Cell uptake and cytotoxicity of a novel cyclometalated iridium(III) complex and its octaarginine peptide conjugate

Article abstract of DOI:10.1016/j.jinorgbio.2012.11.001

The synthesis and characterisation of iridium(III) bis(2-(2,4- difluorophenyl)pyridinato-N, C2′)-2(4-carboxylphenyl)imidazo[4,5-f][1,10] phenanthroline perchlorate, [Ir(dfpp)₂(picCOOH)]⁺ and its octaarginine conjugate [Ir(dfpp)₂(picCONH-Arg₈)] ^{9 +} are reported. Both complex and conjugate exhibit intense and long-lived luminescence, which is O₂ and pH sensitive. Conjugation to the polyarginine peptide renders the complex very water soluble. The uptake of the parent iridium(III) complex and conjugate are compared in two mammalian cell lines; SP2 myeloma and Chinese hamster ovary (CHO). Both complexes internalise into the cytoplasm, however dye uptake rate and distribution vary with peptide conjugation and with cell identity. Whereas transmembrane transport is thought to have been facilitated by the dimethyl sulfoxide (DMSO) used as co-solvent (0.05% v/v) for the parent complex, the octaarginine, the dye-conjugate (iridium-R₈) is membrane permeable in water only. Both complexes exhibit high cytotoxicity, evident through blebbing and vacuole formation within living cells, indicative of apoptosis, within 30 min of exposure to the probe. The IC₅₀ recorded for the cells in the dark was independent, in the case of the parent complex, of the identity of the cell, with IC₅₀ of 84.8 μM and 88 μM respectively for SP2 and CHO cells. The IC₅₀ approximately doubled for the polyarginine conjugate and displayed a significant dependence on cell type with IC₅₀ of 35 μM and 54.1 μM respectively for SP2 and CHO cells. These IC₅₀ values were recorded in the dark. However under irradiation cell death is considerably faster. Evidence from imaging suggests that the conjugate penetrates the nucleus whereas the parent does not, indicating that nuclear penetration may play a role in cytotoxicity.

Full text of DOI:10.1016/j.jinorgbio.2012.11.001

Journal of Inorganic Biochemistry 119 (2013) 65–74
Contents lists available at SciVerse ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Cell uptake and cytotoxicity of a novel cyclometalated iridium(III) complex and its
octaarginine peptide conjugate
Ciarán Dolan a,1, Roisin D. Moriarty a,1, Elena Lestini a, Marc Devocelle b, Robert J. Forster a, Tia E. Keyes a,
⁎
a
School of Chemical Sciences, National Centre for Sensor Research, Dublin City University, Glasnevin, Dublin 9, Ireland
b
The Royal College of Surgeons of Ireland, Dublin 2, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 13 August 2012
Received in revised form 2 November 2012
Accepted 5 November 2012
Available online 10 November 2012
The synthesis and characterisation of iridium(III) bis(2-(2,4-difluorophenyl)pyridinato-N, C2′)-2(4-
carboxylphenyl)imidazo[4,5-f][1,10]phenanthroline perchlorate, [Ir(dfpp)2(picCOOH)]+ and its octaarginine
conjugate [Ir(dfpp)2(picCONH-Arg8)]9+ are reported. Both complex and conjugate exhibit intense and
long-lived luminescence, which is O2 and pH sensitive. Conjugation to the polyarginine peptide renders the com-
plex very water soluble. The uptake of the parent iridium(III) complex and conjugate are compared in two mam-
malian cell lines; SP2 myeloma and Chinese hamster ovary (CHO). Both complexes internalise into the
cytoplasm, however dye uptake rate and distribution vary with peptide conjugation and with cell identity.
Whereas transmembrane transport is thought to have been facilitated by the dimethyl sulfoxide (DMSO) used
as co-solvent (0.05% v/v) for the parent complex, the octaarginine, the dye-conjugate (iridium-R8) is membrane
permeable in water only. Both complexes exhibit high cytotoxicity, evident through blebbing and vacuole forma-
tion within living cells, indicative of apoptosis, within 30 min of exposure to the probe. The IC50 recorded for the
cells in the dark was independent, in the case of the parent complex, of the identity of the cell, with IC50 of
84.8 μM and 88 μM respectively for SP2 and CHO cells. The IC50 approximately doubled for the polyarginine con-
jugate and displayed a significant dependence on cell type with IC50 of 35 μM and 54.1 μM respectively for SP2
and CHO cells. These IC50 values were recorded in the dark. However under irradiation cell death is considerably
faster. Evidence from imaging suggests that the conjugate penetrates the nucleus whereas the parent does not,
indicating that nuclear penetration may play a role in cytotoxicity.
Keywords:
Iridium
Polyarginine
Cellular Imaging
Mammalian cells
Cytotoxicity
Luminescence
© 2012 Elsevier Inc. All rights reserved.
1. Introduction
emission and depending on the application, show a preference for
localisation within a certain organelles of the cell. Until quite recently,
Transition metal polypyridyl complexes have been under intensive
investigation for many years as sensors due to their favourable
photophysical and redox properties including; absorbance and
emission in the visible spectral region, large Stokes shifts, long-lived
intense luminescence, good photostability and useful singlet oxygen
photosensitising properties [1]. These characteristics make them poten-
tially very valuable as biological probes. Their long-lived emission is of
particular interest, as it can permit discrimination from the shorter-
lived auto-fluorescence generated in many biological samples. Their
large Stokes shifts also reduce the possibility of concentration
quenching or self-absorption of the complex and in addition, in the
case of ruthenium(II) complexes, their red emission is isolated from
regions of the spectrum where autofluorescence of biological samples
often occurs. Ideally a luminophore suitable for cellular imaging should
be photostable, have a high quantum yield, good cell permeability and
exhibit low cytotoxicity. Furthermore, it should have visible to NIR
there were relatively few reports on the application of phosphorescent
transition metal complexes in bioimaging. The relatively late graduation
of metal complexes to cell imaging in spite of their attractive optical
properties can in part be attributed to the poor penetrability of inorgan-
ic dyes through the cell membrane in the absence of measures to chem-
ically or physically permeabilize the membrane. The relatively low
cationic charge associated with many of these metal complexes means
that they are unable to effectively use the cell's membrane potential
as a driving force for cellular entry [2]. However, recently metal
complexes are being increasingly explored as possible candidates for
real-time live cell imaging particularly where the complex has been
modified with functionality to improve its aqueous solubility or cell
permeability including those based on zinc(II) [3], rhenium(I) [4],
platinium(II) [5], ruthenium(II) [6–9] and rhodium (III) [10].
In particular, many iridium(III) complexes have shown superior
emission quantum yields and longer lifetimes compared with other
transition metal complexes as a result of their very strong spin-orbit
coupling which promotes efficient intersystem crossing by easing
the spin-forbidden character of the phosphorescent transition
[11,12]. Furthermore, modification of ancillary ligands on iridium(III)
⁎
Corresponding author. Tel.: +353 1 700 8185; fax: +353 1 83608.
E-mail address: tia.keyes@dcu.ie (T.E. Keyes).
1
These authors contributed equally to the work.
0162-0134/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jinorgbio.2012.11.001

66
C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
complexes yields tunable emission from blue to red spectral region
[13]. Such tuneable photophysical characteristics are generally not
seen to the same extent for other transition metal complexes [14].
The addition of ancillary ligands with functional groups such as
ionisable protons, can offer additional environmental sensitivity to
the metal complex which may be manifested in their optical proper-
ties or photophysics. Due to their high phosphorescent quantum
yield and remarkable structure-function relationships, iridium(III)
complexes, particularly cyclometalated complexes, have become
one of the most attractive class of phosphorescent heavy metal
complexes in bio-labelling [13,15–17]. One of the key drawbacks
of cyclometalated iridium complexes, however, is their poor aque-
ous solubility. This has been overcome in many studies involving
their biological application by using DMSO to pre-dissolve the metal
complex and thus solubilise it in water. This is not ideal as DMSO is
not necessarily benign, it can induce cell diffusion, cell differentiation,
enhance the permeability of lipid membranes and can also cause the
cell membrane to become less rigid facilitating membrane diffusion
of exogenous species [18,19]. It is also reported herein that DMSO can
lead to significant cell apoptosis in both SP2 and CHO cell lines.
In addition to their potential as cell imaging agents, some iridium(III)
complexes have been shown to exhibit significant cytotoxicity [20,21].
Although a drawback in cell imaging, this may prove useful in tumour
treatment. The potential to exploit cytotoxicity, as with their applications
in cellular imaging is limited by their poor aqueous solubility and also by
their cell permeability. Lo et al. [55] addressed this recently in a series of
PEGylated iridium(III) cyclometalated complexes which showed aque-
ous solubility and cell permeability, PEGylation also led to a signifi-
cant reduction in cytotoxicity, which was advantageous for imaging
applications.
the probe had permeabilized and distributed broadly throughout
the cell. Such multi-modal imaging capability is almost uniquely en-
abled in metal complexes because of their Stokes shifts [33–36].
Scheme 1 illustrates the structure and steps leading to synthesis of
the novel pH and oxygen sensitive iridium(III) parent dye complex,
which is the focus of the present study. Here, the ligand 2-(4-
carboxylphenyl)imidazo[4,5-f][1,10]phenanthroline) [(picCOOH)] li-
gand, is coordination to the iridium(III) metal centre cyclometalated
to two 2-(2,4-difluorophenyl)pyridine ligands. The [pic(COOH)]
ligand has two ionisable protons located on the imidazole ring giving
rise to well defined acid–base characteristics and a terminal carboxyl
functionality that allows for its efficient conjugation to biomolecules.
The [pic(COOH)] ligand has been exploited previously by our group
coordinated to
a ruthenium metal centre [7,37–40] and the
complexes were found remarkably, to exhibit lifetimes and emission
intensities that were significantly enhanced in aqueous compared
with non-aqueous media. The complexes also showed low cytotoxic-
ity. Also interestingly, iridium(III) complexes coordinated to related
imidazole ligand have also previously been reported in the context
of their anion sensitivity [41].
In this report the cell uptake and cytotoxicity of both the parent
iridium complex and its polyarginine conjugate are examined, in
the case of the polyarginine derivative the complex shows rapid pen-
etration and distribution into the cell and nucleus and in addition its
cytotoxicity was increased by nearly double compared to the parent
complex.
2. Experimental section
2.1. Experimental methods
As an alternative solution to the poor aqueous solubility of
iridium(III) complexes for the purposes of cellular imaging or deliv-
ery of cytotoxic agent to cell or tissue, herein we examine for the
first time, the conjugation of the novel iridium(III) luminophore to
the cell penetrating peptide (CPP), octaarginine (R8). Polypeptides
have been used to deliver a wide variety of molecular cargo to living
cells such as quantum dots [22], drugs [23,24], organic fluorophores
[25,26], proteins [27,28], and other smaller molecules [29,30]. Al-
though some debate remains, it is widely believed that endocytosis
is the most likely mechanism of cellular uptake of arginine rich cell
penetrating peptides [26,31,32]. Cosgrave et al. [39] previously
reported how conjugation of a membrane sensitive ruthenium(II)
probe to octaarginine led to the passive diffusion of the probe across
a cell membrane of SP2 cells without the need for an organic solvent.
Whereas, in contrast, the aqueous soluble parent ruthenium complex
showed no evidence of cell membrane permeability in the absence of
organic solvent. Once inside the cell membrane, the probe only emit-
ted when associated with a membrane, allowing imaging of the exter-
nal and internal membrane structures of the cell. Whereas resonance
Raman, possible without luminescence interference because of the
large Stokes shift of such complexes could be used to confirm that
Alumina chromatography was performed using an automated flash
chromatography system (Analogix-Intelliflash 310) using basic alumina
pre-packed chromatographic columns (Al2O3, 150 mesh size), equipped
with a UV–visible (UV–vis) detector (200–400 nm) and auto collector.
The detection wavelength used was 254 nm with a slope sensitivity of
0.1. 1HNMR and COSY 45 spectra were performed on a Brucker AC
400 MHz using DMSO-d6 or CDCl3 as solvent. Mass spectrometry exper-
iments were carried out using a Bruker Esquire LC_00050 electron spray
interface (ESI) with a positive ion polarity. Elemental analysis was carried
out on an Exador analytical CE440 analyser at UCD microanalysis Labora-
tory. UV–vis spectra were recorded on a Varian Cary 50 spectrophotom-
eter. Emission spectra were recorded on
a Varian Cary Eclipse
fluorescence spectrophotometer with an excitation slit width of 5 nm
and an emission slit width of 5 nm.
Luminescent lifetimes were obtained using a Picoquant Fluotime
100 TCSPC (time correlated single photon counting) system exciting
at 370 nm (picosecond pulse diode lasers) and the band pass dielec-
tric filter was set for detection at 460 nm and above. 10,000 counts
were collected for each lifetime measurement and all measurements
were performed in triplicate using Nanoharp software to confirm
F
F
F
N
Ir
N
Ir
N
F
Cl
Cl
F
F
H
N
H
N
N
N
N
N
EtOH
Ir
COOH
+
COOH
N
N
N
N
N
Reflux 18 hours
F
F
F
F
F
F
2-(4-Carboxylphenyl)imidazo
[4,5.f][1,10]phenanthroline
iridium(III) bis(2-(2,4-difluorophenyl)pyridinato-N,
C2')-2(4-carboxylphenyl)imidazo[4,5-f][1,10]phenanthroline
iridium(III) dichloro-tetrakis(2-(2, 4-
difluorophenyl)pyridinato-N, C2')
Scheme 1. The preparation of iridium(III) bis(2-(2,4-difluorophenyl)pyridinato-N, C2′)-2(4-carboxylphenyl)imidazo[4,5-f][1,10]phenanthroline.(ClO4−) [Ir(dfpp)2(picCOOH)]+
.

C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
67
results. Typical pulse rates of the excitation source were 1×105 s−1
with typical pulse widths of 300 ps. Deaerated samples were pre-
pared by bubbling nitrogen through the sample for 20 min prior to
analysis. The luminescent decays were fit to mono or biexponential
decay functions to extract the lifetime information using FluoFit soft-
ware. As the complexes were long-lived compared to the instrument
response function (IRF), data was fit to exponential decay functions to
the baseline of the decay curve using tail-fitting. Goodness of fit was
determined from visual examination of the residuals of the fit and a
χ2 below 1.1.
Cyclic voltammetry was carried out on CH Instruments CH-600
electrochemical workstation. A three electrode cell was used, with a
glassy carbon electrode (3 mm diameter Teflon shrouded) as the
working electrode, silver wire as reference electrode and platinum
wire as auxiliary electrode. Ferrocene was employed as an
internal standard and calibrated for all results versus Ag/AgCl as
reference. Electrochemistry was conducted using 1 mmol of sample
dissolved in 5 mL of dry acetonitrile (HPLC grade) containing 0.1 M
tetrabutylammonium tetrafluoroborate (TBA) as supporting electro-
lyte. All samples were degassed with nitrogen prior to analysis.
Micro-Raman measurements were carried out in backscattering con-
figuration using an Olympus confocal microscope attached to a HORIBA
Jobin-Yvon Labram HR 1000 spectrometer. Raman scattering was
detected using a Peltier cooled (−70 °C) charge-coupled device (CCD)
camera (255×1024 pixels), excited with vertically polarised excitation
lines from Ar+ laser (458/488/514 nm) or 785 nm diode laser sources.
A 600 grooves/mm grating was used and the spectral resolution was
0.6 cm−1. The area of the laser spot on the samples was 1 μm in diame-
ter. The laser power at the sample was set between 1 and 2 mW using the
inbuilt laser power control. Data acquisition times used in the Raman ex-
periments ranged from 1 to 20 s. The Raman band of a silicon wafer at
520 cm−1 was used to calibrate the spectrometer. The spectral data ac-
quired was analysed using LabSpec software.
FTIR spectra were recorded using a Varian 610-IR microscopy
equipped with a nitrogen-cooled mercury cadmium telluride (MCT)
and DTGs detectors. The samples were dispersed in a KBr matrix
and were recorded at a 2 cm−1 spectral resolution, with KBr used
as a beam splitter. Raw spectra were manipulated by built in func-
tions in Varian Resolution pro software, which performs corrections
to remove atmospheric water vapour and CO2 absorption bands,
baseline and smoothing.
pH titrations were performed in the pH range 0.5 to 12, pH was
monitored with a VWR Symphony SP70P pH meter. pH was adjusted
by the stepwise additions of microlitre aliquots of concentrated
perchloric acid or sodium hydroxide solutions into a 15 mL phosphate
buffered saline solution (PBS solution) containing 20 μM of the complex
under investigation. All PBS solutions were made using P-5368 PBS pH
7.4 from Sigma unless otherwise stated. The dry powder was dissolved
in 1 L of deionised water which yielded a solution containing 0.01 M
PBS solution, 0.138 M NaCl and 0.0027 M KCl at pH 7.4. The resulting
spectra were not corrected for dilution as the volume added to the solu-
tions were negligible compared to its total volume.
culture dishes. Both cell types were grown for 24 h and media was
changed to serum free Leibovitz media (no riboflavin) for 4 h before
imaging. The growth medium was removed by washing with PBS so-
lution (pH 7.4) and 70 μM of the parent complex (in PBS solution
with 0.05% DMSO) or dye-peptide in PBS solution supplemented
with 1.1 mM MgCl2 and 0.9 mM CaCl2, pH 7.4 was added and cells
were imaged immediately. The cell viability dye, DRAQ7 (Biostatus),
was added 1 h after imaging, as it stains the nucleus of dead cells.
DRAQ7 (4 μM final concentration) in PBS solution and excited with
the HeNe 633 nm laser and emission was collected using a long
pass 650 nm filter set. Luminescence images were recorded using a
Zeiss LSM510 Meta confocal microscope using a 63× oil immersion
objective lens (NA 1.4). The 375 nm laser excitation was used for irid-
ium samples and emission was collected using a long pass 505 nm fil-
ter set.
2.3. Cytotoxicity assay
Cells were seeded in a 96-well plate in 100 μL of media for 24 h at
37 °C with 5% CO2 in the dark before addition of compounds. SP2
cells were seeded at 1×105 and CHO cells at 1×104 cells per well.
Metal complex/conjugate were added to yield final concentrations of
140, 70, 35 and 17.5 μM and DMSO and ethanol were added to yield
final concentrations of between 20 and 0.3 % V/V. The cells/solvent/
complex media were left for 16 h at 37 °C in a 5% CO2 incubator in
the dark. 10 μL of either the MTT (3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide) or Resazurin reagents were added
and incubated for a further 4 h for MTT assay and 7 h for Resazurin
assay at 37 °C in the dark. Formazan was solubilised with 150 μL
DMSO per well for the MTT assay and absorbance was measured at
570 nm and corrected at 630 nm for background absorbance. The
resazurin was converted to resorufin in viable cells and its absor-
bance was detected at 570 nm with background subtraction at
600 nm. Absorbance readings were performed using a Tecan 96-well
plate reader.
2.4. Instrumentation used for peptide synthesis
Chromatographic purification was carried out on a Varian HPLC
Chromatography Workstation using a Phenomenex Gemini (C-18)
reverse-phase chromatography column. The Gemini C-18 (250 mm×
2.5 mm) was used at 1 mL/min with linear gradient programs. The
UV–vis detector was monitored for dye-peptides at dual wavelengths
of 214 nm and 350 nm for iridium conjugate. Solvent A consisted of
deionised H2O containing 5% trifluoroacetic acid (TFA) and solvent B
consisted of CH3CN containing 0.1% TFA. The gradient ran over
30 min going from 5% to 65% of solvent A. Mass spectrometry analysis
was performed on a LaserToF by matrix assisted laser desorption
ionisation-time of flight (MALDI-TOF). α-Cyano-4-hydroxy cinnamic
acid matrix was dissolved in 1:1 H2O containing 0.1% TFA and CH3CN
containing 0.1% TFA at a concentration of 10 mg/mL. 2 μL of a 1:1
solution of the dye-peptide to matrix were applied to the maldi plate.
SP2/0-Ag 14 murine Myeloma spleen cells (ATCC no. CRL-1581)
and Chinese hamster ovarian (CHO) (CHO-K1, ATCC no. CCL-61)
were purchased from ATCC Cell Biology Collection (UK). The suspen-
sion cell line SP2 were grown in Dulbecco's modified Eagle's medium
(DMEM) with stable L-glutamine and CHO cells in DMEM/Hams F-12
both supplemented with 10% foetal calf serum at 37 °C with 5% CO2.
Cells were harvested or split when they reached 90% confluency for
CHO cells or 1×106 cells/mL for the suspension SP2 cells. CHO cells
were harvested after trypsinisation (0.25% trypsin for 5 min at 37 °C).
2.5. Materials
All chemicals and reagents were purchased from Sigma-Aldrich
(Ireland) and were used without purification unless otherwise
stated. Protected amino acids, 2-(1 H-benzotriazol-1-yl)-1,1,3,3-
tetramethyluronium hexafluorophosphate (HBTU), benzotriazol-
1-yl-oxytripyrrolidinophosphonium hexafluorophosphate (PyBOP)
and Rink-amide MBHA resin were purchased from Novabiochem
(Merck Biosciences). 1-Hydroxybenzotrizole (HOBt) was purchased
from Iris Biotech GmbH. Synthesizer filters (in-line and reactor) and
solvents (dichloromethane (DCM), N-methylpyrrolidinone (NMP))
were sourced from Applied Biosystems and MTT/Resazurin reagents
2.2. Real-time confocal luminescence imaging
CHO cells were seeded at 8×104 cells in 2 mL media and SP2 cells
were seeded at 5×105 cells in 2 mL media on 35 mm glass bottom

68
C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
were acquired from PromoKine. Cell culture media, serum and penicilin/
streptomycin were purchased from Biosera.
The purified compound showed a single peak on HPLC with retention
time of 30.27 min.
3. Results and discussion
2.6. Synthesis of cyclometalating ligands and iridium(III) chloro bridged
dimer. General Procedure
3.1. Synthesis
The pH sensitive 2-(4-carboxylphenyl)imidazo[4,5-f][1,10]
phenanthroline): [(picCOOH)] ligand [37], 2-(4,6-difluorophenyl)
pyridine: [(dfpp)] ligand [42] and iridium(III) di-μ-chloro-tetrakis(2-
(2, 4-difluorophenyl)pyridinato-N, C2'): [Ir2(dfpp)4(Cl2)] [43] were
synthesised according to literature methods.
The precursor ligands and [Ir2(dfpp)4(Cl2)] were prepared and pu-
rified according to modified literature procedures [37,42,43]. Their
structure and purity was confirmed using mass spectrometry and
1H NMR which conformed to literature values. Coordination of the
[picCOOH] ligand to the iridium(III) metal centre proceeded in good
yield (78%) on overnight reflux of [Ir2(dfpp)4(Cl2)] with the ligand.
The novel iridium(III) complex, [Ir(dfpp)2(picCOOH)]+, was easily
purified by washing without the need for column chromatography.
The octaarginine peptide was prepared using a solid phase peptide
synthesizer (SPPS). Initially a peptide unit is coupled to the solid
support resin using HBTU as the coupling reagent. It is the base, DIEA,
that initiates this coupling reaction. Following Fmoc deprotection, by
piperidine in NMP, the amine group on the peptide unit is available
for amino acid coupling. The peptide is then elongated from the
C-terminus (carboxylic acid functional group) to the N-terminus
(amino functional group) via HBTU/HOBt/DIEA mediated coupling re-
actions. Following SPPS of the octaarginine polypeptide, the remaining
Fmoc protecting group on the hexanoic acid spacer is removed manu-
ally using piperidine. The inorganic molecular probe is then manually
conjugated to the polypeptide using PyBOP/DIEA coupling reagent in
a procedure described in the experimental section. Finally the dye-
peptide is removed from the resin and the amino acid side-chains
(Pbf) groups deprotected using a cleavage cocktail.
Purification of the dye-peptide was performed using semi-
preparative HPLC with the UV detector set for detection at 214 nm
and 350 nm for the iridium conjugate. It was found that a mixture
of 5% TFA in water was a suitable solvent mixture for the HPLC puri-
fication of the [Ir(dfpp)2(picCONH)Arg8]9+ conjugate. MALDI-TOF
mass spec analysis of the product exhibited a molecular ion of
2274 m/z which corresponded to the anticipated molecular weight
of the conjugated complex. HPLC confirmed the purity of the iridium
dye-peptide with a single peak retention time of 30.27 min under the
HPLC conditions described above.
2.7. Synthesis of iridium(III) bis(2-(2,4-difluorophenyl)pyridinato-N,
C2')-2(4-carboxylphenyl)imidazo[4,5-f][1,10]phenanthroline.(ClO4−):
[Ir(dfpp)2(picCOOH)]+
Iridium di-μ-chloro-tetrakis(2-(2, 4-difluorophenyl)pyridinato-N,
C2′) (220 mg, 0.181 mmol), 2-(4-Carboxylphenyl)imidazo[4,5-f]
[1,10]phenanthroline [(picCOOH)] (130 mg, 0.380 mmol) and etha-
nol (30 mL) were added to a round bottomed flask and left refluxing
at 95 °C overnight. The reaction was allowed cool to room tempera-
ture and the solvent removed by warming under reduced pressure.
The solid was suspended in a saturated solution of sodium perchlo-
rate and sonicated. The precipitate was collected by vacuum filtration
and washed with deionised water. The precipitate was then dissolved
in methanol and filtered once again. The filtrate was taken and evap-
orated to dryness to yield a yellow coloured solid (285 mg, 78 %).
1H NMR (400 MHz, DMSO-d6): δ (ppm) 9.22 (d, 2 H, 8 Hz), 8.41
(d, 2 H, J=8 Hz), 8.29 (d, 2 H, J=9.2 Hz), 8.22 (d, 2 H, J=4.4 Hz),
8.17 (d, 2 H, J=8.4 Hz), 8.07 (dd, 2 H, J=5.2 Hz, J=8 Hz), 7.98
(t, 2 H, J=7.6 Hz), 7.58 (d, 2 H, J=5.6 Hz), 7.09–6.99 (m, 4 H), 5.71
(dd, 2 H, J=2 Hz, 2.4 Hz).
Electron spray ionisation-mass spectrometry (ESI-MS): (CH3OH,
m/z): 1012.39, ([M+\ClO4]): 913.1.
Elemental analysis calculated for C42H24F4IrN6O2.(ClO4−).(H2O)2:
C, 48.12; H, 2.69; N, 8.02%. Found: C, 48.35; H, 2.31; N, 7.86%.
Melting point: >300 °C
2.8. Synthesis of [Ir(dfpp)2(picCONH)Arg8]9+ bioconjugate
Eight vials each containing Fmoc-Arg(Pbf)-OH (650 mg, 1 mmol)
and one vial containing a hexanoic acid spacer Fmoc-6-Ahx-OH
(354 mg, 1 mmol) were weighed out and placed into the solid phase
peptide synthesiser (SPPS). The peptide was assembled using Rink
Amide Resin (172 mg, 0.36 mmol) and was performed on the SPPS. Fol-
lowing the reaction, the resin was transferred into dimethylformamide
(DMF) (4 mL) to allow for swelling of the resin. Half of this
solution (2 mL) was added to a vial already containing PyBOP (78 mg,
3.2. Spectroscopic properties of [Ir(dfpp)2(picCOOH)]+
Fig. 1 shows the absorbance and emission spectra of the iridium
parent complex in PBS solution (pH 7.4+1% V/V DMSO) solution
and [Ir(dfpp)2(picCONH)Arg8]9+ conjugate in PBS solution (pH 7.4)
solution respectively. The UV–vis spectrum of [Ir(dfpp)2(picCOOH)]+
is dominated by ligand-centred (LC) difluorophenylpyridine (dfpp)
π–π* electronic transitions giving maximum absorbance peaks at
245 nm and 283 nm. This is consistent with reports of other
iridium(III) complexes containing difluorophenyl pyridine, bipyridyl
or terpyridine ligands [44,45]. The shoulder at 324 nm is assigned to
the intraligand π–π* transition of the [picCOOH] ligand [37,38]. A
weaker, broad feature, centred at 381 nm which extends to roughly
480 nm is attributed to the an iridium dπ to [dfpp] π* metal to ligand
charge transfer (MLCT) transition [42,43,46].
0.15 mmol), HOBt (30 mg, 0.22 mmol), [Ir(dfpp)2(picCOOH)]+
.
(ClO4−
)
(200 mg, 0.20 mmol) and DMF (2 mL). Finally, N,N-
diisopropylethylamine (DIEA) (40 μL, 0.23 mmol) was added and the
solution was left shaking overnight. The resin precipitate was washed
with NMP (5 mL×3 times) and DMF (5 mL×3 times) allowing for
2 min agitation before each filtering. The remaining resin is then left to
dry. The peptide was cleaved by standard methods and then purified by
reverse phase-HPLC. The fractions collected by semi-preparative HPLC
were analysed by mass spectrometry. The synthesis yielded a yellow
coloured solid (3.0 mg, 34%). The purity of the dye-peptide was con-
firmed by mass spectrometry and HPLC analysis. A stock solution
(20 μM) in PBS solution pH 7.4 was then made up and kept in the freezer.
[Ir(dfpp)2(picCOOH)]+ is strongly luminescent, exhibiting an
emission maximum at 521 nm when excited into the MLCT band at
381 nm with a quantum yield of 0.103 in aerated acetonitrile. The
iridium parent complex has a lifetime of 195 2 ns (χ2 =0.879) in
aerated acetonitrile and 674 5 ns (χ2 =1.141) in degassed acetoni-
trile. The lifetimes of the complex were also examined in water. The
iridium(III) parent complex was poorly soluble in PBS solution and
required pre-dissolution in acetonitrile (5%) to aid solubility. The life-
times in aerated aqueous solution (PBS solution containing 5% V/V
acetonitrile) are 207 7 ns at pH 1.0, 168 6 ns at pH 6.9 and 5
ꢀ
ꢁ
Â
Ã
9þ
MAL ꢀ TOF lrðdf ppÞ2ðpicCONHÞArg8
: 2279:4; ð½M−H5ꢁÞ2274:0:
m=z

C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
69
1
1
0.8
0.6
0.4
0.2
0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Absorbance of free Ir(III) complex
Absorbance of Ir-peptide conjugate
Emission of free Ir(III) complex
Emission of Ir-peptide conjugate
225
325
425
525
625
Wavelength (nm)
Fig. 1. Overlaid absorbance and normalized emission spectra (excited at 380 nm, slit widths: 5 nm, normalized to 0.5 arbitrary units) of 20 μM [Ir(dfpp)2(picCOOH)]+ in PBS
solution (pH 7.4+1% DMSO) solution and iridium-R8 in PBS solution (pH 7.4) solution.
1 ns at pH 12.0. In deaerated PBS solution (containing 5% V/V acetoni-
trile) the emission lifetime of the complex increased to 200 4 ns at
pH 6.9.
1.0 had a no significant effect on the absorbance wavelengths of the
iridium complex suggesting that there is no change in the HOMO-
LUMO energy gap or in the distribution of π electrons across the system
upon acidification. However, the absorbance intensities in particular, of
the two LC π–π* transitions at 245 nm and 283 nm vary significantly
with a decrease in pH.
Unlike the parent complex, the peptide conjugate was soluble in
aqueous media and did not require pre-dissolution in an organic
solvent. UV–vis spectra of [Ir(dfpp)2(picCONH)Arg8]9+ is dominated
by a peptide absorbance peak at 262 nm an absorbance at 330 nm
attributed to the intraligand π-π* transition of the [picCOOH] ligand
appears as a shoulder on this transition [37]. The MLCT absorbance
peak at 387 nm is also apparent and as for the complex in the absence
of the peptide, this band is broad, tailing to approximately 480 nm
[46]. The emission spectrum of [Ir(dfpp)2(picCONH)Arg8]9+ exhibits
a maximum at 543 nm in water which is slightly red shifted com-
pared to the parent (537 nm). However, within experimental error,
the lifetime of the parent and conjugate were the same in aqueous
media.
Adjustment of the pH of a solution of [Ir(dfpp)2(picCOOH)]+ to 12.0
as shown in Fig. 2 results in a shift in absorbance wavelengths, in partic-
ular the LC π–π* and intra ligand π–π* [picCOOH] transitions, to lower
energies. This is consistent with behaviour reported for this [picCOOH]
ligand in a ruthenium(II) complex attributed to extension of the π sys-
tem in the anionic imidazoles [37,38]. The effect of pH on the MLCT tran-
sition is less pronounced as this transition is mostly concentrated on LC
cyclometalating ligand transitions at higher energies. The relative
immunity of the MLCT to pH indicates strongly that the LUMO for the
MLCT lies mainly on the [dfpp] ligands, which is also confirmed in
comparing the ground and excited state pKa values, vide infra.
The ground state pKa values of the imidazole were calculated as
3.7 0.1 and 11.5 0.1 respectively by monitoring the spectral
changes at 330 nm over a pH range of 1.0–12.0 and determining the
points of inflection (Fig. S1 supplementary material). It is important
to also note that the acid–base chemistry of both the absorbance
and emission were found to be irreversible as the spectral changes
were not fully recovered on returning from post pKa pHs to neutral
pH.
3.3. pH dependence studies of [Ir(dfpp)2(picCOOH)]+
As the imidazole ligand has three ionizable sites, the effect of pH on
the spectroscopy and photophysics was explored. Fig. 2 shows the
absorbance spectra at various pH values and Fig. S1 (supplementary
material) shows the corresponding pH titration curves from which
the pKa and pHi may be determined respectively. Protonation of the
[Ir(dfpp)2(picCOOH)]+ complex, by adjustment of the pH from 7.0 to
0.6
0.5
0.4
0.3
0.2
0.1
0
pH 1.0
pH 7.0
pH 12.0
230
280
330
380
430
480
530
580
Wavelength (nm)
Fig. 2. UV–vis spectra of 20 μM (15 mL) [Ir(dfpp)2(picCOOH)]+ in PBS solution (containing 5% acetonitrile solution) at pH 1.0, pH 7.0 and pH12.0.

70
C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
The emission spectroscopy of [Ir(dfpp)2(picCOOH)]+ showed
Bracketing experiments confirm that this peak is related to a ligand
reduction peak at −1.02 V as it is not present unless this potential
was scanned first. It is therefore attributed to the oxidation of the ad-
sorption product.
spectral shifts to longer wavelengths upon deprotonation from
536 nm at pH 1.0 to 564 nm at pH 12.0 as shown in Fig. 3. This
suggested an increase in the HOMO–LUMO energy gap at lower pH
values which may be due to the electrostatic repulsion between the
positively charged iridium 4+ metal centre and the extra positive
charge on the imidazolium ligand [37].
The excited state pKa* values were calculated using the maximum
emission intensity to plot the pH titration curves and were fitted to an
expression for the Förster cycle using the following equation:
The irreversibility of the Ir3+/4+ process has been noted before in
cyclometalated iridium complexes and is attributed, when observed,
to strong mixing of the Ir\C σ-bond orbitals. The oxidation potential
is quite anodic compared with other iridium(III) complexes and is
attributed to the strong electron withdrawing nature of the [dfpp]
ligands [50].
pKaꢂ ¼ pHi þ logðτHB=τB−Þ
3.5. Cellular uptake of [Ir(dfpp)2(picCOOH)]+ and [Ir(dfpp)2(picCONH)
Arg8]9+ bioconjugate
where pHi is the point of inflection on the pH emission titration plot
(Fig. 3(A)), τHB is the lifetime of the protonated species and τB− is
the lifetime of the deprotonated species. The excited state pKa* values
were calculated to be 3.1 0.1 and 11.1 0.1 respectively for the (de)
protonation of the ancillary ligand. These values are very similar to
those calculated for ground state pKa indicating that the excited
state lies on the [dfpp] ligands of the iridium complex. If the excited
electron was located on the [picCOOH] ligand, because of the nominal
charge induced by the excited state, would be expected that the ba-
sicity of that ligand in the excited state would increase by as much
as 2–3 pH units compared with the ground state [47]. These values
must be treated with some caution however as spectroscopy recov-
ered only about 80% of its original value following acidification/base
to neutral pH.
The efficacy of uptake of the iridium parent complex and its
octaarginine conjugate ([Ir(dfpp)2(picCONH)Arg8]9+) was examined
in two cells types, chosen for their varied morphology. CHO cells are
adherent cells spreading approximately 50 μm in diameter, whereas,
SP2 cells are round suspension cells approximately 10 μm in diame-
ter. Fig. 4 shows confocal luminescence images of both SP2 and CHO
cells treated with [Ir(dfpp)2(picCOOH)]+ and [Ir(dfpp)2(picCONH)
Arg8]9+ 2 min and 15 min after addition of the dye to the cells.
Both cell types exhibited low levels of uptake immediately upon dye
addition (visible by using the rainbow mode, data not shown). Sur-
prisingly, Fig. 4(B), shows the iridium parent dye was visible in the in-
terior of the cell after approximately 15 min in SP2 cells compared to
just 5 min in CHO cells. For the SP2's the dye is located in the cell's
membrane after approximately 5 min at room temperature in PBS so-
lution (pH 7.4 containing 0.05% DMSO) and continues to permeate
throughout the cells cytoplasm over the following 15 min. Z-stack
images were used to confirm the distribution of the dye throughout
the whole cell (Fig. S6, supplemental materials). However, it is impor-
tant to note that it is likely that the use of DMSO as co-solvent has
permeabilized the cellular membrane rather than that the parent
complex is inherently cell permeable [51,52].
3.4. Electrochemistry
The cyclic voltammogram for [Ir(dfpp)2(picCOOH)]+ in acetonitrile
(containing 0.1 M TBA as electrolyte) is shown in supplementary infor-
mation (Fig. S2). The complex exhibits a quasi reversible Ir3+/4+ redox
couple centred at +1.70 V versus Ag/AgCl and two quasi-reversible re-
ductions at −0.85 V assigned to the [dfpp] ligand, −1.47 V attributed
to the [dfpp] ligand and a further reversible redox peak at −2.09 V at-
tributed to the [picCOOH] ligand versus Ag/AgCl respectively [48]. The
anomalous shape and intensity of the −0.85 V wave is consistent with
adsorption and is likely to arise due to film formation on the electrode
surface as the carboxyl functional group may adsorb to an anodized
glassy carbon electrode surface [49].
Vesicles were observed in both cell lines after a period of approx-
imately 30 min incubation with [Ir(dfpp)2(picCOOH)]+, which is an
indication of cell death. Vesicle formation also occurs in endocytosis,
however, in this instance they are attributed to cell death as the par-
ent iridium(III) complex exhibits high cytotoxicity, vide infra. This
was further confirmed wherein, DRAQ7, a nuclear staining dye
which permeates only dead cells showed evidence of penetration
There is also an oxidation peak at +1.44 V versus Ag/AgCl, which
appears just before the metal redox couple in a full potential sweep.
50
45
40
35
30
25
20
15
10
5
pH 1.0
pH 7.0
pH 12.0
0
460
510
560
610
660
Wavelength (nm)
Fig. 3. Emission spectra of 20 μM (15 mL) [Ir(dfpp)2(picCOOH)]+ in PBS solution (containing 5% acetonitrile solution, excited at 380 nm, slit widths: 5 nm) at pH 1.0, pH 7.0 and pH
12.0.

C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
71
Fig. 4. Uptake of iridium(III) luminophore and iridium peptide conjugate ( [Ir(dfpp)2(picCONH)Arg8]9+) in SP2 and CHO cells. Live cells were seeded at 5×105 cells onto glass bot-
tomed dishes for 24 h before addition of the dyes (70 μM in PBS solution supplemented with 1.1 mM MgCl2 and 0.9 mM CaCl2 and with 0.05% DMSO for the iridium parent com-
plex). Images were acquired using a Zeiss LSM510 Confocal microscope and 63× oil immersion lens. (A) Emission intensity was measured per cell using ImageJ software over
20 min with an average of 95 cells analysed per treatment over three separate experiments SEM. (B) Confocal images measured at room temperature of iridium parent dye
and iridium-R8 at 2 and 15 min, respectively, following the addition of 70 μM of both parent and dye-peptide to SP2 and CHO cells.
into both SP2 and CHO cells when incubated with the iridium
luminophore for 30 min in PBS solution (pH 7.4 containing 0.05%
DMSO) as shown in Fig. S5 in supplementary material.
lines, but a key difference between them is that SP2 is a suspension
cell whereas CHO is an adherent cell. The greater membrane protein
concentration expected at the adherent cell line may play a role in
the greater permeability of this cell line.
The conjugation of octaarginine peptide has been shown previous-
ly [2,31,39,53] to enhance cellular uptake of metal complex dyes
without the requirement for membrane permeablization and this is
further exemplified here in both cell lines. Dye uptake was measured
by live-cell time course analysis and quantified by measuring the dye
intensity in each cell type at the same time points after incubation
wherein the concentration of each dye applied to the cell medium
was identical. As the absorption at the excitation wavelength used
in the microscopy and quantum yield of parent and conjugate are
the same, a direct comparison of intensity can be made as the micros-
copy experiments were carried out under identical conditions. As
Fig. 4 shows, the uptake of [Ir(dfpp)2(picCONH)Arg8]9+ peaks earlier
in SP2 cells; at 5 min the maximum average intensity of 4.02 0.01
was reached. For the CHO cells the maximum uptake was reached
at 10 min, however, the final average intensity at 9.85 0.04 was
nearly double that of the SP2 Fig. 4(A). This slower, but ultimately
greater uptake of complex by the CHO cells compared to SP2 cells is
evident in Fig. 4(B). Both CHO and SP2 are mammalian cancer cell
The dyes appear to bind preferentially to the membrane of the cells,
particularly the SP2 cells as indicated by membrane staining evident
for example in Fig. 4. The iridium parent complex, in particular, collects
at the membrane before diffusing into the cells. There is strong evi-
dence for cell death early on after exposure, wherein the membranes
of the death vacuoles and nuclear envelope light up and blebbing of
the cells occurs, this behaviour is evident in data shown in supplemen-
tal materials, Fig. S5 and confirmed by the Z-stack images of iridium
within the cells in Fig. S6 and typically indicates an apoptotic process.
[Ir(dfpp)2(picCONH)Arg8]9+ enters and distributes into the cell more
quickly than the parent and does not appear to localise as strongly in
the cell membrane. Rather it appears to enter endocytic vesicles,
which transport the dye into the cell where it then locates diffusely.
It is not clear if the cytoplasmic pattern of diffusion is located within
cytoplasmic organelles. However co-localisation studies with the nu-
clear stain, DRAQ7 indicate that the parent iridium complex does not
distribute throughout the nucleus. However, there is evidence of the

72
C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
parent complex and particularly the conjugate; iridium-R8 dye within
the nucleolus (as indicated in supplemental materials Fig. S6 and
highlighted by the intensity profile in Fig. S7). Indeed localisation of
the dye within the nucleolus could indicate a possible death pathway.
The nucleolus is the inner, non-membranous part of the nucleus
responsible for ribosomal assembly and could have a role to play in
turning on apoptosis. However, it is still not clear at this point if there
are necrotic effects of the dye also having an effect on the cells (due
to the presence of cellular debris surrounding the SP2 cells treated
with [Ir(dfpp)2(picCONH)Arg8]9+ at 15 min (Fig. 4(B)).
3.6. [Ir(dfpp)2(picCOOH)]+ and [Ir(dfpp)2(picCONH)Arg8]9+ are
cytotoxic to SP2 and CHO cells
Given the presence of blebbing and vesicle formation in both SP2
and CHO cells following incubation with the iridium(III) complex
and it's conjugate, it was important to assess the toxic effects of the
metal complexes on the cell lines. Fig. 5 illustrates the cytotoxic effect
of the parent luminophores and conjugated dye-peptides on SP2 my-
eloma and CHO cell lines respectively using the resazurin assay and
Table 1 reports the IC50 values parent and conjugate for the two cell
lines explored. These cytotoxicity experiments were performed in
the dark, therefore contribution from photosensitized cell death, for
example, as a result of the singlet oxygen generating ability of the ex-
cited state complex is expected to be minimal in these experiments.
The data shows that both the iridium(III) parent complex and the
conjugate are cytotoxic, however interestingly the IC50 values depend
on both the cell line and whether the complex is peptide conjugated.
Interestingly, the [Ir(dfpp)2(picCONH)Arg8]9+ polypeptide conju-
gate shows substantially higher cytotoxicity in both cell lines than
that of the iridium parent complex. IC50 against SP2 was 84.8 μmol
for the parent compared with 35 μmol for [Ir(dfpp)2(picCONH)
Arg8]9+. Interestingly, cytotoxity was lower against CHO than SP2
for the conjugate with an IC50 value of 54.1 μmol. Although, the
cytotoxicty of the parent in both cell lines was approximately the
same. The cytotoxicity of the parent is not due to the DMSO as
co-solvent (at 0.05% V/V DMSO) as confirmed in the control experi-
ments described in the following section. This DMSO concentration
is below the cytotoxic threshold, as indicated in Fig. 6.
The cytotoxicity of iridium(III) complexes has been noted previously
in a number of reports. For example Lo et al. [4] reported IC50 values of
between 1.1 and 6.3 μM for iridium indole complexes. Xiong et al. [54]
reported a 20% decrease in cell viability with 100 μM of iridium(III)
complex [Ir(pba)2(bpy)]+.PF6. Whereas, in contrast, Yu et al. [15]
reported that [Ir(dfpy)2(bpy)]+. PF6− and [Ir(dfpy)2− (quqo)]+.PF6
do not induce cell death at concentrations up to 100 μM in 2% DMSO.
Interestingly Lo et al. reported decreased cytotoxicity in water soluble
pegylated iridium complexes compared with their non-pegylated ana-
logue (ca. 286.5 to 1180.0 μM compared to ca. 4 to 15 μM) [55]. This
was consistent with previous reports on the use of pegylation in both re-
ducing cytotoxicity and promoting aqueous solubility. In contrast, here,
conjugation to polyarginine approximately doubles the IC50 in both cell
lines explored. This might be attributed to the greater permeability and
therefore cell uptake of the [Ir(dfpp)2(picCONH)Arg8]9+ compared
with the parent, but is unlikely, as CHO, the most permeable to this
dye, achieved the highest final internal concentrations of the conjugate
but showed the lower IC50. The molecular origin of the increased
cytotoxicity is unknown but it has been noted previously that the
polyarginine itself exhibits cytotoxicity toward mammalian cells [56].
The above experiments were performed in the dark. To examine the
role, if any, of light in promoting cell death, an initial control experiment
was performed. This demonstrated that when no complex/conjugate is
present under continuous excitation at 375 nm for more than an hour,
there is negligible cell death, as confirmed with viability and nuclear
stain, DRAQ7. Conversely, after continuous 375 nm laser scanning for
1 h with the iridium dye-peptide, [Ir(dfpp)2(picCONH)Arg8]9+, 90% of
Fig. 5. Cytotoxicity studies of iridium complex and iridium dye-peptide conjugate
(iridium-R8) in (A) SP2 and (B) CHO cell lines. Cells were seeded in 96-well plates for
24 h prior to the addition of both iridium compounds and they were incubated in the
dark overnight at 37 °C. Resazurin reagent was added to cells and incubated in the dark
at 37 °C for a further 7 h. The absorbance was measured at room temperature using a
Tecan plate reader at 570 nm with background correction at 600 nm. (n=4 SEM).
the cells were dead confirmed with DRAQ7, compared with different
regions where there was no DRAQ7 staining. In the absence of light
similar levels of cytotoxicity were only achieved (78.8 12.3% and
100 0.4% cytotoxicity for SP2 and CHO cells treated with iridium-R8,
respectively, Fig. 5) after overnight incubation. Excitation of the iridium
centre therefore dramatically increases the rate of cell death. For exam-
ple using the iridium conjugate at 375 nm excitation the DRAQ7 mean
pixel intensity goes from 1.41 to 5.94 after only 4 scans on the confocal
microscope. The origin of the light promoted cell death is considered
likely to be as a result of the generation of singlet oxygen (1O2). This
was further supported by preliminary experiments using ascorbic acid
as a singlet oxygen scavenger [57] which resulted in a reduction in
cell cytotoxicity of the iridium peptide from 50 to 5 % under the same
Table 1
IC50 values for [Ir(dfpp)2(picCOOH)]+ and [Ir(dfpp)2(picCONH)Arg8]9+ in SP2 and
CHO cell lines. Values were calculated using three sets of experimental data standard
deviation using GraphPad Prism software.
IC50 Values
SP2 (Μm)
CHO (Μm)
[Ir(dfpp)2(picCOOH)]+
84.84 0.02
34.95 0.02
87.97 0.01
54.11 0.02
[Ir(dfpp)2(picCONH)Arg8]9+

C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
73
(2,4-difluorophenyl)pyridinato-N, C2')-2(4-carboxylphenyl)imidazo
[4,5-f][1,10]phenanthroline perchlorate, [Ir(dfpp)2(picCOOH)]+ and
its octaarginine conjugate, [Ir(dfpp)2(picCONH-Arg8)]9+
, was de-
scribed. The polypeptide was formed using solid state peptide synthe-
sis and the complex–peptide conjugation was performed on-resin.
Both complex and conjugate were emissive, with λmax of approximate-
ly 540 nm in aqueous media. [Ir(dfpp)2(picCOOH)]+ exhibited a life-
time of 674 5 ns in degassed organic media and a quantum yields
of 0.103. The lifetime of the complex was reduced in aqueous media
at pH 6.9 to 168 ns which increased to 200 ns when dearated. The
location of the excited state was confirmed from excited state pKa to
be located on the cyclometalating [dfpp] ligands and conjugation of
the polypeptide had relatively little impact on the photophysics of
the complex.
Conjugation of [Ir(dfpp)2(picCOOH)]+ to cell penetrating peptide,
CPP, octaarginine, was conducted in an effort to increase aqueous
solubility and promote the complex's penetration across cell mem-
branes. Comparison of the cell uptake of the iridium parent complex
and the polyarginine conjugate was made in two mammalian cell
lines; SP2 myeloma, a suspension cell line and CHO an adherent line.
Uptake by both cell types was observed after approximately 15 min in-
cubation with the complex at room temperature when DMSO was used
as co-solvent. However, once the iridium luminophore was coupled to
the CPP this permitted its dissolution in buffered aqueous media with-
out the need for an organic solvent. Under these conditions, the
dye-peptide conjugate was rapidly and irreversibly transported the
cell membrane of both SP2 and CHO cells at room temperature. Uptake
was greatest for CHO with almost twice the concentration of the pep-
tide conjugate internalised as for SP2. Transmembrane transport of
the parent complex is believed to have been facilitated by the perme-
abilization of the membrane by DMSO used as co-solvent (0.05% v/v).
However, the dye-conjugate (iridium-R8) is membrane permeable in
water alone, without requirement for permeabilization of the cell
membrane.
Fig. 6. Cytotoxicity induced by DMSO and ethanol on SP2 and CHO cell lines. Cells were
treated with DMSO or ethanol in cell culture media and left overnight at 37 °C in the
dark. Viability was measured by the addition of resazurin reagent for 7 h in the dark
at 37 °C. Results are expressed as a percentage of no treatment, n=3 SD. Similar
results were also obtained using the MTT cell viability assay.
experimental conditions outlined here. Overall, the phototoxicity of the
luminophores is significant even under the imaging conditions used
here. Djurovich et al. [58], also reported that the increased cytotoxicity
of the iridium luminophore in both SP2 and CHO cell lines following
photoexcitation, and attributed this to 1O2 generation due to the long
triplet state lifetimes of such complexes.
3.7. Effects of organic solvents on SP2 myeloma and CHO cell viability
Both parent complex and conjugate are cytotoxic. The IC50
recorded for the cells in the dark was independent, in the case of
the parent complex [Ir(dfpp)2(picCOOH)]+, of the identity of the
cell, with IC50 of 84.8 μM and 88 μM respectively for SP2 and CHO
cells. The IC50 approximately doubled for the polyarginine conjugate
and displayed a significant dependence on cell type with IC50 of
35 μM and 54.1 μM respectively for SP2 and CHO cells. This was in
spite of the fact that the concentration of the conjugate was higher
in CHO. Both parent and conjugate showed high photocytotoxicity
with at least an order of magnitude increase in the rate of cell death
on exposure to light. Preliminary evidence points to a mechanism
involving 1O2 generation by the metal complex.
Finally, as DMSO was used in the media to encourage dissolution of
the iridium(III) parent complex any potential effect such an organic sol-
vent might have on the viability of SP2 and CHO cell lines was also exam-
ined. Fig. 6 illustrates the cytotoxic effect on SP2 myeloma cell lines of
ethanol and DMSO using the resazurin assay. The cytotoxicity of DMSO
proved to be far greater towards SP2 cells with 100% cell cytotoxicity at
a DMSO concentration of 20% in PBS solution using the resazurin assay.
These results are confirmed using a MTT assay (results not shown) that
yielded a comparable cell death at a DMSO concentration of 20%. Con-
versely, a solution containing 20% ethanol only exhibited an SP2 cell pop-
ulation death of approximately 60% using the resazurin assay.
Similar cytotoxicity studies were performed using CHO cells. The
cytotoxicity of ethanol proved to be far greater towards CHO cells
when compared to SP2 cell lines, with 100% cell death with an etha-
nol concentration of ~20% as illustrated in Fig. 6. Additionally,
at lower concentrations of DMSO (1%) there is a considerable increase
in the cytotoxicity towards CHO cells to approximately on aver-
age 30% when compared to the SP2 cell line at similar DMSO
concentrations.
Overall these studies indicate the value of polypeptide conjugates
in metal complex cell delivery. Octaarginine effectively increase
aqueous solubility, promotes cell membrane permeability and
enhance cytotoxcity of the metal complex.
List of Abbreviations
Unfortunately, the use of very low concentrations of organic solvent
remains unavoidable for the parent complex, as solubility of the parent
iridium(III) dye in aqueous buffer is extremely poor. Thus a key advan-
tage, as outlined in this paper, of the conjugation of a metal complex to a
polypeptide is the increased water solubility, membrane penetrating
ability and localisation of the dye-conjugate in SP2 and CHO cells such
conjugation confers such conjugation confers.
[Ir(dfpp)2(picCOOH)]+
iridium(III) bis(2-(2,4-difluorophenyl)pyridinato-N, C2')-
2(4-carboxylphenyl)imidazo[4,5-f][1,10]phenanthroline perchlorate.
[(picCOOH)] ligand
phenanthroline).
[(dfpp)] ligand 2-(4,6-difluorophenyl)pyridine.
2-(4-carboxylphenyl)imidazo[4,5-f][1,10]
CHO
CPP
DCM
DIEA
DMF
Chinese hamster ovary.
Cell penetrating peptide.
Dichloromethane.
N,N-diisopropylethylamine.
Dimethylformamide.
4. Conclusions
The synthesis and photophysical characterization of
cyclometalated iridium(III) polypyridyl complex; iridium(III) bis(2-
a
novel

74
C. Dolan et al. / Journal of Inorganic Biochemistry 119 (2013) 65–74
[17] K.Y. Zhang, S.P. Li, N. Zhu, I.W. Or, M.S. Cheung, Y. Lam, K.K. Lo, Inorg. Chem. 49
HBTU
HOBt
2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate.
1-Hydroxybenzotrizole.
(2010) 2530–2540.
[18] R. Notman, M. Noro, B. O'Malley, J. Anwar, J. Am. Chem. Soc. 128 (2006)
13982–13983.
[19] Z. Yu, P.J. Quinn, Mol. Membr. Biol. 15 (1998) 59–68.
[20] M. Dobroschke, Y. Geldmacher, I. Ott, M. Harlos, L. Kater, L. Wagner, R. Gust, W.S.
Sheldrick, A. Prokop, ChemMedChem 4 (2009) 177–187.
[21] K.K. Lo, M.W. Louie, K.S. Sze, J.S. Lau, Inorg. Chem. 47 (2008) 602–611.
[22] H. Yukawa, Y. Kagami, M. Watanabe, K. Oishi, Y. Miyamoto, Y. Okamoto, M.
Tokeshi, N. Kaji, H. Noguchi, K. Ono, M. Sawada, Y. Baba, N. Hamajima, S.
Hayashi, Biomaterials 31 (2010) 4094–4103.
[23] J.B. Rothbard, S. Garlington, Q. Lin, T. Kirschberg, E. Kreider, P.L. McGrane, P.A.
Wender, P.A. Khavari, Nat. Med. 6 (2000) 1253–1257.
[24] P.A. Wender, J.B. Rothbard, T.C. Jessop, E.L. Kreider, B.L. Wylie, J. Am. Chem. Soc.
124 (2002) 13382–13383.
MALDI-TOF matrix assisted laser desorption ionisation-time of flight
MTT
(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide.
NMP
N-methylpyrrolidinone.
PBS solution Phosphate buffered saline solution.
PyBOP
benzotriazol-1-yl-oxytripyrrolidinophosphonium
hexafluorophosphate.
R8
Octaarginine.
SPPS
TBA
TCSPC
TFA
Solid phase peptide synthesiser.
tetrabutylammonium tetrafluoroborate.
Time correlated single photon counting.
trifluoroacetic acid.
[25] P.A. Wender, D.J. Mitchell, K. Pattabiraman, E.T. Pelkey, L. Steinman, J.B. Rothbard,
Proc. Natl. Acad. Sci. 97 (2000) 13003–13008.
[26] S.M. Fuchs, R.T. Raines, Biochemistry (N.Y.) 43 (2004) 2438–2444.
[27] S. Fawell, J. Seery, Y. Daikh, C. Moore, L.L. Chen, B. Pepinsky, J. Barsoum, Proc. Natl.
Acad. Sci. 91 (1994) 664–668.
UV-Vis UV-visible.
[28] P. Ghosh, X. Yang, R. Arvizo, Z. Zhu, S.S. Agasti, Z. Mo, V.M. Rotello, J. Am. Chem.
Soc. 132 (2010) 2642–2645.
[29] O. Vazquez, J. Blanco-Canosa, M.E. Vazquez, J. Martanez-Costas, L. Castedo, J.L.
Mascareas, ChemBioChem 9 (2008) 2822–2829.
[30] J.L. Cohen, A. Almutairi, J.A. Cohen, M. Bernstein, S.L. Brody, D.P. Schuster, J.M.
Frachet, J. Bioconjug. Chem. 19 (2008) 876–881.
Acknowledgments
[31] S. Fuchs, R.T. Raines, Cell. Mol. Life Sci. 16 (2006) 1819–1822.
[32] S. Futaki, M. Niwa, I. Nakase, A. Tadokoro, Y. Zhang, M. Nagaoka, N. Wakako, Y.
Sugiura, Bioconjug. Chem. 15 (2004) 475–481.
[33] M. Schindler, S. Grabski, E. Hoff, S.M. Simon, Biochemistry (N.Y.) 35 (1996)
2811–2817.
[34] E. Švastová, A. Hulíková, M. Rafajová, M. Zat'ovičová, A. Gibadulinová, A. Casini, A.
Cecchi, A. Scozzafava, C.T. Supuran, J. Pastorek, S. Pastoreková, FEBS Lett. 577
(2004) 439–445.
[35] S. Zhang, M. Hosaka, T. Yoshihara, K. Negishi, Y. Iida, S. Tobita, T. Takeuchi, Cancer
Res. 70 (2010) 4490–4498.
[36] U. Neugebauer, L. Cosgrave, Y. Pellegrin, M. Devocelle, R.J. Forster, T.E. Keyes, Proc.
SPIE 8427 (2012) 84270C.
[37] Y. Pellegrin, R.J. Forster, T.E. Keyes, Inorg. Chim. Acta 362 (2009) 1715–1722.
[38] Y. Pellegrin, R.J. Forster, T.E. Keyes, Inorg. Chim. Acta 361 (2008) 2683–2691.
[39] L. Cosgrave, M. Devocelle, R.J. Forster, T.E. Keyes, Chem. Commun. 46 (2010)
103–105.
[40] M. Atmeh, N. Russell, R.J. Forster, T.E. Keyes, J. Incl. Phenom. Macrocycl. Chem.
(2007) 1607–1612.
[41] Q. Zhao, S. Liu, M. Shi, F. Li, H. Jing, T. Yi, C. Huang, Organometallics 26 (2007)
5922–5930.
[42] S. Lamansky, R.C. Kwong, M. Nugent, P. Djurovich, M.E. Thompson, Org. Electron.
2 (2001) 53–62.
[43] S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq, H. Lee, C. Adachi, P.E.
Burrows, S.R. Forrest, M.E. Thompson, J. Am. Chem. Soc. 123 (2001)
4304–4312.
[44] R. Ragni, E.A. Plummer, K. Brunner, J.W. Hofstraat, F. Babudri, G.M. Farinola, F.
Naso, L. De Cola, J. Mater. Chem. 16 (2006) 1161–1170.
[45] L. Flamigni, J. Collin, J. Sauvage, Acc. Chem. Res. 41 (2008) 857–871.
[46] Y. You, J. Seo, S.H. Kim, K.S. Kim, T.K. Ahn, D. Kim, S.Y. Park, Inorg. Chem. 47
(2008) 1476–1487.
The authors gratefully acknowledge the support from the National
Biophotonics and Imaging Platform, Ireland funded by the Irish
Government's Programme for Research in Third Level Institutions,
Cycle 4, Ireland's EU Structural Funds Programmes 2007–2013. This
material is based upon work supported by the Science Foundation
Ireland under grant no. [10/IN.1/B3025]. TEK and CD gratefully
acknowledge The Irish Research Council for Science, Engineering
and Technology (IRCSET) for a postgraduate scholarship. Professor
Richard O'Kennedy is gratefully acknowledged for providing access
to the cell culture.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.jinorgbio.2012.11.001.
References
[1] S. Hirohara, M. Obata, H. Alitomo, K. Sharyo, T. Ando, S. Yano, M. Tanihara,
Bioconjug. Chem. 20 (2009) 944–952.
[2] C.A. Puckett, J.K. Barton, J. Am. Chem. Soc. 131 (2009) 8738–8739.
[3] C.R. Kowol, R. Trondl, V.B. Arion, M.A. Jakupec, I. Lichtscheidl, B.K. Keppler, Dalton
Trans. 39 (2010) 704–706.
[4] K.K. Lo, M. Louie, K.Y. Zhang, Coord. Chem. Rev. 254 (2010) 2603–2622.
[5] C. Koo, K. Wong, C.W. Man, H. Tam, S. Tsao, K. Cheah, M.H. Lam, Inorg. Chem. 48
(2009) 7501–7503.
[6] C.A. Puckett, J.K. Barton, J. Am. Chem. Soc. 129 (2007) 46–47.
[7] U. Neugebauer, Y. Pellegrin, M. Devocelle, R.J. Forster, W. Signac, N. Moran, T.E.
Keyes, Chem. Commun. 42 (2008) 5307–5309.
[8] C.A. Puckett, J.K. Barton, Biochemistry (N.Y.) 47 (2008) 11711–11716.
[9] C. Tan, S. Lai, S. Wu, S. Hu, L. Zhou, Y. Chen, M. Wang, Y. Zhu, W. Lian, W. Peng, L. Ji,
A.J. Xu, Med. Chem. 53 (2010) 7613–7624.
[10] R. Bieda, I. Ott, R. Gust, W.S. Sheldrick, Eur. J. Inorg. Chem. 25 (2009) 3821–3831.
[11] S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq, R. Kwong, I. Tsyba, M.
Bortz, B. Mui, R. Bau, Inorg. Chem. 40 (2001) 1704–1711.
[12] M. Licini, J.A.G. Williams, Chem. Commun. 19 (1999) 1943–1944.
[13] Q. Zhao, M. Yu, L. Shi, S. Liu, C. Li, M. Shi, Z. Zhou, C. Huang, F. Li, Organometallics
29 (2010) 1085–1091.
[14] C. Ulbricht, C. Friebe, A. Winter, U.S. Schubert, Adv. Mater. 21 (2009) 4418–4441.
[15] M. Yu, Q. Zhao, L. Shi, F. Li, Z. Zhou, H. Yang, T. Yia, C. Huang, Chem. Commun. 18
(2008) 2115–2117.
[47] J.G. Vos, Polyhedron 11 (1992) 2285–2299.
[48] J.L. Kahl, K.W. Hanck, K. DeArmond, J. Phys. Chem. 82 (1978) 540–545.
[49] H. Maeda, Y. Yamauchi, M. Yoshida, H. Ohmori, Anal. Sci. 11 (1995) 947–952.
[50] F. Neve, A. Crispini, S. Campagna, S. Serroni, Inorg. Chem. 38 (1999) 2250–2258.
[51] T.J. Anchordoguy, J.F. Carpenter, J.H. Crowe, J.L.M. Crowe, L.M. Biochim, Biophys.
Acta 1104 (1992) 117–122.
[52] Q. Zhao, C. Huang, F. Li, Chem. Soc. Rev. 40 (2011) 2508–2524.
[53] Y. Takechi, H. Yoshii, M. Tanaka, T. Kawakami, S. Aimoto, H. Saito, Langmuir 27
(2011) 7099–7107.
[54] L. Xiong, Q. Zhao, H. Chen, Y. Wu, Z. Dong, Z. Zhou, F. Li, Inorg. Chem. 49 (2010)
6402–6408.
[55] S. Li, H. Liu, K. Zhang, K.K. Lo, Chem. Eur. J. 16 (2010) 8329–8339.
[56] S.W. Jones, R. Christison, K. Bundell, C. Voyce, S.M.V. Brockbank, P. Newham, M.A.
Lindsay, J. Pharmacol. 145 (2005) 1093–1102.
[57] R.S. Bodannes, P.C. Chan, FEBS Lett. 105 (1979) 195–196.
[58] P.I. Djurovich, D. Murphy, M.E. Thompson, B. Hernandez, R. Gao, P.L. Hunt, M.
Selke, Dalton Trans. 34 (2007) 3763–3770.
[16] W. Jiang, Y. Gao, Y. Sun, F. Ding, Y. Xu, Z. Bian, F. Li, J. Bian, C. Huang, Inorg. Chem.
49 (2010) 3252–3260.