Vesicles and micelles from amphiphilic Zinc(II)-cyclen complexes as highly potent promoters of hydrolytic DNA cleavage

Article abstract of DOI:10.1021/ja209247w

Phosphate esters are essential to any living organism and their specific hydrolysis plays an important role in many metabolic processes. As phosphodiester bonds can be extraordinary stable, as in DNA, great effort has been put into mimicking the active sites of hydrolytic enzymes which can easily cleave these linkages and were often found to contain one or more coordinated metal ions. With this in mind, we report micellar and vesicular Zn(II)-cyclen complexes which considerably promote the hydrolytic cleavage of native DNA and the activated model substrate bis(4-nitrophenyl)phosphate (BNPP). They are formed by self-assembly from amphiphilic derivatives of previously employed complexes in aqueous solution and therefore allow a simple and rapid connection of multiple active metal sites without great synthetic effort. Considering the hydrolytic cleavage of BNPP at 25 °C and pH 8, the micellar and vesicular metal catalysts show an increase of second-order rate constants (k₂) by 4-7 orders of magnitude compared to the unimolecular complexes under identical conditions. At neutral pH, they produce the highest k₂ values reported so far. For pBR322 plasmid DNA, both a conversion of the supercoiled to the relaxed and linear form, and also a further degradation into smaller fragments by double strand cleavages could be observed after incubation with the vesicular Zn(II)-complexes. Finally, even the cleavage of nonactivated single-stranded oligonucleotides could be considerably promoted compared to background reaction.

Full text of DOI:10.1021/ja209247w

Communication
pubs.acs.org/JACS
Vesicles and Micelles from Amphiphilic Zinc(II)−Cyclen Complexes as
Highly Potent Promoters of Hydrolytic DNA Cleavage
Benjamin Gruber, Evgeny Kataev, Jana Aschenbrenner, Stefan Stadlbauer, and Burkhard Konig*
̈
̈
̈
Institut fur Organische Chemie, Universitat Regensburg, D-93040 Regensburg, Germany
*
S Supporting Information
Zn(II) ions were occasionally reported to be less efficient, they
are the choice of hydrolytic enzymes. Therefore, the design of
zinc-based catalysts for phosphodiester hydrolysis is a favorable
ABSTRACT: Phosphate esters are essential to any living
organism and their specific hydrolysis plays an important
role in many metabolic processes. As phosphodiester
bonds can be extraordinary stable, as in DNA, great effort
has been put into mimicking the active sites of hydrolytic
enzymes which can easily cleave these linkages and were
often found to contain one or more coordinated metal
ions. With this in mind, we report micellar and vesicular
Zn(II)−cyclen complexes which considerably promote the
hydrolytic cleavage of native DNA and the activated model
substrate bis(4-nitrophenyl)phosphate (BNPP). They are
formed by self-assembly from amphiphilic derivatives of
previously employed complexes in aqueous solution and
therefore allow a simple and rapid connection of multiple
active metal sites without great synthetic effort. Consid-
ering the hydrolytic cleavage of BNPP at 25 °C and pH 8,
the micellar and vesicular metal catalysts show an increase
6
biomimetic strategy. Covalent connection of two or more
Zn(II)-catalytic centers by a defined spacer has allowed to
achieve higher catalytic rates for the cleavage of nonactivated
phosphate ester bonds than those observed only for a single
6
a
Zn(II)-center. Combination of catalytic centers through
7
noncovalent interactions in a metallomicelle as well a9s
8
immobilization on polymer supports or on nanoparticles
10
are other successful strategies to increase hydrolytic activity.
We report here self-assembled catalytic systems which are
able to promote the hydrolysis of both native DNA and
activated model substrates. The catalysts are based on
amphiphilic derivatives of the Zn(II)-complex of 1,4,7,10-
tetraazacyclododecane, which was previously found to promote
11
the hydrolytic cleavage of phosphodiester substrates. In
buffered aqueous solution the amphiphilic catalysts Zn1 and
of second-order rate constants (k ) by 4−7 orders of
2
Zn 2 spontaneously assemble either into homomicelles or into
2
magnitude compared to the unimolecular complexes under
identical conditions. At neutral pH, they produce the
vesicular membranes if mixed with a suitable lipid component
like 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC;
Scheme 1).
highest k values reported so far. For pBR322 plasmid
2
DNA, both a conversion of the supercoiled to the relaxed
and linear form, and also a further degradation into smaller
fragments by double strand cleavages could be observed
after incubation with the vesicular Zn(II)−complexes.
Finally, even the cleavage of nonactivated single-stranded
oligonucleotides could be considerably promoted com-
pared to background reaction.
Syntheses and preparation procedures have been previously
described by us or are given in the Supporting Information
SI). A series of mono- and bis-cyclen-based zinc complexes
12
(
(
Zn1−Zn 5) have been tested for the hydrolysis of the simple
2
DNA model substrate bis-(4-nitrophenyl) phosphate (BNPP)
in the pH range from 7 to 9 by following the absorption (Abs)
11
increase of released nitrophenolate (NP) at 408 nm. We have
employed our amphiphilic catalyst derivatives as micellar and
vesicular solutions by adding them to 1−10 mM solutions of
BNPP and recorded kinetics using an initial slopes method in
TRIS buffer (50 mM) at a slightly basic pH of 8.0, as these are
common conditions for DNA hydrolysis experiments and allow
a comparison of rate constants with reported values, and in
HEPES buffer (25 mM) at neutral pH (7.4), which renders the
hydrolysis conditions more challenging.
hosphate esters are ubiquitously found in nature and
P
essential to any living being as DNA with its
phosphodiester backbone contains the genetic code of an
organism. Therefore, these biomolecules need to possess a
sufficient stability under physiological conditions to prevent
premature degradation or accidental altering. DNA has an
estimated half-life of 30 million years concerning the
While at pH 8 Zn 4 represents one of the most active Zn(II)
1
2
spontaneous hydrolysis of a single phosphodiester bond.
7d
complexes for this reaction, we observed that hydrolytic rates
Nevertheless, the specific hydrolysis of such very stable
are exceeded by more than 4 orders of magnitude using Zn 2
2
1
3
phosphodiester bonds plays an important role in metabolic and
micelles (at 2.3 mM BNPP and 0.045 mM Zn 2; Table 1).
2
2
regulatory processes and is easily performed by specific3
The rate constant for Zn1-micelles under these conditions is 1
order of magnitude lower. However, compared to Zn3 as
reference, the micellar system outperforms the unimolecular
catalyst by 7 orders of magnitude at identical experimental
enzymes, which often contain metal ions in their active sites.
Understanding and mimicking these enzymes still represents a
challenge for chemists. A variety of different synthetic catalysts
for hydrolysis of DNA and corresponding model substrates
4
have been reported, usually based on lanthanide or transition
Received: September 30, 2011
Published: November 28, 2011
5
metal ions like Cu(II), Ni(II), Co(II), and Zn(II). Although
©
2011 American Chemical Society
20704
dx.doi.org/10.1021/ja209247w | J. Am. Chem.Soc. 2011, 133, 20704−20707

Journal of the American Chemical Society
Communication
Scheme 1. (Amphiphilic) Zn(II)-Complexes and Vesicular/
Micellar Catalyst Systems
Table 1. Second-Order Rate Constants (k ) for BNPP
Hydrolysis at Neutral pH and 25 °C unless Otherwise Stated
and Second-Order Rate Constants Relative to the Hydroxide
2
Promoted Reaction (krel)
T
1
entry
catalyst
OH−
Zn1-vesicles
(° C)
pH
k (M− s−1
)
krel
ref
20a
this
2
1
2
35
25
7
8
2.4 × 10−5
1
1.5
n.d.
(
DSPC/Zn1
:1)
work
9
3
4
5
Zn1-vesicles
DSPC/Zn1
:1)
25
25
25
7.4
8
1.8
24
75000
n.d.
this
work
(
9
Zn 2-vesicles
this
work
2
(DSPC/Zn22
9:1)
Zn 2-vesicles
7.4
9.4
391667 this
work
2
(DSPC/Zn22
9:1)
6
7
8
9
Zn1-micelles
25
25
25
25
8
7.9
n.d.
this
work
Zn1-micelles
7.4
8
7.4
308333 this
work
Zn 2-micelles
2
68
n.d.
this
work
Zn 2-micelles
2
7.4
7.9
329167 this
work
6.3 × 10−
7
n.d.
11
this
work
1
0
1
Zn3
Zn3
25
25
8
a
a
1
7.4
-
-
1.3 × 10−
3
n.d.
11
1
2
3
Zn24
Zn24
25
25
8
a
a
1
7.4
-
-
this
work
4.3 × 10−
1.5
4
n.d.
7a
14
15
16
17
Zn(II)-micelles
Zn(II)-MPGNs
Ce(IV)-micelles
Ce(IV)-MPGNs
35
40
37
40
10
7
conditions. At pH 7.4, Zn1-micelles give about the same rate
62500
25000
38750
9
constants as at pH 8, whereas the rate constant of Zn 2-micelles
2
7
0.6
20b
20b
drops by about 1 order of magnitude. As a result, both micellar
7
0.93
catalysts give similar k -values under these conditions and still
show a remarkable promotion of BNPP hydrolysis, whereas the
2
a
Below detection limit.
unimolecular complexes Zn3 and Zn 4 are not able to produce
significant amounts of NP here.
2
We explain the dramatically increased reaction rates of our
self-assembled catalyst systems by the high local concentration
of coordinated Zn(II) ions either in the metallomicelles or in
tightly packed domains of metal complexes embedded in the
After observing the remarkable rate increases for the micellar
catalysts, we used the amphiphilic zinc complexes in vesicular
catalysts optimizing a number of parameters: We prepared
vesicles in sizes ranging from about 60−150 nm and compared
their rate constants which were all in the same order of
magnitude; however, the maximum activity was obtained at
about 100 nm (see SI, Table S2 and Figure S9). As a result, we
used these for all further experiments. Next we varied particle
composition in terms of Zn(II)-content of the DSPC vesicle
membrane. The peak activity here was found at loading levels
14
15
bilayer membranes and the decreased polarity at the vesicle
bilayer−water interface, which can be compared to MeOH/
CHCl -mixtures (Figure S5) and therefore facilitates a
3
nucleophilic attack of the phosphodiester substrate compared
to bulk water. Despite the high reaction rates, however, no
complete conversion of the BNPP substrate could be achieved
by micelles or vesicles as indicated by the saturation Abs values
and the produced NP equivalents (see also SI). Although this
of 10 mol % Zn 2 as shown by pseudo-first-order rate constants
2
was not observed in the previous studies with Zn3 and Zn 4,
kobs in Table S1. Higher amounts of metal complexes led to
increasing particle destabilization (indicated by DLS, size
exclusion experiments and shifted emission wavelengths, cf.
Figure S4) and reduced catalytic efficiency. Generally best
results were achieved at BNPP concentrations of 2.3 mM and
2
product inhibition might explain limited turnovers as much
2
+
lower amounts of Zn (0.02 equiv) were used for micellar and
vesicular catalysis compared to the unimolecular complexes
11
which were usually employed stoichiometrically. A large
excess of BNPP and NP(P) might replace counterions (like
0
.02 equiv of Zn(II) (see also Figures S7 and S8). At these
−
reactive OH ) in the surrounding layer of the densely packed
conditions, the vesicular catalysts derived from Zn1 and Zn22
2
+
Zn aggregates and therefore reduce their catalytic activity or
even lead to particle destabilization due to charge neutralization
(cf. Figure S8).
yielded second-order rate constants in the same order of
magnitude as the Zn 2-micelles (Table 1).
2
All kinetic data showed excellent reproducibility (cf. Figure
S14) and control experiments with pure DSPC vesicles
containing no metal complexes and buffer solution were
performed to determine background hydrolysis. No significant
amounts of the hydrolysis product nitrophenolate could be
detected here (Figure S13).
The high hydrolysis rates observed with the activated
phosphate diester BNBP encouraged us to expand the study
to the hydrolysis of less activated oligonucleotides. pBR322
plasmid DNA is a widely used test substrate for artificial
hydrolases. The dsDNA is transformed by strand cleavage from
2
0705
dx.doi.org/10.1021/ja209247w | J. Am. Chem.Soc. 2011, 133, 20704−20707

Journal of the American Chemical Society
Communication
its initial supercoiled strained form (I) into its relaxed circular
form (II) and further on to its open linear form (III). All forms
16
are distinguishable by agarose gel electrophoresis. Supercoiled
pBR322 (100 ng) with a size of 4.3 kb was incubated with
−4
Zn 2-vesicles (1 × 10 Zn(II)) at pH 7.4 and 40 °C for
2
different periods of time before the reactions were stopped with
an excess of SDS and EDTA. The obtained samples were then
separated on agarose mini gels (0.7% w/w) and stained with
ethidium bromide. As shown in Figure 1, the supercoiled DNA
−8
Figure 2. Emission increase of TaqMan probe (1 × 10 M) incubated
−
5
with Zn 4, Zn 2-vesicles and -micelles (1 × 10 M Zn(II)) and buffer
2
2
only at 40 °C and pH 8.0 for 24 and 72 h. Inset: Kinetics for Zn 2-
2
vesicles and background hydrolysis for comparison.
Figure 1. Agarose gel (0.7%) electrophoresis of pBR322 (100 ng per
−4
lane) vs Zn 2-vesicles (1 × 10 mM) incubated at 40 °C at pH 7.4 for
comparison of a FAM emission increase after 24 and 72 h.
According to the saturation fluorescence of the vesicular
2
(
a) 0, (b) 4, (c) 11, (d) 24, (e) 33, (f) 38, (g) 42, (h) 48 h.
19
kinetics, it was assumed that about 75% of the probe
molecules have been cleaved at least at a single site which is
remarkable for a stable nonactivated phosphodiester substrate.
As control, the kinetics were again repeated with pure DSPC
vesicles and buffer solution which resulted in no fluorescence
increase compared to background reaction.
The better performance of vesicles compared to micelles in
the catalysis of polyphosphates may be explained by multi-
valence effects resulting in a higher substrate affinity of the
vesicular catalyst. This hypothesis is supported by the much
higher affinity of fructose-1,6-bisphosphate, used as a non-
form (I) is completely degraded with increasing incubation
times of up to 38 h. However, it is not only transformed to
forms II/III, but also to smaller DNA fragments (IV) with a
17
size of approximately 2.0 kb corresponding to about half the
size of the plasmid. The band of these fragments appears on the
gel even before the bands of form II/III indicating that double-
strand cleavages take place. The cleavage is not selective and all
bands broaden with progressing incubation times; after 24 h, all
four forms are present. The bands become subsequently weaker
with time and convert into a smear around band IV. Such
degradation is remarkable under these conditions and usually
hydrolyzable model substrate, to coumarin-labeled amphiphilic
18
only reported for oxidative degradation conditions. Control
−5
Zn(II)-complex Zn 5 (10 M in HEPES) if embedded in
2
experiments using vesicles without Zn 2 and only buffer
2
DSPC vesicles (10 mol % Zn 5) compared to homomicelles. Its
2
solution showed that the supercoiled DNA remained stable
under the otherwise unchanged experimental conditions.
Additionally, separate samples of the vesicular catalysts were
applied to the gels without DNA to exclude the occurrence of
fluorescent bands by vesicle−ethidium bromide interactions
attached fluorescence label previously allowed the direct
observation of phosphate ion binding to the zinc complex in
12a
a vesicular environment by emission quenching
and is
assumed to respond in a similar way if assembled in micelles
see Figure S19, left). Emission titrations with fructose-1,6-
bisphosphate revealed a binding constant toward the vesicles in
(
(Figure S17).
After observing high hydrolysis rates for BNPP and on
12a
the micromolar range, while no significant zinc-phosphate
coordination could be detected in micelles under these
conditions (Figure S19, right).
strained plasmide dsDNA, we tested our catalytic systems on a
single-stranded oligonucleotide as a more challenging substrate.
We chose the commercially available TaqMan probe consisting
of a 24mer oligonucleotide with the random sequence
CAAGTTTGACCAAGTCACAACGGC and a 5′ fluorescence
label (FAM; λexc = 495 nm, λem = 520 nm) and 3′ quencher
Micellar and vesicular catalysts that promote the hydrolysis
of phosphate esters were self-assembled from amphiphilic
Zn(II)−cyclen complexes and commercial lipids. The approach
avoids laborious synthesis to covalently connect multiple
catalytic centers or attaching them to polymers or nano-
particles. Vesicles with embedded domains of dinuclear Zn(II)
complexes showed a remarkable promotion of phosphodiester
cleavage reactions. Although substrate conversion is limited
here due to considerable product inhibition, the highest so far
reported activities for Zn(II) based catalysts in terms of second-
order rate constants for the hydrolysis of BNPP at neutral pH
and 25 °C were observed, exceeding even those of lanthanide
(BHQ-1 λmax = 534 nm). The intact probe shows only very
weak fluorescence intensity, but upon its cleavage at an
arbitrary site, the effect of the quencher on the fluorophore
diminishes and an increasing emission intensity can be
−8
monitored. For our hydrolysis assay, this probe (10 M) was
incubated with Zn 2-micelles and -vesicles at 40 °C for
2
increasing periods of time. The reactions were stopped by
addition of SDS and EDTA and fluorescence emission
intensities at 520 nm were recorded and plotted against
incubation time as shown in Figure 2 (inset). In contrast to the
20
complexes. pBR322 plasmid DNA is readily degraded by
double strand cleavage, which is difficult to achieve with
results for BNPP, the vesicular catalysts of Zn 2 clearly
2
6a
outperformed the micellar catalyst, which gave only a small
enhancement over background hydrolysis similar to the
nonenzymatic catalysts. The catalytic vesicles also consid-
erably enhance the cleavage of a nonactivated single-strand
oligonucleotide compared to background reaction control.
monomeric water-soluble complex Zn 3 illustrated by the
2
2
0706
dx.doi.org/10.1021/ja209247w | J. Am. Chem.Soc. 2011, 133, 20704−20707

Journal of the American Chemical Society
Communication
In conclusion, vesicles with embedded metal complexes as
catalytic active sites show a remarkable catalytic performance in
phosphodiester hydrolysis. Turnover numbers need to be
improved to enable an effective applicability of these systems.
Their preparation, however, is easy and very versatile, which
greatly facilitates further improvement of properties. In
particular, the hydrolysis of nonactivated oligophosphates is
promoted, which might be due to the presence of dense
clusters of bis-Zn(II) cyclen complexes as catalytic centers on
the vesicle surface. Mechanistic details will be investigated
further. Combining such catalytically active sites with binding
sites for sequence recognition by co-embedding might pave the
way to functionalized vesicles, which are able to discriminate in
their hydrolysis between different substrates or target specific
DNA sequences mimicking the properties of hydrolytic
enzymes more closely.
(11) Subat, M.; Woinaroschy, K.; Gerstl, C.; Sarkar, B.; Kaim, W.;
Ko
12) (a) Gruber, B.; Stadlbauer, S.; Woinaroschy, K.; Ko
Biomol. Chem. 2010, 8, 3704. (b) Turygin, D. S.; Subat, M.; Raitman,
O. A.; Arslanov, V. V.; Konig, B.; Kalinina, M. A. Angew. Chem., Int. Ed.
006, 45, 5340.
̈
nig, B. Inorg. Chem. 2008, 47, 4661.
(
̈
nig, B. Org.
̈
2
13) CMC was determined to be less than 1.5 × 10−4 M (see Figure
S1).
14) Gruber, B.; Stadlbauer, S.; Spat
Konig, B. Angew. Chem., Int. Ed. 2010, 49, 7125.
(
(
̈
h, A.; Weiss, S.; Kalinina, M.;
̈
(15) Saito, H.; Araiso, T.; Shirahama, H.; Koyama, T. J. Biochem.
1991, 109, 559.
(
16) (a) Yuan, C. X.; Wei, Y. B.; Yang, P. Chin. J. Chem. 2006, 24,
006. (b) Tjioe, L.; Meininger, A.; Joshi, T.; Spiccia, L.; Graham, B.
Inorg. Chem. 2011, 50, 4327.
1
(17) According to 1 kb DNA Ladder molecular weight standard.
(18) Zhao, Y. M.; Zhu, J. H.; He, W. J.; Yang, Z.; Zhu, Y. G.; Li, Y. Z.;
Zhang, J. F.; Guo, Z. J. Chem.Eur. J. 2006, 12, 6621.
19) Estimated from a standard curve of free FAM measured under
similar conditions.
20) (a) Yatsimirsky, A. K. Coord. Chem. Rev. 2005, 249, 1997.
b) Bonomi, R.; Scrimin, P.; Mancin, F. Org. Biomol. Chem. 2010, 8,
622.
(
ASSOCIATED CONTENT
Supporting Information
■
S
(
*
(
Synthesis of Zn1, detailed data and experimental conditions for
BNPP, oligonucleotide and plasmid DNA hydrolysis. This
material is available free of charge via the Internet at http://
pubs.acs.org.
2
AUTHOR INFORMATION
Corresponding Author
Burkhard.Koenig@chemie.uni-regensburg.de
■
ACKNOWLEDGMENTS
■
We thank the Deutsche Forschungs gemeinschaft for financial
support. E.K. thanks the Alexander von Humboldt foundation
for a research fellowship.
REFERENCES
■
(
(
(
1) Fekry, M. I.; Gates, K. S. Nat. Chem. Biol. 2009, 5, 710.
2) Sreedhara, A.; Cowan, J. A. J. Biol. Inorg. Chem. 2001, 6, 337.
3) (a) Livieri, M.; Mancin, F.; Saielli, G.; Chin, J.; Tonellato, U.
Chem.Eur. J. 2007, 13, 2246. (b) Jik, C. Curr. Opin. Chem. Biol.
997, 1, 514.
4) (a) Gomez-Tagle, P.; Yatsimirsky, A. K. Inorg. Chem. 2001, 40,
786. (b) Komiyama, M.; Takeda, N.; Shigekawa, H. Chem. Commun.
999, 1443.
5) (a) Koike, T.; Kimura, E. J. Am. Chem. Soc. 1991, 113, 8935.
b) Zulkefeli, M.; Suzuki, A.; Shiro, M.; Hisamatsu, Y.; Kimura, E.;
1
(
3
1
(
(
Aoki, S. Inorg. Chem. 2011, 50, 10113. (c) Vichard, C.; Kaden, T. A.
Inorg. Chim. Acta 2002, 337, 173. (d) Yang, G.-m.; Li, J.; Ren, Y.-w.;
Guo, H.; Duan, M.-y.; Zhang, F.-x.; Zhang, X. Transition Met. Chem.
2
(
(
(
1
009, 34, 191.
6) (a) Mancin, F.; Tecilla, P. New J. Chem. 2007, 31, 800.
b) Niittymaki, T.; Lonnberg, H. Org. Biomol. Chem. 2006, 4, 15.
7) (a) Kimura, E.; Hashimoto, H.; Koike, T. J. Am. Chem. Soc. 1996,
18, 10963. (b) Menger, F. M.; Can, L. H.; Johnson, E.; Durstt, D. H.
J. Am. Chem. Soc. 1987, 109, 2800. (c) Jiang, F. B.; Huang, L. Y.; Meng,
X. G.; Du, J.; Yu, X. Q.; Zhao, Y. F.; Zeng, X. C. J. Colloid Interface Sci.
2
006, 303, 236. (d) Tecilla, P.; Mancin, F.; Scrimin, P.; Tonellato, U.
Coord. Chem. Rev. 2009, 253, 2150.
8) (a) Singh, V.; Zharnikov, M.; Gulino, A.; Gupta, T. J. Mater.
Chem. 2011, 21, 10602. (b) Walenzyk, T.; Konig, B. Inorg. Chim. Acta
005, 358, 2269.
9) Bonomi, R.; Selvestrel, F.; Lombardo, V.; Sissi, C.; Polizzi, S.;
Mancin, F.; Tonellato, U.; Scrimin, P. J. Am. Chem. Soc. 2008, 130,
5744.
10) (a) Manea, F.; Pasquato, L.; Prins, L. J.; Scrimin, P. Nucleic Acids
(
̈
2
(
1
(
Symp. Ser. 2007, 51, 67. (b) Buurma, N. J. Annu. Rep. Prog. Chem., Sect.
B: Org. Chem. 2011, 107, 328.
2
0707
dx.doi.org/10.1021/ja209247w | J. Am. Chem.Soc. 2011, 133, 20704−20707