Synthesis and antiproliferative activity of A-ring aromatised and conduritol-like steroidal compounds

Article abstract of DOI:10.1016/j.steroids.2005.07.001

A simple approach to aromatization of steroidal quinols and epoxyquinols using a catalytic amount of TMSOTf is reported. Beside acetylation of the angular OH, the acid-catalyzed (TfOH) dienone-phenol rearrangement occurred affording "para" products, or in the case of blocked position 4, the acetoxy group 1,2-migration leads to the formation of "meta" products. Using epoxyquinol derivative as a substrate, the acetoxy group elimination was observed, followed by acid-catalyzed epoxy-ring opening and subsequent double bond migration, giving as a final product Δ^9,11A-ring aromatized compounds. Synthesis of conduritol-like compounds and structure confirmation by X-ray crystallography of the precursor of steroidal conduritol is also described. In addition, the results of extensive antiproliferative screening against a panel of 60 cancer cell lines are presented.

Full text of DOI:10.1016/j.steroids.2005.07.001

Steroids 70 (2005) 922–932
Synthesis and antiproliferative activity of A-ring aromatised
and conduritol-like steroidal compounds
a,∗
a
b
´
´
ˇ ´ a
Dragana Milic , Tatjana Kop , Zorica Juranic , Miroslav J. Gasic ,
c
d
a,∗
ˇ
Bernard Tinant , Gabriella Pocsfalvi , Bogdan A. Solaja
a
Faculty of Chemistry University of Belgrade, Studentski trg 12-16, POB 158, 11000 Belgrade, Serbia and Montenegro
b
Institute for Oncology and Radiology of Serbia, Pasterova 14, 11000 Belgrade, Serbia and Montenegro
c
Unite´ CSTR, De´partement de chimie, 1, Place Pasteur, B-1348 Louvain-la-Neuve, Belgium
d
Centro di Spettrometria di massa Proteomica e Biomolecolare, Istituto di Scienze dell’Alimentazione,
Consiglio Nazionale delle Ricerche, Avellino, Italy
Received 18 April 2005; received in revised form 11 July 2005; accepted 13 July 2005
Available online 1 September 2005
Abstract
AsimpleapproachtoaromatizationofsteroidalquinolsandepoxyquinolsusingacatalyticamountofTMSOTfisreported. Besideacetylation
of the angular OH, the acid-catalyzed (TfOH) dienone-phenol rearrangement occurred affording “para” products, or in the case of blocked
position 4, the acetoxy group 1,2-migration leads to the formation of “meta” products. Using epoxyquinol derivative as a substrate, the acetoxy
group elimination was observed, followed by acid-catalyzed epoxy-ring opening and subsequent double bond migration, giving as a final
product ꢀ9,11A-ring aromatized compounds. Synthesis of conduritol-like compounds and structure confirmation by X-ray crystallography of
the precursor of steroidal conduritol is also described. In addition, the results of extensive antiproliferative screening against a panel of 60
cancer cell lines are presented.
© 2005 Elsevier Inc. All rights reserved.
Keywords: Aromatization; Conduritol; Rearrangement; Anticancer activity
1. Introduction
cancer versus the benefits to coronary heart disease and osteo-
porosis suggested that the benefits associated with long-term
estrogen usage decreased after 5–10 years [1]. Thus, the risks
of hormone replacement therapy may outweigh the benefits
after 5–10 years of treatment, especially if the individuals are
predisposed to breast cancer [2].
Phenolic compounds can be modified by cellular sys-
tems in a number of ways and it is considered that the
metabolic oxidation to a quinone moiety could be one of
bioactivating steps. Thus, catechol estrogens are among the
major metabolites of estrogens. If these compounds are oxi-
dized to corresponding quinones, they may react with DNA
forming different types of adducts. Catechol estrogen-2,3-
quinones form stable DNA-adducts, while 3,4-quinones lead
to formation of depurinated adducts causing oncogenic muta-
tions, thereby initiating cancer [3]. At the same time, it has
been shown that endogenous metabolite 2-methoxyestradiol
Estrogens have a number of preventive and therapeutic
effects, including reduced risk of osteoporosis, Alzheimer’s
and ischemic heart disease, improving HDL cholesterol and
fibrinogen level. For many women in the postmenopausal
phase, hormonal replacement therapy has been the treatment
of choice for relief of postmenopausal symptoms and osteo-
porotic fractures. However, estrogens are also associated with
side effects such as bleeding and abdominal bloating, and
more importantly, they have also been shown to significantly
increase the risk of some typesof cancerdevelopment. Recent
analysis of the risks associated with breast and endometrial
∗
Corresponding authors. Tel.: +381 11 638606; fax: +381 11 638606.
E-mail addresses: dmilic@chem.bg.ac.yu (D. Milic´),
bsolaja@chem.bg.ac.yu (B.A. Solaja).
ˇ
0039-128X/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2005.07.001

D. Milic´ et al. / Steroids 70 (2005) 922–932
923
inhibits tumor growth, metastasis and angiogenesis in animal
models in vivo [4]. It is also known that 4-halogen substi-
tuted estradiol derivatives exhibit antitumor activity [5]. As a
result, numerous synthetic efforts are currently under way to
circumvent undesirable effects of estrogens while retaining
the beneficial ones [6]. In addition, it has been also shown
that A-ring estrogens can act as “chemical shield” against
hydroxyl radicals (•OH) by traping them to produce quinols,
which are rapidly converted back to the parent estrogen via an
enzyme-catalyzed reduction in the presence of endogenous
reducing agents NADH and NADPH [7,8].
In our efforts to prepare novel, structurally diverse poly-
oxygenated steroids with different biological properties, we
found that phenols can be easily transformed into the quinols
and epoxyquinols using MCPBA as an oxidant [9,10]. Since
structures with a quinol [11,12] and epoxyquinol moieties
[13–15] represent an important class of biologically active
compounds [7], we expect that transformation products of
these compounds could afford suitable models for further
studies on structure-activity relationships.
1640 medium supplemented with l-glutamine (3 mmol/L),
streptomycin (100 ␮g/mL), and penicillin (100 IU/mL),
10% heat inactivated fetal bovine serum, FBS and 25 mM
Hepes, adjusted to pH 7.2 by bicarbonate solution). The
cells were grown at 37 ◦C in a humidified air atmosphere
with 5% CO2. HeLa and Fem-x cells were seeded in 0.1 mL
of nutrient medium into one group of 96-well microtiter
plates, 2000 cells per well. After 20 h, to one series of wells
various dilutions of investigated compounds were added.
All samples were done in triplicate. Nutrient medium with
corresponding agent concentrations, but without target cells,
was used as blank, also in triplicate. Cells K562 were grown
and treated in the same manner as two other lines, except
that 3000 cells per well were seeded and tested compounds
were added after 4 h. Cell survival was determined by MTT
test according to the method of Mosmann [17], modified
by Ohno and Abe [18], 48 or 72 h after the drug addition.
Briefly, 20 ␮L of MTT solution (5 mg/mL PBS) were added
to each well. Samples were incubated for 4 h more at 37 ◦C
in a humidified atmosphere with 5% CO2. Then, 100 ␮L
of 10% SDS in 0.01 M HCl were added to the wells. The
optical density (OD) at 570 nm was recordered on the next
day. To get cell survival percentage (%), the optical density
at 570 nm of a sample with cells grown in the presence
of various concentrations of agent (OD), was divided with
control optical density ODc (the OD of cells grown only
in nutrient medium) ×100. Concentration IC50 was defined
as the concentration of a drug required for inhibiting cell
survival by 50% compared to vehicle-treated control.
Here, we present a simple approach for the aromatization
of steroidal quinols and epoxyquinols using trimethylsilyl tri-
fluoromethanesulfonate (TMSOTf), and we describe the syn-
thesis of conduritol-like compounds. In addition, the initial
biological screening results and the antiproliferative activity
of some synthesized compounds is also presented.
2. Experimental
2.1. General
2.3. X-ray analysis
Melting points were determined on a Boetius PMHK
apparatus and were not corrected. Specific rotations were
measured on a Perkin-Elmer 141 MC and Karl Zeiss Pola-
mat A polarimeters at the given temperatures. IR spectra
were recorded on Perkin-Elmer spectrophotometer FT-IR
1725X. UV spectra were recorded on a Beckman DU-50
Compound 17 was crystallized by slow evaporation from
a DIPE-acetone solution. The X-ray intensity data were col-
lected at room temperature with a MAR345 image plate
˚
using Mo K␣ (λ = 0.71069 A) radiation from a RU200 rotat-
ing anode. 60 images, with ꢀϕ = 3◦ at a crystal to detector
distance of 130 mm were measured. After integration and
merging this give a total of 10,303 reflections of which 1656
are unique. The unit cell parameters were refined using all
the collected spots after the integration process.
1
spectrophotometer. H NMR and 13C NMR spectra were
recorded on Varian Gemini-200 spectrometer (at 200 and
50 MHz, respectively) in the indicated solvent using TMS
as internal standard. Chemical shifts are expressed in ppm
(δ) values and coupling constants (J) in Hz. Mass spec-
tra were taken on a Finnigan-MAT 8230 and QStar Pulsar
(Applied Biosystems) quadrupole-orthogonal time of flight
(QqQ-TOF), hybrid instrument (LRMS and HRMS, respec-
tively).
The structures were solved by direct methods with
SHELXS97 [19] and refined by full-matrix least-squares on
F2 using SHELXL97 [19]. All the non-hydrogen atoms were
refined with anisotropic temperature factors. All the hydro-
gen atoms were localized by Fourier difference synthesis and
included in the refinement with a common isotropic tempera-
60 cell line assay was done at NIH-NCI (Development
Therapeutics Program) [16].
2
˚
ture factor (Ueq = 0.065 A ). Final R index converge to 0.037
for all the data. Complete X-ray data have been deposited
with the Cambridge Crystallographic Data Centre (Nr CCDC
267384).
2.2. Antiproliferative activity testing towards HeLa,
Fem-x and K562 cells
Human cervix carcinoma, HeLa cells, human melanoma
Fem-x cells, were maintained as a monolayer culture, while
human myelogenous leukemia, K562 cells were grown as
a suspension culture, in the same nutrient medium (RPMI
2.3.1. 10β-Acetoxyestra-1,4-dien-3,17-dione (2) and
1,4-diacetoxyestra-1,3,5(10)-trien-17-one (3)
To an ice bath cooled solution of quinol 1 [9] (200 mg,
0.70 mmol) in dichloromethane (10 mL), Ac2O (100 ␮L,

924
D. Milic´ et al. / Steroids 70 (2005) 922–932
1.06 mmol) and TMSOTf (10 ␮L, 55.2 ␮mol) were added
dropwise and reaction mixture was stirred at 0 ◦C for 5 min
(A), or at room temperature for 45 min (B). After wash-
ing with saturated NaHCO3 (5 × 20 mL) and water (10 mL),
the solution was dried over anh. Na2SO4, filtered and
evaporated.
ture of the reflux (C). After washing with saturated NaHCO3
(4 × 15 mL), the solution was dried over anh. Na2SO4,
filtered and evaporated. Dry-flash chromatography on SiO2
column using PhMe/EtOAc 7/3 as an eluant and subsequent
crystallization from DIPE afforded 147 mg (50%) (A),
161 mg (55%) (B), 133 mg (45%) (C) of 1,3-diacetoxy-
4-bromoestra-1,3,5(10)-triene-17-one (5) as colorless
needles.
A: chromatography on Lobar B, SiO2 column, using
PhMe/EtOAc 7/3 as an eluant, and subsequent crystal-
lization from MeOH, afforded 10␤-acetoxyestra-1,4-dien-
3,17-dione (2; 28 mg, 12%) as colorless plates, and 1,4-
diacetoxyestra-1,3,5(10)-trien-17-one (3; 116 mg, 45%) as
colorless needles.
5: mp = 174–176 ◦C (colorless needles, DIPE). IR(KBr):
3462(m, broad), 2944(m), 2880(m), 2360(w), 1775(s),
1737(s), 1635(w), 1592(w), 1447(m), 1371(m), 1192(s),
1149(m) cm−1. MS (EI, 70 eV, m/z (%)): 452 (MH++2
(100)), 450 (MH+ (100)). Anal. calcd. for C22H25BrO5
(449.35): C 58.81, H 5.61, found C 58.82, H 5.62. 1H
NMR(200 MHz, CDCl3), δ:6.74(s, H–C(1)), 3.20–2.40(m,
6H), 2.34 (s, CH3COO–C(Ar)), 2.27 (s, CH3COO–C(Ar)),
0.93 (s, H3C–C(13)). 13C NMR (50 MHz, CDCl3), δ:
220.21 (C(17)), 168.51 (CH3COO–C(1), CH3COO–C(3)),
148.95 (C(3)), 146.29 (C(1)), 140.47 (C(5)), 131.91
(C(10)), 116.80 (C(4)), 115.85, 50.04, 48.20, 44.74,
39.17, 35.56, 32.87, 32.08, 25.77, 24.69, 22.80, 21.45,
21.09, 20.76 (CH3COO–C(1), CH3COO–C(3)), 14.21
(C(18)).
B: after filtration through a short plug of silica gel and crys-
tallization from MeOH, 163 mg (63%) of diacetate 3 was
obtained.
2: mp = 258–260 ◦C (colorless plates, MeOH). IR (KBr):
2942(m), 1737(s), 1663(s), 1630(s), 1606(m), 1250(m),
1236(s) cm−1
.
λ
= 204(2115), 246 nm (5900).
MeOH
max
[α]2D2 = +36 (c = 1.2, chl). HRMS: Calcd. for MH+
1
329.1753, found 329.1746. H NMR (200 MHz, CDCl3),
δ: 6.94 (d, J = 10.4 Hz, H–C(1)), 6.31 (dd, J1,2 = 10.4 Hz,
J2,4 = 2.2 Hz, H–C(2)), 6.15 (t, J = 1.5 Hz, H–C(4)),
2.60–2.40 (m, 2H), 2.13 (s, CH3COO–C(10)), 0.98 (s,
H3C–C(13)). 13C NMR (50 MHz, CDCl3), δ: 219.73
(C(17)), 185.15 (CH3COO–C(10)), 168.68 (C(3)), 162.10
(C(5)), 147.68 (C(1)), 129.33 (C(2)), 123.92 (C(4)), 76.95
(C(10)), 56.44 (C(9)), 49.64 (C(14)), 47.50 (C(13)),
35.36 (C(16)), 34.79 (C(8)), 32.76 (C(7)), 31.58 (C(6)),
30.73 (C(11)), 22.52 (C(15)), 21.74 (C(12)), 20.85
(CH3COO–C(10)), 13.52 (C(18)).
2.3.3. 10β-Acetoxy-1β,2β-epoxyestr-4-en-3,17-dione
(7) and 2,3-diacetoxyestra-1,3,5(10),9(11)-tetraen-
17-one (8)
To an ice bath cooled solution of epoxyquinol 6
[21] (50 mg, 0.16 mmol) in dichloromethane (3 mL), Ac2O
(24 ␮L, 0.25 mmol) and TMSOTf (2.4 ␮L, 13.2 ␮mol) were
added dropwise and the reaction mixture was stirred at 0 ◦C
for5 min(A), oratroomtemperatureover4 h(B). Afterwash-
ing with saturated NaHCO3 (5 × 10 mL) and water (10 mL),
the solution was dried over anh. Na2SO4, filtered and evapo-
rated to dryness.
3: mp = 162–164 ◦C (colorless needles, MeOH). IR (KBr):
3010(w), 2939(m), 1762(s), 1741(s), 1236(s) cm−1
.
MeOH
λ
= 205(6270), 266 (230), 271 (200), 319 nm (200).
max
[α]2D7 = +293.8 (c = 0.8, chl). HRMS: calcd. for MNa+
393.1678, found393.1682. Anal. calcd. forC22H26O5 × 1/4
MeOH (378.45): C 70.61, H 7.19, found: C 70.98, H
6.87. 1H NMR (200 MHz, CDCl3), δ: 6.94, 6.82 (AAꢀBBꢀ,
J = 8.7 Hz, H–C(2) and H–C(3)), 2.85–2.65 (m, 2H),
2.65–2.35 (m, 3H), 2.31, 2.28 (both s, CH3COO–C(1)
and CH3COO–C(4)), 0.93 (s, H3C–C(13)). 13C NMR-
DEPT (50 MHz, CDCl3), δ: 220.18 (C(17)), 169.09, 168.94
(CH3COO–C(1), CH3COO–C(4)), 147.40, 146.53 (C(5),
C(10)), 133.17, 132.08 (C(2), C(3)), 121.00, 119.97 (C(1),
C(4)), 49.98 (CH), 48.11 (C(13)), 44.66 (CH), 49.20 (CH),
35.43 (CH2), 32.03 (CH2), 25.42 (CH2), 24.95 (CH2),
23.93 (CH2), 21.36 (CH2), 21.02, 20.63 (CH3COO–C(1),
CH3COO–C(4)), 14.12 (C(18)).
A: 10␤-Acetoxy-1␤,2␤-epoxyestr-4-en-3,17-dione (7;
33.9 mg, 59%) was obtained as colorless oil.
B: chromatography on Lobar A SiO2 column with
PhMe/EtOAc 7/3 as an eluant afforded 10␤-acetoxy-1␤,2␤-
epoxyestr-4-en-3,17-dione (7; 18.7 mg, 34%), and 2,3-
diacetoxyestra-1,3,5(10),9(11)-tetraen-17-one (8; 5.9 mg,
10%, crystallized from MeOH/EtOH).
7: colorless oil IR(KBr): 3025(w), 1735(s), 1675(s),
MeOH
1610(w), 1283(w), 1230(s), 847(m) cm−1. λ
=
max
225 (9800), 231 (10000), 230 nm (10050). MS (DCIMS,
150 eV, m/z (%)): 345 (MH+, 100), 285 (MH+-AcOH,
30). Anal. calcd. for C20H24O5: C 69.75 H 7.02,
found: C 69.72 H 7.14. 1H NMR (200 MHz, CDCl3),
δ: 5.81 (d, J = 2.2 Hz, H–C(4)), 4.09 (d, J = 3.6 Hz,
H␣–C(1)), 3.44 (dd, J1,2 = 3.6 Hz, J2,4 = 2.2 Hz, H␣–C(2)),
2.60–2.20 (m, 4H), 2.20 (s, CH3COO–C(10)), 0.97 (s,
H3C–C(13)). 13C NMR (50 MHz, CDCl3), δ: 219.69
(C(17)), 191.92 (C(3)), 168.58 (CH3COO–C(10)),162.57
(C(5)), 119.86 (C(4)), 78.98 (C(10)), 56.13, 55.19 (C(1)),
53.53 (C(2)), 49.85, 47.23 (C(13)), 35.43, 35.09, 33.23,
2.3.2. 1,3-Diacetoxy-4-bromoestra-1,3,5(10)-trien-
17-one (5)
To an ice bath cooled solution of bromoquinol 4 [20]
(240 mg, 0.66 mmol) in dichloromethane (25 mL), Ac2O
(153 ␮L, 1.62 mmol) and TMSOTf (31 ␮L, 171.2 ␮mol)
were added dropwise and reaction mixture was stirred for
20 min at 0 ◦C (A), room temperature (B) or at the tempera-

D. Milic´ et al. / Steroids 70 (2005) 922–932
925
32.96, 30.68, 21.80(CH3COO–C(10)), 21.47, 20.83, 13.66
(C(18)).
NaHCO3 (5 × 20 mL) and water (10 mL), the solution was
dried over anh. Na2SO4, and filtered through a short silica gel
column. Crystallization from MeOH/benzene afforded 3,4-
diacetoxyestra-1,3,5(10),9(11)-tetraen-17-one (11; 170 mg,
70%) as colorless needles.
8: mp = 158–161 ◦C (colorless needles, MeOH/EtOH).
IR(KBr): 3014(w), 1767(s), 1741(s), 1623(w),
chl
1207(s) cm−1
.
λ
= 259(15260), 294 nm (4390).
max
[α]2D7 = +198.1 (c = 0.1, chl). MS (DCIMS, 150 eV,
m/z (%)): 369(MH+, 100). Anal. calcd. for C22H24O5
(368.43): C 71.72 H 6,57, found: C 71.67 H 6.54.. 1H
NMR (200 MHz, CDCl3), δ: 7.37 (s, H–C(4)), 6.92
(s, H–C(1)), 6.23–6.15 (m, H–C(11)), 2.88–2.84 (m,
2H), 2.63–2.43 (m, 2H), 2.29 (s, CH3COO–C(2)), 2.28
(s, CH3COO–C(3)), 0.92 (s, H3C–C(13)). 13C NMR
(50 MHz, CDCl3), δ: 221.29 (C(17)), 168.67, 168.51
(CH3COO–C(2), CH3COO–C(3)), 140.61 (C(2)), 140.16
(C(3)), 134.64 (C(10)), 133.06 (C(5)), 123.50 (C(4)),
120.15 (C(1)), 118.68 (C(9)), 117.00 (C(8)), 47.61 (C(14)),
46.03 C(13)), 37.58, 36.11, 33.90, 29.04, 27.31, 22.38,
20.58, 20.54 (CH3COO–C(2), CH3COO–C(3)), 14.38
(C(18)).
11: mp = 187–189 ◦C (colorless needles, MeOH/PhH).
IR(KBr): 3027(w), 1785(s), 1735(s), 1624(w), 1200(s)
MeOH
cm−1. λ
= 212(25630), 224 (33210), 258 (23820),
max
319 nm (13). [α]2D7 = +258.3 (c = 0.8, chl). MS (DCIMS,
150 eV, m/z (%)): 369 (MH+, 100). Anal. calcd. for
C22H24O5 × 1/3 MeOH: C 70.76 H 6,74, found: C 70.83 H
1
6.66. H NMR (200 MHz, CDCl3), δ: 7.51 (d, J = 8.9 Hz,
H–C(2)), 7.00 (d, J = 8.9 Hz, H–C(1)), 6.32–6.24 (m,
H–C(11)), 3.55–2.87 (m, 1H), 2.70–2.37 (m, 2H), 2.40
(s, CH3COO–C(3)), 2.32 (s, CH3COO–C(4)), 0.92 (s,
H3C–C(13)). 13CNMR–DEPT(50 MHz, CDCl3), δ:221.20
(C(17)), 168.51, 168.03 (CH3COO–C(3), CH3COO–C(4)),
140.90(C(4)), 140.18(C(3)), 134.81(C(10)), 133.62(C(9)),
130.07 (C(5)), 122.32 (C(1)), 120.48 (C(11)), 120.06
(C(2)), 47.62 (C(14)), 46.01 C(13)), 37.16 (CH), 36.09
(CH2), 33.94 (CH2), 26.57 (CH2), 23.25 (CH2), 22.36
(CH2), 20.58, 20.21 (CH3COO–C(3), CH3COO–C(4)),
14.30 (C(18)).
2.3.4. 10β-Acetoxy-4β,5β-epoxyestr-1-en-3,17-dione
(10)
To an ice bath cooled solution of epoxyquinol 9 [9]
(200 mg, 0.66 mmol) in dichloromethane (10 mL), Ac2O
(94 ␮L, 1.00 mmol) and TMSOTf (9.4 ␮L, 51.9 ␮mol) were
added dropwise and the reaction mixture was stirred at 0 ◦C
for5 min. AfterwashingwithsaturatedNaHCO3 (5 × 20 mL)
and water (10 mL), the organic layer was dried over anh.
Na2SO4, filtered and evaporated to dryness. Crystallization
from MeOH/THF afforded 10␤-acetoxy-4␤,5␤-epoxyestr-1-
en-3,17-dione (10; 220 mg, 95%) as colorless plates.
2.3.6. 10β-Acetoxy-2-bromo-4β,5β-epoxyestr-1-en-3,
17-dione (13)
To a solution of bromoepoxyquinol 12 [20] (27.3 mg,
0.072 mmol) in dichloromethane (1 mL), at room tem-
perature, Ac2O (19.6 ␮L, 0.144 mmol) and TMSOTf
(0.17 ␮L, 2.0 ␮mol) were added dropwise and the reac-
tion mixture was stirred for 5 min, and then diluted
with dichloromethane (5 mL). After washing with satu-
rated NaHCO3 (3 × 3 mL), the solution was dried over
anh. Na2SO4, filtered and evaporated. Crystallisation
from DCM/DIPE afforded 10␤-acetoxy-2-bromo-4␤,5␤-
epoxyestr-1-en-3,17-dione (13; 27.1 mg, 89%) as colorless
needles.
10: mp = 276–283 ◦C (dec., colorless plates, MeOH/THF).
IR(KBr): 3023(w), 1738(s), 1680(s), 1614(w), 1285(w),
MeOH
1230(s), 843(m) cm−1. λ
= 223(10600), 228 (11140),
max
230 nm (11250). [α]D22 = +43, 5 (c = 0.4, MeOH). MS
(DCIMS, 150 eV, m/z (%)): 345 (MH+, 100), 285 (MH+-
AcOH, 30). Anal. calcd. for C20H24O5: C 69.75 H 7.02,
found: C 69.73 H 7.16. 1H NMR (200 MHz, CDCl3),
δ: 6.83 (d, J = 11.0 Hz, H–C(1)), 5.92 (dd, J1,2 = 11.0 Hz,
J2,4 = 2.2 Hz, H–C(2)), 3.26 (d, J = 2.2 Hz, H␣–C(4)),
2.60–2.25 (m, 3H), 2.20 (s, CH3COO–C(10)), 0.96 (s,
H3C–C(13)). 13C NMR (50 MHz, CDCl3), δ: 219.70
(C(17)), 194.20 (C(3)), 168.92 (CH3COO–C(10)),148.66
(C(1)), 122.26 (C(2)), 80.22 (C(10)), 62.87 (C(5)), 59.27
(C(4)), 56.15, 49.85, 47.41 (C(13)), 35.43, 34.60 (C(8)),
30.56, 28.52 (C(6)), 28.39, 22.00, 21.90 (CH3COO–C(10)),
13.73 (C(18)).
13: mp = 235–238 ◦C softens, 300 ◦C dec. (colorless nee-
dles, DCM/DIPE). IR(KBr): 3451 (m), 3034(w), 2920 (s),
1734(s), 1695(s), 1233(s) cm−1. MS (DCIMS, 150 eV,
m/z (%)): 450 (MH++2 (100)), 448 (MH+ (100)). Anal.
calcd. for C20H23BrO5: C 56.74 H 5.49, found: C
56.25 H 5.47. 1H NMR (200 MHz, CDCl3), δ: 7.24 (s,
H–C(1)), 3.43 (s, H–C(4)), 2.21 (s, CH3COO–C(10)),
0.95 (s, H3C–C(13)). 13C NMR (50 MHz, CDCl3), δ:
219.39 (C(17)), 187.56 (CH3COO–C(10)), 168.73 (C(3)),
148.69 (C(1)), 115.42 (C(3)), 81.12 (C(10)), 62.92 (C(5)),
58.31 (C(4)), 56.06, 49.75, 47.33, 35.38, 34.61, 30.49,
28.37, 28.07, 21.65, 21.36, 21.14 (CH3COO–C(10)), 13.61
(C(18)).
2.3.5. 3,4-Diacetoxyestra-1,3,5(10),9(11)-tetraen-
17-one (11)
To an ice bath cooled solution of epoxyquinol 9 [9]
(200 mg, 0.66 mmol) in dichloromethane (10 mL), Ac2O
(94 ␮L, 1.00 mmol) and TMSOTf (9.4 ␮L, 51.9 ␮mol) were
added dropwise and the reaction mixture was stirred at
room temperature for 6 h. After washing with saturated
2.3.7. 3,4-Diacetoxy-2-bromoestra-1,3,5(10),9(11)-
tetraen-17-one (14)
To a refluxing solution of bromoepoxyquinol 12 [20]
(110 mg, 0.347 mmol) in dichloromethane (7 mL), TMSOTf

926
D. Milic´ et al. / Steroids 70 (2005) 922–932
(13 ␮L, 0.168 mmol) was added in one portion. Ac2O
(82.5 ␮L, 0.608 mmol) was added after 15 min of reflux and
the reaction mixture was maintained at the same tempera-
ture for a further 2.5 h. Cooling to room temperature and
dilution with dichloromethane (10 mL), was followed by
washing with saturated NaHCO3 (4 × 7 mL), drying over
anh. Na2SO4, filtering and evaporation. Chromatography on
Lobar A SiO2 column with PhMe/EtOAc 7/3 as an eluant
and subsequent crystallization from MeOH afforded 3,4-
diacetoxy-2-bromoestra-1,3,5(10),9(11)-tetraen-17-one (14;
100 mg, 77%) as colorless plates.
2.3.9. 4β,5β,10β-Trihydroxyestr-1-en-3,17-dione (17)
To a solution of quinol 1 [9] (250 mg, 0.87 mmol)
and N-methylmorpholine-N-oxide monohydrate (470 mg,
3.48 mmol) in acetone/water mixture (4/1, 100 mL), the
solution of osmium-tetraoxide in tert-butanol (11.0 mg,
0.04 mmol, 0.44 mL of 2.5% solution in tBuOH) was added
dropwise, and the reaction mixture was stirred at room tem-
perature, under Ar for 5 days. After evaporation of acetone,
the residue was diluted with brine and extracted with CHCl3
(3 × 50 mL). The combined extracts were dried over anh.
Na2SO4 and filtered through short plug of silica gel. Chro-
matography on Lobar B SiO2 column using PhMe/EtOAc 7/3
as an eluant, afforded educt (72.5 mg, 29%) and 4␤,5␤,10␤-
trihydroxyestr-1-en-3,17-dione (17; 140.8 mg, 50%), crys-
tallised from acetone/DIPE as colorless needles.
14: mp = 168 ◦C (dec., colorless plates, MeOH). IR(KBr):
3640 (s), 2965(m), 1775(s), 1739(s), 1205 (s) cm−1. MS
(EI, 70 eV, m/z (%)): 447 (M+), 405, 363, 303, 43(100).
Anal. calcd. for C22H23BrO5 × 1/2MeOH: C 58.32 H 5.45,
17: mp = 211–220 ◦C (colorless needles, acetone/DIPE).
1
found: C 58.16 H 5.20. H NMR (200 MHz, CDCl3), δ:
IR(KBr): 3680–3160(s), 1745(s), 1702(s), 1645(w) cm−1
.
7.74 (s, H–C(1)), 6.35–6.23 (m, H–C(11)), 2.90–2.70 (m,
1H), 2.60–2.40 (m, 2H), 2.34, 2.32 (both s, CH3COO–C(3),
CH3COO–C(4)), 0.91 (s, H3C–C(13)). 13C NMR-DEPT
(50 MHz, CDCl3), δ: 220.92 (C(17)), 167.78, 167.56
(CH3COO–C(3), CH3COO–C(4)), 141.76, 139.30, 134.81,
133.95, 129.62, 126.03, 121.48, 114.43, 68.28, 47.58,
46.01, 37.15, 36.07, 34.00, 26.35, 23.24, 22.78, 22.36,
20.30, 20.21 (CH3COO–C(3), CH3COO–C(4)), 14.32
(C(18)).
MeOH
λ
= 202(570), 218 nm (780). [α]2D7 = +110, 4
max
(c = 0.6, chl). HRMS: Calcd. for MH+ 321.1702, found
321.1697; calcd. for MNa+ 343.1521, found 343.1523. 1H
NMR (200 MHz, CDCl3), δ: 6.92 (d, J = 10.4, H–C(1)),
6.10 (d, J = 10.4, H–C(2)), 4.44 (s, H␣–C(4)), 3.75 (br.
s, exchangeable with D2O, HO–C(4)), 3.27, 2.53 (both
s, exchangeable with D2O, HO–C(5) and HO–C(10)),
2.57–2.38(m, 1H), 2.25–2.05(m, 1H), 0.96(s, H3C–C(13)).
13C NMR (50 MHz, CDCl3), δ: 220.55 (C(17)), 198.40
(C(3)), 155.10 (C(1)), 124.10 (C(2)), 80.34, 74.32 (C(5),
C(10)), 73.31 (C(4)), 50.11 (CH), 49.81 (CH), 47.58
(C(13)), 35.55 (CH2), 34.05 (CH), 31.09 (CH2), 29.80
(CH2), 26.30 (CH2), 21.78 (CH2), 21.10 (CH2), 13.71
(C(18)).
2.3.8. 2,3-Dihydroxyestr-5(10)-en-1,4,17-trione (16)
To a solution of quinone 15 [9] (600 mg, 2.11 mmol)
and N-methylmorpholine-N-oxide monohydrate (1.14 g,
8.44 mmol) in acetone/water mixture (4/1, 150 mL), the
solution of osmium-tetraoxide in tert-butanol (26.8 mg,
0.10 mmol, 1.08 mL of 2.5% solution in tBuOH) was added
dropwise, and the reaction mixture was stirred at room tem-
perature, under Ar for 6 h. After evaporation of acetone, the
residue was diluted with brine and extracted with EtOAc
(5 × 25 mL). The combined extracts were dried over anh.
Na2SO4, filtered through short plug of silica gel and chro-
matographed on Lobar B SiO2 column using PhMe/EtOAc
7/3 as an eluant. Crystallisation from acetone/cyclohexane
gave 2,3-dihydroxyestr-5(10)-en-1,4,17-trione (16; 160 mg,
22%) as colorless needles.
2.3.10. 3β,4β,17β-Triacetoxyestr-1-en-5β,10β-diol (18)
A flask dried under a flame-vacuum system was charged
with a solution of triol 17 (450 mg, 1.40 mmol) in dry
THF (20 mL) under Ar. The reaction mixture was cooled
at −15 ◦C, and a solution of DIBAL-H (9.36 mL of 1.5 M
solution in toluene, 14.0 mmol) was added dropwise and
with vigorous stirring. After 15 min, the formed aluminate
complex was carefully destroyed with aqueous methanol
(MeOH/H2O 1/4, 10 mL) and the reaction mixture stirred
at room temperature for additional 2 h. The obtained suspen-
sion was filtered, washed with cold MeOH (5 mL) and THF
(5 mL), and the filtrate was evaporated to dryness and dried.
The crude product was dissolved in Ac2O/Py mixture (2/1,
18 mL) and stirred at room temperature for 12 h, poured onto
ice/HCl, extracted with CH2Cl2 (3 × 10 mL), washed with
10% NaHCO3 (2 × 5 mL), water (15 mL), and dried over
anh. Na2SO4. Chromatography on Lobar B SiO2 column
with PhMe/EtOAc 7/3 as an eluant, and subsequent crys-
tallization from acetone/DIPE mixture afforded 3␤,4␤,17␤-
triacetoxyestr-1-en-5␤,10␤-diol (18; 160 mg, 25%) as color-
less needles.
16:
mp = 172–175 ◦C
(colorless
needles,
ace-
tone/cyclohexane). IR(KBr): 3630–3210(m), 1730(s),
MeOH
1702(s), 1603(w) cm−1. λ
= 204(210), 254 nm
max
(230). [α]2D7 = +386, 3 (c = 0.6, chl). HRMS: Calcd. for
MH+ 319.1545; found 319.1530. 1H NMR (200 MHz,
CDCl3), δ: 4.62, 4.47 (both d, J = 3.1, H–C(2) and
H–C(3)), 2.65–2.50 (m, 2H), 2.50–2.35 (m, 2H), 0.92
(s, H3C–C(13)). 13C NMR (50 MHz, CDCl3), δ: 220.01
(C(17)), 197.42 (C(4)), 195.78 (C(1)), 150.82 (C(10)),
144.36 (C(5)), 79.02 (C(2)), 76.09 (C(3)), 49.71 (C(14)),
48.22 (C(13)), 43.10 (C(9)), 37.49 (C(8)), 35.57 (C(16)),
31.62 (C(12)), 24.54 (C(11)), 23.83 (C(6)), 23.64 (C(7)),
21.39 (C(15)), 14.09 (C(18)).
18: mp = 192–193 ◦C (colorless needles, acetone/DIPE).
IR(KBr): 3680–3120(s), 1745(s), 1715(s), 1256(s), 1226(s),

D. Milic´ et al. / Steroids 70 (2005) 922–932
927
performed at low temperature (0 ◦C) afforded only the acety-
lated epoxyquinol 7 (59%), while prolonging the reaction
time at room temperature led to a mixture of products, acetate
7 and diacetate 8 (34 and 10%, respectively; Scheme 1). In
order to obtain the aromatized compound as the only prod-
uct, increasing the reaction temperature resulted in formation
of a unresolvable complex mixture containing diacetate 8.
The reaction with regioisomeric epoxyquinol 9 proceeded
cleanly, affording only acetylated product 10 at low tempera-
ture (Scheme 1), while at ambient temperature the rearranged
product, the diacetate 11, was isolated in good yield. This
indicates that the epoxy ring at position 4,5 is more suscep-
tible to acid catalyzed opening as compared with the same
functionality in position 1,2. It can be suggested that the rear-
ranged diacetate 11 was formed in analogy to compound 8
(see Fig. 3). Upon elimination of the substituent at C(10)
(OH or OAc), the acid catalyzed opening of the oxirane ring
led to elimination of ␥-hydrogen affording the ꢀ9,11 dou-
ble bond. Tautomerization and subsequent acetylation finally
gave diacetate 11.
Introduction of bromine at position 2 leads to decreased
substrate reactivity towards acetylation, as well as to the
aromatization via epoxy ring opening and subsequent dou-
ble bond migration, while the reaction selectivity remains.
Namely, no reaction was observed when bromoepoxyquinol
12 was treated with acetic anhydride/TMSOTf at low tem-
perature. At the ambient temperature, only simple acetylation
has occurred yielding the compound 13 in good yield (89%),
while at reflux, the rearranged diacetate 14, formed in anal-
ogy to compounds 8 and 11, was isolated as the sole product
(Scheme 1, 77%).
In addition, we examined the synthesis of conduritol-like
steroidal compounds. Conduritols are simple tetrahydroxy-
cyclohexenes, two of six possible isomers are natural prod-
ucts, and all of them were synthesized. They are glycosidase
inhibitors [27], consequently they can be used as therapeutic
agents in metabolic processes involving glycoproteins. Also,
they exhibit antibiotic, anti-leukemic, and tumor-inhibiting
activity [28], so their incorporation into a steroidal frame-
work could be of interest for structure-activity studies.
Conduritol-like compounds were prepared from quinol
1, (a) by quinone synthesis followed by dihydroxylation to
obtain 16, and (b) by dihydroxylation of quinol 1 and subse-
quent reduction to yield 18 (Scheme 2).
Steroidal quinone 15 was prepared in an excellent yield
according to our previously published procedure [9]. Treat-
ing of compound 15 with a catalytic amount of OsO4 in the
presence of NMO gave as a sole product dihydroxyentrione
16 in modest yield (22%). Obtained spectral data indicate
that hydroxylation occurred at the less substituted double
bond. However, the stereochemistry of the formed compound
could not be determined using NOE experiment, due to long
distance of hydroxyl protons (as well as of C(2)- and C(3)-
adjacent hydrogens) and all NOE-related ones.
1645(w) cm−1. [α]D27 = +95.9 (c = 1.2, chl). HRMS:
Calcd. for MH+ 451,2332, found 451.2369. 1H NMR
(200 MHz, CDCl3), δ: 6.07 (d, J = 9.6, H–C(1)), 5.75–5.61
(m, H–C(2) and H–C(3)), 5.51 (d, J = 4.6, H–C(4)),
4.57 (dd, J = 7.5, 1.4, H␣–C(17)), 2.12, 2.10, 2.04 (all
three s, CH3COO–C(3), CH3COO–C(4), and CH3COO–
C(17)), 0.96 (s, H3C–C(13)). 13C NMR (62.9 MHz,
CDCl3), δ: 171.14 (CH3COO–C(17)), 169.82, 169.51
(CH3COO–C(3), CH3COO–C(4)), 139.27 (C(1)), 120.58
(C(2)), 82.56, 75.02, 74.67, 67.05, 65.56 (C(3), C(4), C(5),
C(10), C(17)), 50.34, 49.32, 42.54, 36.46, 34.12, 28.21,
27.36, 26.42, 23.36, 21.24, 21.09, 20.84, 20.46, 13.71
(C(18)).
3. Results
3.1. Chemistry
TMSOTf is often used as a catalyst in synthetic trans-
formations, including formation of acetals from silylated
alcohols [22], stereoselective allylation [23] and imino-ene
reaction [24]. Also, the efficient acylation of alcohols, includ-
ing tertiary and sterically hindered ones, can be accomplished
using TMSOTf-catalyzed processes [25].
When the TMSOTf-catalyzed acetylation of steroidal
quinol 1 [10] was performed at low temperature (0 ◦C), even
in 5 min the mixture of products was formed consisting of
a small amount of expected acetate 2 (12%) and the rear-
ranged diacetate 3 (45%, Scheme 1). In a prolonged reaction
at the same temperature (45 min), the rearranged compound
3 was formed as a single product in good yield (63%).
We assume that the liberated trifluoromethanesulfonic acid
(TfOH) [25] catalyzes dienone-phenol rearrangement.1 In
this way, instead of two low yielding steps2, hydroquinone
diacetate 3 was obtained directly from the parent quinol.
This result prompted us to investigate this process in more
detail, and to that purpose, several steroidal quinols and
epoxyquinols were exposed to TMSOTf catalyzed acetyla-
tion (Scheme 1).
Under the same reaction conditions, 4-substituted quinol
4, due to the good migration ability of the formed acetoxy
group at C(10), underwent the dienone-phenol rearrange-
ment affording the “meta” product 5. At all examined reaction
temperatures (0 ◦C, ambient and reflux), the same aromatic
compound was formed as the only product. In the case of
epoxyquinol 6 [21], products and their distribution obvi-
ously depend on reaction temperature. Thus, the reaction
1
Quinol 1 was treated with cat. amount of TfOH affording corresponding
hydroquinone (not isolated; its Rf value (TLC) was identical to the authentic
sample), which was oxidized into the well-known quinone 15 (Scheme 2).
2
Until now, the usual pathway for hydroquinone diacetate synthesis was
acid-catalyzed dienone-phenol rearrangement of a quinol, followed by sub-
sequent alkanoylation [26]. According to this procedure (Ac2O/Py, r.t.), only
small amount of corresponding diacetate 3 can be isolated from the complex
reaction mixture.
Steroidal analogue of conduritol-D was successfully syn-
thesized according to the second proposed reaction pathway

928
D. Milic´ et al. / Steroids 70 (2005) 922–932
Scheme 1. TMSOTf-catalyzed transformations of steroidal quinols and epoxyquinols. Ac2O, TMSOTf/DCM (A) 0 ◦C; (B) r.t.; (C) rfl.
(Scheme 2). Treating quinol 1 with a catalytic amount of
OsO4 in the presence of NMO for 5 days afforded all-cis-triol
17 in 50% yield. The ability of the hydroxyl group induced
approach of an incoming reagent by hydrogen bonding, and
consecutive formation of all-cis-triol in polar protic solvent
(acetone/water) should not be expected [29]. In addition,
semiempirical calculations indicate that in the most stable
conformation, the ␣-side of the molecule was more hindered
[30], suggesting that the stereochemistry of this process is
determined by the reagent approach from the less hindered,
␤-side The structure of compound 17 was confirmed by X-
ray crystallography. Reaction was not completed (29% of
educt was isolated), probably due to reduced reactivity of
conjugated enones. Reduction of diketotriol 17 afforded as
main product the steroidal analogue of conduritol-D, char-
acterized as its diacetate 18. Hydride approach from the less

D. Milic´ et al. / Steroids 70 (2005) 922–932
929
Scheme 2. Synthesis of steroids with A-ring incorporated conduritol subunit. (i) (a) HCl/AcOH/H2O (5/15/2), rfl, 30ꢀ; (b) Ag2O/THF, r.t., 1 h. (ii)
OsO4/NMO/acetone/H2O, Ar, r.t., 6 h (A), 5 days (B); (iii) DIBAL-H/THF, Ar, −15 ◦C, 15ꢀ; (b) Ac2O/Py (1/2), r.t., 12 h.
hindered ␣-side led to the formation of 3␤-alcohol and, con-
sequently, to coresponding acetate. Expected ␤-configuration
of acetoxy group at C(3) was confirmed by NMR spec-
troscopy: doublet at 6.07 ppm (J = 9.6 Hz) belongs to H–C(1)
and its J-value is a measure of scalar coupling with adja-
cent, vinylic hydrogen. The lack of allylic coupling between
H–C(1) and H–C(3) indicates perpendicular orientation of
the ␴-orbital of the later bond and ␲-electronic double bond
system, provingtheequatorial, ␣-orientationofH–C(3). Mul-
tiplet at 5.75–5.61 ppm represents superposition of peaks
belonging to the vinylic H–C(2) and allylic H–C(3). Dou-
blet at 5.51 ppm (J = 4.6 Hz) belongs to the axially oriented
H␣–C(4) (configuration of which was confirmed by the X-
ray crystallography of the parent compound) coupled to
H–C(3). Small J-value is the measure of a,e-vicinal cou-
pling, confirming once more the equatorial, ␣-orientation of
H–C(3).
3.2. X-ray analysis of compound 17
Ring A has a half-chair conformation with ꢀC2 = 4.0◦
[31], while ring D is an envelope with C14 at the flap,
ꢀCs = 3.6◦ [31]. Three intermolecular hydrogen bonds are
observed, with following geometry:
D–H
d(D–H) d(H· · ·A) <DHA d(D· · ·A)
A
(◦)
(A)
˚
(A)
˚
(A)
˚
O18–H18 0.90
O21–H21 0.95
1.87
1.84
159
168
2.734
2.784
O21 [x + 1, y, z]
O22
[x − 1/2,
−y + 1/2, −z]
O22–H22 0.86
2.15
128
2.768
O20
[x + 1/2,
−y + 1/2, −z]
Crystal data collection and refinement parameters are
summarized in Table 1, while the Fig. 1 represents ORTEP
[32] plot of compound 17.
Fig. 1. ORTEP [32] plot of compound 17.

930
D. Milic´ et al. / Steroids 70 (2005) 922–932
Table 1
Table 2
Data collection and refinement parameters for compound 17
Antiproliferative activity of compound 11 towards 60 cell lines
Formula
C18H24O5
Panel/cell line
Activity (␮M)
a
b
Mr
320.37
Orthorhombic
P212121
6.425(3)
12.520(2)
20.565(8)
1654(1)
1.286
GI50
LC50
System
Space group
a (A)
b (A)
c (A)
V (A )
Dx (g cm−3
Z
Leukemia
CCRF-CEM
HL-60 (TB)
K-562
MOLT-4
RPMI-8226
SR
3.4
6.9
14.0
3.3
1.0
6.6
>100
>100
>100
>100
˚
˚
˚
˚
3
46.4
>100
4
˚
Non-small cell lung cancer
A549/ATCC
EKVX
λ (A)
0.71069
688
0.093
16.0
10.5
19.7
5.2
21.4
21.7
16.4
1.8
71.1
85.9
F(000)
␮ (mm−1
HOP-62
HOP-92
NCI-H23
NCI-H322M
NCI-H460
>100
56.3
>100
>100
99.1
Approximate crystal size (mm)
Θmax for data collection (◦)
Range of hkl
0.32 × 0.15 × 0.10
24.7
0 ≤ h ≤ 7; 0 ≤ k ≤ 14; 0 ≤ l ≤ 24
Number of measured reflections
Number of unique reflections (R int)
Number of observed reflections
[I ≥ 2σ(I)]
10303
1656 (0.067)
1532
NCI-H522
18.0
Colon cancer
COLO 205
HCT-116
HCT-15
HT29
KM12
Number of parameters
R (all data)
283
14.4
4.9
10.7
17.6
16.0
3.2
53.2
45.7
52.0
61.6
69.6
53.1
0.0345 (0.0374)
0.086
1.075
ωR
S
Extinction coefficient
(ꢀ/σ)
0.083
0.000(1)
0.153, −0.153
SW-620
−3
˚
ꢀρ(max, min) (eA
)
CNS cancer
SF-268
SF-295
SF-539
SNB-19
12.1
13.0
14.2
12.5
64.7
68.2
59.9
86.0
3.3. Activity
Selected number of synthesized compounds were
screened in vitro at the National Cancer Institute (NCI,
Bethesda) in one-dose prescreen assay against three can-
cer cell lines—NCI-H460 (lung), MCF7 (breast) and SF-
268 (CNS) [33]. Even on a high concentration (10−4 M),
compounds 2, 3, 10, 16, 17 and 18 proved to be non-
cytotoxic, inducing the growth of the examinated cells in the
range 34–116%. Thus, assaying the p-hydroquinonediacetate
3 resulted in the population growth by 34, 54 and 37%
(NCI-H460, MCF7 and SF-268, respectively). The action
of epoxyquinolacetate 10 on tested cells induced percentage
growth (PG) of 82, 53 and 58% (lung, breast and CNS cancer
cells, respectively), while assaying the quinolacetate 2 and
polyhydroxylated compounds 16, 17 and 18 led to PG of ca.
100 for all three tested cell lines. At applied concentrations,
only the compound 11 showed significant antiproliferative
activity, and appeared to be effective in lung and CNS can-
cer cell lines (PG = 6 and 13, respectively) and was cytotoxic
for selected breast cancer cell line (PG = −58); consequently,
it was processed in 60 cell lines assay. Results of extensive
anticancer activity testing are presented in the Table 2.
From the data in Table 2, it appears that the tested
compound is inactive towards leukemia cell lines, while
against most others it exhibits moderate activity. Strong
anticancer activity was recorded against lung NCI-H522,
ovarian OVCAR-3 and melanoma SK-MEL-5 tumor cells
(LC50 = 18, 22 and 15 ␮M, respectively). The average sensi-
Prostate cancer
PC-3
DU-145
3.2
15.7
43.2
57.7
Melanoma
MALME-3M
M14
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-257
UACC-62
13.8
12.5
15.1
9.1
<0.01
25.6
12.3
59.1
54.3
57.4
46.5
14.9
>100
52.6
Ovarian cancer
IGROV1
4.7
2.4
3.4
15.1
14.3
20.7
44.2
22.0
40.1
61.1
53.8
OVCAR-3
OVCAR-4
OVCAR-5
OVCAR-8
SK-OV-3
>100
Renal cancer
786-0
ACHN
CAKI-1
SN12C
TK-10
2.5
11.8
6.1
14.1
13.9
9.6
34.1
49.9
44.9
55.8
55.2
47.6
UO-31
Breast cancer
MCF7
NCI/ADR-RES
MDA-MB-231/ATCC
14.2
27.6
14.3
68.6
>100
56.1

D. Milic´ et al. / Steroids 70 (2005) 922–932
931
Fig. 2. Dienone-phenol rearrangement of 4-substituted p-quinol acetate 4.
Table 2 (Continued)
4. Discussion
Panel/cell line
Activity (␮M)
The obtained results reveal that TMSOTf-catalyzed acety-
lation is accompanied with substrate aromatization, indi-
cating that reaction media is sufficiently acidic to promote
dienone-phenol rearrangement. In the presence of a simple p-
quinol subunit (compound 1) this process leads to the “para”
product (i.e., p-hydroquinone), while with 4-substituted sub-
strate 4 dienone-phenol rearrangement affords the “meta”
product. After acid-promoted enolyzation of the 3-keto group
and the acetate migration, followed by aromatization and
acetylation of the remaining hydroxyl, bromoquinol 4 was
transformed into the corresponding aromatic bromodiacetate
5, as proposed in Fig. 2.
Performing the same reaction with epoxy-derived sub-
strates 6, 9, and 12, beside the simple acetylation, the opening
of the oxirane ring was observed (Scheme 1). It is reasonable
to propose that product 8 (obtained from 6) was formed after
the elimination of the C(10) acetate3, acid catalyzed (TfOH)
epoxy-ring opening, and subsequent migration of the double
bond (Fig. 3), followed by acetylation of phenolic hydroxyls.
It can be suggested that the other A-ring aromatized
estrone derivatives 11 and 14 (Scheme 1) were formed in
analogy with compound 8. It is important to note that using
our new procedure, beside aromatization, C ring is addition-
ally functionalized at C(9) by introduction of ꢀ9,11 double
bond, giving further opportunities for structure transforma-
tions.
a
b
GI50
LC50
>100
HS 578T
MDA-MB-435
MDA-N
BT-549
T-47D
19.6
4.4
14.2
15.6
9.5
49.7
78.8
60.6
>100
a
The 50% growth-inhibitory concentration, the conc. of the compound
causing 50% inhibition of net cell growth.
b
The50%lethalconcentration, theconcentrationofthecompoundleading
to 50% net cell death.
tivityofalltestedcelllines(meantoxicitygivenasmeangraph
midpoint value, MG MDI) of 8.31 and 61.6 ␮M (derived
from GI50 and LC50 values, respectively) was reached. Com-
paring MG MDI data with the ones from Table 2, it can be
seen that, at GI50 level, aromatic diacetate 11 exhibits higher
activity against most of leukemia and half of tested ovar-
ian cell lines. Strong negative deviation from MG MDI, i.e.,
higher activity, towards lung NCI-H522, ovarian OVCAR-3
and melanoma SK-MEL-5 cell lines observed at GI50 level,
was also maintained on LC50 level. The former results indi-
cate strong, as well as selective anticancer activity of com-
pound 11. Additionaly, the antiproliferative activity of two
bromo A-ring aromatized structures 14 and 5 was also tested
(Table 3). Both compounds were screened against human
malignant HeLa, Fem-x and K562 cells, and it was found
that compound 14 exhibits strong antiproliferative activity
towards leukemia and melanoma cells (IC50 ≈ 10 ␮M), while
the activity of compound 5 was moderate towards all three
tested types of cancer.
After a series of the experiments, it could be concluded
that: (a) during acetylation with Ac2O/TMSOTf system, lib-
erated TfOH induces aromatization by the dienone-phenol
rearrangement or by the acetoxy-group migration; (b) replac-
ing one double bond in the cross-conjugated system with the
epoxy-ring, as in 6, 9 and 12, the aromatization takes place
with oxirane ring-opening; (c) aromatization of epoxyquinols
leads also to the introduction of the double bond into position
9(11).
Table 3
Antiproliferative activity of compounds 14 and 5 after 48 h of continuous
action
In vitro antiproliferative assays showed that o-
hydr‘oquinone diacetates 11 and 14, as well as aromatic
bromodiacetate 5, exhibit significant activity, and they
appear to be active against different cancer cell lines, while
p-hydroquinone diacetate 3 was inactive. It is important
to note that structures 11 and 14 represent a protected
a
Compound
IC50 (␮M)
HeLab
Fem-xc
K562d
14
5
40.70
74.99
13.00
53.56
12.46
44.27
a
The 50% inhibitory concentration, the concentration of a drug required
for inhibiting cell survival by 50%.
b
Human cervix carcinoma.
Human melanoma.
c
3
An additional, qualitative experiment showed that rearranged diacetate
d
Human myelogenous leukemia.
is formed from the acetate 7, as well.

932
D. Milic´ et al. / Steroids 70 (2005) 922–932
Fig. 3. Aromatization of epoxyquinols.
form of catehol estrogen-3,4-quinones wich usually exhibit
cancerogenic action [3].
[13] Johnson CR, Miller MW. Enzymic resolution of a C2 symmetric diol
derived from p-benzoquinone: synthesis of (+)- and (−)-bromoxone.
J Org Chem 1995;60:6674–5.
[14] Wipf P, Kim Y. Synthesis of the antitumor antibiotic LL-
C10037.alpha. J Org Chem 1994;59:3518–9.
[15] Milic DR, Kapor A, Markov B, Ribar B, Strumpel M, Juranic Z,
et al. X-Ray crystal structure of 10␤-hydroxy-4␤,5␤-epoxyestr-1-
en-3,17-dione and antitumor activity of its congeners. Molecules
1999;4:338–52.
Acknowledgment
This research was supported by Serbian Ministry of Sci-
ence and Environmental Protection, grants no 101579 and
101586.
[16] http://dtp.nci.nih.gov./branches/btb/pdf/In Vitro Cancer Method
Human Cell Lines.pdf.
[17] Mosmann T. Rapid colorimetric assay for cellular growth and sur-
vival: application to proliferation and cytotoxicity assay. J Immunol
Methods 1983;65:55–63.
References
[18] Ohno M, Abe T. Rapid colorimetric assay for the quantification of
leukemia inhibitory factor (LIF) and interleukin-6 (IL-6). J Immunol
Methods 1991;145:199–203.
[1] Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B,
et al. Randomized trial of estrogen plus progestin for secondary
prevention of coronary heart disease in postmenopausal women. J
Am Med Assoc 1998;280:605–13.
[2] Sato M, Grese TA, Dodge JA, Bryant HU, Turner CH. Emerging
therapies for the prevention or treatment of postmenopausal osteo-
porosis. J Med Chem 1999;42:1–24.
[3] Cavalieri EL, Stack DE, Devanesan PD, Todorovic R, Dwivedy I,
Higginbotham S, et al. Molecular origin of cancer: catechol estrogen-
3,4-quinones as endogenous tumor inhibitors. Proc Natl Acad Sci
USA 1997;94:10937–42.
[19] Sheldrick GM. SHELXS-97 and SHELXL-97. Program for crystal
structure refinement, University of Go¨ttingen, Germany, 1997.
ˇ
[20] Milic´ D, Kop T, Juranic´ Z, Gasˇic´ MJ, Solaja B. Synthesis and
antiproliferative activity of epoxy and bromo compounds derived
from estrone. Bioorg Med Chem Lett 2001;1116:2197–200.
[21] Adam W, Lupon P. Quinol epoxides from p-cresol and estrone by
photooxigenation and titanium(IV)- or vanadium(V)-catalyzed oxy-
gen transfer. Chem Ber 1988;121:21–5.
[22] Kurihara M, Hakamata W. Convenient preparation of cyclic acetals,
using diols, TMS-source, and a catalytic amount of TMSOTf. J Org
Chem 2003;68:3413–5.
[23] Beignet J, Cox LR. Using a temporary silicon connection in stere-
oselective allylation with allylsilanes: application to the synthesis of
stereodefined 1,2,4-triols. Org Lett 2003;5:4231–4.
[24] Yamanaka M, Nishida A, Nakagawa M. Imino-ene reaction catalyzed
by ytterbium (III) triflate and TMSCl or TMSOTf. J Org Chem
2003;68:3112–20.
[25] Procopiou P, Baugh SPD, Flack SS, Inglis GGA. An extremely
powerful acylation reaction of alcohols with acid anhydrides cat-
alyzed by trimethylsilyl trifluoromethanesulfonate. J Org Chem
1998;63:2342–7.
[4] Cushman M, Mohanakrisnan AK, Hollingshead M, Hamel E. The
effect of exchanging various substituents at the 2-position of 2-
methoxyestradiol on cytotoxicity in human cancer cell cultures and
inhibition of tubulin polymerisation. J Med Chem 2002;45:4748–54.
[5] Liu X, Zhang F, Liu H, Burdette JE, Li Y, Overk CR, et al. Effect of
halogenated substituents on the metabolism and estrogenic effects on
the equine estrogen, equilenin. Chem Res Toxicol 2003;16:741–9.
[6] Zhang Z, Chen Q, Li Y, Doss GA, Dean BJ, Ngui JS, et al. In
vitro bioactivation of dihydrobenzoxathiin selective estrogen receptor
modulators by cytochrome P450 3A4 in human liver microsomes:
formation of reactive iminium and quinone type metabolites. Chem
Res Toxicol 2005;18:675–85.
[7] Prokai L, Prokai-Tatrai K, Perjesi P, Zharikova AD, Simpkins JW.
Quinol-based metabolic cycle for estrogens in rat liver microsomes.
Drug metabolism and disposition 2003;31:701–4.
[26] Gold AM, Schwenk E. Synthesis and reactions of steroidal quinols.
J Am Chem Soc 1958;80:5683–7.
[27] Legler G. Glycoside hydrolases: mechanistic information from stud-
ies with reversible and irreversible inhibitors. Adv Carbohydr Chem
Biochem 1990;48:319–84.
[28] Yoshizaki H, Ba¨ckvall J. Efficient synthesis of ( )-, (+)-, and
(−)-conduritol C via palladium(II)-catalyzed 1,4-diacetoxylation in
combination with enzymatic hydrolysis. J Org Chem 1998;63:9339–
41.
[8] Prokai L, Prokai-Tatrai K, Perjesi P, Zharikova AD, Evelyn J, Perez
EJ, et al. Quinol-based cyclic antioxidant mechanism in estrogen
neuroprotection. Proc Natl Acad Sci USA 2003;100:11741–6.
[9] Milic DR, Gasic MJ, Muster W, Csanadi J, Solaja BA. The synthesis
and biological evaluation of A-ring substituted steroidal p-quinones.
Tetrahedron 1997;53:14073–84.
[10] Solaja BA, Milic DR, Gasic MJ.
A novel m-CPBA oxida-
[29] Donohoe TJ, Blades K, Helliwell M, Waring MJ. The synthesis of
(+)-pericosine B. Tetrahedron Lett 1998;39:8755–8.
tion: p-quinols and epoxyquinols from phenols. Tetrahedron Lett
1996;37:3765–8.
[11] Wada H, Fujita H, Murakami T, Saiki Y, Chen CM. Chemical and
chemotaxonomical studies of ferns. LXXIII. New flavonoids with
modified B-ring from the genus Pseudophegopteris (Thelypterida-
cae). Chem Pharm Bull 1987;35:4757–62.
[12] Wells G, Berry JM, Bradshaw TD, Burger AM, Seaton A, Wang B, et
al. 4-Substituted 4-hydroxycyclohexa-2,5-dien-1-ones with selective
activities against colon and renal cancer cell lines. J Med Chem
2003;46:532–41.
ˇ
[30] Markovic´ Z, Solaja B, Milic´ D, Juranic´ I, Gasˇic´ MJ. Molecular
orbital study of the oxidation of steroidal phenols into quinols and
epoxyquinols. J Serb Chem Soc 2000;65:491–6.
[31] Duax WL, Weeks CM, Rohrer DC. Crystal structures of steroids.
Top Stereochem 1976;9:271–383.
[32] Spek AL. PLATON Molecular geometry program. The Netherlands:
University of Utrecht; 1998.
[33] http://dtp.nci.nih.gov/branches/btb/ivclsp.html.