Ferrocenyl Paclitaxel and Docetaxel Derivatives: Impact of an Organometallic Moiety on the Mode of Action of Taxanes

Article abstract of DOI:10.1002/chem.201601809

A series of ferrocenyl analogues and derivatives of paclitaxel and docetaxel were synthesised and assayed for their antiproliferative/cytotoxic effects, impact on the cell cycle distribution and ability to induce tubulin polymerisation. The replacement of the 3′-N-benzoyl group of paclitaxel with a ferrocenoyl moiety, in particular, led to formation of an analogue that was at least one order of magnitude more potent in terms of antiproliferative activity than the parent compound (IC₅₀values of 0.11 versus 1.11 μm, respectively), but still preserved the classical taxane mode of action, that is, microtubule stabilisation leading to mitotic arrest. Molecular docking studies revealed an unexpected binding pocket in the tubulin structure for the ferrocenoyl group introduced in the paclitaxel backbone.

Full text of DOI:10.1002/chem.201601809

DOI: 10.1002/chem.201601809
Full Paper
&
Drug Discovery
Ferrocenyl Paclitaxel and Docetaxel Derivatives: Impact of an
Organometallic Moiety on the Mode of Action of Taxanes
[
a]
[b]
[b]
[c]
˙
Anna Wieczorek, Andrzej Błau z˙ , Aleksandra Zal, Homayon John Arabshahi,
[c]
[c]
[b]
[a]
Jóhannes Reynisson, Christian G. Hartinger, Bła z˙ ej Rychlik,* and Damian Pla z˙ uk*
Abstract: A series of ferrocenyl analogues and derivatives of
paclitaxel and docetaxel were synthesised and assayed for
their antiproliferative/cytotoxic effects, impact on the cell
cycle distribution and ability to induce tubulin polymeri-
sation. The replacement of the 3’-N-benzoyl group of pacli-
taxel with a ferrocenoyl moiety, in particular, led to forma-
tion of an analogue that was at least one order of magni-
tude more potent in terms of antiproliferative activity than
the parent compound (IC50 values of 0.11 versus 1.11 mm, re-
spectively), but still preserved the classical taxane mode of
action, that is, microtubule stabilisation leading to mitotic
arrest. Molecular docking studies revealed an unexpected
binding pocket in the tubulin structure for the ferrocenoyl
group introduced in the paclitaxel backbone.
Introduction
of the paclitaxel side chain strongly affects its antiproliferative
activity, in some cases increasing the anticancer activity in
[1]
[6a,f,j]
The antineoplastic agent paclitaxel (Taxol) 1 promotes poly-
comparison to paclitaxel
(e.g., esterification of the 2’-hy-
merisation of tubulin and stabilises microtubules, thus disturb-
droxy group with polyunsaturated fatty acids increases its ac-
tivity against drug-resistant colon and drug-sensitive ovarian
[
2,3]
ing cellular division and, consequently, leading to cell death.
[
17]
It was approved by the FDA for treatment of advanced ovarian
and breast cancer in 1992. Since that time, paclitaxel and its
semisynthetic analogue docetaxel 2, which was approved by
the FDA in 1994, are widely used chemotherapeutics for treat-
ment of many types of neoplastic diseases, including, but not
limited to lung, ovarian, breast, head and neck cancers, mela-
tumours in mice, and replacement of the 3’-phenyl group by
alkyl, alkenyl or aryl groups or modification of 3’-N-acyl group
[
7,8]
also improves cytotoxicity ). On the other hand, the influence
of modifying the paclitaxel skeleton on its cytotoxicity strongly
depends on the modification site, for example, acylation or ep-
imerisation of the 7-OH group usually do not influence or even
lead to a lack of activity of the resulting compound, whereas
modification of the 3-benzoyloxy group usually improves cyto-
[4,5]
noma and Kaposi’s sarcoma (Figure 1).
In the last two decades, a large number of structure–activity
relationship (SAR) studies of modified paclitaxel have been
[
9]
toxicity.
[
6]
published. Two main types of structural modifications of
were carried out: modifications performed at the side chain
of paclitaxel and at the taxol skeleton. Usually, the modification
The success of platinum-based complexes, such as cisplatin,
in anticancer therapy stimulated to search for new metal-con-
taining anticancer drug candidates. Tremendous progress in
bioorganometallic chemistry has led to the development of an-
1
[
10]
[10a,11]
[12]
ticancer, antimalarial
and antibacterial compounds in
[
a] A. Wieczorek, Dr. D. Pla z˙ uk
Department of Organic Chemistry
Faculty of Chemistry
University of Łódz´
Tamka 12, 41-403 Łód z´ (Poland)
E-mail: damplaz@uni.lodz.pl
[
b] Dr. A. Błau z˙ , A. Z˙ al, Dr. B. Rychlik
Cytometry Lab
Department of Molecular Biophysics
Faculty of Biology and Environmental Protection
University of Łódz´
ul. Pomorska 141/143, 90-236 Łód z´ (Poland)
E-mail: brychlik@biol.uni.lodz.pl
[
c] H. J. Arabshahi, Dr. J. Reynisson, Prof. Dr. C. G. Hartinger
School of Chemical Sciences
The University of Auckland
Private Bag 92019, Auckland 1142 (New Zealand)
Supporting information and the ORCID identification number(s) for the au-
thor(s) of this article can be found under http://dx.doi.org/10.1002/
chem.201601809.
Figure 1. The structures of paclitaxel 1 and docetaxel 2.
11413 ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2016, 22, 11413 – 11421

Full Paper
[
24]
recent years. A metal atom in various oxidation states gives
access to structural diversity and biological properties that are
ylic acid and slight excess of oxalyl chloride at RT ). The N-
acylation of 5 with 4-ferrocenebutyric acid using an excess of
diisopropylcarbodiimide (DIC) as a coupling agent in the pres-
ence of a catalytic amount of 4-dimethylaminopyridine (DMAP)
in dichloromethane at RT led to N-(4-ferrocenylbutyryl)-3-trie-
thylsilyloxyazetidin-2-one (7) in 56% yield (Scheme 2).
[14]
different from those of typical organic compounds. Depend-
ing on the nature of the metal–ligand bond, bioactive organo-
metallic compounds may feature covalent metal–carbon
bonds and/or labile ligands. Ferrocene is a redox active metal-
locene that is non-cytotoxic and stable in biological media.
However, when incorporated in or conjugated with biologically
active compounds, it provides access to promising cytotoxic
agents with non-conventional modes of action. Many ferrocen-
yl compounds have recently been prepared through conjuga-
[13]
tion of a ferrocene moiety with nucleobases, peptides, vita-
[
14]
[15]
[16]
mins or phenols, which exhibited significant anticancer,
[
11a,17]
[18]
antimalarial
or antibacterial activity. Some of these com-
pounds exhibited unexpected novel properties, for example,
ferrocenyl curcuminoids interfered with microtubule polymeri-
Scheme 1. Synthesis of compound 4
pyridine, RT, 5 min; (b) LiHMDS, then CH
.
Reagents and conditions: (i) (a) TESCl,
COCl, THF, À408C, 30 min.
[
19]
3
sation.
In our continuous efforts to develop organometallic anti-
cancer agents, we developed and studied the biological prop-
erties of ferrocenyl-functionalised taxanes, such as paclitaxel
and docetaxel, which are low-molecular mass microtubule de-
polymerisation inhibitors, as new antimitotic drug candidates.
Our preliminary study showed that a simple esterification of
paclitaxel with ferrocenecarboxylic acid and 3-ferrocenoylpro-
pionic acid at the 2’-O-position leads to highly cytotoxic ferro-
[
20]
cenyl conjugates. Thus, we decided to investigate in a more
detailed way the mode of action of ferrocenyl-modified tax-
anes focusing on their interference with the cellular microtu-
bule network. We prepared two sets of ferrocenyl analogues
and conjugates of paclitaxel and docetaxel having the organo-
metallic group at the side chain of paclitaxel and at the taxol
skeleton. In particular, we describe the synthesis and studies of
the interaction with tubulin, the impact on the cell cycle distri-
bution and the cytotoxic activity. As several ferrocenyl deriva-
tives show increased activity against multidrug resistant cells
Scheme 2. Synthesis of N-ferrocenyl-substituted azetidin-2-ones 6 and 7. Re-
agents and conditions: (i) FcCOCl (Fc=ferrocenyl), Et N, DMAP, DCM, 0–RT,
2 h; (ii) Fc(CH COOH, DIC, DMAP, DCM, RT, 24 h.
3
2 3
)
[21]
(
e.g., a ferrocenyl plinabulin analogue) the antiproliferative
In the next step, 13-O-acylation of 4 with azetidin-2-ones 6
[
22a]
activity of the organometallic taxanes in cells exhibiting elevat-
ed expression of specific ABC transporters responsible for the
MDR (multidrug resistance) phenotype were also assayed.
and 7 under typical conditions
using LiHMDS as a base at
À408C gave the corresponding paclitaxels 8 and 9 in good
yields of up to 84%. Further deprotection of hydroxy groups
with an excess of HF·pyridine (HF·Py) in a solution of pyridine
and acetonitrile at RT for 24 h gave the desired ferrocenyl ana-
logues of paclitaxel 10 and 11 in good yields of up to 90%
(Scheme 3).
Results and Discussion
To synthesise ferrocenyl analogues of paclitaxel bearing a ferro-
cenoyl group instead of the benzoyl group at the 3’-N position,
an established procedure was adopted, starting from optical-
To introduce a ferrocenyl moiety at the 2’-O-position of pa-
clitaxel and docetaxel, an established procedure was used for
selective acylation of 1 and 2 with various ferrocenecarboxylic
acids (Scheme 4). The desired 2’-O-ferrocenyl substituted pa-
clitaxels 12–15 and docetaxel derivatives 16–19 were obtained
in good to excellent yields.
[22]
ly pure (3R,4S)-3-triethylsilyloxy-4-phenylazetidin-2-one (5) and
[
20]
1
0-deacetylbaccatin III (3). In the reaction of 3 with triethylsilyl-
[23]
chloride (TESCl) in pyridine, the 7-OH group was selectively
protected as a triethylsilyl ether, followed by a selective O-ace-
tylation of the 10-OH with LiHMDS and acetyl chloride in THF
at À408C for 30 min leading to the desired compound 4 in
Paclitaxel and docetaxel conjugates bearing a ferrocenyl
substituent at the 7-O-position were synthesised in three
steps, starting from 1 and 2 (Scheme 5). In the first step, 1 and
2 were selectively protected at the 2’-O-position as tert-butyldi-
methylsilyl ethers with tert-butyldimethylsilyl chloride (TBSCl)
8
5% overall yield (Scheme 1). Introduction of the (3R,4S)-phe-
nylisoserine moiety at the 13-OH position of 4 required the
use of optically pure N-ferroceneazetidin-2-ones. (3R,4S)-N-Fer-
rocenoyl-4-phenyl-3-triethylsilyloxyazetidin-2-one (6) was pre-
pared in 52% yield in an N-acylation reaction of 5 with freshly
prepared ferrocenoyl chloride (prepared from ferrocenecarbox-
[
25]
in the presence of imidazole in DMF at RT.
The resulting 2’-O-TBS-paclitaxel 20 was then selectively 7-O-
acylated with 3-ferrocenoylpropionic acid or 4-ferrocenylbuty-
Chem. Eur. J. 2016, 22, 11413 – 11421
www.chemeurj.org
11414
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 3. Synthesis of N-debenzoyl-N-ferrocenoylpaclitaxel derivatives 10 and 11. Reagents and conditions: (i) 6 or 7, LiHMDS, THF, À408C, 40 min, up to
84% yield; (ii) HF·Py, pyridine/MeCN, RT, 24 h, up to 90% yield.
22 and 23, respectively, were isolated in good or excellent
yields by chromatography on silica column (Scheme 6). A simi-
lar reaction of 2’-O-TBS-docetaxel 21 with 5-ferrocenoylpenta-
noic and 6-ferrocenylhexanoic acids as acylating agents carried
out at 08C gave the desired 7-O-acylated-2’-O-TBS-docetaxel
derivatives 24 and 25, respectively, as major products. In this
reaction, small amounts of 10-O-isomers were also formed (less
than 5%). The deprotection of hydroxy groups in 22–25, car-
ried out using HF·Py, gave the desired 7-O-ferrocenyl-substitut-
ed taxanes 26–29, respectively, in good to excellent yields
(
Scheme 7).
Scheme 4. Synthesis of 2’-O-ferrocene-substituted paclitaxel 12–15 and do-
3
cetaxel 16–19 derivatives. Reagents and conditions: (i) FcR COOH, DIC
Cytotoxic activity
(
1.5 equiv), DMAP (0.1 equiv), DCM, RT, 24 h.
SW620 cells originating from human colon adenocarcinoma
were chosen to screen the synthesised ferrocenyl taxanes for
their in vitro antiproliferative/cytotoxic activity in comparison
to 1 and 2. The SW620 cell line is characterised by a relatively
high sensitivity towards various chemotherapeutics as estab-
[
26]
lished by cytotoxicity assays,
thus being well-suited for
studying the activity of novel compounds of an uncertain
mechanism of action. To investigate whether ferrocenyl tax-
anes are able to overcome multidrug resistance (MDR) result-
ing from overexpression of various ABC proteins, we further
employed a panel of five SW620-derived drug resistant cancer
cell lines (SW620C, D, E, M and V) obtained by a stepwise se-
lection with the classical anticancer agents cisplatin, doxorubi-
Scheme 5. Synthesis of 2’-O-TBS ethers of paclitaxel 20 and docetaxel 21.
Reagents and conditions: (i) TBSCl, imidazole, RT, 24 h.
[
20,21]
cin, etoposide, methotrexate and vincristine, respectively.
ric acid using DIC as a coupling agent at 08C, whereas at-
tempts to use ferrocenecarboxylic acid as an acylation agent
under various conditions failed. The corresponding products
The MDR cell lines were fully characterised with regards to
drug cross-resistance, ABC transporter protein expression and
the subcellular localisation and activity of the drugs. These cell
2
Scheme 6. Synthesis of paclitaxel derivatives 26 and 27 substituted in the 7-position with a ferrocene moiety. Reagents and conditions: (i) FcR COOH, DIC
(
1.5 equiv), DMAP (0.1 equiv), DCM, RT, 24 h; (ii) HF·Py/pyridine/MeCN, RT, 24 h.
Chem. Eur. J. 2016, 22, 11413 – 11421 www.chemeurj.org
11415
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
2
Scheme 7. Synthesis of docetaxel derivatives 28 and 29 substituted in the 7-position with ferrocene. Reagents and conditions: (i) FcR COOH, DIC (1.5 equiv),
DMAP (0.1 equiv), DCM, RT, 24 h; (ii) HF·Py/pyridine/MeCN, RT, 24 h.
lines are ranked in the following order with regard to the ex-
pression of the ABC transporter ABCB1 (either at mRNA or pro-
and 28–29) exhibit approximately one order of magnitude
lower activity than the parent compound, except for 28, which
is only 2.7 times less active.
tein
level):
SW620V>SW620D>SW620E@SW620C>
SW620M=SW620. ABCB1 was recognised as a high contributor
When analysing the response of drug resistant cells to the
new compounds, it is clear that none of them are potent
enough to overcome the MDR barrier. All substances were
active against SW620M cells, which is not surprising, as they
do not overexpress ABCB1 compared to the parental cell line.
However, the IC50 values determined for this cell line were on
average four–five times higher than those for SW620 cells
(except for 10 and 28, for which they were 60 and 110 times
higher, respectively). It may suggest an impact of ABCC1,
which is a highly overexpressed ABC transporter in SW620M
[27]
to taxane resistance.
The IC50 values and corresponding
9
5%-confidence intervals for all of the synthesised compounds
are summarised in Table 1. All compounds were analysed at
a concentration range between 3 nm and 30 mm, whereas
higher concentrations are unlikely to be obtained in any bodily
fluid in vivo (peak paclitaxel serum concentration achievable
after parenteral administration of a maximum allowable dose
[
28]
is only 4.5 mm).
All compounds demonstrated antiproliferative/cytotoxic
action against the parental cell line SW620 in the micromolar
or even submicromolar concentration range. It is noteworthy
that the simplest ferrocenyl analogue of taxol (10) obtained by
replacing the N-benzoyl group with a ferrocenoyl moiety, ex-
erted increased antiproliferative/cytotoxic effects compared to
paclitaxel (IC50 value of 0.11 vs 1.11 mm for 10 and 1, respec-
tively). Insertion of a butyryl spacer between the ferrocenyl
moiety and the amine group, as in 11, results in a four-fold al-
leviation of the cytotoxic activity. The cytotoxicity of com-
pounds bearing a ferrocenyl moiety attached to the 2’-OH
group of paclitaxel (12–15) strongly depends on the linker
type. Compound 13 is as active as paclitaxel against SW620
cells (IC =0.84 mm), but its activity is lower than that of 10. All
[
29]
cells in conferring low-level resistance to the analysed tax-
anes. It should also be mentioned that the paclitaxel deriva-
tives 11 and 27 exhibited activity against SW620C cells (ex-
pressing ABCG2), whereas only 18 of the ferrocenyl derivatives
of the docetaxel series exerted a moderate cytotoxic activity
against SW620E (expressing both ABCB1 and ABCC1).
It is interesting to compare the viability of SW620 cells in
the presence of different concentrations of the synthesised
compounds. The viability of SW620 cells at 10 nm concentra-
tion of 10 is as low as 61% (Figure S1 in the Supporting Infor-
mation), whereas all other compounds are slightly, if at all,
active at this concentration. At 100 nm, the simplest ferrocenyl
analogue of paclitaxel 10 was still the most active derivative
but also 1, 2 and 28 showed some activity (41% viability com-
pared to 64, 60 and 63%, respectively). At 1 mm, only two of
the synthesised compounds (14 and 27, both from the pacli-
taxel series) did not significantly affect the viability of the
SW620 cells (84 and 102%, respectively).
50
other compounds of this series (12, 14 and 15) are markedly
two–ten times, taking into account the mean values) less
(
active than paclitaxel 1. Additionally, the presence of a ferro-
cenyl moiety attached to the 7-OH group, further decreases
the antiproliferative activity of the synthesised compounds
compared to 1. All synthesised derivatives of docetaxel (16–19
Chem. Eur. J. 2016, 22, 11413 – 11421
www.chemeurj.org
11416
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 1. Cytotoxicity of the ferrocenyl compounds 10–19 and 26–29 in
reference to paclitaxel 1 and docetaxel 2 as determined by the MTT-re-
duction viability assay (MTT=3-(4,5-dimethylthiazol-2-yl)-2,5-diphenylte-
Table 2. Influence of the synthesised compounds on the cell cycle distri-
bution of SW620 cells at 10 nm concentration.
[a]
[a]
trazolium bromide).
Cell cycle phase distribution [%]
Compound
sub-G0/G1
G0/G1
S
G2/M
IC50 [mm]
SW620C
[
b]
control
DMSO
1
1.6Æ0.1
1.6Æ0.5
21.8Æ5.9
41.0Æ12.3
5.1Æ1.4
42.7Æ3.5
42.3Æ5.5
33.6Æ3.5
17.7Æ7.4
41.9Æ5.3
41.7Æ4.0
32.3Æ9.8
48.4Æ4.5
42.6Æ2.6
43.2Æ5.9
43.8Æ6.8
5.0Æ4.0
47.7Æ2.6
47.9Æ3.8
40.1Æ2.4
31.5Æ3.2
44.9Æ4.4
46.9Æ1.3
41.7Æ4.0
42.3Æ1.2
45.9Æ2.3
46.7Æ5.3
45.6Æ5.8
19.2Æ3.8
44.2Æ0.8
28.1Æ6.8
40.1Æ5.7
42.3Æ5.8
45.8Æ4.6
43.9Æ3.0
8.0Æ0.9
8.0Æ2.4
7.5Æ5.4
13.5Æ2.6
8.5Æ2.3
8.7Æ1.1
10.2Æ0.2
9.7Æ1.4
8.8Æ2.8
7.1Æ0.8
8.6Æ0.7
36.8Æ3.2
9.9Æ4.2
23.4Æ8.0
10.3Æ5.0
10.4Æ4.6
9.8Æ2.2
9.1Æ1.9
Compound
SW620
1.11
SW620M
[
c]
1
N.D.
N.D.
6.72
[4.26–10.60]
6.65
[2.97–14.86]
7.83
[4.84–12.67]
21.32
[14.63–31.07]
3.36
[2.48–4.53]
29.55
[18.05–48.39]
13.09
[8.63–19.87]
19.74
[14.11–27.62]
22.47
[14.05–35.92]
1.49
[1.02–2.19]
6.73
[4.16–11.49]
7.16
[4.74–10.80]
21.18
[10.36–42.04]
26.48
[19.07–36.77]
91.38
[32.61–256.1]
23.05
[
0.46–1.49]
0.11
0.07–0.19]
5.18
3.38–7.94]
7.88
5.33–11.66]
0.84
0.69–1.02]
12.71
10.33–15.62]
2.53
2.00–3.21]
4.42
3.76–5.19]
6.91
5.80–8.23]
0.31
0.22–0.43]
1.47
1.11–1.95]
3.22
1.92–5.40]
2.30
1.87–3.56]
1.99
1.79–3.07]
0.83
0.60–1.15]
4.27
3.36–6.58]
10
11
12
13
14
15
26
27
2
16
17
18
19
28
29
1
0
[
2.6Æ1.3
18.1Æ14.7
2.7Æ1.9
3.2Æ0.8
3.0Æ0.3
2.2Æ0.8
38.3Æ4.5
11.3Æ2.0
33.1Æ23.1
7.2Æ6.0
7.8Æ8.1
7.8Æ5.5
4.0Æ4.5
11
45.26
[25.8–79.39]
N.D.
[
1
1
1
1
2
2
2
1
1
1
1
2
2
2
3
4
5
6
7
[
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
[
37.3Æ1.8
19.7Æ15.2
45.7Æ4.2
43.1Æ3.0
38.4Æ1.2
44.6Æ2.5
[
[
[
[
a] Averaged data ÆSD calculated from three independent experiments.
[
Cell cycle fractions were calculated by FlowJo software using Watson
pragmatic algorithm. [b] Cells in a complete medium. [c] Cells in a com-
plete medium supplemented with DMSO at the concentration used for
investigated compounds.
[
6
7
8
9
8
9
[
[
1
1 and 16 (5.1% and 11.3%, respectively). In general, under
the experimental conditions applied (48-hour exposure), pacli-
taxel 1 and its ferrocenyl analogues turned out to be potent
inducers of apoptosis rather than of mitotic arrest, with 10
being again the most effective. On the other hand, both doce-
taxel and 17 induced cell death, as manifested by an abundant
sub-G0/G1 fraction and G2/M blockage, which was inferred
from alterations of specific cell cycle phases (at least half re-
duction of G0/G1 and S phase cell number, and simultaneous
two–three fold increase in the G2/M phase fraction).
[
[
[
[
[15.39–33.18]
[
a] 95%-confidence intervals are given in brackets (please note that due
to the log-transformation of the data required to perform IC50 calcula-
tions, these are asymmetrical). Calculations are based on results of three
independent experiments. N.D. denotes situations in which the character
of the viability data, due to low cytotoxicity, did not allow for proper
curve fitting and IC50 calculation. This also applies to data for cell lines
SW620D, SW620E, and SW620V
Tubulin polymerisation assay
The basis for microtubule stabilisation is a direct interaction of
taxane and b-tubulin molecules, thus leading to microtubule
stabilisation and prevention of their rearrangement. To check
whether ferrocenyl derivatives are able to induce tubulin poly-
merisation, a fluorescence real-time assay was applied. As
shown in Figure 2, the paclitaxel analogues 10 and 11 promote
tubulin polymerisation much more efficiently than their parent
compound 1. Compound 26 induces tubulin polymerisation to
a lesser extent than paclitaxel but still faster than DMSO, which
was used as the solvent control. All other ferrocenyl derivatives
of paclitaxel, that is, 12–15 and 27, tend to decrease the tubu-
lin polymerisation rate. When analysing the docetaxel series
(Figure S2 in the Supporting Information), all derivatives are
much weaker inducers of tubulin polymerisation than 2. The
most active of this series is 28, which turned out to be as ef-
fective as paclitaxel. On the other hand, 18 was the only doce-
taxel analogue, which seemed to be an inhibitor of tubulin
polymerisation.
Cell cycle distribution studies
The taxane-induced inability to rearrange microtubules pre-
vents mitosis and results in cell death executed mainly through
the mitotic catastrophe. Thus, taxane treatment significantly
alters the cell cycle and leads to accumulation of G2/M cells
(
due to mitotic arrest) and increase in sub-G0/G1 fraction indi-
cative of DNA fragmentation. Results of the SW620 cell cycle
analysis following 48-hour incubation with the ferrocenyl
taxane analogues at 10 nm are presented in Table 2.
Surprisingly, significant mitotic arrest, as indicated by elevat-
ed G2/M levels, was only found for 10, 13 and 17. It was ac-
companied by a high number of cells undergoing cell death,
as indicated by elevation of sub-G0/G1 fraction (from approx.
1
.5% in control samples to over 40% in the case of 10). The
latter effect, although to a lesser extent, was also observed for
Chem. Eur. J. 2016, 22, 11413 – 11421
www.chemeurj.org
11417
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
most active ferrocenyl analogue 10 were conducted and com-
[
30]
pared to 1 and 2. Gold score (GS) was the only scoring func-
tion able to treat the metal complexes. GS gives arbitrary num-
bers with higher values predicting better binding.
Docking of paclitaxel 1 resulted in two predicted hydrogen-
bonding interactions (Figure S4a in the Supporting Informa-
tion) with Gly370 and Thr276 and the compound gave a GS of
6
8. The binding site has hydrophobic pockets that the 3’ and
the benzoyl amide phenyl rings of the ligand are predicted to
occupy (Figure S4b). In proximity to His229, the 3’ and the
benzoyl amide phenyl rings fill the hydrophobic pockets,
whereas the tetracyclic fragment sits partially above the sur-
face of the protein.
When docetaxel 2 was docked, it did not fully overlap with
the conformation of paclitaxel (GS=67). A hydrogen bonding
interaction with Thr276 is predicted (Figure S5a in the Sup-
porting Information), which was also seen for 1. Similar pre-
dicted poses of the compounds resulted in comparable protein
ligand hydrophobic interactions with the phenyl group in the
Figure 2. Tubulin polymerisation induced by ferrocenyl analogues of pacli-
taxel at 1 mm concentration. Results of a representative study from of
a series of three independent experiments is presented.
3
’ position directed into the pocket (Figure S5b).
The influence of the synthesised compounds on tubulin
polymerisation and microtubule stabilisation was also investi-
gated for SW620 cells using confocal microscopy. The results
obtained for the most active compounds 10, 11 and 13 com-
pared to paclitaxel and vincristine as positive and negative
controls, respectively, are summarised in Figure 3. Paclitaxel in-
duces formation of microtubule bundles, which can be easily
recognised in Figure 3b, whereas vincristine prevents tubulin
dimer polymerisation, which results in uniform distribution of
tubulin within the cytoplasm (Figure 3c). Cell treatment with
both 10 and 11 significantly intensifies tubulin staining, visual-
ised as a dense microtubule network (Figures 3d and 3e), al-
though paclitaxel-specific microtubule bundles are seen only
in the case of 13 (Figure 3 f).
The ferrocenyl analogue of paclitaxel 10 has a good fit in
the binding pocket of 1. With a GS of 69, the scoring of the
compound was very similar to 1 and 2. Despite the structural
similarity to 1, a different hydrogen bonding pattern was ob-
served, that is, a hydrogen bond with His229 (see Figure 4 and
Docking studies with tubulin
To explain the findings from cytotoxicity assays and tubulin
polymerisation assays and to estimate the likelihood of binding
to tubulin, docking studies of tubulin (PDB: 1JFF) with the
Figure 4. The docked configuration of 10 in the binding site of tubulin.
Figure S6a in the Supporting Information). The hydrophobic
interactions are similar to paclitaxel. The 2-benzoyl side-chain
phenyl and ferrocenyl moieties (Figure S6b) occupy the hydro-
phobic pockets. The similarity in docking between compound
10, paclitaxel and docetaxel suggests a plausible binding to tu-
bulin. Also, the hydrophobic interaction between the ferrocen-
yl moiety and the binding pocket supports this assumption.
However, note that due to a small data set, it is difficult to
draw ultimate conclusions.
Conclusions
Figure 3. Influence of the synthesised compounds on microtubule formation
in SW620 cells: a) untreated cells, b) paclitaxel 1, c) vincristine, d) 10, e) 11,
f) 13; left panel: nuclear staining with Hoechst 33342; middle panel: anti b-
tubulin antibody staining; right panel: superposition of images from the left
and middle panels.
A series of ferrocenyl derivatives of paclitaxel and docetaxel
were synthesised and evaluated for both inducing tubulin
polymerisation and their antiproliferative activity. The most
potent compound, the simplest ferrocenyl analogue of pacli-
Chem. Eur. J. 2016, 22, 11413 – 11421
www.chemeurj.org
11418
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
taxel 10, was demonstrated to be more active toward SW620
cells and to induce tubulin polymerisation more efficiently
than paclitaxel 1. Compound 10 was able to induce apoptosis
and arrest the cell cycle in the G2/M phase at least as effective
as 1. The high activity of 10 may be explained by its augment-
ed interactions with b-tubulin as suggested by docking stud-
ies. Lower cytotoxic activity of 11 than that of 1, together with
its higher ability to induce polymerisation of tubulin to micro-
tubules may suggest that the mechanism of action of 11 is dif-
ferent from that of paclitaxel and 10. The tubulin polymeri-
sation induced by this compound in a cellular system seems to
be insufficient to exert its antiproliferative activity in a living
cell. On the other hand, 13, 15 and 16 appear to utilise mecha-
nisms other than microtubule stabilisation for their cytotoxic/
cytostatic activity.
(300 mL of silica gel, n-hexane/ethyl acetate 4:1) gave pure 6 as
1
a dark red oil (430 mg, 52%). H NMR (600 MHz, CDCl ): d=7.34–
3
7
6
.41 (m, 4H), 7.29–7.33 (m, 1H), 5.44–5.47 (m, 1H), 5.29 (d, J=
.4 Hz, 1H), 5.17–5.20 (m, 1H), 5.08 (d, J=6.0 Hz, 1H), 4.50–4.53
(
6
m, 2H), 4.15 (s, 5H), 0.82 (t, J=7.9 Hz, 9H), 0.43–0.56 ppm (m,
13
1
H); C{ H} NMR (151 MHz, CDCl ): d=169.4, 164.5, 134.4, 128.4,
3
1
4
28.3, 128.1, 75.2, 72.3, 72.3, 72.1, 71.8, 70.7, 70.0, 60.8, 6.3,
.5 ppm; HRMS (EI) calculated for C H FeNO Si 489.14176, found
26
31
3
489.14218.
(3R,4S)-1-(4-Ferrocenylbutyryl)-4-phenyl-3-((triethylsilyl)oxy)aze-
tidin-2-one (7): DIC (91 mg, 112 mL, 0.721 mmol) was added to a so-
lution of (3R,4S)-4-phenyl-3-((triethylsilyl)oxy)azetidin-2-one (5,
100 mg,
0.378 mmol), DMAP (2.2 mg, 0.018 mmol) in DCM (5 mL). After
4 h of stirring, the solvent was evaporated and pure 7 (first frac-
0.360 mmol),
4-ferrocenylbutyric
acid
(103 mg,
2
tion) was isolated by chromatography on silica gel (50 mL, n-
hexane/ethyl acetate 3:2) as an orange solid (111 mg, 56.3%).
Ferrocenyl analogues of taxanes are an interesting class of
organometallic compounds. The bioactivity of the synthesised
compounds described here demonstrates that the ferrocenyl
moiety has a positive impact on the activity of a taxane as
both an activator of tubulin polymerisation and an antiprolifer-
ative agent. The preliminary data on these organometallic
compounds, which are able to induce or inhibit polymerisation
of tubulin and act as anticancer drug candidates, warrants fur-
ther investigation. Additionally, the exact mode of action of
the most active ferrocenyl taxanes is currently unknown and
requires further studies. Other ferrocene derivatives, for exam-
1
H NMR (600 MHz, CDCl ): d=7.35–7.40 (m, 2H), 7.31–7.35 (m, 1H),
3
7
4
.25–7.30 (m, 2H), 5.16 (d, J=5.9 Hz, 1H), 5.13 (d, J=5.9 Hz, 1H),
.13 (s, 5H), 4.10 (s, 2H), 4.08 (s, 2H), 2.84–2.90 (m, 1H), 2.77–2.83
(
m, 1H), 2.42 (t, J=7.8 Hz, 2H), 1.86–1.96 (m, 2H), 0.82 (t, J=
13
1
8.0 Hz, 9H), 0.42–0.55 ppm (m, 6H); C{ H} NMR (151 MHz, CDCl ):
3
d=171.0, 166.6, 133.5, 128.3, 128.1, 127.9, 88.1, 77.0, 68.5, 68.2,
68.1, 67.3, 67.2, 61.0, 36.7, 28.9, 25.4, 6.2, 4.4 ppm; HRMS (EI) calcu-
lated for C H FeNO Si 531.18872, found 531.18899.
29 37
3
2’,7-O-Bis(trietylsilyl)-N-debenzoyl-N-ferrocenoylpaclitaxel (8): 7-
O-TriethylsilylbaccatinIII (4, 352 mg, 0.503 mmol) and (3R,4S)-1-fer-
rocenoyl-4-phenyl-3-((triethylsilyl)oxy)azetidin-2-one (6, 430 mg,
0.879 mmol) in anhydrous THF (20 mL) were placed in a Schlenk
tube. LiHMDS (1.0m in THF, 0.820 mL, 0.820 mmol) was added at
À408C and the resulting solution was stirred at this temperature
for 40 min. The reaction was quenched by addition of 40 mL of sa-
turated ammonium chloride and the product was extracted with
ethyl acetate. The organic solution was washed with water, brine
and the solvents were evaporated. The pure product was isolated
[
31]
[32]
ple, ferrocifen, ferroquine and ferrocenyl-based antifungal
[
33]
agents
are thought to act by producing reactive species
through a Fenton-type reaction. Thus, the toxicity of ferrocene
conjugates can be explained by the fact that an organic ligand
acts as a trafficking agent to place ferrocene in the vicinity of
a particular cellular target, where oxidation of the iron ion and
production of hydroxyl and superoxide radicals induces
damage. Further studies to explain the mechanism of cytotox-
icity of ferrocenyl taxanes combining the measurement of
redox potential and intracellular generation of reactive oxygen
species (ROS) production, especially in comparison to rutheno-
cenyl derivatives, are planned and will be performed in the
near future.
by chromatography on silica gel (70 mL, n-hexane/ethyl acetate
1
2:1) as a yellow solid (445 mg, 74.5%). H NMR (600 MHz, CDCl
):
3
d=8.16 (d, J=7.6 Hz, 2H), 7.63 (t, J=7.9 Hz, 1H), 7.55 (t, J=7.7 Hz,
2
1
5
1
1
7
1
1
3
H), 7.37–7.42 (m, 3H), 7.32 (t, J=7.3 Hz, 1H), 6.70 (d, J=8.9 Hz,
H), 6.48 (s, 1H), 6.31 (t, J=8.8 Hz, 1H), 5.73 (d, J=7.1 Hz, 1H),
.63 (d, J=8.6 Hz, 1H), 4.96 (dd, J=1.6, 7.9 Hz, 1H), 4.70 (d, J=
.8 Hz, 1H), 4.66 (brs, 1H), 4.61 (brs, 1H), 4.50 (dd, J=6.7, 10.6 Hz,
H), 4.33 (brs, 3H), 4.23 (d, J=8.4 Hz, 1H), 4.18 (s, 5H), 3.86 (d, J=
.0 Hz, 1H), 2.53–2.57 (m, 1H), 2.52 (s, 3H), 2.44 (dd, J=9.6,
5.2 Hz, 1H), 2.20–2.26 (m, 1H), 2.18 (s, 3H), 2.04 (d, J=0.9 Hz, 3H),
.92 (quint, J=2.1, 10.7 Hz, 1H), 1.87 (brs, 1H), 1.72 (s, 3H), 1.23 (s,
H), 1.20 (s, 3H), 0.94 (t, J=7.9 Hz, 9H), 0.88 (t, J=7.9 Hz, 9H),
Experimental Section
(
3R,4S)-1-Ferrocenoyl-4-phenyl-3-((triethylsilyl)oxy)azetidin-2-
0.57–0.65 (m, 6H), 0.51–0.57 (m, 3H), 0.44–0.51 ppm (m, 3H);
C{ H} NMR (151 MHz, CDCl ): d=201.7, 171.5, 170.0, 169.7, 169.3,
3
13
1
one (6): Oxalyl chloride (525 mg, 350 mL, 4.14 mmol) and 1 drop of
DMF as catalyst were added to a slurry of ferrocenecarboxylic acid
167.0, 140.2, 139.0, 133.8, 133.6 (CH), 130.3 (CH), 129.3, 128.8 (CH),
128.6 (CH), 127.9 (CH), 126.5 (CH), 84.2 (CH), 81.2, 78.8, 76.6 (CH2),
75.7, 75.0 (CH), 75.0 (CH), 74.8 (CH), 72.2 (CH), 71.4 (CH), 70.6 (CH),
70.4 (CH), 69.8 (CH), 68.5 (CH), 67.9 (CH), 58.4, 55.1 (CH), 46.7 (CH),
43.4, 37.3 (CH ), 35.7 (CH ), 26.6 (CH), 23.0 (CH), 21.5 (CH), 20.8
(
506 mg, 2.2 mmol) in 10 mL of anhydrous dichloromethane. The
resulting solution was stirred at RT for 1 h and the volatile materi-
als were removed by evaporation. The crude ferrocenoyl chloride
was dried prior to use for 30 min under vacuum (at 0.01 mbar).
2
2
(
3R,4S)-4-phenyl-3-((triethylsilyl)oxy)azetidin-2-one
(5,
470 mg,
(CH), 14.1 (CH), 10.1 (CH), 6.7 (CH), 6.5 (CH), 5.3 (CH ), 4.5 ppm
2
1
.694 mmol), DMAP (200 mg, 1.637 mmol) and anhydrous triethyla-
(CH ); MALDI calculated for C H FeNO Si 1189.47, found 1189.35.
2
63 83
14
2
mine (607 mg, 836 mL, 6.0 mmol) in anhydrous DCM (5 mL) were
placed in a Schlenk tube. A solution of freshly prepared ferrocenoyl
chloride in anhydrous DCM (5 mL) was added dropwise at 08C and
the resulting mixture was stirred at RT for 2 h. Then 50 mL of satu-
rated sodium bicarbonate was added and the product was extract-
ed with ethyl acetate. The organic phase was washed with brine,
dried and the solvent removed. Chromatography on silica gel
N-Debenzoyl-N-ferrocenoylpaclitaxel (10): A large excess of hy-
drogen fluoride·pyridine complex (3.7 mL) was added to a solution
of 8 (445 mg, 0.374 mmol) in a mixture of anhydrous acetonitrile
(15 mL) and anhydrous pyridine (30 mL) placed in a Teflon flask.
The solution was stirred at RT for 20 h, the reaction was then
quenched by addition of saturated sodium bicarbonate (400 mL)
and the product was extracted with ethyl acetate. The organic
Chem. Eur. J. 2016, 22, 11413 – 11421
www.chemeurj.org
11419
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
phase was washed with water, brine and the solvents were evapo-
rated. The pure product (323 mg, 90%) was isolated by chroma-
(CH), 20.9 (CH), 14.2 (CH), 10.0 ppm (CH); MALDI calculated for
C H FeNO 1061.39, found 1061.40.
57
67
15
tography on silica gel (300 mL, dichloromethane/methanol 97:3).
2
’-O-(tert-Butyldimethylsilyl)-7-O-(3-ferrocenoylpropioyl)paclitax-
el (22): DIC (21 mg, 26 mL, 0.165 mmol) was added to a solution of
0 (100 mg, 0.103 mmol), 3-ferrocenoylpropionic acid (44 mg,
.155 mmol) and DMAP (5 mg, 0.041 mmol) in anhydrous dichloro-
1
H NMR (600 MHz, CDCl ): d=8.15 (dd, J=1.3, 7.2 Hz, 2H), 7.66–
3
7
.63 (m, 1H), 7.54 (t, J=7.8 Hz, 2H), 7.50 (d, J=7.4 Hz, 2H), 7.44 (t,
2
0
J=7.7 Hz, 2H), 7.36 (t, J=7.4 Hz, 1H), 6.48 (d, J=8.9 Hz, 1H), 6.30
(
s, 1H), 6.28 (d, J=9.0 Hz, 1H), 5.72 (dd, J=2.5, 8.9 Hz, 1H), 5.70 (d,
methane (1 mL). The resulting solution was stirred at RT for 5 h
and the solvent evaporated. The pure product was isolated by
J=7.0 Hz, 1H), 4.96 (dd, J=2.0, 9.6 Hz, 1H), 4.78 (dd, J=2.7,
5
1
8
1
.1 Hz, 1H), 4.65–4.64 (m, 1H), 4.64–4.63 (m, 1H), 4.44–4.40 (m,
H), 4.35 (t, J=1.6 Hz, 2H), 4.31 (d, J=8.5 Hz, 1H), 4.22 (d, J=
.3 Hz, 1H), 4.16 (s, 5H), 3.82 (d, J=7.0 Hz, 1H), 3.62 (d, J=5.2 Hz,
H), 2.56 (ddd, J=5.3, 6.6, 6.8 Hz, 1H), 2.43 (d, J=4.1 Hz, 1H), 2.39
chromatography on silica gel (100 mL, n-hexane/ethyl acetate 2:1)
1
as an orange solid (105 mg, 82%). H NMR (600 MHz, CDCl ): d=
3
8
.12–8.17 (m, 2H), 7.73–7.78 (m, 2H), 7.58–7.65 (m, 1H), 7.52–7.56
(
(
m, 2H), 7.50 (td, J=1.5, 7.5 Hz, 1H), 7.41–7.45 (m, 2H), 7.37–7.41
m, 2H), 7.29–7.36 (m, 3H), 7.08 (d, J=8.9 Hz, 1H), 6.31 (s, 1H),
(
s, 3H), 2.35 (dd, J=9.2, 12.2 Hz, 1H), 2.25 (s, 3H), 1.92–1.84 (m,
1
3
1
H), 1.84 (d, J=1.1 Hz, 1H), 1.70 (s, 3H), 1.33–1.28 (m, 1H), 1.27 (s,
6.24–6.30 (m, 1H), 5.75 (dd, J=1.7, 8.9 Hz, 1H), 5.72 (d, J=7.0 Hz,
1H), 5.67 (dd, J=7.1, 10.6 Hz, 1H), 5.00 (d, J=8.1 Hz, 1H), 4.82–
1
3
1
H), 1.16 ppm (s, 3H); C{ H} NMR (151 MHz, CDCl ): d=203.6,
3
72.7, 171.2, 170.3, 167.0, 142.0, 138.5, 133.7 (CH), 133.2, 130.2
4
1
.87 (m, 2H), 4.69 (d, J=2.1 Hz, 1H), 4.50 (t, J=1.9 Hz, 2H), 4.34 (s,
H), 4.25 (s, 5H), 4.23 (d, J=8.7 Hz, 1H), 3.99 (d, J=6.9 Hz, 1H),
(
8
CH), 129.2, 129.0 (CH), 128.8 (CH), 128.3 (CH), 127.1 (CH), 84.4 (CH),
1.2, 79.0, 77.2 (CH), 76.5 (CH ), 75.6 (CH), 75.0 (CH), 73.4 (CH), 72.3
2
3.15 (td, J=7.5, 17.7 Hz, 1H), 3.00–3.08 (m, 1H), 2.62–2.78 (m, 3H),
.58 (s, 3H), 2.44 (dd, J=9.5, 15.2 Hz, 1H), 2.20 (s, 3H), 2.14–2.21
(
3
CH), 72.2 (CH), 70.5 (CH), 69.0 (CH), 58.6, 54.6 (CH), 45.6 (CH), 43.2,
5.7 (CH ), 35.6 (CH ), 26.9 (CH), 22.7 (CH), 21.8 (CH), 20.8 (CH), 14.8
2
2
2
(m, 1H), 2.00 (d, J=1.2 Hz, 3H), 1.93 (ddd, J=2.0, 10.8, 14.5 Hz,
(
CH), 9.6 ppm (CH). MALDI calculated for C H FeNO 961.30,
51 55 14
1H), 1.85 (s, 3H), 1.71 (s, 1H), 1.23 (s, 3H), 1.19 (s, 3H), 0.81 (s, 9H),
found 961.39.
13
1
À0.03 (s, 3H), À0.29 ppm (s, 3H); C{ H} NMR (151 MHz, CDCl ):
3
d=202.5, 202.0, 172.2, 171.5, 169.8, 169.0, 167.0, 167.0, 140.9,
2
’-O-(4-Ferrocenylbutyryl)paclitaxel (14): This compound was syn-
[20]
138.3, 134.2, 133.7 (CH), 132.7, 131.8 (CH), 130.2 (CH), 129.2, 128.8
thesised in 64% yield (71 mg) as described previously, starting
from paclitaxel (1, 85 mg, 0.100 mmol), 4-ferrocenylbutyric acid
(
7
CH), 128.7 (CH), 128.0 (CH), 127.0 (CH), 126.4 (CH), 84.1 (CH), 81.1,
8.7, 78.7, 76.4 (CH ), 75.4 (CH), 75.1 (CH), 74.6 (CH), 72.1 (CH), 72.1
2
(
30 mg 0.110 mmol), DMAP (13 mg, 0.106 mmol) and DIC (25 mg,
1
(CH), 71.3 (CH), 69.9 (CH), 69.3 (CH), 69.3 (CH), 56.2, 55.7 (CH), 47.0
3
7
3
1
5
1 mL, 0.200 mmol). H NMR (CDCl3, 600 MHz): d=8.15 (d, J=
.5 Hz, 2H), 7.73 (brs, 2H), 7.58–7.65 (m, 1H), 7.53 (t, J=7.6 Hz,
H), 7.42 (brs, 4H), 7.36 (d, J=6.5 Hz, 3H), 6.80 (brs, 1H), 6.32 (s,
H), 6.27 (t, J=8.8 Hz, 1H), 5.98 (brs, 1H), 5.70 (d, J=7.1 Hz, 1H),
.51 (brs, 1H), 4.99 (d, J=8.4 Hz, 1H), 4.43–4.50 (m, 1H), 4.33 (dd,
(
CH), 43.4, 35.6 (CH ), 34.3 (CH ), 33.4 (CH ), 28.3 (CH ), 26.4 (CH),
2 2 2 2
2
5.5 (CH), 23.0 (CH), 21.4 (CH), 20.8 (CH), 18.1, 14.6 (CH), 14.2, 10.9
(
1
CH), À5.2 (CH), À5.8 ppm; MALDI calculated for C H FeNO Si
67
77
16
235.44, found 1235.35.
J=1.0, 8.4 Hz, 1H), 4.53–4.11 (bs, 9H), 4.22 (d, J=8.4 Hz, 1H), 3.84
d, J=7.0 Hz, 1H), 2.54–2.62 (m, 1H), 2.42–2.52 (m, 5H), 2.38 (dd,
J=9.3, 15.3 Hz, 1H), 2.23 (s, 3H), 2.10–2.20 (m, 1H), 1.96 (s, 3H),
(
Acknowledgements
1
3
1
1
.86–1.93 (m, 1H), 1.75 (s, 2H), 1.70 (s, 3H), 1.61 (s, 3H), 1.25 (s,
1
3
1
This work was financially supported by the National Science
Centre Poland (NCN) based on the decision DEC-2012/05/B/
ST5/00300.
H), 1.15 ppm (s, 3H); C{ H} NMR (151 MHz, CDCl ): d=203.8,
3
71.2, 169.8, 168.1, 167.1, 167.0, 142.8, 137.1, 133.8, 133.7 (CH),
32.8, 132.0 (CH), 130.2 (CH), 129.3, 129.1 (CH), 128.8 (CH), 128.8
(
(
3
3
2
CH), 128.4 (CH), 127.2 (CH), 126.7 (CH), 84.5 (CH), 81.1, 79.3, 76.5
CH ), 75.6 (CH), 75.2 (CH), 74.0 (CH), 72.1 (CH), 71.8 (CH), 69.7 (bs,
2
Keywords: bioorganometallic chemistry · cancer · medicinal
chemistry · microtubule · tubulin polymerisation
”CH, 2Cp), 58.6, 52.8 (CH), 45.6 (CH), 43.2, 35.6 (CH ), 35.6 (CH ),
2 2
3.8 (CH ), 30.5 (CH ), 29.4, 28.9 (CH ), 26.9 (CH), 24.7, 22.8 (CH),
2
2
2
2.1 (CH), 20.8 (CH), 14.8 (CH), 9.6 ppm (CH); MALDI calculated for
[1] M. C. Wani, H. L. Taylor, M. E. Wall, P. Coggon, A. T. McPhail, J. Am. Chem.
C H FeNO 1107.37, found 1107.45.
61
65
15
Soc. 1971, 93, 2325–2327.
2
’-O-(4-Ferrocenylbutyryl)docetaxel (18): This compound was syn-
[2] a) P. B. Schiff, J. Fant, S. B. Horwitz, Nature 1979, 277, 665–667; b) M. A.
Jordan, Curr. Med. Chem. Anti-Cancer Agents 2002, 2, 1–17.
[3] M. L. Miller, I. Ojima, Chem. Rec. 2001, 1, 195–211.
thesised in 44% yield (47 mg) by the method described previously,
[
20]
starting from docetaxel (2, 81 mg, 0.100 mmol), 4-ferrocenylbu-
tyric acid (30 mg, 0.110 mmol), DMAP (13 mg, 0.106 mmol) and DIC
[4] M. Rooseboom, J. N. Commandeur, N. P. Vermeulen, Pharmacol. Rev.
1
2004, 56, 53–102.
(
25 mg, 31 mL, 0.200 mmol). H NMR (600 MHz, CDCl ): d=8.13 (d,
3
[
[
5] E. K. Rowinsky, Annu. Rev. Med. 1997, 48, 353–374.
6] a) I. Ojima, J. C. Slater, E. Michaud, S. D. Kuduk, P.-Y. Bounaud, P.
Vrignaud, M.-C. Bissery, J. M. Veith, P. Pera, R. J. Bernacki, J. Med. Chem.
J=7.5 Hz, 2H), 7.62 (t, J=7.3 Hz, 1H), 7.51 (t, J=7.7 Hz, 2H), 7.37–
7
1
.43 (m, 2H), 7.27–7.36 (m, 3H), 6.26 (brs, 1H), 5.70 (d, J=7.2 Hz,
H), 5.48 (brs, 1H), 5.39 (brs, 1H), 5.34 (m., 1H), 5.22 (s, 1H), 4.98
1996, 39, 3889–3896; b) S. M. Ali, M. Z. Hoemann, J. AubØ, G. I. Georg,
(
d, J=9.0 Hz, 1H), 4.33 (d, J=8.3 Hz, 1H), 4.27 (brs, 9H), 4.21 (d,
L. A. Mitscher, L. R. Jayasinghe, J. Med. Chem. 1997, 40, 236–241;
c) D. G. I. Kingston, A. G. Chaudhary, M. D. Chordia, M. Gharpure, A. A. L.
Gunatilaka, P. I. Higgs, J. M. Rimoldi, L. Samala, P. G. Jagtap, P. Giannaka-
kou, Y. Q. Jiang, C. M. Lin, E. Hamel, B. H. Long, C. R. Fairchild, K. A. John-
ston, J. Med. Chem. 1998, 41, 3715–3726; d) T. J. Altstadt, C. R. Fairchild,
J. Golik, K. A. Johnston, J. F. Kadow, F. Y. Lee, B. H. Long, W. C. Rose,
D. M. Vyas, H. Wong, M.-J. Wu, M. D. Wittman, J. Med. Chem. 2001, 44,
J=8.7 Hz, 1H), 4.18 (s, 2H), 3.95 (d, J=7.2 Hz, 1H), 2.56–2.64 (m,
1
1
1
(
1
(
7
(
3
H), 2.37–2.48 (m, 4H) 2.34 (brs, 1H), 2.18 (brs, 2H), 1.97 (s, 3H),
.86 (t, J=12.2 Hz, 1H), 1.77 (s, 3H), 1.68–1.74 (m, 2H), 1.66 (s, 1H),
1
3
1
.35 (s, 9H), 1.23–1.27 (m, 5H), 1.11–1.17 ppm (m, 3H); C{ H} NMR
151 MHz, CDCl ): d=211.6, 172.5, 169.7, 168.1, 167.1, 139.2, 137.6,
3
35.6, 133.6 (CH), 130. (CH), 129.3, 128.9 (CH), 128.7 (CH), 128.1
CH), 126.3 (CH), 84.2 (CH), 81.0, 80.4, 77.2 (CH), 79.0, 76.6 (CH2),
5.1 (CH), 74.5 (CH), 74.2 (CH), 71.9 (CH), 71.9 (CH), 69.6 (CH), 69.0
CH), 68.3 (CH), 57.6, 54.1, 46.4 (CH), 43.1, 37.0 (CH ), 35.6 (CH ),
4577–4583; e) I. Ojima, C. P. Borella, X. Wu, P.-Y. Bounaud, C. F. Oderda,
M. Sturm, M. L. Miller, S. Chakravarty, J. Chen, Q. Huang, P. Pera, T. A.
Brooks, M. R. Baer, R. J. Bernacki, J. Med. Chem. 2005, 48, 2218–2228;
f) I. Ojima, J. Chen, L. Sun, C. P. Borella, T. Wang, M. L. Miller, S. Lin, X.
Geng, L. Kuznetsova, C. Qu, D. Gallager, X. Zhao, I. Zanardi, S. Xia, S. B.
2
2
3.2 (CH ), 29.3, 28.6 (CH ), 28.2 (CH), 26.4 (CH), 25.8 (CH ), 22.6
2 2 2
Chem. Eur. J. 2016, 22, 11413 – 11421
www.chemeurj.org
11420
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Horwitz, J. Mallen-St. Clair, J. L. Guerriero, D. Bar-Sagi, J. M. Veith, P.
Pera, R. J. Bernacki, J. Med. Chem. 2008, 51, 3203–3221; g) L. V. Kuznet-
sova, A. Pepe, I. M. Ungureanu, P. Pera, R. J. Bernacki, I. Ojima, J. Fluorine
Chem. 2008, 129, 817–828; h) M. Hidaka, T. Koga, H. Kiyota, T. Horigu-
chi, Q.-W. Shi, K. Hirose, T. Uchida, Biosci., Biotechnol., Biochem. 2012,
[17] a) C. Biot, W. Daher, C. M. Ndiaye, P. Melnyk, B. Pradines, N. Chavain, A.
Pellet, L. Fraisse, L. Pelinski, C. Jarry, J. Brocard, J. Khalife, I. Forfar-Bares,
D. Dive, J. Med. Chem. 2006, 49, 4707–4714; b) F. Dubar, G. Anquetin, B.
Pradines, D. Dive, J. Khalife, C. Biot, J. Med. Chem. 2009, 52, 7954–7957.
[18] a) M. Patra, K. Ingram, V. Pierroz, S. Ferrari, B. Spingler, J. Keiser, G.
Gasser, J. Med. Chem. 2012, 55, 8790–8798; b) E. M. Lewandowski, J.
Skiba, N. J. Torelli, A. Rajnisz, J. Solecka, K. Kowalski, Y. Chen, Chem.
Commun. 2015, 51, 6186–6189.
7
6, 349–352; i) A. R. Wohl, A. R. Michel, S. Kalscheuer, C. W. Macosko, J.
Panyam, T. R. Hoye, J. Med. Chem. 2014, 57, 2368–2379; j) R. Matesanz,
C. Trigili, J. Rodríguez-Salarichs, I. Zanardi, B. Pera, A. Nogales, W. S.
Fang, J. Jímenez-Barbero, . Canales, I. Barasoain, I. Ojima, J. F. Díaz,
Bioorg. Med. Chem. 2014, 22, 5078–5090.
[19] A. Arezki, G. G. Chabot, L. Quentin, D. Scherman, G. Jaouen, E. BrulØ,
MedChemComm 2011, 2, 190–195.
[
7] a) E. J. Roh, C. E. Song, D. Kim, H. O. Pae, H. T. Chung, K. S. Lee, K. B.
Chai, C. O. Lee, S. U. Choi, Bioorg. Med. Chem. 1999, 7, 2115–2119;
b) E. J. Roh, D. Kim, C. O. Lee, S. U. Choi, C. E. Song, Bioorg. Med. Chem.
[20] D. Pla z˙ uk, A. Wieczorek, A. Błau z˙ , B. Rychlik, MedChemComm 2012, 3,
498–501.
[21] A. Wieczorek, A. Błau z˙ , J. Zakrzewski, B. Rychlik, D. Pla z˙ uk, ACS Med.
Chem. Lett. 2016, 7, 612–617.
2002, 10, 3145–3151.
[
[
8] L. Kuznetsova, J. Chen, L. Sun, X. Wu, A. Pepe, J. M. Veith, P. Pera, R. J.
Bernacki, I. Ojima, Bioorg. Med. Chem. Lett. 2006, 16, 974–977.
9] O. N. Zefirova, E. V. Nurieva, A. N. Ryzhov, N. V. Zyk, N. S. Zefirov, Russ. J.
Org. Chem. 2005, 41, 315–351.
[22] a) G. I. Georg, G. C. B. Harriman, M. Hepperle, J. S. Clowers, D. G. Vander
Velde, R. H. Himes, J. Org. Chem. 1996, 61, 2664–2676; b) I. Ojima, S. D.
Kuduk, S. Chakravarty in Advances in Medicinal Chemistry, Vol. 4 (Eds.:
E. M. Bruce, B. R. Allen), Elsevier, 1999, 69–124; c) I. Ojima, S. Lin, T.
Inoue, M. L. Miller, C. P. Borella, X. Geng, J. J. Walsh, J. Am. Chem. Soc.
2000, 122, 5343–5353.
[
10] a) M. F. R. Fouda, M. M. Abd-Elzaher, R. A. Abdelsamaia, A. A. Labib, Appl.
Organomet. Chem. 2007, 21, 613–625; b) E. MelØndez, Inorg. Chim. Acta
2
012, 393, 36–52; c) S. S. Braga, A. M. S. Silva, Organometallics 2013, 32,
[23] I. Paterson, G. J. Naylor, T. Fujita, E. Guzman, A. E. Wright, Chem.
Commun. 2010, 46, 261–263.
5
626–5639; d) S. Moon, M. Hanif, M. Kubanik, H. Holtkamp, T. Sçhnel,
S. M. F. Jamieson, C. G. Hartinger, ChemPlusChem 2015, 80, 231–236.
11] a) C. Biot, F. Nosten, L. Fraisse, D. Ter-Minassian, J. Khalife, D. Dive, Para-
site 2011, 18, 207–214; b) W. A. Wani, E. Jameel, U. Baig, S. Mumtazud-
din, L. T. Hun, Eur. J. Med. Chem. 2015, 101, 534–551.
12] a) M. Wenzel, M. Patra, C. H. R. Senges, I. Ott, J. J. Stepanek, A. Pinto, P.
Prochnow, C. Vuong, S. Langklotz, N. Metzler-Nolte, J. E. Bandow, ACS
Chem. Biol. 2013, 8, 1442–1450; b) Y. Liu, H. Zhang, D. Yin, D. Chen, Res.
Chem. Intermed. 2015, 41, 3793—3801; c) S. Mishra, V. Tirkey, A. Ghosh,
H. R. Dash, S. Das, M. Shukla, S. Saha, S. M. Mobin, S. Chatterjee, J. Mol.
Struct. 2015, 1085, 162–172.
13] a) P. Meunier, I. Ouattara, B. Gautheron, J. Tirouflet, D. Camboli, J. Besan-
Åon, Eur. J. Med. Chem. 1991, 26, 351–362; b) A. A. Simenel, G. A. Doku-
chaeva, L. V. Snegur, A. N. Rodionov, M. M. Ilyin, S. I. Zykova, L. A. Ostrov-
skaya, N. V. Bluchterova, M. M. Fomina, V. A. Rikova, Appl. Organomet.
Chem. 2011, 25, 70–75; c) K. Kowalski, A. Koceva-Chyła, A. Pieni a˛ z˙ ek, J.
Bernasi n´ ska, J. Skiba, A. J. Rybarczyk-Pirek, Z. Jó z´ wiak, J. Organomet.
Chem. 2012, 700, 58–68.
[24] N. M. Saied, N. Mejri, R. E. Aissi, E. Benoist, M. Saidi, Eur. J. Med. Chem.
2015, 97, 280–288.
[
[
[25] a) N. F. Magri, D. G. I. Kingston, J. Nat. Prod. 1988, 51, 298–306; b) B. J.
Turunen, H. Ge, J. Oyetunji, K. E. Desino, V. Vasandani, S. Güthe, R. H.
Himes, K. L. Audus, A. Seelig, G. I. Georg, Bioorg. Med. Chem. Lett. 2008,
18, 5971–5974; c) P. Wender, E. A. Dubikovskaya, S. H. Thorne, C. H.
Contag, Leland Stanford Junior University, USA, 2009, WO2009099636
A1d) D. M. Nevarez, Y. A. Mengistu, I. N. Nawarathne, K. D. Walker, J. Am.
Chem. Soc. 2009, 131, 5994–6002.
[26] B. Rychlik, Ł. Pułaski, Ł. Piotrowski, P. Guzenda, E. Macierzy n´ ska, A. Grze-
lak, G. Bartosz, FEBS Special Meeting ATP-Binding Cassette (ABC) Proteins:
From Multidrug Resistance to Genetic Diseases, Innsbruck, Austria, March
4–10, 2006.
[27] P. Wils, V. Phung-Ba, A. Warnery, D. Lechardeur, S. Raeissi, I. J. Hidalgo,
D. Scherman, Biochem. Pharmacol. (Amsterdam, Neth.) 1994, 48, 1528–
1530.
[28] Taxol injection label on FDA homepage, reference ID: 2939751 (http://
www.accessdata.fda.gov/drugsatfda docs/label/2011/
[
[
[
14] B. Long, S. Liang, D. Xin, Y. Yang, J. Xiang, Eur. J. Med. Chem. 2009, 44,
2
572–2576.
020262s049Lbl.pdf).
[29] A. Błau z˙ , B. Rychlik unpublished results.
15] a) D. Pla z˙ uk, A. Vessi›res, E. A. Hillard, O. Buriez, E. LabbØ, P. Pigeon, M.-
A. Plamont, C. Amatore, J. Zakrzewski, G. Jaouen, J. Med. Chem. 2009,
[30] G. Jones, P. Willett, R. C. Glen, A. R. Leach, R. Taylor, J. Mol. Biol. 1997,
267, 727–748.
5
2, 4964–4967; b) E. Hillard, A. Vessi›res, G. Jaouen in Medicinal Organ-
ometallic Chemistry, Vol. 32 (Eds.: G. Jaouen, N. Metzler-Nolte), Springer,
Berlin Heidelberg, 2010, pp. 81–117.
16] a) S. Top, J. Tang, A. Vessieres, D. Carrez, C. Provot, G. Jaouen, Chem.
Commun. 1996, 955–956; b) S. Top, B. Dauer, J. Vaissermann, G. Jaouen,
J. Organomet. Chem. 1997, 541, 355–361; c) E. Hillard, A. Vessieres, F.
Le Bideau, D. Plazuk, D. Spera, M. Huche, G. Jaouen, ChemMedChem
[31] a) G. Jaouen, S. Top in Advances in Organometallic Chemistry and Cataly-
sis: The Silver/Gold Jubilee International Conference on Organometallic
Chemistry Celebratory Books (Ed.: A. J. L. Pombeiro), John Wiley & Sons,
Inc., Hoboken 2013, pp. 563–580; b) H. Z. S. Lee, O. Buriez, F. Chau, E.
LabbØ, R. Ganguly, C. Amatore, G. Jaouen, A. Vessi›res, W. K. Leong, S.
Top, Eur. J. Inorg. Chem. 2015, 4217–4226.
[
2
006, 1, 551–559; d) N. Chavain, C. Biot, Curr. Med. Chem. 2010, 17,
[32] J. Keiser, M. Vargas, R. Rubbiani, G. Gasser, C. Biot, Parasites Vectors
2014, 7, 424.
2
729–2745; e) G. Gasser, I. Ott, N. Metzler-Nolte, J. Med. Chem. 2011,
5
4, 3–25; f) D. Pla z˙ uk, S. Top, A. Vessi›res, M.-A. Plamont, M. HuchØ, J.
[33] R. Rubbiani, O. Blacque, G. Gasser, Dalton Trans. 2016, 45, 6619–6626.
Zakrzewski, A. Makal, K. Wo z´ niak, G. Jaouen, Dalton Trans. 2010, 39,
444–7450; g) D. Pla z˙ uk, B. Rychlik, A. Błau z˙ , S. Domagała, J. Organo-
7
met. Chem. 2012, 715, 102–112; h) G. Gasser, N. Metzler-Nolte, Curr.
Opin. Chem. Biol. 2012, 16, 84–91; i) A. Vessi›res, J. Organomet. Chem.
Received: April 18, 2016
2013, 734, 3–16.
Published online on July 4, 2016
Chem. Eur. J. 2016, 22, 11413 – 11421
www.chemeurj.org
11421
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim