Syntheses of two 5-hydroxymethyl-2′-deoxycytidine phosphoramidites with TBDMS as the 5-hydroxymethyl protecting group and their incorporation into DNA

Article abstract of DOI:10.1021/jo200566d

5-Hydroxymethylcytosine (5-hmC) is a newly discovered DNA base modification in mammalian genomic DNA that is proposed to be a major epigenetic mark. We report here the syntheses of two new versions of phosphoramidites III and IV from 5-iodo-2′-deoxyuridine in 18% and 32% overall yields, respectively, with TBDMS as the 5-hydroxyl protecting group. Phosphoramidites III and IV allow efficient incorporation of 5-hmC into DNA and a "one-step" deprotection procedure to cleanly remove all the protecting groups. A "two-step" deprotection strategy is compatible with ultramild DNA synthesis, which enables the synthesis of 5hmC-containing DNA with additional modifications.

Full text of DOI:10.1021/jo200566d

NOTE
pubs.acs.org/joc
Syntheses of Two 5-Hydroxymethyl-20-deoxycytidine
Phosphoramidites with TBDMS as the 5-Hydroxymethyl Protecting
Group and Their Incorporation into DNA
Qing Dai,*,† Chun-Xiao Song,† Tao Pan,‡ and Chuan He†
†Department of Chemistry and ‡Department of Biochemistry & Molecular Biology, The University of Chicago, 929 E. 57th St., Chicago,
Illinois, 60637 United States
S Supporting Information
b
ABSTRACT: 5-Hydroxymethylcytosine (5-hmC) is a newly
discovered DNA base modification in mammalian genomic
DNA that is proposed to be a major epigenetic mark. We report
here the syntheses of two new versions of phosphoramidites III
and IV from 5-iodo-20-deoxyuridine in 18% and 32% overall
yields, respectively, with TBDMS as the 5-hydroxyl protecting
group. Phosphoramidites III and IV allow efficient incorpora-
tion of 5-hmC into DNA and a “one-step” deprotection
procedure to cleanly remove all the protecting groups. A
“two-step” deprotection strategy is compatible with ultramild
DNA synthesis, which enables the synthesis of 5hmC-containing DNA with additional modifications.
ethylcytosine (5-mC) is an important DNA modifi-
5-Mcation found in eukaryotes and is referred to as the
fifth base besides dA, dC, dG, and dT. It constitutes ∼2ꢀ8% of
the total cytosine in human genomic DNA and impacts a broad
range of biological functions including gene expression, main-
tenance of genome integrity, parental imprinting, X-chromo-
some inactivation, regulation of development, aging, cancer, and
so on.1 In 2009, 5-hydroxymethylcytosine (5-hmC), an oxidized
Figure 1. Two reported phosphoramidite building blocks of 5-hydro-
form of 5-mC, was discovered in substantial amounts in mam-
xymethyl-20-deoxycytidine.
malian genome of certain cell types as the “sixth” base.2 Recent
studies have shown that 5-hmC is a widespread DNA modifica-
So far, two phosphoramidite building blocks (I7 and II,8
tion in the brain tissues and stem cells, and its abundance is
dependent on tissue type.3 A group of Tet dioxygenases have
been shown to utilize dioxygen to oxidize 5-mC to 5-hmC in the
mammalian genome and display important functions in the
maintenance and normal myelopoiesis of embryonic stem cells
(ES cells).2b,4 We have recently developed an efficient method to
specifically label and pull down 5-hmC-containing DNA frag-
ments for subsequent deep sequencing to reveal the genome-
wide distribution of 5-hmC, which in mouse cerebellum appears
to be enriched in a gene expression level-dependent manner.5 All
results so far suggested that 5-hmC is an important epigenetic
modification.6 In order to facilitate the biological study of 5-hmC
in DNA, an efficient synthesis of 5-hmC-containing DNA as
model substrates is a prerequisite. In addition, DNA with other
labeling such as fluorophores or biotin is usually needed for
biochemical characterizations, whose incorporation into DNA
often requires or prefers the use of ultramild DNA synthesis.
Therefore, it is also highly desirable to develop 5-hmC phos-
phoramidite building blocks that are compatible with ultramild
DNA synthesis.
Figure 1) have been developed for the incorporation of 5-hmC
into DNA. These methods are currently used in the synthesis of
5-hmC-containing DNAs for biological studies. However, both
methods pose limitations in the postsynthetic removal of the
protecting groups.
Phosphoramidite I uses Ac as the protecting group for the
5-hydroxymethyl (5-CH2OH) group. After its incorporation
into DNA, the postsynthetic treatment with NH4OH generates
an amide byproduct, which is produced by the SN2 reaction of
ammonia attacking the pseudobenzylic carbon with the OAc
group as the leaving group to give an amine intermediate
followed by migration of the benzoyl group.7 To avoid the
formation of this byproduct, strong base treatment (0.1 M
NaOH in dioxane/H2O) followed by ammonolysis is required.
Phosphoramidite II with a cyanoethyl as the protecting group of
5-CH2OH is more widely used since it is commercially available.
However, the postsynthetic removal of the cyanoethyl group
Received: March 14, 2011
Published: April 04, 2011
r
2011 American Chemical Society
4182
dx.doi.org/10.1021/jo200566d J. Org. Chem. 2011, 76, 4182–4188
|

The Journal of Organic Chemistry
NOTE
Scheme 1. Synthesis of Intermediate 5
turned out to be troublesome.8 Treatment of synthetic DNA
with NH4OH at 65 °C for 60 h could not completely remove the
protecting group. An additional treatment with stronger base
(NaOMe in MeOH) is necessary. Obviously, neither of the
phosphoramidites reported is suitable for synthesizing 5-hmC-
containing DNA with additional base-labile modifications.
Therefore, it is highly desirable to develop new 5-hmC phos-
phoramidites using protecting groups that can be cleanly re-
moved under mild conditions. Herein we report the efficient
syntheses of two 5-hmC phosphoramidite building blocks (III
and IV) and demonstrated advantages in the removal of the
protecting groups following oligo synthesis.
We first chose to synthesize phosphoramidite III with
TBDMS and Bz as 5-CH2OH and exocyclic amino protecting
groups, respectively (Scheme 1). Unlike OAc at the pseudo-
benzylic position, O-TBDMS is not a good leaving group and
thus will not result in the formation of the SN2 byproduct during
NH4OH treatment. TBDMS can be readily removed using by
fluoride treatment such as TBAF.
to explore more efficient synthesis of 5. Crouch et al. reported
that 5-iodo-30,50-di-O-TBDMS-20-deoxyuridine could be con-
verted to the corresponding 5-formyl-dU analogue in 82% yield
using standard Stille coupling conditions.14 We reasoned that the
reduction of the aldehyde analogue should generate the corre-
sponding alcohol. Our new synthesis started from a commercial
reagent, 5-iodo-20-deoxyuridine (6, Scheme 2). We chose to
protect the 30 and 50-hydroxyls with di-tert-butylsilyl group so
that it can be selectively removed after 5-CH2OH is protected by
TBDMS. Thus, 6 was converted to 7 in 92% yield. Stille reaction
generated the corresponding 5-formyl-dU analogue 8 in 85%
yield. Reduction of 8 with NaBH4 in the presence of CeCl3 gave
the corresponding 5-hmdC analogue 9 in 78% yield. Protection
of the hydroxyl group with TBDMS provided 5 in 95% yield. The
product for each step could be easily purified using low polarity
eluents. The overall yield of 5 was improved to 58%.
To prepare 5-hmC phosphoramidite III, we next converted 5
into the 5-hmdC analogue by first treating 5 with POCl3 and
1,2,4-triazole to generate a 4-triazolyl intermediate, followed by
reacting with NH4OH in dioxane to generate the amine 10 in
65% yield in two steps. Protection of the exocyclic amine
provided 11 in 80% yield. Selective removal of the 30,50-silyl
protecting group of 11 with HF in pyridine generated 12 in 84%
yield. Selective protection of the 50-OH of 12 with DMTr (13,
82%) and subsequent phosphitylation of the 30-OH by standard
procedure gave phosphoramidite III (87%). The synthesis en-
tails 9 steps in 18% overall yield from 6.
So far, all of the syntheses of phosphoramidite building blocks
I, II, and III required a multiple-step transformation of convert-
ing 5-hmU derivatives to the corresponding 5-hmC analogues in
two steps and an additional step to protect the resulting exocyclic
amine with Bz. To simplify the synthesis, we chose to synthesize
another version of phosphoramidite IV that bears 4-triazoylide
instead of 4-BzNH. Since 4-triazolide can be converted simulta-
neously into 4-NH2 during postsynthetic NH4OH treatment,
two steps of reaction could therefore be spared. The same
strategy has been successfully used for synthesizing nucleotides
containing 4-NH2, 4-MeO, 4-EtO, 4-Me2N-NH, and 4-SH
analogues.15 Besides allowing for the incorporation of 5-hmdC
into DNA, phosphoramidite IV may also find applications in
synthesizing DNA containing 4-substituted 5-hydroxymethyl-
pyrimidine derivatives.
5-Hydroxymethyl-20-deoxyuridine (5-hmdU, 2) is commer-
cially available but relatively expensive;9 therefore, we chose to
start from 20-deoxyuridine (dU, 1) and converted it to 2 in 60%
yield (Scheme 1).10 Both the 5- and 50-hydroxyl of 2 are primary
alcohols; however, the 5-hydroxyl is more reactive than 50-
hydroxyl since it is in the pseudobenzylic position.11 Conte
et al. reported that 5-CH2OH of 2 could be selectively protected
with TBDMS. When 2 was treated with AgNO3 and TBDMS-Cl
in the presence of pyridine and THF, a mixture of the desired 3
and its regioisomer 4 was obtained in a 4:1 ratio in 82% combined
yield.12 Unfortunately, these two isomers could not be resolved
by silica gel chromatography using various eluent systems.13
Instead we successfully resolved the two isomers by treating the
mixture of 3 and 4 with di-tert-butylsilyl bistriflate in DMF. A
cyclic six-membered ring was formed for 3 to generate 5; in
contrast, a cyclic silyl ring between 30 and 5 hydroxyls of 4 could
not form. Thus, compound 5 was obtained in pure form in 40%
yield from 2 after silica gel chromatography (overall yield from 1
was 24%).
Although we were able to prepare the key intermediate 5 using
the above method, several drawbacks such as the difficulty in
purifying 2 due to its high polarity, the formation of the undesired
regioisomer 4, and the unsatisfactory overall yield prompted us
4183
dx.doi.org/10.1021/jo200566d |J. Org. Chem. 2011, 76, 4182–4188

The Journal of Organic Chemistry
NOTE
Scheme 2. Synthesis of Phosphoramidite III from 6
Scheme 3. Synthesis of Phosphoramidite IV
The synthesis of phosphoramidite IV started from intermedi-
ate 5 (Scheme 3). Selective removal of the 30,50-silyl protecting
group of 5 with HF in pyridine generated 14 in 88% yield.
Selective protection of the 50-OH of 14 with DMTr (15, 85%)
and subsequent phosphitylation of the 30-OH gave phosphor-
amidite 16 in 92% yield. Phosphoramidite 16 was quantitatively
converted to phosphoramidite IV, although it has a silylated
hydroxymethyl group in the adjacent 5-position.8a Phosphora-
midite IV was also stable during silica gel chromatography, and it
was isolated in 81% yield. The synthesis entails 8 steps, and the
overall yield from 1 reached 32%.
Phosphoramidites III and IV were then incorporated into a
short model sequence 50- TCXGA (X = 5-hmC) using ultramild
reagents with the modified phosphoramidites using double
4184
dx.doi.org/10.1021/jo200566d |J. Org. Chem. 2011, 76, 4182–4188

The Journal of Organic Chemistry
NOTE
allow the use of ultramild deprotection of the synthetic DNA. Thus,
after treating the solid resin bearing newly synthesized ODN-2 with
NH4F (0.5 M in MeOH) overnight at room temperature to remove
the TBDMS group and rinsing with MeOH to get rid of excess
NH4F, the ultramild deprotection condition was applied (NH4OH
treatment at rt for 2 h). HPLC analysis showed that all of the
protecting groups were cleanly removed and the desired product
(peak b) was produced as the major product (Figure 2D). The same
treatment of ODN-1 also gave fully deprotected oligo. This is
surprising because 4-benzoyl protecting group of cytidine is usually
not compatible with ultramild deprotection conditions; probably
the 5-CH2OH facilitate the removal of benzoyl group in the
presence of NH4OH.
In summary, we have achieved the syntheses of two novel
5-hmC phosphoramidites III and IV with TBDMS as the
5-Hydroxymethyl protecting group from 5-iodo-20-deoxyuridine
in good overall yields (18% for III and 32% for IV). Complete
removal of the TBDMS protection group was accomplished
conveniently by a “one-step” procedure (overnight treatment
with NH4OH at 65 °C) or a “two-step” procedure (NH4F
treatment followed by the ultramild NH4OH treatment). These
two deprotection procedures are complementary. While the
“one-step” procedure can be used conveniently for the synthesis
of DNA with the 5-hmC modification, the “two-step” procedure
is more suitable for ultramild synthesis of DNA containing
5-hmC and additional base-labile modifications. The application
of the new phosphoramidite building blocks will greatly facilitate
the biological study of 5-hmC in DNA. While the current work
was going on, another 5-hmC phosphoramidite using a carba-
mate to protect simultaneously the amino and hydroxyl groups
was reported.16 The synthesis derives from 5-iodo-deoxycytidine
and entails seven steps of reaction and gave the phosphoramidite
in 7.5% overall yield. Although a significant advance compared to
the commercial available phosphoramidite, the synthesis is still
not compatible with ultramild deprotection conditions.
Figure 2. (A) ODN-2 was deprotected by treatment of NH4OH
overnight at rt. Peak a is a failed sequence dimer, 50-CT (Maldi ms
[MH]þ = 531.1); peak b is the fully deprotected 5mer 50-GAXCT
(Maldi ms [MH]þ = 1493); peak c is the 5mer with TBDMS protecting
group on 5hmC (Maldi ms [MH]þ = 1607). (B) ODN-2 was
deprotected by treatment of NH4OH overnight at 65 °C. (C) ODN-3
was deprotected by treatment of NH4OH overnight at 65 °C; peak d is
the 5mer with cyanoethyl protecting group on 5hmC (Maldi ms [MH]þ
=
1546). (D) ODN-2 was deprotected by treatment of 0.5 M NH4F in
MeOH overnight followed by NH4OH treatment at rt for 2 h.
coupling to give oligodeoxynucleotides ODN1 and ODN2,
respectively. As comparion, phosphoramidite II from Glen
Research was incorporated into the same sequence under the
same conditions to give ODN3. After the postsynthetic treat-
ment of ODN2 with NH4OH at room temperature overnight,
reverse phase HPLC analysis (Figure 2A) showed that, besides
DNA with a TBDMS protecting group (peak c, tR = 34 min),
over 20% of DNA without TBDMS group was also produced
(peak b, tR = 22 min). It is interesting to note that the 5-CH2O-
TBDMS group is much more labile to the NH4OH treatment
than the regular 20-O-TBDMS protecting group in RNA synth-
esis, which can survive the overnight treatment at 55 °C. This
observation is consistent with the enhanced reactivity of the
5-CH2OH in 2 as indicated by selective protection of 5-CH2OH
by TBDMS or Ac against the 50-hydroxyl.11,12 The liability of the
5-CH2O-TBDMS ether to the NH4OH treatment suggested that
it is possible to fully remove it from the oligo at elevated
temperature so that the extra fluoride treatment might be spared.
Indeed, when ODN 2 was treated with NH4OH at 65 °C for 16 h,
HPLC analysis indicated that TBDMS was close to fully removed
(Figure 2B). The same treatment of ODN 1 also gave the same
major desired product. In contrast, ODN-3 only gave about 56%
of the desired product, in addition to another 44% with the
5-cyanoethyl protecting group still on (Figure 2C), suggesting
that the TBDMS is a superior protecting group to the cyano-
theyl group.
’ EXPERIMENTAL SECTION
30,50-O-Di-tert-butylsilyl-5-iodo-20-deoxyuridine (7). To a
cloudy solution of 6 (2.05 g, 5.79 mmol) in DMF (15 mL) at 0 °C was
added di-tert-butylsilyl-bis(trifluoromethanesulfonate) (2.39 mL, 1.1
equiv). After 10 min of stirring at rt, imidazole (0.98 g, 2.5 equiv) was
added, and the mixture was stirred at rt for 0.5 h. After removal of DMF
under high vacuum, the residue was dissolved in ethyl acetate, washed
with water, 5% NaHCO3 solution, and brine, and dried over Na2SO4.
After evaporation of the solvent under reduced pressure, the residue was
purified by silica gel chromatography, eluting with 1ꢀ3% MeOH in
CH2Cl2, to give 7 (2.63 g, 92%) as white foam. 1H NMR (500 MHz)
(CDCl3) δ: 9.86 (br., 1H), 7.66 (s, 1H), 6.13 (m, 1H), 4.46 (m, 1H),
4.20 (m, 1H), 4.01 (m, 1H), 3.71 (m, 1H), 2.41 (m, 2H), 1.08 (s, 9H),
0.99 (s, 9H). 13C NMR (125.8 MHz) (CDCl3) δ: 160.2, 150.0, 144.2,
84.8, 78.3, 74.6, 69.0, 67.3, 39.0, 27.5, 27.2, 22.8, 20.2. HRMS calculated
for C17H28IN2O5Si, [MHþ] 495.0812 (calcd), 495.0807 (found).
30,50-O-Di-tert-butylsilyl-5-formyl-20-deoxyuridine (8). To
a solution of 7 (2.01 g, 4.06 mmol) in anhydrous THF (50 mL) in a flask
with a self-contained glass coupling apparatus equipped with a pressure
equalizing addition funnel were added triphenylphosphine (0.64 g, 0.6
equiv) and Pd2(dba)3 CHCl3 (0.42 g, 0.40 mmol, 0.10 equiv). The
3
apparatus was charged with 50 psi of CO and heated to 70 °C, and
Bu3SnH (1.19 mL, 1.05 equiv) was added slowly with a syringe within 1
h. After that, the mixture was kept stirring for 2 h at 70 °C. After cooled
to rt, the small amount of solid was removed by filtration. The solvent of
the filtrate was removed under reduced pressure, and the residue was
The above one-step deprotection procedure is very convenient
for the synthesis of DNA containing 5-hmC. For DNA containing
additional labeling modifications that requires milder deprotection,
a prior treatment with fluoride to remove the TBDMS group may
4185
dx.doi.org/10.1021/jo200566d |J. Org. Chem. 2011, 76, 4182–4188

The Journal of Organic Chemistry
NOTE
purified by silica gel chromatography, eluting with 2ꢀ3% MeOH in
removal of the solvents under the reduced pressure, the residue was
dissolved in CH2Cl2 (100 mL) and washed with water, 5% NaHCO3,
and brine. The organic phase was dried over Na2SO4 and concentrated
to dryness. The residue was purified by silica gel chromatography,
eluting with 1ꢀ2% MeOH in CH2Cl2, to give 11 (769 mg, 80%) as white
foam. 1H NMR (500 MHz) (CDCl3) δ: 8.24 (m, 2H), 7.58 (s, 1H), 7.54
(m, 1H), 7.45 (m, 2H), 6.29 (m, 1H), 4.70 (m, 2H), 4.48 (m, 1H), 4.22
(m, 1H), 4.04 (m, 1H), 3.85 (m, 1H), 2.41 (m, 2H), 1.08 (s, 9H), 1.06
(s, 9H), 0.96 (s, 9H), 0.18 (s, 6H). 13C NMR (125.8 MHz) (CD3OD) δ:
158.4, 148.7, 137.9, 136.5, 133.6, 131.6, 130.8, 129.9, 129.1, 116.3, 85.3,
79.2, 75.7, 68.2, 59.4, 39.9, 28.4, 28.1, 27.0, 23.6, 21.1, 19.3, ꢀ4.36,
ꢀ4.38. HRMS calculated for C31H50N3O6Si2, [MHþ] 616.3238
(calcd), 616.3233 (found).
1
CH2Cl2, to give 8 (1.37 g, 85%) as pale foam. H NMR (500 MHz)
(CDCl3) δ: 10.03 (s, 1H), 8.60 (br., 1H), 8.21 (s, 1H), 6.13 (m, 1H),
4.49 (m, 1H), 4.17 (m, 1H), 4.08 (m, 1H), 3.77 (m, 1H), 2.45 (m, 2H),
1.07 (s, 9H), 1.04 (s, 9H). 13C NMR (125.8 MHz) (CDCl3) δ: 186.7,
162.1, 149.7, 145.3, 112.1, 86.5, 79.6, 75.1, 68.1, 40.1, 28.4, 28.1, 23.6,
21.1. HRMS calculated for C18H29N2O6Si, [MHþ] 397.1795 (calcd),
397.1789 (found).
30,50-O-Di-tert-butylsilyl-5-hydroxymethyl-20-deoxyuridine
(9). To a solution of 8 (1.50 g, 3.79 mmol) in MeOH (40 mL) was
added CeCl3.7H2O (4.24 g, 3.0 equiv) under argon, and the mixture was
cooled to 0 °C with an ice-bath. NaBH4 (144 mg, 1.0 equiv) was added
slowly within 15 min. After that, the ice-bath was removed, the mixture
was stirred at rt for 0.5 h, and TLC showed that 8 was completely
consumed. Silica gel (5.0 g) was added, and MeOH was removed under
reduced pressure. The residue was purified by silica gel chromatography,
eluting with 8ꢀ12% MeOH in CH2Cl2, to give 9 as pale foam (1.18 g,
78%). 1H NMR (500 MHz) (CDCl3) δ: 10.10 (br., 1H), 7.34 (s, 1H),
6.20 (t, J = 5.0 Hz, 1H), 4.34ꢀ4.45 (m, 3H), 4.21 (m, 1H), 4.01 (m, 1H),
3.68 (m, 1H), 2.36 (m, 2H), 1.06 (s, 9H), 1.00 (s, 9H). 13C NMR (125.8
MHz) (CDCl3) δ: 164.7, 151.1, 138.1, 115.3, 85.0, 79.0, 75.7, 68.2, 59.2,
39.6, 28.4, 28.1, 23.6, 21.1. HRMS calculated for C18H31N2O6Si,
[MHþ] 399.1951 (calcd), 399.1946 (found).
5-tert-Butyldimethylsiloxymethyl-N4-benzoyl-20-deoxy-
cytidine (12). To a solution of 11 (720 mg, 1.17 mmol) in THF
(20 mL) was added HF in pyridine (1.0 equiv) under argon, and the
mixture was stirred at rt for 1 h. Silica gel (4.0 g) was added, and the
solvent was removed under reduced pressure. The residue was purified
by silica gel chromatography, eluting with 3ꢀ7% MeOH in CH2Cl2, to
give 12 (476 mg, 84%) as white foam. 1H NMR (500 MHz) (acetone-
d6) δ: 8.26 (m, 2H), 8.15 (s, 1H), 7.56 (m, 1H), 7.47 (m, 2H), 6.34 (m,
1H), 4.73 (s, 2H), 4.52 (m, 1H), 4.03 (m, 1H), 3.80 (m, 1H), 2.39 (m,
1H), 2.30 (m, 1H), 0.97 (s, 9H), 0.18 (s, 6H). 13C NMR (125.8 MHz)
(acetone-d6) δ: 159.7, 148.6, 139.6, 138.2, 133.3, 130.6, 129.0, 114.9,
89.1, 87.0, 72.3, 63.0, 59.4, 41.47, 26.5, 19.1, ꢀ5.0. HRMS calculated for
C23H34N3O6Si, [MHþ] 476.2217 (calcd), 476.2211 (found).
50-O-(4,40-Dimethoxytrityl)-5-tert-butyldimethylsiloxy-
methyl-N4-benzoyl-20-deoxycytidine (13). To a solution of 12
(412 mg, 0.87 mmol) in pyridine (12 mL) was added DMTr-Cl (352
mg, 1.04 mmol, 1.2 equiv) under argon. The mixture was stirred
overnight at rt. MeOH (1 mL) was added to quench the reaction. After
concentrated to dryness, the residue was dissolved in CH2Cl2 (100 mL),
and the mixture was washed with water, 5% NaHCO3, and brine, dried
over Na2SO4, and concentrated to dryness. The residue was purified by
silica gel chromatography, eluting with 1ꢀ3% MeOH in CH2Cl2, to give
13 (553 mg, 82%) as white foam. 1H NMR (500 MHz) (CD3CN) δ:
8.24 (d, J = 8.0 Hz, 2H), 7.71 (s, 1H), 7.24ꢀ7.58 (m, 12H), 6.87 (m,
4H), 6.22 (m, 1H), 4.57 (m, 1H), 4.40 (m, 1H), 4.28 (m, 1H), 3.98 (m,
1H), 3.76 (s, 6H), 3.34 (m, 1H), 3.29 (m, 1H), 3.22 (m, 1H), 2.39 (m,
1H), 2.18 (m, 1H), 0.83 (s, 9H), 0.05 (s, 3H), 0.04 (s, 3H). 13C NMR
(125.8 MHz) (CD3CN) δ: 159.7, 146.0, 139.3, 137.8, 136.9, 136.8,
133.5, 131.0, 130.9, 130.4, 129.2, 129.0, 128.9, 127.9, 114.1, 87.3, 86.8,
86.6, 71.9, 64.8, 59.5, 55.9, 40.8, 26.3, 18.9, ꢀ5.1. HRMS calculated for
C44H52N3O8Si, [MHþ] 778.3524 (calcd), 778.3525 (found).
50-O-(4,40-Dimethoxytrityl)-5-tert-butyldimethylsiloxy-
methyl-N4-benzoyl-20-deoxycytidine 30-O-(2-cyanoethyl-N,
N-diisopropyl)phosphoramidite (III). To a stirring solution of
13 (120 mg, 0.15 mmol) in CH2Cl2 (8 mL) were added 1-methylimi-
dazole (3.40 mg, 41.5 μmol) and N,N-diisopropylethylamine (0.2 mL)
under argon followed by 2-cyanoethyl N,N-diisopropylchlorophosphor-
amidite (71.7 mg, 0.30 mmol). After the reaction mixture was stirred at
room temperature for 0.5 h, CH2Cl2 (80 mL) was added. The mixture
was washed with 5% aqueous NaHCO3 and brine, dried over Na2SO4,
and concentrated to dryness. The residue was purified by silica gel
chromatography, eluting with 6ꢀ9% acetone in CH2Cl2 containing
0.2% Et3N, to give III (131 mg, 87%) as white foam. 1H NMR
(500 MHz) (CD3CN) δ: 8.30 (m, 2H), 7.31ꢀ7.72 (m, 13H),
6.87 (m, 4H), 6.18ꢀ6.28 (m, 1H), 4.62 (m, 1H), 4.40ꢀ4.50 (m, 2H),
4.15 (m, 1H), 3.76 (m, 6H), 3.65 (m, 4H), 3.32 (m, 2H), 2.65 (m, 2H),
2.54 (m, 2H), 2.28 (m, 1H), 2.16 (m, 1H), 1.17 (m, 9H), 1.05 (m, 3H),
0.84 (s, 9H), 0.04 (s, 6H). 31P NMR (202.5 MHz) (CD3CN) 148.44
ppm. HRMS calcd for C53H69N5O9PSi, [MH]þ 978.4602 (calcd),
978.4611 (found).
30,50-O-Di-tert-butylsilyl-5-tert-butyldimethylsiloxymethyl-
20-deoxyuridine (5). To a solution of 9 (1.00 g, 2.51 mmol) in
DMF (20 mL) were added TBDMS-Cl (454 mg, 1.2 equiv) and
imidazole (427 mg, 2.5 equiv). The mixture was heated to 60 °C for 2
h under argon. DMF was then removed under high vacuum, and to the
residue was added ethyl acetate (100 mL). The solution was washed
with water, 5% NaHCO3, and brine and dried over Na2SO4. After
filtration, the solvent of the filtrate was removed under reduced pressure,
and the residue was purified by silica gel chromatography, eluting with
1
1ꢀ2% MeOH in CH2Cl2, to give 5 (1.22 g, 95%) as white foam. H
NMR (500 MHz) (CDCl3) δ: 9.69 (br., 1H), 7.33 (s, 1H), 6.30 (m,
1H), 4.39ꢀ4.43 (m, 3H), 4.20 (m, 1H), 3.99 (m, 1H), 3.68 (m, 1H),
2.36 (m, 2H), 1.05 (s, 9H), 1.01 (s, 9H), 0.93 (s, 9H), 0.08 (s, 6H). 13C
NMR (125.8 MHz) (CDCl3) δ: 163.3, 151.2, 135.6, 116.1, 84.6,
78.9, 75.9, 68.2, 59.1, 39.6, 28.4, 28.1, 26.9, 23.6, 21.1, 19.2, ꢀ4.48,
ꢀ4.49. HRMS calculated for C24H45N2O6Si, [MHþ] 485.3047 (calcd),
485.3041 (found).
30,50-O-Di-tert-butylsilyl-5-tert-butyldimethylsiloxymethyl-
20-deoxycytidine (10). To a solution of 5 (1.00 g, 2.51 mmol) in
CH3CN (20 mL) were added Et3N (8.56 mL, 55.3 mmol, 22 equiv) and
1,2,4-triazole (3.47 g, 50.5 mmol, 20.0 equiv), and the mixture was
cooled to 0 °C. POCl3 (0.60 mL, 6.27 mmol, 2.5 equiv) was added, and
the mixture was stirred at 0 °C for 0.5 h and at rt for 1 h. CH2Cl2
(100 mL) was added, and the mixture was washed with water, 5%
NaHCO3, and brine, dried over Na2SO4, and concentrated to dryness.
The residue was dissolved in 1,4-dioxane (10 mL) and cooled to 0 °C,
and NH4OH (1 mL) was then added. The mixture was stirred at 0 °C for
1 h and then concentrated to dryness. The residue was purified by silica
gel chromatography, eluting with 2ꢀ4% MeOH in CH2Cl2, to give 10
(835 mg, 65%) as white foam. 1H NMR (500 MHz) (CDCl3) δ: 7.28 (s,
1H), 6.17 (m, 1H), 4.36ꢀ4.46 (m, 3H), 4.06 (m, 1H), 3.97 (m, 1H),
3.71 (m, 1H), 2.36 (m, 2H), 1.02 (s, 9H), 0.98 (s, 9H), 0.86 (s, 9H), 0.06
(s, 6H). 13C NMR (125.8 MHz) (CDCl3) δ: 165.9, 156.3, 138.0, 106.5,
85.9, 79.0, 75.5, 68.5, 61.4, 40.3, 28.4, 28.1, 26.8, 23.6, 21.0, 19.2, ꢀ4.3.
HMRS calculated for C24H46N3O5Si2, [MHþ] 512.2976 (calcd),
512.2971 (found).
30,50-O-Di-tert-butylsilyl-5-tert-butyldimethylsiloxymethyl-
N4-benzoyl-20-deoxycytidine (11). To a solution of 10 (800
mg, 1.56 mmol) in pyridine (12 mL) was added benzoyl chloride
(0.22 mL, 1.87 mmol, 1.2 equiv) under argon. The mixture was stirred
overnight at rt. MeOH (1 mL) was added to quench the reaction. After
4186
dx.doi.org/10.1021/jo200566d |J. Org. Chem. 2011, 76, 4182–4188

The Journal of Organic Chemistry
NOTE
5-tert-Butyldimethylsiloxymethyl-20-deoxyuridine (14). To a
solution of 5 (440 mg g, 0.86 mmol) in THF (20 mL) was added HF in
pyridine (1.0 equiv) under argon, and the mixture was stirred at rt for 1 h.
Silica gel (4.0 g) was added and the solvent was removed under reduced
pressure. The residue was purified by silica gel chromatography, eluting with
3ꢀ7% MeOH in CH2Cl2, to give 14 (281 mg, 88%) as pale foam. 1H NMR
(500 MHz) (acetone-d6) δ: 10.07 (br., 1H), 7.83 (s, 1H), 6.34 (m, 1H),
4.50 (s, 1H), 4.42 (s, 3H), 4.12 (m, 1H), 3.95 (m, 1H), 3.75 (m, 2H), 2.26
(m, 2H), 0.92 (s, 9H), 0.12 (s, 6H). 13C NMR (125.8 MHz) (acetone-d6)
δ: 163.0, 151.3, 137.7, 114.7, 88.7, 85.9, 72.5, 63.2, 59.2, 41.0, 26.4, 19.0,
ꢀ5.1. HRMS calculated for C16H29N2O6Si, [MHþ] 373.1795 (calcd),
373.1789 (found).
50-O-(4,40-Dimethoxytrityl)-5-tert-butyldimethylsiloxy-
methyl-20-deoxyuridine (15). To a solution of 14 (260 mg, 0.70
mmol) in pyridine (8 mL) was added DMTr-Cl (284 mg, 0.84 mmol, 1.2
equiv) under argon. The mixture was stirred overnight at rt. MeOH
(1 mL) was added to quench the reaction. The solvents were removed
under reduced pressure, and the residue was dissolved with CH2Cl2
(100 mL). The mixture was washed with water, 5% NaHCO3, and brine,
dried over Na2SO4, and concentrated to dryness. The residue was
purified by silica gel chromatography, eluting with 1ꢀ3% MeOH in
CH2Cl2, to give 15 (400 mg, 85%) as pale foam. 1H NMR (500 MHz)
(CD3CN) δ: 9.34 (br., 1H), 7.46 (m, 3H), 7.24ꢀ7.35 (m, 7H), 6.87 (m,
4H), 6.23 (m, 1H), 4.29 (m, 2H), 4.12 (m, 1H), 3.95 (m, 1H), 3.77 (s,
6H), 3.29 (m, 1H), 3.20 (m, 1H), 2.30 (m, 1H), 2.19 (m, 1H), 0.84 (s,
9H), 0.02 (s, 3H), 0.01 (s, 3H). 13C NMR (125.8 MHz) (CD3CN) δ:
163.3, 159.7, 151.3, 146.0, 137.6, 136.8, 131.0, 128.98, 128.90, 127.89,
118.3, 115.0, 114.1, 87.3, 86.0, 85.4, 72.1, 65.0, 59.0, 55.9, 40.5, 26.3,
18.9, 7.5. HRMS calculated for C37H47N2O8Si, [MHþ] 675.3102
(calcd), 675.3096 (found).
1H), 2.25 (m, 2H), 1.17 (m, 9H), 1.07 (d, 3H), 0.79 (s, 9H), ꢀ0.04,
ꢀ0.03 (s, 3H), 0.00 (s, 3H). 13C NMR (125.8 MHz) (CD3CN) δ: 159.7,
154.5, 154.5, 147.1, 147.0, 145.6, 136.6, 131.0, 131.0, 129.0, 128.9,
127.98, 127.95, 114.2, 89.9, 88.8, 89.0, 88.1, 88.0, 74.6, 74.4, 64.2, 60.9,
60.9, 59.5, 59.3, 55.9, 55.9, 44.1, 44.02, 44.00, 40.1, 40.1, 26.3, 26.2,, 25.0,
24.9, 24.8, 21.0, 18.6, ꢀ5.15, ꢀ5.16. 31P NMR (202.5 MHz) (CD3CN)
148.31, 148.24 ppm. HMRS calculated for C48H65N7O8PSi, [MHþ]
926.4402 (calcd), 926.4396 (found).
’ ASSOCIATED CONTENT
S
Supporting Information. Analytical data for compounds
b
5, 7ꢀ13, III, and IV including their 1H and 13C NMR spectra.
This material is available free of charge via the Internet at http://
pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: daiqing@uchicago.edu.
’ ACKNOWLEDGMENT
This work is supported by NIH (GM071440 and GM088599)
and the Chicago Biomedical Consortium with support from The
Searle Funds at The Chicago Community Trust.
’ REFERENCES
50-O-(4,40-Dimethoxytrityl)-5-tert-butyldimethylsiloxy-
methyl-20-deoxyuridine 30-O-(2-cyanoethyl-N,N-diisopro-
pyl)phosphoramidite (16). To a stirring solution of 15 (120 mg,
0.18 mmol) in CH2Cl2 (8 mL) were added 1-methylimidazole (3.40 mg,
41.5 μmol) and N,N-diisopropylethylamine (0.2 mL) under argon
followed by 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (60
μL, 0.27 mmol, 1.5 equiv). After the reaction mixture was stirred at room
temperature for 0.5 h, CH2Cl2 (80 mL) was added. The mixture was
washed with 5% aqueous sodium carbonate and brine, dried over sodium
sulfate, and concentrated to dryness. The residue was purified by silica
gel chromatography, eluting with 6ꢀ9% acetone in CH2Cl2 containing
(1) (a) Goll, M. G.; Bestor, T. H. Annu. Rev. Biochem. 2005, 74, 481.
(b) Klose, R. J.; Bird, A. P. Trends Biochem. Sci. 2006, 31, 89. (c) Reik, W.
Nature 2007, 447, 425. (d) Weber, M.; Schubeler, D. Curr. Opin. Cell
Biol. 2007, 19, 273. (e) Gal-Yam, E. N.; Saito, Y.; Egger, G.; Jones, P. A.
Annu. Rev. Med. 2008, 59, 267.
(2) (a) Kriaucionis, S.; Heintz, N. Science 2009, 324, 929. (b) Tahiliani,
M.; Koh, K. P.; Shen, Y.; Pastor, W. A.; Bandukwala, H.; Brudno, Y.;
Agarwal, S.; Iyer, L. M.; Liu, D. R.; Aravind, L.; Rao, A. Science 2009,
324, 930.
(3) (a) M€unzel, M.; Globisch, D.; Br€uckl, T.; Wagner, M.; Welzmil-
ler, V.; Michalakis, S.; M€uller, M.; Biel, M.; Carell, T. Angew. Chem., Int.
Ed. 2010, 49, 5375. (b) Szwagierczak, A.; Bultmann, S.; Schmidt, C. X.;
Spada, F.; Leonhardt, H. Nucleic Acids Res. 2010, 38, 181.
(4) (a) Ito, S.; D’Alessio, A. C.; Taranova, O. V.; Hong, K.; Sowers,
L. C.; Zhang, Y. Nature 2010, 466, 1129. (b) Ko, M.; Huang, Y.;
Jankowska, A. M.; Pape, U. J.; Tahiliani, M.; Bandukwala, H. S.; An, J.;
Lamperti, E. D.; Koh, K. P.; Ganetzky, R.; Liu, X. S.; Aravind, L.;
Agarwal, S.; Maciejewski, J. P.; Rao, A. Nature 2010, 468, 839.
(5) Song, C.-X.; Szulwach, K. E.; Fu, Y.; Dai, Q.; Yi, C.; Li, X.; Li, Y.;
Chen, C.-H.; Zhang, W.; Jian, X.; Wang, J.; Zhang, L.; Looney, T. J.;
Zhang, B.; Godley, L. A.; Hicks, L. M.; Lahn, B. T.; Jin, P.; He, C. Nat.
Biotechnol. 2011, 29, 68.
1
0.2% Et3N, to give 16 (143 mg, 92%) as white foam. H NMR (500
MHz) (CD3CN) δ: 9.04 (br., 1H), 7.03ꢀ7.46 (m, 10H), 6.89 (m, 4H),
6.23 (m, 1H), 4.50 (m, 1H), 4.28 (m, 1H), 4.08 (m, 2H), 3.78 (s, 3H),
3.77 (s, 3H), 3.62 (m, 4H), 3.28 (m, 2H), 2.63 (m, 1H), 2.53 (m, 1H),
2.44 (m, 1H), 2.26 (m, 1H), 1.17 (m, 9H), 1.05 (d, 3H), 0.82 (s, 9H),
0.02 (s, 3H), 0.00 (s, 3H). 31P NMR (202.5 MHz) (CD3CN) 148.34,
148.27 ppm. HRMS calculated for C46H66N4O9PSi, [MHþ] 877.4337
(calcd), 877.4331 (found).
50-O-(4,40-Dimethoxytrityl)-5-tert-butyldimethylsiloxy-
methyl-4-triazolothymidine 30-O-(2-cyanoethyl-N,N-diiso-
propyl)phosphoramidite (IV). To a solution of 16 (110 mg, 0.13
mmol) in CH3CN (5 mL) were added Et3N (0.43 mL) and 1,2,4-
triazole (174 mg), and the mixture was cooled to 0 °C. POCl3 (30 μL)
was added, and the mixture was stirred at 0 °C for 0.5 h and at rt for 1 h.
Added another portion of POCl3 (16 μL), and the mixture was stirred at
rt for 5 h. CH2Cl2 (100 mL) was added, and the mixture was washed
with water, 5% NaHCO3, and brine, dried over Na2SO4, and concen-
trated to dryness. The residue was purified by silica gel chromatography,
eluting with 6ꢀ10% acetone in CH2Cl2 containing 0.2% Et3N, to give
IV (94.5 mg, 81%) as white foam. 1H NMR (500 MHz) (CD3CN) δ:
9.18 (s, 1H), 8.50 (s, 0.5H), 8.38 (s, 0.5H), 8.15 (s, 1H), 7.24ꢀ7.42 (m,
9H), 6.87 (m, 4H), 6.18 (m, 1H), 4.92 (m, 1H), 4.55 (m, 2H), 4.22 (m,
1H), 3.75 (s, 6H), 3.73 (m, 4H), 3.40 (m, 2H), 2.66 (m, 1H), 2.56 (m,
(6) Loenarz, C.; Schofield, C. J. Chem. Biol. 2009, 16, 580.
(7) De Kort, M.; De Visser, P. C.; Kurzeck, J.; Meeuwenoord, N. J.;
Van der Marel, G. A.; Ruger, W.; Van Boom, J. H. Eur. J. Org. Chem.
2001, 11, 2075.
(8) (a) Tardy-Planechaud, S.; Fujimoto, J.; Lin, S. S.; Sowers, L. C.
Nucleic Acids Res. 1997, 25, 553. (b) Hansen, A. S.; Thalhammer, A.;
El-Sagheer, A. H.; Brown, T.; Schofield, C. J. Bioorg. Med. Chem. Lett.
2011, 21, 1181.
(9) 5-Hydroxymethyl-20-deoxyuridine is commercially available
from Berry Associate at the price of $450/0.25 g while 20-deoxyuridine
is commercially available at the price of $45/5 g.
(10) Shiau, G. T.; Schinazi, R. F.; Chen, M. S.; Prusoff, W. H. J. Med.
Chem. 1980, 23, 127.
(11) Sowers, L.; Beardsley, P. G. J. Org. Chem. 1993, 58, 1664.
4187
dx.doi.org/10.1021/jo200566d |J. Org. Chem. 2011, 76, 4182–4188

The Journal of Organic Chemistry
NOTE
(12) Conte, M. R.; Galeone, A.; Avizonis, D.; Hsu, V. L.; Mayol, L.;
Kearns, D. R. Bioorg. Med. Chem. Lett. 1992, 2, 79.
(13) Conte, M. R. reported that the two isomers could only be
resolved by reverse phase HPLC, thus it is difficult to scale up. We
also tried to resolve them by converting the mixture of 3 and 4 to their
50-DMTr ether analogues; however, they were still not separable by silica
gel chromatography.
(14) Crouch, G. J.; Eaton, B. E. Nucleoside Nucleotide 1994, 13, 939.
(15) Xu, Y.-Z.; Zheng, Q.; Swann, P. F. J. Org. Chem. 1992, 57, 3839.
(16) Munzel, M.; Globisch, D.; Trindler, C.; Carell, T. Org. Lett.
2010, 12, 5671.
4188
dx.doi.org/10.1021/jo200566d |J. Org. Chem. 2011, 76, 4182–4188