Design, synthesis and biological evaluation of novel imidazopyridines as potential antidiabetic GSK3β inhibitors

Article abstract of DOI:10.1016/j.bmcl.2012.05.060

Design, synthesis and biological evaluation of the imidazopyridine analogs as novel GSK3β inhibitors for treatment of type 2 diabetes mellitus are described. Most of the analogs exhibited excellent inhibitory activities (IC50 < 44 nM) against glycogen synthase kinase 3β (GSK3β). The structure-activity relationship (SAR) of the imidazopyridine analogs and the binding mode of analog 23 in the catalytic domain of GSK3β, based on our X-ray crystallography study, are described. In particular, analog 28, which was selected as a potential drug candidate for treatment of type 2 diabetes mellitus, exhibited excellent GSK3β inhibition, pharmacokinetic profiles and blood glucose lowering effect in mouse.

Full text of DOI:10.1016/j.bmcl.2012.05.060

Bioorganic & Medicinal Chemistry Letters 22 (2012) 4221–4224
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and biological evaluation of novel imidazopyridines as
potential antidiabetic GSK3b inhibitors
Seung-Chul Lee a,b, Hyun Tae Kim a, Choul-Hong Park a, Do Young Lee a, Ho-Jin Chang a, Soobong Park a,
b,
Joong Myung Cho a, Sunggu Ro a, , Young-Ger Suh
⇑
⇑
a CrystalGenomics, Inc., 5F, Tower A, Korea Bio Park 694-1, Sampyeong-dong, Bundang-gu Seongnam-si Gyeonggi-do 463-400, Republic of Korea
b College of Pharmacy, Seoul National University, 599 Gwanak-ro, Gwanak-gu, Seoul 151-742, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Design, synthesis and biological evaluation of the imidazopyridine analogs as novel GSK3b inhibitors for
treatment of type 2 diabetes mellitus are described. Most of the analogs exhibited excellent inhibitory
activities (IC50 < 44 nM) against glycogen synthase kinase 3b (GSK3b). The structure–activity relation-
ship (SAR) of the imidazopyridine analogs and the binding mode of analog 23 in the catalytic domain
of GSK3b, based on our X-ray crystallography study, are described. In particular, analog 28, which was
selected as a potential drug candidate for treatment of type 2 diabetes mellitus, exhibited excellent
GSK3b inhibition, pharmacokinetic profiles and blood glucose lowering effect in mouse.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 10 April 2012
Revised 14 May 2012
Accepted 15 May 2012
Available online 19 May 2012
Keywords:
Glycogen synthase kinase 3
Imidazopyridine
Type 2 diabetes mellitus
Oral glucose tolerance test
Diabetes is a chronic disease characterized by peripheral insulin
resistance and high concentrations of glucose in the blood. It gen-
erally results in decreased insulin secretion which could poten-
been previously developed by us. In this context, we have initially
attempted to modify the C7-hydroxy group, which seemed to func-
tion as both hydrogen-bonding donor and acceptor interacting
with Asp133 and Val135 of the catalytic domain of GSK3b
(Fig. 2).5 It is well known that phenol bearing compounds undergo
the phase 2 enzymatic metabolism. Therefore, we have decided to
change the 7-hydroxy benzimidazole scaffold to the imidazopyri-
tially
cause
serious
debilitating
conditions
including
cardiovascular morbidity, nerve damage, and mortality.1 Although
significant progress has been achieved in the treatment of type 2
diabetes mellitus (T2DM) over the last decade (Fig. 1), the bio-
chemical and molecular mechanisms underlying insulin resistance
have not been clearly defined yet.2 In recent years, the biological
and pharmacological researches revealed many targets for treat-
ment of type 2 diabetes including glycogen synthase kinase 3
(GSK3). GSK3 is serine/threonine kinase originally identified as
the enzyme responsible for inactivation of glycogen synthase.3 It
is involved in many different biological processes including glucose
homeostasis, cell survival, tumorigenesis, Alzheimer’s disease, and
developmental patterning.4
CH3
NH NH
NH2
O
O
O
H3C
S
O
N
N
N
N
H
N
H
H
CH3
O
N
H
Metformin
Glimepiride
We have recently reported the 7-hydroxy benzimidazole ana-
logs, which exhibited excellent potency in both enzyme and cell
based assay and high selectivity as GSK3b inhibitor.5 However,
the previous scaffold turned out unsatisfactorily for therapeutic
implication in terms of physicochemical properties including polar
surface area (PSA) value. Our strategy involved identifications of
the potent and orally available GSK3b inhibitors via structural
modification of the 7-hydroxy benzimidazole scaffold, which had
CH3
N
CF3
N
N
S
F
F
N
O
N
HN
N
O
O
NH2
O
F
⇑
Sitagliptin
Rosiglitazone
Corresponding authors. Tel.: +88 31 628 2801; fax: +88 31 628 2701 (S.R.); tel.:
+88 2 880 7875; fax: +82 2 888 0649 (Y.-G.S.).
E-mail addresses: sgro@cgxinc.com (S. Ro), ygsuh@snu.ac.kr (Y.-G. Suh).
Figure 1. Examples of known hypoglycemic agents.
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.05.060

4222
S.-C. Lee et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4221–4224
by palladium catalyzed reduction. Condensation of 1 with a variety
(a)
(b)
of aldehyde under microwave irradiation at 200 °C afforded the
corresponding imidazopyridine intermediate 2. Oxidation of 2 in
the presence of KMnO4 and NaOH in H2O under the refluxing con-
ditions afforded acid 3. Amidation of 3 with an appropriate amine
in the presence of Pybop and Et3N in DMF yielded the correspond-
ing imidazopyridine 4. The GSK3b kinase inhibitory activity was
determined using the known method7 and the result is summa-
rized in Table 1.
ASP133
O
H
O
Hydrogen
N
O
Donor
Val135
Pro136
R2
HN R2
HN
Hydrogen
acceptor
N
N
N
N
H
H
R1
R1
Interestingly, the imidazopyridine analog 6 (IC50 14 nM) was
more potent than the 7-hydroxy benzimidazole analog 5 (IC50
47 nM). The other imidazopyridine analogs (7, 8, 9, 10 and 11)
exhibited excellent inhibitory activities in a range 1–12 nM.
The results implied that the imidazopyridine scaffold main-
tained a hydrophobic interaction even with the loss of a hydro-
gen-bonding interaction. In addition, our X-ray crystallographic
data of the complex of GSK3b with analog 23 also revealed that
the imidazopyridine interacted as a hydrogen-bonding donor and
as a hydrogen-bonding acceptor, respectively with a carbonyl
and an amino group of Val135.8 The binding mode was quite sim-
ilar to the results from recently reported our literatures.5,9 The
interaction of Asp133 with the hydroxy group shown in the benz-
imidazole scaffold disappeared as anticipated. Moreover, the inter-
action of C1-NH of the imidazole moiety with Pro136 changed to
the interaction with Val135 (Fig. 3).
Figure 2. Proposed interaction of (a) the 7-hydroxy benzimidazole and (b) the
imidazopyridine with GSK3b.
NO2
NH2
NH2
NH2
N
a
b
R1
N
H
N
N
N
1
2
H
N
O
R2
N
O
OH
N
d
c
R1
R1
N
H
N
H
N
N
3
4
Based on the X-ray crystallographic analysis and the initial
structure–activity relationship (SAR) of the imidazopyridine ana-
logs, the R1 and R2-diversified analogs were synthesized and tested
for the mouse and human microsomal stability. The selected ana-
logs were incubated with human liver microsomes (HLMs) and
mouse liver microsomes (MLMs) at 37 °C for 45 min. The inhibitory
activities, the microsomal stability and the physicochemical prop-
erties of the imidazopyridine analogs are summarized in Table 2.
All analogs exhibited excellent inhibitory activities (IC50s 9–
44 nM) and acceptable cellular potencies (EC50s 56–300 nM). From
the mouse microsomal stability study, it was demonstrated that
analogs 13 and 18 were metabolically stable (70% and 70.2%,
respectively). However, analogs 16, 17, 19, 20, 21 and 18 showed
polar characteristics with PSAs (125–156 Å2), which seemed to
unsatisfactory for oral absorption.10
Thus, we turned our attention to analog 13, which not only had
an acceptable cellular potency (EC50 130 nM) and an excellent
inhibitory activity (IC50 8 nM) but also optimal physicochemical
properties with a clogP (2.73) and a PSA (83.56 Å2). Then, the ana-
logs based on the structure of 13 were synthesized again. The sul-
fonamido and alkyl substituents were avoided in the second series
of the imidazopyridine analogs because these increase PSA and
molecular weight. The inhibitory activities, microsomal stabilities
and physicochemical properties are summarized in Table 3.
The analogs bearing 3-diethylaminophenyl (25), 3-methoxy
phenyl (32) and 3-pyridine (31) substituents exhibited excellent
inhibitory activities (IC50s 3 nM) as anticipated. However, they
were unstable in the presence of both mouse microsome (0%,
6.4% and 6.3%) and human microsome (0%, 23.8% and 6.5%). In gen-
eral, the imidazopyridine analogs showed low metabolic stability
when the phenyl system with an electron-donating group at the
meta-position or the 3-pyridinyl system was introduced as the R1
moiety.
Scheme 1. Reagents and conditions: (a) Pd/C, H2 balloon, MeOH, 5 h, 90–95%; (b)
R1-CHO, nitrobenzene, microwave irradiation, 50 W, 200 °C, 20 min; 70–95%; (c)
KMnO4, NaOH, H2O, reflux, 12 h, 50–90%; (d) Pybop, Et3N, R2-NH2, DMF, 2–10 h, 50–
95%.
Table 1
Inhibitory activities of a 7-hydroxy benzimidazole analog and the imidazopyridine
analogs against GSK3b
O
H
N
S
H
N
O
O
N
R1
N
H
A
Analogs
A
R1
IC50, nM
EC50, nMa
Ref. 14
5
6
7
—
C–OH
N
—
Phenyl
Phenyl
70
47
14
4
ND
ND
62
N
4-F-phenyl
31
8
9
10
11
N
N
N
N
2,4-di-F-phenyl
2-Thiophenyl
3-Thiophenyl
3-Furanyl
2
12
3
290
50
10
1
290
ND, not determined.
a
EC50s were determined by duplicate experiments using Rd cells and expressed
as the mean.
dine scaffold, which has less molecular weight and is tolerable to
potential glucuronidation.6
We have commenced the development of the novel GSK3b
inhibitors by replacing the 7-hydroxy benzimidazole scaffold of
the representative GSK3b inhibitor with the imidazopyridine
scaffold. Representative synthetic route for variation of the R1
and R2-substituent is shown in Scheme 1. Analog 5 consisting of
7-hydroxy benzimidazole scaffold was synthesized to compare
its GSK3b inhibitory activity with that of the corresponding analog
consisting of the imidazopyridine scaffold. Commercially available
2-amino-3-nitro-4-methyl pyridine was converted into diamine 1
Although it exhibited lower potency than analog 27 in the cel-
lular assay (EC50 9 nM vs EC50 1.5 nM), analog 26 consisting of
the 4-methoxy-substituted pyridinyl system as R2 moiety with
metabolic stabilities of 87% (human) and 71% (mouse), was more
stable than analog 27 consisting of the 4-ethoxy-substituted pyrid-
inyl system with metabolic stabilities of 36% (human) and 38%
(mouse). Analogs (24, 26, 28, 29, 33 and 34), which exhibited po-
tent GSK3b inhibition, were subjected to pharmacokinetic study.

S.-C. Lee et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4221–4224
4223
Figure 3. The X-ray crystallographic analysis of the interaction of analog 23 with the catalytic domain of GSK3b. The resolution is 2.6 Å, R value is 0.24 and R_free value is
0.304. The analog is illustrated in sticks with carbon atoms in white color. (a) Binding mode of the analog 23 in the ATP binding pocket of GSK3b was illustrated. (b) Hydrogen-
bonding interaction is described. Interaction of amine of pyridine and NH of the imidazopyridine with NH of Val135, carbonyl of Val135 with distances of 3.0, 2.6 Å are shown.
And carbonyl of amide linker interact with NH of Asp200 through water with distances of 2.5, 3.3 Å.
Table 2
Inhibitory activities against GSK3b, microsomal stabilities and physicochemical properties of the imidazopyridine analogs
6
5
H
B
4
O
N
(CH2)n
N
1
A
3
2
R1
N
N
H
Analogs
R1
R2
IC50, nM
EC50, nM (Rd cells)
Metabolic stabilitya (microsome,%, 45 min)
Mouse Human
clogPb
PSAc, Å2
n
A
B
12
13
14
15
16
17
18
19
20
21
22
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-Cl-phenyl
3-Furanyl
2
0
3
2
2
2
2
2
2
2
2
N
N
N
C
C
C
C
C
C
C
N
4-F
—
4-Cl
3-OH
4-NHSO2CH3
4-NHSO2N(CH3)2
4-NHSO2CH3
4-NHSO2CH3
4-NHSO2CH3
4-NHSO2N(CH3)2
2-Cl
17
8
300
130
250
150
190
210
300
180
150
56
2.9
70
9.4
ND
25.7
12.1
13
7.4
86.6
47
36.8
13.3
26.5
3.34
2.37
3.63
3.28
1.95
1.95
2.42
0.95
1.59
1.58
2.19
83.56
83.56
83.56
17
14
44
27
10
3
12
9
9
4.2
11.1
0.8
1.0
70.2
5.2
2.2
1.6
1.5
90.90
125.22
128.46
125.22
138.36
153.46
156.70
96.70
2-Thiophenyl
3-Thiophenyl
3-Furanyl
310
ND, not determined.
a
Metabolic stability was determined by mass spectroscopy.
clogP.
PSA (polar surface area) was determined using MarvinSketch program.
b
c
Table 3
Inhibitory activities against GSK3b, microsomal stabilities and physicochemical properties of the imidazopyridine analogs based on 13
H
O
N
R2
N
R1
N
H
N
Analogs
R1
R2
IC50, nM
EC50, nM (Rd cells)
Metabolic stability (microsome, %, 45 min)
Mouse Human
clogPa
PSAb, Å2
23
24
25
26
27
28
29
30
31
32
33
34
3-Thiophenyl
4-Phenylpyrrolidine
4-(CH3)2N-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-OMe-phenyl
3-OMe-phenyl
3-Pyridinyl
3-OMe-phenyl
3-Cl-phenyl
Pyridin-3-yl
13
30
3
3
3
8
2
3
3
240
40
23
9
1.5
80
ND
ND
10
10
10
20
ND
28.9
0
87.0
36.0
47.4
56.8
6.6
6.3
6.4
46.7
27.0
ND
36.5
0
71.0
38.0
33.3
51.4
4.0
2.01
2.60
2.20
2.21
2.57
2.50
1.93
1.93
1.21
2.27
3.03
1.21
111.80
86.80
86.80
92.79
92.79
83.56
92.79
92.79
105.68
102.02
92.79
105.68
4-Me-pyridin-3-yl
4-Me-pyridin-3-yl
4-OMe-pyridin-3-yl
4-OEt-pyridin-3-yl
6-Me-pyridin-3-yl
4-Me-pyridin-3-yl
4-Me-pyridin-3-yl
4-OEt-pyridin-3-yl
4-OEt-pyridin-3-yl
4-OMe-pyridin-3-yl
4-OMe-pyridin-3-yl
6.5
7
2
2
23.8
36.5
96.0
4-Pyridine
ND, not determined.
a
clogP.
b
PSA (polar surface area) was determined using MarvinSketch program.

4224
S.-C. Lee et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4221–4224
Table 4
Pharmacokinetic profiles of the selected analogsa
Analogs
AUCb (ng/ml h)
Cmax (ng/mL)
Tmax (h)
24
26
28
29
33
34
39
—
—
1
3
0.5
—
—
545
2398
645
27
228
843
496
—
LLOQc
—
Dose 27.3 mg kgÀ1, ICR (Institute of cancer research) mice (n = 2, male), HCl salt formulation, Oral administration.
a
Date were determined using WinNonline 5.3 program.
AUC, area under curve.
LLOQ, lower limit of quantitation.
b
c
Figure 4. In vivo efficacy data of analog 28 in male mice OGTT (n = 5). The glucose was orally injected 30 min after oral administration of the suspension of analog 28 in
PEG400. (a) Plasma glucose lowering effect and (b) glucose AUC reduction of analog 28. ⁄Values of p < 0.05 were considered statistically significant for vehicle group.
They were formulated as HCl salt for oral administration in ICR
mice. The evaluation results are summarized in Table 4. Analog 28
showed the best oral exposure in the study. However, oral expo-
sures of analogs 24, 33 and 34 were unsatisfactory despite of their
optimal physiochemical properties.
Based on many literature publications, GSK3 had been known as
one of key mediators of insulin signaling pathway as it induces the
glucose uptake, glycogen synthesis and lipid synthesis.11 Therefore,
we have further confirmed the pathophysiological relationship of
the GSK3b inhibition through systemic glucose lowering effect of
analog 28 in the oral glucose tolerance test (OGTT) using the ICR
mice. The results are summarized in Figure 4. The glucose was
orally injected after 30 min of administration of analog 28 and as
expected, analog 28 effectively lowered the plasma glucose on
the doses of 100 mg/kg and 10 mg/kg.12
These data suggested that GSK3b inhibitors could provide a po-
tential therapeutic approach for treating type-2 diabetes.
In summary, we have designed and synthesized series of novel
imidazopyridine analogs as excellent GSK3b inhibitors. We estab-
lished the structure-activity relationship of the synthesized imidazo-
pyridine analogs. The binding mode of the imidazopyridine analogs
in the catalytic domain of GSK3b using analog 23 as a representative
was elucidated via X-ray crystallography study. We were also able to
identify potent GSK3b inhibitors that exhibited acceptable metabolic
stabilities, physicochemical properties and promising pharmacoki-
netic profiles. In particular, analog 28 demonstrated effective blood
glucose lowering effect in the OGTT using ICR mice. Currently, fur-
ther biological and pharmacological studies on the selected analogs
are in progress for the therapeutic implication.
References and notes
1. (a) Saltiel, A. R.; Kahn, C. R. Nature 2001, 414, 799; (b) Amos, A. F.; McCarty, D.
J.; Zimmet, P. Diabet. Med. 1997, 14, 7; (c) King, H.; Aubert, R. E.; Herman, W. H.
Diabetes Care 1998, 21, 1414.
2. (a) Williams-Herman, D.; Johnson, J.; Teng, R.; Luo, E.; Davies, M. J.; Kaufman, K.
D. Curr. Med. Res. Opin. 2009, 25, 569; (b) Yoo, D. Y.; Kim, W.; Nam, S. M.; Yoo,
M. H.; Hwang, I. K.; Yoon, Y. S. Diabetes Care 2010, 36, 2401; (c) Kus, V.; Flachs,
P.; Kuda, O.; Bardova, K. PLoS One 2011, 6, e27126; (d) Sangle, G. V.; Lauffer, L.
M.; Grieco, A. Endocrinology 2012, 153, 564.
3. Cohen, P.; Yellowlees, D.; Aitken, A.; Donella-Deana, A.; Hemmings, B. A.;
Parker, P. j. Eur. J. Biochem. 1982, 124, 21.
4. Martinez, A.; Castro, A.; Dorronsoro, i.; Alonso, m. Med. Res. Rev. 2002, 22, 373.
5. Shin, D.; Lee, S. C.; Heo, Y. S.; Lee, W. Y. Bioorg. Med. Chem. Lett. 2007, 17, 5686.
6. Mutlib, A.; Goosen, T. C.; Bauman, J. N. Chem. Res. Toxicol. 2006, 19, 701.
7. (a) Gong, L.; Grupe, A.; Peltz, G. A. PCT Int. App. WO0210158, 2002.; (b) Cho, J.
M.; Ro, S.; PCT Int. App. WO2004065370, 2004.
8. PDB ID code: 4DIT, title: Crystal structure of GSK3beta in complex with an
imidazopyridine inhibitor.
9. Lee, S. C.; Shin, D.; Cho, J. M.; Ro, S.; Suh, Y. G. Bioorg. Med. Chem. Lett. 1891,
2012, 22.
10. (a) Ertl, P.; Rohde, B.; Selzer, P. J. Med. Chem. 2000, 43, 3714; (b) Lipinski, C. A.;
Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery Rev. 2001, 46, 3.
11. Mohler, M. L.; He, Y.; Wu, Z.; Hwang, D. J.; Miller, D. D. Med. Res. Rev. 2009, 29,
125.
12. As demonstrated in many literatures, analog 28 also showed non-linear
pharmacodynamics. It seems that the inconsistent solubility of analog 28
influenced the intestinal absorption, which resulted in an induction of non-
linear pharmacodynamics. There is also a possibility of effects by the mouse
efflux enzymes (e.g., Pgp). Currently, we try to solve the problem of non-linear
pharmacodynamics via appropriate formulation. (a) Sparreboom, A.; Tellingen,
O. V.; Nooijen, w. j. Cancer Res. 1996, 56, 2112; (b) Mueller, E. A.; Kovarik, J. M.;
Bree, J. B.; Tetzloff, W.; Grevel, J. Pharm Res. 1994, 2, 301; (c) Kodell, R. L.;
Turturro, A. Nonlinearity Biol. Toxicol. Med. 2004, 2, 35.