Organoruthenium and osmium anticancer complexes bearing a maleimide functional group: Reactivity to cysteine, stability, and cytotoxicity

Article abstract of DOI:10.1002/cplu.201402390

Targeted delivery provides a means to overcome the systemic toxicity of cancer chemotherapeutics. Maleimide functionalization of anticancer-active metal complexes allows their site-selective covalent conjugation to human serum albumin (HSA) to exploit passive targeting to the tumor site through the enhanced permeability and retention (EPR) effect. By incorporating the maleimide moiety in the monodentate N-donor coligands, a series of [M^II(cym)X₂] (M = Ru, Os; cym = η⁶-pcymene; X = Cl₂, Br₂, I₂, oxalate) complexes was synthesized and characterized. The influence of the metal center, η⁶-arene, and leaving group X on the chemical and biological properties was studied. The compounds exhibited low stability in dimethyl sulfoxide (DMSO) but were more stable in the 5% aqueous DMSO solution used in biological assays. Incubation with cysteine as a model for thiol-containing biomolecules demonstrated high reactivity of the maleimide with the thiol group. In in vitro anticancer assays in human colorectal, non-small-cell lung, and cervical carcinoma cells the complexes were good to moderately active with the lowest IC₅₀ value at 8 μm for an osmium-iodido complex.

Full text of DOI:10.1002/cplu.201402390

DOI: 10.1002/cplu.201402390
Full Papers
Organoruthenium and Osmium Anticancer Complexes
Bearing a Maleimide Functional Group: Reactivity to
Cysteine, Stability, and Cytotoxicity
Sally Moon,[a] Muhammad Hanif,*[a, b] Mario Kubanik,[a] Hannah Holtkamp,[a] Tilo Sçhnel,[a]
Stephen M. F. Jamieson,[c] and Christian G. Hartinger*[a]
Dedicated to Professor L. James Wright on the occasion of his 60th birthday
Targeted delivery provides a means to overcome the systemic
toxicity of cancer chemotherapeutics. Maleimide functionaliza-
tion of anticancer-active metal complexes allows their site-se-
lective covalent conjugation to human serum albumin (HSA) to
exploit passive targeting to the tumor site through the en-
hanced permeability and retention (EPR) effect. By incorporat-
ing the maleimide moiety in the monodentate N-donor co-
ligands, a series of [MII(cym)X2] (M=Ru, Os; cym=h6-p-
cymene; X=Cl2, Br2, I2, oxalate) complexes was synthesized
and characterized. The influence of the metal center, h6-arene,
and leaving group X on the chemical and biological properties
was studied. The compounds exhibited low stability in dimeth-
yl sulfoxide (DMSO) but were more stable in the 5% aqueous
DMSO solution used in biological assays. Incubation with cys-
teine as a model for thiol-containing biomolecules demonstrat-
ed high reactivity of the maleimide with the thiol group. In in
vitro anticancer assays in human colorectal, non-small-cell
lung, and cervical carcinoma cells the complexes were good to
moderately active with the lowest IC50 value at 8 mm for an
osmium–iodido complex.
Introduction
Among non-platinum anticancer metallodrugs, complexes
based on group 8 metal ions such as iron (Fe), ruthenium (Ru),
and osmium (Os) have emerged as the most promising candi-
dates for drug development. Two ruthenium complexes, NAMI-
A and KP1019/KP1339, are already in clinical trials, and several
organometallic compounds of these metals are at an advanced
stage of preclinical development.[1] In particular, half-sandwich
{MII(h6-arene)} complexes of Ru and Os have been widely rec-
ognized as potential anticancer chemotherapeutics.[2] However,
like many other small molecules, they lack a mechanism to
induce selectivity for tumor tissue, which often results in collat-
eral damage to healthy tissue.[3] Several strategies have been
suggested for targeted drug accumulation in tumor tissue,
among which systemic delivery using macromolecular carrier
systems (e.g., nanoparticles, micelles, liposomes, and protein
biomolecules) has been demonstrated to be efficacious. This
approach exploits the passive targeted delivery of cytotoxic
drugs based on the enhanced permeability and retention (EPR)
effect, which has been observed for macromolecular carrier
systems.[4] One of the carriers suggested to promote selective
accumulation at tumor sites owing to the EPR effect is human
serum albumin (HSA). HSA may be selectively functionalized
on cysteine-34 with a maleimide-carrying prodrug moiety. This
approach was demonstrated by Kratz and co-workers for
INNO-206 (DOXO-EMCH), the (6-maleimidocaproyl)hydrazone
of doxorubicin (Scheme 1), which showed site-selective and co-
[a] S. Moon, Dr. M. Hanif, M. Kubanik, H. Holtkamp, Dr. T. Sçhnel,
Prof. C. G. Hartinger
University of Auckland
Private Bag 92019, Auckland 1142 (New Zealand)
E-mail: m.hanif@auckland.ac.nz
c.hartinger@auckland.ac.nz
[b] Dr. M. Hanif
Department of Chemistry
COMSATS Institute of Information Technology
Abbottabad 22060 (Pakistan)
[c] Dr. S. M. F. Jamieson
Auckland Cancer Society Research Centre
University of Auckland, Private Bag 92019
Auckland 1142 (New Zealand)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201402390.
This article is part of the “Early Career Series”. To view the complete
series, visit: http://chempluschem.org/earlycareer.
Scheme 1. Examples of maleimide-functionalized bioactive molecules.
ChemPlusChem 0000, 00, 0 – 0
1
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

Full Papers
valent introduction of the maleimide-bearing moiety to cys-
teine-34 of circulating albumin after intravenous administra-
tion. DOXO-EMCH has entered phase II clinical trials since the
in vivo results demonstrated enhanced antitumor activity in
cancer models relative to doxorubicin.[5] Kratz also developed
analogues of maleimide-bearing platinum complexes with the
aim to enhance the antitumor efficacy, reduce side effects as
well as to overcome resistance in cisplatin-resistant cancer cells
by increased accumulation.[6]
We extended the approach of conjugating a maleimide
functional group to organometallic anticancer compounds,
specifically RAPTA complexes with the general formula [Ru(h6-
arene)(PTA)Cl2]
(PTA=1,3,5-triaza-7-phosphatricyclodecane).
Scheme 2. Synthesis of maleimide-functionalized indazole and pyridine
derivatives using two pathways to yield I–III using (i) sodium acetate and
acetic anhydride; or (ii) triethylamine and trimethylsilyl chloride in aceto-
nitrile.
The RAPTA derivatives are among the most widely studied or-
ganometallic anticancer agents owing to their pronounced an-
timetastatic activity in in vivo tumor models.[7] The RAPTA de-
rivatives were functionalized with a maleimide on the arene
moiety (Mal-RAPTA; Scheme 1), as the arene was found to
have less impact on the overall biological activity than the PTA
co-ligand.[7b,8] The reactivity of Mal-RAPTA towards thiol-con-
taining biomolecules, such as l-cysteine, glutathione, and
human serum albumin, was studied by using NMR spectrosco-
py, mass spectrometry, and size-exclusion chromatography
(SEC) coupled with inductively coupled plasma mass spectrom-
etry (ICP-MS). These experiments demonstrated chemoselec-
tive covalent anchoring of Mal-RAPTA to cysteine-34 on HSA
without altering significantly the in vitro anticancer effect of
the RAPTA pharmacophore. In aqueous solution, upon extend-
ed exposure to thiol-containing biomolecules, the arene ligand
releases the metal center thereby allowing delivery to tumor
cells.
maleic acid intermediate featured, in addition to the four
proton signals from the indazole rings, the two maleic acid CH
proton signals in the form of doublets at d=6.52 and
6.32 ppm. The ring closing to yield I and II was confirmed by
the appearance of a sharp singlet at d=6.91 ppm for H12 and
H13 (Scheme 2). The methyl group of the acetyl substituent of
I was observed at d=2.79 ppm. In the ESI-mass spectra of
both I and II, [M+Na]+ peaks were observed at the respective
accurate m/z values of 278.0536 (m/z calcd: 278.0529) and
236.0430 (m/z calcd: 236.0424) in positive-ion mode, respec-
tively.
Because of this side reaction occurring on the pyrazole-NH
group, we used 4-aminomethylpyridine for comparison pur-
poses. The conversion of 4-aminomethylpyridine into the re-
spective maleimide resulted in the ring-closed product III,
which was soluble in the reaction mixture of water and acetic
anhydride. All attempts to isolate the product failed, and sig-
nificant ring opening occurred during purification by column
chromatography.
Herein, we report the functionalization of N-donor indazole
and pyridine ligands with maleimides and the preparation of
their ruthenium and osmium(arene) organometallics with dif-
ferent mono- and bidentate leaving groups to study the effect
of such modification on both the reactivity to thiols and bio-
logical activity.
Since the standard ring-closing procedure with acetic anhy-
dride and sodium acetate[10] led to the acetylation of the endo-
cyclic indazole-NH group and purification failed in the case of
the pyridine derivative, we looked for an alternative method
applicable to both indazole and pyridine derivatives. Diederich
et al. reported a procedure to prepare N-(pyridin-4-ylmethyl)-
maleimide (III),[11] by employing trimethylsilyl chloride to
induce the ring-closing process. The same approach was ap-
plied to prepare II (and also III) starting from the ring-opened
intermediate. This was suspended in acetonitrile and triethyla-
mine was added (2 equiv), which resulted in immediate disso-
lution. Trimethylsilylchloride (1 equiv) was added dropwise to
the yellow solution, which turned orange over the course of
one hour at 808C. Filtration through a plug of silica gel and
evaporation of the solvent gave II as a yellow solid in 32%
yield.
Results and Discussion
Nitrogen-donor co-ligands have been widely used in the devel-
opment of metal-based anticancer agents. They have been
found to yield potent platinum and ruthenium complexes and
also organometallic compounds.[1,2b,c,9] Clinically evaluated
KP1019/KP1339 feature, for example, indazole (Hind) ligands,
and 6-aminoindazole is a commercially available precursor
used to prepare maleimide derivatives.
The introduction of a maleimide group in 6-aminoindazole
(Scheme 2) was achieved by reaction with maleic anhydride in
dry dichloromethane to afford the ring-opened intermediate
as a solid. This intermediate was employed in the next step
without further purification. The standard ring-closing proce-
dure[10] using acetic anhydride in conjunction with sodium ace-
tate yields in the case of 6-aminoindazole a mixture of the pyr-
azole-N-acetylated product I in addition to the desired malei-
mide II. Such acylation has been reported for similar com-
In contrast to N-(1-acetylindazol-6-yl)maleimide (I), the
1H NMR spectrum of II did not feature methyl proton signals of
the acetyl group, but the pyrazole amine proton was observed
pound types such as indoles.[10] The H NMR spectrum of the
1
&
ChemPlusChem 0000, 00, 0 – 0
www.chempluschem.org
2
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
at d=10.25 ppm as a broad singlet and the maleimide protons
were present as a singlet at d=6.90 ppm.
Maleimides I–III were used to prepare a series of {RuII(cym)}
and {OsII(cym)} complexes by adding I–III (2 equiv) to a solution
of the respective [{M(cym)X2}2] dimer (M=Ru or Os; X=Cl,
Br, I) in absolute dichloromethane (Scheme 3). To synthesize
tives the signal was shifted by 5–7 ppm from d=134 ppm in II
to d=139–141 ppm in 2a–2c, 2e, and 2 f. For the [M(cym)(ox-
alato)] complexes 2d and 2g in addition to a signal at about
d=169 ppm for C11 and C14, another peak at d=167.2 ppm
was observed, which was assigned to the oxalato carbon
atoms. In case of 3a–3d, the carbon atoms C2 and C6, located
next to the coordinated nitrogen atom, were deshielded after
complexation and appeared 3–6 ppm downfield in the 13C{1H}
NMR spectrum. The H2 and H6 proton signals of the pyridine
1
ligand III in the H NMR spectra are impacted by coordination
and the signals shifted slightly downfield for 3a–3c (d=0.33–
0.84 ppm) and upfield for 3d (d=0.10 ppm). This is compara-
ble to C3ÀH in II and the respective complexes.
In positive-ion mode ESI-MS, all halido complexes gave m/z
values assignable to [MÀX]+ or [M+MeCNÀX]+ ions, whereas
for the oxalato derivatives protonated or sodiated pseudomo-
lecular ions were detected.
The molecular structures of complexes 2c and 3c were de-
termined by single-crystal X-ray diffraction analysis.[13a] The
crystals were obtained by slow diffusion of diethyl ether into
solutions of 2c or 3c in dichloromethane. Both 2c and 3c crys-
tallized in the monoclinic space groups P21/c and P21/n, re-
spectively, and displayed a piano-stool geometry at the metal
center with a p-cymene ring coordinated in h6 fashion to the
metal center (Figure 1; selected bond lengths and angles are
Scheme 3. Synthetic route to [M(h6-p-cymene)] complexes of N-(indazol-6-
yl)maleimide (II) and N-(pyridin-4-ylmethyl)maleimide (III) and the NMR spec-
troscopy numbering scheme.
the oxalato complexes, the [{M(cym)Cl2}2] dimer (M=Ru or Os)
was stirred with silver oxalate in water for 12 hours, which ab-
stracted the chlorido ligands from the metal centers and yield-
ed [M(cym)(oxalate-k2O,O)(H2O)] (d (M=Ru) and g (M=Os)),
which were further treated with II or III (1 equiv). Both of these
procedures gave, after stirring for four hours at room tempera-
ture and workup, the respective mononuclear complexes 2a–
2g and 3a–3d in 57–90% yields. The complexes were charac-
1
terized by melting-point analysis, H and 13C{1H} NMR spectros-
copy, mass spectrometry, and elemental analysis.
While the conversion of II and III with the metal precursors
gave the expected complexes, the reaction of
I with
[{Ru(cym)Cl2}2] (a) interestingly resulted in the loss of the acetyl
1
group. This was clearly visible in the H NMR spectrum, which
was identical to that of 2a. Acetyl cleavage from indazole was
reported before, for example, by Shuttleworth et al. who dem-
onstrated such cleavage by the reaction of indazole in metha-
nol in the presence of HCl.[12] Upon coordination of the inda-
zole-N2 to the metal center, the N1ÀH proton signals of 2a–
2c, 2e, and 2 f were shifted significantly by 1.2–1.9 ppm to
Figure 1. Molecular structures of 2c (top) and 3c (bottom) drawn at the
50% probability level.
1
lower field in the H NMR spectrum relative to II. However, this
given in Table 1). The coordination sphere around the metal
center was completed by the monodentate N-(indazol-6-yl)ma-
leimide (II) or N-(pyridin-4-ylmethyl)maleimide (III) ligands, and
two iodido leaving groups. The distance between the metal
center and the cym centroid was 1.675 and 1.680 ꢁ for 2c and
3c, respectively. The RuÀI bond lengths in 2c were 2.7153(4)
and 2.7409(4) ꢁ, and in the same range as the OsÀI bond
lengths in 3c with 2.7368(4) and 2.7223(4) ꢁ, respectively. The
RuÀN and OsÀN distances were 2.125(3) and 2.150(4) ꢁ, re-
signal was absent in the spectra of the oxalato derivatives 2d
and 2g. The maleimide protons of both the ligands and com-
plexes appeared as a sharp singlet at about d=6.9 ppm. In the
13C{1H} NMR spectra, the most pronounced downfield shifts of
signals were observed for the indazole C3 atoms due to elec-
tron density transferred from the ligands to the metal centers.
For oxalato complexes 2d and 2g, this peak was observed
only about 1 ppm downfield, whereas for the halido deriva-
ChemPlusChem 0000, 00, 0 – 0
www.chempluschem.org
3
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
To investigate the effect of the leaving group on the stability
of the complexes, the Ru–oxalato complexes 2d and 3d were
studied under the same conditions as the analogous chlorido
derivatives. The substitution of the labile chlorido ligands with
a chelating oxalate group in 2d and 3d provided significantly
enhanced stability (Figure 3 for 3d). A similar observation was
reported for P-derived ligand systems that undergo side reac-
tions on the ligand group.[16] However, in the case of these N-
donor ligands, the substitution resulted in increased stability of
Table 1. Bond lengths [ꢁ] and angles [8] of 2c (M=Ru) and 3c (M=Os).
Compound
2c
1.675
2.7153(4)
2.7409(4)
2.125(3)
3c
1.680
2.7223(4)
2.7368(4)
2.150(4)
MÀcymcentroid
MÀI1
MÀI2
MÀN
I1-M-I2
N-M-I1
N-M-I2
90.20(1)
88.12(8)
87.21(7)
87.151(14)
86.60(11)
86.79(11)
1
the RuÀN bond to the coligand. In the case of 2d, the H NMR
spectively, which are comparable to those reported
for similar complexes.[13b,c] The NÀNH bond length in
2c is slightly longer after coordination to the Ru
center than in [H2ind]+ (Hind=indazole) but similar
to Hind in both cis- and trans-configured [Ru(Hind)-
(NO)Cl4](H2ind) complexes.[14]
The maleimide is twisted out of plane at 51.888
from the indazole ring in 2c, and at 94.588 nearly
perpendicular to the pyridine ring in 3c, due to the
methylene spacer separating the maleimide from the
aromatic ring systems. In 2c a hydrogen bond be-
tween a maleimide oxygen and the indazole NH
group of a second molecule is formed (NÀH···O
2.063(3) ꢁ), whereas no hydrogen-bonding interac-
tions were observed in the structure of 3c. This re-
sults in a zigzag pattern-type packing of 2c along
a (Figure S1 in the Supporting Information), whereas
the packing of 3c appears dominated by a coplanar
Figure 2. 1H NMR spectra of [{Ru(cym)Cl2}2] (a), uncoordinated N-(indazol-6-yl)maleimide
II and complex 2a recorded in [D6]DMSO.
arrangement of the pyridine and maleimide ring systems re-
sulting in twofold bands in the (ac) plane (Figure S2 in the
Supporting Information).
Monodentate versus bidentate anionic ligands: Complex
stability in DMSO
Often metal(arene) complexes have limited aqueous solubility,
and thus stock solutions in DMSO are prepared for physico-
chemical and in vitro biological experiments. In some cases,
monodentate nitrogen heterocyclic or amine-based ligand sys-
tems dissociate immediately from the {Ru(h6-arene)} arene frag-
ment once dissolved in DMSO, which results in formation of
[Ru(h6-arene)(DMSO)X2].[15] Therefore, stability tests of the mal-
eimide-bearing complexes 2a and 3a in DMSO were conduct-
Figure 3. 1H NMR spectra of [Ru(cym)(oxalate)(H2O)] (d), uncoordinated N-
(pyridyl-4-ylmethyl)maleimide (III), and complex 3d recorded in [D6]DMSO.
1
1
ed by H NMR spectroscopy (Figure 2). The H NMR spectra of
the complexes were recorded immediately after dissolution in
1
[D6]DMSO. The spectra were then compared with the H NMR
spectrum revealed that approximately 50% of 2d was immedi-
ately converted into the DMSO complex, that is, [Ru(h6-p-cy-
mene)(DMSO)(oxalato)]. In contrast, no immediate dissociation
of III from the Ru center was observed for 3d, but after
1.5 hours minor signals indicative of a reaction with DMSO
spectra of uncoordinated ligands and of [{RuCl2(cym)}2] dis-
solved in [D6]DMSO. It is clear that the dissociation of ligands II
and III from the metal center proceeded immediately. The
proton signals of the N-(indazol-6-yl)maleimide fragment in the
spectrum recorded for 2a were identical to those of II. In addi-
tion, the protons assigned to the {Ru(cym)} fragment displayed
a virtually identical set of signals to that found for the
[{Ru(cym)Cl2}2] dimer a, present then as the DMSO complex
[Ru(cym)(DMSO)Cl2].
1
were detected in the H NMR spectrum (Figure 3). This differ-
ence for the pyridine- and indazole-derived ligands may be ex-
plained by the p-acceptor properties of the pyridine ligand,
which decreases the leaving-group lability relative to the inda-
zole derivative.
&
ChemPlusChem 0000, 00, 0 – 0
www.chempluschem.org
4
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Stability in aqueous solution
Both complexes 2a and 3a immediately dissociated in DMSO
to release II and III, which is, however, less relevant for biologi-
cal studies as diluted aqueous solutions with low DMSO con-
tent are required. Therefore, the stability of the complexes in
1
5% [D6]DMSO/D2O was investigated by H NMR spectroscopy.
In contrast to studies in DMSO, 2a underwent chlorido/solvent
ligand-exchange reactions. In addition, the leaving group was
found to impact the stability as 2a (with ClÀ leaving groups)
hydrolyzed faster than 2c bearing iodido ligands. Notably,
about 50% of 2a hydrolyzed virtually immediately upon disso-
lution and after 24 hours an equilibrium established and a simi-
lar picture was observed for the bromido complex 2b. In con-
trast, the iodido complex 2c demonstrated comparably slower
ligand-exchange reactions than those seen for 2a and 2b.
Compound 2c degraded slowly and after seven days about
90% of the original compound was still detectable. Similar ob-
servations were reported by Sadler et al. for a series of [Ru(h6-
arene)(ethylenediamine)X]n+ complexes (X=Cl, Br, I). When the
h6-arene system was identical in the complexes, the hydrolysis
rate decreased in the order ClꢀBr>I.[17] In the case of the oxa-
lato complex 2d with its bidentate ligand, an intermediate
degree of stability was observed and after about two days still
75% of the initial complex peak was detectable, while an addi-
tional set of signals appeared, possibly due to ligand-exchange
reactions occurring in the solution containing water and 5%
DMSO. The same experiments with the pyridyl compound 3a
showed a higher degree of stability than the indazole ana-
logues. This may be rationalized by the fact that pyridine is
a better p-electron acceptor than the indazole group. Indazole
forms a bond to Ru that has therefore higher s character,
which results in decreased RuÀCl bond stability.[18]
Figure 4. (A) 1H NMR spectra recorded after different time intervals for the
reaction of 2a with l-cysteine in 5% [D6]DMSO/D2O. (B) Decrease of the mal-
eimide-CH proton signals during the reaction of 2a with l-cysteine in 5%
[D6]DMSO/D2O within 25 min.
Binding to thiol-containing biomolecules
Table 2. In vitro anticancer activity (mean IC50 valuesÆstandard devia-
tions) of 2a–2g and 3a–3c in human colorectal (HCT116), non-small-cell
lung (NCI-H460) and cervical (SiHa) carcinoma cell lines (exposure time
72 h).
The thiol-binding ability of 2a and 3a was investigated by
using l-cysteine (Cys) as a model system for larger biomole-
cules. Complex 2a reacted very quickly with Cys, as indicated
by the disappearance of the peak at about d=7.0 ppm in the
1H NMR spectrum (Figure 4A). In fact, the spectrum recorded
immediately after addition of Cys to 2a showed only 18% un-
reacted maleimide (Figure 4B), whereas after 25 minutes of re-
action complete conversion was observed. The pyridyl com-
plex 3a showed a similar reactivity, which indicates that the re-
action is very specific for maleimide functional groups, as was
also observed for {Ru(arene)} complexes functionalized with
a maleimide functional group.[8a]
Ligand
Compound
IC50 values [mm]
HCT116
NCI-H460
SiHa
indazole
2a
2b
2c
2d
2e
2 f
82Æ29
29Æ4
37Æ7
17Æ1
17Æ0.3
14Æ1
25Æ4
105Æ1
47Æ11
32Æ3
164Æ13
66 Æ2
38Æ1
20Æ6
54Æ11
38Æ3
57Æ7
11Æ0.3
25Æ0.6
46Æ14
32Æ6
2g
pyridine
3a
3b
3c
11 Æ1
9Æ0.8
8Æ1
15Æ3
13Æ3
14Æ5
35Æ3
28Æ1
29Æ5
Evaluation for in vitro anticancer activity
The tumor-inhibiting potential of 2a–2g and 3a–3c was de-
termined in human colorectal carcinoma HCT116, non-small-
cell lung carcinoma NCI-H460, and cervical carcinoma SiHa
cells. All the complexes exhibited moderate to low micromolar
activity in a similar concentration range against all tested cell
lines (Table 2). This is not surprising as the complexes are
known to undergo ligand-exchange reactions in even low con-
centrations of DMSO (Figure 4) which, in the presence of multi-
dentate ligand systems, such as amino acids in the cell culture
medium, may eventually lead to cleavage of the N-donor
ligand. However, the biological activity seems to correlate with
ChemPlusChem 0000, 00, 0 – 0
www.chempluschem.org
5
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[1] G. Gasser, I. Ott, N. Metzler-Nolte, J. Med. Chem. 2011, 54, 3–25.
[2] a) M. Hanif, M. V. Babak, C. G. Hartinger, Drug Discovery Today 2014, 19,
1640–1648; b) C. G. Hartinger, N. Metzler-Nolte, P. J. Dyson, Organome-
tallics 2012, 31, 5677–5685; c) C. G. Hartinger, P. J. Dyson, Chem. Soc.
Rev. 2009, 38, 391–401.
[3] F. Marcucci, F. Lefoulon, Drug Discovery Today 2004, 9, 219–228.
[4] a) D. Peer, J. M. Karp, S. Hong, O. C. Farokhzad, R. Margalit, R. Langer,
Nat. Nanotechnol. 2007, 2, 751–760; b) T. M. Allen, Nat. Rev. Cancer
2002, 2, 750–763; c) F. Kratz, J. Controlled Release 2008, 132, 171–183.
[5] a) R. Graeser, N. Esser, H. Unger, I. Fichtner, A. Zhu, C. Unger, F. Kratz,
Invest. New Drugs 2010, 28, 14–19; b) A. Zhu, I. Fichtner, F. Kratz, A.
Craig, Mol. Cancer Ther. 2007, 6, 3421s–3421s.
the stability of the complexes observed in 5% DMSO/water
mixtures (Table 2). The highly stable oxalato derivative 2d was
the most active Ru complex. Os–I complexes such as 2 f are
known to undergo slower ligand-exchange reactions,[16b] and
they were in general more active than their Ru counterparts. In
the case of the oxalato complexes, the Os derivative 2g was
slightly less active than the Ru compound 2d. The nature of
the metal center and the leaving group had hardly any influ-
ence on the cytotoxic activity of the pyridine complexes 3a–
3c.
[6] D. Garmann, A. Warnecke, G. V. Kalayda, F. Kratz, U. Jaehde, J. Controlled
Release 2008, 131, 100–106.
[7] a) C. S. Allardyce, P. J. Dyson, D. J. Ellis, S. L. Heath, Chem. Commun.
2001, 1396–1397; b) C. Scolaro, A. Bergamo, L. Brescacin, R. Delfino, M.
Cocchietto, G. Laurenczy, T. J. Geldbach, G. Sava, P. J. Dyson, J. Med.
Chem. 2005, 48, 4161–4171; c) P. J. Dyson, Chimia 2007, 61, 698–703.
[8] a) M. Hanif, A. A. Nazarov, A. Legin, M. Groessl, V. B. Arion, M. A. Jakupec,
Y. O. Tsybin, P. J. Dyson, B. K. Keppler, C. G. Hartinger, Chem. Commun.
2012, 48, 1475–1477; b) C. Scolaro, T. J. Geldbach, S. Rochat, A. Dorcier,
C. Gossens, A. Bergamo, M. Cocchietto, I. Tavernelli, G. Sava, U. Rothlis-
berger, P. J. Dyson, Organometallics 2006, 25, 756–765.
[9] a) M. Galanski, M. A. Jakupec, B. K. Keppler, Curr. Med. Chem. 2005, 12,
2075–2094; b) C. G. Hartinger, S. Zorbas-Seifried, M. A. Jakupec, B.
Kynast, H. Zorbas, B. K. Keppler, J. Inorg. Biochem. 2006, 100, 891–904;
c) J. M. Rademaker-Lakhai, D. van den Bongard, D. Pluim, J. H. Beijnen,
J. H. Schellens, Clin. Cancer Res. 2004, 10, 3717–3727.
Conclusion
MetalII(h6-arene) complexes are extensively investigated as or-
ganometallic cancer therapeutics. One of the promising ap-
proaches to the design of the next-generation chemothera-
peutics involves the incorporation of functional groups to
obtain compounds that target tumors more selectively than
established chemotherapeutics. By introducing a maleimide
functional group through the co-ligands coordinated to {Ru(h6-
p-cymene)} or {Os(h6-p-cymene)} fragments with chlorido, bro-
mido, iodido, or oxalato leaving groups, we aimed for com-
pounds that would selectively react with thiols to utilize pro-
teins such as HSA with its free Cys34 for selective delivery. The
complexes reacted rapidly and selectively with Cys, which was
here employed as a model compound for HSA. Stability studies
revealed that, like other complexes with monodentate nitro-
gen-donor ligands, they underwent immediate cleavage of
their indazole or pyridyl ligands in pure DMSO. In contrast, in
aqueous solutions the ligand-exchange reactions at the metal
centers were significantly decelerated and allowed the com-
plexes to retain their main pharmacophore intact. The decom-
position can be slowed down by replacing chloride ligands
with those such as iodide or oxalate. The role of the coligand
also became apparent with the pyridine complexes being
more stable than the indazole analogues. This was also shown
to have an impact in the biological studies where especially
the more kinetically inert osmium complexes exhibited IC50
values as low as 8 mm (3c) compared with 11 mm for the analo-
gous ruthenium derivative 3a in the human colorectal carcino-
ma cell line HCT116.
[10] P. Haquette, B. Talbi, S. Canaguier, S. Dagorne, C. Fosse, A. Martel, G.
Jaouen, M. Salmain, Tetrahedron Lett. 2008, 49, 4670–4673.
[11] E. Schweizer, A. Hoffmann-Roder, J. A. Olsen, P. Seiler, U. Obst-Sander, B.
Wagner, M. Kansy, D. W. Banner, F. Diederich, Org. Biomol. Chem. 2006,
4, 2364–2375.
[12] A. J. Folkes, K. Ahmadi, W. K. Alderton, S. Alix, S. J. Baker, G. Box, I. S.
Chuckowree, P. A. Clarke, P. Depledge, S. A. Eccles, L. S. Friedman, A.
Hayes, T. C. Hancox, A. Kugendradas, L. Lensun, P. Moore, A. G. Olivero,
J. Pang, S. Patel, G. H. Pergl-Wilson, F. I. Raynaud, A. Robson, N. Saghir,
L. Salphati, S. Sohal, M. H. Ultsch, M. Valenti, H. J. A. Wallweber, N. C.
Wan, C. Wiesmann, P. Workman, A. Zhyvoloup, M. J. Zvelebil, S. J. Shut-
tleworth, J. Med. Chem. 2008, 51, 5522–5532.
[13] a) CCDC 1038302 (2c) and 1038303 (3c) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif; b) C. A. Vock, C. Scolaro, A. D.
Phillips, R. Scopelliti, G. Sava, P. J. Dyson, J. Med. Chem. 2006, 49, 5552–
5561; c) C. A. Vock, W. H. Ang, C. Scolaro, A. D. Phillips, L. Lagopoulos, L.
Juillerat-Jeanneret, G. Sava, R. Scopelliti, P. J. Dyson, J. Med. Chem. 2007,
50, 2166–2175.
[14] G. E. Bꢂchel, A. Gavriluta, M. Novak, S. M. Meier, M. A. Jakupec, O.
Cuzan, C. Turta, J.-B. Tommasino, E. Jeanneau, G. Novitchi, D. Luneau,
V. B. Arion, Inorg. Chem. 2013, 52, 6273–6285.
[15] M. Patra, T. Joshi, V. Pierroz, K. Ingram, M. Kaiser, S. Ferrari, B. Spingler, J.
Keiser, G. Gasser, Chem. Eur. J. 2013, 19, 14768–14772.
[16] a) M. Hanif, S. M. Meier, W. Kandioller, A. Bytzek, M. Hejl, C. G. Hartinger,
A. A. Nazarov, V. B. Arion, M. A. Jakupec, P. J. Dyson, B. K. Keppler, J.
Inorg. Biochem. 2011, 105, 224–231; b) M. Hanif, A. A. Nazarov, C. G.
Hartinger, W. Kandioller, M. A. Jakupec, V. B. Arion, P. J. Dyson, B. K. Kep-
pler, Dalton Trans. 2010, 39, 7345–7352; c) W. H. Ang, E. Daldini, C. Sco-
laro, R. Scopelliti, L. Juillerat-Jeannerat, P. J. Dyson, Inorg. Chem. 2006,
45, 9006–9013.
[17] S. J. Dougan, P. J. Sadler, Chimia 2007, 61, 704–715.
[18] S. J. Dougan, M. Melchart, A. Habtemariam, S. Parsons, P. J. Sadler, Inorg.
Chem. 2006, 45, 10882–10894.
Acknowledgements
We thank the University of Auckland, Genesis Oncology Trust
(GOT-1263-RPG), the Austrian Science Fund (Schrçdinger fellow-
ship to M.H.), and COST CM1105 for financial support. We are
grateful to Tanya Groutso for collecting the X-ray crystal data
and Nick Lloyd for ESI-MS analyses.
Keywords: bioorganometallics
maleimide · osmium · ruthenium
·
cancer
·
cytotoxicity
·
Received: November 11, 2014
Published online on && &&, 0000
&
ChemPlusChem 0000, 00, 0 – 0
www.chempluschem.org
6
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
Right on target: Functionalization of
anticancer-active metal complexes with
a maleimide moiety is aimed to facili-
tate the selective reaction with human
serum albumin to exploit passive target-
ing of tumors through the enhanced
permeability and retention (EPR) effect
(see figure).
S. Moon, M. Hanif,* M. Kubanik,
H. Holtkamp, T. Sçhnel, S. M. F. Jamieson,
C. G. Hartinger*
&& – &&
Organoruthenium and Osmium
Anticancer Complexes Bearing
a Maleimide Functional Group:
Reactivity to Cysteine, Stability, and
Cytotoxicity
ChemPlusChem 0000, 00, 0 – 0
www.chempluschem.org
7
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ