Novel dehydroepiandrosterone derivatives with antiapoptotic, neuroprotective activity

J. Med. Chem. 2009, 52, 6569–6587 6569

DOI: 10.1021/jm900468p

Novel Dehydroepiandrosterone Derivatives with Antiapoptotic, Neuroprotective Activity

Theodora Calogeropoulou,*^,†Nicolaos Avlonitis,^†Vassilios Minas,^‡Xanthippi Alexi,^§Athanasia Pantzou,^†

Ioannis Charalampopoulos,^‡Maria Zervou,^†Varvara Vergou,^‡Efrosini S. Katsanou,^§Iakovos Lazaridis,^‡

Michael N. Alexis,^§and Achille Gravanis^‡

^†Institute of Organic and Pharmaceutical Chemistry and ^§Institute of Biological Research and Biotechnology, National Hellenic Research

Foundation, 48 Vassileos Constantinou Avenue 11635, Athens, Greece, and ^‡Department of Pharmacology, School of Medicine,

University of Crete, Heraklion 71110, Greece

Received April 11, 2009

DHEA analogues with modifications at positions C3 or C17 were synthesized and evaluated for

neuroprotective activity against the neural-crest-derived PC12 cell model of serum deprivation-induced

apoptosis. The most potent compounds were the spiro-epoxy derivatives 17β-spiro[5-androstene-17,

2⁰-oxiran]-3β-ol (20), (20S)-3β,21-dihydroxy-17β,20-epoxy-5-pregnene (23), and (20R)-3β,21-dihydroxy-

17R,20-epoxy-5-pregnene (27) with IC₅₀values of 0.19 ( 0.01, 99.0 ( 4.6, and 6.4 ( 0.3 nM, respectively.

Analogues 20, 23, and 27, up to the micromolar range of concentrations, were unable to activate estrogen

receptor R and β (ERR and ERβ) or to interfere with ER-dependent gene expression significantly. In

addition, they were unable to stimulate the growth of Ishikawa, MCF-7, and LNCaP cells. Our results

suggest that the spiro-epoxyneurosteroid derivatives 20, 23, and 27 may prove to be lead molecules for the

synthesis of novel neuroprotective agents.

Introduction

stroke.⁶Neuroprotective effects include recovery of motor

function after spinal cord injury,⁷regulation of myelination,⁸

proliferation of neuronal stem cells,^9,8fneurogenesis in the

hippocampus,^8f,9,10and induction of analgesia.¹¹Dehydro-

epiandrosterone (DHEA)^a(Figure 1) and its sulfate ester

(DHEAS) are produced in the brain as well as the adrenals,^12,13

with brain DHEA acting locally in a paracrine fashion and

adrenal DHEA having a systemic role. DHEA is one of the

most potent neuroactive neurosteroids, and the precipitous

decline of both brain and circulating DHEA with advancing

age has been associated to neuronal dysfunction and degene-

ration.¹⁴It is now experimentally supported that DHEA

protects neurons from anoxia,¹⁵that it transiently increases

NGF mRNA level,¹⁶that it exhibits neuroprotective proper-

ties in various types of CNS insults,¹⁷and that it induces a

protective effect against spinal cord injury.¹⁸

Recent studies investigating the physiopathological signifi-

cance of neurosteroids in Alzheimer’s disease (AD) have shown

a significant decline of neurosteroid concentrations in indivi-

dual brain regions of AD patients compared to age-matched

nondemented controls.^19,20Additionally, a significant negative

correlation was found between the levels of phosphorylated

tau proteins and DHEAS in the hypothalamus.^20,21These

studies strongly suggest a possible neuroprotective role for

Neuronal cell death by apoptosis is the “end point”of many

human neurological disorders, including Alzheimer’s, Parkin-

son’s, and Huntington’s diseases, stroke/trauma, and multiple

and amyotrophic lateral sclerosis.^1,2Apoptotic death of hip-

pocampal and cortical neurons is responsible for the symp-

toms of Alzheimer’s disease, while death of midbrain neurons

that use the neurotransmitter dopamine underlies Parkinson’s

disease. Huntington’s disease involves the death of neurons in

the striatum, which control body movements, and death of

lower motor neurons manifests as amyotrophic lateral sclero-

sis. Additionally, brain ischemia and trauma induce necrosis

of a small brain area, which then propagates neuronal cell loss

by apoptosis to a larger brain area, due to the neurotoxic

material released by the necrotic cells. Apoptotic neuronal cell

loss is also observed in the aging brain, as a physiological

process.¹Currently, there is little or no treatment for most

neurodegenerative diseases. At best, available treatments are

symptomatic in nature and do not prevent or slow the

progression of disease.³A fundamental approach for reduc-

ing the burden of neurodegenerative diseases is to slow or

halt progression and, ultimately, to prevent the onset of the

disease process. Thus, strategies for neuroprotection, prevent-

ing apoptotic neuronal cell loss may offer new therapeutic

interventions.

^aAbbreviations: DHEA, dehydroepiandrosterone; DHEAS, dehydro-

epiandrosterone sulfate ester; ERR, estrogen receptor R; ERβ, estrogen

receptor β; E2, estradiol; AD, Alzheimer’s disease; MTT, 3-(4,5-di-

methylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; CSA, camphor-10-

sulfonic acid; PCC, pyridinium chlorochromate; BSA, bovine serum

albumin; PBS, phosphate-buffered saline; AlkP, alkaline phosphatase;

mDBS, membrane DHEA-binding site(s); NGF, nerve growth factor;

MPTP, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; NF-κB, nuclear

factor κB; PMSF, phenylmethanesulfonyl fluoride; FBS, fetal bovine

serum; CNS, central nervous system.

Neurosteroids are synthesized in the central and peripheral

nervous system, particularly in myelinating glial cells but also

in astrocytes and neurons. Neuroactive steroids, which act on

the nervous system,⁴affect neuronal function and differentia-

tion⁵and provide for neuroprotection against ischemia and

*To whom correspondence should be addressed: phone, þ 30210 7273833;

fax, þ30210 7273831; e-mail, tcalog@eie.gr.

r

2009 American Chemical Society

Published on Web 10/21/2009

pubs.acs.org/jmc

6570 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

Scheme 1^a

Figure 1. Structures of dehydroepiandrosterone (DHEA) and allo-

pregnanolone.

endogenous neurosteroids in AD. It is also of interest that

DHEA is able to potentiate locomotor activity of hemi-

Parkinsonian monkeys, alleviating impairment symptoms of the

moderately and severely affected MPTP-treated animals.^14b,22

In addition, DHEA affects the catecholaminergic system,

upregulating the expression of tyrosine hydroxylase and, thus,

increasing catecholamine de novo synthesis and release.²³

The adult central nervous system (CNS) is classically

known as a structure with very limited regenerative capacity.

However, several pathological conditions, e.g., ischemia,

epilepsy and trauma, have been shown to upregulate neural

stem cell activity in the subventricular zone and the dentate

gyrus. These findings suggest that signals are present through-

out the adult brain, which allow limited neuronal regeneration

to occur. This fundamental observation changes our view on

neurodegeneration and the brain’s regenerative capacity,²³

giving us the potential ability to regenerate specific brain

areas. The neurosteroids DHEA and allopregnanolone

(Figure 1) have recently been shown to induce neurogenesis

in various experimental models.^9,24

^aReagents and conditions: (a) ethylene glycol, CSA, cyclohexane, 80 °C;

(b) Dess-Martin periodinane, DCM, room temperature; (c) n-BuLi, Me₃SI,

THF, -40 °C.

Scheme 2^a

The lack of effective treatment for devastating neuro-

degenerative diseases has stimulated great interest in the deve-

lopment of neuroprotective means that can prevent or treat

progressive loss of neural function. There is a sustained need

for the development of new compounds for neural cell

protection, repair, and rescue by targeting neural cell apop-

tosis and survival or neurogenesis. Natural neurosteroids such

as DHEA possess important neuroprotective and neurogenic

properties in vitro and in vivo, in experimental animals, as

described above. However, naturally occurring neurosteroids

are metabolized in humans into estrogens, androgens, or

progestins which are known to exert important genera-

lized endocrine side effects, including hormone-dependent

neoplasias,^4dthus limiting their clinical use.

^aReagents and conditions: (a) MeONa, MeOH, reflux; (b) KCN,

EtOH, reflux; (c) NaN₃, NH₄Cl, MeOH/H₂O (8:1), reflux; (d) HCl/

acetone (1:1), THF, room temperature.

Scheme 3^a

As a continuation of our ongoing studies on neuro-

active steroids^25,26we were interested in obtaining structure-

activity relationships for antiapoptotic-neuroprotective

steroid derivatives. Thus, in the present report we describe

the synthesis and activity of DHEA analogues modified at

positions C3 or C17 with the aim of improving their anti-

apoptotic-neuroprotective activity and inhibiting conversion

to estrogens or androgens. The neuroprotective activity was

evaluated using the neural-crest-derived PC12 cell model,

which has been extensively employed to study neuronal cell

apoptosis and survival.

^aReagents and conditions: (a) n-BuLi, THF, room temperature;

(b) MeLi, THF, room temperature; (c) HCl/acetone (1:1), THF, room

temperature.

the unconjugated ketone 2 in 92% overall yield for the two

steps. Compound 2 has been previously reported in 50% yield

using PCC in the presence of Celite and KOAc.²⁷Corey-

Chaykovsky epoxidation of 2 using trimethylsulfonium iodi-

de and n-BuLi afforded a mixture of the two C3 epimeric

spiro-epoxides, which were separated by flash column chro-

matography to yield the desired epoxide 3. The stereochem-

istry of C-3 of compound 3 and the position of the double

Chemistry

The synthetic strategy followed for the preparation of the

neurosteroid analogues 5, 7, 9, 11, 14, and 15 is depicted in

Schemes 1, 2, and 3. Thus, protection of the 17-keto group of

DHEA using ethylene glycol in the presence of CSA to yield

the ketal 1, followed by oxidation of the 3-hydroxy group in 1

using Dess-Martin periodinane in dichlomethane, afforded

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6571

Scheme 4^a

^aReagents and conditions: (a) TBDPSiCl, imidazole, DMF, 50 °C; (b) LiAlH₄, THF, room temperature; (c) NaH, MeI, THF, room temperature;

(d) TBAF, THF, room temperature.

bond were determined by 1D and 2D NMR spectroscopy

and more specifically from the correlations observed in the

HMBC, COSY, and NOESY spectra (cf.Supporting Infor-

derivative being less polar (higher R_fvalue).²⁸In addition to

their characteristic chromatographic properties the stereo-

isomers were identified by the ¹³C NMR signal of the tertiary

alcohol at C-3, which is lower in parts per million for the

3β-isomer (axial 3R-OH) than the 3R isomer (equatorial

3β-OH), as has been observed previously.²⁸Thus, the C-3

carbon of 10 resonates at 71.9 ppm while its C-3 epimer

resonates at 73.4 ppm (cf. Supporting Information). Finally,

NOE enhancement between the H₂-1⁰and the H₂-2⁰protons

of the 3β-butyl substituent in 10 and protons H-2R, H-2β,

H-4R, and H-4β is in agreement with a 3R-hydroxy-3β-butyl

derivative. Conversely, NOE enhancement between the pro-

tons H₂-1⁰and H-1R, H₂-2⁰and H-6, and H₂-3⁰and H-6 in the

C-3 epimer of 10 supports a 3R-butyl substituent (cf. Support-

ing Information). Deprotection of the carbonyl group at

C-17 in 10 using a mixture of 10% HCl/acetone (1:1) in

THF gave analogue 11 (Scheme 3).

Addition of MeLi to the C-3-keto group of 2 afforded the

two diasteromeric alcohols 12 and 13 in a 1:2 ratio, which were

separated by column chromatography, since the 3R-hydroxy

derivative 13 (axial hydroxyl) is less polar than the 3β-

congener 12 (equatorial hydroxyl).²⁸The position of the double

bond in 12 and 13 is at C-5:C-6 as indicated by the resonance

of the vinylic proton in the ¹H NMR spectra which appears as

a doublet at 5.30 ppm (J = 5.49 Hz) for 12 and at 5.39 ppm

(J = 5.49) for 13. Deprotection of the carbonyl group at C-17

in alcohols 12 and 13, using a mixture of 10% HCl/acetone

(1:1) in THF, afforded analogues 14 and 15, respectively

(Scheme 3). Spectroscopic characterization of compounds

14 and 15 using 1D and 2D NMR spectroscopy was in

agreement with 5-ene derivatives. In addition, the C-3 carbon

of 14 resonates at 72.1 ppm (equatorial hydroxyl) while of its

C-3 epimer 15 resonates at 70.1 ppm (axial hydroxyl).²⁸The

3R-methyl substitution for 14 was also supported by the

observed NOE correlations between the methyl protons

H₃-1⁰and protons H-2R and H-4R. Conversely, the 3β-methyl

substitution for 15 was supported by the observed NOE

correlations between the methyl protons H₃-1⁰and protons

H-2R, H-2β, H-4R, and H-4β (cf. Supporting Information).

The synthesis of the 17β-methoxy derivative 19 is described

in Scheme 4. Thus, protection of the C3-hydroxyl group as the

tert-butyldiphenylsilyl ether²⁹followed by reduction of the

ketone at C17 using LiAlH₄yielded the 17β-hydroxy deriva-

tive 16, which was subsequently alkylated with MeI using

NaH as base, to give the methyl ether 18. Finally, removal of

the silyl protecting group using TBAF afforded analogue 19.

The synthetic route followed for the preparation of deriva-

tives 20 and 23 is depicted in Scheme 5. Epoxidation of the

17-keto group in DHEA using trimethylsulfonium methylide

1

mation). The peak of the vinylic proton in the H NMR

spectrum of 3 appears as a doublet (J = 5.4 Hz) at 5.30 ppm

and shows correlations in the COSY spectrum with the H-7

protons (H-7R at 2.0 ppm and H-7β at 1.58 ppm) and the

H-4R allylic proton at 1.62 ppm. These observations are in

agreement with a C-5:C-6 double bond. The other allylic

proton at 2.88 ppm also exhibited correlations with carbons

C-3, C-10, C-5, C-6, and C-2 and the epoxide methylene which

support its position at C-4. The NOE enhancement between

the epoxide proton at 2.58 ppm and the allylic proton H-4R

(1.62 ppm) and between the epoxide proton at 2.56 ppm and

H-2R at 1.19 ppm are in agreement with a 3β-epoxide.

Additional proof of the stereochemistry of carbon C-3 of

compound 3 was obtained by reaction with LiAlH₄to af-

ford the corresponding 3β-hydroxy-3R-methyl derivative. The

3R-methyl signal in the ¹H NMR spectrum resonates at

1.09 ppm as expected from literature reports and is in accor-

dance with the spectra of compound 12.²⁸Nucleophilic open-

ing of epoxide 3 by NaOMe, NaN₃, or KCN afforded com-

pounds 4, 6, and 8, respectively. The vinylic proton H-6

appears as a doublet at 5.29 ppm (J = 5.49 Hz) in 4 and at

5.40 ppm (J = 5.49 Hz) in 6 and as dd at 5.33 ppm (J = 4.89,

2.46 Hz) in 8, which are in agreement with 5-ene derivatives.

Deprotection of the carbonyl groupatC-17of4, 6, and 8using

a mixture of 10% HCl/acetone (1:1) in THF led to the

neurosteroid analogues 5, 7, and 9, respectively. Use of 2D

NMR spectroscopy in compounds 5, 7, and 9 was in agree-

ment with 3β-hydroxy-substituted 5-ene derivatives. In all

three compounds we observed NOE enhancement between

the H₂-1⁰methylene and the proton H-4R (cf. Supporting

Information).

Addition of n-BuLi to ketone 2 yielded exclusively the

3R-hydroxy-3β-butyl derivative 10, since the bulky n-BuLi

approaches the C-3-carbonyl from the equatorial face of the

steroid scaffold. We did not observe double bond isomeriza-

tion to afford the 4-ene analogue as attested by the peak of the

vinylic proton in the ¹H NMR spectrum of 10, which appears

as a doublet at 5.39 ppm (J = 5.49 Hz) and shows correlations

in the COSY spectrum with the H-7 protons (H-7R at 2.0 ppm

and H-7β at 1.60 ppm) and the H-4 allylic protons and NOE

enhancement with the H-4R proton. Concerning the stereo-

chemistry of C-3 in compound 10 it was deciphered through

the synthesis of both epimers by addition of n-BuMgBr to

ketone 2 (cf. Supporting Information for experimental

procedure). Compound 10 can be differentiated from its C-3

epimer by TLC (silica gel normal phase), the 3β-substituted

6572 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

Scheme 5^a

^aReagents and conditions: (a) t-BuOK, (CH₃)₃SI, DMF, room temperature; (b) vinyl MgBr, THF, -78 °C to room temperature; (c) VO(acac)₂,

t-BuOOH, CH₂Cl₂, -10 °C; (iv) K₂CO₃, MeOH, room temperature.

Scheme 6^a

^aReagents and conditions: (a) t-BuOK, (EtO)₂P(O)CH₂CN, THF, 0 °C to room temperature; (b) DIBAL-H, CH₂Cl₂, -78 °C; (c) NaBH₄,

CeCl₃7H₂O, MeOH, room temperature; (d) K₂CO₃, m-CPBA, CH₂Cl₂, 0 °C.

3

afforded the 17,20-spiro-epoxide 20.³⁰The synthesis of ep-

oxide 23 was effected by addition of vinylmagnesium bromide

to the C17-keto group of DHEA to afford the quaternary

alcohol 21, followed by epoxidation using t-BuOOH in the

presence of vanadyl acetonylacetonate as catalyst, to give a

mixture of the (20R)- and (20S)-20,21-epoxides 22a and 22b

(2:1), respectively. The assignment of the 20R configuration

was based on previously reported analogous systems.³¹Payne

rearrangement of the less hindered epoxide 22a using K₂CO₃

in methanol afforded the desired epoxide 23.³¹Derivative 27

was obtained following the synthetic strategy presented in

Scheme 6. Horner-Emmons olefination of DHEA with

diethyl cyanomethylphosphonate in the presence of t-BuOK

afforded predominantly the (E)-17-cyanomethylene unsatu-

rated nitrile 24 (ratio of 24 and (Z)-17-cyanomethylene nitrile

(5:1)). Several reaction conditions weretriedin ordertoobtain

the E isomer selectively (i.e., different ratios of base and

phosphonate reagent, temperature, or even performing the

reaction on 3-acetyl-DHEA) but resulted either in equimolar

formation of the two geometrical isomers or in favor of the

undesired Z derivative. The previously reported Horner-

Emmons olefination of DHEA using NaH as base in

DME affords also a mixture of the E unsaturated nitrile 24

along with the Z isomer.³²However, compound 24 could

be obtained after purification by column chromatography.

Treatment of 24 with DIBAL-H in dichloromethane at

-78 °C followed by reduction of the resulting aldehyde 25³³

using the Luche methodology³⁴yielded the allylic alcohol 26.

Finally, epoxidation of 26 using m-CPBA in the presence of

K₂CO₃afforded the desired analogue 27 in 40% yield. Epoxi-

dation using t-BuOOH in the presence of a catalytic amount

of vanadyl acetonylacetonate afforded a mixture of the

diastereomeric epoxides 27 and 23 in a 2:1 ratio and in 70%

yield. However, epoxides 27 and 23 could not be separated by

column chromatography.

Results and Discussion

The antiapoptotic activity of the new analogues was eval-

uated using the neural-crest-derived PC12 cell model, which

has been extensively used to study neuronal cell apoptosis and

survival. PC12 cells do not express functional GABA_Aor

NMDA receptors and cannot metabolize DHEA to estrogens

and androgens.^8gCulture of PC12 cells in the absence of

serum results in a strong induction of apoptosis, compared to

cell cultures supplemented with serum. The antiapoptotic

effect of the new compounds at 100 nM was evaluated using

the APOPercentage apoptosis assay (Biocolor Ltd., Belfast,

N. Ireland), and the results are shown in Figure 2.

More specifically, the new 3β-hydroxy-3R-substituted ster-

oid derivatives 5, 7, 9, and 14 were not capable of protecting

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6573

Figure 2. Effect of new neurosteroids against serum deprivation-

induced apoptosis of PC12 cells. Cells were cultured for 24 h either

in complete media (serum) or in serum-free media supplemented

with 1% BSA in the presence of various steroids at 100 nM.

Apoptosis was quantified by the APOPercentage assay, depicted

as percentage of control (serum-free media in the absence of

steroids). Values represent mean ( SEM of at least four indepen-

dent experiments, each performed in triplicate (*, p < 0.05).

Figure 3. Analogues 20, 23, and 27 protected PC12 cells against

serum deprivation-induced apoptosis in a dose-dependent manner.

Cells were cultured for 24 h either in complete media (serum

supplemented) or serum-free media supplemented with 1% BSA

in the presence of 20, 23, or 27 at concentrations ranging from 10^-11

to 10^-5M. Apoptosis was quantified by the APOPercentage assay,

depicted as ratio to control (serum-free media in the absence of

steroids). Values represent mean ( SEM of at least four indepen-

dent experiments, each performed in triplicate.

PC12 cells from apoptosis. This was also observed for the

3R-hydroxy-3β-substituted analogues 11 and 15. Replace-

ment of the C17-ketone by a 17β-methoxy group in 19 again

resulted in loss of antiapoptotic activity. Interestingly, how-

ever, 17,20-spiro-epoxide derivatives 20, 23, and 27 were

capable of protecting neuronal cells against serum depriva-

tion-induced apoptosis, withIC₅₀valuesof 0.19 ( 0.01, 99.0 (

4.6, and 6.4 ( 0.3 nM, respectively (Figure 3). This effect at

concentrations of 100 nM was also confirmed with flow

cytometry (FACS) analysis of annexin V-propidium iodide-

stained PC12 cells, as shown in Figure 4.

derivatives 5,16-androstadiene-17β-cyano-3β-ol, 5-androstene-

3β-ol-16-one, and 5-androstene-16R-bromo-3β-ol-17-one

were also inactive, which was also the case for testosterone

or testosterone:BSA. Surprisingly, the 17β-carboxylic acid

derivative, 5-androsten-3β-ol-17β-carboxylic acid, and the

C17-unsubstituted derivative, 5,16-androsten-3β-ol, exhibi-

ted moderate antiapoptotic activity (Figure 5).

To further our understanding of the role of substitution of

the steroid backbone in the antiapoptotic actions of neuro-

active steroids, we also examined the effect of a series of 24

commercially available Δ⁵and Δ⁴androstenes at 100 nM

concentration (Figure 5). The Δ⁵androstene series comprised

DHEA analogues modified at C3, C7, C10, and C17. More

specifically, replacement of the C3-hydroxy by a 3β-chlorine

atom, inversion of the hydroxyl group at C3, or oxidation

of the 3β-hydroxyl deprived the respective derivatives (3β-

chloro-5-androstene-17-one, 3R-hydroxy-5-androstene-17-one,

and 5-androstene-3,17-dione) from antiapoptotic activity.

Introduction of a 7R-hydroxy or a 7β-hydroxy group in

DHEA (3β,7R-dihydroxy-5-androsten-17-one and 3β,7β-

dihydroxy-5-androsten-17-one, respectively) abolished the

antiapoptotic activity of DHEA. The modifications at C17

included the replacement of the C17-keto group by a N-O-

carboxymethyl oxime, by an ethylene ketal, by an electron-

donating 17β-amino group, by a 17β-hydroxyl group, by a

17β-hydroxy-17R-ethynyl group, or by a 17β-hydroxy-17R-

methyl group (3β-hydroxy-5-androstene-17-carboxymethyl

oxime, 3β-hydroxy-5-androstene-17-ethylene ketal, 17β-amino-

5-androsten-3β-ol, 5-androstene-3β-17β-diol, 5-androstene-

3β-17R-diol, 17R-ethynyl-5-androstene-3β-17β-diol, and 17R-

methyl-5-androstene-3β,17β-diol, respectively) abolished the

antiapoptotic activity. Isomerization of the Δ⁵to the Δ⁴deri-

vative 3β-hydroxy-4-androsten-17-one was detrimental to the

antiapoptotic effect of DHEA, which was also the case for

5-androstene-3β,11β-diol-17-one, 5-androstene-3β,11R-diol-

17-one, and 3β-hydroxy-5-androstene-11,17-dione. DHEA

We further focused our attention on the active synthetic

spiro-epoxides 20, 23, and 27 in an effort to obtain informa-

tion on their mechanism of action in comparison to DHEA.

It is evident from the previously described SAR that the

requirements for neuroprotective activity of DHEA deriva-

tives are very stringent, which may indicate that this effect

is mediated through interaction with a specific receptor or

enzyme. This is in line with our previous findings that

the antiapoptotic effect of DHEA is mediated by specific

G-protein-associated membrane binding sites, independent

of NMDA or GABA_Areceptors.³⁵Indeed, DHEA conju-

gated to bovine serum albumin (BSA), a molecule with no cell

penetrating ability, was found to reverse serum deprivation-

induced apoptosis by almost 50%, and to protect PC12 cells

from apoptosis, with an apparent IC₅₀of 1.5 nM, i.e., in a

manner similar to that of nonconjugated DHEA (1.8 nM).³⁵

Furthermore, DHEA:BSA effectively mimicked DHEA ac-

tions on antiapoptotic Bcl-2 proteins by preventing their

downregulation by serum deprivation. Competition of

[³H]DHEA binding to isolated PC12 cell membranes using

increasing concentrations of radioinert DHEA revealed neuro-

steroid high-affinity binding to membrane DHEA-binding

site(s) (mDBS) with an apparent IC₅₀of 1.65 ( 0.2 nM. We

were pleased to find that the active synthetic spiro-epoxides

20, 23, and 27 displaced [³H]DHEA with apparent IC₅₀values

of 0.04 ( 0.002, 0.11 ( 0.01, and 18.5 ( 2.1 nM, respectively,

indicating a strong interaction of these compounds with

mDBS (Figure 6).

6574 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

Figure 4. FACS analysis of the antiapoptotic effects of DHEA and compounds 20, 23, and 27. PC12 cells were cultured in 24-well plates (10⁶cells

per well) for 24 h either in complete media (serum supplemented) or in serum-free media (supplemented with 1% BSA) in the presence of DHEA, 20,

23, or 27 at 100 nM. The X axis represents annexin V-FITC while the Y axis represents the number of events. A total of 10000 cells were assigned per

treatment. Values represent mean ( SEM of at least four independent experiments, each performed in triplicate (*, p < 0.05).

We recently reported that DHEA at low nanomolar con-

centrations (1 nM) protects PC12 cells against apoptosis,

activating withinminutesthe prosurvival transcriptionfactors

NF-κB and CREB, two upstream effectors of antiapoptotic

Bcl-2 proteins.³⁶Serum deprivation of PC12 cells resulted in

strong suppression of the levels of antiapoptotic Bcl-2 and

Bcl-xL proteins. Analogues 20, 23, and 27, in a similar fashion

to DHEA, maintained antiapoptotic protein expression in

serum-deprived PC12 cells to levels comparable to those of

serum supplementation (Figure 7).

Furthermore, DHEA and DHEAS (at nanomolar con-

centrations) increase within 10 min the secretion of catechol-

amines from dopaminergic neural-crest-derived PC12 cells via

actin filament disassembly and stimulate the expression of

tyrosine hydroxylase, thus stimulating de novo synthesis of

catecholamines.²³PC12 cells were exposed to analogues 20

and 27 (10^-7M) or to DHEA for short periods of time (from

5 to 30 min), and the concentration of dopamine in the culture

media was measured using a radioimmunoassay. We found

that the spiro-neurosteroids provoked a fast and statistically

significant 2-fold increase of dopamine secretion within 10 min

(Figure 8A). We also tested the effect of analogues 20 and 27

for longer periods of time and, more specifically, for 3-48 h.

We observed that incubation of PC12 cells with DHEAS or

with analogues 20 and 27 resulted in an increase of dopamine

levels peaking at 24 h (Figure 8B). These long-term effects

of neurosteroids 20 and 27 suggest that these compounds,

in addition to their acute effects on dopamine secretion

(Figure 8A), may also affect the de novo production of

dopamine in dopaminergic neurons like DHEAS. Indeed,

neurosteroids 20 and 27 provoked a strong induction of the

mRNA of the rate-limiting enzyme of dopamine biosynthesis,

tyrosine hydroxylase (Figure 8C).

The risk to develop endocrine-related cancer, in particular

breast, endometrial, or prostate cancer, is distinctly influenced

by endogenous and exogenous hormones, since hormonal

induction of cell proliferation has been associated with a

higher probability of random mutation errors.³⁷The use,

for instance, of estrogen to alleviate postmenopausal syn-

dromes is associated with a higher risk of developing breast or

endometrial cancer.³⁸Endometrial and prostate cancer cells

are also highly responsive to estrogen,^39,40and the risk to

develop prostate cancer is reportedly dependent on estrogens

as well as androgens.^37,41It is therefore necessary to scrutinize

potentially neuroprotective agents of steroidal structure for

possible hormonal effects on the growth of breast, prostate,

and uterine adenocarcinoma cells. Therefore, it was impera-

tive to examine the active spiro-neurosteroids 20, 23, and 27

for such properties.

Initially, we evaluated the effects of compounds 20, 23, and

27 on the expression of ER-regulated genes. Agonism through

ERR was assessed by way of induction of ERE-dependent

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6575

Figure 5. Effects of Δ⁵and Δ⁴androstenes on serum deprivation-induced apoptosis of PC12 cells. Cells were cultured for 24 h either in

complete media (serum supplemented) or in serum-free media (supplemented with 1% BSA) in the presence of various steroids at 100 nM.

Apoptosis was quantified by the APOPercentage assay, depicted as percentage of control (serum-free media in the absence of steroids). Values

represent mean ( SEM of at least four independent experiments, each performed in triplicate (*, p < 0.05).

Figure 6. Displacement of [³H]DHEA from membranes isolated

from PC12 cells by DHEA and spiro-neurosteroids 20, 23, and 27.

Membranes (at a final concentration of 2 mg/mL) from PC12 cells

were incubated in the presence of 5 nM [³H]DHEA in the presence

or absence of various concentrations of analogues 20, 23, and 27,

ranging from 10^-12to 10^-6M. Following a 30 min incubation in a

water bath at 37 °C, membranes were collected on GF/B filters,

prewet in 0.5% PEI solution at 4 °C. The filters were washed three

times with ice-cold Tris-HCl, dried, and counted in a β-scintillation

Figure 7. Induction of the expression of the antiapoptotic Bcl-2 pro-

counter.

teins in serum-deprived PC12 cells by DHEA and compounds 20, 23,

and 27. Cells were cultured for 12 h either in complete or serum-free

luciferase gene expression in MCF-7:D5L cells. In accordance

media containing 10 nM DHEA, 20, 23, or 27. Cellular extracts

with previous findings^42,43treatment of estrogen-free MCF-7:

D5L cells with 1 nM estradiol results in full (4.2-fold) induc-

containing total proteins were collected, and levels of Bcl-2 and Bcl-xL

proteins were measured by Western blot, as described in the Experi-

tion of luciferase, while treatment with 23 or 27 (1 μM) had

no effect (Table 1, column 2). Treatment with analogue 20

mental Section, and normalized per actin protein content. Values

represent mean ( SEM of three independent experiments (*, p < 0.05).

6576 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

Figure 8. Effect of DHEAS and spiro-neurosteroids 20 and 27 on dopamine secretion and production. PC12 cells were exposed to 10 nM

DHEAS, 20, or 27 for 5-30 min (A) or for 3-48 h (B). At the end of the incubation, dopamine levels were measured in the culture media as

described in the Experimental Section. Data are expressed as percentage of control, i.e., cells not exposed to the steroids. Values represent mean

( SEM of four independent experiments. (*, p < 0.01 vs control). (C) DHEAS, 20, and 27 increased the mRNA levels of tyrosine hydroxylase

(TH). PC12 cells were incubated for 6 h in the absence (control) or the presence of steroids at 10 nM, and the mRNA levels of TH were measured

in cell extracts with RT-PCR analysis. Levels of the TH mRNA were compared to GAPDH mRNA levels.

Table 1. Regulation of the Expression of ER-Dependent Reporter Genes by Spiro-Neurosteroids 20, 23, and 27^a

luciferase expression (MCF-7:D5L cells) luciferase expression (HEK:ERβ cells) alkaline phosphatase expression (Ishikawa cells)

ERR agonism

ERR antagonism

(% of 1 μM ICI)^c

ERβ agonism

ERβ antagonism

ER agonism

(% of 0.1 nM E2)^b

ER antagonism

(% of 1 μM ICI)^c

compd (1 μM) (% of 1 nM E2)^b

(% of 1 nM E2)^b(% of 1 μM ICI)^c

20

23

27

weak (29 ( 1)

ns

weak (22 ( 4)

ns

weak (31 ( 8)

weak (24 ( 6)

^aStatistically significant agonism or antagonism (mean ( SEM; n g 3) was classified as full, partial, or weak depending on whether it was

67-100%, 34-66%, and e33%, respectively, of the agonism of E2 or the antagonism of ICI 182,780; ns = nonsignificant. ^bAgonism, expressed as

% of that of estradiol (E2), was calculated by (expression in the presence of neurosteroid - E_vehicle) ꢀ 100/(E_E2- E_vehicle). ^cAntagonism of the agonist

effect of E2, expressed as % of ICI 182,780 antagonism , was calculated by (E_E2- E_{E2 þ neurosteroid}) ꢀ 100/(E_E2- E_{E2 þ ICI}).

at 1 μM resulted in a weak induction which was ERR-

dependent, since it was fully inhibited by ICI 182,780, an

antiestrogen known to knock out ER expression.⁴⁴Treatment

of estradiol (1 nM) supplemented MCF-7:D5L cells with

1 μM ICI 182,780 fully suppressed estradiol induction of

luciferase, whereas treatment with any of 20, 23, or 27 had

no effect up to 1 μM (Table 1, column 3).

Agonism through estrogen receptor β (ERβ) was assessed

by way of induction of ERE-dependent luciferase gene ex-

pression in HEK:ERβ cells.⁴⁵Treatment of estrogen-free

HEK:ERβ cells with 1 nM estradiol induced luciferase by

3.4-fold, while analogues 20, 23, and 27 were ineffective up to

1 μM (Table 1, column 4). Treatment of estradiol (1 nM)

supplemented HEK:ERβ cells with 1 μM ICI 182,780 fully

inhibited estradiol induction of luciferase. Analogue 27 was

weakly inhibitory at 1 μM, while 20 and 23 were ineffective

(Table 1, column 5).

We also tested the ability of analogues 20, 23, and 27 to

induce the expression of AlkP in Ishikawa cells, known to

express ERβ as well as ERR.³⁹Treatment of estrogen-free

Ishikawacells withpostmenopausal levels of estadiol (0.1nM)

resulted in full (5-fold) induction of AlkP,⁴⁶treatment with 23

or 27 (1 μM) had no effect, while with compound 20 (1 μM) a

weakER-dependent induction of AlkP was obtained(Table1,

column 6). Treatment of estradiol (0.1 nM) supplemented

Ishikawa cells with 1 μM ICI 182,780 fully inhibited estradiol

induction of AlkP. In comparison, treatment with 27 (1 μM)

resulted in weak suppression of estradiol induction of AlkP,

while 20 and 23hadno effectup tothisconcentration (Table1,

column 7). Taken together, these data suggest that analogues

20 and 27 may impact AlkP expression of Ishikawa cells

through ERR and ERβ, respectively, albeit weakly and in

opposite directions.

The effects of spiro-neurosteroids 20, 23, and 27 on the

expression of luciferase suggest that, at concentrations e1 μM,

20 is a weak ERR agonist, 27 is a weak ERβ antagonist, and

23 a nonagonist/nonantagonist of either receptor subtype

with regard to ER-mediated ERE-dependent regulation of

gene expression. In line with these findings we observed that,

in Ishikawa cells, 20 induced AlkP expression weakly, pre-

sumably through ERR, 27 inhibited estradiol induction of

AlkP expression weakly, presumably through ERβ, and 23

was ineffective in either respect, as one would expect from a

nonagonist/nonantagonist of ER.

Subsequently, we examined whether 20, 23, and 27 could

stimulate the growth of MCF-7, LNCaP, and Ishikawa cells.

MCF-7 cells are known to express ERR and to proliferate

in an estrogen- and ERR-dependent manner.^42,43Estradiol

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6577

at concentrations g0.1 nM is known to fully induce the

ERR-dependent growth of MCF-7 cells, whereas 1 μM ICI

172,780 is known to fully inhibit the hormonal effect.⁴²We

found that at 1 μM derivative 20, but not 23 or 27, weakly

stimulated estrogen-free MCF-7 cells to grow in a manner

dependent on ER (Table 2, column 2), since the stimulation

was fully inhibited by 1 μM ICI 182,780. These data indicate

that 20 is a weak agonist of the ER-dependent growth of

MCF-7 cells.

Treatment of estrogen-free Ishikawa cells with 0.1 nM

estradiol for 3 days stimulated the growth of these cells

marginally (15%), and this effect was inhibited by the pre-

sence of 1 μM ICI 182,780. Treatment of Ishikawa cells with

any of 20, 23, and 27 failed to impact on cell growth

significantly (Table 2, column 3).

Finally, we examined the effect of 20, 23, and 27 on the

growth of LNCaP cells, which are known to express ERR,

ERβ, and a mutant form of the androgen receptor (AR).^47,48

The growth of these cells is promoted by estrogens as well as

androgens,^40,48and ERβ is reported to be requisite for

stimulation of cell proliferation by either hormone.⁴⁸How-

ever, the mechanism of estrogen action and the ER subtype(s)

involved in mediating estrogen responsiveness of these cells

remain elusive.^40,47-49The growth of LNCaP cell was signi-

ficantly stimulated by 10 nM estradiol in a manner fully

suppressed by 10 μM ICI 182,780. However, we found that

analogues 20, 23, and 27 could not stimulate the growth of

estrogen-free LNCaP cells (Table 2, column 4). In light of the

above it is clear that none of the three compounds had a

substantial stimulatory effect on the growth of MCF-7,

Ishikawa, or LNCaP cells, even at 1 μM, which is in accor-

dance with the weak ER agonist properties of these com-

pounds, described above.

Table 2. Stimulation of the Growth of Endocrine Cancer Cells by

Spiro-Neurosteroids 20, 23, and 27^a

MCF-7 cell

growth induction

(% of 0.1 nM E2)

Ishikawa cell

growth induction

(% of 0.1 nM E2)

LNCaP cell

growth induction

(% of 10 nM E2)

The effect of the most potent steroids 20, 23, and 27 on cell

viability was examined in three different cell lines, namely,

PC12 (rat adrenal pheochromocytoma), HEK293 (human

embryonic kidney), and HT22 (mouse hippocampal neuro-

blastoma) cells, using the MTT assay. All three compounds

were applied to the cell cultures in concentrations 0.1, 1, and

10 μM, and cell survival was assessed after 24, 48, or 72 h. As

shown in Figure 9, all three compounds failed to significantly

affect cell viability at all concentrations tested.

compd

(1 μM)

20

23

27

weak (33 ( 11)

ns

^aInduction of cell growth, as assessed spectrophotometrically at

550 nm using the MTT assay and expressed as % of that of estradiol

(E2), was calculated by (OD at 550 nm in neurosteroid - OD_vehicle) ꢀ

100/(OD_E2- OD_vehicle)]. Statistically significant growth-inducing ef-

fects (mean ( SEM; n g 3) were classified as full, partial, or weak

depending on whether they were 67-100%, 34-66%, and e33%,

respectively, of the effects of E2; ns = nonsignificant.

As an initial step to correlate the in vitro results with the

active conformations of compounds 20, 23, and 27, a combina-

tion of minimization algorithms with stochastic and systematic

Figure 9. Effects of compounds 20, 23, and 27 on cell viability. PC12, HEK293, and HT22 cells (2ꢀ10⁵cells/mL) were cultured in serum-

supplemented culture media for 24, 48, or 72 h in the absence or the presence of various 0.1, 1, and 10 μM synthetic steroids. At the end of

incubation viable cell numbers were determined using the MTT assay. Results (mean ( SEM of three replicate values of three independent

experiments) are expressed as percentage of untreated control cultures.

6578 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

conformational analysis was employed. The initial conforma-

tion of the compounds was built using the crystal structure of

DHEA,⁵⁰assuming a chair conformation for the A and C rings,

a nearly ideal half-chair conformation for ring B, and a 14a

envelope conformation for ring D. In a step further, the

orientation of the flexible CH₂OH moiety at C21 of 23 and

27 was systematically explored. Thus, grid scan search around

torsion angles τ₁(C17-C20-C21-O) and τ₂(C20-C21-O-H)

resulted in five distinct areas, and the letters A-E and A⁰-E⁰

indicate the corresponding conformational local minima for 23

and 27, respectively. The derived conformers A-E and A⁰-E⁰

are depicted in Figure 10, while their energy values and torsion

angles are shown in Table 3. The energy map as a function of

the dihedral angles τ₁and τ₂is plotted in Figure 11, and the

isoenergetic curves, plotted with a step of 1.5 kcal mol^-1, define

the energetic barriers between conformations. The derived

minima were further examined for their consistency with

experimental NOE data in order to obtain the most probable

conformations. The most informative experimentally observed

dipole interactions were those between proton H20 of the epoxy

ring and protons H12 of the steroid C ring as well as between

both methylene protons H21a and H21b and protons H16 of

the D ring (Figure 10). Conformers E and E⁰adopting an

equatorial configuration for the C21-OH group and arranging

H21 protons in spatial vicinity with H16 are in full agreement

with the NOE data. Analogue 20 was subjected to energy

minimization algorithms, and only one lowest energy confor-

mer was found with 35.5 kcal mol^-1. Subsequently, the pre-

ferred conformations for 23 and 27, E and E⁰, respectively, were

superimposed with that of 20 and DHEA (energy value

calculated at 41.4 kcal mol^-1) (Figure 12). The steroid back-

bone and the 3β-OH group are successfully aligned in all four

compounds, which leads to the overlay of the epoxy groups of

20 and 23 in a pseudoequatorial configuration, while 27

arranges the epoxy group below the steroid plane. In DHEA

the 17-carbonyl group is positioned in an equatorial orientation

with the oxygen atom being equidistant from the epoxide

oxygen either in the 17R or 17β configuration.

Concerning the interaction of analogues 20, 23, and 27 with

the mDBS, and taking into account the apparent K_dvalues, it

is evident that the 17β configuration for the oxirane oxygen is

more advantageous (cf. compounds 20 and 23) than the 17R

with both 20 and 23 having higher affinity for the mDBS than

DHEA. Furthermore, compounds 20, 23, and 27 lack a

phenyl hydroxyl, and thus, they cannot effectively mimic

17β-estradiol in forming the key hydrogen bonds with

Glu353 and Arg394 that underlie high-affinity binding to

ERR or with Glu305/Arg346 for ERβ.^42,51,52As a result,

productive interaction through hydrogen bond formation

between the 17 oxygen of the oxirane moiety and His524 of

ERR or His475 of ERβ, and the ensuing stabilization of the

agonist conformation of the receptor, could be reduced or

even abolished. This is more likely in the case of ERβ, due to

the inherently lower stability of its agonist conformation and

the stricter constraints for optimal fitting into the consider-

ably less spacious binding cavity of this ER subtype.⁵²These

considerations could account for 20 being a weak agonist

through ERR and a nonagonist through ERβ. The structural

difference between 20and23is the presenceofaC21-hydroxy-

methylene group which could be prohibitory for binding

and, thus, result in activation or inhibition of either form of

ER by 23. Concerning 27, the opposite configuration of the

oxirane with respect to 23 may account for it being a weak

antagonist through ERβ and a nonantagonist through ERR.

Our rationale concerning the design of the new DHEA

analogues was based on obtaining active compounds that

would be devoid of hormonal side effects. Thus, we intro-

duced substituents at position C3 of DHEA in order to inhibit

the enzymatic oxidation of the 3β-hydroxyl group to a ketone

or the aromatization of the A ring (DHEA f androstene-

dione f testosterone f estradiol). However, all of the res-

pective compounds (5, 7, 9, 11, 14, and 16), irrespectively

Figure 10. Conformational minima for compound 23 (A-E) and

27 (A⁰-E⁰) derived from a grid scan search around dihedral angles τ₁

(C17-C20-C21-O) and τ₂(C20-C21-O-H).

Table 3. Energy Values and Dihedral Angles of the Conformational Minima Obtained from Grid Scan Search for Compounds 23 and 27

compd 23

(conformers)

energy value

(kcal mol^-1

dihedral

angle τ₁(deg)

dihedral

angle τ₂(deg)

compd 27

(conformers)

energy value

(kcal mol^-1

dihedral

angle τ₁(deg)

dihedral

angle τ₂(deg)

)

A

B

C

D

E

24.7

37.9

31.5

38.5

36.2

-58

-48

82

-13

-161

-20

180

A⁰

B⁰

C⁰

D⁰

E⁰

24.9

38.7

31.2

35.5

36.5

59

43

14

166

-75

-86

173

15

-175

171

95

-164

165

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6579

Figure 11. Contour maps for compounds 23 and 27 generated by rotating around dihedral angles τ₁and τ₂in increments of 10°. The contour

levels which are up to 15 kcal mol^-1higher than the lower energy minimum are plotted with a step of 1.5 kcal mol^-1. Important NOE inter-

actions are indicated by the pink arrows.

apoptosis, with IC₅₀in the range of 0.1-100 nM. These com-

pounds bind with high affinity to membrane DHEA binding

sites³⁵(K_dat nanomolar concentration) and mimic endogen-

ous neurosteroids in inducing prosurvival antiapoptotic Bcl-2

proteins. The new potent spiro-neurosteroid analogues were

found to be practically unable to activate ERR or ERβ up to a

concentration of 1 μM and to interfere with estrogen induc-

tion of ERE-dependent gene expression, whether in the pre-

sence of pre- or postmenopausal levels of estradiol. In addi-

tion, they were found to be unable to impact the growth of

Ishikawa, MCF-7, and LNCaP cells to any substantial extent.

Figure 12. Favorable conformations of 23 (conformer E) and 27

(conformer E⁰), derived from a grid scan search, are superimposed

with energy-optimized structures of 20 and DHEA.

Our results suggest that the spiro-epoxyneurosteroid deriva-

tives 20, 23, and 27 might prove to be lead molecules for the

synthesis of novel neuroprotective agents.

of the nature of the 3R-substituent (aliphatic or polar), were

devoid of antiapoptotic activity, at least at 100 nM concentra-

tion. Subsequently, we focused our search on modifications at

C17, which is also an important position for the metabolic

transformation of DHEA to androgens or estrogens. The

epoxide group was chosen as a substituent at C17 since it is

relatively small, thus disturbing binding to the DHEA mem-

brane receptor sites to a lesser extent, and it has relatively low

reactivity, which should reduce the potential for unselective

alkylation during passage to the target cells. In addition, the

C21-hydroxyl group was introduced since it can participate

in hydrogen bond formation and, thus, potentially improve

binding. As described above, the spiro-neurosteroid ana-

logues 20, 23, and 27 were not only very potent but devoid

of hormonal side effects. The order of antiapoptotic potencies

(Figure 3: 20 > 27 > 23) and that of binding affinities for the

membrane DHEA binding sites (Figure 6: 20 > 23 > 27) are

similar, which could imply that efficient binding to the DHEA

membrane binding sites is the factor predominantly influen-

cing the antiapoptotic activity of these steroid derivatives.

Experimental Section

NMR spectra were recorded on a Varian 300 spectrometer

operating at 300 MHz for H and 75.43 MHz for ¹³C or on a

1

Varian 600 operating at 600 MHz for ¹H. ¹H NMR spectra are

reported in units of δ relative to the internal standard of signals

of the remaining protons of deuterated chloroform at 7.24 ppm.

¹³C NMR shifts are expressed in units of δ relative to CDCl₃at

77.0 ppm. ¹³C NMR spectra were proton noise decoupled. All

NMR spectra were recorded in CDCl₃. Silica gel plates (Merck

F254) were used for thin-layer chromatography. Chromato-

graphic purification was performed with silica gel (200-400 mesh).

Analyses, indicated by the symbols of the elements, were carried

out by the microanalytical section of the Institute of Organic and

Pharmaceutical Chemistry of the National Hellenic Research

Foundation.

17,17-Ethylenodioxy-5-androsten-3β-ol (1). To a solution of

DHEA (1 g, 3.46 mmol) in cyclohexane (100 mL) was added

ethylene glycol (0.6 mL, 10.38 mmol) and camphor-10-sulfonic

acid (CSA) (10 mg, 0.04 mmol), and the resulting mixture was

heated at reflux for 4 h using a Dean-Stark apparatus. After

completion of the reaction, the mixture was allowed to cool to

room temperature and was extracted with ethyl acetate and

saturated aqueous NaHCO₃. The organic layer was washed with

brine and was dried over Na₂SO₄. The solvent was removed

in vacuo to afford ketal 1 (1 g, 92%) as a white solid which was

Conclusions

A number of highly effective compounds were developed,

which protect neuronal cells against serum deprivation induced

6580 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

used in the next step without purification: mp 162-165 °C;

[R]²⁰_D=-93.42° (c=0.00152 g/mL, CHCl₃); ¹H NMR (CDCl₃)

δ 0.85 (s, 3H, CH₃), 1.00 (s, 3H, CH₃), 1.07-2.27 (m, 19H),

3.47-3.57 (m, 1H, H-3R), 3.83-3.94 (m, 4H, OCH₂CH₂O), 5.34

(d, J=5.49 Hz, 1H, H-6); ¹³C NMR (CDCl₃) δ 14.4, 19.6, 20.6,

22.7, 24.3, 30.5, 31.2, 31.6, 32.2, 34.2, 36.5, 37.2, 42.2, 45.7, 49.9,

50.6, 65.2, 71.6, 119.5, 121.4, 140.7. Anal. (C₂₁H₃₂O₃) C, H.

17,17-Ethylenodioxy-5-androsten-3-one (2). To a solution of

alcohol 1 (50 mg, 0.15 mmol) in dry dichloromethane (10 mL)

was added at 0 °C Dess-Martin periodinane (130 mg, 0.30

mmol), and the mixture was stirred at room temperature for

1 h. Then it was diluted with ether and was quenched with a

mixture of saturated aqueous NaHCO₃/Na₂S₂O₃(1:3). The

organic layer was washed with brine and was dried over Na₂SO₄,

and the solvent was evaporated in vacuo to afford unsaturated

ketone 2 (49 mg, quantitative), which was pure enough to be

used for the next step without purification: mp 141-144 °C; ¹H

NMR (CDCl₃) δ 0.87 (s, 3H, CH₃), 1.17 (s, 3H, CH₃), 1.02-2.77

(m, 17H), 2.80 (dd, J=2.44, 16.48 Hz, 1H), 3.26 (dd, J=3.05,

16.48 Hz, 1H), 3.82-3.92 (m, 4H, OCH₂CH₂O), 5.32 (dd, J =

5.49, 2.0 Hz, 1H, H-6); ¹³C NMR (CDCl₃) δ 14.3, 19.2,

20.4, 22.6, 30.4, 31.1, 32.8, 34.1, 35.7, 36.8, 37.6, 45.7, 48.3,

49.0, 49.7, 50.4, 64.5, 65.2, 119.3, 122.6, 138.5, 210.1. Anal.

(C₂₁H₃₀O₃) C, H.

160-164 °C; [R]²⁰_D= 25° (c = 0.001 g/mL, CHCl₃); ¹H NMR

(CDCl₃) δ 0.88 (s, 3H, CH₃), 1.06 (s, 3H, CH₃), 1.02-2.12

(m, 16H), 2.13 (dd, J=1.2, 13.7 Hz, 1H, H-4R), 2.40 (dd, J=2.0,

13.7 Hz, 1H, H-4β), 2.40-2.44 (m, 1H), 3.17 (d J=9.15 Hz, half

of AB system, 1H), 3.24 (d, J=9.15 Hz, half of AB system, 1H),

3.36 (s, 3H, CH₃O-), 5.32 (d, J = 5.49 Hz, 1H, H-6). Anal.

(C₂₁H₃₂O₃) C, H.

(17,17-Ethylenedioxy-3β-hydroxy-5-androsten-3r-yl)acetonitrile (6).

To a solution of epoxide 3 (60 mg, 0.17 mmol) in absolute

ethanol (12 mL) was added potassium cyanide (14 mg, 0.21

mmol) at room temperatue, and the mixture was stirred for 48 h.

The reaction mixture was refluxed for 5 h, followed by stirring at

room temperature for 24 h. Subsequently, water and ether were

added to the reaction mixture, the organic layer was washed

with brine and was dried over anhydrous Na₂SO₄, and the

solvent was evaporated in vacuo. The crude product was purified

by flash column chromatography using petroleum ether, 40-

60 °C, and ethyl acetate (8:2) as elution solvent to yield com-

pound 6 (40 mg, 61%); ¹H NMR (CDCl₃) δ 0.84 (s, 3H, CH₃),

1.02 (s, 3H, CH₃), 1.20-2.30 (m, 19H), 2.47 (br s, 2H, 3R-

CH₂CN), 3.85-3.88 (m, 4H, OCH₂CH₂O), 5.40 (d, J=5.49 Hz,

1H, H-6); ¹³C NMR (CDCl₃) δ 14.0, 19.4, 20.2, 21.9, 27.9, 30.6,

31.1, 31.9, 34.2, 35.2, 36.2, 36.5, 44.5, 47.9, 50.6, 51.3, 65.3, 71.7,

117.4, 119.3, 121.2, 140.4.

17,17-Ethylenodioxy-5-androstene-3,2⁰-oxirane (3). To a solu-

tion of trimethylsulfonium iodide (370 mg, 1.82 mmol) in dry

THF (5 mL) at 0 °C was added n-BuLi (1.6 M solution in

hexanes, 1.15 mL, 1.82 mmol), and the resulting mixture was

stirred at 0 °C for 30 min. The reaction was cooled at -40 °C,

and a solution of ketone 2 (400 mg, 1.21 mmol) in THF (10 mL)

was added. The resulting mixture was stirred at -40 °C for 2 h

and then at room temperature for 12 h. The solvent was

evaporated in vacuo, the residue was diluted with ether, the

organic layer was extracted with water and brine and was dried

over anhydrous Na₂SO₄, and the solvent was removed in vacuo.

The crude residue was purified by flash column chromato-

graphy using petroleum ether, 40-60 °C, and ethyl acetate (92:8)

as eluting solvent to afford the desired epoxide 3 (130 mg, 32%)

as colorless solid: mp 110-113 °C; R_f= 0.60 (petroleum ether,

40-60 °C/ethyl acetate, 8/2); ¹H NMR (CDCl₃) δ 0.87 (s, 3H,

CH₃), 1.06 (s, 3H, CH₃), 1.04-2.18 (m, 18H), 2.56 (d, J=4.8 Hz,

1H, CH₂, epoxide), 2.58 (d, J=4.8 Hz, 1H, CH₂, epoxide), 2.88

(d, J=13.7 Hz, 1H), 3.85-3.91 (m, 4H, OCH₂CH₂O), 5.30 (d,

J =5.0 Hz, 1H, H-6). Anal. (C₂₂H₃₂O₃) C, H.

(3β-Hydroxy-5-androsten-17-oxo-3r-yl)acetonitrile (7). Fol-

lowing the procedure for analogue 5, ketone 7 was obtained

from compound 6 (40 mg, 0.11 mmol) after purification by flash

column chromatography (petroleum ether, 40-60 °C/ethyl

acetate (8:2)) as a colorless solid, 30 mg (86% yield): mp

1

190-194 °C; [R]²⁰_D= 21.74° (c = 0.00115 g/mL, CHCl₃); H

NMR (CDCl₃) δ 0.88 (s, 3H, CH₃), 1.06 (s, 3H, CH₃), 1.02-2.22

(m, 16H), 2.20 (dd, J = 2.7, 13.7 Hz, 1H, H-4R), 2.43-2.49

(m, 1H), 2.48 (s, 2H, -CH₂CN), 2.50 (dd, J = 1.7, 13.7 Hz,

1H, H-4β), 5.45 (d, J =5.13 Hz, 1H, H-6). Anal. (C₂₁H₂₉NO₂)

C, H, N.

17,17-Ethylenedioxy-3r-azidomethyl-5-androsten-3β-ol (8). To

a solution of epoxide 3 (80 mg, 0.23 mmol) in a mixture of

methanol and water (8:1) (18 mL) was added sequentially

sodium azide (75 mg, 1.15 mmol) and ammonium chloride

(28 mg, 0.51 mmol), and the resulting mixture was refluxed

overnight. The mixture was filtered from a silica gel pad, and the

filtrate was evaporated in vacuo. The residue was purified by

flash column chromatography (petroleum ether, 40-60 °C/

ethyl acetate (8:2)) to obtain product 8 (60 mg, 67% yield): ¹H

NMR (CDCl₃) δ 0.83 (s, 3H, CH₃), 1.02 (s, 3H, CH₃), 1.02-

2.13 (m, 18H), 2.40 (dd, J = 2.44, 13.43 Hz, 1H), 3.23 (d, J =

12.21 Hz, 1H, AB system, 3R-CH₂-N₃), 3.29 (d, J = 12.21 Hz,

AB system, 1H, 3R-CH₂-N₃), 3.82-3.92 (m, 4H, OCH₂CH₂O),

5.33 (dd, J=2.46, 4.89 Hz, 1H, H-6); ¹³C NMR (CDCl₃) δ 13.3,

19.2, 20.5, 21.3, 30.5, 31.7, 31.5, 31.8, 35.5, 36.1, 36.5, 42.8, 47.7,

50.3, 51.6, 56.7, 65.4, 73.1, 118.2, 121.4, 140.7.

17,17-Ethylenodioxy-3r-methoxymethyl-5-androsten-3β-ol (4).

To a solution of epoxide 3 (50 mg, 0.14 mmol) in anhydrous

methanol (5 mL) was added sodium methoxide (24 mg, 0.44

mmol), and the mixture was refluxed for 5 h. The solvent was

evaporated in vacuo, and the crude residue was purified by flash

column chromatography using petroleum ether, 40-60 °C, and

ethyl acetate (8:2) as eluting solvent to afford the desired

1

compound 4 (40 mg, 74%). H NMR (CDCl₃) δ 0.84 (s, 3H,

3β-Hydroxyl-3r-azidomethyl-5-androsten-17-one (9). Follow-

ing the procedure for analogue 5, using ketal 8 (60 mg, 0.15

mmol), compound 9 was obtained after purification by flash

column chromatography using petroleum ether, 40-60 °C/ethyl

acetate (8:2) as a colorless solid: yield 45 mg, 85%; mp 168-

19-CH₃), 1.02 (s, 3H, 18-CH₃), 1.02-2.12 (m, 19H), 2.39 (dd,

J = 2.44, 13.43 Hz, 1H), 3.21 (q, J = 9.76 Hz, 2H, 3R-CH₂O),

3.36 (s, 3H, CH₃O), 3.83-3.93 (m, 4H, OCH₂CH₂O), 5.29

(d, J = 5.49 Hz, 1H, H-6); ¹³C NMR (CDCl₃) δ 13.3, 19.2,

20.5, 21.2, 30.6, 31.2, 31.5, 31.8, 35.6, 36.5, 36.8, 42.1, 47.5, 50.7,

51.8, 59.6, 65.3, 71.6, 75.1, 119.3, 121.2, 140.5.

3β-Hydroxy-3r-methoxymethyl-5-androsten-17-one (5). To a

solution of 17,17-ethylenodioxy-3R-methoxymethyl-5-andro-

sten-3β-ol (4) (40 mg, 0.11 mmol) in THF (10 mL) was added

a mixture of aqueous 1 M HCl/acetone (1:1) (2 mL), and the

reaction mixture was stirred at room temperature for 12 h.

Subsequently, the reaction mixture was neutralized by addition

of aqueous 1 M NaOH, the organic layer was extracted with

ethyl acetate, washed with brine, and dried over anhydrous

Na₂SO₄, and the solvent was evaporated in vacuo. The crude

product was purified by flash column chromatography using

petroleum ether, 40-60 °C, and ethyl acetate (7:3) as elution

solvent to yield ketone 5 (30 mg, 86%) as a white solid: mp

1

172 °C; [R]²⁰_D= 22.22° (c = 0.0018 g/mL, CHCl₃); H NMR

(CDCl₃) δ 0.88 (s, 3H, CH₃), 1.06 (s, 3H, CH₃), 1.02-2.17 (m,

17H), 2.45 (m, 2H), 3.26 (d, J = 12.40 Hz, 1H, AB system, 3R-

CH₂-N₃), 3.30 (d, J = 12.40 Hz, AB system, 1H, 3R-CH₂-N₃),

5.38 (d, J=5.15 Hz, 1H, H-6). Anal. (C₂₀H₂₉N₃O₂) C, H, N.

3β-n-Butyl-17,17-ethylenodioxy-5-androsten-3r-ol (10). To a

solution of ketone 2 (100 mg, 0.30 mmol) in dry THF (3 mL) was

added at 0 °C n-BuLi (1.6 M solution in hexanes, 0.38 mL, 0.60

mmol), and the reaction mixture was stirred at room temperature

overnight. The mixture was quenched with saturated aqueous

NH₄Cl solution and was extracted with ethyl acetate. The organic

layer was washed with brine and dried over Na₂SO₄, and the

solvent was removed in vacuo. The residue was purified by flash

column chromatography (petroleum ether, 40-60 °C/ethyl acetate

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6581

(95:5)) to obtain analogue 10 (64 mg, 55%): R_f=0.60 (petroleum

(m, 19H), 3.47-3.54 (m, 1H, 3a-H), 5.13 (d, J=4.89 Hz, 1H, H-

6), 7.35-7.38 (m, 6H, ArH), 7.65-7.67 (m, 4H, ArH); ¹³C NMR

(CDCl₃) δ 14.3, 15.9, 19.3, 20.1, 20.3, 20.5, 27.3, 29.7, 31.6, 32.2,

34.2, 36.8, 38.8, 39.5, 42.2, 44.0, 58.6, 71.2, 121.2, 128.1, 129.9,

133.4, 135.8, 218.9. Anal. (C₃₅H₄₆O₂Si) C, H.

1

ether, 40-60 °C/ethyl acetate (8/2)); H NMR (CDCl₃) δ 0.85

(s, 3H, CH₃), 0.90 (t, J = 7.10 Hz, 3H, CH₃), 0.97 (s, 3H, CH₃),

1.02-2.10 (m, 24H), 2.36 (dd, J=2.30, 14.20 Hz, 1H), 3.91-3.96

(m, 4H, OCH₂CH₂O), 5.39 (d, J=5.49 Hz, 1H, H-6).

3β-n-Butyl-3r-hydroxy-5-androsten-17-one (11). Compound

11 was obtained from ketal 10 (64 mg, 0.16 mmol) following the

procedure for analogue 5 after purification by flash column

chromatography (petroleum ether, 40-60 °C/ethyl acetate

(8:2)) as a colorless solid, 40 mg (71% yield): mp 151-154 °C;

¹H NMR (CDCl₃) δ 0.87 (m, 3H, CH₃), 0.90 (t, J=7.10 Hz, 3H,

CH₃), 0.99 (s, 3H, CH₃), 1.12-2.14 (m, 24H), 2.36-2.50 (m,

2H), 5.41 (d, J=5.49 Hz, 1H, 6-CH); ¹³C NMR (CDCl₃) δ 14.5,

15.1, 19.5, 21.1, 22.7, 24.1, 26.2, 31.6, 32.1, 32.2, 33.6, 35.6, 36.5,

37.7, 43.1, 44.5, 47.9, 50.8, 52.1, 72.0, 122.3, 139.4, 218.9. Anal.

(C₂₃H₃₆O₂) C, H.

17,17-Ethylenodioxy-3r-methyl-5-androsten-3β-ol (12) and

17,17-Ethylenodioxy-3β-methyl-5-androsten-3R-ol (13). To a sol-

ution of ketone 2 (100 mg, 0.30 mmol) in dry THF (5 mL) was

added MeLi (1.6 M solution in ether, 0.41 mL, 0.60 mmol) at

0 °C, and the resulting mixture was stirred at room temperature

for 3 h. The reaction was quenched with saturated aqueous

NH₄Cl and was extracted with ethyl acetate. The organic layer

was washed with brine and dried over Na₂SO₄, and the solvent

was removed in vacuo. The residue was purified by flash column

chromatography (petroleum ether, 40-60 °C/ethyl acetate

(85:15)) to afford products 12 (29 mg) and 13 (58 mg) (75%

overall yield).

3β-(tert-Butyldiphenylsilyloxy)-17β-hydroxy-5-androstene (17).

To a suspension of LiAlH₄(36 mg, 0.95 mmol) in anhydrous

THF (2 mL) was added dropwise at 0 °C a solution of ketone 16

(250 mg, 0.47 mmol) in anhydrous THF (5 mL), and the

resulting mixture was stirred at room temperature for 3 h. Sub-

sequently, the reaction was quenched at 0 °C by a mixture of

H₂O/THF (1:1) (3 mL) followed by addition of ethyl acetate

(30 mL). The resulting mixture was dried over Na₂SO₄, the solid

was filtered, and the filtrate was evaporated in vacuo. The

residue was purified by flash column chromatography

(petroleum ether, 40-60 °C/ethyl acetate (8:2)) to afford alcohol

17 as a white solid in 92% yield (230 mg): mp 161-164 °C;

[R]²⁰_D=37.21° (c=0.00043 g/mL, CHCl₃); ¹H NMR (CDCl₃) δ

0.73 (s, 3H, CH₃), 1.00 (s, 3H, CH₃), 1.06 (s, 9H, (CH₃)₃C-Si),

0.90-2.39 (m, 19H), 3.52-3.62 (m, 2H, 3R-H and 17R-H), 5.12

(d, J = 4.89 Hz, 1H, H-6), 7.35-7.39 (m, 6H, ArH), 7.66-7.70

(m, 4H, ArH); ¹³C NMR (CDCl₃) δ 10.9, 19.1, 19.4, 20.6, 23.4,

26.9, 30.5, 31.4, 31.8, 31.9, 36.6, 37.2, 42.4, 42.7, 50.1, 51.3, 73.1,

81.9, 120.8, 127.4, 129.4, 134.8, 135.6, 135.7, 141.4. Anal.

(C₃₅H₄₈O₂Si) C, H.

3β-(tert-Butyldiphenylsilyloxy)-17β-methoxy-5-androstene (18).

To NaH (60% dispersion in mineral oil, 46 mg, 1.14 mmol) in

anhydrous THF (1 mL) was added at 0 °C a solution of alcohol

17 (300 mg, 0.57 mmol) in anhydrous THF (4 mL), and the

resulting mixture was stirred for 30 min at room temperature.

Subsequently, MeI (70 μL, 1.14 mmol) was added, and the

reaction mixture was stirred at room temperature overnight.

The reaction was quenched with saturated aqueous NH₄Cl and

was extracted with ethyl acetate. The organic layer was washed

with brine and was dried over Na₂SO₄, and the solvent was

evaporated in vacuo. The residue was purified by flash column

chromatography (petroleum ether, 40-60 °C/ether (9:1)) to

afford analogue 18 (150 mg, 81%) as a white solid: mp

149-152 °C; [R]²⁰_D= -46.49° (c =0.00114 g/mL, CHCl₃); ¹H

NMR (CDCl₃) δ 0.76 (s, 3H, CH₃), 1.00 (s, 3H, CH₃), 1.07

(s, 9H, Si-C(CH₃)₃), 0.80-2.39 (m, 19H), 3.20 (t, J=7.9 Hz, 1H,

H- 17R), 3.33 (s, 3H, CH₃O), 3.54-3.58 (m, 1H, H-3a), 5.13 (d,

J = 4.89 Hz, 1H, H-6), 7.36-7.39 (m, 6H, ArH), 7.67-7.70

(m, 4H, ArH); ¹³C NMR (CDCl₃) δ 19.1, 20.7, 23.3, 27.0, 27.6,

27.9, 31.4, 31.7, 31.8, 36.5, 37.2, 37.9, 42.5, 42.6, 50.1, 51.6, 57.8,

73.1, 90.7, 120.8, 127.4, 129.4, 134.8, 135.7, 141.2. Anal.

(C₃₆H₅₀O₂Si) C, H.

17,17-Ethylenodioxy-3r-methyl-5-androsten-3β-ol (12). R_f=

0.23 (petroleum ether, 40-60 °C/acetone (9/1)); ¹H NMR

(CDCl₃) δ 0.87 (s, 3H, CH₃), 1.00 (s, 3H, CH₃), 1.09 (s, 3H,

3R-CH₃), 1.02-2.13 (m, 18H), 2.45 (m, 1H), 3.82-3.91 (m, 4H,

OCH₂CH₂O), 5.30 (d, J =5.49 Hz, 1H, H-6).

17,17-Ethylenodioxy-3β-methyl-5-androsten-3r-ol (13). R_f=

0.30 (petroleum ether, 40-60 °C/acetone (9/1));¹H NMR

(CDCl₃) δ 0.84 (s, 3H, CH₃), 0.97 (s, 3H, CH₃), 1.20 (s, 3H,

3β-CH₃), 1.02-2.37 (m, 18H), 2.40 (dd, J=2.44, 14.04 Hz, 1H),

3.83-3.92 (m, 4H, OCH₂CH₂O), 5.39 (d, J= 5.49, 1H, H-6).

3β-Hydroxy-3r-methyl-5-androsten-17-one (14). Compound

14 was prepared following the method for 5 using the steroidal

ketal 12 (29 mg, 0.08 mmol) after purification by flash column

chromatography using (petroleum ether, 40-60 °C/acetone

(85:15)) in 87% yield (21 mg) as a white solid: mp 130-

133 °C; [R]²⁰_D= 12.55° (c = 0.00255 g/mL, CHCl₃); ¹H NMR

(CDCl₃) δ 0.87 (s, 3H, CH₃), 1.03 (s, 3H, CH₃), 1.12 (s, 3H, 3R-

CH₃), 1.02-2.14 (m, 17H), 2.45 (m, 2H), 5.33 (d, J = 5.13 Hz,

1H, H-6). Anal. (C₂₀H₃₀O₂) C, H.

17β-Methoxy-5-androsten-3β-ol (19). A solution of analogue

18 (140 mg, 0.26 mmol) in anhydrous THF (2 mL) was treated

with a solution of (n-Bu)₄N^þF^-(1 M solution in THF, 6.5 mL,

6.5 mmol), and the resulting mixture was stirred at room

temperature overnight. The reaction was quenched with satu-

rated aqueous NH₄Cl and was extracted with ethyl acetate. The

organic layer was washed with brine and dried over Na₂SO₄, and

the solvent was removed in vacuo. The residue was purified by

flash column chromatography (dichloromethane/ethyl acetate

(9:1)) to yield alcohol 19 (78 mg, 99%) as a white solid: mp

165-168 °C; [R]²⁰_D= -70.92° (c =0.00141 g/mL, CHCl₃); ¹H

NMR (CDCl₃) δ 0.75 (s, 3H, 19-CH₃), 0.99 (s, 3H, 18-CH₃),

0.90-2.25 (m, 19H), 3.21 (t, J = 8.55 Hz, 1H, H-17R), 3.33

(s, 3H, CH₃O), 3.47-3.54 (m, 1H, H-3a), 5.31 (d, J = 4.89 Hz,

1H, H-6); ¹³C NMR (CDCl₃) δ 11.4, 19.4, 20.7, 23.3, 27.6, 31.4,

31.6, 31.7, 36.5, 37.2, 37.8, 42.2, 42.6, 50.2, 51.6, 57.8, 71.6, 90.7,

121.2, 140.8. Anal. (C₂₀H₃₂O₂) C, H.

3r-Hydroxy-3β-methyl-5-androsten-17-one (15). Following

the procedure for 5 using ketal 13 (58 mg, 0.17 mmol), we

obtained ketone 15 (21 mg, 87%) after purification by flash

column chromatography (petroleum ether, 40-60 °C/acetone

(85:15)): [R]²⁰_D=24.79° (c=0.00117 g/mL, CHCl₃); ¹H NMR

(CDCl₃) δ 0.88 (s, 3H, CH₃), 1.00 (s, 3H, CH₃), 1.22 (s, 3H,

3β-CH₃), 1.02-2.15 (m, 17H), 2.42 (m, 1H, H-4β), 2.45 (m, 1H),

5.42 (d, J = 5.49 Hz, 1H, H-6). Anal. (C₂₀H₃₀O₂) C, H.

3β-(tert-Butyldiphenylsilyloxy)-5-androsten-17-one (16). To a

solution of DHEA (500 mg, 1.73 mmol) in anhydrous DMF

(15 mL) was added at 0 °C imidazole (295 mg, 4.32 mmol), and

the mixture was stirred for 15 min. Subsequently was added tert-

butyldiphenylsilyl chloride (1.1 mL, 4.32 mmol), and the result-

ing mixture was stirred at 50 °C overnight. The reaction mixture

was quenched with saturated aqueous NH₄Cl, and the mixture

was stirred for an additional 30 min at room temperature. The

solvent was evaporated in vacuo, the residue was diluted with

ethyl acetate, and the organic layer was washed with water and

brine and was dried over Na₂SO₄. The solvent was removed

in vacuo followed by recrystallization from ethanol and petro-

leum ether, 40-60 °C, to yield analogue 16 (720 mg, 79%) as

a white solid: mp 117-120 °C; ¹H NMR (CDCl₃) δ 0.85 (s, 3H,

CH₃), 1.00 (s, 3H, CH₃), 1.05 (s, 9H, (CH₃)₃C-Si), 1.10-2.47

17β-Spiro[5-androstene-17,2⁰-oxiran]-3β-ol (20). To a solution

of DHEA (500 mg, 1.73 mmol) in anhydrous DMF (10 mL) were

added at 0 °C trimethylsulfonium iodide (530 mg, 2.60 mmol) and

t-BuOK (292 mg, 2.60 mmol), and the resulting mixture was stir-

red at room temperature for 2 h. Subsequently, the reaction was

quenched by addition of water, and the resulting mixture was

6582 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

extracted with diethyl ether. The organic layer was washed with

brine and dried over anhydrous Na₂SO₄, and the solvent was

evaporated in vacuo. The residue was purified by flash column

chromatography (elution solvent: petroleum ether, 40-60 °C/

acetone (8:2)) to obtain the desired epoxide 20 (310 mg, 59%) as

a white crystalline solid: mp 170-173 °C; [R]²⁰_D= -72.80° (c =

0.00125 g/mL, CHCl₃); ¹H NMR (CDCl₃) δ 0.88 (s, 3H, 19-CH₃),

1.00 (s, 3H, 18-CH₃), 1.02-2.28 (m, 19H), 2.59 (d, J = 4.88 Hz,

1H), 2.89 (d, J = 4.88 Hz, 1H), 3.47-3.54 (m, 1H, H-3a), 5.35

(d, J = 5.49 Hz, 1H, H-6);¹³C NMR (CDCl₃) δ 14.1, 19.4, 20.4,

23.6, 29.0, 31.3, 31.5, 32.0, 33.8, 36.6, 37.2, 39.9, 42.2, 50.1, 53.1,

53.6, 70.5, 71.6, 121.2, 140.9. Anal. (C₂₀H₃₀O₂) C, H.

3.76 (dd, J=4.0, 12.0 Hz, 1H, H-21), 5.35 (d, J=5.10 Hz, 1H,

H-6). Anal. (C₂₁H₃₂O₃) C, H.

(E)-(3β-Hydroxy-5-androsten-17-ylidene)acetonitrile (24). To

a solution of t-BuOK (0.93 g, 8.28 mmol) in dry THF (10 mL) at

0 °C was added diethyl cyanomethylphosphonate (0.87 mL, 5.52

mmol), and the reaction was stirred for 1 h. To the above

mixture was added dropwise a solution of DHEA (0.4 g, 1.38

mmol) in dry THF (10 mL), and the mixture was stirred at room

temperature until completion of the reaction. The reaction was

quenched by addition of saturated aqueous NH₄Cl and was

extracted with ethyl acetate, the organic layer was washed with

brine and was dried over anhydrous Na₂SO₄, and the solvent

was evaporated in vacuo. The residue was purified by flash

column chromatography (elution solvent: cyclohexane/ethyl

acetate (8:2)) to afford the desired compound 24 (0.28 gr,

65%): mp 164-167 °C; [R]²⁰_D=-60.10° (c = 0.0099 g/mL,

17r-Vinyl-5-androstene-3β,17β-diol (21).

A solution of

DHEA (1.3 g, 4.5 mmol) in anhydrous THF (13 mL) was added

dropwise at 0 °C to a solution of vinylmagnesium bromide

(1.0 M in THF, 45 mL, 45 mmol), and the resulting mixture was

stirred at room temperature for 12 h. After completion of the

reaction saturated aqueous NH₄Cl was added, and the resulting

mixture was extracted with ethyl acetate. The organic layer was

washed with brine and dried over anhydrous Na₂SO₄, and the

solvent was evaporated in vacuo. The residue was purified by

flash column chromatography (elution solvent: cyclohexane/

AcOEt (7:3)) to obtain 17R-vinyl-5-androstene-3β,17β-diol (21)

(992 mg, 69%) as a white crystalline solid: mp 180-183 °C; ¹H

NMR (CDCl₃) δ 0.93 (s, 3H), 1.02 (s, 3H), 1.10-2.31 (m, 19H),

3.47-3.56 (m, 1H, H-3R), 5.10 (dd, J=0.90, 10.80 Hz, 1H), 5.15

(dd, J=0.90, 17.7 Hz, 1H), 5.34 (d, J=4.92 Hz, 1H), 6.04 (dd,

J = 10.80, 17.3 Hz, 1H); ¹³C NMR (CDCl₃) δ 14.0, 19.4, 20.6,

23.6, 31.5, 31.6, 32.1, 32.6, 36.0, 36.5, 37.2, 42.2, 46.1, 49.9, 50.3,

71.6, 84.2, 111.9, 121.3, 140.8, 143.02. Anal. (C₂₁H₃₂O₂) C, H.

3β,17β-Dihydroxy-20,21-epoxy-5-androstene (22). To a solu-

tion of compound 21 (0.992 g, 3.13 mmol) in anhydrous

dichloromethane (33 mL) were sequentially added at -10 °C

vanadyl acetylacetonate (36 mg, 0.137 mmol) and tert-butyl

hydroperoxide (∼2.35 M in 1,2-dichloroethane, prepared by

extracting tert-butyl hydroperoxide (70%, 38.8 mL) with 65.9 mL

of 1,2-dichloroethane) (4 mL, 9.39 mmol), and the resulting

mixture was stirred at room temperature for 12 h. The reaction

mixture was diluted with dichloromethane, the organic layer

was extracted with water, saturated Na₂SO₃, and brine and was

dried over anhydrous Na₂SO₄, and the solvent was evaporated

in vacuo. The residue was purified by flash column chromato-

graphy (elution solvent: dichloromethane/ethyl acetate (6:1))

to obtain 3β,17β-dihydroxy-20,21-epoxy-5-androstene (22) (2:1

mixture of 22a and 22b) (0.71 g, 68%) as a white crystalline solid:

mp 165-168 °C; [R]²⁰_D=-53.10° (c=0.00113 g/mL, CHCl₃);

¹H NMR (CDCl₃) δ 0.88 (s, 0.99H, -CH₃compound 22b), 0.93

(s, 1.99H, -CH₃compound 22a), 1.02 (s, 3H, CH₃), 0.96-2.32

(m, 19H), 2.69 (dd, J = 4.2, 5.1 Hz, 0.33H compound 22b),

2.74-2.77 (m, 1H), 2.87 (dd, J = 2.97, 4.80 Hz, 0.66H com-

pound 22a), 3.09 (t, J = 4.27 Hz, 0.66H compound 22a), 3.21

(dd, J=3.0, 3.90 Hz, 0.33H compound 22b), 3.45-3.55 (m, 1H),

5.34 (d, J=5.13 Hz, 1H, H-6); ¹³C NMR (CDCl₃) δ 13.9, 19.4,

20.5, 24.0, 31.6, 32.4, 36.0, 36.6, 37.3, 42.2, 43.2, 45.5, 50.1, 51.4,

51.8, 54.8, 56.2, 71.7, 79.7, 121.2, 140.8. Anal. (C₂₁H₃₂O₃) C, H.

(20S)-3β,21-Dihydroxy-17β,20-epoxy-5-pregnene (23). To a

solution of 3β,17β-dihydroxy-20,21-epoxy-5-androstene (22)

(0.71 g, 2.13 mmol) in anhydrous MeOH (35 mL) was added

K₂CO₃(0.737 g, 5.33 mmol), and the resulting mixture was

stirred at room temperature for 12 h. The reaction mixture was

diluted with chloroform, and the organic layer was extracted

with water and brine and was dried over anhydrous Na₂SO₄.

The solvent was evaporated in vacuo, and the residue was puri-

fied by flash column chromatography (elution solvent: petro-

leum ether/acetone (85:15)) to obtain (20S)-3β,21-dihydroxy-17β,

20-epoxy-5-androstene (23) (0.35 g, 48%) as a white crystalline

1

CHCl₃); H NMR (CDCl₃) δ 0.85 (s, 3H, CH₃), 1.02 (s, 3H,

CH₃), 1.03-2.33 (m, 17H), 2.58-2.79 (m, 2H), 3.49-3.56

(m, 1H, 3R-H), 5.00 (t, J = 2.46 Hz, 1H, 20-CH), 5.36 (d, J =

5.12 Hz, 1H, H-6); ¹³C NMR (CDCl₃) δ 17.8, 19.4, 20.7, 23.9,

26.9, 30.3, 31.5, 31.5, 31.6, 34.6, 36.6, 37.2, 42.2, 45.9, 49.9, 54.1,

71.6, 87.9, 121.1, 140.8, 180.8. Anal. (C₂₁H₂₉NO) C, H.

(E)-(3β-Hydroxy-5-androsten-17-ylidene)acetaldehyde (25). To

a solution of (E)-3β-hydroxy-5-androsten-17-ylidene)acetonitrile

(24) (150 mg, 0.48 mmol) in dry dichloromethane (15 mL) was

added at -78 °C a solution of DIBAL-H (1 M in dichloromethane,

1.44 mmol), and the reaction mixture was stirred for 30 min at

-78 °C and for an additional 5 h at room temperature. The

reaction mixture was diluted with dichloromethane, and a satu-

rated aqueous solution of sodium potassium tartrate was added.

The organic layer was extracted with brine and was dried over

anhydrous Na₂SO₄, and the solvent was evaporated in vacuo to

afford (E)-(3β-hydroxy-5-androstene-17-ylidene)acetaldehyde (25),

which was pure enough to be used as such in the next step: yield

151 mg, 93%; mp 169-172 °C; ¹H NMR (CDCl₃) δ 0.88 (s, 3H,

CH₃), 1.04 (s, 3H, CH₃), 1.08-2.33 (m, 17H), 2.78-3.01 (m, 2H),

3.47-3.57 (m, 1H, H-3R), 5.37 (d, J=5.10 Hz, 1H, H-6), 5.75 (d,

J=8.10 Hz, 1H, H-20), 9.87 (d, J=7.80 Hz, 1H, CHO); ¹³CNMR

(CDCl₃) δ 17.9, 19.4, 20.8, 24.3, 27.7, 31.58, 31.63, 34.8, 36.6, 37.2,

42.2, 46.3, 50.2, 53.4, 71.6, 119.5, 121.1, 140.8, 180.0, 192.3. Anal.

(C₂₁H₃₀O₂) C, H.

(E)-17-(2-Hydroxyethylidene)-5-androsten-3β-ol (26). To a

solution of (E)-3β-hydroxy-5-androsten-17-ylidene)acetal-

dehyde (25) (75 mg, 0.24 mmol) in MeOH (2.5 mL) were sequen-

tially added CeCl₃7H₂O (178 mg, 0.48 mmol) and NaBH₄

3

(20 mg, 0.48 mmol). After completion of the reaction saturated

aqueous NH₄Cl was added until pH 7. The mixture was diluted

with ethyl acetate, the organic layer was extracted with brine and

was dried over anhydrous Na₂SO₄, and the solvent was evapo-

rated in vacuo. The residue was purified by flash column chro-

matography (elution solvent: petroleum ether, 40-60 °C/

acetone (8:2)) to afford alcohol 26 (70 mg, 92%): mp 128-

131 °C; ¹H NMR (CDCl₃) δ 0.78 (s, 3H, CH₃), 1.02 (s, 3H, CH₃),

1.08-2.38 (m, 19H), 3.49-3.53 (m, 1H, H-3R), 4.05-4.19

(m, 2H, CH₂OH), 5.23-5.27 (m, 1H, H-20), 5.36 (d, J=5.18 Hz,

1H, H-6); ¹³C NMR (CDCl₃) δ 18.8, 19.7, 21.2, 24.6, 26.4, 29.9,

31.8, 31.9, 35.9, 36.9, 37.5, 42.5, 44.1, 50.7, 54.7, 60.6, 71.9,

115.8, 121.7, 141.1, 155.9. Anal. (C₂₁H₃₂O₂) C, H.

(20R)-3β,21-Dihydroxy-17r,20-epoxy-5-pregnene (27). To a

solution of compound 26 (70 mg, 0.22 mmol) in dry dichloro-

methane (2.2 mL) were added at 0 °C K₂CO₃(36 mg, 0.26 mmol)

and m-CPBA 55% (61 mg, 0.22 mmol), and the mixture was

stirred at 0 °C for 2 h. The solid was filtered off, and the filtrate

was evaporated in vacuo and was purified by flash column chro-

matography (elution solvent: hexane/acetone/25% NH₄OH

(5/3/0.1)) to afford epoxide 27 (29 mg, 40%): mp 138-141 °C;

[R]²⁰_D= -77.58° (c = 0.0097/mL, MeOH); ¹H NMR (CDCl₃)

δ 0.82 (s, 3H, CH₃), 1.02 (s, 3H, CH₃), 0.99-2.34 (m, 19H), 3.05

(dd, J=3.6, 6.8 Hz, 1H, 20-CH), 3.54-3.60 (m, 1H, 3R-H), 3.60

(m, 1H, 21-CH), 3.85 (ddd, J = 3.5, 7.0, 11.5 Hz, 1H, 21-CH),

solid: mp 195-197 °C; [R]²⁰= -70.00° (c = 0.0009 g/mL,

D

1

CHCl₃); H NMR (CDCl₃) δ 0.87 (s, 3H, CH₃), 1.00 (s, 3H,

CH₃), 1.03-2.31 (m, 19H), 3.19 (dd, J=4.1, 6.2 Hz, 1H, H-20),

3.48-3.54 (m, 1H, H-3R), 3.58 (dd, J=6.3, 12.0 Hz, 1H, H-21),

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6583

5.37 (d, J=5.1 Hz, 1H, 6-CH); ¹³C NMR δ 15.9, 19.4, 20.5, 24.0,

27.5, 30.4, 31.5, 31.9, 36.6, 37.2, 41.9, 42.2, 50.0, 52.5, 57.3, 62.7,

71.7, 73.9, 121.3, 140.8. Anal. (C₂₁H₃₂O₂) C, H.

100 μL of 2 μg/mL annexin V-FITC was added in a staining

buffer (10 mM HEPES, 140 mM NaCl, 5 mM KCl, 1 mM

MgCl₂, 2.5 mM CaCl₂, pH 7.4) and incubated for 10 min in the

dark. Then, 1 μg per tube of propidium iodide was added, and

cells were analyzed within 20 min by flow cytometry using a

Beckton-Dickinson FACSArray apparatus (Beckton-Dickinson,

Franklin Lakes, NJ) and analyzed with the CELLQuest

(Beckton-Dickinson) and ModFit LT (Verify Software, Topsham,

MN) software.

Molecular Modeling and Conformational Analysis. Computer

calculations were performed on a Silicon Graphics O2 work-

station using QUANTA 97 software purchased from MSI,

applying the CHARMm energy function and force field. The

initial structures were energy minimized using steepest descents,

conjugate gradient and Powell algorithms. The minimized struc-

tures were subjected to grid scan search around torsion angles τ₁

and τ₂from 0° to 360°, applying a torsion step variation of 10°.

The derived conformations were optimized by applying the

steepest descents algorithm with 2000 steps and an energy

gradient tolerance of 0.001 kcal mol^-1A^-1as convergence

criterion. The dielectric constant was set to ε = 1 to simulate

CDCl₃environment. Visualization of the 3D structures has been

enabled by the use of Accelrys DS visualizer s/w.

Displacement of Membrane Binding of [³H]DHEA with Syn-

thetic Spiro-Neurosteroids. PC12 cells were cultured in 225 cm

flasks, and after being washed twice with PBS, they were

detached from the flasks by vigorous shaking. After a centrifu-

gation at 1500g, they were homogenized by sonication in a

50 mM Tris-HCl buffer, pH 7.4 at 4 °C, containing freshly added

protease inhibitors (1 mM PMSF and 1 μg/mL aprotinin).

Unbroken cells were removed by centrifugation at 1500g (10 min

at 4 °C), and membranes were collected by centrifuging at

102000g for 1 h at 4 °C. Membranes were washed once with

Tris-HCl, briefly acidified at 4 °C (for 3 min) with 50 mM glycine

(pH 5.0), and resuspended in the same buffer. Protein content

was assayed by the method of Bradford using reagents from Bio-

Rad (Hercules, CA).

Membranes (at a final concentration 2 mg/mL) were incu-

bated with 5 nM [³H]DHEA in the absence or the presence

of DHEA or the new spiro-neurosteroids at concentrations

varying from 10^-12to 10^-6M, at a final volume of 100 μL, in

Tris-HCl buffer (50 mM, pH 7.4). After a 30 min incubation in a

water bath at 37 °C, membranes were collected on GF/B filters,

prewet in 0.5% PEI solution at 4 °C. The filters were washed

three times with ice-cold Tris-HCl, dried, and counted in a

β-scintillation counter (PerkinElmer, Foster City, CA) with

60% efficiency for tritium. All experiments were performed in

triplicates.

Western Blot Analysis. PC12 cells were harvested and lysed in

lysis buffer supplemented with freshly added proteinase inhibi-

tors phenylmethanesulfonyl fluoride (10 μg/mL) and 1 μg/mL

aprotinin for the detection of Bcl-2 proteins (antibodies from

Cell Signaling Technologies, Beverly, MA). Membranes were

processed according to standard Western blotting procedures.

To detect protein levels, membranes were incubated with the

appropriate antibodies. Anti-actin monoclonal antibody was

supplied by Chemicon (Temecula, CA). Secondary antibodies

used were horseradish peroxidase-conjugated anti-mouse IgG

(Chemicon) and horseradish peroxidase-conjugated anti-rabbit

IgG (Immunotech, France). Then, the membranes were exposed

to Kodak X-Omat AR films. A PC-based Image Analysis

program was used to quantify the intensity of each band

(Image Analysis, Inc., Ontario, Canada).

Measurement of Dopamine. PC12 cells were grown in six-well

plates coated with poly(^L-lysine) at a concentration of 10⁶cells per

well. Cells were incubated with neurosteroids for several time

periods; for the short-term experiments the incubation time ranged

from 5 to 30 min and for the long-term experiments from 3 to 48 h.

One milliliter of supernatants was transferred to tubes contain-

ing 200 μL of 0.1 M HCI for measurement of catecholamines.

It should be noted that dopamine is stable in acidic environment.

Dopamine was measured by radioimmunoassay (TriCatRIA,

RE29395; IBL Immuno Biological Lab, Hamburg, Germany)

using ¹²⁵I as tracer. Dopamine was extracted by a cis-diol-specific

substrate, converted to 3-methoxytyramine using COMT (freshly

prepared every time) as enzyme and S-adenosyl-^L-methionine as

coenzyme, and simultaneously acylated to N-acyl-3-methoxytyr-

amine. After separation of the bound from the free labeled antigen

by precipitation and centrifugation (bound ¹²⁵I-labeled antigen

was precipitated with a second antibody), the amount of bound

radioactivity of the precipitates was measured in a gamma counter

(1275 minigamma; LKB Wallac). The analytical sensitivity of the

method was 30 pg/mL for dopamine, its intraassay CV was 9.5%,

and its interassay CV was 16.7%.

NMR Spectroscopy. NMR spectra were recorded on a Varian

600 MHz spectrometer. The sample concentration was approxi-

mately 10 mM. Compounds were dissolved in CDCl₃. The

1

DQF-COSY, H-¹³C-edited HSQC, and H-¹³C HMBC, all

run with the use of field gradients, facilitated the assignment.

The optimum mixing time for the NOESY experiment was

found to be 150 ms. Data processing, including sine-bell apodi-

zation, Fourier transformation, and phasing, were performed

with the use of VNMR routines.

Cell Cultures. MCF-7 (human breast adenocarcinoma) cells

and LNCaP (human prostate adenocarcinoma) cells were ob-

tained from ATCC (American Tissue Culture Collection),

whereas Ishikawa (human endometrial adenocarcinoma) cells

were obtained from ECACC (European Collection of Cell

Cultures). All cells were maintained as recommended by the

supplier. If not stated otherwise, fine chemicals and culture

media were purchased from Sigma-Aldrich, and sera were from

Gibco (Invitrogen). ICI 182,780 was purchased from Tocris

Bioscience.

PC12 cells were obtained from Dr. M. Greenberg (Children’s

Hospital, Boston, MA). Cells were maintained in culture at 5%

CO₂, at 37 °C, in RPMI 1640 containing 2 mM L-glutamine,

15 mM HEPES, 100 units/mL penicillin, 0.1 mg/mL strepto-

mycin, 10% horse serum, and 5% fetal calf serum (horse serum

and fetal calf serum were charcoal-treated for removing endo-

genous steroids). Serum-free medium was supplemented with

0.1% bovine serum albumin (BSA) and 10 nM sigma1 receptor

specific antagonist haloperidol to avoid potential interactions of

neurosteroids with sigma1 receptors. Conjugate DHEA-BSA

was initially diluted in phosphate-buffered saline (PBS). Prior to

experiments, stock solutions of DHEA-BSA were mixed with

3% charcoal-0.3% dextran for 30 min, centrifuged at 3000g for

10 min, and finally passed through a 0.22 μm filter in order to

remove any free DHEA. In all experiments cells were incubated

with neurosteroids diluted in serum-free medium.

Measurement of Apoptosis. Apoptosis in PC12 cells was

assayed using two different methods.

(a) The APOPercentage apoptosis assay (Biocolor Ltd., Belfast,

N. Ireland) was used to quantify apoptosis, according to the

manufacturer’s instructions. The assay uses a dye that stains red

the apoptotic cells undergoing the membrane “flip-flop” event

when phosphatidylserine is translocated to the outer leaflet.

Detection of apoptosis can be readily observed under inverted

microscopy. For apoptosis quantitation, the amount of dye

within the labeled cells can subsequently be released into solu-

tion, and the concentration is measured at a wavelength of

550 nm (reference filter 620 nm) using a color filter microplate

colorimeter (Dynatech MicroElisa reader, Chantilly, VA).

Apoptosis was measured at 24 h.

(b) Flow cytometry (FACS) of annexin V-propidium iodide-

stained cells was also used. PC12 cells were transferred to a

staining tube and washed with 4 mL of PBS, containing 1%

BSA, at 4 °C. After medium removal (200 rpm, 10 min, 4 °C),

6584 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

RT-PCR. Total RNA was extracted from PC12 cells using

Trizol reagent (Invitrogen Life Technologies, CA). One micro-

gram of total RNA was reverse transcribed by the Thermo-

Script RT-PCR system (Invitrogen) using random hexamers in a

total volume of 20 μL. Two microliters of the RT product was

used as a template, amplified by PCR using 2 mM MgCl₂, one

strength PCR buffer, 0.2 mM sense and antisense primers, 0.2 mM

dNTPs, and 2.5 units of AmpliTaq Gold DNA polymerase

(Perkin-Elmer ABD, Foster City, CA) in a final reaction volume

of 50 μL. PCR was performed in a Perkin-Elmer DNA thermal

cycler. Primers for TH were 5⁰-TCGCCACAGCCCAAGGGCTT-

CAGAA-3⁰(sense) and 5-CCTCGAAGCGCACAAAATAC-3

(antisense) and for GAPDH were 5⁰-TGAAGGTCGGAGT-

CAACGGATTTGGT-3⁰(sense) and 5⁰-CATGTGGGCCAT-

GAGGTCCACCAC-3⁰(antisense). Oligonucleotides were syn-

thesized by MWG-Biotech AG (Munchen, Germany). After

reverse transcription, the cDNA product was amplified by PCR

at 33 cycles. The cycle number (33) was chosen such that

amplification of the products was in the linear range with respect

to the amount of input cDNA. PCR for GAPDH was performed

in parallel to ensure good quality of RNA and cDNA prepara-

tions. Each cycle consisted of 60 s at 92 °C for denaturation,

120 s at 53 °C for annealing, and 180 s at 72 °C for extension

(60 s at 98 °C, 90 s at 55 °C, and 150 s at 72 °C for GAPDH,

respectively). Ten microliters of the amplified products (368 bp

for TH and 983 bp for GAPDH) was separated on a 2% agarose

gel and visualized by ethidium bromide staining.

Effects on the Growth of Endocrine Cancer Cells. The effect of

test compounds on the growth of MCF-7 cells was assessed as

recently described⁵⁴with minor modifications. Briefly, the cells

were plated in 96-flat-bottomed-well microplates at a density of

8000 cells per well in phenol red-free MEM supplemented with

1 μg/mL insulin and 5% DCC-FBS. After plating (72 h), the

cells were exposed for 96 h to test compounds, and relative

numbers of viable cells were determined by monitoring MTT

[3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide]

conversion to formazan at 550 nm. Cells exposed only to vehicle,

1 μM ICI 182,780, and/or 0.1 nM estradiol served as controls.

Statistically significant effects of the test compounds on cell

growth were classified as full, partial, or weak as described in the

legend to Table 2.

The effect of test compounds on the growth of LNCaP cells

was assessed as described above, except that cells were plated in

polylysine-coated 96-flat-bottomed-well microplates at a den-

sity of 5000 cells per well in phenol red-free RPMI supplemented

with 10% DCC-FBS. After plating (72 h), the cells were exposed

for 96 h to test compounds, and the effects of test compounds on

cell growth, as assessed using MTT, were classified as described

above. Cells exposed only to vehicle, 10 μM ICI 182,780, and/or

10 nM estradiol served as controls.

The effect of test compounds on the growth of Ishikawa cells

was determined as described for MCF-7 cells, except that test

compounds were added to the cells 24 h after plating, and the

effects of test compounds on cell growth were assessed and

classified as above.

Effects on the Expression of Estrogen Receptor- (ER-) Regu-

lated Genes. The neurosteroids were assessed for induction of

ERE-dependent gene expression through ERR and ERβ using

MCF-7:D5L and HEK:ERβ cells, respectively. MCF-7:D5L cells,

an ERR-expressing clone of MCF-7 cells that is stably transfected

with the estrogen-responsive reporter plasmid pERE-Gl-lucifer-

ase, have been described.⁴²HEK:ERβ cells, a clone of human

embryonic kidney (HEK293) cells that is stably transfected with

the estrogen-responsive reporter plasmid pERE-Tk-luciferase and

an expression plasmid coding for the full-length human ERβ, have

also been described.⁴⁵Assessment of neurosteroid regulation of

luciferase expression in MCF-7:D5L and HEK:ERβ cells was

carried out as already described^42,45with minor modifications.

Briefly, the cells were plated in 96-well microculture plates at a

density of 12000 cells per well in DMEM (Dulbecco’s modified

minimum essential medium) devoid of phenol red and supplemen-

ted with 1 μg/mL insulin and 10% DCC-FBS, i.e., fetal bovine

serum (FBS) that was treated with 10% dextran-coated charcoal

(DCC) to remove endogenous steroids as already described.⁵³

After plating (72 h), the cells were exposed to increasing concen-

trations of test compounds (stock solutions were prepared using

tissue culture grade DMSO as vehicle) in the absence (vehicle was

added in this case up to a final concentration e0.2%) or presence

of 1 nM estradiol or 1 μM ICI 182,780 for 16 h, and the induction

of luciferase was assessed using the Steady-Glo luciferase assay

system (Promega). Controls for full agonism (cells exposed only to

1 nM estradiol), full antagonism (exposed to 1 μM ICI 182,780 as

well as to 1 nM estradiol), and nonagonism/antagonism (exposed

only to vehicle) served to classify the test compounds that signi-

ficantly affected luciferase expression as full, partial, or weak ERR

or ERβ agonists or antagonists as described in the legend to

Table 1.

MTT Assay for Cell Viability. PC12 (rat adrenal pheochromo-

cytoma), HEK293 (human embryonic kidney), or HT22 (mouse

hippocampal neuroblastoma) cells were seeded in 96-well plates

at a density of 2 ꢀ 10⁴cells/mL and maintained for 24 h in

RPMI 1640 supplemented with 10% horse serum and 5% fetal

bovine serum (PC12) or DMEM supplemented with 10% fetal

bovine serum (HEK293 and HT22). They were then incu-

bated for various time periods (24-72 h) in the absence or

presence of various concentrations of compounds 20, 23, and 27

(0.1-10 μM). At the end of incubation cells were treated with

5 mg/mL MTT for 4 h at 37 °C, then the medium was removed,

and the cells were dissolved with dimethyl sulfoxide. The for-

mazan reduction product was measured by reading the absor-

bance at 570 nm in a plate reader. Results (mean ( SEM of three

replicate values in three independent experiments) are expressed

as percentage of untreated control cultures. This assay is based

on the cellular reduction of a tetrazolium salt (MTT) into a

colored formazan product by mitochondrial enzymes, and

therefore, it is a reliable means to indirectly assess the number

of viable cells.⁵⁵

Statistical Analysis. Results (OD in apoptosis measurements or

arbitrary densitometric units for Western blot proteins norma-

lized to total protein or actin) were expressed as the percentage of

parallel estimates for control cells, treated only with vehicle. Data

were subjected to analysis of variance, post hoc comparison of

means followed by the Fisher’s least significance difference test.

For percent changes we have used the nonparametric Kruskal-

Wallis test for several independent samples. Differences were

considered significant for values of p<0.05.

Acknowledgment. This work was supported in part by the

Greek General Secretariat for Research and Technology

(Programme PENED 2001 ED258) and a grant from Bio-

nature EA Ltd. and EmergoMed Co.

Test neurosteroids were assessed on the alkaline phosphatase

(AlkP) expression of Ishikawa cells using 96-well microcul-

ture plates that were plated with 12000 cells per well in phenol

red-free MEM supplemented with 1 μg/mL insulin and 5%

DCC-FBS. After plating (24 h), cells were exposed to test

compounds in the absence or presence of 0.1 nM estradiol or

1 μM ICI 182,780 for 72 h, and compound effects on AlkP activity

were assessed as recently described.⁴⁶Cells exposed only to vehicle,

ICI 182,780, and/or estradiol served as controls, and those of test

neurosteroids that significantly affected AlkP expression were

classified as agonists or antagonists as described above.

Supporting Information Available: ¹H and ¹³C data, HMBC

correlations, and COSY and NOE interactions for compounds

3, 5, 7, 9, 10, 14, 15, 23, and 27 (Tables 1-9); experimental

procedure for the synthesis of 3R-n-butyl-17,17-ethyleno-

dioxy-5-androstene-3β-ol and spectroscopic data thereof

(Table 10); conformational minima of compounds 3, 5, 7, 9,

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6585

10, 14, and 15 (Figure 1); ¹H NMR spectra of compounds 3, 5,

7, 9, 10, 14, 15, 23, 27, and 28; elemental analysis data for

compounds 1-3, 5, 7, 9, 11, and 14-27. This material is

available free of charge via the Internet at http://pubs.acs.org.

basic protein expression in organotypic slice cultures of rat cerebellum.

J. Neurochem. 2003, 86, 848–859. (d) Ghoumari, A. M.; Baulieu, E. E.;

Schumacher, M. Progesterone increases oligodendroglial cell prolifera-

tion in rat cerebellar slice cultures. Neuroscience 2005, 135, 47–58.

(e) Le Goascogne, C.; Eychenne, B.; Tonon, M. C.; Lachapelle, F.;

Baumann, N.; Robel, P. Neurosteroid progesterone is up-regulated

in the brain of jimpy and shiverer mice. Glia 2000, 29, 14–24.

(f) Schumacher, M.; Akwa, Y.; Guennoun, R.; Robert, F.; Labombarda,

F.; Desarnaud, F.; Robel, P.; De Nicola, A. F.; Baulieu, E. E. Steroid

synthesis and metabolism in the nervous system: Trophic and protec-

tive effects. J. Neurocytol. 2000, 29, 307–326. (g) Schumacher, M.;

Weill-Engerer, S.; Liere, P.; Robert, F.; Franklin, R. J.; Garcia-Segura,

L. M.; Lambert, J. J.; Mayo, W.; Melcangi, R. C.; Parducz, A.; Suter, U.;

Carelli, C.; Baulieu, E. E.; Akwa, Y. Steroid hormones and neuroster-

oids in normal and pathological aging of the nervous system. Prog.

Neurobiol. 2003, 71, 3–29. (h) Schumacher, M.; Guennoun, R.; Robert,

F.; Carelli, C.; Gago, N.; Ghoumari, A.; Gonzalez Deniselle, M. C.;

Gonzalez, S. L.; Ibanez, C.; Labombarda, F.; Coirini, H.; Baulieu, E. E.;

De Nicola, A. F. Local synthesis and dual actions of progesterone in the

nervous system: Neuroprotection and myelination. Growth Horm. IGF

Res. 2004, 14 (Suppl. A), S18-S33.

References

(1) Mattson, M. Apoptosis in neurodegenerative disorders. Nat. Rev.

Mol. Cell. Biol. 2000, 1, 120–129.

(2) Kratnic, S.; Mechawar, N.; Reix, S.; Quirion, R. Molecular basis of

programmed cell death involved in neurodegeneration. Trends

Neurosci. 2006, 28, 670–676.

(3) Forman, M. S.; Trojanowski, J.; Lee, V. Neurodegenerative dis-

eases: A decade of discoveries paves the way for therapeutic

breakthroughs. Nat. Med. 2004, 10, 1055–1063.

(4) (a) Paul, S. M.; Purdy, R. H. Neuroactive steroids. FASEB J. 1992,

6, 2311–2322. (b) Robel, R.; Baulieu, E. E. Neurosteroids biosynthesis

and function. Trends Endocrinol. Met. 1994, 5, 1–8. (c) Kulkarni,

S. K.; Reddy, D. S. Neurosteroids: A new class of neuromodulators.

Drugs Today 1995, 31, 433–455. (d) Compagnone, N. A.; Mellon,

S. H. Neurosteroids: Biosynthesis and Function of These Novel

Neuromodulators. Front. Neuroendocrinol. 2000, 21, 1–56.

(9) Wang, J. M.; Johnston, P. B.; Ball, B. G.; Brinton, R. D. The

neurosteroid allopregnanolone promotes proliferation of rodent

and human neural progenitor cells and regulates cell-cycle gene and

protein expression. J. Neurosci. 2005, 25, 4706–4718.

(5) Brinton, R. D.; Wang, J. M. Therapeutic potential of neurogenesis

for prevention and recovery from Alzheimer’s disease: Allopreg-

nanolone as a proof of concept neurogenic agent. Curr. Alzheimer

Res. 2006, 3, 185–190.

(10) Keller, E. A.; Zamparini, A.; Borodinsky, L. N.; Gravielle, M. C.;

Fiszman, M. L. Role of allopregnanolone on cerebellar granule

cells neurogenesis. Brain Res. Dev. Brain Res. 2004, 153, 13–17.

(11) (a) Pathirathna, S.; Brimelow, B. C.; Jagodic, M. M.; Krishnan,

K.; Jiang, X.; Zorumski, C. F.; Mennerick, S.; Covey, D. F.;

Todorovic, S. M.; Jevtovic-Todorovic, V. New evidence that both

T-type calcium channels and GABA_Achannels are responsible for

the potent peripheral analgesic effects of 5alpha-reduced neuro-

active steroids. Pain 2005, 114, 429–443. (b) Todorovic, S. M.;

Pathirathna, S.; Brimelow, B. C.; Jagodic, M. M.; Ko, S. H.; Jiang,

X.; Nilsson, K. R.; Zorumski, C. F.; Covey, D. F.; Jevtovic-Todorovic,

V. 5beta-reduced neuroactive steroids are novel voltage-dependent

blockers of T-type Ca^2þchannels in rat sensory neurons in vitro and potent

peripheral analgesics in vivo. Mol. Pharmacol. 2004, 66, 1223–1235.

(12) Weill-Engerer, S.; David, J. P.; Sazdovitch, V.; Liere, P.; Eychenne,

B.; Pianos, A.; Schumacher, M.; Delacourte, A.; Baulieu, E. E.;

Akwa, Y. Neurosteroid quantification in human brain regions:

comparison between Alzheimer’s and nondemented patients.

J. Clin. Endocrinol. Metab. 2002, 87, 5138–5143.

(13) Schumacher, M.; Weill-Engerer, S.; Liere, P.; Robert, F.; Franklin,

R. J.; Garcia-Segura, L. M.; Lambert, J. J.; Mayo, W.; Melcangi,

R. C.; Parducz, A.; Suter, U.; Carelli, C.; Baulieu, E. E.; Akwa, Y.

Steroid hormones and neurosteroids in normal and pathological

aging of the nervous system. Prog. Neurobiol. 2003, 71, 3–29.

(14) Baulieu, E. E.; Robel, P. Dehydroepiandrosterone (DHEA) and

dehydroepiandrosterone sulfate (DHEAS) as neuroactive neuro-

steroids. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 4089–4091.

(15) Compagnone, N. A.; Mellon, S. H. Neurosteroids: Biosynthesis

and function of these novel neuromodulators. Front. Neuroendo-

crinol. 2000, 21, 1–56.

(16) (a) Orentreich, N.; Brind, J. L.; Rizer, R. L.; Vogelman, J. H. Age

changes and sex differences in serum dehydroepiandrosterone

sulfate concentrations throughout adulthood. J. Clin. Endocrinol.

Metab. 1984, 59, 551–555. (b) Belanger, N.; Gregoire, L.; Bedard, P. J.;

Di Paolo, T. DHEA improves symptomatic treatment of moderately and

severely impaired MPTP monkeys. Neurobiol. Aging 2006, 27, 1684–

1693. (c) Bologa, L.; Sharma, J.; Roberts, E. Dehydroepiandrosterone

and its sulphated derivative reduce neuronal death and enhance astro-

cytic differentiation in brain cell cultures. J. Neurosci. Res. 1987, 17,

225–234. (d) Straub, R. H.; Lehle, K.; Herfarth, H.; Weber, M.; Falk,

W.; Preuner, J; Scholmerich, J. Dehydroepiandrosterone in relation to

other adrenal hormones during an acute inflammatory stressful disease

state compared with chronic inflammatory disease: Role of interleukin-6

and tumour necrosis factor. Eur. J. Endocrinol. 2002, 146, 365–374.

(e) Sapolsky, R. M. Stress, the aging brain and the mechanism of

neuron death; MIT Press: Cambridge, MA, 1992. (f) Xilouri, M.;

Papazafiri, P. Anti-apoptotic effects of allopregnanolone on P19

neurons. Eur. J. Neurosci. 2006, 23, 43–54. (g) Xilouri, M.; Avlonitis,

N.; Calogeropoulou, C.; Papazafiri, P. Neuroprotective effects of steroid

analogues on P19-N neurons. Neurochem. Int. 2007, 50, 660–670.

(h) Leskiewicz, M.; Regulska, M.; Budziszewska, B.; Jantas, D.;

Jaworska-Feil, L.; Basta-Kaim, A.; Kubera, M.; Jagla, G.; Nowak,

W.; Lason, W. Effects of neurosteroids on hydrogen peroxide- and

staurosporine-induced damage of human neuroblastoma SHSY5Y

cells. J. Neurosci. Res. 1987, 86, 1361–1370.

(6) (a) Cutler, S. M.; Pettus, E. H.; Hoffman, S. W.; Stein, D. G.

Tapered progesterone withdrawal enhances behavioral and mole-

cular recovery after traumatic brain injury. Exp. Neurol. 2005, 195,

423–429. (b) Djebaili, M.; Guo, Q.; Pettus, E. H.; Hoffman, S. W.; Stein,

D. G. The neurosteroids progesterone and allopregnanolone reduce cell

death, gliosis, and functional deficits after traumatic brain injury in rats.

J. Neurotrauma 2005, 22, 106–118. (c) Hoffman, S. W.; Virmani, S.;

Simkins, R. M.; Stein, D. G. The delayed administration of dehydro-

epiandrosterone sulfate improves recovery of function after traumatic

brain injury in rats. J. Neurotrauma 2003, 20, 859–870. (d) Lapchak,

P. A. The neuroactive steroid 3-alpha-ol-5-betapregnan- 20-one hemi-

succinate, a selective NMDA receptor antagonist improves behavioral

performance following spinal cord ischemia. Brain Res. 2004, 997,

152–158. (e) Meffre, D.; Pianos, A.; Liere, P.; Eychenne, B.; Cambourg,

A.; Schumacher, M.; Stein, D. G.; Guennoun, R. Steroid profiling in

brain and plasma of male and pseudopregnant female rats after

traumatic brain injury: Analysis by gas chromatography/mass spectro-

metry. Endocrinology 2007, 148, 2505–2517. (f) Shear, D. A.; Galani,

R.; Hoffman, S. W.; Stein, D.G. Progesterone protects against necrotic

damage and behavioural abnormalities caused by traumatic brain

injury. Exp. Neurol. 2002, 178, 59–67. (g) VanLandingham, J. W.;

Cutler, S. M.; Virmani, S.; Hoffman, S. W.; Covey, D. F.; Krishnan, K.;

Hammes, S. R.; Jamnongjit, M.; Stein, D. G. The enantiomer of

progesterone acts as a molecular neuroprotectant after traumatic brain

injury. Neuropharmacology 2006, 51, 1078–1085.

(7) (a) Di Michele, F.; Lekieffre, D.; Pasini, A.; Bernardi, G.;

Benavides, J.; Romeo, E. Increased neurosteroids synthesis after

brain and spinal cord injury in rats. Neurosci. Lett. 2000, 284, 65–

68. (b) Fiore, C.; Inman, D. M.; Hirose, S.; Noble, L. J.; Igarashi, T.;

Compagnone, N. A. Treatment with the neurosteroid dehydroepian-

drosterone promotes recovery of motor behaviour after moderate

contusive spinal cord injury in the mouse. J. Neurosci. Res. 2004,

75, 391–400. (c) Labombarda, F.; Pianos, A.; Liere, P.; Eychenne, B.;

Gonzalez, S.; Cambourg, A.; De Nicola, A. F.; Schumacher, M.;

Guennoun, R. Injury elicited increase in spinal cord neurosteroid

content analyzed by gas chromatography mass spectrometry. Endocri-

nology 2006, 147, 1847–1859. (d) Patte-Mensah, C.; Penning, T. M.;

Mensah-Nyagan, A. G. Anatomical and cellular localization of

neuroactive 5alpha/3alpha-reduced steroid-synthesizing enzymes in

the spinal cord. J. Comp. Neurol. 2004, 477, 286–299. (e) Pomata, P.

E.; Colman-Lerner, A. A.; Baranao, J. L.; Fiszman, M. L. In vivo

evidences of early neurosteroid synthesis in the developing rat central

nervous system and placenta. Brain Res. Dev. Brain Res. 2000, 120,

83–86.

(8) (a) Chavez-Delgado, M. E.; Gomez-Pinedo, U.; Feria-Velasco, A.;

Huerta-Viera, M.; Castaneda, S. C.; Toral, F. A.; Parducz, A.;

Anda, S. L.; Mora-Galindo, J.; Garcia- Estrada, J. Ultrastructural

analysis of guided nerve regeneration using progesterone- and

pregnenolone-loaded chitosan prostheses. J. Biomed. Mater.

Res., Part B 2005, 74, 589–600. (b) Gago, N.; Akwa, Y.; Sananes,

N.; Guennoun, R.; Baulieu, E. E.; El-Etr, M.; Schumacher, M. Proges-

terone and the oligodendroglial lineage: Stage-dependent biosynthesis

and metabolism. Glia 2001, 36, 295–308. (c) Ghoumari, A. M.; Ibanez,

C.; El-Etr, M.; Leclerc, P.; Eychenne, B.; O'Malley, B. W.; Baulieu, E.

E.; Schumacher, M. Progesterone and its metabolites increase myelin

(17) Marx, C. E.; Jarskog, L. F.; Lauder, J. M.; Gilmore, J. H.;

Lieberman, J. A.; Morrow, A. L. Neurosteroid modulation of

6586 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Calogeropoulou et al.

embryonic neuronal survival in vitro following anoxia. Brain Res.

2000, 871, 104–112.

(18) Gubba, E. M.; Fawcett, J. W.; Herbert, J. The effects of cortico-

sterone and dehydroepiandrosterone on neurotrophic factor mrna

expression in primary hippocampal and astrocyte cultures. Brain

Res. Mol. Brain Res. 2004, 127, 48–59.

J. Chem. Soc., Chem. Commun. 1979, 156–157. (b) Hanessian, S.;

Lavalee, P. The preparation and synthetic utility of tert-butyldiphenyl-

silyl ethers. Can. J. Chem. 1975, 53, 2975–77. (c) Corey, E. J.;

Venkateswarlu, A. Protection of hydroxyl groups as tert-butyldimethyl-

silyl derivatives. J. Am. Chem. Soc. 1972, 94, 6190–6191.

(30) (a) Summit Gal, G.; Sletzinger, M. 20-Spirox-4,6-dien-3-one

aldosterone inhibitors. U.S. Patent 3,492,292, 1970; (b) Gravanis,

A.; Calogeropoulou, T.; Castanas, E.; Margioris, A.; Charalambopoulos,

I.; Avlonitis, N.; Minas, V.; Alexaki, V.-I.; Tsatsanis, C.;; Alexis, M. N.;

Remboutsika, E.; Vergou, V.; Neophytou, C. Neurosteroid compounds.

WO2008155534 (A2), 2008.

(31) Gill, J. C.; Lockey, P. M.; Marples, B. A.; Traynor, J. R. 3,17β-

Dihydroxy-20,21-epoxy-19-norpregna-1,3,5(10)-trienes: Synthesis,

rearrangement, cytotoxicity and estrogen-receptor binding. J. Med.

Chem. 1986, 29, 1537–1540.

(32) Takano, S.; Yamada, S.; Numata, H.; Ogasawara, K. A new

synthesis of a steroid side chain via stereocontrolled protonation:

Synthesis of (-)-Desmosterol. J. Chem. Soc., Chem. Commun.

1983, 760–761.

(19) (a) Charalampopoulos, I.; Remboutsika, E.; Margioris, A.;

Gravanis, A. Neurosteroids as modulators of neurogenesis and

neuronal survival. Trends Endocrinol. Metab. 2008, 19, 300–307.

(b) Loria, R. M.; Padgett, D. A.; Huynh, P. N. Regulation of the immune

response by dehydroepiandrosterone and its metabolites. J. Endocrinol.

1996, 150 (Suppl.), S209-S220. (c) Mclachlan, J. A.; Serkin, C. D.;

Bakouche, O. Dehydroepiandrosterone modulation of lipopolysacchar-

ide-stimulated monocyte cytotoxicity. J. Immunol. 1996, 156, 328–

335. (d) Kimonides, V. G.; Khatibi, N. H.; Svendsen, C. N.; Sofroniew,

M. V.; Herbert, J. Dehydroepiandrosterone (dhea) and dhea-sulfate

(dheas) protect hippocampal neurons against excitatory amino acid-

induced neurotoxicity. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 1852–

1857. (e) Bastianetto, S.; Ramassamy, C.; Poirier, J.; Quirion, R.

Dehydroepiandrosterone (DHEA) protects hippocampal cells from

oxidative stress-induced damage. Brain Res. Mol. Brain Res. 1999,

66, 35–41. (f) Barger, S. W.; Chavis, J. A.; Drew, P. D. Dehydroepian-

drosterone inhibits microglial nitric oxide production in a stimulus-

specific manner. J. Neurosci. Res. 2000, 62, 503–509. (g) Cardounel,

A.; Regelson, W.; Kalimi, M. Dehydroepiandrosterone protects hippo-

campal neurons against neurotoxin-induced cell death: Mechanism of

action. Proc. Soc. Exp. Biol. Med. 1999, 222, 145–149. (h) Lapchak,

P. A.; Chapman, D. F.; Nunez, S. Y.; Zivin, J. A. Dehydroepiandroster-

one sulfate is neuroprotective in a reversible spinal cord ischemia

model: Possible involvement of GABA(A) receptors. Stroke 2000,

31, 1953–1956. (i) Milman, A.; Zohar, S.; Maayan, R.; Weizman, R.;

Pick, C. G. DHEAS repeated treatment improves cognitive and beha-

vioral deficits after mild traumatic brain injury. Eur. Neuropsycho-

pharmacol. 2008, 18, 181–187.

(33) Parker, K. A. Process for the preparation of 17(20)ene-21-steroid

aldehydes U.S. Patent 4,059,575, 1977.

(34) Luche, J.-L. Lanthanides in organic chemistry. 1. Selective 1,2

reductions of conjugated ketones. J. Am. Chem. Soc. 1978, 100,

2226–2227.

(35) Charalampopoulos, I.; Alexaki, V. I.; Lazaridis, I.; Dermitzaki, E.;

Avlonitis, N.; Tsatsanis, C.; Calogeropoulou, T.; Margioris, A. N.;

Castanas, E.; Gravanis, A. G protein-associated, specific mem-

brane binding sites mediate the neuroprotective effect of dehydro-

epiandrosterone. FASEB J. 2006, 20, 577–579.

(36) Charalampopoulos, I.; Tsatsanis, C.; Dermitzaki, E.; Alexaki,

V. I.; Castanas, E.; Margioris, A. N.; Gravanis, A. Dehydroepian-

drosterone and allopregnanolone protect sympathoadrenal medul-

la cells against apoptosis via antiapoptotic Bcl-2 proteins. Proc.

Natl. Acad. Sci. U.S.A. 2004, 101, 8209–8214.

(20) Compagnone, N. A. Treatments for spinal cord injury: Is there

hope in neurosteroids? J. Steroid Biochem. Mol. Biol. 2008, 109,

307–313.

(21) Weill-Engerer, S.; David, J. P.; Sazdovitch, V.; et al. Neurosteroid

quantification in human brain regions: Comparison between

Alzheimer’s and nondemented patients. J. Clin. Endocrinol. Metab.

2002, 87, 5138–5143.

(22) Belanger, N.; Gregoire, L.; Bedard, P.; Di Paolo, T. Estradiol and

dehydroepiandrosterone potentiate levodopa-induced locomotor

activity in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine monkeys.

Endocrine 2003, 21, 97–101.

(23) Charalampopoulos, I.; Dermitzaki, E.; Vardouli, L.; Tsatsanis, C.;

Stournaras, C.; Margioris, A. N.; Gravanis, A. Dehydroepiandro-

sterone sulfate and allopregnanolone directly stimulate catechola-

mine production via induction of tyrosine hydroxylase and secre-

tion by affecting actin polymerization. Endocrinology 2005, 146,

3309–3318.

(37) Henderson, B. E.; Feigelson, H. S. Hormonal carcinogenesis.

Carcinogenesis 2000, 21, 427–433.

(38) Chen, W. Y.; Manson, J. E.; Hankinson, S. E.; Rosner, B.; Holmes,

M. D.; Willett, W. C.; Colditz, G. A. Unopposed estrogen therapy

and the risk of invasive breast cancer. Arch. Intern. Med. 2006, 166,

1027–1032.

(39) Johnson, S. M.; Maleki-Dizaji, M.; Styles, J. A.; White, I. N.

Ishikawa cells exhibit differential gene expression profiles in re-

sponse to oestradiol or 4-hydroxytamoxifen. Endocr. Relat. Cancer

2007, 14, 337–350.

(40) Takahashi, Y.; Hursting, S. D.; Perkin, S. N.; Wang, T. C.; Wang,

T. T. Genistein affects androgen-responsive genes through both

androgen- and estrogen-induced signaling pathways. Mol. Carci-

nog. 2006, 45, 18–25.

(41) Guerini, V.; Sau, D.; Scaccianoce, E.; Rusmini, P.; Ciana, P.;

Maggi, A.; Martini, P. G.; Katzenellenbogen, B. S.; Martini, L.;

Motta, M.; Poletti, A. The androgen derivative 5alpha-androstane-

3beta,17beta-diol inhibits prostate cancer cell migration through

activation of the estrogen receptor beta subtype. Cancer Res. 2005,

65, 5445–53.

(42) Fokialakis, N.; Lambrinidis, G.; Mitsiou, D. J.; Aligiannis, N.;

Mitakou, S.; Skaltsounis, A. L.; Pratsinis, H.; Mikros, E.; Alexis,

M. N. A new class of phytoestrogens; evaluation of the estrogenic

activity of deoxybenzoins. Chem. Biol. 2004, 11, 397–406.

(43) Lazennec, G.; Alcorn, J. L.; Katzenellenbogen, B. S. Adenovirus-

mediated delivery of a dominant negative estrogen receptor gene

abrogates estrogen-stimulated gene expression and breast cancer

cell proliferation. Mol. Endocrinol. 1999, 13, 969–980.

(44) Wittmann, B. M.; Sherk, A.; McDonnell, D. P. Definition of

functionally important mechanistic differences among selective

estrogen receptor down-regulators. Cancer Res. 2007, 67, 9549–

9560.

(45) Skretas, G.; Meligova, A. K.; Villalonga-Barber, C.; Mitsiou, D. J.;

Alexis, M. N.; Micha-Screttas, M.; Steele, B. R.; Screttas, C. G.;

Wood, D. W. Engineered chimeric enzymes as tools for drug

discovery: Generating reliable bacterial screens for the detection,

discovery, and assessment of estrogen receptor modulators. J. Am.

Chem. Soc. 2007, 129, 8443–8457.

(46) Kasiotis, K. M.; Mendorou, C.; Haroutounian, S. A.; Alexis, M. N.

High affinity 17alpha-substituted estradiol derivatives: Synthesis

and evaluation of estrogen receptor agonist activity. Steroids 2006,

71, 249–255.

(47) He, D.; Falany, C. N. Inhibition of SULT2B1b expression alters

effects of 3beta-hydroxysteroids on cell proliferation and steroid

hormone receptor expression in human LNCaP prostate cancer

cells. Prostate 2007, 67, 1318–1329.

(24) Suzuki, M.; Wright, L. S.; Marwah, P.; Lardy, H. A.; Svendsen,

C. N. Mitotic and neurogenic effects of dehydroepiandrosterone

(DHEA) on human neural stem cell cultures derived from the fetal

cortex. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 3202–3207.

(25) Souli, C.; Avlonitis, N.; Calogeropoulou, T.; Tsotinis, A.; Maksay,

ꢀ

G.; Bıro, T.; Politi, A.; Mavromoustakos, T.; Makriyannis, A.;

´

Reis, H.; Papadopoulos, M. Novel 17beta-substituted conforma-

tionally constrained neurosteroids that modulate GABA_Arecep-

tors. J. Med. Chem. 2005, 48, 5203–5214.

ꢀ

(26) Maksay, G.; Fodor, L.; Bıro, T.; Avlonitis, N.; Calogeropoulou,

´

T. A 17beta-derivative of allopregnanolone is a neurosteroid ant-

agonist at a cerebellar subpopulation of GABA_Areceptors with

nanomolar affinity. Br. J. Pharmacol. 2007, 151, 1078–1086.

(27) Chu, G.-H.; Jagannathan, S.; Li, P.-K. Synthesis of 17-oxoandro-

sta-3,5-dien-3-methyl sulfonate as stable analog of dehydroepian-

drosterone sulfate. Steroids 1998, 63, 214–217.

(28) (a) Hogenkamp, D. J.; Tahir, S. H.; Hawkinson, J. E.; Upasani,

R. B.; Alauddin, M.; Kimbrough, C. L.; Acosta-Burruel, M.;

Whittemore, E. R.; Woodward, R. M.; Lan, N. C.; Gee, K. W.;

Bolger, M. B. Synthesis and in vitro activity of 3β-substituted-

3R-hydroxypregna-20-ones: allosteric modulators of the GABA_A

ꢀ

receptor. J. Med. Chem. 1997, 40, 61–72. (b) Tchedam Ngatcha,

B.; Luu-The, V.; Labrie, F.; Poirier, D. Androsterone 3R-ether-

3β-substituted and androsterone 3β-substituted derivatives as inhibitors

of type 3 17β-hydroxysteroid dehydrogenase: Chemical synthesis and

structure-activity relationship. J. Med. Chem. 2005, 48, 5257–5268.

ꢀ

(c) Tchedam Ngatcha, B.; Poirier, D. Diastereoselective addition of

organomagnesium reagents to 17β-TBDMS-dihydrotestosterone.

Synth. Commun. 1999, 29, 1065–1074.

(29) (a) Hart, T. W.; Metcalfe, D. A.; Scheinmann, F. Total synthesis

(48) Maggiolini, M.; Recchia, A. G.; Carpino, A.; Vivacqua, A.;

ꢁ

Fasanella, G.; Rago, V.; Pezzi, V.; Briand, P. A.; Picard, D.; Ando,

of (()-prostagladin D₁: Uuse of triethylsilyl protecting groups.

Article

S. Oestrogen receptor beta is required for androgen-stimulated

proliferation of LNCaP prostate cancer cells. J. Mol. Endocrinol.

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6587

D. A.; Greene, G. L. Structural characterization of a subtype-

selective ligand reveals a novel mode of estrogen receptor antago-

nism. Nat. Struct. Biol. 2002, 9, 359–364.

2004, 32, 777–791.

€

(49) Vihko, P.; Herrala, A.; Harkonen, P.; Isomaa, V.; Kaija, H.;

Kurkela, R.; Pulkka, A. Control of cell proliferation by steroids:

The role of 17HSDs. Mol. Cell. Endocrinol. 2006, 248, 141–148.

(50) Bhacca, N. S.; Fronczek, F. R.; Sygula, A. Investigations of

dehydroepiandrosterone. Part I: Crystal structure of sublimed

DHEA. J. Chem. Crystallogr. 1996, 26, 483–487.

(53) Gritzapis, A. D.; Baxevanis, C. N.; Missitzis, I.; Katsanou, E. S.;

Alexis, M. N.; Yotis, J.; Papamichail, M. Quantitative fluorescence

cytometric measurement of estrogen and progesterone receptors:

correlation with the hormone binding assay. Breast Cancer Res.

Treat. 2003, 80, 1–13.

(54) Katsanou, E. S.; Halabalaki, M.; Aligiannis, N.; Mitakou, S.;

Skaltsounis, A. L.; Alexi, X.; Pratsinis, H.; Alexis, M. N. Cytotoxic

effects of 2-arylbenzofuran phytoestrogens on human cancer cells:

modulation by adrenal and gonadal steroids. J. Steroid Biochem.

Mol. Biol. 2007, 104, 228–236.

(55) Plumb, J. A.; Milroy, R.; Kaye, S. B. Effects of the pH dependence

of 3-(4,5-dimethylthiazol- 2-yl)-2,5-diphenyltetrazolium bromide-

formazan absorption on chemosensitivity determined by a novel

tetrazolium-based assay. Cancer Res. 1989, 49, 4435–4440.

(51) Nettles, K. W.; Bruning, J. B.; Gil, G.; Nowak, J.; Sharma, S. K.;

Hahm, J. B.; Kulp, K.; Hochberg, R. B.; Zhou, H.; Katzenellenbogen,

J. A.; Katzenellenbogen, B. S.; Kim, Y.; Joachmiak, A.; Greene,

G. L. NFkappaB selectivity of estrogen receptor ligands revealed

by comparative crystallographic analyses. Nat. Chem. Biol. 2008, 4,

241–247.

(52) Shiau, A. K.; Barstad, D.; Radek, J. T.; Meyers, M. J.; Nettles,

K. W.; Katzenellenbogen, B. S.; Katzenellenbogen, J. A.; Agard,

Article Doi

Novel dehydroepiandrosterone derivatives with antiapoptotic, neuroprotective activity

DOI: 10.1021/jm900468p

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm900468p

Full text of DOI:10.1021/jm900468p

Products guided by the article

R&D Labs maybe for 1195795-77-7

Relevant to this article

Hot Product