (s, 2H), 2.48 (s, 3H), 2.29 (s, 3H), 2.26 (s, 3H), 2.19 (s, 3H); C
NMR (CDCl ) δ 169.44, 150.07, 141.17, 139.51, 136.01, 133.46, 128.93 (2C), 128.82, 128.21
3
+
(2C), 127.80, 76.13, 24.06, 16.37, 13.20, 12.30; HRMS (ESI) m/z calculated for C H N O
17 21 2 3
+
[
M+H] 301.1547, found 301.1536.
2
-Acetamido-5-hydroxy-3,4,6-trimethylpyridine 1-oxide (11)
To a suspension of 2-acetamido-5-(benzyloxy)-3,4,6-trimethylpyridine 1-oxide (30, 50 mg,
.17 mmol) in MeOH (3 mL) was added Pd/C (palladium, 10 wt.% on activated carbon, 10
0
mg) then stirred under hydrogen atmosphere at room temperature for 1 h. The reaction mixture
was filtered through a pad of Celite and concentrated. The concentrated crude was purified by
silica gel column chromatography (DCM/MeOH = 20/1) to afford 11 (28 mg, 80%). White
+
1
solid; TLC R 0.24 (DCM/MeOH = 9/1); m.p. 206 ℃; LRMS (ESI) m/z 211 [M+H] ; H NMR
f
1
3
(
DMSO-d ) δ 9.64 (s, 1H), 2.31 (s, 3H), 2.12 (s, 3H), 2.04 (s, 3H), 2.01 (s, 3H); C NMR
6
(
DMSO-d ) δ 169.11, 148.50, 135.34, 135.01, 128.27, 123.21, 22.62, 14.95, 12.40, 11.67;
6
+
+
HRMS (ESI) m/z calculated for C H N O [M+H] 211.1077, found 211.1078.
1
0
15
2
3
2
-Bromo-5-methoxy-3,4,6-trimethylpyridine (31)
To a suspension of 6-bromo-2,4,5-trimethylpyridin-3-ol (25, 200 mg, 0.93 mmol) in mixed
THF-H O (1:1, 5 mL) solvent was added K CO (1.3 g, 9.26 mmol) and iodomethane (288 µL,
2
2
3
4
.63 mmol) at room temperature then stirred for 21 h. The reaction mixture was diluted with
DCM and H O, the aqueous layer was extracted with DCM. The combined DCM layer was
2
dried over MgSO , filtered and concentrated. The concentrated crude was purified by silica gel
4
column chromatography (DCM only) to afford 31 (176 mg, 83%). Pale yellow solid; TLC Rf
+
1
0
3
.40 (DCM/MeOH = 100/1); m.p. 38 ℃; LRMS (ESI) m/z 230 [M+H] ; H NMR (CDCl ) δ
3
1
3
.69 (s, 3H), 2.45 (s, 3H), 2.31 (s, 3H), 2.26 (s, 3H); C NMR (CDCl ) δ 152.90, 150.20,
3

1
41.42, 138.01 , 132.08, 60.45, 19.00, 18.90, 13.62; HRMS (ESI) m/z calculated for
+
+
C H BrNO [M+H] 230.0175, found 230.0181.
9
13
N-(5-Methoxy-3,4,6-trimethylpyridin-2-yl)-1,1-diphenylmethanimine (32)
To a mixture of 2-bromo-5-methoxy-3,4,6-trimethylpyridine (31, 160 mg, 0.70 mmol) in
toluene (5 mL) was added benzophenone imine (128 µL, 0.76 mmol),
tris(dibenzylideneacetone)dipalladium [Pd (dba) ] (8 mg, 0.01 mmol), (R)-(+)-2,2'-
2
3
bis(diphenylphosphino)-1,1'-binaphthalene (9 mg, 0.014 mmol) and sodium tert-butoxide (100
mg, 1.04 mmol) at room temperature and refluxed for 6 h. The reaction mixture was diluted
with EtOAc and washed with brine. The EtOAc layer was dried over MgSO , filtered and
4
concentrated. The concentrated crude was purified by silica gel column chromatography
(
(
(
(
Hex/EtOAc = 5/1) to afford 32 (221 mg, 96%). Pale yellow solid; TLC R 0.28
f
+
1
hexanes/EtOAc = 4/1); m.p. 101 ℃; LRMS (ESI) m/z 331 [M+H] ; H NMR (CDCl ) δ 7.80
3
d, J = 7.3 Hz, 2H), 7.50–7.31 (m, 1H), 7.25–7.11 (m, 5H), 3.61 (dd, J = 2.2, 0.9 Hz, 3H), 2.31
1
3
s, 3H), 2.09 (s, 3H), 1.92 (s, 3H); C NMR (CDCl ) δ 169.48, 157.07, 149.43, 146.91, 139.68,
3
1
39.35, 137.02, 130.92, 129.71 (2C), 129.01 (2C), 128.67, 128.10 (2C), 127.64 (2C), 119.69,
+
+
6
0.47, 18.85, 13.90, 12.48; HRMS (ESI) m/z calculated for C H N O [M+H] 331.1805,
22 23 2
found 331.1818.
5
-Methoxy-3,4,6-trimethylpyridin-2-amine (33)
To a mixture of N-(5-methoxy-3,4,6-trimethylpyridin-2-yl)-1,1-diphenylmethanimine (32, 100
mg, 0.30 mmol) in THF-MeOH (1:1, 5 mL) was added HCl in MeOH (2 M, 1.5 mL) dropwise
and stirred at room temperature for 2 h. The reaction mixture was concentrated then the
concentrated crude was diluted with EtOAc and saturated NaHCO solution. The aqueous layer
3
was extracted with EtOAc, the combined EtOAc layer dried over MgSO filtered and
4
,