Structural analysis and biomedical potential of novel salicyloyloxy estrane derivatives synthesized by microwave irradiation

Article abstract of DOI:10.1007/s11224-015-0678-5

New estrane salicyloyloxy or D-homo derivatives were synthesized under microwave (MW) or conventional heating from estrane precursors and methyl salicylate. The MW technique provides advantages regarding product yield and reaction time, and represents a more environmentally friendly approach than conventional heating. Considering the biomedical potential of estrane compounds, we evaluated the antioxidant activity and cytotoxicity of synthesized estrane derivatives in a series of in vitro tests, as well as their 3β-hydroxysteroid dehydrogenase/Δ⁵ → Δ⁴ isomerase (3βHSD) and 17β-hydroxysteroid dehydrogenase types 1, 2 and 3 (17βHSD1, 17βHSD2 and 17βHSD3) inhibition potentials. In DPPH tests, 3-methoxyestra-1,3,5(10)-trien-17β-yl salicylate displayed antioxidant potential, while all compounds exhibited OH radical neutralization activity. 3-Oxoestr-4-en-17β-yl salicylate showed strong cytotoxicity against MDA-MB-231 breast cancer cells, while 17-oxoestra-1,3,5(10)-trien-3-yl salicylate, estra-1,3,5(10)-triene-3,17β-diyl 3-benzoate 17-salicylate and 3-benzyloxy-17-salicyloyloxy-16,17-secoestra-1,3,5(10)-triene-16-nitrile showed the strongest inhibition of PC-3 prostate cancer cell growth. 3-Hydroxyestra-1,3,5(10)-trien-17β-yl salicylate was the best inhibitor of 17βHSD2, suggesting potential use in treating pathological conditions associated with estrogen depletion. For 3-methoxyestra-1,3,5(10)-trien-17β-yl salicylate and 3-oxoestr-4-en-17β-yl salicylate, X-ray crystal structure analysis and molecular energy optimization were performed to define their conformations and energy minima. Very good overlap in the region of the steroidal nucleus was observed for the molecular structures of each analyzed molecule in the crystalline state and after energy optimization, while conformer analysis indicates conformational flexibility in the form of rotation around the C17···O2 bond. Structural geometry analysis for these compounds shows that the region of ring A in steroids, and especially the C3 atom functional group, is important structural features concerning antiproliferative activity against MDA-MB-231 cells.

Full text of DOI:10.1007/s11224-015-0678-5

Struct Chem (2016) 27:947–960
DOI 10.1007/s11224-015-0678-5
ORIGINAL RESEARCH
Structural analysis and biomedical potential of novel salicyloyloxy
estrane derivatives synthesized by microwave irradiation
1
Olivera R. Klisuric Mihaly Szecsi Evgenija A. Djurendic Szabo Nikoletta
2
3
2
•
•
•
•
´
´
´
´
´
Bianka Edina Herman2 Suzana S. Jovanovic Santa Sanja V. Dojcinovic Vujaskovic
3
3
•
ˇ
•
•
´
Andrea R. Nikolic Ksenija J. Pavlovic Jovana J. Ajdukovic Aleksandar M. Okljesa
ˇ
´
ˇ
´
3
3
3
3
•
•
•
•
´
´
´
ˇ
Edward T. Petri4 Vesna V. Kojic Marija N. Sakac Katarina M. Penov Gasi
5
3
´
ˇ3
ˇ
•
•
•
Received: 26 June 2015 / Accepted: 7 September 2015 / Published online: 19 September 2015
Ó Springer Science+Business Media New York 2015
Abstract New estrane salicyloyloxy or D-homo deriva-
tives were synthesized under microwave (MW) or con-
ventional heating from estrane precursors and methyl
salicylate. The MW technique provides advantages
regarding product yield and reaction time, and represents a
more environmentally friendly approach than conventional
heating. Considering the biomedical potential of estrane
compounds, we evaluated the antioxidant activity and
cytotoxicity of synthesized estrane derivatives in a series of
in vitro tests, as well as their 3b-hydroxysteroid dehydro-
genase/D5 ? D4 isomerase (3bHSD) and 17b-hydroxys-
teroid dehydrogenase types 1, 2 and 3 (17bHSD1,
17bHSD2 and 17bHSD3) inhibition potentials. In DPPH
tests, 3-methoxyestra-1,3,5(10)-trien-17b-yl salicylate dis-
played antioxidant potential, while all compounds exhib-
ited OH radical neutralization activity. 3-Oxoestr-4-en-
17b-yl salicylate showed strong cytotoxicity against MDA-
MB-231 breast cancer cells, while 17-oxoestra-1,3,5(10)-
trien-3-yl salicylate, estra-1,3,5(10)-triene-3,17b-diyl
3-benzoate 17-salicylate and 3-benzyloxy-17-salicyloy-
loxy-16,17-secoestra-1,3,5(10)-triene-16-nitrile showed the
strongest inhibition of PC-3 prostate cancer cell growth.
3-Hydroxyestra-1,3,5(10)-trien-17b-yl salicylate was the
best inhibitor of 17bHSD2, suggesting potential use in
treating pathological conditions associated with estrogen
depletion. For 3-methoxyestra-1,3,5(10)-trien-17b-yl sali-
cylate and 3-oxoestr-4-en-17b-yl salicylate, X-ray crystal
structure analysis and molecular energy optimization were
performed to define their conformations and energy min-
ima. Very good overlap in the region of the steroidal
nucleus was observed for the molecular structures of each
analyzed molecule in the crystalline state and after energy
optimization, while conformer analysis indicates confor-
mational flexibility in the form of rotation around the
C17ÁÁÁO2 bond. Structural geometry analysis for these
compounds shows that the region of ring A in steroids, and
especially the C3 atom functional group, is important
structural features concerning antiproliferative activity
against MDA-MB-231 cells.
´
& Olivera R. Klisuric
olivia@uns.ac.rs
Keywords Estrane salicylic acid derivatives Á
Microwave-assisted (MW) synthesis Á Biomedical
potential Á X-ray structural analysis Á Energy optimization
1
Department of Physics, Faculty of Sciences, University of
´
Novi Sad, Trg Dositeja Obradovica 4, 21000 Novi Sad,
Serbia
2
3
´
1st Department of Medicine, University of Szeged, Koranyi
fasor 8-10, 6720 Szeged, Hungary
Introduction
Department of Chemistry, Biochemistry and Environmental
Protection, Faculty of Sciences, University of Novi Sad, Trg
´
Dositeja Obradovica 3, 21000 Novi Sad, Serbia
Hormone-stimulated cell proliferation, which increases the
number of cell divisions and the opportunity for genetic
mutations, is associated with carcinogenesis in hormone-
related cancers. In target tissues, steroid hormones are
interconverted between their potent, high-affinity forms and
inactive, low-affinity forms for their respective receptors.
Enzymes partly responsible for these interconversions are
4
5
Department of Biology and Ecology, Faculty of Sciences,
´
University of Novi Sad, Trg Dositeja Obradovica 4,
21000 Novi Sad, Serbia
Oncology Institute of Vojvodina, Put Dr Goldmana 4,
21204 Sremska Kamenica, Serbia
123

948
Struct Chem (2016) 27:947–960
the hydroxysteroid dehydrogenases (HSDs). 3b-Hydroxys-
teroid dehydrogenase/D5 ? D4 isomerase (3bHSD) is pre-
sent in many tissues [1]. This enzyme catalyzes both
conversion of the hydroxyl group on carbon 3 (C3) to a keto
group and double bond isomerization during the conversion
of pregnenolone to progesterone or other substrates. Since
3b-HSD is involved in the biosynthesis of nearly all steroid
hormones, inhibition of 3b-HSD could represent a potential
target for the treatment of hormone dependent cancers. 17b-
Hydroxysteroid dehydrogenases catalyze either oxidation of
hydroxyl groups or reduction of keto groups at steroid
position C17 [2, 3]. 17b-Hydroxysteroid dehydrogenase
type 1 (17bHSD1) activates the less active estrone to 17b-
estradiol, a potent ligand for estrogen receptors; thus, inhi-
bitors of this enzyme are highly interesting potential thera-
peutic agents for the control of estrogen-dependent diseases
such as endometriosis, as well as breast and ovarian cancers.
17bHSD1 is inhibited by some estradiol derivatives [4–7].
17bHSD2 catalyzes the reverse process (17b-estradiol to
estrone); thus, inhibitors of this isoform could enable better
regulation of the levels of active estrogens. Therapeutically,
17bHSD2 inhibitors could be used in the treatment of
osteoporosis. Some estrane derivatives proved to be
17bHSD 2 inhibitors. [8, 9]. 17bHSD3, synthesized in testis,
activates androst-4-ene-3,17-dione into testosterone. Thus,
17bHSD 3 inhibitors could be therapeutics for the treatment
of prostate cancer or similar diseases. Some steroidal com-
pounds expressed good inhibitory potential for this isoform
[10].
we examined the effectiveness of newly synthesized sali-
cyloyloxy estrane derivatives at free radical scavenging
and inhibition of tumor cell proliferation, as well as their
inhibitory potential against 3bHSD, 17bHSD1, 17bHSD2
and 17bHSD3.
Herein we report X-ray crystal structures of 3-methox-
yestra-1,3,5(10)-trien-17b-yl salicylate (11) and 3-oxoestr-
4-en-17b-yl salicylate (14), as well as their molecular
energy optimization. Because of their interesting biological
activities, we also compared the structural features of new
crystal structures (11 and 14) with several previously
reported structures [28].
Experimental
Chemical synthesis
General
Infrared spectra (wave numbers in cm-1) were recorded on
a Nexus 670 FT-IR spectrometer. Nuclear magnetic reso-
nance (NMR) spectra were recorded on a Bruker AC 250
apparatus operating at 250 MHz (proton) and 62.9 MHz
(carbon), using standard Bruker software, with tetram-
ethylsilane as the internal standard. Chemical shifts are
given in ppm (d-scale); coupling constants (J) are given in
Hz. High-resolution mass spectra (TOF) were recorded on
a 6210 Time-of-Flight LC/MS Agilent Technologies
(ESI?) instrument. Absorbances of reaction mixtures in
free radical scavenging tests were recorded on a CECIL
CE2021 spectrophotometer. The microwave reactor was a
monomode system (Microwave Synthesis System—Dis-
cover Bench Mate from CEM) with focused waves. Melt-
ing points were determined using a Bu¨chi SMP 20
apparatus and are uncorrected. Organic solutions were
dried over Na2SO4 and evaporated on a rotary evaporator
under reduced pressure. Column chromatography was
performed on Merck grade 60 silica gel (0.063–0.2 mm).
Estrane compounds decrease tumor cell proliferation
in vitro by different mechanisms, enabling alternative
approaches to cancer therapy [11–17]. Some estrane
derivatives affect tumor cells via estrogen receptors, while
others induce cell death or cell cycle arrest, or prevent
cancer progression by scavenging free radicals.
Estrogen hormones have well-known effects on female
reproductive tissues, but also display beneficial effects in
many other tissues, largely due to their antioxidant poten-
tial. In addition, among the family of steroidal molecules,
estrogens have shown the most potential for preventing
neuronal cell death caused by increased oxidative damage
[18–20].
General procedure for the preparation of compounds 7–14
On the other hand, some phenolic substances act as
powerful antioxidants and/or cytotoxic agents. Salicylic
acid derivatives are phenolic compounds exhibiting such
activities [21–28].
Microwave-assisted (MW) synthesis A mixture consisting
of methyl salicylate (18 mmol), the corresponding steroidal
compound (1 mmol) and sodium (3 mmol) was heated to
110 °C. When reaction with sodium was completed
(5–10 min), toluene (3 ml) was added and then the mixture
was irradiated for 30 min at 130–200 °C, using a 200-W
microwave source. After cooling the reaction mixture to
room temperature, water (100 mL) and HCl (1:1, to pH 7)
were added and crude product was extracted with
dichloromethane (3 9 50 mL). The organic phase was
Combining the structural characteristics of both estrane
compounds and salicylic acid, we recently reported some
new biomedically potent steroidal derivatives [29]. In the
present study, we present the synthesis, structural analysis
and biological activity of some newly synthesized salicy-
loyloxy estrane derivatives. Since the relationship between
oxidative stress and carcinogenesis is well established [30],
123

Struct Chem (2016) 27:947–960
949
128.46; 129.57; 129.76; 130.07; 133.44; 135.51; 137.67;
138.27; 148.64; 161.57(C-2); 165.38 (C=O from benzoate);
170.08 (C=O from salicyloyl group). HRMS (TOF) m/z:
C32H32O5 [M ? H]? calculated: 497.23225, found:
497.23250.
dried and evaporated, resulting in an oily product. Chro-
matographic separation of crude product on a silica gel
column (toluene–EtOAc, 9:1 for 7; petroleum ether–ace-
tone, 20:1 for 8 and 9; toluene–EtOAc, 27:1 for 10; pet-
roleum ether–acetone, 10:1 for 11; toluene–EtOAc, 24:1
for 12 and 13; toluene–EtOAc, 24:1 for 14) gave the pure
products 7–9 and 11–14.
3-Methoxyestra-1,3,5(10)-trien-17b-yl salicylate (11) White
crystals, mp 190–194 °C from n-hexane–dichloromethane.
IR (KBr): 3139, 2934, 1669, 1613, 1583, 1500, 1325, 1252,
1159, 1142, 1094, 1034, 995, 759, 737, 704. 1H NMR
(CDCl3): 0.98 (s, 3H, H-18); 3.78 (s, 3H, OCH3); 4.95 (t, 1H,
H-17, J = 7.62 Hz); 6.64–7.88 (group of signals, 7H, H–
Ar); 10.91 (s, 1H, OH phenolic group from salicyloyl resi-
due). 13C NMR (CDCl3): 12.35 (C-18); 23.37; 26.17; 27.22;
27.70; 29.76; 36.94; 38.55; 43.35; 43.75; 49.69; 55.18;
83.74; 111.47; 112.83; 113.78; 117.55; 119.06; 126.35;
129.82; 132.31; 135.53; 137.84; 157.46; 161.61; 170.15
(C=O). HRMS (TOF) m/z: C26H30O4 [M–H]- calculated:
405.20713, found: 405.20783.
Conventionally heated synthesis A mixture consisting of
methyl salicylate (18 mmol), the corresponding steroidal
compound (1 mmol), sodium (3 mmol) and toluene (3 mL)
was refluxed for 5.5–24 h, depending on the substrate.
After reaction completion, the reaction mixture was cooled
to room temperature, water (100 mL) and HCl (1:1, to pH
7) were added, and crude product was extracted with
dichloromethane (3 9 50 mL). The extract was dried and
solvent was evaporated to dryness. Pure compounds 7, 8
and 10–14 were obtained from crude mixture after chro-
matography on a silica gel column (toluene–EtOAc, 18:1
for 7; n-hexane–acetone, 19:1 for 8; n-hexane–acetone,
45:1 for 10; n-hexane–acetone, 15:1 for 11; petroleum
ether–acetone, 9:1 for 12 and 13; toluene–EtOAc, 12:1 for
14).
3-Benzyloxy-17-salicyloyloxy-16,17-secoestra-1,3,5(10)-tri-
ene-16-nitrile (12) Yellow oil IR (film): 3187, 3033, 2929,
2243, 1676, 1613, 1500, 1299, 1249, 1158, 1089, 758, 735,
1
699. H NMR (CDCl3): 1.09 (s, 3H, H-18); 4.06 (d, 1H,
H-17a, J = 11.50 Hz), 4.33 (d, 1H, H-17b, J = 11.52 Hz);
5.03 (s, 2H, CH2 from Bn); 6.73–7.85 (group of signals, H–
Ar); 10.71 (s, 1H, OH phenolic group from salicyloyl resi-
due). 13C NMR (CDCl3): 15.92 (C-15), 16.24 (C-18); 25.94;
27.02; 29.93; 36.25; 38.03; 39.37; 42.18; 42.87; 69.90 (CH2
from Bn); 71.77 (C-17); 112.08; 112.72; 114.38; 117.81;
119.10 (C:N); 119.38; 126.41; 127.43; 127.88; 128.54;
129.50; 131.57; 136.05; 137.12; 137.42; 156.95 (C-3);
161.79; 169.86 (C=O from salicyloyl group); HRMS (TOF)
m/z: C32H33O4N [M ? H]? calculated: 496.24824, found
496.24763.
3-Hydroxyestra-1,3,5(10)-trien-17b-yl salicylate (7) Light
yellow crystals, mp 81–84 °C from acetone–hexane [[29],
mp 81–84 °C].
17-Oxoestra-1,3,5(10)-trien-3-yl salicylate (8) White
crystals, mp 230 °C from n-hexane–acetone. IR (KBr):
3205, 2933, 2871, 1737, 1686, 1615, 1583, 1301, 1247,
1208, 1157, 1068, 759, 699. 1H NMR (CDCl3): 0.92 (s, 3H,
H-18); 6.93-8.07 (group of signals, 7H, H–Ar); 10.55 (s,
1H, OH phenolic group from salicyloyl residue). 13C NMR
(CDCl3): 13.69 (C-18); 21.46; 25.62; 26.17; 29.30; 31.40;
35.72; 37.81; 44.01; 47.79; 50.24; 111.72; 117.67; 118.65;
119.25; 121.47; 126.51; 130.20; 136.32; 137.89; 138.21;
147.81; 162.02; 169.09 (C=O from salicyloyl group);
220.60 (C-17). HRMS (TOF) m/z: C25H26O4 [M ? H]?
calculated: 391.19039, found: 391.18963.
3-Benzyloxy-17-oxa-D-homoestra-1,3,5(10)-trien-16-one
(13) White crystals, mp 167–169 °C from n-hexane–
dichloromethane [32, mp 162 °C].
3-Oxoestr-4-en-17b-yl salicylate (14) White crystals, mp
213–214 °C from n-hexane–dichloromethane. IR (KBr):
3117, 2945, 2860, 1663, 1614, 1583, 1485, 1302, 1252,
Estrone 3-methyl ether (9) White crystals, mp 170 °C
from hexane–dichloromethane [[31], mp 169 °C].
1
1214, 1157, 1093, 795. H NMR (CDCl3): 0.99 (s, 3H,
H-18); 4.85 (t, 1H, H-17, J = 8.27 Hz), 5.83 (s, 1H, H-4);
6.84–7.85 (group of signals, 4H, H–Ar); 10.87 (s, 1H, OH
phenolic group from salicyloyl residue). 13C NMR
(CDCl3): 12.22 (C-18); 23.31; 25.82; 26.46; 27.45; 30.43;
35.26; 36.40; 36.52; 39.97; 42.32; 42.97; 49.18; 49.29;
83.41 (C-17); 112.61 (C-10); 117.46 (C-30); 118.98 (C-60);
124.57 (C-4); 129.67; 135.49 (C-40); 161.49 (C-20); 166.29
(C-5); 169.97 (C=O from salicyloyl group); 199.85 (C-3).
HRMS (TOF) m/z: C25H30O4 [M ? H]? calculated:
395.22169, found: 395.22172.
Estra-1,3,5(10)-triene-3,17b-diyl 3-benzoate 17-salicylate
(10) White crystals, mp 190–194 °C from n-hexane–
dichloromethane. IR (KBr): 3148, 2930, 1730, 1669, 1614,
1
1485, 1303, 1248, 1213, 1158, 1063, 763, 704. H NMR
(CDCl3): 1.03 (s, 3H, H-18); 4.98 (t, 1H, H-17,
J = 8.26 Hz); 6.89–8.25 (group of signals, 12H, H–Ar);
10.96 (s, 1H, OH phenolic group from salicyloyl residue).
13C NMR (CDCl3): 12.28 (C-18); 23.32; 25.96; 26.97;
27.64; 29.46; 36.85; 38.08; 43.24; 43.90; 49.66; 83.62 (C-
17); 112.74; 117.51; 118.69; 119.03; 121.60; 126.44;
123

950
Struct Chem (2016) 27:947–960
Biological tests
calculated using Eq. (1). Three replicates were recorded for
each sample. Estradiol, methyl salicylate, BHT and BHA
were used as reference compounds. IC50 values (the con-
centration at which 50 % of HO is neutralized) were
determined by linear regression analysis from obtained
RSC values.
DPPH radical scavenging assay
The free radical scavenging capacity (RSC) of selected
salicyloyloxy estrane derivatives (7, 11 and 14) and ref-
erence compounds (commercial synthetic anti-oxidants:
3,5-di-tert-butyl-4-hydroxytoluene/BHT Aldrich, Ger-
many, and 3-tert-butyl-4-hydroxyanisole/BHA Fluka,
Germany), estradiol, compound 5 and methyl salicylate
were evaluated by measuring their ability to neutralize 2,2-
diphenyl-1-picrylhydrazyl (DPPH) and hydroxyl (HO)
radicals.
Cytotoxicity
The cytotoxicity of synthesised compounds 7 and 9–14 was
evaluated by the colorimetric sulforhodamine B (SRB)
assay [29, 33]. The chemotherapy drug doxorubicin (DOX)
was used as a control. Three human cancerous cell lines
and one normal cell line were used in this study: estrogen
receptor positive human breast adenocarcinoma (ER?,
MCF-7), estrogen receptor negative human breast adeno-
carcinoma (ER-, MDA-MB-231), human prostate cancer
(PC-3) and normal fetal lung fibroblasts (MRC-5). Cells
were grown in Dulbecco’s modified Eagle’s medium with
4.5 % glucose. Media were supplemented with 10 % of
fetal calf serum and antibiotics (100 IU mL-1 of penicillin
and 100 lg mL-1 of streptomycin; ICN Galenika). All
cells were cultured in flasks (Costar, 25 cm2) at 37 °C in an
atmosphere of 100 % humidity and 5 % CO2 in an incu-
bator. Only viable cells were used in the assay. Viability
was determined by dye exclusion assay with trypan blue.
Cytotoxicity was evaluated by the colourimetric sul-
forhodamine B (SRB) assay [33]. Briefly, a single-cell
suspension (5 9 103 cells) was plated into 96-well micro-
titre plates (Costar, flat bottom). The plates were pre-in-
cubated at 37 °C in a 5 % CO2 incubator for 24 h. The
substances tested (final concentrations ranging from 10-8
to 10-4 M) were transferred into all wells except for con-
trols. After the incubation period (48 h/37 °C/5 % CO2),
the cytotoxicity assay was carried out as follows: 50 lL of
80 % trichloroacetic acid (TCA) was added to all wells;
1 h later, plates were washed with distilled water, and
75 lL of 0.4 % SRB was added to all wells; after 30-min
incubation, the plates were washed with citric acid (1 %)
and dried at room temperature. Finally, 200 lL of 10 mM
Tris (pH 10.5) base was added to all wells. Absorbance was
measured using a microplate reader. Wells without cells,
containing only complete medium, served as blanks.
Cytotoxicity was calculated according to the following
equation:
The DPPH assay was performed as described [29].
Different aliquots (0.1–2.0 mL) of 0.01 M sample solution
(compounds in dichloromethane) were added to DPPH in
methanol (90 lM, 1 mL; Sigma, St. Louis, MO) and
diluted with 95 vol% of methanol to a final volume of
4 mL. The same reaction mixture without test compounds
was used as a control. Absorbance of the reaction mixtures
(Asample) and control (Acontrol) was recorded at 515 nm after
1 h. For each sample, three replicates were recorded. The
percentage of DPPH radical scavenging capacity (DPPH
RSC) was calculated using the following equation:
À
Á
RSC ð%Þ ¼ 100 Â Acontrol À Asample=Acontrol
ð1Þ
IC50 values (the concentration of test compound in the
reaction mixture which causes 50 % of RSC) were deter-
mined by linear regression analysis from RSC values
obtained.
Hydroxyl radical scavenging assay
The hydroxyl radical scavenging capacity (HO RSC) of
selected compounds was evaluated by measuring the
degradation of 2-deoxy-D-ribose (Aldrich, Germany) dur-
ing reaction with OH radicals, generated in situ in Fenton’s
reaction [29]. Different aliquots (0.005–0.5 mL) of a
sample solution in dichloromethane were added to test
tubes (final concentration in the range of 0.01–8 mM, each
containing 0.1 mL of 5 mM H2O2, 0.1 mL of 10 mM
FeSO4, 0.1 mL of 0.05 M 2-deoxy-D-ribose, and 0.067 M
KH2PO4–K2HPO4 buffer of pH 7.4 to a final volume of
3 mL). The same reaction mixture without sample was
used as control. After an incubation period of 1 h at 37 °C,
2 mL of TBA reagent (10.4 mL of 60 vol% HClO4, 3 g of
TBA, and 120 g of trichloroacetic acid (Sigma, St. Louis,
MO, USA)) and 0.2 mL of 0.1 M EDTA (Sigma, St. Louis,
MO, USA) were added to the reaction mixture and the
tubes were heated at 100 °C for 20 min. After cooling,
absorbance of the reaction mixtures and control was
recorded at 532 nm. The percentage of HO RSC was
À
Á
CI ð%Þ ¼ 1 ÀAsample=Acontrol  100
ð2Þ
Three independent experiments were conducted in
quadruplicate for each concentration of test compound. The
IC50 (dose of a compound that inhibits cell growth by
50 %) of test compounds was determined by median effect
analysis.
123

Struct Chem (2016) 27:947–960
951
˚
Determination of 3bHSD, 17bHSD1, 17bHSD2
and 17bHSD3 activities and their inhibition
MoKa radiation (k = 0.7107 A). Data reductions were
performed with the program package CrysAlis RED [36].
Space group determinations were based on analysis of the
Laue class and systematically absent reflections. Structures
were solved by direct methods using SHELXT [37] for 11
and SIR92 [38] for 14. Structures were refined by full-
matrix least squares procedures on F2 using SHELXL-
2014/6 program [37]. For both compounds, non-hydrogen
atoms were refined anisotropically and C–H hydrogen
atoms were included on calculated positions riding on their
Inhibitory effects exerted on steroidogenic enzymes by the
newly synthesized compounds 7, 10–12 and 14 were
investigated with in vitro radio-labeled substrate incuba-
tions (Table 1). Our previously published methods for
3bHSD [28, 34, 35], 17bHSD1 [7], 17bHSD2 [27, 28] and
17bHSD3 [28] were used with minor modifications. During
procedures, tissue preparations serving as enzyme sources
were incubated with 1 lM [14C]-labeled substrate steroids
in the presence of 0.1 mM coenzymes at 37 °C. Buffer
medium consisted of 0.1 M HEPES (pH 7.3), 1 mM EDTA
and 1 mM dithiothreitol. The appropriate substrate was
added in 10 lL of 25 v/v % propylene glycol in HEPES
buffer solution, whereas test compounds were applied in
10 lL of dimethyl sulfoxide solution. In a final reaction
volume of 200-lL incubation medium, these organic sol-
vents did not reduce enzyme activity substantially. After
incubation, enzymatic reactions were stopped by addition
of EtOAc and cooling. Unlabeled carriers of the substrate
and product steroids were added, and samples were then
extracted. Substrates and products were separated by TLC
on Kieselgel-G (Merck Si 254 F) layers (0.25 mm thick)
with the solvent system dichloromethane-diisopropyl ether-
EtOAc (75:15:10 v/v), and UV spots were used to trace the
separated steroids. Spots were cut out and radioactivity was
measured by means of liquid scintillation counting. Test
compounds were applied at 50 lM concentration, and
controls without test substances were also prepared in
every series. At least two experiments were performed with
each test compound. IC50 values were determined for more
potent inhibitors. In this case, conversion was measured at
five or six different concentrations of the test compound
between 0.1 and 50 lM.
˚
attached atoms with fixed distances of 0.93 A (CH; some
hydrogen atoms from CH groups were identified using
˚
difference electron density maps), 0.97 A (CH2) and
˚
0.96 A (CH3). At the final stage of refinement, H atoms
from the hydroxyl group were identified by difference
electron density maps and isotropically refined.
In the absence of significant anomalous scattering effects,
Friedel pairs for both compounds were merged. The absolute
configuration of 11 and 14 can be assumed without risk based
on the absolute configuration of the starting material, con-
sidering that the synthetic transformations carried out do not
affect the chirality in the naturally occurring steroid frame-
work, although the refined Flack parameters x = 0.3(6) for
11 and x = 0.3(5) for 14 [39] are not definitive.
All calculations were performed using PARST [40] and
PLATON [41], as implemented in the WINGX [42] system
of programs. The crystal data and refinement parameters
are summarized in Table 2.
Results and discussion
Steroidal derivatives 7–14 were synthesized by reaction of
methyl salicylate with compounds 1–6 (Table 3). Reac-
tions were performed by MW irradiation and conventional
heating, in the presence of sodium and toluene as solvent.
Substrates 1, 4 and 6 gave corresponding salicyloyl esters
7, 11 and 14 by transesterification reaction under conven-
tional heating or MW irradiation. However, substrates 2, 3
and 5 showed different reactivity. Namely, by conventional
heating substrate 2 was converted into the corresponding
salicyloyl ester 8, but in MW-assisted reaction conditions,
X-ray crystallographic analysis of compounds 11 and 14
Diffraction data for compounds 11 and 14 were collected at
room temperature on an Oxford Diffraction (Agilent
Technologies) Gemini S diffractometer using the program
package CrysAlis CCD [36] with graphite-monochromated
Table 1 Description of radiosubstrate enzyme incubation methods used for inhibition tests
Enzyme
3bHSD
17bHSD1
17bHSD2
17bHSD3
Enzyme source
Substrate
Rat testicular homogenate
Dehydroepi-androsterone
Androst-4-ene-3,17-dione
NAD
Human placental cytosol
Rat liver microsomes
Rat testicular homogenate
Androst-4-ene-3,17-dione
Testosterone
Estrone
17b-Estradiol
NADH
Testosterone
Product
Androst-4-ene-3,17-dione
Coenzyme
NAD
20
NADPH
Incubation time (min)
20
2.5
20
123

952
Struct Chem (2016) 27:947–960
Table 2 Experimental details: crystallographic data and refinement parameters
11
14
Chemical formula
Mr
C26H30O4
C25H30O4
394.49
406.50
Crystal system, space group
Monoclinic, P21
Orthorhombic, P212121
˚
a, b, c (A)
8.7591(7), 7.3572(6), 16.8520(13)
7.1646(3), 11.5003(5), 25.5462(10)
b (°)
V (A )
90.759(6)
–
3
˚
1085.89(15)
2104.88(15)
Z
2
4
l (mm-1
)
0.08
0.08
Crystal size (mm)
Diffractometer
0.57 9 0.21 9 0.15
0.41 9 0.36 9 0.15
Gemini S; Agilent Technologies
(Oxford Diffraction) diffractometer
Absorption correction
Analytical [50]
Multi-scan
CrysAlis RED [36]
CrysAlis RED [36]
Numeric absorption correction using
a multifaceted crystal model
Empirical absorption correction using
spherical harmonics, implemented in
SCALE3 ABSPACK scaling algorithm
Tmin, Tmax
0.958, 0.989
0.949, 1.000
No. of measured, independent and
observed [I [ 2r(I)] reflections
Rint
5183, 3332, 1915
5894, 3288, 2232
0.018
0.017
R[F2 [ 2r(F2)], wR(F2), S
No. of parameters
No. of restraints
H-atom treatment
0.033, 0.072, 0.85
0.035, 0.069, 0.90
293
1
279
0
H atoms treated by a mixture of independent and constrained refinement
-3
)
˚
Dqmax, Dqmin (e A
0.09, -0.12
0.11, -0.10
in addition to esterification reaction, methylation of the
phenol moiety of compound 2 occurred predominantly,
leading to methylether 9. In the case of substrate 3, con-
ventional methods afforded a mixture of salicyloyl esters 7
and 10, while the MW method provided only ester 7. In
both methods, the C3-benzoyloxy group of compound 3
was converted, partially or completely, into a C3-hydroxyl
group. Lactone 13, previously reported [32], was obtained
in addition to salicyloyl ester 12 from D-seco substrate 5 by
both conventional and MW-assisted methods.
(4 9 30 min) resulted in the synthesis of estrone 3-methyl
ether (9, 17 %) as the main product and 3-salicyloyloxy
derivative 8 as a by-product (2 %). Starting from 3-ben-
zoyloxy-17b-hydroxyestra-1,3,5(10)-triene (3), 17b-sali-
cyloyloxy derivative 10 (17.5 %) and compound 7
(10.8 %) were obtained using conventional heating, while
transesterification of methyl salicylate with compound 3,
performed by MW irradiation for 30 min, afforded a
complex mixture from which only compound 7 could be
isolated (18.4 %). In the reaction of estradiol 3-methyl
ether (4) with methyl salicylate, 3-methoxy 17b-salicy-
loyloxy derivative 11 was obtained. This reaction lasted
5.5 h under conventional heating, while the same reaction
was completed in only 30 min using MW irradiation.
Moreover, the MW irradiation method led to an almost
twofold increase in yield of compound 11(68.5 %).
In order to compare the efficiency of the two reaction
conditions, we compared reaction time and yields of newly
synthesized compounds (Table 3) for both MW-assisted
and conventional syntheses.
During transesterification of methyl salicylate with
estradiol 1 under conventional heating (13 h), the salicy-
loyl ester 7 was obtained at 36 % yield [29]. MW-assisted
transesterification (30 min) afforded the desired compound
7 at higher yield (50 %), while the reaction time was
shortened 26-fold compared to conventional heating.
Conventional heating of estrone (2) with methyl salicylate
gave 3-salicyloyloxy derivative 8 at 32 % yield. MW
irradiation of the mixture of methyl salicylate and estrone
D-Secocyanoalcohol 5 [43] reacted with methyl salicy-
late by conventional heating (13 h), yielding 17-salicy-
loyloxy derivative 12 (12.6 %) and D-homo-lactone 13 as
by-product (3.8 %). In our earlier paper [32], lactone 13
was synthesized in high yield (57 %) by the treatment of
the starting secocyanoalcohol 5 with catalytic amounts of
p-toluenesulfonic acid in benzene. MW irradiation of the
123

Struct Chem (2016) 27:947–960
953
Table 3 The structures of starting and product compounds and reaction conditions for both MW-assisted and conventional synthesis
Substrates
Products
Conventional
heating
MW irradiation
(200W)
[time (h)/yield (%)]
[temp (oC)/time
(min)/yield (%)]
13/36
24/32
200/30/50
170/4x30/2
170/4x30/17
-
5.5/17.5
Only 7
7 (10.8%)
(200/30/18.4)
5.5/37
130/30/68.5
160/30/85
13/12.6
13/3.8
160/30/4
8.5/5.8
200/30/57
reaction mixture of compound 5 and methyl salicylate for
30 min provided compound 12 at almost sevenfold higher
yield (85 %) than conventional heating conditions
(12.6 %), while the yield of compound 13 was practically
the same (4 %). The reaction time was shortened 26-fold.
Transesterification reaction of methyl salicylate with
19-nortestosterone (6) by conventional heating gave 3-ox-
oestra-4-en-17b-yl salicylate (14) at only 5.8 % yield,
while MW irradiation resulted in the synthesis of 14 at
57 % yield. Product yield increased almost tenfold during
123

954
Struct Chem (2016) 27:947–960
of a salicyloyl group at C-17. The presence of C=O from
the salicyloyloxy group in 8, 10, 11, 12 and 14 is confirmed
by the signals at 169.09, 170.08, 170.15, 169.86 and
169.97 ppm, respectively, in the corresponding 13C NMR
spectra.
Table 4 Scavenger activity of the selected steroidal compounds
Compound
DPPH (IC50 mM)
OH (IC50 mM)
5
2.85 0.12
0.55 0.06
0.16 0.03
1.30 0.07
0.497 0.01
0.570 0.02
0.04 0.01
0.012 0.002
5.70 0.20
0.20 0.04
0.95 0.07
0.35 0.04
0.004 0.001
0.011 0.007
1.94 0.06
2.13 0.05
7 [29]
11
The antioxidant activity of selected salicylic acid ster-
oidal derivatives was evaluated in in vitro tests and com-
pared with those of parent molecules (estradiol, compound
5 and methyl salicylate), as well as commercial antioxi-
dants BHT and BHA. In the DPPH assay, the ability of the
tested compounds to act as hydrogen or electron donors in
transforming DPPHÁ to its reduced, stabile form, DPPH-H,
was measured spectrophotometrically [29]. The hydroxyl
radical scavenging activity of the examined compounds
was measured using a deoxyribose assay [29]. The pro-
tective effects of the tested compounds on 2-deoxy-D-ri-
bose were assessed as their ability to remove hydroxyl
radicals (formed in the Fenton reaction) from the test
solution and prevent deoxyribose degradation. The OH
radical scavenging activity of the tested compounds was
determined indirectly, by measuring the absorbance of the
resulting pink colored solutions.
14
Estradiol
Methyl salicylate
BHT
BHA
Mean antioxidant activity results with standard deviations are pre-
sented (mean SD)
the microwave-assisted reaction, while reaction times
decreased 17-fold.
The structures of the newly synthesized salicyloyl esters
8, 10, 11, 12 and 14 were deduced on the basis of their IR,
NMR and TOF spectra. In IR spectra of 8, 10, 11, 12 and
14, broad shallow bands associated with phenolic OH
stretching vibrations appear at 3205, 3148, 3139, 3187 and
3117 cm-1, respectively, along with strong bands of C=O
stretching vibrations at 1686 cm-1 (8), 1669 cm-1 (10),
1669 cm-1 (11), 1676 cm-1 (12) and 1663 cm-1 (14),
indicating an intramolecular hydrogen bond between the
phenolic OH group and C=O group (O–HÁÁO=C) within the
salicyloyloxy moiety in each molecule. This intramolecular
hydrogen bond is also indicated by sharp singlets at
10.55 ppm (8), 10.96 ppm (10), 10.91 ppm (11),
All tested compounds neutralized DPPH radical
(Table 4), while the strongest scavenger activity was dis-
played by the 17b-salicyloyloxy derivative 11, which was
more effective than estradiol and methyl salicylate. 17b-
Salicyloyloxy derivative 14, which has no typical (aro-
matic) estrane structure of A ring or phenolic group,
showed lower DPPH radical scavenging activity than
compounds 7 and 11 with 1,3,5(10)-estratriene structures.
All tested salicyloyl esters were more effective in OH
radical neutralization than commercially used antioxidants
BHT and BHA. 3-Methoxy17-salicyloyloxyestrane
derivative 11 expressed fivefold lower activity in OH
radical neutralization than its 3-hydroxy counterpart 7.
D-Seco estrane compound 5, which lacks a phenolic
1
10.71 ppm (12) and 10.87 ppm (14) in the appropriate H
NMR spectrum (OH phenolic group from salicyloyl resi-
due). The chemical shifts of H-17 in 10, 11 and 14 (triplets
at 4.98, 4.95, 4.85 ppm) and those of H-17a and H-17b in
12 (two doublets at 4.06 and 4.33 ppm) prove the presence
Table 5 In vitro cytotoxicity of
the tested compounds
Compound
IC50 (lM)
MCF-7
MDA-MB-231
PC-3
MRC-5
5
[100
[100
[100
[100
7 [29]
8
[100
[100
[100
[100
57.42 2.53
[100
[100
15.54 1.03
26.51 1.98
37.68 2.22
16.99 0.98
[100
[100
10
[100
[100
11
51.26 3.68
[100
48.78 3.76
91.03 4.69
[100
[100
12
[100
13
[100
[100
14
[100
0.75 0.03
8.99 1.05
0.12 0.02
[100
95.61 4.36
[100
0.12 0.01
DOX
Mean cytotoxicity results with standard deviations are presented (mean SD). Compounds with
IC50 [ 100 lM were considered as non-toxic
123

Struct Chem (2016) 27:947–960
955
hydroxyl function, was less active than salicyloyl deriva-
tives. Thus, specifically, the structures of rings A, B and D
in the tested salicyloyloxy derivatives affect their DPPH
and OH radical neutralization potential.
approximately four times more active than doxorubicin.
Doxorubicin was extremely toxic to normal noncancerous
MRC-5 cells, consistent with its nonspecific cellular
cytotoxicity. In contrast, none of the tested steroidal com-
pounds were toxic to healthy non-cancerous cells (MRC-
5).
The estrane compounds 5, 7, 8 and 10–14 were evalu-
ated for their in vitro cytotoxicity against human breast
adenocarcinoma (MCF-7, ER? and MDA-MB-231, ER-)
and prostate cancer cells (PC-3), as well as normal fetal
lung fibroblasts (MRC-5). Cytotoxicity was determined
using the standard SRB assay, after exposure of cells to test
compounds for 48 h [33]. Doxorubicin served as reference
compound and was used as positive control for general
toxicity. The results of cytotoxicity assay for the tested
compounds are presented in Table 5.
Against ER—human breast adenocarcinoma cells
(MDA-MB-231), only 17b-salicyloyloxy 19-nor-testos-
terone 14 (IC50 8.99 lM), with a 4-ene-3-one moiety in A
ring, exhibited fairly strong activity. However, salicyloy-
loxy derivatives with a planar estra-1,3,5(10)-triene moiety
in the A ring (7, 8, 10 and 12) are inactive against MDA-
MB-231 cells with the exception of compound 11 which
showed weak activity.
Prostate cancer PC-3 cells were the most sensitive to the
tested salicyloyloxy estrane compounds (Table 5). Based
on a comparison of IC50 values of 17b-salicyloyloxy
derivatives of estradiol with free (7), 3-benzoyloxy- (10) or
3-methoxy functions (11), higher cytotoxicity against PC-3
cells was observed for compounds 10 and 11, indicating the
favorable influence of substituents at the C3 position.
Steroidal esters of salicylic acid 8, 10 and 12 showed good
cytotoxicity against PC-3 cells, while compound 11
expressed moderate activity against all malignant cells.
Significantly increased antiproliferative activity against
PC-3 cells was observed for the D-seco salicyloyloxy
derivative 12, compared to its precursor 5, probably due to
the presence of the salicyloyl function in 12.
Estrane derivatives without a salicyloyl group, 5 and 13,
did not exhibit cytotoxic activity against neoplastic cells
indicating that the salicyloyl functional group has a posi-
tive influence on the cytotoxicity of these compounds.
Hydroxysteroid dehydrogenase enzymes (HSDs) are
involved in the biosynthetic pathways of steroid hormones,
so their inhibition represents an interesting approach for the
treatment of steroid-dependent diseases. The inhibitory
effects of specific steroidal compounds on selected HSDs
were tested, in order to assess their potential to prevent the
peripheral interconversion of steroid hormones between
their active and inactive forms.
The inhibitory effects of selected compounds on human
placental 17bHSD1 and rat testicular 17bHSD3 activity
were investigated by in vitro radio-substrate incubation
methods. Inhibition activities of salicyloyloxy estrane
derivatives 7, 10–12 and 14 against human 17bHSD1 and
rat 17bHSD3 were very weak (Table 6). 17b-Salicyloyl
ester of estradiol 7 efficiently inhibited rat testicular
3bHSD (IC50 9.3 lM), while other tested salicyloyloxy
Most of the tested compounds exhibited higher activity
against PC-3 cells than doxorubicin (DOX). Note that
DOX controls showed low cytotoxicity against PC-3-cells
(IC50 95.61 lM), in agreement with reported values [44].
The most active compounds against this cell line were
salicyloyloxy derivatives
8
and 12, which were
Table 6 In vitro inhibition of 3bHSD, 17bHSD1, 17bHSD2 and 17bHSD3 by selected steroidal compounds; non-inhibited controls are set as
100 %
Compound
r3bHSD
Relative conversion SD
h17bHSD1
Relative conversion SD
r17bHSD2
Relative conversion SD
r17bHSD3
Relative conversion SD
(% at 50 lM)
(% at 50 lM)
(% at 50 lM)
(% at 50 lM)
7
14 1.5
52
5
4.3 1.4
95
6
IC50 = 9.3 1.3 lM
IC50 = 0.59 0.10 lM
87 14
10
11
12
NI
92 11
81 17
56 13
NI
95
83
91
85
1
2
90
53
6
5
4
5
IC50 = 51 2 lM
34
IC50 = 13 1.5 lM
14
87
5
76
5
3
NI
Mean enzyme activity results, expressed as relative conversion of substrate, with standard deviations are presented (mean SD)
NI no inhibition
123

956
Struct Chem (2016) 27:947–960
Fig. 1 ORTEP [45] drawings of the molecular structures of com-
pounds 11 (a), 14 (b) with labeling of non-H atoms. Displacement
ellipsoids are shown at the 50 % probability level, and H atoms are
drawn as spheres of arbitrary radii. Intramolecular hydrogen bonds
are shown as dashed lines
derivatives only weakly inhibited this enzyme. Three of the
five tested substances expressed significant inhibitory
effects on 17bHSD2 activity. 17b-Salicyloyl ester of
nortestosterone 14 significantly inhibited this 17bHSD
isoform, D-secoestrane salicyloyloxy derivative 12 dis-
played moderate inhibitory activity, whereas the 17b-sali-
cyloyl ester of estradiol 7 strongly inhibited this isozyme
(IC50 0.59 lM). These results suggest the potential of 7 as
a lead compound for modeling and synthesis of therapeu-
tics for the treatment of pathological conditions caused by
estrogen depletion, e.g., osteopenia or osteoporosis.
The best rotational axis bisects C7–C8 and C5–C10 bonds
with asymmetry parameters [47] DC2 = 3.3(4)°. The
weighted average absolute torsion angle is 39.68(14)°, and
˚
weighted average ring bond distance is 1.4988(16) A. Ring
C in 11 has an almost perfect 8b,12a-chair (1C4) confor-
mation where the calculation starts from C8 to C14 and
proceeds in a clockwise direction (Table 8) with selected
asymmetry parameters: DCs(C11) = 0.7(2)°, DC2(C9–
C11) = 2.4(3)°, DC2(C11–C12) = 3.5(3)°, DC2(C12–
C13) = 5.0(3)° and a weighted average absolute torsion
angle of 55.70(13)° and weighted average ring bond dis-
˚
ORTEP [45] drawings of the molecular structures of 11
and 14 are depicted in Fig. 1, while selected bond dis-
tances, bond angles and torsion angles within these com-
pounds are given in Table 7. Compound 11 crystallizes in
the monoclinic non-centrosymmetric P21 space group with
two molecules in the unit cell, while compound 14 crys-
tallizes in the orthorhombic P212121 space group with four
molecules in the unit cell. Intramolecular geometry anal-
ysis showed that bond lengths, angles and torsion angles in
molecule of compound 11 are in agreement with those in
the same region (excluding the C3 functional groups) of
compound 14 (Table 7).
tance of 1.5263(16) A. Ring D in 11 has a 13b-envelope
(1E) conformation deformed toward a 13b,14a-halfchair
(1T2) conformation, where the calculation starts from C13
to C17 and proceeds in a counter-clockwise direction
(Table 8) with the best mirror plane passing through C13
with DCs = 7.8(3)° and rotational symmetry with
DC2(C13–C14) = 13.2(3)°. The weighted average absolute
torsion angle for ring D in 11 is 30.28(13)°, and the
˚
weighted average ring bond distance is 1.5274(18) A.
The keto group bonded to C3 disturbs the A ring con-
formation in 14. The A ring in 14 has a 1a-envelope (E1)
conformation slightly deformed toward a 1a,2b-halfchair
(2H1) (puckering parameters presented in Table 8 where the
calculation starts from C1 to C10 and proceeds in a counter-
clockwise direction). Distortion from the ideal form could be
expressed through asymmetry parameters due to the loss of
mirror symmetry through C1 atom [DCs(C1) = 7.8(3)°] and
rotational symmetry [DC2(C1–C2) = 19.3(3)°]. The
weighted average absolute torsion angle is 29.79(13)°, and
The ABCD steroid backbone of 11 has a methyl group
attached to C13 and shows trans C/D ring junctions. The
molecular region consisting of the methoxy group and
aromatic A ring in 11 is almost perfectly planar with the
torsion angle C2-C3-O1-C19 equal to 175.0(3)° and C4-
C3-O1-C19 equal to -3.4(4)°. The B ring in 11 has a
7a,8b-halfchair (3H4) conformation (puckering parameters
[46] presented in Table 8 where the calculation starts from
C5 to C10 and proceeds in a counter-clockwise direction).
˚
weighted average ring bond distance is 1.4904(14) A. The B
ring in 14 has a 5a,8b-chair (1C4) conformation (puckering
123

Struct Chem (2016) 27:947–960
957
˚
˚
Table 7 Selected bond lengths (A) and angles (°)
Table 9 Hydrogen bond parameters (A, °)
Compound 11
Compound 14
D–HÁÁÁA
D–H
HÁÁÁA
DÁÁÁA
D–HÁÁÁA
Bond
Bond
Compound 11
O2–C20
1.324(3)
1.465(3)
1.217(3)
1.350(4)
1.384(3)
O2–C19
1.333(2)
1.453(3)
1.225(2)
1.356(3)
1.222(3)
O4–H4ÁÁÁO3
1.02(5)
0.96
1.70(5)
2.43
2.605(3)
3.374(4)
146(4)
166
O2–C17
O2–C17
C19–H19BÁÁÁO4
[1 ? x, y, 1 ? z]
O3–C20
O3–C19
O4–C22
O4–C21
Compound 14
O4–H4ÁÁÁO3
O1–C3
O1–C3
0.91(3)
1.78(3)
2.609(3)
150(3)
Angle
Angle
C20–O2–C17
O3–C20–O2
118.8(2)
123.0(3)
124.2(3)
112.7(3)
117.6(3)
122.6(3)
113.9(2)
109.3(2)
C19–O2–C17
O3–C19–O2
117.97(17)
122.0(2)
a clockwise direction). The best rotational axis bisects C9–
C11 and C13–C14 bonds with DC2 = 3.1(2)°. The best
mirror plane passes through C11 and C14 with DCs =
2.31(16)°. The weighted average absolute torsion angle in C
ring is 54.92(8)°, and weighted average ring bond distance is
O3–C20–C21
O2–C20–C21
O4–C22–C23
O4–C22–C21
O2–C17–C16
O2–C17–C13
Torsion angle
O1–C3–C2–C1
C5–C4–C3–O1
C17–O2–C20–O3
C17–O2–C20––C21
O3–C20–C21–C22
O3–C20–C21–C26
C20–C21–C22–O4
C26–C21–C22–O4
C20–O2–C17–C13
C20–O2–C17–C16
C12–C13–C17–O2
C14–C13–C17–O2
C18–C13–C17–O2
O2–C17–C16–C15
O3–C19–C20
O2–C19–C20
O4–C21–C22
O4–C21–C20
O2–C17–C16
O2–C17–C13
Torsion angle
C1–C2–C3–O1
O1–C3–C4–C5
C17–O2–C19–O3
C17–O2–C19–C20
C21–C20–C19–O3
C25–C20–C19–O3
C19–C20–C21–O4
C25–C20–C21–O4
C19–O2–C17–C13
C19–O2–C17–C16
C12–C13–C17–O2
C14–C13–C17–O2
C18–C13–C17–O2
O2–C17–C16–C15
124.6(2)
113.40(19)
118.1(2)
122.0(2)
112.57(18)
110.18(15)
˚
1.5273(12) A. Five-membered ring D in 14 has a 13b,14a-
halfchair (1T2) distorted to a 13b-envelope (1E) conforma-
tion (puckering parameters presented in Table 8 where the
calculation starts from C13 to C17 and proceeds in a
counter-clockwise direction) with asymmetry parameters
DCs(C13) = 13.2(2)° and DC2(C13–C14) = 6.0(3)° with a
weighted average absolute torsion angle of 34.40(10)° and
-177.2(2)
176.7(2)
-7.3(4)
174.5(2)
-8.5(4)
170.1(3)
0.7(4)
-155.8(2)
-173.9(2)
4.8(3)
-173.48(17)
-8.6(3)
˚
weighted average ring bond distance of 1.5354(14) A.
172.67(19)
1.4(3)
Information on intramolecular hydrogen bonding is very
important for understanding various molecular properties,
including molecular geometry and the stability of pre-
dominant conformations and, consequently, the biological
activity of a compound of interest. As can be seen in Fig. 1,
both 11 and 14 contain intramolecular O4–H4ÁÁÁO3
hydrogen bonds. Hydrogen bond parameters are given in
Table 9.
-178.0(3)
-168.4(2)
73.4(3)
-179.76(19)
-176.58(16)
65.5(2)
77.2(3)
83.6(2)
-165.8(2)
-48.4(3)
142.8(3)
-161.09(16)
-43.2(2)
138.72(19)
The crystal packings of 11 and 14 are illustrated in
Fig. 2. The crystal packing of 11 is arranged by C19–
H19BÁÁÁO4 (Table 9) contacts where these intermolecular
interactions lead to infinite planes of head to tail connected
molecules. Atom O4 in 11 acts as acceptor in both
intramolecular O4–H4ÁÁÁO3 hydrogen bond and inter-
molecular C19–H19BÁÁÁO4 contact (Fig. 3). The crystal
packing of 14 is dominantly arranged by van der Waals
forces and corresponds to a discrete arrangement of
molecules (Fig. 2). We have not found classic hydrogen
bonding or other contacts in the intermolecular space of 14.
Comparing the molecular structures of 11 and 14, we
observed minor differences in conformation among rings C
parameters presented in Table 8 where the calculation starts
from C5 to C10 and proceeds in a counter-clockwise
direction). Rotational symmetry is presented with the best
rotational axis bisecting C7–C8 and C10–C5 bonds with
DC2 = 3.6(3)°. The best mirror plane passes through C5 and
C8 with DCs = 4.6(2)°. The weighted average absolute
torsion angle is 49.75(12)°, and weighted average ring bond
˚
distance is 1.5244(13) A. Ring C in 14 has a 8b,12a-chair
(1C4) conformation that is typically distorted in the crys-
talline state (puckering parameters presented in Table 8
where the calculation starts from C8 to C14 and proceeds in
Table 8 The puckering
Compound 11
Compound 14
parameters for compounds 11
and 14
B ring
C ring
D ring
A ring
B ring
C ring
D ring
˚
Q (A)
0.495(3)
45.2(3)
0.568(3)
2.5(3)
0.458(3)
–
0.450(3)
54.6(4)
8.2(4)
0.521(3)
15.6(3)
0.565(2)
6.3(2)
0.455(2)
–
h (°)
u (°)
154.7(5)
299(6)
188.1(4)
165.5(11)
279(2)
193.4(3)
123

958
Struct Chem (2016) 27:947–960
Fig. 2 MERCURY [49] drawings showing the crystal packing of: a compound 11 (along b axis) and b compound 14 (along a-axis)
rings (methoxy group bonded to C3 in aromatic A ring in
11 and keto group bonded to C3 in 14) produce some
conformational changes in the steroid framework.
To complement our investigation of the cytotoxicity of
compounds 11 and 14, we determined the conformation of
the molecules released from the influence of the crystalline
field, i.e., the next step was to define the conformations of
11 and 14 in terms of energy minima. Both X-ray crystal
structures of compounds 11 and 14 were subjected to
exhaustive conjugate gradient energy minimization using
an MMFF94 force field (convergence setting of
10e-7 kJ/mol), followed by molecular mechanics con-
former analysis using a systematic rotomer search in the
program Avogadro [48].
For compound 11 with four rotatable bonds, 899 con-
formers were tested. For compound 14 with three rotatable
bonds, 149 conformers were tested. For both 11 and 14,
conformer analysis indicates conformational flexibility in
the form of rotation around the C17ÁÁÁO2 bond (indicated
by black arrows in Fig. 3) resulting in conformers of nearly
equal energy (range 410.25–520.47 kJ/mol for 11 and
357.38–461.34 kJ/mol for 14). On the other hand, very
good overlap in the region of the steroidal nucleus (ex-
cluding the C3 functional group) was observed for the
molecular structures of molecules 11 and 14 in the crys-
talline state and after energy optimization (Fig. 3). The
O4–H4ÁÁÁO3 hydrogen bond (indicated by a red dashed line
Fig. 3 Superimposed fit of the molecules after conjugate energy
in Fig. 3) observed in the X-ray crystal structures of
minimization (C atoms in green) and the molecule in the crystalline
compounds 11 and 14 is maintained following energy
state (C atoms in white): a compound 11 and b compound 14.
Hydrogen bonds observed during energy minimization for 11 and 14
are shown as red dashed lines (Color figure online)
minimization and conformer analysis.
In our previous study [28], we reported the crystal
structure of (3-oxo-5a-androstan-17b-yl)-2-methoxyben-
zoate (15) (Fig. 4a) as well as its strong cytotoxicity
against MDA-MB-231 cells (IC50 3.45 lM). Structural
superposition of 14 with previously reported compound 15
(with comparable antiproliferative activity against MDA-
and D across 11 and 14 which may be attributed to the
different intermolecular interactions. On the other hand,
when rings A and B in 11 and 14 are compared, it is clear,
as expected, that the modifications introduced in the A
123

Struct Chem (2016) 27:947–960
959
Fig. 4 a Structural formulas of
the compounds whose structural
superpositions are studied.
b Structural superposition of 14
with a previously reported
compound 15. c Structural
superposition of 11 with
compound 15. Hydrogen bonds
observed during energy
minimization for 11 and 14 are
shown as red dashed lines
(Color figure online)
MB-231 breast cancer cells, IC50 8.99 lM vs. 3.45 lM)
showed rotation around the C17ÁÁÁO2 bond during con-
former analysis (indicated by black arrows in Fig. 4b),
while a very good overlap in the region of steroidal nucleus
was observed.
antiproliferative activity against MDA-MB-231 cell lines.
On the other hand, new compounds were selective in their
inhibition of steroidogenesis enzymes. Because of the
selectivity of some tested salicyloyloxy estrane derivatives
toward different HSDs, inhibitors of 17b-HSD2 described
here may prove beneficial for conditions in which the
concentration of active steroid is too low.
Structural superposition of 11, which has moderate
antiproliferative activity against MDA-MB-231 (IC50
48.78 lM), with compound 15 revealed significant struc-
tural differences between 11 and 15 in the A ring (Fig. 4c),
and conformational flexibility around the C17ÁÁÁO2 bond was
also observed for 11 and 15 during conformer analysis
(indicated by black arrows in Fig. 4c). It seems that the
structural geometry in the region of ring A and especially the
C3 functional group is important structural features influ-
encing antiproliferative activity against MDA-MB-231.
Supplementary data
CCDC 1408660-1408661 contains supplementary crystal-
lographic data for this paper. These data are available free
of charge at www.ccdc.cam.ac.uk/conts/retrieving.html [or
from the Cambridge Crystallographic Data Centre
(CCDC), 12 Union Road, Cambridge CB2 1EZ, UK; fax:
?44(0)1223-336033; email: deposit@ccdc.cam.ac.uk].
Conclusions
Acknowledgments Authors would like to thank the Hungary-Serbia
IPA Cross-border Co-operation Programme (Project No. HUSRB/
1002/214/133 RECODAC), Provincial Secretariat for Science and
Technological Development of Vojvodina (Grant No. 114-451-946/
2015-03) and Ministry of Education, Science and Technological
Development of the Republic of Serbia for financial support (Grant
Based on a comparison of transesterification reactions of
methyl salicylate with steroidal alcohols or phenols 1–6 by
MW irradiation versus conventional heating, the MW-as-
sisted method considerably accelerated reactions 11- to 26-
fold. In the case of starting steroidal compounds 1, 4, 5 and
6, MW-assisted reactions also resulted in considerably
higher total product yields.
´
No. 172021). The work of N. Szabo was supported by a PhD Fel-
lowship of the Talentum Fund of Richter Gedeon Plc.
New estrane salicyloyloxy derivatives expressed
antioxidative and/or antiproliferative potential. It can be
concluded based on X-ray analysis that the structural
geometry in the region of ring A and especially the struc-
ture of the C3 functional group in the synthesized com-
pounds is important structural features concerning
References
1. Rasmussen MK, Ekstrand B, Zamaratskaia G (2013) Int J Mol
Sci 14:17926–17942
¨
2. Vihko P, Herrala A, Harkonen P, Isomaa V, Kaija H, Kurkela R,
Li Y, Patrikainen L, Pulkka A, Soronen P, Torn S (2005) J
¨
¨
Steroid Biochem 93:277–283
123

960
Struct Chem (2016) 27:947–960
´ ˇ ´ ˇ ´ ˇ ´
25. Djurendic E, Dojcinovic-Vujaskovic S, Sakac M, Ajdukovic J,
3. Marchais-Oberwinkler S, Henn C, Mu¨ller G, Klein T, Negri M,
´
Gakovic A, Kojic V, Bogdanovic G, Klisuric O, Penov-Gasi K
(2011) Arkivoc 2:83–102
´
´
´
ˇ
Oster A, Spadaro A, Werth R, Wetzel M, Xu K, Frotscher M,
Hartmann RW, Adamski J (2011) J Steroid Biochem Mol Biol
125:66–82
ˇ
´
26. Djurendic E, Savic M, Jovanovic-Santa S, Sakac M, Kojic V,
Szecsi M, Okljesa A, Posa M, Penov Gasi K (2014) Acta Period
´
´
ˇ
´
´
ˇ
ˇ
ˇ
4. Tremblay MR, Lin SX, Poirier D (2001) Steroids 66:821–831
5. Maltais R, Ayan D, Trottier A, Barbeau X, Lagu¨e P, Bouchard
JE, Poirier D (2014) J Med Chem 57:204–222
6. Maltais R, Trottier A, Delhommea A, Barbeau X, Lagu¨e P,
Poirier D (2015) Eur J Med Chem 93:470–480
Technol 45:173–189
ˇ
´
27. Djurendic E, Klisuric O, Szecsi M, Sakac M, Jovanovic-Santa S,
Ignath I, Kojic V, Okljesa A, Savic M, Penov Gasi K (2014)
´
´
´
ˇ
´
´
ˇ
´
ˇ
Struct Chem 25:1747–1758
ˇ ´ ´
28. Penov Gasi KM, Djurendic EA, Szecsi M, Gardi J, Csanadi JJ,
´
´
¨
7. Szabo J, Bacsa I, Wolfling J, Schneider G, Zupko I, Varga M,
´
´
´
´ ˇ ´ ˇ ´ ´ ´
Klisuric OR, Dojcinovic-Vujaskovic SV, Nikolic AR, Savic MP,
Herman BE, Kalmar L, Szecsi M, Mernyak E (2015) J Enzyme
Inhib Med Chem. doi:10.3109/14756366.2015.1050008
8. Sam KM, Labrie F, Poirier D (2000) Eur J Med Chem
35:217–225
´
Ajdukovic JJ, Okljesa AM, Kojic VV, Sakac MN, Jovanovic-
Santa SS (2015) Steroids 94:31–40
ˇ
´
ˇ
´
ˇ
ˇ
29. Penov Gasi K, Djurendic E, Dojcinovic Vujaskovic S, Gakovic
A, Jovanovic-Santa S, Kojic V, Sakac M (2012) Chem Pap
´
ˇ
´
ˇ
´
´
ˇ
´
´
ˇ
9. Bydal P, Auger S, Poirier D (2004) Steroids 69:325–342
10. Spires TE, Fink BE, Kick EK, You D, Rizzo CA, Takenaka I,
Lawrence MR, Ruan Z, Salvati ME, Vite GD, Weinmann R,
Attar RM, Gottardis MM, Lorenzi MV (2005) Prostate
65(2):159–170
66(4):284–294
30. Klaunig JE, Kamendulis LM, Hocevar BA (2010) Toxicol Pathol
38:96–109
¨
31. Butenandt A, Stromer I, Westphal V (1932) Z Physiol Chem
11. Salvador JAR, Carvalho JFS, Neves MAC, Silvestre SM, Leita˜o
208:149–172
ˇ
´ ´ ˇ ´
32. Jovanovic-Santa S, Petri E, Klisuric O, Szecsi M, Kovacevic R,
´
AJ, Silva MMC, Sa e Melo ML (2013) Nat Prod Rep 30:324–374
12. Okljesa A, Jovanovic-Santa S, Klisuric O, Sakac M, Djurendic E,
ˇ
ˇ
´
´
ˇ
´
´
Petrovic J (2015) Steroids 97:45–53
´ ˇ
Jakimov D, Aleksic L, Penov Gasi K (2013) J Braz Chem Soc
24(10):1613–1622
ˇ
33. Skehan P, Storeng R, Scudiero D, Monks A, McMahon J, Vistica
D, Warren TJ, Bokesch H, Kenney S, Boyd RM (1990) J Natl
Cancer Inst 82:1107–1112
´
´
13. Sakac MN, Gakovic AR, Csanadi JJ, Djurendic EA, Klisuric O,
Kojic V, Bogdanovic G, Penov Gasi KM (2009) Tetrahedron Lett
´
´
´
ˇ
´
´
34. Toth I, Szecsi M, Julesz J, Faredin I (1997) Skin Pharmacol
10(3):160–168
50:4107–4109
ˇ
14. Jovanovic-Santa SS, Andric S, Andric N, Bogdanovic G, Petrovic
´
´
´
JA (2011) Med Chem Res 20:1102–1110
´
´
´ ´
´
´ ´
35. Darvas B, Szekacs A, Fonagy A, Szecsi M, Toth I (1997) Gen
Comp Endocrinol 107(3):450–460
´
15. Penov Gasi KM, Okljesa AM, Petri ET, Celic AS, Djurendic EA,
ˇ
ˇ
´
´
Klisuric OR, Csanadi JJ, Batta G, Nikolic AR, Jakimov DS,
36. Oxford Diffraction CrysAlis CCD and CrysAlis RED (2008)
Versions 1.171. Oxford Diffraction Ltd, Abingdon, Oxfordshire,
England
37. Sheldrick GM (2008) Acta Crystallogr A 64:112–122
38. Altomare A, Cascarano G, Giacovazzo C, Guagliardi A (1993) J
Appl Cryst 26:343–350
39. Bernardinelli G, Flack HD (1985) Acta Cryst A41:500–511
40. Nardelli MJ (1995) Appl Cryst 28:659–662
41. Spek AL (1998) PLATON, a multipurpose crystallographic tool.
University of Utrecht, The Netherlands
´
Sakac MN (2013) MedChemComm 4:317–323
´
ˇ
16. Mernyak E, Fiser G, Szabo J, Bodnar B, Schneider G, Kovacs I,
´
Ocsovszki I, Zupko I, Wolfling J (2014) Steroids 89:47–55
´
´
´
´
17. Mernyak E, Szabo J, Huber J, Schneider G, Minorics R, Bozsity
¨
´
N, Zupko I, Varga M, Bikadi Z, Hazai E, Wolfling J (2014)
´
´
´
Steroids 87:128–136
´
¨
¨
18. Romer W, Oettel M, Menzenbach B, Droescher P, Schwarz S
(1997) Steroids 62:688–694
19. Behl C, Skutella T, Lezoualc’h F, Post A, Widmann M, Newton
CJ, Holsboer F (1997) Mol Pharmacol 51:535–541
20. Perez E, Cai ZY, Covey DF, Simpkins JW (2006) Drug Dev Res
66:78–92
21. Ott I, Schmidt K, Kircher B, Schumacher P, Wiglenda T, Gust R
(2005) J Med Chem 48:622–629
42. Farrugia LJ (1999) J Appl Cryst 32:876–881
ˇ
43. Jovanovic-Santa S, Andric S, Kovacevic R, Pejanovic V (2000)
´
´
ˇ
Collect Czech Chem Commun 65:77–82
´
´
44. David-Beabes GL, Overman MJ, Petrofski JA, Campbell PA, de
Marzo AM, Nelson WG (2000) Int J Oncol 17:1077–1086
45. Farrugia LJ (1997) J Appl Crystallogr 30:565
22. Ekinci D, Sentu¨rk M, Ku¨frevioglu OI (2011) Expert Opin Ther
Pat 21(12):1831–1841
23. Fletcher S, Singh J, Zhang X, Yue P, Page BDG, Sharmeen S,
Shahani VM, Zhao W, Schimmer AD, Turkson J, Gunning PT
(2009) ChemBioChem 10(12):1959–1964
46. Cremer D, Pople JA (1975) J Am Chem Soc 97:1354–1358
47. Duax WL, Weeks CM, Rohrer DC (1976) In: Eliel EL, Allinger N
(eds) Topicsin stereochemistry, vol2. Wiley, NewYork, pp271–283
48. Hanwell MD, Curtis DE, Lonie DC, Vandermeersch T, Zurek E,
Hutchison GR (2012) Avogadro: an advanced semantic chemical
editor, visualization, and analysis platform. J Cheminform 4:17
49. Macrae CF, Edgington PR, McCabe P, Pidcock E, Shields GP,
Taylor R, van der Streek T (2006) J Appl Cryst 39:453–457
50. Clark RC, Reid JS (1995) Acta Cryst A51:887–897
´
ˇ
´
24. Djurendic EA, Dojcinovic-Vujaskovic SV, Sakac MN, Jovin
EDJ, Kojic VV, Bogdanovic GM, Klisuric OR, Stankovic SM,
ˇ
´
ˇ
´
´
´
´
ˇ
Lazar DV, Fabian L, Penov-Gasi KM (2010) Struct Chem
21:67–78
123