Synthesis and biological evaluation of some 17-picolyl and 17-picolinylidene androst-5-ene derivatives

Article abstract of DOI:10.1016/j.steroids.2006.10.002

Starting from dehydroepiandrosterone (1) 17-picolyl (2), 17-picolinylidene (7), 17-picolinylidene-16-one (10 and 11), and 17-picolyl-16-one (15) derivatives of androst-5-ene were synthesized in one, two, four and five steps respectively. By the Oppenauer oxidation or dehydration of 2, 7, 10, and 11 with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), the corresponding A and B ring modified derivatives 3, 5, 6, 8, 9, and 12-14 were obtained. The structure of 2 was unambiguously proved by the appropriate X-ray structural analysis. Compounds 3, 5, 9, 12-14 showed inhibitory activity against the enzyme aromatase. Antibacterial activity, toxicity to brine shrimp Artemia salina, antitumor activity against three tumor cell lines (human cervix carcinoma HeLa cells, human melanoma FemX cells, and human myelogenous leukemia K562 cells) and toxicity against peripheral blood mononuclear cells were evaluated. Three tested compounds, namely 11, 13, and 15, showed strong activity against all three cell lines, the IC₅₀ values being in the range of 4-10 μM.

Full text of DOI:10.1016/j.steroids.2006.10.002

steroids 7 2 ( 2 0 0 7 ) 31–40
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/steroids
Synthesis and biological evaluation of some 17-picolyl
and 17-picolinylidene androst-5-ene derivatives
Katarina M. Penov Gasˇia,∗, Maja Dj. Djurendic´ Brenesela, Evgenija A. Djurendic´ a,
a
a
a
b
ˇ
Marija N. Sakacˇ , Janosˇ J. Canadi , Jovana J. Daljev , Thomas Armbruster ,
Silvana Andric´ c, Dusˇan M. Sladic´ d, Tatjana T. Bozˇic´ d,
Irena T. Novakovic´ e, Zorica D. Juranic´ f
a
´
Department of Chemistry, Faculty of Sciences, University of Novi Sad, 21000 Novi Sad, Trg Dositeja Obradovica 3, Serbia
b
c
d
e
f
Laboratory for Chemical and Mineralogical Crystallography, Bern University, Freierstr. 3, CH-3012 Bern, Switzerland
´
Department of Biology, Faculty of Sciences, University of Novi Sad, 21000 Novi Sad, Trg Dositeja Obradovica 2, Serbia
Faculty of Chemistry, University of Belgrade, Studentski trg 16, P.O. Box 158, 11000 Belgrade, Serbia
ˇ
Center for Chemistry ICTM, Njegoseva 12, P.O. Box 473, 11000 Belgrade, Serbia
Department of Experimental Oncology, Institute for Oncology and Radiology of Serbia, Pasterova 14, 11000 Belgrade, Serbia
a r t i c l e i n f o
a b s t r a c t
Article history:
Starting from dehydroepiandrosterone (1) 17-picolyl (2), 17-picolinylidene (7), 17-
picolinylidene-16-one (10 and 11), and 17-picolyl-16-one (15) derivatives of androst-5-ene
were synthesized in one, two, four and five steps respectively. By the Oppenauer oxidation or
dehydration of 2, 7, 10, and 11 with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), the
corresponding A and B ring modified derivatives 3, 5, 6, 8, 9, and 12–14 were obtained. The
structure of 2 was unambiguously proved by the appropriate X-ray structural analysis. Com-
pounds 3, 5, 9, 12–14 showed inhibitory activity against the enzyme aromatase. Antibacterial
activity, toxicity to brine shrimp Artemia salina, antitumor activity against three tumor cell
lines (human cervix carcinoma HeLa cells, human melanoma FemX cells, and human myel-
ogenous leukemia K562 cells) and toxicity against peripheral blood mononuclear cells were
evaluated. Three tested compounds, namely 11, 13, and 15, showed strong activity against
all three cell lines, the IC50 values being in the range of 4–10 ␮M.
Received 17 April 2006
Received in revised form
11 October 2006
Accepted 16 October 2006
Published on line 21 November 2006
Keywords:
Androst-5-ene derivatives
Picolyl and picolinylidene
derivatives
Aromatase inhibition
Antibacterial activity
Antitumor activity
X-ray analysis
© 2006 Elsevier Inc. All rights reserved.
1.
Introduction
dependent breast cancer. Over the past two decades, highly
potent and specific aromatase inhibitors have been developed
Aromatase is the rate-limiting enzyme that catalyzes the
conversion of androgens to estrogens. Blockade of this
step allows treatment of diseases that are dependent
upon estrogen. Compounds that inhibit enzyme aromatase
have applications in the treatment of advanced estrogen-
[1]. An important recent question is whether aromatase
inhibitors are superior to the antiestrogens for treatment
of hormone-dependent breast cancer. The third generation
aromatase inhibitors (anastrozole, letrozole, and exemestane)
have been compared to tamoxifen for the treatment of breast
∗
Corresponding author.
E-mail addresses: kpg@ih.ns.ac.yu, marijas@ih.ns.ac.yu (K.M.P. Gasˇi).
0039-128X/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2006.10.002

32
steroids 7 2 ( 2 0 0 7 ) 31–40
cancer in the advanced, adjuvant, and neoadjuvant settings
[2–4]. All of these studies suggest the superiority of aromatase
inhibitors over tamoxifen. The mechanism responsible
for the superiority of the aromatase inhibitors relates to
the estrogen agonistic effects of tamoxifen. Data from a
large adjuvant therapy trial provide evidence that the aro-
matase inhibitors may also be superior for breast cancer
prevention [5,6]. The recent advances in development of the
aromatase inhibitors clearly demonstrate the utility of these
agents for treatment of breast cancer.
Bearing in mind that the presence of a 4-en-3-one system
and expanded conjugation in the A and B rings leads to the
change of inhibitory activity of some steroids, as well as that
the non-steroid inhibitor of aromatase rogletimid (a pyridyl
analogue of aminoglutetimide) contains a pyridine ring as an
important constitutive element [7] we have synthesized sev-
eral 17-picolyl- and 17-picolinylidene-3␤-hydroxy-androst-5-
ene derivatives and their 4-en-3-one and 1,4,6-triene-3-one
analogues with the aim of continuing on our study of the inhi-
bition effect of D-modified steroids on the enzymes that take
part in the biosynthesis of steroids [8–11].
Since aromatase inhibition can be related to antitumor
activity, evaluation of activity of the synthesized derivatives
against three tumor cell lines was performed. The tested
cell lines were human cervix carcinoma HeLa cells, human
melanoma FemX and human myelogenous leukemia K562
cells. HeLa cells are known to be aromatase negative [12]. On
the other hand, aromatase activity was found in melanoma
tissue [13]. As for leukemia cells, aromatase activity was found
in some cell lines [14], but there are no data on aromatase
status in K562 cells. Therefore, results might give information
whether antitumor activity, if any, is related to aromatase inhi-
bition, or some other mechanisms might be implicated. Toxi-
city to brine shrimp Artemia salina, which is considered to be
well correlated to antitumor activity [15], was also determined.
In A. salina no aromatase activity was detected [16], so that
possible effect should be aromatase-independent. Besides,
toxicity of the derivatives to normal human peripheral blood
mononuclear cells was determined in order to better estimate
potential application of these compounds. Antibacterial activ-
ity against Gram-positive bacteria Staphylococcus aureus and
Gram-negative bacteria Escherichia coli was also evaluated. An
increase in resistance of bacteria to available antimicrobial
agents is continually developing. An approach to this prob-
lem might be the synthesis of antibacterial agents structurally
unrelated to clinically used antimicrobials. It was reported
that the steroidal amide 3␤-acetoxy-17␤-(l-prolyl)amino-5␣-
androstane showed activity to Gram-positive bacteria [17].
Since androst-5-ene derivatives synthesized in this work also
have a basic nitrogen atom, investigation of their antibacterial
activity seems reasonable.
ter. UV spectra (wave lengths in nm) were recorded on CECIL
CE 2021 spectrophotometer at room temperature. NMR spec-
tra were taken on a Bruker AC 250E spectrometer operating
at 250 MHz (proton) and 62.9 MHz (carbon), using standard
Bruker software; the tetramethylsilane peak (ı 0.00) was used
as reference for 1H NMR, whereas the central carbon line of
chloroform-d was set at 77.0 ppm for carbon-13 NMR. GC/MS
analyses were recorded on an Agilent Technologies GC 6890N
instrument with Mass Selective Detector 5973; the first num-
ber denotes m/z value, and the ion abundances are given in
parentheses. Chromatographic separations were performed
on silica gel columns (Kieselgel 60, 0.040–0.063 mm, Merck).
The term “dried” refers to the use of anhydrous sodium sul-
fate. All reagents used were of analytical grade.
2.2.
Chemical synthesis
2.2.1. 17ˇ-Hydroxy-17˛-picolyl-androst-4-ene-3-one (3)
A mixture of compound 2 [25,26] (0.30 g, 0.79 mmol), aluminum
tert-butoxide (0.32 g, 1.32 mmol) and cyclohexanone (10 ml,
96.6 mmol) was refluxed at constant stirring for 3 h. After
that, cyclohexanone was removed by steam distillation. The
residue was extracted with ethyl acetate. Removal of dried sol-
vents gave a residue (0.86 g), which was separated by column
chromatography (30 g silica gel). Eluting with toluene–AcOEt,
7:2 gave compound 3 (0.118 g, 40%, mp 184–185 ◦C) and 4-
androstene-3,17-dione (4; 16 mg, 6%, mp 170–171 ◦C, lit. [18]
mp 174 ◦C).
Compound 3: IR (KBr): 3423, 1673, 1616, 1596, 1569, 1052,
1024, 765, 513. UV (MeOH): ꢀmax (ε) 206.3 (13,472); 242.6 (13,947);
300.1 (379). 1H NMR (CDCl3): 1.00 (3H, s, H-18); 1.21 (3H, s, H-19);
2.79 (1H, d, Jgem = 14.6 Hz, CH2-Py); 3.06 (1H, d, Jgem = 14.6 Hz,
CH2-Py); 5.74 (1H, s, H-4); 6.70 (1H, bs, OH-17); 7.15 (d, 1H,
J = 7.7 Hz, H-3ꢀ, Py), 7.17 (m, 1H, H-5ꢀ, Py), 7.63 (td, 1H, J1 = 7.7 Hz,
J2 = 1.8 Hz, H-4ꢀ, Py); 8.47 (m, 1H, H-6ꢀ, Py). 13C NMR (CDCl3):
199.7 (C-3); 171.5 (C-5); 160.6 (C-2ꢀ, Py); 147.9 (C-6ꢀ, Py); 136.8 (C-
4ꢀ, Py); 124.7 (C-3ꢀ, Py); 123.8 (C-4); 121.4 (C-5ꢀ, Py); 83.3 (C-17);
53.9; 50.1; 46.4; 43.0; 38.7; 36.3; 35.9; 35.7; 34.0; 32.9; 32.0; 31.8;
23.8; 20.7; 17.4; 14.1. MS: 286 (100; M+-CH2Py); 244 (55); 124 (53).
Anal. calcd. for C25H33NO2 0.5 H2O: C, 77.32; H, 8.76; N, 3.61;
found: C, 77.62; H, 8.50; N, 3.49.
2.2.2. 17ˇ-Hydroxy-17˛-picolyl-androsta-1,4,6-triene-3-
one (5)
Compound 2 (0.30 g, 0.79 mmol) was dissolved in dry dioxan
(15 ml), then DDQ (0.59 g, 2.59 mmol) was added and the reac-
tion mixture was stirred at reflux for 24 h, and then filtered.
The precipitate was treated with methylene chloride several
times and the combined extracts were washed first with sat-
urated solution of NaHCO3, then with water. Removal of dried
solvent gave a residue (0.19 g), which was separated by column
chromatography (20 g silica gel). Eluting with benzene–AcOEt,
9:1, gave compound 5 (0.04 g, 12%; mp 169–170 ◦C).
IR (KBr): 3430, 1650, 1610, 1590, 1580, 900, 760. 1H NMR
(CDCl3): 1.08 (3H, s, H-18); 1.23 (3H, s, H-19); 2.43 (1H, m, H-8);
2.78 (1H, d, Jgem = 14.6 Hz, CH2-Py); 3.08 (1H, d, Jgem = 14.6 Hz,
CH2-Py); 6.01(1H, d, J2,4 ∼ 1 Hz, H-4); 6.06 (1H, dd, J6,7 = 9.9 Hz,
J6,8 = 1.9 Hz, H-6); 6.25 (1H, dd, J1,2 = 10.2 Hz, J2,4 ∼ 1 Hz, H-2); 6.26
(1H, dd, J6,7 = 9.9 Hz, J7,8 = 2.5 Hz, H-7); 7.08 (1H, d, J1,2 = 10.2 Hz,
H-1); 7.10 (1H, d, H-3ꢀ, Py); 7.18 (1H, dd, H-5ꢀ, Py), 7.63 (1H, td,
2.
Experimental
2.1.
General
Melting points were determined using a Bu¨ chi SMP 20 appa-
ratus and are uncorrected. IR spectra (wave numbers in cm−1
were recorded on a Perkin-Elmer FT-IR 1725 X spectrome-
)

steroids 7 2 ( 2 0 0 7 ) 31–40
33
H-4ꢀ, Py); 8.47 (1H, dd, H-6ꢀ, Py). 13C NMR (CDCl3): 186.33 (C-3);
162.61 (C-5); 160.28; 153.03 (C-1); 147.81 (C-6ꢀ, Py); 138.08 (C-
6); 137.07 (C-4ꢀ, Py); 128.03 (CH); 127.61 (CH); 124.73 (C-3ꢀ, Py);
123.74 (C-4); 121.49 (C-5ꢀ, Py); 82.97 (C-17); 48.58 (CH); 48.03 (CH);
46.93; 42.91; 41.27; 38.74 (CH); 35.83; 32.05; 23.29; 21.58; 20.7
(CH3); 14.1 (CH3). MS: 282 (90; M+-CH2Py); 157 (74); 134 (100).
Anal. calcd. for C25H29NO2: C, 78.96; H,. 7.78; N, 3.73; found: C,
79.09; H, 7.69; N, 3.66.
157.43; 149.14; 135.85; 123.92; 122.80; 120.27; 118.13; 54.02 (CH);
53.09 (CH); 45.71; 38.69; 35.71; 35.65; 35.45; 33.95; 32.83; 31.83;
29.75; 24.90; 21.08; 18.77(C-19); 17.40(C-18). MS: 361 (100; M+);
342 (42). Anal. calcd. for C25H31NO 0.5 H2O: C, 81.08; H, 8.65;
found: C, 81.64; H, 8.63.
2.2.6. (E)- and (Z)-3ˇ-Hydroxy-17-picolinylidene-
androst-5-ene-16-one (12 and 13)
Method A. Compound 10 or 11 (0.04 g, 0.1 mmol) and sodium
ethoxide in ethanol (0.1 M, 1.5 ml) were stirred at 25 ◦C for
4 h in the dark. The reaction mixture was poured into water
and extracted with CH2Cl2. Removal of dried solvents gave
a crude product 12 (0.05 g) or 13 (0.06 g). Crystallization from
the CH2Cl2–n-hexane 2:1 mixture afforded pure compound 12
(28 mg, 78%; mp 159–160 ◦C) or 13 (23 mg, 64%; mp 193–194 ◦C).
Method B. Compound 11 (0.63 g, 1.52 mmol) was dissolved
in methanol (32 ml), and KOH (0.34 g, 6.08 mmol) was added.
The reaction mixture was vigorously stirred at reflux in the
absence of light for 10 min. After pouring it into water (20 ml),
the precipitate of products 12 and 13 (0.53 g) was obtained.
The mixture of 12 and 13 was separated by preparative chro-
matography on a silica gel G (benzene–AcOEt, 7:3). Compound
12 was obtained in a yield of 29%, and compound 13 in a yield
of 34%.
Compound 12. IR (KBr): 3420, 1725, 1630, 1585, 1470, 1435,
1380, 1060, 780, 750. 1H NMR (CDCl3): 1.07 (3H, s, H-19); 1.31
(3H, s, H-18); 3.52 (1H, m, H-3); 5.37 (1H, m, H-6); 7.32 (1H, s, H-
20); 7.20, 7.40, 7.67 and 8.65 (4H, Py). 13C NMR (CDCl3): 207.47
(C-16); 154.53 (qC); 150.31 (qC); 149.24 (CH); 141.04 (qC); 135.79
(CH); 129.38 (CH); 126.16 (CH); 122.53 (CH); 120.93 (CH); 71.59 (C-
3); 50.36 (CH); 49.76 (CH); 43.88; 42.17; 37.87; 36.87; 36.62; 34.37;
31.74; 31.51; 30.86 (CH); 20.62; 19.37 (C-18); 17.00 (C-19). MS: 379
2.2.3. 3ˇ,17ˇ-Dihydroxy-17˛-picolyl-androst-5-ene
hydrobromide (6)
Compound 2 (0.10 g, 0.26 mmol) was dissolved in acetone
(2 ml). The solution was kept at 50 ◦C while conc. hydrobromic
acid (1 ml) in acetone (5 ml) was added. The solution was
cooled to room temperature and then left for 1 h. The removal
of acetone and crystallization of the residue from acetonitrile
afforded pure compound 6.
IR (KBr): 2800–2300, 1650, 1550, 1050, 780. 1H NMR (DMSO-
d6): 0.84 (3H, s, H-18); 0.98 (3H, s, H-19); 3.06 (2H, s, CH2-Py);
3.27 (1H, m, H-3); 5.29 (1H, m, H-6); 7.90 (1H, t, H-5ꢀ, Py); 8.0 (1H,
d, H-3ꢀ, Py); 8.48 (1H, t, H-4ꢀ, Py); 8.79 (1H, d, H-6ꢀ, Py). 13C NMR
(DMSO-d6): 155.49; 145.05 (C-5); 141.54; 129.60; 124.84; 120.58
(C-6); 82.69 (C-17); 70.28 (C-3); 50.1; 49.8; 46.8; 42.3; 37.2; 36.4;
33.1; 32.8; 31.5; 23.7; 20.7; 19.38 (C-19); 14.61 (C-18). MS: 462 (2,
M+); 288 (100); 272 (53); 255 (79); 203 (40); 91 (38). Anal. calcd. for
C25H36BrNO2 0.5 H2O: C, 63.69; H, 7.91; N, 2.97; found: C, 63.91;
H, 7.93; N, 3.04.
2.2.4. 3ˇ-Hydroxy-17-picolinylidene-androst-5-ene (8)
Compound
7 (1.0 g, 2.47 mmol) and sodium ethoxide in
ethanol (0.1 M, 25 ml) were stirred at room temperature for 4 h.
The reaction mixture was poured into water (50 ml). The pre-
cipitate was separated by filtration, washed with water and
dried. The crude product (0.91 g) was crystallized from MeOH,
affording pure compound 8 (0.84 g, 94%; mp 175 ◦C).
(4; M+); 377 (100); 360 (43); 144 (40). Anal. calcd. for C25
H31NO2
0.5 H2O: C, 77.72; H, 8.29; N, 3.63; found: C, 77.43; H, 8.89;
N, 3.63.
Compound 13. IR (KBr): 3350, 1720, 1630, 1590, 1060, 785.
1H NMR (CDCl3): 1.07 (6H, s, H-18 and H-19); 3.54 (1H, m, H-3);
5.36 (1H, m, H-6); 6.55 (1H, s, H-20); 7.19, 7.69, 8.32 and 8.59
(4H, m, Py). 13C NMR (CDCl3): 206.33 (C-16); 153.52 (qC); 150.56
(qC); 149.01 (CH); 141.13 (C-5); 135.96 (CH); 131.52 (CH); 125.11
(CH); 122.97 (CH); 120.75 (CH); 71.45 (C-3); 49.89 (CH); 49.09 (CH);
44.56; 42.17; 39.39; 36.98; 36.36; 36.67; 35.65; 31.49 (2C atom);
31.11 (CH); 20.79; 19.43 (C-18); 19.15 (C-19). MS: 379 (4; M+); 377
(100); 360 (45); 144 (41). Anal. calcd. for C25H31NO2 0.5 H2O: C,
77.72; H, 8.29; N, 3.63; found: C, 77.40; H, 8.42; N, 3.63.
IR (KBr): 3220, 3070, 1650, 1590, 1570, 1470, 1440, 1370, 1350,
1080, 780. 1H NMR (CDCl3): 0.91 (3H, s, H-18); 1.05 (3H, s, H-
19); 2.73 (1H, m, H-16a); 2.89 (1H, m, H-16b); 3.55 (1H, m, H-3);
5.36 (1H, m, H-6); 6.24 (1H, t, J20,16a = J20,16b = 2.4 Hz, H-20); 7.04,
7.30, 7.62, and 8.56 (4H, Py). 13C NMR (CDCl3): 160.4 (qC); 157.6
(qC); 149.2 (CH); 140.9 (qC); 135.8 (CH); 122.8 (CH); 121.4 (CH);
120.2 (CH); 118.0 (CH); 71.7 (C-3); 54.0 (CH); 50.4 (CH); 45.7 (C-
13); 42.3; 37.3; 36.6 (C-10); 35.8; 31.8; 31.7; 31.6; 29.8; 25.1; 21.1;
19.5 (CH3); 18.6 (CH3). MS: 363 (100; M+); 348 (46). Anal. calcd.
for C25H33NO: C, 82.60; H, 9.15; N, 3.85; found: C, 82.70; H, 9.12;
N, 3.90.
2.2.7. (E)-17-Picolinylidene-androst-4-ene-3,16-dione (14)
A mixture of compound 12 (0.23 g, 0.54 mmol) or 13 (0.29 g,
0.79 mmol), aluminum tert-butoxide (0.47 g, 1.91 mmol for 12
or 0.76 g, 3.15 mmol for 13) and cyclohexanone (25 ml for 12,
and 7.2 ml for 13) was refluxed for 35 h (for 12) or 28 h (for
13). After that, cyclohexanone was removed by steam distil-
lation. The residue was extracted with CH2Cl2. Removal of
dried solvents gave a residue which was separated by col-
umn chromatography (100 g or 78 g silica gel). Eluting with
toluene–AcOEt, 20:1, or toluene–AcOEt, 30:1, gave compound
14 (0.074 g, 36% from 12 or 0.102 g, 49% from 13) in the form of
light-yellow oil.
2.2.5. 17-Picolinylidene-androst-4-ene-3-one (9)
A mixture of compound 8 (0.51 g, 1.4 mmol), cyclohexanone
(10 ml) and aluminum isopropoxide (0.44 g, 2.15 mmol) was
vigorously stirred at reflux for 4 h. After that, cyclohex-
anone was removed by steam distillation and the residue
was extracted with CH2Cl2. Removal of dried solvents gave
a residue (0.67 g), which was crystallized from a CH2Cl2–n-
hexane mixture (0.45 g, 89%; mp 166–167 ◦C).
IR (KBr): 1673, 1652, 1614, 1581, 1559, 864, 782. UV (MeOH):
ꢀmax (ε) 200.9 (20,124); 246.5 (41,765); 285.2 (13,467). 1H NMR
(CDCl3): 0.95 (3H, s, H-18); 1.23 (3H, s, H-19); 2.60-3.00 (2H, m,
H-16a and H-16b); 5.75 (1H, s, H-4); 6.23 (1H, s, H-20); 7.04, 7.28,
7.61, 8.57 (4H, Py). 13C NMR (CDCl3): 199.49 (C-3); 171,12; 159.74;
IR (KBr): 1715, 1675, 1650, 1620, 1585. UV (MeOH): ꢀmax (ε)
201.5 (2,768); 235.7 (2,266); 295.2 (833); 355.1 (164). 1H NMR

34
steroids 7 2 ( 2 0 0 7 ) 31–40
(CDCl3): 1.23 (3H, s, H-19); 1.33 (3H, s, H-18); 5.75 (1H, s, H-4);
7.32 (1H, s, H-20); 7.21, 7.36, 7.68, 8.65 (4H, Py). 13C NMR (CDCl3):
208.73 (C-16); 199.85 (C-3); 163.12; 149,17; 141.62; 136.0; 129.45;
127.22; 126.29; 124.04; 122.68; 53.31; 49.59; 42.29; 37.53; 34.43;
31.51; 28.83; 27.44; 24.68; 22.66; 21.52; 20.53; 17.23 (CH3); 17.02
(CH3). MS: 377 (4; M+); 375 (100); 360 (70); 144 (49).
2.4.
Biological methods
2.4.1. Anti-aromatase activity
2.4.1.1. Chemicals. Antitestosterone-11-BSA serum No. 250,
antiprogesterone-11-BSA serum No. 337 were supplied by G.
D. Niswender (Colorado State University). Medium 199 (M199)
and Hanks were purchased from GIBCO Laboratories; bovine
serum albumine (BSA, Fraction V), collagenase (Type I), testos-
terone, dihydrotestosterone, pregnenolone, progesterone and
ꢄ4-androstenendione, Trypan blue, NADPH and ␤NAD+
(98%) were obtained from Sigma; [1,2,6,7-3H(N)]-testosterone,
2.3.
X-ray crystallographic study of 2
A single colorless crystal of the compound 2 selected for data
collection was mounted on a Bruker PLATFORM three-circle
goniometer equipped with SMART 1K CCD detector mounted
at a crystal-to-detector distance of 5.4 cm. The data were col-
lected using graphite monochromated Mo K␣ X-radiation and
frame widths of 0.3◦ in ω, with 10 s used to acquire each frame.
More than a hemisphere of three-dimensional data were col-
lected. Additional information regarding data collection and
structure refinement is given in Table 1. The data were reduced
using the Bruker program SAINT. A semiempirical absorption-
correction based upon the intensities of equivalent reflections
was applied (program XPREP), and the data were corrected
for Lorentz, polarization, and background effects. The Bruker
SHELXTL Version 5.1 system of programs was used for the
refinement of the crystal structure. The positions of all non-
H atoms were located by direct methods. The positions of
hydrogen atoms were found from the inspection of the differ-
ence Fourier maps. The high value of the Flack parameter (1.4
(8)) indicates that the absolute configuration cannot reliably
be resolved. The final refinement included atomic positional
and displacement parameters for all atoms. The non-H atoms
were refined anisotropically, while all H sites were refined with
isotropic displacement parameters. The refinement converged
at a final agreement index (R1) of 0.0388, calculated for 4591
unique observed reflections (|Fo| > 4ꢁF) and a goodness-of-fit
(S) of 0.792 (732 refined parameters).
[1,2,6,7-3H(N)]-estradiol
and
[1,2,6,7-3H(N)]-progesterone
were obtained from New England Nuclear; hCG (Pregnyl,
3000 IU/mg) was purchased from Organon Inc.; Pregnant
Mares Serum Gonadotrophin (PMSG) was obtained from
the Veterinary Institute Subotica (Serbia), Dextran T70 was
obtained from Pharmacia, Uppsala, Sweden, and charcoal
Norit A from Serva, Heidelberg.
2.4.1.2. Animals (male rats) and cell dispersion. Adult male
Wistar rats, random bred in our laboratory, were raised under
controlled environmental conditions (temperature 22 2 ◦C
and 14 h light:10 h dark), with food and water ad libitum. All
experiments were approved by the Local Ethical Committee
of the University of Novi Sad and were performed and con-
ducted in accordance with the principles and procedures of
the NIH Guide for Care and Use of Laboratory Animals. Adult
male rats (around 3 months old) were killed by decapitation,
testes quickly removed, decapsulated and enzymatically dis-
persed with collagenase (1.2 mg/ml) as described previously
[19]. Dye exclusion test (DET) using 0.2% trypan blue was used
to determine total cell counts and viable cell number.
2.4.1.3. Testosterone production in vitro. hCG stimulated,
progesterone (P) and ꢄ4-androstenedione (ꢄ4-A)-supported
androgen production in vitro was done as described previously
[20]. The desired concentrations of compounds was prepared
by evaporating the necessary amount of stock solution and
dissolving it in M199-BSA. At the end of incubation period
tubes were centrifuged for 10 min at 1500 g at 4 ◦C, and super-
natant was stored at −20 ◦C prior to measurement of testos-
terone by RIA. Since the antitestosterone-11-BSA serum No.
250 (from G. D. Niswender) shows high cross-reactivity with
dihydrotestosterone (DHT), and not with other steroids, assay
values for androgen production are referred to as “T + DHT”
concentrations.
Table 1 – Crystallographic data, data collection and
structure refinement for compound 2
Formula
Mw
(C25 H35 N O2)2·H2O
390.5
Crystal size (mm3)
Crystal description
T (K)
0.41 × 0.30 × 0.51
Colorless prism
300
Crystal system, space group
Orthorhombic, P212121
12.759(2); 90
12.860(2); 90
26.254(5); 90
4307.1(8); 4
1.204
1704.0; 0.08
2.3–28.5; 100
−16/16, −15/16, −28/33
9693
◦
˚
a (A); ˛ ( )
◦
˚
˚
b (A); ˇ ( )
◦
c (A); ꢂ ( )
2.4.1.4. Estimation of 3ˇHSD activity. The 3␤HSD activity was
estimated in suspension of adult rat Leydig cells as described
previously [21], following the conversion of pregnenolone to
progesterone.
˚
V (A); Z
Dc (g cm−3
)
F(000); ꢃ (Mo K␣) (mm−1
)
 range (◦); data completeness (%)
Range of h, k, l
2.4.1.5. Aromatase activity—animals (female rats), treatment,
and assays. Preparation of denucleated ovarian fraction from
PMSG pretreated rats and determination of aromatase activity
in ovarian homogenate was carried out as described previ-
ously [8].
For a preliminary assessment of potential anti-aromatase
activity of the synthesized compounds, the given compound
was added in a single concentration (1 ␮M) to the incubation
No. of unique diffractions
No. of observed diffractions
No. of parameters
4591
732
R, wR for observed diffractions (%)
R for all data (%)
3.9, 6.9
11.5
GOF for all data
0.792
(
/ꢁ)max
0.007
0.12, −0.11
−3
˚
Residual electron density (e A
)

steroids 7 2 ( 2 0 0 7 ) 31–40
35
Ca2+, 145 mM Cl− and 35 gl−1 gelatine polymers, pH 7.4), were
counted and resuspended in nutrient medium.
mixture containing 500 nM of testosterone as a substrate (sat-
urated concentration; the estimated Km for testosterone was
49.17 nM and Vmax 5.76 pmol/min/mg protein).
The screening assay for determining the IC50 values con-
tained 500 nM of testosterone and various concentrations of
tested compounds. The results were analyzed by the least-
squares (log concentration of tested compounds) and IC50 val-
ues were calculated.
2.4.4.3. Treatment of adherent neoplastic cells. Neoplastic HeLa
cells, or FemX cells, were seeded (2000 cells per well) into 96-
well microtiter plates, and 20 h later, after the cell adherence,
five different, double diluted, concentrations of investigated
extracts were added to the wells. Maximal final concentra-
tion extract applied to target cells was 100 ␮M. Only nutrient
medium was added to the cells in the control wells.
2.4.1.6. Statistics. The statistical significance was evaluated
by two-tailed non-parametric Mann–Whitney test.
2.4.4.4. Treatment of cells grown as suspension culture. PBMC
were seeded (150,000 cells per well) into nutrient medium or in
nutrient medium enriched with (5 ␮g/ml) phytohemagglutinin
(PHA) (Wellcome) in 96-well microtiter plates. Leukemia K562
cells were seeded (3000 cells per well) only in nutrient medium.
Two hours later, the investigated extracts were added to the
wells, in triplicate, to five final concentrations, except for the
control wells, where nutrient medium only was added to the
cells. Nutrient medium with corresponding concentrations of
compounds, but void of cells was used as the blank.
2.4.2. Brine shrimp test [15]
A teaspoon of lyophilized eggs of the brine shrimp A. salina was
added to 1 l of artificial sea water, and air was passed through
the suspension thermostated at 28 ◦C, under illumination. The
tested substances were dissolved in DMSO at the initial con-
centrations of 0.05, 0.1, 0.2, and 0.3 mg/ml. In a glass vial, 50 ␮l
of a DMSO solution was diluted to 5 ml and subsequently 1–2
drops of yeast extract solution (3 mg in 5 ml of distilled water)
and 10–20 hatched nauplii were added. For each concentra-
tion, two determinations were performed. The vials were left
at room temperature under illumination for 24 h, and after-
wards live and dead nauplii were counted. LC50 was defined
as the concentration of a drug that causes death of 50% nau-
plii. DMSO was inactive under applied conditions.
2.4.4.5. Determination of HeLa and FemX cell survival. Cell
survival was determined indirectly by measuring total cellu-
lar protein by the Kenacid Blue R (KBR) dye binding method
according to Clothier [23]. Briefly, after 72 h of continuous com-
pound action, medium was discarded and target cells were
washed twice with warm (37 ◦C) phosphate buffered saline
(PBS). Then target cells were fixed for 20 min with 150 ␮l of a
mixture of methanol and acetic acid (3:1) and stained 2–3 h
with 0.04% Coomassie Brilliant Blue R-250 in 25% ethanol
and 12% glacial acetic acid, washed, and the bound dye was
dissolved in desorbing solution (1 M potassium acetate, 70%
ethanol). Absorbance (A) at 570 nm was measured 2 h later. To
get cell survival (%), absorbance of a sample with cells grown
in the presence of various concentrations of the investigated
agent was divided with control optical density (the A of control
cells grown only in nutrient medium), and multiplied by 100. It
was implied that A of the blank was always subtracted from A
of the corresponding sample with target cells. IC50 concentra-
tion was defined as the concentration of an agent inhibiting
cell survival by 50%, compared with a vehicle-treated control.
2.4.3. Determination of antibacterial activity
Antibacterial activity was evaluated using agar diffusion
method [22]. Gram-positive bacteria S. aureus ATCC 25923 and
Gram-negative bacteria E. coli ATCC 35218 were cultivated in
nutrient agar on a Petri dish. Into wells of 8 mm diameter in
the agar 50 ␮l of acetonitrile solutions of the tested substances
(1 mg/ml) were added. After 24 h cultivation at 37 ◦C, diame-
ters of zones of inhibition were determined. Acetonitrile was
inactive under applied conditions.
2.4.4. Antiproliferative activity
Stock solutions of examined compounds were made in DMSO,
at concentration 10 mM, and then various dilutions were pre-
pared in nutrient medium. Nutrient medium was RPMI 1640
medium, supplemented with l-glutamine (3 mM), strepto-
mycin (100 ␮g/ml), and penicillin (100 IU/ml), 10% heat inac-
tivated (56 ◦C) fetal bovine serum (FBS) and 25 mM Hepes, and
was adjusted to pH 7.2 by bicarbonate solution.
2.4.4.6. Determination of PBMC and K562 cell survival. Deter-
mination of PBMC and K562 cell survival was performed
by MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide test, according to Ohno and Abe [24], 72 h upon addi-
tion of the drug, as described earlier. Briefly, 20 ␮l of MTT
solution (5 mg/ml PBS) were added to each well. Samples were
incubated for further 4 h at 37 ◦C in 5% CO2 and humidified
air atmosphere. Then, 100 ␮l of 10% SDS were added to the
wells. Absorbance at 570 nm was read the next day. To get cell
survival (%), absorbance of a sample with cells grown in the
presence of various concentrations of extract was divided with
control absorbance, Ac, (the absorbance of sample with cells
grown only in nutrient medium) × 100. (It was implied that A
of the blank was always subtracted from A of the correspond-
ing sample with target cells.) Concentration IC50 was defined
as the concentration of a drug which inhibits cell survival by
50%, compared with vehicle-treated control.
2.4.4.1. Cell culture. Human cervix carcinoma HeLa cells, and
human melanoma FemX cells were cultured as monolayers in
the nutrient medium. The cells were grown at 37 ◦C in 5% CO2
and humidified air atmosphere.
Human myelogenous leukemia K562 cells were grown as
suspension culture in the nutrient medium. The cells were
grown at 37 ◦C in 5% CO2 and humidified air atmosphere.
2.4.4.2. Preparation of peripheral blood mononuclear cells.
Peripheral blood mononuclear cells (PBMC) were separated
from whole heparinized blood of healthy volunteers by
LymphoprepTM (Oslo, Norway) gradient centrifugation. Inter-
face cells, washed three times with Haemaccel (aqueous
solution supplemented with 145 mM Na+, 5.1 mM K+, 6.2 mM

36
steroids 7 2 ( 2 0 0 7 ) 31–40
group and water molecule. The position of the 17␤-hydroxyl
group was also confirmed by the NOE experiment, by irradiat-
ing the H-18 proton of the angular methyl group at 0.82 ppm,
which produced a NOE effect on the hydroxyl group signal at
5.27 ppm.
With the aim of studying the intramolecular hydrogen
bond in some similar 17␤-hydroxy-17␣-picolyl-systems, as
well as of the effect of free electron pair of the pyridine N-atom
on biological activity, compound 6 was obtained by treating 2
with hydrobromic acid (Scheme 1).
The presence of the intramolecular hydrogen bond was
also proved by comparing the NMR spectra of compounds 2
and 6. Namely, the appearance of the –CH2Py signal as an AB
quartet in the molecule of 2 may reduce the possibility of free
rotation about the corresponding bonds because of the pres-
ence of the intramolecular hydrogen bond. However, in the
case of compound 6 this signal appeared as a singlet, which
is the result of the impossibility of forming hydrogen bond
because of the protonation of the N-atom and potential rota-
tion.
Fig. 1 – Asymmetric unit cell of compound 2.
The picolyl derivative
2 was transformed into 17-
picolinylidene derivative 7 in the boiling acetic anhydride
[25,26]. Compound 8 was obtained from the 3␤-acetoxy deriva-
tive 7 by reaction with sodium ethoxide in ethanol (Scheme 2).
The Oppenauer oxidation of compound 8 with cyclohexanone
and aluminum(III) tert-butoxide yielded 17-picolinylidene-4-
androstene-3-one (9).
3.
Results and discussion
3.1.
Chemistry
The first stage in the preparation of the new 17-picolyl
and 17-picolinylidene androstene derivatives consisted of the
addition of ␣-picolyllithium to the 17-oxo group of dehy-
droepiandrosterone (1), resulting in the 17␣-picolyl deriva-
tive 2 [25,26]. The Oppenauer oxidation of compound 2 with
cyclohexanone and aluminum(III) tert-butoxide afforded 17␤-
hydoxy-17␣-picolyl-4-androstene-3-one (3) in a yield of 40%
and 4-androstene-3,17-dione (4) in a yield of 6% (Scheme 1).
Dehydrogenation of compound 2 with DDQ in dioxane at
the boiling temperature for 24 h yielded 1,4,6-triene-3-oxo-
derivative 5.
The X-ray structural analysis of compound 2 confirmed
the ␣-orientation of the 17-picolyl group, as well as the pres-
ence of the intramolecular hydrogen bond between the C17
OH and N-atom of the pyridine ring (Fig. 1). Besides, there is
an intermolecular hydrogen bond between the same C17-OH
In our previous works [25,27] we described the preparation
of 17-picolinylidene-16-oxo derivatives 10 and 11 (Scheme 3).
In this work we investigated the reaction conditions for
the preparation of their 3␤-hydoxy derivatives 12 and 13,
bearing especially in mind that the ketones 10 and 11 are
prone to isomerization [25,28]. It was found that the hydrol-
ysis of 10 and 11 with potassium hydroxide in methanol
gives a mixture of isomeric ketones 12 and 13, irrespec-
tive of whether the reaction is performed at room temper-
ature or at the boiling temperature in the absence of light.
Thus the ketone 11 in the presence of potassium hydrox-
ide in boiling methanol, afforded in 10 min a mixture of
isomeric compounds 12 (29%) and 13 (34%), which were
separated by preparative chromatography. However, treat-
ment of the ketone 10 with sodium ethoxide in ethanol at
-
Scheme 1 – Reagents (i) ␣-PyCH2Li, THF; (ii) cyclohexanone, Al(tert-BuO)3; (iii) DDQ, dioxane; (iv) HBr, acetone.

steroids 7 2 ( 2 0 0 7 ) 31–40
37
Scheme 2 – Reagents (i) Ac2O; (ii) EtONa, EtOH; (iii) cyclohexanone, Al(tert-BuO)3.
room temperature (also in the absence of light) gave its 3␤-
hydroxy derivative 12 in a yield of 78%. The same procedure
involving the ketone 11 gave its derivative 13 in a yield of
64%.
Compound 12 was transformed to 4-en-3-one-derivative
14 by the Oppenauer oxidation with cyclohexanone and
aluminum(III) tert-butoxide at the boiling temperature of
the reaction mixture in the course of 35 h. Under the
same reaction conditions, the (Z)-isomer 13 afforded the
(E)-isomer 14.
Finally, to investigate the effect of the conjugated sys-
tem in the D-ring on the inhibitory action on steroidogenesis
enzymes, 17␤-picolyl derivative 15 was also synthesized by
the catalytic dehydrogenation of the ketones 10 and 11 in the
presence of Pd/C in ethanol as solvent [28].
3.2.
Biological properties
Screening assay procedures were used to assess the poten-
tial inhibitory effects of the synthesized compounds on
aromatase.
It is evident from Table 2 that 17-picolyl derivatives exhib-
ited a higher inhibitory activity against aromatase if the 3␤-
OH-5-ene system (2, −28.9%) was replaced with the 4-en-3-one
system (3, 51.5%), and especially when it was transformed to
1,4,6-trien-3-one system (5, 86.9%). Blocking of the nitrogen
electron pair (6, −59.2%) potentiated additionally aromatase
activity with respect to compound 2.
Table 2 – Inhibitory effects of 5-androstene derivatives
on the aromatase activity in the denucleated fraction of
ovaries from PMSG pretreated rats
Compound
Percent of inhibition vs. control
(50 ␮M)
Testosterone
Testosterone
0.05 ␮M
0.5 ␮M
2
3
5
6
7
8
9
−28.9 15.7
51.5 0.3**
86.9 0.4**
−59.2 11.5*
−18.2 16.3
−24.5 12.9
67.9 0.9**
−13.7 15.3
14.2 3.7
86.6 1.2**
90.0 0.4**
93.3 3.8**
−0.7 14.6
38.2 7.8**
63.9 1.0**
32.4 4.7**
10
11
12
13
14
15
62.2 3.7**
70.2 3.3**
74.5 3.4**
Purified denucleated fraction of ovaries (0.366 mg of proteins) from
PMSG pretreated female rats was incubated in the medium with
subsaturated (0.05 ␮M) or saturated (0.5 ␮M) solution of substrate
testosterone (T) and NADPH (1 mM) and in the absence (control)
or presence of different 5-androstene derivatives (50 ␮M) for mea-
suring aromatase activity. Estradiol level was determined by RIA.
Results are presented as percent of inhibition vs. control. Num-
bers represent mean SEM of 6 replicates. Significance: **p < 0.005;
*p < 0.05 vs. control (Mann–Whitney non-parametric test).
Scheme 3 – Reagents (i) EtONa, EtOH or KOH, MeOH; (ii)
cyclohexanone, Al(tert-BuO)3; (iii) H2/Pd, EtOH.

38
steroids 7 2 ( 2 0 0 7 ) 31–40
Table 3 – Inhibitory effects of 5-androstene derivatives on the activity of steroidogenic enzymes in postmitochondrial
testicular fraction of adult rat in vitro
Compound (50 ␮M)
Percent of inhibition vs. control
P450c17 activity
3␤HSD activity
17␤HSD activity
3
5
9
−25.3 11.2*
34.3 5.8**
5.2 7.1
−10.9 8.9
−11.6 13.5
14.1 5.0
37.6 1.1**
−18.0 11.8
18.9 4.8*
12
13
16.0 6.7
−0.2 4.2
31.6 3.7**
24.9 3.3**
28.3 9.9**
25.4 0.7**
Purified postmitochondrial testicular fraction (0.301 mg proteins) was incubated in the medium with pregnenolone (25 ␮M) and NAD+ (135 ␮M),
in the absence (control) or presence of different 5-androstene derivatives (50 ␮M), using progesterone production as an index of 3␤HSD activity.
Progesterone level was determined by RIA. To estimate P450c17 or 17␤HSD activity purified postmitochondrial testicular fraction (0.412 mg of
proteins) was incubated in the presence of progesterone or 4-androstenedione (10 ␮M; respectively) and NADPH (1 mM) and in the absence
(control) or presence of different 5-androstene derivatives (50 ␮M). Testosterone plus dihydrotestosterone (T + DHT) levels were determined by
RIA. Results are presented as percent of inhibition vs. control. Numbers represent mean SEM of 6 replicates. Significance: **p < 0.005; *p < 0.05
vs. control (Mann–Whitney non-parametric test).
The 17-picolinylidene derivatives are also more effective
Table 4 – Antibacterial activity of 5-androstene
derivatives
inhibitors of aromatase if the 3␤-OAc-5-ene system (7, −18.2%)
or 3␤-OH-5-ene system (8, −24.5%) is transformed to a 4-en-3-
Compound
Zone diameter (mm)
S. aureus
one system (9, 67.9%).
In the case of 17-picolinylidene-16-one derivative, the
replacement of the 3␤-OAc group in compounds 10 (−13.7%)
and 11 (14.2%) with the 3␤-OH group increased the inhibitory
activity against aromatase (12, 86.6%; 13, 90.0%), and this
increase was especially pronounced by introducing the 4-
en-3-one system (14, 93.3%). The break of conjugation in the
D-ring did not influence essentially the inhibitory activity
(15, −0.7%). Generally, 17-picolinylidene-16-one derivatives
showed a more pronounced inhibitory activity than 17-picolyl
and 17-picolinylidene derivatives. The compounds that
showed inhibitory activity against aromatase (3, 5, 9, 12 and
13) were tested against other steroidogenic enzymes 3␤HSD,
P450c17, and 17␤HSD (Table 3). A comparison of the results
presented in Tables 2 and 3 shows that the highest selectivity
against aromatase exhibited compounds 9 and 13.
Bearing in mind the results obtained for the
inhibitory activity of 17-picolyl, 17-picolinylidene and 17-
picolinylidene16-one derivatives of 5-androstene against
above steroidogenesis enzymes, further investigations of
antibacterial and cytotoxic activities of the synthesized
compounds were carried out.
A weak activity against Gram-positive bacteria Staphyloco-
cus aureus exhibited three derivatives, 10, 11 and 15, a prereq-
uisite for the activity being the presence of the carbonyl C-16
and 3-acetoxy substituent. Similarly, a weak activity against
Gram-negative bacteria E. coli showed the derivatives 3, 7, 8
and 15 (Table 4). However, no clear structure-activity rela-
tionship pattern was observed for E. coli. As for S. aureus, a
prerequisite for the activity is the 16-keto group. It is inter-
esting to note that, like 3␤-acetoxy-17␤-(L-prolyl)amino-5␣-
androstane, for which activity against Gram-positive bacteria
has been established [17], active compounds from this study
also have a basic nitrogen atom in the side chain. There-
fore, further investigation of such compounds in this direction
might be justified.
E. coli
2
3
6
–
10
–
–
–
–
–
–
7
8
9
10
12
–
–
10
11
13
15
–
–
–
14
10
10
–
10
considered to show a good correlation with cytotoxicity tests
[15]. Two derivatives showed a moderate activity, namely 13
and 15 (Table 5), both with 16-keto substituent, but, since 10
and 11 are practically inactive, other structural requirements
must obviously be satisfied. A. salina does not express aro-
matase [16], so that the activity in this test is not related to
aromatase inhibition.
The compounds were also tested against three tumor cell
lines, human cervical carcinoma HeLa cells, human malignant
melanoma FemX cell line, and human myelogenous leukemia
Table 5 – Brine shrimp test results of 5-androstene
derivatives
Compound
LC50 (␮M)
2
3
6
8
9
10
11
13
15
92.4
85.4
99.3
79.1
89.3
88.9
91.4
13.7
11.6
In the Brine shrimp test, general toxicity against nauplii of
the brine shrimp A. salina was evaluated. This simple test is

steroids 7 2 ( 2 0 0 7 ) 31–40
39
Table 6 – Cytotoxicity of the tested compounds
Compound
IC50 (␮M)
K562
HeLa
Fem-X
PBMC-PHA
PBMC + PHA
2
3
6
8
9
11
13
15
83.5 7.8
78.4 1.1
96.8 0.4
>100
>100
8.3 0.3
58.6 10.5
73.2 6.3
72.2 2.9
93.8 3.2
27.5 0.4
7.5 0.7
5.1 0.7
8.2 1.3
67.1 6.1
63.9 7.2
59.7 2.7
52.3 1.7
42.8 2.1
6.7 0.4
5.3 0.6
4.0 0.4
>100
>100
>100
NDa
ND
ND
16.9 1.1
9.4 0.4
>100
>100
>100
ND
ND
ND
6.3 1.7
9.5 1.5
10.0 1.4
7.5 0.6
a
ND, not determined.
r e f e r e n c e s
K562 cells (Table 6). Three compounds, namely 11, 13 and 15,
showed strong activity against all three cell lines, with IC50
values in the range of 4–10 ␮M. A prerequisite for the activity
is the presence of carbonyl at C-16, while the substituent at C-
3 could be either acetoxy or hydroxy group, and the C17–C20
bond either single or double. The antitumor activity of these
compounds is probably not related to aromatase inhibition,
since there is no correlation between aromatase inhibition
and cytotoxicity, and neither there is a difference in activ-
ity between cell lines that express aromatase [13] and those
which are devoid of it [12].
[1] For recent reviews of aromatase inhibitors see:
(a) Santen RJ. Inhibition of aromatase insights from recent
studies. Steroids 2003;68:559–67;
(b) Johnston SRD, Dowsett M. Aromatase inhibitors for breast
cancer: lessons from the laboratory. Nature Rev Cancer
2003;3:821–31;
(c) Brueggemeier RW, Hackett JC, Diaz-Cruz ES. Aromatase
inhibitors in the treatment of breast. Cancer Endocrin Rev
2005;26:331–45;
(d) Freedman OC, Verma S, Clemons MJ. Pre-menopausal breast
cancer and aromatase inhibitors: treating a new generation of
women. Breast Cancer Res Treat 2006;99:241–7;
(e) Bria E, Ciccarese M, Giannarelli D, et al. Early switch with
aromatase inhibitors as adjuvant hormonal therapy for
postmenopausal breast cancer: pooled-analysis of 8794 patients.
Cancer Treat Rev 2006;32:325–32.
[2] Bonneterre JB. Anastrozole is superior to tamoxifen as
first-line therapy in hormone receptor positive advanced
breast carcinoma. Cancer 2001;92:2247–58.
[3] Ellis Mj, Coop A, Singh B, Mauriac L, Llombert-Cussac A,
Janicke F. Letrozole is more effective neoadjuvant endocrine
therapy than tamoxifen for ErbB-1- and/or ErbB-2-positive,
estrogen receptor-positive primary breast cancer: evidence
from a phase III randomized trial. J Clin Oncol
Cytotoxicity was also tested against normal human PBMC.
The derivatives 2, 3 and 6, which showed weak activity against
all three tumor cell lines, appeared to be completely nontoxic
to PBMC. Unfortunately, active derivatives 13 and 15 showed
toxicity against PBMC. Nevertheless, the IC50 of 13 against the
tumor cell lines was approximately three times lower than
against PBMC, whereas the IC50 of 15 against K562 leukemia
cell line was more than two times lower than against PBMC. So,
the derivatives displayed certain amount of selectivity, thus
justifying further study
A comparison of cytotoxicity of the three compounds
(11, 13 and 15) that showed highest activity (Table 6) with
their activity obtained in Brine shrimp test (Table 5) and
anti-aromatase activity (Table 2) shows that compound 13
exhibited signficant inhibition of aromatase (90.0%), moder-
ate activity in Brine shrimp test (LC50 13.7 ␮M) and, at the
same time, very strong activity against all three cell lines:
HeLa (IC50 6.3 ␮M), FemX (IC50 5.1 ␮M), and K562 (IC50 5.3 ␮M).
On the other hand, compound 3, although showing a moder-
ate inhibitory activity against aromatase (51.5%), is nontoxic
against PBMC, and compound 9, which shows slightly higher
inhibitory activity against aromatase (67.9%) compared to 3,
exhibits certain selectivity with respect to melanoma cells.
These compounds (3, 9 and 13) could serve as the basis for
potential aromatase inhibitors and active agents against some
tumor cell lines.
2001;19:3808–16.
[4] Paridaens R, Dirix L, Lohrisch C, et al. Promising activity and
safety of exemestane as first-line hormonal therapy in
metastatic breast cancer patients: final results of an EORTC
randomised phase II trial. Breast Cancer Res Treat
2000;64:52.
[5] Goss PE, Strasser K. Aromatase inhibitors in the treatment
and prevention of breast cancer. J Clin Oncol 2001;19:881–94.
[6] Houssami N, Cuzick J, Dixon JM. The prevention, detection,
and management of breast cancer. MJA 2006;184:230–4.
[7] McCague R, Rowlands MG. Conformational Analysis of the
Aromatase Inhibitor 3-Ethyl-3(4-pyridyl)piperidine-2.6-dione
(Rogletimide) and Discovery of Potent 5-Alkyl Derivatives. J
Med Chem 1992;35:3699–704.
[8] Andric´ S, Kostic´ T, Sakacˇ M, Medic´-Mijacˇevic´ LJ, Penov Gasˇi
K, Kovacˇevic´ R. Biological characterization of some novel
5-androstene derivatives as potential antiandrogens. In:
Proceedings for Natural Sciences, vol. 94. 1998.
Acknowledgments
p. 43–51.
[9] Penov Gasˇi K, Stankovic´ S, Csanadi J, Ðurendic´ E, Sakacˇ M,
Medic´-Mijacˇevic´ LJ, Arcson O, Stojanovic´ S, Andric´ S, Molnar
Gabor D, Kovacˇevic´ R. New D-Modified androstane
derivatives as aromatase inhibitors. Steroids 2001;66:645–53.
We would like to thank the Ministry of Science and Envi-
ronmental Protection of the Republic of Serbia for financial
support (Grant No. 142052).

40
steroids 7 2 ( 2 0 0 7 ) 31–40
[10] Penov Gasˇi K, Stojanovic´ S, Sakacˇ M, Popsavin M, Jovanovic´
[20] Kovacˇevic´ R, Vojinovic´-Miloradov M, Teodorovic´ I, Andric´ S.
Effect of PCBs on androgen production by suspension of
adult rat Leydig cells in vitro. J Steroid Biochem Mol Biol
1995;52:595–7.
ˇ
Santa S, Stankovic´ S, Klisuric´ O, Andric´ N, Kovacˇevic´ R.
Synthesis and anti-aromatase activity of some new steroidal
D-lactones. Steroids 2005;70:47–53.
[11] Penov Gasˇi K, Miljkovic´ D, Medic´-Mijacˇevic´ LJ, Ðurendic´ E,
Stojanovic´ S, Sakacˇ M, Ðurendic´ M, Stankovic´ S, Lazar D,
Andric´ S, Kovacˇevic´ R. Synthesis, X-ray crystal structures
and biological activity of 16-amino-17-substituted-D-homo
steroid derivatives. Steroids 2003;68:667–76.
[21] Maric´ D, Kostic´ T, Kovacˇevic´ R. Effects of acute and chronic
immobilization stress on rat Leydig cell steroidogenesis. J
Steroid Biochem Mol Biol 1996;58:351–5.
[22] Ketchum PA. Microbiology: concepts and applications. New
York: John Wiley and Sons; 1988. p. 512–3.
[12] Yamada K, Harada N, Honda S, Takagi Y. Regulation of
placenta-specific expression of the aromatase cytochrome
P-450 gene. J Biol Chem 1995;270:25064–9.
[23] Clothier RH. The FRAME cytotoxicity test. In: O’Hare S,
Atterwill CK, editors. Methods in molecular biology, vol. 43.
Totowa, NJ: Humana Press; 1995. p. 109–18.
[13] Santen RJ, Santner SJ, Harvey HA, Lipton A, Simmonds M,
Feil PD, Manders E, Davis TS. Marked heterogeneity of
aromatase activity in human malignant melanoma tissue.
Eur J Cancer Clin Oncol 1988;24:1811–6.
[24] 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.
[14] Fiorelli G, Frediani U, Martineti V, Franchi A, Gori F,
Franceschelli F, Tanini A, Serio M, Brandi ML. Aromatase
expression and activity in the human leukaemic cell line
FLG 29.1. J Steroid Biochem Mol Biol 1998;66:105–12.
[15] Meyer BN, Ferrigni NR, Putnam JE, Jacobsen LB, Nichols DE,
McLaughlin JL. Brine shrimp: A convenient general bioassay
for active plant constituents. Planta Med 1982;45:31–4.
[16] Swevers L, Lambert JGD, DeLoof A. Metabolism of
vertebrate-type steroids by tissues of three crustacean
species. Comp Biochem Physiol 1991;99B:35–41.
[25] Miljkovic´ D, Gasˇi K, Kindjer M, Stankovic´ S, Ribar B. Partial
syntheses of 21,27-bisnordemissidine from epiandrosterone
and dehydroepiandrosterone acetates. Crystal and
molecular structure of 21,27-bisnordemissidine
hydrobromide. Croat Chem Acta 1985;58:721–36.
[26] Miljkovic´ D, Gasˇi K. A novel synthetic approach to
3␤-acetoxy-17-picolinylidene-5-androstene-16-one. Bull Soc
Chim Beograd 1982;47:173–81.
[27] Miljkovic´ D, Gasˇi K.
3␤-Acetoxy-17-picolinylidene-5-androstene-16-one—the
key intermediate in 21,27-bisnorsolanidine synthesis. Bull
Soc Chim Beograd 1981;46:263–8.
[17] Pettit RK, Cage GD, Pettit GR, Liebman JA. Antimicrobial and
cancer cell growth inhibitory activities of
3␤-acetoxy-17␤-(l-prolyl)amino-5␣-androstane in vitro. Int J
Antimicrob Agents 2000;15:299–304.
[18] Fieser LF, Fieser M. Steroids. New York: Reinhold Publishing
Co.; 1967. p. 512.
[19] Kovacˇevic´ R, Saracˇ M. Bromocriptine-induced inhibition of
hydroxylase/lyase activity of adult rat Leydig cells. J Steroid
Biochem Mol Biol 1993;46:841–5.
[28] Medic´ Mijacˇevic´ LJ, Miljkovic´ D, Sakacˇ M, Gasˇi K. Study on
catalytic hydrogenation of isomeric (Z)- and
(E)-3␤-acetoxy-17-picolinylidene-5-androstene-16-ones.
Proceedings for Natural Sciences, Matica Srpska
1993;84:69–72.