Modulating retinoid X receptor with a series of (E)-3-[4-hydroxy-3-(3- alkoxy-5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-yl)phenyl]acrylic acids and their 4-alkoxy isomers

Article abstract of DOI:10.1021/jm900096q

Rexinoids are ligands for the retinoid X receptor (RXR) that have great promise for both the prevention and treatment of cancer and metabolic diseases. In this regard, synthetic, functional, and structural investigations into the structure-activity relati

Full text of DOI:10.1021/jm900096q

3150
J. Med. Chem. 2009, 52, 3150–3158
Articles
Modulating Retinoid X Receptor with a Series of
(E)-3-[4-Hydroxy-3-(3-alkoxy-5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-yl)phenyl]acrylic
Acids and Their 4-Alkoxy Isomers†
Efre´n Pe´rez Sant´ın,| Pierre Germain,§, Fabien Quillard,‡ Harshal Khanwalkar, Fa´tima Rodr´ıguez-Barrios,|
,|
,‡,§
Hinrich Gronemeyer,*, Angel R. de Lera,* and William Bourguet*
´
INSERM, U554, 34090 Montpellier, France, UniVersite´ Montpellier 1 et 2, CNRS, UMR5048, Centre de Biochimie Structurale,
34090 Montpellier, France, Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica, UniVersidade de Vigo, 36310 Vigo, Spain, Institut de
Ge´ne´tique et de Biologie Mole´culaire et Cellulaire (IGBMC), BP 10142, 67404 Illkirch Cedex, C. U. de Strasbourg, France
ReceiVed January 23, 2009
Rexinoids are ligands for the retinoid X receptor (RXR) that have great promise for both the prevention and
treatment of cancer and metabolic diseases. In this regard, synthetic, functional, and structural investigations
into the structure-activity relationships of derivatives of the potent RXR agonist (E)-3-[3-(3,5,5,8,8-
pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-4-hydroxyphenyl]acrylic acid (CD3254, 9) have been
conducted. We recently reported on the characterization of a series of C3′-substituted alkyl ether analogues
of 9 (10a-f), which display activities ranging from partial agonists to pure antagonists. The importance of
the position of the alkoxy side chain for ligand activity has been further explored with the synthesis of
C4′-substituted analogues (11a-f). Here we describe the synthesis of compounds 11a-f, which appear
functionally different from their isomeric counterparts, as judged from transactivation assays and fluorescence
anisotropy experiments. We also report on the 2.0 Å resolution structure of RXR in complex with the
parent compound 9, which helps understanding of the impact of the alkyl side chain location on ligand
activity.
Introduction
gene regulation by pure RXR ligands (termed rexinoids) depends
on the nature and the status (ligand-bound or unbound) of the
RXR heterodimerization partner. Thus, in the context of retinoic
acid receptor (RAR)-RXR heterodimers, RAR agonists can
autonomously activate transcription but RXR responds to
rexinoids only in the presence of a RAR ligand.7 Recently,
RXRs have been shown to bind ligands and recruit coactivators
in unliganded (apo)-RAR heterodimers but are unable to activate
transcription because the corepressor does not dissociate (RXR
“subordination”).8 The mechanism of RXR subordination in
RAR-RXR heterodimers does not apply to a group of nuclear
receptors (FXR, LXR, PPAR,...), which form so-called “permis-
sive” heterodimers.9,10 For example, both RXR as well as
peroxisome proliferator activated receptor (PPAR) agonists
activate PPAR-RXR heterodimers, which enables rexinoids to
function as insulin sensitizers in rodent models of noninsulin-
dependent diabetes mellitus.11 This illustrates the potential
therapeutic implications of RXR-selective ligands in combina-
tion with other modulators of the nuclear receptor superfamily
to activate-inactivate the heterodimers and regulate diverse
hormonal pathways.12,13
Retinoid X receptors (RXRs), comprising isotypes R, ꢀ, and
γ, are members of the nuclear receptor (NRa) superfamily of
transcriptional regulators.1,2 Although they are activated by
9-cis-retinoic acid and other ligands such as docosahexaenoic
acid (DHA),3 the true physiological ligand, if existing, remains
to be identified.4 Ligand binding regulates cognate gene
transcription by a still incompletely understood sequence of
events that are initiated upon NR binding to cis-acting DNA
response elements, involve chromatin modification and remod-
eling, and ultimately lead to recruitment of the transcription
machinery. RXR plays a central role in nuclear receptor
signaling because it is the common heterodimerization partner
of multiple nuclear receptors.5 The mechanism of gene regula-
tion through heterodimers is highly complex, and functional
details are emerging only slowly.6 Recent studies revealed that
†
PDB ID Code: The atomic coordinates of the RXRR/CD3254 complex
have been deposited in the Protein Data Bank under the accession code
3FUG.
* To whom correspondence should be addressed. For W.B.: phone, +(33)
4 67 41 7702; fax, +(33) 4 67 41 7913; E-mail: bourguet@cbs.cnrs.fr. For
H.G.: phone, +(33) 3 88 65 3473; fax, +(33) 3 88 65 3437; E-mail,
hg@igbmc.u-strasbg.fr. A.R.de L.: phone, +(34) 9 86 81 2316; fax, +(34)
9 86 81 1940; E-mail, qolera@uvigo.es.
In contrast to the fairly large collection of RXR agonists
known to date,14,15 only a few RXR antagonists have been
reported.16 (2E,4E,6Z)-7-[5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-
2-(n-propoxy)naphthalen-3-yl]-3-methylocta-2,4,6-trienoic acid
(LG100754) acts as an RXR homodimer antagonist but is also
an agonist of several RXR heterodimers (Figure 1).17 (2E,4E,6Z)-
7-[3,5-Di-tert-butyl-2-(2,2-difluoroethoxy)phenyl]-3-methylocta-
2,4,6-trienoic acid (LG101506) shows a homodimer antagonistic
profile as well and synergizes with PPARγ agonists to enhance
‡
INSERM.
§
Universite´ Montpellier 1 et 2, CNRS, UMR5048, Centre de Biochimie
Structurale.
|
Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica, Universidade
de Vigo.
Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire (IGBMC).
a Abbreviations: NR, nuclear receptor; RXR, retinoid X receptor; RAR,
retinoic acid receptor; SRC-1, steroid receptor coactivator 1; TIF-2,
transcriptional intermediary factor 2.
10.1021/jm900096q CCC: $40.75
2009 American Chemical Society
Published on Web 05/01/2009

Modulating RXR
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3151
heterodimers, and their potential clinical application has been
highlighted.25,12 We recently reported26 a new series of RXR
modulators 10a-f based on the structure of CD3254 (9),27
a
potent and selective RXR agonist (Figure 1). Docking studies
suggested that replacing the methyl substituent at the C3′
position of 9 by an extended alkyl ether side chain might prevent
helix H12 of the receptor from adopting its agonist conforma-
tion, in line with current mechanistic views of the antagonistic
action of nuclear receptor ligands.16,28,29 Indeed, a series of
alkoxy analogues of 9 with chains from one to six carbon atoms
(10a-f) exhibited properties ranging from agonist to partial
agonist to antagonist (in particular, (E)-3-[3-(5,5,8,8-tetramethyl-
5,6,7,8-tetrahydro-3-(pentyloxy)naphthalen-2-yl]-4-hydroxyphe-
nyl]acrylic acid, UVI3003) depending on the length of the alkyl
chain. It was thus reasonable to assume that the biological
activities of 10a-f resulted from a modulation of H12 position-
ing induced by the size of the substituent. Indeed, crystal-
lographic and fluorescence anisotropy analyses fully supported
this concept. Either mediated through a water molecule (10a,b)
or by a steric clash with the longer alkyl chains (10b-d), the
reorientation of the H11 residue L436 was proposed to affect
the dynamics of helix H12 and thereby to account for the mixed
agonist/antagonist activity of the corresponding ligands. In
contrast, a direct interaction between the side chains of UVI3003
(10e) and 10f and H12 was considered to give rise to the
antagonistic profile of these compounds.26
Here we report on the chemical synthesis and receptor
activation profiles of a isomeric series of tetramethyltetrahy-
dronaphthalene cinnamic acids 11a-f in which the alkoxy
chain is attached to the C4′ position (Figure 1). A new
synthesis of the parent compound 9 is also described.
Moreover, we disclose the crystal structure of RXRR LBD
bound to 9 that helps explain the divergent activities of the
C3′ and C4′-substituted series of ligands.
Figure 1. RXR antagonists. The series of RXR modulators discussed
in the present report (10a-f and 11a-f), which are derived from the
parent agonist 9, are also depicted.
Results and Discussion
activation at the PPARγ-RXR heterodimer but does not
synergize with the RAR ligand (E)-4-(2-(5,5,8,8-tetramethyl-
5,6,7,8-tetrahydronaphthalen-2-yl)prop-1-enyl)benzoic acid (TT-
NPB) to enhance activation at the RAR-RXR heterodimer.18,19
The polyene structures of LG100754 (1) and LG101506 (2) is
replaced by more rigid skeletons in 4-(5,7,7,10,10-pentamethyl-
2-nitro-7,8,9,10-tetrahydro-5H-benzo[b]naphtho[2,3-e][1,4]dia-
zepin-12-yl)benzoic acid (HX531) and 4-(7,7,10,10-tetramethyl-
5-propyl-7,8,9,10-tetrahydro-5H-benzo[b]naphtho[2,3-
e][1,4]diazepin-12-yl)benzoic acid (HX603), which are inhibitors
of RXR heterodimers, but also inhibit the activation of RARs
by agonists.20 Two other diazepinyl benzoic acids related to
HX531 (3) and HX603 (4) are sulfonamide (5)21 and cyano
derivative (6), which showed improved oral bioavailability and
potency.22 2-[(3-Pentyloxy-5,5,8,8-tetramethyl-5,6,7,8-tetra-
hydronaphthalen-2-yl)(methylamino)]pyrimidine-5-carboxylic acid
(PA451) and 2-[(3-hexyloxy-5,5,8,8-tetramethyl-5,6,7,8-tetrahy-
dronaphthalen-2-yl)(methylamino)]pyrimidine-5-carboxylic acid
(PA452) are specific inhibitors of retinoid synergism in
RAR-RXR heterodimers.23 A rationale for the antagonism of
compounds 3, 4, PA451 (7a), and PA452 (7b) based on the
inhibition of folding of H12 has been put forward.16 In contrast,
cyclopropylidene derivative (8) was shown to block RARꢀ
expression induced by the PPARγ-RXRR heterodimer via
inhibition of 9-cis-retinoic acid induced recruitment of coacti-
vator SRC-1 to RXRR.24
Chemistry. Synthesis of the biaryl30 bond of agonist 9 was
based on the Suzuki-Miyaura cross-coupling, known to be
tolerant of steric hindrance in the aryl components.31 Treatment
of known32 boronic acid 12 and 3-bromo-4-hydroxybenzalde-
hyde 13 yielded aldehyde 14 in low yield (25%). The alternative
route involving phenol protection (Ac2O, Py, DMAP) followed
by olefin formation (NaH, triethylphosphonoacetate, DME, -30
°C) of acetate 15 was more efficient and provided compound
16 in high yield (95%). Heating boronic acid 12 and bromide
16 to 150 °C in the presence of Pd(OAc)2, aqueous Na2CO3,
and the phase transfer agent TBAB33 resulted in a mixture of
phenol 17 and acetate 18 in 97% yield (based on recovered
16). These compounds were separated and characterized at this
stage, but for synthetic convenience, the mixture was carried
on to provide 9 by saponification (Scheme 1).
The synthesis of ethers 11a-f started with the etherification
of known27 bromotetrahydronaphthol 19 under classical condi-
tions (NaH, alkyliodide, DMF, 25 °C). Bromine-lithium
exchange of 20 (n-BuLi, -78 °C),34 followed by trapping each
organolithium with triisopropylborate, furnished boronic acids
21.27 These were characterized and stored as the crystalline
diethanolamine boronates 22,35 and the boronic acids were
released from derivatives 22 by treatment with HCl in THF prior
to use. The coupling of 21 to cinnamic ester 16 required
refluxing in DME with catalytic quantities of Pd(PPh3)4 and
excess Na2CO3 and provided the entire skeleton of the rexinoids
as free phenols 23. Saponification of the acrylic esters gave the
RXR-selective antagonists are useful tools for the elucidation
of the complex gene networks governed by the nuclear hormone

3152 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10
Sant´ın et al.
Scheme 1a
10e (Figure 1), which was previously characterized as a full-
RXR antagonist,26 destabilized holo-H12 significantly more than
11e, suggesting that the latter is a less potent antagonist (or a
more potent partial agonist) than the corresponding isomeric
ligand 10e (Figure 2A). To unambiguously differentiate partial
agonists from full antagonists, we previously demonstrated that
monitoring H12 dynamics in the presence of both ligands and
a coactivator fragment is required.26 We added increasing
concentrations of a 13-residue peptide corresponding to the
nuclear receptor box 2 region of the transcriptional intermediary
factor 2 (TIF-2 NR2) and measured the resulting anisotropy
for the various RXR/ligand complexes. In the presence of the
agonist 9, the addition of TIF-2 NR2 rapidly increased anisot-
ropy and helix H12 appears fully stabilized at a peptide
concentration of 1 µM (Figure 2B). By contrast, with ligands
11a-f, anisotropy values increased gradually upon TIF-2 NR2
addition and helix H12 remained more dynamic, even at the
highest peptide concentration used. Comparison of these data
with those obtained with the isomeric series 10a-f (Figure 1)
revealed that the trends in compound behavior across the two
homologous series were clearly different. Figure 2C shows that
ligands 10a-f generate large variations of anisotropy in response
to the addition of TIF-2 NR2. The partial agonists 10a-d induce
graded receptor dynamics as indicated by the peptide concentra-
tion required for full stabilization of H12, which correlates
inversely with the length of the aliphatic side chain. The
antagonists 10e and 10f display a very different profile as even
the highest doses of peptide fail to stabilize H12 completely.
Conversely, ligands 11a-f induced a more restricted range of
receptor dynamics and higher peptide concentrations are required
to reach the stabilization level obtained with the corresponding
compounds in the 10a-f series. Finally, neither 11e nor 11f
displayed the dynamic profile observed for the full-antagonists
10e-f. Together, these fluorescence anisotropy data reveal that
compounds 11a-f exhibit the dynamics signature of weak
partial agonists whose degree of agonism slightly varies
according to the side-chain length.
a Reagents and reaction conditions: (a) Pd(PPh3)4, aq Na2CO3, 60 °C,
25%; (b) Ac2O, Py, DMAP, 4 h, 25 °C, 99%; (c) NaH, (EtO)2POCH2CO2Et,
DME, -30 °C, 1.5 h, 95%; (d) boronic acid 12, Pd(OAc)2, aq Na2CO3,
TBAB, 150 °C, 97% (brsm); (e) LiOH, dioxane/H2O, 60 °C, 85%.
desired compounds 11a-f in high yields (Scheme 2). The
synthesis of series 10a-f has been described following a similar
sequence.26
Fluorescence Anisotropy Studies. Using fluorescence ani-
sotropy measurements of a fluorescein moiety that had been
attached selectively to the C-terminus of RXR helix H12, we
previously reported experimental evidence for a correlation
between the pharmacological activity of modulators 10a-f
(Figure 1) and their impact on the structural dynamics of the
activation helix H12.26 Using the same approach, we observed
that the novel series of compounds 11a-f increases the mobility
of helix H12, as revealed by the decreased anisotropy in the
presence of these compounds relative to that seen for the full-
agonist 9 (Figure 2A).
Functional Studies. Fluorescence anisotropy studies revealed
that all compounds of the new series do bind to RXR and
impinge on H12 dynamics. To assess the impact of this binding
on RXR-mediated transcription activation in intact cells, we used
These data indicate that compounds 11a-f fail to efficiently
stabilize the active receptor conformation and suggest that they
may act as partial agonists or antagonists. However, compound
Scheme 2a
a Reagents and reaction conditions: (a) NaH, RI, DMF, 25 °C (20a, 92%; 20b, 92%; 20c, 93%; 20d, 97%; 20e, 96%; 20f, 98%); (b) n-BuLi, TMEDA,
B(OiPr)3, THF, 1 h (21a, 69%; 21b, 84%; 21c, 79%; 21d, 78%; 21e, 71%; 21f, 81%); (c) diethanolamine, THF, 25 °C, 1 h, quant; (d) Pd(PPh3)4, 3M
Na2CO3, DME, reflux (23a, 57%; 23b, 71%; 23c, 72%; 23d, 68%; 23e, 63%; 23f, 64%); (e) 2 M KOH, MeOH, reflux (11a, 99%; 11b,88%; 11c, 100%; 11d,
100%; 11e, 93%; 11f, 79%).

Modulating RXR
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3153
Figure 2. Ligand-induced RXRR helix H12 dynamics monitored by fluorescence anisotropy. (A) Using an RARR-RXRR LBD heterodimer in
which a fluorescent dye is specifically attached to the C-terminus of RXRR, we measured anisotropy values in the presence of saturating concentrations
of the series of mixed agonists/antagonists 11a-f. For comparison, we also measured the effect of both the full-RXR agonist 9 and the previously
reported full-antagonist 10e. (B) Similar experiments were carried out in the presence of increasing concentrations of the NR interaction motif 2
peptide of the coactivator TIF-2 (LxxLL). (C) Same experiment as in (B) but using the previously described isomeric series of compounds 10a-f.
Figure 3. RXR agonist/antagonist potential of new compounds. HeLa cells stably transfected with the reporter recombinant 5xGal4-ꢀGlo-Luc and
Gal4-hRXRꢀ were incubated with increasing concentrations of compounds to assess their RXR agonist potential (A) or with 3 nM of 9 and increasing
concentrations of the compounds to assess their RXR antagonist potential (B).
reporter assays with genetically engineered HeLa cell lines.
These “reporter cells” express a chimeric receptor comprising
the ligand binding domain of RXRꢀ fused to the DNA binding
domain of the yeast GAL4 transcription factor. In addition, they
contain a stably integrated chimeric promoter luciferase reporter,
composed of five GAL4 response elements in front of a minimal
ꢀ-globin promoter.
By establishing dose-response curves with this cellular model
the relative potencies of the newly synthesized compounds for
activating transcription of a cognate target gene were compared
to that of 9 and 10e; note that the latter have been characterized
previously as pure RXR agonist and antagonist, respectively,
with a high RXR binding affinity.26 In all cases, the new
compounds acted as weak agonists compared to 9 (Figure 3A).

3154 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10
Sant´ın et al.
Table 1. Data Collection and Refinement Statistics
RXRR/CD3254/TIF-2
Data Collection
P43212
Hence addition of the alkyl ether in the C4′ position provokes
a major loss of transactivation efficacy of the corresponding
holo-RXR. 11a, which harbors the smallest substitution (Figure
1), is the most active of the series, displaying about 30% of the
9-induced transactivation. Further extension of the side chain
reduced the transactivation capacity of the rexinoid such that
11c-f induced only about 10% of the reporter activity seen
with 9. However these C4′-substituted compounds acted always
as partial agonists as they induced significant RXR activity,
while the corresponding C3′-substituted 10e was always inactive.
space group
cell dimensions
a, b, c (Å)
resolution (Å)
Rsym
65.76, 65.76, 110.86
31.53-2.00 (2.11-2.00)a
0.077 (0.377)
I/Iσ
8.6 (2.0)
99.9 (100.0)
6.0 (6.2)
completeness (%)
redundancy
Competition curves derived from challenging 3 nM 9 with
increasing ligand concentrations confirmed these results (Figure
3B). While 10e is a potent full antagonist for 9-induced
transcription through Gal4-RXR, the new compounds provided
very similar competition curves and showed that all these
molecules antagonize 9-induced transcription for RXR. Taken
together the observations that (i) the activation curves of
compounds 11a-f are shifted toward higher concentrations
(Figure 3A), (ii) 9-induced RXR activation is quantitatively
inactivated (Figure 3B) only at compound concentrations at or
above 10 µM, and (iii) 10e induced this level of antagonism
already at a 10-fold lower concentration, it is highly likely that
11a-f possess significantly lower affinity to RXR.
Refinement
resolution (Å)
no. reflections
Rwork/Rfree
no. atoms
protein
ligand
water
B factors
protein
ligand
water
rms deviations
bond lengths (Å)
bond angles (deg)
31.53-2.00
16134
0.205/0.241
1752
27
112
18.94
14.84
24.82
0.010
1.166
With respect to its abilities to act as partial agonists, i.e., to
exhibit both weak agonist and antagonist potential (relative to
the natural or a pure agonist), this series of C4′ substituted
derivatives yielded partial RXR agonists that are poorly
distinguishable from each other. Hence the alkoxy chains
decrease in RXR transcriptional potential and generate ligands
with a modest RXR affinity. While the previous analysis of C3′
alkyl ether-substituted rexinoids revealed a progressive transition
from agonist via mixed agonist/antagonist to full antagonist
which depended on the length of the aliphatic chain, no such
effect is observed for the present C4′ substitutions. Apparently,
the length of the alkyl ether attached to the C4′ position is not
critical and poorly discriminatory.
a Highest resolution shell.
Structural Studies. To provide structural evidence that can
help to understand the different functionalities of the two
isomeric series of RXR modulators, the RXRR LBD was
crystallized with the parent compound 9 and the 13-residue
peptide comprising the nuclear receptor-binding surface NR2
of the p160 coactivator family TIF-2 (also termed GRIP1 and
SRC2). The structure refined at 2.0 Å resolution (Table 1)
reveals the canonical ternary fold of NR LBDs in the active
conformation (Figure 4A) and an unambiguous experimental
2Fo - Fc electron density map for the agonist 9 (Figure 4B).
Figures 4A and 5A show that 9 occupies a similar location in
the ligand-binding pocket (LBP) and adopts a similar binding
mode as other RXR agonists.36-38 The methyl group (C3′
position), which was replaced by a linear alkoxy substituent in
the 10a-f series (Figure 1), points toward a small cavity formed
by residues C269, A272, L436, I447, and L451 (Figure 5A).
As exemplified by the structure of the RXR LBD/10c
complex,26 this cavity can accommodate medium-size aliphatic
side-chains (1-4 carbon atoms), provided that L436 (H11)
undergoes a significant conformational change (Figure 5B).
However, the repositioning of L436 has been shown to account
for the weak agonist activity of compounds 10a-d by generating
a steric clash with L455 from holo-H12.26 In the same line, the
structure of RXR LBD bound by 9 reaffirms the pivotal role of
L436 in stabilizing holo-H12, as the conformation of this residue
is identical to that observed in all structures of RXR complexed
with a full agonist. In the 11a-f series, the alkoxy chain is
attached to the C4′ position as compared to the C3′ position in
Figure 4. X-ray structure of RXRR LBD bound by 9. (A) Overall
structure of RXRR LBD in complex with 9. Helices are numbered from
N- to C-terminus. Together, helices H3, H4, and H12 define the
activation function 2 (AF-2) surface to which the TIF-2 peptide is
bound. (B) Experimental 2Fo - Fc electron density of ligand 9
contoured at 1σ.
the 10a-f isomers (Figure 1). Modeling studies reveal that the
C4′ substitution points toward a space in the LBP that is partially
occupied by residues V265, C269, L436, and F439 (Figure 5C).
To accommodate the aliphatic extension, several residue side
chains (F313, F346, H435, L436, F439) must therefore undergo
significant conformational changes. It is thus predicted that,
similarly to what was observed for the 10a-f series, the
displacement of the L436 side chain toward L455 (H12)
accounts for the weak agonist activity of the novel compounds
by lowering the interaction strength between holo-H12 and the
LBD surface. However, the difference in the directionality of
the alkoxy side chains and the more constrained environment
of the C4′ substitution most likely explain the lower binding
affinities and the weaker agonistic activities of compounds

Modulating RXR
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3155
conformation of L436 (H11) observed in the RXRR LBD/9
structure for full agonism. Perturbation of this conformation by
compounds 10a-f and 11a-f resulted in a decline of agonistic
activity. However, it appears that L436 repositioning alone is
not sufficient to confer full antagonist activity as compounds
of the 11a-f series retain some degree of agonist activity. Thus,
full antagonism is only observed for compounds 10e,f, whose
long hydrophobic extensions induce a steric blockade of holo-
H12 packing.26 Moving the alkoxy side chain from C3′ to C4′
increases the distance to H12 and most likely prevents com-
pounds 11e,f from interfering directly with this helix.
This study provides important additional SAR information
which contributes to the exploration of the chemical space of
this type of rexinoids and reveals the subtle chemical and
structural properties required to confer full or partial RXR
agonistic or antagonistic activities. Accumulation of such
structure-based knowledge will facilitate the design of RXR
modulators optimized for the prevention and treatment of cancer
or metabolic diseases.
Experimental Section
For general procedures, see Supporting Information. All com-
pounds were purified by flash chromatography and g95% purity
was established by combustion analysis.
Ethyl 3-(4-Acetoxy-3-bromophenyl)acrylate (16). A solution
of ethyl 2-phosphonoacetate (1.31 g, 7.24 mmol) in DME (13 mL)
was added to NaH (0.29 g, 7.24 mmol) at -30 °C, and the mixture
was stirred for 30 min. A solution of 4-acetoxy-3-bromobenzalde-
hyde 15 (1.60 g, 6.58 mmol) was slowly added, and stirring was
continued for 1.5 h at the same temperature. The mixture was
poured onto H2O and extracted with ether (3×). The organic extracts
were washed with an aqueous saturated NH4Cl solution (3×) and
with brine, dried over Na2SO4, filtered, and concentrated to dryness.
The residue was purified by chromatography (silicagel, 90:10
hexane/EtOAc) to afford 1.96 g (95% yield) of ester 16 as a white
1
solid (mp 71-72 °C, hexane/EtOAc). H NMR (CDCl3, 400.13
MHz) δ 1.33 (t, J ) 7.0 Hz, 3H, CH2CH3), 2.36 (s, 3H, CH3), 4.25
(q, J ) 7.0 Hz, 2H, CH2CH3), 6.39 (d, J ) 16.0 Hz, 1H, H2), 7.15
(d, J ) 8.4 Hz, 1H, H5′), 7.46 (dd, J ) 8.4, 2.0 Hz, 1H, H6′), 7.58
(d, J ) 16.0 Hz, 1H, H3), 7.76 (d, J ) 2.0 Hz, 1H, H2′). 13C NMR
(CDCl3, 100.62 MHz) δ 14.3 (q), 20.8 (q), 60.7 (t), 116.9 (s), 119.9
(d), 124.1 (d), 128.0 (d), 132.6 (d), 134.0 (s), 141.9 (d), 149.4 (s),
166.4 (s), 168.3 (s). IR (NaCl): υ 2975 (m, C-H), 1772 (s, CdO),
1710 (s, CdO), 1640 (m), 1488 (m), 1177 (s) cm-1. MS (EI+):
m/z (%) 314 ([M]+, 5), 312 ([M]+, 5), 272 (98), 270 (100), 244
(20), 242 (22), 227 (94), 225 (97), 200 (61), 198 (77), 147 (21),
146 (94), 145 (25), 118 (94), 117 (18), 90 (13), 89 (92). HRMS:
calcd for C13H1379BrO4 [M]+: 311.9997; found: 311.9984.
Figure 5. Structural basis of partial agonist action. (A) Close-up stereo
view showing the ligand-binding pocket of RXRR bound to 9. For
clarity, C269 and I447 are not displayed. (B) Superposition of the RXRR
ligand-binding pocket bound by 9 (yellow) and 10c (blue, PDB code
2p1v). (C) Superposition of the RXRR ligand-binding pocket bound
by 9 (yellow) and a docking model of 11a bound to RXRR.
Ethyl 3-[4-Hydroxy-3-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahy-
dronaphthalen-2-yl)-phenyl]acrylate (18). In a Schlenk flask,
Pd(OAc)2 (2 mg, 0.006 mmol) was added to a degassed solution
of bromide 16 (50 mg, 0.159 mmol), boronic acid 12 (40 mg, 0.159
mmol), TBAB (52 mg, 0.159 mmol), and Na2CO3 (51 mg, 0.479
mmol) in H2O (0.32 mL), and the resulting solution was stirred
for 15 min. The reaction mixture was heated to 150 °C for 5 min
to afford a dark solution. After cooling down to 25 °C, water was
added and the solution was extracted with EtOAc (3×). The
combined organic extracts were washed with an aqueous saturated
NaHCO3 solution, dried over Na2SO4, and evaporated to dryness.
The residue was purified by chromatography (silicagel, 85:15
hexane/EtOAc) to afford 31 mg of 18, 15 mg of acetate 17, and 15
mg of unreacted starting material (97% yield based on recovered
11a-f as compared to those of the equivalent compounds in
the C3′ series 10a-f (Figure 3A,B).
Conclusion
Chemical modifications at the C3′ or C4′ positions of 9 have
provided two series of analogues for which the receptor
activation profiles have been determined. C3′ derivatives26
display a wide panel of activities ranging from partial agonists
with a significant residual agonist activity (10a) to full antago-
nists (10e,f). Repositioning the alkyl ether chain to the C4′
position provided a much more restricted range of functional
profiles and led to a decrease in both the binding affinity and
the overall activity of compounds (11a-f). Indeed, all members
of this isomeric series behave as weak or very weak partial
agonists as judged from anisotropy and transient transactivation
experiments. The present study confirms the importance of the
1
starting material). Data for 18: H NMR (CDCl3, 400.13 MHz) δ
1.32 (s, 6H, 2 × CH3), 1.33 (t, J ) 7.4 Hz, 3H, CH2CH3), 1.33 (s,
6H, 2 × CH3), 1.71 (s, 4H, 2 × CH2), 2.11 (s, 3H, CH3), 4.23 (q,
J ) 7.4 Hz, 2H, CH2CH3), 5.12 (br s, 1H, OH), 6.28 (d, J ) 16.0
Hz, 1H, H2), 7.00 (d, J ) 8.5 Hz, 1H), 7.14 (s, 1H), 7.24 (s, 1H),
7.33 (d, J ) 2.2 Hz, 1H), 7.47 (dd, J ) 8.5, 2.2 Hz, 1H), 7.67 (d,

3156 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10
Sant´ın et al.
J ) 16.0 Hz, H3). 13C NMR (CDCl3, 100.62 MHz) δ 14.3 (q),
19.4 (q), 31.8 (q, 2×), 31.9 (q, 2×), 34.0 (s), 34.1 (s), 35.0 (t, 2×),
60.3 (t), 115.7 (d), 115.8 (d), 127.0 (s), 128.4 (d), 128.6 (s), 128.9
(d), 129.0 (d), 130.5 (d), 131.7 (s), 133.8 (s), 143.5 (s), 144.3 (d),
145.7 (s), 154.8 (s), 167.3 (s). IR (NaCl): υ 3500-3100 (br, O-H),
2960 (s, C-H), 2924 (s, C-H), 2859 (w), 1690 (s, CdO), 1631
(m), 1499 (m), 1271 (s), 1168 (s) cm-1. MS (EI+): m/z (%) 392
([M]+, 48), 378 (27), 377 (100). HRMS: calcd for C26H32O3 [M]+:
392.2351; found: 392.2365.
white powder. It was fully characterized as its diethylamino adduct
as indicated below.
2-(4-Methoxy-5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-
2-yl)-1,3,6,2-dioxazaborocane (22a). General Procedure for Forma-
tion of the Diethanolamine Boronates. A solution of diethanolamine
in THF (0.32 mL, 0.45 M, 0.14 mmol) was added dropwise to
boronic acid 21a (30 mg, 0.11 mmol) at 25 °C. After stirring for
1 h, a white precipitate formed, which was filtered off, washed
with hexane, and kept under vacuum for 2 h, to afford 22a (38
mg, 100% yield) as a white powder, mp 253-254 °C (hexane/
EtOAc). 1H NMR (CDCl3, 400.13 MHz) δ 1.26 (s, 6H, 2 × CH3),
1.33 (s, 6H, 2 × CH3), 1.60 (m, 4H, 2 × CH2), 2.5-2.6 (br s, 2H,
CH2), 3.0-3.2 (br s, 2H, CH2), 3,7-3.9 (br s, 3H, -O-CH3, 4H, 2
× CH2), 5.2-5.3 (br s, 1H, NH), 6.86 (s, 1H), 7.13 (s, 1H). 13C
NMR (CDCl3, 100.62 MHz) δ 28.7 (q, 2×), 32.2 (q, 2×), 34.1 (s),
34.5 (s), 35.4 (t), 38.1 (t), 51.2 (t), 55.1 (q), 63.2 (t), 112.9 (d),
123.4 (d, 2×), 132.1 (s), 145.9 (s), 158.2 (s). IR (NaCl): ν
3200-3000 (br, N-H), 2951 (s, C-H), 2924 (s, C-H), 2858 (s,
C-H), 2361 (m), 1458 (m), 1388 (m), 1273 (s), 1217 (s), 1102
(s), 1066 (s) cm-1. MS (EI+): m/z (%) 331 ([M]+, 34), 316 (43),
315 (12), 300 (28), 218 (21), 204 (16), 203 (100), 161 (15), 114
(63), 113 (16), 69 (20). HRMS: calcd for C19H30BNO3 [M]+:
331.2319; found: 331.2311.
3-[4-Hydroxy-3-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydronaph-
thalen-2-yl)-phenyl]acrylic Acid (9). A solution of LiOH ·H2O (82
mg, 2.0 mmol) in dioxane (2 mL) was added to ester 18 (77 mg,
0.2 mmol), and the mixture was heated to 60 °C for 3.5 h. After
cooling down to 25 °C, 10% HCl was added and the mixture was
extracted with EtOAc (3×). The combined organic layers were
washed with H2O and brine, dried over Na2SO4, filtered, and
evaporated to dryness. The residue was purified by chromatography
(silicagel, 60:40 hexane/EtOAc) to afford acid 9 (60 mg, 85% yield)
1
as a white solid (mp 275 °C, hexane/EtOAc). H NMR (CDCl3,
400.13 MHz) δ 1.16 (s, 6H, 2 × CH3), 1.23 (s, 6H, 2 × CH3),
1.61 (s, 4H, 2 × CH2), 2.01 (s, 3H, CH3), 6.21 (d, J ) 15.9 Hz,
1H), 6.92 (d, J ) 8.4 Hz, 1H), 7.04 (s, 1H), 7.15 (s, 1H), 7.25 (d,
J ) 2.2 Hz, 1H), 7.38 (dd, J ) 8.4, 2.2 Hz, 1H), 7.66 (d, J ) 15.9
Hz, 1H). 13C NMR (CDCl3, 100.62 MHz) δ 19.4 (q), 31.8 (q, 2×),
31.9 (q, 2×), 34.0 (s), 34.1 (s), 35.0 (t, 2×), 114.7 (d), 115.9 (d),
126.7 (s), 128.4 (d), 128.7 (s), 128.9 (d), 129.5 (d), 130.8 (d), 131.6
(s), 133.9 (s), 143.6 (s), 145.8 (s), 146.8 (d), 155.3 (s), 172.6 (s).
IR (NaCl): υ 3500-3100 (br, -OH), 2960 (s, C-H), 2926 (s,
C-H), 2862 (m, C-H), 1684 (s, CdO), 1627 (m), 1601 (m), 1495
(m), 1424 (m), 1271 (s), 1174 (m), 1128 (w), 758 (m) cm-1. MS
(FAB+): m/z (%) 364 ([M]+, 47), 350 (25), 349 (100). HRMS: calcd
for C24H28O3 [M]+: 364.2038; found: 364.2037.
(E)-Ethyl 3-[4-Hydroxy-3-(4-methoxy-5,5,8,8-tetramethyl-5,6,7,8-
tetrahydronaphthalen-2-yl)phenyl]acrylate (23a). General Proce-
dure for Suzuki Cross-Coupling. In a Schlenk flask, Pd(PPh3)4 (19
mg, 0.016 mmol) was added to a degassed solution of bromide 16
(0.17 mg, 0.54 mmol), boronic acid 21a (0.21 g, 0.81 mmol), and
Na2CO3 (0.99 mL, 3 M in H2O, 2.97 mmol) in DME (9 mL), and
the resulting mixture was stirred for 15 min at 25 °C and then heated
to reflux for 23 h. After cooling down to 25 °C, a 10% aqueous
HCl solution was added until pH 1, and the aqueous layer was
extracted with EtOAc (3×). The combined organic layers were
washed with an aqueous saturated NaHCO3 solution, dried over
Na2SO4, and evaporated. The residue was purified by flash
chromatography (silica gel, 85:15 hexane/EtOAc) to afford 23a
7-Bromo-5-methoxy-1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaph-
thalene (20a). General Procedure for the Williamson Ether
Synthesis. A solution of naphthol 19 (0.40 g, 1.41 mmol) in DMF
(1.8 mL) was added to NaH (85 mg, 60% in mineral oil, 2.12 mmol)
at 0 °C. After stirring for 30 min, a solution of iodomethane (0.13
mL, 2.12 mmol) in DMF (0.5 mL) was added. The reaction mixture
was allowed to warm up to 25 °C, and stirring was maintained for
2 h. The reaction mixture was poured over water and extracted
with ether (3×). The organic extracts were washed with brine, dried
over Na2SO4, filtered, and evaporated to dryness. The residue was
purified by flash chromatography (silica gel, 98:2 hexane/EtOAc)
to afford 20a (0.39 g, 92% yield) as a white powder, mp 92-94
°C (hexane/EtOAc). 1H NMR (CDCl3, 400.13 MHz) δ 1.26 (s, 6H,
2 × CH3), 1.34 (s, 6H, 2 × CH3), 1.62 (m, 4H, 2 × CH2), 3.79 (s,
3H, -O-CH3), 6.79 (s, 1H), 7.07 (s, 1H). 13C NMR (CDCl3, 100.62
MHz) δ 28.2 (q, 2×), 31.8 (q, 2×), 34.1 (s), 34.7 (t), 34.9 (s), 37.8
(t), 55.2 (t), 112.2 (d), 119.6 (s), 122.4 (d), 132.1 (s), 149.1 (s),
159.4 (s). IR (NaCl): ν 2956 (s, C-H), 2928 (s, C-H), 2862 (m,
C-H), 2361 (m), 1569 (s), 1453 (s), 1362 (s), 1268 (m), 1203 (s),
1056 (s) cm-1. MS (EI+): m/z (%) 298 ([M]+, 24), 296 ([M]+, 25),
284 (15), 283 (99), 282 (15), 281 (100), 202 (23), 187 (15), 173
(15), 160 (26). HRMS: calcd for C15H2179BrO [M]+: 296.0776;
found: 296.0768. Elemental anal. calcd (%) C 60.61, H 7.12, Br
26.88, O 5.38; found: C 60.64, H 7.31.
1
(0.12 g, 57% yield) as a white foam. H NMR (CDCl3, 400.13
MHz) δ 1.31 (s, 6H, CH3), 1.34 (t, J ) 7.0 Hz, 3H, CH3), 1.43 (s,
6H, 2 × CH3), 1.6-1.7 (m, 4H, 2 × CH2), 3.84 (s, 3H, OCH3),
4.26 (q, J ) 7.2 Hz, 2H), 5.87 (br s, 1H, OH), 6.34 (d, J ) 15.9
Hz, 1H, H2), 6.73 (d, J ) 1.7 Hz, 1H), 7.01 (d, J ) 9.2 Hz, 1H),
7.03 (d, J ) 1.7 Hz, 1H), 7.4-7.5 (m, 2H), 7.68 (d, J ) 15.9 Hz,
1H, H2). 13C NMR (CDCl3, 100.62 MHz) δ 14.3 (q), 28.3 (q, 2×),
31.9 (q, 2×), 34.3 (s), 34.7 (s), 34.9 (t), 37.8 (t), 55.2 (q), 60.3 (t),
109.1 (d), 115.8 (d), 116.2 (d), 119.6 (d), 122.1 (s), 127.2 (s), 128.9
(d), 129.1 (s), 130.0 (d), 133.4 (s), 133.7 (s), 144.3 (d), 148.7 (s),
154.6 (s), 159.6 (s), 167.4 (s). IR (NaCl): ν 3500-3100 (br, O-H),
2957 (s, C-H), 2929 (s, C-H), 2862 (m, C-H), 1684 (s), 1632
(s), 1599 (s), 1396 (m), 1283 (s), 1180 (s), 1048 (m) cm-1. MS
(EI+): m/z (%) 409 ([M + H]+, 28), 408 ([M]+, 100), 395 (11),
394 (66), 393 (89), 393 (75), 363 (11). HRMS: calcd for C28H36O4
[M]+: 408.2304; found: 408.2301.
(E)-3-[4-Hydroxy-3-(4-methoxy-5,5,8,8-tetramethyl-5,6,7,8-tet-
rahydronaphthalen-2-yl)phenyl]acrylic Acid (11a). General Pro-
cedure for Ester Hydrolysis. A 2 M solution of KOH in MeOH (8
mL) was added to ester 23a (100 mg, 0.24 mmol), and the mixture
was heated to reflux for 2 h. After cooling down to 25 °C, a 10%
aqueous HCl solution was added and the mixture was extracted
with CH2Cl2 (3×). The combined organic layers were washed with
H2O and brine, dried over Na2SO4, filtered, and evaporated to
dryness. The residue was purified by flash chromatography (silica
gel, 90:10 CH2Cl2/MeOH) to afford 11a (93 mg, 99% yield) as a
white powder, mp 215-217 °C (CH2Cl2/hexane). 1H NMR (CDCl3,
400.13 MHz) δ 1.29 (s, 6H, 2 × CH3), 1.41 (s, 6H, 2 × CH3),
1.6-1.7 (m, 4H, 2 × CH2), 3.83 (s, 3H, OCH3), 6.34 (d, J ) 16.0
Hz, 1H, H2), 6.69 (s, 1H), 6.9-7.1 (m, 2H), 7.4-7.5 (m, 2H), 7.75
(d, J ) 16.0 Hz, 1H, H3). 13C NMR (CDCl3, 100.62 MHz) δ 28.3
(q, 2×), 32.0 (q, 2×), 34.3 (s), 34.8 (s), 34.9 (t), 37.9 (t), 55.2 (t),
109.0 (d), 115.2 (d), 116.3 (d), 119.5 (d), 127.0 (s), 129.0 (s), 129.3
(d), 130.3 (d), 133.5 (s), 133.6 (s), 146.5 (d), 148.9 (s), 154.9 (s),
159.7 (s), 172.7 (s). IR (NaCl): ν 3500-2600 (br, O-H), 2956 (s,
4-Methoxy-5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-
ylboronic Acid (21a). General Procedure for Boronic Acid Forma-
tion. n-BuLi (1.19 mL, 1.22 M in hexane, 1.46 mmol) was slowly
added to a solution of bromide 20a (0.39 g, 1.32 mmol) and
TMEDA (0.44 mL, 5.29 mmol) in THF (4 mL) at -78 °C. After
stirring the mixture for 10 min at -78 °C, a solution of B(OiPr)3
(0.92 mL, 3.97 mmol) in THF (0.5 mL) was slowly added via
cannula, and stirring was maintained for 2 h at the same temperature.
Then 10% HCl (5 mL) was added, and the resulting mixture was
stirred for 2 h before addition of CH2Cl2. The aqueous layer was
extracted with CH2Cl2 (3×), and the combined organic extracts were
washed with brine, dried over Na2SO4, filtered, and evaporated to
dryness. The residue was purified by flash chromatography (silica
gel, 70:30 hexane/EtOAc) to afford 21a (0.24 g, 69% yield) as a

Modulating RXR
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3157
C-H), 2927 (s, C-H), 2860 (m), 1683 (s, CdO), 1628 (m), 1601
(m), 1501 (s), 1428 (m), 1360 (m), 1283 (s), 1184 (s) cm-1. MS
(EI+): m/z (%) 380 ([M]+, 33), 366 (25), 365 (100). HRMS: calcd.
for C24H28O4 [M]+: 380.1988; found: 380.1982. Elem. anal. calcd
for C26H32O4 ·H2O (%) C 72.34, H 7.59; found: C 72.18, H 7.17.
Steady-State Fluorescence Anisotropy. The RARR-RXRR LBD
fluorescent heterodimer has been prepared as previously de-
scribed.26 Fluorescence anisotropy assays were performed using
a Safire2 microplate reader (TECAN) at a protein concentration
of 0.120 µM. The excitation wavelength was set at 470 nm, with
emission measured at 530 nm. The TIF-2 NR2 coactivator
peptide (686-KHKILHRLLQDSS-698) was added to protein
samples containing 10 µM of ligand to a final concentration of
10 µM and then the sample was diluted successively with buffer
C supplemented with 0.120 µM of heterodimer and 10 µM of
ligand. At least three independent measurements were made for
each sample.
Cell Culture and Determination of RXR Activity. Gal4-hRXRꢀ
engineered HeLa cells (stably transfected with (Gal4)5-ꢀGlo-
Luc-Neo reporter and Gal4-hRXRꢀ plasmid) were maintained
in DMEM containing 5% FCS, supplemented with Geneticin
G418 (0.8 mg/mL), puromycin (0.3 µg/mL), hygromycin (0.2
mg/mL), and gentamycin (40 µg/mL). To determine the tran-
scriptional potential of the compounds, equal aliquots of cells
were seeded in a 96-well plate and were incubated at 37 °C and
5% CO2 and exposed to the ligands for 16 h (overnight). The
cells were washed (PBS) and lysed (50 µL of lysis buffer: 25
mM Tris phosphate (pH 7.8), 2 mM EDTA, 1 mM DTT, 10%
glycerol, and 1% Triton X-100) for 15 min. Equal aliquots (25
µL) of the cell lysates were transferred in Optiplate-96, and the
luminescence in RLU was determined on a MicroLumat LB96P
luminometer (Berthold) after automatic injection of 50 µL of
luciferin buffer (20 mM Tris phosphate (pH 7.8), 1.07 mM
MgCl2, 2.67 mM MgSO4, 0.1 mM EDTA, 33.3 mM DTT, 0.53
mM ATP, 0.47 mM luciferin, and 0.27 mM CoA).
Crystal Structure of the RXRr/9/TIF-2 Complex. Protein
expression and purification have been described previously.26
Briefly, the histidine-tagged LBD of human RXRR (residues
223-462 in a pET15b vector) was purified with a Ni2+-affinity
column followed by a gel filtration step. Fractions containing
RXRR LBD were pooled, concentrated, and mixed with a 3-fold
molar excess of CD3254 and a 5-fold molar excess of the TIF-2
NR2 coactivator peptide (686-KHKILHRLLQDSS-698). Crystals
were obtained by vapor diffusion at 20 °C. The well buffer
contained 20% PEG 4000; 0.1 M Tris.HCl, pH 8.0; 1.0 M
ammonium acetate. Crystals were of space group P43212. A
single crystal was mounted from mother liquor onto a cryoloop
(Hampton research), soaked in the reservoir solution containing
an additional 20% glycerol, and quickly frozen in liquid nitrogen.
Diffraction data were collected using an ADSC Quantum Q210
detector at the ID14-EH1 beamline of ESRF (France) at 2.0 Å
resolution. Diffraction data were processed using MOSFLM39
and scaled with SCALA from the CCP4 program suite.40 The
structures was solved by using the previously reported structure
1MVC37 of which the ligand and the coactivator peptide were
omitted. Initial Fo - Fc difference maps had strong signals for
the ligand and the TIF-2 NR2 peptide, which could be fitted
accurately into the electron density. The structure was modeled
with COOT41 and refined with REFMAC540 using rigid body,
least-squares, and individual B-factor refinements. The final
model exhibit very good geometry with 95.9% of the residues
in the most favored regions of the Ramachandran plot and no
residue in the disallowed regions.
B.; SAF07-63880), the Institut National du Cancer (H.G.), the
Ligue contre le Cancer (“Equipe labellise´e La Ligue”) (H.G.),
and the French National Research Agency (ANR-07-PCVI-
0001-01 to W.B.).
Note Added after ASAP Publication. This paper was released
ASAP on May 1, 2009 with an error in Figure 1. The revised
version was published on May 6, 2009.
Supporting Information Available: General experimental
procedures, synthesis, and characterization of intermediate com-
pounds (14, 19, 20b-f, 21b-f, 22b-f, 23b-f, 11b-f) and
molecular modeling. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
(1) Laudet, V.; Gronemeyer, H. The Nuclear Receptor Facts Book;
Academic Press: San Diego, 2002.
(2) Germain, P.; Staels, B.; Dacquet, C.; Spedding, M.; Laudet, V.
Overview of Nomenclature of Nuclear Receptors. Pharmacol. ReV.
2006, 58, 685–704.
(3) de Urquiza, A. M.; Liu, S.; Sjo¨berg, M.; Zetterstro¨m, R. H.; Griffiths,
W.; Sjo¨vall, J.; Perlmann, T. Docosahexaenoic Acid, A Ligand for
the Retinoid X Receptor in Mouse Brain. Science 2000, 290, 2140–
2144.
(4) Calle´ja, C.; Messaddeq, N.; Chapellier, B.; Yang, H.; Krezel, W.; Li,
M.; Metzger, D.; Mascrez, B.; Ohta, K.; Kagechika, H.; Endo, Y.;
Mark, M.; Ghyselinck, N. B.; Chambon, P. Genetic and Pharmacologi-
cal Evidence that a Retinoic Acid Cannot be the RXR-activating
Ligand in Mouse Epidermis Keratinocytes. Genes DeV. 2006, 20,
1525–1538.
(5) Mangelsdorf, D. A.; Evans, R. M. The RXR Heterodimers and Orphan
Receptors. Cell 1995, 83, 841–850.
(6) Gronemeyer, H.; Gustafsson, J.-A.; Laudet, V. Principles for Modula-
tion of the Nuclear Receptor Superfamily. Nat. ReV. Drug DiscoVery
2004, 3, 950–964.
(7) Chen, J.-Y.; Clifford, J.; Zusi, C.; Starret, J.; Tortolani, D.; Ostrowski,
J.; Reczec, P. R.; Chambon, P.; Gronemeyer, H. Two Distinct Actions
of Retinoid-Receptor Ligands. Nature 1996, 382, 819–822.
(8) Germain, P.; Iyer, J.; Zechel, C.; Gronemeyer, H. Co-regulator
Recruitment and the Mechanism of Retinoic Acid Receptor Synergy.
Nature 2002, 415, 187–192.
(9) Nagy, L.; Schwabe, J. W. R. Mechanism of the Nuclear Receptor
Molecular Switch. Trends Biomed. Sci. 2004, 29, 317–324.
(10) Shulman, A. I.; Larson, C.; Mangelsdorf, D. J.; Ranganathan, R.
Structural Determinants of Allosteric Ligand Activation in RXR
Heterodimers. Cell 2004, 116, 417–429.
(11) Mukherjee, R.; Davies, P. J. A.; Crombie, D. L.; Bischoff, E. D.;
Cesario, R. M.; Jow, L.; Hamann, L. G.; Boehm, M. F.; Mondon,
C. E.; Nadzan, A. M.; Paterniti, J. R.; Heyman, R. A. Sensitization of
Diabetic and Obese Mice to Insulin by Retinoid X Receptor Agonists.
Nature 1997, 386, 407–410.
(12) Altucci, L.; Leibowitz, M. D.; Ogilvie, K. M.; de Lera, A. R.;
Gronemeyer, H. RAR and RXR Modulation in Cancer and Metabolic
Disease. Nat. ReV. Drug DiscoVery 2007, 6, 793–810.
(13) Liby, K. T.; Yore, M. M.; Sporn, M. B. Triterpenoids and Rexinoids
as Multifunctional Agents for the Prevention and Treatment of Cancer.
Nat. ReV. Cancer 2007, 7, 357–369.
(14) Dawson, M. I.; Zhang, X. -k. Discovery and Design of Retinoic Acid
Receptor and Retinoid X Receptor Class- and Subtype-Selective
Synthetic Analogs of All-trans-retinoic Acid and 9-cis-Retinoic Acid.
Curr. Med. Chem. 2002, 9, 623–637.
(15) Kagechika, H.; Shudo, K. Synthetic Retinoids: Recent Developments
Concerning Structure and Clinical Utility. J. Med. Chem. 2005, 48,
5875–5883.
(16) Hashimoto, Y.; Miyachi, H. Nuclear Receptor Antagonists Designed
Based on the Helix-folding Inhibition Hypothesis. Bioorg. Med. Chem.
2005, 13, 5080–5093.
(17) Lala, D. S.; Mukherjee, R.; Shulman, I. G.; Canan Koch, S. S.;
Dardashti, L. J.; Nadzan, A. M.; Croston, G. E.; Evans, R. M.;
Heyman, R. A. Activation of Specific RXR Heterodimers by an
Agonist of RXR Homodimers. Nature 1996, 383, 450–453.
(18) Gernert, D. L.; Ajamie, R.; Ardecky, R. A.; Bell, M. G.; Leibowitz,
M. D.; Mais, D. A.; Mapes, C. M.; Michellys, P. Y.; Rungta, D.;
Reifel-Miller, A.; Tyhonas, J. S.; Yumibe, N.; Grese, T. A. Design
and Synthesis of Fluorinated RXR Modulators. Bioorg. Med. Chem.
Lett. 2003, 13, 3191–3195.
Acknowledgment. We thank Claudine Gaudon and Audrey
Bindler for the RXR reporter cells and assistance in the analysis.
This work was supported by funds from the EC (QLK-2002-
02029 “Anticancer Retinoids” and LSHC-CT-2005-518417
“EPITRON” to A.R.de L. and H.G.), the Spanish MEC (SAF04-
07131, FEDER to A.R.de L.; Juan de la Cierva contract to F.R.-
(19) Leibowitz, M. D.; Ardecky, R. J.; Boehm, M. F.; Broderick, C. L.;
Carfagna, M. A.; Crombie, D. L.; D’Arrigo, J.; Etgen, G. J.; Faul,
M. M.; Grese, T. A.; Havel, H.; Hein, N. I.; Heyman, R. A.; Jolley,

3158 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10
Sant´ın et al.
D.; Klausing, K.; Liu, S.; Mais, D. E.; Mapes, C. M.; Marschke, K. B.;
Michellys, P. Y.; Montrose-Rafizadeh, C.; Ogilvie, K. M.; Pascual,
B.; Rungta, D.; Tyhonas, J. S.; Urcan, M. S.; Wardlow, M.; Yumibe,
N.; Reifel-Miller, A. Biological Characterization of a Heterodimer-
selective Retinoid X Receptor Modulator: Potential Benefits for the
Treatment of Type 2 Diabetes. Endocrinology 2006, 147, 1044–1053.
(20) Vivat, V.; Zechel, C.; Wurtz, J.-M.; Bourguet, W.; Kagechika, H.;
Umemiya, H.; Shudo, K.; Moras, D.; Gronemeyer, H.; Chambon, P.
A Mutation Mimicking Ligand-induced Conformational Change Yields
a Constitutive RXR that Senses Allosteric Effects in Heterodimers.
EMBO J. 1997, 16, 5697–5709.
(21) Sakaki, J.; Konishi, K.; Kishida, M.; Gunji, H.; Kanazawa, T.;
Uchiyama, H.; Fukaya, H.; Mitani, H.; Kimura, M. Synthesis and
Structure-Activity Relationship of RXR Antagonists Based on the
Diazepinylbenzoic Acid Structure. Bioorg. Med. Chem. Lett. 2007,
17, 4808–4811.
(22) Sakaki, J.; Kishida, M.; Konishi, K.; Gunji, H.; Toyao, A.; Matsumoto,
Y.; Kanazawa, T.; Uchiyama, H.; Fukaya, H.; Mitani, H.; Arai, Y.;
Kimura, M. Synthesis and Structure-Activity Relationship of Novel
RXR Antagonists: Orally Active Antidiabetic and Antiobesity Agents.
Bioorg. Med. Chem. Lett. 2007, 17, 4804–4807.
(23) Takahashi, B.; Ohta, K.; Kawachi, E.; Fukasawa, H.; Hashimoto, Y.;
Kagechika, H. Novel Retinoid X Receptor Antagonists: Specific
Inhibition of Retinoid Synergism in RXR-RAR Heterodimers. J. Med.
Chem. 2002, 45, 3327–3329.
(24) Cavasotto, C. N.; Liu, G.; James, S. Y.; Hobbs, P. D.; Peterson, V. J.;
Bhattacharya, A. A.; Kolluri, S. K.; Zhang, X.-k.; Leid, M.; Abagyan,
R.; Liddington, R. C.; Dawson, M. I. Determinants of Retinoid X Receptor
Transcriptional Antagonism. J. Med. Chem. 2004, 47, 4360–4372.
(25) Yamauchi, T.; Kamon, J.; Waki, H.; Terauchi, Y.; Kubota, N.; Hara,
K.; Mori, Y.; Ide, T.; Murakami, K.; Tsuboyama-Kasaoka, N.; Ezaki,
O.; Akanuma, Y.; Gavrilova, O.; Vinson, O.; Reitman, M. L.;
Kagechika, H.; Shudo, K.; Yoda, M.; Nakano, Y.; Tobe, K.; Nagai,
R.; Kimura, S.; Tomita, M.; Froguel, P.; Kadowaki, T. The Fat-derived
Hormone Adiponectin Reverses Insulin Resistance Associated with
Both Lipoatrophy and Obesity. Nat. Med. 2001, 7, 941–946.
(26) Nahoum, V.; Pe´rez, E.; Germain, P.; Rodr´ıguez-Barrios, F.; Manzo,
F.; Kammerer, S.; Lemaire, G.; Hirsch, O.; Royer, C. A.; Gronemeyer,
H.; de Lera, A. R.; Bourguet, W. Modulators of the Structural
Dynamics of RXR to Reveal Receptor Function. Proc. Natl. Acad.
Sci. U.S.A. 2007, 104, 17323–17328.
tions and Pharmacological Implications. Trends Pharmacol. Sci. 2000,
21, 381–388.
(29) de Lera, A. R.; Bourguet, W.; Altucci, L.; Gronemeyer, H. Design of
Selective Nuclear Receptor Modulators: RAR and RXR as a Case
Study. Nat. ReV. Drug DiscoVery 2007, 6, 811–820.
(30) Anderson, J. C.; Namli, H.; Roberts, C. A. Investigations into Ambient
Temperature Biaryl Coupling Reactions. Tetrahedron 1997, 53, 15123–
15134.
(31) Kotha, S.; Lahiri, K.; Kashinath, D. Recent Applications of the
Suzuki-Miyaura Cross-coupling Reaction in Organic Synthesis.
Tetrahedron 2002, 58, 9633–9695.
(32) Faul, M. M.; Ratz, A. M.; Sullivan, K. A.; Trankle, W. G.; Winneroski,
L. L. Synthesis of Novel Retinoid X Receptor-Selective Retinoids. J.
Org. Chem. 2001, 66, 5772–5782.
(33) Leadbeater, N. E.; Marco, M. Rapid and Amenable Suzuki Coupling
Reaction in Water Using Microwave and Conventional Heating. J.
Org. Chem. 2003, 68, 888–892.
(34) Tan, Y.-L.; White, A. J. P.; Widdowson, D. A.; Wilhelm, R.; Williams,
D. J. Dilithiation of Arenetricarbonylchromium(0) Complexes with
Enantioselective Quench: Application to Chiral Biaryl Synthesis.
J. Chem. Soc., Perkin Trans. 1 2001, 3269–3280.
(35) Tripathy, P. B.; Matteson, D. S. Asymmetric Synthesis of the Four
Stereoisomers of 4-Methyl-3-heptanol via Boronic Esters: Sequential
Double Stereodifferentiation Leads to very High Purity. Synthesis 1990,
200–206.
(36) Egea, P. F.; Mitschler, A.; Rochel, N.; Ruff, M.; Chambon, P.; Moras,
D. Crystal Structure of the Human RXRR Ligand-binding Domain
Bound to its Natural Ligand: 9-Cis-retinoic Acid. EMBO J. 2000, 19,
2592–2601.
(37) Egea, P. F.; Mitschler, A.; Moras, D. Molecular Recognition of Agonist
Ligands by RXRs. Mol. Endocrinol. 2002, 16, 987–997.
(38) Pogenberg, V.; Guichou, J.-F.; Vivat-Hannah, V.; Kammerer, S.; Pe´rez,
E.; Germain, P.; de Lera, A. R.; Gronemeyer, H.; Royer, C. A.;
Bourguet, W. Characterization of the Interaction Between RAR/RXR
Heterodimers and Transcriptional Coactivators Through Structural and
Fluorescence Anisotropy Studies. J. Biol. Chem. 2005, 280, 1625–
1633.
(39) Leslie, A. G. The Integration of Macromolecular Diffraction Data.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 2006, 62, 48–57.
(40) The CCP4 Suite: Programs for Protein Crystallography. Acta Crys-
tallogr., Sect. D: Biol. Crystallogr. 1994, 50, 760.
(41) Emsley, P.; Cowtan, K. Coot: Model-building Tools for Molecular
Graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004, 60,
2126–2132.
(27) Bernardon, J.-M. Preparation of Bicyclic Aromatic Compounds and
their Use in Cosmetic or Dermatological Compositions. Patent EP
947496 A1 19991006, 1999.
(28) Bourguet, W.; Germain, P.; Gronemeyer, H. Nuclear Receptor Ligand-
Binding Domains: Three-Dimensional Structures, Molecular interac-
JM900096Q