Synthesis of 4,4′-Diaminotriphenylmethanes with Potential Selective Estrogen Receptor Modulator (SERM)-like Activity

Article abstract of DOI:10.1002/cmdc.201500148

In this study, a series of new 4,4′-diaminotriphenylmethanes was efficiently synthesized from aromatic aldehydes and 2,5-dimethoxybenzenamine under microwave irradiation in the presence of Sc(OTf)₃ as a catalyst. Antiproliferative activity was

Full text of DOI:10.1002/cmdc.201500148

DOI: 10.1002/cmdc.201500148
Full Papers
Synthesis of 4,4’-Diaminotriphenylmethanes with Potential
Selective Estrogen Receptor Modulator (SERM)-like Activity
Gema Guedes,[a] ngel Amesty,[a] Roberto JimØnez-Monzón,[b] Jorge Marrero-Alonso,[c]
Mario Díaz,[c] Leandro Fernµndez-PØrez,*[b] and Ana EstØvez-Braun*[a]
Dedicated to Prof. ngel G. Ravelo on the occasion of his recent retirement.
In this study, a series of new 4,4’-diaminotriphenylmethanes
was efficiently synthesized from aromatic aldehydes and 2,5-di-
methoxybenzenamine under microwave irradiation in the pres-
ence of Sc(OTf)3 as a catalyst. Antiproliferative activity was as-
sessed by using the MCF-7 estrogen receptor (ER)-positive
breast cancer cell line, and antagonist/agonist transcriptional
activities were determined. Docking studies and competition
studies of triphenylmethanes and radiolabeled estradiol deter-
mined that these compounds do not bind the ER, indicating
that triphenylmethane-induced changes in proliferative and
transcriptional activities differ from conventional mechanisms
of action triggered by other selective ER modulators.
Introduction
Triarylmethanes are an important group of organic com-
pounds, useful as precursors for the preparation of printing
inks and in the dyeing of ceramics, leather, and polyacrylonit-
rile fibers.[1–6] They have been widely used in analytical fields,
for example, detection of hydrogen peroxide in medical diag-
nostic kits, biotechnology process control, analyses of biologi-
cal fluids, and wastewater treatment.[7] The triarylmethine
moiety can also be found in bioactive natural products such as
cassigarol B,[8] mohsenone, and chamaechromone.[9–11] Further-
more, these compounds are attractive targets because of their
applications in medicinal chemistry as promising antitu-
mor,[12–15] antibacterial,[16] and anti-HIV agents.[17,18]
tral nervous system, and the cardiovascular system. It has been
identified that ERs are responsible of a variety of estrogen-re-
lated diseases, including osteoporosis, breast cancer, prostate
cancer, and inflammation.[23] In addition to the main clinical ap-
plication of breast cancer therapy, ER modulators are also used
in hormone replacement therapy (HRT) to treat postmeno-
pausal diseases. However, clinical studies suggest that long-
term usage of these steroid mimics could increase patient risks
of developing breast and uterine cancer.[24]
Selective estrogen receptor modulators (SERMs) refer to
a class of compound showing tissue-specific estrogenic activity,
thereby granting the possibility to selectively exert agonistic or
antagonistic estrogen-like action in various tissues.[25] These
tissue-specific estrogenic actions could be beneficial for clinical
application. Raloxifene, tamoxifen, and ormeloxifene are exam-
ples of marketed ER modulators.[25] Although these marketed
SERMs have achieved remarkable success in clinical therapies,
insufficient subtype selectivity causing adverse effects and
drug resistance has appeared in the last decades. Therefore,
discovery of new SERMs is of great importance for the clinical
treatment of breast cancer.
On the other hand, estrogen receptors (ERs) regulate mam-
malian hormonal and physiological processes through binding
with their common endogenous steroid hormone, 17b-estra-
diol (E2). This natural estrogen controls a number of physiolog-
ical processes within, but not limited to, the reproductive
organs in both females and males.[19–21] There are two known
subtypes of ERs, namely ERa and ERb, each with distinct tissue
expression.[22] ERa is mainly distributed in breast, uterus, and
bone, while ERb is widely expressed in prostate, lungs, the cen-
Taking into account the aforementioned evidence, together
with the antecedent that triarylmethanes are endowed with
estrogenic activity,[14,15] we describe herein the preparation and
assessment of estrogenic activity of a set of new 4,4’-diamino-
triphenylmethanes. These compounds were obtained by the
reaction of aromatic aldehydes with 2,5-dimethoxybenzena-
mine under microwave (MW) irradiation in the presence of
Sc(OTf)3 as a catalyst. The results showed that some of these
novel compounds displayed mixed SERM-like activities, al-
though their mechanisms of action might differ from those re-
ported for other established SERMs[26] in that they do not di-
rectly interact with the ligand-binding domain of ER.
[a] Dr. G. Guedes, . Amesty, Dr. A. EstØvez-Braun
Instituto Universitario de Bio-Orgµnica Antonio Gonzµlez (CIBICAN)
Departamento de Química Orgµnica, Universidad de La Laguna
Avda. Astrofísico Fco. Sµnchez 2, 38206, La Laguna, Tenerife (Spain)
E-mail: aestebra@ull.es
[b] R. JimØnez-Monzón, Dr. L. Fernµndez-PØrez
Instituto Universitario de Investigaciones BiomØdicas y Sanitarias (IUIBS)
Departamento de Ciencias Clínicas, BIOPHARM
Universidad de Las Palmas de Gran Canaria
Las Palmas de Gran Canaria, 35001 (Spain)
[c] J. Marrero-Alonso, Dr. M. Díaz
Departamento de Biología Animal
Universidad de La Laguna, 38206 Tenerife (Spain)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201500148.
ChemMedChem 2015, 10, 1403 – 1412
1403
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Results and Discussion
Table 1. Reaction conditions for the formation of 4,4’-diaminetriphenylmethanes.
Chemistry
One of the simplest and most
straightforward approaches for
the synthesis of 4,4’-diamino-
triarylmethanes is the Baeyer
condensation, which involves
the direct reaction of arylalde-
hydes with N,N-dimethylaniline.
A variety of reagents, such as p-
toluenesulfonic acid,[27] aniline
hydrochloride,[28] polymer-sup-
ported sulfonic acid,[29] mont-
morillonite K-10,[30] ZrOCl2,[31] n-
Entry
R
Cat. [mol%]
Conditions
Compd
Yield [%][a]
1
2
3
4
5
6
7
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
Br
100
100
100
20
10
10
10
1
RT, 180 min
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3b
3a
3a
3b
94
84
82
84
92
92
92
66
91
56
0
90
82
87
98
95
MW, 1008C, 30 min
MW, 1008C, 15 min
MW, 1008C, 30 min
MW, 1008C, 30 min
MW, 1008C, 15 min
MW, 1008C, 5 min
RT, 240 min
MW, 1008C, 5 min
MW, 1008C, 5 min
MW, 1008C, 30 min
MW,808C, 5 min
MW, 808C, 5 min
MW, 608C, 5 min
MW, 808C, 10 min
MW, 808C, 10 min
butylpyridinium
chloroalumi-
8[b]
9
nate,[32] NbCl5,[33] and atomized
sodium/THF under sonic condi-
tions[34] have been employed to
accomplish this transformation.
However, some of the reported
methods suffer from serious
drawbacks like longer reaction
times, drastic reaction condi-
tions, unsatisfactory yields, or
the use of toxic solvents. Thus,
there is a need for new versatile
5
1
10
11
12
13
14
15
16
–
5
5
5
5
NO2
NO2
Br
5
[a] Isolated yields. [b] Yb(OTf)3 was used following the same reaction conditions described previously.[36]
and simple protocols for the preparation of these compounds.
We recently published the preparation of complex disubsti-
tuted naphthoimidazoles from 1,4-dimethoxynaphthalen-2-
amine, imines obtained from this amine, and aromatic alde-
hydes in the presence of Sc(OTf)3.[35] When we carried out the
reaction of 4-nitrobenzaldehyde with 2,5-dimethoxybenzenea-
mine, we found that a triarylmethane was obtained via an
ABB’ domino reaction, instead of the expected benzoimidazole
(Scheme 1).
We also employed the reaction conditions used by Genovese
et al.[36] with methoxybenzene and aromatic aldehydes
(entry 8, neat, 1 mol% Yb(OTf)3), but only a 66% yield was ob-
tained.
Higher percentage loading of the catalyst lowered the reac-
tion time but did not increase the yield. Thus, we selected the
conditions of entry 15 in Table 1 in order to analyze the forma-
tion of 4,4’-diaminotriphenylmethanes using other aldehydes.
Various mono- and di-substituted aromatic aldehydes contain-
ing both electron-withdrawing and electron-donating
substituents were used, and the results are shown in
Table 2. Good yields (93–98%) were achieved with al-
dehydes having electron-withdrawing groups or with
benzaldehyde (entries 1–6). In the case of aldehydes
with electron-donating groups (entries 7–8) or the
heteroaromatic aldehyde pyridine-4-carbaldehyde
(entry 9), higher temperatures (120–1608C) and
a longer reaction time (30 min) were necessary to
obtain good yields. From a mechanistic point of view,
this may be due to the high oxophilicity of ScIII,
which coordinates the oxygen atom of the aldehyde,
thereby enhancing the electrophilicity of the carbonyl group
and facilitating attack of the p-electron-rich arene. The triaryl-
methanes were regioselectively obtained, with the new
carbon–carbon bond in a para position to the amino group as
determined by 1H NMR analysis.
Scheme 1. Formation of triarylmethanes.
Thus, we decided to study and to optimize this reaction,
which was carried out with anilines instead of N,N-dialkyl ani-
line and without an excess of this reagent, as is employed in
most of the mentioned procedures. Table 1 shows the results
obtained in the reaction of 2,5-dimethoxybenzeneamine (1)
with 4-nitro- or 4-bromobenzaldehyde under different condi-
tions. The best yield was obtained using 5 mol% of Yb(OTf)3
under MW irradiation at 808C for 10 min (entries 15 and 16).
Due to the antecedents of estrogenic activity of some triaryl-
methanes,[14,15] and because our triarylmethanes have hydro-
gen bond donors (ÀNH2) that resemble phenolic groups pres-
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1404
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
hibited E2-induced cell prolifera-
tion (Figure 1C). Taken together,
these results indicated agonist
activity of 3c, whereas 3b and
3e had antagonistic activity
toward MCF-7 cell growth. Nota-
bly, the anti-proliferative effects
of 3a, 3b, and 3e reached vehi-
cle values. These effects were
not attributed to compound-in-
duced cellular toxicity, because
we did not detect significant
changes in cell morphology,
membrane permeability (as as-
sessed by YOYO DNA staining),
or MTT cell metabolism, suggest-
ing that their anti-proliferative
effects are associated with an-
tagonism of ER-mediated MCF-7
cell growth.
Table 2. Scope of the reaction with aldehydes 2.
Entry
R
T [8C]
t [min]
Compd
Yield [%][a]
1
2
3
4
5
6
7
8
9
4-NO2-C6H4
4-Br-C6H4
4-F-C6H4
3-F-C6H4
4-Cl-C6H4
80
80
80
80
80
10
10
10
10
10
10
30
30
30
3a
3b
3c
3d
3e
3 f
3g
3h
3i
98
95
88
99
98
94
93
70
76
phenyl
80
4-OCH3-C6H4
3,4-methylenedioxyphenyl
3-pyridyl
120
120
160
[a] Isolated yields.
ent in many estrogenic compounds, such as 4-hydroxytamoxi-
fen, we decided to test them for their potential SERM-like ac-
tivity.
Figure 1D shows that 3 f, 3g, 3h, and 3i did not markedly
counteract the increase in E2-induced cell proliferation, sug-
gesting that these agents displayed antagonism toward ER-
mediated cell growth that could be reversed by E2. Interesting-
ly, 3 f (Figure 1D) enhanced E2-induced cell growth at concen-
trations as low as 0.01 mm, suggesting that very low concentra-
tions of triphenylmethane could cause cell growth, but higher
concentrations of this agent were inhibitory.
Biological activity
In vitro antiproliferative activity
MCF-7 cells were exposed at tri-
phenylmethane concentrations
ranging from 0.01 mm to 10 mm
in the absence or presence of E2
(0.1 nm), followed by dynamic
monitoring of changes in expo-
nentially growing cells for a fur-
ther 4 days. The triphenylme-
thane derivatives shown in Fig-
ure 1A (3a, 3b, 3d, and 3e) and
Figure 1B (3 f, 3g, 3h, and 3i)
led to a significant (25–75%) de-
crease in basal (vehicle)-treated
MCF-7 cell growth at concentra-
tions as low as 0.01 mm. In con-
trast, triphenylmethane deriva-
tive 3c (Figure 1A) was able to
increase cell proliferation above
basal cell growth in a dose-de-
pendent manner, reaching a max-
imal effect (Emax
)
similar to
0.1 nm E2. Figure 1C also shows
that 3a, 3b, and 3e counteract-
ed the increase in E2-induced
MCF-7 cell growth in a dose-de- Figure 1. Effects of triphenylmethanes on MCF-7 cell proliferation. MCF-7 cells were incubated for 4 days with in-
creased concentrations (0.01 mm–10 mm) of triphenylmethanes alone (A and B, estrogenic approach) or pretreated
pendent manner, with IC50
with triphenylmethanes for 2 h before addition of 0.1 nm E2 (C and D, anti-estrogenic approach). Data are pre-
values of 0.013Æ0.002, 2.8Æ0.1,
sented as the mean ÆSEM of cell proliferation change from vehicle-treated cells (basal level, A and B) or from E2-
and 0.7Æ0.09 mm, respectively.
induced maximal cell proliferation (Emax, C and D) of three assays for agonist or antagonist experiments, respec-
In contrast, neither 3c nor 3d in- tively. At least three replicated assays were used in all experiments; *p<0.05 vs. basal level or Emax
.
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1405
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Transcriptional activation assay
duction in the amount of total protein or altered MTT metabo-
lism in T47D-KBluc cells, but rather indicates that they inhibit-
ed ER-mediated transcriptional processes. In contrast, agents
such as 3 f, 3g, or 3h did not counteract the increase in E2-in-
duced transcription transcriptional activity (Figure 2D). Finally,
neither 4OH-TX nor synthesized triphenylmethanes displayed
cross-activation of luciferase expression under the control of
androgen or glucocorticoid receptors, as demonstrated in the
hormone-responsive MDA-kb2 cell line (data not shown).
From these results, we were able to observe how the type
of substituent on the non-amino aromatic ring modulates the
estrogenic/anti-estrogenic activity. Thus, compound 3 f (with-
out a substituent) was inactive, while compounds 3a (with
a nitro group) and 3i (with a pyridine ring) behaved as moder-
ate agonists. The presence of an electron-donating group led
to weak agonists 3g and 3h. The type and position of the hal-
ogen determined whether each
Agonistic and antagonistic ER-mediated transcriptional activity
was assayed using stably transfected T47D-kbluc cells, which
contain an estrogen response element coupled to a luciferase
reporter gene.[37] Treatment of T47D-kbluc cells with E2 led to
increased transcriptional activity in a dose-dependent manner
(EC50 =14Æ1.1 pm), and the agonist activity of E2 could be
abolished by co-incubation with the ER antagonist ICI-182780
(EC50 =0.0002Æ0.00008 mm)
or
4OH-TX
(EC50 =0.38Æ
0.009 mm). Instead of ICI, the agonistic approach showed that
4OH-TX caused a slight, yet significant, incremental increase in
activity (2–3 fold) above vehicle-treated T47D-KBluc cells.
Next, the synthesized triphenylmethanes were tested at dif-
ferent doses for the assessment of estrogenic and anti-estro-
genic responses. Figures 2A and 2B show that 3a, 3b, 3c, 3e,
compound exhibited antagonis-
tic or dual behavior. Compounds
3b and 3c (with halogens (Br
and F) in the para position) act
as antagonists, and compounds
3d (with a chlorine substituent
in the para position) and 3e
(with a fluorine in the meta posi-
tion) have dual profiles.
Docking studies
Given the effects of synthesized
triphenylmethanes on cellular
proliferation and ER-mediated
transcriptional activation, we
next used a docking computa-
tional approach to assess the
potential interaction of different
triphenylmethanes behaving as
agonists and/or antagonists with
the ligand binding domain of
ERa.
Figure 2. Transcriptional activities of triphenylmethanes on transfected T47D-KBluc cells with an estrogen re-
sponse element coupled to a luciferase reporter gene. Dose–response curves for agonism (A and B) and antago-
nism (C and D) of ER-mediated transcriptional activities were determined. Data are presented as the mean ÆSEM
of ER-mediated transcriptional activity change from vehicle-treated cells (basal level, A and B) or from E2-induced
maximal transcription (Emax, C and D) of three assays for agonist or antagonist experiments, respectively. At least
When we tried to carry out
flexible docking simulations
studies with the Glide soft-
ware,[38] we found that the 4,4’-
diaminotriphenylmethanes 3a–
3i were too large to fit into the
three replicated assays were used in all experiments; *p<0.05 vs. basal level or Emax
.
or 3i caused a moderate yet significant increase (20–40%) in
luciferase transcriptional activity above vehicle-treated T47D-
KBluc cells. To assess anti-estrogenic activity, compounds were
assayed against 0.1 nm E2, the lowest concentration that pro-
duced a maximal estrogenic response. Figures 2C and 2D
show that 3b, 3c, 3d, or 3i counteracted the increase in E2-in-
duced transcription, suggesting that these agents displayed
partial antagonism on ER-induced transcription. The decreasing
effects in luciferase expression were not attributed to com-
pound-induced cellular toxicity, because we did not detect re-
usual binding site of ERa complexed with genistein, as the size
of the binding cavity is estimated to be 490 Š3.[39] Thus, we
studied the binding mode of our compounds against both
conformations of the ERa and ERb ligand-binding domains
(LBD) using the following PDB codes: 1A52, 1ERR, 1X7J, 1X7R,
and 3ERT. We obtained docking results only in the cases of
1A52 for agonist conformation (ERa LBD co-crystallized with
estradiol (E2)) and 3ERT for antagonist conformation (ERa LBD
co-crystallized with 4-hydroxytamoxifen (4OHT)), but low XP
Gscores were achieved (i.e., 3ERT Gscores in kcalmolÀ1: 3a
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1406
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
À3.40, 3b À3.74, 3c À5.67, 3d À5.55, 3e À4.40, 3 f À4.92, 3g
À3.92, 3h À4.50, 3i À5.28, TX À10.28, 4OH-TX À12.58).
These data strongly indicate that the effects observed for tri-
phenylmethane on MCF-7 cell proliferation and ER-dependent
transcriptional activation are not due to direct binding of these
compounds to the ligand binding pocket of ER, which in part
agrees with the results from docking studies, but rather to
non-canonical modulation of ER signaling. Such unconvention-
al modulation of ER has been observed in different prepara-
tions and cell models and different ligand-independent path-
ways to modulate ERs have been described.[26,41,42] For in-
stance, activation of signaling leading to stimulation of kinases,
that is, mitogen-activated protein kinase or cyclin A/Cdk2 com-
plex, phosphorylate different do-
Competition studies on ER
Competition experiments were performed using rat uterus cy-
tosol, which upon ovariectomy upregulates ER expression.[40]
The extracts rich in ER were initially saturated with 5 nm of la-
beled E2 in the presence of increasing concentrations (1 pm–
10 mm) of unlabeled E2 or candidate competitors 3b and 3c.
As can be seen in Figure 3A, left panel, the presence of E2
mains in ER (AF-1, AF-2, or even
hinge domain) and have been
reported to activate ERs in the
absence of ligand.[41–43] Interest-
ingly, these growth factor acti-
vated pathways are thought to
significantly contribute to hor-
mone-independent growth in
some tumors.[44,45] Other possibil-
ities include regulation of coacti-
vators and corepressors that
exist in multifunctional protein
complexes associated to ER sig-
naling,[26] especially those involv-
ing domain interactions outside
LBD that modify the ER coactiva-
tors (such as the p160 family) in-
teraction or those that alter in-
teractions with the CoRNR
box.[26,44] Interestingly, disruption
of the ER–SRC protein–protein
interaction has been reported to
explain the anti-estrogenic ef-
fects of novel synthetic anti-es-
3
Figure 3. ER binding competition assays. A) Left: Uterine cytosolic extracts were saturated with 5 nm of H-labeled
estradiol in the presence of increasing concentrations of unlabeled competitors (1 pm–10 mm) for 18 h at 48C.
Right: Summary of the effects of 3b and 3c (10 mm) and TX (5 mm) in the presence of 5 nm of [3H]estradiol (con-
trol). B) Effects of compounds 3a, 3b, 3c, 3d, 3e, 3i (10 mm), TX (5 mm), and E2 (100 nm) on ER binding in the
presence of 0.5 nm of [3H]E2 (control). Data are presented as the mean ÆSEM for three different assays for each
compound and concentration; *p<0.01 statistically different from control value.
trogens ER5 and ER7, which do
not bind the LBD of ER.[45] The
potential involvement of any of
these mechanisms in the re-
sponse to triphenylmethane
compounds analyzed here re-
quires further investigation.
competed off the binding of [3H]E2 in a dose-dependent
manner, with an IC50 value of ~1.1 nm. Conversely, compounds
3b and 3c failed to compete the [3H]E2 binding to ER in the
Conclusions
whole range of concentrations, indicating that these triphenyl-
methane derivatives do not bind to ER (Figure 3A). As positive
control of competition, we used tamoxifen (5 mm), which
indeed was able to competitively displace the binding of
[3H]E2 from uterine ER (Figure 3A, right panel).
A set of new 4,4’-diaminotriphenylmethanes has been synthe-
sized through an efficient protocol using aromatic aldehydes
and 2,5-dimethoxybenzenamine under MW irradiation in the
presence of Sc(OTf)3 as a catalyst. These compounds were
tested for antiproliferative activity against ER-positive MCF-7
cell lines, and the T47D-Kb-luc transcriptional activation activity
was also determined. From the results obtained, the type of
substituent on the non-amino aromatic ring seems to modu-
late the estrogenic/anti-estrogenic activity. Docking studies in-
dicated that 4,4’-diaminotriphenylmethanes 3a–3i are too
large to fit into the usual binding cavity of ERa complexed
These results encouraged us to modify the assay conditions
by decreasing the amount of [3H]E2 down to 0.5 nm (half the
IC50 determined above), in order to facilitate the potential com-
petition by triphenylmethane derivatives. The results shown in
Figure 3B demonstrate that, with the exception of unlabeled
E2 or TX, neither compound was capable to compete off the
binding of radioactive estradiol to ER.
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1407
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
4-((4-Amino-2,5-dimethoxyphenyl)(4-nitrophenyl)methyl)-2,5-di-
methoxybenzene amine (3b): Following the general procedure
with genistein. In agreement with this, we observed that these
derivatives failed to displace the cognate ligand, 17b-estradiol,
from the LBD of ER, indicating that their effects on cell prolifer-
ation and ER-dependent transcriptional activity involve uncon-
ventional, yet unidentified, modulation of ER signaling. The
compounds used in this study might serve as starting points
for the development of novel SERMs.
described above,
a
solution of 2,5-dimethoxybenzenamine
(50.2 mg, 0.33 mmol), 4-bromobenzaldehyde (30.3 mg, 0.16 mol),
and Sc(OTf)3 (4.0 mg) in dry MeCN (2 mL) was irradiated for 10 min
at 808C. After elimination of the solvent, the residue was purified
by flash chromatography using hexane/EtOAc (2:3) to yield
73.8 mg (95%) of compound 3b as an amorphous white solid:
1
mp: 170–1718C; H NMR (400 MHz, CDCl3): d=3.59 (s, 6H, 2’-MeO,
2’’-MeO), 3.61 (s, 6H, 5’-MeO, 5’’-MeO), 3.71 (bs, 4H, 4’-NH2, 4’’-
NH2), 5.93 (s, 1H, 1-H), 6.31 (s, 2H, 6’-H, 6’’-H), 6.33 (s, 2H, 3’-H, 3’’-
H), 6.93 (d, J=8.2 Hz, 2H, 2’’’-H, 6’’’-H), 7.32 ppm (d, J=8.2 Hz, 2H,
3’’’-H, 5’’’-H); 13C NMR (100 MHz, CDCl3): d=42.3 (CH, C-1), 56.5 (2
CH3), 56.7 (2 CH3), 100.8 (2 CH, C-3’, C-3’’), 113.9 (2 CH, C-6’, C-6’’),
119.3 (C, C-4’’’), 122.1 (2 C, C-1’, C-1’’), 130.9 (4 CH, C-2’’’, C-3’’’, C-
5’’’, C-6’’’), 135.2 (2 C, C-4’, C-4’’), 141.2 (2 C, C-5’, C-5’’), 144.6 (C, C-
1’’’), 152.0 ppm (2 C, C-2’, C-2’’); HRESIMS 497.0880 (calcd for
Experimental Section
General methods: NMR spectra were recorded in CDCl3 or C6D6 at
400 MHz for 1H NMR and 100 or 150 MHz for 13C NMR. Chemical
shifts are given in (d) parts per million and coupling constants (J)
1
in hertz (Hz). H and 13C spectra were referenced using the solvent
signal as an internal standard. Microwave (MW) reactions were con-
ducted in sealed glass vessels (capacity 10 mL) using a Biotage mi-
crowave reactor. EIMS and HREIMS were recorded using an ion
trap mass spectrometer. Polygram-SIL G/UV254 analytical thin-layer
chromatography plates were used. Flash column chromatography
was carried out using Merck silica gel 60 (particle size less than
0.020 mm), using an appropriate mixtures of ethyl acetate and hex-
anes. All solvents and reagents were purified by standard tech-
niques reported[46] or used as supplied from commercial sources.
All compounds were named using the ACD40 Name-Pro program,
which is based on IUPAC rules. 2,5-Dimethoxybenzeneamine was
synthesized following the procedure described in the literature.[35]
17b-Estradiol (E2), testosterone (T), genestein (GEN), 4-hydroxy-ta-
moxifen (OHTX), dexamethasone (DEX), and testosterone were pur-
chased from Sigma–Aldrich Chemicals (St. Louis, MO, USA).
ICI 182,780 (ICI) was purchased from Tocris Bioscience (Bristol, UK).
C23H25N2O4Na81Br
[M]+
497.0875),
495.0884
(calcd
for
C23H25N2O4Na79Br [M]+ 495.0895); Anal. calcd for C23H25BrN2O4: C
58.36, H 5.32, N 5.92, found: C 57.94, H 5.22, N 5.87.
4-((4-Amino-2,5-dimethoxyphenyl)(4-fluorophenyl)methyl)-2,5-
dimethoxybenzene amine (3c): Following the general procedure
described above,
a
solution of 2,5-dimethoxybenzenamine
(50.2 mg, 0.33 mmol), 4-fluorobenzaldehyde (18.0 mL, 0.16 mol),
and Sc(OTf)3 (4.0 mg) in dry MeCN (2 mL) was irradiated for 10 min
at 808C. After elimination of the solvent, the residue was purified
by flash chromatography using hexane/EtOAc (1:1) to yield
59.7 mg (88%) of compound 3c as an amorphous green solid: mp:
1
133–1348C; H NMR (400 MHz, CDCl3): d=3.60 (s, 6H, 2’-MeO, 2’’-
MeO), 3.61 (s, 6H, 5’-MeO, 5’’-MeO), 3.61 (bs, 4H, 4’-NH2, 4’’-NH2),
5.96 (s, 1H, 1-H), 6.31 (s, 2H, 6’-H, 6’’-H), 6.34 (s, 2H, 3’-H, 3’’-H),
6.90 (t, J=8.5 Hz, 2H), 6.99–7.03 ppm (m, 2H); 13C NMR (100 MHz,
CDCl3): d=42.0 (CH, C-1), 56.5 (2 CH3), 56.7 (2 CH3), 100.9 (2 CH, C-
3’, C-3’’), 113.9 (2 CH, C-6’, C-6’’), 114.6 (2 CH, d, J2CÀF =20.9 Hz, C-
3’’’, C-5’’’), 122.7 (2 C, C-1’, C-1’’), 130.4 (2 CH, d, J3CÀF =7.6 Hz, C-
2’’’, C-6’’’), 135.1 (2 C, C-4’, C-4’’), 141.0 (C, C-1’’’), 141.2 (2 C, C-5’, C-
5’’), 152.0 (2 C, C-2’, C-2’’), 161.1 ppm (C, d, J1CÀF =241.4 Hz, C-4’’’);
IR (CH2Cl2): n˜ =3458, 3368, 3197, 2997, 2937, 2834, 1622, 1600,
1511, 1464, 1415, 1326, 1246, 1209, 1042, 854, 831, 735 cmÀ1; HRE-
SIMS: 435.1705 (calcd for C23H25N2O4NaF [M]+ 435.1696).
General procedure for the preparation of 4,4’-diaminotriphenyl-
methanes (3a–3i): A solution of aldehyde (30.0 mg, 0.1 mmol),
2,5-dimethoxybenzeneamine (2.0 equiv), and 5 mol% of Yb(OTf)3 in
CH3CN (2 mL) was placed in a microwave-special closed vial, and
the solution was irradiated in a single-mode microwave oven. The
reaction mixture was then cooled to room temperature. After re-
moving the solvent under reduced pressure, the product was puri-
fied by flash chromatography.
4-((4-Amino-2,5-dimethoxyphenyl)(4-nitrophenyl)methyl)-2,5-di-
4-((4-Amino-2,5-dimethoxyphenyl)(3-fluorophenyl)methyl)-2,5-
methoxybenzene amine (3a): Following the general procedure
dimethoxybenzen amine (3d): Following the general procedure
described above,
a
solution of 2,5-dimethoxybenzenamine
described above,
a
solution of 2,5-dimethoxybenzenamine
(50.2 mg, 0.33 mmol), 4-nitrobenzaldehyde (24.8 mg, 0.16 mol),
and Sc(OTf)3 (4.0 mg) in dry MeCN (2 mL) was irradiated for 10 min
at 808C. After elimination of the solvent, the residue was purified
by flash chromatography using hexane/EtOAc (1:1) to yield
70.6 mg (98%) of compound 3a as an amorphous red solid: mp:
(50.2 mg, 0.33 mmol), 3-flurobenzaldehyde (18.0 mL, 0.17 mol), and
Sc(OTf)3 (4.0 mg) in dry MeCN (2 mL) was irradiated for 10 min at
808C. After elimination of the solvent, the residue was purified by
flash chromatography using hexanes/EtOAc (1:1) to yield 66.8 mg
(99%) of compound 3d as an amorphous green solid: mp: 141–
1
1
180–1828C; H NMR (400 MHz, CDCl3): d=3.60 (s, 6H, 2’-MeO, 2’’-
1438C; H NMR (400 MHz, CDCl3): d=3.60 (s, 6H, 2’-MeO, 2’’-MeO),
MeO), 3.61 (s, 6H, 5’-MeO, 5’’-MeO), 3.78 (bs, 4H, 4’-NH2, 4’’-NH2),
6.02 (s, 1H, 1-H), 6.30 (s, 2H, 6’-H, 6’’-H), 6.35 (s, 2H, 3’-H, 3’’-H),
7.20 (d, J=8.5 Hz, 2H, 2’’’-H, 6’’’-H), 8.06 ppm (d, J=8.5 Hz, 2H, 3’’’-
H, 5’’’-H);13C NMR (100 MHz, CDCl3): d=43.3 (CH, C-1), 56.5 (4 CH3,
2’-MeO, 2’’-MeO, 5’-MeO, 5’’-MeO), 100.5 (2 CH, C-3’, C-3’’), 113.8 (2
CH, C-6’, C-6’’), 120.7 (2 C, C-1’, C-1’’), 123.2 (2 CH, C-3’’’, C-5’’’),
129.7 (2 CH, C-2’’’, C-6’’’), 135.7 (2 C, C-4’, C-4’’), 141.3 (2 C, C-5’, C-
5’’), 146.1 (C, C-4’’’), 152.0 (2 C, C-2’, C-2’’), 154.2 ppm (C, C-1’’’);
ESIMS m/z (%): 440 (26) [M+1]+, 439 (100) [M]+, 424 (12), 408 (19),
317 (26), 271 (14), 153 (36), 138 (71); IR (CH2Cl2): n˜ =3465, 3374,
3205, 3074, 3000, 2941, 2838, 1626, 1601, 1518, 1468, 1419, 1348,
1252, 1213, 1045, 835, 740, 710 cmÀ1; HRESIMS: 439.1746 (calcd for
C23H25N3O6 [M]+ 439.1740); Anal. calcd for C23H25N3O6: C 62.86, H
5.73, N 9.56, found: C 62.44, H 5.61, N 9.27.
3.61 (s, 6H, 5’-MeO, 5’’-MeO), 3.69 (bs, 4H, 4’-NH2, 4’’-NH2), 5.99 (s,
1H, 1-H), 6.33 (s, 2H, 6’-H, 6’’-H), 6.35 (s, 2H, 3’-H, 3’’-H), 6.75 (d,
J
oF =10.4 Hz, 1H, 2’’’-H), 6.80–8.84 (m, 1H, 4’’’-H), 6.85 (d, J=7.6 Hz,
1H, 6’’’-H), 7.16 ppm (dd, J=7.6 Hz, 1H, 5’’’-H); 13C NMR (100 MHz,
CDCl3): d=42.5 (CH, C-1), 56.5 (2 CH3), 56.7 (2 CH3), 100.9 (2 CH, C-
3’, C-3’’), 112.4 (CH, d, J2CÀF =21.1 Hz), 113.9 (2 CH, C-6’, C-6’’), 115.9
(CH, d, J2CÀF =21.3 Hz), 122.2 (2 C, C-1’, C-1’’), 124.9 (CH, C-6’’’),
129.1 (CH, d, J3CÀF =8.1 Hz, C-5’’’), 135.1 (2 C, C-4’, C-4’’), 141.3 (2 C,
C-5’, C-5’’), 148.4 (C, d, J3CÀF =6.6 Hz, C-1’’’), 152.0 (2 C, C-2’, C-2’’),
162.9 ppm (C, d, J1CÀF =242.6 Hz, C-3’’’); IR (CH2Cl2): n˜ =3460, 3370,
3201, 3052, 2997, 2937, 2869, 2834, 1620, 1590, 1515, 1464, 1416,
1326, 1251, 1209, 1042, 891, 829, 736 cmÀ1; HRESIMS: 435.1697
(calcd for C23H25N2O4FNa [M]+ 435.1696); Anal. calcd for
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1408
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
C23H25FN2O4: C 66.98, H 6.11, N 6.79, found: C 66.74, H 6.34,
N 6.23.
1299, 1246, 1209, 1177, 1041, 830, 735 cmÀ1; HRESIMS: 447.1893
(calcd for C24H28N2O5Na [M]+ 447.1896); Anal. calcd for C24H28N2O4:
C 67.91, H 6.65, N 6.60, found: C 67.56, H 6.43, N 6.38.
4-((4-Amino-2,5-dimethoxyphenyl)(4-chlorophenyl)methyl)-2,5-
dimethoxy benzenamine (3e): Following the general procedure
4-((4-Amino-2,5-dimethoxyphenyl)(3,4-methylenedioxyphenyl)-
methyl)-2,5-dimethoxy benzenamine (3h): Following the general
procedure described above, a solution of 2,5-dimethoxybenzena-
mine (50.2 mg, 0.33 mmol), piperonal (24.6 mg, 0.16 mol), and
Sc(OTf)3 (4.0 mg) in dry MeCN (2 mL) was irradiated for 30 min at
1208C. After elimination of the solvent, the residue was purified by
flash chromatography using mixtures of hexane/EtOAc (1:1) to
yield 50.4 mg (70%) of compound 3h as an amorphous brown
solid: mp: 166–1688C; 1H NMR (400 MHz, CDCl3): d=3.61 (s, 6H, 2’-
MeO, 2’’-MeO), 3.62 (s, 6H, 5’-MeO, 5’’-MeO), 3.72 (bs, 4H, 4’-NH2,
4’’-NH2), 5.89 (s, 2H, 7’’’-H), 5.92 (s, 1H, 1-H), 6.34 (s, 2H, 3’-H, 3’’-H
o 6’-H, 6’’-H), 6.35 (s, 2H, 3’-H, 3’’-H o 6’-H, 6’’-H), 6.51 (d, J=8.0 Hz,
1H, 5’’’-H), 6.58 (s, 1H, 2’’’-H), 6.67 ppm (d, J=8.0 Hz, 1H, 6’’’-H);
13C NMR (100 MHz, CDCl3): d=42.2 (CH, C-1), 56.6 (2 CH3), 56.9 (2
CH3), 100.8 (CH2), 101.0 (2 CH, C-3’, C-3’’), 107.7 (CH), 109.9 (CH),
114.0 (2 CH, C-6’, C-6’’), 122.0 (CH), 123.1 (2 C, C-1’,C-1’’), 135.0 (2 C,
C-4’,C-4’’), 139.4 (C), 141.3 (2 C, C-5’, C-5’’), 145.4 (C), 147.3 (C),
152.0 ppm (2 C, C-2’,C-2’’); IR (CH2Cl2): n˜ =3462, 3368, 3199, 2996,
2936, 2833, 1621, 1513, 1486, 1466, 1416, 1325, 1248, 1208, 1040,
described above,
a
solution of 2,5-dimethoxybenzenamine
(50.2 mg, 0.33 mmol), 4-chlorobenzaldehyde (23.8 mg, 0.16 mol),
and Sc(OTf)3 (4.0 mg) in dry MeCN (2 mL) was irradiated for 10 min
at 808C. After elimination of the solvent, the residue was purified
by flash chromatography using hexanes/EtOAc (3:2) to yield
1
69.0 mg (98%) of compound 3e as a yellow oil: H NMR (400 MHz,
CDCl3): d=3.53 (bs, 4H, 4’-NH2, 4’’-NH2), 3.59 (s, 6H, 2’-MeO, 2’’-
MeO), 3.61 (s, 6H, 5’-MeO, 5’’-MeO), 5.95 (s, 1H, 1-H), 6.31 (s, 2H,
6’-H, 6’’-H), 6.34 (s, 2H, 3’-H, 3’’-H), 6.99 (d, J=8.1 Hz, 2H, 2’’’-H, 6’’’-
H), 7.17 ppm (d, J=8.1 Hz, 2H, 3’’’-H, 5’’’-H); 13C NMR (100 MHz,
CDCl3): d=42.2 (CH, C-1), 56.5 (2 CH3), 56.7 (2 CH3), 100.8 (2 CH, C-
3’, C-3’’), 113.9 (2 CH, C-6’, C-6’’), 122.2 (2 C, C-1’, C-1’’), 127.9 (2
CH), 130.5 (2 CH), 131.1 (C, C-4’’’), 135.2 (2 C, C-4’, C-4’’), 141.2 (2 C,
C-5’, C-5’’), 144.1 (C, C-1’’’), 152.0 ppm (2 C, C-2’, C-2’’); IR (CH2Cl2):
n˜ =3456, 3368, 3199, 2996, 2936, 2833, 1622, 1514, 1463, 1415,
1325, 1246, 1209, 1042, 894, 847, 735 cmÀ1; HRESIMS: 451.1394
(calcd for C23H25N2O4Na35Cl [M]+ 451.1401); Anal. calcd for
C23H25ClN2O4: C 64.41, H 5.88, N 6.53, found: C 63.98, H 5.74, N
6.28.
930, 893, 825, 734 cmÀ1
;
HRESIMS: 461.1685 (calcd for
C24H26N2O6Na [M]+ 461.1689); Anal. calcd for C24H26N2O6: C 65.74, H
4-((4-Amino-2,5-dimethoxyphenyl)(phenyl)methyl)-2,5-dime-
thoxybenzenamine (3 f): Following the general procedure de-
scribed above, a solution of 2,5-dimethoxybenzenamine (50.2 mg,
0.33 mmol), benzaldehyde (17.0 mL, 0.16 mol), and Sc(OTf)3 (4.0 mg)
in dry MeCN (2 mL) was irradiated for 10 min at 808C. After elimi-
nation of the solvent, the residue was purified by flash chromatog-
raphy using mixtures of hexane/EtOAc from 4:1 to 2:3 to yield
60.8 mg (94%) of compound 3 f as an amorphous white solid: mp:
5.98, N 6.39, found: C 65.37, H 5.80, N 5.98.
4-((4-Amino-2,5-dimethoxyphenyl)(3-pyridyl)methyl)-2,5-dime-
thoxy benzenamine (3i): Following the general procedure de-
scribed above, a solution of 2,5-dimethoxybenzenamine (50.2 mg,
0.33 mmol), 3-pyridylcarboxaldehyde (16.0 mL, 0.17 mol), and
Sc(OTf)3 (4.0 mg) in dry MeCN (2 mL) was irradiated for 30 min at
1608C. After elimination of the solvent, the residue was purified by
preparative TLC using EtOAc to yield 49.4 mg (76%) of compound
1
148–1508C; H NMR (400 MHz, CDCl3): d=3.59 (s, 6H, 2’-MeO, 2’’-
1
MeO), 3.60 (s, 6H, 5’-MeO, 5’’-MeO), 3.71 (bs, 4H, 4’-NH2, 4’’-NH2),
6.00 (s, 1H, 1-H), 6.34 (s, 4H, 3’-H, 3’’-H, 6’-H, 6’’-H), 7.06 (d, J=
7.2 Hz, 2H, 2’’’-H, 6’’’-H), 7.13 (t, J=7.2 Hz, 1H, 4’’’-H), 7.21 ppm (t,
J=7.2 Hz, 2H, 3’’’-H, 5’’’-H); 13C NMR (100 MHz, CDCl3): d=42.7
(CH, C-1), 56.5 (2 CH3), 56.9 (2 CH3), 101.0 (2 CH, C-3’, C-3’’), 114.1 (2
CH, C-6’, C-6’’), 123.1 (2 C, C-1’, C-1’’), 125.5 (CH, C-4’’’), 127.9 (2
CH), 129.2 (2 CH), 134.9 (2 C, C-4’, C-4’’), 141.3 (2 C, C-5’, C-5’’),
145.3 (C, C-1’’’), 152.1 ppm (2 C, C-2’, C-2’’). IR (CH2Cl2): n˜ =3456,
3368, 3054, 2996, 2936, 2869, 2833, 1622, 1599, 1514, 1462, 1415,
1326, 1247, 1209, 1042, 839, 736, 703 cmÀ1; HRESIMS 417.1791
(calcd for C23H26N2O4Na [M]+ 417.1790); Anal. calcd for C23H26N2O4:
C 70.03, H 6.64, N 7.10, found: C 69.98, H 6.26, N 6.69.
3i as a green oil: H NMR (400 MHz, CDCl3): d=3.60 (s, 6H, 2’-MeO,
2’’-MeO), 3.61 (s, 6H, 5’-MeO, 5’’-MeO), 3.75 (bs, 4H, 4’-NH2, 4’’-
NH2), 5.97 (s, 1H, 1-H), 6.32 (s, 2H, 6’-H, 6’’-H), 6.34 (s, 2H, 3’-H, 3’’-
H), 7.14 (dd, J=4.8 Hz, 1H, 5’’’-H), 7.34 (d, J=7.7 Hz, 1H, 4’’’-H),
8.34 (s, 1H, 2’’’-H), 8.39 ppm (d, J=4.3 Hz, 1H, 6’’’-H); 13C NMR
(100 MHz, CDCl3): d=40.9 (CH, C-1), 56.5 (4 CH3, 2’-MeO, 2’’-MeO,
5’-MeO, 5’’-MeO), 100.6 (2 CH, C-3’, C-3’’), 113.9 (2 CH, C-6’, C-6’’),
121.1 (2 C, C-1’, C-1’’), 122.9 (CH), 135.5 (2 C, C-4’, C-4’’), 136.6 (CH),
141.0 (C, C-1’’’), 141.2 (2 C, C-5’, C-5’’), 146.6 (CH), 150.5 (CH),
151.9 ppm (2 C, C-2’, C-2’’); IR (CH2Cl2): n˜ =3451, 3363, 3195, 2996,
2936, 2833, 1622, 1514, 1463, 1416, 1326, 1248, 1209, 1042, 893,
844, 732 cmÀ1; HRESIMS: 418.1746 (calcd for C22H25N3O4Na [M]+
418.1743); Anal. calcd for C22H25N3O4: C 66.82, H 6.37, N 10.63,
found: C 66.42, H 6.74, N 10.28.
4-((4-Amino-2,5-dimethoxyphenyl)(4-methoxyphenyl)methyl)-
2,5-dimethoxybenzen amine (3g): Following the general proce-
dure described above, a solution of 2,5-dimethoxybenzenamine
(50.2 mg, 0.33 mmol), benzaldehyde (20.0 mL, 0.16 mol), and
Sc(OTf)3 (4.0 mg) in dry MeCN (2 mL) was irradiated for 30 min at
1208C. After elimination of the solvent, the residue was purified by
flash chromatography using a mixture of hexane/EtOAc (1:1) to
yield 64.5 mg (93%) of compound 3g as an amorphous brown
Biological methods
Mammalian cells: The human breast cancer cell lines T47D-kbluc
(ATCC) and MDA-kb2 (ATCC) were maintained in 75 cm2 culture
flasks (Nunclon) in RPMI (Lonza) growth media, without phenol
red, supplemented with 10% fetal bovine serum (FBS) (Lonza),
2 mm glutamine, 10 mm HEPES, and 1 mm sodium pyruvate. MCF-
7 cells (ATCC) were maintained in Dulbecco’s modified Eagle’s
medium (DMEM) supplemented with 10% FBS and 20 mmL-1 d-
glucose. The medium was supplemented with 100 UImLÀ1 penicil-
lin and 100 mgmLÀ1 streptomycin. Cells were grown in a humidified
incubator with 5% CO2 at 378C.
1
solid: mp: 141–1438C; H NMR (400 MHz, CDCl3): d=3.38 (bs, 4H,
4’-NH2, 4’’-NH2), 3.60 (s, 6H, 2’-MeO, 2’’-MeO), 3.61 (s, 6H, 5’-MeO,
5’’-MeO), 3.77 (s, 3H, 4’’’-MeO), 5.95 (s, 1H, 1-H), 6.34 (s, 4H, 3’-H,
3’’-H, 6’-H, 6’’-H), 6.76 (d, J=8.4 Hz, 2H, 2’’’-H, 6’’’-H), 6.97 ppm (d,
J=8.4 Hz, 2H, 3’’’-H, 5’’’-H); 13C NMR (100 MHz, CDCl3): d=41.8
(CH, C-1), 55.3 (CH3, 4’’’-MeO), 56.5 (2 CH3), 56.9 (2 CH3), 101.1 (2
CH, C-3’, C-3’’), 113.3 (2 CH), 114.0 (2 CH), 123.4 (2 C, C-1’, C-1’’),
130.0 (2 CH), 134.9 (2 C, C-4’,C-4’’), 137.3 (C, C-1’’’), 141.3 (2 C, C-5’,
C-5’’), 152.0 (2 C, C-2’,C-2’’), 157.5 ppm (C, C-4’’’); IR (CH2Cl2): n˜ =
3456, 3367, 3199, 2996, 2936, 2834, 1621, 1512, 1463, 1415, 1326,
Transcriptional activity studies: Compound stock solutions were pre-
pared in 100% DMSO in glass amber vials and Teflon-lined caps
and stored at À808C. Dosing solutions were prepared by diluting
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1409
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
compound stocks in fresh 5% dextran-coated charcoal-stripped
FBS (DCC-FBS) (Hyclone) RPMI to desired concentrations. DMSO ve-
hicle did not exceed 0.06%. Transcriptional activity was measured
using a luciferase-based reporter gene assay in two breast cancer
cell lines expressing different nuclear receptors: a) T47D-kbluc cells
(ATCC), which naturally express both ERa and ERb and are stably
transfected with a triplet estrogen-responsive element (ERE) pro-
moter–luciferase reporter plasmid (EREx3-Luc),[37] and b) MDA-kb2
cells (ATCC), which naturally express both glucocorticoid and an-
drogen receptors and are stably transfected with an MMTV.neo
promoter–luciferase reporter gene construct.[37] Cells were main-
tained in growth media, modified by replacement of 10% FBS with
10% DCC-FBS without antibiotic supplement four days prior to the
assay to remove all steroids from the culture. After incubation in
low-steroid media, T47D-kb cells or MDA-kb cells were seeded
onto 96-well plates (Nunclon) at 6”104 cells per well or 3”104
cells per well, respectively, in 5% DCC-FBS RPMI and were allowed
to attach overnight. Media was then replaced with 100 mL per well
of dosing media and the test compound and incubated for 24 h.
Then, cells were washed with cold phosphate-buffered saline, har-
vested in 80mL Passive Lysis Buffer (Promega) per well, and lucifer-
ase activity was determined with LAR (Promega) using a fluores-
cence Ascent FL multiplate reader (Thermo Fisher) and quantified
as relative light units (RLU). Each compound was assayed inde-
pendently at least three times with a minimum of three wells per
each replicate. T47D-kb cells were screened with compounds using
estrogen-positive (E2), -negative (vehicle only), antagonist (E2 plus
ICI), and background (vehicle plus ICI) controls on every plate. Each
compound was tested both alone and in the presence of an ap-
propriate competitor, such as 0.1 nm E2 or ICI. MDA-kb cells were
screened with compounds using agonist positive (testosterone or
dexamethasone) or negative (vehicle only) controls on every plate.
Agonist positive controls were monitored over time as an assess-
ment of the stability of the cell lines. In those instances where cy-
totoxicity of a compound was suspected, duplicate plates were
dosed in parallel.
Estrogen receptor competitive binding assay: Estrogen receptor was
obtained from the uterine cytosol fraction from mature ovariec-
tomized female Sprague-Dawley rats.[40] Aliquots of uterine cytosol
(100 mL) were incubated with 5 nm [3H]E2 and increasing concen-
trations of unlabeled competitors (0.1 nm–10 mm) for 18 h at 48C.
Then, 200 mL of dextran-coated charcoal suspension (0.8% char-
coal:0.08% dextran; DCC) in TRIS-EDTA-glycerol-Mg buffer was
added to each tube and incubated for 10 min. The DCC was then
removed by centrifugation at 3000 g for 10 min. The supernatant
was measured for radioactivity in a 4 mL scintillation cocktail Opti-
phase Hisafe 2 (PerkinElmer) by LKB Wallac 1214-Rackbeta counter
(LKB Instrument). Corrections were made for non-specific binding.
The relative binding affinity (RBA) of agents was calculated as the
ratio of agents and E2 or tamoxifen values as derived from dose–
response curves.
Cytotoxicity assays: Cytotoxicity was evaluated by determining the
mitochondrial function of the cells using the tetrazolium dye 3-
[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT)
following treatment with each compound. MTT is a yellow vital
dye that is actively converted by mitochondrial oxidation–reduc-
tion reactions into blue formazan crystals. The formation of the
blue MTT crystals within the cell decreases in direct proportion to
the viability of cells.[47]
Data and statistical analysis: RLUs per mg protein for each run
were normalized by converting values to fold induction by dividing
the RLUs per mg protein for each well by the average RLUs per
mg protein for the vehicle control. Fold induction values were
then log transformed to control for heterogeneity of variance.
Log10-transformed fold induction values for each plate were further
converted to a percentage of the maximal E2 response using the
following procedure: The concurrent E2 curve for each run in log-
fold (y-axis) was plotted in GraphPad 5.01 (GraphPad Software, San
Diego, CA) versus compound concentration. Concentration values
(x-axis) were log transformed, and then a nonlinear regression
curve fit analysis (four-parameter logistic regression) were per-
formed on the E2 data using a constrained bottom parameter held
constant equal to 0.0. The top parameter (Emax =maximal response)
of the nonlinear regression analysis for E2 was used to convert the
log-fold data into a percent response for each test well. Specifically,
the log-fold response was divided by the top parameter (in log
fold) and then multiplied by 100. Once each run was converted to
percent of the maximal E2 response, data from all runs for each
compound were combined.
Dynamic monitoring of chemical-induced changes in cell proliferation
by IncuCyte ZOOM: Prior to beginning the assay, MCF-7 cells were
seeded in a 96-well plate at 10000 cells per well and cultured over-
night. Compounds were serially diluted with 0.5% DCC-FBS DMEM
growth media containing YOYO-1 (Life Technologies) to a final con-
centration of 0.05 mm. Phase-contrast and fluorescent images were
collected to detect morphological cell changes and plasma mem-
brane permeability (by YOYO-1 DNA staining), both indicators of in
vitro cytotoxicity. The YOYO-1 concentration did not affect prolifer-
ation or cell morphology of the cell types used in this study rela-
tive to identical cells placed in an IncuCyte with a 10” objective in
a standard cell culture incubator at 378C and 5% CO2. Two images
per well were collected every 2 h in both phase-contrast and fluo-
rescence. The assay was considered complete when a maximal re-
sponse was achieved as determined by image analysis. The inte-
grated object counting algorithm was used to isolate the fluores-
cent nuclear signal from background. Specifically, images were seg-
mented in order to identify individual objects, counted, and report-
ed on a per-area (mm2) basis for each time point. For endpoint
assays, in order to correct for differential proliferation of cells, the
total number of DNA containing objects was counted at the final
time point using Triton X-100 (0.0625%) to permeabilize the cells.
Cells were incubated at 378C for 1 h to allow nuclear DNA staining
by YOYO-1 prior to endpoint imaging. This number was used to
calculate the “cytotoxic index”, defined as the number of YOYO-1-
positive objects divided by the total number of DNA-containing
objects (fluorescent objects counted post Triton X-100 treatment).
For ER competition assays, intra-assay data were normalized by ref-
erence to 100% binding of 5 nm [3H]E2. Data were then expressed
as a percentage of competition by increasing concentrations of un-
labeled competitors (1 pm–10 mm). Dose–response curves were
fitted to four-parameter logistic equations by nonlinear regression
analyses.
Docking studies
Docking simulations were performed with Glide software (Glide,
version 6.2; Schrçdinger LLC, New York, NY (USA), 2014). The X-ray
structures of ERa LBD to be used for docking were taken from the
Protein Databank (www.rcsb.org).[48] The 3D coordinates of the cor-
responding ERa and ERb complexes were extracted from PDB en-
tries 1A52, 1ERR, 1X7J, 1X7R, and 3ERT. The PDB structures were
prepared for docking using the Protein Preparation Workflow
(Schrçdinger) accessible from within the Maestro program (Maes-
tro, version 9.7; Schrçdinger). Shortly, the hydrogens were properly
added to the complexes, water molecules more than 5 Š from
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1410
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[12] N. Mibu, K. Sumoto, Chem. Pharm. Bull. 2000, 48, 1810–1813.
[13] R. A. Al-Qawasmeh, Y. Lee, M. Y. Cao, X. Gu, A. Vassilakos, J. A. Wright, A.
Young, Bioorg. Med. Chem. Lett. 2004, 14, 347–350.
[14] R. D. Bindal, J. T. Golab, J. A. Katzenellbogen, J. Am. Chem. Soc. 1990,
112, 7861–7868.
[15] R. D. Bindal, J. A. Katzenellenbogen, J. Med. Chem. 1988, 31, 1978–1983.
[16] M. K. Parai, G. Panda, V. Chaturvedi, Y. K. Manju, S. Shina, Bioorg. Med.
Chem. Lett. 2008, 18, 289–292.
[17] P. Wang, J. Kozlowski, M. Cushman, J. Org. Chem. 1992, 57, 3861–3866.
[18] Y. Q. Long, X. H. Jiang, R. Dayam, T. Sanchez, R. Shoemaker, S. Sei, N.
Neamati, J. Med. Chem. 2004, 47, 2561–2573.
[19] E. R. Simpson, M. Misso, K. N. Hewitt, R. A. Hill, W. C. Boon, M. E. Jones,
A. Kovacic, J. Zhou, C. D. Clyne, Endocr. Rev. 2005, 26, 322–330.
[20] R. P. Barros, J. A. Gustafsson, Cell Metab. 2011, 14, 289–299.
[21] F. Mauvais-Jarvis, D. J. Clegg, A. L. Hevener, Endocr. Rev. 2013, 34, 309–
338.
a heteroatom were deleted, and bond corrections were applied to
the co-crystallized ligands. The receptor was optimized in Maes-
tro 9.2 using the OPLS_2005 force field prior to the docking study.
In the final stage, optimization and minimization of the ligand–pro-
tein complexes were carried out with the OPLS_2005 force field,
and the default values for an RMSD of 0.30 Š for non-hydrogen
atoms were used. The receptor grids were generated using the
prepared proteins, with the docking grids centered on the center
of the bound ligand for each receptor. A receptor grid was gener-
ated using a 1.00 van der Waals (vdW) radius scaling factor and
a 0.25 partial charge cutoff. The binding sites were enclosed in
3
a grid box of 20 Š with default parameters and without con-
straints. The 3D structures of the ligands to be docked were gener-
ated and prepared for docking using the LigPrep as implemented
in Maestro 9.7 (LigPrep, version 2.9; Schrçdinger). In this stage,
a series of treatments are applied to the structures. Finally the geo-
metries are optimized using an OPLS_2005 force field. The ligands
were docked using the extra precision mode (XP)[49] without using
any constraints and a 0.80 vdW radius scaling factor and a 0.15
partial charge cutoff. The generated ligand poses were evaluated
with an empirical scoring function, GlideScore a modified version
of ChemScore,[50] was used to estimate binding affinity and rank li-
gands.[51] 3–5 poses per ligand were generated, and post-docking
minimization was carried out. XP Pose Rank was used to select the
best docked pose for each ligand.
[22] G. G. Kuiper, E. Enmark, M. Peltro-Huikko, S. Nilsson, J. A. Gustafsson,
Proc. Natl. Acad. Sci. USA 1996, 93, 5925–5930.
[23] Z. Liu, H. Shen, J. Huang, S. Chen, L. Chen, J. Chen, G. Liu, H. Jiang, X.
Shen, J. Steroid Biochem. Mol. Biol. 2008, 110, 150–156.
[24] R. Chesworth, M. D. Wessel, L. Heyden, F. M. Mangano, M. Zawistoski, L.
Gegnas, D. Galluzzo, B. Lefker, K. O. Cameron, J. Tickner, B. Lu, T. A. Cas-
tleberry, D. N. Peterson, A. Braultt, P. Perry, O. Ng, T. A. Owen, L. Pa, H. Z.
Ke, T. A. Brown, D. D. Thompson, P. DaSilva-Jardine, Bioorg. Med. Chem.
Lett. 2005, 15, 5562–5566.
[25] J. A. McKie, S. S. Bhagwat, H. Brady, M. Doubleday, L. Gayo, L. Hickmam,
R. K. Jalluri, S. Khammungkhune, A. Kois, D. Mortensen, N. Richard, J. Sa-
pienza, G. Shevlin, B. Stein, M. Sutherland, Bioorg. Med. Chem. Lett.
2004, 14, 3407–3410.
[26] N. Heldring, A. Pike, S. Andersson, J. Matthews, G. Cheng, J. Hartman,
M. Tujague, A. Strçm, E. Treuter, M. Warner, J. A. Gustafsson, Physiol. Rev.
2007, 87, 905–931.
Acknowledgements
[27] A. R. Khosropour, K. Esmaeilpoor, A. Moradie, J. Iran. Chem. Soc. 2006, 3,
81–84.
[28] D. Guzmµn-Lucero, J. Guzmµn, D. Likhatchev, R. Martínez-Palou, Tetrahe-
dron Lett. 2005, 46, 1119–1122.
[29] L.-T. An, F.-Q. Ding, J.-P. Zou, Dyes Pigm. 2008, 77, 478–480.
[30] Z. H. Zhang, F. L. Yang, C. G. Fu, Synth. Commun. 1997, 27, 3823–3828.
[31] S. Reddy, A. Nagaraj, A. Srinivas, G. P. C. Reddy, Ind. J. Chem. 2009, 48B,
248–254.
[32] A. M. Paul, A. C. Khandekar, B. M. Khadikar, J. Chem. Res. 2003, 168–169.
[33] J.-T. Hou, J.-W. Gao, Z.-H. Zang, Monatsh. Chem. 2011, 142, 495–499.
[34] M. A. Pasha, S. Nagashree, Int. J. Res. Chem. Environ. 2013, 3, 54–58.
[35] G. Guedes, M. López-Rodríguez, A. G. Ravelo, A. EstØvez-Braun, Eur. J.
Org. Chem. 2012, 5757–5766.
The authors gratefully acknowledge financial support from the
Spanish Ministry of Economy and Finance (MINECO) SAF2012-
37344-C03-C01 and SAF2012-37344-C03-C02. These projects were
also co-founded by the European Regional Development Fund
(ERDF). The authors also thank the European Union (EU) Re-
search Potential (FP7-REGPOT-2012-61367-IMBRAIN). L.F.-P. was
supported by Grants-in-Aid from the Canarian Agency for Re-
search (ACIISI) (PI2010/0110) and the Alfredo Martin-Reyes Foun-
dation (Arehucas)-FICIC. G.G. thanks GAC for a pre-doctoral
grant. R.J.-M. was supported by a grant from the University of
Las Palmas de Gran Canaria (ULPGC).
[36] S. Genovese, F. Epifano, C. Pelucchini, M. Curini, Eur. J. Org. Chem. 2009,
1132–1135.
[37] V. Wilson, K. Bobseine, L. G. Gray, Jr., Toxicol. Sciences 2004, 81, 69–77.
[38] Schrçdinger, Schrçdinger LLC, New York, 2005, www.schrodinger.com.
[39] A. C. W. Pike, A. M. Brzozowski, R. E. Hubbard, T. Bonn, A. G. Thorsell, O.
Engstrom, J. Ljunggren, J. A. Gustafsson, M. Carlquist, EMBO J. 1999, 18,
4608–4618.
Keywords: molecular docking
estrogen receptors · triarylmethanes
·
selective modulation
·
[40] J. Marrero-Alonso, A. Morales, B. García-Marrero, A. Boto, R. Marín, D.
Cury, T. Gómez, L. Fernµndez-PØrez, F. Lahoz, M. Díaz, Eur. J. Pharm. Bio-
pharm. 2013, 85, 898–910.
[41] J. M. Trowbridge, I. Rogatsky, M. J. Garabedian, Proc. Natl. Acad. Sci. USA
1997, 94, 10132–10137.
[1] R. Aldag, Photochrominm: Molecules and Systems (Eds: H. Dürr, H.
Bouas-Laurent), Elsiever, London, 1990.
[2] P. Rys, H. Zollinger, Fundamentals of the Chemistry and Application of
Dyes, Willey-Interscience, New York, 1972.
[3] R. Muthyala, A. R. Katritzky, X. Lan, Dyes Pigm. 1994, 25, 303–324.
[4] R. Muthyala, Chemistry and Applications of Leuco Dyes, Plenum, New
York, 1997.
[42] W. S. Shim, M. Conaway, S. Masamura, W. Yue, J. P. Wang, R. Kmar, R. J.
Santen, Endocrinology 2000, 141, 396–405.
[43] S. Kato, H. Endoh, Y. Masuhiro, T. Kitamoto, S. Uchiyama, H. Sasaki, S.
Masushige, Y. Gotoh, E. Nishida, H. Kawashima, D. Metzger, P. Chambon,
Science 1995, 270, 1491–1494.
[5] D. F. Duxbury, Chem. Rev. 1993, 93, 381–433.
[6] S. Brasselet, F. Cherioux, P. Audebert, J. Zyss, Chem. Mater. 1999, 11,
1915–1920.
[7] F. Jafarpour, G. R. Bardejee, H. Pirenlahi, V. Oroojpour, H. Dehnamaki, S.
Rahmdel, Chin. J. Chem. 2009, 27, 1415–1419.
[8] K. Baba, K. Maeda, Y. Tabata, M. Doi, M. Kosawa, Chem. Pharm. Bull.
1988, 36, 2977–2983.
[44] X. Hu, M. A. Lazar, Nature 1999, 402, 93–96.
[45] D. Shao, T. J. Berrodin, E. Manas, D. Hauze, R. Powers, A. Bapat, D.
Gonder, R. C. Winneker, D. E. Frail, J. Steroid Biochem. Mol. Biol. 2004, 88,
351–360.
[9] L. Bai, N. Masukawa, M. Yamaki, S. Takagi, Phytochemistry 1998, 47,
1637–1640.
[46] D. D. Perrin, W. L. F. Amarego, Purification of Laboratory Chemicals,
3rd ed., Pergamon Press, Oxford, 1988.
[10] C. Jin, R. G. Michetich, M. Daneshtalab, Phytochemistry 1999, 50, 505–
508.
[11] O. Seligmann, H. Wagner, Tetrahedron 1981, 37, 2601–2606.
[47] T. Mosmann, J. Immunol. Methods 1983, 65, 55–63.
[48] H. M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T. N. Bhat, H. Weissig,
I. N. Shindyalov, P. E. Bourne, Nucleic Acids Res. 2000, 28, 235–242.
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1411
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[49] R. A. Friesner, R. B. Murphy, M. P. Repasky, L. L. Frye, J. R. Greenwood,
T. A. Halgren, P. C. Sanschagrin, D. T. Mainz, J. Med. Chem. 2006, 49,
6177–6196.
[51] R. A. Friesner, J. L. Banks, R. B. Murphy, T. A. Halgren, J. J. Klicic, D. T.
Mainz, M. P. Repasky, E. H. Knoll, M. Shelley, J. K. Perry, D. E. Shaw, P.
Francis, P. S. Shenkin, J. Med. Chem. 2004, 47, 1739–1749.
[50] M. D. Eldridge, C. W. Murray, T. R. Auton, G. V. Paolini, R. P. J. Mee, J.
Comput.-Aided Mol. Des. 1997, 11, 425–445.
Received: April 6, 2015
Published online on June 15, 2015
ChemMedChem 2015, 10, 1403 – 1412
www.chemmedchem.org
1412
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim