Straightforward approach to efficient oxidative DNA cleaving agents based on Cu(ii) complexes of heterosubstituted cyclens

Article abstract of DOI:10.1039/c3dt32857k

The Cu(ii) complexes of cyclen and two of its heterosubstituted analogues were shown to be efficient oxidative DNA cleavers. The reactivity strongly depends on the heteroatom inserted into the macrocycle (O > S > N). The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3dt32857k

Dalton
Transactions
View Article Online
View Journal | View Issue
COMMUNICATION
Straightforward approach to efficient oxidative DNA
cleaving agents based on Cu(II) complexes of
heterosubstituted cyclens†
Cite this: Dalton Trans., 2013, 42, 4357
Received 28th November 2012,
Accepted 30th January 2013
Jan Hormann,a Chrischani Perera,a Naina Deibel,a,b Dieter Lentz,a Biprajit Sarkara
and Nora Kulak*a
DOI: 10.1039/c3dt32857k
www.rsc.org/dalton
The Cu(II) complexes of cyclen and two of its heterosubstituted Different concepts were used to develop DNA-cleaving agents
analogues were shown to be efficient oxidative DNA cleavers. The based on the cyclen ligand. Hydrolytically cleaving complexes
reactivity strongly depends on the heteroatom inserted into the based on Zn(II) and Co(III) as well as oxidatively cleaving com-
macrocycle (O > S > N).
plexes based on Cu(II) were synthesised.4,7–9 It is well known
that Cu(II) cyclen complexes are able to cleave DNA via an oxi-
dative pathway, thus they may represent potential anticancer
drug candidates.10 Recent efforts have focused on further deri-
vatisation of the simple Cu(II) cyclen complex to increase the
affinity to DNA. This was achieved by introducing DNA-inter-
calating agents, adding positively charged groups or synthesising
multinuclear complexes.7–9 The straightforward approach of
changing the tetraaza ligand itself, however, by replacing the
donor atoms and its impact on DNA cutting activity has never
been studied to our knowledge. This is in contrast to appli-
cations in protein and RNA cleavers where the oxygen ana-
logue of cyclen is a well established scaffold.11 Our approach
would possibly lead to changes in the redox behaviour of the
copper centre, thus resulting in a change of cutting activity
depending on the introduced heteroatom.
The three ligands [12]aneN4, [12]aneN3O and [12]aneN3S
were synthesised using previously described protocols.12 The
corresponding Cu(II) complexes 1, 2 and 3 (Fig. 1) were syn-
thesised and characterised (cf. S-1, S-2†). While the crystal
structure of [Cu([12]aneN4)(NO3)](NO3) 1 has already been pub-
lished,13 the crystal structures of the nitrato complexes 2 and 3
are presented in this communication (Fig. 2). All three com-
plexes show a distorted square-pyramidal environment with
Cu–N bond lengths of nearly 2 Å. The Cu–O bond (2.231 Å) of
complex 2 and the Cu–S bond (2.328 Å) of complex 3 are,
however, elongated. Yet another difference exists in the C–N,
The key to success of organic life is the stability of deoxyribo-
nucleic acid (DNA). The half-life of its phosphodiester bond
under physiological conditions is estimated to be in between
ten to one hundred billion years.1 This stability is due to the
repulsion of potential nucleophiles by its negatively charged
backbone. Nevertheless, cleavage of DNA is of high impor-
tance: during replication and transcription of DNA the enzyme
topoisomerase cleaves one or both of the DNA strands, rever-
sing supercoiling and enabling the reading process of DNA.2
Restriction enzymes guard the genome against viral DNA by
cutting the alien DNA out.3 Manipulation of DNA by mankind
remains a big challenge, though. Up to the present developed
artificial nucleases can by far not catch up with naturally
occurring enzymes, although DNA cleaving agents promise
numerous applications in medicine and biotechnology.4
Hydrolytically and oxidatively cleaving nucleases can be distin-
guished. Whereas the first group may be used to design restric-
tion agents in biotechnology that act via hydrolysis of the
phosphodiester bond, the latter ones would be suitable as chemo-
therapeutic agents cleaving nuclear DNA by catalysing oxi-
dation reactions that break down the sugar moiety of DNA.4,5
Since their first synthesis in 1961 macrocyclic polyamines
such as 1,4,7,10-tetraazacyclododecane, the so called cyclen
ligand or [12]aneN4, have been widely used in bioinorganic
and medicinal chemistry due to their ability to form stable
complexes with transition metals and rare earth metals.6
aInstitut für Chemie und Biochemie, Freie Universität Berlin, Fabeckstr. 34/36,
D-14195 Berlin, Germany. E-mail: nora.kulak@fu-berlin.de; Fax: +49 30 838 52440;
Tel: +49 30 838 54697
bUniversität Stuttgart, Institut für Anorganische Chemie, Pfaffenwaldring 55,
70569 Stuttgart, Germay
†Electronic supplementary information (ESI) available: Experimental details and
analytical data. CCDC 911568 and 911569. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c3dt32857k
Fig. 1 Structures of complexes 1, 2 and 3.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 4357–4360 | 4357

View Article Online
Communication
Dalton Transactions
Fig. 2 ORTEP14 diagrams of the molecular structure of 2 (left) and 3 (right).
Thermal ellipsoids are drawn at a probability level of 50%. The counter ions are
omitted for clarity.‡
C–O and the C–S bond lengths. While the C–N bond length of
complex 1 (1.453 Å) and the C–O bond length of complex 2
(1.435 Å) differ only slightly, the C–S bond length of complex 3
(1.818 Å) is considerably longer. The hapticity of the nitrato
Fig. 3 Cleavage activities of complexes 1, 2 and 3 (0.04 mM) on pBR322
(0.025 μg μL−1) in Tris-HCl buffer (100 mM, pH 7.4) and ascorbic acid (0.32 mM)
at 37 °C for 2 h. Illustrated is the average of two measurements, the standard
deviation is shown as error bars.
ligand of complexes
1 and 3, where it coordinates in
η1-fashion with a typical apical Cu–O bond length15 of 2.183 and
2.160 Å, respectively, differs from complex 2, where it shows
shows a maximum in cutting activity at 0.02 mM and the
η2-like coordination with Cu–O bond lengths of 2.031 and Cu([12]aneN3S) complex 3 at 0.16 mM. After passing these con-
2.470 Å. Such an anisodentate chelation of nitrate in Cu(II)
centrations cleavage activities of 1 and 3 decrease. A similar
behaviour has been described for Cu([9]aneN3) complexes
cyclen analogues has been observed by Spiccia et al.15 before.
They have assumed that the long Cu–O bond indicates an elec- prior to our study by Burstyn et al. and was attributed to the
formation of bis-(μ-hydroxo)-bridged dimers.16 As hydroxo-
trostatic rather than a coordinative interaction.
The cleavage activities of 1, 2 and 3 toward pBR322 super-
bridged dimer formation should also be favoured by a higher
coiled plasmid DNA were studied under approximate physio- hydroxide concentration we conducted a pH dependent study,
logical conditions at 37 °C and ascorbic acid at 0.32 mM
concentration as a reducing agent. Agarose gel electrophoresis
was used to monitor the conversion of supercoiled (I) pBR322
into its nicked (II) and linear (III) form. Fig. 3 shows a com-
parison between the DNA cleavage activities of the complexes
1, 2 and 3.
proving that the DNA cleavage activity decreases with increas-
ing pH values (cf. S-3,2†).
Furthermore, the effect of reaction time and temperature
on the cleavage of pBR322 plasmid DNA was studied. Increas-
ing DNA cleavage activity was observed when both temperature
and reaction time were raised (cf. S-3,3 and S-3,4†).
After two hours incubation and at a complex concentration
of 0.04 mM gel electrophoresis shows that complex 2 was more
To elucidate the mechanism of DNA cleavage promoted by
the complexes 1, 2 and 3 the incubation was conducted under
reactive than complexes 1 and 3. Complex 2 cleaved super- an argon atmosphere and under addition of scavengers for
coiled pBR322 plasmid DNA almost completely to form II and
additionally it was able to generate about 4% of form III. Com-
reactive oxygen species (ROS) like hydroxyl radicals (t-BuOH
and DMSO),17 singlet oxygen (NaN3),18 hydrogen peroxide (cata-
pared to complex 1, the DNA cleavage efficiency of complex 3 lase)19 and superoxide (superoxide dismutase, SOD).18 Under
was two times and of complex 2 even four times higher.
Further investigations of the cleaving ability were carried out
anaerobic conditions cleavage was reduced, indicating that
oxygen plays a role in the cleavage mechanism (Fig. 5A). In the
performing a series of optimisation experiments including case of 2 even traces of oxygen are enough to promote DNA
variations of concentration, pH value and reaction time. The
concentration dependent investigation shows that complex 2
cleavage indicating a high catalytic activity of 2. As shown
exemplarily for complex 3 in part B of Fig. 5 DMSO and cata-
exhibited an approximately linear relationship between lase show a strong and NaN3 a weak inhibition effect on DNA
complex concentration and DNA cleavage activity (Fig. 4). At
0.08 mM concentration form I DNA was cleaved into the
cleavage. These results prove the mechanism of DNA cleavage
to be oxidative involving hydroxyl radicals, hydrogen peroxide
nicked form almost completely. As opposed to that complexes and to some extent singlet oxygen as ROS.
1 and 3 did not cleave plasmid DNA to that extent and did not
To gain further insight into the redox reactions involved
show a linear relationship between complex concentration and
electrochemical studies were carried out. Complexes 1–3
cleavage activity. In contrast the Cu([12]aneN4) complex 1 display
a
one-electron reduction wave in their cyclic
4358 | Dalton Trans., 2013, 42, 4357–4360
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Communication
Fig. 4 Effect of different concentrations of complexes 1, 2 and 3 on pBR322 (0.025 μg μL−1) cleavage activity in Tris-HCl buffer (100 mM, pH 7.4) and ascorbic acid
(0.32 mM) at 37 °C for 2 h. Illustrated is the average of two measurements, the standard deviation is shown as error bars (top). As an example the agarose gel of
complex 2 is shown (bottom). See ESI† for gels of complexes 1 and 3 (cf. S-3,1).
Fig. 5 Cleavage of pBR322 plasmid DNA (0.025 μg μL−1) in Tris-HCl buffer (100 mM, pH 7.4) and ascorbic acid (0.32 mM) for 2 h at 37 °C. The average of two
measurements is illustrated, the standard deviation is shown as error bars. (A) Cleavage by complexes 1, 2 and 3 (0.04 mM) under aerobic and anaerobic conditions.
As an example the agarose gel of complex 1 is shown (bottom). (B) Cleavage by complex 3 in the presence of ROS scavengers. See ESI† for a complete overview of
gels (cf. S-3,5 and S-3,6).
voltammogram (cf. S-4,1†). This process is attributed to the that is related to the higher electronegativity of “O” compared
reduction of Cu(II) to Cu(I). Whereas this wave is reversible for to “N”. The reduction of 3 at even lower negative potential
3, it is electrochemically irreversible for 1 and 2. The irreversi- than 2 is attributed to the better π-donor ability of oxygen com-
bility of this wave is attributed to the release of Cu(I) from the pared to sulphur. Such a trend has precedence in the
cyclen ligands after reduction. This phenomenon has been literature.21
observed earlier20 and in the present case, proof comes from
The release of Cu(I) from macrocycles 1 and 2 as suggested
scanning the cyclic voltammograms multiple times, which by the cyclovoltammetric results implies that the free Cu(I)/Cu(II)
shows an increase in the intensity of the wave attributed to could be the reactive species for the generation of ROS.22 This
redox processes at the released copper centre (cf. S-4,2†). The might explain the high reactivity of complex 2 despite the
reversible nature of the reduction step for 3 is readily explained unfavourable potential for reduction in comparison to com-
by the higher affinity of Cu(I) for a sulphur donor as for an plexes 1 and 3. Electrochemistry, however, displays an outer-
oxygen or a nitrogen donor. As would be expected, 2 is reduced sphere electron transfer process, whereas the reduction of O2
at a lower negative potential compared to 1, an observation generating ROS must be attributed to inner sphere electron
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 4357–4360 | 4359

View Article Online
Communication
Dalton Transactions
transfer.23 Therefore, activity in DNA cleavage does not necess-
arily correlate with redox potentials. The attenuated cleavage
activity of 1 might be explained by hydroxo-bridged dimer for-
mation16 which blocks the coordination sites of Cu(II) leaving
no space for coordination of reducing agents, thus inhibiting
the reduction process and the release of Cu(I).
5 W. Knapp Pogozelski and T. D. Tullius, Chem. Rev., 1998,
98, 1089.
6 (a) H. Stetter and K.-H. Mayer, Chem. Ber., 1961, 94, 1410;
(b) J. E. Richman and T. J. Atkins, J. Am. Chem. Soc., 1974,
96, 2268; (c) M. Suchý and R. H. E. Hudson, Eur. J. Org.
Chem., 2008, 4847.
In conclusion, we present a series of simple macrocyclic
ligands for Cu(II)-based DNA cleavers. In contrast to other
cyclen-based systems, where activity is increased by conju-
gation with moieties enhancing DNA affinity, activity increase
is achieved here by simply changing one out of four donor
atoms in the system. Cu(II) oxacyclen is the most efficient
complex that exceeds the literature-known system Cu(II) cyclen
by a factor of four to ten depending on the concentration. The
high reactivity of this derivative might be a good starting point
for developing more efficient DNA cleavers for potential medi-
cinal applications.
7 (a) J. Li, Y. Yue, J. Zhang, Q.-S. Lu, K. Li, Y. Huang,
Z.-W. Zhang, H.-H. Lin, N. Wang and X.-Q. Yu, Transition
Met. Chem., 2008, 33, 759; (b) Q. Liu, J. Zhang, M.-Q. Wang,
D.-W. Zhang, Q.-S. Lu, Y. Huang, H.-H. Lin and X.-Q. Yu,
Eur. J. Med. Chem., 2010, 45, 5302.
8 Q.-L. Li, J. Huang, Q. Wang, N. Jiang, C.-Q. Xia,
H.-H. Lin, J. Wua and X.-Q. Yu, Bioorg. Med. Chem.,
2006, 14, 4151.
9 (a) A. Bencini, E. Berni, A. Bianchi, C. Giorgi, B. Valtancoli,
D. K. Chand and H.-J. Schneider, Dalton Trans., 2003, 793;
(b) Y. Huang, S.-Y. Chen, J. Zhang, X.-Y. Tan, N. Jiang,
J.-J. Zhang, Y. Zhang, H.-H. Lin and X.-Q. Yu, Chem. Bio-
divers., 2009, 6, 475.
10 J. Stubbe and J. W. Kozarich, Chem. Rev., 1987, 87, 1107.
11 (a) S. W. Jang and J. Suh, Org. Lett., 2008, 10, 481;
(b) C. S. Rossiter, R. A. Mathews and J. R. Morrow, Inorg.
Chem., 2005, 44, 9397.
12 See ESI† for references of ligand synthesis.
13 R. Clay, P. Murray-Rust and J. Murray-Rust, Acta Cryst. B,
1979, 35, 1894.
Acknowledgements
We thank Dr Kai Licha from mivenion GmbH for generous
donations of cyclen.
Notes and references
14 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
15 G. Gasser, M. J. Belousoff, A. M. Bond and L. Spiccia, Inorg.
Chem., 2007, 46, 3876.
16 K. M. Deck, T. A. Tseng and J. N. Burstyn, Inorg. Chem.,
2002, 41, 669.
17 E. L. Hegg and J. N. Burstyn, Inorg. Chem., 1996, 35,
7474.
18 A. Sreedhara, J. D. Freed and J. A. Cowan, J. Am. Chem. Soc.,
2000, 122, 8814.
‡Crystal data for 2: C8H19CuN5O7, M = 360.82, monoclinic, a = 7.921(19) Å, b =
11.979(3) Å, c = 14.757(3) Å, α = 90°, β = 92.587(5)°, γ = 90°, V = 1398.9(6) Å3, T =
133(2) K, space group P21/n, Z = 4, μ(MoKα) = 1.604 mm−1, 17 378 reflections
measured, 4238 independent reflections (Rint = 0.0253). The final R1 value was
0.0290 (I > 2σ(I)). The final wR2 value was 0.0683 (I > 2σ(I)). The final R1 value
was 0.0399 (all data). The final wR2 value was 0.0731 (all data). The goodness of
fit on F2 was 1.089.
Crystal data for 3: C8H19CuN5O6S, M = 376.88, monoclinic, a = 8.532(16) Å, b =
11.903(2) Å, c = 14.168(3) Å, α = 90°, β = 94.368 (4)°, γ = 90°, V = 1434.7(5) Å3, T =
133(2) K, space group P21/c, Z = 4, μ(MoKα) = 1.703 mm−1, 17 105 reflections
measured, 4381 independent reflections (Rint = 0.0143). The final R1 value was
0.0241 (I > 2σ(I)). The final wR2 value was 0.0639 (I > 2σ(I)). The final R1 value
was 0.0264 (all data). The final wR2 value was 0.0649 (all data). The goodness of
fit on F2 was 1.129.
19 C. A. Detmer III, F. V. Pamatong and J. R. Bocarsly, Inorg.
Chem., 1996, 35, 6292.
20 L. M. P. Lima, D. Esteban-Gómez, R. Delgado, C. Platas-
Iglesias and R. Tripier, Inorg. Chem., 2012, 51, 6916.
21 (a) G. A. Fox and C. G. Pierpont, Inorg. Chem., 1992, 31,
3718; (b) K. Ray, T. Weyhermüller, F. Neese and
K. Wieghardt, Inorg. Chem., 2005, 44, 5345.
1 F. H. Westheimer, Science, 1987, 235, 1173.
2 J. C. Wang, Nat. Rev. Mol. Cell Biol., 2002, 3, 430.
3 H. W. Boyer, Annu. Rev. Microbiol., 1971, 25, 153.
22 S.-H. Chiou, J. Biochem., 1983, 94, 1259.
4 (a) F. Mancin, P. Scrimin, P. Tecilla and U. Tonellato, Chem. 23 J. Schnödt, M. Sieger, B. Sarkar, S. Strobel, J. Fiedler,
Commun., 2005, 2540; (b) F. Mancin, P. Scrimin and
P. Tecilla, Chem. Commun., 2012, 48, 5545.
J. S. Manzur, C.-Y. Su and W. Kaim, Z. Anorg. Allg. Chem.,
2011, 637, 930.
4360 | Dalton Trans., 2013, 42, 4357–4360
This journal is © The Royal Society of Chemistry 2013