Toxicity of (22R,23R)-22,23-dihydroxystigmastane derivatives to cultured cancer cells

Article abstract of DOI:10.1016/j.steroids.2010.01.006

Toxicity of eight 22,23-dihydroxystigmastane derivatives (four pairs of (22R,23R)- and (22S,23S)-isomers differing in steroid backbone structure) to human breast carcinoma MCF-7 cells was compared. For every pair of structurally related compounds, (22R,23

Full text of DOI:10.1016/j.steroids.2010.01.006

Steroids 75 (2010) 287–294
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Toxicity of (22R,23R)-22,23-dihydroxystigmastane derivatives to cultured
cancer cells
a
a
b
b
Alexander Yu. Misharin , Arif R. Mehtiev , Vladimir N. Zhabinskii , Vladimir A. Khripach ,
Vladimir P. Timofeev , Yaroslav V. Tkachevc
V.N. Orekhovich Institute of Biomedical Chemistry RAMS, Moscow, Russia
Institute of Bioorganic Chemistry NASB, Minsk, Belarus
V.A. Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, 32, Vavilov Street, Moscow 119991, Russia
c,∗
a
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 July 2009
Received in revised form 7 January 2010
Accepted 11 January 2010
Available online 21 January 2010
Toxicity of eight 22,23-dihydroxystigmastane derivatives (four pairs of (22R,23R)- and (22S,23S)-
isomers differing in steroid backbone structure) to human breast carcinoma MCF-7 cells was
compared. For every pair of structurally related compounds, (22R,23R) isomer was found to be
significantly more toxic than (22S,23S) isomer. Computational analysis showed that side chain
of (22R,23R)-22,23-dihydroxystigmastane derivatives is rigid, whereas that of (22S,23S)-isomers is
rather flexible. Structure of steroid backbone significantly affects cytotoxicity of (22R,23R)-22,23-
dihydroxystigmastane derivatives to human breast carcinoma MCF-7 cells, human ovary carcinoma
CaOv cells, and human prostate carcinoma LnCaP cells. (22R,23R)-3␤,22,23-trihydroxystigmast-5-ene
and (22R,23R)-3␤,22,23-trihydroxystigmast-5-en-7-one, both comprising equatorial 3␤-hydroxyl group,
exhibited the highest cytotoxicity, while the most polar 28-homobrassinolide and 28-homocastasterone,
both comprising 2␣,3␣-dihydroxy groups, exhibited the lowest toxicity. Binding of (22R,23R)-
Keywords:
Oxysterols
Brassinosteroids
Side chain conformation
Cytotoxicity
2
2,23-dihydroxystigmastane derivatives to plasmatic membrane was suggested to be important for
cytotoxicity.
©
2010 Elsevier Inc. All rights reserved.
1
. Introduction
A number of side chain oxygenated sterols isolated from
essential for their brassinolide-like activity. Activity of brassino-
lide and castasterone derivatives with varied side chain moieties
in plant derived assays is dependent on the degree of freedom of
the side chain [15,16]. The analysis of conformation–activity rela-
tionship indicated that the presence of C21 methyl group with
a proper stereochemistry dramatically decreases the side chain
conformational flexibility and increases brassinolide-like activity
of castasterone analogs [16]. Stereochemical configuration of C22
and C23 atoms, structure and stereochemical configuration of alkyl
substitutes at C24 affects side chain conformational flexibility,
brassinolide-like activity [17] and may be important for the tox-
icity of brassinolide and castasterone derivatives to several cancer
cell lines [14].
plants and marine organisms are toxic to mammalian cells,
especially to fast proliferating tumor cells. Some of aforemen-
tioned compounds attract attention as potentially antitumor
and anticancer pharmacological agents [1–13]. Specifically, plant
phytohormons brassinosteroids, namely naturally occurring 28-
homocastasterone and synthetic 24-epibrassinolide were recently
found to inhibit cell viability and growth, arrested cells in G phase
of the cell cycle, and induced apoptosis in some cancer cell lines
1
[
14]. At present anticancer and cytotoxic activity of brassinos-
teroids and related 22,23-oxygenated steroids is poorly intelligible
on the molecular level.
It is known that physiological effects of brassinosteroids on plant
growth and improvement of stress tolerance (brassinolide-like
activity), caused by their interactions with specific transmembrane
receptor kinase BRI1, are associated with brassinosteroid structure.
Toxicity of synthetic 22,23-oxygenated stigmastane derivatives
(22R,23R)-3␤,22,23-trihydroxystigmast-5-ene and (22R,23R)-
3␤,22,23-trihydroxystigmast-5-en-7-one
to
human
breast
carcinoma MCF-7 cells was about 10-fold higher compared
with that of 28-homocastasterone and 24-epibrassinolide [14,18].
(22R,23R)-3␤-Bromo-5␣,22,23-trihydroxystigmastan-6-one,
(
22R,23R)-configuration of 22,23-diol group in brassinosteroids is
(
(
22R,23R)-3␤-fluoro-5␣,22,23-trihydroxystigmastan-6-one and
22R,23R)-2␣,3␣,5␣,22,23-pentahydroxystigmastan-6-one exhib-
ited noticeable toxicity towards Vero cells, J774.A1 murine
macrophage cell line, L929 murine fibroblast cell line, and
∗ Corresponding author. Tel.: +7 499 135 9859; fax: +7 499 135 1405.
E-mail address: tim@eimb.ru (V.P. Timofeev).
0
039-128X/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2010.01.006

2
88
A.Yu. Misharin et al. / Steroids 75 (2010) 287–294
Fig. 1. (22,23)-Dihydroxystigmastane derivatives:
1
–
28-homobrassinolide;
(22R,23R)-22,23-dihydroxystigmast-2-en-6-one;
␣,5-cyclo-5␣-stigmastan-6-one; (22S,23S)-22,23-dihydroxy-3␣,5-cyclo-5␣-stigmastan-6-one;
2
–
–
(22S,23S)-28-homobrassinolide;
3
–
28-homocastasterone;
4 – (22S,23S)-
2
8-homocastasterone;
5
–
6
(22S,23S)-22,23-dihydroxystigmast-2-en-6-one; 7 –
(22R,23R)-22,23-dihydroxy-
3
8
–
9
–
(22R,23R)-22,23-dihydroxystigmast-4-en-3,6-dione; 10
–
3
(
(
22R,23R)-22,23-dihydroxystigmast-4-en-3-one; 11 – (22R,23R)-3␤,22,23-trihydroxystigmast-5-ene; 11a – (22R,23R)-[3,4- H2]-3␤,22,23-trihydroxystigmast-5-ene; 12 –
22R,23R)-3␤,22,23-trihydroxystigmast-5-en-7-one.
IOBA-NHC human conjunctive cell line [19–22]. All these data
indicate that toxicity of 22,23-dihydroxy steroids to cultured
mammalian cells is dependent on the number and stereochemical
configuration of polar substitutes in steroid backbone.
stigmasterol according to published procedure [18], based on
regio- and stereoselective Woodward hydroxylation of (22E)-
3␣,5-cyclo-6␤-methoxy-5␣-stigmast-22-ene resulted in major
product comprising required (22R,23R)-configuration.
3
The goal of the present study was the evaluation of effects of
twelve 22,23-dihydroxystigmastane derivatives (formula are given
in Fig. 1) on viability of cultured cancer cells. Specific aims were: (i)
comparison of cytotoxicity of four pairs (22R,23R)- and (22S,23S)-
structurally related compounds (1–8) to MCF-7 cells; (ii) searching
links between cytotoxicity of tested compounds and spatial struc-
ture of their side chain; (iii) selection of compounds exhibiting
highest toxicity to three lines of cancer cells (MCF-7, CaOv, LnCaP)
among (22R,23R)-22,23-dihydroxystigmastane derivatives differ-
ing in steroid backbones.
(22R,23R)-[3,4- H ]-3␤,22,23-trihydroxystigmast-5-ene 11a
2
was synthesized according to procedure [27,28] used ear-
3
lier for preparation of [ H]-labeled 25-hydoxycholesterol and
27-hydroxycholesterol.
(22R,23R)-22,23-Dihydroxystigmast-
4-en-3one 10 (4.4 mg, 10 ␮mol) was treated with 0.5 mL of
the mixture consisting of trimethyl bromosilane, 1,1,1,3,3,3-
hexamethyl disilazane, and pyridine (1:5:4, by vol.) for 20 h at
room temperature. The mixture was extracted with petroleum
ether (3× 3 mL), the extract was evaporated to dryness, the residue
3
was dissolved in isopropanol (1 mL), NaB[ H] (2 mg, 50 ␮mol,
4
8
specific activity 2.5 × 10 cpm/mg) was added, and the mixture
2
. Materials and methods
was stored at a room temperature for 20 h. Then the mixture was
diluted with 2 mL of water, acidified with HCl to pH value of 2.0,
stirred for 10 min, and extracted with ethyl acetate (3× 3 mL).
The extract was dried over Na SO and evaporated, the radio-
2.1. Materials
2
4
3
Construction
of
the
side
chain
of
(22R,23R)-28-
labeled (22R,23R)-[3,4- H2]-3␤,22,23-trihydroxystigmast-5-ene
8
homobrassinolide 1 and (22R,23R)-28-homocastasterone 3 as
well as diols 5 and 6 was carried out via the corresponding
11a (2.8 mg, 6.4 ␮mol, 64%, specific activity 2 × 10 cpm/␮mol)
was isolated by TLC (HPTLC Silica Gel UV-254 plates from “Merck”)
in a petroleum ether: acetone (10:3, by vol.) system. The product
11a was completely identical to the authentic sample 11 according
to TLC and 1H NMR data.
2
2
4
2,23-epoxides [25]. A key step in the preparation of (22S,23S)-
8-homobrassinolide 2 [23] and (22S,23S)-28-homocastasterone
[24] was osmium tetroxide dihydroxylation of the appropri-
2
2
Phosphate-buffered saline (PBS), propidium iodide, 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT),
ate stigmastane ꢀ -olefin. A pair of isomeric alcohols 7 and 8
was obtained by Sharpless dihydroxylation of (22E)-3␣,5-cyclo-
ꢀ
5
␣-stigmast-22-en-6-one in the presence of dihydroquinidine
4 ,6-diamidino-2-phenylindole hydrochloride (DAPI), and egg yolk
p-chlorobenzoate [26]. Compounds 9–12 were synthesized from
phosphatidyl choline (PC) were purchased from ‘Sigma’; culture

A.Yu. Misharin et al. / Steroids 75 (2010) 287–294
289
media and fetal calf serum (FCS) – from ‘Gibco BRL’ and ‘HyClone’;
culture plastics – from ‘Greiner’, ‘Costar’, and ‘Corning’. Small unil-
amellar vesicles PC were prepared by known method [29].
total energies of conformers using Boltzmann distribution at room
temperature (293 K).
2.5. Microphotographs of MCF-7 cells
2.2. Cell cultures
Microphotographs of MCF-7 cells incubated with compound
Human breast carcinoma MCF-7 cells, human ovary carci-
11 at a concentration of 30 ␮M for 30 h in a 35-mm beakers, fol-
lowed by washing with PBS were obtained on a ‘Axiovert 200M’
(Zeiss) instrument. For evaluation of nuclei fragmentation cells
noma CaOv cells, and human prostate carcinoma LnCaP cells were
obtained from American Cell and Tissue Collection (ACTC). Cells
were grown in suspension in culture medium consisting of RPMI
◦
were stained with DAPI solution (1 ␮g/mL) for 15 min at 20 C.
1
640, l-glutamine (2 mM), and antibiotics (100 U/mL penicillin,
1
00 mg/mL streptomycin) supplemented with 10% (v/v) heat inac-
2.6. Flow cytometry
tivated FCS. Cells were cultured in a 96-well plates, or 6-well plates,
◦
or 35-mm beakers at 37 C in an atmosphere containing 5% CO .
MCF-7 cells incubated with compound 11 at a various concen-
trations in 6-well plates were successively treated with 2 mL of
Versen solution and 0.7 mL of 0.25% trypsin solution and, after keep-
2
The tested compounds at various concentrations were added to the
culture medium in ethanolic solutions, the ethanol concentration
in all experiments (including corresponding controls) was 0.5% by
vol.
◦
ing at 37 C for 1 min, suspended in 1 mL of the medium. Cells were
centrifuged for 3 min at 1500 rpm, washed with PBS (2× 3 mL),
◦
treated with 70% ethanol (0.5 mL), and stored at −20 C. Before
2
.3. Interaction of compounds 1–12 with cells and their
experiment, the cells were washed with PBS (2× 3 mL) and vortexed
for 30 s with 0.5 mL of PBS, containing 25 ␮g propidium iodide, 0.5%
FCS, 25 ␮g RNase A, 15 ␮g detergent NP 40. Cells were maintained
cytotoxicity evaluation
◦
Before the experiments the cells in a 96-well plates were incu-
bated for 24 h in a serum-free medium. Toxicity of compounds
for 30 min at 4 C, thereafter were analysed on a “BD FACSAria” flow
cytometer using FL1 protocol.
1
–12 was measured by MTT assay [30]. Cells were incubated with
◦
tested compounds at 37 C in serum-free medium for 48 h, there-
3. Results and discussion
after the medium was aspirated, 125 ␮L of MTT solution in PBS
(
1 mg/mL) was added to each well, and plates were incubated
The first task was to find links between stereochemical con-
figuration of C22 and C23 atoms in 22,23-dihydroxystigmastane
derivatives and cytotoxicity of these compounds to human
breast carcinoma MCF-7 cells. We compared viability of MCF-7
cells incubated with either (22R,23R)-22,23-dihydroxystigmastane
derivatives (1, 3, 5, 7), or with related (22S,23S)-isomers (2, 4,
6, 8) at various concentrations for 48 h in serum-free media
(Fig. 2). Cells viability was assayed by MTT test [28], which
usually shows high correlation with number of living cells,
cell proliferation and release of mitochondrial matrix enzymes
[35–37]. All tested (22R,23R)-22,23-dihydroxystigmastane deriva-
tives (1, 3, 5, 7) exhibited significant toxicity to MCF-7 cells,
while only two compounds: (22S,23S)-28-homobrassinolide 2 and
(22S,23S)-28-homocastasterone 4 were toxic to MCF-7 cells among
(22S,23S)-22,23-dihydroxystigmastane derivatives.
◦
at 37 C for 4 h more. Then 125 ␮L of dimethyl sulfoxide was
added to each well, the plates were stored at room temperature
until the crystals dissolved, thereafter the absorbance at 560 nm
in each well was measured using ‘Multiskan spectrum (Thermo
scientific)’ microplate reader. The values for each point were calcu-
lated from six wells; all experiments were carried out in triplicate.
The toxicity of compounds tested was calculated from plot: cell
viability (% from control) versus concentration of compounds in
medium.
Effects of PC on toxicity of compound 11 to MCF-7 cells were
measured in the same conditions, except the incubation was car-
ried out for 30 h in the presence of PC at concentration of 100 and
3
00 ␮M.
Effects of PC on binding of compound 11 to MCF-7 cells was mea-
sured as follows. The cells in a 12-well plates were incubated for
Earlier 28-homocastasterone 3 was evaluated for cytotoxicity to
MCF-7 cells [14]. Reported IC50 value (40 ± 1.5 ␮M) was just little
higher than that obtained in our hands (32 ± 4 ␮M). These differ-
ences may be explained by distinct experimental conditions: we
carried out incubation in serum-free medium for 48 h, rather than
in medium supplemented with 10% FCS for 72 h as in [14].
Fig. 2 indicates that each (22R,23R) isomer was more toxic
than (22S,23S) isomer among the pairs of structurally related com-
pounds (1 and 2; 3 and 4; 5 and 6; 7 and 8, respectively). Therefore,
cytotoxicity of 22,23-dihydroxystigmastane derivatives is depen-
dent on the stereochemical configuration of C22 and C23 atoms
and apparently by spatial structure of the side chain.
8
3
0 h with compound 11a (10 ␮M, specific activity 10 cpm/␮mol)
in the presence of PC at concentration of 100 ␮M and 300 ␮M, then
◦
medium was aspirated, cells were washed with cold PBS at 4 C,
and treated with the 100 ␮L of buffer consisting of 0.15 M NaCl, 1%
triton X-100, 0.5% sodium deoxycholate and 0.1% sodium dodecyl
sulfate. The 50 ␮L aliquots were used for the measuring of radioac-
tivity on a ‘Tri-Carb 2800 TR (PerkinElmer)’ instrument and cell
protein content [31].
2.4. Computation of side chain conformation
Side chain conformations structures were computed by means
To figure out how the stereochemical configuration of C22 and
C23 atoms affects conformation of the side chain in (22R,23R)- and
(22S,23S)-22,23-dihydroxystigmastane derivatives, we performed
calculation of its low energy conformations. This task was accom-
plished by conformational search procedure, and resulting energies
were used to calculate the local energy minima population densi-
ties, yieldingestimates forrelativecontent ofindividualconformers
in the equilibrium mixture.
of ‘usage directed’ conformational search procedure [32] within the
HyperChem [33] molecular modeling software. The full-parameter
MM+ force field has been utilized for energy evaluations. Struc-
ture parameters varied during the search were six dihedral angles
along the side chain. All searches involve at least 10,000 energy
minimizations, and no more than 2000 conformations with energy
1
6 kcal/mol above best have been kept on each step. Visual molec-
ular dynamics (VMD) molecular graphics software [34] has been
used for structure analysis of conformers and measurements of
side chain spatial parameters (O–C22 C23–O dihedral angle, ꢁ22,23).
Local energy minima population densities were calculated from the
Calculated preferable side chain conformations and their ener-
getic distribution were quite similar for all used (22R,23R)-isomers
(compounds 1, 3, 5, 7) and significantly differ from those for
(22S,23S)-isomers (compounds 2, 4, 6, 8). The distribution of side

2
90
A.Yu. Misharin et al. / Steroids 75 (2010) 287–294
Fig. 2. (A–D) Effects of (22,23)-dihydroxystigmastane derivatives 1–8 on MCF-7 cells viability (details are given in Section 2).
chain conformations in (22R,23R)-isomers was much narrower
than that in (22S,23S)-isomers. Namely, about 96% of (22R,23R)-
isomers in equilibrium mixture lived in just several structurally
close energy minima; on other hand, (22S,23S)-epimeric side chain
accepted many various conformations in small energy range (in
order of 1 kcal/mol) above the global minimum.
It was shown that the presence of 21(S)-methyl group,
as well as the size and S-configuration of C24-substitutes
decrease conformational flexibility of 22,23-dihydroxilated side
chain in brassinosteroid analogs [16,17]. Both (22R,23R)- and
These data indicated that higher cytotoxicity of (22R,23R)-
22,23-dihydroxystigmastane derivatives, compared with that of
related (22S,23S)-isomers, correlates with their narrower distribu-
tion of ꢂ22,23 dihedral angle, and more rigid side chain.
Cytotoxic effects of (22R,23R)-22,23-dihydroxystigmastane
derivatives, varied in the steroid backbone structure, were eval-
uated in three cell lines: human breast carcinoma MCF-7 cells,
human ovary carcinoma CaOv cells, and human prostate carci-
noma LnCaP cells. Effects of compounds 1, 3, 5, 7 on viability
of MCF-7 cells are shown in Fig. 2, those of compounds 9, 10,
11, and 12 were reported earlier [18]. Cytotoxicity of (22R,23R)-
22,23-dihydroxystigmastane derivatives 1, 3, 5, 7, 9, 10, 11, 12
to CaOv cells and LnCaP cells is presented in Fig. 4. Cells were
exposed to each compound at various concentrations for 48 h in
serum-free medium and the proportions of surviving cells were
then estimated by MTT method. Cytotoxicity response of cell
lines to (22R,23R)-22,23-dihydroxystigmastane derivatives varied
in order: LnCaP > MCF-7 > CaOv; the relative toxicity of compounds
tested varied in order: 11 ≥ 12 > 5 ≥ 7 ≈ 9 > 10 > 3 > 1. Compounds
11 and 12, both comprising equatorial 3␤-hydroxyl group, exhib-
ited the highest cytotoxicity; the most polar 28-homobrassinolide
1 and 28-homocastasterone 3, both comprising 2␣,3␣-dihydroxy
groups exhibited the lowest activity; low polar compounds with-
out hydroxyl groups in steroid backbone 5, 7, 9, and 10 possessed
intermediate toxicity towards all used cell lines.
Fig. 5 illustrates characteristic features of MCF-7 caused by
incubation with the most toxic compound of this series (22R,23R)-
3␤,22,23-trihydroxystigmast-5-ene 11. Incubation of cells with
compound 11 led to disturbances of cell monolayer, partial cell
detachment, formation of “round” cells with “swollen” nuclei. We
did not see noticeable apoptosis in treated MCF-7 cells neither by
flow cytofluorimetry method, nor with DAPI staining [38] (data
not shown). Apparently, compound 11 causes necrosis rather than
apoptosis in MCF-7 cells.
(
22S,23S)-22,23-dihydroxystigmastane derivatives used in this
study comprised 21(S)-methyl and 24(S)-ethyl groups, differed
only in configurations of hydroxyls at C22 and C23, nevertheless
varied one another in conformational freedom of side chain. The
structure of steroid backbone did not affect low energy confor-
mation distribution either in (22R,23R)-, or in (22S,23S)-isomers.
Calculated low energy conformations for 28-homobrassinolide
1
and epimeric (22S,23S)-28-homobrassinolide 2 are shown in
Fig. 3A as examples.
Calculated side chain flexibility in brassinosteroids and related
compounds was presented earlier in terms of spatial area occu-
pied by either C25 atom [16], or C26 and C27 atoms [17]. Fig. 3A
shows that spatial area occupied by C22 and C23 hydroxyls groups
in (22R,23R)- and (22S,23S)-22,23-dihydroxystigmastane deriva-
tives are distinct one another and may be used for evaluation of
side chain flexibility as well. Thus we calculated the values for
O–C22–C23–O dihedral angle (ꢂ22,23) in preferable conformers for
both (22R,23R) and (22S,23S) isomers (Fig. 3B). The distributions of
◦
ꢂ22,23 for (22R,23R)-isomer with single sharp peak at about −60
indicates that most of the energy minima near to global one are
populated with conformers comprising that value of ꢂ22,23. On the
other hand, at least three different values of ꢂ22,23 are accepted by
the (22S,23S)-isomer, and the overall distribution of this parameter
is much less localized than that for (22R,23R).

A.Yu. Misharin et al. / Steroids 75 (2010) 287–294
291
Fig. 3. Calculated low energy conformers for (22R,23R)- and (22S,23S)-22,23-dihydroxystigmastane derivatives (A); calculated population density of low energy conformers
for (22R,23R)- and (22S,23S)-22,23-dihydroxystigmastane derivatives versus corresponding values of dihedral angle ꢂ22,23 (B) (28-Homobrassinolide 1 and (22S,23S)-28-
homobrassinolide 2 are shown as examples for (22R,23R)- and (22S,23S)-isomers.)
We have found that the presence of phospholipids decreases
cytotoxic effects of (22R,23R)-22,23-dihydroxystigmastane deriva-
tives. Viability of MCF-7 cells incubated with compound 11 at
a concentration of 10 and 30 ␮M for 30 h either in the absence
or in the presence of PC small unilamellar vesicles is shown
in Fig. 6A. The same effect of PC was observed when MCF-7
cells were incubated with compound 12. The number of sur-
vived cells reversely correlated with the concentration of PC in the
incubation media. We speculate that affinity of (22R,23R)-22,23-
dihydroxystigmastane derivatives 11 and 12 to phospholipids
bilayer could affect their distribution between cells and medium. It
is well known, that sterols and oxysterols, being slightly soluble
in aqueous medium, presumably bind to cell membrane, how-
ever the presence of the PC vesicles, as sterol acceptors in culture
medium, decreases this binding [39,40]. Consequently, the lower-
ing of cytotoxic effect of compounds 11 and 12 in the presence
of PC may be explained as decreasing of its concentration in cell
membrane.
Fig. 4. Effects of (22R,23R)-22,23-dihydroxystigmastane derivatives 1, 3, 5, 7, 9, 10, 11, 12 on LnCaP cells viability (A) and CaOv cells viability (B) (*these compounds at a
concentration higher than 10 ␮M caused cells detachment).

2
92
A.Yu. Misharin et al. / Steroids 75 (2010) 287–294
Fig. 5. Microphotographs of control MCF-7 cells (A) and MCF-7 cells exposed to (22R,23R)-3␤,22,23-trihydroxystigmast-5-ene 11 at concentration of 30 ␮M for 30 h (B).
Incubation of MCF-7 cells with [3H]-labeled (22R,23R)-
␤,22,23-trihydroxystigmast-5-ene 11a in the medium containing
PC vesicles, with subsequent measuring of radioactivity in the
cells (Fig. 6B), proved that the presence of PC in the incu-
bation medium indeed decreases binding of compound 11 to
cells. Obviously, compound 11 has affinity to phospholipids
bilayers and apparently to lipid components of cell mem-
branes.
3
Fig. 6. (A) Effects of PC on toxicity of (22R,23R)-3␤,22,23-trihydroxystigmast-5-ene 11 to MCF-7 cells. Cells were incubated for 30 h with compound 11 at indicated PC
3
concentrations in serum-free medium, followed by determination of percent of survival cells by MTT assay. (B) Effects of PC on binding of (22R,23R)-[3,4- H2]-3␤,22,23-
trihydroxystigmast-5-ene 11a to MCF-7 cells. Cells were incubated for 30 h with compound 11a (10 ␮M) at indicated PC concentrations in serum-free medium, followed by
measuring of radioactivity contents within the cells.

A.Yu. Misharin et al. / Steroids 75 (2010) 287–294
293
Interaction of some oxysterols and steroid sapogenins with
cell membrane is known to be important for the manifesta-
tion of their cytotoxic effects [41–49], since many regulatory
mechanisms controlling cell death and proliferation are linked
to the membrane signaling pathways. Recent finding indicated
that (22R,23R)-22,23-dihydroxystigmastane derivatives, namely
compounds 9, 11, and 12 are potent antagonists of Hedgehog sig-
naling pathway, their effects resembling those for cyclopamine,
the known inhibitor of Hedgehog signaling affecting the plasmatic
membrane (Strandt MF and Krauss S, personal communica-
tion).
We speculate that some sterols and their derivatives compris-
ing either rigid oxygenated side chain, or flexible one, may cause
different perturbation of cell membrane. To check this hypoth-
esis the interaction of 22,23-dihydroxystigmastane derivatives
with artificial bilayer phospholipid membranes is under study at
present.
[15] Uesusuki S, Watanabe B, Yamamoto S, Otsuki J, Nakagawa Y, Miyagawa
H. Synthesis of brassinosteroids of varying acyl side chains and evalua-
tion of their brassinolide-like activity. Biosci Biotechnol Biochem 2004;68:
1097–105.
[16] Yamamoto S, Watanabe B, Otsuki J, Nakagawa Y, Akamatsu M, Miya-
gawa H. Synthesis of 26,27-bisnorcastasterone analogs and analysis of
conformation–activity relationship for brassinolide-like activity. Bioorg Med
Chem 2006;14:1761–70.
[17] Drosihn S, Porzel A, Brandt W. Determination of preferred conformations of
brassinosteroids by means of NMR investigations and Boltzmann statistical
analysis of simulated annealing calculations. J Mol Model 2001;7:34–42.
[
18] Misharin AYu, Mehtiev AR, Morozevich GE, Tkachev YaV, Timofeev VP. Synthe-
sis and cytotoxicity evaluation of 22,23-oxygenated stigmastane derivatives.
Bioorg Med Chem 2008;16:1460–73.
[
19] Michelini FM, Ramirez JA, Berra A, Galagovsky LR, Alche LE. Anti-herpetic and
anti-inflammatory activities of two new synthetic 22,23-dihydroxylated stig-
mastane derivatives. J Steroid Biochem Mol Biol 2008;111:111–6.
[20] Michelini FM, Ramirez J, Berra A, Galagovsky L, Alche LE. In vitro and in vivo
antiherpetic activity of three new synthetic brassinosteroid analogues. Steroids
2004;69:713–20.
[
[
21] Ramirez JA, Teme Centurion OM, Gros EG, Galagovsky LR. Synthesis and bioac-
tivity evaluation of brassinosteroid analogs. Steroids 2000;65:329–37.
22] Ramırez JA, Bruttomesso AC, Michelini FM, Acebedo SL, Alche LE, Galagovsky LR.
Syntheses of immunomodulating androstanes and stigmastanes: comparison
of their TNF-␣ inhibitory activity. Bioorg Med Chem 2007;15:7538–44.
23] Akhrem AA, Lakhvich FA, Khripach VA, Kovganko NV, Zhabinskii VN. New
synthesis of (22S, 23S)-28-homobrassinolide. Doklady AN USSR 1985;283:
Acknowledgments
[
We are very grateful to Drs. Alexander Shtil and Olga Susova
130–3.
(
from N.N. Blokhin Cancer Center, Moscow) for kind gift of CaOv
and LnCaP cells, and helpful discussion; to Dr. Ivan Cheglakov
from V. N. Orekhovich Institute of Biomedical Chemistry RAMS,
[24] Akhrem AA, Lakhvich FA, Khripach VA, Kovganko NV, Zhabinskii VN.
New synthesis of (22S, 23S)-28-homocastasterone. Doklady AN USSR
1984;275:1089–91.
(
[
[
[
25] Khripach VA, Litvinovskaya RP, Raiman ME, Drach SV, Zhabinskii VN,
Sviridov OV, et al. Synthesis and immunochemical determination of 28-
homobrassinosteroids. Vesti NAN Belarusi, ser khim nav 2008;3:47–58.
26] Khripach VA, Zhabinskii VN, Ivanova DE, Khripach NB. Assymetric Sharpless
hydroxylation of ꢀ22-steroids of stigmastane family. Vesti NAN Belarusi, ser
khim nav 2007;1:46–51.
Moscow) for the help in flow cytometry analysis. This study was
supported by the Russian Foundation for Basic Research (Grants
RFBR 07-04-01155-a, 09-04-000454-a, and 08-04-90023-Bel a),
Belorussian Republican Foundation for Basic Research (Grant
BRFBR X08P-031), and ISTC (Project 1332) and program “Molec-
ular and Cellular Biology” of the Presidium of Russian Academy of
Sciences.
5
27] Aringer L, Eneroth P. Synthesis of 3␤-hydroxy-ꢀ -steroids by sodium borohy-
4
dride reduction of dienol trimethylsilyl ether derivatives of 3-oxo-ꢀ -steroids.
Steroids 1971;18:381–8.
[28] Zhang J, Larsson O, Sjovall J. 7a-Hydroxylation of 25-hydroxycholesterol
and 27-hydroxycholesterol in human fibroblasts. Biochim Biophys Acta
1995;1256:353–9.
References
[29] Batzri S, Korn DE. Single bilayer liposomes prepared without sonication.
Biochim Biophys Acta 1973;298:1015–8.
[
30] Mosmann T. Rapid colorimetric assay for cellular growth and survival: appli-
cation to proliferation and cytotoxicity assays. J Immunol Methods 1983;65:
[
[
[
1] D’Auria MV, Minale L, Ricco R. Polyoxygenated steroids of marine origin. Chem
Rev 1993;93:1839–95.
55–63.
2] Mukherjee AK, Basu S, Sarkar N, Ghosh AC. Advances in cancer therapy with
plant based natural products. Curr Med Chem 2001;8:1467–86.
3] Zhou Y, Garcia-Prieto C, Carney DA, Xu R, Pelicano H, Kang Y, et al. OSW-1: a
natural compound with potent anticancer activity and a novel mechanism of
action. J Natl Cancer Inst 2005;97:1781–5.
[
31] Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano
MD, et al. Measurement of protein using bicinchoninic acid. Anal Biochem
1985;150:76–85.
[
[
32] Chang G, Guida WC, Still W. J Am Chem Soc 1989;111:4379–86.
33] HyperChem , Hypercube, Inc., 1115 NW 4th Street, Gainesville, FL 32601, USA,
TM
[
4] Bok JW, Lermer L, Chilton J, Klingeman HG, Towers GH. Antitumor sterols from
the mycelia of Cordyceps sinensis. Phytochemistry 1999;51:891–8.
5] Matsuda H, Murakami T, Kishi A. Structures of withanosides I, II, III, IV, V, VI, and
VII, new withanolide glycosides, from the roots of Indian Withania somnifera
DUNAL and inhibitory activity for tachyphylaxis to clonidine in isolated guinea-
pig ileum. Bioorg Med Chem 2001;9:1499–507.
2000.
[
[
34] Humphrey W, Dalke A, Schulten K. J Mol Graphics 1996;14:33–8.
35] Morishita M, Iwasaki Y, Onishi A, Asai M, Mutsuga N, Yoshida M, et al. The
[
ꢀ
effects of GH-releasing hormone/somatostatin on the 5 -promoter activity of
the GH gene in vitro. J Mol Endocrinol 2003;31:441–8.
[
36] Robe PA, Nguyen-Khac M, Jolois O, Rogister B, Merville M-P, Bours V. Dexam-
ethasone inhibits the HSV-tk/ganciclovir bystander effect in malignant glioma
cells. BMC Cancer 2005;5:32, doi:10.1186/1471-2407r-r5-32.
[
6] Masuda T, Oyama Y, Yamamoto N, Umebayashi C, Nakao H, Toi Y, et al. Cytotoxic
screening of medicinal and edible plants in Okinava, Japan, and identification
of the main toxic constituent of Rhodea japonica (Omoto). Biosci Biotechnol
Biochem 2003;67:1401–4.
[37] Baratta M, Grolli S, Poletti A, Ramoni R, Motta M, Tamanini C. Role of andro-
gens in proliferation and differentiation of mouse mammary epithelial cell line
HC11. J Endocrinol 2000;167:53–60.
[
7] Mimaki Y, Watanabe K, Sakagami H, Sashida Y. Steroidal glycosides from the
leaves of Cestrum nocturnum. J Nat Prod 2002;65:1863–8.
[
8] Mellado GG, Zubıa E, Ortega MJ, Lopez-Gonzalez PJ. New polyoxygenated
steroids from the Antarctic octocoral Dasystenella acanthina. Steroids
[38] Goldberg JL, Zanella CL, Janssen YM, Timblin CR, Jimenez LA, Vacek P, et al.
Novel cell imaging techniques show induction of apoptosis and proliferation
in mesothelial cells by asbestos. Am J Respir Cell Mol Biol 1997;17:265–71.
[39] Phillips MC, Johnson WJ, Rothblat GH. Mechanisms and consequences of
cellular cholesterol exchange and transfer. Biochim Biophys Acta 1987;906:
223–76.
2
004;69:291–9.
[
9] Nakata T, Yamada T, Taji S, Ohishi H, Wada S, Tokuda H, et al. Structure
determination of inonotsuoxides A and B and in vivo anti-tumor promoting
activity of inotodiol from the sclerotia of Inonotus obliquus. Bioorg Med Chem
2
007;15:257–64.
[40] Rothblat GH, Mahlberg FH, Johnson WJ, Phillips MC. Apolipoproteins, mem-
brane cholesterol domains and the regulation of cholesterol efflux. J Lipid Res
1992;33:1091–7.
[41] Wang JW, Megha LE. Relationship between sterol/steroid structure and partici-
pation in ordered lipid domains (lipid rafts): implications for lipid raft structure
and function. Biochemistry 2004;43:1010–8.
[
[
10] Jiang B, Shi H-P, Xu M, Wang W-J, Zhou W-S. Stereoselective synthesis of Cer-
tonardolsterol D3. Tetrahedron 2008;64:9738–44.
11] Hovenkamp E, Demonty I, Plat J, Lutjohann D, Mensink RP, Trautwein EA. Bio-
logical effects of oxidized phytosterols: a review of the current knowledge. Prog
Lipid Res 2008;47:37–49.
[
12] Mandal C, Dutta A, Mallick A, Chandra S, Misra L, Sangwan RS, et al. With-
aferin A induces apoptosis by activating p38 mitogen-activated protein kinase
signaling cascade in leukemic cells of lymphoid and myeloid origin through
mitochondrial death cascade. Apoptosis 2008;13:1450–64.
[42] Massey JB, Pownall HJ. Structures of biologically active oxysterols deter-
mine their differential effects on phospholipid membranes. Biochemistry
2006;45:10747–58.
[43] Massey JB. Membrane and protein interactions of oxysterols. Curr Opin Lipidol
2006;17:296–301.
[44] Javitt NB, Oxysterols:. Novel biologic roles for the 21st century. Steroids
2008;73:149–57.
[45] Corcoran RB, Scott MP. Oxysterols stimulate Sonic hedgehog signal trans-
duction and proliferation of medulloblastoma cells. Proc Natl Acad Sci USA
2006;106:8408–13.
[
[
13] Wang T-M, Hojo T, Ran F-X, Wang R-F, Wang R-Q, Chen H-B, et al. Cardeno-
lides from Saussurea stella with cytotoxicity toward cancer cells. J Nat Prod
2
007;70:1429–33.
14] Malıkova J, Swaczynova J, Kolar Z, Strnad M. Anticancer and antipro-
liferative activity of natural brassinosteroids. Phytochemistry 2008;69:
4
18–26.

2
94
A.Yu. Misharin et al. / Steroids 75 (2010) 287–294
[
46] Dwyer JR, Sever N, Carlson M, Nelson SF, Beachy PA, Parhami F. Oxysterols are
novel activators of the Hedgehog signaling pathway in pluripotent mesenchy-
mal cells. J Biol Chem 2007;282:8959–68.
on Epstein–Barr virus activation and two-stage mouse skin carcinogenesis.
Bioorg Med Chem 2005;13:1403–8.
[49] Ito S-I, Ihara T, Tamura H, Tanaka S, Ikeda T, Kajihara H, et al. ␣-Tomatine,
the major saponin in tomato, induces programmed cell death mediated by
reactive oxygen species in the fungal pathogen Fusarium oxysporum. FEBS Lett
2007;581:3217–22.
[
[
47] Rao AV, Sung M-K. Saponins as anticarcinogens. J Nutr 1995;125:717–24.
48] Sakurai N, Kozuka M, Tokuda H, Mukainaka T, Enjo F, Nagai M, et al. Antitumor
agents 220. Antitumor-promoting effects of Cimigenol and related compounds