Synthesis of HyBeacons and dual-labelled probes containing 2'-fluorescent groups for use in genetic analysis.

Article abstract of DOI:10.1039/b302855k

An FMOC-protected 2'-hydroxyethyl uridine phosphoramidite has been used to synthesise fluorescein-labelled HyBeacon probes and "FAM-ROX" dual-labelled fluorogenic oligonucleotides.

Full text of DOI:10.1039/b302855k

Synthesis of HyBeacons and dual-labelled probes containing
2A-fluorescent groups for use in genetic analysis†
Neil Dobson,a David G. McDowell,b David J. French,b Lynda J. Brown,c John M. Mellora and Tom
Brown*a
a Department of Chemistry, University of Southampton, Highfield, Southampton, UK SO17 1BJ.
E-mail: tb2@soton.ac.uk; Fax: +44 (0)23 80592991; Tel: +44 (0)23 80592974
b LGC Ltd, Queens Road, Teddington, UK TW11 0LY
c
Oswel Research Products Ltd, Biological and Medical Sciences Building, University of Southampton,
Boldrewood, Bassett Crescent East, Southampton, UK SO16 7PX
Received (in Cambridge, UK) 13th March 2003, Accepted 3rd April 2003
First published as an Advance Article on the web 7th May 2003
An FMOC-protected 2A-hydroxyethyl uridine phosphor-
amidite has been used to synthesise fluorescein-labelled
HyBeacon probes and “FAM-ROX” dual-labelled fluoro-
genic oligonucleotides.
cially available dye phosphoramidites.10 The synthesis of the
required phosphoramidite monomer is shown in Scheme 1‡.
N(3)-BOM uridine 111 was protected with Markiewicz’s
reagent to give 2, which was reacted with methyl bromoacetate
in the presence of sodium hydride to give 3 in 72% yield.
Reduction of the ester with sodium borohydride proved
unsatisfactory with our substrate (28% yield), the major product
being a tetrahydrouridine derivative obtained by reduction of
the heterocycle. However, removal of the BOM protecting
group from the nucleobase before reduction prevented this side-
reaction. The BOM-protecting group was cleaved with Pd/C
and reduction of the ester using sodium borohydride gave 4
(65% over 2 steps). Reprotection of N-3 was necessary to give
a clean reaction in the next step in which the 2A-hydroxy group
was protected with FMOC to give 5 (86%). Deprotection of the
BOM group, and removal of TIPDS using HF–pyridine in THF,
followed by tritylation gave 6 in 80% yield. Finally, phosphity-
lation of the 3A-hydroxy group under argon afforded the
phosphoramidite 7 in good yield.
Monomer 7 can be introduced into synthetic oligonucleotides
at thymine sites and used as a point of attachment for a variety
of fluorescent dyes amidites. In order to evaluate its utility, the
HyBeacon probe§ 2DC64C* was synthesised4 (Scheme 2).
Standard protocols were employed, with an extended reaction
time of 5 min for 7, which coupled at 98%. We did not remove
the 5A-DMT protecting group from the oligonucleotide at this
stage. After oligonucleotide assembly the synthesis column was
removed and treated with piperidine (20% in DMF) for 10 min
to remove the FMOC group from the 2A-protected uridine
residue. The column was then returned to the DNA synthesiser,
6-carboxyfluorescein phosphoramidite was coupled to the free
2A-hydroxy group and the 5A-DMT group was removed. The
oligonucleotide was then cleaved from the resin, deprotected in
Many genetic diseases are linked to single nucleotide polymor-
phisms occurring at a significant frequency within the popula-
tion (1% or more), so the ability to rapidly detect SNPs is of
great value. Commonly used homogeneous real-time PCR-
based methods of genetic analysis such as Taqman™1 Scorpi-
ons2 and Molecular Beacons3 are used to analyse SNPs on
platforms such as the Roche LightCycler™. In a drive towards
inexpensive, high throughput genetic analyses we have devel-
oped the novel “HyBeacon” system.4,5 HyBeacons™ are
hybridisation probes consisting of a single-stranded oligonu-
cleotide containing a fluorophore attached to a nucleoside
within the DNA sequence. When the probe anneals to a
complementary target and forms a duplex, an increase in
fluorescence is observed, the precise origin of which is the
subject of ongoing research (Fig. 1).
As the probe possesses no secondary structure, hybridisation
to the target is faster and more efficient than in other genetic
analysis systems, and post-PCR melting analysis becomes
possible. Furthermore, HyBeacons do not require enzymic
activation, so they can be used in conjunction with rapid PCR
cycling conditions.6 Finally, HyBeacon synthesis is relatively
simple and inexpensive, as the probe does not contain a
fluorescence quencher.
In previous work in this field4,5 we incorporated fluorescent
dyes in the major groove of DNA, via the 5-position of
deoxyuridine. The next logical step was to investigate the
properties of HyBeacons with the fluorophore in the minor
groove. The results of the major groove studies4,5 indicated that
the most efficient HyBeacons would result from the use of a
short linker between the nucleotide and the dye moiety, so we
based our strategy on 2A-hydroxyethyluridine. In previous
studies oligonucleotides have been labelled at the 2A-position of
uridine using specific dye-labelled phosphoramidites,7–9 or by
solution-phase derivatisation of a 2A-amino dU residue after
solid-phase synthesis.9 However, our aim was to develop
efficient general solid-phase approaches based on phosphor-
amidites that can be prepared in large quantities and used in
oligonucleotide synthesis in conjunction with other commer-
Scheme 1 (i) 1,3-Dichloro-1,1,3,3-tetraisopropyldisiloxane (1.2 eq.) imida-
zole (1.1 eq.), pyridine rt 7 h, 91%; (ii) BrCH2CO2Me (5 eq.), NaH (2.2 eq.),
DMF, 25 °C, 2 h, 72%; (iii) 20 wt% Pd/C, THF : methanol (1 : 1), H2(g),
RT, 86%; (iv) NaBH4, methanol, tert-butanol, rt, 76%; (v) benzyl
chloromethyl ether, DBU, DMF, rt, 82%; (vi) FmocCl, pyridine, rt, 86%;
(vii) 20 wt% Pd/C, 2 M HCl (aq), THF : methanol (1 : 1), H2(g), rt, 62%;
(viii) HF–pyridine, THF, rt, 54%; (ix) DMTrCl (3 eq.), pyridine, rt, 3 h,
80%; (x) 2-cyanoethyldiisopropyl chlorophosphoramidite, DIPEA, THF,
72%.
Fig. 1 HyBeacon.
† Electronic supplementary information (ESI) available: experimental
details and real-time PCR. See http://www.rsc.org/suppdata/cc/b3/
b302855k/
1234
CHEM. COMMUN., 2003, 1234–1235
This journal is © The Royal Society of Chemistry 2003

aqueous ammonia (5 h, 55 °C) and purified by reversed-phase
HPLC (octyl) with a gradient of 0 to 35% CH3CN in 0.1 M
NH4OAc.
The resultant HyBeacon probe (2DC64C*) was designed to
differentiate between a fully complementary DNA target and
one containing a single C.A mismatch. The 2A-FAM ethoxy dU
is adjacent to the polymorphic site, and is therefore sensitive to
changes in Tm caused by mispairing at this locus. Real-time
PCR†¶ indicated that good discrimination was achieved. The
magnitude of the change in fluorescence on hybridisation of
2DC64C* was similar to that for HyBeacons with major groove
fluorophores.4 Post-PCR fluorescence melting analysis (Fig. 2),
a particular advantage of the HyBeacon system, enabled us to
discriminate very clearly between wild-type, heterozygote and
homozygous mutant. These results prove that the fluorescence
enhancement, characteristic of HyBeacon probes, can be readily
achieved using oligonucleotides with fluorophores attached to
the minor groove.
This novel synthetic methodology can be used to incorporate
different fluorescent dyes in the minor groove of DNA to
produce “multi-coloured” HyBeacons, and such probes could
be used in high-throughput genetic analysis to detect several
different SNPs in a single tube. However, the majority of
current genetic analysis platforms have only one fluorescence
excitation source at 495 nm, and in order to utilise all the
available detection channels, fluorescence resonance energy
transfer (FRET) must be used. This can be achieved by
incorporating FAM (FRET donor) and a second fluorescent dye
(FRET acceptor) in the same oligonucleotide.12–14 To explore
the utility of monomer 7 in FRET applications we prepared
dual-labelled oligonucleotide probes P1 & P2 (Fig. 3) with an
internal FAM and 5A-ROX (Scheme 2).
Fig. 3 FRET from FAM to ROX.
to 47 °C for P1 and 44 °C for P2 (1 mM duplex, 100 mM NaCl,
10 mM Na phosphate, pH 7.0). The origin of these differences
is probably twofold; destabilisation due to the presumed C3A-
endo conformation of the modified ribose sugar, and steric
hindrance caused by the bulky 2A-fluorescein moiety. It is clear
from the excellent discrimination in Fig. 2 that the slight
instability does not confer any limitations on the use of our
nucleotide in genetic probes. Similar effects were observed by
Yamana et al.7,8
Energy transfer from FAM to ROX was recorded for the P1
and P2 duplexes in aqueous buffer at RT, pH 7.0 (data not
shown). The fluorescence intensity of ROX (606 nm) was more
than 4-fold greater in P2 than in P1, demonstrating the need to
separate the two dyes to avoid collisional quenching. FRET in
P2 is efficient with a ROX : FAM emission ratio of 8.2 : 1 These
results are most encouraging and indicate that monomer 7 can
be used in the synthesis of dual-labelled probes for a variety of
applications. The properties of dual-labelled hybridisation
probes are currently under further investigation. In conclusion,
we have synthesised a HyBeacon functionalised with 2A-
fluorescein and shown it to be an effective genetic probe. We
have also demonstrated that the 2A-modified FAM dU nucleo-
side can be used in the synthesis of dual-labelled FRET
probes.
We wish to thank BBSRC and JREI for funding for the Roche
LightCycler, TCS for financial support, and Dr Nicola Thelwell
(Oswel) for helpful comments.
A FRET study was carried out on P1 and P2 hybridised to a
fully complementary oligodeoxynucleotide. but before doing
this we determined the effect of the 2A-modification on duplex
stability. The Tm of the native DNA duplex is 48 °C compared
Notes and references
‡ Compounds were characterised by 1H-NMR, 13C-NMR, MS, IR and mp/
CHN microanalysis or HRMS.
§ The HyBeacon 2DC64C* sequence is dGGGCGFCTGGGGGTGX,
where C = polymorphic site, F = 2A-FAM ethoxy dU, and X = 3A-octanol
PCR stopper). It targets the human CYP2D6 4 mutation in which there is a
G to A transition at position 1846. It matches the wild-type sequence (*1,
C.G base pair) and is mismatched to the mutant (*4, C.A base pair).
¶ The protocols for PCR were taken from previous work on HyBeacons.4,6
Aminolink and 6-FAM Phosphoramidites were purchased from Trans-
genomics Ltd, West of Scotland Science Park, Glasgow, UK. NHS 6-ROX
was purchased from Molecular Probes Inc (www.probes.com).
1 K. J. Livak, S. A. J. Flood, J. Marmaro, W. Giusti and K. Deetz, PCR
Methods and Applications, 1995, 357.
2 D. Whitcombe, J. Theaker, S. P. Guy, T. Brown and S. Little, Nature
Biotech., 1999, 17, 804.
3 S. Tyagi and F. R. Kramer, Nature Biotech., 1995, 14, 303.
4 D. J. French, C. L. Archard, T. Brown and D. G. McDowell, Mol. Cell
Probes, 2001, 15, 363.
5 D. J. French, D. G. McDowell and T. Brown, International patent
application WO 0173118, 2001.
6 D. J. French, C. L. Archard, M. T. Andersen and D. G. McDowell, Mol.
Cell Probes, 2002, 16, 319.
Scheme 2 Oligonucleotide synthesis.
7 K. Yamana, T. Mitsui, H. Hayashi and H. Nakano, Tetrahedron Lett.,
1997, 38, 5815.
8 K. Yamana, Y. Ohashi, K. Nunota and H. Nakano, Tetrahedron, 1997,
53, 4265.
9 K. Yamana, T. Mitsui and H. Nakano, Tetrahedron, 1999, 55, 9143.
10 L. J. Brown, J. P. May and T. Brown, Tetrahedron Lett., 2001, 42,
2587.
11 M. Krecmerova, H. Hrebabecky and H. Antonin, Collect. Czech. Chem.
Commun., 1996, 61, 645.
12 A. Solinas, L. J. Brown, C. McKeen, J. M. Mellor, J. T. G. Nicol, N.
Thelwell and T. Brown, Nucleic Acids Res., 2001, 29, e96.
13 M. A. Ali and S. A. Ahmed, J. Chem. Phys., 1988, 90, 1484.
14 R. A. Cardullo, S. Agrawal, C. Flores, P. C. Zamecnik and D. E. Wolf,
Proc. Natl. Acad. Sci. U. S. A., 1988, 85, 8790.
Fig. 2 Post-PCR fluorescence melting analysis using 2DC64C* HyBeacon.
Data acquired over 15 min. Red: fully complementary homozygous
template; Green: heterozygote; Blue: mutant; Black: negative control (no
template DNA).
CHEM. COMMUN., 2003, 1234–1235
1235