Synthesis, nucleic acid hybridization properties and molecular modelling studies of conformationally restricted 3′-O,4′-C-methylene-linked α-l-ribonucleotides

Article abstract of DOI:10.1016/j.carres.2006.04.010

Nucleotides with conformationally restricted carbohydrate rings such as locked nucleic acid (LNA), α-L-LNA or 2′,5′-linked 3′-O,4′-C-methyleribonucleotides exhibit significant potential as building blocks for antigene and antisense strategies. 2′,5′-Linke

Full text of DOI:10.1016/j.carres.2006.04.010

Carbohydrate Research 341 (2006) 1398–1407
Synthesis, nucleic acid hybridization properties and
molecular modelling studies of conformationally restricted
30-O,40-C-methylene-linked a-L-ribonucleotides
Andreas S. Madsen, Patrick J. Hrdlicka, T. Santhosh Kumar and Jesper Wengel*
Nucleic Acid Center, Department of Chemistry, University of Southern Denmark, DK-5230 Odense M, Denmark
Received 14 March 2006; received in revised form 7 April 2006; accepted 8 April 2006
Available online 18 May 2006
Abstract—Nucleotides with conformationally restricted carbohydrate rings such as locked nucleic acid (LNA), a-L-LNA or 20,50-
linked 30-O,40-C-methyleribonucleotides exhibit significant potential as building blocks for antigene and antisense strategies. 20,50-
Linked a-L-ribo configured monomer X (termed a-L-ONA) was designed as a potential structural mimic of a-L-LNA. The corre-
sponding phosphoramidite building block of monomer X was obtained in five steps (10% overall yield) from the easily obtainable
thymine derivative 1. Incorporation of monomer X into oligodeoxyribonucleotides (ONs) results in dramatically decreased thermal
stabilities with DNA/RNA complements (DTm/mod = ꢀ11.5 to ꢀ17.0 ꢀC) compared to unmodified reference ONs. Less pro-
nounced decreases (DTm/mod = ꢀ4.5 to ꢀ8.5 ꢀC) are observed when monomer X is incorporated into triplex forming ONs and tar-
geted against double-stranded DNA (parallel orientation, pyrimidine motif). This biophysical data, together with modelling studies,
suggest that 20,50-linked a-L-ONA is a poor structural mimic of a-L-LNA.
ꢁ 2006 Elsevier Ltd. All rights reserved.
Keywords: LNA; 20,50-Linked nucleic acids; Conformationally restricted nucleosides; Oligonucleotides; Oxetane
1. Introduction
The flexible furanose ring of unmodified nucleosides
adopts many different conformations in solution,
although they tend to cluster in two regions that are clas-
sified as belonging to either the North (N) or South (S)
part of the pseudorotational cycle.8 The dioxabicyclo-
[2.2.1]heptane skeletons of LNA (locked nucleic acid,
b-D-ribo configuration, Fig. 1)9–13 and its diastereomer
a-L-LNA (a-L-ribo configuration, Fig. 1)14–16 efficiently
lock the furanose ring in an N-type conformation. (The
furanose ring of a-L-LNA is, according to definitions,
conformationally restricted in an N-type conformation
as a result of the L-configuration; however, a-L-LNA
does not overlay onto a typical N-type framework but
rather overlays an S-type framework.) A single incorpo-
ration of an LNA or a-L-LNA building block into ONs
induces, depending on sequence length and composition,
large increases in thermal stability towards complemen-
tary DNA or RNA of up to +10 ꢀC relative to unmodi-
fied ONs.9–16 NMR studies have suggested that the
increases in thermal stability are related to the
The antigene and antisense strategies, that is, targeting of
double-stranded DNA and single-stranded RNA with
exogenous probes, respectively, continue to be highly
desirable approaches to achieve specific control of gene
expression.1 During the past two decades, substantial ef-
forts have been directed towards developing modified
oligonucleotide probes with improved antigene or anti-
sense characteristics such as increased target affinity
and mismatch discrimination and enhanced stability to-
wards enzymatic degradation.2,3 A particularly success-
ful approach to realize this has been to modify
oligodeoxyribonucleotides (ONs) with nucleotides that
have conformationally restricted carbohydrate rings.4–7
A research center funded by the Danish National Research Foun-
dation for studies on nucleic acid chemical biology.
Corresponding author. Tel.: +45 6550 2510; fax: +45 6615 8780;
e-mail: jwe@chem.sdu.dk
*
0008-6215/$ - see front matter ꢁ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2006.04.010

A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
1399
Base
Base
O
Base
T
O
O
O
O
O
O
O
O
O
P
O
O
P
O
O
O
P
O
O
O
O
P
O
O
O
O
O
LNA
α-L-LNA
ONA
"3',4'-BNA"
α-L-ONA
Monomer X
Figure 1. Structures of LNA,9–13 a-L-LNA14–16 and ONA (previously termed 30,40-BNA)38–40 along with the novel a-L-ONA (30-O,40-C-methylene-
linked a-L-ribonucleotide, monomer X) reported herein. T = Thymin-1-yl.
36,37
0
0
˚
predisposition of LNA to behave as an A-type (RNA)
mimic with regard to helical structure17,18 whereas a-L-
LNA behaves as an B-type (DNA) mimic.19,20 The po-
tential of LNA and a-L-LNA as triplex forming
oligonucleotides (TFOs) is obvious, as a single incorpo-
ration of LNA21–24 or a-L-LNA25 nucleotides into
triplex forming ONs generally induces very prominent
increases in thermal stability of up to +10.0 and
+6.0 ꢀC, respectively.
tance above 7 A) similar to that of S-type 3 ,5 -NAs.
The preference for complexation with RNA comple-
ments and prominent stabilization of triplexes with
20,50-NAs has therefore, and in the light of the proper-
ties of N-type nucleosides such as LNA, been suggested
to arise from a tendency of 20,50-NAs to adopt compact
S-type furanose conformations.35
The synthesis of bicyclic 20,50-linked nucleosides in
which the furanose ring is conformationally biased to-
wards a compact S-type conformation, has recently been
reported.38–41 Incorporation of 30-O,40-C-methyleribo-
nucleotides (Fig. 1, herein termed ONA for ‘oxetane
nucleic acid’) into ONs results in significantly decreased
thermal affinities towards DNA complements while the
affinity towards RNA complements is virtually un-
changed compared to unmodified ONs.38 The potential
of ONA as TFOs has been evaluated in a single study
but, arguably due to suboptimal sequence design, ONA
was found to destabilize triplexes,40 and a more thorough
evaluation of ONA as TFOs is therefore awaited.
The inverse relationship of nucleotide geometry and
phosphodiester linkage between 20,50- and 30,50-NAs
(Fig. 2) suggests that the 20,50-linked ONA mimics the
structure of LNA and that a 20,50-linked bicyclic nucle-
oside with a-L-ribo configuration such as monomer X
(Fig. 1) would behave as a structural mimic of a-L-
LNA, which is known to adopt an extended back-
bone.19,20 The incorporation of a 20,50-linked nucleoside
with inverted stereochemistry at the C20-, C30- and C40-
positions (relative to normal RNA) into ONs has not
been reported and we therefore set out to synthesize
monomer X and to evaluate this missing member of
the bicyclic nucleoside family.
20,50-Linked nucleic acids (20,50-NAs)26–28 have been
suggested as potential antisense and antigene probes
due to the following properties: (a) 20,50-NAs form du-
plexes with normal RNA (i.e., 30,50-linked RNA) albeit
they bind with lower affinity than isosequential DNA,
(b) 20,50-NAs display a remarkable preference for com-
plexation with RNA complements, (c) homopyrimidine
chimeras of 20,50-linked DNA and normal DNA exhibit
significantly increased thermal affinity towards double-
stranded DNA compared to an unmodified ON and
(d) 20,50-NAs exhibit high resistance to nuclease diges-
tion.29–35
20,50- and 30,50-NAs have been suggested to exhibit an
interesting inverse relationship between nucleotide
geometry and phosphodiester linkage (Fig. 2), that is,
S-type 20,50-NAs adopt a compact backbone (intra-
strand distance between two phosphorous atoms is
0
0
˚
below 6 A) similar to that of N-type 3 ,5 -NAs, whereas
N-type 20,50-NAs adopt an extended backbone (P–P dis-
β-D-ribo
N-type
S-type
3′,5′-linked
2′,5′-linked
Compact Extended
Extended Compact
2. Results and discussion
Figure 2. Nucleotide conformation of b-D-ribo configured nucleotides
having N- and S-type sugar puckering, emphasizing the inverse
relationship between intrastrand P–P distance (compact vs extended)
and the phosphodiester linkage (30,50 vs 20,50).
Phosphoramidite 6 was identified as a suitable building
block for incorporation of monomer X using automated
DNA synthesis. Retrosynthetic analysis revealed bicyclic

1400
A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
nucleoside 4 as a key intermediate, which was expected to
be obtainable from known thymine derivative 142 after
appropriate blocking of the C20-hydroxy group with
a base stable protecting group to avoid competing
a-L-LNA nucleoside formation during ring closure
(Scheme 1). Tetrahydropyranylation of alcohol 1 was
therefore attempted using 3,4-dihydro-2H-pyran
(DHP) in anhydrous dioxane and pyridinium p-toluene-
sulfonate (PPTS) as catalyst. However, the desired
O20-THP protected nucleoside was only obtained in an
unsatisfactory yield of 20% (data not shown), which
did not improve by changing the solvent and/or cata-
lyst.43,44 Protection of the C20-hydroxy group proved
to be notoriously difficult as introduction of the well-
established MOM or TBDMS groups or the recently
described dimethylthiocarbamate group45 also were
unsuccessful, most likely as a consequence of the steric
hindrance exerted on the C20-position by the 1,3-cis
relationship of the thymine and C30-benzyloxy moieties.
These difficulties prompted us to investigate an alter-
native route to phosphoramidite 6, which is based on
selective ring closure (Scheme 1). Debenzylation of alco-
hol 1 by catalytic hydrogenation using Pd(OH)2/C as
catalyst afforded diol 2 in satisfying 92% yield. Ring
closure to give bicyclic nucleosides has typically been
carried out via intramolecular nucleophilic substitution
under alkaline conditions.9,10,46,47 Because both hydr-
oxyl groups of nucleoside 2 may act as nucleophiles,
two products must be expected, that is, the desired a-
L-ONA nucleoside and the corresponding a-L-LNA
nucleoside. Previous examples of ring closure in related
b-D-ribo configured compounds with two potentially
nucleophilic hydroxy groups include a bicyclic deriva-
tive of 30-C-ethynyluridine47 and 30-O,40-C-methyleneri-
bonucleotides,46 which afforded the desired oxetane
derivatives in good yields.
Among the various conditions attempted, including
LiHMDS in THF,46 NaH in THF9,10 or saturated meth-
anolic ammonia,47 the most satisfactory was treatment
of nucleoside 2 with dilute aqueous sodium hydroxide.
The method afforded an inseparable mixture (ꢁ3:2 by
1H NMR) of two compounds, tentatively assigned as
(1S,3R,4S,5R)-4-hydroxy-1-methanesulfonyloxymethyl-
3-(thymin-1-yl)-2,6-dioxabicyclo[3.2.0]heptane and (1S,
3R,4S,7R)-7-hydroxy-1-methanesulfonyloxymethyl-3-
(thymin-1-yl)-2,5-dioxabicyclo[2.2.1]heptane, that is, the
a-L-ONA and a-L-LNA intermediates, respectively.
The mixture was subsequently subjected to sodium ben-
zoate in DMF from which a-L-ONA nucleoside 3 was
isolated in 25% yield (over two steps from 2). Subse-
quent O50-deacylation of bicyclic nucleoside 3 with
saturated methanolic ammonia furnished diol 4 in 84%
yield. The primary C50-hydroxy group of diol 4 was pro-
tected as a 4,40-dimethoxytrityl (DMTr) ether using
O
O
O
NH
NH
NH
N
O
N
O
N
O
MsO
MsO
OR
OH
O
OH
O
OH
ii
iii
O
O
O
BzO
HO
142 R = Bn
R = H
3
4
i
2
O
O
NH
NH
N
O
N
O
O
OR
O
O
O
iv
vi
O
O
P
DMTrO
O
O
5
R = H
R = P(O(CH2)2CN)N(i-Pr)2
Monomer X
v
6
Scheme 1. Reagents, conditions and yields: (i) H2/(20% Pd(OH)2/C), EtOAc, rt (92%); (ii) (a) aq NaOH, 0–4 ꢀC, (b) NaOBz, DMF, 60–75 ꢀC (25%
over two steps); (iii) satd NH3/CH3OH, rt (84%); (iv) DMTrCl, pyridine, rt (85%); (v) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2, rt (58%); (vi)
DNA synthesizer.

A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
1401
standard conditions to furnish nucleoside 5 in 85% yield.
Taking the difficulties associated with O20-protection of
nucleoside 1 into account, nucleoside 5 underwent stan-
dard phosphitylation to give phosphoramidite 6 in 58%
yield surprisingly easily.
the oxetane ring protons H500 appear as two doublets
(AB-system) at 4.12 and 4.69 ppm. Furthermore NOE
difference spectroscopy identified key contacts between
H10/H20 (14%) and H20/H30 (11%), indicating a cis rela-
tionship between these protons (Fig. 3; NOE-enhance-
ments are given as an average of the two values
observed upon individual irradiation of the protons in
question). In addition, the observed NOE contact
between H6=H50A0 (7%) (H50A0 is tentatively assigned as
the H500 proton closest to H6), indicates a cis relationship
between the nucleobase and the oxymethylene bridge.
2.1. Structural verification of bicyclic nucleoside 4
The following NMR observations ascertained the pro-
posed 2,6-dioxabicyclo[3.2.0]heptane skeleton of nucleo-
side 4 and ruled out a 2,5-dioxabicyclo[2.2.1]heptane
skeleton corresponding to an a-L-LNA nucleoside: (a)
1
in the H NMR spectrum H20 couples to H10, 20-OH
2.2. Synthesis of ONs
and H30 with coupling constants J >5 Hz, while these
coupling constants are very small for a-L-LNA nucleo-
sides (J <2 Hz),16 (b) the signal from 20-OH is observed
as an exchangeable doublet at ꢁ5.8 ppm, (c) no signal
corresponding to a 30-OH group is observed and (d)
Incorporation of phosphoramidite 6 into oligodeoxyri-
bonucleotides (ONs) was performed on a 0.2 lmol scale
using an automated DNA synthesizer. Standard proce-
dures were used resulting in stepwise coupling yields
>98% of phosphoramidite 6 (20 min coupling time)
and >99% of unmodified deoxyribonucleotide phosphor-
amidites (2 min coupling time). The composition and
purity of the resulting a-L-ONAs ON3–ON5, ON7–
ON8, ON10–ON11 and ON13–ON14 (Tables 1 and 2)
were verified by MALDI-MS (Table S1) and ion-ex-
change HPLC, respectively.
O
7%
27%
NH
H
H
H
N
H
O
O
H
OH
H
O
2.3. Duplex thermal denaturation studies
HO
The effect on duplex stability upon incorporation of
monomer X into ONs was evaluated by UV-thermal du-
plex denaturation studies using medium or high salt buf-
fer conditions ([Na+] = 110 and 710 mM, respectively,
pH 7.0, Table 1). Unless otherwise noted, the UV-ther-
mal denaturation curves displayed smooth sigmoidal
14%
11%
Figure 3. Key NOE contacts in bicyclic nucleoside 4. (NOE-enhance-
ments are an average of the two values observed upon individual
irradiation of the protons in question.)
Table 1. Thermal denaturation studies of duplexes at different salt concentrationsa
Tm (DTm/mod)/ꢀC
Medium salt
High salt
DNA
RNA
DNA
RNA
ON1
ON2
ON3
ON4
ON5
50-GTG ATA TGC
50-GCA TAT CAC
50-GTG AXA TGC
50-GCA XAX CAC
50-GXG AXA XGC
30.0b
30.0b
27.5
27.0
14.5 (ꢀ13.0)
<10
<10
36.0b
36.0b
34.0
34.5
14.0 (ꢀ16.0)
21.5c (ꢀ14.5)
22.0d (ꢀ12.0)
<10
<10
<10
12.0 (ꢀ11.5)
<10
<10
ON6
ON7
ON8
50-GCG TTT TTT GCT
50-GCG TTT TTX GCT
50-GCG TTT TXX GCT
46.5
42.5
nd
nd
nd
nd
nd
nd
29.5 (ꢀ17.0)
31.0 (ꢀ11.5)
16.5 (ꢀ15.0)
22.5 (ꢀ10.0)
a Thermal denaturation temperatures (Tm-values) of ON1–ON8 towards DNA/RNA complements [Tm values/ꢀC (DTm/mod = change in Tm value
per incorporation of monomer X relative to DNA:DNA or DNA:RNA reference duplex)] measured as the maximum of the first derivative of the
melting curve (A260 vs temperature) recorded in medium salt buffer (100 mM NaCl, 0.1 mM EDTA, adjusted to pH 7.0 by 10 mM NaH2PO4/5 mM
Na2HPO4) or high salt buffer (as for medium salt buffer except 700 mM NaCl) using 1.0 lM concentrations of the two complementary strands and a
temperature ramp of 1.0 ꢀC/min. Tm values are averages of at least two measurements; A = adenine-9-yl DNA monomer, C = cytosine-1-yl DNA
monomer, G = guanine-9-yl DNA monomer, T = thymin-1-yl DNA monomer, see Scheme 1 for the structure of monomer X. nd = not determined.
b ON1 and ON2 are complementary.
c Tm values (ꢀC) towards DNA strands containing a single mismatch in the central position C/G/T: <10/21.0/<10.
d Tm values (ꢀC) towards RNA strands containing a single mismatch in the central position C/G/U: <10/24.5/<10.

1402
A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
Table 2. Thermal denaturation studies of triplexes at different pH valuesa
Tm/ꢀC
pH 6.0
pH 6.8
Tm1
Tm2
Tm1
Tm2
ON9
ON10
ON11
50-TTT TCT TTT CCC CCC T
50-TTT XCX TTT CCC CCC T
50-TXT XCX TXT CCC CCC T
25.0/22.5b
16.5/15.5b
<10/<10
69.0
69.0
69.0
11.5/<10c
<10/<10
<10/<10
69.0
69.0
69.0
ON12
ON13
ON14
50-CTC TTC TTT TCT TTC
50-CTC TTC XTT XCT TTC
50-CTC XTC XTT XCT XTC
43.0/37.5b,d
26.0/17.5b,d
<10/<10d
73.0
73.0
73.0
26.0/<10c
<10/<10
<10/<10
72.5
72.5
72.5
a Thermal denaturation temperatures (Tm-values) of ON9–ON11 towards DNA duplex 50-CCA CTT TTT AAA AGA AAA GGG GGG ACT GG:30-
GGT GAA AAA TTT TCT TTT CCC CCC TGA CC and of ON12–ON14 towards DNA duplex 50-GCG CGA GAA GAA AAG AAA GCC
GG:30-CGC GCT CTT CTT TTC TTT CGG CC measured as the maximum of the first derivative of the melting curve (A260 vs temperature)
recorded at pH 6.0 [10 mM sodium cacodylate, 150 mM NaCl and 10 mM MgCl2] and pH 6.8 [10 mM sodium cacodylate, 200 mM NaCl and
2 mM MgCl2] using 1.0 lM concentrations of all three strands and a temperature ramp of 0.5 ꢀC/min. Tm values are averages of at least two
measurements.
b Hysteresis was observed. Left Tm-value observed during heating, right Tm-value observed during cooling.
c No hyperchromicity was observed during cooling.
d Lowering the temperature ramp to 0.2 ꢀC/min resulted in the following Tm1 values (ꢀC) at pH 6.0: ON12: 41.5/37.0; ON13: 24.5/20.0; ON14: <10/
<10. Tm2 values were unaffected.
monophasic transitions similar to the unmodified refer-
ence duplexes (see Fig. S1 for a representative example).
Incorporation of a single monomer X into the mixed se-
quence 9-mer ONs ON3–ON5 or thymidine-rich 12-mer
ONs ON7–ON8 previously used to evaluate LNA, a-L-
LNA or ONA monomers,10,15,38 resulted in dramatically
decreased thermal affinities towards DNA complements,
compared to unmodified reference ONs (DTm/mod =
ꢀ17.0 to ꢀ14.5 ꢀC). Less pronounced destabilization
was observed for duplexes with complementary RNA
(DTm/mod = ꢀ13.0 to ꢀ11.5 ꢀC). Multiple incorpora-
tions of monomer X into ONs resulted in roughly addi-
tive destabilizations of duplexes with DNA/RNA
complements.
Thermal denaturation studies were carried out for
ON3 towards DNA and RNA strands with a centrally
placed mismatched nucleoside, to investigate if the
thymine moiety of monomer X base pairs with its
complementary nucleotide. Excellent discrimination
was observed for T:C and T:T/U mismatches while very
poor discrimination of T:G mismatches was observed
(Table 1, footnotes c and d).
rated in 15-mer and 16-mer polypyrimidine sequences
(ON10–ON11 and ON13–ON14, Table 2) previously
used to assess the potential of LNA23, a-L-LNA25 and
ONA40 as TFOs. Binding to 29-mer and 26-mer dou-
ble-stranded DNA (for ON10–ON11 and ON13–
ON14, respectively) with internal target sequences in
parallel orientation (pyrimidine motif) via Hoogsteen
base pairing was evaluated using two different sets of
conditions, that is, at pH 6.0 (10 mM sodium cacodyl-
ate, 150 mM NaCl and 10 mM MgCl2) and at pH 6.8
(10 mM sodium cacodylate, 200 mM NaCl and 2 mM
MgCl2). All ternary complexes exhibit a smooth sigmoi-
dal transition at high temperature corresponding to dis-
sociation of the duplex (Tm2), while no triplex to duplex
transition (Tm1) above 10 ꢀC could be observed for
ON10–ON11, ON13 or ON14 at pH 6.8.
At pH 6.0, the thermal denaturation curves with refer-
ence ON9 and ON12 as well as of the doubly modified
ON10 and ON13 displayed biphasic transitions. Incor-
poration of two X monomers into TFOs resulted in
significantly decreased thermal affinities compared to
unmodified reference TFOs (DTm/mod = ꢀ4.5 to
ꢀ8.5 ꢀC). Incorporation of four monomers X into TFOs
accentuated this trend even further as highly destabilized
triplexes with Tm1 values below 10 ꢀC were observed. It
is noteworthy that all ternary complexes displaying tri-
plex transitions, including the reference strands (also
at pH 6.8), exhibited significant thermal hysteresis (see
e.g., ON12, Table 2), which indicates very slow associa-
tion kinetics. Even when the temperature ramp was de-
creased from 0.5 to 0.2 ꢀC/min, hysteresis was observed
albeit less pronounced (Table 2, footnote d).
Comparison of a-L-ONA with LNA, a-L-LNA or
ONA (Fig. 1),10,15,38 clearly demonstrates that a-L-
ONA (Fig. 1) has the lowest affinity towards DNA/
RNA complements in this series of bicyclic nucleosides.
In fact, the mismatch data for a-L-ONA suggest that the
thymine moiety of monomer X is not base pairing with
its counterpart per se but rather recognizes its ‘binding’
partner merely by size discrimination.
2.4. Triplex thermal denaturation studies
Compared to the properties of LNA and a-L-LNA,
the present results indicate that a-L-ONA has a very lim-
ited potential as a modification for antisense or antigene
strategies.
Due to the potential of LNA,21–24 a-L-LNA25 and 20,50-
NAs35 as TFOs, it was decided to evaluate monomer X
in this context as well. Thus, X monomers were incorpo-

A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
1403
2.5. Molecular modelling
A series of force field calculations were performed to
rationalize the observed trends in thermal affinity of
ONs with incorporations of monomer X, towards
DNA/RNA complements, and furthermore, to place
these results in perspective to the reported properties
of LNA, a-L-LNA and ONA (Fig. 1) and the suggested
inverse relationship between nucleotide geometry (N- vs
S-type) and the nature of phosphodiester linkage (20,50-
vs 30,50-NAs).
Standard B-type DNA:DNA duplexes (50-GTG ATA
TGC:30-CAC TAT ACG) with a single central insertion
(italics) of an LNA, a-L-LNA, ONA or X monomer,
were subjected to molecular dynamics simulation and
minimization using the AMBER force field as imple-
mented in MacroModel V9.1.48 Subsequently, the lowest
energy structures of the modified and unmodified du-
plexes were compared. In the lowest energy structures,
LNA and a-L-LNA monomers adopt conformations
identical to published NMR solution structures,19,49
thereby validating the applied modelling protocol. The
inverted configurations of the C20, C30 and C40-atoms
of a-L-LNA and a-L-ONA, relative to normal DNA,
render comparisons based on furanose configurations
irrelevant. Instead, a comparison of the obtained struc-
tures was made by superimposing the following atoms
onto the corresponding unmodified DNA monomer of
ON1:ON2 (Fig. 4): (a) the two nucleobase atoms of the
modified nucleotide participating in hydrogen bonding
(N3 and0O4), (b) the C30-atom of the preceding nucleo-
tide (C3iꢀ1, i designating the modified monomer) and
(c) the C40-atom of the subsequent nucleotide ðC40iþ1Þ.
The central segment of the lowest energy structure of
the a-L-ONA modified DNA duplex is significantly per-
turbed relative to the reference duplex (Fig. 4). Thus, a
Figure 4. Section of the lowest energy structure (from C30iꢀ1 to C4i0þ1
)
obtained from molecular modelling of LNA, a-L-LNA, ONA and a-L-
ONA (Monomer X) (black line), including overlay of the corresponding
DNA structure (The furanose of the unmodified central thymidine in
ON1:ON2 is in an E-type conformation (P = 97ꢀ, O40-endo) and not in
the expected predominant S-type conformation) (grey line).
rules of compact versus extended nucleotide conforma-
tions may not be applicable when considering a-L-ribo
configured nucleotides. 20,50-Linked a-L-ribo configured
nucleotides seem to adopt an extended conformation
˚
(P–P distance ꢁ 7.0 A) when having a formal N-type
furanose pucker (overlaying an S-type furanose pucker
in b-D-ribo configured nucleotides). In contrast, they adopt
an even further extended conformation (P–P distance
˚
very extended intrastrand P–P distance (7.8 A) and a
parallel displacement of the nucleobase in radial direc-
tion away from the helical axis is observed.
˚
>7.5 A) when having a formal S-type furanose pucker.
Molecular modelling suggests that the oxetane moiety
of monomer X drives the nucleotide towards a more
extended conformation (P = 265ꢀ and P–P distance
To investigate if the a-L-ribo configuration per se leads
to the poor accommodation in the helix and to the highly
extended nucleotide conformation of a-L-ONA, a com-
parison was made with the a-L-LNA monomer. How-
ever, as previously observed,15 O50, O30 and N1, as
well as the C50-atom and the attached phosphodiester
group of a-L-LNA superimpose seamlessly onto the ref-
erence DNA strand, despite the inversion of configura-
tion at the C40-position of a-L-LNA (Fig. 4).
Incorporation of a-L-ONA or a-L-LNA in a b-D-
DNA duplex, alters the torsion angle c (O50–C50–C40–
C30) from +sc (ꢁ60ꢀ), as observed in b-D-ribo configured
nucleotides,8 to ap (ꢁ180ꢀ). This has a profound effect on
the intrastrand P–P distance, as can be observed with a-L-
ONA. In an attempt to further rationalize the effect of
changing from b-D-ribo to a-L-ribo configuration, scru-
tiny of molecular models indicates that the established
˚
ꢁ 7.8 A, thus overlaying an E-type furanose pucker in
b-D-ribo configured nucleotides) rather than a more com-
pact conformation as might be anticipated from the estab-
lished inverse relationship between nucleotide geometry
and phosphodiester linkage (Fig. 2). The very extended
conformation adopted by a-L-ONA (Fig. 4) necessitates
major rearrangements of the backbone, leading to a
strained backbone with unfavourable torsion angles,
which may explain the observed dramatic decreases in
thermal stability of duplexes modified with monomer X.
Similarly, the oxetane moiety of ONA disfavours a
perfect S-type furanose configuration, whereby the con-
formation is driven away from a conformation mimick-
ing LNA and towards an extended nucleotide
˚
conformation (P = 137ꢀ and P–P ꢁ 7.4 A).

1404
A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
To conclude, a-L-ONA and ONA are not 20,50-linked
of C4 substituents is numbered C500 in nucleoside deriva-
tives. Similar conventions apply for the corresponding
hydrogen atoms. Systematic compound names for bicy-
clic nucleosides are given according to the von Baeyer
nomenclature.50 MALDI-HRMS were recorded in posi-
tive ion mode on a IonSpec FT mass spectrometer using
2,5-dihydroxy benzoic acid as a matrix.
structural mimics of a-L-LNA and LNA, respectively.
In addition, intrastrand P–P distances are longer than
might have been expected, and these divergences may
suggest why ONA and particularly a-L-ONA do not ex-
hibit the same thermal stabilization of modified
DNA:DNA duplexes as observed with LNA and a-L-
LNA, respectively.
3.2. 1-[50-O-Methanesulfonyl-40-C-methanesulfonyl-
oxymethyl-a-L-erythro-pentofuranosyl]thymine (2)
3. Experimental
To a solution of alcohol 142 (6.40 g, 11.98 mmol) in
EtOAc (120 mL) was added 20% Pd(OH)2/C (2.01 g).
The resulting mixture was evacuated and a H2 atmo-
sphere was applied, whereafter the mixture was stirred
under a H2 atmosphere for 7 h at rt. The mixture was
then filtered through a Celite pad, which was thoroughly
washed with EtOAc. The combined organic phase was
evaporated to afford diol 2 (4.92 g, 92%) as a white
foam. Rf = 0.2 (15% i-PrOH in CH2Cl2, v/v); 1H
NMR (DMSO-d6): d 11.33 (br s, 1H, ex, NH), 7.57 (d,
J = 1.1 Hz, 1H, H6), 6.09 (d, J = 4.0 Hz, 1H, H10),
5.80 (d, J = 4.8 Hz, 1H, ex, 20-OH), 5.72 (d, J =
5.9 Hz, 1H, ex, 30-OH), 4.72 (d, J = 11.0 Hz, 1H,
H500), 4.42 (dd, J = 5.9, 4.7 Hz, 1H, H30), 4.29–4.35
(m, 3H, H500/H50), 4.21 (m, 1H, H20), 3.26 (s, 3H,
CH3SO2), 3.24 (s, 3H, CH3SO2), 1.76 (d, J = 1.1 Hz,
3H, CH3); 13C NMR (DMSO-d6): d 163.7, 150.4,
138.0 (C6), 107.4, 84.5 (C10), 81.5, 71.8 (C30), 70.2
(C20), 69.1 (C500), 68.6 (C50), 36.9 (CH3SO2), 36.6
3.1. General methods
All reagents and solvents were of analytical grade and
used without further purification as obtained from com-
mercial suppliers, except for CH2Cl2, which was distilled
before use. Petroleum ether of the distillation range 60–
80 ꢀC was used. Anhydrous DMF and pyridine were
used directly as obtained from suppliers. Anhydrous
acetonitrile was dried through storage over activated
˚
3 A molecular sieves. 1,2-Dichloroethane, CH2Cl2,
N,N-diisopropylethylamine, dioxane and THF for use
in anhydrous reactions were dried through storage over
˚
activated 4 A molecular sieves. Reactions were con-
ducted under an atmosphere of argon whenever anhyd-
rous solvents were used. All reactions were monitored
by thin-layer chromatography (TLC) using silica gel
coated plates (analytical SiO2-60, F-254). The TLC
plates were visualized under UV light and by dipping
in either (a) a 5% concd sulfuric acid in absolute ethanol
(v/v) or (b) a solution of molybdato-phosphoric acid and
cerium(IV) sulfate (5 g/L) in 3% concd sulfuric acid in
water (v/v) followed by heating with a heat gun. Column
chromatography using moderate pressure (pressure ball)
was performed with Silica Gel 60 (particle size 0.040–
0.063 mm, Merck). Evaporation of solvents was carried
out under reduced pressure at a temperature below
50 ꢀC. After column chromatography, appropriate frac-
tions were pooled and dried at high vacuum for at least
12 h to give obtained products in high purity (>95%) un-
less otherwise stated. 13C NMR or 31P NMR ascertained
sample purity. 1H NMR, 13C NMR and 31P NMR were
recorded at 300, 75.5 and 121.5 MHz, respectively.
Chemical shifts are reported in parts per million relative
to deuterated solvent as internal standard (dH: DMSO-d6
2.50 ppm; dC: DMSO-d6 39.43 ppm) or external stan-
dard (dP: 85% H3PO4 0.00 ppm). Designation in spectra
is as follows: singlet (s), doublet (d), multiplet (m) and
broad (br). Exchangeable (ex) protons were detected by
disappearance of peaks upon addition of D2O. Assign-
ments of NMR spectra are based on correlation spec-
troscopy 2D spectra (H–H correlated (COSY) and H–
C correlated (HETCOR)) and follow the standard nucle-
oside nomenclature. In general, quaternary carbons in
13C NMR spectra were not assigned. The carbon atom
1
(CH3SO2), 12.1 (CH3). Assignment of H NMR signals
of H50 and H500, and of the corresponding 13C NMR sig-
nals, may be interchanged. MALDI-HRMS m/z calcd
for [C13H20N2O11S2]Na+: 467.0384. Found 467.0401.
3.3. (1S,3R,4S,5R)-1-Benzoyloxymethyl-4-hydroxy-3-
(thymin-1-yl)-2,6-dioxabicyclo[3.2.0]heptane (3)
A solution of diol 2 (4.54 g, 10.22 mmol) in water
(200 mL) was cooled to 0 ꢀC in an ice-bath and aqueous
NaOH (2.0 M, 30 mL) was added dropwise. The result-
ing mixture was stirred at 4 ꢀC for 22 h during which the
otherwise clear solution turned brown. The reaction
mixture was neutralized by addition of saturated aque-
ous NH4Cl, extracted with EtOAc (8 · 150 mL), and
the combined organic phase was evaporated to dryness.
The resulting residue was adsorbed on Kieselguhr and
purified by silica gel column chromatography (0–15%
i-PrOH in CHCl3, v/v) affording a mixture of two com-
pounds tentatively assigned as (1S,3R,4S,5R)-4-hydro-
xy-1-methanesulfonyloxymethyl-3-(thymin-1-yl)-2,6-dio-
xabicyclo[3.2.0]heptane and (1S,3R,4S,7R)-7-hydroxy-
1-methanesulfonyloxymethyl-3-(thymin-1-yl)-2,5-dioxa-
bicyclo[2.2.1]heptane (2.12 g, ꢁ3:2 ratio by 1H NMR) as
a white foam.

A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
1405
The obtained mixture was dissolved in anhydrous
DMF (110 mL) and NaOBz (1.68 g, 11.6 mmol) was
added. The resulting mixture was stirred at 60 ꢀC for
18 h, whereafter a second portion of NaOBz (400 mg,
2.78 mmol) was added and the temperature raised to
75 ꢀC. After stirring for an additional 19 h, the reaction
mixture was cooled to rt and insoluble residues were fil-
tered off and washed extensively with EtOAc. The com-
bined filtrate was evaporated to near dryness and the
residue was partitioned between EtOAc (75 mL) and
H2O (50 mL). The phases were separated, and after
extraction of the aqueous phase with EtOAc
(4 · 50 mL), the combined organic phase was evaporated
to dryness. The resulting residue was adsorbed on Kies-
elguhr and purified by silica gel column chromatography
(0–10% i-PrOH in CH2Cl2, v/v) to afford alcohol 3
(0.97 g, 25% over two steps) as a white foam. Rf = 0.4
J = 7.0 Hz, 1H, H10), 5.73 (d, J = 5.1 Hz, 1H, ex, 20-
OH), 5.05 (t, J = 5.9 Hz, 1H, ex, 50-OH), 5.02 (d,
J = 5.1 Hz, 1H, H30), 4.69 (d, J = 7.7 Hz, 1H, H50B0 ),
4.41–4.49 (m, 1H, H20), 4.12 (d, J = 7.7 Hz, 1H, H50A0 ),
3.42–3.58 (m, 2H, H50), 1.77 (d, J = 0.9 Hz, 3H, CH3);
13C NMR (DMSO-d6): d 163.3, 151.4, 140.0 (C6),
107.6, 86.6, 86.4 (C10), 85.5 (C30), 75.5 (C500), 71.3
(C20), 60.9 (C50), 12.3 (CH3); MALDI-HRMS m/z calcd
for [C11H14N2O6]Na+: 293.0744. Found 293.0758.
3.5. (1S,3R,4S,5R)-1-(4,40-Dimethoxytrityloxymethyl)-4-
hydroxy-3-(thymin-1-yl)-2,6-dioxabicyclo[3.2.0]heptane (5)
Nucleoside diol 4 (0.38 g, 1.40 mmol) was co-evaporated
with anhydrous pyridine (2 · 10 mL), dissolved in anhy-
drous pyridine (15 mL) and 4,40-dimethoxytrityl chlo-
ride (0.55 g, 1.61 mmol) was added. The resulting
mixture was stirred at rt for 100 min, whereafter a sec-
ond portion of 4,40-dimethoxytrityl chloride (160 mg,
0.47 mmol) was added. After stirring at rt for additional
70 min, CH3OH (2.0 mL) was added and the resulting
mixture was evaporated to near dryness. The residue
was co-evaporated with toluene/absolute EtOH
(10 mL, 1:1 v/v), and EtOAc (30 mL) was added. The
organic phase was washed successively with satd aq
NaHCO3 (2 · 20 mL) and brine (15 mL), and the com-
bined aqueous phases were extracted with EtOAc
(15 mL). The combined organic phase was evaporated
to near dryness, and the residue co-evaporated with tol-
uene/absolute EtOH (10 mL, 50:50, v/v) and absolute
EtOH (2 · 10 mL). The resulting residue was purified
by silica gel column chromatography (0–3% CH3OH
in CH2Cl2, v/v containing 1 v% pyridine) and co-evapo-
rated with toluene/absolute EtOH (10 mL, 50:50, v/v)
and absolute EtOH (2 · 10 mL) to afford alcohol 5
(0.68 g, 85%) as a white foam. Rf = 0.3 (5% CH3OH in
1
(10% i-PrOH in CH2Cl2, v/v); H NMR (DMSO-d6): d
11.36 (br s, 1H, ex, NH), 8.09 (d, J = 0.9 Hz, 1H, H6),
7.94–8.00 (m, 2H, Ph), 7.65–7.72 (m, 1H, Ph), 7.51–
7.59 (m, 2H, Ph), 6.48 (d, J = 7.0 Hz, 1H, H10), 5.87
(br s, 1H, ex, 20-OH), 5.23 (d, J = 5.1 Hz, 1H, H30),
4.85 (d, J = 7.9 Hz, 1H, H50B0 ), 4.57–4.65 (m, 1H, H20),
4.45–4.55 (m, 2H, H50), 4.27 (d, J = 7.9 Hz, 1H, H50A0 ),
1.78 (d, J = 0.9 Hz, 3H, CH3); 13C NMR (DMSO-d6):
d 165.3, 163.6, 151.4, 139.9 (C6), 133.5 (Ph), 129.2
(Ph), 129.0 (Ph), 128.8 (Ph), 107.9, 86.6 (C10), 85.6
(C30), 83.4, 75.8 (C500), 71.2 (C20), 64.2 (C50), 12.3
(CH3); MALDI-HRMS m/z calcd for [C18H18N2O7]-
Na+: 397.1006. Found 397.0988.
Physical data for the mixture tentatively assigned as
(1S,3R,4S,5R)-4-hydroxy-1-methanesulfonyloxymethyl-
3-(thymin-1-yl)-2,6-dioxabicyclo[3.2.0]heptane
(a-L-
ONA derivative) and (1S,3R,4S,7R)-7-hydroxy-1-meth-
anesulfonyloxymethyl-3-(thymin-1-yl)-2,5-dioxabicyclo-
[2.2.1]heptane (a-L-LNA derivative): Selected signals
1H0 NMR (DMSO-d6):
d
6.44 (d, J = 7.0 Hz,
CH2Cl2, v/v); H NMR (DMSO-d6): d 11.37 (br s, 1H,
1
H1a-L-ONAÞ, 6.17 (d, J = 4.0 Hz, ex, 30-OHa-L-LNAÞ, 5.95
(s, H10a-L-LNAÞ, 5.85 (d, J = 5.1 Hz, ex, 20-OHa-L-ONAÞ,
5.12 (d, J = 5.1 Hz, H30a-L-ONAÞ, 4.60–4.38 (m, incl.
H2a0 -L-ONA, H30a-L-LNAÞ, 4.27 (s, H2a0 -L-LNAÞ; MALDI-
HRMS m/z calcd for [C12H16N2O8S]Na+: 371.0520.
Found 371.0514.
ex, NH), 8.03 (s, 1H, H6), 7.19–7.41 (m, 10H, Ar),
6.86–6.95 (m, 4H, Ar), 6.53 (d, J = 7.0 Hz, 1H, H10),
5.80 (d, J = 5.5 Hz, 1H, ex, 20-OH), 5.01 (d,
J = 5.5 Hz, 1H, H30), 4.53–4.62 (m, 2H, H20, H50B0 ),
4.16, (d, J = 7.3 Hz, 1H, H50A0 ), 3.74 (s, 6H, CH3O),
3.22 (s, 2H, H50), 1.78 (s, 3H, CH3); 13C NMR
(DMSO-d6): d 163.6, 158.1, 151.3, 144.5, 139.8 (C6),
135.2, 135.1, 129.6 (Ar), 127.8 (Ar), 127.5 (Ar), 126.7
(Ar), 113.2 (Ar), 107.8, 86.7 (C10), 85.8, 85.5 (C30),
84.6, 75.9 (C500), 71.5 (C20), 63.3 (C50), 54.9 (CH3O),
12.3 (CH3); MALDI-HRMS m/z calcd for
[C13H20N2O11S2]Na+: 595.2051. Found 595.2059.
3.4. (1R,3R,4S,5R)-4-Hydroxy-1-hydroxymethyl-3-
(thymin-1-yl)-2,6-dioxabicyclo[3.2.0]heptane (4)
Alcohol 3 (0.92 g, 2.47 mmol) was dissolved in satd
NH3/CH3OH (20 mL) and the resulting mixture was
stirred at rt for 28 h, whereupon the solvent was evapo-
rated. Adsorption of the residue on Kieselguhr and puri-
fication by silica gel column chromatography (0–10%
CH3OH in CH2Cl2, v/v) afforded diol 4 (0.56 g, 84%)
as a white solid material. Rf = 0.3 (15% CH3OH in
3.6. (1S,3R,4S,5R)-4-[2-Cyanoethoxy(diisopropylamino)-
phosphinoxy]-1-(4,40-dimethoxytrityloxymethyl)-3-
(thymin-1-yl)-2,6-dioxabicyclo[3.2.0]heptane (6)
1
CH2Cl2, v/v); H NMR (DMSO-d6): d 11.31 (br s, 1H,
Alcohol 5 (247 mg, 0.43 mmol) was co-evaporated
with anhydrous CH2Cl2 (2 · 2 mL), and dissolved in
ex, NH), 8.07 (d, J = 0.9 Hz, 1H, H6), 6.40 (d,

1406
A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
anhydrous CH2Cl2 (2.0 mL). Anhydrous diisopropyl-
ethylamine (0.5 mL) and 2-cyanoethyl N,N-diisopropyl-
phosphoramidochloridite (150 lL, 0.67 mmol) were
added. After stirring at rt for 3 h, additional 2-cyano-
ethyl N,N-diisopropylphosphoramidochloridite (150
lL, 0.67 mmol) was added. After stirring for a further
28 h, absolute EtOH (2 mL) was added and the resulting
mixture was stirred for 25 min. CH2Cl2 (20 mL) was
added and the organic phase was washed successively
with satd aq NaHCO3 (10 mL) and brine (10 mL). The
combined aqueous phase was extracted with CH2Cl2
(10 mL), and the combined organic phase evaporated
to dryness. The resulting residue was purified by silica
gel column chromatography (0–40% acetone in petro-
leum ether, v/v containing 1 v% pyridine) affording 6
(195 mg, 58%) as a white solid material. Rf = 0.4 (50%
acetone in petroleum ether, v/v); 31P NMR (CDCl3) d
153.6, 151.5.
pH 6.8 adjusted by addition of dilute aq HCl) or cacodyl-
ate buffer at pH 6.0 (10 mM sodium cacodylate, 150 mM
NaCl, 10 mM MgCl2 and pH 6.0 adjusted by addition of
dilute aq HCl). For studies of duplexes as well as tri-
plexes, ONs (1.0 lM each strand) were thoroughly
mixed, denatured by heating to 90 ꢀC and subsequently
cooled to the starting temperature of the experiment.
The absorbance was measured at 0.5 ꢀC intervals using
1 mL quartz optical cells with a path length of 1.0 cm.
The temperature of the denaturation experiments ranged
from at least 15 ꢀC below Tm to 15 ꢀC above Tm
(although not below 5 ꢀC or above 85 ꢀC). A temperature
ramp of 1.0 ꢀC/min was used in all experiments studying
duplex transitions, whereas a temperature ramp of 0.5 or
0.2 ꢀC/min was for triplex studies. Reported thermal
denaturation temperatures are an average of at least
two measurements within 1.0 ꢀC.
3.9. Molecular modelling
3.7. Synthesis of modified oligonucleotides
Duplex ON1:ON2 was built with a standard B-type heli-
cal geometry using the Amber software suite and subse-
quently modified to the duplexes of interest within the
MacroModel V9.1 Suite of programs.48 The charge of
the phosphodiester backbone was neutralized with
Synthesis of ONs containing monomer X was performed
in 0.2 lmol scale using an Applied Biosystems auto-
mated DNA synthesizer. Standard procedures were used
resulting in stepwise coupling yields >98% of phospho-
ramidite 6 (20 min coupling time) and >99% of unmodi-
fied deoxyribonucleotide phosphoramidites (2 min
coupling time) using 1 H-tetrazole as catalyst. Removal
of nucleobase protecting groups and cleavage from solid
support was effected using 32% aq ammonia for 12 h at
55 ꢀC. Detritylation was effected using 80% aq acetic acid
for 20 min at rt. Crude ONs were subsequently precipi-
tated from absolute EtOH (ꢀ18 ꢀC, 12 h) and washed
with 96% EtOH. Composition and purity (>80%) of
the synthesized ONs were verified by MALDI-MS anal-
ysis and analytical ion-exchange HPLC, respectively.
˚
sodium ions, which were placed 3.0 A from the nega-
tively charged oxygen atoms in the plane described by
the phosphorus and the non-bridging oxygen atoms.
During the following molecular dynamics simulation
and minimization, the sodium–oxygen distances were
2
˚
˚
restrained to 3.0 A by a force constant of 418 kJ/molA .
Employing these partial constrictions, the duplex struc-
tures were minimized using the Polak–Ribiere conjugate
gradient method, the all-atom AMBER force field51,52
and GB/SA solvation model53 as implemented in
MacroModel V9.1. Non-bonded interactions were trea-
˚
ted with extended cut-offs (van der Waals 8.0 A and
electrostatics 20.0 A). The minimized structures were
˚
3.8. Thermal denaturation studies
then (using the same constraints as described above)
submitted to 5 ns of stochastic dynamics (simulation
temperature 300 K, time step 2.2 fs, SHAKE all bonds
to hydrogen), during which 500 structures were sampled
and subsequently minimized.
Concentrations of ONs were calculated using the follow-
ing extinction coefficients (OD260/lmol): dG, 12.0; dA,
15.2; T, 8.4; dC, 7.1; G, 13.7; A, 15.4; U, 10.0; C, 9.0.
Thermal denaturation temperatures (Tm values/ꢀC) were
measured on a Perkin–Elmer Lambda 35 UV/VIS spec-
trometer equipped with a PTP-6 Peltier temperature pro-
grammer and determined as the maximum of the first
derivative of the thermal denaturation curve (A260 vs
temperature). Thermal denaturation studies of duplexes
were recorded in either medium salt phosphate buffer
(100 mM NaCl, 0.1 mM EDTA and pH 7.0 adjusted
with 10 mM NaH2PO4/5 mM Na2HPO4) or high salt
phosphate buffer (700 mM NaCl, 0.1 mM EDTA and
pH 7.0 adjusted with 10 mM NaH2PO4/5 mM Na2H-
PO4). Thermal denaturation studies of triplexes were per-
formed using cacodylate buffer at either pH 6.8 (10 mM
sodium cacodylate, 200 mM NaCl, 2 mM MgCl2, and
Acknowledgements
We greatly appreciate funding from The Danish
National Research Foundation, Villum Kann Rasmus-
sen Fonden and The Oticon Foundation.
Supplementary data
Copies of 13C NMR spectra of nucleotides 1–5, 31P
NMR spectrum of phosphoramidite 6, MALDIMS data

A. S. Madsen et al. / Carbohydrate Research 341 (2006) 1398–1407
1407
24. Sun, B.-W.; Babu, B. R.; Sørensen, M. D.; Zakrzewska,
K.; Wengel, J.; Sun, J.-S. Biochemistry 2004, 43, 4160–
4169.
25. Kumar, N.; Nielsen, K. E.; Maiti, S.; Petersen, M. J. Am.
Chem. Soc. 2006, 128, 14–15.
of synthesized ONs (Table S1), representative denatur-
ation curves of ON3 with complementary DNA/RNA
(Fig. S1) and of ON13 as TFO illustrating the observed
thermal hysteresis (Fig. S2). Supplementary data associ-
ated with this article can be found, in the online version,
at doi:10.1016/j.carres.2006.04.010.
26. Kierzek, R.; He, L.; Turner, D. H. Nucleic Acids Res.
1992, 20, 1685–1690.
27. Dougherty, J. P.; Rizzo, C. J.; Breslow, R. J. Am. Chem.
Soc. 1992, 114, 6254–6255.
28. Hashimoto, H.; Switzer, C. J. J. Am. Chem. Soc. 1992,
114, 6255–6256.
References
29. Giannaris, P. A.; Damha, M. J. Nucleic Acids Res. 1993,
21, 4742–4749.
30. Alul, R.; Hoke, G. Antisense Res. Dev. 1995, 5, 3–11.
31. Sheppard, T. L.; Breslow, R. C. J. Am. Chem. Soc. 1996,
118, 9810–9811.
1. Opalinska, J. B.; Gewirtz, A. M. Nat. Rev. Drug Disc.
2002, 1, 503–514.
2. Buchini, S.; Leumann, C. J. Curr. Opin. Chem. Biol. 2003,
7, 717–726.
3. Kurreck, J. Eur. J. Biochem. 2003, 270, 1628–1644.
4. Kværnø, L.; Wengel, J. Chem. Commun. 2001, 1419–
1424.
32. Kandimalla, E. R.; Manning, A.; Zhao, Q.; Shaw, D. R.;
Byrn, R. A.; Sasisekharan, V.; Agrawal, S. Nucleic Acids
Res. 1997, 25, 370–378.
5. Leumann, C. J. Bioorg. Med. Chem. 2002, 10, 841–
854.
6. Meldgaard, M.; Wengel, J. J. Chem. Soc., Perkin Trans. 1
2000, 3539–3554.
7. For a recent example, see: Pradeepkumar, P. I.; Cheruku,
P.; Plashkevych, O.; Acharya, P.; Gohil, S.; Chat-
topadhyaya, J. J. Am. Chem. Soc. 2004, 126, 11484–
11499.
8. Saenger, W. Principles of Nucleic Acid Structure; Springer:
Berlin, 1984.
33. Wasner, M.; Arion, D.; Borkow, G.; Noronha, A.; Uddin,
A. H.; Parniak, M. A.; Damha, M. J. Biochemistry 1998,
37, 7478–7486.
34. Damha, M. J.; Noronha, A. Nucleic Acids Res. 1998, 26,
5152–5156.
35. Obika, S.; Hiroto, A.; Nakagawa, O.; Imanishi, T. Chem.
Commun. 2005, 2793–2795.
36. Lalitha, V.; Yathindra, N. Curr. Sci. 1995, 68, 68–75.
37. Premraj, B. J.; Yathindra, N. J. Biomol. Struc. Dyn. 1998,
16, 313–328.
9. Singh, S. K.; Nielsen, P.; Koshkin, A. A.; Wengel, J.
Chem. Commun. 1998, 455–456.
38. Obika, S.; Morio, K.; Hari, Y.; Imanishi, T. Bioorg. Med.
Chem. Lett. 1999, 9, 515–518.
10. Koshkin, A. A.; Singh, S. K.; Nielsen, P.; Rajwanshi, V.
K.; Kumar, R.; Meldgaard, M.; Olsen, C. E.; Wengel, J.
Tetrahedron 1998, 54, 3607–3630.
11. Obika, S.; Nanbu, D.; Hari, Y.; Andoh, J.; Morio, K.;
Doi, T.; Imanishi, T. Tetrahedron Lett. 1998, 39, 5401–
5404.
39. Obika, S.; Morio, K.; Nanbu, D.; Hari, Y.; Itoh, H.;
Imanishi, T. Tetrahedron 2002, 58, 3039–3049.
40. Savy, P.; Benhida, R.; Fourrey, J.-L.; Maurisse, R.; Sun,
J.-S. Bioorg. Med. Chem. Lett. 2000, 10, 2287–2289.
41. Hrdlicka, P. J.; Kumar, T. S.; Wengel, J. Eur. J. Org.
Chem. 2005, 5184–5188.
12. Wengel, J. Acc. Chem. Res. 1999, 32, 301–310.
13. Petersen, M.; Wengel, J. Trends Biotechnol. 2003, 21, 74–
81.
42. Babu, B. R.; Raunak; Poopeiko, N. E.; Juhl, M.; Bond, A.
D.; Parmar, V. S.; Wengel, J. Eur. J. Org. Chem. 2005,
2297–2321.
43. Greene, T. W.; Wuts, P. G. M. Protective Groups in
Organic Synthesis, 3rd ed.; Wiley: New York, 1999, pp 49–
54, and references cited therein.
44. Kocienski, P. J. Protecting Groups, 3rd ed.; Georg Thieme:
Stuttgart, 2005, pp 315–320, and references cited therein.
45. Barma, D. K.; Bandyopadhyay, A.; Capdevila, J. H.;
Falck, J. R. Org. Lett. 2003, 5, 4755–4757.
46. Obika, S.; Morio, K.; Hari, Y.; Imanishi, T. Chem.
Commun. 1999, 2423–2424.
47. Hrdlicka, P. J.; Andersen, N. K.; Jepsen, J. S.; Hansen, F.
G.; Haselmann, K. F.; Nielsen, C.; Wengel, J. Bioorg.
Med. Chem. 2005, 13, 2597–2621.
48. MacroModel, version 9.1, Schro¨dinger, LLC, New York,
NY, 2005.
49. Nielsen, C. B.; Singh, S. K.; Wengel, J.; Jacobsen, J. P. J.
Biomol. Struct. Dyn. 1999, 17, 175–191.
50. Moss, G. P. Pure Appl. Chem. 1999, 71, 513–529.
51. 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.
52. Weiner, S. J.; Kollman, P. A.; Nguyen, D. T.; Case, D. A.
J. Comput. Chem. 1986, 7, 230–252.
˚
14. Rajwanshi, V. K.; Hakansson, A. E.; Dahl, B. M.;
Wengel, J. Chem. Commun. 1999, 1395–1396.
˚
15. Rajwanshi, V. K.; Hakansson, A. E.; Sørensen, M. D.;
Pitsch, S.; Singh, S. K.; Kumar, R.; Nielsen, P.; Wengel, J.
Angew. Chem., Int. Ed. 2000, 39, 1656–1659.
˚
16. Sørensen, M. D.; Kværnø, L.; Bryld, T.; Hakansson, A.
E.; Verbeure, B.; Gaubert, G.; Herdewijn, P.; Wengel, J. J.
Am. Chem. Soc. 2002, 124, 2164–2176.
17. Nielsen, K. E.; Singh, S. K.; Wengel, J.; Jacobsen, J. P.
Bioconjug. Chem. 2000, 11, 228–238.
18. Petersen, M.; Bondensgaard, K.; Wengel, J.; Jacobsen, J.
P. J. Am. Chem. Soc. 2002, 124, 5974–5982.
˚
19. Nielsen, K. M. E.; Petersen, M.; Hakansson, A. E.;
Wengel, J.; Jacobsen, J. P. Chem. Eur. J. 2002, 8, 3001–
3009.
20. Nielsen, J. T.; Stein, P.; Petersen, M. Nucleic Acids Res.
2003, 31, 5858–5867.
21. Obika, S.; Hari, Y.; Sugimoto, T.; Sekiguchi, M.; Iman-
ishi, T. Tetrahedron Lett. 2000, 41, 8923–8927.
22. Obika, S.; Uneda, T.; Sugimoto, T.; Nanbu, D.; Minami,
T.; Doi, T.; Imanishi, T. Bioorg. Med. Chem. 2001, 9,
1001–1011.
23. Torigoe, H.; Hari, Y.; Sekiguchi, M.; Obika, S.; Imanishi,
T. J. Biol. Chem. 2001, 276, 2354–2360.
53. Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson,
T. J. Am. Chem. Soc. 1990, 112, 6127–6129.