Effects of the ancillary ligands of polypyridyl ruthenium(II) complexes on the DNA-binding behaviors

Article abstract of DOI:10.1039/b212826h

A new polypyridyl ligand PMIP {PMIP = 2-(4-methylphenyl)imidazo[4,5-f]1,10-phenanthroline}, and its Ru(II) complexes, [Ru(bpy)₂PMIP]²⁺ 1 (bpy = 2,2′-bipyridine), [Ru(phen)₂PMIP]²⁺ 2 (phen = 1,10-phenanthroline)

Full text of DOI:10.1039/b212826h

View Article Online / Journal Homepage / Table of Contents for this issue
Effects of the ancillary ligands of polypyridyl ruthenium(II) complexes
on the DNA-binding behaviors
Hong Xu,ab Kang-Cheng Zheng,*a Hong Deng,a Li-Jun Lin,a Qian-Ling Zhangb and
Liang-Nian Ji*a
a
Department of Chemistry, Zhongshan University, Guangzhou 510275, P. R. China.
E-mail: cesjln@zsu.edu.cn; Fax: 86-20-8403-5497; Tel: 86-20-8411-0115
Department of Chemistry and Biology, Normal College, Shenzhen University, Shenzhen
518060, P. R. China. E-mail: xuhong@szu.edu.cn; Fax: 86-755-2653-4605;
Tel: 86-755-2626-8581
b
Received (in Montpellier, France) 2nd January 2003, Accepted 9th April 2003
First published as an Advance Article on the web 2nd July 2003
A new polypyridyl ligand PMIP {PMIP ¼ 2-(4-methylphenyl)imidazo[4,5-f]1,10-phenanthroline}, and its
Ru(II) complexes, [Ru(bpy)2PMIP]2+ 1 (bpy ¼ 2,20-bipyridine), [Ru(phen)2PMIP]2+ 2 (phen ¼ 1,10-
phenanthroline) and [Ru(dmp)2PMIP]2+ 3 (dmp ¼ 2,9-dimethyl-1,10-phenanthroline), have been synthesized
and characterized. The binding of the three complexes to calf thymus DNA (CT DNA) has been investigated
with spectrophotometic methods, viscosity measurements, as well as equilibrium dialysis and circular dichroism
spectroscopy. Complexes 1 and 2 can bind to CT DNA through intercalation. Complex 3 binds to CT DNA via
partial intercalative mode. All the three complexes can enantioselectively interact with CT DNA in a way.
Complex 3 is a much better candidate as an enantioselective binder to CT DNA than complex 1 and complex 2.
The D enantiomer of complex 1 is slightly predominant for binding to CT DNA to the L enantiomer. The
experimental results suggest that the ancillary ligands of polypyridyl Ru(II) complexes have significant effects on
the spectral properties and the DNA-binding behaviors of the complexes. On the basis of the experiments, the
theoretical calculations applying the density functional theory (DFT) on the level of B3LYP/LanL2DZ basis
set for these three complexes have been used to further discuss the trend in the binding strength or binding
constants (Kb) of the complexes to DNA, as well as predict roughly some of their spectral properties.
intercalative ligand because of the steric hindrance of the
methyl group, as the intercalative ligand.
Introduction
The interaction between transition metal complexes and DNA
has been extensively studied.1 Binding studies of small mole-
cules to DNA are very important in the development of new
therapeutic reagents and DNA molecular probes.2 Polypyridyl
Ru(II) complexes can bind to DNA in a non-covalent interac-
tions fashion such as electrostatic binding, groove binding,3
intercalative binding and partial intercalative binding.4 Many
useful applications of these complexes require that the com-
plexes bind to DNA through an intercalative mode. Therefore
much work has been done on modifying the intercalative
ligand,5 and the influence of the ancillary ligands of the com-
plexes has received little attention. Since the octahedral poly-
pyridyl Ru(II) complexes bind to DNA in three dimensions,
the ancillary ligands can also play an important role in govern-
ing DNA-binding of these complexes. So it is significant and
interesting to find the effects of ancillary ligands on interaction
and the binding mode of the complexes to DNA. To more
clearly study the effects of ancillary ligands on DNA-binding
behaviors of the polypyridyl Ru(II) complexes, the selection
of intercalative ligand is also very important. An appropriate
intercalative ligand is helpful to distinguishing the small differ-
ences of interaction of the complexes containing different ancil-
lary ligands with DNA. In our laboratory, a series of similar
Ru(II) complexes with PIP (PIP ¼ 2-phenylimidazo[4,5-f]1,10-
phenanthroline), a comparatively strong intercalative ligand,4b
as the parent compound have been synthesized. In this
paper, we select PMIP ligand {PMIP ¼ 2-(4-methylphenyl)-
imidazo[4,5-f]1,10-phenanthroline}, which is likely a moderate
In addition, Ru(II) polypyridine complexes have also
recently attracted the attention of many theoretical chemists.6
More and more theoretical computations, especially for the
computations applying the density functional theory (DFT)
method7 on Ru(II) and other transition metal complexes were
reported,8 because DFT can better consider electron correla-
tion energies, obviously reduces the computation expenses
and suits such kinds of singlet state complexes.8d–8g Recently,
D. P. Rillema et al. suggested that the HOMO and LUMO dis-
tributions for Ru(II) two ring diimine complex cations from
DFT calculations support the idea that the lowest energy tran-
sitions are metal-to-ligand charge transfer and that the LUMO
energy for the mixed ligand complexes is mainly located on the
ligand with the lowest LUMO energy in the corresponding
complex [RuL3]2+ 8a
Y. Zhang et al. reported hydrolysis the-
.
ory for cisplatin and its analogues based on density functional
studies.8b N. Kurita and K. Kobayashi further reported den-
sity functional MO calculation for stacked DNA base-pairs
with backbones.6e We also reported the studies on di-substitu-
tion effects, electron structures and related properties of some
Ru(II) polypyridyl complexes with the DFT method.8d–8g
These direct theoretical efforts on the level of molecular elec-
tronic structures of the complexes are also very significant in
guiding experimental work.
We report herein the syntheses and characterization of a
new polypyridyl ligand, PMIP and its Ru(II) complexes, [Ru-
(bpy)2PMIP]2+ 1 (bpy ¼ 2,20-bipyridine), [Ru(phen)2PMIP]2+
2 (phen ¼ 1,10-phenanthroline) and [Ru(dmp)2PMIP]2+
3
DOI: 10.1039/b212826h
New J. Chem., 2003, 27, 1255–1263
1255
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003

View Article Online
(dmp ¼ 2,9-dimethyl-1,10-phenanthroline), the examination
of their different DNA-binding behaviors, and the application
of the DFT method to the explanation of the obtained experi-
mental regularities or trends.
56.03; H, 3.80; N, 10.87%. Calc. for C48H38Cl2N8O8Ru: C,
56.14; H, 3.70; N, 10.92%). H NMR (DMSO-d6): d 8.90 (d,
2H), 8.80 (d, 2H), 8.43 (t, 4H), 8.24 (d, 2H), 8.15 (d, 2H),
1
7.97 (d, 2H), 7.38 (m, 4H), 7.32 (d, 2H), 7.22 (br, 2H), 2.29
]
ꢂ 2+).
(s, 3H), 1.83 (s, 6H), 1.70 (s, 6H). m/z 827 ([M ꢂ 2ClO4
Experimental
Measurements
Syntheses
Elemental analyses (C, H and N) were carried out with a Per-
kin-Elmer 240C elemental analyzer. 1H NMR spectra were
recorded on a Bruker DRX-500 NMR spectrometer with
(CD3)2SO as solvent and SiMe4 as an internal standard. An
LCQ electrospray mass spectrometer (ESMS, Finnigan) was
employed for the investigation of charged metal complex spe-
cies in methanol solvent. UV-Vis spectra were recorded on a
Shimadzu UV-2501PC spectrophotometer. Emission spectra
were determined with a Shimadzu RF-5301PC fluorescence
spectrometer. The circular dichroism spectrum of the dialyzate
was measured on a JASCO J-715 spectropolarimeter.
1,10-phenanthroline-5,6-dione,9 cis-[Ru(bpy)2Cl2]ꢀ2H2O, cis-
[Ru(phen)2Cl2]ꢀ2H2O10 and cis-[Ru(dmp)2Cl2]ꢀ2H2O11 were
prepared according to literature procedures. Other reagents
and solvents were purchased commercially and used without
further purification unless otherwise noted.
Caution. Perchlorate salts of metal complexes with organic
ligands are potentially explosive, and only small amounts of
the material should be prepared and handled with great care.
All the experiments involving interaction of the complex
with DNA were conducted in twice distilled buffer containing
tris(hydroxymethyl)-aminomethane (Tris, 5 mM) and NaCl
(50 mM) and adjusted to pH 7.2 with hydrochloric acid. A
solution of calf thymus DNA in the buffer gave a ratio of
UV absorbance at 260 and 280 nm of about 1.8–1.9:1, indicat-
ing that the DNA was sufficiently free of protein.13 The DNA
concentration per nucleotide was determined by absorption
spectroscopy using the molar absorption coefficient (6600
Mꢂ1ꢀcmꢂ1) at 260 nm.14
PMIP. PMIP was synthesized by a method similar to the
one described previously.12 A mixture of 4-methylbenzalde-
hyde (3.5 mmol, 0.41 cm3), 1,10-phenanthroline-5,6-dione
(2.5 mmol, 0.525 g), ammonium acetate (50 mmol, 3.88 g)
and glacial acetic acid (10 cm3) was refluxed for about 2 h, then
cooled to room temperature and diluted with water (ca. 25
cm3). Dropwise addition of concentrated aqueous ammonia
gave yellow precipitates, which were collected and washed with
water. The crude products were purified by silica gel filtration
(60–100 mesh, ethanol). The principal yellow band was col-
lected. The solvent was removed by rotary evaporation, the
products were collected, then were dried at 50 ꢁC in vacuo.
Yield 0.588 g, 76% (Found: C, 77.26; H, 4.63; N, 17.92%. Calc.
for C20H14N4 : C, 77.42; H, 4.52; N, 18.06%). 1H NMR
(DMSO-d6): d13.56 (br, 1H), 9.03 (d, 2H), 8.95 (d, 2H), 8.21
(d, 2H), 7.82 (q, 2H), 7.41 (d, 2H), 2.42 (s, 3H). m/z 311
([M + 1]+).
Viscosity measurements were carried out using an Ubb-
dlodhe viscometer maintained at a constant temperature of
28.0 (ꢃ0.1) ꢁC in a thermostatic bath. DNA samples with an
approximate average length of 200 base pairs were prepared
by sonication in order to minimize complexities arising from
DNA flexibility.15 Flow time was measured with a digital stop-
watch. Each sample was measured three times and an average
flow time was calculated. Data were presented as (Z/Z0)1/3
versus binding ratio ([Ru]/[DNA]),16 where Z is the viscosity
of DNA in the presence of complex and Z0 is the viscosity
of DNA alone. Viscosity values were calculated from the
observed flow time of DNA-containing solutions (t > 100 s)
corrected for the flow time of buffer alone (t0), Z ¼ t ꢂ t0 .
For the absorption spectra, equal solution of DNA was
added to both complex solution and reference solution to elim-
inate the absorbance of DNA itself. The intrinsic binding con-
.
[Ru(bpy)2PMIP](ClO4)2 H2O 1. This complex was prepared
by the following method. A mixture of cis-[Ru(bpy)2Cl2]ꢀ2H2O
(0.5 mmol, 0.260 g), PMIP (0.5 mmol, 0.155 g), ethanol (10
cm3) and water (5 cm3) was refluxed under argon for 2 h to
give a clear red solution. After most of the ethanol solvent
was removed under reduced pressure, a red precipitate was
obtained by dropwise addition of a saturated aqueous NaClO4
solution. The product was purified by column chromatography
on alumina using acetonitrile–toluene (1:1 v/v) as eluent and
then dried in vacuo. Yield, 0.334 g, 71% (Found: C, 51.22;
H, 3.69; N, 11.67%. Calc. for C40H32Cl2N8O9Ru: C, 51.06;
H, 3.40; N, 11.92%). 1H NMR (DMSO-d6): d 9.08 (d, 2H),
8.85 (q, 4H), 8.21 (t, 4H), 8.10 (t, 2H), 8.02 (d, 2H), 7.90 (q,
2H), 7.85 (d, 2H), 7.59 (m, 4H), 7.45 (d, 2H), 7.35 (t, 2H),
stant Kb of the complex to CT DNA was determined from the
½DNAꢄ
eqn. (1)17 through a plot of
versus [DNA].
ea ꢂ ef
½DNAꢄ ½DNAꢄ
1
¼
þ
ð1Þ
ea ꢂ ef
eb ꢂ ef Kbðeb ꢂ ef Þ
2.43 (s, 3H). m/z 723.4 ([M ꢂ 2ClO4
]
ꢂ 2+), 823.1 ([M ꢂ
where [DNA] is the concentration of DNA in base pairs,
ea , ef and eb are respectively the apparent extinction coeffi-
Aobsd
ClO4ꢂ]+).
ꢀ
ꢁ
cient
, the extinction coefficient for free metal (M) complex
.
[Ru(phen)2PMIP](ClO4)2 H2O 2. This complex (red) was
synthesized in a manner identical to the one described for com-
plex 1, with 0.5 mmol, 0.284 g cis-[Ru(phen)2Cl2]ꢀ2H2O in
place of cis-[Ru(bpy)2Cl2]ꢀ2H2O. Yield, 0.316 g, 64% (Found:
C, 53.68; H, 3.19; N, 11.13%. Calc. for C44H32Cl2N8O9Ru:
½Mꢄ
and the extinction coefficient for the metal (M) complex in the fully
½DNAꢄ
bound form. In plots of
versus [DNA], Kb is given by the
e ꢂ ef
ratio of slope to intercept. a
According to the classical Stern–Volmer equation:18
1
C, 53.44; H, 3.24; N, 11.34%). H NMR (DMSO-d6): d 9.06
(d, 2H), 8.77 (d, 4H), 8.39 (d, 4H), 8.21 (d, 2H), 8.12 (d,
2H), 8.07 (d, 2H), 8.00 (d, 2H), 7.79 (m, 2H), 7.76 (m, 4H),
7.47 (d, 2H), 2.43 (s, 3H). m/z 772 ([M ꢂ 2ClO4ꢂ + 1]2+),
871 ([M ꢂ ClO4ꢂ]+).
I0=I ¼ 1 þ Kr
ð2Þ
For the steady-state emission quenching experiment using
[Fe(CN)6]4ꢂ as quencher, where I0 and I are the fluorescence
intensities in the absence and presence of [Fe(CN)6]4ꢂ, respec-
tively. K is a linear Stern–Volmer quenching constant depen-
dent on the ratio of the bound concentration of the Ru(II)
complex to the concentration of DNA. r is the concentration
of the quencher [Fe(CN)6]4ꢂ. For the steady-state competitive
[Ru(dmp)2PMIP](ClO4)2 3. This complex (red) was synthe-
sized in an identical manner to that described for complex 1,
with 0.5 mmol, 0.321 g cis-[Ru(dmp)2Cl2]ꢀ2H2O in place of
cis-[Ru(bpy)2Cl2]ꢀ2H2O. Yield, 0.344 g, 67% (Found: C,
1256
New J. Chem., 2003, 27, 1255–1263

View Article Online
binding experiment, I0 and I are the fluorescence intensities in
the absence and presence of complex, respectively. Stern–
Volmer quenching constant K is dependent on the ratio of
the bound concentration of EB to the concentration of
DNA. r is the ratio of total concentration of the complex to
that of DNA ([Ru]/[DNA]). For both the emission quenching
experiment and the competitive binding experiment, the Stern–
Volmer quenching constant K is given by the ratio of the slope
to intercept in the plot of I0/I versus r.
The dialysis membrane was purchased from Union Carbide
Co. and treated by means of the general procedure before
use.19 Equilibrium dialysis was carried out in the dark and held
at 4 ꢁC for 48 hours with 5 cm3 of calf thymus DNA (1.0 mM)
sealed in a dialysis bag and 10 cm3 of the complexes (100 mM)
outside the bag.
interactions, the stereographs of some related frontier mole-
cular orbitals of the complexes were drawn with Molden
v3.6 program23 based on the obtained computational results.
Results and discussion
Viscosity studies
Hydrodynamic measurements that are sensitive to length
change (i.e. viscosity and sedimentation) are regarded as the
least ambiguous and most critical tests of a binding model in
solution in the absence of crystallographic structural data.24,25
A classical intercalation model demands that the DNA helix
lengthens as base pairs are separated to accommodate the
bound ligand, leading to the increase of DNA viscosity. In
contrast, a partial, non-classical intercalation of ligand could
bend (or kink) the DNA helix, reduce its effective length
and, concomitantly, its viscosity.24,25
Theoretical
Each of these three complexes forms from a Ru(II) ion, one
main ligand or intercalated ligand (PMIP) and two ligands
called ancillary ones (bpy, phen or dmp). The structural mod-
els and the corresponding 1H NMR spectra of the studied
complexes are shown in Fig. 1. The full geometry optimization
computations were performed for them, applying the DFT-
B3LYP method7 and LanL2DZ basis set,20 and taking singlet
state of the complexes.21 All computations were performed
with G98 quantum chemistry program-package.22 In order to
vividly depict the details of the frontier molecular orbital
The effects of complexes 1, 2 and 3 on the viscosity of DNA
are shown in Fig. 2. For complexes 1 and 2, the viscosity of
DNA increases steadily with increasing concentration of the
complex, and the extent of increase observed for complex 1
is smaller than that for complex 2. By contrast, on increasing
the amount of complex 3, the relative viscosity of DNA
decreases. The experimental results suggest that complexes 1
and 2 bind to DNA through a classical intercalation mode,
complex 3 binds to DNA not by the classical intercalation
mode but by the partial, non-classical intercalation model,
Fig. 1 1H NMR spectra of the aromatic region of complex 1 (top), complex 2 (middle) and complex 3 (bottom).
New J. Chem., 2003, 27, 1255–1263
1257

View Article Online
Fig. 2 Effect of increasing amounts of complex 1 (˘), complex 2 (L)
and complex 3 (G) on the relative viscosities of CT DNA at
28.0 ꢃ 0.1 ꢁC. [DNA] ¼ 0.5 mM.
and that complex 2 intercalated more strongly than complex 1.
This difference of DNA-binding mode between complexes 1, 2
and complex 3 should be caused by their different ancillary
ligands. Complex 1 or 2 containing the ancillary ligand bpy
or phen can intercalate deeply into adjacent DNA base pairs.
This would cause an extension in the helix and increase the
viscosity of DNA. For complex 3, due to the steric hindrance
of the methyl groups located at the 2- and 9-positions of the
phen ring in dmp ancillary ligand, the complex could not com-
pletely intercalate into DNA base pairs. The partial intercala-
tion may act as a ‘‘wedge’’ to pry apart one side of a base-pair
stack, as observed for the D-[Ru(phen)3]2+ 24,25
but not fully
,
separate the stack as required by the classical intercalation
mode. This would cause a static bend or kink in the helix
and decrease the viscosity of DNA.
Absorption spectroscopic studies
The electronic absorption spectra of complexes 1, 2 and 3
mainly consist of two resolved bands. The low energy absorp-
tion band centered at 455–472 nm is assigned to metal-to-
ligand charge transfer (MLCT) transition and the other band
centered at 263–286 nm is attributed to intraligand (IL) p–p*
transition by comparison with the spectrum of other polypyr-
idyl Ru(II) complexes.21 Fixed amounts (10 mM) of complexes
1, 2 and 3 were titrated with increasing amounts of DNA. The
electronic spectral traces are given in Fig. 3. For complex 1, the
absorption spectra show clearly that the addition of DNA to
the complexes yields hypochromism about 28.9%, 20.6% and
a red shift of 2, 4 nm at the ratio of [DNA]/[Ru] of 8 in the
IL and MLCT band, respectively. As the DNA concentration
is increased, for complex 2, the hypochromism in the IL band
reaches as high as 30.3% at 263 nm with 2 nm red shift at the
ratio of [DNA]/[Ru] of 3. The MLCT band at 455 nm shows
hypochromism about 14.2% and a red shift of 11 nm under the
same experimental conditions. For complex 3, upon addition
of DNA, the IL band at 271.5 nm displays about 25.4% hypo-
chromism with 1.5 nm red shift at the ratio of [DNA]/[Ru]
of 10. The MLCT band at 466.5 nm exhibits hypochromism
about 8.8% and a red shift of 6 nm under the same experimen-
tal conditions. Obviously, these spectral characteristics suggest
that all the three complexes interact with DNA most likely
through a mode that involves a stacking interaction between
the aromatic chromophore and the base pairs of DNA.
Fig. 3 Absorption spectral of complex 1 (top), complex 2 (middle)
and complex 3 (bottom) in Tris-HCl buffer upon addition of CT
DNA with subtraction of the DNA absorbance. Arrow shows the
absorbance changes upon increasing DNA concentrations. Inset: plots
of [DNA]/(ea–ef) vs. [DNA] for the titration of DNA with complexes;
L, experimental data points; full lines, linear fitting of the data.
2.03 ꢅ 104 Mꢂ1, 5.70ꢅ 104 Mꢂ1 and 1.17 ꢅ 104 Mꢂ1, respectively.
The results indicate that, as the ancillary ligand varies from
phen, bpy to dmp, the DNA binding affinity of polypyridyl
Ru(II) complexes declines. Similar case was observed in pre-
vious literature.26 Comparing the intrinsic binding constant
of the three complexes with that of those so-called DNA-
To compare quantitatively the affinity of the three com-
plexes bound to DNA, the intrinsic binding constants Kb of
these complexes to CT DNA were determined by monitoring
the changes of absorbance in the IL band with increasing
concentration of DNA using eqn. (1).17 The intrinsic binding
constant Kb of complexes 1, 2 and 3 were obtained about
intercalative Ru(II) complexes (1.1 ꢅ 104 ꢂ 4.8 ꢅ 104
M
ꢂ1),27
these results of absorption spectroscopic studies support that
of above viscosity studies, that is, complexes 1 and 2 bind to
DNA by intercalation and complex 3 is close to the border
between a classical intercalation and non-classical intercala-
tion. These results may be explained by the fact that, for the
1258
New J. Chem., 2003, 27, 1255–1263

View Article Online
ancillary ligand, on going from bpy to phen, the plane area
and hydrophobicity increase, leading to a greater binding
affinity to DNA. On the other hand, for dmp ligand, substi-
tution on the 2- and 9-positions of phen may cause severe
steric constraints near the core of Ru(II) when the complex
intercalates into the DNA base pairs. The methyl groups
may come into close proximity of the base pairs at the inter-
calation site. These steric clashes then prevent the complex
from intercalating effectively, and thus cause a diminution
of the intrinsic binding constant.
ment inside the DNA helix reduces the accessibility of solvent
water molecules to the complex and the complex mobility is
restricted at the binding site, leading to decrease of the vibra-
tional modes of relaxation.
We can also derive the binding constants of the two com-
plexes interacting with DNA from the emission spectra using
the luminescence titration method.24 The binding data
obtained from the emission spectra were fitted using the
McGhee and Von Hippel equation28 to acquire the binding
parameters. The intrinsic binding constants Kb of 2.2(ꢃ0.1) ꢅ
105 Mꢂ1 for complex 1 and 4.7(ꢃ0.3) ꢅ 105 Mꢂ1 for complex
2 were determined, which are comparable to that observed
for the similar complex [Ru(bpy)2MCP]2+ (1.8 ꢅ 105 Mꢂ1 12
)
Fluorescence spectroscopic studies
by the same method. Comparing with that obtained from
absorption spectra, although the binding constants obtained
from fluorescence with the McGhee–von Hippel method are
different from those obtained from absorption with the method
suggested by Wolfe et al.,17 both sets of binding constants
show that complex 2 binds to DNA more avidly than does
complex 1, and this difference between the two sets of binding
constants should be caused by the different spectroscopy and
the different calculation method. The binding size, n, was
1.85(ꢃ0.05) base pairs for complex 1 and 2.56(ꢃ0.05) base
pairs for complex 2, which are also similar to those of many
Ru(II) polypyridine complexes.29
Steady-state emission quenching experiments using
[Fe(CN)6]4ꢂ as quencher are also used to observe the binding
of the complexes to DNA. As illustrated in Fig. 5, in the
absence of DNA, complex 1 and complex 2 were efficiently
quenched by [Fe(CN)6]4ꢂ, resulting in a linear Stern–Volmer
plot (slopes 4.18 and 5.39, respectively). In the presence of
DNA, however, the slope of the plot is remarkably decreased
and nearly equal to zero (slopes 0.11 and 0.03, respectively).
This may be explained by the fact that the bound cations of
complexes 1 and 2 are protected from the anionic water-bound
quencher by the array of negative charges along the DNA
phosphate backbone. The curvature reflects different degrees
of protection or relative accessibility of bound cations. A lar-
ger slope for the Stern–Volmer curve parallels poorer protec-
tion and low binding. Therefore complex 2 binds to DNA
more strongly than does complex 1. Such a trend is consistent
with the previous viscosity studies and absorption spectral
results.
Complexes 1 and 2 can emit luminescence in Tris buffer. Fixed
amounts (10 mM) of complexes 1 and 2 were respectively
titrated with increasing amounts of DNA, over a range of
DNA concentrations from 10 to 200 mM. An excitation wave-
length of 460 nm was used, and the total emission intensity was
monitored at 590 nm. The results of the emission titrations for
complex 1 and 2 with DNA are illustrated with the titration
curves in Fig. 4. Upon addition of DNA, for complexes 1
and 2, the emission intensity grows to around 1.53 and 1.78
times larger than that in the absence of DNA and saturates
at a [DNA]/[Ru] ratio of 20, respectively. This implies that
both the complexes can strongly interact with DNA and be
protected by DNA efficiently, since the hydrophobic environ-
For complex 3, no emission is observed either in the
Tris buffer or in the presence of CT DNA. It also shows no
photoluminescence examined in any of the organic solvents.
Fig. 4 Emission spectra of complex 1 (top) and complex 2 (bottom)
in Tris-HCl buffer in the absence and presence of CT DNA. Arrow
shows the intensity changes upon increasing DNA concentrations.
Inset: plots of relative integrated emission intensity versus [DNA]/
[Ru].
Fig. 5 Emission quenching curves of complex 1 and complex 2 with
increasing concentration of quencher [Fe(CN)6]4ꢂ in the absence (for
complex 1 shown as G and complex 2 shown as L) and presence
(for complex 1 shown as ` and complex 2 shown as K) of CT
DNA. [Ru] ¼ 4 mM, [DNA]/[Ru] ¼ 40.
New J. Chem., 2003, 27, 1255–1263
1259

View Article Online
This could be caused by a reduced ligand field making the
excited states accessible. The reduced ligand field is firmed by
the computed coordination bond lengths (Ru–N), which are
0.2124 nm (main-ligand) and 0.2157 nm (co-ligand) for com-
plex 3, and 0.2105 nm and 0.2106 nm for complex 2 and
0.2108 nm and 0.2097 nm for complex 1 respectively. Some
computed frontier molecular orbital energies were also listed
in Fig. 6. They show that the energy of LUMO for complex
3 is much higher than that for complexes 1 and 2, so the
excited state of complex 3 should be unstable, and can easily
return to the ground state through nonradiative energy dissi-
pation. Similar cases were observed for this type of Ru(II) com-
plex.26,30 Steady-state competitive binding experiments using
complex 3 as quencher may give further information about
the binding of the complex to DNA. As we know, ethidium
bromide (EB) emits intense fluorescence light in the presence
of DNA. It was previously reported that the enhanced fluores-
cence can be quenched, at least partially by the addition of a
second molecule.18,31 The extent of quenching fluorescence of
EB bound to DNA is used to determine the extent of binding
between the second molecule and DNA. The emission spectra
of EB bound to DNA in the absence and the presence of com-
plex 3 are shown in Fig. 7. The fluorescence quenching curve of
DNA-bound EB by complex 3 (inset in Fig. 7) illustrates that
the quenching of EB bound to DNA by complex 3 is in good
agreement with the linear Stern–Volmer equation. This also
proves that the partial replacement of EB bound to DNA by
[Ru(dmp)2PMIP]2+ results in a decrease of the fluorescence
intensity.
Fig. 7 Emission spectra of EB bound to DNA in the presence of
complex 2. [EB] ¼ 20 mM, [DNA] ¼ 100 mM, [Ru]/[DNA] ¼ 0, 0.05,
0.10, 0.15, 0.20, 0.25 and 0.30. Arrow shows the intensity changes upon
increasing concentrations of the complexes. Inset: Fluorescence
quenching curve of DNA-bound EB by complex 2.
cates that the D form is slightly predominant for binding to CT
DNA to the L enantiomer. According to the proposed binding
model,30,32 the D enantiomer of the complex, a right-handed
propeller-like structure, will display a greater affinity than
Enantioselective binding
Equilibrium dialysis experiments may offer the opportunity
to examine the enantioselectivity of complexes binding to
DNA. Racemic solutions of the three complexes were dialyzed
against CT DNA for 48 hours and then subjected to CD ana-
lysis. The CD spectra in the UV region of the dialysates of
complexes 1, 2 and 3 are shown in the top of Fig. 8. All the
dialysates of complexes 1, 2 and 3 show CD signals. The CD
signals for complexes 1 and 2 are weak, which shows that
the extents of preferential enantiomeric binding for complexes
1 and 2 are small.
Complex 1 has been resolved into its pure enantiomers. The
CD spectra of the enantiomers of complex 1 in the UV region
are shown in the bottom of Fig. 8. For complex 1, the exhib-
ited CD spectra were assigned to the L form, as judged by
comparison with the CD spectra of its enantiomers. This indi-
Fig. 8 Circular dichroism spectra of the dialysates of complex 1 (full
lines), complex 2 (broken lines) and complex 3 (dotted lines) after 48 h
dialysis against calf thymus DNA ([Ru] ¼ 100 mM, [DNA] ¼ 1.0 mM)
(top). Circular dichroism spectra of the D form (full lines) and the L
form (dotted lines) of complex 1 (bottom).
Fig. 6 Schematic diagram of energies of complexes 1, 2, 3 and related
1MLCT transition.
1260
New J. Chem., 2003, 27, 1255–1263

View Article Online
the L enantiomer with the right-handed calf thymus DNA
helix, due to the appropriate steric matching. Our results
may give evidence to support this proposal.
such a purpose. The energies of the HOMO and 6 occupied
MOs lying near the HOMO for the CG/CG stacking cal-
culated by the authors were ꢂ1.27, ꢂ1.33, ꢂ1.69, ꢂ1.79,
ꢂ1.98, ꢂ2.06 and ꢂ2.08 eV, respectively, and the components
of HOMO and NHOMO were mainly distributed on base-
pairs. We have also performed some calculations for the Ru(II)
complexes using the DFT method in this field, and the calcu-
lated energies of their LUMOs and 14 unoccupied MOs lying
near the LUMOs are within in the range of ꢂ7.42 to ꢂ4.71 eV
for [Ru(bpy)2PMIP]2+, ꢂ7.26 to ꢂ4.67 eV for [Ru(phen)2P-
MIP]2+ and ꢂ7.1 to ꢂ4.66 eV for [Ru(dmp)2PMIP]2+, and
the components of these MOs distribute predominantly on
ligands, in particular, on main-ligands or called intercalative
ligands. It can be clearly seen that the energies of the HOMO
and the occupied orbitals lying near the HOMO of the DNA
model are all much higher than those of the LUMO and 14
unoccupied MOs lying near the LUMO of every one of the
three complex cations. We believe such a trend in the relative
energies to be kept even though the orbital energies for the
CG/CG stacking were replaced by those for the other base-
pair stacking, because the attraction of metal complex cations
with high positive charges for electrons in the frontier MOs
is much stronger than that of various DNA. Lower LUMO
energy of the complex in the intercalation mode should be
advantageous to accept the electron from base-pairs of
DNA, so that the LUMO energies of the complexes should
be an important factor (but not the only one) correlating to
their DNA-binding constants. The experimental results show
that the trend in DNA-binding constants (Kb) is Kb(2) >
Kb(1) > Kb(3). It can be explained as follows: Kb(3) is the
smallest because of eL(3) is the greatest. As for Kb(2) > Kb(1),
although eL(2) > eL(1) to certain extent, the hydrophobicity of
the phen ligand is greater than that of the bpy ligand, and
thus it makes the area of the main ligand intercalating into
DNA increase for the corresponding complex. As a result,
Kb(2) > Kb(1) when synthetically considering both the LUMO
energy factor and the hydrophobicity effect of the ancillary
ligand.
From Fig. 9, we can also clearly see that for the complexes 1
and 2, the max (l) singlet metal to ligand charge-transfer
(1MLCT) bands should correspond to the electron transitions
from their HOMOs-2 to LUMOs respectively, whereas that for
complex 3 should correspond to the electron transition from its
NHOMO to LUMO according to the molecular orbital com-
ponents of the complexes.
Since in absorption and emission spectra of the Ru(II) com-
plexes, the corresponding energy change between presence and
absence of DNA is smaller, and no special pattern changes in
the absorption or emission spectra of Ru(II) complexes in the
presence of DNA have been detected, except increases in lumi-
nescence intensity for emission spectra and hypochromism for
absorption spectra, we consider that there is not a greater effect
on the corresponding orbitals for the complexes binding to
DNA (relative to free complexes). This further suggests that
the interaction between the series of complexes and DNA
should be a weak one between molecules, so that the properties
of 1MLCT electron transition bands in these Ru(II) complexes
binding to DNA should be unchanged, in accordance with
experiment.
The CD signals of complex 3 are much stronger than
those of complexes 1 and 2, and the CD signals of complex
2 are in opposition to those of complex 3. The stronger CD
signals of complex 3 suggest the larger difference between
interactions of its two isomers with DNA. It is likely that,
for the weakly DNA-binding complex 3, since the steric hin-
drances of the methyl groups substituted at dmp ligand have
a very significant effect on one of the isomers binding to
DNA, this leads to a special preference for the other isomer
binding to DNA. On the basis of the previous reported
document,33 we speculate that, for the complexes that have
analogous structures, the same type of isomers of those
complexes should have analogous CD signals. The CD spectra
of complex 2 and complex 3 suggest that, as compared with
complex 2, the different type of isomer of complex 3 strongly
interacts with DNA enantioselectively. This should also be
caused by the steric hindrances of the methyl groups in
dmp ligand.
Although complexes 2 and 3 have not been resolved into
their pure enantiomers, and we can not determine which enan-
tiomer binds to DNA enantioselectivly, for each of the three
complexes, it is certain that complex 3 is a much better candi-
date as an enantioselective binder to CT DNA than complexes
1 and 2, and that the ancillary ligands have a significant influ-
ence on the interaction of the enantioselectivity of polypyridyl
Ru(II) complexes with DNA.
Theoretical explanation of trends in DNA-binding and spectral
properties of complexes
As well-known, there are p–p interactions in the DNA-bind-
ing of these complexes in intercalation mode. But up to
now, a whole DNA molecule, in particular, the supermole-
cular system formed from DNA and a metal complex is
too large in size to be calculated, therefore, it is very impor-
tant and necessary to theoretically estimate the interaction
between both from their individual electronic structural
characteristics.
The above-mentioned trends in DNA-binding and spectral
properties can be explained by our theoretical computations
with DFT method and the frontier molecular orbital theory.34
Some computed frontier molecular orbital energies, the sche-
matic diagram of the max (l) 1MLCT transition, and the
stereographs of the related frontier molecular orbitals of the
three complexes were given in Table 1, Fig. 6 and Fig. 9 respec-
tively. According to the frontier molecular orbital theory, a
higher HOMO energy of one reactant molecule and a lower
LUMO energy of the other are advantageous to the reaction
between the two molecules, because electrons more easily
transfer from the HOMO of one reactant to the LUMO of
the other in the orbital interaction. A simple calculation model
and computed results by the DFT method for stacked DNA
base-pairs with backbones have been reported by Noriguki
Kruita etc.6e It should be a better simplified approximation
model for DNA, and thus should be useful and feasible for
Table 1 Some frontier molecular orbital energies (eI/atomic unit) of the complexes (1 atomic unit ¼ 27.21 eV)
Complex
Occa
Occ
Occ
HOMO
LUMO
Virb
Vir
DeL–H
DeL–NH
DeL–(H-2)
1
2
3
ꢂ0.4000
ꢂ0.3960
ꢂ0.3888
ꢂ0.3953
ꢂ0.3919
ꢂ0.3880
ꢂ0.3890
ꢂ0.3874
ꢂ0.3852
ꢂ0.3643
ꢂ0.3624
ꢂ0.3628
ꢂ0.2726
ꢂ0.2667
ꢂ0.2613
ꢂ0.2693
ꢂ0.2632
ꢂ0.2602
ꢂ0.2627
ꢂ0.2600
ꢂ0.2563
0.0917
0.0957
0.1015
0.1146
0.1207
0.1239
0.1227
0.1252
—
a
b
Occ: Occupied molecular orbital; HOMO (or H): The highest Occ. Vir: Virtual molecular orbital; LUMO (or L): The lowest Vir.
New J. Chem., 2003, 27, 1255–1263
1261

View Article Online
Fig. 9 Some related frontier molecular orbital stereographs of complexes 1, 2 and 3.
The experimental results suggest that the ancillary ligands of
polypyridyl Ru(II) complexes have significant effects on the
spectral properties and the DNA-binding behavior of the com-
plexes. The increase of ancillary ligand size increases its hydro-
phobicity, facilitating interaction with DNA. However, if
the ancillary ligand is too large, its steric hindrance would inter-
fere with the depth of intercalation into the base pair of DNA.
The theoretical calculation results applying DFT method can
also be used to further discuss the trend in the binding strength
or binding constants (Kb) of the complexes to DNA as well as
to roughly predict some of their spectral properties.
Conclusions
Spectroscopic studies, viscosity measurements together with
equilibrium dialysis and circular dichroism spectroscopy show
that complex 1 and complex 2 can bind to CT DNA through
intercalation, complex 3 binds to CT DNA via a partial interca-
lative mode, all the three complexes can interact with CT DNA
enantioselectively in a way, complex 3 is a much better candi-
date as an enantioselective binder to CT DNA than complexes
1 and 2, and that the D enantiomer of complex 1 is slightly pre-
dominant over the L enantiomer for binding to CT DNA.
1262
New J. Chem., 2003, 27, 1255–1263

View Article Online
A, 1998, 102, 3382; (k) S. Tobisch, T. Nowak and H. Bo¨gel,
J. Organomet. Chem., 2001, 619, 24.
Acknowledgements
9
E. Amouyal, A. Homsi, J. C. Chambron and J. P. Sauvage,
J. Chem. Soc., Dalton Trans., 1990, 14, 1841.
We are grateful for the support of the National Natural
Science Foundation of China, the Research Fund of Royal
Society of Chemistry U.K., the Natural Science Foundation
of Guangdong Province, the State Key Laboratory of Coordi-
nation Chemistry in Nanjing University and the State Key
Laboratory of Bio-organic, Natural Products Chemistry in
Shanghai Institute of Organic Chemistry.
10 B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978,
17, 3334.
11 J. P. Collin and J. P. Sauvage, Inorg. Chem., 1986, 25, 135.
12 Y. Xiong, X. H. Zou, J. H. Wu, L. N. Ji, R. H. Li, J. Y. Zhou and
K. B. Yu, Transition Met. Chem., 1999, 24, 263.
13 J. Marmur, J. Mol. Biol., 1961, 3, 208.
14 M. E. Reichmann, S. A. Rice, C. A. Thomas and P. Doty, J. Am.
Chem. Soc., 1954, 76, 3047.
15 J. B. Chaires, N. Dattagupta and D. M. Crothers, Biochemistry,
1982, 21, 3933.
References
1
2
L. N. Ji, X. H. Zou and J. G. Liu, Coord. Chem. Rev., 2001, 216–
217, 513.
16 G. Cohen and H. Eisenberg, Biopolymers, 1969, 8, 45.
17 A. Wolfe, G. H. Shimer and T. Meehan, Biochemistry, 1987, 26,
6392.
18 J. R. Lakowicz and G. Webber, Biochemistry, 1973, 12, 4161.
19 J. K. Barton, J. J. Dannenberg and A. L. Raphael, J. Am. Chem.
Soc., 1984, 106, 2172.
B. A. Jackson, V. Y. Alekseyev and J. K. Barton, Biochemistry,
1999, 38, 4655; P. J. Dandlier, R. E. Holmlin and J. K. Barton,
Science, 1997, 274, 1465; D. B. Hall, R. E. Holmlin and J. K.
Barton, Nature, 1996, 382, 731; C. J. Murphy and J. K. Barton,
Methods Enzymol., 1993, 226, 576; D. S. Sigman, A. Mazumder
and D. M. Perrin, Chem. Rev., 1993, 93, 2295; F. M. O’Reilly
20 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270; P. J.
Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299.
21 A. Juris, V. Balzani, F. Barigelletti,Campagna, S. Campagna,
P. Belser and A. V. Zelewsky, Coord. Chem. Rev., 1988, 84, 85.
22 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery,
Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam,
A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala,
Q. Cui, K. Morokuma, N. Rega, P. Salvador, J. J. Dannenberg,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman,
J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.
Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-
Gordon, E. S. Replogle and J. A. Pople, Gaussian 98, Revision
A.11.4, Gaussian, Inc., Pittsburgh PA, 2002.
´
and J. M. Kelly, New J. Chem., 1998, 22, 215; E. Bernal-Mendez,
´
´
J. S. Sun, F. Gonzalez-Vılchez and M. Leng, New J. Chem., 1998,
22, 1479; L. Mishra, A. K. Yadaw, S. Srivastava and A. B. Patel,
New J. Chem., 2000, 24, 505.
3
4
G. Yang, J. Z. Wu, L. Wang, L. N. Ji and X. Tian, J. Inorg.
Biochem., 1997, 66, 141.
(a) J. G. Liu, B. H. Ye, H. Li, L. N. Ji, R. H. Li and J. Y. Zhou,
J. Inorg. Biochem., 1999, 73, 117; (b) J. Z. Wu, B. H. Ye, L. Wang,
L. N. Ji, J. Y. Zhou, R. H. Li and Z. Y. Zhou, J. Chem. Soc., Dal-
ton Trans., 1997, 21, 1395; (c) Y. Xiong, X. F. He, X. H. Zou, J. Z.
Wu, X. M. Chen, L. N. Ji, R. H. Li, J. Y. Zhou and K. B. Yu,
J. Chem. Soc., Dalton Trans., 1999, 23, 19.
Y. Xiong and L. N. Ji, Coord. Chem. Rev., 1999, 189, 1.
(a) N. V. Ivanova, O. V. Sizov, A. B. Nikolskii and A. I. Panin,
J. Struct. Chem., 1999, 40, 620; (b) O. V. Sizova, A. I. Panin,
N. V. Ivanova and V. I. Batanovskii, J. Struct. Chem., 1997, 38,
366; (c) M. K. Nazeeruddin, S. M. Zakeeruddin, R. Humphry-
Baker, S. I. Gorelsky, A. B. P. Lever and M. Gratzel, Coord.
Chem. Rev., 2000, 208, 213; (d ) M. Ziegler and A. V. Zelewsky,
Coord. Chem. Rev., 1998, 177, 257; (e) N. Kurita and K.
Kobayashi, Comput. Chem., 2000, 24, 351.
5
6
23 G. Schaftenaar, Molden v3.6 program, CMBI, Faculty of Science,
University of Nijmegen, The Netherlands, 1999.
24 S. Satyanarayana, J. C. Dabrowiak and J. B. Chaires, Biochem-
istry, 1992, 31, 9319.
7
8
P. Hohenberg and W. Kohn, Phys. Rev. B, 1964, 136, 864; A. D.
Becke, J. Chem. Phys., 1993, 98, 1372; A. Gorling, Phys. Rev.,
1996, A54, 3912; J. B. Foresman and Æ. Frisch, Exploring Chem-
istry with Electronic Structure Methods, 2nd edn., Gaussian Inc.,
Pittsburgh PA, USA, 1996.
25 S. Satyanarayana, J. C. Dabrowiak and J. B. Chaires, Biochem-
istry, 1993, 32, 2573.
26 J. G. Liu, Q. L. Zhang, L. N. Ji, Y. Y. Cao and X. F. Shi, Tran-
sition Met. Chem., 2001, 26, 733.
27 X. H. Zou, B. H. Ye, H. Li, Q. L. Zhang, H. Cao, J. G. Liu,
L. N. Ji and X. Y. Li, J. Biol. Inorg. Chem., 2001, 6, 143.
28 J. D. McGhee and P. H. Von Hippel, J. Mol. Biol., 1974, 86, 469.
29 R. J. Morgan, S. Chatterjee, A. D. Baker and T. C. Strekas, Inorg.
Chem., 1991, 30, 2687.
30 J. G. Liu, Q. L. Zhang, X. F. Shi and L. N. Ji, Inorg. Chem., 2001,
40, 5045.
31 B. C. Baguley and M. Lebret, Biochemistry, 1984, 23, 937.
32 X. H. Zou, B. H. Ye, H. Li, J. G. Liu, Y. Xiong and L. N. Ji,
J. Chem. Soc., Dalton Trans., 1999, 1423.
33 J. G. Liu, B. H. Ye, Q. L. Zhang, X. H. Zou, Q. X. Zhen, X. Tian
and L. N. Ji, J. Biol. Inorg. Chem., 2000, 5, 119.
34 K. Fukui, T. Yonezawa and H. Shingu, J. Chem. Phys., 1952, 20,
722; G. Klopman, J. Am. Chem. Soc., 1968, 90, 223; I. Fleming,
Frontier Orbitals and Organic Chemical Reactions, Wiley, New
York, 1976.
(a) S. R. Stoyaner, J. M. Villegas and D. P. Rillema, Inorg. Chem.,
2002, 41, 2941; (b) Y. Zhang, Z. J. Guo and X. Z. You, J. Am.
Chem. Soc., 2001, 123, 9378; (c) D. Reha, M. Kabelac, F.
Ryjacek, J. Sponer, J. E. Sponer, M. Elstner, S. Suhai and P.
Hobza, J. Am. Chem. Soc., 2002, 124, 3366; (d ) K. C. Zheng,
J. P. Wang, W. L. Peng, X. W. Liu and F. C. Yun, J. Phys. Chem.
A, 2001, 105, 10 899; (e) K. C. Zheng, J. P. Wang, X. W. Liu, Y.
Shen and F. C. Yun, J. Mol. Struct.: TheoChem., 2002, 577, 95;
( f ) K. C. Zheng, J. P. Wang, Y. Shen, W. L. Peng and F. C.
Yun, Dalton Trans., 2002, 111; (g) K. C. Zheng, Y. Shen, J. P.
Wang, X. W. Liu and F. C. Yun, Inorg. Chim. Acta, 2002,
335C, 100; (h) A. Pfletschinger, W. Koch and H. G. Schmalz,
Chem. Eur. J., 2001, 7, 5325; (i) V. B. Javoslav, S. Jiri and L.
Jerzy, Phys. Chem. Chem. Phys., 2001, 3, 4404; ( j) N. H.
Damrauer, B. T. Weldon and J. K. McCusker, J. Phys. Chem.
New J. Chem., 2003, 27, 1255–1263
1263