Platinum(iv)-azido monocarboxylato complexes are photocytotoxic under irradiation with visible light

Article abstract of DOI:10.1039/d1dt01730f

Complexestrans,trans,trans-[Pt(N₃)₂(OH)(OCOR)(py)₂] where py = pyridine and where OCOR = succinate (1); 4-oxo-4-propoxybutanoate (2) andN-methylisatoate (3) have been synthesized by derivation oftrans,trans,trans-[Pt(OH)₂(N₃)₂(py)₂] (4) and characterised by NMR and EPR spectroscopy, ESI-MS and X-ray crystallography. Irradiation of1-3with green (517 nm) light initiated photoreduction to Pt(ii) and release of the axial ligands at a 3-fold faster rate than for4. TD-DFT calculations showed dissociative transitions at longer wavelengths for1compared to4. Complexes1and2showed greater photocytotoxicity than4when irradiated with 420 nm light (A2780 cell line IC₅₀values: 2.7 and 3.7 μM) and complex2was particularly active towards the cisplatin-resistant cell line A2780cis (IC₅₀3.7 μM). Unlike4, complexes1-3were phototoxic under green light irradiation (517 nm), with minimal toxicity in the dark. A pK_a(H₂O) of 5.13 for the free carboxylate group was determined for1, corresponding to an overall negative charge during biological experiments, which crucially, did not appear to impede cellular accumulation and photocytotoxicity.

Full text of DOI:10.1039/d1dt01730f

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Platinum(IV)-azido monocarboxylato complexes
are photocytotoxic under irradiation with
visible light†
Cite this: Dalton Trans., 2021, 50,
10593
Evyenia Shaili,a Marίa J. Romero,a,b Luca Salassa, a,c,d,e Julie A. Woods,‡f
a
Jennifer S. Butler,a Isolda Romero-Canelón, a,g Guy Clarkson,
a,h
Abraha Habtemariam,a Peter J. Sadler *a and Nicola J. Farrer
*
Complexes trans,trans,trans-[Pt(N3)2(OH)(OCOR)(py)2] where py = pyridine and where OCOR = succinate
(1); 4-oxo-4-propoxybutanoate (2) and N-methylisatoate (3) have been synthesized by derivation of trans,
trans,trans-[Pt(OH)2(N3)2(py)2] (4) and characterised by NMR and EPR spectroscopy, ESI-MS and X-ray
crystallography. Irradiation of 1–3 with green (517 nm) light initiated photoreduction to Pt(II) and release
of the axial ligands at a 3-fold faster rate than for 4. TD-DFT calculations showed dissociative transitions
at longer wavelengths for 1 compared to 4. Complexes 1 and 2 showed greater photocytotoxicity than 4
when irradiated with 420 nm light (A2780 cell line IC50 values: 2.7 and 3.7 μM) and complex 2 was par-
ticularly active towards the cisplatin-resistant cell line A2780cis (IC50 3.7 μM). Unlike 4, complexes 1–3
were phototoxic under green light irradiation (517 nm), with minimal toxicity in the dark. A pKa(H2O) of
5.13 for the free carboxylate group was determined for 1, corresponding to an overall negative charge
during biological experiments, which crucially, did not appear to impede cellular accumulation and
photocytotoxicity.
Received 27th May 2021,
Accepted 11th July 2021
DOI: 10.1039/d1dt01730f
rsc.li/dalton
Introduction
Approximately 40% of the cancer patients treated with chemo-
therapy receive a platinum(II)-based medicine such as cispla-
tin, carboplatin or oxaliplatin.1 Despite the wide-spread use of
aDepartment of Chemistry, University of Warwick, Gibbet Hill Road, Coventry,
CV4 7AL, UK. E-mail: p.j.sadler@warwick.ac.uk
platinum-based drugs2 – particularly in combination therapies
– disadvantages exist, including the development of resistance
and serious side-effects of treatment.3 Octahedral, Pt(IV) pro-
drugs with a d6 electronic configuration typically demonstrate
greater kinetic inertness than Pt(II) complexes, as well as
offering additional ligand sites for derivation.4
bDepartamento de Didácticas Aplicadas, Facultade de Formación do Profesorado,
Universidade de Santiago de Compostela, 27002 Lugo, Spain
cDonostia International Physics Center, Paseo Manuel de Lardizabal 4, Donostia,
20018, Spain
dPolimero eta Material Aurreratuak: Fisika, Kimika eta Teknologia, Kimika
Fakultatea, Euskal Herriko Unibertsitatea UPV/EHU, Paseo Manuel de Lardizabal 3,
Donostia, 20018, Spain
eIkerbasque, Basque Foundation for Science, Bilbao, 48011, Spain
fPhotobiology Unit, Department of Dermatology and Photobiology, Ninewells
Hospital, Dundee, DD1 9SY, UK
gSchool of Pharmacy, Institute of Clinical Sciences, College of Medical and Dental
Sciences, Sir Robert Aitken Institute for Medical Research, University of Birmingham,
Birmingham, B15 2TT, UK
hChemistry Research Laboratory, University of Oxford, 12 Mansfield Road, Oxford,
OX1 3TA, UK. E-mail: Nicola.Farrer@chem.ox.ac.uk; Tel: +44 (0)1865 285131
†Electronic supplementary information (ESI) available: Materials and methods,
synthesis and characterisation data, X-ray crystallographic tables, biological
data. CCDC 1998067, 1998068 and 1998069. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/d1dt01730f
‡This paper describes experimental work performed by Dr Julie A Woods but
she is unaware that it has been submitted for publication. Dr Julie A Woods
therefore, does not take any responsibility for the submission.
Pt(IV) prodrugs are classically regarded to exert their cyto-
toxic effect following reduction to Pt(II) species in vivo,
although Pt(IV) complexes may also form adducts with bio-
molecules.5 Photoactivated chemotherapy (PACT) provides
both spatial and temporal control over Pt(IV) prodrug
reduction. Photoreduction of Pt(IV) prodrugs of carboplatin,6
oxaliplatin7 and cisplatin8 – either by direct irradiation of the
Pt(IV) complex, or by employing separate photosensitisers –
have been reported. The ligands coordinated to the Pt(IV)
prodrug influence the solubility, lipophilicity and ease of
reduction of the complex, whilst the ligands themselves have
the potential to be biologically active, through mechanisms
which are distinct from the resultant platinum species, follow-
ing dissociation. The reduction of Pt(IV) complexes is greatly
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 10593–10607 | 10593

View Article Online
Paper
Dalton Transactions
affected by the coordinated ligands; for a series of complexes
with chlorido, hydroxido and carboxylato anionic axial ligands
[Pt(en)Cl2(X)2] (where X = axial ligand), cathodic reduction
potentials varied by more than 650 mV, with the rates of
−
reduction following the order X = Cl− > OC(O)CH3 > OH−.5 It
is notable that for several Pt(IV) complexes – including
Satraplatin (ctc-[Pt(NH3)(CHA)(OAc)2Cl2] where CHA = cyclo-
hexylamine) – equatorial anionic ligand hydrolysis can occur
more readily than axial anionic ligand hydrolysis.9 The most
important biological reductants, the mechanism(s) and the
rate of Pt(IV) reduction in a biological environment for a given
complex cannot readily be predicted from the reduction poten-
tial.10 A range of Pt(IV) complexes undergo photoreduction,
typically requiring short wavelengths of light i.e. UVA.11 The
coordination of azido (N3) ligands to Pt(IV) results in highly
photosensitive complexes, which can be kinetically inert in the
dark and – depending on judicious ligand choice – can be
photoactivated with longer (visible) wavelengths of light.12
Penetration of light into tissue is wavelength-dependent, and
extending the wavelength of light which can initiate photoreac-
tion of Pt(IV) complexes – whilst retaining stability in the dark
– is therefore a key goal. Replacing the axial OH ligands of Pt
(IV) diazido complexes with acetato (OAc) ligands has been
demonstrated by TD-DFT to lower the energy of the LUMO and
LUMO+1, reducing the energy gap between the HOMO and
LUMO.13 This has the effect of extending the strongest absorp-
tion band to longer wavelengths. Since this modification also
has the potential to lead to more facile spontaneous reduction
in the dark, evaluation of stability of the complex in the
absence of irradiation is important. Modification of the axial
ligands of photoactivatable Pt(IV) diazido complexes to incor-
porate biotin,14 guanidinoneomycin,15 coumarins16 and
naphthalimides,17 as well as an RGD peptide to target integ-
rin18 have been reported. However, the synthesis and extensive
characterisation of 1 itself has not been reported previously,
and we report it here for the first time along with two other
complexes. Hydroxido groups are common Pt(IV) axial ligands
because oxidation of Pt(II) to Pt(IV) is readily accomplished
through the use of H2O2 and also due to the favourable solubi-
lity and stability properties imparted by inclusion of the hydro-
Fig. 1 Platinum(IV) azido complexes reported here (axial group): 1 (suc-
cinate),
2 (4-oxo-4-propoxybutanoate), 3 (N-methylisatoate), and 4
(hydroxide, previously reported).29
tanoate (2) and trans,trans,trans-[Pt(N3)2(OH)(N-MI)(py)2]
where N-MI = N-N-methylisatoate (3), in comparison with
trans,trans,trans-[Pt(N3)2(OH)2(py)2] (4), which we have
reported previously29 (Fig. 1). The N-MI ligand was chosen
since it introduces an additional absorption band (ca. 350 nm)
and increased absorption towards longer wavelengths. We
therefore anticipated that complex (3) might display increased
photosensitivity on irradiation with visible light.
Results
Synthesis and characterisation
Precursor compounds N-methylisatoic acid (N-MIA) (5), trans-
[PtCl2(py)2] (6), trans-[Pt(N3)2(py)2] (7) and complex 4 were syn-
thesized as reported previously; characterisation was consist-
ent with previous reports.29 Complexes 1–3 were synthesised
1
from 4 and characterised by H, 13C, 195Pt-NMR spectroscopy,
ESI-MS and UV-vis spectroscopy (ESI) and single crystal X-ray
diffraction.
X-ray diffraction
xido ligand. Derivatisation of one hydroxido ligand of a dihy- Single crystals suitable for X-ray diffraction were obtained for
droxido Pt(IV) complex to a carboxylate can be achieved complexes trans,trans,trans-[Pt(N3)2(OH)(succ)(py)2] 1, trans,
through a number of methods19 including reaction with trans,trans-[Pt(N3)2(OH)(succ-Pr)(py)2] 2 and trans,trans,trans-
anhydrides,20–23 acid chlorides,24 and reaction with carboxylic [Pt(N3)2(OH)(N-MI)(py)2] 3 as described in the Experimental
acids in acetic acid solvent.25 Terminal carboxylic acids which section. ORTEP diagrams for the molecular structures are
have been produced through reaction with anhydrides can be shown in Fig. 2 and the crystallographic data, including
used to introduce targeting peptides.26 Other functional selected bond distances and angles are summarized in
groups, such as carbamates can also be axially introduced.27,28 Table S2.†
The aim of the present work was to explore the effect of
axial ligand modification on the photochemistry and photo- try around the Pt(IV) ion and differ only in the nature of one of
biology of platinum(IV) azido complexes.12
the oxygen donor ligands coordinated in an axial position. The
Compounds 1, 2 and 3 show a distorted octahedral geome-
We report the synthesis, characterisation, photochemical equatorial plane is defined by four nitrogen atoms from two
and photobiological studies of the mono-carboxylato com- pyridine molecules trans to each other (N1 and N2) and two
pounds: trans,trans,trans-[Pt(N3)2(OH)(succ)(py)2] (1) where azido anions trans (N3 and N6), with an oxygen atom (O2)
succ
= succinate and py = pyridine; trans,trans,trans-[Pt from a hydroxido ligand located in one axial position. The
(N3)2(OH)(succ-Pr)(py)2] where succ-Pr = 4-oxo-4-propoxybu- sixth position is occupied by an oxygen atom (O1) from a
10594 | Dalton Trans., 2021, 50, 10593–10607
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
and 13C NMR spectral resonances were assigned in accordance
with previously reported complexes (labelling shown in
Fig. S13†).
ESI-MS. Pt(IV) diazido complexes commonly form a series of
sodiated adducts when analysed by ESI-MS.31 Complex 1 gave
rise to several species, including [M − N3]+ at 529.0793 m/z, [M
+ Na]+ at 594.0783 m/z and [2M + Na]+ at 1165.1673 m/z.
Complex 2 was detected as the [M + Na]+ adduct at 636.2 m/z.
The formation of [2M + Na]+ adducts was less pronounced for
complex 3 than for complex 1; complex 3 was detected predo-
minantly as [M − N3]+ at 562.1210 m/z, with other adducts and
species including [M + H]+ at 605.1338 m/z (Fig. S1†).
pKa of complex 1
The pKa* of complex 1 (1.1 mM in D2O) was determined by
1
plotting the change in the H NMR chemical shift of the pyri-
dine protons against pH*, with dioxane as an internal refer-
ence. The pH* was measured at ambient temperature, directly
in the NMR tube. The pH* was adjusted to the desired value
using DClO4. 1H NMR spectra were recorded at 25 °C. The data
were fitted to the Henderson–Hasselbalch equation (r2 = 0.99),
and the pKa* of the free carboxyl group was determined to be
5.07 (Fig. S5†) in D2O, which was converted to a final value of
pKa(H2O) 5.13.32
Fig. 2 ORTEP diagrams of the complexes: (A) trans,trans,trans-[Pt
(N3)2(OH)(succ)(py)2] 1; (B) trans,trans,trans-[Pt(N3)2(OH)(succ-Pr)(py)2]
2; and (C) trans,trans,trans-[Pt(N3)2(OH)(N-MI)(py)2] 3. The ellipsoids are
set to 50% probability level. Atom labels do not represent the numbering
in the cif files (CCDC1998067(1); CCDC1998069 (2) and CCDC1998068
(3)†) but are used to facilitate comparison between the complexes. All
hydrogen atoms have been omitted for clarity, except for the axial
hydroxido hydrogen atoms.
UV-vis spectroscopy
UV-vis spectra of complexes 1–3 (in H2O for 1 and 2; in 5%
DMSO/95% H2O for 3 due to solubility constraints) were
measured, with extinction coefficients determined by ICP-MS
monoanionic ligand: a succinate anion in 1, a 4-oxo-4-propoxy- analysis of platinum concentration (Fig. S6†).
butanoate anion in 2 and a N-methylisatoate anion in the case
DFT and TD-DFT calculations
of 3. The distortion from octahedral geometry is demonstrated
by the O2–Pt–O1 bond angle being 174.21(7) Å in 1, 175.52(9) DFT and TD-DFT calculations were performed on complex 1 to
Å in 2 and 176.50(11) Å in 3, which are all considerably less characterise the electronic transitions giving rise to the UV-vis
linear than the corresponding angle in complex 4 at 179.58(11) absorption spectrum and to rationalize the photoactivity of 1,
Å. The Pt–O1 (carboxylato) bond lengths for the monocarboxy- particularly in comparison to complex 4.29 Ground-state geo-
lato, monohydroxido complexes 1–3 are all elongated in com- metry and lowest-lying triplet geometry optimizations were per-
parison to 4 and similar to values for previously reported dicar- formed in the deprotonated state, since the pKa determined
boxylato complexes trans,trans,trans-[Pt(N3)2(OAc)2(py)2] (2.007 (above) for 1 (5.13, H2O) suggests it would be largely deproto-
(5) Å)30 and trans,trans,trans-[Pt(N3)2(CF3COO)2(py)2] (2.0093 nated at physiological pH. The singlet excited states, as well as
(15) Å).15 In complexes 1–3 the Pt–O1 (carboxyl) bond is longer the orbital contributions (%) in each case are summarised in
than the Pt–O2 (hydroxido) bond. However, a difference Table S3,† along with molecular orbital representations and
between both axial Pt–OH bonds was also observed in the crys- Electron Difference Density Maps (EDDMS) of singlet excited-
tallographic structure of 4 (Pt–O1 1.990(3) Å and Pt–O2 2.027 state transitions in H2O (Table S4†) and molecular orbitals for
(3) Å). In 4, one of the hydroxido ligands is involved in two the ground state (Table S5†). The UV-vis spectra were simu-
hydrogen bonds (as a donor and acceptor) with hydroxido and lated by calculation of 32 singlet states using water as a solvent
azido groups of a neighbouring molecule, which correlates (cpcm)33,34 and the TD-DFT method (Table 1) and are overlaid
with the lengthened Pt–OH bond, whereas the second hydro- with the experimental data in Fig. 3.
xido ligand acts only as a donor. The intermolecular inter-
actions in complexes 1, 2 and 3 are shown in the ESI (Fig. S2– complex extend up to 469 nm, whereas in 4 they reach to
S4†).
The longest wavelength singlet–singlet transitions in this
414 nm.29 The transitions S1, S2 and S3 of 1 show that there is
NMR spectroscopy. 195Pt NMR spectroscopic resonances a transfer of electron density from the succinate to the plati-
were observed for 1 (D2O), 2 (D2O) and 3 (DMSO-d6) at num centre (1LMCT) and also contain 1IL (inter-ligand tran-
1059 ppm, 1065 ppm and 990 ppm, respectively, in the sitions) character. Such transitions (S1, S2, S3) as well as some
expected low field Pt(IV) region. These are comparable with the of the higher energy involve the LUMO, which is antibonding
resonance reported for compound 4 (D2O) at 942 ppm.29 1
H
in character with regard to the four equatorial ligands but
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 10593–10607 | 10595

View Article Online
Paper
Dalton Transactions
Table 1 Selected experimental (X-ray crystallography) and calculated
bond lengths for complex 1 in both the ground state and the lowest-
lying triplet state
Solution reactivity
Complexes 1–3 showed no signs of decomposition in aqueous
solution in the absence of light for at least 1 h at ambient
temperature.
Bond length
X-ray
structure (Å)
Ground-state
(deprotonated)
Lowest
lying triplet
Irradiation studies
Bond
UV-Vis spectroscopy. Solutions of compounds 1–4 (ca.
50 μM, H2O) were irradiated with green light (517 nm LED,
29 mW cm−2; see Fig. S7† for spectral output of light sources)
at ambient temperature, and the photoactivity was monitored
by UV-vis spectroscopy (Fig. S8A–D†). The photoactivity of
complex 4 has been reported previously,29 and it was therefore
included here for direct comparison. Consistent with previous
reports, a decrease in the intensity of the N3 → Pt ligand-to-
metal charge-transfer band (the LMCT band in the absorption
spectrum of complexes at ca. 300 nm) was observed during
irradiation of complexes 1–4, due to dissociation of azido
ligand(s). The maximum absorbance for each graph was nor-
malized, such that the first spectrum (t = 0 h, prior to
irradiation) has a normalized absorbance of 100%. Data points
corresponding to the initial data loss of the LMCT band (up to
ca. 38%) were fitted to an exponential decay and the half-life of
each complex was determined (Fig. 4, 4).34 The rates of loss of
the LMCT band upon irradiation with 517 nm light ordered
from fastest to slowest are (t12, min): complex 2 (169) > complex
3 (214) > complex 1 (220) ≫ complex 4 (634).
Pt–N3(azide)
Pt–N6(azide)
Pt–O2 (OH)
Pt–O1(CO)
Pt–N1(py)
Pt–N2(py)
N6–N7
N3–N4
N7–N8
N4–N5
2.049(2)
2.045(2)
1.9852(15)
2.0196(16)
2.040(2)
2.032(2)
1.206(3)
1.202(3)
1.137(3)
1.126(4)
2.088
2.069
2.015
2.048
2.068
2.027
1.213
1.211
1.149
1.148
2.359
2.508
2.015
2.065
2.053
2.040
1.200
1.189
1.164
1.170
Complex 1 was also irradiated with longer wavelength green
light (550 nm, filtered below 530 nm with GG530 filter, 4 nm
slits, 2.5 mW cm−2) to investigate the wavelength limit of photo-
activation of the complex (Fig. S8E†). No significant change in
the UV-vis spectrum of 1 was observed under these conditions.
NMR spectroscopic irradiation studies. The photoactivity of
complex 1 in comparison to 4 was investigated in the presence
of the nucleotide guanosine 5′-monophosphate (5′-GMP) by 1H
Fig. 3 Experimental (black, H2O) and theoretical (blue) UV-vis spectra
of complex 1 in the deprotonated state. Vertical (red) bars represent the
oscillator strengths of the excited states.
NMR spectroscopy. Irradiation (10 min, 463 nm, 64 mW cm−2
)
mainly the azides. Interestingly the LUMO+1, to which electron
density is promoted in the higher wavelength transitions (e.g.
S8, S14) is orthogonal with respect to the LUMO, making it
strongly antibonding mostly towards the pyridines and also
the hydroxido and the succinate ligand.
of 1 resulted in 23% coordinated 5′-GMP in the species trans-
This suggests that occupation of the LUMO or LUMO+1
could potentially lead to dissociation of two out of the six
ligands. These two ligands would be in a trans position to each
other. Singlet–singlet transitions in the UV region (e.g. S14,
S19) occur from lower energy HOMOs and can be assigned as
mixed 1LMCT/1IL. In general, all transitions for complex 1
show some d–d character, as metal orbitals always contribute
to the frontier molecular orbitals.
As is commonly observed in DFT calculations,44–46 we
found that the bond lengths of the ground state of complex 1
are overestimated compared to X-ray, the largest discrepancy
being the two Pt–azido bonds, with differences of 0.035 and
0.028 Å (Table 1). In the lowest lying excited-state geometry,
Pt–azido bonds are significantly elongated with respect of the
ground state (up to 0.44 Å), while variations much less pro-
nounced for the other ligands (Table 1).
Fig. 4 Plot of the decrease in intensity of the N3 → Pt LMCT band of
compounds 1–4 over time, when irradiated with green light (λirr
517 nm).
10596 | Dalton Trans., 2021, 50, 10593–10607
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
[Pt(N3)(5′-GMP)(py)2], compared to only 9% of 5′-GMP co- light. Pyridine release was also observed from 4 under these
ordinated for 1 (Fig. S9†). Under continued irradiation of 1 conditions (Fig. S11,† bottom), which was not observed when
(463 nm), followed by a final irradiation with shorter wave- 4 was irradiated in D2O or H2O solutions with no MeOH-d4
length (420 nm) light, the formation of two main photo- present.29 The release of the axial N-methylisatoate (N-MIA)
products became evident from the 1H NMR spectra (Fig. 5). ligand from complex 3 in aqueous solution was also monitored
One of these can be identified as the mono-azido 5′-GMP by fluorescence emission spectroscopy (Fig. S12†), upon exci-
adduct, trans-[Pt(py)2(N3)(5′-GMP)], with assignment con- tation of complex 3 at 320 nm, which demonstrated an
firmed by LC-MS and 195Pt NMR spectroscopy.29
increase in fluorescence consistent with ligand release.
The photoactivity of complex 1 was also investigated
LC-MS data. LC-MS was employed to characterise the photo-
through monitoring the release of the succinate ligand in the products formed following irradiation of complex 1 in the
presence and absence of 5′-GMP. Photorelease of the succinate presence of 5′-GMP (Fig. 6). For comparison, the mass spec-
ligand from 1 occurred ca. 3× more slowly in the presence of trum of complex 1 in the dark is given in Fig. S1.† Complex 1
5′-GMP than in its absence (Fig. S10†). A precipitate was also was irradiated in the presence of 2 mol. eq. of 5-GMP in either
observed in the absence – but not in the presence – of 5′-GMP, H2O (Fig. 6, red trace) or under similar conditions but in PBS
consistent with previous reports.29
(0.140 M [Cl−], Fig. 6, blue trace). Comparable adducts were
Due to the limited aqueous solubility of 3, irradiation seen under both conditions, including [PtII(N3)(py)2(5′-GMP)]+
studies in the presence of 5′-GMP were carried out in 75% (peak 6) and [PtI2I (µ-5′-GMP-H)(N3)2(py)4]+ (peak 10) species,
MeOH-d4/25% D2O solution. In contrast to complex 1, very although irradiation in H2O resulted in detection of greater
little 5′-GMP binding was observed for 3 following irradiation, amounts of [PtII(5′-GMP)(ACN)(py)2]2+ and irradiation in the
and pyridine release was clearly observed (Fig. S11,† top).
To establish whether the release of pyridine was solvent- additional formation of [Pt(py)2(5′-GMP)Cl]+ (peak 5).
dependent, complex 4 was irradiated in 75% d4-MeOH/25% The photoproducts from irradiation (λirr 420 nm, 45 min) of
presence of chloride-containing PBS solution lead to the
D2O in the presence of 5′-GMP, with either 420 nm or UVA complex 3 (1.5 mM) plus 5′-GMP (2 mol. eq) in MeOH : H2O
Fig. 5 1H NMR spectra (500 MHz, PBS in D2O, pH 7.4, dioxane) of complex 1 and 5’-GMP (2 mol eq.) under the following conditions: (A) dark; then
following irradiation for increasing (total) doses of 463 nm light: (B) 10.2 J cm−2; (C) 35.4 J cm−2; (D) 70.8 mJ cm−2; followed by an additional dose
of shorter wavelength (420 nm) light (E) 30.5 J cm−2
.
●
◀
and denote the two main photoproducts. The sharp singlet at 2.40 ppm corresponds to
free succinate. (F) 195Pt NMR spectrum (129 MHz) following irradiation, showing resonance at −2210 ppm, assignable to the previously reported
trans-[Pt(N3)(5’-GMP)(py)2].29
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 10593–10607 | 10597

View Article Online
Paper
Dalton Transactions
EPR spectroscopy
Compounds 1, 2 and 4 (1 mM, in PBS) were irradiated with
green light (517 nm, 33 mW cm−2) in the presence of the spin-
trap 5,5-dimethyl-pyrroline N-oxide (DMPO) (2 mM). The yield
of trapped azidyl radicals (DMPO-N3) was determined by the
double integration of the EPR signals, quantified by comparison
to Tempol (Fig. 8A). EPR spectra of the solutions were also
recorded in the dark, during which time no azido radical trap-
ping was observed. Analysis of the DMPO-N3 production at
regular time-points (0, 7, 21, 28, 35 min) during the irradiation
demonstrated that the two compounds with carboxylate ligands
(1 and 2) produced azido radicals when irradiated with 517 nm
light, whereas the dihydroxido compound 4 did not produce an
appreciable concentration of azidyl radicals. Complexes 1–4
were then investigated for their ability to generate azidyl radicals
in DMF/H2O (75%/25% v/v), since it was a suitable solvent
system for allowing direct comparison of all complexes.
The yield of azido radicals was determined at regular time-
points during irradiation (0, 7, 14, 21, 28, 35 min), following a
similar trend to the PBS solution, with compounds 1 and 2
giving rise to more intense EPR signals than compound 4,
which, with the inclusion of DMF, did produce azidyl radicals
under green light irradiation.
Fig. 6 Chromatogram (λdetection 240 nm, ACN/H2O, 0.1% formic acid)
for complex 1 in the dark (black trace) or after irradiation in the presence
of 2 mol. eq of 5’-GMP in H2O (red trace, 1 (0.5 mM), λirr 420 nm, 30.5 J
cm−2) and in PBS (blue trace, 1 (1.5 mM) λirr 463 nm, 153.6 J cm−2) with
identifiable peaks assigned. Corresponding mass spectral data for num-
bered peaks are given in Table S6.†
(75% : 25%, v/v) were also analysed by LCMS (Fig. 7). In
addition to the platinum photoproducts, photoproducts
corresponding to loss of pyridine were observed, and free N-MI
ligand was also detected (peak 7) at 152.0715 m/z.
Fig. 7 Chromatogram (λdetection 240 nm, ACN/H2O, 0.1% formic acid)
for complex 3 in the dark (black line) and after irradiation with 420 nm
light in the presence of 2 mol. eq of 5’-GMP in H2O (red line) with ident-
ifiable peaks assigned. Corresponding mass spectral data for numbered
peaks are given in Table S7.†
Fig. 8 Time dependent trapping of azidyl radicals with DMPO (2 eq.)
upon continuous irradiation of complexes with green light (517 nm,
33 mW cm−2); (A) complexes 1, 2 and 4 PBS (1 mM) and (B) complexes
1–4 in DMF/H2O (75%/25% v/v, 2 mM).
10598 | Dalton Trans., 2021, 50, 10593–10607
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
Compound 3 produced only a very low yield of azidyl rad- the complexes in the dark. Instead the mean viability obtained
icals (Fig. 8B). To investigate the effect of the N-MI ligand on in the dark at the maximum administered dose is given.
the yield of azidyl radicals, a solution of complex 4 (2 mM, Cytotoxicity experiments to determine IC50 values also
DMF/H2O (75%/25% v/v)) was irradiated with shorter wave- compare the percentage of cell viability of untreated cells
length light (blue, 463 nm, 64 mW cm−2) in the presence of between the irradiated and dark samples, in all cases there is
free N-MI (2 mol eq., 4 mM) and DMPO (2 eq., 4 mM) and the no statistical difference between these populations. Such con-
azido radical yield was measured at regular time-points. This trols guarantee that the reduction in cell viability observed in
demonstrated a marked decrease in azido radical trapping in the samples is as a results of the complex activation rather
presence of the free ligand compared to in its absence (Fig. 9).
than a light effect. Poorly soluble complex 3 was solubilised in
Phototoxicity and cell uptake studies. The cytotoxicity and DMSO and diluted into buffered salt solution (0.7% v/v) before
photocytotoxicity of complexes 1–4 (IC50 values, Table 2) were administering to the cells. Therefore, it was tested at a
determined in three cancer cell lines (A2780 (ovarian), maximum of 20 µg ml−1 (33 µM) instead of 100 µg ml−1. In the
A2780cis (ovarian, cisplatin-resistant), and OE19 (oesophageal) absence of a (dark) cytotoxic IC50 value, the PI of 3 cannot
as described in the ESI.† Cells were exposed to the complexes therefore be directly compared to 1, 2 and 4, as it will be artifi-
for 1 h in the dark, and then either irradiated or sham-irra- cially lower. Replacement of a hydroxide ligand (4) with either
diated with blue light (5 J cm−2 λmax: 420 nm). IC50 values were a succinate (1) or 4-oxo-4-propoxybutanoate (2) ligand slightly
determined 24 h post-exposure, and photocytotoxicity indices enhanced the PI for A2780 cells, and increased the PI of the
(PI = IC50(dark)/IC50(light)) were calculated, where a large PI is complex in the platinum-resistant A2780cis cell line. Inclusion
desirable, indicating low toxicity in the dark and high photo- of either succinate (1) or 4-oxo-4-propoxybutanoate (2) did not
toxicity. In most cases it was not possible to determine an IC50 increase the PI in OE19 oesophageal cells, in fact, the 4-oxo-4-
value for the sham-irradiated cells, due to the low toxicity of propoxybutanoate ligand appeared to decrease it, relative to
complex 4. Complexes 2 and 3 were similarly phototoxic in the
platinum-sensitive and platinum-resistant ovarian cancer cell
lines, whereas complexes 1 and 4 were slightly less phototoxic
in the A2780cis cell line. Complex 3 was the least phototoxic of
the four complexes, but was as active against the oesophageal
line as towards the ovarian cells. The potential protective effect
of tryptophan (a known azido radical scavenger)35,36 was inves-
tigated; irradiation of complexes 1 and 4 in the presence of
1 mM tryptophan reduced the phototoxicity by 3-fold and
4-fold respectively (Table 3). To investigate the phototoxicity of
the complexes under irradiation with longer wavelength light,
complex-treated A2780 cells were irradiated with either green
(517 27 nm), or yellow (570 27 nm) light, or sham-irra-
diated from a light guide in stirred solution. Viability was
determined 72 h post-exposure, to interrogate anti-proliferative
activity. The phototoxicity of complex 1 with green light was
investigated in both a complex- and light-dose dependent
Fig. 9 [DMPO-N3] signal upon continuous irradiation (463 nm) of
manner, and it increased with both increasing concentration
complex 4 (2 mM) alone, or in the presence of 2 mol eq. of the N-MI
ligand in DMF/H2O (75%/25%) at ambient temperature.
and light dose (Fig. 10A). Complexes 1 (33 µM) and 2 (35 µM)
Table 2 Phototoxicity of 1, 2, 3 and 4 in human carcinoma cell lines 24 h post-activation with 5 J cm−2 visible blue light (λmax: 420 nm)
Cell linea
A2780
A2780cis
OE19
Viability SE
at MAD (%)
Viability SE
at MAD (%)
Viability SE
at MAD (%)
Complex IC50/µM 95% CI
PI
IC50 /µM 95% CI
PI
IC50/µM 95% CI
PI
1
2
3
4
2.7
3.7
13.4
5.1
2.2–3.4
3.0–4.6
6.0–30.3 79.5 5.6
3.7–7.0 98.7 8.4
65.5 13.8
62.2 5.1
>49
>44
>2
5.4
4.0
13.0
8.0
4.3–6.7
2.9–5.4
6.8–24.6 77.5 1.6
5.2–12.3 81.4 5.3
75.4 2.3
59.9 6.8
>37
>41 10.3
>2
>26
8.2
5.3–12.7
Wide
10.3–29.8 85.5 2.1
6.5–10.8 98.6 5.1
89.3 16.4
74.4 3.5
>21
>15
>1.9
>25
17.5
8.4
>41
a IC50: concentration of drug that inhibited dye uptake by 50%. 95% CI: 95% confidence interval of the IC50 value. MAD: maximum administered
dose. In the absence of a cytotoxic IC50 value, the reported PI values are ‘at least’ the value stated (denoted by ‘>’). Results are the mean of 2–5
independent experiments performed in triplicate.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 10593–10607 | 10599

View Article Online
Paper
Dalton Transactions
Table 3 Phototoxicity of 1 and 4 in A2780 carcinoma cells 24 h post-
activation with 5 J cm−2 visible blue light (λmax: 420 nm) in the presence
of 1 mM L-tryptophan
respectively). No reduction in cell viability by complex 1 was
observed under irradiation with even longer-wavelength
light (570
27 nm, 175 µM complex 1: Fig. S14†). These
experiments also demonstrate the lack of dark toxicity of the
complexes compared to the control cells treated with light
only.
Without
With 1 mM
tryptophana
tryptophana
Protection
factor
Complex
IC50 (µM)
95% CI
IC50 (µM)
95% CI
The phototoxicity of complex 1 correlated with the results
of a comet assay (which was not performed with complex 2 as
well, due to time constraints), which showed increased DNA
migration when A2780 cells were irradiated (100 J cm−2
420 nm light) in the presence of the drug (Fig. 11A). This
increased migration might be due to reactive species produced
on photoactivation, but is more likely due to cell death, as this
combination of drug and light is cytotoxic. DNA migration was
not increased when the same concentration of drug was acti-
vated with longer-wavelength (100 J cm−2 517 nm) light
(Fig. 11B) which could be due to reduced toxicity at this wave-
length (ca. 50% cell death) or the nature of the reactive species
produced.
1
4
2.4
5.3
2.1–2.7
3.1–9.2
7.3
22.5
2.3–23.3
7.6–67.0
3.0
4.2
a IC50: concentration of drug that inhibited dye uptake by 50%. 95%
CI: 95% confidence interval of the IC50 value. Results are the mean of
2 independent experiments performed in triplicate.
Fig. 10 Anti-proliferative activity of complexes 1–4 after photoactiva-
tion with green light (517 27 nm) in A2780 cells: (A) in the presence of
increasing concentrations of complex 1; (B) in the presence of com-
plexes 1 or 3 and (C) in the presence of complexes 2 or 4. Data points
represent the mean SD of two independent experiments performed in
triplicate.
showed similar phototoxicity under irradiation with 517 nm
light; 50% cell viability 72 h post-irradiation was observed at
light doses of 112 J cm−2 [86–146 J cm−2] for 1 (Fig. 10B) and
105 J cm−2 [74–150 J cm−2] for complex 2 (Fig. 10C).
Fig. 11 DNA migration in A2780 cells following photoactivation of
complex 1 (35 µM) with 10 J cm−2 of either (A) blue 420 nm or (B) green
Complexes 3 (33 µM) and 4 (45 µM) demonstrated negli-
gible photoactivity under these conditions (Fig. 10B and C 517 nm light.
10600 | Dalton Trans., 2021, 50, 10593–10607
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
nates, and isocyanates to form carboxylates, carbonates and
carbamates, respectively.27,37–39 Three novel complexes 1–3
have been synthesised, axial ligation was achieved through
reaction of 4 with the corresponding anhydride (either succi-
nic anhydride or N-methylisatoic anhydride) through a ring-
opening reaction. Complex 2 was then obtained from 1 via
esterification of the succinate ligand with sodium propanolate,
resulting in formation of a 4-oxo-4-propoxybutanoate group.
Incorporation of the N-methylisatoate ligand in complex 3 was
chosen as it was anticipated that it could impart greater photo-
sensitivity to the complex.
NMR spectroscopy and X-ray crystallography
The higher 195Pt NMR chemical shifts (D2O) for complexes 1
(1059 ppm) and 2 (1065 ppm) in comparison to 4 (942 ppm)
indicate that the 195Pt nucleus is more deshielded following
carboxylation; complex 3 was measured in DMSO-d6 due to
solubility constraints and therefore a direct comparison for
3 is less helpful. Lengthening of the Pt–P bond in the case
of platinum complexes containing phosphine ligands has
been reported to increase the 195Pt NMR chemical shift.40
Here, the Pt–O bonds to carboxylate ligands are systemati-
cally marginally longer than the Pt–OH bond for 1–3
(Table S2†), which may account for the increase in the
chemical shift.
Activation of prodrugs with longer-wavelength light is desir-
able to achieve greater penetration into tissues. Therefore, the
longer wavelength absorption bands in the UV-vis spectra are
of particular interest for photoactivatable metallodrugs. Green
light has greater clinical relevance, penetrating ca. 50% more
deeply into tissues than shorter wavelength blue light.
Complexes 1–4 show two absorption bands at ca. 260 nm and
300 nm (Fig. S6 and S8†), composed mainly of transitions with
1LMCT character. The absorption bands in compounds 1–3
are red-shifted in comparison with those in the precursor 4,
indicating that modification of the axial group has the poten-
tial to enhance the photolability of the Pt(IV)-diazido prodrugs
when irradiated at longer wavelengths. The nature of the axial
ligand also affects the pharmacological properties of the
drugs, such as lipophilicity, water solubility and stability.
Biological reducing agents in the cell and in the blood plasma
include ascorbate, glutathione as well as cysteine-rich metal-
lothioneins.41 Glutathione levels vary widely, but in healthy
adults, mean GSH concentrations are reported as 849 63 μM
Fig. 12 Top: cellular uptake studies of compounds 1–4 (ng Pt/106 cells)
at 37 °C with cisplatin for comparison. Bottom: temperature-depen-
dence on the cellular uptake of complexes 3 and 4, after 1 h drug incu-
bation in the dark. Each experimental condition was carried out in tripli-
cate in the A2780 human ovarian cancer cell line.
Cellular uptake studies
Cellular uptake studies were carried out for complexes 1–4
with the established Pt(II) anti-cancer complex cisplatin as a
control, in the A2780 cell line. The concentration used for the
uptake of all of the complexes was 20 μM, since no dark tox-
icity was observed at this concentration for any of the com-
pounds. Uptake of Pt from the compounds at 37 °C followed
the order (from highest to lowest): 3 (300 ng Pt/106 cells) ≫ 4
(32 ng Pt/106 cells) > cisplatin (29 ng Pt/106 cells) > 2 (25 ng Pt/
106 cells) > 1 (15 ng Pt/106 cells) (Fig. 12, top). The tempera-
ture-dependence of cellular uptake was studied by measuring
uptake at 4 °C, 18 °C and 37 °C (Fig. 12, bottom) for com-
pounds 3 and 4. Both compounds followed the same trend,
with uptake increasing with increasing temperature.
in whole blood and 3.39
1.04 μM in blood plasma.42
Intracellular levels are typically higher, around 2–8 mM in the
cytosol.43 Compared to disease-free tissue, glutathione levels
tend to be elevated in ovarian tumours.44 It is known that com-
plexes with axial acetate ligands have the potential to be more
readily reduced than those with hydroxido ligands: the mono-
carboxylato Pt(IV) analogue of oxoplatin (cis,trans,cis-
[PtCl2(OH)2(NH3)2]) is reduced faster in the presence of gluta-
thione than oxoplatin.41 We determined that 1 showed good
stability in the presence of 2 mol eq. of GSH, which is impor-
Discussion
Synthesis and characterisation
trans-Dihydroxido Pt(IV) complexes can be axially functiona- tant for its potential development as a photoactivatable Pt(IV)
lised through use of anhydrides, acid chlorides, pyrocarbo- prodrug, since these need to be relatively stable in a biological
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 10593–10607 | 10601

View Article Online
Paper
Dalton Transactions
environments following administration before being activated EPR spectroscopy
by irradiation.
The formation of azidyl radicals when Pt(IV) diazido complexes
are irradiated is thought to play a key role in the mechanism of
action of these prodrugs.35 Therefore the photoproduction of
azidyl radicals from 1–4 was studied by EPR spectroscopy.
Under green (517 nm) light irradiation, PBS solutions of
complex 4 did not generate trapped radicals, whereas 1 and 2
produced radicals at concentrations up to 68 μM and 55 μM,
respectively. The identity of the trapped radical was confirmed
as DMPO-N3 by matching the hyperfine splitting constants
(HFSC) to the theoretical data reported previously.47,48
Hydroxyl or carboxylate radicals, which could potentially be
released from the Pt(IV) complex, would give a 1 : 2 : 2 : 1
pattern when trapped by DMPO; this signal was not detected
in the spectrum. Initial production of radicals (after 7 min of
irradiation) was lower for 1 than for 2, although the azidyl
radical yield after 14 min of irradiation was slightly higher for
1 than for 2. Both complexes 1 and 2 gave rise to significantly
greater azido radical production under green light irradiation
than the parent dihydroxido complex 4. A similar experiment
was carried out in DMF/H2O (75%/25% v/v) to enable compari-
son between all four complexes and to investigate the effect of
solvent. Compounds 1, 2 and 4 in DMF/H2O (75%/25% v/v)
gave broadly similar results as seen in PBS, following pro-
longed irradiation, with 1 giving the highest and 4 the lowest
azidyl radical yield, although at the first measurement (7 min)
in DMF/H2O 1 showed greater azide radical production than 2,
whereas 2 had showed greater initial azide radical production
than 1 in PBS. Irradiation of compounds 3 and 4 resulted in a
similar, although modest, concentration of azide radicals
being trapped (which for 4 was higher than the radical pro-
duction observed in PBS), despite the rate of photodegradation
of 3 being ∼3-fold higher than that of 1 (Fig. 4). The
N-methylisatoate ligand (axially bound in 3) may act as a
radical scavenger, which could explain the low levels of azido
radical trapping for 3. In order to confirm this hypothesis,
complex 4 was irradiated (λ = 463 nm, 64 mW cm−2) in the
presence of N-methylisatoic acid (2 mol eq.). The production
of azidyl radicals was entirely quenched by the presence of
N-methylisatoate acid, suggesting that upon release, this
DFT and TD-DFT
The theoretical UV-vis maximum appears at 286 nm
(33 748 mol−1 cm−1) compared to the experimental maximum
at 299 nm (ε = 18 796 mol−1 cm−1). TD-DFT calculations there-
fore slightly underestimate the absorption maximum by ca.
10 nm and overestimate the value of the extinction coefficient.
The method used for these TD-DFT calculations is the same as
that used for 4, for which a closer match between the theore-
tical and the experimental extinction coefficient was found.30
The larger difference between calculation and experiment for 1
is therefore ascribed to the inclusion of the succinate ligand.
Although functional and basis set benchmarking testing
might improve the agreement between calculated and experi-
mental transitions, the results obtained are still satisfactory to
characterise the nature of the UV-Vis absorption bands.
Overall, our results show that light excitation would results in
the formation of low-energy singlet states of mixed nature, in
which the strong 1LMCT component and the nature of the
LUMOs involved is consistent with a variability in the
photoproducts.
Besides, transition metal complexes such as d6-Pt(IV) deriva-
tives usually undergo intersystem crossing (IC) to triplet states
upon light-excitation. Therefore, the nature of the lowest-lying
triplet state is crucial to understand their photochemical be-
havior, since such a state is significantly populated before the
molecule relaxes to the ground state. Elongation of bonds in
the triplet state indicates that they could break upon photo-
excitation, and therefore suggests which ligands are most
likely to dissociate. The difference between the lowest-lying
excited-state geometry and the ground state demonstrates that
for complex 1, both Pt-azido bonds show the biggest elonga-
tion, the Pt–OH bond does not elongate, the Pt-succinate bond
is elongated, and one of the Pt-pyridine bonds is elongated
whereas the other is shortened (Table S2†). This is consistent
with the experimental observations which include release of
the succinate ligand (Fig. 5) and azido radicals (Fig. 8).
Solution reactivity
•
ligand may readily react with N3 Fig. 9. Methyl anthranilate,
Photoreductions of 1–4 were monitored by UV-vis spectroscopy
(spectral output of the light sources are given in Fig. S7†),
showing a decrease of the N3 → Pt ligand–to–metal charge-
transfer (LMCT) band (at ca. 300 nm) due to the loss of an
azido ligand(s) (Fig. S8†), following an exponential decay
profile (Fig. 4, 4). The photoreduction occurred faster for the
which is structurally related to the N-methylisatoate ligand,
has been shown to quench singlet oxygen due to the presence
of the anilinic moiety.49
Biological data
monocarboxylato complexes 1–3 than for the hydroxido The phototoxicity of complexes 1, 2 and 4 towards A2780 cells
complex 4, indicating that the carboxylate functionality after irradiation with green 517 nm light did not correlate with
enhances the photoreactivity of these complexes. Furthermore, levels of Pt uptake by the cells (Fig. 12), but did correlate with
complex 2, showing the fastest photodegradation kinetics, azidyl radical production in PBS (Fig. 8), implicating the azide
shows the longest Pt–O (carboxylate) bond, which photodis- radicals in the phototoxic mechanism of action of the com-
sociates upon irradiation. The increased rate of photodegrada- plexes. If the azide radicals are scavenged by the
tion for the monocarboxylate complexes correlates with the N-methylsatoate ligand this could explain the lower relative
extinction coefficient, which is higher at longer wavelengths phototoxicity of complex 3, despite cellular uptake studies
(Fig. S6†).
demonstrating a Pt content in A2780 cells for 3 at least 10-fold
10602 | Dalton Trans., 2021, 50, 10593–10607
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
that detected for 1, 2, and 4. Furthermore, co-incubating PBS (139 mM chloride) resulted in a chloride readily binding
A2780 cells with 1 mM tryptophan protected against blue-light to give [PtII(Cl)(py)2(5′-GMP)]+. This is consistent with a report
induced phototoxicity in cells which had been treated with that the amount of platinated DNA was significantly reduced
either 1 or 4, with slightly more protection afforded to the cells when 4 was irradiated in the presence of 100 mM chloride.51
treated with 4. Tryptophan is an efficient scavenger of singlet The [M − N3]+ species observed for 1 in the mass spectrum
oxygen50 and can also quench azidyl radicals.36 It should be before irradiation (529.0750 m/z, Fig. S1,† top) was also
noted that blue light itself will photosensitise endogenous cel- observed post-irradiation, but only at low intensity (529.0790
lular chromophores, a process which could be affected by m/z, peak 8, Fig. 6 and Table S6†). We suggest that in-source
tryptophan, but the sensitivity of the cells themselves to blue fragmentation and loss of azide may occur in the mass
light was insignificant (survival 103 5.6%) and this was not spectrometer in the absence of irradiation, but this does not
significantly altered by the presence of tryptophan.
seem to be significantly enhanced by irradiation. The corres-
The irradiation of both 1 and 4 in the presence of 5′-GMP ponding [M − N3]+ species was also detected for 3, before
(2 mol eq.) in PBS produced the mono-5′-GMP Pt(II) adduct (562.1210 m/z, Fig. S1,† bottom) and following irradiation
trans-[Pt(py)2(5′-GMP)(N3)] (Fig. S9†). However, 23% of 5′-GMP (562.1171 m/z, peak 8, Fig. 7 and Table S7†). This could be a Pt
was bound to platinum in the case of 1, whereas only 9% was (III) complex or
a metallacyclic species in which the
bound in the case of compound 4, as judged by analysis of the N-methylanthranilate ligand is doubly deprotonated and coor-
starting/product 5′-GMP resonance S1′/P1′. This result, along dinates to Pt(IV) in a bidentate chelating mode. Pt(III) inter-
with the UV-Vis and EPR spectroscopy studies demonstrates mediates are not unusual in mass spectrometry, but the
the enhanced photoreactivity of the carboxylate complex 1 elution of the species from the HPLC column suggests that it
versus the dihydroxido complex 4. TD-DFT calculations gave an is stable, and therefore may be more likely to correspond to a
insight into the enhanced photoactivity, showing that 1 has metallacyclic Pt(IV) complex. Very little 5′-GMP binding was
dissociative transitions which can extend up to 469 nm. The observed for 3 in 75% MeOH-d4/25% D2O (v/v) following
photoproducts from the irradiation of 1 with blue light irradiation, and photoproducts of
3 detected by LC-MS
(463 nm) were characterised by H NMR and 195Pt-NMR spec- included the axial ligand (N-MI) (Fig. 7 and Table S7†),
troscopy as well as LCMS. The release of succinate as well as whereas for 1, no assignable photoproducts retained the succi-
azidyl radicals was observed (Fig. S10†). The reduction of Pt(IV) nate ligand. In general, the photoproducts detected by LCMS
to Pt(II) requires the gain of two electrons, however, the only were consistent with those previously seen following photopla-
radical detected by EPR spectroscopy was the azidyl radical. tination of the synthetic oligodeoxynucleotide 5′dCATGGCT
Since the major product generated after irradiation is a mono- (containing two consecutive guanine nucleobases) by 1.15
1
azido Pt(II) complex, the loss of both azides as azidyl radicals
The attachment of fluorescent tags to platinum compounds
cannot be the main photochemical pathway to products in is useful to study their temporal and spatial distribution in
PBS. The second electron could originate from either the living cells. Frequently, the attachment of fluorescent tags to
hydroxido or the carboxylato ligands, which could react rapidly Pt(IV) results in fluorescence quenching, when reduced to Pt
with another radical or solvent molecule and thus be (II), within the cell, fluorescence enhancement occurs.52,53 The
quenched.
fluorescence emission spectrum of 3 – which incorporates a
The assignment of trans-[Pt(py)2(5′-GMP)(N3)] was readily fluorescent ligand (N-MI) in the axial position – showed a
1
confirmed as the H and 195Pt-NMR spectral data match those single broad band with maximum at 420 nm, similar to the
chemical shifts assigned previously for trans-[Pt(py)2(5′-GMP) maximum emission wavelength of the free ligand N-MIA, and
(N3)].29 Irradiation of 1 with slightly shorter wavelength light repeated excitation at 320 nm caused an enhancement in the
(420 nm vs. 463 nm) did not produce in any significant change fluorescence emission at 420 nm, which is consistent with the
in the photoproducts, the change in irradiation wavelength detection of free N-MIA ligand by LCMS (Fig. S12†).
appeared to only affect the rate of formation of products.
Pyridine release was not observed when complex 4 was irra-
Similar to the results obtained upon excitation at 463 nm, diated with UVA light in aqueous solution.29 However,
apart from the major product trans-[Pt(py)2(5′-GMP)(N3)], only irradiation of either 3 or 4 in 75% MeOH-d4/25% D2O (v/v),
very low intensity signals corresponding to other 5′-GMP released of pyridine, particularly upon UVA irradiation
adducts were visible in the 1H-NMR spectra.
(Fig. S11†). This highlights the importance of the solvent and
LC-MS analysis was carried out after the irradiation of com- the environment on the photodecomposition pathways of
pound 1 in the presence of 2 mol eq. of 5′-GMP, with the these complexes; if these drugs accumulate in a hydrophobic,
majority of the species successfully assigned (Tables S1 and non-aqueous environment within the cell (e.g. membranes)
S6†). In H2O, in addition to trans-[Pt(py)2(5′-GMP)(N3)], other different photoproducts may be produced. The release of pyri-
5′-GMP adducts were observed including trans-[Pt(ACN)(GMP) dine is consistent with TD-DFT calculations performed on
(py)2]2+ trans-[Pt(ACN)(GMP-H)(py)2]+, [Pt(N3)(py)3]+ and the compounds 1 and 4 29 which showed that the LUMO and
di-platinum species [Pt2(µ-5′-GMP-H)(N3)2(py)4]+. The ACN LUMO+1 are strongly antibonding towards all ligands.
adducts form due to use of acetonitrile as HPLC solvent, but
Complexes 1 and 2 showed promising photocytotoxicity
in biological media, it is anticipated this coordination site towards ovarian cancer cells (A2780) with low-dose blue light
would be readily occupied by another ligand. Irradiation in (420 nm, 5 J cm−2). The phototoxicity of 1 and 2 are 2-fold
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 10593–10607 | 10603

View Article Online
Paper
Dalton Transactions
higher compared to compound 4. Complex 3, was the least N-methylisatoate (1, 2 and 3, respectively), were synthesized
phototoxic with an IC50 value of 13.4 μM. This is despite and characterised by NMR spectroscopy, MS, UV-vis spec-
complex 3 showing ca. 10-fold greater uptake in comparison to trometry and X-ray crystallography. Photoactivation studies
the rest of the compounds (Fig. 12, top). The comparatively comparing complexes 1–3 to their dihydroxido synthetic pre-
modest IC50 for 3 may be related to the low yield of azidyl rad- cursor trans,trans,trans-[Pt(N3)2(OH)2(py)2] 4 indicate a 3-fold
icals or other factors, such as sub-cellular distribution. Cell faster rate of photodegradation for 1–3 through monitoring
uptake of 2 is ca. 2-fold higher than that of 1 in A2780 cells loss of the LMCT band in the UV-vis absorbance spectrum
(from 15 to 25 ng/106), attributable to the increased hydropho- when irradiated with green light (517 nm). This is consistent
bicity arising from the propyl ester functionality in 2. However, with TD-DFT transitions indicating greater absorption at
only a slight improvement in the cytotoxicity was observed longer wavelengths for 1 in comparison to 4. Consistent with
(IC50 values of 2.7 μM for 1 and 3.7 μM for 2). Complex 4 this, complex 1 also showed significantly greater (2.6-fold) 5′-
showed greater accumulation than both 1 and 2, but lower GMP complexation than 4 when irradiated with the same dose
photocytotoxicity, underlining the importance of triggered of blue (463 nm) light. EPR spectroscopic studies of 1, 2 and 4
reduction and production of azido radicals in the biological demonstrated that considerably higher quantities of azidyl rad-
effects.
icals were generated under green light (517 nm) irradiation of
The transport of molecules across the membrane of a cell 1 and 2, with correspondingly lower (DMF/H2O) or minimal
can be either active (energy-dependent) or passive (energy- (PBS) radical production for 4. Complex 3 incorporates the
independent).54 Platinum(II) complexes are taken up by both fluorescent ligand N-methylisatoate (N-MIA). Irradiation of 3
passive and active transport. For cisplatin, copper transporters also resulted in an increase in fluorescence, consistent with
(e.g. Ctr1) play a major role.55–57 Organic cation transporters platinum(IV) reduction and N-MIA release. LC-MS studies upon
constitute another class of important transporters, responsible blue light irradiation (420 nm) resulted in the release of the
for the uptake of picoplatin in mice.58 Cisplatin shows axial ligands (succinate for 1 and N-MIA for 3) and formation
minimal cellular uptake at 4 °C (0.005 ng of Pt per 106 cells) of a variety of photoproducts. Irradiation of 3 and 4 in non-
consistent with predominantly energy-dependent transport.59 aqueous solvent (75% MeOH-d4/25% D2O) in the presence of
In contrast, complexes 3 and 4 accumulated to a greater extent 5′-GMP also resulted in release of pyridyl ligand(s) from the
at this temperature, with accumulation increasing with temp- complexes, which is not observed for 4 in aqueous or buffered
erature (Fig. 12, bottom). The uptake of these complexes is solutions. The pKa of the pendant carboxylate of complex 1
therefore likely to be through a combination of both passive (5.13), suggests it may be negatively charged under biological
and active transport, the passive transport appears to be conditions, which could explain the lower cellular uptake of 1
enhanced greatly by the incorporation of an aromatic ligand in in comparison to 2 and 4. Although complex 3 showed high
the axial position, as shown by complex 3.
levels of cellular accumulation, it showed poor aqueous solubi-
Complexes 1 and 2 were also more effective against the Pt- lity in comparison to 1 and 2, and also lower azido radical trap-
resistant A2780cis cell line compared to complex 4. These three ping upon illumination with green light, possibly as a result of
complexes all showed activity against OE19 oesophageal carci- quenching by the N-MIA ligand. Cellular uptake studies indi-
noma cells with blue light (420 nm). Blue light is already in cated that complexes 3 and 4 may enter cells through a combi-
widespread clinical use for treating actinic keratoses – precan- nation of passive diffusion and active transport.
cerous lesions on the skin, and has been shown to be more
appropriate for treating superficial oesophageal cancerous A2780 cells showed that complexes 1 and 2 (IC50 = 2.7 and
lesions in animal models than green light, since it causes less 3.7 μM) were ∼2-fold more photocytotoxic than 4 (IC50
Phototoxicity studies with 420 nm (blue) irradiation on
=
damage to the underlying healthy tissue.60 For other appli- 5.1 μM). These complexes were also more effective towards the
cations however, longer-wavelength light activation is more platinum-resistant A2780cis cell line in comparison to 4, par-
desirable, and therefore the response of 1 and 2 to irradiation ticularly complex 2. The propyl derivative 2 was synthesised to
with green (517 nm) light and photocytotoxic activity towards produce a neutral compound, potentially also with greater lipo-
both in ovarian cancer (and cisplatin-resistant ovarian cancer) philicity than 1, and although this demonstrated greater cellular
cell lines is encouraging. The photocytotoxicities (λmax 420 nm) accumulation in A2780 cells, this did not translate to greater
of 1–3 are broadly consistent with guanidinoneomycin-deriva- photocytotoxicity than 1 in this cell line. Crucially, both com-
tised succinate complexes in SK-MEL-28 cancer cell lines; in the plexes 1 and 2 also efficiently killed A2780 cells when activated
SK-MEL-28 line the succinate complex 1 was less active than by green (517 nm) light. Of the novel complexes, 3 was the least
complex 4,15 whereas complex 1 was slightly more active than 4 photocytotoxic (13.4 μM). Although the photoproducts formed
in the A2780, A2780cis and OE19 cell lines reported here.
by complexes 1, 2 and 4 are similar, the improved photoactivity
that the carboxylate ligand confers to 1 and 2 results in rapid
generation of Pt(II) photoproducts and higher concentrations of
azide radicals for a given irradiation dose.
Conclusions
The derivatisation of one axial position of platinum(IV)
Complexes trans,trans,trans-[Pt(N3)2(OH)(OCOR)(py)2] where azido complexes with carboxylate ligands enables them to be
OCOR succinate, 4-oxo-4-propoxybutanoate and activated with green (517 nm) light, whilst remaining stable in
=
10604 | Dalton Trans., 2021, 50, 10593–10607
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
the presence of biological reductants (GSH). Complexes 1 and propanolate (2 mL), formed by the dissolution of sodium
2 have the potential to be further modified or developed in metal (0.02 g) in absolute propanol (20 mL), was added and
their own right as phototherapeutics, particularly for super- the reaction carried out at room temperature for 48 h. DMF
ficial cancers; complex 2 in particular showed promising was removed under vacuum and the crude product was puri-
activity towards the cisplatin-resistant A2780cis cell line. The fied by column chromatography (eluent: EtOAc : MeOH, 11 : 1).
production of azido radicals by the complexes is clearly of After solvent removal from the combined fractions, the solid
crucial importance to their photocytotoxic mechanism of was recrystallised from DCM/diethyl ether (yield = 0.060 g,
action.
56%). Crystals suitable for X-ray crystallography were grown by
slow evaporation from H2O at ambient temperature. Purity
(HPLC, UV-Vis absorbance): 96%.
3
1H-NMR (D2O, 600 MHz, dioxane, ppm): δ = 8.76 (dd, J1H1H
Experimental
3
3
= 5.64 Hz, J
= 25.40 Hz, 4H, H6), δ = 8.28 (t, J1H1H = 7.73
195Pt1H
3
Synthesis and characterisation of synthetic precursors 5, 6 and
7 and 4 are given in the ESI.† The NMR spectroscopic assign-
ment of specific hydrogens and carbons for complexes 1–3 is
illustrated in Fig. S13.†
Caution! No problems were encountered during this work,
however heavy metal azides are known to be shock-sensitive
detonators, therefore it is essential that platinum azides are
handled with care. The Pt-diazido complexes were synthesised
and handled under dim lighting conditions.
Hz, 2H, H7), δ = 7.81 (t, J1H1H = 7.14 Hz, 4H, H8), δ = 3.95 (t,
3J1H1H = 6.72, 1H, H3), δ = 2.64, 2.63 (m, 2H, H1,2), δ = 1.54 (s,
3J1H1H = 7.28 Hz, 1H, H4), δ = 0.84 (t, J1H1H = 7.50 Hz, 1H, H5).
3
13C-NMR (D2O, 150 MHz, dioxane, ppm): δ = 179.29 (Cd), δ =
176.22 (Ca), δ = 149.84 (Ch), δ = 143.31 (Cj), δ = 127.64 (Ci), δ =
67.90 (Ce), δ = 31.78 (Cb), δ = 30.88 (Cc), δ = 21.91 (Cf), δ = 10.17
(Cg). 195Pt-NMR (D2O, 107 MHz, dioxane): 1065 ppm. ESI-MS:
[M + Na]+ (m/z) calc. (C17H22N8NaO5Pt). 636.1210; found, 636.2.
trans,trans,trans-[Pt(N3)2(OH)(N-MI)(py)2](3)
trans,trans,trans-[Pt(N3)2(OH)(succ)(py)2](1)
trans,trans,trans-[Pt(N3)2(py)2(OH)2] 3 (0.086 g, 18.25 mmol)
was dissolved in anhydrous DMSO (2 mL) and N-methylisatoic
anhydride (0.162 g, 91.23 mmol, 5 mol eq.) was added. The
reaction was stirred for 72 h at 40 °C under nitrogen with an
outlet to allow the release of CO2(g). The isolation of the
product was achieved by the addition of water (10 mL) at
which point the compound precipitated. For the removal of
excess ligand, repeated washing with diethyl ether was necess-
ary and carried out via centrifugation (yield = 0.020 g, 18%).
Crystals suitable for X-ray diffraction were grown by slow evap-
oration of a solution in DCM/MeOH at ambient temperature.
Carboxylation of one of the axial ligands of 4 was achieved by
modifying
a
published method.61 trans,trans,trans-[Pt
(N3)2(py)2(OH)2] 3 (0.150 g, 0.318 mmol) was dissolved in
anhydrous DMSO (1.8 mL) under N2, in the presence of acti-
vated molecular sieves. Succinic anhydride (1.1 mol eq.,
0.035 g) was added and the reaction stirred in the dark at
40 °C for 48 h. The suspension was filtered and water (∼2 mL)
was added. The yellow solution was allowed to stand at room
temperature until a crystalline solid precipitated (3–4 h). The
remaining solution was decanted and the crystalline precipi-
tate was washed with diethyl ether and then filtered (yield =
0.096 g, 52%). Crystals suitable for single-crystal X-ray diffrac-
tion were obtained by slow diffusion of water into the DMSO
reaction mixture at room temperature. Purity (HPLC, UV-vis
absorbance): 94%.
Purity (HPLC, UV-vis absorbance): 94%.
3
1H-NMR (DMSO-d6, 500 MHz, ppm): δ = 8.86 (dd, J1H1H
=
3
3
5.54 Hz, J
= 27.00 Hz, 4H, H7), δ = 8.30 (t, J1H1H = 7.56
195Pt1H
Hz, 2H, H9), δ = 7.88 (m, 5H, 4H8, H4), δ = 7.62 (q, 3J1H1H = 5.07
Hz, 1H, H5), δ = 7.26 (t, J1H1H = 7.78 Hz, 1H, H3), δ = 6.53 (d,
3
1H-NMR (D2O, 600 MHz, pH 5.18, ppm): δ = 8.77 (dd, 3J1H1H
= 5.78 Hz, 3J195Pt1H = 11 Hz, 4H, Ho), δ = 8.27 (t, 3J1H1H = 7.64 Hz,
3
3J1H1H = 7.53 Hz, 1H, H1), δ = 6.51 (t, J1H1H = 7.53 Hz, 1H, H2),
δ = 3.88 (d, 2J195Pt1H = 14.8 Hz, H10), δ = 2.70 (d, 3J1H1H = 5.00 Hz,
3H, H6). 13C-NMR (DMSO-d6, 125 MHz, ppm): δ = 171.26 Ca, δ
= 150.82 Cc, δ = 149.73 Cb, δ = 149.38 Ci, δ = 142.18 Ck, δ =
132.86 Cg, δ = 132.09 Cf, δ = 126.42 Cj, δ = 113.74 Cd, 110.00 Ce,
δ = 29.07 Ch. 195Pt-NMR (DMSO-d6, 107 MHz,): δ = 990 ppm.
ESI-MS: [M + H]+ (m/z) calc. (C18H20N9O3Pt) 605.1288; found,
605.1338.
3
2H, Hp), δ = 7.80 (t, J1H1H = 7.04 Hz, 4H, Hm,), δ = 2.58 (t,
3
3J1H1H = 6.90 Hz, 2H, H1), δ = 2.48 (t, J1H1H = 6.90 Hz, 2H, H2).
13C-NMR (d4-MeOD, 150 MHz, ppm): δ = 178.7, (Ca), δ = 177.57
(Cd), δ = 150.89, (Ce), δ = 143.15 (Cg), δ = 127.37 (Cf). 195Pt-NMR
(D2O, 129 MHz): δ = 1059 ppm.
ESI-MS: [M + Na]+ (m/z) calc. (C14H16N8NaO5Pt), 594.0741;
found, 594.0783.
trans,trans,trans-[Pt(N3)2(OH)(succ-Pr)(py)2](2)
Conflicts of interest
trans,trans,trans-[Pt(N3)2(OH)(Succ-Pr)(py)2]
4
(0.100
g,
There are no conflicts to declare.
0.176 mmol) was pre-dried using P2O5 and then dissolved in
dry DMF (1 mL) under argon. A stock solution of CDI (1,1′-car-
bonyldiimidazole) in DMF (0.194 M) was prepared under dry
conditions and 1.2 mL was added (1.2 mol eq.) to the solution
Acknowledgements
of the platinum complex and allowed to stir at 50 °C for NF thanks the Wellcome Trust (201406/Z/16/Z); Cancer
20 min, under continuous purging with argon. Then, sodium Research UK (CR-UK) grant number C5255/A18085 through
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 10593–10607 | 10605

View Article Online
Paper
Dalton Transactions
the Cancer Research UK Oxford Centre and the John Fell Fund 20 Q. Zhang, G. Kuang, S. He, H. Lu, Y. Cheng, D. Zhou and
for funding. NF thanks Profs. Stephen Faulkner for support. Y. Huang, Nano Lett., 2020, 20(5), 3039.
PJS and NF thank the EPSRC (for grant EP/P030572/1 and stu- 21 M. Ravera, E. Gabano, I. Zanellato, I. Bonarrigo,
dentship for ES), PJS also thanks the ERC (grant 247450). L. S.
performed this work under the Severo Ochoa Centres of
E. Escribano, V. Moreno, M. Font-Bardia, T. Calvet and
D. Osella, Dalton Trans., 2012, 41, 3313.
Excellence Programme run by the Spanish State Research 22 A. Alvarez-Valdes, J. M. Perez, I. Lopez-Solera, R. Lannegrand,
Agency, grant no. CEX2018-000867-S (DIPC). L. S. also thanks
the Spanish Multi-MetDrugs network (RED2018-102471-T) for
fruitful discussion.
J. M. Continente, P. Amo-Ochoa, M. J. Camazon, X. Solans,
M. Font-Bardia and C. Navarro-Ranninger, J. Med. Chem.,
2002, 45, 1835.
23 K. R. Barnes, A. Kutikov and S. J. Lippard, Chem. Biol.,
2004, 11, 557.
24 C. F. Chin, Q. Tian, M. I. Setyawati, W. Fang, E. S. Q. Tan,
D. T. Leong and W. H. Ang, J. Med. Chem., 2012, 55, 7571.
References
1 M. Galanski, Recent Pat. Anticancer Drug Discov., 2006, 1, 285. 25 M. X. Tan, Z. F. Wang, Q. P. Qin, B. Q. Zou and H. Liang,
2 L. Kelland, Nat. Rev. Cancer, 2007, 7, 573. Dalton Trans., 2020, 49, 1613.
3 R. Oun, Y. E. Moussa and N. J. Wheate, Dalton Trans., 2018, 26 S. Mukhopadhyay, C. M. Barnés, A. Haskel, S. M. Short,
47, 6645.
K. R. Barnes and S. J. Lippard, Bioconjugate Chem., 2008,
19, 39.
4 M. D. Hall, C. T. Dillon, M. Zhang, P. Beale, Z. Cai, B. Lai,
A. P. J. Stampfl and T. W. Hambley, J. Biol. Inorg. Chem., 27 C. M. Giandomenico, M. J. Abrams, B. A. Murrer,
2003, 8, 726.
J. F. Vollano, M. I. Rheinheimer, S. B. Wyer, G. E. Bossard
and J. D. Higgins, Inorg. Chem., 1995, 34, 1015.
28 T. Babu, A. Sarkar, S. Karmakar, C. Schmidt and D. Gibson,
Inorg. Chem., 2020, 59, 5182.
5 L. T. Ellis, H. M. Er and T. W. Hambley, Aust. J. Chem.,
1994, 48, 793.
6 S. Alonso-de Castro, A. Terenzi, S. Hager, B. Englinger,
A. Faraone, J. C. Martínez, M. Galanski, B. K. Keppler, 29 N. J. Farrer, J. A. Woods, L. Salassa, Y. Zhao,
W. Berger and L. Salassa, Sci. Rep., 2018, 8, 1.
7 Z. Deng, N. Wang, Y. Liu, Z. Xu, Z. Wang, T.-C. Lau and
G. Zhu, J. Am. Chem. Soc., 2020, 142, 7803.
8 V. E. Y. Lee, C. F. Chin and W. H. Ang, Dalton Trans., 2019,
48, 7388.
K. S. Robinson, G. Clarkson, F. S. Mackay and P. J. Sadler,
Angew. Chem., Int. Ed., 2010, 49, 8905.
30 F. S. Mackay, N. J. Farrer, L. Salassa, H.-C. Tai, R. J. Deeth,
S. A. Moggach, P. A. Wood, S. Parsons and P. J. Sadler,
Dalton Trans., 2009, 2315.
9 A. Kastner, I. Poetsch, J. Mayr, J. V. Burda, A. Roller, 31 E. Shaili, L. Salassa, J. A. Woods, G. Clarkson, P. J. Sadler
P. Heffeter, B. K. Keppler and C. R. Kowol, Angew. Chem.,
Int. Ed., 2019, 58, 7464.
and N. J. Farrer, Chem. Sci., 2019, 10, 8610.
32 A. Krezel and W. Bal, J. Inorg. Biochem., 2004, 98, 161.
10 J. Z. Zhang, E. Wexselblatt, T. W. Hambley and D. Gibson, 33 M. Cossi, N. Rega, G. Scalmani and V. Barone, J. Comput.
Chem. Commun., 2012, 48, 847. Chem., 2003, 24, 669.
11 W. P. Fehlhammer and W. Beck, Z. Anorg. Allg. Chem., 34 B. S. Howerton, D. K. Heidary and E. C. Glazer, J. Am.
2013, 639, 1053–1082. Chem. Soc., 2012, 134, 8324.
12 H. Shi, C. Imberti and P. J. Sadler, Inorg. Chem. Front., 35 J. S. Butler, J. A. Woods, N. J. Farrer, M. E. Newton and
2019, 6, 1623. P. J. Sadler, J. Am. Chem. Soc., 2012, 134, 16508.
13 H.-C. Tai, Y. Zhao, N. J. Farrer, A. E. Anastasi, G. Clarkson, 36 C. Vallotto, E. Shaili, H. Shi, J. S. Butler, C. J. Wedge,
P. J. Sadler and R. J. Deeth, Chem. – Eur. J., 2012, 18, 10630.
14 H. Shi, C. Imberti, H. Huang, I. Hands-Portman and
P. J. Sadler, Chem. Commun., 2020, 56, 2320.
M. E. Newton and P. J. Sadler, Chem. Commun., 2018, 54,
13845.
37 M. Galanski and B. K. Keppler, Inorg. Chem., 1996, 35, 1709.
15 E. Shaili, M. Fernández-Giménez, S. Rodríguez-Astor, 38 W. H. Ang, S. Pilet, R. Scopelliti, F. Bussy, L. Juillerat-
A. Gandioso, L. Sandín, C. García-Vélez, A. Massaguer, Jeanneret and P. J. Dyson, J. Med. Chem., 2005, 48, 8060.
G. J. Clarkson, J. A. Woods, P. J. Sadler and V. Marchán, 39 J. J. Wilson and S. J. Lippard, Inorg. Chem., 2011, 50, 3103.
Chem. – Eur. J., 2015, 21, 18474.
16 S. Huayun, C. Imberti, G. J. Clarkson and P. J. Sadler, Inorg.
Chem. Front., 2020, 7, 3533.
40 B. M. Still, P. G. A. Kumar, J. R. Aldrich-Wright and
W. S. Price, Chem. Soc. Rev., 2007, 36, 665.
41 Y. Shi, S.-A. Liu, D. J. Kerwood, J. Goodisman and
J. C. Dabrowiak, J. Inorg. Biochem., 2012, 107, 6.
17 H. Shi, J. Kasparkova, C. Soulie, G. J. Clarkson, C. Imberti,
N. Olga, M. J. Paterson, V. Brabec and P. J. Sadler, Chem. – 42 F. Michelet, R. Gueguen, P. Leroy, M. Wellman, A. Nicolas
Eur. J., 2021, DOI: 10.1002/chem.202101168. and G. Siest, Clin. Chem., 1995, 41, 1509.
18 A. Gandioso, E. Shaili, A. Massaguer, G. Artigas, 43 L. Qin, H. Bartlett, H. R. Griffiths, F. Eperjesi,
A. González-Cantó, J. A. Woods, P. J. Sadler and
V. Marchán, Chem. Commun., 2015, 51, 9169.
R. A. Armstrong and D. Gherghel, Invest. Ophthalmol. Visual
Sci., 2011, 52, 5029.
19 J. J. Wilson and S. J. Lippard, Chem. Rev., 2014, 114, 44 M. P. Gamcsik, M. S. Kasibhatla, S. D. Teeter and
4470–4495.
O. M. Colvin, Biomarkers, 2012, 17, 671.
10606 | Dalton Trans., 2021, 50, 10593–10607
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
45 L. Salassa, T. Ruiu, C. Garino, A. M. Pizarro, F. Bardelli, 54 S. P. Oldfield, M. D. Hall and J. A. Platts, J. Med. Chem.,
D. Gianolio, A. Westendorf, P. J. Bednarski, C. Lamberti, 2007, 50, 5227.
R. Gobetto and P. J. Sadler, Organometallics, 2010, 29, 6703. 55 M. D. Hall, M. Okabe, D.-W. Shen, X.-J. Liang and
46 A. Vlček and S. Záliš, Coord. Chem. Rev., 2007, 251, 258.
47 M. Branch, Can. J. Chem., 1980, 58, 1592.
M. M. Gottesman, Annu. Rev. Pharmacol. Toxicol., 2008, 48,
495.
48 R. S. Partridge, S. M. Monroe, J. K. Parks, K. Johnson, 56 D. Sinani, D. J. Adle, H. Kim and J. Lee, J. Biol. Chem.,
J. Parker, W. Davis, G. R. Eaton and S. S. Eaton, Arch.
Biochem. Biophys., 1994, 310, 210.
49 C. Gambetta, J. Natera, W. A. Massad and N. A. García,
J. Photochem. Photobiol., A, 2013, 269, 27.
2007, 282, 26775.
57 S. Ishida, J. Lee, D. J. Thiele and I. Herskowitz, Proc. Natl.
Acad. Sci. U. S. A., 2002, 99, 14298.
58 S. S. More, S. Li, S. W. Yee, L. Chen, Z. Xu, D. M. Jablons
and K. M. Giacomini, Mol. Cancer Ther., 2010, 9, 1058.
50 M. J. Davies, Photochem. Photobiol. Sci., 2004, 3, 17.
51 A. F. Westendorf, A. Bodtke and P. J. Bednarski, Dalton 59 I. Romero-Canelón, A. M. Pizarro, A. Habtemariam and
Trans., 2011, 40, 5342. P. J. Sadler, Metallomics, 2012, 4, 1271.
52 J. J. Wilson and S. J. Lippard, Inorg. Chim. Acta, 2012, 389, 60 A. Radu, R. Conde, C. Fontolliet, G. Wagnieres, H. Van den
77. Bergh and P. Monnier, Gastrointest. Endosc., 2003, 57, 897.
53 E. J. New, R. Duan, J. Z. Zhang and T. W. Hambley, Dalton 61 S. Dhar, W. L. Daniel, D. A. Giljohann, C. A. Mirkin and
Trans., 2009, 3092.
S. J. Lippard, J. Am. Chem. Soc., 2009, 131, 14652.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 10593–10607 | 10607