LNA guanine and 2,6-diaminopurine. Synthesis, characterization and hybridization properties of LNA 2,6-diaminopurine containing oligonucleotides

Article abstract of DOI:10.1016/j.bmc.2004.02.008

LNA guanine and 2,6-diaminopurine (D) phosphoramidites have been synthesized as building blocks for antisense oligonucleotides (ON). The effects of incorporating LNA D into ON were investigated. As expected, LNA D containing ON showed increased affinity t

Full text of DOI:10.1016/j.bmc.2004.02.008

Bioorganic & Medicinal Chemistry 12 (2004) 2385–2396
LNA guanine and 2,6-diaminopurine. Synthesis, characterization
and hybridization properties of LNA 2,6-diaminopurine
containing oligonucleotides
Christoph Rosenbohm,a Daniel Sejer Pedersen,a,ꢀ Miriam Frieden,a Flemming R. Jensen,a
Susan Arent,b Sine Larsenb and Troels Kocha,*
aSantaris Pharma A/S, Bøge Allꢀe 3, DK-2970 Hørsholm, Denmark
bCentre for Crystallographic Studies, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen, Denmark
Received 21 August 2003; accepted 3 February 2004
Abstract—LNA guanine and 2,6-diaminopurine (D) phosphoramidites have been synthesized as building blocks for antisense
oligonucleotides (ON). The effects of incorporating LNA D into ON were investigated. As expected, LNA D containing ON showed
increased affinity towards complementary DNA (DTm +1.6 to +3.0 ꢀC) and RNA (DTm +2.6 to +4.6 ꢀC) ON. To evaluate if LNA D
containing ON have an enhanced mismatch sensitivity compared to their complementary LNA A containing ON thermal dena-
turation experiments towards singly mismatched DNA and RNA ON were undertaken. Replacing one LNA A residue with LNA
D, in fully LNA modified ON, resulted in higher mismatch sensitivity towards DNA ON (DDTm )4 to >)17 ꢀC). The same trend was
observed towards singly mismatched RNA ON (DDTm D–a ¼ )8.7 ꢀC and D–g ¼ )4.5 ꢀC) however, the effect was less clearcut and
LNA A showed a better mismatch sensitivity than LNA D towards cytosine (DTm +5.5 ꢀC).
ꢁ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
ciple all other known nucleic acid analogues based on
the phosphoramidite oligomerization approach, for
example, phosphorothioates and 20-O-alkyl modified
RNA.5 These unique properties positions LNA as a
versatile probe in nucleic acid chemistry.
The promising aspects of using analogues of natural
nucleic acids in antisense oligonucleotides (ON) has
stimulated much interest in this field during recent years.
Locked nucleic acid (LNA)ꢁ was introduced in 19981–3
as a novel class of conformationally restricted ON
analogues. It is well established that LNA is the single
nucleic acid modification that contributes the highest
increase in affinity (Tm) ever obtained in Watson–Crick
hydrogen bonding.4 The 20,40-linked bicyclic structure
provide high affinity and high specificity to both fully
and partially LNA modified ON. LNA monomers can
be mixed with DNA and RNA monomers and in prin-
In order to design and synthesize LNA therapeutics it is
essential to have efficient and robust syntheses of the
necessary LNA phosphoramidites. The LNA thymine,
5-methylcytosine and adenine monomers are now rou-
tinely synthesized on a 100þg scale, however the
essential LNA guanine monomer has until recently
caused problems due to several troublesome synthetic
6;7
€
steps. In particular the Vorbruggen coupling of the
guanine nucleobase (or derivatives thereof) to the car-
bohydrate moiety has been problematic, inevitably
leading to mixtures of the N7 and N9 regioisomers.8
Many different strategies have been devised to enhance
the selectivity in favour of the N9 isomer.9–12 We found
a method employing 2-amino-6-chloropurine in the
Keywords: Nucleosides; Antisense agents; Bioorganic chemistry; Oligo-
nucleotides.
* Corresponding author. Tel.: +45-4517-9800; fax: +45-4517-9898;
e-mail: tk@santaris.com
€
Vorbruggen coupling particularly appealing for several
ꢀ
Present address: Cambridge University, University Chemical Labo-
ratory, Lensfield Road, Cambridge CB2 1EW, UK.
LNA is defined as an ON containing one or more 20-O,40-C-
reasons.9;13–15 Besides potentially being able to furnish a
regiospecific coupling a number of other advantages
could be envisaged: (1) guanine and guanine nucleosides
are very polar compounds, complicating synthetic work
ꢁ
methylene-b-D-ribofuranosyl nucleotide monomers (LNA mono-
mers).1
0968-0896/$ - see front matter ꢁ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.02.008

2386
C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
and the use of the less polar 2-amino-6-chloropurine was
likely to be an advantage; (2) 2-amino-6-chloropurine is
cheap and commercially available in large quantities; (3)
2-amino-6-chloropurine can be transformed into a
number of different nucleobase analogues such as
6-arylpurines,16 2-aminopurine,17;18 6-thiopurine19;20 and
2,6-diaminopurine21;22 essentially making it a convergent
synthesis of a whole range of LNA purine monomers.
The potential of 6-substituted purines as nucleoside
drugs23 and the ability to modulate the properties of ON
by introducing modified nucleobases24 are areas of
considerable interest and has recently also turned our
attention towards modified nucleobases in a LNA con-
text.
B
MsO
MsO
MsO
MsO
O
O
OAc
i
OAc
ii
BnO OAc
BnO
1
B
MsO
O
O
BnO
€
Scheme 1. (i) Vorbruggen coupling; (ii) ring closing to form the
bicyclic LNA skeleton. B ¼ nucleobase or nucleobase analogue.
isobutyrylguanine has been reported previously.8
Although conditions favouring the formation of the LNA
guanine N9 isomer are employed, significant amounts
(approximately 10–15%) of the N7 isomer were always
detected in our hands and the N9 and N7 isomers were
very difficult to separate. Moreover, 2-N-isobutyryl-
guanine is tedious to synthesize and purify.12
Here we present the first regio- and stereospecific syn-
thesis of the LNA G phosphoramidite employing
2-amino-6-chloropurine, unambiguously showing that
€
the N9 isomer is the only product of the Vorbruggen
coupling. In addition, the synthesis of the LNA 2,6-
diaminopurine (LNA D) phosphoramidite and ON
containing this novel LNA monomer has been
accomplished. To evaluate LNA D containing ON
potential as probes for antisense ON, mismatch discrim-
ination experiments of LNA D containing ON towards
singly mismatched DNA and RNA sequences have been
performed, and compared to LNA A containing ON.
Moreover, the hybridization properties of LNA D
containing ON, towards complementary RNA and
DNA sequences, have been studied in an effort to gain a
better understanding of what effects govern 2,6-diam-
inopurines ability to stabilize certain duplexes.
We decided to use 2-amino-6-chloropurine in the
€
Vorbruggen coupling with 1 to see if we could improve
on the selectivity (Scheme 2). A detailed analysis of the
obtained nucleoside 2 by HPLC–MS, 1H and 13C NMR
(vide infra) revealed the exclusive formation of one
isomer. The formation of the b-anomer was expected
due to the C2 acetate group in the coupling sugar (1),
allowing only attack from the b-face. X-ray analysis
confirmed that the desired N9 regioisomer was the
exclusive product (Fig. 1). As previously demonstrated
by other researchers9 the regiospecific formation of the
N9 isomer can be achieved when 2-amino-6-chloro-
purine is coupled to a carbohydrate moiety under ther-
modynamic conditions (e.g., TMS-triflate, refluxing
dichloroethane).11
2. Results and discussion
The convergent synthesis of LNA monomers relies on
the coupling of different nucleobases to the same cou-
pling sugar 1 (Scheme 1) followed by ring closure to give
the bicyclic nucleosides.8 The coupling of 1 with 2-N-
Nucleoside 2 is a versatile compound allowing the syn-
thesis of a wide range of LNA nucleobase analogues.
Either the ring closure to form the bicyclic LNA skele-
ton with concomitant formation of the guanine nucleo-
O
Cl
N
N
N
NH
N
N
MsO
MsO
RO
N
NH2
N
NH2
O
O
i
ii
1
R1O
iii
O
BnO OAc
3 R= Ms, R1=Bn
4 R=Bz, R1=Bn
5 R=R1=OH
2
iv
O
O
N
N
N
NH
NH
v
vii
N
N
RO
DMTO
N
N
N
O
O
N
N
R1O
O
P
O
O
6 R=R1=OH
CN
(i-Pr)2N
O
vi
7 R=DMT, R1=OH
8
Scheme 2. Reagents and conditions: (i) 2-Amino-6-chloropurine, N,O-bis(trimethylsilyl)acetamide, TMS-triflate, 1,2-dichloroethane; (ii)
HOCH2CH2CN, NaH, THF; (iii) NaOBz, DMSO, 100 ꢀC; (iv) (a) MsOH, CH2Cl2, (b) Amberlyst A-26 (OHꢀ (aq), EtOH); (v) (CH3O)2CHN(CH3)2,
(vi) DMT-Cl, pyridine (vii) NC(CH2)2OP(N(iPr)2)2, 4,5-dicyanoimidazole, MeCN, CH2Cl2. DMT ¼ 4,40-dimethoxytrityl.

C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
2387
efficient removal of the benzyl group. However, the
published work-up procedure with triethylamine
resulted in inseparable mixture of nucleoside and tri-
ethylammonium salts. The use of a basic anion exchange
resin (Amberlyst A-26) as a scavenger to remove the
acid solved this problem, but the basic conditions
resulted in the premature hydrolysis of the 50-benzoyl
group, giving nucleoside 5. Due to the very polar nature
of nucleoside 5 it precipitated under the conditions
employed and required large volumes of solvent to wash
it off the resin. This problem was however solved by
washing the resin with ethanol prior to use. Now the 50-
benzoyl nucleoside was effectively removed from resin
devoid of all methanesulfonic acid. The EtOH solution
containing the benzoylated nucleoside was then treated
with aqueous NaOH resulting in rapid hydrolysis of the
50-benzoyl group producing nucleoside 5. The structure
of the fully deprotected nucleoside 5 was confirmed by
X-ray crystallographic analysis (Fig. 1).
We have previously experienced problems in the syn-
thesis of ON where the guanine residues have been
protected with the commonly employed isobutyryl
protection group, requiring longer deprotection times,
and as a result being incompatible with the synthesis of
ON containing base sensitive groups. Therefore, we
decided to introduce the dimethylformamidine protec-
tion group, which is cleaved very rapidly using standard
ON deprotection protocols. Hence, the 2-amino group
on the crude nucleoside 5 was protected as the dimeth-
ylformamidine giving nucleoside 6, followed by protec-
tion on the 50-hydroxy group with 4,40-dimethoxytrityl
chloride producing nucleoside 7 in 84% yield from 4.
Phosphitylation of the 30-hydroxy group gave the
phosphoramidite 8 in 88% yield ready for the automated
incorporation into ON.
Figure 1. X-ray structure of nucleoside 5.
base can be performed (Scheme 2) or alternatively
exclusive formation of the bicyclic structure without
affecting the nucleobase can be carried out allowing for
later modification of the purine to a wide range of nu-
cleobase analogues, for example, 2,6-diaminopurine
(Scheme 4).
It has been reported that 2-amino-6-chloropurines upon
treatment with aqueous NaOH undergo substitution of
the chlorine atom to give guanine.25;26 These conditions
are very similar to those employed for the ring closure of
the LNA bicyclic skeleton (aq LiOH or NaOH in
THF).8 As expected, upon treatment with aq NaOH
compound 2 rapidly formed the bicylic system. How-
ever, the substitution reaction on the purine was very
slow even under forcing conditions and was difficult to
drive to completion. To increase the reaction rate we
clearly needed a better nucleophile. 3-Hydroxypropio-
nitrile is a useful nucleophile for this particular reaction,
giving the desired guanine nucleobase after b-elimina-
tion of acrylonitrile.27;28 In most cases triethylamine or
DBU is used to promote the formation of the alkoxide
in situ. We decided to use slightly different conditions
from the previously published procedures27;28 pre-form-
ing 3 equiv of the alkoxide of 3-hydroxypropionitrile by
treatment with NaH, followed by the addition of
nucleoside 2. These conditions gave a very rapid ring
closure, usually complete in less than 15 min, followed
by a somewhat slower conversion of 2-amino-6-chloro-
purine to guanine yielding nucleoside 3 in >80% yield
after 4 h. Substitution of the mesylate on C50 of nucleo-
side 3 with sodium benzoate produced nucleoside 4
that was debenzylated with Pd(OH)2–C and ammonium
formate, conditions that also cleaved the 50-benzoyl
group producing the fully deprotected LNA guanine 5.
Unfortunately, the reaction was sluggish taking 2–3 days
and up to 0.5 equiv Pd-catalyst to go to completion. This
prompted us to investigate other conditions for the
deprotection of the 30-benzyl group. Using methane-
sulfonic acid in dichloromethane29 resulted in a very
2.1. NMR analysis of the N7 and N9 regioisomers
In order to have an easy method for determining the
regiochemistry of guanine nucleosides in the future we
did a comparative NMR study on the 2-N-isobutyryl
protected nucleosides 9 and 10 (Fig. 2). The route based
€
on the Vorbruggen coupling of 2-N-isobutyrylguanine,
as previously published by Koshkin et al.8 was used to
obtain a mixture of 9 (N9 isomer) and 10 (N7 isomer) in
O
7
5
N
O
NH
8
N
MsO
9
N
N
4
O
H
1'
O
BnO
9
H
N
4
5
9
7
N
O
N
N
8
NH
O
MsO
O
1'
O
BnO
10
Figure 2. The N9 isomer (9) and the N7 isomer (10).

2388
C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
Table 1. The most diagnostic NMR signals for compounds 9 and 10
147.9 ppm in the N9 isomer and the signal at 158.0 ppm
in the N7 isomer (Table 1).
9 (N9 isomer)
10 (N7 isomer)
H8
H10
7.74
5.83
8.12
6.24
Using this simple correlation between the chemical shifts
as a guideline 13C NMR spectrometry proves to be a
simple and fast method for determining which regio-
isomer is obtained, and in our experience the trend is
general for a wide variety of guanine nucleosides.
C2, C4 and C6 147.3, 147.9 and 155.2 148.4, 153.2 and 158.0
C5 121.5 110.8
1H NMR run at 400 MHz and 13C NMR run at 100 MHz in CDCl3.
All values are in ppm.
2.2. Synthesis of LNA 2,6-diaminopurine
a 9:1 ratio. A pure sample of the N9 isomer 3 (Scheme 2)
was prepared as described above and then acylated with
isobutyryl anhydride to give 9 (Fig. 2). Comparison of
the NMR spectra of the mixture with the pure sample
confirmed that the major product in the regioselective
coupling of 2-N-isobutyrylguanine was the N9 isomer
(9).
2,6-Diaminopurine nucleosides can be synthesized from
guanine nucleosides according to the procedure by
Gryaznov and Schultz.33 As anticipated this procedure
worked equally well for the conversion of LNA G to
LNA D (Scheme 3). Initially, this gave us the desired
LNA D phosphoramidite 16 for our studies. However,
because the starting material (LNA G) is the product of
a long linear reaction sequence, adding further steps in a
linear fashion to obtain LNA D was undesirable.
The most diagnostic NMR signals for the two regio-
isomers 9 and 10 are shown in Table 1. Comparison of 9
1
and 10 by H NMR spectrometry showed a significant
upfield shift for H10 and H8 in nucleoside 9 (Table 1).
However, using the H10 proton for identifying regio-
isomers is unreliable because the chemical shift is highly
dependent on the carbohydrate moiety, making the
diagnostic value of the signals limited, unless the car-
bohydrate moiety is unchanged.
During our work on the synthesis of the LNA G
nucleoside (vide supra) we discovered that treatment of
nucleoside 2 (Scheme 4) with aqueous NaOH resulted in
a fast ring closing reaction to give the bicyclic nucleoside
17, leaving the 2-amino-6-chloropurine largely unaf-
fected.
Brand et al.20 synthesized a wide range of alkylated
purines and studied the shift of the H8 proton between
the two regioisomers. They found in full agreement with
the observations of Kjellberg and Johansson30 that the
H8 signal of the N9 regioisomers is shifted upfield rel-
ative to the H8 signal of the N7 regioisomers. The same
is observed for 9 and 10 with a difference of approxi-
mately 0.4 ppm. However, the chemical shift values
again cannot be used as a general method for assigning
the regioisomers, because they are highly dependent on
the alkyl group/carbohydrate moiety attached to the
purine.
This in turn became a useful way of synthesizing the
LNA D nucleoside. Treatment of nucleoside 2 with aq
LiOH in THF produced nucleoside 17 in 90% yield.
Displacement of the mesylate on C50 of 17 with sodium
benzoate in DMF was as expected a fast reaction giving
nucleoside 18 in a 91% yield. Treatment of 18 with a
saturated solution of ammonia in MeOH in a sealed
glass tube at 90 ꢀC and 90 psi resulted in the formation
O
N
NH
N
DMTO
i
N
NHR
O
13C NMR proved a more useful tool for assigning the
two regioisomers. According to Chenon et al.31 and
Bailey and Harnden32 the chemical shift difference
between the C4 and C5 atoms is very pronounced for N7
and N9 regioisomers. They reported that shifting the
substituent from N9 to N7 resulted in a significant
downfield shift of the C4 resonance that approximately
equalled the upfield shift of the C5 resonance. In the
case of regioisomers 9 and 10 an upfield shift of
approximately 10 ppm for C5 in nucleoside 9 compared
to 10 was observed. Therefore, we anticipated a down-
field shift of the same magnitude for C4 when going
from nucleoside 9 to 10. In our case it is difficult to
distinguish between C2, C4 and C6 without using
advanced NMR techniques (e.g., INADEQUATE).
However, upon examination it is seen that two of the
signals are largely unaffected and the remaining signal is
observed shifting downfield approximately 10 ppm
(approximately 147–158 ppm, Table 1). On the basis of
the data in the literature31;32 it is reasonable to assume
that C4 corresponds to either the signal at 147.3 or
ii
O
HO
C(O)i Bu
11 R=
12 R= H
NHR
N
N
N
O
DMTO
N
NHR
R1O
O
13 R=R1=H
iii
iv
v
14 R=R1=Bz
15 R=Bz, R1=H
16 R=Bz, R1=PN2
Scheme 3. Reagents and conditions: (i) Satd methanolic ammonia; (ii)
(a) (TFA)2O, pyridine, 0 ꢀC, (b) NH3 (55% from 11); (iii) BzCl, pyr-
idine; (iv) 2 M aq NaOH, THF, MeOH, H2O (6:4:1, v/v/v) (81% from
13); (v) NC(CH2)2OP(N(iPr)2)2, 4,5-dicyanoimidazole, MeCN, DMF
(95%). DMT ¼ 4,40-dimethoxytrityl, PN2 ¼ 2-cyanoethoxy(diisopro-
pylamino)phosphinyl.

C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
2389
Cl
N
R2
N
N
N
N
N
N
R3
RO
ii
RO
i
iii
NH2
N
O
O
2
R1O
R1O
O
O
19 R=H, R1=Bn, R2=R3=NH2
17 R= Ms, R1=Bn
iv
20 R=Bz, R1=Bn, R2=R3= NHBz/NBz2
18 R=Bz, R1=Bn
NHBz
NHBz
N
N
N
N
N
N
RO
DMTO
N
NHBz
N
NHBz
v
O
O
O
vii
O
O
HO
O
P
CN
21 R=H
15 R=DMT
(i-Pr)2N
vi
16
Scheme 4. Reagents and conditions: (i) LiOH (aq), THF; (ii) BzONa, DMF; (iii) satd methanolic ammonia (80 ꢀC/60 psi); (iv) BzCl, pyridine; (v) (a)
MsOH, CH2Cl2, (b) Amberlyst A-26 (OHꢀ, aq); (vi) DMT-Cl, pyridine (vii) NC(CH2)2OP(N(iPr)2)2, 4,5-dicyanoimidazole, MeCN, DMF.
DMT ¼ 4,40-dimethoxytrityl.
of the 2,6-diaminopurine nucleoside 19 together with the
undesired 6-O-methyl nucleoside by substitution with
methoxide. To favour the formation of nucleoside 19
and suppress the formation of the 6-O-methyl substi-
tuted nucleoside we investigated the reaction at a range
of temperatures and pressures. Lowering the reaction
temperature to 60 ꢀC and the pressure to 36 psi resulted
in the exclusive formation of the desired nucleoside 19
but the reaction was too slow (30% conversion after
12 h) to be of practical use. Eventually, we found that
a temperature of 80 ꢀC and a pressure of 60 psi for 4 h
resulted in the formation of nucleoside 19 in an
acceptable yield with a minimum of the 6-O-methyl
nucleoside and some unreacted starting material 18.
Longer reaction times did not increase the conversion of
starting material 18 to diaminopurine 19, but resulted in
degradation of the product. Due to the very polar nature
of 19 it was used without further purification in the
following step where it was treated with benzoyl chloride
in pyridine to give a mixture the tetra- and penta-
benzoylated nucleosides 20 in a 1:1 ratio (65% yield
from 18). In order to confirm the identity of the material
a sample of the product was treated with diluted
ammonia in methanol for 5 min resulting in the removal
of one/two benzoyl group(s). The tribenzoylated prod-
uct was identified as the desired nucleoside 20 (Scheme
4, R2 ¼ R3 ¼ NHBz) by NMR spectrometry. Because we
expected the imide benzoyl groups to be hydrolyzed
during the work-up after the debenzylation step (vide
infra) it was deemed unnecessary to remove them at this
point and the mixture 20 was used without further
purification.
acid, as described above, gave nucleoside 21. Treating
crude 21 with untreated Amberlyst A-26 resin hydro-
lyzed both the 50-benzoyl group and the imide benzoyl
groups, as expected, without affecting the two amides on
the purine. Crude 21 was then protected with 4,40-di-
methoxytrityl chloride to give nucleoside 15 in 74% yield
from 20. In the final step the 30-hydroxy group was
phosphitylated to produce the phosphoramidite 16 in a
95% yield ready for the automated incorporation into
ON.
Benzoyl protection groups on the nucleobase were
employed, to make the synthesis of the monomers easier,§
even though it has been reported that they can be diffi-
cult to remove from the synthesized ON.34 Slower
deprotection for the LNA D containing ON was indeed
observed, but complete deprotection was always
achieved within 18 h at 65 ꢀC.
2.3. Oligonucleotides containing LNA D
It has been established that D–T basepairs compared to
A–T basepairs stabilize some duplexes and destabilize
others.35 In general, the D–T basepair contributes a
larger increase in duplex stability when incorporated
into RNA duplexes than observed for the complemen-
tary DNA duplexes. It has been suggested that the lower
increase in stability in DNA duplexes is due to the 2-
amino groups effect on structural features such as
hydration of the minor groove, helical conformations
During the work-up of the product from the debenzyl-
ation of the LNA G nucleoside (vide supra) we dis-
covered that it was necessary to wash the Amberlyst
resin (employed as an acid scavenger) to avoid hydrol-
ysis of the 50-benzoyl group. We hoped we could use this
to our advantage in the synthesis of nucleoside 21.
Debenzylation of nucleoside 20 with methanesulfonic
§
Attempts at preparing the dimethylformamidine and dibutylform-
amidine protected 2,6-diaminopurine phosphoramidites were unsuc-
cessful, due to the loss of the formamidine protection groups during
the course of reactions and purifications. We were not able to obtain
any pure formamidine protected diaminopurine nucleoside at any
time, due to the loss of the protection groups during chromatogra-
phy, and we were unable to crystallize any of the nucleosides.

2390
C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
and base stacking.36;37 As expected, this effect is clearly
manifested when 2,6-diaminopurine monomers are
incorporated into ON that promote formation of A-type
duplexes, such as, for example, phosphoramidates,38
hexitol nucleic acids (HNA)37 and threofurano-
syl(30 fi 20)oligonucleotides (TNA),39 showing a higher
affinity towards RNA complements than towards DNA
complements. The observed effect is attributed to the
fact that the carbohydrates are in the north conforma-
tion thus producing A-type duplexes, which may be
essential for formation of the third hydrogen bonding
interaction in the D–T base pair, as well as producing a
more favourable helical conformation, and thus
improved base stacking and hydration of the duplex.
Table 3. Duplex melting temperatures (Tm)
Sequence
50-d(act aXa cg)-30
X ¼ t
X ¼ a
<10
X ¼ c
<10
X ¼ g
9.8
1
2
3
4
5
6
50-cgt ata gt-30
50-cgt dta gt-30
50-cgt AtA gt-30
50-cgt DtA gt-30
23.0
26.8
42.2
43.8
5.6
9.0
5.9
6.1
17.0
22.0
27.3
<10
18.0
7.0
<10
27.0
<10
50-CGT ATA GT-30 58.8
50-CGT DTA GT-30 61.0
33.7
31.4
See Table 2 legend for details.
A and D containing ON versus mismatched DNA sequences.
Table 4. Duplex melting temperatures (Tm)
Sequence
50-r(act aXa cg)-30
It is well known that LNA nucleosides adopt an extreme
north conformation,40 and LNA would therefore be the
perfect nucleoside analogue to test this hypothesis. We
decided to compare the duplex stability of ON con-
taining LNA D towards complementary DNA and
RNA sequences to see if we would obtain higher thermal
stabilities towards RNA sequences compared to DNA
sequences (Table 2).
X ¼ t
X ¼ a
X ¼ c
X ¼ g
1
2
3
4
5
6
50-cgt ata gt-30
50-cgt dta gt-30
50-cgt AtA gt-30
50-cgt DtA gt-30
20.8
26.0
45.4
48.0
<10
<10
17.0
27.0
44.7
40.6
<10
<10
24.0
28.0
46.7
56.8
<10
<10
25.8
25.5
48.9
49.0
50-CGT ATA GT-30 66.8
50-CGT DTA GT-30 71.4
See Table 2 legend for details.
A and D containing ON versus mismatched RNA sequences.
The results in Table 2 clearly show a significant increase
in duplex stability when substituting adenine with
diaminopurine, as expressed by higher Tm values, both
towards DNA and RNA complements. In perfect
agreement with the literature (vide supra) all duplexes
showed the largest increases in thermal stability towards
RNA complements. However, to be able to conclusively
attribute this to an effect arising solely from the carbo-
hydrates extreme north conformation (resulting in an
A-type duplex) comparison of iso-sequential ON con-
taining a nucleoside analogue that adopts an extreme
south conformation (resulting in a B-type duplex) is
necessary. To this end we are currently synthesizing the
bility is expected when this monomer is incorporated,
and this would demonstrate the close relationship
between sugar conformation and the ability of the 2,6-
diaminopurine nucleobase to hybridize satisfactorily to
complementary DNA and RNA strands, giving us a
better understanding of what governs the ability of 2,6-
diaminopurine to stabilize helices.
One of the properties of the 2,6-diaminopurine nucleo-
base, compared to adenine, is a better mismatch dis-
crimination,35 which in particular has been applied
successfully in technologies working by the principle of
non-self complementarity.42 To investigate if this prin-
ciple could be applied using LNA D, thermal denatur-
ation experiments against all three singly mismatch
DNA sequences (Table 3) and singly mismatched RNA
sequences (Table 4) were undertaken.
a-L-LNA 2,6-diaminopurine monomer that Wengel and
co-workers have shown adopt an extreme south type
confirmation.41 A much lower increase in duplex sta-
Table 2. Duplex melting temperatures (Tm). D containing ON versus
complementary DNA and RNA sequences
The Tm determinations of DNA sequences (1 and 2,
Table 3) and DNA sequences containing LNA mono-
mers (3 and 4, Table 3) show no detectable effect upon
replacement of adenine for 2,6-diaminopurine.– How-
ever, for the fully LNA modified ON 5 and 6 a large
effect is seen, showing a much greater mismatch sensi-
tivity towards all singly mismatched sequences for ON 6,
with DTm (ON 5 vs 6) )4 to >)17 ꢀC.
Sequence
Tm=DTm (ꢀC)
50-d(act ata cg)-30 50-r(act ata cg)-30
1
2
3
50-cgt ata gt-30
50-cgt dta gt-30
50-cgt dtd gt-30
23.0
26.8/3.8
31.3/4.1
20.8
26.0/5.2
33.9/6.6
4
5
6
50-cgt AtA gt-30
50-cgt DtA gt-30
50-cgt DtD gt-30
42.2
43.8/1.6
48.2/3.0
45.4
48.0/2.6
54.5/4.5
50-CGT ATA GT-30 58.8
50-CGT DTA GT-30 61.0/2.2
50-CGT DTD GT-30 63.4/2.3
66.8
71.4/4.6
75.6/4.4
The Tm determinations of DNA sequences (1 and 2,
Table 4) show no detectable effect upon replacement of
7
8
9
Lower case bases are DNA or RNA; a ¼ adenine, c ¼ cytosine,
t ¼ thymine, g ¼ guanine, d ¼ 2,6-diaminopurine. Upper case bases are
LNA: A ¼ adenine, C ¼ 5-methyl cytosine, T ¼ thymine, G ¼ guanine,
D ¼ 2,6-diaminopurine. ON were synthesized by phosphoramidite
chemistry. Buffer for melting temperature (Tm) determination: 100 mM
phosphate buffer, 1 M NaCl, 4 lM for each ON strand, melting
starting at 5 ꢀC, under argon flow. DTm is the increase per a substituted
for d or A substituted for D.
–
The low Tm values for sequences 1 and 2 (X ¼ t, Tables 3 and 4)
makes it impossible to detect any difference between the mismatched
sequences (X ¼ a, c and g, Tables 3 and 4). Making the DNA
sequences longer resulted in a very high Tm of the iso-sequential LNA
ON making it impossible to determine them accurately. Hence, it was
necessary to make short sequences, to be able to determine Tm values
for the fully LNA modified ON.

C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
2391
adenine for 2,6-diaminopurine, because of the low Tm of
4. Experimental
the mismatched sequences.– ON 3 and 4 also do not
show any improved discrimination upon introduction
of D.
For reactions conducted under anhydrous conditions
glassware was dried overnight in an oven at 150 ꢀC and
was allowed to cool in a dessicator over anhydrous
KOH. Anhydrous reactions were carried out under an
atmosphere of argon using anhydrous solvents. Solvents
were HPLC grade, of which DMF, pyridine, acetonitrile
and dichloromethane were dried over molecular sieves
However, in the case of the fully LNA modified ON (5
and 6, Table 4) an improved discrimination is observed
for X ¼ a (DDTm ¼ )8.7 ꢀC) and X ¼ g ðDDTm
¼
)4.5 ꢀC). Somewhat unexpected, in the case of X ¼ c,
ON 5 has a better mismatch sensitivity than ON 6
(DDTm ¼ +5.5 ꢀC), showing that the result of intro-
ducing D in place of A does not necessarily mean an
improved mismatch discrimination. The overall effects
on the specificity of exchanging adenine for diaminop-
urine is, at least not in this RNA model, pronounced.
ꢁ
(4 A from Grace Davison) and THF was freshly distilled
from Na Æ benzophenone to a water content below
20 ppm. TLC was run on Merck silica 60 F254 aluminium
sheets. Dry column vacuum chromatography (DCVC)
was performed according to the published procedure.43
Synthesis of nucleoside 13 was carried out in a heavy-
wall borosilicate pressure glass tube from Ace glass fit-
ted with a #15 Ace thred PTFE plug and a pressure
gauge. 1H and 13C NMR spectra were recorded at,
respectively, 400 and 100 MHz with solvents as internal
standard (dH: CDCl3 7.26 ppm, DMSO-d6 2.50 ppm; dC:
CDCl3 77.0 ppm, DMSO-d6 39.4 ppm. D2O with 1% 1,4-
In general the substitution of LNA A with LNA D
results in enhanced affinity towards DNA and more
pronounced towards RNA. The apparent higher speci-
ficity of LNA D towards mismatches in DNA could
make the monomer a suitable tool in DNA targeting.
dioxane as internal standard dH: 3.75 ppm; dC: 67.2). 31
P
NMR was run at 121 MHz with 85% H3PO4 as external
standard. J values are given in Hz. Assignments of
NMR spectra are based on 2D spectra and follow the
standard carbohydrate/nucleoside nomenclature (the
carbon atom of the 40-C-substituent is numbered C100 )
even though the systematic compound names of the
bicyclic nucleoside derivatives are given according to the
von Baeyer nomenclature. Crude compounds were used
without further purification if they were P95% pure by
TLC and HPLC–MS. HPLC was performed with an
Agilent 1100 series HPLC on a Xterra column from
Waters (MS C18, 3.5 lm, 2.1 · 150 mm), with the fol-
lowing gradient 0–0.5 min, 50% A in B; 0.5–3.0 min, 50–
100% A in B; 3.0–7.0 min, 100% B, 7.0–8.0 min, 100–
50% A in B, 8.0–12.0 min, 50% B (solvent A: 0.1% concd
NH4OH in H2O. Solvent B: 20% A in acetonitrile).
Elemental analyzes were obtained from the University
of Copenhagen, Microanalytical Department. Crystal
3. Conclusion
An efficient synthesis of the LNA G phosphoramidite
has been developed taking advantage of the regiospecific
Vorbruggen coupling of 2-amino-6-chloropurine to the
carbohydrate moiety, eliminating the issue of traces of
the N7 regioisomer in the final product. The guanine
nucleobase was protected with the dimethylformamidine
protection group allowing for the use of fast ON
deprotection protocols when this is required. Moreover,
the NMR spectra of both the N7 and N9 regioisomers
€
€
after the Vorbruggen coupling (9 and 10) have been
studied in detail, providing a simple tool for the
assignments LNA guanine nucleosides by NMR spec-
trometry.
The 2-amino-6-chloropurine nucleoside (2) obtained in
€
the Vorbruggen coupling proved to be a versatile syn-
X-ray crystallographic data of nucleoside 5 was
obtained from University of Copenhagen, Centre for
thon for the preparation of a wide range of LNA ana-
logues with modified nucleobases, as exemplified by the
synthesis of the LNA 2,6-diaminopurine (D) phosphor-
amidite.
Crystallographic Studies.||
4.1. 9-(2-O-Acetyl-3-O-benzyl-5-O-methanesulfonyl-4-C-
In full agreement with the literature, the extreme north
conformation of the LNA residues resulted in a higher
increase in thermal stability towards complementary
RNA strands rather than towards complementary DNA
strands. This effect is probably due to a more favourable
helical conformation, and thus improved base stacking,
hydration of the duplex and base-pairing between D and
T.
methanesulfonyloxymethyl-b- -erythro-furanosyl)-2-
D
amino-6-chloropurine (2)
1,2-Di-O-acetyl-5-O-methanesulfonyl-4-C-methane-
sulfonyloxymethy-3-O-benzyl- -erythro-pentofuranose
(1) (19.0 g, 37.2 mmol) was dissolved in 1,2-dichloro-
D
ethane (250 mL) and 2-amino-6-chloropurine (6.9 g,
The mismatch sensitivity of LNA D compared to LNA
A containing ONꢂs has been evaluated towards DNA
and RNA. Towards RNA no clear difference was seen
between LNA A and LNA D. Towards DNA the fully
modified LNA ON showed enhanced mismatch dis-
crimination towards singly mismatched DNA when
LNA A was replaced with LNA D.
||
Crystallographic data (excluding structure factors) for the structures
in this paper have been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication no CCDC
216314. Copies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK
(fax: +44-(0)1223-336033 or e-mail: deposit@ccdc.cam.ac.uk).

2392
C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
40.9 mmol) was added followed by N,O-bis(trimethylsi-
lyl)acetamide (18.2 mL, 74.4 mmol). The reaction mix-
ture was refluxed until it became a clear solution (1 h)
and cooled to rt. Trimethylsilyl triflate (13.5 mL,
74.4 mmol) was added over 15 min and the reaction
mixture was refluxed for 3 h. The reaction mixture was
allowed to cool to rt and was poured into satd aq
NaHCO3 (500 mL). CHCl3 (300 mL) was added and the
mixture was stirred for 30 min and transferred to a
separatory funnel. The phases were separated and the aq
phase was extracted with CHCl3 (3 · 250 mL). The
combined organic phases were washed with satd aq
NaHCO3/brine (1:1, 500 mL), dried (Na2SO4), filtered
and evaporated in vacuo to give a red foam. The
product was purified by DCVC (i.d. 10 cm; 100 mL
fractions; 50–100% EtOAc in n-heptane (v/v) 10%
increments; 1–10% MeOH in EtOAc (v/v) 1% incre-
ments). Fractions containing nucleoside 2 were pooled
and evaporated in vacuo to give a white foam (20.7 g,
90%). Rf ¼ 0.75 (5% MeOH in EtOAc, v/v). ESI-MS
m=z found 620.1 ([MH]þ, calcd 620.1). 1H NMR
(CDCl3): d 7.76 (s, 1H, H8), 7.40–7.32 (m, 5H, Ar–H),
6.00 (d, 1H, J 3.3, H10), 5.84 (dd, 1H, J 5.9 and 3.3,
H20), 5.44 (br s, 2H, NH2), 5.19 (d, 1H, J 6.0, H30), 4.78
(d, 1H, J 10.4, CH2), 4.68–4.61 (m, 3H, H20, CH2), 4.41–
4.38 (m, 2H, CH2), 3.03 (s, 3H, Ms), 3.00 (s, 3H, Ms),
2.11 (s, 3H, Ac). 13C NMR (CDCl3): d 169.6 (CH3C(O)),
159.1 (C2), 152.4, 152.1 (C4, C6), 141.5 (C8), 136.5,
128.6, 128.5, 128.3 (Ph), 125.9 (C5), 88.3 (C10), 84.2
(C40), 78.0 (C30), 74.6 (PhCH2), 74.0 (C20), 67.5, 67.2
(C50 and C100 ), 37.8, 37.5 (2 · Ms), 20.6 (CH3C(O)).
Elemental analysis: calcd for C22H26ClN5O10S2: C, 42.6;
H, 4.2; N, 11.3. Found: C, 42.8; H, 4.2; N, 10.9.
H30), 4.68 (s, 2H, PhCH2), 4.59 (d, 1H, J 11.9, H50), 4.41
(s, 1H, H20 or H30), 4.03 (d, 1H, J 8.1, H100 ), 3.89 (d, 1H,
J 8.1, H100 ), 3.24 (s, 3H, Ms). 13C NMR (DMSO-d6): d
156.7 (C6), 154.0 (C2), 150.5 (C4), 137.6 (Ph), 134.1
(C8), 128.4, 127.8, 127.7 (Ph), 116.7 (C5), 85.0 (C40),
84.5 (C10), 77.7, 77.0 (C20 and C30), 71.7, 71.4 (PhCH2
and C100 ), 66.1 (C50), 36.9 (Ms). Elemental analysis:
calcd for C19H21N5O7S Æ 0.25H2O: C, 48.8; H, 4.6; N,
15.0. Found: C, 48.6; H, 4.4; N, 14.7.
4.3. (1R,3R,4R,7S)-3-(Guanin-9-yl)-7-benzyloxy-1-ben-
zoyloxymethyl-2,5-dioxabicyclo-[2.2.1]heptane (4)
Compound 3 (14.92 g, 32.2 mmol) was dissolved in
DMSO (400 mL) and NaOBz (14.02 g, 96.6 mmol) was
added. The reaction mixture was heated to 100 ꢀC with
stirring. After 3.5 h the reaction was allowed to cool to rt
and EtOAc (400 mL) and H2O/satd aq NaHCO3 (3:1,
500 mL) was added. The solution was transferred to a
separatory funnel and the phases were separated. The aq
phase was extracted with dichloromethane (3 · 300 mL).
The combined organic phases were dried (Na2SO4), fil-
tered and evaporated in vacuo to give a brown oil. The
product was diluted with EtOH (96%, 300 mL) and
boiled for 10 min with decolorizing charcoal. The slurry
was filtered through a plug of Celite and hot water
(500 mL) was added to the solution until it became
turbid. The mixture was allowed to cool to rt and was
then cooled to 5 ꢀC. The product 4 was filtered off,
washed with cold water and ether to give a tan solid
(14.97 g, 95%). Rf ¼ 0.49 (10% MeOH in EtOAc v/v).
ESI-MS m=z found 490.2 ([MH]þ, calcd 490.2). 1H
NMR (DMSO-d6): d 10.66 (br s, 1H, NH), 7.93 (d, 2H,
J 7.1, Bz), 7.70 (s, 1H, H8), 7.67 (d, 1H, J 7.5, Bz), 7.53
(d, 2H, J 7.8, Bz), 7.35–7.24 (m, 5H, Ar–H), 6.57 (br s,
2H, NH2), 5.82 (s, 1H, H10), 4.80–4.65 (m, 5H, H20 or
H30, H50 and PhCH2), 4.50 (s, 1H, H20 or H30), 4.11 (d,
1H, J 8.1, H100 ), 3.97 (d, 1H, J 8.1, H100 ). 13C NMR
(DMSO-d6): d 165.3 (PhC(O)), 156.7 (C6), 153.9 (C2),
150.4 (C4), 137.7 (Ph), 133.9 (Bz), 133.7 (C8), 129.3,
129.1, 128.9, 128.3, 127.7, 127.6 (Ar), 116.7 (C5), 85.0,
84.9 (C10 and C40), 77.6, 77.0 (C20 and C30), 71.9, 71.3
(PhCH2 and C100 ), 60.5 (C50). Elemental analysis: calcd
for C25H23N5O6 Æ 0.5H2O: C, 60.2; H, 4.8; N, 14.0.
Found: C, 60.0; H, 4.7; N, 14.2.
4.2. (1R,3R,4R,7S)-7-Benzyloxy-1-methanesulfonyl-
oxymethyl-3-(guanin-9-yl)-2,5-dioxabicyclo[2.2.1]heptane
(3)
3-Hydroxypropionitrile (3.65 mL, 53.5 mmol) was dis-
solved in THF (75 mL) and cooled to 0 ꢀC. Sodiumhy-
dride (60% in mineral oil, 2.61 g, 65.3 mmol) was added
in portions and the temperature was allowed to reach rt.
The mixture was stirred for 30 min at rt and cooled to
0 ꢀC. Compound 2 (7.37 g, 11.9 mmol) was dissolved in
THF (75 mL) and added dropwise over 20 min and the
temperature was allowed to reach rt. After 4 h the
reaction was quenched by addition of HCl (1 M, aq)/
brine (1:9, 200 mL) and the mixture was transferred to a
separatory funnel. The phases were separated and the
aqueous phase was extracted with EtOAc (3 · 100 mL).
The combined organic phases were dried (Na2SO4), fil-
tered and evaporated in vacuo to give a yellow oil. The
product was purified by DCVC (i.d. 6 cm; 100 mL
fractions; 0–20% MeOH in EtOAc (v/v) 1% increments).
Fractions containing nucleoside 3 were pooled and
evaporated in vacuo to give an off-white solid (4.74 g,
86%). Rf ¼ 0.31 (10% MeOH in EtOAc v/v). ESI-MS
m=z found 464.1 ([MH]þ, calcd 464.1). 1H NMR
(DMSO-d6): d 10.75 (br s, 1H, NH), 7.78 (s, 1H, H8),
7.36–7.26 (m, 5H, Ar–H), 6.64 (br s, 2H, NH2), 5.80 (s,
1H, H10), 4.78 (d, 1H, J 12.1, H50), 4.75 (s, 1H, H20 or
4.4. (1S,3R,4R,7S)-3-(Guanin-9-yl)-7-hydroxy-1-hydr-
oxymethyl-2,5-dioxabicyclo-[2.2.1]heptane (5)
Compound 4 (489 mg, 1.0 mmol) was suspended in
dichloromethane (4 mL) and cooled 0 ꢀC with stirring.
Methanesulfonic acid (1.30 mL, 20 mmol) was added to
give a clear solution. After 30 min the reaction mixture
was allowed to reach rt. After stirring for an additional
3.5 h the reaction was diluted with dichloromethane
(4 mL) and cooled to 0 ꢀC. Amberlyst A-26 resin (20 mL)
pre-washed with NaOH (1 N (aq), 3 · 20 mL) and EtOH
(30 mL) was suspended in EtOH (30 mL), cooled to 0 ꢀC
and the reaction mixture was added slowly. After stir-
ring for 30 min the resin was removed by filtration
and washed with warm EtOH (4 · 20 mL). Fractions

C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
2393
containing the product were combined and the pH was
adjusted to 11 with NaOH (1 N, aq). After 10 min the
solution was neutralized with HCl (1 N, aq). The EtOH
was removed in vacuo and the remaining aqueous
solution was lyophilized to give nucleoside 5. The crude
product was purified by HPLC (Xterra column from
Waters (PrepRP18, 5 lm, 7.8 · 50 mm), with the follow-
ing gradient 0–6.0 min, 0–2% B in A; 6.0–8.0 min, 2–4%
B in A; 8.0–10.0 min, 4–100% B in A (Solvent A: 0.1 M
NH4OAc in H2O, pH ¼ 8. Solvent B: acetonitrile) and
recrystallized from water to give 5 as white needles
(145 mg, 49%). Rf ¼ 0.04 (10% MeOH in EtOAc v/v).
ESI-MS m=z found 296.3 ([MH]þ, calcd 296.1). 1H
NMR (D2O): d 7.86 (s, 1H, H8), 5.85 (s, 1H, H10), 4.60
(s, 1H, H20 or H30), 4.42 (s, 1H, H2000or H30), 4.03 (d, 1H,
J 8.6, H100 ), 3.97 (d, 1H, J 8.5, H1 ), 3.96 (s, 2H, H50).
13C NMR (D2O): d 160.8 (C6), 153.7 (C2), 150.6 (C4),
136.1 (C8), 114.5 (C5), 88.3 (C40), 84.9 (C10), 79.4 (C20),
71.4 (C30), 70.1 (C100 ), 56.7 (C50). Elemental analysis:
calcd for C11H13N5O5 Æ H2O: C, 42.2; H, 4.8; N, 22.4.
Found: C, 41.9; H, 4.8; N, 22.2.
evaporated in vacuo to give a white foam (2.48 g, 84%).
Rf ¼ 0.43 (1%, Et3N and 10% MeOH in dichloro-
methane, v/v/v). ESI-MS m=z found 653.3 ([MH]þ, calcd
653.3). 1H NMR (DMSO-d6): d 11.36 (s, 1H, NH), 8.57
(s, 1H, N@CH), 7.89 (s, 1H, H8), 7.43–7.39 (m, 2H, Ar–
H), 7.34–7.22 (m, 7H, Ar–H), 6.92–6.87 (m, 4H, Ar–H),
5.90 (s, 1H, H10), 5.73 (d, 1H, J 4.7, H30), 4.47 (s, 1H,
OH), 4.40 (d, 1H, J 4.6, H20), 4.35 (d, 1H, J 7.1, H50),
4.31 (d, 1H, J 7.1, H50), 3.74 (s, 6H, OCH3), 3.51 (d, 1H,
J 10.8, H100 ), 3.32 (d, 1H, J 10.8, H100 ), 3.14 (s, 3H,
NCH3), 3.03 (s, 3H, NCH3). 13C NMR (DMSO-d6):
d 158.2 (Ar), 158.2, 157.6, 157.4 (C2, C6 and N@CH),
149.0 (C4), 144.8 (Ar), 135.5 (C8), 135.3, 129.8, 129.7,
127.9, 127.7, 126.8 (Ar), 119.9 (C5), 113.3 (Ar), 86.7,
(C40), 85.5 (C10),0085.1 (CPh3), 79.3 (C30), 71.7 (C50), 70.7
(C20), 59.9 (C1 ), 55.1, 55.0 (2 · OCH3), 40.7, 34.7
(N(CH3)2). Elemental analysis: calcd for C35H36N6O7:
C, 64.4; H, 5.6; N, 12.9. Found: C, 64.2; H, 5.5; N, 13.0.
4.6. (1R,3R,4R,7S)-1-(4,40-Dimethoxytrityloxymethyl)-7-
(2-cyanoethoxy(diisopropylamino)phosphinoxy)-3-[2-N-
((N0,N0-dimethylamino)methylidene)-guanin-9-yl]-2,5-
dioxabicyclo[2.2.1]heptane (8)
4.5. (1R,3R,4R,7S)-1-(4,40-Dimethoxytrityloxymethyl)-7-
hydroxy-3-(2-N-(dimethylamino)methylidene)-guanin-9-
yl)-2,5-dioxabicyclo[2.2.1]heptane (7)
Compound 7 (3.20 g, 4.90 mmol) was dissolved in
dichloromethane–DMF (2:1, 70 mL) and 4,5-dicyano-
imidazole in acetonitrile (1.0 M, 3.43 mL, 3.43 mmol)
was added. 2-Cyanoethyl-N,N,N0,N0-tetraisopropyl-
phosphorodiamidite (1.72 mL, 5.39 mmol) was added
dropwise to the reaction mixture. After 4 h the reaction
was diluted with dichloromethane (70 mL) and trans-
ferred to a separatory funnel and extracted with satd aq
NaHCO3 (2 · 50 mL) and brine (50 mL). The combined
aqueous phases were extracted with dichloromethane
(100 mL). The combined organic phases were dried
(Na2SO4), filtered and concentrated in vacuo to give a
yellow foam. Purification by DCVC (i.d. 4 cm; 100 mL
fractions; the column was pretreated with 1% Et3N in
n-heptane (v/v); 80–100% EtOAc in n -heptane (v/v) 5%
increments; 1–10% MeOH in EtOAc (v/v) 1% incre-
ments) gave nucleoside 8 (3.68 g, 88%) as a white solid.
Rf ¼ 0.37 (1%, Et3N and 10% MeOH in CH2Cl2, v/v/v).
ESI-MS m=z found 853.2 ([MH]þ, calcd 853.4). 31P
NMR (CDCl3) d 149.7, 147.6.
Compound 4 (2.23 g, 4.56 mmol) was suspended in
dichloromethane (18 mL) and cooled to 0 ꢀC with stir-
ring. Methanesulfonic acid (5.9 mL, 91.2 mmol) was
added to give a clear solution. After 30 min the reaction
mixture was allowed to reach rt and the reaction was
stirred for an additional 3.5 h. The reaction mixture was
diluted with dichloromethane (18 mL) and cooled to
0 ꢀC. Amberlyst A-26 resin (100 mL) pre-washed with
NaOH (1 N (aq), 3 · 100 mL) and EtOH (150 mL) was
suspended in EtOH (100 mL), cooled to 0 ꢀC and the
reaction mixture was added slowly. After stirring for
30 min the resin was removed by filtration and washed
with warm EtOH (4 · 50 mL). Fractions containing the
product were combined and the pH was adjusted to 11
with NaOH (1 N, aq). After stirring for 10 min the
solution was neutralized with HCl (1 N, aq). The EtOH
was removed in vacuo and the remaining aqueous
solution was lyophilized to give compound 5. The crude
product was suspended in DMF (30 mL) and N,N-
dimethylformamide dimethyl acetal (1.23 mL, 9.1 mmol)
was added. The reaction was heated to 55 ꢀC for 30 min
and the DMF was removed in vacuo. The resulting
slurry was co-evaporated from pyridine (2 · 25 mL) and
suspended in pyridine (50 mL). 4,40-Dimethoxytrityl
chloride (2.32 g, 6.84 mmol) was added and the reaction
mixture was stirred for 3 h at rt. Most of the pyridine
was removed in vacuo and the residue was dissolved in
EtOAc (200 mL) and washed with half satd aq NaHCO3
(2 · 100 mL) and brine (100 mL). The combined aqueous
phases were extracted with EtOAc (2 · 100 mL). The
combined organic phases were dried (Na2SO4), filtered
and evaporated in vacuo to give a brown foam. The
product was purified by DCVC (i.d. 4 cm; 100 mL
fractions; 2 · 50% dichloromethane in n-heptane (v/v);
0–10% MeOH in dichloromethane (v/v) 1% increments).
Fractions containing nucleoside 7 were pooled and
4.7. (1R,3R,4R,7S)-7-Benzyloxy-1-methanesulfonyl-
oxymethyl-3-(2-N-isobutyrylguanin-9-yl)-2,5-dioxabi-
cyclo[2.2.1]heptane (9)
1H NMR (CDCl3): d 12.13 (s, 1H, NH), 9.62 (s, 1H,
NH), 7.74 (s, 1H, H8), 8.12 (s, 1H, H8), 7.35–7.24 (m,
5H), 5.83 (s, 1H, H10), 4.71–4.45 (m, 6H, H20, H30, H50,
PhCH2–), 4.12 (d, 1H, J 7.9, H100 ), 3.93 (m, 1H, H100 ),
3.06, (s, 3H, Ms), 2.77 (heptet, 1H, J 6.9, (CH3)2CH–),
1.27, 1.25 (s, 6H, (CH3)2CH–). 13C NMR (CDCl3): d
179.4 (NHC@O), 155.2 (C6), 147.9, 147.3 (C2, C4),
136.2 (C8), 128.5, 128.2, 127.9 (Ph), 121.5 (C5), 86.9
(C40), 85.4 (C10), 77.6 (PhCH2), 72.53, 72.45 (C20, C30),
72.1 (C100 ), 64.6 (C50), 37.7 (Ms), 36.2 ((CH3)2CH–),
19.0, 18.9 ((CH3)2CH–).

2394
C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
4.8. (1R,3R,4R,7S)-7-Benzyloxy-1-methanesulfonyl-
oxymethyl-3-(2-N-isobutyrylguanin-7-yl)-2,5-dioxabi-
cyclo[2.2.1]heptane (10)
product was purified by DCVC (i.d. 10 cm; 100 mL
fractions; 0–100% EtOAc in n-heptane (v/v) 10%
increments; 0.5–3% MeOH in EtOAc (v/v) 0.5%
increments). Fractions containing nucleoside 18 were
pooled and evaporated in vacuo to give a white foam
(6.63 g, 91%). Rf ¼ 0.53 (20% n-heptane in EtOAc v/v).
ESI-MS m=z found 508.0 ([MH]þ, calcd 508.1). 1H
NMR (CDCl3): d 7.56 (br d, 2H, J 7.1, Bz), 7.78 (s, 1H,
H8), 7.60 (br t, 1H, J 7.4, Bz), 7.46 (br t, 2H, J 7.7, Bz),
7.30–7.24 (m, 5H, Ph), 5.93 (s, 1H, H10), 5.25 (br s, 2H,
NH2), 4.80 (d, 1H, J 12.8, CH2), 4.77 (s, 1H, H20 or
H30), 4.67 (d, 1H, J 11.9, CH2), 4.65 (d, 1H, J 12.6,
CH2), 4.56 (d, 1H, J 11.9, CH2), 4.27 (d, 1H, J 7.9,
CH2), 4.25 (s, 1H, H20 or H30), 4.08 (d, 1H, J 8.1, CH2).
13C NMR (CDCl3): d 166.0 (PhC(O)), 159.2, 152.4,
151.8 (C2, C4 and C6), 139.1 (C8), 136.7, 133.7, 129.7,
129.3, 128.8, 128.7, 128.4, 127.9 (Ar), 125.8 (C5), 86.7
(C10), 86.0 (C40), 77.5, 77.4 (C20 and C30), 72.8, 72.6,
59.8 (3 · CH2). Elemental analysis: calcd for
C25H22ClN5O5: C, 59.1; H, 4.4; N, 13.8. Found: C, 58.8;
H, 4.3; N, 13.5.
1H NMR (CDCl3): d 12.47 (s, 1H, NH), 10.80 (s, 1H,
NH), 7.35–7.20 (m, 5H), 6.24 (s, 1H, H10), 4.84–4.48 (m,
6H, H20, H30, H50, PhCH2–), 4.16 (d, 1H, J 8.0, H100 ),
3.97 (d, 1H, J 8.1, H100 ), 3.13 (s, 3H, Ms), 2.94 (heptet,
1H, J 6.9, (CH3)2CH–), 1.31–1.24 (m, 6H, (CH3)2CH–).
13C NMR (CDCl3): d 180.6 (NHC@O), 158.0 (C4),
153.2 (C6), 148.4 (C2), 140.5 (C8), 128.8, 128.5, 128.1
(Ph), 110.8 (C5), 88.7 (C40), 86.1 (C10), 77.8 (PhCH2),
72.5, 72.22, 72.25 (C100 , C20, C30), 64.8 (C50), 37.9 (Ms),
36.2 ((CH3)2CH–), 19.4, 19.3 ((CH3)2CH–).
4.9. (1R,3R,4R,7S)-3-(2-Amino-6-chloropurine)-7-benz-
yloxy-1-methanesulfonyloxymethyl-2,5-dioxabi-
cyclo[2.2.1]heptane (17)
To a solution of compound 2 (10.1 g, 16.3 mmol) in
THF (160 mL) aq LiOH (1.0 M, 82 mL, 82 mmol) was
added. The reaction was stirred at ambient temperature
for 45 min and glacial acetic acid (5 mL, 87 mmol) was
added. The reaction mixture was transferred to a
separatory funnel with brine (200 mL) and extracted
with EtOAc (200 + 2 · 100 mL). The combined organic
phases were dried (Na2SO4), filtered and evaporated in
vacuo to give a yellow liquid. The product was dissolved
in dichloromethane and toluene was added. The solution
was evaporated to dryness in vacuo to give a yellow
foam that was purified by DCVC (i.d. 10 cm; 100 mL
fractions; 50–100% EtOAc in n-heptane (v/v) 10%
increments; 1–8% MeOH in EtOAc (v/v) 1% incre-
ments). Fractions containing nucleoside 17 were com-
bined and evaporated in vacuo to give a white foam
(7.03 g, 90%). Rf ¼ 0.41 (100% EtOAc). ESI-MS m=z
4.11. (1R,3R,4R,7S)-3-(2,6-Tribenzoyl-2,6-diamino-
purine-9-yl)-7-benzyloxy-1-benzoyloxymethyl-2,5-dioxa-
bicyclo-[2.2.1]heptane and (1R,3R,4R,7S)-3-(2,6-tetra-
benzoyl-2,6-diaminopurine-9-yl)-7-benzyloxy-1-benzoyl-
oxymethyl-2,5-dioxabicyclo-[2.2.1]heptane (20)
Compound 18 (1.0 g, 1.97 mmol) was dissolved in
MeOH saturated with ammonia (50 mL) and transferred
to a heavy-wall borosilicate pressure glass tube fitted
with a threaded PTFE plug and a pressure gauge. The
reaction was heated to 80 ꢀC (60 psi) for 4 h. The reac-
tion mixture was cooled to 0 ꢀC before opening the
pressure tube. The reaction mixture was transferred to a
round bottom flask and the solvent removed in vacuo to
give a brown oil. The crude product was evaporated
from pyridine (2 · 25 mL) and dissolved in pyridine
(50 mL). Benzoyl chloride (1.37 mL, 11.8 mmol) was
added and the reaction stirred at rt for 16 h. Most of the
solvent was removed in vacuo, the residue dissolved in
EtOAc (100 mL) and transferred to a separatory funnel.
The organic phase was extracted with half satd aq
NaHCO3 (2 · 100 mL) and brine (100 mL). The com-
bined aqueous phases were extracted with EtOAc
(2 · 100 mL). The combined organic phases were dried
(Na2SO4), filtered and evaporated in vacuo to give a
brown oil. The product was purified by DCVC (i.d.
4 cm; 100 mL fractions; 25–75% EtOAc in n-heptane
(v/v) 5% increments). Fractions containing nucleosides
20 were pooled and evaporated in vacuo to give a pink
foam (1.1 g, 65% 1:1 mixture of the tetra- and penta
benzoylated nucleosides).
1
found 482.0 ([MH]þ, calcd 482.1). H NMR (CDCl3): d
7.83 (s, 1H, H8), 7.33–7.26 (m, 5H, Ar–H), 5.91 (s, 1H,
H10), 5.28 (br s, 2H, NH2), 4.73 (s, 1H), 4.68–4.59 (m,
4H), 4.31 (s, 1H), 4.18 (d, 1H, J 7.9), 4.00 (d, 1H, J 8.1)
(3 · CH2, H20 and H30), 3.10 (s, 3H, Ms). 13C NMR
(CDCl3): d 159.0, 152.3, 151.6, 139.2 (C2, C4, C6 and
C8), 136.5, 128.6, 128.4, 127.9 (Ar), 125.5 (C5), 88.6,
85.4, 76.9, 72.6, 72.2, 64.0, 53.8 (3 · CH2, C10, C20, C30
and C40), 37.8 (Ms). Elemental analysis: calcd for
C19H20ClN5O6S Æ 0.5EtOAc: C, 48.0; H, 4.6; N, 13.3.
Found: C, 48.4; H, 4.5; N, 13.3.
4.10. (1R,3R,4R,7S)-3-(2-Amino-6-chloropurine-9-yl)-7-
benzyloxy-1-benzoyloxymethyl-2,5-dioxabicyclo-
[2.2.1]heptane (18)
Compound 17 (6.9 g, 14.3 mmol) was dissolved in DMF
(300 mL) and NaOBz (4.13 g, 28.6 mmol) was added.
The reaction mixture was heated to 100 ꢀC with stirring.
After 4 h the warm reaction mixture was poured into
water (250 mL) and allowed to cool. The solution was
transferred to a separatory funnel with brine (200 mL)
and extracted with dichloromethane (200 + 2 · 100 mL).
The combined organic phases were dried (Na2SO4), fil-
tered and evaporated in vacuo to give a yellow gum. The
4.12. (1R,3R,4R,7S)-3-(2,6-Tribenzoyl-2,6-diamino-
purine-9-yl)-7-benzyloxy-1-benzoyloxymethyl-2,5-dioxa-
bicyclo-[2.2.1]heptane
Rf ¼ 0.57 (30% n-heptane in EtOAc, v/v). HPLC: TR
¼
6.49 min. ESI-MS m=z found 801.2 ([MH]þ, calcd
801.3).

C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
2395
4.13. (1R,3R,4R,7S)-3-(2,6-Tetrabenzoyl-2,6-diamino-
purine-9-yl)-7-benzyloxy-1-benzoyloxymethyl-2,5-dioxa-
bicyclo-[2.2.1]heptane
were pooled and evaporated in vacuo to give a white
foam (452 mg, 74%). Rf ¼ 0.18 (EtOAc). ESI-MS m=z
1
found 805.2 ([MH]þ, calcd 805.3). H NMR (DMSO-
d6): d 11.23 (s, 1H, NH), 11.03 (s, 1H, NH), 8.40 (s, 1H,
H8), 8.07 (d, 2H, J 7.3, Bz), 7.99 (d, 2H, J 7.3, Bz), 7.68–
7.50 (m, 6H, Bz), 7.44–7.40 (m, 2H, Ar–H), 7.35–7.21
(m, 7H, Ar–H), 6.92–6.88 (m, 4H, Ar–H), 6.08 (s, 1H,
H10), 5.75 (d, 1H, J 4.6, H30), 4.65 (s, 1H, OH), 4.42 (d,
1H, J 4.7, H20), 4.02 (d, 1H, J 7.9, H50), 3.97 (d, 1H, J
7.9, H50), 3.73 (s, 6H, OCH3), 3.59 (d, 1H, J 11.9, H100 ),
3.31 (d, 1H, J 11.0, H100 ). 13C NMR (DMSO-d6): d
165.5, 165.4 (C@O), 158.0 (Ar), 152.3, 152.1, 150.7 (C2,
C4 and C6), 144.6 (Ar), 140.9 (C8), 135.4, 135.1, 134.1,
133.1, 129.6, 129.5, 128.3, 128.0, 127.8, 127.5, 126.6
(Ar), 123.3 (C5), 113.1 (Ar), 87.0 (C40), 85.4, 85.2 (C10
and OCPh3), 78.9 (C30), 71.7 (C50), 70.8 (C20), 59.9
(C100 ), 54.9 (OCH3). Elemental analysis: calcd for
C46H40N6O8 Æ H2O: C, 67.1; H, 5.1; N, 10.2. Found: C,
66.9; H, 4.9; N, 10.1.
Rf ¼ 0.67 (30% n-heptane in EtOAc, v/v). HPLC: TR ¼
7.37 min. ESI-MS m=z found 905.1 ([MH]þ, calcd
905.3).
4.14. (1R,3R,4R,7S)-3-(2,6-N,N0-Dibenzoyl-2,6-diamino-
purine-9-yl)-7-benzyloxy-1-benzoyloxymethyl-2,5-dioxa-
bicyclo-[2.2.1]heptane
The mixture 20 was treated with diluted ammonia in
MeOH for 5 min producing the tribenzoylated nucleo-
side. Rf ¼ 0.37 (30% n-heptane in EtOAc, v/v). HPLC:
TR ¼ 5.60 min. ESI-MS m=z found 697.2 ([MH]þ, calcd
697.2). 1H NMR (DMSO-d6): d 11.22 (s, 1H, NH), 11.07
(s, 1H, NH), 8.40 (s, 1H, H8), 8.07 (d, 2H, J 7.3, Bz), 8.0
(d, 2H, J 7.5, Bz), 7.94 (d, 2H, J 7.1, Bz), 7.68–7.47 (m,
9H, Bz), 7.31–7.22 (m, 5H, Ar–H), 6.14 (s, 1H, H10),
5.07 (s, 1H), 4.88 (d, 1H) (H20, H30), 4.80 (s, 2H,
PhCH2), 4.74 (d, 2H, J 4.6, H50), 4.18 (d, 1H, J 8.2,
H100 ), 4.05 (d, 1H, J 8.1, H100 ). 13C NMR (DMSO-d6): d
165.6, 165.3 (NHC@O), 152.4, 152.2, 150.9 (C2, C4 and
C6), 142.0, 141.9 (Ph), 137.8 (C8), 134.2, 133.6, 132.5,
132.0, 129.3, 128.8, 128.5, 128.2, 128.0, 127.5 (Ar), 123.5
(C5), 85.9, 85.2 (C10, C40), 78.2, 77.1 (C20, C30), 72.0,
71.3 (C100 , PhCH2), 60.7 (C50).
4.16. (1R,3R,4R,7S)-1-(4,40-Dimethoxytrityloxymethyl)-
7-(2-cyanoethoxy(diisopropylamino)phosphinoxy)-3-[2,6-
N,N0-dibenzoylaminopurin-9-yl]-2,5-dioxabicyclo-
[2.2.1]heptane (16)
Compound 15 (490 mg, 0.60 mmol) was dissolved in
DMF (10 mL) and 4,5-dicyanoimidazole in acetonitrile
(1.0 M, 0.42 mL, 0.42 mmol) was added with stirring.
2-Cyanoethyl-N,N,N0,N0-tetraisopropylphosphorodiami-
dite (0.21 mL, 0.64 mmol) was added dropwise to the
reaction mixture. After 3 h the reaction was diluted with
EtOAc (50 mL) and transferred to a separatory funnel
and extracted with half satd aq NaHCO3 (2 · 50 mL)
and brine (50 mL). The combined aq phases were
extracted with EtOAc (50 mL). The organic phases were
pooled and dried (Na2SO4). After filtration the organic
phase was concentrated in vacuo to give a yellow foam.
Purification by DCVC (i.d. 2 cm; 20 mL fractions; the
column was pretreated with 1% Et3N in n-heptane (v/v);
50–100% EtOAc in n-heptane (v/v) 5% increments) gave
nucleoside 16 (570 mg, 95%) as a white solid. Rf ¼ 0.45
(EtOAc). ESI-MS m=z found 1005.3 ([MH]þ, calcd
1005.4). 31P NMR (CDCl3). d 150.4, 149.8.
4.15. (1R,3R,4R,7S)-1-(4,40-Dimethoxytrityloxymethyl)-
7-hydroxy-3-(2,6-N,N0-dibenzoylaminopurin-9-yl)-2,5-di-
oxabicyclo[2.2.1]heptane (15)
Compound 20 (650 mg, 0.76 mmol) was dissolved in
dichloromethane (3 mL) and cooled to 0 ꢀC. Methane-
sulfonic acid (1.00 mL, 20 mmol) was added and after
1 h the temperature was allowed to reach rt. After an
additional 3 h the reaction was diluted with dichloro-
methane (4 mL) and cooled to 0 ꢀC. Amberlyst A-26
resin (40 mL) washed with NaOH (1 N (aq), 3 · 20 mL)
was suspended in EtOH (30 mL), cooled to 0 ꢀC and the
reaction mixture was added slowly. After stirring for
10 min the resin was filtered off and washed with EtOH
(3 · 25 mL). The solution was neutralized with HCl
(1.0 N, aq) and concentrated in vacuo. The residue was
lyophilized to give crude 21. The product was evapo-
rated from pyridine (2 · 25 mL) and dissolved in pyr-
idine (35 mL). 4,40-Dimethoxytrityl chloride (515 mg,
1.52 mmol) was added and the reaction mixture stirred
overnight. Most of the pyridine was removed in vacuo
and the residue dissolved in EtOAc (50 mL) and washed
with half satd aq NaHCO3 (2 · 50 mL) and brine
(50 mL). The combined aqueous phases were extracted
with EtOAc (2 · 50 mL). The combined organic phases
were dried (Na2SO4), filtered and evaporated in vacuo to
give a brown foam. The product was purified by DCVC
(i.d. 2 cm; 50 mL fractions; 75–100% EtOAc in n-hep-
tane (v/v) 5% increments; 1–10% MeOH in EtOAc (v/v)
1% increments). Fractions containing nucleoside 15
4.17. Oligonucleotide synthesis
ON were synthesized using phosphoramidite chemistry
on an Expedite 8900/MOSS synthesizer (Multiple Oli-
gonucleotide Synthesis System) on a 1 lmol scale. The
b-D-LNA monomers were obtained from Proligo
(Boulder, CO). After completed ON synthesis (DMT–
ON), the ON were cleaved from the solid support using
aqueous ammonia for 1–2 h at rt, and further depro-
tected for 4 h at 65 ꢀC. ON containing 2,6-diamino-
purine required deprotection for 18 h at 65 ꢀC. The
crude ON were purified by reverse phase HPLC. After
removal of the DMT-group, the ON were characterized
by RP-HPLC and the structure was further confirmed
by ESI-MS.

2396
C. Rosenbohm et al. / Bioorg. Med. Chem. 12 (2004) 2385–2396
17. Vince, R.; Brownell, J.; Beers, S. A. Nucleosides Nucleo-
tides 1995, 14, 39–44.
4.18. UV melting analysis
ꢀ
18. Zagorowska, I.; Adamiak, R. W. Biochimie 1996, 78, 123–
130.
Equimolar solutions of ON and their complementary
DNA or RNA (final concentration, 4 lM) were mixed
and dissolved in 100 mM phosphate buffer, 1 M NaCl,
pH ¼ 7, heated to 95 ꢀC for 2 min and then cooled down
to rt. Meltings were recorded on a Lambda 20 Perkin–
Elmer UV–vis spectrometer attached to a PTP6 Peltier
Temperature Programmer in 1 cm path-length quartz
cells at 260 nm using a heating rate of 1 ꢀC/min, a
response of 0.2 s and a slit of 2 nm with a starting tem-
perature of 5 ꢀC. Tm values were obtained from the
maxima of the first derivatives of the melting curves.
19. Jankowski, A. J.; Wise, D. S.; Townsend, L. B. Nucleo-
sides Nucleotides 1989, 8, 339–348.
20. Brand, B.; Reese, C. B.; Song, Q.; Visintin, C. Tetrahedron
1999, 55, 5239–5252.
21. Trivedi, B. K. Nucleosides Nucleotides 1988, 7, 393–402.
22. Konkel, M. J.; Vince, R. Nucleosides Nucleotides 1995, 14,
2061–2077.
23. Chen, X.; Kern, E. R.; Drach, J. C.; Gullen, E.; Cheng,
Y.-C.; Zemlicka, J. J. Med. Chem. 2003, 46, 1531–1537.
24. Luyten, I.; Herdewijn, P. Eur. J. Med. Chem. 1998, 33,
515–576.
25. Brown, B.; Hegedus, L. S. J. Org. Chem. 2000, 65, 1865–
1872.
26. Rhee, H.; Yoon, D.; Jung, M. E. Nucleosides Nucleotides
Nucleic Acids 2000, 19, 619–628.
References and notes
27. Reitz, A. B.; Rebarchak, M. C. Nucleosides Nucleotides
1992, 11, 1121.
28. Ashwell, M.; Bleasdale, C.; Golding, B. T.; OꢂNeill, I. K.
J. Chem. Soc., Chem. Commun. 1990, 955–956.
29. Matteson, D. S.; Man, H.-W.; Ho, O. C. J. Am. Chem.
Soc. 1996, 118, 4560–4566.
1. 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.
2. Singh, S. K.; Nielsen, P.; Koshkin, A. A.; Wengel, J.
Chem. Commun. 1998, 455–456.
3. Obika, S.; Nanbu, D.; Hari, Y.; Andoh, J.-I.; Morio,
K.-I.; Doi, T.; Imanishi, T. Tetrahedron Lett. 1998, 39,
5401–5404.
4. Wengel, J. LNA (Locked Nucleic Acid). In Antisense Drug
Technology; Principles, Strategies and Applications;
Crooke, S. T., Ed.; Marcel Dekker: New York, Basel,
2001; pp 339–357.
30. Kjellberg, J.; Johansson, N. G. Tetrahedron 1986, 42,
6541–6544.
31. Chenon, M.-T.; Pugmire, R. J.; Grant, D. M.; Panzica,
R. P.; Townsend, L. B. J. Am. Chem. Soc. 1975, 97, 4627–
4636.
32. Bailey, S.; Harnden, M. R. Nucleosides Nucleotides 1987,
6, 555–574.
5. Cook, P. D. Nucleosides Nucleotides 1999, 18, 1141–1162.
€
6. Vorbruggen, H.; Bennua, B. Chem. Ber. 1981, 114, 1279–
33. Gryaznov, S.; Schultz, R. G. Tetrahedron Lett. 1994, 35,
2489–2492.
1286.
7. Vorbruggen, H.; Hofle, G. Chem. Ber. 1981, 114, 1256–
1268.
8. Koshkin, A. A.; Fensholdt, J.; Pfundheller, H. M.;
Lomholt, C. J. Org. Chem. 2001, 66, 8504–8512.
9. Parel, S. P.; Leumann, C. J. Helv. Chim. Acta 2000, 83,
2514–2526.
10. Geen, G. R.; Kincey, P. M.; Spoors, P. G. Tetrahedron
Lett. 2001, 42, 1781–1784.
11. Robins, M. J.; Zou, R.; Guo, Z.; Wnuk, S. F. J. Org.
Chem. 1996, 61, 9207–9212.
12. Jenny, T. F.; Schneider, K. C.; Benner, S. A. Nucleosides
Nucleotides 1992, 11, 1257–1261.
13. Kjellberg, J.; Johansson, N. G. Nucleosides Nucleotides
1989, 8, 225–256.
14. Geen, G. R.; Grinter, T. J.; Kincey, P. M.; Jarvest, R. L.
Tetrahedron 1990, 46, 6903–6914.
34. Cano, A.; Goodman, M. F.; Eritja, R. Nucleosides
Nucleotides 1994, 13, 501–509.
35. Cheong, C.; Tinoco, I.; Chollet, A. Nucleic Acids Res.
1988, 16, 5115–5122.
36. Gaffney, B. L.; Marky, L. A.; Jones, R. A. Tetrahedron
1984, 40, 3–13.
37. Boudou, V.; Kerremans, L.; Bouvere, B. D.; Lescrinier, E.;
Schepers, G.; Busson, R.; Aerschot, A. V.; Herdewijn, P.
Nucleic Acids Res. 1999, 27, 1450–1456.
38. Matray, T.; Gamsey, S.; Pongracz, K.; Gryaznov, S.
Nucleosides Nucleotides Nucleic Acids 2000, 19, 1553–
1567.
39. Wu, X.; Delgado, G.; Krisnamurthy, R.; Eschenmoser, A.
Org. Lett. 2002, 4, 1283–1286.
40. Petersen, M.; Bondensgaard, K.; Wengel, J.; Jacobsen,
J. P. J. Am. Chem. Soc. 2002, 124, 5974–5982.
€
€
ꢁ
41. Nielsen, K. E.; Petersen, M.; Hakansson, A. E.; Wengel,
15. Cooperwood, J. S.; Boyd, V.; Gumina, G.; Chu, C. K.
Nucleosides Nucleotides Nucleic Acids 2000, 19, 219–236.
16. Lakshman, M. K.; Hilmer, J. H.; Martin, J. Q.; Keeler, J.
C.; Dinh, Y. Q. V.; Ngassa, F. N.; Russon, L. M. J. Am.
Chem. Soc. 2001, 123, 7779–7787.
J.; Jacobsen, J. P. Chem. Eur. J. 2002, 8, 3001–3009.
42. Woo, J.; Meyer, R. B.; Gamper, H. B. Nucleic Acids Res.
1996, 24, 2470–2475.
43. Pedersen, D. S.; Rosenbohm, C. Synthesis 2001, 2431–
2434.