Synthesis, oxidation chemistry and cytotoxicity studies on ferrocene derivatives of diethylstilbestrol

Article abstract of DOI:10.1039/b913570g

A series of compounds is described in which one of the ethyl groups in diethylstilbestrol has been replaced by a ferrocenyl substituent. Only those derivatives incorporating phenol moieties underwent isomerisation from the Z to the E form, and some of the

Full text of DOI:10.1039/b913570g

View Online / Journal Homepage / Table of Contents for this issue
This paper is published as part of a Dalton Transactions themed issue on:
Metal Anticancer Compounds
Guest Editor Peter Sadler
University of Warwick, UK
Published in issue 48, 2009 of Dalton Transactions
Image reproduced with permission of Chi-Ming Che
Articles published in this issue include:
PERSPECTIVES:
Non-traditional platinum compounds for improved accumulation, oral bioavailability, and tumor
targeting
Katherine S. Lovejoy and Stephen J. Lippard, Dalton Trans., 2009, DOI: 10.1039/b913896j
Metal complexes as photochemical nitric oxide precursors: Potential applications in the treatment
of tumors
Alexis D. Ostrowski and Peter C. Ford, Dalton Trans., 2009, DOI: 10.1039/b912898k
Novel and emerging approaches for the delivery of metallo-drugs
Carlos Sanchez-Cano and Michael J. Hannon, Dalton Trans., 2009, DOI: 10.1039/b912708a
HOT ARTICLE:
Iron(III) complexes of fluorescent hydroxamate ligands: preparation, properties, and cellular
processing
Antonia J. Clarke, Natsuho Yamamoto, Paul Jensen and Trevor W. Hambley, Dalton Trans., 2009,
DOI: 10.1039/b914368h
Visit the Dalton Transactions website for more cutting-edge inorganic and bioinorganic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Synthesis, oxidation chemistry and cytotoxicity studies on ferrocene
derivatives of diethylstilbestrol†
Yong Leng Kelvin Tan,a Pascal Pigeon,a Elizabeth A. Hillard,*a Siden Top,a Marie-Aude Plamont,a
Anne Vessie`res,a Michael J. McGlinchey,b Helge Mu¨ller-Bunzb and Ge´rard Jaouen*a
Received 8th July 2009, Accepted 1st October 2009
First published as an Advance Article on the web 27th October 2009
DOI: 10.1039/b913570g
A series of compounds is described in which one of the ethyl groups in diethylstilbestrol has been
replaced by a ferrocenyl substituent. Only those derivatives incorporating phenol moieties underwent
isomerisation from the Z to the E form, and some of them could be chemically oxidized to a quinone
species. The compounds were less cytotoxic against hormone-independent MDA-MB-231 breast cancer
cell lines than their corresponding ferrocenyl phenyl or phenol isomers in which the ferrocene and ethyl
moieties are linked to the same carbon atom. The biochemical results were evaluated in conjunction
with information obtained from electrochemical and chemical oxidation experiments.
the synthesis and biological activity of the hydroxyferrocifens5,6
and some ferrocenyl phenols,7–10 as well as metal complexes of
1. Introduction
meso-hexestrol,11 the dihydro derivative of diethylstilbestrol. The
rationale behind the hydroxyferrocifens, designed by replacing the
b phenyl group of hydroxytamoxifen with ferrocene, was to merge
the cytotoxicity associated with the ferrocenyl (or ferricenium)
moiety with the anti-oestrogenicity of the hydroxytamoxifen
skeleton on a single molecule. This idea was further translated into
the creation of ferrocenyl phenols by removing the side chain of
hydroxytamoxifen, or substituting it with a second hydroxyl group.
These molecules display cytotoxic activity towards hormone-
dependent (MCF-7) and hormone-independent (MDA-MB-231)
breast cancer cell lines at much lower concentrations than does
hydroxytamoxifen, although the effects of the compounds lacking
the side chain on MCF-7 cell lines represent a combination of
oestrogenicity and cytotoxicity. At present we are investigating
the mechanism of action for such molecules, and have proposed
an activation pathway involving the in situ oxidation of the phenol
moiety via the ferrocene group to generate an active quinone
methide (QM) species.12,13
The selective oestrogen receptor modulator (SERM) tamoxifen, 2-
[4-(1,2-diphenyl-1-butenyl)phenoxy]-N,N-dimethylethanamine, is
the most common drug used to treat patients diagnosed with breast
cancer in which the oestrogen receptor (ER) is present (ER+). Its
active metabolite is hydroxytamoxifen (Chart 1).1–4 Unfortunately,
some breast cancer cells remain resistant to tamoxifen either
because they do not express ER (ER–) or have developed resistance
as a result of long-term exposure to the drug.
Our work has focused on the creation of organometallic com-
plexes that display not only the anti-oestrogenic effect of tamox-
ifen, but also oestrogen receptor-independent cytotoxicity to treat
tamoxifen-resistant breast cancer. We have previously reported
aLaboratoire Charles Friedel, UMR CNRS 7223, Ecole Nationale
Supe´rieure de Chimie de Paris, 11 rue Pierre et Marie Curie, 75231, Paris
Cedex 05, France. E-mail: gerard-jaouen@chimie-paristech.fr, elizabeth-
hillard@chimie-paristech.fr; Fax: +33 1 43 25 79 75; Tel: +33 1 44 27 66
98
bUCD School of Chemistry and Chemical Biology, University College Dublin,
Belfield, Dublin 4, Ireland
In the course of our systematic study on the role of ferrocene
as well as the structure-activity relationships within the ferrocenyl
phenol family, it was of interest to examine whether changing the
position of the ferrocene moiety relative to the ethyl group would
affect the ability of such compounds to generate QM species, and
hence influence the cytotoxicity of these molecules. In this article,
† Electronic supplementary information (ESI) available: Proposed solution
state structures and tentative 1H NMR assignments for 4 and 5c; 1H
NMR spectrum for 4, 5c, and 5h; COSY spectrum for 4 and 5c; NOESY
spectrum for 4 and 5c. CCDC reference numbers 738753–738757. For
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/b913570g
Chart 1
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 10871–10881 | 10871
©

we present the synthesis, electrochemistry, oxidation studies and
biochemical evaluation for a series of ferrocene derivatives based
on diethylstilbestrol, 2, in which the ferrocene and ethyl groups
reside on each of the olefinic carbon atoms, and compare these
results to those of the ferrocenyl phenyl and phenol compounds
1, in which the ferrocene and ethyl moieties are linked to the same
carbon atom.
verified by the A2B2 proton resonances of the –C6H4OH ring for
Z-2b (d 6.79 and 7.06 ppm, respectively) which are deshielded
relative to those of its E-isomer (d 6.44 and 6.86 ppm, respectively).
A similar observation can be made for 2e. Furthermore, the
shielding effect appears to extend to the ethyl group as well. It
is noted that the proton signals attributable to the ethyl group of
the Z-isomers appear at lower frequency of those observed for the
E-isomers. These assignments were also confirmed by comparing
the X-ray crystal structures of Z-2a, Z-2g, Z-2h, and Z-2i with
the NMR spectra of the obtained crystals.
2. Results and discussion
The McMurry coupling reaction between 4-hydroxybenzoyl
ferrocene and 4-hydroxypropiophenone did not afford the ex-
pected product 2c, resulting instead in the formation of 3.
Compound 3 has been completely characterized, with the 13C
NMR spectrum exhibiting a signal at d 209.1 ppm indicating
the presence of a ketone functional group. A single-crystal
X-ray crystallographic study confirmed the molecular structure of
3, and the ORTEP plot together with pertinent bond parameters
is given in Fig. 1. Compound 3 appears to be the outcome of
a pinacol-pinacolone type rearrangement. Such products derived
from McMurry coupling reactions involving ferrocenyl moieties as
well as cobalt sandwich complexes have been previously observed,
and compounds of a similar nature have also been detected in
Clemmensen reduction reactions.16–18
2.1 Synthesis and characterization
A Friedel–Crafts acylation of ferrocene to obtain benzoyl fer-
rocene, followed by a McMurry cross-coupling of benzoyl fer-
rocene with the appropriate propiophenone, afforded the dark
orange compounds 2a, 2b and 2d–i in moderate to high yields (41
to 71%), as depicted in Scheme 1. It is well known that products of
the McMurry reaction are usually obtained as a mixture of Z and
E-isomers. In our case we were unable to separate the two isomers
of each compound by chromatographic methods, and the presence
of both isomers was observed in the solution state 1H NMR spectra
for 2. It has been reported that for tamoxifen and ferrocifen the
a¢ aromatic ring located between two other rings is subjected to
a shielding effect.5,14,15 Accordingly, the signals ascribable to the
cyclopentadienyl protons of the ferrocenyl units for the Z-isomers
of 2 should resonate at lower frequency in relation to those of
the corresponding E-isomers. Consequently, we have assigned the
major isomers of 2 in solution to be the Z-isomers. This is further
Fig. 1 ORTEP diagram (50% probability thermal ellipsoids) and
◦
˚
selected bond lengths (A) and angles ( ) for 3. Cp(centroid)–Fe =
1.652(1); Cp(centroid)–Fe 1.647(1); C10–C11 1.530(2);
C11–C12 1.5318(19); C11–C18 1.5399(18); C11–C24
=
=
=
=
=
1.553(2); Cp(centroid)–Fe–Cp(centroid) = 176.09(1); C10–C11–C18 =
110.00(11); C10–C11–C24 = 112.50(12); C12–C11–C24 = 109.97(11);
C18–C11–C24 = 105.15(11).
We have previously reported the synthesis for the Z-isomer
of 2c via a McMurry coupling between 4-tert-butyldimethyl-
siloxyphenyl ethyl ketone and 4-methoxyphenyl ferrocenyl ketone,
in which the coupled product was obtained almost exclusively
as the Z-isomer. This was followed by simultaneous elimination
of the two protecting groups, 4-tert-butyldimethylsiloxy and
methoxy, to afford the Z-isomer of 2c.19 In light of the present
difficulty in obtaining 2c via a direct McMurry coupling between
Scheme 1
10872 | Dalton Trans., 2009, 10871–10881
This journal is
The Royal Society of Chemistry 2009
©

Table 1 Common atomic numbering scheme and selected bond parame-
ters for Z-2a, Z-2g, Z-2h·0.5H2O and Z-2i
Table 2 NMR study on the isomerisation of 2a, 2c, 2d and 2f
Time
After
Compound (%) Solvent
10 min After 1 d After 3 d After 1 week
Z-2a : E-2a
Z-2c : E-2c
Z-2d : E-2d
Z-2f : E-2f
CDCl3
100 : 0 100 : 0
100 : 0
100 : 0
—
94 : 6
89 : 11
89 : 11
100 : 0
100 : 0
100 : 0
100 : 0
—
91 : 9
88 : 12
88 : 12
100 : 0
100 : 0
(CD3)2SO 100 : 0 100 : 0
a
CDCl3
(CD3)2SO 100 : 0 99 : 1
CDCl3 100 : 0 90 : 10
(CD3)2SO 100 : 0 90 : 10
CDCl3 100 : 0 100 : 0
(CD3)2SO 100 : 0 100 : 0
—
—
Z-2a
Z-2g
Z-2h·0.5H2Oa
Z-2i
˚
a 2c was found to be sparingly soluble in CDCl3.
Bond lengths/A
C1–C2
C1–C11
C1–C17
C2–C3
C2–C5
1.3438(19) 1.3660(15) 1.354(5); 1.367(5)
1.4983(19) 1.5082(14) 1.513(5); 1.510(4)
1.4835(19) 1.4947(14) 1.488(4); 1.493(5)
1.5114(19) 1.5266(15) 1.509(5); 1.520(5)
1.352(3)
1.512(3)
1.493(3)
1.530(3)
1.510(3)
atom of the double bond. While the planarity of the ethylenic
skeleton is largely preserved, the two phenyl moieties in each
molecule are tilted with respect to the olefinic plane, ranging from
66.2(2)◦ in Z-2g to almost perpendicular [89.1(2)◦] in Z-2i. It is
also noted that although the cyclopentadienyl ring attached to the
double bond is almost co-planar with the latter plane for Z-2a
[1.8(2)◦] and Z-2g [3.2(2)◦], a larger twist angle for the said ring is
observed for Z-2h and Z-2i [16.8(3)◦ and 34.6(2)◦, respectively].
1.498(2)
1.5023(15) 1.505(5); 1.501(5)
Bond angles/◦
C11–C1–C17
C3–C2–C5
C5–C2–C1–C17 4.3(2)
C3–C2–C1–C11 1.7(2)
114.58(12) 114.68(9) 115.8(3); 115.6(3)
114.15(12) 114.69(9) 113.4(3); 112.8(3)
115.12(16)
114.01(16)
2.6(2)
4.4(2)
8.3(2)
1.9(5); -0.1(5)
3.9(5); 1.2(5)
0.6(2)
C5–C2–C1–C11 -117.63(13) -172.9(2) -175.6(3); -177.9(3) -171.8(2)
2.2 Isomerisation studies
a There are two molecules in the asymmetric unit; bond lengths and angles
are listed for both.
The interconversion between Z- and E-isomers is commonly found
in the stilbene series. It has been found that generally the Z-
isomer displays a greater affinity for the specific oestrogen receptor
than the E-isomer, and hence is expected to be more biologically
active. The isomerisation of DES and hydroxytamoxifen have been
studied, and a proposed mechanism involves the interaction and
transfer of a phenolic proton onto the central double bond.20,21
We obtained the pure Z-isomers of 2a, 2c, 2d and 2f by
fractional crystallization, and have studied their isomerisation
the two constituent ketones, we utilized a similar strategy to obtain
a mixture of both the Z- and E-isomers of 2c, by eliminating the
protecting methoxy group from 2i with boron tribromide, in 68%
yield.
Compounds 2a–i have been characterized spectroscopically and
analytically, and for Z-2a, Z-2g, Z-2h and Z-2i, by single-crystal
X-ray crystallographic studies as well. The ORTEP diagram of
Z-2g, which is representative, is illustrated in Fig. 2. A common
atomic numbering scheme, together with selected bond parameters
for all four compounds are collected in Table 1. For each molecule,
the ferrocenyl moiety is oriented towards the phenyl group linked
to the same carbon atom of the olefinic bond, presumably to
minimize steric interaction with its cis-disposed aryl neighbour.
Similarly, there is a narrowing of the bond angle between the ethyl
substituent and the aromatic ring connected to the same carbon
1
by H NMR spectroscopy, utilizing deuterated chloroform and
DMSO as solvents, via the following method: the Z-isomer of
each compound was dissolved in the appropriate solvent in an
NMR tube and left at ambient temperature, and formation of
the E-isomer was monitored. The percentage of E-isomer formed
was determined by measuring the intensity of the triplet signal
corresponding to the methyl protons of the ethyl group for each
isomer. Table 2 summarises the results obtained.
After one week, only 2c and 2d were found to have undergone
isomerisation while for 2a and 2f no conversion of the Z-isomer
was observed in either solvent. It was previously reported that the
isomerisation of DES and hydroxytamoxifen could be catalyzed by
acids.22 We therefore added one molar equivalent of phenol to the
CDCl3 solutions of 2a and 2f. Nevertheless, no isomerisation could
be found in either case even after monitoring for another week. We
have previously studied the isomerisation for Z-hydroxyferrocifen,
and found that the rate of isomerisation could be correlated to
the acidity of the solvent; conversion to the E-isomer was found
to occur rapidly in CDCl3 while no isomerisation took place
in deuterated DMSO after one day.6 In contrast, conversion to
the E-isomer for Z-2c and Z-2d appear to be independent of
the two solvents used. These results suggest that the existence
of a phenol group on the molecule is an important criterion
for conversion between isomers for this series of compounds,
Fig. 2 ORTEP diagram (50% probability thermal ellipsoids, organic
hydrogen atoms omitted) for Z-2g.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 10871–10881 | 10873
©

Table 3 Standard oxidation potentials for 2a–i in methanol (n = 0.1 V s-1)
Compound
E◦, (Fc/Fc+) [V vs. Ag/AgCl]
2a
2b
2c
2d
2e
2f
2g
2h
2i
+0.435
+0.414
+0.397
+0.416
+0.421
+0.421
+0.399
+0.406
+0.407
although the presence of additional phenol groups, or the acidity
of the solvent utilized, does not necessarily translate into a faster
isomerisation process.
2.3 Electrochemistry studies
Cyclic voltammograms were obtained for compounds 2a–i, and
the standard potentials (E◦ vs. Ag/AgCl) are shown in Table 3.
In methanol and methanol-pyridine (50 : 1 v/v) solutions,
all the compounds except for 2c show conventional, reversible
ferrocene/ferricenium couples with redox potentials ranging from
+0.399 to +0.435 V. For these compounds, the influence of the
substituent on the ferrocene redox potential is consistent with the
reported Hammett constants of the substituents.23 For example,
the unsubstituted compound 2a has the highest redox couple and
2g, with two methoxy groups, has the lowest redox potential.
Interestingly, 2c shows the presence of two waves (Fig. 3).
At low scan rates the first wave is largely irreversible, but
gains reversibility at higher scan rates, indicative of an EC
(electrochemical–chemical) mechanism. At low scan rates, we
observed that the first peak represents the abstraction of more
than one electron. In analogy with our previous proposition for
the activation of ferrocenyl phenols; we suggest the following
mechanism (Scheme 2).12 The intensity of the first wave is inversely
proportional to the scan rate, and thus likely corresponds to
the two-electron process of diquinone formation. The second
wave therefore probably represents the reversible oxidation of the
ferrocene on the quinone molecule, accounting for the positive
shift in potential.
Fig. 3 Cyclic voltammogram of 2c (1 mM in 0.1 M Bu4NBF4/MeOH)
in the absence (solid line) and presence (dashed line) of pyridine in 1 : 50
volume ratio. Scan rate 0.05 V s-1. Pt electrode of 0.5 mm diameter.
Scheme 3
of two hydrogen atoms from 2c (m/z 424). Assignment of the
spectral features for 4 and 5c was aided by 2D COSY and NOESY
experiments.
For the 13C NMR spectrum of 4, the resonances at d 187.0
=
and 186.9 ppm indicate the presence of two C O groups. In the
proton NMR spectrum, there are eight distinct doublet of doublet
signals in the d 6–9 ppm range, integrating to one proton each;
these are assigned to the protons of the two quinone rings. It
was also noted that the two CH2 protons of the ethyl group are
non-equivalent, appearing as two independent doublet of quartet
resonances. Further evidence for the non-equivalent nature of the
latter protons was also found in the NOESY spectrum, where only
one of the two CH2 protons displays a crosspeak with the nearest
proton on the quinone ring. For 5c, only one peak is present in the
carbonyl region of the 13C NMR spectrum. A singlet resonance
2.4 Chemical oxidation reactions
Treatment of 2d, 2e, 2f and 2i with silver oxide in acetone afforded
only unreacted starting compounds. On the other hand, chemical
oxidation of 2c resulted in the initial formation of the diquinone
4, which underwent further isomerisation to the phenol–quinone
5c (Scheme 3); the latter two compounds are green in colour.
The molecular ion peaks of 4 and 5c (m/z 422) indicate the loss
Scheme 2
10874 | Dalton Trans., 2009, 10871–10881
This journal is
The Royal Society of Chemistry 2009
©

at d 8.44 ppm in the proton NMR spectrum may be ascribable
to the –OH proton of the phenol ring. Two doublet signals at d
7.14 and 6.76 ppm, integrating to two hydrogens each, have been
assigned to the protons on the phenol moiety. The three methyl
hydrogens on the end of the carbon chain appear as a doublet
resonance, indicating that they are equivalent and coupled to only
one neighbouring proton; this is in accord with the appearance of
a quartet signal at d 6.23 ppm, which corresponds to the single
vinylic proton.
Intrigued by our results on the oxidation of 2c, we also decided
to subject the analogous compounds 2b and 2h (those capable of
forming quinone methide structures) to chemical oxidation. For
2h, this led to a colour change from orange to dark green, indicative
of the formation of the quinone 5h. It was surprising, therefore,
to discover that 2b remained inert to chemical oxidation. Even
after six hours of stirring with silver oxide only starting materials,
together with traces of decomposition products, were detected in
the 1H NMR spectrum; the reasons for this currently are unclear
(Scheme 4). A second attempt to obtain 5b, this time using NOBF4,
was also unsuccessful.
of doublet signals integrating to one proton each, attributable to
the protons of the quinone ring, are displayed in the H NMR
1
spectrum. The two doublet resonances at d 7.23 and 6.85 ppm
integrating to two hydrogens each, as well as a singlet peak at d
3.75 ppm, have been assigned to the protons of the methoxyphenyl
ring and the methoxy group, respectively. In addition, the three
methyl hydrogens on the end of the carbon chain for 5h appears
as a doublet signal at d 1.79 ppm, coupled to the single vinylic
proton which appears as a quartet resonance at d 6.28 ppm.
We have also studied 2c and 5c by UV-vis spectrometry (Fig. 4).
For the parent compound 2c, the bands at 278 and at 448 nm
are typical of Fe (e2g) → Cp (e1g) charge transfer and symmetry-
forbidden Fe (a1g) → Fe (e1g) transitions, respectively.24–28 On the
other hand, the spectrum of 5c displayed an intense shoulder at
242 nm, which may be a quinone (n, p → p*) transition.29 In
addition, a very strong band at 392 nm and a strong broad band
at 590 nm are also present. The band at lower energy was observed
to be solvatochromic, and may be ascribed to an intramolecular
charge transfer transition between the ferrocenyl donor and
quinonyl acceptor centres [ep(HOMO-Fc) → ep*(LUMO-q)].30,31
Fig. 4 UV-vis spectra of 2c (solid line) and 5c (dashed line) in acetonitrile.
Scheme 4
The oxidation of 2c to 4 and subsequent isomerisation of
the latter to 5c is reminiscent of previous observations for
diethylstilbestrol (DES) and its fluorinated derivative.32,33 The
oxidation of DES resulted in the formation of the diquinone
DES Q, of which a crystal structure was obtained (Scheme 5).34
A carbonyl stretch at 1712 cm-1 is present in the infrared
spectrum of 5h. The CI mass spectrum shows the highest mass
ion at m/z 437 (MH+), corresponding to the lost of two hydrogen
atoms from the parent compound 2h (m/z 438). Four doublet
Scheme 5
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 10871–10881 | 10875
©

Table 4 Effect of compounds 1 and 2 on growth of hormone-independent
Remarkably, it was found that while solid DES Q underwent
isomerisation readily at room temperature to Z, Z-dienestrol
(DIES), presumably via a phenol-quinone intermediate, a solution
of DES Q in chloroform remained unchanged for three days under
ambient conditions. The facile rearrangement of crystalline DES
Q to DIES was rationalized by the intermolecular transfer of
one of the methylene hydrogens to a carbonyl oxygen atom on a
neighbouring DES Q molecule, as both the said atoms were found
to be in close contact in the crystal structure. Bond migration
would result in the formation of the phenol-quinone intermediate,
and a corresponding hydrogen transfer to the other quinonoid
group would give DIES. The workers were unable to demonstrate
the formation of the phenol-quinone intermediate in solution,
and suggested that the stability of DES Q in solution could be
explained by the difficult occurrence of intermolecular hydrogen
transfer, given that the DES Q molecules are dispersed in solution.
In contrast, our NMR monitoring experiments indicated that
a CDCl3 solution of the diquinone 4 isomerised overnight to the
phenol–quinone 5c, which subsequently decomposed gradually
in solution over one week. Solid 5c was found to be stable over
a few weeks. The isolation and relative stability of 5c may be
explained by the replacement of one the ethyl groups in DES by a
ferrocene group in 2c, which prevents a second hydrogen transfer
and ensuing bond migration from taking place to give rise to a
diene analogous to DIES.
breast cancer cells
Chemically
oxidized to QM 10 mM on MDA-MB-231
Effect on growth of cellsa
Compound R
1a
R¢
H
H
H
OH
H
H
OH
OH
H
No
Yesb
Yesb
No
No
Yes
No
No
37
11
6
(Z + E)-1b OH
1c
OH
H
Z-2a
Z-2b
Z-2c
Z-2d
Z-2e
Z-2f
Z-2g
Z-2h
Z-2i
53
40
16
49
53
93
90
9
OH
OH
H
OCH3
H
OCH3 No
OCH3 OCH3 No
OH OCH3 Yes
OCH3 OH No
73
a Control = cells without added compounds after 5 d of culture, set at
100%. b Unpublished results.
2.5 Effect on growth of hormone-independent breast cancer cells
MDA-MB-231
It is also surprising that 1a and 2a, neither of which contains
any phenol groups, display anti-proliferative effects on the cells.
We hypothesize that these compounds may be metabolized to
their phenol analogues, viz. 1b and 2b, respectively, in the cell.
It is likely that this process is somewhat similar to the aromatic
hydroxylation of tamoxifen to generate hydroxytamoxifen, as
cytochrome P450 enzymes have been detected in MDA-MB-231
breast cancer cells.36–39
The effect of 2a–i at a concentration of 10 mM was studied on
hormone-independent (MDA-MB-231) breast cancer cells and the
results are reported in Table 4; the values for 1a–c have also been
included for comparison. On these cell lines, the observed anti-
proliferative effect can be attributed only to a cytotoxic effect
potentially induced by the ferrocenyl unit.
The most cytotoxic complexes amongst 2a–i were found to be 2h
followed by 2c. These two compounds were able to be chemically
oxidized to quinone species, while the other complexes in the series,
which remained inert to chemical oxidation, displayed markedly
lower anti-proliferative effects on the cells. This observation is
in line with our hypothesis that oxidative activation to quinone
methide species could be the key to the biological activity of these
types of compounds. In addition, the methoxy compounds 2e, 2f
and 2g were found to be less efficacious than their corresponding
phenol analogues 2b, 2d and 2c, suggesting that intracellular
hydrolysis to the free phenol is not favoured.35
The preliminary results indicate that 1a–c have stronger biolog-
ical activity than their corresponding isomers 2a–c, implying that
the position of the phenol group relative to the ferrocene moiety
has an influence on the cytotoxicity of these types of compounds.
These findings are consistent with our proposition that ferrocene
can act as an intramolecular oxidation catalyst for phenol, with
the conjugated p-system linking the two functionalities playing
the role of an electron tunnel, to facilitate the production of a
cytotoxic quinone methide species. However, this is not the case
for compound 2h, where the ferrocene and hydroxyl groups are not
conjugated to each other, thus preventing the role of ferrocene as
an intramolecular oxidation catalyst. Nonetheless, this compound
is quite toxic and must undergo further mechanistic study.
3. Experimental
3.1 General procedures
All reactions and manipulations were carried out under an argon
atmosphere using standard Schlenk techniques. THF was distilled
over sodium/benzophenone prior to use. Column chromatogra-
phy was performed on silica gel 60 M. The semi-preparative HPLC
separations were performed on a Shimadzu instrument with a
Nucleodur C18 column (length of 25 cm, diameter of 3.2 cm,
particle size of 10 mm) using acetonitrile as eluent. IR spectra
1
were obtained on a BOMEM Michelson-100 spectrometer. H
and 13C NMR spectra were acquired on a Bruker 300 MHz
spectrometer, while 2D NMR spectra were obtained on a Bruker
400 MHz machine. The solvent used was deuterated chloroform
unless otherwise stated. Mass spectrometry was carried out at
the Service de Spectrome´trie de Masse at ENSCP, Paris. High
resolution mass spectra (HRMS) were obtained by the Groupe
de Spectroscopie de Masse of the Laboratoire de Structure et
Fonction de Mole´cules Bioactives at the University of Pierre et
Marie Curie, Paris. Microanalyses were performed by the Service
de Microanalyse ICSN at Gif sur Yvette, France.
10876 | Dalton Trans., 2009, 10871–10881
This journal is
The Royal Society of Chemistry 2009
©

Table 5 Amount of reagents used, products and yields
Reaction
Zinc
TiCl4
Benzoyl ferrocene
Propiophenone
Product
2a
Yield
1
2
1.96 g (30 mmol)
1.96 g (30 mmol)
3.79 g, 2.2 mL (20 mmol)
3.79 g, 2.2 mL (20 mmol)
Benzoyl ferrocene
1.45 g (5 mmol)
4-Hydroxybenzoyl
ferrocene 1.53 g
(5 mmol)
Propiophenone 0.67 g, 0.66 mL
(5 mmol)
Propiophenone 0.67 g, 0.66 mL
(5 mmol)
1.39 g (71%)
0.98 g (48%)
2b
3
4
1.96 g (30 mmol)
1.96 g (30 mmol)
3.79 g, 2.2 mL (20 mmol)
3.79 g, 2.2 mL (20 mmol)
Benzoyl ferrocene
1.45 g (5 mmol)
4-Methoxybenzoyl
ferrocene 1.60 g
(5 mmol)
4-Hydroxypropiophenone
0.75 g (5 mmol)
Propiophenone 0.67 g, 0.66 mL
(5 mmol)
2d
2e
0.91 g (45%)
1.17 g (56%)
5
6
1.96 g (30 mmol)
2.40 g (37 mmol)
3.79 g, 2.2 mL (20 mmol)
3.73 g, 2.2 mL (20 mmol)
Benzoyl ferrocene
1.45 g (5 mmol)
4-Methoxybenzoyl
ferrocene 1.92 g
(6 mmol)
4-Methoxypropiophenone
0.82 g, 0.88 mL (5 mmol)
4-Methoxypropiophenone
0.99 g, 1.06 mL (6 mmol)
2f
1.04 g (49%)
1.09 g (41%)
2g
7
8
9
1.60 g (25 mmol)
11.37 g (174 mmol)
1.96 g (30 mmol)
2.49 g, 1.5 mL (13 mmol)
17.66 g, 10.2 mL (93 mmol)
3.79 g, 2.2 mL (20 mmol)
4-Hydroxybenzoyl
ferrocene 1.22 g
(4 mmol)
4-Methoxybenzoyl
ferrocene 9.10 g
(28 mmol)
4-Hydroxybenzoyl
ferrocene 1.53 g
(5 mmol)
4-Methoxypropiophenone
0.66 g, 0.70 mL (4 mmol)
2h
2i
1.01 g (58%)
5.59 g (45%)
1.04 g (47%)
4-Hydroxypropiophenone
4.27 g (28 mmol)
4-Hydroxypropiophenone
0.75 g (5 mmol)
3a
a Melting point: 246 ◦C.
Benzoyl ferrocene, 4-hydroxybenzoyl ferrocene and 4-
methoxybenzoyl ferrocene were synthesized according to pro-
cedures found in the literature.40 All other reagents were from
commercial sources and used as supplied.
the solution was poured into water and extracted with CH2Cl2. The
combined organic layers were dried on MgSO4, filtered and the
solvent was removed under reduced pressure. Chromatographic
separation on a silica gel column with hexane/CH2Cl2 as the eluent
afforded 2c (1.73 g, 68%).
3.2 Reactions of benzoyl ferrocenes with propiophenones
In a typical reaction, TiCl4 was added dropwise to a suspension
of zinc powder in 80 mL of THF and the mixture was heated
at reflux for 2 h. A second solution was prepared by dissolving
the benzoyl ferrocene and propiophenone in THF (30 mL). The
latter solution was added dropwise to the first solution and the
resulting mixture was heated at reflux for a further 2 h. After
cooling to room temperature, the mixture was stirred with water
and CH2Cl2. The mixture was acidified with dilute HCl until the
dark colour disappeared and was decanted. The aqueous layer
was extracted with CH2Cl2 and the combination of organic layers
dried on MgSO4. This was followed by filtering the mixture and
removing the solvent under reduced pressure. Chromatographic
separation on a silica gel column with hexane/CH2Cl2 as the
eluent afforded compounds 2a–b, 2d–i or 3, respectively. The
reaction conditions and yields are summarized in Table 5, and
spectroscopic and analytical data for compounds 2 and 3 are
given in Tables 6 and 7. For the biological tests, each compound
was re-purified on semi-preparative HPLC with acetonitrile as the
eluent. The acetonitrile was removed under reduced pressure and
the product was recrystallized from either a pentane/CH2Cl2 or a
hexane/CH2Cl2 solution.
3.4 Chemical oxidation reactions
3.4.1 Oxidation of 2c. Silver oxide (0.20 g, 0.9 mmol) was
added to a solution of 2c (0.12 g, 0.3 mmol) in 8 mL acetone
and the mixture was stirred at room temperature in the dark. At
hourly intervals a sample was withdrawn, filtered and the solvent
removed under reduced pressure, after which NMR spectroscopic
analysis was carried out. The presence of 4 was observed after
2 h; quantitative formation of 5c occurred after 4 h. 4: 1H NMR
3
4
[(CD3)2CO] d/ppm: 8.82 (dd, J = 10.2 Hz, J = 2.8 Hz, 1 H,
quinone ring), 7.90 (dd, 3J = 10.2 Hz, 4J = 2.8 Hz, 1 H, quinone
ring), 7.40 (dd, 3J = 10.1 Hz, 4J = 2.8 Hz, 1 H, quinone ring), 7.10
(dd, 3J = 10.1 Hz, 4J = 2.8 Hz, 1 H, quinone ring), 6.57 (dd, 3J =
10.2 Hz, 4J = 2.1 Hz, 1 H, quinone ring), 6.44 (dd, 3J = 10.2 Hz,
4J = 2.1 Hz, 1 H, quinone ring), 6.37 (dd, J = 10.1 Hz, J =
2.1 Hz, 1 H, quinone ring), 6.24 (dd, 3J = 10.1 Hz, 4J = 2.1 Hz, 1
H, quinone ring), 5.00 (m, 1 H, C5H4), 4.86 (m, 1 H, C5H4), 4.69
(m, 1 H, C5H4), 4.38 (s, 5 H, C5H5), 4.22 (m, 1 H, C5H4), 3.04
3
4
2
3
(doublet of quartet, J = 14.7 Hz, J = 7.6 Hz, 1 H, CH2), 2.52
2
3
(doublet of quartet, J = 14.7 Hz, J = 7.6 Hz, 1 H, CH2), 1.02
(t, J = 7.6 Hz, 3 H, CH3). 13C NMR [(CD3)2CO] d/ppm: 187.0,
186.9 (C O), 160.0, 157.4, 139.8, 139.7 (Cquat), 138.6, 137.1, 135.3,
=
130.3, 129.2, 128.8, 128.0, 116.3 (C6H4), 87.5 (Fcipso), 81.0, 77.8,
74.3, 72.5 (C5H4), 71.9 (C5H5), 28.4 (CH2), 13.9 (CH3). MS (CI):
3.3 Demethylation of 2i
m/z 423 (MH+), 451 (M + C2H5 ), calcd for M+: 422. HRMS
+
Compound 2i (2.63 g, 6 mmol) was dissolved in dry CH2Cl2
(150 mL) and boron tribromide (1.2 mL, 13 mmol) was added. The
mixture was stirred overnight at ambient temperature, after which
(C26H22FeO2): m/z 422.0961 (M+), calcd for M+: 422.0969. 5c: IR
-1
1
=
(n/cm ): 3591 (OH), 1716 (C O). H NMR [(CD3)2CO] d/ppm:
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 10871–10881 | 10877
©

Table 6 NMR data for compounds 2 and 3
Compound 1H NMR, d/ppm
13C NMR, d/ppm
2a
7.39–7.16 (m, 10 H, C6H5), 3.92 (s, 5 H, C5H5), 3.83 (t, J = 1.9 Hz, 2 H, C5H4), 3.20 (t,
143.4, 142.3, 139.9, 134.7 (Cquat), 129.6,
129.1, 128.4, 128.0, 126.6, 126.4 (C6H5),
86.0 (Fcipso), 69.3 (C5H4), 69.1 (C5H5), 68.1
(C5H4), 30.7 (CH2), 12.7 (CH3)
J = 1.9 Hz, 2 H, C5H4), 1.99 (q, J = 7.4 Hz, 2 H, CH2), 0.70 (t, J = 7.4 Hz, 3 H, CH3)
[major isomer]
7.01–6.85 (m, 10 H, C6H5), 4.14 (t, J = 1.9 Hz, 2 H, C5H4), 4.13 (t, J = 1.9 Hz, 2 H,
C5H4), 3.97 (s, 5 H, C5H5), 2.84 (q, J = 7.4 Hz, 2 H, CH2), 0.97 (t, J = 7.4 Hz, 3 H, CH3)
[minor isomer]
2b
7.39–7.16 (m, 5 H, C6H5), 7.06 (d, J = 7.9 Hz, 2 H, C6H4), 6.79 (d, J = 7.9 Hz, 2 H,
152.1, 141.5, 138.6, 132.9, 132.1 (Cquat),
C6H4), 4.87 (s, br, 1 H, OH), 4.02 (s, 5 H, C5H5), 3.95 (t, J = 1.9 Hz, 2 H, C5H4), 3.35 (t, 128.7, 127.3, 126.5, 124.7, 112.8 (C6H4,
J = 1.9 Hz, 2 H, C5H4), 2.03 (q, J = 7.5 Hz, 2 H, CH2), 0.70 (t, J = 7.5 Hz, 3 H, CH3)
C6H5), 86.3 (Fcipso), 68.1 (C5H5), 67.8
[major isomer]
(C5H4), 66.9 (C5H4), 28.7 (CH2), 10.9 (CH3)
7.01–6.92 (m, 5 H, C6H5), 6.86 (d, J = 7.9 Hz, 2 H, C6H4), 6.44 (d, J = 7.9 Hz, 2 H,
C6H4), 4.57 (s, br, 1 H, OH), 4.17 (s, 5 H, C5H5), 4.08 (t, J = 1.9 Hz, 2 H, C5H4), 3.49 (t,
J = 1.9 Hz, 2 H, C5H4), 2.86 (q, J = 7.4 Hz, 2 H, CH2), 0.97 (t, J = 7.4 Hz, 3 H, CH3)
[minor isomer]
2c
2d
7.10 (d, J = 8.3 Hz, 2 H, C6H4), 7.02 (d, J = 8.5 Hz, 2 H, C6H4), 6.84 (d, J = 8.5 Hz, 2
157.1, 156.8, 140.6, 135.5, 135.3, 134.6
H, C6H4), 6.83 (d, J = 8.3 Hz, 2 H, C6H4), 4.88 (s, 1 H, OH), 4.86 (s, 1 H, OH), 3.95 (s, 5 (Cquat), 131.4, 130.9, 116.1, 115.6 (C6H4),
H, C5H5), 3.90 (t, J = 1.9 Hz, 2 H, C5H4), 3.30 (t, J = 1.9 Hz, 2 H, C5H4), 1.99 (q, J =
7.3 Hz, 2 H, CH2), 0.69 (t, J = 7.3 Hz, 3 H, CH3) [major isomer]
87.7 (Fcipso), 70.0 (C5H4), 69.7 (C5H5), 68.5
(C5H4), 31.5 (CH2), 13.1 (CH3)
6.79 (d, J = 8.7 Hz, 2 H, C6H4), 6.73 (d, J = 8.7 Hz, 2 H, C6H4), 6.50 (d, J = 8.5 Hz, 2
H, C6H4), 6.48 (d, J = 8.5 Hz, 2 H, C6H4), 4.62 (s, 1 H, OH), 4.61 (s, 1 H, OH), 4.14 (t,
J = 1.9 Hz, 2 H, C5H4), 4.12 (t, J = 1.9 Hz, 2 H, C5H4), 3.99 (s, 5 H, C5H5), 2.82 (q, J =
7.5 Hz, 2 H, CH2), 0.96 (t, J = 7.5 Hz, 3 H, CH3) [minor isomer]
7.37–7.20 (m, 5 H, C6H5), 7.02 (d, J = 7.9 Hz, 2 H, C6H4), 6.82 (d, J = 7.9 Hz, 2 H,
154.4, 142.4, 139.7, 135.6, 134.8 (Cquat),
C6H4), 5.37 (s, br, 1 H, OH), 3.97 (s, 5 H, C5H5), 3.91 (t, J = 1.9 Hz, 2 H, C5H4), 3.33 (t, 130.4, 129.6, 127.9, 126.4, 115.4 (C6H4,
J = 1.9 Hz, 2 H, C5H4), 1.97 (q, J = 7.5 Hz, 2 H, CH2), 0.68 (t, J = 7.5 Hz, 3 H, CH3)
C6H5), 87.3 (Fcipso), 69.6 (C5H5), 69.4
(C5H4), 68.5 (C5H4), 30.7 (CH2), 12.8 (CH3)
[major isomer]
6.97–6.91 (m, 5 H, C6H5), 6.73 (d, J = 7.9 Hz, 2 H, C6H4), 6.43 (d, J = 7.9 Hz, 2 H,
C6H4), 5.00 (s, br, 1 H, OH), 4.13 (t, J = 1.9 Hz, 2 H, C5H4), 4.12 (t, J = 1.9 Hz, 2 H,
C5H4), 4.00 (s, 5 H, C5H5), 2.81 (q, J = 7.5 Hz, 2 H, CH2), 0.96 (t, J = 7.5 Hz, 3 H, CH3)
[minor isomer]
2e
7.40–7.17 (m, 5 H, C6H5), 7.12 (d, J = 8.3 Hz, 2 H, C6H4), 6.89 (d, J = 8.3 Hz, 2 H,
C6H4), 4.02 (s, 5 H, C5H5), 3.94 (t, J = 1.9 Hz, 2 H, C5H4), 3.81 (s, 3 H, OCH3), 3.25 (t,
J = 1.9 Hz, 2 H, C5H4), 2.03 (q, J = 7.5 Hz, 2 H, CH2), 0.70 (t, J = 7.5 Hz, 3 H, CH3)
[major isomer]
158.1, 149.7, 143.5, 140.4, 134.6 (Cquat),
130.5, 129.2, 128.4, 126.5, 113.3 (C6H4,
C6H5), 86.4 (Fcipso), 70.0 (C5H5), 69.7
(C5H4), 68.8 (C5H4), 55.3 (OCH3), 30.7
(CH2), 12.8 (CH3)
7.10–6.98 (m, 5 H, C6H5), 6.82 (d, J = 8.5 Hz, 2 H, C6H4), 6.55 (d, J = 8.5 Hz, 2 H,
C6H4), 4.17 (s, 5 H, C5H5), 4.07 (t, J = 1.9 Hz, 2 H, C5H4), 3.64 (s, 3 H, OCH3), 3.50 (t,
J = 1.9 Hz, 2 H, C5H4), 2.88 (q, J = 7.5 Hz, 2 H, CH2), 0.98 (t, J = 7.5 Hz, 3 H, CH3)
[minor isomer]
2f
7.36–7.19 (m, 5 H, C6H5), 7.10 (d, J = 7.9 Hz, 2 H, C6H4), 6.92 (d, J = 7.9 Hz, 2 H,
C6H4), 4.00 (s, 5 H, C5H5), 3.95 (t, J = 1.9 Hz, 2 H, C5H4), 3.81 (s, 3 H, OCH3), 3.37 (t,
J = 1.9 Hz, 2 H, C5H4), 1.99 (q, J = 7.5 Hz, 2 H, CH2), 0.68 (t, J = 7.5 Hz, 3 H, CH3)
[major isomer]
158.4, 142.4, 139.7, 135.4, 134.7 (Cquat),
130.2, 129.5, 127.9, 126.3, 113.9 (C6H4,
C6H5), 87.5 (Fcipso), 69.9 (C5H5), 69.7
(C5H4), 68.7 (C5H4), 55.3 (OCH3), 30.7
(CH2), 12.8 (CH3)
7.04–6.90 (m, 5 H, C6H5), 6.78 (d, J = 8.5 Hz, 2 H, C6H4), 6.53 (d, J = 8.5 Hz, 2 H,
C6H4), 4.15 (s, 5 H, C5H5), 4.09 (t, J = 1.9 Hz, 2 H, C5H4), 3.71 (t, J = 1.9 Hz, 2 H,
C5H4), 3.62 (s, 3 H, OCH3), 2.83 (q, J = 7.4 Hz, 2 H, CH2), 0.96 (t, J = 7.4 Hz, 3 H,
CH3) [minor isomer]
2g
2h
7.16 (d, J = 8.7 Hz, 2 H, C6H4), 7.04 (d, J = 8.7 Hz, 2 H, C6H4), 6.91 (d, J = 8.7 Hz, 2
H, C6H4), 6.90 (d, J = 8.7 Hz, 2 H, C6H4), 3.88 (s, 5 H, C5H5), 3.79 (t, J = 1.9 Hz, 2 H,
157.3, 157.1, 138.8, 134.7, 133.8, 133.3
(Cquat), 129.6, 129.1, 112.8, 112.2 (C6H4),
C5H4), 3.74 (s, 3 H, OCH3), 3.73 (s, 3 H, OCH3), 3.17 (t, J = 1.9 Hz, 2 H, C5H4), 1.94 (q, 85.6 (Fcipso), 68.3 (C5H4), 68.0 (C5H5), 67.0
J = 7.5 Hz, 2 H, CH2), 0.62 (t, J = 7.5 Hz, 3 H, CH3) [major isomer]
(C5H4), 54.2 (OCH3), 29.8 (CH2), 11.7
(CH3)
6.94–6.88 (m, 4 H, C6H4), 6.76 (d, J = 8.9 Hz, 2 H, C6H4), 6.51 (d, J = 8.7 Hz, 2 H,
C6H4), 4.17 (t, J = 1.9 Hz, 2 H, C5H4), 4.09 (t, J = 1.9 Hz, 2 H, C5H4), 3.90 (s, 5 H,
C5H5), 3.73 (s, 3 H, OCH3), 3.55 (s, 3 H, OCH3), 2.40 (q, J = 7.5 Hz, 2 H, CH2), 0.80 (t,
J = 7.5 Hz, 3 H, CH3) [minor isomer]
7.13 (d, J = 8.7 Hz, 2 H, C6H4), 7.01 (d, J = 8.7 Hz, 2 H, C6H4), 6.90 (d, J = 8.7 Hz, 2
158.3, 154.0, 139.9, 135.7, 135.0, 134.2
H, C6H4), 6.84 (d, J = 8.7 Hz, 2 H, C6H4), 4.75 (s, 1 H, OH), 3.92 (s, 5 H, C5H5), 3.86 (t, (Cquat), 130.8, 130.1, 114.8, 113.9 (C6H4),
J = 1.9 Hz, 2 H, C5H4), 3.81 (s, 3 H, OCH3), 3.25 (t, J = 1.9 Hz, 2 H, C5H4), 1.99 (q,
J = 7.5 Hz, 2 H, CH2), 0.69 (t, J = 7.5 Hz, 3 H, CH3) [major isomer]
86.6 (Fcipso), 69.3 (C5H4), 69.1 (C5H5), 68.1
(C5H4), 55.3 (OCH3), 30.9 (CH2), 12.8
(CH3)
10878 | Dalton Trans., 2009, 10871–10881
This journal is
The Royal Society of Chemistry 2009
©

Table 6 (Contd.)
Compound 1H NMR, d/ppm
13C NMR, d/ppm
6.80 (d, J = 8.6 Hz, 2 H, C6H4), 6.79 (d, J = 8.5 Hz, 2 H, C6H4), 6.55 (d, J = 8.6 Hz, 2
H, C6H4), 6.50 (d, J = 8.5 Hz, 2 H, C6H4), 4.49 (s, 1 H, OH), 4.13 (t, J = 1.9 Hz, 2 H,
C5H4), 4.11 (t, J = 1.9 Hz, 2 H, C5H4), 3.98 (s, 5 H, C5H5), 3.64 (s, 3 H, OCH3), 2.82 (q,
J = 7.5 Hz, 2 H, CH2), 0.95 (t, J = 7.5 Hz, 3 H, CH3) [minor isomer]
2i
7.17 (d, J = 8.9 Hz, 2 H, C6H4), 7.05 (d, J = 8.7 Hz, 2 H, C6H4), 6.91 (d, J = 8.9 Hz, 2
158.1, 154.2, 139.8, 135.9, 134.8, 134.4
H, C6H4), 6.84 (d, J = 8.7 Hz, 2 H, C6H4), 5.12 (s, 1 H, OH), 3.92 (s, 5 H, C5H5), 3.86 (t, (Cquat), 130.6, 130.3, 115.4, 113.3 (C6H4),
J = 1.7 Hz, 2 H, C5H4), 3.81 (s, 3 H, OCH3), 3.26 (t, J = 1.7 Hz, 2 H, C5H4), 1.98 (q,
J = 7.5 Hz, 2 H, CH2), 0.69 (t, J = 7.5 Hz, 3 H, CH3) [major isomer]
86.6 (Fcipso), 69.3 (C5H4), 69.0 (C5H5), 68.0
(C5H4), 55.3 (OCH3), 30.8 (CH2), 12.7
(CH3)
6.84 (d, J = 8.9 Hz, 2 H, C6H4), 6.74 (d, J = 8.7 Hz, 2 H, C6H4), 6.58 (d, J = 8.9 Hz, 2
H, C6H4), 6.48 (d, J = 8.7 Hz, 2 H, C6H4), 4.77 (s, 1 H, OH), 4.13 (t, J = 1.9 Hz, 2 H,
C5H4), 4.10 (t, J = 1.9 Hz, 2 H, C5H4), 3.98 (s, 5 H, C5H5), 3.66 (s, 3 H, OCH3), 2.82 (q,
J = 7.5 Hz, 2 H, CH2), 0.96 (t, J = 7.5 Hz, 3 H, CH3) [minor isomer]
8.27 (s, 2 H, OH), 6.81 (d, J = 8.8 Hz, 4 H, C6H4), 6.64 (d, J = 8.8 Hz, 4 H, C6H4), 4.11
(t, J = 1.9 Hz, 2 H, C5H4), 3.97 (s, 5 H, C5H5), 3.72 (t, J = 1.9 Hz, 2 H, C5H4), 2.55 (q,
J = 7.2 Hz, 2 H, CH2), 0.91 (t, J = 7.2 Hz, 3 H, CH3)
3a
209.1 (C O), 157.0, 135.9 (Cquat), 131.8,
=
115.0 (C6H4), 92.3 (Fcipso), 71.5 (C5H4), 70.1
(C5H5), 68.7 (C5H4), 66.2 (C), 35.1 (CH2),
9.4 (CH3)
a In (CD3)2CO.
Table 7 Infrared, MS and analytical data for compounds 2 and 3
MS, m/z found
HRMS, m/z found (calculated
Elemental analysis: found
(calculated)%
Compound
IR (n/cm-1)
(calculated for M+)
for M+)
2a
2b
2c
2d
2e
2f
2g
2h
2i
2973, 2929, 2869 (CH2, CH3)
392 (392)
C26H24Fe: 392.1222 (392.1227)
C26H24FeO: 408.1171 (408.1177)
C26H24Fe·0.25C5H12: C 80.12
(79.76), H 6.22 (6.63)a
C26H24FeO·0.25H2O: C 75.71
(75.65), H 5.82 (5.98)b
C26H24FeO2·0.25H2O: C 72.91
(72.82), H 5.61 (5.76)b
C26H24FeO·0.75H2O: C 74.15
(74.03), H 6.27 (6.09)b
C27H26FeO: C 76.75 (76.78), H 6.11
(6.21)
C27H26FeO: C, 76.67 (76.78); H, 6.06
(6.21)
C28H28FeO2: C 74.59 (74.34), H 6.31
(6.24)
3310 (OH), 2957, 2927 (CH2,
CH3)
3333 (OH), 2972, 2938 (CH2,
CH3)
408 (408)
424 (424)
C26H25FeO2: 425.1199 [MH+]
(425.1204)
C26H24FeO: 408.1166 (408.1177)
3368 (OH), 2970 (CH2, CH3)
408 (408)
2966 (CH2, CH3)
422 (422)
—
—
—
2957, 2929 (CH2, CH3)
2966, 2933, 2837 (CH2, CH3)
3428 (OH), 2964 (CH2, CH3)
422 (422)
452 (452)
439 [MH+] (439)
438 (438)
C27H27FeO2: 439.1360 [MH+]
(439.1361)
—
C27H26FeO2·0.5H2O: C 72.61
(72.49), H 5.99 (6.08)b
C27H26FeO2: C 73.92 (73.98), H 5.81
(5.98)
3232 (OH), 2969, 2933 (CH2,
CH3)
3
3410, 3333 (OH), 2969, 2940
(CH2, CH3) 1736 (C O)
440 (440)
C26H24FeO3: 440.1069
(440.1075)
C26H24FeO3·0.25CH2Cl2: C 68.72
(68.31), H 5.33 (5.35)c
=
a Presence of pentane in the analytical sample was verified by 1H NMR spectroscopy. b Presence of water in the analytical sample was verified by 1H NMR
spectroscopy. c Presence of dichloromethane in the analytical sample was verified by 1H NMR spectroscopy.
8.76 (dd, 3J = 10.0 Hz, 4J = 2.1 Hz, 1 H, quinone ring), 8.44 (s, 1 H,
OH), 7.33 (dd, 3J = 10.0 Hz, 4J = 2.1 Hz, 1 H, quinone ring), 7.14
(d, J = 8.7 Hz, 2 H, phenol ring), 6.76 (d, J = 8.7 Hz, 2 H, phenol
ring), 6.53 (dd, 3J = 10.0 Hz, 4J = 1.8 Hz, 1 H, quinone ring), 6.26
(dd, 3J = 10.0 Hz, 4J = 1.8 Hz, 1 H, quinone ring), 6.23 (q, J =
7.0 Hz, 1 H, CHCH3), 4.76 (m, 1 H, C5H4), 4.67 (m, 1 H, C5H4),
4.54 (m, 1 H, C5H4), 4.50 (m, 1 H, C5H4), 4.25 (s, 5 H, C5H5),
1.77 (d, J = 7.0 Hz, 3 H, CH3). 13C NMR (CDCl3) d/ppm: 186.9
H 5.46%. Presence of water in the analytical sample was verified
by 1H NMR spectroscopy.
3.4.2 Oxidation of 2h
Silver oxide (0.20 g, 0.9 mmol) was added to a solution of 2h
(0.12 g, 0.3 mmol) in 8 mL acetone and the mixture was stirred
at room temperature in the dark. At hourly intervals a sample
was withdrawn, filtered and the solvent removed under reduced
pressure, after which NMR spectroscopic analysis was carried
(C O), 162.1, 157.8, 141.3 (Cquat), 139.8, 137.5 (C6H4), 132.6,
=
128.8 (Cquat), 128.0, 127.9 (C6H4), 127.8 (CHCH3), 123.6, 116.1
(C6H4), 81.0 (Fcipso), 76.0, 73.2, 72.7, 71.9 (C5H4), 71.3 (C5H5),
15.8 (CH3). MS (EI): m/z 422 (M+), calcd for M+: 422. HRMS
(C26H22FeO2): m/z 423.1038 (MH+), calcd for MH+: 423.1047.
Found for C27H22FeO2·1.5H2O: C 70.41, H 5.64. Calcd: C 70.29,
out. The formation of 5h was observed after 2 h; after 4 h
-1
=
decomposition of 5h was detected. 5h: IR (n/cm ): 1712 (C O).
1H NMR [(CD3)2CO] d/ppm: 8.77 (dd, 3J = 9.9 Hz, 4J = 2.8 Hz, 1
H, quinone ring), 8.32 (dd, 3J = 9.9 Hz, 4J = 2.8 Hz, 1 H, quinone
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 10871–10881 | 10879
©

ring), 7.23 (d, J = 9.0 Hz, 2 H, methoxyphenyl ring), 6.85 (d, J =
kanamycine, at 37 ◦C in a 5% CO2 air humidified incubator.
For proliferation assays, cells were plated in 24-well sterile plates
with 1.5 ¥ 104 cells in 1 mL of DMEM without phenol red,
supplemented with 9% of fetal calf serum desteroided on dextran
charcoal, 0.9% Glutamax I and 0.9% kanamycine, and were
incubated for 24 h. The following day (D0), 1 mL of the same
medium containing the compounds to be tested diluted in DMSO,
was added to the plates (final volumes of DMSO: 0.1%; 3 wells for
each condition). After three days (D3), the incubation medium was
removed and 2 mL of fresh medium containing the compounds
was added. At different days (D3, D4, D5 and D6), the protein
content of each well was quantified by methylene blue staining as
follows. Cell monolayers were fixed and stained for 1 h in methanol
with methylene blue (2 mg mL-1), and then washed thoroughly
with water. Two mL of HCl (0.1 M) was then added, and the
plate was incubated for 1 h at 37 ◦C. The absorbance of each well
was then measured at 655 nm with a Biorad microplate reader.
The results are expressed as the percentage of proteins versus the
control. Experiments were performed at least in duplicate.
3
4
9.0 Hz, 2 H, methoxyphenyl ring), 6.53 (dd, J = 9.9 Hz, J =
2.1 Hz, 1 H, quinone ring), 6.28 (q, J = 7.0 Hz, 1 H, CHCH3),
6.25 (dd, 3J = 9.9 Hz, 4J = 2.1 Hz, 1 H, quinone ring), 4.76 (m,
1 H, C5H4), 4.67 (m, 1 H, C5H4), 4.54 (m, 1 H, C5H4), 4.50 (m, 1
H, C5H4), 4.26 (s, 5 H, C5H5), 3.75 (s, 3 H, OCH3), 1.79 (d, J =
7.0 Hz, 3 H, CH3). MS (CI): m/z 437 (MH+), calcd for M+: 436.
The instability of this compound precluded characterization by
13C NMR.
3.5 Electrochemistry
Cyclic voltammograms (CVs) were obtained using a three elec-
trode cell with a 0.5 mm Pt working electrode, stainless steel rod
counter electrode, and Ag/AgCl ethanol reference electrode, with
an m-Autolab 3 potentiostat driven by GPES software (General
Purpose Electrochemical System, Version 4.8, EcoChemie B.V.,
Utrecht, the Netherlands). Solutions consisted of 10 mL MeOH,
approximately 1 mM analyte, and 0.1 M TBABF4 supporting
electrolyte.
3.7 X-Ray crystal structure determinations
3.6 Biochemical experiments
Crystal data for Z-2a and 3 were collected using a Bruker SMART
APEX CCD area detector diffractometer. Crystal data for 2g,
2h·0.5H2O and 2i were collected using an Enraf-Nonius Kappa-
CCD diffractometer equipped with a CCD two-dimensional
detector. Measurement and refinement details are listed in Table 8.
A full sphere of the reciprocal space was scanned. Data were
corrected for Lorentz and polarization effects, and a pseudo-
empirical absorption correction based on redundant reflections
3.6.1 Materials. Stock solutions (1 ¥ 10-2 M) were prepared
in DMSO. Dulbecco’s modified eagle medium (DMEM), fetal calf
serum, glutamine and kanamycine were obtained from Invitrogen.
MDA-MB-231 cells were from ATCC-LGC Standards.
3.6.2 Culture conditions. Cells were maintained in mono-
layer culture in DMEM with phenol red/Glutamax I, supple-
mented with 9% of decomplemented fetal calf serum and 0.9%
Table 8 Crystal data for Z-2a, Z-2g, Z-2h·0.5H2O, Z-2i and 3
Compound
Z-2a
Z-2g
Z-2h·0.5H2O
Z-2i
3
1
2
Formula
C26H24Fe
392.30
Orthorhombic
C28H28FeO2
452.38
Triclinic
¯
P1
C27H26FeO2· H2O
C27H26FeO2
438.35
Monoclinic
C26H24O3Fe
440.30
Monoclinic
FW/g mol-1
Crystal system
Space group
447.36
Orthorhombic
P212121
Pcab
P21/n
P21/n
˚
a/A
7.6189(6)
15.2824(11)
16.6026(12)
90
90
90
1933.1(2)
4
1.348
0.787
824
9.0998(8)
10.4180(11)
12.3362(12)
97.949(7)
96.147(7)
96.436(6)
1142.0(2)
2
1.315
0.682
476
250
0.08 ¥ 0.11 ¥ 0.15
1.680 to 39.995
36 861
7611 (0.0347)
99.3 (39.995)
1.0–0.87
10.3231(10
17.354(2)
50.204(6)
90
90
90
8993.7(17)
16
1.321
0.694
3728
5.9954(4)
24.330(2)
14.9313(16)
90
95.613(6)
90
2167.6(3)
4
1.343
9.1008(4)
10.2954(4)
21.5897(9)
90
101.042(1)
90
1985.43(14)
4
1.473
˚
b/A
˚
c/A
a/◦
b/◦
g /◦
3
˚
V/A
Z
rc/mg m-3
m(MoKa)/mm-1
F(000)
0.716
920
200
0.786
920
100
T/K
100
200
Crystal size/mm3
0.8 ¥ 0.3 ¥ 0.3
1.81 to 32.00
23 546
6259 (0.0294)
100.0 (30.00)
0.7981–0.6015
6259/0/340
1.048
R1 = 0.0322
wR2 = 0.0783
R1 = 0.0342
wR2 = 0.0794
0.839 and -0.362
-0.006(10)
0.06 ¥ 0.09 ¥ 0.20
1.622 to 28.338
8396
8390 (0.1116)
85.5 (28.338)
0.96–0.92
8390/0/551
0.9897
R1 = 0.1116
wR2 = 0.1559
R1 = 0.1117
wR2 = 0.1561
0.99 and -0.75
n/a
0.08 ¥ 0.14 ¥ 0.30
2.164 to 30.014
6246
6242 (0.0870)
99.6 (30.014)
0.94–0.89
6242/0/273
0.9573
R1 = 0.0870
wR2 = 0.1133
R1 = 0.0870
wR2 = 0.1133
1.01 and -0.81
n/a
0.50 ¥ 0.50 ¥ 0.40
1.92 to 31.60
22 874
6227 (0.0217)
99.8 (30.00)
0.7439–0.6206
6227/0/274
1.055
R1 = 0.0440
wR2 = 0.1175
R1 = 0.0502
wR2 = 0.1221
0.865 and -0.230
n/a
q range/◦
Reflections collected
Independent reflections (Rint
Completeness(%) (to q/◦)
Transmission range
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2s(I)]
)
7611/0/281
1.1069
R1 = 0.0347
wR2 = 0.0401
R1 = 0.0798
wR2 = 0.0812
0.48 and -0.57
n/a
R indices (all data)
-3
˚
Largest diffraction peak and hole/e A
Flack parameter
10880 | Dalton Trans., 2009, 10871–10881
This journal is
The Royal Society of Chemistry 2009
©

was performed by the program SADABS.41 The structures were
solved by direct methods using SHELXS-97,42 and refined by
full matrix least-squares on F2 for all data using SHELXL-97.43
In Z-2a, all hydrogen atoms were located in the difference
Fourier map and allowed to refine freely with isotropic thermal
displacement factors. All other hydrogen atoms were added at
calculated positions and refined using a riding model. Their
isotropic temperature factors were fixed to 1.2 times (1.5 times for
methyl groups) the equivalent isotropic displacement parameters
of the carbon atom the H-atom is attached to. Anisotropic
temperature factors were used for all non-hydrogen atoms.
9 G. Jaouen, S. Top, A. Vessie`res, G. Leclercq and M. J. McGlinchey,
Curr. Med. Chem., 2004, 11, 2505–2517.
10 A. Nguyen, A. Vessie`res, E. A. Hillard, S. Top, P. Pigeon and G. Jaouen,
Chimia, 2007, 61, 716–724.
11 G. Jaouen, S. Top, A. Vessie`res, B. S. Sayer, C. S. Frampton and M. J.
McGlinchey, Organometallics, 1992, 11, 4061–4068.
12 E. A. Hillard, A. Vessie`res, L. Thouin, G. Jaouen and C. Amatore,
Angew. Chem., Int. Ed., 2006, 45, 285–290.
13 E. A. Hillard, P. Pigeon, A. Vessie`res, C. Amatore and G. Jaouen,
Dalton Trans., 2007, 5073–5081.
14 D. J. Collins, J. J. Hobbs and C. W. Emmens, J. Med. Chem., 1971, 14,
952–957.
15 G. R. Bedford and D. N. Richardson, Nature, 1966, 212, 733–734.
16 P. Ha¨rter, K. Latzel, M. Spiegler and E. Herdtweck, Polyhedron, 1998,
17, 1141–1148.
17 Y. Ortin, J. Grealis, C. Scully, H. Mu¨ller-Bunz, A. R. Manning and
M. J. McGlinchey, J. Organomet. Chem., 2004, 689, 4683–4690.
18 P. Denifl, A. Hradsky, B. Bildstein and K. Wurst, J. Organomet. Chem.,
1996, 523, 79–91.
19 A. Vessie`res, S. Top, P. Pigeon, E. A. Hillard, L. Boubeker, D. Spera
and G. Jaouen, J. Med. Chem., 2005, 48, 3937–3940.
20 V. W. Winkler, M. A. Nyman and R. S. Egan, Steroids, 1971, 17, 197–
207.
4. Conclusion
In our ongoing investigations on the cytotoxicity of ferrocenyl
phenols, we have found further evidence that quinoid structures
are likely to be the active species in hormone-independent breast
cancer cells. Organometallic moieties analogous to DES Q as well
as an intermediate in the isomerisation from DES Q to DIES,
viz. 4 and 5c, respectively, have been isolated and characterized.
Compounds 1, in which the ferrocene and phenol groups are
linked to separate carbons on the olefinic bond, are more
efficient in inhibiting the proliferation of MDA-MB-231 breast
cancer cells than their isomers 2, which have the ferrocene and
phenol functionalities attached to the same carbon atom on the
double bond. This study thus enhances our understanding on the
structure-activity relationship of such molecules for the design of
more potent redox active cancer drug candidates.
21 D. W. Robertson and J. A. Katzenellenbogen, J. Org. Chem., 1982, 47,
2387–2393.
22 M. Jarman and R. McCague, J. Chem. Res., 1985, 4, 116–117.
23 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91, 165–195.
24 S. Barlow, H. E. Bunting, C. Ringham, J. C. Green, G. U. Bublitz, S. G.
Boxer, J. W. Perry and S. R. Marder, J. Am. Chem. Soc., 1999, 121,
3715–3723.
25 S. Barlow and D. Ohare, Chem. Rev., 1997, 97, 637–669.
26 K. R. Gordon and K. D. Warden, Inorg. Chem., 1978, 17, 987–994.
27 P. A. Dowben, D. C. Driscoll, R. S. Tate and N. M. Boag,
Organometallics, 1988, 7, 305–308.
28 M. Rosenblum, J. O. Santer and W. G. Howells, J. Am. Chem. Soc.,
1963, 85, 1450–1456.
29 T. Itoh, Chem. Rev., 1995, 95, 2351–2368.
30 S. B. Colbran, S. T. Lee, D. G. Lonnon, F. J. D. Maharaj, A. M.
McDonagh, K. A. Walker and R. D. Young, Organometallics, 2006,
25, 2216–2224.
31 H. Tukada, Tetrahedron Lett., 2008, 49, 3766–3769.
32 J. G. Liehr, B. B. Dague, A. M. Ballatore and J. Henkin, Biochem.
Pharmacol., 1983, 32, 3711–3718.
33 J. G. Liehr, A. M. Bellatore, J. A. McLachlan and D. A. Sirbasku,
Cancer Res., 1983, 43, 2678–2682.
34 Y. J. Abul-Hajj, K. Tabakovic, W. J. Ojala and W. B. Gleason, J. Am.
Chem. Soc., 1995, 117, 5600–5601.
35 J. Axelrod, Biochem. J., 1956, 63, 634–639.
Acknowledgements
We thank P. Herson for crystallographic assistance, Dr M.-N.
Rager for the 2D NMR experiments, as well as the Agence Na-
tionale de la Recherche (no. ANR-06-BLAN-0384-01, “FerVect”)
and COST (D39 WG001) for financial support. Y. L. K. T. ac-
knowledges a postdoctoral fellowship from the Agence Nationale
de la Recherche.
36 D. C. Spink, F. Zhang, M. M. Hussain, B. H. Katz, X. Liu, D. R. Hilker
and J. L. Bolton, Chem. Res. Toxicol., 2001, 14, 572–581.
37 O. Do¨hr, C. Vogel and J. Abel, Arch. Biochem. Biophys., 1995, 321,
405–412.
38 D. C. Spink, B. C. Spink, J. Q. Cao, J. A. DePasquale, B. T. Pentecost,
M. J. Fasco, Y. Li and T. R. Sutter, Carcinogenesis, 1998, 19, 291–298.
39 W. G. R. Angus, M. C. Larsen and C. R. Jefcoate, Carcinogenesis, 1999,
20, 947–955.
References
1 M. M. Buckley and K. L. Goa, Drugs, 1989, 37, 451–490.
2 V. C. Jordan, J. Med. Chem., 2003, 46, 883–908.
3 K. Dhingra, Invest. New Drugs, 1999, 17, 285–311.
4 J. I. MacGregor and V. C. Jordan, Pharmacol. Rev., 1998, 50, 151–196.
5 S. Top, B. Dauer, J. Vaissermann and G. Jaouen, J. Organomet. Chem.,
1997, 541, 355–361.
40 A. C. Benyei, C. Glidewell, P. Lightfoot, B. J. L. Royles and D. M.
Smith, J. Organomet. Chem., 1997, 539, 177–186.
41 G. M. Sheldrick, SADABS, 2000, Bruker AXS Inc, Madison, WI,
53711.
42 G. M. Sheldrick, SHELXS-97, 1997, University of Go¨ttingen.
43 G. M. Sheldrick, SHELXL-97-2, 1997, University of Go¨ttingen.
6 S. Top, A. Vessie`res, G. Leclercq, J. Quivy, J. Tang, J. Vaissermann,
M. Huche´ and G. Jaouen, Chem.–Eur. J., 2003, 9, 5223–5236.
7 G. Jaouen, S. Top, A. Vessie`res, G. Leclercq, J. Quivy, L. Jin and
A. Croisy, C. R. Acad. Sci. Paris Chemie, 2000, 3, 89–95.
8 S. Top, A. Vessie`res, C. Cabestaing, I. Laios, G. Leclercq, C. Provot and
G. Jaouen, J. Organomet. Chem., 2001, 637–639, 500–506.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 10871–10881 | 10881
©