Synthesis, structure, and screening of estrogenic and antiestrogenic activity of new 3,17-substituted-16,17-seco-estratriene derivatives

Article abstract of DOI:10.1016/S0045-2068(03)00101-9

The starting compound for synthesis of new 16,17-seco-estratriene derivatives was 3-benzyloxy-17-hydroxy-16,17-secoestra-1,3,5(10)-triene-16- nitrile (1b), obtained from estrone in several synthetic steps. 17-Tosyl, -chloro-, bromo-, and -iodo- derivatives 2b, 4b, 5b, and 6b were prepared directly from secocyanoalcohol 1b, while the 17-fluoro-derivative 3b was obtained from tosylate 2b in the reaction with tetrabutyl ammonium fluoride. The corresponding 3-hydroxy derivatives of these compounds were produced by action of hydrogen in presence of Pd/C, except the 3-hydroxy-17-iodo derivative 6a, which was obtained from 3-hydroxy-17-tosyloxy derivative 2a. All the newly synthesized compounds in biological tests on experimental animals exhibited an almost total loss of estrogenic activity, while most of them even prevented the action of endogenous estrogens. On the other hand, most of them, except compounds 3a and 6b, partially hindered the action of estradiol benzoate, behaving as moderate antagonists.

Full text of DOI:10.1016/S0045-2068(03)00101-9

BIOORGANIC
CHEMISTRY
Bioorganic Chemistry 31 (2003) 475–484
www.elsevier.com/locate/bioorg
Synthesis, structure, and screening
of estrogenic and antiestrogenic activity of new
3,17-substituted-16,17-seco-estratriene derivatives
a,
Suzana Jovanovic-Santa, Julijana Petrovic, Silvana Andric,
ꢀ a
ꢀ b
*
ꢁ
ꢀ
ꢁ
ꢀ b ꢁ a
ꢀ a
Radmila Kovacevic, Evgenija Ðurendic, Marija Sakac,
ꢁ
c
Dusan Lazar, and Slobodanka Stankovic
ꢀc
a
Department of Chemistry, Faculty of Science, University of Novi Sad, Trg Dositeja Obradovicꢀa 3,
Serbia and Montenegro, 21000 Novi Sad, Yugoslavia
b
Department of Biology, Faculty of Science, University of Novi Sad, Trg Dositeja Obradovicꢀa 2,
Serbia and Montenegro, 21000 Novi Sad, Yugoslavia
c
Department of Physics, Faculty of Science, University of Novi Sad, Trg Dositeja Obradovicꢀa 4,
Serbia and Montenegro, 21000 Novi Sad, Yugoslavia
Received 3 June 2003
Abstract
The starting compound for synthesis of new 16,17-seco-estratriene derivatives was 3-ben-
zyloxy-17-hydroxy-16,17-secoestra-1,3,5(10)-triene-16-nitrile (1b), obtained from estrone in
several synthetic steps. 17-Tosyl, -chloro-, bromo-, and -iodo- derivatives 2b, 4b, 5b, and 6b
were prepared directly from secocyanoalcohol 1b, while the 17-fluoro-derivative 3b was ob-
tained from tosylate 2b in the reaction with tetrabutyl ammonium fluoride. The corresponding
3-hydroxy derivatives of these compounds were produced by action of hydrogen in presence of
Pd/C, except the 3-hydroxy-17-iodo derivative 6a, which was obtained from 3-hydroxy-17-to-
syloxy derivative 2a. All the newly synthesized compounds in biological tests on experimental
animals exhibited an almost total loss of estrogenic activity, while most of them even pre-
vented the action of endogenous estrogens. On the other hand, most of them, except com-
pounds 3a and 6b, partially hindered the action of estradiol benzoate, behaving as
moderate antagonists.
Ó 2003 Published by Elsevier Inc.
* Corresponding author. Fax: +381-21-454-065.
E-mail address: santas@ih.ns.ac.yu (S. Jovanovic-Santa).
ꢁ
ꢀ
0045-2068/$ - see front matter Ó 2003 Published by Elsevier Inc.
doi:10.1016/S0045-2068(03)00101-9

ꢁ
476
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
Keywords: Seco-steroids;
D
-Seco-estratriene derivatives; 3,17-Substituted-16,17-seco-estratriene deriva-
tives; Synthesis; biological activity; X-ray studies
1. Introduction
A broader project, directed towards preparation of potential antiestrogens, re-
sulted in the synthesis of series of new 16,17-seco-estrone derivatives [1–5]. Among
them the behaviour of 3-benzyloxy-17-hydroxy-16,17-secoestra-1,3,5(10)-triene-16-
nitrile (1b, Scheme 1), and its corresponding 3-hydroxy derivative (1a) were of special
interest. Namely, compound 1b showed a very low estrogenic activity, exhibiting a
significant antihormonal property. On the other hand, its 3-hydroxy equivalent 1a
showed a complete loss of estrogenic activity, but a less antiestrogenic effect [5,6].
These facts prompted us to further chemical transformations of 1, in order to study
the influence of the nature of functional groups at C-3 and C-17 on biological action
of the new compounds. Accordingly, the aim of this paper was the synthesis of a se-
ries of new 3-benzyloxy-16,17-secoestratriene derivatives with different substituents
at C-17, as well as their 3-hydroxy analogs, and testing their biological activity on
experimental animals.
Scheme 1.

ꢁ
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
477
2. Materials and methods
2.1. Synthesis and characterization of the newly synthesized compounds
2.1.1. General
€
Melting points were determined in open capillary tubes on a Buchi SMP appara-
tus and the values are uncorrected. Infrared spectra (m in cmꢀ1) were recorded in KBr
pellets or as films on a Perkin–Elmer M457 or Carl Zeiss Specord 75 spectrophotom-
eter. NMR-spectra were taken on a Bruker AC 250E spectrometer operating at
250 Hz (proton) and 62.9 Hz (carbon), using standard Bruker software. The signals
are reported in parts per million downfield from a tetramethylsilane internal stan-
dard (d 0.00); symbols s, bs, d, dd, q, and m denote singlet, broad singlet, doublet,
double doublet, quartet, and multiplet, respectively. Mass spectra were recorded on a
Finnigan-Math 8230 instrument, using electron impact (70 eV) or chemical ioniza-
tion (iso-butane) techniques; the first number denotes m/z value, and the ion abun-
dances are given in parentheses. The diffraction data of the compound 1a were
collected using Siemens P4 diffractometer and the structure was solved by the molec-
ular-mechanic calculations (MMC), using PCMODEL (Serena Software, 1989).
Crystallographic data were deposited at the Cambridge Crystallographic Data Cen-
tre (CCDC 205270). The starting compound, 1b, was synthesized by known proce-
dure [2], as well as compound 1a, whose crystal structure we present. All reagents
used were of analytical grade commercially available substances.
2.1.2. 3-Benzyloxy-17-tosyloxy-16,17-secoestra-1,3,5(10)-triene-16-nitrile (2b)
To the solution of 3-benzyloxy-17-hydroxy-16,17-secoestra-1,3,5(10)-triene-16-ni-
trile (1b, 1 g, 2.67 mmol) in absolute pyridine (10 mL) p-toluenesulphonyl chloride
(0.8 g, 4.2 mmol) was added and the mixture was kept at room temperature for
50 h. After completion of the reaction, the resulting solution was poured into dilute
HCl (1:1, 100 mL), crude product was filtered off, washed with water until neutral,
and air-dried. Column chromatography on silica gel (100 g, dichloromethane–n-hex-
ane 9:1) of the crude p-toluenesulphonate ester 2b afforded 0.84 g (84%) of pure
3-benzyloxy-17-tosyloxy-16,17-secoestra-1,3,5(10)-triene-16-nitrile (2b) in the form
of an amorphous mass.
IR: 3060, 2980, 2300, 1420, 1375, 1260, 1180, 890, 750.
1H NMR spectrum (CDCl3): 0.96 (s, 3H, CH3, H-18); 2.50 (s, 3H, CH3 from Ts);
3.67 (d, 1H, Ha–C-17); 3.92 (d, 1H, Hb–C-17, Jgem ¼ 10.1 Hz); 5.08 (s, 2H, O–CH2–
C6H5); 6.75–7.88 (group of signals, 12H, aromatic protons).
13C NMR spectrum (CDCl3): 15.36 (C-15); 15.91 (C-18); 22.57 (CH3 from Ts);
76.26 (O–CH2–C6H5); 118.77 (CꢁN); 156.92 (C-3).
Mass spectrum: 587 (8; (M+i-Bu)þ); 586 (13; (M+i-Bu-1)þ); 530 (4; (M+1)þ); 415
(47); 358 (54; (M+i-Bu-OTs)þ); 324 (100); 268(2).
2.1.3. 3-Benzyloxy-17-fluoro-16,17-secoestra-1,3,5(10)-triene-16-nitrile (3b)
3-Benzyloxy-17-tosyloxy-16,17-secoestra-1,3,5(10)-triene-16-nitrile (2b, 1 g, 1.90
mmol) and tetrabutyl ammonium fluoride trihydrate (3 g, 9.52 mmol) were dissolved

ꢁ
478
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
in ethylmethyl ketone (30 mL) and the reaction mixture was refluxed during 12 h.
Reaction mixture was then poured into ice-cold water (100 mL) and the formed
emulsion was extracted with ether (3 ꢂ 30 mL). The collected extracts were dried
over anhydrous sodium sulphate and evaporated to dryness. The dark-brown oily
product was purified by column chromatography on silica gel (100 g, dichlorome-
thane–n-hexane 9:1), whereby 0.36 g (70.4%) of pure 3-benzyloxy-17-fluoro-16,
17-secoestra-1,3,5(10)-triene-16-nitrile (3b, mp 127–129 °C) was obtained, and 0.29 g
(29%) of p-toluenesulphonate ester 2b was recovered.
IR: 2920, 1600, 1500, 1450, 1375, 1310, 1180, 940, 670, 550.
1H NMR spectrum (CDCl3): 1.05 (s, 3H, CH3, H-18); 2.53 (dd, 1H, Ha–C-15,
Jgem ¼ 21.01 Hz, J15a;14 ¼ 3.25 Hz); 2.72 (dd, 1H, Hb–C-15, J15b;14 ¼ 5.0 Hz); 4.16
(dd, 1H, Ha–C-17, Jgem ¼ 9.6 Hz, J17a;F ¼ 39.66 Hz); 4.36 (dd, 1H, Hb–C-17,
J17b;F ¼ 40.79 Hz); 5.10 (s, 2H, O–CH2–C6H5); 6.81–7.52 (group of signals, 8H, aro-
matic protons).
13C NMR spectrum (CDCl3): 14.67 (d, C-15, JC;F ¼ 6.29 Hz); 15.78 (d, C-18,
JC;F ¼ 2.96 Hz); 34.99 (d, C-12, JC;F ¼ 3.90 Hz); 38.34 (d, C-13, JC;F ¼ 15.91 Hz);
41.91 (d, C-14, JC;F ¼ 1.76 Hz); 69.82 (O–CH2–C6H5); 90.68 (d, CH2–F,
JC;F ¼ 176.37 Hz); 119.19 (CꢁN); 156.87 (C-3).
Mass spectrum: 434 (42; (M+i-Bu-1)þ); 378 (100; (M+1)þ).
Anal. Calcd for C25H28FNO: C, 79.54; H, 7.47; N, 3.71. Found: C, 79.26; H, 7.52;
N, 3.59.
2.1.4. 3-Benzyloxy-17-chloro-16,17-secoestra-1,3,5(10)-triene-16-nitrile (4b) and
3-benzyloxy-17-bromo-16,17-secoestra-1,3,5(10)-triene-16-nitrile (5b)
3-Benzyloxy-17-hydroxy-16,17-secoestra-1,3,5(10)-triene-16-nitrile (1b, 1 g, 2.67
mmol) was dissolved in absolute pyridine (30 mL) at room temperature. The solu-
tion was cooled to 0 °C and triphenylphosphine (1.72 g, 6.6 mmol) and carbon
tetrachloride or tetrabromide (3.25 mmol) were added in several portions. The re-
action mixture was heated for 2.5 h at 60 °C, then cooled and methanol (10 mL)
was added, to eliminate the excess of reagents, and poured into icy 2 M HCl
(100 mL). The emulsion was extracted with dichloromethane (2 ꢂ 50 mL), collected
extracts were washed with water, dried over anhydrous sodium sulphate and evap-
orated to dryness. The oily crude products were purified by column chromatogra-
phy on silica gel (100 g, toluene–ethyl acetate 95:5), whereby pure compounds 4b,
i.e. 5b were obtained.
3-Benzyloxy-17-chloro-16,17-secoestra-1,3,5(10)-triene-16-nitrile (4b). Yield: 53.6%,
mp 148.5–149.5 °C.
IR: 2925–2860, 2250, 1605, 1500, 1240, 1020, 740.
1H NMR spectrum (CDCl3): 1.09 (s, 3H, CH3, H-18); 2.35 (m, 3H, Ha–C-15 and
2H from the skeleton); 2.61 (dd, 1H, Hb–C-15, Jgem ¼ 17.9 Hz, J15b;14 ¼ 5.2 Hz); 3.42
(d, 1H, Ha–C-17, Jgem ¼ 10.95 Hz); 3.59 (d, 1H, Hb–C-17); 5.08 (s, 2H, O–CH2–
C6H5); 6.78–7.52 (group of signals, 8H, aromatic protons).
13C NMR spectrum (CDCl3): 15.18 (C-15); 18.00 (C-18); 69.85 (O–CH2–C6H5);
118.93 (CN); 156.90 (C-3).

ꢁ
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
479
Mass spectrum: 452 (37; (M+i-Bu)þ); 451 (52; (M+i-Bu-1)þ); 450 (100; (M+i-Bu-
2)þ); 394 (74; Mþ); 360 (17).
Anal. Calcd for C25H28ClNO: C, 76.34; H, 7.12; N, 3.56. Found: C, 75.88; H, 7.18;
N, 4.01.
3-Benzyloxy-17-bromo-16,17-secoestra-1,3,5(10)-triene-16-nitrile (5b). Yield:
84%, mp 157–157.5 °C.
IR: 3030, 2930, 2250, 1605, 1503, 1240, 1015.
1H NMR spectrum (CDCl3): 1.13 (s, 3H, CH3, H-18); 2.38 (m, 3H, Ha–C-15 and
2H from the skeleton); 2.62 (dd, 1H, Hb–C-15, Jgem ¼ 17.8 Hz, J15b;14 ¼ 5.2 Hz); 3.35
(d, 1H, Ha–C-17, Jgem ¼ 10.9 Hz); 3.45 (d, 1H, Hb–C-17); 5.06 (s, 2H, O–CH2–C6H5);
6.75–7.44 (group of signals, 8H, aromatic protons).
13C NMR spectrum (CDCl3): 15.15 (C-15); 18.29 (C-18); 69.83 (O–CH2–C6H5);
118.88 (CN); 156.87 (C-3).
Mass spectrum: 438 (16; Mþ); 437 (18); 91 (100).
Anal. Calcd for C25H28BrNO: C, 68.50 H, 6.39, N, 3.19. Found: C, 68.39; H, 6.53;
N, 3.23.
2.1.5. 3-Benzyloxy-17-iodo-16,17-secoestra-1,3,5(10)-triene-16-nitrile (6b)
The mixture of compound 1b (1g, 2.67 mmol), triphenylphosphine (2 g, 7.5 mmol),
imidazole (0.53 g, 0.8 mmol), and iodine (1.32 g, 5.2 mmol) in toluene (70 mL) was
refluxed during 3 h. After cooling, methanol (10 mL) was added, and the resulting
solution was poured into water (50 mL). The organic layer was separated, washed
with water, dried over anhydrous sodium sulphate, and evaporated to dryness. Col-
umn chromatography of the dark-brown oily crude product on silica gel (100 g,
dichloromethane–n-hexane 9:1), yielded the pure 3-benzyloxy-17-iodo-16,17-secoes-
tra-1,3,5(10)-triene-16-nitrile (6b, 1.24 g, 96%, mp 139–140 °C).
IR: 3040, 2920, 2250, 1605, 1500, 1240, 1020.
1H NMR spectrum (CDCl3): 1.17 (s, 3H, CH3, H-18); 2.35 (m, 3H, Ha–C-15 and
2H from the skeleton); 2.58 (dd, 1H, Hb–C-15, Jgem ¼ 17.8 Hz, J15b;14 ¼ 5.1 Hz); 3.30
(s, 1H, CH2–I); 5.07 (s, 2H, O–CH2–C6H5); 6.78–7.44 (group of signals, 8H, aro-
matic protons).
13C NMR spectrum (CDCl3): 15.21 (C-15); 18.11 (C-18); 23.96 (CH2–I); 69.89 (O–
CH2–C6H5); 118.80 (CꢁN); 156.93 (C-3).
Mass spectrum: 543 (100; (M+i-Bu)þ); 542 (48; (M+i-Bu-1)þ); 487 (22; (M+1)þ).
Anal. Calcd for C25H28INO: C, 61.86 H, 5.82, N, 2.89. Found: C, 61.42; H, 5.95;
N, 3.31.
2.1.6. General procedure for deprotection of the 3-hydroxy function of compounds
2b–5b
To the solution of the corresponding benzyl ether 2b–5b (1 mmol) in ethyl acetate
(10 mL) or in the dichloromethane–methanol mixture (2:1, 10 mL), 10% Pd/C (10%)
was added. The suspensions were stirred at room temperature for 12–24 h in an at-
mosphere of hydrogen. After the removal of the catalyst, the solvent was evaporated
to dryness and the crude products were purified by column chromatography (25 g,

ꢁ
480
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
toluene–ethyl acetate, 4:1), yielding the appropriate 3-hydroxy-17-substituted deriv-
atives (2a–5a).
3-Hydroxy-17-iodide-16,17-secoestra-1,3,5(10)-triene-16-nitrile (6a) could not be
obtained by this procedure. It was synthesized from 3-hydroxy-17-tosyloxy-16,17-
secoestra-1,3,5(10)-triene-16-nitrile (2a), in the reaction with tetrabutyl ammonium
iodide in ethylmethyl ketone.
2.2. Biological tests
Immature Wistar strain female rats (21–23 days old) were randomly divided into
groups of six to eight animals each. The animals were treated by subcutaneous injec-
tion once a day for 3 days with 0.1 mL of a solution of the test compound in olive oil,
either solely or in combination with estradiol benzoate (EB). The control group ob-
tained the vehicle only. The total administered amount of tested compounds was
5 mg/kg of body weight (b.w.), except in case of tamoxifen, used as a comparator,
where doses were 5 mg/kg b.w., i.e. 25 mg/kg b.w., whereas the EB dose was 30 lg/
kg b.w. The animals were sacrificed on the fourth day. The uteri were removed, dis-
sected free of adhering fat and blotted dry after expulsion of uterine fluid and the wet
weights were recorded.
The uterotrophic assay was carried out in six different experiments, where every
experiment had its own control and EB treated group. The means of the relative uteri
weights in control groups from different experiments were comparable, as well as in
the EB treated groups.
Percentage of agonist and antagonist activity in immature rat uterine weight as-
says were calculated from the ratio of values recorded in the treated and control an-
imals, thus
% agonism ¼ ðC ꢀ AÞ ꢂ 100=ðB ꢀ AÞ;
% antagonism ¼ ðB ꢀ DÞ ꢂ 100=ðB ꢀ AÞ;
where A, B, C, and D are uterine wet weights, corrected for differences in body
weights, i.e. (mg/100 g body weight) for vehicle alone, EB, test compound alone, or
test compound plus EB, respectively.
3. Results and discussion
The starting compound, 3-benzyloxy-17-hydroxy-16,17-secoestra-1,3,5(10)-triene-
16-nitrile (1b, Scheme 1) was synthesized in several synthetic steps, starting from
estrone [1,2].
Action of tosyl chloride in dry pyridine upon compound 1b afforded the corre-
sponding 17-tosyl ester 2b, in 84% yield. The substitution reaction of the 17-tosyloxy
function of 2b with tetrabutyl ammonium fluoride in refluxing ethylmethyl ketone
afforded the 17-fluoro derivative 3b, in a 70.4% yield.

ꢁ
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
481
On the other hand, 17-chloro-, bromo-, and -iodo-derivatives were obtained in
high yields directly from secocyanoalcohol 1b, in the substitution reaction of the
17-hydroxy function, with the appropriate reagents. Namely, the chloro-(4b) and
bromo-(5b) derivatives were prepared from 1b by the action of carbon tetrachloride,
i.e., tetrabromide, in the presence of triphenylphosphine [3]. The iodo-(6b) derivative
was formed by treating of compound 1b with a mixture of triphenylphosphine, im-
idazole and iodine in toluene at reflux temperature [4].
The deprotection of the 3-hydroxy function of compounds 2b–5b was performed
by hydrogenolysis at room temperature and low hydrogen pressure, using Pd/C as a
catalyst, whereby 3-hydroxy-16,17-seco-estrone derivatives 2a–5a were obtained in
satisfactory yields [5].
On the other hand, 3-hydroxy-17-iodo derivative 6a could not be obtained by hy-
drogenolysis from the corresponding benzyl ether 6b, probably due to some kind of
long-range effect. Namely, this compound did not react at all under the mentioned
reaction conditions, even if the reaction mixture was acidified (perchloric acid)
and/or heated. Therefore, the derivative 6a was synthesized by substitution of the to-
syloxy function of compound 2a with tetrabutyl ammonium iodide in ethyl methyl
ketone [5].
The estrogenic and antiestrogenic effects of compounds 2a–6a and 2b–6b were
tested on experimental animals, using the uterotrophic and antiuterotrophic assay
[6]. The differences in uteri weights of treated and control animals served for the cal-
culation of the agonistic and antagonistic effects [7], which are presented in Table 1.
As can be seen from Table 1, all the newly synthesized compounds exhibited an
almost total loss of estrogenic activity, while many of them even prevented the action
of endogenous estrogens.
It is known that the estrogens act at the level of estrogen receptors (ER), whereby
the presence of free hydroxyl group at the aromatic moiety is necessary [8,9].
Table 1
Agonistic and antagonistic effects of compounds 2a–6a, 2b–6b, and tamoxifen
Compound
Dose
(mg/kg)
n
Estrogenic effect
(%, mean ꢃ SEM)
n
Antiestogenic effect
(%, mean ꢃ SEM)
1a
1b
5
5
6
7
6
7
6
6
6
7
7
7
6
6
7
8
)2.06 ꢃ 1.06
0.71 ꢃ 0.90
4.52 ꢃ 5.72
)3.51 ꢃ 0.36
0.48 ꢃ 2.08
)4.21 ꢃ 2.07
1.71 ꢃ 0.70
)2.97 ꢃ 0.88
0.80 ꢃ 0.84
)0.19 ꢃ 0.79
3.55 ꢃ 4.77
)5.64 ꢃ 1.29
39.78 ꢃ 3.13
39.36 ꢃ 4.09
8
8
7
7
8
8
7
7
7
7
7
7
—
7
21.13 ꢃ 2.05
31.47 ꢃ 2.26
26.34 ꢃ 3.49
23.37 ꢃ 3.05
0.04 ꢃ 6.44
25.08 ꢃ 2.26
32.95 ꢃ 5.37
25.13 ꢃ 2.61
32.02 ꢃ 3.52
22.28 ꢃ 4.08
16.08 ꢃ 2.95
3.62 ꢃ 5.70
—
2a
2b
5
5
3a
3b
5
5
4a
4b
5
5
5a
5b
5
5
6a
6b
5
5
Tamoxifen
Tamoxifen
5
25
62.80 ꢃ 2.13

ꢁ
482
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
Many of steroidal and nonsteroidal compounds with a phenolic function, even simple
molecules, such as tetrahydronaphtol [9], show high affinity for binding to ER. How-
ever, this binding does not lead to the activation of the receptor, i.e., such derivatives do
not express estrogenic action. For activation of the receptor, besides 3-OH function,
the steroidal skeleton, as well as the presence of a 17-hydroxy group, is crucial.
It was expected that compound 1a, fulfilling the mentioned requirements, will
show estrogenic activity, but, surprisingly, it even hindered the action of endogenous
estrogens. In order to explain such behaviour, we submitted the crystals of 1a to X-
ray structural analysis. The stereo diagram of the molecular structure of 1a and its
relevant crystal data are given in Fig. 1, i.e., Table 2. Selected torsion angles are gi-
ven in Table 3. As it can be seen, in compound 1a the 17-OH function has an oppo-
site orientation in comparison to the 17b-OH group of estradiol, which is essential
for ER activation.
The crystal packing arrangement of this compound could be seen in Table 4. It
consists of coils formed by O1ꢄ ꢄ ꢄN and O2ꢄ ꢄ ꢄO1W hydrogen bonds along b axis.
The coils are linked by O1Wꢄ ꢄ ꢄO1 hydrogen bond.
On the other hand, the synthesized compounds, except 3a and 6b, expressed mod-
erate antiestrogenic effects, which were regularly higher in the case of compounds
with a free 3-OH function, than those bearing the 3-benzyloxy group. The biological
action of 3-benzyloxy derivatives could be explained in two ways. Namely, it can be
assumed that the animal etherases slowly deprotect the 3-OH group, giving the cor-
responding more active compound, or that the benzyl function is capable to bind to
an allosteric site of ER via hydrophobic interaction, leading to conformational
changes of the ER, which slightly hinder the binding of estradiol to the binding site.
With an aim to clarify these presumptions, experiments on in vitro systems will be
undertaken. Furthermore, on the basis of molecular modelling data, synthesis of new
derivatives of compound 1a, directed towards obtaining of compounds with higher
antagonistic effect will be performed.
Fig. 1. The stereo diagram of the molecular structure of compound 1a with the atomic labelling scheme.
Displacement ellipsoids are drawn at the 30% probability level and the H atoms are shown as small spheres
of arbitrary radii.

ꢁ
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
483
Table 2
Crystal data, data collection, and refinement parameters for compound 2a
Crystal data
Data collection
Refinement parameters
C18H23NO2 + H2O
Mr ¼ 303:39
Siemens P4 diffractometer
2h=x scans
4392 Measured reflections
Refinement on F 2
R½F 2 > 2rðF 2Þꢅ ¼ 0:0349
Orthorhombic, P212121
wRðF 2Þ ¼ 0:0848;
2
w ¼ 1=½r2ðFo2Þ þ ð0:0443PÞ þ 0:3945Pꢅ,
where P ¼ ðFo2 þ 2Fc2Þ=3
ꢂ
a ¼ 8:0576ð6Þ A
3654 Independent reflections
3287 Reflections with
I > 2rðIÞ
ꢂ
b ¼ 9:3546ð7Þ A
S ¼ 1:014
ꢂ
c ¼ 21:1293ð14Þ A
hmin ¼ 2:38
3654 Reflections
299 Parameters
H-atom parameters constrained
3
ꢂ
V ¼ 1592:6ð2Þ A
Z ¼ 4
hmax ¼ 27:50
h ¼ ꢀ1 ! 10
Dx ¼ 1:265 Mg mꢀ3
Mo Ka radiation
Cell parameters from 50
reflections
k ¼ 0 ! 12
ðD=rÞ
ðDqÞ
¼ 0:001
max
3
ꢂ
l ¼ ꢀ27 ! 27
¼ 0:162 e A
3
ꢂ
Three standard reflections
ðDqÞmax ¼ ꢀ0:131 e A
min
h ¼ 7:677°–27:263°
l ¼ 0:085 mmꢀ1
T ¼ 293ð1ÞK
Frequency: 197 min
Intensity decay: none
Irregular ellipsoid,
colourless
0.046 ꢂ 0.044 ꢂ 0.044 mm
Table 3
Selected torsion angles (°)
C
C
13–C14–C15–C16
18–C13–C17–O2
88.2(2)
171.5(1)
Table 4
Hydrogen-bonding geometry
ꢂ
ꢂ
D–Hꢄ ꢄ ꢄA
D–H (A)
Hꢄ ꢄ ꢄA (A)
2.24
1.79
1.91
Dꢄ ꢄ ꢄA (°)
3.004(2)
2.653(2)
2.765(2)
D–Hꢄ ꢄ ꢄA (°)
154(2)
179(3)
172(2)
O1–H1ꢄ ꢄ ꢄNi
0.83
0.83
0.83
O2–H2ꢄ ꢄ ꢄO1Wii
O1W–H2Wꢄ ꢄ ꢄO1iii
Symmetry code: (i) 1 ꢀ x; 1=2 þ y; 1=2 ꢀ z; (ii) 1 ꢀ x; ꢀ1=2 þ y; 1=2 ꢀ z; and (iii) ꢀx; ꢀ1=2 þ y;
1=2 ꢀ z.
Acknowledgments
This work was supported by a research grant from the Ministry of Science, Tech-
nologies and Development of the Republic of Serbia (Grant No. 1896).

ꢁ
484
S. Jovanoviꢀc-Santa et al. / Bioorganic Chemistry 31 (2003) 475–484
References
ꢁ
ꢀ
[1] S. Jovanovic-Santa, M.D. Thesis, University of Novi Sad, Novi Sad, 2000.
ꢁ
ꢀ
ꢀ
ꢁ
ꢀ
ꢀ
[2] S. Jovanovic-Santa, S. Andric, R. Kovacevic, V. Pejanovic, Collect. Czech. Chem. C 65 (2000) 77–82.
[3] A. Kashem, M. Anisuzzaman, R.L. Whisler, Carbohydr. Res. 61 (1978) 511–518.
[4] P.J. Garegg, B. Samuelson, J. Chem. Soc., Perkin I (1980) 2866–2869.
ꢁ
ꢀ
ꢀ
ꢀ
[5] S. Jovanovic-Santa, V. Pejanovic, J. Petrovic, J. Serb. Chem. Soc. 64 (1999) 393–396.
[6] C.W. Emmens, in: C.W. Emmens (Ed.), Hormone Assay, Academic Press, New York, 1950, pp. 408–
410.
[7] A.E. Wakeling, B. Valcaccia, E. Newboult, L.R. Green, J. Steroid Biochem. 20 (1984) 111–120.
[8] G.M. Anstead, K.E. Carlson, J.A. Katzenellenbogen, Steroids 62 (1997) 268–303.
[9] W.L. Duax, J.F. Griffin, D.C. Rohrer, C.M. Weeks, R.H. Ebright (Eds.), Biochemical Actions of
Hormones, Academic Press, New York, 1984, pp. 187–206.