"Post-it" type connected DNA created with a reversible covalent cross-link

Article abstract of DOI:10.1002/anie.201407854

We report the development of a new heterobase that is held together through reversible bonding. The so-formed cross-link adds strong stabilization to the DNA duplex. Despite this, the cross-link opens and closes through reversible imine bonding. Moreover,

Full text of DOI:10.1002/anie.201407854

Angewandte
Chemie
DOI: 10.1002/anie.201407854
Covalent Reversible Cross-Linking
Hot Paper
“Post-It” Type Connected DNA Created with a Reversible Covalent
Cross-Link**
Marꢀa Tomꢁs-Gamasa, Sascha Serdjukow, Meng Su, Markus Mꢂller, and Thomas Carell*
Dedicated to Prof. Josꢃ Barluenga on the occasion of his 75th birthday
Abstract: We report the development of a new heterobase that
is held together through reversible bonding. The so-formed
cross-link adds strong stabilization to the DNA duplex.
Despite this, the cross-link opens and closes through reversible
imine bonding. Moreover, even enzymatic incorporation of the
cross-link is possible. The new principle can be used to stabilize
DNA duplexes and nanostructures that otherwise require high
salt concentrations, which may hinder biological applications.
D
NA nanotechnology uses the superb self-organizing
properties of oligonucleotides to assemble complex two-
and three-dimensional objects.[1] Although controlling the
complex assembly process was initially the focus of the
research,[1f,2] we are now entering a phase where the
construction of functional systems is desired.[2a,b,3] Examples
include DNA nanostructures decorated with enzymes,[4] and
nanocontainers which can release cargo in response to an
outside stimulus.[5] In particular, the latter application
requires the DNA-based nanostructures to be delivered to
cells and living organisms. However, we encounter a funda-
mental problem with more complex DNA nanoobjects in
living systems: DNA-based nanostructures require high salt
concentrations for their formation and stability, and it is
difficult to create stable structures under low salt concen-
trations, for example, inside cells. Thus far, DNA structures
have been stabilized using disulfide linkages,[6] alkylating
cross-links,[7] and unspecific UV-induced psoralene cross-
links.[4a,8]
Figure 1. A) The aniline:(hydroxy)benzaldehyde cross-link (To:SII).
B) Space-filling model of the To:SII cross-link compared to the natural
A:T base pair.
reported cross-link represents a “post-it”-type glue for DNA
duplexes and, consequently, also for nanostructures in low salt
concentrations.
The cross-link To:SII is created from an aromatic amine
and a benzaldehyde base with a hydroxy group in the ortho-
position (SII). During assembly, a reversible interstrand cross-
link is formed between the nucleobases through an imine
bond. The presence of the ortho-OH group is likely important
to stabilize the structure, similar to the enzyme-bound
pyridoxal phosphate.[9] A benzaldehyde base lacking the
hydroxy group (SI) has been used to verify this hypothesis. It
is interesting that imine bonding also occurs naturally
between an adenine and a sugar of an opposite abasic site
(A:AP).[10]
We report here the development of an enzymatically
processable cross-link that provides DNA structures with
high, but also reversible stability. The cross-linking “bases”
(Figure 1A) form a covalent connection over time that can,
however, be reopened by heating. In this sense, the here
The synthesis of the amine nucleoside To was accom-
plished (Scheme 1) from 1,4-dibromobenzene (1), furolac-
tone 2, and the NH-protected vinylaniline 4. The C-glyco-
sidation was performed by addition of the monolithiated
compound prepared from 1 to the TBS-protected 2’-deoxy-
ribose furolactone 2, followed by reduction to afford the
b-arylfuranose derivative 3.[11] Then, Heck coupling of N-
trifluoroacetyled vinylaniline 4 (obtained in two steps from
commercially available 2-bromoaniline) with the bromoarene
3 provided the E-alkene 5. The trifluoroacetyl protecting
group was found to be fully compatible with both the
nucleoside preparation and the subsequent solid-phase
DNA synthesis. Finally, compound 5 was transformed into
the corresponding phosphoramidite building block for solid-
[*] Dr. M. Tomꢀs-Gamasa, M. Sc. S. Serdjukow, M. Sc. M. Su,
Dr. M. Mꢁller, Prof. Dr. T. Carell
Department of Chemistry and Pharmacy
Ludwig-Maximilians-Universitꢂt Mꢁnchen
Butenandtstrasse 5–13, 81377 Mꢁnchen (Germany)
E-mail: Thomas.Carell@lmu.de
Homepage: http://www.carellgroup.de
[**] M.T.-G. thanks the Alexander von Humboldt Foundation for
a postdoctoral fellowship. We also owe gratitude to S. Schneider for
the X-ray measurements. This work was supported by the Deutsche
Forschungsgemeinschaft, SFB1032 (TP-A5), and SFB749 (TP-
A4).Further support was obtained from the Cluster of Excellence
Nanosystems Initiative Munich (NIM).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201407854.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Scheme 1. Synthesis of phosphoramidite 8 and the amine nucleobase
6. a) 1. nBuLi (1.63 equiv), THF, 30 min, À788C; 2. 2 (1.00 equiv),
THF, 1 h, À788C; 3. Et3SiH (3.28 equiv), BF3·OEt2 (1.23 equiv), CH2Cl2,
2 h, À788C, 23%; b) 4 (1.25 equiv), 8 mol% Pd(OAc)2, 16 mol%
P(o-tol)3, NEt3, 708C, 70%; c) HF·py, NEt3, 6 h, 96%; d) DMT-Cl
(1.1 equiv), pyridine, 3 h, 76%; e) CED-Cl (1.1 equiv), NEtiPr2, THF,
2 h, 92%. TBS=tert-butyldimethylsilyl, DMT=4,4’-dimethoxytrityl,
CED-Cl=2-cyanoethyl N,N-diisopropylchlorophosphoramidite.
Figure 2. A) Melting profiles obtained from duplex D1c, showing the
denaturing curve (blue solid line) and renaturing curve (blue dashed
line), and duplex D1d, showing the denaturing curve (red solid line)
and renaturing curve (red dashed line). Experiments were monitored
at 260 nm in 150 mm NaCl, 10 mm CHES pH 9 with 2 mm concen-
tration for duplexes D1a–d. B) Model proposed for the melting and
the annealing processes.
phase DNA synthesis. Cleavage of the tert-butylsilyl groups
(compound 6) and subsequent selective DMT-protection to
afford nucleoside 7, was followed by generation of the
phosphoramidite 8.
The aldehyde SI was accessed by using a similar modular
synthesis. It required the coupling of furolactone 2 with
protected 4-bromobenzaldehyde followed by sequential
deprotection and protection steps, and phosphoramidite
formation (see p. 9 in the Supporting Information). The
hydroxy aldehyde nucleobase SII and its phosphoramidite
were prepared according to published procedures.[12,18]
The To- and S-containing DNA oligonucleotides were
then synthesized following standard solid-phase phosphor-
amidite protocols for oligonucleotide synthesis (Table 1). In
the case of the To phosphoramidite, the capping step had to be
avoided because of side reactions.[13] This was no limitation,
because the coupling yields of To and all the subsequent bases
were sufficiently high. Finally, treatment of the assembled
DNA with aqueous NH2Me/NH3 (1:1) at 658C removed the
trifluoroacetyl group and all other protecting groups and
afforded the oligonucleotide cleavage from the solid sup-
port.[14] We subsequently evaluated the biophysical properties
of the cross-links To:SI and To:SII. First, the melting temper-
atures of all the synthesized duplexes (Table 1) were deter-
mined by UV spectroscopy. As an example, Figure 2A shows
the typical melting curves measured for D1c and D1d.
Under the chosen conditions, the reference duplex D1a,
(A:T) had a melting point of 488C.[15] The melting temper-
ature dropped to Tm = 358C when the central A:T unit was
replaced by a A:C mismatch (D1b). The duplex with a To:SI
(D1c) had a Tm value of 428C. This result shows that To:SI
lacking the phenolic OH group creates a rather destabilizing
situation, with the imine linkage not established (Figure 2A).
In sharp contrast, the duplex D1d containing the To:SII cross-
link (with the phenolic OH group) displayed a high melting
point of 798C (Figure 2A). This result confirms our hypoth-
esis of the hydroxy group stabilizing the formed imine
through an intramolecular hydrogen bond.[16]
Table 1: Sequences of the oligonucleotides examined for Tm and CD
experiments.
Duplex
Oligonucleotide sequences
Tm [8C]
D1a
5’- CACATTAATGTTGTA À3’
3’- GTGTAATTACAACAT À5’
5’- CACATTACTGTTGTA À3’
3’- GTGTAATAACAACAT À5’
5’- CACATTAToTGTTGTA À3’
3’- GTGTAATSIACAACAT À5’
5’- CACATTAToTGTTGTA À3’
3’- GTGTAATSIIACAACAT À5’
48
35
D1b
D1c
D1d
It is interesting that, while the curves obtained from the
melting and annealing of natural oligonucleotides are super-
imposable in typical melting experiments, substantial hyste-
resis was observed for the duplex D1d (Figure 2A). We
explain this feature of To:SII with a model that involves a slow
rate of imine hydrolysis and formation (Figure 2B). Upon
38[a]
42[b]
38[a]
79[b]
[a] Tm value for renaturing curve. [b] Tm value for denaturing curve.
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
heating, the canonical base pairs melt, but the duplex is still
held together by the central To:SII cross-link (c) until the
imine bond is also hydrolyzed (d). During cooling, the
flanking natural bases might assemble again first, while the
central noncanonical cross-link represents initially a typical
mismatch (a), which is slowly converted into a stabilizing
imine-connected covalent linkage (b).
To gain support for the proposed mechanism, we studied
the UV-melting properties on variation of the heating and
cooling rates. Indeed, the size of the hysteresis effect showed
a critical rate dependence.[17] When we decreased the heating/
cooling steps from 0.58CminÀ1 (DTm = 31.5) to 0.18CminÀ1
(DTm = 18.5) we observed a strongly reduced hysteresis, while
increased hysteresis was detected when the heating rate was
increased from 0.18CminÀ1 to 1.08CminÀ1 (DTm = 39.0).
These results support the idea that imine formation, and
hence the establishment of the covalent bond, is the rate-
determining step in the assembly process of the duplex.
We next investigated whether the To:SII cross-link can be
created enzymatically. This is of interest when the construc-
tion of longer DNA strands is required. Therefore, we
inserted the To base into a 30-mer oligonucleotide template
and hybridized it to a 23-mer primer strand ending one base
before the To unit (Figure 3A). The deoxy-SII triphosphate
control, in which the To:SII cross-link is replaced by an A:T
pair, the dSII-incorporated 24-mer migrates slower in the
polyacrylamide gel electrophoresis (PAGE) experiments
(Figure 3B, lanes 3 and 5). This is in accordance with
observations already reported for the SII homopair.[18] We
next studied the enzymatic selectivity of the dSIITP incorpo-
ration opposite the To base. Kf was found to provide sufficient
selectivity. Only dATP was misincorporated to some extent
(see Figure S7 in the Supporting Information). Although
dSIITP yielded the + 1 elongated primer almost completely
(Figure 3B, lane 3), all the dNTPs applied to the single
nucleotide insertion resulted in mostly non-elongated primer
(Figure 3B, lane 4).
Despite denaturing conditions during PAGE (7m urea)
and extensive heating of the samples before loading (958C,
10 min), we detected bands in the gel that originate from the
duplex (Figure 3B). The cross-link clearly stabilizes the
double helix to an extent that some of the duplexes do not
melt, in agreement with the intended stabilizing effect.
We next studied the possibility of creating fully elongated
primers. We first experimented with all five dXTPs (four
dNTPs and dSIITP) and the Kf(exoÀ) polymerase. In this case,
a very faint band of the fully elongated 30-mer product was
detected in addition to a prominent 24-mer and a 25-mer
elongated primer (Figure 3B, lane 5). Although Kf(exoÀ) is
clearly able to provide some full extension, the best results
were finally obtained with a combination of the polymerases
Kf and Bst Pol I in the presence of all the dXTPs (Figure 3B,
lane 6).[19] In this case, the fully extended primer is the main
product.
Again, some of the duplexes even survive the denaturing
gel electrophoresis conditions. When we next performed
primer extension studies with the dToTP (for the synthesis see
p. 14 in the Supporting Information) using a SII base in the
template, we found that incorporation of dTo opposite the
templating SII is also possible using the polymerases Kf, Bst
Pol I, and others (see Figure S9 in the Supporting Informa-
tion). Quantitative single nucleotide insertion was observed
for Kf in 20 min (see Figure S10 in the Supporting Informa-
tion). The selectivity of the enzymatic dToTP incorporation
resembles the results of dSIITP incorporation opposite To.
Again, dA misinsertion was obtained to some extent, while
only the artificial dToTP quantitively yields the + 1 elongated
primer (see Figure S11 in the Supporting Information). In this
case, full extension proved to be more difficult and only small
amounts of the full length product are formed (see Figure S12
in the Supporting Information).
Figure 3. A) Sequence of the template and primer strands. The primer
was labeled at the 5’ end with fluorescein. B) PAGE from primer
extension experiments using 1 pmol template, 200 mm dNTPs, 400 mm
dSIITP, 2U polymerase. M: marker, lane 1: negative control, lane 2: +1
positive control (24-mer); lane 3: single nucleotide insertion dSIITP
only (10 min, Kf, 378C); lane 4: similar to lane 3 but addition of
dNTPs only; lane 5: dXTPs (6 h, Kf (exoÀ); lane 6: dXTPs two
polymerases (1. Kf, then Bst Pol I; 2. 10 min, 378C; 3. 6 h, 608C). Note
that a sequential addition of components is not required.
To learn about the bioisoteric nature of the To:SII cross-
link, we cocrystallized a DNA double strand containing the
cross-link at position nÀ5 in complex with Bst Pol I (Fig-
ure 4A, PDB code 4UQG). In this experiment the cross-link
is outside the active site of the polymerase, which served just
a crystallization scaffold. The structure shows that the To:SII
bases face each other and that they are covalently linked
through the expected imine interaction.
An overlay with a structure of a canonical dG:dC base
pair at this position shows that the To:SII cross-link provokes
only small structural perturbations in the duplex (Figure 4B).
The distance of 10.8 ꢀ between the C1’ atoms in the
(dSIITP) was synthesized (see p. 13 in the Supporting
Information), and various polymerases and conditions were
systematically screened for their ability to create the To:SII
cross-link. Initial kinetic studies showed that the Klenow
polymerase (Kf) was able to incorporate dSTP opposite To
quantitatively in only 10 min (see Figure S6 in the Supporting
Information). Notably, in comparison to the + 1 positive
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
template strand. The crystal structure reveals that To:SII
mimics the shape of the canonical base pairs. We believe
that the “post-it”-type cross-link will be helpful in the
assembly of stabilized DNA nanostructures.
Received: August 1, 2014
Published online: && &&, &&&&
Keywords: DNA · DNA cross-linking · DNA nanostructures ·
.
imines · reversible bonding
[1] a) F. Zhang, J. Nangreave, Y. Liu, H. Yan, J. Am. Chem. Soc.
2014, 136, 11198 – 11211; b) J.-L. H. A. Duprey, Y. Takezawa, M.
Shionoya, Angew. Chem. Int. Ed. 2013, 52, 1212 – 1216; Angew.
Chem. 2013, 125, 1250 – 1254; c) B. Saccꢁ, C. M. Niemeyer,
Angew. Chem. Int. Ed. 2012, 51, 58 – 66; Angew. Chem. 2012, 124,
60 – 69; d) E. S. Andersen, M. Dong, M. M. Nielsen, K. Jahn, R.
Subramani, W. Mamdouh, M. M. Golas, B. Sander, H. Stark,
C. L. P. Oliveira, J. S. Pedersen, V. Birkedal, F. Besenbacher,
K. V. Gothelf, J. Kjems, Nature 2009, 459, 73 – 76; e) S. M.
Douglas, H. Dietz, T. Liedl, B. Hçgberg, F. Graf, W. M. Shih,
Nature 2009, 459, 414 – 418; f) P. W. K. Rothemund, Nature 2006,
440, 297 – 302; g) W. M. Shih, J. D. Quispe, G. F. Joyce, Nature
2004, 427, 618 – 621; h) N. C. Seeman, Nature 2003, 421, 427; i) J.
Chen, N. C. Seeman, Nature 1991, 350, 631 – 633.
[2] a) T. Tørring, N. V. Voigt, J. Nangreave, H. Yan, K. V. Gothelf,
Chem. Soc. Rev. 2011, 40, 5636 – 5646; b) A. V. Pinheiro, D. Han,
W. M. Shih, H. Yan, Nat. Nanotechnol. 2011, 6, 763 – 772; c) F. A.
Aldaye, A. L. Palmer, H. F. Sleiman, Science 2008, 321, 1795 –
1799; d) K. V. Gothelf, T. H. LaBean, Org. Biomol. Chem. 2005,
3, 4023 – 4037; e) J. Wengel, Org. Biomol. Chem. 2004, 2, 277 –
280.
[3] a) Z.-G. Wang, B. Ding, Adv. Mater. 2013, 25, 3905 – 3914; b) Y.
Zhang, F. Lu, K. G. Yager, D. van der Lelie, O. Gang, Nat.
Nanotechnol. 2013, 8, 865 – 872; c) M. Endo, Y. Yang, H.
Sugiyama, Biomater. Sci. 2013, 1, 347 – 360; d) O. I. Wilner, I.
Willner, Chem. Rev. 2012, 112, 2528 – 2556; e) Y. Krishnan, F. C.
Simmel, Angew. Chem. Int. Ed. 2011, 50, 3124 – 3156; Angew.
Chem. 2011, 123, 3180 – 3215.
Figure 4. A) Crystal structure of Bst Pol I in complex with a To:SII
cross-link-containing DNA. B) The To:SII cross-link overlaid with a
canonical dG:dC base pair. The C1’-C1’ distance of 10.8 ꢃ for the
unnatural cross-link is identical with the distance observed for the
canonical base pair. C) Rear view of the To:SII cross-link, showing
a twist out of the plane. To is depicted in yellow, SII in orange. For the
electron density of B) und C) see Figure S14.
[4] a) M. R. Hartman, D. Yang, T. N. N. Tran, K. Lee, J. S. Kahn, P.
Kiatwuthinon, K. G. Yancey, O. Trotsenko, S. Minko, D. Luo,
Angew. Chem. Int. Ed. 2013, 52, 8699 – 8702; Angew. Chem.
2013, 125, 8861 – 8864; b) S. H. Park, P. Yin, Y. Liu, J. H. Reif,
T. H. LaBean, H. Yan, Nano Lett. 2005, 5, 729 – 733.
[5] A. Banerjee, D. Bhatia, A. Saminathan, S. Chakraborty, S. Kar,
Y. Krishnan, Angew. Chem. Int. Ed. 2013, 52, 6854 – 6857;
Angew. Chem. 2013, 125, 6992 – 6995.
[6] a) M. Ye, J. Guillaume, Y. Liu, R. Sha, R. Wang, N. C. Seeman,
J. W. Canary, Chem. Sci. 2013, 4, 1319 – 1329; b) M. Endo, T.
Majima, Angew. Chem. Int. Ed. 2003, 42, 5744 – 5747; Angew.
Chem. 2003, 115, 5922 – 5925; c) M. Endo, T. Majima, J. Am.
Chem. Soc. 2003, 125, 13654 – 13655; d) S. Alefelder, S. T.
Sigurdsson, Bioorg. Med. Chem. 2000, 8, 269 – 273; e) G. D.
Glick, Biopolymers 1998, 48, 83 – 96; f) S. A. Wolfe, A. E.
Ferentz, V. Grantcharova, M. E. A. Churchill, G. L. Verdine,
Chem. Biol. 1995, 2, 213 – 221; g) N. C. Chaudhuri, E. T. Kool, J.
Am. Chem. Soc. 1995, 117, 10434 – 10442.
[7] a) J. T. Millard, S. Raucher, P. B. Hopkins, J. Am. Chem. Soc.
1990, 112, 2459 – 2460; b) S. E. Sherman, D. Gibson, A. H. J.
Wang, S. J. Lippard, Science 1985, 230, 412 – 417.
[8] A. Rajendran, M. Endo, Y. Katsuda, K. Hidaka, H. Sugiyama, J.
Am. Chem. Soc. 2011, 133, 14488 – 14491.
[9] a) B. A. Yard, L. G. Carter, K. A. Johnson, I. M. Overton, M.
Dorward, H. Liu, S. A. McMahon, M. Oke, D. Puech, G. J.
Barton, J. H. Naismith, D. J. Campopiano, J. Mol. Biol. 2007,
deoxyribose moieties of the artificial cross-link is practically
identical to that in the natural base pair. In contrast, the
previously reported metal- and ethylenediamine-mediated SII
homobase pair has a distance of 11.4 ꢀ (see Figure S15 in the
Supporting Information). The improved fit of the To:SII cross-
link is achieved because To adopts a conformation, which we
did not expect, with the stilbene double bond rotated away
from the counterbase. It could be that the need to populate
this rotameric conformation is the reason for the sluggish full
elongation after dToTP incorporation opposite a SII template.
The To:SII cross-link is not perfectly planar, with SII
slightly twisted from the To plane (Figure 4C). The oxygen
atom from the SII base and the imine bond are in the same
plane, similar to a bicyclic system, trying to avoid steric clash
with the alkene.
In conclusion, we report a new To:SII cross-linking unit
that allows the formation of selective, reversible, and most
important, highly thermostable DNA interstrand cross-links,
based on the concept of reversible imine chemistry. Primer
extensions showed that enzymatic incorporation of the cross-
link is also possible, particularly with the To base in the
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
370, 870 – 886; b) A. E. Beattie, D. J. Clarke, J. M. Wadsworth, J.
Lowther, H.-L. Sin, D. J. Campopiano, Chem. Commun. 2013,
49, 7058 – 7060.
results are similar to the experiments performed at pH 9. The
denaturing/renaturing behavior of duplex D1d under variation
of the pH conditions is presented in Figures S2 and S3 as well as
Table S4 in the Supporting Information.
[10] N. E. Price, K. M. Johnson, J. Wang, M. I. Fekry, Y. Wang, K. S.
Gates, J. Am. Chem. Soc. 2014, 136, 3483 – 3490.
[11] N. Joubert, M. Urban, R. Pohl, M. Hocek, Synthesis 2008, 1918 –
1932.
[12] G. H. Clever, K. Polborn, T. Carell, Angew. Chem. Int. Ed. 2005,
44, 7204 – 7208; Angew. Chem. 2005, 117, 7370 – 7374.
[13] In that case, the amino group underwent cleavage of the
trifluoroacetyl group and further acetyl protection. The acetyl
group could not be removed during final deprotection.
[14] Some other protecting groups were tested before, such as the
benzoyl and the acetyl group. Unfortunately, the deprotection
strategies always failed.
[16] Melting point experiments to study the selectivity of the To
towards the canonical bases have been performed. The results
are shown in Figure S1 in the Supporting information.
[17] The feature of hysteresis as well as the same tendency of rate
dependence are observed under different pH conditions. How-
ever, at pH 7 and pH 8 the hysteresis is always smaller
(independently of the buffer employed). The maximum value
is achieved at pH 9. The results are shown in Figures S2 and S3 in
the Supporting Information.
[18] C. Kaul, M. Mꢂller, M. Wagner, S. Schneider, T. Carell, Nat.
Chem. 2011, 3, 794 – 800.
[15] Buffer CHES pH 9 was used according to previous studies in the
group (see Refs. [12] and [18]). Nevertheless, pH 7 and pH 8
conditions have also been studied (with different buffers). The
[19] E. L. Tae, Y. Wu, G. Xia, P. G. Schultz, F. E. Romesberg, J. Am.
Chem. Soc. 2001, 123, 7439 – 7440.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Covalent Reversible Cross-Linking
M. Tomꢁs-Gamasa, S. Serdjukow, M. Su,
M. Mꢂller, T. Carell*
&&&& — &&&&
“Post-It” Type Connected DNA Created
with a Reversible Covalent Cross-Link
Reversible bond formation was used as
a design principle to create a “post-it”-
type cross-linking unit. The aldehyde–
amine DNA cross-link endows DNA
duplexes with a reversible but unusual
high stability.
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!