Design, synthesis, and estrogenic activity of a novel estrogen receptor modulator - A hybrid structure of 17β-estradiol and vitamin E in hippocampal neurons

Article abstract of DOI:10.1021/jm070546x

We recently discovered that ICI 182,780 (1), an antagonist of estrogen receptor (ER)-dependent proliferation in reproductive tissues, functions as an estrogenic agonist in primary neurons. The present study investigated whether the agonist properties of 1 in neurons could be translated into structural analogs. 7α-[(4R,8R)-4,8,-12-trimethyltridecyl]estra-1,3,5-trien-3, 17β-diol (2), a hybrid structure of 17β-estradiol and vitamin E, was synthesized and found to bind to both ERα and ERβ. In vitro analyses demonstrated that 2 was neuroprotective and effective in activating molecular mechanisms associated with estrogenic agonist activity in rat primary hippocampal neurons. Collectively, the data support an estrogenic agonist profile of 2 action comparable to 1 in primary neurons, confirming that estrogenic activity of 1 in neurons is not a unique phenomenon. These results provide support for the development of a brain-selective ER modulator, with potential as an efficacious and safe estrogen alternative to prevent Alzheimer's disease and cognitive decline in postmenopausal women.

Full text of DOI:10.1021/jm070546x

J. Med. Chem. 2007, 50, 4471-4481
4471
Design, Synthesis, and Estrogenic Activity of a Novel Estrogen Receptor ModulatorsA Hybrid
Structure of 17â-Estradiol and Vitamin E in Hippocampal Neurons
Liqin Zhao,† Chunyang Jin,‡ Zisu Mao,† Madathil B. Gopinathan,‡ Kenneth Rehder,*,‡ and Roberta D. Brinton*,†
Department of Pharmacology and Pharmaceutical Sciences, School of Pharmacy, UniVersity of Southern California,
Los Angeles, California 90089, and Center for Organic and Medicinal Chemistry, RTI International,
Research Triangle Park, North Carolina 27709
ReceiVed May 9, 2007
We recently discovered that ICI 182,780 (1), an antagonist of estrogen receptor (ER)-dependent proliferation
in reproductive tissues, functions as an estrogenic agonist in primary neurons. The present study investigated
whether the agonist properties of 1 in neurons could be translated into structural analogs. 7R-[(4R,8R)-4,8,-
12-trimethyltridecyl]estra-1,3,5-trien-3,17â-diol (2), a hybrid structure of 17â-estradiol and vitamin E, was
synthesized and found to bind to both ERR and ERâ. In Vitro analyses demonstrated that 2 was neuroprotective
and effective in activating molecular mechanisms associated with estrogenic agonist activity in rat primary
hippocampal neurons. Collectively, the data support an estrogenic agonist profile of 2 action comparable to
1 in primary neurons, confirming that estrogenic activity of 1 in neurons is not a unique phenomenon.
These results provide support for the development of a brain-selective ER modulator, with potential as an
efficacious and safe estrogen alternative to prevent Alzheimer’s disease and cognitive decline in
postmenopausal women.
Introduction
Compound 1 is a 7R-alkylsulfinyl analog of 17â-estradiol
and binds to both ERR and ERâ with an affinity comparable to
17â-estradiol.17 By competing for the ER ligand binding site
and subsequently driving the receptor to reorient into a
characteristic antagonistic conformation,18 1 abolishes the
nuclear ER-dependent proliferative actions of estrogen in
reproductive organs such as the breast and uterus.19 In addition,
1 is devoid of estrogen-like agonist activity in these organs, in
contrast to the partial estrogen agonist/antagonist tamoxifen
(TMX), and TMX-resistant tumors are sensitive to 1 treatment.19
As a result, the FDA has approved the use of 1 as an adjuvant
endocrine therapy to treat ER-positive metastatic breast cancers
in postmenopausal women with disease progression following
first-line antiestrogen therapy (i.e., for TMX-resistant breast
cancers).20 Therefore, from a therapeutic development perspec-
tive, our finding that 1 acts as an estrogenic agonist in neurons
provides a promising translational opportunity for the develop-
ment of a brain-selective SERM mimicking the beneficial effects
of an estrogen agonist in the brain, while lacking or antagonizing
activation of estrogenic proliferative responses in reproductive
organs.10
Our initial efforts to develop a brain-selective SERM have
two objectives: design and synthesize structural analogs of 1
to (a) replicate the promising in vitro activity of 116 and (b)
cross the blood-brain barrier (BBB) because 1 does not.19
Herein, we report our efforts to design and synthesize the first
candidate compound, 7R-[(4R,8R)-4,8,12-trimethyltridecyl]estra-
1,3,5-trien-3,17â-diol (2), and first-stage analyses of its estro-
genic activity in cell-based assays in primary hippocampal
neurons.
Alzheimer’s disease (ADa), a devastating neurodegenerative
condition associated with impaired memory and cognitive
function, affects an estimated 4.5 million people in the United
States.1 Of those affected with AD, 68% are female and 32%
are male.2-4 The greater female vulnerability to AD has been
associated with the marked decrease in the level of estrogen
circulating in postmenopausal women.5,6 In addition to its
multifaceted health-promoting effects on a woman’s body, such
as counteraction of postmenopausal symptoms and preservation
of bone density, research over the past two decades has
supported the use of estrogen therapy (ET) for the prevention
of AD and other age-related neurodegenerative insults when
timely initiated7-9 based on the “healthy cell bias” of estrogen
actions in neurons.10-12 However, side effects of the currently
available ET, such as neoplasm13 and thrombogenesis,14,15
remain serious risks to patients. Therefore, there is an unmet
medical need for estrogen alternatives that are free of long-
term side effects and are perceived as safe by physicians and
patients.
In previous studies, we examined whether existing clinically
relevant selective estrogen receptor modulators (SERMs) would
exert estrogenic agonist effects comparable to the endogenous
estrogen 17â-estradiol in the brain. Using neuronal responsive
predictors associated with estrogen actions in neurons as target
outcomes, we unexpectedly discovered that the full antagonist
to the nuclear estrogen receptor (ER), ICI 182,780 (1), acted as
a full estrogenic agonist in primary neurons at clinically relevant
concentrations.16
* To whom correspondence should be addressed. Tel.: 323-4421430,
Fax: 323-2247473, E-mail: rbrinton@usc.edu (R.D.B.); Tel.: 919-5416681,
Fax: 919-5416499, E-mail: krehder@rti.org (K.R.).
Results and Discussions
Computer Modeling and Design. Using computer-aided
molecular modeling approaches, we first analyzed the three-
dimensional (3D) structural features of 1 in complex with ER.
Analyses demonstrated that, in comparison with the ER full
agonist 17â-estradiol, the nuclear ER full antagonist 1 has two
structural domains that carry out distinct functions. First, the
† University of Southern California.
‡ RTI International.
a Abbreviations: AD, Alzheimer’s disease; ET, estrogen therapy; ER,
estrogen receptor; SERM, selective estrogen receptor modulator; TMX,
tamoxifen; RAL, raloxifene; AF, activation function; BBB, blood-brain
barrier; DIV, days in vitro; ERK, extracellular signal-regulated kinase; Akt,
protein kinase B; PDB, Protein Data Bank.
10.1021/jm070546x CCC: $37.00 © 2007 American Chemical Society
Published on Web 08/14/2007

4472 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18
Zhao et al.
Figure 1. 3D-bound confirmations of 17â-estradiol and 1 in human ERR (hERR). The complex model of 17â-estradiol/hERR was generated based
on the crystallographic structural coordinates downloaded from the Protein Data Bank (PDB ID: 1ERE). The complex model of 1/hERR was
generated from docking calculations with GOLD, followed by energy optimization with InsightII/Discovery. The complex models of 17â-estradiol/
hERR and 1/hERR were superimposed by aligning the 3D coordinates of hERR. Both 17â-estradiol and 1 are depicted in “ball and stick” format
and colored based on the atom types designated in InsightII: red, oxygen atoms; yellow, sulfur atoms; aqua, fluorine atoms; for distinction, the
carbon atoms in 17â-estradiol and 1 are colored in blue and green, respectively. hERR is rendered by its solvent accessible surface depicted in dots
or solid format. These models revealed that the “head moiety”, also referred to as the “ER binding domain” in 1, the 17â-estradiol core structure,
lies in the center of the ligand binding pocket on hERR, in a similar mode to that of 17â-estradiol, although the 13â-methyl groups in both
structures are situated in opposite orientations. The “tail moiety” (7R-alkylsulfinyl side chain), also referred to as the “effector domain” in 1,
protrudes out of the ligand binding pocket and nestles in a relatively hydrophobic groove. This is proposed to be the driving force for the remodeling
of helix 12, leading to impaired ER dimerization and the resultant blockage of nuclear uptake. Images were generated with InsightII 2000 on a SGI
Octane graphical workstation equipped with the IRIX 6.5 operating system.
“head moiety”, represented by the 17â-estradiol core structure,
is accommodated into the ligand binding pocket on ER in a
similar mode to that of 17â-estradiol, although the 13â-methyl
groups on both structures are situated in opposite orientations
(Figure 1). Owing to its driving role in the ligand binding to
ER, as revealed in our previous analyses,10,21 we refer to this
structural moiety as the “ER binding domain” in 1. Second is
the “tail moiety”, represented by the bulky side chain substituted
on the 7R position of 17â-estradiol core structure, which
protrudes out of the ligand binding pocket and occupies the same
channel on ER as occurs with ICI 164,384 (3), a structural
analog of 1 (Figure 1).18 In comparison with many ER partial
agonists/antagonists, such as TMX and raloxifene (RAL) with
relatively shorter “tail moieties” attached to their ER binding
core structures,10 it has been suggested that the extended terminal
portion of the “tail moiety” in 1 nestles within the activation
function-2 (AF-2) cleft that precludes helix 12, a structural
component on ER that is crucial for ER dimerization, from
staying in the same cleft.22 This indicates a unique reorientation
distinct from either its characteristic agonist orientation, as seen
Figure 2. Compound 2 is a hybrid structure of 17â-estradiol and
vitamin E.
with 17â-estradiol, in which helix 12 is aligned over the ligand
binding pocket,22 or its partial antagonist orientation, as seen
with TMX or RAL, in which helix 12 orients along the
coactivator binding surface in AF-2.22 As a result, such a
reorientation of helix 12 induced by 1 leads to a full abolition
of ER transcription regulated by both AF-1 and AF-2, which
accounts for its “pure” antiestrogenic effect, particularly in
reproductive tissues.17,19 In addition, it has been suggested that
this displacement of helix 12 impairs ER dimerization and
inhibits the translocation of ER into the nucleus.23 In view of
the significant role of the “tail moiety” in 1 in shifting the
direction of its impact on the nuclear ER function, we refer to
this structural moiety as its “effector domain”. Overall, these
comparative analyses provide important insights into the rational
design of novel ICI mimics that are anticipated to induce a
pharmacological profile consistent with 1, that is, estrogenic
agonist activity in the brain, while lacking or antagonizing
estrogenic proliferative action in reproductive tissues.
compounds, suggesting the toxicity of the analogs may be less
problematic. One representative compound, 2, is a hybrid
structure of 17â-estradiol and vitamin E (Figure 2), both of
which are brain permeable and widely used in humans.24 While
a structure composed of two BBB-penetrable moieties, estro-
genic “head moiety” and vitamin-like “tail moiety”, cannot be
guaranteed to cross the BBB, it would presumably have greater
potential than combinations composed of brain-inaccessible
moieties. In addition, replacement of the “tail moiety” in 1
(cLogP of 8), with a vitamin E-like hydrophobic side chain,
increases the overall lipid solubility of 2 (cLogP of 12),
comparable to that of vitamin E (clogP of 11). While this
lipophilicity falls out of the range defined by the “Lipinski rule
of five” for druggability and brain penetration,25 vitamin E has
a similar high lipophilicity, yet, can readily enter the brain.24
In view of the complexity of the biological features of BBB26
and the multiple factors that contribute to BBB penetration,27,28
by deliberately mimicking the physicochemical properties of
vitamin E that may jointly impact its brain entry, including
lipophilicity, molecular shape and associated conformational
Guided by insights derived from the above analyses, we
proposed a number of novel ICI analogs predicted to possess
improved brain penetration compared to 1. These analogs have
structural moieties associated with endogenously occurring

Estrogen Receptor Modulator in Hippocampal Neurons
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18 4473
impurities, which could not be removed by chromatographic
purification. Further reaction optimization was successfully
achieved through the use of bis(trimethylsilyl)peroxide and tin
tetrachloride catalysis,42 and (4R,8R)-1-acetoxy-4,8,12-trimeth-
yltridcane (6) was obtained in 62% yield. Hydrolysis of 6
proceeded quantitatively with p-toluenesulfonic acid in refluxing
methanol to yield 7, which was converted to (4R,8R)-1-iodo-
4,8,12-trimethyltridecane (8) in 85% yield upon treatment with
iodine, triphenylphosphine, and imidazole.
With the side chain precursor 8 achieved, focus was next
directed toward the synthesis of a suitably protected steroid core
and subsequent alkylation with 8 (Scheme 2). Oxidation of 3,-
17â-estradiol diacetate (9) with chromium trioxide and 3,5-
dimethylpyrazole gave ketone 10 in 44% yield.43 Hydrolysis
of 10 with potassium hydroxide afforded 6-oxo-3,17â-estradiol
(11), quantitatively. Crude 11, used without further purification,
was then reprotected with tetrahydropyranoxy (THP) groups by
reaction with dihydropyran (DHP) to provide 12 in 98% yield.44
The critical alkylation on the C7-position of the steroid was
achieved by treatment of the potassium enoxyborate of 12,37
generated using potassium t-butoxide and triethylborane, with
(4R,8R)-1-iodo-4,8,12-trimethyltridecane (8) in dimethoxyethane
(DME). The 7R-substituted steroid 13 was obtained as a single
diastereomer in 34% yield. To our knowledge, this is one of
the highest yields reported for the alkylation with this size alkyl
substituent. Finally, deprotection of the THP groups and
deoxygenation of the 6-keto functionality were simultaneously
accomplished with boron trifluoride etherate and triethylsilane37
to produce 2 in 83% yield.
Figure 3. 3D-bound conformations of 1 and 2 in hERR. Models were
derived from docking calculations with GOLD. Both 1 and 2 are
depicted in “stick” format and colored based on the atom types
designated in InsightII: red, oxygen atoms; yellow, sulfur atoms; aqua,
fluorine atoms; except that for distinction, the carbon atoms in 1 and
2 are colored in green and sky blue, respectively. Residues, Glu353
and His524 on hERR, which form hydrogen bond interactions with
both ligands depicted in dashed green lines, are shown in “stick” format
and colored based on the atom types. These models revealed that 2
binds to hERR in a mode similar to 1, suggesting the functional
similarity between two compounds. Image was generated with InsightII
2000 on a SGI Octane graphical workstation equipped with the IRIX
6.5 operating system.
flexibility, and specifically, distribution of a hydrophilic (“head”)/
hydrophobic (“tail”) structural balance that may impact the
interaction with the BBB membrane-water complex, as re-
vealed by recent membrane-interaction quantitative structure-
activity relationship (MI-QSAR) models,29,30 2 is anticipated
to have a similar BBB penetrative ability to vitamin E.
Moreover, 2 has a smaller molecular mass of 496 than 1 at 552
and, therefore, falls below the suggested threshold of 500 for a
brain-permeable molecule.31,32 Most importantly, as revealed
in Figure 3, GOLD docking analyses indicated that 2 binds to
hERR in an energy-favorable fashion, similar to 1. In addition,
hydrogen bond interactions were observed in both compounds
between 3, 17â-OH groups, and the residues, glu353 and
His524, respectively, along the ligand binding site in hERR.
The similar binding modes and ligand-receptor intermolecular
interactions exhibited by 1 and 2 suggested that 2 would exert
a tissue-selective modulation of ER that is consistent with 1.
Chemistry. Synthesis of 7R-substituted estradiol derivatives
has been generally achieved by copper-promoted nucleophilic
1,6-conjugate addition of Grignard reagents to steroidal dienones
such as 6-dehydrotestosterone 17-acetate.33,34 However, such
additions usually lead to a mixture of 7R- and 7â-epimers and,
subsequently, require rearomatization to generate the aromatic
A-ring. More recently, another synthetic approach using alkyl
iodides to alkylate 3,17-diprotected 6-ketoestradiols was found
to be more efficient and versatile for synthesis of 7-substituted
estradiols with high stereoselectivity.35-40 Accordingly, our
synthesis of 7R-[(4R,8R)-4,8,12-trimethyltridecyl]estra-1,3,5-
trien-3,17â-diol (2) was accomplished following the second
synthetic strategy, starting from preparation of the side chain
intermediate, (4R,8R)-1-iodo-4,8,12-trimethyltridecane (8; Scheme
1). Ozonolysis41 of phytol (4) with ozone at -78 °C followed
by a reductive workup gave phytone (5) in 76% yield. Baeyer-
Villiger oxidation of 5 was first attempted using 5 equiv of
3-chloroperbenzoic acid (m-CPBA) in refluxing chloroform to
afford 6 in 54% yield. However, the product contained trace
The R-configuration for the C7-substituent of 2 was assigned
based on two-dimensional NMR (COSY and ROESY) spectral
analyses. Strong correlation was observed between H(7) and
H(8), indicating that these two protons were proximal (Figure
4). Based on the known â-stereochemistry of H(8), the H(7) is
also assigned the â position. No correlation between H(7) and
H(9) was detected. The C6-benzylic protons were clearly evident
at δ 2.86 and 2.72, with coupling constants of 4.8, 16.9, and
16.9 Hz, respectively, which is also consistent with the structure.
Estrogenic Activity of 2. We first determined the binding
affinity and specificity of 2 to ERR and ERâ. A fluorescent
polarization competitive binding assay was used, which is
composed of purified baculovirus-expressed human ERR or ERâ
and a fluorescent estrogen ligand EL Red. Progesterone was
used as a negative control, and known ER ligands, 17â-estradiol,
1, and genistein, were used as positive controls. The ability of
the test compounds at serially diluted concentrations (100 pM
to 10 µM) to compete with the estrogen ligand EL Red for
binding to ERR or ERâ was assessed by a change in polarization
values at 535 nm/590 nm excitation/emission. Figure 5 shows
the competition binding curves of the test compounds for both
ERR and ERâ. As expected, the negative control compound,
progesterone, does not bind to either ER. The IC50 determined
from the binding curves for positive estrogen controls, 17â-
estradiol (4.7 and 16.7 nM for ERR and ERâ, respectively) and
1 (4.9 and 44.1 nM for ERR and ERâ, respectively), are
consistent with the previously reported values.21 Moreover, the
assay was sufficiently sensitive to differentiate the ERâ-binding
preference of the phytoestrogen, genistein, with a 46.8-fold
binding selectivity over ERR, which is consistent with results
derived from alternative methods such as the radioligand assay.45
These comparative analyses demonstrate the reliability of this
assay in determining the binding profiles of small compounds
to both ERs. Compound 2 comparably bound to both ERR, with
a binding IC50 of 193 nM, and ERâ, with a binding IC50 of 267

4474 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18
Zhao et al.
Scheme 1a
a Reagents and conditions: (a) ozone, MeOH-CH2Cl2, -78 °C, then (CH3)2S, room temperature, 2.5 h, 76%; (b) SnCl4, (TMSO)2, CH2Cl2, room temperature,
4 h, 62%; (c) TsOH, MeOH, reflux, 6 h, 100%; (d) Ph3P, imidazole, I2, CH2Cl2, room temperature, 2 h, 85%.
Scheme 2a
a Reagents and conditions: (a) CrO3, 3,5-dimethylpyrazole, -20 °C, 5 h, 44%; (b) KOH, MeOH-H2O, room temperature, 4 h, 100%; (c) dihydropyran,
PPTS, CH2Cl2, reflux, 3 h, 98%; (d) KOt-Bu, Et3B, DME, room temperature, 1 h, then 8, overnight, 34%; (e) Et3SiH, BF3‚Et2O, CH2Cl2, room temperature,
overnight, 83%.
Figure 4. Key ROESY correlations of 2.
nM. Although the binding affinity of 2 to both ERs is
approximately 10- to 50-fold lower than those for 17â-estradiol
and 1, they are well within the therapeutic development range.
To determine whether 2 would act as an estrogenic agonist
in neurons comparable to 17â-estradiol and 1, we first evaluated
the activity of 2 to protect neurons against the neurodegenerative
insult, a supraphysiological concentration of glutamate-induced
neurotoxicity in cultured rat primary hippocampal neurons.
Neuronal viability was assessed by dual measurements of lactate
dehydrogenase (LDH) release in the culture medium, which
Figure 5. Compound 2 binds to both ERR and ERâ, with comparable
IC50 values, 192 nM for ERR and 267 nM for ERâ.
served as an indicator of neuronal membrane integrity, and
calcein acetoxymethyl ester (AM) staining, which served as an
indicator of neuronal metabolic viability. Data shown in Figure
6 demonstrated that 2 promoted neuronal survival in a concen-
tration-dependent manner. The amount of LDH released in the
culture medium induced by 200 µM glutamate was significantly
reduced by 2 at all test concentrations (1-1000 nM), while the
efficacy induced by 100-1000 nM was significantly greater
than that induced by 1-10 nM of 2 (++P < 0.01). There were
no significant differences in LDH release between cultures
treated with 1 nM and 10 nM (Figure 6A, 38.6 ( 3.8% and

Estrogen Receptor Modulator in Hippocampal Neurons
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18 4475
Figure 6. Compound 2 promoted neuronal survival against supraphysiological glutamate-induced neurotoxicity in a concentration-dependent manner
in rat primary hippocampal neurons. Neurons grown for 7 DIV were pretreated with vehicle alone or 2 at serially diluted concentrations (1, 10, 100,
and 1000 nM) for 48 h prior to exposure to 200 µM glutamate for 5 min, followed by a 24 h recovery. Neuroprotective activity was determined
by (A) LDH release in the culture medium, an indicator of neuronal membrane integrity, and (B) calcein AM staining of the live neurons in the
culture, an indicator of neuronal metabolic viability, from the same population of neurons. Results are presented as neuroprotective efficacy (NE),
which is defined as the percentage of glutamate-induced neurotoxicity prevented by 2 treatment and quantitated by the equation: NE ) (V2 -
V
glutamate)/(Vcontrol - Vglutamate) × 100%, where V2 ) the individual value from 2-treated cultures, Vglutamate ) the mean value from glutamate alone-
treated cultures, and Vcontrol ) the mean value from vehicle alone-treated control cultures. Data, expressed as mean ( SEM, n g 6, are derived from
a single experiment and are representative of three separate experiments. ##P < 0.01 compared to vehicle alone treated control cultures, **P < 0.01
compared to glutamate alone-treated cultures. ++P < 0.01 indicates the differences between 1-10 nM and 100-1000 nM of 2 treatment groups.
44.1 ( 3.8% increase in neuronal membrane integrity compared
with glutamate alone-treated cultures, respectively, **P < 0.01),
and between cultures treated with 100 nM and 1000 nM of 2
(Figure 6A, 67.0 ( 4.0% and 59.1 ( 3.7% increase in neuronal
membrane integrity compared with glutamate alone-treated
cultures, respectively, **P < 0.01). Data shown in Figure 6B,
derived from calcein AM staining of metabolically live neurons
in the cultures, revealed a similar trend in neuronal response to
serially diluted concentrations of 2. A significant increase in
neuronal viability was observed in cultures treated with 100-
1000 nM of 2 (Figure 6B, 28.4 ( 3.2% and 25.6 ( 4.4%
increase in neuronal metabolic viability compared with glutamate
alone-treated cultures, respectively, **P < 0.01). In contrast, 2
at 1-10 nM was insufficient to prevent the loss of neuronal
metabolic activity induced by glutamate insult (Figure 6B, 2.7
( 3.2% and 9.3 ( 3.7% increase in neuronal metabolic viability
compared with glutamate alone-treated cultures, respectively),
although 1-10 nM was effective in protecting neurons against

4476 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18
Zhao et al.
Figure 8. Compound 2 upregulated the anti-apoptotic proteins Bcl-2
and Bcl-XL expression in rat primary hippocampal neurons. Neurons
grown for 7 DIV were treated with vehicle alone, E2 (10 nM), or 2
(100 nM) for 48 h followed by Western immunoblotting analyses on
Bcl-2 and Bcl-XL expression in the whole cell lysate preparation of
neurons. â-Tubulin was used as an internal loading control protein.
Results are presented as percent increase in Bcl-2 and Bcl-XL expression
compared to vehicle alone treated control cultures and expressed as
mean ( SEM, n g 3, *P < 0.05. E2: 17â-estradiol.
Figure 7. Compound 2 rapidly increased ERK2 and Akt phosphoryl-
ation in rat primary hippocampal neurons. Neurons grown for 7 DIV
were treated with vehicle alone, E2 (10 nM) or 2 (100 nM) for 30 min
followed by Western immunoblotting analyses on phosphorylated and
total ERK2 and Akt expression in the whole cell lysate preparation of
neurons. pERK2 and pAkt levels were normalized against the levels
of total ERK2 and Akt, respectively. Results are presented as percent
increase in pERK2 and pAkt expression compared to vehicle alone-
treated control cultures and expressed as mean ( SEM, n ) 4, *P <
0.05. E2: 17â-estradiol.
neuronal survival.49-52 Our previous analyses demonstrated that
1 was effective in promoting both ERK and Akt phosphorylation
in hippocampal neurons.16 In this experiment, we sought to
determine whether 2 would activate these same signaling
mechanisms. Rat hippocampal neurons grown for 7 DIV were
B27 supplement-deprived for 45 min prior to incubation with
vehicle alone, 17â-estradiol (10 nM), or 2 (100 nM) for 30 min
prior to harvesting of proteins for detection of phosphorylated
ERK and Akt expression by Western immunoblotting analyses.
Total ERK and Akt expression levels in the same protein
samples were detected and used as loading controls. Results of
these analyses indicated that exposure of neurons to 2 rapidly
induced a significant increase in phosphorylation of both ERK2
and Akt (Figure 7, 46.3 ( 14.7% and 139.1 ( 33.4% increase
compared to vehicle alone treated control cultures, respectively,
*P < 0.05), with efficacy slightly lower than but not signifi-
cantly different from that induced by 17â-estradiol (Figure 7,
38.0 ( 7.7% and 88.0 ( 27.2% increase compared to vehicle
alone-treated control cultures, respectively, *P < 0.05).
Estrogen upregulation of Bcl-2 family anti-apoptotic proteins
Bcl-2 and Bcl-XL has been proposed as one critical component
underlying estrogen promotion of neuronal survival.47,49,53
Upregulation of both Bcl-2 and Bcl-XL expression by 1 was
previously observed as well.16 Accordingly, we evaluated
whether 2 regulated these proteins in rat primary hippocampal
neurons. Neurons grown for 7 DIV were treated with vehicle
alone, 17â-estradiol (10 nM) or 2 (100 nM), for 48 h followed
by Western immunoblotting analyses. Results of these analyses
indicated that 2 induced a significant increase in both Bcl-2
and Bcl-XL expression in neurons (Figure 8, 23.4 ( 6.8% and
58.0 ( 22.2% increase compared to vehicle alone treated control
cultures, respectively, *P < 0.05), with efficacy comparable to
glutamate-induced neuronal membrane damage. These differ-
ences in outcomes derived from measurements of different
biochemical indicators from the same population of neurons
suggested that neuronal membrane damage may be more easily
protected and repaired than damage to neuronal metabolic
function or it may be due to interassay sensitivity. Results of
these analyses are consistent with our previous reports for
multiple estrogens and 1.16,46 Moreover, a 10-fold less potency
associated with 2 than 17â-estradiol and 1, which exhibited the
maximal neuroprotection at 10 nM,16 is consistent with the
differences between the ER binding affinity of 2 and 17â-
estradiol and 1, suggesting that estrogen-inducible neuropro-
tective activity is associated with ER-mediated signaling
cascades. As 2 exhibited a full and optimal neuroprotective
activity at 100 nM, this concentration was used in the following
mechanistic studies. In summary, these data provided the first
line of evidence for an estrogenic agonist profile of 2 action in
neurons consistent with 17â-estradiol and 1.16
Mechanistically, the characterization of signal transduction
pathways activated by estrogen has led to the identification of
proteins that are critical mediators of estrogen activity in
neurons. First, estrogen activation of extracellular signal-
regulated kinase (ERK) leads to activation of the transcription
factor cyclic AMP response element-binding protein, CREB,
which results in increased transcription of multiple neuropro-
tective genes such as Bcl-2 family anti-apoptotic genes47 and
neurotrophic genes such as spinophilin.48 In parallel, through a
unified upstream mechanism, estrogen binding to ER and
subsequent interaction with the phosphatidylinositide-3′-OH
kinase (PI3′K) activate the protein kinase B (also known as
Akt), which phosphorylates the Bcl-2 family pre-apoptotic
member BAD, thereby suppressing apoptosis and promoting

Estrogen Receptor Modulator in Hippocampal Neurons
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18 4477
Investigation of 2 induction of neuronal responses provides two
lines of evidence indicating an estrogenic agonist profile of 2
action in primary hippocampal neurons. First, 2 was found to
be a potent neuroprotectant as evidenced by its promotion of
both neuronal membrane integrity and metabolic viability against
neurodegenerative insult in neurons. Second, 2 effectively
activated signaling cascades and increased expression of proteins
consistent with estrogenic agonist function in neurons and
required for estrogen promotion of neuronal defense and
synaptogenesis. These data indicate that our previous discovery
of the estrogenic agonist activity of the nuclear ER full
antagonist, 1, in primary neurons,16 is not an exclusive feature
of 1 alone and can be achieved through structural analogs that
may have greater BBB penetration potential. Moreover, as
proposed for 1,16 2 is likely acting through a membrane-
associated ER leading to induction of the downstream singling
cascades and neuronal agonistic responses as presented here-
inbefore. However, at this time it cannot be ruled out for
potential non-ER-mediated sites of action for 2 and similar future
compounds which include a phenolic ring as reported by
Simpkins and colleagues.54 Collectively, results of these analyses
provide in vitro proof of principle that development of brain-
selective SERMs could provide a safe and efficacious thera-
peutic alternative to existing hormone therapies to promote
neurological function and health through the postmenopausal
years. In vivo analyses to determine the BBB penetration and
impact on the CNS and periphery of 2 are currently underway.
Figure 9. Compound 2 upregulated the dendritic spine protein,
spinophilin expression in rat primary hippocampal neurons. Neurons
grown for 7 DIV were treated with vehicle alone, E2 (10 nM), or 2
(100 nM) for 48 h followed by Western immunoblotting analyses on
spinophilin expression in the whole cell lysate preparation of neurons.
â-tubulin was used as an internal loading control protein. Results are
presented as percent increase in spinophilin expression compared to
vehicle alone treated control cultures and expressed as mean ( SEM,
n ) 4, *P < 0.05, **P < 0.01, and ++P < 0.01. E2: 17â-estradiol.
17â-estradiol (Figure 8, 20.7 ( 1.8% and 46.6 ( 11.6% increase
compared to vehicle alone-treated control cultures, respectively,
*P < 0.05).
Experimental Section
Chemistry. Melting points were determined on a MEL-TEMP
II capillary melting point apparatus and are uncorrected. NMR (1H,
13C, gHSQC, gHMBC, gCOSY, and ROESY) spectra were obtained
using a Bruker Avance DPX-300 MHz or a Varian Unity Inova
500 MHz NMR spectrometer. Chemical shifts are reported in parts
per million (ppm) with reference to internal solvent. HRMS were
recorded on a Waters Autospec Ultima mass spectrometer and were
performed at the University of Michigan, Ann Arbor, MI. Elemental
analysis was done by Atlantic Microlab Inc., Norcross, GA.
Analytical thin-layer chromatography (TLC) was carried out using
EMD silica gel 60 F254 TLC plates. Flash column chromatography
was done on silica gel 60 (230-400 mesh) or on a CombiFlash
Companion system using Isco prepacked silica gel columns.
Reagents were obtained from Aldrich Chemical Co. and used as
received unless otherwise noted.
In addition to regulating Bcl-2 family anti-apoptotic proteins,
estrogen activation of CREB leads to increased expression of
spinophilin, a protein that is enriched in the heads of neuronal
dendritic spines in hippocampal neurons and which is predictive
of estrogen-inducible promotion of neuronal morphogenesis and
synaptoplasticity.48 Compound 1 was found to be comparably
effective to 17â-estradiol in upregulating spinophilin expression
in primary neurons.16 Based on these earlier findings, we
evaluated the impact of 2 on the expression level of spinophilin,
as an indicator of its neurotrophic potential, in comparison with
17â-estradiol. Data shown in Figure 9 indicated that exposure
of hippocampal neurons to 2 (100 nM) for 48 h induced a
significant increase in spinophilin expression. (Figure 9, 61.7
( 8.2% increase compared to vehicle alone-treated control
cultures, **P < 0.01). Under the same experimental conditions,
17â-estradiol (10 nM) induced a moderate increase (Figure 9,
26.6 ( 6.5% increase compared to vehicle alone-treated control
cultures, *P < 0.05), which was significantly less than that
induced by 2 (++P < 0.01). The present finding is in agreement
with our earlier study that showed a similar trend in the
difference in magnitude of change in spinophilin expression
between 17â-estradiol- and 1-treated neurons.16 These data are
promising in that they suggest that 2 and its structural analogs,
a distinct category of ER ligands from the nuclear ER full
agonist as represented by 17â-estradiol, have a greater potential
in activating mechanisms of neuronal synaptoplasticity and
associated memory function.
(6R,10R)-6,10,14-Trimethylpentadecan-2-one (5).41 A stirred
solution of phytol (4, 12.9 g, 440 mmol) in MeOH (25 mL) and
CH2Cl2 (110 mL) at -78 °C was bubbled with a stream of ozone
until a blue color persisted in the solution. Afterward, nitrogen was
bubbled through the solution until the blue color disappeared.
Dimethyl sulfide (25 mL) was added and the mixture was allowed
to stir at room temperature for 2.5 h. The solvent was removed
under reduced pressure and the residue was partitioned between
H2O (100 mL) and benzene (300 mL). The organic phase was
separated, washed with brine (150 mL), dried (Na2SO4), and
concentrated under reduced pressure. Flash column chromatography
on silica gel (300 g) using 10% EtOAc in hexane gave 6.49 g of
pure 5 and 4.12 g of less pure product. Further purification of the
less pure product on a silica gel column (120 g) using 5 f 10%
EtOAc in hexane afforded another 2.41 g of pure 5, for a combined
yield of 76%, as an oil. 1H NMR (300 MHz, CDCl3) δ 0.83-0.88
(m, 12H), 1.03-1.64 (m, 19H), 2.13 (s, 3H), 2.40 (t, J ) 7.4 Hz,
2H); 13C NMR (75 MHz, CDCl3) δ 19.5, 19.6, 21.3, 22.5, 22.6,
24.1, 24.7, 27.9, 29.6, 32.6, 32.7, 36.4, 37.1, 37.2, 37.3, 39.3, 44.0,
208.8.
Taken together, results of mechanistic analyses provide the
second line of evidence for the estrogenic activity of 2 in
neurons.
(4R,8R)-1-Acetoxy-4,8,12-trimethyltridecane (6). To a stirred
solution of 5 (7.00 g, 26.1 mmol) and bis(trimethylsilyl)peroxide
(4.70 g, 26.3 mmol, Gelest Inc., Morrisville, PA) in anhydrous CH2-
Cl2 (260 mL) at 0 °C under argon was slowly added SnCl4 (6.90
Conclusion
A rationally designed novel mimic of 1, 2, with a hybrid
structure of 17â-estradiol and vitamin E, was synthesized.

4478 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18
Zhao et al.
g, 26.5 mmol). After addition, the solution was stirred at 0 °C for
15 min, warmed to room temperature, and stirred for 1 h. An
additional amount of bis(trimethylsilyl)peroxide (4.70 g, 26.3 mmol)
was added and the mixture was stirred for 3 h. The reaction mixture
was poured into a 10% Na2S2O3 solution (260 mL) and extracted
with ether (350 mL and 2 × 250 mL). The combined organic
extracts were washed with a saturated NaHCO3 solution (200 mL),
dried (Na2SO4), and concentrated under reduced pressure. Flash
column chromatography on silica gel (330 g) using 5% EtOAc in
hexane gave 3.18 g of pure 6 and 2.31 g of less pure product.
Further purification of the less pure product on a silica gel column
(120 g) using 5% EtOAc in hexane afforded another 1.43 g of pure
the reaction mixture was concentrated under reduced pressure. The
residue was taken up in 1 N HCl (50 mL) and extracted with EtOAc
(3 × 50 mL). The combined EtOAc extracts were dried (Na2SO4).
Removal of the solvent under reduced pressure afforded crude 11
(1.95 g, 100%) as a white foam, which was used in the next step
without further purification. 1H NMR (300 MHz, DMSO-d6) δ 0.66
(s, 3H), 1.15-1.62 (m, 8H), 1.70-1.95 (m, 2H), 2.20-2.50 (m,
3H), 3.48-3.60 (m, 1H), 4.48 (d, J ) 4.8 Hz, 1H), 6.99 (dd, J )
8.5, 2.8 Hz, 1H), 7.27 (d, J ) 2.8 Hz, 1H), 7.31 (d, J ) 8.5 Hz,
1H), 9.59 (s, 1H); 13C NMR (75 MHz, DMSO-d6) δ 11.0, 22.4,
25.1, 28.3, 29.7, 36.1, 42.1, 42.5, 43.4, 49.1, 79.8, 111.7, 121.2,
126.7, 132.9, 137.8, 155.6, 197.1.
3,17â-Bis(2-tetrahydropyranyloxy)estra-1,3,5(10)-triene-6-
one (12).44 To a stirred suspension of crude 11 (1.95 g, 6.50 mmol)
in anhydrous CH2Cl2 (65 mL) at room temperature under nitrogen
was added dihydropyran (5.89 mL, 65.0 mmol) and pyridinium
p-toluenesulfonate (33 mg). The mixture was refluxed for 4 h,
cooled to room temperature, and diluted with CH2Cl2 (100 mL).
The solution was washed with brine (3 × 30 mL), dried (Na2SO4),
and concentrated under reduced pressure. Flash column chroma-
tography on silica gel (40 g Isco prepacked column) using 0 f
20% EtOAc in hexane afforded 12 (2.88 g, 98%) as a white foam.
1H NMR (300 MHz, CDCl3) δ 0.81, 0.82 (2s, 3H), 1.20-2.28 (m,
22H), 2.30-2.50 (m, 2H), 2.73 (dd, J ) 16.8, 3.3 Hz, 1H), 3.45-
3.65 (m, 2H), 3.73 (q, J ) 8.4 Hz, 1H), 3.80-4.00 (m, 2H), 4.60-
4.70 (m, 1H), 5.46-5.52 (m, 1H), 7.23 (dd, J ) 8.5, 2.7 Hz, 1H),
7.33, 7.35 (2d, J ) 8.5 Hz, 1H), 7.71, 7.72 (2d, J ) 2.7 Hz, 1H);
HRMS (ESI) calcd for C28H38O5 [M + Na]+, 477.2617; found,
477.2611.
3,17â-Bis(2-tetrahydropyranyloxy)-7r-(4R,8R,12-trimethyl-
tridecyl)estra-1,3,5(10)-triene-6-one (13). To a stirred solution of
12 (0.46 g, 1.01 mmol) in DME (5 mL) at room temperature under
nitrogen was added a 1 M solution of potassium t-butoxide (1.11
mL, 1.11 mmol) in THF. After stirring for 10 min, a 1 M solution
of triethyl-borane (1.26 mL, 1.26 mmol) in THF was added and
stirring was continued for 1 h at room temperature. A solution of
8 (444 mg, 1.26 mmol) in DME (2 mL) was then added to the
resulting enolate. After stirring for 40 min, an additional equivalent
of potassium t-butoxide solution was added and the mixture was
stirred for another 16 h. The reaction was quenched with H2O (20
mL) and extracted with EtOAc (4 × 50 mL). The combined EtOAc
extracts were dried (Na2SO4) and concentrated under reduced
pressure. Flash column chromatography on silica gel (40 g Isco
prepacked column) using of 0 f 20% EtOAc in hexane afforded
13 (232 mg, 34%) as an oil. 1H NMR (300 MHz, CDCl3) δ 0.75-
0.91 (m, 15H), 0.95-1.76 (m, 36H), 1.78-2.18 (m, 6H), 2.28-
2.50 (m, 2H), 2.52-2.68 (m, 1H), 3.45-3.68 (m, 2H), 3.75 (q, J
) 8.4 Hz, 1H), 3.85-4.00 (m, 2H), 4.62-4.73 (m, 1H), 5.44-
5.52 (m, 1H), 7.20 (dd, J ) 8.4, 2.4 Hz, 1H), 7.31, 7.32 (2d, J )
8.4 Hz, 1H), 7.69 (s, 1H); 13C NMR (75 MHz, CDCl3) δ 11.7,
19.0, 19.1, 19.6, 19.8, 19.9, 22.5, 22.6, 22.8, 22.9, 24.1, 24.2, 24.7,
25.0, 25.4, 25.8, 26.9, 27.3, 28.2, 28.9, 30.5, 31.2, 31.3, 32.9, 33.0,
37.2, 37.3, 37.5, 37.6, 37.7, 39.6, 42.5, 42.6, 42.7, 43.0, 43.5, 45.5,
45.6, 48.9, 49.0, 49.1, 62.1, 62.4, 62.5, 62.7, 84.2, 85.6, 96.6, 96.8,
96.9, 99.5, 114.9, 122.4, 122.5, 127.1, 127.2, 132.6, 139.6, 139.8,
155.7, 155.8, 201.0, 201.1; HRMS (ESI) calcd for C44H70O5 [M +
Na]+, 701.5121; found, 701.5125.
7r-[(4R,8R)-4,8,12-Trimethyltridecyl]estra-1,3,5(10)-trien-3,-
17â-diol (2). To a stirred solution of 13 (207 mg, 0.30 mmol) in
anhydrous CH2Cl2 (13 mL) at room temperature under nitrogen
was added triethylsilane (3.35 mL, 21.0 mmol). The mixture was
cooled to 0 °C and boron trifluoride etherate (11.3 mL, 90.0 mmol)
was added dropwise. The yellow solution was warmed to room
temperature and stirred for 16 h. The reaction was carefully
hydrolyzed with a 10% K2CO3 solution (60 mL) and then passed
through a plug of silica gel. The filtrate was extracted with CH2-
Cl2 (4 × 50 mL). The combined CH2Cl2 extracts were dried (Na2-
SO4) and concentrated under reduced pressure. Flash column
chromatography on silica gel (40 g Isco prepacked column) using
0 f 20% EtOAc in hexane afforded 2 (126 mg, 83%) as a white
foam. 1H NMR (500 MHz; CDCl3) δ 0.79 (s, 3H), 0.81-0.90 (m,
1
6, for a combined yield of 62%, as an oil. H NMR (300 MHz,
CDCl3) δ 0.83-0.88 (m, 12H), 1.02-1.71 (m, 19H), 2.05 (s, 3H),
4.04 (t, J ) 6.8 Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 19.5, 19.7,
20.9, 22.5, 22.6, 24.4, 24.7, 26.1, 27.9, 32.4, 32.7, 33.0, 37.1, 37.2,
37.3, 39.3, 64.8, 170.9; HRMS (ESI) calcd for C18H36O2 [M +
Na]+, 307.2613; found, 307.2613.
(4R,8R)-4,8,12-Trimethyltridecanol (7). To a stirred solution
of 6 (4.10 g, 14.4 mmol) in MeOH (90 mL) was added p-TsOH
(110 mg, 0.58 mmol) and the reaction mixture was refluxed for 6
h. After cooling to room temperature, the solvent was removed
under reduced pressure. The residue was dissolved in CH2Cl2 (50
mL) and the CH2Cl2 solution was washed with a saturated NaHCO3
solution (50 mL). The aqueous layer was extracted with CH2Cl2
(2 × 50 mL). The combined CH2Cl2 extracts were dried (Na2SO4).
Removal of the solvent under reduced pressure afforded crude 7
(3.50 g, 100%) as an oil, which was used in the next step without
further purification. 1H NMR (300 MHz, CDCl3) δ 0.83-0.88 (m,
12H), 1.01-1.66 (m, 19H), 3.63 (m, 2H); 13C NMR (75 MHz,
CDCl3) δ 19.6, 19.7, 22.5, 22.6, 24.4, 24.7, 27.9, 30.3, 32.6, 32.7,
32.9, 37.2, 37.3, 37.4, 39.3, 63.1.
(4R,8R)-1-Iodo-4,8,12-trimethyltridecane (8). To a stirred
solution of triphenylphosphine (1.30 g, 4.96 mmol) in anhydrous
CH2Cl2 (17 mL) at room temperature under argon was added
imidazole (0.35 g, 5.13 mmol) followed by iodine (1.25 g, 4.94
mmol). After stirring for 15 min, a solution of 7 (1.00 g, 4.12 mmol)
in anhydrous CH2Cl2 (6 mL) was added to the mixture and the
stirring was continued for another 1.75 h. The mixture was filtered
and the solid was washed with CH2Cl2. The filtrate was concentrated
under reduced pressure. Flash column chromatography on silica
1
gel (120 g) using hexane afforded 8 (1.23 g, 85%) as an oil. H
NMR (300 MHz, CDCl3) δ 0.84-0.88 (m, 12H), 1.02-1.89 (m,
19H), 3.17 (dt, J ) 7.0, 1.5 Hz, 2H); 13C NMR (75 MHz, CDCl3)
δ 7.3, 19.7, 19.8, 22.6, 22.7, 24.4, 24.8, 28.0, 31.3, 32.1, 32.8, 37.2,
37.3, 37.4, 38.0, 39.4; HRMS (EI) calcd for C16H33I [M+],
352.1627; found, 352.1634.
6-Oxo-3,17â-estradiol Diacetate (10).43 To a stirred mixture of
CrO3 (37.4 g, 375 mmol) in anhydrous CH2Cl2 (235 mL) at -20
°C under nitrogen was added 3,5-dimethylpyrazole (35.6 g, 375
mmol). After stirring for 20 min, 3,17â-estradiol diacetate (9, 5.34
g, 15.0 mmol) was added and the reaction mixture was stirred at
-20 to -15 °C for 5 h. Thereafter, 5 N NaOH (150 mL) was added
and the mixture was stirred at -10 °C for 45 min. The mixture
was diluted with H2O (200 mL) and the CH2Cl2 phase was
separated. The aqueous phase was extracted with CH2Cl2 (3 × 200
mL). The combined CH2Cl2 extracts were washed with 1 N HCl
(2 × 100 mL) and brine (200 mL), dried (Na2SO4), and concentrated
under reduced pressure. Flash column chromatography on silica
gel (200 g) using 10 f 30% EtOAc in hexane afforded 10 (2.43 g,
1
44%) as a white foam. H NMR (300 MHz, CDCl3) δ 0.84 (s,
3H), 1.35-1.80 (m, 6H), 1.95-2.05 (m, 2H), 2.07 (s, 3H), 2.20-
2.30 (m, 2H), 2.31 (s, 3H), 3.26-3.38 (m, 1H), 2.51-2.61 (m,
1H), 2.76 (dd, J ) 16.8, 3.3 Hz, 1H), 4.72 (dd, J ) 9.0, 7.8 Hz,
1H), 7.26 (dd, J ) 8.7, 2.7 Hz, 1H), 7.44 (d, J ) 8.7 Hz, 1H), 7.75
(d, J ) 2.7 Hz, 1H); 13C NMR (75 MHz, CDCl3) δ 11.5, 20.6,
20.7, 22.6, 24.9, 27.1, 36.0, 39.1, 42.3, 43.3, 49.2, 81.7, 119.5,
126.4, 126.6, 133.2, 143.8, 149.0, 168.8, 170.4, 196.0.
6-Oxo-3,17â-estradiol (11). To a stirred solution of 10 (2.40 g,
6.50 mmol) in MeOH (100 mL) and THF (50 mL) at room
temperature was added 2 N KOH (20 mL). After stirring for 4 h,

Estrogen Receptor Modulator in Hippocampal Neurons
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18 4479
12H), 1.00-1.58 (m, 26H), 1.58-1.66 (m, 2H), 1.72-1.78 (m, 1H),
1.90 (d, J ) 12.5 Hz, 1H), 2.09-2.18 (m, 1H), 2.26-2.35 (m,
2H), 2.72 (d, J ) 16.9 Hz, 1H), 2.86 (dd, J ) 16.9, 4.8 Hz, 1H),
3.76 (t, J ) 8.5 Hz, 1H), 5.24 (br s, 1H), 6.56 (s, 1H), 6.63 (d, J
) 8.5 Hz, 1H), 7.14 (d, J ) 8.5 Hz, 1H); 13C NMR (125 MHz;
CDCl3) δ 11.3, 19.9, 22.8, 22.9, 24.7, 25.0, 25.7, 26.0, 27.4, 28.2,
30.7, 33.0, 33.3, 34.7, 37.0, 37.4, 37.6, 37.7, 38.2, 39.5, 42.1, 43.6,
46.6, 82.3, 113.0, 116.3, 127.2, 132.0, 137.3, 153.6; HRMS (ESI)
calcd for C34H56O2[M + Na]+, 519.4178; found, 519.4182. Anal.
(C34H56O2‚0.25 H2O) C, H.
fitness score of 74.73, computed from individual scores of, 6.16,
62.67, 0.00, and -17.60 for Shb_ext, Svdw_ext, Shb_int, and Svdw_int
,
respectively.
ER Competitive Binding. The ER binding affinity and selectiv-
ity of 2 and control compounds were determined by a fluorescent
polarization competitive binding assay using purified baculovirus-
expressed human ERR or ERâ and a fluorescent estrogen ligand
EL Red (Invitrogen, Calsbad, CA). Test compounds were serially
diluted to a 2× concentration in assay buffer (20 µM to 200 pM).
A total of 40 µL of assay buffer mixed with 2× test compounds
was added to a 384-well nonbinding surface black microplate
(Corning Life Sciences, Acton, MA), followed by the addition of
40 µL of preincubated 2× complex of ERR (30 nM) or ERâ (60
nM) and EL Red (2 nM) for a final volume of 80 µL. Negative
controls containing ER and EL Red (equivalent to 0% inhibition)
and positive controls containing only free EL Red (equivalent to
100% inhibition) were included. After a 6-hour incubation period
at room temperature, the polarization values were measured using
a GENios Pro microplate reader (Tecan, Mannedorf/Zurich, Swit-
zerland) at excitation/emission 535/590 nm and plotted against the
logarithm of the test compound concentration. The ER/EL Red
complex has a high polarization value yielded from a slow
stumbling due to its large size. If the test compound competes for
estrogen binding, it displaces the EL Red from the ER/EL Red
complex and causes a reduced polarization value. A noncompetitor
does not displace the EL Red from the ER/EL Red complex, and
the polarization value remains high. IC50 value, the concentration
of the test compound that displaces half of the EL Red from ER,
was determined from the plot using a nonlinear least-squares
analysis using GraphPad Prism, version 4.03 (GraphPad Software,
San Diego, CA).
Rat Primary Hippocampal Neuronal Cultures. The use of
animals was approved by the Institutional Animal Care and Use
Committee (IACUC) at the University of Southern California.
Primary cultures of rat hippocampal neurons were prepared
according to the method described previously.57 Briefly, hippocampi
were dissected from the brains of fetuses derived from embryonic
day 18 Sprague-Dawley rats (Harlan Sprague Dawley, Indianapo-
lis, IN) treated with 0.02% trypsin in Hank’s balanced salt solution
(137 mM NaCl, 5.4 mM KCl, 0.4 mM KH2PO4, 0.34 mM Na2-
HPO4‚7H2O, 10.0 mM glucose, and 10.0 mM HEPES) at 37 °C
for 5 min and dissociated by repeated passage through a series of
fire-polished constricted Pasteur pipettes. Approximately 106 cells/
mL were seeded onto poly-D-lysine-coated solid black and clear
bottom 96-well culture plates for neuroprotection and 60 mm petri
dishes for Western immunoblotting analyses. Cells were grown in
phenol-red free Neurobasal medium (NBM, Invitrogen), supple-
mented with B27, 5 U/mL penicillin, 5 µg/mL streptomycin, 0.5
mM glutamine, and 25 µM glutamate at 37 °C in 10% CO2 for the
first 3 d and NBM without glutamate afterward. Cultures grown in
serum-free NBM yield approximately 99.5% neurons and 0.5%
glial cells.
Computer Modeling. All calculations were performed on a SGI
Octane graphical workstation equipped with the IRIX 6.5 operating
system (Silicon Graphics Inc., Mountain View, CA), using a
comprehensive suite of molecular modeling and simulation program,
InsightII 2000 (Accelrys Inc., San Diego, CA), and an automated
ligand docking program, GOLD 3.0 (Genetic Optimization for
Ligand Docking), distributed by CCDC (Cambridge Crystal-
lographic Data Center). InsightII provides a 3D molecular graphics
interface integrated with sets of modeling tools for building,
visualizing, and analyzing molecular structures. GOLD provides a
docking tool for predicting the binding modes of small molecules
into protein binding sites, and it has been highly regarded for its
accuracy and reliability.55,56 Using a genetic algorithm (GA), GOLD
explores a full range of conformational flexibility of the ligand and
partial flexibility of the protein. In the present study, compounds 1
and 2 were first built and energy minimized with the Build module
in InsightII, followed by docking analyses using GOLD. The 3D
structural coordinate data of hERR LBD with a defined ligand
docking site, which was used as the target protein in the docking
runs, were derived from a template crystal structure available in
the Protein Data Bank (PDB entry: 1HJ1).18 1HJ1 contains the
3D coordinate data, resolved by X-ray diffraction, of rat ERâ LBD
complexed with 3, a structural analog of 1.18 Briefly, the 3D
structures of the LBDs of hERR (PDB entry: 1ERR)22 and rERâ
in complex with 3,18 were superimposed based upon the R-atoms
in the backbone chains of two receptors. rERâ was removed from
the complex and the remaining hERR and 3 were associated as a
new assembly. Following removal of water molecules included in
hERR and hydrogenation at a pH of 7.0, the resultant complex was
energetically optimized with the Discover module in InsightII under
a consistent-valence forcefield. Then, based on the binding position
of 3 in hERR, a ligand docking site that includes all atoms within
a probe radius of 15 Å centered at the coordinate of a carbon atom
located at the middle of the 7R-side chain in 3, was defined. Fifty
GA runs were carried out with standard default settings, a population
size of 100, a maximum of 100 000 operations, and a mutation
and crossover of 95. GOLD offers a choice of fitness functions:
GoldScore and ChemScore. GoldScore, which has been optimized
for the prediction of ligand binding positions, was used in the
present study. GoldScore is made up of four components and
quantitated by the following equation: f ) Shb_ext + Svdw_ext + Shb_int
+ Svdw_int, where Shb_ext is the protein-ligand hydrogen bonding
score and Shb_int is the internal hydrogen bonding of the ligand.
Usually, the best result is obtained by letting the internal hydrogen
bonding tend to zero. Svdw_ext and Svdw_int are the scores arising from
weak van der Waals forces. Svdw_ext is multiplied by a factor of
1.375 when the total fitness score is computed, which is an empirical
correction to encourage protein-ligand hydrophobic contact. Three
top-ranked solutions were collected and analyzed in the context of
hERR in InsightII. Taking into account the importance of the
hydrogen bonding interactions between the polar residues, glu353
and his524, along the ER binding site, and the hydroxyl groups
attached to the core structural moiety in ER ligands for a strong
binding, as revealed in our previous analyses,21 solutions with the
highest Svdw_ext were specified as the final bound conformations
for both 1 and 2. In detail, the bound conformation for 1 has a
fitness score of 79.79, computed from individual scores of 6.95,
Glutamate Exposure. Primary hippocampal neurons grown on
solid black and clear bottom 96-well culture plates for 7 d in vitro
(DIV) were pretreated with vehicle alone or 2 at serially diluted
concentrations (1, 10, 100, and 1000 nM) for 48 h, followed by
exposure to 200 µM glutamate at room temperature for 5 min in
HEPES-buffered saline solution (HBS, containing 100 mM NaCl,
2.0 mM KCl, 2.5 mM CaCl2, 1.0 mM MgSO4, 1.0 mM NaH2PO4,
4.2 mM NaHCO3, 10.0 mM glucose, and 12.5 mM HEPES).
Immediately following glutamate exposure, cultures were washed
once with HBS and replaced with fresh NBM containing vehicle
alone or 2. Cultures were returned to the culture incubator and
allowed to incubate for an additional 24 h prior to neuronal viability
measurements on the following day.
Neuronal Viability. LDH Release Measurement: LDH is a
stable cytoplasmic enzyme present in all cells including neurons.
It is rapidly released into the cell culture supernatant when the cell
plasma membrane is damaged. Thus, the LDH level in the culture
medium is a reliable biochemical index for neuronal plasma
membrane damage. In this study, LDH released from the cytosol
61.48, 0.00, and -11.69 for each of four components, Shb_ext, Svdw_ext
,
Shb_int, and Svdw_int, included in the fitness function, GoldScore.
Compound 2 binds to hERR in an analogous mode to 1, with a

4480 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18
Zhao et al.
of damaged neurons into the culture medium following glutamate
exposure was measured using a CytoTox-ONE homogeneous
membrane integrity assay (Promega Corp., Madison, WI), which
determines the LDH activity in the culture medium to enzymaticly
convert the lactate and NAD+ to pyruvate and NADH. In the
enzymatic reaction, the substrate resazurin is reduced to fluorescent
resorufin in the presence of diaphorase, thereby allowing a
fluorometric detection and the signal detected is proportional to
the amount of LDH. Briefly, 40 µL of culture medium from each
well of the culture plate was transferred to a 384-well nonbinding
surface black microplate (Corning Life Sciences) and 40 µL of the
assay reagent was added to incubate at room temperature for 10
min. The fluorescent signal was measured on a SpectraMax
microplate spectrofluorometer (Molecular Devices, Sunnyvale, CA)
at excitation/emission 560/590 nm.
Calcein AM Staining: Calcein AM (Molecular Probes, Eugene,
OR) is a fluorogenic esterase substrate that enters live cells through
permeability and is enzymaticly hydrolyzed to form the polyanionic
dye calcein, which is well-retained within live cells due to the intact
plasma membranes and produces an intense uniform green fluo-
rescence at 530 nm. Thus, calcein AM serves as a reliable indicator
of intact and metabolically viable cells in the cultures. Following
glutamate exposure, after the culture medium was taken out for
LDH release measurement, cultures were rinsed with warm PBS
once and incubated with 1 µM calcein AM in PBS at room
temperature for 30 min. The fluorescent signal was measured on a
SpectraMax microplate spectrofluorometer (Molecular Devices) at
excitation/emission 485/530 nm.
(Vector Laboratories). Relative intensities of the immunoreactive
bands were quantified by optical density analysis using an image
digitizing software, Un-Scan-It version 5.1 (Silk Scientific, Orem,
UT).
Statistical Analyses. Data are presented as group means ( SEM.
Statistically significant differences were determined by a one-way
analysis of variance (ANOVA) followed by a Student-Newman-
Keuls post hoc analysis.
Acknowledgment. This work was supported by grants from
National Institute of Aging (P01 AG14751-10, Project 2), the
Kenneth T. and Eileen L. Norris Foundation, L. K. Whittier
Foundation, Stanley Family Trust, and the NIMH Chemical
Synthesis and Drug Supply Program to R.D.B. The authors also
wish to thank Dr. Anantha Reddy for helpful discussions
regarding the steroid side chain synthesis and Dr. Jason Burgess
for assistance with NMR spectroscopy.
Supporting Information Available: NMR spectra (1H, 13C,
gHSQC, gHMBC, gCOSY, and ROESY), HRMS, HPLC, and
elemental analysis of 2. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) Fillit, H. M. The pharmacoeconomics of Alzheimer’s disease. Am.
J. Manage. Care 2000, 6, S1139-S1144.
(2) Brookmeyer, R.; Gray, S.; Kawas, C. Projections of Alzheimer’s
disease in the United States and the public health impact of delaying
disease onset. Am. J. Public Health 1998, 88, 1337-1342.
(3) Gao, S.; Hendrie, H. C.; Hall, K. S.; Hui, S. The relationship between
age, sex, and the incidence of dementia and Alzheimer disease: A
meta-analysis. Arch. Gen. Psychiatry 1998, 55, 809-815.
(4) Zandi, P. P.; Carlson, M. C.; Plassman, B. L.; Welsh-Bohmer, K.
A.; Mayer, L. S.; Steffens, D. C.; Breitner, J. C. Hormone replacement
therapy and incidence of Alzheimer disease in older women: The
Cache County Study. JAMA, J. Am. Med. Assoc. 2002, 288, 2123-
2129.
Western Immunoblotting. Primary hippocampal neurons grown
on poly-D-lysine-coated culture dishes for 7 DIV were treated with
vehicle alone or test compounds for 30 min (for analyses on pERK2/
ERK2 and pAkt/Akt expression) or 48 h (for analyses on Bcl-2,
Bcl-XL, spinophilin, and â-tubulin expression). Cultures were
washed with cold PBS once and scraped off the dish in 1 mL PBS.
Cells were then centrifuged at 5000 rpm for 5 min, and the pellets
were dissolved in the RIPA lysis buffer (PBS, 1% Triton, 0.2%
SDS, and protease and phosphatase inhibitor cocktail set I (Cal-
BioChem, San Diego, CA)) and suspended by passage through a
200 µL pipet tip. Following incubation at 4 °C for 45 min, the
samples were centrifuged at 12 000 rpm for 10 min, and the
supernatants were collected as the whole cell protein extracts.
Protein concentration was determined by the Bicinchoninic Acid
method. A total of 20 µg of protein samples were diluted in 15 µL
2× SDS containing sample buffer and the final volume was made
30 µL with water. After denaturation on a hot plate at 95-100 °C
for 5 min, 20 µL of the mixture were loaded per lane on 10% SDS-
polyacrylamide mini-gels followed by electrophoresis at 90V. A
high-range Precision Protein Standard (Bio-Rad Laboratories,
Hercules, CA) was used to determine the protein size. The proteins
were then electrotransferred to polyvinylidene difluoride membranes
(Millipore Corp., Bedford, MA) from the gels. Nonspecific binding
sites were blocked with 5% nonfat dry milk in PBS containing
0.05% Tween-20 (PBS-Tween). Membranes were incubated with
(5) Neurgaren, B. L.; Kraines, R. J. Menopausal symptoms in women
of various ages. Psychosom. Med. 1965, 27, 266-273.
(6) Sherwin, B. B., Impact of changing hormonal mileu on psychological
functioning. In Treatment of the postmenopausal woman: Basic and
clinical aspects. Lobo, R. A., Ed.; Lippincott Williams and Wilkins:
Philadelphia, 1999; pp 179-188.
(7) Brinton, R. D. A women’s health issue: Alzheimer’s disease and
strategies for maintaining cognitive health. Int. J. Fertil. Women’s
Med. 1999, 44, 174-185.
(8) Fillit, H. M.; O’Connell, A. W.; Refolo, L. M. Strategies for drug
discovery for cognitive aging and Alzheimer’s disease. J. Mol.
Neurosci. 2002, 19, 1-3.
(9) Henderson, V. W. Estrogen replacement therapy for the prevention
and treatment of Alzheimer’s disease. CNS Drugs 1997, 8, 343-
351.
(10) Zhao, L.; O’Neill, K.; Brinton, R. D. Selective estrogen receptor
modulators (SERMs) for the brain: Current status and remaining
challenges for developing NeuroSERMs. Brain Res. ReV. 2005, 49,
472-493.
(11) Brinton, R. D. Estrogen therapy for prevention of Alzheimer’s disease
not for rehabilitation following onset of disease: the healthy cell
bias of estrogen action. In Synaptic Plasticity: Basic Mechanisms
to Clinical Applications. Baudry, M., Bi, R., Schreiber, S. S., Ed.;
Taylor & Francis Group: New York, 2004; pp 131-157.
(12) Chen, S.; Nilsen, J.; Brinton, R. D. Dose and temporal pattern of
estrogen exposure determines neuroprotective outcome in hippoc-
ampal neuons: therapeutic implications. Endocrinology 2006, 147,
5303-5313.
(13) Hammond, C. B. Women’s concerns with hormone replacement
therapy-compliance issues. Fertil. Steril. 1994, 62, 157S-160S.
(14) Wassertheil-Smoller, S.; Hendrix, S. L.; Limacher, M.; Heiss, G.;
Kooperberg, C.; Baird, A.; Kotchen, T.; Curb, J. D.; Black, H.;
Rossouw, J. E.; Aragaki, A.; Safford, M.; Stein, E.; Laowattana, S.;
Mysiw, W. J. Effect of estrogen plus progestin on stroke in
postmenopausal women: The Women’s Health Initiative: A random-
ized trial. JAMA, J. Am. Med. Assoc. 2003, 289, 2673-2684.
(15) Manson, J. E.; Hsia, J.; Johnson, K. C.; Rossouw, J. E.; Assaf, A.
R.; Lasser, N. L.; Trevisan, M.; Black, H. R.; Heckbert, S. R.;
Detrano, R.; Strickland, O. L.; Wong, N. D.; Crouse, J. R.; Stein,
E.; Cushman, M. Estrogen plus progestin and the risk of coronary
heart disease. N. Engl. J. Med. 2003, 349, 523-534.
primary antibodies directed against phospho-ERK1/2 (pTpY185/187
,
1:1000 dilution, polyclonal, Biosource International, Camarillo,
CA), phospho-Akt (pS,473 1:2000 dilution, monoclonal, Cell Signal-
ing Technology, Beverly, MA), Bcl-2 (1:250 dilution, monoclonal,
Zymed Laboratories, San Francisco, CA), Bcl-XL (1:500 dilution,
monoclonal, Zymed Laboratories), or spinophilin (1:500 dilution,
polyclonal, Upstate, Charlottesville, VA) in PBS-Tween at 4 °C
overnight and then incubated with the horseradish peroxidase
(HRP)-conjugated secondary antibodies (Vector Laboratories, Bur-
lingame, CA) at room temperature for 2 h. After scanning,
membranes were stripped and reprobed with the primary antibodies
against ERK2 (1:5000 dilution, polyclonal, Santa Cruz Biotech-
nology), Akt (1:1,000 dilution, polyclonal, Cell Signaling Technol-
ogy), or â-tubulin (1:5,000 dilution, polyclonal, Abcam, Cambridge,
MA) at room temperature for 2 h, followed by incubation with the
HRP-conjugated secondary antibodies (Vector Laboratories) at room
temperature for 2 h. Immunoreactive bands were detected by a
3′,3′,5′,5′-tetramethyl-benzidine (TMB) peroxidase substrate kit

Estrogen Receptor Modulator in Hippocampal Neurons
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 18 4481
(16) Zhao, L.; O’Neill, K.; Brinton, R. D. Estrogenic agonist activity of
ICI 182,780 (Faslodex) in hippocampal neurons: implications for
basic science understanding of estrogen signaling and development
of estrogen modulators with a dual therapeutic profile. J. Pharmacol.
Exp. Ther. 2006, 319, 1124-1132.
(39) Blazejewski, J. C.; Wilmshurst, M. P.; Popkin, M. D.; Wakselman,
C.; Laurent, G.; Nonclercq, D.; Cleeren, A.; Ma, Y.; Seo, H. S.;
Leclercq, G. Synthesis, characterization and biological evaluation of
7R-perfluoroalkylestradiol derivatives. Bioorg. Med. Chem. 2003, 11,
335-345.
(40) Jiang, X. R.; Sowell, J. W.; Zhu, B. T. Synthesis of 7R-substituted
derivatives of 17â-estradiol. Steroids 2006, 71, 334-342.
(41) Suga, T.; Ohta, S.; Nakai, A.; Munesada, K. Glycinoprenols: novel
polyprenols possessing a phytyl residue from the leaves of soybean.
J. Org. Chem. 1989, 54, 3390-3393.
(42) Matsubara, S.; Takai, K.; Nozaki, H. Baeyer-Villiger oxidation with
Me3SiOOSiMe3 under assistance of SnCl4 or BF3‚OEt2. Bull. Chem.
Soc. Jpn. 1983, 56, 2029-2032.
(43) Akanni, A. O.; Marples, B. A. Improved preparation of 3,17â-
diacetoxyestra-1,3,5(10)-trien-6-one. Syn. Commun. 1984, 14, 713-
715.
(44) Mitra, K.; Marquis, J. C.; Hillier, S. M.; Rye, P. T.; Zayas, B.; Lee,
A. S.; Essigmann, J. M.; Croy, R. G. A rationally designed genotoxin
that selectively destroys estrogen receptor-positive breast cancer cells.
J. Am. Chem. Soc. 2002, 124, 1862-1863.
(45) Kuiper, G. G.; Carlsson, B.; Grandien, K.; Enmark, E.; Haggblad,
J.; Nilsson, S.; Gustafsson, J. A. Comparison of the ligand binding
specificity and transcript tissue distribution of estrogen receptors alpha
and beta. Endocrinology 1997, 138, 863-870.
(17) Wakeling, A. E.; Dukes, M.; Bowler, J. A potent specific pure
antiestrogen with clinical potential. Cancer Res. 1991, 51, 3867-
3873.
(18) Pike, A. C.; Brzozowski, A. M.; Walton, J.; Hubbard, R. E.; Thorsell,
A. G.; Li, Y. L.; Gustafsson, J. A.; Carlquist, M. Structural insights
into the mode of action of a pure antiestrogen. Structure 2001, 9,
145-153.
(19) Howell, A.; Osborne, C. K.; Morris, C.; Wakeling, A. E. ICI 182,-
780 (Faslodex): Development of a novel, “pure” antiestrogen. Cancer
2000, 89, 817-825.
(20) Bross, P. F.; Cohen, M. H.; Williams, G. A.; Pazdur, R. FDA drug
approval summaries: fulvestrant. Oncologist 2002, 7, 477-480.
(21) Zhao, L.; Brinton, R. D. Structure-based virtual screening for plant-
based ERâ-selective ligands as potential preventative therapy against
age-related neurodegenerative diseases. J. Med. Chem. 2005, 48,
3463-3466.
(22) Brzozowski, A. M.; Pike, A. C.; Dauter, Z.; Hubbard, R. E.; Bonn,
T.; Engstrom, O.; Ohman, L.; Greene, G. L.; Gustafsson, J. A.;
Carlquist, M. Molecular basis of agonism and antagonism in the
oestrogen receptor. Nature 1997, 389, 753-758.
(23) Fawell, S. E.; Lees, J. A.; White, R.; Parker, M. G. Characterization
and colocalization of steroid binding and dimerization activities in
the mouse estrogen receptor. Cell 1990, 60, 953-962.
(24) Navarro, A.; Gomez, C.; Sanchez-Pino, M. J.; Gonzalez, H.; Bandez,
M. J.; Boveris, A. D.; Boveris, A. Vitamin E at high doses improves
survival, neurological performance, and brain mitochondrial function
in aging male mice. Am. J. Physiol. 2005, 289, R1392-1399.
(25) Lipinski, C. A.; Dominy, B. W.; Feeney, P. J. Experimental and
computational approaches to estimate solubility and permeability in
drug discovery and development settings. AdV. Drug DeliVery ReV.
1997, 23, 3-25.
(46) Zhao, L.; Brinton, R. D. Select estrogens within the complex
formulation of conjugated equine estrogens (Premarin) are protective
against neurodegenerative insults: implications for a composition
of estrogen therapy to promote neuronal function and prevent
Alzheimer’s disease. BMC Neurosci. 2006, 7, 24.
(47) Wu, T.-W.; Wang, J. M.; Chen, S.; Brinton, R. D. 17â-Estradiol
induced Ca2+ influx via L-type calcium channels activates the Src/
ERK/cyclic-AMP response element binding protein signal pathway
and BCL-2 expresion in rat hippocanpal neurons: A potential
initiation mechanism for estrogen-induced neuroprotection. Neuro-
science 2005, 135, 59-72.
(48) Zhao, L.; Chen, S.; Wang, J. M.; Brinton, R. D. 17â-Estradiol induces
Ca2+ influx, dendritic and nuclear Ca2+ rise and subsequent cyclic
AMP response element-binding protein activation in hippocampal
neurons: a potential initiation mechanism for estrogen neurotrophism.
Neuroscience 2005, 132, 299-311.
(49) Mannella, P.; Brinton, R. D. Estrogen receptor protein interaction
with phosphatidylinositol 3-kinase leads to activation of phospho-
rylated Akt and extracellular signal-regulated kinase 1/2 in the same
population of cortical neurons: A unified mechanism of estrogen
action. J. Neurosci. 2006, 26, 9439-9447.
(50) Singh, M. Ovarian hormones elicit phosphorylation of Akt and
extracellular-signal regulated kinase in explants of the cerebral cortex.
Endocr. J. 2001, 14, 407-415.
(51) Znamensky, V.; Akama, K. T.; McEwen, B. S.; Milner, T. A.
Estrogen levels regulate the subcellular distribution of phosphorylated
Akt in hippocampal CA1 dendrites. J. Neurosci. 2003, 23, 2340-
2347.
(52) Datta, S. R.; Dudek, H.; Tao, X.; Masters, S.; Fu, H.; Gotoh, Y.;
Greenberg, M. E. Akt phosphorylation of BAD couples survival
signals to the cell-intrinsic death machinery. Cell 1997, 91, 231-
241.
(53) Nilsen, J.; Brinton, R. D. Mechanism of estrogen-mediated neuro-
protection: regulation of mitochondial calcium and Bcl-2 expression.
Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 2842-2847.
(26) Abbott, N. J.; Ronnback, L.; Hansson, E. Astrocyte-endothelial
interactions at the blood-brain barrier. Nat. ReV. Neurosci. 2006, 7,
41-53.
(27) Habgood, M. D.; Begley, D. J.; Abbott, N. J. Determinants of passive
drug entry into the central nervous system. Cell. Mol. Neurobiol.
2000, 20, 231-253.
(28) Atkinson, F.; Cole, S.; Green, C.; van de Waterbeemd, H. Lipophi-
licity and other parameters affecting brain penetration. Curr. Med.
Chem. 2002, 2, 229-240.
(29) Iyer, M.; Mishru, R.; Han, Y.; Hopfinger, A. J. Predicting blood-
brain barrier partitioning of organic molecules using membrane-
interaction QSAR analysis. Pharm. Res. 2002, 19, 1611-1621.
(30) Ma, X. L.; Chen, C.; Yang, J. Predictive model of blood-brain barrier
penetration of organic compounds. Acta Pharmacol. Sin. 2005, 26,
500-512.
(31) Loscher, W.; Potschka, H. Role of drug efflux transporters in the
brain for drug disposition and treatment of brain diseases. Prog.
Neurobiol. 2005, 76, 22-76.
(32) Misra, A.; Ganesh, S.; Shahiwala, A.; Shah, S. P. Drug delivery to
the central nervous system: A review. J. Pharm. Sci. 2003, 6, 252-
273.
(33) Bowler, J.; Lilley, T. J.; Pittam, J. D.; Wakeling, A. E. Novel steroidal
pure antiestrogens. Steroids 1989, 54, 71-99.
(34) French, A. N.; Wilson, S. R.; Welch, M. J.; Katzenellenbogen, J. A.
A synthesis of 7 alpha-substituted estradiols: Synthesis and biological
evaluation of a 7 alpha-pentyl-substituted BODIPY fluorescent
conjugate and a fluorine-18-labeled 7 alpha-pentylestradiol analog.
Steroids 1993, 58, 157-169.
(35) Tedesco, R.; Katzenellenbogen, J. A.; Napolitano, E. An expeditious
route to 7a-substituted estradiol derivatives. Tetrahedron Lett. 1997,
38, 7997-8000.
(36) Adamczyk, M.; Johnson, D. D.; Reddy, R. E. A stereoselective
synthesis of 7 alpha-(3-carboxypropyl) estradiol from a noncontrolled
substance. Steroids 1997, 62, 771-775.
(37) Skaddan, M. B.; Wust, F. R.; Katzenellenbogen, J. A. Synthesis and
binding affinities of novel recontaining 7R-substituted estradiol
complexes: Models for breast cancer imaging agents. J. Org. Chem.
1999, 64, 8108-8121.
(38) Hussey, S. L.; He, E.; Peterson, B. R. Synthesis of chimeric 7R-
substituted estradiol derivatives linked to cholesterol and cholesteryl-
amine. Org. Lett. 2002, 4, 415-418.
(54) Green, P. S.; Yang, S. H.; Simpkins, J. W. Neuroprotective effects
of phenolic A ring oestrogens. NoVartis Found. Symp. 2000, 230,
202-213.
(55) Kellenberger, E.; Rodrigo, J.; Muller, P.; Rognan, D. Comparative
evaluation of eight docking tools for docking and virtual screening
accuracy. Proteins 2004, 57, 225-242.
(56) Kontoyianni, M.; McClellan, L. M.; Sokol, G. S. Evaluation of
docking performance: Comparative data on docking algorithms. J.
Med. Chem. 2004, 47, 558-565.
(57) Zhao, L.; Wu, T.-W.; Brinton, R. D. Estrogen receptor subtypes alpha
and beta contribute to neuroprotection and increased Bcl-2 expression
in primary hippocampal neurons. Brain Res. 2004, 1010, 22-34.
JM070546X