Selective synthesis of the two main progesterone metabolites, 3α-hydroxy-5α-pregnanolone (allopregnanolone) and 3α-hydroxypregn-4-en-20-one, and an assessment of their effect on proliferation of hormone-dependent human breast cancer cells

Article abstract of DOI:10.1007/s11172-020-2797-4

A directed synthesis of two progesterone metabolites, allopregnanolone and 3a-hydroxy-pregn-4-en-20-one, from Δ¹⁶-pregnanolone and progesterone, respectively, was carried out by a reduction of the carbonyl groups in positions 3 and subsequent inversion of the configuration of the resulting alcohols by the Mitsunobu reaction. The selectivity of the reduction of the conjugated carbonyl group in position 3 of progesterone with sodium borohydride in the presence of cerium(III) chloride (Luche reduction) was demonstrated. The ef ect of the obtained metabolites on the proliferation of breast cancer cells of the MCF-7 and T47D lines under normal and steroid-free conditions was studied. It is shown that the ef ect of these compounds on the proliferation depends on the presence of additional steroids in the culture medium. Metabolites exerted small cytostatic ef ects on the growth of the MCF-7 cells under standard conditions, while the transfer of the cells to a steroid-free medium weakened these cytotoxic ef ects. In the experiments with the T47D line cells, the cell growth was stimulated under both standard and steroid-free conditions. Allopregnanolone and progesterone stimulate the growth to a greater extent under steroid-free conditions than under standard ones.

Full text of DOI:10.1007/s11172-020-2797-4

Russian Chemical Bulletin, International Edition, Vol. 69, No. 3, pp. 552—557, March, 2020
552
Selective synthesis of the two main progesterone metabolites,
3α-hydroxy-5α-pregnanolone (allopregnanolone) and
3α-hydroxypregn-4-en-20-one, and an assessment of their effect
on proliferation of hormone-dependent human breast cancer cells
M. O. Tserfas,a I. S. Levina,a Y. V. Kuznetsov,a A. M. Scherbakov,b E. I. Mikhaevich,b and I. V. Zavarzina
aN. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences,
47 Leninsky prosp., 119991 Moscow, Russian Federation.
Fax: +7 (499) 135 6390. E-mail: islevina@gmail.com
bN. N. Blokhin National Medical Research Center of Oncology of the Ministry of Health of the Russian Federation,
24 bld. 15, Kashirskoe shosse, 115522 Moscow, Russian Federation.
E-mail: alex.scherbakov@gmail.com
A directed synthesis of two progesterone metabolites, allopregnanolone and 3-hydroxy-
pregn-4-en-20-one, from 16-pregnanolone and progesterone, respectively, was carried out by
a reduction of the carbonyl groups in positions 3 and subsequent inversion of the configuration
of the resulting alcohols by the Mitsunobu reaction. The selectivity of the reduction of the
conjugated carbonyl group in position 3 of progesterone with sodium borohydride in the pres-
ence of cerium(III) chloride (Luche reduction) was demonstrated. The effect of the obtained
metabolites on the proliferation of breast cancer cells of the MCF-7 and T47D lines under
normal and steroid-free conditions was studied. It is shown that the effect of these compounds
on the proliferation depends on the presence of additional steroids in the culture medium.
Metabolites exerted small cytostatic effects on the growth of the MCF-7 cells under standard
conditions, while the transfer of the cells to a steroid-free medium weakened these cytotoxic
effects. In the experiments with the T47D line cells, the cell growth was stimulated under both
standard and steroid-free conditions. Allopregnanolone and progesterone stimulate the growth
to a greater extent under steroid-free conditions than under standard ones.
Key words: progesterone, allopregnanolone, 3-hydroxypregn-4-en-20-one, steroid
metabolites, cytotoxicity, luminal cancer, Luche reduction, MTT assay.
Steroid hormones belong to the main regulators of
rivatives with reduced 4,5-double bonds, 20-dihydro de-
rivatives, 5()-pregnan-3()-ol-20-ones, and their
20-dihydroxy derivatives, 5()-pregnan-3,20-diols.2,3
In particular, two main metabolites of progesterone in the
human mammary gland tissue are 3-hydroxyprogesterone
(3-HP) and 5-dihydroprogesterone (5-P).4 The process
of metabolism of progesterone (1) in the human mammary
gland tissue under the influence of 3-hydroxysteroid
oxidoreductase (3-HSO) and 5-reductase (5-R),
which leads to 3-HP and 5-P, is shown5 in Scheme 1.
animal and human vital activity. The natural hormone
progesterone (1) plays an essential role in the female body,
regulating ovulation, preparation of the endometrium to
implantation, pregnancy preservation, and many other
physiological functions.1 It is known that progesterone
quickly transforms into various metabolites in different
body tissues, and these metabolites can have specific
physiological effects, thereby affecting the efficiency of
the hormone. The main metabolites are 5- and 5-de-
Scheme 1
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 3, pp. 0552—0557, March, 2020.
1066-5285/20/6903-0552 © 2020 Springer Science+Business Media LLC

Synthesis of two main progesterone metabolites
Russ. Chem. Bull., Int. Ed., Vol. 69, No. 3, March, 2020
553
Scheme 2
Reagents and conditions: a. NaBH4, CeCl3•6H2O, MeOH, –20 C. b. BzOH, PyPPh2, diisopropyl azodicarboxylate (DIAD),
toluene, 80 C. c. KOH—MeOH.
In the early 2000s, it was revealed that in the normal
tissue of a human mammary gland, progesterone is meta-
bolized mainly with participation of reversible 20- and
3-hydroxysteroid oxidoreductases, resulting in the 4-preg-
nene derivatives, viz., 20- and 3-hydroxyprogesterones,
respectively. These metabolites have valuable biological
properties, exhibiting antiproliferative effect. The meta-
bolism of progesterone in a tumor significantly differs from
that in the normal tissue. In malignant cells, progesterone
is predominantly metabolized by 5-reductase to a 5-re-
duced metabolite exhibiting undesirable proliferative
activity in a breast tumor.4,6—8
Recently we studied biotransformation of pentacyclic
progesterone analogs in vivo and showed that their meta-
bolic pathways are identical to those of endogenous pro-
gesterone.9 In continuation of our research in the field of
the synthesis and investigation of biological effects of
progestins and their metabolites, we have carried out
a selective synthesis of two progesterone metabolites, viz.,
3-hydroxypregn-4-en-20-one (2) and 3-hydroxy-5-
pregnan-20-one (allopregnanolone) (3), and studied their
effect on proliferation of the T47D and MCF-7 cells of
hormone-dependent human breast cancer. Metabolites 2
and 3 were isolated earlier10 in low yields (15 and 30%)
from products of reduction of progesterone and 5-pre-
gnan-3,20-dione, respectively, with KS-selectride.
Results and Discussion
We have developed a convenient preparative synthesis
of 3-hydroxypregn-4-en-20-one (2) by a selective reduc-
tion of the 4-3-keto group of progesterone (1) according
to the Luche method11 by the action of NaBH4 in meth-
anol in the presence of CeCl3•6H2O followed by the in-
version of the center in position 3 of the formed steroid 4
by the Mitsunobu reaction. At the stage of preparation of
benzoate 5, diphenyl(2-pyridyl)phosphine12 was used
instead of triphenylphosphine to achieve more efficient
removal of unreacted phosphine and its oxide by simple
washing with diluted hydrochloric acid. The desired
3-hydroxypregn-4-en-20-one (2) was obtained by alka-
line hydrolysis of benzoate 5 with methanol solution of
potassium hydroxide (Scheme 2).
3-Hydroxy-5-pregnan-20-one (3) was obtained
from 16-pregnanolone acetate (6) (Scheme 3). The double
bond in the molecule of steroid 6 was quantitatively hydro-
genated in the presence of 10% Pd/C, then the 3-acetoxyl
group in obtained compound 7a was hydrolyzed, produc-
ing 3-hydroxysteroid 7b, which resulted in the formation
of 3-hydroxypregnane 3 in high yield under the condi-
tions of the Mitsunobu reaction (see Scheme 3).
We have studied the influence of the prepared com-
pounds 2 and 3, as well as progesterone 1 on the growth
Scheme 3
Reagents and conditions: a. H2, Pd/C. b. KOH—MeOH. c. BzOH, PyPPh2, DIAD, toluene, 80 C.

Tserfas et al.
554
Russ. Chem. Bull., Int. Ed., Vol. 69, No. 3, March, 2020
(steroid-free conditions) is additionally used.13 Under the
steroid-free conditions, it is possible to model direct com-
petition between two certain steroids or to investigate the
effect of particular steroid on a hormone receptor without
the background created by endogenous serum steroids.
Moreover, the steroid-free conditions can be considered
as an adequate model for studying tumors in menopausal
women (when the steroid hormone level is decreased).
The steroid-free conditions make it possible to estimate
the ability of non-steroid molecules to act as potential
agonists of hormone receptors. For instance, it was de-
monstrated that clodronic acid, which is a clinically ap-
proved bone resorption inhibitor, exhibits estrogenic ef-
fects.13 Thus, the use of steroid-free conditions expand
the possibilities of in vitro studies of hormone metabolites
and compounds with hormone-like properties.
The MCF-7 cells were transferred to steroid-free
conditions (see the Experimental). It was shown the ste-
roid-free conditions weaken the cytostatic effect of pro-
gesterone (1) and completely neutralize the effects of
metabolites 2 and 3 (see Fig. 1, b). Moreover, weak
stimulating effect on proliferation of MCF-7 cells was
revealed under the steroid-free conditions for compound 3
in low concentrations (15% in comparison with the
control cells).
The T47D cells, like the MCF-7 cells, belong to lu-
minal cancer cells. This type of mammary gland tumors
is characterized by expression of estrogen and progesterone
receptors and a relatively good clinical prognosis. One of
differences between these lines consists in higher expres-
sion of progesterone receptors in the T47D cells compared
to the MCF-7 cells. We revealed a minor stimulating effect
of allopregnanolone (3) in a concentration of 12 mol L–1
on growth of the T47D cells under standard conditions of
culturing (Fig. 2, a). Like in the experiments with the
MCF-7 cells, progesterone (1) in high concentrations
caused cytostatic influence on the T47D cells. When the
steroid-free conditions were used, it was found that this
compound stimulates the T47D cell growth (see Fig. 2, b).
Low (up to 25 mol L–1) concentrations of allopregna-
nolone (3) and progesterone (1) caused substantial stimu-
lation of cell growth. The effect of metabolite 2 on the cell
growth stimulation was less pronounced. In a concentra-
tion of 50 mol L–1, the stimulating effect of compound
3 decreased, whereas progesterone in the same concentra-
tion inhibited the cell growth by 44%.
Survival rate (%)
a
120
100
80
2
3
1
60
*
40
10
20
30
40 C/mol L–1
Survival rate (%)
140
b
120
100
80
3
2
1
#
60
#
10
20
30
40 C/mol L–1
Fig. 1. The effect of progesterone (1) and its metabolites 2 and
3 on the survival rate of the MCF-7 breast cancer cells in 72 h
of incubation under the standard (a) and steroid-free (b) condi-
tions (for purification of the serum from steroids see the
Experimental). Symbols "*" and "#" correspond to p < 0.05 in
comparisons of progesterone with compounds 2 and 3.
of cells of hormone-dependent breast cancer. The ex-
periments were carried out with two lines of luminal breast
cancer cells: MCF-7 and T47D. Progesterone 1 and its
metabolites 2 and 3 in concentrations up to 25 mol L–1
did not cause significant cytostatic or stimulating effects
on the MCF-7 cells (Fig. 1, a). At steroid concentrations
of 25 mol L–1, the cell growth inhibition did not exceed
25% compared to the control cells. Upon going to higher
concentrations (50 mol L–1), the cytostatic effect of
progesterone became more pronounced than that of its
metabolites 2 and 3. Under these conditions, 3-hydr-
oxypregn-4-en-20-on (2) and allopregnanolone (3) in-
hibited the MCF-7 cell growth by less than 30%, whereas
progesterone (1) resulted in the 60% proliferation sup-
pression.
Incubation of cells with a studied compound for 72 h
in a medium with 7—10% fetal serum is the generally ac-
cepted standard for preclinical experiments. These ex-
periments are usually aimed at characterization of anti-
proliferative potential of newly developed drugs or assess-
ment of the capability of physiological metabolites to
promote proliferation of malignant cells or normal epi-
thelium. Universal protocols of MTT assay and other
analogous procedures make it possible to compare data
obtained in different laboratories. For a more detailed
characterization of steroids, serum purified from steroids
It was shown earlier5 that in mammary gland tissues,
progesterone practically completely transforms into meta-
bolites, the major of which are 5-Р and 3-HР (see
Scheme 1). For both cell lines, metabolite 5-Р signifi-
cantly prevails over other progesterone metabolites.14,15
However, in the MCF-10A cells of normal epithelium of
mammary gland, progesterone is metabolized to a lesser
extent than in tumor cells, and the 5-Р : 3-HР ratio is
smaller by 10—30 times than in the latter.14,16

Synthesis of two main progesterone metabolites
Russ. Chem. Bull., Int. Ed., Vol. 69, No. 3, March, 2020
555
Survival rate (%)
140
their growth was found. Almost no cytostatic effect was
observed under the standard conditions for the compounds
under study. Under the steroid-free conditions of cultur-
ing, progesterone (1) and allopregnanolone (3) exhibited
marked stimulating influence on the cell growth, while the
stimulating effect of compound 2 was not exceeded 20%.
Unfortunately, the 5-Р and 3-HР concentration de-
pendences shown in an earlier study7 correspond to values
not exceeding 1 mol L–1, which limits the possibility of
a direct comparison of those data with ours. The authors
studied the effect of 5-Р and 3-HР on incorporation of
tritium-labeled thymidine in DNA. This method is not
a direct analog of the MTT assay, which provides an ad-
ditional explanation of partial discrepancy between the
data obtained for 3-HР.
a
*
120
100
80
2
3
60
1
40
10
20
30
40 C/mol L–1
Survival rate (%)
200
b
Thus, even inside the cells of breast cancer of luminal
type, the action of progesterone metabolites can vary. In
particular, their influence on proliferation depends on the
presence of additional steroids in the medium for cultur-
ing. The transition to steroid-free conditions in the case
of MCF-7 decreases the cytostatic effect of metabolites 2
and 3, whereas in the case of the T47D cells, it enhances
the capability of allopregnanolone (3) and progesterone
(1) to stimulate the cancer growth.
On the whole, metabolism of progesterone (and syn-
thetic progestins as anticancer drugs) is a highly important
factor of progression of breast tumor. Further more de-
tailed studies of the status of progesterone receptors and
the activity of progestin metabolism enzymes are needed
to correct the hormonal treatment of patients with breast
cancer. Patients with breast cancer of luminal type with
high level of progesterone receptors or/and progestin
metabolism enzymes can be singled out into a separate group
for the creation of personalized treatment protocols.
180
160
140
120
100
80
3
2
*
*
#
1
60
10
20
30
40 C/mol L–1
Fig. 2. The effect of progesterone (1) and its metabolites 2 and
3 on the survival rate of the T47D breast cancer cells in 72 h of
incubation under the standard (a) (symbol "*" corresponds to
p < 0.05 in pairwise comparison of the compounds) and steroid-
free (b) conditions (symbol "*" corresponds to p < 0.05 in com-
parisons of progesterone with compounds 3 and 1, symbol "#"
corresponds to p < 0.05 in comparisons of compound 1 with
compounds 2 and 3).
At the same time it was noted that metabolites 5-Р
and 3-HР themselves have the opposite effect on prolif-
eration of cells of both tumor and normal epithelium of
mammary gland: 5-Р considerably stimulates proliferation,
while 3-HР inhibits it.4,6 Recently17 it was suggested that
the effects of metabolites 5-Р and 3-HР are mediated
by corresponding membrane-associated receptors, however
the mechanism of the opposite action of these metabolites
at the cellular level has not yet been established. Our ex-
perimental data partly support the effects described in the
mentioned studies. Steroids 2 and 3 had small cytostatic
effects on the growth of the MCF-7 cells of breast cancer
under the standard conditions, and no significant differ-
ence between the actions of these compounds was found.
Progesterone in a concentration of 50 mol L–1 was more
active than the synthesized metabolites. The use of steroid-
free conditions weakened the cytostatic effects of the
compounds, and a tendency to stimulation of growth of
the MCF-7 cells was noted for allopregnanolone.
Experimental
Melting points were obtained using a Boetius melting-point
apparatus. Specific optical rotation was measured using a JASCO
P-2000 digital polarimeter in CHCl3 at 22 C. 1H NMR spectra
were recorded in CDCl3 at 30 С with the use of a Bruker AM-300
spectrometer (300.13 MHz). The signal of residual CHCl3
(H 7.27) in CDCl3 was used as a standard. High resolution mass
spectra were obtained with a Bruker micrOTOF II mass spec-
trometer using electrospray ionization (ESI). Analytical TLC
was carried out on Silica gel 60 F254 plates (Merck). Compounds
were revealed by the treatment of the plates with 1% solution of
Ce(SO4)2 in 10% aqueous H2SO4 and subsequent warming up.
The preparative separation was carried out by column chrom-
atography using a column with Kieselgel 60 (0.063—0.200 m)
silica gel (Merck) at the compound : sorbent ratio = 1 : 40.
Solvents were purified according to standard procedures. Yields
of products are given for recrystallized samples. Starting 3-acet-
oxy-5-pregn-16-en-20-one was kindly provided by G. S. Gri-
nenko (All-Union Scientific Research Chemical-Pharmaceutical
Institute), progesterone was obtained from Fisher, other reagents
were obtained from Sigma.
In the experiments with the luminal breast cancer cells
of another line (T47D), more pronounced stimulation of

Tserfas et al.
556
Russ. Chem. Bull., Int. Ed., Vol. 69, No. 3, March, 2020
3β-Hydroxypregn-4-en-20-one (4). A solution of progester-
C(18)Н3); 0.82 (s, 3 H, C(19)Н3); 1.12—1.17 (m, 1 H, H(5));
2.10, 2.15 (both s, 2×3 H, 3-OAc, C(21)Н3); 2.50—2.55 (m, 1 H,
H(17)); 4.55—4.59 (m, 1 H, H(3)) (see Ref. 19).
one (1) (314 mg, 1 mmol) and CeCl3•6H2O (374 mg, 1 mmol)
in methanol (10 mL) was cooled to –20 С, then NaBH4 (19 mg,
0.5 mmol) was added under stirring. In 10 min, acetone (0.37 mL)
was added to the mixture, and cooling was ceased. The reaction
mixture was poured into cold water (30 mL) and extracted with
ethyl acetate (3×10 mL). The organic layer was washed with
a saturated NaCl solution (15 mL) and dried over anhydrous
Na2SO4. The obtained product was chromatographed on a col-
umn. By elution with CH2Cl2—petroleum ether mixtures (2 : 1)
with additions of 0.21% acetone, 3-hydroxypregn-4-en-20-
one 4 was obtained (210 mg, 67%), m.p. 153—155 С (see Ref. 18:
m.p. 159 С). 1H NMR, : 0.66 (s, 3 H, C(18)H3); 1.07 (s, 3 H,
C(19)H3); 2.13 (s, 3 H, C(21)H3); 2.50—2.55 (m, 1 H, H(17));
4.17 (br.s, 1 H, H(3)); 5.31 (s, 1 Н, H(4)) (see Ref. 10).
3β-Hydroxy-5α-pregnan-20-one (7b). A suspension of ac-
etate 7a (790 mg, 2.2 mmol) and an 1.8 N KOH solution (5.6 mL)
in MeOH (22 mL) was heated at 60 С for 1 h. The reaction
mixture was poured out into cold water (40 mL) acidified with
concentrated HCl to pH 1—2 and extracted with ethyl acetate
(3×10 mL). The organic layer was washed with water (20 mL),
dried over anhydrous Na2SO4, the solvent was removed in vacuo
to give 3-hydroxysteroid 7b (670 mg, 96%), m.p. 194—195 С
(ethanol) (see Ref. 20: m.p. 194—195 С). 1H NMR, : 0.68
(s, 3 H, C(18)Н3); 0.82 (s, 3 H, C(19)Н3); 0.99—1.30 (m, 1 H,
H(5)); 2.13 (s, 3 H, C(21)Н3); 2.50—2.55 (m, 1 H, H(17));
3.57—3.63 (m, 1 H, H(3)) (see Ref. 21).
3α-Benzoyloxypregn-4-en-20-one (5). A solution of DIAD
(0.36 mL, 1.7 mmol) in dry toluene (1.8 mL) was added dropwise
to a cooled (ice bath) mixture of alcohol 4 (350 mg, 1.1 mmol),
benzoic acid (212 mg, 1.7 mmol), and diphenyl(2-pyridyl)phos-
phine (457 mg, 1.7 mmol) in toluene (8 mL). The reaction
mixture was heated at 80 С for 16 h, then cooled to room tem-
perature and washed with an 1 N solution of HCl (4×10 mL),
a saturated NaHCO3 solution (3×10 mL), and a saturated NaCl
solution (10 мл). The organic fraction was separated, dried over
anhydrous Na2SO4, and the solvent was removed in vacuo. The
rest was chromatographed on a column with silica gel. 3-Benzo-
yloxypregn-4-en-20-one (5) was obtained in the form of dense
oil (330 mg, 70%) by elution with CH2Cl2—petroleum ether
mixtures (2 : 1) with additions of 0.21% acetone. The product
was further used without additional purification. 1H NMR,
: 0.68 (s, 3 H, C(18)H3); 1.12 (s, 3 H, C(19)H3); 2.13 (s, 3 H,
C(21)H3); 2.55—2.59 (m, 1 H, H(17)); 5.39 (br.s, 1 H, H(3));
5.59 (s, 1 H, H(4)); 7.49 (t, 2 H, Ph, J = 7.7 Hz); 7.59 (t, 1 H,
Ph, J = 7.4 Hz); 8.09 (d, 2 H, Ph, J = 7.2 Hz).
3α-Hydroxypregn-4-en-20-one (2). A suspension of benzo-
ate 5 (330 mg, 0.78 mmol) and a 3 N KOH solution (1.3 mL) in
MeOH (6 mL) was heated at 60 С for 2.5 h. The reaction mix-
ture was poured out into cold water (23 mL) acidified with
concentrated HCl to pH 1—2 and extracted with ethyl acetate
(3×10 mL). The organic layer was washed with a saturated
NaHCO3 solution (15 mL) and dried over anhydrous Na2SO4,
the solvent was then removed in vacuo. The rest was chromato-
graphed on a silica gel column. After elution with CH2Cl2—pe-
troleum ether mixtures (2 : 1) containing 0.21% acetone and
recrystallization, 3-hydroxypregn-4-en-20-one (2) was obtained
(150 mg, 64%), m.p. 127—129 C (hexane—acetone) (see Ref. 10:
m.p. 127—129 С), []D22 +229 (c 1.00) (see Ref. 10: []D26 +231
(c 1.80, CHCl3)). 1H NMR, : 0.65 (s, 3 H, C(18)H3); 0.99
(s, 3 H, C(19)H3); 2.12 (s, 3 H, C(21)H3); 2.51—2.55 (m, 1 H,
H(17)); 4.08 (br.s, 1 H, H(3)); 5.48 (d, 1 H, H(4), J = 4.5 Hz)
(see Ref. 10). High resolution mass spectrum, found: m/z
339.2293 [M + Na]+. C21H32NaO2. Calculated: 339.2295.
3-Acetoxy-5α-pregnan-20-one (7a). A solution of 3-acet-
oxy-5-pregn-16-en-20-one (6) (1.52 g, 4.2 mmol) in dioxane
(25 mL) was hydrogenated in the presence of Pd/C (0.07 g, 10%)
until complete conversion of the starting compound according
to TLC (development with an aqueous solution of KMnO4). The
catalyst was filtered, the solvent was evaporated in vacuo, the
rest was recrystallized from a hexane—acetone mixture to give
3-acetoxy-5-pregnan-20-one (7a) (1.5 g, 98%), m.p. 137—142 С
(see Ref. 19: m.p. 146—147 С). 1H NMR, : 0.67 (s, 3 H,
3α-Benzoyloxy-5α-pregnan-20-one (8). A solution of DIAD
(0.7 mL, 3.31 mmol) in dry toluene (3.5 mL) was added dropwise
to a cooled (ice bath) mixture of alcohol 7b (670 mg, 2.10 mmol),
benzoic acid (404 mg, 3.31 mmol), and diphenyl(2-pyridyl)
phosphine (876 mg, 3.33 mmol) in toluene (17 mL). The reaction
mixture was heated at 80 С for 16 h, cooled to room temperature,
and washed with an 1 N HCl solution (4×10 mL), a saturated
NaHCO3 solution (3×10 mL), and a saturated NaCl solution
(10 mL). The organic fraction was separated and dried over
anhydrous Na2SO4, the solvent was removed in vacuo. The rest
was chromatographed on a silica gel column. 3-Benzoyloxy-
5-pregnan-20-one (8) was obtained in the form of dense oil
(750 mg, 84%) by elution with CH2Cl2—petroleum ether mixtures
(1 : 1) containing 0.21% acetone and was used further without
additional purification. 1H NMR, : 0.64 (s, 3 H, C(18)Н3);
0.87 (s, 3 H, C(19)Н3); 2.13 (s, 3 H, C(21)Н3); 2.55—2.59
(m, 1 H, H(17)); 5.32 (br.s, 1 H, H(3)); 7.48 (t, 2 H, Ph, J = 7.7 Hz);
7.58 (t, 1 H, Ph, J = 7.4 Hz); 8.09 (d, 2 H, Ph, J = 7.2 Hz).
Hydroxy-5α-pregnan-20-one (allopregnanolone) (3). A sus-
pension of benzoate 8 (750 mg, 1.77 mmol) and a 7 N KOH
solution (2.3 mL) in MeOH (45 mL) was heated at 60 С for
2.5 h. The reaction mixture was poured into cold water (37 mL)
acidified with concentrated HCl to pH 1—2 and extracted
with ethyl acetate (3×10 mL). The organic layer was washed
with a saturated NaHCO3 solution (15 mL), water (15 mL), and
dried over anhydrous Na2SO4, the solvent was removed in vacuo.
The rest was chromatographed on a column with silica gel.
Allopregnanolone (3) was obtained (320 mg, 57%) by elution
with CH2Cl2—petroleum ether mixtures (1 : 1)) containing
0.21% acetone, m.p. 177 С (ethanol) (see Ref. 22: m.p.
22
16
174—176 С), []D +103.9 (c 1.00) (see Ref. 23: []D +96
(c 0.5, CHCl3)). 1H NMR, : 0.62 (s, 3 H, C(18)Н3); 0.80 (s, 3
H, C(19)Н3); 2.13 (s, 3 H, C(21)Н3); 2.52—2.61 (m, 1 H, H(17));
4.07 (br.s, 1 H, H(3)) (see Ref. 10).
Biological studies. For biological testing, the compounds were
dissolved in DMSO to a concentration of 5 mmol L–1 and kept
at –20 С until the use. Human breast cancer cells MCF-7 and
T47D (ER- and PR-positive) were obtained from the ATCC
collection. The MCF-7 and T47D cells were cultured in vitro in
the standard DMEM medium (Gibco) with additions of sodium
pyruvate (0.1 mg mL–1, ChemCruz). The medium contained
a 10% fetal calf serum (HyClone), penicillin (50 IU mL–1), and
streptomycin (50 g mL–1) (PanEco). Incubation was carried
out in a 5% СО2 atmosphere at 37 С and relative humidity of
80—90% (a NU-5840E CO2 incubator, NuAir). The effects of
steroids which are present in the fetal serum on the activities of

Synthesis of two main progesterone metabolites
Russ. Chem. Bull., Int. Ed., Vol. 69, No. 3, March, 2020
557
the compounds under study were assessed under steroid-free
conditions. In this case, DMEM without addition of phenol red
(Gibco) and fetal serum purified with charcoal with dextran were
used. The MTT assay was employed to determine the number of
surviving cells. It consisted in the transformation of MTT
(3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide)
into crystalline formazan. Cells were seeded in 24-well plates
(Corning) by 40000 MCF-7 cells or 60000 T47D cells per one
well under the standard conditions and by 50000 MCF-7 cells
or 70000 T47D cells per one well under the steroid-free condi-
tions. The use of different seeding densities made it possible to
compensate for the difference in their proliferation rates and
create the conditions required for comparing the activities of the
compounds. The compounds under analysis were added to cells
in 12 h. Just before the experiment, the DMSO solutions of the
compounds were mixed with the culture medium in separate
tubes. Then, the working solutions were transferred into wells
with cells. The corresponding amount of the solvent (DMSO)
mixed with the culture medium was added to control cells. The
volume of the added solvent was equal to 1% of the total volume
of the medium for incubation. The concentration of the com-
4. J. P. Wiebe, Endocr. Relat. Cancer., 2006, 13, 717.
5. J. P. Wiebe, G. Zhang, I. Welch, H. T. Cadieux-Pitre, Breast
Cancer Res., 2013, 15, R38, https://doi.org/10.1186/bcr3422.
6. J. P. Wiebe, D. Muzia, J. Hu, D. Szwajcer, S. A. Hill, J. L.
Seachrist, Cancer Res., 2000, 60, 936.
7. J. P. Wiebe, M. Beausoleil, G. Zhang, V. Cialacu, J. Steroid
Biochem. Mol. Biol., 2010, 118, 125.
8. J. P. Wiebe, M. A. Rivas, M. F. Mercogliano, P. V. Elizalde,
R. Schillaci, J. Steroid Biochem. Mol. Biol., 2015, 149, 27.
9. I. S. Levina, A. K. Nazarov, G. V. Nazarov, L. E. Kulikova,
Y. V. Kuznetsov, A. S. Dmitrenok, D. V. Minin, A. V. Aksenov,
I. V. Zavarzin, J. Steroid Biochem. Mol. Biol., 2019, 194,
105436.
10. J. P. Wiebe, C. Deline, K. D. Buckingham, Steroids, 1985,
45, 39.
11. J.-L. Luche, L. Rodriguez-Hahn, P. Crabbe, J. Chem. Soc.,
Chem. Commun., 1978, 601.
12. F. Balssa, M. Fischer, Y. Bonnaire, Steroids, 2011, 76, 667.
13. F. Journe, C. Chaboteaux, J.-C. Dumon, G. Leclercq,
G. Laurent, J.-J. Body, Brit. J. Cancer, 2004, 91, 1703.
14. J. P. Wiebe, M. J. Lewis, BMC Cancer, 2003, 3, 9.
15. J. P. Wiebe, L. Souter, G. Zhang, J. Steroid Biochem. Mol.
Biol., 2006, 100(4—5), 129.
16. M. J. Lewis, J. P. Wiebe, J. G. Heathcote, BMC Cancer, 2004,
4, 27.
17. J. P. Wiebe, K. J. Pawlak, A. Kwok, J. Steroid Biochem. Mol.
Biol., 2016, 155, 166.
pounds in the culture medium was varied from 6 to 50 mol L–1
,
the content of the organic solvent did not exceed 1%. In 72 h of
incubation with an added compound, the medium was sepa-
rated, and the MTT reagent (AppliChem) was added to cells,
and they were incubated for 2 h. Then the cells were lysed in
DMSO, and formazan accumulated by them was dissolved by
light shaking of the plate. Optical density of the obtained solu-
tions was measured with a MultiScan FC spectrophotometer
(ThermoFisher) at 571 nm. The survival rate of the cells treated
with the solvent was taken as 100%. Comparison of the compound
activities was carried out using the ANOVA statistics realized in
the GraphPad program.
18. M. Stefanovic, S. Lajsic, Tetrahedron Lett., 1967, 1777.
19. P. Bunyathaworn, S. Boonananwong, B. Kongkathip, N. Kong-
kathip, Steroids, 2010, 75, 432.
20. The Merck Index, 11th ed., 1989, Merck and Co., Inc., 275.
21. C. J. MacNevin, F. Atif, I. Sayeed, D. G. Stein, D. C. Liotta,
J. Med. Chem., 2009, 52, 6012.
22. R. H. Purdy, A. L. Morrow, J. R. Blinn, S. M. Paul, J. Med.
Chem., 1990, 33, 1572.
23. V. Prelog, E. Tagmann, S. Liebermann, L. Ruzicka, Helv.
Chim. Acta, 1947, 30, 1080.
This work was financially supported by the Russian
Foundation for Basic Research (Project No. 18-29-09017).
References
1. R. Sitruk-Ware, Climacteric., 2018, 21, 315.
2. Steroid Analysis, Eds J. Makin, D. B. Gower, Springer Sci.
and Business Media, 2010, 559.
3. M. Sinreih, S. Zukunft, I. Sosič, J. Cesar, S. Gobec,
J. Adamski, T. L. Rižner, PLoS ONE, 2015, 10(2), e0117984.
Received October 21, 2019;
in revised form November 29, 2019;
accepted December 10, 2019