An α-D-configured bicyclic nucleoside restricted in an E-type conformation: Synthesis and parallel RNA recognition

Article abstract of DOI:10.1021/jo0500380

An α-D-arabino configured bicyclic nucleoside strongly restricted in an E-type conformation by a 2′-3′-fused oxetane ring is synthesized. Several synthetic strategies toward the target compound are described, and the successful preparation from a D-xylose

Full text of DOI:10.1021/jo0500380

An r-D-Configured Bicyclic Nucleoside Restricted in an E-type
Conformation: Synthesis and Parallel RNA Recognition
Pawan K. Sharma,† Michael Petersen, and Poul Nielsen*
Nucleic Acid Center,‡ Department of Chemistry, University of Southern Denmark,
5230 Odense M, Denmark
pon@chem.sdu.dk
Received January 7, 2005
An R-D-arabino configured bicyclic nucleoside strongly restricted in an E-type conformation by a
2′-3′-fused oxetane ring is synthesized. Several synthetic strategies toward the target compound
are described, and the successful preparation from a D-xylose derivative is based on a ruthenium-
mediated cleavage of a double bond, an SN2-inversion at the 2-position to give an arabino-
configuration, nucleobase coupling, and finally ring closure to give the oxetane ring. The E-type
conformation is confirmed by molecular modeling and NMR. The nucleoside is incorporated into
short R-DNA sequences. In a mixed pyrimidine context, these recognize complementary parallel
RNA-sequences with mainly increased affinity and complementary parallel DNA-sequences with
decreased affinity. The present bicyclic analogue represents the first conformationally restricted
R-DNA-analogue to improve nucleic acid recognition in mixmers with R-DNA monomers.
Introduction
monomer 1, Figure 1) are locked in N-type conforma-
tions,3,5 hereby dictating an overall A-type duplex forma-
tion.6,7 Thus, the incorporation of single LNA monomers
into an oligodeoxynucleotide induces a conformational
shift in the neighboring 2′-deoxynucleotides toward the
N-type conformation leading to strongly stabilized du-
plexes with antiparallel DNA and RNA complements.5,6
R-Configured 2′-deoxy nucleic acid sequences (R-DNA) are
well-known for their recognition of complementary DNA
and RNA sequences forming stable parallel stranded
duplexes.8 This rather unusual structural motif can be
Conformational restriction of synthetic nucleic acids
by the use of nucleoside analogues with bi- or tricyclic
carbohydrate moieties has been an excellent tool in
nucleic acid chemical biology.1,2 The induced control of
geometry has led to stabilized nucleic acid duplexes and
other structural motifs with great potential in the
development of therapeutics1-3 and diagnostics,3 and for
the design of functional nucleic acid architectures.4 The
most intriguing example is LNA (locked nucleic acid) in
which the nucleoside building blocks (as the thymine
(5) (a) Singh, S. K.; Nielsen, P.; Koshkin, A. A.; Wengel, J. Chem.
Commun. 1998, 455-456. (b) 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. (c) Obika, S.; Nanbu, D.; Hari,
Y.; Andoh, J.; Morio, K.; Doi, T.; Imanishi, T. Tetrahedron Lett. 1998,
39, 5401-5404.
(6) (a) Bondensgaard, K.; Petersen, M.; Singh, S. K.; Rajwanshi, V.
K.; Kumar, R.; Wengel, J.; Jacobsen, J. P. Chem. Eur. J. 2000, 6, 2687-
2695. (b) Petersen, M.; Bondensgaard, K.; Wengel, J.; Jacobsen, J. P.
J. Am. Chem. Soc. 2002, 124, 5974-5982.
(7) For definitions and nomenclature of nucleoside and nucleic acid
structure and conformations, see: Saenger, W. Principles of Nucleic
Acid Structure; Springer: New York, 1984.
† On leave from Kurukshetra University, Kurukshetra 136119,
India.
‡ The Nucleic Acid Center is funded by the Danish National
Research Foundation for studies on nucleic acid chemical biology.
(1) (a) Kool, E. T. Chem. Rev. 1997, 97, 1473-1487. (b) Herdewijn,
P. Biochim. Biophys. Acta 1999, 1489, 167-179. (c) Meldgaard, M.;
Wengel, J. J. Chem. Soc., Perkin Trans 1 2000, 3539-3554. (d)
Leumann, C. J. Bioorg. Med. Chem. 2002, 10, 841-854.
(2) Kurreck, J. Eur. J. Biochem. 2003, 270, 1628-1644.
(3) Petersen, M.; Wengel, J. Trends. Biotechnol. 2003, 21, 74-81.
(4) Wengel, J. Org. Biomol. Chem. 2004, 2, 277-280.
10.1021/jo0500380 CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/18/2005
4918
J. Org. Chem. 2005, 70, 4918-4928

R-D-Configured Bicyclic Nucleosides
duplexes with complementary DNA displaying overall
B-type conformation and with complementary RNA
displaying an intermediate conformation.12,13 Therefore,
we recently envisioned and proved that the anomer of
R-L-LNA, i.e., â-L-LNA is a good overall mimic of R-DNA.11
In a complete mirror-image study, â-L-LNA sequences
were proved to recognize both complementary RNA and
DNA sequences in a parallel stranded mode and with
affinities comparable to what was found for the corre-
sponding parallel R-LNA:RNA duplex.11 However, syn-
thesis of â-L-LNA monomers has not been realized and
mixmers of â-L-LNA and R-DNA have not been studied.
Recently, Keller and Leumann introduced a true S-type
R-DNA mimic with a perfect γ-torsion7 due to a tricyclic
carbohydrate skeleton.14 However, in mixmers with
R-DNA this was found to give slightly decreased affinity
toward complementary DNA and RNA.14
As a third alternative to nucleoside analogues with
strong restrictions in N- or S-type conformations, bicyclic
nucleosides restricted toward the intermediate E-type (or
O4′-endo) conformations7 have been introduced.15-17 As
the prime example, the nucleoside 3 with an 2′-O to 3′-C
oxetane ring (Figure 1) has been shown to increase the
affinity toward complementary RNA and DNA sequences
in both fully modified sequences and in mixmers with
DNA.16,18 Recently, we developed an improved synthesis
of 3 in order to undertake further biological studies.19
Also the 3′-azido-3′-deoxy analogue of 3 as well as its
R-configured epimer have been synthesized and investi-
gated as potential antiviral compounds.20 Herein, we
introduce the corresponding epimer of 3, i.e., the [3.2.0]-
bicyclic R-D-configured nucleoside 4 as the first E-type
mimic of R-DNA. Here, we expected a nucleoside ana-
logue that, in contrary to the N-type mimic 2, can work
in mixmers with R-DNA to increase the scope of parallel
nucleic acid recognition. For immediate comparison with
other studies and for synthetic convenience, we decided
to concentrate on exclusively the thymidine derivative 4
for this study and postpone the preparation of other
pyrimidine or purine analogues to subsequent studies.
FIGURE 1. R- and â-configured bicyclic thymidine analogues.
further stabilized by replacing phosphordiester moieties
in R-DNA with cationic phosphoramidate analogues.9
Conformational restriction of the sugar moieties of
R-DNA has not led to the same conclusions as with
ordinary DNA. We have introduced the R-D-configured
analogue of LNA (R-LNA, see thymine monomer 2,
Figure 1) as an R-DNA analogue with locked N-type
nucleoside conformations.10 Sequences of pyrimidine
monomers were synthesized and studied,10 whereas
mixed sequences including purines have been evaluated
recently in a mirror-image study.11 Fully modified R-LNA
sequences displayed an unprecedented recognition of
parallel complementary RNA sequences whereas comple-
mentary DNA-sequences were not recognized.10,11 On the
other hand, mixmers of R-LNA and R-DNA displayed a
low affinity for RNA.10 This observation can most prob-
ably be attributed to the inability of R-DNA monomers
to adopt N-type conformations, thus creating incoherency
between N- and S-type conformations within the same
sequence.10 This led to the conclusion that a locked
N-type conformation is not the solution for making a
stabilized parallel duplex formation with the option of
fine-tuning the thermal stability in mixmer sequences.
Results
Retrosynthetic Considerations. Several possible
routes toward the R-configured [3.2.0]bicyclic nucleoside
4 were evaluated, as indicated in Figure 2 on the level
of an abridged retrosynthetic analysis. In principle, the
The R-L-configured analogue of LNA (R-L-LNA) has
been demonstrated to be an excellent mimic of DNA.12
Thus, R-L-LNA mixmers formed high-affinity antiparallel
(8) (a) Morvan, F.; Rayner, B.; Imbach, J.-L.; Thenet, S.; Bertrand,
J.-R.; Paoletti, J.; Malvy, C.; Paoletti, C. Nucleic Acids Res. 1987, 15,
3421-3437. (b) Thuong, N. T.; Asseline, U.; Roig, V.; Takasugi, M.;
He´le`ne, C. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 5129-5133. (c)
Gagnor, C.; Bertrand, J.-R.; Thenet, S.; Lemaˆıtre, M.; Morvan, F.;
Rayner, B.; Malvy, C.; Lebleu, B.; Imbach, J.-L.; Paoletti, C. Nucleic
Acids Res. 1987, 15, 10419-10436. (d) Gagnor, C.; Rayner, B.; Leonetti,
J.-P.; Imbach, J.-L.; Lebleu, B. Nucleic Acids Res. 1989, 17, 5107-
5114.
(9) (a) Peyrottes, S.; Vasseur, J.-J.; Imbach, J.-L.; Rayner, B.
Tetrahedron Lett. 1996, 37, 5869-5872. (b) Debart, F.; Meyer, A.;
Vasseur, J.-J.; Rayner, B. Nucleic Acids Res. 1998, 26, 4551-4556. (c)
Laurent, A.; Naval, M.; Debart, F.; Vasseur, J.-J.; Rayner, B. Nucleic
Acids. Res. 1999, 27, 4151-4159. (d) Michel, T.; Martinand-Mari, C.;
Debart, F.; Lebleu, B.; Robbins, I.; Vasseur, J.-J. Nucleic Acids. Res.
2003, 31, 5282-5290.
(12) Sørensen, M. D.; Kværnø, L.; Bryld, T.; Håkansson, A. E.;
Verbeure, B.; Gaubert, G.; Herdewijn, P.; Wengel, J. J. Am. Chem.
Soc. 2002, 124, 2164-2176.
(13) (a) Nielsen, K. M. E.; Petersen, M.; Håkansson, A. E.; Wengel,
J.; Jacobsen, J. P. Chem. Eur. J. 2002, 8, 3001-3009. (b) Nielsen, J.
T.; Stein, P. C.; Petersen, M. Nucleic Acid Res. 2003, 31, 5858-5867.
(14) Keller, B. M.; Leumann, C. J. Synthesis 2002, 789-796.
(15) (a) Nielsen, P.; Pfundheller, H. M.; Olsen, C. E.; Wengel, J. J.
Chem. Soc., Perkin Trans. 1 1997, 3423-3433. (b) Minasov, G.;
Teplova, M.; Nielsen, P.; Wengel, J.; Egli, M. Biochemistry 2000, 39,
3525-3532. (c) Nielsen, P.; Petersen, M.; Jacobsen, J. P. J. Chem. Soc.,
Perkin Trans. 1 2000, 3706-3713.
(16) Christensen, N. K.; Petersen, M.; Nielsen, P.; Jacobsen, J. P.;
Olsen, C. E.; Wengel, J. J. Am. Chem. Soc. 1998, 120, 5458-5463.
(17) Kværnø, L.; Wengel, J. J. Org. Chem. 2001, 66, 5498-5503.
(18) Tømmerholt, H. V.; Christensen, N. K.; Nielsen, P.; Wengel,
J.; Stein, P. C.; Jacobsen, J. P.; Petersen, M.; Org. Biomol. Chem. 2003,
1, 1790-1797.
(19) Sharma, P. K.; Nielsen, P. J. Org. Chem. 2004, 69, 5742-5745.
(20) Sørensen, M. H.; Nielsen, C.; Nielsen, P. J. Org. Chem. 2001,
66, 4878-4886.
(10) Nielsen, P.; Dalskov, J. K. Chem. Commun. 2000, 1179-1180.
(b) Nielsen, P.; Christensen, N. K.; Dalskov, J. K. Chem. Eur. J. 2002,
8, 712-722.
(11) Christensen, N. K.; Bryld, T.; Sørensen, M. D.; Arar, K.; Wengel,
J.; Nielsen, P. Chem. Commun. 2004, 282-283.
J. Org. Chem, Vol. 70, No. 13, 2005 4919

Sharma et al.
FIGURE 2. Retrosynthetic analysis of 4. R ) varying protons or protecting groups. PG ) stable protecting groups for 5′- and
3′-alcohols. LG ) leaving group. T ) thymin-1-yl.
target compound might be accomplished by two funda-
mentally different final steps, either with the oxetane
ring formation by a nucleophilic ring-closure as the last
step on the appropriately prepared R-nucleoside A, or
alternatively, nucleobase coupling on the appropriately
prepared bicyclic furanose derivative B. The latter route
was immediately refused as attempts on Vorbru¨ggen type
coupling reactions21 with conformationally restricted
bicyclic carbohydrate substrates on several occasions
have been shown to give ring-opening reactions and
release of ring strain rather than glycosidic coupling
reactions.22 Furthermore, mixtures of R- and â-configured
products would be expected in this case. Therefore, we
concentrated on the possible routes toward the key
intermediates of type A, and the initial route to be
considered (route A, Figure 2) was modeled after the
previously published synthesis of the â-configured ana-
logue 3.16 Thus, the leaving group in A precedes a 3′-C-
hydroxymethyl group that might be obtained by oxidative
cleavage of a vinyl group in C. This compound should be
conveniently accessible by nucleobase coupling from D.
However, the compound D should have arabino-config-
uration in order to take advantage of anchimeric as-
sistance from a suitable 2-O-protecting group. Opening
of the isopropylidene ring followed by inversion of con-
figuration at C-2 of E was anticipated to give the required
D. E is well-known to be readily obtainable by the
stereoselective Grignard addition of a vinyl group to an
appropriately protected 3-ulofuranose of D-xylose.16 Al-
ternatively, C could be obtained from E via the 1,2-
anhydro-â-D-arabinofuranose derivative F based on simi-
lar transformations reported by Chow and Danishefsky.23
Thus, the salient feature of route A is the use of the vinyl
group as a precursor for the hydroxymethyl group as well
as stable protecting groups of the 3′-O and 5′-O positions
until the synthesis of nucleoside A.
Route B differs strategically from route A in the sense
that the oxidative cleavage of the double bond precedes
the nucleobase coupling. Thus, compound A should be
available by stereoselective base coupling from G. How-
ever, G must have arabino-configuration and should be
accessible with a protected hydroxymethyl group at C-3
in correct stereochemistry from H by inversion of con-
figuration at C-2. The hydroxymethyl group of H could
again be formed via oxidative cleavage of the double bond
in the precursor E. As a last alternative, a route through
the 3-methylene furanose I was considered. This would
be accessible in fewer steps starting from D-arabinose,24
and hereby the route would resemble the successful route
toward the 3′-azido-3′-deoxy analogues of 3 and 4.20
However, the route through I was not pursued as it is
known that dihydroxylation of olefin I (PG ) TBDPS)
proceeds exclusively from the â-face of the ring24 rather
than (more expectedly) from the R-face that is required
in our case. Finally, it might be considered if H could be
obtained directly from D-xylose, i.e., through the 3-ulo-
furanose thereof, with a nucleophilic attack of another
hydroxymethyl precursor rather than the vinyl group.
Nevertheless, any potential ROCH2M organometallic
reagent or anything like it would lead to an alternative
set of protecting groups to be handled. The vinyl group
is an obvious and inert precursor for the hydroxymethyl
(21) (a) Vorbru¨ggen, H.; Krolikiewicz, K.; Bennua, B. Chem. Ber.
1981, 114, 1234-1255. (b) Vorbru¨ggen, H.; Ho¨fle, G. Chem. Ber. 1981,
114, 1256-1268.
(22) Lewis acid mediated ring opening reactions in connection to
nucleobase couplings have been reported with a [3.2.0]bicyclic methyl
furanoside [(a) Obika, S.; Morio, K.; Hari, Y.; Imanishi, T. Chem.
Commun. 1999, 2423-2424] and with a [2.2.1]bicyclic methyl fura-
noside: (b) Nielsen, P.; Wengel, J. Chem. Commun. 1998, 2645-2646.
(23) Chow, K.; Danishefsky, S. J. Org. Chem. 1990, 55, 4211-4214.
(24) Houseknecht, J. B.; Lowary, T. L. J. Org. Chem. 2002, 67,
4150-4164.
4920 J. Org. Chem., Vol. 70, No. 13, 2005

R-D-Configured Bicyclic Nucleosides
SCHEME 1. Synthesis of Nucleoside 4a
a Reagents and conditions: (a) RuCl3‚xH2O, NaIO4, H2O, EtOAc, CH3CN, then NaBH4, H2O, THF, then NaIO4, then NaBH4, ref 19
(72%); (b) MsCl, pyridine (94% 7); (c) Ac2O, pyridine (98%, 14); (d) BzCl, pyridine (16); (e) CH3COCl in CH2Cl2/CH3OH/H2O (66% 8; 81%
15); (f) Tf2O, pyridine, CH2Cl2, -30 to 0 °C; (g) CsOAc, 18-crown-6, DMF, 95 °C, (46% 18 over four steps from 6); (h) thymine, N,O-
bis(trimethylsilyl)acetamide, CH3CN, TMS-triflate (68% 19); (i) NaOCH3, MeOH (90% 11 from 19); (j) MsCl, pyridine, -40 °C to rt (84%
from pure 11; 10% over five steps from 8); (k) NaH, dioxane, 50 °C (98%); (l) H2, Pd(OH)2/C (98%); (m) 4,4′-dimethoxytrityl chloride,
pyridine; (n) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(iPr)2, CH2Cl2 (79% over two steps).
group, and any other opportunities were postponed until
full coverage of the convenient starting material E.
Initial Attempts. Our first approach for the synthesis
of 4 was highly motivated by the original synthesis of
the â-analogue 316 and, therefore, followed the retrosyn-
thetic route A. Hence, the dibenzylated 3′-C-vinyl com-
pound 5 (corresponding to E (PG ) Bn), Figure 2) was
obtained from a D-xylose derivative via a stereoselective
Grignard reaction in a well-known and high yielding five-
step procedure.16 Three different strategies for the inver-
sion of C-2 configuration were tried, namely (i) an SN2
reaction, (ii) 1,2-anhydro formation (through F), and (iii)
an oxidation-reduction procedure. The latter strategy
gave the best results, and a material corresponding to D
was obtained. However, after successful coupling to give
C, the oxidative cleavage to give A was troublesome, and
in the end, 4 could be obtained in a maximum of only
2.7% overall yield from 5. All aspects of our efforts toward
4 through Route A are covered in the electronic Support-
ing Information.
Synthesis of 4 by Route B. After the efforts following
route A, it became clear that the major bottleneck here
(even more problematic than it was in the original
synthesis of the â-analogue 316), was the oxidative
cleavage of the vinyl group to a hydroxymethyl group
using a two-step OsO4-NaIO4/NaBH4 protocol.25 Hence,
we decided to perform the problematic oxidative cleavage
step in the beginning of the reaction sequence with the
conveniently obtained compound 516 as the substrate
(Scheme 1). Our attempts toward oxidative cleavage of
the double bond of 5 using the standard osmium tetraox-
ide mediated protocols (OsO4-NaIO4/NaBH4 or OsO4-
NMO followed by NaIO4/NaBH4)25,26 as well as ozonolysis-
reduction protocols27 were unsuccessful.28 Nevertheless,
we were successful in developing a new general and
efficient ruthenium-catalyzed protocol for the oxidative
cleavage of monosubstituted alkenes to yield exclusively
(one carbon shorter) primary alcohols. Thus, using ru-
thenium tetraoxide (made in situ from RuCl3‚xH2O and
NaIO4 as a co-axidant) for the oxidation, followed by an
additional sodium borohydride reduction step before the
diol-cleavage and reduction successfully afforded the
conversion of 5 to the hydroxymethyl product 6 in 72%
yield (Scheme 1). This reaction also paved the way for a
highly improved synthesis of the known â-configured
bicyclic nucleoside 3, and we have recently published the
general ruthenium protocol and the significantly im-
proved synthesis of 3 in a note.19
Having achieved the desired goal of efficient conversion
of 5 to 6, we attempted the synthesis of the target
nucleoside 4 as depicted in Scheme 1. We decided to
mesylate the hydroxy group of 6 considering our need
for a leaving group for the creation of an oxetane ring as
shown in Figure 2. Based on previous experiences, we
hoped that the mesylate group would be stable in the
conditions needed for inversion of C-2 configuration and
for the coupling of the nucleobase. Thus, in the prepara-
tion of R-L-LNA monomers, mesylates on a 5-O position
(26) VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron Lett. 1976,
17, 1973-1976.
(27) (a) Sousa, J. A.; Bluhm, A. L. J. Org. Chem. 1960, 25, 108-
111. (b) Flippin, L. A.; Gallagher, D. W.; Araghi, K. J. J. Org. Chem.
1989, 54, 1430-1432.
(28) After our project was completed, an improved procedure for the
oxidative cleavage of olefins by OsO4-NaIO4 was published. However,
we attempted this procedure and found it unsuccessful on our
substrate. Yu, W.; Mei, Y.; Hua, Z.; Jin, Z. Org. Lett. 2004, 6, 3217-
3219.
(25) Pappo, R.; Allen, D. S.; Lemieux, R. U.; Johnson, W. S. J. Org.
Chem. 1956, 21, 478-479.
J. Org. Chem, Vol. 70, No. 13, 2005 4921

Sharma et al.
and a 4-C-hydroxymethyl position have been found to be
stable under the conditions needed for an SN2 inversion
of C-2 configuration by the use of a 2-O-trifluoromethane-
sulfonate.12,29 Accordingly 6 was mesylated in pyridine
to give monomesylate 7 in a high yield. Treatment of 7
with methanolic hydrochloric acid afforded the anomeric
mixture (â/R ) 2.3:1) of methyl furanosides 8 in 66%
yield. The inversion of configuration at C-2 was then
attempted by conversion of 8 into the corresponding 2-O-
trifluoromethanesulfonate and subsequent reaction with
cesium acetate in DMF in the presence of 18-crown-6.
This, however, resulted in a complex mixture of com-
pounds which was thought to be a mixture of arabino-
configured diacetate 9 and an unidentified compound.
Thus, under these reaction conditions, the mesylate
group participated in an SN2 reaction with cesium
acetate. Efforts to further separate the mixture by
column chromatography were not successful and coupling
of this mixture with thymine using the Vorbru¨ggen
method21 was performed to give an inseparable mixture
of R-nucleoside 10 and an unidentified nucleoside. Deacety-
lation of the diacetate 10 to give 11 along with an
unidentified product, followed by selective mesylation of
the primary hydroxyl group gave a complex mixture.
Fortunately, we were able to separate this complex
mixture by column chromatography and obtained pure
required monomesylate 12 in 10% overall yield from 8.
Ring closure by the use of sodium hydride gave 13 and
debenzylation afforded the target bicyclic nucleoside 4
in 96% yield over the two steps. Thus, we were successful
in completing the synthesis of target nucleoside 4 but the
reaction sequence was marred by the formation of
unexpected products presumably due to the use of mesyl
group to protect (and later activate) the primary hydroxyl
group resulting in a very low overall yield of 2.9% of 4
from 5. As the mesyl group got replaced by an acetyl
group during the inversion step, we decided to protect
the primary hydroxyl group of 6 by an acetyl group. Thus,
6 was converted to monoacetate 14 in 98% yield and
subsequent treatment of 14 with methanolic hydrochloric
acid resulted in required methanolysis to yield the methyl
ribofuranoside. However, the acetyl group could not
survive the reaction conditions and we got deacetylated
methyl ribofuranosides 15 along with unidentified side
products presumably due to alternative furanose forma-
tion. Hence, it became clear that neither the acetyl was
the right protecting group for further elaborations and
this route was therefore not pursued further.
into considation that R-trisubstituted secondary alcohols
(a neopentyl-type system) are among the poorest candi-
dates for an SN2 inversion process due to the steric
hindrance and the competing SN1 reaction, this was a
very satisfying result. In our first strategy (route A,
Figure 2), low yields were obtained with the 3-vinyl
analogue of 17 (â/R ) 2.5:1) as the substrate for this
process, as only the R-anomer participated in the inver-
sion reaction (see the Supporting Information). Hence,
the yield at the inversion step improved considerably by
replacing the vinyl group with the benzoyloxymethyl
group, and moreover, 18 was found to be an anomeric
mixture approximately in an R:â ratio of 1.1:1 indicating
clearly that both the anomers of 17 responded to the
inversion conditions. Coupling of 18 with thymine using
the Vorbru¨ggen method21 yielded the R-D-arabino-con-
figured nucleoside 19 in 68% yield. Deacetylation and
debenzoylation was performed in a single step using
sodium methoxide to give the diol 11 in 90% yield. The
selective mesylation of 11 to give 12, oxetane ring
formation to give 13 and debenzylation afforded the
target bicyclic nucleoside 4 in 81% yield over the three
steps. The overall yield of 4 from 5 by this final strategy
was 16.3% as compared to overall yields of 1.3-2.7%
(route A, see the Supporting Information) and 2.9% (by
using a mesyl group instead of a benzoyl group in route
B).
Conformation of Nucleoside 4. Nucleoside confor-
mations are conveniently defined by the pseudoration
angle.7,31 Thus, the N-type conformation corresponds to
a pseudoration angle P ≈ 18°, the S-type to P ≈ 162°,
whereas E- and W-type (O4′-endo and O4′-exo) conforma-
tions correspond to P ≈ 90° and 270°, respectively.7,31
A
pseudorotation energy profile for 4 was calculated by ab
initio calculations at the MP2/cc-pVTZ level (see Figure
3). Two local energy minima were found in the pseudoro-
tation profile for this nucleoside (P ≈ 110° and 270°)
corresponding to E- and W-type conformations, respec-
tively, and the minimum at P ≈ 110° is 6.0 kcal/mol lower
in energy than the P ≈ 270° minimum. Using a general-
ized Karplus equation,32-34 we found JH1′-H2′ ) 0.6 and
3.9 Hz for the two low energy sugar puckers, respectively.
1
Experimentally, H-1′ displays a singlet in the H NMR
spectrum of 4 corresponding to a JH1′-H2′ < 1.0 Hz, and
this shows that 4 exclusively adopts an E-type furanose
conformation in solution without repuckering to the
W-conformation. The pseudorotation profile is character-
ized by high energy barriers rendering pseudorotation
impossible. The conformation with a planar furanose is
4.9 kcal/mol higher in energy than the P ≈ 110° geom-
etry. An unconstrained geometry optimization (MP2/6-
31G*) gave a structure with P ) 97° and a puckering
amplitude Φmax ) 33° (see Figure 3, structure inserted).
Therefore, it is confirmed that the bicyclic nucleoside 4
Careful perusal of the reaction sequence revealed that
a benzoyl group should be able to survive the acidic
methanolysis as well as the inversion conditions. Accord-
ingly, benzoylation of hydroxyl group of 6 to give 16
followed by acidic methanolysis afforded the anomeric
mixture of methyl furanosides 17. NMR of the crude 17
revealed the anomeric mixture to be approximately in a
â/R ratio of 2.7:1. Inversion of C-2 configuration was again
attempted in a two step procedure converting the 2-hy-
droxyl group to a strong leaving group followed by an
SN2 reaction by cesium acetate taking advantage of the
“cesium effect”.30 Hereby, we got the desired arabino-
furanoside 18 in 46% yield over four steps from 6. Taking
(30) (a) Kijkstra, G.; Kruizinga, W. H.; Kellogg, R. M. J. Org. Chem.
1987, 4230-4234. (b) Hawryluk, N. A.; Snider, B. B. J. Org. Chem.
2000, 65, 8379-8380.
(31) Altona, C.; Sundaralingam, M. J. Am. Chem. Soc. 1972, 94,
8205.
(32) Donders, L. A.; de Leeuw, F. A. A. M.; Altona, C. Magn. Reson.
Chem. 1989, 27, 556.
(33) Altona, C.; Ippel, J. H.; Westra Hoekzema, A. J. A.; Erkelens,
C.; Groesbeek, M.; Donders, L. A. Magn. Reson. Chem. 1989, 27, 564.
(34) Altona, C.; Francke, R.; de Haan, R.; Ippel, J. H.; Daalmans,
G. J.; Westra Hoekzema, A. J. A.; van Wijk, J. Magn. Reson. Chem.
1994, 32, 670.
(29) Raunak; Babu, B. R.; Sørensen, M. D.; Parmar, V. S.; Harrit,
N. H.; Wengel, L. Org. Biomol. Chem. 2004, 2, 80-89.
4922 J. Org. Chem., Vol. 70, No. 13, 2005

R-D-Configured Bicyclic Nucleosides
TABLE 1. Hybridization Data of Modified and
Unmodified Oligothymidylate r-DNA Sequences
DNA (dA14
sequencesa Tm (∆Tm)/°Cb
22 5′-RT14 32.0
)
RNA (rA14
Tm (∆Tm)/°Cb
43.0
)
ODN
23 5′-RT7XT6
24 5′-RT5X4T5
25 5′-RT10
30.5 (-1.5)c (-6.5)e 46.0 (+3.0)c (-8.0)e
24.0 (-2.0)c ( -1.5)e 33.0 (-2.5)c (-4.6)e
18.0
<10
33.5
26 5′-RX10
22.0 (-2.1)d (+1.2)e
a R-DNA sequences with X ) 4. b Melting temperatures obtained
from the maxima of the first derivatives of the melting curve (A260
vs temperature) recorded in a buffer containing 10 mM Na2HPO4,
100 mM NaCl, 0.1 mM EDTA, pH 7.0 using 1.0 µM concentrations
of each strand. Values in parentheses show the changes in Tm
values per modification compared with the reference strands.
c Compared with 22. d Compared with 25. e These values refer to
the corresponding changes in Tm values per modification obtained
with R-LNA, i.e., X ) 2.10
DNA and RNA sequences were studied by thermal
denaturation examination (Tables 1 and 2). The oligo-
thymidylate sequences 22-26 were mixed with comple-
mentary dA14 and rA14 and melting temperatures of the
complexes were determined. The affinity of a 14-mer
R-DNA sequence 22 toward DNA was slightly decreased
by the incorporation of the bicyclic monomer as in 23 and
24, for which a drop in melting temperature of 1.5-2.0
°C for each modification was obtained. A fully modified
decamer R-DNA sequence 26 cannot recognize comple-
mentary DNA. With complementary RNA the situation
is different. Thus, with one bicyclic monomer incorpo-
rated in an otherwise unmodified R-DNA sequence, 23,
the affinity increases significantly with 3 °C. However,
a fully modified sequence, 26, displays a decreased
affinity toward RNA, and the sequence with four bicyclic
entities in the middle of a 14-mer R-DNA sequence, 24,
displays a similar decrease in affinity.
A mixed pyrimidine sequence allows a verification of
a parallel recognition mode. Hence, the unmodified
R-DNA sequence 27 displays the earlier established8,10
high affinity for complementary parallel DNA and RNA.
One bicyclic moiety, 28, leads to a small decrease in the
affinity for the parallel DNA sequence whereas three or
six bicyclic moieties, 29 or 30, result in a more pro-
nounced destabilization. On the other hand, when the
same sequences were mixed with complementary RNA,
a stabilization of the parallel duplex is seen with one or
six bicyclic moieties, 28 or 30, but peculiarly, a small
destabilization with three bicyclic moieties, 29. All mixed
pyrimidine sequences 27-30 were also mixed with
complementary antiparallel DNA and RNA sequences,
and as expected,8,10 no melting temperatures above the
detection limit of 10 °C were observed.
FIGURE 3. MP2/cc-pVTZ potential energies as a function of
pseudorotation angle P for 4. All energies are given relative
to the P ) 110° conformation. Inset: the geometry-optimized
(MP2/6-31G*) structure displaying an E-type conformation
with P ) 97°.
like its â-anomer 316 is strongly conformationally re-
stricted in an E-type (O4′-endo) conformation and only
samples conformations near this global energy minimum.
Preparation of Oligonucleotides and Thermal
Denaturation Studies. To incorporate the [3.2.0]bicy-
clic nucleoside monomer into oligonucleotides, 4 was
converted to the appropriately protected phosphoramidite
derivative (Scheme 1). Thus, the primary alcohol function
of 4 was easily protected with the DMT (4,4′-dimethox-
ytrityl) group to afford 20, which was eventually trans-
formed into the corresponding 3′-O-phosphoramidite
building block 21. For the automated solid-phase syn-
thesis of oligonucleotides by the phosphoramidite ap-
proach,35 the bicyclic R-nucleoside monomeric building
block 21 was used in combination with the corresponding
unmodified R-2′-deoxynucleoside thymine and 5-methyl-
cytosine phosphoramidites, which were obtained accord-
ing to literature methods.36,37 Hence, modified and un-
modified R-DNA sequences (22-30, Tables 1 and 2) were
obtained with >98% stepwise coupling yields by using
tetrazole activation and coupling times of 10-15 min for
21. All sequences were obtained on universal CPG
support (Biogenex/Glen Research) using the DMT-ON
mode. This allowed synthesis of fully R-configurated
sequences after cleavage from the solid support using
LiCl in aqueous ammonia. The oligomers 22-30 were
purified by using reversed-phase HPLC yielding products
with >90% purity. The compositions of all sequences were
verified from their MALDI mass spectra.
Tables 1 and 2 display also the corresponding changes
in melting temperature obtained with R-LNA monomers
(i.e., 2) as the synthetic monomer in R-DNA instead of
the [3.2.0]bicyclic monomer 4 in similar sequences. As
evident from a direct comparison, the [3.2.0]bicyclic
moiety leads to a less pronounced destabilization of the
parallel R-DNA:DNA duplexes than does the R-LNA
monomer. The effect of the [3.2.0]bicyclic moieties on
R-DNA/RNA duplexes is even more favorable. Thus, the
bicyclic monomer 4 increases the RNA-affinity though,
apparently, dependent on sequence composition. In other
words, the cooperativity between 4 and the R-DNA
For the synthetic convenience and the direct compari-
son with the results obtained for R-LNA (monomer 2),
only 10-mer and 14-mer oligothymidylate sequences
(Table 1) and a mixed 10-mer pyrimidine sequence (Table
2) were employed for this study. The hybridization
between the modified R-DNA sequences and unmodified
(35) Caruthers, M. H. Acc. Chem. Res. 1991, 24, 278-284.
(36) Chassignol, M.; Thuong, N. T. C. R. Acad. Sci. Paris Ser. II
1987, 305, 1527-1530.
(37) Kurfu¨rst, R.; Roig, V.; Chassignol, M.; Asseline, U.; Thuong,
N. T. Tetrahedron 1993, 49, 6975-6990.
J. Org. Chem, Vol. 70, No. 13, 2005 4923

Sharma et al.
TABLE 2. Hybridization Data of Modified and Unmodified Mixed Pyrimidine r-DNA Sequences
DNA (p) dGAGGAAGAAA RNA (p) rGAGGAAGAAA
Tm (∆Tm)/°Cb
ODN sequencesa
Tm (∆Tm)/°Cb
27
28
29
30
5′-R-mCTmCmCTTmCTTT
5′-R-mCTmCmCTXmCTTT
5′-R-mCXmCmCXTmCXTT
5′-R-mCXmCmCXXmCXXX
43.0
32.0
42.0 (-1.0)
34.0 (-3.0)
19.0 (-4.0)
(-12.5)c
(<-5.5)c
35.0 (+3.0)
30.0 (-0.7)
37.0 (+0.8)
(-6.0)c
(-1.7)c
a R-DNA sequences with X ) 4. b See Table 1. Values in parentheses show the changes in Tm values per modification compared with
the reference strand 27. c These values refer to the corresponding changes in Tm values per modification obtained with R-LNA, i.e., X )
2.10
monomers is functional and mixmers with high RNA-
affinity can be obtained, whereas mixmers between
R-LNA and R-DNA monomers let to largely destabilized
duplexes.10 On the other hand, the fully modified se-
quence of 10 [3.2.0]bicyclic monomers 26 has a low RNA-
affinity, whereas the corresponding R-LNA sequence
displayed an RNA-affinity much higher than the corre-
sponding R-DNA,10 as evident also in mixed sequences.10,11
The present hybridization data of modified R-DNA
sequences can lead to some conclusions about the status
of nucleic acid recognition by parallel duplex formation.
Thus, with 4 as the first E-type conformational R-DNA
mimic, some immediate comparisons between different
conformations in R-DNA can be performed. From our
former studies with R-LNA it is evident that the confor-
mational behavior of the nucleosides in R-DNA and in
DNA is very different.10,11 For standard â-configured
oligonucleotides and antiparallel duplexes, the strong
N-type mimic LNA always leads to duplex stabilization
also in mixmers.3,5,6 In other words, the LNA monomers
can force neighboring 2′-deoxynucleotides into N-type
conformations and the overall duplex toward an A-type
duplex.5,6 This was absolutely not possible in parallel
duplexes of R-DNA/DNA or R-DNA/RNA. In mixmers
with R-DNA, the strong N-type mimic R-LNA always
leads to significant duplex destabilization.10 On the other
hand, strong parallel recognition was found with a
complete R-LNA sequence, however, in the formation of
a very different extended duplex structure.40 From the
present study, it can be concluded that the E-type
conformation is compatible with R-DNA nucleosides in
mixmers. Thus, the incorporation of 4 into otherwise
unmodified R-DNA sequences leads to stabilization of the
R-DNA/RNA duplex in most cases. On the other hand,
the corresponding R-DNA/DNA parallel duplex is desta-
bilized with the E-type mimic, and fully modified du-
plexes (sequence 26 and complements) are in general not
stable. Also, this result is opposed to the corresponding
results with the â-anomeric E-type mimic 3 incorporated
in standard DNA/DNA or DNA/RNA duplexes, where
stabilizations were generally observed with both mixmers
of 3 and 2′-deoxynucleotides and with fully modified
sequences.16,18 Finally, conformationally restricted or
even locked S-type mimics have not given the same
stabilization of the standard â-configured and antipar-
allel duplexes as it might have been expected. However,
all bi- or tricyclic â-nucleoside S-type mimics studied41,42
suffers from other problems, including sterical problems
with standard (B-type) duplex geometry.42,43 The same
Discussion
Several different approaches were examined toward
the synthesis of the target [3.2.0]bicyclic R-nucleoside 4.
In the end, the route B (see Figure 2 and Scheme 1)
starting from E (i.e., 5, originally obtained from D-
xylose)16 through intermediates H, G and A afforded 4
in a high 16% overall yield. All investigated strategies
applied the vinyl group as a strategically “protected”
hydroxymethyl group. In the end, the oxidative cleavage
of the vinyl group was the major bottleneck for the
successful preparation of 4, but this key point was
successfully passed by our development of a ruthenium-
based four step reaction sequence.19 It is our expectation
that this sequence being generally applicable for the
conversion of terminal alkenes to one carbon shorter
primary alcohols can be a general tool and a good
alternative to existing methods in other aspects of
synthetic organic chemistry. The limiting step of our
synthesis of 4 was hereafter the inversion of C-2-
configuration relying on a two-step procedure giving the
arabino-furanoside 18 from the corresponding ribo-fura-
noside 17. On the other hand, this is by far the most
convenient of the inversion reactions investigated in the
present study. Thus, the arabino-furanoside 18 was
formed as the only product and the handling of stereoi-
someric mixtures of nucleosides (as in route A; see the
Supporting Information) was thereafter avoided. Fur-
thermore, this synthetic strategy is not based on the
pyrimidine-mediated conversion of C-2′ configuration
following the so-called anhydro approach38 that has been
used in many occasions for the preparation of arabino-
configured nucleoside derivatives with pyrimidines as the
nucleobase39 including several bicyclic nucleosides.12,15,16,19
That strategy leaves behind the preparation of the purine
analogues needing tedious conversions of C-2′-configu-
ration as the one present herein. In our strategy, there-
fore, the purine analogues could be expected to be formed
by more or less equal efficiency, and compound 18 is
therefore a key compound for the preparation of all
analogues of 4 with different nucleobases.
(40) Christensen, N. K.; Petersen, M.; Vester, B.; Nielsen, P.
Nucleosides, Nucleotides & Nucleic Acids 2003, 22, 1143-1145.
(41) (a) Tarko¨y, M.; Bolli, M.; Schweizer, B.; Leumann, C. Helv.
Chim. Acta 1993, 76, 481-510. (b) Renneberg, D.; Leumann, C. J. J.
Am. Chem. Soc. 2002, 124, 5993-6002.
(42) (a) Thomasen, H.; Meldgaard, M.; Freitag, M.; Petersen, M.;
Wengel, J.; Nielsen, P. Chem. Commun. 2002, 1888-1889. (b) Freitag,
M.; Thomasen, H.; Christensen, N. K.; Petersen, M.; Nielsen, P.
Tetrahedron 2004, 60, 3775-3786.
(43) (a) Altmann, K.-H.; Imwinkelried, R.; Kesselring, R.; Rihs, G.
Tetrahedron Lett. 1994, 35, 7625-7628. (b) Ravn, J.; Thorup, N.;
Nielsen, P. J. Chem. Soc., Perkin Trans. 1 2001, 1855-1861. (c)
Kværnø, L.; Wightman, R. H.; Wengel, J. J. Org. Chem. 2001, 66,
5106-5112.
(38) Fox, J. J.; Miller, N. C. J. Org. Chem. 1963, 28, 936-941.
(39) Schinazi, R. F.; Chen, M. S.; Prusoff, W. H. J. Med. Chem. 1979,
22, 1273-1277.
4924 J. Org. Chem., Vol. 70, No. 13, 2005

R-D-Configured Bicyclic Nucleosides
recorded at 200 or 300 MHz for 1H NMR, 75 MHz for 13C NMR,
and 121.49 MHz for 31P NMR. The δ values are in ppm relative
to tetramethylsilane as internal standard (for 1H and 13C
NMR) and relative to 85% H3PO3 as external standard (for
31P NMR). Assignments of NMR spectra when given are based
on 2D spectra and follow standard carbohydrate and nucleo-
side style; i.e., the carbon atom next to a nucleobase is assigned
C-1′, etc. In nucleosides, C-1′′ designates the carbon atom in
the 3′-C branch; otherwise, C-1′ designates the carbon atom
in the 3-C branch. Compound names given in this section for
the bicyclic compounds are given according to the von Baeyer
nomenclature. HRMALDI and EI mass spectra were recorded
in positive-ion mode except for the oligonucleotides where these
were recorded in negative-ion mode.
3,5-Di-O-benzyl-1,2-di-O-isopropylidene-3-C-(methane-
sulfonyloxymethyl)-r-D-ribofuranose (7). A solution of 619
(113 mg, 0.28 mmol) in anhydrous pyridine (5 mL) was cooled
to 0 °C. Methanesulfonyl chloride (44 µL, 0.56 mmol) was
added dropwise at 0 °C, and the mixture was stirred at room
temperature for 2 h. The reaction mixture was quenched by
adding ice-water (20 mL), and the resulting mixture was
extracted with CH2Cl2 (3 × 15 mL). The combined organic
phase was washed with saturated aqueous NaHCO3, dried
(Na2SO4), and concentrated under reduced pressure to yield 7
(127 mg, 94%) as a clear oil which was used without further
purification in the next step: Rf 0.65 (19:1 CH2Cl2/CH3OH);
1H NMR (300 MHz, CDCl3) δ 7.37-7.24 (m, 10H, Ph), 5.86 (d,
1H, JH1-H2 ) 3.6 Hz, H-1), 4.84-4.72 (m, 3H), 4.58-4.53 (m,
2H), 4.41 (s, 2H), 4.34 (m, 1H, H-4), 3.76 (m, 1H, H-5a), 3.66
(m, 1H, H-5b), 2.84 (s, 3H, SO2CH3), 1.61 (s, 3H, CH3), 1.38
(s, 3H, CH3); 13C NMR (75 MHz, CDCl3) δ 138.4, 137.8, 128.5,
128.4, 128.0, 127.9, 127.8, 127.7 (Ph), 113.2 (C(CH3)2), 104.1
(C-1), 83.7, 79.6, 78.7, 73.8, 69.6, 68.9, 66.9 (C-3, C-2, C-4, 2 ×
CH2Ph, C-5, C-1′), 37.3 (SO2CH3), 26.7 (CH3), 26.7 (CH3);
HRMALDI MS m/z 501.1556 ([M + Na]+, C24H30O8SNa+ calcd
501.1553).
might be evident for the few R-configured S-type mimics
studied so far, i.e., the R-D-arabino-2′-O-methylthymi-
dine,44 a [4.3.0]bicyclic nucleoside45 as well as the bi-
cyclic46 and tricyclic nucleoside analogues14 from Leu-
mann and co-workers. In the latter example, however,
this was met by a strong restriction of not only the sugar
conformation but also of the torsion angle γ.14 Neverthe-
less, and despite the obvious need for experimental
evidence, one might argue that the gain of a strong
conformational restriction of R-DNA into an S-type
conformation could be relatively small, as the unmodified
R-DNA monomers are already strongly restricted toward
this conformation by the combined contributions from
anomeric and gauche stereoelectronic effects.47,48
Finally, the parallel nucleic acid recognition found for
â-L-LNA is interesting. Though not displaying S-type
sugar conformations, this LNA-stereoisomer is certainly
an R-DNA and not an R-RNA mimic.11 Therefore, mixmers
of â-L-LNA and R-DNA, perhaps even combined with our
E-type mimic 4, might be favorable for parallel duplex
formation and fine-tuning of duplex stability. The com-
bination of several modifications, for instance the bicyclic
E-type mimic present herein and the positively charged
phosphoramidate R-DNA,9 might increase the scopes for
parallel nucleic acid recognition to an even higher level,
and might lead to therapeutically interesting oligonucle-
otides. On the other hand, regarding the low DNA and
RNA affinity found for sequence 26, it seems unlikely
that the thermal stability of parallel nucleic acid duplexes
can surpass the level found for fully modified R-LNA10,11
and â-L-LNA sequences.11 Thereby, parallel duplexes
might not ever reach the same level of thermal stability
as antiparallel duplexes.
Methyl
3,5-Di-O-benzyl-3-C-(methanesulfonyloxy-
methyl)-D-ribofuranoside (8). Anhydrous CH3OH (4 mL)
was stirred at -30 °C, and acetyl chloride (0.68 mL) was added
slowly. The mixture was stirred at -30 °C for 30 min and
warmed to room temperature. A solution of 7 (204 mg, 0.43
mmol) in CH2Cl2/CH3OH/H2O (2:2:1 (v/v/v), 5 mL) was stirred
at 0 °C, CH3OH was added dropwise. The mixture was stirred
at room temperature for 90 h, and water (5 mL) was added.
The mixture was neutralized with saturated aqueous NaHCO3
and extracted with CH2Cl2 (3 × 5 mL). The combined organic
extracts were dried (Na2SO4) and evaporated to dryness under
reduced pressure. The residue was purified by silica gel column
chromatography (eluent: gradient of CH3OH (0-2%) in CH2-
Cl2) to give 8 (127 mg, 66%, R/â ) 1:2.3) as a colorless viscous
oil: Rf 0.50 (19:1 CH2Cl2/CH3OH); 1H NMR (300 MHz, CDCl3)
δ 7.36-7.25 (m, Ph), 5.00-4.44 (m), 4.13 (s), 3.7-3.59 (m), 3.49
(s, OCH3R), 3.39 (s, OCH3â), 2.93 (s, SO2CH3â), 2.90 (s, SO2-
CH3R); 13C NMR (300 MHz, CDCl3) δ 138.4, 137.5, 137.5, 137.3,
128.8, 128.7, 128.6, 128.5, 128.3, 128.2, 128.1, 128.0, 127.9,
127.8, 127.7, 127.3, 109.0, 102.0, 82.7, 81.4, 80.5, 80.0, 77.9,
76.5, 75.2, 74.0, 73.9, 72.9, 69.9, 69.0, 68.8, 68.4, 67.3, 67.2,
66.9, 56.0, 55.8, 37.5, 37.5; HRMALDI MS m/z 475.1405 ([M
+ Na]+, C22H28O8SNa+ calcd 475.1397).
3-C-Acetoxymethyl-3,5-di-O-benzyl-1,2-di-O-iso-
propylidene-r-D-ribofuranose (14). To a stirred solution of
6 (579 mg, 1.45 mmol) in anhydrous pyridine (5 mL) was added
acetic anhydride (0.68 mL, 7.24 mmol), and the reaction
mixture was stirred for 2 h at room temperature. The reaction
was quenched with ice-cold water (20 mL) and extracted with
CH2Cl2 (2 × 25 mL). The combined organic phase was washed
with saturated aqueous sodium hydrogen carbonate (3 × 25
mL), dried (Na2SO4), and concentrated under reduced pressure
to give 14 (628 mg, 98%) as an oily residue which was used
without further purification in the next step: Rf 0.70 (19:1 CH2-
Cl2/CH3OH); 1H NMR (300 MHz, CDCl3) δ 7.37-7.24 (m, 10H,
Ph), 5.83 (d, 1H, JH1-H2 ) 3.9 Hz, H-1), 4.77, 4.65 (AB, 2H,
Conclusion
A bicyclic R-D-configured nucleoside has been efficiently
synthesized and proved to be strongly restricted in an
E-type conformation. By incorporation of this into R-D-
configured oligodeoxynucleotides, the scopes of parallel
nucleic acid duplex formation has been illuminated. The
ability of R-DNA to form strong parallel duplexes with
complementary DNA seems to be closely related to its
strong tendency to adopt S-type conformations. On the
other hand, we have demonstrated that the parallel
duplex formed between R-DNA and complementary RNA
can be slightly stabilized by mixmers with a conforma-
tionally restricted E-type mimic dependent on sequence
constitution.
Experimental Section
General Methods. Reactions were performed under an
atmosphere of nitrogen when anhydrous solvents were used.
Column chromatography was carried out on glass columns
using silica gel 60 (0.040-0.063 mm). NMR spectra were
(44) Gotfredsen, C. H.; Jacobsen, J. P.; Wengel, J. Bioorg. Med.
Chem. 1996, 4, 1217-1225.
(45) Zou, R.; Matteucci, M. Bioorg. Med. Chem. Lett. 1998, 8, 3049-
3052.
(46) Bolli, M.; Lubini, P.; Leumann, C. Helv. Chim. Acta 1995, 78,
2077-2096.
(47) Aramini, J. M.; Mujeeb, A.; Germann, M. W. Nucleic Acids Res.
1998, 26, 5644.
(48) Thibaudeau, C.; Fo¨ldesi, A.; Chattopadhyaya, J. Tetrahedron
1998, 54, 1867-1900.
J. Org. Chem, Vol. 70, No. 13, 2005 4925

Sharma et al.
Jgem ) 10.8 Hz, 3-OCH2Ph), 4.61-4.51 (m, 3H, H-2, 5-OCH2-
Ph), 4.39 (m, 1H, H-4), 4.35, 4.18 (AB, 2H, Jgem ) 12.6 Hz,
H-1′), 3.75 (m, 1H, H-5a), 3.60 (m, 1H, H-5b), 2.04 (s, 3H,
COCH3), 1.61 (s, 3H, CH3), 1.37 (s, 3H, CH3); 13C NMR (75
MHz, CDCl3) δ 170.5 (CO), 138.6, 138.0, 128.4, 128.3, 127.9,
127.8, 127.7, 127.7, 127.6 (Ph), 113.0 (C(CH3)2), 104.3 (C-1),
83.4 (C-3), 80.4 (C-4), 79.3 (C-2), 73.7 (5-OCH2Ph), 68.2 (3-
OCH2Ph), 67.8 (C-5), 63.1 (C-1′), 26.9 (CH3), 26.8 (CH3), 21.0
(COCH3); HRMALDI MS m/z 465.1868 ([M + Na]+, C25H30O7-
Na+ calcd 465.1883).
Methyl 3,5-Di-O-benzyl-3-C-(hydroxymethyl)-D-ribo-
furanoside (15). The same procedure as in the preparation
of 8 was used with the ribofuranose 14 (588 mg, 1.33 mmol)
affording a mixture of 15 and an unidentified material (471
mg): Rf 0.35 (19:1 CH2Cl2/CH3OH); HRMALDI MS m/z
397.1604 ([M + Na]+, C21H26O6Na+ calcd 397.1621).
(m), 3.44 (s, OCH3 R), 3.40 (s, OCH3 â), 2.03 (s, COCH3 â),
2.01 (s, COCH3 R); 13C NMR (75 MHz, CDCl3) δ 169.6, 169.4,
166.2, 166.2, 138.1, 137.9, 137.8, 137.8, 133.4, 133.3, 130.0,
130.0, 129.9, 129.9, 129.8, 129.7, 128.7, 128.6, 128.6, 128.5,
128.5, 128.4, 128.3, 128.0, 128.0, 127.9, 127.9, 127.9, 127.8,
127.6, 127.5, 127.5, 107.2 (C-1 R), 102.1 (C-1 â), 85.3, 84.2,
82.6, 81.3, 81.3, 73.7, 69.9, 68.7, 66.84, 66.2, 60.7, 60.1, 56.6,
55.4, 20.9, 20.8; HRMALDI MS m/z 543.1973 ([M + Na]+,
C30H32O8Na+ calcd 543.1989).
N1-(2-O-Acetyl-3-C-(benzoyloxymethyl)-3,5-di-O-benzyl-
r-D-arabinofuranosyl)thymine (19). To a stirred solution
of 18 (470 mg, 0.90 mmol) and thymine (228 mg, 1.81 mmol)
in anhydrous CH3CN (15 mL) was added N,O-bis(trimethyl-
silyl)acetamide (1.8 mL, 7.23 mmol). The reaction mixture was
stirred at reflux for 30 min. After cooling to 0 °C, trimethylsilyl
triflate (0.82 mL, 4.52 mmol) was added dropwise, and the
solution was stirred for 24 h at 70 °C. After cooling to room
temperature, the reaction mixture was diluted with ethyl
acetate (50 mL) and washed with saturated aqueous NaHCO3
solution (3 × 30 mL). The combined organic phase was dried
(Na2SO4) and evaporated under reduced pressure. The residue
was purified by silica gel column chromatography (eluent:
gradient of CH3OH (0-3.0%) in CH2Cl2) to give nucleoside 19
(380 mg, 68%) as a white foam: Rf 0.50 (9:1 CH2Cl2/CH3OH);
1H NMR (300 MHz, CDCl3) δ 8.69 (s, 1H, NH), 8.01-7.98 (m,
Methyl 2-O-Acetyl-3-C-(benzoyloxymethyl)-3,5-di-O-
benzyl-D-arabinofuranoside (18). To a stirred solution of
6 (1.661 g, 4.15 mmol) in anhydrous pyridine (6 mL) at 0 °C
was added benzoyl chloride (1.204 mL, 10.38 mmol). The
reaction mixture was allowed to warm to room temperature
and stirred for 2 h. The mixture was concentrated to dryness
under reduced pressure and coevaporated with toluene (3 ×
15 mL) under reduced pressure. The residue was partitioned
between ethyl acetate (15 mL) and water (9 mL). The sepa-
rated organic phase was washed with a saturated aqueous
solution of NaHCO3 (3 × 15 mL), dried (Na2SO4), and
concentrated under reduced pressure to give 16 as an oily
residue. In a separate flask, anhydrous CH3OH (33 mL) was
stirred at -30 °C, and acetyl chloride (6.6 mL) was added
slowly. The mixture was stirred at -30 °C for 30 min and
warmed to room temperature. A solution of the oily intermedi-
ate 16 in CH2Cl2/ CH3OH /H2O (2:2:1 (v/v/v), 41.5 mL) was
stirred at 0 °C, and the first solution of acetyl chloride in CH3-
OH was added dropwise. The mixture was stirred at room
temperature for 96 h, and water (41 mL) was added. The
mixture was neutralized with saturated aqueous NaHCO3 and
extracted with CH2Cl2 (3 × 25 mL). The combined organic
extracts were dried (Na2SO4) and evaporated to dryness under
reduced pressure to afford 17 as an anomeric mixture (â/R )
2H, Ph), 7.61-7.25 (m, 14H, H-6, Ph), 6.14 (d, 1H, JH1′-H2′
)
2.4 Hz, H-1′), 5.48 (d, 1H, JH1′-H2′ ) 2.4 Hz, H-2′), 4.88-4.54
(m, 7H), 3.77-3.66 (m, 2H), 1.95 (s, 3H, COCH3), 1.49 (d, 3H,
J ) 1.2 Hz, CH3); 13C NMR (75 MHz, CDCl3) δ 168.9 (COCH3),
166.1 (COPh), 163.7 (C-4), 150.4 (C-2), 137.2 (Ph), 136.5 (C-
6), 136.2, 133.6, 129.8, 129.3, 128.8, 128.7, 128.7, 128.4, 128.3,
128.3, 128.2 (Ph), 111.2 (C-5), 90.1, 86.1, 83.3, 79.8, 74.0, 68.4,
66.3, 59.4 (C-1′, C-2′, C-3′, C-4′, C-5′, C-1′′, 2 × CH2Ph), 20.6
(COCH3), 12.2 (CH3); HRMALDI MS m/z 637.2150 ([M + Na]+,
C34H34N2O9Na+ calcd 637.2156). Anal. Calcd for C34H34N2O9:
C, 66.44; H, 5.57; N, 4.56. Found: C, 66.28; H, 5.52; N, 4.59.
N1-(3,5-Di-O-benzyl-3-C-(hydroxymethyl)-r-D-arabino-
furanosyl)thymine (11). Sodium methoxide (134 mg, 2.48
mmol) was added to a solution of 19 (380 mg, 0.62 mmol) in
anhydrous CH3OH (5 mL), and the reaction mixture was
stirred for 2 h at room temperature. Excess sodium methoxide
was neutralized with dilute aqueous hydrochloric acid. The
mixture was extracted with CH2Cl2 (2 × 20 mL), and the
combined extract was washed with saturated aqueous NaH-
CO3 solution (3 × 15 mL), dried (Na2SO4), and evaporated
under reduced pressure. The residue was purified by silica gel
column chromatography (eluent: gradient of CH3OH (0-
3.25%) in CH2Cl2) to give nucleoside 11 as a white foam (260
1
2.7:1 on the basis of a crude H NMR): 1H NMR (300 MHz,
CDCl3) δ 8.02-7.99 (m), 7.59-7.21 (m), 4.99-4.90 (m), 4.83-
4.43 (m), 4.27-4.14 (m), 3.77-3.56 (m), 3.50 (s, OCH3 R), 3.41
(s, OCH3 â), 3.16-3.11 (m); 13C NMR (75 MHz, CDCl3) δ 166.2,
166.2, 138.7, 137.7, 137.6, 133.6, 133.5, 133.4, 133.3, 130.0,
129.9, 129.8, 129.8, 128.8, 128.6, 128.6, 128.6, 128.5, 128.4,
128.2, 128.2, 128.0, 127.9, 127.8, 127.7, 127.5, 127.2, 109.3,
102.1, 83.0, 81.6, 80.8, 80.4, 80.1, 78.5, 75.5, 75.3, 74.5, 73.9,
73.8, 73.5, 71.7, 69.6, 68.9, 67.0, 66.8, 64.3, 62.6, 56.0, 55.7;
HRMALDI MS m/z 501.1861 ([M + Na]+, C28H30O7Na+ calcd
501.1883). The residue of crude 17 was dissolved in a mixture
of anhydrous CH2Cl2 (45 mL) and anhydrous pyridine (9 mL).
After the mixture was cooled to -30 °C, trifluoromethane-
sulfonic anhydride (1.47 mL, 8.90 mmol) was added dropwise.
The reaction mixture was allowed to warm to 0 °C. After being
stirred for an additional 30 min, the mixture was diluted with
CH2Cl2 (100 mL), washed with a saturated aqueous solution
of NaHCO3 (3 × 125 mL), and dried (Na2SO4). After evapora-
tion of the solvents, the residue was coevaporated with
anhydrous toluene (2 × 50 mL) and dissolved in anhydrous
DMF (24 mL). CsOAc (4.23 g, 22.1 mmol) and 18-crown-6
(2.120 g, 8.02 mmol) were added, and the mixture was stirred
for 30 min at room temperature and then heated to 95 °C for
2 h. After cooling to room temperature the mixture was diluted
with ethyl acetate (150 mL) and washed with water (2 × 100
mL) and brine (2 × 50 mL), dried (Na2SO4) and evaporated to
dryness under reduced pressure. The residue was purified by
silica gel column chromatography (eluent: gradient of CH3-
OH (0-0.1%) in CH2Cl2) to give 18 as a white foam (973 mg,
46% over four steps, R/â ) 1.1:1): 1H NMR (300 MHz, CDCl3)
δ 7.98-7.93 (m), 7.60-7.15 (m), 5.56 (s), 5.30 (d, J ) 4.8 Hz),
5.21 (d, J ) 4.8 Hz), 4.92-4.87 (m), 4.77-4.38 (m), 3.87-3.62
1
mg, 90%): Rf 0.40 (9:1 CH2Cl2/CH3OH); H NMR (300 MHz,
CDCl3) δ 10.14 (s, 1H, NH), 7.40-7.16 (m, 11H, H-6, Ph), 5.86
(s, 1H, H-1′), 5.29 (d, 1H, JH2′-OH ) 6.3 Hz, OH), 4.69-4.42
(m, 6H), 4.16-4.11 (m, 2H), 3.88 (m, 1H, H-5′a), 3.74 (m, 1H,
H-5′b), 3.00 (m, 1H, OH), 1.51 (d, 3H, J ) 1.2 Hz, CH3); 13C
NMR (75 MHz, CDCl3) δ 164.5 (C-4), 151.0 (C-2), 137.1, 136.9,
136.6, 128.8, 128.7, 128.4, 128.2, 128.1 (C-6, Ph), 109.8 (C-5),
94.4, 88.0, 86.6, 79.3, 74.2, 69.7, 66.0, 58.3 (C-1′, C-2′, C-3′,
C-4′, C-5′, C-1′′, 2 × CH2Ph), 12.3 (CH3); HRMALDI MS m/z
491.1766 ([M + Na]+, C25H28N2O7Na+ calcd 491.1788). Anal.
Calcd for C25H28N2O7‚1/4H2O: C, 63.48; H, 6.07; N, 5.92.
Found: C, 63.35; H, 5.94; N, 5.86
N1-(3,5-Di-O-benzyl-3-C-(methanesulfonyloxymethyl)-
r-D-arabinofuranosyl)thymine (12). A solution of 11 (373
mg, 0.82 mmol) in anhydrous pyridine (3 mL) was cooled to
-40 °C. Methanesulfonyl chloride (64 µL, 0.82 mmol) was
added dropwise, and the mixture was stirred at -40 °C for 2
h and then at room temperature for 1 h. The reaction mixture
was cooled to 0 °C and water (25 mL) was added followed by
a saturated aqueous solution of NaHCO3 (25 mL). The result-
ing mixture was extracted with CH2Cl2 (3 × 25 mL), and the
organic phase was dried (Na2SO4). The solvent was evaporated
under reduced pressure, and the residue was purified by silica
gel column chromatography (eluent: gradient of CH3OH (0-
4926 J. Org. Chem., Vol. 70, No. 13, 2005

R-D-Configured Bicyclic Nucleosides
3.0%) in CH2Cl2) to give nucleoside 12 as a white foam (364
2-cyanoethyl N,N-diisopropylphosphoramidochloridite (0.21
mL, 0.94 mmol) were added, and the mixture was stirred at
room temperature for 23 h. The reaction was quenched with
CH3OH (3 mL), and the mixture was dissolved in ethyl acetate
(70 mL), washed with saturated aqueous NaHCO3 (3 × 50 mL)
and brine (3 × 50 mL), dried (Na2SO4), and was evaporated
under reduced pressure. The residue was purified by silica gel
column chromatography (eluent: gradient of ethyl acetate (0-
60%) in a solution of 0.5% pyridine in n-hexane) to give 21 as
a white foam (286 mg, 79%): 31P NMR (CDCl3) δ 146.31,
145.95; HRMALDI MS: m/z 795.3113 ([M + Na]+, C41H49N4O9-
PNa+ calcd 795.3129).
1
mg, 84%): Rf 0.45 (9:1 CH2Cl2/CH3OH); H NMR (300 MHz,
CDCl3) δ 9.70 (br s, 1H, NH), 7.42-7.18 (m, 11H, H-6, Ph),
5.93 (s, 1H, H-1′), 5.31 (m, 1H, OH), 4.74-4.40 (m, 8H, H-4′,
2 × CH2Ph, 2 × H-1′′, H-2′), 3.85-3.74 (m, 2H, 2 × H-5′), 2.94
(s, 3H, SO2CH3), 1.45 (s, 3H, CH3); 13C NMR (75 MHz, CDCl3)
δ 164.4 (C-4), 150.6 (C-2), 136.9, 136.4, 136.3, 128.9, 128.8,
128.5, 128.4, 128.2 (C-6, Ph), 109.9 (C-5), 94.1, 86.1, 85.3, 79.3,
74.2, 69.2, 66.2, 64.7 (C-1′, C-2′, C-3′, C-4′, C-5′, C-1′′, 2 × CH2-
Ph), 37.4 (SO2CH3), 12.2 (CH3); HRMALDI MS: m/z 569.1577
([M + Na]+, C26H30N2O9SNa+ calcd 569.1564).
(1R,2R,4S,5S)-1-(Benzyloxy)-2-(benzyloxymethyl)-4-
(thymin-1-yl)-3,6-dioxabicyclo[3.2.0]heptane (13). A solu-
tion of nucleoside 12 (364 mg, 0.67 mmol) in anhydrous
dioxane (5 mL) was cooled to 10 °C. A 60% oily dispersion of
NaH (67 mg, 1.67 mmol) was added in one portion. The
reaction mixture was stirred at 50 °C for 2 h and then
quenched with saturated aqueous NH4Cl solution (20 mL). The
reaction mixture was extracted with CH2Cl2 (4 × 15 mL). The
combined organic phase was dried (Na2SO4) and evaporated
under reduced pressure. The residue was purified by silica gel
column chromatography (eluent: gradient of CH3OH 0-1.5%
in CH2Cl2) to afford nucleoside 13 (293 mg, 98%) as a white
foam: Rf 0.45 (9:1 CH2Cl2/CH3OH); 1H NMR (300 MHz, CDCl3)
δ 8.98 (s, 1H, NH), 7.34-7.25 (m, 10H, Ph), 7.03 (d, 1H, J )
1.2 Hz, H-6), 5.88 (s, 1H, H-1′), 5.50 (s, 1H, H-2′), 4.86-4.58
(m, 6H, 2 × H-1′′, 2 × CH2Ph), 4.52 (m, 1H, H-4′), 3.86 (m,
1H, H-5′a), 3.74 (m, 1H, H-5′b), 1.86 (d, 3H, J ) 1.2 Hz, CH3);
13C NMR (75 MHz, CDCl3) δ 163.8 (C-4), 150.6 (C-2), 137.5,
137.4 (Ph), 137.2 (C-6), 128.7, 128.6, 128.2, 128.1, 128.0, 127.5
(Ph), 111.4 (C-5), 91.1 (C-1′), 91.1 (C-2′), 86.0 (C-3′), 84.4 (C-
4′), 73.9, 73.3, 68.9, 68.7 (2 × CH2Ph, C-5′, C-1′′), 12.6 (CH3);
HRMALDI MS m/z 473.1666 ([M + Na]+, C25H26N2O6Na+ calcd
473.1683). Anal. Calcd for C25H26N2O6: C, 66.65; H, 5.81; N,
6.21. Found: C, 66.50; H, 5.86; N, 6.10.
Synthesis of Oligodeoxynuclotides: Oligonucleotide
synthesis was carried out by using an automated DNA
synthesizer following the phosphoramidite approach. Synthesis
of R-oligonucleotides 22-30 was performed on a 0.2 µmol scale
by using R-thymidine 2-cyanoethyl phosphoramidite,36 N-
benzoyl-protected R-5-methylcytosine 2-cyanoethyl phosphora-
midite,37 as well as compound 21. The synthesis followed the
regular protocol for DNA synthesizer and a universal CPG
support (Biogenex/Glen Research) was employed. However, for
compound 21, a prolonged coupling time of 10 min was used.
Coupling yields for all 2-cyanoethyl phosphoramidites were
>98%. The 5′-O-DMT-ON oligonucleotides were removed from
the universal solid support by treatment with 2% LiCl in
concentrated ammonia at 55 °C for 20 h, which also removed
the protecting groups. Subsequent purification using reversed-
phase HPLC (column xterra C18 (10 µm) 7.8 × 300 mm, with
a gradient of buffers A and B (A, 90% 0.1 M NH4HCO3 + 10%
CH3CN; B, 25% 0.1 M NH4HCO3 + 75% CH3CN), 1 mL/min.),
5′-O-detritylation by acetic acid, and another HPLC purifica-
tion (same conditions) afforded the > 90% pure oligonucle-
otides. MALDI-MS [M - H]- gave the following results (found/
calcd): 22 (4196.9/4196.8); 23 (4222.5/4224.8); 24 (4306.7/
4308.8); 25 (2976.4/2980.0); 26 (3259.6/3260.1); 27 (2975.3/
2976.0); 28 (3003.6/3004.0); 29 (3058.9/3060.0); 30 (3141.8/
3144.1).
Thermal Denaturation Experiments. The thermal de-
naturation studies were performed in a medium salt buffer
containing Na2HPO4 (10 mM), NaCl (100 mM) and EDTA (0.1
mM), pH 7.0 with 1 µM concentrations of the two complemen-
tary sequences. The extinction coefficients were calculated
assuming the extinction coefficients for all thymine nucleotides
to be identical. The increase in absorbance at 260 nm as a
function of time was recorded while the temperature was
increased linearly from 10 to 65 °C at a rate of 0.5 °C/min by
means of a temperature programmer. The melting tempera-
ture was determined as the local maximum of the first
derivatives of the absorbance versus temperature curve. All
melting curves were found to be reversible.
(1R,2R,4S,5S)-1-Hydroxy-2-(hydroxymethyl)-4-(thymin-
1-yl)-3,6-dioxabicyclo[3.2.0]heptane (4). To a stirred solu-
tion of nucleoside 13 (280 mg, 0.62 mmol) in ethanol (5 mL)
was added 20% palladium hydroxide over carbon (140 mg).
The mixture was degassed several times with argon and placed
under a hydrogen atmosphere. The reaction mixture was
stirred at room temperature for 6 h and then filtered through
Celite. The filtrate was evaporated under reduced pressure,
and the residue was purified by silica gel column chromatog-
raphy (eluent: gradient of CH3OH (0-4.0%) in CH2Cl2) to give
nucleoside 4 as a white foam (165 mg, 98%): Rf 0.25 (17:3 CH2-
1
Cl2/CH3OH); H NMR (300 MHz, CD3OD) δ 7.36 (d, 1H, J )
1.2 Hz, H-6), 5.73 (s, 1H, H-1′), 5.27 (s, 1H, H-2′), 4.80 (m, 1H,
H-1′′a), 4.58 (m, 1H, H-1′′b), 4.15 (m, 1H, H-4′), 3.91 (m, 1H,
H-5′a), 3.77 (m, 1H, H-5′b), 1.88 (d, 3H, J ) 1.2 Hz, CH3); 13
C
Quantum Mechanical Calculations. All ab initio quan-
tum mechanical calculations were carried out using the
Gaussian 98 program.49 To determine the potential energy of
pseudorotation, sugar torsion angles ν2 and ν4 were calculated
using the theory of Altona and Sundaralingam with a maximal
puckering amplitude of 38°.31 For each point along the
pseudorotation pathway, a full geometry optimisation (HF/6-
31G) was carried out while maintaining the desired ν2 and ν4
angles. To ensure that no hydrogen bonds were formed
NMR (75 MHz, CD3OD) δ 166.4 (C-4), 152.7 (C-2), 139.7 (C-
6), 111.6 (C-5), 95.4 (C-2′), 92.7 (C-1′), 87.3 (C-4′), 81.3 (C-3′),
77.6 (C-1′′), 61.9 (C-5′), 12.4 (CH3); HRMALDI MS m/z
293.07470 ([M + Na]+, C11H14N2O6Na+ calcd 293.07441).
(1R,2R,4S,5S)-1-(2-Cyanoethoxy(diisopropylamino)-
phosphinoxy)-2-(4,4′-dimethoxytrityloxymethyl)-4-(thy-
min-1-yl)-3,6-dioxabicyclo[3.2.0]heptane (21). To a solu-
tion of diol 4 (160 mg, 0.593 mmol) in anhydrous pyridine (4
mL) was added 4,4′-dimethoxytrityl chloride (502 mg, 1.482
mmol), and the mixture was stirred at room temperature for
18 h. The reaction was quenched with CH3OH (1.5 mL), and
the mixture was evaporated under reduced pressure. The
residue was purified by silica gel column chromatography
(eluent: gradient of CH3OH (0-2.5%) in a solution of 0.5%
pyridine in CH2Cl2) to give 20 as a white foam: Rf ) 0.80 (17:3
(49) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. Gaussian 98, revision A.11.3; Gaussian, Inc.: Pittsburgh,
PA, 1998.
1
CH2Cl2/CH3OH); H NMR (200 MHz, CD3OD) δ 9.21 (s, 1H,
NH), 7.47-7.15 (m, 9H, Ar), 7.02 (d, 1H, J ) 1.0 Hz, H-6),
6.82 (d, 4H, J ) 8.0 Hz), 5.34 (s, 1H, H-1′), 5.24 (s, 1H, H-2′),
4.60-4.36 (m, 3H, H-1′′, H-4′), 3.78 (s, 6H, OCH3), 3.52 (m,
1H, H-5′a), 3.22 (m, 1H, H-5′b), 1.95 (d, 3H, J ) 1.0 Hz, CH3).
Intermediate 20 was dissolved in anhydrous CH2Cl2 (5 mL).
N,N-Diisopropylethylamine (0.32 mL, 1.89 mmol) followed by
J. Org. Chem, Vol. 70, No. 13, 2005 4927

Sharma et al.
between the 3′-hydroxy group and the nucleobase, the ꢀ torsion
angle was constrained to -75°.47 The â, γ, and ø angles were
not constrained but their initial values before optimization
were taken from the NMR structure of Aramini et al.47 In the
optimized structures there is no steric interactions between
the 5′-hydroxy group, the sugar and the nucleobase. Single
point energies were determined using second-order Møller-
Plesset theory (MP2) with the cc-pVTZ basis set. An unre-
strained geometry optimization using MP2/6-31G* was also
performed.
Acknowledgment. The Danish National Research
Foundation, the Danish Centre for Scientific Computa-
tion, and the Danish Natural Science Research Council
are thanked for financial support. Ms. Birthe Haack,
Malene Larsson, Suzy W. Lena, and Kirsten Østergaard
are thanked for synthetic and technical assistance. Ms.
Britta M. Dahl, Department of Chemistry, University
of Copenhagen, is thanked for synthesizing oligonuc-
leotide sequences.
Karplus Equation. A Karplus relationship correlating
JH1′-H2′ and the ν1 torsion angle was constructed using a state-
of-the-art Karplus equation (eq 1) for nucleosides and nucle-
otides developed by Altona and co-workers,32-34 where the
electronegativity of the HCCH-fragment substituents is ac-
counted for in the coefficients Cm and Sn.
Supporting Information Available: Melting curves sup-
porting the thermal denaturation studies. HPLC profiles for
oligonucleotides. Full coverage of the efforts toward 4 following
route A, Figure 2 including experimental details for new
compounds S1-S5, S7, S8, and S10-S15. Experimental
details for the preparation of 6. Experimental details for the
alternative preparation of 12 from 8. 1H and 13C NMR spectra
for compound 4. This material is available free of charge via
the Internet at http://pubs.acs.org.
3
3
JHH (θ) )
CMcos(mθ) +
Sn(nθ)
(1)
∑
∑
m)0
)1
JO0500380
4928 J. Org. Chem., Vol. 70, No. 13, 2005