Steroidal esters of the aromatic nitrogen mustard 2-[4-N,N-bis(2- chloroethyl)amino-phenyl]butanoic acid (2-PHE-BU): Synthesis and in-vivo biological evaluation

Article abstract of DOI:10.1097/CAD.0b013e328357f687

On the basis of the results of in-silico predictions and in an effort to extend our structure-activity relationship studies, the aromatic nitrogen mustard 2-[4-N,N-bis(2-chloroethyl) amino-phenyl]butanoic acid (2-PHE-BU) was synthesized and conjugated with various steroidal alcohols. The resulting steroidal esters were evaluated for their in-vivo toxicity and antileukemic activity in P388-leukemia-bearing mice. The new derivatives showed significantly reduced toxicity and marginally improved antileukemic activity compared with free 2-PHE-BU. Nevertheless, they did not prove to be superior either to the template steroidal ester used for in-silico predictions or to previously synthesized steroidal esters of aromatic nitrogen mustards. The results obtained indicate that in-silico design predictions may guide the design and synthesis of new bioactive steroidal esters, but further parameters should be considered aiming at the discovery of compounds with optimum activity.

Full text of DOI:10.1097/CAD.0b013e328357f687

52 Preclinical report
Steroidal esters of the aromatic nitrogen mustard
2-[4-N,N-bis(2-chloroethyl)amino-phenyl]butanoic acid
(2-PHE-BU): synthesis and in-vivo biological evaluation
Ioanna C. Papaconstantinoua, Manolis A. Fousterisa, Anna I. Koutsoureaa,
Georgios N. Pairasa, Athanasios D. Papageorgioub and
Sotiris S. Nikolaropoulosa
On the basis of the results of in-silico predictions and in an
effort to extend our structure–activity relationship studies,
the aromatic nitrogen mustard 2-[4-N,N-bis(2-chloroethyl)
amino-phenyl]butanoic acid (2-PHE-BU) was synthesized
and conjugated with various steroidal alcohols. The
esters, but further parameters should be considered
aiming at the discovery of compounds with optimum
ꢀc
activity. Anti-Cancer Drugs 24:52–65
2012 Wolters
Kluwer Health | Lippincott Williams & Wilkins.
Anti-Cancer Drugs 2013, 24:52–65
resulting steroidal esters were evaluated for their in-vivo
toxicity and antileukemic activity in P388-leukemia-bearing
mice. The new derivatives showed significantly reduced
toxicity and marginally improved antileukemic activity
compared with free 2-PHE-BU. Nevertheless, they did not
prove to be superior either to the template steroidal ester
used for in-silico predictions or to previously synthesized
steroidal esters of aromatic nitrogen mustards. The results
obtained indicate that in-silico design predictions may
guide the design and synthesis of new bioactive steroidal
Keywords: antileukemic activity, aromatic nitrogen mustards, in-silico drug
design, ( )-2-phenylbutyric acid, steroidal esters
aLaboratory of Medicinal Chemistry, Department of Pharmacy, School of Health
Sciences, University of Patras, Patras and bLaboratory of Experimental
Chemotherapy, Theagenion Anticancer Institute, Thessaloniki, Greece
Correspondence to Manolis A. Fousteris, PhD, Laboratory of Medicinal
Chemistry, Department of Pharmacy, School of Health Sciences, University
of Patras, GR-26500 Rion, Patras, Greece
Tel: + 30 2610 969391; fax: + 30 2610 992776; e-mail: manolisf@upatras.gr
Received 19 February 2012 Revised form accepted 10 July 2012
Introduction
mustards. Among them, the generation of DNA-directed
alkylating agents through the chemical linkage of
nitrogen mustards with molecules of increased DNA-
binding affinity [10] and the synthesis of nitrogen
mustard prodrugs [11] led to interesting results. Another
strategy utilizes the conjugation of nitrogen mustards
with steroids, exploiting the role of the nuclear receptors
in the effective transport and localization of the alkylating
agents into the nucleus or the preferential sensitivity of
hormone-dependent tumors to relevant hormone-drug
conjugates [12,13]. Estramustine and prednimustine are
steroidal derivatives of alkylating agents that have been
used in combination therapies for the treatment of
various types of cancer [14].
Several chemotherapeutics act as DNA-damaging agents,
resulting in cell cycle arrest and cell death of the
uncontrollably proliferating cancer cells [1]. Among them,
bifunctional DNA alkylating agents [2], such as the nitrogen
mustards cyclophosphamide, chlorambucil (CHL), and
melphalan, have been used for years in the clinic, whereas
recently, the US FDA approved bendamustine [3] (Fig. 1)
for the treatment of chronic lymphocytic leukemia and
indolent B-cell non-Hodgkin’s lymphoma [4,5]. They have
an N,N-bis(2-chloroethyl)amine functional group and act as
reactive electrophiles through their intramolecular transfor-
mation into aziridinium cations. As a consequence, they bind
covalently to electron-rich atoms of DNA bases, forming
either intrastrand or interstrand crosslinks, and inhibit DNA
replication and transcription [6]. Although these drugs
remain some of the most commonly prescribed chemothera-
pies for the treatment of various solid and hematological
malignancies, particularly in combination with other classical
or targeted therapeutics in multiagent regimens [7,8], severe
side effects on normal tissues represent drawbacks of their
use. These are attributed to their low alkylation selectivity of
DNA bases because of their high inherent reactivity,
resulting in the nonspecific alkylation of other cellular
nucleophilic species such as amino acid residues or low-
molecular-weight thiols [9].
In this direction, our group has published a series of studies
related to steroidal esters of aromatic nitrogen mustards as
antineoplastic, especially antileukemic, agents [15]. Our
efforts have succeeded in the identification of potent and
promising derivatives (Fig. 2) with enhanced activity and
reduced toxicity compared with the corresponding nitrogen
mustards against in-vitro and in-vivo experimental tu-
mors [15,16]. Extensive structure–activity relationship
(SAR) studies have shown unique structural features of
both the steroidal part and the nitrogen mustard, which
contribute positively to the bioactivity of the target steroidal
esters [16–20]. Furthermore, recent 3D QSAR/CoMFA and
CoMSIA studies have led to the generation of related
models that indicate the influence of stereoelectronic and
Several approaches have been explored to reduce the
toxicity and increase the therapeutic efficacy of nitrogen
ꢀc
0959-4973
2012 Wolters Kluwer Health | Lippincott Williams & Wilkins
DOI: 10.1097/CAD.0b013e328357f687
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

New antileukemic steroidal esters Papaconstantinou et al. 53
Fig. 1
COOH
COOH
N
COOH
O
P
Cl
Cl
N
Cl
NH2
N
N
N
H
Cl
N
O
N
Cl
Cl
Cl
Cl
Chlorambucil
Cyclophosphamide
Melphalan
Bendamustine
DNA alkylating agents.
Fig. 2
(a)
Cl
OH
N
O
O
HO
Cl
O
O
HO
Cl
N
O
O
Cl
Estramustine
NHCOCH3
Prednimustine
(b)
Cl
NHCOCH3
Cl
H
N
Cl
O
N
O
N
NH
O
O
Cl
Cl
O
N
Cl
O
O
O
O
O
SE-2
SE-3
SE-7
LD50 (mg/kg)= 75
aT/CP388 %= 314
bCures: 0/6
LD50 (mg/kg)=90
T/CP388 %=333
Cures: 3/6
LD (mg/kg)=280
T/CP388 %=300
Cures: 4/6
(a) Structures of estramustine and prednimustine. (b) In-vivo toxicity and antileukemic activity of representative steroidal esters of aromatic nitrogen
mustards on P388-leukemia-bearing mice. aT/C%, the percent increase median life span of the drug-treated animals (T) versus corn-oil-treated
animals (C) Â 100; bCure, the number of long-term survival defined as mice alive for 90 days after tumor inoculation.
physicochemical parameters on the antileukemic activity of
target compounds [21]. Particularly, these studies have
indicated that the steric effect and the hydrophobic/
hydrophilic balance especially in the steroidal part of the
molecules probably determined their activity. Notably, it
was observed that the orientation of the alkylating agent
toward the surface of the B ring of the steroidal part was
correlated with increased activity. Moreover, the presence of
hydrophobic groups in rings B and D of the steroidal
skeleton and the insertion of hydrophilic/H-bond acceptor
groups at positions 7 and 17 of the steroidal core enhanced
the antileukemic activity. The reliability of both CoMFA
and CoMSIA models was evaluated and their predictive
ability on the activity of a test set of compounds was found
to be satisfactory. Furthermore, on the basis of the CoMFA
model proposed and using the de-novo ligand design
routine LeapFrog of SYBYL [22], a series of candidate
molecules with potentially optimal bioactivity have been
proposed, creating new challenges in further investigation of
this class of compounds.
On the basis of the aforementioned results and extending
our SAR studies, we aimed to examine whether the
incorporation of an in-silico predictive nitrogen mustard [21]
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

54 Anti-Cancer Drugs 2013, Vol 24 No 1
Fig. 3
and are uncorrected. The infrared (IR) spectra were
recorded on an FT-IR Jasco spectrophotometer (JASCO
Inc., Easton, Maryland, USA). 1H and 13C nuclear magnetic
resonance (NMR) spectra were recorded using a Brucker
DPX 400 MHz spectrometer (Bruker, Karlsruhe, Germany).
Chemical shifts are in parts per million (d) relative to
D
Cl
1
CHCl3 (7.26 and 77.0 for H and 13C-NMR, respectively)
B
N
as the internal standard and the following abbreviations are
used: singlet (s), doublet (d), triplet (t), quartet (q),
multiplet (m), doublet of doublets (dd), and quartet of
doublets (qd). Electron-spray ionization (ESI) mass spectra
were recorded at 30 V on a Micromass-Platform LC spectro-
meter (Waters Corporation, Milford, Massachusetts, USA)
using methanol or acetonitrile as a solvent.
O
Cl
O
(2-PHE-BU)
B, D ring modified or
unmodified steroidal skeletons
New steroidal esters of 2-PHE-BU. 2-PHE-BU, 2-[4-N,N-bis
(2-chloroethyl)amino-phenyl]butanoic acid.
3b-Hydroxy-androst-5-en-17-one was purchased from
Steraloids Inc. (Newport, Rhode Island, USA). 3b-
Hydroxy-cholest-5-en-7-one [24], 3b-hydroxy-androst-5-
en-7,17-dione [25], 3b-hydroxy-17b-acetamide-androst-
5-en-7-one [15], 3b-hydroxy-17a-aza-D-homo-androst-5-
en-7,17-dione [26], 3b-hydroxy-7a-aza-B-homo-cholest-
5-en-7-one [27], 3b-hydroxy-7a-aza-B-homo-androst-5-
en-7,17-dione [28], 3b-hydroxy-7a-aza-B-homo-17b-
acetamide-androst-5-en-7-one [15], 3b-hydroxy-7a,17a-
diaza-B,D-dihomoandrost-5-en-7,17-dione [28], 3b-hy-
droxy-17b-acetamide-androst-5-en [29], and 3b-hydroxy-
17a-aza-D-homo-androst-5-en-17-one [30] were prepared
according to the procedures reported in the literature.
into various simple and modified steroids might yield
steroidal esters with improved antineoplastic activity. The
selection of the nitrogen mustard that was synthesized and
used in this study was made on the basis of the predictive
antileukemic activity of its corresponding steroidal esters
and the synthetic accessibility of the intermediate and target
compounds. Thus, we designed and synthesized a new
series of steroidal esters containing the aromatic nitrogen
mustard 2-[4-N,N-bis(2-chloroethyl)amino-phenyl]butanoic
acid (2-PHE-BU) [23] (Fig. 3). 2-PHE-BU is structurally
related to other aromatic nitrogen mustards that have been
used extensively in our previous studies such as CHL, 4-
methyl-3-N,N-bis(2-chloroethyl)amino-benzoic acid (4-Me-
CABA), and especially with 4-N,N-bis(2-chloroethyl)amino-
phenylacetic acid (PHE). In an effort to evaluate the
predictive reliability of the Leapfrog routine and compare
the biological activity of structurally related compounds, a
variety of simple or modified steroidal skeletons were
conjugated with 2-PHE-BU, yielding a diverse library of
new steroidal esters. Herein, we report a straightforward
synthesis of this alkylating agent and selected target
steroidal esters as well as results of the evaluation of their
in-vivo acute toxicity and antileukemic activity.
2-(4-Nitrophenyl)butanoic acid (2)
A mixture (19.08ml) of concentrated nitric (9.54ml) and
sulfuric acid (9.54 ml) was added slowly to ( )-2-
phenylbutyric acid 1 (20g, 121.80 mmol) at 01C with
stirring. After completion of the addition, the cooling bath
was removed and the mixture was stirred vigorously at room
temperature for 2 h. Then, it was poured into an aqueous
solution of sodium hydrogen sulfate 20% (100 ml) and
extracted with dichloromethane (4 Â 50 ml). The com-
bined organic phases were washed with brine, dried over
sodium sulfate anhydrous, filtered, and evaporated under
reduced pressure to yield 2 as a crude oil (23.94 g), which
was used without further purification.
Materials and methods
Chemicals
All reagents were obtained commercially from Alfa Aesar
(Alfa Aesar GmbH & Co KG, Karlsruhe, Germany), Sigma-
Aldrich (Sigma-Aldrich, Steinheim, Germany), or Merck
(Merck KGaA, Darmstadt, Germany) and used without
further purification. Reactions involving moisture-sensitive
reactants were run in flame-dried glassware under an
atmosphere of argon. Benzene and toluene were purchased
in anhydrous forms and used without further purification.
Analytical TLC was performed on Merck Silica gel 60 F254
on precoated silica gel plates, with visualization under UV
light (254 and 365 nm) or/and by exposure to iodide vapors.
Flash column chromatography was performed on a silica gel
(SDS 60A, 40–63 mm). Melting points were determined on
an Electrothermal IA9200 digital melting point apparatus
(Electrothermal, Rochford, Essex, UK) in capillary tubes
Ethyl 2-(4-nitrophenyl)butanoate (3)
To a crude solution of 2 (23.94 g) in absolute ethanol
(64 ml), concentrated sulfuric acid (2.6 ml) was added
and the mixture was refluxed for 6 h under argon. Then,
the solvent was removed under reduced pressure and the
crude residue was diluted with dichloromethane and
extracted (4 Â 50 ml). The combined organic extracts
were washed with brine, dried over sodium sulfate
anhydrous, filtered, and evaporated under reduced
pressure to yield 3 as a crude oil (25.93 g), which was
used without further purification.
1H-NMR (400 MHz, CDCl3): d 0.90 (t, J = 7.3 Hz, 3H),
1.22 (t, J = 7.1 Hz, 3H), 1.74–1.88 (m, 1H), 2.05–2.22
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

New antileukemic steroidal esters Papaconstantinou et al. 55
(m, 1H), 3.56 (t, J = 7.7 Hz, 1H), 4.04–4.22 (m, 2H),
7.49 (d, J = 8.6 Hz, 2H), 8.18 (d, J = 8.6 Hz, 2H);
13C-NMR (100 MHz, CDCl3): d 12.02, 14.09, 26.81,
53.32, 61.12, 123.73(2 Â C), 128.94 (2 Â C), 146.53,
147.17, 172.74.
(10.5 ml) was added dropwise. After the completion of
the addition, the mixture was refluxed for 1 h protected
from moisture with a calcium chloride trap. Then, it was
carefully poured into ice-water with stirring and the two
phases were separated. The aqueous phase was washed
with benzene (3 Â 50 ml). The combined organic phases
were washed with brine (1 Â 50 ml), dried over sodium
sulfate anhydrous, filtered, and evaporated under reduced
pressure to yield 6 (22.76 g, 82%) as an oily residue,
Ethyl 2-(4-aminophenyl)butanoate (4)
A crude solution of 3 (25.93 g) in methanol (354 ml) was
hydrogenated over Pd 10% on activated carbon (338 mg)
at 50 psi for 1 h. Filtration of the catalyst through a celite
pad and evaporation of the filtrate under reduced
pressure yielded 4 as crude oil (21.52 g), which was used
without further purification.
1
which was used without further purification; H-NMR
(400 MHz, CDCl3): d 0.89 (t, J = 7.4 Hz, 3H), 1.22 (t,
J = 7.1 Hz, 3H), 1.68–1.81 (m, 1H), 2.0–2.11(m, 1H),
3.35 (t, J = 7.7 Hz, 1H), 3.59–3.66 (m, 4H), 3.67–3.74
(m, 4H), 4.02–4.20 (m, 2H), 6.66 (d, J = 8.8 Hz, 2H),
7.20 (d, J = 8.8 Hz, 2H); MS (ESI+ ) (m/z): 393.38
1H-NMR (400 MHz, CDCl3): d 1.07 (t, J = 7.4 Hz, 3H),
1.40 (t, J = 7.1 Hz, 3H), 1.86–2.00 (m, 1H), 2.23 (qt,
J = 15.1, 7.4 Hz, 1H), 3.52 (t, J = 7.7 Hz, 1H), 3.80 (br s,
2H), 4.21–4.38 (m, 2H), 6.84 (d, J = 8.5 Hz, 2H), 7.29 (d,
J = 8.4 Hz, 2H); 13C-NMR (100 MHz, CDCl3): d 12.08,
14.12, 26.68, 52.64, 60.38, 115.26 (2 Â C), 128.72 (2 Â C),
129.33, 145.06, 174.49.
[M + Na + K]+ , 332.28 [M + H]+
.
2-[4-N,N-bis(2-chloroethyl)amino-phenyl]butanoic acid
(2-PHE-BU)
A solution of 6 (22.76 g, 68.5 mmol) in hydrochloric acid
37% (50 ml) was refluxed for 5 h. Ice water was added
and the mixture was extracted with dichloromethane
(4 Â 50 ml). The combined organic phases were washed
with saturated aqueous sodium carbonate (1 Â 50 ml) and
brine (1 Â 50 ml), dried over sodium sulfate anhydrous,
filtered, and evaporated under reduced pressure to give
2-PHE-BU (17.71 g, 85%) as an oil; IR (KBr): nmax 1703,
1612, 1518, 815, 738 cm– 1; 1H-NMR (400 MHz, CDCl3):
d 0.91 (t, J = 7.4 Hz, 3H), 1.70–1.83 (m, 1H), 2.00–2.13
(m, 1H), 3.36 (t, J = 7.7 Hz, 1H), 3.62 (t, J = 7.3 Hz,
4H), 3.71 (t, J = 6.6 Hz, 4H), 6.64 (d, J = 8.8 Hz, 2H),
7.20 (d, J = 8.8 Hz, 2H); 13C-NMR (100 MHz, CDCl3): d
12.09, 26.10, 40.38 (2 Â C), 52.18, 53.48 (2 Â C), 111.98
(2 Â C), 127.33, 129.32 (2 Â C), 145.34, 180.53; MS
Ethyl 2-[4-N,N-bis(2-hydroxyethyl)amino-
phenyl]butanoate (5)
Ethylene oxide (24.2 ml, 484.55 mmol) was added to an
ice-cooled solution of 4 (21.52 g, 103.8 mmol) in a
mixture of glacial acetic acid (9.7 ml) and water
(48.4 ml) in an autoclave. The mixture was allowed to
warm up slowly at room temperature with stirring for
48 h. After cooling in an ice bath, the autoclave was
opened and the mixture was exposed to the air with
stirring overnight. The concentrated solution was poured
into ice-water and the mixture was extracted with
dichloromethane (5 Â 50 ml). The combined organic
phases were washed with saturated aqueous sodium
hydrogen carbonate (1 Â 50 ml) and brine (1 Â 50 ml),
dried over sodium sulfate anhydrous, filtered, and
evaporated under reduced pressure. The crude residue
was purified by silica gel flash column chromatography
(0–2% MeOH in dichloromethane) to yield 5 (24.6 g, 68%
over four steps) as an oil.
(ESI+ ) (m/z): 304.44 [M + H]+ , 268.41 [M – Cl]+
.
General procedure for the synthesis of steroidal esters
of 2-PHE-BU (7a–k)
A solution of 2-PHE-BU (3 mmol) in dry toluene or dry
benzene (15 ml) was treated with 2,4,6-trichlobenzoyl
chloride (3.6 mmol) and triethylamine (3.6 mmol) and
refluxed for 1.5 h under argon. In the above mixture,
a solution of the appropriate steroidal alcohol (2.5 mmol)
in dry toluene or dry benzene (15 ml) and 4-dimethyla-
minopyridine (2.5 mmol) was added. The reaction
mixture was refluxed additionally for the indicated time
under argon. The solvent was removed under reduced
pressure and the residual oil was dissolved in dichloro-
methane and washed successively with 5% aqueous HCl
(1 Â 50 ml), water (1 Â 50 ml), 5% aqueous NaHCO3
(1 Â 50 ml), and water (1 Â 50 ml). The organic phase was
dried over sodium sulfate anhydrous, filtered, and
evaporated under reduced pressure. The crude residue
was purified by silica gel flash column chromatography
using mixtures of MeOH in dichloromethane to yield the
desired steroidal ester.
1H-NMR (400 MHz, CDCl3): d 0.88 (t, J = 7.3 Hz, 3H),
1.21 (t, J = 7.1 Hz, 3H), 1.67–1.79 (m, 1H), 2.04 (ddd,
J = 22.5, 14.3, 6.7 Hz, 1H), 3.33 (t, J = 7.7 Hz, 1H), 3.54
(dd, J = 11.0, 6.4 Hz, 4H), 3.81 (dd, J = 9.7, 4.9 Hz, 4H),
4.00–4.18 (m, 2H), 6.64 (d, J = 8.6 Hz, 2H), 7.15 (d,
J = 8.7 Hz, 2H); 13C-NMR (100 MHz, CDCl3): d 12.00,
13.98, 26.54, 52.27, 55.10 (2 Â C), 60.35 (3 Â C), 112.18
(2 Â C), 126.94, 128.51 (2 Â C), 146.73, 174.66; MS
(ESI+ ) (m/z): 334.24 [M + K]+ , 318.30 [M + H]+
.
Ethyl 2-[4-N,N-bis(2-chloroethyl)amino-
phenyl]butanoate (6)
To an ice-cooled solution of 5 (24.6 g, 83.29 mmol) in dry
benzene (86 ml), a solution of freshly distilled phos-
phorus oxychloride (18 ml, 193.11 mmol) in dry benzene
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

56 Anti-Cancer Drugs 2013, Vol 24 No 1
Cholest-5-en-7-one-3b-yl 2-[4-N,N-bis(2-
chloroethyl)amino-phenyl]butanoate (7a)
J = 8.8 Hz, 2H), 7.16 (d, J = 8.8 Hz, 2H); 13C-NMR
(100 MHz, CDCl3): d 11.98, 12.16, 17.30, 20.74, 23.55,
25.78, 26.78, 27.29, 28.66, 35.85, 37.56, 37.77, 38.31, 40.41
(2 Â C), 43.55, 45.32, 46.48, 49.72, 52.45, 53.46 (2 Â C),
57.92, 72.10, 111.92 (2Â C), 126.43, 128.02, 129.05
(2 Â C), 145.13, 164.49, 169.94, 173.58, 201.15; MS
(ESI+ ) (m/z): 653.16 [M + Na]+, 631.18 [M + H]+,
595.09 [M – Cl]+.
The title compound was obtained using toluene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (1–2% MeOH in dichloromethane) to
yield 7a (370 mg, 62%), white solid; mp 611C; IR (KBr):
nmax 2951, 1730, 1672, 1265, 814, 738 cm-1; 1H-NMR
(400 MHz, CDCl3): d 0.68 (s, 3H), 0.86 (d, J = 1.9 Hz,
3H), 0.87 (d, J = 1.8 Hz, 3H), 0.90–0.93 (m, 6H), 1.19
(s, 3H), 3.33 (t, J = 7.7 Hz, 1H), 3.61–3.65 (m, 4H),
3.70–3.73 (m, 4H), 4.64–4.77 (m, 1H), 5.68 (d, J = 15.3,
1H), 6.64 (d, J = 8.7 Hz, 2H), 7.18 (d, J = 8.8 Hz, 2H);
13C-NMR (100 MHz, CDCl3): d 11.94, 12.16, 17.26,
18.85, 21.15, 22.52, 22.77, 23.81, 26.28, 26.81, 27.39,
27.96, 28.50, 35.68, 36.16, 38.29, 38.66, 39.45, 40.38
(2 Â C), 43.10, 45.39, 49.78, 49.96, 52.51, 53.58 (2 Â C),
54.78, 72.25, 112.12 (2 Â C), 125.11, 126.56, 128.18,
129.09 (2 Â C), 145.09, 163.78, 173.59, 201.84; MS
17a-Aza-D-homo-androst-5-en-7,17-dione-3b-yl 2-[4-
N,N-bis(2-chloroethyl)amino-phenyl]butanoate (7d)
The title compound was obtained using toluene as
a solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (0–2% MeOH in dichloromethane) to
yield 7d (448 mg, 56%), yellow oil; IR (KBr): nmax 3435,
2962, 1728, 1683, 1653, 1273, 817, 734; 1H-NMR
(400 MHz, CDCl3): d 0.88 (t, J = 7.4 Hz, 3H), 1.17 (s,
3H), 1.20 (s, 3H), 2.67–2.77 (m, 1H), 3.32 (t, J = 7.7 Hz,
1H), 3.60–3.63 (m, 4H), 3.69–3.72 (m, 4H), 4.63–4.75
(m, 1H), 5.71 (d, J = 15.5 Hz, 1H), 6.62 (d, J = 8.5 Hz,
2H), 7.05 (br s, 1H), 7.17 (d, J = 8.7 Hz, 2H); 13C-NMR
(100 MHz, CDCl3): d 12.15, 16.73, 20.55, 21.36, 22.58,
26.78, 27.16, 30.58, 35.69, 37.28, 38.23, 38.89, 40.40
(2 Â C), 41.29, 44.55, 48.95, 52.40, 53.43 (2 Â C), 53.79,
72.05, 111.89 (2 Â C), 126.30, 127.91, 129.03 (2 Â C),
145.14, 163.14, 172.89, 173.58, 200.15; MS (ESI+ ) (m/z):
(ESI+ ) (m/z): 710.35 [M + K]+
.
Androst-5-en-7,17-dione-3b-yl 2-[4-N,N-bis(2-
chloroethyl)amino-phenyl]butanoate (7b)
The title compound was obtained using toluene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (0–0.5% MeOH in dichloromethane) to
yield 7b (520 mg, 80%); oil; IR (KBr): nmax 2961, 1736,
641.26 [M + K]+
.
7a-Aza-B-homo-cholest-5-en-7-one-3b-yl 2-[4-N,N-bis(2-
chloroethyl)amino-phenyl]butanoate (7e)
1
1670, 1614, 1261, 815, 736 cm– 1; H-NMR (400 MHz,
CDCl3): d 0.87 (s, 3H), 0.88–0.93 (m, 3H), 1.21 (s, 3H),
2.74–2.84 (m, 1H), 3.32 (t, J = 7.7 Hz, 1H), 3.59–3.63
(m, 4H), 3.68–3.71 (m, 4H), 4.64–4.77 (m, 1H), 5.72 (d,
J = 15.4 Hz, 1H), 6.62 (d, J = 8.7 Hz, 2H), 7.17 (d,
J = 8.7 Hz, 2H); 13C-NMR (100 MHz, CD3OD): d 12.14,
13.67, 17.32, 20.45, 24.08, 26.75, 27.24, 30.58, 35.55,
35.86, 37.59, 37.75, 38.36, 40.31 (2 Â C), 44.24, 45.61,
47.76, 49.81, 52.38, 53.48 (2 Â C), 71.94, 112.03 (2 Â C),
126.39, 128.12, 129.03 (2 Â C), 144.96, 164.76, 173.55,
200.63; MS (ESI+) (m/z): 588.42 [M]+, 552.39 [M–Cl]+.
The title compound was obtained using benzene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (0–2% MeOH in dichloromethane) to
yield 7e (378 mg, 63%), orange foam; mp 851C; IR (KBr):
nmax 3447, 2955, 1730, 1660, 1267, 814, 738 cm– 1
;
1H-NMR (400 MHz, CDCl3): d 0.68 (s, 3H), 0.86 (d,
J = 2.1 Hz, 3H), 0.87 (d, J = 2.0 Hz, 3H), 0.89–0.91 (m,
6H), 1.23 (s, 3H), 3.25–3.38 (m, 2H), 3.61–3.64 (m, 4H),
3.70–3.73 (m, 4H), 4.64–4.77 (m, 1H), 5.69 (s, 1H), 5.79
(d, J = 18.3 Hz, 1H), 6.63 (d, J = 8.0 Hz, 2H), 7.17 (d,
J = 8.7 Hz, 2H); 13C-NMR (100 MHz, CDCl3): d 11.40,
12.17, 18.54, 19.35, 22.52, 22.78, 23.23, 23.78, 25.05,
26.80, 27.78, 27.97, 35.21, 35.52, 35.93, 38.58, 39.41,
40.47 (2 Â C), 41.40, 41.57, 41.87, 43.66, 48.95, 51.78,
52.51, 53.53 (2 Â C), 55.45, 55.72, 72.40, 111.98 (2 Â C),
122.44, 128.19, 129.10 (2 Â C), 145.14, 156.39, 167.30,
173.56; MS (ESI+ ) (m/z): 739.38 [M + K]+ , 723.39
17b-Acetamide-androst-5-en-7-one-3b-yl 2-[4-N,N-bis(2-
chloroethyl)amino-phenyl]butanoate (7c)
The title compound was obtained using toluene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (0–2% MeOH in dichloromethane) to
yield 7c (560 mg, 70%), white solid; mp 1061C; IR (KBr):
nmax 3306, 2949, 1728, 1668, 1614, 1292, 814, 736 cm–1
;
[M + Na]+ , 701.34 [M + H]+
.
1H-NMR (400 MHz, CDCl3): d 0.67 (s, 3H), 0.88 (t,
J = 7.4 Hz, 3H), 1.18 (s, 3H), 1.96 (s, 3H), 3.31 (t,
J = 7.7 Hz, 1H), 3.59–3.63 (m, 4H), 3.74–3.65 (m, 4H),
3.88 (q, J = 9.1Hz, 1H), 4.74–4.63 (m, 1H), 5.48 (d,
J= 9.0Hz, 1H), 5.67 (d, J= 12.9 Hz, 1H), 6.61 (d,
7a-Aza-B-homo-androst-5-en-7,17-dione-3b-yl 2-[4-N,N-
bis(2-chloroethyl)amino-phenyl]butanoate (7f)
The title compound was obtained using benzene as a
solvent. After the addition of the steroidal alcohol, the
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

New antileukemic steroidal esters Papaconstantinou et al. 57
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (0–4% MeOH in dichloromethane) to
yield 7f (637 mg, 80%), yellow oil; IR (KBr): nmax 3454,
2962, 1736, 1660, 1614, 1267, 819, 736; 1H-NMR
(400 MHz, CDCl3): d 0.86–0.90 (m, 6H), 1.25 (s, 3H),
3.31 (t, J=7.7 Hz, 1H), 3.60–3.64(m, 4H), 3.69–3.72 (m,
4H), 4.66–4.76 (m, 1H), 5.81 (d, J=18.8 Hz, 1H), 6.13 (br
s, 1H), 6.62 (d, J=7.4 Hz, 2H), 7.16 (d, J=8.8 Hz, 2H);
13C-NMR (100 MHz, CDCl3): d 11.97, 12.95, 19.06, 22.28,
22.40, 26.43, 26.55, 26.66, 30.24, 35.04, 35.46, 40.30 (2 Â C),
43.90, 46.57, 49.46, 49.78, 50.76, 52.20, 53.20 (2 Â C), 72.18,
111.70 (2 Â C), 122.31, 127.68, 127.76, 128.81 (2 Â C),
144.91, 167.72, 173.27, 218.32; MS (ESI+) (m/z): 641.20
[M + K]+, 625.20 [M + Na]+, 603.28 [M+H]+.
46.78, 49.08, 51.10, 52.48, 53.04, 53.56 (2 Â C), 72.64,
112.09 (2 Â C), 121.70, 121.75, 127.96, 129.08 (2 Â C),
145.12, 168.43, 168.74, 171.94, 173.47; MS (ESI+ ) (m/z):
640.25 [M + Na]+ , 618.33 [M + H]+
.
Androst-5-en-17-one-3b-yl 2-[4-N,N-bis(2-
chloroethyl)amino-phenyl]butanoate (7i)
The title compound was obtained using toluene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (dichloromethane) to yield 7i (470 mg,
78%), yellow foam; mp 591C; IR (KBr): nmax 2962, 1734,
1612, 1267, 814, 736; 1H-NMR (400 MHz, CDCl3): d
0.88–0.92 (m, 6H), 1.03 (s, 3H), 2.45 (dd, J = 19.2,
8.8 Hz, 1H), 3.32 (t, J = 7.7 Hz, 1H), 3.61–3.65 (m, 4H),
3.69–3.72 (m, 4H), 4.56–4.66 (m, 1H), 5.38 (dd, J = 14.8,
4.8 Hz, 1H), 6.63 (d, J = 7.9 Hz, 2H), 7.19 (d, J = 8.4 Hz,
2H); 13C-NMR (100 MHz, CDCl3): d 12.18, 13.52, 19.34,
20.30, 21.85, 26.89, 27.74, 30.75, 31.40, 31.45, 35.81,
36.72, 37.85, 38.09, 40.41 (2 Â C), 47.48, 50.12, 51.69,
52.64, 53.58 (2 Â C), 73.76, 112.06 (2 Â C), 121.79,
128.60, 129.12 (2 Â C), 139.91, 145.01, 173.76, 220.89;
7a-Aza-B-homo-17b-acetamide-androst-5-en-7-one-3b-
yl 2-[4-N,N-bis(2-chloroethyl)amino-phenyl]butanoate
(7g)
The title compound was obtained using benzene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (2–7% MeOH in dichloromethane) to
yield 7g (460 mg, 65%), yellow solid; mp 1851C; IR
(KBr): nmax 3273, 2962, 1728, 1660, 1612, 1273, 814,
MS (ESI+) (m/z): 612.35 [M + K]+ , 596.29 [M + Na]+,
+
574.37 [M + H], 538.35 [M – Cl]
.
1
740 cm-1; H-NMR (400 MHz, CDCl3): d 0.69 (s, 3H),
17b-Acetamide-androst-5-en-3b-yl 2-[4-N,N-bis(2-
chloroethyl)amino-phenyl]butanoate (7j)
0.88 (t, J = 7.4 Hz, 3H), 1.24 (s, 3H), 1.98 (s, 3H), 3.31
(t, J = 7.6 Hz, 1H), 3.60–3.64 (m, 4H), 3.69–3.72 (m,
4H), 4.66–4.77 (m, 1H), 5.32 (d, J = 8.9 Hz, 1H), 5.80 (d,
J = 19.4 Hz, 1H), 6.15 (br s, 1H), 6.62 (d, J = 8.8 Hz,
2H), 7.16 (d, J = 8.7 Hz, 2H); 13C-NMR (100 MHz,
CDCl3): d 11.51, 12.24, 19.44, 22.99, 23.49, 24.25, 26.84,
28.06, 35.27, 35.83, 40.52 (2 Â C), 41.52, 42.43, 44.00,
49.15, 51.56, 52.08, 52.51, 53.53 (2 Â C), 58.33, 72.34,
111.99 (2 Â C), 121.83, 127.94, 128.07, 129.12 (2 Â C),
145.18, 167.76, 170.16, 171.00, 173.62; MS (ESI+ ) (m/z):
The title compound was obtained using toluene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (0–1% MeOH in dichloromethane) to
yield 7j (438 mg, 73%), brown oil; IR (KBr): nmax 3445,
;
2962, 1726, 1649, 1267, 814, 736 cm–1 1H-NMR
(400MHz, CDCl3): d 0.69 (s, 3H), 0.89 (t, J = 7.52 Hz,
3H), 1.01 (s, 3H), 1.98 (s, 3H), 3.32 (t, J= 7.73 Hz, 1H),
3.60–3.65 (m, 4H), 3.69–3.72 (m, 4H), 4.55–4.63(m, 1H),
5.26 (d, J= 9.01 Hz, 1H), 5.35 (dd, J = 14.91, 4.67 Hz,
1H), 6.64 (d, J = 8.62 Hz, 2H), 7.19 (d, J= 8.81 Hz, 2H);
13C-NMR (100MHz, CDCl3): d 11.94, 12.19, 19.35, 20.54,
23.52, 23.63, 26.91, 27.55, 27.76, 28.68, 31.50, 32.03, 36.65,
36.74, 37.87, 38.11, 40.35 (2 Â C), 42.66, 49.97, 52.66,
52.76, 53.68 (2 Â C), 58.90, 73.91, 112.25 (2 Â C), 122.18,
128.19, 129.16 (2 Â C), 139.75, 169.97, 173.72; MS (ESI+)
(m/z): 617.45 [M + H], 605.42 [M + Na + H – Cl]+.
684.28 [M + K]+ , 668.28 [M + Na]+ , 646.17 [M + H]+
.
7a,17a-Diaza-B,D-dihomoandrost-5-en-7,17-dione-3b-yl
2-[4-N,N-bis(2-chloroethyl)amino-phenyl]butanoate
(7h)
The title compound was obtained using benzene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 2 h. The
crude residue was purified by silica gel flash column
chromatography (2–7% MeOH in dichloromethane) to
yield 7h (452 mg, 61%), gray oil; IR (KBr): nmax 3470,
17a-Aza-D-homo-androst-5-en-17-one-3b-yl 2-[4-N,N-
bis(2-chloroethyl)amino-phenyl]butanoate (7k)
1
2962, 1730, 1653, 1625, 1269, 844, 738 cm-1; H-NMR
The title compound was obtained using benzene as a
solvent. After the addition of the steroidal alcohol, the
reaction mixture was refluxed for additional 1.5 h. The
crude residue was purified by silica gel flash column
chromatography (0–2% MeOH in dichloromethane) to
yield 7k (414 mg, 69%), yellow foam; mp 1111C (dec.);
IR (KBr): nmax 3200, 2964, 1726, 1653, 1263, 814,
(400 MHz, CDCl3): d 0.88 (t, J = 8.2 Hz, 3H), 1.15 (s,
3H), 1.23 (s, 3H), 3.31 (t, J = 7.7 Hz, 1H), 3.60–3.64 (m,
4H), 3.69–3.72 (m, 4H), 4.64–4.72 (m, 1H), 5.81 (d,
J = 18.8 Hz, 1H), 6.59 (br s, 1H), 6.63 (d, J = 8.8 Hz,
2H), 7.16 (d, J = 8.7 Hz, 2H), 7.38 (br s, 1H); 13C-NMR
(100 MHz, CDCl3): d 12.17, 19.12, 21.42, 21.84, 21.88,
26.82, 30.80, 35.26, 38.06, 40.43 (2 Â C), 41.58, 44.45,
736 cm– 1
;
1H-NMR (400 MHz, CDCl3):
d
0.87
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

58 Anti-Cancer Drugs 2013, Vol 24 No 1
(t, J = 7.5 Hz, 3H), 0.97 (s, 3H), 1.13 (s, 3H), 3.30 (t,
J = 7.7 Hz, 1H), 3.58–3.61 (m, 4H), 3.66–3.70 (m, 4H),
4.53–4.61 (m, 1H), 5.33 (d, J = 16.8 Hz, 1H), 6.60 (d,
J = 8.6 Hz, 2H), 6.86 (br s, 1H), 7.17 (d, J = 8.6 Hz, 2H);
13C-NMR (100 MHz, CDCl3): d 12.00, 18.98, 19.70,
20.61, 21.63, 26.67, 27.43, 30.42, 30.96, 31.98, 36.54,
37.37, 37.61, 39.06, 40.27 (2 Â C), 47.58, 49.10, 52.37,
53.27 (2 Â C), 53.90, 73.43, 111.75 (2 Â C), 121.46,
128.18, 128.87 (2 Â C), 139.34, 144.82, 171.93, 173.51;
mice alive for 90 days after tumor inoculation. Each drug-
treated group included six mice, whereas the tumor
control group included eight mice; in each group, equal
numbers of male and female mice were used. Experi-
ments were initiated by implanting mice with tumor cells
according to the protocol of the National Cancer Institute
(USA) [31]. Treatments were given as a single dose
(LD10, day 1) and as an intermittent dose (LD10/2 Â 3,
days 1, 3 and 6). The experiments were terminated on
day 90. Statistical evaluation of the experimental data was
carried out using the Wilcoxon test.
MS (ESI+ ) (m/z): 627.28 [M + K]+ , 589.36 [M + Na]+
.
In-vivo experiments
Results and discussion
Chemistry
Compounds
For an intraperitoneal treatment, stock solutions of the
compounds used in this study were prepared immediately
before use. They were suspended in corn oil in the
desired concentration following initial dissolution in 5%
dimethylsulfoxide. This concentration by itself produced
no observable toxic effects.
2-PHE-BU was synthesized using the synthetic route,
which is shown in Scheme 1. Commercially available ( )-
2-phenylbutyric acid 1 was nitrated by treatment with a
mixture of concentrated nitric and sulfuric acid. The desired
p-nitro-derivative 2 was obtained in more than a 70% yield,
as estimated by 1H-NMR analysis of the crude mixture. The
latter was subsequently transformed into the ethyl ester 3
[32], which was immediately hydrogenated over a Pd
catalyst to the corresponding 4-aminophenyl derivative 4
[32]. The installation of the N,N-bis-chloroethyl moiety on
the aniline was carried out following a standard two-step
procedure. Particularly, the crude mixture of 4 obtained
upon prolonged treatment with ethylene oxide in an
autoclave yielded the N,N-bis-hydroxyethyl precursor 5 in
a 68% overall yield over four steps. The latter was
chlorinated by means of phosphorus oxychloride to the
N,N-bis-chloroethyl intermediate 6. Finally, acidic hydrolysis
of the ethyl ester 6 provided the desired aromatic nitrogen
mustard 2-PHE-BU in a very good yield.
Mice: BALB/c, DBA/2, and BDF1 mice of both sexes,
weighing 20–23 g, 6–8 weeks old, were used for toxicity
studies and antitumor evaluation. Mice obtained from the
experimental section of the Research Center of Thea-
genion Anticancer Hospital (Thessaloniki, Greece) were
kept under conditions of constant temperature and
humidity, in sterile cages, with water and food ad libitum.
The animals used in this study were cared for in
accordance with the Declaration of Helsinki and the
guidelines of the local ethics committee.
Tumors
Leukemia P388-bearing BDF1 (DBA/2 Â C57BL) mice
were used to evaluate the cytostatic effect. Lymphocytic
P388 leukemia was maintained in an ascitic form by an
injection of 106 cells at 7-day intervals into the peritoneal
cavity of DBA/2 mice.
Target steroidal esters of 2-PHE-BU were obtained using
the method of mixed anhydrides [33], which has been
used successfully by our group in the synthesis of related
compounds. Accordingly, the treatment of 2-PHE-BU
with 2,4,6-trichlobenzoyl chloride in the presence of
triethylamine in refluxing toluene or benzene yielded the
intermediate mixed anhydride (Scheme 2). Following a
one-pot procedure, the mixed anhydride was treated
subsequently with the appropriate steroidal alcohol in the
presence of 4-dimethylamino pyridine, producing the
target steroidal esters 7a–k in a very good yield.
Estimation of acute toxicity
The acute toxicity of the compounds was determined
following a single intraperitoneal injection into BALB/C
in groups of 10 mice/dose at three different dosages. The
mice were observed for 30 days and the therapeutic dose
of the compounds was determined after a graphical
estimation of the LD50 (30-day curves). The highest dose
used for a single treatment was equal to the LD10 value.
Biological evaluation
The results of the biological evaluation of the newly
synthesized esters of 2-PHE-BU with various 7-keto,
B-lactam, and slightly modified steroidal alcohols are
presented in Table 1. The newly synthesized steroidal
esters were first evaluated for their in-vivo acute toxicity.
As indicated by the LD50 and LD10 values in Table 1, the
target steroidal esters 7a–k (LD50 = 250 to >500 mg/kg)
induced significantly decreased toxicity effects compared
with the free nitrogen mustard 2-PHE-BU (LD50 = 160
mg/kg). Among the tested compounds, derivative 7g was
Antileukemic evaluation
For the survival experiments, the antileukemic activity of
the tested compounds against the above-mentioned
murine tumors was assessed from the oncostatic para-
meter T/C%, which is the increase median life span of
the drug-treated animals (T) excluding long-term survi-
vors versus corn-oil-treated controls (C), expressed as a
percentage. The other index of antileukemic activity
used was the number of long-term survivors defined as
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

New antileukemic steroidal esters Papaconstantinou et al. 59
Scheme 1
O
O
O
O
OH
OH
O
O
i
ii
iii
iv
NO2
NO2
3
NH2
4
1
2
O
O
O
O
O
OH
v
vi
N
N
N
HO
OH
Cl
Cl
Cl
Cl
5
6
2-PHE-BU
Reagents and conditions: (i) HNO3, H2SO4, 01C to rt, 2 h; (ii) CH3CH2OH, H2SO4, reflux, 6 h; (iii) H2, Pd 10% on activated C, CH3CH2OH, 1 h;
(iv) ethylene oxide, CH3COOH, rt, 72 h, 68% (over four steps); (v) POCl3, C6H6, reflux, 1 h, 82%; (vi) HCl 9N, reflux, 5 h, 85%.
Scheme 2
Cl
Cl
Cl
O
O
Cl
N
Cl
O
O
Cl
OH
i
+
N
Cl
O
Cl
Cl
Cl
Cl
2-PHE-BU
Cl
ii
HO
N
Cl
O
O
7a−k
Reagents and conditions: (i) NEt3, dry toluene or benzene, reflux, 1.5 h; (ii) 4-DMAP, dry toluene or benzene, reflux, 1.5–2 h, 56–80%. 4-DMAP,
4-dimethylaminopyridine
the most toxic; derivatives 7a, 7b, 7d, 7h, 7i, 7j, and 7k
showed moderate toxicity, whereas derivatives 7c, 7e, and
7f showed low toxicity values. These results are in
agreement with those of previous studies [19,20,33–35],
indicating that the conjugation of aromatic nitrogen
mustards either on simple or modified steroidal skeletons
results in the limitation of their undesired or toxic
effects, possibly because of the enhancement of their
lipophilicity and effective transport through the cell
membranes to the nucleus, close to DNA-specific sites.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

60 Anti-Cancer Drugs 2013, Vol 24 No 1
Table 1 Toxicity and antitumor activity of 2-PHE-BU and steroidal esters 7a–k on P388-leukemia-bearing mice
a
Compounds
Control
LD50 (mg/kg) LD10 (mg/kg) Treatment schedule (days) Dosage (mg/kg/day) T/Cb (%)
–
–
–
Corn oil
100
160
50
1
50
25
120
128
1, 3, 6
390
240
1
240
120
118
125
1, 3, 6
420
450
410
460
240
450
320
270
1
240
120
135
144
1, 3, 6
1
450
225
114
144
1, 3, 6
1
320
160
114
140
1, 3, 6
1
270
135
115
122
1, 3, 6
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

New antileukemic steroidal esters Papaconstantinou et al. 61
Table 1 (continued)
a
Compounds
LD50 (mg/kg) LD10 (mg/kg) Treatment schedule (days) Dosage (mg/kg/day) T/Cb (%)
> 500
280
1
280
140
128
118
1, 3, 6
250
175
1
175
87.5
100
105
1, 3, 6
310
220
1
220
110
128
147
1, 3, 6
410
250
1
250
125
110
118
1, 3, 6
320
210
1
210
105
125
130
1, 3, 6
310
180
1
180
90
135
138
1, 3, 6
aLD50 values were estimated graphically, where the percentage of deaths because of the toxicity of each dose is shown in the ordinate, whereas the doses administered
are indicated on the abscissae on semilogarithmic paper. For chemotherapy testing, the highest dose used for a single treatment was LD10. Therefore, the drugs in the
following experiments were compared at equitoxic doses.
bT/C = the percent increase median life span of the drug-treated animals (T) versus corn-oil-treated animals (C).
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

62 Anti-Cancer Drugs 2013, Vol 24 No 1
Next, the in-vivo antileukemic activity of the newly
synthesized compounds was assessed using the oncostatic
parameter T/C%. P388 lymphocytic leukemia-bearing
mice were treated with the tested compounds following
two different schedule treatments as presented
in Table 1. However, the intermittent administration of
the compounds at doses equal to LD10/2 on days 1, 3, and
6 after the implantation of the tumor cells was found to
be superior in all cases.
the 7-keto group in the enhancement of the antineoplas-
tic activity of these compounds, possibly through the
interaction of the conjugated steroidal enone system with
specific sites of the DNA that induce effective genotoxic
and cytotoxic effects.
As mentioned previously, the number and structural
diversity of steroidal skeletons that have been tethered to
2-PHE-BU were used to evaluate the predictive reliability
of the Leapfrog routine. Some of them were simple
steroids, whereas others had privileged structural features
that have been shown to be essential for the antineoplastic
activity of many alkylating steroidal esters, such as a
transformed D-six-membered lactam ring, an axial 17-
acetamido group, a conjugated 7-keto moiety, or an
expanded B-seven-membered lactam ring. Nevertheless,
the steroidal esters of 2-PHE-BU present low to moderate
activity, as determined by the corresponding T/C% values,
indicating a reduced antineoplastic potency of the nitrogen
mustard. This is also confirmed by the quite low potency of
the free nitrogen mustard (T/C 128%) compared with
CHL (T/C 155%) [37] or PHE (T/C 113%) [16] as has
already been reported in comparable studies. Furthermore,
2-PHE-BU (LD50 = 160mg/kg) was found to be approxi-
mately one order of magnitude less toxic than CHL
(LD50 = 24 mg/kg) [37] or other aromatic nitrogen mus-
tards such as PHE (LD50 = 20 mg/kg) [17] and 4-Me-
CABA (LD50 = 18 mg/kg) [38], which have been used in
our earlier works. These findings suggest that although
tethering of low potency alkylating agents on steroid
skeletons leads to steroidal esters with improved anti-
neoplastic activity, the unique structural features of the
nitrogen mustard determine crucially the efficacy of the
target steroidal esters. As we have reported earlier, the type
of chemical linkage between the steroidal part and the
nitrogen mustard is crucial for the activity of the target
compounds. An easily hydrolyzed chemical bond between
the alkylating agent and the steroidal skeleton is essential
for the bioactivity of the target steroidal esters, as noneasily
hydrolyzed conjugates such as amide conjugates have been
proved to be nonpotent derivatives [18].
The free nitrogen mustard 2-PHE-BU showed marginally
accepted antileukemic activity with a T/C equal to 128%,
slightly above the value of 125%, which is considered as the
limit for a potential chemotherapeutic agent according to
the National Cancer Institute criteria. However, the
majority of the steroidal esters showed improved antileu-
kemic activity compared with 2-PHE-BU, as indicated by
the corresponding T/C% values, confirming that this type
of chemical linkage between an aromatic nitrogen mustard
and steroids not only reduces the toxic effects but may also
lead to compounds with improved bioactivity.
Furthermore, the B,D-bilactam derivative 7h was found
to be the most potent among the tested compounds. As
we have reported elsewhere, the modification of the B
ring into a seven-membered lactam ring induces con-
formational changes in the steroidal part, which enhances
the genotoxic and cytotoxic activity [36] as well as the in-
vivo antileukemic potency [15] of the target steroidal
esters. Especially, B,D-bilactam derivatives have shown
diverse in-vivo toxicity and moderate to excellent
antileukemic activity depending on the structural fea-
tures of the conjugated nitrogen mustard [16,20,37].
These results possibly indicate an alternative mechanism
of action for the B-lactamic steroidal esters that does not
involve indispensably the interaction of the endocyclic –
NHCO – group in the B-ring with cellular sites crucial for
tumor growth, but with other specific binding sites that
are able to induce antileukemic responses. Nevertheless,
as can be observed from the T/C% values of derivatives
7e, 7f, and 7g, the presence of a B-seven-membered
lactam ring is not adequate to improve the antileukemic
activity of 2-PHE-BU as they appeared less active than
compound 7h. Obviously, the conjugation of an aromatic
nitrogen mustard with a properly modified steroidal
skeleton that evidently contributes to the enhancement
of the antileukemic potency may lead to steroidal esters
with improved antileukemic potency.
Furthermore, according to recent conformational analysis
studies [21], the structure of the side chain between the
carboxylic group and the aromatic ring of the nitrogen
mustard affects the orientation of the alkylating agent
toward the B-ring of steroidal skeleton, which is
considered an important stereochemical feature for the
antineoplastic activity. Possibly, this part of the molecule
plays a pivotal role in the bioactivity, as it might influence
the effective enzymatic hydrolysis of steroidal ester and
the release of nitrogen mustard in vivo. In this context,
comparing the more active steroidal esters of 2-PHE-BU
with the corresponding esters of PHE (SE-1 – SE-6)
(Table 2), it is obvious that the presence of the ethyl
group around the ester bond is detrimental for the
bioactivity, although this chemical modification was
However, steroidal esters 7b and 7c with an allylic 7-keto
group showed an equipotent, but slightly decreased
antileukemic activity compared with compound 7h.
Particularly, comparing the T/C% values for the following
pairs of steroidal esters 7b–7i, 7c–7j, and 7d–7k, it is
obvious that the introduction of a keto group at position
7 of the steroidal skeleton clearly enhances the antileu-
kemic activity. This observation confirms our previous
studies [19,33], which point to the fundamental role of
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

New antileukemic steroidal esters Papaconstantinou et al. 63
Table 2 Comparison of toxicity and antitumor activity of 2-PHE-BU, PHE, and their steroidal esters on P388-leukemia-bearing mice
a
Structure
Compounds
LD50 (mg/kg)
T/Cb (%)
Cures
2-PHE-BU
160
128c
0/6
PHE
20
113d
0/6
7b
SE-1
420
65
144c
321d
0/6
0/6
7c
SE-2
450
75
144c
314d
0/6
0/6
7d
SE-3
410
90
140c
333d
0/6
3/6
7f
SE-4
> 500
64
118c
133d
0/6
0/6
7g
SE-5
250
80
105c
221e
0/6
1/6
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

64 Anti-Cancer Drugs 2013, Vol 24 No 1
Table 2 (continued)
a
Structure
Compounds
LD50 (mg/kg)
T/Cb (%)
Cures
7h
SE-6
310
90
147c
153d
0/6
0/6
2-PHE-BU, 2-[4-N,N-bis(2-chloroethyl) amino-phenyl]butanoic acid; PHE, 4-N,N-bis(2-chloroethyl)amino-phenylacetic acid.
aLD50 values were estimated graphically, where the percentage of deaths because of the toxicity of each dose is shown in the ordinate, whereas the doses administered
are indicated on the abscissae on semilogarithmic paper. For chemotherapy testing, the dose used was LD10/2. Therefore, the drugs in the following experiments were
compared at equitoxic doses.
bT/C = the percent increase median life span of the drug-treated animals (T) versus corn-oil-treated animals (C).
cTreatment was given as an intermittent dose equal to LD10/2 on days 1, 3, and 6 for 2-PHE-BU and its steroidal esters.
dTreatment was given as an intermittent dose equal to LD10/2 on days 1, 5, and 9 for PHE and its steroidal esters.
eTreatment was given as an intermittent dose equal to LD10/2 on days 1, 4, and 7 for PHE and its steroidal esters.
pioneered by the LeapFrog prediction. This provides
evidence that the steric hindrance around the ester bond
might prevent the enzymatic hydrolysis and the release of
the active nitrogen mustard in vivo. It is also in
accordance with the low toxicity of 2-PHE-BU and can
lead to the assumption that a huge percentage of the
administered dose either of the steroidal ester or even of
the free nitrogen mustard bypasses the metabolic
activation without playing its role in vivo.
predictive reliability in the design of more potent
analogues. Thus, further studies that are ongoing and
considering the aforementioned results may shed more
light on the design and synthesis of new steroidal esters
with improved bioactivity.
Conclusion
In continuation of our studies in the field of antineoplastic
agents, new steroidal esters of 2-PHE-BU were designed
and synthesized on the basis of the results of our extended
SAR studies and guided by in-silico predicted simulations
derived from a recent 3D QSAR analysis. Particularly,
2-PHE-BU was synthesized from commercially available
Notably, steroidal ester SE-3 was used as a template
compound in the LeapFrog routine with restraining of the
nitrogen mustard groups [–N(CH2CH2Cl)2], whereas
stereoelectronic contour maps obtained by the CoMFA
analysis were inserted as input for the generation of a
hypothetical receptor cavity [21]. Operating on the
OPTIMIZE mode, which suggests further improvement
of the existing leads and after repeated structural changes
of the template, new optimized compounds were found
with lower binding energy to the receptor in comparison
with compound SE-3. Interestingly, after initial structur-
al modifications of SE-3, compound 7d arose and ranked
first among other optimized compounds. Despite this
promising prediction, compound 7d was found to be
almost more than two times less potent and significantly
less toxic than the parent compound SE-3 (Table 2).
These findings show that the stereoelectronic parameters
that are taken into account by the in-silico design are
probably not adequate to ensure reliable wide predic-
tions, as perhaps other issues such as metabolic
modifications interfere and contribute substantially
towards the bioactivity of the target compounds.
Furthermore, aspects related to the construction and
initial validation of the CoMFA model should be
reconsidered. For example, the number and the chemical
diversity of the inserted molecules in both the training
and the test set of compounds should be enriched, aiming
at the generation of a revised model with optimum
(
)-2-phenylbutyric acid following a straightforward
approach in six steps and an overall yield of 47%.
Subsequently, 2-PHE-BU was conjugated to various
modified or unmodified steroidal alcohols using the well-
established method of mixed anhydrides, producing the
steroidal esters 7a–k in very good yields (56–80%).
In general, the new alkylating steroidal esters showed
reduced toxicity and slightly improved antileukemic
activity against P388-leukemia-bearing mice compared
with the 2-PHE-BU. Nevertheless, they were not super-
ior to the already synthesized and structurally similar
steroidal esters of PHE, indicating that the enzymatic
hydrolysis and the release of the nitrogen mustard in vivo
are possibly disfavored because of the steric hindrance of
the ethyl group around the ester bond formed. This
becomes apparent on comparing the toxicity obtained and
the T/C% values for derivatives 7d and SE-3. It is worth
noting that SE-3 was used initially as a template in de-
novo design by the LeapFrog routine, whereas 7d was
generated during the optimization process as a candidate
compound with improved bioactivity. Nevertheless, the
biological results obtained in the present study clearly
indicate that except for the important stereoelectronic
requirements that are considered in and incorporated as
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

New antileukemic steroidal esters Papaconstantinou et al. 65
data into the LeapFrog routine, other parameters may
possibly be implicated and determine the bioactivity of
the target steroidal esters.
16 Papageorgiou A, Koutsourea AI, Arsenou ES, Fousteris MA, Mourelatos D,
Nikolaropoulos SS. Structure–anti-leukemic activity relationship study of
B- and D-ring modified and non-modified steroidal esters of chlorambucil’s
active metabolite. Anticancer Drugs 2005; 16:1075–1082.
17 Wampler GL, Catsoulacos P. Antileukemic effect of homoaza-steroidal ester
of p-bis(2-chloroethyl)amino phenyl acetic acid. Cancer Treat Rep 1977;
61:37–41.
18 Nikolaropoulos SS, Arsenou ES, Papageorgiou A, Mourelatos D. Antitumor
and cytogenetic effects of esteric (ASE) and amidic (ASA) steroidal
derivatives of p-bis(2-chloroethyl)aminophenylacetic acid (CAPA).
A comparative study. Anticancer Res 1997; 17:4525–4530.
19 Arsenou E, Fousteris M, Koutsourea A, Papageorgiou A, Karayianni V,
Mioglou E, et al. The allylic 7-ketone at the steroidal skeleton is crucial for the
antileukemic potency of chlorambucil’s active metabolite steroidal esters.
Anticancer Drugs 2004; 15:983–990.
We expect that a careful revision of the implicated
parameters such as the chemical diversity and the
number of the used compounds will lead to the
generation of new CoMFA and CoMSIA models that will
subsequently contribute toward the higher predicted
reliability of the in-silico design by software packages
such as LeapFrog. Relevant studies are under way, and
the results will be reported in due course.
20 Fousteris MA, Koutsourea AI, Arsenou ES, Papageorgiou A, Mourelatos D,
Nikolaropoulos SS. Structure–antileukemic activity relationship study of
B- and D-ring modified and nonmodified steroidal esters of 4-methyl-3-N,N-
bis(2-chloroethyl)amino benzoic acid: a comparative study. Anticancer
Drugs 2007; 18:997–1004.
21 Kapou A, Benetis NP, Durdagi S, Nikolaropoulos S, Mavromoustakos T. 3D
QSAR/CoMFA and CoMSIA studies on antileukemic steroidal esters
coupled with conformationally flexible nitrogen mustards. J Chem Inf Model
2008; 48:2254–2264.
Acknowledgements
I. Papaconstantinou was supported by a fellowship from
the ‘Cyprus State Scholarship Foundation’. The authors
thank Dr D. Vachliotis and Dr G. Magoulas (Center of
Instrumental Analysis, University of Patras, Greece) for
recording NMR and MS spectra.
22 Tripos Inc.. Sybyl molecular modeling software packages, ver. 6.8. St Louis,
MO: Tripos Inc.; 2001.
23 Levshina KV, Ivanushkina TA, Sergievskaya SI. Substituted N-bis (b-
chloroethyl)amines. Condensation of acetanilide with g-butyrolactone in the
presence of aluminum chloride. Zh Obshch Khimii 1966; 2:213–216.
24 Parish EJ, Scott AD. Selective oxidation of steroidal allylic alcohols using
3,5-dimethylpyrazole and pyridinium chlorochromate. J Org Chem 1983;
48:4766–4768.
25 Lardy H, Kneer N, Wei Y, Partridge B, Marwah P, Ergosteroids II. Biologically
active metabolites and synthetic derivatives of dehydroepiandrosterone.
Steroids 1998; 63:158–165.
Conflicts of interest
There are no conflicts of interest.
References
1
Skladanowski A, Bozko P, Sabisz M. DNA structure and integrity
checkpoints during the cell cycle and their role in drug targeting and
sensitivity of tumor cells to anticancer treatment. Chem Rev 2009;
109:2951–2973.
26 Arsenou EA, Koutsourea AI, Fousteris MA, Nikolaropoulos SS. Optimization
of the allylic oxidation in the synthesis of 7-keto-5-steroidal substrates.
Steroids 2003; 68:407–414.
27 Singh H, Padmanabhan S. Steroids and related studies. XVII. 3,7a-Diaza-
A,B-bishomo-4a-ene-4,7-dione azasteroid system and related studies.
Indian J Chem 1972; 10:355–357.
28 Koutsourea AI, Arsenou EA, Fousteris MA, Nikolaropoulos SS. Synthetic
approaches for the synthesis of a cytostatic steroidal B-D bilactam. Steroids
2003; 68:659–666.
29 Schmidt-Thome J. The Beckmarnn rearrangement of 20-oxo steroid oximes.
Ber 1955; 88:895–900.
30 Regan BM, Hayes FN. 17 and 17a-aza-D-homo steroids. J Am Chem Soc
1956; 78:639–643.
31 Goldin A, Sofina Z, Syrkin A. Experimental evaluation of antitumor drugs in
the USA and USSR and clinical correlations. Natl Cancer Inst Monogr
1980; 55:25–26.
32 Hamacher H, Bermann B. Potential antineoplastics, V: synthesis of N,N-
bis(2-chloroethyl)-5-[1-ethyl-2-(4-methoxyphenyl)-2-butenyl]-2-
methoxyanilin. Arch Pharm (Weinheim) 1981; 314:257–267.
33 Karayianni V, Papageorgiou A, Mioglou E, Iakovidou Z, Mourelatos D,
Fousteris M, et al. 7-Keto hybrid steroidal esters of nitrogen mustard:
cytogenetic and antineoplastic effects. Anticancer Drugs 2002; 13:637–643.
34 Catsoulacos P, Politis D, Wampler GL. A new steroidal alkylating agent with
improved activity in advanced murine leukemias. Cancer Chemother
Pharmacol 1979; 3:67–70.
2
3
4
5
6
7
8
9
Rajski SR, Williams RM. DNA cross-linking agents as antitumor drugs.
Chem Rev 1998; 98:2723–2795.
Leoni LM. Bendamustine: rescue of an effective antineoplastic agent from
the mid-twentieth century. Semin Hematol 2011; 48 (Suppl 1):S4–S11.
Keating MJ, Bach C, Yasothan U, Kirkpatrick P. Bendamustine. Nat Rev Drug
Discov 2008; 7:473–474.
Rummel MJ, Gregory SA. Bendamustine’s emerging role in the management of
lymphoid malignancies. Semin Hematol 2011; 48 (Suppl 1):S24–S36.
Noll DM, Mason TM, Miller PS. Formation and repair of interstrand cross-
links in DNA. Chem Rev 2006; 106:277–301.
Smolej L. Modern concepts in the treatment of chronic lymphocytic
leukemia. Hematology 2009; 14:249–254.
Kalaycio M. Bendamustine: a new look at an old drug. Cancer 2009;
115:473–479.
Suzukake K, Vistica BP, Vistica DT. Dechlorination of L-phenylalanine mustard
by sensitive and resistant tumor cells and its relationship to intracellular
glutathione content. Biochem Pharmacol 1983; 32:165–167.
10 Kapuriya N, Kapuriya K, Dong H, Zhang X, Chou T-C, Chen Y-T, et al. Novel
DNA-directed alkylating agents: design, synthesis and potent antitumor
effect of phenyl N-mustard-9-anilinoacridine conjugates via a carbamate or
carbonate linker. Bioorg Med Chem 2009; 17:1264–1275.
11 Hu L, Wu X, Han J, Chen L, Vass SO, Browne P, et al. Synthesis and structure-
activity relationships of nitrobenzyl phosphoramide mustards as nitroreductase-
activated prodrugs. Bioorg Med Chem Lett 2011; 21:3986–3991.
12 Rank P, Peter R, Depenbrock H, Eisenbrand G, Schmid P, Pitzl H, et al.
Preclinical activity of 17-[N-[N’-(2-chloroethyl)-N’-nitrosocarbamoyl]-L-
alanyl]-5-dihydrotestosterone (E91) against tumour colony forming units and
haematopoietic progenitor cells. Eur J Cancer 1999; 35:1009–1013.
13 Marquis JC, Hillier SM, Dinaut AN, Rodrigues D, Mitra K, Essigmann JM,
et al. Disruption of gene expression and induction of apoptosis in prostate
cancer cells by a DNA-damaging agent tethered to an androgen receptor
ligand. Chem Biol 2005; 12:779–787.
14 Ravery V, Fizazi K, Oudard S, Drouet L, Eymard JC, Culine S, et al. The use
of estramustine phosphate in the modern management of advanced prostate
cancer. BJU Int 2011; 108:1782–1786.
15 Koutsourea A, Fousteris MA, Arsenou ES, Papageorgiou A, Pairas GN,
Nikolaropoulos SS. Rational design, synthesis, and in vivo evaluation of the
antileukemic activity of six new alkylating steroidal esters. Bioorg Med Chem
2008; 16:5207–5215.
35 Anastasiou A, Catsoulacos P, Papageorgiou A, Margariti E. J Heterocycl
Chem 1994; 31:367–374.
36 Fousteris MA, Koutsourea AI, Lagonikakos NP, Arsenou EA, Spyridonidou
C, Mourelatos D, et al. Rational design, synthesis and in vitro evaluation of
three new alkylating steroidal esters. Med Chem 2006; 2:569–576.
37 Fousteris MA, Koutsourea AI, Arsenou ES, Papageorgiou A, Mourelatos D,
Nikolaropoulos SS. Structure–anti-leukemic activity relationship study of
B- and D-ring modified and non-modified steroidal esters of chlorambucil.
Anticancer Drugs 2006; 17:511–519.
38 Fousteris MA, Koutsourea AI, Arsenou ES, Papageorgiou A, Mourelatos D,
Nikolaropoulos SS. Antileukemic and cytogenetic effects of modified and
non-modified esteric steroidal derivatives of 4-methyI-3-bis(2-
chloroethyl)aminobenzoic acid (4-Me-CABA). Anticancer Res 2002;
22:2293–2300.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.