To determine the cytotoxicity of chlorambucil and one of its nitro-derivatives, conjugated to prasterone and pregnenolone, towards eight human cancer cell-lines

Article abstract of DOI:10.1016/j.ejmech.2008.11.014

Four ester prodrugs derived from the bifunctional alkylating agent chlorambucil, and one of its nitro-derivatives, 3-nitrochlorambucil conjugated to prasterone and pregnenolone, were synthesized and tested for their cytotoxic activity against eight human cell lines, using the standard MTT assay. A comparison between the esters and the controls, namely chlorambucil and 3-nitrochlorambucil would suggest that all four esters possess to varying degrees, specificity towards the breast adenocarcinoma cell line (MDA-mb468) than the other seven cells' lines tested. The overall findings are encouraging since it infers that these lipophilic esters not only have the ability to traverse specific cell membranes but also exhibit cytotoxicity towards most of the cell lines tested.

Full text of DOI:10.1016/j.ejmech.2008.11.014

European Journal of Medicinal Chemistry 44 (2009) 2944–2951
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
To determine the cytotoxicity of chlorambucil and one of its nitro-derivatives,
conjugated to prasterone and pregnenolone, towards eight human cancer
cell-lines
,
b
c
b
c
Leroy A. Shervington a , Nigel Smith , Emma Norman , Timothy Ward , Roger Phillips ,
*
Amal Shervington d
a School of Pharmacy and Pharmaceutical Sciences, University of Central Lancashire, Corporation Street, Preston PR1 2HE, UK
b Paterson Institute for Cancer Research, The University of Manchester, Manchester#M20 4BX, UK
c Cancer Research Unit, Tom Connors Cancer Centre, University of Bradford, Bradford#BD7 1DP, UK
d School of Forensic and Investigative Science, University of Central Lancashire, Preston PR1 2HE, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Four ester prodrugs derived from the bifunctional alkylating agent chlorambucil, and one of its nitro-
derivatives, 3-nitrochlorambucil conjugated to prasterone and pregnenolone, were synthesized and
tested for their cytotoxic activity against eight human cell lines, using the standard MTT assay. A
comparison between the esters and the controls, namely chlorambucil and 3-nitrochlorambucil would
suggest that all four esters possess to varying degrees, specificity towards the breast adenocarcinoma cell
line (MDA-mb468) than the other seven cells’ lines tested. The overall findings are encouraging since it
infers that these lipophilic esters not only have the ability to traverse specific cell membranes but also
exhibit cytotoxicity towards most of the cell lines tested.
Received 21 August 2007
Received in revised form
16 September 2008
Accepted 28 November 2008
Available online 9 December 2008
Keywords:
Bifunctional alkylating agents
Cytotoxicity
Ó 2008 Elsevier Masson SAS. All rights reserved.
Pregnenolone
Prasterone
Ester prodrugs
MTT assay
1. Introduction
The aim of this study was to determine the cytotoxicity of four
steroidal esters of prasterone and pregnenolone, linked to bifunc-
Chlorambucil, also known as 4-{4-[di(2-chloroethyl)amino]
phenyl}butyric acid, was first synthesised in 1953. It is an aromatic
nitrogen mustard and its cytotoxicity as a bifunctional alkylating
agent is due to its ability to cross-link between the bases of DNA [1].
Chlorambucil is still used as one of the front-line drugs in the
treatment of chronic lymphocytic leukaemia, malignant
lymphomas and advanced ovarian and breast carcinomas. Clinical
use of chlorambucil is however limited by its toxic side-effects [2,3].
At physiological pH chlorambucil is predominantly ionized [4] and
therefore may not rely solely on passive diffusion [5,6] in order to
traverse biological membranes and reach the targeted tumour cells.
It is also known to bind to plasma constituents such as albumin
which reduces its ability to effectively target tumour cells [7].
tional alkylating agents, chlorambucil and its nitro-derivative, 3-
nitrochlorambucil, against three breast adenocarcinoma cell lines,
namely, the estrogen-positive breast adenocarcinoma (MDA-
mb468) the estrogen-negative breast adenocarcinoma (MCF-7),
and the transfected breast adenocarcinoma (MDA NQ01). In addi-
tion to these cell lines, the study was also carried out to investigate
the cytotoxic effects of the esters on a broader spectrum of cell
types, namely, colon adenocarcinoma (Widr), brain posterior fossa
medullablastoma (Daoy), lung large cell carcinoma (H460), ovarian
adenocarcinoma (OVCA-3) and skin malignant melanoma (A375).
The presence of the strong withdrawing nitro-group, ortho to the
N,N-bis(2-chloroethyl) moiety [8] should reduce the reactivity of
alkylating moiety by limiting the formation of the highly reactive
aziridine intermediate [9]. The masking of the ionisable carboxylic
moiety of chlorambucil and the corresponding nitro-derivative, via
an ester bridge to a steroidal moiety, should increase the overall
lipophilicity of the nitrogen mustard and possibly enhance the
compounds ability to traverse cell membranes [10]. Furthermore,
* Corresponding author. Tel.: þ44 (0) 1772 893519.
E-mail address: lashervington@uclan.ac.uk (L.A. Shervington).
0223-5234/$ – see front matter Ó 2008 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2008.11.014

L.A. Shervington et al. / European Journal of Medicinal Chemistry 44 (2009) 2944–2951
2945
Table 1
the steroidal moiety could possibly enhance the selectivity of the
alkylating moiety by targeting tumour cells that have hormone
receptors.
Steroid esters of 4-{4-[di(2-chloroethyl)amino]phenyl}butanoic acid (chlorambucil).
Cl
Cl
R
Prednimustine and bestrabucil are two clinically available
hormonal alkylating agents that were specifically designed to
enhance selectivity of the chlorambucil moiety. Prednimustine,
derived from the combination of chlorambucil and prednisolone
linked via an ester bridge, was designed to mask the carboxylic acid
and to selectively target tumour cells with specific hormone
receptors [11]. It was proved to be effective in the management of
a number of leukaemias and lymphomas [12,13]. Bestrabucil, the
benzoate of the estradiol-chlorambucil conjugate, was initially
developed as a target orientated anticancer agent for the therapy of
estrogen receptor-positive breast cancer [14]. Research data sup-
porting the targeting nature of this conjugate has shown that
concentration of bestrabucil and some of its derivatives, accumu-
late 5–10 times higher in tumour tissue of the sensitive xenografts,
than in blood and muscle tissue [15].
CO2R1
N
(CH2)3
Compound
R1
R
Formula
33H45Cl2NO3
C33H44Cl2N2O5
C35H49Cl2NO3
C35H48Cl2N2O5
C14H19Cl2NO2
C14H18Cl2N2O4
C1
C2
C3
C4
C5
C6
Prasterone
Prasterone
Pregnenolone
Pregnenolone
H
H
NO2
H
NO2
H
C
H
NO2
C1 is chlorambucil ester of prasterone; C2 is 3-nitrochlorambucil ester of prasterone;
C3 is chlorambucil ester of pregnenolone; C4 is 3-nitrochlorambucil ester of preg-
nenolone; C5 is chlorambucil; C6 is 3-nitrochlorambucil.
3.80 ꢁ 0.30 and 3.70 ꢁ 0.35, respectively, and were in close agree-
ment with the estimated values (Table 2). The measured partition
coefficient of chlorambucil was also found to be in close agreement
to previous studies [10] The determined alkylating activity of 3-
nitrochlorambucil ester of prasterone (C2) and 3-nitrochlorambucil
ester of pregnenolone (C4) was significantly lower, presumably due
to the electron withdrawing effect of the nitro-group attached ortho
to the N,N-bis (2-chloroethyl) moiety reducing the activity of the
alkylating moiety. In addition, the determined alkylating activity of
the esters was much lower than that of the alkylating activity of
chlorambucil and 3-nitrochlorambucil.
Although prasterone and pregnenolone (Fig. 1), are known to be
important steroidal intermediates in the biosynthesis of various
hormones [16,17], relatively few studies have explored their
potential as carriers. In addition, because there is growing interest
in both prasterone and pregnenolone [18–26], it would be useful to
explore the potential of these steroidal compounds linked to
chlorambucil and the corresponding nitro-derivative, 3-nitro-
chlorambucil as potential carriers. The four esters in question
(Table 1) were chlorambucil and the 3-nitrochlorambucil esters of
prasterone (C1 and C2), and chlorambucil and the 3-nitro-
chlorambucil esters of pregnenolone (C3 and C4).
A similar phenomenon was observed in earlier studies [10]. It
was found that on increasing the complexity of chlorambucil esters
by the addition of carbons, a sharp decrease in their alkylating
activity resulted. Interestingly, there was a rough correlation
between the alkylating activity and the cytotoxicity of the
compounds with a number of the cell lines tested.
2. Chemistry
The esters were prepared by treating chlorambucil separately
with prasterone and pregnenolone in the presence of DCC and
DMAP at room temperature in dichloromethane [27] (Scheme 1).
The nitro-derivatives were prepared in a similar manner after
obtaining 3-nitrochlorambucil (C6) by treating chlorambucil at 0 ꢀC
with nitronium tetrafluoroborate in acetonitrile [8]. The four esters
were tested in vitro for their cytotoxic effects against eight human
cell lines using a growth inhibition assay. Chlorambucil (C5) and 3-
nitrochlorambucil (C6) were used as controls in the investigation.
Prior to measure the cytotoxic effect of the esters on the eight
human cancer cell lines, measurements involving the determina-
tion of the partition coefficients, the hydrolyzing properties and the
alkylating activities of these four esters together with chlorambucil
and 3-nitrochlorambucil were carried out. The measured partition
coefficients for the four esters were found to be greater than 8.00
and were compared with values obtained by the application of
a computer aided program. The estimated partition coefficients for
the four esters ranged from 8.54 ꢁ 0.46 to 9.18 ꢁ 0.42, confirming
the highly lipophilic nature of the esters. The partition coefficients
of chlorambucil and 3-nitrochlorambucil were found to be
3. Pharmacology and pharmacokinetics
A HPLC method was developed to monitor the hydrolysis of the
esters and to confirm the esters possible role as prodrugs. Although
the present work was centered on chlorambucil and 3-nitro-
chlorambucil esters of both prasterone and pregnenolone, during
the developmental phase of the HPLC method, two additional
esters were also analyzed, namely 3,6-dinitrochlorambucil esters of
prasterone and pregnenolone. The retention times of these esters
together with the esters of interest are shown in Table 3.
The method was initially developed to monitor, separately, the
appearance of either chlorambucil, 3-nitrochlorambucil or 3,6-
dinitrochlorambucil, however, it was found that observing the loss
in peak area of the ester over time was a more accurate and reliable
procedure for monitoring hydrolysis. The validation of the HPLC
method also included the simultaneous separation of all the esters
(see Fig. 2) in addition to the separate analysis of each of the esters.
Fig. 3 shows an example of a chromatogram of 3-nitrochlorambucil
ester of prasterone (C2). Conveniently, the hydrolysis is readily
monitored by the decrease in peak area corresponding to the ester
over time. On storing both the prasterone and pregnenolone
derivatives of the nitrochlorambucil at 37 ꢀC in phosphate buffer
saline (pH 7.4) over a period of forty five days in sealed sample
bottles resulted in 46% and 48% degradations, respectively, of these
esters. This was also supported by infrared spectroscopy involving
the decrease in the carbonyl ester with time. Interestingly, it has
been reported that the half-life of chlorambucil in PBS is approxi-
mately 60 min, with the formation of the dihydroxy derivative 4-
{4-[di(2-hydroxyethyl)amino] phenyl}butyric acid. Fortunately,
O
O
HO
HO
Prasterone
Pregnenolone
Fig. 1. Structures of prasterone and pregnenolone.

2946
L.A. Shervington et al. / European Journal of Medicinal Chemistry 44 (2009) 2944–2951
N(CH2CH2Cl)2
N(CH2CH2Cl)2
NO2
(i)
CO2H
CO2H
Chlorambucil (C5)
3-nitrochlorambucil (C6)
(ii)
(iii)
O
N(CH2CH2Cl)2
O
N(CH2CH2Cl)2
NO2
O
O
O
O
Chlorambucil ester of prasterone (C1)
3-nitrochlorambucil ester of prasterone (C2)
O
N(CH2CH2Cl)2
O
N(CH2CH2Cl)2
NO2
O
O
O
Chlorambucil ester of pregnenolone (C3)
O
3-nitrochlorambucil ester of pregnenolone (C4)
Scheme 1. Synthesis of C1–C4 and C6. Reagents and conditions: (i) NO2BF4/CH3CN (1.2 equiv.) at 0 ꢀC under argon. (ii) Chlorambucil/CH2Cl2/DCC/DMAP in the presence of
prasterone or pregnenolone at room temperature. (iii) 3-Nitrochlorambucil/CH2Cl2/DCC/DMAP in the presence of prasterone or pregnenolone at room temperature.
this degradation is not observed in either blood or plasma [28]. It
was found that on incubating separately, all four esters (C1–C4) at
37 ꢀC in the presence of porcine liver esterase, the rate of hydrolysis
ranged from 24 to 37% and 83 to 94%, over a 2 h and a 24 h periods,
respectively. Grieg et al. reported the hydrolysis of various chlor-
ambucil ester derivatives using non-specific plasma esterases
measured in vivo, using rat plasma, resulted in rapid cleavage
ranging from less than 10 s to 5 min, depending on the size of the
ester [10]. However, Grieg et al. also determined the rate of
hydrolysis of chlorambucil tertiary butyl ester in freshly prepared
rat liver and blood, and found that this ester was relatively stable
with half-lives of approximately 2 h and 7 h, respectively [29]. This
phenomenon was attributed to the steric hindrance of the tertiary
butyl group. Interestingly, the esters involved in this study have
similar stereochemistry’s to the chlorambucil tertiary butyl esters,
and after incubating them in fresh human plasma over a period
24 h, the percent of the prasterone and pregnenolone derivatives
that had hydrolysed ranged from 51 to 55 and 37 to 39%,
respectively.
The chlorambucil ester of prasterone (C1) showed activity
against all the cell-lines with the exception of (Widr). The highest
activity for this ester was against (MDA-mb468) with an IC50 of
Table 2
Log P partition coefficient, and in vitro alkylating activity of chlorambucil and
chlorambucil derivatives.
Table 3
Compound
Determined
Estimated partition
coefficientb
Alkylating activity
The assignment of the peaks in the chromatogram of all the esters.
partition coefficienta
(% of chlorambucil)c
Assignment
of peaks
Compound
Retention
time (min)
C1
C2
C3
C4
C5
C6
>8.00
>8.00
>8.00
>8.00
8.64 ꢁ 0.44
8.57 ꢁ 0.46
9.25 ꢁ 0.41
9.18 ꢁ 0.42
3.70 ꢁ 0.35
3.63 ꢁ 0.37
70
8
18
4
100
38
A
B
C
D
E
F
3,6-Dinitrochlorambucil ester of prasteronea
3-Nitrochlorambucil ester of prasterone
3,6-Dinitrochlorambucil ester of pregnenolonea
Chlorambucil ester of prasterone
3-Nitrochlorambucil ester of pregnenolone
Chlorambucil ester of pregnenolone
5.74
7.17
8.64
10.08
11.39
16.88
3.80 ꢁ 0.30
3.70 ꢁ 0.35
C1 is chlorambucil ester of prasterone; C2 is 3-nitrochlorambucil ester of prasterone;
C3 is chlorambucil ester of pregnenolone; C4 is 3-nitrochlorambucil ester of preg-
nenolone; C5 is chlorambucil; C6 is 3-nitrochlorambucil.
The hydrolysis of esters C1–C4 was monitored separately using porcine liver esterase
a
Determined values using the method described by Leo et al. [45].
Estimated values obtained from a ACD/Log P software package.
Measured in vitro at 37 ꢀC at concentrations ranging between 0.1 and 1.0 mM.
and fresh human plasma.
b
a
These esters were used in another HPLC study and were not analyzed for their
c
cytotoxic properties. In addition, neither of these esters were subjected to hydrolysis
using porcine liver esterase or fresh human plasma.
Chlorambucil is assigned 100% alkylation.

L.A. Shervington et al. / European Journal of Medicinal Chemistry 44 (2009) 2944–2951
2947
Fig. 2. A chromatogram of the simultaneous separation of all the esters indicated in Table 3. RP-HPLC conditions: mobile phase was acetic acid (50 ml of a 2% solution) made upto
a litre with acetonitrile, pH ¼ 5.2 at a flow rate of 1.2 ml/min. A C18 Symmetry Waters column (250 mm ꢂ 4.6 mm ID) fitted with a guard column (3 cm ꢂ 4.6 mm ID) was used and
the wavelength was set at 254 nm.
5.48
cytotoxic against (MDA-mb468) and the transfected (MDA-NQO1)
cell line, with IC50’s of 7.25 and 10.60 M, respectively. The chlor-
ambucil ester of pregnenolone (C3) was cytotoxic to varying
degrees against all of the cell-lines, the least being the MCF-7 cell
line which expresses estrogen and progesterone receptors [30],
however, it was most active against MDA-mb468 cell line with an
IC50 of 10.26 mM. The 3-nitrochlorambucil ester of pregnenolone
(C4) was cytotoxic against (MDA-mb468), the transfected (MDA-
NQO1) and the (MCF-7) cell-lines, with IC50’s of 19.34, 32.56 and
m
M. The 3-nitrochlorambucil ester of prasterone (C2) was only
compounds, since the added flavoenzyme NAD(P)H: quinone
oxidoreductase (DT-diaphorase), via an oxygen independent
pathway [31], catalyzes two-electron reduction of quinones and
nitro-compounds to hydroquinones and nitroso compounds,
respectively [32]. In addition, there is an over expression of DT-
diaphorase throughout many tumor tissues [33,34] and its activity
and gene expression have been found to be up regulated in
comparison to tissue levels in lung, colon, liver and breast tumors
[35,36]. DT-diaphorase is therefore an attractive target for the
development of bioreductively activated chemotherapeutic agents
[37]. Furthermore, DT-diaphorase is also capable of detoxifying
certain potentially carcinogenic xenobiotics [38]. However, the
determined cytotoxicity of the tested compounds was found to be
between 1.5 and 2 fold greater in the (MDA-mb468) compared to
the (MDA-NQO1) cell line. These results did not therefore provide
conclusive evidence that the esters (C2 and C4) were reduced to the
corresponding nitroso or hydroxylamine derivatives but may
suggest that the transfected cell line exhibits greater resistance by
possibly detoxifying the compounds via the DT-diaphorase
dependent pathway, whereas, the (MDA-mb468) cell line, pos-
sessing no functional diaphorase, is less resistant. The esters were
tested for their cytotoxicity against the breast adenocarcinoma
(MCF-7) cell line in order to confirm whether they displayed any
specificity against estrogen dependent cell lines. However, mild
cytotoxicity was only observed with esters (C1, C3 and C4),
m
68.30
were used as controls in the study involving the eight cell-lines, and
afforded IC50’s ranging from 0.65 to 7.13 M and 1.16 to 13.04 M,
mM, respectively. Both chlorambucil and 3-nitrochlorambucil
m
m
respectively. The effect of the nitro-group was clearly evident when
comparing the cytotoxicity of chlorambucil (C5) and 3-nitro-
chlorambucil (C6), where the reduced activity of 3-nitro-
chlorambucil was attributed to the effect of the nitro-group ortho to
N,N-bis(2-chloroethyl) moiety and closely correlated with the in
vitro determination of the alkylating activity using p-nitrobenzyl
pyridine. As a control measure, the cytotoxicity of both prasterone
and pregnenolone was also carried out on the cell lines and was
found to be non-cytotoxic at concentrations below 100
data can be found in Table 4.
mM. The IC50
4. Results and discussion
affording IC50’s of 72.5, 90 and 68.3
mined IC50’s for chlorambucil and nitrochlorambucil were 4.5 and
23 M, respectively and would therefore imply that esters (C1, C3
and C4) do not exhibit any significant specificity for the estrogen
dependent cell line.
mM, respectively. The deter-
A number of interesting findings were obtained from investi-
gating the activity of the compounds against the tested cell lines.
The most significant finding involved the breast adenocarcinoma
(MDA-mb468) cell line, where all the compounds exhibited cyto-
m
A comparison between the IC50 data obtained for prasterone
and pregnenolone esters, would suggest that prasterone esters are
in general, more cytotoxic and the data roughly correlates with that
toxic activity ranging from 0.65 mM to 19.34 mM. The reason for
using the transfected breast adenocarcinoma (MDA-NQO1) cell line
was to assess the possible bioreductive potential of the nitro-
Fig. 3. An example of a chromatogram of one of the esters, namely, 3-nitrochlorambucil ester of prasterone mentioned in Table 3 with a retention time of 7.17 min. RP-HPLC
conditions: mobile phase was acetic acid (50 ml of a 2% solution) made upto a litre with acetonitrile, pH ¼ 5.2 at a flow rate of 1.2 ml/min. A C18 Symmetry Waters column
(250 mm ꢂ 4.6 mm ID) fitted with a guard column (3 cm ꢂ 4.6 mm ID) was used and the wavelength was set at 254 nm.

2948
L.A. Shervington et al. / European Journal of Medicinal Chemistry 44 (2009) 2944–2951
Table 4
The IC50 values for chlorambucil and the chlorambucil derivatives on eight human cell-lines.
Cell line IC50 m
M
Widr
Daoy
MDA468
MDA NQ01
MCF-7
H460
OVCA-3
A375
C1
C2
C3
C4
C5
C6
>100
>100
19.61 (ꢁ6.9)
>100
5.48 (ꢁ5.6)
7.25 (ꢁ3.1)
10.26 (ꢁ6.7)
19.34 (ꢁ3.4)
0.65 (ꢁ3.0)
1.16 (ꢁ6.3)
11.83 (ꢁ1.4)
10.60 (ꢁ5.7)
20.53 (ꢁ2.1)
32.56 (ꢁ4.3)
0.79 (ꢁ2.9)
3.75 (ꢁ4.0)
72.5 (ꢁ2.0)
>100
90 (ꢁ1.5)
68.3 (ꢁ2.5)
4.5 (ꢁ1.3)
23 (ꢁ3.1)
11.85 (ꢁ1.5)
>100
82.50 (ꢁ1.2)
>100
61.42 (ꢁ8.9)
>100
86.30 (ꢁ2.0)
41.90 (ꢁ1.3)
36.16 (ꢁ1.4)
88.80 (ꢁ3.7)
74.21 (ꢁ7.2)
>100
>100
>100
>100
>100
7.13 (ꢁ5.1)
13.04 (ꢁ3.9)
1.36 (ꢁ1.5)
5.30 (ꢁ3.7)
2.19 (ꢁ6.6)
9.53 (ꢁ1.1)
8.92 (ꢁ2.7)
15.56 (ꢁ2.1)
2.58 (ꢁ6.2)
9.46 (ꢁ7.1)
All values are given in mM, however, the values shown in brackets are the standard deviations for the % growth inhibition.
C1 is chlorambucil ester of prasterone; C2 is 3-nitrochlorambucil ester of prasterone; C3 is chlorambucil ester of pregnenolone; C4 is 3-nitrochlorambucil ester of preg-
nenolone; C5 is chlorambucil; C6 is 3-nitrochlorambucil.
obtained for their alkylating activity. Furthermore, due to the
presence of the nitro-group in esters (C2 and C4), the determined
alkylating activity was significantly reduced compared to esters (C1
and C3) which roughly correlated with the activity obtained for
these esters in all but two of the cell lines tested. Generally, when
a strong electron withdrawing group such as a nitro-group is
attached ortho to the N,N-bis (2-chloroethyl) moiety, the potential
to alkylate is reduced. This phenomena was observed when
comparing compounds C1 with C2, C3 with C4, C5 with C6, in
addition to C1 with C4 and C2 with C3. However, there is no
correlation, as one would expect, between the magnitude of
alkylation and the degree of cytotoxicity in the cell lines tested.
6. Experimental section
6.1. Materials
The IR spectra for the compounds were recorded as either
a chloroform solution or as an oil using a Genesis FTIR spectrometer.
1H NMR were recorded in deuterated chloroform using a Bruker
drx500 (500 MHz), a amx360 (360 MHz), and a Avance dpx300
(300 MHz) instrument. Tetramethylsilane was used as internal
standard, and J values are given in Hz. 13C NMR spectra were
recorded using the listed NMR spectrometers at 125.77, 90.55 and
75.47 MHz, respectively. Mass spectra were recorded on a VG ZAB
2SE high resolution mass spectrometer, with Opus V3.1 and DEC
3000 Alpha Station and microanalysis were recorded on a LEEMANS
C E440 Elemental Analyzer. Reactions were monitored, whenever
possible by TLC on silica gel plates (G254) and column chromatog-
5. Conclusion
raphy was performed using C60 silica gel (35–70 mm). Samples were
The findings are encouraging because the four esters possess, to
varying degrees, some specificity towards the breast adenocarci-
noma cell line (MDA-mb468), with chlorambucil esters of pras-
terone (C1 and C2) displaying higher activity. However the
chlorambucil ester of prasterone (C1) and the nitrochlorambucil
ester of pregnenolone (C3) also exhibited relatively significant
cytotoxic activity towards brain posterior fossa, medullablastoma
(Daoy) and lung large cell carcinoma (H460) cell lines. An addi-
tional advantage of linking chlorambucil and its nitro-derivative to
prasterone and pregnenolone is that the resulting esters exhibit
increased lipophilic properties, which should thus increase their
ability to traverse blood-tissue barriers. However, this must be
balanced by their ability to be cleaved at some stage in order to
avoid possible accumulation within the cell membrane and thus be
prevented from entering the cytosol. From the results, one can
conclude that the esters did not show improved cytotoxicity
towards the cell lines tested, compared to the parent drugs, in fact,
it was not the intention to develop compounds with greater activity
but compounds that possess cytotoxicity similar to or close to the
parent drugs, in the hope of being more selective to cancerous cells,
and less damaging to normal cells and tissues. A comparison
between the activity of the esters and the parent drugs, does
however, require some degree of caution, since it is inconclusive as
to whether the difference in activity was due to intrinsic properties
of the esters or whether it was attributed to the partial or complete
bioconversion of the esters into the active parent drugs. It would
certainly be of interest to determine the precise mode of action of
these esters in order to confirm whether or not they truly function
as prodrugs. The esters may also be subject to efflux by P-glyco-
proteins or by ABC transporters, thus limiting their bioavailability
[39–42]. It would be of interest to investigate the cytotoxicities of
these esters in cells that over-express P-glycoproteins, initially
targeting the breast adenocarcinoma cell line (MCF-7) using
matched controls and an MCF-7 cell line that over-expresses P-
glycoprotein. Studies are now underway in order to answer some of
these questions.
centrifuged using a Sanyo MSE Microcentaur and esters were
incubated (hydrolysed) in a Memmert UM200 incubator.
Three HPLC systems were used at various stages of the investi-
gation – GBC LC 1110 HPLC pump connected with a 20 ml loop, a GBC
LC 1210 UV/vis detector and a Hewlett Packard HP3395 integrator.
The second system involved the use of a Varian Vista Series 5000 LC
pump connected to a 20
a Hewlett Packard HP3395 integrator. The third system involved
the use of a Gilson 305 pump, connected to a 20 l loop, a Dionex
ml loop, a Waters 486 UV/vis detector and
m
UVD 340S diode array detector controlled by a Dell Optiplex GX1
computer equipped with Chromeleon (version 6.10). The unit was
connected to a Hewlett Packard DeskJet 890C printer.
Chemicals: dichloromethane, tertiary butanol, N,N-dicyclohex-
ylcarbodiimide, 4-dimethylaminopyridine, nitronium tetrafluo-
roborate, sodium acetate, chlorambucil, isopropanol, n-octanol,
HPLC grade acetonitrile, sodium chloride, potassium dihyhrogen
phosphate, disodium hydrogen phosphate potassium chloride,
porcine liver esterase, suspended in 3.2 M ammonium sulphate.
(340 ml equivalent to 16.5 mg of protein), absolute ethanol and
water were all purchased from Sigma. Ethyl acetate, C60 silica gel,
light petroleum ether (40–60), ethanol (96%), deuterated chloro-
form and acetone (analar) were purchased from Merck. Pregnen-
olone and prasterone were purchased from Acros Organics.
4-(4-Nitrobenzyl) pyridine and 3-amino-1-propanol were
purchased from Lancaster Research Chemicals. The media used for
the tissue culture was RPMI 1460 supplemented with 2 mM
L-
glutamine and 10% foetal bovine serum. The RPMI and -glutamine
L
were purchased from Sigma while the foetal bovine serum was
purchased from PAA Laboratories.
6.2. Methods
6.2.1. Cell lines used in the investigation
The ATCC human cell lines used in the study were cultured at
The Patersons Institute for Cancer Research, Manchester, UK and

L.A. Shervington et al. / European Journal of Medicinal Chemistry 44 (2009) 2944–2951
2949
The Cancer Research Council Unit, Bradford, UK. The cell lines used
were: colon adenocarcinoma (Widr); brain posterior fossa medul-
lablastoma (Daoy); breast adenocarcinoma (MDA-mb468); breast
adenocarcinoma cell line (MDA-NQO1) transfected with fla-
voenzyme NAD(P)H: quinone oxidoreductase (DT-diaphorase);
breast adenocarcinoma (MCF-7); lung large cell carcinoma (H460);
ovarian adenocarcinoma (OVCA-3) and skin malignant melanoma
(A375).
(C18),19.5 (C19), 20.5 (C11), 22.2 (C15), 26.4 (CH2CH2CO2), 27.2 (C2),
29.6 (C12), 31.5 (C7)1, 31.8 (C8)1, 34.1 (ArCH2 and CH2CH2CO2), 36.0
(C16), 36.8 (C10), 37.0 (C1), 38.1 (C4), 40.3 (2 ꢂ CH2Cl), 47.4 (C13),
50.0 (C9), 51.6 (C14), 53.8 (2 ꢂ CH2CH2Cl), 73.5 (C3), 112.4 [2 ꢂ CH,
ortho to N(CH2CH2Cl)2], 121.8 (C6), 129.7 [2 ꢂ CH, meta to
N(CH2CH2Cl)2], 131.0 (C–CH2), 139.8 (C5), 43.8 (C–N), 172.2 (O–
C]O), 220.9 (C]O); m/z (EI) 575.
6.2.2.3. (3b)-3-Hydroxyandrost-5-ene-17-one, 4-{4-[di(2-chloroeth-
6.2.2. General procedures for the synthesis of compounds C1, C2, C3,
C4 and C6
yl)amino]-3-nitrophenyl}butyrate (C2). The final step involved the
removal of ethyl acetate by evaporation under reduced pressure to
give an oil which was chromatographed on a column of silica gel
using ethyl acetate-light petroleum (1:2) as eluent to afford the
ester as a light orange oil (690 mg, 70%). Found: C, 63.82; H, 7.22; N,
4.47; Cl, 11.28%. C33H44Cl2N2O5 requires C, 63.97; H, 7.16; N, 4.52; Cl,
11.44%; vmax (CHCl3)/cmꢃ1 1731 cmꢃ1 (C]O); dH (300 MHz; CDCl3)
0.88 (3H, s, CH3, on C18), 1.05 (3H, s, CH3, on C19), 2.30 (2H, t, CH2,
CH2CH2CO2), 2.36 (1H, m, CH, on C8), 2.41 (2H, m, CH2, on C16), 2.66
(2H, m, ArCH2), 3.40–3.59 (8H, m, 2 ꢂ CH2Cl and 2 ꢂ CH2CH2Cl),
4.60 (1H, m, CH, on C3), 5.40 (1H, d, CH, on C6), 7.30 [1H, d, CH, ortho
to N(CH2CH2Cl)2], 7.35 [1H, dd, CH, meta to N(CH2CH2Cl)2], 7.54 (1H,
d, CH, ortho to NO2, C6H3NO2); dC (CDCl3) 13.5 (C18), 19.5 (C19), 20.3
(C11), 21.8 (C15), 26.1 (CH2CH2CO2), 27.7 (C2), 31.0 (C12), 31.3 (C7)1,
31.4 (C8)1, 33.7 (CH2CO2), 33.9 (ArCH2), 35.8 (C16), 36.7 (C10), 36.9
(C1), 38.0 (C4),41.4 (2 ꢂ CH2Cl), 47.4 (C13), 50.0 (C9), 51.6 (C14), 55.9
(2 ꢂ CH2CH2Cl), 73.8 (C3), 121.9 (C6), 124.8 [CH, ortho to
N(CH2CH2Cl)2], 126.7 (CH, ortho to NO2), 133.3 (CH, para to NO2),
138.5 (C–CH2), 139.8 (C5),140.5 (C–NO2), 146.3 (C–N), 172.4 (O–
C]O), 221.0 (C]O); m/z (EI) 620.
6.2.2.1. Synthesis of 4-{4-[di(2-chloroethyl)amino]-3-nitrophenyl}
butanoic acid (C6). A solution of nitronium tetrafluoroborate (2.0 g,
15.64 mmol) in acetonitrile (60 ml) was stirred at 0 ꢀC under argon,
and after 15 min chlorambucil (4.0 g, 13.16 mmol) in acetonitrile
(60 ml) was added dropwise over a period of 10 min. After being
stirred for an additional 40 min at 0 ꢀC, the reaction mixture was
stirred at room temperature for 1 h and then poured into an excess
of water (150 ml). The organic material was extracted with
dichloromethane (3 ꢂ 60 ml), washed with water (3 ꢂ 60 ml) and
dried over anhydrous sodium sulphate. The sodium sulphate was
removed by filtration and the dichloromethane was removed by
evaporation under reduced pressure to afford the crude product
which was chromatographed on a column of silica gel with ethyl
acetate-light petroleum (2:1) as eluent to give the 3-nitro-
chlorambucil as a dark orange oil (2.51 g, 55%). Found: C, 47.9; H,
5.40; N, 8.20; Cl, 20.20%. C14H18Cl2N2O4 requires C, 48.15; H, 5.20;
N, 8.00; Cl, 20.30%; nmax (CHCl3)/cmꢃ1 1720 (C]O) and 2800–3500
br (OH); dH (300 MHz; CDCl3) 1.98 (2H, m, CH2CH2CO2H), 2.42
(2H, t, J 6.9, CH2CO2H), 2.70 (2H, t, J 7.5, ArCH2), 3.50 [8H, m,
N(CH2CH2Cl)2], 7.30 (1H, d, J 9.0, meta to NO2, C6H3NO2), 7.35 (1H,
dd, J 8.2 and 2.0, para to NO2, C6H3NO2), and 7.54 (1H, d, J 2.0, ortho
to NO2C6H3NO2); dC (CDCl3) 25.8 (CH2CH2CO2H), 32.9 and 33.9
(CH2CH2CH2), 41.5 (2 ꢂ CH2Cl), 55.9 (2 ꢂ H2CN), 124.8 [CH, ortho to
N(CH2CH2Cl)2], 126.8 (CH, ortho to NO2), 133.3 (CH, para to NO2),
138.2 (C–CH2), 140.6 (C–NO2), 146,4 [C–N(CH2CH2 Cl)2], 177.0
(CO2H); m/z (EI) 348.
6.2.2.4. (3b)-3-Hydroxypregn-5-ene-20-one, 4-{4-[di(2-chloroethyl)
amino] phenyl}butyrate (C3). The final step involved the removal of
ethyl acetate under reduced pressure to afford an oil which was
chromatographed on a column of silica gel with ethyl acetate–light
petroleum (1:3) as eluent to give the ester as an off-white oil (1.18 g,
60%). Found: C, 69.36; H, 8.20; N, 2.33; Cl, 11.58%. C35H49Cl2NO3
requires C, 69.75; H, 8.19; N, 2.32; Cl, 11.76%; vmax (CHCl3)/cmꢃ1
1730 cmꢃ1 (O–C]O) and 1704 cmꢃ1 (CH3–C]O); dH (300 MHz;
CDCl3) 0.63 (3H, s, CH3, on C18), 1.01 (3H, s, CH3, on C19), 2.13 (3H, s,
CH3, on C21), 2.28 (1H, m, CH, on C8), 2.30 (2H, t, CH2CH2CO2), 2.50
(1H, m, CH, on C17), 2.54 (2H, t, ArCH2), 3.60–3.72 (8H, m, 2 ꢂ CH2Cl
and 2 ꢂ CH2CH2Cl), 4.60 (1H, m, CH, on C3), 5.37 (1H, d, CH, on C6),
6.64 [2H, d, ortho to N(CH2CH2Cl)2], 7.07 [2H, d, meta to
N(CH2CH2Cl)2]; dC (CDCl3) 13.2 (C18), 19.3 (C19), 21.2 (C11), 22.8
(C16), 24.5 (C15), 26.1 (CH2CH2CO2), 27.7 (C2), 31.5 (C7)1, 31.7 (C21),
31.8 (C8)1, 33.8 (CH2CO2), 34.0 (ArCH2), 36.6 (C10), 36.9 (C1), 38.0
(C4), 38.6 (C12), 40.3 (2 ꢂ CH2Cl), 43.9 (C13), 49.8 (C9), 53.7
(2 ꢂ CH2CH2Cl), 56.7 (C14), 63.6 (C17), 73.6 (C3), 112.4 [2 ꢂ CH,
ortho to N(CH2CH2Cl)2], 122.2 (C6), 129.7 [2 ꢂ CH, meta to
N(CH2CH2Cl)2], 131 (C–CH2), 139.6 (C5), 143.9 (C–N), 172.9 (O–
C]O), 209.6 (CH3–C]O); m/z (EI) 603.
6.2.2.2. Synthesis of (3b)-3-hydroxyandrost-5-ene-17-one, 4-{4-
[di(2-chloroethyl)amino] phenyl}butyrate (C1). General procedure
was used for preparing all four esters.
A solution of chlorambucil (1.0 g, 3.29 mmol) in dichloro-
methane (40 ml) was stirred at room temperature for 10 min, after
which time, prasterone (1.09 g, 3.78 mmol) in dichloromethane
(10 ml) was added dropwise over 5 min. A solution of DCC (780 mg,
3.78 mmol) in dichloromethane (20 ml) was added after which
time DMAP (1.2 mg, 9.8 mmol) was added to catalyze the reaction.
The reaction mixture was sealed and stirred for 20 h at room
temperature. The resulting suspension was treated with acetoni-
trile (30 ml) in order to enhance the precipitation of the by-prod-
ucts. The precipitate was filtered and the solvents were evaporated
under reduced pressure to afford the crude product. The crude
product was redissolved in ethyl acetate (50 ml), leaving behind
undissolved by-product which was removed by filtration. The ethyl
acetate was removed by evaporation under reduced pressure to
give an oil which was chromatographed on a column of silica gel
using ethyl acetate-light petroleum (1:2) as eluent to afford the
ester as an off-white oil (1.19 g, 63%). Found: C, 68.53; H, 7.41; N,
2.40; Cl, 12.19%. C33H45Cl2NO3 requires C, 68.97; H, 7.89; N, 2.43; Cl,
12.33%; vmax (CHCl3)/cmꢃ1 1731 cmꢃ1 (C]O); dH (300 MHz; CDCl3)
0.88 (3H, s, CH3 on C18), 1.05 (3H, s, CH3 on C19), 2.30 (2H, t,
CH2CH2CO2), 2.36 (1H, m, CH, on C8), 2.41 (2H, m, CH2, on C16), 2.56
(2H, m, ArCH2), 3.60–3.72 (8H, m, 2 ꢂ CH2Cl and 2 ꢂ CH2CH2Cl),
4.61 (1H, m, CH, on C3), 5.40 (1H, d, CH, on C6), 6.65 [2H, d, ortho to
N(CH2CH2Cl)2], 7.08 [2H, d, meta to N(CH2CH2Cl)2]; dC (CDCl3) 13.5
6.2.2.5. (3b)-3-Hydroxypregn-5-ene-20-one,4-{4-[di(2-chloroethyl)-
amino]-3-nitro-phenyl}butyrate (C4). After the removal of the ethyl
acetate under reduced pressure, the resulting oil was chromato-
graphed on a column of silica gel using ethyl acetate-light petro-
leum (1:3) as eluent to afford the ester as a light orange oil (780 mg,
66%). (Found: C, 64.55; H, 7.36; N, 4.32; Cl, 10.86%. C35H48Cl2N2O5
requires C, 64.90; H, 7.47; N, 4.32; Cl, 10.94%); vmax (CHCl3)/cmꢃ1
1730 cmꢃ1 (O–C]O) and 1703 cmꢃ1 (CH3–C]O); dH (300 MHz;
CDCl3) 0.63 (3H, s, CH3, on C18), 1.01 (3H, s, CH3, on C19), 2.13 (3H, s,
CH3, on C21), 2.28 (1H, m, CH, on C8), 2.30 (2H, t, CH2CH2CO2), 2.50
1
Possible reversed assignment.

2950
L.A. Shervington et al. / European Journal of Medicinal Chemistry 44 (2009) 2944–2951
(1H, m, CH, on C17), 2.61 (2H, t, ArCH2), 3.39–3.50 (8H, m, 2 ꢂ CH2Cl
and 2 ꢂ CH2CH2Cl), 4.60 (1H, m, CH, on C3), 5.35 (1H, d, CH, on C6),
7.30 [1H, d, CH, ortho to N(CH2CH2Cl)2], 7.35 [1H, dd, CH, meta
to N(CH2CH2Cl)2], 7.54 (1H, d, CH, ortho to NO2, C6H3NO2); dC
(CDCl3) 13.2 (C18), 19.3 (C19), 20.9 (C11), 22.8 (C16), 24,4 (C15),
26.1 (CH2CH2CO2), 27.7 (C2), 31.5 (C7), 31.7 (C21), 31.8 (C), 3.7
(CH2CO2), 34.0(ArCH2), 36.6 (C10), 36.9 (C), 38.0 (C4), 38.(C12), 41.5
(2 ꢂ CH2Cl), 43.9 (C13), 49.8(C9), 55.9 (2 ꢂ CH2CH2Cl), 56.7 (C14),
63.6 (C17), 73.9 (C3), 122.4 (C6), 124 [CH, ortho to N(C2CH2Cl)2],
126.7 (CH, ortho NO2), 133.3 (CH, para to NO2), 139.5 (C5),
140.(C–NO2), 146.3 (C–N), 172.4 (O–C]O), 209.6 (CH3–C]O); m/z
(EI) 648.
6.3.3. General procedure used to determine the hydrolysis of the
esters using porcine liver esterase [47]
The stock solution of the ester was prepared by dissolving the
ester (50 mg) in absolute ethanol (10 ml) and transferring it to
a 25 ml volumetric flask and making upto the mark with absolute
ethanol. The phosphate buffer was prepared by dissolving sodium
chloride (2 g), potassium dihydrogen phosphate (50 mg), disodium
hydrogen phosphate (250 mg) and potassium chloride (50 mg) in
distilled deionised water and making up to the 250 ml mark in
a volumetric flask (pH 7.4). After incubating the esterase enzyme
(20
(200
in buffer at 37 ꢀC for 15 min prior to being analyzed by HPLC. A
sample of the reaction mixture (200 l) was then removed and
transferred to microfuge tubes at set time intervals and treated
with acetonitrile (200 l). The mixture was then vortexed for 30 s
and then centrifuged for 10 min at 11,600 g. A portion of the
m
l) in buffer (2 ml) for approximately 15 min at 37 ꢀC, the ester
l) was added. The zero reaction involved incubating the ester
m
m
6.3. Alkylating activity [29,43,44]
m
The alkylating activity of each compound was determined at
concentrations ranging from 0.1 mM to 1.0 mM, dissolved in 50% (v/
v) acetone/ethanol solution. The compounds (0.2 ml) were added
separately to screwed capped test tubes containing 1 ml of 0.2 M
acetate buffer (pH 5.6). Each tube was treated with 0.5 ml of 5% (w/
v) solution p-nitrobenzyl pyridine in acetone and the mixture was
incubated for 4 h at 37 ꢀC. Each tube was then treated with 3.0 ml of
25% (v/v) 3-amino-1-propanol in tertiary butyl alcohol and the
coloration of the reaction product was measured using a visible
spectrophotometer at 560 nm. The alkylating activity of the five
compounds was compared with that of an equimolar chlorambucil
solution.
supernatant (200
mobile phase (200
m
m
l) was transferred to a microfuge tube and
l) was added and mixed. A standard solution of
the ester was prepared by transferring 5 ml of the stock solution to
a 50 ml volumetric flask, making upto the mark with the mobile
phase, and then transferring 3 ml of this solution to a 25 ml volu-
metric flask and making upto the mark with mobile phase and
finally, 5 ml of this solution was transferred to a 10 ml volumetric
flask and made upto the mark to afford a final concentration of the
standard solution of 1.20 mg%. Both the standard and sample
solutions were analyzed using a validated HPLC method (unpub-
lished data), derived from a modified version of a published
method [48]. The esters were also stored in phosphate buffer saline
at 37 ꢀC and monitored for their stability using the HPLC method,
supported by infrared spectroscopy, over a period of forty five days.
The chromatographic conditions: RP-HPLC, the mobile phase con-
sisted of acetic acid (50 ml of a 2% solution) made upto a litre with
acetonitrile, pH 5.2 and the flow rate was set at 1.2 ml/min. A C18
Symmetry Waters column (250 mm ꢂ 4.6 ID) fitted with a guard
column (3 cm ꢂ 4.6 mm ID) was used and the wavelength was set
at 254 nm.
6.3.1. Log P values
The six compounds of known weight (10 mg) were added
separately to 50 ml volumetric flasks and dissolved in n-octanol
and then made up to the mark [29]. A portion (4 ml) of each of the
solutions was transferred separately to screwed capped centrifuge
tubes and an equal volume of water was added to each tube. The
tubes were capped and shaken on a rotator for 1 h at 20 ꢀC, after
which they were centrifuged for 10 min at 3600 g and the n-octanol
layer was carefully separated from the water layer. Portions of the
original solutions were diluted to obtain a concentration of
0.8 mg%. The shaken solutions were also diluted in a similar
manner and comparisons were made using UV spectroscopy
between 200 and 400 nm. The log P values were then calculated
according to Leo et al. [45].
6.3.4. General procedure used to determine the hydrolysis of the
esters using human plasma
A stock solution of the ester was prepared by dissolving the
ester (50 mg) in absolute ethanol (10 ml) and transferring it to
a 50 ml volumetric flask and making upto the mark with the same
solvent. The ester (50
treated with mobile phase (950
500 l of the solution was transferred to another microfuge tube
and made upto 1 ml with mobile phase and vortexed. Finally, 500 ml
of this solution was transferred to a another microfuge tube, made
upto 1 ml with mobile phase, and mixed in the usual manner, to
afford a 1.25 mg% concentration of standard solution.
m
l) was transferred to a microfuge tube and
6.3.2. In vitro cytotoxicity assay [46]
ml), mixed using a vortex mixer and
The standard assay for cytotoxicity employed a 96 well format
m
using 400–1000 cells per well in 100
represented by three wells. Media (100
dilutions of 2ꢂ drug were added to triplicate sets. Control wells
received 100 l of media alone. The plates were then incubated in
a humidified atmosphere for five days at 37 ꢀC, with 5% carbon
dioxide. After incubation 50 l of MTT (3-(4,5-dimethylthiazol-2-
m
m
l. Each drug dose was
l) containing doubling
m
Freshly prepared plasma (850
treated with phosphate buffer saline (100
the ester (50 l), sealed with parafilm, vortexed for 2 min and then
placed into an incubator set at 37 ꢀC. The incubated sample was
treated in the following manner prior to analysis. 500 l of the
mixture was transferred to a microfuge tube and treated with
500 l of mobile phase, mixed for 2 min using a vortex mixer and
the resulting precipitate was removed by centrifugation at 11,600 g
for 20 min. 500 l of the supernatant was removed and transferred
ml) was transferred to a test tube,
m
ml) and stock solution of
yl)-2,5-diphenyl tetrazolium bromide) (3 mg/ml) was added to
each well and the plate returned to the incubator for 3 h. The
process is known as metabolic reduction and requires an active
mitochondrial function to reduce the salt. The media and excess
MTT were then aspirated from each well and the formazan
m
m
m
crystals solubilized in 200
ml of DMSO. The plates were then read
using multiscan microplate reader (Titretech) at 540 nm
a
m
with subtraction at 620 nm to allow for turbidity. The resultant
output was processed using an excel spreadsheet. The percentage
of growth inhibition absorbance of the drug treated wells/
mean absorbance of the control well ꢂ 100 and the IC50 (concen-
tration of compound required to inhibit 50% growth) were
calculated.
to a microfuge tube and made upto 1 ml with mobile phase.
The content of the tube was then centrifuged at 11,600 g for
a further 10 min.
A comparison between the peak areas of the standard and the
prepared samples were then carried out by the HPLC method in
order to estimate the percent of ester hydrolyzed with time.

L.A. Shervington et al. / European Journal of Medicinal Chemistry 44 (2009) 2944–2951
2951
[20] S. Li, X. Yan, A. Belanger, F. Labrie, Breast Cancer Res. Treat. 29 (1993) 203–217.
[21] E. Osawa, A. Nakajima, S. Yoshida, M. Omura, H. Nagase, N. Ueno, K. Wada,
N. Matsuhashi, M. Ochiai, H. Nakagama, H. Sekihara, Life Sci. 70 (2002)
2623–2630.
[22] B. Zumoff, Hormonal profile in women with breast cancer (review), Anticancer
Res. 8 (1988) 627–636.
[23] G.B. Gordon, T.L. Bush, K.J. Helzlsover, S.R. Miller, G.W. Cumstock, Cancer Res.
50 (1990) 3859–3862.
[24] H. Adlercreutz, E. Hamalainen, S.L. Gorbach, B.R. Goldin, M.N. Woods,
J.T. Dwyer, Am. J. Clin. Nutr. 49 (1989) 433–442.
Acknowledgment
The authors would like to thank the University of Central
Lancashire, School of Pharmacy and Pharmaceutical Sciences UK,
and the Faculty of Pharmacy, Applied Science University (Amman,
Jordan), for their support. A special thanks to Professor David Ross,
the School of Pharmacy, University of Colorado Health Science
Centre, USA, for generously providing the MDA468 NQO1 cell line.
[25] P. Ebeling, V.A. Koivisto, Lancet 343 (1994) 1479–1481.
[26] E. Plassart-Schies, E. Baulieu, Neurosteroids: recent findings, Brain Res. Rev. 37
(2001) 133–140.
[27] S. Saah, W. Wu, K. Eberst, E. Marvanyos, N. Bodor, J. Pharm. Sci. 85 (1996)
496–504.
References
[28] K. Lof, J. Hovinen, P. Reinikainen, L.M. Vilpo, E. Seppala, J.A. Vilpo, Chem. Biol.
Interact. 103 (1997) 187–198.
[1] J.L. Everett, J.J. Roberts, W.C.J. Ross, Aryl-2-halogenoalkylamines, part XII. Some
carboxylic derivatives of N,N-di-2-chloroethylaniline, J. Chem. Soc. Part III
(1953) 2386–2392.
[2] J.E.F. Reynolds, Martindale, The Extra Pharmacopoeia, twentyninth ed. The
Pharmaceutical Press, 1989.
[29] S. Genka, J. Deutsch, U.H. Shetty, P.L. Stahle, V. John, I.M. Lieberburg, F. Ali-
Osmant, S.I. Rapoport, N.H. Greig, Clin. Exp. Metastasis 11 (1993) 131–140.
[30] C.K. Osborne, K. Hobbs, J.M. Trent, Breast Cancer Res. Treat. 9 (1987) 111–121.
[31] S.A. Everett, E. Swann, M.A. Naylor, M.R.L. Stratford, K.B. Patel, N. Tian,
R.G. Newman, B. Vojnovic, C.J. Moody, P. Wardman, Biochem. Pharmacol. 63
(2002) 1629–1639.
[3] S.K. Carter, M.T. Bakowski, K. Hellmann, Chemotherapy of Cancer, third ed.
Wiley Medical, New York, 1987.
[4] N.H. Greig, S.I. Rapoport, Cancer Chemother. Pharmacol. 21 (1988) 1–8.
[5] B.B. Bank, D. Kanganis, L.F. Liebes, R. Silber, Cancer Res. 49 (1989) 554–559.
[6] R. Silvennoinen, K. Malminiemi, O. Malminiemi, E. Seppala, J. Vilpo, Pharma-
col. Toxicol. 87 (2000) 223–228.
[32] R.J. Riley, P. Workman, Biochem. Pharmacol. 43 (1992) 1657–1669.
[33] R.J. Knox, F. Friedlos, M. Boland, Cancer Metastasis Rev. 12 (1993) 195–212.
[34] H.D. Beall, A.M. Murphy, D. Siegel, R.H.J. Hargreaves, J. Butler, D. Ross, Mol.
Pharmacol. 48 (1995) 499–504.
[7] F.Y.F. Lee, P. Coe, P. Workman, Cancer Chemother. Pharmacol. 17 (1986) 21–29.
[8] J. Mann, L.A. Shervington, J. Chem. Soc. Perkin Trans. 1 (1991) 2961–2964.
[9] G.C. Kundu, J.R. Schullek, I.B. Wilson, Pharmacol. Biochem. Behav. 43 (1994)
621–624.
[35] J.J. Schlayer, G. Powis, Int. J. Cancer 45 (1990) 403–409.
[36] T. Cresteil, A.K. Jaiswal, Biochem. Pharmacol. 42 (1991) 1021–1027.
[37] V.M. Kiniene, E. Sergediene, A. Nemeikaite, J. Segura-Aguilar, N. Cenes, Cancer
Lett. 146 (1999) 217–222.
[10] N.H. Greig, S. Genka, E.M. Daly, D.J. Sweeney, S.I. Rapoport, Cancer Chemother.
Pharmacol. 25 (1990) 311–319.
[38] V. Misra, A. Grondin, H.J. Klamut, A.M. Rauth, Br. J. Cancer 83 (2000)
998–1002.
[11] I. Konyves, J. Liljekvist, The steroid molecule as a carrier of cytotoxic groups, in:
W. Davis, C. Maltoni (Eds.), Biological Characterisation of Human Tumours,
375, Elsevier, Amsterdam, 1976, p. 98 [International Congress Series].
[12] L. Brandt, I. Konyves, T.R. Moller, Acta. Med. Scand. 197 (1975) 323–327.
[13] J. Lober, H.T. Mouridsen, I.E. Christiansen, P. Dombernowsky, W. Mattsson,
M. Rorth, Cancer 52 (1983) 1570–1576.
[39] J. Rieger, W. Roth, T. Glaser, S. Winter, L. Rieger, J. Dichgans, M. Weller, Neurol.
Psychiatr. Brain Res. 7 (1999) 37–46.
[40] N. Okamura, T. Sakaeda, K. Okumura, Pharmacogenomics of MDR and MRP
subfamilies, Personalized Med. 1 (2004) 85–104.
[41] A.A. Stavrovskaya, Cellular mechanism of multidrug resistance of tumour cells,
Biochemistry (Mosc). 65 (2000) 95–106.
[14] N. Ohsawa, Z. Yamazaki, T. Wagatsuma, K. Isurugi, Jpn. J. Cancer Chemother. 11
(1984) 2115–2124.
[42] P.K. Smitheman, A.J. Townsend, T.E. Kute, C.S. Morrow, J. Pharmacol. Exp. Ther.
308 (2004) 260–267.
[15] T. Kubota, T. Kawamura, T. Suzuki, T. Yamada, H. Toyoda, T. Miyagawa,
T. Kurokawa, Jpn. J. Clin. Oncol. 16 (1986) 357–364.
[43] J. Epstein, R.W. Rosenthal, R.J. Ess, Anal. Chem. 27 (1955) 1435–1439.
[44] T.J. Bardos, N. Datta-Gupta, P. Hebborn, D.J. Triggle, J. Med. Chem. 8 (1965) 167.
[45] A. Leo, C. Hansch, D. Elkins, Partition coefficients and their uses, Chem. Rev. 71
(1971) 525–616.
[46] T. Mossmann, J. Immunol. Methods 65 (1983) 55–63.
[47] T. Kawaguchi, A.F. Youssef, T. Aboul-Fadl, F.A. Omar, H.H. Farag, T. Hasegawa,
Pharmazie 51 (1996) 717–719.
[16] Goodman, Gilman‘s, Health professional division chapters, The Pharmaco-
logical Basis of Therapeutics, 9th ed. McGraw-Hill, 1996, pp. 57–59.
[17] N.A. Compagnone, S.H. Mellon, Neurosteroids: biosynthesis and Function of
these Novel Neuromodulators, Front. Neuroendocrinol. 21 (2000) 1–56.
[18] S.M. Sirrs, R.A. Bebb, DHEA: panacea or Snake oil, Can. Fam. Physician 45
(1999) 1723–1728.
[48] N.H. Greig, E.H. Daly, D.J. Sweeney, S.I. Rapoport, Cancer Chemother. Phar-
macol. 25 (1990) 320–325.
[19] E. Baulieu, Life Sci. 323 (2000) 513–518.