Synthesis, cytotoxicity and cellular uptake studies of N3 functionalized Re(CO)3 thymidine complexes

Article abstract of DOI:10.1039/c0dt01452d

Nucleoside-derived drugs play an important role in the treatment of cancer. Here, we present the synthesis and characterization of an intriguing series of N3 conjugated Re(CO)₃ thymidine complexes. The complexes were characterized by NMR spectr

Full text of DOI:10.1039/c0dt01452d

View Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
Radiopharmaceuticals for Imaging and Therapy
Guest Editors Stephen Faulkner (Oxford University) and Nick Long (Imperial College)
Published in issue 23, 2011 of Dalton Transactions
Image reproduced with permission of Martin W. Brechbiel
Articles in the issue include:
PERSPECTIVES:
Multimodal radio- (PET/SPECT) and fluorescence imaging agents based on metallo-radioisotopes:
current applications and prospects for development of new agents
Flora L. Thorp-Greenwood and Michael P. Coogan
Dalton Trans., 2011, DOI: 10.1039/C0DT01398F
Radiometallated peptides for molecular imaging and targeted therapy
João D. G. Correia, António Paulo, Paula D. Raposinho and Isabel Santos
Dalton Trans., 2011, DOI: 10.1039/C0DT01599G
Towards translation of 212Pb as a clinical therapeutic; getting the lead in!
Kwon Yong and Martin W. Brechbiel
Dalton Trans., 2011, DOI: 10.1039/C0DT01387K
ARTICLES:
First dinuclear Re/Tc complex as a potential bimodal Optical/SPECT molecular imaging agent
Alexandre Boulay, Marine Artigau, Yvon Coulais, Claude Picard, Béatrice Mestre-Voegtlé and Eric
Benoist
Dalton Trans., 2011, DOI: 10.1039/C0DT01397H
Synthesis, cytotoxicity and cellular uptake studies of N3 functionalized Re(CO)3 thymidine
complexes
Mark D. Bartholomä, Anthony R. Vortherms, Shawn Hillier, John Joyal, John Babich, Robert P.
Doyle and Jon Zubieta
Dalton Trans., 2011, DOI: 10.1039/C0DT01452D
Visit the Dalton Transactions website for more cutting-edge inorganic, organometallic and
bioinorganic research
www.rsc.org/dalton

Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 6216
www.rsc.org/dalton
PAPER
Synthesis, cytotoxicity and cellular uptake studies of N3 functionalized
Re(CO)3 thymidine complexes†
Mark D. Bartholoma¨,‡a Anthony R. Vortherms,a Shawn Hillier,b John Joyal,b John Babich,b Robert P. Doyle*a
and Jon Zubieta*a
Received 22nd October 2010, Accepted 23rd December 2010
DOI: 10.1039/c0dt01452d
Nucleoside-derived drugs play an important role in the treatment of cancer. Here, we present the
synthesis and characterization of an intriguing series of N3 conjugated Re(CO)3 thymidine complexes.
The complexes were characterized by NMR spectroscopy and mass spectrometry and their cytotoxicity
was assessed against A549 cells. A similar dependence on the spacer length and the toxicity has been
found for N3 functionalized complexes as recently reported for their C5¢ counterparts. A remarkable
cytotoxic complex 22, carrying a dodecylene spacer at position N3 with a bis-quinoline metal chelate
moiety, with an IC50 value of 3.4 1.6 mM, has been identified. Addition of a 100-fold excess of
thymidine did not statistically reduce the observed cytotoxicity of all complexes. Cellular uptake studies
of complex 22 have been performed by fluorescent microscopy, showing that compound 22 was clearly
internalized into A549 cells. Temperature dependent uptake studies, blocking experiments with
thymidine, and endosomal co-localization suggest that uptake of 22 occurs via passive diffusion and
endocytosis.
anticancer agents are metabolized and either inhibit enzymes
Introduction
involved in DNA synthesis and/or are incorporated into DNA,
Nucleoside-derived drugs are an important class of anticancer
thus affecting its structural integrity. This latter effect results in
chain termination or deformation inducing apoptosis.8 Prominent
examples of clinically relevant anticancer agents are the purine
analogs fludarabine (2-CdA, 2-chlorodeoxyadenosine) used in the
treatment of lymphoproliferative malignancies and the cytidine
derivative cytarabine (ara-C, 1-b-D-arabinosilfuranosilcytosine)
for acute leukaemia. The pyrimidine analogs gemcitabine (dFdC,
2¢,2¢-difluorodeoxycytidine) and capecitabine are used for the
treatment of solid tumors.2,9 More recently, clofarabine (Cl–F-
araA, 2-chloro-2¢-fluoro-deoxy-9-b-arabinosyladenine) has been
used for relapsed and refractory pediatric acute lymphoblastic
leukemia and troxacitabine (OddC, (-)-2¢-deoxy-3¢-oxacitabine)
has demonstrated efficacy in both solid and hematological malig-
nancies in clinical trials.10,11
The incorporation of a pendant metal complex to nucleosides
has received considerable attention in the development of artificial
nucleobases, mostly as biological markers, but also as novel
pharmaceutical agents.12 In the latter regard, cobalt carbonyl com-
plexes have been investigated as potential anti-tumor agents.13,14
The cytotoxicity of other nucleoside metal complexes has also been
investigated against a range of cell lines.15,16 However, there have
been only limited investigations of the antiproliferative properties
of rhenium(I) complexes and limited studies of IC50 values for
such compounds have been reported.17–20 We recently described
the synthesis and cytotoxicity of rhenium(I) tricarbonyl thymidine
(dT) complexes.21 Based on results of a rhenium(I)–folic acid
conjugate that exhibited greater toxicity than cisplatin in a screen
agents and represent a significant part of chemotherapeutic drugs
approved for clinical use to date.1–5 The main characteristic of
cancer cells is their elevated proliferation compared to quiescent
cells. For a cell to divide it must replicate all of its components
including its genome but unlike major elements, such as RNA
or proteins, DNA replication does not occur to a great extent
in quiescent cells. The treatment with nucleoside-based drugs
thus exploits the fact that the majority of cells are not in the
process of replication, and their mechanism of action is then
based on some interference of DNA replication in proliferating
cells. For nucleoside analogs to exhibit their cytotoxicity, they
need to enter into the intracellular space across the plasma
membrane. Because of their hydrophilic character, however,
they cannot cross cellular membranes by simple diffusion and
transport occurs through carrier proteins in the form of specific
nucleoside transporters.6,7 Once inside the cell, nucleoside-based
aDepartment of Chemistry, 1-014 Center for Science and Tech-
nology, College Place, Syracuse University, Syracuse, NY, 13244.
E-mail: rpdoyle@syr.edu, jazubiet@syr.edu; Fax: +1 315 443-4070; Tel: +1
315 443-4070
bMolecular Insight Pharmaceuticals, 120 Second Street, Cambridge, MA,
02142
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c0dt01452d
‡ Current address: Harvard Medical School, Children’s Hospital Boston,
Division of Nuclear Medicine, Department of Radiology, 300 Longwood
Avenue, Boston, MA 02115.
6216 | Dalton Trans., 2011, 40, 6216–6225
This journal is
The Royal Society of Chemistry 2011
©

against a doxorubicin- and cisplatin-resistant human ovarian
cancer cell line (A2780/AD) over-expressing the folate receptor,
our approach was to develop a nucleoside-based anticancer agent
that takes advantage of the antiproliferative properties of the non-
radioactive [Re(CO)3]+1 core.22 In this previous study of Re(CO)3
dT complexes, we were able to identify C5¢ and N3 conjugated
Re(CO)3 dT complexes that exhibited cytotoxic activity against
A549 lung carcinoma cells. We furthermore found that the toxicity
is dependent on the length and the nature of the spacer. A general
trend was established for position C5¢ of the ribose sugar in
which an increase of toxicity was noted with increasing spacer
length between dT and the [Re(CO)3]+1 core. A C5¢ derivatized
dT complex equipped with a dodecyl spacer and a dipicolylamine
metal chelate is more potent than cisplatin and doxorubicin against
A549 cells. In the same study, a N3 derivatized dT complex with
a hexyl spacer was also of moderate toxicity whereas its shorter
homologue with an ethylene spacer was not toxic indicating that
further elongation of the spacer might also result in an increase of
toxicity.
The [M(CO)3]+1 core with M = 99mTc or Re has received
considerable interest for SPECT imaging (single photon emis-
sion tomography) because of its small size and organometallic
character compared to other Tc core structures.23 While 99mTc
is extensively used in SPECT, the radioisotopes of its group 7
congener, rhenium (186Re and 188Re), are among the most promising
radionuclides for therapeutic applications. The [M(CO)3]+1 core
is chemically robust and resistant to oxidation, and the small
size allows the labeling of low-molecular weight biomolecules.
Because of the d6 low-spin configuration of the transition
metal, the corresponding complexes of the [M(CO)3]+1 core are
chemically inert maintaining their integrity even under harsh
reaction conditions. The corresponding technetium and rhenium
precursors [M(CO)3(H2O)3]+1 can be conveniently accessed by
synthetic procedures, are water soluble, and the three aqua ligands
are labile and can readily undergo ligand exchange.24–26
to cell uptake, an additional derivative with a dodecyl spacer but
with a bis-quinoline metal chelate was prepared to study the cell
internalization of these compounds by fluorescence microscopy.
Results
Synthesis and characterization of thymidine ligands and their
corresponding Re(CO)3 core complexes
From our previous investigations, we learned that the cou-
pling of tosylalkylphthalimide precursors with an alkyl chain
no longer than six CH2 groups occurs at position N3 of the
pyrimidine ring of unprotected dT. Longer chained homologues,
such as tosyloctylphthalimide, however react preferentially with
position C5¢ of the ribose sugar. For the selective coupling of
precursors 2–4 to dT, protection of the C5¢ position of dT
was necessary and consequently compound 1 with positions
C5¢ and C3¢ protected by tert-butyldimethylsilyl groups was
used as the starting material for the synthesis (Scheme 1). The
precursors 2–4 were prepared according to a recently published
method.21 For the ligand synthesis, di-TBDMS-dT 1 was reacted
with the corresponding tosylalkylphthalimide precursor 2–4 and
anhydrous potassium carbonate at 60 ◦C in anhydrous DMF
Encouraged by our previous results of C5¢ functionalized
Re(CO)3 dT complexes, we have pursued the preparation of
a small series of corresponding counterparts and hypothesized
that they might show a similar trend, namely increased toxicity
with increasing spacer length between dT and the metal core.
The potential novelty of such previously described Re(CO)3 dT
complexes and complexes described herein is their potential to
provide a targeted combination of chemo- and radiotherapy in a
single compound in combination with the corresponding rhenium
radionuclides 186Re and 188Re, a major advantage over commonly
used anticancer agents.
Herein, we report the synthesis, cytotoxicity, and confirmed
internalization by endocytosis of a preliminary series of N3
functionalized Re(CO)3 dT complexes. For investigation of the
correlation between the spacer length and the toxicity complexes
of different spacer lengths were prepared, specifically octyl, decyl,
and dodecyl spacer. To date, the major question of Re(CO)3
dT complexes that remains unresolved is their ability to enter
the intracellular space and the mechanism by which cell uptake
occurs. We recently reported that the Re(CO)3 complex with
a bis(quinolin-2-ylmethyl)amino moiety displays luminescence
properties with a long emission lifetime, relatively high quantum
yield, and a large Stokes’ shift making this complex moiety an
ideal fluorescence probe for in vitro imaging studies.27 With regard
Scheme 1 Synthesis of dT precursors. a) K2CO3 anhydrous/DMF/60 ◦C;
b) 0.5 M ethanolic hydrazine, rt.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6216–6225 | 6217
©

Scheme 2 Synthesis of dT ligands and their corresponding Re(CO)3 complexes. a) thiazole-4-carboxaldehyde/NaBH(OAc)3/DCE/rt; b) 3 M methanolic
HCl; c) quinoline-2-carboxaldehyde/NaBH(OAc)3/DCE/rt; d) [Re(CO)3(H2O)3]Br/MeOH/60 ◦C.
Table 1 Inhibition data for Re(CO)3 thymidine complexes from the
under argon atmosphere overnight providing intermediates 5–
7 in moderate yields (47–78%) after workup. The phthalimide
protecting group was removed by overnight reaction of 5–7 in
a 0.5 M ethanolic hydrazine solution at room temperature to
obtain the corresponding free amines 8–10 in 52–57% yield. The
tridentate chelate functionality for binding of the [Re(CO)3]+1 core
was then introduced in excellent yields (82–99%) to the corre-
sponding amine by reductive amination with either thiazole-4-
carboxaldehyde (11–13) or quinoline-2-carboxaldehyde (14) using
sodium triacetoxyborohydride in dichloroethane (Scheme 2). The
tert-butyldimethylsilyl groups were cleaved by reacting 11–14 in 3
M methanolic hydrochloric acid at room temperature to give the
ligands 15–18 in excellent yields (87–91%). All intermediates and
the ligands were characterized by 1H NMR spectroscopy.
cytotoxicity study against A549 cells
Compound
Position
Spacer
IC50 [mM]
42
5-FdU
C5
—
7
19
N3
N3
N3
N3
C5¢
C5¢
C5¢
N3
N3
Octyl
147 74
27 11
46 22
3.4 1.6
160 18a
150 17a
6.6 3.0a
2500 1300a
124 23a
20
Decyl
Dodecyl
Dodecyl
Octyl
Decyl
Dodecyl
Ethyl
21
22
C5¢ (CH2)8 Re(CO)3 dT (23)
C5¢ (CH2)10 Re(CO)3 dT (24)
C5¢ (CH2)12 Re(CO)3 dT (25)
N3 (CH2)2 Re(CO)3 dT (27)
N3 (CH2)6 Re(CO)3 dT (28)
Hexyl
a From Ref. 21.
The corresponding rhenium complexes 19–22 were obtained in
excellent yields (80–90%) by reaction of [Re(CO)3(H2O)3]Br and
a stoichiometric amount of the corresponding ligand 15–18 in
the free ligand which is replaced by two doublets in the spectrum
of the complex as the CH2 protons are no longer equivalent. The
observed AB coupling pattern is furthermore indicative for a facial
coordination of the [Re(CO)3]+1 core to the tridentate chelate.21,36,37
The mass spectrometry data is in agreement with the NMR
spectroscopic data and the highest m/z ratios could be assigned
to the parent complex [Re(CO)3L]+1. The corresponding isotope
patterns are in close agreement with the isotope distribution of
185/187Re complexes (Supporting Information†).21,36,37
◦
methanol at 60 C. The metal precursor [Re(CO)3(H2O)3]Br was
prepared according to the literature procedure.24 In case of 19–21,
the crude products were purified by solid phase extraction using
C18 Sep-Pak cartridges and a water–methanol gradient. Complex
22 was purified by semi-preparative reversed phase HPLC using
a water–methanol gradient. All complexes were characterized
by electrospray mass spectrometry and NMR spectroscopy. The
NMR spectroscopic analyses provide strong evidence of the site
specific coordination of the [Re(CO)3]+1 core to the tridentate
chelate functionalities introduced at position N3 of the pyrimi-
dine base. A pronounced downfield shift of the methylene and
thiazole/quinoline ring protons of the tridentate chelate moiety
has been observed after metal chelation. In contrast, the remaining
proton signals of dT showed no significant shift indicating site spe-
cific coordination of the [Re(CO)3]+1 core. This has been observed
for similar examples in the literature and is also in agreement with
our previous report.21,36,37 In addition, the four equivalent CH2
protons of the bis(thiazol-4-ylmethyl)amino and bis(quinolin-2-
ylmethyl)amino moiety exhibit a singlet in the proton NMR of
In vitro evaluation
Cell cytotoxicity. In our first study, the Re(CO)3 dT complexes
were screened against the A549 lung carcinoma cell line. We
have found that C5¢ and N3 derivatized dT complexes exhibited
moderate toxicity and that the toxicity was dependent on the
spacer length between dT and the metal core. Significantly, a
C5¢ dodecyl Re(CO)3 dT 25 has been identified that is more
potent than cisplatin and doxorubicin with an IC50 value of 6.6
3.0 mM (Table 1). The series of dT complexes described herein
are the extension of the N3 derivatized series of increasing spacer
6218 | Dalton Trans., 2011, 40, 6216–6225
This journal is
The Royal Society of Chemistry 2011
©

length to investigate if a similar trend can be observed for the N3
functionalized dT complexes as seen for the C5¢ analogs. In this
study, 19–22 were thus screened against the A549 cell line by cell
viability assay in accordance with US National Institute of Health
(NIH) guidelines at 48 h time points.28 5-Fluorouridine (5-FdU)
was used as a control showing moderate toxicity against A549
cells with an IC50 of 42 7 mM (Table 1). In accordance with
the C5¢ derivatives, a similar trend has been observed for the N3
analogs where the toxicity increased with the elongation of the
spacer length. Complex 19 with an octyl spacer is of moderate
potency with an IC50 value of 147 74 mM, a value in the range
of its C5¢ derivatized counterpart. Extension of the spacer by two
and four CH2 groups results in further increase of the toxicity
as illustrated by the IC50 values of 27 11 mM and 46 22 mM
for 20 and 21, respectively. However, while the general trend is
apparent, a perfectly linear correlation between toxicity and spacer
length for the N3 derivatized dT complexes does not occur as
illustrated by the break in the trend for 20 and 21. The difference
in toxicity between 21 and 25 is most likely a consequence of the
interplay between the site of modification (C5¢ vs. N3) and the
difference in the chelate moiety (dipicolylamino vs. bis(thiazol-
4-ylmethyl)amino). The identity of the metal chelate therefore
may play an important role in the antiproliferative properties of
rhenium dT complexes. This is further indicated by the difference
in toxicity for complexes 21 and 22 which are both of the same
spacer length but differ in the metal chelate functionality. To our
delight, the change to a bis(quinolin-2-ylmethyl)amino chelate
moiety resulted in a nearly ten-fold increase of toxicity of 22
over 21 with an IC50 value of 3.4 1.6 mM. This is also a two-
fold increase compared to its C5¢ functionalized counterpart 25.
From both series (C5¢ and N3), it furthermore becomes clear
that a minimum spacer length (decyl) is required for rhenium dT
complexes to unfold their pronounced cytotoxicity. To examine the
role of thymidine dependent enzymes the toxicities of 20, 21, and
22 were examined in the presence of 100-fold excess thymidine. The
IC50 values for these complexes did not change significantly with
IC50 values of 34 8, 15 6, and 10 5 mM respectively. This would
suggest that their toxicity cannot be mitigated by thymidine, which
is also consistent with a non dT dependent uptake mechanism
and lack of inhibition of hTK-1 (vide infra). IC50 values of other
metal-based anticancer agents against the A549 cell line have
been reported for organometallic osmium(II) arene complexes
(ª 24 mM at 24 h), chloro half-sandwich osmium(II) complexes
(6.4 mM at 24 h), an alkenyl-substituted ansa-zirconocene complex
(136.4 mM at 96 h), and an octaazacyclotetracosane dizinc(II)
complex (8.8 mM at 48 h).29–34 In contrast, organic chiral salicyl
diamines are potent anticancer molecules, with IC50 values in the
range of 0.8–1.8 mM at 48 h.35
performed uptake studies with 99mTc(CO)3 N3 and C3¢ derivatized
dT complexes and found that hydrophilic dT complexes were not
actively internalized into U251 and LN229 cells.36 In another
report of N3 functionalized dT complexes, only a N3 decylene
dT complex of neutral overall charge and a positive log P value
showed minor cellular uptake into SKNMC cells.37 The authors
proposed that the uptake of the complexes is not actively mediated
by nucleoside transporters and suggested that their primary route
of internalization might occur via passive diffusion. This has also
been suggested by Tjarks and co-workers as the major uptake
mechanism for carboranethymidine derivatives.38
To study the cell internalization, confocal microscopy experi-
ments were performed with 22 on A549 cells. A solution of 22
at a concentration of 20 mM was added to the cells followed
by incubation for 1 h at 37 ◦C. The complex solution was then
removed and cells washed with media. Fig. 1a clearly shows the
cellular internalization of 22 at 37 ◦C as visualized by fluorescence
microscopy with excitation at 488 nm. Cell internalization was
confirmed by scanning through the intracellular regions by slicing
through the cell in 1 mm increments (ESI†). Fluorescence was
uniformly distributed in and throughout the cell, which clearly
indicates the successful uptake of 22. Nuclear counterstaining
was preformed as above with the addition of Nuclear-ID Red
membrane permeable stain following manufacture’s instructions
(Enzo). No co-localization of 22 and Nuclear-ID red was observed
indicating 22 does not get incorporated into the nucleus (see ESI†).
Additional experiments were performed at 4 ◦C initially to check
whether uptake was a consequence of receptor mediated internal-
ization or passive diffusion. As shown in Fig. 1b, internalization of
22 was also observed at 4 ◦C, which indicates that 22 is internalized
by passive diffusion.
To determine whether the presence of the dT moiety has some
influence on the uptake, blocking experiments with a 100-fold
excess of dT over 22 were performed in addition. The pretreatment
with excess of dT resulted in uptake as shown in Fig. 2 indicating
that the presence of the dT moiety may not play a crucial role in
the uptake of 22. From the results of both experiments, one might
thus draw the conclusion that the uptake of 22 occurs through
non-specific passive diffusion.
Additional endosomal co-localization experiments were con-
ducted on A549 cells to further elucidate if the uptake of 22
is based on endocytosis. Fluorescent tagging of endosomes was
carried out using the Organelle Lights kit from Invitrogen. The
media was replaced with PBS containing 20 mM of 22 and the
cells were incubated for 1 h at 37 ◦C. The tagged endosomes were
excited at 543 nm and 22 visualized by excitation at 488 nm. Fig. 3a
shows the internalization of 22 and Fig. 3b shows the fluorescently-
labeled endosomes. As seen in Fig. 3d in the merged image, the
red fluorescence of the dye counterstained with the green of 22
appears mostly yellow. This observation suggests that uptake of
22 might also be occurring, at least in part, by an endocytotic
process at 37 ◦C.
Cell uptake studies. For a nucleoside-derived anticancer agent
to exhibit its antiproliferative properties, it has to be internalized
into the cell. With antiproliferative C5¢ and a N3 derivatized dT
complexes at hand, one of the questions remaining was whether
they are able to penetrate the extracellular membrane to enter
the intracellular space or, in contrast, are not internalized and
aggregateinthecellmembrane. The mainroutes for internalization
are the two energy-dependent mechanisms, endocytosis and active
transport, or the two energy-independent processes, facilitated
diffusion and passive diffusion. Schibli and co-workers have
Thymidine kinase inhibition assays. The primary enzyme in-
volved in the salvage pathway of dT is human thymidine kinase-
1 (hTK-1), which transfers a phosphate group from ATP to
dT.39,40 The corresponding dT monophosphate is then converted
into its di- and triphosphate by other enzymes and finally
incorporated into DNA. The enzyme hTK-1 is over-expressed
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6216–6225 | 6219
©

Fig. 1 Fluorescence images of A549 incubated with 20 mM of complex 22 a) at 37 ◦C and b) at 4 ◦C showing cell internalization.
were not soluble in the assay media. However, the IC50 values of
19 and 20 can be estimated to be 10-fold higher than AZT which
has been assayed previously in a similar experiment.21 For both
21 and 22 with a dodecyl spacer, no inhibition was observed up
to 500 mM, suggesting that further elongation of the spacer does
not result in improved hTK-1 interaction. A significant increase
in inhibition has not then been observed for all of the described
complexes compared to the previously described 28. With respect
to hTK-1 interaction, a spacer length of six to ten methylene
groups thus appears to be optimal for conjugations at position N3.
The low inhibitory effect of complexes 19–22 suggests, however,
that the observed toxicity is most likely not based on hTK-1
interaction.
Conclusions
We successfully prepared and characterized a small series of N3
functionalized Re(CO)3 dT complexes to elucidate if a similar
trend in cytotoxicity against the A549 lung carcinoma cell line
can be observed as seen for their C5¢ counterparts. Indeed, an
Fig. 2 Fluorescence images of A549 cells incubated with 20 mM of
complex 22 with 100 fold excess of dT.
in A549 cells.41 According to the crystal data, hTK-1 accepts
rather large substituents at position N3 of dT.40 This was recently
confirmed by reports of Tjarks and Schibli.36,37,38 In this respect,
we previously found that 28 showed inhibition at a concentration
of 500 mM and thus hypothesized that the toxicity of our N3
derivatized complexes might be correlated to hTK-1 interaction.
For competitive inhibition experiments a constant concentration
of 10 mM [3H]-dT was incubated with hTK-1 at 37 ◦C for 15 min
in the presence of 0–500 mM test compound. Samples were then
spotted on Whatman disks and the retained radioactivity detected.
In this experiment, the IC50 value of the control dT was determined
to 16 2 mM. Complexes 19 and 20 carrying an octyl and a decyl
spacer, respectively, showed convincing inhibition at the highest
concentration (500 mM) but did not reach baseline. An exact
determination of the corresponding IC50 values was not possible
because at concentrations higher than 500 mM, the complexes
extension of the spacer length between dT and the metal chelate
moiety results in an increased toxicity. A very potent compound,
namely 22, which is more toxic than doxorubicin and cisplatin
against A549 cells, has been identified with an IC50 value of
3.4 1.6 mM at 48 h. In our first study, we had been able to
establish a correlation between the toxicity and the site, and
nature of the conjugation as well as the spacer length. In this
contribution, we find that the nature of the metal chelating
moiety might have an impact on the toxicity (bis-quinoline vs. bis-
thiazole). Cell uptake studies by fluorescence microscopy clearly
showed the internalization of complex 22 which was one of the
major obstacles for Re(CO)3 dT complexes. To the best of our
knowledge, this is the first report on the successful internalization
of Re(CO)3 dT complexes. However, blocking and temperature-
dependent experiments suggest that the uptake is not receptor
mediated and rather occurs primarily via passive diffusion, with
6220 | Dalton Trans., 2011, 40, 6216–6225
This journal is
The Royal Society of Chemistry 2011
©

Fig. 3 Fluorescence images of endosomal counter staining with 20 mM of complex 22. A shows the fluorescence and distribution of 22 in the cell (shown
as green), while B shows counter staining of the endosomes (shown in red). C is the transmitted light picture of the cell. D is the overlay of A, B, and C
with the overlap of the endosomes and complex 22 shown in yellow.
non-specific endocytosis also possible. The mechanism of entry
affects the cell-type specificity, the rate of internalization, and
Experimental
Methods and materials
the fate of the complex inside the cell. Passive diffusion is less
cell-type specific while endocytosis often leads to entrapment
of molecules in endosomes and consequently to degradation
by lysosomal enzymes. For these reasons, further development
on Re(CO)3 dT complexes is necessary to ensure cell-type
specificity.
To shed light on the mechanism of action of Re(CO)3 dT
complexes, initial competitive inhibition experiments with hTK-
1, the primary enzyme involved in the salvage pathway of dT,
have been performed. The results, however, do not substantiate
the assumption to a significant extent. All described complexes
comprise a positive overall charge and it seems likely that the
cytotoxicity is related to their positive overall charge. For example,
the positive charged complexes might interact with the negatively
charged DNA backbone after internalization. The mechanism of
action of C5¢ and N3 conjugated Re(CO)3 dT complexes, however,
remains speculative at this point, and further investigations
including the specificity towards particular cell lines, the influence
of the overall charge and detailed insight into the mode of toxicity
are currently being performed in our laboratory.
Chemicals and analysis. Chemicals and dry solvents were
purchased from Fluka, Alfa Aesar or Sigma-Aldrich (USA).
Chemicals and anhydrous solvents were of highest purity com-
mercially available and applied without further purification.
Standard solvents were of reagent grade and obtained from
PharmaCo Products Inc. (USA). Reactions were monitored by
thin layer chromatography using precoated silica gel 60 F254
aluminium sheets (Merck) and precoated alumina oxide 60 F254
PET foils (Sigma), respectively, visualized by UV absorption.
Column chromatography was performed using silica gel 60 (MP
Biomedicals GmbH) and alumina oxide 58 (Alfa Aesar, activated,
basic, Brockmann Grade I). An Agilent 1100 reversed-phase
HPLC instrument with manual injection and automated fraction
collector was fitted with a C18 semipreparative column (Agilent
Eclipse XDB-C18 9.4 ¥ 250 mm) at a flow rate of 2 mL min-1.
Detection was performed by UV monitoring at 260 nm. The
gradient used was 80% 0.1% TFA in H2O and 20% MeOH to
0% 0.1% TFA in H2O and 100% MeOH in 16 min over an entire
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6216–6225 | 6221
©

period of 20 min. 1H and 13C NMR spectroscopy was performed on
a Bruker Avance DPX 300 MHz or Bruker Avance DPX 500 MHz
instrument, respectively. 2D NMR experiments were exclusively
performed on the Bruker Avance DPX 500 MHz spectrometer.
Chemical shifts are given as parts per million (ppm) and are
reported relative to residual solvent peaks or to tetramethylsilane.
Coupling constants are reported in Hertz (Hz). The multiplicity of
the NMR signals are described as follows: s = singlet, d = duplet,
t = triplet, q = quartet, m = multiplet. A Shimadzu LCMS-2010 A
mass spectrometer was used for low resolution ESI MS analysis
of the rhenium tricarbonyl complexes.
colorless oil (4.80 g, 47%). H NMR (CDCl3, 300 MHz): d 7.76
1
(2H, dd, J = 2.97, 5.56), 7.63 (2H, dd, J = 3.14, 5.33), 7.38 (1H,
d, J = 1.18), 6.31 (1H, m), 4.34 (1H, m), 3.87–3.68 (5H, m), 3.59
(2H, m), 2.20 (1H, m), 1.96 (1H, m), 1.85 (3H, d, J = 1.08), 1.58
(4H, m, br), 1.26 (8H, s, br), 0.85 (18H, d, J = 10.38), 0.04 (12H,
m).
Synthesis of 2-(10-(3-((2R,4S,5R)-4-(tert-butyldimethylsily-
loxy)-5-((tert-butyldimethylsilyloxy)methyl)tetrahydrofuran-2-yl) -
5-methyl-2,6-dioxo-2,3-dihydropyrimidin-1(6H)-yl)decyl)isoin-
doline-1,3-dione (6). Compound 6 was prepared by a procedure
similar to that described for compound 5, except that compound
3 was used instead of compound 2. The reaction mixture was
purified by silica gel column chromatography using a gradient
of hexane:ethylacetate 1 : 1 (v/v). The product was obtained as
Ligand synthesis and characterization
Synthesis of 1-((2R,4S,5R)-4-(tert-butyldimethylsilyloxy)-5-
((tert-butyldimethylsilyloxy)methyl)tetrahydrofuran-2-yl)-5-meth-
ylpyrimidine-2,4(1H,3H)-dione (di-TBDMS-dT, 1). Thymidine
(10.00 g, 41.28 mmol), tert-butyldimethylsilyl chloride (13.07 g,
86.68 mmol) and imidazole (11.24 g, 165.09 mmol) were mixed in
80 mL anhydrous DMF under an inert atmosphere (argon) and
stirred for 18 h at room temperature (r.t.). The mixture was then
partitioned between each 60 mL water and dichloromethane. The
organic layer was further washed twice with water. The aqueous
layer was extracted once more with 20 mL dichloromethane and
the combined organic layers dried using anhydrous sodium sulfate
and filtered. The filtrate was concentrated under reduced pressure
to a sticky oil and co-evaporated three times with 20 mL toluene.
The crude product was further purified by silica gel column
chromatography starting with dichloromethane. The product was
finally obtained by elution with dichloromethane: methanol 40 : 1
1
colorless oil (3.19 g, 78%). H NMR (CDCl3, 300 MHz): d 7.75
(2H, dd, J = 3.00, 5.64), 7.64 (2H, dd, J = 3.12, 5.35), 7.36 (1H,
d, J = 1.18), 6.30 (1H, m), 4.34 (1H, m), 3.85–3.64 (5H, m), 3.55
(2H, m), 2.19 (1H, m), 1.95 (1H, m), 1.84 (3H, d, J = 1.08), 1.57
(4H, m, br), 1.26 (12H, s, br), 0.85 (18H, d, J = 10.37), 0.04 (12H,
m).
Synthesis of 2-(12-(3-((2R,4S,5R)-4-(tert-butyldimethylsily-
loxy)-5-((tert-butyldimethylsilyloxy)methyl)tetrahydrofuran-2-yl)-
5-methyl-2,6-dioxo-2,3-dihydropyrimidin-1(6H)-yl)dodecyl)isoin-
doline-1,3-dione (7). Compound 7 was prepared by a procedure
similar to that described for compound 5, except that compound
4 was used instead of compound 2. The reaction mixture was
purified by silica gel column chromatography using a gradient
of hexane:ethylacetate 1 : 1 (v/v). The product was obtained as
1
a colorless oil (7.13 g, 77%). H NMR (CDCl3, 300 MHz): d
1
(v/v) in quantitative yield. H NMR (CDCl3, 300 MHz): d 9.30
7.77 (2H, dd, J = 3.00, 5.60), 7.65 (2H, dd, J = 3.12, 5.31), 7.39
(1H, d, J = 1.18), 6.32 (1H, m), 4.36 (1H, m), 3.90–3.64 (5H, m),
3.60 (2H, m), 2.21 (1H, m), 1.98 (1H, m), 1.86 (3H, d, J = 1.10),
1.58 (4H, m, br), 1.26 (16H, s, br), 0.86 (18H, d, J = 10.41), 0.04
(12H, m).
(1H, s, br), 7.41 (1H, d, J = 1.21), 6.28 (1H, t, J = 5.95), 4.36 (1H,
m), 3.87–3.69 (3H, m), 2.18 (1H, m), 1.96 (1H, m), 1.86 (3H, d,
J = 1.08), 0.85 (18H, d, J = 10.00), 0.03 (12H, m).
Synthesis of Tosylalkylphthalimides (2–4). The tosylalkylph-
thalimide precursors have been prepared according to a previously
published synthesis.21 Briefly, the corresponding chloro/bromo-
alkylalcohol was reacted with 1.1 equivalents potassium ph-
thalimide under reflux in DMF to obtain the corresponding
n-hydroxyalkylphthalimide (Yield: 72–90%), which was in turn
converted into its tosylate by reaction with 1.5 equivalents p-
toluenesulfonyl chloride in a chloroform/pyridine mixture at r.t.
(Yield: 42–55%).
Synthesis of 3-(8-aminooctyl)-1-((2R,4S,5R)-4-(tert-butyldime-
thylsilyloxy)-5-((tert-butyldimethylsilyloxy)methyl)tetrahydrofuran-
2-yl)-5-methylpyrimidine-2,4(1H,3H)-dione (8). Compound 5
(4.80 g, 11.21 mmol) was stirred in a 0.5 M ethanolic hydrazine
solution (50 mL) for 16 h at room temperature. During the course
of the reaction, the formation of a white precipitate was observed.
The suspension was filtered and the solvent and excess hydrazine
was removed from the filtrate under reduced pressure. The crude
reaction mixture was further purified by silica gel chromatography
using a mixture of chloroform–methanol 5 : 1 (v/v). The product
Synthesis of 2-(8-(3-((2R,4S,5R)-4-(tert-butyldimethylsilyloxy)-
5-((tert-butyldimethylsilyloxy)methyl)tetrahydrofuran-2-yl)-5-me-
thyl-2,6-dioxo-2,3-dihydropyrimidin-1(6H)-yl)octyl)isoindoline-1,
3-dione (5). Di-TBDMS-dT 1 (6.61 g, 14.04 mmol), anhy-
drous potassium carbonate (4.31 g, 31.18 mmol) and 6-(1,3-
dioxoisoindolin-2-yl)octyl 4-methylbenzenesulfonate 2 (6.70 g,
15.59 mmol) were suspended in anhydrous DMF (80 mL) and
stirred at 60 ◦C overnight. The solvent was removed in vacuo and
the remaining residue co-evaporated once with toluene (50 mL).
The remaining residue was redissolved in 50 mL dichloromethane
and the organic phase washed twice with brine (50 mL), dried
using anhydrous sodium sulfate, filtered, and concentrated under
reduced pressure. The crude product was further purified by silica
gel chromatography starting with chloroform. The product was
eluted using a chloroform–methanol mixture 60 : 1 to give 5 as
1
was obtained as a colorless oil (2.79 g, 52%). H NMR (CDCl3,
300 MHz): d 7.42 (1H, d, J = 1.18), 6.36 (1H, m), 4.39 (1H, m),
3.90–3.70 (5H, m), 2.66 (2H, t, J = 6.75), 2.26 (1H, m), 1.96 (1H,
m), 1.92 (3H, d, J = 1.08), 1.60 (2H, m, br), 1.46 (2H, m, br), 1.22
(10H, m, br), 0.91 (18H, d, J = 9.85), 0.07 (12H, m).
Synthesis of 3-(10-aminodecyl)-1-((2R,4S,5R)-4-(tert-butyl-
dimethylsilyloxy)-5-((tert -butyldimethylsilyloxy)methyl)tetrahy-
drofuran-2-yl)-5-methylpyrimidine-2,4(1H,3H)-dione (9). Com-
pound 9 was prepared by a procedure similar to that described
for compound 8, except that compound 6 was used instead of
compound 5. The reaction mixture was purified by silica gel
column chromatography using a mixture of chloroform–methanol
5 : 1 (v/v). The product was obtained as a colorless oil in 57% yield.
6222 | Dalton Trans., 2011, 40, 6216–6225
This journal is
The Royal Society of Chemistry 2011
©

1H NMR (CDCl3, 300 MHz): d 7.44 (1H, d, J = 1.16), 6.38 (1H,
m), 4.40 (1H, m), 3.95–3.74 (5H, m), 2.68 (2H, t, J = 6.71), 2.26
(1H, m), 1.99 (1H, m), 1.93 (3H, d, J = 1.06), 1.61 (2H, m, br),
1.46 (4H, m, br), 1.27 (12H, m, br), 0.92 (18H, d, J = 9.72), 0.08
(12H, m).
yield. 1H NMR (CDCl3, 300 MHz): d 8.78 (2H, d, J = 2.05), 7.43
(1H, d, J = 1.15), 7.29 (2H, d, J = 2.05), 6.38 (1H, m), 4.40 (1H,
m), 3.95–3.75 (9H, m), 2.55 (2H, m), 2.26 (1H, m), 2.00 (1H, m),
1.94 (3H, d, J = 1.18), 1.58 (4H, m, br), 1.26 (16H, m, br), 0.92
(18H, d, J = 9.85), 0.09 (12H, m).
Synthesis of 3-(12-aminododecyl)-1-((2R,4S,5R)-4-(tert-butyl-
dimethylsilyloxy)-5-((tert -butyldimethylsilyloxy)methyl)tetrahy-
Synthesis of 3-(12-(bis(quinolin-2-ylmethyl)amino)dodecyl)-
1-((2R,4S,5R)-4-(tert-butyldimethylsilyloxy)-5-((tert-butyldime-
thylsilyloxy)methyl)tetrahydrofuran-2-yl)-5-methylpyrimidine-
2,4(1H,3H)-dione (14). Compound 14 was prepared by a
procedure similar to that described for compound 11, except that
compound 10 and 2-quinolinecarboxaldehyde were used instead
of compound 8 and thiazole-4-carboxaldehyde. The reaction
mixture was purified by basic alumina oxide chromatography
using hexane:ethylacetate 4 : 1 (v/v) as eluent to give 11 as a dark
drofuran-2-yl)-5-methylpyrimidine-2,4(1H,3H)-dione
(10).
Compound 10 was prepared by a procedure similar to that
described for compound 8, except that compound 7 was used
instead of compound 5. The reaction mixture was purified by
silica gel column chromatography using a mixture of chloroform–
methanol 5 : 1 (v/v). The product was obtained as a colorless oil
1
in 55% yield. H NMR (CDCl3, 300 MHz): d 7.45 (1H, d, J =
1
1.15), 6.39 (1H, m), 4.42 (1H, m), 3.98–3.78 (5H, m), 2.70 (2H, t,
J = 6.80), 2.28 (1H, m), 1.99 (1H, m), 1.94 (3H, d, J = 1.10), 1.64
(2H, m, br), 1.49 (4H, m, br), 1.29 (16H, m, br), 0.93 (18H, d, J =
10.00), 0.08 (12H, m).
orange oil in 91% yield. H NMR (CDCl3, 300 MHz): d 8.12
(2H, d, J = 8.53), 8.04 (2H, d, J = 8.46), 7.77 (4H, t, J = 7.92),
7.67 (2H, m), 7.49 (2H, m), 7.44 (1H, s), 6.38 (1H, m), 4.40 (1H,
m), 4.00–3.74 (9H, m), 2.60 (2H, m), 2.25 (1H, m), 1.99 (1H, m),
1.93 (3H, s), 1.59 (4H, m, br), 1.21 (16H, m, br), 0.91 (18H, d, J =
9.92), 0.08 (12H, m).
Synthesis
of
3-(8-(bis(thiazol-4-ylmethyl)amino)octyl)-1-
((2R,4S,5R)-4-(tert-butyldimethylsilyloxy)-5-((tert-butyldimethyl-
silyloxy)methyl)tetrahydrofuran - 2 - yl) - 5 - methylpyrimidine - 2,4 -
(1H,3H)-dione (11). Compound 8 (1.08 g, 1.80 mmol) and
thiazole-4-carboxaldehyde (0.43 g, 3.78 mmol) were mixed in
anhydrous dichloroethane (20 mL) under an inert atmosphere
(argon) and stirred at r.t. for 30 min followed by the addition
of sodium triacetoxyborohydride (1.14 g, 5.39 mmol). The
reaction was stirred for additional 16 h. After completion of the
reaction, the solvent was removed under reduced pressure and
the residue taken up in methanol (50 mL). After the gas evolution
ceased, the solvent was removed again and the crude reaction
mixture purified by basic alumina oxide chromatography using
chloroform as eluent to give 11 as a pale-yellow oil (1.25 g, 82%).
1H NMR (CDCl3, 300 MHz): d 8.75 (2H, d, J = 2.05), 7.42 (1H,
d, J = 1.19), 7.26 (2H, d, J = 2.10), 6.36 (1H, m), 4.39 (1H, m),
3.92–3.56 (9H, m), 2.51 (2H, m), 2.21 (1H, m), 1.97 (1H, m), 1.91
(3H, d, J = 1.04), 1.54 (4H, m, br), 1.24 (8H, m, br), 0.89 (18H, d,
J = 10.02), 0.06 (12H, m).
Synthesis
of
3-(8-(bis(thiazol-4-ylmethyl)amino)octyl)-1-
((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-
5-methylpyrimidine-2,4(1H,3H)-dione (15). Compound 11
(1.25 g, 1.57 mmol) was dissolved in 20 mL methanol and to that
solution 5 mL concentrated hydrochloric acid were added. The
solution was stirred for 16 h at r.t. and concentrated in vacuum
afterwards resulting in a pale-yellow solid. The crude product
was further purified by basic alumina oxide chromatography
using chloroform–methanol 40 : 1 (v/v) to yield compound 15 as
a pale-yellow solid (0.75 g, 85%). 1H NMR (CDCl3, 300 MHz): d
8.84 (2H, d, J = 1.05), 7.53 (2H, s, br), 7.40 (1H, d, J = 1.09), 6.17
(1H, m), 4.62 (1H, m), 4.06 (4H, m), 3.91 (4H, m), 3.73 (1H, m),
2.64 (2H, m), 2.40 (2H), 1.92 (3H, d, J = 1.87), 1.62 (4H, m, br),
1.25 (8H, m, br).
Synthesis
of
3-(10-(bis(thiazol-4-ylmethyl)amino)decyl)-1-
((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-
5-methylpyrimidine-2,4(1H,3H)-dione (16). Compound 16 was
prepared by a procedure similar to that described for compound
15, except that compound 12 was used instead of compound
11. The reaction mixture was purified by basic alumina oxide
chromatography using chloroform–methanol 40 : 1 (v/v) as eluent
to give 16 as a pale-yellow solid in 91% yield. 1H NMR (CDCl3,
300 MHz): d 8.90 (2H, d, J = 2.04), 7.64 (1H, d, J = 1.12), 7.41
(2H, d, J = 2.01), 6.33 (1H; m), 4.40 (1H, m), 3.90–3.70 (8H, m),
3.60 (1H, m), 2.53 (2H, m), 2.25 (2H, m), 1.90 (3H, d, J = 1.02),
1.53 (4H, m, br), 1.23 (12H, m, br).
Synthesis
of
3-(10-(bis(thiazol-4-ylmethyl)amino)decyl)-1-
((2R,4S,5R)-4-(tert -butyldimethylsilyloxy)-5-((tert -butyldime-
thylsilyloxy)methyl)tetrahydrofuran-2-yl)-5-methylpyrimidine-2,4-
(1H,3H)-dione (12). Compound 12 was prepared by a procedure
similar to that described for compound 11, except that compound
9 was used instead of compound 8. The product was obtained as
pale-yellow oil without any further purification in quantitative
yield. 1H NMR (CDCl3, 300 MHz): d 8.78 (2H, d, J = 2.05), 7.43
(1H, d, J = 1.20), 7.29 (2H, d, J = 2.02), 6.38 (1H, m), 4.38 (1H,
m), 3.94-3.73 (9H, m), 2.53 (2H, m), 2.24 (1H, m), 1.98 (1H, m),
1.92 (3H, d, J = 1.14), 1.57 (4H, m, br), 1.29–1.23 (12H, m, br),
0.91 (18H, d, J = 9.86), 0.08 (12H, m).
Synthesis of 3-(12-(bis(thiazol-4-ylmethyl)amino)dodecyl)-1-
((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-
5-methylpyrimidine-2,4(1H,3H)-dione (17). Compound 17 was
prepared and purified by a procedure similar to that described
for compound 15, except that compound 13 was used instead of
Synthesis of 3-(12-(bis(thiazol-4-ylmethyl)amino)dodecyl)-1-
((2R,4S,5R)-4-(tert -butyldimethylsilyloxy)-5-((tert -butyldime-
thylsilyloxy)methyl)tetrahydrofuran-2-yl)-5-methylpyrimidine-2,
4(1H,3H)-dione (13). Compound 13 was prepared by a pro-
cedure similar to that described for compound 11, except that
compound 10 was used instead of compound 8. The reaction
mixture was purified by basic alumina oxide chromatography
using chloroform as eluent to give 13 as a pale-yellow oil in 95%
1
compound 11. Yield 87%. H NMR (CDCl3, 300 MHz): d 8.96
(2H, d, J = 2.04), 7.84 (1H, d, J = 1.14), 7.51 (2H, d, J = 2.02),
6.32 (1H; m), 4.42 (1H, m), 3.92–3.73 (8H, m), 3.62 (1H, m), 2.51
(2H, m), 2.27 (2H, m), 1.90 (3H, d, J = 0.97), 1.57 (4H, m, br),
1.25 (16H, m, br).
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6216–6225 | 6223
©

Synthesis of 3-(12-(bis(quinolin-2-ylmethyl)amino)dodecyl)-1-
((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-
5-methylpyrimidine-2,4(1H,3H)-dione (18). Compound 18 was
prepared and purified by a procedure similar to that described
for compound 15, except that compound 14 was used instead of
30.60, 30.48, 28.66, 28.07, 28.01, 26.35, 13.37. ESI-MS m/z (%):
+
calcd. for C33H45N5O8ReS2 [M+] 890.23, found 889.99 (100%).
Synthesis of [Re(CO)3(18)]Br (22). The product was puri-
fied by semi-preparative C18 HPLC using a gradient of water–
1
methanol. Yield: 80%. H NMR (d4-MeOD, 500 MHz): d 8.56
1
compound 11. Yield 89%. H NMR (CDCl3, 300 MHz): d 8.20
(4H, t, J = 8.48), 8.03 (2H, dd, J = 1.36, 8.14), 7.88 (2H, dt, J =
1.47, 7.04, 8.64), 7.84 (1H, d, J = 1.19), 7.70 (2H, t, J = 7.13), 7.62
(2H, d, J = 8.46), 6.30 (1H, t, J = 6.70), 5.24 (2H, d, J = 17.83),
5.05 (2H, d, J = 17.81), 4.40 (1H, m), 3.90–3.70 (7H, m), 2.23 (2H,
(2H, d, J = 8.63), 8.18 (2H, d, J = 8.50), 7.83 (4H, t, J = 8.02), 7.73
(2H, m), 7.53 (2H, m), 7.49 (1H, s), 6.36 (1H; m), 4.48 (1H, m),
3.95–3.76 (8H, m), 3.66 (1H, m), 2.54 (2H, m), 2.32 (2H, m), 1.92
(3H, d, J = 1.05), 1.60 (4H, m, br), 1.28 (16H, m, br).
m), 2.02 (2H, m), 1.89 (3H, d, J = 1.12), 1.56–1.30 (20H, m). 13
C
NMR (d4-MeOD, 500 MHz): d 195.54, 194.20, 166.62, 165.34,
152.31, 148.16, 142.98, 136.43, 134.12, 130.91, 129.82, 129.53,
129.46, 120.78, 110.72, 88.92, 87.07, 72.16, 70.07, 69.62, 62.80,
42.33, 41.32, 30.50, 30.44, 30.41, 30.30, 28.50, 27.90, 27.85, 27.45,
13.23. ESI-MS m/z (%): calcd. for C33H45N5O8ReS2+ [M+] 978.35,
found 978.06 (100%).
Complex synthesis and characterization
The metal precursor [Re(CO)3(H2O)3]Br was prepared according
to the literature procedure.24 The rhenium dT complexes were
prepared by a similar procedure. A representative synthesis is
described for the complex formation of ligand 15.
In vitro assays
Synthesis
of
[Re(CO)3(15)]Br
(19). The
precursor
[Re(CO)3(H2O)3]Br (100 mg, 0.25 mmol) and ligand 15
(139 mg, 0.25 mmol) were mixed in 10 mL methanol and stirred
at 60 ◦C for 2 h. The solvent was removed under reduced pressure
and the crude complex purified using a C18 Sep-Pak cartridge.
The sample was first desalted by eluting with 20 mL water. The
side products were eluted with 10 mL methanol–water (2 : 8)
followed by 10 mL methanol–water (1 : 1). Finally, the product 19
was eluted with 20 mL methanol–water (8 : 2). After evaporation
of the solvents, 0.20 g (88%) of 19 were obtained as pale-yellow
powder. 1H NMR (d4-MeOD, 500 MHz): d 9.43 (2H, d, J = 2.08),
7.81 (1H, d, J = 1.18), 7.62 (2H, d, J = 1.58), 6.28 (1H, t, J = 6.50),
4.63 (2H, d, J = 15.61), 4.50 (2H, d, J = 16.06), 4.38 (1H, m),
3.89 (3H, m), 3.80–3.70 (4H, m), 2.21 (2H, m), 1.94 (2H, m), 1.88
(3H, d, J = 1.14), 1.58 (2H, m), 1.44 (4H, m), 1.35 (8H, m). 13C
NMR (d4-MeOD, 500 MHz): d 196.92, 194.78, 165.55, 161.27,
157.10, 152.51, 136.61, 118.64, 110.90, 89.09, 87.26, 72.34, 71.96,
62.98, 62.14, 42.39, 41.47, 30.24, 30.08, 28.52, 27.84, 26.19, 13.38.
ESI-MS m/z (%): calcd. for C29H37N5O8ReS2+ [M+] 834.16, found
833.95 (100%).
Cell lines and culture conditions. The A549 lung cancer cell line
(ATCC CCL-185) was cultured as adherent monolayers in F-12
K media. The F-12 K media was supplemented with penicillin
(10 000 U), 10 mg mL-1 streptomycin (Sigma), and 10% (v/v)
FBS (Sigma). Prior to testing, cells were then passed in F12 K
media and penicillin–streptomycin solution. Cells were incubated
and grown in a VWR mammalian incubator at 5% CO2 and 95%
humidity. Cells were cultured in Millipore 250 mL culture bottles
with vented lids.
Cell cytotoxicity. The proliferation of the exponential-phase
cultures of A549 cells was assessed by Cell Counting Kit 8
(CCK8) colorimetric assay. Adherent cell cultures were harvested
by stripping of culture flasks by using a non-enzymatic cell stripper
after a 30 min incubation period. The cells were counted and
plated, 8000 cells per well, into a 96-well plate. After a 24 h
incubation time to facilitate adherence, the media was removed
and replaced with 200 mL fresh media that contained various
concentrations of the dThd–rhenium complexes. The cells were
then incubated for 48 h. After incubation, the drug-containing
media was removed and replaced with fresh media containing
10% CCK8 solution. The plate was incubated for 1.5 h. The
absorbance at l 450 nm was measured by using a plate reader.
The cell viability was expressed and plotted as the percent of the
untreated control microcultures. The IC50 values were calculated
based on an exponential fit by using Prism Graphpad 5 software
with R2 values >0.90 in all cases. All experimental points were
measured in triplicate, and each experiment was performed at
least three times.
Synthesis of [Re(CO)3(16)]Br (20). The product was purified
by Sep-Pak cartridges. Yield: 90%. 1H NMR (d4-MeOD, 500
MHz): d 9.40 (2H, d, J = 2.08), 7.84 (1H, d, J = 1.20), 7.65
(2H, s), 6.30 (1H, t, J = 6.50), 4.68 (2H, d, J = 15.68), 4.53 (2H,
d, J = 15.75), 4.40 (1H, m), 3.92 (3H, m), 3.83–3.72 (4H, m),
2.23 (2H, m), 1.97 (2H, m), 1.90 (3H, d, J = 1.13), 1.63 (2H, m),
1.47 (6H, m), 1.38 (2H, m). 13C NMR (d4-MeOD, 500 MHz):
d 196.89, 194.86, 165.56, 161.26, 157.10, 152.50, 136.59, 118.59,
110.89, 89.07, 87.26, 72.31, 71.98, 62.96, 62.12, 42.48, 41.47, 30.48,
30.36, 28.61, 28.01, 27.93, 26.30, 13.37. ESI-MS m/z (%): calcd.
for C31H41N5O8ReS2 [M+] 862.20, found 861.88 (100%).
+
Confocal microscopy. 200,000 cells were plated onto MatTek
35 mm glass bottom culture dishes and incubated overnight.
The media was then replaced with PBS containing 20 mM of
complex 22. The compound was incubated for 1 h. The cells
were then washed with media and viewed under a Zeiss LSM
710 Pascal confocal microscope with Zen 2009 image analysis
software equipped with argon ion and HeNe lasers. Fluores-
cence was observed with a 63X objective with an excitation at
488 nm. A control plate was used to ensure that no background
fluorescence was observed (from flavins for example) under the
final test conditions used. Uptake at 4 ◦C was examined by
Synthesis of [Re(CO)3(17)]Br (21). The product was purified
by Sep-Pak cartridges. Yield: 89%. 1H NMR (d4-MeOD, 500
MHz): d 9.42 (2H, d, J = 2.06), 7.84 (1H, d, J = 1.06), 7.65
(2H, s), 6.31 (1H, t, J = 6.72), 4.67 (2H, d, J = 15.61), 4.52 (2H,
d, J = 15.81), 4.40 (1H, m), 3.91 (3H, m), 3.82–3.72 (4H, m),
2.23 (2H, m), 1.96 (2H, m), 1.91 (3H, d, J = 0.96), 1.60 (2H, m),
1.45 (4H, m), 1.34 (12H, m). 13C NMR (d4-MeOD, 500 MHz):
d 196.90, 194.77, 165.56, 161.26, 157.09, 152.49, 136.59, 118.62,
110.89, 89.07, 87.23, 72.31, 71.98, 62.96, 62.13, 42.50, 41.48, 30.68,
6224 | Dalton Trans., 2011, 40, 6216–6225
This journal is
The Royal Society of Chemistry 2011
©

pre-cooling the cells and drug solution for 30 min prior to addition.
After addition, the cells were placed back at 4 ◦C for 1 h incubation
period. The cells were washed with cold media and viewed under
the same conditions as above.
Thymidine challenge was performed at 37 C with a 100- fold
excess of thymidine over complex 22.
9 W. Plunkett and V. Gandhi, Drug Dispos., 2002, 41, 93.
10 P. L. Bonate, L. Arthaud Jr., K. Stephenson, J. A. Secrist III and S.
Weitman, Nat. Rev. Drug Discovery, 2006, 5, 855–863.
11 S. Fanderl, V. Gandhi, M. J. Keating, S. Jeha, W. Plunkett and H. M.
Kantarjian, Cancer, 2005, 103, 1985–1995.
◦
12 J. M. Lynam, Dalton Trans., 2008, 4067–4078.
13 N. Esho, B. Davies, J. Lee and R. Dembinski, Chem. Commun., 2002,
332–333.
14 C. D. Sergeant, I. Ott, A. Sniady, S. Meneni, R. Gust, A. L. Rheingold
and R. Dembinski, Org. Biomol. Chem., 2008, 6, 73–80.
15 M. Hocek, P. Stepnicka, J. Luv´ık, I. C´ısarova´, I. Votruba, D. Reha and
P. Hobza, Chem.–Eur. J., 2004, 10, 2058–2066.
16 M. Vra´bel, R. Pohl, B. Klepeta´rova´, I. Votruba and M. Hocek, Org.
Biomol. Chem., 2007, 5, 2849–2857.
17 W. Wang, Y. K. Yan, A. T. S. Hor, J. J. Vittal, J. R. Wheaton and I. H.
Hall, Polyhedron, 2002, 21, 1991–1999.
18 D.-L. Ma, C.-M. Che, F.-M. Siu, M. Yang and K.-Y. Wong, Inorg.
Chem., 2007, 46, 740–749.
Endosomal co-localization. The endosomal study was con-
ducted using Invitrogen organelle lights kit Endosomal-RFP
(O36231) following the manufacturer’s instructions. Briefly,
200,000 cells were plated on a 33 mm glass bottom dish, and
allowed to incubate over night. The media was then replaced with
the Endosomes-RFP reagent diluted to working concentration
with PBS containing 20 mM EDTA. The cells were incubated
at rt for 4 h. The solution was then removed and replaced with
media containing the BacMam enhancer at suggested dilution,
19 D. K. Orsa, G. K. Haynes, S. J. Pramanik, M. O. Iwunze, G. E. Greco,
D. M. Ho, J. A. Krause, D. A. Hill, R. J. Williams and S. K. Mandal,
Inorg. Chem. Commun., 2008, 11, 1054–1056.
◦
and incubated at 37 C overnight. The media was then replaced
20 D.-L. Ma, C.-M. Che, F.-M. Siu, M. Yang and K.-Y. Wong, Inorg.
Chem., 2007, 46, 740–749.
21 M. Bartholoma¨, A. Vortherms, S. Hillier, B. Ploier, J. Joyal, J. Babich,
R. P. Doyle and J. Zubieta, ChemMedChem, 2010, 5, 1513–1529.
22 N. Viola-Villegas, A. E. Rabideau, J. Cesnavicious, J. Zubieta and R. P.
Doyle, ChemMedChem, 2008, 3, 1387–1394.
23 M. Bartholoma¨, J. Valliant, K. P. Maresca, J. Babich and J. Zubieta,
Chem. Commun., 2009, 493–512.
24 N. Lazarova, S. James, J. Babich and J. Zubieta, Inorg. Chem. Commun.,
2001, 4, 1023–1026.
with PBS containing 20 mM of complex 22. The tagged endosomes
were excited using a 543 nm laser.
Thymidine kinase competition assay. The recombinant enzyme
human TK-1 was obtained from Baffin GmbH (Germany). An
amount of 15 ng of hTK-1 was added to reaction mixtures
consisting of 0.19 M Tris (pH 7.6), 1.9 mM MgCl2, 1.9 mM
adenosine triphosphate, 10 mM dithiothreitol, 10 mM NaF,
3 mM phosphocreatine, 0.54 units of creatinine phosphokinase,
25 R. Schibli, R. Schwarzbach, R. Alberto, K. Ortner, H. Schmalle, C.
Dumas, A. Egli and P. A. Schubiger, Bioconjugate Chem., 2002, 13,
750–756.
3
and 10 mM H-deoxythmidine (5 mCi, Perkin Elmer, Waltham,
MA) in the presence of 0–500 mM test compound. Individual
mixtures were incubated at 37 ◦C for 15 min. To sequester
labeled nucleotides, samples were spotted onto Whatman DE-
81 filters, dried, and then washed in 4 mM ammonium formate
twice and once with ethanol to remove nonphosphorylated
substrate. Activity captured by the filters was measured by liquid
scintillation counting on a Wallac 1450 Trilux Microbeta (Perkin
Elmer, Watham, MA). Inhibition of hTK-1 was analyzed by
normalization to the reaction containing 0 mM test compound.
26 G. Jaouen, S. Top, A. Vessieres and R. Alberto, J. Organomet. Chem.,
2000, 600, 23–36.
27 L. Wei, J. W. Babich, W. Ouellette and J. Zubieta, Inorg. Chem., 2006,
45, 3057–3066.
28 A. Monks, D. A. Scudiero, P. Skehan, R. H. Shoemaker, K. D. Paull, D.
T. Vistica, C. Hose, J. Langley, P. Cronice, M. Vaigro-Wolf, M. Gray-
Goodrich, H. Campbell and M. R. Mayo, J. Natl. Cancer Inst., 1991,
83, 757–766.
29 S. H. Van Rijt, A. F. A. Peacock, R. D. L. Johnstone, S. Parsons and P.
J. Sadler, Inorg. Chem., 2009, 48, 1753–1762.
30 A. F. A. Peacock, A. Habtemariam, S. A. Moggach, A. Prescimone, S.
Parsons and P. J. Sadler, Inorg. Chem., 2007, 46, 4049–4059.
31 F. Li, S. Awale, Y. Tezuka and S. Kadota, Bioorg. Med. Chem., 2008,
16, 5434–5440.
Acknowledgements
32 A. F. A. Peacock, S. Parsons and P. J. Sadler, J. Am. Chem. Soc., 2007,
129, 3348–3357.
33 S. Gomez-Ruiz, G. N. Kaluderovic, D. Polo-Ceron, V. Tayurskaya, S.
Prashar, M. Fajardo and R. Paschke, J. Organomet. Chem., 2009, 694,
3032–3038.
Mark Bartholoma¨ thanks Molecular Insight Pharmaceuticals,
Inc. of Cambridge, MA for a postdoctoral fellowship. This work
was supported by funding from Syracuse University.
34 J. Gao, J. H. Reibenspies and A. E. Martell, Org. Biomol. Chem., 2003,
1, 4242–4247.
References
35 J. Gao, Y.-G. Liu, Y. Zhou and R. A. Zingaro, ChemMedChem, 2007,
2, 1723–1729.
1 C. M. Galmarini, F. Popwycz and B. Joseph, Curr. Med. Chem., 2008,
15, 1072–1082.
2 W. Parker, Chem. Rev., 2009, 109(7), 2880–2893.
3 O. Olivero, Environ. Mol. Mutagen., 2007, 48, 215–223.
4 M. D. Lynx, A. T. Bentley and E. E. McKee, Biochem. Pharmacol.,
2006, 71, 1342–1348.
5 J. Balzarini, P. Herdewijn and E. De Clerq, J. Biol. Chem., 1989, 264(11),
6127–6133.
6 I. Huber-Ruano and M. Pastor-Anglada, Curr. Drug Metab., 2009, 10,
347–358.
7 M. Molina-Arcas and M. Pastor-Anglada, Nucleosides, Nucleotides
Nucleic Acids, 2010, 29, 335–346.
8 D. Sampath, V. A. Rao and W. Plunkett, Oncogene, 2003, 22, 9063–
9074.
36 H. Struthers, D. Viertl, M. Kosinski, B. Springler, F. Buchegger and R.
Schibli, Bioconjugate Chem., 2010, 21, 622–634.
37 D. Desbouis, H. Struthers, V. Spiwok, T. Ku¨ster and R. Schibli, J. Med.
Chem., 2008, 51, 6689–6698.
38 Y. Byun, J. H. Yan, A. S. Al-Madhoun, J. Johnsamuel, W. L. Yang, R. F.
Barth, S. Eriksson and W. Tjarks, J. Med. Chem., 2005, 48, 1188–1198.
39 S. Eriksson, B. Munch-Petersen, K. Johansson and H. Eklund, Cell.
Mol. Life Sci., 2002, 59, 1327–1346.
40 M. Welin, U. Kosinka, N.-E. Mikkelsen, C. Carnrot, C. Zhu, L. Wang,
S. Eriksson, B. Munch-Petersen and H. Eklund, Proc. Natl. Acad. Sci.
U. S. A., 2004, 101, 17970–17975.
41 J. S. Rasey, J. R. Grierson, L. W. Wiens, P. D. Kolb and J. L. Schwartz,
J. Nucl. Med., 2002, 43, 1210–1217.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6216–6225 | 6225
©