Reversible DNA condensation induced by a tetranuclear nickel(II) complex

Article abstract of DOI:10.1002/chem.201001457

DNA condensing agents play a critical role in gene therapy. A tetranuclear nickel(II) complex, [Ni^II₄(L-2H)(H₂O) ₆(CH₃CH₂OH)₂]?6NO₃ (L=3,3',5,5'-tetrakis{[(2-hydroxyet

Full text of DOI:10.1002/chem.201001457

FULL PAPER
DOI: 10.1002/chem.201001457
Reversible DNA Condensation Induced by a Tetranuclear Nickel(II)
Complex
Xindian Dong,[a] Xiaoyong Wang,*[b] Yafeng He,[a] Zhen Yu,[a] Miaoxin Lin,[a]
Changli Zhang,[a] Jing Wang,[b] Yajie Song,[b] Yangmiao Zhang,[a] Zhipeng Liu,[a]
Yizhi Li,[a] and Zijian Guo*[a]
Abstract: DNA condensing agents play
98.175(13)8. The DNA condensation
induced by the complex has been in-
vestigated by means of UV/Vis spec-
troscopy, fluorescence spectroscopy, cir-
cular dichroism spectroscopy, dynamic
light scattering, atomic force microsco-
py, gel electrophoresis assay, and zeta
potential analysis. The complex inter-
acts strongly with DNA through elec-
trostatic attraction and induces its con-
densation into globular nanoparticles
at low concentration. The release of
DNA from its compact state has been
achieved using the chelator ethylene-
diaminetetraacetic acid (EDTA) for
the first time. Other essential proper-
ties, such as DNA cleavage inactivity
and biocompatibility, have also been
examined in vitro. In general, the com-
plex satisfies the requirements of a
gene vector in all of these respects.
a critical role in gene therapy. A tetra-
nuclear nickel(II) complex, [NiII -
4
A
ACHTUNGTRENNUNG(H2O)6ACHTUNGTRENNUNG(CH3CH2OH)2]·6NO3
(L=3,3’,5,5’-tetrakis{[(2-hydroxyethyl)-
(pyridin-2-ylmethyl)amino]methyl}bi-
phenyl-4,4’-diol), has been synthesized
as a nonviral vector to induce DNA
condensation. X-ray crystallographic
data indicate that the complex crystalli-
zes in the monoclinic system with
space group P21/n, a=10.291(9), b=
24.15(2), c=13.896(11) ꢀ, and b=
Keywords: DNA
interactions gene therapy
nickel · nonviral vector
· electrostatic
·
·
Introduction
efficiency in transfection and gene expression; however,
some disadvantages, such as host immunogenicity and limit-
ed scale of production, have limited their application.[1a] In
contrast, nonviral vectors can be selected from a variety of
compounds with low cost of production and desirable flexi-
bility of application; moreover, they are relatively non-im-
munogenic and do not induce significant inflammatory re-
sponses.[5] Such vectors often possess positive charges, which
neutralize negative charges on the phosphates of the DNA
backbone and/or reorient water dipoles near DNA surfaces;
besides, multivalent cations may cause localized bending or
distortion of DNA.[3,6] These interactions decrease repul-
sions between DNA segments and induce them to compact
into a tight globular structure. Therefore, the primary func-
tion of a vector is to promote the condensation of DNA.
A wide range of agents, such as cationic lipids, polymers,
dendrimers, polypeptides, and nanoparticles, are known to
provoke the condensation of DNA.[5,7] For example, cationic
polyamines and graft copolymers can condense DNA
through cooperative electrostatic interactions and enable ge-
netic materials to penetrate the plasma cell membrane.[8]
However, cationic polymers and lipids, as well as dendrim-
ers, may induce cytotoxicity and non-biodegradability; in ad-
dition, their poor delivery efficiency and low cell specificity
The essence of gene therapy is to deliver extrinsic genes
into host cells to replace or override the defective genes
causative of genetic abnormalities or deficiencies.[1] DNA
condensation is an essential precondition for transporting a
therapeutic gene to its target position,[2] because compact
structures protect DNA from nuclease[3] and facilitate its
entry into cells by endocytosis.[4] Genes can be delivered by
viral or nonviral vectors. Viral vectors usually display high
[a] X. Dong, Y. He, Z. Yu, M. Lin, C. Zhang, Y. Zhang, Z. Liu,
Prof. Dr. Y. Li, Prof. Dr. Z. Guo
State Key Laboratory of Coordination Chemistry
School of Chemistry and Chemical Engineering
Nanjing University, Nanjing 210093 (P.R. China)
Fax : (+86)25-83224539
E-mail: zguo@nju.edu.cn
[b] Prof. Dr. X. Wang, J. Wang, Y. Song
State Key Laboratory of Pharmaceutical Biotechnology
School of Life Sciences, Nanjing University
Nanjing 210093 (P.R. China)
Fax : (+86)25-83224539
E-mail: boxwxy@nju.edu.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001457.
Chem. Eur. J. 2010, 16, 14181 – 14189
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
14181

as DNA vectors need to be overcome before clinical trials.[8a,9]
Transition-metal complexes are emerging as a new class of
nonviral vectors; some of them, including complexes of
Results and Discussion
Design and synthesis: The biphenol-based ligand L was de-
rived from biphenyl-4,4’-diol by sequential reactions with
formaldehyde, hydrogen chloride, and 2-pyridylmethyl-2-hy-
droxyethylamine (Scheme 1). The obtained ligand contains
cobaltACHTUNGTRENNUNG
(III),[10] copper(II),[11] and ruthenium(II),[12] have dis-
played a remarkable ability to induce DNA condensation.
Moreover, the inducing impact of metal complexes on DNA
condensation seems to be more effective than that of purely
organic condensing agents, in that [CoACTHNUTRGNEUNG
(NH3)6]3+ is four
times more efficient than spermidine, although they have
the same charge (3+).[13] Despite their considerable merit,
some unfavorable properties of metal complexes, such as
toxicity and DNA cleavage potentiality, have not been fully
addressed in the design strategies. On the other hand, overly
stable binding of the above condensing agents to DNA may
preclude or greatly reduce its interaction with intracellular
molecules and thereby decrease the transfection efficiency
relative to that with viral vectors.[14] Indeed, it has been
shown that the property of vector unpacking can limit the
efficiency of gene delivery and expression.[15] The transfec-
tion efficiency of stimuli-responsive delivery vectors is
higher than that of non-stimulus-responsive vectors.[16]
Therefore, conditional disassembly of the compact DNA
should be an additional design principle for nonviral vectors.
In consideration of all of the aforementioned factors, in
this study we have designed a tetranuclear nickel(II) com-
plex (1) as a novel gene vector. The rationale for this design
is that nickel is an essential element for several animal spe-
cies and many biological processes, such as lipid metabolism
and anaerobic metabolism,[17] which implies that the toxicity
of nickel(II) complexes might be tolerable for patients.
More importantly, octahedral nickel(II) complexes do not
cleave DNA under oxidative conditions,[18] suggesting that
genes carried by such vectors may not be damaged in the
delivery. Cations with a valence ꢁ3 are generally required
to induce the condensation of DNA in aqueous solution.
Therefore, we chose a ligand with multicoordination centers
to ligate nickel(II), which may form a polynuclear complex
ion with high positive charges beneficial to DNA condensa-
tion. As expected, complex 1 is able to condense DNA at
concentrations with marginal toxicity, and the condensed
DNA can be released from the vector by treatment with
ethylenediamine tetraacetic acid (EDTA) in aqueous solu-
tion.
Scheme 1. Synthetic route to the ligand. DIEA=diisopropylethylamine.
four potential coordination centers for nickel(II), and there-
fore could form a multinuclear complex with this metal ion.
Since the four alcoholic hydroxyl groups in this ligand are
not liable to lose protons in a neutral medium, the complex
ion is expected to carry high positive charge, which would
strengthen the electrostatic interaction between the complex
and DNA. Biphenyl-4,4’-diol was chosen as a linker between
the bimetallic centers to keep them at an appropriate dis-
tance for more effective interactions with the phosphodiest-
er groups in DNA. In addition, the existence of hydroxyl
groups and flexible amine chains may enhance the solubility
of the complex. Complex 1 was prepared directly by react-
ing L with NiACTHUNTRGNE(UNG NO3)2·6H2O in ethanol/water at room temper-
ature. The complex is highly water soluble, which facilitates
investigations in aqueous solution.
Crystal structure of complex 1: The structure of complex 1
has been characterized by X-
ray crystallography. A perspec-
tive view of the complex is
shown in Figure 1. The crystal-
lographic data are listed in
Table S1 in the Supporting In-
formation;
selected
bond
lengths and bond angles are
given in Table 1.
X-ray diffraction analysis
showed that complex 1 crystalli-
14182
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14181 – 14189

Reversible DNA Condensation by a Tetranuclear NiII Complex
FULL PAPER
bond lengths are in the ranges
2.046(3)–2.166(3) ꢀ
and
2.032(4)–2.075(4) ꢀ, respective-
ly, which are comparable to
those previously reported for
nickel(II) complexes with simi-
lar phenol-based ligands.[19,20]
The distance between two adja-
cent nickel(II) centers is
3.8025(24) ꢀ and the bridging
Ni-O-Ni angle is 133.27(14)8,
which are similar to those in an
analogous nickel(II) complex
with one bridge (3.775 ꢀ,
131.88),[21] but larger than those
in a similar complex with two
bridges (3.35 ꢀ, 1148).[20a] The
distances between the non-
Figure 1. Crystal structure of the cationic core of complex 1. Counterions and hydrogen atoms are omitted for
clarity.
neighboring nickel(II) centers (Ni1···Ni1A, Ni1···Ni2A) are
about 12 ꢀ, which happen to be approximately twice the
distance between two consecutive phosphate groups (6–7 ꢀ)
in a B-DNA strand.[22] This coincidence suggests that when
complex 1 interacts with phosphodiester groups in a DNA
strand, the two non-neighboring nickel(II) centers may in-
teract synchronously with two phosphoryl oxygen atoms be-
longing to every other phosphodiester group, thus facilitat-
ing the electrostatic interaction with DNA.
Table 1. Selected bond lengths (ꢀ) and angles (8) for complex 1.[a]
ꢀ
ꢀ
Ni1 O1
2.111(3)
2.062(3)
2.132(3)
2.051(3)
2.032(4)
2.073(4)
2.080(3)
2.166(3)
2.046(3)
2.105(3)
Ni2 N3
2.039(3)
2.075(4)
3.802(3)
12.122
ꢀ
ꢀ
Ni1 O2
Ni2 N4
ꢀ
ꢀ
Ni1 O4
Ni1 Ni2
ꢀ
ꢀ
Ni1 O5
Ni1 Ni1A
ꢀ
ꢀ
Ni1 N1
Ni1 Ni2A
11.530
ꢀ
Ni1 N2
N1-Ni1-N2
O1-Ni1-N2
N3-Ni2-N4
O3-Ni2-N4
Ni1-O2-Ni2
82.23(14)
82.61(12)
83.47(15)
80.99(13)
133.27(14)
ꢀ
Ni2 O2
ꢀ
Ni2 O3
ꢀ
Ni2 O6
ꢀ
Ni2 O7
UV/Vis spectroscopic studies: The spectroscopic characteris-
tics of L and complex 1 were first investigated by UV/Vis
spectroscopy in aqueous solutions. As exhibited in Figure 2a,
the ligand shows a broad absorption band in the range 250–
320 nm, with lmax at around 265 nm, which could be assigned
to the overlapped a band of the substituted phenyl rings, as
well as the n!p* and p!p* absorption bands of the pyrid-
yl units. This broad band displays a red shift and splitting in
the spectrum of complex 1 in which lmax appears at 268 nm
and a moderate absorption band appears at 310 nm. The red
shift of the absorption band is indicative of the coordination
between L and nickel(II), particularly indicating the coordi-
nation of the phenolic oxygen (-Oꢀ) in the complex. Theo-
retically, an octahedral nickel(II) complex should give rise
to four absorption bands. Of these, three strong bands result
[a] Equivalent atoms were generated by symmetry transformation: A=
1ꢀx, ꢀy, ꢀz.
zes in the monoclinic crystal system with space group P21/n.
The crystal structure consists of a centrosymmetric [NiII -
4
A
ACHTUNGTRENNUNG(H2O)6ACHTUNGTRENNUNG
(CH3CH2OH)2]6+ cationic core with six ni-
ꢀ
through the C12 C12A bond (A: 1ꢀx, ꢀy, ꢀz), forming a
tetranuclear complex. In the half-structure, Ni1-N1-C5-C6-
N2, Ni1-N2-C7-C8-O1, Ni2-N3-C21-C22-N4, and Ni2-N4-
C23-C24-O3 form four five-membered chelate rings. The in-
terior angles of these rings are in the range 80.99(13)–
117.0(4)8, and all of the N-Ni-N and N-Ni-O angles in the
rings are smaller than 908. Two adjacent Ni atoms are
bridged by the phenoxide oxygen, and the bimetallic centers
are linked by the rigid biphenol ligand. In contrast to its di-
nuclear analogue, which has two identical NiII centers,[19] the
coordination environments, bond lengths, and angles of the
two adjacent Ni atoms are somewhat different from each
other in complex 1. Ni1 is coordinated by a pyridine nitro-
gen atom (N1), an amine nitrogen atom (N2), an appended
alcoholic oxygen atom (O1), a bridging phenoxide oxygen
atom (O2), a water oxygen atom (O4), and an alcoholic
oxygen atom (O5), constituting a slightly distorted octahe-
dral geometry. In contrast, although the coordination setting
of Ni2 is similar to that of Ni1, ethanol does not participate
in the coordination of Ni2; as an alternative, another water
3
from spin-allowed transitions from the ground state A2g to
3
3
3
the excited triplet states T2g, T1g(F), and T1g(P), and a weak
band arises from the spin-forbidden transition to the singlet
1
3
state Eg, which is often observed as a shoulder of the A2g!
3T2g transition band at about 800 nm.[23] In Figure 2b, two
broad absorption bands are observed in the regions 550–
700 nm and 950–990 nm, respectively. The former may be at-
3
3
tributed to the A2g! T1g(F) transition, whereas the latter
3
3
3
may be assigned to the A2g! T2g transition. The A2g!
3T1g(P) transition that would be expected to appear at around
360 nm is not observed in the spectrum, which may be due
to superposition of the band by ligand-to-metal charge-
transfer transitions. The characteristic bands and absorption
intensities of complex 1 remained unchanged for more than
ꢀ
ꢀ
oxygen atom (O6) coordinates to Ni2. The Ni O and Ni N
Chem. Eur. J. 2010, 16, 14181 – 14189
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14183

Z. Guo, X. Wang et al.
the variations in the UV/Vis spectrum of calf thymus (CT)-
DNA in response to titration with complex 1. At low com-
plex to DNA molar ratios (0.01–0.05), only a moderate de-
crease in the absorbance of CT-DNA at 260 nm was ob-
served. A more significant decrease was observed when the
molar ratio of complex to DNA reached 0.1, suggesting that
a considerable amount of free DNA had been condensed by
the complex. When the molar ratio was increased to 0.2,
DNA deposits were observed after centrifugation of the re-
action mixture.
Ethidium bromide (EB) is a cationic dye that displays a
significant increase in fluorescence when it intercalates into
DNA. The enhancement of fluorescence upon intercalation
of EB between base pairs of the double helix indicates the
presence of native DNA in solution, and also provides the
basis for DNA staining in gel electrophoresis (see the Ex-
perimental Section). Competitive displacement of EB from
DNA into solution by positively charged species can result
in quenching of the fluorescence.[24] In this way, the interac-
tion between DNA and complex 1 was investigated by
means of an EB displacement assay. As shown in Figure 4,
Figure 2. UV/Vis spectra of the ligand (35 mm, 5% ethanol) and complex
1 (35 mm) in aqueous solution (a), and enlargements of the absorption
bands in the range 500–1000 nm (b); solid line (ligand), dotted line (com-
plex 1).
one month, suggesting that the complex is quite stable in
aqueous solution.
Figure 4. Fluorescence changes of the DNA–EB system ([DNA]=[EB]=
5ꢂ10ꢀ5 m) upon addition of aliquots of complex 1 (3ꢂ10ꢀ4 m) in buffer
(5 mm Tris-HCl/50 mm NaCl, pH 7.3). The inset shows the fluorescence
Interaction between DNA and complex 1: The interaction
of DNA with complex 1 was studied by UV/Vis, fluores-
cence, and circular dichroism spectroscopies. Figure 3 shows
variations (%) with the molar ratio of [complex 1]/ACTHUNRTGNEUNG[DNA–EB] at 599 nm.
upon addition of complex 1 to a DNA–EB solution, the
fluorescence intensity decreased dramatically. A great mass
of EB was expelled by the complex, and the fluorescence in-
tensity stopped decreasing when the molar ratio of complex
1 to DNA reached 0.1. At this point, the fluorescence had
been quenched to about 15% of its original intensity (inset
in Figure 4). In the present case, the displacement of EB by
the multivalent complex cations results from charge neutral-
ization on the DNA backbone, which is accompanied by the
collapse of linear DNA to a compact configuration without
a dramatic change in the local structure.[25] This is a highly
cooperative process. Moreover, a decrease in DNA flexibili-
ty because of the condensation can shift the binding equilib-
rium of EB towards the solution phase.[26] The results indi-
cate that complex 1 interacts strongly with DNA and exerts
a remarkable impact on its configuration.
Figure 3. UV/Vis spectra of CT-DNA (0.11 mm) with complex 1 in buffer
(5 mm Tris-HCl/50 mm NaCl, pH 7.3) at different [complex]/ACTHNUGRTNEUNG[DNA] molar
ratios.
14184
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14181 – 14189

Reversible DNA Condensation by a Tetranuclear NiII Complex
FULL PAPER
Circular dichroism (CD) spectroscopy was further used to
investigate the global changes in DNA conformation in-
duced by complex 1. As shown in Figure 5, the CD spectrum
Figure 6. Agarose gel electrophoresis patterns of pBR322 plasmid DNA
(200 ng) after incubation with different concentrations of complex 1 in
buffer (50 mm Tris-HCl/50 mm NaCl, pH 7.3) at 378C for 2 h. Lane 1,
DNA control; lanes 2–13, DNA+complex 1 (5, 10, 15, 17.5, 20, 25, 27.5,
30, 40, 50, 60, and 70 mm, respectively).
centration of 1 reached 60 mm. The retardation suggests that
the negative charges on the DNA backbone were neutral-
ized by the positive complex ion, and that large-sized DNA
particles were formed due to a reduction of electrostatic re-
pulsions, which would impede the migration of DNA in the
gel. The results demonstrate that low concentrations of com-
plex 1 can induce the condensation of supercoiled DNA.
Figure 5. Circular dichroism spectra of CT-DNA (0.1 mm) in the presence
of complex 1 at different [complex]/ACTHNUTRGNEUNG[DNA] molar ratios.
Atomic force microscopy: DNA deposits on an unmodified
mica surface formed from an aqueous solution (50 mm Tris-
HCl, 100 mm NaCl, pH 7.3) of free DNA or DNA plus com-
plex 1 were observed by AFM. Figure 7 shows typical AFM
of CT-DNA features a positive band at 275 nm due to base
stacking and a negative band at 245 nm due to the helicity,
which are characteristic of B-DNA.[27] The ensuing spectra
only displayed a moderate change in ellipticity for both the
positive and negative bands when a small amount of com-
plex 1 ([complex 1]/ACHTUNGTRENNUNG[DNA]=0.01–0.05) was added to DNA,
suggesting that the interaction between complex 1 and CT-
DNA is electrostatic in nature.[28] The CD spectra displayed
an abrupt decrease when this ratio was increased to 0.1, in-
dicating that a large amount of DNA had been induced to
adopt its condensed form by complex 1. When the ratio was
further increased to 0.4, both the positive and negative
bands of CT-DNA almost disappeared, indicating the disap-
pearance of the relaxed DNA and the formation of DNA
condensate. The influence of the ligand on the conformation
of DNA was also evaluated by CD spectroscopy. As Fig-
ure S1 (in the Supporting Information) shows, in contrast to
Figure 7. AFM images of free supercoiled pBR322 DNA (a) and con-
densed DNA induced by complex 1 at 5 mm, 2 h (b), 8 mm, 3.5 h (c), and
10 mm, 4 h (d) at 378C on the surface of mica. Scale bars: a) and
b) 200 nm; c) 500 nm, c) 1 mm.
the case of complex 1, at a [ligand]/ACTHNUTRGNEUNG[DNA] molar ratio of
0.4, the ellipticity of both the positive and negative bands re-
mained nearly unchanged compared with that of free DNA.
The results demonstrate that although complex 1 efficiently
interacts with DNA and induces its condensation at relative-
ly low [complex]/
exerts any influence on the conformation of DNA.
[DNA] molar ratios, the ligand hardly
images reflecting the morphological changes of supercoiled
pBR322 DNA under different conditions. In the absence of
complex 1, DNA existed as relaxed circles with slight twist-
ing of the strands (Figure 7a), which is characteristic of un-
condensed DNA. In the presence of 5 mm complex 1, parti-
cles of condensed DNA began to appear, coexisting with
free circular DNA (Figure 7b). When the concentration of
complex 1 reached 8 mm, the largest size of the particles in-
creased to around 500 nm, while some amount of DNA was
still in free form (Figure 7c). The supercoiled DNA was en-
tirely induced into nanoparticles when the concentration of
complex 1 reached 10 mm (Figure 7d). The condensed DNA
particles were nearly globular and their mean size was about
Gel electrophoresis assay: The DNA condensation induced
by complex 1 at different concentrations was tested by an
agarose gel electrophoresis assay, which reflects the size/
charge ratio of the relevant DNA particles.[29] Retardation
of the DNA bands in the gel is an indication of a decrease
in the negative charge on the plasmid DNA and the forma-
tion of large-sized DNA particles.[30] As shown in Figure 6,
the delay of DNA migration became increasingly evident as
the concentration of complex 1 was increased from 5 to
50 mm. Most DNA was confined to the gel slot when the con-
Chem. Eur. J. 2010, 16, 14181 – 14189
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14185

Z. Guo, X. Wang et al.
500 nm. The AFM images clearly indicate that the conden-
sation of DNA can be induced by concentrations of complex
1 of less than 10 mm.
Dynamic light scattering: The size of the condensed DNA
particles in solution induced by complex 1 was further inves-
tigated using dynamic light scattering (DLS). This measure-
ment does not affect the electric field relevant to the DNA
condensation. The distribution of hydrodynamic diameters
of the DNA particles measured at a scattering angle of 908
is shown in Figure 8. The apparent mean diameter of the
Figure 9. CD spectra of preformed compact CT-DNA (0.1 mm DNA+
0.02 mm complex 1) in the presence of EDTA at different [EDTA]/ACHTUNGTRENNUNG[com-
plex] molar ratios after incubation at 378C and pH 7.3 for 1 h. The
dashed line shows the spectrum of CT-DNA (0.1 mm) in loose form.
Figure 8. Hydrodynamic diameter distribution of the DNA particles de-
termined by DLS at a scattering angle of 908 and 258C. The concentra-
tions of DNA and complex 1 were 20 mgmLꢀ1 (3ꢂ10ꢀ5 mbp) and 10 mm,
respectively.
Figure 10. Agarose gel electrophoresis patterns of pBR322 plasmid DNA
(200 ng) after reaction with complex 1 in buffer (50 mm Tris-HCl/50 mm
NaCl, pH 7.3) at 378C for 2 h and addition of EDTA. Lane 1, DNA con-
trol; lanes 2–13, DNA+complex 1 (1, 3, 6, 10, 15, 20, 25, 30, 60, 120, 180,
and 270 mm, respectively)+EDTA (3 mm).
condensed DNA particles was about 1000 nm, which is
larger than that determined by AFM in the dry state. We
chose 10 mm of complex 1 to carry out the experiment be-
cause at this concentration DNA was completely condensed.
Consistent with the AFM observations, the DLS results also
suggested the existence of condensed DNA particles in solu-
tion.
presence of EDTA, and hence further confirm that the
DNA condensation induced by complex 1 is reversible
under certain conditions.
Disassembly of the compact DNA: Intracellular release of
DNA from its compact state is essential for efficient gene
expression.[7b,31] Various methods, such as increasing pH,[32]
reducing disulfide linkages,[16] and using light-responsive or
redox-active surfactants,[26,33] have been used to trigger
DNA dissociation from the vector. Here, we have used
EDTA for the first time to disassemble the compact DNA
preformed with complex 1. The release process was first ex-
amined by CD spectroscopy. As shown in Figure 9, after in-
cubation with EDTA, both the positive and negative bands
of the CT-DNA spectrum displayed an increase in ellipticity,
suggesting that EDTA can unpack the compact DNA in-
duced by complex 1, and that the amount of relaxed DNA
increases with the incremental addition of EDTA.
An agarose gel electrophoresis assay was carried out to
verify the CD results. Figure 10 shows the electrophoresis
patterns of plasmid DNA after reaction with different con-
centrations of complex 1 and subsequent incubation with
EDTA. In contrast to the observations in Figure 6, neither
the retardation of the DNA bands in the agarose gel nor the
confinement of DNA in the gel slot was observed in this ex-
periment, even when the concentration of complex 1 was in-
creased to 270 mm. The results demonstrate that the compact
DNA was completely restored to the relaxed form in the
The release of DNA from the compact state by EDTA
may be attributed to two factors. On the one hand, anionic
EDTA may react with the condensed DNA by neutralizing
the positive charge on complex 1, which would weaken the
attraction between 1 and DNA, and enhance the repulsion
between DNA segments, resulting in the release of DNA
from the vector. This hypothesis was borne out by a zeta-po-
tential assay, as described in the following section. On the
other hand, as a strong chelating agent, EDTA may remove
nickel(II) from complex 1 through an ion-exchange reaction,
which would decompose complex 1 and eliminate the elec-
trostatic attraction between 1 and DNA, leading to the dis-
sociation of DNA from the carrier. This assumption has
been verified by ESI-MS (Figure S2 in the Supporting Infor-
mation), with a peak at m/z 394.08 being attributable to
[Na2·EDTAꢀH+Ni]+ (C10H13N2O8Na2Ni, calcd 393.88), sug-
gesting that EDTA could grab a nickel(II) ion from complex
1 mingled with compact DNA. These CD and electrophore-
sis results provide some detailed information about the
mechanism of DNA condensation mediated by complex 1.
Zeta potential: Zeta (z) potential gives an indication of the
surface charge on a particulate species, which can profound-
14186
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14181 – 14189

Reversible DNA Condensation by a Tetranuclear NiII Complex
FULL PAPER
ly affect the aggregation and distribution of the particles.[34]
Table 2 presents the zeta potentials of DNA particles con-
densed at different concentrations of complex 1. As the
tested in vitro against the human hepatocarcinoma cell line
HepG2, the human non-small-cell lung cancer cell line
A549, the human renal epithelial cell line 293T, the rat
pheochromocytoma cell line PC12, and the human breast
cancer cell line MCF-7. As presented in Figure 12, complex
Table 2. Zeta potentials of the condensed DNA particles recorded in the
presence of complex 1 at different concentrations before and after the
addition of EDTA.
Complex 1 [mm]
Zeta potential [mV]
5
8
12.6
15.8
10
21.0
10+EDTA (100 mm)
ꢀ29.2
ratio of complex to DNA was increased, the positive surface
charges of the DNA particles also increased; thus, the z po-
tential increased with increasing concentration of complex 1.
A large surface charge can stabilize the DNA particles and
prevent them from aggregating by electrostatic repulsion.
On the other hand, an increase in the amount of positive
complex ion can enhance its interaction with the negatively
charged DNA and facilitate the DNA condensation. In the
presence of EDTA, the z potential was seen to decrease sig-
nificantly, suggesting that the positive surface charges were
neutralized by the anionic EDTA. Further, the results may
imply that the DNA condensation was reversed and that the
negatively charged DNA was dissociated from the compact
particles.
Figure 12. Cytotoxicity of complex 1 towards HepG2, A549, 293T, PC12,
and MCF-7 cell lines at concentrations of 1 and 100 mm, respectively,
after incubation at 378C for 72 h.
1 did not exhibit obvious cytotoxicity towards the HepG2,
293T, PC12, or MCF-7 cell lines at 1 mm, and only showed a
weak toxicity towards A549 cells. Even when the concentra-
tion of complex 1 was increased to 100 mm, the viabilities of
each of the cell lines were still above 65%. The results indi-
cate that the cytotoxicity of complex 1 is suitably low. Since
the concentration of complex 1 needed for the DNA con-
densation is far below 100 mm, the toxicity of this complex to
organisms might be negligible in biological settings.
Cleavage property: The DNA cleavage property of complex
1 was studied using supercoiled pUC19 plasmid DNA by gel
electrophoresis in a buffer (50 mm Tris-HCl/50 mm NaCl,
pH 7.3) under physiologically relevant conditions. Figure 11
Conclusion
A novel tetranuclear nickel(II) complex has been designed
as a potential nonviral vector for gene therapy. Low concen-
trations of the complex are effective in condensing DNA
into nanoparticles through electrostatic interactions. The
high positive charge carried by the complex ion is the major
factor determining the efficiency of DNA condensation.
EDTA and CD spectroscopy have been used for the first
time to examine the disassembly of compact DNA in vitro.
The condensed DNA can be unpacked by EDTA either
through charge neutralization or ion-exchange reaction, or
both, which opens up a new path for the release of confined
DNA in gene delivery. Most notably, this octahedral com-
plex exhibits no DNA cleavage activity under physiological-
ly relevant conditions, thus ensuring the integrity of DNA
during the condensation. With additional excellent proper-
ties in terms of biocompatibility and water solubility, the
complex meets the various requirements of a gene vector.
Studies of the transfection efficiency and gene expression
level of the complex in cellular and animal models are ongo-
ing.
Figure 11. Agarose gel electrophoresis patterns of pBR322 plasmid DNA
(200 ng) after incubation with ascorbic acid (1 mm) and complex 1 in
buffer (50 mm Tris-HCl/50 mm NaCl, pH 7.3) at 378C for 2 h. Lane 1,
DNA control; lane 2, DNA+Vc; lanes 3–10, DNA+Vc+complex
(0.5, 1, 2.5, 5, 10, 20, 40, and 60 mm, respectively).
1
shows the electrophoresis results for DNA at increasing con-
centrations of complex 1 in the presence of ascorbic acid
(1 mm). The supercoiled DNA (Form I) remained almost
unchanged in the tested concentration range (0.5–60 mm) as
compared with the control. Therefore, this octahedral nick-
el(II) complex has no cleavage activity towards DNA.
Cytotoxicity: Low toxicity is a prerequisite for potential
gene vectors. The cytotoxic profile of complex 1 was thus
Chem. Eur. J. 2010, 16, 14181 – 14189
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14187

Z. Guo, X. Wang et al.
7.42–7.43 (d, 4H; Py), 7.76–7.79 (m, 4H; Py), 8.52–8.53 ppm (d, 4H; Py);
13C NMR (125 MHz, CD3OD, 258C): d=55.64, 56.14, 57.13, 58.05,
120.91, 123.31, 124.02, 128.87, 131.06, 137.74, 148.76, 154.48 ppm; ESI
(MS): m/z: calcd for C48H58N8O6 (L): [L+2H]+: 422.23; [L+H]+: 843.46;
[L+Na]+: 865.44; found: 422.50, 843.58, 865.25.
Experimental Section
Materials and methods: Solvents, such as ethanol, methanol, and diethyl
ether were all of analytical grade and were used as received. Calf thymus
DNA (CT-DNA), pBR322 plasmid DNA, tris(hydromethyl)aminome-
thane (Tris), and ethidium bromide (EB) were purchased from Sigma.
Ethylenediaminetetraacetic acid (EDTA) disodium salt was purchased
from Alfa Aesar. Ultrapure milli-Q water was used in all experiments.
IR spectra (from samples in KBr pellets) were recorded in the range
4000–500 cmꢀ1 on a Bruker VECTOR22 spectrometer. Elemental analy-
sis was performed on a Perkin–Elmer 240C analytical instrument. UV/
Vis absorption spectra were determined on a Shimadzu UV 3600 UV/Vis
spectrophotometer using solutions in 1.0 cm quartz cuvettes. CD spectra
were acquired on a Jasco J-810 automatic recording spectropolarimeter.
1H and 13C NMR spectra were recorded at 298 K on a Bruker DRX-500
spectrometer using standard pulse sequences. Electrospray mass spectra
were recorded using an LCQ Fleet electron-spray mass spectrometer
(Thermo Scientific). X-ray crystallographic raw data were collected on a
Bruker SMART APEX CCD diffractometer using graphite-monochrom-
atized MoKa radiation (l=0.71073 ꢀ) at 291(2) K. CCDC-707620 con-
tains the crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Fluorescence spectra were recorded
on a Perkin–Elmer LS55 luminescence spectrometer using a 1 cm cell.
Hydrodynamic diameters were determined using a BI-200SM dynamic
light-scattering apparatus (DLS, Brookhaven Instruments Co., Holtsville,
NY). Zeta potentials were measured on a Malven Nano-Z instrument.
The concentrations of CT-DNA are expressed with respect to nucleotides
and were determined by UV spectroscopy at 260 nm, taking
6600mꢀ1 cmꢀ1 as the molar absorption coefficient.
Preparation of complex 1: NiACTHNUTRGNEUN(G NO3)2·6H2O (0.14 g, 0.48 mmol) in ethanol/
H2O (10 mL, 8:2, v/v) was added to L (0.10 g, 0.12 mmol) in ethanol
(5 mL) and the mixture was stirred at room temperature for 24 h. The re-
sulting solution was filtered and the filtrate was set aside for crystalliza-
tion by slow evaporation at room temperature. Blue crystals suitable for
X-ray diffraction analysis were obtained. Elemental analysis calcd (%)
for C52H80N14Ni4O32: C 37.90, H 4.89, N 11.90; found: C 37.81, H 4.74, N
11.81; IR (KBr pellet): n˜ =3417 (brm), 2921 (m), 1608 (m), 1460 (m),
1394 (s), 1105 (m), 1026 (m), 877 (s), 767 (s), 609 cmꢀ1 (m); UV/Vis: lmax
(e)=267.5 (2.15757ꢂ104), 310 nm (1.581818ꢂ104 dm3 molꢀ1 cmꢀ1).
X-ray crystal diffraction: An empirical absorption correction was applied
to the raw crystallographic data of complex 1 using the SADABS multi-
scan program.[37] The crystal structure was solved by direct methods and
refined by the full-matrix least-squares technique using the SHELXTL
program package.[38] Anisotropic thermal parameters were assigned to all
non-hydrogen atoms. All hydrogen atoms were placed in calculated posi-
tions and refined using a riding model.
Circular dichroism study: CD spectra of CT-DNA (0.1 mm) in the pres-
ence or absence of complex 1 were collected after incubation in buffer
(5 mm Tris-HCl/50 mm NaCl, pH 7.3). All CD experiments were per-
formed at 258C and involved scanning at a speed of 10 nmminꢀ1 from
220 to 320 nm. The buffer background was subtracted.
EB displacement assay: A solution of complex 1 was dropped into the
EB–DNA ([EB]/ACTHNUTRGNEUNG
[DNA]=1:1, [DNA]=5ꢂ10ꢀ5 m) system in buffer (5 mm
Tris-HCl/50 mm NaCl, pH 7.3). The mixture was allowed to equilibrate
for 1 min. Fluorescence spectra with an excitation wavelength of 526 nm
were recorded in the emission range 530–750 nm.
Synthesis of the ligand: The intermediate 3,3’,5,5’-tetrakis(hydroxyme-
thyl)biphenyl-4,4’-diol was prepared according to a literature method
with some modifications.[35] Cold aqueous KOH solution (6%, 100 mL)
was added to biphenyl-4,4’-diol (5.0 g, 26.8 mmol) and the resulting solu-
tion was stirred at room temperature for 20 min. An aqueous solution of
formaldehyde (37%, 9 mL) was then added dropwise to the above solu-
tion over 2 h. After reacting for 15 days at room temperature, the result-
ing solution was neutralized with dilute hydrochloric acid to form a pre-
cipitate, which was collected and redissolved in ethanol. The product was
purified by column chromatography (silica gel; EtOAc/EtOH, 6:1, v/v)
and was obtained as a white powder in 30% yield. 1H NMR (500 MHz,
[D6]DMSO, 258C): d=8.55 (s, 2H; OH), 7.40 (s, 4H; Ar), 4.60 ppm (s,
8H; CH2); 13C NMR (125 MHz, [D6]DMSO, 258C): d=59.84, 123.86,
129.01, 131.96, 151.01 ppm; ESI (MS): m/z: calcd for C16H18O6 (M):
[MꢀH]ꢀ: 305.11; found: 305.17. HCl gas, after drying by passage through
concentrated H2SO4, was passed through a suspension of 3,3’,5,5’-tetra-
kis(hydroxymethyl)biphenyl-4,4’-diol (2.50 g, 8.17 mmol) in ethyl acetate
(50 mL). When the reaction solution became clear and yellow, it was
stirred under a flow of dry HCl gas at room temperature for 2 h. After
removing the solvent under reduced pressure, the intermediate 3,3’,5,5’-
tetrakis(chloromethyl)biphenyl-4,4’-diol was obtained in 55% yield by re-
crystallizing the crude product from hexane and ethyl acetate. 1H NMR
(500 MHz, (CD3)2CO, 258C): d=4.88 (s, 8H; CH2), 7.69 ppm (s, 4H;
Ph); 13C NMR (125 MHz, (CD3)2CO, 258C): d=42.64, 126.88, 130.23,
132.78, 153.53 ppm. A solution of 3,3’,5,5’-tetrakis(chloromethyl)biphen-
yl-4,4’-diol (0.62 g, 1.63 mmol) and N,N-diisopropylethylamine (DIEA) in
EtOAc/CHCl3 (10 mL, 1:1, v/v) was cooled to 08C, whereupon 2-pyridyl-
methyl-2-hydroxyethylamine was added dropwise.[36] The resulting mix-
ture was stirred at room temperature, and the reaction was monitored by
thin-layer chromatography until it had reached completion. The solvent
was removed from the mixture under reduced pressure and the residue
was diluted with CHCl3. The solution obtained was washed several times
with brine, dried with anhydrous Na2SO4, and concentrated. The crude
product was purified by column chromatography (CHCl3/CH3OH, 8:1, v/
v) to afford yellow oily 3,3’,5,5’-tetrakis{[(2-hydroxyethyl)(pyridin-2-ylme-
thyl)amino]methyl}biphenyl-4,4’-diol (L) in a yield of 20%. 1H NMR
(500 MHz, CD3OD, 258C): d=3.03 (s, 8H; CH2), 3.80 (s, 8H; CH2), 4.14
(s, 8H; CH2), 4.20 (s, 8H; CH2), 7.31–7.33 (m, 4H; Py), 7.38 (s, 4H; Ph),
Agarose gel electrophoresis: Supercoiled pBR322 DNA (200 ng) was
treated with gradient concentrations of complex 1 in buffer (50 mm Tris-
HCl/50 mm NaCl, pH 7.3) until the total added volume reached 10 mL.
The mixtures were incubated at 378C for 2 h, and then loading buffer
(36% glycerol, 0.05% xylene cyanol FF, 0.05% bromophenol blue), with-
out or with EDTA (3 mm), was added to the solutions. The resulting solu-
tions were loaded onto agarose gel (1%) and subjected to electrophore-
sis in a buffer (40 mm Tris/acetate). DNA bands were stained with EB, vi-
sualized under UV light, and photographed. The DNA cleavage experi-
ment was carried out according to a similar method, except that ascorbic
acid (1 mm) was added to the reaction system as a cleaving initiator.
Atomic force microscopy: Samples of pBR322 DNA (200 ng) were pre-
pared and incubated with gradient concentrations of complex 1 in buffer
(50 mm Tris-HCl/50 mm NaCl, pH 7.3) and the total volume was adjusted
to 10 mL. Following the digestion, each sample was applied to a mica sur-
face and the volatiles were evaporated. The remaining salts were rinsed
with ultrapure water (50 mL) and then dried with N2. AFM images were
obtained on a Nanoscope V multimode scanning probe workstation using
etched silicon nano-probes (probe model AC160TS, Digital Instruments,
Olympus) under ambient conditions. Nanoscope VII software provided
by the manufacturer of the AFM instrument was used to measure the
volume distributions (height, width, and length) of the DNA deposits on
the mica. Five random spots in the entire area were analyzed in each
case.
Dynamic light-scattering analysis: DNA solutions (20 mgmLꢀ1
) were
mixed with complex 1 (10 mm) in Tris-HCl buffer (pH 7.3), and each mix-
ture (2 mL) was transferred to a standard quartz cuvette. The samples
were allowed to stand for 2 min at room temperature before measure-
ment. Dynamic light scattering was used to determine the size distribu-
tion of the DNA particles condensed by complex 1. The scattering angle
was set at 908.
Zeta potential measurement: DNA solution (6 mgmLꢀ1) was mixed with
complex 1 (5, 8, 10 mm, respectively) or with 1 in the presence of EDTA
(100 mm). Each mixture was incubated at 378C for 60 min. Zeta potential
14188
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14181 – 14189

Reversible DNA Condensation by a Tetranuclear NiII Complex
FULL PAPER
was measured in aqueous solution at pH 7.1. The mean of triplicate
measurements was taken as the final result.
[13] J. Pelta, F. Livolant, J. L. Sikorav, J. Biol. Chem. 1996, 271, 5656–
5662.
[14] D. Lechardeur, G. L. Lukacs, Curr. Gene Ther. 2002, 2, 183–194.
[15] a) D. V. Schaffer, N. A. Fidelman, N. Dan, D. A. Lauffenburger, Bio-
technol. Bioeng. 2000, 67, 598–606; b) M. L. Read, K. H. Bremner,
Cytotoxicity assay: The cytotoxicities of complex 1 towards HepG2,
A549, 293T, PC12, and MCF-7 cell lines were tested by MTT assays
(MTT=3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide).[39]
Cells of each type were cultured in Dulbeccoꢃs modified Eagleꢃs medium
(DMEM, Gibco) supplemented with fetal bovine serum (10%, v/v),
streptomycin (0.1 mgmLꢀ1), and penicillin (100 UmLꢀ1) in a humid at-
mosphere containing 5% CO2 at 378C. The cell lines were seeded, re-
spectively, in 96-well plates at 5ꢂ103 cells per well in DMEM medium,
incubated overnight, and then treated thrice with fresh medium contain-
ing different concentrations of complex 1. The cells were incubated at
378C under 5% CO2 for 72 h and then incubated with MTT solution
(5 mgmLꢀ1, 10 mL) in PBS buffer (8 g NaCl, 0.2 g KCl, 1.44 g Na2HPO4,
0.24 g KH2PO4 per liter) for 4 h. DMSO (150 mL) was added to each well
after the medium had been removed. The absorbance of the purple for-
mazan was recorded at 490 nm using an ELISA plate reader. The viabili-
ty of the cells was calculated based on the data of three parallel tests.
´
D. Oupicky, N. K. Green, P. F. Searle, L. W. Seymour, J. Gene Med.
2003, 5, 232–245.
[16] Y. Z. You, Z. Q. Yu, M. M. Cui, C. Y. Hong, Angew. Chem. 2010,
122, 1117–1120; Angew. Chem. Int. Ed. 2010, 49, 1099–1102.
[17] a) D. G. Barceloux, Clin. Toxicol. 1999, 37, 239–258; b) M. J. Maro-
ney, Curr. Opin. Chem. Biol. 1999, 3, 188–199; c) P. E. Carrington,
P. T. Chivers, F. Al-Mjeni, R. T. Sauer, M. J. Maroney, Nat. Struct.
Biol. 2003, 10, 126–130.
[18] X. Y. Chen, S. E. Rokita, C. J. Burrows, J. Am. Chem. Soc. 1991,
113, 5884–5886.
[19] J. W. Chen, B. Liu, X. Q. Chen, Y. Z. Li, Acta Crystallogr. Sect. C
2007, 63, m42–m44.
[20] a) H. Sakiyama, T. Suzuki, K. Ono, R. Ito, Y. Watanabe, M. Yama-
saki, M. Mikuriya, Inorg. Chim. Acta 2005, 358, 1897–1903; b) A.
Asokan, B. Varghese, A. Caneschi, P. T. Manoharan, Inorg. Chem.
1998, 37, 228–232.
Acknowledgements
[21] M. A. de Brito, A. J. Bortoluzzi, A. Greatti, A. S. Ceccato, A. C.
Joussef, S. M. Drechsel, Acta Crystallogr. Sect. C 2000, 56, 1188–
1190.
[22] a) C. Sissi, P. Rossi, F. Felluga, F. Formaggio, M. Palumbo, P. Tecilla,
C. Toniolo, P. Scrimin, J. Am. Chem. Soc. 2001, 123, 3169–3170;
b) Y. An, M. L. Tong, L. N. Ji, Z. W. Mao, Dalton Trans. 2006,
2066–2071.
We are grateful to the financial support from the National Natural Sci-
ence Foundation of China (Nos.: 20631020, 90713001, 20721002, and
30870554) and the Natural Science Foundation of Jiangsu Province
(BK2008015).
[23] C. Ochs, F. E. Hahn, T. Lꢅgger, Eur. J. Inorg. Chem. 2001, 1279–
1285.
[24] V. A. Izumrudov, M. V. Zhiryakova, A. A. Goulko, Langmuir 2002,
18, 10348–10356.
[25] a) G. S. Manning, Biopolymers 1980, 19, 37–59; b) G. S. Manning,
Biopolymers 1981, 20, 1261–1270.
[26] A.-L. M. Le Ny, C. T. Lee, Jr., Biophys. Chem. 2009, 142, 76–83.
[27] W. C. Johnson, in Circular Dichroism: Principles and Applications
(Eds.: K. Nakanishi, N. Berova, R. W. Woody), VCH, Weinheim,
1994, pp. 523–540.
[1] a) D. Ferber, Science 2001, 294, 1638–1642; b) Y. Liu, Z. L. Yu,
Y. M. Zhang, D. S. Guo, Y. P. Liu, J. Am. Chem. Soc. 2008, 130,
10431–10439; c) C. E. Pedraza, D. C. Bassett, M. D. McKee, V.
Nelea, U. Gbureck, J. E. Barralet, Biomaterials 2008, 29, 3384–3392.
[2] I. W. Hamley, V. Castelletto, Angew. Chem. 2007, 119, 4524–4538;
Angew. Chem. Int. Ed. 2007, 46, 4442–4455.
[3] V. A. Bloomfield, Biopolymers 1997, 44, 269–282.
[4] A. Rꢄmy-Kristensen, J. P. Clamme, C. Vuilleumier, J. G. Kuhry, Y.
Mꢄly, Biochim. Biophys. Acta 2001, 1514, 21–32.
[28] S. P. Tang, L. Hou, Z. W. Mao, L. N. Ji, Polyhedron 2009, 28, 586–
592.
[29] C. F. Ke, S. Hou, H. Y. Zhang, Y. Liu, K. Yang, X. Z. Feng, Chem.
Commun. 2007, 3374–3376.
[30] S. J. Sun, W. G. Liu, N. Cheng, B. Q. Zhang, Z. Q. Cao, K. D. Yao,
D. C. Liang, A. J. Zuo, G. Guo, J. Y. Zhang, Bioconjugate Chem.
2005, 16, 972–980.
[5] M. A. Mintzer, E. E. Simanek, Chem. Rev. 2009, 109, 259–302.
[6] N. Korolev, N. V. Berezhnoy, K. D. Eom, J. P. Tam, L. Nordenskiçld,
Nucleic Acids Res. 2009, 37, 7137–7150.
[7] a) J. P. Behr, Acc. Chem. Res. 1993, 26, 274–278; b) V. Vijayanathan,
T. Thomas, T. J. Thomas, Biochemistry 2002, 41, 14085–14094; c) K.
Fant, E. K. Esbjçrner, P. Lincoln, B. Nordꢄn, Biochemistry 2008, 47,
1732–1740.
[31] L. de Laporte, J. C. Rea, L. D. Shea, Biomaterials 2006, 27, 947–954.
[32] N. Makita, K. Yoshikawa, Biophys. Chem. 2002, 99, 43–53.
[33] M. E. Hays, C. M. Jewell, D. M. Lynn, N. L. Abbott, Langmuir 2007,
23, 5609–5614.
[34] R. M. Xing, X. Y. Wang, C. L. Zhang, Y. M. Zhang, Q. Wang, Z.
Yang, Z. J. Guo, J. Inorg. Biochem. 2009, 103, 1039–1044.
[35] W. F. Veldhuyzen, P. Pande, S. E. Rokita, J. Am. Chem. Soc. 2003,
125, 14005–14013.
[36] J. W. Chen, X. Y. Wang, Y. G. Zhu, J. Lin, X. L. Yang, Y. Z. Li, Y.
Lu, Z. J. Guo, Inorg. Chem. 2005, 44, 3422–3430.
[37] SADABS, Version 2.03, G. M. Sheldrick, University of Gçttingen
(Germany), 2002.
[38] a) SHELXS97 and SHELXL97, G. M. Sheldrick, University of Gçt-
tingen (Germany), 1997; b) SHELXTL, Version 6.10, G. M. Shel-
drick, Bruker AXS Inc., Madison, Wisconsin, 2002.
[39] M. B. Hansen, S. E. Nielsen, K. Berg, J. Immunol. Methods 1989,
119, 203–210.
[8] a) D. Putnam, Nat. Mater. 2006, 5, 439–451; b) H. K. Nguyen, P. Le-
mieux, S. V. Vinogradov, C. L. Gebhart, N. Guꢄrin, G. Paradis, T. K.
Bronich, V. Y. Alakhov, A. V. Kabanov, Gene Ther. 2000, 7, 126–
138; c) T. H. Kim, H. L. Jiang, D. Jere, I. K. Park, M. H. Cho, J. W.
Nah, Y. J. Choi, T. Akaike, C. S. Cho, Prog. Polym. Sci. 2007, 32,
726–753; d) J. H. Yang, P. Zhang, L. Tang, P. Sun, W. G. Liu, P. Sun,
A. J. Zuo, D. C. Liang, Biomaterials 2010, 31, 144–155.
[9] S. K. Kim, K. M. Park, K. Singha, J. Kim, Y. Ahn, K. Kim, W. J.
Kim, Chem. Commun. 2010, 46, 692–694.
[10] a) A. Rodger, K. J. Sanders, M. J. Hannon, I. Meistermann, A. Par-
kinson, D. S. Vidler, I. S. Haworth, Chirality 2000, 12, 221–236; b) S.
He, P. G. Arscott, V. A. Bloomfield, Biopolymers 2000, 53, 329–341;
c) B. I. Kankia, V. Buckin, V. A. Bloomfield, Nucleic Acids Res.
2001, 29, 2795–2801; d) C. C. Conwell, I. D. Vilfan, N. V. Hud, Proc.
Natl. Acad. Sci. USA 2003, 100, 9296–9301.
[11] X. G. Meng, L. Liu, C. S. Zhou, L. Wang, C. L. Liu, Inorg. Chem.
2008, 47, 6572–6574.
[12] a) E. Musatkina, H. Amouri, M. Lamoureux, T. Chepurnykh, C.
Cordier, J. Inorg. Biochem. 2007, 101, 1086–1089; b) B. Sun, J. X.
Guan, L. Xu, B. L. Yu, L. Jiang, J. F. Kou, L. Wang, X. D. Ding, H.
Chao, L. N. Ji, Inorg. Chem. 2009, 48, 4637–4639.
Received: May 26, 2010
Published online: October 22, 2010
Chem. Eur. J. 2010, 16, 14181 – 14189
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14189