Half-Sandwich Ruthenium(II) Biotin Conjugates as Biological Vectors to Cancer Cells

Article abstract of DOI:10.1002/chem.201403974

Ruthenium(II)-arene complexes with biotin-containing ligands were prepared so that a novel drug delivery system based on tumor-specific vitamin-receptor mediated endocytosis could be developed. The complexes were characterized by spectroscopic methods and their in vitro anticancer activity in cancer cell lines with various levels of major biotin receptor (COLO205, HCT116 and SW620 cells) was tested in comparison with the ligands. In all cases, coordination of ruthenium resulted in significantly enhanced cytotoxicity. The affinity of RuII-biotin complexes to avidin was investigated and was lower than that of unmodified biotin. Hill coefficients in the range 2.012-2.851 suggest strong positive cooperation between the complexes and avidin. To estimate the likelihood of binding to the biotin receptor/transporter, docking studies with avidin and streptavidin were conducted. These explain, to some extent, the in vitro anticancer activity results and support the conclusion that these novel half-sandwich ruthenium(II)-biotin conjugates may act as biological vectors to cancer cells, although no clear relationship between the cellular Ru content, the cytotoxicity, and the presence of the biotin moiety was observed.

Full text of DOI:10.1002/chem.201403974

DOI: 10.1002/chem.201403974
Full Paper
&
Drug Discovery
Half-Sandwich Ruthenium(II) Biotin Conjugates as Biological
Vectors to Cancer Cells
[a, b, f]
[c]
[b, d]
[f]
Maria V. Babak,
Damian Pla z˙ uk, Samuel M. Meier,
Homayon John Arabshahi,
[f]
[e]
[e]
[e]
Jꢀhannes Reynisson, Bła z˙ ej Rychlik, Andrzej Błau z˙ , Katarzyna Szulc,
[f]
[a]
[a]
[a, b]
Muhammad Hanif, Sebastian Strobl, Alexander Roller, Bernhard K. Keppler,
and
[a, b, f]
Christian G. Hartinger*
Abstract: Ruthenium(II)–arene complexes with biotin-con-
taining ligands were prepared so that a novel drug delivery
system based on tumor-specific vitamin-receptor mediated
endocytosis could be developed. The complexes were char-
acterized by spectroscopic methods and their in vitro anti-
cancer activity in cancer cell lines with various levels of
major biotin receptor (COLO205, HCT116 and SW620 cells)
was tested in comparison with the ligands. In all cases, coor-
dination of ruthenium resulted in significantly enhanced cy-
Hill coefficients in the range 2.012–2.851 suggest strong pos-
itive cooperation between the complexes and avidin. To esti-
mate the likelihood of binding to the biotin receptor/trans-
porter, docking studies with avidin and streptavidin were
conducted. These explain, to some extent, the in vitro anti-
cancer activity results and support the conclusion that these
novel half-sandwich ruthenium(II)–biotin conjugates may act
as biological vectors to cancer cells, although no clear rela-
tionship between the cellular Ru content, the cytotoxicity,
and the presence of the biotin moiety was observed.
II
totoxicity. The affinity of Ru –biotin complexes to avidin was
investigated and was lower than that of unmodified biotin.
Introduction
ment of ulcers. As a result, a number of patients refuse or stop
[1,2]
adjuvant chemotherapy.
Therefore, a key feature of new
The most frequently used chemotherapeutic agents indiscrim-
inately target rapidly dividing cells, such as cancer cells, but
also healthy epithelial cells. This leads to low selectivity and
severe side effects, such as nausea, loss of hair, and develop-
chemotherapeutic agents is selectivity for cancerous cells over
healthy tissue. This can be achieved by exploiting intrinsic dif-
ferences between healthy and tumor cells. One strategy is to
take advantage of the fact that cancerous cells overexpress
various receptors on their surface that are less prevalent in
healthy cells. If an anticancer agent can selectively interact
with one of those receptors, a cytotoxin may be accumulated
more effectively in tumor cells. Consequently, the therapeutic
effect will be enhanced and adverse effects will be reduced.
Cancer cells require significant amounts of vitamins to sus-
tain their rapid growth. Moreover, there is a strong correlation
between the level of over-expression of vitamin receptors and
the stage of tumor growth, with the highest levels for tumors
[a] Dr. M. V. Babak, S. Strobl, A. Roller, Prof. Dr. B. K. Keppler,
Prof. Dr. C. G. Hartinger
Institute of Inorganic Chemistry, University of Vienna
Waehringer Strasse 42, 1090 Vienna (Austria)
[b] Dr. M. V. Babak, Dr. S. M. Meier, Prof. Dr. B. K. Keppler,
Prof. Dr. C. G. Hartinger
Research Platform, “Translational Cancer Therapy Research”
University of Vienna, Waehringer Strasse 42, 1090 Vienna (Austria)
[
c] Dr. D. Pla z˙ uk
Department of Organic Chemistry, Faculty of Chemistry
University of Łꢀd z´ , Tamka 12, 91-403 Łꢀd z´ (Poland)
[3]
at the terminal stage. The attachment of vitamins, such as vi-
tamin B12, folic acid, and biotin to anticancer prodrugs is
[d] Dr. S. M. Meier
[4]
Institute of Analytical Chemistry, University of Vienna
Waehringer Strasse 38, 1090 Vienna (Austria)
therefore a worthwhile strategy to enhance tumor targeting.
Various cancer cell types express higher levels of biotin recep-
tors than of folate or vitamin B12 receptors and consequently
display higher levels of uptake of biotin-modified molecules. It
was reported that biotin derivatives were also more effective
in killing cancer cells, which makes biotin a particularly promis-
[
e] Dr. B. Rychlik, Dr. A. Błau z˙ , K. Szulc
Cytometry Laboratory, Department of Molecular Biophysics
Faculty of Biology and Environmental Protection
University of Łꢀd z´ , 12/16 Banacha St., 90-237 Łꢀd z´ (Poland)
[f] Dr. M. V. Babak, H. J. Arabshahi, Dr. J. Reynisson, Dr. M. Hanif,
[5]
Prof. Dr. C. G. Hartinger
ing vector.
School of Chemical Sciences, University of Auckland
Private Bag 92019, Auckland 1142 (New Zealand)
E-mail: c.hartinger@auckland.ac.nz
The main biotin uptake system in human intestinal epithelial
cells is the sodium-dependent multivitamin transporter (SMVT)
system. SMVT is a protein (635 amino acids) encoded by the
SLC5A6 gene, which was found to be activated in various ag-
Homepage: http://hartinger.wordpress.fos.auckland.ac.nz/
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201403974.
[6,7]
gressive cancer cell lines.
Consequently, biotinylation of
Chem. Eur. J. 2015, 21, 1 – 9
1
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
compounds converts them into biological vectors to SMVT-
overexpressing cells. This has been the subject of numerous
studies with organic drugs and drug candidates (camptothe-
[
8]
[5,9]
[5]
[10–12]
cin, metothrexate,
doxorubicin, paclitaxel,
gemcita-
and TAT
[
13,14]
[15–17]
bine,
peptides).
polyamidoamine dendrimers (PAMAM),
[
18,19]
However, the targeted delivery of analogous
metal-based compounds remains largely unexplored. Although
a number of metal-based biotin compounds have been report-
ed, only a few have been tested for their biological properties
in cells. Lo and co-workers evaluated the anticancer activity of
[
20,21]
[22,23]
[23]
Re,
Ir,
and Rh complexes with a range of polyaro-
II
matic diimine ligands with a differing number of attached
biotin moieties. The cytotoxicity of the complexes was strongly
dependent on the number of pendant biotin groups, with
lower IC50 values corresponding to a lower number of biotin
Scheme 1. Biotinylation of half-sandwich Ru complexes (anhydrous DMF,
6
2 2
ratio of [{Ru(h -p-cymene)Cl } ] to ligand=1:2, 3 h, RT).
were used. For comparison, the structurally similar ligands 4
and 5, with an incorporated aminohexanoyl linker, were pre-
pared. Published data on the effect of the spacer between the
active component and targeting moiety on the biological ac-
tivity of compounds is controversial. The incorporation of an
aminohexanoyl spacer between biotin and a ferrocene com-
[
21]
residues. However, nonbiotinylated analogues with a similar
structure exhibited the same order of cytotoxicity as the bio-
tinylated derivatives. Similarly, the cytotoxicity of a biotin-ap-
6
pended RAPTA complex ([Ru(h -arene)Cl (PTA)], in which PTA is
2
1
,3,5-triaza-7-phosphaadamantane) was not higher than of un-
6
[24]
[25]
modified RAPTA-C, [Ru(h -p-cymene)Cl (PTA)].
In contrast,
plex had a detrimental effect on the cytotoxicity. Other stud-
2
[
25]
biotin–ferrocene conjugates
and biotinylated cisplatin-
ies revealed that such spacers dramatically increase the stabili-
[
15]
II
loaded nanoparticles showed a significant increase of cyto-
toxicity compared with the respective nonbiotinylated deriva-
tives. However, much of the data was collected on different
cell lines with different affinity and capacity of the SMVT trans-
ty of Ru (polypyridine)–biotin complexes and their interactions
[28]
with cell membranes. Moreover, it was demonstrated that
spatial separation between biotin and the metal-based com-
[29]
plex is vital for biorecognition of both units.
[
26]
porter.
Compounds of the biotin–linker–N-heterocycle type, such as
2–5, have been used extensively for a range of purposes and
their syntheses are well established. To synthesize ruthenium
complexes with these ligands, it is essential that they do not
contain any trace of unreacted precursors, which could also co-
ordinate to the ruthenium fragment. Therefore, all ligands
were additionally purified by reversed phase HPLC, and their
purity was confirmed by analytical HPLC and elemental analy-
sis. However, this resulted in low yield of the target com-
pounds. 6-(Methylamino)indazole can be used for coupling
with biotin derivatives without HPLC purification. All ligands
Herein, we present the results of a systematic investigation
II
of novel biotin-conjugated half-sandwich Ru compounds in
terms of their antiproliferative activity in cancer cells. We have,
in particular, aimed to evaluate the correlation between the cy-
totoxicity, ruthenium compounds’ uptake, and the SLC5A6
gene (SMVT) expression, as the latter is an important parame-
ter for biotin to function as a vector. These studies were com-
plemented by the determination of the binding affinity of the
biotin derivatives to avidin and by molecular docking experi-
ments.
1
13
1
were characterized by H and C{ H} NMR (for the atom num-
bering scheme, see Figure 4 in the Experimental Section) as
well as ESI-MS and elemental analysis.
Results and Discussion
To test the effect of using ruthenium–biotin conjugates on the
viability of cancer cells and to assess the correlation with levels
of SLC5A6 gene expression, we synthesized biotin-containing li-
gands 2–5, the corresponding half-sandwich ruthenium(II)
complexes 7, 8, 10, and 11, and the structurally similar nonbio-
tinylated ruthenium(II) analogues 6 and 9 (Scheme 1). The
SMVT transporter can recognize biotin when the thiophane
To coordinate the biotin ligands to the Ru center, we adapt-
ed the procedure of Vock et al., who developed the synthesis
II
of half-sandwich Ru complexes with a number of N-heterocy-
[30]
clic ligands. By following this method, the nonbiotinylated
II
6
[Ru (h -p-cymene)] complexes with pyridine (6) and indazole
6
(9) ligands were obtained by heating [{Ru(h -p-cymene)Cl } ] to
2
2
reflux with the corresponding ligand in anhydrous toluene for
[
25,27]
II
and keto fragments of the latter are not modified.
There-
3 h. However, the analogous synthesis of Ru –biotin conjugates
fore, these moieties remained unaltered, and the valeric car-
boxylic acid residue of biotin was used for further derivatiza-
tion. Consequently, 2 and 3 were prepared by standard amide
coupling of the biotin targeting unit with primary amino
groups of N-heterocycles to allow coordination to the metal
center. 4-Aminopyridine and 6-aminoindazole were found to
be inappropriate for such coupling and therefore commercially
available 4-(aminomethyl)pyridine and 6-(aminomethyl)ind-
azole from the reduction of 6-cyano-1H-indazole with LiAlH4
was seriously limited by the solubility of the biotin derivatives
and by the stability of the corresponding complexes. Ligands
2–5 were moderately soluble in ethanol and effectively solvat-
ed only in dimethylformamide (DMF) and dimethyl sulfoxide.
We have recently discovered that incubation of half-sandwich
II
arene Ru complexes in dimethyl sulfoxide resulted in loss of
[24]
the arene moiety. Therefore, we investigated the reactions of
6
[{Ru(h -p-cymene)Cl } ] with 2 equiv of ligands 2–5 in ethanol
2
2
or DMF by means of ESI-MS. In the present study, ethanol was
&
&
Chem. Eur. J. 2015, 21, 1 – 9
www.chemeurj.org
2
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
a less suitable solvent for such complexation reactions due to
The molecular structure of the nonbiotinylated complex 6
was determined by X-ray diffraction analysis (Figure S6; for
crystallographic data see Table S1). Single crystals were grown
by slow diffusion of diethyl ether into a saturated dichlorome-
thane solution at 277 K. The p-cymene ring is coordinated to
the marked formation of dinuclear Ru species, especially for 10
(for the mass spectrum see Figure 1) and 11 (Figure S1). The
formation of ruthenium complexes 7, 8, 10, and 11 was more
selective in DMF (see Figure 1 and S1–S5), although either di-
6
the ruthenium center in a h -manner. The rest of the coordina-
tion sphere is filled by two chlorido and a pyridine ligand. The
geometrical parameters of the complex are in agreement with
previously determined structures of compounds, which vary
[32,33]
only in the nature of the N-donor ligands.
The average
RuÀC bond length in 6 is 2.186 ꢂ, whereas RuÀCl1, RuÀCl2 and
RuÀN distances are 2.4194(4), 2.4026(4), and 2.1368(12) ꢂ, re-
6
spectively (in comparison for [Ru(h -p-cymene)Cl (NH )], RuÀC
2
3
av
2
2
.170(8), RuÀCl1 2.4157(6), RuÀCl2 2.4157(6), and RuÀN
[32]
.130(3) ꢂ). The crystal structure of 6 was employed as the
starting point for subsequent docking studies.
Figure 1. Full ESI-IT mass spectra of reaction mixtures giving 10 after 3 h.
The reactions were carried out in DMF or in EtOH. The reaction using DMF
as a solvent leads to more selective formation of 10.
Cytotoxicity and cellular accumulation
The in vitro anticancer activity of the ruthenium complexes
and ligands was determined in COLO205, SW620, and HCT116
colon carcinoma cells by means of the colorimetric MTT assay
with an exposure time of 70 h (see Table 1 for the IC50 values;
meric hydrolysis products or DMF adducts accompanied by
ligand release were detected for the complexes without ami-
nohexanoic spacers; that is, 7 and 10, respectively. The detec-
tion of complexes with coordinated monodentate N-donor li-
gands can be challenging with ESI-MS and indicates limited
[
31,32]
complex stability under the ESI spraying conditions.
How-
Table 1. Cytotoxicity of the Ru complexes 6–11 in COLO205 (colon ade-
nocarcinoma), HCT116 (colon carcinoma), and SW620 (colon adenocarci-
noma) cells determined by means of the MTT assay. Calculations are
based on results of three independent experiments. 95% confidence in-
tervals are given in parentheses to enable better comparison of the re-
sults.
ever, this does not reflect the situation in solution. Interesting-
ly, such ions were not observed for the complexes with amino-
hexanoic spacers (i.e., 8 and 11). These findings suggest that
once the ligand is coordinated to the ruthenium center, the
spacer is beneficial for the stability of the resulting complexes.
Consequently, the complex stability increases when the ligand
contains indazole and/or a spacer.
Compound
IC50 [mm]
HCT116
COLO205
SW620
SMVT (low)
SMVT (low)
SMVT (high)
The reactions were analyzed by ESI-MS after 3, 6, and 24 h
6
7
8
ꢀ100
3.4 (2.2–5.1)
6.7 (4.6–9.6)
2.5 (1.4–4.5)
4.1 (2.3–5.5)
1.1 (0.8–1.7)
2.1 (1.3–3.4)
(
Figures S2–S5). Whereas 8, 10, and 11 did not reveal any
pyridine
indazole
4.6 (2.1–10.2)
17 (9–36)
changes within 24 h of incubation, complex 7 was significantly
less stable and started to decompose after 6 h (Figure S2).
Based on the results of the ESI-MS experiments, we modified
[
a]
9
10
11
23 (18–31)
ꢀ100
33 (18–60)
14 (12–17)
19 (15–24)
29 (22–37)
ꢀ100
6
the procedure of Vock et al. to stirring a mixture of [{Ru(h -p-
ꢀ100
6.7 (4.6–9.7)
cymene)Cl } ] with the corresponding ligand in anhydrous DMF
2
2
[
a] Extrapolated value.
at room temperature for 3 h with subsequent removal of DMF
by freeze-drying. This generic method was used for the synthe-
sis of biotinylated complexes 7, 8, 10, and 11 (Scheme 1). Non-
biotinylated 6 was also obtained by this method and its char-
concentration-effect curves are shown in Figure S7). These cell
lines were chosen because of their different SLC5A6 gene ex-
pression levels coding for SMVT. Therefore, the response of
cancer cells to the Ru–biotin conjugates may be correlated
with the relative level of SMVT expression. SLC5A6 mRNA is
most abundant in SW620 cells, and almost three times higher
than in COLO205 and HCT116 cells, which have comparable
[
30]
acterization was consistent with published data.
We as-
sumed that lyophilization of the products did not affect their
integrity, because the ESI-MS of the lyophilized and redissolved
complexes did not change. The novel complexes 7, 8, 10, and
11 were highly hygroscopic and moisture-sensitive, and pro-
[25]
longed exposure to air resulted in their decomposition. There-
fore, for subsequent biological studies, the lyophilized com-
plexes were incubated in situ with cell extracts. It was previ-
ously reported that addition of more than one biotin moiety
SLC5A6 gene expression levels.
With the exception of 7, COLO205 cells were markedly less
chemosensitive than HCT116 and SW620 to 6–11 (Table 1).
Complex 6 was reported to be inactive in TS/A murine adeno-
[
21]
[30]
to the substrate resulted in decreased cytotoxicity and an
carcinoma cells (IC =757 mm); however, it was significantly
50
[
19]
aggravated SMVT transporter recognition. Accordingly, only
a single biotin-functionalized ligand was coordinated to the
ruthenium center.
more active in the human cell lines HCT116 and SW620, with
IC50 values of 3.4 and 4.1 mm, respectively, although different
conditions were used in these experiments. In general, cancer
Chem. Eur. J. 2015, 21, 1 – 9
www.chemeurj.org
3
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
cells were more sensitive towards the pyridine complexes 6–8
than their indazole analogues 9–11. Whereas the activities of 6
and 8 were similar in cell lines with high and low levels of
SMVT expression, complexes 7 and 10 were significantly more
active in the SMVT-overexpressing SW620 cell line. In contrast,
troduced by any of the investigated compounds because inter-
actions among biotin-binding sites remain unchanged. The
recognition of biotin-appended complexes by avidin and SMVT
transporter indicates that thiophane and keto fragments of
[
25,27]
biotin ligands were not modified upon complexation.
11 was four times more active in the cell line with a low level
of SMVT expression (6.7 and 29 mm in HCT116 and SW620, re-
spectively). Interestingly, biotin derivatives with pyridine were
more potent in the SW620 cell line (IC =1.1 mm and 2.1 mm
Docking studies with strept(avidin)
To explain the findings from the cytotoxicity assays and to esti-
mate the likelihood of binding to the SMVT transporter, dock-
ing studies with crystal structures of avidin (PDB: 1LDQ) and
streptavidin (PDB: 3RY2) were conducted using the GOLD soft-
50
for 7 and 8, respectively) than the parent nonbiotinylated com-
pound 6 (IC =4.1 mm). However, biotin complexes with an
50
indazole fragment were less active in SW620 cells than their
nonbiotinylated analogue 9. From these results, no clear corre-
lations between the cytotoxicity, the SLC5A6 expression levels,
and the structures of the complexes can be drawn. Notably,
none of the corresponding ligands was toxic in the investigat-
ed concentration range (up to 30 mm). This is consistent with
previous investigations that showed that there is no clear
effect of biotin-conjugation on the antiproliferative activi-
[34]
ware. Goldscore (GS) was the only scoring function able to
treat ruthenium complexes. GS gives arbitrary numbers with
higher values predicting better binding. The docking studies
were conducted with streptavidin after initial experiments with
avidin. However, when docking 10 to avidin, the biotin moiety
was found to be situated outside the binding pocket of avidin,
thus giving no reason to suggest any binding between the
ligand and protein. The docking experiments of biotin with the
crystal structure of streptavidin gave a good overlap of mod-
eled biotin and co-crystallized biotin (GS 71), therefore this
was used as a model system.
[
15,21,24–26]
ty.
It seems that the activity of the investigated drug
itself also adds to the cytotoxicity of the conjugate.
The Ru content in the cancer cells was analyzed in depend-
ence of the supplementation with biotin (Tables S2–S4). A simi-
lar picture was obtained to that found in the cytotoxicity stud-
ies: no clear relationship was identified between the presence
of the biotin moiety and the concentration of Ru in the cells.
This supports the assumption that alternative pathways must
exist for cellular uptake of this compound class and that there
is possibly more than one uptake mechanism in place.
Docking of pyridine derivatives 6–8 to streptavidin was com-
pared with that of biotin. The biotin moieties of 7 and 8
showed good overlap with the co-crystallized biotin in the
streptavidin crystal structure, reproducing also the hydrogen-
bonding pattern. The top three results for 7 had scores of 70Æ
2 (Figure 2). This is comparable with the scoring of biotin, and
indicates similar binding energies of 7 and biotin. Compound
8
7
gave top scores of 80Æ1, predicting higher binding than for
Interactions with avidin
and biotin. This may be related to the increase in the molec-
ular size. The addition of the linker provides greater flexibility
to the ruthenium moiety, whereas the biotin was anchored
deep in the pocket. In contrast, the GS for the pyridine com-
plex 6 was only 36Æ1 and was found in the binding pocket
where biotin is normally located (Figure S8). The low score
coupled with the observed lack of hydrogen bonding suggests
The relative affinity of compounds 7, 8, 10, and 11 with avidin
was determined by using Biotective Green reagent, which is
a fluorescent derivative of avidin, complexed with 2-(4’-hy-
droxyazobenzene)benzoic acid as a quencher. The apparent
equilibrium dissociation constants for the investigated com-
pounds with avidin were found to be higher than for biotin,
which was used as a reference. Accordingly, the affinity of all
test compounds towards avidin was lower than for the natural
ligand. The affinity for pyridine derivatives was higher than for
the indazole analogues. Hill coefficients in the range of 2.012–
that the complex binding in the pocket is not favored. In addi-
6
tion, a model compound with the {Ru(h -p-cymene)Cl } frag-
2
ment coordinated to the biotin-sulfur atom was studied (Fig-
ure S9) and a score of 55Æ2 was obtained. The biotin moiety
of the top three docking results showed some overlap with
the co-crystallized biotin, resulting in a partial reproduction of
the hydrogen-bonding pattern. The top three hits also showed
good overlap with each other.
2.851 indicate positive cooperativity in the binding event
(Table 2). The values do not differ significantly among the com-
pounds tested. This suggests no major steric hindrance was in-
The same set of experiments was conducted with the ind-
azole-substituted biotin derivatives. For 10, the top three hits
showed overlap between each other, but not in the expected
position, resulting in no overlap with the co-crystallized biotin
in the protein structure (Figures 3 and S10). The biotin head of
the docked compounds was positioned outside the binding
pocket. The scoring for these molecules was 57Æ1, suggesting
less favorable binding in comparison to biotin or analogue 7.
Even more surprising was the result of the docking studies fea-
turing 11, for which the top three scores from the screening
showed no overlap at all. Due to the high level of flexibility,
Table 2. Apparent equilibrium dissociation constants and Hill coefficients
for investigated compounds. Calculations were based on results of at
least three independent experiments. 95% confidence intervals are given
in parentheses to enable better comparison of the results.
Compound
K
d
[nm]
Hill coefficient
biotin
pyridine
680 (610–751)
895 (823–967)
1124 (1050–1198)
1194 (1112–1275)
1980 (1144–2816)
2.752 (1.985–3.520)
2.851 (2.257–3.445)
2.785 (2.390–3.179)
2.371 (2.085–2.658)
2.012 (1.030–2.994)
7
8
10
indazole
11
&
&
Chem. Eur. J. 2015, 21, 1 – 9
www.chemeurj.org
4
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
showed much lower scores than especially 7 and 8 in the
streptavidin docking experiments, which is consistent with the
higher in vitro activity of the latter derivatives in the SMVT-
high cell line SW620. The same set of compounds showed
much lower activity in the SMVT-low cells and therefore in this
case the (strept)avidin binding ability may be indicative of an
SMVT-mediated mode of action. On the other hand, the ind-
azole derivatives also show significant cytotoxicity in some of
the cell lines, which may be explained with a mode of action
not involving the SMVT transporter. The favorable interaction
of the indazole derivatives with avidin in the in vitro experi-
ment may be related to the extended p-system of indazole
rather than the biotin moiety.
Conclusions
The high demand of tumors for vitamin H (biotin), which is re-
[3,4]
quired to sustain their rapid growth,
means that the tumor
cells overexpress biotin receptors (e.g., SMVT) on the cell sur-
face. This provides an opportunity to selectively accumulate
chemotherapeutics with high affinity for SMVT or other biotin
receptors by designing novel drug delivery systems. We linked
a biotin-receptor-targeting moiety through a spacer to a bio-
logically active metal fragment. Biotin derivatives featuring the
natural binding moiety to the biotin receptor were used as the
basis for the vector to exploit tumor-specific vitamin-receptor
mediated endocytosis. As a proof of concept, a series of half-
sandwich ruthenium(II) complexes with ligands functionalized
with biotin were prepared. Their biological activity was con-
firmed by cytotoxicity assays in cell lines with differing SMVT
expression profiles. Whereas the ligands were not toxic in the
investigated concentration range, the complexes exhibited
marked cytotoxicity, and 7 and 10 showed cell-specificity
based on the level of SMVT expression, although no direct cor-
relation with the cellular uptake was identified. The relative
binding affinity of the complexes to avidin was determined
and it was shown that all complexes were recognized by
avidin, although to a lower extent than biotin but still in the
same dimension. This indicates that the structural components
of biotin that are essential for recognition were not significant-
ly altered upon complexation. The interactions of the com-
plexes with (strept)avidin were investigated by docking studies
and considering the antiproliferative activity of the complexes
confirmed the likelihood of binding of the complexes to SMVT
transporter. However, other transport mechanisms might con-
tribute to Ru accumulation in the cell. Overall, the findings
suggest that the complexation of these biotin-functionalized li-
gands to Ru centers results in cytotoxic compounds with some
degree of selectivity for cancer cells with higher biotin receptor
levels.
Figure 2. A) Overlay of the docking results of 7 and 8 in the binding site of
streptavidin indicating overlap of the biotin moieties of both molecules, and
B) the hydrogen-bond network formed between the biotin moiety of 7 and
the amino acids Asp128, Asn23, Ser27, Tyr43 and Ser45 in streptavidin.
Figure 3. Complex 10 docked with streptavidin (the charge distribution sur-
face of the protein is shown) was predicted to be on the surface rather than
the biotin moiety in the biotin binding site.
the molecules were spread out in various positions. Similar to
6
, complex 9 occupied the binding pocket without any hydro-
Experimental Section
gen bonding. The scores for the top three configurations were
Synthesis
40Æ1, slightly higher than that of 6.
The results of the docking experiments explain to some
6-(Methylamino)indazole (1): LiAlH (12 mmol, 456 mg) was sus-
4
extent the in vitro anticancer activity data. Compounds 9–11
pended in anhydrous THF (80 mL) at 08C under argon. A solution
Chem. Eur. J. 2015, 21, 1 – 9
www.chemeurj.org
5
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
of 6-cyano-1H-indazole (3 mmol, 430 mg) in anhydrous THF (5 mL)
under argon atmosphere was cautiously added to this suspension
at 08C. After the color of the reaction mixture changed from light
yellow to dark brown, it was allowed to warm to RT and heated at
reflux for 4 h. The solution turned almost colorless and a light
yellow precipitate formed. The reaction mixture was cooled to 08C
and a minimum amount of water was added in 50 mL aliquots with
the last aliquot following termination of gas evolution. The suspen-
sion was filtered and the pale yellow solution was evaporated to
dryness and dried in vacuo. The crude product was purified by re-
versed-phase HPLC (H O/MeOH/0.1% CF COOH, gradient 5–95%
a white solid. The purity of the product was confirmed by analyti-
1
cal HPLC (retention time: 12.8 min). H NMR (500.10 MHz,
3
5
[D ]DMSO): d=8.50 (dd, J(H,H)=4.4 Hz, J(H,H)=1.6 Hz, 2H; H ),
6
v,s
3
3
5
8.41 (t, J(H,H)=5.9 Hz, 1H; NH), 7.24 (dd, J(H,H)=4.4 Hz, J(H,H)=
1.5 Hz, 2H; H ), 6.43 (s, 1H; NH-biotin), 6.37 (s, 1H; NH-biotin),
w,r
3
4.32 (m, 1H, H ), 4.29 (d, J(H,H)=5.9 Hz, 2H; H ), 4.14 (m, 1H; H ),
g
p
f
2
3
3.12 (m, 1H; H ), 2.84 (dd, J(H,H)=12.6 Hz, J(H,H)=5.1 Hz, 1H;
H ), 2.59 (d, J(H,H)=12.5 Hz, 1H; H ), 2.19 (t, J(H,H)=7.4 Hz, 2H;
Ha), 1.68–1.28 ppm (m, 6H; Hb,c,d);
e
2
3
h
h’
13
C NMR (125.75 MHz,
[D ]DMSO): d=172.9 (C), 163.2 (C ), 150.0 (C ), 149.2 (C ), 122.5
6
i
x
s,v
q
(C ), 61.5 (C ), 59.7 (C ), 55.9 (C ), 41.5 (C ), 40.4 (overlap with
2
3
r,w
f
g
e
p
MeOH, 20 min) followed by evaporation under reduced pressure to
DMSO) (Ch,h’), 35.6 (C ), 28.7 (Cb/c/d), 28.5 (Cb/c/d), 25.7 ppm (Cb/c/d);
MS (ESI+): m/z: 335.14 [M+H] (calcd 335.15); elemental analysis
a
+
give 1 (243 mg, 54%) as a white solid. The purity was assessed by
1
analytical HPLC (retention time 11.6 min). H NMR (500.10 MHz,
calcd (%) for C H N O S·1.1H O (354.25): C 54.25, H 6.89, N 15.82,
16
22
4
2
2
+
[
8
7
D ]DMSO): d=13.24 (s, 1H; NH-indazole), 8.22 (brs, 3H; NH ),
S 9.05; found: C 54.47, H 6.52, N 15.55, S 8.89
6
3
3
.11 (s, 1H; H ), 7.82 (d, J(H,H)=8.4 Hz, 1H; H ), 7.66 (s, 1H; H ),
u s w
.20 (d, J(H,H)=8.4 Hz, 1H; H ), 4.19 ppm (s, 2H; H ); MS (ESI+):
Biotinyl-(N-e-amidocaproic)-(6-methylamido)indazole (4): Biotin-
amidohexanoic acid N-hydroxysuccinimidyl ester (3; 545 mg,
3
r
p
+
+
m/z: 130.8 [MÀNH2] (calcd 131.1), 148.4 [M+H] (calcd 148.1).
1
.2 mmol) was dissolved in a minimum amount of DMF (ca.
Biotin(6-methylamido)indazole (2): Biotin N-hydroxysuccinimidyl
ester (240 mg, 0.7 mmol) and 6-(methylamino)indazole (1; 103 mg,
20 mL). Triethylamine (334 mL, 2.4 mmol, 242 mg) and 6-(methyla-
mino)indazole (1; 176 mg, 1.2 mmol) in DMF (10 mL) were added.
The mixture was stirred for 20 h at RT, filtered, and evaporated
under reduced pressure. The residue was dissolved in a minimum
amount of methanol and filtered. Diethyl ether (50 mL) was added
to the filtrate, and the precipitate was removed by filtration,
washed with diethyl ether, and dried in vacuo to give a pale
yellow solid. The crude product was purified by reversed phase
0
.7 mmol) were dissolved in
a minimum amount of DMF
(
ca. 10 mL). Triethylamine (292 mL, 212 mg, 2.1 mmol) was added to
the solution and the color changed to light yellow. The reaction
mixture was stirred overnight at RT. DMF was removed under re-
duced pressure and the residue was dissolved in a minimum
amount of methanol and filtered. Diethyl ether (50 mL) and petro-
leum ether (50 mL) were added to the filtrate. The formed precipi-
tate was collected by filtration, washed with petroleum ether, and
dried in vacuo to give a pale yellow powder. The crude product
was purified by reversed-phase HPLC (H O/MeOH/0.1% CF COOH,
HPLC (H O/MeOH/0.1% CF COOH, gradient 5–95% MeOH, 20 min)
2 3
followed by evaporation under reduced pressure to give 4
(128 mg, 22%) as a white solid. The purity of the product was con-
1
firmed by analytical HPLC (retention time 18.7 min). H NMR
2
3
gradient 5–95% MeOH, 20 min) followed by evaporation under re-
(500.10 MHz, [D ]DMSO; Figure 4): d=12.98 (s, 1H; NH-indazole),
6
duced pressure to obtain 2 (191 mg, 73%) as a white solid. The
purity of the product was confirmed by analytical HPLC (retention
1
time 17.8 min). H NMR (500.10 MHz, [D ]DMSO): d=12.97 (s, 1H;
6
3
NH-indazole), 8.36 (t, J(H,H)=6.1 Hz, 1H; NH), 8.02 (s, 1H; H ), 7.69
d, J(H,H)=8.4 Hz, 1H; H ), 7.38 (s, 1H; H ), 7.01 (d, J(H,H)=
s w
u
3
3
(
8
.4 Hz, 1H; H ), 6.41 (s, 1H; NH-biotin), 6.36 (s, 1H; NH-biotin), 4.39
r
3
(
d, J(H,H)=6.0 Hz, 2H; H ), 4.31 (m, 1H; H ), 4.13 (m, 1H; H ), 3.10
p g f
2
3
(
m, 1H; H ), 2.83 (dd, J(H,H)=12.5 Hz, J(H,H)=5.0 Hz, 1H; H ),
e h
2
3
2
1
.59 (d, J(H,H)=12.6 Hz, 1H; H ), 2.17 (t, J(H,H)=7.4 Hz, 2H; H ),
h’ a
.68–1.28 ppm (m, 6H; Hb,c,d); C NMR (125.75 MHz, [D ]DMSO):
13
6
d=172.5 (C), 163.2 (C ), 140.6 (C ), 138.4 (C ), 133.7 (C ), 122.3 (C ),
i
x
v
q
u
t
Figure 4. NMR numbering scheme used for 4.
1
20.8 (C ), 120.8 (C ), 108.5 (C ), 61.5 (C ), 59.7 (C ), 55.9 (C ), 40.4
r s w f g e
(
(
overlap with DMSO) (Ch,h’), 42.8 (C ), 35.7 (C ), 28.7 (Cb/c/d), 28.5
p a
19
3
3
Cb/c/d), 25.8 ppm (Cb/c/d); F NMR (470.56 MHz, [D ]DMSO): d=
8.36 (t, J(H,H)=5.9 Hz, 1H; NH), 8.02 (m, 1H; H ), 7.76 (t, J(H,H)=
6
u
+
3
À76.0 ppm (s, CF COOH); MS (ESI+): m/z: 374.45 [M+H] (calcd
5.6 Hz, 1H; NH), 7.69 (d, J(H,H)=8.3 Hz, 1H; H ), 7.37 (s, 1H; H ),
3
s
w
3
3
74.16); elemental analysis calcd (%) for C H N O S·1.25CF COOH
18 23 5 2 3
7.01 (d, J(H,H)=8.3 Hz, 1H; H ), 6.41 (s, 1H; NH-biotin), 6.35 (s, 1H;
r
3
(
1
516.00): C 47.72, H 4.74, N 13.57, S 6.21; found: C 47.58, H 4.89, N
3.58, S 6.42%.
NH-biotin), 4.39 (d, J(H,H)=5.9 Hz, 2H; H ), 4.31 (m, 1H; H ), 4.13
m, 1H; H ), 3.40 (m,1H; H ), 3.02 (dt, J(H,H)=12.8 Hz, J(H,H)=
p
g
3
3
(
f
e
2
3
6
2
2
.8 Hz, 2H; H ), 2.82 (dd, J(H,H)=12.4 Hz, J(H,H)=5.0 Hz, 1H; H ),
j h
Biotin(4-methylamido)pyridine (3): Biotin N-hydroxysuccinimidyl
ester (512 mg, 1.5 mmol) and 4-(aminomethyl)pyridine (243 mg,
2
1
2
3
.58 (d, J(H,H)=12.4 Hz, 1H; H ), 2.16 (t, J(H,H)=7.4 Hz, 2H; H ),
h’
a
3
.05 (t, J(H,H)=7.4 Hz, 2H; H ),1.67–1.28 (m, 12H; Hb,c,d,m,l,k);
C NMR (125.75 MHz, [D ]DMSO; Figure 4): d=172.6 (C ), 172.4 (C),
.25 mmol) were dissolved in a minimum amount of DMF (ca.
0 mL). Triethylamine (418 mL, 303 mg, 3 mmol) addition to the so-
n
1
3
6
o
i
1
1
63.2 (C ), 140.6 (C ), 138.4 (C ), 133.7 (C ), 122.3 (C ), 120.8 (C ),
x v q u t r
lution caused a color change to light yellow. The reaction mixture
was stirred overnight at RT. DMF was removed under reduced pres-
sure and the residue was subsequently washed with diethyl ether,
dissolved in a minimum amount of methanol, and filtered. Diethyl
ether (50 mL) and petroleum ether (50 mL) were added to the fil-
trate, and the precipitate was removed by filtration, washed with
petroleum ether, and dried in vacuo to give a pale yellow powder.
20.8 (C ), 108.4 (C ), 61.5 (C ), 59.7 (C ), 55.9 (C ), 40.4 (overlap
s
w
f
g
e
with DMSO) (Ch,h’), 42.7 (C
), 38.8 (C
p
), 35.8 (C
j
), 35.7 (C ), 29.5
n
a
(Cb/c/d/m/l/k), 28.7 (Cb/c/d/m/l/k), 28.5 (Cb/c/d/m/l/k), 26.6 (Cb/c/d/m/l/k), 25.8
19
(Cb/c/d/m/l/k), 25.6 ppm (Cb/c/d/m/l/k); F NMR (470.56 MHz, [D ]DMSO):
6
+
d=À75.2 ppm (s, CF
COOH); MS (ESI+): m/z: 486.93 [M+H]
S·H O·0.3
3
(calcd 487.25); elemental analysis calcd (%) for C24
H
N
6
O
3
34
2
The crude product was purified by reversed phase HPLC (H O/
CF COOH (538.85): C 54.83, H 6.79, N 15.60, S 5.95; found: C 54.51,
3
2
MeOH/0.1% CF COOH, gradient 5–95% MeOH, 20 min) followed
H 6.90, N 15.91, S 5.76.
3
by evaporation under reduced pressure to give 3 (331 mg, 66%) as
&
&
Chem. Eur. J. 2015, 21, 1 – 9
www.chemeurj.org
6
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
6
Biotinyl-(N-e-amidocaproic)-(4-methylamido)pyridine (5): The
[Ru(h -p-cymene)(biotin(6-methylamido)-kN2-H1-indazole)Cl ]
2
2+
procedure of Lo et al. was used with additional purification
(10): MS (ESI-IT, pos. mode): m/z: 608.25 [MÀ2ClÀH]
(calcd
[35]
+
steps.
Biotinamidohexanoic acid N-hydroxysuccinimidyl ester
608.16); m/z: 644.31 [MÀCl] (calcd 644.14).
(
(
182 mg, 0.4 mmol) was dissolved in a minimum amount of DMF
ca. 5 mL). Triethylamine (226 mL, 164 mg, 1.6 mmol) and a solution
6
[
Ru(h -p-cymene)(biotinyl(N-e-amidocaproic)-(6-methylamido)-
kN2-H1-indazole)Cl ] (11): MS (ESI-IT, pos. mode): m/z: 360.33
2
of 4-(aminomethyl)pyridine (65 mg, 0.6 mmol) in DMF (5 mL) were
added. The mixture was stirred overnight at RT and evaporated to
dryness under reduced pressure to give a pale yellow oil, which
was suspended in methanol and the insoluble residue was re-
moved by filtration. The filtrate was concentrated under reduced
pressure and dichloromethane (7 mL) and diethyl ether (50 mL)
were added. The precipitate was removed by filtration, washed
with dichloromethane and diethyl ether, and dried in vacuo to
give an off-white solid. The crude product was purified by reversed
phase HPLC (H O/MeOH/0.1% CF COOH, gradient 5–95% MeOH,
2+
2+
[
MÀ2Cl]
(calcd 360.62); m/z: 721.40 [MÀ2ClÀH]
(calcd
+
7
21.25); m/z: 757.35 [MÀCl] (calcd 757.22).
Acknowledgements
This work was supported by the Platform Austria for Chemical
Biology of the Genome Research Program Austria (GEN-AU,
BMWF-70.081/0018-II/1a/2008), the University of Auckland,
Genesis Oncology Trust, the Royal Society of New Zealand and
COST CM1105. We gratefully acknowledge Professor Vladimir
Arion for the refinement of X-ray diffraction data, Dr. Manuel
Brehs for help with reversed phase HPLC, and Mr. Janusz
Mazur for AAS measurements.
2
3
2
0 min) followed by evaporation under reduced pressure to give 5
(
66 mg, 37%) as a white solid. The purity of the product was con-
1
firmed by analytical HPLC (retention time 17.8 min). H NMR
(
1
5
6
6
3
5
500.10 MHz, [D ]DMSO): d=8.59 (dd, J(H,H)=6.0 Hz, J(H,H)=
6
3
3
.4 Hz, 2H; H ), 8.46 (t, J(H,H)=5.8 Hz, 1H; NH), 7.74 (t, J(H,H)=
v,s
3
5
.6 Hz, 1H; NH), 7.39 (dd, J(H,H)=6.1 Hz, J(H,H)=1.4 Hz, 2H; H ),
w,r
3
.42 (s, 1H; NH-biotin), 6.36 (s, 1H; NH-biotin), 4.35 (d, J(H,H)=
.0 Hz, 2H; H ), 4.31 (m, 1H; H ), 4.14 (m, 1H; H ), 3.11 (m, 1H; H ),
p
g
f
e
Keywords: antitumor agents
ruthenium · sandwich complexes
· cancer · drug delivery ·
3
3
3
.02 (dt, J(H,H)=12.8 Hz, J(H,H)=6.9 Hz, 2H; Hj), 2.83 (dd,
J(H,H)=12.5 Hz, J(H,H)=5.1 Hz, 1H; H ), 2.59 (d, J(H,H)=12.5 Hz,
H; H ), 2.19 (t, J(H,H)=7.5 Hz, 2H; H ), 2.05 (t, J(H,H)=7.4 Hz,
h’ a
H; H ), 1.67–1.28 ppm (m, 12H; Hb,c,d,m,l,k); C NMR (125.75 MHz,
2
3
2
h
3
3
1
2
13
n
[1] R. M. Bremnes, K. Andersen, E. A. Wist, Eur. J. Cancer 1995, 31A, 1955–
1959.
[
D ]DMSO): d=173.0 (C ), 172.3 (C), 163.2 (C ),152.3 (C ), 147.9
6 o i x q
(
C ), 123.2 (C ), 61.5 (C ), 59.7 (C ), 55.9 (C ), 41.7 (C ), 40.4 (overlap
[2] M. L. Slevin, L. Stubbs, H. J. Plant, P. Wilson, W. M. Gregory, P. J. Armes,
S. M. Downer, Br. Med. J. 1990, 300, 1458–1460.
s,v
r,w
f
g
e
p
with DMSO) (Ch,h’), 38.8 (C), 35.7 (C ), 35.7 (C ), 29.5 (Cb/c/d/m/l/k), 28.7
(
2
j
a
n
[
3] J. A. Reddy, P. S. Low, Crit. Rev. Ther. Drug Carrier Syst. 1998, 15, 587–
Cb/c/d/m/l/k), 28.5 (C
5.6 ppm (Cb/c/d/m/l/k);
), 26.6 (Cb/c/d/m/l/k), 25.8 (Cb/c/d/m/l/k),
b/c/d/m/l/k
19
627.
F NMR (470.56 MHz, [D ]DMSO): d=
6
+
[4] L. Bildstein, C. Dubernet, P. Couvreur, Adv. Drug Delivery Rev. 2011, 63,
–23.
À75.15 ppm (s, CF COOH); MS (ESI+): m/z: 448.20 [M+H] (calcd
3
3
4
48.24); elemental analysis calcd (%) for C H N O S·0.9CF COOH
22 33 5 3 3
[
5] G. Russell-Jones, K. McTavish, J. McEwan, J. Rice, D. Nowotnik, J. Inorg.
Biochem. 2004, 98, 1625–1633.
(
550.22): C 51.95, H 6.21, N 12.73, S 5.83, O 13.96; found: C 51.86,
H 6.50, N 12.81, S 5.56, O 14.31.
[6] H. M. Said, J. Nutr. 2009, 139, 158–162.
6
[7] H. Wang, W. Huang, Y. J. Fei, H. Xia, T. L. Yang-Feng, F. H. Leibach, L. D.
Devoe, V. Ganapathy, P. D. Prasad, J. Biol. Chem. 1999, 274, 14875–
14883.
[
Ru(h -p-cymene)Cl (kN2-H1-indazole)] (9): The synthetic method
2
[30]
was adapted from a reported procedure.
.3 mmol, 2 equiv) was added to a suspension of [{Ru(h -p-cy-
mene)Cl } ] (92 mg, 0.15 mmol) in toluene (15 mL) at RT (an orange
Indazole (35 mg,
6
0
[
8] T. Minko, P. V. Paranjpe, B. Qiu, A. Lalloo, R. Won, S. Stein, P. J. Sinko,
Cancer Chemother. Pharmacol. 2002, 50, 143–150.
2
2
precipitate formed immediately). The resulting mixture was heated
to reflux for 3 h. After the mixture cooled, the precipitate was fil-
tered, washed with petroleum ether (3ꢃ10 mL) and diethyl ether,
[
9] K. Na, T. B. Lee, K.-H. Park, E.-K. Shin, Y.-B. Lee, H.-K. Choi, Eur. J. Pharm.
Sci. 2003, 18, 165–173.
[10] S. Chen, X. Zhao, J. Chen, J. Chen, L. Kuznetsova, S. S. Wong, I. Ojima,
Bioconjugate Chem. 2010, 21, 979–987.
[11] I. Ojima, Acc. Chem. Res. 2008, 41, 108–119.
[12] S. Lin, K. Fang, M. Hashimoto, K. Nakanishi, I. Ojima, Tetrahedron Lett.
2000, 41, 4287–4290.
[13] S. Bhuniya, M. H. Lee, H. M. Jeon, J. H. Han, J. H. Lee, N. Park, S. Maiti, C.
Kang, J. S. Kim, Chem. Commun. 2013, 49, 7141–7143.
and dried in vacuo, affording 9 (93 mg, 73%) as an orange micro-
1
crystalline solid. H NMR (500.10 MHz, CDCl ): d=11.78 (s, 1H; NH-
3
3
indazole), 8.41 (s, 1H; H ), 7.71 (d, J(H,H)=8.4 Hz, 1H; H ), 7.41 (m,
H; H ), 7.19 (m, 1H; H ), 5.59 (d, J(H,H)=6.0 Hz, 2H; CHcym),
r,w q
.41 (d, J(H,H)=6.0 Hz, 2H; CH ), 3.00 (sept, J(H,H)=7.0 Hz, 1H;
CHMe ), 2.28 (s, 3H; C H (CH )), 1.29 ppm (d, J(H,H)=7.0 Hz, 6H;
u
s
3
2
5
3
3
cym
3
2
6
4
3
[
14] S. Maiti, N. Park, J. H. Han, H. M. Jeon, J. H. Lee, S. Bhuniya, C. Kang, J. S.
Kim, J. Am. Chem. Soc. 2013, 135, 4567–4572.
CH(CH ) ); elemental analysis calcd (%) for C H Cl N Ru (424.33): C
3
2
17 20
2
2
4
8.12, H 4.75, N 6.60; found: C 48.19, H 4.69, N 6.54.
[
15] V. K. Yellepeddi, A. Kumar, D. M. Maher, S. C. Chauhan, K. K. Vangara, S.
Palakurthi, Anticancer Res. 2011, 31, 897–906.
Preparation of complexes 7, 8, 10, and 11; general procedure:
6
[16] V. K. Yellepeddi, A. Kumar, S. Palakurthi, Anticancer Res. 2009, 29, 2933–
943.
17] W. Yang, Y. Cheng, T. Xu, X. Wang, L.-p. Wen, Eur. J. Med. Chem. 2009,
4, 862–868.
[
{Ru(h -p-cymene)Cl } ] (4.72 mg, 0.01 mmol) and 2–5 (2 equiv,
2 2
2
0
.02 mmol) were dissolved in anhydrous DMF (2 mL). The reaction
[
[
[
[
mixture was stirred for 3 h, freeze-dried, and incubated in situ with
the cancer cells.
4
18] S. Ramanathan, S. Pooyan, S. Stein, P. D. Prasad, J. Wang, M. J. Leibowitz,
V. Ganapathy, P. J. Sinko, Pharm. Res. 2001, 18, 950–956.
19] S. Ramanathan, B. Qiu, S. Pooyan, G. Zhang, S. Stein, M. J. Leibowitz,
P. J. Sinko, J. Controlled Release 2001, 77, 199–212.
6
[Ru(h -p-cymene)(biotin(4-methylamido)-kN1-pyridine)Cl ]
(7):
2
2
+
MS (ESI-IT, pos. mode): m/z: 285.06 [MÀ2Cl] (calcd 285.08); m/z:
+
6
05.12 [MÀCl] (calcd 605.13).
20] K. K.-W. Lo, M.-W. Louie, K.-S. Sze, J. S.-Y. Lau, Inorg. Chem. 2008, 47,
6
[Ru(h -p-cymene)(biotinyl(N-e-amidocaproic)-(4-methylamido)-
602–611.
kN1-pyridine)Cl ] (8): MS (ESI-IT, pos. mode): m/z: 341.58
2
[21] M.-W. Louie, M. H.-C. Lam, K. K.-W. Lo, Eur. J. Inorg. Chem. 2009, 4265–
4273.
[22] K. Y. Zhang, K. K.-W. Lo, Inorg. Chem. 2009, 48, 6011–6025.
2
+
2+
[
6
MÀ2Cl]
(calcd 341.62); m/z: 682.20 [MÀ2ClÀH]
(calcd
+
82.24); m/z: 718.14 [MÀCl] (calcd 718.21).
Chem. Eur. J. 2015, 21, 1 – 9
www.chemeurj.org
7
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[
[
23] S.-K. Leung, K. Y. Kwok, K. Y. Zhang, K. K.-W. Lo, Inorg. Chem. 2010, 49,
984–4995.
[30] C. A. Vock, C. Scolaro, A. D. Phillips, R. Scopelliti, G. Sava, P. J. Dyson, J.
Med. Chem. 2006, 49, 5552–5561.
[31] W. Kandioller, C. G. Hartinger, A. A. Nazarov, C. Bartel, M. Skocic, M. A. Ja-
kupec, V. B. Arion, B. K. Keppler, Chem. Eur. J. 2009, 15, 12283–12291.
[32] M. V. Babak, S. M. Meier, A. A. Legin, M. S. A. Razavi, A. Roller, M. A. Jaku-
pec, B. K. Keppler, C. G. Hartinger, Chem. Eur. J. 2013, 19, 4308–4318.
4
24] M. V. Babak, S. M. Meier, K. Huber, J. Reynisson, A. A. Legin, M. A. Jaku-
pec, A. Roller, A. Stukalov, M. Gridling, K. L. Bennett, J. Colinge, W.
Berger, P. J. Dyson, G. Superti-Furga, B. K. Keppler, C. G. Hartinger, 2014,
unpublished data.
[
[
[
33] S. Betanzos-Lara, A. Habtemariam, G. J. Clarkson, P. J. Sadler, Eur. J.
Inorg. Chem. 2011, 3257–3264.
34] G. Jones, P. Willet, R. C. Glen, A. R. Leach, R. Taylor, J. Mol. Biol. 1997,
[
[
[
25] D. Plazuk, J. Zakrzewski, M. Salmain, A. Blauz, B. Rychlik, P. Strzelczyk, A.
Bujacz, G. Bujacz, Organometallics 2013, 32, 5774–5783.
26] A. D. Vadlapudi, R. K. Vadlapatla, A. K. Mitra, Curr. Drug Targets 2012, 13,
2
67, 727–748.
9
94–1003.
27] A. D. Vadlapudi, R. K. Vadlapatla, D. Pal, A. K. Mitra, Int. J. Pharm. 2013,
41, 535–543.
35] K. K.-W. Lo, W.-K. Hui, Inorg. Chem. 2005, 44, 1992–2002.
4
[
[
28] K. K.-W. Lo, T. K.-M. Lee, Inorg. Chem. 2004, 43, 5275–5282.
29] M. Slim, N. Durisic, P. Grutter, H. F. Sleiman, ChemBioChem 2007, 8, 804–
Received: June 15, 2014
Revised: January 12, 2015
Published online on && &&, 0000
812.
&
&
Chem. Eur. J. 2015, 21, 1 – 9
www.chemeurj.org
8
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
On target: Ruthenium(II) arene com-
plexes with biotin-containing ligands
were prepared so that drug delivery
through tumor-specific vitamin-receptor
mediated endocytosis could be exploit-
ed (see figure). Complex formation with
&
Drug Discovery
M. V. Babak, D. Pla z˙ uk, S. M. Meier,
H. J. Arabshahi, J. Reynisson, B. Rychlik,
A. Błau z˙ , K. Szulc, M. Hanif, S. Strobl,
A. Roller, B. K. Keppler, C. G. Hartinger*
6
a {Ru(h -p-cymene)} fragment resulted
&
& – &&
in significantly enhanced cytotoxicity in
comparison with free ligands. These
novel half-sandwich ruthenium(II) biotin
conjugates may act as biological vectors
to cancer cells.
Half-Sandwich Ruthenium(II) Biotin
Conjugates as Biological Vectors to
Cancer Cells
Chem. Eur. J. 2015, 21, 1 – 9
www.chemeurj.org
9
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ