Fast copper-free click DNA ligation by the ring-strain promoted alkyne-azide cycloaddition reaction

Article abstract of DOI:10.1039/c1cc10743g

Templated DNA strand ligation by the ring-strain promoted alkyne-azide [3+2] cycloaddition reaction is very fast; with dibenzocyclooctyne, the reaction is essentially complete in 1 min. It is inhibited by the presence of a single mismatched base pair sugg

Full text of DOI:10.1039/c1cc10743g

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 6257–6259
www.rsc.org/chemcomm
COMMUNICATION
Fast copper-free click DNA ligation by the ring-strain promoted
alkyne-azide cycloaddition reactionw
Montserrat Shelbourne,a Xiong Chen,a Tom Brownz*a and Afaf H. El-Sagheerz*ab
Received 8th February 2011, Accepted 12th April 2011
DOI: 10.1039/c1cc10743g
Templated DNA strand ligation by the ring-strain promoted
alkyne–azide [3+2] cycloaddition reaction is very fast; with
dibenzocyclooctyne, the reaction is essentially complete in 1 min.
It is inhibited by the presence of a single mismatched base pair
suggesting applications in genetic analysis.
non-templated mode. This should be achievable, as DNA
templation accelerates reaction rates by several orders of
magnitude.17 To achieve this objective we have synthesised
two activated cyclooctyne derivatives (Fig. 2) for conjugation
to amino-functionalised oligonucleotides; the non-substituted
cyclooctyne (NSCO, 1) and the previously reported dibenzo-
cyclooctynol (DIBO, 2).25 We anticipated that alkyne 2 would
be more reactive towards azides than 1 because the aromatic
rings are expected to impose additional ring strain and electron
withdrawing properties.25 For the method to be suitable for
the controlled simultaneous ligation of several oligonucleotides it
is important that the ligation reaction occurs only in the
presence of a complementary DNA template so that only
the desired products are formed. Therefore it was necessary
to compare reactivity under both templated and non-
templated conditions.
The CuAAC ‘click reaction’ (CuI-catalysed [3+2] alkyne azide
cycloaddition) has been used in a vast range of applications
due to its high efficiency and orthogonality with almost all
other functional groups and solvents.1 It has found favour in
the nucleic acids field2 as a method of joining together single
strands of DNA, cross-linking complementary strands,2,3
cyclising single and double strands,4,5 labelling oligonucleo-
tides with reporter groups,2,6–9 attaching DNA to surfaces,2
producing analogues of DNA with modified nucleobases2,7,10,11
and backbones,2,12–14 synthesising large chemically modified
RNA constructs15 and creating biochemically active PCR
templates.12 However, CuI has undesirable cytotoxicity even
at low concentrations16 so it is not fully compatible with in vivo
applications. Unfortunately, the uncatalysed DNA-templated
AAC reaction with terminal alkynes is exceedingly slow in
comparison17 unless highly activated alkynes are used, and
these are unstable in water. Clearly, it would be advantageous
to have the option to carry out click chemistry on DNA in the
absence of metal ion catalysis. For carbohydrates this problem
has been elegantly solved16,18–22 by the introduction of the ring
strain-promoted azide–alkyne [3+2] cycloaddition reaction
(SPAAC).23,24 This involves the reaction between an azide
and a strained alkyne such as a cyclooctyne. Here, we have
adopted this approach for nucleic acids and describe the use of
cyclooctyne derivatives in CuI-free templated DNA ligation
(Fig. 1).
Initially both alkynes were attached post-synthetically to the
50-end of an aminohexyl-labelled oligonucleotide (ODN-1, 2,
Table 1). The NHS ester of 6-azidohexanoic acid was added to
a 30-amino alkyl labelled oligonucleotide to provide the azide
oligonucleotide (ODN-3) which has a fluorescein dye at the
50-end to allow visualisation at low concentrations. HPLC
purification was carried out on all oligonucleotides and the
products were characterised by mass spectrometry (ESI).
Templated and non-templated ligation reactions between
azide ODN-3 and alkyne ODN-1 and ODN-2 were carried
out in the absence of CuI.
Of the two cyclooctynes tested, DIBO (2) was much faster in
templated ligation (Fig. 3). Reactions with this alkyne
proceeded cleanly and were essentially complete within
1 min at 2 mM DNA concentration. NSCO (1) also reacted
cleanly but required more than 30 min for complete reaction.
In both cases the non-templated reactions gave little or
no product under otherwise identical conditions (Fig. 3).
Importantly, introduction of a single mismatch base pair
between template ODN-11 and DIBO-labelled ODN-1 was
sufficient to inhibit the ligation reaction (Fig. 3 lanes 17, 18),
pointing to future applications in genetic analysis.
Interestingly the addition of CuI strongly inhibited the
SPAAC reaction (Fig. 4a lanes 4, 5, 8, 9 and ESIw). There
are two possibilities here: either the alkyne is deactivated on
complexation to CuI, or degrades when exposed to the metal
ion. Treatment of the cyclooctyne ODNs with CuI indicated
that no degradation occurred (analysis by mass spec and gel
The aim of this study was to discover alkynes that would
react quickly with azides at low concentration under
DNA-templated conditions, but remain unreactive in a
a School of Chemistry, University of Southampton, Highfield,
Southampton, SO17 1BJ, UK. E-mail: ahes@soton.ac.uk,
tb2@soton.ac.uk
b Chemistry Branch, Dept. of Science and Mathematics,
Faculty of Petroleum and Mining Engineering,
Suez Canal University, Suez 43721, Egypt
w Electronic supplementary information (ESI) available: Experimental
details, gel electrophoresis, UV melting studies, oligonucleotide mass
spectral data. See DOI: 10.1039/c1cc10743g
z AHE-S and TB are joint main authors.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 6257–6259 6257

View Online
Fig. 1 SPAAC click DNA ligation between azide-labelled and cyclooctyne-labelled oligonucleotides. (a) schematic of reaction (FAM =
6-carboxyfluorescein). (b, c) chemical structures of NSCO and DIBO alkynes respectively at the ligation point.
Fig. 3 SPAAC templated and non-templated reactions comparing
Fig. 2 (1, 2) Activated cyclooctynes for labelling amino-modified
oligonucleotides. (3) DIBO phosphoramidite for insertion during
solid-phase oligonucleotide synthesis.
DIBO and NSCO alkyne oligonucleotides. Lanes 1–3: templated
reactions using DIBO ODN-1; 0 min, 5 min, 30 min, RT, lanes 4–6:
non-templated reactions using DIBO ODN-1; 0 min, 5 min, 30 min,
RT, lanes 7–9: templated reactions using NSCO ODN-2; 0 min, 5 min,
30 min, RT, lanes 10–12: non-templated reactions using NSCO
ODN-2; 0 min, 5 min, 30 min, RT, lane 13: azide ODN-3, lanes 14–16:
templated reactions using DIBO ODN-1; 1 min, 3 min, 5 min, RT, lanes
17 and 18 discrimination between fully matched and mismatched
templated reactions using DIBO ODN-1 and templates ODN-5, (fully
matched, lane 17) and ODN-11 (single base pair mismatch, lane 18);
5 min, 45 1C. All reactions performed at 2 mM oligonucleotide conc. in
0.2 M aq. NaCl for the specified time then mixed with formamide and
loaded directly onto a 20% polyacrylamide gel.
Table 1 Oligonucleotide sequences. F = fluorescein, z = amino C7
labelled with 6-azidohexanoic acid, CK = aminohexyl labelled with
NSCO (1), DIBOK = aminohexyl labelled with DIBO (2), DIBOK1
=
dibenzocyclooctynyl derived from phosphoramidite (3). X, Y and
Z = ligated triazole products derived from ODN-3 with ODN-1,
ODN-2 and ODN-4 respectively. FzDIBO = 6-fluoresceinamidopro-
pylazide + DIBO
Code
Oligonucleotide sequences
ODN-1 DIBOK-GCGATCAATCAGACG
ODN-2 CK-GCGATCAATCAGACG
ODN-3 F-CTTTCCTCCACTGTTGCz
ODN-4 DIBOK1-GCGATCAATCAGACG
ODN-5 TTTATTGATCGCGCAACAGTGTTT
ODN-6 F-CTTTCCTCCACTGTTGCXGCGATCAATCAGACG
ODN-7 F-CTTTCCTCCACTGTTGCYGCGATCAATCAGACG
ODN-8 F-CTTTCCTCCACTGTTGCZGCGATCAATCAGACG
ꢀ
ꢀ
ꢀ
ODN-9 FzDIBOK1-GCGATCAATCAGACG
ODN-10 CTTTCCTCCACTGTTGCGCGATCAATCAGACG
ODN-11 TTTATTCATCGCGCAACAGTGTTT
electrophoresis). To test the stability of the putative complex
with cuprous ion, the alkyne oligonucleotides were treated
with CuI then all small molecules were removed by sephadex
gel-filtration before carrying out the SPAAC reaction. This
process did not restore the reactivity of the alkyne suggesting
that the alkyne-CuI complex survived gel-filtration (Fig. 4a). It
is known that CuI readily forms a complex with cyclooctyne26
(Fig. 4b) and we assume that such a complex prevents
the SPAAC reaction from occurring. The complex was not
sufficiently stable to be observed by mass spectrometry of the
oligonucleotides. Ethylenediamine tetraacetic acid (EDTA)
(100 equ.) partially restored the reactivity of the cyclooctyne
oligonucleotide (ESIw).
Fig. 4 (a) SPAAC templated and non-templated reactions using
alkyne oligonucleotides pre-treated with CuI. Lane 1: azide ODN-3,
lanes 2 and 3: control, templated and non-templated SPAAC reactions
using the DIBO alkyne oligonucleotide (ODN-1), lanes 4 and 5:
templated and non-templated reactions using DIBO oligonucleotide
(ODN-1) pre-treated with CuI, lanes 6 and 7: control; templated and
non-templated SPAAC reactions using the NSCO alkyne oligonucleo-
tide (ODN-2), lanes 8 and 9: templated and non-templated reactions
using NSCO oligonucleotide (ODN-2) pre-treated with CuI. All
reactions carried out at 2 mM DNA conc. in 0.2 M NaCl for 30 min
at room temperature before loading directly onto a 20% polyacrylamide
gel. (b) Cu(I)-cyclooctyne complex.26
c
6258 Chem. Commun., 2011, 47, 6257–6259
This journal is The Royal Society of Chemistry 2011

View Online
Ultraviolet melting studies on duplexes containing the
alkynes NSCO (1) and DIBO (2) showed a decrease of
B7 1C in melting temperature (Tm) compared with the
unmodified duplex which had a Tm of 66.1 1C (ESIw). The
similarity in Tm for both ligated oligonucleotides was expected
as both triazole linkers are quite long. If the linker was short
there might be a difference in stability between duplexes
containing the two different alkynes as DIBO is more bulky,
and might also participate in aromatic stacking interactions
with the nucleobases.
CuAAC reaction, does not require catalysis with toxic metals.
Importantly the cyclooctyne and azide oligonucleotides are
both stable in aqueous buffer. The SPAAC reaction on DNA
has potential for applications in biology, genomics and
nanotechnology.
AHE-S and TB have received funding from the European
Union’s Seventh Framework Programme (FP7/2007-2013)
under grant agreement no 201418 (READNA).
Notes and references
It would be more convenient if the alkyne functionality
could be added to oligonucleotides during solid-phase
synthesis rather than post-synthetically. This would provide
the necessary orthogonality to facilitate the synthesis of
cyclooctyne oligonucleotides which also contain sensitive
fluorescent dyes or other labels which have to be added as
active esters after solid-phase synthesis. With this in mind, and
to evaluate the effect of the length of cyclooctyne linker on
click reactivity, we prepared the phosphoramidite derivative 3
of DIBO (Fig. 2 and ESIw). The cyclooctyne was completely
stable to oligonucleotide synthesis and deprotection conditions,
including heating in conc. aqueous ammonia for 5 h at 55 1C.
Pure alkyne-labelled oligonucleotides were readily obtained
and these gave efficient templated ligation with azide oligo-
nucleotides (ESIw). UV melting of the duplex containing the
ligated product (ODN-8) from phosphoramidite 3 (propyl-
carbamoyl linker) gave a similar Tm to that obtained from the
ligated product (ODN-6) derived from the alkyne active ester
(amidohexyl linker) (ESIw).
Oligonucleotide labelling with fluorophores is of major
importance for DNA diagnostics, sequencing and related
applications.27 Some fluorophores can be incorporated into
oligonucleotides during solid-phase synthesis but many are
labile, and therefore unsuitable for direct insertion. In such
cases an alternative strategy is to introduce fluorophores post-
synthetically by the use of amine/NHS ester, thiol/iodoacetamide
or thiol/maleimide chemistry. These methods have limitations,
as the electrophiles rapidly decompose in aqueous media.
In recent years click chemistry has become an important
alternative; it provides the highest conjugation efficiency and
uses very stable, robust alkynes and azides. Oligonucleotide
labelling using the copper-free SPAAC reaction would
constitute a significant advance and could in principle be
carried out in vivo. To evaluate this chemistry, labelling
of ODN-4 with 6-carboxyfluorescein azide (10 eq.) was
performed at 37 1C for 1 h, resulting in complete conversion
to labelled oligonucleotide ODN-9 as two isomeric fluorescent
products (ESIw). While working on this manuscript two
independent reports have appeared on the use of this
methodology for adding tags to DNA.28,29
1 P. Wu and V. V. Fokin, Aldrichimica Acta, 2007, 40, 7–17.
2 A. H. El-Sagheer and T. Brown, Chem. Soc. Rev., 2010, 39,
1388–1405.
3 P. Kocalka, A. H. El-Sagheer and T. Brown, ChemBioChem, 2008,
9, 1280–1285.
4 R. Kumar, A. El-Sagheer, J. Tumpane, P. Lincoln,
L. M. Wilhelmsson and T. Brown, J. Am. Chem. Soc., 2007, 129,
6859–6864.
5 A. H. El-Sagheer, R. Kumar, S. Findlow, J. M. Werner,
A. N. Lane and T. Brown, ChemBioChem, 2008, 9, 50–52.
6 J. Gierlich, G. A. Burley, P. M. E. Gramlich, D. M. Hammond and
T. Carell, Org. Lett., 2006, 8, 3639–3642.
7 F. Seela and X. Ming, Helv. Chim. Acta, 2008, 91, 1181–1200.
8 F. Seela, V. R. Sirivolu and P. Chittepu, Bioconjugate Chem., 2008,
19, 211–224.
9 S. Berndl, N. Herzig, P. Kele, D. Lachmann, X. H. Li,
O. S. Wolfbeis and H. A. Wagenknecht, Bioconjugate Chem.,
2009, 20, 558–564.
10 F. Seela, H. Xiong, P. Leonard and S. Budow, Org. Biomol. Chem.,
2009, 7, 1374–1387.
11 F. Seela and V. R. Sirivolu, Chem. Biodiversity, 2006, 3, 509–514.
12 A. H. El-Sagheer and T. Brown, J. Am. Chem. Soc., 2009, 131,
3958–3964.
13 H. Isobe, T. Fujino, N. Yamazaki, M. Guillot-Nieckowski and
E. Nakamura, Org. Lett., 2008, 10, 3729–3732.
14 T. Fujino, N. Yamazaki and H. Isobe, Tetrahedron Lett., 2009, 50,
4101–4103.
15 A. H. El-Sagheer and T. Brown, Proc. Natl. Acad. Sci. U. S. A.,
2010, 107, 15329–15334.
16 J. C. Jewett, E. M. Sletten and C. R. Bertozzi, J. Am. Chem. Soc.,
2010, 132, 3688–3690.
17 A. H. El-Sagheer and T. Brown, Pure Appl. Chem., 2010, 82,
1599–1607.
18 Z. J. Gartner and C. R. Bertozzi, Proc. Natl. Acad. Sci. U. S. A.,
2009, 106, 4606–4610.
19 P. V. Chang, J. A. Prescher, E. M. Sletten, J. M. Baskin,
I. A. Miller, N. J. Agard, A. Lo and C. R. Bertozzi, Proc. Natl.
Acad. Sci. U. S. A., 2010, 107, 1821–1826.
20 S. T. Laughlin, J. M. Baskin, S. L. Amacher and C. R. Bertozzi,
Science, 2008, 320, 664–667.
21 S. T. Laughlin and C. R. Bertozzi, ACS Chem. Biol., 2009, 4,
1068–1072.
22 N. J. Agard and C. R. Bertozzi, Acc. Chem. Res., 2009, 42,
788–797.
23 J. C. Jewett and C. R. Bertozzi, Chem. Soc. Rev., 2010, 39,
1272–1279.
24 M. F. Debets, C. W. J. van der Doelen, F. Rutjes and F. L. van
Delft, ChemBioChem, 2010, 11, 1168–1184.
25 X. Ning, J. Guo, M. A. Wolfert and G. J. Boons, Angew. Chem.,
Int. Ed., 2008, 47, 2253–2255.
26 G. Groger, U. Behrens and F. Olbrich, Organometallics, 2000, 19,
3354–3360.
27 R. T. Ranasinghe and T. Brown, Chem. Commun., 2005,
5487–5502.
28 P. van Delft, N. J. Meeuwenoord, S. Hoogendoorn, J. Dinkelaar,
H. S. Overkleeft, G. A. van der Marel and D. V. Filippov, Org.
Lett., 2010, 12, 5486–5489.
29 K. N. Jayaprakash, C. G. Peng, D. Butler, J. P. Varghese,
M. A. Maier, K. G. Rajeev and M. Manoharan, Org. Lett.,
2010, 12, 5410–5413.
In summary, two cyclooctynes have been incorporated into
oligonucleotides and used in SPAAC reactions. Templated
DNA ligation was very fast and a single base pair mismatch
was sufficient to strongly inhibit the reaction. It should be
possible to use this approach for multiple simultaneous
templated DNA ligation reactions if participating oligo-
nucleotides are labelled with either two alkynes or two azides.
The SPAAC reaction is orthogonal to amide bond formation
and the Diels–Alder reaction on DNA30 and, unlike the
30 A. H. El-Sagheer, V. V. Cheong and T. Brown, Org. Biomol.
Chem., 2011, 9, 232–235.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 6257–6259 6259