Selective DNA purine base photooxidation by bis-terdentate iridium(III) polypyridyl and cyclometalated complexes

Article abstract of DOI:10.1021/ic402476b

Two bis-terdentate iridium(III) complexes with polypyridyl and cyclometalated ligands have been prepared and characterized. Their spectroscopic and electrochemical properties have been studied, and a photophysical scheme addressing their properties is proposed. Different types of excited states have been considered to account for the deactivation processes in each complex. Interestingly, in the presence of mono- or polynucleotides, a photoinduced electron-transfer process from a DNA purine base (i.e., guanine or adenine) to the excited complex is shown through luminescence quenching experiments. For the first time, this work reports evidence for selective DNA purine bases oxidation by excited iridium(III) bis-terdentate complexes.

Full text of DOI:10.1021/ic402476b

Article
pubs.acs.org/IC
Selective DNA Purine Base Photooxidation by Bis-terdentate
Iridium(III) Polypyridyl and Cyclometalated Complexes
Alexandre Jacques, Andree Kirsch-De Mesmaeker,† and Benjamin Elias*
́
̀ ́ ́
Institut de la Matiere Condensee et des Nanosciences, Universite Catholique de Louvain, Place Louis Pasteur 2, Box L4.01.02, B-1348
Louvain-la-Neuve, Belgium
S
* Supporting Information
ABSTRACT: Two bis-terdentate iridium(III) complexes with poly-
pyridyl and cyclometalated ligands have been prepared and
characterized. Their spectroscopic and electrochemical properties
have been studied, and a photophysical scheme addressing their
properties is proposed. Different types of excited states have been
considered to account for the deactivation processes in each complex.
Interestingly, in the presence of mono- or polynucleotides, a
photoinduced electron-transfer process from a DNA purine base
(i.e., guanine or adenine) to the excited complex is shown through
luminescence quenching experiments. For the first time, this work
reports evidence for selective DNA purine bases oxidation by excited
iridium(III) bis-terdentate complexes.
complexes obtained from the coordination of either two
terpyridyl-like ligands {i.e., [Ir(4′-COOH-Phtpy)(4′-MeO-
Phtpy)]3+ [4′-COOH-Phtpy = 4′-(4-carboxyphenyl)-
2,2′:6′,2″-terpyridine and 4′-MeO-Phtpy = 4′-(4-methoxyphen-
yl)-2,2′:6′,2″-terpyridine] called hereafter [IrN6]3+ for the sake
of clarity; Figure 1} or one terpyridyl and one triphenylpyridyl
ligands {i.e., [Ir(4′-COOH-Phtpy)(4′-MeO-tppy)]+ [4′-MeO-
tppy = 2,6-diphenyl-4-(4′-methoxyphenyl)pyridine] called
hereafter [IrN4C2]+ for the sake of clarity; Figure 1}. Their
photophysics and photochemistry have been investigated.
Although both complexes display very similar structural
features, their photoredox behavior toward DNA bases and
the excited state involved in the electron transfer process are
quite different depending on the coordination sphere. Using
luminescence measurements in the absence and presence of
several DNA building blocks, synthetic or natural DNA, we
were able to show that these complexes selectively photooxidize
purine nucleobases (i.e., adenine and guanine), whereas no
nucleobase reduction occurs under irradiation. This paper
represents, to the best of our knowledge, one of the very few
studies addressing photoinduced electron-transfer processes
between excited bis-terdentate iridium(III) complexes and
DNA bases.
INTRODUCTION
■
Polyazaaromatic transition-metal complexes have gained
considerable interest these past decades. Their use as catalysts
in the water splitting process,1 therapeutic agents in several
cancer treatments,2 or photoprobes of biological media3 is well-
established. Among all transition metals available, iridium(III)
is one of the most interesting elements for building complexes.4
Indeed, it allows the synthesis of multiple polypyridyl and
cyclometalated complexes with bidentate or terdentate ligands.
Iridium(III) is capable of forming stable mono-, bis-, and tris-
cyclometalated complexes, and up to three C-donor sites can be
coordinated onto the metal center. The C/N ratio present in
the first coordination sphere can thus be varied from 0/6 to 3/
3, hence providing an excellent means to tune the photoredox
properties of the resulting complexes. For instance, tris-
bidentate iridium(III) complexes made of two PPY (PPY =
2-phenylpyridine) and an imidazo[4,5-f ][1,10]phenanthroline
have C/N = 2/4 and are specific anion sensors.5 Upon
replacement of the imidazolo ligand by an intercalative ligand
dppz (dppz = dipyrido[3,2-a:2′,3′-c]phenazine), the resulting
luminescent iridium(III) has been used to study DNA-
mediated hole and electron transfer.6 Recently, it has been
shown that a luminescent bis-terdentate iridium(III) complex
made exclusively of terpyridine-type-like ligands (C/N = 0/6)
exhibits intercalation between DNA base pairs.7 The
luminescence of this complex is quenched in the presence of
a guanine moiety via a photoinduced electron-transfer process,
showing the high oxidizing ability of the excited state. Although
displaying spectroelectrochemical properties enabling photo-
reactions toward DNA, iridium(III) complexes have only been
scarcely used as such in the literature. Therefore, in this paper,
we present the comprehensive study of two novel iridium(III)
EXPERIMENTAL SECTION
■
Materials and Instrumentation. 4′-(Methoxyphenyl)-2,2′:6′,2″-
terpyridine8 (4′-MeO-Phtpy), [(4′-COOMe-Phtpy)IrCl3],9 [(4′-
COOH-Phtpy)IrCl3],10 and 2,6-diphenyl-4-(4′-methoxyphenyl)-
pyridine11 (4′-MeO-tppy) were prepared as previously described.
Deoxynucleotides (dGMP, dAMP, dCMP, and dTMP) were
Received: October 1, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic402476b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(acetone/water, from 100/0 to 90/10), which yielded the complex
−
[IrN6]3+ (PF6 salt) as a yellow solid (40 mg, 82%). When the
chloride salt was required for the experiments, counteranion exchange
−
was achieved upon the addition of tBu4N+Cl− to the PF6 salt
1
dissolved in acetone. H NMR (CD3CN, 500 MHz): δ 9.11 (s, 2H,
H3′ and H5′ of tpy-C6H4-COOH), 9.03 (s, 2H, H3′ and H5′ of tpy-
C6H4-OMe), 8.72 (d, 2H, H3 and H3″ of tpy-C6H4-COOH, JH3−H4
=
8.1 Hz), 8.70 (d, 2H, H3 and H3″ of tpy-C6H4-OMe, JH3−H4 = 8.1 Hz),
8.40 (d, 2H, Ha of tpy-C6H4-COOH, JHa−Hb = 8.4 Hz), 8.28 (d, 2H, Hb
of tpy-C6H4-COOH, JHb−Ha = 8.4 Hz), 8.25−8.20 (m, 6H, H4 and H4″
of both tpy’s, Hb of tpy-C6H4-OMe), 7.70 (dd, 2H, H6 and H6″ of tpy-
C6H4-OMe, JH6−H5 = 4.9 Hz, JH6−H4 = 0.9 Hz), 7.68 (dd, 2H, H6 and
H6″ of tpy-C6H4-OMe, JH6−H5 = 4.9 Hz, JH6−H4 = 0.9 Hz), 7.51−7.46
(m, 4H, H5 and H5″ of both tpy’s), 7.34 (d, 2H, Ha of tpy-C6H4-
COOH, JHa−Hb = 8.8 Hz), 3.99 (s, 3H, OMe). HRMS (ESI). Calcd for
[C44H32N6O3193IrP2F12 − PF6]+: m/z 1175.1443. Found: m/z
1175.1429. Calcd for [C44H32N6O3193Ir − 3PF6]3+: m/z 295.0716.
Found: m/z 295.0723.
[Ir(4′-COOH-Phtpy)(4′-MeO-tppy)](NO3) ([IrN4C2](NO3)). A
mixture of [(4′-COOMe-Phtpy)IrCl3] (98 mg, 0.15 mmol), 4′-
MeO-tppy (54 mg, 0.16 mmol), and AgNO3 (125 mg, 0.74 mmol)
was refluxed in degassed ethylene glycol (13 mL) in the dark and
under argon for 22 h. After cooling, the mixture was filtered through
Celite to remove AgCl. The Celite was washed with MeOH and
subsequently evaporated. Water was added, and the aqueous layer was
extracted with CHCl3 (three times). The organic solutions were
combined and evaporated to dryness. The crude product was purified
by column chromatography on SiO2 [CH3CN/saturated aqueous
KNO3 (10/1)]. The first eluted orange band was recovered and the
solvent evaporated. The resulting red solid was saponified in a mixture
MeOH (30 mL)/1 M NaOH (20 mL), and the solution was heated to
80 °C for 3 h. After cooling, the solution was adjusted to pH = 7 with
2 M HCl and solvents were evaporated. The solid was purified by
column chromatography on SiO2 [CH3CN/saturated aqueous KNO3
(10/1)]. The red powder was washed with water and dried under
vacuum to afford [IrN4C2]+ with 40% yield (63 mg). 1H NMR
(CD3CN, 300 MHz): δ 9.00 (s, 2H, H3′ and H5′ of tpy-C6H4-
COOH), 8.67 (d, 2H, H3 and H3″ of tpy-C6H4-COOH, JH3−H4 = 7.8
Hz), 8.36 (d, 2H, Ha of tpy-C6H4-COOH, JHa−Hb = 8.6 Hz), 8.29 (s,
2H, H3′ and H5′ of tppy-OMe), 8.28 (d, 2H, Hb of tpy-C6H4-COOH,
JHb−Ha = 8.6 Hz), 8.13 (d, 2H, Hb of tppy-OMe, JHb−Ha = 8.4 Hz),
8.03−7.96 (m, 4H, H4 and H4′′ of tpy-C6H4-COOH and H3 and H3′′
Figure 1. Structures of the iridium(III) complexes: (a) [IrN6]3+; (b)
[IrN4C2]+.
purchased from Sigma-Aldrich (purity ≥99%) and used without
further purification. Calf thymus DNA (CT-DNA) was purchased
from Sigma-Aldrich, and [poly(dG-dC)]2, [poly(dA-dT)]2, and
[poly(dA)-poly(dT)] were purchased from Amersham Bioscience.
Polynucleotide phosphate concentrations were determined spectro-
scopically (ε260 nm = 6600 M−1 for CT-DNA,12 ε262 nm = 6600 M−1 for
[poly(dA-dT)]2,13 ε260 nm = 6000 M−1 for [poly(dA)-poly(dT)],14 and
ε254 nm = 8400 M−1 for [poly(dG-dC)]2).15 1H NMR experiments
were performed in CD3CN on a Bruker AC-300 Avance II (300 MHz)
or on a Bruker AM-500 (500 MHz). The chemical shifts (given in
ppm) were measured versus the residual peak of the solvent as the
internal standard. High-resolution mass spectrometry (HRMS) were
recorded on a Q-Extractive orbitrap from ThermoFisher. Samples
were ionized by electrospray ionization (ESI; capillary temperature =
250 °C, vaporizer temperature = 250 °C, sheath gas flow rate = 20).
UV−vis absorption spectra were recorded on a Shimadzu UV-1700.
Room temperature fluorescence spectra were recorded on a Varian
Cary Eclipse instrument. Luminescence lifetimes were measured with a
modified Applied Photophysics laser kinetic spectrometer by exciting
the samples at 355 nm with a Nd:YAG pulsed laser (Continuum NY
61-10). Emitted light as a function of time was detected with a R-928
Hamamatsu photomultiplier tube whose output was applied to a
digital oscilloscope (Hewlett-Packard HP 54200A) interfaced with a
Dell Dimension DE051 computer. Signals were averaged over at least
16 shots and corrected for baseline. Igor 6.1 software was used for
decay analyses. Cyclic voltammetry was carried out in a one-
compartment cell, using a glassy carbon disk working electrode
(approximate area = 0.03 cm2), a platinum wire counter electrode, and
an Ag/AgCl reference electrode. The potential of the working
electrode was controlled by an Autolab PGSTAT 100 potentiostat
through a PC interface. The cyclic voltammograms were recorded with
a sweep rate of 100 mV s−1, either in dried acetonitrile (Acros, HPLC
grade) or in dried N,N-dimethylformamide (DMF; Sigma-Aldrich,
HPLC grade). Tetrabutylammonium perchlorate (0.1 M) was used as
the supporting electrolyte, and the samples were purged by argon
before each measurement.
of tppy-OMe), 7.89 (d, 2H, H6 and H6′′ of tpy-C6H4-COOH, JH6−H5
5.1 Hz), 7.31−7.24 (m, 4H, H5 and H5′′ of tpy-C6H4-COOH and Ha
of tppy-OMe), 7.01 (td, 2H, H4 and H4′′ of tppy-OMe, JH4−H5 (H3)
=
=
7.4 Hz, JH4−H6 = 1.1 Hz), 6.78 (td, 2H, H5 and H5′′ of tppy-OMe,
JH5−H4 (H6) = 7.3 Hz, JH5−H3 = 1.0 Hz), 6.28 (dd, 2H, H6 and H6′′ of
tppy-OMe, JH6−H5 = 7.3 Hz, JH6−H4 = 0.9 Hz), 3.97 (s, 3H, OMe).
HRMS (ESI). Calcd for [C46H32N4O3193Ir − NO3]+: m/z 881.2098.
Found: m/z 881.2088.
RESULTS
■
Synthesis. The synthesis of both complexes was achieved in
two steps by heating iridium(III) chloride salts with the desired
ligand in ethylene glycol according to methodologies previously
described.9,16 It is worth mentioning that the synthesis of
[IrN4C2]+ requires harsher conditions than that of [IrN6]3+, as
expected from the greater inertness of the cyclometalated
ligand. [IrN6]3+ and [IrN4C2]+ display very similar structural
features. Thanks to the derivatization on the 4′ position, they
have several symmetry elements, hence circumventing the
problem of the occurrence of geometrical isomers as
encountered for tris-bidentate arrangements. Both complexes
have thus been unambiguously characterized by 1H NMR
spectroscopy and HRMS (see the Experimental Section).
Electrochemistry. The redox potentials of both complexes
were determined by cyclic voltammetry measurements. The
data are gathered in Table 1. [IrN6]3+ displays two irreversible
[Ir(4′-COOH-Phtpy)(4′-MeO-Phtpy)](PF6)3 ([IrN6](PF6)3). A sus-
pension of [(4′-COOH-Phtpy)IrCl3] (24 mg, 0.037 mmol) and 4′-
MeO-Phtpy (12 mg, 0.037 mmol) was heated to 170 °C in degassed
ethylene glycol (2.5 mL) in the dark and under argon for 30 min. After
cooling, aqueous NH4PF6 was added to precipitate the crude product.
The pure complex was isolated by column chromatography on Al2O3
B
dx.doi.org/10.1021/ic402476b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Electrochemical Data Measured at Room
Temperature in DMF for [IrN6]3+ and in CH3CN for
[IrN4C2]+ with 0.1 M Bu4NClO4 as the Supporting
Table 2. Absorption and Luminescence Data for the
Iridium(III) Complexes
[IrN6]3+
[IrN4C2]+
a
Electrolyte
λmax Abs, nm (ε,
×10−3 M−1
CH3CN
CH3CN
382 (28.1), 319 (40.8)
, 279 (52.7),
512 (5.2),
[IrN6]3+
[IrN4C2]+
439 (8.4),
325 (37.4),
281 (102.8)
cm−1
)
252 (59.0)
Eox (V vs Ag/AgCl)
Ered (V vs Ag/AgCl)
Eox* (V vs Ag/AgCl)
>2.2
1.14 (r, ΔEp = 58 mV)
a
−0.81 (ir), −1.22 (ir) −1.24 (r, ΔEp = 54 mV)
−1.22
λmax Em (nm)
568
528
697
under air at
298 K
Ered* (V vs Ag/AgCl) 1.38
1.12
b
H2O
560
505
514
505
682
663
a
r = reversible; ir = irreversible; ΔEp (the peak-to-peak separation) is
a
λmax Em (nm) at EtOH/
close to 59 mV.
77 K
MeOH
(4/1)
τ (μs) under air CH3CN
0.47
1.00
0.26
one-electron-reduction waves at −0.81 and −1.22 V vs Ag/
AgCl (Figure S1 in the SI). No oxidation process could be
detected up to +2.2 V vs Ag/AgCl, suggesting that the
oxidation is highly energetic and probably metal-centered. In
contrast, [IrN4C2]+ is characterized by one reversible reduction
wave and one reversible oxidation wave, at −1.24 and +1.14 V
vs Ag/AgCl, respectively (Figures S2 and S3 in the SI). The
oxidation of [IrN4C2]+ is probably due to the cyclometalated
iridium fragment (Ir−C^N^C). The potentials in the excited
state (namely, Eox* and Ered*)17 for both complexes have been
estimated from the ground-state redox potentials and the
energy of the excited state corresponding to the maximum of
the emission spectrum at 298 K, i.e., λmax Em = 568 nm (2.19
eV) for [IrN6]3+ and λmax Em = 528 nm (2.36 eV) for
[IrN4C2]+. Not surprisingly, [IrN6]3+ is strongly oxidizing
(Ered* = +1.38 V vs Ag/AgCl) in its excited state.
at 298 K
b
H2O
1.31
1.76
0.39
c
Φ
CH3CN
0.0010
0.0019
0.0055
0.0291
5.5
0.0006
0.0227
2.3
b
H2O
d
kr (×10−3 s−1
)
CH3CN
2.2
a
b
Excitation wavelength = 350 nm. [IrN4C2]+ was dissolved in a
H2O/CH3CN (90/10) mixture because of its low solubility in water.
c
The quantum yield ( 5%) was measured under air using [Ru-
(bpy)3]2+ as a reference; ϕref = 0.028 in CH3CN,18 λExc = 420 nm, and
293 K. Radiative rate constant at 298 K (kr = Φ/τ).
d
luminescence intensity with respect to degassed solutions is
observed, showing a sensitization of 1O2 by excited iridium(III)
complexes. Because all of our experiments have been carried
out under the same conditions, the influence of oxygen is kept
constant and is therefore not further investigated in the present
manuscript.
Absorption and Emission Spectroscopy. Absorption
and emission spectra in CH3CN under air at room temperature
for both complexes are shown in Figure 2. Table 2 gathers the
corresponding data. For both complexes, absorption (ε >
30000 M−1 cm−1) bands in the UV region tailing into the
visible are observed.
Luminescence Behavior of Iridium(III) Complexes in
the Presence of Deoxynucleotides. With respect to the
excited-state potentials of iridium(III) complexes and the ability
of some DNA bases to act as an electron donor or acceptor, a
photooxidative or photoreductive electron-transfer process
could be thermodynamically possible between the excited
complex and some DNA building blocks (deoxynucleotides).
Thanks to the ability of iridium complexes to emit in aqueous
solution, monitoring the luminescence in the presence of
increased concentration of a given nucleotide can be
performed. It is worth mentioning that all of the experiments
presented here were carried out with nucleobase concentrations
far below the aggregation and self-association limit.19 In
addition, because of the poor solubility of [IrN4C2]+ in pure
water, 10% CH3CN (v/v) was added for each experiment with
deoxynucleotides. Figure 3 shows the relative emission decrease
of both [IrN6]3+ and [IrN4C2]+ upon increased concentrations
of deoxyguanosine-5′-monophosphate (dGMP). It is striking
that solely the higher-energy emission band (λem = 514 nm) for
[IrN4C2]+ is quenched, whereas the other band (λem = 682 nm)
is not affected by the presence of dGMP. A linear Stern−
Volmer relationship in the luminescence intensity can be
obtained from these data (Figure 3, inset), suggesting that pure
dynamic quenching20 of the excited state of both [IrN6]3+ and
[IrN4C2]+ is occurring in the presence of dGMP.
Figure 2. Absorption (solid lines) and emission (dashed lines; λexc
=
350 nm) for [IrN6]3+ (green) and [IrN4C2]+ (red) complexes in
CH3CN under air at room temperature.
[IrN6]3+ and [IrN4C2]+ emit in CH3CN under air (Φ ∼
10−3). A broad-band emission centered at around 568 nm
(τRT,air = 470 ns) is observed for [IrN6]3+, whereas [IrN4C2]+
exhibits a dual-band emission, centered at 528 nm (τRT,air = 1
μs) and 697 nm (τRT,air = 260 ns). Although there is almost no
emission band shift between CH3CN and water, the excited-
state lifetimes are affected upon solvent change and increase by
a factor of 2 or 3 in water for both complexes. It is worth
mentioning that, in an air-equilibrated solvent, a decrease in the
In the presence of deoxyadenosine-5′-monophosphate
(dAMP), a luminescence quenching and a decrease of the
excited-state lifetime (τ0/τ > 1) are observed only for [IrN6]3+.
However, in that case, a negative deviation from the Stern−
Volmer linear relationship in intensity is observed (Figure 4).
This presumably shows that the luminescence quenching of
[IrN6]3+ in the presence of dAMP is more complicated than a
C
dx.doi.org/10.1021/ic402476b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
[IrN4C2]+ in aqueous buffer, we have decided to limit our study
of the influence of polynucleotides on the spectroscopic
properties of the complex to [IrN6]3+.
Although no significant change in the absorption spectra of
[IrN6]3+ is observed when interacting with the polynucleotide,
its luminescence is dramatically altered. Figure 5 shows the
Figure 5. Relative luminescence intensity of [IrN6]3+ measured at λem
= 560 nm in the presence of increased concentrations of [poly(dA-
dT)] (blue circle), [poly(dA)-poly(dT)] (green circle), CT-DNA
2
(black circle), and [poly(dG-dC)]2 (red circles). All of the
measurements were performed with a constant complex concentration
of 32 μM in a 50 mM Tris-HCl buffer aqueous solution (pH = 7.4)
with 50 mM NaCl (λexc = 350 nm).
relative luminescence intensity changes I/I0 for [IrN6]3+ upon
the addition of selected polynucleotides. Two distinct behaviors
are observed. In the presence of GC base pairs containing
polynucleotides (i.e., [poly(dG-dC)2], 100% GC base pairs;
CT-DNA, 42% GC base pairs22), a luminescence quenching is
observed. In contrast, in the presence of AT base pairs
containing polynucleotides (i.e., [poly(dA-dT)2] or [poly(dA)-
poly(dT)], 100% AT base pairs), a somewhat more complex
behavior is observed with a strong luminescence increase
followed by a plateau value of I/I0 ≈ 3 for [poly(dA-dT)2] or
followed by a decrease for [poly(dA)-poly(dT)].
Figure 3. Luminescence spectra (λexc = 350 nm) of (a) [IrN6]3+ and
(b) [IrN4C2]+ in the presence of increased concentration of dGMP
(50 mM Tris buffer, pH = 7.4, RT).21 Inset: Stern−Volmer plots
obtained upon the addition of dGMP for (a) [IrN6]3+ and (b)
[IrN4C2]+ measured at λem = 560 and 514 nm, respectively.
DISCUSSION
■
Photophysics of the Complexes Alone in Solution.
Iridium(III) complexes are well-known in the literature for their
rich photophysics and photochemistry. Indeed, because of a
strong spin−orbit coupling and an important influence of the
ancillary ligands, a mixture of different excited states usually
participates in the radiative deactivation of the complex. In the
case of [Ir(tpy)2]3+ and parent compounds, it has been shown
that the emission arises from a ligand-centered (3LC) excited
state, with a small but nonnegligible contribution of a metal-to-
ligand charge-transfer (3MLCT) excited state.16
Figure 4. Stern−Volmer plots measured using the excited-state
luminescence lifetime (dashed line) and luminescence intensity (solid
line) of [IrN6]3+ upon increased concentration of dAMP (50 mM Tris
buffer, pH = 7.4, RT) at λem = 560 nm (λexc = 350 nm).
For [IrN6]3+, a change in the solvent polarity (i.e., from
CH3CN to water) does not influence the emission maximum,
in agreement with a 3LC excited state. However, the
luminescence at room temperature in a polar fluid solvent is
unstructured, decays in about 1 μs, and is blue-shifted at 77 K
(Table 2), which is more in favor of a 3MLCT excited state. In
addition, the kr value (≈103 s−1) is relatively high, and usually
simple dynamic diffusion between the excited complex and
dAMP monomers.20 The addition of dCMP or dTMP to
[IrN6]3+ or [IrN4C2]+ did not affect their luminescence
behavior, excluding thus any photoinduced process involving
these nucleobases.
23
3
such values are much smaller for pure LC states. Therefore,
Luminescence Behavior of [IrN6]3+ in the Presence of
Polynucleotides. Because of the relatively low solubility of
we propose that the lowest excited state in [IrN6]3+
corresponds to a LC centered on the electron-poorest ligand,
3
D
dx.doi.org/10.1021/ic402476b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
3
3
i.e., [π → π*(4′-COOH-Phtpy)], mixed with some MLCT
usually arises when static quenching is operative, but in that
particular case, an upward curvature is observed. Instead, for
[IrN6]3+, a downward curvature indicates that other processes
have to be taken into account. A similar behavior has been
reported in the literature for the fluorescence quenching of
doxorubicin in the presence of AMP.28 Two conformers (i.e.,
with one or two intramolecular hydrogen-bonded hydro-
phenolic moieties) of doxorubicin in the ground state with
different accessibility to the quencher were shown to be
responsible for the emission quenching behavior. In the present
case, (de)protonation of the carboxylic acid moiety could
generate two ground states with a different quenching constant
by dAMP. Experiments at different pH values in buffered
aqueous solution showed that the proton concentration of the
medium has only a small influence on the quenching behavior
(Figure S5 in the SI). Moreover, converting the acid function
into an ester has no effect (Figures S6 and S7 in the SI).
Therefore, it is likely that the behavior of [IrN6]3+ in the
presence of dAMP originates from the complex mixture of
excited states participating in the emission, with each state
being successively reductively quenched by dAMP with
different rate constants. In the presence of oligonucleotides,
the change of luminescence of [IrN6]3+ strongly depends on
the nature of the nucleobase content. In agreement with the
behavior observed in the presence of dGMP, quenching of
excited [IrN6]3+ with [poly(dG-dC)]2 or CT-DNA would arise
from the photoinduced electron-transfer process from a G unit
of the polynucleotide to the excited complex, hence generating
oxidative damage within the DNA double helix. The plateau
value, reflecting the quenching efficiency, does not seem to
depend on the percentage of GC of the polynucleotide, in
contrast to G photooxidizing ruthenium(II) complexes.22 This
shows that the quenching of the excited state of [IrN6]3+ is
highly efficient, in agreement with its very high oxidation power
when irradiated by light.
3
character, [dπ(Ir) → π*(4′-COOH-Phtpy)].
In contrast, the cyclometalated compound [IrN4C2]+
displays a different photophysical behavior. Compared to
[IrN6]3+, the absorption is shifted to the red, in agreement with
the greater donating ability of the cyclometalated ligand. In
addition, a dual-band emission is observed. According to
experimental24 and theoretical25 data, iridium(III) complexes
made from terdentate cyclometalated ligands usually have an
emitting state corresponding to a ligand-to-ligand charge
transfer (LLCT), often combined with other 3LC/3MLCT
excited states.
In the case of [IrN4C2]+, the emission at λem ≈ 520 nm is
similar to that of [IrN6]3+ described above. Therefore, we
ascribe this transition to a mixture LC/3MLCT, i.e., [dπ(Ir)
→ π*(4′-COOH-Phtpy) and π(4′-COOH-Phtpy) → π*(4′-
COOH-Phtpy)]. The most bathochromic band (λem = 697 nm)
3
3
3
3
is believed to correspond to a LLCT, i.e., [cyclometalated
iridium fragment (Ir−C^N^C) → π*(4′-COOH-Phtpy)].
Indeed, the emission at 77 K is poorly resolved, in agreement
with a CT process.23 Moreover, the addition of protons (e.g., 2
M HNO3; Figure S4 in the SI) in the solution leads to a
complete disappearance of this 3LLCT emission, probably
arising from protonation of the −OMe group, hence inhibiting
any CT processes involving the electron-donating Ir−C^N^C
fragment.
Photochemical Behavior in the Presence of Mono-
and Polynucleotides. Although the total emission of
[IrN6]3+ is quenched in the presence of a G unit, solely one
emission band of [IrN4C2]+ undergoes a quenching process
upon the addition of GMP (Figure 3b). On the basis of
thermodynamic considerations, this luminescence quenching
for both complexes could be ascribed to a photoinduced
electron-transfer process from the G unit to the excited
complexes (eqs 1 and 2).
In contrast, in the presence of adenine-rich polynucleotides
such as [poly(dA-dT)]2 or [poly(dA)-poly(dT)], a lumines-
cence increase is recorded, in spite of the sufficiently reducing
power of A for quenching the excited state of [IrN6]3+ (vide
supra; Figure 4). The luminescence of the complex in the
presence of these specific polynucleotides probably originates
from two antagonist processes: (i) protection of the complex
from the solvent molecules and from oxygen by the DNA
hydrophobic environment (via intercalation or groove binding,
for instance), resulting in a luminescence increase, and (ii)
quenching by the A moieties, resulting in a luminescence
decrease. In the case of the alternating copolymeric [poly(dA-
dT)]2, the luminescence enhancement is monotonic, reflecting
a constant involvement of both processes at any [poly]/
[IrN6]3+ ratio, with the protection effect being predominant.
For the homopolymeric [poly(dA)-poly(dT)], a different
behavior is observed; this could originate from a different
type of interaction with the double helix, depending on the
polynucleotide loading level by the complex (Figure 5). Thus,
because of these structural changes, at [poly]/[IrN6]3+ > 40, the
quenching by the A units would become predominant
compared to their protection, which would account for the
inflection observed in the titration curve.
[IrN6]3+* + G → [IrN6]2+ + G•+
(1)
[IrN C2]+* + G → [IrN C2] + G•+
(2)
4
4
Indeed, although the G oxidation potential has been debated
for a long time in the literature, it is generally accepted that Eox
of isolated G is close to +1.10 V vs Ag/AgCl and even less
positive when stacked to other G units (Eoxpoly(dG-dC)2] <
+0.85 V vs Ag/AgCl).26 Considering the oxidation power of the
excited state of [IrN6]3+ and using the empirical Rehm−Weller
equation,27 it is expected that process (1) will be exergonic by
about 0.3 eV.17 In the case of [IrN4C2]+, solely the most
energetic emission band (λem = 514 nm) is quenched via a
photoreductive electron transfer (eq 2), favored by less than 0.1
eV based on thermodynamic assumptions. The low oxidizing
power of the less energetic excited state (LLCT) does not allow
any G oxidation. Efficient quenching in the presence of G units
is also in agreement with the high value of the quenching rate
constant (kq) obtained from the Stern−Volmer plot of [IrN6]3+
and [IrN4C2]+ with dGMP (vide supra; Stern−Volmer plots,
Figure 3, inset; kq = 3.9 × 109 M−1 s−1 for [IrN6]3+ and 5.3 ×
108 M−1 s−1 for [IrN4C2]+), both close to the diffusion limit.
The behavior of [IrN6]3+ in the presence of A units is less
common. The emission lifetimes dramatically decrease upon
the addition of dAMP, evidencing an efficient quenching of the
excited state with kq = 5.3 × 108 M−1 s−1 (Figure 4). However,
a deviation from the linearity of the Stern−Volmer relationship
for the emission intensity (I0/I) is observed (Figure 4). This
CONCLUSION
■
We managed to synthesize two new bis-terdendate iridium(III)
complexes. Both have been obtained thanks to successive
addition of the ligands on the metal center. The spectroscopic
E
dx.doi.org/10.1021/ic402476b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Kelly, J. M. Coord. Chem. Rev. 2011, 255, 2666−2675. (f) Lo, K. K-W.;
Zhang, K. Y. RSC Adv. 2012, 2, 12069−12083.
(4) (a) Dixon, I. M.; Collin, J.-P.; Sauvage, J.-P.; Flamigni, L.;
Encinas, S.; Barigelletti, F. Chem. Soc. Rev. 2000, 29, 385−391.
(b) Flamigni, L.; Collin, J.-P.; Sauvage, J.-P. Acc. Chem. Res. 2008, 41,
857−871. (c) Baranoff, E.; Collin, J.-P.; Flamigni, L.; Sauvage, J.-P.
Chem. Soc. Rev. 2004, 33, 147−155.
(5) (a) Zhao, Q.; Liu, S.; Shi, M.; Li, F.; Jing, H.; Yi, T.; Huang, C.
Organometallics 2007, 26, 5922−5930. (b) Lo, K. K-W.; Louie, M.-W.;
Zhang, K. Y. Coord. Chem. Rev. 2010, 254, 2603−2622.
(6) (a) Elias, B.; Genereux, J. C.; Barton, J. K. Angew. Chem., Int. Ed.
2008, 47, 9067−9070. (b) Shao, F.; Elias, B.; Barton, J. K. Inorg. Chem.
2007, 46, 10187−10199. (c) Shao, F.; Barton, J. K. J. Am. Chem. Soc.
2007, 129, 14733−14738.
(7) Campagna, S.; Cavazzini, M.; Cusumano, M.; Di Pietro, M. L.;
Giannetto, A.; Puntoriero, F.; Quici, S. Inorg. Chem. 2011, 50, 10667−
10672.
(8) Wang, J.; Hanan, G. S. Synlett 2005, 8, 1251−1254.
(9) Shinpuku, Y.; Inui, F.; Nakai, M.; Nakabayashi, Y. J. Photochem.
Photobiol. A 2011, 222, 203−209.
(10) Goldstein, D. C.; Chen, Y. Y.; Schmidt, T. W.; Bhadbhade, M.;
Thordarson, P. Dalton Trans. 2011, 40, 2053−2061.
(11) Adib, M.; Tahermansouri, H.; Koloogani, S. A.; Mohammadi, B.;
Bijanzadehb, H. R. Tetrahedron Lett. 2006, 47, 5957−5960.
(12) Felsenfeld, G.; Hirschman, S. Z. J. Mol. Biol. 1965, 13, 407−427.
(13) Inman, R. B.; Baldwin, R. L. J. Mol. Biol. 1962, 172−184.
(14) Cui, T.; Wei, S.; Brew, K.; Leng, F. J. Mol. Biol. 2005, 352, 629−
645.
and electrochemical data revealed that several excited states
contribute to the overall luminescence. Oxidative quenching of
the [IrN6]3+ emission in the presence of dGMP or dAMP
confirms the high oxidizing ability of the excited complex. The
behavior of [IrN4C2]+ is less common. Indeed, a selective
quenching of the most energetic excited state is observed in the
presence of dGMP. No quenching in the presence of dTMP or
dCMP was evidenced for both complexes. This shows that
these two iridium(III) complexes are not sufficiently reducing
(or oxidizing) in their excited state to reduce (or oxidize)
thymine or cytosine derivatives. Therefore, selective oxidative
electron-transfer processes involving DNA purine bases can be
triggered upon irradiation of these new iridium(III) complexes.
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthetic details and spectroscopic properties of esterified
[IrN6]3+, luminescence experiments in the presence of dGMP
for esterified [IrN6]3+, luminescence experiments in the
1
presence of dAMP at different pH values for [IrN6]3+, H
NMR spectra and HRMS data for all complexes, disappearance
of luminescence of [IrN4C2]+ upon the addition of HNO3, and
cyclic voltammograms. This material is available free of charge
via the Internet at http://pubs.acs.org.
(15) Wells, R. D.; Larso, J. E.; Grant, R. C.; Shortle, B. E.; Cantor, C.
AUTHOR INFORMATION
■
R. J. Mol. Biol. 1970, 54, 465−497.
Corresponding Author
*E-mail: Benjamin.Elias@uclouvain.be.
(16) Collin, J.-P.; Dixon, I. M.; Sauvage, J.-P.; Williams, G. A.;
Barigelletti, F.; Flamigni, L. J. Am. Chem. Soc. 1999, 121, 5009−5016.
(17) ΔGET
= Eo x(deoxynucleobase) − Ered*, where
Present Address
†A.K-D.: Service de Chimie Organique et Photochimie,
Eox(deoxynucleobase) is the oxidation potential of a selected
deoxynucleobase and Ered* is the reduction potential of [IrN6]3+
and [IrN4C2]+ in their excited states. Therefore, we have obtained
ΔGET = 1.10 − 1.38 = −0.28 eV and ΔGET = 1.10 − 1.12 = −0.02 eV
respectively for [IrN6]3+ and [IrN4C2]+.
́
Universite libre de Bruxelles, CP160/08, 50 Av F. D. Roosevelt,
B-1050 Brussels, Belgium.
Author Contributions
(18) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.;
Von Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85−277.
(19) Wong, A.; Ida, R.; Spindler, L.; Wu, G. J. Am. Chem. Soc. 2005,
127, 6990−6998.
All authors have given approval to the final version of the
manuscript.
Notes
The authors declare no competing financial interest.
(20) Balzani, V.; Scandola, F. Supramolecular Photochemistry; Ellis
Horwood: Chichester, U.K., 1991.
(21) The low solubility of [IrN4C2]+ prevents its dissolution in a
pure aqueous Tris buffer. Therefore, 10% of acetonitrile has been
added to the buffer in order to have complete dissolution of the
complex.
ACKNOWLEDGMENTS
■
The authors gratefully acknowledge the Fonds National pour la
Recherche Scientifique (F.R.S.-F.N.R.S.), the Fonds pour la
Formation a
(F.R.I.A.), the Reg
Louvain, and the Prix Pierre et Colette Bauchau for financial
support. We are also grateful to L. Marcelis and L. Troian-
̀
la Recherche dans l’Industrie et dans l’Agriculture
(22) Ortmans, I.; Elias, B.; Kelly, J. M.; Moucheron, C.; Kirsch-De
́
ion Wallonne, the Universite catholique de
́
Mesmaeker, A. Dalton Trans. 2004, 4, 668−676.
(23) Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletti,
F. Top. Curr. Chem. 2007, 281, 143−203.
́
Gautier for their scientific help.
(24) Polson, M.; Fracasso, S.; Bertolasi, V.; Ravaglia, M.; Scandola, F.
Inorg. Chem. 2004, 43, 1950−1956.
(25) Polson, M.; Ravaglia, M.; Fracasso, S.; Garavelli, M.; Scandola,
F. Inorg. Chem. 2005, 44, 1282−1289.
REFERENCES
■
(1) (a) Du, P.; Knowles, K.; Eisenberg, R. J. Am. Chem. Soc. 2008,
130, 12576−12577. (b) DiSalle, B. F.; Bernhard, S. J. Am. Chem. Soc.
2011, 133, 11819−11821. (c) Ozawa, H.; Sakai, K. Coord. Chem. Rev.
2007, 251, 2753−2766.
(26) Steenken, S.; Jovanovics, V. J. Am. Chem. Soc. 1997, 119, 617−
618.
(27) Rehm, D.; Weller, A. Ber. Bunsen Ges. 1969, 73, 834−839.
(28) Htun, T. J. Fluoresc. 2004, 14, 217−222.
(2) (a) Gasser, G.; Ott, I.; Metzler-Nolte, N. J. Med. Chem. 2011, 54,
3−25. (b) Le Gac, S.; Rickling, S.; Gerbaux, P.; Defrancq, E.;
Moucheron, C.; Kirsch-De Mesmaeker, A. Angew. Chem., Int. Ed. 2009,
48, 1122−1125. (c) Gianferrara, T.; Bratsos, I.; Alessio, E. Dalton
Trans. 2009, 37, 7588−7598.
(3) (a) Erkkila, K. E.; Odom, D. T.; Barton, J. K. Chem. Rev. 1999,
99, 2777−2796. (b) Elias, B.; Kirsch-De Mesmaeker, A. Coord. Chem.
Rev. 2006, 250, 1627−1641. (c) McKinley, A. W.; Lincoln, P.; Tuite,
E. M. Coord. Chem. Rev. 2011, 255, 2676−2692. (d) Moucheron, C.
New J. Chem. 2009, 33, 235−245. (e) Smith, J. A.; George, M. W.;
F
dx.doi.org/10.1021/ic402476b | Inorg. Chem. XXXX, XXX, XXX−XXX