Novel fatty acid binding protein 4 (FABP4) inhibitors: Virtual screening, synthesis and crystal structure determination

Article abstract of DOI:10.1016/j.ejmech.2014.11.020

Fatty acid binding protein 4 (FABP4) is a potential drug target for diabetes and atherosclerosis. For discovering new chemical entities as FABP4 inhibitors, structure-based virtual screening (VS) was performed, bioassay demonstrated that 16 of 251 tested compounds are FABP4 inhibitors, among which compound m1 are more active than endogenous ligand linoleic acid (LA). Based on the structure of m1, new derivatives were designed and prepared, leading to the discovery of two more potent inhibitors, compounds 9 and 10. To further explore the binding mechanisms of these new inhibitors, we determined the X-ray structures of the complexes of FABP4-9 and FABP4-10, which revealed similar binding conformations of the two compounds. Residue Ser53 and Arg126 formed direct hydrogen bonding with the ligands. We also found that 10 could significantly reduce the levels of lipolysis on mouse 3T3-L1 adipocytes. Taken together, in silico, in vitro and crystallographic data provide useful hints for future development of novel inhibitors against FABP4.

Full text of DOI:10.1016/j.ejmech.2014.11.020

European Journal of Medicinal Chemistry 90 (2015) 241e250
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Novel fatty acid binding protein 4 (FABP4) inhibitors: Virtual
screening, synthesis and crystal structure determination
a
,
b
,
1
a
,
c
,
1
d
,
1
a
a
Haiyan Cai
, Qiufeng Liu
, Dingding Gao , Ting Wang , Tiantian Chen ,
Guirui Yan , Kaixian Chen , Yechun Xu , Heyao Wang a , Yingxia Li
a
a
a,
*
,
*
d, *
,
a, *
Weiliang Zhu
a
Drug Discovery and Design Center, Key Laboratory for Receptor Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555
Zuchongzhi Road, Shanghai 201203, China
Department of Pathophysiology, Key Laboratory of Cell Differentiation and Apoptosis of Chinese Ministry of Education, Shanghai JiaoTong University,
b
School of Medicine, Shanghai, China
c
School of Life Science and Technology, Shanghai Tech University, Shanghai 200031, China
School of Pharmacy, Fudan University, 826 Zhangheng Road, Shanghai 201203, China
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 July 2014
Received in revised form
Fatty acid binding protein 4 (FABP4) is a potential drug target for diabetes and atherosclerosis. For
discovering new chemical entities as FABP4 inhibitors, structure-based virtual screening (VS) was per-
formed, bioassay demonstrated that 16 of 251 tested compounds are FABP4 inhibitors, among which
compound m1 are more active than endogenous ligand linoleic acid (LA). Based on the structure of m1,
new derivatives were designed and prepared, leading to the discovery of two more potent inhibitors,
compounds 9 and 10. To further explore the binding mechanisms of these new inhibitors, we determined
the X-ray structures of the complexes of FABP4-9 and FABP4-10, which revealed similar binding con-
formations of the two compounds. Residue Ser53 and Arg126 formed direct hydrogen bonding with the
ligands. We also found that 10 could significantly reduce the levels of lipolysis on mouse 3T3-L1 adi-
pocytes. Taken together, in silico, in vitro and crystallographic data provide useful hints for future
development of novel inhibitors against FABP4.
8
November 2014
Accepted 10 November 2014
Available online 11 November 2014
Keywords:
FABP4
Inhibitor
Diabetes
Virtual screening
Crystal structure
© 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
FABP4 expression in adipose tissue in human being showed lower
serum triglyceride levels and significantly reduced risk of type 2
Fatty acids function both as energy source and as signals for
metabolic regulation in mammalian cells. Chaperones of fatty acids
are commonly known as fatty acid binding proteins (FABPs) that act
as fatty acids shuttles and play important roles in metabolic and
inflammatory diseases [1e3]. Since the initial discovery of FABPs in
diabetes [9]. Increasing evidences indicate that pharmacological
agents that modify FABP4 function may offer therapeutic oppor-
tunities for metabolic syndrome. Indeed, several series of small-
molecule inhibitors of FABP4 have been reported over the last
decade [10e17]. For example, BMS309403, a highly active inhibitor
of FABP4, demonstrated to be effective to improve glucose meta-
bolism, enhance insulin sensitivity in both dietary and genetic
mouse models of obesity and diabetes, and ameliorate the symp-
tom of atherosclerosis [18]. The above studies indicated the po-
tential application of FABP4 inhibitors in the treatment of diabetes
and atherosclerosis.
1972, at least nine members have been identified.
FABP4, also known as A-FABP or aP2, was first detected in
mature adipocytes and adipose tissues. It plays a significant role in
many aspects of metabolic syndrome [4,5]. Disruption of FABP4 in
mice prevents them from diet-induced insulin resistance [6,7],
while macrophage-specific deletion of FABP4 leads to protection of
atherosclerosis in apolipoprotein E-deficient mice [8]. Reduced
To date, several crystal structures of human FABP4 have been
determined for apo and holo forms (bound with various ligands)
[
13e16,19,20]. Structurally, FABP4 is composed of a well-known
assembly of ten antiparallel -strands and a helix-loop-helix cap
in the N-terminal. The complex structures of FABP4 and various
b
*
Corresponding authors.
E-mail addresses: ycxu@simm.ac.cn (Y. Xu), wlzhu@mail.shcnc.ac.cn (W. Zhu).
These authors contributed equally to the work.
1
http://dx.doi.org/10.1016/j.ejmech.2014.11.020
223-5234/© 2014 Elsevier Masson SAS. All rights reserved.
0

242
H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
ligands provide insight into the structural foundation underlying
the binding modes of endogenous ligands and small molecule in-
hibitors in the active pocket of FABP4.
value of 16.8 mM (Table 1, Fig. 1), which was slightly more active
than the endogenous ligand linoleic acid (LA, 17.7 mM, Table 1).
These results inspire us to discover and design novel inhibitors
of FABP4 via the current state-of-the-art structure-based approach.
We reported a novel types of FABP4 inhibitors discovered by virtual
screening [11]. Here, we report another type of small-molecule
inhibitors. Starting from the structure of m1 (Fig. 1), which was
discovered via virtual screening, we designed and synthesized a
series of new 2,4,6-triisopropylbenzene derivatives, leading to the
discovery of more potent compounds 9 and 10. X-ray crystallog-
raphy was then applied to study the binding mechanisms between
FABP4 and the new compounds.
2.2. Compound design and synthesis
Starting from the molecular structure of m1 and the docked
complex structure of FABP4-m1, we designed and synthesized a
new series of 2,4,6-triisopropylbenzene derivatives mainly focusing
on the replacement of imidazole with benzimidazole and benz-
imidazole analogs (Schemes 1e3). In general, treatment of appro-
priate commercially available substituted diaminobenzene with
triethyl orthoformate or triethyl orthoacetate and CH OH under the
3
Table 1
Inhibitory activities of newly designed and synthesized benzimidazole analogs.
2
. Results and discussion
2.1. Virtual screening and biological evaluation
In an effort to discover novel inhibitors of FABP4, two chemical
databases: Specs (http://www.specs.net) and Maybridge (http://
www.maybridge.com) were virtually screened against the target
protein FABP4. The X-ray structure of FABP4 binding with
BMS309403 (PDB code: 2NNQ [14]) was retrieved from the Protein
Data Bank (http://www.rcsb.org/pdb) for docking calculation. Specs
compounds (~200,000 compounds) were first filtered with our in-
house script Druglk, leading to 80,000 compounds left. Then
similar procedure to our previous work was applied for the virtual
screening of FABP4 inhibitors in this study [11]. Briefly, the data-
bases Maybridge and Specs were filtered by software DOCK4.0 [21],
the top 10% ranked compounds with the highest score were
selected by software GLIDE [22,23] (www.schr o€ inger.com). Then
the top 800 and 300 compounds from Maybridge and Specs,
respectively, were rescored by AUTODOCK3 program [24]. Taking
into account both Glide and AUTODOCK3 scores, 52 and 199 mol-
ecules from Maybridge and Specs were finally selected and pur-
chased from the vendors. The inhibitory activity of the purchased
compounds were determined using the 1,8-ANS ligand displace-
ment assay as described in our previous study [11]. Primary
Compound
R
X
Y
Inhibition
IC50
a
at 100
m
M
(%) (mean ± SE (mM))
Linoleic acid (LA)
m1
1
e
e
e
H
H
H
e
85.0
87.8
<10.0
36.3
<10.0
<10.0
<10.0
53.6
17.7 ± 0.20
16.8 ± 0.18
NT
>100.0
NT
NT
NT
82.8 ± 0.91
14.8 ± 0.29
NT
NT
NT
NT
NT
14.8 ± 0.26
NT
NT
NT
NT
NT
NT
>100.0
>100.0
e
e
H
CH
H
H
H
CH
H
CH
H
2
2
a
c
3
3
2d
2e
CH3 CH3
CH
CH
H
3
3
H
3
3
3
3
3
a
b
c
Br
Br
H
H
92.2
H
Br
Br
H
<10.0
<10.0
<10.0
<10.0
<10.0
90.8
<10.0
<10.0
<10.0
<10.0
<10.0
<10.0
10.0
d
CH
CH
H
3
3
H
4a
Cl
Cl
H
H
OCH
OCH3
H
H
NO
H
NH
4
4
4
5
b
c
d
a
H
H
Cl
Cl
H
CH
CH
H
H
CH
H
3
3
3
bioassay was performed at 100
mM, 11 compounds showed inhib-
5b
5c
H
itory activity against FABP4 (6 for Specs and 5 for Maybridge). Based
on these hits, a structural similarity-based search in the two da-
tabases was performed. Another 36 compounds were purchased
from Maybridge, leading to the discovery of another 5 hits. How-
ever, none were purchased from Specs because no highly similar
structures were identified in the database. Thus, 16 FABP4 in-
hibitors in total were discovered through the two-step virtual
screening and bioassay approach. One series of the active com-
pounds (10 compounds) have been reported in our previous study
OCH
OCH
H
NO
H
3
5d
6b
6c
7
8
3
3
2
H
H
2
2
13.3
NH
S
2
O
O
[11]. The rest 6 active compounds are sulfonamides, among which
compound m1 demonstrated highest FABP4 inhibition with an IC50
9
1
92.1
96.0
7.9 ± 0.086
OH
S
O
O
0
4.0 ± 0.043
COOH
NT: not determined.
a
Values are means of triplicate experiments with relative standard deviations
Fig. 1. Structure of m1.
<10%, the data were analyzed with GraphPad Prism software.

H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
243
promotion of ZrCl
triisopropylbenzenesulfonyl chloride and triethylamine to give 2e5
Scheme 1). Because benzimidazole ring has the phenomenon of
tautomerism in the solution, compounds 11aeh and 6-
nitrobenzimidazole reacted with 2,4,6-
4
afforded 11aeh. Then 11aeh reacted with 2,4,6-
N
R
N
N
S
(
O
S
Et N, THF, rt.
R
3
O
O
+
Cl
N
H
O
triisopropylbenzenesulfonyl chloride to give both 5-substituted
and 6-substituted products. The ratio of them was about 1:1.
The target compounds 1 and 2a were synthesized by commer-
cially available 2-substituted benzimidazole reacting with 2,4,6-
triisopropylbenzenesulfonyl chloride in the presence of triethyl-
amine as shown in Scheme 2. Compound 6b and 6c were prepared
in a similar manner as described for compound 1 started from
commercially available 6-nitrobenzimidazole. A suspension of 6b
1
2
R=H;
a R=CH3;
Scheme 2. Scheme for the synthesis of 1 and 2a.
2
However, 3a with a methyl group at the position C of the benz-
imidazole group, showed reduced activity (IC50 82.8 M).
m
and Pd/C in CH
2
Cl
2
eMeOH under H
2
at room temperature to give
The most active inhibitors of m1 analogs, namely 3b and 4d, had
similar activity to m1. Thus, the replacement of imidazole with
benzimidazole analogs has no significant effect on activity. Studies
have revealed that small molecules with acid groups showed high
inhibitory activity against FABP4 [13,14], we then replaced the
benzimidazole group with a hydroxyl group (9). Indeed, the
inhibitory activity of 9 increased remarkably with an IC50 value of
7.9 mM. When the sulfonic acid group was substituted by carboxylic
acid group (10), the inhibitory activity increased nearly 2-fold again
(Table 1).
compound 7 as shown in Scheme 3.
The preparation of the target compounds 8, 9 and 10 was per-
formed as shown in Scheme 4. The target compound 8 was pre-
pared from 2,4,6-triisopropylbenzene-sulfonyl chloride by reaction
with NH
chloride in H
3
$H
2
O in THF. Reflux of 2,4,6-triisopropylbenzenesulfonyl
O/dioxane (v/v,1:1) to provide compound 9. Com-
2
pound 10 was produced by the reaction of 2,4,6-
triisopropylbenzoyl chloride with triethylamine in acetone in 99%
yield.
FABP3-deficient mice showed reduced exercise tolerance, and
developed regional cardiac hypertrophy at old age, indicating that
FABP3 played an important role in metabolic homeostasis [25].
Therefore, the activity of 9 and 10 against FABP3 were also deter-
mined (Table 2). 9 displayed 11.5-fold preferences over FABP3,
while 10 showed 18.3-fold preferences over FABP3.
Fabp4 and Fabp5 double-knockout mice demonstrated strong
protection from diet-induced obesity (DIO), insulin resistance and
type 2 diabetes [26]. Dual inhibition by both FABP4 and FABP5 was
reported to be potentially useful for the treatment of dyslipidemia
and/or diabetes. Thus, we also determined the inhibitory activities
of 9 and 10 on FABP5. 9 demonstrated modest activity on FABP5
2
.3. In vitro FABP4 inhibition assay
The synthesized compounds were evaluated for their ability to
inhibit FABP4. As shown in Table 1, 8 of the newly synthesized
compounds showed inhibitory activities against FABP4 at 100 M.
Notably, compounds 9 and 10 were the most potent inhibitors
against FABP4 with IC50 values of 7.9 and 4.0 M, respectively. The
bioassay result showed that the substitution of imidazole with
benzimidazole (1) caused complete loss of inhibition (Table 1).
Benzimidazole C -substituted compounds of 1 (2c, 3c, 4c, 5c, 6c)
exhibited no activities against FABP4 at 100 mM, either. However,
increased inhibitory activity was observed when a methyl group
was introduced into the position C of the benzimidazole group (2a,
m
m
5
with an IC50 value of 16
mM while 10 is weak with the IC50 value
higher than 50 M. Thus, further structural optimization of these
m
compounds is still needed for the development of more effective
candidates to treat type 2 diabetes without side effects.
2
Table 1). Moreover, compound 4d, with a methyl group at the po-
sition C2 of the benzimidazole group of 4c, showed pretty high
potency against FABP4 with an IC50 value of 14.8
mM (Table 1).
2.4. Lipolysis inhibition in 3T3-L1 adipocytes
FABP4 inhibitory activity of the analogs was significantly
6
affected by the type of chemical modifications on the position C
(
As targeted deletion or chemical inhibition of FABP4 decreased
lipolysis in adipocytes [12,27], we accessed whether 9 and 10 could
modulate the levels of lipolysis on adipocytes. We used mouse 3T3-
L1 adipocytes in this work, as FABP4 sequences and tertiary
6
3b, 4b, 5b, 6b, 7). The bioassay results revealed that C -amino
substituted 7 showed weak inhibitory activity, while 3b with
halogen substituent exhibited high potency (IC50 14.8 M).
m
R
3
4
N
R
2
H
N
R
1
NH
2
2
OEt
R
N
S
ZrCl
4
3
, CH OH, rt.
Et N, THF, rt.
3
R
2
O
O
+
R
2
C
OEt
N
R
1
NH
OEt
11a-h
2-5
2
2
2
3
3
3
3
c R =R =H, R =R =CH ;
4a R
4b R
2
=CH
=R
=R
3
, R
=H, R
=H, R
3
=H, R
4
=R
1
=Cl;
2
4
3
1
3
d R =R =R =CH , R =H;
2
3
4
=R
=R
=R
1
=Cl;
=Cl;
2
4
1
3
3
e R =R =R =CH , R =H;
4c R
a R =CH , R =H, R =R =Br; 4d R
2
4
3
1
2
3
1
3
4
=CH
=CH
, R
=Cl, R
=H, R =R
=R
=R
d R =CH , R =R =OCH , R =H;
=H;
2
3
3
4
1
2
2
2
3
3
1
4
b R =R =H, R =R =Br;
5a R
5b R
, R
3
3
4
1
=OCH ;
3
2
3
4
1
c R =R =H, R =R =Br;
=R
=R
3
=H, R
=H, R
4
1
=OCH
=OCH ;
3
3
;
2
4
3
1
d R =CH , R =R =Br, R =H; 5c R
2
3
3
1
4
2
4
3
1
5
2
3
3
1
3
4
Scheme 1. Scheme for the synthesis of 2e5.

244
H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
N
R
1
N
H
N
H N
N
S
2
N
S
2 2 2
10% Pd/C,H ,MeOH/CH Cl ,
O
rt.
R2
Et N, THF, rt.
O
O
3
O
N
2
O N
7
6
b or 6c
6
6
7
b R
1
2 2
=H, R =NO ;
c R =NO , R =H;
1
2
2
R
1
=H, R
2
2
=NH ;
Scheme 3. Scheme for the synthesis of 6 and 7.
Cl
S
NH2
O
O
O
S
O
THF, NH ·H O, rt., 0.5 h
3
2
8
OH
Cl
S
O
S
O
O
O
H O/dioxane, reflux 18 h
2
9
O
O
OH
Cl
Fig. 2. Effect of 9 (50
mM) and 10 (50 mM) on forskolin-stimulated lipolysis in mouse
C
C
3T3-L1 adipocytes. Control: DMSO.
Et N 1eq, H O 0.5 eq,
3
2
acetone, reflux 18 h
determined by X-ray crystallization. The FABP4 conformations that
we obtained can align well with the previous reported structures.
The complex structures revealed similar binding modes for the two
compounds in the pocket of FABP4 (Fig. 3(A) and (B)). The 2,4,6-
triisopropylbenzene groups lie in the hydrophobic region of the
pocket defined by Met20, Val25, Ala33, Ala36, Phe57, Ala75 and
Ile104 (Fig. 3(A) and (B)). In addition, the acid groups of both
compounds (sulfonic acid group of 9 and carboxylic acid group of
1
0
Scheme 4. Scheme for the synthesis of 8e10.
structures are highly conserved between human (PDB code: 2NNQ
14]) and mouse (PDB code: 3HK1 [12]). The result showed 10 could
significantly reduce forskolin-stimulated lipolysis at dose level of
M (Fig. 2). Although no effect was observed at 50 M (Fig. 2), 9
reduced approximately 9.0% of the forskolin-stimulated lipolysis at
00 M. These results are in agreement with our inhibitory activity
[
10) formed hydrogen bonds with the protein and four structural
5
0
m
m
water molecules (Ser53, Thr60, Arg106, Arg126, Tyr128; Fig. 3).
Although the X-ray structures provided detailed binding infor-
mation between FABP4 and the two compounds, the mechanisms
for activity differences between the benzimidazole analogs were
still unclear. To further explore the binding mechanism between
other benzimidazole analogs with FABP4, 4d was docked into the
binding pocket of FABP4 by AUTODOCK4 [24]. The crystal results
demonstrated that the acid groups of 9 and 10 formed main polar
interactions with 3 key residues of FABP4 (Arg106, Arg126, Tyr128,
Fig. 3). The docking result showed that sulfonyl group formed polar
interactions with residue Arg126 and Tyr128 of FABP4, but not
Arg106 (Fig. 4). This might be one of the reasons that 4d and other
benzimidazole analogs had weaker inhibitory activities than 9 or
1
m
data on FABP4 (Table 1). It confirmed that 9 and 10 are FABP4
inhibitors.
2
.5. Binding mechanisms revealed by X-ray crystallography
To explore the binding mechanisms between FABP4 and the
inhibitors, the complex structures of FABP4-9 and FABP4-10 were
Table 2
Selectivity of the tested FABP4 inhibitors over FABP3.
10. The docking result also revealed hydrophobic interactions be-
Compound
FABP4 IC50a
(
m
M) FABP3 IC50b
(
mM)
Selectivity index
FABP3/FABP4
tween residue Met40, Ile115 and Arg128 and the methyl group at
2
the position C of the benzimidazole. This methyl group might help
Linoleic acid (LA)
9
1.0
0.45
0.23
1.0
5.16
4.20
1.0
11.5
18.3
to stabilize the conformation of the ligand. Thus when a methyl
group was introduced into the position C of the benzimidazole
2
10
group (2a, Table 1), it showed moderate activity, and those without
a
2
Values are means of triplicate experiments with relative standard deviations
methyl group at the position C (2c, 3c, 4c, 5c, 6c) showed no ac-
<
10%, presented as the fold of LA on FABP4.
tivities except 3b. Moreover, polar groups with small volume, such
b
Values are means of triplicate experiments with relative standard deviations
5
as Cl, might be favored at the position C as shown in Fig. 4. The
<
10%, presented as the fold of LA on FABP3.

H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
245
Fig. 3. X-ray crystal structures of 9 (A) and 10 (B) binding to the active site of FABP4. The dashed lines in red represent hydrogen bonds. Compounds 9 (slate) and 10 (green) are
shown in sticks. Residues of FABP4 are shown in yellow sticks. Waters are shown as red balls. (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
above results should also be helpful for further structural optimi-
zation of FABP4 inhibitors.
between the ligands and FABP4. Our results provide useful infor-
mation for further development of FABP4 inhibitors.
3
. Conclusion
4. Experimental
In summary, we carried out virtual screening to identify novel
4.1. Chemistry
inhibitors of FABP4. Based on the structure of the most active hit
m1, structural modifications by the replacement of imidazole with
benzimidazole and benzimidazole analogs were carried out. The
bioassay demonstrated eight new compounds had inhibitory ac-
tivities against FABP4. Among them, compounds 9 and 10 were the
Melting points were measured on an SGW X-4 melting point
apparatus without correction. The type of analytical thin-layer
chromatography (TLC) was HSGF 254 (0.15e0.2 mm thickness,
Yantai Jiangyou Company, China). H NMR and C NMR spectra
were taken on a Varian/Mercury Plus 400 spectrometer with tet-
ramethylsilane as an internal standard. Chemical shifts were re-
1
13
most potent compounds against FABP4, with IC50 7.9
.0 M, respectively. Moreover, 10 could reduce forskolin-
stimulated lipolysis at dose level of 50 M. X-ray crystallography
mM and
4
m
m
ported in parts per million (ppm,
d
)
downfield from
and further docking results demonstrated the binding mechanisms
tetramethylsilane. Proton coupling patterns were described as
singlet (s), doublet (d), triplet (t), quartet (q), multiplet (m), and
broad (br). Mass spectra were given with electric (EI) produced by
Agilent 1100LC/MSD mass spectrometer.
4.1.1. 1-(2,4,6-Triisopropylphenylsulfonyl)-1H-benzo[d]imidazole
(1)
A solution of benzimidazole (550 mg, 4.66 mmol) in anhydrous
THF (10 mL) and triethylamine (0.67 mL, 4.66 mmol) was stirred for
.5 h, then treated with 2,4,6-triisopropylbenzenesulfonyl chloride
1.41 g, 4.66 mmol) in anhydrous THF (5 mL), and refluxed for 12 h
0
(
until TLC indicated that the reaction was complete. The resulting
precipitate of triethylammonium chloride was filtered off. The so-
lution was concentrated under reduced pressure. The residue was
purified by silica gel column chromatography (8:1, v/v, petroleum
ether/EtOAc) to give 1 (1.467 g, 82%) as a white solid; R
f
¼ 0.6 (1:4,
EtOAc:Petroleum ether); mp 142e144 C; H NMR (DMSO): 8.80
s, 1H, CHeN), 7.75e7.72 (m, 1H, AreH), 7.33 (s, 2H, AreH),
ꢀ
1
d
(
7.31e7.26 (m, 2H, AreH), 7.10e7.07 (m, 1H, AreH), 4.06e3.99 (m,
2
H, 2 ꢁ CH(CH
3 2 3 2
) ), 2.92e2.88 (m, 1H, CH(CH ) ), 1.14 (d, 6H,
13
J ¼ 6.7 Hz, 2 ꢁ CH
3
), 1.01 (d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
); C NMR
(DMSO):
d
147.3, 146.8, 140.7, 130.6, 126.0, 121.4, 114.5, 33.3, 28.0,
þ
2
4.8; ESI-MS m/z: 385.1 ([MþH] ).
Fig. 4. Predicted binding conformation of 4d (magenta) aligned with 10 (green) in the
crystal structure of FABP4. The dashed lines in red represent hydrogen bonds. Com-
pounds 4d and 10 are shown in sticks. Residues of FABP4 are shown in yellow sticks.
4
.1.2. 2-Methyl-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
imidazole (2a)
Compound 2a was prepared as a white solid in a similar manner
(
For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
as described for compound 1. Yield: 35%;
R
f
¼ 0.6 (1:4,

246
H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
EtOAc:Petroleum ether); mp 104e105 C; 1H NMR (DMSO):
7.62e7.59 (m, 1H, AreH), 7.56e7.54 (m, 1H, AreH), 7.32 (s, 2H,
AreH), 7.30e7.28 (m, 2H, AreH), 3.90e3.83 (m, 2H, 2 ꢁ CH(CH ),
.94e2.91 (m, 1H, CH(CH ), 2.28 (s, 3H, CH ), 1.15 (d, 6H,
ꢀ
4.1.4. General procedure for the preparation of 2,5-trimethyl-1-
(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]imidazole or 2,6-
trimethyl-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
imidazole (2ce5d)
Compounds 2ce5d were prepared in a similar manner as
described for compound 1 started from appropriate substituted
benzimidazole 11aeh.
d
3 2
)
2
3
)
2
3
13
J ¼ 6.7 Hz, 2 ꢁ CH
3
), 0.93 (d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
); C NMR
(
CDCl
3
):
d
155.5, 151.6, 151.1, 141.4, 134.2, 132.2, 124.7, 124.5, 119.8,
þ
1
13.3, 34.5, 29.7, 24.4; ESI-MS m/z: 400.0 ([MþH] ).
4.1.3. General procedure for the preparation of 2,5-substituted-1H-
4.1.4.1. 5-Methyl-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
benzo[d]imidazole or 2,6-substituted-1H-benzo[d]imidazole
11aeh)
A mixture of o-phenylenediamine (1.0 mmol), triethyl ortho-
formate (1.2 mmol) and ZrCl (0.1 mmol) in 10 mL MeOH was
stirred at room temperature for 3 h. After completion of the reac-
tion, as indicated by TLC, the solvent was concentrated and the
resulting product was directly purified by silica gel column chro-
matography (4:1 / 1:1, v/v, petroleum ether/EtOAc) to afford
compound 11aeh.
imidazole (2c). A white solid, yield 85%; R ¼ 0.6 (1:4, EtOAc:Pe-
f
ꢀ
1
(
troleum ether); mp 154e155 C; H NMR (CDCl ):
d
8.28 (s, 0.3H,
3
CHeN), 8.26 (s, 1H, CHeN), 7.64 (d, 1H, J ¼ 8.2 Hz, AreH), 7.56 (s,
0.3H, AreH), 7.22 (s, 2H, AreH), 7.21 (s, 0.7H, AreH), 7.15 (d, 1H,
J ¼ 5.9 Hz, AreH), 7.11 (d, 1H, J ¼ 6.7 Hz, AreH), 4.24e4.15 (m, 2H,
2 ꢁ CH(CH ) ), 2.94e2.91 (m, 1H, CH(CH ) ), 2.46 (s, 1H, CH ), 2.38
4
3
2
3 2
3
(s, 3H, CH ), 1.25 (d, 6H, J ¼ 7.0 Hz, 2 ꢁ CH ), 1.14 (d, 12H, J ¼ 6.7 Hz,
3
3
13
4 ꢁ CH ); C NMR (CDCl ):
3
3
d
155.5, 155.5, 151.7, 151.6, 141.5, 140.5,
140.0, 135.3, 131.4, 126.6, 125.8, 124.5, 120.8, 120,4, 111.7, 34.3, 29.6,
þ
2
4.4, 23.4; ESI-MS m/z: 400.0 ([MþH] ).
4
.1.3.1. 6-Methyl-1H-benzo[d]imidazole (11a). A white solid, yield
1
100%; R
f
¼ 0.4 (1:20, CH
2 2 3 3
Cl :CH OH); H NMR (CDCl ): d 8.1 (s, 1H,
4.1.4.2. 2,6-Trimethyl-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo
CHeN), 7.58 (d, 1H, J ¼ 8.3 Hz, AreH), 7.46 (s, 1H, AreH), 7.13 (dd,
[d]imidazole
(2d);
2,5-trimethyl-1-(2,4,6-triisopropyl-phenyl-
1
3
1
d
H, J ¼ 1.2, 8.3 Hz, AreH), 2.49 (s, 3H, AreCH
3 3
); C NMR (CDCl ):
sulfonyl)-1H-benzo[d]imidazole (2e). A white solid of mixture of 2d
140.5, 137.5, 136.4, 132.7, 124.3, 115.4, 114.8, 21.6.
and 2e, yield 85%; R
f
¼ 0.6 (1:4, EtOAc:Petroleum ether); mp
1
ꢀ
125e127 C (mixture of 2d and 2e); H NMR (CDCl
3
):
d
7.65 (dd, 1H,
4
.1.3.2. 2,6-Dimethyl-1H-benzo[d]imidazole (11b). A white solid,
J ¼ 0.8, 7.8 Hz, AreH), 7.52 (d, 2H, J ¼ 6.3 Hz, AreH), 7.43 (s, 1H,
AreH), 7.27 (d, 2H, J ¼ 2.4 Hz, AreH), 7.18 (s, 2H, AreH), 7.11 (d, 2H,
J ¼ 7.8 Hz, AreH), 4.07e4.03 (m, 4H, 4 ꢁ CH(CH ) ), 2.94e2.90 (m,
1
yield 96%; R
d
f
¼ 0.2 (1:1, EtOAc:Petroleum ether); H NMR (CDCl
3
):
7.45 (d, 1H, J ¼ 8.3 Hz, AreH), 7.33 (s, 1H, AreH), 7.04 (d, 1H,
3
2
1
3
J ¼ 7.9 Hz, AreH), 2.62 (s, 3H, CH
3
), 2.46 (s, 3H, CH
3
); C NMR
2H, 2 ꢁ CH(CH ) ), 2.45 (s, 6H, 2 ꢁ CH ), 2.43 (s, 3H, CH ), 2.39 (s,
3
2
3
3
(
CDCl ): 150.9, 138.5, 137.1, 131.9, 123.5, 114.3, 114.0, 21.6, 14.9.
3
d
3H, CH ), 1.25 (d, 12H, J ¼ 7.0 Hz, 4 ꢁ CH ), 1.04 (d, 24H, J ¼ 6.7 Hz,
3
3
13
8
ꢁ CH
3
); C NMR (CDCl
3
):
d
155.2155.1, 151.3, 150.7, 141.4, 139.2,
4.1.3.3. 5-Bromo-1H-benzo[d]imidazole (11c). A brown solid, yield
134.4, 134.0, 133.6, 132.0, 131.9, 125.7, 125.2, 124.2, 119.5, 119.0,
1
9
5%; R
f
¼ 0.1 (1:1, EtOAc:Petroleum ether); H NMR (CDCl
3
):
d
8.10
113.0, 112.5, 34.2, 29.4, 24.1, 21.8, 21.3, 20.7, 16.0, 15.9; ESI-MS m/z:
þ
(
(
s, 1H, CHeN), 7.87 (s, 1H, AreH), 7.52 (d, 1H, J ¼ 8.6 Hz, AreH), 7.46
414.0 ([MþH] ).
d, 1H, J ¼ 8.2 Hz, AreH).
4.1.4.3. 6-Bromo-2-methyl-1-(2,4,6-triisopropylphenylsulfonyl)-1H-
4
.1.3.4. 5-Bromo-2-methyl-1H-benzo[d]imidazole (11d). A light gray
benzo[d]imidazole (3a). A white solid, yield 17%; R ¼ 0.7 (1:4,
f
1
ꢀ
1
solid, yield 87%; R
CDCl ):
H, J ¼ 8.2 Hz, AreH), 2.62 (s, 3H, CH
f
¼ 0.1 (1:1, EtOAc:Petroleum ether); H NMR
7.67 (s, 1H, AreH), 7.39 (d, 1H, J ¼ 8.6 Hz, AreH), 7.32 (d,
).
EtOAc:Petroleum ether); mp 106e107 C; H NMR (CDCl ): d 7.80
3
(
1
3
d
(d, 1H, J ¼ 1.8 Hz, AreH), 7.69 (d, 1H, J ¼ 8.9 Hz, AreH), 7.42 (dd, 1H,
J ¼ 2.0, 8.9 Hz, AreH), 7.20 (s, 2H, AreH), 4.02e3.96 (m, 2H,
3
2
ꢁ CH(CH
3
)
2
), 2.96e2.90 (m, 1H, CH(CH
3
)
2
), 2.38 (s, 3H, ]CCH
3
),
C
1
3
4.1.3.5. 5(6)-Chloro-1H-benzo[d]imidazole (11e). A white solid,
1.26 (d, 6H, J ¼ 6.7 Hz, 2 ꢁ CH
3
),1.04 (d,12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
);
1
yield 100%; R
d, 1H, J ¼ 9.8 Hz, AreH), 8.11 (s, 2H, CH), 7.71 (d, 1H, J ¼ 8.6 Hz,
AreH), 7.66 (d, 2H, J ¼ 2.0 Hz, AreH), 7.58 (d, 2H, J ¼ 8.6 Hz, AreH),
f
¼ 0.3 (20:1, CH
2
Cl
2
:CH
3
OH); H NMR (CDCl
3
):
d
8.36
NMR (CDCl
3
):
d
155.6, 151.4, 142.5, 133.0, 131.6, 127.4, 124.3, 122.6,
þ
(
117.0, 114.3, 34.3, 29.5, 24.1, 15.8; ESI-MS m/z: 477.9 ([MþH] ).
13
4
.98 (br s, 2H, 2 ꢁ NH); C NMR (CDCl
3
):
d
142.3, 141.7, 138.1, 136.5,
4.1.4.4. 6-Bromo-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
135.7, 128.7, 126.4, 125.8, 116.4, 115.4, 29.6, 23.6.
imidazole (3b). A white solid, yield 13%; R ¼ 0.5 (1:4, EtOAc:Pe-
f
ꢀ
1
3
troleum ether); mp 168e169 C; H NMR (CDCl ): d 8.32 (s, 1H,
4
.1.3.6. 5-Chloro-2-methyl-1H-benzo[d]imidazole (11f). A white
CHeN), 7.64 (d, 1H, J ¼ 8.6 Hz, AreH), 7.44 (s, 1H, AreH), 7.42 (dd,
1H, J ¼ 2.0, 8.2 Hz, AreH), 7.24 (s, 2H, AreH), 4.19e4.12 (m, 2H,
2 ꢁ CH(CH ) ), 2.95e2.91 (m, 1H, CH(CH ) ), 1.26 (d, 6H, J ¼ 7.1 Hz,
1
solid, yield 100%; R
d
f
¼ 0.3 (20:1, CH
2
Cl
2
:CH
3
OH); H NMR (CDCl
3
):
7.53 (d, 1H, J ¼ 2.0 Hz, AreH), 7.44 (d, 1H, J ¼ 8.6 Hz, AreH), 7.20
3
2
3 2
13
1
3
(
d
dd, 1H, J ¼ 2.0, 8.2 Hz, AreH), 2.64 (s, 3H, CCH
3
); C NMR (CDCl
3
):
2 ꢁ CH ), 1.15 (d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH ); C NMR (CDCl ):
d 156.1,
3
3
3
152.0, 128.0, 122.9, 115.2, 114.5, 109.7, 15.0.
151.9, 141.0, 129.5, 127.7, 124.7122.2, 118.6, 115.0, 34.3, 29.6, 24.4;
þ
ESI-MS m/z: 463.9 ([MþH] ).
4
6
.1.3.7. 6-Methoxy-1H-benzo[d]imidazole (11g). A black solid, yield
1
8%; R
f
¼ 0.1 (1:1, EtOAc:Petroleum ether); H NMR (CDCl
3
):
d
8.03
4.1.4.5. 5-Bromo-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
(
s, 1H, CHeN), 7.56 (d, 1H, J ¼ 8.7 Hz, AreH), 7.11 (d, 1H, J ¼ 1.6 Hz,
imidazole (3c). A white solid, yield 20%; R ¼ 0.6 (1:4, EtOAc:Pe-
f
ꢀ
1
AreH), 6.94 (dd, 1H, J ¼ 2.0, 8.7 Hz, AreH), 3.84 (s, 3H, OCH ).
3
troleum ether); mp 166e167 C; H NMR (CDCl ):
d
8.29 (s, 1H,
3
CHeN), 7.94 (d, 1H, J ¼ 1.6 Hz, AreH), 7.39 (dd, 1H, J ¼ 1.6, 8.3 Hz,
4.1.3.8. 6-Methoxy-2-methyl-1H-benzo[d]imidazole (11h). A black
AreH), 7.22 (s, 2H, AreH), 7.18 (d, 1H, J ¼ 8.7 Hz, AreH), 4.13e4.09
1
solid, yield 92%; R
f
¼ 0.1 (1:1, EtOAc:Petroleum ether); H NMR
(m, 2H, 2 ꢁ CH(CH
3 2 3 2
) ), 2.94e2.90 (m, 1H, CH(CH ) ), 1.25 (d, 6H,
13
(
CDCl
3
):
d
7.42 (d, 1H, J ¼ 8.7 Hz, AreH), 7.02 (d, 1H, J ¼ 2.0 Hz,
J ¼ 7.1 Hz, 2 ꢁ CH
3
), 1.12 (d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
); C NMR
AreH), 6.86 (dd, 1H, J ¼ 2.4, 8.7 Hz, AreH), 3.83 (s, 3H, OCH
3
), 2.60
(CDCl
3
):
d
155.9, 151.7, 144.7, 141.6, 130.2, 128.3, 124.6, 117.5, 112.9,
þ
(
s, 3H, CH ).
3
34.3, 29.7, 24.4; ESI-MS m/z: 463.9 ([MþH] ).

H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
247
4
.1.4.6. 5-Bromo-2-methyl-1-(2,4,6-triisopropylphenylsulfonyl)-1H-
benzo[d]imidazole (3d). A white solid, yield 13%; R
¼ 0.6 (1:4,
EtOAc:Petroleum ether); mp 130e132 C; H NMR (CDCl ): 7.88
d, 1H, J ¼ 1.6 Hz, AreH), 7.51 (d, 1H, J ¼ 8.6 Hz, AreH), 7.42 (dd, 1H,
CHeN), 7.63 (d, 1H, J ¼ 8.6 Hz, AreH), 7.22 (s, 2H, AreH), 6.91 (dd,
1H, J ¼ 2.3, 8.6 Hz, AreH), 6.67 (d, 1H, J ¼ 2.3 Hz, AreH), 4.16e4.12
f
ꢀ
1
3
d
(m, 2H, 2 ꢁ CH(CH
CH(CH
4 ꢁ CH
3 2 3
) ), 3.68 (s, 3H, OCH ), 2.94e2.90 (m, 1H,
(
3
)
2
), 1.25 (d, 6H, J ¼ 6.7 Hz, 2 ꢁ CH
3
), 1.12 (d, 12H, J ¼ 6.7 Hz,
13
J ¼ 1.6, 8.6 Hz, AreH), 7.21 (s, 2H, AreH), 4.03e3.97 (m, 2H,
); C NMR (CDCl ): d 157.9, 155.6, 151.7, 139.3, 137.6, 132.0,
3 3
2
ꢁ CH(CH
3
)
2
), 2.95e2.92 (m, 1H, CH(CH
3
)
2
), 2.41 (s, 3H, ]CCH
3
),
C
130.2, 124.5, 121.4, 113.7, 95.0, 55.6, 34.3, 29.7, 24.4; ESI-MS m/z:
1
3
þ
1
.23 (d, 6H, J ¼ 7.1 Hz, 2 ꢁ CH
3
),1.06 (d,12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
);
415.9 ([MþH] ).
NMR (CDCl
3
):
d
155.7, 151.4, 140.2, 134.6, 131.4, 127.3, 124.4, 120.7,
þ
117.7, 116.1, 34.3, 29.5, 24.2, 23.4, 16.0; ESI-MS m/z: 477.9 ([MþH] ).
4.1.4.13. 5-Methoxy-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo
[
d]imidazole (5c). A gray solid, yield 38%; R
f
¼ 0.3 (1:5, EtOAc:Pe-
ꢀ
1
4
.1.4.7. 6-Chloro-2-methyl-1-(2,4,6-triisopropylphenylsulfonyl)-1H-
¼ 0.6 (1:4,
EtOAc:Petroleum ether); mp 140e141 C; H NMR (CDCl ): 7.74
d, 1H, J ¼ 9.0 Hz, AreH), 7.64 (d, 1H, J ¼ 2.0 Hz, AreH), 7.28 (dd, 1H,
troleum ether); mp 134e135 C; H NMR (CDCl ): 8.27 (s, 1H, ]
3
d
benzo[d]imidazole (4a). A light yellow solid, yield 12%; R
f
CHeN), 7.24 (d, 1H, J ¼ 2.3 Hz, AreH), 7.21 (s, 2H, AreH), 7.16 (d, 1H,
J ¼ 9.0 Hz, AreH), 6.89 (dd, 1H, J ¼ 2.3, 9.0 Hz, AreH), 4.17e4.13 (m,
ꢀ
1
3
d
(
2H, 2 ꢁ CH(CH
3
)
2
), 3.83 (s, 3H, OCH
3
), 2.94e2.90 (m, 1H, CH(CH
3
)
2
),
C
1
3
J ¼ 2.0, 9.0 Hz, AreH), 7.20 (s, 2H, AreH), 4.03e3.96 (m, 2H,
1.25 (d, 6H, J ¼ 7.0 Hz, 2 ꢁ CH
3
), 1.13 (d, 12H, J ¼ 7.0 Hz, 4 ꢁ CH
3
);
2
1
ꢁ CH(CH
.26 (d, 6H, J ¼ 7.0 Hz, 2 ꢁ CH
NMR (CDCl ):
3
)
2
), 2.95e2.91 (m, 1H, CH(CH
3
)
2
), 2.38 (s, 3H, ]CCH
3
),
C
NMR (CDCl ):
3
d
157.3, 155.5, 151.7, 144.5, 141.0, 130.2, 125.6, 124.5,
1
3
3
), 1.04 (d,12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
);
114.8, 112.0, 103.2, 55.7, 34.3, 29.6, 24.4; ESI-MS m/z: 415.9
þ
3
d
152.3, 151.4, 142.1, 132.6, 131.6, 129.6, 124.7, 124.3,
([MþH] ).
119.5, 113.4, 112.3, 34.3, 29.7, 24.2, 15.9; ESI-MS m/z: 433.9
þ
(
[MþH] ).
4.1.4.14. 5-Methoxy-2-methyl-1-(2,4,6-triisopropylphenyl-sulfonyl)-
1
H-benzo[d]imidazole (5d). A gray solid, yield 25%; R
f
¼ 0.4 (1:5,
ꢀ
1
4
.1.4.8. 6-Chloro-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
imidazole (4b). A light yellow solid, yield 27%; R
¼ 0.6 (1:4,
EtOAc:Petroleum ether); mp 155e157 C; H NMR (CDCl ): 8.31 (s,
H, CHeN), 7.77 (d, 1H, J ¼ 0.8 Hz, AreH), 7.25 (d, 1H, J ¼ 1.6 Hz,
AreH), 7.24 (s, 1H, AreH), 7.22 (s, 2H, AreH), 4.14e4.08 (m, 2H,
EtOAc:Petroleum ether); mp 103e104 C; H NMR (CDCl ): 7.51
3
d
f
(d, 1H, J ¼ 8.7 Hz, AreH), 7.19 (s, 2H, AreH), 7.18 (d, 1H, J ¼ 2.4 Hz,
AreH), 6.91 (dd, 1H, J ¼ 2.4, 8.7 Hz, AreH), 4.05e4.01 (m, 2H,
ꢀ
1
3
d
1
2 ꢁ CH(CH
3
)
2
), 3.78 (s, 3H, OCH
3
), 2.95e2.91 (m, 1H, CH(CH
3
)
2
),
), 1.05 (d, 12H,
3 3
); C NMR (CDCl ): d 157.4, 155.2, 151.3, 149.6,
2.42 (s, 3H,]CCH
3
), 1.27 (d, 6H, J ¼ 6.7 Hz, 2 ꢁ CH
3
13
2
2
1
ꢁ CH(CH
ꢁ CH ), 1.12 (d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH
00 MHz): 155.9,151.7,144.3,141.8,130.1,129.8,125.7,124.6,120.9,
3
)
2
), 2.94e2.90 (m, 1H, CH(CH
3
)
2
), 1.25 (d, 6H, J ¼ 7.0 Hz,
J ¼ 6.7 Hz, 4 ꢁ CH
13
3
3
); C NMR (CDCl
3
,
135.4, 134.5, 124.2, 119.9, 112.8, 97.0, 55.8, 34.2, 29.5, 24.2, 16.1; ESI-
þ
d
MS m/z: 430.0 ([MþH] ).
þ
112.5, 34.3, 29.7, 24.4; ESI-MS m/z: 420.0 ([MþH] ).
4.1.5. 6-Nitro-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
4
.1.4.9. 5-Chloro-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
¼ 0.5 (1:4,
EtOAc:Petroleum ether); mp 150e151 C; H NMR (CDCl ): 8.31 (s,
H, ]CHeN), 7.69 (d, 1H, J ¼ 9.0 Hz, AreH), 7.31e7.29 (m, 2H,
),
), 1.15
156.0, 151.9, 142.1,
imidazole (6b)
imidazole (4c). A light yellow solid, yield 29%; R
f
Compound 6b was prepared as a yellow solid in a similar
manner as described for compound 1 started from commercially
available 6-nitrobenzimidazole. Yield 46%; R
ꢀ
1
3
d
1
f
¼ 0.6 (1:4, EtOAc:-
ꢀ
1
AreH), 7.24 (s, 2H, AreH), 4.18e4.12 (m, 2H, 2 ꢁ CH(CH
3
)
2
Petroleum ether); mp 178e179 C; H NMR (CDCl ): 8.60 (s, 1H,
3
d
2
.95e2.92 (m, 1H, CH(CH
3
)
2
), 1.26 (d, 6H, J ¼ 6.7 Hz, 2 ꢁ CH
3
CHeN), 8.25 (dd, 1H, J ¼ 2.1, 8.9 Hz, AreH), 8.13 (d, 1H, J ¼ 1.8 Hz,
AreH), 7.89 (d, 1H, J ¼ 8.9 Hz, AreH), 7.27 (s, 2H, AreH), 4.20e4.13
13
(
1
d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
3
); C NMR (CDCl ): d
41.1, 131.1, 129.6, 125.1, 124.7, 121.8, 112.1, 34.3, 29.6, 24.4; ESI-MS
(m, 2H, 2 ꢁ CH(CH
3 2 3 2
) ), 2.97e2.90 (m, 1H, CH(CH ) ), 1.25 (d, 6H,
þ
13
m/z: 420.0 ([MþH] ).
J ¼ 7.0 Hz, 2 ꢁ CH
3
), 1.17 (d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
); C NMR
(
CDCl
3
):
d
156.7, 152.0, 147.7, 145.2, 144.7, 130.6, 129.0, 125.0, 121.3,
þ
4
.1.4.10. 5-Chloro-2-methyl-1-(2,4,6-triisopropylphenylsulfonyl)
120.0, 108.4, 34.3, 29.8, 24.4; ESI-MS m/z: 430.9 ([MþH] ).
-
(
d
1H-benzo[d]imidazole (4d). A light yellow solid, yield 15%; R ¼ 0.5
1:4, EtOAc:Petroleum ether); mp 111e112 C; H NMR (CDCl ):
7.76 (d, 1H, J ¼ 2.0 Hz, AreH), 7.56 (d, 1H, J ¼ 8.6 Hz, AreH), 7.29
f
ꢀ
1
3
4.1.6. 5-Nitro-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo[d]
imidazole (6c)
Compound 6c was prepared as a yellow solid in a similar
manner as described for compound 1 starded from commercially
available 6-Nitrobenzimidazole. Yield 42%; R
Petroleum ether); mp 158e160 C; H NMR (CDCl ): 8.70 (s, 1H,
(
2
dd, 1H, J ¼ 2.0, 8.2 Hz, AreH), 7.21 (s, 2H, AreH), 4.04e3.97 (m, 2H,
ꢁ CH(CH
3
)
2
), 2.95e2.92 (m, 1H, CH(CH
3
)
2
), 2.40 (s, 3H, ]CCH
3
),
C
1
3
1
.26 (d, 6H, J ¼ 7.1 Hz, 2 ꢁ CH
3
), 1.06 (d, 12H, J ¼ 7.1 Hz, 4 ꢁ CH
3
);
f
¼ 0.7 (1:4, EtOAc:-
ꢀ
1
NMR (CDCl
3
):
d
155.7, 151.6, 151.4, 124.6, 124.4, 120.3, 113.3, 34.3,
3
d
þ
2
9.5, 24.2, 16.0; ESI-MS m/z: 433.9 ([MþH] ).
CHeN), 8.44 (s,1H, AreH), 8.23 (d,1H, J ¼ 8.9 Hz, AreH), 7.45 (d,1H,
J ¼ 9.2 Hz, AreH), 7.25 (s, 2H, AreH), 4.12e4.05 (m, 2H,
4
1
.1.4.11. 6-Methoxy-2-methyl-1-(2,4,6-triisopropylphenyl-sulfonyl)-
H-benzo[d]imidazole (5a). A gray solid, yield 26%; R
¼ 0.3 (1:5,
): 7.65
d, 1H, J ¼ 9.1 Hz, AreH), 7.19 (s, 2H, AreH), 7.13 (d, 1H, J ¼ 2.8 Hz,
2 ꢁ CH(CH
3
)
2
), 2.97e2.91 (m, 1H, CH(CH
3
)
2
), 1.25 (d, 6H, J ¼ 7.0 Hz,
13
f
2 ꢁ CH
3
), 1.13 (d, 12H, J ¼ 6.7 Hz,4 ꢁ CH
3 3
); C NMR (CDCl ): d 156.5,
ꢀ
1
EtOAc:Petroleum ether); mp 109e110 C; H NMR (CDCl
(
3
d
151.9, 145.0, 143.6, 143.2, 135.2, 129.4, 124.8, 120.7, 117.5, 112.0, 34.3,
þ
29.8, 24.4; ESI-MS m/z: 430.9 ([MþH] ).
AreH), 6.91 (dd, 1H, J ¼ 2.8, 9.1 Hz, AreH), 4.05e4.01 (m, 2H,
2
2
ꢁ CH(CH
3
)
2
), 3.85 (s, 3H, OCH
3
), 2.93e2.89 (m, 1H, CH(CH
), 1.04 (d, 12H,
156.9, 155.2, 151.3, 142.2,
3
)
2
),
4.1.7. 1-(2,4,6-Triisopropylphenylsulfonyl)-1H-benzo[d]imidazole-
6-amine (7)
A suspension of 6b (97 mg, 0.23 mmol) and Pd/C (80 mg, 10%) in
2 2 2
CH Cl eMeOH (9:1, 20 mL) was stirred under H for 2 h at room
temperature and then filtered and concentrated. The residue was
purified by silica gel column chromatography (5:1, v/v, petroleum
ether/EtOAc) to give compound 7 (64.3 mg, 71%) as a yellow solid
.38 (s, 3H, ]CCH
3
), 1.27 (d, 6H, J ¼ 7.1 Hz, 2 ꢁ CH
3
13
J ¼ 6.7 Hz, 4 ꢁ CH
3
); C NMR (CDCl
3
):
d
128.3, 124.2, 113.5, 113.3, 102.3, 55.6, 34.2, 29.7, 24.2, 15.9; ESI-MS
þ
m/z: 430.0 ([MþH] ).
4
.1.4.12. 6-Methoxy-1-(2,4,6-triisopropylphenylsulfonyl)-1H-benzo
d]imidazole (5b). A gray solid, yield 40%; R
¼ 0.4 (1:5, EtOAc:Pe-
troleum ether); mp 152e153 C; H NMR (CDCl ): 8.23 (s, 1H,
ꢀ
1
[
f
with R
f
0.4 (1:4, EtOAc:Petroleum ether); mp 152e153 C; H NMR
8.06 (s, 1H, CHeN), 7.53 (d, 1H, J ¼ 8.6 Hz, AreH), 7.26 (s,
ꢀ
1
3
d
(CDCl ): d
3

248
H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
2
2
H, AreH), 6.69 (s, 1H, AreH), 6.66 (s, 1H, AreH), 4.18e4.14 (m, 2H,
ꢁ CH(CH ), 3.72 (s, 2H, NH ), 2.94e2.90 (m, 1H, CH(CH ), 1.26
by using the Protein Preparation and Grid Preparation tools in the
3
)
2
2
3
)
2
Schr o€ inger Maestro interface. In the grid preparation process, the
default settings were adopted for the cutoff, neutralization, scaling
and dimensions of the binding pocket. The standard precision (SP)
mode of GLIDE was used to explore the favorable binding poses.
Ligand conformation is allowed to be flexible while the protein is
held as a rigid structure during the docking process. After
screening, 300 compounds with the highest scores were selected.
Then these molecules were filtered by AUTODOCK3 program [24].
The center of the ligand BMS309403 was defined as the grid box
center. The grid size was set to 40 ꢁ 40 ꢁ 40 points with grid
spacing of 0.375 Å. The Lamarckian genetic algorithm was applied
to account for proteineligand interactions. According to the pre-
dicted binding conformations and their binding scores from GLIDE
and AUTODOCK3, 199 compounds were purchased.
13
(
d, 6H, J ¼ 6.7 Hz, 2 ꢁ CH
3
), 1.14 (d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH
3
);
C
NMR (CDCl
3
):
d
155.4, 151.6, 144.6, 138.5, 136.6, 132.5, 126.0, 124.5,
þ
121.4, 113.5, 97.1, 34.2, 29.5, 24.5; ESI-MS m/z: 400.9 ([MþH] ).
4.1.8. 2,4,6-Triisopropylbenzenesulfonamide (8)
To a solution of 2,4,6-triisopropylbenzenesulfonyl chloride
(
3 2
50 mg, 0.165 mmol) in THF (2 mL), NH $H O (0.127 mL, 25%) was
added under stirring. The mixture was stirred for 0.5 h at room
temperature until TLC indicated that the reaction was complete.
The solvent was concentrated under reduced pressure. The residue
was purified by silica gel column chromatography (1:5, v/v,
EtOAc:Petroleum ether) to give 8 (48.3 mg, 98%) as a white solid;
R
ꢀ
f
¼ 0.3 (1:4, EtOAc:Petroleum ether); mp 121e122 C(lit. [1]
ꢀ
1
1
19e119.6 C); H NMR (CDCl
NH
), 4.14e4.07 (m, 2H, 2 ꢁ CH(CH
3 3
CH(CH ); C NMR (CDCl ): d 152.8,
3
):
d
7.16 (s, 2H, AreH), 4.83 (s, 2H,
2
3 2
)
), 2.94e2.87 (m, 1H,
4.3. FABP4 inhibitory activity assays
1
3
3
)
2
), 1.29e1.24 (m, 18H, 6 ꢁ CH
ꢂ
149.1, 123.7, 34.2, 29.8, 24.7; ESI-MS m/z: 282.3 ([MꢂH] ).
The methods for expression, purification of FABP4 and the assay
for FABP4 inhibition have been described in our previous studies
11,28]. The compounds were prepared as 100 mM stock solution in
DMSO. Some of the compounds didn't dissolve well, then they were
prepared in DMSO as 10 mM stock solution. On the day of activity
assay, the compounds were all diluted in PBS to 100 mM for the first
round activity test. Compounds that showed inhibition rate higher
than 50% were selected for further IC50 values determination.
4
.1.9. 2,4,6-Triisopropylbenzenesulfonic acid (9)
[
2
,4,6-Triisopropylbenzenesulfonyl chloride (50 mg, 0.165 mmol)
was dissolved in a mixture of dioxane (5 mL) and water (10 mL) and
heated under reflux for 24 h. The solvent was concentrated under
reduced pressure. The residue was purified by silica gel column
chromatography (20:1, v/v, CH
as a white solid; R
¼ 0.2 (20:1, CH
NMR (DMSO):
2
2
d
2
Cl
2
:MeOH) to give 9 (41.6 mg, 89%)
ꢀ
1
f
2
2 3
Cl :CH OH); mp 260e262 C; H
d
6.93 (s, 2H, AreH), 4.59e4.54 (m, 2H,
), 2.80e2.76 (m, 1H, CH(CH
), 1.22 (d, 6H, J ¼ 7.0 Hz,
), 1.09 (d, 12H, J ¼ 6.7 Hz, 4 ꢁ CH
147.2, 146.8, 141.8, 121.4, 33.3, 28.0, 24.8; ESI-MS m/z: 283.1 ([M-
4.4. Analysis of adipocyte lipolysis
ꢁ CH(CH
3
)
2
3 2
)
13
ꢁ CH
3
3
); C NMR (DMSO):
The assay for evaluating the lipolysis of the compounds on 3T3-
-
L1 adipocytes was described in our previous study [28]. Briefly,
after mouse 3T3-L1 preadipocytes differentiated to adipocytes, the
compounds or DMSO (final concentration 0.2%) with different
concentrations were added to the 48-well plates for 24 h. On the
day of experimentation, cells were incubated in Krebs Ringer Hepes
H] ).
4
.1.10. 2,4,6-Triisopropylbenzoic acid (10)
To a solution of 2,4,6-triisopropylbenzoyl chloride (100 mg,
.375 mmol) in acetone (20 mL), Et N (54.2 L, 0.375 mmol) and
L, 0.188 mmol) were added under stirring. The mixture
0
3
m
buffer for 2 h, with or without 20
supernatants were measured using a commercial Glycerol assay kit
Applygen Technologies Inc, China) for lipolysis determination.
mM forskolin. Glycerol levels in
water (3.4
m
was refluxed for 18 h until TLC indicated that the reaction was
complete. The solvent was concentrated under reduced pressure.
The residue was purified by silica gel column chromatography
(
(
1:10 / 1:5, v/v, EtOAc:Petroleum ether) to give 10 (92 mg, 99%) as
a white solid;
¼ 0.8 (1:5, EtOAc:Petroleum ether); mp
86e187 C; H NMR (DMSO): 7.04 (s, 2H, AreH), 3.07e3.03 (m,
), 2.92e2.89 (m, 1H, CH(CH ), 1.30e1.25 (m, 18H,
4.5. Crystal structure determination
R
f
ꢀ
1
1
d
The purification and crystallization of human FABP4 was
referred to the protocol of Eric Marr et al. [19] with slight modifi-
cation. The cDNA of the full length FABP4 was cloned into pET28a
vector with the Nde I/Xho I restriction sites. The recombinant
protein was isolated by nickel-affinity chromatography, eluted with
2
6
3
H, 2 ꢁ CH(CH
3
)
2
3 2
)
1
3
ꢁ CH
3
); C NMR (DMSO):
d
171.4, 149.0, 143.3, 132.0, 120.4, 33.6,
ꢂ
0.8, 23.8; ESI-MS m/z: 247.3 ([MꢂH] ).
4.2. Virtual screening
2 4
50 mM NaH PO pH8.0, 300 mM NaCl, 80 mM imidazole and then
buffer exchanged into 20 mM HEPES pH7.5, 50 mM NaCl.
Crystals of FABP4 were obtained at 1.6 M Na-citrate at pH 6.5.
Apo-FABP4 was soaked with 1 mM inhibitor in the reservoir solu-
tion overnight. Crystals were flash frozen in liquid nitrogen in the
presence of well solution supplemented with 20% ethylene glycol
(EG).
The Data sets were collected at 100 K on the beamline BL17U at
the Shanghai Synchrotron Radiation Facility (SSRF), and were pro-
cessed with the XDS software packages [29]. The structures were
solved by molecular replacement, using the program PHASER [30]
with the search model of PDB code 3P6C. The structures were
refined using PHENIX [31]. Using the program Coot [32], inhibitor,
Virtual screening was employed to search for novel FABP4 in-
hibitors. Two chemical databases: Specs (http://www.specs.net)
and Maybridge (http://www.maybridge.com) were virtually
screened against the target protein FABP4. The X-ray structure of
FABP4 binding with BMS309403 (PDB code: 2NNQ [14]) was
retrieved from the Protein Data Bank (http://www.rcsb.org/pdb) for
docking calculation. For the screening of database Maybridge, the
procedure has been reported in our previous study [11]. 52 com-
pounds were finally purchased from the vendor.
For Specs, similar methods were used. First, we filtered the
200,000 compounds with our in-house script Druglk, leading to
~
8
0,000 compounds left. Then, software DOCK4.0 [21] was
water molecules were fitted into to the initial F
o
ꢂ F
c
map. The
employed, for constructing the grids for docking, residues within a
radius of 3.5 Å around BMS309403 were used. The top 7800
compounds were then selected for the second round screening by
GLIDE 4.0 [22,23] (www.schr o€ inger.com). Receptor was prepared
refined structures were deposited to Protein Data Bank with
accession codes 4NNS (FABP4-9) and 4NNT (FABP4-10). The com-
plete statistics, as well as the quality of the solved structure, is
shown in Table 3.

H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
249
Table 3
[4] B.M. Spiegelman, M. Frank, H. Green, Molecular cloning of mRNA from 3T3
adipocytes. Regulation of mRNA content for glycerophosphate dehydrogenase
and other differentiation-dependent proteins during adipocyte development,
J. Biol. Chem. 258 (16) (1983) 10083e10089.
Crystallographic data collection and refinement statistics.
PDB code
Space group
Cell dimensions: a (Å)
b (Å)
c (Å)
Wavelength (Å)
Reflections (unique)
Resolution range (Å)
Highest-resolution shell (Å)
Redundancy
I/s(I)
Completeness (%)
R merge
4NNS (FABP4-9)
P2
32.49
53.86
75.32
4NNT (FABP4-10)
P2
32.49
53.87
75.44
1
2
1
2
1
1
2
1
2
1
[5] C.R. Hunt, J.H. Ro, D.E. Dobson, H.Y. Min, B.M. Spiegelman, Adipocyte P2 gene:
0
developmental expression and homology of 5 -flanking sequences among fat
cell-specific genes, Proc. Natl. Acad. Sci. U. S. A. 83 (11) (1986) 3786e3790.
[6] G.S. Hotamisligil, R.S. Johnson, R.J. Distel, R. Ellis, V.E. Papaioannou,
B.M. Spiegelman, Uncoupling of obesity from insulin resistance through a
targeted mutation in aP2, the adipocyte fatty acid binding protein, Science 274
0.9793
0.9793
19,991(1441)
1.53e43.81
1.53e1.57
5.14(5.36)
27.95(9.73)
96.9(96.7)
4.3(21.5)
20,579(1482)
1.53e29.84
1.53e1.57
7.03(7.10)
21.54(6.31)
99.6(100)
(5291) (1996) 1377e1379.
[
7] K.T. Uysal, L. Scheja, S.M. Wiesbrock, S. Bonner-Weir, G.S. Hotamisligil,
Improved glucose and lipid metabolism in genetically obese mice lacking aP2,
Endocrinology 141 (9) (2000) 3388e3396.
[
8] L. Makowski, J.B. Boord, K. Maeda, V.R. Babaev, K.T. Uysal, M.A. Morgan,
R.A. Parker, J. Suttles, S. Fazio, G.S. Hotamisligil, et al., Lack of macrophage
fatty-acid-binding protein aP2 protects mice deficient in apolipoprotein E
against atherosclerosis, Nat. Med. 7 (6) (2001) 699e705.
10.0(50.3)
19.00/21.01
R
work/Rfree (%)
18.77/20.65
RMS values
Bond length (Å)
Bond angle ( )
Number of non-hydrogen atoms
[9] G. Tuncman, E. Erbay, X. Hom, I. De Vivo, H. Campos, E.B. Rimm,
G.S. Hotamisligil, A genetic variant at the fatty acid-binding protein aP2 locus
reduces the risk for hypertriglyceridemia, type 2 diabetes, and cardiovascular
disease, Proc. Natl. Acad. Sci. U. S. A. 103 (18) (2006) 6970e6975.
10] H. Lan, C.C. Cheng, T.J. Kowalski, L. Pang, L. Shan, C.C. Chuang, J. Jackson,
A. Rojas-Triana, L. Bober, L. Liu, et al., Small-molecule inhibitors of FABP4/5
ameliorate dyslipidemia but not insulin resistance in mice with diet-induced
obesity, J. Lipid Res. 52 (4) (2011) 646e656.
[11] H. Cai, G. Yan, X. Zhang, O. Gorbenko, H. Wang, W. Zhu, Discovery of highly
selective inhibitors of human fatty acid binding protein 4 (FABP4) by virtual
screening, Bioorg. Med. Chem. Lett. 20 (12) (2010) 3675e3679.
12] A.V. Hertzel, K. Hellberg, J.M. Reynolds, A.C. Kruse, B.E. Juhlmann, A.J. Smith,
M.A. Sanders, D.H. Ohlendorf, J. Suttles, D.A. Bernlohr, Identification and
characterization of a small molecule inhibitor of fatty acid binding proteins,
J. Med. Chem. 52 (19) (2009) 6024e6031.
[13] T. Barf, F. Lehmann, K. Hammer, S. Haile, E. Axen, C. Medina, J. Uppenberg,
S. Svensson, L. Rondahl, T. Lundback, N-Benzyl-indolo carboxylic acids: design
and synthesis of potent and selective adipocyte fatty-acid binding protein (A-
FABP) inhibitors, Bioorg. Med. Chem. Lett. 19 (6) (2009) 1745e1748.
[14] R. Sulsky, D.R. Magnin, Y. Huang, L. Simpkins, P. Taunk, M. Patel, Y. Zhu,
T.R. Stouch, D. Bassolino-Klimas, R. Parker, et al., Potent and selective biphenyl
azole inhibitors of adipocyte fatty acid binding protein (aFABP), Bioorg. Med.
Chem. Lett. 17 (12) (2007) 3511e3515.
0.005
1.080
0.006
1.075
ꢀ
[
Protein
1045
19
119
1034
18
127
Inhibitor
Water Oxygen
Others
Mean temperature factors (Å )
Protein
2
14.76
14.80
15.31
15.10
Inhibitor
[
Ramachandran plot
Residues in most-favored regions
Residues in allowed regions
Residues in disallowed regions
130(98.48%)
2(1.52%)
0(0.00%)
130(98.48%)
2(1.52%)
0(0.00%)
4.6. The possible binding mode analysis of compound 4d using
AUTODOCK4
The compound 4d was docked by AUTODOCK4 program [24].
[15] R. Ringom, E. Axen, J. Uppenberg, T. Lundback, L. Rondahl, T. Barf, Substituted
benzylamino-6-(trifluoromethyl)pyrimidin-4(1H)-ones: a novel class of se-
lective human A-FABP inhibitors, Bioorg. Med. Chem. Lett. 14 (17) (2004)
The center of the ligand 10 was defined as the grid box center. The
grid size was set to 60 ꢁ 60 ꢁ 60 points with grid spacing of
4449e4452.
0
.375 Å. The docking parameters were set as follows:
[16] F. Lehmann, S. Haile, E. Axen, C. Medina, J. Uppenberg, S. Svensson,
T. Lundback, L. Rondahl, T. Barf, Discovery of inhibitors of human adipocyte
fatty acid-binding protein, a potential type 2 diabetes target, Bioorg. Med.
Chem. Lett. 14 (17) (2004) 4445e4448.
[17] M.J. van Dongen, J. Uppenberg, S. Svensson, T. Lundback, T. Akerud,
M. Wikstrom, J. Schultz, Structure-based screening as applied to human
FABP4: a highly efficient alternative to HTS for hit generation, J. Am. Chem.
Soc. 124 (40) (2002) 11874e11880.
ga_pop_size
¼
150 (number of individuals in population),
ga_run ¼ 100 (the number of dockings that were performed), other
parameters were set to the software's default values. The La-
marckian genetic algorithm was applied to account for pro-
teineligand interactions.
[
[
[
18] M. Furuhashi, G. Tuncman, C.Z. G o€ rgün, L. Makowski, G. Atsumi,
E. Vaillancourt, K. Kono, V.R. Babaev, S. Fazio, M.F. Linton, et al., Treatment of
diabetes and atherosclerosis by inhibiting fatty-acid-binding protein aP2,
Nature 447 (7147) (2007) 959e965.
19] E. Marr, M. Tardie, M. Carty, T. Brown Phillips, I.K. Wang, W. Soeller, X. Qiu,
G. Karam, Expression, purification, crystallization and structure of human
adipocyte lipid-binding protein (aP2), Acta Crystallogr. Sect. F Struct. Biol.
Cryst. Commun. 62 (Pt 11) (2006) 1058e1060.
4
.7. Statistical analysis
The IC50 value for compound against FABPs was determined by
nonlinear regression using GraphPad Prism software (San Diego,
CA, USA). Statistical significance of the data was performed using a
one-way analysis of variance (ANOVA) followed by Dunnett's test.
p < 0.05 is considered statistically significant.
20] K. Hellberg, P.A. Grimsrud, A.C. Kruse, L.J. Banaszak, D.H. Ohlendorf,
D.A. Bernlohr, X-ray crystallographic analysis of adipocyte fatty acid binding
protein (aP2) modified with 4-hydroxy-2-nonenal, Protein Sci. 19 (8) (2010)
1480e1489.
[
[
21] T.J.A. Ewing, I.D. Kuntz, Critical Evaluation of Search Algorithms for Auto-
mated Molecular Docking and Database Screening, 1997.
22] R.A. Friesner, J.L. Banks, R.B. Murphy, T.A. Halgren, J.J. Klicic, D.T. Mainz,
M.P. Repasky, E.H. Knoll, M. Shelley, J.K. Perry, et al., Glide: a new approach for
rapid, accurate docking and scoring. 1. Method and assessment of docking
accuracy, J. Med. Chem. 47 (7) (2004) 1739e1749.
Acknowledgments
This work was supported by grants from National Natural Sci-
ence Foundation of China (81302699), the National Science and
Technology Major Project “Key New Drug Creation and
Manufacturing Program” (2012ZX09301-001-004), and the Na-
tional 863 High Performance Computing Project (2012AA01A305).
[23] Glide: A New Approach for Rapid, Accurate Docking and Scoring. 2. Enrich-
ment Factors in Database Screening.
[
24] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew,
A.J. Olson, Automated docking using a Lamarckian genetic algorithm and an
empirical binding free energy function, J. Comput. Chem. 19 (14) (1998)
1
639e1662.
References
[25] B. Binas, H. Danneberg, J. McWhir, L. Mullins, A.J. Clark, Requirement for the
heart-type fatty acid binding protein in cardiac fatty acid utilization, FASEB J.
13 (1999) 805e812.
[26] K. Maeda, H. Cao, K. Kono, C.Z. Gorgun, M. Furuhashi, K.T. Uysal, Q. Cao,
G. Atsumi, H. Malone, B. Krishnan, Y. Minokoshi, B.B. Kahn, R.A. Parker,
G.S. Hotamisligil, Adipocyte/macrophage fatty acid binding proteins control
integrated metabolic responses in obesity and diabetes, Cell Metab. 1 (2005)
107e119.
[
[
[
1] G.S. Hotamisligil, Inflammation and metabolic disorders, Nature 444 (7121)
2006) 860e867.
2] A. Chmurzynska, The multigene family of fatty acid-binding proteins (FABPs):
function, structure and polymorphism, J. Appl. Genet. 47 (1) (2006) 39e48.
3] N.H. Haunerland, F. Spener, Fatty acid-binding proteins e insights from ge-
netic manipulations, Prog. Lipid Res. 43 (4) (2004) 328e349.
(

250
H. Cai et al. / European Journal of Medicinal Chemistry 90 (2015) 241e250
[
27] N.R. Coe, M.A. Simpson, D.A. Bernlohr, Targeted disruption of the adipocyte
lipid-binding protein (aP2 protein) gene impairs fat cell lipolysis and increases
cellular fatty acid levels, J. Lipid. Res. 40 (1999) 967e972.
[30] A.J. McCoy, R.W. Grosse-Kunstleve, P.D. Adams, M.D. Winn, L.C. Storoni, R.J. Read,
Phaser crystallographic software, J. Appl. Crystallogr. 40 (Pt 4) (2007) 658e674.
[31] P.D. Adams, P.V. Afonine, G. Bunkoczi, V.B. Chen, I.W. Davis, N. Echols,
J.J. Headd, L.W. Hung, G.J. Kapral, R.W. Grosse-Kunstleve, et al., PHENIX: a
comprehensive Python-based system for macromolecular structure solution,
Acta Crystallogr. D Biol. Crystallogr. 66 (Pt 2) (2010) 213e221.
[
28] H.Y. Cai, T. Wang, J.C. Zhao, P. Sun, G.R. Yan, H.P. Ding, Y.X. Li, H.Y. Wang,
W.L. Zhu, K.X. Chen, Benzbromarone, an old uricosuric drug, inhibits human
fatty acid binding protein 4 in vitro and lowers the blood glucose level in db/
db mice, Acta Pharmacol Sin. 34 (11) (2013) 1397e1402.
[32] P. Emsley, B. Lohkamp, W.G. Scott, K. Cowtan, Features and development of
Coot, Acta Crystallogr. D Biol. Crystallogr. 66 (Pt 4) (2010) 486e501.
[29] W. Kabsch, XDS, Acta Crystallogr. D66 (2010a) 125e132.