Structural tuning of organoruthenium compounds allows oxidative switch to control ER stress pathways and bypass multidrug resistance

Article abstract of DOI:10.1039/c6sc00268d

Multidrug resistance (MDR) is a major impediment to the success of chemotherapy in many cancer types. One particular MDR mechanism is the inherent or acquired adaptation of the cellular survival pathways that render malignant cells resistant to apoptotic

Full text of DOI:10.1039/c6sc00268d

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
Structural tuning of organoruthenium compounds
allows oxidative switch to control ER stress
pathways and bypass multidrug resistance†
Cite this: DOI: 10.1039/c6sc00268d
ab
cd
be
cd
Mun Juinn Chow, Cynthia Licona, Giorgia Pastorin, Georg Mellitzer,
ab
cd
Wee Han Ang* and Christian Gaiddon*
Multidrug resistance (MDR) is a major impediment to the success of chemotherapy in many cancer types.
One particular MDR mechanism is the inherent or acquired adaptation of the cellular survival pathways that
render malignant cells resistant to apoptotic cell death. Since most drugs act through apoptosis,
compounds capable of inducing alternative forms of programmed cell death (PCD) can potentially be
harnessed to bypass MDR. We investigated two organoruthenium complexes, RAS-1H and RAS-1T, and
demonstrated that although they both induced non-apoptotic PCD through ER stress pathways, their
modes-of-action were drastically different despite modest structural variations. RAS-1T acted through
ROS-mediated ER stress while RAS-1H was ROS-independent. We further showed that they were more
efficacious against apoptosis-resistant cells compared to clinical drugs including oxaliplatin. This work
provides the basis for underpinning ER stress modulation using metal complexes to bypass apoptosis
resistance.
Received 20th January 2016
Accepted 25th February 2016
DOI: 10.1039/c6sc00268d
www.rsc.org/chemicalscience
strategies to overcome this mechanism of MDR involves the
restoration of the expression or function of the pro-apoptotic
Introduction
Chemotherapy remains one of the major treatment options for gene in cancer cells through chemical or genetic modula-
1
17,18
many cancer types. However, the effectiveness of chemothera- tion.
A more attractive strategy would involve directly
2
peutic treatments is frequently diminished due to the multi- bypassing this mechanism of MDR entirely to induce cancer cell
19
drug-resistance (MDR) phenotype found in many cancers that death via non-apoptotic programmed cell death (PCD).
are associated with a poor clinical outcome such as gastric
With the discovery of the antitumoural activity of cisplatin
cancer, the third and h leading cause of cancer mortality in (CDDP), a widely used Pt -based clinical drug for cancer
One dominant chemotherapy, metallodrugs have received revived interest and
MDR mechanism in these cancers was determined to be the attention in recent years. Yet, as with most clinical drugs
defective or selective adaptation of apoptotic pathways, which including CDDP, the majority of the metallodrugs investigated
II
3
–6
men and women worldwide, respectively.
20
7
–11
results in resistance to apoptosis.
In addition, the over- act via apoptotic pathways and are therefore subject to classical
expression of the membrane-bound ‘efflux’ transporter P-gp, MDR limitations. There have only been a handful of metal-
one of the main causes of MDR, is also known to inhibit locomplexes shown to exert cytotoxicity via other alternative
1
2–14
apoptosis by preventing caspase activation.
clinically used drugs act by inducing apoptosis as their primary exhibited the ability to induce type II autophagic cell death.
Given that most forms of PCD. Several Au-, Ru- and Fe-based complexes have
21–23
II
I
15,16
mode-of-action,
it is not surprising that the success rates of A class of Cu thioxotriazole complexes and a Cu triazole–
chemotherapy in these cancers have been poor. One of the few phosphine complex have demonstrated the ability to induce
24,25
V
paraptosis in several cancer cell lines.
More recently, two Re
oxo complexes were reported to induce cell death via a form of
programmed necrosis called necroptosis. However, their
ability to overcome MDR mechanisms has never been validated.
a
Department of Chemistry, National University of Singapore, 3 Science Drive 3, 117543
26
Singapore. E-mail: chmawh@nus.edu.sg; Tel: +65 6516 5131
b
NUS Graduate School for Integrative Sciences and Engineering, Singapore
c
U1113 INSERM, 3 Avenue Moli `e re, Strasbourg 67200, France. E-mail: gaiddon@ Therefore, new metallocomplexes with well-delineated modes
unistra.fr; Tel: +33 68 52 53 56
of action, particularly via alternative PCD, could constitute
a new strategy to overcome MDR.
d
Section Oncology, FMTS, Strasbourg University, Strasbourg, France
e
Department of Pharmacy, National University of Singapore, 18 Science Drive 4,
We earlier reported the combinatorial synthesis and
evaluation of a new class of water-soluble/stable half-sandwich
1
17543 Singapore
†
Electronic supplementary information (ESI) available: Preparation,
II
Ru arene Schiff-base (RAS) complexes via a coordination-
characterization protocol and data for RAS-1H and RAS-1T, and the protocol
27
directed 3-component assembly.
RAS-1T, which bore
and data for in vitro studies. See DOI: 10.1039/c6sc00268d
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Scheme 1 Synthesis route for the preparation of Ruthenium(II) Schiff-
base (RAS) complexes.
Fig. 1 Differential ER stress pathway activation by RAS complexes
leads to an alternative (non-apoptotic) PCD that bypasses drug resis- the desired [Ru(arene)Cl
]
precursor (0.5 equiv., arene ¼ HMB
2
2
tance mechanisms.
or TIPB) in MeOH. Purication via ash column chromatog-
raphy gave both compounds in good yields. The complexes were
1
characterized using H NMR and ESI-MS and their purity was
conrmed using RP-HPLC and elemental analysis.
We compared the physicochemical properties of both
complexes. Stability studies using UV-Vis spectroscopy showed
that both complexes were stable to aquation, and were not
prone to react with sulfur- or nitrogen-containing biomolecules
triisopropylbenzene (TIPB) and iminoquinoline ligands, was
identied as a lead candidate as it was highly efficacious against
several cancer cell lines yet exhibited attributes distinct from
classical alkylating agents, such as CDDP and reported anti-
II
cancer Ru complexes. For instance, RAS-1T was stable against
hydrolysis and did not interact directly with dGMP nucleotides.
In addition, it did not induce p53 expression in treated cells,
commonly associated with DNA damage. Using gastric cancer
as a model in this present study and in comparison with a newly
synthesized hexamethylbenzene (HMB) analogue, RAS-1H, we
demonstrate for the rst time that varying the facially-bound
arene ligands can have drastic effects on the compounds' mode-
of-action leading to cell death, activating either ROS-dependent
or ROS-independent ER stress pathways (Fig. 1). We show that
(
Fig. S4†). The lipophilicity of RAS-1H and RAS-1T was deter-
mined from their extent of partitioning between n-octanol and
water. The log POW for RAS-1H and RAS-1T was determined to
be ꢀ1.40 and ꢀ0.85, respectively, showing that both
compounds were relatively hydrophilic despite having highly
hydrophobic ligands, presumably due to the cationic nature of
the compounds.
the independent activation of both pathways leads to non- Distinct cytotoxicity proles of RAS-1H and RAS-1T
apoptotic PCD in treated cells and that this strategy could be
harnessed to overcome apoptosis-resistance.
We investigated the potentially contrasting mode-of-action of
RAS-1H and RAS-1T by rst testing their efficacy against a panel
of gastric and colorectal cancer cell lines (Table S1†). As ex-
Results and discussion
pected, RAS-1T displayed a low micromolar IC50 value in all 4
Synthesis and characterization
We previously reported the combinatorial synthesis of 450
distinct RAS complexes via the co-ordination of in situ assem-
27
bled Schiff base ligands to organoruthenium scaffolds.
Screening of this RAS library revealed RAS-1T to exhibit p53-
independent activity and to be a lead candidate for a metal-
locomplex with antiproliferative activity that is distinct from
classical alkylating agents, e.g. cisplatin, for further mode-of-
action elucidation. We postulated that minor structural varia-
tions could inuence cellular pathway activation, regardless of
the similarities in their physicochemical properties. In order to
discern how the structural variation of the arene ligand could Fig. 2 Varying the arene ligand on RAS complexes changes their
antiproliferative profiles. Cell viability curves of compounds RAS-1H
inuence the mode-of-action, we included the HMB-analogue
RAS-1H in the present study. Both RAS-1H and RAS-1T were
synthesized in good yields and purity (Scheme 1). The chelate
ligand was prepared directly from 4-methoxylaniline and 2-
and RAS-1T. The curve represents the mean ꢁ s.e.m. of three inde-
pendent experiments. RAS-1H displayed a biphasic cytotoxicity curve
unlike RAS-1T, hinting at a different mode-of-action. Two equipotent
concentrations at low-dose (LD) and high-dose (HD) were chosen for
formylquinoline via imine condensation and was treated with cell treatment in proteins and mRNA expression studies.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
cell lines tested. The IC50 of RAS-1T was also 34 times and 7 species (ROS) generation has been implicated in the mode-of-
II
times lower than that of cisplatin in the gastric cancer cell lines action of many Ru complexes given the low energy barrier of
II
III
AGS and KATOIII, respectively. In comparison, RAS-1H the Ru /Ru redox states, although it is not always clear that
28–30
demonstrated slightly more modest activity with IC50 values 11 ROS generation was the cause of the observed cytotoxicity.
times lower than cisplatin in AGS and 1.5 times higher in Cellular ROS levels were determined using a commercial cell-
KATOIII. Closer scrutiny of the cytotoxicity curve of RAS-1H in permeable ROS probe, carboxy-H DCFDA. The ROS levels were
2
AGS revealed a bi-phasic prole that was different from that of markedly increased in AGS cells treated with both RAS
RAS-1T (Fig. 2). This suggested different modes-of-action for complexes in a time- and concentration-dependent manner. In
RAS-1H and RAS-1T. We determined two equipotent concen- cells treated with both complexes, ROS induction peaked at an
trations at low dose (LD) and high dose (HD) for each complex early time-point of 3 h at similar levels before decreasing as
for cell treatment to further determine the difference in their indicated from the quantitative uorescence measurements
mode-of-action.
(Fig. 3a) and uorescence microscopy images (Fig. 3b and S5†).
To gain an insight into the impact of ROS on the cells at
relevant time-points (6 h and 24 h post-treatment), we examined
the cellular response by looking into the antioxidant defense
mechanism (Scheme S1†). Central to the antioxidant defense is
RAS-1H and RAS-1T induced early time-point ROS and
activated the antioxidant defense mechanism
To unravel the mechanism of action of RAS-1H and RAS-1T,
given the complex interplay of various possible pathways, we
examined oxidative stress levels in treated AGS gastric cancer
cells and the subsequent cellular response. Reactive oxygen
31
the transcription factor Nrf-2, which is responsible for the
3
2
regulation of several downstream target genes such as gclc,
33
34
mrp2, and nqo1, each having various roles in the mediation
Fig. 3 Complexes RAS-1H and RAS-1T induce early time-point ROS and activate the cellular antioxidant defence mechanism. (a) Detection
of ROS with carboxy-H
2
DCFDA (20 mM) after treatment with RAS-1H and RAS-1T for 3 h, 6 h and 9 h using a microplate assay. Mean ꢁ s.e.m.
*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; Student's t test). (b) Detection of ROS with a fluorescence microscope after treatment for 6 h.
c) Western blot analysis of Nrf-2, a central protein in cellular antioxidant defence and (d) expression levels of the Nrf-2 target gene in AGS cells
(
(
after treatment with RAS-1H, RAS-1T and cisplatin at LD and HD for 6 h and 24 h. Homogeneous protein loading determined with reference to
actin and gene expression normalized against tbp levels.
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
of oxidative stress. When ROS levels are elevated, Nrf-2 is acti- investigated if RAS-1H and RAS-1T induced ER stress biomarkers
vated and its target genes expression increases. Increased and the implications to their antiproliferative activity. RAS-1H
expression of Nrf-2 was observed aer 6 h of treatment at HD for and RAS-1T induced ER stress via the IRE1a/XPB-1s pathway as
both RAS-1H and RAS-1T (Fig. 3c). Similarly, mRNA levels for all seen from the increased accumulation of XBP-1s aer 6 h of
three Nrf-2 target genes increased aer 6 h of treatment at HD treatment at HD (Fig. S6†). The induction of the downstream
(
Fig. 3d). This was consistent with the induction of ROS target CHOP aer both 6 h and 24 h of HD treatment also
39
observed at early-time points. In contrast, cells treated with conrmed the ER stress induction. While RAS-1H induced high
CDDP remained at basal levels. However, there were differences levels of CHOP expression, RAS-1T favored XBP-1s splicing,
in the Nrf-2 expression and activity patterns for both alluding to different signaling pathways.
compounds. Firstly, the protein levels for Nrf-2 were higher aer
We therefore examined whether ROS production and the
RAS-1H treatment compared to RAS-1T. Secondly, Nrf-2 induc- subsequent ER stress were critical to the antiproliferative
tion remained elevated aer 24 h for both concentrations of activity of RAS-1H and RAS-1T on AGS cells. We employed
RAS-1H while Nrf-2 returned to basal levels aer 24 h for N-acetylcysteine (NAC) as a ROS quencher and determined the
RAS-1T. Lastly, the RNA level of gclc was more signicantly cell viability and protein level expression of Nrf-2, XBP-1s and
increased by RAS-1H than RAS-1T. Taken together, the results CHOP under various conditions, in the absence or presence of
indicate that the antioxidant defense for RAS-1H was switched NAC. In the case of RAS-1T, cytotoxicity up to 2.5 mM (HD) was
on for an extended duration, underscoring the mechanistic completely removed by the co-treatment with NAC. At higher
differences in oxidative stress induction between the RAS levels of RAS-1T exposure, i.e. 10 mM which typically led to the
complexes despite similar early time-point ROS production.
100% loss of viable cells, 50% cell viability was restored with
NAC co-treatment (Fig. 4a). Furthermore, Nrf-2 levels were low
in cells co-treated with NAC (Fig. 4b) indicating that the anti-
oxidant defense was not switched on. In keeping with these
observations, ER stress markers XBP-1s and CHOP were also
Differential induction of the ER stress pathway by RAS-1H and
RAS-1T
The relationship between ROS and endoplasmic reticulum (ER) suppressed in cells co-treated with NAC, suggesting a low UPR
stress is well-established and ROS could occur either upstream and reduced ER stress. Taken together, the results point toward
35,36
(cause) or downstream (product) of ER stress.
ER stress can be ROS production being critical for the antiproliferative activity
characterized by the unfolded protein response (UPR), which can of RAS-1T (causal) and it exerting its mode-of-action via
37
lead to recovery, cellular dysfunction or cell death. Three ROS-mediated ER stress. In contrast, ROS quenching did not
distinct UPR signaling pathways have been identied, namely the suppress RAS-1H-induced cell death at treatments of less than
38
PERK/eif2a, IRE1a/XPB-1s and ATF6 pathways (Scheme S2†).
30 mM (HD) (Fig. 4a). The NAC treatment did not block RAS-1H-
Since elevated ROS levels point towards ER stress and UPR, we induced Nrf-2 protein levels aer 6 h of treatment and had only
Fig. 4 Differential activation of ROS-independent and ROS-mediated ER stress pathways by RAS-1H and RAS-1T. (a) Cell viability (%) of AGS cells
treated with RAS-1H and RAS-1T for 48 h with and without NAC (2 mM). Mean ꢁ s.e.m. (**p < 0.01, ***p < 0.001, ****p < 0.0001; two-tailed
Student's t-test). (b) Western blot analysis of ER stress protein markers in AGS cells after treatment with RAS-1H and RAS-1T at HD, with and
without NAC (2 mM). Homogeneous protein loading determined with reference to actin.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
a partial effect at the 24 h time-point. Moreover, ROS quenching cells treated with RAS-1H and RAS-1T did not exhibit such
also did not suppress XBP-1s or CHOP induction (Fig. 4b). In expression proles, regardless of the concentration or duration
fact, the co-treatment of RAS-1H with NAC increased ER stress of treatment. Furthermore, the two distinct hallmarks of
as indicated by the marked increase in XBP-1s and CHOP apoptosis, namely (i) the activation of apoptosis-executor cas-
accumulation. This suggested that the cell death induced by pases, and (ii) morphological changes consistent with apoptosis
42
RAS-1H was not dependent on the elevated ROS levels (Fig. 3a) were also absent in cells treated with RAS-1H and RAS-1T. The
and that ROS production was not critical for its mode-of-action co-treatment of cells with a broad-spectrum caspase-inhibitor,
(
by-product).
Z-VAD-FMK, did not reduce the efficacy of either RAS-1H or
RAS-1T, as seen from their unchanged IC50 values (Fig. 5c) but it
provided cytoprotection from CDDP. In addition, the cell death
morphology induced by RAS-1H and RAS-1T aer 24 h was
markedly different from that induced by CDDP (Fig. 5d and
S7†). For cells treated with cisplatin, the ‘budding’ and forma-
tion of smaller apoptotic bodies were observed. In contrast, cells
treated with RAS-1H and RAS-1T did not display the same
morphological changes commonly seen in apoptotic cell death.
To rule out the possibility of necrotic cell death, we also tested
the activity of RAS-1H and RAS-1T in the absence and presence
of IM-54, an inhibitor of oxidative stress-induced necrosis. The
cell viability of cells treated with various concentrations of
RAS-1H and RAS-1T did not change signicantly in the presence
of IM-54, indicating that necrosis was unlikely to be the mode of
RAS-1H and RAS-1T induce non-apoptotic programmed cell
death
The contrasting yet well-dened antiproliferative pathways
brought about by RAS-1H and RAS-1T, both of which invoked
ROS production, could be harnessed to bypass the conventional
apoptotic pathways induced by classical drugs such as CDDP.
We therefore compared the apoptosis biomarkers commonly
40,41
induced by CDDP treatment with the RAS complexes.
Western blot analyses of AGS cells treated with CDDP revealed
signicant upregulation in p53 expression and cleavage of
caspase 3 and PARP-1 (Fig. 5a). A concomitant increase of bax
and decrease of bcl-2 gene expression was also observed aer
2
4 h (Fig. 5b), indicative of apoptosis induction. In contrast,
Fig. 5 Complexes RAS-1H and RAS-1T induce non-apoptotic cell death. (a) Western blot analysis of proteins related to the apoptosis pathway
and (b) expression levels of pro-apoptotic and anti-apoptotic genes in AGS cells after treatment with RAS-1H, RAS-1T and cisplatin at LD and HD
for 6 h and 24 h. Homogeneous protein loading determined with reference to actin and gene expression normalized against tbp levels. (c) IC50
values of RAS-1H, RAS-1T and cisplatin after 48 h of treatment with and without an apoptosis inhibitor, Z-VAD-FMK (5 mM). Mean ꢁ s.e.m.
(
1
*p < 0.05; two-tailed Student's t-test). (d) Microscope images of AGS cells treated with RAS-1H, RAS-1T and cisplatin for 24 h (scale bar ¼
00 mm). Examples of smaller apoptotic bodies due to ‘budding’ are pointed out with white arrows.
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
cell death (Fig. S8†). The lack of caspase involvement and the compared to the clinical drugs oxaliplatin (OXP), etoposide
marked difference in cell death morphology compared to (ETOP), 5-uorouracil (5-FU) and doxorubicin (DOX) using their
CDDP-treated cells suggested that RAS-1H and RAS-1T induced resistance factor (RF), expressed as a factor of IC50 [TC7] against
a form of non-apoptotic PCD.
either IC50 [HT-29] or IC50 [HCT116]. OXP was used instead of
CDDP as it is the leading drug for the treatment of colorectal
cancer for which CDDP is poorly efficacious.
Both RAS-1H and RAS-1T were the least affected by TC7
compared to the frequently used clinical drugs (Fig. 6 and S9,
and Table S2†). In particular, RAS-1T exhibited the lowest
resistance factors (RFs) amongst all tested compounds of 4.5
RAS-1H and RAS-1T bypass the apoptosis resistance
mechanism in colorectal cancer cell lines
In order to validate the ability of the RAS complexes to bypass
apoptosis resistance, we employed the apoptosis-resistant TC7
cell line as a functional cell model. TC7 is a cell line cloned from
parental colorectal adenocarcinoma Caco-2 cells using limited
dilution and is found to be highly resistant to conventional drug
(TC7/HT-29) and 2.8 (TC7/HCT116) while the RAS-1H values
were 9.2 and 9.0, respectively. In contrast, the RF values for OXP
were 26.9 and 10.2 while 5-FU exhibited RFs exceeding 111 and
43
treatment. It has a p53-null status and a higher basal expres-
123. Furthermore, RAS-1T also exhibited low micromolar IC50
sion of anti-apoptotic Bcl-2 and Bcl-x , as well as a lower
L
values in TC7, highlighting its high efficacy in the apoptosis-
resistant cell lineage. These results suggested that the
apoptosis-resistance of TC7 did not signicantly impact the
effectiveness of both RAS complexes, unlike the tested clinical
drugs, thereby validating in a functional cell model the
hypothesis that RAS-1H and RAS-1T could bypass apoptosis
resistance via the induction of alternative PCDs (Scheme S3†).
expression of pro-apoptotic Bax. A previous study done on the
activity of 5-uorouracil on a panel of colorectal cancer cell lines
including TC7 showed a positive correlation between their
resistance to 5-uorouracil treatment and their basal (Bcl-2 +
44
L
Bcl-x )/Bax expression ratio, suggesting that the drug-resis-
tance observed in TC7 was due in part to apoptosis-resistance.
The colorectal cancer cell lines HCT116 and HT-29 were used as
non-resistant models. The activities of RAS-1H and RAS-1T were
Conclusion
The variation of the arene ligand in half-sandwich Ru
complexes is commonly used as a means to modulate physical
properties such as hydrophobicity and solubility. Limited
studies have been done on how arene ligands could inuence
the cellular mode-of-action. Two separate studies on a class of
Ru(II)–arene complexes bearing ethylenediamine ligands have
demonstrated that arene ligands, with varying hydrophobicity
and p-acidity, could be used to modulate the rate of hydrolysis
45,46
and extent of p-stacking of DNA bases.
A separate study
showed that the functionalization of the arene with a maleimide
moiety allowed for the selective delivery of the complexes via
47
selective binding with thiol-containing biomolecules.
However, these studies focused on how arene modication
could be used to modulate the physicochemical properties and
target-binding ability without demonstrating any differential
activation of cellular pathways.
In contrast, our current study represents one of the few such
studies, demonstrating that a subtle change in the arene
ligands on the RAS complexes had a drastic effect on its mode-
of-action, switching the ability of the drug to induce cell death
from a ROS-mediated ER stress pathway to a ROS-independent
pathway. Since the variation of the arene ligand did not
signicantly change the physiochemical properties such as
stability, reactivity to biomolecules or hydrophobicity, the
observed difference is most likely inuenced directly by the site-
specic structural changes of the arene ligand. Although the
induction of ER stress by other metal complexes has been
Fig. 6 RAS-1H and RAS-1T are least affected by the resistance
mechanism of drug-resistant TC7 cells, compared to other clinically
approved drugs. (a) Resistance factors are calculated by taking the ratio previously reported,
of IC50 from 3 independent experiments in apoptosis-resistant TC7
3
0,48,49
mechanistic insights into ER stress
activation and the subsequent implications of structural factors
were not discussed. Our current study provides the rst
molecular basis for ER stress activation, highlighting the
complex relationship between the structure of the compound
and its impact on the mode-of-action via ER stress. It is
over the mean IC50 in non-resistant HT29 and HCT116. Mean ꢁ s.e.m.
(
*p < 0.05, **p < 0.01, ***p < 0.001; one-way ANOVA test with Tukey
#
post-hoc test). IC50 in TC7 > 1000 mM in all 3 experiments and the
resistance factor shown is the lowest possible. (b) Comparison of the
cell viability curves for RAS-1T and 5-FU in TC7 and HT-29.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
noteworthy that the structural tuning also impacted its ability to 17 K. Friedrich, T. Wieder, C. Von Haefen, R. J ¨a nicke,
bypass cancer cell MDR mechanisms, as seen in the two-fold
difference in the resistance factor between RAS-1H and RAS-1T.
K. Schulze-Osthoff, B. D ¨o rken and P. T. Daniel, Oncogene,
2001, 20, 2749–2760.
These factors should be taken into consideration when 18 H. Ehrhardt, F. Wachter, M. Maurer, K. Stahnke and
designing such compounds as anticancer agents to improve the I. Jeremias, Clin. Cancer Res., 2011, 17, 7605–7613.
treatment outcome for MDR cancers. Based on the current 19 S. Fulda and K. M. Debatin, Oncogene, 2006, 25, 4798–4811.
study, we hypothesized that the structural change to the RAS 20 K. D. Mjos and C. Orvig, Chem. Rev., 2014, 114, 4540–4563.
complexes could have affected the selectivity of binding to their 21 M. Librizzi, A. Longo, R. Chiarelli, J. Amin, J. Spencer and
target, especially if the cellular target contains several homol-
ogous isoforms. This could have caused the observed differ- 22 S. Tian, F. M. Siu, S. C. Kui, C. N. Lok and C. M. Che, Chem.
ences in the mode-of-action between the two complexes. Commun., 2011, 47, 9318–9320.
Nevertheless, more studies are required to identify the cellular 23 J. Yuan, Z. Lei, X. Wang, F. Zhu and D. Chen, Metallomics,
C. Luparello, Chem. Res. Toxicol., 2012, 25, 2608–2616.
target(s) of RAS-1H and RAS-1T before this can be validated.
2015, 7, 896–907.
2
4 S. Tardito, I. Bassanetti, C. Bignardi, L. Elviri, M. Tegoni,
C. Mucchino, O. Bussolati, R. Franchi-Gazzola and
L. Marchio, J. Am. Chem. Soc., 2011, 133, 6235–6242.
Conflict of interest
The authors declare no competing nancial interest.
25 C. Marzano, M. Pellei, D. Colavito, S. Alidori, G. G. Lobbia,
V. Gandin, F. Tisato and C. Santini, J. Med. Chem., 2006,
4
9, 7317–7324.
Acknowledgements
2
6 K. Suntharalingam, S. G. Awuah, P. M. Bruno,
T. C. Johnstone, F. Wang, W. Lin, Y. R. Zheng, J. E. Page,
M. T. Hemann and S. J. Lippard, J. Am. Chem. Soc., 2015,
137, 2967–2974.
The authors acknowledge ARC, Ligue contre le cancer, CNRS,
European COST action CM1105 and the Singapore Ministry of
Education (R143-000-638-112) for funding as well as CMMAC-
NUS for performing elemental analysis and ICP-OES analysis.
27 M. J. Chow, C. Licona, D. Y. Q. Wong, G. Pastorin,
C. Gaiddon and W. H. Ang, J. Med. Chem., 2014, 57, 6043–
6059.
Notes and references
2
8 Z. Luo, L. Yu, F. Yang, Z. Zhao, B. Yu, H. Lai, K. H. Wong,
S. M. Ngai, W. Zheng and T. Chen, Metallomics, 2014, 6,
1480–1490.
1
2
3
4
Cancer Treatment & Survivorship Facts and Figures 2014-2015,
American Cancer Society, Alanta, 2014.
G. Morgan, R. Ward and M. Barton, Clin. Oncol., 2004, 16, 29 C. Tan, S. Lai, S. Wu, S. Hu, L. Zhou, Y. Chen, M. Wang,
549–560.
Y. Zhu, W. Lian, W. Peng, L. Ji and A. Xu, J. Med. Chem.,
2010, 53, 7613–7624.
30 V. Vidimar, X. Meng, M. Klajner, C. Licona, L. Fetzer,
S. Harlepp, P. Hebraud, M. Sidhoum, C. Sirlin,
J. P. Loeffler, G. Mellitzer, G. Sava, M. Pfeffer and
C. Gaiddon, Biochem. Pharmacol., 2012, 84, 1428–1436.
Global Cancer Facts and Figures 3rd Editions, American
Cancer Society, Atlanta, 2015.
D. Zhang and D. Fan, Expert Rev. Anticancer Ther., 2007, 7,
1369–1378.
5
6
R. Agarwal and S. B. Kaye, Nat. Rev. Cancer, 2003, 3, 502–516.
A.-M. Florea and D. B u¨ sselberg, J. Loc. Glob. Health Sci., 2013, 31 W. Li and A. N. Kong, Mol. Carcinog., 2009, 48, 91–104.
, DOI: 10.5339/jlghs.2013.2. 32 S. C. Lu, Mol. Aspects Med., 2009, 30, 42–59.
F. H. Igney and P. H. Krammer, Nat. Rev. Cancer, 2002, 2, 33 V. Vollrath, A. M. Wielandt, M. Iruretagoyena and
77–288. J. Chianale, Biochem. J., 2006, 395, 599–609.
E. Hervouet, M. Cheray, F. M. Vallette and P. F. Cartron, 34 D. Ross, J. K. Kepa, S. L. Winski, H. D. Beall, A. Anwar and
2
7
8
9
2
Cells, 2013, 2, 545–573.
S. Fulda, Int. J. Cancer, 2009, 124, 511–515.
D. Siegel, Chem.–Biol. Interact., 2000, 129, 77–97.
35 J. D. Malhotra and R. J. Kaufman, Antioxid. Redox Signaling,
2007, 9, 2277–2293.
1
1
0 D. M. O'Gorman and T. G. Cotter, Leukemia, 2001, 15, 21–34.
1 Y. Pommier, O. Sordet, S. Antony, R. L. Hayward and 36 B. Bhandary, A. Marahatta, H. R. Kim and H. J. Chae, Int. J.
K. W. Kohn, Oncogene, 2004, 23, 2934–2949. Mol. Sci., 2012, 14, 434–456.
2 Y. Zu, Z. Yang, S. Tang, Y. Han and J. Ma, Molecules, 2014, 19, 37 C. Hetz, Nat. Rev. Mol. Cell Biol., 2012, 13, 89–102.
1
13061–13075.
38 D. Ron and P. Walter, Nat. Rev. Mol. Cell Biol., 2007, 8, 519–
13 K. M. Tainton, M. J. Smyth, J. T. Jackson, J. E. Tanner,
529.
L. Cerruti, S. M. Jane, P. K. Darcy and R. W. Johnstone, 39 S. Oyadomari and M. Mori, Cell Death Differ., 2004, 11, 381–
Cell Death Differ., 2004, 11, 1028–1037.
389.
4 M. J. Smyth, E. Krasovskis, V. R. Sutton and R. W. Johnstone, 40 Z. H. Siddik, Oncogene, 2003, 22, 7265–7279.
1
1
1
Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 7024–7029.
5 U. Fischer and K. Schulze-Osthoff, Cell Death Differ., 2005,
41 H. H. Sigurethsson, C. W. Olesen, R. Dybboe, G. Lauritzen
and S. F. Pedersen, Mol. Cancer Res., 2015, 13, 63–77.
1
2, 942–961.
6 D. Wang and S. J. Lippard, Nat. Rev. Drug Discovery, 2005, 4,
07–320.
42 A. Saraste and K. Pulkki, Cardiovasc. Res., 2000, 45, 528–537.
3
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
4
3 I. Chantret, A. Rodolosse, A. Barbat, E. Dussaulx, E. Brot-
Laroche, A. Zweibaum and M. Rousset, J. Cell Sci., 1994,
P. J. Sadler, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 18269–
18274.
1
07, 213–225.
47 M. Hanif, A. A. Nazarov, A. Legin, M. Groessl, V. B. Arion,
M. A. Jakupec, Y. O. Tsybin, P. J. Dyson, B. K. Keppler and
C. G. Hartinger, Chem. Commun., 2012, 48, 1475–1477.
48 K. Suntharalingam, T. C. Johnstone, P. M. Bruno, W. Lin,
M. T. Hemann and S. J. Lippard, J. Am. Chem. Soc., 2013,
135, 14060–14063.
4
4 S. Violette, L. Poulain, E. Dussaulx, D. Pepin, A. M. Faussat,
J. Chambaz, J. M. Lacorte, C. Staedel and T. Lesuffleur, Int. J.
Cancer, 2002, 98, 498–504.
5 H. Chen, J. A. Parkinson, R. E. Morris and P. J. Sadler, J. Am.
Chem. Soc., 2003, 125, 173–186.
4
4
6 F. Wang, A. Habtemariam, E. P. L. van der Geer, R. Fernandez, 49 R. Cao, J. Jia, X. Ma, M. Zhou and H. Fei, J. Med. Chem., 2013,
M. Melchart, R. J. Deeth, R. Aird, S. Guichard, F. P. A. Fabbiani,
P. Lozano-Casal, I. D. H. Oswald, D. I. Jodrell, S. Parsons and
56, 3636–3644.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016