Organometallic iridium(III) cyclopentadienyl anticancer complexes containing C,N-Chelating ligands

Article abstract of DOI:10.1021/om2005468

Organometallic Ir^III cyclopentadienyl complexes [(η⁵-Cp^x)Ir(C^{^}N)Cl] {Cp^x = Cp*, C^{^}N = 2-(p-tolyl)pyridine (1), 2-phenylquinoline (2), 2-(2,4-difluorophenyl)pyridine (3), Cp^x = te

Full text of DOI:10.1021/om2005468

ARTICLE
pubs.acs.org/Organometallics
Organometallic Iridium(III) Cyclopentadienyl Anticancer Complexes
Containing C,N-Chelating Ligands
Zhe Liu, Abraha Habtemariam, Ana M. Pizarro, Guy J. Clarkson, and Peter J. Sadler*
Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, U.K.
S Supporting Information
b
ABSTRACT: Organometallic IrIII cyclopentadienyl complexes [(η5-Cpx)Ir-
(C∧N)Cl] {Cpx = Cp*, C∧N = 2-(p-tolyl)pyridine (1), 2-phenylquinoline
(2), 2-(2,4-difluorophenyl)pyridine (3), Cpx = tetramethyl(phenyl)-
cyclopentadienyl (Cpxph), C∧N = 2-phenylpyridine (4), and Cpx = tetra-
methyl(biphenyl)cyclopentadienyl (Cpxbiph), C∧N = 2-phenylpyridine
(5)} have been synthesized and characterized. The X-ray crystal structures
of 2 and 5 have been determined and show typical “piano-stool” geometry.
All the complexes hydrolyzed rapidly in aqueous solution (<5 min) even at
278 K. The pKa values of the aqua adducts 1Aꢀ5A are in the range
8.31ꢀ8.87 and follow the order 1A > 2A > 4A > 5A ≈ 3A. Hydroxo-bridged
dimers {[(η5-Cpx)Ir]2(μ-OD)3}+ (Cpx = Cp*, 6; Cpxph, 7; Cpxbiph, 8) are readily formed during pH titrations at ca. pH 8.7.
Complexes 1 and 3ꢀ5 bind strongly to 9-ethylguanine (9-EtG), moderately strongly to 9-methyladenine (9-MeA), and hence
preferentially to 9-EtG when in competition with 9-MeA. The extent of guanine and adenine binding to complex 2 was
significantly lower for both purines due to steric hindrance from the chelating ligand. All complexes showed potent cytotoxicity,
with IC50 values ranging from 6.5 to 0.7 μM toward A2780 human ovarian cancer cells. Potency toward these cancer cells
increased with additional phenyl substitution on Cp*: Cpxbiph > Cpxph > Cp*. Cpxbiph with complex 5 exhibited submicromolar
activity (2ꢁ as active as cisplatin). These data demonstrate how the aqueous chemistry, nucleobase binding, and anticancer
activity of C,N-bound IrIII cyclopentadienyl complexes can be controlled and fine-tuned by the modification of the chelating and
cyclopentadienyl ligands.
’ INTRODUCTION
charged C,N-bound 2-phenylpyridine (phpy) ligand in a Cp* IrIII
chlorido complex can switch on biological activity.3b This finding
has led us to make detailed investigations of Cp* IrIII complexes
with C,N-bound ligands and to study the influence of phenyl- or
biphenyl-substituted Cp* on their chemical and biological
properties.
We report studies of the aqueous chemistry (hydrolysis,
acidity of the resultant aqua adducts), nucleobase binding, and
cancer cell toxicity of IrIII complexes containing C,N-chelating
ligands based on 2-phenylpyridine with electron-donating or
electron-withdrawing substituents and with Cp* and substituted
Cp* ligands, Chart 1. The results suggest that this new class of
organometallic Ir(III) complexes is well suited for development
as anticancer agents.
Organometallic complexes offer enormous scope for the
design of anticancer candidates due to their versatile structures,
potential redox features, and wide range of ligand substitution
rates.1 For examples, titanocenyl, ferrocenyl, and RuII arene
anticancer complexes are attracting current attention.2 How-
ever, there are only a limited number of reported studies on the
anticancer activity of organometallic iridium complexes.3 In
general iridium(III) complexes are usually thought to be
relatively inert.4
The negatively charged pentamethylcyclopentadienyl ligand
(Cp*) is often an excellent stabilizing ligand for organometallic
iridium(III) complexes. However, a number of Cp* IrIII com-
plexes of the type [(η5-C5Me5)Ir(XY)Cl]0/+, where XY = N,N-
bound 1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane(PTA),3i
ethylenediamine (en), 2,20-bipyridine (bpy), 1,10-phenanthro-
line (phen), or N,O-bound picolinate (pico),3a have been
reported to be inactive toward A2780 human ovarian cancer
cells. However, the electronic and steric properties of the ligands
can have a major effect on the chemical and biological activities of
transition metal complexes.5 Recently we have demonstrated
that introduction of phenyl or biphenyl substituents on Cp* can
improve cancer cell cytotoxicity of N,N-chelated IrIII complexes
significantly.3a In addition, we have found that replacement of
the neutral N,N-bound chelating ligand bpy by the negatively
’ EXPERIMENTAL SECTION
Materials. 2-Phenylpyridine, 2-(2,4-difluorophenyl)pyridine, 2-(p-
tolyl)pyridine, 2-phenylquinoline, 9-ethylguanine, and 9-methyladenine
were purchased from Sigma-Aldrich. Methanol was distilled over
magnesium/iodine prior to use. Dimers [(η5-C5Me5)IrCl2]2,6 [(η5-
C5Me4C6H5)IrCl2]2,3a and [(η5-C5Me4C6H4C6H5)IrCl2]23a were pre-
pared according to reported methods.
Received: June 24, 2011
Published: August 05, 2011
r
2011 American Chemical Society
4702
dx.doi.org/10.1021/om2005468 Organometallics 2011, 30, 4702–4710
|

Organometallics
ARTICLE
Chart 1. Iridium Cyclopentadienyl Complexes Studied in
This Work
130.87, 130.33, 127.97, 126.34, 125.30, 122.11, 116.56, 89.11, 77.35,
76.71, 9.26. Anal. Calcd for C25H25ClNIr (567.13): C, 52.94; H, 4.44; N,
2.47. Found: C, 53.06; H, 4.41; N, 2.42. MS: m/z 531 [M ꢀ Cl]+.
Crystals suitable for X-ray diffraction were obtained by slow evaporation
of a methanol/diethyl ether solution at ambient temperature.
[(η5-C5Me5)Ir(dfphpy)Cl] (3). The synthesis was performed as for 1
using [(η5-C5Me5)IrCl2]2 (48 mg, 0.06 mmol), 2-(2,4-difluorophe-
nyl)pyridine (23 mg, 0.12 mmol), and sodium acetate (20 mg, 0.24
mmol). Yield: 46 mg (70%). 1H NMR (CDCl3): δ 8.71 (d, 1H, J = 6.0
Hz), 8.19 (d, 1H, J = 8.8 Hz), 7.69 (t, 1H, J = 8.0 Hz), 7.31 (d, 1H, J = 8.8
Hz), 7.10 (t, 1H, J = 6.3 Hz), 6.49 (t, 1H, J = 9.5 Hz), 1.67 (s, 15H). 13C
NMR (CDCl3): δ 151.54, 143.53, 137.52, 122.7, 117.28, 98.20. 89.02,
77.35, 8.78. Anal. Calcd for C21H21ClF2NIr (553.10): C, 45.60; H, 3.83;
N, 2.53. Found: C, 45.76; H, 3.71; N, 2.46. MS: m/z 517 [M ꢀ Cl]+.
[(η5-C5Me4C6H5)Ir(phpy)Cl] (4). A solution of [(η5-C5Me4C6H5)-
IrCl2]2 (46 mg, 0.05 mmol), 2-phenylpyridine (15 mg, 0.10 mmol), and
sodium acetate (16 mg, 0.20 mmol) in CH2Cl2 (15 mL) was heated
under reflux in an N2 atmosphere for 24 h. The solution was filtered
through Celite. The filtrate was evaporated to dryness on a rotary
evaporator and washed with diethyl ether. The product was recrystal-
lized from CHCl3/hexane. Yield: 37 mg (57%). 1H NMR (MeOD-d4):
δ 8.60 (d, 1H, J = 5.3 Hz), 8.04 (d, 1H, J = 8.3 Hz), 7.84 (m, 2H), 7.65 (d,
1H, J = 7.8 Hz), 7.38 (m, 3H), 7.33 (m, 2H), 7.16 (t, 1H, J = 6.1 Hz),
7.13 (t, 1H, J = 7.2 Hz), 7.09 (t, 1H, J = 7.3 Hz), 1.85 (s, 3H), 1.74 (s,
3H),1.72 (s, 3H), 1.56 (s, 3H). 13C NMR (CDCl3): δ 151.53, 137.13,
135.65, 131.14, 128.77, 127.33, 123.90, 122.20, 118.96, 77.35, 9.66. Anal.
Calcd for C26H25ClNIr (579.16): C, 53.92; H, 4.35; N, 2.42. Found: C,
53.77; H, 4.31; N, 2.41. MS: m/z 543 [M ꢀ Cl]+.
[(η5-C5Me4C6H4C6H5)Ir(phpy)Cl] (5). The synthesis was performed
as for 4 using [(η5-C5Me4C6H4C6H5)IrCl2]2 (53 mg, 0.05 mmol),
2-phenylpyridine (15 mg, 0.10 mmol), and sodium acetate (16 mg, 0.20
mmol). Yield: 37 mg (57%). 1H NMR (CDCl3): δ 8.51 (d, 1H, J = 5.3
Hz), 7.81 (d, 1H, J = 7.3 Hz), 7.72 (m, 2H), 7.64 (m, 5H), 7.51 (m, 4H),
7.37 (d, 1H, J = 7.6 Hz), 7.16 (t, 1H, J = 7.3 Hz), 7.05 (t, 1H, J = 6.0 Hz),
6.94 (t, 1H, J = 7.3 Hz), 1.92 (s, 3H), 1.82 (s, 3H),1.79 (s, 3H), 1.67 (s,
3H). 13C NMR (CDCl3): δ 151.45, 139.90, 136.98, 135.57, 131.05,
129.01, 127.22, 123.93, 122.51, 118.94, 77.34, 9.88. Anal. Calcd for
C32H29ClNIr (655.25): C, 58.66; H, 4.46; N, 2.14. Found: C, 58.46; H,
4.35; N, 2.18. MS: m/z 619 [M ꢀ Cl]+. Crystals suitable for X-ray
diffraction were obtained by slow evaporation of a methanol/diethyl
ether solution at ambient temperature.
’ METHODS AND INSTRUMENTATION
Syntheses. [(η5-C5Me5)Ir(tpy)Cl] (1). A solution of [(η5-C5Me5)-
IrCl2]2 (48 mg, 0.06 mmol), 2-(p-tolyl)pyridine (20 mg, 0.12 mmol), and
sodium acetate (20 mg, 0.24 mmol) in CH2Cl2 (15 mL) was stirred for
2 h at ambient temperature. The solution was filtered through Celite. The
filtrate was evaporated to dryness on a rotary evaporator and washed with
diethyl ether. The product was recrystallized from CHCl3/hexane. Yield:
45 mg (70%). 1H NMR (CDCl3): δ 8.65 (d, 1H, J = 5.7 Hz), 7.75 (d, 1H,
J = 8.3 Hz), 7.62 (m, 2H), 7.57 (d, 1H, J = 8.0 Hz), 7.03 (t, 1H, J = 6.3
Hz), 6.86 (d, 1H, J = 7.8 Hz), 1.68 (s, 15H). 13C NMR (CDCl3): δ
151.29, 140.79, 136.74, 128.35, 123.27, 121.65, 118.43, 88.41, 77.36, 8.83.
Anal. Calcd for C22H25ClNIr (531.14): C, 49.75; H, 4.74; N, 2.64.
Found: C, 49.66; H, 4.65; N, 2.68. MS: m/z 496 [M ꢀ Cl]+.
X-ray Crystallography. All diffraction data were obtained on an
Oxford Diffraction Gemini four-circle system with a Ruby CCD area
detector using Mo KR radiation. Absorption corrections were applied
using ABSPACK.7 The crystals were mounted in oil and held at 100(2)
K with the Oxford Cryosystem Cobra. The structures were solved by
direct methods using SHELXS (TREF)8 with additional light atoms
found by Fourier methods. Complexes 2 and 5 were refined against F2
using SHELXL,9 and hydrogen atoms were added at calculated positions
and refined riding on their parent atoms.
X-ray crystallographic data for complexes 2 and 5 have been
deposited in the Cambridge Crystallographic Data Centre under the
accession numbers CCDC 829525 and 829524, respectively.
[(η5-C5Me5)Ir(phq)Cl] (2). The synthesis was performed as for 1 using
[(η5-C5Me5)IrCl2]2 (48 mg, 0.06 mmol), 2-phenylquinoline (25 mg,
0.12 mmol), and sodium acetate (20 mg, 0.24 mmol). Yield: 43 mg
(75%). 1H NMR (CDCl3): δ 8.71 (d, 1H, J = 8.8 Hz), 8.02 (d, 1H, J =
8.7 Hz), 7.93 (d, 2H, J = 8.8 Hz), 7.77 (m, 2H), 7.69 (t, 1H, J = 8.1 Hz),
7.53 (t, 1H, J = 6.7 Hz), 7.24 (t, 1H, J = 7.8 Hz), 7.07 (t, 1H, J = 7.7 Hz),
1.57 (s, 15H). 13C NMR (CDCl3): δ 156.22, 137.76, 136.49, 131.36,
1
NMR Spectroscopy. H NMR spectra were acquired in 5 mm
NMR tubes at 298 K (unless stated otherwise) on either Bruker DPX
400 (1H = 400.03 MHz) or AVA 600 (1H = 600.13 MHz) spectrom-
1
eters. H NMR chemical shifts were internally referenced to CHCl3
(7.26 ppm) for chloroform-d1, CHD2OD (3.33 ppm) for methanol-d4,
or 1,4-dioxane (3.75 ppm) for aqueous solutions. All data processing was
carried out using XWIN-NMR version 3.6 (Bruker UK Ltd.).
4703
dx.doi.org/10.1021/om2005468 |Organometallics 2011, 30, 4702–4710

Organometallics
ARTICLE
Figure 1. X-ray crystal structures and atom-numbering schemes for complexes [(η5-C5Me5)Ir(phq)Cl] (2) and [(η5-C5Me4C6H4C6H5)Ir(2-
phpy)Cl] (5).
Mass Spectrometry. Electrospray ionization mass spectra (ESI-
MS) were obtained on a Bruker Esquire 2000 ion trap spectrometer.
Samples were prepared in 50% CH3CN and 50% H2O (v/v). The mass
spectra were recorded with a scan range of m/z 50ꢀ1000 for positive ions.
Elemental Analysis. CHN elemental analyses were carried out on
a CE-440 elemental analyzer by Exeter Analytical (UK) Ltd.
pH Measurement. pH* values (pH meter reading without correc-
tion for the effect of deuterium on the glass electrode) of NMR samples
in D2O were measured at ca. 298 K directly in the NMR tube, before and
after recording the NMR spectra, using a Corning 240 pH meter
equipped with a micro combination electrode calibrated with Aldrich
buffer solutions of pH 4, 7, and 10.
Determination of pKa Values. To generate the aqua complexes,
chlorido complexes were dissolved in D2O and 0.98 molar equiv of
AgNO3 was added. The solution was stirred for 24 h at 298 K, and AgCl
was removed by filtration. For determinations of pKa* values (pKa values
for solutions in D2O), the pH* values of solutions of the aqua complexes
in this study were varied from ca. pH* 2 to 10 by the addition of dilute
NaOD and DClO4, and 1H NMR spectra were recorded. The chemical
shifts of the chelating ligand protons and/or of the methyl protons of
Cpx were plotted against pH*. The pH* titration curves were fitted to the
HendersonꢀHasselbalch equation, with the assumption that the ob-
served chemical shifts are weighted averages according to the popula-
tions of the protonated and deprotonated species. These pKa* values can
be converted to pKa values by use of the equation pKa = 0.929pKa* +
0.42, as suggested by Krezel and Bal10 for comparison with related values
in the literature.
Interactions with Nucleobases. The reaction of complexes 1ꢀ5
(ca. 1 mM) with nucleobases typically involved addition of a solution
containing 1 molar equiv of nucleobase in D2O to an equilibrium
solution of complexes 1ꢀ5 in 20% MeOD-d4/80% D2O (v/v). 1H NMR
spectra of these solutions were recorded at 310 K after various time
intervals.
Cytotoxicity. The A2780 human ovarian cancer cell line was
obtained from the ECACC (European Collection of Animal Cell
Cultures, Salisbury, UK). The cells were maintained in RPMI 1640
media (supplemented with 10% fetal calf serum, 1% L-glutamine, and 1%
penicillin/streptomycin). All cells were grown at 310 K in an humidified
atmosphere containing 5% CO2. Stock solutions of the IrIII complexes
were first prepared in DMSO to assist dissolution (maximum final
DMSO concentration 2%) and then diluted into 0.9% saline and
medium (1:1). After plating 5000 A2780 cells per well on day 1, IrIII
complexes were added to the cancer cells on day 3 at concentrations
ranging from 0.05 to 50 μM. Cells were exposed to the complexes for 24
h, washed with PBS, supplied with fresh medium, and allowed to grow
for three doubling times (72 h). Protein content (proportional to cell
survival) was then measured using the sulforhodamine B assay.11 The
standard errors are based on two independent experiments carried out in
triplicate.
’ RESULTS
Five IrIII half-sandwich complexes of the type [(η5-Cpx)Ir-
(C∧N)Cl], where Cpx is pentamethylcyclopentadienyl Cp*, or
its phenyl (Cpxph) or biphenyl (Cpxbiph) derivatives, and the C,
N-chelating ligands are 2-(p-tolyl)pyridine (tpy, 1), 2-phenyl-
quinoline (phq, 2), 2-(2,4-difluorophenyl)pyridine (dfphpy, 3),
or 2-phenylpyridine (phpy, 4 and 5), were synthesized in good
yields by reaction of the chelating ligand with the appropriate
dimer [(η5-Cpx)IrCl2]2 in CH2Cl2. All the synthesized com-
1
plexes were fully characterized by H NMR and 13C NMR
spectroscopy, ESI-MS, and CHN elemental analysis. All the
complexes in this study are chiral, but no attempt was made to
resolve them, and racemates are used in the following studies.
X-ray Crystal Structures. The X-ray crystal structures of [(η5-
C5Me5)Ir(phq)Cl] (2) and [(η5-C5Me4C6H4C6H5)Ir(phpy)Cl]
(5) were determined. The structures and atom-numbering
schemes are shown in Figure 1. Crystallographic data are
shown in Table 1, and selected bond lengths and angles are
listed in Table 2. Complexes 2 and 5 adopt the expected half-
sandwich pseudo-octahedral “three-legged piano-stool” geometry
with the iridium bound to a η5-cyclopentadienyl ligand (Ir to
ring centroid distances of 1.829 and 1.825 Å, respectively). The
IrꢀCl bond distances are 2.3989(16) and 2.3886(8) Å for 2
and 5, respectively. The IrꢀC(chelating ligand) bond lengths
in 2 and 5 [2.045(6) and 2.057(3) Å, respectively] are sign-
ificantly shorter than the IrꢀN bond lengths [2.128(5) and
2.080(3) Å, respectively]. For complex 5, the twist angle
between the bound cyclopentadienyl ring and the central
phenyl ring is 48.93°, while that between the bound ring and
the terminal phenyl ring is 28.50°. The phenyl rings are twisted
by 21.55°. No intermolecular π-ring stacking in the unit cell is
observed in the two crystal structures.
Hydrolysis Studies. The hydrolysis of complexes 1ꢀ5 in 20%
1
MeOD-d4/80% D2O (v/v) was monitored by H NMR spec-
troscopy at different temperatures. The presence of methanol
ensured the solubility of the complexes. All these IrIII complexes
undergo rapid hydrolysis. Equilibrium was reached by the time
the first 1H NMR spectrum was acquired (∼5 min) even at 278
K. At equilibrium 20ꢀ50% of complexes 1ꢀ5 was in the
hydrolyzed form, based on 1H NMR peak integrals.
To confirm the hydrolysis of the complexes, NaCl was added
to equilibrium solutions containing the chlorido complexes and
4704
dx.doi.org/10.1021/om2005468 |Organometallics 2011, 30, 4702–4710

Organometallics
ARTICLE
Table 1. Crystallographic Data for [(η5-C5Me5)Ir(phq)Cl]
(2) and [(η5-C5Me4C6H4C6H5)Ir(2-phpy)Cl] (5)
Table 2. Selected Bond Lengths (Å) and Angles (deg) for
[(η5-C5Me5)Ir(phq)Cl] (2) and [(η5-C5Me4C6H4C6H5)-
Ir(2-phpy)Cl] (5)
2
5
bond(s)
2
5
formula
MW
C25H25ClIrN
567.11
C32H29ClIrN
655.21
IrꢀC(Cpx)
2.139(6)
2.157(6)
2.167(5)
2.243(6)
2.304(6)
1.829
2.151(3)
2.163(3)
2.183(3)
2.240(3)
2.243(3)
1.825
cryst color
cryst size (mm)
λ (Å)
orange block
0.40 ꢁ 0.40 ꢁ 0.10
0.71073
orange block
0.22 ꢁ 0.18 ꢁ 0.12
0.71073
temp (K)
cryst syst
space group
a (Å)
100
100
IrꢀC(centroid)
IrꢀC
IrꢀN
IrꢀCl
CꢀIrꢀN
CꢀIrꢀCl
NꢀIrꢀCl
orthorhombic
P2(1)2(1)2(1)
8.05090(13)
15.9169(3)
16.0586(3)
90
monoclinic
P2(1)/n
10.0094(4)
22.9497(6)
11.2128(4)
90
2.045(6)
2.128(5)
2.3989(16)
77.4(2)
2.057(3)
2.080(3)
2.3886(8)
78.27(13)
88.20(9)
86.34(8)
b (Å)
c (Å)
89.66(17)
87.23(13)
R (deg)
β (deg)
γ (deg)
vol (Å3)
Z
90
103.473(3)
90
90
2057.84(6)
4
2504.84(14)
4
of the coordinated water were determined. This gave pKa values
between 8.31 and 8.87 (Table 3), with the Cpxbiph aqua complex
5A and fluoro-substituted phenylpyridine Cp* complex 3A
having the lowest pKa values (8.31 and 8.32, respectively) and
the methyl-substituted phenylpyridine complex 1A having the
highest (8.87).
density(calc)
1.830
1.737
(Mg mꢀ3
)
3
abs coeff (mmꢀ1
)
6.628
5.459
F(000)
1104
1288
During the pH titrations for aqua complexes 1Aꢀ5A, the
appearance of a new set of peaks was detected with increasing
pH* (>8.7). The new peaks are attributable to the free C,N-
chelating ligands and to the hydroxo-bridged dimers {[(η5-
Cpx)Ir]2(μ-OD)3}+ (Cpx = Cp*, 6; Cpxph, 7; Cpxbiph, 8; see
Chart 1). The 1H NMR peaks for hydroxo-bridged dimers 6ꢀ8
increased in intensity with increase in pH*. For complex [(η5-
C5Me5)Ir(dfphpy)(D2O)]+ (3A), the amount of dimer 6 in-
creased from 23% at pH* 9.0 to 50% at pH* 9.6, Figure 3. ESI-MS
studies on the diluted sample (0.2 mM) gave a major peak at m/z
709.2, consistent with the presence of {[(η5-C5Me5)Ir]2(μ-
OD)3}+ (calcd m/z 710.2).
θ range (deg)
index ranges
3.11 to 30.45
3.10 to 29.33
ꢀ11 e h e 6, ꢀ20 e ꢀ13 e h e 12, ꢀ31 e k
k e 22, ꢀ22 e l e 11 e 31, ꢀ14 e l e 15
reflns collected
indep reflns
10 123
23 877
5496 [R(int) = 0.0266] 6211 [R(int) = 0.0542]
data/restraints/params 5496/0/253
6211/0/320
R1 = 0.0293,
wR2 = 0.0680
R1 = 0.0377,
wR2 = 0.0699
0.999
final R indices
[I > 2σ(I)]
R1 = 0.0368,
wR2 = 0.0856
R1 = 0.0407,
wR2 = 0.0886
1.053
R indices (all data)
GOF
Interactions with Nucleobases. Since DNA is a potential
target for transition metal anticancer drugs,13 nucleobase binding
reactions of complexes [(η5-C5Me5)Ir(tpy)Cl] (1), [(η5-
C5Me5)Ir(phq)Cl] (2), [(η5-C5Me5)Ir(dfphpy)Cl] (3), [(η5-
C5Me4C6H5)Ir(phpy)Cl] (4), and [(η5-C5Me4C6H4C6H5)Ir-
(phpy)Cl] (5) with 9-ethylguanine (9-EtG) and 9-methylade-
nine (9-MeA) were investigated. Solutions of 1ꢀ5 (ca. 1 mM,
containing an equilibrium mixture of 1ꢀ5 and their respective
aqua adducts 1Aꢀ5A) and 1 molar equiv of 9-EtG or 9-MeA in
largest diff peak
and hole (e Åꢀ3
6.037 and ꢀ1.728
1.754 and ꢀ2.149
)
their aqua adducts to final concentrations of 4, 23, and 104 mM
NaCl, mimicking the chloride concentrations in cell nucleus, cell
cytoplasm, and blood plasma, respectively.12 1H NMR spectra
were then recorded within 10 min of the Clꢀ additions at 298 K.
Upon addition of NaCl, the 1H NMR peaks corresponding to the
chlorido complexes increased in intensity, while peaks for the
aqua adducts decreased, Figures 2 and S1. These data confirm the
formation of the aqua adducts and the reversibility of the process.
1
20% MeOD-d4/80% D2O (v/v) were prepared, and H NMR
spectra were recorded at different time intervals at 310 K. The
percentages of nucleobase adducts formed by the complexes after
1
1
On the basis of H NMR data, reaction of aqua complexes
24 h reaction, based on H NMR peak integrals, are shown in
1Aꢀ5A was almost complete in 104 mM [Clꢀ] or in 23 mM
[Clꢀ], and ca. 5% of aqua complex was observed at 4 mM [Clꢀ]
after 10 min with no further change after 24 h.
Table S1 and Figure 4.
In the 1H NMR spectrum of a solution containing 4 (0.8 mM)
and 1 molar equiv of 9-ethylguanine (20% MeOD-d4/80% D2O,
pH* 7.4, 310 K), one set of new peaks assignable to the 9-EtG
adduct 4G appeared, showing that 100% of 4 had reacted after 10
min (Figure S3). A significant change in chemical shift from 8.62
ppm for the chlorido complex 4 to 9.28 ppm for 9-EtG adduct 4G
for the CHdN (phpy ligand) proton was observed. A new 9-EtG
H8 peak appeared at 7.44 ppm (singlet), shifted by 0.34 ppm to
high field relative to that of free 9-EtG. After 24 h, no further
change was observed. The ESI-MS of an equilibrium solution
pKa* Determination. Changes in the 1H NMR chemical shifts
of the methyl protons of Cp* or protons of the coordinated
chelating ligands in aqua complexes 1Aꢀ5A and [(η5-C5Me5)Ir-
(phpy)(D2O)]+ (for comparison) were followed with change in
pH* over a range of 2ꢀ10 (Figure S2). 1H NMR peaks assigned
to aqua complexes gradually shifted to high field with increase in
pH*. The resulting pH titration curves were fitted to the
HendersonꢀHasselbalch equation, from which the pKa* values
4705
dx.doi.org/10.1021/om2005468 |Organometallics 2011, 30, 4702–4710

Organometallics
ARTICLE
Figure 2. Confirmation of hydrolysis of IrIII complex [(η5-C5Me5)Ir-
Figure 4. Bar chart showing the extent of binding of complexes 1ꢀ5
(ca. 1 mM in 20% MeOD-d4/80% D2O) to the nucleobases 9-EtG and
9-MeA at equilibrium (24 h), based on 1H NMR peak integrals.
1
(tpy)Cl] (1). (A) H NMR spectrum of an equilibrium solution of 1
1
(1 mM) in 20% MeOD-d4/80% D2O (v/v) at 298 K. (B) H NMR
spectrum recorded 10 min after addition of NaCl (final concentration,
4 mM) to the equilibrium solution of 1. Complex 1A corresponds to the
aqua complex [(η5-C5Me5)Ir(tpy)(D2O)]+. The peaks for the chlorido
complex 1 increased in intensity, while peaks for the aqua complex 1A
decreased (ca. 5% 1A remaining) upon addition of NaCl.
[(η5-C5Me5)Ir(tpy)Cl] showed an exceptionally high affinity for
9-MeA, with 90% adduct formation after 24 h. Complex 2,
containing 2-phenylquinoline, formed almost no 9-MeA adduct
(<5%). Except for 2, two adenine nucleobase adducts were
formed in the reactions between complexes 1 and 3ꢀ5 with
9-MeA, most likely through iridium binding to N1 or N7 of
adenine in a ratio typically of 1:5.
Table 3. pKa* and pKa Valuesa for the Deprotonation of the
Coordinated D2O in Complexes 1Aꢀ5A and [(η5-C5Me5)-
Ir(phpy)(D2O)]+
Competition reactions between equimolar amounts of 9-EtG
and 9-MeA and complexes 1 and 3ꢀ5 (ca. 1 mM) in 20%
MeOD-d4/80% D2O (v/v, pH* ca. 7.4) at 310 K were investi-
gated. In the competitive experiment, 9-EtG adduct 3G, 4G, or
5G was observed as the only product for complexes 3ꢀ5 and as
80% 9-EtG adduct for complex 1 (20% 9-MeA adduct), Table S1.
aqua complex
pKa*
pKa
[(η5-C5Me5)Ir(tpy)(D2O)]+ (1A)
[(η5-C5Me5)Ir(phq)(D2O)]+ (2A)
[(η5-C5Me5)Ir(dfphpy)(D2O)]+ (3A)
[(η5-C5Me4C6H5)Ir(phpy)(D2O)]+ (4A)
[(η5-C5Me4C6H4C6H5)Ir(phpy)(D2O)]+(5A)
[(η5-C5Me5)Ir(phpy)(D2O)]+
9.10
8.86
8.51
8.64
8.50
8.97
8.87
8.65
8.32
8.45
8.31
8.75
1
The H NMR spectra for the competition reaction for 3 are
shown in Figure 5.
Reactions of complexes [(η5-C5Me4C6H5)Ir(phpy)Cl] (4)
and [(η5-C5Me4C6H4C6H5)Ir(phpy)Cl] (5) with 9-EtG were
also investigated in the presence of 4 mM NaCl, Figure 6. A
solution of 4 (ca. 0.8 mM) in 20% MeOD-d4/80% D2O (v/v)
containing an equilibrium mixture of 4 and aqua adduct 4A was
a pKa values calculated from pKa* according to Krezel and Bal.10
1
prepared, and its H NMR spectrum was recorded, Figure 6A.
Then NaCl was added to give a 4 mM [Clꢀ] solution. The aqua
adduct 4A was suppressed to ca. 5%, Figure 6B. Addition of ca. 1
molar equiv of 9-EtG resulted in complete formation of 9-EtG
adduct after 24 h at 310 K, Figure 6C. Complex 5 also formed a
guanine adduct quantitatively in the presence of 4 mM [Clꢀ].
Cytotoxicity. The cytotoxicity of complexes 1ꢀ5 toward
A2780 human ovarian cancer cells was investigated, Table 4.
The IC50 value (concentration at which 50% of the cell growth is
inhibited) for Cp* IrIII complexes 1ꢀ3 is comparable with that of
cisplatin. Complexes 4 and 5, containing Cpxph or Cpxbiph, were
even more potent, especially complex 5, with an IC50 value of 0.7
μM (ca. twice as active as cisplatin). Overall, the cytotoxic
potency increases with phenyl substitution on Cp*: Cpxbiph
Cpxph > Cp*, Table 4 and Figure S4.
>
Figure 3. Methyl region of 1H NMR spectra from the pH titration of
the aqua complex [(η5-C5Me5)Ir(dfphpy)(D2O)]+ (3A), showing an
increase in intensity of the peak for the hydroxo-bridged dimer {[(η5-
C5Me5)Ir]2(μ-OD)3}+ (6) with increase in pH*.
’ DISCUSSION
X-ray Crystal Structures. A search of the Cambridge Crystal-
lographic Database showed that the crystal structure of
complex 5 is only the second example to be reported of a
metal complex containing the Cpxbiph ligand. The only other
example is the complex [(η5-C5Me4C6H4C6H5)Ir(bpy)Cl]-
PF6 (where bpy = bipyridine).3a The two complexes are
structurally very similar. In complex 5, the chelating ligand is
closer to the IrIII center than in [(η5-C5Me4C6H4C6H5)Ir-
(bpy)Cl]PF6 since the IrꢀC bond length [2.057(3) Å] in
contained a major peak at m/z 723.2, confirming the formation of
the 9-EtG adduct 4G, [(η5-C5Me4C6H5)Ir(phpy)(9-EtG)]+
(calcd m/z 722.9). Similarly, complexes 1, 3, and 5 also formed
9-EtG adducts to the extent of 100% after 24 h. Only complex 2,
containing 2-phenylquinoline, showed less strong binding to
9-EtG (45%, Figure 4 and Table S1).
Complexes 3ꢀ5 formed moderately strong 9-MeA adducts
(35ꢀ63% at equilibrium after 24 h, Table S1). Only complex 1
4706
dx.doi.org/10.1021/om2005468 |Organometallics 2011, 30, 4702–4710

Organometallics
ARTICLE
Figure 5. Low-field region of the 1H NMR spectra for the competitive reaction between 9-EtG and 9-MeA and complex 3 [(η5-C5Me5)Ir(dfphpy)Cl].
(A) Equilibrium solution of 3 (1.0 mM) in 20% MeOD-d4/80% D2O (v/v, pH* 7.4) at 310 K, containing both the chlorido complex 3 and its aqua
adduct 3A. (B) 10 min after addition of equimolar amounts of 9-EtG and 9-MeA, showing the complete formation of 9-EtG adduct 3G.
1
Figure 6. Low-field region of the H NMR spectra for the reaction between 9-EtG and complex 4 [(η5-C5Me4C6H5)Ir(phpy)Cl] at 310 K. (A)
Equilibrium solution of 4 (0.8 mM) in 20% MeOD-d4/80% D2O (v/v, pH* 7.4), containing both the chlorido complex 4 and its aqua adduct 4A. (B) 10
min after addition of NaCl (final concentration, 4 mM) to the equilibrium solution. (C) The complete formation of 9-EtG adduct 4G after addition of 1
molar equiv of 9-EtG.
The bond lengths and bond angles in complexes [(η5-
Table 4. Inhibition of Growth of A2780 Human Ovarian
C5Me5)Ir(phq)Cl] (2) and [(η5-C5Me5)Ir(phpy)Cl]14 are si-
Cancer Cells by Complexes 1ꢀ5 and Comparison with
Cisplatin
milar, except that the IrꢀN bond for 2 [2.128(5) Å] is longer
than that of [(η5-C5Me5)Ir(phpy)Cl] [2.080(2) Å], implying
complex
IC50a (μM)
weaker σ donation from N to the iridium center due to the
electron-withdrawing phenyl ring in the quinoline moiety.
Hydrolysis and pKa of Aqua Adducts. Since MꢀOH2 (M =
metal) aqua complexes are often more reactive than the equivalent
chlorido complexes,3a,12,15 hydrolysis of the MꢀCl bonds can repre-
sent an activation step for transition metal anticancer complexes.16
There are only a few previous studies of the aquation (substitution
of Cl by H2O) of organometallic IrIII complexes.3a,4a,17 In the work
reported here all the [(η5-Cpx)Ir(C∧N)Cl] complexes 1ꢀ5 hydro-
lyzed too rapidly for determination of their hydrolysis rates by
NMR, even for the biphenyl-substituted Cpxbiph complex 5 [(η5-
C5Me4C6H4C6H5)Ir(2-phpy)Cl]. We have previously reported
half-lives for the hydrolysis of some Cpxph or Cpxbiph IrIII complexes
containing N,N-bound 2,20-bipyridine (bpy) or 1,10-phenanthro-
line (phen) chelating ligands of ca. 4 min at 310 K.3a The electron-
donor methyl groups on the Cp ring and the negatively charged C,
N-chelating ligands may together contribute to the fast hydrolysis
observed for the complexes reported here since the increased
effective charge on the Ir center may facilitate chloride loss. Fast
hydrolysis rates also have been reported for some hexamethylben-
zene RuII complexes,18 acetylacetonate RuII and OsII complexes,19
and picolinate IrIII complexes.3a These results illustrate that IrIII
[(η5-C5Me5)Ir(tpy)Cl] (1)
[(η5-C5Me5)Ir(phq)Cl] (2)
[(η5-C5Me5)Ir(dfphpy)Cl] (3)
[(η5-C5Me4C6H5)Ir(phpy)Cl] (4)
[(η5-C5Me4C6H4C6H5)Ir(phpy)Cl] (5)
[(η5-C5Me5)Ir(phpy)Cl] b
3.28 ( 0.14
2.55 ( 0.03
6.53 ( 0.50
2.14 ( 0.50
0.70 ( 0.04
10.78 ( 1.72
1.19 ( 0.12
cisplatin
a Drug-treatment period was 24 h. b Ref 3b.
complex 5 is significantly shorter than the IrꢀN bond length
[2.091(5) Å]3a in the latter complex. The short IrꢀC-
(phenylpyridine) distance causes a slight elongation of the
Irꢀcyclopentadienyl (centroid) bond with a distance of 1.825
Å in complex 5 (Table 2), compared to 1.787 Å in complex
[(η5-C5Me4C6H4C6H5)Ir(bpy)Cl]PF6. The IrꢀCl bond
lengths are similar in the two complexes, with distances of
2.3886(8) Å (Table 2) and 2.3840(14) Å,3a respectively. This
behavior is similar to that observed for the C,N-chelated
complex [(η5-C5Me5)Ir(phpy)Cl]14 when compared to the
N,N-chelated complex [(η5-C5Me5)Ir(bpy)Cl]Cl.3b
4707
dx.doi.org/10.1021/om2005468 |Organometallics 2011, 30, 4702–4710

Organometallics
ARTICLE
complexes are not always inert and that IrꢀCl bonds can
be labile.
In this study, the interactions of model nucloebases, 9-EtG and
9-MeA, with complexes 1ꢀ5 were investigated (Figure 4 and
Table S1). All the C,N-chelated IrIII complexes, except 2, showed
an exceptionally high nucleobase affinity, with 100% guanine
adduct formation for 9-EtG, which may contribute to their high
cytotoxicity. Complexes [(η5-C5Me4C6H5)Ir(phpy)Cl] (4) and
[(η5-C5Me4C6H4C6H5)Ir(phpy)Cl] (5) still bind strongly to
guanine in the presence of 4 mM [Clꢀ] (typically the chloride
concentration in the cell nucleus) (Figure 6), which suggests that
this class of iridium complexes might interact with DNA in the
cell nucleus. Compared with N,N-chelated IrIII complexes,3a
complexes containing a C,N-bound chelating ligand bind more
significantly to 9-EtG, which may be due to their inherent
advantage of having higher pKa values. Under similar pH
conditions, hydrolyzed C,N-complexes are more likely to be
present as the reactive aqua form compared to the N,N-chelated
complexes.
Complexes 1ꢀ5 bind more weakly to adenine compared to
guanine. The competition between 9-EtG and 9-MeA for the C,
N-bound IrIII complexes give rise to 9-EtG adducts as the only
product for 3ꢀ5 and as the major product for 1 (Figure 4 and
Table S1), confirming that binding to guanine is stronger than to
adenine. This may be due to the steric hindrance caused by the
NH2 group at the 6-position of the adenine ring. In addition,
guanine is usually considered to be a stronger electron donor
than adenine.21 The widely used anticancer drug cisplatin also
prefers guanine over adenine.22 Organometallic RuII, OsII, and
IrIII complexes containing N,O-chelating ligands or O,O-chelat-
ing ligands such as picolinate and acetylacetonate, which possess
oxygen as an H-bond acceptor for adenine C6NH2, also bind to
both guanine and adenine residues.3a,19a,20b,23
At chloride concentrations typical of blood plasma (104 mM)
and cell cytoplasm (23 mM), complexes 1ꢀ5 were found to be
almost all present in solution as the relatively unreactive chlorido
species, and only about 5% of complexes 1ꢀ5 were present as the
reactive aqua species at a chloride concentration of 4 mM close to
that of the cell nucleus. These data suggest that aquation is
suppressed almost totally at the saline concentration in blood.
However, complexes 1ꢀ5 might be activated by aquation in the
cell nucleus. Another possibility is that the complexes might react
by direct substitution of chloride by nucleobases (DNA).
The aqua complexes [(η5-C5Me5)Ir(tpy)(D2O)]+ (1A), [(η5-
C5Me5)Ir(phq)(D2O)]+ (2A), [(η5-C5Me5)Ir(dfphpy)(D2O)]+
(3A), [(η5-C5Me4C6H5)Ir(phpy)(D2O)]+ (4A), [(η5-C5Me4-
C6H4C6H5)Ir(phpy)(D2O)]+ (5A), and [(η5-C5Me5)Ir(phpy)-
(D2O)]+ have similar pKa values, ranging from 8.31 to 8.87
(Table 3). Although the substituents on the Cp* ring and the
2-phenylpyridine chelating ligand do not significantly affect the
acidity of the bound water, the observed trend caused by these
substituents is clear, following the order 1A > 2A > 4A > 5A ≈ 3A.
The presence of phenyl and biphenyl substituents on the Cp* ring
lower the pKa value by ca. 0.3 to 0.4 units, consistent with the
electron-withdrawing properties of these groups. Substitution of
cyclometalated 2-phenylpyridine by the fluorinated chelating
ligand 2-(2,4-difluorophenyl)pyridine leads to a decrease in pKa
by 0.4 unit. However, replacing cyclometalated 2-phenylpyridine
by 2-(p-tolyl)pyridine or by the more π-delocalized 2-phenylqui-
noline has little effect on the pKa value (ca. 0.1 unit).
However, the aqua complexes [(η5-C5Me5)Ir(phpy)(D2O)]+,
[(η5-C5Me4C6H5)Ir(phpy)(D2O)]+ (4A), and [(η5-C5Me4-
C6H4C6H5)Ir(phpy)(D2O)]+ (5A), containing the C,N-che-
lated 2-phenylpyridine ligand, have significantly higher pKa
values (average 1.9 units higher) than those for the structurally
similar IrIII Cpx analogues bearing the N,N-bound 2,20-bipyr-
idine (bpy) ligand.3a Therefore, the replacement of the neutral
bpy ligand by the anionic phpy ligand plays a significant role in
decreasing the acidity of the aqua complexes, consistent with
previous reports.3a,19a The high pKa values of 1Aꢀ5A thus
ensure that most of the hydrolyzed complexes would be
present in the active aqua form at physiological pH.
The pH titration also showed that additional species were
formed for all complexes studied here above pH 8.7 (Figure 3),
indicating the formation of the hydroxo-bridged dimers {[(η5-
Cpx)Ir]2(μ-OD)3}+ (6ꢀ8, Chart 1). However, this does not
occur when the chelating ligand is an N,N-chelated bipyridine
ligand.3a Clearly the IrꢀC bond shows less stability with respect
to the formation of dimers 6ꢀ8 than the IrꢀN bond. It seems
likely that the mechanism of formation of 6ꢀ8 involves initial
cleavage of the IrꢀC bond, followed by IrꢀN bond breakage.
Some OsII and RuII complexes containing N,O-bound or O,O-
bound ligands have been reported to form hydroxo-bridged
dimers readily during their hydrolysis, resulting in their inactivity
toward cancer cell lines.20 In this work, no hydroxo-bridged
dimers of 6ꢀ8 were observed during hydrolysis, and all the
complexes are active toward A2780 human ovarian cancer
cells. This indicates that the C,N-chelated IrIII complexes are
stable in aqueous solution and the dimers 6ꢀ8 formed only
appear at high pH and have no negative effect on their cytotoxicity at
physiological pH.
IrIII cyclopentadienyl complexes containing a neutral N,N-
chelating ligand (phen, bpy, or ethylenediamine) bind selectively
to 9-EtG, but not to adenine.3a The formation of adenine adducts
in this work may be due to the interaction between NH2 of
9-MeA and negatively charged carbons on the C,N-chelating
ligand.3b Thus complexes 1ꢀ3 containing different substituents
on the phenylpyridine bind to 9-MeA differently. The electron-
donating methyl group on the phenyl ring in complex [(η5-
C5Me5)Ir(tpy)Cl] (1) increases electron density and may facil-
itate the interaction with 9-MeA (>90%). In contrast, only 35%
of complex [(η5-C5Me5)Ir(dfphpy)Cl] (3) formed 9-MeA ad-
ducts, which may be due to the presence of the electron-with-
drawing fluoro group. Complex 2 [(η5-C5Me5)Ir(phq)Cl] has
the lowest affinity for both model nucleobases among the five
complexes, with 45% and <5% binding to 9-EtG and 9-MeA,
respectively. The weaker binding is most likely due to the steric
hindrance caused by the quinoline ligand.
Cytotoxicity. We have reported that some Cp* IrIII com-
plexes containing N,N- or N,O-chelating ligands are inactive
toward A2780 human ovarian cancer cells.3a However, Cp*
complexes [(η5-C5Me5)Ir(tpy)Cl] (1), [(η5-C5Me5)Ir-
(phq)Cl] (2), and [(η5-C5Me5)Ir(dfphpy)Cl] (3) studied here
showed promising activity toward the human ovarian A2780
cancer cell line with IC50 values ranging from 2.5 to 6.5 μM
(Table 4), close to the value we reported recently for [(η5-
C5Me5)Ir(phpy)Cl].3b Thus the introduction of C,N-chelating
ligands is an effective method for switching on the cancer cell
cytotoxicity of Cp* IrIII complexes. The strong binding of Ir to
nucleobases, especially to guanine bases, may provide an
important contribution toward the cytotoxicity. Also the neu-
tral C,N-complexes display more hydrophobic character than
Interactions with Nucleobases. Interaction with DNA is
often associated with the cytotoxicity of metal anticancer drugs.13
4708
dx.doi.org/10.1021/om2005468 |Organometallics 2011, 30, 4702–4710

Organometallics
ARTICLE
the positively charged N,N-complexes3b and therefore possess
enhanced cellular uptake, which may also contribute to the
cytotoxicity. The introduction of substituents on the phenyl-
pyridine ring enhanced the cytotoxicity of [(η5-C5Me5)Ir-
(phpy)Cl], Table 4. Particularly active is the complex [(η5-
C5Me5)Ir(phq)Cl] (2), which is 4 times more potent than[(η5-
C5Me5)Ir(phpy)Cl]. The ability of the phq ligand to inter-
calate into DNA24 may contribute to this enhanced potency.
The introduction of phenyl or biphenyl substituents onto the
tetramethylcyclopentadienyl ring to give complexes [(η5-
C5Me4C6H5)Ir(phpy)Cl] (4) and [(η5-C5Me4C6H4C6H5)Ir-
(phpy)Cl] (5) results in a dramatic increase in cytotoxicity
compared to the parent Cp* complex [(η5-C5Me5)Ir-
(phpy)Cl], Table 4 and Figure S4. The activity of RuII arene
complexes also increases with the size of the coordinated
arene.25 This suggests that the phenyl groups may play a crucial
role in the mechanism of action of these phenylpyridine
complexes. First, the phenyl or biphenyl ring increases the
hydrophobicity of the molecule, which may assist with passage
across cell membranes. In addition, the extended phenyl rings
can intercalate into DNA, thus causing distortion of DNA
structure. We have reported that the intercalative ability of
1,10-phenanthroline IrIII chlorido complexes increases in the
order of Cpxbiph > Cpxph > Cp*.3a Complexes 4 and 5, contain-
ing phenyl or biphenyl substitutions, may interact with DNA by
a dual mode: nucleobase binding to iridium accompanied by
intercalation of the phenyl groups, which is a different mecha-
nism of action from that of cisplatin.
complexes studied due to steric hindrance from the chelating
ligand 2-phenylquinoline.
All the C,N-complexes showed very promising anticancer
activity toward A2780 human ovarian cancer cells. The Cp*
complexes possess activity comparable to cisplatin. The intro-
duction of a phenyl or biphenyl substituent significantly im-
proved their cytotoxicity, especially for the Cpxbiph complex 5,
which has submicromolar activity against A2780 cancer cells.
The strong binding to guanine bases and hydrophobicity may
contribute to their high activity. This study shows that desirable
features can be introduced into this class of complexes to
optimize their design as anticancer agents.
’ ASSOCIATED CONTENT
S
Supporting Information. Details of the extent of nucleo-
b
base binding (Table S1), aqueous chemistry (Figures S1 and S2),
nucleobase studies (Figure S3), and IC50 comparison (Figure
S4). This material is available free of charge via the Internet at
http://pubs.acs.org. X-ray crystallographic data in CIF format
are available from the Cambridge Crystallographic Data Centre
(http://www.ccdc.cam.ac.uk).
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: P.J.Sadler@warwick.ac.uk.
’ ACKNOWLEDGMENT
’ CONCLUSIONS
Z.L. was supported by a University of Warwick Research
Scholarship (WPRS). We thank the ERC (grant no. 247450 for
P.J.S.), EPSRC, ORSAS, ERDF, and AWM for Science City
funding, and members of COST Action D39 for stimulating
discussions.
Iridium-based anticancer agents, including organometallic
iridium complexes, are currently attracting attention as potential
anticancer agents with novel mechanisms of action. We have
studied here the effects of changing the Cpx ligand and negatively
charged C,N-chelating ligand of IrIII cyclopentadienyl complexes
of the type [(η5-Cpx)Ir(C∧N)Cl] on the hydrolysis of the
chlorido complexes, acidity of the aqua adducts, nucleobase
binding, and cancer cell cytotoxicity.
All the complexes undergo rapid hydrolysis (<5 min at 278 K)
due to the strongly electron-donating methyl groups on the Cp
ring and negatively charged C,N-chelating ligand. However, the
complexes are likely to be present in their unhydrolyzed forms in
the extracellular fluid and cell cytoplasm (typically 104 and
23 mM [Clꢀ]), whereas they are likely to be activated by
hydrolysis in the cell nucleus ([Clꢀ] ca. 4 mM). Generally, the
aqua adducts of the C,N-complexes studied here possess low
acidity, with pKa values 1.9 units higher than N,N-analogues,
which ensures that the active aqua adduct is the major form after
hydrolysis at physiological pH and may contribute to the strong
binding to guanine. The substituents on both the Cp* ring and on
2-phenylpyridine can fine-tune the acidity of aqua adduts accord-
ing to their electronic effects. Hydroxo-bridged dimers {[(η5-
Cpx)Ir]2(μ-OD)3}+ (6ꢀ8) are observed at high pH (>8.7),
which suggests the stability of C,N-complexes is not as high as N,
N-analogues under strongly basic conditions.
’ REFERENCES
(1) (a) Medicinal Organometallic Chemistry (Topics in Organometallic
Chemistry), 1st ed.; Jaouen, G.; Metzler-Nolte, N., Eds.; Springer-Verlag:
Heidelberg, Germany, 2010; Vol. 32. (b) Gasser, G.; Ott, I.; Metzler-
Nolte, N. J. Med. Chem. 2011, 54, 3–25. (c) Fish, R. H. Aust. J. Chem. 2010,
63, 1505–1513. (d) Suss-Fink, G. Dalton Trans. 2010, 39, 1673–1688.
(e) Dyson, P. J.; Sava, G. Dalton Trans. 2006, 1929–1933. (f) Bruijnincx,
P. C. A.; Sadler, P. J. In Advances in Inorganic Chemistry; Rudi van, E.,
Hubbard, C. D., Eds.; Academic Press: New York, 2009; Vol. 61,
pp 1ꢀ62.
(2) (a) Allen, O. R.; Gott, A. L.; Hartley, J. A.; Hartley, J. M.; Knox,
R. J.; McGowan, P. C. Dalton Trans. 2007, 5082–5090. (b) Tan, Y. L. K.;
Pigeon, P.; Hillard, E. A.; Top, S.; Plamont, M.-A.; Vessieres, A.;
McGlinchey, M. J.; Muller-Bunz, H.; Jaouen, G. Dalton Trans.
2009, 10871–10881. (c) van Rijt, S. H.; Sadler, P. J. Drug Discovery
Today 2009, 14, 1089–1097. (d) Strohfeldt, K.; Tacke, M. Chem. Soc.
Rev. 2008, 37, 1174–1187. (e) Eger, S.; Immel, T. A.; Claffey, J.; Muller-
Bunz, H.; Tacke, M.; Groth, U.; Huhn, T. Inorg. Chem. 2010,
49, 1292–1294. (f) Vessiꢀeres, A.; Top, S.; Pigeon, P.; Hillard, E.;
Boubeker, L.; Spera, D.; Jaouen, G. J. Med. Chem. 2005,
48, 3937–3940. (g) Chatterjee, S.; Biondi, I.; Dyson, P.; Bhattacharyya,
A. J. Biol. Inorg. Chem. 2011, 1–10. (h) Hanif, M.; Schaaf, P.; Kandioller,
W.; Hejl, M.; Jakupec, M. A.; Roller, A.; Keppler, B. K.; Hartinger, C. G.
Aust. J. Chem. 2010, 63, 1521–1528.
Complexes 1 and 3ꢀ5 show strong binding to the nucleobase
guanine, even in the presence of 4 mM [Clꢀ]. Unlike the IrIII
complexes containing N,N-chelating ligands,3a the C,N-com-
plexes can also bind to adenine. However, they show a strong
preference for binding to guanine over adenine. Complex 2
displayed the lowest extent of nucleobase binding among the
(3) (a) Liu, Z.; Habtemariam, A.; Pizarro, A. M.; Fletcher, S. A.;
Kisova, A.; Vrana, O.; Salassa, L.; Bruijnincx, P. C. A.; Clarkson, G. J.;
Brabec, V.; Sadler, P. J. J. Med. Chem. 2011, 54, 3011–3026. (b) Liu, Z.;
Salassa, L.; Habtemariam, A.; Pizarro, A. M.; Clarkson, G. J.; Sadler, P. J.
4709
dx.doi.org/10.1021/om2005468 |Organometallics 2011, 30, 4702–4710

Organometallics
ARTICLE
Inorg. Chem. 2011, 50, 5777–5783. (c) Sch€afer, S.; Sheldrick, W. S. J.
Organomet. Chem. 2007, 692, 1300–1309. (d) Sliwinska, U.; Pruchnik,
F. P.; Ulaszewski, S.; Latocha, M.; Nawrocka-Musial, D. Polyhedron
2010, 29, 1653–1659. (e) Leung, S.-K.; Kwok, K. Y.; Zhang, K. Y.; Lo, K.
K.-W. Inorg. Chem. 2010, 49, 4984–4995. (f) Ali Nazif, M.; Bangert, J.-A.;
Ott, I.; Gust, R.; Stoll, R.; Sheldrick, W. S. J. Inorg. Biochem. 2009,
103, 1405–1414. (g) Amouri, H.; Moussa, J.; Renfrew, A. K.; Dyson,
P. J.; Rager, M. N.; Chamoreau, L.-M. Angew. Chem., Int. Ed. 2010,
49, 7530–7533. (h) Wirth, S.; Rohbogner, C.; Cieslak, M.; Kazmierczak-
Baranska, J.; Donevski, S.; Nawrot, B.; Lorenz, I.-P. J. Biol. Inorg. Chem.
2010, 15, 429–440. (i) Casini, A.; Edafe, F.; Erlandsson, M.; Gonsalvi,
L.; Ciancetta, A.; Re, N.; Ienco, A.; Messori, L.; Peruzzini, M.; Dyson,
P. J. Dalton Trans. 2010, 39, 5556–5563. (j) Gras, M.; Therrien, B.; S€uss-
Fink, G.; Casini, A.; Edafe, F.; Dyson, P. J. J. Organomet. Chem. 2010,
695, 1119–1125. (k) Hartinger, C. G. Angew. Chem., Int. Ed. 2010,
49, 8304–8305.
(23) Melchart, M.; Habtemariam, A.; Parsons, S.; Sadler, P. J. Inorg.
Biochem. 2007, 101, 1903–1912.
(24) (a) Atwell, G. J.; Bos, C. D.; Baguley, B. C.; Denny, W. A. J. Med.
Chem. 1988, 31, 1048–1052. (b) Atwell, G. J.; Baguley, B. C.; Denny,
W. A. J. Med. Chem. 1989, 32, 396–401.
(25) (a) Morris, R. E.; Aird, R. E.; del Socorro Murdoch, P.; Chen,
H.; Cummings, J.; Hughes, N. D.; Parsons, S.; Parkin, A.; Boyd, G.;
Jodrell, D. I.; Sadler, P. J. J. Med. Chem. 2001, 44, 3616–3621. (b) Aird,
R. E.; Cummings, J.; Ritchie, A. A.; Muir, M.; Morris, R. E.; Chen, H.;
Sadler, P. J.; Jodrell, D. I. Br. J. Cancer 2002, 86, 1652–1657.
(4) (a) Helm, L.; Merbach, A. E. Coord. Chem. Rev. 1999,
187, 151–181. (b) Wilbuer, A.; Vlecken, D. H.; Schmitz, D. J.; Kr€aling,
K.; Harms, K.; Bagowski, C. P.; Meggers, E. Angew. Chem., Int. Ed. 2010,
49, 3839–3842.
(5) (a) van Rijt, S. H.; Peacock, A. F. A.; Johnstone, R. D. L.; Parsons,
S.; Sadler, P. J. Inorg. Chem. 2009, 48, 1753–1762. (b) Kostrhunova, H.;
Florian, J.; Novakova, O.; Peacock, A. F. A.; Sadler, P. J.; Brabec, V.
J. Med. Chem. 2008, 51, 3635–3643.
(6) White, C. Y., A.; Maitlis, P. M. Inorg. Chem. 1992, 29, 228–234.
(7) CrysAlis PRO; Oxford Diffraction Ltd.: Abington, Oxfordshire,
U.K., 2007.
(8) Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467–473.
(9) Sheldrick, G. M. SHELXL-97; University of G€ottingen:
G€ottingen, Germany, 1997.
(10) Krezel, A.; Bal, W. J. Inorg. Biochem. 2004, 98, 161–166.
(11) Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J.;
Vistica, D.; Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J. Natl.
Cancer Inst. 1990, 82, 1107–1112.
(12) Martin, R. B. In Cisplatin: Chemistry and Biochemistry of a
Leading Anticancer Drug; Lippert, B., Ed.; VHCA & Wiley-VCH: Z€urich,
Switzerland, 1999; pp 181ꢀ205.
(13) (a) Zhang, C. X.; Lippard, S. J. Curr. Opin. Chem. Biol. 2003,
7, 481–489. (b) Deubel, D. V.; Lau, J. K.-C. Chem. Commun. 2006,
2451–2453.
(14) Li, L.; Brennessel, W. W.; Jones, W. D. J. Am. Chem. Soc. 2008,
130, 12414–12419.
(15) Hohmann, H.; Hellquist, B.; van Eldik, R. Inorg. Chem. 1992,
31, 345–351.
(16) Pizarro, A. M.; Habtemariam, A.; Sadler, P. J. In Medicinal
Organometallic Chemistry (Topics in Organometallic Chemistry), 1st ed.;
Jaouen, G., Metzler-Nolte, N., Eds.; Springer-Verlag: Heidelberg,
Germany, 2010; Vol. 32, pp 21ꢀ56.
(17) Koelle, U. Coord. Chem. Rev. 1994, 135ꢀ136, 623–650.
(18) Wang, F.; Habtemariam, A.; van der Geer, E. P. L.; Fernꢁandez,
R.; Melchart, M.; Deeth, R. J.; Aird, R.; Guichard, S.; Fabbiani, F. P. A.;
Lozano-Casal, P.; Oswald, I. D. H.; Jodrell, D. I.; Parsons, S.; Sadler, P. J.
Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 18269–18274.
(19) (a) Fernꢁandez, R.; Melchart, M.; Habtemariam, A.; Parsons, S.;
Sadler, P. J. Chem.—Eur. J. 2004, 10, 5173–5179. (b) Peacock, A. F. A.;
Melchart, M.; Deeth, R. J.; Habtemariam, A.; Parsons, S.; Sadler, P. J.
Chem.—Eur. J. 2007, 13, 2601–2613.
(20) (a) Peacock, A. F. A.; Habtemariam, A.; Fernꢁandez, R.;
Walland, V.; Fabbiani, F. P. A.; Parsons, S.; Aird, R. E.; Jodrell, D. I.;
Sadler, P. J. J. Am. Chem. Soc. 2006, 128, 1739–1748. (b) Peacock,
A. F. A.; Parsons, S.; Sadler, P. J. J. Am. Chem. Soc. 2007, 129, 3348–3357.
(c) Kljun, J.; Bytzek, A. K.; Kandioller, W.; Bartel, C.; Jakupec, M. A.;
Hartinger, C. G.; Keppler, B. K.; Turel, I. Organometallics 2011,
30, 2506–2512.
(21) Pullman, B.; Pullman, A. Biochim. Biophys. Acta 1959, 36,
343–350.
(22) Baik, M.-H.; Friesner, R. A.; Lippard, S. J. J. Am. Chem. Soc.
2003, 125, 14082–14092.
4710
dx.doi.org/10.1021/om2005468 |Organometallics 2011, 30, 4702–4710