Nucleic acid secondary structures containing the double-headed nucleoside 5′(S)-C-(2-(thymin-1-yl)ethyl)thymidine

Article abstract of DOI:10.1039/b810930c

Oligodeoxynucleotides containing the double-headed nucleoside 5′(S)-C-(2-(thymin-1-yl)ethyl)thymidine were prepared by standard solid phase synthesis. The synthetic building block for incorporating the double-headed moiety was prepared from thymidine, whi

Full text of DOI:10.1039/b810930c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Nucleic acid secondary structures containing the double-headed nucleoside
5¢(S)-C-(2-(thymin-1-yl)ethyl)thymidine
Charlotte Andersen,a Pawan K. Sharma,b Mikkel S. Christensen,a Signe I. Steffansen,a Charlotte M. Madsena
and Poul Nielsen*a
Received 27th June 2008, Accepted 5th August 2008
First published as an Advance Article on the web 18th September 2008
DOI: 10.1039/b810930c
Oligodeoxynucleotides containing the double-headed nucleoside 5¢(S)-C-(2-(thymin-1-
yl)ethyl)thymidine were prepared by standard solid phase synthesis. The synthetic building block for
incorporating the double-headed moiety was prepared from thymidine, which was stereoselectively
converted to a protected 5¢(S)-C-hydroxyethyl derivative and used to alkylate the additional thymine by
a Mitsunobu reaction. The oligodeoxynucleotides were studied in different nucleic acid secondary
structures: duplexes, bulged duplexes, three-way junctions and artificial DNA zipper motifs. The
thermal stability of these complexes was studied, demonstrating an almost uniform thermal penalty of
incorporating one double-headed nucleoside moiety into a duplex or a bulged duplex, comparable to
the effects of the previously reported double-headed nucleoside 5¢(S)-C-(thymin-1-yl)methylthymidine.
The additional base showed only very small effects when incorporated into DNA or RNA three-way
junctions. The various DNA zipper arrangements indicated that extending the linker from methylene to
ethylene almost completely removed the selective minor groove base–base stacking interactions
observed for the methylene linker in a (-3)-zipper, whereas interactions, although somewhat smaller,
were observed for the ethylene linker in a (-4)-zipper motif.
Introduction
Chemically modified oligonucleotides can serve in nucleic acid
nanotechnology,1,2 oligonucleotide-based diagnostics3 and as anti-
sense drugs.4 Altered or improved selective nucleic acid recognition
is a key aim, as is varying the properties of DNA as a skeleton
for nanoscale engineering. The idea of using double-headed 2¢-
deoxynucleosides, in which additional nucleobases have been
attached to 2¢-deoxynucleotides, for these purposes has recently
been introduced by us and others.5–9 The additional nucleobase
constitutes a potential recognition site on the outside of a duplex
or another nucleic acid secondary structure. The nucleobase can
participate in specific hydrogen bonding as well as stacking and
hereby pave the way for a second coding function, for interstrand
communication and for thermal stabilisation of a secondary
structure.
As our first example of a double-headed nucleoside, we intro-
duced 1 (see Fig. 1) containing an additional nucleobase in the 2¢-
position.6 The hybridisation properties of oligodeoxynucleotides
(ODNs) containing this building block showed a slight thermal
destabilisation of both DNA : RNA and DNA : DNA duplexes as
compared with unmodified duplexes. By using the same ODNs and
targets containing a hairpin structure, 1 was also incorporated in
the branching point of a three-way junction, and this arrangement
demonstrated a stabilising effect of the additional nucleobase as
Fig. 1 Double-headed nucleosides. T = thymin-1-yl, U = uracil-1-yl, A =
adenin-9-yl.
◦
judged from an increase in melting temperature (Tm) of +4.2 C
in a DNA context and +2.5 ◦C with an RNA complement.6
Herdewijn and co-workers introduced 2 as another example
of a double-headed nucleoside with either thymine or adenine as
the additional nucleobase pointing into the minor groove when
incorporated into duplexes.7 With the purpose of finding base–
base interactions, preferably a classic A–T base-pairing in the
minor grove, the most significant results were found when using
solely the adenine moiety, and these interactions were rather weak
and indicated to be a result of stacking and not of Watson–Crick
type base-paring.7
Another recent example from our laboratory was the double-
headed nucleoside 3, where the additional thymine has been
attached to the 5¢-position via a methylene linker.8 By using
the 5¢(S)-configuration, the substituent points towards the minor
groove in a B-type duplex. By incorporating 3 into ODNs, the
effect of the additional base was studied in different secondary
nucleic acid structures. In standard duplexes the modification
caused a constant drop in Tm of 4–6 ◦C in a DNA : DNA and
3.3 ◦C in a DNA : RNA duplex. However, interesting base–base
aNucleic Acid Center†, Department of Chemistry and Physics, University
of Southern Denmark, 5230 Odense M, Denmark. E-mail: pon@ifk.sdu.dk;
Fax: +45 66158780; Tel: +45 65502565
bDepartment of Chemistry, Kurukshetra University, Kurukshetra-136 119,
India
† The Nucleic Acid Center is funded by the Danish National Research
Foundation for studies on nucleic acid chemical biology.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3983–3988 | 3983
©

View Article Online
interactions in the minor groove were detected in an artificial
DNA zipper arrangement of 3. Hence, hybridisation studies and
molecular modelling demonstrated a specific stacking interaction
between the additional nucleobases when arranged in the (-3)-
zipper arrangement, that is positioned on each complementary
strand in the duplex with two interspacing base-pairs. Thus, the
destabilisation of the duplex, which in all other cases was additive
with two or more modifications, was counteracted by 6–7 ◦C with
the (-3)-zipper arrangement.8
On the other hand, hybridisation studies of ODNs with 3
incorporated in the branching point of a three-way junction in the
same way as for 1 did not show any thermal stabilisation in either
a DNA : DNA or a DNA : RNA context. Molecular modelling of
a DNA three-way junction with a CC bulge adjacent to the strand
exchange site indicated that the additional base is positioned near
the region of the bulge pyrimidines but with no specific contact.8
Although the thermal studies of this type of structure did not
indicate a clear interaction, i.e. neither a significant destabilisation
or stabilisation, it is known that this type of structure exhibits
folding isomerism with regard to the site of bulging out, i.e.
between the junction and either the 3¢ end or the 5¢ end.10 It is
therefore entirely possible that the modelling structure and the
experimental system display different isomeric conformations. To
investigate this end further, we reasoned that a longer linker such as
homologue 4 (Fig. 1) would increase the possibility of interaction
between the extra base and the bulge.
to investigate the effect of the longer linker on DNA secondary
structures including modified DNA zippers.
Results and discussion
Chemical synthesis
The double-headed nucleoside 4 was synthetically approached
by a linear strategy with thymidine as the starting material.
The first attempted strategy was designed using a Michael-type
addition as the key step for introducing the additional thymine
(Scheme 1). Previously, several acyclic nucleosides have been
synthesised using a Michael-type strategy.11–13 Using a known
method, the 3¢-protected derivative 5 was converted to the 5¢-
C-vinylthymidine derivative 6 as an epimeric mixture by oxida-
tion, Grignard reaction affording the 5¢-C-ethynyl derivative and
Lindlar hydrogenation.14 The Michael substrate 7 was obtained
by oxidation of the secondary alcohol in 6 using Dess–Martin
periodinane. The Michael-type addition was attempted with either
thymine or 3-N-benzoylthymine using numerous different condi-
tions including heating in DMSO with or without the addition
of different bases or treatment in ionic liquid.13 Unfortunately,
no double-headed nucleoside was isolated, probably due to the
reversibility of the Michael-type reaction.
Also the influence of linker length on the base–base interactions
observed in a DNA zipper should be investigated. Hence, a
modelling study of 4 similar to the homologue 38 with a (-3)-
zipper arrangement in a B-DNA duplex suggested that the extra
nucleobases stack in the minor groove (Fig. 2). However, whereas
the shorter methylene linker places the two additional thymine
bases in an arrangement almost co-planar to the bases in the core
strands (Fig. 2a),8 the longer ethylene linker cannot accommodate
such a conformation and places the extra nucleobases in a
perpendicular arrangement (Fig. 2b).
Scheme 1 Reagents and conditions: a, ref. 14: (i) DMP, DCM, 76%,
(ii) HCCMgBr, THF, 70%, (iii) Lindlar cat., H2, quinoline, EtOAc, 92%;
b, DMP, DCM, 76%; TBS = tert-butyldimethylsilyl.
The synthetic strategy was then changed, and the introduction
of the additional thymine was approached via a Mitsunobu
reaction as applied in the preparation of 1. Firstly, attempts were
made to transform compound 6 to the 1,3-dihydroxyl derivative by
hydroboration of the double bond, but only with limited success.
Therefore, we decided to approach the target by introducing a 5¢-
C-allyl group, which could be subjected to an oxidative cleavage
(Scheme 2). We have previously reported15 that the 5¢-C-allylation
can be performed with complete stereoselectivity by the use of the
TBDPS protecting group instead of the TBS group.16 Hence, the
5¢-hydroxy group of the thymidine derivative 8 was oxidised and
the allyl linker was introduced via a Lewis acid catalyzed reaction
to give exclusively the 5¢(S)-configured product 9.15 The secondary
alcohol of compound 9 was protected with the pixyl group to
give 10 in 96% yield. The pixyl group has proven superior to the
more traditional DMT (4,4¢-dimethoxytrityl) group for protection
of the secondary 5¢-hydroxy groups and equally efficient in the
subsequent automatic DNA-synthesis.8,15,17 The double bond was
Fig. 2 Average structures from MD-simulation at 298 K. (a) Duplex
with 3 incorporated in each strand in a (-3)-zipper arrangement.8 (b) The
corresponding duplex with 4.
Herein we report the synthesis and evaluation of 4 as a
homologue of 3 with the linker one carbon extended, in order
3984 | Org. Biomol. Chem., 2008, 6, 3983–3988
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
phase DNA-synthesis with 1H-tetrazole as the activator and a
prolonged coupling time for 13 of 20 minutes. The coupling yield
for 13 was >90% and standard acidic treatment after the coupling
also removed the pixyl group. The constitution and purity of
the ODNs were verified by MALDI-MS and ion exchange/RP-
HPLC, respectively. The ODN sequences applied in this study
(ON1–ON9) are shown in Table 1 and 2. The hybridisation
properties of the ODNs with different complementary DNA and
RNA sequences were evaluated by thermal stability experiments,
and the results from these experiments are further depicted in
Tables 1 and 2.
Firstly, the modified sequence ON3, containing one central
double-headed nucleoside 4, was studied and compared with
the unmodified sequences ON1 and ON2. The effect of the
modification was evaluated in duplexes, bulged duplexes and
three-way junctions formed with complementary DNA and RNA
(Table 1). The duplexes formed between ON3 and complementary
DNA and RNA were destabilised in both cases with DTm of
-5 ◦C and -2.5 ◦C, respectively, compared with unmodified ON1.
This indicates that incorporation of the additional nucleobase
in the minor groove is not favourable, but on the other hand is
reasonably well accepted in the duplex. This result is similar to the
previously reported results with the shorter linker in the double-
headed nucleoside 3.8
Scheme 2 Reagents and conditions: a, ref. 15: (i) DMP, DCM, 81%,
(ii) allyltrimethylsilane, BF3·OEt2, DCM, 75%; b, pixylchloride, pyridine,
96%; c, (i) OsO4, NaIO4 THF, (ii) NaBH4, THF, H2O, 53% over 2
steps; d, (i) TBz, PPh3, DEAD, (ii) TBAF, THF, (iii) NH3(aq), CH3OH,
80% over 3 steps; e, NC(CH2)2OP(N(iPr)2)2, diisopropylammonium
tetrazolide, DMF, CH3CN, 45%; TBDPS = tert-butyldiphenylsilyl, pixyl =
9-phenylxanthen-9-yl.
Subsequently, the hybridisation properties of ON3 with com-
plementary bulged DNA and RNA were studied. The penalty
for the incorporation of 4 is seen in all the formed duplexes as a
drop in Tm of around 4 ◦C, indicating that the bulges are willing to
accommodate the additional thymine in the same way as duplexes.
On the other hand, no positive influence is indicated by the
additional base in the bulged duplexes. It is clear when comparing
the Tm values of ON2 against the complementary strand forming a
regular duplex and ON3 with the same complement forming an A-
bulge that the large drop in Tm of 12.5 ◦C strongly indicates that the
double-headed nucleoside does not behave as a TT-dinucleotide.
This also indicates that the additional nucleobase is not likely
intercalated in the duplex core. Similar destabilisation is seen in
duplexes with GA bulges opposing the double-headed nucleoside.
For the three-way junction studies (Table 1), two different
sequences containing a hairpin-loop were used. Thus, the DNA
complement contains an additional CC-bulge as compared with
cleaved to give the primary alcohol 11 in 53% yield via a stan-
dard procedure consisting of dihydroxylation by OsO4, oxidative
cleavage by NaIO4 and reduction using NaBH4. Introduction of
the additional nucleobase was accomplished using the Mitsunobu
reaction with 3-N-benzoylthymine as the nucleophile, giving the
desired double-headed nucleoside. Subsequent treatment with
TBAF deprotected the 3¢-hydroxy group, and cleavage of the
benzoyl group was obtained by treatment with 25% aqueous
ammonia and methanol. Compound 12 was obtained in 80%
yield over the 3 steps and subsequently converted to the desired
phosphoramidite 13 using a standard protocol.
Preparation and evaluation of oligodeoxynucleotides
The double-headed nucleoside phosphoramidite 13 was incor-
porated into DNA sequences using standard automated solid
Table 1 Hybridisation data for duplexes, bulged duplexes and three-way junctionsa
5¢-GCT CAC XCT CCC A
X = T
(ON1)
X = TT
(ON2)
X = 4
(ON3)
DTm 1b
DTm 2c
DNA
RNA
3¢- d(CGA GTG AGA GGG T)
51.0
58.5
43.0
49.5
42.5
49.5
23.5
36.0
—
—
46.0
56.0
39.0
46.0
38.0
46.0
22.5
35.0
-5.0
-2.5
-4.0
-3.5
-4.5
-3.5
-1.0
-1.0
—
—
3¢- r(CGA GUG AGA GGG U)
DNA, A-bulge
RNA, A-bulge
DNA, GA-bulge
RNA, GA-bulge
DNA hairpin
RNA hairpin
3¢- d(CGA GTG AAG AGG GT)
52.0
58.5
42.0
48.0
25.0
36.5
-13.0
-12.5
-4.0
-2.0
-2.5
-1.5
3¢- r(CGA GUG AAG AGG GU)
3¢- d(CGA GTG AGA GAG GGT)
3¢- r(CGA GUG AGA GAG GGU)
3¢- d(CGA GTG ACC CGC GTT TTC GCG AGA GGG T)
3¢- r(CGA GUG ACG CGU UUU CGC GAG AGG GU)
a Melting temperatures (Tm values/◦C) obtained from the maxima of the first derivatives of the melting curves (A260 vs. temperature) recorded in a medium
salt buffer (Na2HPO4 (7.5 mM), NaCl (100 mM), EDTA (0.1 mM), pH 7.0) using 1.0 mM concentrations of each strand. b DTm1 = Tm (ON3) - Tm
(ON1), c DTm2 = Tm (ON3) - Tm (ON2).
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3983–3988 | 3985
©

View Article Online
Table 2 Hybridisation data for the DNA zipper motifsa
Duplex sequence
Tm
DTm
DTm3
Duplex sequence
Tm
DTm
-9.5
DDTm DTm3
DDTm3
5¢-CGC ATA TTC GC
3¢-GCG TAT AAG CG
45.5
ON7
ON4
5¢-CGC ATA TXC GC
3¢-GCG XAT AAG CG
36.0
+1.2
+2.7
0.0
-9.8
-3.8
0.0
5¢-CGC ATA TTC GC
3¢-GCG XAT AAG CG
39.0
40.0
33.5
-6.5
-5.5
-5.4
-4.7
ON8
ON4
5¢-CGC ATA XTC GC
3¢-GCG XAT AAG CG
37.5
35.8
35.5
31.0
31.3
31.0
29.8
24.5
-8.0
-9.7
+6.5
-0.9
-0.8
+7.0
-1.5
-1.7
+6.8
+6.1
ON4
ON5
ON6
5¢-CGC ATA TTC GC
3¢-GCG TAX AAG CG
ON7
ON5
5¢-CGC ATA TXC GC
3¢-GCG TAX AAG CG
-10.0
-10.4
-10.4
-18.2
-17.0
-9.0
5¢-CGC ATA TTC GC
3¢-GCG XAX AAG CG
-12.0
-10.9
ON8
ON5
5¢-CGC ATA XTC GC
3¢-GCG TAX AAG CG
-10.0
-14.5
-14.2
-14.5
-15.7
-21.0
-0.3
+1.0
+0.3
+1.7
+0.5
0.0
ON9
ON4
5¢-CGC AXA XTC GC
3¢-GCG XAT AAG CG
ON7
ON8
ON9
5¢-CGC ATA TXC GC
3¢-GCG TAT AAG CG
41.3
41.3
36.5
-4.2
-4.2
-9.0
-4.4
-4.9
ON9
ON5
5¢-CGC AXA XTC GC
3¢-GCG TAX AAG CG
5¢-CGC ATA XTC GC
3¢-GCG TAT AAG CG
ON7
ON6
5¢-CGC ATA TXC GC
3¢-GCG XAX AAG CG
5¢-CGC AXA XTC GC
3¢-GCG TAT AAG CG
-12.0
ON8
ON6
5¢-CGC ATA XTC GC
3¢-GCG XAX AAG CG
ON9
ON6
5¢-CGC AXA XTC GC
3¢-GCG XAX AAG CG
-16.8
a Melting temperatures (Tm values/◦C) obtained from the maxima of the first derivatives of the meltng curves (A260 vs. temperature) recorded in a
medium salt buffer (Na2HPO4 (7.5 mM), NaCl (100 mM), EDTA (0.1 mM), pH 7.0) using 1.0 mM concentrations of each strand. X corresponds to 4
by the incorporation of 13. DDTm values for duplex x : y are defined as DTm(x : y) - (DTm(x : ref) + DTm(y : ref)). DTm3 and DDTm3 correspond to the similar
measurements for sequences containing 3 taken from ref. 8.
the RNA-complement in order to form a more well-defined
secondary structure,10 as explained in our former study.8 The
hybridisation studies of ON3 show, however, that the double-
headed nucleoside 4 is not able to stabilise either of the two
three-way junctions when compared to the dinucleotide in ON2,
or the single thymidine in ON1, as small destabilisations of 1–
tends to be higher for 4 than for 3, whereas it tends to be opposite
when the complementary sequence is modified. Nevertheless, the
decrease in Tm is relatively uniform in all cases. Hereafter, the
zipper motifs in which the additional nucleobases are positioned
in the opposite sequences with distances of 0–3 interspacing base-
pairs (corresponding to (-1) to (-4)-zippers) were studied. The
earlier results with 3, shown in the right hand column of Table 2,
demonstrate the very selective thermally stable formation of (-3)-
zippers. Only in that specific positioning was a relative increase
in Tm (defined as DDTm) of 6–7 ◦C observed, most likely as a
result of a strong stacking interaction in the minor groove as
supported by modelling experiments (Fig. 2).8 In the case of 4,
the results are completely different, demonstrating only a much
smaller effect when positioned in a (-3)-zipper, as seen by a DDTm
of +2.7 ◦C in the case of ON8 : ON4, decreasing to +1.0 ◦C,
+0.5 C and even 0 C in ON9 : ON4, ON8 : ON6, and ON9 :
ON6, respectively. Therefore, although the initial modelling study
indicates the two extra bases to be stacked (Fig. 2), this particular
arrangement leads to significantly poorer interactions than seen
with the simple methylene linker arrangement. On the other hand,
a small increase is also seen with the (-4)-zipper in ON7 : ON4 and
ON7 : ON6, with a DDTm of +1.2 ◦C and +1.7 ◦C, respectively. In
all other cases, Tm indicated an additive destabilisation when 4 was
incorporated multiple times. These results demonstrate the unique
ability of 3 to position the additional nucleobases in the minor
groove in a position allowing efficient stacking in a (-3)-zipper
motif, and suggests that the extension of the linker in 4 induces
a flexibility that is not accepted. Due to the same flexibility, the
2
◦C are seen in all cases. Comparing this destabilisation with
the penalty observed with incorporation into duplexes, it is clear
that the additional base is better accommodated in the three-
way junctions, possibly due to the more flexible system. On the
other hand, no stabilisation is observed, and this indicates that
the linker extension on 4 as compared with 3 is not leading
the additional base in the DNA : DNA-complex into a contact
with the CC-bulge. The results with 4 are not different from the
results with 3,8 and this may also suggest that the isomerism
is different for the modelling system and the experimental
system.
In order to examine the possibility of minor groove base–base
interactions in an artificial DNA zipper motif, the incorporation of
4 into the same series of sequences as 3 was successfully performed.
These results are illustrated in Table 2 in combination with the
results from the previous study of 3 for direct comparison. Firstly,
single incorporations were studied with the sequences ON4–ON9
hybridised to their unmodified complements, and in all cases
decreases in melting temperatures of 4.2–6.5 ◦C were detected
with the expected additive effect of the two modifications in ON6
and ON9. When comparing the two modifications 4 and 3 (DTm
and DTm3, respectively), the thermal penalty paid in one sequence
◦
◦
3986 | Org. Biomol. Chem., 2008, 6, 3983–3988
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
indicated base–base contact in a (-4)-zipper is much smaller and
less conclusive.
Preparation of 5¢(S)-C-(2-hydroxyethyl)-5¢-O-(9-phenylxanthen-9-
yl)-3¢-O-(tert-butyldiphenylsilyl)thymidine (11)
To a solution of compound 1015 (2.00 g, 2.58 mmol) in THF
(20 mL) stirred at room temperature was added a solution of
NaIO4 (2.10 g, 9.79 mmol) in H2O (20 mL) and a 2.5% solution
of OsO4 in tert-butyl alcohol (0.26 mL, 26 mmol). The reaction
mixture was stirred at room temperature for 26 h. Ethylene glycol
(2 mL, 144 mmol) was added, and the resulting mixture was stirred
for 15 min whereupon H2O (40 mL) was added. The mixture was
extracted by CH2Cl2 (2 ¥ 50 mL), and the organic phase was
dried (Na2SO4) and concentrated under reduced pressure. The
residue was dissolved in a mixture of THF (20 mL) and H2O
(20 mL), and NaBH4 (195 mg, 5.16 mmol) was added. After being
stirred at room temperature for 1 h, the reaction mixture was
extracted with CH2Cl2 (2 ¥ 50 mL). The organic phase was washed
with a saturated aqueous solution of NaHCO3 (2 ¥ 50 mL), dried
(Na2SO4) and concentrated under reduced pressure. The residue
was purified by column chromatography (0–2.5% CH3OH and
0.1% pyridine in CH2Cl2) to give the product 11 as a white foam
(1.06 g, 53%): Rf 0.45 (50% ethyl acetate in petroleum ether). 1H
NMR (300 MHz, CDCl3) d 8.36 (br s, 1H, NH) 7.81 (d, 1H, J =
1.2 Hz, H-6), 7.60–7.10 (m, 17H, Ph), 7.05–6.80 (m, 4H, Ph), 6.70
(m, 1H, Ph), 6.45 (m, 1H, Ph), 6.35 (dd, 1H, J = 4.8, 9.3 Hz, H-1¢),
3.55–3.50 (m, 2H, H-3¢, H-5¢), 3.10–2.90 (m, 3H, H-4¢, H-7¢), 2.28
(dd, 1H, J = 4.8, 12.3 Hz, H-2¢), 2.01 (d, 3H, J = 1.2 Hz, CH3),
1.92 (m, 1H, H-2¢), 1.25 (m, 1H, H-6¢), 1.18 (m, 1H, OH), 0.98 (s,
9H, SiC(CH3)3) 0.75 (m, 1H, H-6¢); 13C NMR (75 MHz, CDCl3)
d 163.7 (C-4), 151.7, 151.6, 150.1, 147.3 (C-2, Ph), 135.9, 135.7,
133.5, 133.1, 131.7, 130.4, 130.0, 129.8, 128.1, 127.9, 127.6, 127.2,
123.5, 123.3, 123.2, 123.1, 116.7, 116.5 (C-6, Ph), 110.7 (C-5), 89.8,
85.7, 75.2, 71.3, 58.7 (C-1¢, C-3¢, C-4¢, C-5¢, C-7¢), 41.3 (C-2¢), 35.4
(C-6¢), 26.9 (C(CH3)3), 18.7 (CCH3)3), 12.8 (CH3). HiRes MALDI
MS m/z (M + Na+) found/calc. 803.3104/873.3128.
Conclusion
The synthesis of the double-headed nucleoside 5¢(S)-C-(2-
(thymin-1-yl)ethyl)thymidine 4 has been accomplished and its
incorporation into oligodeoxynucleotides was straightforward.
The relatively small thermal penalty paid by incorporating this
moiety and positioning the additional nucleobase on the surface
of a nucleic acid duplex showed that this is well accommodated
and available for recognition. On the other hand, no stabilisation
by the additional base in combination with the ethylene linker was
detected in the secondary structures studied. The data proved, first
of all, that the specific base–base contact in a DNA (-3)-zipper
positioning of the two opposite moieties of 3 is unique and lost by
increasing the linker length. Therefore, we are concentrating our
future studies on this motif by varying the additional nucleobase
moiety.
Experimental section
All commercial reagents were used as supplied. All reactions were
conducted under a nitrogen atmosphere. Column chromatography
was carried out using silica gel 60 (0.040–0.063 mm). HiRes
MALDI mass spectra were recorded on an Ionspec Ultima Fourier
Transform mass spectrometer using 2,5-dihydroxybenzoic acid as
a matrix. NMR spectra were recorded on a 300 MHz Varian spec-
trometer. The values for d are in ppm relative to tetramethylsilane
as an internal standard for the 1H NMR assignments, relative to
the solvent signal (DMSO-d6 39.52; CDCl3 77.16) for the 13C-
NMR assignments, and relative to 85% H3PO4 as an external
standard for the 31P NMR spectra. Assignments of NMR spectra
are based on 2D and/or DEPT spectra and follow nucleoside
naming style, i.e. the anomeric carbon is C-1¢. No distinction has
been made between the two thymine nucleobases.
Preparation of 5¢(S)-C-(2-(thymin-1-yl)ethyl)-5¢-O-(9-
phenylxanthen-9-yl)thymidine (12)
Preparation of 5¢-C-vinyl-5¢-oxo-3¢-O-(tert-
butyldimethylsilyl)thymidine (7)
A solution of 11 (1.00 g, 1.28 mmol), 3-N-benzoylthymine (649 mg,
2.82 mmol) and PPh3 (1.01 g, 3.85 mmol) in anhydrous THF
(30 mL) was stirred at 0 ◦C, and DEAD (1.56 mL, 3.59 mmol)
was added dropwise. The solution was stirred for 6 h at room
temperature and concentrated under reduced pressure. The residue
was diluted with ethyl acetate (50 mL) and washed with a
saturated aqueous solution of NaHCO3 (50 mL). The aqueous
layer was extracted with ethyl acetate (2 ¥ 20 mL), and the
combined organic phase was dried (MgSO4) and concentrated
under reduced pressure. The residue was dissolved in anhydrous
THF (30 mL) and a 1.0 M solution of tetrabutylammonium
fluoride in THF (2.20 mL) was added. The solution was stirred at
room temperature for 2 h and then partitioned between water
(50 mL) and ethyl acetate (50 mL). The aqueous phase was
extracted with ethyl acetate (2 ¥ 25 mL), and the combined
organic phase was washed with brine (50 mL), dried (Na2SO4) and
concentrated under reduced pressure. The residue was dissolved
in methanol (30 mL) and a 25% aqueous solution of ammonia
(3.1 mL) was added dropwise. The reaction mixture was stirred for
16 h at room temperature, concentrated under reduced pressure
and purified by silica gel column chromatography (1–9% CH3OH
A solution of compound 6 (100 mg, 0.26 mmol) in anhydrous
CH2Cl2 was stirred at room temperature. Dess–Martin periodi-
nane (128 mg, 0.30 mmol) was added and the mixture was stirred
for 20 h. The reaction mixture was diluted with CH2Cl2 (20 mL)
and washed with a saturated aqueous solution of Na2S2O3 (10 mL)
and a saturated aqueous solution of NaHCO3 (10 mL), dried
(Na2SO4) and concentrated under reduced pressure. The residue
was purified by column chromatography (0–80% ethyl acetate in
petroleum ether) to give 7 as a white foam (76 mg, 76%). Rf 0.71
(75% ethyl acetate in petroleum ether). 1H NMR (CDCl3): d 9.03
(1H, br s, NH), 8.09 (1H, d, J = 1.2, H6), 6.59–6.42 (3H, m,
H1¢, H7¢), 6.02 (1H, dd, J = 1.5, 9.6, H6¢), 4.82 (1H, d, J =
1.8, H4¢), 4.48 (1H, m, H3¢), 2.17–1.95 (2H, m, H2¢), 1.98 (3H, s,
CH3), 0.94–0.85 (9H, m, C(CH3)3), 0.14–0.07 (6H, m, SiCH3). 13
C
NMR (CDCl3): d 197.0 (C5¢), 164.1 (C4), 150.5 (C2), 136.4 (C6),
133.0, 131.3 (C6¢, C7¢), 111.2 (C5), 88.9 (C1¢), 86.2 (C4¢), 74.1
(C3¢), 39.8 (C2¢), 25.6 (C(CH3)3), 17.9 (CCH3)3), 12.6 (CH3), -4.6,
-4.8 (SiCH3). HiRes MALDI MS m/z (M + Na+) found/calc.
403.1667/403.1660.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3983–3988 | 3987
©

View Article Online
and 0.1% pyridine in CH2Cl2) to give the product 12 as a white
foam (663 mg, 80%): Rf 0.15 (7.5% CH3OH in ethyl acetate). 1H
NMR (300 MHz, DMSO-d6) d 11.32 (br s, 1H, NH), 11.12 (br s,
1H, NH), 7.81 (d, 1H, J = 1.2 Hz, H-6), 7.40–6.94 (m, 14H, Ph,
H-6), 6.00 (dd, 1H, J = 6.6, 7.2 Hz, H-1¢), 5.08 (d, 1H, J = 4.5 Hz,
OH), 3.76 (m, 1H, H-3¢), 3.63 (t, 1H, J = 3.6 Hz, H-4¢), 3.50–3.15
(m, 3H, H-5¢, H-7¢), 2.05–1.90 (m, 2H, H-2¢), 1.78 (s, 3H, CH3),
1.72 (d, 3H, J = 1.2 Hz, CH3), 1.40–1.25 (m, 2H, H-6¢); 13C NMR
(75 MHz, DMSO-d6) d 164.2 (C-4), 163.73 (C-4), 150.9, 150.7,
150.5, 150.3, 148.0 (C-2, Ph), 137.7, 135.6, 131.2, 131.1, 129.8,
129.7, 127.7, 127.0, 126.8, 123.4, 123.2, 123.1, 122.5, 116.3, 115.9
(C-6, Ph), 109.7, 108.5 (C-5), 86.5, 83.1, 76.2, 71.1, 70.3 (C-1¢,
C-3¢, C-4¢, C-5¢, CPh3), 54.9 (C-7¢), 43.5 (C-2¢), 30.8 (C-6¢), 12.3
(CH3), 11.9 (CH3). HiRes ESI MS m/z (M + Na+) found/calc.
673.2269/673.2274.
buffer, 38 min; 70–100% buffer, 7 min; 100% buffer, 10 min.
All fractions containing 5¢-O-DMT protected oligonucleotide
(retention time 20–30 minutes) were collected and concentrated.
The products were detritylated by treatment with an 80% aqueous
solution of acetic acid for 20 min, and finally isolated by
precipitation with ethanol at -18 ◦C overnight. MALDI-MS m/z
(found/calcd); ON3 (3979.1/3982.6), ON4 (3532.3/3533.4), ON5
(3530.2/3533.4), ON6 (3682.4/3685.4), ON7 (3442.8/3444.3),
ON8 (3444.3/3444.3), ON9 (3594.4/3596.3). After dissolution
in double distilled water, the concentrations were determined
spectrometrically at 260 nm in the pH 7.0 buffer, assuming an
extinction coefficient for the modified double-headed nucleoside
of equal to twice that of thymidine. The UV-melting curves were
acquired at 260 nm with a scan rate of 0.2 ◦C min-1 cycled between
15 and 75 ◦C after denaturation at 80 ◦C. All melting curves were
found to be reversible. The melting temperatures were taken as
the first derivatives of the absorbance versus temperature up-curve
and reported as the average of two measurements.
Preparation of 5¢(S)-C-(2-(thymin-1-yl)ethyl)-5¢-O-
(9-phenylxanthen-9-yl)-3¢-O-(P-b-cyanoethoxy-N,N -
diisopropylaminophosphinyl)thymidine (13)
Modelling experiment
A solution of compound 12 (500 mg, 0.923 mmol) and diiso-
propylammonium tetrazolide (236 mg, 1.39 mmol) in anhydrous
DMF (10 mL) and anhydrous CH3CN (10 mL) was stirred at
room temperature and N,N,N ¢,N¢-tetraisopropylphosphoramidite
(0.44 mL, 1.38 mmol) was added. The reaction mixture was stirred
for 10 h and diluted with ethyl acetate (20 mL). The solution
was washed with brine (2 ¥ 20 mL). The combined aqueous
phase was extracted with ethyl acetate (20 mL), and the combined
organic phase was dried (Na2SO4) and concentrated under reduced
pressure. The residue was purified by column chromatography (1–
6% CH3OH and 0.1% pyridine in CH2Cl2) to give the product 13
as a white foam (291 mg, 45%): Rf 0.25 (7.5% CH3OH in ethyl
acetate). 31P NMR (121.5 MHz, CDCl3) d 150.26, 149.88. HiRes
ESI MS m/z (M + Na+) found/calc. 873.3347/873.3353.
The MD-simulations in Fig. 2b were obtained by the same
protocol as used in ref. 8.
Acknowledgements
The Danish National Research Foundation, The Danish Research
Agency and Møllerens Fond are thanked for financial support.
Mrs Birthe Haack is thanked for technical assistance.
References and Notes
1 H. Yan, Science, 2004, 306, 2048.
2 J. Wengel, Org. Biomol. Chem., 2004, 2, 277.
3 P. Mouritzen, A. Toftgaard Nielsen, H. M. Pfundheller, Y. Choleva, L.
Kongsbak and S. Møller, Expert Rev. Mol. Diagn., 2003, 3, 89.
4 J. Kurreck, Eur. J. Biochem., 2003, 270, 1628.
5 T. Wu, M. Froeyen, G. Schepers, K. Mullens, J. Rozenski, R. Busson,
A. Van Aerscot and P. Herdewijn, Org. Lett., 2004, 6, 51.
6 S. L. Pedersen and P. Nielsen, Org. Biomol. Chem., 2005, 3, 3570.
7 T. Wu, K. Nauwelaerts, A. Van Aerschot, M. Froeyen, E. Lescrinier
and P. Herdewijn, J. Org. Chem., 2006, 71, 5423.
8 M. S. Christensen, C. M. Madsen and P. Nielsen, Org. Biomol. Chem.,
2007, 5, 1586.
9 M. S. Christensen, A. D. Bond and P. Nielsen, Org. Biomol. Chem.,
2008, 6, 81.
10 N. B. Leontis, W. Kwok and J. S. Newman, Nucleic Acids Res., 1991,
19, 759.
11 H. B. Lazrek, H. Kha¨ıder, A. Rachdi, J.-L. Barascut and J.-L. Imbach,
Tetrahedron Lett., 1996, 37, 4701.
12 S. Guaillarme, S. Legoupy, A.-M. Aubertin, C. Olicard, N. Bour-
gougnon and F. Huet, Tetrahedron, 2003, 59, 2177.
13 J.-M. Xu, C. Qian, B.-K. Liu, Q. Wu and X.-F. Lin, Tetrahedron, 2007,
63, 986.
14 A. M. Sørensen, K. E. Nielsen, B. Vogg, J. P. Jacobsen and P. Nielsen,
Tetrahedron, 2001, 57, 10191.
Oligonucleotide synthesis and hybridisation experiments
The oligodeoxynucleotides were synthesised using an automated
Expedite 8909 nucleic acid synthesis system following the phos-
phoramidite approach. Synthesis of oligonucleotides ON3–ON9
was performed on a 0.2 mmol scale by using 2-cyanoethyl phos-
phoramidites of standard 2¢-deoxynucleosides in combination
with the modified phosphoramidite 13. The synthesis followed
the regular protocol employing standard CPG supports and 4,5-
dicyanoimidazole as the activator except in the case of the modified
amidites, which were manually coupled using 0.05 M amidite and
0.5 M tetrazole as the activator in CH3CN for 20 min. The coupling
yields for 13 in combination with the following unmodified
amidites were in the range of 91–100%. The 5¢-O-DMT oligonu-
cleotides were removed from the solid support by treatment with
concentrated aqueous ammonia at 55 ◦C for 16–24 h, which also
removed the protecting groups. The oligonucleotides were purified
by reversed-phase HPLC on a Waters 600 system using an Xterra
prep MS C18; 10 mm; 7.8 ¥ 150 mm column; gradient of buffer
(0.05M triethylammonium acetate) in 75% CH3CN(aq): 0–70%
15 P. K. Sharma, B. H. Mikkelsen, M. S. Christensen, K. E. Nielsen,
C. Kirchhoff, S. L. Pedersen, A. M. Sørensen, K. Østergaard and P.
Nielsen, Org. Biomol. Chem., 2006, 4, 2433.
16 V. Banuls and J.-M. Escudier, Tetrahedron, 1999, 55, 5831.
17 H. Trafelet, E. Stulz and C. Leumann, Helv. Chim. Acta, 2001, 84, 87.
3988 | Org. Biomol. Chem., 2008, 6, 3983–3988
This journal is
The Royal Society of Chemistry 2008
©