Design of novel tyrosine-nitrogen mustard hybrid molecules active against uterine, ovarian and breast cancer cell lines

Article abstract of DOI:10.1016/j.steroids.2011.12.021

L-para-Tyrosine was linked to ortho-hydroxyaniline, meta-hydroxyaniline and para-hydroxyaniline giving three distinct tyrosinamide molecules. The new extended amino acid derivatives were constructed to imitate, in part, the estradiol (E₂, the natural female sex hormone) nucleus. The resulting tyrosinamides were then linked to chlorambucil either directly, or via a 5 and 10 carbon atoms spacer chain. This was done in an attempt to target cancerous cells expressing the estrogen receptor alpha (ERα) and to obtain a more specific chemotherapeutic agent. The tyrosinamide-chlorambucil molecules were designed and synthesized in good yields, according to two different approaches. The novel compounds were evaluated for their anticancer efficacy in hormone-dependent and hormone-independent (ER+; MCF-7 and ER-; MDA-MB-231) breast cancer cell lines. Interestingly, the meta-hydroxyphenyl-tyrosinamide- chlorambucil derivatives were more active than the ortho- and para- analogs. The molecules bearing a 5 carbon atoms spacer were selected for additional biological study using a panel of female cancerous cells; breast (ZR-75-1, MDA-MB-436, MDA-MB-468), ovarian (OVCAR-3, A2780) and uterine (Ishikawa, HEC-1A). It was discovered that for breast cancer cells, the new compounds were up to 4.2 times more active than chlorambucil itself.

Full text of DOI:10.1016/j.steroids.2011.12.021

Steroids 77 (2012) 403–412
Contents lists available at SciVerse ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Design of novel tyrosine-nitrogen mustard hybrid molecules active against
uterine, ovarian and breast cancer cell lines
⇑
Caroline Descôteaux, Kevin Brasseur, Valérie Leblanc, Sophie Parent, Éric Asselin, Gervais Bérubé
Groupe de Recherche en Oncologie et Endocrinologie Moléculaires, Département de Chimie-Biologie, Université du Québec à Trois-Rivières, C.P. 500, Trois-Rivières,
Québec, Canada G9A 5H7
a r t i c l e i n f o
a b s t r a c t
Article history:
L-para-Tyrosine was linked to ortho-hydroxyaniline, meta-hydroxyaniline and para-hydroxyaniline giving
Received 12 November 2011
Received in revised form 14 December 2011
Accepted 17 December 2011
Available online 29 December 2011
three distinct tyrosinamide molecules. The new extended amino acid derivatives were constructed to
imitate, in part, the estradiol (E2, the natural female sex hormone) nucleus. The resulting tyrosinamides
were then linked to chlorambucil either directly, or via a 5 and 10 carbon atoms spacer chain. This was
done in an attempt to target cancerous cells expressing the estrogen receptor alpha (ERa) and to obtain a
more specific chemotherapeutic agent. The tyrosinamide–chlorambucil molecules were designed and
synthesized in good yields, according to two different approaches. The novel compounds were evaluated
for their anticancer efficacy in hormone-dependent and hormone-independent (ER+; MCF-7 and ERÀ;
MDA-MB-231) breast cancer cell lines. Interestingly, the meta-hydroxyphenyl-tyrosinamide-chlorambu-
cil derivatives were more active than the ortho- and para- analogs. The molecules bearing a 5 carbon
atoms spacer were selected for additional biological study using a panel of female cancerous cells; breast
(ZR-75-1, MDA-MB-436, MDA-MB-468), ovarian (OVCAR-3, A2780) and uterine (Ishikawa, HEC-1A). It
was discovered that for breast cancer cells, the new compounds were up to 4.2 times more active than
chlorambucil itself.
Keywords:
Ortho-, meta- and para-hydroxyphenyl-
tyrosinamide-chlorambucil derivatives
Female cancers
Estrogen receptor alpha
Chlorambucil
Biological activity
Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction
be overexpressed [7]. Hence, ER
has been recognized as a useful tool for endocrine therapy.
The ER has already been targeted in an attempt to accumulate
a known chemotherapeutic agent to breast cancer cells [4]. Endog-
enous hormones have been used as potential carrier ligands for the
a is an attractive target which
Chemotherapy is a very common anticancer treatment which
uses anti-proliferative molecules to kill cancerous cells. These che-
motherapeutic agents are highly active drugs which also affect
healthy cells, leading to severe side effects [1]. Drug targeting is
a challenging research subject which can be applied to various dis-
eases such as cancer. In recent years, many cytotoxic compounds
have been designed in order to get more specific anticancer drugs
and as a result, to minimize toxic side effects [2,3]. Targeting strat-
egies have been applied to cases where characteristic physiological
differences exist between diseased and healthy tissues [4].
a
ER
a
. These have been modified and designed to create ligand-
overexpressing cells [8]. Anti-
cytotoxic molecules targeting ER
a
proliferative drugs have been linked to the steroidal ligand. E2-
based anticancer molecules reported have shown cytocidal activity
against hormone-dependent cancer cells at a higher level than the
parent drug given alone [9–12]. Therefore, the use of a shuttle li-
gand such as E2 has demonstrated enhanced effectiveness of
known anticancer drug giving further evidence of the transport
role of E2 [13]. Moreover, such steroidal-anticancer molecules have
Female organs (breast, ovary and uterus) are known to express
a specific protein, the estrogen receptor alpha (ERa) [5]. The ERa is
a member of the nuclear receptor superfamily which is responsible
for the concentration of physiological levels of estradiol (E2) (1,
Fig. 1), the most active steroidal female hormone [6]. In most cases,
female cancers are classified as hormone-dependent, which means
that the tumor development is stimulated by the presence of the
shown to reach the ERa and then, induce E2-like genomic effects
through the targeted protein [14].
In an attempt to avoid the promotion of cancer in response to
estrogenic activity, non-steroid ligands conjugated to cytotoxic
compounds were also designed [8]. There are numerous reports
of compounds with unusual structural skeleton binding to the
ERa. Also, when normal cells become cancerous, ERa is found to
ER
high binding affinity for the ER
can bind to the ER , even if their structures are quite dissimilar to
a
[15]. Among these molecules, some are selective and have
a
. Hence, many different molecules
a
⇑
the natural hormone steroidal backbone. This can also be explained
by the known mobility of protein segments upon ligand binding
Corresponding author. Tel.: +1 819 376 5011x3353; fax: +1 819 376 5084.
E-mail address: Gervais.Berube@uqtr.ca (G. Bérubé).
0039-128X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2011.12.021

404
C. Descôteaux et al. / Steroids 77 (2012) 403–412
OH
Cl
N
O
Cl
HO
HO
Chlorambucil
1
Cl
O
R
O
R
N
O
Cl
HO
Cl
HO
OH
HN
HN
3
N
H
NH2
m
3
HO
O
N
Cl
3
2
4
R = CO2CH3 or CH2OH
Fig. 1. Structure of 17b-estradiol (1), chlorambucil, tyrosine–chlorambucil hybrid molecules (2 and 3) and
L
-para-tyrosine (4).
[16]. Therefore, the binding cavity of the ER
modate various non-steroidal synthetic ligands.
a
is known to accom-
tyrosine–chlorambucil analogs, previously reported [19]. In addi-
tion, adding an aromatic moiety to the ligand structure would cer-
tainly increase the lipophilicity of the final derivatives. Such
molecule would be less polar than tyrosine and consequently,
more comparable to the steroid backbone. Hence, tyrosine–
hydroxyaniline conjugates would be more susceptible to imitate
the natural steroid, estradiol.
Several structural characteristics need to be considered in order
to build an estrogen-like ligand. A phenol ring, similar to the A-ring
of E2, must be a part of the molecule [17]. Moreover, two polar
functional groups, such as hydroxyl, must be also present. The 3-
OH and 17b-OH located at the two extremities of E2 skeleton are
involved in the E2–ER binding [18]. The interatomic distance of
the hydroxyl groups in E2 is 10.9 Å. Thus, in order to bind to the
In this particular study, ortho, meta and para-hydroxyaniline
moieties have been linked to the L-para-tyrosine (4), in attempt
corresponding residue into the ER
lar groups far away from each other. Other molecules, with polar
functional groups from 10.3 to 12.1 Å apart, have also shown inter-
esting binding affinity for the ERa [18].
Recently, several tyrosine–chlorambucil derivatives (2 and 3,
with R = CO2CH3 or CH2OH, Fig. 1) have been reported [9,19].
Chlorambucil was linked, directly or via a tether chain, to tyrosine
ester (R = CO2CH3) or tyrosine hydroxymethyl (R = CH2OH) mole-
cules. Chlorambucil is a known anticancer agent which structure
is shown in Fig. 1. In this article, tyrosine (4, Fig. 1) was selected
as a template to mimic, at least in part, the A-ring phenol of the
estradiol nucleus. Tyrosine is a natural amino acid reported to have
a structure similar to that of the phenol group of E2 [18]. Molecular
modeling study showed that the phenol group of tyrosine can
a
, a molecule most have two po-
to obtain ligands with different binding abilities. The position of
a second hydroxyl group, as a part of the hydroxyaniline moiety,
was varied. The tyrosinamide regioisomers were linked to chlor-
ambucil, directly or via a 5 and a 10 carbon atoms spacer chain.
This manuscript gives the detailed synthesis of all the tyrosina-
mide–chlorambucil regioisomers made (5 and 6, Fig. 2). Two meth-
odologies were used and compared; the linear and the convergent
synthesis. Then, the biological activity of all the cytotoxic mole-
cules synthesized was evaluated on hormone-dependent (MCF-7)
and hormone-independent (MDA-MB-231) breast cancer cell lines.
The influence of hydroxyl group location (ortho, meta and para) on
the tyrosinamide–chlorambucil hybrids was studied. Next, the
molecules bearing a 5 carbon atoms spacer were selected for a
more thorough biological evaluation. These L-para-tyrosinamide–
interact into the ER
a
binding cavity in the same manner as the
chlorambucil derivatives were further tested on a panel of hor-
mone-dependent and hormone-independent female cancer cell
lines; breast: ZR-75-1 (ER+), MDA-MB-468 (ERÀ), MDA-MB-436
(ERÀ); ovarian: OVCAR-3 (ER+), A2780 (ERÀ); uterine: Ishikawa
(ER+), HEC-1A (ERÀ).
A-ring phenol of E2 does [19]. The tyrosine–chlorambucil hybrid
molecules 2 and 3 showed enhanced cytotoxicity when compared
to chlorambucil itself and this can possibly indicate the role of
tyrosine as a cytotoxic agent carrier.
In spite of the non negligible cytotoxicity observed, molecular
docking calculations showed that the tyrosine skeleton does not
fully occupy the ER
fit correctly (and completely) into the ER
a
binding cavity. Tyrosine seems too small to
binding pocket [19].
2. Experimental
a
Hence, there is a reason to believe that perhaps, tyrosine-based
molecules with extended skeleton could interact more adequately
into the receptor network and consequently, be more active.
The new tyrosine-based molecules were designed and synthe-
sized keeping in mind the minimal structural requirements needed
2.1. Chemistry
All reactions were performed with ACS Fisher solvents. In some
cases, solvent, as well as starting materials and reactants, were first
purified and dried by standard means [20]. Anhydrous reactions
required an inert atmosphere of dry nitrogen. The 6-aminohexa-
noic acid, 11-aminoundecanoic acid, N-Boc-tyrosine and all
hydroxyaniline isomers (ortho, meta and para) were purchased
from Sigma–Aldrich Canada Ltd., Oakville, Ontario, Canada. All
reactions were monitored by UV fluorescence or staining with io-
dine on Sigma T 6145 commercial TLC plates (polyester silica gel
60 Å, 0.25 mm). Purifications were done using flash column chro-
matography according to the method of Still et al. [21] on Silicycle
to create ER
a ligands. Tyrosine was modified and coupled to a
hydroxyaniline moiety in order to obtain an extended backbone
bearing two phenol groups distant from each other. Then, tyrosine
ester (R = CO2CH3) and tyrosine hydroxymethyl (R = CH2OH) deriv-
atives previously synthesized were replaced by a tyrosinamide en-
tity. The two phenol groups of the tyrosinamide unit could possibly
create hydrogen bonding interactions with the same residue than
E2 does. It is hypothesized that these molecules with an extended
shape could fit more effectively into the receptor than the smaller
UltraPure Flash Silica Gel, 40–63 lm mesh. Hexanes and acetone

C. Descôteaux et al. / Steroids 77 (2012) 403–412
405
OH
OH
O
O
O
Cl
N
N
H
N
O
Cl
H
HN
HN
HO
HO
3
m
N
H
3
Cl
O
N
Cl
6, made with ortho-, meta- and para-hydroxyaniline
5, made with ortho-, meta- and para-hydroxyaniline
m = 5 or 10
Fig. 2. Structures of the tyrosinamide–chlorambucil hybrid molecules (5 and 6).
were distilled before their use as chromatography eluent. Chloro-
form and methanol were used without any prior purification.
The infrared spectra were taken on a Nicolet Impact 420 FT-IR
and a Thermo Nicolet IS10 FT-IR with ATR. Sodium chloride, potas-
sium bromide pellets or ATR diamond accessory were used for
analysis. Mass spectral assays were obtained using a MS model
6210, Agilent technology instrument. The high resolution mass
spectra (HRMS) were obtained by TOF (time of flight) using ESI
(electrospray ionization) using the positive mode (ESI+) (analysis
at Université du Québec à Montréal).
Nuclear magnetic resonance (NMR) spectra were recorded on a
Varian 200 MHz NMR apparatus. Samples were dissolved in deute-
roacetone (acetone-d6) for data acquisition using tetramethylsilane
as internal standard (TMS, d 0.0 ppm for 1H NMR and 13C NMR).
Chemical shifts (d) are expressed in parts per million (ppm), the
coupling constants (J) are expressed in hertz (Hz). Multiplicities
are described by the following abbreviations: s for singlet, d for
doublet, dd for doublet of doublets, dt for doublet of triplets, t
for triplet, q for quartet, m for multiplet, #m for several multiplets,
br s for broad singlet, br d for broad doublet and br t for broad
triplet.
131.1 (4-C AP), 130.6 (2C, 3-C tyr), 128.4 (4-C tyr), 121.8 (2C, 3-C
CLL), 115.3 (4C, 2-C tyr and 2-C CLL), 78.8 (C(CH3)3), 57.0 (CHNH),
37.8 (CH2CHNH), 27.9 (3C, 3Â CH3). ESI + HRMS: [M+Na]+ calcu-
lated for C20H24N2NaO5 = 395.1577; found = 395.1572; [M+H]+ cal-
culated for C20H25N2O5 = 373.1758; found = 373.1756.
Spectral data for N-Boc-meta-hydroxyphenyl-
(8, made with meta-hydroxyaniline). IR (KBr,
L
-para-tyrosinamide
m
max, cmÀ1): 3100–
3600 (2Â O–H and 2Â N–H), 1671 (2Â C@O, NHCOO and NHCO),
1514 and 1240 (C–N–H). 1H NMR (Acetone-d6, d ppm): 4. 9.20
(1H, s, OH AP), 8.39 (1H, br s, OH tyr), 8.27 (1H, br s, NH AP),
7.34 (1H, t, J = 2.0 Hz, 4-CH AP), 7.10 (2H, d, J = 8.2 Hz, 3-CH tyr),
7.06 (1H, d, J = 7.8 Hz, 2-CH AP), 6.97 (1H, dt, J = 1.2 Hz and
J = 8.6 Hz, 5-CH AP), 6.74 (2H, d, J = 8.6 Hz, 2-CH tyr), 6.56 (1H,
dq, J = 1.2 Hz and J = 7.8 Hz, 6-CH AP), 6.16 (1H, d, J = 7.4 Hz, NH-
Boc), 4.45 (1H, m, CHNH),2.85–3.15 (2H, m, CH2CHNH), 1.36 (9H,
s, 3Â CH3). 13C NMR (Acetone-d6, d ppm): 170.8 (CONH), 158.0
(1-C AP), 156.3 (1-C tyr), 155.8 (NHCOO), 140.1(3-C AP), 130.6
(2C, 3-C tyr), 129.6 (5-C AP), 128.2 (4-C tyr), 115.4 (2C, 2-C tyr),
111.1 (2C, 4-C AP and 6-C AP), 107.2 (2-C AP), 78.6 (C(CH3)3),
57.2 (CHNH), 37.7 (CH2CHNH), 27.9 (3C, 3Â CH3). ESI + HRMS:
[M+Na]+
calculated
for
C20H24N2NaO5 = 395.1577;
found = 395.1578; [M+H]+ calculated for C20H25N2O5 = 373.1758;
found = 373.1760.
2.1.1. Preparation of the hydroxyphenyl-L-para-tyrosinamide–
chlorambucil molecules with the linear methodology
2.1.1.1. General procedure for the preparation of the N-Boc-hydroxy-
Spectral data for N-Boc-ortho-hydroxyphenyl-
(8, made with ortho-hydroxyaniline). IR (KBr,
L
-para-tyrosinamide
m
max, cmÀ1): 3100–
3500 (2Â O–H and 2Â N–H), 1686 (2Â C@O, NHCOO and NHCO),
1519 and 1257 (C–N–H). 1H NMR (Acetone-d6, d ppm): 9.21 (2H,
br s, 2Â OH), 8.26–9.19 (1H, br s, NH AP), 7.73 (1H, dd, J = 1.6 Hz
and J = 7.8 Hz, 3-CH AP), 7.15 (2H, d, J = 8.6 Hz, 3-CH tyr), 6.78
(2H, d, J = 8.6 Hz, 2-CH tyr), 6.74–7.04 (3H, #m, 4-CH AP, 5-CH
AP and 6-CH AP), 6.41 (1H, d, J = 7.0 Hz, NH-Boc), 4.55 (1H, m,
CHNH), 2.92–3.26 (2H, m, CH2CHNH), 1.37 (9H, s, 3Â CH3). 13C
NMR (Acetone-d6, d ppm): 171.8 (CONH), 156.4 (1-C tyr), 156.0
(NHCOO), 148.1 (1-C AP), 130.6 (2C, 3-C tyr), 128.3 (4-C tyr),
126.5 (2-C AP), 125.6 (5-C AP), 121.9 (3-C AP), 120.0 (4-C AP),
116.9 (6-C AP), 115.5 (2C, 2-C tyr), 79.3 (C(CH3)3), 57.3 (CHNH),
37.2 (CH2CHNH), 27.9 (3C, 3Â CH3). ESI + HRMS: [M+Na]+ calcu-
lated for C20H24N2NaO5 = 395.1577; found = 395.1582.
phenyl- -para-tyrosinamide (8). The hydroxyaniline (ortho, meta
L
and para) (3.09 mmol, 1.05 eq.) was dissolved in dimethylformam-
ide (5.0 mL) and triethylamine (2.94 mmol, 1.00 eq.) was added.
Separately, N-Boc-tyrosine 7 (2.94 mmol, 1.00 eq.) derivative was
dissolved in dimethylformamide (5.0 mL), and DCC (3.24 mmol,
1.10 eq.) followed by HOBt (3.24 mmol, 1.10 eq.) were added to
activate the acid function. Then, the hydroxyaniline solution was
added to the activated tyrosine solution. The resulting mixture
was stirred at room temperature for 25 h. A precipitate of dicyclo-
hexylurea was formed during the course of the reaction. The reac-
tion mixture was diluted with ethyl acetate (30 mL) and water
(20 mL), and then washed with water (4Â 20 mL). The organic
phase was dried with sodium sulfate, filtered and evaporated.
The product was further purified by flash chromatography (hex-
anes:acetone, 7:3) to give a pure compound in 92–96% yield.
2.1.1.2. General procedure for the preparation of the N-chlorambucil-
Spectral data for N-Boc-para-hydroxyphenyl-
L
-para-tyrosinamide
hydroxyphenyl-
L
-para-tyrosinamide (5). The N-Boc-hydroxyphenyl-
(8, made with para-hydroxyaniline). IR (KBr,
m
max, cmÀ1): 3100–
L-para-tyrosinamide 8 (0.16 mmol, 1.00 eq.) was dissolved in
3600 (2Â O–H and 2Â N–H), 1669 (2Â C@O, NHCOO and NHCO),
1516 and 1246 (C–N–H). 1H NMR (Acetone-d6, d ppm): 9.05 (1H,
s, OH AP), 8.24 (2H, br s, OH tyr and NH AP), 7.41 (2H, d,
J = 9.0 Hz, 3-CH AP), 7.10 (2H, d, J = 8.2 Hz, 3-CH tyr), 6.76 (2H, d,
J = 9.0 Hz, 2-CH AP), 6.74 (2H, d, J = 8.6 Hz, 2-CH tyr), 6.11 (1H, d,
J = 8.2 Hz, NH-Boc), 4.42 (1H, m, CHNH), 2.80–3.16 (2H, m,
CH2CHNH), 1.36 (9H, s, 3Â CH3). 13C NMR (Acetone-d6, d ppm):
170.2 (CONH), 156.3 (1-C tyr), 155.7 (NHCOO), 154.0 (1-C AP),
dichloromethane (5.0 mL) and trifluoroacetic acid (1.60 mmol,
10.00 eq.) was added. The solution was stirred at room tempera-
ture for 24 h. After evaporation, the resulting trifluoroacetic salt
was dissolved in dimethylformamide (2.5 mL) and neutralized
with triethylamine (0.16 mmol, 1.00 eq.). Simultaneously, chlor-
ambucil (0.24 mmol, 1.50 eq.) was dissolved in dimethylformam-
ide (2.5 mL) and DCC (0.25 mmol, 1.60 eq.) followed by HOBt
(0.25 mmol, 1.60 eq.) were added. Then, the hydroxyphenyltyrosa-

406
C. Descôteaux et al. / Steroids 77 (2012) 403–412
mide solution was added to the activated chlorambucil solution.
The resulting mixture was stirred at room temperature for 27 h.
A precipitate of dicyclohexylurea was formed during the course
of the reaction. The reaction mixture was diluted with ethyl ace-
tate (30 mL) and water (20 mL), and then washed with water (4Â
20 mL). The organic phase was dried with sodium sulfate, filtered
and evaporated. The product was further purified by flash chroma-
tography (hexanes:acetone, 7:3) to give a pure compound in 37–
44% yield.
2.1.1.3. General procedure for the preparation of the N-Boc-amino acid
alkyl chains (10, m = 5 or 10). A solution of a commercially available
aminoacid alkyl chain 9 (7.77 mmol, 1.00 eq.) in water (10.0 mL)
and dioxane (15.0 mL) was stirred at room temperature. Trielthy-
amine (11.65 mmol, 1.50 eq.) was added slowly followed by Boc-
ON (8.54 mmol, 1.10 eq.). The solution was kept for 23 h under
nitrogen. Afterwards, the mixture was diluted with diethyl ether
(75 mL) and water (50 mL) and then washed with water (6Â
50 mL). The aqueous layers were then combined and a few drops
of 10% aqueous HCl solution were slowly added in order to reach
pH 2.5. The resulting solution was then washed with dichloro-
methane (3Â 60 mL). The organic layers were washed with satu-
rated sodium chloride salt solution (1Â 50 mL). The organic
phase was dried with magnesium sulfate, filtered and evaporated.
Flash chromatography (hexanes:acetone, 4:1) was performed to
yield a colorless yellow viscous oil in 63–97% yield.
Spectral data for N-chlorambucil-para-hydroxyphenyl-
L
-para-tyro-
sinamide (5, made with para-hydroxyaniline). IR (KBr,
m
max, cmÀ1):
3100–3600 (2Â O–H and 2Â N–H), 1654 (2Â C@O, NHCO), 1517
and 1244 (C–N–H). 1H NMR (Acetone-d6, d ppm): 9.15 (1H, s, OH
AP), 8.18 and 8.20 (2H, 2s, OH tyr and NH AP), 7.41 (1H, d, partly
hidden, NH tyr), 7.40 (2H, d, J = 9.0 Hz, 3-CH AP), 7.11 (2H, d,
J = 8.6 Hz, 3-CH tyr), 7.01 (2H, d, J = 9.0 Hz, 3-CH CLL), 6.66–6.77
(6H, 3d, J = 9.0 Hz, J = 8.6 Hz and J = 9.0 Hz, 2-CH AP, 2-CH tyr and
2-CH CLL), 4.78 (1H, m, CHNH), 3.73 (8H, m, 2Â CH2Cl and 2Â
NCH2), 2.84–3.15 (2H, m, CH2CHNH), 2.44 (2H, t, J = 7.6 Hz,
CH2CH2Ph), 2.21 (2H, t, J = 7.2 Hz, NHCOCH2), 1.76–1.87 (2H, m,
CH2CH2CH2Ph). 13C NMR (Acetone-d6, d ppm): 172.9 (CO CLL),
170.0 (CO tyr), 156.3 (1-C tyr), 154.0 (1-C AP), 144.8 (1-C CLL),
131.2 (4-C AP), 130.9 (4-C tyr), 130.6 (2C, 3-C tyr), 129.7 (2C, 3-C
CLL), 128.5 (4-C CLL), 121.6 (2C, 3-C AP), 115.3 (4C, 2-C tyr and
2-C AP), 112.4 (2C, 2-C CLL), 55.6 (CHNH), 53.3 (2C, NCH2CH2Cl),
41.0 (2C, NCH2CH2Cl), 37.5, 35.2, 34.1, 27.7. ESI + HRMS: [M+H]+
calculated for C29H34Cl2N3O4 = 558.1921; found = 558.1929.
Spectral data for 6-Boc-amino-hexanoic acid (10, m = 5). IR (NaCl,
m
max, cmÀ1): 3339 (N–H), 2860–3100 (O–H), 1711 (2Â C@O), 1527
and 1250 (C–N–H), 1173 (C–O–C). 1H NMR (CDCl3, d ppm): 8.91
(1H, br s, COOH), 4.74 (1H, s, NH), 3.02 (2H, m, CH2NH), 2.27
(2H, t, J = 7.4 Hz, CH2COOH), 1.24–1.68 (6H, #m, 3Â CH2), 1.37 (s,
9H, 3Â CH3). 13C NMR (CDCl3, d ppm): 178.7 (COOH), 156.4
(OCONH), 79.4 ((CH3)3C), 40.5 (NHCH2), 34.1 (CH2COOH), 29.8
(NHCH2CH2), 28.6 (3Â CH3), 26.4 (CH2CH2CH2CH2CH2), 24.5
(CH2CH2COOH).
11H21NNaO4 = 254.1363; found = 254.1360.
Spectral data for 11-Boc-amino-undecanoic acid (10, m = 10). IR
(NaCl,
max, cmÀ1): 3365 (N–H), 2850–3100 (O–H), 1704 (C@O,
ESI + HRMS:
[M+Na]+
calculated
for
C
Spectral data for N-chlorambucil-meta-hydroxyphenyl-
L-para-
m
tyrosinamide (5, made with meta-hydroxyaniline). IR (NaCl, mmax
,
COOH), 1683 (C@O OCONH), 1526 (C–N–H), 1171 (C–O–C). 1H
NMR (CDCl3, d ppm): 10.98 (1H, br s, COOH), 4.55 (1H, s, NH),
3.06 (2H, m, CH2NH), 2.32 (2H, t, J = 7.4 Hz, CH2COOH), 1.61 (2H,
m, CH2), 1.43 (s, 11H, 3Â CH3 and 1Â CH2), 1.26 (br s, 12H, 6Â
CH2). 13C NMR (CDCl3, d ppm): 179.7 (COOH), 156.3 (OCONH),
79.3 ((CH3)3C), 40.8 (NHCH2), 34.3 (CH2COOH), 30.2, 29.6, 29.5,
29.4 (2C), 29.2, 28.6 (3Â CH3), 27.0, 24.9. ESI + HRMS: [M+Na]+ cal-
culated for C16H31NNaO4 = 324.2145; found = 324.2145.
cmÀ1): 3100–3500 (2Â O–H and 2Â N–H), 1656 (2Â C@O, NHCO),
1516 and 1239 (C–N–H). 1H NMR (Acetone-d6, d ppm): 9.30 (1H, s,
OH AP), 8.40 (1H, s, OH tyr), 8.22 (1H, s, NH AP), 7.44 (1H, d,
J = 8.2 Hz, NH tyr), 7.34 (1H, t, J = 2.1 Hz, 4-CH AP), 7.11 (2H, d,
J = 8.6 Hz, 3-CH tyr), 7.01 (2H, d, J = 9.0 Hz, 3-CH CLL), 6.96–7.14
(2H, m apparent, partly hidden, 2-CH and 5-CH AP), 6.73 (2H, d,
J = 8.6 Hz, 2-CH tyr), 6.68 (2H, J = 8.6 Hz, 2-CH CLL), 6.55 (1H, dq,
J = 1.2 Hz and J = 7.8 Hz, 6-CH AP), 4.80 (1H, m, CHNH), 3.73 (8H,
m, 2Â CH2Cl and 2Â NCH2), 2.84–3.17 (2H, m, CH2CHNH), 2.44
(2H, t, J = 7.6 Hz, CH2CH2Ph), 2.22 (2H, t, J = 7.2 Hz, NHCOCH2),
1.73–1.87 (2H, m, CH2CH2CH2Ph). 13C NMR (Acetone-d6, d ppm):
173.1 (CO CLL), 170.4 (CO tyr), 158.0 (1-C AP), 156.3 (1-C tyr),
144.8 (1-C CLL), 140.2 (3-C AP), 130.8 (4-C tyr), 130.5 (2C, 3-C
tyr), 129.7 (2C, 3-C CLL), 129.6 (5-C AP), 128.4 (4-C CLL), 115.3
(2C, 2-C tyr), 112.4 (2C, 2-C CLL), 110.9 (2C, 4-C AP and 6-C AP),
107.0 (2-C AP), 55.8 (CHNH), 53.3 (2C, NCH2CH2Cl), 41.0 (2C,
NCH2CH2Cl), 37.3, 35.2, 34.0, 27.7. ESI + HRMS: [M+H]+ calculated
for C29H34Cl2N3O4 = 558.1921; found = 558.1925.
2.1.1.4. General procedure for the preparation of the N-((N-Boc-
amino)alcanoyl)-p-hydroxyphenyl-L-para-tyrosinamide (12). The N-
Boc-p-hydroxyphenyl- -para-tyrosinamide
L
derivative
8
(0.57 mmol, 1.00 eq.) was dissolved in dichloromethane (3.0 mL)
and trifluoroacetic acid (5.70 mmol, 10.00 eq.) was added. The
solution was stirred at room temperature for 29 h. After evapora-
tion, the resulting trifluoroacetic salt was dissolved in dimethyl-
formamide (2.0 mL) and neutralized with triethylamine
(0.57 mmol, 1.00 eq.). Separately, N-Boc-amino acid alkyl chain
10 (m = 5, 0.60 mmol, 1.05 eq.) was dissolved in dimethylformam-
ide (2.0 mL), and DCC (0.63 mmol, 1.10 eq.) followed by HOBt
(0.63 mmol, 1.10 eq.) were added. Then, the p-hydroxyphenylty-
rosamide solution was added to the activated amino acid chain
solution. The resulting mixture was stirred at room temperature
for 28 h. A precipitate of dicyclohexylurea was formed during the
course of the reaction. The reaction mixture was diluted with ethyl
acetate (30 mL) and water (20 mL), and then washed with water
(4Â 20 mL). The organic phase was dried with sodium sulfate, fil-
tered and evaporated. The product was further purified by flash
chromatography (hexanes:acetone, 7:3) to give a pure compound
in 45–46% yield.
Spectral data for N-chlorambucil-ortho-hydroxyphenyl-
L-para-
tyrosinamide (5, made with ortho-hydroxyaniline). IR (NaCl, mmax
,
cmÀ1): 3100–3400 (2Â O–H and 2Â N–H), 1704 and 1656 (2Â
C@O, NHCO), 1523 and 1257 (C–N–H). 1H NMR (Acetone-d6, d
ppm): 9.16 (1H, s, OH AP), 8.29 (1H, d, J = 7.5 Hz, NH tyr), 8.16
(1H, s, OH tyr), 7.58 (1H, d, J = 7.5 Hz, 3-CH AP), 6.97–7.17 (7H,
#m apparent, 3-CH CLL, 3-CH tyr, 4-CH AP, 5-CH AP and NH AP),
6.66–6.80 (5H, #m apparent, 2-CH tyr, 2-CH CLL and 6-CH AP),
4.72 (1H, m, CHNH), 3.73 (8H, m, 2Â CH2Cl and 2Â NCH2), 2.79–
3.24 (2H, m, CH2CHNH), 2.63 (2H, t, J = 7.6 Hz, CH2CH2Ph), 2.20
(2H, t, J = 7.4 Hz, CHNHCOCH2), 1.76–1.87 (2H, m, CH2CH2CH2Ph).
13C NMR (Acetone-d6, d ppm): 174.0 (CONH), 171.6 (CO tyr),
156.3 (1-C tyr), 145.0 (1-C AP), 144.9 (1-C CLL), 130.7 (4-C tyr),
130.5 (2C, 3-C tyr), 129.7 (2C, 3-C CLL), 128.6 (4-C CLL), 126.0 (2-
C AP), 123.8 (5-C AP), 122.9 (3-C AP), 121.5 (4-C AP), 115.4 (2C,
2-C tyr), 112.5 (6-C AP), 112.4 (2C, 2-C CLL), 55.9 (CHNH), 53.3
(2C, NCH2CH2Cl), 41.0 (2C, NCH2CH2Cl), 35.2, 34.1, 34.0, 27.8.
ESI + HRMS: [M+H]+ calculated for C29H34Cl2N3O4 = 558.1921;
found = 558.1925.
Spectral data for N-((N-Boc-amino)hexanoyl)-para-hydroxy-
phenyl- -para-tyrosinamide (12, m = 5). IR (KBr, mmax,
cmÀ1):
L
3150–3450 (2Â O–H and 3Â N–H), 1666 (3Â C@O, NHCOO and
2Â NHCO). 1H NMR (Acetone-d6, d ppm): 9.36 (1H, s, OH AP),
8.44 (1H, br s, OH tyr), 7.71 (2H, d, J = 8.2 Hz, NH AP), 7.41 (2H,
d, J = 8.4 Hz, 3-CH AP), 7.10 (2H, d, J = 8.2 Hz, 3-CH tyr), 6.76 and
6.74 (4H, 2d, J = 9.0 Hz and J = 8.6 Hz, 2-CH AP and 2-CH tyr),
5.97 (1H, br t, J = 5.9 Hz, NH-Boc), 4.86 (1H, m, CHNH), 2.85–3.18

C. Descôteaux et al. / Steroids 77 (2012) 403–412
407
(4H, 2 m, CH2CHNH and CH2NHCO), 2.19 (2H, t apparent, partly
hidden, NHCOCH2), 1.41 (9H, s, 3Â CH3), 1.21–1.51 (6H, #m, 3Â
CH2). 13C NMR (Acetone-d6, d ppm): 173.4 (CONH), 170.3 (CONH),
156.3 (2C, 1-C tyr and NHCOO), 154.1 (1-C AP), 131.0 (4-C AP),
130.6 (2C, 3-C tyr), 128.4 (4-C tyr), 121.8 (2C, 3-C CLL), 115.4
(4C, 2-C tyr and 2-C CLL), 78.1 (C(CH3)3), 59.9 (CHNH), 40.4
(CH2NH), 37.8 (CH2CHNH), 35.9 (NHCOCH2), 29.6, 28.1 (3C, 3Â
CH3), 26.4, 25.5. ESI + HRMS: [M+Na]+ calculated for
tyr), 130.5 (2C, 3-C tyr), 129.7 (2C, 3-C CLL), 128.5 (4-C CLL),
121.6 (2C, 3-C AP), 115.4 (2C, 2-C tyr), 115.3 (2C, 2-C AP), 112.5
(2C, 2-C CLL), 55.5 (CHNH), 53.3 (2C, NCH2CH2Cl), 41.0 (2C,
NCH2CH2Cl), 39.1, 37.5, 35.9, 35.6, 34.2, 27.9, 26.4, 25.5, (1C hid-
den).
35H45Cl2N4O5 = 671.2762; found = 671.2754.
Spectral data for N-((N-chlorambucilamino)undecanoyl)-para-
hydroxyphenyl- -para-tyrosinamide (6, m = 10). IR (KBr, mmax
ESI + HRMS:
[M+H]+
calculated
for
C
L
,
C
26H35N3NaO6 = 508.2418; found = 508.2422; [M+H]+ calculated
cmÀ1): 3100–3500 (2Â O–H and 3Â N–H), 1643 (3Â C@O, NHCO),
1517 and 1227 (C–N–H). 1H NMR (Acetone-d6, d ppm): 9.11 (1H, s,
OH AP), 8.34 (2H, br s, OH tyr and NH AP), 7.41, (2H, d, J = 9.0 Hz, 3-
CH AP), 7.36 (1H, d, partly hidden, NH tyr), 7.15 (1H, br t, J = 6.0 Hz,
CH2NHCO), 7.09 (2H, d, J = 8.6 Hz, 3-CH tyr), 7.06 (2H, d, J = 8.6 Hz,
3-CH CLL), 6.69–6.78 (6H, 3d, J = 9.0 Hz, J = 8.6 Hz and J = 8.6 Hz, 2-
CH AP, 2-CH tyr and 2-CH CLL), 4.73 (1H, m, CHNH), 3.74 (8H, m,
2Â CH2Cl and 2Â NCH2), 2.83–3.25 (4H, 2 m, CH2CHNH and
CH2NHCO), 2.52 (2H, t, J = 7.6 Hz, CH2CH2Ph), 2.18 (4H, 2t over-
lapped, J = 7.2 Hz, CH2NHCOCH2 and CHNHCOCH2), 1.82–1.93
(2H, m, CH2CH2CH2Ph), 1.20–1.56 (16H, m, 8Â CH2). 13C NMR (Ace-
tone-d6, d ppm): 172.98 (CONH), 172.71 (CONH), 169.8 (CONH tyr),
156.4 (1-C tyr), 154.0 (1-C AP), 145.0 (1-C CLL), 131.2 (4-C AP),
130.9 (4-C tyr), 130.5 (2C, 3-C tyr), 129.7 (2C, 3-C CLL), 128.5 (4-
C CLL), 121.5 (2C, 3-C AP), 115.3 (4C, 2-C tyr and 2-C AP), 112.5
(2C, 2-C CLL), 55.6 (CHNH), 53.3 (2C, NCH2CH2Cl), 41.0 (2C,
NCH2CH2Cl), 39.0, 37.3, 35.9, 35.6, 34.2, 29.8 (2C), 29.4 (3C), 27.9,
26.8, 25.7, (1C hidden). ESI + HRMS: [M+H]+ calculated for
for C26H36N3O6 = 486.2599; found = 486.2606.
Spectral data for N-((N-Boc-amino)undecanoyl)-para-hydroxy-
phenyl- -para-tyrosinamide (12, m = 10). IR (KBr, mmax,
cmÀ1):
L
3100–3400 (2Â O–H and 3Â N–H), 1667 and 1681 (3Â C@O,
NHCOO and 2Â NHCO). 1H NMR (Acetone-d6, d ppm): 9.05 (1H, s,
OH AP), 8.18 (2H, s, OH tyr and NH), 7.40 (2H, d, J = 9.0 Hz, 3-CH
AP), 7.28 (1H, br d, J = 7.8 Hz, NH), 7.09 (2H, d, J = 8.6 Hz, 3-CH
tyr), 6.73 and 6.75 (4H, 2d, J = 8.6 Hz and J = 9.0 Hz, 2-CH tyr and
2-CH AP), 5.93 (1H, br s, NH-Boc), 4.70 (1H, m, CHNH), 2.94–3.10
(4H, 2 m, CH2CHNH and CH2NHCO), 2.19 (2H, t, J = 7.2 Hz,
NHCOCH2), 1.40 (9H, s, 3Â CH3), 1.26–1.50 (16H, #m, 8Â CH2).
13C NMR (Acetone-d6, d ppm): 172.7 (CONH), 169.6 (CONH),
156.3 (2C, 1-C tyr and NHCOO), 153.9 (1-C AP), 131.3 (4-C AP),
130.5 (2C, 3-C tyr), 128.5 (4-C tyr), 121.5 (2C, 3-C CLL), 115.2
(4C, 2-C tyr and 2-C CLL), 77.7 (C(CH3)3), 55.5 (CHNH), 40.4
(CH2NH), 37.3 (CH2CHNH), 35.9 (NHCOCH2), 32.0, 26.8, 25.7, 22.7
(7C
C
hidden).
ESI + HRMS:
[M+Na]+
calculated
for
31H45N3NaO6 = 578.3201; found = 578.3200; [M+H]+ calculated
C40H55Cl2N4O5 = 741.3544; found = 741.3548.
for C31H46N3O6 = 556.3381; found = 556.3381.
2.1.2. Preparation of the hydroxyphenyl- -para-tyrosinamide–
L
2.1.1.5. General procedure for the preparation of the N-((N-chloram-
chlorambucil molecules with the convergent methodology
bucilamino)alcanoyl)-p-hydroxyphenyl-
N-((N-Boc-amino)alcanoyl)-para-hydroxyphenyl-
L
-para-tyrosinamide (6). The
2.1.2.1. General procedure for the preparation of the N-chlorambucil-
amino acid alkyl chain (14, m = 5 or 10). The amino acid 9
(6.57 mmol, 2.00 eq.) was dissolved in dichloromethane (2.0 mL).
Hexamethyldisilazane (21.69 mmol, 6.60 eq.) and concentrated
sulfuric acid (1 drop, cat) were added. The mixture was stirred
and heated to reflux under nitrogen atmosphere until complete
dissolution. The solution was kept to reflux for 0.5 h. After cooling
down, benzene (2.0 mL), triethylamine (4.91 mmol, 0.67 eq.) and
chlorotrimethylsilane (4.91 mmol, 0.67 eq.) were added. The
resulting mixture was stirred at room temperature for 12 h. In an-
other flask, chlorambucil (3.29 mmol, 1.00 eq.) was dissolved in
dichloromethane (2.0 mL) at 0 °C. Triethylamine (3.62 mmol,
1.10 eq.) and isobutylchloroformate (3.62 mol, 1.10 eq) were
added and the mixture was kept at 0 °C for 1 h. Then, the chloram-
bucil solution was added to the activated amino acid solution and
the mixture was stirred at room temperature for 4 h. Work-up was
done by diluting with ethyl acetate (20 mL) and by washing the or-
ganic phase with HCl 10% solution (2Â 15 mL) and with a saturated
sodium chloride solution (2Â 15 mL). The organic phase was dried
with anhydrous magnesium sulfate, filtered and evaporated. The
product was purified by flash chromatography (hexanes:acetone,
7:3) to give the desired material in 68–98% yield.
L
-para-tyrosina-
mide 12 (0.17 mmol, 1.10 eq.) was dissolved in dichloromethane
(3.0 mL) and treated with trifluoroacetic acid (1.70 mmol,
10.00 eq.). The solution was stirred at room temperature for
22.5 h. After evaporation, the resulting trifluoroacetic salt was dis-
solved in dimethylformamide (1.0 mL) and neutralized with trieth-
ylamine (0.18 mmol, 1.05 eq.). Simultaneously, chlorambucil
(0.17 mmol, 1.00 eq.) was dissolved in dimethylformamide
(1.0 mL) and treated with DCC (0.19 mmol, 1.10 eq.) and with HOBt
(0.19 mmol, 1.10 eq.). Then, the N-alkylated tyrosamide solution
was added to the activated chlorambucil solution. The resulting
mixture was stirred at room temperature for 44 h. A precipitate
of dicyclohexylurea was formed during the course of the reaction.
The reaction mixture was diluted with ethyl acetate (20 mL) and
water (15 mL), and then washed with water (4Â 15 mL). The or-
ganic phase was dried with sodium sulfate, filtered and evapo-
rated. The product was further purified by flash chromatography
(chloroform:methanol, 97:3) to give a pure compound in 17–19%
yield.
Spectral data for
N-((N-chlorambucilamino)hexanoyl)-para-
hydroxyphenyl- -para-tyrosinamide (6, m = 5). IR (KBr,
L
m
max, cmÀ1):
3100–3500 (2Â O–H and 3Â N–H), 1639–1613 (3Â C@O, NHCO),
1516 and 1230 (C–N–H). 1H NMR (Acetone-d6, d ppm): 9.13 (1H,
s, OH AP), 8.24 and 8.34 (2H, 2s, OH tyr and NH AP), 7.41 (2H, d,
J = 9.0 Hz, 3-CH AP), 7.34 (1H, d, J = 8.20 Hz, NH tyr), 7.13 (1H, br
t, J = 6.0 Hz, CH2NHCO), 7.09 (2H, d, J = 8.6 Hz, 3-CH tyr), 7.06
(2H, d, J = 8.6 Hz, 3-CH CLL), 6.69–6.78 (6H, 3d, J = 9.0 Hz,
J = 8.6 Hz and J = 8.6 Hz, 2-CH AP, 2-CH tyr and 2-CH CLL), 4.74
(1H, m, CHNH), 3.74 (8H, m, 2Â CH2Cl and 2Â NCH2), 2.80–3.20
(4H, 2 m, CH2CHNH and CH2NHCO), 2.51 (2H, t, J = 7.6 Hz,
CH2CH2Ph), 2.17 (2H, t, J = 7.6 Hz, CH2NHCOCH2), 2.11–2.19 (2H,
t partly hidden, CHNHCOCH2), 1.81–1.92 (2H, m, CH2CH2CH2Ph),
1.12–1.54 (6H, #m, 3Â CH2). 13C NMR (Acetone-d6, d ppm):
172.74 (CONH), 172.70 (CONH), 170.0 (CONH tyr), 156.4 (1-C
tyr), 154.0 (1-C AP), 145.0 (1-C CLL), 131.2 (4-C AP), 130.9 (4-C
Spectral data for N-6-chlorambucil-hexanoic acid (14, m = 5). IR
(NaCl,
m
max, cmÀ1): 3300 (N–H), 3200–2600 (O–H), 1716 (C@O,
COOH), 1622 (C@O, NHCOO), 1522 and 1255 (C–N–H). 1H NMR
(Acetone-d6, d ppm): 7.26 (1H, br s, NH), 7.06 (2H, d, J = 8.6 Hz,
3-CH CLL), 6.70 (2H, d, J = 8.6 Hz, 2-CH CLL), 3.73 (8H, m, 2Â CH2Cl
and 2Â NCH2), 3.19 (2H, m, CH2NHCO), 2.51 (2H, t, J = 7.4 Hz,
CH2CH2Ph), 2.27 (2H, t, J = 7.2 Hz, CH2COOH), 2.18 (2H, t,
J = 7.4 Hz, CH2NHCOCH2), 1.78–1.96 (2H, m, CH2CH2CH2Ph),
1.28–1.67 (6H, m, 3Â CH2), (1H, COOH hidden). 13C NMR (Ace-
tone-d6, d ppm): 174.3 (COOH), 173.0 (CONH), 144.8 (1-C CLL),
130.9 (4-C CLL), 129.7 (2C, 3-C CLL), 112.5 (2C, 2-C CLL), 53.3 (2C,
2Â NCH2CH2Cl), 41.0 (2Â C, 2Â NCH2CH2Cl), 39.1, 35.6, 34.3,
33.6, 29.4, 27.9, 26.5, 24.7. ESI + HRMS: [M+H]+ calculated for
C
20H31Cl2N2O3 = 417.1706; found = 417.1707.

408
C. Descôteaux et al. / Steroids 77 (2012) 403–412
Spectral data for N-6-chlorambucil-undecanoic acid (14, m = 10).
IR (NaCl,
max, cmÀ1): 3300 (N–H), 3200–2600 (O–H), 1714 (C@O,
34.2, 27.9, 26.3, 25.5, (1C hidden). ESI + HRMS: [M+H]+ calculated
for C35H45Cl2N4O5 = 671.2762; found = 671.2757.
m
COOH), 1618 (C@O, NHCOO), 1526 and 1250 (C–N–H). 1H NMR
(Acetone-d6, d ppm): 7.07 (2H, d, J = 8.6 Hz, 3-CH CLL), 7.03 (1H,
br s, NH), 6.72 (2H, d, J = 9.0 Hz, 2-CH CLL), 3.75 (8H, m, 2Â CH2Cl
and 2Â NCH2), 3.17 (2H, m, CH2NHCO), 2.90 (1H, br s, COOH), 2.51
(2H, t, J = 7.4 Hz, CH2CH2Ph), 2.27 (2H, t, J = 7.4 Hz, CH2COOH), 2.14
(2H, t, J = 7.2 Hz, CH2NHCOCH2), 1.80–1.92 (2H, m, CH2CH2CH2Ph),
1.41–1.62 (4H, m, 2Â CH2), 1.30 (12H, 6Â CH2). 13C NMR (Acetone-
d6, d ppm): 174.0 (COOH), 172.1 (CONH), 144.9 (1-C CLL), 131.0 (4-
C CLL), 129.7 (2C, 3-C CLL), 112.5 (2C, 2-C CLL), 53.3 (2C, 2Â
NCH2CH2Cl), 41.0 (2C, 2Â NCH2CH2Cl), 39.0, 35.5, 34.2, 33.6, 29.8,
29.5 (2C), 29.3 (3C), 27.9, 27.0, 25.0. ESI + HRMS: [M+H]+ calculated
for C25H41Cl2N2O3 = 487.2489; found = 487.2485.
Spectral data for N-((N-chlorambucilamino)undecanoyl)-meta-
hydroxyphenyl- -para-tyrosinamide (isomer of 6, m = 10). IR (NaCl,
L
m
max, cmÀ1): 3100–3500 (2Â O–H and 3Â N–H), 1704, 1649 and
1622 (3Â C@O, NHCO), 1526 and 1222 (C–N–H). 1H NMR (Ace-
tone-d6, d ppm): 9.34 (1H, s, OH AP), 8.55 (1H, s, OH tyr), 8.45
(1H, s, NH AP), 7.47 (1H, d, J = 8.2 Hz, NH tyr), 7.33 (1H, t,
J = 2.1 Hz, 4-CH AP), 7.21 (1H, br t, J = 5.7 Hz, CH2NHCO), 7.10
(2H, d, J = 8.2 Hz, 3-CH tyr), 7.06 (2H, d, J = 8.2 Hz, 3-CH CLL),
6.96–7.12 (2H, m apparent, partly hidden, 2-CH and 5-CH AP),
6.74 (2H, d, J = 8.2 Hz, 2-CH tyr), 6.70 (2H, J = 8.6 Hz, 2-CH CLL),
6.56 (1H, dq, J = 1.2 Hz and J = 7.8 Hz, 6-CH AP), 4.78 (1H, m,
CHNH), 3.73 (8H, m, 2Â CH2Cl and 2Â NCH2), 3.20 (2H, q,
J = 5.9 Hz, CH2NHCO), 2.83–3.15 (2H, m, CH2CHNH), 2.52 (2H, t,
J = 7.6 Hz, CH2CH2Ph), 2.19 (4H, 2t overlapped, J = 7.2 Hz,
2.1.2.2. General procedure for the preparation of the N-((N-chloram-
CH2NHCOCH2
and
CHNHCOCH2),
1.79–1.96
(2H,
m,
bucilamino)alcanoyl)-hydroxyphenyl-L-para-tyrosinamide (6). The
CH2CH2CH2Ph), 1.22–1.59 (16H, #m and s, 8Â CH2). 13C NMR (Ace-
tone-d6, d ppm): 173.3 (CONH), 172.9 (CONH), 170.4 (CO tyr),
158.1 (1-C AP), 156.5 (1-C tyr), 144.9 (1-C CLL), 140.3 (3-C AP),
130.9 (4-C tyr), 130.5 (2C, 3-C tyr), 129.7 (2C, 3-C CLL), 129.6 (5-
C AP), 128.3 (4-C CLL), 115.3 (2C, 2-C tyr), 112.4 (2C, 2-C CLL),
110.9 (2C, 4-C AP and 6-C AP), 107.1 (2-C AP), 55.8 (CHNH), 53.3
(2C, NCH2CH2Cl), 41.0 (2C, NCH2CH2Cl), 39.1, 35.9, 35.7, 34.3,
29.4 (3C), 29.1 (2C), 28.0, 28.0 (2C), 26.8, 25.7. ESI + HRMS:
appropriate N-Boc-hydroxyphenyl-L-para-tyrosinamide
8
(0.38 mmol, 1.00 eq.) was dissolved in dichloromethane (3.0 mL)
and treated with trifluoroacetic acid (3.80 mmol, 10.0 eq.). The
solution was stirred at room temperature for 24 h. After evapora-
tion, the resulting trifluoroacetic salt was dissolved in dimethyl-
formamide (1.5 mL) and neutralized with triethylamine
(0.40 mmol, 1.05 eq.). At the same time, the N-chlorambucil-amino
acid derivative 6 (0.40 mmol, 1.05 eq.) was dissolved in dimethyl-
formamide (1.5 mL), and DCC (0.42 mmol, 1.10 eq.) followed by
HOBt (0.42 mmol, 1.10 eq.) were added. Then, the hydrox-
yphenyltyrosamide solution was added to the activated chloram-
bucil-amino acid derivative (14) solution. The resulting mixture
was stirred at room temperature for 29 h. A precipitate of dicyclo-
hexylurea was formed during the course of the reaction. The reac-
tion mixture was diluted with ethyl acetate (30 mL) and water
(20 mL), and then washed with water (4Â 20 mL). The organic
phase was dried with sodium sulfate, filtered and evaporated.
The product was further purified by flash chromatography (hex-
anes:acetone, 7:3) to give a pure compound in 25–51% yield.
[M+H]+
found = 741.3541.
Spectral data for N-((N-chlorambucilamino)hexanoyl)-ortho-
calculated
for
C40H55Cl2N4O5 = 741.3544;
hydroxyphenyl- -para-tyrosinamide (isomer of 6, m = 5). IR (NaCl,
L
m
max, cmÀ1): 3100–3400 (2Â O–H and 3Â N–H), 1646 (3Â C@O,
NHCO), 1526 and 1246 (C–N–H). 1H NMR (Acetone-d6, d ppm):
9.60 (1H, br s, OH AP), 9.17 (1H, br s, OH tyr), 7.90 (1H, d,
J = 7.8 Hz, NH tyr), 7.68 (1H, d, J = 7.8 Hz, 3-CH AP), 7.45 (1H, br t,
J = 5.7 Hz, CH2NHCO), 7.13 (2H, d, J = 8.6 Hz, 3-CH tyr), 7.06 (2H,
d, J = 8.6 Hz, 3-CH CLL), 6.96 (2H, 2Â d overlapped, J = 3.9 Hz, 4-
CH AP and 5-CH AP), 6.78 (2H, d, J = 8.6 Hz, 2-CH tyr), 6.70 (2H,
d, J = 8.6 Hz, 2-CH CLL), 6.68–6.83 (2H, m apparent, partly hidden,
6-CH AP and NH AP), 4.86 (1H, m, CHNH), 3.72 (8H, m, 2Â CH2Cl
and 2Â NCH2), 3.16 (2H, m, CH2NHCO), 2.86–3.29 (2H, m,
CH2CHNH), 2.51 (2H, t, J = 7.4 Hz, CH2CH2Ph), 2.22 (4H, m appar-
ent, CH2NHCOCH2 and CHNHCOCH2), 1.80–1.95 (2H, m,
CH2CH2CH2Ph), 1.11–1.63 (6H, #m, 3Â CH2). 13C NMR (Acetone-
d6, d ppm): 173.7 (CONH), 173.6 (CONH), 170.8 (CO tyr), 156.5
(1-C tyr), 147.7 (1-C AP), 144.8 (1-C CLL), 130.9 (4-C tyr), 130.5
(2C, 3-C tyr), 129.7 (2C, 3-C CLL), 128.4 (4-C CLL), 126.7 (2-C AP),
125.0 (5-C AP), 121.2 (3-C AP), 120.0 (4-C AP), 116.9 (6-C AP),
115.5 (2C, 2-C tyr), 112.5 (2C, 2-C CLL), 55.9 (CHNH), 53.3 (2C,
NCH2CH2Cl), 41.0 (2C, NCH2CH2Cl), 39.3, 36.7, 35.9, 35.6, 34.2,
33.6, 27.9, 26.3, 25.6. ESI + HRMS: [M+H]+ calculated for
C35H45Cl2N4O5 = 671.2762; found = 671.2763.
Spectral data for
N-((N-chlorambucilamino)hexanoyl)-para-
hydroxyphenyl- -para-tyrosinamide (6, m = 5). The spectral data of
L
this derivative were identical to those previously described for
the same derivative made by the linear synthesis.
Spectral data for N-((N-chlorambucilamino)undecanoyl)-para-
hydroxyphenyl- -para-tyrosinamide (6, m = 10). The spectral data
L
of this derivative were identical to those previously described for
the same derivative made by the linear synthesis.
Spectral data for
hydroxyphenyl-
m
N-((N-chlorambucilamino)hexanoyl)-meta-
L
-para-tyrosinamide (isomer of 6, m = 5). IR (NaCl,
max, cmÀ1): 3100–3600 (2Â O–H and 3Â N–H), 1619 (3Â C@O,
NHCO), 1515 and 1239 (C–N–H). 1H NMR (Acetone-d6, d ppm):
9.20 (1H, s, OH AP), 8.49 (1H, s, OH tyr), 8.34 (1H, s, NH AP),
7.29–7.35 (2H, d, J = 8.2 Hz, NH tyr and t apparent, J = 2.34 Hz, 4-
CH AP), 7.13 (1H, br t, J = 5.1 Hz, CH2NHCO), 7.09 (2H, d,
J = 8.2 Hz, 3-CH tyr), 7.07 (2H, d, J = 8.6 Hz, 3-CH CLL), 6.99–7.11
(2H, m apparent, partly hidden, 2-CH and 5-CH AP), 6.74 (2H, d,
J = 8.2 Hz, 2-CH tyr), 6.71 (2H, J = 9.0 Hz, 2-CH CLL), 6.54 (1H, dq,
J = 1.2 Hz and J = 7.8 Hz, 6-CH AP), 4.73 (1H, m, CHNH), 3.74 (8H,
m, 2Â CH2Cl and 2Â NCH2), 2.79–3.21 (2H, m, CH2CHNH), 3.15
(2H, m, CH2NHCO), 2.52 (2H, t, J = 7.6 Hz, CH2CH2Ph), 2.18 (4H,
2t overlapped, J = 7.2 Hz, CH2NHCOCH2 and CHNHCOCH2), 1.82–
1.93 (2H, m, CH2CH2CH2Ph), 1.18–1.59 (6H, #m, 3Â CH2). 13C
NMR (Acetone-d6, d ppm): 172.9 (CONH), 172.8 (CONH), 170.3
(CO tyr), 158.1 (1-C AP), 156.4 (1-C tyr), 144.9 (1-C CLL), 140.2
(3-C AP), 130.9 (4-C tyr), 130.5 (2C, 3-C tyr), 129.7 (2C, 3-C CLL),
129.6 (5-C AP), 128.4 (4-C CLL), 115.4 (2C, 2-C tyr), 112.4 (2C, 2-
C CLL), 110.9 (2C, 4-C AP and 6-C AP), 107.0 (2-C AP), 55.7 (CHNH),
53.3 (2C, NCH2CH2Cl), 41.0 (2C, NCH2CH2Cl), 39.1, 37.1, 35.8, 35.6,
Spectral data for N-((N-chlorambucilamino)undecanoyl)-ortho-
hydroxyphenyl- -para-tyrosinamide (isomer of 6, m = 10). IR (NaCl,
L
m
max, cmÀ1): 3100–3600 (2Â O–H and 3Â N–H), 1690, 1646 and
1615 (3Â C@O, NHCO), 1526 and 1250 (C–N–H). 1H NMR (Ace-
tone-d6, d ppm): 9.21 (1H, br s, OH tyr), 7.70 (1H, dd, J = 1.4 Hz
and J = 7.8 Hz, 3-CH AP), 7.49 (1H, d, J = 8.2 Hz, NH tyr), 7.13 (2H,
d, J = 8.6 Hz, 3-CH tyr), 7.11–7.15 (1H, hidden, CH2NHCO), 7.06
(2H, d, J = 8.6 Hz, 3-CH CLL), 6.83–7.02 (2H, #m, 4-CH AP and 5-
CH AP), 6.76 (2H, d, J = 8.6 Hz, 2-CH tyr), 6.71 (2H, d, J = 9.0 Hz,
2-CH CLL), 6.68–6.80 (2H, m apparent, partly hidden, 6-CH AP
and NH AP), 4.83 (1H, m, CHNH), 3.74 (8H, m, 2Â CH2Cl and 2Â
NCH2), 3.20 (2H, m, CH2NHCO), 2.88–3.25 (2H, m, CH2CHNH),
2.52 (2H, t, J = 7.4 Hz, CH2CH2Ph), 2.22 (4H, 2Â d, J = 7.2 Hz and
J = 7.4 Hz, CH2NHCOCH2 and CHNHCOCH2), 1.82–1.93 (2H, m,
CH2CH2CH2Ph), 1.24–1.56 (16H, #m and s, 8Â CH2), (1H, OH AP,
hidden). 13C NMR (Acetone-d6, d ppm): 173.2 (CONH), 172.7

C. Descôteaux et al. / Steroids 77 (2012) 403–412
409
(CONH), 171.2 (CO tyr), 156.5 (1-C tyr), 148.1 (1-C AP), 144.9 (1-C
CLL), 130.9 (4-C tyr), 130.5 (2C, 3-C tyr), 129.7 (2C, 3-C CLL), 128.3
(4-C CLL), 126.7 (2-C AP), 125.4 (5-C AP), 121.7 (3-C AP), 119.8 (4-C
AP), 116.9 (6-C AP), 115.4 (2C, 2-C tyr), 112.5 (2C, 2-C CLL), 55.6
(CHNH), 53.3 (2C, NCH2CH2Cl), 41.0 (2C, NCH2CH2Cl), 39.0, 36.7,
35.9, 35.8, 35.6, 34.2, 33.6, 29.0 (2C), 28.0, 26.7, 25.7, (2C hidden).
ESI + HRMS: [M+H]+ calculated for C40H55Cl2N4O5 = 741.3544;
found = 741.3531.
on the treatment day, and the growth inhibition percentage was
calculated for each drug contact period.
Ortho-, meta- and para-hydroxyphenyl-tyrosinamide–chloram-
bucil derivatives bearing a 5 carbon atoms spacer were used for
a more exhaustive biological analysis. The three selected molecules
were submitted to the same MTT assay as described earlier, on a
panel of female hormone-dependent (ER+) and hormone-indepen-
dent (ERÀ) breast, uterine and ovarian cancer cell lines. The human
cancerous cell lines used were MCF-7 and ZR-75-1 (breast; ER+),
MDA-MB-231, MDA-MB-436 and MDA-MB-468 (breast; ERÀ),
Ishikawa (uterine; ER+), HEC-1A (uterine; ERÀ), OVCAR-3 (ovarian;
ER+) and A2780 (ovarian; ERÀ).
2.2. Biology
2.2.1. In vitro cytotoxic activity
The cytotoxicity of all the hydroxyphenyl-tyrosinamide–chlor-
ambucil regioisomers (5 and 6, m = 5 or 10) was evaluated on
MCF-7 (ER+) and MDA-MD-231 (ERÀ) breast cancer cell lines.
Chlorambucil was used for control. MTT (3-(4,5-dimethylthiazol-
2-yl)-phenyl-tetrazolium bromide) assay, a standard colorimetric
test, was used for measuring cellular proliferation [22]. Briefly, tu-
mor cell lines were added into 96-well tissue culture plates in cul-
ture medium and incubated at 37 °C in a 5% CO2 atmosphere.
Dilutions were done using cremophore:ethanol (1:1) solution.
The cremophore:ethanol solution was used in line with our stan-
dard protocol for the in vivo biological assays that we might per-
form in the future [14]. Cells were incubated with or without
drugs for 72 h. Culture plates were processed using MTT for 3.5 h
afterwards SDS solubilization solution (HCl 0.010 M, sodium dode-
cyl sulfate solution 10%) was added. The absorbance was read
using a scanning multiwell spectrophotometer (FLUOStar OPTIMA)
at 565 nm. All measurements were carried in triplicates. The re-
sults were compared with those of a control reference plate fixed
3. Results and discussion
3.1. Synthesis of ortho-, meta- and para-hydroxyphenyl-
tyrosinamide–chlorambucil hybrid molecules (5 and 6)
As shown in Scheme 1, the chlorambucil entity was directly
linked to ortho, meta- and para-hydroxyphenyl-
Commercially available N-Boc- -tyrosine (7) was coupled with
hydroxyaniline (ortho-, meta- and para-) upon treatment with
HOBt, DCC in DMF to give derivative 8 with a 96% optimal yield.
This compound was readily deprotected under acidic conditions
and coupled with chlorambucil using standard reaction conditions
(HOBt, DCC) to give derivative 5 with a 44% optimal yield.
L-para-tyrosine.
L
Chlorambucil was also linked to the para-hydroxyphenyl-L-
para-tyrosinamide by means of a 5 or a 10 carbon atoms spacer.
These molecules were designed and synthesized according to
two different methodologies; the linear synthesis (Scheme 1) and
the convergent synthesis (Scheme 2). Initially, following Scheme
OH
OH
O
O
O
(b) (c)
N
H
(a)
OH
NHtBoc
7
N
H
HN
NHtBoc
8
HO
HO
HO
3
O
Cl
N
5
(b)
Cl
t-Boc
(d)
O
HO2C
N
m
HO2C
NH2
m
H
9, m = 5 or 10
10
OH
OH
O
m = 5 or m = 10
(e)
N
H
N
H
NH3+CF3CO2
-
HN
HO
HO
m
NHBoc
O
11 (with para)
(b)
12
OH
OH
O
O
O
Cl
Cl
N
(c)
N
H
N
H
HN
HN
NH3+CF3CO2
-
HO
HO
N
m
m
3
H
O
13
O
6
Scheme 1. Synthesis of the ortho-, meta- and para-hydroxyphenyl-
L-para-tyrosinamide–chlorambucil molecules by the linear path. Reagents: (a) hydroxyaniline, HOBt, DCC,
DMF (92–96%); (b) TFA, CH2Cl2, 10 min (100%); (c) CLL, HOBt, DCC, triethylamine, DMF (37–44% for 5; 17–19% for 6); (d) 2-(t-butoxycarbonyloxyimino)-2-phenylacetonitrile
(BOC-ON), 1,4-dioxane, water, triethylamine, 22 °C (63–97%); (e) BocNH(CH2)mCO2H (10), HOBt, DCC, triethylamine, DMF (45–46%).

410
C. Descôteaux et al. / Steroids 77 (2012) 403–412
Cl
N
O
Cl
m = 5 or m = 10
(a)
HO2C
m NH2
HO2C
N
H
m
3
14
9, m = 5 or 10
OH
OH
O
O
O
Cl
(b)
(c)
N
N
N
Cl
O
H
H
NHtBoc
HN
HO
HO
N
m H
3
8
(ortho, meta or para)
6 (ortho, meta or para)
Scheme 2. Synthesis of the hydroxyphenyl-
L-para-tyrosine–chlorambucil molecules by the convergent path. Reagents: (a) 1. HMDS, H2SO4 cat., TMSCl, DCM, triethylamine;
2. CLL, HOBt, DCC, triethylamine, DMF (66–98%); (b) TFA, CH2Cl2, 10 min (100%); (c) 14, HOBt, DCC, triethylamine, DMF (25–51%).
para-hydroxyphenyl-L-para-tyrosinamide derivative (8), prepared
Table 1
Inhibitory concentrations (IC50, l
Ma) of chlorambucil and the hybrids (5, 6; m = 5 and
as previously described, was treated with TFA. The resulting triflu-
oroacetic ammonium salt was coupled with 14 using standard
reaction conditions (HOBt, DCC) to give hybrid 6 with a 51% opti-
mal yield.
The optimal overall yield for the synthesis of 6 (para-hydroxy-
phenyl) was 8% via the linear synthesis (Scheme 1) and 24% via
the convergent synthesis (Scheme 2). Moreover, considering N-
10) on both ER+ and ERÀ breast cancer cell lines.
Compound
MCF-7 (ER+)
MDA-MB-231 (ERÀ)
Chlorambucil
ortho-5
meta-5
130.36 2.92
155.11 19.31
31.25 2.29
136.85 6.79
115.07 7.61
48.61 6.28
100.66 4.62
para-5
118.59 3.90
ortho-6, m = 5
meta-6, m = 5
para-6, m = 5
90.24 6.03
57.57 2.04
137.61 6.92
73.99 3.80
53.70 6.26
69.31 4.15
Boc-L-tyrosine (7) as the starting material, the same final com-
pounds were prepared in only two chemical steps with the conver-
gent path as compared to five chemical steps with the linear path.
The convergent synthesis is, as anticipated, more efficient (3 times
more) and more rapid than the linear synthesis. Of note, all the
new compounds were fully characterized by their respective IR,
1H NMR, 13C NMR and mass spectra.
ortho-6, m = 10
meta-6, m = 10
para-6, m = 10
NR
NR
NR
NR
NR
NR
a
Inhibitory concentration (IC50, lM) as obtained by the MTT assay. Experiments
were performed in duplicates and the results represent the mean SEM of three
independent experiments. The cells were incubated for a period of 72 h. NR: Not
reached.
3.2. In vitro cytotoxic activity
The cytotoxicity of the newly synthesized molecules was evalu-
ated on several tumor cell lines using the MTT colorimetric assay
[22]. Chlorambucil, the parent drug, was tested as the control on
all human mammary carcinomas.
Initially, all the regioisomers (5 and 6, ortho-, meta- and para-
hydroxyphenyl, with or without a spacer chain) were tested
against estrogen-receptor positive (ER+, MCF-7) and estrogen-
receptor negative (ERÀ, MDA-MB-231) breast cancer cells. As
shown by the MTT assays, in most cases, the tyrosinamide–chlor-
ambucil hybrid molecules (ortho-, meta- and para-5 and -6,
m = 5) were more cytotoxic than chlorambucil itself (up to 4 times)
(Table 1). For example, the IC50 for compound meta-5 was
1, the linear synthesis was applied. The amino acids, 6-aminohexa-
noic acid 9 (m = 5) and 11-aminoundecanoic acid 9 (m = 10), were
protected using BOC-ON given the corresponding N-Boc-protected
molecules (10). Beside, derivative 8 was deprotected with TFA to
give trifluoroacetic ammonium salt intermediate 11. Next, the
tyrosine based compound (11) was easily coupled with N-Boc-6-
aminohexanoic acid or N-Boc-11-aminoundecanoic acid (10,
m = 5 or 10) to give derivative 12 with a 46% optimal yield. Subse-
quent deprotection of compound 12, with TFA in dichloromethane,
given the trifluoroacetic ammonium salt intermediate 13, followed
by coupling with chlorambucil, using HOBt, DCC and triethylamine
in DMF, gave hybrid 6 with a 19% optimal yield.
31.25 lM as compared to an IC50 of 130.36 lM for chlorambucil.
Alternatively, the final molecules bearing a 5 or 10 carbon
atoms spacer (6) were synthesized using a convergent and more
efficient approach. This method was used for the preparation of
Thus, the tyrosine entity contributes to the higher activity of the
novel molecules, either by concentrating the drug to the estrogen
receptor site or by its own antioxidant effect contributing to the
overall activity of the hybrid.
all the regioisomers, the ortho-, meta- and para-hydroxyphenyl-L-
para-tyrosinamide–chlorambucil derivatives. As shown in Scheme
2, 6-aminohexanoic acid (9, m = 5) and 11-aminoundecanoic acid
(9, m = 10) were initially coupled with chlorambucil to give cyto-
toxic chains (14) with a 98% optimal yield. This transformation
was performed upon treatment of the corresponding amino acid
with 1,1,1,3,3,3-hexamethyldisilazane (HMDS) in the presence tri-
methylchlorosilane (TMSCl) and catalytic sulfuric acid in DCM and
triethylamine to give the sililated amino acid intermediate. The lat-
ter was then added to activated chlorambucil to produce derivative
14 with excellent yields. Then, the N-protected ortho-, meta- or
According to the in vitro biological assays, the newly synthe-
sized molecules were active on both hormone-dependent (MCF-
7) and hormone-independent (MDA-MB-231) breast cancer cell
lines. Even if the ERa content is much greater in MCF-7 cells, no
specificity was observed with the tysoninamide–chlorambucil
compounds. It is possible that the tyrosinamide part of the mole-
cules interact with other estrogen receptors, which might be pres-
ent in both cancer cell lines. However the specificity could be
noticed more clearly in vivo, as it was previously reported for other
studies [23,24]. Further in vivo studies will be performed.

C. Descôteaux et al. / Steroids 77 (2012) 403–412
411
Table 2
Inhibitory concentrations (IC50
,
l
Ma) of chlorambucil and the hybrids (6, m = 5) on both ER+ and ERÀ breast, ovarian and uterine cancer cell lines.
Compound
ZR-75-1 (breast,
ER+)
MDA-MB-468 (breast, MDA-MB-436 (breast, OVCAR-3 (ovarian, A2780 (ovarian,
Ishikawa (uterine, HEC-1A (uterine,
ERÀ)
ERÀ)
ER+)
ERÀ)
ER+)
ERÀ)
Chlorambucil 100.48 7.01
131.83 7.87
53.24 5.14
146.88 10.69
63.87 5.30
66.11 3.43
61.43 2.55
63.17 6.32
30.57 1.62
198.03 814
103.78 2.31
244.98 26.61
116.55 3.93
ortho-6,
65.87 3.21
m = 5
meta-6, m = 5
para-6, m = 5
52.10 4.76
58.51 4.03
35.42 2.73
45.03 0.92
NA
31.79 2.91
58.27 4.05
48.91 4.82
24.10 2.13
118.10 6.91
NR
219.01 7.66
131.37 11.32
69.67 1.47
a
Inhibitory concentration (IC50, lM) as obtained by the MTT assay. Experiments were performed in duplicates and the results represent the mean SEM of three
independent experiments. The cells were incubated for a period of 72 h. NR: Not reached.
Unlike other tyrosine–chlorambucil derivatives, the molecules
bearing a 10 carbon atoms chain (6, m = 10) did not present any
activity on both cancer cell types. This lack of cytotoxicity could
be explained by a decrease in solubility resulting from the longer
alkyl chain. Interestingly, this study shows a superior biological
activity for the meta-hydroxyphenyl derivatives (meta-5, m = 5
and meta-6, m = 5) when compared to their analogs, ortho- and
para-hydroxyphenyl (ortho-5, para-5, and ortho-6, para-6, m = 5).
On both ER+(MCF-7) and ERÀ (MDA-MB-231) cells lines, the meta
derivatives (meta-5, and meta-6, m = 5) were from 1.3 to 5.0 times
more active than the two other regioisomers (ortho-5, para-5 and
ortho-6, para-6, m = 5). For example, on MCF-7 cancer cells lines,
The various L-para-tyrosinamide regioisomers were made using
two different synthetic paths: linear and convergent methodolo-
gies were applied. The convergent synthesis exceeds the linear
synthesis with a global overall yield of 24% compared to 8% for
the latter. In vitro cytotoxicity assays revealed that all the tyrosina-
mide–chlorambucil molecules synthesized (ortho-, meta- para-5
and -6, with m = 5) were generally more active than chlorambucil
itself on breast, ovarian and uterine cancer cells. Hence, the tyrosi-
namide entity contributes to the enhanced cytotoxicity of the new-
ly synthesized molecules. In addition, all the cytotoxic compounds
were active on both hormone-dependent and hormone-indepen-
dent cancer cells tested. Further biological in vivo studies will be
performed in order to elucidate the mechanism of action of these
new cytotoxic molecules.
the meta-5 regioisomer showed an IC50 of 31.25
to 155.11 and 118.59 M for the ortho-5 and para-5 regioisomers,
respectively. A previous study reported that hybrids made from
meta- -tyrosine isomer (see 2 and 3) produced compounds exhib-
lM as compared
l
Moreover, the meta-hydroxyphenyl derivative was more cyto-
toxic than it’s ortho and para analogs on breast and ovarian cancer
cells. These results are in line with an earlier reported which
showed that derivatives made of meta-tyrosine as the starting
material were more active than those made with ortho-tyrosine
L
iting higher cytotoxic activity [19]. Then, according to these new
data, it is possible to confirm that the meta-hydroxyphenyl deriva-
tives (meta-5, and meta-6, m = 5) also provide more active
molecules.
and para-tyrosine. So, the specific combination of L-meta-tyrosine
In order to explore the biological activity of the ortho-, meta-
and para-hydroxyphenyl-tyrosinamide molecules on other female
cancer types, derivatives bearing a 5 carbon atoms spacer (ortho-
6, meta-6 and para-6 with m = 5) were selected and tested for a
more complete study using a panel of female cancer cell lines.
The choice of these molecules was based on their interesting bio-
logical activity on breast cancer cells (Table 1). The MTT colorimet-
ric assays were performed using ZR-75-1 (breast, ER+), MDA-MB-
468 and MDA-MB-436 (breast, ERÀ), OVCAR-3 (ovarian, ER+),
A2780 (ovarian, ERÀ) Ishikawa (uterine, ER+) and HEC-1A (uterine,
ERÀ) cancer cell lines. Once again, in most cases, on all the female
cancer cell lines used, the novel hybrid molecules were more cyto-
toxic than their parent drug, chlorambucil (Table 2). When looking
at the meta-hydroxyphenyl compound (meta-6, m = 5), its en-
hanced activity (when compared to the two others regioisomers;
ortho-6 and para-6, m = 5) was still observed with breast and ovar-
ian cancer cell lines. Derivative meta-6, m = 5 possesses an IC50 of
and meta-hydroxyaniline will be investigated in our laboratory in
an attempt to obtain even more powerful compounds.
Acknowledgments
The authors wish to thank the Fonds de Recherche sur la Nature
et les Technologies du Québec (FQRNT) for key financial support.
The NSERC student fellowship for Caroline Descôteaux is gratefully
acknowledged.
References
[1] Sachdeva MS. Drug targeting systems for cancer chemotherapy. Expert Opin
Invest Drugs 1998;7(11):1849–64.
[2] Dubowchik GM, Walker MA. Receptor-mediated and enzyme-dependent
targeting of cytotoxic anticancer drugs. Pharmacol Ther 1999;83(2):67–123.
[3] Jaracz S, Chen J, Kuznetsova LV, Ojima I. Recent advances in tumor-targeting
anticancer drug conjugates. Bioorg Med Chem 2005;13(17):5043–54.
[4] Keely N, Meegan M. Targeting tumors using estrogen receptor ligand
conjugates. Curr Cancer Drug Targets 2009;9(3):370–80.
31.79 lM as compared to 66.11 lM for chlorambucil on OVCAR-3
ovarian cancer cells lines. This marked cytotoxicity for the meta-
hydroxyphenyl molecules (meta-6, m = 5) was more obvious with
the ER+ breast (MCF-7) and ER+ ovarian (OVCAR-3) cancer cell
[5] Kuiper GGJM, Carlsson BO, Grandien KAJ, Enmark GGJM, Häggblad J, Nilsson S,
et al. Comparison of the ligand binding specificity and transcript tissue
distribution of estrogen receptors {alpha} and {beta}. Endocrinology
1997;138(3):863–70.
[6] Belcher SM. Rapid signaling mechanisms of estrogens in the developing
cerebellum. Brain Res Rev 2008;57(2):481–92.
lines, which express the ERa protein.
[7] Shanle EK, Xu W. Selectively targeting estrogen receptors for cancer treatment.
Adv Drug Deliv Rev 2010;62(13):1265–76.
[8] Bai Z, Gust R. Breast cancer, estrogen receptor and ligands. Arch Pharm
2009;342(3):133–49.
4. Conclusion
[9] Descôteaux C, Leblanc V, Bélanger G, Parent S, Asselin É, Bérubé G. Improved
synthesis of unique estradiol-linked platinum (II) complexes showing potent
cytocidal activity and affinity for the estrogen receptor alpha and beta. Steroids
2008;73(11):1077–89.
[10] Gupta A, Saha P, Descôteaux C, Leblanc V, Asselin É, Bérubé G. Design,
synthesis and biological evaluation of estradiol–chlorambucil hybrids as
anticancer agents. Bioorg Med Chem Lett 2010;20(5):1614–8.
[11] Kuduk SD, Zheng FF, Sepp-Lorenzino L, Rosen N, Danishefsky SJ. Synthesis and
evaluation of geldanamycin-estradiol hybrids. Bioorg Med Chem Lett
1999;9(9):1233–8.
In conclusion, the natural amino acid, L-para-tyrosine, was mod-
ified in order to get a structure similar to the steroid backbone.
Ortho-, meta- and para-hydroxyaniline were added to the tyrosine
entity via an amide bonding. Chlorambucil was linked directly or
via a carbon atom spacer (5 carbon or 10 carbon atoms) to create
cytotoxic molecules. The design, the synthesis and the biological
evaluation were presented in this manuscript.

412
C. Descôteaux et al. / Steroids 77 (2012) 403–412
[12] Devraj R, Barrett JF, Fernandez JA, Katzenellenbogen JA, Cushman M. Design,
synthesis, and biological evaluation of ellipticine–estradiol conjugates. J Med
Chem 1996;39(17):3367–74.
[18] Anstead GM, Carlson KE, Katzenellenbogen JA. The estradiol pharmacophore:
ligand structure-estrogen receptor binding affinity relationships and a model
for the receptor binding site. Steroids 1997;62(3):268–303.
[13] Gagnon V, St-Germain ME, Descôteaux C, Provencher-Mandeville J, Parent S,
Mandal SK, et al. Biological evaluation of novel estrogen-platinum (II) hybrid
molecules on uterine and ovarian cancers–molecular modeling studies. Bioorg
Med Chem Lett 2004;14(23):5919–24.
[14] Van Themsche C, Parent S, Leblanc V, Descôteaux C, Simard AM, Bérubé G, et al.
VP-128, a novel oestradiol-platinum(II) hybrid with selective anti-tumour
activity towards hormone-dependent breast cancer cells in vivo. Endocr Relat
Cancer 2009;16(4):1185–95.
[15] Sun J, Huang YR, Harrington WR, Sheng S, Katzenellenbogen JO,
Katzenellenbogen BS. Antagonists selective for estrogen receptor {alpha}.
Endocrinology 2002;143(3):941–7.
[16] Sampson NS, Knowles JR. Segmental motion in catalysis: investigation of a
hydrogen bond critical for loop closure in the reaction of triosephosphate
isomerase. Biochemistry 1992;31(36):8488–94.
[19] Descôteaux C, Brasseur K, Leblanc V, Parent S, Asselin É, Bérubé G. SAR study of
tyrosine chlorambucil hybrid regioisomers; synthesis and biological
evaluation against breast cancer cell lines. Amino Acids 2011;20(24):7388–92.
[20] Perrin D, Armarego W. Purification of Laboratory Chemicals. third
ed. Oxford: Pergamon Press; 1988.
[21] Still W, Kahn M, Mitra A. Rapid chromatographic techniques for preparative
separation with moderate resolution. J Org Chem 1978;43(14):2923–5.
[22] Carmichael J, DeGraff WG, Gazdar AF, Minna JD, Mitchell JB. Evaluation of a
tetrazolium-based semiautomated colorimetric assay: assessment of
chemosensitivity testing. Cancer Res 1987;47(4):936–42.
[23] Karl J, Gust R, Spruss T, Schneider MR, Schoenenberger H, Engel J, et al. Ring-
substituted [1,2-bis (4-hydroxyphenyl) ethylenediamine] dichloroplatinum
(II) complexes: compounds with a selective effect on the hormone-dependent
mammary carcinoma. J Med Chem 1988;31(1):72–83.
[17] Muthyala RS, Carlson KE, Katzenellenbogen JA. Exploration of the bicyclo
[3.3.1] nonane system as a template for the development of new ligands for
the estrogen receptor. Bioorg Med Chem Lett 2003;13(24):4485–8.
[24] Otto AM, Faderl M, Schönenberger H. Dissociation of estrogenic and cytotoxic
properties of an estrogen receptor-binding platinum complex in human breast
cancer cell lines. Cancer Res 1991;51(12):3217–23.