Novel modified steroid derivatives of androstanolone as chemotherapeutic anti-cancer agents

Article abstract of DOI:10.1016/j.ejmech.2009.04.020

The aim of the present study is to synthesize and evaluate new potential chemotherapeutic anti-tumor agents. Several thiazolo-, pyrido-, pyrano- and lactam steroid derivatives were obtained using 17β-hydroxy-5α-androstan-3-one (androstanolone) 1 as starti

Full text of DOI:10.1016/j.ejmech.2009.04.020

European Journal of Medicinal Chemistry 44 (2009) 3936–3946
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Novel modified steroid derivatives of androstanolone
as chemotherapeutic anti-cancer agents
,
b
Mohamed El-Far a, Gamal A. Elmegeed b , Emad F. Eskander ,
*
Hanaa M. Rady c, Mohamed A. Tantawy b
a Division of Biochemistry, Faculty of Science, Mansoura University, Mansoura, Egypt
b Hormones Department, National Research Center, Dokki, Giza, Egypt
c Chemistry of Natural Compounds Department, National Research Center, Dokki, Giza, Egypt
a r t i c l e i n f o
a b s t r a c t
Article history:
The aim of the present study is to synthesize and evaluate new potential chemotherapeutic anti-tumor
agents. Several thiazolo-, pyrido-, pyrano- and lactam steroid derivatives were obtained using
17b-hydroxy-5a-androstan-3-one (androstanolone) 1 as starting steroid. The structure of the novel
steroid derivatives was confirmed using the analytical and spectral data. The most pure and structurally
promising compounds 7a, 10a, 12b, 18 and 23 were evaluated as anti-tumor agents. The in vitro cytotoxic
activity was evaluated against hepatoma cell lines using MTT assay. Also the in vivo anti-tumor activity
was evaluated against Ehrlich ascites carcinoma (EAC). The results of the in vitro study showed that at
incubation time 72 h, in olive oil, compound 7a was the most effective cytotoxic compound with IC50 of
Received 15 January 2009
Received in revised form
5 April 2009
Accepted 9 April 2009
Available online 17 April 2009
Keywords:
Anti-tumor
Steroids
Heterocycles
30
35
m
M, while the effects of compounds 18 and 23 were approximately similar with IC50 of 37
mM and
m
M respectively. While the tested compounds when dissolved in DMSO showed approximately the
Hepatoma (HepG2)
Ehrlich ascites carcinoma (EAC)
Cytotoxicity
same IC50 at both 48 and 72 h incubation period, compound 23 was the most effective cytotoxic with IC50
42 M at 48 h and 40 M at 72 h. The results of the in vivo study showed that all the tested novel
m
m
compounds at 25 mg/kg were effective against EAC. Our novel steroid derivatives are promising candi-
dates as anti-cancer agents, none of the mice treated with our novel derivatives showed any toxic
symptoms, but they also completely inhibited tumor growth and retained the hemoglobin content, body
weight, and WBCs near normal values and similar to what obtained for the standard drug 5-flurouracil.
Ó 2009 Elsevier Masson SAS. All rights reserved.
1. Introduction
mutlifocal nature of liver carcinoma, most cancer patients are
considered non-resectable. In these patients, chemotherapy is the
Hepatocellular carcinoma (HCC) is one of the most common
types of malignancies that carries poor prognosis worldwide; it is
the fifth most common cancer and the third leading cause of cancer
death [1]. The incidence of HCC varies greatly worldwide, reflecting
the prevalence of its common etiologic factors such as hepatitis B
(HBV) and hepatitis C (HCV) virus infections, heavy alcohol
consumption and high dietary exposure to aflatoxins [2]. In Egypt,
HCC is the second most common malignancy in males and the fifth
in females. The major risk factor is infection by HCV, which
accounts for 20% of acute hepatitis, 70% of chronic hepatitis, 40% of
cirrhosis and up to 90% of HCC cases [3]. Egypt has the highest
prevalence of hepatitis C virus (HCV) in the world with an average
of approximately 13.8% in the general population [4]. Because of the
only choice of treatment. Unfortunately, development of drug
resistance in tumor after treatment is always a major obstacle to the
successful management of liver cancer [5]. Thus, developing new
therapeutic agents that can overcome drug resistance becomes an
urgent need for cancer patients.
Hybrid anti-canceragents, which combine two active compounds
in one, such as steroidal alkylators, contain steroidal moiety as bio-
logicalvectorsforcytotoxic agents in ordertodiminishtoxicityandto
enhance specificity was recently demonstrated [6]. These agents
attain duplicate effects on cancer cells. A variety of steroids with
unusual and interesting structures have been synthesized and eval-
uated for their anti-tumor activity [7,8]. Also the anti-tumor activity
of many compounds containing heterocyclic ring has been reviewed
[9,10].
In view of the above facts, we set as a goal to prepare some new
steroid derivatives by combining heterocyclic moiety of potent
cytotoxic activity. We report here the facile synthesis of new
* Corresponding author. Tel.: þ20 235682070; fax: þ20 23370931.
E-mail address: gamalae@hotmail.com (G.A. Elmegeed).
0223-5234/$ – see front matter Ó 2009 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2009.04.020

M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
3937
steroidal heterocyclic derivatives. Furthermore, the more structur-
ally promising compounds were examined for their cytotoxicity in
vitro against a panel of hepatoma cell lines (HepG2) using MTT
[3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide]
assay. Also, the new compounds as anti-cancer agents were
examined in vivo against Ehrlich ascites carcinoma (EAC) in mice.
containing amino groups which were identified, according to the
analytical and spectral data (cf. Materials and methods), as
60-aminopyrano[30,40:2,3]androstane derivatives 7a,b (Scheme 1).
The formation of these compounds was assumed to proceed via the
non-isolable aldol adduct intermediates 6a,b followed by cycliza-
tion (Scheme 1).
The reaction of compound 4 with either phenyl isothiocyanate
8a or phenyl isocyanate 8b in dimethylformamide using triethyl-
amine as a catalyst afforded the corresponding non-isolable inter-
mediates 9a,b which readily underwent intramolecular cyclization
to give the 60-aminopyrido[30,40:2,3]androstane derivatives 10a,b
respectively (Scheme 2). Elucidation of the proposed structures
7a,b and 10a,b was based on their correct elemental analyses
and compatible IR, 1H NMR, 13C NMR and mass spectral data
(cf. Materials and methods). The reactivity of enaminonitrile
moiety of compounds 7a and 7b towards active methylene reagents
was studied in the aim of further confirmation of their structures
and to form annulated aminopyridopyrano-steroids with potential
biological activities. Thus, the reaction of compound 7a or 7b with
equimolar amounts of malononitrile 3 in boiling dioxane solution
containing a catalytic amount of triethylamine afforded the corre-
sponding diaminopyrido[200,300:60,50]pyrano[30,40:2,3]androstane
derivatives 12a,b respectively (Scheme 2). The reaction takes place
via an initial addition of the active methylene group of malononi-
trile to the cyano group of compounds 7a,b to give the non-isolable
intermediates 11a,b which undergo intramolecular cyclization to
afford the isolable products 12a,b in 78% and 72% yield respectively
(Scheme 2). The mass spectrum of compound 12a revealed
2. Results and discussion
2.1. Chemistry
Pyrane, pyridine, thiazole, thiophene and lactam rings represent
molecular frameworks that serve as a platform for developing
pharmaceutical agents for various applications. Many derivatives of
these rings proved as anti-tumor agents [11–13]. Simple acetylation
of 17
17 -acetoxy-5
malononitrile 3 in the presence of ammonium acetate to form the
Knoevenagel condensed product, 3-dicyanomethylene-5 -andros-
b
-hydroxy-5a-androstan-3-one (androstanolone 1) afforded
b
a
-androstan-3-one 2. Compound 2 was fused with
a
tane derivative 4 (Scheme 1) in 75% yield. The mass spectrum of
compound 4 (EIMS) revealed a molecular ion peak at m/z ¼ 380
(28%) and the IR spectrum showed two cyano groups stretching at
n
2220 and 2225 cmꢀ1
The presence of the
.
a,
b-unsaturated nitrile moiety in compound
4 showed pronounced activity towards several chemical reagents.
The reaction of compound 4 with either benzaldehyde 5a or ani-
saldehyde 5b in refluxing ethanol in the presence of sodium acetate
did not give the corresponding arylidine but afforded compounds
OH
H3C
OAc
H3C
H3C
H
CH3COCl
O
R
1
2
OAc
R
H3C
H3C
H
CHO
H3C
CN
CN
NH4Ac
fusion
HO
NC
5a, R=H; b, R=OCH3
2
+
EtOH / NaOAc
NC
1
H
CN
3
CN
4
4
6a,b
R
18
H3C
17 OAc
12
11
19
13
16
15
H3C
9
2'
1
2
1'
14
O
8
10
3'
6'
4'
7
3
4
6
H2N
5'
H
CN
7a, R=H
b, R=OCH3
Scheme 1.

3938
M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
OAc
H3C
Y
H3C
PhNCY
Ph
H3C
H
NH
NC
8a, Y=S; b, Y=O
DMF/Et3N
H
NC
CN
4
CN
9a,b
Y
H3C
Ph
N
H2N
H
CN
10a, Y=S
b, Y=O
R
R
OAc
H3C
H3C
H
H3C
H
O
CN
O
dioxane / Et3N
+
H2N
CN
H2N
NC
CN
NH2
11a,b
3a
7a, R=H
b, R=OCH3
NC
R
H3C
H
O
N
H2N
NH2
NC
12a, R=H
b, R=OCH3
Scheme 2.
a molecular ion peak (Mþ ꢀ 1) at m/z ¼ 551 (25%) and its IR spec-
compound 2 reacted with hydrazine hydrate 13 by fusion in the
presence of glacial acetic acid drops to form the 3-hydrazono-5
androstane derivative 14 in 80% yield (Scheme 3). The reaction of
amines with carbon disulfide followed by cyclization with -hal-
oketones to yield the corresponding thiazole with potential bio-
logical activities has been reported [14,15]. Thus, the reaction of
compound 14 with carbon disulfide in dimethylformamide, con-
taining potassium hydroxide, afforded the non-isolable interme-
diates, N-potassium thiocarbamate salt 15 (Scheme 3). The latter
reacted in situ with phenacyl bromide 16 to give the phenyl thia-
zolyl androstane derivative 18 in 73% yield via the non-isolable
intermediate 17. The mass spectrum of compound 18 revealed
trum showed absorption peaks at
n
3450–3230 cmꢀ1 (2NH2) and
n
a-
2220 cmꢀ1 (CN). The 1H NMR spectrum of compound 12a revealed
the presence of two D2O-exchangeable singlets (4H) at
d
¼ 4.89 and
a
5.62 ppm which are characteristic for the NH2 groups. Also, all the
analytical and spectral data of compound 12b are in agreement
with the proposed structure (cf. Materials and methods). Mixture of
distereoisomers at position C-20 in compounds 7a,b and 12a,b was
isolated. The isolation and identification of these isomers were
beyond the scope of this study.
The reactivity of 17b-acetoxy-5a-androstan-3-one 2 towards
hydrazines was studied in the aim to form aminosteroids. Thus,

M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
3939
H3C
CH3COOH
CS2 - KOH
2 + NH2NH2. H2O
fusion
DMF
H2NN
13
H
14
H3C
H
H3C
H
PhCOCH2Br
S
S
C
-
+
16
PhOC
N
S
KS
NH
NH
N
15
17
CH3COCH2Cl
19
H3C
H3C
S
S
S
S
N
N
N
H
N
H
Ph
H3C
18
20
Scheme 3.
molecular ion peak at m/z ¼ 522 (33%) and its IR spectrum showed
4-aza-A-homoandrostan-3-one 23, in 72% yield (Scheme 4) free
from the isomer 3-aza-A-homoandrostan-3-one. The IR spectrum
absorption peak at
n
¼ 1195 cmꢀ1 for the C]S group. Moreover, the
1H NMR spectrum revealed the disappearance of NH2 group singlet
of compound 23 revealed cyclic carbonyl group at
n
¼ 1682 cmꢀ1 in
and the presence of singlet at
d
6.47 for the thiazole 5-H and
addition to the acetate-carbonyl at
n
¼ 1734 cmꢀ1 and its 1H NMR
a multiplet (5H) at 7.46–7.84 for the aromatic protons. Similarly,
d
spectrum showed broad singlet at 8.03 ppm for the amide proton.
All the analytical and spectral data of compound 23 are in accor-
dance with the proposed structures.
the reaction of the non-isolable intermediate 15 with chlor-
oacetone 19 under the same experimental conditions provided the
methyl thiazolyl androstane derivative 20 in 68% yield (Scheme 3).
The structure of compound 20 was confirmed by their elemental
analysis and spectral data (cf. Materials and methods).
2.2. In vitro evaluation of the cytotoxic activity
Taken into account the structural requirements for the cyto-
toxicity, several new hydroximinosteroid derivatives had been
prepared and evaluated for their cytotoxic activity [16,17].
Compound 2 reacted with hydroxylamine hydrochloride 21 in
ethanolic sodium acetate solution to give (E)-3-hydroxyiminoan-
drostane derivative 22 in 80% yield (Scheme 4). The structure of
compound 22 was determined on the basis of its analytical and
spectral data. IR spectra bands at 3267 cmꢀ1 for the (E)-isomer
indicate the presence of a hydroximine group. In the 1H NMR
spectrum of compound 22, the signal for Hb–C(4) was shifted
downfield, appearing at 3.25 ppm as a doublet of doublets due to
the deshielding influence of the hydroxy oxygen of the oxime
which confirmed the (E) configuration. Also, the 1H NMR revealed
In this study, the most structurally promising compounds 7a,
10a, 12b, 18 and 23 were investigated individually as anti-tumor
agents against Hepatoma (HepG2) cell lines. The inhibition of
proliferation of HepG2 cells was determined using MTT assay. The
OAc
H3C
H3C
NaOAc
NH2OH.HCl
2
+
HO
21
N
H
22
the presence of broad singlet (1H) at
group.
d
¼ 9.36 ppm for the NOH
OAc
H3C
In view of the importance of steroid containing alkylating
agents, modified steroids, such as aza-homosteroids, characterized
by the amide group in the steroidal nucleus as biological carrier
that transports the alkylating agent to the tumor site was reported
[18]. It is well known that Beckman rearrangement of steroidal
keto-oximes in the E-configuration usually leads to the formation of
enamine-type lactams [19,20]. Thus, treatment of (E)-3-hydrox-
iminoandrostane derivative 22 with thionyl chloride in dioxane
solution gave, as a single product, an enamide-type lactam,
SO2Cl
dioxane
H3C
O
N
H
H
23
Scheme 4.

3940
M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
and 100
(P > 0.01) inhibited the cell proliferation of HepG2 cells. At 72 h
incubation, and concentration 50 M/ml: compounds 7a, 18 and 23
significantly (P > 0.01) inhibited the cell proliferation of HepG2
cells, while compounds 10a and 12b showed very weak cytotoxic
effect on HepG2 cells compared with that obtained at 48 h of
mM/ml concentration, all tested compounds significantly
C -ve C+ve olive oil C+veDMSO
m
0.4
0.35
0.3
0.25
0.2
incubation time. At a concentration of 100 mM/ml, compounds 7a,
18 and 23 showed highly significant (P > 0.01) cytotoxic effect on
HepG2 cells with percent of inhibition 69, 55 and 68% respectively.
Also compounds 10a and 12b could overcome the cells resistance
and significantly (P > 0.01) inhibited the cell growth of HepG2 cells
with the rates 38 and 41% respectively.
0.15
0.1
0.05
0
Table 3 revealed the IC50 of the tested compounds. Actually, all
tested compounds gave the best results when dissolved in olive oil
24h
48h
Incubation time
72h
in 50 and 100
the most effective one among the other tested compounds
(IC50 ¼ 30 M), while the effect of compounds 18 and 23 was
M and 35 M respectively). In the
mM/ml and incubation time 72 h. Compound 7a was
Fig. 1. Effect of DMSO and olive oil (100
volume, at different time intervals; C ¼ control.
m
l) on growth of HepG2 cells with 100 ml
m
approximately similar (IC50 ¼ 37
m
m
case of using DMSO as solvent, IC50 of all tested compounds were
elevated as compared to that obtained in the case of olive oil.
However, compound 23 showed also good inhibition of HepG2 cells
usage of olive oil or DMSO as a solvent at a volume of 100 ml (the-
maximum volume used to dissolve the tested compounds) had
non-significant effect on the viability of HepG2 cells when treated
for 24 h, 48 h and 72 h (Fig. 1). Data in Tables 1 and 2 are expressed
as percentage of cell growth inhibition of the treated cells vs con-
trols ꢁ S.D. calculated on the average of the experiments performed
when dissolved in DMSO in 50 or 100
mM/ml at incubation time
72 h (IC50 ¼ 40 M). The results revealed that olive oil is the suitable
m
vehicle for anti-tumor drugs and is better than DMSO. Olive oil has
not synergic effect with anti-tumor activity [21] as shown in Fig. 1.
These results were confirmed also via the indirect effect of the
tested compounds on morphology of HepG2 cells. The cells
underwent dramatic morphological changes, shrunken and the
ratio of cytolysis increased after exposure to tested compounds that
were dissolved in olive oil, while the cells that were treated with
compounds dissolved in DMSO still have some viable cells and less
cytolysis compared to that obtained in the case of olive oil
(Photograph 1 for compound 7a, as example).
in triplicate. The concentration 10 mM/ml did not show any signif-
icant effect on viability of hepatoma cells at 24, 48, and 72 h incu-
bation time for any of the tested compounds in olive oil or in DMSO.
The results in Table 1 and Fig. 2 showed the effect of compounds 7a,
10a, 12b, 18 and 23, dissolved in olive oil, on HepG2 cells at 24, 48
and 72 h incubation time. At 24 h incubation and concentration
50 mM/ml: compound 7a increased HepG2 cells growth (40%,
P < 0.05) while compound 10a showed significant inhibition of
HepG2 cells growth (26%, P < 0.05). The other tested compounds
12b, 18 and 23 showed non-significant cell growth inhibition of
2.2.1. Structure–activity relationship
HepG2 cells. At a concentration of 100 mM/ml, compound 10a
The analysis of the structure–activity relationship indicates that
the cytotoxicity of the tested compounds seems to be linked to the
presence of heterocyclic moiety fused to the steroid moiety. This is
identified clearly in comparison with the published cytotoxic
activity of some steroids [22]. In olive oil at 72 h incubation time the
aminopyranoandrostane derivative 7a is the most cytotoxic
compound followed by the lactam derivative 23 and thiazolyl
androstane 18. Compound 7a is more effective than the amino-
pyridoandrostane derivative 10a and this may be attributed to the
presence of pyrane ring with its oxygen atom at position 2. Also, the
addition of pyridine ring to the pyranoandrostane moiety to form
compound 12b decreased the cytotoxic activity of the pyranoan-
drostane moiety of compound 7a. Also, 4-aza-A-homoandrostan-3-
one derivative 23 is the most cytotoxic derivative in DMSO at 48
and 72 h incubation time. This compound is characterized by the
presence of the amide group in the steroid nucleus, which acts as
a biological carrier that transports the alkylating agent to the tumor
site. The presence of amide group is important in order to lower
significantly inhibited the proliferation of HepG2 cells (38%,
P > 0.01), while the other tested compounds showed non-signifi-
cant effect on HepG2 cells. At 48 h incubation, there was significant
cell growth inhibition of hepatoma cells by all the tested
compounds with the same effect at both 50 and 100 mM/ml
concentration. The cytotoxic effect of all tested compounds at 72 h
incubation was highly significant (P > 0.001) and more than that
obtained at 48 h incubation, the percent of cell growth of
compounds 7a, 10a, 12b, 18 and 23 was 66, 74, 65, 64 and 65%
respectively.
The results in Table 2 and Fig. 3 showed the effect of compounds
7a, 10a, 12b, 18 and 23 dissolved in DMSO on HepG2 cells at 24, 48
and 72 h incubation time. At 24 h incubation, and concentration
50
inhibited the cell proliferation of HepG2 cells. Also at a concentra-
tion of 100 M/ml, all tested compounds non-significantly inhibi-
mM/ml: compounds 7a, 18 and 23 non-significantly (P < 0.05)
m
ted the cell growth of HepG2 cells. At 48 h incubation, at both 50
Table 1
Effect of tested compounds using olive oil as solvent on HepG2 cell line proliferation, data expressed as percent of cell growth ꢁ S.D. *P < 0.05 was considered as statistically
significant. **P < 0.01 was considered as highly significant.
Treatment
Conc.
24 h
10
48 h
10
72 h
10
m
M
50
m
M
100
mM
m
M
50
mM
100
m
M
m
M
50
mM
100 mM
7a
99.05 ꢁ 0.27
99.7 ꢁ 0.00
139.5 ꢁ 0.06**
74.20 ꢁ 0.32*
80.66 ꢁ 0.01
99.5 ꢁ 0.04
112.72 ꢁ 0.08
62.98 ꢁ 0.31**
95.61 ꢁ 0.01
112.72 ꢁ 0.04
115.45 ꢁ 0.03
100.04 ꢁ 0.05
97.5 ꢁ 0.01
46.13 ꢁ 0.04**
47.49 ꢁ 0.27**
47.91 ꢁ 0.02**
45.71 ꢁ 0.02**
46.27 ꢁ 0.03**
45.45 ꢁ 0.09**
43.07 ꢁ 0.55**
50.86 ꢁ 0.02**
46.72 ꢁ 0.00**
47.01 ꢁ 0.07**
97.04 ꢁ 0.01
96.34 ꢁ 0.02
98.56 ꢁ 0.01
95.37 ꢁ 0.01
95.88 ꢁ 0.00
29.85 ꢁ 0.006**
26.87 ꢁ 0.00**
35.25 ꢁ 0.06**
32.80 ꢁ 0.07**
35.97 ꢁ 0.04**
33.63 ꢁ 0.00**
25.89 ꢁ 0.00**
35.25 ꢁ 0.00**
34.53 ꢁ 0.00**
35.25 ꢁ 0.02**
10a
12b
18
100.04 ꢁ 0.00
99.02 ꢁ 0.07
98.97 ꢁ 0.04
98.88 ꢁ 0.03
99.67 ꢁ 0.04
100.34 ꢁ 0.00
23
99.09 ꢁ 0.01

M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
3941
Table 2
Effect of tested compounds using DMSO as solvent on HepG2 cell line proliferation, data were expressed as percent of cell growth ꢁ S.D. *P < 0.05 was considered as statistically
significant. **P < 0.01 was considered as highly significant.
Treatment
Conc.
24 h
10
48 h
10
72 h
10
m
M
50
m
M
100
m
M
m
M
50
mM
100
m
M
m
M
50
m
M
100 mM
7a
100.04 ꢁ 0.00
98.97 ꢁ 0.04
99.05 ꢁ 0.27
100.6 ꢁ 0.07
100.02 ꢁ 0.063
90.00 ꢁ 0.00
103.33 ꢁ 0.00
107.33 ꢁ 0.00
86.66 ꢁ 0.00
79.33 ꢁ 0.00
91.33 ꢁ 0.01
97.33 ꢁ 0.01
94.00 ꢁ 0.00
84.00 ꢁ 0.00
72.00 ꢁ 0.00
98.04 ꢁ 0.01
99.34 ꢁ 0.03
98.56 ꢁ 0.07
99.39 ꢁ 0.03
100.88 ꢁ 0.00
62.98 ꢁ 0.01**
77.51 ꢁ 0.00**
74.30 ꢁ 0.00**
66.66 ꢁ 0.00**
37.56 ꢁ 0.00**
41.98 ꢁ 0.00**
59.55 ꢁ 0.00**
47.67 ꢁ 0.01**
41.79 ꢁ 0.00**
38.12 ꢁ 0.00**
98.05 ꢁ 0.04
100.04 ꢁ 0.05
99.26 ꢁ 0.04
97.53 ꢁ 0.00
96.74 ꢁ 0.078
67.21 ꢁ 0.02**
98.34 ꢁ 0.07
89.21 ꢁ 0.05
71.36 ꢁ 0.01**
34.56 ꢁ 00**
31.21 ꢁ 0.00**
62.24 ꢁ 0.028**
58.50 ꢁ 0.01**
45.22 ꢁ 0.02**
31.95 ꢁ 0.00**
10a
12b
18
23
systemic toxicity and improve specificity in cancer treatment [18].
The results verified the importance of the presence of pyrane,
thiazole and lactam moieties as pharmacophores for the anti-
cancer activity.
12b, 18 and 23 at dose level of 25 mg/kg completely inhibited the
tumor growth in the experimental model and showed zero tumor
volume at the end of in vivo experiment, while EAC control showed
4.56 ꢁ 0.37 ml (Table 4, Photograph 2). In cancer chemotherapy one
of the major problems is anemia which is mainly due to reduction
in RBCs or hemoglobin percentage. Treatment with all novel
synthesized compounds retained the hemoglobin content, body
2.3. In vivo evaluation of the anti-tumor activity
The anti-tumor activity was evaluated on EAC bearing mice
using animal model. All the tested novel steroid derivatives 7a, 10a,
24h
110
100
90
80
70
7a
10a
12b
18
60
50
40
30
20
10
0
23
10
10
10
50
conc. µM/ml
100
100
100
48h
110
100
90
80
70
60
50
40
30
20
10
0
7a
10a
12b
18
23
50
conc. µM/ml
72h
110
100
90
80
70
60
50
40
30
20
10
0
7a
10a
12b
18
23
50
conc. µM/ml
Fig. 2. Effect of tested compounds dissolved in olive oil on HepG2 cells at 24, 48 and
Fig. 3. Effect of tested compounds dissolved in DMSO on HepG2 cells at 24, 48 and
72 h.
72 h.

3942
M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
Table 3
and 125 MHz respectively, in DMSO-d6 or CDCl3 as solvent and
chemical shifts were recorded in ppm relative to TMS. Mass spectra
were recorded on a GCMS-QP 1000 Ex spectra mass spectrometer
operating at 70 eV. Elemental analyses were carried out by Micro-
analytical Data Unit at National Research Center, Giza, Egypt.
Reactions were monitored on Merck aluminum thin layer chro-
matography (TLC) plates and visualized by UV light (254 nm). For
the nomenclature of steroid derivatives, we used the definitive
rules for the nomenclature of steroids published by the Joint
Commission on the Biochemical Nomenclature (JCBN) of IUPAC
[23,24].
The in vitro cytotoxic activity of tested compounds on HepG2 cells (IC50 in mM).
Compounds
Olive oil
48 h
DMSO
48 h
72 h
72 h
73
>100
>100
90
7a
46
48
48
47
47
30
40
41
37
35
81
>100
95
85
42
10a
12b
18
23
40
weight, hematocrit % and WBCs beyond normal values and similar
to values observed with the most commonly clinically used drug,
5-flurouracil (Table 4, Fig. 4). Furthermore, none of the treated mice
with novel compounds exhibited any abnormal behavioral or any
toxicity symptoms of dose used during this study. They were all
active as anti-tumor agents with no loss of appetite and showed no
dizziness or erection of hairs or hypothermia (Photograph 2). This
indicated the safety and high activity of all tested compounds at the
specified low dose as chemotherapeutic agents.
3.1.1. 17
b
-Acetoxy-3-dicyanomethylen-5a-androstane (4)
To a mixture of 17
b
-acetoxy-5 -androstan-3-one 2 (0.33 g,
a
1 mmol) and ammonium acetate (0.5 g), equimolar amount of
malononitrile 3 (0.06 g, 1 mmol) was added. The reaction mixture
was heated in an oil bath at 130 ꢂC for 15 min. The solid product
formed upon cooling at room temperature was triturated with
ethanol, collected by filtration and crystallized from ethanol to
yield dark yellow crystals of compound 4, yield 0.28 g (75%), mp
178–180 ꢂC, IR (KBr, cmꢀ1):
(2CN), 1732 (acetate C]O), 1643 (C]C). 1H NMR (CDCl3, ppm):
¼ 0.78 (s, 3H, CH3-19), 1.03 (s, 3H, CH3-18), 2.13 (s, 3H, COCH3),
n
¼ 2935, 2873 (CH3, CH2), 2220, 2225
2.4. Conclusion
d
3.35–3.54 (m, 1H, C5-a
H). MS (EI): m/z (%): 380 (Mþ, 28), 320
In conclusion, we have described a facile synthesis of novel
promising anti-cancer steroid derivatives and investigated also the
importance of incorporating heterocyclic moiety to the steroid
nucleus to form new effective anti-cancer hybrid molecules. The
three compounds namely 7a, 18, and 23 dissolved in olive oil
showed promising broad spectrum cytotoxic activity. All tested
compounds completely inhibited the tumor growth of EAC in
experimental model and brought back the hemoglobin content,
body weight, hematocrit % and WBCs beyond normal values.
Finally, the broad spectrum anti-tumor activity displayed by these
compounds may be of interest for further derivatization, further in
vivo and clinical studies in the hope of finding more active and
selective anti-tumor agents.
(Mþ ꢀ CH3COOH, 33), 273 (C19H29O, 100), 257 (C19H29, 40). Calc. for
C24H32N2O2 (380.5230): C, 75.75; H, 8.48; N, 7.36; found: C, 75.49;
H, 8.32; N, 7.17%.
3.1.2. 17b
-Acetoxy-60-amino-20-phenyl-20H-
pyrano[30,40:2,3]androstan-50-carbonitrile (7a); 17 -acetoxy-60-
b
amino-20-(p-methoxyphenyl)-20H-pyrano[30,40:2,3]-androstan-50-
carbonitrile (7b): General procedure
To a solution of compound 4 (0.76 g, 2 mmol) in absolute
ethanol (30 ml) containing sodium acetate (1.0 g) either benzal-
dehyde 5a (0.21 g, 2 mmol) or anisaldehyde 5b (0.27 g, 2 mmol)
was added. The reaction mixture in each case was heated under
reflux for 3–5 h until all the reactants had disappeared as indicated
by TLC. Sodium acetate was isolated on hot by filtration and the
filtrate was concentrated under vacuum. The isolated product upon
cooling overnight at room temperature in each case was collected
by filtration, dried and crystallized from the appropriate solvent.
3. Materials and methods
3.1. Synthetic methods, spectral and analytical data
Starting steroid 17
b-hydroxy-5a-androstan-3-one 1 (androsta-
3.1.2.1. Compound 7a. Brown crystals from dioxane, yield 0.69 g
nolone) was purchased from Sigma Company, USA. All solvents
were dried by distillation prior to use. All melting points were
measured using an Electrothermal apparatus and are uncorrected.
The IR spectra were recorded in (KBr discs) on a Shimadzu FT-IR
(72%), mp 138–140 ꢂC, IR (KBr, cmꢀ1):
n
¼ 3480 (NH2), 3050
(CH-aromatic), 2935, 2850 (CH3, CH2), 2225 (CN), 1735 (acetate-
C]O), 1615 (C]C). 1H NMR (DMSO-d6, ppm):
¼ 0.75 (s, 3H,
CH3-19), 0.96 (s, 3H, CH3-18), 2.03 (s, 3H, COCH3), 3.39–3.55
(m, 1H, C5- H), 4.49 (s, 2H, NH2, D2O-exchangeable), 7.15–7.45
d
8201 PC spectrometer and expressed in cmꢀ1. The 1H NMR and 13
C
a
NMR spectra were recorded with Jeol instrument (Japan), at 270
Photograph 1. The morphology of HepG2 cells after 72 h incubation: (a) control without any treatment, (b) conc. 100 mM of 7a using olive oil as solvent, (c) conc. 100 mM of 7a using
DMSO as solvent.

M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
3943
Table 4
Effect of the tested compounds on hematological parameters and on body weight of EAC bearing mice. n ¼ 5. Mean ꢁ SE. *P > 0.05 vs normal group. **P > 0.01 vs normal group.
Parameters
Normal
EAC control
(2 ꢃ 106
7a
10a
12b
18
23
Standard
(25 mg/kg) þ EAC (25 mg/kg) þ EAC (25 mg/kg) þ EAC (25 mg/kg) þ EAC (25 mg/kg) þ EAC 5-flurouracil
cells/mouse)
(20 mg/kg) þ EA
Hemoglobin/g %
Haematocrit %
13.63 ꢁ 0.75 6.86 ꢁ 2.25**
42.36 ꢁ 1.00 25.9 ꢁ 2.29**
12.50 ꢁ 0.20
36.6 ꢁ 2.81**
12.60 ꢁ 0.14
33.8 ꢁ 0.70**
5.75 ꢁ 0.49**
18.00 ꢁ 1.41*
Nil
12.65 ꢁ 0.91
38.3 ꢁ 2.12*
7.35 ꢁ 1.48*
21.5 ꢁ 2.12
Nil
10.70 ꢁ 0.72*
42.7 ꢁ 2.15
5.66 ꢁ 1.15**
20.00 ꢁ 1.50
Nil
12.96 ꢁ 0.30
38.56 ꢁ 1.15*
6.46 ꢁ 0.76**
20.33 ꢁ 0.57
Nil
10.73 ꢁ 0.15**
40.86 ꢁ 2.17
6.40 ꢁ 0.52**
23.0 ꢁ 1.0
Nil
Total WBC/103 mmꢀ3 9.16 ꢁ 0.70
11.16 ꢁ 0.25** 5.9 ꢁ 0.30**
Body weight/g
Tumor volume/ml
22.66 ꢁ 0.57 27.33 ꢁ 2.51** 18.66 ꢁ 2.30*
Nil
4.56 ꢁ 0.37
Nil
(m, 5H, C6H5). 13C NMR (DMSO-d6, ppm):
d
¼ 34.7 (C-1), 138.9, 137.5
(2C]O), 1598 (C]C). 1H NMR (CDCl3, ppm):
19), 0.98 (s, 3H, CH3-18), 2.18 (s, 3H, COCH3), 3.27–3.63 (m, 1H, C5-
d
¼ 0.78 (s, 3H, CH3-
(fused, C-2, C-3), 35.3 (C-4), 49.9 (C-5), 27.2 (C-6), 28.7 (C-7), 36.4
(C-8), 52.7 (C-9), 46.8 (C-10), 21.21 (C-11), 37.0 (C-12), 42.2 (C-13),
52.3 (C-14), 23.1 (C-15), 27.5 (C-16), 82.0 (C-17), 21.1 (C-18), 21.7
(C-19), 170 (C]O), 20.9 (CH3-acetyl), 74.4 (C-20), 53.2 (C-50), 182.0
(C-60), 117.0 (CN), 127.7, 129.0, 129.2, 141.8 (C-phenyl). MS (EI): m/z
(%): 486 (Mþ, 43), 426 (Mþ ꢀ CH3COOH, 15), 409 (Mþ ꢀ C6H5, 20),
261 (C19H33, 100), 77 (C6H5, 38). Calc. for C31H38N2O3 (486.6450):
C, 76.51; H, 7.87; N, 5.76; found: C, 76.33; H, 7.67; N, 5.52.
a
H), 5.82 (s, 2H, NH2, D2O-exchangeable), 7.23–7.63 (m, 5H, C6H5).
MS (EI): m/z (%): 498 (Mþ ꢀ 1, 62), 438 (Mþ ꢀ CH3COOH, 50), 77
(C6H5, 100). Calc. for C31H37N3O3 (499.6437): C, 74.52; H, 7.46; N,
8.41; found: C, 74.30; H, 7.23; N, 8.65.
3.1.4. 17
b
-Acetoxy-400,600-diamino-200-
phenylpyrido[200,300:60,50]pyrano[30,4 :2,3]-5 -androstan-500-
a
carbonitrile (12a); 17b
-acetoxy-400,600-diamino-20-(p-
3.1.2.2. Compound 7b. Yellow crystals from methanol, yield 0.72 g
methoxyphenyl)pyrido[200,300,60,50]-pyrano[30,40:2,3]-5
a-androstan-
(70%), mp 128–130 ꢂC, IR (KBr, cmꢀ1):
n
¼ 3452 (NH2), 3030
(CH-aromatic), 2948, 2852 (CH3, CH2), 2220 (CN), 1726 (C]O), 1642
(C]C). 1H NMR (DMSO-d6, ppm):
¼ 0.78 (s, 3H, CH3-19), 1.06
(s, 3H, CH3-18), 2.18 (s, 3H, COCH3), 3.32–3.50 (m, 1H, C5- H), 5.86
500-carbonitrile (12b): General procedure
A solution of equimolar amounts of compound 7a (0.97 g,
2 mmol) or compound 7b (1.03 g, 2 mmol) and malononitrile 3
(0.13 g, 2 mmol) in dioxane (30 ml) containing triethylamine
(0.5 ml) was heated under reflux for 4–5 h until all the reactants
had disappeared as indicated by TLC. The reaction mixture was then
evaporated under vacuum and the remaining product was tritu-
rated with ethanol. The solid product that formed in each case was
collected by filtration and crystallized from the appropriate solvent.
d
a
(s, 3H, OCH3), 5.12 (s, 2H, NH2, D2O-exchangeable), 7.30 (dd, 2H-
aromatic, JHH 9 Hz), 7.50 (dd, 2H-aromatic, JHH 9 Hz). MS (EI): m/z
(%): 516 (Mþ, 28), 456 (Mþ ꢀ CH3COOH, 61), 409 [Mþ ꢀ (C6H4–
OCH3), 55], 107 (C6H4–OCH3, 100). Calc. for C32H40N2O4 (516.6713):
C, 74.39; H, 7.80; N, 5.42; found: C, 74.50; H, 7.98; N, 5.63.
3.1.3. 17
b
-Acetoxy-60-amino-10-phenyl-20-thioxopyrido[30,40:2,3]-
3.1.4.1. Compound 12a. Brown crystals from ethanol, yield 0.86 g
5
a
-androstan-50-carbonitrile (10a); 17
b
-acetoxy-60-amino-10-
(78%), mp 183–185 ꢂC, IR (KBr, cmꢀ1):
n
¼ 3345–3230 (2NH2), 3015
(CH-aromatic), 2220 (CN),1730 (C]O),1642 (C]N),1580 (C]C). 1H
NMR (DMSO-d6, ppm):
¼ 0.74 (s, 3H, CH3-18), 0.89 (s, 3H, CH3-19),
2.04 (s, 3H, OCH3), 3.32–3.50 (m, 1H, C5- H), 4.89, 5.62 (2s, 4H,
phenyl-20-oxopyrido[30,40:2,3]-5
General procedure
a
-androstan-50-carbonitrile (10b):
d
To a solution of compound 4 (0.76 g, 2 mmol) in dime-
thylformamide (20 ml) containing a catalytic amount of triethyl-
amine (1 ml) either phenyl isothiocyanate 8a (0.27 g, 2 mmol) or
phenyl isocyanate 8b (0.23 g, 2 mmol) was added. The reaction
mixture, in each case, was heated under reflux for 5–7 h until all the
reactants had disappeared as indicated by TLC. The reaction mixture
was poured over an ice/water mixture and neutralized with dilute
hydrochloric acid. The solid product that formed in each case was
filtered off, dried and crystallized from the appropriate solvent.
a
2NH2, D2O-exchangeable), 7.35–7.67 (m, 5H, C6H5). MS (EI): m/z
(%): 551 (Mþ ꢀ 1, 25), 492 (Mþ ꢀ CH3COOH, 33), 475 (Mþ ꢀ C6H5,
42), 273 (C19H29O, 18), 257 (C19H29, 100), 77 (C6H5, 37). Calc. for
C34H40N4O3 (552.7064): C, 73.88; H, 7.29; N, 10.14; found: C, 73.62;
H, 7.04; N, 10.35.
3.1.3.1. Compound 10a. Brown crystals from ethanol, yield 0.78 g
(76%), mp 163–165 ꢂC, IR (KBr, cmꢀ1):
n
¼ 3376 (NH2), 3025 (CH-
aromatic), 2930, 2825 (CH3, CH2), 2207 (CN), 1730 (C]O), 1597
(C]C), 1195 (C]S). 1H NMR (DMSO-d6, ppm):
¼ 0.75 (s, 3H, CH3-
19), 0.96 (s, 3H, CH3-18), 2.04 (s, 3H, COCH3), 3.22–3.56 (m, 1H, C5-
H), 6.12 (s, 2H, NH2, D2O-exchangeable), 6.95–7.35 (m, 5H, C6H5).
13C NMR (DMSO-d6, ppm):
¼ 38.8 (C-1), 140.7, 148.9 (fused, C-2, C-
d
a
d
3), 31.7 (C-4), 50.1 (C-5), 27.4 (C-6), 28.9 (C-7), 36.6 (C-8), 53.4 (C-9),
42.6 (C-10), 21.1 (C-11), 38.5 (C-12), 42.3 (C-13), 52.3 (C-14), 23.3 (C-
15), 27.9 (C-16), 82.2 (C-17), 20.2 (C-18), 21.2 (C-19), 170.3 (C]O),
21.0 (CH3-acetyl), 180.2 (C-20), 62.8 (C-50), 157.2 (C-60), 115.0 (CN),
134.0, 126.5, 129.1, 124.8 (C-phenyl). MS (EI): m/z (%): 515 (Mþ, 43),
455 (Mþ ꢀ CH3COOH, 27), 77 (C6H5, 100). Calc. for C31H37N3O2S
(515.7093): C, 72.20; H, 7.23; N, 8.15; S, 6.22; found: C, 72.40; H,
7.45; N, 8.35; S, 6.46.
3.1.3.2. Compound 10b. Yellow crystals from ethanol, yield 0.72 g
Photograph 2. Effect of 7a at dose 25 mg/kg b.wt. on mice with EAC (right), compared
to an EAC control – non treated-animal (left). Substantial difference observed on
treated animals indicating complete response.
(73%), mp 158–160 ꢂC, IR (KBr, cmꢀ1):
n
¼ 3385 (NH2), 3030
(CH-aromatic), 2942, 2835 (CH3, CH2), 2220 (CN), 1735, 1710

3944
M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
CBC
Hemoglobin
Haematocrit
Total WBC
Body weight
45
40
35
30
25
20
15
10
5
0
Normal EAC control
7a
10a
12b
18
23
Standard
5-flurouracil
Fig. 4. Effect of the tested compounds on hematological parameters and on body weight of EAC bearing mice.
3.1.4.2. Compound 12b. Dark brown crystals from ethanol, yield
0.84 g (72%), mp 194–196 ꢂC, IR (KBr, cmꢀ1):
¼ 3382–3320 (2NH2),
3020 (CH-aromatic), 2215 (CN), 1720 (C]O), 1640 (C]N), 1571
(C]C). 1H NMR (CDCl3, ppm):
¼ 0.73 (s, 3H, CH3-18), 0.91 (s, 3H,
CH3-19), 2.09 (s, 3H, COCH3), 3.28–3.45 (m, 1H-C5– H), 3.80 (s, 3H,
mixture was heated in water bath at 80 ꢂC for 1 h, then left to cool
to 20 ꢂC. The
-halogenated compound, phenacyl bromide 16
n
a
(0.39 g, 2 mmol) or chloroacetone 19 (0.18 g, 2 mmol), was added
dropwise to the cooled reaction mixture with stirring. The reaction
mixture, in each case, was stirred at room temperature for 24 h. The
solid product formed upon pouring the reaction mixture into ice
containing few drops of hydrochloric acid was collected by filtra-
tion, dried and crystallized from the proper solvent.
d
a
OCH3), 4.67, 5.23 (2s, 4H, 2NH2, D2O-exchangeable), 7.42 (dd, 2H-
aromatic, JHH 9 Hz), 7.62 (dd, 2H-aromatic, JHH 9 Hz). 13C NMR
(DMSO-d6, ppm):
d
¼ 33.7 (C-1), 142.8, 130.7 (fused, C-2, C-3), 31.4
(C-4), 48.5 (C-5), 29.4 (C-6), 28.9 (C-7), 37.8 (C-8), 52.7 (C-9), 48.7
(C-10), 21.7 (C-11), 30.5 (C-12), 45.6 (C-13), 50.3 (C-14), 23.5 (C-15),
27.9 (C-16), 82.7 (C-17), 22.1 (C-18), 17.3 (C-19), 170.1 (C]O), 20.7
(CH3-acetyl), 82.7 (C-20), 102.3, 162.3 (fused C-50, C-60), 162.5 (C-400),
67.2 (C-500), 161.1 (C-600), 117.8 (CN). MS (EI): m/z (%): 582 (Mþ, 15),
522 (Mþ ꢀ CH3COOH, 28), 475 [Mþ ꢀ C6H4–OCH3, 100], 107 (C6H4–
OCH3, 38). Calc. for C35H42N4O4 (582.7323): C, 72.14; H, 7.26; N,
9.61; found: C, 72.39; H, 7.49; N, 9.80.
3.1.6.1. Compound 18. Dark yellow crystals from ethanol, yield
0.76 g (73%), mp 99–100 ꢂC, IR (KBr, cmꢀ1):
n
¼ 3030
(CH-aromatic), 2970, 2848 (CH3, CH2), 1732 (C]O), 1657 (C]N),
1591 (C]C), 1195 (C]S). 1H NMR (CDCl3, ppm):
¼ 0.65 (s, 3H,
CH3-19), 0.87 (s, 3H, CH3-18), 2.15 (s, 3H, COCH3), 3.32–3.62 (m,
1H, C5–
H), 6.47 (s, 1H, thiazole C5–H), 7.46–7.84 (m, 5H, C6H5). 13
NMR (DMSO-d6, ppm):
¼ 37.5 (C-1), 22.9 (C-2), 162.0 (C-3), 28.1
d
a
C
d
(C-4), 44.0 (C-5), 27.0 (C-6), 30.7 (C-7), 35.2 (C-8), 49.8 (C-9), 45.8
(C-10), 22.3 (C-11), 34.5 (C-12), 42.1 (C-13), 52.8 (C-14), 23.5 (C-
15), 27.6 (C-16), 81.8 (C-17), 20.3 (C-18), 16.9 (C-19), 170.2 (C]O),
20.7 (CH3, acetate), 210.3 (C-20, C]S), 158.0 (C-40), 107.6 (C-50),
134.0, 126.4, 128.7, 127.8 (C-phenyl). MS (EI): m/z (%): 522 (Mþ, 33),
462 (Mþ ꢀ CH3COOH, 21), 316 (Mþ ꢀ C9H6N2S2, 86), 77 (C6H5, 100).
Calc. for C30H38N2O2S2 (522.7649): C, 68.93; H, 7.33; N, 5.36; S,
12.27; found: C, 68.74; H, 7.56; N, 5.17; S, 12.54.
3.1.5. 17b-Acetoxy-3-hydrazono-5a-androstane (14)
To a dry mixture of compound 2 (1.66 g, 5 mmol) and hydrazine
hydrate 13 (0.25 g, 5 mmol), few drops of glacial acetic acid were
added. The reaction mixture was heated in an oil bath at 120 ꢂC for
30 min and then left to cool at room temperature. The solid product
obtained upon triturating the fused result with ethanol followed by
pouring onto ice/water mixture was collected by filtration, dried and
crystallized from 1,4-dioxane to afford compound 14 as white crys-
tals, yield 1.38 g (80%), mp 239–240 ꢂC, IR (KBr, cmꢀ1):
(NH2), 2985, 2878 (CH3, CH2), 1725 (C]O), 1670 (C]N). 1H NMR
(DMSO-d6, ppm):
¼ 0.82 (s, 3H, CH3-19), 1.07 (s, 3H, CH3-18), 2.05
(s, 3H, COCH3), 3.27–3.53 (m, 1H, C5- H), 5.57 (s, 2H, NH2, D2O-
exchangeable). 13C NMR (DMSO-d6, ppm):
¼ 37.8 (C-1), 30.4 (C-2),
n
¼ 3498
3.1.6.2. Compound 20. Yellow crystals from ethanol, yield 0.62 g
(68%), mp 108–110 ꢂC, IR (KBr, cmꢀ1):
n
¼ 2934, 2851 (CH3, CH2),
1730 (C]O), 1643 (C]N), 1515 (C]C), 1192 (C]S). 1H NMR (CDCl3,
ppm):
¼ 0.78 (s, 3H, CH3-19), 0.91 (s, 3H, CH3-18), 2.03 (s, 3H,
COCH3), 2.85 (s, 3H, CH3), 3.32–3.54 (m, 1H, C5– H), 6.57 (s, 1H,
d
a
d
d
a
161.6(C-3),23.3(C-4),42.4(C-5),28.5(C-6),31.1(C-7),36.0(C-8),53.6
(C-9), 45.7 (C-10), 20.4 (C-11), 34.9 (C-12), 40.4 (C-13), 40.3 (C-14),
53.8 (C-15), 27.3 (C-16), 82.4 (C-17), 22.8 (C-18), 16.5 (C-19), 170.6
(C]O), 21.0 (CH3-acetate). MS (EI): m/z (%): 347 (Mþ þ 1, 25), 286
(Mþ ꢀ CH3COOH, 36), 257 (C19H29, 100). Calc. for C21H34N2O2
(346.5068): C, 72.79;H, 9.89; N, 8.08; found: C, 72.52;H,10.12;N, 7.87.
thiazole C5–H). MS (EI): m/z (%): 460 (Mþ, 37), 400 (Mþ ꢀ CH3COOH,
21), 254 (Mþ ꢀ C9H6N2S2, 28), 77 (C6H5, 100). Calc. for
C25H36N2O2S2 (460.6955): C, 65.18; H, 7.88; N, 6.08; S, 13.92; found:
C, 65.35; H, 8.05; N, 5.92; S, 13.75.
3.1.7. 17b-Acetoxy-3-hydroximinoandrostane (22)
To a solution of compound 2 (1.66 g, 5 mmol) in absolute ethanol
(30 ml) containing anhydrous sodium acetate (1 g), a solution of
hydroxylamine hydrochloride 21 (0.41 g, 5 mmol) in water (5 ml)
was added. The reaction mixture was stirred at 25 ꢂC for 30 min
then poured into ice/water mixture. The solid product formed was
collected by filtration and crystallized from ethanol to afford white
crystals of compound 22 in 80% yield (1.38 g), mp 122–124 ꢂC, IR
3.1.6. 17b
-Acetoxy-3-N-(40-phenyl-20-thioxo-10,30-thiazol-30-
yl)androstane (18); 17b
-acetoxy-3-N-(40-methyl-20-thioxo-1,3-
thiazol-30-yl)androstane (20): General procedure
To a solution of compound 14 (0.69 g, 2 mmol) in DMF (30 ml)
a solution of KOH (0.11 g, 2 mmol) in 5 ml H2O was added with
stirring followed by carbon disulfide (0.15 g, 2 mmol). The reaction

M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
3945
(KBr, cmꢀ1):
1648 (C]N), 1542 (C]C). 1H NMR (DMSO-d6, ppm):
CH3-19), 1.06 (s, 3H, CH3-18), 2.13 (s, 3H, COCH3), 3.25 (dd, J ¼ 3.6,
n
¼ 3267 (N–OH), 2931, 2851 (CH3, CH2), 1736 (C]O),
¼ 0.79 (s, 3H,
cells were incubated for an additional 4 h. The medium was dis-
carded; 100 l of DMSO was added to dissolve the purple formazan
d
m
crystals formed. The optical density (OD) of solubilized formazan
was measured at 570 nm (reference filter 690 nm) using an auto-
matic microplate reader (Wako, Japan). The results are expressed as
percent of cell growth inhibition compared with the control. The
effect of compounds 7a, 10a, 12b, 18 and 23 on the morphology of
treated hepatocellular carcinoma cells was investigated by the light
microscope and then photographed by SONY CYBER-SHORT [26].
1H, Hb–C4), 3.38–3.52 (m, 1H, C5–aH), 9.36 (br s, 1H, NOH). MS (EI):
m/z (%): 347 (Mþ, 28), 287 (Mþ ꢀ CH3COOH), 257 (C19H29, 100). Calc.
for C21H33NO3 (347.4916): C, 72.58; H, 9.57; N, 4.03; found: C, 72.35;
H, 9.36; N, 3.87.
3.1.8. 17b-Acetoxy-4-aza-A-homoandrostan-3-one (23)
To a solution of compound 22 (0.69 g, 2 mmol) in dioxane
(20 ml), thionyl chloride (0.15 ml) was added dropwise with stir-
ring. The reaction mixture was stirred for further 30 min at room
temperature. Water was added to this solution and the reaction
mixture was extracted with diethyl ether (3 ꢃ 20 ml). The extract
was washed with water, 5% NaHCO3 and water, dried over CuSO4.
Removal of the solvent gave a crude compound 23, which was
crystallized from methanol to afford white crystals, yield 0.50 g
3.3. In vivo anti-tumor activity
3.3.1. Tumor transplantation
A line of Ehrlich ascites carcinoma (EAC) was supplied from the
National Cancer Institute, Cairo, Egypt. Each mouse was injected
intraperitoneally (i.p.) with 0.2 ml of EAC, which contained 2 ꢃ 106
cells.
(72%), mp 273–275 ꢂC, IR (KBr, cmꢀ1):
(CH3, CH2), 1734, 1682 (2C]O), 1532 (C]C). 1H NMR (CDCl3, ppm):
¼ 0.72 (s, 3H, CH3-19), 0.98 (s, 3H, CH3-18), 2.13 (s, 3H, COCH3),
3.38–3.57 (m, 1H, C5–
H), 8.03 (br s, 1H, NH, D2O-exchangable). 13
NMR (DMSO-d6, ppm):
¼ 39.2 (C-1), 33.5 (C-2), 179.2 (C-3), 43.3
n
¼ 3432 (NH), 2928, 2852
3.3.2. Experimental animals
d
Forty Swiss albino female mice weighing 20 ꢁ 2 g were obtained
from the Animal House Colony of the National Research Center,
Cairo, Egypt. Five mice/cage were housed in plastic cages of
dimensions 42L ꢃ 26W ꢃ 22H centimeters. The animals were
maintained under controlled conditions of humidity, temperature,
and diurnal environment of light and dark. The mean ambient
temperature in the housing facility was 28 ꢂC. The animals were
randomly assigned to 8 groups (n ¼ 5) as follows, Group 1 (vehicle
control) was left without any treatment for 14 days. Group 2
(negative control) injected intraperitoneally (i.p.) with 0.2 ml of
EAC, which contained 2 ꢃ 106 cells for tumor induction and left for
14 days. Group 3 (positive control) injected (i.p.) with 0.2 ml of EAC
and treated with 20 mg/kg b.wt. of reference drug 5-flurouracil for
14 days. Groups 4, 5, 6, 7 and 8 received either one of the tested
compounds 7a, 10a, 12b, 18, or 23 at doses 25 mg/kg b.wt. dissolved
in DMSO (i.p.) day after day for 14 days. After administration of the
last dose followed by 18 h fasting, the mice were then sacrificed for
observation of anti-tumor activity. The effect of the tested
compounds on the growth of transplantable murine tumor, and
simultaneous alterations in the hematological profile were
estimated.
a
C
d
(C-4), 49.2 (C-5), 26.3 (C-6), 27.7 (C-7), 35.7 (C-8), 50.9 (C-9), 44.7
(C-10), 22.7 (C-11), 36.3 (C-12), 43.6 (C-13), 50.6 (C-14), 23.5 (C-15),
27.8 (C-16), 82.3 (C-17), 20.0 (C-18), 16.2 (C-19), 171.3 (C]O,
acetate), 20.5 (CH3, acetate), 168.2 (C]O, amide). MS (EI): m/z (%):
348 (Mþ þ 1, 100), 287 (Mþ ꢀ CH3COOH, 18). Calc. for C21H33NO3
(347.4916): C, 72.58; H, 9.57; N, 4.03; found: C, 72.38; H, 9.39; N,
3.94.
3.2. In vitro cytotoxic activity
3.2.1. Materials
Fetal calf serum (FCS), L-glutamine and dimethyl sulfoxide
(DMSO) were purchased from Sigma Chemical Co. (St. Louis, MO,
USA). Penicillin G sodium and streptomycin sulfate were obtained
from Bio-Waste Co. (Wexford, Ireland). MTT was purchased from
Duchefa-Biochemie (Haarlem, Amsterdam, Netherlands). Sodium
bicarbonate was obtained from Merck Co. Inc. (USA).
3.2.2. Cell culture
Human hepatoma cell lines (HepG2) were supplied by Naval
American Research Unit, Egypt (NAMRU). Cells were propagated
3.4. Statistics
and maintained in RPMI-1640 medium with L-glutamine and sup-
plemented with 10% fetal calf serum for growth and 2% of the
maintenance medium [1% of 4% sodium bicarbonate and 1% anti-
Data were assessed by the method of analysis of ANOVA fol-
lowed by t-test, P < 0.05 was considered as statistically significant
(*), P < 0.01 was considered as statistically highly significant (**).
biotic mixture (1,000,000 units of penicillin
G sodium and
1,000,000 mg streptomycin sulfate in 100 ml deionized water)] in
References
75 cm2 tissue culture flasks.
[1] P. Pisani, D.M. Parkin, F. Bray, J. Ferlay, Int. J. Cancer 83 (1999) 18–29.
[2] G.D. Kirk, O.A. Lesi, M. Mendy, A.O. Akano, O. Sam, J.J. Goedert, P. Hainaut,
A.J. Hall, H. Whittle, R. Montesano, Hepatology 39 (2004) 211–219.
[3] G. Hosny, N. Farahat, H. Tayel, P. Hainaut, Cancer Lett. 264 (2008) 201–208.
[4] S. Deuffic-Burban, M.K. Mohamed, B. Larouze, F. Carrat, A. Valleron, J. Hepatol.
44 (2006) 455–461.
[5] T.H. Wu, R.L. Yang, L.P. Xie, H.Z. Wang, L. Chen, S. Zhang, Y. Zhao, R.Q. Zhang,
Cancer Lett. 232 (2006) 199–205.
[6] D.T.P. Trafalis, G.D. Geromichalos, C. Koukoulitsa, A. Papageorgiou,
P. Karamanakos, C. Camoutsis, Breast Cancer Res. Treat. 97 (2006) 17–31.
[7] D. Thibeault, J. Roy, P.D. Roy, D. Poirier, Bioorg. Med. Chem.16 (2008) 5062–5077.
3.2.3. Growth inhibition assay
The cytotoxic effect of compounds 7a, 10a, 12b, 18 and 23 was
done using MTT assay [25]. The human hepatocarcinoma (HepG2
cells) at approximately 80% confluence (i.e. logarithmically growing
cells) were selected for trypsinization, and stained with trypan blue
and their numbers were recorded. The percentage of cells that
resisted staining ought to be above 97%. The cells were seeded in
96-well microplates, after the cell concentrations were adjusted to
ˇ
ˇ
´
[8] Evgenija A. Djurendı´c, Marija N. Sakac, M.P. Zavis, A.R. Gakovıc, J.J. Canadia,
3 ꢃ103 cells/well in 100
ml RPMI-1640 culture medium and incu-
ꢀ
S.A. Andric, O.R. Klisuric, V.V. Kojic, G.M. Bogdanovı´c, K.M. Penovgasi, Steroids
73 (2008) 681–688.
bated at 37 ꢂC and 5% CO2 overnight. The cells were treated with
[9] M.N. Preobrazhenskaya, Chem. Heterocycl. Comp. 21 (1985) 13–24.
[10] M. Kidwai, R. Venkataramanan, R. Mohan, P. Sapra, Curr. Med. Chem. 9 (2002)
1209–1228.
[11] M.T. Cocco, C. Congiu, V. Lilliu, V. Onnis, Bioorg. Med. Chem. 15 (2007)
1859–1867.
either one of compounds 7a, 10a, 12b, 18 and 23 dissolved in olive
oil or DMSO in three concentrations (10, 50 and 100
incubated for 24 h, 48 h and 72 h. Then the cells were washed with
sterile phosphate buffer saline (PBS) and 100 l of the tetrazolium
dye (MTT) (0.5 mg/ml) solution was added to each well, and the
mg/ml) and re-
m
[12] A.E. Amr, A.M. Mohamed, S.F. Mohamed, N.A. Abdel-Hafez, A.G. Hammam,
Bioorg. Med. Chem. 14 (2006) 5481–5488.

3946
M. El-Far et al. / European Journal of Medicinal Chemistry 44 (2009) 3936–3946
[13] L. Brault, E. Migianu, A. Ne´guesque, E. Battaglia, D. Bagrel, G. KirschMaria, Eur.
J. Med. Chem. 40 (2007) 757–763.
[14] G.A. Elmegeed, W.W. Wardakhan, A.R. Baiuomy, Pharmazie 60 (2005)
[20] C. Camoutsis, P. Catsoulacos, J. Heterocycl. Chem. 25 (1988) 1617–1619.
[21] S. Kunihiro, Eur. Patent, EP0209078, 1987, http://www.freepatentsonline.
com/EP0209078A2.html.
328–333.
[22] S. Yoshida, A. Honda, Y. Matsuzaki, S. Fukushima, N. Tanak, A. Takagiwa,
Y. Fujimoto, H. Migazaki, G. Salen, Steroids 68 (2003) 73–83.
[23] IUPAC, Joint Commission on Biochemical Nomenclature (JCBN), Pure Appl.
Chem. 61 (1989) 1783–1822.
[24] IUPAC, Joint Commission on Biochemical Nomenclature (JCBN), Eur. J.
Biochem. 186 (1989) 429–458.
[15] R.M. Mohareb, S.I. Aziza, N.I. Abdel-Sayed, A.H. El-banna, J. Chem. Res. (1999)
101–128.
[16] J. Poza, M. Rega, V. Paz, B. Alonso, J. Rodrıguez, N. Salvador, A. Fernandez,
C. Jime`nez, Bioorg. Med. Chem. 15 (2007) 4722–4740.
[17] J. Cui, L. Huang, L. Fan, A. Zhou, Steroids 73 (2008) 252–256.
[18] C. Camoutsis, D. Trafalis, G. Pairas, A. Papageorgiou, Farmaco 60 (2005) 826–829.
[19] N. Krstic, M. Bjelakovic, M.M. Dabovic, L.B. Lorenc, V.D. Pavlovic, J. Serb. Chem.
Soc. 69 (2004) 413–420.
´
´
[25] T. Mosmann, J. Immunol. Methods 65 (1983) 55–63.
ˇ
[26] M. Theiszova, S. Jantova, J. Dragnova, P. Grznarova, M. Palouc, Biomed. Pap.
Med. Fac. Univ. Palacky Olomouc. Czech. Repub. 149 (2005) 393–396.