The 4-(N-dichloroacetyl-N-methylamino)benzyloxymethyl group for 2′-hydroxyl protection of ribonucleosides in the solid-phase synthesis of oligoribonucleotides

Article abstract of DOI:10.1021/jo702717g

(Chemical Equation Presented) Emerging RNA-based technologies for controlling gene expression have triggered a high demand for synthetic oligoribonucleotides and have motivated the development of ribonucleoside phosphoramidites that would exhibit coupling

Full text of DOI:10.1021/jo702717g

The 4-(N-Dichloroacetyl-N-methylamino)benzyloxymethyl Group for
2′-Hydroxyl Protection of Ribonucleosides in the Solid-Phase
Synthesis of Oligoribonucleotides
Jacek Cies´lak,† Andrzej Grajkowski,† Jon S. Kauffman,‡ Robert J. Duff,‡ and
Serge L. Beaucage*,†
DiVision of Therapeutic Proteins, Center for Drug EValuation and Research, Food and Drug
Administration, 8800 RockVille Pike, Bethesda, Maryland 20892, and Lancaster Laboratories,
2425 New Holland Pike, Lancaster, PennsylVania 17605-2425
serge.beaucage@fda.hhs.goV
ReceiVed December 20, 2007
Emerging RNA-based technologies for controlling gene expression have triggered a high demand for
synthetic oligoribonucleotides and have motivated the development of ribonucleoside phosphoramidites
that would exhibit coupling kinetics and coupling efficiencies comparable to those of deoxyribonucleoside
phosphoramidites. To fulfill these needs, the novel 4-(N-dichloroacetyl-N-methylamino)benzyloxymethyl
group for 2′-hydroxyl protection of ribonucleoside phosphoramidites 9a-d has been implemented (Schemes
1 and 2). The solid-phase synthesis of AUCCGUAGCUAACGUCAUGG was then carried out employing
9a-d as 0.2 M solutions in dry MeCN and 5-benzylthio-1H-tetrazole as an activator. The coupling
efficiency of 9a-d averaged 99% within a coupling time of 180 s. Following removal of all base-
sensitive protecting groups, cleavage of the remaining 2′-[4-(N-methylamino)benzyl] acetals from the
RNA oligonucleotide was effected in buffered 0.1 M AcOH (pH 3.8) within 30 min at 90 °C. RP-HPLC
and PAGE analyses of the fully deprotected AUCCGUAGCUAACGUCAUGG were comparable to those
of a commercial RNA oligonucleotide sharing an identical sequence. Enzymatic digestion of the RNA
oligomer catalyzed by bovine spleen phosphodiesterase and bacterial alkaline phosphatase revealed no
significant amounts of RNA fragments containing (2′f5′)-internucleotidic phosphodiester linkages or
noteworthy nucleobase modifications.
Introduction
in mammalian cells.2,3 Typically, this process entails the
incorporation of the siRNA into a protein complex including
the endonuclease Argonaute 2, which mediates the cleavage of
the sense strand of the siRNA during activation of the RNA-
induced-silencing complex (RISC).4 The antisense strand of the
siRNA guides the RISC to complementary sequences in target
mRNAs; RISC then cleaves the targeted mRNA at a single site
With the discovery of RNA interference (RNAi) as a cellular
process for silencing the expression of specific genes,1 small-
interfering double-stranded RNAs (siRNAs) of 21-25 base pairs
in length have been efficient in targeting and cleaving mRNAs
* To whom correspondence should be addressed. Tel: (301) 827-5162. Fax:
(301) 480-3256.
† Food and Drug Administration.
(2) Elbashir, S. M.; Harborth, J.; Lendeckel, W.; Yalcin, A.; Weber, K.;
Tuschl, T. Nature 2001, 411, 494-498.
(3) Dorsett, Y.; Tuschl, T. Nat. ReV. Drug DiscoVery 2004, 3, 318-
329.
‡ Lancaster Laboratories.
(1) Fire, A.; Xu, S.; Montgomery, M. K.; Kostas, S. A.; Driver, S. E.;
Mello, C. C. Nature 1998, 391, 806-811.
10.1021/jo702717g CCC: $40.75 © 2008 American Chemical Society
Published on Web 03/08/2008
2774
J. Org. Chem. 2008, 73, 2774-2783

2′-Hydroxyl Protection of Ribonucleosides.
between nucleobases 10 and 11 relative to the 5′-end of the
siRNA antisense strand,4a,c,d thereby resulting in a decreased
expression level of the encoded protein. The RNAi machinery
also serves as an effector for endogenous, noncoding RNAs
known as microRNAs (miRNAs).5 These miRNAs are abundant
and are believed to be important in many biological processes
through regulation of gene expression by controlling the
efficiency of mRNA translation. Unlike an mRNA, a pre-
miRNA is embedded in a ∼70-nucleotide (nt) hairpin structure,
which is cleaved by the RNase III enzyme Dicer to produce a
“mature” miRNA as a ∼21-nt double-stranded RNA.4d,6 Like
siRNAs, miRNAs are assembled into the RISC and guide the
complex to partially complementary mRNA sites by Watson-
Crick base-pairing. The expression of targeted mRNAs is
therefore blocked either by translational repression or by
degradation.7 Although miRNAs do not encode proteins and
the precise molecular function of these biomolecules in mam-
mals is largely unknown, each miRNA has the potential to
regulate hundreds of mRNAs. In order to better understand the
function of miRNAs in vivo, chemically engineered single-
stranded RNA oligonucleotides, termed “antagomirs”, were
developed.8 The silencing of endogenous miRNAs with an-
tagomirs in mice was specific, efficient, and long-lasting. Thus,
the use of antagomirs may lead to a viable therapeutic strategy
in the treatment of cancer, hepatitis, and diabetes considering
the involvement of miRNAs in these diseases has been
demonstrated.8 These emerging RNA-based strategies in the
control of gene expression have created a high demand for
synthetic oligoribonucleotides and has spurred a “renaissance”
in the development of rapid and efficient methods for solid-
phase RNA synthesis. The design and implementation of 2′-
hydroxyl protecting groups that would provide ribonucleoside
phosphoramidites with coupling kinetics and coupling efficien-
cies comparable to those of deoxyribonucleoside phosphora-
midites are key to the production of RNA oligonucleotides in
sufficient quantity and purity for pharmaceutical applications.
Technically, a 2′-hydroxyl protecting group must be stable to
the reagents and conditions used during solid-phase RNA
synthesis, and to the conditions employed for nucleobase and
phosphate deprotection. The 2′-hydroxyl protecting group must,
ultimately, be cleaved from the oligonucleotides under condi-
tions that would not induce formation of (2′f5′)-internucleotidic
phosphodiester linkages and/or RNA chain cleavage. The search
for an optimal 2′-hydroxyl protecting group for ribonucleosides
has been ongoing for decades.9 Particularly noteworthy is the
progress made in the synthesis of RNA oligonucleotides over
the past decade through ribonucleoside phosphoramidites func-
tionalized with the 2-nitrobenzyloxymethyl (1),10 4-nitroben-
zyloxymethyl (2),11 1-[2-(trimethylsilyl)ethoxy]methyl,12 1-(4-
chlorophenyl)-4-ethoxypiperidin-4-yl,13 triisopropylsilyloxymethyl
(3),14 substituted 1-(benzyloxy)ethyl,15 1-(2-cyanoethoxy)]-
ethyl,16 (2-cyanoethoxy)methyl (4),17 levulinyl (5),18 2-tert-
butyldithiomethyl (6),19 acyloxymethyl,20 acylthiomethyl,20 bis-
(2-acetoxyethoxy)methyl (7),21 2-(4-tolylsulfonyl)ethoxymethyl
(8),22 and allyl23 groups for 2′-hydroxyl protection. Although
ribonucleoside phosphoramidites 1-8 exhibit reaction kinetics
and coupling efficiencies in solid-phase RNA synthe-
sis10,11,14,17-19,21,22 comparable to those of DNA phosphoramid-
ites in typical solid-phase DNA synthesis, many ribonucleoside
phosphoramidites do not, presumably because of steric hindrance
created by bulky 2′-O-protecting groups in the vicinity of the
activated phosphoramidite entities. It should also be emphasized
that the implementation of a number of 2′-O-protected ribo-
nucleoside phosphoramidites in solid-phase RNA synthesis
required nucleobase, 5′-hydroxyl, and phosphate protecting
groups different from those typically employed in standard solid-
phase DNA synthesis. These requirements demanded the
development of laborious and often costly strategies for the
preparation of RNA phosphoramidite building blocks. Further-
more, the stability of some 2′-O-protecting groups during solid-
phase synthesis of RNA oligonucleotides is questionable and
certain 2′-O-protecting groups generated side products upon
deprotection, which had to be deactivated to prevent nucleobase
alkylation. In an attempt to circumvent these drawbacks and
(10) (a) Schwartz, M. E.; Breaker, R. R.; Asteriadis, G. T.; deBear, J.
S.; Gough, G. R. Bioorg. Med. Chem. Lett. 1992, 2, 1019-1024. (b) Miller,
T. J.; Schwartz, M. E.; Gough, G. R. In Current Protocols in Nucleic Acid
Chemistry; Beaucage, S. L., Bergstrom, D. E., Glick, G. D., Jones, R. A.,
Eds.; John Wiley & Sons: New York, 2000; Vol. I, pp 2.5.1-2.5.36 and
pp 3.7.1-3.7.8.
(11) Gough, G. R.; Miller, T. J.; Mantick, N. A. Tetrahedron Lett. 1996,
37, 981-982.
(12) (a) Wincott, F. E.; Usman, N. Tetrahedron Lett. 1994, 35, 6827-
6830. (b) Wada, T.; Tobe, M.; Nagayama, T.; Furusawa, K.; Sekine, M.
Tetrahedron Lett. 1995, 36, 1683-1684.
(13) Lloyd, W.; Reese, C. B.; Song, Q.; Vandersteen, A. M.; Visintin,
C.; Zhang, P.-Z. J. Chem. Soc., Perkin Trans. 1 2000, 165-176.
(14) (a) Pitsch, S.; Weiss, P. A.; Jenny, L.; Stutz, A.; Wu, X. HelV. Chim.
Acta 2001, 84, 3773-3795. (b) Pitsch, S.; Weiss, P. A. In Current Protocols
in Nucleic Acid Chemistry; Beaucage, S. L., Bergstrom, D. E., Glick, G.
D., Jones, R. A., Eds.; John Wiley & Sons: New York, 2001; Vol. I, pp
2.9.1-2.9.14 and 3.8.1-3.8.15.
(15) Matysiak, S.; Pfleiderer, W. HelV. Chim. Acta 2001, 84, 1066-
1085.
(16) Umemoto, T.; Wada, T. Tetrahedron Lett. 2004, 45, 9529-9531.
(17) (a) Ohgi, T.; Masutomi, Y.; Ishiyama, K.; Kitagawa, H.; Shiba, Y.;
Yano, J. Org. Lett. 2005, 7, 3477-3480. (b) Shiba, Y.; Masuda, H.;
Watanabe, N.; Ego, T.; Takagaki, K.; Ishiyama, K.; Ohgi, T.; Yano, J.
Nucleic Acids Res. 2007, 35, 3287-3296.
(4) For excellent reviews on the mechanistic aspects of RNAi, see: (a)
Bumcrot, D.; Manoharan, M.; Koteliansky, V.; Sah, D. W. Y. Nature Chem.
Biol. 2006, 2, 711-719. (b) Manoharan, M. Curr. Opin. Chem. Biol. 2004,
8, 570-579. (c) Chiu, Y.-L.; Rana, T. M. Mol. Cell 2002, 10, 549-561.
(d) Kim, D. H.; Rossi, J. J. Nature ReV. Genet. 2007, 8, 173-184.
(5) Bartel, D. P. Cell 2004, 116, 281-297.
(6) Lee, Y.; Ahn, C.; Han, J.; Choi, H.; Kim, J.; Yim, J.; Lee, J.; Provost,
P.; Radmark, O.; Kim, S.; Kim, V. N. Nature 2003, 425, 415-419.
(7) Hammond, S. M. Trends Mol. Med. 2006, 12, 99-101.
(8) Kru¨tzfeldt, J.; Rajewsky, N.; Braich, R.; Rajeev, K. G.; Tuschl, T.;
Manoharan, M.; Stoffel, M. Nature 2005, 438, 685-689 and references
therein.
(9) . Reese, C. B. Org. Biomol. Chem. 2005, 3, 3851-3868. (b) Mu¨ller,
S.; Wolf, G.; Ivanov, S. A. Curr. Org. Synth. 2004, 1, 293-307. (c)
Marshall, W. S.; Kaiser, R. J. Curr. Opin. Chem. Biol. 2004, 8, 222-229.
(d) Reese, C. B. Tetrahedron 2002, 58, 8893-8920. (e) Reese, C. B. In
Current Protocols in Nucleic Acid Chemistry; Beaucage, S. L., Bergstrom,
D. E., Glick, G. D., Jones, R. A., Eds.; John Wiley & Sons: New York,
2000; Vol. I, pp 2.2.1-2.2.24. (f) Pitsch, S. Chimia 2001, 55, 320-324.
(g) Beaucage, S. L.; Iyer, R. P. Tetrahedron 1992, 48, 2223-2311.
(18) Lackey, J. G.; Sabatino, D.; Damha, M. J. Org. Lett. 2007, 9, 789-
792.
(19) Semenyuk, A.; Fo¨ldesi, A.; Johansson, T.; Estmer-Nilsson, C.;
Blomgren, P.; Bra¨nnvall, M.; Kirsebom, L. A.; Kwiatkowski, M. J. Am.
Chem. Soc. 2006, 128, 12356-12357.
(20) Parey, N.; Baraguey, C.; Vasseur, J.-J.; Debart, F. Org. Lett. 2006,
8, 3869-3872.
(21) (a) Marshall, W. S.; Kaiser, R. J. Curr. Opin. Chem. Biol. 2004, 8,
222-229. (b) Scaringe, S. A.; Kitchen, D.; Kaiser, R. J.; Marshall, W. S.
In Current Protocols in Nucleic Acid Chemistry; Beaucage, S. L., Bergstrom,
D. E., Glick, G. D., Jones, R. A., Eds.; John Wiley & Sons: New York,
2004; Vol. I, pp. 2.10.1-2.10.16. (c) Scaringe, S. A.; Wincott, F. E.;
Caruthers, M. H. J. Am. Chem. Soc. 1998, 120, 11820-11821.
(22) (a) Zhou, C.; Honcharenko, D.; Chattopadhyaya, J. Org. Biomol.
Chem. 2007, 5, 333-343. (b) Zhou, C.; Pathmasiri, W.; Honcharenko, D.;
Chatterjee, S.; Barman, J.; Chattopadhyaya, J. Can. J. Chem. 2007, 85,
293-301.
(23) Tanaka, S.; Hirakawa, T.; Oishi, K.; Hayakawa, Y.; Kitamura, M.
Tetrahedron Lett. 2007, 48, 7320-7322.
J. Org. Chem, Vol. 73, No. 7, 2008 2775

Cies´lak et al.
phosphoramidite derivatives (9a,b) is outlined in Scheme 1.
Commercial uridine (10a) and N4-benzoylcytidine (10b) were
mixed with the Markiewicz reagent26 in anhydrous pyridine to
produce the 3′,5′-disilylated ribonucleosides 11a,b in near
quantitative yields. Without further purification, 11a,b were
reacted with DMSO, acetic anhydride and acetic acid under
conditions described in Scheme 1 to give the ribonucleoside
2′-thioacetals 12a,b in isolated yields ranging from 73% to 85%
after silica gel chromatography. The yields of 12a,b are based
on the amounts of 10a,b employed in the preparation of 11a,b
and are somewhat dependent on the ratio of the reactants used
in the thioacetalization of 11a,b. The ratios of DMSO, acetic
anhydride and acetic acid, recommended by Semenyuk et al.,19
were found near optimal in the preparation of 12a,b.
Transacetalization of 12a,b was performed under rigorously
anhydrous conditions, via activation of the thioacetal function
with N-iodosuccinimide and fresh trifluoromethanesulfonic acid
(TfOH),27 in the presence of 4-(N-dichloroacetyl-N-methylami-
no)benzyl alcohol and powdered 5A molecular sieves to produce
the ribonucleoside 2′-acetals 13a,b. Additional TfOH was often
necessary to complete the transacetalization reaction. Crude
13a,b were completely desilylated within 16 h by treatment with
ammonium fluoride19,28 in methanol. The acetals 14a,b were
purified by silica gel chromatography and isolated as pure
amorphic solids in yields averaging 65% relative to the amounts
of 12a,b utilized in the preparation of 13a,b. 5′-O-Dimethox-
ytritylation of 14a,b was carried out in anhydrous pyridine using
a slight molar excess (1.2 equiv) of 4,4′-dimethoxytrityl chloride.
The reaction was monitored by TLC and was found complete
within 2 h. After routine workup, crude 15a,b were purified by
chromatography on silica gel and were isolated as lyophilized
powders in yields of ca. 90%. The phosphinylation of 15a,b
was effected within 2 h when employing 2-cyanoethyl N,N-
diisopropylchlorophosphoramidite and dry Et3N in anhydrous
CH2Cl2. After an aqueous workup, the crude ribonucleoside
phosphoramidites 9a,b were purified by silica gel chromatog-
raphy using C6H6:Et3N (9:1 v/v) as the eluent. The presence of
Et3N in the eluent is necessary to neutralize the inherent acidity
of silica gel and thus prevent decomposition of 9a,b during
chromatography. Purified 9a,b were freed from residual Et3N
by precipitation in cold (-78 °C) hexane. The white precipitates
were dissolved in dry C6H6 and the resulting solutions were
frozen and then lyophilized under high vacuum affording
moisture and Et3N-free 9a,b in yields averaging 80%. However,
depending upon the original amount of residual moisture and
Et3N contaminating 9a,b, an additional round of precipitation
in cold hexane and lyophilization from dry benzene may be
inspired by the work of Gough et al.10,11 and Pfleiderer et al.15,24
on the use of benzyl acetals as 2′-hydroxyl protecting groups
in RNA synthesis, we recently reported the 4-(N-dichloroacetyl-
N-methylamino)benzyloxymethyl group for 2′-hydroxyl protec-
tion of ribonucleoside phosphoramidite 9a, which was employed
in the solid-phase synthesis of the chimeric polyuridylic acid
U19dT.25 RP-HPLC analysis of unpurified [2′-O-[4-(N-methy-
lamino)benzyloxy]methyl U]19dT revealed a high quality prod-
uct. Thus, activation of 9a with 5-ethylthio-1H-tetrazole pro-
duced within 180 s coupling efficiencies comparable to those
obtained with deoxyribonucleoside phosphoramidites. Further-
more, RP-HPLC analysis of unpurified U19dT indicated that
hydrolysis of the 2′-acetals from the oligoribonucleotide pro-
ceeded smoothly under mild acidic conditions without significant
internucleotidic chain cleavage. These encouraging results
prompted us to synthesize the ribonucleoside phosphoramidites
9b-d and assess the coupling kinetics and coupling efficiencies
of these monomers in the solid-phase synthesis of AUC-
CGUAGCUAACGUCAUGG. The results of our investigations
will be the focus of this report.
Results and Discussion
The synthesis of 2′-O-[4-(N-dichloroacetyl-N-methylamino)-
benzyloxy]methyl ribonucleosides (14a,b and 15a,b) and their
(26) (a) Markiewicz, W. T. J. Chem. Res. (S) 1979, 24-25. (b)
Markiewicz, W. T. J. Chem. Res., Miniprint 1979, 181-197.
(27) Veeneman, G. H.; Van Der Marel, G. A.; Van Den Elst, H.; Van
Boom, J. H. Tetrahedron 1991, 47, 1547-1562.
(24) Matysiak, S.; Fitznar, H.-P.; Schnell, R.; Pfleiderer, W. HelV. Chim.
Acta 1998, 81, 1545-1566.
(25) Cies´lak, J.; Kauffman, J. S.; Kolodziejski, M. J.; Lloyd, J. R.;
Beaucage, S. L. Org. Lett. 2007, 9, 671-674.
(28) Mu¨nch, U.; Pfleiderer, W. HelV. Chim. Acta 2003, 86, 2546-2565.
2776 J. Org. Chem., Vol. 73, No. 7, 2008

2′-Hydroxyl Protection of Ribonucleosides.
SCHEME 1. Synthesis of 2′-O-Protected Ribonucleosides 14a,b and 15a,b and Ribonucleoside Phosphoramidites 9a,ba
a
Keys: BP, uracil-1-yl (a) and N4-benzoylcytosin-1-yl (b); R, 4-(N-dichloroacetyl-N-methylamino)benzyl; TIPSiCl2, 1,3-dichloro-1,1,3,3-tetraisopropy-
ldisiloxane; NIS, N-iodosuccinimide; TfOH, trifluoromethanesulfonic acid; DCE, 1,2-dichloroethane; DMTrCl, 4,4′-dimethoxytrityl chloride; 2-CE-DIPCP,
2-cyanoethyl-N,N-diisopropylchlorophosphoramidite.
SCHEME 2. Synthesis of 2′-O-Protected Ribonucleosides 14c,d and 15c,d and Ribonucleoside Phosphoramidites 9c,da
a
Keys: Ade, adenin-9-yl; AdeBz, N6-benzoyladenin-9-yl; GuaiBu, N2-isobutyrylguanin-9-yl; R1, N-dichloroacetyl-N-methylamino; R, 4-(N-dichloroacetyl-
N-methylamino)benzyl.
required to rigorously ensure the absence of moisture and base
contaminants, and guarantee high coupling efficiency of phos-
phoramidite monomers.
condensed with 19c′,d to produce a regioisomeric mixture of
ribonucleoside 3′- and 2′-acetals. The ribonucleoside 2′-acetals
14c′,d were separated by silica gel chromatography29 and were
isolated in yields of 30-35%. Both 3′- and 5′-hydroxyls of 14c′
were transiently protected within 2 h upon reaction with
trimethylsilyl chloride in dry pyridine.30 Benzoylation of the
exocyclic amino group of adenine was then achieved by adding
benzoyl chloride to the reaction mixture. After 2 h, the crude
product was subjected to aqueous workup and the material was
desilylated by treatment with conc. NH4OH at 0 °C for 30 min
to afford the ribonucleoside 2′-acetal 14c. Regioselective 5′-
O-dimethoxytritylation of 14c,d was accomplished in 2 h when
employing a slight molar excess (1.2 equiv) of dimethoxytrityl
chloride in pyridine; 5′-O-protected ribonucleoside 2′-acetals
15c,d were isolated in yields of 90-95%% after purification
Although the preparation of ribonucleosides 14a,b proceeded
satisfactorily according to the reaction pathway shown in
Scheme 1, we experienced difficulties in obtaining reproducible
results in the synthesis of 14c,d (Scheme 2). Despite our
attempts at improving the synthetic process, reproducibility
issues were reoccurring and forced us to adopt a different
approach to the preparation of 14c,d. We opted for the method
of Gough et al.,10b which was reported for the synthesis of
ribonucleoside precursors leading to the preparation of phos-
phoramidites 1 and 2. As shown in Scheme 2, thioacetalization
of 4-(N-dichloroacetyl-N-methyl)aminobenzyl alcohol (16)25 in
DMSO, acetic anhydride and acetic acid was carried out
affording the thioacetal 17 in 65% yield within 24 h. Stanny-
lation of adenosine (10c′) and N2-isobutyryl guanosine (10d)
upon reaction with dibutyltin oxide in methanol produced the
respective stannylidene derivatives 19c′,d, which were used in
the next synthetic step without further purification. The thio-
acetal 17 was treated with sulfuryl chloride in CH2Cl2 to
generate the chloromethyl ether intermediate 18, which was
(29) This purification step is critically important. One must ensure that
the desired ribonucleoside 2′-acetal is not contaminated with any 3′-acetal
to prevent subsequent incorporation of (2′f5′)-internucleotidic phosphodi-
ester linkages into the oligoribonucleotide during solid-phase synthesis.
(30) (a) Ti, G. S.; Gaffney, B. L.; Jones, R. A. J. Am. Chem. Soc. 1982,
104, 1316-1319. (b) Jones, R. A. In Olignucleotide synthesis - a practical
approach; Gait, M. J., Ed.; IRL Press: Oxford, 1984; pp 23-34.
J. Org. Chem, Vol. 73, No. 7, 2008 2777

Cies´lak et al.
on silica gel. Phosphinylation of 15c,d was carried out under
conditions similar to those employed for 15a,b. The purification
and post-purification processing of the resulting ribonucleoside
phosphoramidites 9c,d were performed under near identical
conditions to those used for the production of 9a,b. The
phosphinylation of 15a,b and 15c,d produced phosphoramidites
9a-d in yields varying from 62 to 85%. The key ribonucleoside
2′-acetals 14a-d were characterized by 1H and 13C NMR
spectroscopies and by high-resolution time-of-flight mass
spectrometry using atmospheric pressure electron spray ioniza-
tion (APESI-HRMS), whereas phosphoramidites 9a-d were
characterized by 31P NMR spectroscopy and APESI-HRMS
(data shown in the Experimental Section and in the Supporting
Information).
The 2′-O-[4-(N-methylamino)benzyloxy]methyl ribonucleo-
side phosphoramidites 9a-d were then employed in the solid-
phase synthesis of AUCCGUAGCUAACGUCAUGG as a
model oligoribonucleotide to evaluate the coupling kinetics and
efficiencies of these phosphoramidites. Solid-phase synthesis
of the 20-mer RNA oligonucleotide was carried out on a 0.2
µmol scale using 9a-d as 0.2 M solutions in anhydrous MeCN.
It should be emphasized that the effective concentration of
activated 9a-d in the synthesis column was ∼0.1 M. Phos-
phoramidite activators, such as 1H-tetrazole, 5-ethylthio-1H-
tetrazole, and 5-benzylthio-1H-tetrazole were tested in the
preparation of the RNA oligonucleotide; 0.25 M 5-benzylthio-
1H-tetrazole in MeCN produced the best result in terms of
phosphoramidite coupling efficiency, which averaged 99%
within a coupling time of 180 s. The stepwise coupling yields
were determined spectrophotometrically at 498 nm through the
standard dimethoxytrityl cation assay. After completion of the
RNA oligonucleotide chain assembly, the solid-phase-linked 20-
mer was exposed to concd NH4OH for 10 h at 55 °C to cleave
nucleobase and phosphate protecting groups concurrently with
the dichloroacetyl group of each 2′-acetal function, and release
the 2′-O-protected oligoribonucleotide from the solid support.
The RP-HPLC profile of the unpurified 2′-O-[4-(N-methylami-
no)benzyloxy]methyl (4-MABOM) RNA oligomer, shown in
Figure 1A, is reflective of the high coupling efficiency of
phosphoramidites 9a-d on the basis of the relatively low level
of failure sequences clustered at the base of the single major
peak.
FIGURE 1. RP-HPLC profiles of unpurified 2′-O-protected/depro-
tected AUCCGUAGCUAACGUCAUGG. (A) AUCCGUAGCUAA-
CGUCAUG[2′-O-(4-MABOM)]19G. (B) AUCCGUAGCUAACGU-
CAUG[2′-O-(4-MABOM)]19G in 0.1 M AcOH/TEMED, pH 3.8,
90 °C, 30 min. (C) Material of profile B that was desalted using a
PD-10 (Sephadex G-25M) column. RP-HPLC analyses were performed
employing a 5 µm Supelcosil LC-18S column (25 cm × 4.6 mm)
according to the following conditions: starting from 0.1 M triethy-
lammonium acetate pH 7.0, a linear gradient of 1% MeCN/min was
pumped at a flow rate of 1 mL/min for 40 min and was then held,
isocratically, for 20 min. Peak heights are normalized to the highest
peak, which is set to 1 arbitrary unit.
Desalting the fully deprotected RNA oligonucleotide through
a size exclusion column gave a high quality product, the RP-
HPLC profile of which, is shown in Figure 1C. The RP-HPLC
profile is comparable to that of a commercial oligonucleotide
of identical RNA sequence (data shown in the Supporting
Information). Polyacrylamide gel electrophoresis analysis of the
unpurified and desalted RNA oligonucleotide was also compared
with that of the commercial RNA oligonucleotide reference.
As shown in Figure 2, the electrophoretic mobility and the purity
of the RNA oligonucleotide prepared from ribonucleoside
phosphoramidites 9a-d are comparable to those of the com-
mercial RNA oligonucleotide reference.
The purity of the fully deprotected oligoribonucleotide in
terms of nucleobase modifications and integrity of the (3′f5′)-
internucleotidic phosphodiester linkages throughout the RNA
sequence was also assessed through enzymatic hydrolysis.
Specifically, the unpurified and desalted RNA 20-mer was
subjected to snake venom phosphodiestrase (SVP) and bacterial
alkaline phosphatase (BAP) for 16 h at 37 °C. As displayed in
Figure 3, the RP-HPLC profile of the enzymatic digest indicates
that the RNA oligonucleotide is completely degraded to the
respective nucleosides without significant amounts of nucleobase
modification.
While the cleavage of the 2′-O-(4-MABOM) group is
performed under acidic conditions, vicinal migration of each
(3′f5′)-internucleotidic phosphodiester linkage may occur and
may also lead to RNA chain cleavage.33 To evaluate the extent
of (2′f5′)-internucleotidic phosphodiester linkage formation
under these conditions, the crude fully deprotected RNA
oligonucleotide was purified by RP-HPLC and then desalted
by size-exclusion chromatography. The pure RNA oligomer was
then treated with bovine spleen phosphodiesterase (BSP) and
BAP for 16 h at 37 °C. BSP, unlike SVP, does not cleave
Removal of the 2′-O-(4-MABOM) groups from the RNA
oligomer was effected by dissolving the oligonucleotide in
0.1 M AcOH adjusted to pH 3.8 with N,N,N′,N′-tetramethyl-
ethylenediamine (TEMED)21b and heating the solution at 90 °C
for 30 min. The RP-HPLC profile of the deprotection reaction
is presented in Figure 1B. Aside from benzamide, which was
generated during nucleobase deprotection, 4-(N-methylamino)-
benzyl alcohol (29)31 was produced through hydrolysis of the
N-methyliminoquinone methide intermediate 2832 that is pre-
sumably generated during the cleavage of the 2′-acetals.
(31) The RP-HPLC retention time of 29 (tR ) 19.6 min) is identical to
that of an authentic sample of 4-(N-methylamino)benzyl alcohol that was
prepared from the reduction of 4-(N-methylamino)benzoic acid (see ref 25).
(32) Carl, P. L.; Chakravarty, P. K.; Katzenellenbogen, J. A. J. Med.
Chem. 1981, 24, 479-480.
(33) Reese C.B. Org. Biomol. Chem. 2005, 3, 3851-3868 and references
therein.
2778 J. Org. Chem., Vol. 73, No. 7, 2008

2′-Hydroxyl Protection of Ribonucleosides.
oligomer35 that was exposed to acidic conditions (pH 3.8) at
elevated temperature (90 °C) during the time allocated for
complete 2′-O-deprotection (30 min). RNA fragments containing
at least one or several (2′f5′)-internucleotidic phosphodiester
linkages are expected to exhibit longer retention times than those
of individual ribonucleosides but shorter than the retention time
of the full length 20-mer (17.5 min) under identical chromato-
graphic conditions.
Amplification of the baseline corresponding to retention times
in the range of 12 to 18 min (Figure 4B′) revealed a number of
peaks, the relative intensity of which, increased upon increasing
exposure to acidic conditions by an additional 30 min at 90 °C
(Figure 4A′). This trend is consistent with increased formation
of (2′f5′)-internucleotidic phosphodiester linkages with in-
creasing exposure time to acidic conditions. The formation of
(2′f5′)-internucleotidic phosphodiester linkages under acidic
conditions is not detected in the SVP/BAP digestion reaction
since it has been demonstrated that synthetic (2′f5′)UpU is
completely hydrolyzed to uridine by the enzymes (data shown
in the Supporting Information). One can nonetheless safely state
that the amount of RNA fragments containing (2′f5′)-inter-
nucleotidic phosphodiester linkages, produced under the depro-
tection conditions defined above for 2′-O-(4-MABOM) RNA
oligonucleotides, is negligible.
In summary, RNA phosphoramidites protected with a 4-(N-
dichloroacetyl-N-methylamino)benzyloxymethyl group for 2′-
O-protection allow rapid and efficient solid-phase RNA syn-
thesis without a mandatory requirement for nucleobase, 5′-
hydroxyl, and phosphate protecting groups to differ from those
traditionally employed in standard solid-phase DNA synthesis.
Consequently, the conditions used for the cleavage of nucleobase
and phosphate protecting groups are essentially identical to those
used for standard DNA oligonucleotide deprotection, and
resulted in the release of a 2′-O-(4-MABOM)-protected RNA
oligonucleotide of high quality (Figure 1A). Removal of the
2′-O-(4-MABOM) group from the RNA oligonucleotide pro-
ceeded smoothly, in a buffered acetic acid solution kept at
90 °C, without producing significant amounts of (2′f5′)-
internucleotidic phosphodiester linkages and nucleobase modi-
fications (Figures 3 and 4). The overall quality of the native
unpurified RNA oligonucleotide after full deprotection was
comparable to that of a commercial oligonucleotide of identical
RNA sequence on the basis of chromatographic (data shown in
the Supporting Information) and electrophoretic data (Figure
2). Although the 4-(N-dichloroacetyl-N-methylamino)benzy-
loxymethyl group is perfectly suitable for 2′-hydroxyl protection
in solid-phase RNA synthesis, an improved method for 2′-O-
protection of ribonucleoside purines with this group is necessary
for the commercialization of phosphoramidites 9a-d. This
objective is currently under development in our laboratories.
FIGURE 2. Polyacrylamide gel analysis (20% polyacrylamide-7 M
urea, 1X Tris Borate EDTA buffer, pH 8.3) of unpurified and desalted
AUCCGUAGCUAACGUCAUGG. Left lane: commercial RNA oli-
gomer (0.25 OD260). Right lane: RNA oligomer (0.25 OD260) obtained
from the deprotection of AUCCGUAGCUAACGUCAUG[2′-O-(4-
MABOM)]19G under acidic conditions. Oligonucleotides are visualized
as blue bands upon staining the gel with Stains-all. Bromophenol blue
is used as a marker and shows as a large band, in each lane, at the
bottom of the gel.
FIGURE 3. RP-HPLC chromatogram of unpurified and desalted
AUCCGUAGCUAACGUCAUGG after treatment with snake venom
phosphodiesterase and bacterial alkaline phosphatase. RP-HPLC analy-
ses were performed under conditions identical to those described in
the caption of Figure 1. Key: C, cytidine; U, uridine; G, guanosine;
A, adenosine.
Experimental Section
3′,5′-O-(1,1,3,3-Tetraisopropyldisiloxane-1,3-diyl)uridine (11a).
Uridine (10a, 15 mmol) was dried by coevaporation with anhydrous
pyridine (3 × 20 mL) under reduced pressure. The dry ribonucleo-
(2′f5′)-internucleotidic phosphodiester linkages34 and thus
provides a means to assess the level of (2′f5′)-internucleotidic
phosphodiester bond formation among the native (3′f5′)-
internucleotidic phosphodiester linkages. Figure 4B shows near
complete digestion of the RP-HPLC purified and desalted RNA
(35) Commercial bovine spleen phosphodiesterase contains some (1%
w/w) adenosine deaminase (ADA) as a contaminant. Such a low concentra-
tion of ADA in the digestion reaction was sufficient to quantitatively convert
adenosine to inosine. This conversion was unambiguously demonstrated
by mixing adenosine with BSP/BAP under the conditions used for the
digestion of the RNA oligomer. RP-HPLC analysis of the reaction indicated
complete conversion of adenosine to inosine, which had a retention time
identical to that of a commercial sample of inosine (data shown in the
Supporting Information).
(34) Synthetic (3′f5′)UpU is a substrate for BSP but synthetic (2′f5′)-
UpU is not. Data are presented in the Supporting Information. See also:
(a) Giannaris, P. A.; Damha, M. J. Nucleic Acids Res. 1993, 21, 4742-
4749. (b) Thayer, J. R.; Rao, S.; Puri, N.; Burnett, C. A.; Young, M. Anal.
Biochem. 2007, 361, 132-139.
J. Org. Chem, Vol. 73, No. 7, 2008 2779

Cies´lak et al.
FIGURE 4. RP-HPLC profiles of purified and desalted AUCCGUAGCUAACGUCAUGG after treatment with bovine spleen phosphodiesterase
and bacterial alkaline phosphatase. (A) The RP-HPLC-purified RNA oligomer had been exposed to pH 3.8 at 90 °C for a period of time of 30 min
after 2′-O-deprotection and RP-HPLC purification. (A′) 100-Fold baseline amplification of chromatogram A corresponding to retention times in the
range of 12-18 min. (B) The RNA oligomer had been exposed to pH 3.8 at 90 °C for 30 min during 2′-O-deprotection prior to purification. (B′)
100-Fold baseline amplification of chromatogram B corresponding to retention times in the range of 12-18 min. RP-HPLC analyses were performed
under conditions identical to those described in the caption of Figure 1. Key: C, cytidine; U, uridine; I, inosine; G, guanosine.
side was redissolved in anhydrous pyridine (70 mL), and 1,3-
dichloro-1,1,3,3-tetraisopropyldisiloxane26 (5.76 mL, 18 mmol) was
added, dropwise, over a period of 30 min. The reaction mixture
was then stirred at ∼25 °C until 10a had completely converted to
a new product (∼3 h) as indicated by TLC (CHCl3/MeOH (9:1
v/v)). The reaction mixture was concentrated to an oil, which was
then dissolved in CH2Cl2 (200 mL). The solution was extracted
with satd NaHCO3 (150 mL). The organic layer was dried over
anhydrous Na2SO4, filtered, and rotoevaporated to dryness under
vacuum. The product was dried further by coevaporation with
toluene (3 × 20 mL). Crude 11a19 was isolated as a white foam
and used without purification in the next synthetic step.
3′,5′-O-(1,1,3,3-Tetraisopropyldisiloxane-1,3-diyl)-2′-O-meth-
ylthiomethyluridine (12a). To a solution of crude 11a in DMSO
(15 mL) were added glacial AcOH (23 mL) and Ac2O (15 mL).
The solution was stirred at ∼25 °C for 16 h and then at 45 °C until
completion of the reaction (∼5 h), which was monitored by TLC
(CHCl3/MeOH (95:5 v/v)). The solution was transferred to a 2 L
Erlenmeyer flask to which was added, under vigorous stirring, a
solution of K2CO3 (31 g) in water (200 mL). The precipitated
material was isolated either by filtration or decantation and was
redissolved in a minimum volume of THF (∼15-20 mL). The
resulting solution was then poured into water (250 mL) to give the
crude product as a gummy material. Most of the water was
decanted; the crude product was carefully dried by consecutive
coevaporation with pyridine (30 mL), toluene (3 × 30 mL) and
dichloromethane (30 mL). The crude ribonucleoside 12a was
purified by chromatography on silica gel (150 g in a 6.5 cm i.d.
column) using a gradient of MeOH (0 f 3%) in CH2Cl2. Fractions
containing pure 12a (TLC) were collected, rotoevaporated to a foam
under low pressure, and dissolved in dry C6H6 (∼20 mL); the
solution was frozen and then lyophilized under high vacuum to
produce a white powder (7 g, 12.8 mmol, 85%). Characterization
data obtained from 1H and 13C NMR analysis of 12a are in
agreement with those reported by Semenyuk et al.19
transferred to a 2 L Erlenmeyer flask and was vigorously stirred
while adding slowly a solution of K2CO3 (35 g) in water (270 mL).
The precipitated material was purified and processed as described
for the preparation of 12a. Purified 12b was isolated as a white
powder (7.1 g, 11 mmol, 73%). Characterization data obtained from
1H and 13C NMR analysis of 12b are in agreement with those
reported by Semenyuk et al.19
3′,5′-O-(1,1,3,3-Tetraisopropyldisiloxane-1,3-diyl)-2′-O-[4-(N-
dichloroacetyl-N-methylamino)benzyloxy]methyluridine (13a).
To a cooled (0 °C) suspension of thoroughly dried 12a (5.00 mmol),
4-(N-dichloroacetyl-N-methylamino)benzyl alcohol25 (16, 766 mg,
5.00 mmol), and powdered 5A molecular sieves (∼200 mg) in
anhydrous 1,2-dichloroethane (DCE, 50 mL) was added a freshly
prepared solution of N-iodosuccinimide (NIS, 1.12 g, 5.00 mmol)
and fresh trifluoromethanesulfonic acid (TfOH, 64 µL, 0.72 mmol)
in DCE/Et2O ((1:1 v/v), 25 mL). Progress of the reaction was
assessed by monitoring the disappearance of 12a by TLC (CHCl3/
MeOH (93:7 v/v)). The addition of more TfOH (2 up to 4 × 0.72
mmol) may be necessary to complete the reaction. The suspension
was filtered; the filtrate was diluted with CH2Cl2 (150 mL) and
washed, consecutively, with aqueous 1 M sodium bisulfite (100
mL) and satd NaHCO3 (100 mL). The organic layer was dried over
anhydrous Na2SO4, filtered and, then, evaporated to a foam under
reduced pressure. Crude 13a was used in the next synthetic step
without further purification.
N4-Benzoyl-3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)-
2′-O-[4-(N-dichloroacetyl-N-methylamino)benzyloxy]methyl-
cytidine (13b). This compound was prepared from 12b under
conditions similar to those employed for the synthesis of 13a. Crude
13b was used without purification in the next synthetic step.36
2′-O-[4-(N-Dichloroacetyl-N-methylamino)benzyloxy]methy-
luridine (14a). Crude 13a (5.00 mmol) was dissolved in methanol
(30 mL), and ammonium fluoride19,28 (741 mg, 20.0 mmol) was
added. The reaction mixture was stirred at ∼25 °C until cleavage
of the disiloxyl group was complete (∼16 h) as indicated by TLC
(CHCl3/MeOH (9:1 v/v)). The reaction mixture was concentrated
in vacuo, and then partitioned between CH2Cl2 (100 mL) and satd
NaHCO3 (50 mL). The aqueous layer was extracted further with
CH2Cl2 (2 × 100 mL). The organic extracts were pooled together
and were dried over anhydrous Na2SO4, filtered, and evaporated
N4-Benzoyl-3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)-
cytidine (11b). This compound was prepared under conditions
similar to those employed for the synthesis of 11a. Crude 11b19
was also used without purification in the next synthetic step.
N4-Benzoyl-3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)-
2′-O-methylthiomethylcytidine (12b). To a solution of crude 11b
in DMSO (15 mL) were added glacial AcOH (15 mL) and Ac2O
(10 mL). The solution was stirred at ∼25 °C for 16 h and then at
45 °C until completion of the reaction (∼5 h), which was monitored
by TLC (CHCl3/MeOH (95:5 v/v)). The solution was then
(36) Silica gel chromatography may occasionally be necessary to separate
13b from unreacted 12b to avoid subsequent purification problems.
Typically, 13b is adequately resolved from 12b when a gradient of MeOH
(0f2%) in CH2Cl2 is employed as an eluent.
2780 J. Org. Chem., Vol. 73, No. 7, 2008

2′-Hydroxyl Protection of Ribonucleosides.
to dryness under reduced pressure. Crude 14a was purified by
chromatography on silica gel (20 g in a 2.5 cm i.d. column) using
a gradient of MeOH (0 f 5%) in CH2Cl2. Fractions containing
pure 14a (TLC) were collected and rotoevaporated to a foam under
low pressure. The yield of pure ribonucleoside 14a was 70%
(1.75 g, 3.5 mmol) relative to the amount of 12a that was employed
in the preparation of 13a. 1H NMR (300 MHz, DMSO-d6): δ 11.31
(s, 1H), 7.94 (d, J ) 8.0 Hz, 1H), 7.41-7.33 (m, 4H), 6.22 (s,
1H), 5.95 (d, J ) 5.8 Hz, 1H), 5.60 (d, J ) 8.0 Hz, 1H), 5.26 (d,
J ) 5.8 Hz, 1H), 5.17 (t, J ) 5.1 Hz, 1H), 4.83 (s, 2H), 4.61 (d,
2J ) 12.4 Hz, 1H), 4.55 (d, 2J ) 12.4 Hz, 1H), 4.24 (t, J ) 5.1 Hz,
1H), 4.14 (dt, J ) 5.1, 4.4 Hz, 1H), 3.91 (m, 1H), 3.70-3.55 (m,
2H), 3.23 (s, 3H). 13C NMR (75 MHz, DMSO-d6): δ 162.9, 162.8,
150.5, 140.3, 140.2, 138.4, 128.9, 126.9, 101.9, 93.4, 86.1, 85.2,
78.2, 68.5, 67.9, 64.7, 60.5, 38.2. ESI/TOF-HRMS: calcd for
C20H22Cl2N3O8 (M - H)- 502.0784, found 502.0780.
N6-Benzoyl-2′-O-[4-(N-dichloroacetyl-N-methylamino)benzyl-
oxy]methyladenosine (14c). To a cold (-78 °C) stirred solution
of 17 (4.9 g, 16 mmol) in anhydrous CH2Cl2 was added dropwise,
within 5 min, 1 M SO2Cl2 (16 mL, 16 mmol) in CH2Cl2.10b The
reaction mixture was stirred at -78 °C for an additional 30 min
and was then rotoevaporated under reduced pressure. The residue
(18) was coevaporated with dry CH2Cl2 (3 × 30 mL) under low
pressure, and anhydrous DMF (20 mL) was added. The clear
solution of 18 was immediately added, dropwise over a period of
5 min, to a stirred solution of 19c′ (8 g, 16 mmol) and dry tetra-
n-butylammonium bromide (7.7 g, 16 mmol) in dry DMF (65 mL)
kept at 60 °C. The reaction mixture was allowed to stir for 60 min
at 60 °C, whereupon pyridine:H2O (3:2 v/v, 20 mL) was added.
The solution was stirred for 25 min at ∼25 °C and was then
rotoevaporated under high vacuum to remove DMF. The crude
material was dissolved in a minimum volume of pyridine/H2O (3:2
v/v); the solution was dispersed in silica gel (20 g) and allowed to
dry overnight inside a fume hood. The coated silica gel was
pulverized and layered on the top of a chromatography column
(6.5 cm i.d.) packed with silica gel (150 g). The desired 2′-acetal
14c′ eluted faster than the 3′-acetal when a gradient of MeOH (0
f 5%) in CH2Cl2 was used as the eluent.29 Fractions containing
14c′ were pooled together and the solvents were removed by
rotoevaporation under reduced pressure to give pure 14c′ (2.65 g,
5.00 mmol, 31%). Trimethylchlorosilane (3.2 mL, 25 mmol) was
added to a solution of 14c′ (2.65 g, 5.00 mmol) in anhydrous
pyridine (20 mL). The reaction mixture was stirred at ∼25 °C for
2 h, upon which, benzoyl chloride (2.9 mL, 25 mmol) was added.
After 2 h, the reaction mixture was cooled to ∼0 °C; water (5 mL)
was added and was followed, 10 min later, by the addition of concd
NH4OH (10 mL). The open-flask reaction was left stirring at
∼25 °C for 30 min. Solvents and excess NH4OH were removed
by rotoevaporation under low pressure. The crude product (14c)
was dissolved in EtOAc (250 mL) and the solution was extracted
with 1 M NaHCO3 (3 × 200 mL) and then with 2 M NaCl (3 ×
200 mL). The organic phase was dried over anhydrous Na2SO4,
filtered, and rotoevaporated to dryness under reduced pressure. The
material left was purified by chromatography on silica gel (150 g
in a 6.5 cm I.D. column) using a gradient of MeOH (0% f 2%) in
CH2Cl2 as the eluent. Fractions containing pure 14c were collected
and rotoevaporated to dryness affording a white solid (2.8 g, 4.4
N4-Benzoyl-2′-O-[4-(N-dichloroacetyl-N-methylamino)benzyl-
oxy]methylcytidine (14b). This compound was prepared on a 5
mmol scale and purified under conditions similar to those employed
for the synthesis and purification of 14a. Pure ribonucleoside 14b
was isolated in a yield of 60% (1.8 g, 3.0 mmol) relative to the
1
amount of 12b that was employed in the preparation of 13b. H
NMR (300 MHz, DMSO-d6): δ 11.30 (s, 1H), 8.57 (d, J ) 7.4
Hz, 1H), 8.01 (d, J ) 7.4 Hz, 2H), 7.6 (m, 1H), 7.52 (d, J ) 8.0
Hz, 2H), 7.46 (d, J ) 9.1 Hz, 2H), 7.35 (m, 3H), 6.19 (s, 1H),
5.97 (d, J ) 2.5 Hz, 1H), 5.03 (d, J ) 6.6 Hz, 1H), 4.92 (d, J )
6.6 Hz, 1H), 4.66 (m, 2H), 4.24 (dd, J ) 3.0, 2.7 Hz, 1H), 4.17
(m, 1H), 3.99 (m, 1H), 3.83 (dd, J ) 12.3, 2.2 Hz, 1H), 3.67 (dd,
J ) 12.3, 2.6 Hz, 1H) 3.19 (s, 3H). 13C NMR (300 MHz, DMSO-
d6): δ 167.3, 163.1, 162.7, 154.4, 145.0, 140.4, 138.6, 133.0, 132.7,
129.5, 128.4, 126.9, 96.1, 93.3, 88.6, 84.3, 78.7, 68.1, 67.5, 74.7,
59.4, 38.2. APESI-HRMS: calcd for C27H28Cl2N4O8 (M)+ 606.1284,
found 606.1277.
Preparation of 2′,3′-O-(Dibutylstannylene) Ribonucleosides
19c′,d. Stannylation of adenosine (10c′) or N2-isobutyrylguanosine
(10d) was performed according to a method described in the
literature.10b To a solution of 10c′ or 10d (20 mmol) in MeOH
(1 L) was added dibutyltin oxide (20 mmol). The reaction mixture
was refluxed for 1 h and then allowed to cool to room temperature.
The solvent was removed by rotoevaporation under reduced
pressure. The crystalline material was left over phosphorus pen-
toxide for 16 h in a desiccator, under high vacuum, prior to be
used without purification in the next synthetic step.
1
mmol, 88%). H NMR (300 MHz, DMSO-d6): δ 11.21 (s, 1H),
8.77 (s, 1H), 8.71 (s, 1H), 8.05 (d, J ) 7.1 Hz, 2H), 7.67-7.47
(m, 3H), 7.29 (d, J ) 8.1 Hz, 2H), 7.18 (d, J ) 8.1 Hz, 2H), 6.26
(d, J ) 6.0 Hz, 1H), 6.20 (s, 1H), 4.90 (t, J ) 5.5 Hz, 1H), 4.84
4-(N-Dichloroacetyl-N-methylamino)benzyl
Methylthio-
2
2
methyl Ether (17). A modified literature procedure10b was em-
ployed for the preparation of 17. Specifically, 4-(N-dichloroacetyl-
N-methylamino)benzyl alcohol (16,25 10 g, 40 mmol) was dissolved
in dry DMSO (40 mL). Ac2O (30 mL) and AcOH (20 mL) were
added to the solution, which was left standing in the dark for 24 h
at ∼25 °C. The yellow solution was then added dropwise during
2 h to a vigorously stirred slurry of NaHCO3 (88 g) in H2O
(133 mL) to control the release of carbon dioxide. The reaction
mixture was left stirring in the dark for an additional 24 h. The
oily material was allowed to settle and the aqueous supernatant
was carefully decanted. The crude oil was dissolved in EtOAc/
hexane (1:1 v/v, 100 mL) and the solution was extracted with satd
NaHCO3 (3 × 100 mL) and once with 2 M NaCl (100 mL). The
organic phase was dried over anhydrous Na2SO4 and the solvent
was removed by rotoevaporation under reduced pressure. The
material left was purified by chromatography on silica gel (150 g
in a 6.5 cm I.D. column) using a gradient of AcCN (0% f 15%)
in CH2Cl2 as the eluent. Fractions containing the desired product
(TLC) were pooled together and rotoevaporated under low pressure
(2d J ) 12.3 Hz, 2H), 4.47 (d, J ) 11.9 Hz, 1H), 4.40 (dd, J )
4.1, 3.8 Hz, 1H), 4.32 (d, 2J ) 11.9 Hz, 1H), 4.05 (dt, J ) 3.8, 3.6
Hz, 1H), 3.74 (dd, J ) 11.9, 4.1 Hz, 1H), 3.61 (dd, J ) 11.9, 3.8
Hz, 1H), 3.20 (s, 3H). 13C NMR (75 MHz, DMSO-d6): δ 165.5,
162.7, 152.0, 151.6, 150.4, 143.1, 140.3, 138.2, 133.2, 132.4, 128.9,
128.4, 126.9, 125.8, 93.8, 86.3, 86.1, 78.4, 69.1, 68.1, 64.8, 61.1,
38.2. APESI-HRMS: calcd for C28H28Cl2N6O7 (M)+ 630.1397,
found 630.1388.
N2-Isobutyryl-2′-O-[4-(N-dichloroacetyl-N-methylamino)ben-
zyloxy]methylguanosine (14d). The preparation of this compound
is essentially identical to that of 14c′ with the exception of the
following modification: the stannylated ribonucleoside 19d
(6.5 g, 11 mmol) is mixed with a solution of 18 (16 mmol) in DMF
(20 mL) in the absence of tetra-n-butylammonium bromide.10b The
reaction mixture is stirred at 60 °C and processed under conditions
identical to those used in the preparation of 14c′. Crude 14d was
also obtained as a mixture of regioisomeric acetals. Purification of
14d was achieved by chromatography on silica gel (150 g in a 6.5
cm I.D. column) using a gradient of MeOH (2% f 8%) in CH2Cl2
as the eluent.29 Fractions containing pure 14d were collected and
rotoevaporated to dryness under reduced pressure affording a white
solid (2 g, 3.3 mmol, 30%).1H NMR (300 MHz, DMSO-d6): δ
12.07 (s, 1H), 11.66 (s, 1H), 8.27 (s, 1H), 7.49 (d, J ) 7.9 Hz,
2H), 7.39 (d, J ) 7.9 Hz, 2H), 6.20 (s, 1H), 5.85 (d, J ) 6.6 Hz,
1
to give 17 (8 g, 26 mmol, 65%) as a yellow oil. H NMR (300
MHz, DMSO-d6): δ 7.45 (d, J ) 8.0 Hz, 2H), 7.38 (d, J ) 8.0
Hz, 2H), 6.21 (s, 1H), 4.75 (s, 2H), 4.60 (s, 2H), 3.23 (s, 3H), 2.14
(s, 3H). 13C NMR (75 MHz, DMSO-d6): δ 162.7, 140.4, 138.5,
129.1, 127.0, 74.1, 68.1, 64.6, 38.1, 13.4.
J. Org. Chem, Vol. 73, No. 7, 2008 2781

Cies´lak et al.
1H), 5.62 (d, J ) 5.8 Hz, 1H), 5.14 (t, J ) 5.3 Hz, 1H), 4.92 (d,
N6-Benzoyl-5′-O-(4,4′-dimethoxytrityl)-3′-O-[(N,N-diisopro-
pylamino)(2-cyanoethyloxy)]phosphinyl-2′-O-[4-(N-dichloroacetyl-
N-methylamino)benzyloxy]methyladenosine (9c). 31P NMR (121
MHz, C6D6): δ 152.1, 150.7. APESI-HRMS: calcd for C58H63-
Cl2N6O10P (M)+ 1132.3782, found 1132.3770.
N2-Isobutyryl-5′-O-(4,4′-dimethoxytrityl)-3′-O-[(N,N-diisopro-
pylamino)(2-cyanoethyloxy)]phosphinyl-2′-O-[4-(N-dichloroacetyl-
N-methylamino)benzyloxy]methylguanosine (9d). 31P NMR (121
MHz, C6D6): δ 151.4, 151.3. APESI-HRMS: calcd for C55H65-
Cl2N8O11P (M)+ 1114.3887, found 1114.3880.
2
2J ) 12.9 Hz, 1H), 4.91 (d, J ) 12.9 Hz, 1H), 4.68 (s, 2H), 4.62
(m, 1H), 4.28 (2d, J ) 4.7 Hz, 1H), 4.04 (m, 1H), 3.65 (dd, J )
12.0 , 4.8 Hz, 1H), 3.57 (dd, J ) 12.0, 4.8 Hz, 1H), 3.23 (s, 3H),
2.77 (h, J ) 6.9 Hz, 1H), 1.12 (d, J ) 6.9 Hz, 6H). 13C NMR (75
MHz, DMSO-d6):δ 180.0, 154.7, 148.9, 148.1, 140.4, 138.7, 137.2,
129.2, 127.0, 120.0, 93.8, 86.0, 83.6, 75.2, 73.2, 68.2, 64.6, 60.9,
54.8, 38.2, 34.6, 18.73, 18.7. APESI-HRMS: calcd for C25H30-
Cl2N6O8 (M)+ 612.1502, found 612.1493.
General Procedure for the Preparation of 5′-O-(4,4′-Dimeth-
oxytrityl)-2′-O-[4-(N-dichloroacetyl-N-methylamino)benzyloxy]-
methyl Ribonucleosides 15a-d. To a solution of a dry ribonu-
cleoside (14a-d, 3.5 mmol) in anhydrous pyridine (15 mL) was
added 4,4′-dimethoxytrityl chloride (1.4 g, 4.2 mmol). The solution
was stirred at ∼25 °C and progress of the reaction was monitored
by TLC (CHCl3/MeOH (95:5 v/v)) until complete disappearance
of 14a-d (∼2 h). The reaction mixture was then poured into satd
NaHCO3 (200 mL) and the aqueous solution was extracted with
CH2Cl2 (3 × 150 mL). The combined organic layers were dried
over anhydrous Na2SO4. Following filtration, the filtrate was
evaporated under reduced pressure and the material left was
coevaporated with toluene (3 × 100 mL). The crude product was
purified by chromatography on silica gel (20 g in a 2.5 cm i.d.
column) using a gradient of CH3OH (0 f 2%) in CH2Cl2 containing
0.2% (v/v) triethylamine as the eluent. Fractions containing the pure
5′-O-DMTr ribonucleoside were collected and rotoevaporated to a
foam under low pressure. Ribonucleosides 15a, 15b, 15c, and 15d
were then dissolved each in dry C6H6 (10 mL); the resulting
solutions were frozen and then lyophilized under high vacuum to
produce yellowish powders in the following overall yields, relative
to the amounts of the respective ribonucleosides 10a, 10b, 10c′,
and 10d used as starting materials: 15a, 55%; 15b, 42%; 15c, 26%;
15d, 28%.
General Procedure for the Preparation of 5′-O-(4,4′-Dimeth-
oxytrityl)-3′-O-[(N,N-diisopropylamino)(2-cyanoethyloxy)]-
phosphinyl-2′-O-[4-(N-dichloroacetyl-N-methylamino)benzyloxy]-
methyl Ribonucleosides (9a-d). To a solution of a suitably pro-
tected ribonucleoside (15a-d, 3.0 mmol) in anhydrous CH2Cl2 (15
mL) containing Et3N (2 mL, 15 mmol) was added 2-cyanoethyl
N,N-diisopropylchlorophosphoramidite (1.15 mL, 5.10 mmol). The
reaction mixture was stirred at ∼25 °C under argon until complete
disappearance of 15a-d (∼2 h) was observed by TLC (C6H6/Et3N
(9:1 v/v)). The reaction mixture was then poured into water
(50 mL) and was extracted with CH2Cl2 (100 mL). The organic
layer was dried over anhydrous Na2SO4 and then filtered. The
filtrate was rotoevaporated to dryness under reduced pressure. The
crude phosphoramidite product was purified by chromatography
on silica gel (20 g in a 2.5 cm i.d. column) using C6H6/Et3N (9:1
v/v) as the eluent (9d required C6H6/CH2Cl2/Et3N (45:45:10 v/v/v)
to elute off the column). Fractions containing the pure product
(TLC) were pooled together and rotoevaporated to dryness under
vacuum. The material was dissolved in dry C6H6 (3 mL), and the
resulting solution was added to cold (-78 °C) stirred hexane (100
mL). The pure ribonucleoside phosphoramidite precipitated im-
mediately as a white solid. After careful decantation of hexane,
the solid was dissolved in dry C6H6 (10 mL); the solution was frozen
and then lyophilized under high vacuum. Et3N-free 9a-d were
isolated each as a powder in yields varying from 62 to 85%.
Solid-Phase Oligonucleotide Synthesis. Solid-phase synthesis
of AUCCGUAGCUAACGUCAUGG was conducted on a scale of
0.2 µmole in the “trityl-off” mode using a succinyl long chain
alkylamine controlled-pore glass (Succ-LCAA-CPG) support func-
tionalized with N2-dimethylaminomethylene-5′-O-(4,4′-dimethox-
ytrityl)-2′-/3′-O-acetylguanosine as the leader nucleoside. The solid
support was mixed with 0.5 mL of each Cap A solution (Ac2O/
pyridine/THF) and Cap B solution (10% 1-methylimidazole in THF)
for a period of 60 s to acetylate any unprotected hydroxyls of the
leader nucleoside. RNA oligonucleotide synthesis was carried out
using a DNA/RNA synthesizer and phosphoramidites 9a-d, which
were dissolved in dry MeCN to give 0.2 M solutions. 5-Benzylthio-
1H-tetrazole was employed for phosphoramidite activation and used
as a 0.25 M solution in dry MeCN. All other ancillary reagents
necessary for the preparation of oligonucleotides, with the exception
of trichloroacetic acid (TCA), were purchased and utilized as
recommended by the manufacturer. The reaction time for each
phosphoramidite coupling step was set to 180 s. The dedimethox-
17b
ytritylation step was effected using 4% TCA in CH2Cl2 during
a period of 60 s. The capping step of each synthesis cycle was also
programmed for a duration of 60 s.
Characterization and Deprotection of AUCCGUAGC-
UAACGUCAUG[2′-O-(4-MABOM)]19G. The solid-phase-linked
5′-dedimethoxytritylated oligonucleotide was placed into a 4 mL
screw-capped glass vial to which was added concd NH4OH (1 mL).
The suspension was shaken occasionally over a period of 30 min.
The ammoniacal solution was then transferred to another 4 mL
glass screw-capped and was heated at 55 °C for 10 h. A sample of
the ammoniacal solution (0.9 OD260) was evaporated to dryness
using a stream of air. The oligonucleotide was then dissolved in
0.1 M triethylammonium acetate, (pH 7.0, 100 µL) and was
analyzed by RP-HPLC. The analysis was performed using a 5 µm
Supelcosil LC-18S column (25 cm × 4.6 mm) according to the
following conditions: starting from 0.1 M triethylammonium acetate
pH 7.0, a linear gradient of 1% MeCN/min was pumped at a flow
rate of 1 mL/min for 40 min and was then held, isocratically, for
20 min. Peak heights are normalized to the highest peak, which is
set to 1 arbitrary unit. The RP-HPLC chromatogram of unpurified
AUCCGUAGCUAACGUCAUG[2′-O-(4-MABOM)]19G (tR ) 33.7
min) is shown in Figure 1A. The remaining portion of the
ammoniacal solution of AUCCGUAGCUAACGUCAUG[2′-O-(4-
MABOM)]19G (15 OD260) was evaporated to dryness employing a
stream of air. The oligonucleotide was dissolved in 0.1 M AcOH
buffered to pH 3.8 with N,N,N′,N′-tetramethylethylenediamine
(TEMED).21b The solution was kept at 90 °C for 30 min. A sample
(0.8 OD260) of the fully deprotected AUCCGUAGCUAACGU-
CAUGG was analyzed by RP-HPLC under chromatographic
conditions identical to those used for the analysis of AUC-
CGUAGCUAACGUCAUG[2′-O-(4-MABOM)]19G. The RP-HPLC
chromatogram of unpurified and fully deprotected AUCCGUAGC-
UAACGUCAUGG (tR ) 17.5 min) is shown in Figure 1B. The
solution of the fully deprotected oligoribonucleotide was applied
on the top of a 10-mL PD-10 (Sephadex G-25M) column. The
oligonucleotide was eluted from the column using DEPC-treated
H2O as the eluent. Fractions of 1 mL were collected and those
containing the RNA oligomer (A260) were pooled together and
evaporated to dryness using a stream of air. The unpurified
oligoribonucleotide was dissolved in DEPC-treated H2O and its
5′-O-(4,4′-Dimethoxytrityl)-3′-O-[(N,N-diisopropylamino)(2-
cyanoethyloxy)]phosphinyl-2′-O-[4-(N-dichloroacetyl-N-meth-
ylamino)benzyloxy]methyluridine (9a). 31P NMR (121 MHz,
C6D6): δ 151.9, 150.2. APESI-HRMS: calcd for C50H58Cl2N5O11P
(M)+ 1005.3248, found 1005.3248.
N4-Benzoyl-5′-O-(4,4′-dimethoxytrityl)-3′-O-[(N,N-diisopro-
pylamino)(2-cyanoethyloxy)]phosphinyl-2′-O-[4-(N-dichloroacetyl-
N-methylamino)benzyloxy]methylcytidine (9b). 31P NMR (121
MHz, C6D6): δ 152.7, 150.6. APESI-HRMS: calcd for C57H63-
Cl2N6O11P (M)+ 1108.3670, found 1108.3668.
2782 J. Org. Chem., Vol. 73, No. 7, 2008

2′-Hydroxyl Protection of Ribonucleosides.
concentration was determined by UV spectroscopy at 260 nm. RP-
HPLC analysis of the desalted unpurified oligomer was performed
and the chromatogram resulting from this analysis is shown in
Figure 1C. MALDI-TOF MS: calcd for C190H217N75O138P19 [M -
H]- 6346, found 6346.
of 14a-d and 17; 31P NMR spectra of 9a-d; mass spectra of
14a-d and 9a-d; expanded Figures 1 and 4A,B; RP-HPLC
chromatograms comparing AUCCGUAGCUAACGUCAUGG ob-
tained commercially and through solid-phase synthesis employing
phosphoramidites 9a-d; RP-HPLC analysis of synthetic (3′f5′)-
UpU and (2′f5′)UpU; RP-HPLC chromatograms of the enzymatic
digestion of synthetic (3′f5′)UpU and (2′f5′)UpU catalyzed by
either snake venom phosphodiesterase or bovine spleen phosphodi-
esterase; RP-HPLC chromatogram of the conversion of adenosine
to inosine by bovine spleen phosphodiesterase. This material is
available free of charge via the Internet at http://pubs.acs.org.
Supporting Information Available: Materials and methods;
procedure for polyacrylamide gel electrophoresis analysis of
unpurified and desalted AUCCGUAGCUAACGUCAUGG; pro-
cedures for hydrolysis of AUCCGUAGCUAACGUCAUGG cata-
lyzed by both snake venom phosphodiesterase/bacterial alkaline
phosphatase and bovine spleen phosphodiesterase/bacterial alkaline
phosphatase; 1H NMR spectra of 14a-d and 17; 13C NMR spectra
JO702717G
J. Org. Chem, Vol. 73, No. 7, 2008 2783