Peptide-bridged dinuclear Ru(II) complex for mitochondrial targeted monitoring of dynamic changes to oxygen concentration and ROS generation in live mammalian cells

Article abstract of DOI:10.1021/ja508043q

A novel mitochondrial localizing ruthenium(II) peptide conjugate capable of monitoring dynamic changes in local O₂ concentrations within living cells is presented. The complex is comprised of luminescent dinuclear ruthenium(II) polypyridyl complex bridged across a single mitochondrial penetrating peptide, FrFKFrFK-CONH₂ (r = d-arginine). The membrane permeability and selective uptake of the peptide conjugate at the mitochondria of mammalian cells was demonstrated using confocal microscopy. Dye co-localization studies confirmed very precise localization and preconcentration of the probe at the mitochondria. This precision permitted collection of luminescent lifetime images of the probe, without the need for co-localizing dye and permitted semiquantitative determination of oxygen concentration at the mitochondria using calibration curves collected at 37 °C for the peptide conjugate in PBS buffer. Using Antimycin A the ability of the probe to respond dynamically to changing O₂ concentrations within live HeLa cells was demonstrated. Furthermore, based on lifetime data it was evident that the probe also responds to elevated reactive oxygen species (ROS) levels within the mitochondria, where the greater quenching capacity of these species led to luminescent lifetimes of the probe at longer Antimycin A incubation times which lay outside of the O₂ concentration range. Although both the dinuclear complex and a mononuclear analogue conjugated to an octaarginine peptide sequence exhibited some cytotoxicity over 24 h, cells were tolerant of the probes over periods of 4 to 6 h which facilitated imaging. These metal-peptide conjugated probes offer a valuable opportunity for following dynamic changes to mitochondrial function which should be of use across domains in which the metabolic activity of live cells are of interest from molecular biology and drug discovery.

Full text of DOI:10.1021/ja508043q

Article
pubs.acs.org/JACS
Peptide-Bridged Dinuclear Ru(II) Complex for Mitochondrial
Targeted Monitoring of Dynamic Changes to Oxygen Concentration
and ROS Generation in Live Mammalian Cells
Aaron Martin, Aisling Byrne, Christopher S. Burke, Robert J. Forster, and Tia E. Keyes*
†School of Chemical Sciences, National Biophotonics and Imaging Platform, Dublin City University, Dublin 9, Ireland
*
S Supporting Information
ABSTRACT: A novel mitochondrial localizing ruthenium(II) peptide conjugate capable
of monitoring dynamic changes in local O concentrations within living cells is presented.
2
The complex is comprised of luminescent dinuclear ruthenium(II) polypyridyl complex
bridged across a single mitochondrial penetrating peptide, FrFKFrFK-CONH (r = D-
2
arginine). The membrane permeability and selective uptake of the peptide conjugate at the
mitochondria of mammalian cells was demonstrated using confocal microscopy. Dye co-
localization studies confirmed very precise localization and preconcentration of the probe
at the mitochondria. This precision permitted collection of luminescent lifetime images of
the probe, without the need for co-localizing dye and permitted semiquantitative
determination of oxygen concentration at the mitochondria using calibration curves
collected at 37 °C for the peptide conjugate in PBS buffer. Using Antimycin A the ability of
the probe to respond dynamically to changing O concentrations within live HeLa cells
2
was demonstrated. Furthermore, based on lifetime data it was evident that the probe also
responds to elevated reactive oxygen species (ROS) levels within the mitochondria, where
the greater quenching capacity of these species led to luminescent lifetimes of the probe at longer Antimycin A incubation times
which lay outside of the O concentration range. Although both the dinuclear complex and a mononuclear analogue conjugated
2
to an octaarginine peptide sequence exhibited some cytotoxicity over 24 h, cells were tolerant of the probes over periods of 4 to 6
h which facilitated imaging. These metal-peptide conjugated probes offer a valuable opportunity for following dynamic changes
to mitochondrial function which should be of use across domains in which the metabolic activity of live cells are of interest from
molecular biology and drug discovery.
INTRODUCTION
Both because triplet states tend to be long-lived and because
■
O is triplet in the ground state, most efficient quenching is
2
Oxygen concentration within living cells and tissue is an
important marker of metabolic status. Intracellular O is the key
achieved via energy transfer to a triplet excited state acceptor.
2
The main probes reported for in-cell O sensing have therefore
2
metabolite and energy source in mammalian cells and is also a
substrate for numerous enzymatic reactions which are vital for
normal cell function. There are relatively few reliable ways to
quantitatively measure oxygen or oxygen species within cells
and fewer again capable of measuring such species at an
organelle specific level. Oxygen electrodes remain a key means
been phospors. Typically these have been metal complexes or
organic triplet state emitters such as porphyrins, both molecular
7
−13
and particle based.
However, there are relatively few probes
that have been demonstrated to respond to changing oxygen
concentration in a quantitative way and which are also capable
of being transported across the cell membrane, without the
need for permeablisation e.g. through the use of DMSO,
of measuring O in tissue and cell culture medium and although
2
quantitative, such electrodes are invasive and not suited to
1
4,15
1
ethanol or detergent in the media.
Our group have focused
cells. Although experiments are conducted using O sensing
2
on the application of Ru(II) polypyridyl complexes for in-cell
electrodes on individual organelles, they typically must be
extracted from the cell, reducing biorelevance. Electron
paramagnetic resonance (EPR) is suitable for both cells and
tissues but is highly specialized. Luminescence is currently the
method of choice in cells as it is the most practical approach.
O is typically detected through its quenching of the excited
16−18
2
sensing, including, O , membrane and protein binding.
2
The key advantages of the Ru(II) polypyridyl probes are their
photostability, oxygen sensitivity and redox and optical/
photophysical tuneability. Their emission is formally a
3
3
phosphorescence, from a MLCT excited state which is
2
populated with unity efficiency from the associated singlet
state of the given luminophore, reducing its emission intensity
19
4
,5
absorbance. This is one of their key advantages over organic
and lifetime. However, as luminescence intensity is affected
by probe concentration it is a parameter unsuited to
quantitative oxygen measurement. Therefore most lumines-
cence-based assays must rely on ratiometric approaches or
probes which usually also deactivate through fluorescence and
Received: August 6, 2014
6
measurement of the luminescence lifetime of the probe.
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1. Synthesis of Novel 5-Phenanthroline Benzoic Acid (3) and Its Ru(II) Complex Followed by Conjugation to a
a
Mitochondrial Penetrating Peptide
a
(
i) 4-Ethoxyphenylboronic acid, K CO , Pd(dppf)Cl ·DCM, dioxane/H O (3:1), 6 h, 70%. (ii, iii) DCM/MeOH (9:1), NaOH; HCl. (iv, v)
2 3 2 2
EtOH/H O (1:1), LiClO
. (vi, vii) NH -AHX-peptide HATU, DIPEA; LiClO4 (aq).
2
4 (aq)
2
tend, without metal coordination, at room temperature to have
low quantum yields of phosphorescence. Although aqueous
solubility of metal polypyridyl complexes depends on ligands
and counterion identity, relatively lipophilic complexes can be
rendered water-soluble by conjugation to peptides, PEGylation,
changes and the production of reactive oxygen species within
the mitochondria in real time can therefore provide key insights
into disease progression, cell death and the influence of therapy
2
6,27
on cellular metabolism.
in the treatment of multiple disease states.
This organelle is also a prime target
28
20−22
and other biocompatible functional groups.
Although most
The key challenge to directing a probe to the mitochondria,
which has arisen in drug delivery, is the relatively impervious
nature of the hydrophobic inner membrane. Theoretical and
experimental studies have revealed that for optimal mitochon-
drial penetration a combination of lipophilicity and multi-
reports of in-cell O sensing have focused on probes in the
2
cytoplasm, there is a strong need for probes which can be
selectively directed to the organelles. Probes which distribute
throughout cytoplasm along with various organelles are not
very suitable as it is difficult, particularly in lifetime imaging, to
identify the target organelle against the broader background of
luminophore emission. In such a situation there is a
requirement, for example, for a co-localization probe to identify
the region of interest. By using a probe capable of accumulating
in the organelle of choice, the need for such a co-localizing
probe can be avoided. This is important, as in sensing such dye
co-localization can lead to unwanted cross-talk such as
coabsorption of the same wavelength or quenching. The key
challenges therefore, are to direct probes predictably to target
29
cationic charge is required. However, the relationship
between chemical structure and entry remains somewhat ill-
30,31
defined.
There have been a number of examples of luminescent
platinum group metal luminophores which have been reported
to reach the mitochondria through manipulation of the
32−34
complex structure and their appendages.,
For example,
Barton et al. reported a systematic study of the effect of
changing the lipophilicity of rhodium-based metallo-DNA
inserters to target DNA in nucleus or mitochondria selectively
key organelles so that O can be monitored in real time at
2
23
in mammalian cells. Lo et al., recently described a family of
precise organelle locations, without the need for destructive
permeablization of the cell/organelle and without requirement
for co-localizing probe to precisely identify the organelle against
PEGylated cyclometalated Iridium complexes, these phosphor-
escent materials reached the mitochondria and exhibited strong
35
23
photocytoxocity. There have been a small number of reports
on dinuclear ruthenium complexes applied to live cells studies,
background probe emission. Mitochondria are one of the key
targets for O sensing as they are the centers of oxygen
2
36−38
both imaging and cytotoxity.
Thomas et al. recently
metabolism in the cell. There, oxygen is consumed by
cytochrome oxidase leading ultimately to ATP production. O2
concentrations in this organelle directly reflect the health of the
cell. Furthermore, the role of mitochondria across a wide range
of debilitating diseases from cancer and diabetes to neurological
disorders such as Alzheimer’s, and, Multiple Sclerosis is being
reported on dinculear probes as in-cell sensors for DNA
39
detection-based on two photon luminescent lifetime imaging.
In a very recent study reported by Keene et al., on dinuclear
ruthenium complexes bridged by n-alkyl chains showed cellular
34
uptake into L1210 murine leukemia cells. This demonstrated
24,25
increasingly recognized.
Understanding O concentration
for the first time selective uptake of a dinuclear complex where
2
B
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
it was found to preferentially locate within the mitochondria.
These dinuclear complexes were found to be cytotoxic.
aminohexanoic acid) in the presence of HATU (1-[bis-
(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]-
pyridinium 3-oxid hexafluorophosphate) and DIPEA (diisopro-
pylethylamine) in DMF.
In parallel we have been developing a targeted binuclear
probe which is bridged across a mitochondrial penetrating
peptide (MPP) with the aim of intramitochondrial oxygen
sensing. An overarching objective of our work is to exploit
peptides for truly predictable organelle targeting of the metal
complex for in-cell sensing. The peptide sequence we used,
Following the reaction, the solution was placed on ice, and a
saturated aqueous solution of LiClO was added until a bright
4
orange precipitate formed which was filtered and washed with
H O. The product was then dissolved in EtOH/H O 1:1 v/v
2
2
FrFKFrFK-CONH , was reported previously by Kelley et al. to
and the EtOH removed in vacuo. The remaining H O was
2
2
30
drive organic fluorophores to the mitochondria, and we
demonstrate here that this peptide also possesses the
physiochemical properties needed to selectively translocate a
dinuclear Ru(II) polypyridyl complex to the mitochondria,
without the need for use of solvent or other membrane
permeablization. Across the previous luminescent metal
polypyridyl complexes which have reached the mitochondria,
there have to our knowledge been no examples of their
removed by lyophilization to yield the pure dimeric conjugate
1
(HPLC, Figure S24). H NMR analysis of the conjugate
confirmed the conjugate contained two Ru(bpy) phen-Ar
2
moieties. Single mass analysis showed the molecular weight
was 2975.8765 which corresponded to the calculated mass of
1
2975.8643 g/mol for C144H150Cl N O Ru -ClO . H NMR
3
30 23
2
4
nuclear Overhauser effect spectroscopy (NOESY) was used to
determine the conjugation points of the complex to the
peptide. In order to determine which branched amine group
the second ruthenium moiety was coupled to, we studied the
aromatic signals on the phenylalanines. They are in close
application in O sensing. We report here on the first example
2
of a Ru(II) complex to report on O in the mitochondria and
2
demonstrate using Antimycin A as a mitochondrial inhibitor,
that the probe responds dynamically to O concentration
proximity to all but one of the remaining free NH groups
2
2
changes and also, it seems, to production of reactive oxygen
species (ROS). This targeted binuclear complex is potentially
of significant value across applications in cell biology and drug
discovery where understanding the metabolic status of the
mitochondria is required. This work further supports growing
evidence that peptides are a truly powerful approach to
achieving predictable cell permeability and organelle targeting
of metal complexes for in-cell sensing.
within the sequence. From NOESY, no positive contacts to any
other aromatic signals associated with the coordinating ligands
around the ruthenium center were observed (Figure S16).
Therefore, we can conclude that the attachment must be via the
least sterically hindered free amine on the terminal lysine unit
(
Scheme 2). The complexes and the conjugate were both
prepared as perchlorate salts, and both showed excellent water
solubility.
RESULTS AND DISCUSSION
Syntheses and Characterization of Ligands and
Complexes. The synthesis of 5-bromo-1,10-phenanthroline
■
Scheme 2. Schematic of the Structure of Mitochondrial
7
+
Localizing Conjugate [(Ru(bpy) phen-Ar) -MPP] ·
2 2
4(ClO4)−
(
1) was achieved by modification of a procedure developed by
40
Eisenberg et al. A high-pressure sealed reaction vessel
containing 1,10-phenanthroline, bromine, and fuming sulfuric
acid was heated at 135 °C for 23 h. Following this a Suzuki−
Miyaura protocol was implemented so as to isolate ethyl 4-
(
1,10-phenanthrolin-6-yl) benzoate (2). The conditions used
combined (1) with 4-ethoxyphenylboronic acid (1.5 equiv) and
Pd(dppf)Cl . DCM (10 mol %) as catalyst and K CO (2
2
2
3
equiv) as a base in dioxane:H O (3:1) heated to reflux for ca. 6
2
h (Scheme 1). This gave ethyl 4-(1,10-phenanthrolin-6-yl)
benzoate (2) in a 70% yield.
Typically, this reaction came to completion with minimal
byproducts in solution (confirmed by TLC). The desired
product was isolated by flash column chromatography (DCM/
MeOH 95:5) as a yellow oil which was solidified via trituration
from hot CHCl with cold pentane followed by co-solvent
3
recrystallization from MeCN/H O (1:1) to yield a white
2
powder. Base catalyzed hydrolysis of the ethyl ester
41
functionality yielded (3) as outlined in Scheme 1. The
identity of 3 was confirmed by high-resolution positive ion
mass spectrometry which revealed an m/z 301.0988, which
An analogous coupling protocol was used to conjugate a
simple cell penetrating peptide to [Ru(bpy) phen-Ar-
2
+
2+
corresponded to the peak [M + H] (calculated for [M] , m/z
COOH]
by reacting the complex with NH -AHX-
RRRRRRRR-CONH2 to yield [Ru(bpy) phen-Ar-Arg ]
2
1
0+
300.0977).
.
2
8
The ligand was coordinated to cis-Ru(bpy) Cl (4) in mixed
The octarginine polypeptide sequence has been shown to be
effective in transporting Ru(II) complexes across the cell
membrane into cytoplasmic regions with nuclear exclusion
without organelle specificity and was therefore used here to
2
2
solvent H O/EtOH (1:1 v/v), to yield compound [Ru-
2
2
+
(
bpy) phen-Ar-COOH] 2(ClO ) as a bright orange solid
2 4 2
in a 79% yield.
16
An amide coupling protocol was employed to conjugate the
compare uptake and targeting ability of the peptide.
parent complex to the peptide by reacting [Ru(bpy) phen-Ar-
Spectroscopic Properties. Figure 1 shows the absorbance
and emission spectra of the ruthenium parent complex
2
COOH]2 with NH -AHX-FrFKFrFK-CONH2 (AHX =
+
2
C
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Absorbance and emission spectra of [Ru(bpy) phen-Ar-
2
COOH]2 (red line) and [(Ru(bpy) phen-Ar) -MPP] (black line),
+
7+
2
2
respectively. Spectra were recorded at 25 μM (PBS) with emission slit
widths of 5 nm and emission excitation wavelength of 460 nm.
2
+
[
Ru(bpy) phen-Ar-COOH] and [(Ru(bpy) phen-Ar) -
2 2 2
7+
MPP] , both in PBS solution (pH 7.4). The electronic
absorption spectra for each compound show an absorption
band at 460 nm (ε ∼ 17 × 10 and 41 × 10 L mol cm
respectively) assigned to a metal to ligand charge transfer
3
3
−1
−1
(
(
MLCT) transition. The larger molar absorptivity for [(Ru-
7
+
bpy) phen-Ar) -MPP] is attributed to the presence of two
2
2
Figure 2. Stern−Volmer plots for quenching of [(Ru(bpy) phen-Ar) -
ruthenium centers at the peptide bridge. The conjugate,
2
2
MPP]7 by molecular oxygen. (A) Emission intensity versus [O ] from
+
1
0+
[
Ru(bpy) phen-Ar-Arg ] , also displays a band at 460 nm
2
2
8
3
−1
−1
saturation to deaeration. Stern−Volmer plot of emission intensity as a
but with a similar ε (19 × 10 L mol cm ) to that of the
parent presursor. When excited into the MLCT absorbance, an
intense emission band for [Ru(bpy) phen-Ar-COOH] was
function of [O ] at 18 °C. (B) Stern−Volmer plot of emission lifetime
2
as a function of [O ] at (i) 18 °C and (ii) 37 °C.
2+
2
2
observed centered around 604 nm with a quantum yield of
3
0
.046 ± 0.003 which is slightly lower than that of the peptide
thermally accessible triplet metal centered eg* state which is
3
bound species [Φ = 0.056 ± 0.002 for [(Ru(bpy) phen-Ar) -
promoted from the MLCT state in Ru(II) complexes
containing polypyridyl ligands, though it can be alleviated by
the introduction of strong σ-donating ligands.
2
2
MPP]7 and 0.067 ± 0.005 for [Ru(bpy) phen-Ar-R ]
+
10+
.
2
8
2
+
42,43
Under aerated conditions, [Ru(bpy) phen-Ar-COOH]
As an
2
exhibits a lifetime of 455 ± 11 ns in aqueous PBS buffer
example of the sensitivity of the photophysical properties of
7+
which is increased to 780 ± 9 ns after deaeration by N at room
[(Ru(bpy) phen-Ar) -MPP] , the emission lifetime is 698 ± 6
2
2 2
temperature.
ns in aerated PBS and 1.23 μs ± 10 ns in deaerated PBS at
room temperature (18 °C), whereas at 37 °C, in aerated media,
Temperature-Dependent Response of [(Ru-
7+
7+
(
bpy) phen-Ar) -MPP] Luminescence to Oxygen Con-
the lifetime [(Ru(bpy) phen-Ar) -MPP] was substantially
2
2
2
2
7+
centration. The response of [(Ru(bpy) phen-Ar) -MPP] to
lower at 458 ± 7 ns in aerated and 948 ± 6 ns in deaerated
2
2
O concentration was explored by examining the emission
media. Comparison of the intensity and lifetime based Stern−
2
intensity of this complex with varying [O ] in PBS buffer. The
Volmer plots at room temperature confirms K (the slope) is
2
SV
O concentration of each solution was measured independently
the same in each case, consistent with dynamic quenching.
2
using a Presense fiber optic O probe which recorded saturation
From KSV and the lifetime values in the absence of O , we
2
2
9
−1
O concentration in PBS buffer (pH 7.4) as 220 μM. The
obtain oxygen quenching rate constants, k , of 2.5 × 10 M s
q
9
2
−1
emission spectrum was recorded at O saturation, and a 3 mL
and 4.64 × 10 M s for 18 and 37 °C, respectively. The latter
2
solution was then deaerated under N for 15 min, wherein a 0
value is slightly lower than, but consistent with, the O
2
2
9
−1
μM concentration of O was recorded within the sample. As
quenching rate reported by Demas et al. (3.2 × 10 M s )
2+
44
2
expected there was an increase in emission intensity (Figure
at 20 °C for O quenching of [Ru(bpy) ] . The somewhat
2 3
S38). The solution was gradually reaerated, the O concen-
lower O sensitivity seen here may be attributed to lower access
2
2
tration recorded by fiber optic probe, and the emission spectra
of O to the complex because of the bulky peptide as noted
2
7+
18
and lifetime collected from [(Ru(bpy) phen-Ar) -MPP] as a
previously. From the 37 °C plot we now have a calibration
2
2
function of O concentration to yield the linear Stern−Volmer
curve from which we can estimate O concentration within the
2
2
plot (eq 1, Figure S39).
cell from fluorescence lifetime imaging.
Figure 2B shows the Stern−Volmer plots of luminescent
Cell Uptake and Mitochondrial Staining. The uptake of
2
+
lifetime versus O concentration for [(Ru(bpy) phen-Ar) -
[Ru(bpy) phen-Ar-COOH] and [(Ru(bpy) phen-Ar) -
2
2
2
2
2
2
MPP]7 by O at 18 °C (as for the I Figure 2A) and also at 37
+
MPP] by HeLa and CHO cells were compared across a
range of dye concentrations (10−100 μM) in PBS buffer. Both
cell lines failed to take up the parent precursor across any of
these concentrations. Such poor cellular penetrability by
ruthenium complexes, in the absence of permeablization, e.g.,
7+
2
°
C. It is important to note the temperature sensitivity of the
Stern−Volmer plot. Ruthenium polypyridyl complexes typically
exhibit temperature-dependent luminescence, with decreases in
quantum yield and lifetime commonly observed with increasing
temperature. This behavior is attributed to the population of a
4
5
with organic solvent has been noted previously. In contrast
D
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
7
+
Figure 3. (A,B) Confocal luminescence Imaging of [(Ru(bpy) phen-Ar) -MPP]
in HeLa cells. (C,D) Confocal luminescence Images of
2
2
10+
[
Ru(bpy) phen-Ar-Arg ] in HeLa cells. (E) Co-localization studies were carried out using 500 nM of the mitochondrial marker MitoTracker Deep
Red. Cells were treated with [(Ru(bpy) phen-Ar) -MPP] for 2 h and MitoTracker Deep Red for 30 min. [(Ru(bpy) phen-Ar) -MPP] is shown in
green, [(Ru(bpy) phen-Ar) -MPP] in red, and co-stained regions are shown in yellow on the left. (F) Cross-section image describing the co-
localization of [(Ru(bpy) phen-Ar) -MPP] and MitoTracker Deep Red in HeLa cell shown in green, and its corresponding plot (G)
2 8
7+
7+
2
2
2
2
7+
2
2
7+
2
2
7+
demonstrating both compounds localize within the mitochondria and are nuclear excluding. [(Ru(bpy) phen-Ar) -MPP] was excited using 488
2
2
nm argon laser and emission collected using long pass 630 nm filter set. MitoTracker Deep Red was excited with 633 nm HeNe laser and emission
collected using long pass 630 nm filter (G); (scale bar: A, C, D, and F 10 μM; and B 5 μM).
7
+
[
(Ru(bpy) phen-Ar) -MPP] was taken up rapidly and
cytoplasm, furthermore the conjugate is nuclear excluding
(Figure 3A,B). Similar nuclear exclusion was observed for
2
2
irreversibly by the cells in a concentration-dependent manner.
5 μM in the contacting solution was found to be optimum
10+
7
[Ru(bpy) phen-Ar-Arg ] when incubated with HeLa cells at
2 8
concentration for the uptake and imaging in terms of a balance
between emission intensity and cytotoxicity. Interestingly,
conjugate uptake was found to be temperature dependent. It
was found for example that when incubated with HeLa cells at
75 μM for 2 h. It is strongly membrane permeable but
distributes within the cytoplasm, excluding the nucleus (Figure
3C,D) and with no evidence for mitochondrial uptake. To
conclusively confirm localization of [(Ru(bpy) phen-Ar) -
MPP] at the mitochondria, HeLa cells incubated with 75
μM of the conjugate for 2 h at 37 °C were dual stained with
MitoTracker Deep Red. The MitoTracker showed strong co-
2
2
7
+
7+
4
°C, [(Ru(bpy) phen-Ar) -MPP] failed to cross the cell
2
2
membrane. Instead, the conjugate appears to localize in the
membrane even after extensive incubation (Figure S37). This
result indicates that uptake is an activated process such as
endocytosis.
localization with the conjugate, in Figure 3E, [(Ru(bpy) phen-
2
7+
Ar) -MPP] is seen in green, MitoTracker in red, and their co-
2
Detailed uptake and localization studies were carried out on
localization in yellow. Figure 3F expands a single cell which
clearly shows superimposed emissions of [(Ru(bpy) phen-Ar) -
7
+
[
(Ru(bpy) phen-Ar) -MPP] using confocal laser scanning
2 2
2
2
7+
microscopy (CLSM). The initial uptake is fast; the conjugate
enters the cells within 5 min of its addition to the contacting
solution (Figure S25). After an incubation period of 1 h it
appeared initially to concentrate within the cytoplasm and then
localizes strongly in the mitochondria, and localization is
complete at 37 °C following in 2 h incubation, both in the
presence and absence of light. It is evident that the dye strongly
preconcentrates inside the mitochondria, although lumines-
cence imaging shows that some complex remains within the
MPP] and the MitoTracker at a single cell and confirms they
are co-localized. Indeed an x-scan of emission intensity from
each probe, shown in Figure 3F, indicates that both are at a
coincident location and nuclear excluding.
7
+
However, it appears that [(Ru(bpy) phen-Ar) -MPP] is
2
2
even more narrowly confined to the mitochondrial region than
MitoTracker, suggesting perhaps that Mitotracker is more
confined to the membrane region of the mitochondria than its
interior.
E
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
This result indicates that this mitochondrial directing peptide
sequence can be used to direct metal complexes to the
mitochondria selectively with a much lower background
concentration of the probe remaining in the cytoplasm. This
is an important result as the background contributions from
In order to assess if the probes could respond dynamically to
changes in mitochondrial function, HeLa cells were loaded with
7+
[(Ru(bpy) phen-Ar) -MPP] and then treated with Antimycin
2
2
A. Antimycin A is a product comprised of Antimycin A1 and A3
derived from Streptomyces kitazawensis which has been widely
7
+
46−49
[
(Ru(bpy) phen-Ar) -MPP] residing in cytoplasm is readily
studied using HeLa cells.
It is a classical mitochondrial
2
2
distinguished from the mitochondria making localized measure-
uncoupler, which predictably changes oxygen levels within the
mitochondria through inhibition of electron transport between
cytochromes b and c at Complex III in mitochondria. The
consequences of this inhibition are an increase in O2
concentration at the mitochondria to ambient levels as its
consumption at the mitochondria is terminated and eventually
ment of luminescent lifetime and therefore O concentration at
2
the mitochondria potentially relatively straightforward.
Fluorescence Lifetime Imaging and Oxygen Depend-
ence Studies in HeLa cells. This was borne out in the
fluorescence lifetime imaging experiments shown in Figure 4,
5
0,51
an increase in ROS and decrease in ATP production.
Here, Antimycin A was added to live HeLa cells following
7
+
their incubation with [(Ru(bpy) phen-Ar) -MPP] for 2 h.
2
2
Using FLIM, the lifetime of the conjugate was monitored at 10
min intervals over 90 min following Antimycin administration.
Examples of the FLIM images collected over this period along
7
+
with the lifetime distributions of [(Ru(bpy) phen-Ar) -MPP]
2
2
from the mitochondria are shown in Figure 5. Figure 6 shows a
Figure 4. FLIM image (A) and its corresponding lifetime distribution
7+
(
B) of [(Ru(bpy) phen-Ar) -MPP] , in the mitochondria in HeLa
2 2
cells. Cells were incubated with 75 μM of the conjugate in the dark for
2
h. FLIM image and lifetime were recorded exciting with 405 nm laser
line and emission collected using 530 nm filter set.
left, where the mitochondria are clearly distinguished from the
FLIM signal without the need for costaining. At 37 °C in
aqueous, air saturated PBS solution, τ of [(Ru(bpy) phen-Ar) -
Figure 5. Luminescence lifetime distributions of [(Ru(bpy) phen-
2
7+
Ar) -MPP] from luminescence lifetime imaging of HeLa cells in
2
2
2
response to the addition of Antimycin A. HeLa cells were incubated
MPP]7 was recorded as 458 ± 6 ns. Figure 4 shows the
+
7+
with [(Ru(bpy) phen-Ar) -MPP] for 2 h in the dark (A), and
2
2
distribution of luminescent lifetimes of [(Ru(bpy) phen-Ar) -
2
2
Antimycin A (200 μM/mL) was added. Cells were incubated for 10
min with Antimycin, and FLIM recorded (B). Cells were incubated for
a further 90 min, where total exposure time was 100 min, and FLIM
lifetime recorded (C) (n = 2). FLIM was carried out by accumulating
photon counts over 10 min in each case under identical, wavelength,
and laser intensity conditions.
MPP]7 within the mitochondria of live HeLa cells. From the
color distribution chart, the lifetime of the conjugate within the
mitochondria appears to be 525 ± 10 ns which, based on the
calibration plot at 37 °C, represents an intracellular O2
concentration of ≈183 μM. (The surrounding solution was
+
air saturated, confirmed by Fiber optic O probe and the
2
lifetime of the complex in this solution). The lower O2
concentration at the mitochondria is expected for normally
metabolizing cells.
Monitoring the Respiratory Responses of Cells to
Mitochondrial Uncoupler Antimycin A. Such precise
plot of the luminescent lifetime from the mitochondria as a
function of incubation time for HeLa cells with Antimycin A as
well as a control, in which the lifetime of [(Ru(bpy) phen-Ar) -
2
2
MPP]7+ at the mitochondria over time was measured from
HeLa cells without addition of Antimycin A. The control
shows, within experimental error, no change in luminescent
lifetime over time. The results from the Antimycin A treated
cells show an initial decrease in τaverage from the cell
mitochondria from 525 to 423 ns after 10 min incubation of
the cells with Antimycin A (200 μg/mL) to the cells. This
mitochondrial localization with O quantitation has, to our
2
knowledge, not been demonstrated previously. Others have
achieved cellular penetration with an O sensitive probe and
2
realized perinuclear localization but with a broad cellular
14
diffusion pattern including mitochondrial uptake.
F
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
To assess if the toxicity was specifically due to the properties
of the metal complex or due to specifically to its localization
2
+
within the mitochondria, [Ru(bpy) phen-Ar-COOH] was
2
conjugated to a simple cell penetrating peptide (CPP) octa-
10+
arginine to produce [Ru(bpy) phen-Ar-Arg ] . Polyarginine
2
8
peptides have been shown to deliver complexes to the
cytoplasmic region of the cell, with no specific target.
10+
[
Ru(bpy) phen-Ar-Arg ] was incubated at varying concen-
2 8
trations with HeLa cells for 24 h and cell viability assessed using
the Resazurin (Alamar Blue) reagent. Figure S27 shows that
10+
[
Ru(bpy) phen-Ar-Arg ] was cytotoxic, there was 0% cell
2 8
viability at concentrations in HeLa cells incubated with
concentrations of 150 and 200 μM following 24 h (IC50 = 44
±
0.7 μM). This suggests that it is the parent complex which is
fundamentally toxic toward the cells and not necessarily a
feature of its subcellular localization. As the polyarginine
delivers the complex in a nonselective manner to the cytoplasm
and distributes it widely without specific preconcentration
within organelles such as the endoplasmic reticulum, lysosomes,
and mitochondria. (Figures 3C,D). Furthermore, the behavior
of this complex contrasts strongly with the behavior of peptide
Figure 6. Effects of Antimycin A on the lifetime of [(Ru(bpy) phen-
2
7+
Ar) -MPP] over 100 min at 37 °C. FLIM images were acquired every
2
1
0 min following exposure of HeLa cells to the drug. The control
7+
sample of [(Ru(bpy) phen-Ar) -MPP] only was measured without
2
2
any addition of mitochondrial uncoupling reagent. Sample excited
using 405 nm laser and emission collected using 530 nm filter set.
2
+
conjugates of [Ru(bpy) PIC] complex (PIC is 2-(4-
decrease is indicative of O diffusion from the cytoplasm/media
2
2
carboxyphenyl)-imidazo-[4,5-f ][1,10]-phenanthroline). The
which in the absence of O consumption by Complex III
2
octa-arginine conjugate of this complex showed very similar
returns O close to saturation levels. However, this decrease in
2
10+
distribution to [Ru(bpy) phen-Ar-Arg ]
cytotoxicity.
but showed low
lifetime continued until a plateau was reached at approximately
2
8
17
228 ns (τaverage) after 100 min (Figures 5 and 6). It is important
For the purposes of imaging/O determination, we found
to note that in a control experiment Antimycin A had no effect
2
that when incubated in the dark for 2 h, the cells are viable,
toxicity is induced only over periods exceeding 4 h. We probed
on the luminescence intensity or lifetime of [(Ru(bpy) phen-
2
7+
Ar) -MPP] in solution. Given the dramatic decrease in
2
7
+
the effect of exposure of [(Ru(bpy) phen-Ar) -MPP]
luminescent lifetime of the conjugate over extended incubation
2
2
incubated cells to a light source to see if light induced
cytotoxicity, Figure 7.
times with Antimycin A, and the fact that [O ] cannot exceed
2
saturation, there must exist a separate quenching pathway for
7
+
phen-Ar)
-MPP]7+ was
2 2
[
(Ru(bpy) phen-Ar) -MPP] . This quenching pathway is
The photocytoxicity of [(Ru(bpy)
followed in HeLa cells by assessing the uptake of DRAQ 7 in
2
2
attributed to the ROS generated, induced by inhibition of
mitochondrial function. Superoxide anion O2• , H O , and
−
7+
[(Ru(bpy)
phen-Ar) -MPP] loaded cells over time under
2
2
2
2
hydroxyl radical have all been shown to be increased in
visible irradiation, Figure 7A−D shows the results. After 20 min
50
7+
Antimycin A treated mitochondria. All three, on the basis of
their redox potentials, are expected to be capable of quenching
continuous irradiation of [(Ru(bpy) phen-Ar) -MPP] loaded
2 2
cells at 488 nm we observed significant cell death, even before
the complex had time to accumulate strongly within the
mitochondria. This is consistent with the fundamental
photocytoxicity of the complex rather than cytotoxicity
associated with its localization. An area of the same sample
that was not exposed to light was examined as a control, and no
damage to the cells in this region was detected, indicating that
the uptake of DRAQ 7, i.e., cytotoxicty was indeed induced by
irradiation, (Figure 7E,F). Further control experiments show
the cells are still viable at the scanning intensities used during
illumination in the absence of dye. No DRAQ 7 entered the
cells up to and after 50 min of scanning (Figure S26). This
confirmed that the cell death is due to the presence of
7
+
the excited of [(Ru(bpy) phen-Ar) -MPP] , although the
2
2
quenching rate by hydrogen peroxide would be expected to be
5
2
relatively slow. It is therefore reasonable to assume that in the
mitochondria in addition to responding to O concentration
2
the conjugate is also reporting on ROS generation.
Cytotoxicity and Phototoxicity. Consideration of the
7
+
cytotoxicity of [(Ru(bpy) phen-Ar) -MPP] is important in
2
2
assessing their suitability as O /ROS probes. Cytotoxicity
2
studies were carried out using the Resazurin (Alamar Blue)
assay to assess cell viability after 24 h incubation with
2
+
[
Ru(bpy) phen-Ar-COOH] and [(Ru(bpy) hen-Ar) -
2
2
p
2
7+
MPP] over a concentration range of 0−200 μM. The data
7
+
are shown in Figure S27. It was found that [Ru(bpy) phen-Ar-
[(Ru(bpy)
2
phen-Ar)
2
-MPP] .
2
COOH]2 exhibited no cytotoxic effects on the cells, with 75%
cell viability up to 200 μM. This is expected; as described
above, this complex is not effective at penetrating the cell
membrane, and uptake is negligible. However, viability of cells
was only 15% when incubated with [(Ru(bpy) phen-Ar) -
+
-MPP]7+ is not toxic to the cells
As [(Ru(bpy)
phen-Ar)
2
2
when incubated for 2 h in the dark, we were interested to assess
if phototoxicity was even greater when the dye had accumlated
in the mitochondria. HeLa cells were incubated with [(Ru-
7+
(bpy) phen-Ar) -MPP] (75 μM) for 2 h in the dark and then
2
2
2
2
7+
MPP] , at 200 μM for 24 h. The cytotoxicity was dose
dependent with 40% viability observed after 24 h at our
working concentration of 75 μM (IC = 47 ± 1.1 μM). A time-
exposed to light source. The cells were continuously irradiated
at 488 nm under 0.64 μW/cm− laser power. Cellular toxicity
under these conditions was evident after 20 min exposure time
(Figure 7E,F), which is comparable to the time taken for
photocytoxicity before the dye reached the mitochondria.
Again, the photodamage was only observed in the areas
exposed to light. This was similar to the effects seen during
3
5
0
dependent cytotoxicity study was carried out on [(Ru-
7
+
(
bpy) phen-Ar) -MPP] at 75 μM over 1−7 h to assess cell
viability during our working incubation time. After 5 h 40% of
the cells remained viable (Figure S28).
2
2
G
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
22
death. Similarly an increase in luminescence intensity was also
recorded after the complex induced photo damage to the cells.
It is important to note that although the photocytotoxicty of
the complexes represents an issue for conventional confocal
imaging, FLIM uses low energy, pulsed irradiation, and so
photocytotoxity is not a significant issue in such measurements.
CONCLUSIONS
■
A luminescent dinuclear ruthenium(II) probe bridged across a
mitochondrial penetrating peptide sequence, FrFKFrFK,
exhibited excellent cell permeability with rapid, irreversible
uptake, and accumulation in test cells in common growth media
over approximately 20 min. The complex concentrates very
precisely inside the mitochondria over a 2 h period. This was
confirmed from colocalisation studies which showed that the
complex was strongly confined to the mitochondria with much
weaker staining of the remainder of the cell and exclusion from
the nucleus.
The probe responds dynamically to changes in O2
concentration as demonstrated by a series of in vitro calibration
studies of both emission intensity and lifetime versus [O2]
(
μM) at 18 °C and at 37 °C. As expected for ruthenium
complexes with modestly σ donating ligands, the calibration
curves were temperature sensitive. The ability of the probes to
respond to O changes within the mitochondria of live HeLa
2
7
+
Figure 7. Phototoxicity of [(Ru(bpy) phen-Ar) -MPP] in HeLa
cells was explored by artificially altering the metabolic state of
the cells using a classical mitochondrial uncoupling reagent,
Antimycin A, and assessing the lifetime profile of the probe
2
2
cells. In both experiments, row 1 shows both channels for
(Ru(bpy) phen-Ar) -MPP] and DRAQ 7, and row 2 shows the
DRAQ 7 channel only to record toxicity (scale bar 20 μM). Immediate
uptake of [Ru(bpy) phen-Ar] -MPP is evident in images collected
after 5 min exposure in (A), but no toxicity to cells is observed. After
0 min some toxicity is seen (B), and after 25 min DRAQ 7 has
entered the nucleus of all the cells indicating cell death (C). After 35
min of exposure to [Ru(bpy) phen-Ar] -MPP in a new location of
the same sample that has not been exposed to the laser no entry of
DRAQ 7 observed (D). Lastly, HeLa cells were incubated with 75 μM
(Ru(bpy) phen-Ar) -MPP] for 2 h in the dark, washed, and DRAQ
added (E). Cells were exposed to continuous irradiation and DRAQ
entered the nucleus after 20 min scanning (F). Laser power used was
7+
[
2 2
within the mitocondria using fluorescence lifetime imaging.
7+
7+
2
2
[
(Ru(bpy) phen-Ar) -MPP] responded rapidly to changing
2
2
O concentration, and after 10 min exposure to Antimycin A
2
2
the [O ] returned to what seemed to be air saturated levels.
2
7+
Further incubation to a 100 min exposure time led to τ of 228
ns, which is shorter than the luminescence lifetime in air
saturated solution. This indicates that the complex is being
quenched by ROS generated from the action of the Antimycin
A.
The complexes were moderately cytotoxic in the dark, and
under conditions used for imaging they remained more than
2
2
7+
[
2 2
7
7
0
−
3
.64 μW/cm .
4
0% viable over 24 h. However, under continuous photo-
uptake phototoxicity studies prior to probe localization. As the
cells were viable for up to 7 h of incubation in the dark (Figure
irradiation at 488 nm they showed significant photocytotoxicity.
Interestingly, conjugating the parent complex to cell penetrat-
ing octaarginine, results in comparable membrane permeability
but a wider distribution of the dye within the cytoplasmic
region of the cell. It was also nuclear excluding, and there was
7+
S28), these results suggest that [(Ru(bpy) phen-Ar) -MPP]
2
2
is photocytotoxic. Furthermore, it is evident that [(Ru-
7
+
(
bpy) phen-Ar) -MPP] does not need to be located within
2 2
the mitochondria to induce photo damage upon the cells.
no evidence for specific localization of [Ru(bpy) phen-Ar-
2
7+
10+
Interestingly luminescence from [(Ru(bpy) phen-Ar) -MPP]
Arg8] within the mitochondria. Yet, this conjugate was found
2
2
7
+
showed a 3-fold emission intensity increase upon photodamage
to the cell. Likewise, after illumination on live uptake,
phototoxicity toward the cell resulted in a 5-fold increase in
to be more cytotoxic than the [(Ru(bpy) phen-Ar) -MPP]
2 2
under the same conditions which suggests it is not specifically
damage to the mitochondria which is responsible for
cytotoxicity/photocytotoxcity of these complexes.
7
+
emission intensity of [(Ru(bpy) phen-Ar) -MPP] from
2
2
within the cell. This is compared to the intensity when
Overall this work highlights the value in combining
luminescent metal complexes with the targeting capability of
MPPs in directing metal complexes to the mitochondria and
incubated without light exposure. A similar effect was noted for
2+
[
Ru(phen) ] , which when illuminated in live cells caused cell
3
Table 1. Photophysical Data of All Compounds Recorded in PBS (pH 7.4)
compound
Φ (%)
τ (ns) 18 °C aerated τ (ns) 18 °C deaerated τ (ns) @ 37 °C aerated τ (ns) @ 37 °C deaerated
2
+
0.046 ± 0.003b
[
[
[
Ru(bpy) phen-Ar-COOH]
455 ± 11
579 ± 11
698 ± 6
780 ± 9
1.02 μs ± 9 ns
1.23 μs ± 10 ns
−
−
−
−
2
1
0+
0.067 ± 0.005b
Ru(bpy) phen-Ar-Arg ]
2
8
7
+
a
(Ru(bpy) phen-Ar) -MPP]
0.053 ± 0.002
458 ± 7
948 ± 6
2
2
aStandard used [Ru(bpy)3]2+ 0.04 ± 0.002
H
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
confirms overall the value of peptides in targeted imaging by
metal complexes. Although toxicity may be an issue over longer
times, the probes are potentially powerful tool for short-term
dynamic studies of mitochondrial, cellular function both for
monitoring of disease progression, and in in vitro studies of
therapy, e.g., in assessment of mitochondrial effectors.
(17) Cosgrave, L.; Neugebauer, U.; Forster, R. J.; Keyes, T. E. Chem.
Commun. 2010, 46, 103−105.
(
18) Adamson, K.; Dolan, C.; Moran, N.; Forster, R. J.; Keyes, T. E.
Bioconj. Chem. 2014, 25, 928−944.
(19) Balzani, V., Bergamini, G.; Campagna, S.; Puntoriero, F. Topics
in Current Chemistry, Photochemistry and Photophysics of Coordination
Compounds; Balzani, V., Campagna, S., Eds.; Springer-Verlag: Berlin,
2
(
007; Vol. 280, pp 1−36.
20) Dolan, C.; Moriarty, R.; Lestini, E.; Devocelle, M.; Forster, R. J.;
Keyes, T. E. Inorg. Biochem. 2013, 119, 65−74.
21) Louie, M. W.; Liu, H. W.; Lam, M.H-C.; Lam, Y.-W.; Lo, K.K-
W. Chem.Eur. J. 2011, 17, 8304−8308.
22) Jankte, D.; Cokoja, M.; Pothig, A.; Herrmann, W. A.; Kuhn, F.
ASSOCIATED CONTENT
■
*
S
Supporting Information
(
All experimental details and characterization of novel materials,
spectra, HPLC traces, and additional biological data. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(
E. Organometallics 2013, 32, 741−744.
(
(
(25) Tiede, L. M.; Cook, E. A.; Morsey, B.; Fox, H. S. Cell Death Dis.
2
23) Dobrucki, J.W. J. Photochem. Photobiol. 2001, 127, 136−144.
24) Duchen, M. R.; Szabadkai, G. Essays Biochem. 2010, 47, 115−37.
AUTHOR INFORMATION
■
011, 2, e246.
26) Galluzzi, L.; Larochette, N.; Zamzami, N.; Kroemer, G.
Corresponding Author
Tia.Keyes@dcu.ie
(
Oncogene 2006, 25, 4812−4830.
Notes
(27) Green, K.; Brand, M. D.; Murphy, M.p. Diabetes 2004, 53, 110−
1
18.
The authors declare no competing financial interest.
(
(
28) Armstrong, J. S. J. Pharmacol. 2007, 151, 1154−1165.
29) Smith, R. A. J.; Porteous, C. M.; Gane, A. M.; Murphy, M. P.
ACKNOWLEDGMENTS
■
Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 5407−5412.
Based upon works supported by Science Foundation Ireland
under [10/IN.1/B3025] and [12/TIDA/B2382]. The National
Biophotonics and Imaging Platform, Ireland, was established
through funding received by the Irish Government’s PRTLI,
Cycle 4, and EU Structural Funds Programmes 2007−2013.
C.S.B. and T.E.K. gratefully acknowledge the Irish Research
Council for PhD scholarship funding. Professor R O’Kennedy,
DCU, is gratefully acknowledged for access to cell culture
facilities.
(30) Horton, K. L.; Stewart, K. M.; Fonseca, S. B.; Guo, Q.; Kelley, S.
O. Chem. Biol. 2008, 15, 375−382.
(31) Carneado-Fernandez, J.; Van Gool, M.; Martos, V.; Castel, S.;
Prados, P.; Mendoza, J. D.; Giralt, E. J. Am. Chem. Soc. 2005, 127,
8
69−874.
32) Komor, A. C.; Schneider, C. J.; Weidmann, A. G.; Barton, J. K. J.
Am. Chem. Soc. 2012, 134, 19223−19233.
33) Chen, T.; Mei, W.-J.; Wong, Y.-S.; Liu, J.; Liu, Y.; Xie, H.-S.;
Zheng, W.-J. Med. Chem. Comm. 2010, 1, 73−75.
34) Pisani, M. J.; Weber, D. K.; Heimann, K.; Collins, J. G.; Keene,
(
(
(
R. F. Metallomics 2010, 2, 393−396.
REFERENCES
(35) Li, P. Y.; Lau, C. T. S.; Louie, M. W.; Lam, Y. W.; Cheng, S. H.;
Lo, K. K. W. Biomaterials 2013, 34, 7519−7532.
■
(
(
(
1) Yotter, R. A.; Wilson, D. M. IEEE Sens. J. 2004, 4, 412.
2) Li, Z.; Graham, B. H. Methods Mol. Biol. 2012, 837, 63−72.
3) Diepart, C.; Verrax, J.; Calderon, P. B.; Feron, O.; Jordan, B. F.;
(36) Baggaley, E.; Gill, M. R.; Green, N. H.; Turton, D.; Sazanovich,
I. V.; Botchway, S. T.; Smythe, C.; Haycock, J. W.; Weinstein, J. A.;
Thomas, J. A. Angew. Chem. 2014, 53, 3367−3371.
Gallez, B. Anal. Biochem. 2010, 396, 250−256.
4) Morris, K. J.; Roach, M. S.; Xu, W.; Demas, J. N.; DeGraff, B. A.
Anal. Chem. 2007, 79, 9310−9314.
5) Abdel-Shafi, A. A.; Ward, M. D.; Schmidt, R. Dalton Trans 2007,
517−2527.
6) Fercher, A.; Borisov, S. M.; Zhdanov, A. V.; Klimant, I.;
Papkovsky, D. B. ACS Nano 2011, 7, 5499−5508.
7) Vanderkooi, J. M.; Wright, W. W.; Erecinska, M. Biochemistry
990, 29, 5332−5338.
8) Yoshihara, R.; Yamaguchi, T.; Hosaka, Y.; Takeuchi, M.; Tobita,
T. S. Angew. Chem. 2012, 51, 4148−4151.
9) Zhdanov, A. V.; Dmitriev, R. I.; Papkovsky, D. B. Cell. Mol. Life
Sci. 2010, 68, 903−917.
10) Dmitriev, R. I.; Papkovsky, D. B. Cell. Mol. Life Sci. 2012, 69,
025−2039.
11) Choi, N. W.; Verbridge, S. S.; Williams, R. M.; Chen, J.; Kim, J.
(37) Gill, M. R.; Garcia-Lara, J.; Foster, S. J.; Smythe, C.; Battaglia,
(
G.; Thomas, J. A. Nat. Chem. 2009, 1, 662−667.
(38) McDonnell, U.; Kerchoffs, J. M. C. A.; Castineiras, Rosa P. M.;
(
2
(
Hicks, M. R.; Hotze, Anna C. G.; Hannon, M.; Rodger, A. Dalton
Trans. 2008, 667−675.
(39) Baggaley, E1; Gill, M. R.; Green, N. H.; Turton, D.; Sazanovich,
I. V.; Botchway, S. W.; Smythe, C.; Haycock, J. W.; Weinstein, J. A.;
(
1
(
Thomas, J. A. Angew. Chem., Int. Ed. Engl. 2014, 24;53 (13), 3367−71.
(40) Hissler, M.; Connick, W. B.; Geiger, D. K.; McGarrah, J. E.;
Lipa, D.; Lachicotte, R. J.; Eisenberg, R. Inorg. Chem. 2000, 39, 447−
457.
(41) Martin, A.; Moriarty, R. D.; Long, C.; Forster, R. J.; Keyes, T. E.
Asian J. Org. Chem. 2013, 9, 763−778.
(42) Van Houten, J.; Watts, R. J. J. Am. Chem. Soc. 1976, 98, 4853−
4858.
(43) Keyes, T. E.; Vos, J. G.; Kolnaar, J. A.; Haasnoot, J.; Reedijk, J.;
Hage, R. Inorg. Chim. Acta 1996, 245, 237−242.
(44) Demas, J. N.; Adamson, A. W. J. Am. Chem. Soc. 1973, 95, 5159.
(45) Blackmore, L.; Moriarty, R.; Dolan, C.; Adamson, K.; Forster, R.
J.; Devocelle, M.; Keyes, T. K. Chem. Commun. 2013, 49, 5658−2660.
(46) Park, W.-H.; Han, Y.-W.; Kim, S.-K.; Kim, S.-Z. J. Cell. Biochem.
2007, 102, 98−109.
(47) Chen, Q.; Vazquez, E. D.; Moghaddas, S.; Hoppel, C. L.;
Lesnefsky, E. J. J. Biol. Chem. 2003, 38, 36027−36031.
(48) Drose, S.; Brandt, U. J. Biol. Chem. 2008, 31, 21649−21654.
(49) Han, Y.-W.; Moon, H.-J.; You, B.-R.; Kim, S.-Z.; Kim, S.-H.;
Park, W.-H. Anticancer Res. 2009, 29, 4423−4432.
(50) Balaban, R. S.; Nemoto, S.; Finkel, T. Cell 2005, 120, 483−495.
(
(
2
(
Y.; Schmehl, R.; Farnum, C. E.; Zipfel, W. R.; Fischbach, C.; Stroock,
A. D. Biomaterials 2012, 33, 2710−2722.
12) Zhang, S.; Hosaka, M.; Yoshihara, T.; Negishi, K.; Iida, Y.;
(
Tobita, S.; Takeuchi, T. Cancer Res. 2010, 70, 4490−4498.
13) Korzeniowska, B.; Nooney, R.; Wencel, D.; McDonagh, C.
Nanotechnology 2013, 24, 442002.
14) Koren, K.; Dmitriev, R. I.; Borisov, S. M.; Papkovsky, D. B.;
Klimant, I. ChemBioChem 2012, 13, 1184−1190.
15) Baggaley, E.; Botchway, S. W.; Haycock, J. W.; Morris, H.;
Sazanovich, I. V.; Williams, J. A. G.; Weinstein, J. A. Chem. Sci. 2014, 5,
79−886.
16) Neugeburger, U.; Pellegrin, Y.; Moran, N.; Signac, W.; Forster,
R. J.; Keyes, T. E. Chem. Commun. 2008, 5307−5309.
(
(
(
8
(
I
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(
51) Maguire, J. J.; Kagan, V. E.; Packer, L. Biochem. Biophys. Res.
Commun. 1992, 188, 190−197.
52) Choi, J. P.; Bard, A. J. Anal. Chim. Acta 2005, 541, 143−150.
(
J
dx.doi.org/10.1021/ja508043q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX