Insights into the antiproliferative mechanism of (C^N)-chelated half-sandwich iridium complexes

Article abstract of DOI:10.1039/d0dt03414b

Transition metal-based anticancer compounds, as an alternative to platinum derivatives, are raising scientific interest as they may present distinct although poorly understood mechanisms of action. We used a structure-activity relationship-based methodology to investigate the chemical and biological features of a series of ten (C^N)-chelated half-sandwich iridiumIII complexes of the general formula [IrCp?(phox)Cl], where (phox) is a 2-phenyloxazoline ligand forming a 5-membered metallacycle. This series of compounds undergoes a fast exchange of their chlorido ligand once solubilised in DMSO. They were cytotoxic to HeLa cells with IC50 values in the micromolar range and induced a rapid activation of caspase-3, an apoptosis marker. In vitro, the oxidative power of all the complexes towards NADH was highlighted but only the complexes bearing substituents on the oxazoline ring were able to produce H2O2 at the micromolar range. However, we demonstrated using a powerful HyPer protein redox sensor-based flow cytometry assay that most complexes rapidly raised intracellular levels of H2O2. Hence, this study shows that oxidative stress can partly explain the cytotoxicity of these complexes on the HeLa cell line and gives a first entry to their mechanism of action. This journal is

Full text of DOI:10.1039/d0dt03414b

Dalton
Transactions
View Article Online
View Journal
PAPER
Insights into the antiproliferative mechanism of
(C^N)-chelated half-sandwich iridium complexes†
Cite this: DOI: 10.1039/d0dt03414b
a
Robin Ramos, a,b Jérémy M. Zimbron,a Serge Thorimbert,
Lise-Marie Chamoreau, a Annie Munier,b Candice Botuha, a Anthi Karaiskou,
b
Michèle Salmain *a and Joëlle Sobczak-Thépot
*
b
Transition metal-based anticancer compounds, as an alternative to platinum derivatives, are raising scien-
tific interest as they may present distinct although poorly understood mechanisms of action. We used a
structure–activity relationship-based methodology to investigate the chemical and biological features of
a series of ten (C^N)-chelated half-sandwich iridiumIII complexes of the general formula [IrCp*(phox)Cl],
where (phox) is a 2-phenyloxazoline ligand forming a 5-membered metallacycle. This series of com-
pounds undergoes a fast exchange of their chlorido ligand once solubilised in DMSO. They were cytotoxic
to HeLa cells with IC50 values in the micromolar range and induced a rapid activation of caspase-3, an
apoptosis marker. In vitro, the oxidative power of all the complexes towards NADH was highlighted but
only the complexes bearing substituents on the oxazoline ring were able to produce H2O2 at the micro-
molar range. However, we demonstrated using a powerful HyPer protein redox sensor-based flow cyto-
metry assay that most complexes rapidly raised intracellular levels of H2O2. Hence, this study shows that
oxidative stress can partly explain the cytotoxicity of these complexes on the HeLa cell line and gives a
first entry to their mechanism of action.
Received 2nd October 2020,
Accepted 13th November 2020
DOI: 10.1039/d0dt03414b
rsc.li/dalton
transfer (MLCT) transitions, making them popular imaging
probes as well as photodynamic therapy agents.2,3 The second
Introduction
Platinum-based compounds in association with other cytotoxic series, namely the so-called half-sandwich complexes, presents
agents are still among the most frequently prescribed anti- a different coordination pattern with a pseudo-tetrahedral geo-
cancer drugs, despite their severe side effects and resistance metry. The iridium centre is coordinated by a penta-substi-
developed by many tumours. Alternative metal-based tuted cyclopentadienyl ligand (typically pentamethyl cyclopen-
compounds, among which are various rutheniumII and tadienyl, Cp*), a chelating bidentate ligand and often a chlor-
rutheniumIII complexes,1 are raising scientific interest as they ido ligand occupying the last coordination position. The labi-
may present original multi-targeting mechanisms of action lity of this last ligand makes the complexes reactive towards
(MoA) that stand out from the classical DNA-damage inducers various substrates, hence their initial application as hydrogen-
and antimitotics.
ation and C–H activation catalysts.4–6
Iridium complexes provide other potentially useful alterna-
The antiproliferative properties of half-sandwich complexes
tives to platinum-based anticancer drugs. In the iridium(III) bearing a C^N chelating ligand were first reported in 2011.7
family, the octahedral, bis-cyclometallated complexes of the Subsequent reports allowed to delineate the structural para-
general formula [IrIII(C^N)2(N^N)]+, X− are well known for their meters (charge, nature of the chelating ligand, degree of
luminescence properties owing to metal-to-ligand charge lability of the third ligand) governing their cytotoxicity.8–11 For
some complexes, a relationship between cytotoxicity and rapid
raise of total intracellular reactive oxygen species (ROS) was
aInstitut Parisien de Chimie Moléculaire (IPCM), Sorbonne Université, CNRS,
highlighted12 with concomitant upregulation of antioxidant
4 place Jussieu, F-75005 Paris, France.
responses.13 More generally, the ability of molecules to modu-
late the oxidation stress represents an innovative strategy to
preferentially target cancer cells owing to their unique meta-
bolic properties.14,15
Because the relationship between cytotoxicity, in vitro gene-
ration of H2O2 and intracellular H2O2 levels had never been
experimentally demonstrated, we undertook to investigate it
E-mail: michele.salmain@sorbonne-universite.fr
bCentre de Recherche Saint Antoine (CRSA), Sorbonne Université, INSERM,
184 rue du Faubourg Saint Antoine, F-75012 Paris, France.
E-mail: joelle.sobczak@inserm.fr
†Electronic supplementary information (ESI) available: Experimental pro-
cedures; additional NMR and biological studies; videomicroscopy. CCDC
1995094. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/d0dt03414b
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
on a library of half-sandwich iridium(III) complexes. For this 40–99%. Following procedures from the previously reported
purpose, ten complexes Ir1–10 comprising variously substi- structures Ir1,16 Ir217 and Ir8,18 sodium acetate was used to
tuted 2-phenyloxazolines (phox) L1–10 as C^N chelating produce Ir1–10 from [Cp*IrCl2]2 with the corresponding
ligands were synthesized and characterized. The phox scaffold (phox) ligands (L1–10) in moderate to good yields (38–97%).
was selected owing to the readiness to introduce a wide range The phenyl group coordinates the iridium atom from a C–H
of electron-donating or withdrawing substituents on either of bond activation reaction which rate was proven to depend on
the 2 rings and a second stereogenic centre on the oxazoline the electronic nature of the ligand.18 The oxazoline’s hydro-
ring. A structure–activity relationship study was systematically gens in the 1H NMR spectra of all the complexes were magneti-
performed regarding their antiproliferative properties on the cally inequivalent owing to the chiral environment around the
HeLa cell model together with their ability to elicit the pro- iridium atom.
duction of hydrogen peroxide in vitro and in vivo. Using a
Owing to the presence of another stereogenic centre on
robust and selective flow cytometric approach with the chiral ligands L7–10, complexes Ir7 to Ir10, were obtained as
endogenously expressed protein redox sensor HyPer, we mixtures of diastereomers. Ir7–8 were obtained in a 9 : 1 dia-
showed that some of the complexes induce a fast and long- stereomeric ratio18 whereas Ir9–10 were obtained in a 1 : 1
1
lasting perturbation of intracellular H2O2 levels. This behav- ratio measured by H NMR in DMSO-d6. However, a coordinat-
iour was correlated to their ability to catalyse the production of ing environment like DMSO-d6 induces epimerization of the
H2O2 from NADH in aerated physiological environment.
(phox)-chelated complexes at the iridium centre and the dia-
stereomeric excess quantification is no longer relevant. The
molecular structure of Ir1 solved by X-ray diffraction analysis
(Fig. 1, Tables S1 and 2†) showed the expected “piano stool”
geometry typical of this family of complexes.
Results and discussion
Synthesis and characterization
The synthetic route to the ten half-sandwich complexes Ir1–10
with the general structure [Cp*(C^N)IrCl] is reported in
Scheme 1. The substitution patterns aimed at examining the
influence of steric effects and stereochemistry (R2,3) as well as
electronic effects (R1) on the chemical and biological pro-
perties of the complexes.
Ligand L1 is commercially available and compounds L2–10
were obtained in two or three steps in moderate to high yields.
The condensation of β-ethanolamines with para-substituted
benzoyl chlorides was followed by an oxidation step using
excess SOCl2. For L5 and L6 this led to a simple chlorination
of the N-(2-hydroxyethyl) benzamide intermediate. Hence,
another step was required and NaH was employed to induce
cyclization from the nitrogen’s lone pair (procedure B). Global
yields for the synthesis of L2–10 were comprised between
Reactivity towards the DMSO vehicle
DMSO is a widely used administration solvent for metal com-
plexes owing to its ability to solubilise highly lipophilic struc-
tures and its convenient miscibility with water. Despite its rela-
tively low toxicity, DMSO has been shown to exert a negative
effect on the activation process of cisplatin (CDDP)19 and a
rapid formation of a solvento adduct was reported for ruthe-
nium half-sandwich complexes.20
The vehicle used for further biological tests being DMSO,
the stability of the complexes in this solvent was investigated
1
by H NMR spectroscopy. The characteristic chemical shift of
the Cp* ligand was used to monitor the conversion of Ir1–10
to [Ir-DMSO-d6]+ (Fig. S1†). At 297 K, most complexes undergo
a very fast chlorido to DMSO ligand exchange with half-lives
below 12 min (Ir1, Ir2, Ir4, Ir6, Ir9–10) albeit with an electronic
effect observed for Ir3 and Ir5 linked to the σ-I Hammett con-
stant of fluoro and nitro substituents. Ir5′s half-life in DMSO-
d6 was the longest of the series, that is 270 min, thus 20-fold
Scheme 1 Synthetic route used to access the (phox) ligands L1–10 and
the corresponding half-sandwich iridium chlorido complexes Ir1–10,
chlorido ligand exchange in DMSO. a Procedure A. (a) CH2Cl2, r.t.; (b)
SOCl2 (4 eq.), CH2Cl2, r.t. b Procedure B. (a) CH2Cl2, r.t.; (b) SOCl2 (4 eq.),
CH2Cl2, r.t.; (c) NaH (1 eq.), THF, r.t. Procedure C. NaOAc (6 eq.), CH2Cl2,
r.t.
Fig. 1 Molecular structure of Ir1 determined by X-ray crystallography
with thermal ellipsoids drawn at 50% probability. The hydrogen atoms
have been omitted for clarity. Selected bond lengths (Å): Ir–Cl 2.4084
(9), Ir–N 2.076(3), Ir–C 2.054(3), C–Ir–N 77.89(13).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
longer than that of Ir3 (Table S3†). For comparison, the half- equilibrium between the neutral and the cationic forms was
sandwich iridium complex carrying the 2-phenylpyridine reached for each complex. Their IC50 on the HeLa cancer cell
ligand showed a much lower rate of exchange with a half-life line were evaluated using the MTT cell viability assay. The
of 1125 min.21 This structure-based approach demonstrates obtained values are comprised between 3 and 6 µM (Table 1),
that the Ir–Cl bond strength is correlated to the electron-with- in good agreement with values previously on the same cell line
drawing character of the substituent on the (phox) ligand. for half-sandwich iridium complexes bearing anionic C^O che-
Furthermore, the steric effect induced by the substitutions R2,3 lating ligands.22 This rather narrow range depicts the minor
on the oxazoline ring is another parameter affecting the influence of the substituents carried by the chelating ligand
exchange rate since Ir7–8 also present an extended half-life. At on the antiproliferative activity as well as the stereochemistry
equilibrium, the ligand exchange is total for most complexes of the phox ligand. This result dramatically contrasts with pre-
of the series except for Ir2 and Ir7–8 for which the establish- vious reports on structurally related complexes for which the
ment of a (1 : 9) ratio was observed between the neutral and substitution pattern greatly influenced the IC50 values.23–25
cationic forms (Fig. 2).
The cell response to Ir2, Ir4, Ir8 and Ir10 was also investi-
To check whether the DMSO adducts could further undergo gated by a complementary label-free viability assay based on
aquation once diluted in the cell-culture medium, a [Ir1- impedance readout, using the xCELLigence Real Time Cell
DMSO]+ complex was synthesized and isolated as a nitrate salt Analysis (RTCA) system that allows the dynamic monitoring of
1
(see procedure in Fig. S2†). Its H NMR spectrum in CD3OD/ live cell adhesion and proliferation. All compounds induced a
D2O (2 : 1) showed that no water/DMSO exchange took place. dose-dependent and immediate loss of cell adhesion. As illus-
On the other hand, addition of a saturating amount of NaCl trated in Fig. S4† for Ir8, cells treated at concentrations close
(1.1 M) resulted in the coalescence of the two singlets around to the IC50 were able to re-adhere and then to proliferate
2.89 ppm attributed to two methyl groups, marking a very fast although at lower rates, as a consequence of either inhibition
exchange between DMSO and Cl− at the NMR timescale. The of cell division or cell death or both (see videomicroscopy
high chloride concentration also favoured the partial displace- ESI†). The resulting time-dependent IC50 was found to reach a
ment of DMSO by Cl− affording a mixture of cationic and minimum after 48 h of treatment and Ir8 was the most active
neutral complexes in 9 : 1 ratio at equilibrium. This experiment compound of the series with an IC50 of 3.6 µM (Table 1).
proved the stability of the DMSO adduct, ruling out the for-
mation of an aqua complex and supported the hypothesis that
the cationic DMSO complexes are the true administered mole-
cules in the biological assays.
The ability of Ir2 to undergo aquation was also examined by
1H NMR in MeOD-d4/D2O (3 : 2). Addition of D2O caused peak
broadening but the chemical shift of all the protons remained
the same. Further addition of NaCl restored the initial peaks
width (Fig. S3†). Therefore, no hydrolysis of the Ir–Cl bond of
Ir2 took place in these conditions.
Apoptosis assay
Morphological changes induced by Ir1–10 were suggestive of
regulated cell death. Apoptosis is characterised by hierarchical
activation of caspases by proteolytic processing, with the
effector caspase-3 being a critical executioner that degrades
many cellular proteins. We thus used caspase-3 cleavage as a
readout of apoptosis in treated samples. Fig. 3 shows the pres-
ence of cleaved caspase-3 in iridium-treated samples starting
12 h after drug administration, a fast kinetics compared to
conditions of etoposide-induced DNA damage suggesting a
different MoA. Overall, these results point to regulated cell
death induced by the complexes regardless of their substi-
tution pattern.
Cytotoxicity assays
The library of iridium complexes was examined for toxicity.
Prior to cell assays, stock solutions of Ir1–10 in DMSO were
incubated at room temperature for 24 h to ensure that the
Table 1 Antiproliferative activity of Ir1–10 towards the HeLa cell line
Compound
MTT IC50a (µM)
RTCA IC50b (µM)
Ir1
Ir2
Ir3
Ir4
Ir5
Ir6
Ir7
Ir8
Ir9
Ir10
4.17 0.11
4.05 0.47
3.56 0.68
3.73 0.99
4.47 0.29
5.81 0.61
3.46 0.29
3.06 0.38
3.35 0.24
3.36 0.57
NDc
5.06 0.23
ND
4.72 0.82
ND
ND
ND
3.59 0.15
ND
4.41 0.59
a Median inhibiting concentration after 96 h of treatment assessed by
the MTT cell viability assay. b Median inhibiting concentration after
48 h of treatment assessed by the xCELLigence RTCA system. c Not
determined.
Fig. 2 Comparison of the rates of solvolysis of three representative
complexes in DMSO-d6. Quantification was done by integration of the
characteristic 1H NMR peak of the Cp* ligand of the two forms (Fig. S1†).
Half-lives of Ir1–10 are presented in Table S3.†
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 3 Immunoblot showing accumulation of cleaved caspase-3 frag-
ments at 19 and 17 kDa for the indicated time (h) after addition of 10 µM
compound or vehicle (DMSO). Three representative complexes of the
series are presented. Etoposide, a clinically approved topoisomerase II
inhibitor that induces single and double strand DNA breaks, was used as
positive control and showed the presence of the cleaved form of
caspase-3 after 24 h. Equal loadings were assessed by Ponceau S stain-
ing (Fig. S5†).
Fig. 4 Amounts of remaining NADH and H2O2 produced in model
physiological environment. Data are shown as mean
SD of three
experiments performed at 37 °C in 5 mM phosphate buffer pH 7.4 con-
taining 1% MeOH in the presence of 10 mol% iridium complex.
Hydrogen peroxide was quantified using a multiplex measurement of tri-
plicates and standard for each experiment.
of SOD to prevent the formation of radical intermediates that
spontaneously oxidize the Amplex substrate, following rec-
ommendations from the literature.36 Although all of the com-
plexes were able to accept a hydride from coenzyme NADH,
only the oxazoline-substituted complexes (Ir2, Ir7–10) were
able to transfer their hydride to dissolved O2, producing
detectable amounts of hydrogen peroxide (Fig. 4). This trend
may be due to differences of hydricity, i.e. hydride donor
strength, among the complexes37 which may be linked to the
electron-donating property of the alkyl/aryl substituent on the
oxazoline ring of the chelating ligand or to steric hindrance
around the labile position. Reassuringly, the couples of
isomers (Ir7–8 and Ir9–10) displayed a similar potency to cata-
lyse the production of H2O2.
In vitro catalytic activity
The reaction of various half-sandwich iridium complexes with
NADH is widely described in the literature.26 Some structures
are able to form metal hydride complexes with concomitant
formation of NAD+ following a dissociative mechanism accord-
ing to DFT calculation.27 Hydride can then be transferred to
various acceptors such as flavin mononucleotide (FMN),28 pyr-
uvate to form lactate,29 proton to form dihydrogen,30 alde-
hydes to form alcohols31,32 or imines to yield amines.33 It was
also shown that dioxygen could act as hydride acceptor to yield
hydrogen peroxide which presence after 48 h incubation of
NADH and complexes in aerated solutions was qualitatively
detected using a peroxides test stick.12,34
The ability of Ir1–10 to catalyse NADH oxidation was firstly
investigated by monitoring the absorbance at 340 nm of a
100 µM solution of NADH in 5 mM phosphate buffer/MeOH
Intracellular H2O2 production
(99 : 1) in the presence of 10 mol% complex during 150 min. Among cellular reactive oxygen species (ROS), hydrogen per-
Reactions were performed with stock solutions of chlorido oxide is involved in various metabolic pathways and is sus-
complexes in MeOH as we found that the Ir-DMSO adduct was pected of having a signalling role in natural biological pro-
less active (Fig. S7†), which is consistent with the poor lability cesses.38 According to the literature, iridium half-sandwich
of the DMSO ligand (Fig. S2†). As NADH oxidation was pre- complexes including 2-phenylpyridine as chelating ligand were
viously shown to generate various side-products,35 results are shown to induce oxidative stress (i.e. total ROS and RNS levels
expressed in concentration of remaining substrate after increase) in cancer cells and this was suggested to result from
150 min (Fig. 4). All the complexes catalysed the oxidation of the catalytic production of H2O2.12 We thus investigated
NADH. Moreover, the moderately contrasted data show again a whether the ability of Ir1–10 to catalyse the production of
modest influence of the ligand substitution pattern on this H2O2 in vitro could be correlated to their cytotoxicity and their
process. Resulting TONs at 2.5 h were calculated for Ir1–10 capacity to elevate intracellular H2O2.
(Table S4†) and all values are higher than 1, confirming the
catalytic nature of this reaction.
Instead of employing classical but non-selective chemical
ROS probes such as H2DCFDA, we used a fluorescent protein-
Secondly, we examined whether oxidation of NADH was based H2O2-selective redox sensor, the HyPer ratiometric
accompanied by production of H2O2 in the presence of O2, the probe developed by Belousov et al.39 The spectral properties of
iridium complexes thus mimicking NADH oxidase activity (see the HyPer sensor change upon its reversible oxidation by H2O2
catalytic pathway in Fig. S8†). For this, the production of H2O2 and reduction by glutathione-dependent reductases, thus
was monitored for the first time using the HRP-linked Amplex allowing the indirect measurement of intracellular H2O2. A
Red® assay over 150 min. Since NADH autooxidation is known stable HeLa cell line that constitutively expresses HyPer40 was
to give false positive results in this assay, we added 40 U mL−1 exposed to Ir1–10, then reduced and oxidized forms of HyPer
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Fig. 5 (a) HyPer index derived from the median values of HeLa cells exposed to 10 µM Ir1–10 for 60 min (at least 5000 gated events). Statistical sig-
nificance was compared to the negative control with a Mann–Whitney non-parametric t-test (p < 0.0001) (b). Evolution of the HyPer ratios over
time after addition of a saturating amount of hydrogen peroxide. (c). Evolution of the HyPer ratios over time after addition of 20 µM Ir2, similar
results were obtained with lower concentrations and are reported in Fig. S6.† Cells treated for 90 min with DMSO were used as negative control.
were quantified by flow cytometry as the ratio of fluorescence
emission at 520 nm after excitation at 405 and 488 nm respect-
Conclusions
ively (Fig. S6†).
In summary, we have reported the synthesis and characteriz-
The median ratios M of the HyPer-expressing cells were ation of a set of half-sandwich iridium complexes comprising
used to calculate a HyPer-index41 H which was quantified in % a phenyloxazoline chelating ligand with a wide range of substi-
according to eqn (1):
tuents. We examined their outcome in the vehicle and the
stability of the resulting DMSO adducts in the presence of
water. These adducts are cationic, more lipophilic and thus
susceptible to display an enhanced penetration across cellular
membranes. The strong coordinating power of DMSO to
iridium suggests that the lifetimes of the cationic iridium com-
plexes may be long in the intracellular context. Altogether, it
ðM ꢀ MDMSO
Þ
H ¼
ð1Þ
ðMH O ꢀ MDMSO
Þ
2
2
where MDMSO and MH2O2 correspond to the median ratio appears necessary to properly determine the stability of the
values obtained for the negative and positive controls, respect- drug in the vehicle, a feature scarcely described in the litera-
ively. This index directly shows the significance of the observed ture, since it can dramatically influence its biological activity.
variations upon a 60 min treatment with 10 µM Ir1–10 This study also highlighted the minor influence of the struc-
(Fig. 5a). Dramatic differences between the complexes were tural modifications of the (phox) ligand on the overall toxicity
immediately observed with Ir2, Ir4, Ir7 and Ir10 displaying of the complexes, all of them inducing cell death by apoptosis.
high HyPer indexes while Ir1 and Ir5 were fully unable to In vitro, all complexes were able to accept a hydride from coen-
induce H2O2 production in HeLa cells. Since Ir2 complex dis- zyme NADH, but they were not equally able to transfer it to
played the highest HyPer index, the rate of this increase was dioxygen to produce H2O2. We quantified for the first time the
also monitored over a time course of 80 min. As a positive production of hydrogen peroxide elicited by half-sandwich
control, cells exposed to a saturating amount of hydrogen per- iridium complexes in two experimental settings, both in vitro
oxide (Fig. 5b, Fig. S6†) exhibited a fast recovery with the ratio- and in vivo. We found that the complexes carrying substituents
metric median parameter decreasing to a basal value within on the oxazoline ring (Ir2, Ir7–10) were the most efficient cata-
an hour, as a result of cellular detoxification processes. In con- lysts of H2O2 production in vitro. Strikingly, among these mole-
trast upon treatment with Ir2 in the same time lapse (Fig. 5c), cules, Ir2, Ir7 and Ir10 were also proven to generate high levels
H2O2 levels rose progressively, suggesting an uncompensated of H2O2 in living cells during the first hour of treatment.
production of H2O2 that is presumably catalytic. Moreover, the Moreover, the cells’ detoxification machinery was not able to
rate of H2O2 production was also concentration-dependent in compensate this production over time unlike with a simple
the 5–20 µM range (Fig. S6†).
hydrogen peroxide addition. While this mode of action pro-
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
vides a rationale for the fast cytotoxicity of the compounds
that induced intracellular accumulation of H2O2, it cannot be
excluded that other mechanisms be at stake, leading to the
fast activation of caspase-3 and cell death by apoptosis.
Indeed, neither Ir1 nor Ir5 were efficient in both analyses
despite a comparable antiproliferative activity. Recent studies
have shown that a structurally related iridium complex exhibi-
ted a mitochondrial tropism and induced distortions of the
inner membrane.42 A future challenge will be to decipher
whether this accumulation of H2O2 results from mitochondrial
disruption or is a direct consequence of intracellular transfer
hydrogenation catalysed by the iridium complexes.
6 J. Liu, X. Wu, J. A. Iggo and J. Xiao, Half-sandwich iridium
complexes—Synthesis and applications in catalysis, Coord.
Chem. Rev., 2008, 252, 782–809.
7 Z. Liu, L. Salassa, A. Habtemariam, A. M. Pizarro,
G. J. Clarkson and P. J. Sadler, Contrasting Reactivity
and
Cancer
Cell
Cytotoxicity
of
Isoelectronic
Organometallic Iridium(III) Complexes, Inorg. Chem., 2011,
50, 5777–5783.
8 R. M. Lord and P. C. McGowan, Organometallic Iridium
Arene Compounds: The Effects of C -Donor Ligands on
Anticancer Activity, Chem. Lett., 2019, 48, 916–924.
9 Y. Yang, X. Ge, L. Guo, T. Zhu, Z. Tian, H. Zhang, Q. Du,
H. Peng, W. Ma and Z. Liu, Zwitterionic and cationic half-
sandwich iridium(III) ruthenium(II) complexes bearing sul-
fonate groups: synthesis, characterization and their
different biological activities, Dalton Trans., 2019, 48, 3193–
3197.
Conflicts of interest
There are no conflicts to declare.
10 J. M. Zimbron, K. Passador, B. Gatin-Fraudet,
C.-M. Bachelet, D. Plażuk, L.-M. Chamoreau, C. Botuha,
S. Thorimbert and M. Salmain, Synthesis, Photophysical
Properties, and Living Cell Imaging of Theranostic Half-
Sandwich Iridium–4,4-Difluoro-4-bora-3 a, 4 a -diaza- s
-indacene (BODIPY) Dyads, Organometallics, 2017, 36,
3435–3442.
11 Z. Liu, I. Romero-Canelón, A. Habtemariam, G. J. Clarkson
and P. J. Sadler, Potent Half-Sandwich Iridium(III)
Anticancer Complexes Containing C ^ N-Chelated and
Pyridine Ligands, Organometallics, 2014, 33, 5324–5333.
12 Z. Liu, I. Romero-Canelón, B. Qamar, J. M. Hearn,
A. Habtemariam, N. P. E. Barry, A. M. Pizarro,
G. J. Clarkson and P. J. Sadler, The Potent Oxidant
Anticancer Activity of Organoiridium Catalysts, Angew.
Chem., Int. Ed., 2014, 53, 3941–3946.
Acknowledgements
This research was supported by the IPV doctoral program of
Sorbonne Université. We wish to acknowledge basic support
from CNRS, INSERM and Sorbonne Université, to UMR_8232
and UMR_S938 and competitive funding from LabEx MiChem
part of French state funds managed by the ANR within
the Investissements d’avenir programme under reference
ANR-11-IDEX-004-02 and from INCA-DGOS-Inserm_12560
SiRIC CURAMUS financially supported by the French National
Cancer Institute, the French Ministry of Solidarity and Health
and Inserm. We thank Dr Sophie Vriz for the gift of HeLa-
HyPer stable cell line. We are indebted to Blaise Gatin-Fraudet
and Kévin Passador for supporting technical assistance and
data acquisition and to Romain Morichon for videomicroscopy
experiments.
13 J. M. Hearn, G. M. Hughes, I. Romero-Canelón,
A. F. Munro, B. Rubio-Ruiz, Z. Liu, N. O. Carragher and
P. J. Sadler, Pharmaco-genomic investigations of organo-
iridium anticancer complexes reveal novel mechanism of
action, Metallomics, 2018, 10, 93–107.
References
1 L. Zeng, P. Gupta, Y. Chen, E. Wang, L. Ji, H. Chao and 14 D. Trachootham, J. Alexandre and P. Huang, Targeting
Z.-S. Chen, The development of anticancer ruthenium(ii)
complexes: from single molecule compounds to nano-
materials, Chem. Soc. Rev., 2017, 46, 5771–5804.
cancer cells by ROS-mediated mechanisms: a radical thera-
peutic approach?, Nat. Rev. Drug Discovery, 2009, 8, 579–
591.
2 C. Caporale and M. Massi, Cyclometalated iridium(III) com- 15 C. Gorrini, I. S. Harris and T. W. Mak, Modulation of oxi-
plexes for life science, Coord. Chem. Rev., 2018, 363, 71–91.
3 P. Zhang, H. Huang, S. Banerjee, G. J. Clarkson, C. Ge,
dative stress as an anticancer strategy, Nat. Rev. Drug
Discovery, 2013, 12, 931–947.
C. Imberti and P. J. Sadler, Nucleus-Targeted 16 J. H. Barnard, C. Wang, N. G. Berry and J. Xiao, Long-range
Organoiridium-Albumin Conjugate for Photodynamic
Cancer Therapy, Angew. Chem., Int. Ed., 2019, 58, 2350–
2354.
metal–ligand bifunctional catalysis: cyclometallated
iridium catalysts for the mild and rapid dehydrogenation
of formic acid, Chem. Sci., 2013, 4, 1234.
4 X. Li, W. Ouyang, J. Nie, S. Ji, Q. Chen and Y. Huo, Recent 17 D. L. Davies, O. Al-Duaij, J. Fawcett, M. Giardiello,
Development
on
Cp*Ir(III)–Catalyzed
C−H
Bond
S. T. Hilton and D. R. Russell, Room-temperature cyclome-
Functionalization, ChemCatChem, 2020, 12, 2358–2384.
tallation of amines, imines and oxazolines with [MCl
2
5 C. Michon, K. MacIntyre, Y. Corre and F. Agbossou-
Cp*] 2 (M = Rh, Ir) and [RuCl 2 (p-cymene)] 2, Dalton Trans.,
2003, 4132–4138.
Niedercorn,
Pentamethylcyclopentadienyl
Iridium(III)
Metallacycles Applied to Homogeneous Catalysis for Fine 18 Y. Boutadla, D. L. Davies, R. C. Jones and K. Singh, The
Chemical Synthesis, ChemCatChem, 2016, 8, 1755–1762.
Scope of Ambiphilic Acetate-Assisted Cyclometallation with
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Half-Sandwich Complexes of Iridium, Rhodium and
Ruthenium, Chem. – Eur. J., 2011, 17, 3438–3448.
Accompanied by Generation and Consumption of
Hydrogen with
a Water-Soluble Iridium Complex at
19 M. D. Hall, K. A. Telma, K.-E. Chang, T. D. Lee,
J. P. Madigan, J. R. Lloyd, I. S. Goldlust, J. D. Hoeschele
Ambient Pressure and Temperature, J. Am. Chem. Soc.,
2012, 134, 367–374.
and M. M. Gottesman, Say No to DMSO: Dimethylsulfoxide 31 A. H. Ngo, M. Ibañez and L. H. Do, Catalytic
Inactivates Cisplatin, Carboplatin, and Other Platinum
Complexes, Cancer Res., 2014, 74, 3913–3922.
20 M. Patra, T. Joshi, V. Pierroz, K. Ingram, M. Kaiser,
Hydrogenation of Cytotoxic Aldehydes Using Nicotinamide
Adenine Dinucleotide (NADH) in Cell Growth Media, ACS
Catal., 2016, 6, 2637–2641.
S. Ferrari, B. Spingler, J. Keiser and G. Gasser, 32 S. Bose, A. H. Ngo and L. H. Do, Intracellular Transfer
DMSO-Mediated Ligand Dissociation: Renaissance for
Hydrogenation Mediated by Unprotected Organoiridium
6
Biological Activity of N -Heterocyclic-[Ru(η -arene)Cl
Drug Candidates, Chem. – Eur. J., 2013, 19, 14768–14772.
]
Catalysts, J. Am. Chem. Soc., 2017, 139, 8792–8795.
2
5
33 M. Soetens, F. Drouet and O. Riant, (η
21 A. C. Carrasco, V. Rodríguez-Fanjul, A. Habtemariam and
A. M. Pizarro, Structurally Strained Half-Sandwich Iridium
(III) Complexes As Highly Potent Anticancer Agents, J. Med.
Chem., 2020, 63, 4005–4021.
-Pentamethylcyclopentadienyl)iridium Complex Catalyzed
Imine Reductions Utilizing the Biomimetic 1,4-NAD(P)H
Cofactor and N -Benzyl-1,4-dihydronicotinamide as the
Hydride-Transfer Agent, ChemCatChem, 2017, 9, 929–933.
22 Q. Du, L. Guo, M. Tian, X. Ge, Y. Yang, X. Jian, Z. Xu, Z. Tian 34 Z. Liu and P. J. Sadler, Formation of glutathione sulfenate
and Z. Liu, Potent Half-Sandwich Iridium(III) and
Ruthenium(II) Anticancer Complexes Containing
and sulfinate complexes by an organoiridium(III) anti-
cancer complex, Inorg. Chem. Front., 2014, 1, 668–672.
35 A. Bucci, S. Dunn, G. Bellachioma, G. Menendez
Rodriguez, C. Zuccaccia, C. Nervi and A. Macchioni, A
Single Organoiridium Complex Generating Highly Active
Catalysts for both Water Oxidation and NAD+/NADH
Transformations, ACS Catal., 2017, 7, 7788–7796.
a
P^O-Chelated Ligand, Organometallics, 2018, 37, 2880–2889.
23 J. Yellol, S. A. Pérez, A. Buceta, G. Yellol, A. Donaire,
P. Szumlas, P. J. Bednarski, G. Makhloufi, C. Janiak,
A. Espinosa and J. Ruiz, Novel C,N-Cyclometalated
Benzimidazole Ruthenium(II) and Iridium(III) Complexes as
Antitumor and Antiangiogenic Agents: A Structure–Activity 36 T. V. Votyakova and I. J. Reynolds, Detection of hydrogen
Relationship Study, J. Med. Chem., 2015, 58, 7310–7327.
24 Y. Yang, L. Guo, Z. Tian, Y. Gong, H. Zheng, S. Zhang,
Z. Xu, X. Ge and Z. Liu, Novel and Versatile Imine-
peroxide with Amplex Red: interference by NADH and
reduced glutathione auto-oxidation, Arch. Biochem.
Biophys., 2004, 431, 138–144.
N-Heterocyclic
Carbene
Half-Sandwich
Iridium(III) 37 A. H. Ngo and L. H. Do, Structure–activity relationship
Complexes as Lysosome-Targeted Anticancer Agents, Inorg.
Chem., 2018, 57, 11087–11098.
25 A. J. Millett, A. Habtemariam, I. Romero-Canelón,
study of half-sandwich metal complexes in aqueous trans-
fer hydrogenation catalysis, Inorg. Chem. Front., 2020, 7,
583–591.
G. J. Clarkson and P. J. Sadler, Contrasting Anticancer 38 C. Rampon, M. Volovitch, A. Joliot and S. Vriz, Hydrogen
Activity of Half-Sandwich Iridium(III) Complexes Bearing
Functionally Diverse 2-Phenylpyridine Ligands,
Organometallics, 2015, 34, 2683–2694.
26 Y. Okamoto and T. R. Ward, Transfer Hydrogenation
Peroxide and Redox Regulation of Developments,
Antioxidants, 2018, 7, 159.
39 D. S. Bilan and V. V. Belousov, HyPer Family Probes: State
of the Art, Antioxid. Redox Signal., 2016, 24, 731–751.
Catalyzed by Organometallic Complexes Using NADH as a 40 C. Gauron, F. Meda, E. Dupont, S. Albadri, N. Quenech’Du,
Reductant in a Biochemical Context, Biochemistry, 2017, 56,
5223–5224.
E. Ipendey, M. Volovitch, F. Del Bene, A. Joliot, C. Rampon
and S. Vriz, Hydrogen peroxide (H 2 O 2) controls axon
pathfinding during zebrafish development, Dev. Biol., 2016,
414, 133–141.
27 I. Ritacco, N. Russo and E. Sicilia, DFT Investigation of the
Mechanism of Action of Organoiridium(III) Complexes As
Anticancer Agents, Inorg. Chem., 2015, 54, 10801–10810.
28 S. Shibata, T. Suenobu and S. Fukuzumi, Direct Synthesis of
Hydrogen Peroxide from Hydrogen and Oxygen by Using a
Water-Soluble Iridium Complex and Flavin Mononucleotide,
Angew. Chem., Int. Ed., 2013, 52, 12327–12331.
41 O. G. Lyublinskaya, Ju. S. Ivanova, N. A. Pugovkina,
I. V. Kozhukharova, Z. V. Kovaleva, A. N. Shatrova,
N. D. Aksenov, V. V. Zenin, Yu. A. Kaulin, I. A. Gamaley and
N. N. Nikolsky, Redox environment in stem and differen-
tiated cells: A quantitative approach, Redox Biol., 2017, 12,
758–769.
29 S. Betanzos-Lara, Z. Liu, A. Habtemariam, A. M. Pizarro,
B. Qamar and P. J. Sadler, Organometallic Ruthenium and 42 J. J. Conesa, A. C. Carrasco, V. Rodríguez-Fanjul, Y. Yang,
Iridium Transfer-Hydrogenation Catalysts Using Coenzyme
NADH as a Cofactor, Angew. Chem., Int. Ed., 2012, 51, 3897–
3900.
J. L. Carrascosa, P. Cloetens, E. Pereiro and A. M. Pizarro,
Unambiguous Intracellular Localization and Quantification
of a Potent Iridium Anticancer Compound by Correlative
3D Cryo X-Ray Imaging, Angew. Chem., 2020, 59, 1270–
1278.
30 Y. Maenaka, T. Suenobu and S. Fukuzumi, Efficient
Catalytic Interconversion between NADH and NAD+
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.