Fine tuning of electrostatics around the internucleotidic phosphate through incorporation of modified 2′,4′-carbocyclic-LNAs and -ENAs leads to significant modulation of antisense properties

Article abstract of DOI:10.1021/jo8016742

(Chemical Equation Presented) In the antisense (AS) and RNA interference (RNAi) technologies, the native single-stranded 2′-deoxyoligonucleotides (for AS) or double-stranded RNA (for RNAi) are chemically modified to bind to the target RNA in order to give improved downregulation of gene expression through inhibition of RNA translation. It is shown here how the fine adjustment of the electrostatic interaction by alteration of the substituents as well as their stereochemical environment around the internucleotidic phosphodiester moiety near the edge of the minor grove of the antisense oligonucleotides (AON)-RNA heteroduplex can lead to the modulation of the antisense properties. This was demonstrated through the synthesis of various modified carbocyclic-locked nucleic acids (LNAs) and -ethylene-bridged nucleic acids (ENAs) with hydroxyl and/or methyl substituents attached at the carbocyclic part and their integration into AONs by solid-phase DNA synthesis. The target affinity toward the complementary RNA and DNA, nuclease resistance, and RNase H elicitation by these modified AONs showed that both the nature of the modification (-OH versus -CH₃) and their respective stereochemical orientations vis-à-vis vicinal phosphate play a very important role in modulating the AON properties. Whereas the affinity to the target RNA and the enzymatic stability of AONs were not favored by the hydrophobic and sterically bulky modifications in the center of the minor groove, their positioning at the edge of the minor groove near the phosphate linkage resulted in significantly improved nuclease resistance without loss of target affinity. On the other hand, hydrophilic modification, such as a hydroxyl group, close to the phosphate linkage made the internucleotidic phosphodiester especially nucleolytically unstable, and hence was not recommended. The substitutions on the carbocyclic moiety of the carba-LNA and -ENA did not affect significantly the choice of the cleavage sites of RNase H mediated RNA cleavage in the AON/RNA hybrid duplex, but the cleavage rate depended on the modification site in the AON sequence. If the original preferred cleavage site by RNase H was included in the 4-5nt stretch from the 3′-end of the modification site in the AON, decreased cleavage rate was observed. Upon screening of 52 modified AONs, containing 13 differently modified derivatives at C6′ and C7′ (or C8′) of the carba-LNAs and -ENAs, two excellent modifications in the carba-LNA series were identified, which synergistically gave outstanding antisense properties such as the target RNA affinity, nuclease resistance, and RNase H activity and were deemed to be ideal candidates as potential antisense or siRNA therapeutic agents.

Full text of DOI:10.1021/jo8016742

Fine Tuning of Electrostatics around the Internucleotidic Phosphate
through Incorporation of Modified 2′,4′-Carbocyclic-LNAs and
-ENAs Leads to Significant Modulation of Antisense Properties
Chuanzheng Zhou, Yi Liu, Mounir Andaloussi, Naresh Badgujar, Oleksandr Plashkevych,
and Jyoti Chattopadhyaya*
Department of Bioorganic Chemistry, Box 581, Biomedical Center, Uppsala UniVersity,
SE-751 23 Uppsala, Sweden
jyoti@boc.uu.se
ReceiVed July 28, 2008
In the antisense (AS) and RNA interference (RNAi) technologies, the native single-stranded 2′-deoxyoligonucleotides
(for AS) or double-stranded RNA (for RNAi) are chemically modified to bind to the target RNA in order to give
improved downregulation of gene expression through inhibition of RNA translation. It is shown here how the fine
adjustment of the electrostatic interaction by alteration of the substituents as well as their stereochemical environment
around the internucleotidic phosphodiester moiety near the edge of the minor grove of the antisense oligonucleotides
(AON)-RNA heteroduplex can lead to the modulation of the antisense properties. This was demonstrated through
the synthesis of various modified carbocyclic-locked nucleic acids (LNAs) and -ethylene-bridged nucleic acids (ENAs)
with hydroxyl and/or methyl substituents attached at the carbocyclic part and their integration into AONs by solid-
phase DNA synthesis. The target affinity toward the complementary RNA and DNA, nuclease resistance, and
RNase H elicitation by these modified AONs showed that both the nature of the modification (-OH versus -CH3)
and their respective stereochemical orientations vis-a`-vis vicinal phosphate play a very important role in modulating
the AON properties. Whereas the affinity to the target RNA and the enzymatic stability of AONs were not favored
by the hydrophobic and sterically bulky modifications in the center of the minor groove, their positioning at the
edge of the minor groove near the phosphate linkage resulted in significantly improved nuclease resisitance without
loss of target affinity. On the other hand, hydrophilic modification, such as a hydroxyl group, close to the phosphate
linkage made the internucleotidic phosphodiester especially nucleolytically unstable, and hence was not recommended.
The substitutions on the carbocyclic moiety of the carba-LNA and -ENA did not affect significantly the choice of
the cleavage sites of RNase H mediated RNA cleavage in the AON/RNA hybrid duplex, but the cleavage rate
depended on the modification site in the AON sequence. If the original preferred cleavage site by RNase H was
included in the 4-5nt stretch from the 3′-end of the modification site in the AON, decreased cleavage rate was
observed. Upon screening of 52 modified AONs, containing 13 differently modified derivatives at C6′ and C7′ (or
C8′) of the carba-LNAs and -ENAs, two excellent modifications in the carba-LNA series were identified, which
synergistically gave outstanding antisense properties such as the target RNA affinity, nuclease resistance, and RNase
H activity and were deemed to be ideal candidates as potential antisense or siRNA therapeutic agents.
118 J. Org. Chem. 2009, 74, 118–134
10.1021/jo8016742 CCC: $40.75 2009 American Chemical Society
Published on Web 12/04/2008

Modulation of Antisense Properties
Introduction
of the sugar puckering.24 Oligonucleotides containing LNA or
ENA are preorganized in the A-type canonical structure,25 which
shows unusually high affinity toward complementary RNA
strand (3-8 °C per modification).21,26 Unfortunately, though
LNA and ENA containing AONs are more nucleolytically stable
than the native PO AONs, they do not have nuclease resistance
as good as that of the PS AONs.27
Recently, the synthesis of the conformationally-locked five-
membered 2′,4′-carbocyclic-LNA (carba-LNA) and six-mem-
bered 2′,4′-carbocyclic-ENA (carba-ENA) thymidine analogs
through free-radical cyclization28 was reported by our group.
The homo- and heteroduplexes of these carba-LNA or -ENA
substituted AONs showed thermal stabilities comparable to those
of their LNA and ENA counterparts. they also showed
unprecedented blood serum stability (>48 h) and uncompro-
mised RNase H recruitment capability, which was comparable
to those of the native counterparts.28 We argued that the rigidity
of the 2′,4′-carbocycle in the carba-LNA or -ENA molecules,
together with the unique antisense properties they confer to their
AONs, make them very attractive scaffolds for further func-
tionalization. This in turn might allow us to modulate important
properties such as steric clash,29 hydration pattern,30-35 or other
electrostatic effects36-38 around phosphodiester linkage in its
homo- or heteroduplex form.
The antisense technology1,2 and RNA interference (RNAi)3
are important approaches to downregulate gene expression
through inhibition of RNA translation. In these approaches,
single-stranded oligonucleotides (for antisense) or double-
stranded RNA (for RNAi) are designed to bind to target RNA
through Watson-Crick base paring, followed by degradation
of target RNA, ideally, through the RNase H (for antisense)4
pathway or through a RNA-induced silencing complex (RISC,
for RNAi) mediated mechanism.5 Native oligoribonucleotides
do not have acceptable pharmacokinetic and pharmacodynamic
properties6,7 mainly because of lack of nuclease resistance. This
has stimulated enormous chemical efforts to produce therapeuti-
cally significant modified oligonucleotides to provide improved
stability, while maintaining high target affinity and specificity
and eliciting RNase H activity or RISC recruitment properties
in both antisense8 and siRNA technologies.7,9
AONs with direct chemical modification of the phosphodi-
ester linkage to the phosphorothioate have shown improved
nuclease resistance10,11 but reduced affinity towards a comple-
mentary RNA sequence.11 The second generation of AON 12,13
involve various 2′-modifications of the ribose moiety with
appropriate electronegative substituents. This modification can
confer an RNA-like 3′-endo sugar puckering, which results in
a considerable improvement of binding affinity to the target
RNA.14 The size, electronegativity, and configuration of the 2′-
substituents are all very important factors for the target affinity
and nuclease resistance.15,16 The new third generation of
antisense oligonucleotides, containing conformationally con-
strained LNA17 (also called BNA18) and its 6′-alkylated
derivatives19,20 or ENA21 modifications,22,23 have rigid bicyclic
ring systems that result in typical locked 3′-endo conformation
It is shown here that it is possible to carefully engineer the
chemical (electrostatic) environment around the phosphodiester
moiety by modification of the carbocyclic part of the carba-
LNA and -ENA with bulky, hydrophobic methyl and/or hydro-
philic hydroxyl groups. These specific modifications have
subsequently led to successful modulation of both the nuclease
resistance and duplex thermal stability without compromising
the RNase H activity.
Results and Discussion
(1) Zamecnik, P. C.; Stephenson, M. L. Proc. Natl. Acad. Sci. U.S.A. 1978,
75, 280–284.
(2) Stephenson, M. L.; Zamecnik, P. C. Proc. Natl. Acad. Sci. U.S.A. 1978,
75, 285–288.
(3) Fire, A.; Xu, S.; Montgomery, M. K.; Kostas, S.; Driver, S. E.; Mello,
C. C. Nature 1998, 391, 806–811.
(4) Kurreck, J. Eur. J. Biochem. 2003, 270, 1628–1644.
(5) Meister, G.; Tuschl, T. Nature 2004, 431, 343–349.
(6) Crooke, S. Annu. ReV. Med. 2004, 55, 61–95.
(7) Bumcrot, D.; Manoharan, M.; Koteliansky, V.; Sah, D. W. Y. Nat. Chem.
Biol. 2006, 2, 711–719.
1. Synthesis of the Modified Carba-LNA and ENA
Thymidine Building Blocks. As shown in the retrosynthetic
analysis in Scheme 1, the C6′ site in 7′-methyl carba-LNA or
8′-methyl carba-ENA is closest (d C6′-P ≈ 4 Å) to the adjacent
phosphodiester linkage (compound 1). This spatial proximity
(24) Obika, S.; Nanbu, D.; Hari, Y.; Morio, K.; In, Y.; Ishida, T.; Imanishi,
T. Tetrahedron Lett. 1997, 38, 8735–8738.
(8) Uhlmann, E.; Peyman, A. Chem. ReV. 1990, 90, 543–584.
(9) Manoharan, M. Curr. Opin. Chem. Biol. 2004, 8, 570–579.
(10) Campbell, J. M.; Bacon, T. A.; Wickstrom, E. J. Biochem. Biohphys.
Methods 1990, 20, 259–267.
(11) Levin, A. A. Biochim. Biophys. Acta 1999, 1489, 69–84.
(12) Manoharan, M. Biochim. Biophys. Acta 1999, 1489, 117–130.
(13) Freier, S. M.; Altmann, K. H. Nucleic Acids Res. 1997, 25, 4429–4443.
(14) Kawasaki, A. W.; Casper, M. D.; Freier, S. M.; Lesnik, E. A.; Zounes,
M. C.; Cummins, L. L.; Gonzalez, C.; Cook, P. D. J. Med. Chem 1993, 36,
831–841.
(15) Cummins, L. L.; Owens, S. S.; Risen, L. M.; Lesnik, E. A.; Freier,
S. M.; McGee, D.; Guinosso, C. J.; Cook, P. D. Nucleic Acids Res. 1995, 23,
2019–2024.
(16) Lesnik, E. A.; Guinosso, C. J.; Kawasaki, A. W.; Sasmor, H.; Zounes,
M.; Cummins, L. L.; Ecker, D.; Cook, P. D.; Freier, S. M. Biochemistry 1993,
32, 7832–7838.
(17) Singh, S. K.; Nielsen, P.; Koshkin, A. A.; Wengel, J. Chem. Commun.
1998, 455–456.
(18) Obika, S.; Morio, K.; Nanbu, D.; Imanishi, T. Chem. Commun. 1997,
1643–1644.
(19) Swayze, E. E.; Seth, P. P. Patent WO2007090071, 2007.
(20) Enderlin, G.; Nielsen, P. J. Org. Chem. 2008, 73, 6891–6894.
(21) Morita, K.; Hasegawa, C.; Kaneko, M.; Tsutsumi, S.; Sone, J.; Ishikawa,
T.; Imanishi, T.; Koizumi, M. Bioorg. Med. Chem. Lett. 2002, 12, 73–76.
(22) Albak, N.; Petersen, M.; Nielsen, P. J. Org. Chem. 2006, 71, 7731–
7740.
(25) Petersen, M.; Bondensgaard, K.; Wengel, J.; Jacobsen, J. P. J. Am. Chem.
Soc. 2002, 124, 5974–5982.
(26) Koshkin, A. A.; Singh, S. K.; Nielsen, P. R. V. K.; Kumar, R. M. M.;
Olsen, C. E.; Wengel, J. Tetrahedron 1998, 54, 3607–3630.
(27) Morita, K.; Takagi, M.; Hasegawa, C.; Kaneko, M.; Tsutsumi, S.; Sone,
J.; Ishikawa, T.; Imanishi, T.; Koizumi, M. Bioorg. Med. Chem. 2003, 11, 2211–
2226.
(28) Srivastava, P.; Barman, J.; Pathmasiri, W.; Plashkevych, O.; Wenska,
M.; Chattopadhyaya, J. J. Am. Chem. Soc. 2007, 129, 8362–8379.
(29) Pattanayek, R.; Sethaphong, L.; Pan, C.; Prhavc, M.; Prakash, T. P.;
Manoharan, M.; Egli, M. J. Am. Chem. Soc. 2004, 126, 15006–15007.
(30) Teplova, M.; Minasov, G.; Tereshko, V.; Inamati, G. B.; Cook, P. D.;
Manoharan, M.; Egli, M. Nat. Struct. Biol. 1999, 6, 535–539.
(31) Egli, M.; Minasov, G.; Teplova, M.; Kumar, R.; Wengel, J. Chem.
Commun. 2001, 651–652.
(32) Schneider, B.; Patel, K.; Berman, H. M. Biophys. J 1998, 75, 2422–
2434.
(33) Tereshko, V.; Gryaznov, S.; Egli, M. J. Am. Chem. Soc. 1998, 120,
269–283.
(34) Auffinger, P.; Westhof, E. Angew. Chem., Int. Ed. 2001, 40, 4648–
4650.
(35) Ball, P. Chem. ReV. 2008, 108, 74–108.
(36) Griffey, R. H.; Monia, B. P.; Cummins, L. L.; Freier, S. M.; Greig,
M. J.; Guinosso, C. J.; Lesnik, E.; Manalili, S. M.; Mohan, V.; Owens, S.; Ross,
B. R.; Sasmor, H.; Wancewicz, E.; Weiler, K.; Wheeler, P. D.; Cook, P. D.
J. Med. Chem. 1996, 39, 5100–5109.
(23) Abdur Rahman, S. M.; Seki, S.; Obika, S.; Yoshikawa, H.; Miyashita,
K.; Imanishi, T. J. Am. Chem. Soc. 2008, 130, 4886–4896.
(37) Prhavc, M.; Prakash, T. P.; Minasov, G.; Cook, P. D.; Egli, M.;
Manoharan, M. Org. Lett. 2003, 5, 2017–2020.
J. Org. Chem. Vol. 74, No. 1, 2009 119

Zhou et al.
SCHEME 1
suggests that the hydration pattern around the adjacent phosphate
could be modulated by introducing steric and/or polar effect
through introduction of a simple substituent such as a hydroxyl
and/or methyl group at C6′ (R1 and R2 in compound 1 in Scheme
1). This is particularly attractive because 6′-OH and/or 6′-Me
can easily be derived from the cyclic ketone 2. This cyclic
ketone, on the other hand, can be obtained by free radical
cyclization28 of the hydroxyl olefin 3. The key cyclization
precursor 3 can be easily synthesized from aldehyde 528,39 with
Grignard reagent. In order to shorten our synthetic route and
avoid side reactions, we decided to build up the 4′-alkyl side
chain through the Grignard reaction before the glycosylation
step with thymine.
they have the C6′S stereochemistry, which also proved that the
Grignard reaction took place exclusively to give 7 (6 f7) with
C6′S stereochemistry. The 12a and 12b were oxidized to the
corresponding ketones 13a and 13b with Dess-Martin perio-
dinane, followed by reduction with NaBH4. For 12b, this
oxidization-reduction sequence [12b f 13b (73%) f 14b
(90%)] resulted in complete inversion of configuration at C6′
in 14b, while the reduction of 13a gave the two isomers 14a
and 12a in 6/4 ratio. This differential stereoselectivity for
reduction of 13a and 13b can be easily understood from the
fact that in 13b, the bulkyl 7′-methyl and 3′-O-Bn pointed to
the same side of the carbocycle (i.e., above the ring) and hence
hydride attacked predominantly from below the plane of the
carbocycle ring owing to the steric hindrance at the top. On the
other hand, in 13a the orientation of 3′-O-Bn (located above
the plain of the carbocycle) and 7′-methyl (located below the
plain of the carbocycle) were opposite, and therefore the hydride
attack was relatively less selective, suggesting that both the top
and the bottom sides of the carba-ring may have similar steric
hindrance.
The synthetic route to the modified carba-LNAs is shown in
Scheme 2. The synthesis started from the known precursor 6,39
which was oxidized through Swern oxidation to the correspond-
ing aldehyde 5. The crude aldehyde was subjected to Grignard
reaction with vinylmagnesium bromide at room temperature
overnight to give the C6′-hydroxyl olefin 7 in high yield (84%
in two steps). TLC and NMR showed that the olefin 7 is a pure
isomer; no configuration at C6′ could be confirmed by nuclear
Overhauser effect (NOE) experiment at this stage. This could
however be confirmed only after the free-radical cyclization step
(see below), which subsequently showed the C6′-S stereochem-
istry of 7 after the Grignard reaction. Acetylation of 7 using a
mixture of acetic anhydride, acetic acid, and triflic acid gave
the corresponding triacetate 8 as an anomeric mixture. The crude
8 was subjected to a modified Vorbruggen reaction involving
in situ silylation of thymine and subsequent trimethylsilyl
triflate-mediated coupling to give exclusively ꢀ-D-ribofuranosyl
thymine derivative 9.39 Deacetylation of 9 using 30% methy-
lamine in ethanol at room temperature overnight gave compound
10 quantitatively, which was esterified with phenyl chlorothio-
formate to give the key precursor 11 for the free-radical
cyclization. The esterification of 10 to 11 was fast, highly
selective, and high yielding (92%). The 2′-O-phenoxythiocar-
bonyl (PTC) derivative 11 was found to be the sole product
even when 2.5 equiv of phenyl chlorothioformate was used.
The key free-radical cyclization was carried out in refluxing
anhydrous toluene with the Bu3SnH, using AIBN as the initiator,
which were added dropwise in order to avoid the side reactions.
Just as reported before,28 the 5-hexenyl radical cyclization
proceeded in the 5-exo cyclization mode to give 12a (with C7′
methyl at the pseudoequatorial position, 7′S) as the major
product along with the minor product 12b (with C7′ methyl
being pseudoaxial, 7′R). In addition, 1D NOE experiments
(Figures SIII.1 and SIII.2) of 12a and 12b also showed that
To synthesize the phosphoramidites of 12a, 12b, 14a, and
14b for solid-support DNA synthesis, the 6′-OH should be
protected with an acetyl group. Unfortunately, it was found
that 6′-O-acetate was unstable during the debenzylation by
hydrogenolysis. The 6′-O-(p-toluoyl) group, however, proved
to be a good choice to circumvent this. Thus, compounds
15a and 15b could be debenzylated completely using 20%
Pd(OH)2/C (500 mg/mmol substrate) and ammonium formate
(20 equiv) under reflux in methanol for 4 h, without any
toluoyl ester cleavage being detected, followed by 5′-
dimethoxytritylation to give 16a and 16b in 70% and 63%
yields, respectively. However, under the same condition,
debenzylation of 15c gave a byproduct beside the desired
product 15c′. The amount of this byproduct increased with
prolongation of the reaction time. NMR analysis of the
byproduct showed it was formed by transfer of the 6′-O-(p-
toluoyl) group to the primary 5′-OH in 15c′ (see Figures
SI.49-53 for NMR characterization of the side product 15c′′).
A simple conformational check showed that in compound
15c′ the vicinal C5′ and 6′-O- were cis to each other with
the torsion φC5′-C4′-C6′-O6′ ) 42° (Figure SIII.25). This cis
comformational disposition made it possible for primary 5′-
OH to attack the CdO of the 6′-O-(p-toluoyl) group to form
a relatively stable six-membered carboxonium ion transition
state, whereas in compounds 15a′ and 15b′, φC5′-C4′-C6′-O6′
was about 75° so formation of such a six-membered transition
state was disfavored, and hence no isomeric product was
formed owing to migration. Compound 15d′ had the same
6′-R configuation as in compound 15c′ and was also
accompanied by this type of isomeric side reaction during
debenzylation. Fortunately, this side reaction could efficiently
(38) Prakash, T. P.; Puschl, A.; Lesnik, E.; Mohan, V.; Tereshko, V.; Egli,
M.; Manoharan, M. Org. Lett. 2004, 6, 1971–1974.
(39) Varghese, O.; Barman, J.; Pathmasiri, W.; Plashkevych, O.; Hon-
charenko, D.; Chattopadhyaya, J. J. Am. Chem. Soc. 2006, 128, 15173–15187.
120 J. Org. Chem. Vol. 74, No. 1, 2009

Modulation of Antisense Properties
SCHEME 2
be suppressed by increasing the hydrogenolysis reagents (20%
Pd(OH)2/C, 1500 mg/mmol substrate and ammonium formate,
60 equiv). Under this condition, the reaction was completed
in a shorter time (2 h) to give pure desired product 15d′,
which could be subjected directly to 5′-dimethoxytritylation
to give 16d in high yield. The 3′-phosphitylation of 16a-16d
gave the respective phosphoramidites 17a-17d using stan-
dard conditions in moderate yields 60-77%.
anhydrous THF for 3 h to give two isomers, 18a (43%) and
18b (18%). The two isomers have a large Rf difference on TLC
and could easily be separated by short column chromatography.
The isomer with lower Rf was the major product, and confor-
mational analysis by 1D NOE (Figure SIII.5) proved it to
possess a 6′-S configuration (18a). The ratio of 18a/18b was
2.5:1, which means attacking the carbon center from the side
of the 7′-methyl substituent by methyl anion had more steric
hindrance. In view of the stecric hindrance and poorer reactivity
of the tertiary 6′-OH, no attempt was made to protect it. Hence,
compounds 18a and 18b were debenzylated through catalytic
hydrogenolysis followed by 5′-dimethoxytritylation to give 19a
For methylation40 of C6′ of carba-LNA (Scheme 3), the
ketone 13a was treated with methyl magnesium iodide in
(40) Kakefuda, A.; Yoshimura, Y.; Sasaki, T.; Matsuda, A. Tetrahedron 1993,
49, 8513–8528.
J. Org. Chem. Vol. 74, No. 1, 2009 121

Zhou et al.
SCHEME 3
SCHEME 4
and 19b in 84% and 63% yields, respectively. Despite the
presence of two hydroxyl groups (6′-OH and 3′-OH) in
compounds 19a and 19b, the 3′-OH could be phosphitylated
selectively by treating 19a and 19b with 2-cyanoethyl-N,N-
diisopropylphosphoramidochloridite under standard conditions,
giving phosphoramidites 20a and 20b in 77% and 67% yields,
respectively.
To deoxygenate the 6′-OH in compound 18a, a radical
deoxygenation procedure41 was employed. Thus, 18a was treated
with methyl(chlorocarbonyl) formate42 in acetonitrile in the
presence of DMAP to give 21 in 94% yield. Compound 21 was
subjected to a standard radical deoxygenation condition to give
an intractable mixture of two isomers 22a and 22b (total yield
33%), which could not be separated by column chromatography.
The mixture of 22a and 22b was debenzylated followed by 5′-
dimethoxytritylation to give an intractable mixture of 23a (6′S,
80%)/23b (6′R, 20%) (total yield 73%), which was phosphity-
lated to give phosphoramidites 24a and 24b as a diastereomeric
mixture (63%) that was used directly for DNA synthesis.
The synthetic routes to the modified carba-ENA compounds
are shown in Schemes 4 and 5, which are very closely similar
to the ones used for the synthesis of carba-LNA derivatives. A
brief description of these two schemes, with the detailed
experimental procedures, product isolation, and spectroscopic
characterization can be found in Supporting Information (SI Part
IV).
2. NMR Characterization of Key Intermediates Involved
in the Synthesis of the Functionalized Carba-LNA and
-ENA Thymidines. All key intermediates 12a/b, 14a/b, 18a/
(41) Hartwig, W. Tetrahedron 1983, 39, 2609–2645.
(42) Dolan, S. C.; MacMillan, J. J. Chem. Soc., Chem. Commun. 1985, 1588–
1589.
1
b, 22a/b, 30, 31, 33, 35a/b, and 36 were characterized by H,
122 J. Org. Chem. Vol. 74, No. 1, 2009

Modulation of Antisense Properties
SCHEME 5
13C, COSY, 1H-13C HMQC, and HMBC NMR and mass
spectroscopy.
3. Molecular Structures of Functionalized Carba-LNA-T
Based on NMR and ab Initio Calculations. Comparison of
experimental vicinal coupling constants with those back calcu-
lated from the corresponding theoretical torsion angles employ-
ing the Haasnoot-de Leeuw-Altona generalized Karplus
equation43,44 have shown satisfactory convergence of the
experimental and theoretical results (Table SIII.1). This con-
vergence was a good indicator of the rigidity of both the
furanose sugar and the carbocycle moiety of carba-LNA
derivatives. The chemical nature and the chirality of the
substituents at C6′ and C7′ have been found to dictate the degree
of puckering of the constraining 2′,4′ carbocyclic five-membered
ring, which has been found to be ranging from near-planar
puckering in compound 14b (C2′-C7′-C6′-C4′ ≈ -0.7°) to
nearly 8° for the C2′-C7′-C6′-C4′ torsion in compounds 14a
and 18b (Table SIII.2). The chemical nature and the chirality
of the substituents have also been found to influence puckering
of the furanose sugar ring, resulting in the pseudorotational
angles ranging from 16° (compound 12b) to 20° (compound
18b, Table SIII.2).
4. Thermal Denaturation Studies of Carba-LNA and
Carba-ENA Modified AONs 1-52. The phosphoramidites
17a-d, 20a/b, 24a/b, and 44-48 (Schemes 2, 3, and 5) were
incorporated into a 15mer DNA sequence through solid-phase
DNA synthesis protocol45 on an automated RNA/DNA synthe-
sizer. The discussion about oligonucleotides synthesis and
purification can be found in Supporting Information (SI Part
IV). The Tm values of duplexes formed by AONs 1-52 with
the complementary RNA or DNA are listed in Table 1.
Comparison of Tm’s of AONs containing different carba-LNA
derivatives clearly suggested how the nature of the modification
(-OH versus -CH3) and their respective stereochemical orienta-
tions at C6′ and C7′ in the carbocylic moiety of carba-LNA
and carba-ENA modulates the AONs affinity toward the
complemenary RNA. This Tm effect owing to -OH versus -CH3
substitution in the carba moiety gives us an insight as to how
the spine of hydration is affected around the internucleotidic
phosphate, thereby either minimizing the internucleotidic phos-
phodirester repulsion or attenuating the stabilization of the
phosphate charges through additional salt bridge formation.
4.1. Comparison of Type II with Type III and Type IV
with Type V Modifications in Carba-LNA. The [6′S-OH,7′R-
Me]-carba-LNA modified (Type III) AONs 10-12 have
The correlation between H2′ with H7′ for 12a/b (Figures
SI.11 and SI.16) or H2′ and H8′ for 30 (Figure SII.11) in COSY
experiments showed that bicyclic systems have indeed been
formed during radical cyclization. This evidence was further
corroborated by the observation of long range connectivity of
H7′ with C2′ (for compound 12a/b, Figures SI.13 and SI.18)
or H8′ with C2′ (for compound 30, Figure SII.13) in the HMBC
experiments.
The orientation of substituents in carbocyclic moiety of
compound 12a/b, 14a/b, 18a/b, 22a/b, 30, 33, 35a/b, and 36
were assigned by NOE experiments (see Figures SIII.1-SIII.13
and discussion in SI Part IV). The coupling constants were also
used to derive information about configurations of various chiral
centers in these molecules (Tables SIII.1 and SIII.3). The 3JH2′,H7′
of compounds 12b and 14b were close to zero, which suggested
that the torsion angles H2′-C2′-C7′-H7′ were nearly 90°,
corroborating their 7′R-configuration, whereas compounds 12a,
3
14a, 18a/b, and 22a/b, which had a JH2′,H7′ of ∼4-5 Hz
(Figures SIII.14-SIII.21), were in a 7′S-configuration. The
observation of the four-bond W-coupling (wJ) between H3′ and
H6′ in compounds 14a/b (Figures SI.25 and SI.30) and 22b
(Figure SI.96) suggested that the 6′-H point at O4′. The 3JH6′,H7′
in compounds 12a, 14b, and 22a were 8-11 Hz, suggesting
that H6′ and H7′ were in cis configuration, whereas in
3
compounds 12b, 14a, and 22b, the JH6′,H7′ were 3.4-6 Hz
(Figures SIII.14-SIII.21), and hence H3′ and H6′ were in trans
configuration.Thecarba-ENAadoptsatypicalchairconformation.22,28
3
In compounds 30, 31, 33, 35a, 35b, and 36, the large JH8′,H7′′
(12-13.5 Hz) suggested that the H8′ and H7′′ (axial proton)
were in trans orientation, and hence the H8′ should also be axial
in these compounds. The H6′ in compounds 30, 31, and 35a
3
should also be axial, given the large JH6′,H7′′ (10.5-12.5 Hz).
On the contrary, in compounds 35b and 36, H6′ should be
equatorial (3JH6′,H7′′ ) 7.5 and 5.0 Hz in compounds 35b and
36, respectively). Thus the configurational information obtained
from coupling constant analysis found to corroborate very well
with those obtained from the NOE experiments.
In compounds 19a/b and 41, selective phosphitylation at 3′-
OH was confirmed by the NMR in that (i) the comparison of
the 1H and 1H{31P} NMR spectra showed that the H3′ of
compounds 20a/b and 46 were coupled with 31P with J≈ 10
Hz (Figures SI.80, SI.84, SI 88, SI 91, SII.62, and SII.63), and
(ii) the 13C NMR experiment showed the C3′ were also coupled
to 31P (JC3′,P ≈ 15-18 Hz, Figure SI.81 and SI.85) but not to
the C6′-Me.
(43) Haasnoot, C. A. G.; de Leeuw, F. A. A. M.; Altona, C. Tetrahedron
1980, 36, 2783–2792.
(44) Altona, C.; Sundaralingam, M. J. Am. Chem. Soc. 1972, 94, 8205–1822.
(45) Beaucage, S. L.; Caruthers, M. H. Current Protocols in Nucleic Acid
Chemistry; John Wiley & Sons, Inc.: New York, 2003.
J. Org. Chem. Vol. 74, No. 1, 2009 123

Zhou et al.
TABLE 1. Thermal Denaturation of Duplexes of Native and Carba-LNA and -ENA Modified AONs with Complementary RNA or DNAa
a Tm values measured as the maximum of the first derivative of the melting curve (A260nm vs temperature) in medium salt buffer (60 mM Tris-HCl at
pH 7.5, 60 mM KCl, 0.8 mM MgCl2) with temperature range 20 to 70 °C using 1 µM concentrations of the two complementary strands. The value of
Tm given is the average of two or three independent measurements. If the error of the first two measurements exceeded (0.3 °C, the third measurement
was carried out to confirm if the error is indeed within (0.3°C. b Modification I and IX have been reported,28 and here they are standards for
comparison. c For carba-LNA derivatives modified AONs 6-32, the Tm were compared with (7′-Me)-carba-LNA28 (Type I) modified AONs 2-5,
whereas carba-ENA derivatives modified AONs 37-52 were compared with (8′S-Me)-carba-ENA28 (Type IX) modified AONs 33-36 to give the ∆Tm
values.
124 J. Org. Chem. Vol. 74, No. 1, 2009

Modulation of Antisense Properties
significantly decreased Tm’s (-1.5 to -2°C) compared to [6′S-
OH,7′S-Me]-carba-LNA modified (Type II) AONs 6-8, which
suggests that when the 7′-methyl group is pointed toward the
vicinal 3′-phosphate, it destabilizes the AON/RNA duplex
greatly. This result can be further corroborated by the result
that [6′R-OH,7′R-Me]-carba-LNA modified (Type V) AONs
18-21 also have lower Tm’s by -1.5 to -2°C compared to
[6′R-OH,7′S-Me]-carba-LNA modified (Type IV) AONs 14-17.
4.2. Comparison of Type II with Type IV and Type III
with Type V Modifications in Carba-LNA. [6′R-OH,7′S-Me]-
Carba-LNA modified (Type IV) AONs 14-17 have similar Tm’s
as [6′S-OH,7′S-Me]-carba-LNA modified (Type II) AONs 6-9.
[6′R-OH,7′R-Me]-Carba-LNA modified (Type V) AONs 18-21
have also similar Tm’s as [6′S-OH,7′R-Me]-carba-LNA modified
(Type III) AONs 10-12, thereby suggesting that the stereo-
chemical orientation of C6′-OH in carba-LNA does not exert a
marked effect on the AON/RNA duplex thermal stability.
4.3. Comparison of Type II with Type VII and Type
IV with Type VIII modifications in Carba-LNA. Type VII
([6′S-OH/Me,7′S-Me]-carba-LNA) modificaion has one ad-
ditional C6′-methyl group, which points at the vicinal 3′-
phosphate compared to that in Type II ([6′S-OH,7′S-Me]-carba-
LNA) modification, and as a result, the Type VII modified
AONs 26-29 have slightly higher (∼0.5 °C) Tm’s, at least not
less than those in Type II modified AONs 6-9. Similarly, [6′R-
OH/Me,7′S-Me]-carba-LNA (Type VIII) modified AONs
30-32 have similar or slightly higher Tm’s compared with those
of [6′R-OH,7′S-Me]-carba-LNA (Type IV) modified AONs
30-32. This means that the orientation of the C6′-Me in either
S or R configuration in carba-LNA does not impair the AON/
RNA duplex thermal stability, and in some cases, it can even
have a stabilization effect.
4.4. Comparison of Type IX with Type X, Type XI,
Type XII, and Type XIII Modifications in Carba-ENA. It
seems that the hydroxyl and methyl subsitutents in the carbocy-
clic moiety of carba-ENA exert different kinds of effect on the
thermal stability of AON/RNA duplex compared to what is
observed in carba-LNAs.
4.4.1. Orientation of C6′-OH in Carba-ENA Plays
Important role. In Type X modification ([6′S-OH,8′S-Me]-
carba-ENA), C6′-OH points away from the 3′-phosphate, and
thus Type X modified AONs 37-40 have decreased (-1 to
-1.5°C) Tm values compared with those in [6′R-OH,8′S-Me]-
carba-ENA modified AONs 41-44 in which the C6′-OH points
at the 3′-phosphate.
when the methyl group is pointed toward the vicinal 3′-
phosphate. The carba-ENA uridine was reported to be stabilizing
the AON/RNA duplex by 3.5-4.5 °C/per modification com-
pared to the native counterpart,22 whereas the (8′-methyl)-carba-
ENA thymidine modified AONs 33-36 show only 0.5-1°C
increase in Tm’s. It should be mentioned that the sequences
reported here are different from the literature,22 and so it is not
clearly possible here to distinguish the real effect from the
sequence-specific effect. Secondly, when the 6′-methyl in carba-
LNA or carba-ENA is pointed toward the vicinal 3′-phosphate,
it does not impair the thermal stabliltiy of AON/RNA duplex.
The effects on thermal stability of AON/DNA duplex
imparted by the carbocyclic substitution follow the same trend
as those observed for the AON/RNA duplex, but in general the
stabilization or destabilization effect in the AON/DNA homo-
duplex is more pronounced than in the AON/RNA heteroduplex.
In summary, the electronic and steric influence of substituents
and their relative stereochemical configuration in the AON
strand on its target RNA affinity is subtle. An additional
discussion about how the position and stereochemical orientation
of various substitutents on the carbocyclic moiety in the AON
strand affects the Tm of the duplex can be found in Supporting
Information (SI Part IV).
5. 3′-Exonuclease Stability of the Carbocylic-Modified
AONs. New Insights into the Mechanism of the Cleavage.
The metabolism of AONs in vivo involves degradation by exo-
and endocucleases, and the predominate nuclease activity was
of 3′-exonuclease.46 Here the AONs with a single modification
at position T13 (position 3 from 3′-end, see Figure 1D) were
chosen to test the 3′-exonuclease resistance of the modified
oligonucleotides. Thus, the AONs (32P-labeled at 5′-end) were
incubated with phosphodiesterase I from Crotalus adamanteus
venom (SVPDE) [SVPDE 6.7 ng/µL, AON 3 µM, 100 mM
Tris-HCl (pH 8.0), 15 mM MgCl2, total volume 30 µL) at 21°C.
Aliquots were taken out at appropriate time intervals and
analyzed by 20% denaturing PAGE. The gel pictures obtained
upon autoradiography are shown in Figures SIII.24 and SIII.25.
The native DNA (AON 1) did not show any 3′-exonuclease
resistance and was completely degraded in 10 min, whereas for
the modified AONs, the last nucleotide T15 was removed
quickly to give 14mer oligonucleotides. Because of the presence
of T13 modification, phosphate P14 (see Figure 1D for
phosphate (P) numbering) has considerably improved stability
toward 3′-exonuclease. T13 modification also improves the
stability of the phosphate P13 to some extent, so after slow
digestion of P14, the formed 13mer AON was also relatively
stable (see Figures SIII.24 and SIII.25). However, once the P13
cleaved, AON was degraded to the monomer blocks quickly,
and thus no bands corresponding to 12mer to dimer oligos could
be observed.
4.4.2. Orientation of C6′-Me in Carba-ENA Is Also
Important. [6′S-Me,8′S-Me]-Carba-ENA modified AONs
45-48 give distinctly decreased Tm’s compared with those of
AONs 33-36, but [6′R-Me,8′S-Me]-carba-ENA modified AONs
49-52 do not have any negative effect on the AON/RNA
thermal stability compared to AONs 33-36. This result suggests
that when the C6′ methyl group points away from the vicinal
3′-phosphate in carba-ENA, it destabilizes the AON/RNA
duplex significantly, but when the C6′ methyl group points at
the 3′-phosphate, its role is more subtle.
In Figure 1, total percentages of integrated 14mer and 13mer
AONs were plotted against time points to give SVPDE digestion
curves for each AON, and pseudo-first-order reaction rates could
be obtained by fitting the curves to single-exponential decay
functions. A comparison of digestion rates of AONs with
different types of modifications showed the following results:
(i) 6′-Methyl always considerably improved enzymatic stability
of vicinal phosphate linkage. Among the carba-LNA derivatives
modified AONs, [6′R/S-Me,7′S-Me]-carba-LNA modified (Type
VI) AON 22 (k ) 0.0218 ( 0.0013 h-1) showed the best
On the basis of the above comparison, it can be concluded
that there are two common features for carba-LNA and -ENA
as to how the orientation of hydroxyl and methyl group in the
carba moiety influnces the thermal stabliltiy of AON/RNA
duplex. First, both the 7′-Me in carba-LNA and 8′-Me in carba-
ENA that are located at the center of the minor groove play a
marked role in destabilizing the AON/RNA duplex, especially
(46) Shaw, J. P.; Kent, K.; Bird, J.; Fishback, J.; Froehler, B. Nucleic Acids
Res. 1991, 19, 747–750.
J. Org. Chem. Vol. 74, No. 1, 2009 125

Zhou et al.
FIGURE 1. Amount of remaining initial oligonucleotide (taken 14mer and 13mer together in %) during 3′-exonuclease (SVPDE) promoted digestion.
(A) AONs with functionalized carba-LNA modifications. (B) AONs with functionalized carba-ENA modifications. For panels (A) and (B), following
digestion condition was used: AON 3µM (5′-end 32P labeled with specific activity 80 000cpm), 100 mM Tris-HCl (pH 8.0), 15 mM MgCl2, SVPDE
6.7 ng/µL, reaction temperature 21°C, total reaction volume was 30 µL. (C) Comparison of the stabilities of AON 6 and AON 14 upon SVPDE
digestion using similar digestion condition as in panels A and B, but using lower SVPDE concentration (2.2 ng/µL). (D) Molecular structure of T13
modified AONs. R1, R2, R3, R4 ) Me, OH or H. n ) 0 or 1. The full sequence is 3′-d(CTT13 CTT TTT TAC TTC-32P)-5′.
126 J. Org. Chem. Vol. 74, No. 1, 2009

Modulation of Antisense Properties
enzymatic stability, and was 7 times more stable than [7′-Me]-
carba-LNA (Type I) modified AON 2 (k ) 0.1525 ( 0.0097
h-1); [6′S-Me,8′S-Me]-carba-ENA (Type XII) modified AONs
45 and [6′R-Me,8′S-Me]-carba-ENA (Type XIII) modified AON
49 were also shown to have the best enzymatic stability amongst
the carba-ENA derivatives modified AONs. (ii) When 6′-OH
pointed at the vicinal 3′-phosphate, it was renderded more
susceptible to enzymatic degradation. [6′R-OH,7′S-Me]-carba-
LNA modified (Type IV) AON 14 was the least enzymatically
stable amongst the carba-LNA modified AONs and was
degraded completely in 10 min. The [6′R-OH,8′S-Me]-carba-
resistance of AONs. The lipophilic nature of the methyl group
conferred nuclease resistance by two effects: First, the methyl
could decrease the magnitude of hydration around the scissile
phosphate (hydrophobic effect). Second, the bulky methyl could
also disturb the interaction between nuclease and phosphate
linkage owing to the steric clash. This effect was also illustrated
by Imanishi’s group recently.23 All of the modifications of Type
II, III, IV, and V in AONs 6, 10, 14, and 18 have a 7′-methyl
substituent, but [6′S-OH,7′R-Me]-carba-LNA modified (Type
III) AON 10 and [6′R-OH,7′R-Me]-carba-LNA modified (Type
V) 18 were more stable than [6′S-OH,7′S-Me]-carba-LNA
modified (Type II) AON 6 and [6′ R-OH,7′S-Me]-carba-LNA
modified (Type IV) 14, respectively (if only the stability of P14
is considered, see Figure SIII.27). This result suggested that as
the scissile phosphate came closer to the lipophilic substituent,
it became more stable toward the nuclease degradation. Thus,
the case that carba-ENA do not have significantly improved
nuclease resistance compared to carba-LNA could be plausibly
explained by the fact that the six-membered carbocycle of carba-
ENA adopted a perfect chair conformation,22,47 and thus the
additional endocyclic methylene group orientated away from
the 3′-phosphate (see the model in Figure SIV.4). On the
contrary, an additional methyl group at C6′ of carba-LNA, which
is spatially located at the edge of the minor groove, came much
closer to the vicinal scissile phosphate linkage, resulting in its
greater improvement in the nuclease stability.
ENA (Type XI) modified AON 41 (k ) 0.6701 ( 0.0437 h-1
)
was also degraded 6 times faster than the [8′S-Me]-carba-ENA
(Type IX) modified AON 33 (k ) 0.1145 ( 0.0030 h-1). (iii)
When the 6′-OH pointed away from the vicinal 3′-phosphate,
its role was more complex. For example, [6′S-OH,8′S-Me]-
carba-ENA (Type X) modified AON 37 (k ) 0.2119 ( 0.0096
h-1) was less enzymatically stable than [8′S-Me]-carba-ENA
modified AON 33, while [6′S-OH,7′S-Me]-carba-LNA (Type
II) modified AON 6 (k ) 0.0532 ( 0.0044 h-1) and [6′S-
OH,7′R-Me]-carba-LNA (Type III) modified AON 10 (k )
0.0605 ( 0.0032 h-1) were 2-3 times more stable than [7′-
Me]-carba-LNA modified AON 2. One interesting observation
is that when the 6′-OH points away from the vicinal 3′-phosphate
(C6′-S), as in modifications II, III, VII, and V in AON 6, 10,
26, and 37, it can improve the nuclease resistance of 5′-
phosphate (P13 in Figure 1D) and so the band corresponding
to 13mer could be observed on PAGE. (iv) By comparing
degradation kinetics and digestion pattern of [6′R-OH,7′S-Me]-
carba-LNA modified (Type IV) AON 14 with [6′R-OH,7′R-
Me]-carba-LNA modified (Type V) AON 18, [6′S-OH,7′S-Me]-
carba-LNA modified (Type II) AON 6 with [6′S-OH,7′R-Me]-
carba-LNA modified (Type III) AON 10, it could be concluded
that the 7′-methyl, when it pointed at the vicinal 3′-phosphate
(C7′-R), had more positive effect on the enzymatic stability of
the vicinal 3′-phosphate than when it pointed away from(C7′-
S) the phosphate. (v) Comparing digestion rates (Figure SIII.26)
of carba-LNA modifications with carba-ENA modifications that
have the same substitution (same fuctional group, location and
orientation), it could be seen that although carba-ENA deriva-
tives have an additional endocyclic methylene group, they did
not show significantly better nuclease resistance compared to
that of the carba-LNA counterparts.
5.1. Lipophilic Substituents Improve Exonuclease
Resistance of the Vicinal Scissile Phosphate, Depending
upon Their Stereochemical Orientation with Respect
Vicinal Phosphate. The most interesting conclusion one can
derive from the above results is that the exonuclease resistance
is dependent not only on the lipophilic Versus hydrophilic nature
of the substituent but also significantly on the stereochemical
orientation of its chirality vis-a`-vis scissile phosphate. In our
previous study,28 comparison of nuclease resistance of different
2′,4′-constrained modifications with different lipophilic or
hydrophilic nature of the C2′-substituent (oxo vs aza vs carba)
led to the conclusion that hydration around the C2′-substituent
was an important factor for promoting the nuclease-promoted
hydrolysis of phosphate. That study showed that the lipophilic
C2′ carba-substituent can improve the nuclease resistance of
the vicinal scissile phosphate. In the present study, this
conclusion was further corroborated by the fact that the methyl
substitution in the carbocycle moiety of the carba-LNA or -ENA,
independent of its location, can always improve nuclease
5.2. Studies on
a Pair of Pure Diastereomers of
C6′-Hydroxyl (R and S)-Substituted Carba-LNA Shows
the Chirality-Dependent Stabiliy of the Scissile Phosphate
toward Exonuclease. Hydroxyl substitution on the carbocyclic
moiety generally leads to loss of enzymatic stability of vicinal
phosphate, as expected from the above discussion. Thus, AONs
37 and 41 with the hydrophilic 6′-OH substituents are less
enzymatic stable than the unsubstituted AON 33. Similarly,
AONs 14 and 18 are less enzymatically stable than AON 2 (see
Figure 1A and B). The orientation of hydrophilic 6′-OH had
even more pronounced effect on the nuclease resistance of the
vicinal scissile phosphate. The 6′-OH in [6′S-OH,7′S-Me]-carba-
LNA (Type II) modified AON 6 and [6′R-OH,7′S-Me]-carba-
LNA modified (Type IV) AON 14 orientate with opposite
chirality. As a result, AON 6 and AON 14 had completely
different stability toward SVPDE incubation. In the upper
experiment, AON 14 with 6′-OH pointing toward the 3′-
phosphate was degraded so fast that no accurate degradation
rate could be obtained. Hence, AON 6 and 14 were incubated
under a very low concentration of SVPDE [SVPDE 2.2 ng/µL,
AON 3µM, 100 mM Tris-HCl (pH 8.0), 15 mM MgCl2, 21°C,
total volume 30 µL]. The gel pictures obtained by autoradiog-
raphy are shown in Figure SIII.28, and the degradation curves
are shown in Figure 1C. Under this condition, AON 6 (k )
0.0001 min-1), with modification Type II in which 6′-OH
pointed away from the 3′-phosphate (C6′-S, d6′O-P ≈ 5.5 Å,
the molecular model is shown in Figure SIV.2) is 270 times
more stable than AON 14 (k ) 0.027 ( 0.0019 min-1) with
modification IV in which C6′-OH pointing toward the 3′-
phosphate (C6′-R, d6′O-P ≈ 4.4 Å). On the other hand, if AON
14 was incubated in the same buffer but without enzyme, it
was totally stable. This result ruled out any possibility that a
pH 8.0 buffer used in the SVPDE digestion can mediate any
(47) Zhou, C.; Plashkevych, O.; Chattopadhyaya, J. Org. Biomol. Chem. 2008,
Epub ahead of print; DOI: 10.1039/B813870B.
J. Org. Chem. Vol. 74, No. 1, 2009 127

Zhou et al.
FIGURE 2. Autoradiograms of 20% denatured PAGE showing the stability of 5′-32P-labeled AONs in human blood serum. (A) Carba-LNA derivatives
modified AONs. (B) Carba-ENA derivatives modified AONs.
cleavage of phosphate in Type IV modified AON 14 by
intramolecular transesterification.
modified AON 49 is more stable than all other AONs reported
here (Figure 2). It therefore appeared that in human blood serum
the relative order of stability of modified AONs was nearly the
same as that found upon treatment with 3′-exonuclease.
7. Lessons from Modifications in the AON versus
Target Affinity vis-a`-vis Nuclease Resistance. From combina-
tion of the target affinity and nuclease resistance of modified
AONs 2-52, the following conclusions can be obtained: (i) In
the center of the minor groove, hydrophobic and steric
modification is not favored. This type of modification causes
obvious loss of target affinity toward RNA. (ii) At the edge of
the minor groove, hydrophobic and steric modification are
attractive in that this type of modification can result in
considerable improvement on enzymatic stability, at the same
time without loss of target affinity. (iii) Hydrophilic modification
close to phosphate linkage makes the phosphate too labile
toward the nuclease, and so this type of modification should be
avoided.
6. Stability of Carba-LNA and -ENA Modified AONs
in Human Blood Serum. Blood serum stability of AONs with
a single modification at position T13 (Table 1) has also been
tested. The AONs (32P-labeled at 5′-end) were incubated with
human blood serum (male, Type AB) up to 48 h at 21°C and
aliquots were taken out at regular time intervals and analyzed
by 20% denaturing PAGE. The gel pictures obtained by
autoradiography are shown in Figure 2. Because of the presence
of alkaline phosphatase in serum that removed the 5′-end 32P-
label gradually, it was impossible to get the quantified data from
the gel picture to compare their relative blood serum stability.
Comparison of the gel pictures shows that the [6′S-OH,7′S-Me]-
carba-LNA modified (Type II) AONs 6, [6′S-OH,7′R-Me]-
carba-LNA modified (Type III) AON 10, [6′R/S-Me,7′S-Me]-
carba-LNA modified (Type VI) AON 22, and [6′S-OH/Me,7′S-
Me]-carba-LNA modified (Type VII) AON 26 were more stable
than [7′-Me]-carba-LNA modified (Type I)28 AON 2 in blood
serum, whereas [6′R-OH,7′S-Me]-carba-LNA modified (Type
IV) AONs 14 and [6′R-OH,7′R-Me]-carba-LNA modified (Type
III) AON 18 were much less stable than AON 2 upon blood
serum treatment. For AONs containing carbocylic-ENA deriva-
tives, their relative stabilities in human blood serum were as
follows: [6′R-Me,8′S-Me]-carba-ENA modified AON 49, [6′S-
Me,8′S-Me]-carba-ENA modified AON 45 > [8′S-Me]-carba-
LNA modified AON 33 > [6′S-OH,8′S-Me]-carba-ENA modi-
fied AON 37, [6′R-OH,8′S-Me]-carba-ENA modified 41 >
native DNA. Comparison of all modified craba-LNA and carba-
ENA modified AONs, however, showed that the carba-ENA
For AONs with [6′S-OH,7′S-Me]-carba-LNA (Type II) and
[6′S-OH/Me,7′S-Me)-carba-LNA (Type VII) modifications, their
affinity toward complementary RNA (Tm, +4 to +5 °C/
modification compared to the native counterpart) are similar to
that of AONs that have LNA modification (Tm, +5°C/modifica-
tion under the same conditions).28 At the same time, they are
enzymatically 2-3 times more stable than AONs with the [7′-
Me]-carba-LNA (Type I) modification, which are known to be
much more stable than LNA modified AONs.28 Hence, Type
II and VII modified AONs are deemed to be excellent
candidates for antisense therapeutics.
128 J. Org. Chem. Vol. 74, No. 1, 2009

Modulation of Antisense Properties
FIGURE 3. Escherichia coli RNase H1 promoted cleavage pattern of AONs 1-52/RNA duplexes. Vertical arrows show the RNase H cleavage
sites, with the relative length of the arrow showing the extent of the cleavage. The square boxes show the stretch of the modification, which is
resistant to RNase H1 cleavage, thereby giving footprints.
8. RNase
H
Activity of Carba-LNA and -ENA
curves to single-exponential decay functions gave the reaction
rates that are shown as bar plots (Figure 4). The main
observations are as follows: (i) cleavage rate is site-dependent.
Exclusively, the AONs/RNA having the modification at posi-
tions 6 and 8 from the 3′-end in the AON strands have lower
cleavage rates. Moreover, AONs with the modification at
position 8 generally give the lowest degradation rates. On the
contrary, cleavage rates for AONs/RNA with modification at
the positions 3 and 10 from the 3′-end of the AON strands
generally were similar to or even faster than that of the native
counterpart. (ii) Cleavage efficiency is not very sensitive to the
nature of substituents in the carbocyclic moiety especially for
the carba-ENA derivatives. Cleavage of RNA in AONs 37-52/
RNA was as efficient as the cleavage of RNA in AONs 33-36/
RNA. For carba-LNA derivatives, [6′S-OH,7′S-Me]-carba-LNA
(Type II) modfied AONs 6-9, [6′S-OH,7′R-Me]-carba-LNA
(Type III) modified AONs 10-13, and [6′S-OH/Me,7′S-Me]-
carba-LNA (Type VII) modified AONs 26-29 elicit slightly
less recruitment of RNase H.
Modified AONs with the Complementary RNA Is Very
Similar to That of the Native Counterpart. In the antisense
strategy, RNase H recruitment is a very important property when
the modified AONs are bound to the target RNA in heteroduplex
form. The [7′-methyl]-carba-LNA and [8′S-methyl]-carba-ENA
were reported to be able to elicit RNase H activity and the
relative RNA cleavage rates were very similar to that of the
native counterpart.28 It was known that RNase H binds to
the DNA/RNA hybrid in the minor groove.48 In the present
study, all of the modified compounds have their substituents
located in the minor groove and so are ideal substrates to
illustrate how the lipophilic versus hydrophilic or steric nature
of the substituents in the modified AONs affects the efficiency
of RNase H elicitation. Hence, native AON 1 and all of the
modified AONs 2-52 were hybridized with 5′-32P labeled
15mer complementary RNA at 21 °C and incubated with
Escherichia coli RNase H1. Aliquots were taken after 5, 10,
15, 30, and 60 min and analyzed by denaturing PAGE. The
results are shown in Fig SIII.29.
For cleavage of AON/RNA hybrids by RNase H, the cleavage
footprint pattern is affected primarily by the global conformation
of AON/RNA hybrid. AONs containing 2′,4′-bridged modifica-
tion, such as LNA, carba-LNA, aza-ENA, and carba-ENA,
generally showed identical cleavage footprint patterns.28 This
is presumably caused by the local conformational change
imparted by the constrained nature of the modifications. This
conformational change leads to a stretch of ca. 5 bp, starting
from the modification site toward the 5′-end in the AON strand,
not accessible to RNase H. In the present study, all of the
modified AON/RNA hybrids have shown a similar cleavage
pattern, suggesting that the hydroxyl or methyl substitution in
the carbocyclic moiety of carba-LNA and carba-ENA does not
lead to significant conformational disruption.
All modified AONs 2-52/RNA hybrids were found to be
excellent substrates for RNase H. The cleavage pattern was not
affected by the nature of different substituents but depended
upon the site of the modification. For example, despite the fact
that all AONs 6/10/14/18/22/26/33/37/41/45/49 have different
substituents in the carbocyclic moiety at position 3 from the
3′-end, they showed cleavage patterns similar to those of [7′-
Me]-carba-LNA (Type I) modified AON 2 and [8′S-Me]-carba-
ENA (Type IX) modified AON 33 (Figure 3). For all of the
modified AONs 2-52, the cleavage activity of RNase H was
suppressed within a 4-5 base pairs long region that starts from
the base opposite the modified T nucleotide, towards the 3′-
end of the RNA strand. The RNase H promoted cleavage has a
preference at the middle site A8 of the RNA strand for native
AON 1/RNA duplex. If A8 is included within the suppressed
region, the major cleavage site shifts to the edges of the
suppressed region.
In the native AON 1/RNA hybrid, the RNase H1 cleavage
between position 5 and position 13 from the 5′-end of the RNA
strand and have a strong preference at position 8 (Figure 3).
However, if position 8 is included in the stretch of modification,
a new preferred cleavage site appeared in the end of the
modification stretch. Though the sequence in the new preferred
cleavage site is different from that at position 8, this sequence
difference should contribute limitedly to cleavage rate variation
The cleavage rates were determined by autoradiography of
gels and subsequently by plotting the uncleaved RNA fraction
as a function of time (Figure SIII.29). Fitting the degradation
(48) Zamaratski, E.; Pradeepkumar, P. I.; Chattopadhyaya, J. J. Biochem.
Biophys. Methods 2001, 48, 189–208.
J. Org. Chem. Vol. 74, No. 1, 2009 129

Zhou et al.
FIGURE 4. Bar plots of the observed cleavage rates of the RNase H1 promoted degradation of RNA in AON 1-52/RNA hybrid duplexes. The
degradation curves are shown in Figure SIII.33.
becauseRNaseH1onlydisplaysminimalsequencepreference.49,50
The observation that AONs with the modification at position 6
and especially at position 8 gave much lower degradation rates
by RNase H1 (see Figure 4) suggests that if the original
preferred cleavage site is located in the stretch of modification,
the RNA degradation rates by RNase H1 will be affected.
also highlight the key role of stereochemical orientation of the
substituents in the minor groove of the heteroduplex in
engineering the antisense properties of the carba-modified
AONs.
3. For the RNase H1 mediated RNA cleavage in an AON/
RNA hybrid, the cleavage pattern and rate is not very sensitive
to the nature of substitution in the carbocyclic moiety of the
carba-LNA and carba ENA. The RNase H has strong positional
preference for a given active DNA/RNA hybrid. During design
of the modified antisense oligonucleotides by gapmer technol-
ogy, in order to ensure high cleavage efficiency by RNase H,
the original preferred cleavage site in the stretch of modification
should be avoided.
Conclusions
1. To study how the fine modulation around the phosphodi-
ester moiety in the AON can affect its antisense property such
as target affinity, nuclease resistance, and RNase H elicitation,
carba-LNA and carba-ENA with hydroxyl and/or methyl
substituents attached at the carbocyclic moiety were synthesized.
Various NMR experiments including 1H, 13C, DEPT, one bond
4. Type II and VII modified AONs have better target affinity
and nuclease stability than the Type I carba-LNA. In addition,
Type II and VII modified AONs can also elicit RNase H activity
efficiently, which make them excellent candidates as potential
antisense or RNAi therapeutic agents.
1
1H-13C correlation (HMQC), and long range H-13C HMBC
experiments have been employed to characterize all synthesized
compounds. The configuration of subsituents in the key
intermediates was determined by NOE and was also cor-
3
1
roborated by the JH, coupling constants obtained from H
H
Experimental Section
homodecoupling experiments as well as by ab initio calcula-
tions.
Note: All individual reaction steps, work-up, and product
characterization for compounds 13b, 14b, 15b-d, 16b-d, 17a-d,
19a/b, 20a/b, 23a/b, 24a/b, and 25-48 are given in Supporting
Information (SI Part IV).
2. The relationship between substitution and target affinity
and nuclease resistance of AONs can be summarized as follows:
(i) In the center of the minor groove, the hydrophobic and steric
substituents are not favored. This type of modification decreases
the AON′s target affinity obviously but without significant
improvement on the nuclease resistance. (ii) Steric substitution
on the edge of the minor groove is favored. Modified AONs
with these substituents retain the improved target affinity that
carbocyclic LNA/ENA provide and have much better nuclease
resistance. At the same time, they do not cause any loss of
RNase H elicitation. (iii) Hydrophilic modification close to the
internucleotide phosphate linkage should be avoided because it
makes AONs especially enzymatically unstable. These conclu-
sions confirm the importance of the chemical characteristics of
the substituent-type such as lipophilic vs hydrophobic. They
3,5-Di-O-benzyl-4-C-(1-hydroxy-allyl)-1,2-O-isopropylidene-
r-D-ribofuranose (7). Oxalyl chloride (4.1mL, 46 mmol) was added
to precooled CH2Cl2 (1200 mL) at -78 °C. Then DMSO (5.3 mL,
74 mmol) was added dropwise to the solution over 30 min. After
stirring for 20 min, a solution of 6 (7.4 g, 18.5 mmol) in CH2Cl2
(35 mL) was added dropwise over about 20 min, and the mixture
was kept stirring at -78 °C for another 30 min. Then reaction was
cooled with ice, and DIPEA (22 mL) was added. The mixture was
then allowed to warm to room temperature, and water was added.
The organic layer was separated, washed with water and brine, dried
over MgSO4, and concentrated under reduced pressure to give the
corresponding aldehyde 5 as oil. This oil was dissolved in THF
(220 mL) under nitrogen and cooled to 0 °C. Vinylmagnesium
bromide (1.0 M solution in THF, 39 mL, 37 mmol) was added,
and then it was allowed to warm to room temperature and kept
stirring at this temperature overnight. The reaction was quenched
with water (600 mL) slowly. After recovery of THF, the residue
was diluted with CH2Cl2, washed with saturated NaHCO3, dried
(49) Crooke, S. T.; Lemonidis, K. M.; Neilson, L.; Griffey, R.; Lesnik, E. A.;
Monia, B. P. Biochem. J. 1995, 312, 599–608.
(50) Wu, H.; Lima, W. F.; Crooke, S. T. J. Biol. Chem. 1999, 274, 28270–
28278.
130 J. Org. Chem. Vol. 74, No. 1, 2009

Modulation of Antisense Properties
over MgSO4, and concentrated under reduced pressure. The crude
material was purified by column chromatography on silica gel (0-5
% ethyl acetate in cyclohexane, v/v) to give 7 as a colorless oil
10 (6.7g, 13.6 mmol) was coevaporated thrice with anhydrous
pyridine and dissolved in the same solvent (250 mL). The solution
was cooled to -5 °C, and then phenyl chlorothionoformate was
added dropwise (2.43 mL, 17.6 mmol), while temperature was
maintained at ∼-5 °C during addition. After 3 h of stirring at room
temperature, pyridine was recovered under reduced pressure. The
residue was dissoloved in CH2Cl2 and washed with saturated
solution of NaHCO3 twice. The organic layer was separated, dried
over MgSO4, and concentrated. Crude product was chromato-
graphed over silica gel (0-2 % methanol in CH2Cl2, v/v) to give
1
(7.1 g, 84%). H NMR (600 MHz, CDCl3): δ 7.4-7.2 (10H, m,
Bn), 6.05 (1H, m, H7), 5.81 (1H, d, JH1,
) 4.2 Hz, H1), 5.37
H2
(1H, dd, J ) 2.4 Hz, 17.4 Hz, H8), 5.19 (1H, dd, J ) 2.4 Hz,
11.4Hz, H8), 4.98 (1H, m, H6), 4.87(1H, d, Jgem ) 11.4 Hz,
CH2Bn), 4.70 (1H, dd, JH1,H2 ) 4.2 Hz, JH2,H3 ) 5.4 Hz, H2), 4.50
(2H, dd, Jgem ) 11.4 Hz, 11.4Hz, CH2Bn), 4.47 (1H, d, JH3,H2
)
5.4 Hz, H3) 4.37 (1H, d, Jgem ) 11.4 Hz, CH2Bn), 3.76 (1H, d,
Jgem ) 10.8 Hz, H5), 3.56 (1H, m, 6-OH), 3.52 (1H, d, Jgem ) 10.8
Hz, H5), 1.60 (3H, s, CH3), 1.36 (3H, s, CH3); 13C NMR (150 MHz,
CDCl3): δ 139.9 (Bn), 138.7 (Bn), 137.0 (C7) 130.4, 130.1, 129.8,
129.3 (aromatic), 117.0 (C8), 115.0 (isopropyl), 106.7 (C1), 88.9
(C4), 80.7 (C2), 80.4 (C3), 75.3 (CH2Bn), 74.7 (CH2Bn), 73.9 (C6),
71.1 (C5), 28.6 (CH3), 28.2 (CH3). MALDI-TOF m/z: [M + Na]+
449.1, calcd 449.2.
1
11 as white foam (8.1 g, 92%). H NMR (600 MHz, CDCl3): δ
9.18 (1H, s, NH), 7.54 (1H, s, H6), 7.27-7.40 (13H, m, aromatic),
7.12 (2H, d, J ) 7.80 Hz, aromatic), 6.43 (1H, d, JH1′,H2′ ) 5.4 Hz,
H1′), 5.96 (1H, m, H7′), 5.49 (1H, app t, J ) 5.4 Hz, H2′), 5.39
(1H, dd, J ) 1.80 Hz, 17.4 Hz, H8′), 5.26 (1H, dd, J ) 1.80 Hz,
10.8Hz, H8′), 4,91(1H, d, Jgem ) 11.4 Hz, CH2Bn), 4.72 (1H, m,
H6′), 4.66 (1H, d, JH3′,H2′ ) 5.4 Hz, H3′), 4.47-4.56 (3H, m,
CH2Bn), 3.82 (1H, d, Jgem ) 10.8 Hz, H5′), 3.71 (1H, d, Jgem
)
1-[4-C-(1-Acetyl-allyl)-2,6-di-O-acetyl-3,5-di-O-benzyl-ꢀ-D-ri-
bofuranosyl]-thymine (9). Acetic anhydride (36 mL, 380 mmol)
and acetic acid (175 mL) were added to 7 (12.5 g, 29 mmol). The
mixture was cooled, and triflic acid (0.2 mL, 1.4 mmol) was added.
After stirring for 2 h at room temperature, the reaction was quenched
with cold saturated NaHCO3 solution and extracted with CH2Cl2.
The organic layer was separated, dried over MgSO4, and evaporated
to give crude product 8, which was coevaporated with anhydrous
CH3CN thrice and dissolved in the same solvent. Thymine (4.31
g, 34 mmol) and N,O-bis(trimethylsilyl)acetamide (16 mL, 65
mmol) were added to the solution and refluxed for 45 min until
the suspension became a clear solution. Then reaction mixture was
cooled to 0 °C, and TMSOTf (6.18 mL, 34 mmol) was added
dropwise, followed by reflux overnight. CH3CN was evaporated
under reduced pressure. To the residue was added saturated NH4Cl
solution and the mixture was extracted with CH2Cl2. The organic
layer was dried with MgSO4, evaporated, and chromatographed over
silica gel (1-3 % methanol in CH2Cl2, v/v) to give 9 as a white
foam (14 g, 24 mmol, 91%). 1H NMR (270 MHz, CDCl3): δ 8.73
(1H, s, NH), 7.42 (1H, s, H6), 7.22-7.35 (10H, m, Bn), 6.31 (1H,
d, JH1′,H2′ ) 6.56 Hz, H1′), 5.94 (1H, m, H7′), 5.79 (1H, d, JH6′,H7′
) 4.80 Hz, H6′), 5.48 (1H, app t, J ) 6.31 Hz, 12.0 Hz, H2′),
5.14-5.28 (2H, m, H8′), 4,52 (5H, m, 2 x CH2Bn, H3′), 3.76 (2H,
s, H5′, H5′′), 2.04 (3H, s, acetyl-CH3), 1.90 (3H, s, acetyl-CH3),
1.47 (3H, s, T-CH3). 13C NMR (67.5 MHz, CDCl3): δ 170.4 (CdO,
acetyl), 168.6 (CdO, acetyl), 163.5 (C4), 150.5 (C2), 137.0
(aromatic), 135.7 (C6), 132.0 (C7′), 128.8, 128.6, 128.2, 128.1,127.8,
127.6 (aromatic), 117.6 (C5), 111.6, (C8′), 88.0 (C4′), 87.0 (C1′),
79.1 (C3′), 75.3 (CH2Bn), 75.0 (C2′), 74.0 (CH2Bn), 72.9 (C6′),
72.1 (C5′), 20.8 (acetyl-CH3), 20.7 (acetyl-CH3), 12.0 (T-CH3).
MALDI-TOF m/z: [M + H]+ 579.2, calcd 579.2.
10.8 Hz, H5′′), 2.92 (1H, m, 6′-OH), 1.47 (3H, s, T-CH3). 13C NMR
(150 MHz): δ163.7 (C4), 152.9 (CdS), 150.5, 150.4 (C2), 136.9
(Bn), 136.4 (Bn), 135.3 (C6), 134.4 (C7′) 129.6, 128.9, 128.6, 128.2,
127.6, 126.4, 120.8, (aromatic), 116.4 (C8′), 111.5 (C5), 87.4 (C4′),
85.8 (C1′), 78.6 (C2′), 78.1 (C3′), 75.1 (CH2Bn), 73.6 (CH2Bn),
71.7 (C6′), 69.3 (C5′), 12.0 (T-CH3). MALDI-TOF m/z: [M + Na]+
653.1, calcd 653.2.
(1R,3R,4R,5S,6S,7S)-7-Benzyloxy-1-benzyloxmethyl-6-hydroxyl-
5-methlyl-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]heptane (12a) and
(1R,3R,4R,5R,6S,7S)-7-Benzyloxy-1-benzyloxmethyl-6-hydroxyl-
5-methyl-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]heptane (12b). Com-
pound 11 (4.2 g, 5.5 mmol) was dissolved in 200 mL of anhydrous
toluene to which N2 was purged for half an hour. The mixture was
heated to reflux and Bn3SnH (1.85 mL in 20 mL anhydrous toluene),
AIBN (0.55 g in 20 mL anhydrous toluene) were added dropwise
in 2 h. After 0.5 h, the reaction was found to be incomplete (TLC).
Hence, a second portion of Bn3SnH (0.9 mL in 10 mL toluene)
and AIBN (0.25 g in 10 mL toluene) was added dropwise in 1 h,
and reflux was continued for 1 h. Solvent was evaporated, and
residue was applied to silica short column chromatography (EtOAc/
cyclohexane, 2/8 to 6/4) to give 1.3 g of compound 12a (49%),
1
0.32 g of 12b (12%), and 0.7 g of recovered substrate. 12a: H
NMR (600 MHz, CDCl3): δ 8.87 (1H, broad, NH), 7.72 (1H, s,
H6), 7.34-7.23 (10H, m, aromatic), 5.75 (1H, s, H1′), 4.61-4.42
(4H, m, BnCH2), 4.15 (1H, t, J6′H,6′OH ) 11.6 Hz, J6′H,7′H ) 9.2 Hz,
H6′), 4.05 (1H, s, 3′H), 3.91(1H, d, Jgem ) 11.0 Hz, 5′H), 3.81
(1H, d, Jgem ) 11.0 Hz, 5′′H), 2.72 (1H, m, H7′), 2.66 (1H, d,
J2′,H7′ ) 4.5 Hz, 2′H), 2.25 (1H, d, J6′H,6′OH ) 11.6 Hz, 6′OH), 1.50
(s, 3H, T-CH3), 1.15 (3H, d, J7′-CH3, H7′ ) 7.6 Hz, 7′CH3).13C NMR
(150 MHz, CDCl3): δ 164.0 (C4), 149.9(C2), 137.6, 137.0, 136.0
(C6), 128.6, 128.5, 128.1, 127.9, 127.6, 109.6 (C5), 89.1 (C4′),
83.8 (C1′), 76.8 (C3′), 73.9 (BnCH2), 72.1 (BnCH2), 71.9 (C6′),
66.2 (C5′), 47.8 (C2′), 33.1 (C7′), 12.1 (T-CH3), 8.4 (7′CH3).
1-[3,5-Di-O-benzyl-2,6-di-hydorxy-4-C-(1-hydroxy-allyl)-ꢀ-D-
ribofuranosyl]-thymine (10). Compound 9 (6.9 g, 12 mmol) was
treated with 30% methylamine solution in ethanol (240 mL) at room
temperature overnight. Then reaction mixture was extracted with
CH2Cl2, washed with NaHCO3 solution, and dried over MgSO4.
After evaporation of solvent, the residue was chromatographed over
silica gel (0-2.5 % methanol in CH2Cl2, v/v) to give 10 (6.8 g,
100%) as a white solid. 1H NMR (600 MHz, CDCl3): δ 9.92 (1H,
s, NH), 7.55 (1H, s, H6), 7.26-7.36 (10H, m, Bn), 6.08 (1H, d,
1
MALDI-TOF m/z: [M + Na]+ 501.221, calcd 501.200. 12b: H
NMR (600 MHz, CDCl3): δ 8.71 (1H, s, NH), 7.75 (1H, s, H6),
7.22-7.34 (10H, m, aromatic), 5.50 (1H, s, H1′), 4.40-4.62 (4H,
m, BnCH2), 4.04 (1H, s, 3′H), 4.00 (1H, dd, J6′H,6′OH ) 11.0 Hz,
J6′H,7′H ) 3.7 Hz, 6′H), 3.93 (1H, d, Jgem ) 11.0 Hz, 5′H), 3.83
(1H, d, Jgem ) 11.0 Hz, 5′′H), 2.55 (1H, s, 2′H), 2.23 (1H, d, J6′H,6′OH
) 11.0 Hz, 6′OH), 1.77 (m, 1H, 7′H), 1.47 (3H, s, T-CH3), 1.31
(3H, d, J7′CH3,7′H ) 7.3 Hz, 7′CH3).13C NMR (150 MHz, CDCl3): δ
163.8 (C4) 149.9 (C2), 137.5, 136.8, 135.9, 128.6, 128.5, 128.1,
128.0, 127.9, 127.5, 109.5 (C5), 88.9 (C4′), 88.6 (C1′), 81.0 (C6′),
79.0 (C3′), 73.9 (BnCH2), 72.4 (BnCH2), 65.8 (C5′), 45.3 (C2′),
42.1 (C7′), 18.1 (7′CH3), 12.0 (T-CH3). MALDI-TOF m/z: [M +
Na]+ 501.222, calcd 501.200.
JH1′,H2′ ) 5.4 Hz, H1′), 5.94 (1H, m, H7′), 5.32(1H, dt, JH7′,H8′
)
17.4 Hz, H8′), 5.19(1H, dt, JH7′,H8′′ ) 10.6 Hz, H8′′), 5.01 (1H, d,
CH2Bn), 4.71 (1H, m, H6′), 4.70 (1H, d, CH2Bn), 4.52 (3H, s, H2′,
CH2Bn), 4.40 (1H, d, JH2′,H3′ ) 5.4 Hz, H3′), 3.80(1H, d, Jgem
)
10.8 Hz, H5′), 3.66(1H, d, Jgem ) 10.8 Hz, H5′′), 2.89 (d, 1H, J )
4.2 Hz, OH), 1.47(3H, s, T-CH3). 13C NMR (150 MHz, CDCl3): δ
163.9 (C4), 151.3 (C2), 137.4, 137.2, 135.9, 135.3, 128.81, 128.7,
128.4, 128.3, 128.1, 127.6, 116.2, 111.0 (C5), 89.4 (C1′), 87.7 (C4′),
79.4 (C3′), 75.3 (C2′), 74.7 (Bn), 73.8 (Bn), 72.6 (C6′), 70.5 (C5′),
12.10 (T-CH3). MALDI-TOF m/z: [M + Na]+ 517.2, calcd 517.2.
1-[3,5-Di-O-benzyl-4-C-(1-hydroxy-allyl)-6-hydroxyl-2-O-phe-
noxythiocarbonyl-ꢀ-D-ribofuranosyl]-thymine (11). Compound
(1R,3R,4R,5S,7S)-7-Benzyloxy-1-benzyloxmethyl-5-methyl-6-
one-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]heptane (13a). Com-
pound 12a (551 mg, 1.15 mmol) was dissolved in anhydrous CH2Cl2
(3 mL), and 15% Dess-Martin periodinane in CH2Cl2 (3.2 mL,
1.5 mmol) was added dropwise under nitrogen. After stirring at
J. Org. Chem. Vol. 74, No. 1, 2009 131

Zhou et al.
room temperature for 2 h, the reaction mixture was diluted with
CH2Cl2 and filtered through celite. The filtrate was washed with
saturated Na2S2O3 solution and NaHCO3 successively and dried
over MgSO4. After evaporation of solvent under reduced pressure
pure ketone 13a was obtained quantitatively. 1H NMR (500 MHz,
CDCl3): δ 8.72 (1H, broad, NH), 7.70 (1H, s, H6), 7.35-7.19 (10H,
m, aromatic), 5.76 (1H, s, H1′), 4.64-4.49 (4H, m, BnCH2), 4.26
(1H, s, H3′), 4.00 (1H, d, Jgem ) 11.5 Hz, H5′), 3.85 (1H, d, Jgem
) 11.5 Hz, H5′′), 3.08 (1H, m, H2′), 2.86 (1H, m, H7′), 1.54 (3H,
s, T-CH3), 1.33 (3H, d, J7′,7’Me ) 7.5 Hz, 7′-Me). 13C-NMR (125
MHz, CDCl3): δ 206.6 (CdO), 163.8 (C4), 149.8 (C2), 137.3,
136.4, 135.7, 128.7, 128.6, 128.3, 128.2, 127.9, 127.6, 109.9 (C5),
86.3 (C4′), 84.0(C1′), 75.7 (C3′), 74.1, 72.3 (BnCH2), 63.2 (C5′),
46.4 (C2′), 40.3 (C7′), 12.1 (T-Me), 9.5 (7′-Me). MALDI-TOF m/z:
[M + Na]+ 499.208, calcd 499.184.
and the mixture was refluxed for 2 h. The suspension was filtered
over celite, and the organic phase was evaporated to give crude
15a′, which was co-evaporated twice with anhydrous pyridine and
dissolved in the same solvent (5 mL). 4,4′-Dimethoxytrityl chloride
(178 mg, 0.53 mmol) was added, and the mixture was stirred for
2 h at room temperature. Then solvent was removed, and the residue
was diluted with CH2Cl2, washed with saturatured NaHCO3, and
dried over MgSO4. After evaporation of solvent, the residue was
subjected to column chromatography on silica gel (methanol in
CH2Cl2 containing 1% pyridine, 0.5-1.5%, v/v) to obtain 16a (240
1
mg, 70 %). H NMR (500 MHz, CDCl3): δ 7.93 (2H, d, J ) 8.5
Hz, aromatic), 7.78 (1H, s, H6), 7.47-6.82 (15H, m, aromatic),
5.89 (1H, s, H1′), 5.41 (1H, d, J ) 9.5 Hz, H6′), 4.31 (1H, d, J )
1.5 Hz, H3′), 3.78 (6H, s, MeO), 3.69 (1H, dd, Jgem ) 11.5 Hz,
H5′), 3.52 (1H, dd, Jgem ) 11.5 Hz, H5′′), 3.09 (1H, m, H7′), 2.64
(1H, m, H2′), 2.44 (3H, s, tol-CH3), 1.45 (3H, s, T-CH3), 1.15 (3H,
d, J ) 7.5 Hz, 7′-CH3). 13C NMR (125 MHz, CDCl3): δ 166.2,
163.9, 158.6, 149.6 (C2), 144.2, 139.8, 135.4, 135.3, 135.2, 129.9,
129.7, 129.1, 128.0, 127.9, 127.1, 113.3, 109.6 (C5), 88.6, 87.0,
83.7, 73.9, 72.5, 59.6, 55.1, 46.5, 32.4, 21.5, 12.2, 8.5 (7′-Me).
MALDI-TOF m/z: [M + K]+ 757.3, calcd 757.3.
(1R,3R,4R,5S,6S,7S)-7-Benzyloxy-1-benzyloxmethyl-6-hydroxyl-
5,6-dimethyl-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]heptane (18a)
and (1R,3R,4R,5S,6R,7S)-7-Benzyloxy-1-benzyloxmethyl-6-hy-
droxyl-5,6-dimethyl-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]hepta-
ne (18b). MeMgI (3 M in ether, 3.48 mL, 10.45 mmol) was added
dropwise to the solution of compound 13a (0.87 g, 1.82 mmol) in
anhydrous THF (30 mL) at 0 °C under nitrogen. After stirring at
room temperature for 3 h, the reaction was quenched with saturated
NH4Cl solution and extracted with CH2Cl2. The organic layer was
separated, dried over MgSO4, and concentrated. Residue was
chromatographed on silica gel (methanol in CH2Cl2, 0-2%, v/v)
to give 18a (391 mg, 43%) and 18b (157 mg, 18%). 18a: 1H NMR
(500 MHz, CDCl3): δ 8.62 (1H, brs, NH), 7.71 (1H, s, H6), 7.36-
7.25 (10H, m, aromatic), 5.72 (1H, s, H1′), 4.63-4.45 (4H, m,
CH2Bn), 4.06 (1H, s, H3′), 3.89 (2H, s, H5′, H5′′), 2.81 (1H, s,
6′-OH), 2.62(1H, d, JH2′,H7′ ) 3.7 Hz, H2′), 2.41(1H, m, H7′), 1.51
(3H, s, T-CH3), 1.33 (3H, s, 6′-Me), 1.13 (3H, d, J ) 7.6 Hz, 7′-
CH3). 13C NMR (125 MHz, CDCl3): δ 163.8 (C4), 149.7 (C2),
137.4, 136.9, 136.0 (C6), 128.6-127.6 (aromatic), 109.5 (C5), 89.8
(C4′), 82.9 (C1′), 77.2, 76.7 (C3′, C6′), 74.0, 72.2 (CH2Bn), 64.6
(C5′), 48.2 (C2′), 40.1 (C7′), 23.6 (6′-CH3), 12.0 ( T-CH3), 8.3
(7′-CH3). MALDI-TOF m/z: [M + Na]+ found 515.231, calcd
515.215. 18b: 1H NMR (500 MHz, CDCl3): δ 8.66 (1H, brs, NH),
7.69 (1H, s, H6), 7.34-7.23 (10H, m, aromatic), 5.53 (1H, s, H1′),
(1R,3R,4R,5S,6R,7S)-7-Benzyloxy-1-benzyloxmethyl-6-hydroxyl-
5-methyl-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]heptane (14a). The
ketone 13a (548 mg, 1.15 mmol) was dissolved in 95% ethanol
(15 mL), and NaBH4 (47 mg, 1.24 mmol) was added in portions.
The mixture was allowed to stir at room temperature for 2 h. Then
solvent was removed, and the residue was extracted with CH2Cl2.
The organic layer was washed with saturated NaHCO3, dried over
MgSO4, and evaporated to give crude product, which was subjected
to short column chromatography on silica gel (ethyl acetat in
cyclohexane, 20-60% v/v) to give 14a (297 mg, 54%) and 219
1
mg of 12a (40%). H NMR (500 MHz, CDCl3): δ 8.75 (broad,
H3), 7.72 (s, 1H, H6), 7.36-7.22 (10H, m, aromatic), 5.59 (1H, s,
H1′), 4.65-4.44 (4H, m, BnCH2), 4.17 (s, 1H, H3′), 4.04 (1H, d,
Jgem ) 11.3 Hz, H5′), 3.96 (1H, d, Jgem ) 11.3 Hz, H5′′), 3.32 (1H,
dt, 3J6′H,6′OH ) 12.2 Hz, WJ6′H,3′H ) 1.8 Hz, 3J6′H,7′H ) 3.4 Hz, H6′),
3
2.82 (1H, d, 3J2′H,7′H ) 3.4 Hz, H2′), 2.46 (1H, d, J6′H,6′OH ) 12.2
Hz, 6′OH), 2.31 (1H, m, H7′), 1.52 (3H, s, T-CH3), 1.36 (3H, d,
J7′,7′’Me ) 7.3 Hz, 7′-Me). 13C NMR (125 MHz, CDCl3): δ 163.9
(C4), 149.9 (C2), 137.6, 136.3, 136.1, 128.7, 128.6, 128.5, 128.1,
128.0, 127.9, 109.5 (C5), 86.9 (C4′), 83.7(C1′), 81.0 (C6′), 79.5
(C3′), 73.9, 72.5 (BnCH2), 65.7 (C5′), 47.2 (C2′), 42.2 (C7′), 13.4
(7′-Me), 12.1 (T-Me). MALDI-TOF m/z: [M + H]+ 479.217, calcd
479.218.
(1R,3R,4R,5S,6S,7S)-7-Benzyloxy-1-benzyloxymethyl-5-meth-
yl-6-(4-methylbenzoate)-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]hep-
tane (15a). The compound 12a (291 mg, 0.6 mmol) was coevap-
orated with anhydrous pyridine twice and dissolved in anhydrous
CH2Cl2 (5 mL), to which was added anhydrous pyridine 0.2 mL.
The mixture was cooled to -5 °C, and 4-methyl benzoyl chloride
(0.1 mL, 0.72 mmol) was added. The mixture was allowed to stir
at room temperature for 2 h. Pyridine was recovered under reduced
pressure, and the residue was dissoloved in CH2Cl2. The obtained
solution was washed with a saturated solution of NaHCO3, dried
with MgSO4, and concentrated under reduced pressure to give crude
product, which was subjected to short column chromatography on
silica gel (20-30 % ethyl acetate in cyclohexane, v/v) to give 15a
(278 mg, 78%). 1H NMR (500 MHz, CDCl3): δ 8.67 (s, 1H, NH),
7.99 (2H, d, J ) 8.5 Hz, aromatic), 7.84 (1H, s, H6), 7.38-7.27
(12H, m, aromatic), 5.93 (1H, s, H1′), 5.50 (1H, d, J ) 9.5 Hz,
H6′), 4.63-4.51 (4H, m, CH2Bn), 4.17 (1H, s, H3′), 3.78 (2H, dd,
Jgem ) 11.5 Hz, H5′, H5′′), 3.04 (1H, m, H7′), 2.76 (1H, d, J ) 3.5
Hz, H2′), 2.45 (3H, s, Tol-CH3), 1.51 (3H, s, T-CH3), 1.08 (3H, d,
J ) 8.0 Hz, 7′-CH3). 13C NMR (125 MHz, CDCl3): δ 166.5 (CdO),
163.9 (C4), 149.8 (C2), 144.2, 137.6, 136.9, 136.1, 129.8, 129.2,
128.6, 128.5, 128.7, 128.0, 127.8, 127.6, 126.9, 109.5 (C5), 88.7
(C4′), 83.8 (C1′), 77.5 (C3′), 74.0(C6′), 73.9 (CH2Bn), 72.3
(CH2Bn), 66.3 (C5′), 48.1 (C2′), 32.8 (C7′), 21.7(Tol-Me), 11.9
(T-Me), 8.9 (7′-Me). MALDI-TOF m/z: [M + H]+ 597.2, calcd
597.2.
4.64-4.48 (4H, m, CH2Bn), 4.16 (1H, s, H3′), 4.06 (1H, d, Jgem
)
11.0 Hz, H5′), 3.87 (1H, d, Jgem ) 11.0 Hz, H5′′), 3.35 (1H, s,
6′-OH), 2.71(1H, d, JH2′,H7′ ) 4.3 Hz, H2′), 2.46(1H, m, H7′), 1.51
(3H, s, T-CH3), 1.20 (3H, d, JH7′,7′Me ) 7.6 Hz, 7′-CH3), 1.14 (3H,
s, 6′-Me). 13C-NMR (125 MHz, CDCl3): δ 164.3 (C4), 150.2 (C2),
138.1, 136.6, 136.4 (C6), 129.1, 128.9, 128.8, 128.4, 128.2, 109.8
(C5), 88.3 (C4′), 83.3 (C1′), 79.3 (C6′), 78.8 (C3′), 74.4, 73.1
(CH2Bn), 65.2 (C5′), 48.5 (C2′), 45.2 (C7′), 16.8 (6′-CH3), 12.0 (
T-CH3), 10.6 (7′-CH3). MALDI-TOF m/z: [M + Na]+ 515.249,
calcd 515.215.
(1R,3R,4R,5S,6S,7S)-7-Benzyloxy-1-benzyloxmethyl-5,6-di-
methyl-6-methoxalyloxy-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]hep-
tane (21). Methyl oxalyl chloride (0.1 mL, 1.08 mmol) was added
to the solution of compound 18a (351 mg, 0.72 mmol) in anhydrous
MeCN (5 mL) containing DMAP (175 mg, 1.44 mmol) under
nitrogen. After stirring at room temperature for 2 h, another portion
of methyl oxalyl chloride (0.1 mL, 1.08 mmol) and DMAP (175
mg, 1.44 mmol) was added, and the mixture was stirred for further
2 h. Solvent was removed, and the residue obtained was dissolved
in CH2Cl2, which was washed with saturated NaHCO3 solution,
dried over MgSO4, and evaporated. The residue was purified by
column chromatography on silica gel (20-35 % ethyl acetate in
(1R,3R,4R,5S,6S,7S)-1-(4,4′-Dimethoxytrityloxymethyl)-7-hy-
roxyl -5-methyl-6-(4-methylbenzoate)-3-(thymin-1-yl)-2-oxa-
bicyclo[2.2.1]heptane (16a). To a solution of compound 15a (285
mg, 0.48 mmol) in anhydrous methanol (5 mL) was added 20%
Pd(OH)2/C (0.72 g) and ammonium formate (1.8 g, 28.7 mmol),
1
cyclohexane, v/v) to obtain 21 (393 mg, 94%) as white solid. H
NMR (500 MHz, CDCl3): δ 8.44 (1H, brs, H3), 7.64 (1H, s, H6),
132 J. Org. Chem. Vol. 74, No. 1, 2009

Modulation of Antisense Properties
7.34-7.26 (10H, m, aromatic), 5.77 (1H, s, H1′), 4.67 (4H, m,
CH2Bn), 4.39 (1H, d, Jgem ) 11.5 Hz, H5′), 4.24 (1H, s, H3′), 4.03
(1H, d, Jgem ) 11.5 Hz, H5′′), 3.90 (3H, s, CH3- methyl oxalyl),
2.64 (1H, m, H7′), 2.52 (1H, s, H2′), 1.72 (3H, s, 6′-CH3), 1.54
(3H, s, T-CH3), 1.33 (3H, d, J7′-Me,7′ ) 7.0 Hz, 7′-CH3). 13C NMR
(125 MHz, CDCl3): δ 163.7 (C4), 158.3, 156.5 (CdO methyl
oxalyl), 149.7 (C2), 137.8, 136.9, 135.9 (C6), 128.5-127.7 (aro-
matic), 109.6 (C5), 89.8, 89.7, 83.1 (C1′), 78.7 (C3′), 73.7, 72.5
(CH2Bn), 66.4 (C5′), 53.6 (CH3- methyl oxalyl), 48.1 (C2′), 42.9
(C7′), 20.6 (6′CH3), 12.1 (T-CH3), 8.9 (7′-CH3). MALDI-TOF m/z:
[M + H]+ 579.2, calcd 579.2.
PAGE (20% polyacrylamide with 7 M urea), extracted with 0.3 M
NaOAc, and desalted with a C-18 reverse phase cartridge to give
AONs in >99% purity, and correct masses have been obtained by
MALDI-TOF mass spectroscopy for each of them (Table 1).
RNA was also synthesized by a solid-supported phosphoramidite
approach based on the 2′-O-TEM strategy.51,52
UV Melting Experiments. Determination of the Tm of the AON/
RNA hybrids or AON/DNA duplex was carried out in the following
buffer: 60 mM Tris-HCl (pH 7.5), 60 mM KCl, 0.8 mM MgCl2.
Absorbance was monitored at 260 nm in the temperature range
from 20 to 70 °C using an UV spectrophotometer equipped with a
Peltier temperature programmer with the heating rate of 1 °C/min.
Prior to measurements, the samples (1 µM AON and 1 µM
complementary DNA or RNA mixture) were preannealed by heating
to 80 °C for 5 min followed by slow cooling to 21 °C and 30 min
equilibration at this temperature. The value of Tm is the average of
two or three independent measurements. If error of the first two
measurements was greater than (0.3 °C, the third measurement
was carried out to check if the error is indeed within (0.3 °C;
otherwise it was repeated again.
(1R,3R,4R,5S,6S,7S)-7-Benzyloxy-1-benzyloxmethyl-5,6-di-
methyl-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1] heptane (22a) and
(1R,3R,4R,5S,6R,7S)-7-Benzyloxy-1-benzyloxmethyl-5,6-dimeth-
yl-3-(thymin-1-yl)-2-oxa-bicyclo[2.2.1]heptane (22b). Compound
21 (210 mg, 0.36 mmol) was dissolved in anhydrous toluene (10
mL) and purged with anhydrous nitrogen for 30 min. AIBN (30
mg, 0.18 mmol) and Bu3SnH (0.29 mL, 1.08 mmol) were added to
the mixture, which was refluxed for 0.5 h. The reaction mixture
was cooled to room temperature and evaporated. The residue was
chromatographed over silica gel (20-35% ethyl acetate in cyclo-
hexane, v/v) to obtain 22a and 22b as a mixture (57mg, 33.3%,
32P Labeling of Oligonucleotides. The oligoribonucleotides and
oligodeoxyribonucleotides were 5′-end labeled with 32P using T4
polynucleotide kinase, [γ-32P]ATP, and the standard protocol.
Labeled AONs and the target RNA were purified by QIAquick
Nucleotide Removal Kit, and specific activities were measured using
a Beckman LS 3801 counter.
1
22a/22b ) 4/3). 22a: H NMR (500MHz, CDCl3): δ 8.19 (1H,
brs, NH), 7.81(1H, s, H6), 7.38-7.27 (10H, m, aromatic), 5.71 (1H,
s, H1′), 4.65-4.43 (4H, m, CH2Bn), 4.06 (1H, s, H3′), 3.79 (1H, d,
Jgem ) 11.0 Hz, H5′), 3.70 (1H, d, Jgem ) 11.0 Hz, H5′′), 2.72 (1H,
m, H7′), 2.57(1H, d, JH2′,H7′ ) 4.0 Hz, H2′), 2.30 (1H, m, H6′),
1.54 (3H, s, T-CH3), 1.14 (3H, d, J7′-Me,7′ ) 7.6 Hz, 7′-CH3), 0.92
(3H, d, JH6′, 6′-CH3 ) 7.3 Hz, 6′-CH3). 13C NMR (125 MHz, CDCl3):
δ 163.8 (C4), 149.73 (C2), 137.9, 137.6, 136.5 (C6), 128.5-127.5
(aromatic), 109.1 (C5), 89.8 (C4′), 83.7 (C1′), 77.6 (C3′), 73.8,
71.9 (CH2Bn), 67.2 (C5′), 48.4 (C2′), 35.4 (C6′), 31.4 (C7′), 12.2
SVPDE Degradation Studies. Stability of the AONs toward
3′-exonucleases was tested using phosphodiesterase I from Crotalus
adamanteus (obtained from USB corporation, Cleveland, OH). All
reactions were performed at 3 µM DNA concentration (5′-end 32
P
labeled with specific activity 80 000 cpm) in 100 mM Tris-HCl
(pH 8.0) and 15 mM MgCl2 at 21 °C. An exonuclease concentration
of 6.7 or 2.2 ng/µL was used for digestion of oligonucleotides.
Total reaction volume was 30 µL. Aliquots (3 µL) were taken at
proper time points and quenched by the addition of stop solution
(4 µL) [containing 0.05 M EDTA, 0.05% (w/v) bromophenol blue,
and 0.05% (w/v) xylene cyanole in 80% formamide]. Reaction
progress was monitored by 20% denaturing (7 M urea) PAGE and
autoradiography.
1
(T-CH3), 10.1 (7′-CH3), 8.0 (6′-CH3). 22b: H NMR (500 MHz,
CDCl3): δ 8.19 (1H, brs, NH), 7.72 (1H, s, H6), 7.38-7-27 (10H,
m, aromatic), 5.72 (1H, s, H1′), 4.65-4.43 (4H, m, CH2Bn), 4.03
(1H, s, H3′), 3.85 (2H, s, H5′, H5′′), 2.64 (1H, d, JH2′,H7′ ) 4.0 Hz,
H2′), 2.21 (1H, m, H7′), 1.58 (3H, s, T-CH3), 1.55 (1H, m, H6′),
1.28 (3H, d, J7′-Me,7′ ) 7.3 Hz, 7′-CH3), 1.07 (3H, d, J6′,6′-CH3 ) 7.3
Hz, 6′-CH3). 13C NMR (125 MHz, CDCl3) : δ 163.8 (C4), 149.70
(C2), 137.8, 137.5, 136.4 (C6), 128.5-127.5 (aromatic), 109.1 (C5),
89.1 (C4′), 83.8 (C1′), 79.6 (C3′), 73.8, 71.9 (CH2Bn), 66.8 (C5′),
47.2 (C6′), 38.2 (C7′), 15.3 (6′-CH3), 13.9 (7′-CH3), 12.2 (T-CH3).
MALDI-TOF m/z: [M + Na]+ 499.236, calcd 499.221.
The total percentage of integrated AONs (13-15mer) were
plotted against time points to give the digestion curves, and the
pseudo-first-order reaction rates could be obtained by fitting the
curves to single-exponential decay functions.
General Procedure for Phosphoramidite Synthesis. The
substrate (1 equiv) was dissolved in anhydrous THF to which was
added DIPEA (5 equiv) followed by 2-cyanoethyl-N,N-diisopropyl
phosphoramidochloridite (2 equiv) in an ice bath. The reaction
mixture was allowed to warm to room temperature and stirred at
this temperature for 2-4 h. After being quenched with methanol,
the reaction mixture was extracted with ethyl acetate and washed
with saturated NaHCO3 solution, dried over MgSO4, and concen-
trated. The residue was subjected to short column chromatography
on silica gel to give pure phosphoramidite, which was first
precipitated in n-hexane and then dried over P2O5 on vacuum for
3 days before it was used for DNA synthesis.
Oligonucleotide Synthesis and Purification. All AONs were
synthesized using an automated DNA/RNA synthesizer based on
phosphoramidite chemistry. For native A, G, and C building blocks,
fast deprotecting phosphoramidites (Ac for C, iPr-Pac for G, Pac
for A) were used. Standard DNA synthesis reagents and cycles were
used except that 0.25 M 5-ethylthio-1H-tetrazole was used as the
activator and Tac2O as cap A. For incorporating modified nucle-
otides, extended coupling time (10 min comparing to 25 s for native
nucleotides) was used. For AONs 6-13 and 37-40 the deprotec-
tions were carried out in 33% aqueous NH3 for 16 h at 55 °C. For
AONs 14-21 and 41-44 the deprotections were carried out in
33% aqueous NH3 for 72 h at 55 °C. Other oligoes were deprotected
at room temperature by treatment with 33% aqueous NH3 for 12h.
After deprotection, all crude oligoes were purified by denaturing
Stability Studies in Human Blood Serum. AONs at 2 µM
concentration (5′-end 32P-labeled with specific activity 80 000 cpm)
were incubated in 10 µL of human blood serum (male AB, obtained
from Sigma-Aldrich) at 21 °C (total reaction volume was 36 µL).
Aliquots (3 µL) were taken at the proper time points, quenched
with 4 µL of stop solution [containing 0.05 M EDTA, 0.05% (w/
v) bromophenol blue, and 0.05% (w/v) xylene cyanole in 80%
formamide], resolved in 20% polyacrylamide denaturing (7 M urea)
gel electrophoresis, and visualized by autoradiography.
RNase H Digestion Assay. Target 0.1 µM RNA (specific activity
80 000 cpm) and AON (2 µM) were incubated in a buffer containing
20 mM Tris-HCl (pH 7.5), 20 mM KCl, 10 mM MgCl2, 0.1 mM
EDTA, and 0.1 mM DTT at 21 °C in the presence of 0.08 U of E.
coli RNase H (obtained from USB corporation, Cleveland, OH).
Prior to the addition of the enzyme, reaction components were
preannealed in the reaction buffer by heating at 80 °C for 5 min
followed by slow cooling to 21°C and 30 min equilibration at this
temperature. Total reaction volume was 30 µL. Aliquots of 3 µL
were removed after 5, 10, 15, 30, and 60 min, and the reactions
were terminated by mixing with stop solution [containing 0.05 M
EDTA, 0.05% (w/v) bromophenol blue, and 0.05% (w/v) xylene
(51) Zhou, C.; Honcharenko, D.; Chattopadhyaya, J. Org. Biomol. Chem.
2007, 5, 333–343.
(52) Zhou, C.; Pathmasiri, W.; Honcharenko, D.; Chatterjee, S.; Barman, J.;
Chattopadhyaya, J. Can. J. Chem 2007, 85, 293–301.
J. Org. Chem. Vol. 74, No. 1, 2009 133

Zhou et al.
cyanole in 80% formamide]. The samples were subjected to 20%
7 M urea PAGE and visualized by autoradiography. Pseudo-first-
order reaction rates could be obtained by fitting the digestion curves
to single-exponential decay functions.
Theoretical Calculations. Methyl- and hydroxyl-substituted
carba-LNA nucleosides were investigated in silico using ab initio
technique. The geometry optimizations of the modified nucleosides
were carried out using the GAUSSIAN 98 program package53 at
the Hartree-Fock level using 6-31G**. Relevant vicinal proton 3JH,H
coupling constants have been back-calculated from the correspond-
ing theoretical torsions employing the Haasnoot-de Leeuw-Altona
generalized Karplus equation43,44 taking into account ꢀ substituent
correction in the form
the Swedish Foundation for Strategic Research (Stiftelsen fo¨r
Strategisk Forskning), and the EU-FP6 funded RIGHT project
(Project no. LSHB-CT-2004-005276) is gratefully acknowl-
edged. Individual authors’ contributions on this work can be
found in SI Part IV.
1
Supporting Information Available: H, 13C, 31P, DEPT,
COSY, HMQC, and HMBC of carba-LNA derivatives (SI part
I) and carba-ENA derivatives (SI part II); 1D NOE and 2D
NOESY spectra, experimental and calculated coupling constants
of key intermediates 12, 14, 18, 22, 30, 31, 33, 35, and 36;
autoradiograms of 20% denaturing PAGE of digestion of AONs
in SVPDE. Autoradiograms of 20% denaturing PAGE as well
as degradation curves, showing the cleavage kinetics of target
RNA in AON/RNA hybrides by E. coli RNase H1(SI Part III);
a brief description of Schemes 4 and 5; discussion about
configuration determination on chiral center by NOE experiment,
how the position and stereochemical orientation of various
substitutents on the carbocyclic moiety affects the Tm’s; RNA/
DNA duplex model to show the location of subsitituents in the
carbocylic moeity in the duplex (SI Part IV). This material is
available free of charge via the Internet at http://pubs.acs.org.
3J ) P1 cos2(ꢁ) + P2 cos(ꢁ) + P3 +
∆ꢂgroup P +P cos2 ꢃ ꢁ + P |∆ꢂgroup
| (1)
(
(
))
∑
i
4
5
i
6
i
where P1 ) 13.70, P2 ) -0.73, P3 ) 0.00, P4 ) 0.56, P5 ) -2.47,
P6 ) 16.90, P7 ) 0.14 (parameters from ref 43), and ∆ꢂigroup
)
∆ꢂiR-substituent - P7 Σ∆ꢂiꢀ-substituent where ∆ꢂi are taken as Huggins
electronegativities.54
Acknowledgment. Generous financial support from the
Swedish Natural Science Research Council (Vetenskapsrådet),
(53) Frisch, M. J., et al. Gaussian 98 (ReVision A.6); Gaussian, Inc: Pittsburgh,
PA, 1998.
(54) Huggins, M. L. J. Am. Chem. Soc. 1953, 75, 4123–4126.
JO8016742
134 J. Org. Chem. Vol. 74, No. 1, 2009