Coupling across a DNA helical turn yields a hybrid DNA/organic catenane doubly tailed with functional termini

Article abstract of DOI:10.1021/ja8041096

We describe the synthesis of a hybrid DNA/organic macrocycle that is prepared by formation of an amide linkage across one full turn of DNA. Formation of a catenane proved that the linkage crossed a turn rather than running along the phosphodiester backbone contour. The product, a doubly tailed catenane, contains 5′- and 3′-termini that can be functionalized further or used to incorporate the catenane structure into other DNA assemblies. Copyright

Full text of DOI:10.1021/ja8041096

Published on Web 07/29/2008
Coupling Across a DNA Helical Turn Yields a Hybrid DNA/Organic Catenane
Doubly Tailed with Functional Termini
Yu Liu, Akinori Kuzuya, Ruojie Sha, Johan Guillaume, Risheng Wang, James W. Canary,* and
Nadrian C. Seeman*
Department of Chemistry, New York UniVersity, New York, New York 10003
Received May 31, 2008; E-mail: james.canary@nyu.edu; ned.seeman@nyu.edu
Catenanes constructed from organic building blocks have been
used in the demonstration of molecular devices such as positional
switches, unidirectional motors, and other nanoscale functional
materials.1 In the nucleic acid world, catenanes have long been
known; they are found in nature and are common in structural DNA
nanotechnology.2 Designed, all-DNA catenanes have been used as
topological labels.3 Cross-links with organic linkages have been
utilized to probe the structures and properties of DNA duplexes,
hairpins, and higher order structures, such as t-RNA and ribozymes,4
as well as to build DNA nanostructures and nanodevices.5 In this
paper, we describe the synthesis of a macrocycle that is prepared
by formation of an amide linkage across one full turn of DNA,
forming a tailed DNA/organic catenane containing 5′- and 3′-termini
that are available for further functionalization.
In prior work, we showed that 2′-pendent amines and carboxy-
lates could be linked to form nylon along the phosphodiester
backbone contour.6 Our inquiry here began when we asked whether
longer linkers could be used to attach polymeric components parallel
to the helix axis. This target requires bridging approximately 35 Å
across a nucleic acid turn, which had not been reported previously
(Figure 1A), although there is now a report by Gothelf et al.7
Tetraethylene glycol was used as a spacer into a carboxylate-
functionalized uridine, and undecaethylene glycol was incorporated
into an amine-functionalized uridine. The linker distances have not
been optimized, although studies with tetraethylene glycol as a
Figure 1. Schematic depiction of (A) templated synthesis of the tailed
catenane, (B) cleavage, and labeling of the tailed catenane constructed from
strand 4. The catenane stereoisomer is established by the antiparallel pairing
of DNA before closure.
spacer for both amine and caboxylate were unsuccessful.
Three 16-mer ODNs 1, 2, and 3 were synthesized, where two
modified uridines were separated by 8, 9, and 10 unmodified
nucleotides, respectively. Cross-turn coupling using a DNA hairpin
template gave >97% yields of ODN 1 and 2, as estimated by
MALDI-TOF spectra and denaturing gel electrophoresis, whereas
the coupling yield of 3 was about 62% (Table 1, Supporting
Information). For further characterization, the coupled product was
subjected to complete nuclease digestion followed by analysis to
detect linked nucleotides. The detection of a polyethylene glycol
(PEG)-amide linked uridine dimer by LCMS corroborated the
formation of coupled products (Supporting Information).
demonstrated that the cross-turn coupling strategy produced a
linkage parallel to the DNA helix axis.
To test for the catenated structure, the synthesized catenane was
first treated with exonuclease; no dissociation was observed,
although the tails of the hybrid macrocycle were degraded. To
release the DNA circular template by cleaving the hybrid macro-
cycle, a catenane was constructed from ODN 4 containing a
photocleavable linker (PCL) (Figure 1B). It was then restricted with
BamH I, which cleaved the 78-mer DNA circular template as
indicated in Figure 1B to produce an 18-mer and a 60-mer and
released the PEG-amide-oligonucleotide hybrid circle as shown by
its electrophoretic mobility (Figure 2, lane 6). Identical products
were obtained from cleavage of the DNA template alone (Figure
2, lane 7) and coupled ODN 4 (lane 8). Upon irradiating the
catenane with near-UV light (350 nm), the PCL was cleaved,
resulting in linearization of the hybrid macrocycle and release of
the template circle. The reaction products appear in Figure 2 (lane
4) and can be compared with intact catenane and the DNA circular
template (lanes 5 and 3) and photocleavage products of coupled
strand 4 (lane 2).
These data established the formula of the reaction product but
did not distinguish between a newly formed linkage parallel to the
helix axis (i.e., across the turn) versus along the phosphodiester
backbone contour. However, templation of the reaction by a circular
DNA template (Figure 1A) would yield an interlocked complex
only if the coupling reaction occurred across the turn. ODNs 1 and
2 paired with a 78-mer DNA circular template were subjected to
the same reaction procedure as the hairpin-templated coupling.
Almost-quantitive yields of catenanes were produced from the
coupling of both 1 and 2, which was established by denaturing gel
analysis. Thus, the linkage across a full turn of DNA helix generated
a doubly tailed catenane with special features discussed below. This
9
10882 J. AM. CHEM. SOC. 2008, 130, 10882–10883
10.1021/ja8041096 CCC: $40.75
2008 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Oligonucleotides Used in This Study and MALDI-TOF MS Analysisa
uncoupled
coupled
found
ODNs
sequence
calcd
found
calcd
1
2
3
4
5′-TG U3N6ACGTGCGAU1C6TTCG
5′-TG U3N6ACGTGCGATU1C6TCG
5′-TG U3N6ACGTGCGATTU1C6CG
5′-TG U3N6ACGTG(PCL)CGATU1C6TCG
5729.2
5729.2
5729.2
5988.4
5729.4
5728.4
5728.9
5988.4
5711.2
5711.2
5711.2
5970.4
5711.3
5710.5
5711.0, 5728.1
5969.1
b
a Cross-turn coupling trials were conducted with a DNA hairpin as template. b PCL: Photocleavable linker.
proaches that employ 5′- and 3′-terminal linkers.5a,9 To test the
feasibility of postsynthetic modification of these tails, the catenane
containing ODN 4 was subjected to 5′- and 3′-32P labeling in two
experiments. The generation of 3′-labeled (Figure 3, lane 1) and
5′-labeled product (lane 3) are readily visible in the autoradiogram
of the denaturing gel. Thus, this cross-turn coupling strategy may
be advantageous for nucleic acid labeling. In principle, nucleic acids
of any sequence could be targeted and labeled by this approach.
In conclusion, coupling across a DNA helical turn was achieved
utilizing ODNs decorated with 2′-functionalized linkers. The
formation of a doubly tailed catenane based on this strategy
demonstrates intrastrand cross-linking. The free ends of the
catenated PEG-amide-oligonucleotides provide useful handles for
postsynthetic modification or labeling.
Acknowledgment. We gratefully acknowledge support by the
NSF (CTS-0608889) to N.C.S. and J.W.C., Grant GM-076202 from
NIGMS to J.W.C., and by Grants GM-29554 from NIGMS, Grant
CCF-0726378 from the NSF, Grants 48681-EL and MURI W911NF-
07-1-0439 from ARO, and a Grant from the W. M. Keck
Foundation to N.C.S.
Figure 2. Denaturing gel analysis of the dissociation of the tailed catenane
by BamH I digestion and UV cleavage. Lane 1: 10bp DNA ladder marker.
Lane 2: UV-cleaved products of coupled strand 4. Lane 3: 78-mer DNA
circle. Lane 4: UV-cleaved products of the tailed catenane. Lane 5: The
tailed catenane. Lane 6: BamH I digested products of the tailed catenane.
Lane 7: BamH I digested products of 78-mer DNA circle. Lane 8: Coupled
strand 4. Lane 9: Uncoupled strand 4. L: Linear DNA. C: Circular DNA.
Supporting Information Available: Full experimental details
including: Syntheses of phosphoramidites, MALDI-TOF MS of ODNs,
LCMS analysis of complete nuclease digestion, denaturing gel analyses
of catenane synthesis, exonuclease digestion, and dissociation by
restriction enzyme treatment. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) For recent reviews, see: (a) Arico, F.; Badjic, J. D.; Cantrill, S. J.; Flood,
A. H.; Leung, K. C. F.; Liu, Y.; Stoddart, J. F. Top. Curr. Chem. 2005, 249
(Templates in Chemistry II), 203-259. (b) Champin, B.; Mobian, P.;
Sauvage, J. P. Chem. Soc. ReV. 2007, 36, 358–366. (c) Kay, E. R.; Leigh,
D. A.; Zerbetto, F. Angew. Chem., Int. Ed. 2007, 46, 72–191. (d) Lankshear,
M. D.; Beer, P. D. Acc. Chem. Res. 2007, 40, 657–668.
(2) Seeman, N. C. In Synthetic DNA Topology, Molecular Catenanes, Rotaxanes
and Knots; Sauvage, J.-P., Dietrich-Buchecker, D., Eds.; Wiley-VCH:
Weinheim, 1999; pp 323-356..
(3) Liang, X. G.; Kuhn, H.; Frank-Kamenetskii, M. D. Biophys. J. 2006, 90,
2877–2889.
(4) (a) El-Sagheer, A. H.; Kumar, R.; Findlow, S.; Werner, J. M.; Lane, A. N.;
Brown, T. ChemBioChem 2008, 9, 50–52. (b) Walter, N. G.; Burke, J. M.
Curr. Opin. Chem. Biol. 1998, 2, 24–30. (c) Glick, G. D. Biopolymers 1998,
48, 83–96.
(5) (a) Kumar, R.; El-Sagheer, A.; Tumpane, J.; Lincoln, P.; Wilhelmsson, L. M.;
Brown, T. J. Am. Chem. Soc. 2007, 129, 6859–6864. (b) Endo, M.; Uegaki,
S.; Majima, T. Chem. Commun. 2005, 3153–3155. (c) Endo, M.; Seeman,
N. C.; Majima, T. Angew. Chem., Int. Ed. 2005, 44, 6074–6077.
(6) (a) Zhu, L.; Lukeman, P. S.; Canary, J. W.; Seeman, N. C. J. Am. Chem.
Soc. 2003, 125, 10178–10179. (b) Liu, Y.; Wang, R.; Ding, L.; Sha, R.;
Lukeman, P. S.; Canary, J. W.; Seeman, N. C. ChemBioChem 2008, 9, 1641–
1648. (c) Liu, Y., Sha, R., Wang, R., Ding, L., Canary, J. W., Seeman, N.
C. Tetrahedron, in press.
Figure 3. Autoradiogram of a denaturing gel showing both 5′ and 3′
labeling of the tailed catenane. Lane 1: 3′-32P labeling of the tailed catenane.
Lane 2: 3′-32P labeling of 78-mer DNA circle. Lane 3: 5′-32P labeling of
the tailed catenane. Lane 4: 5′-32P labeling of 78-mer DNA circle. Lane 5:
10bp DNA ladder marker.
(7) Andersen, C. S., Yan, H., Gothelf, K. Angew. Chem., Int. Ed. 2008, 47,
5569–5572.
The catenane formation reaction may be viewed as a “padlock”
function as the oligonucleotide acted as a probe to recognize the
target circular DNA, and interlocking was accomplished by the
pendent linkers. The 2′-localization of the linkers creates a
bibracchial lariat structure,8 leaving the free 5′- and 3′-ends of the
oligonucleotides available to serve as accessible sites for further
modification or labeling, which is an advantage over prior ap-
(8) Gokel, G. W.; Barbour, L. J.; Ferdani, R.; Hu, J. Acc. Chem. Res. 2002, 35,
878–886.
(9) (a) Ryan, K.; Kool, E. T. Chem. Biol. 1998, 5, 59–67. (b) Escude, C.;
Garestier, T.; Helene, C. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 10603–
10607. (c) Demidov, V. V.; Kuhn, H.; Lavrentieva-Smolina, I. V.; Frank-
Kameneteskii, M. D. Methods 2001, 23, 123–131.
JA8041096
9
J. AM. CHEM. SOC. VOL. 130, NO. 33, 2008 10883