A new approach to oligonucleotide labelling using Diels-Alder cycloadditions and detection by SERRS

Article abstract of DOI:10.1039/b204790j

Diels-Alder cycloaddition has been used to add a benzotriazole azo dye to a diene tagged oligonucleotide to generate unique SERRS signals at 10 attomoles.

Full text of DOI:10.1039/b204790j

A new approach to oligonucleotide labelling using Diels–Alder
cycloadditions and detection by SERRS†
Ljiljana Fruk, Antonio Grondin, W. Ewen Smith and Duncan Graham*
Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow, UK
G1 1XL. E-mail: duncan.graham@strath.ac.uk; Fax: 0141 552 0876; Tel: 0141 548 4701
Received (in Cambridge, UK) 17th May 2002, Accepted 29th July 2002
First published as an Advance Article on the web 8th August 2002
Diels–Alder cycloaddition has been used to add a benzo-
triazole azo dye to a diene tagged oligonucleotide to generate
unique SERRS signals at 10 attomoles.
species and shows differences to that of the developing reagent.
Unlike fluorescence, this enables detection without separa-
tion.
Diels–Alder cycloaddition was chosen as the chemistry for
labeling, as it is a fast, quantitative reaction that occurs in
aqueous solution. A diene was used as the tag and a dienophile
as the developing reagent. Diels–Alder cycloaddition has been
used previously to add fluorescent labels to oligonucleotides
highlighting the simplicity and ease of the approach for direct
labeling but unlike our approach isolation of the labeled probe
Modern DNA detection makes use of a number of physical
techniques that utilize detection of a label covalently added to
an oligonucleotide probe. In particular fluorescence spectros-
copy is widely used.1 The addition of an external label such as
a fluorophore can be a disadvantage since most labels have
appreciable biological activity in that they can associate with
particular biomolecules in a non-covalent fashion. This can lead
,2
9
was required prior to analysis. In this study the approach of
3,4
to reduced and in some cases total probe inactivity.
using Diels–Alder cycloaddition to produce a SERRS active
species is demonstrated by using oligonucleotides. The diene
chosen was furan which has been shown to react with a
maleimide to produce different SERS signals10 and is easily
added to the 5’-terminus of a synthetic oligonucleotide (Scheme
1). (This method has been used to prove the principle of
cycloaddition. Synthesis of a furan phosphoramidite is currently
under investigation to allow more efficient addition.) Two
different oligonucleotides were used to ensure that there were
no sequence specific effects from subsequent experiments.
A new, non-fluorescent dye was designed as the developing
agent using a maleimide as the dienophile, an azo and a
benzotriazole group to complex to the silver surface. (Scheme
2).
The synthesis of a benzotriazole dye maleimide (BDM),
which incorporated the maleimide as part of the chromophore
so that the SERRS spectrum of the cycloadduct was different
from the BDM, was a substantial synthetic challenge. A number
of alternatives were investigated before the route shown in
Scheme 2 was chosen. The most difficult step was the
intramolecular cyclisation to form the maleimide. A number of
standard methods were attempted, however, due to low yields or
Additionally any excess labeled probe has to either be removed
from the assay or internally masked prior to analysis. This leads
to extra separation procedures or the use of complicated and
expensive probes. In an attempt to overcome these problems we
have designed a new approach to detection based on Diels–
Alder cycloadditions and surface enhanced resonance Raman
scattering, SERRS which provides unique signals of the desired
product using a simple probe without use of separation steps at
ultra low concentrations (Scheme 1).
SERRS requires the presence of a coloured moiety that will
adsorb onto a metal surface.5 This produces enhanced Raman
scattering at ultra low concentration levels and with a high
degree of molecular specificity.7 In this new approach a small
chemical tag is added to a probe molecule. The tag is invisible
to the detection technique of SERRS and is only made SERRS
active by use of a developing agent that reacts specifically with
the tag to produce the label. As the new species has a different
molecular structure, the signal produced is unique to that
,6
,8
Scheme 1 Reagents and conditions: (i) DCM, (ii) CDI, DMF, 40 °C, (iii)
4
NH OH, 55 °C.
†
Electronic supplementary information (ESI) available: full experimental
details on the synthesis and analysis of the reported compounds. See http://
www.rsc.org/suppdata/cc/b2/b204790j/
Scheme 2 Reagents and conditions: (i) HCl, NaNO
2
, NaOAc pH 6.0, (ii)
MalAn, DCM, (iii) Co naphthenate, DMA, Ac O.
2
2
100
CHEM. COMMUN., 2002, 2100–2101
This journal is © The Royal Society of Chemistry 2002

side reactions the method of choice was the use of cobalt(II
)
naphthenate with acetic anhydride.11
The oligonucleotides were then exposed to the BDM (3)
under a variety of conditions and the reaction initially followed
at high concentration by HPLC to ensure the reaction was
occurring. The optimal conditions were found to be similar to
9
those reported by Hill et al. except that the reaction time and
number of equivalents of dienophile required could be reduced
by use of a copper(II) nitrate catalyst. Equimolar amounts of
furan oligo and BDM could be used to give complete reaction
with the catalyst. It also reduced the reaction time to less than
one hour. Both oligonucleotides gave the same result which
proved that the reaction was not a sequence dependent process.
Fig. 1 shows the cycloaddition as followed by HPLC using a 25
mM phosphate buffer at pH 5.5 and a three-fold excess of
BDM.
The reaction was then followed by SERRS at a concentration
lower than could be detected by a UV-HPLC system to evaluate
the sensitivity. In this case a 1+1 ratio of BDM to furan
oligonucleotide was used and no separation steps were
performed prior to analysis. Two methods of SERRS analysis
were investigated. One was the addition of 10 ml of the reaction
mixture to a silver film produced in situ from a silver nitrate/
PVA matrix and the other was the addition to citrate reduced
silver colloid. The PVA films have been used previously to give
good SERRS12 and in this case direct deposition of the mixture
followed by an aqueous wash after 10 min resulted in the
excellent spectra shown in Fig. 2. The cycloadduct showed
distinct differences from the BDM. The spectra were more
intense, new bands formed at 1204, 1263 and 1387 (azo stretch)
2
8
Fig. 2 SERRS of the cycloaddition from a silver film; 10 ml of a 6.3 3 10
2
1
cm and there were changes in relative intensity in several
other bands. The most obvious features to be used for detection
M solution of reaction mixture was used for each analysis.
2
1
are the shift in frequency of the band at 1399 to 1387 cm and
oligonucleotide influencing the surface adsorption properties of
the cycloadduct. The sensitivity of these oligonucleotides is
excellent, with the sensitivity of the PVA films being better than
that of the colloid. The exciting light is focused onto a small area
2
1
the change in intensity of the band at 1434 cm on production
of the cycloadduct. DFT on related compounds indicated that
these bands are predominantly stretching modes of the aromatic
rings with a contribution from the azo stretch.
2
of the film (350 objective d = 5 mm, ≈ 20 pm ) and we
When the silver colloidal suspension was used, direct
addition of the mixture to the suspension resulted in very poor
signals. However, as found previously for oligonucleotides,
addition of spermine allowed improved surface adsorption1
and resulted in similar spectra to those obtained from the PVA
films. Surprisingly the signal to noise ratio of the oligonucleo-
tide cycloadducts is significantly better than that of the
cycloadduct alone. We attribute this to the presence of the
estimate the actual number of moles under examination as being
in the region of 10 attomoles.
This is a totally new approach to biological labeling and
shows excellent promise in terms of selectivity, sensitivity and
simplicity. It is a convenient method of labeling oligonucleo-
tides in an aqueous environment to produce distinctive species
that do not need separation from starting materials prior to
analysis. This approach to labeling of biomolecules will be
effective for many other targets and a number of alternatives are
under investigation.
3,14
The authors wish to the BBSRC for the award of a David
Phillips Fellowship to D. G.
Notes and references
1
2
3
A. Castro and J. G. K. Williams, Anal. Chem., 1997, 69, 3915.
L. J. McBride and M. D. Oneill, Am. Lab., 1991, 23, 52 et seq.
Z. R. Zhu, J. Chao, H. Yu and A. S. Waggoner, Nucleic Acids Res.,
1
994, 22, 3418.
4
5
6
7
C. Wojczewski, K. Stolze and J. W. Engels, Synlett, 1999, 1667.
P. Hildebrandt and M. Stockburger, J. Phys. Chem., 1984, 88, 5935.
A. M. Stacy and R. P. Vanduyne, Chem. Phys. Lett., 1983, 102, 365.
C. H. Munro, W. E. Smith, M. Garner, J. Clarkson and P. C. White,
Langmuir, 1995, 11, 3712.
8
9
D. Graham, B. J. Mallinder and W. E. Smith, Angew. Chem., Int. Ed.,
2
000, 39, 1061.
K. W. Hill, J. Taunton-Rigby, J. D. Carter, E. Kropp, K. Vagle, W.
Pieken, D. P. C. McGee, G. M. Husar, M. Leuck, D. J. Anziano and D.
P. Sebesta, J. Org. Chem., 2001, 66, 5352.
1
0 A. Grondin, D. C. Robson, W. E. Smith and D. Graham, J. Chem. Soc.,
Perkin Trans. 2, 2001, 2136.
1
1
1 J. E. T. Corrie, J. Chem. Soc., Perkin Trans. 1, 1994, 2975.
2 M. Volkan, D. L. Stokes and V. D. Tuan, Appl. Spectrosc., 2000, 54,
1
842.
1
3 D. Graham, W. E. Smith, A. M. T. Linacre, C. H. Munro, N. D. Watson
and P. C. White, Anal. Chem., 1997, 69, 4703.
Fig. 1 HPLC trace of the cycloaddition. Peak 1 = furan oligo peak 2 =
oligo cycloadduct.
14 D. Graham, B. J. Mallinder and W. E. Smith, Biopolym. (Biospectrosc.),
2000, 57, 85.
CHEM. COMMUN., 2002, 2100–2101
2101