Development of Tetrachlorophthalimides as Liver X Receptor β (LXRβ)-Selective Agonists

Article abstract of DOI:10.1002/cmdc.201600305

Liver X receptor (LXR) agonists are candidates for the treatment of atherosclerosis via induction of ABCA1 (ATP-binding cassette A1) gene expression, which contributes to reverse cholesterol transport (RCT) and to cholesterol efflux from the liver and intestine. However, LXR agonists also induce genes involved in lipogenesis, such as SREBP-1c (sterol regulatory binding element protein 1c) and FAS (fatty acid synthase), thereby causing an undesirable increase in plasma and hepatic triglyceride (TG) levels. Recent studies indicate that LXRα contributes to lipogenesis in liver, and selective LXRβ activation improves RCT in mice. Therefore, LXRβ-selective agonists are promising candidates to improve atherosclerosis without increasing plasma or hepatic TG levels. However, the ligand-binding domains in the two LXR isoforms α/β share high sequence identity, and few LXR ligands show subtype selectivity. In this study we identified a tetrachlorophthalimide analogue as an LXRβ-selective agonist. Structural development led to (E)-4,5,6,7-tetrachloro-2-(2-styrylphenyl)isoindoline-1,3-dione (24 a), which shows potent and selective LXRβ agonistic activity in reporter gene assays. In binding assays, compound 24 a bound to LXRβ preferentially over LXRα. It also induced the expression of ABCA1 mRNA but not SREBP-1c mRNA in cells. Compound 24 a appears to be a promising lead compound for therapeutic agents to treat atherosclerosis without the side effects induced by LXRα/β dual agonists.

Full text of DOI:10.1002/cmdc.201600305

DOI: 10.1002/cmdc.201600305
Full Papers
Development of Tetrachlorophthalimides as Liver X
Receptor b (LXRb)-Selective Agonists
Sayaka Nomura,[a] Kaori Endo-Umeda,[b] Makoto Makishima,[b] Yuichi Hashimoto,[a] and
Minoru Ishikawa*[a]
Liver X receptor (LXR) agonists are candidates for the treat-
ment of atherosclerosis via induction of ABCA1 (ATP-binding
cassette A1) gene expression, which contributes to reverse
cholesterol transport (RCT) and to cholesterol efflux from the
liver and intestine. However, LXR agonists also induce genes
involved in lipogenesis, such as SREBP-1c (sterol regulatory
binding element protein 1c) and FAS (fatty acid synthase),
thereby causing an undesirable increase in plasma and hepatic
triglyceride (TG) levels. Recent studies indicate that LXRa con-
tributes to lipogenesis in liver, and selective LXRb activation
improves RCT in mice. Therefore, LXRb-selective agonists are
promising candidates to improve atherosclerosis without in-
creasing plasma or hepatic TG levels. However, the ligand-
binding domains in the two LXR isoforms a/b share high se-
quence identity, and few LXR ligands show subtype selectivity.
In this study we identified a tetrachlorophthalimide analogue
as an LXRb-selective agonist. Structural development led to (E)-
4,5,6,7-tetrachloro-2-(2-styrylphenyl)isoindoline-1,3-dione (24a),
which shows potent and selective LXRb agonistic activity in re-
porter gene assays. In binding assays, compound 24a bound
to LXRb preferentially over LXRa. It also induced the expres-
sion of ABCA1 mRNA but not SREBP-1c mRNA in cells. Com-
pound 24a appears to be a promising lead compound for
therapeutic agents to treat atherosclerosis without the side ef-
fects induced by LXRa/b dual agonists.
Introduction
Liver X receptors (LXRs) are members of the nuclear receptor
(NR) superfamily.[1,2] They are ligand-activated transcription fac-
tors involved in the regulation of cholesterol and lipid metabo-
lism. The physiological ligands for LXRa/b are oxysterols, in-
cluding 22(R)-hydroxycholesterol (1) and 24(S),25-epoxycholes-
terol (2) (Figure 1).[3] Synthetic ligands include T0901317 (3)
and GW3965 (4).[4,5]
LXRs bind to LXR response elements (LXREs), which are pres-
ent within gene promoter regions, as heterodimers with reti-
noid X receptors (RXR) to regulate the expression of target
genes. LXR agonists activate LXRs and induce the expression
of ABCA1/G1 (ATP-binding cassette A1/G1) in macrophages[6,7]
and ABCG5/G8 in the liver and small intestine.[8] They contrib-
ute to reverse cholesterol transport (RCT) and cholesterol
efflux from the liver (as bile acid)[9] and intestine.[10] Therefore,
LXRs are attractive therapeutic targets for the treatment of
atherosclerosis.[11] However, LXR agonists also induce genes in-
volved in lipogenesis, such as SREBP-1c (sterol regulatory bind-
ing element protein 1c)[12] and FAS (fatty acid synthase),[11] re-
sulting in increased plasma and hepatic triglyceride (TG)
levels,[2] which in turn might lead to fatty liver and atheroscle-
rosis as possible side effects.
LXRs include two subtypes, LXRa and LXRb, which have dif-
ferent tissue distributions. LXRa is highly expressed in the liver,
intestine, adipose tissue, and macrophages, whereas LXRb is
present ubiquitously in organs and tissues. LXRa contributes
to lipogenesis in liver, while selective LXRb activation improves
RCT in LXRa-knockout mice.[13,14] Therefore, LXRb-selective ago-
nists are expected to induce RCT and cholesterol efflux from
the liver and to improve atherosclerosis without increasing
plasma or hepatic TG levels. However, the ligand-binding do-
mains (LBDs) in the two LXR isoforms a/b share high sequence
identity (78%), especially in the vicinity of the ligand-binding
pocket.[15] Consequently, most LXR ligands, including T0901317
(3) and GW3965 (4), do not show subtype selectivity. LXR pan-
agonists also activate TG synthesis in the liver by upregulation
of SREBP-1c and FAS, and this activity limits their clinical utility.
A few LXRb-selective agonists have been reported,[16] as
listed in Table 1. Compound 6 showed 4.9-fold selectivity for
LXRb (IC50 =14 nm) in an LXR binding assay,[17] and potently in-
duced ABCA1 in a human whole-blood assay (EC50 =1.2 mm;
55% efficacy relative to full pan-agonist 5b). In mice, com-
pound 6 caused significant ABCA1 mRNA induction in blood
cells, with 2.7-fold induction of SREBP-1c, but no significant ele-
vation of plasma or hepatic triglycerides was observed. It ex-
hibited partial and nonselective agonistic activity for LXRb
(EC50 =250 nm; 72% efficacy relative to full pan-agonist 5a)
[a] S. Nomura, Prof. Y. Hashimoto, Dr. M. Ishikawa
Institute of Molecular and Cellular Biosciences, The University of Tokyo,
1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-0032 (Japan)
E-mail: m-ishikawa@iam.u-tokyo.ac.jp
[b] Dr. K. Endo-Umeda, Prof. M. Makishima
Nihon University School of Medicine,
30-1 Oyaguchi-kamicho, Itabashi-ku, Tokyo 173-8610 (Japan)
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cmdc.201600305: Compound names, characteristics,
yields, purity and spectral information (Table S2), along with Figures S1–2
and RT-PCR primer sequences.
ChemMedChem 2016, 11, 1 – 15
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Figure 1. Physiological and synthetic LXR ligands.
Table 1. Reported LXRb-selective agonists.
6
7
8a: racemate;
9
8b: R enantiomer
Reporter gene assay (LXRa/LXRb)
1.7/0.58
HEK293
>1/>1
Huh7
0.23/0.25
CV-1
0.346/0.063
HepG2
0.244/0.021
HEK293
EC50 [mm]
7/45[c]
HEK293
1/3[c]
38/72[a]
41/98[d]
HepG2
60/56[c]
HEK293
Eff [%]*
CV-1
Huh7
Gene expression (ABCA1)
EC50 [mm]
1.2
hWBA
55[b]
1.09
THP-1
79[c]
>10
THP-1
60[d] at 10 mm
THP-1
0.004
THP-1
101[c]
Eff [%]*
hWBA
THP-1
THP-1
Binding assay (LXRa/LXRb)
IC50 [nm]
Ki [nm]
IC50 [nm]
Ki [nm]
68/14
>1000/53
5000/66
81/3
[ ] Percent efficacy: percent of maximal activity relative to full pan-agonists [a] 5a, [b] 5b, [c] 3, and [d] 4.
*
and LXRa (EC50 =230 nm; 38% efficacy) in a reporter gene
assay in CV-1 (monkey kidney) cells. Compound 7 showed
>20-fold selectivity for LXRb (IC50 =53 nm) in LXR binding
assays.[18] Compound 7 also showed potent agonistic activity in
THP-1 (human acute monocytic leukemia) cells in terms of in-
creasing ABCA1 gene expression (EC50 =1.09 mm) and cholester-
ol efflux from THP-1 foam cells. In the duodenum, compound
7 significantly induced ABCA1, ABCG5, and ABCG8 gene expres-
sion and exhibited weak induction of SREBP-1c gene expres-
sion in LDLRÀ/À mice. It showed selective partial agonistic activ-
ity for LXRb in HEK293 (human embryonic kidney) cells (EC50 =
0.58 mm; 45% efficacy), but did not activate LXRb in Huh7
(human hepatoma cellular carcinoma) cells (EC50 >1 mm; 3.1%
efficacy). Thus, compound 7 appears to be a tissue-selective
&
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
(kidney and macrophages versus liver) LXR ligand. Compound
8b showed 76-fold selectivity for LXRb (IC50 =66 nm) in an LXR
binding assay, and significantly increased ABCA1 gene expres-
sion in THP-1 cells (60% at 10 mm).[19] The racemate 8a also in-
duced higher levels of ABCA1 gene expression in LXRa-null
than in LXRb-null macrophages, and 8a and 8b induced Apo-
A1-dependent cholesterol efflux from primary mouse macro-
phages. These results indicate that 8b is functionally an LXRb-
selective agonist. However, racemate 8a failed to elicit induc-
tion of ABCA1 and SREBP-1c in liver or other tissues of mice,
probably because of a high level of protein binding. Com-
pound 8 exhibited five-fold selectivity for LXRb in a reporter
gene assay with HepG2 (human liver hepatocellular carcinoma)
cells: LXRb EC50 =0.063 mm (98% efficacy) and LXRa EC50 =
0.346 mm (41% efficacy). Compound 9 showed 27-fold selectiv-
ity for LXRb (IC50 =3 nm) in an LXR binding assay, and induced
ABCA1 gene expression in HEK293 cells.[20] Compound 9 also
induced SREBP-1c gene expression (91% relative to 3) in
HepG2 cells. Therefore, overall, these reported agonists show
insufficient LXRb selectivity in reporter gene assays or have
problematic pharmacokinetics. Herein we present tetrachlor-
o(styrylphenyl)phthalimides as a new chemical class of LXRb-
selective agonists that exhibit high selectivity in reporter gene
assays.
We have reported several LXR antagonists based on the
phenylphthalimide structure,[21] and we recently described
a new chemical class of potent and transrepression-selective
LXR modulators.[22] During that work, we found that 29b
(Figure 1), which did not show LXR agonistic activity in a Gal4–
LXRb (LBD) reporter gene assay (data not shown), showed
LXRb-selective agonistic activity, with an EC50 value of 1.38 mm
in a full-length LXRb reporter gene assay system, but did not
show full-length LXRa-agonistic activity even at 30 mm. We
therefore selected 29b as a lead compound for further devel-
opment of LXRb-selective agonists, and in this work we syn-
thesized and evaluated various analogues to clarify the SAR.
Results and Discussion
Synthesis
Styrylphenylphthalimide analogues were synthesized as shown
in Scheme 1. Bromination of alcohols 10c,d yielded benzyl bro-
mides 11 c,d. Benzyl chlorides 11 e,j, and bromides 11 a–d, g–
i and 12l were treated with PPh3 to generate phosphonium
ylides 13a–e, g–j, and 14l. Wittig reaction of ylides 13a–e, g–j
and aldehydes 15l–n, and ylide 14l and aldehydes 16 f,k af-
forded ortho-stilbenes 17a–k, meta-stilbene 18b, and para-stil-
Scheme 1. Reagents and conditions: a) PBr3, CH2Cl2, 08C; b) PPh3, CH3CN, reflux, 38–100%; c) benzaldehydes, 18-crown-6, K2CO3, CH2Cl2, reflux, 74–100%;
d) SnCl2·2H2O, EtOAc, reflux, 33–95%; e) Pd/C, H2, EtOAc, RT, 66–86%; f) phthalic anhydrides, AcOH, reflux or 2008C, 4–92%.
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
bene 19b. Reduction of the nitro group of 17a–k with
SnCl2·2H2O, cyclization with tetracholorophthalic anhydride,
and separation of the E and Z isomers gave the E isomers
24a–k. The nitro group and olefin of 17a–d,18b, and 19b
were reduced with Pd/C, and cyclized with various phthalic an-
hydrides to give 25–33.
Table 2. Agonist SAR: tetrachlorophthalimide analogues.
Other types of tetrachlorophenylphthalimide analogues
38b, 47b, and 48b were synthesized as shown in Scheme 2.
Sonogashira coupling of 34b gave 35b, and subsequent Sono-
gashira coupling of 35b with 36 afforded 37b. Compounds
45b and 46b were prepared from 39b and 40b according to
the same procedure shown in Scheme 1. Cyclization of 37b,
45b and 46b with tetracholorophthalic anhydride gave 38b,
47b and 48b, respectively.
Compd
X
R1
R2
R3
LXRa
LXRb
EC50 [mm][a] Eff [%][b]
EC50 [mm][a] Eff [%][b]
3
–
H
H
H
H
–
H
H
–
H
OMe
–
H
H
H
0.006
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
100
0.021
N.A.
100
25a
25b
25c
25d
29a Cl
29b Cl
N.A.
N.A.
N.A.
OMe OMe
OMe OMe OMe
H
H
Structure–activity relationships
H
OMe
H
H
H
4.98Æ0.29
1.38Æ0.51
8.91Æ0.01
N.A.
117
97
57
First, we investigated the effect of the tetrachloro substituents
on the phthalimide (Table 2). Reporter gene assays were con-
ducted by using full-length LXRs instead of Gal4–LXRs (LBD).
This is why T0901317 (3) showed different agonistic activities
from those of our previously reported data.[22] Compounds
29a–c showed modest LXRb agonistic activity, although com-
pounds 25a–d were inactive. This result indicated that the tet-
rachloro substituents play an important role in LXRb-agonistic
activity. Removal of the methoxy group (29a) decreased the
LXRb-agonistic activity relative to the lead compound 29b.
Compounds 29c and 29d, possessing two or three methoxy
substituents on the terminal benzene ring of the phenethyl
moiety, showed weaker activity than 29b. This result led us to
fix the substituent on the terminal benzene ring as a para-me-
thoxy group, as it afforded the most potent activity among
these compounds (29b, EC50 =1.38 mm).
29c Cl OMe OMe
29d Cl OMe OMe OMe
[a] N.A.: no activity at 30 mm; data are the meanÆSD for all groups.
[b] Percent efficacy: percent of maximal activity relative to 3.
activity than the corresponding tetrarchloro analogues (29a,
EC50 =4.98 mm; 29b, EC50 =1.38 mm). These results indicate that
bulky substituents (Cl and Br) on phthalimide are important,
but the degree of electronegativity is not critical.
Next, the effect of the number and position of chloro sub-
stituents on phthalimide was examined. All the tested com-
pounds possessing only one or two chloro substituents (26a–
28a, 26b–28b) were inactive, suggesting that three or more
chloro substituents on phthalimide (or possibly dichloro sub-
stituents at the X1 and X2 or X1 and X3 positions) are required
for LXRb-agonistic activity. None of the compounds in Table 3
exhibited LXRa-agonistic activity at 30 mm.
We next synthesized tetrafluoro- and tetrabromophthalimide
analogues (Table 3). Tetrafluoro analogues 30a and 30b did
not show LXRb agonistic activity, and tetrabromo analogues
31a (EC50 =8.82 mm) and 31b (EC50 =8.53 mm) showed weaker
Scheme 2. Reagents and conditions: a) 1. Pd(PPh3)2Cl2, CuI, trimethylsilylacetylene, NEt3, 608C, 2. K2CO3, CH2Cl2, RT, 30%; b) Pd(PPh3)2Cl2, CuI, 36, NEt3, 608C,
47%; c) tetrachlorophthalic anhydride, AcOH, reflux, 22–34%; d) PPh3, CH3CN, reflux, 11–35%; e) benzaldehydes, 18-crown-6, K2CO3, CH2Cl2, reflux; f) Pd/C, H2,
EtOAc, RT.
&
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
ligand-binding pocket to accommodate a more elongated
linker, and the rigidity of spacer could contribute to LXRb-ago-
nistic activity. We fixed the linker as a vinyl group because of
the poor solubility of phenylethynyl analogue 38b.
Table 3. Agonist SAR: substituents on phthalimide.
We next explored the effect of substituents on the terminal
benzene ring of the styrylphenyl moiety (Table 5). To examine
the effect of a para-methoxy group, we synthesized several an-
alogues 24e–i possessing various substituents at the para po-
sition. Analogues bearing Me (24e) > OMe (24b) > SMe (24i)
Compd X1 X2 X3 X4
R
LXRa
EC50 [mm][a]
LXRb
EC50 [mm][a] Eff [%][b]
Table 5. Agonist SAR: substituent effects on terminal benzene ring.
29a
29b
30a
30b
31a
31b
26a
26b
27a
27b
28a
28b
Cl Cl Cl Cl
Cl Cl Cl Cl OMe
H
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
4.98Æ0.29
1.38Æ0.51
N.A.
117
97
F
F
F
F
F
F
F
F
H
OMe
H
N.A.
Br Br Br Br
Br Br Br Br OMe
8.82Æ1.51
8.53Æ3.89
N.A.
135
62
H
H
H
H
Cl
Cl
Cl
Cl
Cl Cl
Cl Cl
H
H
H
H
H
H
H
H
Cl
H
OMe
H
OMe
H
N.A.
N.A.
N.A.
N.A.
H
H
Compd
R1
R2
R3
R4
LXRa
EC50 [mm][a]
LXRb
Cl OMe
N.A.
EC50 [mm][a] Eff [%][b]
[a] N.A.: no activity at 30 mm; data are the meanÆSD. [b] Percent efficacy:
percent of maximal activity relative to 3.
24e
24 f
24g
24h
24i
24a
24k
24j
H
H
H
H
H
H
OMe
H
H
H
H
H
H
H
H
Me
CF3
nBu
tBu
SMe
H
H
H
OMe
H
H
H
H
H
H
H
H
H
H
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
0.824Æ0.07
9.83Æ0.59
N.A.
169
80
N.A.
2.97Æ0.48
0.27Æ0.02
0.64Æ0.21
0.62Æ0.06
2.65Æ0.14
0.21Æ0.12
N.A.
114
143
111
120
143
12
We then varied the position of the phenethyl substituent,
spacer length and rigidity (Table 4). Meta (32b) and para (33b)
analogues lacked LXRb-agonistic activity, so we fixed the
ortho-position, and next investigated the effects of length and
rigidity of the spacer. Phenylpropyl (47b) and phenylbutyl
(48b) analogues lacked LXRb-agonistic activity. On the other
hand, styryl (24b, 143% efficacy) and phenylethynyl (38b,
170% efficacy) analogues showed higher percent efficacy
values than the phenethyl analogue (29b, 97% efficacy). These
results indicated that there is no further space available in the
H
H
OMe
H
OMe OMe
24b
24c
24d
H
OMe OMe OMe
[a] N.A.: no activity at 30 mm; data are the meanÆSD. [b] Percent efficacy:
percent of maximal activity relative to 3.
> CF3 (24 f) @ nBu (24g), tBu (24h) showed potency (EC50)
and percent efficacy in the indicated order. We found that the
analogues with bulky substituents 24g (nBu) and 24h (tBu)
lacked LXRb-agonistic activity, whereas small and electron-do-
nating substituents enhanced LXRb-agonistic activity.
Table 4. Agonist SAR: phenethyl group position, spacer length and rigidi-
ty.
We also investigated the substitution pattern of methoxy
group(s). Unsubstituted styrylphenyl analogue 24a showed
the most potent LXRb-agonistic activity, with an EC50 value of
0.27 mm. It exhibited at least 100-fold selectivity over LXRa, be-
cause it had no LXRa-agonistic activity at 30 mm. Among the
monomethoxy compounds, the ortho (24k, EC50 =0.64 mm)
and meta (24j, EC50 =0.62 mm) analogues showed potent
LXRb-agonistic activity compared with the para analogue
(24b, EC50 =2.65 mm). In contrast, di-/trimethoxy analogues
24c,d almost lacked LXRb-agonistic activity. Dimethoxy ana-
logue 24c showed potent partial agonistic activity. These re-
sults indicated that there is also no extra space in the LXRb
ligand-binding pocket in the direction of the para position of
the terminal benzene ring. Finally, we selected the most
potent compound, 24a (EC50 =0.27 mm, 143% efficacy), and in-
vestigated its biological activities in greater detail.
Compd Position
X
LXRa
LXRb
EC50 [mm][a]
EC50 [mm][a] Eff [%][b]
29b
32b
33b
47b
48b
24b
38b
2
3
4
2
2
2
2
(CH2)2
(CH2)2
(CH2)2
(CH2)3
(CH2)4
(E)-CH=CH
CꢀC
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
1.38Æ0.51
N.A.
97
N.A.
N.A.
N.A.
2.65Æ0.14
2.41Æ0.37
143
170
[a] N.A.: no activity at 30 mm; data are the meanÆSD. [b] Percent efficacy:
percent of maximal activity relative to 3.
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Biological evaluations
an LXR agonist binds to the LXR LBD, fluorescein-conjugated
coactivator peptide is recruited, and terbium emission
(l 490 nm) is converted into bound fluorescein emission
(l 520 nm) via FRET. Thus, the ratio Em520/Em490 represents
the agonistic activity. Compound 24a showed dose-dependent
LXRb-partial binding activity (EC50 =98.7 nm, 18% efficacy rela-
tive to T0901317 (3) at 3 mm). On the other hand, compound
24a showed very weak binding activity toward LXRa (6% effi-
cacy at 0.3 mm compared with T0901317 (3) 3 mm). Compound
24a showed the most potent activity at 0.3 mm in the range of
0.03–30 mm. These results may indicate that 24a binds directly
to both the LXRb LBD and LXRa LBD, but recruits the coactiva-
tor peptide preferentially to LXRb rather than LXRa, at least
under our conditions.
Compound 7 has been reported to show different activity de-
pending on the cell type (HEK293, Huh7).[18] To investigate the
generality of the LXRb-selective agonistic activity of 24a, we
chose Caco-2 (human colon cancer) cells for LXR reporter gene
assays (Figure 2), because it has been reported that T0901317
increases cholesterol efflux in Caco-2 cells.[23] Compound 24a
showed LXRb-agonistic activity in Caco-2 cells, but had 59%
maximum efficacy, which is less than that of T0901317 (3).
Compound 24a did not show any LXRa-agonistic activity at
10 mm, indicating that it is an LXRb-selective agonist not only
in HEK293 cells, but also in Caco-2 cells. Next, we considered
the difference of agonistic activities toward human and mouse
LXRs. Compound 24a showed LXRb-agonistic activity with an
EC50 value of 0.30 mm (26% efficacy) in a mouse full-length LXR
reporter gene assay (Figure 3), but showed no LXRa-agonistic
activity at 10 mm. Thus, 24a is an LXRb-selective agonist for
both human and mouse.
To understand the binding mode between 24a and LXR,
24a was computationally docked into the co-crystal structures
of LXRa LBD and LXRb LBD taken from the complexes with
GW3965 (4) (Figure 5). The most favorable conformation had
a
free energy of binding of 12.81 kcalmolÀ1 (LXRa) or
The selectivity of 24a over other NRs (PPARa/g, RARa/b/g,
RORa/d/g, RXRa/b/g, FXR, AR, GR) was evaluated. Compound
24a showed weak agonistic activity toward FXR (8% efficacy)
and RARa (6% efficacy), and inverse agonistic activity toward
RORg (33% inhibition at 10 mm) but did not show activity
toward other NRs. (Supporting Information Figure S1).
12.72 kcalmolÀ1 (LXRb), indicating that 24a would bind to
both LXRs. However, the docking model predicts no hydrogen-
bonding interaction between the LXRs and 24a. In addition,
the figure indicates that 24a would bind similarly to the bind-
ing pockets of both LXRs. Thus, the docking study failed to ex-
plain the selectivity. However, both the docking study and TR-
FRET LXRa/b binding assay indicated that 24a would bind
both LXRs. These results are consistent with the idea that the
LXRb-selective agonistic activity of 24a might result from
We next investigated whether 24a binds directly to LXRa/
b LBD by means of time-resolved fluorescence resonance
energy transfer (TR-FRET) assay (Figure 4). In this assay, when
Figure 2. Human LXRa/b reporter gene assay in Caco-2 cells. RLU: relative luciferase units. Data are the meanÆSD; *p<0.05, **p<0.01, N.S.=not significant
relative to DMSO control.
Figure 3. Mouse LXRa/b reporter gene assay in HEK293 cells. RLU: relative luciferase units. Data are the meanÆSD; *p<0.05, **p<0.01, N.S.=not significant
relative to DMSO control.
&
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 4. Results of TR-FRET LXRa/b binding assays. Data are the meanÆSD; *p<0.05, **p<0.01, N.S.=not significant relative to DMSO control.
Figure 5. Docking simulation of LXRs and 24a. Compound 24a (magenta) was docked into the X-ray crystal structures of complexes between GW3965 (4,
yellow) and the LBDs (white mesh) of a) LXRa and b) LXRb.
a post-binding conformational change and/or differential coac-
tivator recruitment.[19]
Conclusions
We next examined the agonistic activity of 24a by means of
mRNA expression analysis of ABCA1 and SREBP-1c. In THP-
1 cells, 24a induced expression of only ABCA1 mRNA (Fig-
ure 6a), but not SREBP-1c mRNA expression (Supporting Infor-
mation Figure S2). In addition, compound 24a did not induce
SREBP-1c mRNA in HepG2 cells in the concentration range of
0.1–10 mm (Figure 6b). This is consistent with the idea that
24a works as an LXRb-specific partial agonist, which causes
the main effect in macrophages without lipogenic side effects
in liver.
We have identified a series of LXRb-selective agonists with
a tetrachlorophthalimide skeleton and investigated the SAR by
means of LXR reporter gene assay in HEK293 cells. Among
them, compound 24a showed >100-fold selectivity for LXRb
(EC50 =0.27 mm) over LXRa (inactive at 30 mm). Compound 24a
also exhibited clear LXRb selectivity in a reporter gene assay
with another cell line (Caco-2) and with mouse LXRs, and
showed high selectivity over other NRs. In a TR-FRET LXR bind-
ing assay, 24a showed potent and partial LXRb binding activi-
ty, and exhibited high percent efficacy toward LXRb relative to
Figure 6. Real-time PCR analysis of a) ABCA1 mRNA expression in THP-1 cells and b) SREBP-1c mRNA expression in HepG2 cells. Data are the meanÆSD;
*p<0.05, N.S.=not significant relative to DMSO control.
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Triphenyl-(3,4,5-trimethoxybenzyl)phosphonium bromide (13d):
To a solution of 3,4,5-trimethoxybenzyl alcohol (10d) (1.98 g,
10.0 mmol) in CH2Cl2 (15.0 mL) was added dropwise PBr3 (658 mL,
7.00 mmol) at 08C under an argon atmosphere. The mixture was
stirred for 1 h, then poured into ice water (50 mL) and neutralized
with sat. NaHCO3 (50 mL). The aqueous mixture was extracted with
CH2Cl2 (20 mLꢁ3). The organic layer was washed with H2O
(20 mLꢁ2) and then dried over MgSO4. The solvent was removed
under reduced pressure. To a solution of the intermediate 11 d in
toluene (30 mL) was added PPh3 (3.74 g, 14.3 mmol). The mixture
was held at reflux for 30 min, and the precipitate was collected by
filtration to give 13d (2.01 g, 38% in two steps) as a white solid.
1H NMR (CDCl3): d=7.79–7.75 (9H, m), 7.63–7.62 (6H, m), 6.49 (2H,
d, J=2.9 Hz), 5.41 (2H, d, J=14.3 Hz), 3.77 (3H, s), 3.51 ppm (6H,
s).
LXRa. Docking simulations indicated that 24a binds similarly
to both LXRs. We speculate that the LXRb-selective agonistic
activity of 24a arises from a post-binding conformational
change or differential coactivator recruitment, rather than
binding preference. In real-time PCR analysis, compound 24a
induced only ABCA1 mRNA expression (a pharmacological
effect of LXR agonists), but not SREBP-1c mRNA expression (an
undesired effect of LXR pan-agonists). Compound 24a seems
to be a promising lead compound for the development of
agents to treat atherosclerosis without the side effects caused
by LXRa activation.
Experimental Section
(4-Methylbenzyl)triphenylphosphonium chloride (13e): Com-
pound 13e (3.25 g, 8.07 mmol) was prepared from 4-methylbenzyl
chloride (11 e) by means of GP-A in 100% yield as a white solid.
1H NMR (CDCl3): d=7.77–7.73 (9H, m), 7.64–7.63 (6H, m), 6.95–6.93
(4H, m), 5.38 (2H, d, J=14.3 Hz), 2.25 ppm (3H, s).
Chemistry
1
General methods: H NMR spectra were recorded on a JEOL JNM-
ECA500 (500 MHz) spectrometer, and 13C NMR spectra were record-
ed on a JEOL JNM-ECA500 (125 MHz) spectrometer. Chemical shifts
(d) are expressed in parts per million (ppm) relative to tetramethyl-
silane (TMS) as an internal reference, with coupling constants in
Hz. The abbreviations s, d, t, q, br, and m signify singlet, doublet,
triplet, quartet, broad, and multiplet, respectively. Fast atom bom-
bardment mass spectra (FAB-MS) were recorded on a JEOL JMS-
HX110 spectrometer with m-nitrobenzyl alcohol. Electrospray ioni-
zation mass spectra (ESI-MS) were recorded on a Bruker micrOTO-
F II spectrometer. Analytical and preparative thin-layer chromatog-
raphy (TLC and PTLC) was performed on Merck 60 F254 pre-coated
silica gel plates. Flash column chromatography was performed on
silica gel 60 Kanto Kagaku (40–50 mm).
(4-Butylbenzyl)triphenylphosphonium bromide (13g): Com-
pound 13g (2.86 g, 5.91 mmol) was prepared from 4-butylbenzyl
bromide (11 g) by means of GP-A in 100% yield as a white solid.
1H NMR (CDCl3): d=7.79–7.75 (3H, m), 7.74–7.69 (6H, m), 7.67–7.61
(6H, m), 6.97–6.94 (4H, m), 5.30–5.25 (2H, m), 2.55–2.46 (2H, m),
1.51–1.50 (2H, m), 1.29–1.28 (2H, m), 0.95–0.82 ppm (3H, m).
(4-tert-Butylbenzyl)triphenylphosphonium bromide (13h): Com-
pound 13h (1.24 g, 8.55 mmol) was prepared from 1-(bromometh-
yl)-4-(tert-butyl)benzene (11 h) by means of GP-A in 100% yield as
1
a white solid. H NMR (CDCl3): d=7.77–7.72 (9H, m), 7.64–7.63 (6H,
m), 7.18–7.14 (4H, m), 5.31 (2H, d, J=13.7 Hz), 1.23 ppm (9H, s).
General procedure A (GP-A) (Triphenylphosphonium salt): To
a solution of substituted benzyl halide in acetonitrile was added
PPh3 (1.5 equiv). The mixture was heated at reflux for 4.5 h and
then concentrated. The residue was taken up in toluene, and the
product was collected by filtration.
(4-(Methylthio)benzyl)triphenylphosphonium bromide (13i):
Compound 13i (900 mg, 1.87 mmol) was prepared from (4-(bro-
momethyl)phenyl)(methyl)sulfane (11 i) by means of GP-A in 94%
yield as a white solid. H NMR (CDCl3): d=7.79–7.74 (9H, m), 7.66–
7.62 (6H, m), 7.04–6.99 (4H, m), 5.41 (2H, d, J=14.3 Hz), 2.40 ppm
1
Benzyltriphenylphosphonium bromide (13a): Compound 13a
(2.31 g, 5.33 mmol) was prepared from benzyl bromide (11 a) by
means of GP-A in 90% yield as a white solid. H NMR (CDCl3): d=
7.76–7.70 (9H, m), 7.64–7.61 (6H, m), 7.21–7.07 (5H, m), 5.41 ppm
(2H, d, J=14.3 Hz).
(3H, s).
1
(3-Methoxybenzyl)triphenylphosphonium chloride (13j): Com-
pound 13j (3.43 g, 8.18 mmol) was prepared from 1-(chlorometh-
yl)-3-methoxybenzene (11 j) by means of GP-A in 81% yield as
1
a white solid. H NMR (CDCl3): d=7.78–7.76 (9H, m), 7.65–7.62 (6H,
(4-Methoxybenzyl)triphenylphosphonium chloride (13b): Com-
pound 13b (3.77 g, 9.00 mmol) was prepared from 4-methoxyben-
zyl chloride (11 b) by means of GP-A in 90% yield as a white solid.
1H NMR (CDCl3): d=7.78–7.74 (9H, m), 7.65–7.61 (6H, m), 7.01 (2H,
d, J=8.7, Hz), 6.66 (2H, d, J=8.7 Hz), 5.45 (2H, d, J=13.7 Hz),
3.73 ppm (3H, s).
m), 7.02 (1H, dd, J=7.3, 7.3 Hz), 6.76–6.75 (2H, m), 6.65 (1H, d, J=
7.3 Hz), 5.50 (2H, d, J=14.3 Hz), 3.54 ppm (3H, s).
(2-Nitrobenzyl)triphenylphosphonium bromide (14l): Compound
14l (6.82 g, 14.3 mmol) was prepared from 2-nitrobenzyl bromide
(12l) by means of GP-A in 100% yield as a pale-yellow solid.
1H NMR (CDCl3): d=8.16–8.13 (1H, m), 7.94 (1H, d, J=8.6 Hz),
7.81–7.78 (3H, m), 7.71–7.64 (13H, m), 7.50–7.48 (1H, m), 6.15 ppm
(2H, d, J=14.9 Hz).
(3,4-Dimethoxybenzyl)triphenylphosphonium bromide (13c): To
a
solution of 3,4-dimethoxybenzyl alcohol (10c) (2.00 mL,
13.5 mmol) in CH2Cl2 (15.0 mL) was added dropwise PBr3 (658 mL,
7.0 mmol) at 08C under an argon atmosphere. The mixture was
stirred for 1 h, then poured into ice water (50 mL) and neutralized
with sat. NaHCO3 (50 mL). The aqueous mixture was extracted with
CH2Cl2 (20 mLꢁ3). The organic layer was washed with H2O
(20 mLꢁ2) and then dried over MgSO4. The solvent was removed
under reduced pressure. To a solution of the intermediate 11 c in
toluene (30 mL) was added PPh3 (3.74 g, 14.25 mmol). The mixture
was held at reflux for 30 min. The precipitate was collected by fil-
tration to give 13c (4.46 g, 80% in two steps) as a white solid.
1H NMR (CDCl3): d=7.77–7.60 (15H, m), 6.84 (1H, s), 6.62–6.59 (2H,
m), 5.34 (2H, d, J=13.8 Hz) 3.79 (3H, s) 3.54 ppm (3H, s).
(4-Methoxyphenethyl)triphenylphosphonium bromide (41b):
Compound 41b (790 mg, 1.65 mmol) was prepared from 1-(2-bro-
moethyl)-4-methoxybenzene (39b) by means of GP-A in 35% yield
as a white solid. H NMR (CDCl3): d=7.89–7.78 (9H, m), 7.70–7.68
(6H, m), 7.25 (2H, d, J=8.6 Hz), 6.77 (2H, d, J=8.6 Hz), 4.25–4.19
1
(2H, m), 3.75 (3H, m), 3.04–2.98 ppm (2H, m).
(3-(4-Methoxyphenyl)propyl)triphenylphosphonium
bromide
(42b): Compound 42b (230 mg, 0.468 mmol) was prepared from
1-(3-bromopropyl)-4-methoxybenzene (40b) by means of GP-A in
1
11% yield as a white solid. H NMR (CDCl3): d=7.78–7.74 (9H, m),
&
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
7.67–7.64 (6H, m), 7.12 (2H, d, J=8.6 Hz), 6.79 (2H, d, J=8.6 Hz),
3.90–3.85 (2H, m), 3.76 (3H, m), 2.97 (2H, t, J=6.8 Hz), 1.94–
1.86 ppm (2H, m)
7.8, 1.1 Hz), 7.43–7.42 (4H, m), 7.21–7.19 (1H, m), 7.15 (2H, d, J=
8.0 Hz), 7.09 (1H, J=16.0 Hz), 7.03 (1H, J=12.0 Hz), 6.79 ppm (1H,
J=12.0 Hz).
General procedure B (GP-B) (Wittig reaction): To a solution of al-
dehyde in dehydrated CH2Cl2 were added benzyltriphenylphospho-
nium salt (1.0 equiv), potassium carbonate (1.1 equiv) and 18-
crown-6 (0.18 equiv). The mixture was stirred and held at reflux for
6 h, then filtered. The filtrate was concentrated and the residue
was purified by silica gel column chromatography (eluent: n-
hexane/EtOAc=10:1 to 3:1, v/v) to give the product as an E/Z mix-
ture.
1-(4-Butylstyryl)-2-nitrobenzene (17g): Compound 15h (2.13 g,
7.56 mmol, as an E/Z mixture) was prepared from 13g and 15l by
means of GP-B in 100% yield as a yellow oil. 1H NMR (CDCl3) (as
a mixture: E/Z=1:1): d=8.10–8.07 (1H, m), 7.95 (1H, dd, J=8.0,
1.1 Hz), 7.77 (1H, dd, J=8.0, 1.1 Hz), 7.59–7.20 (3H, m), 7.55 (1H, d,
J=16.0 Hz), 7.08 (1H, d, J=16.0 Hz), 6.84 (1H, d, J=12.0 Hz), 6.73
(1H, d, J=12.0 Hz), 2.20–1.24 (12H, m), 0.95–0.88 ppm (6H, m).
1-(4-tert-Butylstyryl)-2-nitrobenzene (17h): Compound 17h
(1.06 g, 3.77 mmol, as an E/Z mixture) was prepared from 13h and
15l by means of GP-B as a yellow oil. It was used for the next reac-
tion without further purification.
1-Nitro-2-styrylbenzene
(17a):
Compound
17a
(1.36 g,
6.00 mmol, as an E/Z mixture) was prepared from 13a and 15l by
1
means of GP-B in 100% yield as a yellow solid. H NMR (CDCl3) (as
a mixture: E/Z=4:6): d=8.10–8.07 (1H, m), 7.96 (1H, dd, J=8.2,
1.2 Hz), 7.77 (1H, dd, J=8.2, 1.2 Hz), 7.61–7.15 (14H, m), 7.09–7.06
(3H, m), 6.89 (1H, d, J=12.3 Hz), 6.77 ppm (1H, d, J=12.3 Hz).
Methyl(4-(2-nitrostyryl)phenyl)sulfane (17i): Compound 17i
(336 mg, 1.24 mmol, as an E/Z mixture) was prepared from 13i and
15l by means of GP-B in 74% yield as a yellow oil. H NMR (CDCl3)
1
1-(4-Methoxystyryl)-2-nitrobenzene (17b): Compound 17b
(1.40 g, 5.48 mmol, as an E/Z mixture) was prepared from 13b and
15l by means of GP-B in 91% yield as a yellow oil. H NMR (CDCl3)
(as a mixture: E/Z=3:5): d=8.08 (1H, d, J=8.0 Hz), 7.94 (1H, d, J=
7.5 Hz), 7.76 (1H, d, J=7.5 Hz), 7.58 (1H, dd, J=7.5, 7.5 Hz), 7.47
(1H, d, J=16.1 Hz), 7.51–7.32 (6H, m), 7.06 (1H, d, J=16.1 Hz), 6.99
(2H, d, J=9.0 Hz), 6.92 (2H, d, J=9.0 Hz), 6.78 (1H, d, J=12.0 Hz),
6.73–6.68 (3H, m), 3.84 (3H, s), 3.75 ppm (3H, s).
(as a mixture: E/Z=1:2): d=8.11–8.07 (1H, m), 7.96 (1H, d, J=
8.6 Hz), 7.76 (1H, d, J=8.6 Hz), 7.61–7.58 (1H, m), 7.56 (1H, d, J=
16.0 Hz), 7.45 (2H, d, J=8.6 Hz), 7.43–7.24 (6H, m), 7.06–7.04 (3H,
m), 6.97 (2H, d, J=8.6 Hz), 6.86 (1H, d, J=12.0 Hz), 6.70 (1H, d, J=
12.0 Hz), 2.51 (3H, s), 2.43 ppm (3H, s).
1
1-(3-Methoxystyryl)-2-nitrobenzene (17j): Compound 17j (1.71 g,
6.70 mmol, as an E/Z mixture) was prepared from 13j and 15l by
means of GP-B in 100% yield as a yellow oil. 1H NMR (CDCl3) (as
a mixture: E/Z=1:1): d=8.10–8.08 (1H, m), 7.98–7.96 (1H, m),
7.78–7.76 (1H, m), 7.61–7.58 (2H, m), 7.43–7.38 (3H, m), 7.31–7.30
(2H, m), 7.14–7.04 (4H, m), 6.90 (1H, d, J=12.0 Hz), 6.89–6.87 (1H,
m), 6.74 (1H, d, J=12.0 Hz), 6.73–6.71 (1H, m), 6.65–6.64 (1H, m),
6.57 (1H, s), 3.86 (3H, s), 3.61 ppm (3H, s).
1,2-Dimethoxy-4-(2-nitrostyryl)benzene (17c): Compound 17c
(2.81 g, 9.84 mmol, as an E/Z mixture) was prepared from 13c and
1
15l by means of GP-B in 94% yield as a yellow oil. H NMR (CDCl3)
(as a mixture: E/Z=1:1): d=8.08 (1H, dd, J=8.0, 1.7 Hz), 7.95 (1H,
dd, J=8.2, 1.3 Hz), 7.76 (1H, dd, J=8.2, 1.3 Hz), 7.61–7.57 (1H, m),
7.47 (1H, d, J=16.0 Hz), 7.41–7.36 (4H, m), 7.10–7.09 (2H, m), 7.05
(1H, d, J=16.0 Hz), 6.89–6.87 (1H, m), 6.81 (1H, d, J=12.0 Hz),
6.71–6.66 (3H, m), 6.50 (1H, d, J=1.7 Hz), 3.95 (3H, s), 3.92 (3H, s),
3.83 (3H, s), 3.54 ppm (3H, s).
1-(2-Methoxystyryl)-2-nitrobenzene (17k): Compound 17k
(1.46 g, 5.71 mmol, as an E/Z mixture) was prepared from 14l and
16k by means of GP-B in 95% yield as a yellow oil. It was used for
the next reaction without further purification.
1,2,3-Trimethoxy-5-(2-nitrostyryl)benzene (17d): Compound 17d
(1.42 g, 4.50 mmol, as an E/Z mixture) was prepared from 13d and
15l by means of GP-B in 75% yield as a yellow solid. 1H NMR
(CDCl3) (as a mixture: E/Z=1:1): d=8.07 (1H, d, J=9.0 Hz), 7.97
(1H, d, J=8.0 Hz), 7.75 (1H, d, J=8.0 Hz), 7.61 (1H, dd, J=8.0,
8.0 Hz), 7.49 (1H, d, J=16.0 Hz), 7.41 (1H, dd, J=8.0 Hz), 7.45 (1H,
d, J=7.5 Hz), 7.40 (1H, dd, J=7.5, 7.5 Hz), 7.33 (1H, d, J=7.5 Hz),
7.02 (1H, d, J=16.0 Hz), 6.86 (1H, d, J=12.0 Hz), 6.66 (1H, d, J=
12.0 Hz), 6.23 (2H, s), 6.75 (2H, s), 3.91 (6H, s), 3.87 (3H, s), 3.78
(3H, s), 3.58 ppm (6H, s).
1-(4-Methoxystyryl)-3-nitrobenzene (18b): Compound 18b was
obtained according to ref. [21].
1-Methoxy-4-(4-nitrostyryl)benzene (19b): Compound 19b was
obtained according to ref. [21].
1-(3-(4-Methoxyphenyl)prop-1-en-1-yl)-2-nitrobenzene
(43b):
Compound 43b (255 mg, 0.946 mmol, as an E/Z mixture) was pre-
pared from 41b by means of GP-B as a yellow oil. It was used for
the next reaction without further purification.
1-(4-(4-Methoxyphenyl)but-1-en-1-yl)-2-nitrobenzene
(44b):
1-(4-Methylstyryl)-2-nitrobenzene (17e): Compound 17 f (2.18 g,
9.10 mmol, as an E/Z mixture) was prepared from 13e and 15l by
means of GP-B in 100% yield as a yellow oil. For characterization,
Compound 44b (129 mg, 0.455 mmol, as an E/Z mixture) was pre-
pared from 42b by means of GP-B as a yellow oil. It was used for
the next reaction without further purification.
1
a portion was purified by PTLC. E isomer: H NMR (CDCl3): d=7.93
(1H, dd, J=7.8, 1.1 Hz), 7.75 (1H, dd, J=7.8, 1.1 Hz), 7.59 (1H, ddd,
J=7.8, 7.8, 1.1 Hz), 7.55 (1H, d, J=16.0 Hz), 7.44 (2H, d, J=8.0 Hz),
7.40–7.37 (1H, m), 7.20 (2H, d, J=8.0 Hz), 7.07 (1H, d, J=16.0 Hz),
2.38 ppm (3H, s). Z isomer: H NMR (CDCl3): d=8.09–8.04 (1H, m),
7.40–7.26 (3H, m), 6.95 (2H, d, J=8.0 Hz), 6.92 (2H, d, J=8.0 Hz),
6.82 (1H, d, J=12.0 Hz), 6.71 (1H, d, J=12.0 Hz), 2.25 ppm (3H, s).
General procedure C (GP-C) (Reduction with SnCl2·2H2O): To a so-
lution of substituted nitrobenzene in EtOAc was added SnCl2·2H2O
(5.0 equiv). The mixture was held at reflux for 4 h, and then the re-
action was terminated by addition of sat. NaHCO3. The resulting
mixture was filtered through a pad of Celite and extracted with
EtOAc. The organic layer was washed with H2O and brine, and
then dried over MgSO4. The solvent was removed under reduced
pressure. The residue was purified by silica gel column chromatog-
raphy (eluent: n-hexane/EtOAc=10:1 to 3:1, v/v) to give the target
compound as an E/Z mixture.
1
2-Nitro-2-(4-(trifluoromethyl)styryl)benzene (17 f): Compound
17 f (1.60 g, 5.45 mmol, as an E/Z mixture) was prepared from com-
pound 14l and 16 f by means of GP-B in 91% yield as a yellow
1
solid/oil mixture. H NMR (CDCl3) (as a mixture: E/Z=1:2): d=8.15–
8.11 (1H, m), 8.01 (1H, dd, J=7.8, 1.1 Hz), 7.76 (1H, d, J=7.8 Hz),
7.69 (1H, d, J=16.0 Hz), 7.64–7.61 (5H, m), 7.46 (1H, ddd, J=7.8,
2-Styrylaniline (20a): Compound 20a (640 mg, 3.26 mmol, as a E/
Z mixture) was prepared from 17a (870 mg, 3.86 mmol) by means
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
of GP-C in 85% yield as a red oil/solid mixture. For characterization,
a portion was purified by PTLC. E isomer: H NMR (CDCl3): d=7.51
2-(4-(methylthio)styryl)aniline (20i): Compound 20i (784 mg,
3.24 mmol) was prepared from 17i by means of GP-C as a yellow
oil. It was used for the next reaction without further purification.
1
(2H, d, J=7.4 Hz), 7.40 (1H, dd, J=7.7, 1.4 Hz), 7.36 (2H, d, J=
7.4 Hz), 7.28–7.23 (1H, m), 7.16 (1H, d, J=16.0 Hz), 7.11 (1H, dd,
J=7.4, 7.4 Hz), 6.99 (1H, d, J=16.0 Hz), 6.81 (1H, dd, J=7.4,
2-(3-Methoxystyryl)aniline (20j): Compound 20j (501 mg,
2.22 mmol, as an E/Z mixture) was prepared from 17j by means of
GP-C. It was used for the next reaction without further purification.
1
7.4 Hz), 6.72 ppm (1H, d, J=7.4 Hz). Z isomer: H NMR (CDCl3): d=
7.27–7.16 (5H. m) 7.07 (2H, d, J=7.5 Hz), 6.69 (2H, d, J=7.5 Hz),
6.64 (1H, d, J=12.0 Hz), 6.50 ppm (1H, d, J=12.0 Hz).
2-(2-Methoxystyryl)aniline (20k): Compound 20k (760 mg,
3.37 mmol, as an E/Z mixture) was prepared from 17k by means of
GP-C. It was used for the next reaction without further purification.
2-(4-Methoxystyryl)aniline (20b): Compound 20b (630 mg,
2.80 mmol, as an E/Z mixture) was prepared from 17b by means of
GP-C in 87% yield as a red solid/oil mixture. 1H NMR (CDCl3) (as
a mixture: E/Z=6:4): d=7.45 (2H, d, J=8.6 Hz), 7.38 (1H, dd, J=
7.7, 1.4 Hz), 7.16 (2H, d, J=7.3 Hz), 7.10–7.07 (3H, m), 7.02 (1H, d,
J=16.3 Hz), 6.93 (1H, d, J=16.3 Hz), 6.92–6.67 (8H, m), 6.59 (1H, d,
J=12.0 Hz), 6.40 (1H, d, J=12.0 Hz), 3.83 (3H, s), 3.76 ppm (3H, s).
General procedure D (GP-D) (Reduction by Pd/C): Substituted ni-
trobenzene was dissolved in EtOAc and hydrogenated with 10%
Pd/C (catalytic amount). The reaction mixture was stirred at room
temperature (RT) and filtered through a pad of Celite. The filtrate
was evaporated under reduced pressure. The residue was purified
by silica gel column chromatography (eluent: n-hexane/EtOAc=
10:1 to 3:1, v/v).
2-(3,4-Dimethoxystyryl)aniline (20c): Compound 20c (E isomer:
763 mg, 3.00 mmol, Z isomer: 459 mg, 1.80 mmol) was prepared
from 17c by means of GP-C in 85% yield as a pale-yellow solid. E
2-Phenethylaniline (21a): Compound 21a (950 mg, 4.82 mmol)
was prepared from 17a by means of GP-D in 86% yield as a color-
1
isomer: H NMR (CDCl3): d=7.38 (1H, dd, J=7.7, 1.1 Hz), 7.10 (1H,
1
ddd, J=7.7, 7.7, 1.2 Hz), 7.07–7.04 (2H, m), 7.03 (1H, d, J=16.0 Hz),
6.93 (1H, d, J=16.0 Hz) 6.87 (1H, d, J=7.7 Hz), 6.81 (1H, ddd, J=
7.7, 7.7, 1.2 Hz), 6.73 (1H, d, J=7.7 Hz), 3.94 (3H, s), 3.91 ppm (3H,
less oil. H NMR (CDCl3): d=7.31–7.27 (m, 2H), 7.23–7.20 (m, 3H),
7.07–7.04 (m, 2H), 6.75 (td, 1H, J=7.3, 1.2 Hz), 6.67 (dd, 1H, J=7.3,
1.2 Hz), 3.5 (brs, 2H), 2.96–2.92 (m, 2H), 2.81–2.78 ppm (m, 2H).
1
s). Z isomer: H NMR (CDCl3): d=7.12 (1H, d, J=7.7 Hz), 7.07 (1H,
2-(4-Methoxyphenethyl)aniline (21b): Compound 21b (1.01 g,
4.44 mmol) was prepared from 17b (1.40 g, 5.48 mmol) by means
of GP-D in 66% yield as a pale-brown solid. 1H NMR (CDCl3): d=
7.11 (2H, d, J=8.9 Hz), 7.05–7.03 (2H, m), 6.83 (2H, d, J=8.9 Hz),
6.77- 6.60 (2H, m), 3.79 (3H, s), 2.94–2.68 ppm (4H, m).
dd, J=7.7, 7.7, Hz), 6.81–6.77 (2H, m), 6.74–6.69 (3H, m), 6.59 (1H,
d, J=12.0 Hz), 6.44 (1H, d, J=12.0 Hz), 3.51 (3H, s), 3.83 ppm (3H,
s).
2-(3,4,5-Trimethoxystyryl)aniline (20d): Compound 20d (E
isomer: 305 mg, 1.07 mmol, Z isomer: 560 mg, 1,96 mmol) was pre-
pared from 17d by means of GP-C in 95% yield as a yellow paste
2-(4-Methoxyphenethyl)aniline (21c): Compound 21c was ob-
tained according to ref. [21].
1
(E) and a red oil (Z). E isomer: a yellow paste. H NMR (CDCl3): d=
7.38 (1H, dd, J=7.6, 1.3 Hz), 7.11 (1H, ddd, J=7.6, 7.6, 1.3 Hz), 7.07
(1H, d, J=16.0 Hz), 6.91 (1H, d, J=16.0 Hz), 6.82 (1H, dd, J=7.6,
7.6 Hz), 6.74–6.72 (3H, m), 3.91 (6H, s), 3.87 ppm (3H, s). Z isomer:
a red oil. H NMR (CDCl3): d=7.11 (1H, d, J=7.4 Hz), 7.07 (1H, ddd,
J=7.4, 7.4, 1.3 Hz), 6.72–6.70 (2H, m), 6.55 (1H, d, J=12.0 Hz), 6.49
(1H, d, J=12.0 Hz), 6.46 (2H, s), 3.80 (3H, s), 3.60 ppm (6H, s).
2-(4-Methoxyphenethyl)aniline (21d): Compound 21d was ob-
tained according to ref. [21].
1
3-(4-Methoxyphenethyl)aniline (22b): Compound 22b was ob-
tained according to ref. [24].
4-(4-Methoxyphenethyl)aniline (23b): Compound 23b was ob-
tained according to ref. [24].
2-(4-Methylstyryl)aniline (20e): Compound 20e (240 mg,
1.15 mmol, as an E/Z mixture) was prepared from 17e by means of
GP-C as an orange solid. It was used for the next reaction without
further purification.
2-(3-(4-Methoxyphenyl)propyl)aniline (45b): Compound 45b was
prepared from 43b (72.0 mg, 0.267 mmol) by means of GP-D as
a pale-yellow oil. It was used for the next reaction without further
purification.
2-(4-(Trifluoromethyl)styryl)aniline
(20 f):
Compound
20 f
(307 mg, 1.17 mmol, as an E/Z mixture) was prepared from 17 f by
means of GP-C in 33% yield as a yellow solid/oil mixture. For char-
acterization, a portion was purified by PTLC. E isomer: 1H NMR
(CDCl3): d=7.62–7.58 (4H, m), 7.41 (1H, dd, J=7.8, 1.2 Hz), 7.26
(1H, d, J=16.0 Hz), 7.14 (1H, ddd, J=7.8, 7.8, 1.2 Hz), 7.01 (1H, d,
J=16.0 Hz), 6.83 (1H, dd, J=7.8, 7.8 Hz), 6.74 (1H, dd, J=7.8,
1.2 Hz), 3.83 ppm (2H, s). Z isomer: H NMR (CDCl3): d=7.43 (2H,
d, J=8.0 Hz), 7.31 (2H, d, J=8.0 Hz), 7.10 (1H, dd, J=7.4, 7.4 Hz),
7.01 (1H, d, J=7.4 Hz), 6.72–6.64 (4H, m), 3.71 ppm (2H, s).
2-(4-(4-Methoxyphenyl)butyl)aniline (46b): Compound 46b was
prepared from 44b (50.4 mg, 0.178 mmol) by means of GP-D as
a pale-yellow oil. It was used for the next reaction without further
purification.
General procedure E (GP-E) (Formation of phthalimide): To a solu-
tion of substituted aniline in AcOH was added phthalic anhydride
(1.5 equiv). The mixture was held at reflux for 3 h, and then the re-
action was terminated by the addition of sat. NaHCO3. The result-
ing mixture was extracted with CHCl3. The organic layer was dried
over MgSO4. The solvent was removed under reduced pressure,
and the residue was purified by PTLC to afford the E-isomer.
1
2-(4-Butylstyryl)aniline (20g): Compound 20g (170 mg,
0.68 mmol, as an E/Z mixture) was prepared from 17g by means of
GP-C as a yellow paste. It was used for the next reaction without
further purification.
(E)-4,5,6,7-Tetrachloro-2-(2-styrylphenyl)isoindoline-1,3-dione
(24a): Compound 24a was prepared from 20a (100 mg,
0.51 mmol) by means of GP-E in 48% yield as a pale-yellow solid.
1H NMR (CDCl3): d=7.82 (1H, d, J=7.3 Hz), 7.52 (1H, dd, J=
7.3 Hz), 7.43–7.41 (1H, m), 7.39 (2H, d, J=7.4 Hz), 7.32–7.20 (4H,
m), 7.11 (1H, d, J=16.0 Hz), 6.84 ppm (1H, d, J=16.0 Hz); 13C NMR
(CDCl3): d=162.8, 140.8, 136.8, 136.4, 132.9, 130.4, 130.3, 129.1,
128.8, 128.7, 128.6, 128.3, 127.5, 127.1, 127.0, 122.8 ppm.
2-(4-tert-Butylstyryl)aniline (20h): Compound 29h (680 mg,
2.70 mmol, as an E/Z mixture) was prepared from 17h by means of
GP-C as an orange solid. It was used for the next reaction without
further purification.
&
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
(E)-4,5,6,7-Tetrachloro-2-(2-(4-methoxystyryl)phenyl)isoindoline-
1,3-dione (24b): Compound 24b was prepared from 20b (50 mg,
0.195 mmol) by means of GP-E in 10% yield as a yellow solid.
1H NMR (CDCl3): d=7.79 (1H, d, J=7.3 Hz), 7.49 (1H, dd, J=
7.3 Hz), 7.38 (1H, dd, J=7.3 Hz), 7.32 (2H, d, J=8.8 Hz), 7.18 (1H,
d, J=7.3 Hz), 7.05 (1H, d, J=16.0 Hz), 6.83 (2H, d, J=8.8 Hz), 6.69
(1H, d, J=16.0 Hz), 3.79 ppm (3H, s).
6.99 (1H, d, J=7.7 Hz), 6.92 (1H, s), 6.83 (1H, d, J=16.0 Hz), 6.81–
6.80 (1H, m), 3.79 ppm (3H, s).
(E)-4,5,6,7-Tetrachloro-2-(2-(2-methoxystyryl)phenyl)isoindoline-
1,3-dione (24k): Compound 24k was prepared from 20k (118 mg,
0.523 mmol) by means of GP-E in 4.8% yield as a yellow solid.
1H NMR (CDCl3): d=7.85 (1H, d, J=7.7 Hz), 7.50 (1H, d, J=7.7 Hz),
7.44–7.38 (2H, m), 7.42 (1H, d, J=16.0 Hz), 7.24–7.19 (2H, m),
6.90–6.84 (2H, m), 6.88 (1H, d, J=16.0 Hz), 3.79 ppm (3H, s).
(E)-4,5,6,7-Tetrachloro-2-(2-(3,4-dimethoxystyryl)phenyl)isoindo-
line-1,3-dione (24c): Compound 24c was obtained according to
ref. [21].
2-(2-Phenethylphenyl)isoindoline-1,3-dione (25a);[4] 2-(2-(4-Me-
thoxyphenethyl)phenyl)isoindoline-1,3-dione (25b); 2-(2-(3,4-Di-
methoxyphenethyl)phenyl)isoindoline-1,3-dione (25c); 2-(2-
(3,4,5-Trimethoxyphenethyl)phenyl)isoindoline-1,3-dione (25d):
Compounds 25a–d were obtained according to ref. [22].
(E)-4,5,6,7-Tetrachloro-2-(2-(3,4,5-trimethoxystyryl)phenyl)isoin-
doline-1,3-dione (24d): Compound 24d was prepared from E-
isomer of 20d (140 mg, 0.490 mmol) by means of GP-E in 92% as
pale-red solid. 1H NMR (CDCl3): d=7.76 (1H, dd, J=7.7 Hz), 7.51
(1H, dd, J=7.7, 7.7 Hz), 7.42 (1H, dd, J=7.7, 7.7 Hz), 7.22 (1H, d,
J=7.7 Hz), 7.00 (1H, d, J=16.0 Hz), 6.74 (1H, d, J=16.0 Hz), 6.60
(2H, s), 3.84 (6H, s), 3.83 ppm (3H, s); MS (FAB) 553 [M]+, 554 [M+
1]+.
5-Chloro-2-(2-phenethylphenyl)isoindoline-1,3-dione (26a): Com-
pound 26a was prepared from 21a (19.6 mg, 0.10 mmol) by
means of GP-E in 42% yield as a brown oil. 1H NMR (CDCl3): d=
7.93 (1H, d, J=1.7 Hz), 7.89 (1H, d, J=8.0 Hz), 7.76 (1H, dd, J=8.0,
1.7 Hz), 7.43–7.39 (1H, m), 7.37–7.35 (2H, m), 7.22–7.17 (3H, m),
7.15–7.12 (1H, m), 7.07–7.05 (2H, m), 2.86 (2H, td, J=8.7, 3.6 Hz),
2.79 ppm (2H, td, J=8.7, 3.6 Hz); MS (FAB) 391 [M]+ 362 [M+1]+
363 [M+2]+.
(E)-4,5,6,7-Tetrachloro-2-(2-(4-methylstyryl)phenyl)isoindoline-
1,3-dione (24e): Compound 24e was prepared from 20e
(24.0 mg, 0.114 mmol) by means of GP-E in 54% yield as a yellow
1
solid. H NMR (CDCl3): d=7.78 (1H, d, J=7.5 Hz), 7.49 (1H, dd, J=
5-Chloro-2-(2-(4-methoxyphenethyl)phenyl)isoindoline-1,3-dione
(26b): Compound 26b was prepared from 21b (18.2 mg,
0.10 mmol) by means of GP-E in 24% yield as a colorless oil.
1H NMR (CDCl3): d=7.92 (1H, d, J=1.4 Hz), 7.89 (1H, d, J=8.0 Hz),
7.76 (1H, dd, J=8.0, 1.4 Hz), 7.40 (1H, dd, J=7.4, 7.4 Hz), 7.35 (1H,
dd, J=7.4, 7.4 Hz), 7.33 (1H, d, J=7.4 Hz), 7.17 (1H, d, J=7.4 Hz),
6.97 (2H, d, J=8.9 Hz), 6.73 (2H, d, J=8.9 Hz), 3.75 (3H, s), 2.81–
2.78 (2H, m), 2.77–2.73 ppm (2H, m); MS (FAB) 391 [M]+ 392 [M+
1]+ 393 [M+2]+.
7.5 Hz), 7.39 (1H, dd, J=7.5 Hz), 7.27–7.25 (2H, m), 7.19 (1H, d, J=
7.5 Hz), 7.10 (2H, d, J=8.0 Hz), 7.05 (1H, d, J=16.0 Hz), 6.79 (1H,
dd, J=16.0 Hz), 2.31 ppm (3H, s).
(E)-4,5,6,7-Tetrachloro-2-(2-(4-(trifluoromethyl)styryl)phenyl)i-
soindoline-1,3-dione (24 f): Compound 24 f was prepared from
20 f (32.0 mg, 0.121 mmol) by means of GP-E in 18% yield as
1
a white solid. H NMR (CDCl3): d=7.80 (1H, d, J=7.5 Hz), 7.547.43
(6H, m), 7.20 (1H, d, J=6.9 Hz), 7.11 (1H, d, J=16.0 Hz), 6.99 ppm
(1H, dd, J=16.0 Hz).
5,6-Dichloro-2-(2-phenethylphenyl)isoindoline-1,3-dione (27a):
Compound 27a was prepared from 21a (19.6 mg, 0.10 mmol) by
means of GP-E in 20% yield as white solid. 1H NMR (CDCl3): d=
8.02 (2H, s), 7.44–7.40 (1H, m), 7.37–7.35 (2H, m), 7.20 (2H, dd, J=
6.9, 6 Hz), 7.17–7.13 (2H, m), 7.05 (2H, d, J=6.9 Hz), 2.87–2.84 (2H,
m), 2.78–2.75 ppm (2H, m); MS (FAB) 396 [M]+ 397 [M+1]+ 398
[M+2]+ 399 [M+3]+.
(E)-2-(2-(4-Butylstyryl)phenyl)-4,5,6,7-tetrachloroisoindoline-1,3-
dione (24g): Compound 24g was prepared from 20g (23.9 mg,
0.10 mmol) by means of GP-E in 31% yield as a pale-yellow solid.
1H NMR (CDCl3): d=7.81 (1H, d, J=7.2 Hz), 7.51 (1H, dd, J=7.2,
7.2 Hz), 7.40 (1H, dd, J=7.2, 7.2 Hz), 7.30 (2H, d, J=8.0 Hz), 7.19
(1H, d, J=7.2 Hz), 7.11 (2H, d, J=8.0 Hz), 7.08 (1H, d, J=16.0 Hz),
6.79 (1H, d, J=16.0 Hz), 2.57 (2H, t, J=7.7 Hz), 1.57–1.56 (2H, m),
1.37–1.28 (2H, m), 0.91 ppm (3H, t, J=7.4 Hz).
5,6-Dichloro-2-(2-(4-methoxyphenethyl)phenyl)isoindoline-1,3-
dione (27b): Compound 27b was prepared from 21b (15.6 mg,
1
(E)-2-(2-(4-(tert-Butyl)styryl)phenyl)-4,5,6,7-tetrachloroisoindo-
line-1,3-dione (24h): Compound 24h was prepared from 20h
(25.6 mg, 0.102 mmol) by means of GP-E in 22% yield as a pale-
0.07 mmol) by means of GP-E in 27% yield as white solid. H NMR
(CDCl3): d=8.02 (2H, s), 7.43–7.40 (1H, m), 7.36–7.34 (2H, m), 7.16
(1H, d, J=8.0 Hz), 6.96 (2H, d, J=8.6 Hz), 6.73 (2H, d, J=8.6 Hz),
3.74 (3H, s), 2.80–2.77 (2H, m), 2.74–2.72 ppm (2H, m); MS (FAB)
426 [M]+ 427 [M+1]+ 428 [M+2]+.
1
yellow solid. H NMR (CDCl3): d=7.81 (1H, d, J=7.7 Hz), 7.51 (1H,
dd, J=7.7 Hz), 7.41 (1H, dd, J=7.7, 7.7 Hz), 7.34–7.33 (4H, s), 7.20
(1H, d, J=7.7 Hz), 7.09 (1H, d, J=16.0 Hz), 6.80 (1H, d, J=16.0 Hz),
1.29 ppm (9H, s).
4,7-Dichloro-2-(2-phenethylphenyl)isoindoline-1,3-dione (28a):
Compound 28a was prepared from 21a (20.0 mg, 0.10 mmol) by
means of GP-E in 55% yield as a yellow oil. 1H NMR (CDCl3): d=
7.63 (2H, d, J=1.7 Hz), 7.42 (1H, dd, J=7.8, 7.8 Hz), 7.36–7.33 (2H,
m), 7.22–7.12 (4H, m), 7.10 (2H, d, J=7.8 Hz), 2.90–2.86 (2H, m),
2.83–2.78 ppm (2H, m).
(E)-4,5,6,7-Tetrachloro-2-(2-(4-(methylthio)styryl)phenyl)isoindo-
line-1,3-dione (24i): Compound 24i (49.0 mg, 0.09 mmol) was pre-
pared from 20i by means of GP-E in 18% yield as a yellow solid.
1H NMR (CDCl3): d=7.81 (1H, d, J=7.7 Hz), 7.51 (1H, dd, J=
7.7 Hz), 7.41 (1H, dd, J=7.7, 7.7 Hz), 7.30 (2H, d, J=8.6 Hz), 7.20
(1H, d, J=7.7 Hz), 7.17 (2H, d, J=8.6 Hz), 7.06 (1H, d, J=16.0 Hz),
6.78 (1H, d, J=16.0 Hz), 2.46 ppm (3H, s).
4,7-Dichloro-2-(2-(4-methoxyphenethyl)phenyl)isoindoline-1,3-
dione (28b): Compound 28b was prepared from 21b (20.0 mg,
0.09 mmol) by means of GP-E in 70% yield as a pale-yellow solid.
1H NMR (CDCl3): d=7.63 (2H, s), 7.46–7.39 (1H, m), 7.35–7.32 (2H,
m), 7.16 (1H, d, J=8.0 Hz), 7.00 (2H, d, J=8.0 Hz), 6.73 (2H, d, J=
8.0 Hz), 3.74 (3H, s), 2.81–2.76 ppm (4H, m).
(E)-4,5,6,7-Tetrachloro-2-(2-(3-methoxystyryl)phenyl)isoindoline-
1,3-dione (24j): Compound 24j was prepared from 20j (117.5 mg,
0.520 mmol) by means of GP-E in 3.9% yield as a white solid.
1H NMR (CDCl3): 7.81 (1H, d, J=7.7 Hz), 7.52 (1H, d, J=7.7 Hz), 7.42
(1H, dd, J=7.7, 7.7 Hz), 7.23–7.21 (2H, m), 7.07 (1H, d, J=16.0 Hz),
4,5,6,7-Tetrachloro-2-(2-(4-methoxyphenethyl)phenyl)isoindo-
line-1,3-dione (29a): Compound 29a was obtained by the proce-
dure according to ref. [22].
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
4,5,6,7-Tetrachloro-2-(2-(4-methoxyphenethyl)phenyl)isoindo-
line-1,3-dione (29b): Compound 29b was prepared from 21b
(100 mg, 0.36 mmol) by means of GP-E in 77% yield as a white
solid. 1H NMR (CDCl3): d=7.42 (1H, d, J=7.6 Hz), 7.36–7.34 (2H,
m), 7.15–7.14 (1H, m), 6.98 (2H, d, J=8.6 Hz), 6.72 (2H, d, J=
8.6 Hz), 3.74 (3H, s), 2.82–2.79 (2H, m), 2.77–2.74 ppm (2H, m); MS
(FAB) 495 [M]+ 496 [M+1]+.
2-((4-Methoxyphenyl)ethynyl)aniline (37b): Compound 37b was
prepared from 35b and 2-iodoaniline (36) (219 mg, 1.00 mmol) by
means of GP-F in 47% yield as a brown solid. H NMR (CD3OD): d=
7.46 (2H, d, J=9.2 Hz), 7.35 (1H, dd, J=7.5, 1.3 Hz), 7.12 (1H, dd,
J=7.5, 1.3 Hz), 6.88 (2H, d, J=9.2 Hz), 6.73–6.70 (2H, m), 4.27 (2H,
brs), 3.83 ppm (3H, s).
1
4,5,6,7-Tetrachloro-2-(2-((4-methoxyphenyl)ethynyl)phenyl)isoin-
doline-1,3-dione (38b): Compound 38b (16.8 mg, 0.03 mmol) was
prepared from 37b by means of GP-F in 34% yield as pale-yellow
solid. 1H NMR (CDCl3): d=7.69–7.68 (1H, m), 7.50–7.46 (2H, m),
7.36–7.32 (1H, m), 7.23 (2H, d, J=9.2 Hz), 6.79 (2H, d, J=9.2 Hz),
3.78 ppm (3H, s).
4,5,6,7-Tetrachloro-2-(2-(3,4-dimethoxyphenethyl)phenyl)isoin-
doline-1,3-dione (29c); 4,5,6,7-Tetrachloro-2-(2-(3,4,5-trimethox-
yphenethyl)phenyl)isoindoline-1,3-dione (29d): Compounds
29c,d were obtained according to ref. [24].
4,5,6,7-Tetrafluoro-2-(2-phenethylphenyl)isoindoline-1,3-dione
(30a): Compound 30a was prepared from 21a (31.0 mg,
0.157 mmol) by means of GP-E in 28% yield as a pale-yellow solid.
1H NMR (CDCl3): d=7.45–7.41 (1H, m), 7.37–7.33 (2H, m), 7.22–7.20
(2H, m), 7.16–7.12 (2H, m), 7,08–7,07 (2H, m), 2.88–2.85 (2H, m),
2.79–2.76 ppm (2H, m); MS (FAB) 399 [M]+ 400 [M+1]+ 401 [M+
2]+.
4,5,6,7-Tetrachloro-2-(2-(3-(4-methoxyphenyl)propyl)phenyl)i-
soindoline-1,3-dione (47b): Compound 47b was prepared from
41b by means of GP-E in 22% yield (three steps) as a pale-yellow
solid. 1H NMR (CDCl3): d=7.43–7.38 (2H, m), 7.30 (1H, d, J=
7.5 Hz), 7.05 (1H, d, J=8.0 Hz), 6.91 (2H, d, J=8.6 Hz), 6.51 (2H, d,
J=8.6 Hz), 3.66 (3H, s), 2.49 (2H, t, J=6.3 Hz), 2.41 (2H, t, J=
8.6 Hz), 1.75–1.69 ppm (2H, m).
4,5,6,7-Tetrafluoro-2-(2-(4-methoxyphenethyl)phenyl)isoindo-
line-1,3-dione (30b): Compound 30b was prepared from 21b
(32.0 mg, 0.14 mmol) by means of GP-E. The target compound was
4,5,6,7-Tetrachloro-2-(2-(4-(4-methoxyphenyl)butyl)phenyl)isoin-
doline-1,3-dione (48b): Compound 48b was prepared from 42b
by means of GP-E in 28% yield (three steps) as a pale-yellow solid.
1H NMR (CDCl3): d=7.42–7.30 (3H, m), 7.11 (1H, d, J=6.9 Hz), 7.01
(2H, d, J=8.3 Hz), 6.75 (2H, d, J=8.3 Hz), 3.76 (3H, s), 2.51–2.46
(4H, m), 1.60–1.56 ppm (4H, m).
1
obtained in 24% yield as a pale-yellow solid. H NMR (CDCl3): d=
7.44–7.40 (1H, m), 7.35–7.33 (1H, m), 7.13–7.11 (2H, m, 7.29), 6.98
(2H, d, J=8.6 Hz), 6.74 (2H, d, J=8.6 Hz) 3.76 (3H, s), 2.82–2.79
(2H, m), 2.77–2.73 ppm (2H, m); MS (FAB) 429 [M]+ 430 [M+1]+
431 [M+2]+
4,5,6,7-Tetrabromo-2-(2-phenethylphenyl)isoindoline-1,3-dione
(31a): Compound 31a was prepared from 21a (10.0 mg,
0.05 mmol) by means of GP-E in 14% yield as a pale-yellow solid.
1H NMR (CDCl3): d=7.44–7.41 (1H, m), 7.36–7.34 (2H, m), 7.22–7.20
(2H, m), 7.16–7.11 (2H, m), 7.08 (2H, d, J=6.9 Hz), 2.89–2.86 (2H,
m), 2.79–2.76 ppm (2H, m).
Docking simulations
In silico molecular docking simulation was done with AutoDock
(version 4.2). The target in this study was the X-ray structure of
compound 24a bound to LXRa (PDB ID: 3IPQ) and LXRb (PDB ID:
1PQ6). The grid covered the binding site of 24a. The Lamarckian
genetic algorithm was used for the molecular docking simulation.
Dockings were automatically ranked by AutoDock according to the
lowest calculated binding energies (kcalmolÀ1).
4,5,6,7-Tetrabromo-2-(2-(4-methoxyphenethyl)phenyl)isoindo-
line-1,3-dione (31b): Compound 31b was prepared from 21b
(10.0 mg, 0.04 mmol) by means of GP-E in 16% yield as a yellow
paste. 1H NMR (CDCl3): d=7.42–7.41 (1H, m), 7.35–7.34 (2H, m),
7.15–7.14 (1H, m), 6.99 (2H, d, J=7.5 Hz), 6.73 (2H, d, J=7.5 Hz),
3.74 (3H, s), 2.82–2.80 (2H, m), 2.77–2.74 ppm (2H, m).
Biology
Cell culture conditions: Human acute monocytic leukemia THP-
1 cells were cultured in RPMI 1640 containing 10% fetal bovine
serum (FBS), penicillin, and streptomycin mixture at 378C in a hu-
midified atmosphere of 5% CO2 in air. Human embryonic kidney
(HEK) 293 cells were cultured in DMEM containing 5% FBS, penicil-
lin, and streptomycin mixture at 378C in a humidified atmosphere
of 5% CO2 in air. Human liver cancer HepG2 cells were cultured in
DMEM containing 10% FBS, penicillin, and streptomycin mixture at
378C in a humidified atmosphere of 5% CO2 in air. Caco-2 cells
were cultured in EMEM containing 10% FBS, penicillin, and strepto-
mycin mixture at 378C in a humidified atmosphere of 5% CO2 in
air.
4,5,6,7-Tetrachloro-2-(3-(4-methoxyphenethyl)phenyl)isoindo-
line-1,3-dione (32b): Compound 32b was obtained according to
ref. [22].
4,5,6,7-Tetrachloro-2-(4-(4-methoxyphenethyl)phenyl)isoindo-
line-1,3-dione (33b): Compound 33b was obtained according to
ref. [22].
1-Ethynyl-4-methoxybenzene (35b): General procedure F (GP-F)
(Sonogashira coupling): To a solution of 1-iodo-4-methoxyben-
zene (34b) (3.0 g, 12.8 mmol), Pd(PPh3)2Cl2 (180 mg, 0.26 mmol),
and CuI (100 mg, 0.52 mol) in triethylamine (10 mL) was added tri-
methylsilylacetylene (2.20 mL, 15.4 mmol) at RT The solution was
stirred for 4 h at 608C, then extracted with CH2Cl2. The organic
layer was washed with brine, dried over MgSO4 and filtered. The
solvent was removed under reduced pressure. To the obtained oil
was added K2CO3 (1.78 g, 12.8 mmol) and CH2Cl2 (10 mL). The mix-
ture was stirred for 1 h at RT, diluted with CH2Cl2 and washed with
H2O. The organic layer was dried over MgSO4, filtered, and concen-
trated to afford 35b as a brown oil (500.1 mg) in 30% yield.
1H NMR (CD3OD): d=7.42 (2H, d, J=8.6 Hz), 6.84 (2H, d, J=
8.6 Hz), 3.81 (3H, s), 2.99 ppm (1H, s).
Transient transfection assays: HEK293 cells or Caco-2 cells were
plated at a density corresponding to 10% confluence in a 96-well
plate 24 h prior to transfection. Cells were co-transfected with
15 ng of expression plasmid of each NR (CMX-hLXRa/b, CMX-
mouse LXRa/b, CMX-GAL4 PPARa/g, CMX-GAL4 RARa/b/g, CMX-
GAL4 CMX-RORa/d/g, CMX-GAL4 RXRa/b/g, CMX-GAL4 FXR, CMX-
AR, CMX-GR), 50 ng of reporter plasmid (CMX-TK-BILE TE-lucꢁ3,
CMX-RORE luc, CMX-ARE luc, CMX-MTV luc), and 10 ng of CMX-b-
galactosidase expression vector. Transfections were performed ac-
cording to the calcium phosphate co-precipitation method (for
HEK293 cells) or with Lipofectamine LTX Reagent (Thermo) (for
&
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
12
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Blanchard, T. A. Spencer, T. M. Willson, J. Biol. Chem. 1997, 272, 3137–
3140.
[4] J. L. Collins, A. M. Fivush, M. A. Watson, C. M. Galardi, M. C. Lewis, L. B.
Moore, D. J. Parks, J. G. Wilson, T. K. Tippin, J. G. Binz, K. D. Plunket, D. G.
Morgan, E. J. Beaudet, K. D. Whitney, S. A. Kliewer, T. M. Willson, J. Med.
Chem. 2002, 45, 1963–1966.
[5] S. B. Joseph, E. McKilligin, L. Pei, M. A. Watson, A. Collins, B. A. Laffitte,
M. Chen, G. Noh, J. Goodman, G. N. Hagger, J. Tran, T. K. Tippin, X.
Wang, A. J. Lusis, W. A. Hsueh, R. E. Law, J. L. Collins, T. M. Willson, P. Ton-
tonoz, Proc. Natl. Acad. Sci. USA 2002, 99, 7604–7609.
Caco-2 cells). After 24 h, transfected cells were treated with test
compounds or dimethyl sulfoxide (DMSO) for 16 h. Agonistic activi-
ty was measured at a concentration of 0.1–30 mm test compounds.
Treated cells were assayed for luciferase activity in a luminometer.
The luciferase activity of each sample was normalized by the level
of b-galactosidase activity. Each transfection was carried out twice
in triplicate. Error bars represent SD. Statistics were calculated
using the unpaired, two-sided Student’s t-test.
LXR binding assay: LanthaScreen TR-FRET LXR competitive bind-
ing assay kit (Invitrogen, Carlsbad, CA, USA) was used as specified
in the supplier documentation. In brief, 4ꢁ test compound were
made up in TR-FRET LXR assay buffer (1% DMSO). A mixture of 4ꢁ
fluorescein-labeled coactivator peptide, 4ꢁ terbium (Tb)-labeled
anti-GST antibody, and 4ꢁ LXR ligand-binding domain (LBD) was
made up in TR-FRET LXR assay buffer. To each well, 15 mL of mix-
ture was added, followed by 5 mL of 4ꢁ test. Plates were stored in
the dark at room temperature for 1.5 h prior to evaluation with an
EnVision reader (PerkinElmer). The TR-FRET ratio was calculated by
dividing the emission signal at 520 nm by the emission signal at
490 nm. TR-FRET assay was carried out twice and Figure 2 shows
representative data. Data points represent triplicate averages; error
bars represent SD. Statistics were calculated using the unpaired,
two-sided Student’s t-test.
[6] A. Venkateswaran, B. A. Laffitte, S. B. Joseph, P. A. Mak, D. C. Wilpitz,
P. A. Edwards, P. Tontonoz, Proc. Natl. Acad. Sci. USA 2000, 97, 12097–
12102.
[7] M. A. Kennedy, A. Venkateswaran, P. T. Tarr, I. Xenarios, J. Kudoh, N. Shi-
mizu, P. A. Edwards, J. Biol. Chem. 2001, 276, 39438–39447.
[8] J. J. Repa, K. E. Berge, C. Pomajzl, J. A. Richardson, H. Hobbs, D. J. Man-
gelsdorf, J. Biol. Chem. 2002, 277, 18793–18800.
[9] L. Yu, J. York, K. von Bergmann, D. Lutjohann, J. C. Cohen, H. H. Hobbs,
J. Biol. Chem. 2003, 278, 15565–15570.
[10] C. Zhao, K. Dahlman-Wright, J. Endocrinol. 2010, 204, 233–240.
[11] S. B. Joseph, B. A. Laffitte, P. H. Patel, M. A. Watson, K. E. Matsukuma, R.
Walczak, J. L. Collins, T. F. Osborne, P. Tontonoz, J. Biol. Chem. 2002, 277,
11019–11025.
[12] J. J. Repa, D. Mangelsdorf, J. Nat. Med. 2002, 8, 1243–1248.
[13] M. N. Bradley, C. Hong, M. Chen, S. B. Joseph, D. C. Wilpitz, X. Wang,
A. J. Lusis, A. Collins, W. A. Hseuh, J. L. Collins, R. K. Tangirala, P. Tonto-
noz, J. Clin. Invest. 2007, 117, 2337–2346.
Real-time quantitative PCR analysis: Total RNA from samples was
prepared using NucleoSpin RNA (Macherey–Nagel). RNA was re-
verse-transcribed into complementary DNA (cDNA) using a Prime
script RT-PCR kit (Takara). The reaction mixture included MgCl2, re-
verse transcriptase buffer, diethyl pyrocarbonate-treated water,
oligo-dT, AMV, inhibitor, and sample RNA. Sequence-specific pri-
mers of ABCA1, SREBP-1c, and GAPDH (glyceraldehyde-3-phosphate
dehydrogenase) were used for PCR (Table S1), and the amplifica-
tions were accomplished under the following conditions: 428C for
30 min, 958C for 5 min, and 48C to the end. PCR products were
electrophoresed on 2% agarose gels and stained with ethidium
bromide. Real-time PCR was performed on a TP800 thermal cycler
Dice Real Time (Takara) using SYBR Green PCR Master Mix (Takara).
All samples were run in duplicate using the primers shown in
Table S1. The mRNA values were normalized to the expression
level of GAPDH mRNA. Error bars represent SD. Statistics were cal-
culated using the unpaired, two-sided Student’s t-test.
[14] E. G. Lund, L. B. Peterson, A. D. Adams, M. H. N. Lam, C. A. Burton, J.
Chin, Q. Guo, S. Huang, M. Latham, J. C. Lopez, J. G. Menke, D. P. Milot,
L. J. Mitnaul, S. E. Rex-Rabe, R. L. Rosa, J. Y. Tian, S. D. Wright, C. P. Spar-
row, Biochem. Pharmacol. 2006, 71, 453–463.
[15] a) M. Fꢂrnegꢃrdh, T. Bonn, S. Sun, J. Ljunggren, H. Ahola, A. Wilhelms-
son, J. A. Gustafsson, M. Carlquist, J. Biol. Chem. 2003, 278, 38821–
38828, b) S. Williams, R. K. Bledsoe, J. L. Collins, S. Boggs, M. H. Lambert,
A. B. Miller, J. Moore, D. D. McKee, L. Moore, J. Nichols, D. Parks,
M. Watson, B. Wisely, T. M. Willson, J. Biol. Chem. 2003, 278, 27138–
27143.
[16] a) C. M. Tice, P. B. Noto, K. Y. Fan, L. Zhuang, D. S. Lala, S. B. Singh, J.
Med. Chem. 2014, 57, 7182–7205, b) T. Matsuda, A. Okuda, Y. Watanabe,
T. Miura, H. Ozawa, A. Tosaka, K. Yamazaki, Y. Yamaguchi, S. Kurobuchi,
M. Koura, K. Shibuya, Bioorg. Med. Chem. 2015, 25, 1274–1278, c) M.
Koura, T. Matsuda, A. Okuda, Y. Watanabe, Y. Yamaguchi, S. Kurobuchi, Y.
Matsumoto, K. Shibuya, Bioorg. Med. Chem. 2015, 25, 2668–2674, d) B.
Hu, R. Unwalla, M. Collini, E. Quinet, I. Feingold, A. Goos-Nilsson, A. Wi-
helmsson, P. Nambi, J. Wrobel, Med. Chem. 2009, 17, 3519–3527, e) S.
Stachel, C. Zerbinatti, M. T. Rudd, M. Cosden, S. Suon, K. K. Nanda, K.
Wessner, J. DiMuzio, J. Maxwell, Z. Wu, J. M. Uslaner, M. S. Michener, P.
Szczerba, E. Brnardic, E. V. Rada, Y. Kim, R. Meissner, P. Wuelfing, Y. Yuan,
J. Ballard, M. Holahan, D. J. Klein, J. Lu, X. Fradera, G. Parthasarathy
et al., J. Med. Chem. 2016, 59, 3489–3498.
Acknowledgements
[17] E. Kick, R. Martin, Y. Xie, B. Flatt, E. Schweiger, T. L. Wang, B. Busch, M.
Nyman, X. H. Gu, G. Yan, B. Wagner, M. Nanao, L. Nguyen, T. Stout, A.
Plonowski, I. Schulman, J. Ostrowski, T. Kirchgessner, R. Wexler, R.
Mohan, Bioorg. Med. Chem. Lett. 2015, 25, 372–377.
[18] B. Hu, R. J. Unwalla, I. Goljer, J. W. Jetter, E. M. Quinet, T. J. Berrodin,
M. D. Basso, I. B. Feingold, A. G. Nilsson, A. Wilhelmsson, M. J. Evans, J. E.
Wrobel, J. Med. Chem. 2010, 53, 3296–3304.
This work was supported in part by Grants-in Aid for Scientific
Research from The Ministry of Education, Culture, Sports, Science
and Technology (Japan), and from the Japan Society for the Pro-
motion of Science, and the Platform for Drug Discovery, Infor-
matics, and Structural Life Science.
[19] V. Molteni, X. Li, J. Nabakka, F. Liang, J. Wityak, A. Koder, L. Vargas, R.
Romeo, N. Mitro, P. A. Mak, H. M. Seidel, J. A. Haslam, D. Chow, T. Tunt-
land, T. A. Spalding, A. Brock, M. Bradley, A. Castrillo, P. Tontonoz, E.
Saez, J. Med. Chem. 2007, 50, 4255–4259.
Keywords: ABCA1 · agonists · atherosclerosis · liver X receptor
(LXR) · SREBP-1c
[20] Y. Zheng, L. Zhuang, K. Y. Fan, C. M. Tice, W. Zhao, C. Dong, S. D. Lotesta,
K. Leftheris, P. R. Lindblom, Z. Liu, Shimada. J, P. B. Noto, S. Meng, A.
Hardy, L. Howard, P. Krosky, J. Guo, K. Lipinski, G. Kandpal, Y. Bukhtiyar-
ov, Y. Zhao, D. Lala, O. R. Van, J. Zhou, G. Chen, Z. Wu, B. M. McKeever,
G. M. McGeehan, R. E. Gregg, D. A. Claremon, S. B. Singh, J. Med. Chem.
2016, 59, 3264–3271.
[21] a) K. Motoshima, T. Noguchi-Yachide, K. Sugita, Y. Hashimoto, M. Ishika-
wa, Bioorg. Med. Chem. 2009, 17, 5001–5014, b) T. Noguchi-Yachide, A.
Aoyama, M. Makishima, H. Miyachi, Y. Hashimoto, Bioorg. Med. Chem.
Lett. 2007, 17, 3957–3961.
[1] J. S. Annicotte, K. Schoonjans, J. Auwerx, Anat. Rec. Part A 2004, 277,
312–316.
[2] J. R. Schultz, H. Tu, A. Luk, J. J. Repa, J. C. Medina, L. P. Li, S. Schwendner,
S. Wang, M. Thoolen, D. J. Mangelsdorf, K. D. Lustig, B. Shan, Genes Dev.
2000, 14, 2831–2838.
[3] a) B. A. Janowski, M. J. Grogan, S. A. Jones, G. B. Wisely, S. A. Kliewer,
E. J. Corey, D. J. Mangelsdorf, Proc. Natl. Acad. Sci. USA 1999, 96, 266–
271, b) J. M. Lehmann, S. A. Kliewer, L. B. Moore, T. A. Smith-Oliver, D. E.
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
13
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[22] a) A. Aoyama, K. Endo-Umeda, K. Kishida, K. Ohgane, T. Noguchi-Ya-
chide, H. Aoyama, M. Ishikawa, H. Miyachi, M. Makishima, Y. Hashimoto,
J. Med. Chem. 2012, 55, 7360–7377, b) S. Nomura, K. Endo-Umeda, A.
Aoyama, M. Makishima, Y. Hashimoto, M. Ishikawa, ACS Med. Chem. Lett.
2015, 6, 902–907.
[24] K. Motoshima, M. Ishikawa, Y. Hashimoto, K. Sugita, Chem. Pharm. Bull.
2011, 59, 880–884.
Received: June 17, 2016
Revised: August 23, 2016
[23] S. Murthy, E. Born, S. N. Mathur, F. J. Field, J. Lipid Res. 2002, 43, 1054–
1064.
Published online on && &&, 0000
&
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
14
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
Clearing the way: Liver X receptor
(LXR) agonists are candidates for the
treatment of atherosclerosis via induc-
tion of ABCA1 gene expression. Howev-
er, they also induce lipogenic genes
such as SREBP-1c. Recent studies indi-
cate that LXRb-selective agonists im-
prove atherosclerosis without lipogenic
side effects. We developed compound
24a (shown), which showed potent and
selective LXRb agonistic activity in sev-
eral assays. Compound 24a may there-
fore serve as a lead for the develop-
ment of anti-atherosclerotic drugs that
lack the side effects of nonselective LXR
agonists.
S. Nomura, K. Endo-Umeda,
M. Makishima, Y. Hashimoto,
M. Ishikawa*
&& – &&
Development of
Tetrachlorophthalimides as Liver X
Receptor b (LXRb)-Selective Agonists
ChemMedChem 2016, 11, 1 – 15
www.chemmedchem.org
15
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ