Mechanism of retinoid X receptor partial agonistic action of 1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)-1H-benzotriazole-5- carboxylic acid and structural development to increase potency

Article abstract of DOI:10.1021/jm400033f

We have reported that retinoid X receptor (RXR) partial agonist 1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)-1H-benzotriazole-5- carboxylic acid (CBt-PMN, 4a) shows a significant antidiabetes effect in the KK-A^y type 2 diabetes model

Full text of DOI:10.1021/jm400033f

Article
pubs.acs.org/jmc
Mechanism of Retinoid X Receptor Partial Agonistic Action of
1‑(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)‑
1H‑benzotriazole-5-carboxylic Acid and Structural Development To
Increase Potency
Fuminori Ohsawa,† Shoya Yamada,†,‡ Nobumasa Yakushiji,† Ryosuke Shinozaki,† Mariko Nakayama,†
Kohei Kawata,† Manabu Hagaya,† Toshiki Kobayashi,† Kazutaka Kohara,† Yuuki Furusawa,†
Chisa Fujiwara,† Yui Ohta,† Makoto Makishima,§ Hirotaka Naitou,∥ Akihiro Tai,⊥ Yutaka Yoshikawa,#
Hiroyuki Yasui,# and Hiroki Kakuta*,†
†Division of Pharmaceutical Sciences, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences,
1-1-1, Tsushima-Naka, Kita-Ku, Okayama 700-8530, Japan
‡Research Fellowship Division, Japan Society for the Promotion of Science, Sumitomo-Ichibancho FS Bldg., 8 Ichibancho,
Chiyoda-ku, Tokyo 102-8472, Japan
§Division of Biochemistry, Department of Biomedical Sciences, Nihon University School of Medicine, 30-1 Oyaguchi-kamicho,
Itabashi-ku, Tokyo 173-8610, Japan
∥Graduate School of Nutritional and Environmental Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan
⊥Faculty of Life and Environmental Sciences, Prefectural University of Hiroshima, 562 Nanatsuka-Cho, Shobara,
Hiroshima 727-0023, Japan
#Department of Analytical and Bioinorganic Chemistry, Division of Analytical and Physical Chemistry,
Kyoto Pharmaceutical University, 5 Nakauchi-cho, Misasagi, Yamashina-ku, Kyoto 607-8414, Japan
S
* Supporting Information
ABSTRACT: We have reported that retinoid X receptor (RXR)
partial agonist 1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydro-2-naph-
thyl)-1H-benzotriazole-5-carboxylic acid (CBt-PMN, 4a) shows a
significant antidiabetes effect in the KK-Ay type 2 diabetes model
mice, with reduced side effects compared to RXR full agonists.
To elucidate the mechanism of the RXR partial agonist activity of
4a, we synthesized derivatives of 4a, evaluated their RXR agonist
activity, and performed structure−activity relationship analysis.
Reporter gene assay revealed that though 6b, which possesses an
amino group at the 2-position of 5-carboxybenzimidazole, showed
RXR full-agonist activity, compounds 6d and 6e, which possess
an oxygen atom and a sulfur atom at the corresponding position,
respectively, showed weak RXR agonist activity. On the other
hand, 6c, which has a trifluoromethyl group at the corresponding position, acts as an RXR partial agonist, having similar
E
max (67 2%) and lower EC50 (15 0 nM) compared to those of 4a (Emax = 75 4%, EC50 = 143 2 nM). A fluorescence
polarization assay of cofactor recruitment confirmed that fluorescein-labeled D22 coactivator peptide was less efficiently recruited
to RXR by 4a and 6c than by LGD1069 (1), a known RXR full agonist. Electrostatic potential field calculations and
computational docking studies suggested that full agonists show an electrostatic attraction, which stabilizes the holo structure and
favors coactivator recruitment, between the side chain at the benzimidazole 2-position and the α-carbonyl oxygen of asparagine-
306 in helix 4 (H4) of the RXR receptor. However, RXR partial agonists 4a and 6c lack this interaction. Like 4a, 6c showed a
significant antidiabetes effect in KK-Ay type 2 diabetes model mice with reduced levels of the side effects associated with RXR full
agonists. These findings should aid the design of new RXR partial agonists as antitype 2 diabetes drug candidates.
receptors (PPARs), or liver X receptors (LXRs).1−3 Interestingly,
INTRODUCTION
■
so-called permissive heterodimers, such as PPAR/RXR4−6 and
Retinoid X receptors (RXRs) are nuclear receptors that act as
ligand-dependent transcription factors and function as the homo-
dimer or as heterodimers with other nuclear receptors, such as
retinoic acid receptors (RARs), peroxisome proliferator-activated
Received: October 3, 2012
Published: February 7, 2013
© 2013 American Chemical Society
1865
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
LXR/RXR,7−10 which regulate glucose/lipid metabolism and
immune response, can be activated by RXR agonists alone.11
Therefore, RXR agonists are considered to be candidate
therapeutic agents for the treatment of type 2 diabetes and
autoimmune disease. In addition, synthetic RXR agonists are
being used as therapies for other indications. For example,
bexarotene (1) is approved for treatment of subcutaneous
T-cell lymphoma (STCL) in the United States.12 Although several
RXR agonists have been created (Figure 1),13−17 they were
critical for RXR agonistic activity. Therefore, we focused on modi-
fication of the heterocyclic ring of 4a at the 2-position (Figure 2).
Figure 2. Molecular design strategy for creating derivatives of 4a.
Indeed, we found that 6c, which possesses a trifluoromethyl
group at the benzimidazole 2-position, has similar Emax and
lower EC50 values compared to those of 4a, showing that 6c
is a more potent RXR partial agonist than 4a. To interpret
the structure−activity relationship findings, we carried out elec-
trostatic potential field calculations and computational docking
studies. The results indicated that docked RXR full agonists
exhibit an electrostatic attraction between the side chain at
the benzimidazole 2-position and the α-carbonyl oxygen of
asparagine-306 in helix 4 (H4) of the RXR receptor. Such an
interaction would serve to stabilize the holo structure, favoring
coactivator recruitment. However, RXR partial agonists 4a and
6c lack this interaction. We also confirmed that 6c shows a
significant antidiabetes effect in KK-Ay type 2 diabetes model
mice with reduced levels of the side effects associated with RXR
full agonists.
Figure 1. Chemical structures of known RXR ligands 1−5.
found to induce body weight gain,18 hepatomegaly,19 blood
triglyceride (TG) elevation,20 and other adverse effects.
However, it has been reported that RXR agonists show
different patterns of RXR heterodimer activation.21 Further,
when we evaluated the therapeutic and side effects of NEt-
TMN (2), NEt-3IB (3a), and NEt-3IP (3b) in KK-Ay type 2
diabetes model mice, we found similar activation of RXR
homodimer but different patterns of RXR heterodimer
activation.22 These compounds show therapeutic effects, but
also show some of the side effects reported for other RXR
agonists.
All RXR agonists that exhibit severe side effects are RXR full
agonists, which activate RXR completely. Thus, in our previous
work, we hypothesized that there is a difference in RXR
activation threshold for the appearance of the therapeutic
effects and the side effects. In other words, we considered that
RXR partial agonists, whose maximum activation of RXR is
lower than that of RXR full agonists, might induce the desired
medicinal effects without (or at least with reduced levels of) the
side effects that are associated with RXR full agonists. To test
this idea, we designed and synthesized RXR partial agonists
based on 1 or 2 by linking the acidic domain and the linking
domain of 1 or 2, with the aim of restricting the molecular
flexibility. This approach yielded the RXR partial agonist CBt-
PMN (4a), which indeed showed a significant antidiabetes
effect in KK-Ay type 2 diabetes model mice but with reduced
levels of the side effects associated with RXR full agonists.16
However, the reason why 4a acts as an RXR partial agonist was
not established.
CHEMISTRY
■
Compounds were synthesized as illustrated in Schemes 1 and 2.
Compound 10 was synthesized by nitration and hydrogen
reduction from 8, the synthesis of which has already been
reported.16 On the other hand, compound 13 was obtained
from 4-iodobenzoic acid (11) by nitration and protection of the
carboxyl group as the methyl ester. Compound 14 was
synthesized from 10 and 13 by Buchwald coupling reaction,
and the common intermediate 15 was obtained from 14 by
hydrogen reduction. Compounds 16a−e were synthesized by a
ring-closing reaction using formic acid, cyanogen bromide,
trifluoroacetic anhydride, triphosgene, and carbon disulfide,
respectively. Hydrolysis of the ester group under alkaline
conditions afforded the desired products 6a−e (Scheme 1).
Compound 7 was obtained by alkaline saponification from 19,
which was synthesized from 17 (obtained as previously reported23)
and 18 by Ullmann condensation (Scheme 2).
RESULTS AND DISCUSSION
■
The compounds obtained were evaluated by reporter gene
assay using COS-1 cells (Table 1). Among the compounds
bearing a side chain at the 2-position in the benzimidazole ring,
amino derivative 6b activated RXRα as potently as 2 and 4b, as
previously reported.16 Compounds 6d and 6e showed low
efficacy, even at 10 μM. On the other hand, compound 6a,
which has no side chain at the 2-position in the benzimidazole
ring, as well as trifluoromethyl derivative 6c and compound 7
In the present work, we aimed to elucidate the mechanism of
RXR partial agonistic activity of 4a, and for this purpose we
synthesized a series of derivatives of 4a for structure−activity
relationship analysis. We noted that CBiM-PMN (4b),16 which
possesses a methyl group at the benzimidazole 2-position,
shows RXR full agonistic activity, and thus we considered that
the heterocyclic moiety, especially the 2-position, might be
1866
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
a
Scheme 1
a
(a) HNO3, Ac2O, 72%. (b) H2, Pd−C, EtOAc, 84%−qy. (c) HNO3, H2SO4, 90%. (d) H2SO4, MeOH, qy. (e) Pd2(dba)3, rac-BINAP, Cs2CO3,
toluene, qy. (f) HCO2H, 93%. (g) BrCN, THF, 26%. (h) TFAA, TFA, 95%. (i) Triphosgene, TEA, DCE, qy. (j) CS2, DBU, DMF, 93%. (k) (1) 2 N
NaOH, MeOH, THF; (2) 2 N HCl, 43%−qy. (1) 2 N NaOH, MeOH; (2) 2 N HCl, 86%−qy.
a
Scheme 2
a
(a) CuI, DMEDA, K3PO4, toluene, 7.9%. (b) (1) 2 N NaOH, MeOH, THF; (2) 2 N HCl, 92%.
possessing an indole structure, showed lower Emax values than
those of 2, 4b, and 6b.
6a, 6c, and 7 exhibit RXR agonist activity (Figure 3a). How-
ever, since only 4a and 6c behaved as competitive antagonists
in the presence of 1 (Figure 3b), only these two compounds
were considered to be RXR partial agonists.
The ligand-dependent transcription of RXR dimers is
induced by conformational change of the receptor from the
apo-form (a ligand-free form that recruits a corepressor) to the
holo-form (an agonist-binding form that recruits a coactiva-
tor).26 RXR partial agonists are thought to recruit both
coactivators and corepressors, acting also as RXR partial
antagonists, and it is thought that the recruitment of each type
of cofactor by partial agonists is less efficient than that by full
According to the review by Pierre Germain et al., nuclear
receptor partial agonists are defined as ligands that are potent
but exhibit reduced efficacy when compared with full agonists.24
These compounds can also induce antagonist conformation.
For example, PPARγ partial agonist GW0072 is reported to
act as a competitive antagonist of rosiglitazone, a PPAR full
agonist, despite exhibiting weak PPARγ agonist activity.25 Thus,
to examine whether 6a, 6c, and 7 are RXR partial agonists, we
evaluated their competitive antagonistic activity against 1 μM 1.
The dose−response curves in the absence of 1 showed that 4a,
1867
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
Table 1. Cotransfection Data for 2, 4a, 4b, 6a−e, and 7 towards RXRα in COS-1 Cells and Substituent Constants at the
Heterocyclic 2-Position
a,b
efficacy (%)
properties of R
c
d
e
compd
at 0.1 μM
at 1 μM
at 10 μM
EC50 (nM)
3.8 0.2
Emax (%)
Es
π
2
86
2
2
1
96
75
70
52
35
65
NA
NA
67
4
3
4
3
3
1
98
69
80
72
96
71
22
37
77
3
3
5
4
4
6
2
0
4
96
75
94
75
103
67
−
4
4
5
3
−
−
−
−
4a
4b
6a
6b
6c
6d
6e
7
24
143
2
28
367 130
633 33
155 10
−1.24
0.00
−0.61
−2.40
−
0.56
0.00
−1.23
0.88
−
NA
NA
21
2
2
3
15
−
0
2
NA
NA
45
−
−
−
−
5
83
8
73
2
0.00
0.00
a
The relative transactivation activity is based on the luciferase activity of 1 μM 1 (RXR full agonist) taken as 100%. All values represent the mean
b
c
value of at least three separate experiments with triplicate determinations. NA means nonactive (the value is less than 5). EC50 values were
determined from full dose−response curves. Es: Taft’s steric substituent constant (steric effect). π: lipophilicity. These data were cited from
Hansch, C.; Leo, A. Substituent Constants for Correlation Analysis in Chemistry and Biology; Wiley-Interscience: New York, 1979.
d
e
Figure 3. Relative transactivation activities of 2, 4a, 6a, 6c, and 7 toward RXRα. (a) Dose-dependence of RXRα agonist activities of 2 (open circles),
4a (open triangles), 6a (closed circles), 6c (closed triangles), and 7 (closed squares). (b) Dose-dependence of RXRα antagonist activities of 4a
(open triangles), 6a (closed circles), 6c (closed triangles), 7 (closed squares), and PA452 (closed diamond) in the presence of 1 μM 1. The
transactivation activity is shown as relative activity based on the luciferase activity of 1 μM 1 taken as 1.0. Error bars are SEM.
Figure 4. Changes in fluorescence polarization of (a) 5 nM fluorescein-labeled coactivator peptide D22 or (b) 5 nM fluorescein-labeled corepressor
peptide SMRT-ID2 induced by 6c (putative partial agonist), compound 1 (RXR full agonist), and PA452 (RXR antagonist). Fluorescence
polarization values, expressed in mP, are the mean SEM of measurements obtained from triplicate wells.
agonists or full antagonists. Thus, we examined cofactor re-
cruitment by our putative RXR partial agonists. Cofactor
recruitment assay was performed using fluorescence polar-
ization measurements with RXR receptor, each ligand and
coactivator and RXR receptor should result in a higher FP
value than that in its absence.
As coactivator and corepressor peptides, we selected D2229
and silencing mediator for retinoid and thyroid hormone
receptors interaction domain 2 (SMRT-ID2),30 respectively.
D22 has been used for time-resolved fluorescence resonance
energy transfer (TR-FRET) assay with RXR.29 SMRT-ID2 is
reported to bind to apo-RXR.30 First, to determine the appro-
priate RXR concentration for this assay, we evaluated FP values
at various RXR concentrations with each fluorescein-labeled
cofactor (5 nM) in the absence or presence of 1 (1 μM). As the
́
fluorescein-labeled cofactors according to Levy-Bimbot et
al.27,28 If a fluorescein-labeled cofactor does not bind to the
receptor, the degree of fluorescence polarization (FP) will be
low, because the fluorescein-labeled cofactor retains high
mobility. However, binding of the cofactor to the receptor
reduces the mobility and induces a high FP value. Thus,
addition of an agonist to a mixture of fluorescein-labeled
1868
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
Figure 5. Calculated electrostatic potential distribution of 4a, 4b, and 6a−e.
RXR concentration was increased, the FP value also increased,
indicating the binding of each peptide with RXR (Supporting
Information, Figure S1). In the presence of full agonist 1, the
plot for fluorescein-labeled D22 was shifted to the left, in-
dicating that 1 induced interaction of D22 with RXR. On the
other hand, the plot for fluorescein-labeled SMRT-ID2 was
shifted to the right in the presence of 1, indicating that 1
decreased the interaction between SMRT-ID2 and RXR. The
RXR concentration showing the largest change of FP by 1 was
1.0 mg/mL for fluorescein-labeled D22 or 0.1 mg/mL for
SMRT-ID2, and therefore these RXR concentrations were used
in subsequent experiments. Using these conditions, we com-
pared the cofactor recruitment ability of RXR partial agonists
4a and 6c with those of RXR full agonist 1 and RXR antagonist
PA452.31 Each compound was used at 10 μM, because we
found no difference in recruitment of fluorescein-labeled D22
between 10 and 33 μM concentrations (Supporting Information,
Figure S2). As shown in Figures S2 (Supporting Information)
and 4, the coactivator recruitment ability of both RXR partial
agonists was intermediate between those of RXR full agonist 1
and RXR antagonist PA452. On the other hand, the corepressor
recruitment ability of RXR partial agonist 6c was less than that
of PA452 and similar to that of RXR full agonist 1. Taken
together with the results of reporter gene assay and cofactor
recruitment assay, these findings confirm that 6c is a RXR
partial agonist, like 4a.
To examine the effects of substituents at the benzimidazole
2-position, we performed structure−activity relationship
analysis of 2, 4a, 4b, and 6a−c (Table 1). Considering their
Taft’s steric substituent constants (Es) and lipophilicity (π), we
found that these two constants appeared not to be related to
RXR agonistic activity, indicating that electronic properties at the
benzimidazole 2-position are critical for RXR agonistic activity.
Thus, to investigate the spatial electronic properties of these
compounds, we calculated the electrostatic potential field using
Spartan 10 (Figure 5). The RXR full agonists 4b and 6b have
a positive electrostatic potential field at the benzimidazole
2-position, while RXR partial agonist 6c has a weak negative
electrostatic potential field at that position. On the other hand,
6d and 6e, which showed weak RXR agonistic activity, have a
stronger negative electrostatic potential field at the benzimida-
zole 2-position than 6c. These results suggest that a positive
electrostatic potential field at the benzimidazole 2-position, as
in 4b and 6b, is important to stabilize the holo structure,
thereby favoring coactivator recruitment for RXR full agonistic
activity.
Next, to examine the interaction between the electrostatic
potential field around the heterocyclic 2-position and RXRα,
we performed a docking study using AutoDock 4.232 (Figure 6
Figure 6. (a) Docking model of RXR partial agonist 4a in the ligand-
binding pocket of RXRα (PDB code 3H0A). H4, H11, and H12 mean
helix 4 (jade green), helix 11 (orange), and helix 12 (red), respectively.
(b) Docking model of RXR full agonist 4b in the ligand-binding
pocket of RXRα. (c) Docking model of RXR full agonist 6b in the
ligand-binding pocket of RXRα. (d) Docking model of RXR partial
agonist 6c in the ligand-binding pocket of RXRα. The asterisk
indicates electrostatic repulsion.
and Table 2). The results reveal that the 2-position of the
heterocyclic ring in 4a, 4b, 6b, and 6c is close to the α-carbonyl
oxygen of asparagine-306 (Asn306) in helix 4 (H4; jade green).
Thus, full agonists 4b and 6b, which have a positive
electrostatic potential field at the 2-position of the heterocyclic
ring, may have an attractive interaction with Asn306 in H4. In
turn, this interaction may promote the proper folding of helix
12 (H12; red), stabilizing the holo form and thus favoring
recruitment of a coactivator. On the other hand, 6d and 6e,
which have a strong negative electrostatic potential field at the
1869
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
other RXR subtypes, as well as RARs, PPARγ, LXRα, PPARγ/
RXRα, and LXRα/RXRα, by means of reporter gene assay.
Compound 6c showed partial agonistic activities toward RXRβ
and RXRγ, but caused little activation of RARs (Supporting
Information, Figure S3). In addition, 6c showed no PPARγ
agonistic activity but activated LXRα, LXRα/RXRα, and
PPARγ/RXRα (Supporting Information, Figure S4). Com-
pounds 4a and 6c also showed RXR partial agonist activity
toward mouse RXRα (Supporting Information, Figure S5), as
well as human RXRα. Thus, 6c can be expected to show
antidiabetic effects in mice, similar to 4a.
Table 2. Docking Score and Distance between the
α-Carbonyl Oxygen of Asn306 and the Closest Atom
at the Heterocyclic 2-Position of Each Compound
docking
score
mean binding energy
(kcal/mol)
distance
(Å)
compd
atom
2
−
−
−
−
N
H
H
H
F
4a
4b
6a
6b
6c
6d
6e
7
100/100
100/100
100/100
100/100
100/100
100/100
100/100
100/100
−11.92
−12.13
−11.45
−11.72
−11.41
−11.49
−11.97
−11.67
4.42
2.67
3.85
2.66
2.51
3.88
3.39
3.88
Since 6c shows similar receptor activation patterns to 4a,16
we were interested in the anti-type 2 diabetes effect and the side
effects of 6c. First, we confirmed that single oral administration
of 6c at 30 mg/kg resulted in a blood concentration over 6 μM
in ICR male mice (Supporting Information, Figure S6). Then,
to evaluate side effects associated with RXR full agonists,
changes in body weight, hepatomegaly, blood triglyceride
(TG), and blood total cholesterol (TCHO) were assessed
when 6c, RXR full agonist 2, and 4b were administered orally
to ICR mice at 30 mg/kg/day for 7 days (Figure 7). Despite
developing slight hepatomegaly, the group treated with 6c
showed a similar body weight gain to the control group and did
not show blood TG or TCHO elevation, whereas the group
treated with 2 or 4b showed significant increases of body weight
gain and blood TG level compared with the control. We found no
significant differences of renal function parameters between the
treated and the vehicle control groups (Supporting Information,
Table S2). Since liver function parameters (in particular,
alkaline phosphatase, ALP) were elevated, the hepatomegaly
may be associated with hepatotoxicity related to the chemical
O
S
H
2-position of the heterocyclic ring, would exhibit electrostatic
repulsion between the 2-position of the heterocyclic ring and
Asn306. Such electrostatic repulsion may reduce the affinity for
RXR and lead to a steric clash between Asn306 in H4 and
glycine-429 (Gly429) in helix 11 (H11; orange), thereby
unstabilizing the holo form. On the other hand, compounds 4a
and 6c lack the electrostatic attractive force because these
compounds possess a weak negative electrostatic field around
the 2-position of the heterocyclic ring. Thus, RXR partial
agonists might be less able to induce stable folding of H12 in
RXR for coactivator recruitment, as compared to RXR full
agonists. These results suggest that the nature of the interaction
between a ligand and H4 is important for determining whether
the ligand exerts RXR full agonistic or partial agonistic activity.
We found that 6c is a more potent RXR partial agonist
than 4a. Thus, we next examined the activities of 6c toward
Figure 7. Evaluation of adverse effects of repeated oral administration of compounds at 30 mg/kg/day to male ICR mice for 7 consecutive days.
(a) Time course of body weight change. Diamond, circles, open squares, and closed squares indicate vehicle, 2, 4b, and 6c, respectively. Effects of
compounds on (b) liver weight gain, (c) serum triglyceride, and (d) total cholesterol, respectively. The data (n = 7−23) represent the mean SEM.
Statistical analysis was performed by analysis of variance (ANOVA). Significant difference: *p < 0.05 vs vehicle. **p < 0.01 vs vehicle.
1870
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
Figure 8. Evaluation of anti-type 2 diabetes effects of repeated oral administration of 6c or pioglitazone at 10 mg/kg/day to male KK-Ay mice for 14
consecutive days. Time course of (a) body weight change, (b) blood glucose levels. (c) food intake, (d) water intake, and (e) oral glucose tolerance
test (OGTT) results. (f) Liver weight gain in male KK-Ay mice treated with vehicle and compounds. Inverted triangles, closed squares, and asterisks
indicate vehicle, 6c, and pioglitazone, respectively. The data (n = 4−5) represent the mean SEM. Statistical analysis was performed by analysis of
variance (ANOVA). Significant difference: *p < 0.05, 6c vs vehicle. **p < 0.01, 6c vs vehicle. †p < 0.05, pioglitazone vs vehicle. ††p < 0.01,
pioglitazone vs vehicle.
structure of 6c. Nevertheless, the hepatomegaly induced by
6c is clearly less marked than that induced by RXR full
agonist 2.
Table 3. Plasma Parameters of Male KK-Ay Mice after Oral
Administration of Vehicle, 6c, or Pioglitazone at 10 mg/kg/day
for 14 Consecutive Days
a
Further, when 6c was administered orally to male and female
SD rats at 30 mg/kg/day for 28 days, we observed hepato-
toxicity similar to that in the case of administration to ICR mice
for 7 days, together with slight enlargement of the spleen, but
there was no significant difference in body weight change or in the
weights of other organs (Supporting Information, Figure S7 and
Tables S3 and S4). As for blood constituents, some significant
differences from the vehicle were observed, but the data, other
than liver function parameters, were all within the ranges con-
sidered normal by the suppliers.
Next, we examined the antitype 2 diabetes activity in
KK-Ay mice. Compound 6c or pioglitazone was orally admin-
istered at 10 mg/kg/day for 14 days (Figure 8). The group
treated with 6c did not show significant improvement in the
oral glucose tolerance test (OGTT) compared to the control
group (Figure 8e), but the average blood glucose level in the
group treated with 6c was significantly reduced at days 4−6,
9, and 12 (Figure 8b). Compound 6c also reduced water
vehicle
pioglitazone
6c
AST (U/I)
49.6 2.3
17.6 0.9
5.8 0.6
45.3 8.4
21.3 1.9*
7.3 2.3
42.3 2.9
17.3 2.6
6.3 0.3
ALT (U/I)
γ-GTP (U/I)
ALP (U/I)
358.2 43.1
DL
395.7 44.1
DL
554.0 113.2*
DL
CRE (mg/dL)
BUN (mg/dL)
TG (mg/dL)
TCHO (mg/dL)
21.9 1.1
122.4 11.0
153.2 11.1
22.2 1.3
124.0 28.5
147.3 3.8
22.5 1.1
185.0 38.0
155.3 10.7
a
The data (n = 3−5) represent the mean SEM. Statistical analysis
was performed by analysis of variance (ANOVA). Significant
difference: *p < 0.05 vs vehicle. **p < 0.01 vs vehicle. DL means
detection limit (0.2 mg/dL).
intake significantly at days 3−12 compared to vehicle-treated
mice (Figure 8d), thus showing significant anti-type 2 diabetes
effects.
1871
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
according to ref 16. HPLC: 13.8 min; >95% purity (MeOH:AcON-
H4(aq) = 80:20).
2-Methyl-1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydro-2-naph-
thyl)-1H-benzoimidazole-5-carboxylic Acid (4b). This compound
was prepared according to ref 16. HPLC: 14.0 min; >95% purity
(MeOH:AcONH4(aq) = 80:20).
1,1,4,4,6-Pentamethyl-7-nitro-1,2,3,4-tetrahydronaphthalene
(9). To an ice-cooled solution of 8 (2.0 g, 9.9 mmol) in Ac2O (10 mL)
was added dropwise concd HNO3 (0.75 mL). The reaction mixture
was poured into ice and extracted with EtOAc (2 × 50 mL). The
organic layer was washed with H2O (2 × 50 mL) and brine (50 mL),
dried over MgSO4, and evaporated under reduced pressure. The
residue was recrystallized from EtOAc/hexane to yield 4.5 g of 9 as a
pale yellow powder (72%). 1H NMR (300 MHz, CDCl3) δ: 7.96
(s, 1H), 7.21 (s, 1H), 2.56 (s, 3H), 1.70 (s, 4H), 1.30 (s, 6H), 1.29
(s, 6H).
These effects also may result partially or entirely from activa-
tion of PPAR/RXR or LXR/RXR activation by 6c. Moreover,
6c did not elevate blood TG or blood TCHO significantly
(Table 3). These results support our hypothesis of a threshold
difference for the therapeutic and side effects of RXR agonists
and indicate that RXR partial agonists represent a promising
class of candidate anti-type 2 diabetes agents with reduced
levels of the serious adverse effects caused by RXR full agonists.
CONCLUSION
■
In order to elucidate the mechanism of action of the RXR
partial agonist 4a, we synthesized several derivatives of 4a and
performed structure−activity relationship analysis. The results
of electrostatic potential field calculations and computational
docking studies indicated that the electrostatic attraction
between an oxygen atom of Asn306 in H4 and the 2-position
at the benzimidazole ring, which is observed in RXR full
agonists, is absent in the partial agonists 4a (reported
previously) and 6c (newly synthesized in this work). The
presence of this interaction would stabilize the holo form of
RXR and favor coactivator recruitment. Compound 6c showed
similar Emax and lower EC50 values toward RXR compared to
4a. Fluorescence polarization assay of cofactor recruitment
confirmed its partial agonist character. We evaluated its side
effects in ICR mice and SD rats, as well as its therapeutic effects
in the KK-Ay mouse model of type 2 diabetes. Although 6c
induced slight hepatomegaly, the degree of hepatomegaly was
less than that induced by RXR full agonists 2 and 4b. We
confirmed that 6c had a significant antidiabetes effect in KK-Ay
type 2 diabetes model mice with less side effects than RXR full
agonists. These results indicate that RXR partial agonists are
promising candidates for antidiabetes drugs with reduced levels
of the side effects associated with full agonists. Further
structural modification of our compounds may generate
attractive candidates for antitype 2 diabetes agents.
3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-ylamine
(10). To a solution of 9 (2.5 g, 10 mmol) in EtOAc (20 mL) was
added Pd/C (catalytic amount). The mixture was stirred at room
temperature (rt) under a H2 atmosphere for 7.0 h, filtered through
Celite, and evaporated under reduced pressure to yield 2.2 g of 10 as a
1
pale yellow solid (qy). H NMR (300 MHz, CDCl3) δ: 6.97 (s, 1H),
6.61 (s, 1H), 3.45 (br s, 2H), 2.14 (s, 3H), 1.64 (s, 4H), 1.24 (s, 6H),
1.24 (s, 6H).
4-Iodo-3-nitrobenzoic Acid (12). To a solution of 11 (2.5 g,
10 mmol) in concd H2SO4 (14 mL) was added dropwise a solution of
concd HNO3 (4.9 mL) and concd H2SO4 (4.3 mL). The reaction
mixture was stirred at rt overnight, poured onto ice (50 mL), and
filtered. The filtrate was evaporated, and the residue was dried to yield
1
2.6 g of 12 as a pale yellow powder (90%). H NMR (500 MHz,
CDCl3) δ: 8.49 (s, 1H), 8.19 (d, J = 8.0 Hz, 1H), 7.92 (d, J = 8.0 Hz, 2H).
4-Iodo-3-nitrobenzoic Acid Methyl Ester (13). To a solution of 12
(2.6 g, 9.0 mmol) in dry MeOH (10 mL) was added dropwise concd
H2SO4 (1.9 mL). The reaction mixture was stirred at rt overnight,
neutralized with sat. NaHCO3 (10 mL), and extracted with EtOAc
(3 × 40 mL). The organic layer was washed with H2O (40 mL) and
brine (50 mL), dried over MgSO4, and evaporated under reduced
pressure. The residue was recrystallized from MeOH to yield 2.8 g of
1
13 as yellow needles (qy). H NMR (500 MHz, CDCl3) δ: 8.45 (d,
J = 2.0 Hz, 1H), 8.15 (d, J = 8.0 Hz, 1H), 7.88 (dd, J = 8.0, 2.0 Hz,
1H), 3.97 (s, 3H).
EXPERIMENTAL SECTION
■
3-Nitro-4-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphthalen-
2-ylamino)benzoic Acid Methyl Ester (14). To a solution of 10 (0.83 g,
3.8 mmol) and 13 (1.2 g, 3.8 mmol) in dry toluene (4.0 mL) were
added Pd2(dba)3 (170 mg, 0.19 mmol), rac-BINAP (180 mg,
0.28 mmol), and Cs2CO3 (3.1 g, 9.5 mmol). The reaction mixture
was refluxed at 110 °C under an Ar atmosphere overnight and filtered
through Celite. The filtrate was evaporated under reduced pressure.
The residue was purified by flash column chromatography (EtOAc:n-
Chemistry. General. Melting points were determined with a
Yanagimoto hot-stage melting point apparatus and are uncorrected. IR
spectra were recorded on a JASCO FT/IR350 (KBr). 1H NMR
spectra were recorded on a JEOL JNM-AL300 FT-NMR system
(300 MHz), a Varian VXR-300 (300 MHz) spectrometer, or a Varian
VXR-500 (500 MHz) spectrometer. FAB-MS was carried out with a
VG70-SE. The purity of all tested compounds was >95%, as confirmed
by combustion analysis or by HPLC analysis. Elemental analysis was
carried out with a Yanagimoto MT-5 CHN recorder elemental
analyzer, and results were within 0.4% of the theoretical values. The
HPLC system used in this study was a Shimadzu liquid chro-
matographic system (Kyoto, Japan) consisting of an LC-10AD pump,
SPD-10AV UV−vis spectrophotometric detector, CTO-10AS column
oven and C-R5A Chromatopac. The samples (each 20 μL) were
injected and the chromatographic analyses were carried out on an
Inertsil ODS-3 (4.6 i.d. × 250 mm, 5 μm, GL Sciences, Tokyo, Japan)
with a guard column of Inertsil ODS-3 (4.0 i.d. × 10 mm, 5 μm, GL
Sciences) at 40 °C, using methanol:16.7−50 mM ammonium acetate
(final concentration at 5 mM) (adjusted with acetic acid to pH 5.0)
(90:10, 80:20, or 70:30, v/v) as a mobile phase. The flow rate was
0.7 mL/min and the absorbance at 280 nm was monitored.
1
hexane = 1:15) to yield 1.3 g of 14 as a yellow foam (qy). H NMR
(300 MHz, CDCl3) δ: 9.63 (br s, 1H), 8.93 (d, J = 2.0 Hz, 1H), 7.93
(dd, J = 9.0, 2.0 Hz, 1H), 7.24 (s, 1H), 7.17 (s, 1H), 6.82 (d, J = 9.0 Hz,
1H), 3.91 (s, 3H), 2.19 (s, 3H), 1.70 (s, 4H), 1.31 (s, 6H), 1.25 (s, 6H).
3-Amino-4-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphtha-
len-2-ylamino)benzoic Acid Methyl Ester (15). To a solution of 14
(0.50 g, 1.3 mmol) in EtOAc (2.0 mL) was added Pd/C (catalytic
amount). The mixture was stirred at rt under a H2 atmosphere
overnight and then filtered through Celite. The filtrate was evaporated
under reduced pressure to yield 0.39 g of 15 as a white solid (84%).
1H NMR (300 MHz, CDCl3) δ: 7.49 (s, 1H), 7.48 (d, J = 8.0 Hz, 1H),
7.13 (s, 1H), 6.96 (s, 1H), 6.83 (d, J = 8.0 Hz, 1H), 5.39 (br s, 1H),
3.87 (s, 3H), 3.55 (br s, 2H), 2.19 (s, 3H), 1.67 (s, 4H), 1.28 (s, 6H),
1.21 (s, 6H).
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-1H-
benzoimidazole-5-carboxylic Acid Methyl Ester (16a). To com-
pound 15 (150 mg, 0.40 mmol) was added formic acid (1.0 mL). The
reaction mixture was refluxed at 100 °C for 4.0 h, poured into 2 N
NaOH (10 mL), and extracted with EtOAc (3 × 40 mL). The or-
ganic layer was collected, washed with brine (10 mL), and dried
over MgSO4. The solvent was evaporated under reduced pressure.
4-[1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-
ethenyl]benzoic Acid (1). This compound was prepared according to
ref 13. HPLC: 11.6 min; >95% purity (MeOH:AcONH4(aq) = 90:10).
6-[Ethyl(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-
amino]nicotinic Acid (2). This compound was prepared according to
ref 14. HPLC: 24.9 min; >95% purity (MeOH:AcONH4(aq) = 80:20).
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)-1H-ben-
zotriazole-5-carboxylic Acid (4a). This compound was prepared
1872
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
The residue was purified by flash column chromatography (EtOAc:n-
hexane = 1:4) to yield 140 mg of 16a as a colorless powder (93%).
1H NMR (300 MHz, CDCl3) δ: 8.61 (d, J = 2.0 Hz, 1H), 8.06 (s, 1H),
8.03 (dd, J = 9.0, 2.0 Hz, 1H), 7.33 (s, 1H), 7.22 (s, 1H), 7.21 (d, J =
9.0 Hz, 1H), 3.98 (s, 3H), 2.05 (s, 3H), 1.75 (s, 4H), 1.36 (s, 6H),
1.29 (s, 6H).
2-Amino-1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphtha-
len-2-yl)-1H-benzoimidazole-5-carboxylic Acid (6b). To a solution of
16b (95 mg, 0.24 mmol) in MeOH (2.0 mL) and THF (2.0 mL) was
added 2 N NaOH (2.0 mL). The reaction mixture was stirred at 60 °C
for 0.5 h, poured into 2 N HCl (4.0 mL), and extracted with EtOAc
(3 × 70 mL). The organic layer was collected, washed with brine
(30 mL) and H2O (30 mL), dried over MgSO4, and evaporated under
reduced pressure. The residue was purified by flash column chromato-
graphy (CH2Cl2:MeOH = 10:1) to yield 110 mg of 6b as white
crystals (qy). Mp: 226.2−227.9 °C. HPLC: 25 min; >95% purity
2-Amino-1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphtha-
len-2-yl)-1H-benzoimidazole-5-carboxylic Acid Methyl Ester (16b).
To a solution of BrCN (48 mg, 0.45 mmol) in THF (5.0 mL) was
added 15 (110 mg, 0.30 mmol). The reaction mixture was stirred at rt
under an Ar atmosphere for 12 h, poured into 2 N NaOH (10 mL),
and extracted with EtOAc (3 × 60 mL). The organic layer was washed
with brine (20 mL) and H2O (40 mL), dried over MgSO4, and
evaporated under reduced pressure. The residue was purified by flash
column chromatography (CH2Cl2:MeOH = 50:1) to yield 30 mg of
1
(MeOH:AcONH4(aq) = 70:30). H NMR (300 MHz, DMSO-d6) δ:
12.95 (br s, 1H), 8.36 (br s, 2H), 7.98 (s, 1H), 7.79 (dd, J = 8.5, 1.0
Hz, 1H), 7.53 (s, 1H), 7.50 (s, 1H), 6.79 (d, J = 8.5 Hz, 1H), 2.01
(s, 3H), 1.69 (s, 4H), 1.33 (s, 6H), 1.25 (s, 3H), 1.23 (s, 3H). FAB-
MS m/z: 378 [M + H]+.
1
16b as a white powder (26%). H NMR (300 MHz, CDCl3) δ: 8.14
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-2-
trifluoromethyl-1H-benzoimidazole-5-carboxylic Acid (6c). To a
solution of 16c (50 mg, 0.11 mmol) in MeOH (1.0 mL) and THF
(2.0 mL) was added 2 N NaOH (1.0 mL). The reaction mixture was
stirred at 60 °C for 1.0 h and then poured into 2 N HCl (2.0 mL) and
extracted with EtOAc (3 × 40 mL). The organic layer was collected,
washed with brine (10 mL) and H2O (10 mL), dried over MgSO4, and
evaporated under reduced pressure. The residue was recrystallized
from EtOAc/n-hexane to yield 21 mg of 6c as a colorless powder
(d, J = 1.5 Hz, 1H), 7.77 (dd, J = 8.5, 1.5 Hz, 1H), 7.34 (s, 1H), 7.23
(s, 1H), 6.81 (d, J = 8.5 Hz, 1H), 3.92 (s, 3H), 2.03 (s, 3H), 1.74
(s, 4H), 1.36 (s, 3H), 1.34 (s, 3H), 1.28 (s, 3H), 1.27 (s, 3H).
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-2-
trifluoromethyl-1H-benzoimidazole-5-carboxylic Acid Methyl Ester
(16c). To a solution of 15 (520 mg, 1.4 mmol) in TFA (8.0 mL) was
added trifluoroacetic anhydride (1.0 mL, 7.0 mmol). The reaction
mixture was stirred at rt for 1.0 h, poured into sat. NaHCO3 (30 mL),
and extracted with EtOAc (3 × 50 mL). The organic layer was washed
with H2O (50 mL) and brine (10 mL), dried over MgSO4, and
evaporated under reduced pressure. The residue was purified by flash
column chromatography (EtOAc:n-hexane = 1:20) to yield 590 mg of
1
(43%). Mp: 235.8−237.1 °C. H NMR (300 MHz, CDCl3) δ: 8.75
(d, J = 1.5 Hz, 1H), 8.15 (dd, J = 8.5, 1.5 Hz, 1H), 7.30 (s, 1H), 7.21
(s, 1H), 7.15 (d, J = 9.5 Hz, 1H), 1.91 (s, 3H), 1.74 (s, 4H), 1.36 (s, 6H),
1.25 (s, 6H). FAB-MS m/z: 431 [M + H]+. Anal. Calcd for
C24H25F3N2O2: C, 66.96; H, 5.85; N, 6.51. Found: C, 66.93; H, 6.11;
N, 6.41.
1
16c as a white solid (95%). H NMR (300 MHz, CDCl3) δ: 8.68 (d,
J = 1.5 Hz, 1H), 8.10 (dd, J = 9.0, 1.5 Hz, 1H), 7.28 (d, J = 8.5 Hz,
1H), 7.20 (s, 1H), 7.12 (dd, J = 9.0, 0.5 Hz, 1H), 3.97 (s, 3H), 1.89
(s, 3H), 1.73 (s, 4H), 1.35 (s, 6H), 1.25 (s, 6H).
2-Oxo-1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)-
2,3-dihydro-1H-benzoimidazole-5-carboxylic Acid (6d). To a
solution of 16d (120 mg, 0.30 mmol) in MeOH (10 mL) was added
2 N NaOH (10 mL). The reaction mixture was stirred at 60 °C for
30 min, poured into 1 N HCl (20 mL), and extracted with EtOAc
(2 × 30 mL). The organic layer was washed with H2O (40 mL) and
brine (30 mL), dried over MgSO4, and evaporated under reduced
pressure to yield 110 mg of 6d as a pale yellow solid (qy). The residue
was recrystallized from EtOAc/n-hexane to yield 45 mg of 6d as a
2-Oxo-1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)-
2,3-dihydro-1H-benzoimidazole-5-carboxylic Acid Methyl Ester
(16d). To a solution of 15 (110 mg, 0.30 mmol) in 1,2-dichloroethane
(6.0 mL) were added Et3N (70 μL, 0.50 mmol) and triphosgene (59 mg,
0.20 mmol). The mixture was refluxed at 110 °C overnight, poured
into H2O (40 mL), and extracted with EtOAc (2 × 30 mL). The
organic layer was washed with H2O (60 mL), dried over MgSO4, and
evaporated under reduced pressure to yield 120 mg of 16d as a pale
yellow solid (qy). 1H NMR (500 MHz, CDCl3) δ: 11.35 (s, 1H), 7.67
(dd, J = 8.0, 1.5 Hz, 1H), 7.60 (d, J = 1.5 Hz, 1H), 7.39 (s, 1H), 7.27
(s, 1H), 6.66 (d, J = 8.0 Hz, 1H), 3.84 (s, 3H), 2.00 (s, 3H), 1.68 (s, 4H),
1.31 (s, 3H), 1.30 (s, 3H), 1.23 (s, 3H), 1.23 (s, 3H).
1
colorless powder. Mp: > 295 °C; H NMR (300 MHz, DMSO-d6) δ:
12.69 (br s, 1H), 11.29 (s, 1H), 7.64 (dd, J = 8.0, 1.5 Hz, 1H), 7.59 (d, J =
1.5 Hz, 1H), 7.39 (s, 1H), 7.26 (s, 1H), 6.63 (d, J = 8.0 Hz, 1H), 2.01
(s, 3H), 1.68 (s, 4H), 1.31 (s, 3H), 1.30 (s, 3H), 1.24 (s, 3H), 1.23
(s, 3H); IR (KBr): 2959−2926 (OH), 1712 (CO), 1685 (CO) cm−1.
FAB-MS m/z: 379 [M + H]+. Anal. Calcd for C23H26N2O3·1/4H2O:
C, 72.13; H, 6.97; N, 7.31. Found: C, 72.40; H, 7.05; N, 7.18.
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)-2-thio-
xo-2,3-dihydro-1H-benzoimidazole-5-carboxylic Acid (6e). To a
solution of 16e (120 mg, 0.30 mmol) in MeOH (3.0 mL) was
added 2 N NaOH (3.0 mL). The reaction mixture was stirred at 60 °C
for 15 min, poured into 2 N HCl (30 mL), and extracted with EtOAc
(2 × 20 mL). The organic layer was washed with H2O (2 × 30 mL)
and brine (30 mL), dried over MgSO4, and evaporated under reduced
pressure to yield 100 mg of 6e as a pale yellow solid (86%). The
residue was recrystallized from EtOAc/n-hexane to yield 90 mg of
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)-2-thio-
xo-2,3-dihydro-1H-benzoimidazole-5-carboxylic Acid Methyl Ester
(16e). To a solution of 15 (62 mg, 0.17 mmol) in DMF (2.0 mL) were
added CS2 (100 μL, 1.7 mmol) and DBU (25 μL, 0.17 mmol). The
reaction mixture was stirred at 50 °C for 0.5 h, poured into H2O
(60 mL), and extracted with EtOAc (3 × 40 mL). The organic layer
was washed with H2O (3 × 80 mL) and brine (80 mL), dried over
MgSO4, and evaporated under reduced pressure. The residue was
purified by flash column chromatography (EtOAc:n-hexane = 1:5) to
1
yield 64 mg of 16d as a white solid (93%). H NMR (300 MHz,
CDCl3) δ: 12.3 (br s, 1H), 8.00 (s, 1H), 7.89 (dd, J = 8.5, 1.5 Hz, 1H),
7.34 (s, 1H), 7.22 (s, 1H), 6.81 (d, J = 8.5 Hz, 1H), 3.92 (s, 3H), 2.08
(s, 3H), 1.74 (s, 4H), 1.36 (s, 3H), 1.35 (s, 3H), 1.31 (s, 3H), 1.26 (s, 3H).
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-1H-
benzoimidazole-5-carboxylic Acid (6a). To a solution of 16a (140 mg,
0.40 mmol) in MeOH (3.0 mL) and THF (1.0 mL) was added
2 N NaOH (2.0 mL). The reaction mixture was stirred at 60 °C
for 2.0 h, poured into 2 N HCl (2.0 mL), and extracted with EtOAc
(3 × 40 mL). The organic layer was collected, washed with brine (20 mL),
dried over MgSO4, and evaporated under reduced pressure to yield
180 mg of 6a (qy). The residue was recrystallized from EtOAc/
n-hexane to yield 110 mg of colorless powder. Mp: 255.0−257.0 °C.
HPLC: 13 min; 99% purity (MeOH:AcONH4(aq) = 70:30). 1H NMR
(300 MHz, CDCl3) δ: 8.76 (d, J = 2.0 Hz, 1H), 8.17 (s, 1H), 8.11 (dd,
J = 8.0, 2.0 Hz, 1H), 7.33 (s, 1H), 7.27 (d, J = 8.0 Hz, 1H), 7.26
(s, 1H), 7.23 (s, 1H), 2.06 (s, 3H), 1.74 (s, 4H), 1.36 (s, 6H), 1.29
(s, 6H). FAB-MS m/z: 363 [M + H]+.
1
6e as pale yellow needles. Mp: > 295.0 °C. H NMR (300 MHz,
DMSO-d6) δ: 13.20 (br s, 1H), 12.91 (br s, 1H), 7.76 (dd, J = 8.5,
1.5 Hz, 1H), 7.75 (d, J = 1.5 Hz, 1H), 7.41 (s, 1H), 7.24 (s, 1H), 6.68
(d, J = 8.5 Hz, 1H), 1.95 (s, 3H), 1.68 (s, 4H), 1.33 (s, 3H), 1.32
(s, 3H), 1.23 (s, 3H), 1.23 (s, 3H). IR (KBr): 3056 (NH), 2962−2923
(OH), 1693 (CO) cm−1. FAB-MS m/z: 395 [M + H]+. Anal. Calcd for
C23H26N2O2S·1/2H2O: C, 68.48; H, 6.74; N, 6.94. Found: C, 68.39;
H, 6.63; N, 6.82.
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-1H-
indole-5-carboxylic Acid Methyl Ester (19). To a solution of 17
(1.0 g, 3.6 mmol) and 18 (530 mg, 3.0 mmol) in dry toluene (4.0 mL)
were added CuI (28 mg, 0.15 mmol), K3PO4 (1.3 g, 6.3 mmol), KI
(600 mg, 3.6 mmol), and N,N′-dimethylethylenediamine (65 μL,
0.60 mmol). The reaction mixture was refluxed at 160 °C for 2.5 h
under microwave irradiation, filtered through Celite, and evaporated
1873
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
under reduced pressure. The residue was purified by flash column
chromatography (EtOAc:n-hexane = 1:40) to yield 89 mg of 19 as
yellow crystals (7.9%). H NMR (300 MHz, CDCl3) δ: 8.47 (d, J =
2.0 Hz, 1H), 7.88 (dd, J = 9.0, 2.0 Hz, 1H), 7.28 (s, 1H), 7.25 (d, J =
3.0 Hz, 1H), 7.23 (s, 1H), 7.09 (d, J = 9.0 Hz, 1H), 6.75 (d, J = 3.0 Hz,
1H), 3.95 (s, 3H), 2.00 (s, 3H), 1.74 (s, 1H), 1.36 (s, 6H), 1.28
(s, 6H).
1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-1H-
indole-5-carboxylic Acid (7). To a solution of 19 (66 mg, 0.18 mmol)
in MeOH (2.0 mL) and THF (1.0 mL) was added 2 N NaOH (2.0 mL).
The reaction mixture was stirred at 60 °C for 1.5 h, poured into
2 N HCl (2.0 mL), and extracted with EtOAc (3 × 40 mL). The organic
layer was collected, washed with brine (20 mL), dried over MgSO4, and
evaporated under reduced pressure to yield 58 mg of 7 (92%). The
residue was recrystallized from EtOAc/n-hexane to yield 10 mg of 7
as a colorless powder. Mp: 244.0−245.0 °C. HPLC: 33 min; >95%
purity (MeOH:AcONH4(aq) = 80:20). 1H NMR (300 MHz, DMSO)
δ: 12.5 (s, 1H), 8.47 (d, J = 2.0 Hz, 1H), 7.74 (dd, J = 9.0, 2.0 Hz,
1H), 7.58 (d, J = 3.0 Hz, 1H), 7.41 (s, 1H), 7.26 (s, 1H), 7.09 (d, J =
9.0 Hz, 1H), 6.79 (d, J = 3.0 Hz, 1H), 1.94 (s, 3H), 1.68 (s, 4H), 1.31
(s, 6H), 1.24 (s, 6H). FAB-MS m/z: 362 [M + H]+. Anal. Calcd for
C24H27NO2·1/4H2O: C, 78.76; H, 7.57; N, 3.83. Found: C, 78.60; H,
7.36; N, 3.93.
reagents were diluted in phosphate buffer (50 mM sodium phosphate
pH 7.2, 154 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 0.01%
NP40), and the final DMSO concentration in the assay mixtures was
adjusted to 1%. The mixtures containing fluorescein-labeled cofactor
peptide, RXRα, and various RXR ligands in phosphate buffer were
incubated for 1 h at 25 °C. The fluorescence polarization of the
mixtures was measured at an excitation wavelength of 485 nm and an
emission wavelength of 535 nm. Fluorescence polarization measure-
ments were made with a TECAN Polarion. Fluorescence polarization
is the ratio of the difference between the intensities of parallel and
perpendicularly polarized fluorescent light to the total light intensity.
Electrostatic Potential Fields and Molecular Docking. The crystal
structure of human RXRα−ligand binding domain was retrieved from
the Brookhaven Protein Data Bank (http://www.rcsb.org/pdb/
Welcome.do). Polar hydrogen atoms were added to both the protein
and the ligand. United atom Kollman charges were assigned for the
protein. The 3D structures of ligands used for the docking study were
constructed by using Spartan (Wave function, Inc.). After semi-
empirical pm3 calculations, 6-31G* ab initio calculations were
performed to find the lowest energy conformers. The electrostatic
potential fields (ESP) were drawn with Spartan. The AutoDock4.2
molecular docking program32 was employed by using a genetic
algorithm with local search (GALS). One hundred individual GA runs,
150 chromosomes, a crossover ratio of 0.80, a rate of gene mutation of
0.02, and an elitism ratio of 0.10 were used for each ligand. The grid
box was created with dimensions of 40 × 40 × 40 Å3, which encloses
the original ligand. Molfeat (FiatLux Co., Tokyo, Japan) was used for
molecular modeling.
1
Luciferase Reporter Gene Assay. Culture of COS-1 Cells. COS-1
cells were maintained in Dulbecco’s modified Eagle’s medium
supplemented with 10% FBS, NaHCO3 (1.0 g), L-glutamine (0.292 g),
penicillin (25 000 units), and streptomycin (25 000 μg) in a humidified
atmosphere of 5% CO2 in air at 37 °C.
Luciferase Reporter Gene Assay. Luciferase reporter gene assays
were performed using COS-1 cells transfected with three kinds of
vectors: each receptor subtype, a luciferase reporter gene under the
control of the appropriate RXR response element, and secreted
alkaline phosphatase (SEAP) gene33 as a background. CRBPII-tk-Luc,
tk-TREPII-Luc, tk-PPREx3-Luc, and tk-rBARx3-Luc reporters were
used as RXR, RAR, PPAR, and LXR response elements, respectively.
The amounts of each receptor subtype and response element were 1.0
and 4.0 μg, respectively. Transfection was performed with QIA
Effectene transfection reagent according to the supplier’s protocol. In
the case of heterodimer assay, RXRα (0.5 μg), each partner receptor
(PPARγ or LXRα, 0.5 μg), and the partner response element (4.0 μg)
were transfected into COS-1 cells as described above. Test compound
solutions (DMSO concentration below 1%) were added to the
suspension of transfected cells, which were seeded at about 2.0 × 104
cells/well in 96-well white plates. After incubation in a humidified
atmosphere of 5% CO2 at 37 °C for 18 h, 25 μL of the medium was
used for analyzing SEAP activities, and the remaining cells were used
for luciferase reporter gene assays with a Steady-Glo luciferase assay
system (Promega) according to the supplier’s protocol. The luciferase
activities were normalized using SEAP activities. The assays were
carried out in triplicate three times.
Production and Purification of RXRα Protein. Production and
purification of recombinant RXRα protein were done using GATE-
WAY technology.34,35 Destination vectors were generated by insertion
of human RXRα DNA (Ultimate Human ORF Clone, Invitrogen)36
into a pDEST17 vector (Invitrogen) and were transformed into
Escherichia coli BL21-AI cells (Invitrogen) by means of LR reaction.
These cells were used as expression clones. The expression clones
were cultured in LB medium containing 100 μg/mL ampicillin at
37 °C with shaking until the OD600 reached 0.6−1.0 and then were
diluted to OD 0.1 at 600 nm. After addition of 0.2% L-arabinose and
0.1% glucose to the culture during the exponential phase of growth
(OD 0.4 at 600 nm), cells were cultured for 2 h and then harvested.
The cell pellets were resuspended in Lysis buffer (50 mM Tris-HCl,
pH 7.5, 0.15 M NaCl, 0.1% SDS, 1.0% Triton X-100, 1 mM PMSF) at
4 °C. The RXR protein was purified using a His GraviTrap column
(GE Healthcare)37 to give 2.2 mg/L pure protein in the culture
medium.
In Vivo Assays. General. All experiments were conducted in
accordance with the Guidelines for Animal Experiments at Okayama
University Advanced Science Research Center, and all procedures
were approved by the Animal Research Control Committee of
Okayama University.
Measurement of Serum Concentration of Test Compounds after
Oral Administration at 30 mg/kg to Mice. Groups of six-week-old
ICR male mice (n = 6−9 in each) were treated with solutions of test
compound 30 mg/kg (1% ethanol and 0.5% CMC in distilled water)
at a volume of 10 mL/kg of body weight by oral administration. At the
indicated times, 0.6 mL of blood was taken from the inferior vena cava
under diethyl ether anesthesia. Each blood sample was centrifuged at
1900g for 5 min at rt. To 100 μL of the resulting plasma were added
100 μL of ice-cold 5 mM ammonium acetate solution (adjusted with
acetic acid to pH 5.0) and 1 mL of ice-cold ethyl acetate. The resulting
mixture was vortexed for 30 s, kept at room temperature for 10 min,
and centrifuged at 1900g for 30 s at room temperature. An 800 μL
aliquot of the ethyl acetate phase was removed and concentrated to
dryness in a centrifugal evaporator. To the resulting residue was added
100 μL of HPLC-grade methanol. This solution was directly subjected
to HPLC analysis, and the concentration of each compound was
determined from the peak area of the sample with reference to a
calibration plot obtained with the authentic compound.
HPLC Conditions. The HPLC system used in this study was a
Shimadzu liquid chromatographic system (Kyoto, Japan) consisting of
an SCL-10A system controller, LC-10AD pump, SPD-10AV UV−vis
spectrophotometric detector, SIL-10AD autoinjector, CTO-10A
column oven, DGU-14A degasser, and C-R7A Chromatopac. The
samples (each 20 μL) were injected using a refrigerated autosampler
kept at 10 °C. The chromatographic analyses were carried out on an
Inertsil ODS-3 (4.6 i.d. × 250 mm, 5 μm, GL Sciences, Tokyo, Japan)
kept at 40 °C, using methanol:33.3 mM ammonium acetate (adjusted
with acetic acid to pH 5.0) (85:15, v/v) as the mobile phase. The flow
rate was 0.7 mL/min and the absorbance was monitored at 280 nm.
Observation of Side Effects after Once-Daily Oral Administration
at 30 mg/kg for 7 Consecutive Days in Male ICR Mice. Six- to seven-
week-old male ICR mice were purchased from Charles River
Laboratories Japan, Inc. After arrival of the animals, all were group-
housed and acclimated to the colony for 1 or 2 days before the
experiment. Before the experiment, they were housed with four mice
per cage, with free access to water and chow pellets in a light (12 h on,
8:00 a.m. /12 h off, 8:00 p.m.), temperature (23 1 °C), and relative
Fluorescence Polarization Assay. Fluorescein-labeled cofactor
peptides were purchased from Life Technologies. Assays were performed
in 96-well half-area black plates (Greiner) in a final volume of 40 μL. All
1874
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
humidity (50
20%) controlled environment. Before experiments,
(γ-GTP), alkaline phosphatase (ALP), creatinine (CRE), blood urea
nitrogen (BUN), triglyceride (TG), total cholesterol (TCHO), and
glucose (GLU) were measured by using a Fuji Dry Chem system (Dry
Chem 4000 V, Fuji Medical Co., Tokyo, Japan).
mice were assigned to experimental groups so as to minimize the
variance between groups based on the measured weight [four per cage
(17 × 33 × 15 cm3)]. Body weight was measured at approximately
10:00 a.m. every day for 7 days before dosing. Mice were administered
orally with a solution of test compound at a dose of 30 mg/kg or with
the vehicle [1% ethanol and 0.5% carboxymethyl cellulose (CMC) in
distilled water] at a volume of 10 mL/kg of animal at approximately
10:00 a.m. every day for 7 days. On the final day of dosing, animals
were fasted from 17:00 p.m. and given water ad libitum. On the next
day, at approximately 10:00 a.m., animals were weighed and anesthetized
with diethyl ether. Blood and liver were removed immediately, and the
liver was weighed. Approximately 1 mL of blood in an Eppendorf sample
tube was centrifuged to afford a serum sample. Each blood sample was
centrifuged at 1900g for 5 min at rt.
Oral Glucose Tolerance Test. KK-Ay mice treated with each
compound for 14 days were fasted for 17 h and orally given glucose
solution (100 mg glucose in 1 mL distilled water) at a dose of 1 g/kg
body weight. At 0, 15, 30, 45, 60, 90, and 120 min after the glucose
loading, blood glucose level was measured as described above.
ASSOCIATED CONTENT
* Supporting Information
■
S
HPLC charts and in vitro and in vivo data are described. This
information is available free of charge via the Internet at http://
pubs.acs.org.
Observation of Side Effects after Once-Daily Oral Administration
at 30 mg/kg for 28 Consecutive Days in SD Rats. Four-week-old
male and female SD rats were purchased from Charles River
Laboratories Japan, Inc. After arrival of the rats, all were group-
housed and acclimated to the colony for 6 (male) or 7 (female) days
before the experiment. Before the experiment, they were housed with
two rats per cage, with free access to water and chow pellets in a light
(12 h on, 8:00 a.m./12 h off, 8:00 p.m.), temperature (23 1 °C), and
AUTHOR INFORMATION
Corresponding Author
*Tel/Fax: +81-(0)86-251-7963. E-mail: kakuta@pharm.
okayama-u.ac.jp.
■
Author Contributions
relative humidity (50
20%)-controlled environment. Before
H.K. conceived and designed the project. F.O., N.Y., R.S. and
M.H. synthesized compounds. F.O., S.Y., and Y.O. performed
reporter gene assays. M.M. prepared plasmids. H.N. prepared
RXR proteins. S.Y. performed cofactor recruitment assays. A.T.
and H.K. performed HPLC analysis. F.O., M.N., K.K., T.K., K.K.,
Y.F., C.F., Y.Y., H.Y., and H.K. performed in vivo experiments.
The manuscript was written by F.O., A.T., and H.K.
experiments, rats were assigned to experimental groups so as to
minimize the variance between groups based on the measured weight
[two per cage (25.0 × 41.5 × 19.0 cm3)]. Body weight was measured
at approximately 10:00 a.m. every day for 28 days before dosing. Rats
were administered orally with a solution of test compound at a dose of
30 mg/kg or with the vehicle (1% ethanol and 0.5% CMC in distilled
water) at a volume of 5 mL/kg of animal at approximately 10:00 a.m.
every day for 28 days. On the final day of dosing, animals were fasted
from 17:00 p.m. and given water ad libitum. On the next day, at
approximately 10:00 a.m., animals were weighed and anesthetized with
isoflurane. Blood, liver, brain, kidney, spleen, and testis (male only)
were removed immediately. The liver, brain, kidney, spleen, and testis
were weighed and frozen with liquid nitrogen. Approximately 10 mL
of blood in a centrifuge tube was centrifuged at 2000g for 10 min at
4 °C to obtain a serum sample.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Miyachi (Okayama University) for kindly
providing TIPP-703 and carba-T0901317. We also thank
Dr. Kagechika (Tokyo Medical and Dental University) for
kindly providing PA452. The authors are also grateful to Prof.
Yoshio Naomoto, Dr. Takuya Fukazawa (Kawasaki Medical
School), and Mrs. Kaori Endo-Umeda (Nihon University) for
preparing plasmids. The authors gratefully thank Division of
Instrumental Analysis, Okayama University for the NMR
measurements. This work was supported by Health and Labour
Science research grants for Research on Seeds for Publicly
Essential Drugs and Medical Devices (23080401) from the
Ministry of Health, Labour, and Welfare of Japan. We also
thank the Ministry of Education, Science, Culture and Sports of
Japan, Takeda Science Foundation and “Top Young Researcher
Award of Okayama University for 2011” for financial support.
Evaluation of Blood Glucose-Lowering Activities in KK-Ay Mice.
Type 2 diabetic KK-Ay mice, in which the Ay allele at the agouti locus
had been transferred to the inbred KK strain by repetitive
backcrossing, were used as the congenic strain. The introduction of
the Ay allele caused DM and massive hereditary obesity. Four-week-old
male KK-Ay mice were purchased from CLEA Japan Inc. The KK-Ay
mice were allowed free access to solid food and tap water. After arrival
of the animals, all were group-housed and acclimated to the colony for
6 weeks before the experiment. Before the experiment, they were
housed with one mouse per cage, with free access to water and chow
pellets in a light (12 h on, 8:00 a.m./12 h off, 8:00 p.m.), tempera-
ture (23 1 °C), and relative humidity (50 20%) controlled en-
vironment. Before experiments, mice were assigned to experimental
groups so as to minimize the variance between groups based on the
blood glucose level [one per cage (17 × 33 × 15 cm3)]. Body weight
was measured at approximately 10:00 a.m. every day for 14 days before
dosing. Mice were administered orally with a solution of test
compound at a dose of 10 mg/kg or with the vehicle (1% ethanol
and 0.5% CMC in distilled water) at a volume of 10 mL/kg of animal
at approximately 10:00 a.m. every day for 14 days. At day 15, an oral
glucose tolerance test (OGTT) was performed, and animals were
fasted from 17:00 p.m. and given water ad libitum. On the next day, at
approximately 10:00 a.m., animals were weighed and anesthetized with
diethyl ether. Blood was removed immediately and centrifuged in an
Eppendorf sample tube to obtain serum. Each blood sample was
centrifuged at 1900g for 5 min at rt. Samples for measurements of fed
blood glucose level were taken from the tail vein of the mice, and
glucose was measured by using the glucose oxidase method (Medisafe-
mini, TERUMO, Tokyo, Japan).
ABBREVIATIONS USED
■
RXR, retinoid X receptor; RAR, retinoic acid receptor; LXR, liver
X receptor; PPAR, peroxisome proliferator-activated receptors; qy,
quantitative yield; TFAA, trifluoroacetic anhydride; TFA, trifluoro-
acetic acid; TEA, triethylamine; DCE, 1,2-dichloroethane; DBU,
1,8-diazabicyclo[5.4.0]undec-7-ene; DMEDA, N,N′-dimethylethy-
lenediamine; Asn, asparagine; Gly, glycine; TG, triglyceride;
TCHO, total cholesterol; AST, alanine aminotransferase; ALT,
aspartate aminotransferase; γ-GTP, γ-glutamyltranspeptidase; ALP,
alkaline phosphatase; CRE, creatinine; BUN, blood urea nitrogen.
REFERENCES
■
(1) Svensson, S.; Ostberg, T.; Jacobsson, M.; Norstrom, C.;
̈
Measurements of Blood Parameters. Aspartate aminotransferase
(AST), alanine aminotransferase (ALT), γ-glutamyl transpeptidase
Stefansson, K.; Hallen
́
, D.; Johansson, I. C.; Zachrisson, K.; Ogg, D.;
Jendeberg, L. Crystal structure of the heterodimeric complex of LXRα
1875
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
and RXRβ ligand-binding domains in a fully agonistic conformation.
EMBO J. 2003, 22, 4625−4633.
receptor/retinoid X receptor (RAR/RXR) heterodimers and tran-
scriptional coactivators through structural and fluorescence aniso-
tropystudies. J. Biol. Chem. 2005, 280, 1625−1633.
(18) Li, X.; Hansen, P. A.; Xi, L.; Chandraratna, R. A.; Burant, C. F.
Distinct mechanisms of glucose lowering by specific agonists for
peroxisomal proliferator activated receptor gamma and retinoic acid X
receptors. J. Biol. Chem. 2005, 280, 38317−38327.
(2) de Lera, A. R.; Bourguet, W.; Altucci, L.; Gronemeyer, H. Design
of selective nuclear receptor modulators: RAR and RXR as a case
study. Nat. Rev. Drug Discovery 2007, 6, 811−820.
(3) Mangelsdorf, D. J.; Evans, R. M. The RXR heterodimers and
orphan receptors. Cell 1995, 83, 841−850.
(4) Su, C. G.; Wen, X.; Bailey, S. T.; Jiang, W.; Rangwala, S. M.;
Keilbaugh, S. A.; Flanigan, A.; Murthy, S.; Lazar, M. A.; Wu, G. D. A
novel therapy for colitis utilizing PPARγ ligands to inhibit the
epithelial inflammatory response. J. Clin. Invest. 1999, 104, 383−389.
(5) Chao, E. Y.; Caravella, J. A.; Watson, M. A.; Campobasso, N.;
Ghisletti, S.; Billin, A. N.; Galardi, C.; Wang, P.; Laffitte, B. A.;
Iannone, M. A.; Goodwin, B. J.; Nichols, J. A.; Parks, D. J.; Stewart, E.;
Wiethe, R. W.; Williams, S. P.; Smallwood, A.; Pearce, K. H.; Glass, C.
K.; Willson, T. M.; Zuercher, W. J.; Collins, J. L. Structure-guided
design of N-phenyl tertiary amines as transrepression-selective liver X
receptor modulators with anti-inflammatory activity. J. Med. Chem.
2008, 51, 5758−5765.
(6) Pascual, G.; Fong, A. L.; Ogawa, S.; Gamliel, A.; Li, A. C.; Perissi,
V.; Rose, D. W.; Willson, T. M.; Rosenfeld, M. G.; Glass, C. K. A
SUMOylation-dependent pathway mediates transrepression of inflam-
matory response genes by PPARγ. Nature 2005, 437, 759−763.
(7) Mitro, N.; Mak, P. A.; Vargas, L.; Godio, C.; Hampton, E.;
Molteni, V.; Kreusch, A.; Saez, E. The nuclear receptor LXR is a
glucose sensor. Nature 2007, 445, 219−223.
(19) Lenhard, J. M.; Lancaster, M. E.; Paulik, M. A.; Weiel, J. E.; Binz,
J. G.; Sundseth, S. S.; Gaskill, B. A.; Lightfoot, R. M.; Brown, H. R. The
RXR agonist LG100268 causes hepatomegaly, improves glycaemic
control and decreases cardiovascular risk and cachexia in diabetic mice
suffering from pancreatic beta-cell dysfunction. Diabetologia 1999, 42,
545−554.
(20) Davies, P. J.; Berry, S. A.; Shipley, G. L.; Eckel, R. H.; Hennuyer,
N.; Crombie, D. L.; Ogilvie, K. M.; Peinado-Onsurbe, J.; Fievet, C.;
Leibowitz, M. D.; Heyman, R. A.; Auwerx, J. Metabolic effects of
rexinoids: Tissue-specific regulation of lipoprotein lipase activity. Mol.
Pharmacol. 2001, 59, 170−176.
(21) Ohsawa, F.; Morishita, K.; Yamada, S.; Makishima, M.; Kakuta,
H. Modification at the lipophilic domain of RXR agonists differentially
influences activation of RXR heterodimers. ACS Med. Chem. Lett.
2010, 1, 521−525.
(22) Kakuta, H.; Ohsawa, F.; Yamada, S.; Makishima, M.; Tai, A.;
Yasui, H.; Yoshikawa, Y. Feasibility of structural modification of RXR
agonists to separate blood glucose-lowering action from adverse
effects: Studies in KK-Ay type 2 diabetes model mice. Biol. Pharm. Bull.
2012, 35, 629−633.
(23) Qing, F. L.; Yue, X. J. A novel synthesis of 9,13-di-cis double
bonds locked retinoids. Tetrahedron Lett. 1997, 38, 8067−8070.
(24) Germain, P.; Staels, B.; Dacquet, C.; Spedding, M.; Laudet, V.
Overview of nomenclature of nuclear receptors. Pharmacol. Rev. 2006,
58, 685−704.
(25) Oberfield, J. L.; Collins, J. L.; Holmes, C. P.; Goreham, D. M.;
Cooper, J. P.; Cobb, J. E.; Lenhard, J. M.; Hull-Ryde, E. A.; Mohr, C.
P.; Blanchard, S. G.; Parks, D. J.; Moore, L. B.; Lehmann, J. M.;
Plunket, K.; Miller, A. B.; Milburn, M. V.; Kliewer, S. A.; Willson, T.
M. A peroxisome proliferator-activated receptor gamma ligand inhibits
adipocyte differentiation. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 6102−
6106.
(8) Commerford, S. R.; Vargas, L.; Dorfman, S. E.; Mitro, N.;
Rocheford, E. C.; Mak, P. A.; Li, X.; Kennedy, P.; Mullarkey, T. L.;
Saez, E. Dissection of the insulin-sensitizing effect of liver X receptor
ligands. Mol. Endocrinol. 2007, 21, 3002−3012.
(9) Joseph, S. B.; Castrillo, A.; Laffitte, B. A.; Mangelsdorf, D. J.;
Tontonoz, P. Reciprocal regulation of inflammation and lipid
metabolism by liver X receptors. Nat. Med. 2003, 9, 213−219.
(10) Fowler, A. J.; Sheu, M. Y.; Schmuth, M.; Kao, J.; Fluhr, J. W.;
Rhein, L.; Collins, J. L.; Willson, T. M.; Mangelsdorf, D. J.; Elias, P.
M.; Feingold, K. R. Liver X receptor activators display antiin-
flammatory activity in irritant and allergic contact dermatitis models:
Liver-X-receptor-specific inhibition of inflammation and primary
cytokine production. J. Invest. Dermatol. 2003, 120, 246−255.
(11) Shulman, A. I.; Larson, C.; Mangelsdorf, D. J.; Ranganathan, R.
Structural determinants of allosteric ligand activation in RXR
heterodimers. Cell 2004, 116, 417−429.
(12) Gniadecki, R.; Assaf, C.; Bagot, M.; Dummer, R.; Duvic, M.;
Knobler, R.; Ranki, A.; Schwandt, P.; Whittaker, S. The optimal use of
bexarotene in cutaneous T-cell lymphoma. Br. J. Dermatol. 2007, 157,
433−440.
(13) Boehm, M. F.; Zhang, L.; Badea, B. A.; White, S. K.; Mais, D. E.;
Berger, E.; Suto, C. M.; Goldman, M. E.; Heyman, R. A. Synthesis and
structure−activity relationships of novel retinoid X receptor-selective
retinoids. J. Med. Chem. 1994, 37, 2930−2941.
(14) Fujii, S.; Ohsawa, F.; Yamada, S.; Shinozaki, R.; Fukai, R.;
Makishima, M.; Enomoto, S.; Tai, A.; Kakuta, H. Modification at the
acidic domain of RXR agonists has little effect on permissive RXR
heterodimer activation. Bioorg. Med. Chem. Lett. 2010, 20, 5139−5142.
(15) Takamatsu, K.; Takano, A.; Yakushiji, N.; Morohashi, K.;
Morishita, K.; Matsuura, N.; Makishima, M.; Tai, A.; Sasaki, K.;
Kakuta, H. The first potent subtype-selective retinoid X receptor
(RXR) agonist possessing a 3-isopropoxy-4-isopropylphenylamino
moiety, NEt-3IP (RXRα/β-dual agonist). ChemMedChem 2008, 3,
780−787.
(26) Egea, P. F.; Mitschler, A.; Rochel, N.; Ruff, M.; Chambon, P.;
Moras, D. Crystal structure of the human RXRalpha ligand-binding
domain bound to its natural ligand: 9-cis Retinoic acid. EMBO J. 2000,
19, 2592−2601.
́ ́
(27) Levy-Bimbot, M.; Major, G.; Courilleau, D.; Blondeau, J.; Levi,
Y. Tetrabromobisphenol-A disrupts thyroid hormone receptor alpha
function in vitro: Use of fluorescence polarization to assay corepressor
and coactivator peptide binding. Chemosphere 2012, 87, 782−788.
(28) Ozers, M. S.; Ervin, K. M.; Steffen, C. L.; Fronczak, J. A.;
Lebakken, C. S.; Carnahan, K. A.; Lowery, R. G.; Burke, T. J. Analysis
of ligand-dependent recruitment of coactivator peptides to estrogen
receptor using fluorescence polarization. Mol. Endcrinol. 2005, 19, 25−
34.
(29) Stafslien, D. K.; Vedvik, K. L.; Rosier, T.; Ozers, M. S. Analysis
of ligand-dependent recruitment of coactivator peptides to RXRβ in a
time-resolved fluorescence resonance energy transfer assay. Mol. Cell.
Endocrinol. 2007, 264, 82−89.
(30) Ghosh, J. C.; Yang, X.; Zhang, A.; Lambert, M. H.; Li, H.; Xu, H.
E.; Chen, J. D. Interactions that determine the assembly of a retinoid X
receptor/corepressor complex. Proc. Natl. Acad. Sci. U. S. A. 2002, 99,
5842−5847.
(31) Takahashi, B.; Ohta, K.; Kawachi, E.; Fukasawa, H.; Hashimoto,
Y.; Kagechika, H. Novel retinoid X receptor antagonists: Specific
inhibition of retinoid synergism in RXR−RAR heterodimer actions. J.
Med. Chem. 2002, 45, 3327−3330.
(32) Morris, G. M.; Huey, R.; Lindstrom, W.; Sanner, M. F.; Belew,
R. K.; Goodsell, D. S.; Olson, A. J. AutoDock4 and AutoDockTools4:
Automated docking with selective receptor flexibility. J. Comput. Chem.
2009, 30, 2785−2791.
(16) Kakuta, H.; Yakushiji, N.; Shinozaki, R.; Ohsawa, F.; Yamada, S.;
Ohta, Y.; Kawata, K.; Nakayama, M.; Hagaya, M.; Fujiwara, C.;
Makishima, K.; Uno, S.; Tai, A.; Maehara, A.; Nakayama, M.; Oohashi,
T.; Yasui, H.; Yoshikawa, Y. RXR partial agonist CBt-PMN exerts
therapeutic effects on type 2 diabetes without the side effects of RXR
full agonists. ACS Med. Chem. Lett. 2012, 3, 427−432.
(17) Pogenberg, V.; Guichou, J. F.; Vivat-Hannah, V.; Kammerer, S.;
Per
́
ez, E.; Germain, P.; de Lera, A. R.; Gronemeyer, H.; Royer, C. A.;
Bourguet, W. Characterization of the interaction between retinoic acid
1876
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877

Journal of Medicinal Chemistry
Article
(33) Kain, S. R. Use of secreted alkaline phosphatase as a reporter of
gene expression in mammalian cells. Methods Mol. Biol. 1997, 63, 49−
60.
(34) Landy, A. Dynamic, structural, and regulatory aspects of lambda
site-specific recombination. Annu. Rev. Biochem. 1989, 58, 913−949.
(35) Walhout, A. J.; Temple, G. F.; Brasch, M. A.; Hartley, J. L.;
Lorson, M. A.; van den Heuvel, S.; Vidal, M. GATEWAY
recombinational cloning: Application to the cloning of large numbers
of open reading frames or ORFeomes. Methods Enzymol. 2000, 328,
575−592.
(36) Liang, F.; Matrubutham, U.; Parvizi, B.; Yen, J.; Duan, D.;
Mirchandani, J.; Hashima, S.; Nguyen, U.; Ubil, E.; Loewenheim, J.;
Yu, X.; Sipes, S.; Williams, W.; Wang, L.; Bennett, R.; Carrino, J.
ORFDB: An information resource linking scientific content to a high-
quality open reading frame (ORF) collection. Nucleic Acids Res. 2004,
32, D595−D599.
(37) Hengen, P. Purification of His-Tag fusion proteins from
Escherichia coli. Trends Biochem. Sci. 1995, 20, 285−286.
1877
dx.doi.org/10.1021/jm400033f | J. Med. Chem. 2013, 56, 1865−1877