Competitive Binding Assay with an Umbelliferone-Based Fluorescent Rexinoid for Retinoid X Receptor Ligand Screening

Article abstract of DOI:10.1021/acs.jmedchem.9b00995

Ligands for retinoid X receptors (RXRs), "rexinoids", are attracting interest as candidates for therapy of type 2 diabetes and Alzheimer's and Parkinson's diseases. However, current screening methods for rexinoids are slow and require special apparatus or facilities. Here, we created 7-hydroxy-2-oxo-6-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-2H-chromene-3-carboxylic acid (10, CU-6PMN) as a new fluorescent RXR agonist and developed a screening system of rexinoids using 10. Compound 10 was designed based on the fact that umbelliferone emits strong fluorescence in a hydrophilic environment, but the fluorescence intensity decreases in hydrophobic environments such as the interior of proteins. The developed assay using 10 enabled screening of rexinoids to be performed easily within a few hours by monitoring changes of fluorescence intensity with widely available fluorescence microplate readers, without the need for processes such as filtration.

Full text of DOI:10.1021/acs.jmedchem.9b00995

Subscriber access provided by Nottingham Trent University
Article
Competitive Binding Assay with an Umbelliferone-based
Fluorescent Rexinoid for Retinoid X Receptor Ligand Screening
Shoya Yamada, Mayu Kawasaki, Michiko Fujihara, Masaki Watanabe, Yuta
Takamura, Maho Takioku, Hiromi Nishioka, Yasuo Takeuchi, Makoto Makishima,
Tomoharu Motoyama, Sohei Ito, Hiroaki Tokiwa, Shogo Nakano, and Hiroki Kakuta
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 04 Sep 2019
Downloaded from pubs.acs.org on September 4, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Competitive Binding Assay with an Umbelliferone-
based Fluorescent Rexinoid for Retinoid X Receptor
Ligand Screening
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Shoya Yamada,†,‡ Mayu Kawasaki,¶ Michiko Fujihara,†,§ Masaki Watanabe,† Yuta Takamura,†
Maho Takioku,† Hiromi Nishioka,† Yasuo Takeuchi,† Makoto Makishima,‖ Tomoharu Motoyama,¶
Sohei Ito,¶ Hiroaki Tokiwa,#,∆ Shogo Nakano*,¶ 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, Tokyo102-8472, Japan.
§AIBIOS Co. Ltd. Tri-Seven Roppongi 8F 7-7-7 Roppongi, Minato-ku, Tokyo 106-0032 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 Integrated Pharmaceutical and Nutritional Sciences, University of
Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 40
#Department of Chemistry, Rikkyo University, 3-34-1 Nishi-ikebukuro, Toshimaku, Tokyo 171-
1
2
3
4
8501, Japan.
5
6
7
8
∆Research Center of Smart Molecules, Rikkyo University, 3-34-1 Nishi-ikebukuro, Toshima-ku,
9
Tokyo 171-8501, Japan.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS. binding assay, fluorescence, retinoid X receptors, rexinoid, screening,
umbelliferone
ABSTRACT. Ligands for retinoid X receptors (RXRs), “rexinoids”, are attracting interest as
candidates for therapy of type 2 diabetes, Alzheimer’s and Parkinson’s diseases. However,
current screening methods for rexinoids are slow and require special apparatus or facilities.
Here, we created 7-hydroxy-2-oxo-6-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphthalen-2-
yl)-2H-chromene-3-carboxylic acid (10, CU-6PMN) as a new fluorescent RXR agonist and
developed a screening system of rexinoids using 10. Compound 10 was designed based on
the fact that umbelliferone emits strong fluorescence in a hydrophilic environment, but the
fluorescence intensity decreases in hydrophobic environments such as the interior of proteins.
The developed assay using 10 enabled screening of rexinoids to be performed easily within a
few hours by monitoring changes of fluorescence intensity with widely available fluorescence
microplate readers, without the need for processes such as filtration.
ACS Paragon Plus Environment

Page 3 of 40
Journal of Medicinal Chemistry
INTRODUCTION
1
2
3
4
5
6
7
8
9
There are 48 types of human nuclear receptors that regulate gene expression in response to
binding of small-molecular ligands1,2. Among them, retinoid X receptors (RXRs) often act as
heterodimers with a variety of other receptors, and ligands targeting RXR (known as rexinoids3)
can control the activities of these RXR heterodimers. Since RXRs are activated by low
concentrations of 9-cis retinoic acid (1)4 (Figure 1), 1 had been considered as an endogenous
ligand, though it has not been detected in blood5. More recently, 9-cis-13,14-dihydroretinoic
acid (2) (Figure 1) has been suggested to be an endogenous ligand of RXR6. Many synthetic
rexinoids have also been reported. Among them, bexarotene (3) (Figure 1) is an activator
(agonist) of RXR and is used clinically to treat cutaneous T cell lymphoma (CTCL)7. It may also
be effective against diabetes3, Alzheimer's disease8,9, and Parkinson's disease10,11. The basis of
these activities is thought to be activation of RXR heterodimers with nuclear receptors such as
peroxisome proliferator-responsive receptors (PPARs), liver X receptor (LXR), and nuclear
receptor related 1 protein (Nurr1), which are involved in the regulation of sugar and lipid
metabolism, inflammation and other processes. The ability of RXR agonists to regulate these
RXR heterodimers via their action on RXR alone is known as the permissive mechanism of
activation12. On the other hand, polyunsaturated fatty acids such as docosahexaenoic acid
(DHA, 4) (Figure 1) are natural rexinoids13,14. They are reported to improve memory and
metabolic syndrome15, and foods containing them are used as functional foods. Other
compounds, including various environmental pollutants, can have adverse effects on nuclear
receptors, for example serving as endocrine disruptors16. Therefore, there is a great demand
for a simple, low-cost assay system that can identify compounds with RXR ligand activity, not
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
only for drug discovery, but also for studies on functional foods and endocrine disruptors.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Chemical structures of 1–11 and molecular design strategies used here.
Existing rexinoid search technologies are summarized below and in Table 1.
1) Reporter gene assay4: This assay evaluates not only transcriptional activation, but also cell
permeability of test compounds, because it employs living cells transfected with reporter
plasmid encoding a reporter gene such as luciferase under an RXR response element.
However, it requires a long time (3–4 days).
ACS Paragon Plus Environment

Page 5 of 40
Journal of Medicinal Chemistry
2) Time-resolved fluorescence resonance energy transfer (TR-FRET)17: This assay reveals
1
2
3
4
5
6
7
recruitment of a co-activator peptide by test compounds within a short period. However, a
special plate reader is required.
8
9
3) Binding assay using a radioisotope-labeled ligand18: This assay can assess the binding
affinity of test compounds for RXR with high sensitivity, using a small amount of RXR peptide.
However, this method requires a radioactive tracer (such as [3H]1), as well as bound/free (B/F)
separation. The procedure is complicated and involves radioactivity-related safety issues.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4) Fluorescence quenching method (FQM)19,20: This method is based on the quenching of
autofluorescence derived from tryptophan in the ligand-binding site (LBD) of RXR. However,
the autofluorescence of tryptophan is weak and the fluorescence wavelengths (Ex/Em 280/337
nm) are outside the range of usual plate readers.
Table 1. Existing Rexinoid Search Technologies
Measurement Principle
Advantages
Disadvantages
Evaluation of transcriptional Transcription
Reporter gene activation ability of test activation by test
A long time is
required
days).
(3–4
assay4
compounds
overexpressing RXRs
using
cells compounds can
be measured.
Time-resolved
fluorescence
resonance
Activation of RXR
FRET phenomenon between a
terbium-labeled RXR and a
A special plate
by
test
reader
is
compounds can
be measured in a
short period.
coactivator
fluorescein
labeled
with
required for TR-
FRET.
energy transfer
(TR-FRET)17
Binding
using
radiolabelled
assay
Bound/free (B/F)
Highly sensitive.
Based on competition of test
compounds and radio-labeled
separation
necessary,
is
the
Small amount of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 40
compound (ex. compound for binding to RXRs receptor protein is procedure
is
&
1
2
3
4
[3H]1)18
sufficient.
complicated,
safety issues.
5
6
Usual
plate
readers are not
suitable (Ex/Em
280/337 nm are
7
8
9
Measurement can
Binding of test compound be done just by
quenches the autofluorescence mixing RXR-LBD
Fluorescence
quenching
method
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
outside
range
usual
auto
of tryptophan in RXRs.
and
test
(FQM)19,20
&
compound.
fluorescence is
weak).
Other reported RXR ligand screening methods include cell-based RXR assays such as a yeast
reporter assay21; magnetic microbead affinity selection screening (MagMASS) to find RXR
ligands in complex mixtures22,23 and virtual screening approaches for RXR ligands24. However,
cell-based RXR assays require a long time (3–4 days). MagMASS requires HPLC-MS apparatus.
As for virtual screening, it is always necessary to experimentally confirm the binding affinity of
candidate RXR ligands.
Thus, to overcome the disadvantages of existing methods, we set out to design a fluorescent
rexinoid that would be suitable for use in a simple, highly sensitive RXR binding assay. We
previously reported the fluorescent RXR agonist 5 (Figure 1), which has a fluorescent
carbostyril structure 6 in place of the 1,1,4,4-tetramethyltetralin structure of 3. Compound 5
has been used in a fluorescence polarization-based assay targeting RXRs.25 However, the RXR
agonist activity of 5 was much weaker than that of 3 (the EC50 values were 1 µM and 20 nM,
respectively), and its relatively poor binding ability to RXRs was a significant drawback.
ACS Paragon Plus Environment

Page 7 of 40
Journal of Medicinal Chemistry
Therefore, in this study, we aimed to create a new fluorescent RXR agonist that would be
suitable for highly sensitive RXR ligand screening using a standard fluorescence plate reader.
1
2
3
4
5
6
7
8
RESULTS AND DISCUSSION
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Pohl et al. reported fluorescent retinoids targeting cellular retinoid binding protein II
(CRABPII)26,27. Compared to CRBPII, less space is available in the ligand-binding site of RXR25,
and therefore we considered that the acidic domain of RXR agonists would have to be
converted to a fluorophore.
Umbelliferone (7) is a fluorophore that can be detected with a widely available fluorescent
filter set (Ex 360 nm/ Em 465 nm)28. Its derivative 8 with a carboxy group at the 3-position has
a 7-hydroxyl group, which exists in OH form in a hydrophobic environment, but as the anion
(O-) in an aqueous environment, resulting in an increase of fluorescence intensity29. Thus, if the
receptor-bound compound is displaced by a competively binding RXR ligand, an increase of
the fluorescence intensity will be observed. With this in mind, we designed and synthesized 10
(Design 1), which has an umbelliferone skeleton, and its regioisomer 11 (Design 2) based on
the chemical structure of the RXR agonist CD3254 (9).30 Since 9 has a cinnamic acid structure,
we anticipated that this structure could be developed towards an umbelliferone structure.
Compounds 10 and 11 were synthesized according to Schemes 1 and 2.
Scheme 1.a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 40
Me
Me
Me
Me
Me
Me
Me
Me
Me
OH
OH
Me
Me
Me
Me
Cl
Me
Br
Me
a
c
b
1
2
3
Cl
78% for 2 steps
73%
B(OH)2
Me
Me
Me
4
5
12
13
14
15
6
7
8
9
HO
Br
OH
MOMO
Br
OMOM
CHO
HO
OH
e
d
f
28%
92%
86%
CHO
CHO
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
17
18
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
OMOM
Me
OH
Me
OH
g
h
Me
Me
88%
55%
OMOM
OH
O
19
20
10
CHO
CHO
O
CO2H
aReagents and conditions: a) conc. HCl, b) 2-bromotoluene, AlCl3, DCM, 78% for 2 steps. c) 1)
n-BuLi, B(Oi-Pr)3, THF. 2) 2 M HCl, 73%. d) Br2, AcOH, 28%. e) MOMCl, DIPEA, DMF, 92%. f) 4,
Pd(PPh3)4, 2 M Na2CO3, EtOH, toluene, 86%. g) 4 M HCl/EtOAc, 88%. h) Meldrum's acid,
piperidine, EtOH, 55%.
Scheme 2. a
a
b
c
MOMO
OMOM
B(OH)2
23
q.y.
49%
78%
HO
OH
MOMO
OMOM
21
22
Me
Me
Me
Me
Me
Me
Me
Me
OMOM
Me
OH
Me
OH
d
e
Me
Me
Me
Me
Me
39%
54%
MOMO
HO
HO
26
CHO
24
25
Me
Me
Me
Me
Me
f
OH
46%
O
O
CO2H
11
ACS Paragon Plus Environment

Page 9 of 40
Journal of Medicinal Chemistry
aReagents and conditions: a) MOMCl, NaH, DMF, q.y. b) 1) n-BuLi, Et2O, 2) B(OMe)3, 49%. c)
14, Pd(PPh3)4, K3PO4, DME, H2O, 78%. d) 4 M HCl/EtOAc, 54%. e) POCl3, DMF, 39%. f)
Meldrum's acid, piperidine, EtOH, 46%.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In reporter assay, 10 showed potent RXR activation (EC50 = 22 nM toward human RXR), like
3 (Figure 2A). In binding assay using human RXR ligand-binding domain (hRXR-LBD), both
compounds showed concentration-dependent competition with [3H]1 (Figure 2B). The IC50
values were calculated from the binding data, and the inhibition constant (Ki) of each
compound was determined using the Cheng-Prusoff equation.31 The Ki values for 3 and 10
were 150 nM and 230 nM, respectively. However, the behavior of 10 was different from that of
3, in that [3H]1 was not completely displaced from the receptor by 10. In addition, at the lower
concentration of 10, the % of bound [3H]1 was increased compared to that in the absence of
10 as a competitor.
A
B
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
200
150
100
50
0
-9 -8 -7 -6 -5
-8 -7 -6 -5
Concentration (Log M)
Concentration (Log M)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 40
Figure 2. Transactivation of human RXR (hRXR) by 3, 10 and 11 and binding to human
RXR ligand-binding domain (hRXR-LBD) of 3 and 10 in the presence of [3H]1. (A) Dose-
dependent agonistic activity curves of 3 (open circles), 10 (closed circles) and 11 (closed
diamonds) toward hRXR in COS-1 cells. The data are shown as relative transactivation activity
with respect to 1 µM 3, taken as 1. Data are mean ± SD (n = 3). (B) Dose-dependent binding of
[3H]1 with hRXR-LBD in the presence of 3 (open circles) or 10 (closed circles). The data are
shown as % of the value of [3H]1 in the absence of any competitive sample. [3H]1 and hRXR-
LBD were used at concentrations of 10 nM and 50 nM, respectively. Data are mean ± SD (n =
3).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To reveal the reason for these incongruities, we conducted X-ray structure analysis. The
crystal structure of the hRXRα-LBD/10 (green) complex was determined at 2.65 Å resolution
(Figure 3, Table S1) (PDB code 6JNO). The structure turned out to be tetrameric and the
coupling ratio of 10 to RXRα-LBD was predicted as 4 to 2 based on the electron density map
of hRXRα-LBD/10 complex. In addition, this confirmed that 10 binds at the dimer-dimer
interface, in contrast to 1 (PDB ID: 3OAP),32 and this may be the reason why [3H]1 was not
completely displaced from the receptor by 10. Though ligand-binding at the dimer-dimer
interface has already been reported for RXR antagonists rhein (PDB ID: 3R2A),33 K8003 (PDB ID:
5TBP),34 and K8008 (PDB ID: 4N8R)35 (Figure S1), this is the first example of an RXR agonist
binding at this location. Since 10 has RXR agonist activity and showed concentration-
dependent competition with [3H]1, it seems likely that 10 also binds to the ligand-binding
pocket (LBP) of the hRXRα-LBD, like 3, as well as to the dimer-dimer interface. However, it is
ACS Paragon Plus Environment

Page 11 of 40
Journal of Medicinal Chemistry
still not clear why a low concentration of 10 promoted the binding of [3H]1. One possibility
might be that binding of 10 to the interface induces some structural change at the LBP.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Structure of the complex of 10 with the hRXRα-LBD. (A) Comparison of ligand-
binding modes in hRXRα-LBD. The hRXRα-LBD in the complexes with 10 (PDB ID: 6JNO) and 1
(PDB ID: 3OAP)32 is shown in green and orange, respectively. The binding site of 10 to the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 40
hRXRα-LBD is located at the interface between the two domains of the hRXRα-LBD. (B) The
2Fo-Fc electron density map (blue color) was contoured at 0.50 σ. (C) Comparison of ligand-
binding modes in hRXRα-LBD. The binding modes of rhein (PDB ID: 3R2A),33 K8003 (PDB ID:
5TBP),34 K8008 (PDB ID: 4N8R),35 and 10 are shown in cyan, magenta, yellow and green,
respectively.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The fluorescence properties of 10 were examined. The maximum absorbance of 10 in 0.1 N
aqueous solution of sodium hydroxide was 396 nm, the maximum fluorescence wavelength
was 453 nm, and the fluorescence quantum yield was 0.47 (Table 2, Figure S2). In addition,
when the fluorescence intensity was measured with fixed excitation and fluorescence
wavelengths (Ex 360 nm/ Em 465 nm) in several solvents with different polarity, the
fluorescence intensity of 10 decreased significantly as the solvent polarity decreased (Figure
4B). This suggests that a decrease in the fluorescence intensity of 10 can be expected upon
binding to RXR, due to the more hydrophobic environment, as anticipated. Thus, the binding
ability of 10 to RXR was determined based on the change in the fluorescence intensity of 10
upon binding to the hRXR-LBD (Figure 4C). The experimental protocol is described in the
supporting information. After sample mixing at room temperature, the measured intensity
became stable at 2 hours, so all fluorescent measurements were made under these conditions.
The fluorescence intensity of 10 decreased in a sigmoid manner depending on the
concentration of hRXR-LBD (Figure 4C), and a Hill coefficient (n) of 1.9 and dissociation
constant (Kd) of 233 nM were obtained (Figure 4D). A Hill coefficient of >1 indicates positive
cooperativity, i.e., binding of the first ligand facilitates binding of a second ligand at another
ACS Paragon Plus Environment

Page 13 of 40
Journal of Medicinal Chemistry
site on the dimeric receptor complex. Imai et al. reported similar results for a fluorescent ligand
targeting the homodimeric estrogen receptor.36 The fact that the Hill coefficient is greater than
1 implies that binding of 10 to RXRs induces positive cooperativity, i.e., binding of 10 to RXRs
induces further binding of 10.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Absorption and Fluorescence Data of 10 in 0.1 N NaOH and MeOH, respectively
Abs
Em

Solvent [L/(mol·cm)]
0.1 N
[nm]
Φ
0.468 ±
0.003
NaOH
1.78 × 104
396
352
453
453
MeOH
6.71 × 103
0.095 ± 0.02
n = 3, mean ± SD
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 40
1
2
3
4
A
B
120
50000
0.1 N NaOH
100
80
60
40
20
0
40000
30000
20000
10000
0
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
H
M
C
c
y
.
e
H
C
1
e
A
C
0
200
400
600
800
c
l
o
p
O
H
N
e
l
3
N
h
-10000
s
N
e
Wavelength (nm)
(
a
x
p
O
a
H
n
H
e
7
.
9
)
C
D
100
10
1
Hill coefficient
n = 1.93 ± 0.25
Kd = 233 ± 44 nM
10000
5000
0
-5000
-10000
-15000
-20000
-25000
0.1
0.01
0.1
1
10
0.01
0.1
1
10
RXR–LBD
Concentration (µM)
Rf (µM)
Figure 4. Characterization of 10. Relative absorbance and fluorescence spectra of 10 in 0.1 N
NaOH. (A) Fluorescence spectra were obtained at Ex/Em = 396 nm/453 nm. (B) Relative
fluorescent intensity data of 100 nM 10 in various solvents containing 1% DMSO (Ex/Em = 360
nm/465 nm). (C and D) Specific binding curve (C) and Hill plot (D) of 10 with hRXR-LBD. “Rf”
is the concentration of free receptor. Final concentration of 10 was 300 nM in assay buffer
containing 10 mM Hepes, 150 mM NaCl, 2 mM MgCl2, 5 mM DTT (pH7.9). DMSO final
concentration was 1%. Ex/Em=360 nm/465 nm. Data are mean ± SD (n = 4).
ACS Paragon Plus Environment

Page 15 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Next, the affinity of test compounds for RXR was determined using 10 and hRXR-LBD in a
384-well plate. To each well (4 wells per sample), 10 µL of hRXRα-LBD at 1 µM (final
concentration 0.5 µM), 5 μL of the sample at 4 times higher concentration than the final
concentration, and 5 µL of 400 nM 10 (final concentration 100 nM) were added. The mixture
was incubated for 2 hours at room temperature and the fluorescence intensity was measured.
This incubation time was chosen based on evaluation of the Z’-factor at incubation times of
0.5, 1, 2, 3 hours for 1, 10 and hRXRα-LBD (Figure S3). The IC50 was calculated from the
fluorescence intensity data, and the Ki of each compound was determined using the Cheng-
Prusoff equation.31 The experimental protocols and data analysis methods are described in the
supporting information. The Ki values are shown in Table S2. Figure 5A compares the results of
radioisotope (RI) experiments using [3H]1 and the assay using 10 to measure the binding
ability of known rexinoids to hRXRα-LBD (Figures S4 and S5). Because of limited availability, we
used commercially available hRXRα-LBD for the RI experiment, and our prepared protein for
the binding assay using compound 10. The correlation coefficient between Ki values
determined by the two methods was about 0.7, despite the difference in hRXRα-LBD
preparation used. However, rexinoids that gave small Ki values in RI, such as NEt-3IB (Figure
S4), gave Ki values one order of magnitude higher as determined in the assay using 10. The
reason for this is not clear, but might be related to failure of 10 to completely displace [3H]1,
observed in the RI experiments. Because polyunsaturated fatty acids are known to function as
rexinoids13,14, their binding ability to RXR was also examined using the developed method
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(Figure 5B and Table S3). A higher concentration of DHA (4) was required for binding to RXR,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 40
compared to bexarotene (3), while no binding of palmitic acid was observed, in accordance
with previous reports. Interestingly, our system revealed the binding affinity of
eicosapentaenoic acid (EPA), which has not reported before. We also evaluated some readily
available RXR antagonists using this method and confirmed the order of binding affinity
(Figures S6 and S7, and Table S3). Thus, our system provides simple and sensitive methodology
with readily available equipment for the initial screening of RXR ligands, though the agonistic
or antagonistic activity of hits would need to be separately examined.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 17 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
1000
800
600
400
200
0
A
y = 1.1266x - 209.3
R² = 0.7505
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
10 (100 nM)
hRXR-LBD (500 nM)
n = 4
0
200 400 600 800 1000
Ki by 10 (nM)
B
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
Bexarotene
DHA
EPA
Palmitic acid
0.01
0.1
1
10
100
Concentration of test
compound (µM)
Figure 5. Competitive binding assay using 10. (A) Correlation of hRXR-LBD binding affinity
values obtained with 10 and [3H]1. The Ki values obtained with 10 and those obtained with
[3H]1 are plotted on the X and Y axes, respectively. (B) Relative fluorescence intensity curves of
test compounds in the assay using 10. These data were calculated by subtraction of the
fluorescence values for the hRXR-LBD with the test compound in the absence of 10 from
those in the presence of 10. The y-axis is shown as the relative value to 10 µM 3, taken as 1
(positive control). This assay was performed in a 384-well plate (4 wells/sample). Final
ACS Paragon Plus Environment
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Journal of Medicinal Chemistry
Page 18 of 40
concentrations of 10 and hRXR-LBD were 100 nM and 500 nM, respectively. Each compound
except for 3 was added at 0.20, 0.39, 0.78, 1.56, 3.13, 6.25, 12.5, 25, 50 and 100 µM (final
concentration). Assay buffer contained 10 mM Hepes, 150 mM NaCl, 2 mM MgCl2, 5 mM DTT
(pH 7.9). DMSO final concentration was 1%. Ex 360 nm/Em 465 nm. Data are mean ± SD (n =
4).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSIONS
In this study, we designed and synthesized an umbelliferone derivative 10 as a new
fluorescent rexinoid, and applied it to develop a fluorescence-based RXR-binding assay system
for screening rexinoids. This assay system is simple, sensitive, and can be performed in a few
hours with widely available fluorescence microplate readers, without the need for processes
such as filtration. Though there remains an issue concerning the correlation of this assay
system with RI experiments, this assay is expected to be useful for simple, rapid, low-cost
screening studies. Hit compounds can then be followed up by assessing transcriptional activity
by means of reporter gene assay. We believe it will be useful not only for identifying RXR
binders in drug discovery studies, but also for studies of functional foods and endocrine
disruptors, though it should be noted that fluorescence-based assays often suffer from
interference when used to screen natural products. A similar strategy may also be applicable to
other receptors.
EXPERIMENTAL SECTION
Chemistry.
ACS Paragon Plus Environment

Page 19 of 40
Journal of Medicinal Chemistry
General. All solvents and chemicals were used as purchased without further purification. The
progress of all reactions was monitored by thin layer chromatography (TLC) on 0.2 mm thick
TLC plates (Merck, glass-backed, silica gel 60 F245), and spots were detected under UV light.
Silica gel 60 (Kanto Chemical, particle size 0.04–0.05 mm) was used for purification by flash
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
13
column chromatography. H NMR and C NMR spectra were measured on a Varian Mercury-
300 (1H 300 MHz, C 75 MHz) spectrometer or a Varian VXR-400 (1H 400 MHz) spectrometer
13
at room temperature. Deuterated chloroform and dimethyl sulfoxide (CDCl3 and DMSO-d6)
were used as solvents in all routine NMR measurements. Chemical shifts are reported in ppm
relative to the respective deuterated solvent peak, CHCl3 (δ 7.26 ppm), DMSO-d6 (δ 2.50 ppm)
1
13
for H NMR and CDCl3 (δ 77 ppm) for C NMR, and coupling constants are given in Hz. FAB-
MS spectra were measured on a JEOL JMS-700 mass spectrometer. Electrospray ionization (ESI)
mass spectra were measured on a Bruker MicrOTOF. Melting points were determined with a J-
Science RFS-10 hot stage microscope. Elemental analysis was carried out with a Perkin Elmer
2400 Series II CHNS/O Elemental Analyzer and results were within ± 0.4% of the theoretical
values. The purity of all tested compounds was >95%, as confirmed by HPLC.
Preparative HPLC. Preparative HPLC was performed using
a
Shimadzu liquid
chromatographic system (Kyoto, Japan) consisting of an LC-10AS pump, SPD-10AV UV−vis
spectrophotometric detector, CTO-10AS column oven and C-R5A Chromatopac. The samples
(2 mL) were injected and the chromatographic analyses were carried out on an YMC Pack ODS-
AM (10 mm i.d. × 250 mm, 5 µm, YMC Co., Ltd.) at 40 °C, using MeOH:H2O (85/15 + 0.1%
formic acid, v/v) as the mobile phase. The flow rate was 0.7 mL/min and the absorbance at 260
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 40
nm was monitored. All the solvents were of analytical grade and were used directly without
purification.
1
2
3
4
5
6
7
8
9
Analytical HPLC. Analytical HPLC was performed using a Shimadzu liquid chromatographic
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 onto an Inertsil ODS-3 (4.6 i.d. × 100 mm, 3 μm, GL Sciences, Tokyo, Japan) fitted with
a guard column of Inertsil ODS-3 (4.0 mm i.d. × 10 mm, 3 μm, GL Sciences) at 40 °C, using
MeOH:H2O (85/15 + 0.1% formic acid, v/v) as the mobile phase. The flow rate was 0.7 mL/min
and the absorbance at 260 nm was monitored.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7-Hydroxy-2-oxo-6-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-2H-
chromene-3-carboxylic acid (10, CU-6PMN). To a solution of 20 (41 mg, 0.12 mmol) and
Meldrum's acid (17 mg, 0.12 mmol) in ethanol (1 mL) was added piperidine (50 µL). The
reaction mixture was refluxed for 2 h, then poured into water (30 mL), acidified with sat. NH4Cl
aq and extracted with EtOAc (2 × 100 mL). The organic layer was washed with water (2 × 50
mL) and brine (50 mL), dried over Na2SO4, and filtered. The filtrate was evaporated in a rotary
evaporator. The residue was recrystallized from CH2Cl2/MeOH to yield 27 mg of 10 as a yellow
particulate solid (55%). Mp: 229.6−230.1 ºC 1H-NMR (300 MHz, DMSO-d6) δ: 8.70 (s, 1H), 7.63
(s, 1H), 7.18 (s, 1H), 7.03 (s, 1H), 6.85 (s, 1H), 2.05 (s, 3H), 1.65 (s, 4H), 1.27 (s, 6H), 1.22 (s, 6H).
13C-NMR (75 MHz, CDCl3) δ: 165.07, 163.81, 161.22, 156.61, 151.86, 147.10, 144.34, 134.33,
132.46, 130.14, 129.57, 129.03, 128.88, 112.70, 111.17, 103.66, 51.30, 45.42, 35.29, 35.26, 34.62,
34.46, 32.30, 32.17, 19.78. FAB-MS m/z: 407 [M + H]+. HRMS (ESI−) m/z: [M − H]− Calcd for
ACS Paragon Plus Environment

Page 21 of 40
Journal of Medicinal Chemistry
C25H25O5 405.1707; Found 405.1704. Anal. Calcd for C25H26O5·1/4H2O: C, 73.06; H, 6.50. Found:
C, 72.69; H, 6.13. Analytical HPLC: 9.8 min; >95% purity.
1
2
3
4
5
6
7
8
9
7-Hydroxy-2-oxo-8-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-2H-
chromene-3-carboxylic acid (11). To a solution of 26 (18 mg, 0.053 mmol) and Meldrum's
acid (8.0 mg, 0.053 mmol) in ethanol (1 mL) was added piperidine (50 µL). The reaction mixture
was refluxed for 2 h, then poured into water (30 mL). The mixture was acidified with 2 M HCl aq
and extracted with EtOAc (2 × 50 mL). The organic layer was washed with water (2 × 50 mL)
and brine (50 mL), dried over Na2SO4, and filtered. The filtrate was evaporated in a rotary
evaporator. The residue was chromatographed (SiO2, EtOAc/n-hexane = 1/3, v/v) to yield 20
mg of a yellow solid, which was further purified by preparative HPLC (MeOH/H2O = 85/15 +
0.1% formic acid, v/v). The eluate was evaporated under reduced pressure to afford 10 mg of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
11 as a pale-yellow solid (46%). Mp: 221.7–223.0 ºC H-NMR (300 MHz, CDCl3) δ: 8.92 (s, 1H),
7.68 (d, 1H, J = 8.8 Hz), 7.32 (s, 1H), 7.16 (d, 1H, J = 8.8 Hz), 7.10 (s, 1H), 2.06 (s, 3H), 1.72 (s, 4H),
1.37 (s, 3H), 1.33 (s, 3H), 1.26 (s, 6H). 13C-NMR (75 MHz, CDCl3) δ: 164.67, 163.34, 160.63,
155.83, 151.80, 146.97, 144.00, 134.72, 131.23, 129.55, 128.54, 124.10, 116.64, 114.99, 112.47,
110.44, 34.92, 34.83, 34.23, 34.05, 32.02, 31.81, 31.76, 31.70, 19.31. FAB-MS m/z: 407 [M + H]+.
HRMS (ESI−) m/z: [M − H]− Calcd for C25H25O5 405.1707; Found 405.1704. Anal. Calcd for
C25H26O5·3/2H2O: C, 69.26; H, 6.76. Found: C, 69.65; H, 6.79. Analytical HPLC: 8.8 min; >97%
purity.
6-Bromo-1,1,4,4,7-pentamethyl-1,2,3,4-tetrahydronaphthalene (14). This compound was
prepared according to reference 39. 2,5-Dimethyl-2,5-hexanediol (12, 2.0 g, 13.6 mmol) was
dissolved in conc. HCl (20 mL) and the solution was stirred at room temperature for 4 h. The
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 40
mixture was filtered. The filtrate was washed with water and dissolved in CH2Cl2 (150 mL) and
the organic solution was washed with water (50 mL). The resulting solution was dried over
MgSO4, filtered, and evaporated using a rotary evaporator to afford 13. Then 13 was dissolved
in CH2Cl2 again, and 2-bromotoluene (3.3 mL, 27 mmol) was added to the solution. Next, AlCl3
(170 mg) was slowly added at 0 ºC under Ar. The reaction mixture was stirred at room
temperature for 20 h, and then diluted with n-hexane (150 mL). The organic layer was
collected, washed with water (2 × 50 mL) and brine (50 mL), dried over MgSO4, and filtered.
The filtrate was evaporated in a rotary evaporator, and the residue was recrystallized from
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
methanol to yield 3.0 g of 14 as yellow needles (78%). H-NMR (300 MHz, CDCl3) δ: 7.42 (s,
13
1H), 7.14 (s, 1H), 2.34 (s, 3H), 1.65 (s, 4H), 1.25 (s, 12H). C-NMR (75 MHz, CDCl3) δ: 144.60,
144.06, 134.57, 130.14, 128.89, 122.12, 34.91, 34.87, 34.00, 33.91, 31.76, 31.72, 22.53.
(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)boronic
acid
(15).
This
compound was prepared according to reference 39. A solution of 1.6 M n-BuLi in n-hexane
(1.5 mL, 2.4 mmol) was slowly added to a solution of 14 (560 mg, 2 mmol) in THF (6 m) at -78
ºC under an Ar atmosphere. The reaction mixture was stirred at -78 ºC for 20 min and then a
solution of triisopropyl borate (1.4 mL, 6 mmol) in THF (0.5 mL) was added to it. Stirring was
continued at -78 ºC for 2 h. The reaction mixture was poured into 2 M HCl aq (10 mL) and the
mixture was stirred at room temperature for 1 h and diluted with EtOAc (150 mL). The organic
layer was collected, washed with water (2 × 50 mL) and finally with brine (50 mL). The resulting
solution was dried over MgSO4, filtered and evaporated using a rotary evaporator. The residue
was chromatographed using (SiO2, EtOAc/n-hexane = 1/5, v/v) to yield 360 mg of 15 as a
yellow oil (73%). 1H-NMR (300 MHz, CDCl3) δ: 8.28 (s, 1H), 7.21 (s, 1H), 2.81 (s, 3H), 1.72 (s, 4H),
ACS Paragon Plus Environment

Page 23 of 40
Journal of Medicinal Chemistry
13
1.34 (s, 6H), 1.32 (s, 6H). C-NMR (75 MHz, CDCl3) δ: 149.42, 142.89, 141.49, 136.34, 128.59,
1
2
3
4
35.06, 35.03, 34.37, 33.82, 31.87, 31.56, 22.70.
5
6
7
8
9
5-Bromo-2,4-dihydroxybenzaldehyde (17). This compound was prepared according to
reference 40. To a solution of 2,4-dihydroxybenzaldehyde (16, 3.3 g, 24 mmol) in acetic acid
(24 mL) was slowly added bromine (1.24 mL, 24 mmol). The reaction mixture was stirred at
room temperature for 15 h, then poured into water (35 mL), and the whole was filtered. The
crude product was chromatographed (SiO2, EtOAc/n-hexane = 1/4, v/v) and recrystallized from
EtOAc/n-hexane to yield 1.46 g of 17 as pale brown needles (28%). 1H NMR (300 MHz, CDCl3)
δ: 11.25 (s, 1H), 9.70 (d, 1H, J = 1 Hz), 7.66 (s, 1H), 6.63 (s, 1H), 6.11 (s, 1H).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5-Bromo-2,4-bis(methoxymethoxy)benzaldehyde (18). DIPEA (520 µL, 3 mmol) and
chloromethyl methyl ether (324 µL, 4.3 mmol) were added to a solution of 17 (220 mg, 1
mmol) in DMF (3 m) at 0 ºC under an Ar atmosphere. The reaction mixture was stirred at room
temperature for 27 h, then poured into sat. NH4Cl aq (60 mL) and extracted with EtOAc (3 × 40
mL). The organic layer was collected, washed with water (2 × 60 mL) and brine (60 mL), dried
over MgSO4, and filtered. The filtrate was evaporated in a rotary evaporator. The residue was
chromatographed (SiO2, EtOAc/n-hexane = 1/5, v/v) to yield 280 mg of 18 as a white solid
(92%). 1H-NMR (300 MHz, CDCl3) δ: 10.29 (s, 1H), 8.03 (s, 1H), 7.01 (s, 1H), 5.32 (s, 2H), 5.29 (s,
13
2H), 3.53 (s, 6H). C-NMR (75 MHz, CDCl3) δ: 187.01, 160.24, 159.10, 132.36, 120.61, 105.39,
101.92, 94.75, 56.59, 56.52.
2,4-Bis(methoxymethoxy)-5-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphthalen-2-
yl)benzaldehyde (19). To a solution of 15 (200 mg, 0.8 mmol) and 18 (150 mg, 0.49 mmol) in
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 40
toluene (2 mL) and ethanol (1 mL) were added tetrakis(triphenylphosphine)palladium (30 mg,
0.026 mmol) and 2 M Na2CO3 aq (0.5 mL). The reaction mixture was stirred at 100 ºC for 24 h,
then diluted with EtOAc (150 mL). The organic layer was collected, washed with sat. NH4Cl aq
(100 mL), water (100 mL) and brine (100 mL), dried over MgSO4, and filtered. The filtrate was
evaporated in a rotary evaporator. The residue was chromatographed (SiO2, EtOAc/n-hexane =
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1/6, v/v) to yield 180 mg of 19 as a colorless oil (86%). H-NMR (300 MHz, CDCl3) δ: 10.39 (s,
1H), 7.69 (s, 1H), 7.13 (s, 1H), 7.04 (s, 2H), 5.34 (s, 2H), 5.16 (s, 2H), 3.57 (s, 3H), 3.39 (s, 3H), 2.10
(s, 3H), 1.69 (s, 4H), 1.31 (s, 6H), 1.25 (s, 6H). 13C-NMR (75 MHz, CDCl3) δ: 188.31, 160.69,
160.47, 143.81, 141.76, 134.10, 133.38, 131.20, 128.19, 127.41, 126.24, 119.63, 101.02, 94.80,
94.52, 56.55, 56.36, 35.09, 33.87, 33.79, 31.84, 31.81, 31.49, 22.56, 19.66, 14.10.
2,4-Dihydroxy-5-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphthalen-2-
yl)benzaldehyde (20). To a solution of 19 (171 mg, 0.4 mmol) in EtOAc (2 mL) was added 4 M
HCl/EtOAc (2 mL). The reaction mixture was stirred at room temperature for 4 h and
evaporated in a rotary evaporator. The residue was chromatographed (SiO2, EtOAc/n-hexane =
1/5, v/v) to yield 117 mg of 20 as a yellow solid (88%). 1H-NMR (300 MHz, CDCl3) δ: 11.37 (1H,
s), 9.72 (1H, s), 7.32 (1H, s), 7.25 (1H, s), 7.12 (1H, s), 6.56 (1H, s), 5.52 (1H, s), 2.11 (4H, s), 1.71
13
(4H, s), 1.32 (6H, s), 1.27 (6H, s). C-NMR (75 MHz, CDCl3) δ: 194.45, 163.51, 160.93, 145.77,
143.47, 135.99, 134.27, 130.69, 128.80, 128.77, 121.74, 115.19, 102.82, 60.53, 34.91, 34.05, 33.94,
31.84, 31.72, 21.02, 19.38, 14.10.
1,3-Bis(methoxymethoxy)benzene (22). This compound was prepared according to
reference 41. To a suspension of 60% NaH (920 mg, 23 mmol) in DMF (10 mL) was added a
solution of resorcinol (21, 1.1 g, 10 mmol) in DMF (10 mL). The mixture was stirred at 0 ºC
ACS Paragon Plus Environment

Page 25 of 40
Journal of Medicinal Chemistry
under an Ar atmosphere for 15 min, then chloromethyl methyl ether (1.67 mL, 22 mmol) was
added, and stirring was continued at room temperature for 27 h. The reaction mixture was
poured into ice-cold water (50 mL) and extracted with EtOAc (2 × 100 mL). The organic layer
was collected, washed with water (2 × 50 mL) and brine (50 mL), dried over MgSO4, and
filtered. The filtrate was evaporated in a rotary evaporator to yield 1.98 g of 22 as a clear oil
(q.y.). 1H-NMR (400 MHz, CDCl3) δ: 7.19 (t, 1H, J = 8.2 Hz), 6.74 (d, 1H, J = 2.2 Hz), 6.70 (dd, 2H,
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
J = 8.2, 2.2 Hz), 5.16 (s, 4H), 3.48 (s, 6H). C-NMR (75MHz, CDCl3) δ: 158.17, 129.70, 109.31,
104.67, 94.12, 55.66.
(2,6-Bis(methoxymethoxy)phenyl)boronic acid (23). This compound was prepared
according to reference 41. To a solution of 22 (297 mg, 1.5 mmol) in Et2O (6 mL) was slowly
added 1.6 M n-BuLi in n-hexane solution (1.13 mL, 1.8 mmol) at 0 ºC under an Ar atmosphere.
The reaction mixture was stirred at room temperature for 3 h, then trimethyl borate (250 µL,
2.25 mmol) was added, and stirring was continued at room temperature for 1 h. The reaction
mixture was poured into 2 M HCl aq (10 mL). The whole was stirred at room temperature for 1
h. The mixture was filtered, washed with water, and dried under reduced pressure to yield 177
mg of 23 as a pale-yellow solid (49%). 1H-NMR (300 MHz, CDCl3) δ: 7.36 (t, 1H, J = 8.3 Hz), 7.23
13
(s, 2H), 6.88 (d, 2H, J = 8.3 Hz), 5.30 (s, 4H), 3.51 (s, 6H). C-NMR (75 MHz, CDCl3) δ: 163.02,
133.00, 108.23, 94.78, 56.58.
6-(2,6-Bis(methoxymethoxy)phenyl)-1,1,4,4,7-pentamethyl-1,2,3,4-
tetrahydronaphthalene (24). To a solution of 13 (113 mg, 0.4 mmol) and 23 (145 mg, 0.6
mmol) in DME (4 mL) were added tetrakis(triphenylphosphine)palladium (69 mg, 0.06 mmol),
K3PO4 (255 mg, 1.2 mmol) and H2O (1.3 mL). The reaction mixture was refluxed for 30 min,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 40
then poured into water (30 mL) and extracted with EtOAc (3 × 30 mL). The organic layer was
collected, washed with water (2 × 50 mL) and brine (50 mL), dried over MgSO4, and filtered.
The filtrate was evaporated in a rotary evaporator. The residue was chromatographed (SiO2,
EtOAc/n-hexane = 1/30, v/v) to yield 124 mg of 24 as a white solid (78%). 1H-NMR (400 MHz,
CDCl3) δ: 7.24 (t, 1H, J = 8.4 Hz), 7.13 (s, 1H), 7.05 (s, 1H), 6.89 (d, 2H, J = 8.4 Hz), 5.04 (d, 2H, J =
6.6 Hz), 4.96 (d, 2H, J = 6.6 Hz), 3.27 (s, 6H), 2.07 (s, 3H), 1.68 (s, 4H), 1.55 (s, 3H), 1.30 (s, 6H),
1.23 (s, 6H). 13C-NMR (75 MHz, CDCl3) δ: 155.50, 143.21, 141.10, 133.93, 131.01, 129.93, 128.86,
128.34, 127.09, 122.91, 109.69, 109.59, 94.93, 94.43, 55.88, 35.33, 35.26, 34.92, 34.89, 33.93,
33.82, 31.93, 31.90, 31.76, 19.55.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)benzene-1,3-diol (25). To a
solution of 24 (108 mg, 0.27 mmol) was added 4 M HCl/EtOAc (2 mL). The reaction mixture
was stirred at room temperature for 3 h, then poured into water (30 mL) and extracted with
EtOAc (30 mL). The organic layer was collected, washed with water (30 mL) and brine (30 mL),
dried over MgSO4, and filtered. The filtrate was evaporated in a rotary evaporator. The residue
was chromatographed using (SiO2, EtOAc/n-hexane = 1/8, v/v) to yield 45 mg of 25 as a pale-
1
yellow solid (54%). H-NMR (300 MHz, CDCl3) δ: 7.31 (s, 1H), 7.18 (s, 1H), 7.15 (d, 1H, J = 8.0
Hz), 6.59 (d, 2H, J = 8.0 Hz), 4.72 (s, 2H), 2.10 (s, 3H), 1.71 (s, 4H), 1.33 (s, 6H), 1.25 (s, 6H). 13C-
NMR (75 MHz, CDCl3) δ: 153.48, 146.69, 144.66, 144.42, 135.90, 130.16, 129.63, 129.53, 129.35,
128.91, 125.92, 114.76, 107.13, 34.95, 34.89, 34.19, 34.07, 31.88, 31.75, 31.70, 19.05.
2,4-Dihydroxy-3-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaphthalen-2-
yl)benzaldehyde (26). POCl3 (100 µL) was added to DMF (350 µL) at 0 ºC. The reaction
mixture was stirred for 1.5 h, and then a solution of 25 (40 mg, 0.128 mmol) in DMF (1.5 mL)
ACS Paragon Plus Environment

Page 27 of 40
Journal of Medicinal Chemistry
was added to it, and stirring was continued at room temperature for 48 h. The reaction mixture
was poured into ice-cold water (30 mL), adjusted to pH 3 with sat. NaHCO3 aq., and extracted
with EtOAc (2 × 50 mL). The organic layer was washed with water (2 × 50 mL) and brine (50
mL), dried over MgSO4, and filtered. The filtrate was evaporated in a rotary evaporator. The
residue was chromatographed (SiO2, EtOAc/n-hexane = 1/8, v/v) to yield 17 mg of 26 as a
pale-yellow solid (39%). 1H-NMR (300 MHz, CDCl3) δ: 11.71 (s, 1H), 9.76 (s, 1H), 7.48 (d, 1H, J =
8.4 Hz), 7.28 (s, 1H), 7.11 (1H, s), 6.69 (1H, d, J = 8.4 Hz), 5.69 (s, 1H), 2.10 (s, 3H), 1.70 (s, 4H),
1.33 (s, 3H), 1.31 (s, 3H), 1.26 (s, 6H). 13C-NMR (75 MHz, CDCl3) δ: 194.69, 161.13, 160.63,
146.03, 143.59, 135.27, 134.89, 129.18, 128.68, 125.87, 115.72, 115.10, 108.20, 35.01, 34.95,
34.14, 32.07, 31.85, 31.73, 31.69, 19.20, 14.18.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In vitro assays
Reporter gene assay. Luciferase reporter gene assays using COS-1 cells were performed
according to reference 42. Briefly, COS-1 cells were transfected with three kinds of vectors
consisting of human RXR receptor alpha, a luciferase reporter gene under the control of the
appropriate RXR response element (CRBPII-tk-Luc), and secreted alkaline phosphatase (SEAP)
gene as a background. Test compound solutions (DMSO concentration: 1%) were added to a
suspension of transfected cells seeded at about 2.0×104 cells/well in 96-well white plates. After
incubation in a humidified atmosphere of 5% CO2 at 37ºC for 24 h, 25 µL of the medium was
used for analysis of SEAP activity, 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 the SEAP activities. The assays were
carried out three times in triplicate.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 40
Ligand binding assay using radioligand. Competitive binding assay using 9-cis-[11,12-3H]-
retinoic acid ([3H]1) was performed as described in references 43 and 44 with some
modifications. [3H]1, NET1151, 1.11–2.22 TBq/mmol (30–60 Ci/mmol) 7.4 MBq/mL (0.2
mCi/mL)) was purchased from PerkinElmer (USA). hRXRα-LBD (NR2B1, Catalog No: 31135) was
purchased from Active Motif, Inc (USA). Dextran-coated charcoal (DCC) was purchased from
Sigma-Aldrich. Assay buffer contained 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 5
mM dithiothreitol (DTT), 10% glycerol. hRXRα-LBD, 3, and [3H]1 were diluted as required with
the assay buffer described above.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kd determination of [3H]1. To a 1.5 mL Eppendorf tube were added 30 µL of serial dilutions
of hRXRα-LBD (3,000 nM to 3 nM, common ratio 3, 12 final concentration levels), 15 µL of 2%
DMSO assay buffer or 40 µM bexarotene (final concentration 10 µM), and 15 µL of 40 nM [3H]1
(final concentration was 10 nM). Three samples per test condition were used. The mixture was
incubated at 4°C overnight. [3H]1 bound to hRXRα-LBD was separated from free [3H]1 by
adding 20 µL of ice-cold 0.4% DCC suspension to each sample. The samples were placed on
ice for 1 min and the reaction mixture was centrifuged at 12,000 rpm (8,000×g) at 4°C for 5
min. Fifty microliters of each supernatant was poured into a scintillation vial containing 3 mL
scintillation cocktail and bound radioactivity was determined by liquid scintillation counting of
the solution. The specific equilibrium binding constant (Kd) was derived from the specific
binding curve by fitting the data into a sigmoid equation using Excel software. The Kd value of
[3H]1 was obtained as 37 nM. The concentration of hRXRα-LBD was 50 nM.
IC50 determination of test compounds using [3H]1. To a 1.5 mL Eppendorf tube were added
30 µL of 100 nM hRXRα-LBD, 15 µL of serial dilutions of test compounds (10,000 nM to 1 nM,
ACS Paragon Plus Environment

Page 29 of 40
Journal of Medicinal Chemistry
common ratio 3, 9 final concentration levels) in assay buffer containing 8% DMSO, and 15 µL
of 40 nM [3H]1 (final concentration 10 nM). Three samples per test condition were used. The
mixture was incubated at 4°C overnight, and bound radioligand was separated from free as
described above. IC50 was derived from the binding curve by fitting the data to a sigmoid
equation using Excel software. The inhibition constant (Ki) value was calculated as described
below.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ki calculation from IC50. The inhibition constant (Ki) value of each test compound was
calculated by using equation 1.31
Ki = IC50/(1 + [L]/Kd) -------- (1)
where IC50 is the concentration of test compound that inhibits 50% of binding, [L] is the
concentration of [3H]1 or 10 used, and Kd is the Kd value of [3H]1 or 10.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS Publications website at
DOI:
Crystallization and X-ray data collection of hRXRα-LBD/10, UV–vis and fluorescence spectra
measurements, fluorescence quantum yield determination, Kd determination of CU-6PMN
(10), transformation of measured fluorescence intensity into bound receptor concentration,
binding affinity assay of a panel of RXR ligands with hRXRα-LBD using CU-6PMN (10), Z’-
ACS Paragon Plus Environment