High-affinity RNA targeting by oligonucleotides displaying aromatic stacking and amino groups in the major groove. Comparison of triazoles and phenyl substituents

Article abstract of DOI:10.1021/jo4025896

Three 5-modified 2′-deoxyuridine nucleosides were synthesized and incorporated into oligonucleotides and compared with the previously published 5-(1-phenyl-1,2,3-triazol-4-yl)-2′-deoxyuridine monomer W. The introduction of an aminomethyl group on the phenyl group led to monomer X, which was found to thermally stabilize a 9-mer DNA:RNA duplex, presumably through the partial neutralization of the negative charge of the backbone. By also taking advantage of the stacking interactions in the major groove of two or more of the monomer X, an extremely high thermal stability was obtained. A regioisomer of the phenyltriazole substituent, that is the 5-(4-phenyl-1,2,3-triazol-1-yl)- 2′-deoxyuridine monomer Y, was found to destabilize the DNA:RNA duplex significantly, but stacking in the major groove compensated for this when two to four monomers were incorporated consecutively. Finally, the 5-phenyl-2′-deoxyuridine monomer Z was incorporated for comparison, and it was found to give a more neutral influence on duplex stability indicating less efficient stacking interactions. The duplexes were investigated by CD spectroscopy and MD simulations.

Full text of DOI:10.1021/jo4025896

Article
pubs.acs.org/joc
High-Affinity RNA Targeting by Oligonucleotides Displaying
Aromatic Stacking and Amino Groups in the Major Groove.
Comparison of Triazoles and Phenyl Substituents
Pawan Kumar,† Mick Hornum,† Lise J. Nielsen,† Gerald Enderlin,‡ Nicolai Krog Andersen,†
́
Christophe Len,‡ Gwenaelle Herve,‡ Guillaume Sartori,‡ and Poul Nielsen*,†
́
́
̈
†Nucleic Acid Center, Department of Physics, Chemistry and Pharmacy, University of Southern Denmark, 5230 Odense M, Denmark
‡Transformations Integ
́
rees de la Matiere Renouvelable, UTC-ESCOM, Centre de recherche Royallieu, BP 20529, 60205 Compiegne,
́
̀
̀
France
S
* Supporting Information
ABSTRACT: Three 5-modified 2′-deoxyuridine nucleosides
were synthesized and incorporated into oligonucleotides and
compared with the previously published 5-(1-phenyl-1,2,3-
triazol-4-yl)-2′-deoxyuridine monomer W. The introduction of
an aminomethyl group on the phenyl group led to monomer X,
which was found to thermally stabilize a 9-mer DNA:RNA
duplex, presumably through the partial neutralization of the
negative charge of the backbone. By also taking advantage of the
stacking interactions in the major groove of two or more of the
monomer X, an extremely high thermal stability was obtained. A
regioisomer of the phenyltriazole substituent, that is the 5-(4-
phenyl-1,2,3-triazol-1-yl)-2′-deoxyuridine monomer Y, was
found to destabilize the DNA:RNA duplex significantly, but
stacking in the major groove compensated for this when two to four monomers were incorporated consecutively. Finally, the 5-
phenyl-2′-deoxyuridine monomer Z was incorporated for comparison, and it was found to give a more neutral influence on
duplex stability indicating less efficient stacking interactions. The duplexes were investigated by CD spectroscopy and MD
simulations.
INTRODUCTION
■
Since its introduction in the late 1970s,1 antisense technology
has been seen as an attractive strategy to regulate gene
expression by the use of short oligonucleotides (ONs) called
antisense ONs which bind to their target RNA in a sequence
specific manner.2 Introduction of chemically modified nucleo-
tides in antisense ONs has emerged as an important approach
to increase their affinity and specificity toward RNA targets as
well as their stability against nucleases.3,4 One approach in this
regard is the introduction of modified nucleobases capable of
enhancing stacking interactions while maintaining the Watson−
Crick base-pairing properties. For instance, replacement of
cytosine with phenoxazine has been shown to increase the
thermal stability of a DNA:RNA duplex significantly.5 Another
example is the 5-propynylpyrimidine nucleoside modification,
which has also been shown to increase the thermal duplex
stability by taking advantage of better stacking interactions.6,7
Recently, we introduced the modified nucleoside monomer
W (Figure 1) in which a phenyltriazole group is attached to the
5-position of 2′-deoxyuridine.8 This monomer is very
conveniently synthesized by using the Cu(I)-catalyzed
alkyne−azide cycloaddition (CuAAC) reaction,9,10 which has
been recently used intensively in nucleic acid chemistry.11−17
Figure 1. Structure of 5-functionalized 2′-deoxyuridine monomers.
Consecutive incorporations of monomer W led to the
formation of very stable DNA:RNA duplexes through efficient
stacking between the phenyltriazole moieties in the major
Received: November 22, 2013
Published: March 10, 2014
© 2014 American Chemical Society
2854
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
groove.8 Thus, two consecutive incorporations of W gave an
increase in melting temperature of a 9-mer DNA:RNA duplex
of 3.3 °C per modification, and this increased to 5.1 °C per
modification with four consecutive incorporations.8,18 The
scope of this DNA analogue was further extended with the
introduction of the corresponding 2′-deoxycytidine monomer,
which was also found to participate in stacking interactions in
the major groove.19 Furthermore, the ONs modified with these
monomers were found to be resistant toward degradation by
nucleases.19 In an effort to optimize the stacking interactions in
the major groove, we have introduced various substituted
phenyltriazoles.18 A sulfonamide substituent on the phenyl ring
of W was shown to give a slight further increase in the thermal
stabilization of the DNA:RNA duplexes, i.e., a maximum
increase in melting temperature of 6.1 °C per modification with
four consecutive incorporations as compared to the 5.1 °C
obtained with W.18,20 For the present study, we decided to
further explore the scope of the phenyltriazole building block
W by (1) introducing a positively charged substituent to the
phenyltriazole (monomer X, Figure 1), (2) varying the
substitution pattern of the triazole ring (monomer Y, Figure
1), and (3) replacing the phenyltriazole substituent with a
simpler phenyl group (monomer Z, Figure 1).
a
Scheme 1
By the introduction of monomer X (Figure 1), we expected
to stabilize DNA:RNA duplexes even further by combining the
efficient stacking of the phenyltriazole moieties in the major
groove with electrostatic interactions between the positively
charged amino group and the negatively charged phospho-
diester backbone. Hence, previous results with nucleotide
monomers modified with an aminopropynyl group at the 5-
position of 2′-deoxyuridine have been shown to form thermally
and enzymatically stable duplexes through a similar combina-
tion.21,22 However, we have recently shown that stacking
efficiency of triazoles attached to the 5-position of 2′-
deoxyuridine is significantly higher than alkynyl substituents
in the same position,20 and we therefore expected X to reveal
duplexes of very high thermal stability.
a
The monomer Y (Figure 1) is an analogue of W in which the
phenyl group is attached to the 5-position of the 2′-
deoxyuridine through an inversed triazole, i.e., the 5-position
of the uracil is connected to the N-1 position of the triazole ring
rather than the C-4 position. Therefore, a comparison can be
drawn between the normal phenyltriazole (monomer W) and
the inversed phenyltriazole (monomer Y) in terms of their
ability to participate in stacking interactions in the major
groove. Furthermore, it is interesting to explore not only the 5-
ethynyl modified pyrimidine nucleosides as a component for
nucleic acid click chemistry reactions, but also the 5-azido-
modified pyrimidine nucleosides. In order to further under-
stand the role of the triazole in the stacking interactions in the
major groove, we also decided to completely remove the
triazole linker and introduce monomer Z (Figure 1) with a
phenyl group directly attached to the 5-position of 2′-
deoxyuridine.
Reagents and conditions: (a) NaN3, N,N-dimethylethylenediamine,
CuI, EtOH/H2O, MW 100 °C, Na ascorbate, 72%; (b) NC-
(CH2)2OPClN(i-Pr)2, (i-Pr)2NEt, CH2Cl2, 54%; (c) DMTrCl,
pyridine, 42% 5, 47% 10; (d) CuSO4·5H2O, Na ascorbate, pyridine/
H2O/tBuOH (1:2:2, v/v/v), 65%; (e) NC(CH2)2OPClN(i-Pr)2,
EtN(i-Pr)2, (ClCH2)2, 67% 7, 66% 11; (f) Na2PdCl4 (0.1 mol %),
KOH, H2O, MW, 100 °C, 85% (ref 26).
protected 5-ethynyl-2′-deoxyuridine 123 was reacted with N-
(3-bromobenzyl)-2,2,2-trifluoroacetamide using our formerly
published in situ azidation/CuAAC protocol8 to obtain
nucleoside 2. Phosphitylation using standard conditions
afforded the phosphoramidite 3. 5-Azido-2′-deoxyuridine 4
was prepared in three high-yielding steps from 5-bromo-2′-
deoxyuridine using a literature procedure.24 DMTr-protection
of 4 gave 5 and subsequent CuAAC reaction with commercially
available phenylacetylene furnished nucleoside 6 in good yield.
Standard phosphitylation of this nucleoside afforded the
phosphoramidite 7. Finally, 5-phenyl-2′-deoxyuridine 925,26
was prepared by a Suzuki−Miyaura coupling between
commercially available 5-iodo-2′-deoxyuridine 8 and phenyl-
boronic acid.26 Subsequent DMT protection afforded 10, and
phosphitylation gave the required phosphoramidite 11. The
identities of the compounds were fully ascertained by NMR
spectroscopy (1H, 13C, 31P, COSY, HSQC, and/or HMBC)
and high-resolution mass spectrometry.
RESULTS AND DISCUSSION
■
Chemical Synthesis. For the introduction of the
monomers X, Y, and Z, we synthesized the three 5′-O-
DMTr-protected 3′-O-phosphoramidites 3, 7, and 11,
respectively (Scheme 1), as suitable starting materials for
solid phase oligonucleotide synthesis. The first two are based
on the general concept of the Cu(I)-catalyzed azide−alkyne
cycloaddition (CuAAC).9,10 The well-known 5′-O-DMTr-
2855
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
Hybridization Studies. The phosphoramidites 3, 7, and 11
were successfully incorporated into ONs to give the modified
nucleotide monomers X, Y, and Z, respectively, using an
automated DNA synthesizer with tetrazole as the activator.
Standard conditions were employed except for extended
coupling times (15 min) for the modified phosphoramidites.
The ON sequence chosen for the present study is a T-rich 9-
mer sequence similar to our previous studies with monomer
W.8,18 ONs with one to four consecutive incorporations of
monomers X, Y, or Z (Table 1, entries 6−17) and mixed
context (entry 2). This indicates that the thermal penalty due
to a single phenyltriazole group is compensated by the
aminomethyl group in monomer X. Interestingly, very stable
DNA:RNA duplexes were observed for the ONs featuring two
to four consecutive incorporations of X (ΔTm in the range of
+7.0 to +9.3 °C per modification, entries 7−9) compared to the
unmodified duplex. Therefore, the effect was stronger than
what was seen for monomer W (ΔTm from +3.3 to +5.1 °C per
modification, entries 2−4) and other 5-triazole-substituted 2′-
deoxyuridines.8,18,20 This indicates for monomer X that the
efficient stacking of the phenyltriazole moieties in the major
groove are combined with partial neutralization of the
negatively charged backbone by the aminomethyl group to
give the very stable duplexes. When introduced in the center of
a 9-mer dsDNA, monomer X decreases the thermal stability of
the duplex (ΔTm = −2.0 °C, entry 6) compared to the
unmodified duplex. However, this decrease is smaller than that
was seen for W in the similar context (ΔTm = −5.0 °C, entry
2). With three or four consecutive incorporations of monomer
X, stabilized duplexes were observed (entries 8 and 9, ΔTm/
mod =2.7−3.0 °C) in contrast to what was seen for monomer
W in the similar sequences (entries 2−4). These data further
support that the aminomethyl group is increasing the thermal
stability of the duplexes by partially neutralizing the negative
charge of the backbone.
A single incorporation of monomer Y featuring an inversed
triazole moiety causes a penalty of 6.5 °C (entry 10) in the
thermal stability of the modified DNA:RNA duplex compared
to the unmodified duplex (entry 1). This is in contrast to
monomer W which causes a drop of only 2.0 °C (entry 2). This
suggests that phenyltriazoles are better accommodated in
DNA:RNA duplexes when connected to the nucleoside
through the C-4 position of the triazole ring rather than the
N-1 position. A second adjacent incorporation of Y partially
compensates the thermal penalty from the single incorporation
(compare ΔTm/mod of −6.5 °C, entry 10, with ΔTm/mod of
−1.0 °C, entry 11) indicating favorable stacking interactions
between the two inversed phenyltriazole moieties (monomer
Y) in accordance with the property of the regular phenyltriazole
moiety (monomer W). A duplex with intact thermal stability
was observed for three consecutive incorporations of Y (entry
12), and four consecutive incorporations of Y gave a
DNA:RNA duplex with increased thermal stability (entry 13,
ΔTm/mod = +1.4 °C). These observations strongly suggest
that the inversed phenyltriazole group from monomer Y has the
ability to participate in stacking interactions when incorporated
more than once in the center of a DNA:RNA duplex. However,
the thermal penalty caused by the introduction of each
monomer Y (entry 10) counteracts the positive effect of
stacking interactions. When introduced in the center of the 9-
mer dsDNA (entry 10), monomer Y induced a drop of 8.0 °C
in the thermal stability as compared to 5.0 °C for monomer W
(entry 2). Thus, also in dsDNA, monomer W is better
accommodated than Y, and in duplexes with two to four
consecutive incorporations of Y the thermal penalty per
modification decreases only from 8.0 to 2.9 °C (entries 10−
13).
Table 1. Hybridization Data for the Duplexes Featuring the
Monomers W, X, Y, or Z
a
Tm (°C) (ΔTm per modification
b
(°C))
entry
ON sequences
compl RNA
31.0
compl DNA
33.0
1
5′-d(GTGTTTTGC)
5′-d(GTGTWTTGC)
5′-d(GTGTWWTGC)
5′-d(GTGWWWTGC)
5′-d(GTGWWWWGC)
5′-d(GTGTXTTGC)
5′-d(GTGTXXTGC)
5′-d(GTGXXXTGC)
5′-d(GTGXXXXGC)
5′-d(GTGTYTTGC)
5′-d(GTGTYYTGC)
5′-d(GTGYYYTGC)
5′-d(GTGYYYYGC)
5′-d(GTGTZTTGC)
5′-d(GTGTZZTGC)
5′-d(GTGZZZTGC)
5′-d(GTGZZZZGC)
5′-d(GTGWXWTGC)
5′-d(GTGXWXTGC)
5′-d(GTGXWWXGC)
5′-d(GTGWYWTGC)
5′-d(GTGWZWTGC)
c
c
2
29.0 (−2.0)
28.0 (−5.0)
30.5 (−1.5)
30.0 (−1.0)
32.0 (−0.3)
c
c
3
37.5 (+3.3)
43.0 (+4.0)
51.5 (+5.1)
33.5 (+2.5)
45.5 (+7.0)
59.0 (+9.3)
68.0 (+9.2)
c
c
c
4
c
5
6
31.0 (−2.0)
33.5 (+0.2)
41.0 (+2.7)
45.0 (+3.0)
25.0 (−8.0)
23.5 (−5.0)
21.5 (−3.8)
21.5 (−2.9)
30.5 (−3.0)
27.5 (−2.7)
28.0 (−1.7)
25.5 (−1.9)
33.0 (0.0)
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
24.5 (−6.5)
29.0 (−1.0)
31.0 (0.0)
36.5 (+1.4)
26.5 (−4.5)
25.0 (−3.0)
25.0 (−2.2)
27.0 (−1.0)
47.0 (+5.3)
50.0 (+6.3)
57.0 (+6.5)
38.0 (+2.3)
29.5 (−0.5)
35.5 (+1.2)
41.0 (+2.0)
25.5 (−2.5)
25.5 (−2.5)
a
Melting temperatures are obtained from the maxima of the first
derivatives of the melting curves (A260 vs temperature) recorded in a
medium salt buffer (Na2HPO4 (2.5 mM), NaH2PO4 (5 mM), NaCl
(100 mM), EDTA (0.1 mM), pH 7.0) using 1.5 μM concentrations of
each strand. Complementary target sequences are 5′-GCA AAA CAC
b
(RNA or DNA). The changes in melting temperature for each
modification as compared to the unmodified reference duplex (entry
c
1) are shown in parentheses. Data taken from refs 8 and 18.
incorporations of X, Y, or Z with W (entries 18−22) were
prepared. Constitution and purity of the synthesized ONs was
verified by MALDI-TOF MS analysis and ion-exchange
chromatography, respectively. The ONs were mixed in medium
salt buffer with complementary single-stranded RNA or single-
stranded DNA. Melting temperatures (Tm) of the resulting
duplexes were derived from the UV melting curves at neutral
pH 7 and compared with the Tm of the corresponding
unmodified duplex (Table 1, entry 1), and differences in
melting temperature per modification (ΔTm/mod) were
determined (Table 1, entries 2−22).
A single incorporation of monomer X in the center of a 9-
mer DNA:RNA duplex was found to increase the thermal
stability of the duplex by 2.5 °C (entry 6) compared to the
unmodified duplex (entry 1). For comparison, monomer W
decreases the duplex stability by 2.0 °C in the same sequence
The single incorporation of monomer Z led to the
destabilization of the modified DNA:RNA duplex by 4.5 °C
and of the dsDNA duplex by 3.0 °C compared to the
unmodified duplexes (entry 14). Therefore, Z is better
accommodated in both duplexes than Y. On the other hand,
two to four consecutive incorporations of Z in the center of the
2856
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
duplex did not compensate for this to the same extent since the
thermal penalty per modification decreases only from 4.5 to 1.0
°C in the DNA:RNA duplex (entries 14−17). This suggest that
stacking interactions between phenyl groups in the major
groove from monomer Z are present but weaker than for the
triazoles. In general, a similar behavior of Z was observed for
the corresponding dsDNA duplexes.
A graphical presentation of how the ΔTm per modification
varies with one to four incorporations of W, X, Y, or Z in the
center of a 9-mer DNA:RNA duplex is shown in Figure 2. In
monomer X is cooperatively participating in the stacking
interactions with W and that the aminomethyl group is
inducing further stabilization of the duplex, presumably by
partially neutralizing the negative charge of the backbone. In
comparison with the entries 7−9, however, it is also clear that
this additional increase in stability by the amino group is
independent of the placement in the sequence and can be
obtained uncompromised on two consecutive phenyltriazoles.
The ON featuring one Y in between two W’s on hybridization
with the complementary RNA strand forms a duplex with a
thermal stability higher that the unmodified duplex as well as
the duplex with three Ys in the center (compare entry 21 with 1
and 12, respectively). This indicates that the inversed
phenyltriazole from monomer Y can stack with the phenyl-
triazole from monomer W. However, a drop of 5.0 °C in the Tm
was observed when compared to the duplex with three
consecutive incorporations of monomer W (compare entry
21 with entry 4) in accordance to the negative influence of each
individual inverse phenyltriazole on thermal stability. The
combination of a monomer Z in between two W’s leads to a
DNA:RNA duplex with neutral thermal stability indicating that
the phenyl group is hampering the stacking in the major group
significantly.
Finally, we investigated the fidelity in the RNA recognition
for the ONs, which form the most stable duplexes with
complementary RNA targets, i.e., with four consecutive
incorporations of X or with mixed incorporations of W and
X (Table 2). These ONs were hybridized with RNA targets
featuring a centrally placed mismatch nucleotide, and as
expected, mismatched duplexes were found to be less stable
than the matched duplex. The C mismatch nucleotide was best
discriminated (ΔTm in the range of −20.0 to −27.5 °C)
followed by the U mismatch nucleotide (ΔTm in the range of
−15.0 to −20.5 °C) and the G (ΔTm in the range of −8.5 to
−11.0 °C). This is parallel to what was observed for monomer
W and other 5-triazole substituted 2′-deoxyuridine monomers
studied.8,18,20
Figure 2. Graphical illustration of ΔTm vs the number of
incorporations.
going from one incorporation of the monomers W, X, or Y to
two adjacent incorporations, ΔTm/mod increases by around 5
°C. This clearly suggests for the similar strong stacking
efficiency of regular phenyltriazole (monomers W and X) and
inversed phenyltriazole (monomer Y) in the major groove. In a
similar context, monomer Z induced a smaller increase of only
1.5 °C in the ΔTm/mod value indicating a favorable but smaller
stacking interaction between the phenyl groups. In general, the
effect of the third and fourth incorporation was found to be
similar for W, X, or Y, reflecting their similar efficiency to stack.
The ΔTm/mod for X is found to be around 4−5 °C higher than
for W in all the studied duplexes, demonstrating the positive
effect of the aminomethyl group on the thermal stability of the
duplexes.
Next, ONss with mixed consecutive incorporations of
monomer X, Y, or Z with W were studied in order to
illuminate the interplay between the different monomers
(Table 1, entries 18−22). ONs featuring mixed consecutive
incorporations of monomer W and X form duplexes with
complementary RNA or DNA strands, which are generally
more thermally stable than the duplexes featuring the same
number of modifications of W (compare entries 18 and 19 with
entry 4, and entry 20 with 5). These data support that
CD Spectroscopy. The global conformation of DNA:RNA
duplexes featuring four consecutive incorporations of mono-
mers X, Y, or Z (Table 1, entries 9, 13, and 17) and four
consecutive mixed incorporations of X with W (entry 20) were
investigated by CD spectroscopy. CD spectra were recorded at
10 °C in a medium salt buffer by using 3.0 μM of each strand
(Figure 3). As expected, the CD spectrum of the unmodified
DNA:RNA duplex (entry 1) was found to adopt intermediate
A/B-type geometry with characteristic bands from both
geometries. Thus, a positive band at 265 nm (A-type) with a
shoulder at 282 nm (B-type) was observed along with a positive
band at 220 nm (B-type) and negative bands at 247 nm (B-
Table 2. Hybridization Data for the Mismatched Duplexes
a
b
Tm /°C (ΔTm /°C ) 3′-r(CAC ABA ACG)
entry
1
ON sequences
B = A
51.5
B = C B = G
B = U
c
c
c
5′-d(GTGWWWWGC)
24.0
(−27.5)
42.0
42.0
(−9.5)
57.0
31.0
(−20.5)
49.0
2
3
5′-d(GTGXXXXGC)
5′-d(GTGXWWXGC)
68.0
57.0
(−26.0)
37.0
(−11.0)
48.5
(−19.0)
42.0
(−20.0)
(−8.5)
(−15.0)
a
b
c
See Table 1. Mismatch studies. The changes in melting temperature as compared to the matched duplex are shown in parentheses. Data taken
from ref 8.
2857
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
accuracy of calculation (MP2 using 6-31G(d,p) basis sets), the
2′-deoxyribose moieties were replaced by methyl groups, and
the phenyltriazoles were replaced by methyltriazoles, to get the
simplified model compounds W′/X′, Y′, and Z′.
The obtained energy profiles are shown in Figure 5. In
agreement with our former results,8 the energy profile for W′/
Figure 3. CD spectra of selected DNA:RNA duplexes. Entry numbers
refer to Table 1.
type) and 210 nm (A-type). After establishing the geometry of
the unmodified duplex as the mixed A/B type, the effect of the
modified monomers on the duplex geometry was studied. The
duplex featuring four consecutive incorporations of X (entry 9)
displayed a positive band at 265 nm and two negative bands at
251 and 210 nm suggesting the overall A/B type duplex
geometry. However, the positive band at 220 nm observed in
the CD spectrum of the unmodified duplex was very small. This
is in accordance with what was seen for monomer W (entry
5).8 Furthermore, a very broad band with a center at around
311 nm and a new negative band at 230 nm were also observed
(entry 9). These new bands observed for X were not seen in
the CD spectrum of W in the similar sequence context,8 and
their appearance might be due to local changes in the
conformation of the duplex by interactions of the aminomethyl
group with the backbone. The CD spectrum recorded for the
duplex featuring mixed incorporations of W and X (entry 20)
was very similar. The CD spectrum and thereby the overall
duplex geometry of the duplex featuring four consecutive
incorporations of Y (entry 13) was found to be generally similar
to the unmodified duplex with an intense positive band at 265
nm, a shoulder at 280 nm and a negative band at 210 nm. Small
variations were observed as the band at 265 nm is more intense,
the band at 220 nm is smaller and the negative band at 247 nm
is shifted toward shorter wavelength and appeared at 238 nm. It
is interesting to note that for monomers W and X, this band is
shifted toward longer wavelength and appeared at 251 nm. The
CD spectrum of the duplex with four consecutive incorpo-
rations of Z (entry 17) was very similar to the one from the
unmodified duplex indicating no major influence on duplex
geometry from this modification.
Figure 5. Dihedral MP2 scans of the bonds connecting the triazoles or
the phenyl to the pyrimidine (see Figure 4) using 2° increments.
X′ has a global minimum in θ = 180°, corresponding to a
strong preference for the molecule to adopt a coplanar
conformation with a CH−O interaction from the triazole-C5
to the uracil-O4. When the triazole ring is inversed as in Y′, the
preferred conformation surprisingly becomes antiperiplanar,
with the global minimum shifted about 32° away from
coplanarity (φ = 148° and 212°). The coplanar conformation
(φ = 180°) is, however, only slightly higher in energy (+0.86
kJ/mol) and the curve of Y′ is essentially flat from φ = 145° to
φ = 215°, indicating that Y′ has a higher degree of freedom
compared to W′. The calculated barrier to rotation happens to
be almost identical for W′/X′ and Y′ (∼47 kJ/mol). When the
triazole ring is entirely removed and replaced by a phenyl group
as in Z′, the coplanar conformations (ω = 0° and ω = 180°)
become global maxima and the perpendicular conformations
(ω = 90° and ω = 270°) are local maxima. In fact, the lowest
energy structures correspond to an inclination of 45° or 135° of
the phenyl substituent. It is also significant that the rotational
barrier for Z′ is much smaller (∼14 kJ/mol) than that of the
triazolated molecules. Altogether, these results suggest that
while W and X should strongly prefer coplanar conformations,
Y is more flexible around coplanarity, whereas the phenyl ring
of Z is not prone to align in the same plane as the pyrimidine
indicating that it could lead to poor stacking interaction in
duplexes.
Molecular Modeling. In order to get a better under-
standing of the preferred conformations of the monomers W,
X, Y, and Z, we used ab initio calculations on three model
structures to find the torsions θ, φ, and ω (see Figure 4) of the
5-substituents relative to the uracil. The 5-substituent should
align in the same plane as the uracil for π-overlap such that the
most efficient stacking is achieved. In order to reduce the
computational burden to a feasible level while retaining the
To get more information about how well multiple
incorporations of W, X, Y, and Z are accommodated in the
DNA:RNA duplexes, molecular dynamics (MD) simulations
were performed. Specifically, these simulations may provide an
indication of the efficiency of stacking in the major groove as
well as the effects of the modifications on the duplex geometry.
In the case of X, an idea about the positioning of the positive
charges that arise from the methylamines might also be
obtained. Using a simulation time of 5 ns, a reasonable
Figure 4. Structures of model compounds W′/X′, Y′, and Z′.
2858
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
Figure 6. Global minimum structures obtained by 5 ns MD simulations of the DNA:RNA duplexes corresponding to entries 1, 5, 9, 13, and 17 in
Table 1. The last was performed at 10 °C and the other four at 27 °C. The backbone is shown in red, nucleobases in green, and 5-substituents in
space-filling blue, except for the amino groups, which are shown in space-filling green.
estimation of the global minimum for the duplex can be found.
The structure of the 9-mer duplex featuring four consecutive W
residues (entry 5, Table 1) have been modeled before,8 and we
decided to use this minimized structure as a benchmark for the
duplexes containing four residues of either X, Y, or Z (entries 9,
13, and 17, Table 1). The obtained global minima structures are
shown in Figure 6. In addition, the duplex containing three X’s
(entry 8) or two Zs (entry 15) as well as the duplex containing
mixed residues of X and Z (entry 20) were studied (see the
structures in Figure S1, Supporting Information).
neutralize the phosphate charge of the RNA strand. In general,
all simulations of duplexes containing X shows coplanarity of
the triazole and the uracil corresponding to torsional angles θ
(Figures 4 and 5) of 160−181°, which is, however, deviating
more from the optimal 180° than in the case of the W’s (θ =
173−185°). In the case of Y (Figure 6d), there is indeed
stacking between the Y’s as suggested by the Tm values,
although the efficiency of the stacking may not be as clear as in
the case of the W residues (Figure 6b). Specifically, there
appears to be a slight bending of the two lowermost residues.
The torsional angles φ, however, are 173−183°. Most
strikingly, the introduction of four Z’s appears to induce a
slight unwinding of the helix (Figure 6e), probably caused by
the tilting of the phenyl substituents. Hence, the angles of the
phenyluracil are from 211 to 217° not far from the optimal as
seen in Figure 5. Therefore, efficient stacking is probably
prohibited.
In our previous studies,8 four consecutive W’s in the 9-mer
duplex have been shown to give efficient stacking in the major
groove, and this is now augmented by the present MD
simulation (Figure 6b). When four consecutive X’s are
introduced (Figure 6c), the amino groups of the 5-substituents
appear to neutralize the negatively charged phosphates of the
complementary DNA strand (for a close-up, see Figure S2,
Supporting Information). In addition, there appears to be
efficient stacking between the three upmost X’s in a similar
degree to the W’s, but notably, the 5′-end X appears to rotate
around the phenyl−triazole bond and slide to the side in order
to neutralize the backbone charge rather than engage in
stacking interactions. This might explain why the fourth
incorporation does not improve the relative increase in thermal
stability even further (see Figure 2). With three consecutive X’s
(entry 8) (Figure S1, Supporting Information), all three
participate in stacking as well as neutralization of phosphates
in the same way as the first three X’s from the 3′-end of the four
in entry 9. When the XWWX motif was applied to the MD
simulation (Figure S1, Supporting Information), a certain
bending of the substituents was observed that allowed some
degree of stacking. While the 3′-end X residue still neutralizes
the phosphate charge of the DNA strand like in the case of four
X’s (Figure 6c), the 5′-X residue of the XWWX motif
unexpectedly appears to “cross-link” the two strands to
DISCUSSION
■
In this study, we have introduced three new 5-substituted-2′-
deoxyuridine nucleotide monomers, X, Y, and Z, into ONs.
The syntheses of all three are easy and based on well-
established organometallic reactions, either the CuAAC or the
Suzuki−Miyaura reaction, and the phosphoramidite building
blocks are obtained in only four to six steps from commercially
available nucleoside precursors. From the hybridization study it
is clear that monomer X displays by far the highest melting
temperatures when introduced in both DNA:RNA and dsDNA
duplexes. In particular, DNA:RNA duplexes are extremely
stable with consecutive incorporations of X and consistently
found to be more stable than corresponding duplexes modified
with W. This demonstrates that the very strong RNA-targeting
behavior found for W and analogues thereof8,18−20 can be even
further improved by the additive effect of the stacking
2859
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
interactions from the 5-phenyltriazole moiety and of the
aminomethyl group, which presumably leads to charge
neutralization of the backbone. This further consolidates the
potential of the phenyltriazole pyrimidine modification for the
development of antisense oligonucleotides, as the RNA affinity
can be tuned by the number of consecutive incorporations of
monomer W as well as by the number of aminomethyl groups
by using monomer X. It is an important observation that by the
use of X in contrary to W, an increased DNA:RNA duplex
stability can even be obtained by a single incorporation of the
phenyltriazole (see Table 1 and Figure 2). Future structural and
physicochemical studies are needed to finally confirm the
charge neutralization from the aminomethyl group of X. We
have earlier proved that also the cytidine analogue with a
phenyltriazole substituent can increase the duplex stability so
that mixed purine RNA sequences can be targeted.19
Furthermore, the phenyltriazoles significantly stabilize the
ONs against degradation under physiological conditions.19
Obviously, the potential of the regioisomeric monomer Y is
hampered by the large decrease in duplex stability associated
with each incorporation. Even though this is compensated
(Table 1, Figure 2) by efficient stacking interactions, the
potential for antisense purposes is small as compared to W.
Nevertheless, it is a very important observation that the
substitution pattern of the triazole substituents displays a large
impact on the resulting duplexes. Hence, several studies have
used the CuAAC reaction and a triazole at the 5-position of
uracil as the attaching point for, e.g., labeling ONs mostly
through the 5-ethynyluracil27 but recently also through the in
situ generated 5-azidouracil moiety.28 Hence, it is important to
know that the thermal penalty of introducing a single triazole
into a dsDNA or a DNA:RNA duplex is very high if the triazole
is “inversed” (as in Y) and somewhat smaller if not (as in W
and X). Likewise, direct attachment of aromatics, like
substituted phenyl groups, has been applied before for different
purposes, incl. labeling.29 Hereby, it is an interesting
observation that an isolated Z is better accommodated in the
duplexes than Y, and in the case of dsDNA, also better than W.
Therefore, Suzuki−Miyaura couplings of phenyls can be a
better choice than the CuAAC reaction for labeling ONs if the
thermal stability of the subsequent duplexes is an issue.
In the present study, we have applied CD spectra and
molecular modeling in order to enlighten the different behavior
of the monomers X, Y, Z, and W in DNA:RNA duplexes.
Hence, it is very clear that the stacking of the substituents in the
major groove is the key to the high stability of the duplexes. For
the phenyltriazoles in W and X, there is a very strong
preference for a coplanar conformation (see Figure 5) due to a
CH−O interaction, and the MD simulation confirms the
coplanarity also in the duplexes. Furthermore, this is not
hampered by the introduction of aminomethyl groups and the
subsequent neutralization of the phosphates. In fact, the
synergy of stacking and charge neutralization leads to the
extraordinary duplex stability, and this works despite
consecutive incorporations of several amino groups. Monomer
Y is apparently not restricted in the coplanar conformation (see
Figure 5) and prefers in fact a 45° angle between the two
aromatics. This might be the reason why the first introduction
of Y is very unfavorable for the duplex, whereas stacking
interactions between several substituents can force these into a
coplanar conformation and eventually stabilize the duplexes.
The fact that the coplanar conformation of the phenyl uracil is a
local maximum (see Figure 5) is closely connected to the very
imperfect stacking interactions of Z. The MD simulations (see
Figure 6) support this and show a duplex conformation that is
closer to the unmodified DNA:RNA duplex than to the
triazole-modified duplexes, an indication that is fully confirmed
by the CD spectrum. The small compensation in duplex
stability obtained for several incorporations of Z might be due
to hydrophobic effects rather than stacking interactions.
CONCLUSION
■
The combination of stacking interactions from the phenyl-
triazole group and charge neutralization through aminomethyl
substituents in the major groove led to extraordinary stable
DNA:RNA duplexes using consecutive incorporations of
monomer X. This makes the combination of X and W and
analogues thereof a strong and simple tool for the search for
therapeutic oligonucleotides. Furthermore, it shows that with
the right linkage, long arrays of aromatic groups can be
arranged in the duplex major groove without compromising
duplex stability and structure. Although the “inversed” triazole
monomer Y is efficiently synthesized and can partake in similar
stacking interactions in the major groove, the thermal penalty
for a single incorporation is higher. The phenyl group of Z is
better accommodated in the duplexes, although stacking
interactions in the major groove are weaker. Hereby, this
study presented the first direct comparison of the stacking
interactions of the 5-phenyltriazole and 5-phenyl groups
establishing a clear correlation between coplanarity of the
aromats, the ability for stacking and the increase in duplex
stability. The importance of the C-connected triazole group at
the 5-position as in W and X for optimum stacking interactions
is clear. Therefore, the present study demonstrates how
delicately the decoration of nucleic acids is connected to
duplex structure and stability.
EXPERIMENTAL SECTION
■
General Methods. All commercial reagents were used as supplied
except CH2Cl2, which was distilled prior to use. Anhydrous solvents
were dried over 4 Å activated molecular sieves (CH2Cl2, pyridine, and
1,2-dichloroethane) or 3 Å activated molecular sieves (DMF,
CH3CN). Reactions were carried out under argon when anhydrous
solvents were used. All mixtures of solvents were prepared as a volume
to volume ratio (v/v). All reactions were monitored by TLC using
Merck silica gel plates (60 F254). To visualize the plates, they were
exposed to UV light (254 nm) and/or immersed in a solution of 5%
H2SO4 in MeOH (v/v) followed by charring. Column chromatog-
raphy was performed with silica gel 60 (particle size 0.040−0.063 μm,
Merck). Silica gel was pretreated with 1% pyridine in CH2Cl2 (v/v) for
the purification of 4,4′-dimethoxytrityl-protected nucleosides. 1H
NMR, 13C NMR and 31P NMR spectra were recorded at 400 MHz,
101, and 162 MHz, respectively. Chemical shift values (δ) are reported
in ppm relative to either tetramethylsilane (1H NMR) or the
deuterated solvents as internal standard for 13C NMR (δ: CDCl3
77.16 ppm, DMSO-d6 39.52 ppm), and relative to 85% H3PO4 as
external standard for 31P NMR. 2D spectra (1H−1H COSY, and
1H−13C HSQC) have been used in assigning 1H and 13C NMR
signals. High-resolution ESI (quadrupole) mass spectra were recorded
in positive ion mode.
N-(3-Bromobenzyl)-2,2,2-trifluoroacetamide. A solution of 3-
bromobenzylamine (0.63 mL, 5 mmol) and triethylamine (2.5 mL,
15 mmol) in CH2Cl2 (30 mL) was stirred at 0 °C, and a solution of
trifluoroacetic anhydride (1.42 mL; 10 mmol) in CH2Cl2 (5 mL) was
added. The reaction mixture was stirred at room temperature for 18 h
and then washed sequentially with a 1 M aqueous solution of HCl and
a saturated aqueous solution of NaHCO3. The combined organic
fraction was dried (Na2SO4) and then concentrated under reduced
pressure. The residue was recrystallized from petroleum ether to afford
2860
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
the product as white needles (1.06 g; 75%): Rf 0.4 (ethyl acetate/
petroleum ether, 4:1 v/v); mp 59 °C; H NMR (400 MHz, DMSO-
CH2Cl2); 1H NMR (CDCl3, 400 MHz) δ 9.27 (br s, 1H, NH), 7.51 (s,
1H, H6), 7.42−7.40 (m, 2H, DMTr), 7.33−7.23 (m, 7H, DMTr),
6.86−6.83 (m, 4H, DMTr), 6.28 (t, 1H, J = 6.6 Hz, H1′), 4.55 (m, 1H,
H3′), 4.05 (m, 1H, H4′), 3.79 (s, 6H, 2 × OCH3), 3.41 (dd, J = 10.4,
3.2 Hz, 1H, H5′), 3.33 (dd, J = 10.4, 3.2 Hz, 1H, H5′), 2.47 (m, 1H,
H2′), 2.25 (m, 1H, H2′); 13C NMR (CDCl3, 100 MHz) δ 159.5 (C4),
158.8 (DMTr), 149.0 (C2), 144.5, 135.6, 135.4, 130.22, 130.20, 128.1
(DMTr), 127.4, 127.2 (C6, DMTr), 115.6 (C5), 113.4, (DMTr), 87.3
(DMTr), 86.4 (C4′) 85.5 (C1′), 72.2 (C3′), 63.4 (C5′), 55.4
(OCH3), 41.3 (C2′); HR ESI MS m/z 594.1948 ([M + Na]+,
C30H29N5O7Na+ calcd 594.1959).
5′-(4,4′-Dimethoxytrityl)-5-(4-phenyl)-1,2,3-triazol-1-yl)-2′-deox-
yuridine (6). To a stirred solution of nucleoside 5 (170 mg, 0.29
mmol) and phenylacetylene (100 μL, 0.91 mmol) in a mixture of H2O,
t-BuOH, and pyridine (5 mL, 2:2:1, v/v/v) was added sodium
ascorbate (60 mg, 0.30 mmol) and CuSO4·5H2O (25 mg, 0.10 mmol),
and the reaction mixture was stirred at room temperature for 3 h. Ethyl
acetate (30 mL) was added, and the mixture was washed with H2O (20
mL) and a saturated aqueous solution of NaHCO3 (20 mL). The
combined aqueous phase was extracted with ethyl acetate (2 × 15
mL), and the combined organic phase was dried (Na2SO4) and
concentrated under reduced pressure. The residue was purified by
column chromatography (0−5% MeOH in CH2Cl2) to afford the
nucleoside 6 (130 mg, 65%) as a pale yellow solid: 1H NMR (DMSO-
d6, 400 MHz) δ 12.15 (s, 1H, NH), 8.53 (s, 1H, CH triazole), 8.27 (s,
1H, H6), 7.83 (d, J = 7.6 Hz, 2H, Ph), 7.46 (t, J = 7.6 Hz, 2H, Ph),
7.38−7.34 (m, 1H, Ph), 7.29−7.27 (m, 2H, DMTr), 7.23−7.19 (m,
2H, DMTr), 7.19−7.12 (m, 5H, DMTr), 6.82−6.79 (m, 4H, DMTr),
6.18 (t, J = 6.4 Hz, 1H, H1′), 5.37 (d, J = 4.4 Hz, 1H, 3′−OH), 4.27
(m, 1H, H3′), 3.95 (m, 1H, H4′), 3.67 (s, 6H, OCH3), 3.25 (dd, J =
10.4, 5.2 Hz, 1H, H5′), 3.11 (dd, J = 10.4, 2.6 Hz, 2H, H5′), 2.39−
2.30 (m, 2H, H2′); 13C NMR (DMSO-d6, 100 MHz) δ 158.6 (C4),
157.9, (DMTr), 149.1 (C2), 146.2 (C triazole), 144.5 (DMTr), 136.9
(C6), 135.3, 135.1, 130.1, 129.5, 129.4, 128.8, 127.9, 127.7, 127.4,
126.5, (Ph, DMTr), 125.2 (Ph), 123.0 (CH triazole), 113.0 (DMTr,
C5), 85.9, 85.7, 85.5 (C1′, C4′, DMTr), 70.2 (C3′), 63.5 (C5′), 54.8
(OCH3), 40.0 (C2′); HR ESI MS m/z 696.2398 ([M + Na]+,
C38H35N5O7Na+ calcd 696.2429).
3′-O-(P-(2-Cyanoethoxy)-N,N′-diisopropylaminophosphinyl)-5′-
O-(4,4′-dimethoxytrityl)-5-(4-phenyl)-1,2,3-triazol-1-yl)-2′-deoxyuri-
dine (7). The nucleoside 6 (120 mg, 0.17 mmol) was coevaporated
with anhydrous 1,2-dichloroethane (2 × 5 mL) and dissolved in
anhydrous CH2Cl2 (5 mL). N,N-Diisopropylammonium tetrazolide
(72 mg, 0.36 mmol) and 2-cyanoethyl N,N,N′,N′-tetraisopropylphos-
phoramidite (120 μL, 0.35 mmol) were added, and the mixture was
stirred at room temperature for 14 h. The reaction was quenched by
the addition of 99.9% EtOH (4 drops), and the mixture was
concentrated under reduced pressure. The residue was purified by
column chromatography (0−2% MeOH in CH2Cl2) to afford the
product 7 (105 mg) as a white foam containing some hydrolyzed
phosphitylation reagent. The compound was used without further
purification in the ON synthesis: Rf 0.5 (3% MeOH in CH2Cl2); 31P
NMR (CDCl3, 162 MHz) δ 149.1, 148.8; HR ESI MS m/z 874.3685
([M + H]+, C47H53N7O8PH+ calcd 874.3688).
1
d6) δ = 10.04 (br s, 1H, NH), 7.51−7.48 (m, 2H, Ar), 7.36−7.29 (m,
3
2H, Ar), 4.41 (d, JHH = 5.9 Hz, 2H, CH2); 13C NMR (101 MHz,
2
DMSO) δ = 156.5 (q, JCF = 36 Hz, CO), 140.2, 130.7, 130.2,
130.2, 126.4, 121.7 (Ar), 115.9 (q, 1JCF = 288 Hz, CF3), 42.0 (d, 4JCF
=
5 Hz, CH2); 19F NMR (376 MHz, DMSO-d6) δ = −74.3 (s, CF3); HR
ESI MS m/z = 281.9741 ([M + H+, 81Br], C9H8BrF3NO+ calcd
281.9736).
5′-O-(4,4′-Dimethoxytrityl)-5-(1-(3-((2,2,2-trifluoroacetamido)-
methyl)phenyl)-1,2,3-triazol-4-yl)-2′-deoxyuridine (2). To a stirred
solution of N-(3-bromobenzyl)-2,2,2-trifluoroacetamide (423 mg, 1.5
mmol) in a mixture of ethanol and water (7:3, v/v, 3.0 mL) were
added NaN3 (97 mg, 1.5 mmol), CuI (28 mg, 0.15 mmol), N,N′-
dimethylethylenediamine (20 mg, 0.22 mmol), and sodium ascorbate
(15 mg, 0.07 mmol). The mixture was stirred under microwave
irradiation (100 W) at 100 °C for 60 min. N,N′-Dimethylethylenedi-
amine (20 mg, 0.22 mmol), sodium ascorbate (15 mg, 0.07 mmol),
CuI (28 mg, 0.15 mmol), and nucleoside 1 (332 mg, 0.60 mmol) were
added, and the mixture was stirred under microwave irradiation at 100
°C for 30 min. The reaction mixture was extracted with ethyl acetate
(2 × 15 mL), and the combined organic phase was dried (Na2SO4)
and concentrated under reduced pressure. The residue was purified by
silica gel column chromatography (0−5% MeOH in CH2Cl2) and then
coevaporated with toluene (10 mL) to afford the product 2 as a white
1
foam (348 mg, 72%): Rf 0.5 (5% MeOH/CH2Cl2, v/v); H NMR
(DMSO-d6, 400 MHz) δ 11.84 (br, 1H, NH(U)), 10.10 (t, J = 5.9 Hz,
1H, NHCO), 8.83 (s, 1H, CH triazole), 8.42 (s, 1H, H6), 7.89 (s, 1H,
Ar), 7.85 (d, J = 8.0 Hz, 1H, Ar), 7.58 (t, J = 8.0 Hz, 1H, Ar), 7.42−
7.37 (m, 3H, DMTr, Ar), 7.29−7.21 (m, 6H, DMTr), 7.19−7.11 (m,
3H, DMTr), 6.84−6.81 (m, 4H, DMTr), 6.20 (t, J = 6.4 Hz, 1H, H1′),
5.38 (d, J = 4.6 Hz, 1H, 3′−OH), 4.53 (d, J = 5.9 Hz, 2H, CH2NH),
4.22 (m, 1H, H3′), 3.97 (m, 1H, H4′), 3.66 (s, 6H, 2 × OCH3), 3.25−
3.22 (m, 2H, H5′), 2.30−2.26 (m, 2H, H2′); 13C NMR (DMSO-d6,
2
100 MHz) δ 161.1, 158.0, 156.4 (q, JCF = 36 Hz, COCF3), 149.5,
144.8, 139.9, 139.5, 136.6 (C6), 136.3, 135.5, 135.4, 130.1, 129.7,
129.6, 127.7, 127.6, 126.5, 119.9 (CH triazole), 119.1, 119.0, 115.9 (q,
1JCF = 288 Hz, CF3), 113.1, 104.6, 85.7 (C4′), 85.3 (C1′), 70.4 (C3′),
54.9 (OCH3), 42.2 (CH2), 39.9 (C2′); 19F NMR (DMSO-d6, 376
MHz). δ −74.29 (CF3); HR ESI MS m/z 821.2516 ([M + Na]+,
C41H37F3N6O8Na+ calcd 821.2517).
3′-O-(P-(2-Cyanoethoxy)-(N,N-diisopropylamino)phosphinyl)-5′-
O-(4,4′-dimethoxytrityl)-5-(1-(3-((2,2,2-trifluoroacetamido)methyl)-
phenyl)-1,2,3-triazol-4-yl)-2′-deoxyuridine (3). Nucleoside 2 (180
mg, 0.25 mmol) was coevaporated with anhydrous 1,2-dichloroethane
(2 × 5 mL) and dissolved in anhydrous CH2Cl2 (5.0 mL). N,N-
diisopropylethylamine (220 μL, 1.25 mmol), and 2-cyanoethyl-N,N-
diisopropylamino chlorophosphite (180 μL, 0.75 mmol) were added
and the reaction mixture was stirred at room temperature for 2 h. The
reaction was quenched by the addition of 99.9% EtOH (2−3 drops),
and the mixture was concentrated under reduced pressure. The residue
was purified by column chromatography (0−2% MeOH in CH2Cl2) to
afford the product 3 (125 mg, 54%) as a white foam: Rf 0.4 (2%
MeOH in CH2Cl2); 31P NMR (CDCl3, 162 MHz) δ 149.1, 148.7; HR
ESI MS m/z 1021.3625 ([M + Na]+, C50H54F3N8O9PNa+ calcd
1021.3596).
5-Azido-5′-(4,4′-dimethoxytrityl)-2′-deoxyuridine (5). Nucleoside
4 (270 mg, 1.0 mmol) was coevaporated with anhydrous pyridine (2 ×
10 mL) and redissolved in the same solvent (10 mL). 4,4′-
Dimethoxytrityl chloride (410 mg, 1.20 mmol) was added, and the
reaction mixture was stirred at room temperature for 14 h. The
reaction was quenched by the addition of EtOH (99.9%, 3−4 drops),
and the mixture was concentrated under reduced pressure. The residue
was coevaporated with toluene (2 × 10 mL), dissolved in CH2Cl2 (30
mL), and washed with a saturated aqueous solution of NaHCO3 (2 ×
20 mL). The combined aqueous phase was extracted with CH2Cl2 (2
× 20 mL), and the combined organic phase was dried (Na2SO4) and
concentrated under reduced pressure. The residue was purified by
column chromatography (0−5% MeOH in CH2Cl2) to give the
product 5 (240 mg, 42%) as a yellow solid: Rf 0.4 (5% MeOH in
5′-(4,4′-Dimethoxytrityloxymethyl)-5-phenyl-2′-deoxyuridine
(10). To a stirred solution of 5-phenyl-2′-deoxyuridine 9 (268 mg, 0.88
mmol) in a mixture of anhydrous CH3CN and pyridine (17 mL, 1:1,
v/v) was added 4,4′-dimethoxytrityl chloride (462 mg, 1.36 mmol).
The mixture was stirred at room temperature for 4 h and then
quenched by the addition of MeOH (4 mL). The mixture was
concentrated under reduced pressure, and the residue was dissolved in
CH2Cl2 (100 mL). The organic phase was washed with water (20 mL)
and a saturated aqueous solution of NaHCO3 (20 mL). The organic
phase was dried (MgSO4) and concentrated under reduced pressure.
The residue was purified by flash chromatography (10−100% ethyl
acetate in cyclohexane) to afford the product 10 (254 mg, 47%) as a
1
white amorphous solid: Rf 0.30 (5% MeOH in CH2Cl2); H NMR
(400 MHz, DMSO-d6) δ 11.60 (s, 1H, NH), 7.66 (s, 1H, H6), 7.30−
7.15 (m, 14H, Ph, DMTr), 6.77 (m, 4H, DMTr), 6.23 (t, J = 6.8 Hz,
1H, H1′), 5.34 (d, J = 4.4 Hz, 1H, 3′−OH), 4.29 (m, 1H, H3′), 3.93
2861
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
(m, 1H, H4′), 3.69 (s, 6H, 2 × OCH3), 3,19 (dd, J = 10.0 Hz, 4.8 Hz,
1H, H5′), 3.13 (dd, J = 10.0 Hz, 4.8 Hz, 1H, H5′), 2.34 (m, 1H, H2′),
2.23 (m, 1H, H2′); 13C NMR (101 MHz, DMSO-d6) δ 162.0 (C4),
158.0 (DMTr), 149.9 (C2), 144.6 (DMTr), 137.1 (C6), 135.2, 132.7,
129.6, 129.5, 128.2, 127.9, 127.8, 127.5, 127.2, 126.6 (DMTr, Ph),
114.1 (C5), 113.1 (DMTr), 85.7 (C4′), 84.7 (C1′), 70,6 (C3′), 63.8
(C5′), 55.0 (OCH3), 40.0 (C2′). HR ESI MS m/z 629.2248 ([M +
Na]+, C36H34N2O7Na+ calcd 629.2264).
3′-O-(P-(2-Cyanoethoxy)-N,N′-diisopropylaminophosphinyl)-5′-
(4,4′-dimethoxytrityl)-5-phenyl-2′-deoxyuridine (11). Nucleoside 10
(136 mg, 0.224 mmol) was coevaporated with anhydrous 1,2-
dichloroethane (2 × 5 mL) and dissolved in anhydrous CH2Cl2.
N,N-Diisopropylammonium tetrazolide (89 mg, 0.448 mmol) and 2-
cyanoethyl-N,N,N′,N′-tetraisopropylphosphoramidite (142 μL, 0.448
mmol) were added, and the mixture was stirred at room temperature
for 2 h. The reaction was quenched by the addition of 99.9% EtOH (1
mL), and the mixture was concentrated under reduced pressure. The
residue was purified by column chromatography (0−5% MeOH in
CH2Cl2) to give the product 11 (110 mg) as a white foam containing
some hydrolyzed phosphitylation reagent. The compound was used
without further purification in the ON synthesis: Rf 0.70 (10%
CH3OH in CH2Cl2); 31P NMR (CDCl3, 162 MHz) δ 149.01, 148.52;
HR ESI MS m/z 829.3334 ([M + Na]+, C45H51N4O8PNa+ calcd
829.3337).
Oligonucleotide Synthesis. Oligonucleotide synthesis was
carried out on an automated DNA synthesizer following the
phosphoramidite approach. Synthesis of oligonucleotides was
performed on a 0.2 μmol scale using the amidites 3, 7, and 11 as
well as the corresponding commercial 2-cyanoethyl phosphoramidites
of the natural 2′-deoxynucleosides and the amidite for W.8 The
synthesis followed the regular protocol for the DNA synthesizer. For
the modified phosphoramidites, a prolonged coupling time of 15 min
was used. 1H-Tetrazole was used as the activator, and coupling yields
for all 2-cyanoethyl phosphoramidites were >90%. The 5′-O-DMT-
ON oligonucleotides were removed from the solid support by
treatment with concentrated aqueous ammonia at 55 °C for 16 h,
which also removed the protecting groups. The oligonucleotides were
purified by reversed-phase HPLC on a Waters 600 system using a
XBridge OST C18 column, 19 × 100 mm, 5 μm + precolumn:
XBridge 10 × 10 mm, 5 μm. Temperature 50 °C; Buffer A: 0.05 M
triethylammonium acetate pH 7.4. Buffer B: MeCN/H2O (3:1).
Program used: 2 min 100% A, 100%−30%:0%−70% A:B over 17 min,
4 min 100% B, 6 min 100% A. Flow 5 mL/min. All fractions
containing 5′-O-DMTr protected oligonucleotide were collected and
concentrated. The products were detritylated by treatment with 80%
aqueous acetic acid (100 μL) for 30 min at room temperature and
subsequently diluted with water (100 μL). Aqueous solutions of
sodium acetate (3 M, 15 μL) and sodium perchlorate (5 M, 15 μL)
were added followed by acetone (1 mL). Oligonucleotides were
allowed to precipitate overnight at −20 °C. After centrifugation at
12000 rpm, 10 min at 4 °C, the supernatant was removed, and the
pellet was washed with cold acetone (2 × 1 mL), dried by heating (55
°C) under a flow of nitrogen, and dissolved in pure water (1000 μL).
Constitution of the synthesized ONs was verified by ion-exchange
chromatography and MALDI-MS. The concentration was determined
by UV at 260 nm. The extinction coefficients of the modified ONs
were estimated from a standard method using micromolar extinction
coefficients for the monomers. For Y this was determined on a small
amount of detritylated 6; ε260 = 14.0. For X and Z were used the
coefficients of W8 and of dT, respectively.
were found to be reversible. All determinations are averages of at least
duplicates within 0.5 °C.
CD Spectroscopy. CD spectra (200−350 nm) were recorded on a
JASCO J-815 CD-spectrometer as an average of 5 scans using a split of
2.0 nm, and a scan speed of 50 nm/min. Samples were dissolved in a
medium salt buffer containing 2.5 mM Na2HPO4, 5 mM NaH2PO4,
100 mM NaCl, and 0.1 mM EDTA at pH = 7.0 with 3.0 μM
concentrations of the two complementary oligonucleotide sequences,
heated to 80 °C, and cooled to 10 °C. Quartz optical cells with a path
length of 5.0 mm were used.
Ab Initio Calculations. The energy profiles were calculated by
relaxed coordinate scans of the dihedral angles between the uracil ring
and the planar 5-substituents. The computations were performed at
the LMP2 level with the 6-31G(d,p) basis set using the Jaguar V7.8
software package within the Maestro Molecular Modeling interface
V9.2.109. The torsional sampling was performed by a 2° increment of
the dihedral angle (from 0° to 360°) to obtain 181 minimized
structures, the relative potential energies of which were plotted against
the corresponding dihedral angle.
Global Minimum Structures. The global minimum structures
were found from 5 ns molecular dynamics simulations using the all-
atom AMBER* force field and the Polak-Ribiere Conjugate Gradient
(PRCG) method. The duplexes were built in B-type geometries, and
initial Monte Carlo torsional samplings (MCMM) were performed to
generate 1000 structures, which were minimized into local minima.
The lowest energy structures for each of the duplexes were used for
subsequent MD simulations. The MD simulations were performed at
300 K. SHAKE all bonds to hydrogen was imposed in order to
increase the time step to 2.2 fs, and an equilibration time of 100 ps was
used to stabilize the calculations. A multiple minimization of the 500
sampled structures was performed to obtain a converged global
minimum structure.
ASSOCIATED CONTENT
* Supporting Information
■
S
MALDI-TOF data for oligonucleotides. Additional MD-
simulations. Selected NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Fax: +45 66158780. E-mail: pouln@sdu.dk.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The project was supported by The Danish National Research
Foundation, The Danish Council for Independent Research/
Natural Sciences (FNU) and Technology and Production
Sciences (FTP), and Villum Kann Rasmussen Fonden.
REFERENCES
■
(1) Zamecnik, P. C.; Stephenson, M. L. Proc. Natl. Acad. Sci. U.S.A.
1978, 75, 280−284.
(2) Crooke, S. T.; Wickers, T.; Lima, W.; Wu, H. In Antisense Drug
Technology: Principle, Strategies and Applications, 2nd ed.; Crooke, S. T.,
Ed.; CRC: Boca Raton, 2008; pp 3−46.
(3) Kurreck, J. Eur. J. Biochem. 2003, 270, 1628−1644.
(4) Deleavey, G. F.; Damha, M. J. Chem. Biol. 2012, 937−954.
(5) Lin, K. −Y.; Jones, R. J.; Matteucci, M. J. J. Am. Chem. Soc. 1995,
117, 3873−3874.
Thermal Denaturation Experiments. UV melting experiments
were carried out on a UV spectrometer. Samples were dissolved in a
medium salt buffer containing 2.5 mM Na2HPO4, 5 mM NaH2PO4,
100 mM NaCl, and 0.1 mM EDTA at pH = 7.0 with 1.5 μM
concentrations of the two complementary sequences. The increase in
absorbance at 260 nm as a function of time was recorded while the
temperature was increased linearly from 10 to 75 °C at a rate of 1.0
°C/min by means of a Peltier temperature programmer. The melting
temperature was determined as the local maximum of the first
derivatives of the absorbance vs temperature curve. The melting curves
(6) Froehler, B. C.; Wadwani, S.; Terhorst, T. J.; Gerrard, S. R.
Tetrahedron Lett. 1992, 33, 5307−5310.
(7) Freier, S. M.; Altmann, K.-H. Nucleic Acids Res. 1997, 25, 4429−
4443.
(8) Kocalka, P.; Andersen, N. K.; Jensen, F.; Nielsen, P.
ChemBioChem 2007, 8, 2106−2116.
2862
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863

The Journal of Organic Chemistry
Article
(9) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002,
67, 3057−3064.
(10) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596−2599.
(11) Gramlich, P. M. E.; Wirges, C. T.; Manetto, A.; Carell, T.
Angew.Chem., Int. Ed. 2008, 47, 8350−8358.
(12) Amblard, F.; Cho, J. H.; Schinazi, R. F. Chem. Rev. 2009, 109,
4207−4220.
(13) El-Sagheer, A. H.; Brown, T. Chem. Soc. Rev. 2010, 39, 1388−
1405.
(14) 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.
(15) Shaikh, K. I.; Madsen, C. S.; Nielsen, L. J.; Jørgensen, A. S.;
Nielsen, H.; Petersen, M.; Nielsen, P. Chem.Eur. J. 2010, 16,
12904−12919.
(16) Jørgensen, A. S.; Shaikh, K. I.; Enderlin, G.; Ivarsen, E.; Kumar,
S.; Nielsen, P. Org. Biomol. Chem. 2011, 9, 1381−1388.
(17) Kumar, P.; Shaikh, K. I.; Jørgensen, A. S.; Kumar, S.; Nielsen, P.
J. Org. Chem. 2012, 77, 9562−9573.
(18) Andersen, N. K.; Chandak, N.; Brulikova, L.; Kumar, P.; Jensen,
M. D.; Jensen, F.; Sharma, P. K.; Nielsen, P. Bioorg. Med. Chem. 2010,
18, 4702−4710.
(19) Andersen, N. K.; Døssing, H.; Jensen, F.; Vester, B.; Nielsen, P.
J. Org. Chem. 2011, 76, 6177−6187.
(20) Kumar, P.; Chandak, N.; Nielsen, P.; Sharma, P. K. Bioorg. Med.
Chem. 2012, 20, 3843−3849.
(21) Heystek, L. E.; Zhou, H.-Q.; Dande, P.; Gold, B. J. Am. Chem.
Soc. 1998, 120, 12165−12166.
(22) 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.
(23) Hurley, D. J.; Tor, Y. J. Am. Chem. Soc. 1998, 120, 2194−2195.
(24) Gourdain, S.; Petermann, C.; Harakat, D.; Clivio, P. Nucleosides,
Nucleotides Nucleic Acids 2010, 29, 542−546.
(25) (a) Bigge, C. F.; Mertes, M. P. J. Org. Chem. 1981, 46, 1994−
1997. (b) Western, E. C.; Daft, J. R.; Johnson, E. M.; Gannett, P. M.;
Shaughnessy, K. H. J. Org. Chem. 2003, 68, 6767−6774. (c) Fresneau,
N.; Hiebel, M.-A.; Agrofoglio, L. A.; Berteina-Raboin, S. Molecules
2012, 17, 14409−14417.
(26) (a) Sartori, G.; Enderlin, G.; Herve,
44, 767−772. (b) Sartori, G.; Herve, G.; Enderlin, G.; Len, C. Synthesis
2013, 45, 330−333. (c) Gallagher-Duval, S.; Herve, G.; Sartori, G.;
́
G.; Len, C. Synthesis 2012,
́
́
Enderlin, G.; Len, C. New J. Chem. 2013, 37, 1989−1995.
(27) (a) Burley, G. A.; Gierlich, J.; Mofid, M. R.; Nir, H.; Tal, S.;
Eichen, Y.; Carell, T. J. Am. Chem. Soc. 2006, 128, 1398−1399.
(b) Gierlich, J.; Burley, G. A.; Gramlich, P. M. E.; Hammond, D. M.;
Carell, T. Org. Lett. 2006, 8, 3639−3642.
(28) Beyer, C.; Wagenknecht, H.-A. Chem. Commun. 2010, 46,
2230−2231.
(29) (a) Wagner, C.; Wagenknecht, H.-A. Chem.Eur. J. 2005, 11,
1871−1876. (b) Jacobsen, M. F.; Ferapontova, E. E.; Gothelf, K. V.
Org. Biomol. Chem. 2009, 7, 905−908. (c) Fukuda, M.; Nakamura, M.;
Takada, T.; Yamana, K. Tetrahedron Lett. 2010, 51, 1732−1735.
(d) Park, S. M.; Nam, S.-J.; Jeong, H. S.; Kim, W. J.; Kim, B. H.
Chem.Asian J. 2011, 6, 487−492.
2863
dx.doi.org/10.1021/jo4025896 | J. Org. Chem. 2014, 79, 2854−2863