A computational approach to tuning the photochemistry of platinum(IV) anticancer agents

Article abstract of DOI:10.1002/chem.201200782

Diazido Pt^IV complexes are inert stable prodrugs that can be photoactivated to produce Pt^II species with promising anticancer activity. Our studies of the photochemistry of Pt^IV complexes, [Pt(X)₂(Y)₂(Z)₂]^0/-1 (X=N-ligands (NH₃, pyridine, etc.)/S(CH₃)₂/H^-, Y=(pseudo)halogen (N₃^-, I^-), Z=OR^-, R=H, Ac) by time-dependent density functional theory (TDDFT) show close agreement with spectroscopic data. Broad exploration of cis/trans geometries, trans influences, the nature of the OR^- and (pseudo)halogen ligands, electron-withdrawing/donating/delocalising substituents on the N-ligands, and intramolecular H bonds shows that: 1) the design of platinum(IV) complexes with intense bands shifted towards longer wavelengths (from 289 to ～330 nm) can be achieved by introducing intramolecular H bonds involving the OH ligands and 2-hydroxyquinoline or by iodido ligands; 2) mesomeric electron-withdrawing substituents on pyridine result in low-energy absorption with significant intensity in the visible region; and 3) the distinct makeup of the molecular orbitals involved in the electronic transitions for cis/trans-{Pt(N ₃)₂} isomers results in different photoproducts. In general, the comparison of the optimised geometries shows that Pt^IV complexes with longer Pt-L bonds are more likely to undergo photoreduction with longer-wavelength light. The novel complex trans,trans,trans-[Pt(N ₃)₂(OH)₂(NH₃)(4-nitropyridine)] with predicted absorption in the visible region has been synthesised. The experimental UV/Vis spectrum in aqueous solution correlates well with the intense band in the computed spectrum, whereas the overlay in the low-energy region can be improved by a solvent model. This combined computational and experimental study shows that TDDFT can be used to tune the coordination environment for optimising photoactive Pt^IV compounds as anticancer agents. Computer-aided design: Platinum(IV) complexes are inert stable prodrugs that can be photoactivated to produce species with promising anticancer activity. Here a combined computational and experimental study was carried out to show that TDDFT can aid the design of the coordination environment of Pt ^IV complexes to allow optimisation of the photoactivity of these Pt^IV anticancer agents (see figure). Copyright

Full text of DOI:10.1002/chem.201200782

FULL PAPER
DOI: 10.1002/chem.201200782
A Computational Approach to Tuning the Photochemistry of Platinum(IV)
Anticancer Agents
Hui-Chung Tai, Yao Zhao, Nicola J. Farrer, Anna E. Anastasi, Guy Clarkson,
[
a]
Peter J. Sadler,* and Robert J. Deeth*
IV
Abstract: Diazido Pt complexes are
shifted towards longer wavelengths
(from 289 to ~330 nm) can be achieved
by introducing intramolecular H bonds
involving the OH ligands and 2-hy-
droxyquinoline or by iodido ligands;
eral, the comparison of the optimised
geometries shows that Pt complexes
with longer PtꢀL bonds are more likely
to undergo photoreduction with longer-
wavelength light. The novel complex
IV
inert stable prodrugs that can be pho-
toactivated to produce Pt species with
promising anticancer activity. Our stud-
ies of the photochemistry of Pt com-
plexes, [Pt(X) (Y) (Z) ]
gands (NH , pyridine, etc.)/S
Y=(pseudo)halogen (N , I ), Z=
OR , R=H, Ac) by time-dependent
density functional theory (TDDFT)
show close agreement with spectro-
scopic data. Broad exploration of cis/
trans geometries, trans influences, the
II
IV
0
/ꢀ1
(X=N-li-
2) mesomeric
electron-withdrawing
trans,trans,trans-[Pt(N ) (OH)2 ACHTUTGNENRNU(G NH )(4-
3 2 3
2
2
2
ꢀ
A
H
U
G
E
N
N
(CH ) /H ,
substituents on pyridine result in low-
energy absorption with significant in-
tensity in the visible region; and 3) the
distinct makeup of the molecular orbi-
tals involved in the electronic transi-
tions for cis/trans-{Pt(N ) } isomers re-
nitropyridine)] with predicted absorp-
tion in the visible region has been syn-
thesised. The experimental UV/Vis
spectrum in aqueous solution correlates
well with the intense band in the com-
puted spectrum, whereas the overlay in
the low-energy region can be improved
by a solvent model. This combined
computational and experimental study
shows that TDDFT can be used to
tune the coordination environment for
3
3 2
ꢀ
ꢀ
3
ꢀ
3
2
sults in different photoproducts. In gen-
ꢀ
nature of the OR and (pseudo)halo-
gen ligands, electron-withdrawing/do-
nating/delocalising substituents on the
N-ligands, and intramolecular H bonds
shows that: 1) the design of platinu-
m(IV) complexes with intense bands
Keywords: antitumor agents · bio-
inorganic chemistry · computational
chemistry
platinum
·
photochemistry
·
IV
optimising photoactive Pt compounds
as anticancer agents.
[5]
Introduction
apoptosis contributing to tumour resistance to Pt drugs.
Thus, the design of drugs with increased selectivity is of
great importance for significant clinical advantages over cur-
rent drugs.
Since the serendipitous discovery of the inhibition of cell di-
vision by cisplatin in the 1960s, platinum-based drugs have
[1]
been widely used against various tumours. Much effort has
been devoted to the design of novel Pt anticancer drugs to
make chemotherapy safer for patients, in particular, lessen-
Prodrugs to improve therapeutic targeting and controlled
release at the tumour site show promising possibilities for
[6]
increased selectivity. Improved delivery of platinum drugs
with liposomal-based or polymer-based vehicles is an attrac-
[2]
[3]
ing severe side effects, increasing oral bioavailability, and
[4]
[7]
overcoming drug resistance. Platinum-based drugs disrupt
DNA synthesis by mechanisms common to all actively divid-
ing cells, regardless of whether they are cancerous or not,
which generates severe side effects. In addition, the large
volume of damaged cells further causes altered cellular
transport, enhanced repair of distorted DNA, over-expres-
sion of DNA-damage recognition proteins, and decreased
tive strategy and has led to clinical trials. However, creat-
ing delivery systems that target specific sites to achieve
good release and activation is challenging. An additional ap-
proach involves the use of a photosensitising agent in com-
bination with light, which provides a promising avenue to
[8]
achieve accurate targeting.
An approach to platinum-based photochemotherapy of
cancer is to design photoactive Pt complexes that have
a strong absorbance, increased selectivity, and facile metabo-
[9]
II
lism. Compared to Pt species, platinum(IV) complexes
are much more inert to reaction in the ground electronic
state but can readily undergo excited-state photoinduced re-
duction and ligand substitution. Hence, photoexcitation can
be used to trigger specific interactions between the metal
complex and target biomolecules (such as DNA, RNA or
[
a] H.-C. Tai, Y. Zhao, Dr. N. J. Farrer, Dr. A. E. Anastasi,
Dr. G. Clarkson, Prof. P. J. Sadler, Prof. R. J. Deeth
Department of Chemistry, University of Warwick
Gibbet Hill Road, Coventry CV4 7AL (UK)
Fax : (+44)24-7652-3819
E-mail: p.j.sadler@warwick.ac.uk
r.j.deeth@warwick.ac.uk
[10]
proteins). Altering the electronic structure by irradiation
can lead to remarkable differences in chemistry, including
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201200782.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

significant changes in geometry (coordination bond lengths
and angles), dipole moment, redox characteristics and acid-
base properties, and can bring about reactions that normally
focus is on making a systematic investigation of various
chemical influences on the photochemistry of Pt com-
plexes with respect to both the intense band and low-energy
absorptions in order to provide guidelines for future synthe-
sis of this class of compounds.
IV
[11]
do not occur under thermal conditions. Exploration of the
IV
factors that control the photochemistry of Pt complexes
may thus aid the design of effective, new anticancer agents.
Our aim is to design complexes that can be activated with
a variety of wavelengths of light from UVA across the visi-
ble region, since longer wavelengths help to reduce cell
Results and Discussion
[12]
damage and deeper penetration can be achieved. trans,-
trans,trans-[Pt(N ) (OH) (NH )(py)] (1) is highly phototoxic
Influence of chemical modifications: The electronic excita-
tions of Pt complexes were evaluated by the influence of
IV
A
H
U
G
R
N
U
G
3
2
2
3
to cancer cells and, in particular, the pyridine ligand appears
to contribute to the potency giving rise to a mechanism of
various chemical modifications with reference to complex
1 (Figure 1). Calculated excitation energies for the intense
band from TDDFT and theoretical bands generated by the
summation of Gaussian peaks of the TDDFT wavelengths in
simulated spectra (see below) are in good agreement with
experimental measurements with errors less than 0.3 eV.
Electronic trans influences are a well-understood ground-
[10f]
action different from cisplatin.
Therefore, complex 1 has
been used as the starting point and the template for our the-
oretical modelling.
We have studied the effect of the coordination environ-
IV
ment on the electronic excitations of Pt complexes by
ꢀ
[14]
tuning cis/trans geometries, trans influences, the OR (R=
state phenomena, but relatively little is known about their
ꢀ
ꢀ
ꢀ
H, Ac) ligands, (pseudo)halogen (N , I ) ligands, electron-
effects on photophysical properties. OR and halogen li-
3
[15]
withdrawing/donating/delocalising substituents on pyridine
ligands, and intramolecular H bond effects (Figure 1). We
have also evaluated the influence of various chemical modi-
fications on: 1) the electronic excitations of the intense ab-
sorption band (normally found at shorter wavelengths) and
of low-energy absorption bands (the “tail” normally seen at
longer wavelengths), 2) the molecular orbitals involved in
the electronic transitions in the UV/Vis bands, and 3) the
nature of bonding. In addition, trans,trans,trans-
gands have a large effect on the rate of reduction,
and
thus it is interesting to study their influence on the electron-
IV
[9b,15–16]
ic transitions in Pt complexes. As earlier reports
IV
ꢀ
showed that Pt complexes with OAc or halogen ligands
are often easily reduced in cells and are less likely to exhibit
a marked difference between light and dark toxicity, in this
ꢀ
ꢀ
study the design with trans OH and trans N groups is pre-
3
ferred. Pyridine, with its range of possible substituents, is
a useful ligand for structure–property studies, including the
IV
[
Pt(N ) (OH)
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(NH )(4-nitropyridine)] has been synthesised
photochemistry of Pt compounds. These chemical proper-
3
2
2
3
and characterised in order to compare its absorption proper-
ties with the calculations. A theoretical study of the ground
ties led us to the current construction of the theoretical
models in this work.
IV
and excited state properties of photoactive Pt diazido was
Figure 2 summarises the key features of the computed in-
tense and lowest energy absorptions under the influence of
the various chemical modifications given above. The
maxima of the intense bands in Figure 2 were produced by
summing over Gaussian peaks as in the simulated spectra
and thus are slightly different compared to Tables 1 and 2.
[
13]
reported while this manuscript was in preparation.
Our
ꢀ
ꢀ
results on the replacement of OH ligand by aromatic OR
and the formation of azidyl radicals N C upon photoreduc-
3
tion largely agree with those reported in that work, but our
IV
Figure 3 displays simulated UV spectra of selected Pt com-
plexes. Note that for complexes with two intense bands, we
focus the comparison on the bands at longer wavelength.
Comparison between experimental and calculated structures
and absorption spectra: The fully optimised structures of
cis,trans,cis-[Pt(N ) (OH)2 ACHTUNGRTNEUNG( NH ) ], and its all-trans isomer
3 2
3
2
show that bond lengths and angles are in good agreement
with the respective experimental values, with a root-mean-
square deviation (RMSD) of 0.046 ꢁ for PtꢀL bonds and
0
.98 for LꢀPtꢀL angles (Table S1 in the Supporting Informa-
[10g,17]
tion).
The computed absorption maximum for the com-
(NH )(py)] (2; py=pyridine;
plex cis,trans,cis-[Pt(N ) (OH) ACHTNUGTRENNUNG
Figure 1. Some possible chemical modifications for exploring the photo-
chemistry of the photocytotoxic platinum complex, trans,trans,trans-
3
2
2
3
273 nm, Table 2) is 20 nm shorter than that of its all-trans
isomer complex 1 (293 nm, Table 1). The same trend is ob-
[
3 2 2 3
Pt(N ) (OH) ACHTUNGTRENNUNG( NH ) ACHTUNGTRENNUNG( pyridine)] (1). Changes in the coordination geome-
try and trans influence are considered. Substitutions (R) on the imine li-
gands, OR and (pseudo)halogen ligands are also included in the calcula-
tions. The substitutions on the imine ligands vary from withdrawing/do-
nating groups/quinoline to intramolecular H bonding involving the OH
ligand.
served for the cis,trans,cis-[Pt(N ) (OH) ACHTUNGTRENUNG( NH ) ] (271 nm,
2 3 2
3
2
Table 1) and trans,trans,trans-[Pt(N ) (OH) ACHTUNGTENRUNG( NH ) ] (295 nm,
2 3 2
3
2
Table 1), which is consistent with the previously reported ex-
[10g]
perimental data, 256 and 285 nm, respectively (Table 1).
&
2
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Photoactive Platinum Prodrugs
FULL PAPER
The trans influence: To investi-
gate the trans influence on pho-
IV
tophysical properties in Pt
complexes,
ligands
with
a strong trans influence were
ꢀ
placed trans to the N3 ligand,
which can behave as the major
reducing/leaving
ligand.
TDDFT predicts intense bands
at 293 and 308 nm for Me S and
2
ꢀ
H ligands trans to azide in cis,-
trans,cis-[Pt(N ) (OH) CHTUNGTNREUNG( NH )S-
2A
3
2
3
AHCTUNGTERG(NNUN CH ) ] (3) and cis,trans,cis-
3
2
ꢀ
[
PtH(N3)
A
H
U
T
E
N
N
(OAc)
A
H
U
G
R
N
U
G
(4),
2
2
3
respectively, significantly longer
wavelengths than in complex 2
(
274 nm). Comparison of the
main absorbance (lmax for
)
complexes 2, 3 and 4 shows that
the trans influence has a consid-
erable effect on the electronic
IV
excitation energies in Pt com-
pounds (see Figures 2 and 3 for
IV
the comparison with other Pt
complexes).
ꢀ
OR
ACHTUNGTRENNUNG( R=H, Ac) and halogen
ligands: To explore the influ-
ꢀ
ence of OR and halogen li-
gands, the electronic transitions
ꢀ
ꢀ
in complexes with OH , OAc ,
ꢀ
and I ligands were studied.
TDDFT assigns an absorption
maximum (lmax) at longer wave-
length, 318 nm for trans,tran-
s,trans-[Pt(N3)2
ACHTUNGTNRENUNG( OAc) -
2
AHCTUNGERTG(NNUN NH )(py)] (5) and 314 nm for
3
trans,trans,trans-[Pt(N ) I -
3
2 2
AHCTUNGTRNENUG( NH )(py)] (6), compared to
3
complex 1 (297 nm). The exci-
tation energies decrease in the
ꢀ
ꢀ
ꢀ
order OH >OAc ~I , which,
interestingly, reflects the order
of the standard hydrogen elec-
trode reduction potentials for
Figure 2. Calculated wavelength of the intense band and lowest energy absorption for platinum complexes
with various chemical modifications. The wavelengths of the intense bands were produced by summing over
Gaussian peaks as in the simulated spectra, resulting in only a slight difference compared to the values in
Tables 1 and 2. The bands with close energies can be merged into one. There are two intense bands for com-
plexes 4, 5, 7, and 16 given in order of the intensity (from strong to low).
cis,trans,cis-[Pt ACHTGNUTRENN(NUG NH ) (X) (Cl) ]:
3 2 2 2
ꢀ
ꢀ
ꢀ [15–16]
OH >OAc >Cl .
The hy-
droxide complexes thus appear
Compared to the latter two compounds, the intense bands
lmax) for complex 1 and trans,trans,trans-[Pt(N ) (OH) -
ACHTUNGTRENNUNG( NH ) ACHTUNGTRENNUNG( quin)] (13; quin=quinoline) reveal a smaller discrep-
3
to have the highest energy unoccupied frontier orbitals in
excitation, but the lowest energy LUMO for reduction. The
excitation energies follow the same trend as the energy dif-
ference between the products and reactants in reduction for
(
3
2
2
ancy between TDDFT and experimental values. The high
electron density of pyridine and quinoline may increase the
IV
ꢀ
Pt complexes with OR and halogen ligands. The replace-
[18]
ꢀ
ꢀ
accuracy of the modelled frontier orbitals. In addition to
the electronic excitation energies, the computed intensity of
absorption is consistent with the UV/Vis spectrum (Fig-
ment of the N3 ligands in complex 5 by I ligands (complex
7 in Figure 2) results in a red-shift of the intense band at
333 nm. Meanwhile, the “tail” of the absorbance extends
from 506 nm in complex 1 to 744 nm in complex 6 and to
[10g,17]
ure 4).
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

P. J. Sadler, R. J. Deeth et al.
7
32 nm in complex 7. The trend in the influence on the si-
sorption at longer wavelengths can be better achieved by
the latter approach. This concept leads to the exploration of
Pt complexes with an intramolecular hydrogen bond be-
mulated spectra of the studied complexes is displayed in
Figure 3.
IV
tween 2-hydroxypyridine/2-hydroxyquinoline and the hy-
droxyl group. In general, electron-withdrawing functional
groups on pyridine red-shift not only the absorption maxima
by 10 nm but also the lowest energy absorption by 38–
332 nm, whereas intramolecular hydrogen bonding involving
the OH ligands and 2-hydroxyquinoline red-shifts the main
absorption by 17–32 nm and extends the lowest energy ab-
sorption toward the red region by 40–100 nm.
Substituents on pyridine, quinoline, and intramolecular hy-
drogen bonding: To elucidate the relationship between elec-
tronic effects on the pyridine ring and photophysical proper-
IV
ties of Pt complexes, electron-withdrawing (via inductive
or resonance effects), electron-donating and electron-deloc-
ꢀ
alising substituents on pyridine were considered. OR and
halogen ligands show a critical influence on electronic tran-
sitions. However, the relatively higher energies of the unoc-
The calculated wavelengths of the intense bands for com-
IV
ꢀ
cupied frontier orbitals in Pt complexes with OAc or hal-
ogen ligands often leads to rapid reduction in cells. The
lower excitation energies can be achieved by either increas-
ing the energies of the occupied frontier orbitals or decreas-
ing the energies of the unoccupied frontier orbitals, and
thus, the design of photoactivated Pt drugs with greater ab-
plexes
dine)] (8), trans,trans,trans-[Pt(N ) (OH) AHCTUNGTRNEUNG( NH )(2,6-difluoro-
2 3
trans,trans,trans-[Pt(N ) (OH) AHCTUGNERNTUNG( NH )(2-cyanopyri-
3 2 2 3
3
2
pyridine)] (9), and trans,trans,trans-[Pt(N ) (OH) ACHTUNGTRENNUNG( NH )(2-
2 3
3
2
nitropyridine)] (10) are 305, 305 and 307 nm, respectively,
which reveal that electron-withdrawing by induction and
resonance have only a subtle influence on the electronic ex-
3 2 2 3 2 3 2 2 3 2
Table 1. Experimental and calculated absorption properties (intense band) of trans,trans,trans-[Pt(N ) (OH) ACHUTGNTRENUN(NG NH ) ], cis,trans,cis-[Pt(N ) (OH) ACHTUNGTRENNUNG( NH ) ],
trans,trans,trans-[Pt(N
3
)
2
(OH)
2
A
T
N
T
E
N
G
3
)(py)] (1), and trans,trans,trans-[Pt(N
3
)
2
(OH)
Energy [eV]
wavelength [nm])
2 3
ACHTUNGTREUNNN(G NH ) ACHUTNGTRNEGUN( quin)] (13).
ꢀ
1
ꢀ1
l
max [nm] (e [m cm ])
Composition
Oscillator
strength
Assignment
(
trans,trans,trans-[Pt(N
3
)
2
(OH)
2
A
H
N
T
E
N
N
3 2
) ]
HOMO3!LUMO (46%)
HOMO6!LUMO (24%)
HOMO3!LUMO+1 (19%)
HOMO!LUMO+3 (67%)
1
1
4.21 (295)
5.24 (237)
0.341
0.192
LMCT/LLCT
LLCT
2
85 (18900)
1
]
9
3 2 2 3 2
cis,trans,cis-[Pt(N ) (OH) ACHNUTTERGNNUNG( NH )
HOMO5!LUMO+1 (64%)
HOMO4!LUMO (20%)
HOMO4!LUMO+1 (39%)
HOMO5!LUMO (33%)
1
3
4
4.47 (277)
4.58 (271)
0.134
0.156
LMCT/LLCT
LMCT/LLCT
2
56 (13700)
1
trans,trans,trans-[Pt(N
3
)
2
(OH)
2
A
H
U
G
R
N
U
G
3
)(py)] (1)
HOMO6!LUMO (48%)
1
7
1
HOMO3!LUMO (29%)
HOMO3!LUMO+2 (14%)
HOMO6!LUMO (46%)
HOMO3!LUMO (37%)
)(4-nitropyridine)] (11)
3.97 (312)
4.23 (293)
0.119
0.261
LMCT/LLCT
LMCT/LLCT
2
89 (18800)
2
trans,trans,trans-[Pt(N
87 (18200)
trans,trans,trans-[Pt(N
3
3
)
)
2
(OH)
(OH)
2
2
A
H
U
T
E
N
N
(NH
3
2
9
HOMO-10!LUMO (68%)
HOMO-3!LUMO+1 (59%)
HOMO-5!LUMO+1 (13%)
3.99 (311)
4.08 (304)
0.147
0.311
LLCT/MLCT
LLCT/LMCT
2
30
2
A
H
U
G
R
N
N
(NH
3
)
A
H
N
T
E
N
G
HOMO5!LUMO+2 (31%)
HOMO4!LUMO+2 (17%)
HOMO4!LUMO (15%)
HOMO8!LUMO+3 (17%)
HOMO11!LUMO (14%)
3
16 (17800) and 294 (28300)
33 (42600)
20
45
3.98 (311)
5.27 (235)
0.120
0.106
LMCT/LLCT
2
LMCT/LLCT/MLCT/d-d
[
a] Tr=transition number as obtained in the TDDFT calculation.
IV
Table 2. Calculated absorption properties (intense band) of Pt complexes.
[
a]
Tr
Composition Energy [eV] (wavelength [nm])
Oscillator strength
Assignment
cis,trans,cis-[Pt(N
3
)
2
(OH) ACHTUNRTGENNU(GN NH ) AHCTUNGTREG(UNNN pyridine)] (2)
2 3
HOMO-5!LUMO+2 (47%)
HOMO-4!LUMO (18%)
HOMO-4!LUMO+2 (35%)
HOMO-5!LUMO (25%)
HOMO-5!LUMO+2 (10%)
2
6
7
4.50 (276)
0.133
0.205
LMCT/LLCT
LMCT/LLCT
2
4.54 ACHUTNRGNENUG( 273)
3 2 2 3 3 2
cis,trans,cis-[Pt(N ) (OH) ACTHUNGNRTEUNG( NH )S ACTHNGUTRENNGU( CH ) ] (3)
&
4
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Photoactive Platinum Prodrugs
Table 2. (Continued)
FULL PAPER
[
a]
Tr
Composition
Energy [eV] (wavelength [nm])
4.20 (295)
Oscillator strength
0.255
Assignment
HOMO-4!LUMO+1 (29%)
HOMO-4!LUMO (28%)
HOMO-3!LUMO+1 (11%)
HOMO-5!LUMO+1 (61%)
1
2
LMCT/LLCT
LMCT/LLCT
LMCT/LLCT
LMCT/LLCT
1
3
4.48 (277)
4.05 (306)
4.50 (276)
0.114
0.103
0.113
HOMO-5!LUMO (19%)
ꢀ
cis,trans,cis-[PtH(N
3
3
)
2
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(OAc)
2
A
H
U
G
E
N
N
(NH
3
)] (4)
1
HOMO-2!LUMO+1 (72%)
HOMO-8!LUMO (45%)
HOMO-8!LUMO+1 (17%)
HOMO-5!LUMO+1 (14%)
HOMO-6!LUMO+1 (13%)
1
7
3 2 2 3
trans,trans,trans-[Pt(N ) ACHTUNGTRENNUNG( OAc) AHCNUTTGRENNNUG( NH ) ACHNUTTGERNUNN(G pyridine)] (5)
HOMO-4!LUMO (31%)
2
4
HOMO-4!LUMO+1 (28%)
HOMO-5!LUMO+2 (13%)
HOMO-7!LUMO (22%)
3.89 (318)
4.27 (291)
0.243
0.148
LMCT/LLCT
LMCT/LLCT
HOMO-7!LUMO+1 (17%)
HOMO-8!LUMO+1 (17%)
3 2 2 3
trans,trans,trans-[Pt(N ) I CATHUNGTRENUNN(G NH ) AHCUTNGRUENNN(G pyridine)] (6)
HOMO-6!LUMO+1 (33%)
HOMO-7!LUMO (21%)
HOMO-10!LUMO (18%)
HOMO-6!LUMO+1 (27%)
HOMO-7!LUMO+1 (19%)
HOMO-10!LUMO (16%)
HOMO-10!LUMO (48%)
3
1
3.87 (320)
0.221
LMCT/LLCT/MLCT/d-d
3
2
5
3.93 (315)
4.24 (293)
0.292
0.199
LMCT/LLCT/MLCT/d-d
LMCT/LLCT/MLCT/d-d
3
HOMO7!LUMO (11%)
I
2
trans-trans-trans
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
[Pt
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(OAc)
2
A
T
N
T
E
N
N
(NH
3
)
A
H
N
R
N
U
(pyridine)] (7)
HOMO-4!LUMO (31%)
HOMO-9!LUMO (27%)
HOMO-10!LUMO (22%)
HOMO-10!LUMO (43%)
HOMO-4!LUMO (21%)
HOMO-11!LUMO (13%)
2
3
5
4
3.73 (333)
4.28 (290)
0.197
0.349
LMCT/LLCT
LMCT/LLCT/MLCT/d-d
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHTNGRENNUNG( NH )(2-cyanopyridine)] (8)
HOMO-3!LUMO+1 (69%)
2
2
4.05 (306)
0.333
0.344
0.229
LMCT/LLCT
HOMO-3!LUMO+3 (12%)
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHGNTNERNUNG( NH )(2,6-difluoropyridine)] (9)
HOMO-3!LUMO (43%)
HOMO-3!LUMO+2 (33%)
HOMO-6!LUMO (15%)
2
3
4.07 (305)
LMCT/LLCT
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHGTNRENNUNG( NH )(2-nitropyridine)] (10)
HOMO-3!LUMO+1 (37%)
2
8
4.03 (307)
LMCT/LLCT/MLCT
HOMO-5!LUMO+2 (23%)
trans,trans,trans-[Pt(N
3
)
2
(OH)
2
A
H
U
G
R
N
U
G
3
)(4-methoxypyridine)] (12)
HOMO-3!LUMO (46%)
HOMO-7!LUMO (34%)
1
2
4
7
0
2
4.03 (308)
0.184
0.123
0.168
LMCT/LLCT
HOMO-3!LUMO+3 (64%)
HOMO-3!LUMO (19%)
HOMO-7!LUMO (12%)
HOMO-5!LUMO+2 (51%)
HOMO-9!LUMO+1 (21%)
4.26 (291)
LMCT/LLCT
5.39 (230)
LMCT/LLCT/MLCT/d-d
trans,trans,trans-[Pt(N
3
)
2
(OH)
2
A
H
U
T
E
N
N
(NH
3
)(2-hydroxypyridine)] (14)
4.00 (310)
HOMO-3!LUMO (41%)
HOMO-4!LUMO (13%)
1
3
0.253
LMCT/LLCT/MLCT/d-d
trans,trans,trans-[Pt(N
3
)
2
(OH)
2
A
H
U
G
R
N
U
G
3
)(2-hydroxyquinoline)] (15)
HOMO-4!LUMO (43%)
HOMO-5!LUMO (25%)
1
7
3.91 (317)
0.259
0.138
LMCT/LLCT/MLCT/d-d
LLCT
HOMO-6!LUMO+1 (15%)
HOMO-5!LUMO+3 (18%)
HOMO-1!LUMO+3 (15%)
3
8
5.06 (245)
trans,trans,trans-[Pt(N
3
)
2
(OH)
2
(2-hydroxyquinoline)
2
] (16)
HOMO-7!LUMO+1 (35%)
2
9
3.81 (325)
0.159
LMCT/LLCT/MLCT/d-d
HOMO-4!LUMO+1 (21%)
[
a] Tr=transition number as obtained in the TDDFT calculation.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

P. J. Sadler, R. J. Deeth et al.
a similar extent as electron-withdrawing/donating groups.
Moreover, the calculated intense bands of trans,trans,trans-
[
Pt(N ) (OH) ACHTUNTGRNNEG(U NH )(2-hydroxyquinoline)] (15) and of trans,-
3 2 2 3
trans,trans-[Pt(N ) (OH) (2-hydroxyquinoline) ] (16), 314
3
2
2
2
and 329 nm, respectively, are red-shifted compared to com-
plexes 8, 9, 10, 12 and 13, which shows that intramolecular
hydrogen bonding involving OH ligands can have a marked
effect. On the other hand, it is interesting to note that the
low-energy, low-intensity spectroscopic absorption extends
from the green region of the spectrum for complex
1
(506 nm) to the red region for complexes 16 (606 nm), 8
(
625 nm), and 10 (858 nm), yet is blue shifted by 30 nm for
IV
complex 12. The trend of electronic transitions for the Pt
complexes is shown in Figures 2 and 3. The data for trans,-
trans,trans-[Pt(N ) (OH) ACHTUNGTRENUNG( NH )(4-nitropyridine)] (11) will
2 3
3
2
be analysed and discussed later.
Orbital analysis: In addition to exploring the energies of
electronic transitions upon different chemical modifications,
it is highly desirable to understand qualitatively how the
IV
structural changes in Pt complexes relate to spectroscopic
absorption. By identifying and examining the most impor-
tant orbitals that control the electronic excitation, this corre-
lation and thus mechanisms of photoactivation can be eluci-
dated. The results for electronic transitions of the intense
band and electronic transitions of low-energy absorption are
summarised in Tables 2 and 3, respectively. The major orbi-
tal contributions to the main absorption are shown in
Figure 5.
IV
Figure 3. Simulated absorption spectra of selected Pt complexes 1, 2, 3,
, 7, and 16. The oscillator strength has been scaled by 2.5-times for the
5
convenience of comparing with experimental absorption of complex 1.
Electronic transitions of the intense band: The key transi-
tions are a mixture of ligand-to-metal charge transfer
ꢀ
(
LMCT), where electron density migrates from the OR
and (pseudo)halogen ligands to the metal, and ligand-to-
ligand charge transfer (LLCT), where electrons move be-
ꢀ
tween OR and (pseudo)halogen ligands. In comparison
with complex 1 (297 nm), the cis isomer, complex 2
(
274 nm), has a lower nonbonding orbital of N character re-
3
sulting in higher energies for the transitions. The trans influ-
ence tends to increase the probabilities of electronic transi-
tion from N trans to Me S and slightly lower the energies of
3
2
Figure 4. Comparison between calculated and experimental UV/Vis spec-
the unoccupied orbitals compared to complex 2.
tra for trans,trans,trans-[Pt(N
3 2 2 3
) (OH) ACHTUNGTRNENU(G NH ) ACHTNUGTRUNNNE(G pyridine)] (1) and trans,tran-
The profile of electronic transitions in Figure 5 reveals
that electrons in cis isomers tend to move from orbitals of
the (pseudo)halogen ligands to the orbitals of the metal,
whereas those in all-trans isomers are more likely to transfer
s,trans-[Pt(N (OH) (NH )(quinoline)] (13). The oscillator strength of
3
)
2
2
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
3
complexes 1 and 13 has been scaled by 2.5- and 1.8-times, respectively,
for the convenience of comparison.
ꢀ
from both OR and (pseudo)halogen ligands to the d orbi-
IV
citation energies of Pt compounds. Surprisingly, trans,tran-
tals of the metal and thus lead to different photoproducts.
s,trans-[Pt(N ) (OH)
A
H
U
G
R
N
U
G
The NMR spectra for the photolysis of trans,trans,trans-
3
2
2
3
a maximum at 301 nm, similar to complexes with electron-
withdrawing groups on pyridine. Comparison of the predict-
ed bands of complexes 8, 9, 10, and 12 indicates that elec-
tron-withdrawing/donating substituents on pyridine slightly
extend the main absorption to longer wavelength. In com-
plex 13, the band centred at 305 nm suggests that stabilisa-
tion of the unoccupied frontier orbitals by electron delocali-
sation decreases the energies of electronic transitions to
[Pt(N ) (OH)
reported previously correlate with the calculations.
A
H
U
G
R
N
N
(NH ) ] and cis,trans,cis-[Pt(N ) (OH)
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(NH ) ]
3
2
2
3
2
3
2
2
[
3
2
10a,g,i]
It
will be necessary to study a series of cis-trans isomers and
the makeup of the important molecular orbitals in future
work for a complete understanding of the mechanisms that
give rise to different photopathways and photoproducts.
Complexes 8–12 show lower energy of absorption com-
ꢀ
ꢀ
IV
pared to complex 1 (297 nm). OAc and I ligands in Pt
&
6
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Photoactive Platinum Prodrugs
FULL PAPER
IV
Table 3. Calculated low-energy absorption properties of Pt complexes.
[
a]
Tr
Composition
Energy [eV] (wavelength [nm])
Oscillator strength
Assignment
trans,trans,trans-[Pt(N
3
)
2
(OH) ACHTUNRTGENNU(GN NH ) AHCTUNGTRENNUNG( pyridine)] (1)
2 3
1
5
HOMO!LUMO (100%)
HOMO-2!LUMO (71%)
HOMO-1!LUMO+1 (28%)
2.45 (506)
2.93 (423)
0.0
LMCT
0.007
LMCT/LLCT/MLCT/d-d
3 2 2 3
cis,trans,cis-[Pt(N ) (OH) ACHTUNGTERNNU(GN NH ) AHCUTNGRUENNN(G pyridine)] (2)
1
2
HOMO!LUMO (96%)
2.52 (492)
2.64 (469)
0.0
0.007
LMCT
LLCT
HOMO!LUMO+1 (96%)
3 2 2 3 3 2
cis,trans,cis-[Pt(N ) (OH) ACTHUNGNRTEUNG( NH )S ACTHNGUTRENNGU( CH ) ] (3)
1
HOMO!LUMO (100%)
2.37 (523)
2.65 (468)
0.0
LMCT/LLCT
LMCT/LLCT
HOMO-1!LUMO (50%)
2
0.003
HOMO-2!LUMO (48%)
ꢀ
cis,trans,cis-[PtH(N
3
)
2
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(OAc)
2
A
H
U
G
E
N
N
(NH
3
)] (4)
1
HOMO!LUMO (100%)
HOMO-3!LUMO (80%)
2.26 (549)
3.04 (408)
0.0
LMCT/LLCT
LMCT/LLCT
4
0.043
HOMO!LUMO+1 (13%)
3 2 2 3
trans,trans,trans-[Pt(N ) ACHTUNGTRENNUNG( OAc) AHCNUTTGRENNNUG( NH ) ACHNUTTGERNUNN(G pyridine)] (5)
1
HOMO!LUMO (96%)
2.17 (572)
0.001
0.052
LMCT
LMCT
HOMO-2!LUMO+1 (39%)
HOMO-1!LUMO+1 (28%)
HOMO-1!LUMO (18%)
7
2.79 (445)
3 2 2 3
trans,trans,trans-[Pt(N ) I CATHUNGTRENUNN(G NH ) AHCUTNGRUENNN(G pyridine)] (6)
1
HOMO!LUMO (98%)
1.67 (744)
2.20 (564)
0.0
LMCT/LLCT
LMCT/LLCT
HOMO-5!LUMO (54%)
7
0.007
HOMO-1!LUMO+1 (37%)
I
2
trans-trans-trans
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
[Pt
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(OAc)
2
A
H
U
G
R
N
N
(NH
3
)
A
H
U
G
E
N
N
(pyridine)] (7)
1
8
HOMO!LUMO (99%)
1.69 (732)
2.46 (504)
0.0
0.002
LMCT/LLCT
LLCT
HOMO-1!LUMO+2 (99%)
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHTNGRENNUNG( NH )(2-cyanopyridine)] (8)
1
4
HOMO!LUMO (100%)
1.98 (625)
2.46 (504)
0.0
0.003
LLCT
LLCT
HOMO-2!LUMO (98%)
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHGNTNERNUNG( NH )(2,6-difluoropyridine)] (9)
1
4
HOMO!LUMO (99%)
2.28 (544)
2.45 (506)
0.001
0.005
LMCT
MLCT/LLCT
HOMO-1!LUMO+1 (98%)
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHGTNRENNUNG( NH )(2-nitropyridine)] (10)
1
2
3
HOMO!LUMO (99%)
HOMO-1!LUMO (97%)
HOMO-2!LUMO (96%)
1.45 (858)
1.75 (709)
1.94 (639)
0.002
0.002
0.014
LLCT
LLCT/MLCT
LLCT
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHGTNRENNUNG( NH )(4-nitropyridine)] (11)
1
4
1
HOMO!LUMO (100%)
HOMO-3!LUMO (100%)
HOMO-6!LUMO (99%)
1.12 (1111)
1.87 (664)
2.88 (430)
0.0
0.006
0.010
LLCT
LLCT
LLCT/MLCT
0
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHGNTRNENUNG( NH )(4-methoxypyridine)] (12)
1
5
HOMO!LUMO (98%)
2.60 (476)
3.32 (374)
0.0
0.001
LMCT/LLCT
LLCT
HOMO!LUMO+2 (95%)
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACHTUGNTRENNUNG( NH )(quinoline)] (13)
1
6
HOMO!LUMO (97%)
2.37 (524)
2.90 (428)
0.0
0.015
LMCT
LLCT
HOMO-2!LUMO+1 (84%)
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHGNTRNENUNG( NH )(2-hydroxypyridine)] (14)
1
HOMO!LUMO (99%)
HOMO-2!LUMO+1 (68%)
HOMO-4!LUMO (20%)
2.47 (502)
0.0
LMCT/LLCT
7
2.88 (431)
0.001
LMCT/LLCT/MLCT/d-d
3 2 2 3
trans,trans,trans-[Pt(N ) (OH) ACUTHGNTRNENUNG( NH )(2-hydroxyquinoline)] (15)
1
7
HOMO!LUMO (88%)
2.27 (546)
2.88 (431)
0.0
0.001
LMCT/LLCT
LMCT/LLCT
HOMO-2!LUMO+1 (98%)
3 2 2 2
trans,trans,trans-[Pt(N ) (OH) (2-hydroxyquinoline) ] (16)
1
2
HOMO!LUMO (98%)
2.05 (606)
2.19 (565)
0.002
0.007
LMCT/LLCT
LMCT/LLCT
HOMO-1!LUMO (93%)
[
a] Tr=transition number as obtained in the TDDFT calculation.
complexes significantly lower the energies of the LUMO
and LUMO+1 and lead to a red shift of the main absorp-
tion to 318 and 314 nm for complexes 5 and 6. The fairly
long wavelength of complex 7 (333 nm) is mainly due to the
ergies of the unoccupied orbitals are similar to those in com-
plexes 5 and 6. The substituents on the pyridine ligands can
also alter the absorption energy. For example, quinoline and
pyridine with electron-withdrawing substituents tend to
lower the energies of the unoccupied orbitals of complexes
ꢀ
higher occupied orbital with I as leaving ligands, yet the en-
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

P. J. Sadler, R. J. Deeth et al.
electron-donating effects on py
(
544 nm for complex 9, 2,6-di-
fluoro-py ligand, 476 nm in
complex 12, 4-MeO-py ligand).
Selected frontier orbitals for
complex 10 generated by Mole-
[19]
kel v.5.3
are presented in
Figure 5. The main contribu-
tions for low-energy absorption
are from the nonbonding orbi-
tals of azide ligands (HOMO-2
and HOMO) to the LUMO,
which comprise antibonding or-
bitals on the pyridine nitro
group. Some familiar trends are
observed in complexes 8 and
1
6, which feature unusually low
energy absorption. Electron-
withdrawing groups tend to in-
crease the likelihood of ligand-
to-ligand
charge
transfer
(
LLCT), where electrons move
from leaving groups to py/quin,
leading to a different mecha-
Figure 5. Prominent molecular orbital diagram of the main absorption for selected platinum complexes. Orbi-
tal energies are given in eV. Full symbols correspond to vacant orbitals; empty symbols correspond to occu- nism of photoreaction. Like-
pied orbitals. The arrows represent contributions to a single excitation and should not be interpreted as indi- wise, the calculated low-energy
vidual excitations.
absorption for complexes 6 and
7
with iodido ligands are 744
8
–11 to a similar level, resulting in lowering the energy of
and 732 nm, respectively. The results are consistent with the
UV/Vis spectra of iodide-containing complexes obtained in
absorption from 301 to 307 nm. Also, the presence of a meth-
oxy group on pyridine in complex 12 results in increase of
the absorption maximum to 301 nm, but the mechanism is
different from that for complexes 8–11. The electron-donat-
ing substituent reduces the energies of transitions by in-
creasing the contribution from the frontier orbitals that are
closer to the HOMO and LUMO.
[9b]
previous studies. Comparison of the energies of the fron-
tier orbitals in complex 1 reveals that the energy of the
LUMO in complex 6 is lowered whereas the energy of the
HOMO in complex 7 is raised, resulting in a smaller gap be-
tween HOMO and LUMO, and thus a longer wavelength
absorption can be achieved.
Electronic transitions of low energy: The TDDFT results in
Table 3 show that the “tail” of the absorbance extends from
the green region for complex 1 (506 nm, py ligand) to the
red region for complexes 8 (625 nm, 2-CN-py ligand) and 10
Structures and bonding characteristics: The structural fea-
IV
tures and PtꢀL bond lengths of the computed Pt com-
plexes are listed in Figure S1 in the Supporting Information
and the results of PtꢀN bond analysis are shown in
py
(
858 nm, 2-NO -py ligand). These results suggest that elec-
Table 4. The relationship between the structures/bonding
characteristics and photophysical properties provides insight
into the pathway of photoreaction and may aid the design of
more efficient photoactive anticancer compounds.
2
tron-withdrawal by resonance effects has a larger influence
on reducing the energy of the LUMO in Pt complexes com-
pared to electron-withdrawal by the inductive effect and
[
a]
[b]
[c]
IV
Table 4. Energy decomposition of the PtꢀNpy and PtꢀNquin bond in selected Pt complexes.
Complex
1
2
5
8
9
12
13
14
15
DEstr (py)
DEstr (pt)
DEstr
0.7
16.1
16.8
1.0
9.0
10.0
0.4
30.5
31.0
1.6
20.6
22.2
1.4
21.3
22.7
1.2
16.5
17.7
0.8
20.7
21.5
9.7
18.0
27.6
12.0
24.5
36.5
DEPauli
DEelst
DEorb
DEint
DEsum
167.6
ꢀ137.4
ꢀ84.0
ꢀ53.7
ꢀ36.9
0.61
157.0
ꢀ127.6
ꢀ71.9
ꢀ42.5
ꢀ32.5
0.56
159.5
ꢀ131.4
ꢀ82.0
ꢀ53.9
ꢀ23.0
0.62
162.2
ꢀ130.2
ꢀ79.8
ꢀ47.7
ꢀ25.5
0.61
133.9
ꢀ101.4
ꢀ63.6
ꢀ31.1
ꢀ8.4
0.63
178.82
ꢀ148.7
ꢀ88.1
ꢀ58.0
ꢀ40.3
0.59
166.3
ꢀ134.8
ꢀ83.8
ꢀ52.3
ꢀ30.8
0.62
223.8
ꢀ170.1
ꢀ126.8
ꢀ73.0
ꢀ45.4
0.75
215.6
ꢀ158.9
ꢀ127.4
ꢀ70.7
ꢀ34.2
0.80
DEorb/DEelst
ꢀ
1
[
a] Energies in kcalmol ; [b] py=pyridine; [c] quin=quinoline.
&
8
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Photoactive Platinum Prodrugs
FULL PAPER
Relationship between calculated structures and wavelength
of simulated intense absorption band: Due to the varied
ligand atoms and geometries, any relationship between
Energy decomposition of the Ptꢀpy/quin bond: To explore
ꢀ
the effect of geometry, OR and (pseudo)halogen ligands,
and various pyridine substituents on the photochemistry of
IV
IV
structure and spectroscopy of the modelled Pt complexes
Pt complexes, and on the nature of the PtꢀN bond, the
needs to be interpreted with care. However, some limited
correlations between the structures and the lmax of the in-
tense absorption band can be made.
PtꢀN and PtꢀN
bond energies were decomposed into
quin
py
contributions from the strain energy, DE , and the interac-
str
tion energy, DE ; that is, DE=DE +DE . The former is
int
str
int
Firstly, the average PtꢀN , PtꢀO and PtꢀN-ligand bond
the energy required to deform the equilibrium structures of
the metal fragment and py/quin toward their respective geo-
metries in the Pt complexes. DEint is the interaction energy
3
lengths of complexes 8–13 are similar and there is only
a marginal difference in the lmax of the intense band (304–
IV
3
11 nm). Secondly, ligands with a stronger trans influence
between the in-complex metal and py/quin fragments. It was
further partitioned into three components: 1) the Pauli re-
pulsion between the (non-overlapping) charge clouds of the
py/quin and metal fragments at their in-complex positions,
DEPauli; 2) the electrostatic interaction between the two frag-
ments, DEelec; and 3) the stabilising orbital interactions ac-
companying relaxation of the electron density, DEorb. Thus,
elongate the PtꢀL bonds in the trans position, which tends
to decrease the energy gap between the occupied and unoc-
cupied frontier orbitals in the platinum complexes, leading
to a slightly red-shifted intense band in the absorption spec-
trum (Figure 5). Therefore, the lmax of the computed intense
band for complexes 1 (293 nm), 3 (295 nm) and 4 (306 nm)
are longer than for complex 2 (273 nm).
DEint =DEPauli +DEelec +DEorb
.
For complexes 1 and 2, the PtꢀN bond lengths in the
As summarised in Table 4, relative to complex 1, the larg-
est total interaction energies are observed for complex 14
3
trans-N isomer (1; ca. 2.09 ꢁ) are longer than the corre-
3
ꢀ
1
sponding values for the cis-N isomer (2; ca. 2.05 ꢁ) due to
(intramolecular H bond, <M->45.4 kcalmol ) and com-
3
ꢀ
1
the stronger trans influence of azide ligand, yet the average
PtꢀO bond lengths in the two isomers are similar (ca.
plex 12 (electron-donating substituent, ꢀ40.3 kcalmol ).
ꢀ1
Electrostatics provides a 30–60 kcalmol larger stabilising
2
.05 ꢁ). Furthermore, the PtꢀN bonds of SꢀPtꢀN in com-
contribution to the PtꢀN and PtꢀN bond than do orbital
3
3
py
quin
plex 3 (2.08 ꢁ) and of HꢀPtꢀN in complex 4 (2.25 ꢁ) are
interactions. The ratio of the orbital to the electrostatic con-
3
much longer than that of pyꢀPtꢀN in complex 2 because
tributions, DEorb/DEelec, indicates the relative covalent versus
3
ꢀ
IV
the S
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(CH ) and H ligands have stronger trans influence
ionic character of the PtꢀN and PtꢀN
bond. For Pt
quin
3
2
py
than pyridine.
complexes with no intramolecular H bond, the covalent/
Thirdly, the bond lengths of PtꢀOAc are longer than
those of PtꢀOH. For instance, in complexes 5, the PtꢀOAc
bond lengths are 2.08 and 2.11 ꢁ, whereas the PtꢀOH bond
ionic ratio, R , is between 0.56 and 0.63. The PtꢀN and
ci
py
PtꢀN
bond for complexes with an intramolecular H bond
quin
ꢀ
between OH on the py/quin ring and an OH ligand exhibits
lengths for complex 1 and 8–13 are about 2.05 ꢁ. Thus, the
lmax of intense band in the simulated UV/Vis spectrum for
complex 5 (318 nm) is substantially longer than that of com-
plexes 1 and 8–13.
more covalent bond character (R =0.75–0.80). The strain
ci
ꢀ
1
energy is 6.8 kcalmol smaller for complex 2 (10.0 kcal
mol ) with a cis geometry, yet it is 10–20 kcalmol larger
for complex 5 (31.0 kcalmol ) with acetate ligand and com-
plexes 14 (27.6 kcalmol ) and 15 (36.5 kcalmol ) with in-
tramolecular H bonding compared to complex 1 (16.8 kcal
ꢀ
1
ꢀ1
ꢀ
1
ꢀ
1
ꢀ1
Fourthly, complexes 14–16 were designed to form intra-
molecular H bonds by introducing a 2-hydroxypyridine
group or a 2-hydroxyquinoline group. Interestingly, although
in complexes 14 and 15, the bond lengths of PtꢀOH (ca.
ꢀ
1
mol ). In terms of bonding stability, these results suggest
useful features for incorporation into the design of new Pt
IV
2.08 ꢁ) involved in an H bond are longer than all the other
complexes.
free PtꢀOH bonds, the average PtꢀO bond lengths for each
complex (14, 2.05 ꢁ; 15, 2.06 ꢁ; 16, 2.05 ꢁ) are not much
different from that of complex 1 and 13 (1, 2.05 ꢁ; 13,
Photophysics of trans,trans,trans-[Pt(N ) (OH) ACHTUNGRTNEUNG( NH )(4-ni-
2 3
tropyridine)] (11): Tissue-penetration of light is wavelength-
3
2
2
.05 ꢁ). However, in comparisons of complex 14 with 1 and
dependent. Red and longer wavelengths of light (600–
[12,20]
complexes 15 and 16 with 13, the PtꢀN-ligand bond lengths
are either slightly or substantially increased due to the intra-
molecular H bonds (14, 2.07 ꢁ; 1, 2.06 ꢁ; 15, 2.08 ꢁ; 16,
800 nm) penetrate more deeply into tissues,
treatment of larger tumours with photochemotherapy. In the
and allow
calculated
UV/Vis
spectrum
of
trans,trans,trans-
2.14 ꢁ; 13, 2.07 ꢁ). Therefore, the lmax of the computed in-
[Pt(N ) (OH) ACTHUNGTRENNUNG
(NH )(4-nitropyridine)] (complex 11), there is
3
2
2
3
tense band for complex 14 (310 nm) is substantially longer
than that of complex 1. The same trend is found for com-
plexes 15 (317 nm) and 16 (325 nm) compared to complex
a low-energy, low intensity absorption band around 664 nm
(Figure 6) induced by the electron-withdrawing nitro group,
which mainly corresponds to a ligand AHCUTNGRNENUG( azide)-to-ligand(4-ni-
1
3 (311 nm). In conclusion, the comparison of the optimised
tropyridine) charge-transfer transition. Complex 11 was thus
synthesised in order to compare its UV/Vis spectrum with
the computed results and investigate the photoreaction
pathway promoted by the electron-withdrawing substituent.
IV
geometries shows that Pt complexes with longer PtꢀL
bonds lengths tend to have lower electronic excitation ener-
gies and are hence more likely to undergo photodecomposi-
tion with longer-wavelength light.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

P. J. Sadler, R. J. Deeth et al.
Figure 6. Comparison between calculated excited-state transitions and ab-
sorption spectrum of trans,trans,trans-[Pt(N (OH) (NH )(4-nitropyri-
3
)
2
2
A
H
U
T
E
N
N
3
dine)] (11). The oscillator strength has been scaled by 1.84-times for the
convenience of comparing with the experimental absorption. Representa-
tive frontier orbitals are shown on the right. Insert: magnification of the
spectra from 300–800 nm. The darker line corresponds to the TDDFT
COSMO data.
Experimental study: Complex
11 trans,trans,trans-
[
Pt(N ) (OH) ACHTUNGTRENNUNG( NH )(4-nitropyridine)] was synthesised fol-
3
2
2
3
lowing the steps described in Scheme S1 in the Supporting
Information. (The synthesis procedure can be found in the
1
13
195
Supporting Information.) H, C, Pt NMR and ESI-MS
Figure 7. a) UV/Vis spectra recorded for 11 in H O upon irradiation of
2
1
95
ꢀ2
3
65 nm (light dose 0, 0.7, 2.1, 4.2 Jcm ). b) Pseudo-quantum yield, F (*)
were used to characterise all the products. The Pt chemi-
cal shift of complex 11 D O is 857.7 ppm, which is consistent
with the expected value for trans-{Pt N O } species. The
of action spectrum (wavelength-dependent photodecomposition) for 11,
2
in comparison with extinction coefficient, e ( ). The data points are the
&
IV
[21]
4
2
average of two experiments.
characterisation of the intermediates and the single crystal
X-ray diffraction structures of trans-[PtCl
ACHTUNGTNERNUN(G NH )(4-nitropyri-
4 3
dine)]·H O (11a·H O) and trans-[Pt(N )
A
H
U
G
R
N
N
(NH )(4-nitropyri-
Computed and experimental spectra of complex 11: A com-
parison of calculated and experimental spectra of complex
11 is shown in Tables 1 and 3, and Figure 6. Both spectra
show absorption maxima in the UV region, which can be as-
signed to mixed LLCT/LMCT/MLCT (N , OH!N , Pt and
2
2
3
2
3
dine)] (11c) are described in the Supporting Information.
In the experimental UV/Vis spectrum of complex 11 in
aqueous solution (Figure 7a), there is an absorbance maxi-
mum at 287 nm, which is assigned to a ligand ACHUTNGRENNUG( azide)-to-
metal charge-transfer (LMCT) transition (also see the Sup-
3
3
py, Pt!4-nitropyridine) and LLCT/MLCT (py, Pt!4-nitro-
pyridine) transitions by TDDFT. In the investigation of low-
energy absorption, TDDFT predicts two small peaks at
II
porting Information for a comparison with its {Pt (N ) } pre-
3
2
cursor). The absorption intensity at >400 nm is very small
and no absorption was observed above 540 nm. When the
sample was irradiated at 365 nm, the major absorption band
430 nm (LLCT/MLCT; N , Pt, OH!4-nitropyridine) and
3
664 nm (LLCT; N !4-nitropyridine), whereas the UV/Vis
3
IV
decreased, indicating the cleavage of the Pt -azide bonds,
spectrum shows no absorption beyond 540 nm.
probably due to loss of the azide ligands (Figure 7a). For
the purpose of investigating the relationship between the
wavelength of irradiation and photodecomposition of 11 in
The spectrum of complex 11 was therefore recomputed by
TDDFT with the conductor-like screening model
[23]
(COSMO) of solvation, and the overlay with the experi-
mental spectrum shows an improvement in terms of the in-
tense bands and the low-energy absorptions. In particular,
the two small peaks at 430 and 664 nm are flattened and
two blue-shifted peaks at 376 and 531 nm are predicted in-
stead. This increase in energy is mainly due to the stabilisa-
tion of the occupied frontier orbitals and destabilisation of
LUMO by solvent effects and the resulting larger energy
gap between HOMO-3/HOMO-7 and LUMO (Figure 6).
This result matches closely the experimental absorption
spectrum and the action spectrum. It is worth noting that
the comparison of the action spectrum of trans,trans,trans-
[21c,22]
aqueous solution, an “action spectrum” was determined
Figure 7b). The action spectrum recorded the rate of pho-
(
todecomposition of the molecule upon the exposure to light
of wavelength 365 to 540 nm. The pseudo-quantum yield
(
F; ratio of the number of photoreactions to the number of
incident photons) was determined (Figure 7b). The pseudo-
quantum yield, F, for photolysis of complex 11 (upon irradi-
ation from 365 to 520 nm correlated closely with the extinc-
tion coefficient of the bands). At wavelengths >540 nm (in-
cluding 600, 633, 660 and 700 nm) the extent of photolysis
was too small to be detected under the experimental condi-
tions used.
[21c]
[Pt(N ) (OH) (py) ] (l <520 nm) reported previously
3
2
2
2
min
&
10
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Photoactive Platinum Prodrugs
FULL PAPER
with that of complex 11 demonstrates that the electron-with-
drawing group can extend the absorption to a longer wave-
length by about 20 nm by resonance effects. The photophysi-
cal behaviour has been shown experimentally to depend on
solvent polarity. The low-energy low-intensity absorption
in less polar solvents or in the gas phase may be expected to
be in closer agreement with the TDDFT prediction for the
gas-phase.
lation to electronic excitation energies calculated by DFT
and TDDFT. The main findings and the biological relevance
of this work are as follows:
[24]
1) Calculated excitation energies for the intense band ac-
cording to TDDFT with the SAOP functional are in
good agreement with experimental observations; the
theory has been shown to be a useful tool to portray
spectroscopic absorption, to further assist the design of
In addition, the lowest-lying triplet states of transition
metal complexes have been shown experimentally to make
IV
Pt complexes to achieve photoactivation at longer
[25]
prominent contributions to photodissociation processes.
wavelength, and to study the mechanisms of photoreac-
tion (vide infra).
The geometry of the lowest-lying triplet state of complex 11
was thus optimised and is displayed in Figure 8. Relative to
IV
2) Electronic transitions for cis-{Pt (N ) } isomers tend to
3
2
the singlet ground state, the two PtꢀN bond lengths are sig-
arise from the orbitals of the (pseudo)halogen ligands to
3
nificantly elongated from 2.098 and 2.097 ꢁ to 2.387 and
the d orbitals of metal, whereas those for all-trans iso-
ꢀ
2
.343 ꢁ, in the triplet state (Figure 8). Mulliken population
mers are more likely to arise from both OR and (pseu-
[26]
analysis shows that the two nitrogen atoms of azide coor-
dinated to Pt have spin densities of 0.23 and 0.28 e. The re-
sults of the calculation correlate with the release of azido li-
do)halogen ligands to the d orbitals of the metal and
thus lead to different photoproducts.
3) Electron-withdrawing/donating substituents on pyridine
(e.g., complexes 8–12) tend to red-shift the wavelength
of the main absorption slightly compared to the unsubsti-
tuted pyridine (e.g., complex 1) by up to 10 nm. Howev-
er, electron-donating substituents on pyridine reduce the
energies of transitions by increasing the contribution
from frontier orbitals closer to the HOMO and LUMO,
whereas quinoline and electron-withdrawing substituents
on pyridine achieve absorption at longer wavelength by
lowering the energies of the unoccupied orbitals.
[10i]
gands as N C radicals observed experimentally.
3
4
) The design of photoactivated Pt drugs with slower reduc-
tion rates and greater absorption at longer wavelengths
can be better achieved by lowering the energies of the
unoccupied frontier orbital of Pt complexes. Intramolec-
ꢀ
ular hydrogen bonding involving the OH ligand and 2-
hydroxyquinoline red shifts the main absorption by 17–
3
2 nm and extends the lowest energy absorption toward
the red region by 40–100 nm. Moreover, mesomeric elec-
ꢀ
ꢀ
tron-withdrawing groups (CN and NO ) on pyridine
2
also give rise to low-energy absorption with significant
intensity in the visible region.
IV
5
6
) The Pt complexes with longer calculated PtꢀL bond
lengths tend to absorb light of longer wavelengths.
) Comparison of the computed and experimental UV/Vis
spectra of complex 11, trans,trans,trans-[Pt(N ) (OH) -
3
2
2
AHCTUNGTNERNU(G NH )(4-nitropyridine)], shows that the prediction of
3
low-energy absorption in solution can be improved by
using a solvent model, whereas the absorption maximum
of the most intense feature in the spectra is little affect-
ed. The absorption energies of the complexes in biologi-
cal systems may lie between those predicted by TDDFT
with and without a solvent, depending on the extent of
solvation and the local hydrophobicity (dielectric con-
stant) of the medium in the tissue.
Figure 8. The computed geometries of the ground state and the lowest-
lying triplet state for trans,trans,trans-[Pt(N
dine)] (11).
3
)
2
(OH)
2
A
H
N
T
E
N
N
(NH
3
)(4-nitropyri-
7
) The structural and spin density analysis of the lowest-
Conclusion
lying triplet state for complex 11 suggests that the release
of azido ligands as N C radicals is possible, which is con-
3
IV
In this work, we have considered the photochemistry of Pt
complexes in terms of their coordination environment in re-
sistent with the experimental data.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!

P. J. Sadler, R. J. Deeth et al.
was carried out by Topspin 2.0 (Bruker) and MestReC version 4.9.9.9
Computational Details
(
Mestrelab Research S.L.).
ESI-MS: Positive/negative ion electrospray mass spectrometry was per-
formed on a Bruker Esquire 2000 mass spectrometer coupled with an
Agilent 1100 HPLC (without column) as an automatic sample delivery
system. Samples were prepared in 80% acetonitrile/20% water.
The geometries of molecules were optimised with the Amsterdam Densi-
[27]
ty Functional 2007 program (ADF)
at the gradient-corrected DFT
level by using the BP86 functional in combination with the uncontracted
triple-z+polarisation (TZP) STO basis sets. A small frozen core was
used for efficient treatment of the inner atomic shells. Relativistic effects
LC-MS: HPLC coupled mass spectrometry was performed on a Bruker
HCT-Ultra mass spectrometer coupled with an Agilent 1200 HPLC
system with an Agilent ZORBAX Eclipse Plus reverse phase C18
column (4.6ꢃ250 mm, 5 mm).
[
28]
were considered by the zeroth-order regular approximation (ZORA).
Vibrational frequency calculations for a selection of complexes confirmed
that the structures reported are minima on the potential energy surfaces.
Ziegler and Raukꢂs energy decomposition scheme has been used to study
UV/Vis: Electronic absorption spectra were recorded on a Varian Cary
[
29]
the metalꢀligand bond for selected structures. The lowest fifty singlet
3
00 UV/Vis spectrophotometer in a 1 cm path-length quartz cuvette. All
electronic excitation energies and their corresponding oscillator strengths
spectra were referenced to neat solvent and data were processed with
OriginLab Origin 7.0. Extinction coefficients (e) were determined over
a concentration range (Amax ~0.4–1.6) with at least four data points, by
using Pt concentrations determined by ICP-MS.
[
30]
were calculated by the statistical average of orbital potentials (SAOP)
in combination with a TZ2P basis sets (i.e., TZP with an additional polar-
isation function) by using scalar relativistic time-dependent density func-
tional theory (TDDFT) as implemented in the ADF program. Spin-orbit
Action spectrum: A modified Shimadzu RF-551 fluorescence HPLC
monitor was used as the light source to give monochromatic irradiation
with PWHH (peak-width at half-height) of about 15 nm. Filters were
IV
coupling was evaluated in the Pt complexes to enhance the comparison
with available experimental data as previous theoretical study has shown
that relativistic effects can have a significant effect on the simulated elec-
used to eliminate second order diffraction of shorter wavelengths from
[
31]
tronic spectra of heavy metal complexes.
main feature of the spectra is similar with or without spin-orbit coupling
Table S2 in the Supporting Information). Electronic transitions were
also computed by using TDDFT with COSMO for trans,trans,trans-
Pt(N (OH) (NH ] at the previously optimised structure in gas phase.
The results show that the
[33]
the specified longer wavelength monochromatic light.
complex 11 (52 mm) in
0.6 mL) were irradiated upon the monochromatic light. The progress of
Solutions of
a
1 cm path-length UV/Vis quartz cuvette
(
(
the photoreaction and time-course data and the spectral output of the
light source were monitored and recorded with an Ocean Optics
USB4000-UV/Vis spectrophotometer and processed with OriginLab
Origin 7.0. UV/Vis spectra were recorded every minute during irradiation
and the rate of loss of absorption at lmax (average of 10 data points at
[
3
)
2
2
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
3 2
)
The addition of solvent effects does not show improvement in terms of
the intense absorption maxima compared to the experimental data
(
Table S2 in the Supporting Information) and most of the calculations
were carried out by using scalar relativistic TDDFT at a relatively low
computational cost and no solvent effects. The exception was complex 11
for which the effect of solvation was investigated for the low-energy ab-
sorptions. The geometry optimisation of the lowest-lying triplet state of
complex 11 was performed by using a spin-unrestricted BP86/TZP
method with two unpaired electrons. The simulated UV spectra were
produced by the summation of Gaussian peaks, generated for each of the
calculated absorption maximum according to:
2
87(ꢁ1) nm) was monitored. In each case, the photolysis of 11 was taken
to completion (1–18 h depending on lirr), but only data relating to <10%
photodecomposition were considered when analysing the initial rate of
photolysis of this complex. The data of rate were the average of two in-
dependent runs at each lirr. The number of incident photons (Einsteins
was measured and calculated by using a potassium ferrioxalate actinome-
ter. The absorbance at lmax (A287) was plotted versus time (t); the aver-
age photolysis rate (dc/dt) was determined by Equation (3), and pseudo-
quantum yield (F) was calculated from Equation (4):
ꢀ1
)
[34]
2
2
I ¼ AeꢀðxꢀbÞ
=ð2c Þ
ð1Þ
dc dA287
1
¼
ꢂ le
287
ð3Þ
ð4Þ
where I is the peak intensity as a function of wavelength (x), A is the cal-
culated absorption maximum, b is the wavelength at which the calculated
maximum occurs and c is the peak-width parameter related to the full
width at half maximum (FWHM) resolution by:
dt
dt
No: of reactions
No: of incident photons
f ¼
pffiffiffiffiffiffiffiffiffiffiffiffi
FWHM ¼ 2cð 2 ln 2Þ
ð2Þ
The synthesis and characterisations of the intermediates for complex 11
are listed in the Supporting Information.
The model was devised to overcome issues inherent in the comparison of
computed “point spectra” with those generated by experiment. It ac-
counts for coalescence of neighbouring peaks, allowing this to be visual-
ised as a function of peak-width, but takes no account of dynamic peak-
width variations arising from the different types of electronic states in-
volved in each transition.
trans,trans,trans-[Pt(N ) (OH)
A
H
U
G
R
N
U
G
(11):
trans-
3
2
2
3
[Pt(N3) ACHNUTGTRENNUG(N NH ) ACHTUGNTRNE(GUN 4-NO -py)] (30 mg) was placed in H O (12 mL) and H O
2 3
2
2
2
(0.66 mL, 30%) was added and stirred at 313 K for 2–3 h, to give a clear
yellow solution after filtration. Water (50 mL) was added to the solution,
which was then lyophilised to remove all the solvent; yield: 18 mg
(55%).
1
H NMR (600 MHz, 90% H
2
O/10% D
2
O, pH 3.8): d=9.12 (d, H2,6,
3
195
1
1
14
1
J( Pt, H)=24 Hz, 2H), 8.56 (d, H3,5, 2H), 6.18 (t, NH
3
,
J
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
( N, H)=
2
195
1
13
5
(
8
3 Hz,
), 152.6 (C2,6), 120.9 ppm (C3,5); Pt NMR (129.4 MHz, D
57.7 ppm.
J ACHTUNGTERNNUNG( Pt, H)=52 Hz, 3H); C NMR (150.9 MHz, D O): d=156.5
2
Experimental Section
1
95
C
4
2
O): d=
All materials were used as obtained from commercial sources unless oth-
erwise stated.
+
+
LC-ESI-MS (m/z): [2M+H] calcd 909.1, found 909.1; [2M ꢀ OH]
ꢀ
1
ꢀ1
2
calcd 891.1, found 891.1; e287 =18200 Lmol cm (H O).
NMR spectroscopy: NMR spectra were recorded on Bruker DPX-400
1
(
for H) or Bruker AVIII-600 (for other nuclei) spectrometers. NMR
chemical shifts were referenced to residual protio-solvent peaks from
1
13
4
deuterated solvents [D ]MeOD ( H, d=3.31 ppm; C, d=49.00 ppm),
1
13
and [D
6
]acetone ( H, d=2.05 ppm; C, d=29.84, 206.26 ppm). D
O/10% D O chemical shifts are referenced internally to dioxane
H, d=3.75 ppm in D
d=67.19 ppm for both solvents).
nally referenced to K PtCl (15 mm) in D
2
O or
Acknowledgements
9
(
0% H
2
2
1
13
2
O; d=3.764 ppm for 90% H
2
O/10% D
Pt NMR chemical shifts were exter-
O (d=0 ppm). Data processing
2
O; C,
We thank John Bomphrey for help with simulation of UV/Vis spectra,
Dr. Lijiang Song and Philip Aston for help with mass spectrometry and
ICP-MS experiments and Dr. Luca Salassa and members of COST
[
32] 195
2
6
2
&
12
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Photoactive Platinum Prodrugs
FULL PAPER
Action D39 for helpful discussions. Financial support from Warwick Post-
graduate Research Scholarships (WPRS), ORSAS, Science City/AWM/
ERDF, ERC (grant no 247450), and EPSRC (EP/G006792/1) are grate-
fully acknowledged.
[12] L. Brancaleon, H. Moseley, Lasers Med. Sci. 2002, 17, 173–186.
[13] A. Y. Sokolov, H. F. Schaefer III, Dalton Trans. 2011, 40, 7571.
[14] J. E. Huheey, E. A. Keiter, R. L. Keiter, Inorganic Chemistry: Prin-
ciples of Structure and Reactivity, 4th ed., Harper Collins, New York,
1
993.
[
15] M. Hall, T. Hambley, Coord. Chem. Rev. 2002, 232, 49–67.
[
[
1] a) E. Wong, C. M. Giandomenico, Chem. Rev. 1999, 99, 2451–2466;
b) L. Kelland, Nat. Rev. Cancer 2007, 7, 573–584.
2] a) K. Harrap, Cancer Treat. Rev. 1985, 12, 21–33; b) R. J. Knox, F.
Friedlos, D. A. Lydall, J. J. Roberts, Cancer Res. 1986, 46, 1972–
[16] M. D. Hall, H. R. Mellor, R. Callaghan, T. W. Hambley, J. Med.
Chem. 2007, 50, 3403–3411.
[17] F. S. Mackay, S. A. Moggach, A. Collins, S. Parsons, P. J. Sadler,
Inorg. Chim. Acta 2009, 362, 811–819.
1
979.
[18] M. J. G. Peach, P. Benfield, T. Helgaker, D. J. Tozer, J. Chem. Phys.
2008, 128, 044118–044118.
[19] H. Petzold, J. J. Xu, P. J. Sadler, Angew. Chem. 2008, 120, 3050–
3053; Angew. Chem. Int. Ed. 2008, 47, 3008–3011.
[20] M. R. Detty, S. L. Gibson, S. J. Wagner, J. Med. Chem. 2004, 47,
3897–3915.
[
[
3] a) L. Kelland, Cancer Res. 1993, 53, 2581–2586; b) L. Kelland, Exp.
Opin. Invest. Drugs 2000, 9, 1373–1382.
4] a) L. R. Kelland, Cancer Res. 1992, 52, 3857–3864; b) D. Gately, S.
Howell, Br. J. Cancer 1993, 67, 1171–1176; c) S. Ishida, J. Lee, D. J.
Thiele, I. Herskowitz, Proc. Natl. Acad. Sci. USA 2002, 99, 14298–
1
4302; d) P. Mistry, L. R. Kelland, G. Abel, S. Sidhar, K. R. Harrap,
[21] a) B. M. Still, P. G. A. Kumar, J. R. Aldrich-Wright, W. S. Price,
Chem. Soc. Rev. 2007, 36, 665–686; b) P. S. Pregosin, Coord. Chem.
Rev. 1982, 44, 247–291; c) N. J. Farrer, J. A. Woods, L. Salassa, Y.
Zhao, K. S. Robinson, G. Clarkson, F. S. Mackay, P. J. Sadler,
Angew. Chem. 2010, 122, 9089–9092; Angew. Chem. Int. Ed. 2010,
49, 8905–8908.
Br. J. Cancer 1991, 64, 215–220; e) M. Dabholkar, F. Bostick-
Bruton, C. Weber, V. A. Bohr, C. Egwuagu, E. Reed, J. Natl. Cancer
Inst. 1992, 84, 1512–1517; f) G. Gifford, J. Paul, P. Vasey, S. Kaye, R.
Brown, Clin. Cancer Res. 2004, 10, 4420–4426.
5] a) Y. Jung, S. J. Lippard, Chem. Rev. 2007, 107, 1387–1407.
6] F. Kratz, I. Mueller, C. Ryppa, A. Warnecke, ChemMedChem 2008,
[
[
[22] D. Loganathan, H. Morrison, Photochem. Photobiol. 2006, 82, 237–
247.
3
, 20–53.
[
7] a) S. C. White, P. Lorigan, G. P. Margison, J. M. Margison, F. Martin,
N. Thatcher, H. Anderson, M. Ranson, Br. J. Cancer 2006, 95, 822–
[23] A. Klamt, J. Phys. Chem. 1995, 99, 2224–2235.
[24] R. S. Butler, P. Cohn, P. Tenzel, K. A. Abboud, R. K. Castellano, J.
Am. Chem. Soc. 2009, 131, 623–633.
[25] P. S. Wagenknecht, P. C. Ford, Coord. Chem. Rev. 2011, 255, 591–
616.
8
28; b) T. Dragovich, D. Mendelson, S. Kurtin, K. Richardson, D.
Von Hoff, A. Hoos, Cancer Chemother. Pharmacol. 2006, 58, 759–
64; c) J. M. Rademaker-Lakhai, C. Terret, S. B. Howell, C. M.
7
Baud, R. F. de Boer, D. Pluim, J. H. Beijnen, J. H. M. Schellens, J. P.
Droz, Clin. Cancer Res. 2004, 10, 3386–3395.
8] a) P. J. Bednarski, F. S. Mackay, P. J. Sadler, Anti-Cancer Agents
Med. Chem. 2007, 7, 75–93; b) S. B. Brown, E. A. Brown, I. Walker,
Lancet Oncol. 2004, 5, 497–508.
[26] R. S. Mulliken, J. Chem. Phys. 1955, 23, 1833–1840.
[27] G. te Velde, F. M. Bickelhaupt, E. J. Baerends, C. Fonseca Guerra,
S. J. A. van Gisbergen, J. G. Snijders, T. Ziegler, J. Comput. Chem.
2001, 22, 931–967.
[28] a) E. van Lenthe, R. van Leeuwen, E. J. Baerends, J. G. Snijders, Int.
J. Quantum Chem. 1996, 57, 281–293; b) E. van Lenthe, A. Ehlers,
E. J. Baerends, J. Chem. Phys. 1999, 110, 8943–8953.
[29] a) K. Morokuma, J. Chem. Phys. 1971, 55, 1236–1244; b) T. Ziegler,
A. Rauk, Theor. Chem. Acc. 1977, 46, 1–10; c) F. M. Bickelhaupt,
E. J. Baerends, in Reviews in Computational Chemistry, Vol. 15
(Eds.: K. B. Lipkowitz, D. B. Boyd), Wiley-VCH, Weinheim, 2000,
pp. 1–86; d) G. Frenking, N. Frohlich, Chem. Rev. 2000, 100, 717–
774.
[
[
9] N. J. Farrer, P. J. Sadler, Aust. J. Chem. 2008, 61, 669–674.
[
10] a) P. J. Bednarski, R. Grunert, M. Zielzki, A. Wellner, F. S. Mackay,
P. J. Sadler, Chem. Biol. 2006, 13, 61–67; b) J. Kasparkova, P. Herin-
gova, V. Brabec, F. Mackay, P. J. Sadler, Metal Ions in Biology and
Medicine 2008, 10, 463–469; c) J. Kasparkova, F. S. Mackay, V.
Brabec, P. J. Sadler, J. Biol. Inorg. Chem. 2003, 8, 741–745; d) N. A.
Kratochwil, A. I. Ivanov, M. Patriarca, J. A. Parkinson, A. M.
Gouldsworthy, P. D. Murdoch, P. J. Sadler, J. Am. Chem. Soc. 1999,
1
21, 8193–8203; e) F. S. Mackay, J. Woods, H. Moseley, P. J. Sadler,
[30] P. R. T. Schipper, O. V. Gritsenko, S. J. A. van Gisbergen, E. J. Baer-
ends, J. Chem. Phys. 2000, 112, 1344–1352.
[31] F. Wang, T. Ziegler, J. Chem. Phys. 2005, 123, 194102.
[32] H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 62,
7512–7515.
Br. J. Dermatol. 2005, 152, 857–858; f) F. S. Mackay, J. A. Woods, P.
Heringova, J. Kasparkova, A. M. Pizarro, S. A. Moggach, S. Parsons,
V. Brabec, P. J. Sadler, Proc. Natl. Acad. Sci. USA 2007, 104, 20743–
2
0748; g) F. S. Mackay, J. A. Woods, H. Moseley, J. Ferguson, A.
Dawson, S. Parsons, P. J. Sadler, Chem. Eur. J. 2006, 12, 3155–3161;
h) P. Mꢄller, B. Schroder, J. A. Parkinson, N. A. Kratochwil, R. A.
Coxall, A. Parkin, S. Parsons, P. J. Sadler, Angew. Chem. 2003, 115,
[33] L. Mackenzie, W. Frainbel, Br. J. Dermatol. 1973, 89, 251–264.
[34] a) H. J. Kuhn, S. E. Braslavsky, R. Schmidt, Pure Appl. Chem. 2004,
76, 2105–2146; b) S. L. Murov, G. L. Hug, I. Carmichael, Handbook
of Photochemistry, 2nd ed., M. Dekker, New York, 1993.
3
49–353; Angew. Chem. Int. Ed. 2003, 42, 335–339; i) H. I. A. Phil-
lips, L. Ronconi, P. J. Sadler, Chem. Eur. J. 2009, 15, 1588–1596.
11] C. E. Wayne, R. P. Wayne, Photochemistry, 1st ed., Oxford Universi-
ty Press, New York, 1996.
[
Received: March 8, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
13
&
ÞÞ
These are not the final page numbers!

P. J. Sadler, R. J. Deeth et al.
Computer-aided design: Platinum(IV)
complexes are inert stable prodrugs
that can be photoactivated to produce
species with promising anticancer
Antitumor Agents
H.-C. Tai, Y. Zhao, N. J. Farrer,
A. E. Anastasi, G. Clarkson,
activity. Here a combined computa-
tional and experimental study was car-
ried out to show that TDDFT can aid
the design of the coordination environ-
P. J. Sadler,* R. J. Deeth* . . . &&&& — &&&&
A Computational Approach to Tuning
the Photochemistry of Platinum(IV)
Anticancer Agents
IV
ment of Pt complexes to allow opti-
misation of the photoactivity of these
IV
Pt anticancer agents (see figure).
&
14
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!