Duplex and triplex formation of mixed pyrimidine oligonucleotides with stacking of phenyl-triazole moieties in the major groove

Article abstract of DOI:10.1021/jo200919y

5-(1-Phenyl-1,2,3-triazol-4-yl)-2′-deoxycytidine was synthesized from a modified CuAAC protocol and incorporated into mixed pyrimidine oligonucleotide sequences together with the corresponding 5-(1-phenyl-1,2,3- triazol-4-yl)-2′-deoxyuridine. With consecutive incorporations of the two modified nucleosides, improved duplex formation with a complementary RNA and improved triplex formation with a complementary DNA duplex were observed. The improvement is due to π-π stacking of the phenyl-triazole moieties in the major groove. The strongest stacking and most pronounced positive influence on thermal stability was found in between the uridine analogues or with the cytidine analogue placed in the 3′ direction to the uridine analogue. Modeling indicated a different orientation of the phenyl-triazole moieties in the major groove to account for the difference between the two nucleotides. The modified oligonucleotides were all found to be significantly stabilized toward nucleolytic degration.

Full text of DOI:10.1021/jo200919y

ARTICLE
pubs.acs.org/joc
Duplex and Triplex Formation of Mixed Pyrimidine Oligonucleotides
with Stacking of Phenyl-triazole Moieties in the Major Groove
Nicolai Krog Andersen,† Holger Døssing,‡ Frank Jensen,§ Birte Vester,‡ and Poul Nielsen†,*
†Nucleic Acid Center, Department of Physics and Chemistry and ‡Department of Biochemistry and Molecular Biology,
University of Southern Denmark, 5230 Odense M, Denmark
§Department of Chemistry, Århus University, Langelandsgade, 8000 Århus, Denmark
S Supporting Information
b
ABSTRACT: 5-(1-Phenyl-1,2,3-triazol-4-yl)-20-deoxycytidine was
synthesized from a modified CuAAC protocol and incorporated
into mixed pyrimidine oligonucleotide sequences together with the
corresponding 5-(1-phenyl-1,2,3-triazol-4-yl)-20-deoxyuridine. With
consecutive incorporations of the two modified nucleosides, im-
proved duplex formation with a complementary RNA and improved
triplex formation with a complementary DNA duplex were observed.
The improvement is due to πꢀπ stacking of the phenyl-triazole
moieties in the major groove. The strongest stacking and most
pronounced positive influence on thermal stability was found in
between the uridine analogues or with the cytidine analogue placed
in the 30 direction to the uridine analogue. Modeling indicated a
different orientation of the phenyl-triazole moieties in the major
groove to account for the difference between the two nucleotides. The modified oligonucleotides were all found to be significantly
stabilized toward nucleolytic degration.
’ INTRODUCTION
continuous π-system. Two consecutive incorporations of 1 into the
oligonucleotide were necessary to give the positive effect, which
increased even further with three or four consecutive incorporations
proving that πꢀπ stacking of the aromatic substituents is driving the
increase in duplex stability.10,12 Similar or slightly increased melting
temperatures were obtained by the more hydrophilic phenol or
benzensulfonamide substituents (1, R = OH or SO2NH2).12 The
most important limitation to the concept of obtaining improved RNA
recognition by the simple stacking building block 1 is the sequence
context demanding a stretch of at least two 20-deoxyuridine derivatives
in the antisense oligonucleotide. Herein we meet this challenge by the
introduction of the cytidine derivative 2 (Figure 1).13,14 This nucleo-
side building block can be expected to show a different conforma-
tional behavior, and therefore modeling is applied in order to
understand the sequence specific behavior of stacking between 1
and 2. Furthermore, the potential of 1 and 2 in triplex-forming
oligonucleotides, as well as the stability of the oligonucleotides
toward nucleolytic degradation, are investigated.
The idea of regulating gene expression by chemically modified
oligonucleotides has been pursued for three decades.1,2 The ther-
apeutic potential has been formulated in the so-called antisense2,3
and antigene strategies,4 where the molecular target is RNA and
genomic double stranded DNA, respectively. The chemically
modified oligonucleotides should be physiologically stable and form
thermally strong duplexes with RNA or triplexes with DNA.3,4
A
significant amount of chemical modifications have been presented,5
and among the most successful are 20-modified nucleotides, con-
formationally restricted carbohydrate moieties, as in LNA,6 and
base-modified analogues.7 As a prime example of the latter, nucleo-
bases with extended ring systems have been found to give very strong
duplexes due to increased πꢀπ stacking. Especially the phenoxazine
replacement of the cytosine is a good example8 leading to an increase
in thermal stability of a DNA:RNA duplex with up to 5 °C for one
incorporation, but also 5-modified (for instance 5-propynyl-) pyr-
imidine nucleosides have been shown to increase the duplex stability.9
Recently, we demonstrated that simple substituted triazoles intro-
duced in the 5-position of 20-deoxyuridine lead to increased melting
temperatures of DNA:RNA duplexes.10 These nucleosides (like 1,
Figure 1) are easily obtained by using the CuAAC (Cu-catalyzed
alkyn-azide cycloaddition) reaction11 and can be introduced into
oligonucleotides after making the appropriate phosphoramidite build-
ing block in only four synthetic steps.10 The most pronounced results
were obtained with a phenyl-substituent (1,R=H) andherebyalarge
’ RESULTS
Modeling of the TriazoleꢀCytosine Bond. In our former
study, ab initio calculations on a simple model compound
Received: May 9, 2011
Published: June 21, 2011
r
2011 American Chemical Society
6177
dx.doi.org/10.1021/jo200919y J. Org. Chem. 2011, 76, 6177–6187
|

The Journal of Organic Chemistry
ARTICLE
Figure 1. The reason is probably repulsion between the
nitrogen lone pairs possibly in combination with a CHꢀN
hydrogen bond interaction. The tautomers A and B display
similarly shaped rotational profiles, with tautomer B being
19 kJ/mol higher in energy than A. The global minimum for
tautomer B is a nonplanar anti conformation with ω ≈ 150°,
whereas A has a broad global minimum at ω = 0° (essentially
flat energy surface ꢀ30° to 30°) corresponding to the con-
formation drawn for 2 in Figure 1. This is probably stabilized
by an internal (triazole) NꢀHN (cytosine) hydrogen bond.
A nonplanar anti conformation with ω ≈ 150° is a local mini-
mum. These results strongly indicate that the cytidine ana-
logue 2 will prefer the opposite coplanar conformation as
compared to the uridine analogue 1 (Figure 1). This will position
the phenyl group of 2 as compared to 1 with a different orientation
in the major groove. A recent X-ray crystal study on 2 by Hudson
and co-workers also revealed the same nearly coplanar conformation.14
Chemical Synthesis. The synthesis of the cytosine derivative
2 was naturally inspired by the convenient synthesis of 1.10
However, a protection of the exocyclic amine was necessary.
After several strategies had been considered and approached,13
we succeeded in synthesizing the 4-N-acetylated phosphorami-
dite in a series of reactions starting from 5-iodo-20-deoxycytidine
followed by a Sonogashira coupling, cycloaddition, 4-N-protec-
tion, 50-O-tritylation, and 30-O-phosphitylation (Scheme 1).
Hence, the known 5-ethynyl-20-deoxycytidine, 4,14,16 was made
by a slightly modified procedure using an efficient Sonogashira
reaction on unprotected 5-iodo-20-deoxycytidine, 3, and then
reacted in our formerly published in situ azidation/CuAAC
protocol,10 affording 5-(4-phenyl-1,2,3-triazol-1-yl)-20-deoxycy-
tidine, 2, in a good yield. Peracetylation and subsequent selective
hydrolysis of the acetyl esters afforded the 4-N-acetylated
derivative 5. Protection of the 50-O-position with the DMT
(4,40-dimethoxytrityl) group gave compound 6, and subsequent
phosphitylation afforded the phosphoramidite 7, which can be
used in standard automated solid-phase oligonucleotide synthesis.
The phosphoramidite 7 was incorporated in a series of
oligonucleotide sequences in combination with the corresponding
Figure 1. 5-(Phenyltriazole)pyrimidine nucleosides 1 and 2 (sketched
in their preferred conformations).
(1,3-dimethyl-5-(1-methyl-1,2,3-triazole-4-yl)uracil) showed
a strong preference for a coplanar organization of the triazole-
uracil in 1 with the triazole-CH pointing toward O4 of the
uracil (Figure 1) probably through a CHꢀO hydrogen bond-
ing interaction and an N/O lone pair repulsion.10 We ratio-
nalized, however, that this conformation might be unlikely for
2 and that the triazole might induce a different tautomeric
preference for 2 as compared to natural cytitidine analogues.
Therefore, a similar series of ab initio calculations and sub-
sequent conformational analysis was performed on three
tautomers of a simple dimethylated 5-(1,2,3-triazole-4-yl)cytosine
derivative (Figure 2). The torsional energy profiles were
obtained by single point MP2/aug-cc-pVDZ energies on
MP2/6-31G(d,p) geometry relaxed scans15 as a function of
the angle ω. The profiles are shown in Figure 2 and show that
the tautomer A is lower in minimal energy than tautomer B
and C with 19 and 12 kJ/mol, respectively. Tautomer C is
similar to the corresponding uracil analogue, and the rota-
tional profile is almost identical to the one found for trimethy-
lated 5-(1,2,3-triazole-4-yl)uracil10 with a deep global minimum
at 180° corresponding to the anti conformation shown for 1 in
Figure 2. Ab initio calculations of rotational profiles for three tautomers of a model compound of 2. Left, the tautomers with the definition on the
torsional angle ω. The structures drawn corresponds to ω = 0°. Right, torsional rotation profile for each tautomer.
6178
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
phosphoramidite of 110 and unmodified 20-deoxynucleotide
phosphoramidites (Table 1). The standard sequence chosen
for the present study was a 16-mer mixed pyrimidine sequence,
which has been previously studied as a triplex-forming oligonu-
cleotide (TFO).17 A pyrimidine sequence was preferred in order
to allow for a range of combinations of 1 and 2, and we expect this
sequence to give a representative picture of the full sequence
potential of these modifications. Hence, the modified phos-
phoramidites were introduced in 7 modified oligonucleotides
combining 1ꢀ4 consecutive incorporations of 1 and 2 in
different combinations in order to study the effect of stacking
in different sequence contexts in duplexes formed with com-
plementary DNA and RNA as well as in triplexes formed with a
double stranded DNA-target.
Hybridization Studies. The modified oligonucleotide se-
quences of the present study, ON2ꢀ8, represent a series of
replacements of T and dC moieties in the corresponding
unmodified sequence, ON1, with the modified nucleosides 1
and 2 (represented as the monomers X and Y, respectively,
Table 1). ON2 and ON3 contain single incorporations of either
1 or 2, whereas ON4ꢀ6 contains mixed consecutive incorpora-
tions of 1 or 2 and ON7ꢀ8 four consecutive incorporations of
either 1 or 2, respectively. All of these ONs were mixed with
complementary DNA and RNA as well as with a DNA duplex,
and the melting temperatures of the duplexes or triplexes formed
were determined by UV melting experiments (Table 1).
Scheme 1a
When mixed with complementary DNA, the single modified
ON2 displayed a Tm that is decreased by 1.3 °C. This is coherent
with the larger decreases in Tm (3ꢀ5 °C) observed with single
incorporations of 1 centrally in a shorter 9-mer sequence (the
underlined positions of 50-dGTG TTT TGC).10,12 It can be
expected that a single modification has a larger relative impact on
stability in a shorter sequence. With the single incorporation of
the cytidine analogue 2 in ON3, a slightly larger decrease in Tm
with complementary DNA (ΔTm = ꢀ1.9 °C) was observed.
With a consecutive introduction of 1 and 2 in ON4 and ON5,
only slightly smaller decreases per modification were seen (ΔTm
= ꢀ0.6 and ꢀ1.3 °C). This is consistent with our former study of
two consecutive incorporations of 1 in a 9-mer sequence (ΔTm =
ꢀ1.5 °C).12 On the other hand, the three consecutive mixed
stacking moieties in ON6 lead to an increase in duplex stability
(ΔTm = +1.6 °C), which is in contrast to a destabilization by three
consecutive incorporations of 1 in a 9-mer (ΔTm = ꢀ1.0 °C).12
The four incorporations of 1 (ON7) or of 2 (ON8) both display a
smaller increase in duplex stability (ΔTm = +0.6 and +0.8 °C),
which corresponds well to our former study of four incorporations
of 1 in a shorter 9-mer sequence (ΔTm = ꢀ0.3 °C).10 In general,
the effects of introducing 1 and 2 in DNA:DNA duplexes are small
a Reagents and conditions: (a) (i) TMS-CtCH, Pd(PPh3)2Cl2, CuI,
Et3N, DMF; (ii) NH3 (aq), 95%. (b) (i) PhBr, NaN3, N,N-dimethy-
lethylenediamine, CuI, EtOH/H2O, MW 100°C; (ii) 3, TBTA, CuI, Na
ascorbate, 80%. (c) (i) Ac2O, pyridine; (ii) 2 M NaOH, 1,4-dioxane/
H2O, 47%. (d) DMT-Cl, pyridine, 67%. (e) Diisopropylammonium
tetrazolide, NC(CH2)2OP(N(i-Pr)2)2, (ClCH2)2, 85%. TBTA = tris-
((1-benzyl-1,2,3-triazol-4-yl)methyl)amine.
Table 1. Thermal stability data of modified duplexes and triplexes
complementary DNAb
Tm(ΔTm/mod), °Cd
complementary RNAc
Tm(ΔTm/mod), °Cd
complementary DNA duplexe
ON sequencesa
Tm(ΔTm/mod), °C f
ON1
ON2
ON3
ON4
ON5
ON6
ON7
ON8
50-dTTT TCT TTT CCC CCC T
50-dTTT TCT TTX CCC CCC T
50-dTTT TCT TTT YCC CCC T
50-dTTT TYX TTT CCC CCC T
50-dTTT XYT TTT CCC CCC T
50-dTTT XYX TTT CCC CCC T
50-dTTT TCX XXX CCC CCC T
50-dTTT TCT TTT YYY YCC T
51.9
62.7
28.0
50.6 (ꢀ1.3)
50.0 (ꢀ1.9)
50.8 (ꢀ0.6)
49.4 (ꢀ1.3)
55.0 (+1.6)
55.0 (+0.8)
54.3 (+0.6)
64.5 (+1.8)
60.2 (ꢀ2.5)
66.7 (+2.0)
68.7 (+3.0)
70.9 (+2.7)
76.0 (+3.3)
67.6 (+1.2)
26.1 (ꢀ1.9)
23.9 (ꢀ4.1)
24.8 (ꢀ1.6)
25.7 (ꢀ1.2)
30.1 (+0.7)
35.5 (+1.9)
ntg
a Oligodeoxynucleotide sequences with Y = 2 and X = 1 corresponding to the incorporation of 7 and of the phosphoramidite of 1,10 respectively.
c
b Complementary DNA: 50-dAGG GGG GAA AAG AAA A. Complementary RNA: 50-rAGG GGG GAA AAG AAA A. d Melting temperatures
obtained from the maxima of the first derivatives of the melting curves (A260 vs temperature) recorded in a buffer containing 2.5 mM Na2HPO4, 5.0 mM
NaH2PO4, 100 mM NaCl, 0.1 mM EDTA, pH 7.0 using 1.5 μM concentrations of each strand. Values in brackets show the changes in Tm values per
modification compared with the reference strand. e Target DNA duplex: 50-dGGT GAA AAA TTT TCT TTT CCC CCC TGA CC; 30-dCCA CTT
TTT AAA AGA AAA GGG GGG ACT GG. f Melting temperatures of the duplex to triplex transition obtained from the maxima of the first derivatives of
the melting curves (A260 vs temperature) recorded in a buffer containing 10 mM sodium cacodylate, 150 mM NaCl and 10 mM MgCl2, pH = 6.0 using
1.5 μM concentration of the TFO and 1.0 μM concentration of the duplex. g No transition detected above 10 °C.
6179
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
Figure 3. CD spectra of the modified DNA:DNA duplexes.
concerning the duplex stability, and no conclusive differences
between 1 and 2 are seen.
modifications in ON6, some stabilization of the triplex was
demonstrated (ΔTm = +0.7 °C). Interestingly, four consecutive
incorporations of 1 increased the thermal stability significantly
with 7.5 °C (ΔTm = +1.9 °C/mod), whereas ON8 with four
consecutive incorporations of 2 did not form a detectable triplex.
All melting temperatures were determined at pH = 6.0, as no
triplex-to-duplex transition was observed above 10 °C in a pH 7.0
buffer for neither the unmodified ON1 nor any of the modified
ON2ꢀ8.
When mixed with complementary RNA, the modified oligo-
nucleotides revealed more interesting results. Hence, ON2 with
one incorporation of 1 demonstrated a surprising stabilization of
the DNA:RNA duplex (ΔTm = +1.8 °C), which is contrary to
similar single incorporations in the 9-mer sequence (ΔTm
=
ꢀ2.0 °C),10 indicating a dependence on the neighboring un-
modified nucleosides. This is also demonstrated by the incor-
poration of the cytidine analogue 2 in ON3 exhibiting a decrease
in Tm (ΔTm = ꢀ2.5 °C). As expected from the former study,12
consecutive incorporations of stacking phenyl-triazoles by 1 and
2 in ON4ꢀ6 markedly increased the melting temperatures with
ΔTm's of 2.0ꢀ3.0 °C for each modification, which is fully
comparable to the results of 2ꢀ3 incorporations of 1 in the
shorter 9-mer sequences (ΔTm = +3.0ꢀ4.7 °C).12 The differ-
ence between ON4 and ON5 might indicate that a 50-XY motif is
slightly better tolerated than a 50-YX motif probably due to
different stacking proportions. Four consecutive incorporations
of 1 in ON7 have a significantly more pronounced stabilizing
effect than the four consecutive incorporations of 2 in ON8
(ΔTm = +3.3 and +1.2 °C per modification, respectively), compar-
able to the ΔTm of +5.0 °C observed for each modification with four
consecutive incorporations of 1 in a shorter 9-mer sequence.10
Finally, we tested the ability of ON2ꢀ8 as triplex forming
oligonucleotides (TFOs) by mixing these with a 29-mer dsDNA
target containing a 16-nucleotide-long stretch of purines known
as the HIV-1 polypurine tract.17 The target DNA duplex has a Tm
of 69 °C, and the triplex-to-duplex transition with the unmodified
ON1 was detected at 28 °C. Single incorporations of 1 and 2 led
to decreases in melting temperatures of the triplex-to-duplex
transition of 1.9 and 4.1 °C in ON2 and ON3, respectively,
showing that the 20-deoxycytidine analogue 2 in the triplex, as
well as in the duplexes, is not as well accommodated as the 20-
deoxyuridine analogue 1. Two consecutive incorporations of 1
and 2 in ON4ꢀ5 did not change the picture as destabilizations
were observed (ΔTm = ꢀ1.2 and ꢀ1.6 °C), but with the three
CD Spectroscopy. In order to study structural properties,
circular dichroism (CD) spectra were studied for all duplexes and
triplexes formed with ON1ꢀ8. It is well-known that different
duplex types reveal different CD spectra.18 Hence, A-type duplexes
give an intense negative band at ∼210 nm and a positive band
at ∼260 nm, whereas B-type duplexes give a negative band at
∼250 nm and positive bands at ∼220 and ∼280 nm. This can be
connected to the fact that DNA:DNA duplexes adopt a B-type
form in solution, whereas RNA:RNA duplexes adopt an A-type
and DNA:RNA duplexes intermediate A/B-type structures. In
Figure 3, the unmodified DNA:DNA duplex formed by ON1
shows a CD spectrum that is characteristic for a B-type duplex,
whereas the corresponding unmodified ON1:RNA duplex shows
a spectrum for an intermediate A/B duplex type with an
increased positive band at ∼260 nm and an increased negative
band at ∼210 nm. The modified DNA:DNA duplexes of the
present study (Figure 3) demonstrate generally the retention of a
B-type duplex with only slightly increased bands at ∼260 nm.
However, ON7 with the four incorporations of 1 demonstrates a
more pronounced increase in this band as compared to ON8
with four incorporations of 2. This indicates that the uridine
analogue has a larger impact on duplex structure than the cytidine
analogue. A similar difference can be seen for the single incor-
porations in ON2 and ON3, showing a larger increase in the
260 nm band for the former. Less clear are the CD spectra of the
duplexes formed by ON4-6 with DNA, where ON4 shows a
somewhat smaller change in the CD spectrum as compared to
the other two sequences indicating that 50-XY and 50-YX have
6180
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
Figure 4. CD spectra of the modified DNA:RNA duplexes.
different impacts on the duplex. Figure 4 shows the CD spectra
formed by the modified oligonucleotides ON2ꢀ8 with comple-
mentary RNA, and all of the duplex spectra are similar to the
unmodified DNA:RNA duplex formed by ON1. However,
especially ON8 shows some deviation in the form of a much
smaller positive band at ∼260 nm and a larger negative band at
∼250 nm. Compared to ON7, this indicates again a different
change in structure for 2 as compared to 1 in accordance with the
thermal stability data (Table 1). CD spectra for the triplexes
formed by ON1ꢀ8 were also studied, but only small variations
were seen and no results leading to conclusions on structural
impact were obtained (see Supporting Information for the spectra).
Molecular Modeling. Molecular dynamics simulations were
applied to study the DNA:RNA duplexes and the triplexes of the
current study. The DNA:RNA hybrid duplexes formed by the
oligonucleotides with two and four consecutive incorporations of
1 and 2 (ON4ꢀ5 and ON7ꢀ8) as well the two triplexes formed
by ON7ꢀ8 were built and/or modified within the MacroModel
suite of programs19,20 and studied in a molecular dynamics
simulation. The all atom AMBER* force field is modified as
described by Pꢀerez et al.21 and using the GB/SA solvation model
as implemented in the AMBER software suite.22 The initial
DNA:RNA hybrid structures were built in the B-type duplex
conformation, and the incorporated monomers were subjected
to a Monte Carlo conformational search23 verifying the anti
conformation of the torsional angle or the uracil-analogue 1 and
the syn conformation of ω for the cytosine-analogue 2 as found
via ab initio calcultations (se above, Figures 1 and 2). The
obtained lowest energy structure was then subjected to a 5 ns
molecular dynamics simulation during which 500 structures were
sampled. These 500 structures were subsequently minimized,
and the local minimum structure obtained was used for further
analysis. Models of the resulting modified duplexes are shown in
Figure 5. The four modified DNA:RNA hybrid duplexes were all
found to be A/B-type duplexes with the substituents pointing
into the major groove. In all cases, a significant stacking was
observed between both triazoles and phenyl moieties. However,
noteworthy differences can be observed reflecting the different
preferred conformations of the torsional angle ω for the two
monomers as seen in Figures 1 and 2. The top view of the ON4:
RNA duplex looking from the 50-end of the modified sequence
shows the cytosine monomer 2 clearly positioned on the top of
the uracil monomer 1, and in the ON5:RNA duplex the situation
is opposite (Figure 5a and b). The stacking between the two
monomers, however, is clearly different due to the phenyl group
of 2 pointing away from the duplex and the phenyl group of 1
positioned toward the core of the duplex. The total overlap
between the two triazole-phenyl substituents seems somewhat
larger in the case of the 50-YX sequence in the ON4:RNA duplex
than with the 50-XY sequence in the ON5:RNA duplex. This
observation seems in contrast to the thermal stability of the two
duplexes with the latter being slightly more stable (Table 1).
However, the relation between stacking energy and aromatic
overlap is not direct, and the difference in Tm between the two
duplexes is after all only 2 °C. It is also noted that the internal
overlap of the two triazole rings alone seems larger in the ON5:
RNA duplex. The differences are more clear between the two
duplexes with four incorporations of either 1 or 2, i.e., the ON7:
RNA and the ON8.RNA duplexes, respectively (Figure 5c and d).
In both cases a perfect stacking orientation of the aromatic
moieties were observed, but from the top-views of the two
duplexes, the different orientation of the two monomers can be
seen. The phenyl group of 2 is again pointing out into the major
groove, whereas the phenyl group of 1 is pointing into the duplex
core aligning with the nucleobases (as seen in our former study of
this monomer in shorter sequences).10,12 This gives a much
larger overlap and subsequent stacking of the phenyl rings with
the four consecutive 1 monomers than with the four 2 mono-
mers. This result is in line with the thermal denaturation data on
ON7 showing a significantly stronger duplex with the RNA
complement than ON8 (Table 1).
The models of the two modified triplex structures formed by
ON7 and ON8 are shown in Figure 6.20 In the first triplex, the
four monomers 1 in ON7 are clearly showing a coplanar stacking
6181
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
Figure 5. Modeling structures of modified DNA:RNA duplexes. (a) ON4:RNA, (b) ON5:RNA, (c) ON7:RNA, and (d) ON8:RNA. Blue =
5-substituents of monomer X (1); cyan = 5-substituents of monomer Y (2); red = backbone; green = nucleobases.
orientation of the phenyl-triazole substituents with the torsional
angle ω in the expected anti conformation around 180° (Figure 6a).
In the top view, the stacking of the substituents involving both
the triazoles and the phenyls is clearly seen. In the other triplex
formed by ON8, the monomers 2 do not show a perfect coplanarity
of the aromatic and heteroaromatic rings, and the stacking between
the substituents are somewhat hampered (Figure 6b). From the top
view it is clear that the phenyl groups are pointing away from the
triplex core and not participating in stacking. This might explain the
significant difference in stability, where the triplex formed byON8 is
not stable and the triplex formed by ON7 is significantly stabilized as
compared to the unmodified triplex.
Nuclease Resistance. In order to establish the stability of the
modified oligonucleotides against nucleolytic degradation, the
modified oligonucleotides were radiolabeled with 32P and mixed
with a 30-exonuclease as well as with fetal calf serum, and the
degradation was studied by gel electrophoresis (Figure 7 and 8).
ON1ꢀ8 were radiolabeled on the 50-end and digested with snake
venom phosphordiesterase (SVPD). Samples were taken after 1,
5, 15, and 30 min and after 1, 4, 16, 24, and 66 h and resolved on
denaturing (7 M urea) acrylamide gels (Figure 7). Smaller
radioactive fragments migrate faster and are located at the
bottom of the gels. As evident from Figure 7, the unmodified
ON1 is degraded within 15 min, whereas all modified oligonu-
cleotides demonstrate a significant resistance toward SVPD. For
the single modified ON2 and ON3, SVPD stalls at position 9 and
10 (from the 50-end), respectively, after removal of the 30-ends of
the sequences. However, the oligonucleotides were completely
broken down by 24 h. For the sequences with 2ꢀ3 consecutive
incorporations of 1 and 2, ON4ꢀ6, the 30-ends were again
rapidly removed, but SVPD cannot digest past the first modified
positions (position 5 or 6) and stalls indefinitely. Hence, two
consecutive incorporations are much more efficient than one in
protecting the oligonucleotide against the 30-exonucleoase. In a
similar way, the oligonucleotides with four incorporations of 1 or
2, ON7ꢀ8, demonstrate a rapid digestion of the 30-end that stops
at the first (or second) modification. This design too appears to
be stable against SVPD, although completely degraded products
appear to increase over 24 h.
The digestion in fetal calf serum was also studied, and as
evident from Figure 8 this was faster for the modified oligonu-
cleotides than the 30-exonucleolytic degradation by SVPD. As
all oligonucleotides disappeared after 8 h, probably due to a
phosphatase activity removing the 32P-labels, we focused on the
reaction in the first 3 h, where a significant stabilization by the
modified nucleotides was detected. Whereas the single modified
oligonucleotides ON2ꢀ3 were almost fully degraded in 3 h, the
oligonucleotides with 2ꢀ4 consecutive modifications were much
more stable, demonstrating a rapid digestion of the 30-end but
then a stall at the first (or second) modification.
’ DISCUSSION
With the present introduction of the phenyl-triazole substi-
tuted cytidine monomer 2 as a supplement to the corresponding
uridine monomer 1, a much wider range of RNA sequences can
be targeted. Our hybridization data proves that both monomers
can lead to increased thermal stability of the DNA:RNA duplexes
if at least two consecutive incorporations are used. This means
that from a situation with just 1 in hand, where only RNA
sequences with AA stretches are potential targets, we have now
broadened the potential to target RNA sequences with just any of
6182
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
Figure 6. Modeling structures of modified triplexes. (a) ON7:dsDNA and (b) ON8:dsDNA; full structures, and modified sections in side view and top
view. Cyan = backbone of the TFO; blue = 5-substituents of the TFO; red = backbone of the dsDNA; green = nucleobases.
the four possible dipurine stretches. It is clear that the impact on
thermal stability is larger with an increasing number of con-
secutive incorporations and better with consecutive incorpora-
tions of the uridine monomer 1 as compared to the cytidine
monomer 2. Nevertheless, a positive impact is found for both
monomers. Convincing mis-match discrimination has also been
demonstrated for sequences containing consecutive incorpora-
tions of 1.10 The combination of 1 and 2 hereby allows for a new
and very simple approach for the development of potential
antisense oligonucleotides. In this aspect, it is very important
that the modifications also lead to a very significant stabilization
toward nucleolytic degradation, as has been previously observed
also for other 5-modified pyrimidine analogues.24 It adds to this
importance that the stabilization increases with two consecutive
modifications of 1 or 2. For the development of antisense
oligonucleotides, 1 and 2 can easily be combined with other
modifications such as LNA and phosphorthioates that in combi-
nation has proven very promising in antisense research.6 LNA
monomers induce a larger increase in thermal stability of DNA:
RNA duplexes than the nucleobase substituents of 1 and 2, but
the two kinds of modifications show opposite behavior, as the
impact of an LNA monomer is largest by one incorporation and
then decreasing, relatively, by the number of consecutive in-
corporations. Thus, the right balance in affinity needed for
securing a selective binding of the target sequences can be
programmed with the combination of different modifications such
as 1 and 2. It is therefore also important that the phosphoramidites
of both monomers 1 and 2 are easily synthesized with the
convenient in situ azidation/CuAAC reaction as the key step.
In contrary to the RNA binding and thereby antisense
potential, the triplex-forming and thereby antigene-potential of
1 and 2 is less pronounced. Even though some increase in
thermal stability of the triplex can be observed with 3ꢀ4
consecutive incorporations, the increase is relatively small and
possible only with four incorporations of 1 and not with 2.
Furthermore, the increase obtained with 1 is probably too small
as compared to other modifications, for instance, LNA,17 not the
least taking into account that triplexes are in the starting point
not very stable as compared to duplexes.
Modeling and CD spectra show that there is a close connec-
tion between the structural impact of monomers 1 and 2 on
duplexes and the thermal stability. Modeling shows that stacking
between monomers, as expected, is the key to the increase in
duplex stability. The CD spectra show that the duplexes with the
most pronounced relative increases in thermal stability, due to
stacking, are also those with most impact on duplex structures. In
line with the ab initio calculations proving coplanar but opposite
conformations for 1 and 2, Figure 5 shows that the positioning of
the substituents in the duplex is also very different between the
two monomers. This has a very clear connection to the stacking
of consecutive incorporations of either 1 or 2 (as in ON7 and
ON8), but when the two modifications are mixed the difference
decreases, and positive influence of stacking exists in both the 50-XY
as in the 50-YX sequences. The modeling also demonstrates why the
triplex is not stable with four incorporations of 2, whereas a stabilized
triplex is formed with four incorporations of 1. Hence, there is
apparently not room for a sufficient stacking of the phenyls of 2 in
the sterically crowded core of the triplex structure.
6183
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
Figure 7. Gel electrophoresis of 50-32P-labeled oligonucleotides digested by SVPD.
Figure 8. Gel electrophoresis of 50-32P-labeled oligonucleotides incubated with fetal calf serum.
In addition to the obvious therapeutic potential of 1 and 2 in
the antisense approach, the programming of duplex stability by
stacking phenyl-triazole substituents in the major groove might
also find applications in the preparation of nanomaterials. The
delicate correspondence between the substituents, their inter-
dependent stacking and the subsequent influence on structure
gives a new possibility in a design where both structure and
thermal stability can be controlled. Furthermore, 1 and 2 can be
combined with other pyrimidine nucleotide building blocks, on
which the CuAAC reaction has been used to attach functional or
fluorescent moieties to the 5-position through a triazole, for
instance, pyrenes.25 Hereby, the decrease in duplex stability
due to a single triazole modification might be converted to an
increase by flanking this modification with 1 and/or 2.
’ CONCLUSION
Even though LNA and a few other nucleic acid analogues have
set the standard for potential antisense therapeutics, there is still a
need for new simple and easily obtained analogues that can be
used alone or in combination with other modifications. We
believe that the building blocks of the current study, 1 and 2,
6184
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
together forms such a candidate with the simple synthesis, the
increase in RNA recognition due to stacking interactions of
phenyl-triazole substituents in the major groove, the program-
mable RNA-affinity by the combination of the U and C-building
blocks, and the high physiological stability.
H-40), 3.72ꢀ3.58 (m, 2H, H-50), 2.23 (m, 1H, H-20), 2.12 (m, 1H, H-20).
13C NMR (100 MHz, DMSO-d6) δ 162.4, 153.8, 142.6, 140.4, 136.4,
129.9, 129.1, 120.4, 119.4, 96.4, 87.3, 85.4, 69.7, 60.8, 40.7. HR-ESI. m/z
393.1290 [M ꢀ Na]+ calcd C17H18N6O4 Na+ 393.1282.
3
Preparation of 4-N-Acetyl-5-(phenyl-1H-1,2,3-triazol-4-yl)-
20-deoxycytidine (5). The nucleoside 2 (442 mg, 1.19 mmol) was
co-evaporated with pyridine (2 ꢁ 10 mL) and redissolved in the
same solvent (12 mL). The solution was stirred at 0 °C, and acetic
anhydride (371 μL, 3.39 mmol) was added. The mixture was stirred for
4 h at 0 °C, diluted with CH2Cl2 (10 mL) and washed with a saturated
aqueous solution of NaHCO3 (10 mL). The aqueous phase was extracted
with CH2Cl2 (5 ꢁ 20 mL), and the combined organic phase was dried
(MgSO4) and concentrated under reduced pressure. The residue was
purified by flash column chromatography (0ꢀ10% MeOH in CH2Cl2)
to give the peracetylated intermediate as a white solid (355 mg, 60%). Rf
0.9 (10% MeOH in CH2Cl2). 1H NMR (400 MHz, DMSO-d6) δ 10.85
(s, 1H, NH), 9.17 (s, 1H, H-6), 8.42 (s, 1H, triazole CH), 7.94 (m, 2H,
Ph), 7.67 (m, 2H, Ph), 7.56 (m, 1H, Ph), 6.19 (t, 1H, J = 6.8 Hz, H-10),
5.25 (m, 1H, H-30), 4.39ꢀ4.32 (m, 3H, H-40, H-50), 2.61ꢀ2.48 (m, 2H,
H-20), 2.44 (s, 3H, CH3CO), 2.10 (s, 3H, CH3CO), 1.97 (s, 3H,
CH3CO). 13C NMR (100 MHz, DMSO-d6) δ 170.9, 170.2, 169.96,
158.4, 152.6, 142.6, 141.6, 136.2, 130.0), 129.2, 120.2, 98.7, 87.3, 82.4,
74.1, 63.5, 37.2, 25.9, 20.7, 20.4. HR-ESI: m/z 519.1594 [M ꢀ Na]+
’ EXPERIMENTAL SECTION
General. All commercial reagents were used as supplied. Reactions
were performed under an atmosphere of nitrogen when anhydrous
solvents were used. Column chromatography was carried out on glass
columns using silica gel 60 (0.040ꢀ0.063 mm). NMR spectra were
recorded at 300 or 400 MHz for 1H NMR, at 75 or 100 MHz for 13C NMR,
and at 162 MHz for 31P NMR. The δ values are in ppm relative to
tetramethylsilane as internal standard or 85% H3PO4 as external
standard. Assignments of NMR signals are based on 2D spectra and
follow the standard nucleoside convention. HR ESI mass spectra were
recorded in positive-ion mode.
Preparation of 5-Ethynyl-20-deoxycytidine (4). A solution of
5-iodo-20-deoxycytidine, 3, (1.97 g, 5.58 mmol) in anhydrous DMF
(37 mL) and anhydrous triethylamine (19 mL) was degassed and stirred
at room temperature. Trimethylsilylacetylene (4.4 g, 44.6 mmol), Pd-
(PPh3)2Cl2 (392 mg, 0.558 mmol) and CuI (106 mg, 0.558 mmol) were
added, and the mixture was stirred for 1 h. The mixture was concentrated
under reduced pressure, and the residue was co-evaporated with xylene
(2 ꢁ 30 mL) and purified by flash column chromatography (0ꢀ20%
MeOH in CH2Cl2) to afford the crude TMS-protected intermediate
(2.39 g) as a white foam. Rf 0.3 (20% MeOH in CH2Cl2). 1H NMR (400
MHz, DMSO-d6) δ 8.20 (s, 1H, H-6), 7.78 (br s, 1H, NH), 6.60 (br s,
1H, NH), 6.09 (t, J = 6.5 Hz, H-10), 5.20 (d, J = 4.3 Hz, 1H, 30-OH), 5.06
(t, Hz, 1H, 50-OH), 4.20 (m, 1H, H-30), 3.78 (m, 1H, H-40), 3.58 (m, 2H,
H-50), 2.14 (ddd, J = 13.1, 5.8, 3.7 Hz, 1H, H-20), 1.99 (dt, J = 13.2, 6.5
Hz, 1H, H-20), 0.21 (s, 9H, TMS). 13C NMR (101 MHz, DMSO-d6) δ
164.09, 153.44, 145.47 (C-6), 99.84, 97.12, 87.65 (C-40), 85.62 (C-10),
70.17 (C-30), 61.16 (C-50), 40.98 (C-20), 0.11 ((CH3)3Si). HR-ESI-MS:
calcd C23H24N6O7 Na+ 519.1599. The intermediate nucleoside (344
3
mg, 0.69 mmol) was suspended in 1,4-dioxane (15 mL) and H2O
(7.5 mL), and the mixture was stirred at 0 °C. A solution of sodium
hydroxide (110 mg, 2.76 mmol) in H2O (1.4 mL) was added dropwise,
and the reaction mixture was stirred for 2 h at 0 °C. A saturated aqueous
solution of ammonium chloride (10 mL) was added, and the mixture was
concentrated under reduced pressure. The residue was purified by
column chromatography (0ꢀ20% MeOH in CH2Cl2) to give the
product 5 as a white solid (224 mg, 78%). Rf 0.2 (10% MeOH in
CH2Cl2). 1H NMR (400 MHz, DMSO-d6) δ 8.92 (s, 1H, H-6), 8.45
(s, 1H, triazole CH), 7.93 (d, J = 7.8 Hz, 2H, Ph), 7.65 (t, J = 7.8 Hz,
2H, Ph), 7.54 (t, J = 7.8 Hz, 1H, Ph), 6.25 (m, 1H, H-10), 6.13 (br s,
1H, OH), 5.25 (t, J = 2.8 Hz, 1H, 50-OH), 4.08 (d, J = 2.4 Hz, 1H,
H-30), 3.77ꢀ3.38 (m, 3H, H-40, H-50), 2.42 ꢀ 2.28 (m, 2H, H-20),
2.08 (s, 3H, CH3CO). 13C NMR (101 MHz, DMSO) δ 170.1, 162.5,
153.77, 142.3, 140.3, 136.4, 130.0, 129.1, 120.5, 119.8, 96.8, 85.5,
84.9, 74.7, 61.2, 37.8, 20.9. HR-ESI: m/z 435.1383 [M ꢀ Na]+ calcd
m/z 346.1180 [M + Na]+ calcd C14H21N3O4Si Na+ 346.1194. The
3
crude intermediate was dissolved in an 24% aqueous solution of ammonia
(40 mL), and the mixture was stirred at room temperature for 2 h. The
mixture was concentrated under reduced pressure, and the residue was
co-evaporated with 99% ethanol (50 mL) and purified by flash column
chromatography (0ꢀ30% MeOH in CH2Cl2) affording the target nucleo-
side 4 (1.33 g, 95%) as a white solid. Rf 0.2 (20% MeOH in CH2Cl2). 1H
NMR (200 MHz, DMSO-d6) δ 8.25 (s, 1H, H-60), 7.69 (s, 1H, NH),
6.81 (s, 1H, NH), 6.09 (t, J = 6.4 Hz, 1H, H-10), 5.21 (d, J = 2.8 Hz, 1H,
30-OH), 5.09 (t, J = 4.4 Hz, 1H, 50-OH), 4.33 (s, HCtC, 1H), 4.20 (m,
1H, H-30), 3.79 (q, J = 3.2 Hz, 1H, H-40), 3.65 ꢀ 3.51 (m, 2H, H-50), 2.16
(m, 1H, H-20), 1.99 (m, 1H, H-20). 13C NMR (100 MHz DMSO-d6)
δ 164.2, 153.4, 145.4, 88.8, 87.5, 85.8, 85.4, 75.9, 70.0, 60.9, 40.9. HR-
C19H20N6O5 Na+ 435.1388.
3
Preparation of 4-N-Acetyl-50-O-(4,40-dimethoxytrityl)-5-
(phenyl-1H-1,2,3-triazol-4-yl)-20-deoxycytidine (6). The nucleo-
side 5 (192 mg, 0.47 mmol) was co-evaporated with anhydrous pyridine
(4 ꢁ 10 mL) and redissolved in the same solvent (4.7 mL). 4,40-
Dimethoxytrityl chloride (166 mg, 0.49 mmol) was added, and the reaction
mixture was stirred at room temperature for 2 h. Methanol (2 mL) was
added, and the mixture was diluted with CH2Cl2 (20 mL) and washed with
a saturated aqueous solution of NaHCO3 (20 mL). The aqueous phase was
extracted with CH2Cl2 (4 ꢁ 20 mL), and the combined organic phase was
dried (MgSO4) and concentrated under reduced pressure. The residue was
purified by flash column chromatography (0ꢀ20% MeOH and 1% TEA in
CH2Cl2) to give the product 6 as a white solid (225 mg, 67%). Rf 0.5 (10%
ESI-MS: m/z 274.0790 [M + Na]+ calcd C11H13N3O4 Na+ 274.0799.
3
Preparation of 5-(Phenyl-1H-1,2,3-triazol-4-yl)-20-deoxy-
cytidine (2). A mixture of bromobenzene (316 μL, 3.0 mmol), NaN3
(195 mg, 3.0 mmol), CuI (57 mg, 0.3 mmol), and N,N0-dimethylethy-
lenediamine (49 μL, 0.45 mmol) in EtOH/H2O (7:3, v/v, 6 mL) was
stirred under microwave irradiation at 100 °C for 1 h and then filtered
through a 45 μm filter. Nucleoside 4 (300 mg, 1.19 mmol), TBTA (127
mg, 0.24 mmol), and CuI (34 mg, 0.18 mmol) were added, and the
reaction mixture was stirred at room temperature for 96 h and then at
50 °C for 22 h. The mixture was concentrated under reduced pressure,
and the residue was purified by flash column chromatography (0ꢀ20%
MeOH in CH2Cl2) affording the product 2 as a white solid (356 mg,
80%). Rf 0.3 (20% MeOH in CH2Cl2). 1H NMR (400 MHz, DMSO-d6)
δ 8.86 (s, 1H, H-6), 8.49 (s, 1H, triazole CH), 7.94ꢀ7.51 (m, 7H, Ph,
NH2), 6.20 (t, 1H, J = 6.4 Hz, H-10), 5.25 (d, 1H, J = 4.4 Hz, 30-OH),
5.19 (t, 1H, J = 4.8 Hz, 50-OH), 4.27 (m, 1H, H-30), 3.83 (q, 1H, J = 3.6 Hz,
1
MeOH in CH2Cl2). H NMR (400 MHz, DMSO-d6) δ 8.46 (s, 1H,
triazole CH), 8.18 (s, 1H, H-6), 7.93 (br s, 1H, NH), 7.62 (br s, 1H, NH),
7.55ꢀ7.46 (m, 5H, Ph, DMT), 7.32 (d, J = 7.6 Hz, 2H, Ph),
7.22ꢀ7.06 (m, 7H, Ph, DMT), 6.74 (m, 4H, DMT), 6.27 (m, 1H,
H-10), 5.15 (m, 1H, H-30), 4.16 (m, 1H, H-40), 3.62 (s, 3H, OCH3),
3.61 (s, 3H, OCH3), 3.39ꢀ3.25 (m, 2H, H-50), 2.48ꢀ2.39 (m, 2H,
H-20), 2.03 (s, 3H, CH3CO). 13C NMR (101 MHz, DMSO-d6) δ
169.9, 162.7, 158.0, 153.7, 144.5, 141.9, 139.7, 136.1, 135.3, 129.8,
129.5, 128.9, 127.8, 127.6, 126.7, 120.0, 119.6, 113.1, 97.0, 85.9, 85.6,
83.3, 74.4, 63.5, 54.9, 37.7, 20.8. HR-ESI: m/z 737.2694 [M ꢀ Na]+
calcd C40H38N6O7 Na+ 737.2695.
3
6185
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
Preparation of 4-N-Acetyl-30-O-(P-2-cyanoethyl-N,N-diiso-
propylaminophosphinyl)-50-O-(4,40-dimethoxytrityl)-5-(phenyl-
1H-1,2,3-triazol-4-yl)-20-deoxycytidine (7). The nucleoside 6
(150 mg, 0.21 mmol) was co-evaporated with anhydrous 1,2-dichlorethane
(2 ꢁ 4 mL) and redissolved in the same solvent (2.0 mL). Diisopropy-
lammonium tetrazolide (68 mg, 0.4 mmol) and 2-cyanoethyl-N,N,N0,N0-
tetraisopropylphosphoramidite (157 mg, 0.52 mmol) was added, and
the mixture was stirred at room temperature for 25 h. Methanol (2 mL)
was added, and the mixture was purified by flash column chromatogra-
phy (0ꢀ10% MeOH and 1% TFA in CH2Cl2) to give the product 7 as a
white solid (164 mg, 85%). Rf 0.5 (5% MeOH in CH2Cl2). 31P NMR
(162 MHz, CDCl3) δ 149.61, 149.36, 148.89, 148.48 (rotamers). HR-
Molecular Modeling. General Parameters. In all of the calcula-
tions the phosphodiester backbone charge was neutralized with
sodium ions, placed 3.0 Å from the negatively charged oxygen atoms
in the plane described by the phosphorus and the nonbridging oxygen
atoms. The sodium ions were constrained throughout the series of
calculations by a force constant of 418 KJ/molÅ2. The AMBER* force
field supplied with MacroModel V9.1,19 atom type and GB/SA
solvation model was adapted and modified using the parambsc0
parameters.21,22
Generation of Starting Structure Procedure. A standard B-type
DNA:DNA helix I built in the MacroModel V.9.1 suite of programs19
and modified with the appropriate C-5 thymine modification to form
ON1ꢀ10. The model structure was subjected to an torsional rotation
MCMM structure search.23 In the MCMM structure search the rotation
around C10/N1, C5/C100, and N10/C100 bonds are given a 180° rotational
freedom, and during the MCMM 1000 structures are generated. The
obtained structures are subjected to a multiple minimization to reduce
the number of structures generated into a number of local and global
minima. The global minima structure generated is then used in the
subsequent MD simulation.
Molecular Dynamics Parameters. The global energy structure obtained
was subjected to a 5 ns MD simulation (simulation temperature 300 K, time
step 2.2 fs, SHAKE all bonds to hydrogen and the molecular dynamics
setting) during which 500 individual structures were sampled. These sample
structures were subsequently minimized to obtain a converged global
minimum. The duplex structures were minimized using the PolakꢀRibiere
conjugate gradient method, the modified all-atom AMBER force field19,23,26,27
and modified GB/SA solvation model22 implemented in MacroModel
V9.1. Nonbonded interactions were treated with extended cut-offs (van der
Waals 8.0 Å and electrostatics 20.0 Å). The 500 sample structures were
subjected to a multiple minimization to identify the global and local minima
sampled. The global minimum was used for analysis.
Nuclease Resistance Assays. Oligonucleotide stability against
snake venom phosphodiesterase I from Crotalus adamanteus (Pharmacia
Biotech) was evaluated by incubating 3 μM of of 50-32P-labeled
oligonucleotide with 6.7 ng/μL phosphodiesterase I in 100 mM Tri-
s-HCl (pH 8.0), 15 mM MgCl2, at 21 °C. Initial samples (0 min) were
drawn immediately prior to adding the enzyme. Oligonucleotide
stability in fetal calf serum (Biochrom AG) was evaluated by incubating
2 μM of 50-32P-labeled oligonucleotides in a 1:1 mixture of serum and
water. Aliquots were quenched with 1 vol ice-cold 95% formamide with
excess EDTA. Samples were resolved on 13% denaturing polyacrylamide
electrophoresis gels with 7 M urea and visualized by autoradiography.
ESI: m/z 915.3910 [M ꢀ H]+ C49H55N8O8P H+ calcd 915.3953.
3
Oligonucleotide Synthesis. Oligonucleotide synthesis was car-
ried out on an automated DNA synthesizer following the phosphor-
amidite approach. Synthesis of oligonucleotides ON2ꢀ8 was performed
on a 0.2 μmol scale (CPG support) by using the modified nucleoside
phosphoramidites (7 and the corresponding amidite of 110) as well as
the corresponding commercial 2-cyanoethyl phosphoramidites of the
natural 20-deoxynucleosides. The synthesis followed the regular protocol
for the DNA synthesizer. For the modified phosphoramidites a prolonged
coupling time of 20 min was used. 1H-Tetrazole was used as activator.
Coupling yields for all 2-cyanoethyl phosphoramidites were >96%. The 50-
O-DMT-ON oligonucleotides were removed from the solid support by
treatment with concentrated aqueous ammonia at 55 °C for 12 h. The
oligonucleotides were purified by reversed-phase HPLC on a Waters 600
system using Xterra MS C18 10 μm; 7.8 ꢁ 150 mm column + precolumn:
Xterra MS C18 10 μm, 7.8 ꢁ 10 mm. Buffer A: 0.05 M triethyl ammonium
acetate, pH 7.4. Buffer B: 75% MeCN/H2O (3:1, v/v). Program used: 2
min 100% A, 100ꢀ30% A over 38 min, 10 min 100% B, 10 min 100% A. All
oligonucleotides were detritylated by treatment with 80% aqueous acetic
acid for 20 min and neutralized by addition of sodium acetate (3 M, 15 μL),
and then sodium perchlorate (5 M, 15 μL) was added followed by acetone
(1 mL). The pure oligonucleotides precipitated overnight at ꢀ20 °C. The
mixture was then placed in a centrifuge and subjected to 12,000 rpm, 10 min
at 4 °C. The supernatant was removed, and the pellet was washed with cold
acetone (2 ꢁ 1 mL). The pellet was then dried for 30 min under reduced
pressure and dissolved in pure water (500 μL), and the concentration was
measured as OD at 260 nm. The purity was confirmed by IC analysis.
MALDI-TOF MS [M ꢀ H]ꢀ gave the following results (calcd/found):
ON2 (4831.2/4830.8), ON3 (4845.2/4846.3), ON4 (4974.3/4974.4),
ON5 (4974.3/4974.6), ON6 (5105.5/5106.4), ON7 (5218.6/5219.2) and
ON8 (5274.7/5276.3).
Thermal Denaturation Experiments. Extinction coefficients of
the modified oligonucleotides were estimated from a standard method but
calibrated by the micromolar extinction coefficients of the monomeric
’ ASSOCIATED CONTENT
compounds 1 and 2, which were estimated from their UV spectra (1, ε260
=
S
Supporting Information. CD spectra for the studied
b
7.8;10 2, ε260 = 13.5). UV melting experiments were thereafter carried out
on a UV spectrometer. For the duplex studies, samples were dissolved in a
medium salt buffer containing 2.5 mM Na2HPO4, 5 mM NaH2PO4,
100mMNaCl, and0.1mMEDTAat pH=7.0with1.5μM concentrations
of the two complementary oligonucleotide sequences. The increase in
absorbance at 260 nm as a function of time was recorded while the temperature
was increased linearly from 5 to 80 or 90 °C at a rate of 1.0 °C/min by means
of a Peltier temperature programmer. For the triplex studies, heating was
performed at a rate of 0.5 °C/min in a buffer containing 10 mM sodium
cacodylate, 150 mM NaCl and 10 mM MgCl2, pH = 6.0. The melting
temperature was determined as the local maximum of the first derivatives of the
absorbance vs temperature curve. The melting curves were found to be
reversible. All determinations are averages of at least duplicates within (0.5 °C.
Circular Dichroism Spectroscopy. CD spectra were obtained at
5 °C using the same medium salt buffer as in the UV melting
experiments with 1.5 μM concentrations of the two complementary
oligonucleotide sequences.
triplexes and selected NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: pon@ifk.sdu.dk.
’ ACKNOWLEDGMENT
The project was supported by The Danish Research
Agency0s Programme for Young Researchers, Nucleic Acid
Center and The Danish National Research Foundation, The
Danish Council for Independent Research, Technology and
Production Sciences (FTP) and Natural Sciences (FNU), Danish
Center for Scientific Computing and Møllerens Fond.
6186
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187

The Journal of Organic Chemistry
ARTICLE
’ REFERENCES
(25) Østergaard, M. E.; Guenther, D. C.; Kumar, P.; Baral, B.;
Deobald, L.; Paszczynski, A. J.; Sharma, P. K.; Hrdlicka, P. J. Chem.
Commun. 2010, 46, 4929–4931.
(26) Weiner, S. J.; Kollman, P. A.; Case, D. A.; Singh, U. C.; Ghio, C.;
Alagona, G.; Profeta, S.; Weiner, P. J. Am. Chem. Soc. 1984, 106, 765–784.
(27) Weiner, S. J.; Kollman, P. A.; Nguyen, D. T.; Case, D. A.
J. Comput. Chem. 1986, 7, 230–252.
(1) (a) Zamecnik, P. C.; Stephenson, M. L. Proc. Natl. Acad. Sci.
U.S.A. 1978, 75, 280–284. (b) Stephenson, M. L.; Zamecnik, P. C. Proc.
Natl. Acad. Sci. U.S.A. 1978, 75, 285–288.
(2) Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 543–580.
(3) (a) Kurreck, J. Eur. J. Biochem. 2003, 270, 1628–1644. (b) Bell,
N. M.; Micklefield, J. ChemBioChem 2009, 10, 2691–2703.
(4) Buchini, S.; Leumann, C. J. Curr. Opin. Chem. Biol. 2003, 7, 717–726.
(5) (a) Freier, S.; Altmann, K.-H. Nucleic Acid Res. 1997, 25, 4429–4443.
(b) Aboul-Fadl, T. Curr. Med. Chem. 2005, 12, 2193–2214. (c) Cobb, A. J. A.
Org. Biomol. Chem. 2007, 5, 3260–3275. (d) Lebreton, J.; Escudier, J.-M.;
Arzel, L.; Len, C. Chem. Rev. 2010, 110, 3371–3418.
(6) (a) Vester, B.; Wengel, J. Biochemistry 2004, 43, 13233–13241.
(b) Kaur, H.; Babu, B. R.; Maiti, S. Chem. Rev. 2007, 107, 4672–4697.
(c) Nielsen, P.; Wengel, J. In Modified Nucleosides in Biochemistry,
Biotechnology and Medicine; Herdewijn, P., Ed.; Wiley-VCH: Weinheim,
2008; pp 133ꢀ152.
(7) Luyten, I.; Herdewijn, P. Eur. J. Med. Chem. 1998, 33, 515–576.
(8) Lin, K.-Y.; Jones, R. J.; Matteucci, M. J. Am. Chem. Soc. 1995,
117, 3873–3874.
(9) (a) Wagner, R. W.; Matteucci, M. D.; Lewis, J. G.; Gutierrez,
A. J.; Moulds, C. M.; Froehler, B. C. Science 1993, 260, 1510–1513.
(b) Gutierrez, A. J.; Matteucci, M. D.; Grant, D.; Matsumura, S.; Wagner,
R. W.; Froehler, B. C. Biochemistry 1997, 36, 743–748. (c) Gutierrez,
A. J.; Terhorst, T. J.; Matteucci, M. D.; Froehler, B. C. J. Am. Chem. Soc.
1994, 116, 5540–5544. (d) Sinkeldam, R. W.; Greco, N. J.; Tor, Y.
ChemBioChem 2008, 9, 706–709.
(10) Koꢁcalka, P.; Andersen, N. K.; Jensen, F.; Nielsen, P. ChemBio-
Chem 2007, 8, 2106–2116.
(11) (a) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem.
2002, 67, 3057–3064. (b) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.;
Sharpless, K. B. Angew. Chem., Int. Ed. 2002, 41, 2596–2599.
ꢀ
(12) Andersen, N. K.; Chandak, N.; Brulikovꢀa, L.; Kumar, P.; Jensen,
M. D.; Jensen, F.; Sharma, P. K.; Nielsen, P. Bioorg. Med. Chem. 2010,
18, 4702–4710.
(13) A preliminary synthesis of compound 2 has been published in a
conference proceeding: Andersen, N. K.; Spꢀaꢁcilovꢀa, L.; Jensen, M. D.;
Koꢁcalka, P.; Jensen, F.; Nielsen, P. Nucleic Acids Symp. Ser. 2008,
52, 149–150.
(14) Independently, compound 2 has been recently presented as a
fluorescent nucleoside analogue: Dodd, D. W.; Swanick, K. N.; Price,
J. T.; Brazeau, A. L.; Ferguson, M. J.; Jones, N. D.; Hudson, R. H. E. Org.
Biomol. Chem. 2010, 8, 663–666.
(15) Frisch, M. J. et al. ; Gaussian 03; Gaussian Inc.: Wallingford, CT,
2004
(16) (a) Barr, P. J.; Jones, A. S.; Serafinowski, P.; Walker, R. T.
J. Chem. Soc., Perkin Trans. 1 1978, 1263–1267. (b) Marsh, A.; Alcock,
N. W.; Errington, W.; Sagar, R. Tetrahedron 2003, 59, 5595–5601.
(17) Højland, T.; Kumar, S.; Babu, B. R.; Umemoto, T.; Albæk, N.;
Sharma, P. K.; Nielsen, P.; Wengel, J. Org. Biomol. Chem. 2007,
5, 2375–2379.
(18) Saenger, W. Principles of Nucleic Acid Structure; Springer:
New York, 1984.
(19) MacroModel, version 9.1; Schr€odinger, LLC: New York, NY,
2005.
(20) The triplex structure was built via the web page http://w3dna.
rutgers.edu/ allowing full control of sequence design, and the generated
structure was exported as a pdb file and modified accordingly.
(21) Pꢀerez, A.; Marchꢀan, I.; Svozil, D.; Sponer, J.; Cheatham, T. E.;
Laughton, C. A.; Orozco, M. Biophys. J. 2007, 92, 3817–3829.
(22) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T.
J. Am. Chem. Soc. 1990, 112, 6127–6129.
(23) Chang, G.; Guida, W. C.; Still, W. C. J. Am. Chem. Soc. 1989,
111, 4379–4386.
(24) Østergaard, M. E.; Kumar, P.; Baral, B.; Raible, D. J.; Kumar,
T. S.; Anderson, B. A.; Guenther, D. C.; Deobald, L.; Paszczynski, A. J.;
Sharma, P. K.; Hrdlicka, P. J. ChemBioChem 2009, 10, 2740–2743.
6187
dx.doi.org/10.1021/jo200919y |J. Org. Chem. 2011, 76, 6177–6187