A luminescent poly(amidoamine)-iridium complex as a new singlet-oxygen sensitizer for photodynamic therapy

Article abstract of DOI:10.1021/ic502378z

A polymer complex (1_P) was synthesized by binding bis(cyclometalated) Ir(ppy)₂⁺ fragments (ppy = 2-phenylpyridyl) to phenanthroline (phen) pendants of a poly(amidoamine) copolymer (PhenISA, in which the phen pendants involved 6% of the repeating units). The corresponding molecular complex [Ir(ppy)₂(bap)]⁺ (1_M, bap = 4-(butyl-4-amino)-1,10-phenanthroline) was also prepared for comparison. In water solution 1_P gives nanoaggregates with a hydrodynamic diameter of 30 nm in which the lipophilic metal centers are presumed to be segregated within polymer tasks to reduce their interaction with water. Such confinement, combined with the dilution of triplet emitters along the polymer chains, led to 1_P having a photoluminescence quantum yield greater than that of 1_M (0.061 vs 0.034, respectively, in an aerated water solution) with a longer lifetime of the ³MLCT excited states and a blue-shifted emission (595 nm vs 604 nm, respectively). NMR data supported segregation of the metal centers. Photoreaction of O₂ with 1,5-dihydroxynaphthalene showed that 1_P is able to sensitize ¹O₂ generation but with half the quantum yield of 1_M. Cellular uptake experiments showed that both 1_M and 1_P are efficient cell staining agents endowed with two-photon excitation (TPE) imaging capability. TPE microscopy at 840 nm indicated that both complexes penetrate the cellular membrane of HeLa cells, localizing in the perinuclear region. Cellular photodynamic therapy tests showed that both 1_M and 1_P are able to induce cell apoptosis upon exposure to Xe lamp irradiation. The fraction of apoptotic cells for 1_M was higher than that for 1_P (74 and 38%, respectively) 6 h after being irradiated for 5 min, but cells incubated with 1_P showed much lower levels of necrosis as well as lower toxicity in the absence of irradiation. More generally, the results indicate that cell damage induced by 1_M was avoided by binding the iridium sensitizers to the poly(amidoamine).

Full text of DOI:10.1021/ic502378z

Article
pubs.acs.org/IC
A Luminescent Poly(amidoamine)−Iridium Complex as a New
Singlet-Oxygen Sensitizer for Photodynamic Therapy
,
#,&,‡
Marco Galli, Laura D’Alfonso, Donato Inverso, Maria Vittoria Dozzi,#
#
§
†,⊥
#,&
Daniela Maggioni,*
§
†,⊥
§
#,&
Laura Sironi, Matteo Iannacone, Maddalena Collini, Paolo Ferruti, Elisabetta Ranucci,
and Giuseppe D’Alfonso#
,&
#
Dipartimento di Chimica, Universita
̀
degli Studi di Milano, Via Golgi 19, 20133 Milano, Italy
di Milano Bicocca, Piazza della Scienza 3, 20126 Milano, Italy
Division of Immunology, Transplantation and Infectious Diseases, Dynamics of Immune Responses Unit, San Raffaele Scientific
§
Dipartimento di Fisica, Universita
̀
†
Institute, Via Olgettina 58, 20132 Milano, Italy
⊥
Vita-Salute San Raffaele University, Via Olgettina 58, 20132 Milano, Italy
&
Milan Research Unit, Consorzio Interuniversitario Nazionale per la Scienza e Tecnologia dei Materiali, Via Golgi 19, 20133 Milano,
Italy
‡
Institute of Molecular Science and Technologies, CNR, Via Golgi 19, 20133 Milano, Italy
S Supporting Information
*
ABSTRACT: A polymer complex (1 ) was synthesized by
P
binding bis(cyclometalated) Ir(ppy)2+ fragments (ppy = 2-
phenylpyridyl) to phenanthroline (phen) pendants of a poly-
(
amidoamine) copolymer (PhenISA, in which the phen pendants
involved ∼6% of the repeating units). The corresponding
molecular complex [Ir(ppy) (bap)] (1 , bap = 4-(butyl-4-
+
2
M
amino)-1,10-phenanthroline) was also prepared for comparison.
In water solution 1 gives nanoaggregates with a hydrodynamic
P
diameter of 30 nm in which the lipophilic metal centers are
presumed to be segregated within polymer tasks to reduce their
interaction with water. Such confinement, combined with the
dilution of triplet emitters along the polymer chains, led to 1 having a photoluminescence quantum yield greater than that of 1
P
M
3
(0.061 vs 0.034, respectively, in an aerated water solution) with a longer lifetime of the MLCT excited states and a blue-shifted
emission (595 nm vs 604 nm, respectively). NMR data supported segregation of the metal centers. Photoreaction of O with 1,5-
2
1
dihydroxynaphthalene showed that 1 is able to sensitize O generation but with half the quantum yield of 1 . Cellular uptake
P
2
M
experiments showed that both 1 and 1 are efficient cell staining agents endowed with two-photon excitation (TPE) imaging
M
P
capability. TPE microscopy at 840 nm indicated that both complexes penetrate the cellular membrane of HeLa cells, localizing in
the perinuclear region. Cellular photodynamic therapy tests showed that both 1 and 1 are able to induce cell apoptosis upon
M
P
exposure to Xe lamp irradiation. The fraction of apoptotic cells for 1 was higher than that for 1 (74 and 38%, respectively) 6 h
M
P
after being irradiated for 5 min, but cells incubated with 1 showed much lower levels of necrosis as well as lower toxicity in the
P
absence of irradiation. More generally, the results indicate that cell damage induced by 1 was avoided by binding the iridium
M
sensitizers to the poly(amidoamine).
1
. INTRODUCTION
turn generate ROS (type I photochemistry). Alternatively, it
1
can directly interact with molecular oxygen in its ground triplet
Photodynamic therapy (PDT) is receiving an increasing
amount of attention as a noninvasive clinical treatment for
different types of cancer (such as those affecting skin,
bladder, esophagus, lung, and head and neck ) and as an
alternative method for killing pathogens in localized infections.
1
state to produce “in situ” cytotoxic singlet oxygen ( O )
2
through an energy transfer process (eq 1).
2
3
4
4
5
6
3
3
1
PS* + O → PS + O
(1)
2
2
7
PDT is based on the use of a photosensitizer (PS), which
This type II photochemistry is the most relevant mechanism
1
efficiently populates an excited triplet state upon interaction
of PDT in cells, because most PSs are effective O producers.
2
3
with visible light. The triplet state of a PS ( PS*) can produce
The reactive oxygen species generated are capable of causing
toxic reactive oxygen species (ROS), such as singlet oxygen
1
3
(
O ) or free radicals, by two different pathways. PS* can react
Received: September 29, 2014
2
with molecules to generate intermediate free radicals that in
©
XXXX American Chemical Society
A
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
irreversible damage if generated inside cells, particularly inside
specific subcellular organelles (mitochondria, Golgi apparatus,
etc.) where the PSs can localize and accumulate. Indeed, singlet
oxygen has a radius of destruction measured in nanometers
Chart 1. Structure of the PhenISA Copolymer
(
10−60 nm, its lifetime being in the range of 10−320 ns), and
photodynamic damage will then occur only very close to the
intracellular location of the PS. Although dependent on many
8
factors, only one of which is the subcellular location of the PS,
PDT treatment can cause cell death by apoptotic or necrotic
pathways, in dependence of many factors, including the
subcellular location of the PS. Apoptosis is programed cell
death that does not cause inflammation in vivo, whereas
necrosis is a pathological cell death in which cellular content
leaks out, potentially causing lethal damage in nearby cells and
inflammation in vivo. In many cases, PDT has been found to be
highly efficient at inducing apoptosis as a result of a complex
involved ∼6% repeating units, whereas the large majority of the
25
units were those of the ISA23 polymer. PhenISA was able to
bind Re(CO)3+ and Ru(phen)2 fragments, affording
luminescent polymer complexes that were internalized by
2+
24
HEK-293 cells. These results prompted us to investigate the
binding of PhenISA to iridium fragments to obtain a
8
,9
cascade of events. This feature is very important because it
implies that doses lower than those necessary for producing
luminescent triplet emitter useful for both optical imaging
10
1
necrosis may still be effective at killing cells. Furthermore,
efficient induction of apoptosis by PDT implies that PDT may
be able to bypass mechanisms that make cells resistant to
apoptosis in response to chemotherapeutic drugs and ionizing
and producing O for PDT purposes.
2
This Article reports on the synthesis and photochemical
characterization of a water-soluble Ir−PhenISA conjugate,
showing that binding to the polymer improved the photo-
physical properties of the Ir emitters. The photooxidation of
1,5-dihydroxynaphthalene (DHN) was used to check the
1
0
radiation.
PDT would therefore be able to selectively kill diseased cells,
reducing collateral effects on healthy tissues, provided that
selective delivery of the PS to the target could be assured.
capability of this new metallopolymer to act as a sensitizer of
1
O generation. In preliminary tests with HeLa cells, the
2
1
a,11
1a
12
Porphyrins,
phthalocyanines, fullerene derivatives, and
complex was internalized in cells and could be imaged by two-
photon excitation microscopy. Moreover, the complex was able
to induce cell apoptosis upon Xe lamp irradiation, with no
evidence of the intrinsic cell damaging properties typically seen
using the corresponding free Ir complex.
1
3
14
organic dyes, like methylene blue or rose bengal, are valid
1a
PSs, as are some organometallic complexes. Among them,
cyclometalated iridium complexes have attracted much
1
5−17
attention
on the basis of many favorable properties: high
quantum yields of triplet formation (even higher than 0.9 in
18
2
. RESULTS AND DISCUSSION
2.1. Model Complex [Ir(ppy) (bap)] (1 ) (ppy = 2-
deaerated solutions), long lifetimes of the excited triplet state
typically in the microsecond range, which is long enough to
+
(
2
M
3
enable quenching reactions with O2 before spontaneous
decay), and triplet energy high enough to allow for the energy
transfer process (eq 1). Moreover, it has been shown that they
phenylpyridyl, bap = 4-(butyl-4-amino)-1,10-phenan-
throline). The related [Ir(ppy) (phen)] complex was
previously prepared by reacting dinuclear precursor [Ir-
(ppy) Cl] with phen at high temperature in ethylene glycol.
+
2
16
26
are usually resistant to attack by singlet oxygen.
2
2
However, several disadvantages hinder biomedical applica-
tions of these complexes. Their solubility in water is generally
very low. Moreover, the circulating time of small molecules in
biological fluids is generally too short to allow significant
accumulation of a molecular sensitizer proximal to the target for
in vivo applications.
In the polymer complex reported here, phen ligands are
appended to the polymer chain by butylamino substituents in
the 4-position of phenanthrolines (bap ligand, whose synthesis
2
4,27
has been reported elsewhere).
It was therefore necessary to
+
synthesize the [Ir(ppy) (bap)] complex (1 ) to have a reliable
2
M
molecular model of the designed polymer complex. A
previously published protocol was followed with some
modifications (Scheme 1). The precursor [Ir(ppy) Cl] was
The loading of complexes on suitable nanometric carriers can
be exploited to improve their solubility in aqueous media,
increase their plasma residence time, and reduce their
2
2
treated with 2 equiv of bap ligand in a mixed solvent (THF/
1b,19
toxicity.
Actually, stealth nanoparticles (i.e., nanoparticles
H O 1:1) under mild heating (∼70 °C). A clear yellow solution
2
covered by macromolecules that make them invisible to the
reticuloendothelial system) can benefit from a prolonged
circulating time and effectively accumulate in solid tumors,
owing to the so called enhanced permeability and retention
formed upon heating, and the orange photoluminescence of the
Ir complex progressively overcame the blue fluorescence of free
1
phen. Reaction progress was monitored by H NMR
spectroscopy. All resonances were attributed by scalar and
2
0
1
1
1
13
(
EPR) effect.
Linear amphoteric poly(amidoamine)s (PAAs) have well
established properties of water solubility, biocompatibility,
dipolar correlation two-dimensional H− H and H− C NMR
experiments (Figures S1−S3 in the Supporting Information
(SI)).
21
22
+
biodegradability, and stealth-like behavior. Moreover, they
tend to aggregate in water solutions in the form of small
nanoparticles with hydrodynamic diameters in the range of 5−
In contrast to the [Ir(ppy) (phen)] complex, 1 (as a
2
M
chloride salt) is moderately soluble in water due to the presence
of a protonated amino group on the phen ligand (pK = 10.2, as
a
2
3,24
2
0 nm.
They are therefore very attractive as carriers for
determined by potentiometric titration on the bap ligand).
Solubility was high enough to perform photophysical character-
ization in water solution, as well as the biological tests described
in section 2.5. However, through dynamic light scattering
(DLS) and NMR data, a minor fraction of the compound was
molecular complexes. In a previous work, a PAA copolymer
(
(
PhenISA, Chart 1) was synthesized bearing phenanthroline
phen) pendants, which are strong chelating ligands toward a
24
variety of transition-metal fragments. The phen ligand pendants
B
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Synthesis of Complex 1M
found to be present in the form of nanoaggregates, as discussed
below.
luminescence quantum yields (PLQYs, ϕ) in the presence of
oxygen, observed in CH Cl or CH CN solution (Table 1),
2
2
3
The UV−vis absorption spectrum (Figure 1) of 1 shows
indicates that emission occurs from an excited state with a
M
1
strong spin-allowed ligand-centered ( LC) bands in the range
substantially triplet nature. This agrees with literature data that
+
attributed the emission from [Ir(ppy) (phen)] complexes
2
(
phen indicating different phenanthroline-based ligands) as
3
arising from MLCT states involving the π* orbitals of phen
ligands as the acceptor orbitals.
2
6,28,34
In water, 1M photoluminescence is centered at 604 nm, a
position that did not change in the pH range of 3.2−7.2. The
behavior in the presence of oxygen did not fit with the typical
behavior of triplet emitters observed in organic solvents.
Lifetimes and PLQYs were only modestly affected by the
presence of oxygen (Table 1). Moreover, in aerated conditions,
the PLQYs were higher in water than in acetonitrile, contrary to
what would be expected from the polarity trend. A possible
explanation was provided by DLS analysis of the water solution
∼2 × 10− M), which showed the presence of nanoparticles
5
(
with hydrodynamic diameters of ∼200 nm. This suggested that
Figure 1. UV−vis absorption (left) and photoluminescence (right)
spectra of 1 (black traces) and 1 (gray traces) in water at room
temperature with λex = 400 nm.
1
was not completely dissolved but was present, at least in
M
M
P
part, in the form of nanosized aggregates. Such aggregates
cannot account for the entirety of 1M present in solution
because colloids are NMR silent (or give very broad, hardly
35
1
detectable resonances), whereas sharp H NMR signals were
of 200−300 nm and broad weaker absorptions at longer
wavelengths (peaks are recognizable at approximately 378 and
observed for samples of 1 in water (e.g., Figure S1 of the SI).
M
It has been shown (see Experimental Section) that the NMR
4
17 nm (Figure 1); their position did not change significantly
silent fraction of 1 in water corresponds to ∼25% of the total
upon varying the solvent, see Figure S4 in the SI), attributable
to singlet metal-to-ligand charge transfer ( MLCT) transitions,
M
1
sample. Such a fraction of aggregated 1 , even if minor, could
M
dominate the emission features because the complexes inside
the nanoparticles are expected to be more brilliant, being
protected from the deactivating actions of oxygen and the polar
which is in agreement with literature data for similar
1
5,26,28
complexes.
Upon excitation at 400 nm, complex 1M exhibits yellow-
orange photoluminescence at wavelengths that are strongly
sensitive to the nature of the solvent (Table 1 and Figure S4 in
the SI). A blue shift is observed upon decreasing solvent
36
solvent. Therefore, the wavelength, lifetime, and PLQYs of
the emission measured in water should be primarily attributed
to the aggregates, with little contribution from the free
molecules. Interestingly, in water, a double exponential model
was necessary to describe the lifetime decays (Table 1) to
2
9
polarity, which is typical of excited states that are more polar
than their ground states (such as CT states) because they are
preferentially stabilized by polar solvents, provided that the
excited state lifetime is longer than the solvent reorganization
2
achieve a function of merit (χ ) comparable to that obtained for
the organic solvents. The main component (longer and less
sensitive to oxygen) is ascribable to the species inside the
3
2,33
time.
The sharp drop in the lifetimes (τ) and photo-
Table 1. Photoluminescence Data for Molecular Complex 1 in Aerated or Deaerated Solutions in Different Solvents and for
M
a
Polymer Complex 1 in an Aerated Water Solution
P
compound
solvent
CH Cl
λem (nm)
condition
τ (ns)
ϕ
kr (s−1)
3.4 × 105
knr (s−1
)
6
aerated
153
1123
56
0.052
0.35
6.2 × 10
5.8 × 10
1.8 × 10
8.2 × 10
8.3 × 10
7.2 × 10
4.5 × 10
1M
2
2
574
5
5
7
5
6
6
6
deaerated
aerated
3.1 × 10
0.017
0.30
3.1 × 105
1M
CH CN
3
591
5
deaerated
aerated
858
3.5 × 10
2.8 × 105
b
b
b
116 (98%), 2 (2%)
134 (95%), 64 (5%)
212
0.033
0.038
0.061
1M
H O
2
604
595
5
b
deaerated
aerated
2.8 × 10
1P
H O
2
2.9 × 105
a
b
Room temperature, λ = 400 nm; k and k indicate the radiative and nonradiative decay constants of the excited states, respectively. Computed
ex
r
nr
on the most significant lifetime component.
C
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Synthesis of the Polymer Complex 1P
nanoparticles, whereas the minor component (shorter and
relieved, and exposure to the mixed solvent made the
strongly affected by oxygen) is attributable to free molecules.
Ir(ppy) (bap) pendants more mobile and consequently their
H signals more visible.
2
1
2
.2. Preparation and Characterization of the 1P
Complex. Synthesis of the PhenISA copolymer (Chart 1)
The UV−vis absorption spectrum of 1 was superimposable
P
24
has been previously reported. The majority of it is derived
from a Michael addition reaction between piperazine and
bis(acrylamido)acetic acid (BAC), whereas a minority piece
on that of molecular complex 1 , except for the very intense
M
absorption at ∼200 nm due to the amide groups of the polymer
(gray trace in Figure 1). The photoluminescence spectrum
instead showed some significant differences with respect to that
(
∼6%) arises from the analogous reaction between BAC and
the primary amine bap. DLS measurements indicated that this
copolymer self-assembled in aqueous media, giving rise to the
formation of roughly spherical nanoaggregates with a hydro-
of 1 . The emission maximum was blue-shifted by 9 nm (250
M
−1
cm ), and a longer lifetime (212 ns) and higher quantum yield
(0.061) were found. These findings agree with the idea that, in
water, the pendant Ir complexes are segregated into polymer
pockets in which they experience less polar surroundings (thus
giving a blue-shifted emission) and a more rigid environment
than in solution. Moreover, the emitting Ir centers are very
diluted along the polymer chains of 1P because the units
bearing the phen pendants are a very minor component of the
copolymer. This prevents self-quenching and annihilation
effects, which are often encountered for emission from the
triplet state of transition-metal complexes in solids or in films at
24
dynamic diameter of ∼20 nm.
Complexation of Ir to PhenISA was performed by a route
similar to that used for preparing molecular complex 1M
(
Scheme 2), but in a THF/H O mixed solvent containing
2
double the amount of water with respect to the synthesis of 1M
because of the low solubility of PhenISA in nonaqueous media.
On the other hand, the presence of THF was necessary because
starting reagent [Ir(ppy) Cl] is insoluble in water. The
2
2
reaction temperature was lower (50 °C) than that used for
the synthesis of 1M to avoid thermal degradation of the
polymer. After 6.5 h, the mixture was concentrated under
reduced pressure to remove most of the THF and dialyzed
against water for four days to remove unreacted metal
fragments and lower molecular weight polymer fractions, if
present.
37
high dopant concentrations and might also occur in the
nanoaggregates formed by 1M in water. In fact, the best
emitting features are often observed in systems where a rigid
38
environment (rigidochromic effect) is joined to a low dopant
39
concentration.
2.3. Photochemical Stability of 1 . Photochemical
P
The content of Ir bound to the polymer was measured by
ICP-AES analysis. The results (2.45% w/w) indicated that the
majority (88.3%) of the phenanthroline pendants had reacted
stability tests were carried out to determine the robustness of
both the iridium complex and the PAA chain under prolonged
irradiation in the presence of O . This is a key prerequisite for
2
with Ir(ppy) fragments.
using the polymer complex as a photosensitizer. 1P was
2
Binding of the metal centers did not hamper the tendency of
PhenISA to self-aggregate. A DLS measurement, performed on
dissolved in water/methanol (85:15), and the solution was
saturated with oxygen by bubbling O for 10 min. The solution
2
a sample of dialyzed and lyophilized 1 , showed a size
distribution centered at ∼30 nm (Figure S5 in the SI), which is
slightly larger than that of PhenISA alone.
was then exposed to visible light (150 W xenon lamp, 390 nm
cutoff filter), and UV−vis absorption spectra were recorded
every 30 min for 4 h (Figure S7 of the SI). Superposition of the
absorption spectra recorded at different times revealed high
photostability of the compound. The same test was repeated for
P
Another interesting feature of the organization of the
1
polymer coils in solution was revealed by the H NMR
spectrum of the purified polymer complex 1 in D O solution.
complex 1 with analogous results.
P
2
M
In these conditions, the aromatic resonances of the ppy and
phen ligands were particularly broad, and their integrated
intensities did not reflect the true percentage of the minority
component (see the bottom spectrum of Figure S6 in the SI, in
which the aromatic signals are almost undetectable). Pro-
2.4. Photoreaction of 1 and 1 with 1,5-Dihydrox-
P
M
1
ynaphthalene as a Reporter of O Formation. To test the
2
1
capability of 1 and 1 to act as sensitizers for O generation,
M
P
2
photochemical reactions were performed in solutions presatu-
rated with O , employing 1,5-dihydroxynaphthalene (DHN) as
2
1
gressive addition of THF-d to the NMR tube led to the
the O reporter. Indeed, it is known that DHN promptly and
quantitatively reacts with O to give 5-hydroxy-1,4-naphthale-
nedione (juglone, Scheme 3).
8
2
1
recovery of the aromatic resonances at their correct intensities
2
1
5,40
(Figure S6 in the SI). This suggests that, in water, the
complexes, in which the metal is surrounded by a lipophilic
cage formed by the three large aromatic ligands, tend to
segregate within the polymer to avoid contact with the water.
Upon the less polar THF solvent being added, segregation was
Reaction progress was monitored by UV−vis absorption
spectroscopy following the decrease in the DHN band at 297
nm and the concomitant increase of the large juglone band
centered at 427 nm (Figure 2). This reaction occurred without
D
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
Scheme 3. Photochemical Reaction Used To Monitor O
2
a
Formation
a
PS represents the Ir complexes 1M and 1P.
the formation of long-lived intermediates or byproducts as
indicated by the two isosbestic points at 280 and 330 nm
observed in the spectra recorded during the course of the
irradiation. Negligible juglone formation was observed in the
absence of the Ir sensitizers, even after 70 min of irradiation of
O -saturated solutions.
2
Figure 3 shows the first-order semilogarithmic plots for
reaction 2, sensitized by either 1 or 1 . In the case of 1 , the
M
P
M
values of ln(A /A ) decreased linearly over time from the
t
0
Figure 3. Photooxidation of DHN in the presence of sensitizers 1M
beginning up to ∼50 min of irradiation, in agreement with
and 1 (data up to 50 min have been plotted corresponding to the
P
pseudo-first-order kinetics, with rate r = k [DHN] (slope =
linearity time interval). A and A represent the absorbance measured
obs
t
0
kobs = 5.1 × 10− min ). At longer times, progressive deviation
from linearity was observed, possibly related to the fact that
juglone absorbs at the same wavelengths as the Ir sensitizer.
3
−1
at 297 nm (the maximum of the DHN absorption band) at time t and
time 0, respectively. The absorbance of the Ir sensitizer at this
wavelength was subtracted.
The rate of juglone formation compares well with that observed
+
3
for the same reaction sensitized by the [Ir(ppy) (phen)]
of excited triplet sensitizers by O (reaction 1) requires the
2
2
complex in a mixed acetonitrile/2-PrOH solution (k = 6 ×
interaction of the two species, and the mobility of the Ir
obs
−3
−1 40
10
min ).
complexes in 1 is drastically reduced with respect to that of 1M
P
because of their binding to the polymer. Furthermore, the
photophysical and NMR data suggest that the polymer adopts
conformations that favor segregation of the complexes to
1
2
DHN + O → juglone + 3O
1
2
2
(
2)
On the contrary, an induction period was observed for 1P,
reduce its interaction with water. Therefore, the time necessary
1
followed by a linear decrease with a smaller slope (k = 2.1 ×
for the generated O to diffuse in the solution and encounter
obs
2
−3
−1
10
min ). The induction time suggests that at the beginning
the DHN probe must also be considered.
The ratio between the slopes of the two lines reported in
Figure 3, corrected to take into account the different amount of
1
the polymer itself competed with DHN for consuming O ,
2
41
most likely by reaction with the alkene groups constituting
the terminals of the PAA chains (obtained by Michael addition
reactions). Therefore, reaction 2 likely could only start when
these terminal groups had been saturated. The lower rate
radiation absorbed by the two sensitizers during the reaction,
1
allowed for the ratio of the quantum yields of O generation by
2
1 and 1 to be estimated as 2.1 (see Experimental Section for
M
P
relative to 1 is probably attributable to the fact that quenching
details). Thus, binding to the polymer reduces the efficiency of
M
Figure 2. UV−vis absorption spectra recorded at different times of irradiation (λ > 390 nm, xenon lamp) on solutions containing (a) complex 1
M
−5
−5
−4
(
4.1 × 10 M) or (b) complex 1 (3.6 × 10 mol/L of Ir) and DHN (3.7 × 10 M) in 3 mL of H O/MeOH (85:15) bubbled with O for 10 min.
P
2
2
E
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
the Ir complex as a sensitizer for O generation, though not
dramatically.
further confirmed by quantification of dead cells on non-
irradiated samples as described below.
2
2
.5. PDT Treatment of Cells. Before testing the capability
For PDT tests, HeLa cells were incubated with 1 or 1 as
described above for the cellular uptake assays. Subsequently,
multiwell plates were irradiated for 5 min with a Xe lamp and
M
P
1
of the Ir complexes to induce cell death by O generation, we
investigated their cellular uptake. The efficiency of photo-
sensitizer incorporation into the cells can significantly affect the
capability of the sensitizer to induce cell death. Therefore,
HeLa cells were incubated with compounds 1 and 1 at 37 °C
under a 5% CO atmosphere (see Experimental Section) for
differing amounts of time. Images of the incubated cells were
then acquired by two-photon excitation (TPE) microscopy at
40 nm, where both complexes showed the highest two-photon
absorption. TPE allows for less cellular damage and deeper light
penetration in vivo because typical laser wavelengths lie in the
2
then maintained at 37 °C under a 5% CO atmosphere (see
2
14
Experimental Section). The same treatment was applied to
control cells (without sensitizer) to check for adverse effects of
exposure to the Xe lamp. Cell viability was evaluated by flow
cytometry analysis performed 4 and 6 h after irradiation along
with nonirradiated samples (Figure S9 of SI) using annexin V
and 7-aminoactinomycin D (7AAD) staining to denote
apoptotic and necrotic cells, respectively.
M
P
2
8
Cells incubated with either 1M or 1 underwent apoptosis
P
7
00−900 nm range corresponding to the window in which
upon irradiation (Figure 5). The fraction of apoptotic cells for
both the tissue chromophores and water absorb more weakly.
Previous studies had shown that cellular internalization of the
24
Ru−PhenISA complex required several hours. Uptake of Ir−
PhenISA complex 1 was therefore checked after 12 h of
P
incubation. However, uptake of molecular complexes is usually
4
2
faster; therefore, internalization of 1 was monitored after 2 h
M
of incubation.
Figure 4 shows that both molecular complex 1M (upper
panels) and polymer complex 1 (lower panels) are able to
P
Figure 5. Percentages of dead cells resulting from apoptotic or
necrotic events upon PDT treatments (6 h after being irradiated for 5
min) in control samples and samples sensitized by 1M or 1P.
1M was higher than that for 1 , which is in line with the kinetics
P
1
of O generation measured in the photooxidation of DHN.
2
However, an increased number of necrotic events with respect
to the untreated cells were observed in the cells treated with 1M
(
Figure 5 and Figure S9 of the SI).
Next, to investigate dark toxicity (i.e., toxicity in the absence
of irradiation), HeLa cells were incubated with either 1 or 1 ,
M
P
and the number of viable (living cells recovered after
treatment) and dead cells was measured (see Experimental
Section). As shown in Figure S10 of the SI, cells treated with
1M had a significantly higher number of dead cells and a
significant reduction in cell viability relative to both untreated
Figure 4. TPE microscopy images (superposition of the blue and red
channels) of HeLa cells incubated for 2 h with 26 μM 1M (upper
and 1 -treated cells.
P
panels) or for 12 h with 22 μM 1 (lower panels) and counterstained
P
Therefore, taken together, these results indicate that
treatment with free complex 1M has a significant cytotoxic
effect even without exposure to the Xe lamp. Cells incubated
with Hoechst to visualize the nuclei in the blue channel. The red color
is due to the Ir complexes. An image of control HeLa cells (incubation
without Ir complexes) for this experiment is shown in Figure S8 in the
2
with 1 showed much less cytotoxicity (both in the dark and
P
SI. Left panels: 157 × 157 μm ; right panels: enhanced to 79 × 79
2
upon irradiation). Interestingly, binding of the sensitizer to
poly(amidoamine) inhibited necrosis upon irradiation while
preserving its ability to induce apoptosis.
μm .
penetrate the cellular membrane and that both tend to be
located in the perinuclear region. This is in line with literature
data concerning related bis(cyclometalated) Ir complexes, in
which localization in subcellular organelles, such as the Golgi
apparatus, endoplasmic reticulum, lysosomes, and mitochon-
3
. CONCLUSIONS
The new luminescent metallopolymer 1 presented here is of
interest in several respects. Upon light absorption, bis-
(cyclometalated) Ir complexes appended to the polymer give
rise to triplet metal-to-ligand charge transfer excited states that
P
1
7b,d,e,43
dria, was suggested.
3
These preliminary experiments therefore indicated that 1 is
can either radiately decay or react with O and can then be
P
2
an efficient cell staining agent endowed with TPE imaging
capability that is well tolerated by cells. Indeed, cells still
appeared viable after the long incubation time, which was
exploited for either imaging or PDT applications. It has been
found that photoluminescence can be triggered by two-photon
excitation, which offers advantages over traditional one-photon
F
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
excitation in terms of less cellular damage and deeper light
penetration in vivo. Moreover, preliminary tests have shown
that 1P is internalized by living HeLa cells, which is an
important asset for PDT applications. Therefore, from the
DLS measurements were performed using a Malvern Zetasizer nano
ZS instrument at 289 K on samples dissolved in ultrapure water
(
typically 1 mg/mL).
Elemental C, H, N analyses were performed on a PerkinElmer
CHN 2400 instrument. Ir content was determined by ICP-AES (ICAP
300, Thermo Electron) on a known amount of 1 dissolved in 1.5 mL
point of view of imaging applications, 1 provides an efficient
P
6
P
water-soluble cell staining agent capable of TPE imaging that is
well tolerated by cells.
of 30% HCl (Suprapur) and 0.5 mL of 65% HNO3 (Suprapur),
digested overnight at room temperature, and finally diluted to 10 mL
with ultrapure water in a volumetric flask.
Moreover, binding to the polymer improved the photo-
physical properties of the Ir complexes, doubling their lifetimes
and photoluminescence quantum yields, most likely because
the metal centers, being buried within lipophilic ligand cages,
are diluted and segregated into polymer tasks in an environ-
ment that is more rigid and somewhat protected from water
UV−vis absorption spectra were acquired on an Agilent model 8543
spectrophotometer at room temperature. Determination of the molar
1
absorptivity (ε) values of the MLCT transitions for 1 was performed
P
by dissolving an accurately weighted amount of 1P (balance with
readability up to 0.01 mg) in a volumetric flask and calculating the Ir
content on the basis of ICP-AES analyses.
(
and possibly from O ). In fact, several examples reported in
2
Steady state photoluminescence measurements were performed on
a Horiba−Jobin−Yvon Fluorolog spectrometer, correcting the
emission spectra for the spectroscopic sensitivity of the photo-
multiplier tube. Quantum yields were determined using [Ru(bpy) ]Cl
2
the literature have used similar approaches to shield triplet
excited states from quenchers that would prevent applications
44
in bioimaging or photocatalysis.
3
4
7
as a standard (ϕ = 0.04 in aerated water solutions). Values of the
radiative constants k have been obtained from the ϕ/τ ratio, assuming
Notably, in the present case, segregation did not hinder the
1
r
ability of the Ir complexes contained in 1 to sensitize O
P
2
the efficiency of the intersystem crossing process is equal to 1.
Lifetimes were measured by frequency-domain methods. A
frequency modulated phase fluorometer (Digital K2, ISS) was
employed using a laser diode at 378 nm as an excitation source. At
least 15 data points were acquired at logarithmically spaced
frequencies in the range of 0.3−60 MHz with a cross correlation
frequency of 400 Hz. The convenient accuracies for phase angles and
modulation ratios were 0.2° and 0.004, respectively. Lifetime
measurements were performed under the magic angle conditions,
and a 535 nm long pass filter (Andover Co.) was employed to cut light
scattering. A solution of glycogen in doubly distilled water was used as
the reference sample. Lifetime data fitting was accomplished by an ISS
routine based on Marquardt least-squares minimization. In the case of
multiple exponential components in the decay scheme, the fit of the
1
generation. The rate of O generation for 1 was lower than
2
P
that for the free complex 1M as shown by photooxidation of
DHN, but the rate difference was not dramatic. Then, 1P
maintained the capability of inducing apoptosis by irradiation
for a short time in a large fraction of irradiated cells. The
percentage of dead cells was lower than that observed in the
PDT tests sensitized by 1 , but 1 caused much less necrosis
M
P
than 1 . Thus, the experiments reported here indicate that
M
binding of the Ir sensitizers to the PhenISA copolymer allows
them to avoid the significant cellular damage caused by the free
complex 1M.
This is just one of the beneficial effects expected from the
binding of metal emitters to poly(amidoamine)s, whose
fluorescence intensity decay F(t) yields lifetime values (τ ) together
i
i
−
t/τ
22
with corresponding fractional intensities (f ): F(t) = ∑α e
and f =
i
biocompatibility has already been evidenced. Moreover, for
in vivo applications, it is expected that binding of the sensitizers
to a nanometric carrier will increase their circulating time and
favor their effective accumulation into solid tumors by the EPR
effect.
i
i
α τ /∑α τ , where α represent pre-exponential factors. The value of the
fractional intensities for each component is proportional to the molar
i
i
i
i
i
extinction coefficient (ε ), the concentration (c ), and the quantum
i
i
yield (ϕ) of the component, f ∝ ε c ϕ .
i
i
i i i
The typical error affecting lifetime measurements is about 5%, with
the exception of longer lifetimes (i.e., >500 ns), for which the error
rises to ∼10% due to the nonoptimal working frequency range
available for the amplifiers. In organic solvents, lifetime decay data
were satisfactorily described by a simple single exponential model;
The present work therefore provides proof of principle that
binding of Ir triplet emitters to poly(amidoamine)s affords
conjugates with potential as both cell imaging agents and
1
sensitizers of O generation. For improving such potential as it
2
2
fitting of the data yielded χ values on the order of 5 with no trends in
relates to clinical applications, further work will be necessary to
tailor the ligand sphere of the metal and to shed light on the
mechanism of cellular penetration and intracellular localization.
2
the residue plot (deviation of the χ from unity is mainly due to the
poor performance of the amplifiers at 0.3 MHz; their working range
spans the interval 0.1−300 MHz). As stated above, a double
exponential model had to be assumed to describe the lifetime decay
2
in water to achieve a function of merit (χ ) comparable to that
4
. EXPERIMENTAL SECTION
obtained for the organic solvents.
4
.1. Materials. 2-Methylpiperazine was purchased from Fluka and
The photochemical stability test and photoreaction with DHN in
the presence of 1 and 1 employed a Jasco V-650 spectrophotometer.
purified by sublimation, with final purity (98%) determined by
M
P
acidimetric titration. N,N′-Bis(acrylamido)acetic acid (BAC) was
The experiments were performed in a 3 mL quartz cuvette inserted
into a homemade housing that consists of a black box mounted on an
optical bench. The irradiation source was an Osram Powerstar HCI-T
with a 150 W/NDL lamp mounted on a Twin Beam T 150 R reflector
that primarily emits visible light above 400 nm, with a small emission
in the 350−400 nm range that was eliminated after placing a 390 nm
cutoff filter at the black box entrance (Figure S10 in the SI). The lamp
and reactor were separated by a fixed distance of 10 cm. The whole
setup was maintained at ambient temperature by a continuous stream
of air.
4
5
prepared following a method previously published in the literature,
and purity (98%) was determined by NMR spectroscopy and titration.
46
[
Ir(ppy) Cl] was prepared using a published method with IrCl3·
2 2
3
H O as a precursor (BASF). All the other reagents were obtained
2
from Aldrich and used as received, if not otherwise specified. THF was
distilled from Na/benzophenone using standard Schlenk techniques.
−
2
Ultrapure water (Milli-Q, Millipore, resistivity = 18 M Ω cm ) was
used for the preparation of the aqueous solutions. D O (99.9%) was
2
purchased from Aldrich and used as received; THF-d (99.9%) was
8
purchased from CIL.
4.3. Synthesis of Model Complex [Ir(ppy) (bap)]Cl (1 ·Cl). In
2
M
−2
4
.2. Instruments and Methods. NMR experiments were
a Schlenk tube, 34.7 mg of [Ir(ppy) Cl] (3.24 × 10 mmol) was
dissolved in 9 mL of anhydrous, freshly distilled THF. A solution of 4-
(bap, 16.2 mg, 6.45 × 10
mmol) in water (9 mL) was then added at room temperature. The
2 2
performed on a Bruker DRX400 spectrometer equipped with a Bruker
2
4,27
−2
5
5
mm BBI Z-gradient probe head with a maximum gradient strength of
3.5 G/cm.
(butyl-4-amino)-1,10-phenanthroline
G
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
mixture was heated at 70 °C. The starting turbid suspension rapidly
turned to a clear yellow solution, and the color observed under UV
irradiation changed from blue (fluorescence of free phen) to orange
following the formation of the luminescent Ir complex. Reaction
progress was monitored by 1H NMR spectroscopy. Heating was
stopped after 5 h, and the mixture was stirred overnight. The solvent
was evaporated under vacuum, and the product was isolated by
through a cutoff optical filter (>390 nm) for 240 min overall, and UV−
vis absorption spectra were collected after 30, 60, 90, 180, and 240 min
of irradiation. The same procedure was employed for compound 1 in
M
the solution obtained by dissolving 2.6 mg of 1 in 10 mL of an H O/
2
MeOH (85:15) mixture, and then diluting the solution 1:10 (final
−5
[1 ] = 3.3 × 10 M).
M
4.6. Photoreaction of 1M and 1 with DHN. In a volumetric
P
−
3
precipitation from THF/Et O (1:2) and dried under vacuum. Final
flask (10 mL), 3.2 mg of 1M (4.1 × 10 mmol) was dissolved in an
3
2
H
2
O/MeOH (85:15) mixture containing 6.0 mg of DHN (3.7 × 10−
mmol). This solution was diluted 1:10 to obtain a final [1 ] = 4.1 ×
10 M and [DHN] = 3.7 × 10 M. Before irradiation, the solution
was saturated with O by bubbling directly in the cuvette for 10 min.
treatment of the solid with Et O (2 mL × 3) followed by vacuum
2
evaporation to dryness gave 1 ·Cl. Isolated yield: 69%. UV−vis (H O)
M
M
2
−1
−1
−5
−4
MLCT absorptions: λ 378 nm (ε = 4690 M cm ) and 417 nm (ε
2610 M cm ). H NMR (D O, 300 K, 9.4 T): δ 8.60 (1H,
abs
−1
−1
1
=
2
2
CH(9), dd, J = 8.5, 1.0 Hz), 8.33 (1H, CH(5), d, J = 9.5 Hz), 8.30
1H, CH(7), dd, J = 5.2, 1.0 Hz), 8.18 (1H, CH(2), d, J = 5.3 Hz),
.16 (1H, CH(6), d, J = 9.5 Hz), 8.02 (2H, CH(6′), pseudo d, J = 7.9
Hz), 7.85 (2H, CH(3″), d, J = 7.6 Hz), 7.75 (2H, CH(5′), m), 7.70
1H, CH(8), dd, J = 8.5, 5.2 Hz), 7.58 (1H, CH(3), d, J = 5.3 Hz),
.40 (2H, CH(3′), pseudo t, Japp = 6.7 Hz), 7.08 (2H, CH(4″), pseudo
t, J = 7.2 Hz), 6.95 (2H, CH(5″), pseudo t, J = 7.2 Hz), 6.80 (2H,
The solution was then irradiated through a cutoff optical filter (>390
nm) for a total of 150 min, while spectra were recorded every 5 min
for the first hour, then at 90, 120, and 150 min. The same procedure
(
8
was followed for the photoreaction involving 1
(8.9 × 10 mmol of Ir) in 2.5 mL of the H
using 0.7 mg of 1
O/MeOH mixture
2
P
P
−5
(
7
−5
containing DHN ([Ir] bound to PhenISA = 3.6 × 10 M).
1
The ratio between the quantum yields for O
1 was evaluated on the basis of the following considerations. The rate
P
generation by 1 and
app
app
2
M
CH(4′), pseudo t, J = 6.7 Hz), 6.46 (2H, CH(6″), d, J = 7.4 Hz),
app
3
1
.28 (2H, CH(δ), t, J = 7.00 Hz), 2.95 (2H, CH (α), t, J = 7.30 Hz),
of the photochemical reaction (r) is expressed by eq 3, where ϕ is the
quantum yield of the reaction and I indicates the photons absorbed by
2
1
3
.82 (2H, CH (γ), m), 1.72 (2H, CH (β), m). C NMR (D O, 300
2
2
2
4
8
K, 9.4 T): δ 151.4 (CH(5)), 150.5 (CH(2)), 149.0 (CH(3′)), 138.4
the reactant.
(
(
(
(
(
CH(9)), 138.4 (CH(5′)), 131.9 (CH(6″)), 130.5 (CH(5″)), 127.9
CH(6)), 126.2 (CH(8)), 125.9 (CH(8)), 124.9 (CH(3″)), 124.1
CH(7)), 127.9 (CH(6)), 123.3 (CH(4′)), 122.6 (CH(4″)), 119.8
r = ϕI
(3)
The ratio between the quantum yields for 1O2 generation is
therefore given by eq 4.
CH(6′)). FAB-MS m/z: calcd for C H N Ir [M+], 752; found, 752
38
33
5
with the expected isotopic pattern). Elemental analysis: Anal. Calcd
for C H N IrCl: C, 57.97; H, 4.22; N, 8.89. Found: C, 57.52; H,
ϕ1M
r
1
M
I1P
38
33
5
=
×
4
.35; N, 8.73. For determining the amount of NMR silent 1M (i.e., 1M
ϕ1
r
I1M
(4)
1P
present in the form of nanoaggregates), a sample of 1M (∼1 mg) was
P
dissolved in CD Cl2 in an NMR tube, and the intensities of its
2
For monochromatic excitation, the ratio I /I can be computed by
1
P
1
M
resonances were calibrated against an internal standard (CH CN, 1
3
eq 5, where I° indicates the incident photon flux and A(1 ) and A(1 )
μL). The solution was evaporated to dryness in the NMR tube. The
P
M
indicate the absorbance of the two complexes at the excitation
wavelength.
residue was dissolved in D O; 1 μL of standard was added, and the
2
calibration was repeated, showing intensities of the resonances
corresponding to ∼75% with respect to those of the CD Cl solution.
2
2
−A(1P)
I1P
I° (1 − 10
1P
)
4
.4. Preparation and Characterization of 1 . A water solution
P
=
−
3
(
2 mL) of PhenISA (57.3 mg, 7.29 × 10 mmol of bap) was treated
with 1 mL of a THF solution containing 4.6 mg of [Ir(ppy) Cl] (4.29
10 mmol, 1.15 equiv), giving a turbid solution that became clear as
−A(1M)
I
I° (1 − 10
)
1M
1
(5)
M
2
2
−3
×
In our case, multichromatic excitation by a filtered lamp was
employed, and then integration over the entire spectral range (λ =
90−600 nm) was performed on the reduced fraction obtained by
considering that the incident photon fluxes (I°) were identical in the
the temperature rose to 50 °C. The yellow color of the solution did
not change significantly during the reaction, whereas under UV
illumination, the orange luminescence of the complex increased
progressively. After 6.5 h, the mixture was concentrated under reduced
pressure to remove most of the THF prior to purification by dialysis,
which was performed against water with a 50 kDa cutoff dialysis tube
3
two experiments. Thus, the ratio I /I was calculated to be 0.89.
1
P
1
M
As to the r /r term, the starting concentration of DHN was the
1
M
1
P
same in the two experiments (Figure 3), and therefore the ratio
between the initial reaction rates is given simply by the ratio of kobs for
1M and 1 (i.e., 2.4). This ratio should hardly be affected by the fact
that absorbance outside the linear range of the Lambert−Beer
relationship was used in the first-order plots affording k (Figures 2
and 3). In fact, the inaccuracy is expected to be the same for both 1M
and 1 because the reactant was the same (DHN), and the range of
the spanned absorbances was very similar in the two cases.
Therefore, ϕ /ϕ = 2.4 × 0.89 = 2.1.
(
Spectrapore) for four days at room temperature. Finally, the dialyzed
solution was lyophilized, affording a luminescent fluffy yellow solid.
UV−vis (H O) MLCT absorptions: λ 378 nm (ε = 4310 M cm )
and 417 nm (ε = 2300 M cm ). H NMR (D O, 300 K, 9.4 T),
−1
−1
P
2
abs
−1
−1
1
2
majority part signals: δ 8.85 (1H, NHCO(4)), 8.75 (1H, NHCO(7)),
obs
5
3
.49 (1H, CH(5)), 3.49 (1H, CH (1)), 3.46 (1H, CH (12)), 3.40−
2
2
.20 (accidentally overlapped signals: 1H, CH (13); 1H, CH(15); 1H,
P
2
CH(16)), 3.19−3.04 (accidentally overlapped signals: 1H, CH (1);
2
1
H, CH(10)), 3.03 (1H, CH (12)), 2.87 (1H, CH (13)), 2.71 (1H,
1
M
1
P
2
2
13
49
CH (16)), 2.66 (2H, CH (2)), 2.60 (2H, CH (9)). C NMR (D O,
4.7. Cell Culture and Treatment. The HeLa cell line was
cultured at 37 °C with 5% CO in complete DMEM (10% FBS, 2 mM
2
2
2
2
3
00 K, 9.4 T): δ 58.2 (CH(5)), 56.4 (CH(15)), 55.5 (CH (16)), 52.1
2
2
(
(
CH (10)), 49.3 (CH (13)), 48.9 (CH (12)), 48.1 (CH (1)), 30.6
L-glutamine, 10 mM HEPES, 100 μM nonessential amino acids, and
penicillin plus streptomycin). Cultures at ∼80% confluence were
routinely split 1:10 in 10 cm culture dishes. Cell cultures that were
70−80% confluent were incubated with either 1 or 1 in DMEM plus
1% FBS. After incubation, the cells were washed three times in PBS,
and then complete DMEM was added.
2
2
2
2
CH (9)), 30.0 (CH (2)), 13.9 (CH (17)). Elemental analysis: Anal.
2
2
3
Calcd for (C H N O ) (C H N O )
(C H ClN Ir)
22 16 2 0.053
1
3
22
4
4
0.94
24 27
5
4
0.06
(
H O) (CF SO H) : C, 43.58; H, 6.64; N, 14.00; Ir, 2.44. Found:
2 2 3 3 0.3
M
P
C, 43.83; H, 6.77; N, 14.18; Ir, 2.45. The Ir content was measured by
ICP-AES spectroscopy after a known amount of lyophilized polymer
was digested in a mixture of concentrated HCl and HNO in a 1:1
ratio for 24 h at room temperature.
4.8. TPE Measurements and Cellular Uptake. The laser source
was a mode-locked Ti:sapphire laser (Mai Tai HP, Spectra Physics)
with pulses of 120 fs full width at half-maximum and 80 MHz
repetition frequency. The optical setup was built around a confocal
scanning head (FV-300, Olympus) mounted on an upright optical
microscope (BX51, Olympus) equipped with a high working distance
3
4
.5. Photochemical Stability of 1M and 1 . A sample of 1 (0.8
P P
mg) was dissolved in 2.5 mL of an H O/MeOH (85:15) mixture,
2
−5
corresponding to [Ir] = 4.1 × 10 M. After saturation by O (bubbled
2
directly into the cuvette for 10−15 min), the solution was irradiated
H
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
objective (NA = 1.1, wd = 2 mm, 60×, water immersion, Olympus).
Nonconfocal TPE imaging was performed using the FV-300 scanning
unit after removing the largest pinhole from the pinhole wheel. The
objective simultaneously focused the laser beam on the sample and
collected the signal in epifluorescence geometry through the non-
descanned collection channels described hereafter. The non-descanned
detection system (ND unit) collected the emitted light right above the
microscope objective lens, thereby avoiding the complex optical path
back to the photomultipliers in the FV-300 scanning head. The signal
reaching the ND unit is fed to three Hamamatsu analog output
photomultipliers (HC125-02, Hamamatsu) whose 21 mm diameter
photocathode ensured the collection of most of the light during
scanning. The ND unit has been designed to minimize the distance
between the entrance pupil of the objective and the active area of the
detector. The fluorescence signal was filtered by 485/50, 535/50, and
REFERENCES
■
(
2
1) (a) DeRosa, M. C.; Crutchley, R. J. Coord. Chem. Rev. 2002,
33−234, 351−371. (b) Wang, S.; Gao, R.; Zhou, F.; Selke, M. J.
Mater. Chem. 2004, 14, 487−493. (c) Allison, R. R.; Mota, H. C.;
Shibata, C. H. Photodiagn. Photodyn. Ther. 2004, 1, 263−277.
(
2
(
d) Chatterjee, D. K.; Fong, L. S.; Zhang, Y. Adv. Drug Delivery Rev.
008, 60, 1627−1637.
2) Triesscheijn, M.; Baas, P.; Schellens, J. H.; Stewart, F. A.
Oncologist 2006, 11, 1034−1044.
3) Rhodes, L. E.; de Rie, M.; Enstrom, Y.; Groves, R.; Morken, T.;
Goulden, V.; Wong, G. A.; Grob, J. J.; Varma, S.; Wolf, P. Arch.
Dermatol. 2004, 140, 17−23.
(4) Hur, C.; Nishioka, N. S.; Gazelle, G. S. Dig. Dis. Sci. 2003, 48,
1273−1283.
(5) Kato, H. J. Photochem. Photobiol. 1998, B42, 96−99.
(6) Schuller, D. E.; McCaughan, J. S., Jr.; Rock, R. P. Arch.
Otolaryngol. 1985, 111, 351−355.
(7) Kharkwal, G. B.; Sharma, S. K.; Huang, Y. Y.; Dai, T.; Hamblin,
M. R. Lasers Surg. Med. 2011, 43, 755−767.
(8) Agostinis, P.; Berg, K.; Cengel, K. A.; Foster, T. H.; Girotti, A.
W.; Gollnick, S. O.; Hahn, S. M.; Hamblin, M. R.; Juzeniene, A.;
Kessel, D.; Korbelik, M.; Moan, J.; Mroz, P.; Nowis, D.; Piette, J.;
Wilson, B. C.; Golab, J. CA−Cancer J. Clin. 2011, 61, 250−281.
(9) For example: (a) Mroz, P.; Yaroslavsky, A.; Kharkwal, G. B.;
Hamblin, M. R. Cancers 2011, 3, 2516−2539. (b) Allison, R. R.;
Moghissi, K. Clinical Endoscopy 2013, 46, 24−29.
10) Oleinick, N. L.; Morris, R. L.; Belichenko, I. Photochem.
Photobiol. Sci. 2002, 1, 1−21 , and references therein.
11) (a) Murtinho, D.; Pineiro, M.; Pereira, M. M.; Gonsalves, A. M.;
(
6
00/40 band-pass filters to select the fluorescence light and remove
scattering and undesired autofluorescence from the sample, and it was
processed by means of Fluoview 5.0 software (Olympus).
Images were recorded at different times after the addition of the Ir
−
4
complexes (∼2 h for 1 , 100 μL of a 5.5 × 10 M solution affording a
M
−4
2
6 μM concentration in the well; ∼12 h for 1 , 100 μL of a 4.7 × 10
P
M solution affording a 22 μM concentration in the well). Two-photon
excitation at 840 nm was exploited (with an excitation power of 15
mW). Autofluorescent bleed through was verified on nonstained
samples by measuring fluorescence emission in the presence and
absence of the band-pass filter selecting the emission of the samples.
Images shown in this Article are the result of 2 kalman average scans
with 10 or 6 μs of residence time per pixel, depending on the zoom
factor. The field of view of the 512 × 512 pixel images was 157 × 157
(
2
2
μm or 79 × 79 μm , as indicated in the caption of Figure 4.
.9. Flow Cytometry. After treatment for 4 and 6 h, the adherent
cells were harvested from the culture dishes with trypsin-EDTA
Euroclone) and collected in complete DMEM, washed 3 times in
(
4
Arnaut, L. G.; Miguel, M. G.; Burrows, H. D. J. Chem. Soc., Perkin
Trans. 2 2000, 2441−2447. (b) Horiuchi, H.; Tanaka, T.; Yoshimura,
K.; Sato, K.; Kyushin, S.; Matsumoto, H.; Hiratsuka, H. Chem. Lett.
(
fluorescence-activated cell-sorting (FACS) buffer containing phos-
phate buffered solution (PBS) with 2 mM EDTA and 2% fetal bovine
serum (FBS), and then processed for the annexin V/7ADD assay
2
006, 35, 662−663. (c) Maeda, D.; Shimakoshi, H.; Abe, M.; Hisaeda,
Y. Inorg. Chem. 2009, 48, 9853−9860.
12) (a) Ikeda, A.; Sato, T.; Kitamura, K.; Nishiguchi, K.; Sasaki, Y.;
Kikuchi, J.; Ogawa, T.; Yogo, K.; Takeya, T. Org. Biomol. Chem. 2005,
, 2907−2909. (b) Mroz, P.; Tegos, G. P.; Gali, H.; Wharton, T.;
Sarna, T.; Hamblin, M. R. Photochem. Photobiol. Sci. 2007, 6, 1139−
149.
13) Tardivo, J. P.; Giglio, A. D.; Oliveira, C. S.; Gabrielli, D. S.;
(
(
Biolegend) following the manufacturer’s instructions. Stained cells
4
(
∼5 × 10 ) were acquired with a CANTO II flow cytometer (BD
3
Pharmingen) and analyzed with FlowJo software (Treestar).
4
.10. Viability Test in the Dark. HeLa cells at 80% confluency
1
(
were cultured and treated with 1M or 1 as described in section 4.7,
P
and then a trypan blue exclusion test (that selectively stains dead cells)
and a Neubauer cell counting chamber were used to count the total
cell number and the percentage of dead cells. All conditions were
tested in triplicate.
Junqueira, H. C.; Tada, D. B.; Severino, D.; Turchiello, R. F.; Baptista,
M. S. Photodiag. Photodyn. Ther. 2005, 2, 175−191.
14) Ishiyama, K.; Nakamura, K.; Ikai, H.; Kanno, T.; Kohno, M.;
(
Sasaki, K.; Niwano, Y. PLoS One 2012, 7, e37871.
15) Takizawa, S.; Aboshi, R.; Murata, S. Photochem. Photobiol. Sci.
011, 10, 895−903.
16) (a) Djurovich, P. I.; Murphy, D.; Thompson, M. E.; Hernandez,
B.; Gao, R.; Hunt, P. L.; Selke, M. Dalton Trans. 2007, 3763−3770.
b) Gao, R.; Ho, D. G.; Hernandez, B.; Selke, M.; Murphy, D.;
Djurovich, P. I.; Thompson, M. E. J. Am. Chem. Soc. 2002, 124,
(
2
(
ASSOCIATED CONTENT
■
*
S
Supporting Information
The NMR spectra, the absorption and emission data in
different solvents, the DLS size distribution profile, the stability
test, the microscopy image of the control untreated HeLa cells,
the flow-cytometry analysis, the dark toxicity data, and the
emission profile of the Xe lamp. This material is available free of
charge via the Internet at http://pubs.acs.org.
(
1
4828−14829.
(17) (a) Majumdar, P.; Yuan, X.; Li, S.; Le Guennic, B.; Jie Ma, J.;
Zhang, C.; Jacqueminde, D.; Zhao, J. J. Mater. Chem. B 2014, 2, 2838−
2854. (b) Nakagawa, A.; Hisamatsu, Y.; Moromizato, S.; Kohno, M.;
Aoki, S. Inorg. Chem. 2014, 53, 409−422. (c) Ye, R.-R.; Tan, C.-P.; He,
L.; Chen, M.-H.; Ji, L.-N.; Mao, Z.-W. Chem. Commun. (Cambridge,
U.K.) 2014, 50, 10945−10948. (d) Li, S. P.-Y.; Lau, C. T.-S.; Louie,
M.-W.; Lam, Y.-W.; Cheng, S. H.; Lo, K. K.-W. Biomaterials 2013, 34,
AUTHOR INFORMATION
■
*
Corresponding Author
E-mail: daniela.maggioni@unimi.it. Fax: +390250314405.
Phone: +390250314352.
7
519−7532. (e) Moromizato, S.; Hisamatsu, Y.; Suzuki, T.; Matsuo,
Y.; Abe, R.; Aoki, S. Inorg. Chem. 2012, 51, 12697−12706.
18) Sajoto, T.; Djurovich, P. I.; Tamayo, A. B.; Oxgaard, J.;
(
Notes
Goddard, W. A.; Thompson, M. E. J. Am. Chem. Soc. 2009, 131, 9813−
9822.
The authors declare no competing financial interest.
(
19) (a) Larson, N.; Ghandehari, H. Chem. Mater. 2012, 24, 840−
8
53. (b) Soesbe, T. S.; Kiefer, G. E.; Sherry, A. D. J. Am. Chem. Soc.
008, 130, 13854−13855. (c) Liu, L.; Li, X.; Hou, S.; Xue, Y.; Yao, Y.;
ACKNOWLEDGMENTS
■
2
The authors warmly thank Dr. Vladimiro Dal Santo for the
ICP-AES measurements.
Ma, Y.; Feng, X.; He, S.; Lu, Y.; Wang, Y.; Zeng, X. Chem. Commun.
(Cambridge, U.K.) 2009, 44, 6759−6761. (d) Kim, J. H.; Park, K.;
I
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Nam, H. Y.; Lee, S.; Kim, K.; Kwon, I. C. Prog. Polym. Sci. 2007, 32,
(43) Lo, K. K.-W.; Lee, P.-K.; Lau, J. S.-Y. Organometallics 2008, 27,
2998−3006.
1
(
6
4
031−1053.
20) (a) Matzumura, Y.; Maeda, H. Cancer Res. 1986, 46, 6387−
392. (b) Fox, M. E.; Szoka, F. C.; Frechet, J. M. Acc. Chem. Res. 2009,
2, 1141−1151. (c) Peer, D.; Karp, J. M.; Hong, S.; Farokhzad, O. C.;
(44) Strassert, C. A.; Mauro, M.; De Cola, L. Adv. Inorg. Chem. 2011,
63, 47−103.
́
(45) Ferruti, P.; Ranucci, E.; Trotta, F.; Gianasi, E.; Evagorou, G. E.;
Wasil, M.; Wilson, G.; Duncan, R. Macromol. Chem. Phys. 1999, 200,
1644−1654.
Margalit, R.; Langer, R. Nat. Nanotechnol. 2007, 2, 751−760.
(
21) (a) Ferruti, P.; Marchisio, M. A.; Duncan, R. Macromol. Rapid
(46) Sprouse, S.; King, K. A.; Spellane, P. J.; Watts, R. J. J. Am. Chem.
Commun. 2002, 23, 332−355. (b) Ranucci, E.; Ferruti, P.; Lattanzio,
E.; Manfredi, A.; Rossi, M.; Mussini, P. R.; Chiellini, F.; Bartoli, C. J.
Polym. Sci., Part A: Polym. Chem. 2009, 47, 6977−6991.
Soc. 1984, 106, 6647−6653.
47) Ishida, A.; Tobita, S.; Hasegawa, Y.; Katoh, R.; Nozaki, K. Coord.
Chem. Rev. 2010, 254, 2449−2458.
48) Balzani, V.; Carassiti, V. Photochemistry of Coordination
Compounds; Academic Press: London and New York, 1970.
49) Scherer, W. F.; Syverton, J. T.; Jey, G. O. J. Exp. Med. 1953, 97,
95−710.
(
(
22) (a) Cavalli, R.; Bisazza, A.; Sessa, R.; Primo, L.; Fenili, F.;
(
Manfredi, A.; Ranucci, E.; Ferruti, P. Biomacromolecules 2010, 11,
667−2674. (b) Ferruti, P.; Franchini, J.; Bencini, M.; Ranucci, E.;
Zara, G. P.; Serpe, L.; Primo, L.; Cavalli, R. Biomacromolecules 2007, 8,
498−1504.
23) Donghi, D.; Maggioni, D.; D’Alfonso, G.; Amigoni, F.; Ranucci,
2
(
6
1
(
E.; Ferruti, P.; Manfredi, A.; Fenili, F.; Bisazza, A.; Cavalli, R.
Biomacromolecules 2009, 10, 3273−3282.
(
24) Maggioni, D.; Fenili, F.; D’Alfonso, L.; Donghi, D.; Panigati, M.;
Zanoni, I.; Marzi, R.; Manfredi, A.; Ferruti, P.; D’Alfonso, G.; Ranucci,
E. Inorg. Chem. 2012, 51, 12776−12788.
(
25) Richardson, S.; Ferruti, P.; Duncan, R. J. Drug Targeting 1999, 6,
3
91−404.
26) Lowry, M. S.; Hudson, W. R.; Pascal, R. A., Jr.; Bernhard, S. J.
Am. Chem. Soc. 2004, 126, 14129−14135.
27) Wang, Z.; McWilliams, A. R.; Evans, C. E. B.; Lu, X.; Chung, S.;
Winnik, M. A.; Manners, I. Adv. Funct. Mater. 2002, 12, 415−419.
28) Zhao, Q.; Liu, S.; Shi, M.; Li, F.; Jing, H.; Yi, T.; Huang, C.
Organometallics 2007, 26, 5922−5930.
29) The term “solvent’s polarity” is usually employed for describing
the capacity of a solvent for solvating dissolved charged or dipolar
(
(
(
(
30
species. Although this is easy to understand qualitatively, substantial
difficulties are encountered on attempting a more precise/
comprehensive definition of “polarity” or its quantitative measures.
3
0
Many semiempirical solvent polarity scales have been developed, and
some of them are derived from spectroscopic measurements, such as
31
the well-known E (30) scale.
T
(
30) Katritzky, A. R.; Fara, D. C.; Yang, H.; Tam
Karelson, M. Chem. Rev. 2004, 104, 175−198.
31) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry,
rd ed.; Wiley-VCH Publishers: Weinheim, Germany, 2003.
32) Giordano, P. J.; Wrighton, M. J. Am. Chem. Soc. 1979, 101,
888−2897.
33) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.;
Kluwer Academic/Plenum Publishers: New York, 1999; Chapter 6, p
85.
34) Lo, K. K.-W.; Ng, D. C.-M.; Chung, C.-K. Organometallics 2001,
0, 4999−5001, and references therein.
35) For example: Mauro, M.; De Paoli, G.; Otter, M.; Donghi, D.;
D’Alfonso, G.; De Cola, L. Dalton Trans. 2011, 40, 12106−12116.
36) For example: Rampazzo, E.; Bonacchi, S.; Montalti, M.; Prodi,
L.; Zaccheroni, N. J. Am. Chem. Soc. 2007, 129, 14251−14256.
37) Yersin, H.; Rausch, A. F.; Czerwieniec, R.; Hofbeck, T.; Fischer,
T. Coord. Chem. Rev. 2011, 255, 2622−2652.
38) Wrighton, M.; Morse, D. L. J. Am. Chem. Soc. 1974, 96, 998−
003.
39) For example: (a) Lees, A. J. Coord. Chem. Rev. 1998, 177, 3−35.
b) Li, M.-J.; Chen, Z.; Yam, V. W.-W.; Zu, Y. ACS Nano 2008, 2,
̈
m, K.; Tamm, T.;
(
3
(
2
(
1
(
2
(
(
(
(
1
(
(
9
(
05−912.
40) Wu, W.; Yang, P.; Ma, L.; Lalevee
Chem. 2013, 228−231.
41) Stratakis, M.; Orfanopoulos, M. Tetrahedron 2000, 56, 1595−
615.
42) (a) Fernandez-Moreira, V.; Thorp-Greenwood, F. L.; Coogan,
M. P. Chem. Commun. (Cambridge, U.K.) 2010, 46, 186−202.
b) Mari, C.; Panigati, M.; D’Alfonso, L.; Zanoni, I.; Donghi, D.;
́
, J.; Zhao, J. Eur. J. Inorg.
(
1
(
(
Sironi, L.; Collini, M.; Maiorana, S.; Baldoli, C.; D’Alfonso, G.;
Licandro, E. Organometallics 2012, 31, 5918−5928.
J
DOI: 10.1021/ic502378z
Inorg. Chem. XXXX, XXX, XXX−XXX